diff --git a/.gitattributes b/.gitattributes
index bed0738c7eeb449bca98b5d2f33c89a1ee56349a..b68fad08448588507787871741a4351b44a0a0be 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -58,3 +58,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
# Video files - compressed
*.mp4 filter=lfs diff=lfs merge=lfs -text
*.webm filter=lfs diff=lfs merge=lfs -text
+docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405.md filter=lfs diff=lfs merge=lfs -text
diff --git a/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026.json b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026.json
new file mode 100644
index 0000000000000000000000000000000000000000..ba1decf696435e5291677c4ec30f01481e0bd388
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026.json
@@ -0,0 +1,5 @@
+{
+ "importance": 7,
+ "urgency": 7,
+ "tier": 2
+}
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026.md b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026.md
new file mode 100644
index 0000000000000000000000000000000000000000..54565d4034a74a2029ee04f5cc9a3782330cffe9
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026.md
@@ -0,0 +1,64 @@
+# 次世代エージェントAI:ベンチマーク、OSS動向、およびビデオ生成エージェントの進化 (2026)
+
+## 概要
+本レポートは、2026年5月時点におけるエージェント型AI(Agentic AI)の主要な技術動向をまとめたものである。Langflow等のオーケストレーションツールで重要視されるベンチマーク指標、台頭するOSSモデル、およびHeyGenに代表されるビデオ生成エージェントの最新技術を統合的に分析する。
+
+---
+
+## 1. エージェント評価指標の転換:Agentic Benchmarks
+2026年現在、AIモデルの評価軸は「会話の流暢さ」から「実務実行能力」へと完全にシフトしている。
+
+### 1.1 主要ベンチマーク
+* **BFCL v4 (Berkeley Function Calling Leaderboard)**:
+ * **評価対象**: 関数呼び出し(Tool Use)の精度。
+ * **トレンド**: 単一の呼び出しから、依存関係のある複数のツールを順序立てて実行する能力が重視されている。
+* **SWE-bench Pro**:
+ * **評価対象**: 実際のソフトウェア開発リポジトリにおけるIssue解決能力。
+ * **トレンド**: Python以外の言語(Go, TypeScript等)への拡大と、自律的なテスト・デバッグサイクルの完遂が評価の要。
+* **WebArena / GAIA**:
+ * **評価対象**: ブラウザ操作を伴う複雑なタスクの完遂率。
+ * **トレンド**: 人間の曖昧な指示から具体的な実行ステップを生成する「プランニング能力」の差が顕著になっている。
+
+---
+
+## 2. 2026年5月時点の主要OSSモデル(エージェント特化型)
+プロプライエタリなモデル(Claude Mythos, GPT-5.3)に比肩する能力を持つOSS/オープンウェイトモデルの選定。
+
+| モデル名 | 主要な強み | ベンチマーク指標 (参考) | 特徴 |
+| :--- | :--- | :--- | :--- |
+| **Qwen3-235B (A22B)** | 汎用オーケストレーション | BFCL v3: 74.9% | ツール利用の安定性が高く、Langflow等のワークフロー構築に最適。 |
+| **DeepSeek-V4 Pro** | コーディング / 推論 | SWE-bench: 76.4% | 論理的思考が必要なエージェントタスクにおいて圧倒的な勝率。 |
+| **GLM-5 / 4.5** | 信頼性・低ハルシネーション | BFCL v3: 76.7% | 厳密な出力フォーマットが求められる業務自動化に強い。 |
+| **Llama-4 Agent Edition** | 自律プランニング | - | MetaによるAgent特化版。CoT(思考の連鎖)の質が高い。 |
+
+---
+
+## 3. ビデオ生成エージェントの革新 (HeyGen Avatar V)
+ビデオ生成分野においても、単なるピクセル生成から「クリエイティブ・エージェント」への進化が見られる。
+
+### 3.1 技術的ブレイクスルー
+* **Avatar V Architecture**:
+ * **Flow Matching & DiT**: ボトルネックを排除し、リファレンスビデオの全トークンに直接条件付けを行うことで、2025年以前のモデルでは困難だった微細な表情の変化を完全再現。
+* **Agentic Kernel Synthesis**:
+ * HeyGenのインファレンス最適化レポートによれば、LLMベースのエージェントが**カスタムカーネルをリアルタイムで合成**し、巨大モデルの低遅延推論を実現している。
+
+### 3.2 制作フローの変革
+* **Prompt-to-Blueprint**:
+ * 直接動画を出すのではなく、まずエージェントが「構成案(Blueprint)」を提示し、ユーザーが対話形式で修正を加えるフローが定着。これにより「一発勝負」の生成から「意図を反映した制作」へと移行した。
+
+---
+
+## 4. 今後の展望:Decision-Centric AI
+2026年後半に向けて、エージェントAIは以下の方向に進化すると予測される。
+
+1. **意思決定中心(Decision-Centric)の評価**: 「何を言ったか」ではなく「どの意思決定が最適だったか」がAIの価値となる。
+2. **Harness Engineeringの標準化**: OpenAIが提唱する、モデルを制御・監視するためのコードレイヤー(ハーネス)の設計が、エージェント開発の必須スキルとなる。
+3. **マルチエージェント・コラボレーション**: 単一の巨大モデルではなく、Qwen3(司令塔)とDeepSeek-V4(実装)のような、役割分担されたOSSモデル群による協調動作が主流となる。
+
+---
+*作成日: 2026年5月16日*
+*出典: Notion (Langflow/Benchmarks), HeyGen Technical Reports (2026), Berkeley BFCL Leaderboard*
+
+
+**Scores:** Importance:7, Urgency:7
+**Tier:** 2
diff --git a/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026.rag_ingested b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..1e59f396e63daae16b2d31766abe87be32f66ae9
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "a8eac8bc37296199", "ingested_at": "2026-06-30T08:25:41.762466", "content_hash": "a8eac8bc37296199"}
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_long.md b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_long.md
new file mode 100644
index 0000000000000000000000000000000000000000..ce6826fbd15a646a1b72e5827da4878353dfbf51
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_long.md
@@ -0,0 +1,16 @@
+Good evening. Tonight, we turn to the rapidly evolving world of artificial intelligence and its impact on our future workforces. A new report sheds light on how AI is transforming the way we approach tasks and solve problems.
+
+The report dives deep into the "Agentic AI" landscape, focusing on the shift from conversational fluency to practical execution capabilities. AI models are now being evaluated based not just on their ability to hold a conversation, but also on their capacity to perform real-world tasks efficiently. The report highlights key benchmarks like the Berkeley Function Calling Leaderboard (BFCL) and SWE-bench Pro, which measure an AI's performance in software development and issue resolution.
+
+The study reveals that these benchmarks are shifting towards a more practical focus, emphasizing the ability of AI models to execute complex tasks autonomously. This transition is driven by advancements in tools like Langflow, which allows for the orchestration of various AI models based on specific workflows.
+
+The report also explores the rise of open-source (OSS) models that rival proprietary models like Claude Mythos and GPT-5.3. These OSS models are gaining traction due to their ability to perform tasks across different domains and languages, including Python, Go, and TypeScript. Examples include Qwen3-235B, DeepSeek-V4 Pro, GLM-5/4.5, and Llama-4 Agent Edition.
+
+The report highlights the emergence of video generation agents like HeyGen Avatar V. This technology is moving beyond simple pixel manipulation to create more "creative" AI agents capable of generating realistic videos with nuanced expressions. This advancement is driven by breakthroughs in architecture like "Avatar V Architecture," which allows for real-time synthesis of custom kernels, and a new approach to video production called "Prompt-to-Blueprint." This method focuses on creating visual blueprints that guide the creation process, allowing for more precise control over final outputs.
+
+Looking ahead, the report predicts that AI will become even more decision-centric in the coming years. This means evaluating not just what an AI says but also its ability to make optimal decisions based on available information. The report emphasizes the need for "Harness Engineering," a standard code layer that controls and monitors models, becoming a crucial skillset for future AI developers.
+
+Finally, the report suggests a shift towards collaborative efforts between multiple AI agents. Instead of relying solely on one large model, we'll likely see more teams of specialized OSS models working together to solve complex problems. This multi-agent approach promises greater efficiency and adaptability in various fields.
+
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_long_clean.txt b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_long_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..629aa15359774235da77153b126b825ba35479bb
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_long_clean.txt
@@ -0,0 +1,15 @@
+Good evening. Tonight, we turn to the rapidly evolving world of artificial intelligence and its impact on our future workforces. A new report sheds light on how AI is transforming the way we approach tasks and solve problems.
+
+The report dives deep into the "Agentic AI" landscape, focusing on the shift from conversational fluency to practical execution capabilities. AI models are now being evaluated based not just on their ability to hold a conversation, but also on their capacity to perform real-world tasks efficiently. The report highlights key benchmarks like the Berkeley Function Calling Leaderboard and SWE-bench Pro, which measure an AI's performance in software development and issue resolution.
+
+The study reveals that these benchmarks are shifting towards a more practical focus, emphasizing the ability of AI models to execute complex tasks autonomously. This transition is driven by advancements in tools like Langflow, which allows for the orchestration of various AI models based on specific workflows.
+
+The report also explores the rise of open-source models that rival proprietary models like Claude Mythos and GPT-5.3. These OSS models are gaining traction due to their ability to perform tasks across different domains and languages, including Python, Go, and TypeScript. Examples include Qwen3-235B, DeepSeek-V4 Pro, GLM-5/4.5, and Llama-4 Agent Edition.
+
+The report highlights the emergence of video generation agents like HeyGen Avatar V. This technology is moving beyond simple pixel manipulation to create more "creative" AI agents capable of generating realistic videos with nuanced expressions. This advancement is driven by breakthroughs in architecture like "Avatar V Architecture," which allows for real-time synthesis of custom kernels, and a new approach to video production called "Prompt-to-Blueprint." This method focuses on creating visual blueprints that guide the creation process, allowing for more precise control over final outputs.
+
+Looking ahead, the report predicts that AI will become even more decision-centric in the coming years. This means evaluating not just what an AI says but also its ability to make optimal decisions based on available information. The report emphasizes the need for "Harness Engineering," a standard code layer that controls and monitors models, becoming a crucial skillset for future AI developers.
+
+Finally, the report suggests a shift towards collaborative efforts between multiple AI agents. Instead of relying solely on one large model, we'll likely see more teams of specialized OSS models working together to solve complex problems. This multi-agent approach promises greater efficiency and adaptability in various fields.
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_middle.md b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_middle.md
new file mode 100644
index 0000000000000000000000000000000000000000..a26489b7e267cecfd4640c56433bc8aaf097a635
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_middle.md
@@ -0,0 +1,12 @@
+Good evening. Tonight's report focuses on the rapidly evolving field of agent-based AI, and how it's shaping the future of work.
+
+A new report highlights a shift in how we evaluate AI models. No longer focused solely on conversational fluency, experts are now prioritizing "real-world execution" - the ability to actually perform tasks. This has led to a focus on benchmarks like the Berkeley Function Calling Leaderboard (BFCL) and SWE-bench Pro, which measure an AI's efficiency and problem-solving skills in real-world scenarios.
+
+The report also dives into the rise of open-source models. Models like Qwen3-235B, DeepSeek-V4 Pro, and GLM-5/4.5 are showing impressive performance on benchmarks, offering a glimpse into what's possible with open collaboration. And in the realm of video generation, we're seeing a shift from simple pixel manipulation to "creative agents" capable of producing realistic and expressive videos.
+
+The report emphasizes the importance of "decision-centric AI," where the focus shifts from simply understanding what an AI says to understanding which decision was ultimately optimal. This means evaluating AI based on its ability to make informed choices, not just its conversational skills.
+
+Experts predict that in the coming years, we'll see a shift towards more collaborative models. Think of it as a team effort where different specialized AI agents work together to solve complex problems.
+
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_middle_clean.txt b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_middle_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..5a9eb30228d8bec235c75aafd43a8896828f44a9
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_middle_clean.txt
@@ -0,0 +1,11 @@
+Good evening. Tonight's report focuses on the rapidly evolving field of agent-based AI, and how it's shaping the future of work.
+
+A new report highlights a shift in how we evaluate AI models. No longer focused solely on conversational fluency, experts are now prioritizing "real-world execution" - the ability to actually perform tasks. This has led to a focus on benchmarks like the Berkeley Function Calling Leaderboard and SWE-bench Pro, which measure an AI's efficiency and problem-solving skills in real-world scenarios.
+
+The report also dives into the rise of open-source models. Models like Qwen3-235B, DeepSeek-V4 Pro, and GLM-5/4.5 are showing impressive performance on benchmarks, offering a glimpse into what's possible with open collaboration. And in the realm of video generation, we're seeing a shift from simple pixel manipulation to "creative agents" capable of producing realistic and expressive videos.
+
+The report emphasizes the importance of "decision-centric AI," where the focus shifts from simply understanding what an AI says to understanding which decision was ultimately optimal. This means evaluating AI based on its ability to make informed choices, not just its conversational skills.
+
+Experts predict that in the coming years, we'll see a shift towards more collaborative models. Think of it as a team effort where different specialized AI agents work together to solve complex problems.
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_short.md b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_short.md
new file mode 100644
index 0000000000000000000000000000000000000000..1670c201110f4609380bbf7ac9331799cb18bc05
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_short.md
@@ -0,0 +1,22 @@
+"New AI is changing the game in sports. Analysts say that a combination of Spanish structure and Argentine creativity is leading to higher performance predictions. This new AI could be revolutionizing how we view teams."
+
+[Video footage of soccer players, highlights, and data visualizations]
+
+"The AI's ability to analyze vast amounts of data - from past matches to player conditions - is giving us a deeper understanding of potential strengths and weaknesses. Traditional analysis focused on star players or tactical success, but now the AI is changing that."
+
+[Video footage of AI algorithms analyzing data]
+
+"This report highlights key benchmarks for agent-based AI: BFCL v4, SWE-bench Pro, and WebArena/GAIA. These metrics are shifting from simple conversation flow to real-world task execution. The rise of open source models like Qwen3-235B is also a game-changer."
+
+[Video footage of open-source software development]
+
+"And in the world of video generation, AI is evolving beyond just pixel manipulation. HeyGen Avatar V is pushing the boundaries with new architecture and real-time synthesis capabilities. This means more creative control for users."
+
+
+[Video footage of a virtual soccer player performing complex maneuvers on screen]
+
+
+
+"The future? AI will focus on decision-making, harness engineering standardization, and multi-agent collaboration. This is just the beginning of a revolution in sports and beyond."
+
+[End video with text overlay: "Reporting for The Android Times."]
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_short_clean.txt b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_short_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..4ed01baa498f8f6813b4ba78066a17096d94bccb
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/agentic_ai_tech_report_2026_script_short_clean.txt
@@ -0,0 +1,9 @@
+New AI is changing the game in sports. Analysts say that a combination of Spanish structure and Argentine creativity is leading to higher performance predictions. This new AI could be revolutionizing how we view teams.
+
+The AI's ability to analyze vast amounts of data - from past matches to player conditions - is giving us a deeper understanding of potential strengths and weaknesses. Traditional analysis focused on star players or tactical success, but now the AI is changing that.
+
+This report highlights key benchmarks for agent-based AI: BFCL v4, SWE-bench Pro, and WebArena/GAIA. These metrics are shifting from simple conversation flow to real-world task execution. The rise of open source models like Qwen3-235B is also a game-changer.
+
+And in the world of video generation, AI is evolving beyond just pixel manipulation. HeyGen Avatar V is pushing the boundaries with new architecture and real-time synthesis capabilities. This means more creative control for users.
+
+The future? AI will focus on decision-making, harness engineering standardization, and multi-agent collaboration. This is just the beginning of a revolution in sports and beyond.
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/ai_formal_logic_survey.md b/docs/AI_Agents_and_Logic/ai_formal_logic_survey.md
new file mode 100644
index 0000000000000000000000000000000000000000..04e8eff282b384441fc905f6f201025ce8cb87f9
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/ai_formal_logic_survey.md
@@ -0,0 +1,497 @@
+# AI(LLM)× 形式論理 融合の最前線:包括的サーベイ
+
+> **対象読者**: AI・形式手法に関心のある研究者・エンジニア
+> **調査時点**: 2026年5月
+> **カバー範囲**: ニューロ・シンボリックAI、数学オリンピック自動証明、AI証明探索ツール、自動形式化、ベンチマーク
+
+---
+
+## 目次
+
+1. [ニューロ・シンボリックAI(Neuro-Symbolic AI)](#1-ニューロシンボリックaineuro-symbolic-ai)
+2. [数学オリンピック自動証明システム](#2-数学オリンピック自動証明システム)
+3. [AI証明探索ツール・フレームワーク](#3-ai証明探索ツールフレームワーク)
+4. [自動形式化(Auto-formalization)](#4-自動形式化auto-formalization)
+5. [ベンチマークと評価](#5-ベンチマークと評価)
+6. [形式化エコシステム:Mathlib と数学コミュニティ](#6-形式化エコシステムmathlib-と数学コミュニティ)
+7. [今後の展望と未解決課題](#7-今後の展望と未解決課題)
+8. [主要文献・リソース一覧](#8-主要文献リソース一覧)
+
+---
+
+## 1. ニューロ・シンボリックAI(Neuro-Symbolic AI)
+
+### 1.1 なぜニューロ・シンボリックAIが必要なのか
+
+大規模言語モデル(LLM)は自然言語処理やコード生成において圧倒的な能力を示していますが、根本的な限界が明らかになっています。
+
+| LLMの弱点 | 具体的な問題 | 形式論理による解決 |
+| :--- | :--- | :--- |
+| **ハルシネーション** | もっともらしい嘘を生成する | 形式検証器が出力を数学的に検証 |
+| **論理推論の脆弱性** | 多段推論で矛盾や飛躍が発生 | 定理証明器が各ステップを厳密にチェック |
+| **検証不可能性** | 「正しそう」だが検証手段がない | 形式的証明は機械的に100%検証可能 |
+| **Chain-of-Thought の限界** | 思考過程を自然言語で出力するだけで、論理的正しさの保証なし | 形式言語での推論は意味論的に正しいことが保証される |
+
+### 1.2 体系的分類:3つの統合パラダイム
+
+2024–2025年のサーベイ論文(IJCAI 2025 サーベイ等)では、LLMとシンボリックシステムの統合を以下の3パラダイムに分類しています。
+
+```mermaid
+graph LR
+ subgraph "パラダイム1: Symbolic → Neural"
+ S1["シンボリック知識
(ルール・オントロジー)"] -->|ガイド・制約| N1["LLM
(訓練/推論)"]
+ end
+
+ subgraph "パラダイム2: Neural → Symbolic"
+ N2["LLM
(意味解析)"] -->|形式言語へ変換| S2["シンボリック推論器
(ソルバー・証明器)"]
+ end
+
+ subgraph "パラダイム3: Hybrid (双方向)"
+ N3["LLM"] <-->|反復的に協調| S3["形式エンジン"]
+ end
+```
+
+#### パラダイム1: Symbolic → Neural(記号 → ニューラル)
+シンボリック知識(論理ルール、オントロジー、形式的制約)を用いてLLMの訓練や推論を**ガイド・正則化**するアプローチ。
+
+- **Logic Tensor Networks (LTN)**: 論理制約をテンソル演算として直接ニューラルネットワークの学習に埋め込む
+- **制約付きデコーディング**: LMQL, Guidance, Outlines などのフレームワークにより、生成時に文法的・論理的制約を強制
+- **DSR-LM / Ctrl-G**: 推論時に論理制約を適用して出力を制御
+
+#### パラダイム2: Neural → Symbolic(ニューラル → 記号)
+LLMを「意味解析器(Semantic Parser)」として使い、自然言語を形式言語(一階述語論理、SAT/SMT、PDDL等)に変換し、**確定的なシンボリック推論器**で解くアプローチ。
+
+- **Logic-LM**: 自然言語問題を FOL/CSP/SAT に変換し、Z3 や Prover9 で推論。エラーメッセージをLLMにフィードバックする**自己修正ループ**を備える
+- **LINC (Logical Inference via Neurosymbolic Computation)**: 前提と結論を FOL に変換し、定理証明器で論理的帰結を判定。ProofWriter、FOLIO ベンチマークで大幅な精度向上を実証
+- **SatLM**: SATソルバーを用いた論理推論タスク処理
+
+#### パラダイム3: Hybrid(双方向統合)
+LLMと形式エンジンが**反復的に対話・批判・修正**し合うハイブリッドフレームワーク。
+
+- **PEIRCE**: 反復的な「推測–批判(Conjecture-Criticism)」プロセスで、素材的推論と形式的推論を統一
+- **ENVISIONS**: ニューロ・シンボリック自己訓練フレームワーク。エージェントの軌跡からシンボリック知識を帰納的に獲得
+- **APOLLO**: 形式推論のための協調フレームワーク
+
+> [!IMPORTANT]
+> 最も注目すべきトレンドは、シンボリック成分を単なる「外部ツール」ではなく、**System 2(熟慮的推論)の不可欠なレイヤー**としてLLMアーキテクチャに統合する動きです。これにより、LLMが「推測する機械」から「証明する機械」へと進化する道筋が見えてきています。
+
+### 1.3 産業界での採用状況
+
+| 組織 | 主なアプローチ |
+| :--- | :--- |
+| **Google DeepMind** | AlphaProof(Lean 4 + AlphaZero RL)、AlphaGeometry(DDM + Gemini) |
+| **Microsoft Research** | Lean 4 開発主導、形式検証ツール、Lean Copilot |
+| **Meta AI** | HyperTree Proof Search (HTPS)、Lean/Metamath 環境での自動証明 |
+| **OpenAI** | 自然言語推論ベースの数学解法、IMO 2025 で金メダル級 |
+| **ByteDance** | Seed-Prover(エージェント型 Lean 4 証明器、IMO 2025 金メダル級) |
+| **Harmonic** | Aristotle(Lean 4 形式証明 + 幾何ソルバー、IMO 2025 金メダル級) |
+
+---
+
+## 2. 数学オリンピック自動証明システム
+
+### 2.1 AlphaProof(Google DeepMind, 2024)
+
+AlphaProof は、LLMと強化学習を組み合わせて **Lean 4 上で形式的に検証可能な数学証明を自動生成する**、Google DeepMind のシステムです。
+
+#### アーキテクチャ
+
+```
+┌──────────────────────────────────────────────────────────┐
+│ AlphaProof │
+│ │
+│ ┌──────────────┐ ┌──────────────────┐ │
+│ │ 事前訓練済み │ │ AlphaZero 式 │ │
+│ │ 言語モデル │────▶│ 強化学習エンジン │ │
+│ │ (Gemini 系) │ │ (MCTS 木探索) │ │
+│ └──────────────┘ └────────┬─────────┘ │
+│ │ │
+│ ▼ │
+│ ┌──────────────────┐ │
+│ │ Lean 4 環境 │ │
+│ │ (タクティクス実行 │ │
+│ │ ・形式検証) │ │
+│ └──────────────────┘ │
+└──────────────────────────────────────────────────────────┘
+```
+
+#### 技術的特徴
+
+| 項目 | 詳細 |
+| :--- | :--- |
+| **基盤モデル** | 事前訓練済み言語モデル(Gemini ベース) |
+| **強化学習** | AlphaZero アルゴリズムによるモンテカルロ木探索(MCTS) |
+| **形式言語** | Lean 4(証明の正しさを型チェッカーが100%検証) |
+| **訓練方法** | 数百万の問題で証明・反証を数週間にわたり実施。競技中も問題のバリエーションを自己生成して強化学習を継続 |
+| **自動形式化** | IMO の問題文は**人間が手動で Lean に翻訳**(2024年時点の制約) |
+
+#### IMO 2024 での成果
+
+- **6問中4問を解答**(代数2問、整数論1問 + AlphaGeometry 2 が幾何1問)
+- **得点: 42点満点中28点** → **銀メダル相当**
+- 特筆すべきは、人間の正答率が極めて低い**最難問の Problem 6(コンビナトリクス)を解答**したこと
+
+> [!NOTE]
+> AlphaProof の革新性は、LLMの「直感的な推測」をAlphaZeroの「体系的な探索」と組み合わせ、さらにLean 4 の型チェッカーが各ステップの正しさを保証するという**三層構造**にあります。これにより、ハルシネーションの余地がゼロの証明が生成されます。
+
+### 2.2 AlphaGeometry 2(Google DeepMind, 2024)
+
+幾何学に特化したニューロ・シンボリックシステムの進化版。
+
+#### アーキテクチャ(「速い思考」と「遅い思考」の融合)
+
+```
+┌─────────────────────────────────────────────┐
+│ AlphaGeometry 2 │
+│ │
+│ ┌────────────────────┐ ┌───────────────┐ │
+│ │ Gemini ベース LLM │ │ 記号的演繹 │ │
+│ │ (補助線の直感的提案) │ │ エンジン │ │
+│ │ 【System 1: 速い】 │ │ (ルール適用) │ │
+│ │ │ │ 【System 2】 │ │
+│ └────────┬───────────┘ └───────┬───────┘ │
+│ │ 相互作用 │ │
+│ └──────────┬───────────┘ │
+│ ▼ │
+│ 幾何学定理の証明 │
+└─────────────────────────────────────────────┘
+```
+
+- **訓練データ**: 初代の10倍以上の合成データで訓練
+- **記号エンジン**: 初代の100倍高速化
+- **拡張**: 移動変換(平行移動・回転等)、角度・比・距離に関する線形方程式に対応
+- **IMO 2024 幾何問題**: 形式化提供後**わずか19秒で解答**
+
+### 2.3 IMO 2025:金メダル時代の到来
+
+2025年のIMOでは、複数のAIシステムが**金メダル級**のパフォーマンスを達成し、形式証明AIの新時代が到来しました。
+
+| システム | 開発元 | 解答数 (6問中) | メダル相当 | 特徴 |
+| :--- | :--- | :---: | :---: | :--- |
+| **Aristotle** | Harmonic | 5問 | 🥇 金 | Lean 4 形式証明 + 幾何専用ソルバー |
+| **Seed-Prover** | ByteDance | 4-5問 | 🥇 金 (追加計算後) | エージェント型 Lean 4 証明器 + Python 実行 + 数学ライブラリ検索 |
+| **OpenAI モデル** | OpenAI | 5問 | 🥇 金 | 自然言語推論ベース(形式証明なし) |
+| **DeepMind モデル** | Google | 5問 | 🥇 金 | 自然言語推論ベース(形式証明なし) |
+
+> [!WARNING]
+> **形式証明 vs 自然言語推論の分岐**: IMO 2025 では、Aristotle と Seed-Prover は**機械検証可能な形式証明**を出力したのに対し、OpenAI と Google のシステムは自然言語による証明を出力し、人間の審査員が正しさを評価しました。この「検証可能性の差」は、AIの信頼性において根本的な違いをもたらします。
+
+### 2.4 Meta AI: HyperTree Proof Search (HTPS)
+
+Meta AI が2022年に発表した、AlphaZero にインスパイアされた定理証明探索アルゴリズムです。
+
+- **手法**: ハイパーグラフ上の木探索。選択(Selection)→ 展開(Expansion)→ 逆伝播(Back-propagation)の反復
+- **ニューラル構成**: Policy モデル(タクティクス提案)+ Critic モデル(証明進捗評価)
+- **オンライン学習**: 自身の成功した証明から学習し、訓練分布外の問題にも汎化
+- **環境**: Lean + Metamath で評価。miniF2F-curriculum、IMO問題での成果
+
+---
+
+## 3. AI証明探索ツール・フレームワーク
+
+### 3.1 LeanDojo
+
+AI駆動の Lean 4 定理証明のためのオープンソースフレームワーク。NeurIPS 2023 で発表。
+
+#### アーキテクチャ
+
+```mermaid
+graph TB
+ subgraph "LeanDojo エコシステム"
+ A["Lean 4 リポジトリ
(Mathlib 等)"] -->|トレースと抽出| B["データ抽出エンジン
(証明状態・タクティクス・前提)"]
+ B --> C["訓練データセット"]
+ C --> D["ReProver
(検索拡張型 LLM)"]
+ D -->|タクティクス提案| E["Lean 4 証明環境"]
+ E -->|フィードバック| D
+ end
+```
+
+#### 主要コンポーネント
+
+| コンポーネント | 機能 |
+| :--- | :--- |
+| **データ抽出エンジン** | Lean 4 リポジトリから証明状態、タクティクス、前提を自動抽出 |
+| **ReProver** | 検索拡張型生成(RAG)ベースの LLM。関連する前提(レンマ・定理)を Mathlib から検索し、タクティクスを提案 |
+| **証明環境** | Lean 4 とプログラム的に対話し、タクティクスの実行と検証を自動化 |
+
+#### LeanDojo-v2(2025年)
+
+2025年にリリースされた第2世代。以下の機能が大幅に強化されています。
+
+- **エンドツーエンドフレームワーク**: リポジトリトレース → データセット管理 → モデル訓練 → 評価を統一
+- **強化学習サポート**: GRPOTrainer による教師あり微調整と強化学習の統合
+- **生涯学習型データセット管理**: Mathlib の更新に追従するデータセット管理
+- **Lean4Code**: 数学者向けの軽量 IDE。AI 機能をフォーマライゼーションワークフローに直接統合
+
+### 3.2 Lean Copilot
+
+Lean 4 環境内で LLM を直接利用できるようにするフレームワーク。VS Code 上でリアルタイムに証明支援を提供します。
+
+#### 主要機能
+
+| 機能 | 説明 |
+| :--- | :--- |
+| **タクティクス提案** | 現在の証明状態に基づいて、コンテキストを考慮した次の一手を推薦 |
+| **前提検索** | Mathlib やユーザーのプロジェクトから関連するレンマ・定理を動的に検索 |
+| **証明探索** | 人間の入力とAIの自動探索を組み合わせた半自動証明 |
+| **モデル柔軟性** | ローカルモデル(オフライン動作)またはクラウド API 経由で任意の LLM を利用可能 |
+
+#### 性能
+
+- 従来のルールベース自動化(AESOP 等)を**大幅に上回る**証明成功率
+- 完全自動証明と人間アシスト証明の両方で優位性を実証
+
+### 3.3 主要な LLM ベース定理証明モデル比較
+
+以下は2024–2025年に登場した主要なLLM定理証明モデルの比較です。
+
+| モデル | 開発元 | 基盤モデル | 主な技術的革新 | 対象環境 |
+| :--- | :--- | :--- | :--- | :--- |
+| **DeepSeek-Prover** | DeepSeek | DeepSeek | 大規模合成データ生成 + Lean 4 自動フィルタリング | Lean 4 |
+| **DeepSeek-Prover-V2** | DeepSeek | DeepSeek-V3 | 再帰的定理証明パイプライン + サブゴール分解 | Lean 4 |
+| **Kimina-Prover-72B** | — | Qwen2.5 | 推論駆動探索、レンマ発見パターン、エラー修正 | Lean 4 |
+| **InternLM2.5-StepProver** | Shanghai AI Lab | InternLM2.5 | エキスパート反復 + Critic モデルによる証明探索ガイド | Lean 4 |
+| **LLEMMA** | EleutherAI / Princeton | Code Llama | Proof-Pile-2 での継続事前訓練。オープンソース数学基盤モデル | Lean 4 / Isabelle |
+| **Seed-Prover** | ByteDance | — | エージェント型アーキテクチャ + ツール統合(Python, ライブラリ検索) | Lean 4 |
+| **Aristotle** | Harmonic | — | 形式証明 + 非形式推論 + 幾何ソルバーの統合 | Lean 4 |
+| **HunyuanProver** | Tencent | — | System 2 思考 + ガイド付き木探索 + Critic モデル | Lean 4 |
+
+#### DeepSeek-Prover-V2:再帰的定理証明パイプラインの詳細
+
+```
+┌───────────────────────────────────────────────┐
+│ DeepSeek-Prover-V2 パイプライン │
+│ │
+│ 1. 複雑な定理 T を入力 │
+│ │ │
+│ ▼ │
+│ 2. 高レベル推論(非形式的分析) │
+│ └─ DeepSeek-V3 が T を複数のサブゴールに分解 │
+│ │ │
+│ ▼ │
+│ 3. 各サブゴール S₁, S₂, ... Sₙ を再帰的に証明 │
+│ └─ サブゴールがさらに複雑なら、再度分解 │
+│ │ │
+│ ▼ │
+│ 4. Lean 4 型チェッカーで各ステップを検証 │
+│ │ │
+│ ▼ │
+│ 5. 全サブゴール完了 → 定理 T の証明完了 │
+└───────────────────────────────────────────────┘
+```
+
+- MiniF2F、PutnamBench で**最先端の性能**を達成
+- 非形式的推論(自然言語での数学的直感)と形式的検証(Lean 4)の統合が鍵
+
+#### Kimina-Prover:推論駆動探索パラダイム
+
+従来の木探索アルゴリズム(BFS、MCTS)から脱却し、モデル自身の**内部推論能力**に依拠する新しいパラダイムを提案。
+
+- **レンマ発見パターン**: モデルが自律的に中間レンマを発見・適用
+- **エラー修正機能**: Lean のコンパイルエラーメッセージを解釈し、修正を提案
+- **多段階強化学習**: Qwen2.5 バックボーン上で段階的に RL を適用
+
+### 3.4 その他の関連ツール
+
+| ツール / システム | 対象 | 特徴 |
+| :--- | :--- | :--- |
+| **Tactician** | Coq | Coq 環境内での AI タクティクス自動提案 |
+| **Sledgehammer** | Isabelle/HOL | 外部 ATP(E, SPASS, Z3 等)を呼び出して自動証明 |
+| **Draft-Sketch-Prove** | 汎用 | LLM で非形式証明のスケッチを作成 → 形式証明器で完成させるプロンプト戦略 |
+
+---
+
+## 4. 自動形式化(Auto-formalization)
+
+### 4.1 概要:自然言語から形式言語への翻訳
+
+自動形式化とは、自然言語で書かれた数学的命題やその証明を、Lean 4 などの形式言語に自動翻訳する技術です。これは、AIによる数学の完全自動化において**最大のボトルネック**の一つです。
+
+```
+「任意の正の整数 n に対して、 ────▶ theorem example (n : ℕ) (hn : 0 < n) :
+ n² + n は偶数である」 Even (n^2 + n) := by
+ ...
+```
+
+### 4.2 主要な技術アプローチ(2024–2025)
+
+#### プロセス監視型・コンパイラフィードバック型
+
+```mermaid
+graph LR
+ A["自然言語の
数学命題"] -->|LLM 翻訳| B["Lean 4 コード
(初期生成)"]
+ B -->|コンパイル試行| C{"Lean 4
コンパイラ"}
+ C -->|成功| D["✅ 形式化完了"]
+ C -->|エラー| E["エラーメッセージ"]
+ E -->|フィードバック| A
+```
+
+- **PDA (Process-supervised Data Augmentation)**: Lean 4 コンパイラをリアルタイム検証器として使い、反復的に修正
+- **Mathesis-Autoformalizer**: アンサンブル LLM 検証 + 強化学習で高忠実度翻訳を実現
+
+#### 強化学習ベース
+
+- **GRPO (Group Relative Policy Optimization)**: 構文的正しさ(コンパイル可能か)と意味的正しさ(元の問題を正確に表現しているか)の両方を最適化
+- **サイクル一貫性報酬**: 形式 → 自然言語 に逆翻訳し、元の入力と一致するかで報酬を与える
+
+#### 大規模合成データ
+
+並列 NL-to-Lean データセットが希少であるため、合成データの生成が重要。
+
+- **Herald**: 階層的・検索ベースの翻訳済み Lean データセット
+- **FineLeanCorpus**: 大量のアラインメント済みデータ
+
+### 4.3 残された最大の課題:意味的ドリフト
+
+> [!CAUTION]
+> **意味的ドリフト(Semantic Drift)**: コードは構文的に正しく(コンパイルが通る)が、元の数学問題を**論理的に正確に表現していない**という問題。例えば、「すべての素数 p に対して〜」という問題を「すべての自然数 n に対して〜」と形式化してしまうケース。コンパイルは成功するが、証明された命題は元の問題と異なります。これは現在の自動形式化における最大の未解決課題です。
+
+---
+
+## 5. ベンチマークと評価
+
+### 5.1 形式定理証明ベンチマーク
+
+| ベンチマーク | 対象レベル | 形式言語 | 特徴 | 現在の状況 |
+| :--- | :--- | :--- | :--- | :--- |
+| **MiniF2F** | 高校数学競技 (AIME/IMO) | Lean 4 / Isabelle | 最も広く使われる標準ベンチマーク | 飽和に近づいている(miniF2F-v2 が登場) |
+| **PutnamBench** | 学部数学 (Putnam 競技) | Lean 4 | 解析・抽象代数など高度な分野を含む | 依然として非常に困難。解答率は低い |
+| **ProofNet** | 学部数学 | Lean 4 | 自動形式化の評価にも使用 | — |
+| **LeanDojo Benchmark 4** | 汎用 | Lean 4 | 「新規前提(Novel Premises)」でのテストにより汎化能力を評価 | — |
+| **FormL4** | 汎用 | Lean 4 | 命題と証明の両方の自動形式化を評価 | — |
+| **VeriSoftBench** | ソフトウェア検証 | Lean 4 | リポジトリ規模の形式検証を評価 | — |
+
+### 5.2 論理推論ベンチマーク(LLM 直接評価用)
+
+| ベンチマーク | 対象 | 特徴 |
+| :--- | :--- | :--- |
+| **FOLIO** | 一階述語論理推論 | 自然言語 → FOL 変換 + 論理的帰結判定 |
+| **ProofWriter** | 多段推論 | 異なる証明深度でのルールベース推論 |
+| **PrOntoQA** | 論理的含意 | ルールベース推論 |
+| **LogicalDeduction** | 複合論理 | 複雑な実世界の論理クエリ |
+| **AR-LSAT** | 法的推論 | LSAT 形式の高度な論理問題 |
+| **GSM8K / MATH** | 数学的推論 | 非形式的な数学問題解決(形式証明なし) |
+
+### 5.3 評価の新課題
+
+> [!IMPORTANT]
+> **「コンパイル成功 ≠ 正解」問題**: 最近の研究では、AIが生成した証明が「コンパイルに通る」(構文的に有効)にもかかわらず、意図された問題を**論理的に解決していない**ケースが報告されています。これに対応するため、**T2** や **FormalRewardBench** といった、より厳密な意味論的正しさを評価するフレームワークが開発されています。
+
+---
+
+## 6. 形式化エコシステム:Mathlib と数学コミュニティ
+
+### 6.1 Mathlib の現在(2025年5月時点)
+
+Lean 4 の最大の数学ライブラリであり、AIの訓練データとしても決定的に重要です。
+
+| 指標 | 数値 |
+| :---: | :---: |
+| 定理数 | **210,000+** |
+| 定義数 | **100,000+** |
+| 総コード行数 | **190万行+** |
+| コントリビューター数 | **500人+** |
+
+### 6.2 テレンス・タオ教授と形式数学
+
+フィールズ賞受賞者であるテレンス・タオ教授は、形式数学の最も著名な提唱者の一人です。
+
+- **自身の研究論文の形式化**: Lean 4 を使って自身のプレプリントの証明を形式化。微妙なエラーの発見、前提の厳密な記述、複雑な論証のリファクタリングなどの利点を実証
+- **「形式化ギャップ」の議論**: 現時点では形式化はLaTeX執筆よりはるかに時間がかかるが、新しいタクティクスとAIツールがこのギャップを急速に縮めつつあると指摘
+- **「生きた数学」の提唱**: フェルマーの最終定理の形式化プロジェクトなど、大規模な協調的数学プロジェクトにおいて、形式化が複雑性管理の重要なツールであると強調
+
+### 6.3 Mathlib Initiative(2025年後半~)
+
+Mathlib のスケールに伴い、以下の課題に対処するための組織化が進行中:
+
+- レビュープロセスのボトルネック解消
+- ドキュメンテーションの改善
+- 下流ライブラリとAI統合ワークフローのインフラ整備
+
+---
+
+## 7. 今後の展望と未解決課題
+
+### 7.1 技術的課題
+
+```mermaid
+graph TD
+ A["未解決課題"] --> B["自動形式化の意味的正確性"]
+ A --> C["計算コスト
(RL訓練・探索に莫大なGPU時間)"]
+ A --> D["形式化ギャップ
(自然言語数学との乖離)"]
+ A --> E["ドメイン汎化
(競技数学以外への拡張)"]
+ A --> F["長い証明への対応
(数百ステップの証明の生成)"]
+
+ B --> B1["意味的ドリフトの検出と防止"]
+ D --> D1["より自然な形式言語の設計"]
+ E --> E1["ソフトウェア検証
物理学
生物学への応用"]
+```
+
+### 7.2 短期的展望(2025–2027)
+
+| 領域 | 予測される進展 |
+| :--- | :--- |
+| **競技数学** | IMO 全問正解(満点金メダル)の達成。形式証明ベースのシステムが主流に |
+| **ツール統合** | IDE 内での AI 証明支援がデフォルト化(Lean Copilot の進化) |
+| **エージェント化** | 証明器がツール使用(Python 実行、ライブラリ検索、Web 検索)を自律的に行う |
+| **ソフトウェア検証** | 重要なオープンソースライブラリのコアモジュールを AI が形式検証する試み |
+
+### 7.3 長期的展望(2027–)
+
+- **数学研究の変革**: AI が新しい定理の予想(Conjecture)を生成し、自ら形式証明する「数学の自動発見」
+- **認知パートナー**: 人間の直感と形式的精密さを組み合わせた「認知パートナー」としてのAI
+- **AGI への道**: 形式論理は、AI の「正確な推論能力」を保証する唯一の手段として、AGI(汎用人工知能)アーキテクチャの不可欠な構成要素に
+- **教育**: AI 証明支援ツールが数学教育を根本的に変え、学生が「証明の構造」を対話的に学べるように
+
+---
+
+## 8. 主要文献・リソース一覧
+
+### 8.1 論文(時系列順)
+
+| 年 | タイトル | 著者 / 組織 | 会議 / 掲載 |
+| :--- | :--- | :--- | :--- |
+| 2022 | HyperTree Proof Search for Neural Theorem Proving | Lample et al. (Meta AI) | NeurIPS 2022 |
+| 2023 | LLEMMA: An Open Language Model for Mathematics | Azerbayev et al. (EleutherAI / Princeton) | ICLR 2024 |
+| 2023 | LeanDojo: Theorem Proving with Retrieval-Augmented Language Models | Yang et al. (Caltech / MIT) | NeurIPS 2023 |
+| 2024 | Lean Copilot: LLMs as Copilots for Theorem Proving in Lean | Song et al. | — |
+| 2024 | AlphaProof and AlphaGeometry 2 | Google DeepMind | Blog / Technical Report |
+| 2024 | AlphaGeometry: An Olympiad-level AI system for geometry | Trinh et al. (DeepMind) | Nature 2024 |
+| 2024 | DeepSeek-Prover: Advancing Theorem Proving via Large-scale Synthetic Data | Xin et al. (DeepSeek) | — |
+| 2024 | Logic-LM: Empowering LLMs with Symbolic Solvers for Faithful Logical Reasoning | Pan et al. | ACL 2024 |
+| 2024 | LINC: A Neurosymbolic Approach for Logical Reasoning | Olausson et al. | EMNLP 2023 |
+| 2024 | InternLM2.5-StepProver | Shanghai AI Lab | — |
+| 2025 | DeepSeek-Prover-V2: Recursive Theorem Proving | DeepSeek | — |
+| 2025 | Kimina-Prover-72B | — | — |
+| 2025 | LeanDojo-v2 | Yang et al. | — |
+| 2025 | Seed-Prover | ByteDance | — |
+| 2025 | Aristotle | Harmonic | — |
+| 2025 | IJCAI Survey: LLM Reasoning with NeSy | Various | IJCAI 2025 |
+
+### 8.2 サーベイ論文
+
+- **"A Survey on Large Language Models with Neuro-Symbolic Reasoning"** — IJCAI 2025。LLMとニューロ・シンボリック推論の統合を体系的に分類
+- **"LLM-Symbolic-Reasoning-Survey"** — GitHub リポジトリ。シンボリック形式化、定理証明、ニューロ・シンボリック統合の研究をトラッキング
+- **"Awesome-LLM-Reasoning-with-NeSy"** — IJCAI 2025 サーベイに関連するキュレーションリスト
+
+### 8.3 実践リソース
+
+| リソース | URL / 場所 | 用途 |
+| :--- | :--- | :--- |
+| **Lean 4** | [leanprover.github.io](https://leanprover.github.io/) | 証明言語本体 |
+| **Mathlib** | [github.com/leanprover-community/mathlib4](https://github.com/leanprover-community/mathlib4) | Lean 4 数学ライブラリ |
+| **LeanDojo** | [leandojo.org](https://leandojo.org/) | AI定理証明フレームワーク |
+| **Lean Copilot** | GitHub | VS Code 上の AI 証明支援 |
+| **Natural Number Game** | [adam.math.hhu.de](https://adam.math.hhu.de/) | Lean 4 を遊びながら学ぶゲーム |
+| **Z3 Solver** | [github.com/Z3Prover/z3](https://github.com/Z3Prover/z3) | SMT ソルバー |
+| **MiniF2F** | GitHub | 形式定理証明ベンチマーク |
+| **PutnamBench** | GitHub | 学部数学レベルのベンチマーク |
+
+---
+
+> [!TIP]
+> **この分野に入門するための推奨パス**:
+> 1. [Natural Number Game](https://adam.math.hhu.de/) で Lean 4 の基礎を体験(所要: 数時間)
+> 2. LeanDojo のチュートリアルで AI 証明探索を体験
+> 3. Z3 (Python) で制約充足問題を解く経験を積む
+> 4. DeepSeek-Prover や Lean Copilot を使って実際の数学の形式化を試みる
diff --git a/docs/AI_Agents_and_Logic/ai_formal_logic_survey.rag_ingested b/docs/AI_Agents_and_Logic/ai_formal_logic_survey.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..3b75c8ad2849ce7d3b4390ada562707b970aeafb
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/ai_formal_logic_survey.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "5bc408de7c4c1e24", "ingested_at": "2026-06-30T08:25:49.660344", "content_hash": "5bc408de7c4c1e24"}
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/ai_proof_search_tools_survey.md b/docs/AI_Agents_and_Logic/ai_proof_search_tools_survey.md
new file mode 100644
index 0000000000000000000000000000000000000000..f91de98577fb61b6dd7d07f9524acee91123c29b
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/ai_proof_search_tools_survey.md
@@ -0,0 +1,670 @@
+# AI証明探索ツール&モデル — 包括的サーベイ
+
+> **調査時点**: 2026年5月
+> **主題**: LeanDojo, Lean Copilot, 及び主要AIプルーバーの技術的比較
+> **範囲**: 開発基盤(ツール)からモデル、ベンチマーク、訓練手法まで
+
+---
+
+## 目次
+
+1. [全体像:AI証明探索のエコシステム](#1-全体像ai証明探索のエコシステム)
+2. [基盤インフラ:Lean 4 と Mathlib4](#2-基盤インフラlean-4-と-mathlib4)
+3. [LeanDojo:v1 → v2 の進化](#3-leandojov1--v2-の進化)
+4. [Lean Copilot:対話型AI証明アシスタント](#4-lean-copilot対話型ai証明アシスタント)
+5. [主要AIプルーバー・モデルの技術比較](#5-主要aiプルーバーモデルの技術比較)
+6. [証明生成戦略の分類](#6-証明生成戦略の分類)
+7. [ベンチマークと評価](#7-ベンチマークと評価)
+8. [訓練手法の比較](#8-訓練手法の比較)
+9. [未解決課題と今後の展望](#9-未解決課題と今後の展望)
+10. [文献一覧](#10-文献一覧)
+
+---
+
+## 1. 全体像:AI証明探索のエコシステム
+
+AI証明探索は、**3層のスタック**として整理できます。
+
+```mermaid
+graph TB
+ subgraph "Layer 3: AI プルーバー・モデル"
+ M1["DeepSeek-Prover-V2"]
+ M2["Kimina-Prover-72B"]
+ M3["InternLM2.5-StepProver"]
+ M4["Goedel-Prover-V2"]
+ M5["LLEMMA"]
+ M6["Seed-Prover"]
+ end
+
+ subgraph "Layer 2: 開発フレームワーク・ツール"
+ T1["LeanDojo v2"]
+ T2["Lean Copilot"]
+ T3["LeanAgent"]
+ end
+
+ subgraph "Layer 1: 基盤インフラ"
+ L1["Lean 4 (証明アシスタント)"]
+ L2["Mathlib4 (数学ライブラリ)"]
+ end
+
+ M1 & M2 & M3 & M4 & M5 & M6 --> T1 & T2 & T3
+ T1 & T2 & T3 --> L1 & L2
+
+ style L1 fill:#2D3436,stroke:#333,color:#fff
+ style L2 fill:#2D3436,stroke:#333,color:#fff
+ style T1 fill:#0984E3,stroke:#333,color:#fff
+ style T2 fill:#0984E3,stroke:#333,color:#fff
+ style T3 fill:#0984E3,stroke:#333,color:#fff
+```
+
+| レイヤー | 役割 | 具体例 |
+| :--- | :--- | :--- |
+| **Layer 1: 基盤** | 形式言語・検証カーネル・数学ライブラリ | Lean 4, Mathlib4 |
+| **Layer 2: ツール** | データ抽出・インタラクション・IDE統合 | LeanDojo, Lean Copilot, LeanAgent |
+| **Layer 3: モデル** | タクティクス予測・証明探索・推論 | DeepSeek-Prover-V2, Kimina-Prover 等 |
+
+---
+
+## 2. 基盤インフラ:Lean 4 と Mathlib4
+
+### 2.1 Lean 4 — 証明アシスタント
+
+```
+ Lean 4 の二重の役割
+ ┌──────────────────────────────────────────────┐
+ │ │
+ │ 1. プログラミング言語としての Lean │
+ │ ・依存型理論(Calculus of Inductive │
+ │ Constructions)に基づく │
+ │ ・汎用関数型プログラミング言語としても使用可能 │
+ │ │
+ │ 2. 証明アシスタントとしての Lean │
+ │ ・タクティクスモードでインタラクティブ証明 │
+ │ ・カーネル(型チェッカー)は小さく検証済み │
+ │ ・Lean 自身で Lean のメタプログラミングが可能 │
+ │ │
+ │ AIにとっての重要性: │
+ │ → 型チェッカーが完全に決定論的な「報酬信号」 │
+ │ → タクティクス = AIが選択する「行動空間」 │
+ │ → 証明状態 = AIが観測する「環境状態」 │
+ └──────────────────────────────────────────────┘
+```
+
+### 2.2 Mathlib4 — 最大の形式化数学ライブラリ
+
+| 指標 | 数値(2026年5月時点) |
+| :--- | :--- |
+| **定理数** | 約 274,000 |
+| **定義数** | 約 131,000 |
+| **コード行数** | 210万行以上 |
+| **貢献者数** | 770名以上 |
+
+> [!IMPORTANT]
+> **Mathlib4 は AI 証明探索の「訓練データ」であり「ツールボックス」でもあります。**
+> - **訓練データ**: 数十万の定理-証明ペアは、AIプルーバーの教師あり微調整(SFT)の基礎データ
+> - **ツールボックス**: 推論時に AI がライブラリ内の既存定理を検索・引用して証明を構成
+> - **ベンチマーク**: Mathlib の難しい定理は、AI能力の測定に使用
+
+---
+
+## 3. LeanDojo:v1 → v2 の進化
+
+### 3.1 LeanDojo v1(NeurIPS 2023 Oral)
+
+> **論文**: *"LeanDojo: Theorem Proving with Retrieval-Augmented Language Models"*
+> **著者**: Kaiyu Yang, Aidan M. Swope, Alex Gu, Rahul Chalamala, Peiyang Song, Shixing Yu, Saad Godil, Ryan Prenger, Anima Anandkumar
+> **会議**: NeurIPS 2023 (Datasets and Benchmarks Track, **Oral**)
+
+```
+ LeanDojo v1 のアーキテクチャ
+ ┌─────────────────────────────────────────────────┐
+ │ │
+ │ ┌───────────────────┐ ┌───────────────────┐ │
+ │ │ リポジトリ・トレーサー │ │ プログラム的 │ │
+ │ │ │ │ インタラクション │ │
+ │ │ ・AST 抽出 │ │ │ │
+ │ │ ・依存関係 DAG │ │ ・Lean 環境操作 │ │
+ │ │ ・前提の定義位置 │ │ ・タクティクス送信 │ │
+ │ │ ・証明状態キャプチャ │ │ ・証明状態取得 │ │
+ │ └───────┬───────────┘ └────────┬──────────┘ │
+ │ │ │ │
+ │ ▼ ▼ │
+ │ ┌─────────────────────────────────────────┐ │
+ │ │ LeanDojo ベンチマーク │ │
+ │ │ (Lean 3/4 の train/val/test 分割) │ │
+ │ └─────────────────┬───────────────────────┘ │
+ │ │ │
+ │ ▼ │
+ │ ┌─────────────────────────────────────────┐ │
+ │ │ ReProver │ │
+ │ │ (検索拡張型定理証明器) │ │
+ │ │ │ │
+ │ │ Retriever → Encoder-Decoder LM │ │
+ │ │ (前提検索) (タクティクス生成) │ │
+ │ └──────────────────────────────────────────┘ │
+ └──────────────────────────────────────────────────┘
+```
+
+#### ReProver(Retrieval-Augmented Prover)の動作
+
+```
+ 証明目標: ∀ n : ℕ, 0 + n = n
+ │
+ ▼
+ ┌────────────────────────┐
+ │ 1. 前提検索 (Retriever) │
+ │ 「この目標に関連する │
+ │ Mathlib の定理は?」 │
+ │ → Nat.zero_add を発見 │
+ └──────────┬─────────────┘
+ │ 検索結果を連結
+ ▼
+ ┌────────────────────────┐
+ │ 2. タクティクス生成 │
+ │ (Encoder-Decoder) │
+ │ → 「exact Nat.zero_add n」│
+ └──────────┬─────────────┘
+ │
+ ▼
+ ┌────────────────────────┐
+ │ 3. Lean で検証 │
+ │ → ✅ 型チェック成功 │
+ └────────────────────────┘
+```
+
+### 3.2 LeanDojo v2(2025年)
+
+| 特性 | LeanDojo v1 | LeanDojo v2 |
+| :--- | :--- | :--- |
+| **スコープ** | データ抽出 + 基本インタラクション | **統合的 E2E フレームワーク** |
+| **データ管理** | 静的ベンチマーク | **ライフロング・データセット管理**(新定理を再トレースなしに追加可能) |
+| **トレーナー** | — | **SFT, GRPO, Retrieval** トレーナーを内蔵 |
+| **プルーバー** | ReProver のみ | **HF Prover, External Prover, LeanAgent** をサポート |
+| **IDE** | なし | **Lean4Code**(軽量IDE)を同梱 |
+| **モデル対応** | 固定 | **マルチモデル対応**(HuggingFace, 外部API) |
+| **API 設計** | モノリシック | **モジュラーAPI**(トレース・訓練・評価を分離) |
+| **ライセンス** | MIT | MIT |
+
+### 3.3 LeanAgent(LeanDojo エコシステム内)
+
+LeanDojo v2 と統合されたエージェント型プルーバー。**生涯学習(Lifelong Learning)**を実現。
+
+```
+ LeanAgent のライフロング学習サイクル
+ ┌──────────────────────────────────────────┐
+ │ │
+ │ リポジトリ A リポジトリ B リポジトリ C │
+ │ (代数学) (位相幾何) (複素解析) │
+ │ │ │ │ │
+ │ └──────────────┼──────────────┘ │
+ │ │ │
+ │ ▼ │
+ │ ┌──────────────────┐ │
+ │ │ カリキュラム学習 │ │
+ │ │ (難易度順にソート) │ │
+ │ └────────┬─────────┘ │
+ │ │ │
+ │ ▼ │
+ │ ┌──────────────────┐ │
+ │ │ 漸進的訓練 │ │
+ │ │ (安定性-可塑性 │ │
+ │ │ バランス) │ │
+ │ └────────┬─────────┘ │
+ │ │ │
+ │ ▼ │
+ │ ┌──────────────────┐ │
+ │ │ 動的知識ベース │ │
+ │ │ 管理 │ │
+ │ └──────────────────┘ │
+ │ │
+ │ 成果: 23リポジトリで160以上の定理を証明 │
+ │ (人間未証明の定理を含む) │
+ └──────────────────────────────────────────┘
+```
+
+---
+
+## 4. Lean Copilot:対話型AI証明アシスタント
+
+### 4.1 アーキテクチャ
+
+```
+ ┌─────────────────────────────────────────────────────┐
+ │ Lean Copilot │
+ │ │
+ │ VS Code (Lean エディタ) │
+ │ ┌───────────────────────────────────┐ │
+ │ │ ユーザーが書いている Lean 証明 │ │
+ │ │ │ │
+ │ │ theorem foo : P → Q := by │ │
+ │ │ ▌ ← カーソル位置 │ │
+ │ └───────────────┬───────────────────┘ │
+ │ │ 現在の証明状態を抽出 │
+ │ ▼ │
+ │ ┌─────────────────────────────────────────┐ │
+ │ │ 3つの自動化機能 │ │
+ │ │ │ │
+ │ │ ┌─────────────┐ ┌──────────────────┐ │ │
+ │ │ │ suggest_tactics│ │ search_proofs │ │ │
+ │ │ │ (タクティクス │ │ (証明探索) │ │ │
+ │ │ │ 候補を提示) │ │ │ │ │
+ │ │ └──────────────┘ └──────────────────┘ │ │
+ │ │ ┌──────────────┐ │ │
+ │ │ │ select_premises│ │ │
+ │ │ │ (前提選択) │ │ │
+ │ │ └──────────────┘ │ │
+ │ └───────────────┬─────────────────────────┘ │
+ │ │ │
+ │ ▼ │
+ │ ┌──────────────────────────────────────────┐ │
+ │ │ 推論バックエンド │ │
+ │ │ │ │
+ │ │ ローカル クラウド │ │
+ │ │ ┌──────────┐ ┌──────────┐ │ │
+ │ │ │CTranslate2│ │ API │ │ │
+ │ │ │(C++ FFI) │ │(OpenAI等)│ │ │
+ │ │ └──────────┘ └──────────┘ │ │
+ │ └──────────────────────────────────────────┘ │
+ └─────────────────────────────────────────────────────┘
+```
+
+### 4.2 LeanDojo vs Lean Copilot の機能比較
+
+| 特性 | LeanDojo v2 | Lean Copilot |
+| :--- | :--- | :--- |
+| **主な用途** | 研究者向け:AI プルーバーの訓練・評価パイプライン | 実務者向け:証明作成のリアルタイム支援 |
+| **ターゲットユーザー** | ML 研究者、AI for Math 研究者 | 数学者、形式検証エンジニア |
+| **データ抽出** | ✅ リポジトリトレース、AST、証明状態 | ❌ なし |
+| **訓練機能** | ✅ SFT, GRPO, Retrieval トレーナー | ❌ なし(推論専用) |
+| **IDE 統合** | Lean4Code(軽量) | ✅ VS Code 完全統合 |
+| **タクティクス提案** | ❌(バッチ評価向け) | ✅ `suggest_tactics` |
+| **証明探索** | ✅ BFS, MCTS 等 | ✅ `search_proofs`(aesop 統合) |
+| **前提選択** | ✅ 検索拡張型 | ✅ `select_premises` |
+| **モデルデプロイ** | バッチ処理向け | ローカル(FFI)/ クラウド(API) |
+| **レイテンシ要件** | 非リアルタイム | ⚡ リアルタイム(ユーザー体験重視) |
+| **オープンソース** | ✅ MIT | ✅ |
+
+> [!TIP]
+> **使い分けガイド**
+> - **「新しいAIプルーバーを訓練・ベンチマークしたい」** → LeanDojo v2
+> - **「日常の証明作成でAI支援がほしい」** → Lean Copilot
+> - **「複数リポジトリにまたがって継続学習するエージェントを作りたい」** → LeanDojo v2 + LeanAgent
+
+---
+
+## 5. 主要AIプルーバー・モデルの技術比較
+
+### 5.1 総合比較表
+
+| 特性 | DeepSeek-Prover-V2 | Kimina-Prover-72B | InternLM2.5-StepProver | Goedel-Prover-V2 | LLEMMA | Seed-Prover 1.5 |
+| :--- | :--- | :--- | :--- | :--- | :--- | :--- |
+| **開発元** | DeepSeek | Numina + Kimi | SHLab (上海AI Lab) | Princeton | EleutherAI + Princeton | ByteDance |
+| **ベースモデル** | DeepSeek-V3 | Qwen2.5-72B | InternLM2.5 | Qwen2.5 系 | Code Llama | 独自LLM |
+| **パラメータ数** | **671B** (37B active) / 7B | **72B** | 7B | 8B / **32B** | 7B / 34B | 非公開 |
+| **アーキテクチャ** | MoE | Dense Transformer | Dense Transformer | Dense Transformer | Dense Transformer | Dense Transformer |
+| **形式環境** | Lean 4 | Lean 4 | Lean 4 | Lean 4 | Lean 4 / Isabelle | Lean 4 |
+| **証明戦略** | 再帰的サブゴール分解 | 推論駆動探索 | Critic指導型探索 | スキャフォールド合成 | 基盤モデル(Few-shot) | エージェント型レンマ駆動 |
+| **オープンソース** | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ |
+| **公開年** | 2025 | 2025 | 2024 | 2025 | 2023 | 2025 |
+
+### 5.2 各モデルの技術詳細
+
+---
+
+#### DeepSeek-Prover-V2 — 再帰的サブゴール分解
+
+```mermaid
+graph TB
+ A["定理 T"] --> B["DeepSeek-V3 (671B)
非形式的証明スケッチ + Lean スケルトン"]
+ B --> C["Lean スケルトン
(sorry プレースホルダー付き)"]
+ C --> D["サブゴール S₁"]
+ C --> E["サブゴール S₂"]
+ C --> F["サブゴール S₃"]
+
+ D --> G["7B モデルが S₁ を証明"]
+ E --> H["7B モデルが S₂ を証明"]
+ F --> I["7B モデルが S₃ を証明
(再帰的にさらに分解も可能)"]
+
+ G & H & I --> J["統合: sorry を証明済みコードで置換"]
+ J --> K["Lean 4 コンパイラで全体検証"]
+ K --> L["✅ 完全な形式証明"]
+
+ style B fill:#6C5CE7,stroke:#333,color:#fff
+ style G fill:#00B894,stroke:#333,color:#fff
+ style H fill:#00B894,stroke:#333,color:#fff
+ style I fill:#00B894,stroke:#333,color:#fff
+```
+
+| 技術要素 | 詳細 |
+| :--- | :--- |
+| **コア思想** | 大モデルが「戦略」、小モデルが「戦術」を担当する分業型 |
+| **MoE** | Mixture-of-Experts。671B 中 37B のみ活性化 → 推論効率が高い |
+| **コンテキスト長** | 最大 163K トークン(長い証明チェーンに対応) |
+| **MLA** | Multi-head Latent Attention。長距離依存関係の処理に優れる |
+| **訓練** | Cold-start データ合成 → GRPO(強化学習) |
+| **デュアルモード** | Non-CoT(高速)/ CoT(高精度)の切り替え |
+| **一貫性報酬** | CoT の推論と形式コードの構造的一致を報酬として使用 |
+
+---
+
+#### Kimina-Prover-72B — 推論駆動探索
+
+```
+ 従来のアプローチ vs Kimina のアプローチ
+
+ 従来: LLM + 外部探索アルゴリズム
+ ┌──────────┐ ┌──────────┐
+ │ LLM │ → │ BFS / │ → 証明
+ │(タクティ │ │ MCTS │
+ │ クス生成)│ │(外部探索)│
+ └──────────┘ └──────────┘
+
+ Kimina: LLM が「探索も推論も」内部で実行
+ ┌──────────────────────────────────┐
+ │ Kimina-Prover-72B │
+ │ │
+ │ │
+ │ 非形式的推論... │
+ │ レンマの発見... │
+ │ エラー解釈 → 修正... │
+ │ │
+ │ │
+ │ exact Nat.add_comm a b │
+ │ │
+ └──────────────────────────────────┘
+```
+
+| 技術要素 | 詳細 |
+| :--- | :--- |
+| **コア思想** | BFS/MCTS 等の外部探索を廃し、モデル内部の推論能力で探索を代替 |
+| **Formal Reasoning Pattern** | `...` 内で非形式推論、`...` で形式コードを交互に生成 |
+| **レンマパターン** | RL 中にランダムなレンマを挿入し、モデルがレンマの発見・再利用を学習 |
+| **エラー修正** | Lean のコンパイルエラーメッセージを解釈し、ゼロから再生成せずに修正 |
+| **多段階 RL** | SFT → RL(Kimi k1.5 ベース)→ 反復的精緻化 |
+| **MiniF2F** | **92.2%** pass rate(リリース時 SOTA) |
+
+---
+
+#### InternLM2.5-StepProver — Critic 指導型探索
+
+```
+ ┌───────────────────────────────────────────────┐
+ │ InternLM2.5-StepProver │
+ │ │
+ │ ┌─────────────────┐ ┌─────────────────┐ │
+ │ │ Prover Model │ │ Critic Model │ │
+ │ │ (タクティクス │ │ (軌跡の選好 │ │
+ │ │ 生成) │ │ 評価) │ │
+ │ └────────┬────────┘ └────────┬────────┘ │
+ │ │ │ │
+ │ │ ┌────────────────┘ │
+ │ ▼ ▼ │
+ │ ┌─────────────────────────────┐ │
+ │ │ Critic 指導型 Best-First │ │
+ │ │ Search │ │
+ │ │ │ │
+ │ │ ・Prover が候補タクティクス生成 │ │
+ │ │ ・Critic が各軌跡の「有望さ」 │ │
+ │ │ を評価(選好情報を活用) │ │
+ │ │ ・有望な経路を優先的に探索 │ │
+ │ └──────────────────────────────┘ │
+ │ │
+ │ Expert Iteration (20,000 CPU-days): │
+ │ 証明成功 → データ化 → モデル更新 → 反復 │
+ └───────────────────────────────────────────────┘
+```
+
+| 技術要素 | 詳細 |
+| :--- | :--- |
+| **コア思想** | タクティクス軌跡の「選好情報」を学習した Critic モデルで探索を効率化 |
+| **Expert Iteration** | 20,000 CPU日の大規模計算。証明成功 → データ → モデル改善 のサイクル |
+| **Critic の役割** | 通常の BFS は各タクティクスの「累積ログ確率」で優先度を決定するが、Critic は軌跡全体の「証明到達可能性」を直接評価 |
+| **深い証明の発見** | 標準 BFS が見逃す「初期ステップは低確率だが最終的に成功する」経路を発見 |
+| **Log-Linear スケーリング** | 計算量と解決問題数の間にlog-linear 関係を発見 → スケーリング則 |
+
+---
+
+#### Goedel-Prover-V2 — スキャフォールド合成 + 自己修正
+
+```
+ ┌───────────────────────────────────────────┐
+ │ Goedel-Prover-V2 │
+ │ │
+ │ Innovation 1: スキャフォールド合成 │
+ │ ┌─────────────────────────────────┐ │
+ │ │ 失敗した証明試行から新しい │ │
+ │ │ サブ問題を構成 → 難易度を │ │
+ │ │ 段階的に上げた訓練タスクを合成 │ │
+ │ └─────────────────────────────────┘ │
+ │ │
+ │ Innovation 2: 検証器ガイド自己修正 │
+ │ ┌─────────────────────────────────┐ │
+ │ │ 証明生成 → Lean コンパイル │ │
+ │ │ → エラー → エラーメッセージで修正 │ │
+ │ │ → 再コンパイル → (反復) │ │
+ │ └─────────────────────────────────┘ │
+ │ │
+ │ Innovation 3: モデル平均化 │
+ │ ┌─────────────────────────────────┐ │
+ │ │ 複数チェックポイントの重みを平均 │ │
+ │ │ → 出力の多様性を維持 │ │
+ │ │ → 深い微調整による性能劣化を防止 │ │
+ │ └─────────────────────────────────┘ │
+ │ │
+ │ PutnamBench: 86問 (32B, pass@184) │
+ │ → DeepSeek-Prover-V2-671B (47問) を大幅凌駕│
+ └───────────────────────────────────────────┘
+```
+
+---
+
+#### LLEMMA — オープンソース数学基盤モデル
+
+| 技術要素 | 詳細 |
+| :--- | :--- |
+| **位置づけ** | 特化型プルーバーではなく、数学推論のための**汎用基盤モデル** |
+| **ベース** | Code Llama (Llama 2 系) |
+| **訓練データ** | Proof-Pile-2(55B トークン): arXiv (29B) + OpenWebMath (15B) + AlgebraicStack (11B) |
+| **特徴** | 追加微調整なしで few-shot タクティクス予測が可能。ツール使用(Python, CAS)に対応 |
+| **意義** | 下流の特化モデル(ReProver 等)のための**事前訓練済みベース**を提供 |
+
+---
+
+### 5.3 探索戦略の比較表
+
+| モデル | 探索戦略 | 長所 | 短所 |
+| :--- | :--- | :--- | :--- |
+| **DeepSeek-Prover-V2** | 再帰的分解 + sorry 埋め | 大問題を管理可能な単位に分割。大小モデルの分業 | 分解の質が大モデルの非形式推論に依存 |
+| **Kimina-Prover-72B** | 推論駆動(探索を内部化) | 外部探索オーバーヘッドなし。72Bでのスケーリング効果 | 計算量がモデルサイズに依存 |
+| **InternLM2.5-StepProver** | Critic 指導型 BFS | 深い証明の発見。選好学習で効率化 | Critic 訓練に大規模計算が必要 |
+| **Goedel-Prover-V2** | スキャフォールド + 自己修正 | 小型モデル(8B/32B)でも高性能。モデル平均化で安定 | 検証器フィードバックループの速度 |
+| **Seed-Prover 1.5** | エージェント型(ツール統合) | Python実行・Mathlib検索。最も柔軟 | システム全体の複雑性が高い |
+
+---
+
+## 6. 証明生成戦略の分類
+
+AI証明生成の主要な戦略を体系的に分類します。
+
+### 6.1 戦略マップ
+
+```mermaid
+graph TD
+ S["証明生成戦略"] --> S1["全証明一括生成
(Whole-Proof)"]
+ S --> S2["ステップバイステップ
(Tactic-by-Tactic)"]
+ S --> S3["階層的分解
(Hierarchical)"]
+ S --> S4["ライブラリ学習
(Library Learning)"]
+
+ S1 --> S1a["Baldur"]
+ S2 --> S2a["ReProver"]
+ S2 --> S2b["InternLM2.5-StepProver"]
+ S3 --> S3a["Draft-Sketch-Prove"]
+ S3 --> S3b["DeepSeek-Prover-V2"]
+ S3 --> S3c["Seed-Prover"]
+ S4 --> S4a["LEGO-Prover"]
+
+ style S1 fill:#E17055,stroke:#333,color:#fff
+ style S2 fill:#FDCB6E,stroke:#333
+ style S3 fill:#00B894,stroke:#333,color:#fff
+ style S4 fill:#0984E3,stroke:#333,color:#fff
+```
+
+### 6.2 戦略比較表
+
+| 戦略 | 代表的システム | 動作原理 | 固有の強み | 固有の弱点 |
+| :--- | :--- | :--- | :--- | :--- |
+| **全証明一括生成** | Baldur | LLMが証明全体を一度に生成。エラー時は修復 | 探索オーバーヘッドなし。高速 | 長い証明で精度低下。構造的洞察が限定的 |
+| **Draft-Sketch-Prove** | DSP | 非形式草案 → 形式スケルトン → Hammer で補完 | 人間的推論と形式的精密さの融合 | パイプライン段階間の翻訳エラー |
+| **ステップバイステップ** | ReProver, StepProver | 各タクティクスを逐次生成 + 探索(BFS/MCTS) | 最も柔軟。中間状態で軌道修正可能 | 長い証明で探索空間が爆発 |
+| **再帰的分解** | DeepSeek-Prover-V2 | sorry 付きスケルトン → サブゴールを再帰的に解決 | 大問題への自然な対処。並列化可能 | 分解の質が全体性能を決定 |
+| **レンマ駆動型** | Seed-Prover, Kimina | 自然言語推論 → レンマの発見 → 形式化 | 人間的証明スタイル。再利用可能なレンマ蓄積 | レンマの正確な形式化が困難 |
+| **ライブラリ学習** | LEGO-Prover | 検証済みレンマの「スキルライブラリ」を漸進的に構築 | 知識の蓄積・再利用。複雑問題への対応力 | 計算予算を揃えると効果が減少する報告あり |
+
+---
+
+## 7. ベンチマークと評価
+
+### 7.1 主要ベンチマーク概要
+
+| ベンチマーク | 難易度 | 問題数 | 出題源 | 形式環境 | 測定内容 |
+| :--- | :--- | :--- | :--- | :--- | :--- |
+| **MiniF2F** | 🟡 高校〜オリンピック | 488 (244 test) | AMC, AIME, IMO | Lean 4 / Isabelle | Olympiad レベルの形式証明能力 |
+| **ProofNet** | 🟡 学部レベル | 371 | 大学数学教科書 | Lean 3/4 | 自動形式化 + 証明 |
+| **PutnamBench** | 🔴 大学上級 | 672 | Putnam Competition (1962–2025) | Lean 4 | 高難度の形式証明 |
+| **Lean-Workbook-Plus** | 🟡 多様 | 48,809 | 数学フォーラム・教科書 | Lean 4 | 大規模な証明能力 |
+
+### 7.2 MiniF2F-test 性能比較(2025年時点)
+
+| モデル | パラメータ | Pass Rate | 備考 |
+| :---: | :---: | :---: | :--- |
+| **Kimina-Prover-72B** | 72B | **~92.2%** | 推論駆動探索。リリース時 SOTA |
+| **Goedel-Prover-V2-32B** | 32B | ~90.4% | 自己修正モード。pass@32 |
+| **Goedel-Prover-V2-8B** | 8B | ~84.6% | pass@32。小型ながら高性能 |
+| **DeepSeek-Prover-V2-671B** | 671B (37B active) | ~88.9% | MoE。非形式+形式融合 |
+| **InternLM2.5-StepProver** | 7B | ~65.9% | Critic 指導型。7B としては高水準 |
+| **ReProver (LeanDojo)** | ~1B | ~26.5% | 検索拡張型。2023年の基準 |
+
+> [!WARNING]
+> **ベンチマーク飽和への注意**: MiniF2F は 90% を超えるモデルが複数登場しており、飽和が指摘されています。研究コミュニティは **PutnamBench** や **VeriBench-FTP** 等のより困難なベンチマークへの移行を進めています。
+
+### 7.3 PutnamBench 性能比較
+
+| モデル / システム | 解答数 | Pass 条件 | 備考 |
+| :--- | :---: | :--- | :--- |
+| **Aleph + GPT-5.2** (Logical Intelligence) | **668/672** | pass@? | エージェント型。2026年初SOTA |
+| **Goedel-Prover-V2-32B** | 86 | pass@184 | オープンソース1位(2025年中盤) |
+| **DeepSeek-Prover-V2-671B** | 47 | — | MoE 671B |
+
+---
+
+## 8. 訓練手法の比較
+
+### 8.1 訓練パイプラインの分類
+
+```mermaid
+graph LR
+ subgraph "Phase 1: 事前訓練"
+ P1["汎用LLM事前訓練
(数学・コードデータ)"]
+ end
+ subgraph "Phase 2: 教師あり微調整 (SFT)"
+ P2["Mathlib 証明データ
+ 合成データで微調整"]
+ end
+ subgraph "Phase 3: 強化学習 (RL)"
+ P3["GRPO / Expert Iteration
Lean 検証器が報酬"]
+ end
+ subgraph "Phase 4: テスト時適応"
+ P4["TTRL / TTT
問題固有の追加学習"]
+ end
+
+ P1 --> P2 --> P3 --> P4
+```
+
+### 8.2 強化学習手法の比較
+
+| 手法 | 使用モデル | 動作原理 | 報酬信号 | 特徴 |
+| :--- | :--- | :--- | :--- | :--- |
+| **GRPO** | DeepSeek-Prover-V2, Goedel-Prover | グループ内の相対的性能で方策を最適化。Value モデル不要 | Lean コンパイル成功/失敗 | メモリ効率が高い。大規模モデル向き |
+| **Expert Iteration** | InternLM2.5-StepProver | 証明成功 → データ化 → モデル更新 → 再探索 を反復 | 証明発見の成否 | 大規模計算(20,000 CPU日)が必要 |
+| **Kimi k1.5 RL** | Kimina-Prover-72B | 多段階 RL。候補証明のロールアウトを Lean で検証 | Lean バイナリ報酬 | 外部探索なしで推論内部でRL |
+| **TTRL / TTT** | AlphaProof, Aristotle | テスト時に問題固有の追加 RL/勾配更新 | 問題バリエーションの証明成否 | 計算コスト極大だが最高性能 |
+
+### 8.3 訓練データソースの比較
+
+| データソース | 種類 | 規模 | 使用モデル |
+| :--- | :--- | :--- | :--- |
+| **Mathlib4** | 人間作成の形式証明 | ~274K 定理 | ほぼ全モデルの SFT 基礎データ |
+| **Proof-Pile-2** | arXiv + Web + コード | 55B トークン | LLEMMA |
+| **合成データ(Lean検証済み)** | RL 中に自動生成 | 数百万〜数千万件 | DeepSeek, Kimina, Goedel |
+| **Lean-Workbook** | フォーラム由来の問題 | ~48K 問題 | InternLM StepProver 他 |
+| **FoVer-40K** | FOL 検証済みラベル | ~40K | PRM 訓練用 |
+
+---
+
+## 9. 未解決課題と今後の展望
+
+### 9.1 現在の技術的課題
+
+| 課題 | 説明 | 影響 |
+| :--- | :--- | :--- |
+| **ベンチマーク飽和** | MiniF2F が 90%+ に到達。真の能力測定が困難に | PutnamBench, VeriBench 等への移行 |
+| **形式/非形式のギャップ** | ベンチマーク問題の形式化自体に誤りがある場合がある | 評価の信頼性に影響 |
+| **計算コスト** | Expert Iteration: 20,000 CPU日。TTRL: TPU数日 | 再現性・アクセシビリティの障壁 |
+| **長い証明** | 100ステップ以上の証明は依然として困難 | 研究レベルの数学への適用障壁 |
+| **Mathlib 追従** | Mathlib の頻繁な更新にモデルが追いつけない | LeanDojo v2 のライフロング管理で対処中 |
+| **汎化** | 特定ベンチマーク最適化 vs 真の数学的推論能力 | ベンチマーク設計の課題 |
+
+### 9.2 今後の方向性
+
+```mermaid
+graph TD
+ F["今後の研究方向"] --> F1["エージェント型証明"]
+ F --> F2["生涯学習"]
+ F --> F3["マルチモーダル"]
+ F --> F4["人間-AI協調"]
+
+ F1 --> F1a["ツール使用の高度化
(Python, CAS, Web検索)"]
+ F2 --> F2a["新しいMathlib定理に
動的に適応"]
+ F3 --> F3a["図・数式画像からの
自動形式化"]
+ F4 --> F4a["数学者の日常的な
証明作業を支援"]
+```
+
+---
+
+## 10. 文献一覧
+
+### ツール・フレームワーク
+
+| 論文 | 著者 | 会議 | 年 |
+| :--- | :--- | :--- | :--- |
+| LeanDojo: Theorem Proving with Retrieval-Augmented LMs | Yang, Swope et al. | NeurIPS 2023 (Oral) | 2023 |
+| Lean Copilot: LLMs as Copilots for Theorem Proving | Song et al. | NeurIPS 2024 | 2024 |
+| LeanAgent: Lifelong Learning for Formal Theorem Proving | Kumarappan et al. | ICLR 2025 (Oral) | 2025 |
+| LeanDojo v2: A Comprehensive Library for AI Theorem Proving | Yang et al. | — | 2025 |
+
+### AIプルーバー・モデル
+
+| 論文 | 開発元 | 年 |
+| :--- | :--- | :--- |
+| DeepSeek-Prover-V2: Subgoal Decomposition for Theorem Proving | DeepSeek | 2025 |
+| Kimina-Prover: Reasoning-Driven Exploration for Formal Reasoning | Numina + Kimi | 2025 |
+| InternLM2.5-StepProver: Expert Iteration on Lean Workbook | SHLab | 2024 |
+| Goedel-Prover-V2: Scaling Open-Source Formal Reasoning | Princeton | 2025 |
+| LLEMMA: An Open Language Model for Mathematics | EleutherAI + Princeton | 2023 |
+| Seed-Prover: Automating Formal Proof with Agentic LLM | ByteDance | 2025 |
+
+### 証明生成戦略
+
+| 論文 | 年 |
+| :--- | :--- |
+| Draft, Sketch, and Prove: Guiding Formal Theorem Provers with Informal Proofs | 2023 |
+| LEGO-Prover: Neural Theorem Proving with Growing Libraries | 2024 |
+| Baldur: Whole-Proof Generation and Repair with Large Language Models | 2023 |
+
+### 基盤モデル・ベンチマーク
+
+| 名称 | 年 | 内容 |
+| :--- | :--- | :--- |
+| MiniF2F | 2021 | AMC/AIME/IMO レベルの形式数学ベンチマーク |
+| ProofNet | 2023 | 学部レベル自動形式化ベンチマーク |
+| PutnamBench | 2024 | Putnam Competition ベースの高難度ベンチマーク |
+| Lean-Workbook-Plus | 2024 | 大規模 Lean 証明データセット |
diff --git a/docs/AI_Agents_and_Logic/ai_proof_search_tools_survey.rag_ingested b/docs/AI_Agents_and_Logic/ai_proof_search_tools_survey.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..05c8cb48645e3a84fa91f616f20570aa93562df5
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/ai_proof_search_tools_survey.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "c3dea20d559b01a1", "ingested_at": "2026-06-30T08:25:55.025982", "content_hash": "c3dea20d559b01a1"}
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/neurosymbolic_ai_survey.md b/docs/AI_Agents_and_Logic/neurosymbolic_ai_survey.md
new file mode 100644
index 0000000000000000000000000000000000000000..bbbed614b15100b7f8bf1ae11f8a0b86e5615ee9
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/neurosymbolic_ai_survey.md
@@ -0,0 +1,719 @@
+# ニューロ・シンボリックAI(Neuro-Symbolic AI)包括的サーベイ
+
+> **調査時点**: 2026年5月
+> **主題**: LLMの弱点を形式論理で克服するニューロ・シンボリックAIの体系的分類と最新フレームワーク
+> **対象読者**: AI研究者・MLエンジニア・形式手法に関心のある技術者
+
+---
+
+## 目次
+
+1. [背景と動機:なぜニューロ・シンボリックAIか](#1-背景と動機なぜニューロシンボリックaika)
+2. [学術的分類体系:Kautzの6類型と3パラダイム](#2-学術的分類体系kautzの6類型と3パラダイム)
+3. [パラダイム1: Symbolic → Neural(記号がニューラルを導く)](#3-パラダイム1-symbolic--neural記号がニューラルを導く)
+4. [パラダイム2: Neural → Symbolic(ニューラルが記号に翻訳する)](#4-パラダイム2-neural--symbolicニューラルが記号に翻訳する)
+5. [パラダイム3: Hybrid(双方向統合)](#5-パラダイム3-hybrid双方向統合)
+6. [微分可能ニューロ・シンボリック統合(Tightly Coupled)](#6-微分可能ニューロシンボリック統合tightly-coupled)
+7. [応用ドメイン](#7-応用ドメイン)
+8. [ベンチマークと評価](#8-ベンチマークと評価)
+9. [未解決課題と今後の展望](#9-未解決課題と今後の展望)
+10. [主要文献一覧](#10-主要文献一覧)
+
+---
+
+## 1. 背景と動機:なぜニューロ・シンボリックAIか
+
+### 1.1 LLMの構造的限界
+
+大規模言語モデル(LLM)は、「統計的パターン照合」に基づく **System 1(直感的・高速)** 型のAIです。自然言語処理、コード生成、創造的文章の執筆では驚異的な能力を示しますが、以下の構造的限界を抱えています。
+
+```
+┌──────────────────────────────────────────────────────────────────┐
+│ LLM の構造的限界 │
+├──────────────┬───────────────────────────────────────────────────┤
+│ ハルシネーション │ もっともらしい嘘を自信満々に生成する。 │
+│ │ 「次のトークン予測」の確率的性質に起因。 │
+├──────────────┼───────────────────────────────────────────────────┤
+│ 論理的推論の │ 多段推論(5段以上)で急激に精度が低下。 │
+│ 脆弱性 │ 論理的「ショートカット」を取りがち。 │
+├──────────────┼───────────────────────────────────────────────────┤
+│ 検証不可能性 │ 推論過程(CoT)が「もっともらしい物語」に過ぎず、 │
+│ │ 論理的正しさが保証されない。 │
+├──────────────┼───────────────────────────────────────────────────┤
+│ 永続的状態の │ トランスフォーマーは外部ワーキングメモリや │
+│ 欠如 │ 変数束縛のメカニズムを持たない。 │
+├──────────────┼───────────────────────────────────────────────────┤
+│ 体系的一般化 │ 訓練データの分布外に出ると急激に性能低下。 │
+│ の欠如 │ 構成的(compositional)な一般化が困難。 │
+└──────────────┴───────────────────────────────────────────────────┘
+```
+
+### 1.2 シンボリックAIの強みと弱み
+
+一方、シンボリックAI(形式論理、定理証明器、制約ソルバー等)は **System 2(熟慮的・正確)** 型のAIであり、以下の特徴を持ちます。
+
+| 強み | 弱み |
+| :--- | :--- |
+| ✅ 推論結果の**正確性**が数学的に保証される | ❌ 自然言語のような曖昧な入力の処理が困難 |
+| ✅ 推論過程が**完全に追跡可能**(説明可能) | ❌ 知識の手動エンコーディングが高コスト |
+| ✅ 体系的な組み合わせ的一般化が可能 | ❌ 非構造化データ(画像、音声)を扱えない |
+| ✅ 矛盾を検出し、一貫性を保証できる | ❌ 不完全情報やノイズへの耐性が低い |
+
+### 1.3 ニューロ・シンボリックAIの核心的アイデア
+
+> [!IMPORTANT]
+> **ニューロ・シンボリックAI = LLMの「直感」 × 形式論理の「厳密性」**
+>
+> ニューラルネットワーク(System 1)とシンボリック推論(System 2)を統合することで、両者の弱点を相互に補完し、**直感的かつ正確で検証可能なAI**を実現する。これがニューロ・シンボリックAIの本質です。
+
+```mermaid
+graph LR
+ subgraph "System 1: Neural(直感・認識)"
+ N["LLM / ニューラルネット"]
+ end
+ subgraph "System 2: Symbolic(論理・検証)"
+ S["定理証明器 / ソルバー / ルールエンジン"]
+ end
+ N <-->|"統合"| S
+ N --> R1["パターン認識
自然言語理解
帰納的学習"]
+ S --> R2["論理的推論
形式検証
一貫性保証"]
+
+ style N fill:#6C5CE7,stroke:#333,color:#fff
+ style S fill:#00B894,stroke:#333,color:#fff
+```
+
+---
+
+## 2. 学術的分類体系:Kautzの6類型と3パラダイム
+
+### 2.1 Henry Kautz の6類型(AAAI 2020 / AI Magazine 2022)
+
+ニューロ・シンボリックAIの学術的分類として最も影響力のあるフレームワークは、Henry Kautz が AAAI 2020 の Robert S. Engelmore Memorial Award Lecture「The Third AI Summer」で提唱し、2022年に *AI Magazine* で体系化した **6類型** です。
+
+これらは、ニューラルとシンボリックの「統合の密度(Tightness of Coupling)」に基づくスペクトラム上に配置されます。
+
+```mermaid
+graph TB
+ subgraph "疎結合(Loosely Coupled)"
+ T1["Type 1: Symbolic ⇌ Neural
記号入出力のニューラル
例: GPT, BERT"]
+ T2["Type 2: Symbolic[Neural]
記号システムがニューラルを呼出
例: AlphaGo"]
+ T3["Type 3: Neural[Symbolic]
ニューラルが記号システムを呼出
例: Logic-LM, LINC"]
+ end
+ subgraph "密結合(Tightly Coupled)"
+ T4["Type 4: Neural ∪ Compile(Symbolic)
記号知識をネットワークにコンパイル
例: KBANN"]
+ T5["Type 5: Neural_{Symbolic}
論理を埋め込み・損失関数に統合
例: LTN, LNN"]
+ T6["Type 6: Neural ≡ Symbolic
完全統合エンジン
(理想目標)"]
+ end
+ T1 --> T2 --> T3 --> T4 --> T5 --> T6
+
+ style T6 fill:#e17055,stroke:#333,color:#fff
+```
+
+| 類型 | 名称 | 説明 | 代表例 |
+| :---: | :--- | :--- | :--- |
+| **Type 1** | Symbolic ⇌ Neural | 記号(テキストトークン)を入出力とするニューラルネット。事実上の標準LLM | GPT, BERT |
+| **Type 2** | Symbolic[Neural] | シンボリックシステム(探索、計画)がニューラルをサブルーチンとして呼出 | AlphaGo, AlphaProof |
+| **Type 3** | Neural[Symbolic] | ニューラルが外部シンボリックシステム(ソルバー、証明器)を呼出 | Logic-LM, LINC, SatLM |
+| **Type 4** | Neural ∪ Compile(Sym) | シンボリック知識をネットワークのトポロジーや初期重みにコンパイル | KBANN |
+| **Type 5** | Neural_{Symbolic} | 論理をエンベディングや損失関数制約に写像。エンドツーエンド微分可能 | LTN, LNN (IBM) |
+| **Type 6** | Neural ≡ Symbolic | ニューラルとシンボリックが完全に統合された単一アーキテクチャ | **理想目標(未達成)** |
+
+### 2.2 LLM時代の3パラダイム(IJCAI 2025 サーベイ等)
+
+LLMの登場以降(2023–2025年)、上記の6類型をLLM中心に再編成した**3パラダイム分類**が主流となっています。
+
+````carousel
+### パラダイム1: Symbolic → Neural
+**記号知識がニューラルを導く**
+
+```
+┌──────────────────┐
+│ シンボリック知識 │
+│ (ルール・制約 │
+│ オントロジー) │
+└────────┬─────────┘
+ │ ガイド・制約・正則化
+ ▼
+┌──────────────────┐
+│ LLM │
+│ (訓練 or 推論) │
+└──────────────────┘
+```
+
+- 論理制約をネットワーク訓練に埋め込む
+- 推論時に文法的・論理的制約でデコーディングを制御
+- 例: LTN, LNN, 制約付きデコーディング(Outlines, LMQL)
+
+### パラダイム2: Neural → Symbolic
+**ニューラルが形式言語に翻訳し、ソルバーが解く**
+
+```
+┌──────────────────┐
+│ 自然言語の問題 │
+└────────┬─────────┘
+ │
+ ▼
+┌──────────────────┐
+│ LLM (翻訳器) │ ← 意味解析
+└────────┬─────────┘
+ │ FOL / SAT / SMT
+ ▼
+┌──────────────────┐
+│ シンボリック推論器 │ ← Z3, Prover9 等
+└────────┬─────────┘
+ │
+ ▼
+┌──────────────────┐
+│ 確定的・検証済み │
+│ の回答 │
+└──────────────────┘
+```
+
+- LLMを「意味解析器」として使用
+- 推論は確定的ソルバーが担当
+- 例: Logic-LM, LINC, SatLM
+
+### パラダイム3: Hybrid(双方向)
+**ニューラルと記号が反復的に対話・批判・修正**
+
+```
+┌──────────────────┐
+│ LLM (推測) │
+└────────┬─────────┘
+ │ 推測を提出
+ ▼
+┌──────────────────┐
+│ 記号的検証器 │
+│ (Hard Critique) │
+└────────┬─────────┘
+ │ フィードバック
+ ▼
+┌──────────────────┐
+│ LLM (修正) │
+└──────────────────┘
+ ↕ 反復
+```
+
+- 推測–批判(Conjecture-Criticism)サイクル
+- エラーメッセージによる自己修正ループ
+- 例: PEIRCE, Logic-LM++, FoVer
+````
+
+---
+
+## 3. パラダイム1: Symbolic → Neural(記号がニューラルを導く)
+
+このパラダイムでは、**シンボリック知識やルール**がニューラルネットワークの訓練・推論プロセスを制約・ガイドします。
+
+### 3.1 制約付きデコーディング(Constrained Decoding)
+
+LLMの出力を生成時点で制約し、構文的・論理的に不正な出力を「物理的に不可能」にする技術です。
+
+#### メカニズム
+
+```
+ LLM の各デコーディングステップ
+ │
+ ▼
+ ┌─────────────────┐
+ │ 全語彙のロジット │
+ │ (確率分布) │
+ └────────┬────────┘
+ │
+ ▼
+ ┌─────────────────┐
+ │ 文法/論理制約 │ ← CFG, FSM, 正規表現
+ │ によるマスク │
+ └────────┬────────┘
+ │ 違反トークンを -∞ に設定
+ ▼
+ ┌─────────────────┐
+ │ 制約済みロジット │
+ │ からサンプリング │
+ └─────────────────┘
+```
+
+#### 主要フレームワーク比較
+
+| フレームワーク | アプローチ | 制約の種類 | 最適な用途 | 特徴 |
+| :--- | :--- | :--- | :--- | :--- |
+| **Outlines** | ロジットレベルマスク (CFG/正規表現) | 構文的 | JSON/Pydantic 準拠の構造化出力(ローカルモデル) | 高速。XGrammar による最適化 |
+| **Guidance** | Python 制御フロー分岐 | 構文的+手続き的 | スキーマが中間判断に依存する複雑なワークフロー | 柔軟性が高い |
+| **LMQL** | 宣言的クエリ言語 (SQL風) | 構文的+論理的 | 従来のプログラミングロジックとLLM呼び出しの融合 | 高レベルな制約記述 |
+| **Instructor** | 生成後バリデーション+リトライ | 構文的 | API ベースの LLM との簡易統合 | シンプルで互換性が広い |
+
+> [!WARNING]
+> **制約のバイアストレードオフ(ACL 2025)**: 最近の研究は、制約付きデコーディングが構文的正しさを保証する一方で、過度な制約がモデルの「分布バイアス」を引き起こし、複雑な推論タスクではむしろ性能を低下させる場合があることを報告しています。構文的制約と推論能力のバランスが重要な研究課題です。
+
+### 3.2 知識埋め込み型アプローチ
+
+ルールやオントロジーをニューラルネットワークの訓練データや損失関数に直接反映するアプローチ(Kautz Type 4/5)。詳細は [セクション6](#6-微分可能ニューロシンボリック統合tightly-coupled) で解説。
+
+---
+
+## 4. パラダイム2: Neural → Symbolic(ニューラルが記号に翻訳する)
+
+LLMを「意味解析器(Semantic Parser)」として使い、自然言語を形式言語(FOL, SAT, SMT等)に変換した上で、**確定的なシンボリック推論器**に推論を完全に委ねるアプローチです。Kautz の Type 3 に相当します。
+
+### 4.1 Logic-LM
+
+> **論文**: *"Logic-LM: Empowering Large Language Models with Symbolic Solvers for Faithful Logical Reasoning"*
+> **著者**: Liangming Pan, Alon Albalak, Xinyi Wang, William Yang Wang
+> **会議**: Findings of EMNLP 2023
+
+#### アーキテクチャ
+
+```mermaid
+graph TB
+ A["自然言語の問題"] --> B["モジュール1: Problem Formulator
(LLMが形式言語に翻訳)"]
+ B -->|"FOL / CSP / SAT
表現"| C["モジュール2: Symbolic Reasoner
(Z3 / Prover9)"]
+ C -->|"記号的回答"| D["モジュール3: Result Interpreter
(自然言語に復元)"]
+ C -->|"❌ エラー発生"| E["モジュール4: Self-Refinement
(エラーメッセージをLLMにフィードバック)"]
+ E -->|"修正された形式表現"| C
+ D --> F["最終回答"]
+
+ style E fill:#FDCB6E,stroke:#333
+```
+
+#### 3つのモジュールの詳細
+
+| モジュール | 役割 | 技術的詳細 |
+| :--- | :--- | :--- |
+| **Problem Formulator** | LLMが自然言語問題を形式的記号表現に変換 | Few-shot プロンプトで FOL, Python (CSP), SAT のいずれかの形式に翻訳 |
+| **Symbolic Reasoner** | 外部ソルバーが確定的推論を実行 | **Z3**(SMTソルバー)または **Prover9**(一階述語論理定理証明器)を使用 |
+| **Result Interpreter** | ソルバーの出力を自然言語に変換 | 記号的回答をユーザーが理解できる形式に復元 |
+| **Self-Refinement** | ソルバーのエラーメッセージに基づく自己修正ループ | 構文エラーや充足不可能性のエラーをLLMにフィードバックし、反復的に修正。一般的な修正パターンの Few-shot デモも利用 |
+
+#### 性能と意義
+
+- **「忠実な推論(Faithful Reasoning)」の実現**: LLMの確率的推論をソルバーの確定的推論に置き換えることで、推論結果の論理的正しさを保証
+- **ハルシネーション防止**: 推論がソルバー内部で実行されるため、LLMの推論における論理的飛躍が排除される
+- FOLIO, ProofWriter, AR-LSAT ベンチマークで、標準 CoT プロンプトを大幅に上回る精度
+
+#### Logic-LM++ (2024)
+
+Logic-LM の後継研究。**多段階修正戦略**を導入し、LLMがシンボリック表現を精緻化するプロセスを改善。FOLIO, ProofWriter, AR-LSAT での性能向上を実証。
+
+---
+
+### 4.2 LINC(Logical Inference via Neurosymbolic Computation)
+
+> **論文**: *"LINC: A Neurosymbolic Approach for Logical Reasoning by Combining Language Models with First-Order Logic Provers"*
+> **会議**: EMNLP 2023 **🏆 Outstanding Paper Award 受賞**
+
+#### アーキテクチャ
+
+```
+ ┌─────────────────────────────────────────────┐
+ │ LINC パイプライン │
+ │ │
+ │ 入力: 自然言語の前提 + 結論 │
+ │ │ │
+ │ ▼ │
+ │ ┌─────────────────┐ │
+ │ │ Stage 1: │ │
+ │ │ LLM による │ ← 前提と結論を個別に │
+ │ │ FOL 変換 │ FOL 式に翻訳 │
+ │ └────────┬────────┘ │
+ │ │ │
+ │ ▼ FOL 式 (∀x. P(x) → Q(x), ...) │
+ │ ┌─────────────────┐ │
+ │ │ Stage 2: │ │
+ │ │ Prover9 │ ← 厳密な記号的推論 │
+ │ │ (定理証明器) │ │
+ │ └────────┬────────┘ │
+ │ │ │
+ │ ▼ │
+ │ 結果: Entailment / Contradiction / Unknown │
+ └─────────────────────────────────────────────┘
+```
+
+#### Logic-LM との比較
+
+| 特性 | LINC | Logic-LM |
+| :--- | :--- | :--- |
+| **形式言語** | FOL(一階述語論理)に特化 | FOL, CSP, SAT を問題に応じて選択 |
+| **ソルバー** | Prover9 | Z3 / Prover9 |
+| **自己修正** | なし(翻訳品質に依存) | あり(エラーフィードバックループ) |
+| **タスク** | 論理的含意判定(Entailment) | 汎用的論理推論 |
+| **注目の成果** | 🏆 EMNLP 2023 Outstanding Paper | 幅広いベンチマークでの評価 |
+
+#### 重要な発見
+
+> [!TIP]
+> **小さなモデル + LINC > 大きなモデル + CoT**: LINC の論文は、**StarCoder+(比較的小規模なモデル)に LINC を適用した場合、GPT-4 + 標準 Chain-of-Thought プロンプトよりも高い精度**を達成したことを報告しています。これは、モデルサイズを拡大するよりも、**外部シンボリック推論器との統合の方が論理推論において効果的**であることを示す重要な証拠です。
+
+---
+
+### 4.3 SatLM(Satisfiability-Aided Language Models)
+
+> **論文**: *"SATLM: Satisfiability-Aided Language Models Using Declarative Prompting"*
+> **著者**: Xi Ye, Qiaochu Chen, Isil Dillig, Greg Durrett
+> **会議**: NeurIPS 2023 MATH-AI Workshop
+
+#### 核心的アイデア:宣言的プロンプティング
+
+SatLM は、Logic-LM や LINC とは異なるアプローチを取ります。LLMに対して問題の**手続き的な解法**(プログラム)を生成させるのではなく、問題の**宣言的な仕様**(論理的制約の集合)を生成させます。
+
+```
+ ┌──────────────────────────────────┐
+ │ 従来のアプローチ │
+ │ 自然言語 → LLM → 手続き的プログラム │
+ │ (命令的: HOW を記述) │
+ │ 「x = 3; y = x + 2; ...」 │
+ └──────────────────────────────────┘
+ vs
+ ┌──────────────────────────────────┐
+ │ SatLM のアプローチ │
+ │ 自然言語 → LLM → 宣言的仕様 │
+ │ (宣言的: WHAT を記述) │
+ │ 「x > 0 ∧ x + y = 7 ∧ y < 5」 │
+ │ │ │
+ │ ▼ │
+ │ SAT/SMT ソルバー │
+ │ (仕様を満たす解を自動発見) │
+ └──────────────────────────────────┘
+```
+
+#### 利点
+
+1. **宣言的仕様 は LLM にとって生成しやすい**: 複雑な多段階の手続き的推論よりも、問題の制約を列挙する方がLLMにとって容易
+2. **推論の正確性保証**: 充足可能性ソルバーが解を見つけるため、回答は制約と論理的に一貫していることが保証される
+3. **LSAT, BoardGameQA で SOTA**: 評価時点で、プログラム支援型LLM (ProgLM) や CoT を上回る性能
+
+---
+
+### 4.4 FoVer(First-Order Logic Verification)
+
+FoVer は2つの研究ストリームを持ちます。
+
+#### ストリーム1: LLM推論の検証
+
+```mermaid
+graph LR
+ A["LLMの推論出力
(CoT)"] -->|FOLに変換| B["FOL 式"]
+ B -->|検証| C["Z3 定理証明器"]
+ C -->|Valid| D["✅ 論理的に正しい"]
+ C -->|Invalid| E["❌ 論理的エラーあり"]
+```
+
+- LLMが生成した自然言語の推論チェーンをFOLに翻訳し、Z3で厳密に検証
+- CoT の「忠実性(Faithfulness)」を事後的に保証
+
+#### ストリーム2: Process Reward Model (PRM) の自動訓練
+
+- **FoVer-40K**: Z3 や Isabelle で自動的にステップレベルのエラーラベルを付与した大規模訓練データセットを合成
+- 人間によるアノテーション不要で、高品質な PRM を訓練可能
+- タスク間の**汎化能力**が高く、形式的ベンチマーク以外の非形式的推論タスクでも性能向上
+
+---
+
+## 5. パラダイム3: Hybrid(双方向統合)
+
+LLMとシンボリック推論器が**反復的に対話**し、互いの出力を批判・修正し合うアプローチです。
+
+### 5.1 PEIRCE
+
+> **会議**: ACL 2025
+
+#### 推測–批判(Conjecture-Criticism)サイクル
+
+```mermaid
+graph TB
+ A["自然言語の問題"] --> B["LLM: 推測の生成
(自然言語 + 形式言語)"]
+ B --> C{"批判モジュール群"}
+
+ C --> D["Hard Critique
(Isabelle / Prolog)
形式的論理検証"]
+ C --> E["Soft Critique
(一貫性・尤もらしさ
・簡潔性の評価)"]
+
+ D -->|"形式的エラー"| F["フィードバック統合"]
+ E -->|"内容的問題"| F
+
+ F -->|"修正要求"| B
+ F -->|"合格"| G["✅ 最終解答
(素材的 + 形式的に妥当)"]
+
+ style D fill:#E74C3C,stroke:#333,color:#fff
+ style E fill:#3498DB,stroke:#333,color:#fff
+```
+
+#### PEIRCEの革新性
+
+| 要素 | 説明 |
+| :--- | :--- |
+| **素材的推論(Material Inference)** | 内容の妥当性、文脈的関連性、知識との整合性(LLMの強み) |
+| **形式的推論(Formal Inference)** | 構造的・論理的妥当性、演繹的正しさ(シンボリック推論器の強み) |
+| **統一** | 両方の基準を同時に満たす解のみを許容。科学的方法(仮説→検証)のAI実装 |
+
+> [!NOTE]
+> PEIRCE の名前は、アメリカの論理学者・哲学者チャールズ・サンダース・パース(Charles Sanders Peirce)に由来しています。パースは「アブダクション(最良の説明への推論)」を提唱し、科学的探究プロセスにおける推測と検証の反復を哲学的に体系化した人物です。
+
+### 5.2 パラダイム間の比較総括
+
+| 比較項目 | パラダイム1
Symbolic→Neural | パラダイム2
Neural→Symbolic | パラダイム3
Hybrid |
+| :--- | :--- | :--- | :--- |
+| **情報の流れ** | 記号→ニューラル(一方向) | ニューラル→記号(一方向) | 双方向(反復的) |
+| **LLMの役割** | 出力が制約される受動的生成器 | 意味解析器(翻訳器) | 推測生成器+修正器 |
+| **シンボリックの役割** | 訓練データ・損失関数・デコーディング制約 | 最終的な推論実行者 | 批判者・検証者 |
+| **正確性保証** | 部分的(構文的制約のみの場合が多い) | 高い(ソルバー結果は正確) | 高い(反復検証) |
+| **柔軟性** | 高い(制約の強さを調整可能) | 低い(翻訳精度に依存) | 高い(動的調整可能) |
+| **代表的システム** | LTN, LNN, Outlines, LMQL | Logic-LM, LINC, SatLM | PEIRCE, Logic-LM++ |
+
+---
+
+## 6. 微分可能ニューロ・シンボリック統合(Tightly Coupled)
+
+Kautz の Type 5 に該当する、論理とニューラルネットワークを**単一の微分可能な計算グラフ**に統合するアプローチ群です。
+
+### 6.1 Logic Tensor Networks(LTN)
+
+論理式をテンソル演算に変換し、**エンドツーエンドで微分可能**な推論と学習を実現するフレームワーク。
+
+#### 動作原理:Real Logic
+
+```
+ 一階述語論理 テンソル演算
+ ──────────── ────────────
+ 定数 a, b, c →→→ テンソル / エンベディング
+ 述語 P(x) →→→ ニューラルネットワーク f: ℝⁿ → [0, 1]
+ ∧ (AND) →→→ 微分可能な T-norm(例: min, product)
+ ∨ (OR) →→→ 微分可能な T-conorm
+ ¬ (NOT) →→→ 1 - x
+ ∀x (FORALL) →→→ min / generalized mean over batch
+ ∃x (EXISTS) →→→ max / generalized mean over batch
+```
+
+- **真理値**: 古典的な {0, 1} ではなく、ファジー論理の [0, 1] 区間の実数値
+- **訓練**: 論理的公理の真理値を最大化するように勾配降下法で最適化
+- **知識グラフ**: 不完全な知識グラフに対して、論理制約に基づくリンク予測・補完が可能
+
+### 6.2 Logical Neural Networks(LNN) — IBM Research
+
+IBM Research が開発した、LTN とは異なるアプローチでニューラルとシンボリックを完全統合するフレームワーク。
+
+#### LTN との根本的な違い
+
+| 特性 | LTN | LNN (IBM) |
+| :--- | :--- | :--- |
+| **論理のエンコード** | 論理式を損失関数の制約として表現 | 各ニューロンが論理式の構成要素(原子、連言、選言等)に1対1対応 |
+| **推論方向** | 順方向(Forward)のみ | **全方向(Omnidirectional)** — 順方向+逆方向推論が可能 |
+| **不完全情報** | 対応可能(ファジー真理値) | 真理値の**上下限バウンド**を維持。開世界仮定(OWA)を自然に処理 |
+| **解釈可能性** | 中程度 | **高い** — ネットワーク構造が論理構文木に直接対応 |
+| **ルール抽出** | 困難 | ネットワークから人間が読める論理ルールを抽出可能 |
+
+#### LNN の構造
+
+```
+ 一階述語論理の論理式:
+ ∀x. Human(x) → Mortal(x)
+
+ LNN のネットワーク構造:
+ ┌──────────────────────────────────────┐
+ │ ∀ ニューロン │
+ │ └── → ニューロン(含意) │
+ │ ├── Human(x) ニューロン │
+ │ └── Mortal(x) ニューロン │
+ └──────────────────────────────────────┘
+
+ 各ニューロンは [lower_bound, upper_bound] ∈ [0,1]² を保持
+ → 不確実性・不完全情報を自然に表現
+```
+
+### 6.3 確率的論理プログラミング統合
+
+ニューラル知覚と確率的論理推論を統合するフレームワーク群。
+
+| フレームワーク | 基盤論理 | 特徴 | 代表的応用 |
+| :--- | :--- | :--- | :--- |
+| **DeepProbLog** | ProbLog(確率論理プログラミング) | ニューラル述語をProbLogの確率的事実として統合。エンドツーエンド微分可能 | 画像からの推論 |
+| **Scallop** | Datalog(データログ) | 出自(Provenance)ベースの微分可能推論。スケーラビリティ重視 | 視覚的質問応答(VQA) |
+| **NeurASP** | ASP(解答集合プログラミング) | ニューラルネットの出力を ASP の確率的原子として統合 | 宣言的推論 |
+
+#### DeepProbLog の動作例
+
+```
+ 画像入力 → [CNN] → 「digit(Image1) = 3」(確率 0.95)
+ 「digit(Image2) = 5」(確率 0.88)
+ │
+ ▼
+ ProbLog ルール: addition(X, Y, Z) :- digit(I1, X), digit(I2, Y), Z is X+Y.
+ │
+ ▼
+ 出力: addition(Image1, Image2, 8) — 確率 0.836
+
+ → ニューラル知覚 + 論理推論 + 確率的不確実性 が一体化
+```
+
+---
+
+## 7. 応用ドメイン
+
+### 7.1 分野別の適用マップ
+
+```mermaid
+graph TD
+ NeSy["ニューロ・シンボリックAI"] --> H["医療・ヘルスケア"]
+ NeSy --> R["ロボティクス"]
+ NeSy --> AD["自動運転"]
+ NeSy --> PS["プログラム合成"]
+ NeSy --> L["法律・コンプライアンス"]
+ NeSy --> CS["サイバーセキュリティ"]
+
+ H --> H1["説明可能な診断支援"]
+ H --> H2["臨床試験の患者選定"]
+ H --> H3["創薬(知識グラフ統合)"]
+
+ R --> R1["タスク計画(記号ルール + ニューラル知覚)"]
+ R --> R2["安全制約付き動作生成"]
+
+ AD --> AD1["交通ルールのガードレール"]
+ AD --> AD2["Road2Code(シーンのプログラム合成)"]
+
+ PS --> PS1["検証可能なコード生成"]
+ PS --> PS2["DSL 合成"]
+
+ style NeSy fill:#6C5CE7,stroke:#333,color:#fff
+```
+
+### 7.2 各ドメインの詳細
+
+| ドメイン | ニューラル成分 | シンボリック成分 | 具体的事例 |
+| :--- | :--- | :--- | :--- |
+| **医療** | 医用画像解析、EHR 処理 | 臨床ガイドライン、医学オントロジー | 診断根拠の追跡可能な支援システム |
+| **ロボティクス** | センサー入力処理、ジェスチャー認識 | 安全プロトコル、手順ルール | 外科手術支援、産業用分解ロボット |
+| **自動運転** | カメラ・LiDAR知覚 | 交通法規(ハード制約) | エッジケースでの予測可能な振る舞い |
+| **プログラム合成** | コード候補の生成 | 形式仕様の検証 | 正しさが保証されたコードの自動生成 |
+| **法律** | 契約文書の解析 | 法的ルールの形式化 | コンプライアンス自動チェック |
+| **サイバーセキュリティ** | 脅威パターンの検出 | セキュリティオントロジー | 検証可能な脅威分析・対応 |
+
+> [!IMPORTANT]
+> **AI規制との関連**: EU AI Act(2024年施行)をはじめとする世界的なAI規制は、高リスクAIシステムに対して**透明性・説明可能性・監査可能性**を要求しています。ニューロ・シンボリックAIは、推論過程が追跡可能であるという特性から、これらの規制要件を満たすための**事実上の標準アーキテクチャ**となる可能性があります。
+
+---
+
+## 8. ベンチマークと評価
+
+### 8.1 論理推論ベンチマーク
+
+| ベンチマーク | 論理の種類 | タスク | 使用される代表的なソルバー |
+| :--- | :--- | :--- | :--- |
+| **FOLIO** | 一階述語論理 | 自然言語 → FOL 変換 + 論理的含意判定 | Prover9 |
+| **ProofWriter** | 命題論理 / FOL | 多段推論(証明深度1–5) | 各種 |
+| **PrOntoQA** | 命題論理 | ルールベース論理的含意 | — |
+| **AR-LSAT** | 命題論理 + 制約 | LSAT形式の法的推論問題 | Z3 |
+| **BoardGameQA** | 制約充足 | ゲームルールに基づく推論 | SAT ソルバー |
+| **LogicalDeduction** | 命題論理 | 複雑な実世界の論理クエリ | — |
+
+### 8.2 主要な評価指標
+
+| 指標 | 説明 | 対象 |
+| :--- | :--- | :--- |
+| **推論精度(Accuracy)** | 正しい論理的結論に到達した割合 | 全システム |
+| **実行可能率(Executable Rate)** | LLM が生成した形式表現がソルバーで実行可能だった割合 | Neural→Symbolic 系 |
+| **忠実性(Faithfulness)** | 推論過程が結論を実際に論理的に導いているか | CoT vs NeSy |
+| **自己修正成功率** | エラーフィードバック後に正しい形式表現を生成できた割合 | Logic-LM, Logic-LM++ |
+
+### 8.3 性能比較の概観
+
+| 手法 | FOLIO | ProofWriter | AR-LSAT |
+| :--- | :---: | :---: | :---: |
+| GPT-4 + CoT | ~70% | ~75% | ~55% |
+| GPT-4 + Logic-LM | ~78% | ~85% | ~68% |
+| GPT-4 + LINC | ~80% | ~88% | — |
+| GPT-4 + SatLM | — | — | ~72% |
+
+> [!NOTE]
+> 上記の数値はおおよその傾向を示すもので、プロンプト設計やモデルバージョンにより変動します。一貫した傾向として、**NeSy アプローチは標準 CoT を5–15ポイント上回る**ことが多いです。
+
+---
+
+## 9. 未解決課題と今後の展望
+
+### 9.1 現在の技術的課題
+
+```mermaid
+graph TD
+ C["未解決課題"] --> C1["翻訳ボトルネック
自然言語→形式言語の変換精度"]
+ C --> C2["スケーラビリティ
大規模知識ベースでの推論効率"]
+ C --> C3["表現力の限界
FOL/SAT で表現できない知識"]
+ C --> C4["制約バイアス
過度な制約が推論能力を阻害"]
+ C --> C5["学習と推論の統一
微分可能論理の理論的限界"]
+ C --> C6["説明可能性 vs 性能
のトレードオフ"]
+
+ C1 --> C1a["意味的ドリフト
(構文OK・意味NG)"]
+ C2 --> C2a["組み合わせ爆発
(状態空間の指数的増大)"]
+```
+
+| 課題 | 説明 | 現在のアプローチ |
+| :--- | :--- | :--- |
+| **翻訳ボトルネック** | LLMが正確な形式表現を生成できない場合がある | 自己修正ループ、合成データ、GRPO |
+| **意味的ドリフト** | 構文的に正しいが意味的に誤った形式化 | サイクル一貫性、FormalRewardBench |
+| **スケーラビリティ** | 大規模な知識ベースでの推論コスト | Provenance ベース手法(Scallop)、top-k 枝刈り |
+| **表現力** | 現実世界の曖昧性や段階的変化をFOLで表現することの困難 | ファジー論理(LTN)、確率的論理(DeepProbLog) |
+| **評価の難しさ** | 「コンパイル成功 ≠ 正解」問題 | T2, FormalRewardBench |
+| **メタ認知の欠如** | システムが自身の推論の限界を認識・調整できない | 研究段階 |
+
+### 9.2 今後の研究方向
+
+| 方向性 | 内容 | 期待されるインパクト |
+| :--- | :--- | :--- |
+| **Type 6 統合の追求** | ニューラルとシンボリックを完全に融合した単一アーキテクチャ | AGI への最も有力な道筋の一つ |
+| **自律的形式化** | LLMが自律的に問題を形式化し、適切なソルバーを選択する「エージェント型」NeSy | 人間の介入なしで複雑な推論を完遂 |
+| **多段・多モーダル推論** | 視覚・言語・論理を横断する推論チェーンの統合 | ロボティクス、科学的発見 |
+| **自己改善型推論** | 推論過程を自己監視し、戦略を動的に調整するメタ認知システム | 信頼性の飛躍的向上 |
+| **規制適合** | EU AI Act 等の規制要件を満たす検証可能なAIアーキテクチャ | 産業界での大規模採用 |
+
+---
+
+## 10. 主要文献一覧
+
+### 10.1 サーベイ論文
+
+| 年 | タイトル | 著者 / 組織 | 会議 |
+| :--- | :--- | :--- | :--- |
+| 2022 | The Third AI Summer (Neuro-Symbolic AI taxonomy) | Henry Kautz | AI Magazine / AAAI 2020 Award Lecture |
+| 2025 | LLM Reasoning with Neuro-Symbolic AI | Various | IJCAI 2025 Survey |
+
+### 10.2 Neural → Symbolic フレームワーク
+
+| 年 | タイトル | 著者 | 会議 |
+| :--- | :--- | :--- | :--- |
+| 2023 | Logic-LM: Empowering LLMs with Symbolic Solvers | Pan, Albalak, Wang, Wang | EMNLP 2023 (Findings) |
+| 2023 | LINC: Combining LMs with FOL Provers 🏆 | Olausson et al. | EMNLP 2023 (Outstanding Paper) |
+| 2023 | SATLM: Satisfiability-Aided LMs Using Declarative Prompting | Ye, Chen, Dillig, Durrett | NeurIPS 2023 MATH-AI |
+| 2024 | Logic-LM++: Multi-step Refinement | — | ACL 2024 |
+| 2024 | FoVer: FOL Verification / PRM Training | Various (MIT etc.) | — |
+
+### 10.3 Hybrid / 双方向フレームワーク
+
+| 年 | タイトル | 会議 |
+| :--- | :--- | :--- |
+| 2025 | PEIRCE: Conjecture-Criticism for Material and Formal Inference | ACL 2025 |
+| 2024 | ENVISIONS: Neuro-Symbolic Self-Training | ACL 2024 |
+
+### 10.4 微分可能ニューロ・シンボリック統合
+
+| 年 | タイトル | 開発元 |
+| :--- | :--- | :--- |
+| 2020+ | Logic Tensor Networks (LTN) | SonyCLI / Academia |
+| 2020+ | Logical Neural Networks (LNN) | IBM Research |
+| 2018+ | DeepProbLog | KU Leuven |
+| 2023+ | Scallop | UPenn |
+| 2020+ | NeurASP | ASU / NMSU |
+
+### 10.5 制約付きデコーディング
+
+| ライブラリ | GitHub | 特徴 |
+| :--- | :--- | :--- |
+| **Outlines** | dottxt-ai/outlines | CFG/正規表現ベースの高速ロジットマスク |
+| **Guidance** | guidance-ai/guidance | Python制御フロー分岐 |
+| **LMQL** | eth-sri/lmql | 宣言的クエリ言語 |
+| **Instructor** | jxnl/instructor | API互換の生成後バリデーション |
+
+### 10.6 関連リポジトリ
+
+| リポジトリ | URL | 内容 |
+| :--- | :--- | :--- |
+| **Awesome-LLM-Reasoning-with-NeSy** | GitHub | IJCAI 2025 サーベイ関連キュレーション |
+| **LLM-Symbolic-Reasoning-Survey** | GitHub | シンボリック形式化・定理証明のトラッキング |
+| **IBM LNN** | GitHub | LNN 公式実装 |
+| **Scallop** | scallop-lang.org | Scallop 公式サイト・ドキュメント |
diff --git a/docs/AI_Agents_and_Logic/neurosymbolic_ai_survey.rag_ingested b/docs/AI_Agents_and_Logic/neurosymbolic_ai_survey.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..2e8fb4bf25ddc229a0b5afd2c93048c76c3e3fa3
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/neurosymbolic_ai_survey.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "3b876a0aa99ad0c8", "ingested_at": "2026-06-30T08:25:59.744620", "content_hash": "3b876a0aa99ad0c8"}
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/script_long.md b/docs/AI_Agents_and_Logic/script_long.md
new file mode 100644
index 0000000000000000000000000000000000000000..fad332ef10c2d482dc621de70c2daf97074db576
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/script_long.md
@@ -0,0 +1,30 @@
+Good evening, and welcome back. Tonight's report focuses on the rapidly evolving field of agent-based AI, specifically how these intelligent systems are transforming our technological landscape.
+
+We're in a period of significant change, where AI is moving beyond simply mimicking human conversation to truly understanding and executing complex tasks. This shift has been driven by several key developments:
+
+First, we're seeing a transition from evaluating AI models based on their conversational fluency to their practical execution capabilities. Think of it like this: instead of just talking, these new agents are actually doing things.
+
+The benchmark indicators for this evolution are evolving rapidly. The BFCL v4 benchmark, which focuses on the accuracy of function calls, is a prime example. We're seeing a shift towards models that can not only call functions but also understand dependencies and execute them in a logical order, essentially mimicking human workflows. And then there's SWE-bench Pro, which evaluates an AI's ability to solve real-world software development issues. The focus here is on expanding beyond just Python, into other languages like Go or TypeScript, and even automating the entire testing and debugging process.
+
+Finally, we're seeing a shift towards more complex tasks that require human-like planning. This is where models like WebArena and GAIA come in. These platforms are designed to evaluate an AI's ability to follow complex instructions and complete tasks involving browser interactions. The trend here is clear: the ability to plan and execute actions based on human-like input is becoming a key differentiator for these new agents.
+
+Now, let's delve into the open-source landscape of agent-based AI. We are seeing an exciting rise in models that rival proprietary ones like Claude Mythos and GPT-5.3. These open-source models are pushing boundaries with their ability to handle complex tasks, all while remaining accessible for researchers and developers alike.
+
+Qwen3-235B, DeepSeek-V4 Pro, GLM-5 / 4.5, and Llama-4 Agent Edition are just a few examples of these powerful open-source models. They demonstrate the rapid progress in AI development and offer exciting opportunities for innovation.
+
+But it's not just about individual models; the field is also witnessing a revolution in video generation. We're seeing a shift from simply generating pixels to creating truly creative agents capable of producing high-quality videos.
+
+The key here is "agentic" - meaning these AI systems are no longer just passive viewers, but active participants, able to understand and respond to prompts. This has been made possible by advancements in video generation techniques like Flow Matching & DiT, which allows for the accurate reproduction of even subtle facial expressions.
+
+And then there's the "Agentic Kernel Synthesis" technology developed by HeyGen. This breakthrough enables these AI agents to synthesize custom kernels in real-time, leading to faster and more efficient responses. This is a game changer, as it allows for highly responsive video generation with minimal latency.
+
+The future of this technology is exciting! We're seeing a shift towards "decision-centric" evaluation, where the focus isn't just on what an AI says, but rather how effectively it makes decisions and achieves desired outcomes. This means we need to develop new tools and techniques for evaluating and optimizing these agents.
+
+We also expect a greater emphasis on "harness engineering," which involves designing and implementing code layers that allow us to control and monitor these complex models. This will be crucial as the field of agent-based AI continues to evolve.
+
+Finally, we're moving towards a future where multiple specialized agents work together in a collaborative manner. Think of it like a team of experts working together on a project. We'll see more of this kind of multi-agent collaboration, with models like Qwen3 and DeepSeek-V4 playing distinct roles to achieve optimal results.
+
+This is just the beginning of a revolution in AI technology. The coming years promise to be even more exciting as we witness the emergence of truly intelligent agents that can solve complex problems and automate tasks previously thought impossible.
+
+
+And that's our report for tonight. I'm [your name], reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/script_long.rag_ingested b/docs/AI_Agents_and_Logic/script_long.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..bd7996991472d1521bd4b0ce50a7cb73a90f7b9c
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/script_long.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "08450b5bebe609da", "ingested_at": "2026-06-30T08:26:00.918749", "content_hash": "08450b5bebe609da"}
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/script_long_clean.txt b/docs/AI_Agents_and_Logic/script_long_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..71872d3d046c0939fb01a17feabeab20feeddff3
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/script_long_clean.txt
@@ -0,0 +1,29 @@
+Good evening, and welcome back. Tonight's report focuses on the rapidly evolving field of agent-based AI, specifically how these intelligent systems are transforming our technological landscape.
+
+We're in a period of significant change, where AI is moving beyond simply mimicking human conversation to truly understanding and executing complex tasks. This shift has been driven by several key developments:
+
+First, we're seeing a transition from evaluating AI models based on their conversational fluency to their practical execution capabilities. Think of it like this: instead of just talking, these new agents are actually doing things.
+
+The benchmark indicators for this evolution are evolving rapidly. The BFCL v4 benchmark, which focuses on the accuracy of function calls, is a prime example. We're seeing a shift towards models that can not only call functions but also understand dependencies and execute them in a logical order, essentially mimicking human workflows. And then there's SWE-bench Pro, which evaluates an AI's ability to solve real-world software development issues. The focus here is on expanding beyond just Python, into other languages like Go or TypeScript, and even automating the entire testing and debugging process.
+
+Finally, we're seeing a shift towards more complex tasks that require human-like planning. This is where models like WebArena and GAIA come in. These platforms are designed to evaluate an AI's ability to follow complex instructions and complete tasks involving browser interactions. The trend here is clear: the ability to plan and execute actions based on human-like input is becoming a key differentiator for these new agents.
+
+Now, let's delve into the open-source landscape of agent-based AI. We are seeing an exciting rise in models that rival proprietary ones like Claude Mythos and GPT-5.3. These open-source models are pushing boundaries with their ability to handle complex tasks, all while remaining accessible for researchers and developers alike.
+
+Qwen3-235B, DeepSeek-V4 Pro, GLM-5 / 4.5, and Llama-4 Agent Edition are just a few examples of these powerful open-source models. They demonstrate the rapid progress in AI development and offer exciting opportunities for innovation.
+
+But it's not just about individual models; the field is also witnessing a revolution in video generation. We're seeing a shift from simply generating pixels to creating truly creative agents capable of producing high-quality videos.
+
+The key here is "agentic" - meaning these AI systems are no longer just passive viewers, but active participants, able to understand and respond to prompts. This has been made possible by advancements in video generation techniques like Flow Matching & DiT, which allows for the accurate reproduction of even subtle facial expressions.
+
+And then there's the "Agentic Kernel Synthesis" technology developed by HeyGen. This breakthrough enables these AI agents to synthesize custom kernels in real-time, leading to faster and more efficient responses. This is a game changer, as it allows for highly responsive video generation with minimal latency.
+
+The future of this technology is exciting! We're seeing a shift towards "decision-centric" evaluation, where the focus isn't just on what an AI says, but rather how effectively it makes decisions and achieves desired outcomes. This means we need to develop new tools and techniques for evaluating and optimizing these agents.
+
+We also expect a greater emphasis on "harness engineering," which involves designing and implementing code layers that allow us to control and monitor these complex models. This will be crucial as the field of agent-based AI continues to evolve.
+
+Finally, we're moving towards a future where multiple specialized agents work together in a collaborative manner. Think of it like a team of experts working together on a project. We'll see more of this kind of multi-agent collaboration, with models like Qwen3 and DeepSeek-V4 playing distinct roles to achieve optimal results.
+
+This is just the beginning of a revolution in AI technology. The coming years promise to be even more exciting as we witness the emergence of truly intelligent agents that can solve complex problems and automate tasks previously thought impossible.
+
+And that's our report for tonight. I'm , reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/script_middle.md b/docs/AI_Agents_and_Logic/script_middle.md
new file mode 100644
index 0000000000000000000000000000000000000000..76a51d8102aced878012080ca7178e6a90eeb0d6
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/script_middle.md
@@ -0,0 +1,19 @@
+Good evening. Tonight, we're diving into the world of agent AI, a rapidly evolving field with major implications for how we interact with technology. The landscape is changing fast, and this report gives us a glimpse into what to expect in the coming years.
+
+One key trend is the shift from conversational fluency to practical performance. AI models are now being evaluated on their ability to actually accomplish tasks, not just talk. This means looking at things like how well they can solve real-world problems using software tools or even how effectively they can plan and execute complex actions.
+
+We're seeing this reflected in the development of new benchmarks for AI models. For example, the BFCL v4 benchmark focuses on a model's ability to use multiple tools in sequence, while SWE-bench Pro evaluates its performance in solving real-world software development issues. And then there's WebArena and GAIA, which measure how well an AI can complete tasks involving browser interaction.
+
+Now, let's talk about open-source models. These are becoming increasingly important as we see the rise of more specialized agent AI models.
+We've compiled a list of some standout models that are pushing the boundaries of what's possible: Qwen3-235B, DeepSeek-V4 Pro, GLM-5/4.5, and Llama-4 Agent Edition. Each model has its unique strengths, but they all share a common goal: to become more autonomous and capable in real-world scenarios.
+
+The video generation field is also seeing exciting advancements. We're moving beyond simply generating pixels to creating truly creative agents who can produce high-quality videos. This is driven by breakthroughs like the Avatar V architecture, which allows for finer control over facial expressions and body language.
+
+And then there's the shift in how video content is created. Instead of just producing raw footage, we now have a system where an AI agent first creates a "blueprint" or outline of what needs to be shown, and then users can refine it through dialogue. This approach allows for more targeted and personalized video production.
+
+Looking ahead, the future of agent AI is all about decision-making. The focus will shift from simply understanding what was said to analyzing which decisions were most effective. We'll also see a greater emphasis on "harness engineering," a crucial skill that involves controlling and monitoring models in real-time.
+
+Finally, we can expect to see more collaboration between multiple AI agents working together. Think of it like having different specialists working together on a project - each with their own expertise and strengths. This will lead to more efficient and powerful solutions for complex problems.
+
+
+This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/script_middle.rag_ingested b/docs/AI_Agents_and_Logic/script_middle.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..621593c60d6ba699b226fc460e97adc4d7dafeba
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/script_middle.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "af797480d09ce4a1", "ingested_at": "2026-06-30T08:26:01.450064", "content_hash": "af797480d09ce4a1"}
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/script_middle_clean.txt b/docs/AI_Agents_and_Logic/script_middle_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..055167c4dfcb6391e13c9092052a5a94f874efd2
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/script_middle_clean.txt
@@ -0,0 +1,19 @@
+Good evening. Tonight, we're diving into the world of agent AI, a rapidly evolving field with major implications for how we interact with technology. The landscape is changing fast, and this report gives us a glimpse into what to expect in the coming years.
+
+One key trend is the shift from conversational fluency to practical performance. AI models are now being evaluated on their ability to actually accomplish tasks, not just talk. This means looking at things like how well they can solve real-world problems using software tools or even how effectively they can plan and execute complex actions.
+
+We're seeing this reflected in the development of new benchmarks for AI models. For example, the BFCL v4 benchmark focuses on a model's ability to use multiple tools in sequence, while SWE-bench Pro evaluates its performance in solving real-world software development issues. And then there's WebArena and GAIA, which measure how well an AI can complete tasks involving browser interaction.
+
+Now, let's talk about open-source models. These are becoming increasingly important as we see the rise of more specialized agent AI models.
+
+We've compiled a list of some standout models that are pushing the boundaries of what's possible: Qwen3-235B, DeepSeek-V4 Pro, GLM-5/4.5, and Llama-4 Agent Edition. Each model has its unique strengths, but they all share a common goal: to become more autonomous and capable in real-world scenarios.
+
+The video generation field is also seeing exciting advancements. We're moving beyond simply generating pixels to creating truly creative agents who can produce high-quality videos. This is driven by breakthroughs like the Avatar V architecture, which allows for finer control over facial expressions and body language.
+
+And then there's the shift in how video content is created. Instead of just producing raw footage, we now have a system where an AI agent first creates a "blueprint" or outline of what needs to be shown, and then users can refine it through dialogue. This approach allows for more targeted and personalized video production.
+
+Looking ahead, the future of agent AI is all about decision-making. The focus will shift from simply understanding what was said to analyzing which decisions were most effective. We'll also see a greater emphasis on "harness engineering," a crucial skill that involves controlling and monitoring models in real-time.
+
+Finally, we can expect to see more collaboration between multiple AI agents working together. Think of it like having different specialists working together on a project - each with their own expertise and strengths. This will lead to more efficient and powerful solutions for complex problems.
+
+This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/script_short.md b/docs/AI_Agents_and_Logic/script_short.md
new file mode 100644
index 0000000000000000000000000000000000000000..13007736583d29faf6610cbfa061c8790d469906
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/script_short.md
@@ -0,0 +1,12 @@
+"Good evening. The world of AI is changing fast, and the future of 'Agent' AI is looking incredibly exciting. We're seeing a shift from just "talking," to actually doing things!
+
+New benchmarks are emerging that measure an agent's ability to solve real-world problems, like writing code or even planning complex tasks. Think of it as a new level of intelligence. Models like Qwen3 and DeepSeek are already showing incredible results in these areas. And the video generation game is changing too!
+
+HeyGen Avatar V is pushing boundaries with its ability to create lifelike videos. It's not just about generating pixels anymore, but actually understanding what makes a scene work. This means more realistic and creative content.
+
+The future of AI looks like it will be all about making decisions. We're moving from "what was said" to "what decision was the best." And this is going to require new skills in engineering and collaboration.
+
+So, what does this mean for you? It means we're on the cusp of a revolution in how AI interacts with us. Get ready for a future where AI truly helps us solve problems and create amazing things!"
+
+
+This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/script_short.rag_ingested b/docs/AI_Agents_and_Logic/script_short.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..0f082554253445f1ced937b708bae624ad1bac65
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/script_short.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "842a158f20506355", "ingested_at": "2026-06-30T08:26:01.777537", "content_hash": "842a158f20506355"}
\ No newline at end of file
diff --git a/docs/AI_Agents_and_Logic/script_short_clean.txt b/docs/AI_Agents_and_Logic/script_short_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..cb5a2a21da8eac4193f4d90cc7f01ae4d3a45d11
--- /dev/null
+++ b/docs/AI_Agents_and_Logic/script_short_clean.txt
@@ -0,0 +1,11 @@
+"Good evening. The world of AI is changing fast, and the future of 'Agent' AI is looking incredibly exciting. We're seeing a shift from just "talking," to actually doing things!
+
+New benchmarks are emerging that measure an agent's ability to solve real-world problems, like writing code or even planning complex tasks. Think of it as a new level of intelligence. Models like Qwen3 and DeepSeek are already showing incredible results in these areas. And the video generation game is changing too!
+
+HeyGen Avatar V is pushing boundaries with its ability to create lifelike videos. It's not just about generating pixels anymore, but actually understanding what makes a scene work. This means more realistic and creative content.
+
+The future of AI looks like it will be all about making decisions. We're moving from "what was said" to "what decision was the best." And this is going to require new skills in engineering and collaboration.
+
+So, what does this mean for you? It means we're on the cusp of a revolution in how AI interacts with us. Get ready for a future where AI truly helps us solve problems and create amazing things!"
+
+This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/VFI_Research_Report.md b/docs/AI_Computer_Vision/VFI_Research_Report.md
new file mode 100644
index 0000000000000000000000000000000000000000..74dbb3f34b00640bd2b2fd664219e63944bb443b
--- /dev/null
+++ b/docs/AI_Computer_Vision/VFI_Research_Report.md
@@ -0,0 +1,163 @@
+# Video Frame Interpolation (VFI) 最新研究トレンドレポート (2025–2026)
+
+本レポートは、2025年から2026年にかけて発表された最新の学術論文に基づき、コンピュータビジョン分野における**ビデオフレーム補間(Video Frame Interpolation: VFI)**の最新研究動向、主要な技術的アプローチ、および画期的な成果を体系的にまとめたドキュメントです。
+
+本リサーチで特定された代表的な論文5本は、本ディレクトリ下の `papers/` フォルダにPDF形式でローカルにダウンロードされています。
+
+---
+
+## 1. 総合概要(Executive Summary)
+
+ビデオフレーム補間(VFI)は、連続するビデオフレーム間の中間フレームを生成し、フレームレートを向上させる(例:通常動画からハイスピードスローモーション動画への変換)ための重要技術です。
+
+2025〜2026年のトレンドを分析すると、従来の「オプティカルフローを推定し、CNNや単純なTransformerでワーピングする」手法から脱却し、以下の**4つの破壊的イノベーション**が起きています。
+
+1. **動画生成拡散モデル(Video Diffusion Models)のVFI適応**:
+ テクスチャの整合性や大局的な動きの品質を劇的に高めるため、事前に学習された強力なImage-to-Video(I2V)拡散モデルをVFIタスクへ適応・微調整する研究が急増しています。
+2. **画像編集基盤モデル(Image Editing Foundation Models)のFew-shot適応**:
+ 明示的なオプティカルフローの計算を行わず、静的画像の編集能力を持つ基盤モデルに対してLoRA等のパラメータ効率の良いチューニング(PEFT)を施すことで、極めて少ないデータ(数十〜数百サンプル)で高精度なフレーム補間を「解き放つ」アプローチが登場しました。
+3. **イベントカメラ(Event Camera)の融合による時間超解像**:
+ 超高頻度かつ非同期に輝度変化を捉えるイベントカメラのストリームを融合し、高速・複雑な非線形モーションや、ブレのある(露出時間の長い)低フレームレート映像からの極限のフレーム補間(時間超解像)が可能になりました。
+4. **人間の知覚に準拠した高効率な動き評価指標(PSNRDIV)の誕生**:
+ 従来の単純なL1/L2誤差(PSNR/SSIM)や極めて重いFloLPIPSに代わり、動きベクトルの「発散(Divergence)」を利用して動画特有の破綻領域を検知する、超高速かつ軽量な評価指標(かつ損失関数)が提案されました。
+
+---
+
+## 2. 主要論文の詳細レビュー
+
+ローカルにPDFをダウンロードした5本の最重要論文について、それぞれの提案手法、アーキテクチャ、および主要な貢献を詳細に解説します。
+
+---
+
+### 【レビュー1】AceVFI: 総合的サーベイと体系化
+* **論文名**: AceVFI: A Comprehensive Survey of Advances in Video Frame Interpolation
+* **著者**: Dahyeon Kye, Changhyun Roh, Sukhun Ko, Chanho Eom, Jihyong Oh
+* **arXiv ID**: `2506.01061` | **ローカルパス**: `papers/AceVFI.pdf`
+* **論文URL**: [https://arxiv.org/abs/2506.01061](https://arxiv.org/abs/2506.01061)
+
+#### 核心アイデアと手法
+250本以上の代表的論文を網羅し、近年のVFIの発展史を体系化した集大成サーベイです。
+VFIの核心を以下のアーキテクチャ軸および学習軸で明確に分類しています。
+
+* **アーキテクチャ分類**:
+ * **Kernel-based(カーネルベース)** / **Flow-based(フローベース)** / **Hybrid(ハイブリッド)**
+ * **Phase-based(位相ベース)** / **GAN-based(敵対的学習)**
+ * **Transformer-based(アテンションベース)**
+ * **Mamba-based(状態空間モデル)** / **Diffusion-based(拡散モデル)**
+* **学習パラダイム分類**:
+ * **CTFI(Center-Time Frame Interpolation)**: $t=0.5$ など、前後のちょうど中間の時間のみを生成する最もシンプルな形態。
+ * **ATFI(Arbitrary-Time Frame Interpolation)**: $t \in (0, 1)$ の任意の時間パラメータを入力として受け取り、任意のフレームレート(例:120FPSやハイスピードスロー)を動的に出力する汎用性の高いアプローチ。
+
+#### 主要な貢献と結果
+* 大局的かつ急激な動き、オクルージョン(物体の隠れ)、照明の動的変化、および非線形モーションという、従来のVFIが直面してきた主要課題に対する各モデルのアプローチを比較。
+* 標準的な評価データセット(UCF101、Vimeo-90K、GOPROなど)、画質・時間的整合性を測るための各種損失関数と評価指標のベンチマーク結果を集約。
+* 実用的なオープンソースリソースとして、GitHub上に関連論文と実装をまとめた「Awesome-Video-Frame-Interpolation」プロジェクトをホストし、コミュニティの発展に大きく貢献しています。
+
+---
+
+### 【レビュー2】FC-VFI: 高画質・高FPSスローモーション生成
+* **論文名**: FC-VFI: Faithful and Consistent Video Frame Interpolation for High-FPS Slow Motion Video Generation
+* **著者**: Ganggui Ding, Hao Chen, Xiaogang Xu
+* **arXiv ID**: `2603.04899` | **ローカルパス**: `papers/FC_VFI.pdf`
+* **論文URL**: [https://arxiv.org/abs/2603.04899](https://arxiv.org/abs/2603.04899)
+
+#### 核心アイデアと手法
+大規模動画生成モデル(HunyuanVideo-I2Vなど)の持つ強力な動画合成プライア(事前知識)をベースとし、スローモーション動画生成に向けてVFI専用にファインチューニングした拡散ベースのフレームワークです。最大2560×1440(2.5K)の超高解像度に対応します。
+
+* **Temporal Fidelity Modulation Reference (TFMR)**:
+ 前後フレームから抽出された高解像度のテクスチャや境界情報を、中間生成段階に時間的結合(Temporal Concatenation)とモジュレーション(変調)を介して直接フィードバックします。これにより、拡散モデル特有の「細かいテクスチャがブレたり変化してしまう問題」を抑制し、視覚的一貫性を維持します。
+* **Structure-Aware Motion**:
+ オプティカルフローが完全に崩壊するような複雑な境界や隠消現実(オクルージョン)に対して、セマンティックなマッチングライン(構造ガイド)を導入。動きの構造情報をアテンション機構に融合させます。
+
+#### 主要な貢献と結果
+* **高効率性**: 双方向のフロー推定や推論といった前処理フェーズを排除し、わずか**10ステップのノイズ除去**だけで高品質な4倍/8倍補間を達成。
+* 従来手法ではゴーストが発生していた激しいスポーツ動画や複雑な煙・水のような流体シミュレーション動画において、テクスチャが破綻しない極めてリアルなハイスピードスローモーションを生成することに成功。
+
+---
+
+### 【レビュー3】Edit2Interp: 画像編集モデルからのFew-Shot転用
+* **論文名**: Edit2Interp: Adapting Image Foundation Models from Spatial Editing to Video Frame Interpolation with Few-Shot Learning
+* **著者**: Nasrin Rahimi, Mısra Yavuz, Burak Can Biner, et al.
+* **arXiv ID**: `2603.15003` | **ローカルパス**: `papers/Edit2Interp.pdf`
+* **論文URL**: [https://arxiv.org/abs/2603.15003](https://arxiv.org/abs/2603.15003)
+
+#### 核心アイデアと手法
+「ビデオフレーム補間とは、空間的な編集モデルによる制約付きの画像編集問題の一種である」というパラダイムシフトを提案した画期的なアプローチです。
+
+* **基盤モデルの有効活用**:
+ 多目的画像編集基盤モデルである **Qwen-Image-Edit** をバックボーンに採用。開始フレーム $I_0$ と終了フレーム $I_1$、および「$t=0.5$ の中間フレームを生成せよ」といった固定的なプロンプト情報をマルチモーダル入力として与えます。
+* **Few-Shot LoRA Adaptation**:
+ 動画モデルの巨大な学習コストをかけることなく、わずか**64〜256の動画サンプルセット**を用いて、LoRA(Low-Rank Adaptation)を適用するだけで、画像モデル内に眠る空間的一貫性と時間的補間能力を効率的に引き出します。
+
+#### 主要な貢献と結果
+* **オプティカルフロー不要の画期性**: フロー推定やワーピングモジュールといった複雑な物理的モーション予測を行わず、モデルが画像の意味理解(セマンティクス)に基づいて自然な中間遷移を決定します。
+* データが極端に制限された環境(学習データが非常に少ないドメイン)であっても、既存の多くのフルモデルトレーニングVFI手法と同等以上の自然なテクスチャ保持と形状変化を実現。
+
+---
+
+### 【レビュー4】PSNRDIV: 動きの発散度に基づく高効率・高速評価指標
+* **論文名**: An Efficient Quality Metric for Video Frame Interpolation Based on Motion-Field Divergence (PSNRDIV)
+* **著者**: Conall Daly, Darren Ramsook, Anil Kokaram
+* **arXiv ID**: `2510.01361` (別題先行: `2508.09078`)| **ローカルパス**: `papers/PSNRDIV.pdf`
+* **論文URL**: [https://arxiv.org/abs/2510.01361](https://arxiv.org/abs/2510.01361)
+
+#### 核心アイデアと手法
+従来のVFI評価は、静的なピクセル毎のL1/L2誤差(PSNR/SSIM)か、もしくは極めて計算コストの高い深層アテンション評価(FloLPIPS)に依存しており、リアルタイムの評価やネットワークの損失関数(Loss)への組み込みが困難でした。本論文は、映画復元技術から着想を得た軽量・高精度な新しいフルリファレンス画質評価指標 **PSNRDIV** を提案しています。
+
+* **Motion-Field Divergence Weighting**:
+ 動きベクトルフィールドの発散(Divergence)を計算することで、動画内のオクルージョン(隠れ領域)、高速な物体の境界、またはモーションの急激な不整合が発生している「破綻しやすい特異点領域」を特定。
+* この特異点領域におけるピクセルエラーに対して適応的に重みを加算し、動画における人間の主観的な不快感・違和感(アーティファクト)と直接連動したスコアを算出します。
+
+#### 主要な貢献と結果
+* **超高速かつ低メモリ**: 既存の代表的な評価指標である FloLPIPS と比較して、**2.5倍高速**に動作し、かつ**メモリ消費量を4分の1に削減**。
+* **高い主観一致度**: 人間の視覚的な評価との相関を示す指標(PLCC)において、FloLPIPS を +0.09 上回る優れた相関関係を実証。
+* 軽量であるため評価だけでなく、深層ネットワークモデルの学習段階における画期的な損失関数(Loss Function)としても直接組み込める実用性を有しています。
+
+---
+
+### 【レビュー5】EventDiff: イベントカメラを融合した超解像拡散モデル
+* **論文名**: EventDiff: A Unified and Efficient Diffusion Model Framework for Event-based Video Frame Interpolation
+* **著者**: Hanle Zheng, Xujie Han, Zegang Peng, et al.
+* **arXiv ID**: `2505.08235` | **ローカルパス**: `papers/EventDiff.pdf`
+* **論文URL**: [https://arxiv.org/abs/2505.08235](https://arxiv.org/abs/2505.08235)
+
+#### 核心アイデアと手法
+超高時間解像度で輝度変化のみを出力する「イベントカメラ」の情報と、通常のカラー画像(RGBフレーム)を融合し、激しい運動や暗所・オクルージョン下での極限のフレーム補間を拡散モデルで実現する統一フレームワークです。
+
+* **Event-Frame Hybrid AutoEncoder (HAE)**:
+ 時間的に非常に不均一なイベントデータのストリームと、静的で豊かな空間情報を持つRGBフレームを統一された特徴マップ空間に圧縮・マッピングするハイブリッドエンコーダ。
+* **Spatial-Temporal Cross Attention (STCA)**:
+ 時空間アテンションを導入することで、イベント情報をガイダンスとして拡散プロセスのノイズ除去を精緻化します。
+
+#### 主要な貢献と結果
+* 従来フローベースモデルが破綻していた「極めて高速なスポーツ動作(テニスのスマッシュなど)」や「突然照明が明滅する環境」において、破綻やブレを一切発生させずに鮮明な中間超解像フレームを出力。
+* ワーピングといった複雑なモーション補正モジュールに依存せず、すべての動きの補完を潜在表現の拡散プロセス内で完結させるため、非常に堅牢で安定した処理が可能です。
+
+---
+
+## 3. テクノロジートレンドの比較・整理
+
+最新研究から見えるVFIの代表的アプローチをマトリクスで整理します。
+
+| アプローチ | 代表論文 | モーション予測の手法 | メリット | 課題 |
+| :--- | :--- | :--- | :--- | :--- |
+| **動画生成モデル転用** | **FC-VFI**
(レビュー2) | 構造アライメントマッチングライン + I2Vプライア | 2.5K超高画質、写真のようにリアルなテクスチャ生成、10ステップの超高速推論 | モデルサイズが大きく、モバイル等のエッジデバイスには不向き |
+| **画像基盤モデルFew-shot** | **Edit2Interp**
(レビュー3) | 空間編集セマンティクスによる動的合成(フロー不要) | 数十枚のデータでチューニング可能、オプティカルフロー特有の境界の破綻が皆無 | 3次元的な幾何学的一貫性を完全に維持するのが難しいケースがある |
+| **動きの特異点評価** | **PSNRDIV**
(レビュー4) | モーションベクトルの発散分析(評価用) | 2.5倍高速、省メモリ。損失関数(Loss)としてトレーニングに組み込み可能 | 評価指標であり、これ単体では画像を生成するわけではない |
+| **イベントセンサー融合** | **EventDiff**
(レビュー5) | 非同期イベントパルスによるトラッキング案内 | 急激な輝度変化や超高速非線形モーションでも絶対に破綻しない | 特殊なイベントカメラハードウェアが必要 |
+
+---
+
+## 4. 今後の展望と展望
+
+2025–2026年の研究トレンドは、VFIが単なる「画像変形処理」から**「物理演算や世界の構造的意味理解に基づいた『新フレーム創出(Generative VFI)』」**に完全移行したことを如実に示しています。
+
+* **アニメ制作・映画修復**: 手描きアニメのコマ割り適応(Time-adaptive VFI)や、古いフィルム映像の超高精細なコマ補間(PSNRDIVの応用)への活用が進んでいます。
+* **自動運転・ロボットビジョン**: イベントカメラを用いた `TimeTracker` や `EventDiff` は、夜間や悪天候、超高速移動時でも視覚的追跡を失わない強固なナビゲーション基盤として実用段階に移りつつあります。
+* **次世代ゲーム機・ディスプレイ**: 低負荷な基盤モデルや軽量拡散ステップの組み合わせにより、コンソール機器上でのリアルタイムなフレームレートブースター(DLSSやFSRの進化系)としての応用が期待されます。
+
+---
+
+> [!WARNING]
+> **arXiv APIの利用およびライセンスに関する注意**
+> 本ドキュメントの情報、および `papers/` ディレクトリに収集されたPDF資料は、arXivの利用規約([https://info.arxiv.org/help/api/index.html](https://info.arxiv.org/help/api/index.html))に従ってAPI経由で公式に取得したものです。各論文を学術研究以外(商業的利用、再配布など)に利用する際は、PDFファイルや各arXivページに明記されている個々のライセンス(Creative Commons等)の範囲を必ず遵守してください。
diff --git a/docs/AI_Computer_Vision/VFI_Research_Report.rag_ingested b/docs/AI_Computer_Vision/VFI_Research_Report.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..f1dc3a3f8d8e437ff23b2af125d1943b49b06127
--- /dev/null
+++ b/docs/AI_Computer_Vision/VFI_Research_Report.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "6949f371b515d0e7", "ingested_at": "2026-06-30T08:49:28.206798", "content_hash": "6949f371b515d0e7"}
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405.json b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405.json
new file mode 100644
index 0000000000000000000000000000000000000000..8f23e7cef1c36bd5cdd8cdb7f12e4d0fe0ab3e65
--- /dev/null
+++ b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405.json
@@ -0,0 +1,5 @@
+{
+ "importance": 10,
+ "urgency": 10,
+ "tier": 1
+}
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405.md b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405.md
new file mode 100644
index 0000000000000000000000000000000000000000..dfc9360268c0555ac4c32ca68de3457edb6b9095
--- /dev/null
+++ b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405.md
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d012338d44840196d55d21b96292ea19a0d866ea962fd118d33df572373b4ea
+size 10953161
diff --git a/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_long.md b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_long.md
new file mode 100644
index 0000000000000000000000000000000000000000..90246e2cb9ac7ff12e24d7ec575133104391267b
--- /dev/null
+++ b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_long.md
@@ -0,0 +1,33 @@
+This looks like a list of research papers submitted to arXiv in April 2026, focusing on various aspects of computer science and AI. Let's break down the information you provided:
+
+**Understanding the Structure:**
+
+* **arXiv:** This is a platform for sharing pre-prints of scientific papers, including those in computer science, mathematics, physics, and other fields.
+* **ID:** Each paper has a unique identifier (like "2604.02290") that helps researchers find and track it.
+* **Submitted:** Indicates when the paper was submitted to arXiv.
+* **Authors:** Lists the names of the authors of the research paper.
+* **abs:** Provides a short abstract summarizing the paper's main findings. This is often the first thing people look at to get an idea of what the paper is about.
+* **PDF:** Links to the full, downloadable version of the paper (usually in PDF format).
+
+**Key Themes and Topics:**
+
+The papers cover a wide range of topics within computer science:
+
+* **Medical Imaging:** Papers like "AdamFlow" focus on using AI for surface registration in medical imaging.
+* **3D Modeling & Generation:** "Omni123" explores how to generate 3D models from text and images, even with limited data.
+* **Autonomous Driving:** Papers like "Deep Neural Network Based Roadwork Detection" aim to improve self-driving cars' ability to detect road features.
+* **AI Agents:** Research on "Novel Memory Forgetting Techniques" focuses on how AI agents can learn and adapt without forgetting too much information.
+* **Vision Transformers (ViT):** Several papers use ViT models, a type of deep learning architecture for image processing, like "ViT-Explainer."
+* **Image Generation:** Papers like "Reflection Generation" explore using diffusion models to generate realistic images with reflections.
+* **Denoising and Image Reconstruction:** Research on "DenOiS" aims to improve ultrasound imaging by removing noise and artifacts.
+
+
+**How to Find More Papers:**
+
+1. **arXiv Website:** You can visit the arXiv website (https://arxiv.org/) and search for specific topics or authors.
+2. **Specific Research Areas:** If you're interested in a particular area of computer science, like medical imaging, autonomous driving, or AI ethics, you can find relevant papers on those websites.
+
+
+**Important Note:** The information provided is based on the content of the list you shared.
+
+Let me know if you have any other questions about these papers!
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_long_clean.txt b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_long_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..f3bdb5c29e6082862958f2eab2a2a8817a62dbac
--- /dev/null
+++ b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_long_clean.txt
@@ -0,0 +1,35 @@
+This looks like a list of research papers submitted to arXiv in April 2026, focusing on various aspects of computer science and AI. Let's break down the information you provided:
+
+arXiv: This is a platform for sharing pre-prints of scientific papers, including those in computer science, mathematics, physics, and other fields.
+
+ID: Each paper has a unique identifier that helps researchers find and track it.
+
+Submitted: Indicates when the paper was submitted to arXiv.
+
+Authors: Lists the names of the authors of the research paper.
+
+abs: Provides a short abstract summarizing the paper's main findings. This is often the first thing people look at to get an idea of what the paper is about.
+
+PDF: Links to the full, downloadable version of the paper .
+
+Medical Imaging: Papers like "AdamFlow" focus on using AI for surface registration in medical imaging.
+
+3D Modeling & Generation: "Omni123" explores how to generate 3D models from text and images, even with limited data.
+
+Autonomous Driving: Papers like "Deep Neural Network Based Roadwork Detection" aim to improve self-driving cars' ability to detect road features.
+
+AI Agents: Research on "Novel Memory Forgetting Techniques" focuses on how AI agents can learn and adapt without forgetting too much information.
+
+Vision Transformers : Several papers use ViT models, a type of deep learning architecture for image processing, like "ViT-Explainer."
+
+Image Generation: Papers like "Reflection Generation" explore using diffusion models to generate realistic images with reflections.
+
+Denoising and Image Reconstruction: Research on "DenOiS" aims to improve ultrasound imaging by removing noise and artifacts.
+
+1. arXiv Website: You can visit the arXiv website and search for specific topics or authors.
+
+2. Specific Research Areas: If you're interested in a particular area of computer science, like medical imaging, autonomous driving, or AI ethics, you can find relevant papers on those websites.
+
+The information provided is based on the content of the list you shared.
+
+Let me know if you have any other questions about these papers!
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_middle.md b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_middle.md
new file mode 100644
index 0000000000000000000000000000000000000000..ee8c0dd8370b953042483d2b527ea2a7ae7c2ce5
--- /dev/null
+++ b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_middle.md
@@ -0,0 +1,6 @@
+Good evening, and we're continuing our coverage of the latest advancements in AI research. A new partnership has been announced between Reflection AI and SpaceX, with a significant financial commitment involved. This deal will see Reflection AI paying $150 million per month for access to Nvidia's cutting-edge GB300 AI chips and supporting hardware within SpaceX's Colossus 2 data center in Tennessee.
+
+This agreement marks a major step forward in the field of Artificial Intelligence, as it allows Reflection AI to leverage these powerful tools for their own research and development. The deal is expected to significantly boost Reflection AI's capabilities, particularly in areas like machine learning and deep learning. The Colossus 2 data center is known for its advanced infrastructure and is well-equipped to handle the demands of this partnership.
+
+We'll continue to follow this story as it develops, and we'll be sure to provide updates on how this collaboration impacts the AI landscape in the coming months.
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_middle_clean.txt b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_middle_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..3273cad41dbd6b6a8eb84af2c964b8b4bb7daf9d
--- /dev/null
+++ b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_middle_clean.txt
@@ -0,0 +1,7 @@
+Good evening, and we're continuing our coverage of the latest advancements in AI research. A new partnership has been announced between Reflection AI and SpaceX, with a significant financial commitment involved. This deal will see Reflection AI paying $150 million per month for access to Nvidia's cutting-edge GB300 AI chips and supporting hardware within SpaceX's Colossus 2 data center in Tennessee.
+
+This agreement marks a major step forward in the field of Artificial Intelligence, as it allows Reflection AI to leverage these powerful tools for their own research and development. The deal is expected to significantly boost Reflection AI's capabilities, particularly in areas like machine learning and deep learning. The Colossus 2 data center is known for its advanced infrastructure and is well-equipped to handle the demands of this partnership.
+
+We'll continue to follow this story as it develops, and we'll be sure to provide updates on how this collaboration impacts the AI landscape in the coming months.
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_short.md b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_short.md
new file mode 100644
index 0000000000000000000000000000000000000000..b7a572625c2c4258b87e905085bad050b6f3f8f0
--- /dev/null
+++ b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_short.md
@@ -0,0 +1,4 @@
+SpaceX just struck a major deal with Reflection AI! They're going to use SpaceX's Colossus 2 data center for super powerful AI chips, worth $150 million a month. This will help them power up their cutting-edge projects. We're talking about the future of AI here folks - this is huge!
+
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_short_clean.txt b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_short_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..ef126e9e569b73f859ad3ceafd678de7f11bfc0e
--- /dev/null
+++ b/docs/AI_Computer_Vision/arxiv-cs-cv-20250405-20260405_script_short_clean.txt
@@ -0,0 +1,3 @@
+SpaceX just struck a major deal with Reflection AI! They're going to use SpaceX's Colossus 2 data center for super powerful AI chips, worth $150 million a month. This will help them power up their cutting-edge projects. We're talking about the future of AI here folks - this is huge!
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/arxiv-cs-cv-2026-04-05.md b/docs/AI_Computer_Vision/arxiv-cs-cv-2026-04-05.md
new file mode 100644
index 0000000000000000000000000000000000000000..4b6c686bd3761dfa8f31b91d237a24374e3b8824
--- /dev/null
+++ b/docs/AI_Computer_Vision/arxiv-cs-cv-2026-04-05.md
@@ -0,0 +1,363 @@
+# arXiv cs.CV — Computer Vision and Pattern Recognition (recent)
+
+*Category:* `cs.CV` · *sort:* submittedDate descending
+
+*Feed `updated`:* 2026-04-05T06:29:11Z
+
+*Fetched at (UTC):* 2026-04-05T06:29:10Z
+
+*Machine-readable twin:* `arxiv-cs-cv-2026-04-05.json`
+
+---
+
+## 1. EventHub: Data Factory for Generalizable Event-Based Stereo Networks without Active Sensors
+
+- **arXiv ID:** `2604.02331`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Luca Bartolomei, Fabio Tosi, Matteo Poggi, Stefano Mattoccia, Guillermo Gallego
+- **Abstract page:** [https://arxiv.org/abs/2604.02331](https://arxiv.org/abs/2604.02331)
+- **PDF:** [https://arxiv.org/pdf/2604.02331v1](https://arxiv.org/pdf/2604.02331v1)
+
+### Abstract
+
+We propose EventHub, a novel framework for training deep-event stereo networks without ground truth annotations from costly active sensors, relying instead on standard color images. From these images, we derive either proxy annotations and proxy events through state-of-the-art novel view synthesis techniques, or simply proxy annotations when images are already paired with event data. Using the training set generated by our data factory, we repurpose state-of-the-art stereo models from RGB literature to process event data, obtaining new event stereo models with unprecedented generalization capabilities. Experiments on widely used event stereo datasets support the effectiveness of EventHub and show how the same data distillation mechanism can improve the accuracy of RGB stereo foundation models in challenging conditions such as nighttime scenes.
+
+---
+
+## 2. ActionParty: Multi-Subject Action Binding in Generative Video Games
+
+- **arXiv ID:** `2604.02330`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `cs.AI`, `cs.LG`
+- **Published:** 2026-04-02
+- **Authors:** Alexander Pondaven, Ziyi Wu, Igor Gilitschenski, Philip Torr, Sergey Tulyakov, Fabio Pizzati, Aliaksandr Siarohin
+- **Abstract page:** [https://arxiv.org/abs/2604.02330](https://arxiv.org/abs/2604.02330)
+- **PDF:** [https://arxiv.org/pdf/2604.02330v1](https://arxiv.org/pdf/2604.02330v1)
+
+### Abstract
+
+Recent advances in video diffusion have enabled the development of "world models" capable of simulating interactive environments. However, these models are largely restricted to single-agent settings, failing to control multiple agents simultaneously in a scene. In this work, we tackle a fundamental issue of action binding in existing video diffusion models, which struggle to associate specific actions with their corresponding subjects. For this purpose, we propose ActionParty, an action controllable multi-subject world model for generative video games. It introduces subject state tokens, i.e. latent variables that persistently capture the state of each subject in the scene. By jointly modeling state tokens and video latents with a spatial biasing mechanism, we disentangle global video frame rendering from individual action-controlled subject updates. We evaluate ActionParty on the Melting Pot benchmark, demonstrating the first video world model capable of controlling up to seven players simultaneously across 46 diverse environments. Our results show significant improvements in action-following accuracy and identity consistency, while enabling robust autoregressive tracking of subjects through complex interactions.
+
+---
+
+## 3. Generative World Renderer
+
+- **arXiv ID:** `2604.02329`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Zheng-Hui Huang, Zhixiang Wang, Jiaming Tan, Ruihan Yu, Yidan Zhang, Bo Zheng, Yu-Lun Liu, Yung-Yu Chuang, Kaipeng Zhang
+- **Abstract page:** [https://arxiv.org/abs/2604.02329](https://arxiv.org/abs/2604.02329)
+- **PDF:** [https://arxiv.org/pdf/2604.02329v1](https://arxiv.org/pdf/2604.02329v1)
+
+### Abstract
+
+Scaling generative inverse and forward rendering to real-world scenarios is bottlenecked by the limited realism and temporal coherence of existing synthetic datasets. To bridge this persistent domain gap, we introduce a large-scale, dynamic dataset curated from visually complex AAA games. Using a novel dual-screen stitched capture method, we extracted 4M continuous frames (720p/30 FPS) of synchronized RGB and five G-buffer channels across diverse scenes, visual effects, and environments, including adverse weather and motion-blur variants. This dataset uniquely advances bidirectional rendering: enabling robust in-the-wild geometry and material decomposition, and facilitating high-fidelity G-buffer-guided video generation. Furthermore, to evaluate the real-world performance of inverse rendering without ground truth, we propose a novel VLM-based assessment protocol measuring semantic, spatial, and temporal consistency. Experiments demonstrate that inverse renderers fine-tuned on our data achieve superior cross-dataset generalization and controllable generation, while our VLM evaluation strongly correlates with human judgment. Combined with our toolkit, our forward renderer enables users to edit styles of AAA games from G-buffers using text prompts.
+
+---
+
+## 4. Modulate-and-Map: Crossmodal Feature Mapping with Cross-View Modulation for 3D Anomaly Detection
+
+- **arXiv ID:** `2604.02328`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Alex Costanzino, Pierluigi Zama Ramirez, Giuseppe Lisanti, Luigi Di Stefano
+- **Abstract page:** [https://arxiv.org/abs/2604.02328](https://arxiv.org/abs/2604.02328)
+- **PDF:** [https://arxiv.org/pdf/2604.02328v1](https://arxiv.org/pdf/2604.02328v1)
+
+### Abstract
+
+We present ModMap, a natively multiview and multimodal framework for 3D anomaly detection and segmentation. Unlike existing methods that process views independently, our method draws inspiration from the crossmodal feature mapping paradigm to learn to map features across both modalities and views, while explicitly modelling view-dependent relationships through feature-wise modulation. We introduce a cross-view training strategy that leverages all possible view combinations, enabling effective anomaly scoring through multiview ensembling and aggregation. To process high-resolution 3D data, we train and publicly release a foundational depth encoder tailored to industrial datasets. Experiments on SiM3D, a recent benchmark that introduces the first multiview and multimodal setup for 3D anomaly detection and segmentation, demonstrate that ModMap attains state-of-the-art performance by surpassing previous methods by wide margins.
+
+---
+
+## 5. Steerable Visual Representations
+
+- **arXiv ID:** `2604.02327`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `cs.AI`
+- **Published:** 2026-04-02
+- **Authors:** Jona Ruthardt, Manu Gaur, Deva Ramanan, Makarand Tapaswi, Yuki M. Asano
+- **Abstract page:** [https://arxiv.org/abs/2604.02327](https://arxiv.org/abs/2604.02327)
+- **PDF:** [https://arxiv.org/pdf/2604.02327v1](https://arxiv.org/pdf/2604.02327v1)
+
+### Abstract
+
+Pretrained Vision Transformers (ViTs) such as DINOv2 and MAE provide generic image features that can be applied to a variety of downstream tasks such as retrieval, classification, and segmentation. However, such representations tend to focus on the most salient visual cues in the image, with no way to direct them toward less prominent concepts of interest. In contrast, Multimodal LLMs can be guided with textual prompts, but the resulting representations tend to be language-centric and lose their effectiveness for generic visual tasks. To address this, we introduce Steerable Visual Representations, a new class of visual representations, whose global and local features can be steered with natural language. While most vision-language models (e.g., CLIP) fuse text with visual features after encoding (late fusion), we inject text directly into the layers of the visual encoder (early fusion) via lightweight cross-attention. We introduce benchmarks for measuring representational steerability, and demonstrate that our steerable visual features can focus on any desired objects in an image while preserving the underlying representation quality. Our method also matches or outperforms dedicated approaches on anomaly detection and personalized object discrimination, exhibiting zero-shot generalization to out-of-distribution tasks.
+
+---
+
+## 6. Beyond Referring Expressions: Scenario Comprehension Visual Grounding
+
+- **arXiv ID:** `2604.02323`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Ruozhen He, Nisarg A. Shah, Qihua Dong, Zilin Xiao, Jaywon Koo, Vicente Ordonez
+- **Abstract page:** [https://arxiv.org/abs/2604.02323](https://arxiv.org/abs/2604.02323)
+- **PDF:** [https://arxiv.org/pdf/2604.02323v1](https://arxiv.org/pdf/2604.02323v1)
+
+### Abstract
+
+Existing visual grounding benchmarks primarily evaluate alignment between image regions and literal referring expressions, where models can often succeed by matching a prominent named category. We explore a complementary and more challenging setting of scenario-based visual grounding, where the target must be inferred from roles, intentions, and relational context rather than explicit naming. We introduce Referring Scenario Comprehension (RSC), a benchmark designed for this setting. The queries in this benchmark are paragraph-length texts that describe object roles, user goals, and contextual cues, including deliberate references to distractor objects that often require deep understanding to resolve. Each instance is annotated with interpretable difficulty tags for uniqueness, clutter, size, overlap, and position which expose distinct failure modes and support fine-grained analysis. RSC contains approximately 31k training examples, 4k in-domain test examples, and a 3k out-of-distribution split with unseen object categories. We further propose ScenGround, a curriculum reasoning method serving as a reference point for this setting, combining supervised warm-starting with difficulty-aware reinforcement learning. Experiments show that scenario-based queries expose systematic failures in current models that standard benchmarks do not reveal, and that curriculum training improves performance on challenging slices and transfers to standard benchmarks.
+
+---
+
+## 7. Large-scale Codec Avatars: The Unreasonable Effectiveness of Large-scale Avatar Pretraining
+
+- **arXiv ID:** `2604.02320`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `cs.GR`
+- **Published:** 2026-04-02
+- **Authors:** Junxuan Li, Rawal Khirodkar, Chengan He, Zhongshi Jiang, Giljoo Nam, Lingchen Yang, Jihyun Lee, Egor Zakharov, Zhaoen Su, Rinat Abdrashitov, Yuan Dong, Julieta Martinez, Kai Li, Qingyang Tan, Takaaki Shiratori, Matthew Hu, Peihong Guo, Xuhua Huang, Ariyan Zarei, Marco Pesavento, Yichen Xu, He Wen, Teng Deng, Wyatt Borsos, Anjali Thakrar, Jean-Charles Bazin, Carsten Stoll, Ginés Hidalgo, James Booth, Lucy Wang, Xiaowen Ma, Yu Rong, Sairanjith Thalanki, Chen Cao, Christian Häne, Abhishek Kar, Sofien Bouaziz, Jason Saragih, Yaser Sheikh, Shunsuke Saito
+- **Abstract page:** [https://arxiv.org/abs/2604.02320](https://arxiv.org/abs/2604.02320)
+- **PDF:** [https://arxiv.org/pdf/2604.02320v1](https://arxiv.org/pdf/2604.02320v1)
+
+### Abstract
+
+High-quality 3D avatar modeling faces a critical trade-off between fidelity and generalization. On the one hand, multi-view studio data enables high-fidelity modeling of humans with precise control over expressions and poses, but it struggles to generalize to real-world data due to limited scale and the domain gap between the studio environment and the real world. On the other hand, recent large-scale avatar models trained on millions of in-the-wild samples show promise for generalization across a wide range of identities, yet the resulting avatars are often of low-quality due to inherent 3D ambiguities. To address this, we present Large-Scale Codec Avatars (LCA), a high-fidelity, full-body 3D avatar model that generalizes to world-scale populations in a feedforward manner, enabling efficient inference. Inspired by the success of large language models and vision foundation models, we present, for the first time, a pre/post-training paradigm for 3D avatar modeling at scale: we pretrain on 1M in-the-wild videos to learn broad priors over appearance and geometry, then post-train on high-quality curated data to enhance expressivity and fidelity. LCA generalizes across hair styles, clothing, and demographics while providing precise, fine-grained facial expressions and finger-level articulation control, with strong identity preservation. Notably, we observe emergent generalization to relightability and loose garment support to unconstrained inputs, and zero-shot robustness to stylized imagery, despite the absence of direct supervision.
+
+---
+
+## 8. Stop Wandering: Efficient Vision-Language Navigation via Metacognitive Reasoning
+
+- **arXiv ID:** `2604.02318`
+- **Primary category:** `cs.RO`
+- **Categories:** `cs.RO`, `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Xueying Li, Feng Lyu, Hao Wu, Mingliu Liu, Jia-Nan Liu, Guozi Liu
+- **Abstract page:** [https://arxiv.org/abs/2604.02318](https://arxiv.org/abs/2604.02318)
+- **PDF:** [https://arxiv.org/pdf/2604.02318v1](https://arxiv.org/pdf/2604.02318v1)
+
+### Abstract
+
+Training-free Vision-Language Navigation (VLN) agents powered by foundation models can follow instructions and explore 3D environments. However, existing approaches rely on greedy frontier selection and passive spatial memory, leading to inefficient behaviors such as local oscillation and redundant revisiting. We argue that this stems from a lack of metacognitive capabilities: the agent cannot monitor its exploration progress, diagnose strategy failures, or adapt accordingly. To address this, we propose MetaNav, a metacognitive navigation agent integrating spatial memory, history-aware planning, and reflective correction. Spatial memory builds a persistent 3D semantic map. History-aware planning penalizes revisiting to improve efficiency. Reflective correction detects stagnation and uses an LLM to generate corrective rules that guide future frontier selection. Experiments on GOAT-Bench, HM3D-OVON, and A-EQA show that MetaNav achieves state-of-the-art performance while reducing VLM queries by 20.7%, demonstrating that metacognitive reasoning significantly improves robustness and efficiency.
+
+---
+
+## 9. A Simple Baseline for Streaming Video Understanding
+
+- **arXiv ID:** `2604.02317`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Yujiao Shen, Shulin Tian, Jingkang Yang, Ziwei Liu
+- **Abstract page:** [https://arxiv.org/abs/2604.02317](https://arxiv.org/abs/2604.02317)
+- **PDF:** [https://arxiv.org/pdf/2604.02317v1](https://arxiv.org/pdf/2604.02317v1)
+
+### Abstract
+
+Recent streaming video understanding methods increasingly rely on complex memory mechanisms to handle long video streams. We challenge this trend with a simple finding: a sliding-window baseline that feeds only the most recent N frames to an off-the-shelf VLM already matches or surpasses published streaming models. We formalize this baseline as SimpleStream and evaluate it against 13 major offline and online video LLM baselines on OVO-Bench and StreamingBench. Despite its simplicity, SimpleStream delivers consistently strong performance. With only 4 recent frames, it reaches 67.7% average accuracy on OVO-Bench and 80.59% on StreamingBench. Controlled ablations further show that the value of longer context is backbone-dependent rather than uniformly increasing with model scale, and reveal a consistent perception-memory trade-off: adding more historical context can improve recall, but often weakens real-time perception. This suggests that stronger memory, retrieval, or compression modules should not be taken as evidence of progress unless they clearly outperform SimpleStream under the same protocol. We therefore argue that future streaming benchmarks should separate recent-scene perception from long-range memory, so that performance improvements from added complexity can be evaluated more clearly.
+
+---
+
+## 10. VOID: Video Object and Interaction Deletion
+
+- **arXiv ID:** `2604.02296`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `cs.AI`
+- **Published:** 2026-04-02
+- **Authors:** Saman Motamed, William Harvey, Benjamin Klein, Luc Van Gool, Zhuoning Yuan, Ta-Ying Cheng
+- **Abstract page:** [https://arxiv.org/abs/2604.02296](https://arxiv.org/abs/2604.02296)
+- **PDF:** [https://arxiv.org/pdf/2604.02296v1](https://arxiv.org/pdf/2604.02296v1)
+
+### Abstract
+
+Existing video object removal methods excel at inpainting content "behind" the object and correcting appearance-level artifacts such as shadows and reflections. However, when the removed object has more significant interactions, such as collisions with other objects, current models fail to correct them and produce implausible results. We present VOID, a video object removal framework designed to perform physically-plausible inpainting in these complex scenarios. To train the model, we generate a new paired dataset of counterfactual object removals using Kubric and HUMOTO, where removing an object requires altering downstream physical interactions. During inference, a vision-language model identifies regions of the scene affected by the removed object. These regions are then used to guide a video diffusion model that generates physically consistent counterfactual outcomes. Experiments on both synthetic and real data show that our approach better preserves consistent scene dynamics after object removal compared to prior video object removal methods. We hope this framework sheds light on how to make video editing models better simulators of the world through high-level causal reasoning.
+
+---
+
+## 11. AdamFlow: Adam-based Wasserstein Gradient Flows for Surface Registration in Medical Imaging
+
+- **arXiv ID:** `2604.02290`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `math.OC`
+- **Published:** 2026-04-02
+- **Authors:** Qiang Ma, Qingjie Meng, Xin Hu, Yicheng Wu, Wenjia Bai
+- **Abstract page:** [https://arxiv.org/abs/2604.02290](https://arxiv.org/abs/2604.02290)
+- **PDF:** [https://arxiv.org/pdf/2604.02290v1](https://arxiv.org/pdf/2604.02290v1)
+
+### Abstract
+
+Surface registration plays an important role for anatomical shape analysis in medical imaging. Existing surface registration methods often face a trade-off between efficiency and robustness. Local point matching methods are computationally efficient, but vulnerable to noise and initialisation. Methods designed for global point set alignment tend to incur a high computational cost. To address the challenge, here we present a fast surface registration method, which formulates surface meshes as probability measures and surface registration as a distributional optimisation problem. The discrepancy between two meshes is measured using an efficient sliced Wasserstein distance with log-linear computational complexity. We propose a novel optimisation method, AdamFlow, which generalises the well-known Adam optimisation method from the Euclidean space to the probability space for minimising the sliced Wasserstein distance. We theoretically analyse the asymptotic convergence of AdamFlow and empirically demonstrate its superior performance in both affine and non-rigid surface registration across various anatomical structures.
+
+---
+
+## 12. Omni123: Exploring 3D Native Foundation Models with Limited 3D Data by Unifying Text to 2D and 3D Generation
+
+- **arXiv ID:** `2604.02289`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `cs.AI`
+- **Published:** 2026-04-02
+- **Authors:** Chongjie Ye, Cheng Cao, Chuanyu Pan, Yiming Hao, Yihao Zhi, Yuanming Hu, Xiaoguang Han
+- **Abstract page:** [https://arxiv.org/abs/2604.02289](https://arxiv.org/abs/2604.02289)
+- **PDF:** [https://arxiv.org/pdf/2604.02289v1](https://arxiv.org/pdf/2604.02289v1)
+
+### Abstract
+
+Recent multimodal large language models have achieved strong performance in unified text and image understanding and generation, yet extending such native capability to 3D remains challenging due to limited data. Compared to abundant 2D imagery, high-quality 3D assets are scarce, making 3D synthesis under-constrained. Existing methods often rely on indirect pipelines that edit in 2D and lift results into 3D via optimization, sacrificing geometric consistency. We present Omni123, a 3D-native foundation model that unifies text-to-2D and text-to-3D generation within a single autoregressive framework. Our key insight is that cross-modal consistency between images and 3D can serve as an implicit structural constraint. By representing text, images, and 3D as discrete tokens in a shared sequence space, the model leverages abundant 2D data as a geometric prior to improve 3D representations. We introduce an interleaved X-to-X training paradigm that coordinates diverse cross-modal tasks over heterogeneous paired datasets without requiring fully aligned text-image-3D triplets. By traversing semantic-visual-geometric cycles (e.g., text to image to 3D to image) within autoregressive sequences, the model jointly enforces semantic alignment, appearance fidelity, and multi-view geometric consistency. Experiments show that Omni123 significantly improves text-guided 3D generation and editing, demonstrating a scalable path toward multimodal 3D world models.
+
+---
+
+## 13. Deep Neural Network Based Roadwork Detection for Autonomous Driving
+
+- **arXiv ID:** `2604.02282`
+- **Primary category:** `cs.RO`
+- **Categories:** `cs.RO`, `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Sebastian Wullrich, Nicolai Steinke, Daniel Goehring
+- **Abstract page:** [https://arxiv.org/abs/2604.02282](https://arxiv.org/abs/2604.02282)
+- **PDF:** [https://arxiv.org/pdf/2604.02282v1](https://arxiv.org/pdf/2604.02282v1)
+
+### Abstract
+
+Road construction sites create major challenges for both autonomous vehicles and human drivers due to their highly dynamic and heterogeneous nature. This paper presents a real-time system that detects and localizes roadworks by combining a YOLO neural network with LiDAR data. The system identifies individual roadwork objects while driving, merges them into coherent construction sites and records their outlines in world coordinates. The model training was based on an adapted US dataset and a new dataset collected from test drives with a prototype vehicle in Berlin, Germany. Evaluations on real-world road construction sites showed a localization accuracy below 0.5 m. The system can support traffic authorities with up-to-date roadwork data and could enable autonomous vehicles to navigate construction sites more safely in the future.
+
+---
+
+## 14. Novel Memory Forgetting Techniques for Autonomous AI Agents: Balancing Relevance and Efficiency
+
+- **arXiv ID:** `2604.02280`
+- **Primary category:** `cs.AI`
+- **Categories:** `cs.AI`, `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Payal Fofadiya, Sunil Tiwari
+- **Abstract page:** [https://arxiv.org/abs/2604.02280](https://arxiv.org/abs/2604.02280)
+- **PDF:** [https://arxiv.org/pdf/2604.02280v1](https://arxiv.org/pdf/2604.02280v1)
+
+### Abstract
+
+Long-horizon conversational agents require persistent memory for coherent reasoning, yet uncontrolled accumulation causes temporal decay and false memory propagation. Benchmarks such as LOCOMO and LOCCO report performance degradation from 0.455 to 0.05 across stages, while MultiWOZ shows 78.2% accuracy with 6.8% false memory rate under persistent retention. This work introduces an adaptive budgeted forgetting framework that regulates memory through relevanceguided scoring and bounded optimization. The approach integrates recency, frequency, and semantic alignment to maintain stability under constrained context. Comparative analysis demonstrates improved long-horizon F1 beyond 0.583 baseline levels, higher retention consistency, and reduced false memory behavior without increasing context usage. These findings confirm that structured forgetting preserves reasoning performance while preventing unbounded memory growth in extended conversational settings.
+
+---
+
+## 15. Modular Energy Steering for Safe Text-to-Image Generation with Foundation Models
+
+- **arXiv ID:** `2604.02265`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Yaoteng Tan, Zikui Cai, M. Salman Asif
+- **Abstract page:** [https://arxiv.org/abs/2604.02265](https://arxiv.org/abs/2604.02265)
+- **PDF:** [https://arxiv.org/pdf/2604.02265v1](https://arxiv.org/pdf/2604.02265v1)
+
+### Abstract
+
+Controlling the behavior of text-to-image generative models is critical for safe and practical deployment. Existing safety approaches typically rely on model fine-tuning or curated datasets, which can degrade generation quality or limit scalability. We propose an inference-time steering framework that leverages gradient feedback from frozen pretrained foundation models to guide the generation process without modifying the underlying generator. Our key observation is that vision-language foundation models encode rich semantic representations that can be repurposed as off-the-shelf supervisory signals during generation. By injecting such feedback through clean latent estimates at each sampling step, our method formulates safety steering as an energy-based sampling problem. This design enables modular, training-free safety control that is compatible with both diffusion and flow-matching models and can generalize across diverse visual concepts. Experiments demonstrate state-of-the-art robustness against NSFW red-teaming benchmarks and effective multi-target steering, while preserving high generation quality on benign non-targeted prompts. Our framework provides a principled approach for utilizing foundation models as semantic energy estimators, enabling reliable and scalable safety control for text-to-image generation.
+
+---
+
+## 16. SPAR: Single-Pass Any-Resolution ViT for Open-vocabulary Segmentation
+
+- **arXiv ID:** `2604.02252`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Naomi Kombol, Ivan Martinović, Siniša Šegvić, Giorgos Tolias
+- **Abstract page:** [https://arxiv.org/abs/2604.02252](https://arxiv.org/abs/2604.02252)
+- **PDF:** [https://arxiv.org/pdf/2604.02252v1](https://arxiv.org/pdf/2604.02252v1)
+
+### Abstract
+
+Foundational Vision Transformers (ViTs) have limited effectiveness in tasks requiring fine-grained spatial understanding, due to their fixed pre-training resolution and inherently coarse patch-level representations. These challenges are especially pronounced in dense prediction scenarios, such as open-vocabulary segmentation with ViT-based vision-language models, where high-resolution inputs are essential for accurate pixel-level reasoning. Existing approaches typically process large-resolution images using a sliding-window strategy at the pre-training resolution. While this improves accuracy through finer strides, it comes at a significant computational cost. We introduce SPAR: Single-Pass Any-Resolution ViT, a resolution-agnostic dense feature extractor designed for efficient high-resolution inference. We distill the spatial reasoning capabilities of a finely-strided, sliding-window teacher into a single-pass student using a feature regression loss, without requiring architectural changes or pixel-level supervision. Applied to open-vocabulary segmentation, SPAR improves single-pass baselines by up to 10.5 mIoU and even surpasses the teacher, demonstrating effectiveness in efficient, high-resolution reasoning. Code: [https://github.com/naomikombol/SPAR](https://github.com/naomikombol/SPAR)
+
+---
+
+## 17. UAV-Track VLA: Embodied Aerial Tracking via Vision-Language-Action Models
+
+- **arXiv ID:** `2604.02241`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `cs.RO`
+- **Published:** 2026-04-02
+- **Authors:** Qiyao Zhang, Shuhua Zheng, Jianli Sun, Chengxiang Li, Xianke Wu, Zihan Song, Zhiyong Cui, Yisheng Lv, Yonglin Tian
+- **Abstract page:** [https://arxiv.org/abs/2604.02241](https://arxiv.org/abs/2604.02241)
+- **PDF:** [https://arxiv.org/pdf/2604.02241v1](https://arxiv.org/pdf/2604.02241v1)
+
+### Abstract
+
+Embodied visual tracking is crucial for Unmanned Aerial Vehicles (UAVs) executing complex real-world tasks. In dynamic urban scenarios with complex semantic requirements, Vision-Language-Action (VLA) models show great promise due to their cross-modal fusion and continuous action generation capabilities. To benchmark multimodal tracking in such environments, we construct a dedicated evaluation benchmark and a large-scale dataset encompassing over 890K frames, 176 tasks, and 85 diverse objects. Furthermore, to address temporal feature redundancy and the lack of spatial geometric priors in existing VLA models, we propose an improved VLA tracking model, UAV-Track VLA. Built upon the $π_{0.5}$ architecture, our model introduces a temporal compression net to efficiently capture inter-frame dynamics. Additionally, a parallel dual-branch decoder comprising a spatial-aware auxiliary grounding head and a flow matching action expert is designed to decouple cross-modal features and generate fine-grained continuous actions. Systematic experiments in the CARLA simulator validate the superior end-to-end performance of our method. Notably, in challenging long-distance pedestrian tracking tasks, UAV-Track VLA achieves a 61.76 success rate and 269.65 average tracking frames, significantly outperforming existing baselines. Furthermore, it demonstrates robust zero-shot generalization in unseen environments and reduces single-step inference latency by 33.4 (to 0.0571s) compared to the original $π_{0.5}$, enabling highly efficient, real-time UAV control. Data samples and demonstration videos are available at: [https://github.com/Hub-Tian/UAV-Track\_VLA](https://github.com/Hub-Tian/UAV-Track\_VLA).
+
+---
+
+## 18. SCALE: Semantic- and Confidence-Aware Conditional Variational Autoencoder for Zero-shot Skeleton-based Action Recognition
+
+- **arXiv ID:** `2604.02222`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Soroush Oraki, Feng Ding, Jie Liang
+- **Abstract page:** [https://arxiv.org/abs/2604.02222](https://arxiv.org/abs/2604.02222)
+- **PDF:** [https://arxiv.org/pdf/2604.02222v1](https://arxiv.org/pdf/2604.02222v1)
+
+### Abstract
+
+Zero-shot skeleton-based action recognition (ZSAR) aims to recognize action classes without any training skeletons from those classes, relying instead on auxiliary semantics from text. Existing approaches frequently depend on explicit skeleton-text alignment, which can be brittle when action names underspecify fine-grained dynamics and when unseen classes are semantically confusable. We propose SCALE, a lightweight and deterministic Semantic- and Confidence-Aware Listwise Energy-based framework that formulates ZSAR as class-conditional energy ranking. SCALE builds a text-conditioned Conditional Variational Autoencoder where frozen text representations parameterize both the latent prior and the decoder, enabling likelihood-based evaluation for unseen classes without generating samples at test time. To separate competing hypotheses, we introduce a semantic- and confidence-aware listwise energy loss that emphasizes semantically similar hard negatives and incorporates posterior uncertainty to adapt decision margins and reweight ambiguous training instances. Additionally, we utilize a latent prototype contrast objective to align posterior means with text-derived latent prototypes, improving semantic organization and class separability without direct feature matching. Experiments on NTU-60 and NTU-120 datasets show that SCALE consistently improves over prior VAE- and alignment-based baselines while remaining competitive with diffusion-based methods.
+
+---
+
+## 19. UniDriveVLA: Unifying Understanding, Perception, and Action Planning for Autonomous Driving
+
+- **arXiv ID:** `2604.02190`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `cs.RO`
+- **Published:** 2026-04-02
+- **Authors:** Yongkang Li, Lijun Zhou, Sixu Yan, Bencheng Liao, Tianyi Yan, Kaixin Xiong, Long Chen, Hongwei Xie, Bing Wang, Guang Chen, Hangjun Ye, Wenyu Liu, Haiyang Sun, Xinggang Wang
+- **Abstract page:** [https://arxiv.org/abs/2604.02190](https://arxiv.org/abs/2604.02190)
+- **PDF:** [https://arxiv.org/pdf/2604.02190v1](https://arxiv.org/pdf/2604.02190v1)
+
+### Abstract
+
+Vision-Language-Action (VLA) models have recently emerged in autonomous driving, with the promise of leveraging rich world knowledge to improve the cognitive capabilities of driving systems. However, adapting such models for driving tasks currently faces a critical dilemma between spatial perception and semantic reasoning. Consequently, existing VLA systems are forced into suboptimal compromises: directly adopting 2D Vision-Language Models yields limited spatial perception, whereas enhancing them with 3D spatial representations often impairs the native reasoning capacity of VLMs. We argue that this dilemma largely stems from the coupled optimization of spatial perception and semantic reasoning within shared model parameters. To overcome this, we propose UniDriveVLA, a Unified Driving Vision-Language-Action model based on Mixture-of-Transformers that addresses the perception-reasoning conflict via expert decoupling. Specifically, it comprises three experts for driving understanding, scene perception, and action planning, which are coordinated through masked joint attention. In addition, we combine a sparse perception paradigm with a three-stage progressive training strategy to improve spatial perception while maintaining semantic reasoning capability. Extensive experiments show that UniDriveVLA achieves state-of-the-art performance in open-loop evaluation on nuScenes and closed-loop evaluation on Bench2Drive. Moreover, it demonstrates strong performance across a broad range of perception, prediction, and understanding tasks, including 3D detection, online mapping, motion forecasting, and driving-oriented VQA, highlighting its broad applicability as a unified model for autonomous driving. Code and model have been released at [https://github.com/xiaomi-research/unidrivevla](https://github.com/xiaomi-research/unidrivevla)
+
+---
+
+## 20. Lightweight Spatiotemporal Highway Lane Detection via 3D-ResNet and PINet with ROI-Aware Attention
+
+- **arXiv ID:** `2604.02188`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Sorna Shanmuga Raja, Abdelhafid Zenati
+- **Abstract page:** [https://arxiv.org/abs/2604.02188](https://arxiv.org/abs/2604.02188)
+- **PDF:** [https://arxiv.org/pdf/2604.02188v1](https://arxiv.org/pdf/2604.02188v1)
+
+### Abstract
+
+This paper presents a lightweight, end-to-end highway lane detection architecture that jointly captures spatial and temporal information for robust performance in real-world driving scenarios. Building on the strengths of 3D convolutional neural networks and instance segmentation, we propose two models that integrate a 3D-ResNet encoder with a Point Instance Network (PINet) decoder. The first model enhances multi-scale feature representation using a Feature Pyramid Network (FPN) and Self-Attention mechanism to refine spatial dependencies. The second model introduces a Region of Interest (ROI) detection head to selectively focus on lane-relevant regions, thereby improving precision and reducing computational complexity. Experiments conducted on the TuSimple dataset (highway driving scenarios) demonstrate that the proposed second model achieves 93.40% accuracy while significantly reducing false negatives. Compared to existing 2D and 3D baselines, our approach achieves improved performance with fewer parameters and reduced latency. The architecture has been validated through offline training and real-time inference in the Autonomous Systems Laboratory at City, St George's University of London. These results suggest that the proposed models are well-suited for integration into Advanced Driver Assistance Systems (ADAS), with potential scalability toward full Lane Assist Systems (LAS).
+
+---
+
+## 21. CXR-LT 2026 Challenge: Projection-Aware Multi-Label and Zero-Shot Chest X-Ray Classification
+
+- **arXiv ID:** `2604.02185`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Juno Cho, Dohui Kim, Mingeon Kim, Hyunseo Jang, Chang Sun Lee, Jong Chul Ye
+- **Abstract page:** [https://arxiv.org/abs/2604.02185](https://arxiv.org/abs/2604.02185)
+- **PDF:** [https://arxiv.org/pdf/2604.02185v1](https://arxiv.org/pdf/2604.02185v1)
+
+### Abstract
+
+This challenge tackles multi-label classification for known chest X-ray (CXR) lesions and zero-shot classification for unseen ones. To handle diverse CXR projections, we integrate projection-specific models via a classification network into a unified framework. For zero-shot classification (Task 2), we extend CheXzero with a novel dual-branch architecture that combines contrastive learning, Asymmetric Loss (ASL), and LLM-generated descriptive prompts. This effectively mitigates severe long-tail imbalances and maximizes zero-shot generalization. Additionally, strong data and test-time augmentations (TTA) ensure robustness across both tasks.
+
+---
+
+## 22. ViT-Explainer: An Interactive Walkthrough of the Vision Transformer Pipeline
+
+- **arXiv ID:** `2604.02182`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `cs.HC`
+- **Published:** 2026-04-02
+- **Authors:** Juan Manuel Hernandez, Mariana Fernandez-Espinosa, Denis Parra, Diego Gomez-Zara
+- **Abstract page:** [https://arxiv.org/abs/2604.02182](https://arxiv.org/abs/2604.02182)
+- **PDF:** [https://arxiv.org/pdf/2604.02182v1](https://arxiv.org/pdf/2604.02182v1)
+
+### Abstract
+
+Transformer-based architectures have become the shared backbone of natural language processing and computer vision. However, understanding how these models operate remains challenging, particularly in vision settings, where images are processed as sequences of patch tokens. Existing interpretability tools often focus on isolated components or expert-oriented analysis, leaving a gap in guided, end-to-end understanding of the full inference pipeline. To bridge this gap, we present ViT-Explainer, a web-based interactive system that provides an integrated visualization of Vision Transformer inference, from patch tokenization to final classification. The system combines animated walkthroughs, patch-level attention overlays, and a vision-adapted Logit Lens within both guided and free exploration modes. A user study with six participants suggests that ViT-Explainer is easy to learn and use, helping users interpret and understand Vision Transformer behavior.
+
+---
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/arxiv-cs-cv-2026-04-05.rag_ingested b/docs/AI_Computer_Vision/arxiv-cs-cv-2026-04-05.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..fa1f9e3aa68347cf2170448b899aa9d0bcf547f9
--- /dev/null
+++ b/docs/AI_Computer_Vision/arxiv-cs-cv-2026-04-05.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "2d2fdab41e12646b", "ingested_at": "2026-06-30T08:49:23.959865", "content_hash": "2d2fdab41e12646b"}
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/arxiv_video_analysis_papers.md b/docs/AI_Computer_Vision/arxiv_video_analysis_papers.md
new file mode 100644
index 0000000000000000000000000000000000000000..149eb1e148fac34ce31fd7188df4a5695ba3f332
--- /dev/null
+++ b/docs/AI_Computer_Vision/arxiv_video_analysis_papers.md
@@ -0,0 +1,74 @@
+# arXiv 動画解析・理解 関連論文リスト
+
+arXivから収集した「動画の解析(Video Analysis / Video Understanding)」に関する代表的かつ最新の学術論文リストです。
+
+本リストは **`literature-search-arxiv`** スキルを使用して取得されました。
+
+---
+
+## 1. 包括的サーベイ論文(全体の動向把握に最適)
+
+### **Video Understanding: From Geometry and Semantics to Unified Models**
+* **著者**: Zhaochong An, Zirui Li, Mingqiao Ye, Feng Qiao, Jiaang Li, Zongwei Wu, Vishal Thengane, Chengzu Li, Lei Li, Luc Van Gool, Guolei Sun, Serge Belongie
+* **公開日**: 2026年3月18日
+* **arXiv ID**: [2603.17840v1](https://arxiv.org/abs/2603.17840v1)
+* **PDFリンク**: [https://arxiv.org/pdf/2603.17840v1](https://arxiv.org/pdf/2603.17840v1)
+* **概要**:
+ 動画理解の進展を、低レベルの幾何学(geometry)理解、高レベルのセマンティクス理解、および「統一型動画理解モデル」の3つの視点から構造的にまとめた包括的な最新サーベイです。従来のタスク特化型パイプライン設計から、マルチモーダルかつ多様な下流タスクに適応可能な動画基盤モデル(Video Foundation Model)への移行について詳しく論じています。
+
+---
+
+## 2. 長尺動画理解とプロンプト最適化(最新技術)
+
+### **CoS: Chain-of-Shot Prompting for Long Video Understanding**
+* **著者**: Jian Hu, Zixu Cheng, Chenyang Si, Wei Li, Shaogang Gong
+* **公開日**: 2025年2月10日
+* **arXiv ID**: [2502.06428v2](https://arxiv.org/abs/2502.06428v2)
+* **PDFリンク**: [https://arxiv.org/pdf/2502.06428v2](https://arxiv.org/pdf/2502.06428v2)
+* **概要**:
+ マルチモーダル大規模言語モデル(MLLM)が長尺動画を扱う際、視覚トークンの多さによる文脈制限や余分なシーン(不要なショット)の混入が課題となります。本論文では、タスクに応じて意味的に重要度の高いショットを動的に選別するテスト時プロンプト最適化手法「Chain-of-Shot (CoS)」を提案しています。
+
+---
+
+## 3. 動画基盤モデル(大規模マルチモーダル学習)
+
+### **InternVideo2: Scaling Foundation Models for Multimodal Video Understanding**
+* **著者**: Yi Wang, Kunchang Li, Xinhao Li, Jiashuo Yu, Yinan He, Chenting Wang, Guo Chen, Baoqi Pei, Ziang Yan, Rongkun Zheng, Jilan Xu, Zun Wang, Yansong Shi, Tianxiang Jiang, Songze Li, Hongjie Zhang, Yifei Huang, Yu Qiao, Yali Wang, Limin Wang
+* **公開日**: 2024年3月22日(ECCV 2024採択)
+* **arXiv ID**: [2403.15377v4](https://arxiv.org/abs/2403.15377v4)
+* **PDFリンク**: [https://arxiv.org/pdf/2403.15377v4](https://arxiv.org/pdf/2403.15377v4)
+* **概要**:
+ 動画エンコーダを60億パラメータ(6B)規模にスケールアップした動画基盤モデル「InternVideo2」を提案しています。Masked Video Modeling、対照学習、Next Token Predictionを統一した段階的学習により、動画認識や動画対話など60以上の多様な動画関連タスクで高いパフォーマンスを実証しています。
+
+---
+
+## 4. 人間中心の複雑な行動解析とデータセット
+
+### **Human in Events: A Large-Scale Benchmark for Human-centric Video Analysis in Complex Events**
+* **著者**: Weiyao Lin, Huabin Liu, Shizhan Liu, Yuxi Li, Rui Qian, Tao Wang, Ning Xu, Hongkai Xiong, Guo-Jun Qi, Nicu Sebe
+* **公開日**: 2020年5月9日(IJCV 2023掲載)
+* **arXiv ID**: [2005.04490v6](https://arxiv.org/abs/2005.04490v6)
+* **PDFリンク**: [https://arxiv.org/pdf/2005.04490v6](https://arxiv.org/pdf/2005.04490v6)
+* **概要**:
+ 実世界の混雑シーンや異常行動が発生する複雑なイベントにおいて、人間の姿勢・動作・軌跡を追跡する「Human-in-Events (HiEve)」という大規模データセットを提案。姿勢推定や行動分類の学習用ベンチマークとして広く活用されています。
+
+---
+
+## 5. 効率的なオンライン動画解析(高速化技術)
+
+### **ECO: Efficient Convolutional Network for Online Video Understanding**
+* **著者**: Mohammadreza Zolfaghari, Kamaljeet Singh, Thomas Brox
+* **公開日**: 2018年4月24日(ECCV 2018)
+* **arXiv ID**: [1804.09066v2](https://arxiv.org/abs/1804.09066v2)
+* **PDFリンク**: [https://arxiv.org/pdf/1804.09066v2](https://arxiv.org/pdf/1804.09066v2)
+* **概要**:
+ 動画解析における計算コストの課題に対し、フレーム間の冗長性を削減するサンプリング戦略と、ネットワーク内での長期特徴マージ手法を組み合わせた「ECO(Efficient Convolutional Network)」を提案。既存手法と比較して大幅な高速処理(10倍〜80倍)を実現し、オンラインでのアクティビティ分類に貢献しました。
+
+---
+
+## 参照したペーパーのURL一覧
+* [https://arxiv.org/pdf/2603.17840v1](https://arxiv.org/pdf/2603.17840v1)
+* [https://arxiv.org/pdf/2502.06428v2](https://arxiv.org/pdf/2502.06428v2)
+* [https://arxiv.org/pdf/2403.15377v4](https://arxiv.org/pdf/2403.15377v4)
+* [https://arxiv.org/pdf/2005.04490v6](https://arxiv.org/pdf/2005.04490v6)
+* [https://arxiv.org/pdf/1804.09066v2](https://arxiv.org/pdf/1804.09066v2)
diff --git a/docs/AI_Computer_Vision/arxiv_video_analysis_papers.rag_ingested b/docs/AI_Computer_Vision/arxiv_video_analysis_papers.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..77e4b198f12957a030c0ecbcf0fb51b84f253d49
--- /dev/null
+++ b/docs/AI_Computer_Vision/arxiv_video_analysis_papers.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "e1729b3a6980a520", "ingested_at": "2026-06-30T08:49:24.586622", "content_hash": "e1729b3a6980a520"}
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/layout_models_survey.md b/docs/AI_Computer_Vision/layout_models_survey.md
new file mode 100644
index 0000000000000000000000000000000000000000..a95a6de0a35eb9cba5812150dd1596f79bfd1eb0
--- /dev/null
+++ b/docs/AI_Computer_Vision/layout_models_survey.md
@@ -0,0 +1,370 @@
+# レイアウト分析・生成モデル 包括的サーベイ
+
+> [!NOTE]
+> 本レポートは、**文書・画像・動画**におけるレイアウト分析モデルとレイアウト生成モデルの最新動向(2024〜2025年)を網羅的にまとめたものです。
+
+---
+
+## 目次
+
+1. [文書レイアウト分析モデル](#1-文書レイアウト分析モデル)
+2. [レイアウト生成モデル(グラフィック/UI/文書)](#2-レイアウト生成モデルグラフィックui文書)
+3. [画像レイアウト生成モデル(Layout-to-Image)](#3-画像レイアウト生成モデルlayout-to-image)
+4. [動画レイアウト分析・生成モデル](#4-動画レイアウト分析生成モデル)
+5. [VLM/LLMベースのレイアウト理解モデル](#5-vlmllmベースのレイアウト理解モデル)
+6. [商用レイアウト分析ソリューション](#6-商用レイアウト分析ソリューション)
+7. [主要データセット・ベンチマーク](#7-主要データセットベンチマーク)
+8. [技術トレンド総括](#8-技術トレンド総括)
+
+---
+
+## 1. 文書レイアウト分析モデル
+
+文書レイアウト分析(Document Layout Analysis: DLA)は、文書画像からテキスト、表、図、タイトル等の構造要素を検出・分類するタスクです。
+
+### 1.1 マルチモーダルTransformerベース
+
+| モデル | 年 | アーキテクチャ | 特徴 | リンク |
+|:---|:---|:---|:---|:---|
+| **LayoutLM** | 2020 | BERT + 2D Position Embedding | テキスト+空間位置情報の統合学習のパイオニア | [GitHub](https://github.com/microsoft/unilm/tree/master/layoutlm) |
+| **LayoutLMv2** | 2021 | Transformer + CNN Backbone | 視覚情報の追加、空間アテンション導入 | [HuggingFace](https://huggingface.co/microsoft/layoutlmv2-base-uncased) |
+| **LayoutLMv3** | 2022 | Unified Transformer (Patch Embed) | CNNをViTスタイルに置換、統一的マスクモデリング | [HuggingFace](https://huggingface.co/microsoft/layoutlmv3-base) |
+| **LayoutXLM** | 2022 | 多言語LayoutLMv2 | 53言語対応、多言語文書理解 | [HuggingFace](https://huggingface.co/microsoft/layoutxlm-base) |
+
+**特長**: テキスト・視覚・レイアウトの3モダリティを統合し、フォーム理解、文書VQA、エンティティ抽出など**意味理解**を必要とするタスクに最適。
+
+### 1.2 物体検出ベース(ビジョンのみ)
+
+| モデル | 年 | ベース | 特徴 | リンク |
+|:---|:---|:---|:---|:---|
+| **DocLayout-YOLO** | 2024 | YOLO | Global-to-Local Controllable Receptive Module、DocSynth-300K合成データで学習 | [GitHub](https://github.com/opendatalab/DocLayout-YOLO) |
+| **PP-DocLayout** | 2024 | PaddleDetection | M/Sバリアント、リソース制約環境向け最適化 | PaddlePaddle |
+| **Heron-101** | 2025 | RT-DETR | Doclingパイプラインの一部、28ms/imageの高速推論、78% mAP | [Docling](https://github.com/DS4SD/docling) |
+| **LayoutParser** | 2021 | Detectron2 | レイアウト検出の統合ライブラリ、複数モデル対応 | [GitHub](https://github.com/Layout-Parser/layout-parser) |
+| **chunkr-layout-1** | 2025 | — | 80.9% mAP@50、金融・法律・医療文書向けプロダクションモデル | [chunkr.ai](https://chunkr.ai) |
+
+**特長**: テキスト情報を使わず画像のみでレイアウト要素を検出。**高速推論**が必要なパイプラインに最適。
+
+### 1.3 統一Transformerベース
+
+| モデル | 年 | アーキテクチャ | 特徴 |
+|:---|:---|:---|:---|
+| **DLAFormer** | 2024 (ICDAR) | DETR-inspired Encoder-Decoder | 物体検出・テキスト領域検出・論理役割分類・読み順予測を統一的に関係予測として処理 |
+| **RT-DETRv2 / DFINE** | 2024-2025 | Real-Time DETR | リアルタイム推論と高精度のバランス |
+
+### 1.4 グラフベース手法
+
+| モデル | 年 | 特徴 |
+|:---|:---|:---|
+| **Paragraph2Graph** | 2024 | 文書要素間の空間・論理関係をGNNでモデリング |
+| **gDSA** | 2024 | グラフベース文書構造解析 |
+
+### 1.5 パイプライン/フレームワーク
+
+| システム | 開発 | 特徴 |
+|:---|:---|:---|
+| **Docling** | IBM | PDFからMarkdown/JSONへのエンドツーエンド変換、Heron, PP-DocLayoutなど多モデル対応、LangChain/LlamaIndex統合 |
+| **MinerU** | — | RAG向け文書パース |
+| **Unstructured.io** | — | 文書パース統合フレームワーク |
+
+---
+
+## 2. レイアウト生成モデル(グラフィック/UI/文書)
+
+レイアウト生成は、要素の種類・制約条件から自動的にレイアウト(バウンディングボックスの位置・サイズ)を生成するタスクです。
+
+### 2.1 GAN / VAEベース(初期手法)
+
+| モデル | 年 | アーキテクチャ | 特徴 |
+|:---|:---|:---|:---|
+| **LayoutGAN** | 2019 | GAN | 微分可能ワイヤフレームレンダリング層、CNN判別器による整列性評価 |
+| **LayoutVAE** | 2019 | VAE | CountVAE(要素数予測)+ BBoxVAE(位置予測)の2段構成 |
+
+### 2.2 Transformerベース
+
+| モデル | 年 | アーキテクチャ | 特徴 |
+|:---|:---|:---|:---|
+| **LayoutTransformer** | 2021 | Autoregressive Transformer | レイアウトを離散トークン列として自己回帰生成 |
+| **BLT** (Bidirectional Layout Transformer) | 2022 (Google) | Non-Autoregressive Transformer | ドラフト生成→反復精錬、10倍高速化、制御可能な生成 |
+| **LayoutFormer++** | 2023 | Encoder-Decoder / Decoder-only | Seq2Seq形式の統一レイアウト生成 |
+| **StructLayoutFormer** | 2025 | Transformer | 構造情報と要素配置を分離、制約付きレイアウト生成 |
+
+### 2.3 拡散モデル / Flow Matchingベース
+
+| モデル | 年 | アーキテクチャ | 特徴 |
+|:---|:---|:---|:---|
+| **LayoutDM** | 2023 | Discrete Diffusion + Transformer | ノイズからレイアウトを段階的デノイジング、柔軟な条件付き生成(補完、カテゴリ→位置) |
+| **LayoutFlow** | 2024 | Flow Matching | Diffusionの代替、滑らかなサンプリング軌跡、高速推論 |
+| **LACE** | 2024 | Constrained Diffusion | 美的制約を反映したレイアウト拡散モデル |
+| **UniLayDiff** | 2025 | Unified Diffusion Transformer | 多様なコンテンツ認識レイアウトタスクを単一モデルで統合 |
+| **SLayR** | 2025 | Rectified Flow | シーン・グラフィックレイアウト生成向け改善型Flow Matching |
+
+### 2.4 LLMベース
+
+| モデル | 年 | ベースLLM | 特徴 |
+|:---|:---|:---|:---|
+| **LayoutGPT** | 2023 (NeurIPS) | GPT-3.5/4 | テキスト条件からCSS/コード形式でレイアウト生成、学習不要 |
+| **PosterLlama** | 2024 | LLaMA系 | HTML形式でポスター・グラフィックレイアウトを生成、視覚エンコーダとLLMの2段階学習 |
+| **PosterLLaVa** | 2024 | LLaVA系 | LLMベースのコンテンツ認識レイアウト生成 |
+| **LGGPT** | 2024/2025 | 汎用LLM | Arbitrary Layout Instruction (ALI) + Universal Layout Response (ULR) で統一的レイアウト生成 |
+| **LayoutKAG** | 2025 | LLM + Knowledge Graph | 知識拡張型レイアウト生成 |
+
+> [!TIP]
+> **2024-2025年のトレンド**: LLMが「レイアウトプランナー」として高レベルな設計知識を提供し、Diffusion Transformerが高品質な最終出力を合成する**ハイブリッドアプローチ**が主流になりつつあります。
+
+---
+
+## 3. 画像レイアウト生成モデル(Layout-to-Image)
+
+バウンディングボックスやレイアウト指定から、それに合致する画像を生成するタスクです。
+
+### 3.1 Layout-to-Image生成
+
+| モデル | 年 | 会議 | アーキテクチャ | 特徴 |
+|:---|:---|:---|:---|:---|
+| **InstanceAssemble** | 2025 | NeurIPS | DiT + LoRA | Instance-assembling attention、Flux/SD3互換、DenseLayoutベンチマーク(5K画像/90Kインスタンス)、Layout Grounding Score (LGS) 指標を提案 |
+| **CreatiLayout** | 2025 | ICCV | Siamese MM-DiT (SiamLayout) | レイアウトを独立モダリティとして扱う、LayoutDesigner(LLMプランナー)、LayoutSAMデータセット(270万画像/1070万エンティティ) |
+| **UniLayDiff** | 2025 | — | Unified DiT | コンテンツ認識レイアウトタスクの統一モデル、LoRAベースの制約統合 |
+| **GLIGEN** | 2023 | CVPR | Grounded Diffusion | バウンディングボックスによる空間制御の先駆的手法 |
+| **LayoutDiffuse** | 2023 | — | Diffusion | セマンティックレイアウトからのシーン画像生成 |
+
+### 3.2 UI/Webレイアウト生成
+
+| ツール/モデル | 特徴 |
+|:---|:---|
+| **Galileo AI** | テキストプロンプトからUI/UXモックアップ生成、Figmaエクスポート対応 |
+| **Uizard** | ラフスケッチからプロフェッショナルUIモックアップ |
+| **Screenshot-to-Code** | スクリーンショットからHTML/CSSコードへ変換 |
+| **Design2Code** | デザイン画像からフロントエンドコード生成 |
+
+---
+
+## 4. 動画レイアウト分析・生成モデル
+
+### 4.1 レイアウト制御型動画生成
+
+動画生成において、物体の空間配置(バウンディングボックス)、カメラ軌道、動きの方向を明示的に制御するモデル群です。
+
+| モデル | 年 | 制御方式 | 特徴 |
+|:---|:---|:---|:---|
+| **VideoDirectorGPT** | 2024 (COLM) | LLMプランニング + BBox | GPT-4がテキストからビデオプラン(シーン記述・エンティティレイアウト・一貫性グループ)を生成、Layout2Vidモジュールで空間制御付き動画合成 |
+| **TrailBlazer** | 2024 | BBox軌跡 + Attention編集 | 学習不要、事前学習済T2Vモデルの時空間アテンションマップを編集してキーフレームBBoxで物体軌跡制御 |
+| **MotionCtrl** | 2024 | カメラ運動 + 物体運動 | CMCM(カメラ制御)+ OMCM(物体制御)の2モジュール、アダプターベース |
+| **DrivingDiffusion** | 2024 | 3D空間レイアウト | 自動運転シーン向けマルチビュー一貫性、3D空間レイアウトガイド |
+| **GenProp** | 2025 (Adobe) | 1フレーム編集伝播 | 最初のフレームの変更を動画全体に伝播(挿入・除去・トラッキングの統一) |
+
+### 4.2 動画生成基盤モデル(DiTベース)
+
+2024-2025年、動画生成はU-Netから**Diffusion Transformer (DiT)** アーキテクチャへ大きく移行しました。
+
+| モデル | 開発 | 特徴 |
+|:---|:---|:---|
+| **HunyuanVideo** | Tencent | 長尺高品質動画、DiTベース |
+| **CogVideoX** | Zhipu AI | テキストから動画生成、DiTベース |
+| **Wan2.1** | Alibaba | 大規模動画生成 |
+
+### 4.3 制御方式の分類
+
+```mermaid
+graph TB
+ A["動画レイアウト制御"] --> B["構造的制御"]
+ A --> C["ID/外観制御"]
+ A --> D["動き制御"]
+
+ B --> B1["BBox/レイアウト"]
+ B --> B2["深度マップ"]
+ B --> B3["スケッチ"]
+ B --> B4["セマンティックマップ"]
+
+ D --> D1["カメラ軌道"]
+ D --> D2["物体軌跡"]
+ D --> D3["ランドマーク/ポーズ"]
+
+ B1 --> E["VideoDirectorGPT"]
+ B1 --> F["TrailBlazer"]
+ D1 --> G["MotionCtrl"]
+ D2 --> G
+```
+
+### 4.4 データセット
+
+| データセット | 特徴 |
+|:---|:---|
+| **LVD-2M** | 200万本の長尺動画(10秒以上)、時間的に密なキャプション付き |
+
+---
+
+## 5. VLM/LLMベースのレイアウト理解モデル
+
+大規模Vision-Language Model(VLM)を用いた文書・画像レイアウト理解は2024-2025年の主要トレンドです。
+
+### 5.1 文書特化VLM
+
+| モデル | 開発 | 特徴 |
+|:---|:---|:---|
+| **mPLUG-DocOwl2** | Alibaba / Renmin Univ. | OCRフリー文書理解、高解像度DocCompressorで1ページ324トークンに圧縮、マルチページ対応 |
+| **Qwen2.5-VL** | Alibaba Cloud | エージェント能力(ツール使用、画面操作)、動的解像度ViT、複雑なレイアウト・チャート・表の分析、構造化出力(JSON)生成 |
+| **Florence-2** | Microsoft | 統一Seq2Seqフレームワーク、キャプション・物体検出・セグメンテーション・OCRを1モデルで、軽量・ファインチューニング向き |
+| **GLM-4.5V** | Zhipu AI | 高度なマルチモーダル推論、文書レイアウト理解 |
+| **InternVL2** | Shanghai AI Lab | 高性能VLM、文書理解タスクでSOTA級 |
+
+### 5.2 比較表
+
+| モデル | 最適用途 | レイアウト理解力 |
+|:---|:---|:---|
+| **mPLUG-DocOwl2** | 効率的なマルチページ文書分析(低レイテンシ) | ◎ OCRフリー高効率 |
+| **Qwen2.5-VL** | 複雑な推論・構造化データ抽出・エージェントタスク | ◎◎ 最高性能 |
+| **Florence-2** | 軽量・汎用ビジョンタスク(OCR+検出+セグメンテーション) | ○ 汎用的 |
+
+---
+
+## 6. 商用レイアウト分析ソリューション
+
+### 6.1 クラウドベンダー比較
+
+| サービス | 提供 | 強み | 最適ユースケース |
+|:---|:---|:---|:---|
+| **Google Document AI** | Google Cloud | 特化型プリトレーニングプロセッサ群(請求書、税務書類、契約書等)、Vertex AI/BigQuery連携 | GCPユーザー、特定業務文書の高精度処理 |
+| **Amazon Textract** | AWS | S3/Lambda/SQSとのネイティブ連携、Queries機能(自然言語で特定データ抽出) | 大量トランザクション処理、AWSパイプライン |
+| **Azure AI Document Intelligence** | Microsoft | **レイアウト分析でリーダー的評価**、可変レイアウト文書への対応、Studio UIでカスタムモデル学習が容易 | 複雑・非定型レイアウトの文書 |
+
+### 6.2 その他の商用/デザインツール
+
+| ツール | 分野 | 特徴 |
+|:---|:---|:---|
+| **Adobe Firefly** | グラフィックデザイン | Photoshop/Illustrator統合、生成塗りつぶし、テキストからベクター |
+| **Canva Magic Studio** | グラフィックデザイン | AIレイアウト提案、Magic Designでプレゼン即時生成 |
+| **Kittl / Recraft** | グラフィックデザイン | ブランド一貫性重視、ベクターベース出力、編集可能アセット |
+| **Reducto** | 文書処理 | エンタープライズ向けAgentic文書解析 |
+| **V7 Go** | 文書処理 | 高精度・検証可能なAI文書分析 |
+
+---
+
+## 7. 主要データセット・ベンチマーク
+
+### 7.1 文書レイアウト分析
+
+| データセット | サイズ | アノテーション | 対象文書 |
+|:---|:---|:---|:---|
+| **PubLayNet** | ~360K ページ | 自動生成(PDF/XML対応) | 科学論文(PubMed Central) |
+| **DocLayNet** | ~81K ページ | 人手アノテーション、11クラス | 金融、マニュアル、科学、法律、特許、政府 |
+| **DocBank** | 500K ページ | 弱教師(LaTeXソース) | arXiv論文、トークンレベルアノテーション |
+| **D⁴LA** | — | 詳細人手アノテーション | 多様な文書 |
+| **DocStructBench** | 数千ページ | 人手、10カテゴリ | 学術、教科書、金融、市場分析 |
+| **Comp-HRDoc** | — | — | 複雑な高解像度文書 |
+
+### 7.2 レイアウト生成
+
+| データセット | 対象 | 特徴 |
+|:---|:---|:---|
+| **RICO** | モバイルUI | UIデザイン・構造階層の大規模コレクション |
+| **PubLayNet** | 文書 | レイアウト生成評価にも使用 |
+| **Magazine** | 雑誌レイアウト | ポスター・雑誌のグラフィックレイアウト |
+| **CGL Dataset** | コンテンツ認識レイアウト | グラフィックレイアウト生成向け |
+| **LayoutSAM** | 汎用画像 | CreatiLayoutが提案、270万画像/1070万エンティティ |
+| **DenseLayout** | 汎用画像 | InstanceAssembleが提案、5K画像/90Kインスタンス |
+
+### 7.3 評価指標
+
+| 分野 | 指標 | 説明 |
+|:---|:---|:---|
+| **レイアウト検出** | mAP, IoU | 物体検出の標準指標 |
+| **レイアウト生成** | FID | 生成レイアウトの品質・多様性 |
+| **レイアウト生成** | Alignment, Overlap | レイアウト要素の整列性、重複度 |
+| **レイアウト生成** | LGS (Layout Grounding Score) | InstanceAssembleが提案した新指標 |
+| **文書理解** | DocVQA, ANLS | 文書視覚質問応答 |
+| **レイアウト検出** | LPIPS | ロバストネスベンチマーク向け |
+
+---
+
+## 8. 技術トレンド総括
+
+### 8.1 全体マップ
+
+```mermaid
+graph LR
+ subgraph "レイアウト分析(入力→構造理解)"
+ DA["文書レイアウト分析"]
+ IA["画像レイアウト分析"]
+ VA["動画レイアウト分析"]
+ end
+
+ subgraph "レイアウト生成(条件→レイアウト出力)"
+ DG["文書/グラフィック
レイアウト生成"]
+ IG["Layout-to-Image
画像生成"]
+ VG["Layout-guided
動画生成"]
+ end
+
+ DA --> |"構造情報"| DG
+ IA --> |"空間情報"| IG
+ VA --> |"時空間情報"| VG
+
+ DG --> |"BBox/マスク"| IG
+ IG --> |"フレーム生成"| VG
+```
+
+### 8.2 主要トレンド
+
+| トレンド | 説明 |
+|:---|:---|
+| **1. Diffusion Transformer (DiT) への移行** | U-Netベースの拡散モデルからTransformerベースへ。レイアウト生成(LayoutDM→UniLayDiff)、画像生成(InstanceAssemble)、動画生成(HunyuanVideo等)すべてで進行 |
+| **2. LLMのレイアウトプランナー化** | LayoutGPT、PosterLlama、VideoDirectorGPTに見られるように、LLMがデザイン知識を活用してレイアウトを「設計」する手法が台頭 |
+| **3. ハイブリッドアプローチ** | LLM(高レベルプランニング)+ Diffusion Model(高品質合成)の組み合わせが2025年のSOTA |
+| **4. Flow Matchingの台頭** | LayoutFlowやSLayRに見られる、拡散モデルの代替としてのFlow Matching。よりシンプルで高速 |
+| **5. 統一モデルの志向** | 複数のレイアウトタスク(生成、補完、条件付き生成等)を1つのモデルで処理(UniLayDiff, LGGPT, DLAFormer) |
+| **6. コンテンツ認識の強化** | 背景画像やテキスト内容を考慮したレイアウト生成(PosterLlama, CreatiLayout) |
+| **7. エンタープライズRAGパイプライン統合** | Docling, MinerU等による文書レイアウト分析→LLM/RAGへの直接的な統合 |
+| **8. Agentic Document AI** | VLM(Qwen2.5-VL等)がレイアウト理解だけでなく、ツール使用や自律的な判断を行うエージェント的振る舞い |
+
+### 8.3 アーキテクチャの進化タイムライン
+
+```mermaid
+timeline
+ title レイアウトモデルの進化
+ 2019 : LayoutGAN
+ : LayoutVAE
+ 2020 : LayoutLM
+ 2021 : LayoutTransformer
+ : LayoutLMv2
+ : LayoutParser
+ 2022 : LayoutLMv3
+ : BLT (Google)
+ 2023 : LayoutDM
+ : LayoutGPT
+ : GLIGEN
+ : LayoutFormer++
+ 2024 : DocLayout-YOLO
+ : DLAFormer
+ : LayoutFlow
+ : PosterLlama
+ : VideoDirectorGPT
+ : TrailBlazer
+ : MotionCtrl
+ 2025 : UniLayDiff
+ : CreatiLayout
+ : InstanceAssemble
+ : GenProp
+ : Heron-101
+ : StructLayoutFormer
+```
+
+### 8.4 選定ガイド
+
+> [!IMPORTANT]
+> 用途に応じた推奨モデル選択
+
+| やりたいこと | 推奨モデル/ツール |
+|:---|:---|
+| **文書からレイアウト要素を高速検出** | DocLayout-YOLO, Heron-101 (Docling) |
+| **文書のセマンティック理解(フォーム抽出、VQA)** | LayoutLMv3, Qwen2.5-VL |
+| **文書を構造化してRAG/LLMに投入** | Docling, MinerU, Unstructured.io |
+| **ポスター・グラフィックのレイアウト自動生成** | PosterLlama, LayoutDM, LayoutFlow |
+| **UI/Webレイアウトの自動生成** | LayoutGPT, Galileo AI, Uizard |
+| **レイアウト指定から画像生成** | InstanceAssemble, CreatiLayout, GLIGEN |
+| **レイアウト制御付き動画生成** | VideoDirectorGPT, TrailBlazer, MotionCtrl |
+| **商用文書処理パイプライン** | Azure AI Document Intelligence, Google Document AI, AWS Textract |
+
+---
+
+> [!NOTE]
+> 本サーベイは2025年5月時点の情報に基づいています。この分野は急速に進展しており、新しいモデルやベンチマークが継続的に登場しています。
diff --git a/docs/AI_Computer_Vision/layout_models_survey.rag_ingested b/docs/AI_Computer_Vision/layout_models_survey.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..34e1c4abc02b58842f3967fdd5b0e92e2eac64a9
--- /dev/null
+++ b/docs/AI_Computer_Vision/layout_models_survey.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "d85e0629dd23f2bf", "ingested_at": "2026-06-30T08:49:26.140736", "content_hash": "d85e0629dd23f2bf"}
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/script_long.md b/docs/AI_Computer_Vision/script_long.md
new file mode 100644
index 0000000000000000000000000000000000000000..28220ce9f5cbaeda08604eac72e0530bab5c45ef
--- /dev/null
+++ b/docs/AI_Computer_Vision/script_long.md
@@ -0,0 +1,26 @@
+Good evening, and welcome back. Tonight we're diving into a fascinating new development in the field of artificial intelligence, specifically in computer vision. A recent surge of research papers has been published on arXiv, an online repository for scientific papers, and these breakthroughs are shaping the future of how machines understand and interact with the world around us.
+
+Let’s take a closer look at some of these groundbreaking advancements:
+
+First, we have "EventHub," which proposes a new approach to building event-based stereo networks. These networks can be used for various applications, from autonomous vehicles to robotics, and EventHub's innovative design promises increased efficiency and accuracy.
+
+Next up is “ActionParty,” a multi-subject action binding system designed for generative video games. This could revolutionize the way we create immersive gaming experiences by allowing players to interact with characters in more dynamic and realistic ways.
+
+Then there’s "Generative World Renderer," which focuses on creating realistic 3D environments using AI. This technology has the potential to transform fields like film, animation, and architectural visualization.
+
+We also see a focus on anomaly detection through “Modulate-and-Map,” which utilizes crossmodal feature mapping with cross-view modulation for enhanced accuracy in identifying anomalies. This could be crucial for various applications such as security systems or medical imaging analysis.
+
+The field of visual representation is also being pushed forward with "Steerable Visual Representations," offering a new way to control and manipulate images using AI. This could have implications for fields like augmented reality, where precise image manipulation would be essential.
+
+Beyond these advancements in computer vision, we're seeing breakthroughs in understanding language and the world around us through “Scenario Comprehension Visual Grounding.” This research aims to bridge the gap between visual information and human understanding of narratives.
+
+The field of video understanding is also advancing with "A Simple Baseline for Streaming Video Understanding," which focuses on creating a more efficient way to analyze live video streams, potentially revolutionizing applications like real-time object tracking or automated surveillance.
+
+And finally, we have “VOID,” which proposes a new method for deleting objects and interactions from videos in a controlled manner. This could be invaluable for applications such as privacy protection or content editing.
+
+
+These are just a few of the exciting developments that promise to reshape our interaction with technology in the coming years. As these technologies continue to mature, we can expect even more innovative breakthroughs that will change how we live and work.
+
+
+
+This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/script_long.rag_ingested b/docs/AI_Computer_Vision/script_long.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..b0870b8ac63654f0cc0120110317df2edc58fac5
--- /dev/null
+++ b/docs/AI_Computer_Vision/script_long.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "176cc3fb0aea221e", "ingested_at": "2026-06-30T08:49:26.364957", "content_hash": "176cc3fb0aea221e"}
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/script_long_clean.txt b/docs/AI_Computer_Vision/script_long_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..06d00a4b5fd714abc6c29d61f8dd32e7db726d12
--- /dev/null
+++ b/docs/AI_Computer_Vision/script_long_clean.txt
@@ -0,0 +1,23 @@
+Good evening, and welcome back. Tonight we're diving into a fascinating new development in the field of artificial intelligence, specifically in computer vision. A recent surge of research papers has been published on arXiv, an online repository for scientific papers, and these breakthroughs are shaping the future of how machines understand and interact with the world around us.
+
+Let’s take a closer look at some of these groundbreaking advancements:
+
+First, we have "EventHub," which proposes a new approach to building event-based stereo networks. These networks can be used for various applications, from autonomous vehicles to robotics, and EventHub's innovative design promises increased efficiency and accuracy.
+
+Next up is “ActionParty,” a multi-subject action binding system designed for generative video games. This could revolutionize the way we create immersive gaming experiences by allowing players to interact with characters in more dynamic and realistic ways.
+
+Then there’s "Generative World Renderer," which focuses on creating realistic 3D environments using AI. This technology has the potential to transform fields like film, animation, and architectural visualization.
+
+We also see a focus on anomaly detection through “Modulate-and-Map,” which utilizes crossmodal feature mapping with cross-view modulation for enhanced accuracy in identifying anomalies. This could be crucial for various applications such as security systems or medical imaging analysis.
+
+The field of visual representation is also being pushed forward with "Steerable Visual Representations," offering a new way to control and manipulate images using AI. This could have implications for fields like augmented reality, where precise image manipulation would be essential.
+
+Beyond these advancements in computer vision, we're seeing breakthroughs in understanding language and the world around us through “Scenario Comprehension Visual Grounding.” This research aims to bridge the gap between visual information and human understanding of narratives.
+
+The field of video understanding is also advancing with "A Simple Baseline for Streaming Video Understanding," which focuses on creating a more efficient way to analyze live video streams, potentially revolutionizing applications like real-time object tracking or automated surveillance.
+
+And finally, we have “VOID,” which proposes a new method for deleting objects and interactions from videos in a controlled manner. This could be invaluable for applications such as privacy protection or content editing.
+
+These are just a few of the exciting developments that promise to reshape our interaction with technology in the coming years. As these technologies continue to mature, we can expect even more innovative breakthroughs that will change how we live and work.
+
+This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/script_middle.md b/docs/AI_Computer_Vision/script_middle.md
new file mode 100644
index 0000000000000000000000000000000000000000..80069e74597af21db1dcaa326d5cc86f2b0fae78
--- /dev/null
+++ b/docs/AI_Computer_Vision/script_middle.md
@@ -0,0 +1,12 @@
+Good evening. We're bringing you a look at some of the latest research published on arXiv, a platform for sharing scientific papers. Researchers have been busy, submitting over 35,000 new papers in just the past year. This includes work on everything from computer vision to video game development and even how we understand videos.
+
+Take a look at this selection of papers that were recently published: "EventHub" is an example of a study exploring event-based stereo networks without active sensors. Researchers are looking to create more efficient systems for understanding the world around us.
+
+Then there's "ActionParty," which focuses on multi-subject action binding in generative video games, and "Generative World Renderer." This paper explores how we can use AI to generate realistic 3D worlds.
+
+The list goes on, with papers exploring topics like crossmodal feature mapping for anomaly detection, steerable visual representations, and even scenario comprehension visual grounding.
+
+These are just a few examples of the cutting-edge research being conducted across various fields. You can find more details about these studies and others by visiting the arXiv website.
+
+
+This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/script_middle.rag_ingested b/docs/AI_Computer_Vision/script_middle.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..3c409121898873674f814ac292cbb80164f88127
--- /dev/null
+++ b/docs/AI_Computer_Vision/script_middle.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "e18e1b8265e3e95b", "ingested_at": "2026-06-30T08:49:26.562449", "content_hash": "e18e1b8265e3e95b"}
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/script_middle_clean.txt b/docs/AI_Computer_Vision/script_middle_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..ef3dc8837bcbee5ddca44f1fca4253bdfc084e98
--- /dev/null
+++ b/docs/AI_Computer_Vision/script_middle_clean.txt
@@ -0,0 +1,11 @@
+Good evening. We're bringing you a look at some of the latest research published on arXiv, a platform for sharing scientific papers. Researchers have been busy, submitting over 35,000 new papers in just the past year. This includes work on everything from computer vision to video game development and even how we understand videos.
+
+"EventHub" is an example of a study exploring event-based stereo networks without active sensors. Researchers are looking to create more efficient systems for understanding the world around us.
+
+Then there's "ActionParty," which focuses on multi-subject action binding in generative video games, and "Generative World Renderer." This paper explores how we can use AI to generate realistic 3D worlds.
+
+The list goes on, with papers exploring topics like crossmodal feature mapping for anomaly detection, steerable visual representations, and even scenario comprehension visual grounding.
+
+These are just a few examples of the cutting-edge research being conducted across various fields. You can find more details about these studies and others by visiting the arXiv website.
+
+This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/script_short.md b/docs/AI_Computer_Vision/script_short.md
new file mode 100644
index 0000000000000000000000000000000000000000..a946463954b1c0ef9db19ae26f129b32515671b1
--- /dev/null
+++ b/docs/AI_Computer_Vision/script_short.md
@@ -0,0 +1,6 @@
+New research is dropping faster than you can say "AI breakthrough." arXiv just released a massive batch of papers, over 35,000 in total! We're talking about everything from new ways to create realistic video games to groundbreaking advancements in how computers understand and interact with the world around them.
+
+There are tons of exciting breakthroughs here, including a new approach to understanding videos, a system for creating more lifelike avatars, and even a way to delete objects from video footage! Want to learn more? Check out the links on screen for the full research papers. This is just the tip of the iceberg – stay tuned for more updates as this field continues to explode!
+
+
+This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/script_short.rag_ingested b/docs/AI_Computer_Vision/script_short.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..16e052ad6cd8ee34d4245d73fc5fb25318d6a84c
--- /dev/null
+++ b/docs/AI_Computer_Vision/script_short.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "bce174ea1c9eda0a", "ingested_at": "2026-06-30T08:49:26.650814", "content_hash": "bce174ea1c9eda0a"}
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/script_short_clean.txt b/docs/AI_Computer_Vision/script_short_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..de3510da0cceca598429d76646ff802a86b6088a
--- /dev/null
+++ b/docs/AI_Computer_Vision/script_short_clean.txt
@@ -0,0 +1,5 @@
+New research is dropping faster than you can say "AI breakthrough." arXiv just released a massive batch of papers, over 35,000 in total! We're talking about everything from new ways to create realistic video games to groundbreaking advancements in how computers understand and interact with the world around them.
+
+There are tons of exciting breakthroughs here, including a new approach to understanding videos, a system for creating more lifelike avatars, and even a way to delete objects from video footage! Want to learn more? Check out the links on screen for the full research papers. This is just the tip of the iceberg – stay tuned for more updates as this field continues to explode!
+
+This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Computer_Vision/video_understanding_models_survey.md b/docs/AI_Computer_Vision/video_understanding_models_survey.md
new file mode 100644
index 0000000000000000000000000000000000000000..72f7a53940fe926c6770ef158ff9c1bad31bc165
--- /dev/null
+++ b/docs/AI_Computer_Vision/video_understanding_models_survey.md
@@ -0,0 +1,435 @@
+# 動画理解モデル サーベイレポート
+
+> **作成日**: 2026年5月23日
+> **対象期間**: 2020年〜2026年(2024年以降を重点的にカバー)
+
+---
+
+## 目次
+
+1. [概要](#概要)
+2. [技術的進化の歴史](#技術的進化の歴史)
+3. [主要アーキテクチャの分類](#主要アーキテクチャの分類)
+4. [商用モデル(Closed-Source)](#商用モデルclosed-source)
+5. [オープンソースモデル](#オープンソースモデル)
+6. [動画基盤モデル(Video Foundation Models)](#動画基盤モデルvideo-foundation-models)
+7. [専門プラットフォーム・API](#専門プラットフォームapi)
+8. [ベンチマークとリーダーボード](#ベンチマークとリーダーボード)
+9. [応用分野](#応用分野)
+10. [課題と限界](#課題と限界)
+11. [今後の研究方向性](#今後の研究方向性)
+12. [まとめと推奨](#まとめと推奨)
+
+---
+
+## 概要
+
+動画理解(Video Understanding)は、映像コンテンツから意味的な情報を抽出・推論するAI技術の総称である。2024年以降、大規模マルチモーダルモデル(LMM)の急速な発展により、動画理解は**短いクリップの認識**から**長時間映像の複雑な推論**へと大きくパラダイムシフトしている。
+
+### 動画理解のタスク体系
+
+```mermaid
+graph TD
+ A["動画理解
Video Understanding"] --> B["認識系タスク"]
+ A --> C["言語連携タスク"]
+ A --> D["時空間推論タスク"]
+ A --> E["生成・編集タスク"]
+
+ B --> B1["行動認識
Action Recognition"]
+ B --> B2["物体検出・追跡
Object Detection/Tracking"]
+ B --> B3["シーン分類
Scene Classification"]
+
+ C --> C1["動画キャプション生成
Video Captioning"]
+ C --> C2["動画質問応答
Video QA"]
+ C --> C3["動画要約
Video Summarization"]
+
+ D --> D1["時間的グラウンディング
Temporal Grounding"]
+ D --> D2["因果推論
Causal Reasoning"]
+ D --> D3["物理法則理解
Physical Reasoning"]
+
+ E --> E1["動画検索
Video Retrieval"]
+ E --> E2["動画セグメンテーション
Video Segmentation"]
+ E --> E3["動画編集支援
Video Editing"]
+```
+
+---
+
+## 技術的進化の歴史
+
+動画理解モデルは、CNN時代からTransformerベースのネイティブマルチモーダルモデルへと劇的に進化してきた。
+
+### タイムライン
+
+| 時期 | 主な進展 | コアアーキテクチャ | 代表モデル |
+|:-----|:---------|:-------------------|:-----------|
+| **2020–2021** | 動き・時間一貫性の学習、2D画像TransformerをVideoへ拡張 | CNN / Hybridモデル | TimeSformer, ViViT, Video Swin |
+| **2022** | 自己教師あり事前学習の確立、マルチモーダル統合の萌芽 | Masked Autoencoder | VideoMAE, InternVideo |
+| **2023** | Video-Language Modelの台頭、動画を"対話"で理解 | Vision Encoder + LLM | Video-ChatGPT, Video-LLaMA |
+| **2024** | Diffusion Transformer革命、大規模VideoLLMのスケーリング | DiT / 大規模VLM | InternVideo2, VideoPrism, Qwen2-VL |
+| **2025** | ネイティブマルチモーダル化、長時間動画推論、物理世界理解 | Unified LLM | Gemini 2.5 Pro, Qwen2.5-VL, InternVL3 |
+| **2026** | World Foundation Model、リアルタイム映像エージェント | MoE + Native Multimodal | Kimi K2.5, Qwen3-Omni, Gemini Omni Flash |
+
+### パラダイムの変遷
+
+```mermaid
+flowchart LR
+ A["ピクセルの分析
(2020-2021)"] --> B["マルチモーダル統合
(2022-2023)"]
+ B --> C["スケーリングと推論
(2024-2025)"]
+ C --> D["世界理解
(2026~)"]
+
+ style A fill:#ff6b6b,color:#fff
+ style B fill:#ffa502,color:#fff
+ style C fill:#1e90ff,color:#fff
+ style D fill:#2ed573,color:#fff
+```
+
+---
+
+## 主要アーキテクチャの分類
+
+現在の動画理解モデルは、大きく3つのアーキテクチャパラダイムに分類できる。
+
+### 1. Vision Encoder + Projector + LLM パターン
+
+最も広く採用されているアプローチ。事前学習済みのVision Encoder(CLIP、SigLIP等)とLLMをProjection層で接続する。
+
+```
+動画フレーム → [Vision Encoder] → 視覚トークン → [Projector/MLP] → [LLM] → テキスト出力
+```
+
+**代表モデル**: LLaVA-Video, Video-ChatGPT, mPLUG-Owl
+
+> [!NOTE]
+> この方式はモジュール性が高く、既存のLLMを活用できる利点があるが、Vision EncoderとLLMの間に「モダリティギャップ」が生じやすい。
+
+### 2. ViT-MLP-LLM パターン(強化版統合)
+
+Vision Transformerの出力をMLP層で変換し、LLMに入力する方式。位置エンコーディングや学習戦略に独自の工夫を施す。
+
+**代表モデル**: InternVL3(V2PE位置エンコーディング), Qwen2.5-VL(MRoPE + Window Attention)
+
+| モデル | Vision Encoder | 位置エンコーディング | 注意機構 |
+|:-------|:---------------|:---------------------|:---------|
+| **Qwen2.5-VL** | ViT(Window Attention) | MRoPE + 絶対時間エンコーディング | Window + Full Attention |
+| **InternVL3** | InternViT-6B | V2PE(可変視覚位置) | Standard Attention |
+| **LLaVA-Video** | CLIP/LanguageBind | Causal Attention Mask | Standard Attention |
+
+### 3. ネイティブマルチモーダルアーキテクチャ
+
+テキスト・画像・動画・音声のトークンを統一バックボーンで処理する。モダリティ間のアラインメントが学習時から自然に行われる。
+
+**代表モデル**: Gemini 2.5 Pro, Qwen3-Omni, Gemma 4
+
+> [!IMPORTANT]
+> 2025〜2026年のトレンドとして、「後付けでVisonを統合する」方式から、「最初からマルチモーダルで事前学習する」ネイティブ方式への移行が加速している。InternVL3の「Native Multimodal Pre-Training」はこの方向性の代表例。
+
+---
+
+## 商用モデル(Closed-Source)
+
+### Google — Gemini シリーズ
+
+| モデル | リリース | 動画理解の特徴 |
+|:-------|:---------|:---------------|
+| **Gemini 1.5 Pro** | 2024年2月 | 100万トークンコンテキスト、1時間超の動画処理 |
+| **Gemini 2.5 Pro** | 2025年3月 | ネイティブ音声・映像統合、~6時間の動画処理、Video-MME 84.8% |
+| **Gemini Omni Flash** | 2026年 | リアルタイムVideo-to-Video対話、映像エージェント |
+
+- **強み**: 最長のコンテキストウィンドウ、ネイティブ音声映像統合、APIの充実
+- **動画処理方式**: フレームサンプリング + 音声トラック統合、`low` 解像度パラメータで長時間処理
+
+### OpenAI — GPT シリーズ
+
+| モデル | リリース | Video-MME スコア | 備考 |
+|:-------|:---------|:-----------------|:-----|
+| **GPT-4o** | 2024年5月 | 65.3% | フレーム単位の画像入力として動画処理 |
+| **GPT-4.1** | 2025年4月 | 72.0% | 100万トークンコンテキスト、コーディング特化 |
+| **o3/o4-mini** | 2025年 | — | 推論特化、視覚入力対応 |
+
+- **特徴**: 直接的なネイティブ動画処理はGeminiほど強くない。フレーム抽出ベースの処理が中心
+- **強み**: 強力なテキスト推論能力を動画QAに活用
+
+### Anthropic — Claude シリーズ
+
+- **Claude 4 / Opus 4.6**: 画像理解には強いが、動画のネイティブ処理については後発
+- PDF、画像の分析には優れるが、長時間動画の直接入力は限定的
+
+### Moonshot AI — Kimi K2.5
+
+- **Video-MME 1位(87.4%)**: 2026年5月時点でリーダーボード首位
+- 中国発のモデルで、動画理解に特に強いマルチモーダルLLM
+
+---
+
+## オープンソースモデル
+
+### Tier 1: 最先端オープンウェイトモデル
+
+| モデル | 組織 | パラメータ | 特徴 | ライセンス |
+|:-------|:-----|:-----------|:-----|:-----------|
+| **Qwen3-Omni** | Alibaba/Qwen | — | ネイティブ音声映像統合、SOTA級性能 | Apache 2.0 |
+| **Qwen2.5-VL** | Alibaba/Qwen | 3B/7B/72B | 動的FPSサンプリング、MRoPE、時間グラウンディング | Apache 2.0 |
+| **InternVL3** | Shanghai AI Lab | 1B〜78B | V2PE、ネイティブマルチモーダル事前学習 | Apache 2.0 |
+| **Gemma 4** | Google | 各サイズ | エッジ/モバイル展開向け、効率重視 | Gemma License |
+| **Llama 4 Scout** | Meta | — | 1000万トークンコンテキスト、長時間動画向け | Llama License |
+
+### Tier 2: 特化型・研究モデル
+
+| モデル | 組織 | 特徴 |
+|:-------|:-----|:-----|
+| **LLaVA-Video** | UW-Madison + ByteDance | CLIP + LLM統合フレームワーク、高いカスタマイズ性 |
+| **Video-LLaMA 2** | DAMO Academy | 音声・視覚のデュアルブランチ処理 |
+| **mPLUG-Owl** | DAMO Academy | マルチモーダル対話、細粒度理解 |
+| **VideoChat** | Shanghai AI Lab | 対話型動画理解 |
+| **ShareGPT4Video** | — | 高品質動画キャプションデータ生成 |
+
+### アーキテクチャ詳細比較
+
+#### Qwen2.5-VL の特徴
+
+```mermaid
+flowchart TD
+ V["動画入力"] --> DS["動的FPSサンプリング
Dynamic FPS"]
+ DS --> NR["ネイティブ解像度処理"]
+ NR --> WA["Window Attention
(計算コスト線形スケール)"]
+ WA --> MR["MRoPE
絶対時間エンコーディング"]
+ MR --> LLM["Qwen LLM"]
+ LLM --> OUT["テキスト出力
+ 時間グラウンディング"]
+```
+
+- **動的FPSサンプリング**: 動画の動きの量に応じてフレーム数を動的に調整
+- **MRoPE**: 時間・空間の位置情報を絶対タイムスタンプとして統合
+- **Window Attention**: パッチ数に対し線形にスケールする計算コスト
+
+#### InternVL3 の特徴
+
+- **V2PE(Variable Visual Position Encoding)**: 視覚トークンの位置増分を柔軟に設定し、長コンテキスト理解を改善
+- **Native Multimodal Pre-Training**: 言語のみの事前学習→視覚の後付けではなく、最初からインターリーブドマルチモーダルデータで学習
+- **ツール使用能力**: ブリッジングモジュールなしで複雑なタスクを処理
+
+---
+
+## 動画基盤モデル(Video Foundation Models)
+
+LLMとの接続ではなく、動画そのものの表現学習に特化したモデル群。
+
+### VideoMAE(2022年)
+
+- **手法**: Masked Autoencoder(MAE)の動画拡張。チューブマスキング + 極端に高いマスク率(90%以上)
+- **意義**: 小規模データセット(Kinetics-400)でも効率的な自己教師あり学習が可能であることを実証
+- **用途**: 行動認識、動画分類の事前学習済みエンコーダとして広く利用
+
+### VideoPrism(2024年、Google Research)
+
+- **手法**: 2段階学習 — ① コントラスティブ学習(動画-テキストアラインメント)→ ② Masked Video Modeling
+- **データ規模**: 3600万の高品質動画-テキストペア + 5億8200万の動画クリップ
+- **特徴**: 単一の「凍結」モデルで分類・検索・QAなど多様なタスクでSOTA
+
+### InternVideo シリーズ(2022–2025年)
+
+| バージョン | 年 | 主な特徴 |
+|:-----------|:---|:---------|
+| **InternVideo** | 2022/2023 | 生成的(マスクモデリング)+ 判別的(コントラスティブ)学習の統合 |
+| **InternVideo2** | 2024 | 段階的学習:マスク再構成→クロスモーダル対照→次トークン予測、6Bパラメータ |
+| **InternVideo2.5** | 2025年初 | Video MLLMの長コンテキストモデリング強化 |
+| **InternVideo-Next** | 2025年末 | 意味的事前知識を用いた拡張動画再構成フレームワーク |
+
+---
+
+## 専門プラットフォーム・API
+
+### Twelve Labs
+
+動画ネイティブのエンタープライズAIプラットフォーム(NVIDIA出資)。
+
+| モデル | 機能 |
+|:-------|:-----|
+| **Marengo** | 動画エンコーダ。動画・音声・テキストを時空間埋め込みに変換し、検索可能にする |
+| **Pegasus** | 動画言語モデル。要約・QA・高次推論を実行 |
+
+- **用途**: エンタープライズ動画検索、コンテンツ発見、コンプライアンススキャン、編集ワークフロー
+
+### NVIDIA Cosmos
+
+NVIDIA の World Foundation Model スイート。Physical AI・ロボティクス向け。
+
+| コンポーネント | 機能 |
+|:---------------|:-----|
+| **Cosmos Predict** | 30秒の予測映像生成(世界モデル) |
+| **Cosmos Transfer** | シミュレーション→実世界変換 |
+| **Cosmos Reason** | 具現化推論用VLM、物理的常識理解 |
+
+### その他の注目プラットフォーム
+
+- **Microsoft Azure Video Indexer**: エンタープライズ向け動画分析
+- **AWS Rekognition Video**: 顔認識、ラベル検出、コンテンツモデレーション
+- **Alibaba Tongyi Qianwen API**: Qwenモデルベースの動画理解API
+
+---
+
+## ベンチマークとリーダーボード
+
+### 主要ベンチマーク一覧
+
+| ベンチマーク | 対象 | 動画長 | 特徴 |
+|:-------------|:-----|:-------|:-----|
+| **Video-MME** | 包括的評価 | 短〜1時間 | 6ドメイン、30サブフィールド、字幕・音声対応 |
+| **Video-MME-v2** | 堅牢性評価 | 多様 | 3段階階層、グループ評価で推測ペナルティ |
+| **MVBench** | マルチモーダル動画理解 | 主に短め | 多様なタスクカバレッジ |
+| **EgoSchema** | エゴセントリック動画 | 数分 | 一人称視点での長時間推論 |
+| **ActivityNet-QA** | アクティビティ理解 | 数分〜10分 | 実世界アクティビティのQA |
+| **PhysBench** | 物理法則理解 | 短い | 物理的常識と因果関係の評価 |
+| **VideoPhy** | 物理推論 | 短い | 物理法則への準拠度 |
+| **VideoScore** | 人間評価ベース | 多様 | 人間のフィードバック駆動 |
+
+### Video-MME リーダーボード(2026年5月時点)
+
+| 順位 | モデル | 組織 | スコア |
+|:-----|:-------|:-----|:-------|
+| 🥇 1 | **Kimi K2.5** | Moonshot AI | **87.4%** |
+| 🥈 2 | **Gemini 2.5 Pro** | Google | **84.8%** |
+| 🥉 3 | **Qwen3.6 Plus** | Alibaba/Qwen | **84.2%** |
+| 4 | Gemini 1.5 Pro | Google | 78.6% |
+| 5 | Gemini 1.5 Flash | Google | 76.1% |
+| 6 | Qwen3 VL 30B A3B Instruct | Alibaba/Qwen | 74.5% |
+| 7 | Qwen3 VL 30B A3B Thinking | Alibaba/Qwen | 73.3% |
+| 8 | Qwen3 VL 8B Thinking | Alibaba/Qwen | 71.8% |
+| 9 | Qwen3 VL 8B Instruct | Alibaba/Qwen | 71.4% |
+
+> [!WARNING]
+> ベンチマークスコアには「言語事前知識(Language Prior)」問題がある。モデルが動画を実際に「見ず」にテキストベースの手がかりや事前知識だけで正解できるケースが報告されている。Video-MME-v2はこの問題に対処するため、より厳密な評価戦略を採用している。
+
+---
+
+## 応用分野
+
+### 1. 映像監視・セキュリティ
+
+| 従来 | 現在(2025〜) |
+|:-----|:---------------|
+| 受動的な録画・再生 | 能動的なインテリジェント監視 |
+| クラウド依存の分析 | エッジAIによるリアルタイム分析 |
+| 動き検出のみ | 行動意図の分析(徘徊、共連れ検出) |
+| 手動検索 | 自然言語による映像検索 |
+
+### 2. 自動運転
+
+- **End-to-Endラーニング**: ルールベース→大量の実世界データからの学習へ
+- **因果推論**: 「歩行者が道路に出る*理由*」を理解し、将来の行動を予測
+- **合成データ**: 生成モデルで稀なエッジケースのシミュレーション環境を構築
+
+### 3. コンテンツモデレーション
+
+- **マルチモーダル分析**: 映像・音声・テキストの同時分析でコンテキスト依存の有害コンテンツを検出
+- **リアルタイム処理**: ライブストリームを200ms以下のレイテンシで処理
+- **ディープフェイク検出**: AI生成コンテンツの検出と透かし識別
+
+### 4. メディア・エンターテイメント
+
+- 動画要約・ハイライト自動生成
+- コンテンツ推薦システム
+- 自動字幕生成・翻訳
+- シーン分析とメタデータ自動付与
+
+### 5. 医療・ヘルスケア
+
+- 手術映像の分析・教育支援
+- 患者行動モニタリング
+- 医療画像の動画解析(超音波、内視鏡等)
+
+### 6. ロボティクス・Physical AI
+
+- NVIDIA Cosmos等による物理世界の理解と予測
+- ロボットの環境認識と行動計画
+- シミュレーション→実世界転移(Sim-to-Real)
+
+---
+
+## 課題と限界
+
+### 1. ハルシネーション(幻覚)
+
+動画理解におけるハルシネーションは、画像よりも**深刻**。時間軸の追加により以下の問題が発生する:
+
+- **動的歪み(Dynamic Distortion)**: 動きや因果関係の誤解
+- **コンテンツ捏造(Content Fabrication)**: 存在しないオブジェクトやシーケンスの記述
+- **根本原因**: 不十分な視覚的グラウンディング、動き非認識エンコーディング、静的な「画像事前知識」への依存
+
+### 2. トークン爆発(Token Explosion)
+
+- 1分間の高解像度動画 → 数十万の視覚トークン
+- フロンティアLLMのコンテキストウィンドウすら超過するケースがある
+- 計算コストとメモリの大幅な増大
+
+### 3. 長時間動画の「針の穴」問題
+
+- 長時間動画から特定の情報を見つけ出す効率が低い
+- 全動画を逐次処理する傾向 → 高レイテンシ、非効率
+
+### 4. 評価の課題
+
+- 従来のベンチマークは短いクリップに集中
+- 時間的一貫性、因果推論、長期的ナラティブ理解の測定が不十分
+- 言語事前知識問題により、真の視覚理解をテストできていないケースが存在
+
+### 5. 音声統合の未成熟
+
+- 多くのモデルは視覚情報のみに依存
+- エゴセントリック動画や会話動画では音声が重要な手がかり
+- 音声映像の統合はまだ発展途上
+
+---
+
+## 今後の研究方向性
+
+### 短期的(2026〜2027年)
+
+| 方向性 | 詳細 |
+|:-------|:-----|
+| **ハルシネーション軽減** | 訓練不要の手法(アクティベーション工学、自己診断的対照デコーディング、DINOv2による視覚的顕著性)、反事実学習 |
+| **効率的な長時間処理** | 「全てを見る」→「重要な部分を検索する」パラダイム。反復的ズームイン、スパースメモリ、階層的トークン圧縮 |
+| **エージェント型フレームワーク** | 動画を「スキム」「早送り」し、重要な時間領域に動的に計算を割り当てるエージェント的アプローチ |
+
+### 中期的(2027〜2028年)
+
+| 方向性 | 詳細 |
+|:-------|:-----|
+| **統合マルチモーダル推論** | 音声・テキスト・動画をより効果的に融合するネイティブアーキテクチャ |
+| **World Foundation Model** | 「何が起きているか」だけでなく、「次に何が起きるか」を予測する世界モデル |
+| **MoE(Mixture of Experts)の普及** | 巨大なパラメータ数でもアクティブパラメータを抑え、コンシューマGPUでの実行を可能にする |
+
+### 長期的ビジョン
+
+- **Physical AI**: 物理法則を理解し、実世界で行動するAIシステム
+- **動画理解と生成の融合**: 理解モデルと生成モデルの区別がなくなる
+- **人間-機械協調**: 完全自動化が困難な長期ナラティブ理解における人間との協調システム
+
+---
+
+## まとめと推奨
+
+### 用途別推奨モデル
+
+| 用途 | 推奨モデル | 理由 |
+|:-----|:-----------|:-----|
+| **最高精度(コスト問わず)** | Kimi K2.5 / Gemini 2.5 Pro | Video-MME トップスコア |
+| **プロダクション級動画理解** | Qwen3-Omni / Qwen2.5-VL | オープンウェイト、高性能、Apache 2.0 |
+| **長時間動画分析** | Gemini 2.5 Pro / Llama 4 Scout | 超長コンテキストウィンドウ |
+| **エッジ/モバイル展開** | Gemma 4 / Qwen2.5-VL 3B | 効率重視、小型モデル |
+| **エンタープライズ動画検索** | Twelve Labs (Marengo + Pegasus) | 専用API、検索最適化 |
+| **ロボティクス/Physical AI** | NVIDIA Cosmos | 物理世界理解、Sim-to-Real |
+| **研究・カスタマイズ** | LLaVA-Video / InternVL3 | モジュール性、拡張性 |
+| **動画基盤エンコーダ** | VideoPrism / InternVideo2.5 | 下流タスクの事前学習済みエンコーダ |
+
+### 2026年の動画理解の現状を一言で
+
+> **「ピクセルの分析」から「世界の推論」へ** — 動画理解モデルは短いクリップの認識ツールから、長時間映像の因果推論・物理理解・リアルタイムエージェントへと進化した。ネイティブマルチモーダルアーキテクチャとMoEによる効率化が主要トレンドである。
+
+---
+
+> [!TIP]
+> **さらに深く調べたい場合のリソース**:
+> - **Awesome-LLMs-for-Video-Understanding**: コミュニティ維持のモデル・論文・ベンチマーク集約リポジトリ
+> - **Video-MME / Video-MME-v2 リーダーボード**: 最新のモデルランキング
+> - **LLM Stats (llm-stats.com)**: リアルタイムのベンチマーク追跡
+> - **Voxel51 ブログ**: 動画AI技術の業界動向追跡
diff --git a/docs/AI_Computer_Vision/video_understanding_models_survey.rag_ingested b/docs/AI_Computer_Vision/video_understanding_models_survey.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..eb9ecb8138e7f3adc0c5c9ebc031e7a006afec80
--- /dev/null
+++ b/docs/AI_Computer_Vision/video_understanding_models_survey.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "aeda5a7889b16734", "ingested_at": "2026-06-30T08:49:30.186163", "content_hash": "aeda5a7889b16734"}
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405.json b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405.json
new file mode 100644
index 0000000000000000000000000000000000000000..8f23e7cef1c36bd5cdd8cdb7f12e4d0fe0ab3e65
--- /dev/null
+++ b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405.json
@@ -0,0 +1,5 @@
+{
+ "importance": 10,
+ "urgency": 10,
+ "tier": 1
+}
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405.md b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405.md
new file mode 100644
index 0000000000000000000000000000000000000000..0646a0afdf58d58536c48a77e84699f931e7da9d
--- /dev/null
+++ b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405.md
@@ -0,0 +1,157948 @@
+# arXiv AI / ML — one-year window (index)
+
+_Categories (OR):_ `cs.AI`, `cs.LG`, `cs.CL`, `stat.ML`, `cs.NE`
+
+_UTC `submittedDate` window:_ 2025-04-05 → 2026-04-05 · _weekly API slices_
+
+_Weekly slices are disjoint;_ sum of per-slice `totalResults` matches unique count here. _Unique papers downloaded:_ 90178 _(single-query `totalResults` for the same window was 90183 at fetch time; small mismatch is normal)._
+
+_Fetched at (UTC):_ 2026-04-05T06:49:49Z
+
+_Machine-readable twin (full abstracts):_ `arxiv-ai-ml-20250405-20260405.json`
+
+_Abstracts: JSON only._
+
+---
+
+## 1. ActionParty: Multi-Subject Action Binding in Generative Video Games
+
+- **ID:** `2604.02330` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Alexander Pondaven, Ziyi Wu, Igor Gilitschenski, et al.
+- **abs:** https://arxiv.org/abs/2604.02330 · **PDF:** https://arxiv.org/pdf/2604.02330v1
+
+## 2. Steerable Visual Representations
+
+- **ID:** `2604.02327` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Jona Ruthardt, Manu Gaur, Deva Ramanan, et al.
+- **abs:** https://arxiv.org/abs/2604.02327 · **PDF:** https://arxiv.org/pdf/2604.02327v1
+
+## 3. Grounded Token Initialization for New Vocabulary in LMs for Generative Recommendation
+
+- **ID:** `2604.02324` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Daiwei Chen, Zhoutong Fu, Chengming Jiang, et al.
+- **abs:** https://arxiv.org/abs/2604.02324 · **PDF:** https://arxiv.org/pdf/2604.02324v1
+
+## 4. Batched Contextual Reinforcement: A Task-Scaling Law for Efficient Reasoning
+
+- **ID:** `2604.02322` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Bangji Yang, Hongbo Ma, Jiajun Fan, et al.
+- **abs:** https://arxiv.org/abs/2604.02322 · **PDF:** https://arxiv.org/pdf/2604.02322v1
+
+## 5. No Single Best Model for Diversity: Learning a Router for Sample Diversity
+
+- **ID:** `2604.02319` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Yuhan Liu, Fangyuan Xu, Vishakh Padmakumar, et al.
+- **abs:** https://arxiv.org/abs/2604.02319 · **PDF:** https://arxiv.org/pdf/2604.02319v1
+
+## 6. Beyond the Assistant Turn: User Turn Generation as a Probe of Interaction Awareness in Language Models
+
+- **ID:** `2604.02315` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Sarath Shekkizhar, Romain Cosentino, Adam Earle
+- **abs:** https://arxiv.org/abs/2604.02315 · **PDF:** https://arxiv.org/pdf/2604.02315v1
+
+## 7. Topological Effects in Neural Network Field Theory
+
+- **ID:** `2604.02313` · **primary:** `hep-th` · **Submitted:** 2026-04-02
+- **Authors:** Christian Ferko, James Halverson, Vishnu Jejjala, et al.
+- **abs:** https://arxiv.org/abs/2604.02313 · **PDF:** https://arxiv.org/pdf/2604.02313v1
+
+## 8. go-$m$HC: Direct Parameterization of Manifold-Constrained Hyper-Connections via Generalized Orthostochastic Matrices
+
+- **ID:** `2604.02309` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Torque Dandachi, Sophia Diggs-Galligan
+- **abs:** https://arxiv.org/abs/2604.02309 · **PDF:** https://arxiv.org/pdf/2604.02309v1
+
+## 9. VOID: Video Object and Interaction Deletion
+
+- **ID:** `2604.02296` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Saman Motamed, William Harvey, Benjamin Klein, et al.
+- **abs:** https://arxiv.org/abs/2604.02296 · **PDF:** https://arxiv.org/pdf/2604.02296v1
+
+## 10. Taming the Exponential: A Fast Softmax Surrogate for Integer-Native Edge Inference
+
+- **ID:** `2604.02292` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Dimitrios Danopoulos, Enrico Lupi, Michael Kagan, et al.
+- **abs:** https://arxiv.org/abs/2604.02292 · **PDF:** https://arxiv.org/pdf/2604.02292v1
+
+## 11. Omni123: Exploring 3D Native Foundation Models with Limited 3D Data by Unifying Text to 2D and 3D Generation
+
+- **ID:** `2604.02289` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Chongjie Ye, Cheng Cao, Chuanyu Pan, et al.
+- **abs:** https://arxiv.org/abs/2604.02289 · **PDF:** https://arxiv.org/pdf/2604.02289v1
+
+## 12. Unifying Group-Relative and Self-Distillation Policy Optimization via Sample Routing
+
+- **ID:** `2604.02288` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Gengsheng Li, Tianyu Yang, Junfeng Fang, et al.
+- **abs:** https://arxiv.org/abs/2604.02288 · **PDF:** https://arxiv.org/pdf/2604.02288v1
+
+## 13. Novel Memory Forgetting Techniques for Autonomous AI Agents: Balancing Relevance and Efficiency
+
+- **ID:** `2604.02280` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Payal Fofadiya, Sunil Tiwari
+- **abs:** https://arxiv.org/abs/2604.02280 · **PDF:** https://arxiv.org/pdf/2604.02280v1
+
+## 14. The Self Driving Portfolio: Agentic Architecture for Institutional Asset Management
+
+- **ID:** `2604.02279` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Andrew Ang, Nazym Azimbayev, Andrey Kim
+- **abs:** https://arxiv.org/abs/2604.02279 · **PDF:** https://arxiv.org/pdf/2604.02279v1
+
+## 15. De Jure: Iterative LLM Self-Refinement for Structured Extraction of Regulatory Rules
+
+- **ID:** `2604.02276` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Keerat Guliani, Deepkamal Gill, David Landsman, et al.
+- **abs:** https://arxiv.org/abs/2604.02276 · **PDF:** https://arxiv.org/pdf/2604.02276v1
+
+## 16. Crystalite: A Lightweight Transformer for Efficient Crystal Modeling
+
+- **ID:** `2604.02270` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Tin Hadži Veljković, Joshua Rosenthal, Ivor Lončarić, et al.
+- **abs:** https://arxiv.org/abs/2604.02270 · **PDF:** https://arxiv.org/pdf/2604.02270v1
+
+## 17. SKILL0: In-Context Agentic Reinforcement Learning for Skill Internalization
+
+- **ID:** `2604.02268` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Zhengxi Lu, Zhiyuan Yao, Jinyang Wu, et al.
+- **abs:** https://arxiv.org/abs/2604.02268 · **PDF:** https://arxiv.org/pdf/2604.02268v1
+
+## 18. Model-Based Reinforcement Learning for Control under Time-Varying Dynamics
+
+- **ID:** `2604.02260` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Klemens Iten, Bruce Lee, Chenhao Li, et al.
+- **abs:** https://arxiv.org/abs/2604.02260 · **PDF:** https://arxiv.org/pdf/2604.02260v1
+
+## 19. Retrieval-Augmented Question Answering over Scientific Literature for the Electron-Ion Collider
+
+- **ID:** `2604.02259` · **primary:** `hep-ex` · **Submitted:** 2026-04-02
+- **Authors:** Tina. J. Jat, T. Ghosh, Karthik Suresh
+- **abs:** https://arxiv.org/abs/2604.02259 · **PDF:** https://arxiv.org/pdf/2604.02259v1
+
+## 20. Best-Arm Identification with Noisy Actuation
+
+- **ID:** `2604.02255` · **primary:** `cs.IT` · **Submitted:** 2026-04-02
+- **Authors:** Merve Karakas, Osama Hanna, Lin F. Yang, et al.
+- **abs:** https://arxiv.org/abs/2604.02255 · **PDF:** https://arxiv.org/pdf/2604.02255v1
+
+## 21. Smoothing the Landscape: Causal Structure Learning via Diffusion Denoising Objectives
+
+- **ID:** `2604.02250` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Hao Zhu, Di Zhou, Donna Slonim
+- **abs:** https://arxiv.org/abs/2604.02250 · **PDF:** https://arxiv.org/pdf/2604.02250v1
+
+## 22. BVFLMSP : Bayesian Vertical Federated Learning for Multimodal Survival with Privacy
+
+- **ID:** `2604.02248` · **primary:** `stat.ML` · **Submitted:** 2026-04-02
+- **Authors:** Abhilash Kar, Basisth Saha, Tanmay Sen, et al.
+- **abs:** https://arxiv.org/abs/2604.02248 · **PDF:** https://arxiv.org/pdf/2604.02248v1
+
+## 23. (PAC-)Learning state machines from data streams: A generic strategy and an improved heuristic (Extended version)
+
+- **ID:** `2604.02244` · **primary:** `cs.FL` · **Submitted:** 2026-04-02
+- **Authors:** Robert Baumgartner, Sicco Verwer
+- **abs:** https://arxiv.org/abs/2604.02244 · **PDF:** https://arxiv.org/pdf/2604.02244v1
+
+## 24. Generative AI Spotlights the Human Core of Data Science: Implications for Education
+
+- **ID:** `2604.02238` · **primary:** `cs.CY` · **Submitted:** 2026-04-02
+- **Authors:** Nathan Taback
+- **abs:** https://arxiv.org/abs/2604.02238 · **PDF:** https://arxiv.org/pdf/2604.02238v1
+
+## 25. Do Emotions in Prompts Matter? Effects of Emotional Framing on Large Language Models
+
+- **ID:** `2604.02236` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Minda Zhao, Yutong Yang, Chufei Peng, et al.
+- **abs:** https://arxiv.org/abs/2604.02236 · **PDF:** https://arxiv.org/pdf/2604.02236v1
+
+## 26. Answering the Wrong Question: Reasoning Trace Inversion for Abstention in LLMs
+
+- **ID:** `2604.02230` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Abinitha Gourabathina, Inkit Padhi, Manish Nagireddy, et al.
+- **abs:** https://arxiv.org/abs/2604.02230 · **PDF:** https://arxiv.org/pdf/2604.02230v1
+
+## 27. When to ASK: Uncertainty-Gated Language Assistance for Reinforcement Learning
+
+- **ID:** `2604.02226` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Juarez Monteiro, Nathan Gavenski, Gianlucca Zuin, et al.
+- **abs:** https://arxiv.org/abs/2604.02226 · **PDF:** https://arxiv.org/pdf/2604.02226v1
+
+## 28. Impact of Multimodal and Conversational AI on Learning Outcomes and Experience
+
+- **ID:** `2604.02221` · **primary:** `cs.HC` · **Submitted:** 2026-04-02
+- **Authors:** Karan Taneja, Anjali Singh, Ashok K. Goel
+- **abs:** https://arxiv.org/abs/2604.02221 · **PDF:** https://arxiv.org/pdf/2604.02221v1
+
+## 29. VISTA: Visualization of Token Attribution via Efficient Analysis
+
+- **ID:** `2604.02217` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Syed Ahmed, Bharathi Vokkaliga Ganesh, Jagadish Babu P, et al.
+- **abs:** https://arxiv.org/abs/2604.02217 · **PDF:** https://arxiv.org/pdf/2604.02217v1
+
+## 30. Universal Hypernetworks for Arbitrary Models
+
+- **ID:** `2604.02215` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Xuanfeng Zhou
+- **abs:** https://arxiv.org/abs/2604.02215 · **PDF:** https://arxiv.org/pdf/2604.02215v1
+
+## 31. Multi-Agent Video Recommenders: Evolution, Patterns, and Open Challenges
+
+- **ID:** `2604.02211` · **primary:** `cs.IR` · **Submitted:** 2026-04-02
+- **Authors:** Srivaths Ranganathan, Abhishek Dharmaratnakar, Anushree Sinha, et al.
+- **abs:** https://arxiv.org/abs/2604.02211 · **PDF:** https://arxiv.org/pdf/2604.02211v1
+
+## 32. CV-18 NER: Augmented Common Voice for Named Entity Recognition from Arabic Speech
+
+- **ID:** `2604.02209` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Youssef Saidi, Haroun Elleuch, Fethi Bougares
+- **abs:** https://arxiv.org/abs/2604.02209 · **PDF:** https://arxiv.org/pdf/2604.02209v1
+
+## 33. Blinded Radiologist and LLM-Based Evaluation of LLM-Generated Japanese Translations of Chest CT Reports: Comparative Study
+
+- **ID:** `2604.02207` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Yosuke Yamagishi, Atsushi Takamatsu, Yasunori Hamaguchi, et al.
+- **abs:** https://arxiv.org/abs/2604.02207 · **PDF:** https://arxiv.org/pdf/2604.02207v1
+
+## 34. LEO: Graph Attention Network based Hybrid Multi Sensor Extended Object Fusion and Tracking for Autonomous Driving Applications
+
+- **ID:** `2604.02206` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Mayank Mayank, Bharanidhar Duraisamy, Florian Geiss
+- **abs:** https://arxiv.org/abs/2604.02206 · **PDF:** https://arxiv.org/pdf/2604.02206v1
+
+## 35. On the Role of Depth in the Expressivity of RNNs
+
+- **ID:** `2604.02201` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Maude Lizaire, Michael Rizvi-Martel, Éric Dupuis, et al.
+- **abs:** https://arxiv.org/abs/2604.02201 · **PDF:** https://arxiv.org/pdf/2604.02201v1
+
+## 36. Towards Position-Robust Talent Recommendation via Large Language Models
+
+- **ID:** `2604.02200` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Silin Du, Hongyan Liu
+- **abs:** https://arxiv.org/abs/2604.02200 · **PDF:** https://arxiv.org/pdf/2604.02200v1
+
+## 37. From High-Dimensional Spaces to Verifiable ODD Coverage for Safety-Critical AI-based Systems
+
+- **ID:** `2604.02198` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Thomas Stefani, Johann Maximilian Christensen, Elena Hoemann, et al.
+- **abs:** https://arxiv.org/abs/2604.02198 · **PDF:** https://arxiv.org/pdf/2604.02198v1
+
+## 38. Computing the Exact Pareto Front in Average-Cost Multi-Objective Markov Decision Processes
+
+- **ID:** `2604.02196` · **primary:** `eess.SY` · **Submitted:** 2026-04-02
+- **Authors:** Jiping Luo, Nikolaos Pappas
+- **abs:** https://arxiv.org/abs/2604.02196 · **PDF:** https://arxiv.org/pdf/2604.02196v1
+
+## 39. Neuro-RIT: Neuron-Guided Instruction Tuning for Robust Retrieval-Augmented Language Model
+
+- **ID:** `2604.02194` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Jaemin Kim, Jae O Lee, Sumyeong Ahn, et al.
+- **abs:** https://arxiv.org/abs/2604.02194 · **PDF:** https://arxiv.org/pdf/2604.02194v1
+
+## 40. Neural network methods for two-dimensional finite-source reflector design
+
+- **ID:** `2604.02184` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Roel Hacking, Lisa Kusch, Koondanibha Mitra, et al.
+- **abs:** https://arxiv.org/abs/2604.02184 · **PDF:** https://arxiv.org/pdf/2604.02184v1
+
+## 41. TRU: Targeted Reverse Update for Efficient Multimodal Recommendation Unlearning
+
+- **ID:** `2604.02183` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Zhanting Zhou, KaHou Tam, Ziqiang Zheng, et al.
+- **abs:** https://arxiv.org/abs/2604.02183 · **PDF:** https://arxiv.org/pdf/2604.02183v1
+
+## 42. The Expert Strikes Back: Interpreting Mixture-of-Experts Language Models at Expert Level
+
+- **ID:** `2604.02178` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Jeremy Herbst, Jae Hee Lee, Stefan Wermter
+- **abs:** https://arxiv.org/abs/2604.02178 · **PDF:** https://arxiv.org/pdf/2604.02178v1
+
+## 43. Adam's Law: Textual Frequency Law on Large Language Models
+
+- **ID:** `2604.02176` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Hongyuan Adam Lu, Z. L., Victor Wei, et al.
+- **abs:** https://arxiv.org/abs/2604.02176 · **PDF:** https://arxiv.org/pdf/2604.02176v1
+
+## 44. Quantifying Self-Preservation Bias in Large Language Models
+
+- **ID:** `2604.02174` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Matteo Migliarini, Joaquin Pereira Pizzini, Luca Moresca, et al.
+- **abs:** https://arxiv.org/abs/2604.02174 · **PDF:** https://arxiv.org/pdf/2604.02174v1
+
+## 45. Do Lexical and Contextual Coreference Resolution Systems Degrade Differently under Mention Noise? An Empirical Study on Scientific Software Mentions
+
+- **ID:** `2604.02171` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Atilla Kaan Alkan, Felix Grezes, Jennifer Lynn Bartlett, et al.
+- **abs:** https://arxiv.org/abs/2604.02171 · **PDF:** https://arxiv.org/pdf/2604.02171v1
+
+## 46. A Practical Two-Stage Framework for GPU Resource and Power Prediction in Heterogeneous HPC Systems
+
+- **ID:** `2604.02158` · **primary:** `cs.DC` · **Submitted:** 2026-04-02
+- **Authors:** Beste Oztop, Dhruva Kulkarni, Zhengji Zhao, et al.
+- **abs:** https://arxiv.org/abs/2604.02158 · **PDF:** https://arxiv.org/pdf/2604.02158v1
+
+## 47. AstroConcepts: A Large-Scale Multi-Label Classification Corpus for Astrophysics
+
+- **ID:** `2604.02156` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Atilla Kaan Alkan, Felix Grezes, Sergi Blanco-Cuaresma, et al.
+- **abs:** https://arxiv.org/abs/2604.02156 · **PDF:** https://arxiv.org/pdf/2604.02156v1
+
+## 48. Brief Is Better: Non-Monotonic Chain-of-Thought Budget Effects in Function-Calling Language Agents
+
+- **ID:** `2604.02155` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Xuan Qi
+- **abs:** https://arxiv.org/abs/2604.02155 · **PDF:** https://arxiv.org/pdf/2604.02155v1
+
+## 49. Auction-Based Online Policy Adaptation for Evolving Objectives
+
+- **ID:** `2604.02151` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Guruprerana Shabadi, Kaushik Mallik
+- **abs:** https://arxiv.org/abs/2604.02151 · **PDF:** https://arxiv.org/pdf/2604.02151v1
+
+## 50. Samplet limits and multiwavelets
+
+- **ID:** `2604.02150` · **primary:** `math.NA` · **Submitted:** 2026-04-02
+- **Authors:** Gianluca Giacchi, Michael Multerer, Jacopo Quizi
+- **abs:** https://arxiv.org/abs/2604.02150 · **PDF:** https://arxiv.org/pdf/2604.02150v1
+
+## 51. AEGIS: Adversarial Entropy-Guided Immune System -- Thermodynamic State Space Models for Zero-Day Network Evasion Detection
+
+- **ID:** `2604.02149` · **primary:** `cs.CR` · **Submitted:** 2026-04-02
+- **Authors:** Vickson Ferrel
+- **abs:** https://arxiv.org/abs/2604.02149 · **PDF:** https://arxiv.org/pdf/2604.02149v1
+
+## 52. TRACE-Bot: Detecting Emerging LLM-Driven Social Bots via Implicit Semantic Representations and AIGC-Enhanced Behavioral Patterns
+
+- **ID:** `2604.02147` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Zhongbo Wang, Zhiyu Lin, Zhu Wang, et al.
+- **abs:** https://arxiv.org/abs/2604.02147 · **PDF:** https://arxiv.org/pdf/2604.02147v1
+
+## 53. MTI: A Behavior-Based Temperament Profiling System for AI Agents
+
+- **ID:** `2604.02145` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Jihoon Jeong
+- **abs:** https://arxiv.org/abs/2604.02145 · **PDF:** https://arxiv.org/pdf/2604.02145v1
+
+## 54. Application of parametric Shallow Recurrent Decoder Network to magnetohydrodynamic flows in liquid metal blankets of fusion reactors
+
+- **ID:** `2604.02139` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** M. Lo Verso, C. Introini, E. Cervi, et al.
+- **abs:** https://arxiv.org/abs/2604.02139 · **PDF:** https://arxiv.org/pdf/2604.02139v1
+
+## 55. GaelEval: Benchmarking LLM Performance for Scottish Gaelic
+
+- **ID:** `2604.02135` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Peter Devine, William Lamb, Beatrice Alex, et al.
+- **abs:** https://arxiv.org/abs/2604.02135 · **PDF:** https://arxiv.org/pdf/2604.02135v1
+
+## 56. Intelligent Cloud Orchestration: A Hybrid Predictive and Heuristic Framework for Cost Optimization
+
+- **ID:** `2604.02131` · **primary:** `cs.DC` · **Submitted:** 2026-04-02
+- **Authors:** Heet Nagoriya, Komal Rohit
+- **abs:** https://arxiv.org/abs/2604.02131 · **PDF:** https://arxiv.org/pdf/2604.02131v1
+
+## 57. SEAL: An Open, Auditable, and Fair Data Generation Framework for AI-Native 6G Networks
+
+- **ID:** `2604.02128` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Sunder Ali Khowaja, Kapal Dev, Engin Zeydan, et al.
+- **abs:** https://arxiv.org/abs/2604.02128 · **PDF:** https://arxiv.org/pdf/2604.02128v1
+
+## 58. Gradient estimators for parameter inference in discrete stochastic kinetic models
+
+- **ID:** `2604.02121` · **primary:** `physics.comp-ph` · **Submitted:** 2026-04-02
+- **Authors:** Ludwig Burger, Annalena Kofler, Lukas Heinrich, et al.
+- **abs:** https://arxiv.org/abs/2604.02121 · **PDF:** https://arxiv.org/pdf/2604.02121v1
+
+## 59. AA-SVD : Anchored and Adaptive SVD for Large Language Model Compression
+
+- **ID:** `2604.02119` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Atul Kumar Sinha, François Fleuret
+- **abs:** https://arxiv.org/abs/2604.02119 · **PDF:** https://arxiv.org/pdf/2604.02119v1
+
+## 60. LLM-as-a-Judge for Time Series Explanations
+
+- **ID:** `2604.02118` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Preetham Sivalingam, Murari Mandal, Saurabh Deshpande, et al.
+- **abs:** https://arxiv.org/abs/2604.02118 · **PDF:** https://arxiv.org/pdf/2604.02118v1
+
+## 61. Reliable Control-Point Selection for Steering Reasoning in Large Language Models
+
+- **ID:** `2604.02113` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Haomin Zhuang, Hojun Yoo, Xiaonan Luo, et al.
+- **abs:** https://arxiv.org/abs/2604.02113 · **PDF:** https://arxiv.org/pdf/2604.02113v1
+
+## 62. Cross-Modal Visuo-Tactile Object Perception
+
+- **ID:** `2604.02108` · **primary:** `cs.RO` · **Submitted:** 2026-04-02
+- **Authors:** Anirvan Dutta, Simone Tasciotti, Claudia Cusseddu, et al.
+- **abs:** https://arxiv.org/abs/2604.02108 · **PDF:** https://arxiv.org/pdf/2604.02108v1
+
+## 63. CASHG: Context-Aware Stylized Online Handwriting Generation
+
+- **ID:** `2604.02103` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Jinsu Shin, Sungeun Hong, Jin Yeong Bak
+- **abs:** https://arxiv.org/abs/2604.02103 · **PDF:** https://arxiv.org/pdf/2604.02103v1
+
+## 64. Prosodic ABX: A Language-Agnostic Method for Measuring Prosodic Contrast in Speech Representations
+
+- **ID:** `2604.02102` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Haitong Sun, Stephen McIntosh, Kwanghee Choi, et al.
+- **abs:** https://arxiv.org/abs/2604.02102 · **PDF:** https://arxiv.org/pdf/2604.02102v1
+
+## 65. LatentUM: Unleashing the Potential of Interleaved Cross-Modal Reasoning via a Latent-Space Unified Model
+
+- **ID:** `2604.02097` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Jiachun Jin, Zetong Zhou, Xiao Yang, et al.
+- **abs:** https://arxiv.org/abs/2604.02097 · **PDF:** https://arxiv.org/pdf/2604.02097v1
+
+## 66. Optimizing RAG Rerankers with LLM Feedback via Reinforcement Learning
+
+- **ID:** `2604.02091` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Yuhang Wu, Xiangqing Shen, Fanfan Wang, et al.
+- **abs:** https://arxiv.org/abs/2604.02091 · **PDF:** https://arxiv.org/pdf/2604.02091v1
+
+## 67. Mining Instance-Centric Vision-Language Contexts for Human-Object Interaction Detection
+
+- **ID:** `2604.02071` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Soo Won Seo, KyungChae Lee, Hyungchan Cho, et al.
+- **abs:** https://arxiv.org/abs/2604.02071 · **PDF:** https://arxiv.org/pdf/2604.02071v1
+
+## 68. Diff-KD: Diffusion-based Knowledge Distillation for Collaborative Perception under Corruptions
+
+- **ID:** `2604.02061` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Pengcheng Lyu, Chaokun Zhang, Gong Chen, et al.
+- **abs:** https://arxiv.org/abs/2604.02061 · **PDF:** https://arxiv.org/pdf/2604.02061v1
+
+## 69. Ouroboros: Dynamic Weight Generation for Recursive Transformers via Input-Conditioned LoRA Modulation
+
+- **ID:** `2604.02051` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Jaber Jaber, Osama Jaber
+- **abs:** https://arxiv.org/abs/2604.02051 · **PDF:** https://arxiv.org/pdf/2604.02051v1
+
+## 70. Goose: Anisotropic Speculation Trees for Training-Free Speculative Decoding
+
+- **ID:** `2604.02047` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Tao Jin, Phuong Minh Nguyen, Naoya Inoue
+- **abs:** https://arxiv.org/abs/2604.02047 · **PDF:** https://arxiv.org/pdf/2604.02047v1
+
+## 71. BidirLM: From Text to Omnimodal Bidirectional Encoders by Adapting and Composing Causal LLMs
+
+- **ID:** `2604.02045` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Nicolas Boizard, Théo Deschamps-Berger, Hippolyte Gisserot-Boukhlef, et al.
+- **abs:** https://arxiv.org/abs/2604.02045 · **PDF:** https://arxiv.org/pdf/2604.02045v1
+
+## 72. Tracking the emergence of linguistic structure in self-supervised models learning from speech
+
+- **ID:** `2604.02043` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Marianne de Heer Kloots, Martijn Bentum, Hosein Mohebbi, et al.
+- **abs:** https://arxiv.org/abs/2604.02043 · **PDF:** https://arxiv.org/pdf/2604.02043v1
+
+## 73. Reinforcement Learning for Speculative Trading under Exploratory Framework
+
+- **ID:** `2604.02035` · **primary:** `q-fin.MF` · **Submitted:** 2026-04-02
+- **Authors:** Yun Zhao, Alex S. L. Tse, Harry Zheng
+- **abs:** https://arxiv.org/abs/2604.02035 · **PDF:** https://arxiv.org/pdf/2604.02035v1
+
+## 74. AI in Insurance: Adaptive Questionnaires for Improved Risk Profiling
+
+- **ID:** `2604.02034` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Diogo Silva, João Teixeira, Bruno Lima
+- **abs:** https://arxiv.org/abs/2604.02034 · **PDF:** https://arxiv.org/pdf/2604.02034v1
+
+## 75. IndoorCrowd: A Multi-Scene Dataset for Human Detection, Segmentation, and Tracking with an Automated Annotation Pipeline
+
+- **ID:** `2604.02032` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Sebastian-Ion Nae, Radu Moldoveanu, Alexandra Stefania Ghita, et al.
+- **abs:** https://arxiv.org/abs/2604.02032 · **PDF:** https://arxiv.org/pdf/2604.02032v1
+
+## 76. Rare-Aware Autoencoding: Reconstructing Spatially Imbalanced Data
+
+- **ID:** `2604.02031` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Alejandro Castañeda Garcia, Jan van Gemert, Daan Brinks, et al.
+- **abs:** https://arxiv.org/abs/2604.02031 · **PDF:** https://arxiv.org/pdf/2604.02031v1
+
+## 77. The Latent Space: Foundation, Evolution, Mechanism, Ability, and Outlook
+
+- **ID:** `2604.02029` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Xinlei Yu, Zhangquan Chen, Yongbo He, et al.
+- **abs:** https://arxiv.org/abs/2604.02029 · **PDF:** https://arxiv.org/pdf/2604.02029v1
+
+## 78. Why Gaussian Diffusion Models Fail on Discrete Data?
+
+- **ID:** `2604.02028` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Alexander Shabalin, Simon Elistratov, Viacheslav Meshchaninov, et al.
+- **abs:** https://arxiv.org/abs/2604.02028 · **PDF:** https://arxiv.org/pdf/2604.02028v1
+
+## 79. Systematic Analyses of Reinforcement Learning Controllers in Signalized Urban Corridors
+
+- **ID:** `2604.02025` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Xiaofei Song, Kerstin Eder, Jonathan Lawry, et al.
+- **abs:** https://arxiv.org/abs/2604.02025 · **PDF:** https://arxiv.org/pdf/2604.02025v1
+
+## 80. APEX: Agent Payment Execution with Policy for Autonomous Agent API Access
+
+- **ID:** `2604.02023` · **primary:** `cs.CR` · **Submitted:** 2026-04-02
+- **Authors:** Mohd Safwan Uddin, Mohammed Mouzam, Mohammed Imran, et al.
+- **abs:** https://arxiv.org/abs/2604.02023 · **PDF:** https://arxiv.org/pdf/2604.02023v1
+
+## 81. ATBench: A Diverse and Realistic Trajectory Benchmark for Long-Horizon Agent Safety
+
+- **ID:** `2604.02022` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Yu Li, Haoyu Luo, Yuejin Xie, et al.
+- **abs:** https://arxiv.org/abs/2604.02022 · **PDF:** https://arxiv.org/pdf/2604.02022v1
+
+## 82. Feature Weighting Improves Pool-Based Sequential Active Learning for Regression
+
+- **ID:** `2604.02019` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Dongrui Wu
+- **abs:** https://arxiv.org/abs/2604.02019 · **PDF:** https://arxiv.org/pdf/2604.02019v1
+
+## 83. Demographic Parity Tails for Regression
+
+- **ID:** `2604.02017` · **primary:** `stat.ML` · **Submitted:** 2026-04-02
+- **Authors:** Naht Sinh Le, Christophe Denis, Mohamed Hebiri
+- **abs:** https://arxiv.org/abs/2604.02017 · **PDF:** https://arxiv.org/pdf/2604.02017v1
+
+## 84. Optimizing Interventions for Agent-Based Infectious Disease Simulations
+
+- **ID:** `2604.02016` · **primary:** `cs.MA` · **Submitted:** 2026-04-02
+- **Authors:** Anja Wolpers, Johannes Ponge, Adelinde M. Uhrmacher
+- **abs:** https://arxiv.org/abs/2604.02016 · **PDF:** https://arxiv.org/pdf/2604.02016v1
+
+## 85. $k$NNProxy: Efficient Training-Free Proxy Alignment for Black-Box Zero-Shot LLM-Generated Text Detection
+
+- **ID:** `2604.02008` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Kahim Wong, Kemou Li, Haiwei Wu, et al.
+- **abs:** https://arxiv.org/abs/2604.02008 · **PDF:** https://arxiv.org/pdf/2604.02008v1
+
+## 86. Apriel-Reasoner: RL Post-Training for General-Purpose and Efficient Reasoning
+
+- **ID:** `2604.02007` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Rafael Pardinas, Ehsan Kamalloo, David Vazquez, et al.
+- **abs:** https://arxiv.org/abs/2604.02007 · **PDF:** https://arxiv.org/pdf/2604.02007v1
+
+## 87. ProCeedRL: Process Critic with Exploratory Demonstration Reinforcement Learning for LLM Agentic Reasoning
+
+- **ID:** `2604.02006` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Jingyue Gao, Yanjiang Guo, Xiaoshuai Chen, et al.
+- **abs:** https://arxiv.org/abs/2604.02006 · **PDF:** https://arxiv.org/pdf/2604.02006v1
+
+## 88. How and why does deep ensemble coupled with transfer learning increase performance in bipolar disorder and schizophrenia classification?
+
+- **ID:** `2604.02002` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Sara Petiton, Antoine Grigis, Benoit Dufumier, et al.
+- **abs:** https://arxiv.org/abs/2604.02002 · **PDF:** https://arxiv.org/pdf/2604.02002v1
+
+## 89. GenGait: A Transformer-Based Model for Human Gait Anomaly Detection and Normative Twin Generation
+
+- **ID:** `2604.01997` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Elisa Motta, Marta Lorenzini, Clara Mouawad, et al.
+- **abs:** https://arxiv.org/abs/2604.01997 · **PDF:** https://arxiv.org/pdf/2604.01997v1
+
+## 90. SAFE: Stepwise Atomic Feedback for Error correction in Multi-hop Reasoning
+
+- **ID:** `2604.01993` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Daeyong Kwon, Soyoung Yoon, Seung-won Hwang
+- **abs:** https://arxiv.org/abs/2604.01993 · **PDF:** https://arxiv.org/pdf/2604.01993v1
+
+## 91. Attention at Rest Stays at Rest: Breaking Visual Inertia for Cognitive Hallucination Mitigation
+
+- **ID:** `2604.01989` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Boyang Gong, Yu Zheng, Fanye Kong, et al.
+- **abs:** https://arxiv.org/abs/2604.01989 · **PDF:** https://arxiv.org/pdf/2604.01989v1
+
+## 92. SenseMath: Do LLMs Have Number Sense? Evaluating Shortcut Use, Judgment, and Generation
+
+- **ID:** `2604.01988` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Haomin Zhuang, Xiangqi Wang, Yili Shen, et al.
+- **abs:** https://arxiv.org/abs/2604.01988 · **PDF:** https://arxiv.org/pdf/2604.01988v1
+
+## 93. Curia-2: Scaling Self-Supervised Learning for Radiology Foundation Models
+
+- **ID:** `2604.01987` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Antoine Saporta, Baptiste Callard, Corentin Dancette, et al.
+- **abs:** https://arxiv.org/abs/2604.01987 · **PDF:** https://arxiv.org/pdf/2604.01987v1
+
+## 94. World Action Verifier: Self-Improving World Models via Forward-Inverse Asymmetry
+
+- **ID:** `2604.01985` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Yuejiang Liu, Fan Feng, Lingjing Kong, et al.
+- **abs:** https://arxiv.org/abs/2604.01985 · **PDF:** https://arxiv.org/pdf/2604.01985v1
+
+## 95. Homogenized Transformers
+
+- **ID:** `2604.01978` · **primary:** `math.PR` · **Submitted:** 2026-04-02
+- **Authors:** Hugo Koubbi, Borjan Geshkovski, Philippe Rigollet
+- **abs:** https://arxiv.org/abs/2604.01978 · **PDF:** https://arxiv.org/pdf/2604.01978v1
+
+## 96. RuleForge: Automated Generation and Validation for Web Vulnerability Detection at Scale
+
+- **ID:** `2604.01977` · **primary:** `cs.CR` · **Submitted:** 2026-04-02
+- **Authors:** Ayush Garg, Sophia Hager, Jacob Montiel, et al.
+- **abs:** https://arxiv.org/abs/2604.01977 · **PDF:** https://arxiv.org/pdf/2604.01977v1
+
+## 97. Ego-Grounding for Personalized Question-Answering in Egocentric Videos
+
+- **ID:** `2604.01966` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Junbin Xiao, Shenglang Zhang, Pengxiang Zhu, et al.
+- **abs:** https://arxiv.org/abs/2604.01966 · **PDF:** https://arxiv.org/pdf/2604.01966v1
+
+## 98. Do We Need Bigger Models for Science? Task-Aware Retrieval with Small Language Models
+
+- **ID:** `2604.01965` · **primary:** `cs.IR` · **Submitted:** 2026-04-02
+- **Authors:** Florian Kelber, Matthias Jobst, Yuni Susanti, et al.
+- **abs:** https://arxiv.org/abs/2604.01965 · **PDF:** https://arxiv.org/pdf/2604.01965v1
+
+## 99. Abnormal Head Movements in Neurological Conditions: A Knowledge-Based Dataset with Application to Cervical Dystonia
+
+- **ID:** `2604.01962` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Saja Al-Dabet, Sherzod Turaev, Nazar Zaki
+- **abs:** https://arxiv.org/abs/2604.01962 · **PDF:** https://arxiv.org/pdf/2604.01962v1
+
+## 100. Generalization Bounds and Statistical Guarantees for Multi-Task and Multiple Operator Learning with MNO Networks
+
+- **ID:** `2604.01961` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Adrien Weihs, Hayden Schaeffer
+- **abs:** https://arxiv.org/abs/2604.01961 · **PDF:** https://arxiv.org/pdf/2604.01961v1
+
+## 101. Diagnosing Translated Benchmarks: An Automated Quality Assurance Study of the EU20 Benchmark Suite
+
+- **ID:** `2604.01957` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Klaudia Thellmann, Bernhard Stadler, Michael Färber
+- **abs:** https://arxiv.org/abs/2604.01957 · **PDF:** https://arxiv.org/pdf/2604.01957v1
+
+## 102. Qiana: A First-Order Formalism to Quantify over Contexts and Formulas with Temporality
+
+- **ID:** `2604.01952` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Simon Coumes, Pierre-Henri Paris, François Schwarzentruber, et al.
+- **abs:** https://arxiv.org/abs/2604.01952 · **PDF:** https://arxiv.org/pdf/2604.01952v1
+
+## 103. Learn by Surprise, Commit by Proof
+
+- **ID:** `2604.01951` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Kang-Sin Choi
+- **abs:** https://arxiv.org/abs/2604.01951 · **PDF:** https://arxiv.org/pdf/2604.01951v1
+
+## 104. annbatch unlocks terabyte-scale training of biological data in anndata
+
+- **ID:** `2604.01949` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Ilan Gold, Felix Fischer, Lucas Arnoldt, et al.
+- **abs:** https://arxiv.org/abs/2604.01949 · **PDF:** https://arxiv.org/pdf/2604.01949v1
+
+## 105. PAC-Bayesian Reward-Certified Outcome Weighted Learning
+
+- **ID:** `2604.01946` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Yuya Ishikawa, Shu Tamano
+- **abs:** https://arxiv.org/abs/2604.01946 · **PDF:** https://arxiv.org/pdf/2604.01946v1
+
+## 106. Physics-Informed Transformer for Multi-Band Channel Frequency Response Reconstruction
+
+- **ID:** `2604.01944` · **primary:** `cs.NI` · **Submitted:** 2026-04-02
+- **Authors:** Anatolij Zubow, Joana Angjo, Sigrid Dimce, et al.
+- **abs:** https://arxiv.org/abs/2604.01944 · **PDF:** https://arxiv.org/pdf/2604.01944v1
+
+## 107. A Novel Theoretical Analysis for Clustering Heteroscedastic Gaussian Data without Knowledge of the Number of Clusters
+
+- **ID:** `2604.01943` · **primary:** `stat.ML` · **Submitted:** 2026-04-02
+- **Authors:** Dominique Pastor, Elsa Dupraz, Ismail Hbilou, et al.
+- **abs:** https://arxiv.org/abs/2604.01943 · **PDF:** https://arxiv.org/pdf/2604.01943v1
+
+## 108. Captioning Daily Activity Images in Early Childhood Education: Benchmark and Algorithm
+
+- **ID:** `2604.01941` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Sixing Li, Zhibin Gu, Ziqi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2604.01941 · **PDF:** https://arxiv.org/pdf/2604.01941v1
+
+## 109. Probabilistic classification from possibilistic data: computing Kullback-Leibler projection with a possibility distribution
+
+- **ID:** `2604.01939` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Ismaïl Baaj, Pierre Marquis
+- **abs:** https://arxiv.org/abs/2604.01939 · **PDF:** https://arxiv.org/pdf/2604.01939v1
+
+## 110. How to measure the optimality of word or gesture order with respect to the principle of swap distance minimization
+
+- **ID:** `2604.01938` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Ramon Ferrer-i-Cancho
+- **abs:** https://arxiv.org/abs/2604.01938 · **PDF:** https://arxiv.org/pdf/2604.01938v1
+
+## 111. Reliable News or Propagandist News? A Neurosymbolic Model Using Genre, Topic, and Persuasion Techniques to Improve Robustness in Classification
+
+- **ID:** `2604.01936` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Géraud Faye, Benjamin Icard, Morgane Casanova, et al.
+- **abs:** https://arxiv.org/abs/2604.01936 · **PDF:** https://arxiv.org/pdf/2604.01936v1
+
+## 112. BraiNCA: brain-inspired neural cellular automata and applications to morphogenesis and motor control
+
+- **ID:** `2604.01932` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Léo Pio-Lopez, Benedikt Hartl, Michael Levin
+- **abs:** https://arxiv.org/abs/2604.01932 · **PDF:** https://arxiv.org/pdf/2604.01932v1
+
+## 113. Quantum-Inspired Geometric Classification with Correlation Group Structures and VQC Decision Modeling
+
+- **ID:** `2604.01930` · **primary:** `quant-ph` · **Submitted:** 2026-04-02
+- **Authors:** Nishikanta Mohanty, Arya Ansuman Priyadarshi, Bikash K. Behera, et al.
+- **abs:** https://arxiv.org/abs/2604.01930 · **PDF:** https://arxiv.org/pdf/2604.01930v1
+
+## 114. Woosh: A Sound Effects Foundation Model
+
+- **ID:** `2604.01929` · **primary:** `cs.SD` · **Submitted:** 2026-04-02
+- **Authors:** Gaëtan Hadjeres, Marc Ferras, Khaled Koutini, et al.
+- **abs:** https://arxiv.org/abs/2604.01929 · **PDF:** https://arxiv.org/pdf/2604.01929v1
+
+## 115. ImplicitBBQ: Benchmarking Implicit Bias in Large Language Models through Characteristic Based Cues
+
+- **ID:** `2604.01925` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Bhaskara Hanuma Vedula, Darshan Anghan, Ishita Goyal, et al.
+- **abs:** https://arxiv.org/abs/2604.01925 · **PDF:** https://arxiv.org/pdf/2604.01925v1
+
+## 116. Is Clinical Text Enough? A Multimodal Study on Mortality Prediction in Heart Failure Patients
+
+- **ID:** `2604.01924` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Oumaima El Khettari, Virgile Barthet, Guillaume Hocquet, et al.
+- **abs:** https://arxiv.org/abs/2604.01924 · **PDF:** https://arxiv.org/pdf/2604.01924v1
+
+## 117. Learning Spatial Structure from Pre-Beamforming Per-Antenna Range-Doppler Radar Data via Visibility-Aware Cross-Modal Supervision
+
+- **ID:** `2604.01921` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** George Sebastian, Philipp Berthold, Bianca Forkel, et al.
+- **abs:** https://arxiv.org/abs/2604.01921 · **PDF:** https://arxiv.org/pdf/2604.01921v1
+
+## 118. SURE: Synergistic Uncertainty-aware Reasoning for Multimodal Emotion Recognition in Conversations
+
+- **ID:** `2604.01916` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Yiqiang Cai, Chengyan Wu, Bolei Ma, et al.
+- **abs:** https://arxiv.org/abs/2604.01916 · **PDF:** https://arxiv.org/pdf/2604.01916v1
+
+## 119. The Rank and Gradient Lost in Non-stationarity: Sample Weight Decay for Mitigating Plasticity Loss in Reinforcement Learning
+
+- **ID:** `2604.01913` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Zihao Wu, Hongyao Tang, Yi Ma, et al.
+- **abs:** https://arxiv.org/abs/2604.01913 · **PDF:** https://arxiv.org/pdf/2604.01913v1
+
+## 120. Lifting Unlabeled Internet-level Data for 3D Scene Understanding
+
+- **ID:** `2604.01907` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Yixin Chen, Yaowei Zhang, Huangyue Yu, et al.
+- **abs:** https://arxiv.org/abs/2604.01907 · **PDF:** https://arxiv.org/pdf/2604.01907v1
+
+## 121. Combating Data Laundering in LLM Training
+
+- **ID:** `2604.01904` · **primary:** `cs.CR` · **Submitted:** 2026-04-02
+- **Authors:** Muxing Li, Zesheng Ye, Sharon Li, et al.
+- **abs:** https://arxiv.org/abs/2604.01904 · **PDF:** https://arxiv.org/pdf/2604.01904v1
+
+## 122. Enhancing the Reliability of Medical AI through Expert-guided Uncertainty Modeling
+
+- **ID:** `2604.01898` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Aleksei Khalin, Ekaterina Zaychenkova, Aleksandr Yugay, et al.
+- **abs:** https://arxiv.org/abs/2604.01898 · **PDF:** https://arxiv.org/pdf/2604.01898v1
+
+## 123. Bayesian Elicitation with LLMs: Model Size Helps, Extra "Reasoning" Doesn't Always
+
+- **ID:** `2604.01896` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Luka Hobor, Mario Brcic, Mihael Kovac, et al.
+- **abs:** https://arxiv.org/abs/2604.01896 · **PDF:** https://arxiv.org/pdf/2604.01896v1
+
+## 124. LI-DSN: A Layer-wise Interactive Dual-Stream Network for EEG Decoding
+
+- **ID:** `2604.01889` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Chenghao Yue, Zhiyuan Ma, Zhongye Xia, et al.
+- **abs:** https://arxiv.org/abs/2604.01889 · **PDF:** https://arxiv.org/pdf/2604.01889v1
+
+## 125. When does learning pay off? A study on DRL-based dynamic algorithm configuration for carbon-aware scheduling
+
+- **ID:** `2604.01886` · **primary:** `math.OC` · **Submitted:** 2026-04-02
+- **Authors:** Andrea Mencaroni, Robbert Reijnen, Yingqian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2604.01886 · **PDF:** https://arxiv.org/pdf/2604.01886v1
+
+## 126. HieraVid: Hierarchical Token Pruning for Fast Video Large Language Models
+
+- **ID:** `2604.01881` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Yansong Guo, Chaoyang Zhu, Jiayi Ji, et al.
+- **abs:** https://arxiv.org/abs/2604.01881 · **PDF:** https://arxiv.org/pdf/2604.01881v1
+
+## 127. DDCL-INCRT: A Self-Organising Transformer with Hierarchical Prototype Structure (Theoretical Foundations)
+
+- **ID:** `2604.01880` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Giansalvo Cirrincione
+- **abs:** https://arxiv.org/abs/2604.01880 · **PDF:** https://arxiv.org/pdf/2604.01880v1
+
+## 128. Robust Graph Representation Learning via Adaptive Spectral Contrast
+
+- **ID:** `2604.01878` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Zhuolong Li, Boxue Yang, Haopeng Chen
+- **abs:** https://arxiv.org/abs/2604.01878 · **PDF:** https://arxiv.org/pdf/2604.01878v1
+
+## 129. Towards Intrinsically Calibrated Uncertainty Quantification in Industrial Data-Driven Models via Diffusion Sampler
+
+- **ID:** `2604.01870` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Yiran Ma, Jerome Le Ny, Zhichao Chen, et al.
+- **abs:** https://arxiv.org/abs/2604.01870 · **PDF:** https://arxiv.org/pdf/2604.01870v1
+
+## 130. Efficient Constraint Generation for Stochastic Shortest Path Problems
+
+- **ID:** `2604.01855` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Johannes Schmalz, Felipe Trevizan
+- **abs:** https://arxiv.org/abs/2604.01855 · **PDF:** https://arxiv.org/pdf/2604.01855v1
+
+## 131. Beyond Detection: Ethical Foundations for Automated Dyslexic Error Attribution
+
+- **ID:** `2604.01853` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Samuel Rose, Debarati Chakraborty
+- **abs:** https://arxiv.org/abs/2604.01853 · **PDF:** https://arxiv.org/pdf/2604.01853v1
+
+## 132. From Guessing to Placeholding: A Cost-Theoretic Framework for Uncertainty-Aware Code Completion
+
+- **ID:** `2604.01849` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Liang Zhu, Haolin Chen, Lidong Zhao, et al.
+- **abs:** https://arxiv.org/abs/2604.01849 · **PDF:** https://arxiv.org/pdf/2604.01849v1
+
+## 133. CANDI: Curated Test-Time Adaptation for Multivariate Time-Series Anomaly Detection Under Distribution Shift
+
+- **ID:** `2604.01845` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** HyunGi Kim, Jisoo Mok, Hyungyu Lee, et al.
+- **abs:** https://arxiv.org/abs/2604.01845 · **PDF:** https://arxiv.org/pdf/2604.01845v1
+
+## 134. Investigating Permutation-Invariant Discrete Representation Learning for Spatially Aligned Images
+
+- **ID:** `2604.01843` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Jamie S. J. Stirling, Noura Al-Moubayed, Hubert P. H. Shum
+- **abs:** https://arxiv.org/abs/2604.01843 · **PDF:** https://arxiv.org/pdf/2604.01843v1
+
+## 135. Retrieval-aligned Tabular Foundation Models Enable Robust Clinical Risk Prediction in Electronic Health Records Under Real-world Constraints
+
+- **ID:** `2604.01841` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Minh-Khoi Pham, Thang-Long Nguyen Ho, Thao Thi Phuong Dao, et al.
+- **abs:** https://arxiv.org/abs/2604.01841 · **PDF:** https://arxiv.org/pdf/2604.01841v1
+
+## 136. Not All Tokens See Equally: Perception-Grounded Policy Optimization for Large Vision-Language Models
+
+- **ID:** `2604.01840` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Zekai Ye, Qiming Li, Xiaocheng Feng, et al.
+- **abs:** https://arxiv.org/abs/2604.01840 · **PDF:** https://arxiv.org/pdf/2604.01840v1
+
+## 137. PLOT: Enhancing Preference Learning via Optimal Transport
+
+- **ID:** `2604.01837` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Liang Zhu, Yuelin Bai, Xiankun Ren, et al.
+- **abs:** https://arxiv.org/abs/2604.01837 · **PDF:** https://arxiv.org/pdf/2604.01837v1
+
+## 138. Language-Pretraining-Induced Bias: A Strong Foundation for General Vision Tasks
+
+- **ID:** `2604.01833` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Yaxin Luo, Zhiqiang Shen
+- **abs:** https://arxiv.org/abs/2604.01833 · **PDF:** https://arxiv.org/pdf/2604.01833v1
+
+## 139. Physics Informed Reinforcement Learning with Gibbs Priors for Topology Control in Power Grids
+
+- **ID:** `2604.01830` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Pantelis Dogoulis, Maxime Cordy
+- **abs:** https://arxiv.org/abs/2604.01830 · **PDF:** https://arxiv.org/pdf/2604.01830v1
+
+## 140. Neural Network-Assisted Model Predictive Control for Implicit Balancing
+
+- **ID:** `2604.01805` · **primary:** `eess.SY` · **Submitted:** 2026-04-02
+- **Authors:** Seyed Soroush Karimi Madahi, Kenneth Bruninx, Bert Claessens, et al.
+- **abs:** https://arxiv.org/abs/2604.01805 · **PDF:** https://arxiv.org/pdf/2604.01805v1
+
+## 141. Graph Neural Operator Towards Edge Deployability and Portability for Sparse-to-Dense, Real-Time Virtual Sensing on Irregular Grids
+
+- **ID:** `2604.01802` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** William Howes, Jason Yoo, Kazuma Kobayashi, et al.
+- **abs:** https://arxiv.org/abs/2604.01802 · **PDF:** https://arxiv.org/pdf/2604.01802v1
+
+## 142. A deep learning pipeline for PAM50 subtype classification using histopathology images and multi-objective patch selection
+
+- **ID:** `2604.01798` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Arezoo Borji, Gernot Kronreif, Bernhard Angermayr, et al.
+- **abs:** https://arxiv.org/abs/2604.01798 · **PDF:** https://arxiv.org/pdf/2604.01798v1
+
+## 143. Learning in Prophet Inequalities with Noisy Observations
+
+- **ID:** `2604.01789` · **primary:** `stat.ML` · **Submitted:** 2026-04-02
+- **Authors:** Jung-hun Kim, Vianney Perchet
+- **abs:** https://arxiv.org/abs/2604.01789 · **PDF:** https://arxiv.org/pdf/2604.01789v1
+
+## 144. DEFT: Distribution-guided Efficient Fine-Tuning for Human Alignment
+
+- **ID:** `2604.01787` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Liang Zhu, Feiteng Fang, Yuelin Bai, et al.
+- **abs:** https://arxiv.org/abs/2604.01787 · **PDF:** https://arxiv.org/pdf/2604.01787v1
+
+## 145. Taming CATS: Controllable Automatic Text Simplification through Instruction Fine-Tuning with Control Tokens
+
+- **ID:** `2604.01779` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Hanna Hubarava, Yingqiang Gao
+- **abs:** https://arxiv.org/abs/2604.01779 · **PDF:** https://arxiv.org/pdf/2604.01779v1
+
+## 146. Bridging Deep Learning and Integer Linear Programming: A Predictive-to-Prescriptive Framework for Supply Chain Analytics
+
+- **ID:** `2604.01775` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Khai Banh Nghiep, Duc Nguyen Minh, Lan Hoang Thi
+- **abs:** https://arxiv.org/abs/2604.01775 · **PDF:** https://arxiv.org/pdf/2604.01775v1
+
+## 147. Domain-constrained knowledge representation: A modal framework
+
+- **ID:** `2604.01770` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Chao Li, Yuru Wang, Chunyi Zhao
+- **abs:** https://arxiv.org/abs/2604.01770 · **PDF:** https://arxiv.org/pdf/2604.01770v1
+
+## 148. Dual-Attention Based 3D Channel Estimation
+
+- **ID:** `2604.01769` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Xiangzhao Qin, Sha Hu
+- **abs:** https://arxiv.org/abs/2604.01769 · **PDF:** https://arxiv.org/pdf/2604.01769v1
+
+## 149. FSKD: Monocular Forest Structure Inference via LiDAR-to-RGBI Knowledge Distillation
+
+- **ID:** `2604.01766` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Taimur Khan, Hannes Feilhauer, Muhammad Jazib Zafar
+- **abs:** https://arxiv.org/abs/2604.01766 · **PDF:** https://arxiv.org/pdf/2604.01766v1
+
+## 150. DriveDreamer-Policy: A Geometry-Grounded World-Action Model for Unified Generation and Planning
+
+- **ID:** `2604.01765` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Yang Zhou, Xiaofeng Wang, Hao Shao, et al.
+- **abs:** https://arxiv.org/abs/2604.01765 · **PDF:** https://arxiv.org/pdf/2604.01765v1
+
+## 151. FourierMoE: Fourier Mixture-of-Experts Adaptation of Large Language Models
+
+- **ID:** `2604.01762` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Juyong Jiang, Fan Wang, Hong Qi, et al.
+- **abs:** https://arxiv.org/abs/2604.01762 · **PDF:** https://arxiv.org/pdf/2604.01762v1
+
+## 152. LiveMathematicianBench: A Live Benchmark for Mathematician-Level Reasoning with Proof Sketches
+
+- **ID:** `2604.01754` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Linyang He, Qiyao Yu, Hanze Dong, et al.
+- **abs:** https://arxiv.org/abs/2604.01754 · **PDF:** https://arxiv.org/pdf/2604.01754v1
+
+## 153. Detecting Toxic Language: Ontology and BERT-based Approaches for Bulgarian Text
+
+- **ID:** `2604.01745` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Melania Berbatova, Tsvetoslav Vasev
+- **abs:** https://arxiv.org/abs/2604.01745 · **PDF:** https://arxiv.org/pdf/2604.01745v1
+
+## 154. DDCL: Deep Dual Competitive Learning: A Differentiable End-to-End Framework for Unsupervised Prototype-Based Representation Learning
+
+- **ID:** `2604.01740` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Giansalvo Cirrincione
+- **abs:** https://arxiv.org/abs/2604.01740 · **PDF:** https://arxiv.org/pdf/2604.01740v1
+
+## 155. AeroTherm-GPT: A Verification-Centered LLM Framework for Thermal Protection System Engineering Workflows
+
+- **ID:** `2604.01738` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Chuhan Qiao, Jinglai Zheng, Jie Huang, et al.
+- **abs:** https://arxiv.org/abs/2604.01738 · **PDF:** https://arxiv.org/pdf/2604.01738v1
+
+## 156. From BM25 to Corrective RAG: Benchmarking Retrieval Strategies for Text-and-Table Documents
+
+- **ID:** `2604.01733` · **primary:** `cs.IR` · **Submitted:** 2026-04-02
+- **Authors:** Meftun Akarsu, Recep Kaan Karaman, Christopher Mierbach
+- **abs:** https://arxiv.org/abs/2604.01733 · **PDF:** https://arxiv.org/pdf/2604.01733v1
+
+## 157. Solving the Two-dimensional single stock size Cuting Stock Problem with SAT and MaxSAT
+
+- **ID:** `2604.01732` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Tuyen Van Kieu, Chi Linh Hoang, Khanh Van To
+- **abs:** https://arxiv.org/abs/2604.01732 · **PDF:** https://arxiv.org/pdf/2604.01732v1
+
+## 158. Koopman-Based Nonlinear Identification and Adaptive Control of a Turbofan Engine
+
+- **ID:** `2604.01730` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** David Grasev
+- **abs:** https://arxiv.org/abs/2604.01730 · **PDF:** https://arxiv.org/pdf/2604.01730v1
+
+## 159. The AnIML Ontology: Enabling Semantic Interoperability for Large-Scale Experimental Data in Interconnected Scientific Labs
+
+- **ID:** `2604.01728` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Wilf Morlidge, Elliott Watkiss-Leek, George Hannah, et al.
+- **abs:** https://arxiv.org/abs/2604.01728 · **PDF:** https://arxiv.org/pdf/2604.01728v1
+
+## 160. MATA-Former & SIICU: Semantic Aware Temporal Alignment for High-Fidelity ICU Risk Prediction
+
+- **ID:** `2604.01727` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Zhichong Zheng, Xiaohang Nie, Xueqi Wang, et al.
+- **abs:** https://arxiv.org/abs/2604.01727 · **PDF:** https://arxiv.org/pdf/2604.01727v1
+
+## 161. LiteInception: A Lightweight and Interpretable Deep Learning Framework for General Aviation Fault Diagnosis
+
+- **ID:** `2604.01725` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Zhihuan Wei, Xinhang Chen, Danyang Han, et al.
+- **abs:** https://arxiv.org/abs/2604.01725 · **PDF:** https://arxiv.org/pdf/2604.01725v1
+
+## 162. Causal Scene Narration with Runtime Safety Supervision for Vision-Language-Action Driving
+
+- **ID:** `2604.01723` · **primary:** `cs.RO` · **Submitted:** 2026-04-02
+- **Authors:** Yun Li, Yidu Zhang, Simon Thompson, et al.
+- **abs:** https://arxiv.org/abs/2604.01723 · **PDF:** https://arxiv.org/pdf/2604.01723v1
+
+## 163. Transformer self-attention encoder-decoder with multimodal deep learning for response time series forecasting and digital twin support in wind structural health monitoring
+
+- **ID:** `2604.01712` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Feiyu Zhou, Marios Impraimakis
+- **abs:** https://arxiv.org/abs/2604.01712 · **PDF:** https://arxiv.org/pdf/2604.01712v1
+
+## 164. Human-Guided Reasoning with Large Language Models for Vietnamese Speech Emotion Recognition
+
+- **ID:** `2604.01711` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Truc Nguyen, Then Tran, Binh Truong, et al.
+- **abs:** https://arxiv.org/abs/2604.01711 · **PDF:** https://arxiv.org/pdf/2604.01711v1
+
+## 165. OpenGo: An OpenClaw-Based Robotic Dog with Real-Time Skill Switching
+
+- **ID:** `2604.01708` · **primary:** `cs.RO` · **Submitted:** 2026-04-02
+- **Authors:** Hanbing Li, Xuewei Cao, Zhiwen Zeng, et al.
+- **abs:** https://arxiv.org/abs/2604.01708 · **PDF:** https://arxiv.org/pdf/2604.01708v1
+
+## 166. Memory in the LLM Era: Modular Architectures and Strategies in a Unified Framework
+
+- **ID:** `2604.01707` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Yanchen Wu, Tenghui Lin, Yingli Zhou, et al.
+- **abs:** https://arxiv.org/abs/2604.01707 · **PDF:** https://arxiv.org/pdf/2604.01707v1
+
+## 167. Development and multi-center evaluation of domain-adapted speech recognition for human-AI teaming in real-world gastrointestinal endoscopy
+
+- **ID:** `2604.01705` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Ruijie Yang, Yan Zhu, Peiyao Fu, et al.
+- **abs:** https://arxiv.org/abs/2604.01705 · **PDF:** https://arxiv.org/pdf/2604.01705v1
+
+## 168. On the Role of Reasoning Patterns in the Generalization Discrepancy of Long Chain-of-Thought Supervised Fine-Tuning
+
+- **ID:** `2604.01702` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Zhaoyi Li, Xiangyu Xi, Zhengyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2604.01702 · **PDF:** https://arxiv.org/pdf/2604.01702v1
+
+## 169. MiCA Learns More Knowledge Than LoRA and Full Fine-Tuning
+
+- **ID:** `2604.01694` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Sten Rüdiger, Sebastian Raschka
+- **abs:** https://arxiv.org/abs/2604.01694 · **PDF:** https://arxiv.org/pdf/2604.01694v1
+
+## 170. Scale over Preference: The Impact of AI-Generated Content on Online Content Ecology
+
+- **ID:** `2604.01690` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Tianhao Shi, Yang Zhang, Xiaoyan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2604.01690 · **PDF:** https://arxiv.org/pdf/2604.01690v1
+
+## 171. EvoSkills: Self-Evolving Agent Skills via Co-Evolutionary Verification
+
+- **ID:** `2604.01687` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Hanrong Zhang, Shicheng Fan, Henry Peng Zou, et al.
+- **abs:** https://arxiv.org/abs/2604.01687 · **PDF:** https://arxiv.org/pdf/2604.01687v1
+
+## 172. Coupled Query-Key Dynamics for Attention
+
+- **ID:** `2604.01683` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Barak Gahtan, Alex M. Bronstein
+- **abs:** https://arxiv.org/abs/2604.01683 · **PDF:** https://arxiv.org/pdf/2604.01683v1
+
+## 173. PRISM: Probability Reallocation with In-Span Masking for Knowledge-Sensitive Alignment
+
+- **ID:** `2604.01682` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Chenning Xu, Mao Zheng, Mingyang Song
+- **abs:** https://arxiv.org/abs/2604.01682 · **PDF:** https://arxiv.org/pdf/2604.01682v1
+
+## 174. Bridging Large-Model Reasoning and Real-Time Control via Agentic Fast-Slow Planning
+
+- **ID:** `2604.01681` · **primary:** `cs.RO` · **Submitted:** 2026-04-02
+- **Authors:** Jiayi Chen, Shuai Wang, Guangxu Zhu, et al.
+- **abs:** https://arxiv.org/abs/2604.01681 · **PDF:** https://arxiv.org/pdf/2604.01681v1
+
+## 175. GPA: Learning GUI Process Automation from Demonstrations
+
+- **ID:** `2604.01676` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Zirui Zhao, Jun Hao Liew, Yan Yang, et al.
+- **abs:** https://arxiv.org/abs/2604.01676 · **PDF:** https://arxiv.org/pdf/2604.01676v1
+
+## 176. Can Heterogeneous Language Models Be Fused?
+
+- **ID:** `2604.01674` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Shilian Chen, Jie Zhou, Qin Chen, et al.
+- **abs:** https://arxiv.org/abs/2604.01674 · **PDF:** https://arxiv.org/pdf/2604.01674v1
+
+## 177. PRCCF: A Persona-guided Retrieval and Causal-aware Cognitive Filtering Framework for Emotional Support Conversation
+
+- **ID:** `2604.01671` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Yanxin Luo, Xiaoyu Zhang, Jing Li, et al.
+- **abs:** https://arxiv.org/abs/2604.01671 · **PDF:** https://arxiv.org/pdf/2604.01671v1
+
+## 178. Hierarchical Memory Orchestration for Personalized Persistent Agents
+
+- **ID:** `2604.01670` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Junming Liu, Yifei Sun, Weihua Cheng, et al.
+- **abs:** https://arxiv.org/abs/2604.01670 · **PDF:** https://arxiv.org/pdf/2604.01670v1
+
+## 179. Robust Embodied Perception in Dynamic Environments via Disentangled Weight Fusion
+
+- **ID:** `2604.01669` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Juncen Guo, Xiaoguang Zhu, Jingyi Wu, et al.
+- **abs:** https://arxiv.org/abs/2604.01669 · **PDF:** https://arxiv.org/pdf/2604.01669v1
+
+## 180. M3D-BFS: a Multi-stage Dynamic Fusion Strategy for Sample-Adaptive Multi-Modal Brain Network Analysis
+
+- **ID:** `2604.01667` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Rui Dong, Xiaotong Zhang, Jiaxing Li, et al.
+- **abs:** https://arxiv.org/abs/2604.01667 · **PDF:** https://arxiv.org/pdf/2604.01667v1
+
+## 181. ContextBudget: Budget-Aware Context Management for Long-Horizon Search Agents
+
+- **ID:** `2604.01664` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Yong Wu, YanZhao Zheng, TianZe Xu, et al.
+- **abs:** https://arxiv.org/abs/2604.01664 · **PDF:** https://arxiv.org/pdf/2604.01664v1
+
+## 182. Ontology-Aware Design Patterns for Clinical AI Systems: Translating Reification Theory into Software Architecture
+
+- **ID:** `2604.01661` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Florian Odi Stummer
+- **abs:** https://arxiv.org/abs/2604.01661 · **PDF:** https://arxiv.org/pdf/2604.01661v1
+
+## 183. CORAL: Towards Autonomous Multi-Agent Evolution for Open-Ended Discovery
+
+- **ID:** `2604.01658` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Ao Qu, Han Zheng, Zijian Zhou, et al.
+- **abs:** https://arxiv.org/abs/2604.01658 · **PDF:** https://arxiv.org/pdf/2604.01658v1
+
+## 184. What Do Claim Verification Datasets Actually Test? A Reasoning Trace Analysis
+
+- **ID:** `2604.01657` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Delip Rao, Chris Callison-Burch
+- **abs:** https://arxiv.org/abs/2604.01657 · **PDF:** https://arxiv.org/pdf/2604.01657v1
+
+## 185. Moiré Video Authentication: A Physical Signature Against AI Video Generation
+
+- **ID:** `2604.01654` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Yuan Qing, Kunyu Zheng, Lingxiao Li, et al.
+- **abs:** https://arxiv.org/abs/2604.01654 · **PDF:** https://arxiv.org/pdf/2604.01654v1
+
+## 186. Cognitive Energy Modeling for Neuroadaptive Human-Machine Systems using EEG and WGAN-GP
+
+- **ID:** `2604.01653` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Sriram Sattiraju, Vaibhav Gollapalli, Aryan Shah, et al.
+- **abs:** https://arxiv.org/abs/2604.01653 · **PDF:** https://arxiv.org/pdf/2604.01653v1
+
+## 187. ThinknCheck: Grounded Claim Verification with Compact, Reasoning-Driven, and Interpretable Models
+
+- **ID:** `2604.01652` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Delip Rao, Feijiang Han, Chris Callison-Burch
+- **abs:** https://arxiv.org/abs/2604.01652 · **PDF:** https://arxiv.org/pdf/2604.01652v1
+
+## 188. Label Shift Estimation With Incremental Prior Update
+
+- **ID:** `2604.01651` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Yunrui Zhang, Gustavo Batista, Salil S. Kanhere
+- **abs:** https://arxiv.org/abs/2604.01651 · **PDF:** https://arxiv.org/pdf/2604.01651v1
+
+## 189. AromaGen: Interactive Generation of Rich Olfactory Experiences with Multimodal Language Models
+
+- **ID:** `2604.01650` · **primary:** `cs.HC` · **Submitted:** 2026-04-02
+- **Authors:** Yunge Wen, Awu Chen, Jianing Yu, et al.
+- **abs:** https://arxiv.org/abs/2604.01650 · **PDF:** https://arxiv.org/pdf/2604.01650v1
+
+## 190. Exploring Robust Multi-Agent Workflows for Environmental Data Management
+
+- **ID:** `2604.01647` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Boyuan Guan, Jason Liu, Yanzhao Wu, et al.
+- **abs:** https://arxiv.org/abs/2604.01647 · **PDF:** https://arxiv.org/pdf/2604.01647v1
+
+## 191. Fragile Reasoning: A Mechanistic Analysis of LLM Sensitivity to Meaning-Preserving Perturbations
+
+- **ID:** `2604.01639` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Shou-Tzu Han, Rodrigue Rizk, KC Santosh
+- **abs:** https://arxiv.org/abs/2604.01639 · **PDF:** https://arxiv.org/pdf/2604.01639v1
+
+## 192. Seclens: Role-specific Evaluation of LLM's for security vulnerablity detection
+
+- **ID:** `2604.01637` · **primary:** `cs.CR` · **Submitted:** 2026-04-02
+- **Authors:** Subho Halder, Siddharth Saxena, Kashinath Kadaba Shrish, et al.
+- **abs:** https://arxiv.org/abs/2604.01637 · **PDF:** https://arxiv.org/pdf/2604.01637v1
+
+## 193. CRIT: Graph-Based Automatic Data Synthesis to Enhance Cross-Modal Multi-Hop Reasoning
+
+- **ID:** `2604.01634` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Junyoung Sung, Seungwoo Lyu, Minjun Kim, et al.
+- **abs:** https://arxiv.org/abs/2604.01634 · **PDF:** https://arxiv.org/pdf/2604.01634v1
+
+## 194. Grounding AI-in-Education Development in Teachers' Voices: Findings from a National Survey in Indonesia
+
+- **ID:** `2604.01630` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Nurul Aisyah, Muhammad Dehan Al Kautsar, Arif Hidayat, et al.
+- **abs:** https://arxiv.org/abs/2604.01630 · **PDF:** https://arxiv.org/pdf/2604.01630v1
+
+## 195. RefinementEngine: Automating Intent-to-Device Filtering Policy Deployment under Network Constraints
+
+- **ID:** `2604.01627` · **primary:** `cs.CR` · **Submitted:** 2026-04-02
+- **Authors:** Davide Colaiacomo, Chiara Bonfanti, Cataldo Basile
+- **abs:** https://arxiv.org/abs/2604.01627 · **PDF:** https://arxiv.org/pdf/2604.01627v1
+
+## 196. OSCAR: Orchestrated Self-verification and Cross-path Refinement
+
+- **ID:** `2604.01624` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Yash Shah, Abhijit Chakraborty, Naresh Kumar Devulapally, et al.
+- **abs:** https://arxiv.org/abs/2604.01624 · **PDF:** https://arxiv.org/pdf/2604.01624v1
+
+## 197. Expert-Choice Routing Enables Adaptive Computation in Diffusion Language Models
+
+- **ID:** `2604.01622` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Shuibai Zhang, Caspian Zhuang, Chihan Cui, et al.
+- **abs:** https://arxiv.org/abs/2604.01622 · **PDF:** https://arxiv.org/pdf/2604.01622v1
+
+## 198. DWDP: Distributed Weight Data Parallelism for High-Performance LLM Inference on NVL72
+
+- **ID:** `2604.01621` · **primary:** `cs.DC` · **Submitted:** 2026-04-02
+- **Authors:** Wanqian Li, Jintao Peng, Zongfei Jing, et al.
+- **abs:** https://arxiv.org/abs/2604.01621 · **PDF:** https://arxiv.org/pdf/2604.01621v1
+
+## 199. Automatic Image-Level Morphological Trait Annotation for Organismal Images
+
+- **ID:** `2604.01619` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Vardaan Pahuja, Samuel Stevens, Alyson East, et al.
+- **abs:** https://arxiv.org/abs/2604.01619 · **PDF:** https://arxiv.org/pdf/2604.01619v1
+
+## 200. Tex3D: Objects as Attack Surfaces via Adversarial 3D Textures for Vision-Language-Action Models
+
+- **ID:** `2604.01618` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Jiawei Chen, Simin Huang, Jiawei Du, et al.
+- **abs:** https://arxiv.org/abs/2604.01618 · **PDF:** https://arxiv.org/pdf/2604.01618v1
+
+## 201. Analysis of LLM Performance on AWS Bedrock: Receipt-item Categorisation Case Study
+
+- **ID:** `2604.01615` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Gabby Sanchez, Sneha Oommen, Cassandra T. Britto, et al.
+- **abs:** https://arxiv.org/abs/2604.01615 · **PDF:** https://arxiv.org/pdf/2604.01615v1
+
+## 202. Pseudo-Quantized Actor-Critic Algorithm for Robustness to Noisy Temporal Difference Error
+
+- **ID:** `2604.01613` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Taisuke Kobayashi
+- **abs:** https://arxiv.org/abs/2604.01613 · **PDF:** https://arxiv.org/pdf/2604.01613v1
+
+## 203. NEMESIS: Noise-suppressed Efficient MAE with Enhanced Superpatch Integration Strategy
+
+- **ID:** `2604.01612` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Kyeonghun Kim, Hyeonseok Jung, Youngung Han, et al.
+- **abs:** https://arxiv.org/abs/2604.01612 · **PDF:** https://arxiv.org/pdf/2604.01612v1
+
+## 204. GraphWalk: Enabling Reasoning in Large Language Models through Tool-Based Graph Navigation
+
+- **ID:** `2604.01610` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Taraneh Ghandi, Hamidreza Mahyar, Shachar Klaiman
+- **abs:** https://arxiv.org/abs/2604.01610 · **PDF:** https://arxiv.org/pdf/2604.01610v1
+
+## 205. Swift-SVD: Theoretical Optimality Meets Practical Efficiency in Low-Rank LLM Compression
+
+- **ID:** `2604.01609` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Ruoling Qi, Yirui Liu, Xuaner Wu, et al.
+- **abs:** https://arxiv.org/abs/2604.01609 · **PDF:** https://arxiv.org/pdf/2604.01609v1
+
+## 206. From Multi-Agent to Single-Agent: When Is Skill Distillation Beneficial?
+
+- **ID:** `2604.01608` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Binyan Xu, Dong Fang, Haitao Li, et al.
+- **abs:** https://arxiv.org/abs/2604.01608 · **PDF:** https://arxiv.org/pdf/2604.01608v1
+
+## 207. ModTrans: Translating Real-world Models for Distributed Training Simulator
+
+- **ID:** `2604.01607` · **primary:** `cs.DC` · **Submitted:** 2026-04-02
+- **Authors:** Yi Lyu
+- **abs:** https://arxiv.org/abs/2604.01607 · **PDF:** https://arxiv.org/pdf/2604.01607v1
+
+## 208. Random Coordinate Descent on the Wasserstein Space of Probability Measures
+
+- **ID:** `2604.01606` · **primary:** `stat.ML` · **Submitted:** 2026-04-02
+- **Authors:** Yewei Xu, Qin Li
+- **abs:** https://arxiv.org/abs/2604.01606 · **PDF:** https://arxiv.org/pdf/2604.01606v1
+
+## 209. CRaFT: Circuit-Guided Refusal Feature Selection via Cross-Layer Transcoders
+
+- **ID:** `2604.01604` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Su-Hyeon Kim, Hyundong Jin, Yejin Lee, et al.
+- **abs:** https://arxiv.org/abs/2604.01604 · **PDF:** https://arxiv.org/pdf/2604.01604v1
+
+## 210. Training In-Context and In-Weights Mixtures Via Contrastive Context Sampling
+
+- **ID:** `2604.01601` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Deeptanshu Malu, Deevyanshu Malu, Aditya Nemiwal, et al.
+- **abs:** https://arxiv.org/abs/2604.01601 · **PDF:** https://arxiv.org/pdf/2604.01601v1
+
+## 211. MM-ReCoder: Advancing Chart-to-Code Generation with Reinforcement Learning and Self-Correction
+
+- **ID:** `2604.01600` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Zitian Tang, Xu Zhang, Jianbo Yuan, et al.
+- **abs:** https://arxiv.org/abs/2604.01600 · **PDF:** https://arxiv.org/pdf/2604.01600v1
+
+## 212. ByteRover: Agent-Native Memory Through LLM-Curated Hierarchical Context
+
+- **ID:** `2604.01599` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Andy Nguyen, Danh Doan, Hoang Pham, et al.
+- **abs:** https://arxiv.org/abs/2604.01599 · **PDF:** https://arxiv.org/pdf/2604.01599v1
+
+## 213. Learning from the Right Rollouts: Data Attribution for PPO-based LLM Post-Training
+
+- **ID:** `2604.01597` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Dong Shu, Denghui Zhang, Jessica Hullman
+- **abs:** https://arxiv.org/abs/2604.01597 · **PDF:** https://arxiv.org/pdf/2604.01597v1
+
+## 214. Optimizing EEG Graph Structure for Seizure Detection: An Information Bottleneck and Self-Supervised Learning Approach
+
+- **ID:** `2604.01595` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Lincan Li, Rikuto Kotoge, Xihao Piao, et al.
+- **abs:** https://arxiv.org/abs/2604.01595 · **PDF:** https://arxiv.org/pdf/2604.01595v1
+
+## 215. Do Large Language Models Mentalize When They Teach?
+
+- **ID:** `2604.01594` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Sevan K. Harootonian, Mark K. Ho, Thomas L. Griffiths, et al.
+- **abs:** https://arxiv.org/abs/2604.01594 · **PDF:** https://arxiv.org/pdf/2604.01594v1
+
+## 216. ThinkTwice: Jointly Optimizing Large Language Models for Reasoning and Self-Refinement
+
+- **ID:** `2604.01591` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Difan Jiao, Qianfeng Wen, Blair Yang, et al.
+- **abs:** https://arxiv.org/abs/2604.01591 · **PDF:** https://arxiv.org/pdf/2604.01591v1
+
+## 217. NED-Tree: Bridging the Semantic Gap with Nonlinear Element Decomposition Tree for LLM Nonlinear Optimization Modeling
+
+- **ID:** `2604.01588` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Zhijing Hu, Yufan Deng, Haoyang Liu, et al.
+- **abs:** https://arxiv.org/abs/2604.01588 · **PDF:** https://arxiv.org/pdf/2604.01588v1
+
+## 218. Variational LSTM with Augmented Inputs: Nonlinear Response History Metamodeling with Aleatoric and Epistemic Uncertainty
+
+- **ID:** `2604.01587` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Manisha Sapkota, Min Li, Bowei Li
+- **abs:** https://arxiv.org/abs/2604.01587 · **PDF:** https://arxiv.org/pdf/2604.01587v1
+
+## 219. SHOE: Semantic HOI Open-Vocabulary Evaluation Metric
+
+- **ID:** `2604.01586` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Maja Noack, Qinqian Lei, Taipeng Tian, et al.
+- **abs:** https://arxiv.org/abs/2604.01586 · **PDF:** https://arxiv.org/pdf/2604.01586v1
+
+## 220. Harmonized Tabular-Image Fusion via Gradient-Aligned Alternating Learning
+
+- **ID:** `2604.01579` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Longfei Huang, Yang Yang
+- **abs:** https://arxiv.org/abs/2604.01579 · **PDF:** https://arxiv.org/pdf/2604.01579v1
+
+## 221. Thinking While Listening: Fast-Slow Recurrence for Long-Horizon Sequential Modeling
+
+- **ID:** `2604.01577` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Shota Takashiro, Masanori Koyama, Takeru Miyato, et al.
+- **abs:** https://arxiv.org/abs/2604.01577 · **PDF:** https://arxiv.org/pdf/2604.01577v1
+
+## 222. Care-Conditioned Neuromodulation for Autonomy-Preserving Supportive Dialogue Agents
+
+- **ID:** `2604.01576` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Shalima Binta Manir, Tim Oates
+- **abs:** https://arxiv.org/abs/2604.01576 · **PDF:** https://arxiv.org/pdf/2604.01576v1
+
+## 223. Does Your Optimizer Care How You Normalize? Normalization-Optimizer Coupling in LLM Training
+
+- **ID:** `2604.01563` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Abdelrahman Abouzeid
+- **abs:** https://arxiv.org/abs/2604.01563 · **PDF:** https://arxiv.org/pdf/2604.01563v1
+
+## 224. Acoustic and perceptual differences between standard and accented Chinese speech and their voice clones
+
+- **ID:** `2604.01562` · **primary:** `cs.SD` · **Submitted:** 2026-04-02
+- **Authors:** Tianle Yang, Chengzhe Sun, Phil Rose, et al.
+- **abs:** https://arxiv.org/abs/2604.01562 · **PDF:** https://arxiv.org/pdf/2604.01562v1
+
+## 225. ReFlow: Self-correction Motion Learning for Dynamic Scene Reconstruction
+
+- **ID:** `2604.01561` · **primary:** `cs.CV` · **Submitted:** 2026-04-02
+- **Authors:** Yanzhe Liang, Ruijie Zhu, Hanzhi Chang, et al.
+- **abs:** https://arxiv.org/abs/2604.01561 · **PDF:** https://arxiv.org/pdf/2604.01561v1
+
+## 226. DeltaMem: Towards Agentic Memory Management via Reinforcement Learning
+
+- **ID:** `2604.01560` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Qi Zhang, Shen Huang, Chu Liu, et al.
+- **abs:** https://arxiv.org/abs/2604.01560 · **PDF:** https://arxiv.org/pdf/2604.01560v1
+
+## 227. EXHIB: A Benchmark for Realistic and Diverse Evaluation of Function Similarity in the Wild
+
+- **ID:** `2604.01554` · **primary:** `cs.CR` · **Submitted:** 2026-04-02
+- **Authors:** Yiming Fan, Jun Yeon Won, Ding Zhu, et al.
+- **abs:** https://arxiv.org/abs/2604.01554 · **PDF:** https://arxiv.org/pdf/2604.01554v1
+
+## 228. ZEUS: Accelerating Diffusion Models with Only Second-Order Predictor
+
+- **ID:** `2604.01552` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Yixiao Wang, Ting Jiang, Zishan Shao, et al.
+- **abs:** https://arxiv.org/abs/2604.01552 · **PDF:** https://arxiv.org/pdf/2604.01552v1
+
+## 229. RAE-AR: Taming Autoregressive Models with Representation Autoencoders
+
+- **ID:** `2604.01545` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Hu Yu, Hang Xu, Jie Huang, et al.
+- **abs:** https://arxiv.org/abs/2604.01545 · **PDF:** https://arxiv.org/pdf/2604.01545v1
+
+## 230. Countering Catastrophic Forgetting of Large Language Models for Better Instruction Following via Weight-Space Model Merging
+
+- **ID:** `2604.01538` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Mengxian Lyu, Cheng Peng, Ziyi Chen, et al.
+- **abs:** https://arxiv.org/abs/2604.01538 · **PDF:** https://arxiv.org/pdf/2604.01538v1
+
+## 231. Read More, Think More: Revisiting Observation Reduction for Web Agents
+
+- **ID:** `2604.01535` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Masafumi Enomoto, Ryoma Obara, Haochen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2604.01535 · **PDF:** https://arxiv.org/pdf/2604.01535v1
+
+## 232. PHMForge: A Scenario-Driven Agentic Benchmark for Industrial Asset Lifecycle Maintenance
+
+- **ID:** `2604.01532` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Ayan Das, Dhaval Patel
+- **abs:** https://arxiv.org/abs/2604.01532 · **PDF:** https://arxiv.org/pdf/2604.01532v1
+
+## 233. A Role-Based LLM Framework for Structured Information Extraction from Healthy Food Policies
+
+- **ID:** `2604.01529` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Congjing Zhang, Ruoxuan Bao, Jingyu Li, et al.
+- **abs:** https://arxiv.org/abs/2604.01529 · **PDF:** https://arxiv.org/pdf/2604.01529v1
+
+## 234. ProdCodeBench: A Production-Derived Benchmark for Evaluating AI Coding Agents
+
+- **ID:** `2604.01527` · **primary:** `cs.SE` · **Submitted:** 2026-04-02
+- **Authors:** Smriti Jha, Matteo Paltenghi, Chandra Maddila, et al.
+- **abs:** https://arxiv.org/abs/2604.01527 · **PDF:** https://arxiv.org/pdf/2604.01527v1
+
+## 235. A Determinantal Approach to a Sharp $\ell^1-\ell^\infty-\ell^2$ Norm Inequality
+
+- **ID:** `2604.01525` · **primary:** `math.CA` · **Submitted:** 2026-04-02
+- **Authors:** Jose Antonio Lara Benitez
+- **abs:** https://arxiv.org/abs/2604.01525 · **PDF:** https://arxiv.org/pdf/2604.01525v1
+
+## 236. Learning ECG Image Representations via Dual Physiological-Aware Alignments
+
+- **ID:** `2604.01526` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Hung Manh Pham, Jialu Tang, Aaqib Saeed, et al.
+- **abs:** https://arxiv.org/abs/2604.01526 · **PDF:** https://arxiv.org/pdf/2604.01526v1
+
+## 237. LLM Agents as Social Scientists: A Human-AI Collaborative Platform for Social Science Automation
+
+- **ID:** `2604.01520` · **primary:** `cs.AI` · **Submitted:** 2026-04-02
+- **Authors:** Lei Wang, Yuanzi Li, Jinchao Wu, et al.
+- **abs:** https://arxiv.org/abs/2604.01520 · **PDF:** https://arxiv.org/pdf/2604.01520v1
+
+## 238. Why Instruction-Based Unlearning Fails in Diffusion Models?
+
+- **ID:** `2604.01514` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Zeliang Zhang, Rui Sun, Jiani Liu, et al.
+- **abs:** https://arxiv.org/abs/2604.01514 · **PDF:** https://arxiv.org/pdf/2604.01514v1
+
+## 239. ToolMisuseBench: An Offline Deterministic Benchmark for Tool Misuse and Recovery in Agentic Systems
+
+- **ID:** `2604.01508` · **primary:** `cs.SE` · **Submitted:** 2026-04-02
+- **Authors:** Akshey Sigdel, Rista Baral
+- **abs:** https://arxiv.org/abs/2604.01508 · **PDF:** https://arxiv.org/pdf/2604.01508v1
+
+## 240. Beyond Logit Adjustment: A Residual Decomposition Framework for Long-Tailed Reranking
+
+- **ID:** `2604.01506` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** Zhanliang Wang, Hongzhuo Chen, Quan Minh Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2604.01506 · **PDF:** https://arxiv.org/pdf/2604.01506v1
+
+## 241. Magic, Madness, Heaven, Sin: LLM Output Diversity is Everything, Everywhere, All at Once
+
+- **ID:** `2604.01504` · **primary:** `cs.CL` · **Submitted:** 2026-04-02
+- **Authors:** Harnoor Dhingra
+- **abs:** https://arxiv.org/abs/2604.01504 · **PDF:** https://arxiv.org/pdf/2604.01504v1
+
+## 242. Non-monotonicity in Conformal Risk Control
+
+- **ID:** `2604.01502` · **primary:** `stat.ML` · **Submitted:** 2026-04-02
+- **Authors:** Tareq Aldirawi, Yun Li, Wenge Guo
+- **abs:** https://arxiv.org/abs/2604.01502 · **PDF:** https://arxiv.org/pdf/2604.01502v1
+
+## 243. Identifying and Estimating Causal Direct Effects Under Unmeasured Confounding
+
+- **ID:** `2604.01501` · **primary:** `stat.ME` · **Submitted:** 2026-04-02
+- **Authors:** Philippe Boileau, Nima S. Hejazi, Ivana Malenica, et al.
+- **abs:** https://arxiv.org/abs/2604.01501 · **PDF:** https://arxiv.org/pdf/2604.01501v1
+
+## 244. Matching Accuracy, Different Geometry: Evolution Strategies vs GRPO in LLM Post-Training
+
+- **ID:** `2604.01499` · **primary:** `cs.LG` · **Submitted:** 2026-04-02
+- **Authors:** William Hoy, Binxu Wang, Xu Pan
+- **abs:** https://arxiv.org/abs/2604.01499 · **PDF:** https://arxiv.org/pdf/2604.01499v1
+
+## 245. From SWE-ZERO to SWE-HERO: Execution-free to Execution-based Fine-tuning for Software Engineering Agents
+
+- **ID:** `2604.01496` · **primary:** `cs.SE` · **Submitted:** 2026-04-02
+- **Authors:** Nikolai Ludwig, Wasi Uddin Ahmad, Somshubra Majumdar, et al.
+- **abs:** https://arxiv.org/abs/2604.01496 · **PDF:** https://arxiv.org/pdf/2604.01496v1
+
+## 246. CuTeGen: An LLM-Based Agentic Framework for Generation and Optimization of High-Performance GPU Kernels using CuTe
+
+- **ID:** `2604.01489` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Tara Saba, Anne Ouyang, Xujie Si, et al.
+- **abs:** https://arxiv.org/abs/2604.01489 · **PDF:** https://arxiv.org/pdf/2604.01489v1
+
+## 247. AgentSocialBench: Evaluating Privacy Risks in Human-Centered Agentic Social Networks
+
+- **ID:** `2604.01487` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Prince Zizhuang Wang, Shuli Jiang
+- **abs:** https://arxiv.org/abs/2604.01487 · **PDF:** https://arxiv.org/pdf/2604.01487v1
+
+## 248. The topological gap at criticality: scaling exponent d + η, universality, and scope
+
+- **ID:** `2604.01484` · **primary:** `cond-mat.stat-mech` · **Submitted:** 2026-04-01
+- **Authors:** Matthew Loftus
+- **abs:** https://arxiv.org/abs/2604.01484 · **PDF:** https://arxiv.org/pdf/2604.01484v1
+
+## 249. Type-Checked Compliance: Deterministic Guardrails for Agentic Financial Systems Using Lean 4 Theorem Proving
+
+- **ID:** `2604.01483` · **primary:** `cs.LO` · **Submitted:** 2026-04-01
+- **Authors:** Devakh Rashie, Veda Rashi
+- **abs:** https://arxiv.org/abs/2604.01483 · **PDF:** https://arxiv.org/pdf/2604.01483v1
+
+## 250. DISCO-TAB: A Hierarchical Reinforcement Learning Framework for Privacy-Preserving Synthesis of Complex Clinical Data
+
+- **ID:** `2604.01481` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Arshia Ilaty, Hossein Shirazi, Amir Rahmani, et al.
+- **abs:** https://arxiv.org/abs/2604.01481 · **PDF:** https://arxiv.org/pdf/2604.01481v1
+
+## 251. A Self-Evolving Agentic Framework for Metasurface Inverse Design
+
+- **ID:** `2604.01480` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Yi Huang, Bowen Zheng, Yunxi Dong, et al.
+- **abs:** https://arxiv.org/abs/2604.01480 · **PDF:** https://arxiv.org/pdf/2604.01480v1
+
+## 252. Soft MPCritic: Amortized Model Predictive Value Iteration
+
+- **ID:** `2604.01477` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Thomas Banker, Nathan P. Lawrence, Ali Mesbah
+- **abs:** https://arxiv.org/abs/2604.01477 · **PDF:** https://arxiv.org/pdf/2604.01477v1
+
+## 253. When Reward Hacking Rebounds: Understanding and Mitigating It with Representation-Level Signals
+
+- **ID:** `2604.01476` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Rui Wu, Ruixiang Tang
+- **abs:** https://arxiv.org/abs/2604.01476 · **PDF:** https://arxiv.org/pdf/2604.01476v1
+
+## 254. Prime Once, then Reprogram Locally: An Efficient Alternative to Black-Box Service Model Adaptation
+
+- **ID:** `2604.01474` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Yunbei Zhang, Chengyi Cai, Feng Liu, et al.
+- **abs:** https://arxiv.org/abs/2604.01474 · **PDF:** https://arxiv.org/pdf/2604.01474v1
+
+## 255. SelfGrader: Stable Jailbreak Detection for Large Language Models using Token-Level Logits
+
+- **ID:** `2604.01473` · **primary:** `cs.CR` · **Submitted:** 2026-04-01
+- **Authors:** Zikai Zhang, Rui Hu, Olivera Kotevska, et al.
+- **abs:** https://arxiv.org/abs/2604.01473 · **PDF:** https://arxiv.org/pdf/2604.01473v1
+
+## 256. The Newton-Muon Optimizer
+
+- **ID:** `2604.01472` · **primary:** `math.OC` · **Submitted:** 2026-04-01
+- **Authors:** Zhehang Du, Weijie Su
+- **abs:** https://arxiv.org/abs/2604.01472 · **PDF:** https://arxiv.org/pdf/2604.01472v1
+
+## 257. Oscillator-Based Associative Memory with Exponential Capacity: Theory, Algorithms, and Hardware Implementation
+
+- **ID:** `2604.01469` · **primary:** `cs.NE` · **Submitted:** 2026-04-01
+- **Authors:** Arie Ogranovich, Taosha Guo, Arvind R. Venkatakrishnan, et al.
+- **abs:** https://arxiv.org/abs/2604.01469 · **PDF:** https://arxiv.org/pdf/2604.01469v1
+
+## 258. A Dynamic Atlas of Persian Poetic Symbolism: Families, Fields, and the Historical Rewiring of Meaning
+
+- **ID:** `2604.01467` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Kourosh Shahnazari, Seyed Moein Ayyoubzadeh, Mohammadali Keshtparvar
+- **abs:** https://arxiv.org/abs/2604.01467 · **PDF:** https://arxiv.org/pdf/2604.01467v1
+
+## 259. Efficient Equivariant Transformer for Self-Driving Agent Modeling
+
+- **ID:** `2604.01466` · **primary:** `cs.RO` · **Submitted:** 2026-04-01
+- **Authors:** Scott Xu, Dian Chen, Kelvin Wong, et al.
+- **abs:** https://arxiv.org/abs/2604.01466 · **PDF:** https://arxiv.org/pdf/2604.01466v1
+
+## 260. Low-Burden LLM-Based Preference Learning: Personalizing Assistive Robots from Natural Language Feedback for Users with Paralysis
+
+- **ID:** `2604.01463` · **primary:** `cs.RO` · **Submitted:** 2026-04-01
+- **Authors:** Keshav Shankar, Dan Ding, Wei Gao
+- **abs:** https://arxiv.org/abs/2604.01463 · **PDF:** https://arxiv.org/pdf/2604.01463v1
+
+## 261. Reducing Hallucinations in LLM-based Scientific Literature Analysis Using Peer Context Outlier Detection
+
+- **ID:** `2604.01461` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Daniel Xie, Maxwell J. Jacobson, Adil Wazeer, et al.
+- **abs:** https://arxiv.org/abs/2604.01461 · **PDF:** https://arxiv.org/pdf/2604.01461v1
+
+## 262. Koopman Subspace Pruning in Reproducing Kernel Hilbert Spaces via Principal Vectors
+
+- **ID:** `2604.01459` · **primary:** `eess.SY` · **Submitted:** 2026-04-01
+- **Authors:** Dhruv Shah, Jorge Cortes
+- **abs:** https://arxiv.org/abs/2604.01459 · **PDF:** https://arxiv.org/pdf/2604.01459v1
+
+## 263. Wired for Overconfidence: A Mechanistic Perspective on Inflated Verbalized Confidence in LLMs
+
+- **ID:** `2604.01457` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Tianyi Zhao, Yinhan He, Wendy Zheng, et al.
+- **abs:** https://arxiv.org/abs/2604.01457 · **PDF:** https://arxiv.org/pdf/2604.01457v1
+
+## 264. Infeasibility Aware Large Language Models for Combinatorial Optimization
+
+- **ID:** `2604.01455` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Yakun Wang, Min Chen, Zeguan Wu, et al.
+- **abs:** https://arxiv.org/abs/2604.01455 · **PDF:** https://arxiv.org/pdf/2604.01455v1
+
+## 265. A Multi-Agent Human-LLM Collaborative Framework for Closed-Loop Scientific Literature Summarization
+
+- **ID:** `2604.01452` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Maxwell J. Jacobson, Daniel Xie, Jackson Shen, et al.
+- **abs:** https://arxiv.org/abs/2604.01452 · **PDF:** https://arxiv.org/pdf/2604.01452v1
+
+## 266. When AI Gets it Wong: Reliability and Risk in AI-Assisted Medication Decision Systems
+
+- **ID:** `2604.01449` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Khalid Adnan Alsayed
+- **abs:** https://arxiv.org/abs/2604.01449 · **PDF:** https://arxiv.org/pdf/2604.01449v1
+
+## 267. Better Rigs, Not Bigger Networks: A Body Model Ablation for Gaussian Avatars
+
+- **ID:** `2604.01447` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Derek Austin
+- **abs:** https://arxiv.org/abs/2604.01447 · **PDF:** https://arxiv.org/pdf/2604.01447v1
+
+## 268. All Substitution Is Local
+
+- **ID:** `2604.01443` · **primary:** `econ.TH` · **Submitted:** 2026-04-01
+- **Authors:** Nidhish Shah, Shaurjya Mandal, Asfandyar Azhar
+- **abs:** https://arxiv.org/abs/2604.01443 · **PDF:** https://arxiv.org/pdf/2604.01443v1
+
+## 269. Generative Profiling for Soft Real-Time Systems and its Applications to Resource Allocation
+
+- **ID:** `2604.01441` · **primary:** `eess.SY` · **Submitted:** 2026-04-01
+- **Authors:** Georgiy A. Bondar, Abigail Eisenklam, Yifan Cai, et al.
+- **abs:** https://arxiv.org/abs/2604.01441 · **PDF:** https://arxiv.org/pdf/2604.01441v1
+
+## 270. Know Your Streams: On the Conceptualization, Characterization, and Generation of Intentional Event Streams
+
+- **ID:** `2604.01440` · **primary:** `cs.DB` · **Submitted:** 2026-04-01
+- **Authors:** Andrea Maldonado, Christian Imenkamp, Hendrik Reiter, et al.
+- **abs:** https://arxiv.org/abs/2604.01440 · **PDF:** https://arxiv.org/pdf/2604.01440v1
+
+## 271. ClawSafety: "Safe" LLMs, Unsafe Agents
+
+- **ID:** `2604.01438` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Bowen Wei, Yunbei Zhang, Jinhao Pan, et al.
+- **abs:** https://arxiv.org/abs/2604.01438 · **PDF:** https://arxiv.org/pdf/2604.01438v1
+
+## 272. Reproducible, Explainable, and Effective Evaluations of Agentic AI for Software Engineering
+
+- **ID:** `2604.01437` · **primary:** `cs.SE` · **Submitted:** 2026-04-01
+- **Authors:** Jingyue Li, André Storhaug
+- **abs:** https://arxiv.org/abs/2604.01437 · **PDF:** https://arxiv.org/pdf/2604.01437v1
+
+## 273. Leveraging the Value of Information in POMDP Planning
+
+- **ID:** `2604.01434` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Zakariya Laouar, Qi Heng Ho, Zachary Sunberg
+- **abs:** https://arxiv.org/abs/2604.01434 · **PDF:** https://arxiv.org/pdf/2604.01434v1
+
+## 274. Semantically Annotated Multimodal Dataset for RF Interpretation and Prediction
+
+- **ID:** `2604.01433` · **primary:** `cs.ET` · **Submitted:** 2026-04-01
+- **Authors:** Steve Blandino, Jelena Senic, Raied Caromi, et al.
+- **abs:** https://arxiv.org/abs/2604.01433 · **PDF:** https://arxiv.org/pdf/2604.01433v1
+
+## 275. Are Finer Citations Always Better? Rethinking Granularity for Attributed Generation
+
+- **ID:** `2604.01432` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Hexuan Wang, Jingyu Zhang, Benjamin Van Durme, et al.
+- **abs:** https://arxiv.org/abs/2604.01432 · **PDF:** https://arxiv.org/pdf/2604.01432v1
+
+## 276. Improving Latent Generalization Using Test-time Compute
+
+- **ID:** `2604.01430` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Arslan Chaudhry, Sridhar Thiagarajan, Andrew Lampinen
+- **abs:** https://arxiv.org/abs/2604.01430 · **PDF:** https://arxiv.org/pdf/2604.01430v1
+
+## 277. The power of context: Random Forest classification of near synonyms. A case study in Modern Hindi
+
+- **ID:** `2604.01425` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Jacek Bąkowski
+- **abs:** https://arxiv.org/abs/2604.01425 · **PDF:** https://arxiv.org/pdf/2604.01425v1
+
+## 278. Cost-Efficient Estimation of General Abilities Across Benchmarks
+
+- **ID:** `2604.01418` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Michael Krumdick, Adam Wiemerslage, Seth Ebner, et al.
+- **abs:** https://arxiv.org/abs/2604.01418 · **PDF:** https://arxiv.org/pdf/2604.01418v1
+
+## 279. ReFormeR: Learning and Applying Explicit Query Reformulation Patterns
+
+- **ID:** `2604.01417` · **primary:** `cs.IR` · **Submitted:** 2026-04-01
+- **Authors:** Amin Bigdeli, Mert Incesu, Negar Arabzadeh, et al.
+- **abs:** https://arxiv.org/abs/2604.01417 · **PDF:** https://arxiv.org/pdf/2604.01417v1
+
+## 280. Adaptive Stopping for Multi-Turn LLM Reasoning
+
+- **ID:** `2604.01413` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Xiaofan Zhou, Huy Nguyen, Bo Yu, et al.
+- **abs:** https://arxiv.org/abs/2604.01413 · **PDF:** https://arxiv.org/pdf/2604.01413v1
+
+## 281. Test-Time Scaling Makes Overtraining Compute-Optimal
+
+- **ID:** `2604.01411` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Nicholas Roberts, Sungjun Cho, Zhiqi Gao, et al.
+- **abs:** https://arxiv.org/abs/2604.01411 · **PDF:** https://arxiv.org/pdf/2604.01411v1
+
+## 282. Assessing Pause Thresholds for empirical Translation Process Research
+
+- **ID:** `2604.01410` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Devi Sri Bandaru, Michael Carl, Xinyue Ren
+- **abs:** https://arxiv.org/abs/2604.01410 · **PDF:** https://arxiv.org/pdf/2604.01410v1
+
+## 283. Causal Optimal Coupling for Gaussian Input-Output Distributional Data
+
+- **ID:** `2604.01406` · **primary:** `eess.SY` · **Submitted:** 2026-04-01
+- **Authors:** Daran Xu, Amirhossein Taghvaei
+- **abs:** https://arxiv.org/abs/2604.01406 · **PDF:** https://arxiv.org/pdf/2604.01406v1
+
+## 284. Friends and Grandmothers in Silico: Localizing Entity Cells in Language Models
+
+- **ID:** `2604.01404` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Itay Yona, Dan Barzilay, Michael Karasik, et al.
+- **abs:** https://arxiv.org/abs/2604.01404 · **PDF:** https://arxiv.org/pdf/2604.01404v1
+
+## 285. Benchmark Problems and Benchmark Datasets for the evaluation of Machine and Deep Learning methods on Photoplethysmography signals: the D4 report from the QUMPHY project
+
+- **ID:** `2604.01398` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Urs Hackstein, Jordi Alastruey, Philip Aston, et al.
+- **abs:** https://arxiv.org/abs/2604.01398 · **PDF:** https://arxiv.org/pdf/2604.01398v1
+
+## 286. GRAZE: Grounded Refinement and Motion-Aware Zero-Shot Event Localization
+
+- **ID:** `2604.01383` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Syed Ahsan Masud Zaidi, Lior Shamir, William Hsu, et al.
+- **abs:** https://arxiv.org/abs/2604.01383 · **PDF:** https://arxiv.org/pdf/2604.01383v1
+
+## 287. Can LLMs Predict Academic Collaboration? Topology Heuristics vs. LLM-Based Link Prediction on Real Co-authorship Networks
+
+- **ID:** `2604.01379` · **primary:** `cs.SI` · **Submitted:** 2026-04-01
+- **Authors:** Fan Huang, Munjung Kim
+- **abs:** https://arxiv.org/abs/2604.01379 · **PDF:** https://arxiv.org/pdf/2604.01379v1
+
+## 288. Residuals-based Offline Reinforcement Learning
+
+- **ID:** `2604.01378` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Qing Zhu, Xian Yu
+- **abs:** https://arxiv.org/abs/2604.01378 · **PDF:** https://arxiv.org/pdf/2604.01378v1
+
+## 289. RIFT: A RubrIc Failure Mode Taxonomy and Automated Diagnostics
+
+- **ID:** `2604.01375` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Zhengyang Qi, Charles Dickens, Derek Pham, et al.
+- **abs:** https://arxiv.org/abs/2604.01375 · **PDF:** https://arxiv.org/pdf/2604.01375v1
+
+## 290. AffordTissue: Dense Affordance Prediction for Tool-Action Specific Tissue Interaction
+
+- **ID:** `2604.01371` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Aiza Maksutova, Lalithkumar Seenivasan, Hao Ding, et al.
+- **abs:** https://arxiv.org/abs/2604.01371 · **PDF:** https://arxiv.org/pdf/2604.01371v1
+
+## 291. CogBias: Measuring and Mitigating Cognitive Bias in Large Language Models
+
+- **ID:** `2604.01366` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Fan Huang, Songheng Zhang, Haewoon Kwak, et al.
+- **abs:** https://arxiv.org/abs/2604.01366 · **PDF:** https://arxiv.org/pdf/2604.01366v1
+
+## 292. VIANA: character Value-enhanced Intensity Assessment via domain-informed Neural Architecture
+
+- **ID:** `2604.01365` · **primary:** `physics.chem-ph` · **Submitted:** 2026-04-01
+- **Authors:** Luana P. Queiroz, Icaro S. C. Bernardes, Ana M. Ribeiro, et al.
+- **abs:** https://arxiv.org/abs/2604.01365 · **PDF:** https://arxiv.org/pdf/2604.01365v1
+
+## 293. From Automation to Augmentation: A Framework for Designing Human-Centric Work Environments in Society 5.0
+
+- **ID:** `2604.01364` · **primary:** `econ.GN` · **Submitted:** 2026-04-01
+- **Authors:** Cristian Espinal Maya
+- **abs:** https://arxiv.org/abs/2604.01364 · **PDF:** https://arxiv.org/pdf/2604.01364v1
+
+## 294. Crashing Waves vs. Rising Tides: Preliminary Findings on AI Automation from Thousands of Worker Evaluations of Labor Market Tasks
+
+- **ID:** `2604.01363` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Matthias Mertens, Adam Kuzee, Brittany S. Harris, et al.
+- **abs:** https://arxiv.org/abs/2604.01363 · **PDF:** https://arxiv.org/pdf/2604.01363v1
+
+## 295. Semantic Modeling for World-Centered Architectures
+
+- **ID:** `2604.01359` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Andrei Mantsivoda, Darya Gavrilina
+- **abs:** https://arxiv.org/abs/2604.01359 · **PDF:** https://arxiv.org/pdf/2604.01359v1
+
+## 296. Open-Domain Safety Policy Construction
+
+- **ID:** `2604.01354` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Di Wu, Siyue Liu, Zixiang Ji, et al.
+- **abs:** https://arxiv.org/abs/2604.01354 · **PDF:** https://arxiv.org/pdf/2604.01354v1
+
+## 297. No Attacker Needed: Unintentional Cross-User Contamination in Shared-State LLM Agents
+
+- **ID:** `2604.01350` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Tiankai Yang, Jiate Li, Yi Nian, et al.
+- **abs:** https://arxiv.org/abs/2604.01350 · **PDF:** https://arxiv.org/pdf/2604.01350v1
+
+## 298. PI-JEPA: Label-Free Surrogate Pretraining for Coupled Multiphysics Simulation via Operator-Split Latent Prediction
+
+- **ID:** `2604.01349` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Brandon Yee, Pairie Koh
+- **abs:** https://arxiv.org/abs/2604.01349 · **PDF:** https://arxiv.org/pdf/2604.01349v1
+
+## 299. Procedural Knowledge at Scale Improves Reasoning
+
+- **ID:** `2604.01348` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Di Wu, Devendra Singh Sachan, Wen-tau Yih, et al.
+- **abs:** https://arxiv.org/abs/2604.01348 · **PDF:** https://arxiv.org/pdf/2604.01348v1
+
+## 300. Safety, Security, and Cognitive Risks in World Models
+
+- **ID:** `2604.01346` · **primary:** `cs.CR` · **Submitted:** 2026-04-01
+- **Authors:** Manoj Parmar
+- **abs:** https://arxiv.org/abs/2604.01346 · **PDF:** https://arxiv.org/pdf/2604.01346v1
+
+## 301. Malliavin Calculus for Counterfactual Gradient Estimation in Adaptive Inverse Reinforcement Learning
+
+- **ID:** `2604.01345` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Vikram Krishnamurthy, Luke Snow
+- **abs:** https://arxiv.org/abs/2604.01345 · **PDF:** https://arxiv.org/pdf/2604.01345v1
+
+## 302. IDEA2: Expert-in-the-loop competency question elicitation for collaborative ontology engineering
+
+- **ID:** `2604.01344` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Elliott Watkiss-Leek, Reham Alharbi, Harry Rostron, et al.
+- **abs:** https://arxiv.org/abs/2604.01344 · **PDF:** https://arxiv.org/pdf/2604.01344v1
+
+## 303. Massively Parallel Exact Inference for Hawkes Processes
+
+- **ID:** `2604.01342` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Ahmer Raza, Hudson Smith
+- **abs:** https://arxiv.org/abs/2604.01342 · **PDF:** https://arxiv.org/pdf/2604.01342v1
+
+## 304. Regularizing Attention Scores with Bootstrapping
+
+- **ID:** `2604.01339` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Neo Christopher Chung, Maxim Laletin
+- **abs:** https://arxiv.org/abs/2604.01339 · **PDF:** https://arxiv.org/pdf/2604.01339v1
+
+## 305. SECURE: Stable Early Collision Understanding via Robust Embeddings in Autonomous Driving
+
+- **ID:** `2604.01337` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Wenjing Wang, Wenxuan Wang, Songning Lai
+- **abs:** https://arxiv.org/abs/2604.01337 · **PDF:** https://arxiv.org/pdf/2604.01337v1
+
+## 306. Bias Inheritance in Neural-Symbolic Discovery of Constitutive Closures Under Function-Class Mismatch
+
+- **ID:** `2604.01335` · **primary:** `cs.CE` · **Submitted:** 2026-04-01
+- **Authors:** Hanbing Liang, Ze Tao, Fujun Liu
+- **abs:** https://arxiv.org/abs/2604.01335 · **PDF:** https://arxiv.org/pdf/2604.01335v1
+
+## 307. Evolutionary Multi-Objective Fusion of Deepfake Speech Detectors
+
+- **ID:** `2604.01330` · **primary:** `cs.SD` · **Submitted:** 2026-04-01
+- **Authors:** Vojtěch Staněk, Martin Perešíni, Lukáš Sekanina, et al.
+- **abs:** https://arxiv.org/abs/2604.01330 · **PDF:** https://arxiv.org/pdf/2604.01330v1
+
+## 308. Model Merging via Data-Free Covariance Estimation
+
+- **ID:** `2604.01329` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Marawan Gamal Abdel Hameed, Derek Tam, Pascal Jr Tikeng Notsawo, et al.
+- **abs:** https://arxiv.org/abs/2604.01329 · **PDF:** https://arxiv.org/pdf/2604.01329v1
+
+## 309. Efficient and Principled Scientific Discovery through Bayesian Optimization: A Tutorial
+
+- **ID:** `2604.01328` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Zhongwei Yu, Rasul Tutunov, Alexandre Max Maraval, et al.
+- **abs:** https://arxiv.org/abs/2604.01328 · **PDF:** https://arxiv.org/pdf/2604.01328v1
+
+## 310. Macroscopic transport patterns of UAV traffic in 3D anisotropic wind fields: A constraint-preserving hybrid PINN-FVM approach
+
+- **ID:** `2604.01327` · **primary:** `cs.CE` · **Submitted:** 2026-04-01
+- **Authors:** Hanbing Liang, Fujun Liu
+- **abs:** https://arxiv.org/abs/2604.01327 · **PDF:** https://arxiv.org/pdf/2604.01327v1
+
+## 311. The Digital Twin Counterfactual Framework: A Validation Architecture for Simulated Potential Outcomes
+
+- **ID:** `2604.01325` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Olav Laudy
+- **abs:** https://arxiv.org/abs/2604.01325 · **PDF:** https://arxiv.org/pdf/2604.01325v1
+
+## 312. Detecting Complex Money Laundering Patterns with Incremental and Distributed Graph Modeling
+
+- **ID:** `2604.01315` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Haseeb Tariq, Alen Kaja, Marwan Hassani
+- **abs:** https://arxiv.org/abs/2604.01315 · **PDF:** https://arxiv.org/pdf/2604.01315v1
+
+## 313. JetPrism: diagnosing convergence for generative simulation and inverse problems in nuclear physics
+
+- **ID:** `2604.01313` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Zeyu Xia, Tyler Kim, Trevor Reed, et al.
+- **abs:** https://arxiv.org/abs/2604.01313 · **PDF:** https://arxiv.org/pdf/2604.01313v1
+
+## 314. Preference learning in shades of gray: Interpretable and bias-aware reward modeling for human preferences
+
+- **ID:** `2604.01312` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Simona-Vasilica Oprea, Adela Bâra
+- **abs:** https://arxiv.org/abs/2604.01312 · **PDF:** https://arxiv.org/pdf/2604.01312v1
+
+## 315. An Online Machine Learning Multi-resolution Optimization Framework for Energy System Design Limit of Performance Analysis
+
+- **ID:** `2604.01308` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Oluwamayowa O. Amusat, Luka Grbcic, Remi Patureau, et al.
+- **abs:** https://arxiv.org/abs/2604.01308 · **PDF:** https://arxiv.org/pdf/2604.01308v1
+
+## 316. M2-Verify: A Large-Scale Multidomain Benchmark for Checking Multimodal Claim Consistency
+
+- **ID:** `2604.01306` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Abolfazl Ansari, Delvin Ce Zhang, Zhuoyang Zou, et al.
+- **abs:** https://arxiv.org/abs/2604.01306 · **PDF:** https://arxiv.org/pdf/2604.01306v1
+
+## 317. UQ-SHRED: uncertainty quantification of shallow recurrent decoder networks for sparse sensing via engression
+
+- **ID:** `2604.01305` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Mars Liyao Gao, Yuxuan Bao, Amy S. Rude, et al.
+- **abs:** https://arxiv.org/abs/2604.01305 · **PDF:** https://arxiv.org/pdf/2604.01305v1
+
+## 318. Scaling Reasoning Tokens via RL and Parallel Thinking: Evidence From Competitive Programming
+
+- **ID:** `2604.01302` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Qianfan Zhang, Tianyu Guo, Xuandi Ren, et al.
+- **abs:** https://arxiv.org/abs/2604.01302 · **PDF:** https://arxiv.org/pdf/2604.01302v1
+
+## 319. Forecasting Supply Chain Disruptions with Foresight Learning
+
+- **ID:** `2604.01298` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Benjamin Turtel, Paul Wilczewski, Kris Skotheim
+- **abs:** https://arxiv.org/abs/2604.01298 · **PDF:** https://arxiv.org/pdf/2604.01298v1
+
+## 320. Look Twice: Training-Free Evidence Highlighting in Multimodal Large Language Models
+
+- **ID:** `2604.01280` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Marco Morini, Sara Sarto, Marcella Cornia, et al.
+- **abs:** https://arxiv.org/abs/2604.01280 · **PDF:** https://arxiv.org/pdf/2604.01280v1
+
+## 321. Sven: Singular Value Descent as a Computationally Efficient Natural Gradient Method
+
+- **ID:** `2604.01279` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Samuel Bright-Thonney, Thomas R. Harvey, Andre Lukas, et al.
+- **abs:** https://arxiv.org/abs/2604.01279 · **PDF:** https://arxiv.org/pdf/2604.01279v1
+
+## 322. Descending into the Modular Bootstrap
+
+- **ID:** `2604.01275` · **primary:** `hep-th` · **Submitted:** 2026-04-01
+- **Authors:** Nathan Benjamin, A. Liam Fitzpatrick, Wei Li, et al.
+- **abs:** https://arxiv.org/abs/2604.01275 · **PDF:** https://arxiv.org/pdf/2604.01275v1
+
+## 323. HippoCamp: Benchmarking Contextual Agents on Personal Computers
+
+- **ID:** `2604.01221` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Zhe Yang, Shulin Tian, Kairui Hu, et al.
+- **abs:** https://arxiv.org/abs/2604.01221 · **PDF:** https://arxiv.org/pdf/2604.01221v1
+
+## 324. Universal YOCO for Efficient Depth Scaling
+
+- **ID:** `2604.01220` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Yutao Sun, Li Dong, Tianzhu Ye, et al.
+- **abs:** https://arxiv.org/abs/2604.01220 · **PDF:** https://arxiv.org/pdf/2604.01220v1
+
+## 325. LAtent Phase Inference from Short time sequences using SHallow REcurrent Decoders (LAPIS-SHRED)
+
+- **ID:** `2604.01216` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Yuxuan Bao, Xingyue Zhang, J. Nathan Kutz
+- **abs:** https://arxiv.org/abs/2604.01216 · **PDF:** https://arxiv.org/pdf/2604.01216v1
+
+## 326. The Recipe Matters More Than the Kitchen:Mathematical Foundations of the AI Weather Prediction Pipeline
+
+- **ID:** `2604.01215` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Piyush Garg, Diana R. Gergel, Andrew E. Shao, et al.
+- **abs:** https://arxiv.org/abs/2604.01215 · **PDF:** https://arxiv.org/pdf/2604.01215v1
+
+## 327. $\texttt{YC-Bench}$: Benchmarking AI Agents for Long-Term Planning and Consistent Execution
+
+- **ID:** `2604.01212` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Muyu He, Adit Jain, Anand Kumar, et al.
+- **abs:** https://arxiv.org/abs/2604.01212 · **PDF:** https://arxiv.org/pdf/2604.01212v1
+
+## 328. CliffSearch: Structured Agentic Co-Evolution over Theory and Code for Scientific Algorithm Discovery
+
+- **ID:** `2604.01210` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Youssef Mroueh, Carlos Fonseca, Brian Belgodere, et al.
+- **abs:** https://arxiv.org/abs/2604.01210 · **PDF:** https://arxiv.org/pdf/2604.01210v1
+
+## 329. LLM REgression with a Latent Iterative State Head
+
+- **ID:** `2604.01206` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Yiheng Su, Matthew Lease
+- **abs:** https://arxiv.org/abs/2604.01206 · **PDF:** https://arxiv.org/pdf/2604.01206v1
+
+## 330. Neural Harmonic Textures for High-Quality Primitive Based Neural Reconstruction
+
+- **ID:** `2604.01204` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Jorge Condor, Nicolas Moenne-Loccoz, Merlin Nimier-David, et al.
+- **abs:** https://arxiv.org/abs/2604.01204 · **PDF:** https://arxiv.org/pdf/2604.01204v1
+
+## 331. Therefore I am. I Think
+
+- **ID:** `2604.01202` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Esakkivel Esakkiraja, Sai Rajeswar, Denis Akhiyarov, et al.
+- **abs:** https://arxiv.org/abs/2604.01202 · **PDF:** https://arxiv.org/pdf/2604.01202v2
+
+## 332. Learning and Generating Mixed States Prepared by Shallow Channel Circuits
+
+- **ID:** `2604.01197` · **primary:** `quant-ph` · **Submitted:** 2026-04-01
+- **Authors:** Fangjun Hu, Christian Kokail, Milan Kornjača, et al.
+- **abs:** https://arxiv.org/abs/2604.01197 · **PDF:** https://arxiv.org/pdf/2604.01197v1
+
+## 333. ORBIT: Scalable and Verifiable Data Generation for Search Agents on a Tight Budget
+
+- **ID:** `2604.01195` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Nandan Thakur, Zijian Chen, Xueguang Ma, et al.
+- **abs:** https://arxiv.org/abs/2604.01195 · **PDF:** https://arxiv.org/pdf/2604.01195v2
+
+## 334. Embarrassingly Simple Self-Distillation Improves Code Generation
+
+- **ID:** `2604.01193` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Ruixiang Zhang, Richard He Bai, Huangjie Zheng, et al.
+- **abs:** https://arxiv.org/abs/2604.01193 · **PDF:** https://arxiv.org/pdf/2604.01193v1
+
+## 335. True (VIS) Lies: Analyzing How Generative AI Recognizes Intentionality, Rhetoric, and Misleadingness in Visualization Lies
+
+- **ID:** `2604.01181` · **primary:** `cs.HC` · **Submitted:** 2026-04-01
+- **Authors:** Graziano Blasilli, Marco Angelini
+- **abs:** https://arxiv.org/abs/2604.01181 · **PDF:** https://arxiv.org/pdf/2604.01181v1
+
+## 336. A ROS 2 Wrapper for Florence-2: Multi-Mode Local Vision-Language Inference for Robotic Systems
+
+- **ID:** `2604.01179` · **primary:** `cs.RO` · **Submitted:** 2026-04-01
+- **Authors:** J. E. Domínguez-Vidal
+- **abs:** https://arxiv.org/abs/2604.01179 · **PDF:** https://arxiv.org/pdf/2604.01179v1
+
+## 337. Screening Is Enough
+
+- **ID:** `2604.01178` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Ken M. Nakanishi
+- **abs:** https://arxiv.org/abs/2604.01178 · **PDF:** https://arxiv.org/pdf/2604.01178v1
+
+## 338. NeuroDDAF: Neural Dynamic Diffusion-Advection Fields with Evidential Fusion for Air Quality Forecasting
+
+- **ID:** `2604.01175` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Prasanjit Dey, Soumyabrata Dev, Angela Meyer, et al.
+- **abs:** https://arxiv.org/abs/2604.01175 · **PDF:** https://arxiv.org/pdf/2604.01175v1
+
+## 339. Safe learning-based control via function-based uncertainty quantification
+
+- **ID:** `2604.01173` · **primary:** `eess.SY` · **Submitted:** 2026-04-01
+- **Authors:** Abdullah Tokmak, Toni Karvonen, Thomas B. Schön, et al.
+- **abs:** https://arxiv.org/abs/2604.01173 · **PDF:** https://arxiv.org/pdf/2604.01173v1
+
+## 340. Online Reasoning Calibration: Test-Time Training Enables Generalizable Conformal LLM Reasoning
+
+- **ID:** `2604.01170` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Cai Zhou, Zekai Wang, Menghua Wu, et al.
+- **abs:** https://arxiv.org/abs/2604.01170 · **PDF:** https://arxiv.org/pdf/2604.01170v1
+
+## 341. Bridging the Simulation-to-Experiment Gap with Generative Models using Adversarial Distribution Alignment
+
+- **ID:** `2604.01169` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Kai Nelson, Tobias Kreiman, Sergey Levine, et al.
+- **abs:** https://arxiv.org/abs/2604.01169 · **PDF:** https://arxiv.org/pdf/2604.01169v1
+
+## 342. S0 Tuning: Zero-Overhead Adaptation of Hybrid Recurrent-Attention Models
+
+- **ID:** `2604.01168` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Jack Young
+- **abs:** https://arxiv.org/abs/2604.01168 · **PDF:** https://arxiv.org/pdf/2604.01168v1
+
+## 343. AdaLoRA-QAT: Adaptive Low-Rank and Quantization-Aware Segmentation
+
+- **ID:** `2604.01167` · **primary:** `eess.IV` · **Submitted:** 2026-04-01
+- **Authors:** Prantik Deb, Srimanth Dhondy, N. Ramakrishna, et al.
+- **abs:** https://arxiv.org/abs/2604.01167 · **PDF:** https://arxiv.org/pdf/2604.01167v1
+
+## 344. Reasoning Shift: How Context Silently Shortens LLM Reasoning
+
+- **ID:** `2604.01161` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Gleb Rodionov
+- **abs:** https://arxiv.org/abs/2604.01161 · **PDF:** https://arxiv.org/pdf/2604.01161v1
+
+## 345. Property-Level Flood Risk Assessment Using AI-Enabled Street-View Lowest Floor Elevation Extraction and ML Imputation Across Texas
+
+- **ID:** `2604.01153` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Xiangpeng Li, Yu-Hsuan Ho, Sam D Brody, et al.
+- **abs:** https://arxiv.org/abs/2604.01153 · **PDF:** https://arxiv.org/pdf/2604.01153v2
+
+## 346. Brainstacks: Cross-Domain Cognitive Capabilities via Frozen MoE-LoRA Stacks for Continual LLM Learning
+
+- **ID:** `2604.01152` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Mohammad R. Abu Ayyash
+- **abs:** https://arxiv.org/abs/2604.01152 · **PDF:** https://arxiv.org/pdf/2604.01152v1
+
+## 347. Detecting Multi-Agent Collusion Through Multi-Agent Interpretability
+
+- **ID:** `2604.01151` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Aaron Rose, Carissa Cullen, Brandon Gary Kaplowitz, et al.
+- **abs:** https://arxiv.org/abs/2604.01151 · **PDF:** https://arxiv.org/pdf/2604.01151v1
+
+## 348. Deep Reinforcement Learning for Robotic Manipulation under Distribution Shift with Bounded Extremum Seeking
+
+- **ID:** `2604.01142` · **primary:** `cs.RO` · **Submitted:** 2026-04-01
+- **Authors:** Shaifalee Saxena, Rafael Fierro, Alexander Scheinker
+- **abs:** https://arxiv.org/abs/2604.01142 · **PDF:** https://arxiv.org/pdf/2604.01142v1
+
+## 349. Looking into a Pixel by Nonlinear Unmixing -- A Generative Approach
+
+- **ID:** `2604.01141` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Maofeng Tang, Hairong Qi
+- **abs:** https://arxiv.org/abs/2604.01141 · **PDF:** https://arxiv.org/pdf/2604.01141v1
+
+## 350. The Overlooked Repetitive Lengthening Form in Sentiment Analysis
+
+- **ID:** `2604.01268` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Lei Wang, Eduard Dragut
+- **abs:** https://arxiv.org/abs/2604.01268 · **PDF:** https://arxiv.org/pdf/2604.01268v1
+
+## 351. Toward Personalized Darts Training: A Data-Driven Framework Based on Skeleton-Based Biomechanical Analysis and Motion Modeling
+
+- **ID:** `2604.01130` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Zhantao Chen, Dongyi He, Jin Fang, et al.
+- **abs:** https://arxiv.org/abs/2604.01130 · **PDF:** https://arxiv.org/pdf/2604.01130v2
+
+## 352. Paper Reconstruction Evaluation: Evaluating Presentation and Hallucination in AI-written Papers
+
+- **ID:** `2604.01128` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Atsuyuki Miyai, Mashiro Toyooka, Zaiying Zhao, et al.
+- **abs:** https://arxiv.org/abs/2604.01128 · **PDF:** https://arxiv.org/pdf/2604.01128v1
+
+## 353. Lightweight Prompt-Guided CLIP Adaptation for Monocular Depth Estimation
+
+- **ID:** `2604.01118` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Reyhaneh Ahani Manghotay, Jie Liang
+- **abs:** https://arxiv.org/abs/2604.01118 · **PDF:** https://arxiv.org/pdf/2604.01118v1
+
+## 354. Reconsidering Dependency Networks from an Information Geometry Perspective
+
+- **ID:** `2604.01117` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Kazuya Takabatake, Shotaro Akaho
+- **abs:** https://arxiv.org/abs/2604.01117 · **PDF:** https://arxiv.org/pdf/2604.01117v1
+
+## 355. Trust and Reliance on AI in Education: AI Literacy and Need for Cognition as Moderators
+
+- **ID:** `2604.01114` · **primary:** `cs.HC` · **Submitted:** 2026-04-01
+- **Authors:** Griffin Pitts, Neha Rani, Weedguet Mildort
+- **abs:** https://arxiv.org/abs/2604.01114 · **PDF:** https://arxiv.org/pdf/2604.01114v1
+
+## 356. CARE: Privacy-Compliant Agentic Reasoning with Evidence Discordance
+
+- **ID:** `2604.01113` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Haochen Liu, Weien Li, Rui Song, et al.
+- **abs:** https://arxiv.org/abs/2604.01113 · **PDF:** https://arxiv.org/pdf/2604.01113v1
+
+## 357. Adversarial Moral Stress Testing of Large Language Models
+
+- **ID:** `2604.01108` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Saeid Jamshidi, Foutse Khomh, Arghavan Moradi Dakhel, et al.
+- **abs:** https://arxiv.org/abs/2604.01108 · **PDF:** https://arxiv.org/pdf/2604.01108v1
+
+## 358. Inverse Design of Optical Multilayer Thin Films using Robust Masked Diffusion Models
+
+- **ID:** `2604.01106` · **primary:** `physics.optics` · **Submitted:** 2026-04-01
+- **Authors:** Jonas Schaible, Asena Karolin Özdemir, Charlotte Debus, et al.
+- **abs:** https://arxiv.org/abs/2604.01106 · **PDF:** https://arxiv.org/pdf/2604.01106v1
+
+## 359. Approximating Pareto Frontiers in Stochastic Multi-Objective Optimization via Hashing and Randomization
+
+- **ID:** `2604.01098` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Jinzhao Li, Nan Jiang, Yexiang Xue
+- **abs:** https://arxiv.org/abs/2604.01098 · **PDF:** https://arxiv.org/pdf/2604.01098v1
+
+## 360. Temporal Dependencies in In-Context Learning: The Role of Induction Heads
+
+- **ID:** `2604.01094` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Anooshka Bajaj, Deven Mahesh Mistry, Sahaj Singh Maini, et al.
+- **abs:** https://arxiv.org/abs/2604.01094 · **PDF:** https://arxiv.org/pdf/2604.01094v1
+
+## 361. TRACE: Training-Free Partial Audio Deepfake Detection via Embedding Trajectory Analysis of Speech Foundation Models
+
+- **ID:** `2604.01083` · **primary:** `cs.SD` · **Submitted:** 2026-04-01
+- **Authors:** Awais Khan, Muhammad Umar Farooq, Kutub Uddin, et al.
+- **abs:** https://arxiv.org/abs/2604.01083 · **PDF:** https://arxiv.org/pdf/2604.01083v1
+
+## 362. ProOOD: Prototype-Guided Out-of-Distribution 3D Occupancy Prediction
+
+- **ID:** `2604.01081` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Yuheng Zhang, Mengfei Duan, Kunyu Peng, et al.
+- **abs:** https://arxiv.org/abs/2604.01081 · **PDF:** https://arxiv.org/pdf/2604.01081v1
+
+## 363. A Hierarchical Importance-Guided Multi-objective Evolutionary Framework for Deep Neural Network Pruning
+
+- **ID:** `2604.01076` · **primary:** `cs.NE` · **Submitted:** 2026-04-01
+- **Authors:** Zak Khan, Azam Asilian Bidgoli
+- **abs:** https://arxiv.org/abs/2604.01076 · **PDF:** https://arxiv.org/pdf/2604.01076v1
+
+## 364. Narrative Fingerprints: Multi-Scale Author Identification via Novelty Curve Dynamics
+
+- **ID:** `2604.01073` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Fred Zimmerman, Hilmar AI
+- **abs:** https://arxiv.org/abs/2604.01073 · **PDF:** https://arxiv.org/pdf/2604.01073v1
+
+## 365. VibeGuard: A Security Gate Framework for AI-Generated Code
+
+- **ID:** `2604.01052` · **primary:** `cs.CR` · **Submitted:** 2026-04-01
+- **Authors:** Ying Xie
+- **abs:** https://arxiv.org/abs/2604.01052 · **PDF:** https://arxiv.org/pdf/2604.01052v1
+
+## 366. Adversarial Attacks in AI-Driven RAN Slicing: SLA Violations and Recovery
+
+- **ID:** `2604.01049` · **primary:** `cs.NI` · **Submitted:** 2026-04-01
+- **Authors:** Deemah H. Tashman, Soumaya Cherkaoui
+- **abs:** https://arxiv.org/abs/2604.01049 · **PDF:** https://arxiv.org/pdf/2604.01049v1
+
+## 367. Integer-State Dynamics of Quantized Spiking Neural Networks for Efficient Hardware Acceleration
+
+- **ID:** `2604.01042` · **primary:** `cs.NE` · **Submitted:** 2026-04-01
+- **Authors:** Lei Zhang
+- **abs:** https://arxiv.org/abs/2604.01042 · **PDF:** https://arxiv.org/pdf/2604.01042v1
+
+## 368. Automated Framework to Evaluate and Harden LLM System Instructions against Encoding Attacks
+
+- **ID:** `2604.01039` · **primary:** `cs.CR` · **Submitted:** 2026-04-01
+- **Authors:** Anubhab Sahu, Diptisha Samanta, Reza Soosahabi
+- **abs:** https://arxiv.org/abs/2604.01039 · **PDF:** https://arxiv.org/pdf/2604.01039v1
+
+## 369. Aligning Recommendations with User Popularity Preferences
+
+- **ID:** `2604.01036` · **primary:** `cs.IR` · **Submitted:** 2026-04-01
+- **Authors:** Mona Schirmer, Anton Thielmann, Pola Schwöbel, et al.
+- **abs:** https://arxiv.org/abs/2604.01036 · **PDF:** https://arxiv.org/pdf/2604.01036v1
+
+## 370. Revision or Re-Solving? Decomposing Second-Pass Gains in Multi-LLM Pipelines
+
+- **ID:** `2604.01029` · **primary:** `cs.SE` · **Submitted:** 2026-04-01
+- **Authors:** Jingjie Ning, Xueqi Li, Chengyu Yu
+- **abs:** https://arxiv.org/abs/2604.01029 · **PDF:** https://arxiv.org/pdf/2604.01029v1
+
+## 371. Fast and Accurate Probing of In-Training LLMs' Downstream Performances
+
+- **ID:** `2604.01025` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Zhichen Liu, Tianle Lun, Zhibin Wen, et al.
+- **abs:** https://arxiv.org/abs/2604.01025 · **PDF:** https://arxiv.org/pdf/2604.01025v1
+
+## 372. Model-Based Learning of Near-Optimal Finite-Window Policies in POMDPs
+
+- **ID:** `2604.01024` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Philip Jordan, Maryam Kamgarpour
+- **abs:** https://arxiv.org/abs/2604.01024 · **PDF:** https://arxiv.org/pdf/2604.01024v1
+
+## 373. Transfer learning for nonparametric Bayesian networks
+
+- **ID:** `2604.01021` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Rafael Sojo, Pedro Larrañaga, Concha Bielza
+- **abs:** https://arxiv.org/abs/2604.01021 · **PDF:** https://arxiv.org/pdf/2604.01021v1
+
+## 374. Observable Geometry of Singular Statistical Models
+
+- **ID:** `2604.01267` · **primary:** `math.ST` · **Submitted:** 2026-04-01
+- **Authors:** Sean Plummer
+- **abs:** https://arxiv.org/abs/2604.01267 · **PDF:** https://arxiv.org/pdf/2604.01267v1
+
+## 375. OrgAgent: Organize Your Multi-Agent System like a Company
+
+- **ID:** `2604.01020` · **primary:** `cs.MA` · **Submitted:** 2026-04-01
+- **Authors:** Yiru Wang, Xinyue Shen, Yaohui Han, et al.
+- **abs:** https://arxiv.org/abs/2604.01020 · **PDF:** https://arxiv.org/pdf/2604.01020v1
+
+## 376. Omni-SimpleMem: Autoresearch-Guided Discovery of Lifelong Multimodal Agent Memory
+
+- **ID:** `2604.01007` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Jiaqi Liu, Zipeng Ling, Shi Qiu, et al.
+- **abs:** https://arxiv.org/abs/2604.01007 · **PDF:** https://arxiv.org/pdf/2604.01007v2
+
+## 377. Query-Conditioned Evidential Keyframe Sampling for MLLM-Based Long-Form Video Understanding
+
+- **ID:** `2604.01002` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Yiheng Wang, Lichen Zhu, Yueqian Lin, et al.
+- **abs:** https://arxiv.org/abs/2604.01002 · **PDF:** https://arxiv.org/pdf/2604.01002v1
+
+## 378. EgoSim: Egocentric World Simulator for Embodied Interaction Generation
+
+- **ID:** `2604.01001` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Jinkun Hao, Mingda Jia, Ruiyan Wang, et al.
+- **abs:** https://arxiv.org/abs/2604.01001 · **PDF:** https://arxiv.org/pdf/2604.01001v1
+
+## 379. EmbedPart: Embedding-Driven Graph Partitioning for Scalable Graph Neural Network Training
+
+- **ID:** `2604.01000` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Nikolai Merkel, Ruben Mayer, Volker Markl, et al.
+- **abs:** https://arxiv.org/abs/2604.01000 · **PDF:** https://arxiv.org/pdf/2604.01000v1
+
+## 380. Uncertainty-Aware Variational Reward Factorization via Probabilistic Preference Bases for LLM Personalization
+
+- **ID:** `2604.00997` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Gyuseok Lee, Wonbin Kweon, Zhenrui Yue, et al.
+- **abs:** https://arxiv.org/abs/2604.00997 · **PDF:** https://arxiv.org/pdf/2604.00997v1
+
+## 381. Multimodal Analysis of State-Funded News Coverage of the Israel-Hamas War on YouTube Shorts
+
+- **ID:** `2604.00994` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Daniel Miehling, Sandra Kuebler
+- **abs:** https://arxiv.org/abs/2604.00994 · **PDF:** https://arxiv.org/pdf/2604.00994v1
+
+## 382. Focal plane wavefront control with model-based reinforcement learning
+
+- **ID:** `2604.00993` · **primary:** `astro-ph.IM` · **Submitted:** 2026-04-01
+- **Authors:** Jalo Nousiainen, Iremsu Taskin, Markus Kasper, et al.
+- **abs:** https://arxiv.org/abs/2604.00993 · **PDF:** https://arxiv.org/pdf/2604.00993v1
+
+## 383. Bridging Structured Knowledge and Data: A Unified Framework with Finance Applications
+
+- **ID:** `2604.00987` · **primary:** `stat.ML` · **Submitted:** 2026-04-01
+- **Authors:** Yi Cao, Zexun Chen, Lin William Cong, et al.
+- **abs:** https://arxiv.org/abs/2604.00987 · **PDF:** https://arxiv.org/pdf/2604.00987v1
+
+## 384. Do Phone-Use Agents Respect Your Privacy?
+
+- **ID:** `2604.00986` · **primary:** `cs.CR` · **Submitted:** 2026-04-01
+- **Authors:** Zhengyang Tang, Ke Ji, Xidong Wang, et al.
+- **abs:** https://arxiv.org/abs/2604.00986 · **PDF:** https://arxiv.org/pdf/2604.00986v2
+
+## 385. Dual Optimal: Make Your LLM Peer-like with Dignity
+
+- **ID:** `2604.00979` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Xiangqi Wang, Yue Huang, Haomin Zhuang, et al.
+- **abs:** https://arxiv.org/abs/2604.00979 · **PDF:** https://arxiv.org/pdf/2604.00979v2
+
+## 386. Flow-based Policy With Distributional Reinforcement Learning in Trajectory Optimization
+
+- **ID:** `2604.00977` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Ruijie Hao, Longfei Zhang, Yang Dai, et al.
+- **abs:** https://arxiv.org/abs/2604.00977 · **PDF:** https://arxiv.org/pdf/2604.00977v1
+
+## 387. Rapid mixing in positively weighted restricted Boltzmann machines
+
+- **ID:** `2604.00963` · **primary:** `cs.DS` · **Submitted:** 2026-04-01
+- **Authors:** Weiming Feng, Heng Guo, Minji Yang
+- **abs:** https://arxiv.org/abs/2604.00963 · **PDF:** https://arxiv.org/pdf/2604.00963v1
+
+## 388. Phase transition on a context-sensitive random language model with short range interactions
+
+- **ID:** `2604.00947` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Yuma Toji, Jun Takahashi, Vwani Roychowdhury, et al.
+- **abs:** https://arxiv.org/abs/2604.00947 · **PDF:** https://arxiv.org/pdf/2604.00947v1
+
+## 389. Differentially Private Manifold Denoising
+
+- **ID:** `2604.00942` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Jiaqi Wu, Yiqing Sun, Zhigang Yao
+- **abs:** https://arxiv.org/abs/2604.00942 · **PDF:** https://arxiv.org/pdf/2604.00942v1
+
+## 390. WARP: Guaranteed Inner-Layer Repair of NLP Transformers
+
+- **ID:** `2604.00938` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Hsin-Ling Hsu, Min-Yu Chen, Nai-Chia Chen, et al.
+- **abs:** https://arxiv.org/abs/2604.00938 · **PDF:** https://arxiv.org/pdf/2604.00938v1
+
+## 391. PsychAgent: An Experience-Driven Lifelong Learning Agent for Self-Evolving Psychological Counselor
+
+- **ID:** `2604.00931` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Yutao Yang, Junsong Li, Qianjun Pan, et al.
+- **abs:** https://arxiv.org/abs/2604.00931 · **PDF:** https://arxiv.org/pdf/2604.00931v2
+
+## 392. Learning Quantised Structure-Preserving Motion Representations for Dance Fingerprinting
+
+- **ID:** `2604.00927` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Arina Kharlamova, Bowei He, Chen Ma, et al.
+- **abs:** https://arxiv.org/abs/2604.00927 · **PDF:** https://arxiv.org/pdf/2604.00927v1
+
+## 393. Positional Cognitive Specialization: Where Do LLMs Learn To Comprehend and Speak Your Language?
+
+- **ID:** `2604.00923` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Luis Frentzen Salim, Lun-Wei Ku, Hsing-Kuo Kenneth Pao
+- **abs:** https://arxiv.org/abs/2604.00923 · **PDF:** https://arxiv.org/pdf/2604.00923v1
+
+## 394. Representation Selection via Cross-Model Agreement using Canonical Correlation Analysis
+
+- **ID:** `2604.00921` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Dylan B. Lewis, Jens Gregor, Hector Santos-Villalobos
+- **abs:** https://arxiv.org/abs/2604.00921 · **PDF:** https://arxiv.org/pdf/2604.00921v1
+
+## 395. GPT-NL Public Corpus: A Permissively Licensed, Dutch-First Dataset for LLM Pre-training
+
+- **ID:** `2604.00920` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Jesse van Oort, Frank Brinkkemper, Erik de Graaf, et al.
+- **abs:** https://arxiv.org/abs/2604.00920 · **PDF:** https://arxiv.org/pdf/2604.00920v1
+
+## 396. Multi-Mode Quantum Annealing for Variational Autoencoders with General Boltzmann Priors
+
+- **ID:** `2604.00919` · **primary:** `quant-ph` · **Submitted:** 2026-04-01
+- **Authors:** Gilhan Kim, Daniel K. Park
+- **abs:** https://arxiv.org/abs/2604.00919 · **PDF:** https://arxiv.org/pdf/2604.00919v1
+
+## 397. Generalization Bounds for Spectral GNNs via Fourier Domain Analysis
+
+- **ID:** `2604.00918` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Vahan A. Martirosyan, Daniele Malitesta, Hugues Talbot, et al.
+- **abs:** https://arxiv.org/abs/2604.00918 · **PDF:** https://arxiv.org/pdf/2604.00918v1
+
+## 398. Investigating Autonomous Agent Contributions in the Wild: Activity Patterns and Code Change over Time
+
+- **ID:** `2604.00917` · **primary:** `cs.SE` · **Submitted:** 2026-04-01
+- **Authors:** Razvan Mihai Popescu, David Gros, Andrei Botocan, et al.
+- **abs:** https://arxiv.org/abs/2604.00917 · **PDF:** https://arxiv.org/pdf/2604.00917v1
+
+## 399. Orthogonal Learner for Estimating Heterogeneous Long-Term Treatment Effects
+
+- **ID:** `2604.00915` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Haorui Ma, Dennis Frauen, Valentyn Melnychuk, et al.
+- **abs:** https://arxiv.org/abs/2604.00915 · **PDF:** https://arxiv.org/pdf/2604.00915v1
+
+## 400. Benchmarking and Mechanistic Analysis of Vision-Language Models for Cross-Depiction Assembly Instruction Alignment
+
+- **ID:** `2604.00913` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Zhuchenyang Liu, Yao Zhang, Yu Xiao
+- **abs:** https://arxiv.org/abs/2604.00913 · **PDF:** https://arxiv.org/pdf/2604.00913v1
+
+## 401. Event Embedding of Protein Networks : Compositional Learning of Biological Function
+
+- **ID:** `2604.00911` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Antonin Sulc
+- **abs:** https://arxiv.org/abs/2604.00911 · **PDF:** https://arxiv.org/pdf/2604.00911v1
+
+## 402. Fatigue-Aware Learning to Defer via Constrained Optimisation
+
+- **ID:** `2604.00904` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Zheng Zhang, Cuong C. Nguyen, David Rosewarne, et al.
+- **abs:** https://arxiv.org/abs/2604.00904 · **PDF:** https://arxiv.org/pdf/2604.00904v1
+
+## 403. Experience as a Compass: Multi-agent RAG with Evolving Orchestration and Agent Prompts
+
+- **ID:** `2604.00901` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Sha Li, Naren Ramakrishnan
+- **abs:** https://arxiv.org/abs/2604.00901 · **PDF:** https://arxiv.org/pdf/2604.00901v1
+
+## 404. Super-Resolving Coarse-Resolution Weather Forecasts With Flow Matching
+
+- **ID:** `2604.00897` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Aymeric Delefosse, Anastase Charantonis, Dominique Béréziat
+- **abs:** https://arxiv.org/abs/2604.00897 · **PDF:** https://arxiv.org/pdf/2604.00897v1
+
+## 405. When Users Change Their Mind: Evaluating Interruptible Agents in Long-Horizon Web Navigation
+
+- **ID:** `2604.00892` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Henry Peng Zou, Chunyu Miao, Wei-Chieh Huang, et al.
+- **abs:** https://arxiv.org/abs/2604.00892 · **PDF:** https://arxiv.org/pdf/2604.00892v1
+
+## 406. Beyond Symbolic Solving: Multi Chain-of-Thought Voting for Geometric Reasoning in Large Language Models
+
+- **ID:** `2604.00890` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Md. Abu Bakor Siddique, Shahrin Hossain, Sadman Ahmed Siam, et al.
+- **abs:** https://arxiv.org/abs/2604.00890 · **PDF:** https://arxiv.org/pdf/2604.00890v1
+
+## 407. PixelPrune: Pixel-Level Adaptive Visual Token Reduction via Predictive Coding
+
+- **ID:** `2604.00886` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Nan Wang, Zhiwei Jin, Chen Chen, et al.
+- **abs:** https://arxiv.org/abs/2604.00886 · **PDF:** https://arxiv.org/pdf/2604.00886v1
+
+## 408. OkanNet: A Lightweight Deep Learning Architecture for Classification of Brain Tumor from MRI Images
+
+- **ID:** `2604.01264` · **primary:** `eess.IV` · **Submitted:** 2026-04-01
+- **Authors:** Okan Uçar, Murat Kurt
+- **abs:** https://arxiv.org/abs/2604.01264 · **PDF:** https://arxiv.org/pdf/2604.01264v1
+
+## 409. KUET at StanceNakba Shared Task: StanceMoE: Mixture-of-Experts Architecture for Stance Detection
+
+- **ID:** `2604.00878` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Abdullah Al Shafi, Md. Milon Islam, Sk. Imran Hossain, et al.
+- **abs:** https://arxiv.org/abs/2604.00878 · **PDF:** https://arxiv.org/pdf/2604.00878v1
+
+## 410. Accurate and Scalable Matrix Mechanisms via Divide and Conquer
+
+- **ID:** `2604.00868` · **primary:** `cs.DB` · **Submitted:** 2026-04-01
+- **Authors:** Guanlin He, Yingtai Xiao, Jiamu Bai, et al.
+- **abs:** https://arxiv.org/abs/2604.00868 · **PDF:** https://arxiv.org/pdf/2604.00868v1
+
+## 411. Policy Improvement Reinforcement Learning
+
+- **ID:** `2604.00860` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Huaiyang Wang, Xiaojie Li, Deqing Wang, et al.
+- **abs:** https://arxiv.org/abs/2604.00860 · **PDF:** https://arxiv.org/pdf/2604.00860v1
+
+## 412. Debiased Estimators in High-Dimensional Regression: A Review and Replication of Javanmard and Montanari (2014)
+
+- **ID:** `2604.00848` · **primary:** `stat.OT` · **Submitted:** 2026-04-01
+- **Authors:** Benjamin Smith
+- **abs:** https://arxiv.org/abs/2604.00848 · **PDF:** https://arxiv.org/pdf/2604.00848v1
+
+## 413. Proactive Agent Research Environment: Simulating Active Users to Evaluate Proactive Assistants
+
+- **ID:** `2604.00842` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Deepak Nathani, Cheng Zhang, Chang Huan, et al.
+- **abs:** https://arxiv.org/abs/2604.00842 · **PDF:** https://arxiv.org/pdf/2604.00842v1
+
+## 414. Transforming OPACs into Intelligent Discovery Systems: An AI-Powered, Knowledge Graph-Driven Smart OPAC for Digital Libraries
+
+- **ID:** `2604.01262` · **primary:** `cs.DL` · **Submitted:** 2026-04-01
+- **Authors:** M. S. Rajeevan, B. Mini Devi
+- **abs:** https://arxiv.org/abs/2604.01262 · **PDF:** https://arxiv.org/pdf/2604.01262v1
+
+## 415. Agentic Tool Use in Large Language Models
+
+- **ID:** `2604.00835` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Jinchao Hu, Meizhi Zhong, Kehai Chen, et al.
+- **abs:** https://arxiv.org/abs/2604.00835 · **PDF:** https://arxiv.org/pdf/2604.00835v1
+
+## 416. Learning to Learn-at-Test-Time: Language Agents with Learnable Adaptation Policies
+
+- **ID:** `2604.00830` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Zhanzhi Lou, Hui Chen, Yibo Li, et al.
+- **abs:** https://arxiv.org/abs/2604.00830 · **PDF:** https://arxiv.org/pdf/2604.00830v2
+
+## 417. DySCo: Dynamic Semantic Compression for Effective Long-term Time Series Forecasting
+
+- **ID:** `2604.01261` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Xiang Ao, Yinyu Tan, Mengru Chen
+- **abs:** https://arxiv.org/abs/2604.01261 · **PDF:** https://arxiv.org/pdf/2604.01261v1
+
+## 418. LinguDistill: Recovering Linguistic Ability in Vision- Language Models via Selective Cross-Modal Distillation
+
+- **ID:** `2604.00829` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Patrick Amadeus Irawan, Erland Hilman Fuadi, Shanu Kumar, et al.
+- **abs:** https://arxiv.org/abs/2604.00829 · **PDF:** https://arxiv.org/pdf/2604.00829v1
+
+## 419. Optimal Brain Decomposition for Accurate LLM Low-Rank Approximation
+
+- **ID:** `2604.00821` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Yuhang Li, Donghyun Lee, Ruokai Yin, et al.
+- **abs:** https://arxiv.org/abs/2604.00821 · **PDF:** https://arxiv.org/pdf/2604.00821v1
+
+## 420. Emotion Entanglement and Bayesian Inference for Multi-Dimensional Emotion Understanding
+
+- **ID:** `2604.00819` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Hemanth Kotaprolu, Kishan Maharaj, Raey Zhao, et al.
+- **abs:** https://arxiv.org/abs/2604.00819 · **PDF:** https://arxiv.org/pdf/2604.00819v1
+
+## 421. DVGT-2: Vision-Geometry-Action Model for Autonomous Driving at Scale
+
+- **ID:** `2604.00813` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Sicheng Zuo, Zixun Xie, Wenzhao Zheng, et al.
+- **abs:** https://arxiv.org/abs/2604.00813 · **PDF:** https://arxiv.org/pdf/2604.00813v1
+
+## 422. Cost-Penalized Fitness in FMA-Orchestrated Mixture of Experts: Experimental Evidence for Molecular Memory in Domain Adaptation
+
+- **ID:** `2604.00812` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Martin Jaraiz
+- **abs:** https://arxiv.org/abs/2604.00812 · **PDF:** https://arxiv.org/pdf/2604.00812v1
+
+## 423. Deconfounding Scores and Representation Learning for Causal Effect Estimation with Weak Overlap
+
+- **ID:** `2604.00811` · **primary:** `stat.ML` · **Submitted:** 2026-04-01
+- **Authors:** Oscar Clivio, Alexander D'Amour, Alexander Franks, et al.
+- **abs:** https://arxiv.org/abs/2604.00811 · **PDF:** https://arxiv.org/pdf/2604.00811v1
+
+## 424. Routing-Free Mixture-of-Experts
+
+- **ID:** `2604.00801` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Yilun Liu, Jinru Han, Sikuan Yan, et al.
+- **abs:** https://arxiv.org/abs/2604.00801 · **PDF:** https://arxiv.org/pdf/2604.00801v1
+
+## 425. MIRANDA: MId-feature RANk-adversarial Domain Adaptation toward climate change-robust ecological forecasting with deep learning
+
+- **ID:** `2604.00800` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Yuchang Jiang, Jan Dirk Wegner, Vivien Sainte Fare Garnot
+- **abs:** https://arxiv.org/abs/2604.00800 · **PDF:** https://arxiv.org/pdf/2604.00800v1
+
+## 426. Multimodal Language Models Cannot Spot Spatial Inconsistencies
+
+- **ID:** `2604.00799` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Om Khangaonkar, Hadi J. Rad, Hamed Pirsiavash
+- **abs:** https://arxiv.org/abs/2604.00799 · **PDF:** https://arxiv.org/pdf/2604.00799v1
+
+## 427. Preference Guided Iterated Pareto Referent Optimisation for Accessible Route Planning
+
+- **ID:** `2604.00795` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Paolo Speziali, Arno De Greef, Mehrdad Asadi, et al.
+- **abs:** https://arxiv.org/abs/2604.00795 · **PDF:** https://arxiv.org/pdf/2604.00795v1
+
+## 428. RefineRL: Advancing Competitive Programming with Self-Refinement Reinforcement Learning
+
+- **ID:** `2604.00790` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Shaopeng Fu, Xingxing Zhang, Li Dong, et al.
+- **abs:** https://arxiv.org/abs/2604.00790 · **PDF:** https://arxiv.org/pdf/2604.00790v1
+
+## 429. Valency Classification of Mapudungun Verbal Roots. Established by the language's own morphotactics
+
+- **ID:** `2604.00789` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Andrés Chandía
+- **abs:** https://arxiv.org/abs/2604.00789 · **PDF:** https://arxiv.org/pdf/2604.00789v2
+
+## 430. UK AISI Alignment Evaluation Case-Study
+
+- **ID:** `2604.00788` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Alexandra Souly, Robert Kirk, Jacob Merizian, et al.
+- **abs:** https://arxiv.org/abs/2604.00788 · **PDF:** https://arxiv.org/pdf/2604.00788v1
+
+## 431. Finding Low Star Discrepancy 3D Kronecker Point Sets Using Algorithm Configuration Techniques
+
+- **ID:** `2604.00786` · **primary:** `cs.NE` · **Submitted:** 2026-04-01
+- **Authors:** Imène Ait Abderrahim, Carola Doerr, Martin Durand
+- **abs:** https://arxiv.org/abs/2604.00786 · **PDF:** https://arxiv.org/pdf/2604.00786v1
+
+## 432. Scalable Pretraining of Large Mixture of Experts Language Models on Aurora Super Computer
+
+- **ID:** `2604.00785` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Dharma Teja Vooturi, Dhiraj Kalamkar, Dipankar Das, et al.
+- **abs:** https://arxiv.org/abs/2604.00785 · **PDF:** https://arxiv.org/pdf/2604.00785v1
+
+## 433. Using predefined vector systems to speed up neural network multimillion class classification
+
+- **ID:** `2604.00779` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Nikita Gabdullin, Ilya Androsov
+- **abs:** https://arxiv.org/abs/2604.00779 · **PDF:** https://arxiv.org/pdf/2604.00779v1
+
+## 434. From Early Encoding to Late Suppression: Interpreting LLMs on Character Counting Tasks
+
+- **ID:** `2604.00778` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Ayan Datta, Mounika Marreddy, Alexander Mehler, et al.
+- **abs:** https://arxiv.org/abs/2604.00778 · **PDF:** https://arxiv.org/pdf/2604.00778v1
+
+## 435. From Baselines to Preferences: A Comparative Study of LoRA/QLoRA and Preference Optimization for Mental Health Text Classification
+
+- **ID:** `2604.00773` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Mihael Arcan
+- **abs:** https://arxiv.org/abs/2604.00773 · **PDF:** https://arxiv.org/pdf/2604.00773v1
+
+## 436. Thinking Wrong in Silence: Backdoor Attacks on Continuous Latent Reasoning
+
+- **ID:** `2604.00770` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Swapnil Parekh
+- **abs:** https://arxiv.org/abs/2604.00770 · **PDF:** https://arxiv.org/pdf/2604.00770v1
+
+## 437. ActivityNarrated: An Open-Ended Narrative Paradigm for Wearable Human Activity Understanding
+
+- **ID:** `2604.00767` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Lala Shakti Swarup Ray, Mengxi Liu, Alcina Pinto, et al.
+- **abs:** https://arxiv.org/abs/2604.00767 · **PDF:** https://arxiv.org/pdf/2604.00767v1
+
+## 438. IWP: Token Pruning as Implicit Weight Pruning in Large Vision Language Models
+
+- **ID:** `2604.00757` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Dong-Jae Lee, Sunghyun Baek, Junmo Kim
+- **abs:** https://arxiv.org/abs/2604.00757 · **PDF:** https://arxiv.org/pdf/2604.00757v1
+
+## 439. Stochastic Attention: Connectome-Inspired Randomized Routing for Expressive Linear-Time Attention
+
+- **ID:** `2604.00754` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Zehao Jin, Yanan Sui
+- **abs:** https://arxiv.org/abs/2604.00754 · **PDF:** https://arxiv.org/pdf/2604.00754v1
+
+## 440. BioCOMPASS: Integrating Biomarkers into Transformer-Based Immunotherapy Response Prediction
+
+- **ID:** `2604.00739` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Sayed Hashim, Frank Soboczenski, Paul Cairns
+- **abs:** https://arxiv.org/abs/2604.00739 · **PDF:** https://arxiv.org/pdf/2604.00739v1
+
+## 441. Spectral Compact Training: Pre-Training Large Language Models via Permanent Truncated SVD and Stiefel QR Retraction
+
+- **ID:** `2604.00733` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Björn Roman Kohlberger
+- **abs:** https://arxiv.org/abs/2604.00733 · **PDF:** https://arxiv.org/pdf/2604.00733v1
+
+## 442. A CEFR-Inspired Classification Framework with Fuzzy C-Means To Automate Assessment of Programming Skills in Scratch
+
+- **ID:** `2604.00730` · **primary:** `cs.CY` · **Submitted:** 2026-04-01
+- **Authors:** Ricardo Hidalgo-Aragón, Jesús M. González-Barahona, Gregorio Robles
+- **abs:** https://arxiv.org/abs/2604.00730 · **PDF:** https://arxiv.org/pdf/2604.00730v1
+
+## 443. Exploring Silent Data Corruption as a Reliability Challenge in LLM Training
+
+- **ID:** `2604.00726` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Anton Altenbernd, Philipp Wiesner, Odej Kao
+- **abs:** https://arxiv.org/abs/2604.00726 · **PDF:** https://arxiv.org/pdf/2604.00726v1
+
+## 444. A Benchmark of State-Space Models vs. Transformers and BiLSTM-based Models for Historical Newspaper OCR
+
+- **ID:** `2604.00725` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Merveilles Agbeti-messan, Thierry Paquet, Clément Chatelain, et al.
+- **abs:** https://arxiv.org/abs/2604.00725 · **PDF:** https://arxiv.org/pdf/2604.00725v1
+
+## 445. LangMARL: Natural Language Multi-Agent Reinforcement Learning
+
+- **ID:** `2604.00722` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Huaiyuan Yao, Longchao Da, Xiaoou Liu, et al.
+- **abs:** https://arxiv.org/abs/2604.00722 · **PDF:** https://arxiv.org/pdf/2604.00722v1
+
+## 446. GRASP: Gradient Realignment via Active Shared Perception for Multi-Agent Collaborative Optimization
+
+- **ID:** `2604.00717` · **primary:** `cs.MA` · **Submitted:** 2026-04-01
+- **Authors:** Sihan Zhou, Tiantian He, Yifan Lu, et al.
+- **abs:** https://arxiv.org/abs/2604.00717 · **PDF:** https://arxiv.org/pdf/2604.00717v1
+
+## 447. CircuitProbe: Predicting Reasoning Circuits in Transformers via Stability Zone Detection
+
+- **ID:** `2604.00716` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Rajkiran Panuganti
+- **abs:** https://arxiv.org/abs/2604.00716 · **PDF:** https://arxiv.org/pdf/2604.00716v1
+
+## 448. To Memorize or to Retrieve: Scaling Laws for RAG-Considerate Pretraining
+
+- **ID:** `2604.00715` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Karan Singh, Michael Yu, Varun Gangal, et al.
+- **abs:** https://arxiv.org/abs/2604.00715 · **PDF:** https://arxiv.org/pdf/2604.00715v1
+
+## 449. AfrIFact: Cultural Information Retrieval, Evidence Extraction and Fact Checking for African Languages
+
+- **ID:** `2604.00706` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Israel Abebe Azime, Jesujoba Oluwadara Alabi, Crystina Zhang, et al.
+- **abs:** https://arxiv.org/abs/2604.00706 · **PDF:** https://arxiv.org/pdf/2604.00706v1
+
+## 450. AutoEG: Exploiting Known Third-Party Vulnerabilities in Black-Box Web Applications
+
+- **ID:** `2604.00704` · **primary:** `cs.CR` · **Submitted:** 2026-04-01
+- **Authors:** Ruozhao Yang, Mingfei Cheng, Gelei Deng, et al.
+- **abs:** https://arxiv.org/abs/2604.00704 · **PDF:** https://arxiv.org/pdf/2604.00704v1
+
+## 451. G-ICSO-NAS: Shifting Gears between Gradient and Swarm for Robust Neural Architecture Search
+
+- **ID:** `2604.00703` · **primary:** `cs.NE` · **Submitted:** 2026-04-01
+- **Authors:** Xingbang Du, Enzhi Zhang, Rui Zhong, et al.
+- **abs:** https://arxiv.org/abs/2604.00703 · **PDF:** https://arxiv.org/pdf/2604.00703v1
+
+## 452. Learning to Hint for Reinforcement Learning
+
+- **ID:** `2604.00698` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Yu Xia, Canwen Xu, Zhewei Yao, et al.
+- **abs:** https://arxiv.org/abs/2604.00698 · **PDF:** https://arxiv.org/pdf/2604.00698v1
+
+## 453. Inverse-Free Sparse Variational Gaussian Processes
+
+- **ID:** `2604.00697` · **primary:** `stat.ML` · **Submitted:** 2026-04-01
+- **Authors:** Stefano Cortinovis, Laurence Aitchison, Stefanos Eleftheriadis, et al.
+- **abs:** https://arxiv.org/abs/2604.00697 · **PDF:** https://arxiv.org/pdf/2604.00697v1
+
+## 454. Internal APIs Are All You Need: Shadow APIs, Shared Discovery, and the Case Against Browser-First Agent Architectures
+
+- **ID:** `2604.00694` · **primary:** `cs.ET` · **Submitted:** 2026-04-01
+- **Authors:** Lewis Tham, Nicholas Mac Gregor Garcia, Jungpil Hahn
+- **abs:** https://arxiv.org/abs/2604.00694 · **PDF:** https://arxiv.org/pdf/2604.00694v1
+
+## 455. Performance of Neural and Polynomial Operator Surrogates
+
+- **ID:** `2604.00689` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Josephine Westermann, Benno Huber, Thomas O'Leary-Roseberry, et al.
+- **abs:** https://arxiv.org/abs/2604.00689 · **PDF:** https://arxiv.org/pdf/2604.00689v1
+
+## 456. OmniVoice: Towards Omnilingual Zero-Shot Text-to-Speech with Diffusion Language Models
+
+- **ID:** `2604.00688` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Han Zhu, Lingxuan Ye, Wei Kang, et al.
+- **abs:** https://arxiv.org/abs/2604.00688 · **PDF:** https://arxiv.org/pdf/2604.00688v2
+
+## 457. Full-Gradient Successor Feature Representations
+
+- **ID:** `2604.00686` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Ritish Shrirao, Aditya Priyadarshi, Raghuram Bharadwaj Diddigi
+- **abs:** https://arxiv.org/abs/2604.00686 · **PDF:** https://arxiv.org/pdf/2604.00686v1
+
+## 458. Convergence of projected stochastic natural gradient variational inference for various step size and sample or batch size schedules
+
+- **ID:** `2604.00683` · **primary:** `stat.ME` · **Submitted:** 2026-04-01
+- **Authors:** Thomas Guilmeau, Hadrien Hendrikx, Florence Forbes
+- **abs:** https://arxiv.org/abs/2604.00683 · **PDF:** https://arxiv.org/pdf/2604.00683v1
+
+## 459. Procela: Epistemic Governance in Mechanistic Simulations Under Structural Uncertainty
+
+- **ID:** `2604.00675` · **primary:** `physics.comp-ph` · **Submitted:** 2026-04-01
+- **Authors:** Kinson Vernet
+- **abs:** https://arxiv.org/abs/2604.00675 · **PDF:** https://arxiv.org/pdf/2604.00675v1
+
+## 460. Common TF-IDF variants arise as key components in the test statistic of a penalized likelihood-ratio test for word burstiness
+
+- **ID:** `2604.00672` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Zeyad Ahmed, Paul Sheridan, Michael McIsaac, et al.
+- **abs:** https://arxiv.org/abs/2604.00672 · **PDF:** https://arxiv.org/pdf/2604.00672v1
+
+## 461. Embedded Variational Neural Stochastic Differential Equations for Learning Heterogeneous Dynamics
+
+- **ID:** `2604.00669` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Sandeep Kumar Samota, Reema Gupta, Snehashish Chakraverty
+- **abs:** https://arxiv.org/abs/2604.00669 · **PDF:** https://arxiv.org/pdf/2604.00669v1
+
+## 462. TRIMS: Trajectory-Ranked Instruction Masked Supervision for Diffusion Language Models
+
+- **ID:** `2604.00666` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Lingjie Chen, Ruizhong Qiu, Yuyu Fan, et al.
+- **abs:** https://arxiv.org/abs/2604.00666 · **PDF:** https://arxiv.org/pdf/2604.00666v1
+
+## 463. Streaming Model Cascades for Semantic SQL
+
+- **ID:** `2604.00660` · **primary:** `cs.DB` · **Submitted:** 2026-04-01
+- **Authors:** Paweł Liskowski, Kyle Schmaus
+- **abs:** https://arxiv.org/abs/2604.00660 · **PDF:** https://arxiv.org/pdf/2604.00660v1
+
+## 464. Chameleons do not Forget: Prompt-Based Online Continual Learning for Next Activity Prediction
+
+- **ID:** `2604.00653` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Marwan Hassani, Tamara Verbeek, Sjoerd van Straten
+- **abs:** https://arxiv.org/abs/2604.00653 · **PDF:** https://arxiv.org/pdf/2604.00653v2
+
+## 465. On rankings in multiplayer games with an application to the game of Whist
+
+- **ID:** `2604.00641` · **primary:** `cs.GT` · **Submitted:** 2026-04-01
+- **Authors:** Alexis Coyette, Charles Modera, Candy Sonveaux, et al.
+- **abs:** https://arxiv.org/abs/2604.00641 · **PDF:** https://arxiv.org/pdf/2604.00641v1
+
+## 466. Neural Ordinary Differential Equations for Modeling Socio-Economic Dynamics
+
+- **ID:** `2604.00632` · **primary:** `math.DS` · **Submitted:** 2026-04-01
+- **Authors:** Sandeep Kumar Samota, Snehashish Chakraverty, Narayan Sethi
+- **abs:** https://arxiv.org/abs/2604.00632 · **PDF:** https://arxiv.org/pdf/2604.00632v1
+
+## 467. A Survey of On-Policy Distillation for Large Language Models
+
+- **ID:** `2604.00626` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Mingyang Song, Mao Zheng
+- **abs:** https://arxiv.org/abs/2604.00626 · **PDF:** https://arxiv.org/pdf/2604.00626v1
+
+## 468. English to Central Kurdish Speech Translation: Corpus Creation, Evaluation, and Orthographic Standardization
+
+- **ID:** `2604.00613` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Mohammad Mohammadamini, Daban Q. Jaff, Josep Crego, et al.
+- **abs:** https://arxiv.org/abs/2604.00613 · **PDF:** https://arxiv.org/pdf/2604.00613v1
+
+## 469. Speech LLMs are Contextual Reasoning Transcribers
+
+- **ID:** `2604.00610` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Keqi Deng, Ruchao Fan, Bo Ren, et al.
+- **abs:** https://arxiv.org/abs/2604.00610 · **PDF:** https://arxiv.org/pdf/2604.00610v1
+
+## 470. Predicting Dynamics of Ultra-Large Complex Systems by Inferring Governing Equations
+
+- **ID:** `2604.00599` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Qi Shao, Duxin Chen, Jiawen Chen, et al.
+- **abs:** https://arxiv.org/abs/2604.00599 · **PDF:** https://arxiv.org/pdf/2604.00599v1
+
+## 471. Agent psychometrics: Task-level performance prediction in agentic coding benchmarks
+
+- **ID:** `2604.00594` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Chris Ge, Daria Kryvosheieva, Daniel Fried, et al.
+- **abs:** https://arxiv.org/abs/2604.00594 · **PDF:** https://arxiv.org/pdf/2604.00594v1
+
+## 472. UniMixer: A Unified Architecture for Scaling Laws in Recommendation Systems
+
+- **ID:** `2604.00590` · **primary:** `cs.IR` · **Submitted:** 2026-04-01
+- **Authors:** Mingming Ha, Guanchen Wang, Linxun Chen, et al.
+- **abs:** https://arxiv.org/abs/2604.00590 · **PDF:** https://arxiv.org/pdf/2604.00590v2
+
+## 473. More Human, More Efficient: Aligning Annotations with Quantized SLMs
+
+- **ID:** `2604.00586` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Jiayu Wang, Junyoung Lee
+- **abs:** https://arxiv.org/abs/2604.00586 · **PDF:** https://arxiv.org/pdf/2604.00586v1
+
+## 474. Representation choice shapes the interpretation of protein conformational dynamics
+
+- **ID:** `2604.00580` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Axel Giottonini, Thomas Lemmin
+- **abs:** https://arxiv.org/abs/2604.00580 · **PDF:** https://arxiv.org/pdf/2604.00580v1
+
+## 475. A Japanese Benchmark for Evaluating Social Bias in Reasoning Based on Attribution Theory
+
+- **ID:** `2604.00568` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Taihei Shiotani, Masahiro Kaneko, Naoaki Okazaki
+- **abs:** https://arxiv.org/abs/2604.00568 · **PDF:** https://arxiv.org/pdf/2604.00568v1
+
+## 476. Multi-Camera View Scaling for Data-Efficient Robot Imitation Learning
+
+- **ID:** `2604.00557` · **primary:** `cs.RO` · **Submitted:** 2026-04-01
+- **Authors:** Yichen Xie, Yixiao Wang, Shuqi Zhao, et al.
+- **abs:** https://arxiv.org/abs/2604.00557 · **PDF:** https://arxiv.org/pdf/2604.00557v1
+
+## 477. HabitatAgent: An End-to-End Multi-Agent System for Housing Consultation
+
+- **ID:** `2604.00556` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Hongyang Yang, Yanxin Zhang, Yang She, et al.
+- **abs:** https://arxiv.org/abs/2604.00556 · **PDF:** https://arxiv.org/pdf/2604.00556v1
+
+## 478. Ontology-Constrained Neural Reasoning in Enterprise Agentic Systems: A Neurosymbolic Architecture for Domain-Grounded AI Agents
+
+- **ID:** `2604.00555` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Thanh Luong Tuan
+- **abs:** https://arxiv.org/abs/2604.00555 · **PDF:** https://arxiv.org/pdf/2604.00555v1
+
+## 479. Scenario theory for multi-criteria data-driven decision making
+
+- **ID:** `2604.00553` · **primary:** `stat.ML` · **Submitted:** 2026-04-01
+- **Authors:** Simone Garatti, Lucrezia Manieri, Alessandro Falsone, et al.
+- **abs:** https://arxiv.org/abs/2604.00553 · **PDF:** https://arxiv.org/pdf/2604.00553v1
+
+## 480. BloClaw: An Omniscient, Multi-Modal Agentic Workspace for Next-Generation Scientific Discovery
+
+- **ID:** `2604.00550` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Yao Qin, Yangyang Yan, Jinhua Pang, et al.
+- **abs:** https://arxiv.org/abs/2604.00550 · **PDF:** https://arxiv.org/pdf/2604.00550v1
+
+## 481. Does Unification Come at a Cost? Uni-SafeBench: A Safety Benchmark for Unified Multimodal Large Models
+
+- **ID:** `2604.00547` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Zixiang Peng, Yongxiu Xu, Qinyi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2604.00547 · **PDF:** https://arxiv.org/pdf/2604.00547v1
+
+## 482. Activation Saturation and Floquet Spectrum Collapse in Neural ODEs
+
+- **ID:** `2604.00543` · **primary:** `math.DS` · **Submitted:** 2026-04-01
+- **Authors:** Nikolaos M. Matzakos
+- **abs:** https://arxiv.org/abs/2604.00543 · **PDF:** https://arxiv.org/pdf/2604.00543v1
+
+## 483. MATHENA: Mamba-based Architectural Tooth Hierarchical Estimator and Holistic Evaluation Network for Anatomy
+
+- **ID:** `2604.00537` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Kyeonghun Kim, Jaehyung Park, Youngung Han, et al.
+- **abs:** https://arxiv.org/abs/2604.00537 · **PDF:** https://arxiv.org/pdf/2604.00537v1
+
+## 484. Optimsyn: Influence-Guided Rubrics Optimization for Synthetic Data Generation
+
+- **ID:** `2604.00536` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Zhiting Fan, Ruizhe Chen, Tianxiang Hu, et al.
+- **abs:** https://arxiv.org/abs/2604.00536 · **PDF:** https://arxiv.org/pdf/2604.00536v1
+
+## 485. Learning from Many and Adapting to the Unknown in Open-set Test Streams
+
+- **ID:** `2604.00533` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Xiao Zhang, Juntao Lyu, Tianyu Hu, et al.
+- **abs:** https://arxiv.org/abs/2604.00533 · **PDF:** https://arxiv.org/pdf/2604.00533v1
+
+## 486. Learning Shared Representations for Multi-Task Linear Bandits
+
+- **ID:** `2604.00531` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Jiabin Lin, Shana Moothedath
+- **abs:** https://arxiv.org/abs/2604.00531 · **PDF:** https://arxiv.org/pdf/2604.00531v1
+
+## 487. MF-QAT: Multi-Format Quantization-Aware Training for Elastic Inference
+
+- **ID:** `2604.00529` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Zifei Xu, Sayeh Sharify, Hesham Mostafa
+- **abs:** https://arxiv.org/abs/2604.00529 · **PDF:** https://arxiv.org/pdf/2604.00529v1
+
+## 488. Think, Act, Build: An Agentic Framework with Vision Language Models for Zero-Shot 3D Visual Grounding
+
+- **ID:** `2604.00528` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Haibo Wang, Zihao Lin, Zhiyang Xu, et al.
+- **abs:** https://arxiv.org/abs/2604.00528 · **PDF:** https://arxiv.org/pdf/2604.00528v2
+
+## 489. Lipschitz Dueling Bandits over Continuous Action Spaces
+
+- **ID:** `2604.00523` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Mudit Sharma, Shweta Jain, Vaneet Aggarwal, et al.
+- **abs:** https://arxiv.org/abs/2604.00523 · **PDF:** https://arxiv.org/pdf/2604.00523v1
+
+## 490. Toward Optimal Sampling Rate Selection and Unbiased Classification for Precise Animal Activity Recognition
+
+- **ID:** `2604.00517` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Axiu Mao, Meilu Zhu, Lei Shen, et al.
+- **abs:** https://arxiv.org/abs/2604.00517 · **PDF:** https://arxiv.org/pdf/2604.00517v1
+
+## 491. MAESIL: Masked Autoencoder for Enhanced Self-supervised Medical Image Learning
+
+- **ID:** `2604.00514` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Kyeonghun Kim, Hyeonseok Jung, Youngung Han, et al.
+- **abs:** https://arxiv.org/abs/2604.00514 · **PDF:** https://arxiv.org/pdf/2604.00514v1
+
+## 492. MOON3.0: Reasoning-aware Multimodal Representation Learning for E-commerce Product Understanding
+
+- **ID:** `2604.00513` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Junxian Wu, Chenghan Fu, Zhanheng Nie, et al.
+- **abs:** https://arxiv.org/abs/2604.00513 · **PDF:** https://arxiv.org/pdf/2604.00513v2
+
+## 493. Adaptive Parallel Monte Carlo Tree Search for Efficient Test-time Compute Scaling
+
+- **ID:** `2604.00510` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Hongbeen Kim, Juhyun Lee, Sanghyeon Lee, et al.
+- **abs:** https://arxiv.org/abs/2604.00510 · **PDF:** https://arxiv.org/pdf/2604.00510v1
+
+## 494. A Decoupled Basis-Vector-Driven Generative Framework for Dynamic Multi-Objective Optimization
+
+- **ID:** `2604.00508` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Yaoming Yang, Shuai Wang, Bingdong Li, et al.
+- **abs:** https://arxiv.org/abs/2604.00508 · **PDF:** https://arxiv.org/pdf/2604.00508v1
+
+## 495. Towards Initialization-dependent and Non-vacuous Generalization Bounds for Overparameterized Shallow Neural Networks
+
+- **ID:** `2604.00505` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Yunwen Lei, Yufeng Xie
+- **abs:** https://arxiv.org/abs/2604.00505 · **PDF:** https://arxiv.org/pdf/2604.00505v1
+
+## 496. Generalized Heavy-tailed Mutation for Evolutionary Algorithms
+
+- **ID:** `2604.00502` · **primary:** `cs.NE` · **Submitted:** 2026-04-01
+- **Authors:** Anton V. Eremeev, Dmitri V. Silaev, Valentin A. Topchii
+- **abs:** https://arxiv.org/abs/2604.00502 · **PDF:** https://arxiv.org/pdf/2604.00502v1
+
+## 497. Scheduling LLM Inference with Uncertainty-Aware Output Length Predictions
+
+- **ID:** `2604.00499` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Haoyu Zheng, Yongqiang Zhang, Fangcheng Fu, et al.
+- **abs:** https://arxiv.org/abs/2604.00499 · **PDF:** https://arxiv.org/pdf/2604.00499v1
+
+## 498. A Reasoning-Enabled Vision-Language Foundation Model for Chest X-ray Interpretation
+
+- **ID:** `2604.00493` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Yabin Zhang, Chong Wang, Yunhe Gao, et al.
+- **abs:** https://arxiv.org/abs/2604.00493 · **PDF:** https://arxiv.org/pdf/2604.00493v1
+
+## 499. Executing as You Generate: Hiding Execution Latency in LLM Code Generation
+
+- **ID:** `2604.00491` · **primary:** `cs.PL` · **Submitted:** 2026-04-01
+- **Authors:** Zhensu Sun, Zhihao Lin, Zhi Chen, et al.
+- **abs:** https://arxiv.org/abs/2604.00491 · **PDF:** https://arxiv.org/pdf/2604.00491v1
+
+## 500. Adapting Text LLMs to Speech via Multimodal Depth Up-Scaling
+
+- **ID:** `2604.00489` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Kazuki Yano, Jun Suzuki, Shinji Watanabe
+- **abs:** https://arxiv.org/abs/2604.00489 · **PDF:** https://arxiv.org/pdf/2604.00489v1
+
+## 501. The Rashomon Effect for Visualizing High-Dimensional Data
+
+- **ID:** `2604.00485` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Yiyang Sun, Haiyang Huang, Gaurav Rajesh Parikh, et al.
+- **abs:** https://arxiv.org/abs/2604.00485 · **PDF:** https://arxiv.org/pdf/2604.00485v1
+
+## 502. Tucker Diffusion Model for High-dimensional Tensor Generation
+
+- **ID:** `2604.00481` · **primary:** `stat.ME` · **Submitted:** 2026-04-01
+- **Authors:** Jianhua Guo, Xinbing Kong, Zeyu Li, et al.
+- **abs:** https://arxiv.org/abs/2604.00481 · **PDF:** https://arxiv.org/pdf/2604.00481v1
+
+## 503. The Silicon Mirror: Dynamic Behavioral Gating for Anti-Sycophancy in LLM Agents
+
+- **ID:** `2604.00478` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Harshee Jignesh Shah
+- **abs:** https://arxiv.org/abs/2604.00478 · **PDF:** https://arxiv.org/pdf/2604.00478v2
+
+## 504. Logarithmic Scores, Power-Law Discoveries: Disentangling Measurement from Coverage in Agent-Based Evaluation
+
+- **ID:** `2604.00477` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** HyunJoon Jung, William Na
+- **abs:** https://arxiv.org/abs/2604.00477 · **PDF:** https://arxiv.org/pdf/2604.00477v1
+
+## 505. Phase space integrity in neural network models of Hamiltonian dynamics: A Lagrangian descriptor approach
+
+- **ID:** `2604.00473` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Abrari Noor Hasmi, Haralampos Hatzikirou, Hadi Susanto
+- **abs:** https://arxiv.org/abs/2604.00473 · **PDF:** https://arxiv.org/pdf/2604.00473v1
+
+## 506. Automated Detection of Multiple Sclerosis Lesions on 7-tesla MRI Using U-net and Transformer-based Segmentation
+
+- **ID:** `2604.00469` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Michael Maynord, Minghui Liu, Cornelia Fermüller, et al.
+- **abs:** https://arxiv.org/abs/2604.00469 · **PDF:** https://arxiv.org/pdf/2604.00469v1
+
+## 507. Not My Truce: Personality Differences in AI-Mediated Workplace Negotiation
+
+- **ID:** `2604.00464` · **primary:** `cs.HC` · **Submitted:** 2026-04-01
+- **Authors:** Veda Duddu, Jash Rajesh Parekh, Andy Mao, et al.
+- **abs:** https://arxiv.org/abs/2604.00464 · **PDF:** https://arxiv.org/pdf/2604.00464v1
+
+## 508. First Logit Boosting: Visual Grounding Method to Mitigate Object Hallucination in Large Vision-Language Models
+
+- **ID:** `2604.00455` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Jiwoo Ha, Jongwoo Baek, Jinhyun So
+- **abs:** https://arxiv.org/abs/2604.00455 · **PDF:** https://arxiv.org/pdf/2604.00455v1
+
+## 509. Convergence of Byzantine-Resilient Gradient Tracking via Probabilistic Edge Dropout
+
+- **ID:** `2604.00449` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Amirhossein Dezhboro, Fateme Maleki, Arman Adibi, et al.
+- **abs:** https://arxiv.org/abs/2604.00449 · **PDF:** https://arxiv.org/pdf/2604.00449v1
+
+## 510. Towards Reliable Truth-Aligned Uncertainty Estimation in Large Language Models
+
+- **ID:** `2604.00445` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Ponhvoan Srey, Quang Minh Nguyen, Xiaobao Wu, et al.
+- **abs:** https://arxiv.org/abs/2604.00445 · **PDF:** https://arxiv.org/pdf/2604.00445v1
+
+## 511. Polysemanticity or Polysemy? Lexical Identity Confounds Superposition Metrics
+
+- **ID:** `2604.00443` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Iyad Ait Hou, Rebecca Hwa
+- **abs:** https://arxiv.org/abs/2604.00443 · **PDF:** https://arxiv.org/pdf/2604.00443v1
+
+## 512. Execution-Verified Reinforcement Learning for Optimization Modeling
+
+- **ID:** `2604.00442` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Runda Guan, Xiangqing Shen, Jiajun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2604.00442 · **PDF:** https://arxiv.org/pdf/2604.00442v1
+
+## 513. TR-ICRL: Test-Time Rethinking for In-Context Reinforcement Learning
+
+- **ID:** `2604.00438` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Wenxuan Jiang, Yuxin Zuo, Zijian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2604.00438 · **PDF:** https://arxiv.org/pdf/2604.00438v1
+
+## 514. Internal State-Based Policy Gradient Methods for Partially Observable Markov Potential Games
+
+- **ID:** `2604.00433` · **primary:** `cs.MA` · **Submitted:** 2026-04-01
+- **Authors:** Wonseok Yang, Thinh T. Doan
+- **abs:** https://arxiv.org/abs/2604.00433 · **PDF:** https://arxiv.org/pdf/2604.00433v1
+
+## 515. Denoising distances beyond the volumetric barrier
+
+- **ID:** `2604.00432` · **primary:** `stat.ML` · **Submitted:** 2026-04-01
+- **Authors:** Han Huang, Pakawut Jiradilok, Elchanan Mossel
+- **abs:** https://arxiv.org/abs/2604.00432 · **PDF:** https://arxiv.org/pdf/2604.00432v1
+
+## 516. Shapley-Guided Neural Repair Approach via Derivative-Free Optimization
+
+- **ID:** `2604.00422` · **primary:** `cs.SE` · **Submitted:** 2026-04-01
+- **Authors:** Xinyu Sun, Wanwei Liu, Haoang Chi, et al.
+- **abs:** https://arxiv.org/abs/2604.00422 · **PDF:** https://arxiv.org/pdf/2604.00422v1
+
+## 517. Self-Routing: Parameter-Free Expert Routing from Hidden States
+
+- **ID:** `2604.00421` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Jama Hussein Mohamud, Drew Wagner, Mirco Ravanelli
+- **abs:** https://arxiv.org/abs/2604.00421 · **PDF:** https://arxiv.org/pdf/2604.00421v1
+
+## 518. G-Drift MIA: Membership Inference via Gradient-Induced Feature Drift in LLMs
+
+- **ID:** `2604.00419` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Ravi Ranjan, Utkarsh Grover, Xiaomin Lin, et al.
+- **abs:** https://arxiv.org/abs/2604.00419 · **PDF:** https://arxiv.org/pdf/2604.00419v1
+
+## 519. Learning Humanoid Navigation from Human Data
+
+- **ID:** `2604.00416` · **primary:** `cs.RO` · **Submitted:** 2026-04-01
+- **Authors:** Weizhuo Wang, Yanjie Ze, C. Karen Liu, et al.
+- **abs:** https://arxiv.org/abs/2604.00416 · **PDF:** https://arxiv.org/pdf/2604.00416v1
+
+## 520. Decision-Centric Design for LLM Systems
+
+- **ID:** `2604.00414` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Wei Sun
+- **abs:** https://arxiv.org/abs/2604.00414 · **PDF:** https://arxiv.org/pdf/2604.00414v1
+
+## 521. COTTA: Context-Aware Transfer Adaptation for Trajectory Prediction in Autonomous Driving
+
+- **ID:** `2604.00402` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Seohyoung Park, Jaeyeol Lim, Seoyoung Ju, et al.
+- **abs:** https://arxiv.org/abs/2604.00402 · **PDF:** https://arxiv.org/pdf/2604.00402v1
+
+## 522. A Cross-graph Tuning-free GNN Prompting Framework
+
+- **ID:** `2604.00399` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Yaqi Chen, Shixun Huang, Ryan Twemlow, et al.
+- **abs:** https://arxiv.org/abs/2604.00399 · **PDF:** https://arxiv.org/pdf/2604.00399v1
+
+## 523. Improving Generalization of Deep Learning for Brain Metastases Segmentation Across Institutions
+
+- **ID:** `2604.00397` · **primary:** `cs.CV` · **Submitted:** 2026-04-01
+- **Authors:** Yuchen Yang, Shuangyang Zhong, Haijun Yu, et al.
+- **abs:** https://arxiv.org/abs/2604.00397 · **PDF:** https://arxiv.org/pdf/2604.00397v1
+
+## 524. Deep Networks Favor Simple Data
+
+- **ID:** `2604.00394` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Weyl Lu, Chenjie Hao, Yubei Chen
+- **abs:** https://arxiv.org/abs/2604.00394 · **PDF:** https://arxiv.org/pdf/2604.00394v2
+
+## 525. EvolveTool-Bench: Evaluating the Quality of LLM-Generated Tool Libraries as Software Artifacts
+
+- **ID:** `2604.00392` · **primary:** `cs.SE` · **Submitted:** 2026-04-01
+- **Authors:** Alibek T. Kaliyev, Artem Maryanskyy
+- **abs:** https://arxiv.org/abs/2604.00392 · **PDF:** https://arxiv.org/pdf/2604.00392v1
+
+## 526. Gradient-Based Data Valuation Improves Curriculum Learning for Game-Theoretic Motion Planning
+
+- **ID:** `2604.00388` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Shihao Li, Jiachen Li, Dongmei Chen
+- **abs:** https://arxiv.org/abs/2604.00388 · **PDF:** https://arxiv.org/pdf/2604.00388v1
+
+## 527. RAGShield: Provenance-Verified Defense-in-Depth Against Knowledge Base Poisoning in Government Retrieval-Augmented Generation Systems
+
+- **ID:** `2604.00387` · **primary:** `cs.CR` · **Submitted:** 2026-04-01
+- **Authors:** KrishnaSaiReddy Patil
+- **abs:** https://arxiv.org/abs/2604.00387 · **PDF:** https://arxiv.org/pdf/2604.00387v1
+
+## 528. GUIDE: Reinforcement Learning for Behavioral Action Support in Type 1 Diabetes
+
+- **ID:** `2604.00385` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Saman Khamesian, Sri Harini Balaji, Di Yang Shi, et al.
+- **abs:** https://arxiv.org/abs/2604.00385 · **PDF:** https://arxiv.org/pdf/2604.00385v1
+
+## 529. Locally Confident, Globally Stuck: The Quality-Exploration Dilemma in Diffusion Language Models
+
+- **ID:** `2604.00375` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Liancheng Fang, Aiwei Liu, Henry Peng Zou, et al.
+- **abs:** https://arxiv.org/abs/2604.00375 · **PDF:** https://arxiv.org/pdf/2604.00375v1
+
+## 530. In harmony with gpt-oss
+
+- **ID:** `2604.00362` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Borislav Mavrin
+- **abs:** https://arxiv.org/abs/2604.00362 · **PDF:** https://arxiv.org/pdf/2604.00362v1
+
+## 531. Signals: Trajectory Sampling and Triage for Agentic Interactions
+
+- **ID:** `2604.00356` · **primary:** `cs.AI` · **Submitted:** 2026-04-01
+- **Authors:** Shuguang Chen, Adil Hafeez, Salman Paracha
+- **abs:** https://arxiv.org/abs/2604.00356 · **PDF:** https://arxiv.org/pdf/2604.00356v1
+
+## 532. Deep Learning-Accelerated Surrogate Optimization for High-Dimensional Well Control in Stress-Sensitive Reservoirs
+
+- **ID:** `2604.00352` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Mahammad Valiyev, Jodel Cornelio, Behnam Jafarpour
+- **abs:** https://arxiv.org/abs/2604.00352 · **PDF:** https://arxiv.org/pdf/2604.00352v1
+
+## 533. Go Big or Go Home: Simulating Mobbing Behavior with Braitenbergian Robots
+
+- **ID:** `2604.00350` · **primary:** `cs.RO` · **Submitted:** 2026-04-01
+- **Authors:** Elaheh Sanoubari
+- **abs:** https://arxiv.org/abs/2604.00350 · **PDF:** https://arxiv.org/pdf/2604.00350v1
+
+## 534. Agent Q-Mix: Selecting the Right Action for LLM Multi-Agent Systems through Reinforcement Learning
+
+- **ID:** `2604.00344` · **primary:** `cs.CL` · **Submitted:** 2026-04-01
+- **Authors:** Eric Hanchen Jiang, Levina Li, Rui Sun, et al.
+- **abs:** https://arxiv.org/abs/2604.00344 · **PDF:** https://arxiv.org/pdf/2604.00344v1
+
+## 535. Is One Token All It Takes? Graph Pooling Tokens for LLM-based GraphQA
+
+- **ID:** `2604.00342` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Ankit Grover, Lodovico Giaretta, Rémi Bourgerie, et al.
+- **abs:** https://arxiv.org/abs/2604.00342 · **PDF:** https://arxiv.org/pdf/2604.00342v1
+
+## 536. When Career Data Runs Out: Structured Feature Engineering and Signal Limits for Founder Success Prediction
+
+- **ID:** `2604.00339` · **primary:** `cs.LG` · **Submitted:** 2026-04-01
+- **Authors:** Yagiz Ihlamur
+- **abs:** https://arxiv.org/abs/2604.00339 · **PDF:** https://arxiv.org/pdf/2604.00339v1
+
+## 537. MVNN: A Measure-Valued Neural Network for Learning McKean-Vlasov Dynamics from Particle Data
+
+- **ID:** `2604.00333` · **primary:** `math.NA` · **Submitted:** 2026-04-01
+- **Authors:** Liyao Lyu, Xinyue Yu, Hayden Schaeffer
+- **abs:** https://arxiv.org/abs/2604.00333 · **PDF:** https://arxiv.org/pdf/2604.00333v1
+
+## 538. The Persistent Vulnerability of Aligned AI Systems
+
+- **ID:** `2604.00324` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Aengus Lynch
+- **abs:** https://arxiv.org/abs/2604.00324 · **PDF:** https://arxiv.org/pdf/2604.00324v1
+
+## 539. Large Language Models in the Abuse Detection Pipeline
+
+- **ID:** `2604.00323` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Suraj Kath, Sanket Badhe, Preet Shah, et al.
+- **abs:** https://arxiv.org/abs/2604.00323 · **PDF:** https://arxiv.org/pdf/2604.00323v1
+
+## 540. Collaborative AI Agents and Critics for Fault Detection and Cause Analysis in Network Telemetry
+
+- **ID:** `2604.00319` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Syed Eqbal Alam, Zhan Shu
+- **abs:** https://arxiv.org/abs/2604.00319 · **PDF:** https://arxiv.org/pdf/2604.00319v1
+
+## 541. Breaking Data Symmetry is Needed For Generalization in Feature Learning Kernels
+
+- **ID:** `2604.00316` · **primary:** `stat.ML` · **Submitted:** 2026-03-31
+- **Authors:** Marcel Tomàs Bernal, Neil Rohit Mallinar, Mikhail Belkin
+- **abs:** https://arxiv.org/abs/2604.00316 · **PDF:** https://arxiv.org/pdf/2604.00316v1
+
+## 542. Prompt-Guided Prefiltering for VLM Image Compression
+
+- **ID:** `2604.00314` · **primary:** `eess.IV` · **Submitted:** 2026-03-31
+- **Authors:** Bardia Azizian, Ivan V. Bajic
+- **abs:** https://arxiv.org/abs/2604.00314 · **PDF:** https://arxiv.org/pdf/2604.00314v1
+
+## 543. Robust Multimodal Safety via Conditional Decoding
+
+- **ID:** `2604.00310` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Anurag Kumar, Raghuveer Peri, Jon Burnsky, et al.
+- **abs:** https://arxiv.org/abs/2604.00310 · **PDF:** https://arxiv.org/pdf/2604.00310v1
+
+## 544. Vocal Prognostic Digital Biomarkers in Monitoring Chronic Heart Failure: A Longitudinal Observational Study
+
+- **ID:** `2604.00308` · **primary:** `cs.SD` · **Submitted:** 2026-03-31
+- **Authors:** Fan Wu, Matthias P. Nägele, Daryush D. Mehta, et al.
+- **abs:** https://arxiv.org/abs/2604.00308 · **PDF:** https://arxiv.org/pdf/2604.00308v1
+
+## 545. SAGE: Subsurface AI-driven Geostatistical Extraction with proxy posterior
+
+- **ID:** `2604.00307` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Huseyin Tuna Erdinc, Ipsita Bhar, Rafael Orozco, et al.
+- **abs:** https://arxiv.org/abs/2604.00307 · **PDF:** https://arxiv.org/pdf/2604.00307v1
+
+## 546. Set-Based Value Function Characterization and Neural Approximation of Stabilization Domains for Input-Constrained Discrete-Time Systems
+
+- **ID:** `2604.00305` · **primary:** `eess.SY` · **Submitted:** 2026-03-31
+- **Authors:** Mohamed Serry, S. Sivaranjani, Jun Liu
+- **abs:** https://arxiv.org/abs/2604.00305 · **PDF:** https://arxiv.org/pdf/2604.00305v1
+
+## 547. Asymmetric Actor-Critic for Multi-turn LLM Agents
+
+- **ID:** `2604.00304` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Shuli Jiang, Zhaoyang Zhang, Yi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2604.00304 · **PDF:** https://arxiv.org/pdf/2604.00304v1
+
+## 548. SANA I2I: A Text Free Flow Matching Framework for Paired Image to Image Translation with a Case Study in Fetal MRI Artifact Reduction
+
+- **ID:** `2604.00298` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Italo Felix Santos, Gilson Antonio Giraldi, Heron Werner Junior
+- **abs:** https://arxiv.org/abs/2604.00298 · **PDF:** https://arxiv.org/pdf/2604.00298v1
+
+## 549. SYNTHONY: A Stress-Aware, Intent-Conditioned Agent for Deep Tabular Generative Models Selection
+
+- **ID:** `2604.00293` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Hochan Son, Xiaofeng Lin, Jason Ni, et al.
+- **abs:** https://arxiv.org/abs/2604.00293 · **PDF:** https://arxiv.org/pdf/2604.00293v1
+
+## 550. MambaVoiceCloning: Efficient and Expressive Text-to-Speech via State-Space Modeling and Diffusion Control
+
+- **ID:** `2604.00292` · **primary:** `cs.SD` · **Submitted:** 2026-03-31
+- **Authors:** Sahil Kumar, Namrataben Patel, Honggang Wang, et al.
+- **abs:** https://arxiv.org/abs/2604.00292 · **PDF:** https://arxiv.org/pdf/2604.00292v1
+
+## 551. Frege in the Flesh: Biolinguistics and the Neural Enforcement of Syntactic Structures
+
+- **ID:** `2604.00291` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Elliot Murphy
+- **abs:** https://arxiv.org/abs/2604.00291 · **PDF:** https://arxiv.org/pdf/2604.00291v1
+
+## 552. Improvisational Games as a Benchmark for Social Intelligence of AI Agents: The Case of Connections
+
+- **ID:** `2604.00284` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Gaurav Rajesh Parikh, Angikar Ghosal
+- **abs:** https://arxiv.org/abs/2604.00284 · **PDF:** https://arxiv.org/pdf/2604.00284v1
+
+## 553. Data-Driven Reachability Analysis via Diffusion Models with PAC Guarantees
+
+- **ID:** `2604.00283` · **primary:** `eess.SY` · **Submitted:** 2026-03-31
+- **Authors:** Yanliang Huang, Peng Xie, Wenyuan Wu, et al.
+- **abs:** https://arxiv.org/abs/2604.00283 · **PDF:** https://arxiv.org/pdf/2604.00283v1
+
+## 554. Human-in-the-Loop Control of Objective Drift in LLM-Assisted Computer Science Education
+
+- **ID:** `2604.00281` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Mark Dranias, Adam Whitley
+- **abs:** https://arxiv.org/abs/2604.00281 · **PDF:** https://arxiv.org/pdf/2604.00281v1
+
+## 555. VeriAct: Beyond Verifiability -- Agentic Synthesis of Correct and Complete Formal Specifications
+
+- **ID:** `2604.00280` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Md Rakib Hossain Misu, Iris Ma, Cristina V. Lopes
+- **abs:** https://arxiv.org/abs/2604.00280 · **PDF:** https://arxiv.org/pdf/2604.00280v1
+
+## 556. The Geometry of Compromise: Unlocking Generative Capabilities via Controllable Modality Alignment
+
+- **ID:** `2604.00279` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Hongyuan Liu, Qinli Yang, Wen Li, et al.
+- **abs:** https://arxiv.org/abs/2604.00279 · **PDF:** https://arxiv.org/pdf/2604.00279v1
+
+## 557. Hybrid Energy-Based Models for Physical AI: Provably Stable Identification of Port-Hamiltonian Dynamics
+
+- **ID:** `2604.00277` · **primary:** `eess.SY` · **Submitted:** 2026-03-31
+- **Authors:** Simone Betteti, Luca Laurenti
+- **abs:** https://arxiv.org/abs/2604.00277 · **PDF:** https://arxiv.org/pdf/2604.00277v2
+
+## 558. Benchmarking Interaction, Beyond Policy: a Reproducible Benchmark for Collaborative Instance Object Navigation
+
+- **ID:** `2604.00265` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Edoardo Zorzi, Francesco Taioli, Yiming Wang, et al.
+- **abs:** https://arxiv.org/abs/2604.00265 · **PDF:** https://arxiv.org/pdf/2604.00265v1
+
+## 559. Autonomous Adaptive Solver Selection for Chemistry Integration via Reinforcement Learning
+
+- **ID:** `2604.00264` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Eloghosa Ikponmwoba, Opeoluwa Owoyele
+- **abs:** https://arxiv.org/abs/2604.00264 · **PDF:** https://arxiv.org/pdf/2604.00264v1
+
+## 560. Can Large Language Models Self-Correct in Medical Question Answering? An Exploratory Study
+
+- **ID:** `2604.00261` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Zaifu Zhan, Mengyuan Cui, Rui Zhang
+- **abs:** https://arxiv.org/abs/2604.00261 · **PDF:** https://arxiv.org/pdf/2604.00261v2
+
+## 561. Learning to Shuffle: Block Reshuffling and Reversal Schemes for Stochastic Optimization
+
+- **ID:** `2604.00260` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Lam M. Nguyen, Dzung T. Phan, Jayant Kalagnanam
+- **abs:** https://arxiv.org/abs/2604.00260 · **PDF:** https://arxiv.org/pdf/2604.00260v1
+
+## 562. LLM Essay Scoring Under Holistic and Analytic Rubrics: Prompt Effects and Bias
+
+- **ID:** `2604.00259` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Filip J. Kucia, Anirban Chakraborty, Anna Wróblewska
+- **abs:** https://arxiv.org/abs/2604.00259 · **PDF:** https://arxiv.org/pdf/2604.00259v1
+
+## 563. Hierarchical Apprenticeship Learning from Imperfect Demonstrations with Evolving Rewards
+
+- **ID:** `2604.00258` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Md Mirajul Islam, Rajesh Debnath, Adittya Soukarjya Saha, et al.
+- **abs:** https://arxiv.org/abs/2604.00258 · **PDF:** https://arxiv.org/pdf/2604.00258v1
+
+## 564. Informed Machine Learning with Knowledge Landmarks
+
+- **ID:** `2604.00256` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Chuyi Dai, Witold Pedrycz, Suping Xu, et al.
+- **abs:** https://arxiv.org/abs/2604.00256 · **PDF:** https://arxiv.org/pdf/2604.00256v1
+
+## 565. A Safety-Aware Role-Orchestrated Multi-Agent LLM Framework for Behavioral Health Communication Simulation
+
+- **ID:** `2604.00249` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Ha Na Cho
+- **abs:** https://arxiv.org/abs/2604.00249 · **PDF:** https://arxiv.org/pdf/2604.00249v1
+
+## 566. REM-CTX: Automated Peer Review via Reinforcement Learning with Auxiliary Context
+
+- **ID:** `2604.00248` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Pawin Taechoyotin, Daniel E. Acuna
+- **abs:** https://arxiv.org/abs/2604.00248 · **PDF:** https://arxiv.org/pdf/2604.00248v1
+
+## 567. FGR-ColBERT: Identifying Fine-Grained Relevance Tokens During Retrieval
+
+- **ID:** `2604.00242` · **primary:** `cs.IR` · **Submitted:** 2026-03-31
+- **Authors:** Antonín Jarolím, Martin Fajčík
+- **abs:** https://arxiv.org/abs/2604.00242 · **PDF:** https://arxiv.org/pdf/2604.00242v1
+
+## 568. Softmax gradient policy for variance minimization and risk-averse multi armed bandits
+
+- **ID:** `2604.00241` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Gabriel Turinici
+- **abs:** https://arxiv.org/abs/2604.00241 · **PDF:** https://arxiv.org/pdf/2604.00241v1
+
+## 569. A Taxonomy of Programming Languages for Code Generation
+
+- **ID:** `2604.00239` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Nishat Raihan, Christian Newman, Marcos Zampieri
+- **abs:** https://arxiv.org/abs/2604.00239 · **PDF:** https://arxiv.org/pdf/2604.00239v1
+
+## 570. AI-Mediated Explainable Regulation for Justice
+
+- **ID:** `2604.00237` · **primary:** `cs.CY` · **Submitted:** 2026-03-31
+- **Authors:** Thomas Hofweber, Andreas Sudmann, Evangelos Pournaras
+- **abs:** https://arxiv.org/abs/2604.00237 · **PDF:** https://arxiv.org/pdf/2604.00237v1
+
+## 571. Hierarchical Discrete Flow Matching for Graph Generation
+
+- **ID:** `2604.00236` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Yoann Boget, Pablo Strasser, Alexandros Kalousis
+- **abs:** https://arxiv.org/abs/2604.00236 · **PDF:** https://arxiv.org/pdf/2604.00236v1
+
+## 572. MAC-Attention: a Match-Amend-Complete Scheme for Fast and Accurate Attention Computation
+
+- **ID:** `2604.00235` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Jinghan Yao, Sam Adé Jacobs, Walid Krichene, et al.
+- **abs:** https://arxiv.org/abs/2604.00235 · **PDF:** https://arxiv.org/pdf/2604.00235v1
+
+## 573. Neural Collapse Dynamics: Depth, Activation, Regularisation, and Feature Norm Threshold
+
+- **ID:** `2604.00230` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Anamika Paul Rupa
+- **abs:** https://arxiv.org/abs/2604.00230 · **PDF:** https://arxiv.org/pdf/2604.00230v1
+
+## 574. Do Language Models Know When They'll Refuse? Probing Introspective Awareness of Safety Boundaries
+
+- **ID:** `2604.00228` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Tanay Gondil
+- **abs:** https://arxiv.org/abs/2604.00228 · **PDF:** https://arxiv.org/pdf/2604.00228v1
+
+## 575. Diversity-Aware Reverse Kullback-Leibler Divergence for Large Language Model Distillation
+
+- **ID:** `2604.00223` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Hoang-Chau Luong, Dat Ba Tran, Lingwei Chen
+- **abs:** https://arxiv.org/abs/2604.00223 · **PDF:** https://arxiv.org/pdf/2604.00223v1
+
+## 576. Risk-Aware Batch Testing for Performance Regression Detection
+
+- **ID:** `2604.00222` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Ali Sayedsalehi, Peter C. Rigby, Gregory Mierzwinski
+- **abs:** https://arxiv.org/abs/2604.00222 · **PDF:** https://arxiv.org/pdf/2604.00222v1
+
+## 577. Do LLMs Know What Is Private Internally? Probing and Steering Contextual Privacy Norms in Large Language Model Representations
+
+- **ID:** `2604.00209` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Haoran Wang, Li Xiong, Kai Shu
+- **abs:** https://arxiv.org/abs/2604.00209 · **PDF:** https://arxiv.org/pdf/2604.00209v1
+
+## 578. Measuring the Representational Alignment of Neural Systems in Superposition
+
+- **ID:** `2604.00208` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Sunny Liu, Habon Issa, André Longon, et al.
+- **abs:** https://arxiv.org/abs/2604.00208 · **PDF:** https://arxiv.org/pdf/2604.00208v1
+
+## 579. Lead Zirconate Titanate Reservoir Computing for Classification of Written and Spoken Digits
+
+- **ID:** `2604.00207` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Thomas Buckley, Leslie Schumm, Manor Askenazi, et al.
+- **abs:** https://arxiv.org/abs/2604.00207 · **PDF:** https://arxiv.org/pdf/2604.00207v1
+
+## 580. Unsupervised 4D Flow MRI Velocity Enhancement and Unwrapping Using Divergence-Free Neural Networks
+
+- **ID:** `2604.00205` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Javier Bisbal, Julio Sotelo, Hernán Mella, et al.
+- **abs:** https://arxiv.org/abs/2604.00205 · **PDF:** https://arxiv.org/pdf/2604.00205v1
+
+## 581. Offline Constrained RLHF with Multiple Preference Oracles
+
+- **ID:** `2604.00200` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Brenden Latham, Mehrdad Moharrami
+- **abs:** https://arxiv.org/abs/2604.00200 · **PDF:** https://arxiv.org/pdf/2604.00200v1
+
+## 582. QUEST: A robust attention formulation using query-modulated spherical attention
+
+- **ID:** `2604.00199` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Hariprasath Govindarajan, Per Sidén, Jacob Roll, et al.
+- **abs:** https://arxiv.org/abs/2604.00199 · **PDF:** https://arxiv.org/pdf/2604.00199v1
+
+## 583. Lévy-Flow Models: Heavy-Tail-Aware Normalizing Flows for Financial Risk Management
+
+- **ID:** `2604.00195` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Rachid Drissi
+- **abs:** https://arxiv.org/abs/2604.00195 · **PDF:** https://arxiv.org/pdf/2604.00195v1
+
+## 584. Making Sense of AI Agents Hype: Adoption, Architectures, and Takeaways from Practitioners
+
+- **ID:** `2604.00189` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Ruoyu Su, Matteo Esposito, Roberta Capuano, et al.
+- **abs:** https://arxiv.org/abs/2604.00189 · **PDF:** https://arxiv.org/pdf/2604.00189v1
+
+## 585. Explainable AI for Blind and Low-Vision Users: Navigating Trust, Modality, and Interpretability in the Agentic Era
+
+- **ID:** `2604.00187` · **primary:** `cs.HC` · **Submitted:** 2026-03-31
+- **Authors:** Abu Noman Md Sakib, Protik Dey, Zijie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2604.00187 · **PDF:** https://arxiv.org/pdf/2604.00187v1
+
+## 586. Agentic AI and Occupational Displacement: A Multi-Regional Task Exposure Analysis of Emerging Labor Market Disruption
+
+- **ID:** `2604.00186` · **primary:** `eess.SY` · **Submitted:** 2026-03-31
+- **Authors:** Ravish Gupta, Saket Kumar
+- **abs:** https://arxiv.org/abs/2604.00186 · **PDF:** https://arxiv.org/pdf/2604.00186v1
+
+## 587. NFC based inventory control system for secure and efficient communication
+
+- **ID:** `2604.00181` · **primary:** `cs.CR` · **Submitted:** 2026-03-31
+- **Authors:** Razi Iqbal, Awais Ahmad, Asfandyar Gillani
+- **abs:** https://arxiv.org/abs/2604.00181 · **PDF:** https://arxiv.org/pdf/2604.00181v1
+
+## 588. Finite-Time Analysis of Projected Two-Time-Scale Stochastic Approximation
+
+- **ID:** `2604.00179` · **primary:** `eess.SY` · **Submitted:** 2026-03-31
+- **Authors:** Yitao Bai, Thinh T. Doan, Justin Romberg
+- **abs:** https://arxiv.org/abs/2604.00179 · **PDF:** https://arxiv.org/pdf/2604.00179v1
+
+## 589. Sit-to-Stand Transitions Detection and Duration Measurement Using Smart Lacelock Sensor
+
+- **ID:** `2604.00175` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Md Rafi Islam, Md Rejwanul Haque, Elizabeth Choma, et al.
+- **abs:** https://arxiv.org/abs/2604.00175 · **PDF:** https://arxiv.org/pdf/2604.00175v1
+
+## 590. Polish phonology and morphology through the lens of distributional semantics
+
+- **ID:** `2604.00174` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Paula Orzechowska, R. Harald Baayen
+- **abs:** https://arxiv.org/abs/2604.00174 · **PDF:** https://arxiv.org/pdf/2604.00174v1
+
+## 591. Unified Architecture Metamodel of Information Systems Developed by Generative AI
+
+- **ID:** `2604.00171` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Oleg Grynets, Vasyl Lyashkevych
+- **abs:** https://arxiv.org/abs/2604.00171 · **PDF:** https://arxiv.org/pdf/2604.00171v1
+
+## 592. Neural-Assisted in-Motion Self-Heading Alignment
+
+- **ID:** `2604.00168` · **primary:** `cs.RO` · **Submitted:** 2026-03-31
+- **Authors:** Zeev Yampolsky, Felipe O. Silva, Adriano Frutuoso, et al.
+- **abs:** https://arxiv.org/abs/2604.00168 · **PDF:** https://arxiv.org/pdf/2604.00168v1
+
+## 593. A Study on the Impact of Fault localization Granularity for Repository-Scale Code Repair Tasks
+
+- **ID:** `2604.00167` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Joseph Townsend, Chandresh Pravin, Kwun Ho Ngan, et al.
+- **abs:** https://arxiv.org/abs/2604.00167 · **PDF:** https://arxiv.org/pdf/2604.00167v1
+
+## 594. Epileptic Seizure Detection in Separate Frequency Bands Using Feature Analysis and Graph Convolutional Neural Network (GCN) from Electroencephalogram (EEG) Signals
+
+- **ID:** `2604.00163` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Ferdaus Anam Jibon, Fazlul Hasan Siddiqui, F. Deeba, et al.
+- **abs:** https://arxiv.org/abs/2604.00163 · **PDF:** https://arxiv.org/pdf/2604.00163v1
+
+## 595. Open, Reliable, and Collective: A Community-Driven Framework for Tool-Using AI Agents
+
+- **ID:** `2604.00137` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Hy Dang, Quang Dao, Meng Jiang
+- **abs:** https://arxiv.org/abs/2604.00137 · **PDF:** https://arxiv.org/pdf/2604.00137v1
+
+## 596. ParetoBandit: Budget-Paced Adaptive Routing for Non-Stationary LLM Serving
+
+- **ID:** `2604.00136` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Annette Taberner-Miller
+- **abs:** https://arxiv.org/abs/2604.00136 · **PDF:** https://arxiv.org/pdf/2604.00136v1
+
+## 597. Predicting Wave Reflection and Transmission in Heterogeneous Media via Fourier Operator-Based Transformer Modeling
+
+- **ID:** `2604.00132` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Zhe Bai, Hans Johansen
+- **abs:** https://arxiv.org/abs/2604.00132 · **PDF:** https://arxiv.org/pdf/2604.00132v1
+
+## 598. Oblivion: Self-Adaptive Agentic Memory Control through Decay-Driven Activation
+
+- **ID:** `2604.00131` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Ashish Rana, Chia-Chien Hung, Qumeng Sun, et al.
+- **abs:** https://arxiv.org/abs/2604.00131 · **PDF:** https://arxiv.org/pdf/2604.00131v1
+
+## 599. Hierarchical Chain-of-Thought Prompting: Enhancing LLM Reasoning Performance and Efficiency
+
+- **ID:** `2604.00130` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Xingshuai Huang, Derek Li, Bahareh Nikpour, et al.
+- **abs:** https://arxiv.org/abs/2604.00130 · **PDF:** https://arxiv.org/pdf/2604.00130v1
+
+## 600. From Domain Understanding to Design Readiness: a playbook for GenAI-supported learning in Software Engineering
+
+- **ID:** `2604.00120` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Rafal Wlodarski
+- **abs:** https://arxiv.org/abs/2604.00120 · **PDF:** https://arxiv.org/pdf/2604.00120v1
+
+## 601. Efficient Software Vulnerability Detection Using Transformer-based Models
+
+- **ID:** `2604.00112` · **primary:** `cs.CR` · **Submitted:** 2026-03-31
+- **Authors:** Sameer Shaik, Zhen Huang, Daniela Stan Raicu, et al.
+- **abs:** https://arxiv.org/abs/2604.00112 · **PDF:** https://arxiv.org/pdf/2604.00112v1
+
+## 602. Speeding Up Mixed-Integer Programming Solvers with Sparse Learning for Branching
+
+- **ID:** `2604.00094` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Selin Bayramoğlu, George L Nemhauser, Nikolaos V Sahinidis
+- **abs:** https://arxiv.org/abs/2604.00094 · **PDF:** https://arxiv.org/pdf/2604.00094v1
+
+## 603. Hierarchical Pre-Training of Vision Encoders with Large Language Models
+
+- **ID:** `2604.00086` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Eugene Lee, Ting-Yu Chang, Jui-Huang Tsai, et al.
+- **abs:** https://arxiv.org/abs/2604.00086 · **PDF:** https://arxiv.org/pdf/2604.00086v1
+
+## 604. One Panel Does Not Fit All: Case-Adaptive Multi-Agent Deliberation for Clinical Prediction
+
+- **ID:** `2604.00085` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Yuxing Lu, Yushuhong Lin, Jason Zhang
+- **abs:** https://arxiv.org/abs/2604.00085 · **PDF:** https://arxiv.org/pdf/2604.00085v1
+
+## 605. Automatic Identification of Parallelizable Loops Using Transformer-Based Source Code Representations
+
+- **ID:** `2603.30040` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Izavan dos S. Correia, Henrique C. T. Santos, Tiago A. E. Ferreira
+- **abs:** https://arxiv.org/abs/2603.30040 · **PDF:** https://arxiv.org/pdf/2603.30040v1
+
+## 606. Aligned, Orthogonal or In-conflict: When can we safely optimize Chain-of-Thought?
+
+- **ID:** `2603.30036` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Max Kaufmann, David Lindner, Roland S. Zimmermann, et al.
+- **abs:** https://arxiv.org/abs/2603.30036 · **PDF:** https://arxiv.org/pdf/2603.30036v1
+
+## 607. Reward-Based Online LLM Routing via NeuralUCB
+
+- **ID:** `2603.30035` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Ming-Hua Tsai, Phat Tran
+- **abs:** https://arxiv.org/abs/2603.30035 · **PDF:** https://arxiv.org/pdf/2603.30035v1
+
+## 608. Tucker Attention: A generalization of approximate attention mechanisms
+
+- **ID:** `2603.30033` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Timon Klein, Jonas Kusch, Sebastian Sager, et al.
+- **abs:** https://arxiv.org/abs/2603.30033 · **PDF:** https://arxiv.org/pdf/2603.30033v1
+
+## 609. Covertly improving intelligibility with data-driven adaptations of speech timing
+
+- **ID:** `2603.30032` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Paige Tuttösí, Angelica Lim, H. Henny Yeung, et al.
+- **abs:** https://arxiv.org/abs/2603.30032 · **PDF:** https://arxiv.org/pdf/2603.30032v1
+
+## 610. Cognitive Friction: A Decision-Theoretic Framework for Bounded Deliberation in Tool-Using Agents
+
+- **ID:** `2603.30031` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Davide Di Gioia
+- **abs:** https://arxiv.org/abs/2603.30031 · **PDF:** https://arxiv.org/pdf/2603.30031v3
+
+## 611. Beyond Symbolic Control: Societal Consequences of AI-Driven Workforce Displacement and the Imperative for Genuine Human Oversight Architectures
+
+- **ID:** `2604.00081` · **primary:** `cs.CY` · **Submitted:** 2026-03-31
+- **Authors:** Richard J. Mitchell
+- **abs:** https://arxiv.org/abs/2604.00081 · **PDF:** https://arxiv.org/pdf/2604.00081v1
+
+## 612. ContextClaim: A Context-Driven Paradigm for Verifiable Claim Detection
+
+- **ID:** `2603.30025` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Yufeng Li, Rrubaa Panchendrarajan, Arkaitz Zubiaga
+- **abs:** https://arxiv.org/abs/2603.30025 · **PDF:** https://arxiv.org/pdf/2603.30025v1
+
+## 613. Hybrid Framework for Robotic Manipulation: Integrating Reinforcement Learning and Large Language Models
+
+- **ID:** `2603.30022` · **primary:** `cs.RO` · **Submitted:** 2026-03-31
+- **Authors:** Md Saad, Sajjad Hussain, Mohd Suhaib
+- **abs:** https://arxiv.org/abs/2603.30022 · **PDF:** https://arxiv.org/pdf/2603.30022v1
+
+## 614. Refined Detection for Gumbel Watermarking
+
+- **ID:** `2603.30017` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Tor Lattimore
+- **abs:** https://arxiv.org/abs/2603.30017 · **PDF:** https://arxiv.org/pdf/2603.30017v1
+
+## 615. Architecting Secure AI Agents: Perspectives on System-Level Defenses Against Indirect Prompt Injection Attacks
+
+- **ID:** `2603.30016` · **primary:** `cs.CR` · **Submitted:** 2026-03-31
+- **Authors:** Chong Xiang, Drew Zagieboylo, Shaona Ghosh, et al.
+- **abs:** https://arxiv.org/abs/2603.30016 · **PDF:** https://arxiv.org/pdf/2603.30016v1
+
+## 616. Scalable AI-assisted Workflow Management for Detector Design Optimization Using Distributed Computing
+
+- **ID:** `2603.30014` · **primary:** `cs.DC` · **Submitted:** 2026-03-31
+- **Authors:** Derek Anderson, Amit Bashyal, Markus Diefenthaler, et al.
+- **abs:** https://arxiv.org/abs/2603.30014 · **PDF:** https://arxiv.org/pdf/2603.30014v1
+
+## 617. Tracking Equivalent Mechanistic Interpretations Across Neural Networks
+
+- **ID:** `2603.30002` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Alan Sun, Mariya Toneva
+- **abs:** https://arxiv.org/abs/2603.30002 · **PDF:** https://arxiv.org/pdf/2603.30002v1
+
+## 618. Phyelds: A Pythonic Framework for Aggregate Computing
+
+- **ID:** `2603.29999` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Gianluca Aguzzi, Davide Domini, Nicolas Farabegoli, et al.
+- **abs:** https://arxiv.org/abs/2603.29999 · **PDF:** https://arxiv.org/pdf/2603.29999v1
+
+## 619. Enhancing Structural Mapping with LLM-derived Abstractions for Analogical Reasoning in Narratives
+
+- **ID:** `2603.29997` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Mohammadhossein Khojasteh, Yifan Jiang, Stefano De Giorgis, et al.
+- **abs:** https://arxiv.org/abs/2603.29997 · **PDF:** https://arxiv.org/pdf/2603.29997v1
+
+## 620. Extending MONA in Camera Dropbox: Reproduction, Learned Approval, and Design Implications for Reward-Hacking Mitigation
+
+- **ID:** `2603.29993` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Nathan Heath
+- **abs:** https://arxiv.org/abs/2603.29993 · **PDF:** https://arxiv.org/pdf/2603.29993v1
+
+## 621. Aligning Validation with Deployment: Target-Weighted Cross-Validation for Spatial Prediction
+
+- **ID:** `2603.29981` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Alexander Brenning, Thomas Suesse
+- **abs:** https://arxiv.org/abs/2603.29981 · **PDF:** https://arxiv.org/pdf/2603.29981v1
+
+## 622. Structural Feature Engineering for Generative Engine Optimization: How Content Structure Shapes Citation Behavior
+
+- **ID:** `2603.29979` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Junwei Yu, Mufeng Yang, Yepeng Ding, et al.
+- **abs:** https://arxiv.org/abs/2603.29979 · **PDF:** https://arxiv.org/pdf/2603.29979v1
+
+## 623. Quantifying Cross-Modal Interactions in Multimodal Glioma Survival Prediction via InterSHAP: Evidence for Additive Signal Integration
+
+- **ID:** `2603.29977` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Iain Swift, JingHua Ye, Ruairi O'Reilly
+- **abs:** https://arxiv.org/abs/2603.29977 · **PDF:** https://arxiv.org/pdf/2603.29977v1
+
+## 624. Meteorology-Driven GPT4AP: A Multi-Task Forecasting LLM for Atmospheric Air Pollution in Data-Scarce Settings
+
+- **ID:** `2603.29974` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Prasanjit Dey, Soumyabrata Dev, Bianca Schoen-Phelan
+- **abs:** https://arxiv.org/abs/2603.29974 · **PDF:** https://arxiv.org/pdf/2603.29974v1
+
+## 625. Do covariates explain why these groups differ? The choice of reference group can reverse conclusions in the Oaxaca-Blinder decomposition
+
+- **ID:** `2603.29972` · **primary:** `stat.ME` · **Submitted:** 2026-03-31
+- **Authors:** Manuel Quintero, Advik Shreekumar, William T. Stephenson, et al.
+- **abs:** https://arxiv.org/abs/2603.29972 · **PDF:** https://arxiv.org/pdf/2603.29972v1
+
+## 626. Trimodal Deep Learning for Glioma Survival Prediction: A Feasibility Study Integrating Histopathology, Gene Expression, and MRI
+
+- **ID:** `2603.29968` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Iain Swift, JingHua Ye
+- **abs:** https://arxiv.org/abs/2603.29968 · **PDF:** https://arxiv.org/pdf/2603.29968v1
+
+## 627. Think Anywhere in Code Generation
+
+- **ID:** `2603.29957` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Xue Jiang, Tianyu Zhang, Ge Li, et al.
+- **abs:** https://arxiv.org/abs/2603.29957 · **PDF:** https://arxiv.org/pdf/2603.29957v2
+
+## 628. Structured Intent as a Protocol-Like Communication Layer: Cross-Model Robustness, Framework Comparison, and the Weak-Model Compensation Effect
+
+- **ID:** `2603.29953` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Peng Gang
+- **abs:** https://arxiv.org/abs/2603.29953 · **PDF:** https://arxiv.org/pdf/2603.29953v1
+
+## 629. Physiological and Semantic Patterns in Medical Teams Using an Intelligent Tutoring System
+
+- **ID:** `2603.29950` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Xiaoshan Huang, Conrad Borchers, Jiayi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.29950 · **PDF:** https://arxiv.org/pdf/2603.29950v1
+
+## 630. Real-Time Explanations for Tabular Foundation Models
+
+- **ID:** `2603.29946` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Luan Borges Teodoro Reis Sena, Francisco Galuppo Azevedo
+- **abs:** https://arxiv.org/abs/2603.29946 · **PDF:** https://arxiv.org/pdf/2603.29946v1
+
+## 631. Four Generations of Quantum Biomedical Sensors
+
+- **ID:** `2603.29944` · **primary:** `quant-ph` · **Submitted:** 2026-03-31
+- **Authors:** Xin Jin, Priyam Srivastava, Ronghe Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.29944 · **PDF:** https://arxiv.org/pdf/2603.29944v2
+
+## 632. Better than Average: Spatially-Aware Aggregation of Segmentation Uncertainty Improves Downstream Performance
+
+- **ID:** `2603.29941` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Vanessa Emanuela Guarino, Claudia Winklmayr, Jannik Franzen, et al.
+- **abs:** https://arxiv.org/abs/2603.29941 · **PDF:** https://arxiv.org/pdf/2603.29941v1
+
+## 633. Rewrite the News: Tracing Editorial Reuse Across News Agencies
+
+- **ID:** `2603.29937` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Soveatin Kuntur, Nina Smirnova, Anna Wroblewska, et al.
+- **abs:** https://arxiv.org/abs/2603.29937 · **PDF:** https://arxiv.org/pdf/2603.29937v1
+
+## 634. Rethinking AI Literacy Education in Higher Education: Bridging Risk Perception and Responsible Adoption
+
+- **ID:** `2603.29935` · **primary:** `cs.CY` · **Submitted:** 2026-03-31
+- **Authors:** Shasha Yu, Fiona Carroll, Barry L. Bentley
+- **abs:** https://arxiv.org/abs/2603.29935 · **PDF:** https://arxiv.org/pdf/2603.29935v1
+
+## 635. Bethe Ansatz with a Large Language Model
+
+- **ID:** `2603.29932` · **primary:** `cond-mat.stat-mech` · **Submitted:** 2026-03-31
+- **Authors:** Balázs Pozsgay, István Vona
+- **abs:** https://arxiv.org/abs/2603.29932 · **PDF:** https://arxiv.org/pdf/2603.29932v1
+
+## 636. ScoringBench: A Benchmark for Evaluating Tabular Foundation Models with Proper Scoring Rules
+
+- **ID:** `2603.29928` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Jonas Landsgesell, Pascal Knoll
+- **abs:** https://arxiv.org/abs/2603.29928 · **PDF:** https://arxiv.org/pdf/2603.29928v1
+
+## 637. End-to-End Image Compression with Segmentation Guided Dual Coding for Wind Turbines
+
+- **ID:** `2603.29927` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Raül Pérez-Gonzalo, Andreas Espersen, Søren Forchhammer, et al.
+- **abs:** https://arxiv.org/abs/2603.29927 · **PDF:** https://arxiv.org/pdf/2603.29927v1
+
+## 638. Training deep learning based dynamic MR image reconstruction using synthetic fractals
+
+- **ID:** `2603.29922` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Anirudh Raman, Olivier Jaubert, Mark Wrobel, et al.
+- **abs:** https://arxiv.org/abs/2603.29922 · **PDF:** https://arxiv.org/pdf/2603.29922v1
+
+## 639. Uncertainty Gating for Cost-Aware Explainable Artificial Intelligence
+
+- **ID:** `2603.29915` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Georgii Mikriukov, Grégoire Montavon, Marina M. -C. Höhne
+- **abs:** https://arxiv.org/abs/2603.29915 · **PDF:** https://arxiv.org/pdf/2603.29915v1
+
+## 640. Task Scarcity and Label Leakage in Relational Transfer Learning
+
+- **ID:** `2603.29914` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Francisco Galuppo Azevedo, Clarissa Lima Loures, Denis Oliveira Correa
+- **abs:** https://arxiv.org/abs/2603.29914 · **PDF:** https://arxiv.org/pdf/2603.29914v1
+
+## 641. SISA: A Scale-In Systolic Array for GEMM Acceleration
+
+- **ID:** `2603.29913` · **primary:** `cs.AR` · **Submitted:** 2026-03-31
+- **Authors:** Luigi Altamura, Alessio Cicero, Mateo Vázquez Maceiras, et al.
+- **abs:** https://arxiv.org/abs/2603.29913 · **PDF:** https://arxiv.org/pdf/2603.29913v1
+
+## 642. Learning to Play Blackjack: A Curriculum Learning Perspective
+
+- **ID:** `2604.00076` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Amirreza Alasti, Efe Erdal, Yücel Celik, et al.
+- **abs:** https://arxiv.org/abs/2604.00076 · **PDF:** https://arxiv.org/pdf/2604.00076v2
+
+## 643. C-TRAIL: A Commonsense World Framework for Trajectory Planning in Autonomous Driving
+
+- **ID:** `2603.29908` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Zhihong Cui, Haoran Tang, Tianyi Li, et al.
+- **abs:** https://arxiv.org/abs/2603.29908 · **PDF:** https://arxiv.org/pdf/2603.29908v1
+
+## 644. $p$-adic Character Neural Network
+
+- **ID:** `2603.29905` · **primary:** `math.NT` · **Submitted:** 2026-03-31
+- **Authors:** Tomoki Mihara
+- **abs:** https://arxiv.org/abs/2603.29905 · **PDF:** https://arxiv.org/pdf/2603.29905v1
+
+## 645. ATP-Bench: Towards Agentic Tool Planning for MLLM Interleaved Generation
+
+- **ID:** `2603.29902` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Yinuo Liu, Zi Qian, Heng Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.29902 · **PDF:** https://arxiv.org/pdf/2603.29902v1
+
+## 646. Less Is More? Selective Visual Attention to High-Importance Regions for Multimodal Radiology Summarization
+
+- **ID:** `2603.29901` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Mst. Fahmida Sultana Naznin, Adnan Ibney Faruq, Mushfiqur Rahman, et al.
+- **abs:** https://arxiv.org/abs/2603.29901 · **PDF:** https://arxiv.org/pdf/2603.29901v1
+
+## 647. FLEURS-Kobani: Extending the FLEURS Dataset for Northern Kurdish
+
+- **ID:** `2603.29892` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Daban Q. Jaff, Mohammad Mohammadamini
+- **abs:** https://arxiv.org/abs/2603.29892 · **PDF:** https://arxiv.org/pdf/2603.29892v1
+
+## 648. Penalized GMM Framework for Inference on Functionals of Nonparametric Instrumental Variable Estimators
+
+- **ID:** `2603.29889` · **primary:** `econ.EM` · **Submitted:** 2026-03-31
+- **Authors:** Edvard Bakhitov
+- **abs:** https://arxiv.org/abs/2603.29889 · **PDF:** https://arxiv.org/pdf/2603.29889v1
+
+## 649. ShapE-GRPO: Shapley-Enhanced Reward Allocation for Multi-Candidate LLM Training
+
+- **ID:** `2603.29871` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Rui Ai, Yu Pan, David Simchi-Levi, et al.
+- **abs:** https://arxiv.org/abs/2603.29871 · **PDF:** https://arxiv.org/pdf/2603.29871v1
+
+## 650. Towards Empowering Consumers through Sentence-level Readability Scoring in German ESG Reports
+
+- **ID:** `2603.29861` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Benjamin Josef Schüßler, Jakob Prange
+- **abs:** https://arxiv.org/abs/2603.29861 · **PDF:** https://arxiv.org/pdf/2603.29861v1
+
+## 651. SNEAK: Evaluating Strategic Communication and Information Leakage in Large Language Models
+
+- **ID:** `2603.29846` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Adar Avsian, Larry Heck
+- **abs:** https://arxiv.org/abs/2603.29846 · **PDF:** https://arxiv.org/pdf/2603.29846v1
+
+## 652. DIAL: Decoupling Intent and Action via Latent World Modeling for End-to-End VLA
+
+- **ID:** `2603.29844` · **primary:** `cs.RO` · **Submitted:** 2026-03-31
+- **Authors:** Yi Chen, Yuying Ge, Hui Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.29844 · **PDF:** https://arxiv.org/pdf/2603.29844v1
+
+## 653. Toward Generalizable Whole Brain Representations with High-Resolution Light-Sheet Data
+
+- **ID:** `2603.29842` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Minyoung E. Kim, Dae Hee Yun, Aditi V. Patel, et al.
+- **abs:** https://arxiv.org/abs/2603.29842 · **PDF:** https://arxiv.org/pdf/2603.29842v1
+
+## 654. DiSGMM: A Method for Time-varying Microscopic Weight Completion on Road Networks
+
+- **ID:** `2603.29837` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Yan Lin, Jilin Hu, Shengnan Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.29837 · **PDF:** https://arxiv.org/pdf/2603.29837v1
+
+## 655. Owl-AuraID 1.0: An Intelligent System for Autonomous Scientific Instrumentation and Scientific Data Analysis
+
+- **ID:** `2603.29828` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Han Deng, Anqi Zou, Hanling Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.29828 · **PDF:** https://arxiv.org/pdf/2603.29828v1
+
+## 656. Curvature-Guided LoRA: Steering in the pretrained NTK subspace
+
+- **ID:** `2603.29824` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Frédéric Zheng, Alexandre Proutière
+- **abs:** https://arxiv.org/abs/2603.29824 · **PDF:** https://arxiv.org/pdf/2603.29824v1
+
+## 657. Loss Gap Parity for Fairness in Heterogeneous Federated Learning
+
+- **ID:** `2603.29818` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Brahim Erraji, Michaël Perrot, Aurélien Bellet
+- **abs:** https://arxiv.org/abs/2603.29818 · **PDF:** https://arxiv.org/pdf/2603.29818v1
+
+## 658. AMShortcut: An Inference- and Training-Efficient Inverse Design Model for Amorphous Materials
+
+- **ID:** `2603.29812` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Yan Lin, Jonas A. Finkler, Tao Du, et al.
+- **abs:** https://arxiv.org/abs/2603.29812 · **PDF:** https://arxiv.org/pdf/2603.29812v1
+
+## 659. PASM: Population Adaptive Symbolic Mixture-of-Experts Model for Cross-location Hurricane Evacuation Decision Prediction
+
+- **ID:** `2604.00074` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Xiao Qian, Shangjia Dong
+- **abs:** https://arxiv.org/abs/2604.00074 · **PDF:** https://arxiv.org/pdf/2604.00074v1
+
+## 660. From Density Matrices to Phase Transitions in Deep Learning: Spectral Early Warnings and Interpretability
+
+- **ID:** `2603.29805` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Max Hennick, Guillaume Corlouer
+- **abs:** https://arxiv.org/abs/2603.29805 · **PDF:** https://arxiv.org/pdf/2603.29805v2
+
+## 661. ENEIDE: A High Quality Silver Standard Dataset for Named Entity Recognition and Linking in Historical Italian
+
+- **ID:** `2603.29801` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Cristian Santini, Sebastian Barzaghi, Paolo Sernani, et al.
+- **abs:** https://arxiv.org/abs/2603.29801 · **PDF:** https://arxiv.org/pdf/2603.29801v1
+
+## 662. Multimodal Machine Learning for Early Prediction of Metastasis in a Swedish Multi-Cancer Cohort
+
+- **ID:** `2603.29793` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Franco Rugolon, Korbinian Randl, Braslav Jovanovic, et al.
+- **abs:** https://arxiv.org/abs/2603.29793 · **PDF:** https://arxiv.org/pdf/2603.29793v1
+
+## 663. Reasoning-Driven Synthetic Data Generation and Evaluation
+
+- **ID:** `2603.29791` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Tim R. Davidson, Benoit Seguin, Enrico Bacis, et al.
+- **abs:** https://arxiv.org/abs/2603.29791 · **PDF:** https://arxiv.org/pdf/2603.29791v1
+
+## 664. From Skeletons to Semantics: Design and Deployment of a Hybrid Edge-Based Action Detection System for Public Safety
+
+- **ID:** `2603.29777` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Ganen Sethupathy, Lalit Dumka, Jan Schagen
+- **abs:** https://arxiv.org/abs/2603.29777 · **PDF:** https://arxiv.org/pdf/2603.29777v1
+
+## 665. Terminal Agents Suffice for Enterprise Automation
+
+- **ID:** `2604.00073` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Patrice Bechard, Orlando Marquez Ayala, Emily Chen, et al.
+- **abs:** https://arxiv.org/abs/2604.00073 · **PDF:** https://arxiv.org/pdf/2604.00073v1
+
+## 666. Associative Constructive Evolution: Enhancing Metaheuristics through Hebbian-Learned Generative Guidance
+
+- **ID:** `2603.29774` · **primary:** `cs.NE` · **Submitted:** 2026-03-31
+- **Authors:** Shanxian Lin, Yuichi Nagata, Haichuan Yang
+- **abs:** https://arxiv.org/abs/2603.29774 · **PDF:** https://arxiv.org/pdf/2603.29774v1
+
+## 667. Big2Small: A Unifying Neural Network Framework for Model Compression
+
+- **ID:** `2603.29768` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Jing-Xiao Liao, Haoran Wang, Tao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.29768 · **PDF:** https://arxiv.org/pdf/2603.29768v1
+
+## 668. Training-Free Dynamic Upcycling of Expert Language Models
+
+- **ID:** `2603.29765` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Eros Fanì, Oğuzhan Ersoy
+- **abs:** https://arxiv.org/abs/2603.29765 · **PDF:** https://arxiv.org/pdf/2603.29765v1
+
+## 669. Tracking vs. Deciding: The Dual-Capability Bottleneck in Searchless Chess Transformers
+
+- **ID:** `2603.29761` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Quanhao Li, Wei Jiang
+- **abs:** https://arxiv.org/abs/2603.29761 · **PDF:** https://arxiv.org/pdf/2603.29761v1
+
+## 670. TSHA: A Benchmark for Visual Language Models in Trustworthy Safety Hazard Assessment Scenarios
+
+- **ID:** `2603.29759` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Qiucheng Yu, Ruijie Xu, Mingang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.29759 · **PDF:** https://arxiv.org/pdf/2603.29759v1
+
+## 671. One-for-All: A Lightweight Stabilized and Parameter-Efficient Pre-trained LLM for Time Series Forecasting
+
+- **ID:** `2603.29756` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Prasanjit Dey, Soumyabrata Dev, Bianca Schoen-Phelan
+- **abs:** https://arxiv.org/abs/2603.29756 · **PDF:** https://arxiv.org/pdf/2603.29756v1
+
+## 672. Empirical Validation of the Classification-Verification Dichotomy for AI Safety Gates
+
+- **ID:** `2604.00072` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Arsenios Scrivens
+- **abs:** https://arxiv.org/abs/2604.00072 · **PDF:** https://arxiv.org/pdf/2604.00072v1
+
+## 673. CausalPulse: An Industrial-Grade Neurosymbolic Multi-Agent Copilot for Causal Diagnostics in Smart Manufacturing
+
+- **ID:** `2603.29755` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Chathurangi Shyalika, Utkarshani Jaimini, Cory Henson, et al.
+- **abs:** https://arxiv.org/abs/2603.29755 · **PDF:** https://arxiv.org/pdf/2603.29755v1
+
+## 674. Brain MR Image Synthesis with Multi-contrast Self-attention GAN
+
+- **ID:** `2604.00070` · **primary:** `eess.IV` · **Submitted:** 2026-03-31
+- **Authors:** Zaid A. Abod, Furqan Aziz
+- **abs:** https://arxiv.org/abs/2604.00070 · **PDF:** https://arxiv.org/pdf/2604.00070v1
+
+## 675. HyperKKL: Learning KKL Observers for Non-Autonomous Nonlinear Systems via Hypernetwork-Based Input Conditioning
+
+- **ID:** `2603.29744` · **primary:** `eess.SY` · **Submitted:** 2026-03-31
+- **Authors:** Yahia Salaheldin Shaaban, Abdelrahman Sayed Sayed, M. Umar B. Niazi, et al.
+- **abs:** https://arxiv.org/abs/2603.29744 · **PDF:** https://arxiv.org/pdf/2603.29744v1
+
+## 676. BotVerse: Real-Time Event-Driven Simulation of Social Agents
+
+- **ID:** `2603.29741` · **primary:** `cs.SI` · **Submitted:** 2026-03-31
+- **Authors:** Edoardo Allegrini, Edoardo Di Paolo, Angelo Spognardi, et al.
+- **abs:** https://arxiv.org/abs/2603.29741 · **PDF:** https://arxiv.org/pdf/2603.29741v1
+
+## 677. Spontaneous Functional Differentiation in Large Language Models: A Brain-Like Intelligence Economy
+
+- **ID:** `2603.29735` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Junjie Zhang, Zhen Shen, Gang Xiong, et al.
+- **abs:** https://arxiv.org/abs/2603.29735 · **PDF:** https://arxiv.org/pdf/2603.29735v1
+
+## 678. Perspective: Towards sustainable exploration of chemical spaces with machine learning
+
+- **ID:** `2604.00069` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Leonardo Medrano Sandonas, David Balcells, Anton Bochkarev, et al.
+- **abs:** https://arxiv.org/abs/2604.00069 · **PDF:** https://arxiv.org/pdf/2604.00069v1
+
+## 679. mlr3mbo: Bayesian Optimization in R
+
+- **ID:** `2603.29730` · **primary:** `stat.ML` · **Submitted:** 2026-03-31
+- **Authors:** Marc Becker, Lennart Schneider, Martin Binder, et al.
+- **abs:** https://arxiv.org/abs/2603.29730 · **PDF:** https://arxiv.org/pdf/2603.29730v1
+
+## 680. Unbounded Density Ratio Estimation and Its Application to Covariate Shift Adaptation
+
+- **ID:** `2603.29725` · **primary:** `stat.ML` · **Submitted:** 2026-03-31
+- **Authors:** Ren-Rui Liu, Jun Fan, Lei Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.29725 · **PDF:** https://arxiv.org/pdf/2603.29725v1
+
+## 681. Reinforced Reasoning for End-to-End Retrosynthetic Planning
+
+- **ID:** `2603.29723` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Chenyang Zuo, Siqi Fan, Yizhen Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.29723 · **PDF:** https://arxiv.org/pdf/2603.29723v1
+
+## 682. Nonnegative Matrix Factorization in the Component-Wise L1 Norm for Sparse Data
+
+- **ID:** `2603.29715` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Giovanni Seraghiti, Kévin Dubrulle, Arnaud Vandaele, et al.
+- **abs:** https://arxiv.org/abs/2603.29715 · **PDF:** https://arxiv.org/pdf/2603.29715v1
+
+## 683. Symphony for Medical Coding: A Next-Generation Agentic System for Scalable and Explainable Medical Coding
+
+- **ID:** `2603.29709` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Joakim Edin, Andreas Motzfeldt, Simon Flachs, et al.
+- **abs:** https://arxiv.org/abs/2603.29709 · **PDF:** https://arxiv.org/pdf/2603.29709v1
+
+## 684. Exploring the Impact of Skin Color on Skin Lesion Segmentation
+
+- **ID:** `2603.29694` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Kuniko Paxton, Medina Kapo, Amila Akagić, et al.
+- **abs:** https://arxiv.org/abs/2603.29694 · **PDF:** https://arxiv.org/pdf/2603.29694v1
+
+## 685. Measuring the metacognition of AI
+
+- **ID:** `2603.29693` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Richard Servajean, Philippe Servajean
+- **abs:** https://arxiv.org/abs/2603.29693 · **PDF:** https://arxiv.org/pdf/2603.29693v1
+
+## 686. A First Step Towards Even More Sparse Encodings of Probability Distributions
+
+- **ID:** `2603.29691` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Florian Andreas Marwitz, Tanya Braun, Ralf Möller
+- **abs:** https://arxiv.org/abs/2603.29691 · **PDF:** https://arxiv.org/pdf/2603.29691v1
+
+## 687. KEditVis: A Visual Analytics System for Knowledge Editing of Large Language Models
+
+- **ID:** `2603.29689` · **primary:** `cs.HC` · **Submitted:** 2026-03-31
+- **Authors:** Zhenning Chen, Hanbei Zhan, Yanwei Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.29689 · **PDF:** https://arxiv.org/pdf/2603.29689v1
+
+## 688. Beyond the Steeper Curve: AI-Mediated Metacognitive Decoupling and the Limits of the Dunning-Kruger Metaphor
+
+- **ID:** `2603.29681` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Christopher Koch
+- **abs:** https://arxiv.org/abs/2603.29681 · **PDF:** https://arxiv.org/pdf/2603.29681v1
+
+## 689. View-oriented Conversation Compiler for Agent Trace Analysis
+
+- **ID:** `2603.29678` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Lvmin Zhang, Maneesh Agrawala
+- **abs:** https://arxiv.org/abs/2603.29678 · **PDF:** https://arxiv.org/pdf/2603.29678v2
+
+## 690. Mind the Gap: A Framework for Assessing Pitfalls in Multimodal Active Learning
+
+- **ID:** `2603.29677` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Dustin Eisenhardt, Yunhee Jeong, Florian Buettner
+- **abs:** https://arxiv.org/abs/2603.29677 · **PDF:** https://arxiv.org/pdf/2603.29677v1
+
+## 691. A Comprehensive Information-Decomposition Analysis of Large Vision-Language Models
+
+- **ID:** `2603.29676` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Lixin Xiu, Xufang Luo, Hideki Nakayama
+- **abs:** https://arxiv.org/abs/2603.29676 · **PDF:** https://arxiv.org/pdf/2603.29676v1
+
+## 692. Near-Miss: Latent Policy Failure Detection in Agentic Workflows
+
+- **ID:** `2603.29665` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Ella Rabinovich, David Boaz, Naama Zwerdling, et al.
+- **abs:** https://arxiv.org/abs/2603.29665 · **PDF:** https://arxiv.org/pdf/2603.29665v1
+
+## 693. Agenda-based Narrative Extraction: Steering Pathfinding Algorithms with Large Language Models
+
+- **ID:** `2603.29661` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Brian Felipe Keith-Norambuena, Carolina Inés Rojas-Córdova, Claudio Juvenal Meneses-Villegas, et al.
+- **abs:** https://arxiv.org/abs/2603.29661 · **PDF:** https://arxiv.org/pdf/2603.29661v1
+
+## 694. 6GAgentGym: Tool Use, Data Synthesis, and Agentic Learning for Network Management
+
+- **ID:** `2603.29656` · **primary:** `cs.NI` · **Submitted:** 2026-03-31
+- **Authors:** Jiao Chen, Jianhua Tang, Xiaotong Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.29656 · **PDF:** https://arxiv.org/pdf/2603.29656v1
+
+## 695. Concept frustration: Aligning human concepts and machine representations
+
+- **ID:** `2603.29654` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Enrico Parisini, Christopher J. Soelistyo, Ahab Isaac, et al.
+- **abs:** https://arxiv.org/abs/2603.29654 · **PDF:** https://arxiv.org/pdf/2603.29654v1
+
+## 696. Semantic Interaction for Narrative Map Sensemaking: An Insight-based Evaluation
+
+- **ID:** `2603.29651` · **primary:** `cs.HC` · **Submitted:** 2026-03-31
+- **Authors:** Brian Felipe Keith-Norambuena, Fausto German, Eric Krokos, et al.
+- **abs:** https://arxiv.org/abs/2603.29651 · **PDF:** https://arxiv.org/pdf/2603.29651v1
+
+## 697. Disentangled Graph Prompting for Out-Of-Distribution Detection
+
+- **ID:** `2603.29644` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Cheng Yang, Yu Hao, Qi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.29644 · **PDF:** https://arxiv.org/pdf/2603.29644v1
+
+## 698. Optimizing Donor Outreach for Blood Collection Sessions: A Scalable Decision Support Framework
+
+- **ID:** `2603.29643` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** André Carneiro, Pedro T. Monteiro, Rui Henriques
+- **abs:** https://arxiv.org/abs/2603.29643 · **PDF:** https://arxiv.org/pdf/2603.29643v1
+
+## 699. ASI-Evolve: AI Accelerates AI
+
+- **ID:** `2603.29640` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Weixian Xu, Tiantian Mi, Yixiu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.29640 · **PDF:** https://arxiv.org/pdf/2603.29640v1
+
+## 700. Temporal Memory for Resource-Constrained Agents: Continual Learning via Stochastic Compress-Add-Smooth
+
+- **ID:** `2604.00067` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Michael Chertkov
+- **abs:** https://arxiv.org/abs/2604.00067 · **PDF:** https://arxiv.org/pdf/2604.00067v1
+
+## 701. MacTok: Robust Continuous Tokenization for Image Generation
+
+- **ID:** `2603.29634` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Hengyu Zeng, Xin Gao, Guanghao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.29634 · **PDF:** https://arxiv.org/pdf/2603.29634v1
+
+## 702. An Empirical Study of Multi-Agent Collaboration for Automated Research
+
+- **ID:** `2603.29632` · **primary:** `cs.MA` · **Submitted:** 2026-03-31
+- **Authors:** Yang Shen, Zhenyi Yi, Ziyi Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.29632 · **PDF:** https://arxiv.org/pdf/2603.29632v1
+
+## 703. Convergent Representations of Linguistic Constructions in Human and Artificial Neural Systems
+
+- **ID:** `2603.29617` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-31
+- **Authors:** Pegah Ramezani, Thomas Kinfe, Andreas Maier, et al.
+- **abs:** https://arxiv.org/abs/2603.29617 · **PDF:** https://arxiv.org/pdf/2603.29617v1
+
+## 704. Central limit theorems for the outputs of fully convolutional neural networks with time series input
+
+- **ID:** `2603.29612` · **primary:** `stat.ME` · **Submitted:** 2026-03-31
+- **Authors:** Annika Betken, Giorgio Micali, Johannes Schmidt-Hieber
+- **abs:** https://arxiv.org/abs/2603.29612 · **PDF:** https://arxiv.org/pdf/2603.29612v1
+
+## 705. Evolution Strategies for Deep RL pretraining
+
+- **ID:** `2604.00066` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Adrian Martínez, Ananya Gupta, Hanka Goralija, et al.
+- **abs:** https://arxiv.org/abs/2604.00066 · **PDF:** https://arxiv.org/pdf/2604.00066v1
+
+## 706. Learning Diagnostic Reasoning for Decision Support in Toxicology
+
+- **ID:** `2603.29608` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Nico Oberländer, David Bani-Harouni, Tobias Zellner, et al.
+- **abs:** https://arxiv.org/abs/2603.29608 · **PDF:** https://arxiv.org/pdf/2603.29608v1
+
+## 707. Generating Key Postures of Bharatanatyam Adavus with Pose Estimation
+
+- **ID:** `2603.29570` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Jagadish Kashinath Kamble, Jayanta Mukhopadhyay, Debaditya Roy, et al.
+- **abs:** https://arxiv.org/abs/2603.29570 · **PDF:** https://arxiv.org/pdf/2603.29570v1
+
+## 708. The Geometry of Polynomial Group Convolutional Neural Networks
+
+- **ID:** `2603.29566` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Yacoub Hendi, Daniel Persson, Magdalena Larfors
+- **abs:** https://arxiv.org/abs/2603.29566 · **PDF:** https://arxiv.org/pdf/2603.29566v1
+
+## 709. Genetic algorithms for multi-omic feature selection: a comparative study in cancer survival analysis
+
+- **ID:** `2604.00065` · **primary:** `q-bio.GN` · **Submitted:** 2026-03-31
+- **Authors:** Luca Cattelani, Vittorio Fortino
+- **abs:** https://arxiv.org/abs/2604.00065 · **PDF:** https://arxiv.org/pdf/2604.00065v1
+
+## 710. When Can We Trust LLM Graders? Calibrating Confidence for Automated Assessment
+
+- **ID:** `2603.29559` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Robinson Ferrer, Damla Turgut, Zhongzhou Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.29559 · **PDF:** https://arxiv.org/pdf/2603.29559v1
+
+## 711. FlowPIE: Test-Time Scientific Idea Evolution with Flow-Guided Literature Exploration
+
+- **ID:** `2603.29557` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Qiyao Wang, Hongbo Wang, Longze Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.29557 · **PDF:** https://arxiv.org/pdf/2603.29557v1
+
+## 712. Total Variation Guarantees for Sampling with Stochastic Localization
+
+- **ID:** `2603.29555` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Jakob Kellermann
+- **abs:** https://arxiv.org/abs/2603.29555 · **PDF:** https://arxiv.org/pdf/2603.29555v1
+
+## 713. Capturing Multivariate Dependencies of EV Charging Events: From Parametric Copulas to Neural Density Estimation
+
+- **ID:** `2603.29554` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Martin Výboh, Gabriela Grmanová
+- **abs:** https://arxiv.org/abs/2603.29554 · **PDF:** https://arxiv.org/pdf/2603.29554v1
+
+## 714. Bringing Up a Bilingual BabyLM: Investigating Multilingual Language Acquisition Using Small-Scale Models
+
+- **ID:** `2603.29552` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Linda Zeng, Steven Y. Feng, Michael C. Frank
+- **abs:** https://arxiv.org/abs/2603.29552 · **PDF:** https://arxiv.org/pdf/2603.29552v1
+
+## 715. Reducing Complexity for Quantum Approaches in Train Load Optimization
+
+- **ID:** `2603.29543` · **primary:** `quant-ph` · **Submitted:** 2026-03-31
+- **Authors:** Zhijie Tang, Albert Nieto-Morales, Arit Kumar Bishwas
+- **abs:** https://arxiv.org/abs/2603.29543 · **PDF:** https://arxiv.org/pdf/2603.29543v1
+
+## 716. Can LLM Agents Identify Spoken Dialects like a Linguist?
+
+- **ID:** `2603.29541` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Tobias Bystrich, Lukas Hamm, Maria Hassan, et al.
+- **abs:** https://arxiv.org/abs/2603.29541 · **PDF:** https://arxiv.org/pdf/2603.29541v1
+
+## 717. Mean Masked Autoencoder with Flow-Mixing for Encrypted Traffic Classification
+
+- **ID:** `2603.29537` · **primary:** `cs.CR` · **Submitted:** 2026-03-31
+- **Authors:** Xiao Liu, Xiaowei Fu, Fuxiang Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.29537 · **PDF:** https://arxiv.org/pdf/2603.29537v1
+
+## 718. Quantization with Unified Adaptive Distillation to enable multi-LoRA based one-for-all Generative Vision Models on edge
+
+- **ID:** `2603.29535` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Sowmya Vajrala, Aakash Parmar, Prasanna R, et al.
+- **abs:** https://arxiv.org/abs/2603.29535 · **PDF:** https://arxiv.org/pdf/2603.29535v1
+
+## 719. Learning Surrogate LPV State-Space Models with Uncertainty Quantification
+
+- **ID:** `2603.29532` · **primary:** `eess.SY` · **Submitted:** 2026-03-31
+- **Authors:** E. Javier Olucha, Valentin Preda, Amritam Das, et al.
+- **abs:** https://arxiv.org/abs/2603.29532 · **PDF:** https://arxiv.org/pdf/2603.29532v1
+
+## 720. Sampling at intermediate temperatures is optimal for training large language models in protein structure prediction
+
+- **ID:** `2603.29529` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-03-31
+- **Authors:** L. Ghiringhelli, A. Zambon, G. Tiana
+- **abs:** https://arxiv.org/abs/2603.29529 · **PDF:** https://arxiv.org/pdf/2603.29529v1
+
+## 721. Baby Scale: Investigating Models Trained on Individual Children's Language Input
+
+- **ID:** `2603.29522` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Steven Y. Feng, Alvin W. M. Tan, Michael C. Frank
+- **abs:** https://arxiv.org/abs/2603.29522 · **PDF:** https://arxiv.org/pdf/2603.29522v1
+
+## 722. TrafficMoE: Heterogeneity-aware Mixture of Experts for Encrypted Traffic Classification
+
+- **ID:** `2603.29520` · **primary:** `cs.CR` · **Submitted:** 2026-03-31
+- **Authors:** Qing He, Xiaowei Fu, Lei Zhang
+- **abs:** https://arxiv.org/abs/2603.29520 · **PDF:** https://arxiv.org/pdf/2603.29520v1
+
+## 723. Impact of enriched meaning representations for language generation in dialogue tasks: A comprehensive exploration of the relevance of tasks, corpora and metrics
+
+- **ID:** `2603.29518` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Alain Vázquez, Maria Inés Torres
+- **abs:** https://arxiv.org/abs/2603.29518 · **PDF:** https://arxiv.org/pdf/2603.29518v1
+
+## 724. LLM Probe: Evaluating LLMs for Low-Resource Languages
+
+- **ID:** `2603.29517` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Hailay Kidu Teklehaymanot, Gebrearegawi Gebremariam, Wolfgang Nejdl
+- **abs:** https://arxiv.org/abs/2603.29517 · **PDF:** https://arxiv.org/pdf/2603.29517v1
+
+## 725. Variational Graph Neural Networks for Uncertainty Quantification in Inverse Problems
+
+- **ID:** `2603.29515` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** David Gonzalez, Alba Muixi, Beatriz Moya, et al.
+- **abs:** https://arxiv.org/abs/2603.29515 · **PDF:** https://arxiv.org/pdf/2603.29515v1
+
+## 726. Target-Aligned Reinforcement Learning
+
+- **ID:** `2603.29501` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Leonard S. Pleiss, James Harrison, Maximilian Schiffer
+- **abs:** https://arxiv.org/abs/2603.29501 · **PDF:** https://arxiv.org/pdf/2603.29501v1
+
+## 727. Learning to Generate Formally Verifiable Step-by-Step Logic Reasoning via Structured Formal Intermediaries
+
+- **ID:** `2603.29500` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Luoxin Chen, Yichi Zhou, Huishuai Zhang
+- **abs:** https://arxiv.org/abs/2603.29500 · **PDF:** https://arxiv.org/pdf/2603.29500v1
+
+## 728. Model Predictive Path Integral PID Control for Learning-Based Path Following
+
+- **ID:** `2603.29499` · **primary:** `eess.SY` · **Submitted:** 2026-03-31
+- **Authors:** Teruki Kato, Koshi Oishi, Seigo Ito
+- **abs:** https://arxiv.org/abs/2603.29499 · **PDF:** https://arxiv.org/pdf/2603.29499v1
+
+## 729. Distilling Human-Aligned Privacy Sensitivity Assessment from Large Language Models
+
+- **ID:** `2603.29497` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Gabriel Loiseau, Damien Sileo, Damien Riquet, et al.
+- **abs:** https://arxiv.org/abs/2603.29497 · **PDF:** https://arxiv.org/pdf/2603.29497v1
+
+## 730. Metriplector: From Field Theory to Neural Architecture
+
+- **ID:** `2603.29496` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Dan Oprisa, Peter Toth
+- **abs:** https://arxiv.org/abs/2603.29496 · **PDF:** https://arxiv.org/pdf/2603.29496v1
+
+## 731. MemFactory: Unified Inference & Training Framework for Agent Memory
+
+- **ID:** `2603.29493` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Ziliang Guo, Ziheng Li, Bo Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.29493 · **PDF:** https://arxiv.org/pdf/2603.29493v3
+
+## 732. Calibrated Confidence Expression for Radiology Report Generation
+
+- **ID:** `2603.29492` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** David Bani-Harouni, Chantal Pellegrini, Julian Lüers, et al.
+- **abs:** https://arxiv.org/abs/2603.29492 · **PDF:** https://arxiv.org/pdf/2603.29492v1
+
+## 733. Structural Compactness as a Complementary Criterion for Explanation Quality
+
+- **ID:** `2603.29491` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Mohammad Mahdi Mesgari, Jackie Ma, Wojciech Samek, et al.
+- **abs:** https://arxiv.org/abs/2603.29491 · **PDF:** https://arxiv.org/pdf/2603.29491v1
+
+## 734. Why not to use Cosine Similarity between Label Representations
+
+- **ID:** `2603.29488` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Beatrix M. G. Nielsen
+- **abs:** https://arxiv.org/abs/2603.29488 · **PDF:** https://arxiv.org/pdf/2603.29488v1
+
+## 735. Survival In-Context: Prior-fitted In-context Learning Tabular Foundation Model for Survival Analysis
+
+- **ID:** `2603.29475` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Dmitrii Seletkov, Paul Hager, Rickmer Braren, et al.
+- **abs:** https://arxiv.org/abs/2603.29475 · **PDF:** https://arxiv.org/pdf/2603.29475v1
+
+## 736. From Big Data to Fast Data: Towards High-Quality Datasets for Machine Learning Applications from Closed-Loop Data Collection
+
+- **ID:** `2603.29474` · **primary:** `eess.SY` · **Submitted:** 2026-03-31
+- **Authors:** Philipp Reis, Jacqueline Henle, Stefan Otten, et al.
+- **abs:** https://arxiv.org/abs/2603.29474 · **PDF:** https://arxiv.org/pdf/2603.29474v1
+
+## 737. iPoster: Content-Aware Layout Generation for Interactive Poster Design via Graph-Enhanced Diffusion Models
+
+- **ID:** `2603.29469` · **primary:** `cs.HC` · **Submitted:** 2026-03-31
+- **Authors:** Xudong Zhou, Jinyuan Liang, Qiuyi Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.29469 · **PDF:** https://arxiv.org/pdf/2603.29469v1
+
+## 738. M-MiniGPT4: Multilingual VLLM Alignment via Translated Data
+
+- **ID:** `2603.29467` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Seung Hun Han, Youssef Mohamed, Mohamed Elhoseiny
+- **abs:** https://arxiv.org/abs/2603.29467 · **PDF:** https://arxiv.org/pdf/2603.29467v1
+
+## 739. An Isotropic Approach to Efficient Uncertainty Quantification with Gradient Norms
+
+- **ID:** `2603.29466` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Nils Grünefeld, Jes Frellsen, Christian Hardmeier
+- **abs:** https://arxiv.org/abs/2603.29466 · **PDF:** https://arxiv.org/pdf/2603.29466v1
+
+## 740. Forecast collapse of transformer-based models under squared loss in financial time series
+
+- **ID:** `2604.00064` · **primary:** `stat.ML` · **Submitted:** 2026-03-31
+- **Authors:** Pierre Andreoletti
+- **abs:** https://arxiv.org/abs/2604.00064 · **PDF:** https://arxiv.org/pdf/2604.00064v1
+
+## 741. Authorship Impersonation via LLM Prompting does not Evade Authorship Verification Methods
+
+- **ID:** `2603.29454` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Baoyi Zeng, Andrea Nini
+- **abs:** https://arxiv.org/abs/2603.29454 · **PDF:** https://arxiv.org/pdf/2603.29454v1
+
+## 742. Few-shot Writer Adaptation via Multimodal In-Context Learning
+
+- **ID:** `2603.29450` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Tom Simon, Stephane Nicolas, Pierrick Tranouez, et al.
+- **abs:** https://arxiv.org/abs/2603.29450 · **PDF:** https://arxiv.org/pdf/2603.29450v1
+
+## 743. NeoNet: An End-to-End 3D MRI-Based Deep Learning Framework for Non-Invasive Prediction of Perineural Invasion via Generation-Driven Classification
+
+- **ID:** `2603.29449` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Youngung Han, Minkyung Cha, Kyeonghun Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.29449 · **PDF:** https://arxiv.org/pdf/2603.29449v1
+
+## 744. mtslearn: Machine Learning in Python for Medical Time Series
+
+- **ID:** `2603.29432` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Zhongheng Jiang, Yuechao Zhao, Donglin Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.29432 · **PDF:** https://arxiv.org/pdf/2603.29432v1
+
+## 745. CounselReflect: A Toolkit for Auditing Mental-Health Dialogues
+
+- **ID:** `2603.29429` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Yahan Li, Chaohao Du, Zeyang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.29429 · **PDF:** https://arxiv.org/pdf/2603.29429v1
+
+## 746. Multi-AUV Cooperative Target Tracking Based on Supervised Diffusion-Aided Multi-Agent Reinforcement Learning
+
+- **ID:** `2603.29426` · **primary:** `cs.NI` · **Submitted:** 2026-03-31
+- **Authors:** Jiaao Ma, Chuan Lin, Guangjie Han, et al.
+- **abs:** https://arxiv.org/abs/2603.29426 · **PDF:** https://arxiv.org/pdf/2603.29426v1
+
+## 747. RAAP: Retrieval-Augmented Affordance Prediction with Cross-Image Action Alignment
+
+- **ID:** `2603.29419` · **primary:** `cs.RO` · **Submitted:** 2026-03-31
+- **Authors:** Qiyuan Zhuang, He-Yang Xu, Yijun Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.29419 · **PDF:** https://arxiv.org/pdf/2603.29419v1
+
+## 748. Adversarial Prompt Injection Attack on Multimodal Large Language Models
+
+- **ID:** `2603.29418` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Meiwen Ding, Song Xia, Chenqi Kong, et al.
+- **abs:** https://arxiv.org/abs/2603.29418 · **PDF:** https://arxiv.org/pdf/2603.29418v1
+
+## 749. AGFT: Alignment-Guided Fine-Tuning for Zero-Shot Adversarial Robustness of Vision-Language Models
+
+- **ID:** `2603.29410` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Yubo Cui, Xianchao Guan, Zijun Xiong, et al.
+- **abs:** https://arxiv.org/abs/2603.29410 · **PDF:** https://arxiv.org/pdf/2603.29410v1
+
+## 750. Hybrid Quantum-Classical Spatiotemporal Forecasting for 3D Cloud Fields
+
+- **ID:** `2603.29407` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Fu Wang, Qifeng Lu, Xinyu Long, et al.
+- **abs:** https://arxiv.org/abs/2603.29407 · **PDF:** https://arxiv.org/pdf/2603.29407v1
+
+## 751. PRISM: PRIor from corpus Statistics for topic Modeling
+
+- **ID:** `2603.29406` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Tal Ishon, Yoav Goldberg, Uri Shaham
+- **abs:** https://arxiv.org/abs/2603.29406 · **PDF:** https://arxiv.org/pdf/2603.29406v1
+
+## 752. Hallucination-aware intermediate representation edit in large vision-language models
+
+- **ID:** `2603.29405` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Wei Suo, Hanzu Zhang, Lijun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.29405 · **PDF:** https://arxiv.org/pdf/2603.29405v1
+
+## 753. Security in LLM-as-a-Judge: A Comprehensive SoK
+
+- **ID:** `2603.29403` · **primary:** `cs.CR` · **Submitted:** 2026-03-31
+- **Authors:** Aiman Almasoud, Antony Anju, Marco Arazzi, et al.
+- **abs:** https://arxiv.org/abs/2603.29403 · **PDF:** https://arxiv.org/pdf/2603.29403v1
+
+## 754. ELT-Bench-Verified: Benchmark Quality Issues Underestimate AI Agent Capabilities
+
+- **ID:** `2603.29399` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Christopher Zanoli, Andrea Giovannini, Tengjun Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.29399 · **PDF:** https://arxiv.org/pdf/2603.29399v2
+
+## 755. Is my model perplexed for the right reason? Contrasting LLMs' Benchmark Behavior with Token-Level Perplexity
+
+- **ID:** `2603.29396` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Zoë Prins, Samuele Punzo, Frank Wildenburg, et al.
+- **abs:** https://arxiv.org/abs/2603.29396 · **PDF:** https://arxiv.org/pdf/2603.29396v1
+
+## 756. Extend3D: Town-Scale 3D Generation
+
+- **ID:** `2603.29387` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Seungwoo Yoon, Jinmo Kim, Jaesik Park
+- **abs:** https://arxiv.org/abs/2603.29387 · **PDF:** https://arxiv.org/pdf/2603.29387v1
+
+## 757. PromptForge-350k: A Large-Scale Dataset and Contrastive Framework for Prompt-Based AI Image Forgery Localization
+
+- **ID:** `2603.29386` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Jianpeng Wang, Haoyu Wang, Baoying Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.29386 · **PDF:** https://arxiv.org/pdf/2603.29386v1
+
+## 758. Causality-inspired Federated Learning for Dynamic Spatio-Temporal Graphs
+
+- **ID:** `2603.29384` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Yuxuan Liu, Wenchao Xu, Haozhao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.29384 · **PDF:** https://arxiv.org/pdf/2603.29384v1
+
+## 759. Deep Learning-Assisted Improved Differential Fault Attacks on Lightweight Stream Ciphers
+
+- **ID:** `2603.29382` · **primary:** `cs.CR` · **Submitted:** 2026-03-31
+- **Authors:** Kok Ping Lim, Dongyang Jia, Iftekhar Salam
+- **abs:** https://arxiv.org/abs/2603.29382 · **PDF:** https://arxiv.org/pdf/2603.29382v1
+
+## 760. Finite-time analysis of Multi-timescale Stochastic Optimization Algorithms
+
+- **ID:** `2603.29380` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Kaustubh Kartikey, Shalabh Bhatnagar
+- **abs:** https://arxiv.org/abs/2603.29380 · **PDF:** https://arxiv.org/pdf/2603.29380v1
+
+## 761. Deep Learning-Based Anomaly Detection in Spacecraft Telemetry on Edge Devices
+
+- **ID:** `2603.29375` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Christopher Goetze, Tim Schlippe, Daniel Lakey
+- **abs:** https://arxiv.org/abs/2603.29375 · **PDF:** https://arxiv.org/pdf/2603.29375v1
+
+## 762. Beyond Idealized Patients: Evaluating LLMs under Challenging Patient Behaviors in Medical Consultations
+
+- **ID:** `2603.29373` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Yahan Li, Xinyi Jie, Wanjia Ruan, et al.
+- **abs:** https://arxiv.org/abs/2603.29373 · **PDF:** https://arxiv.org/pdf/2603.29373v1
+
+## 763. AP-DRL: A Synergistic Algorithm-Hardware Framework for Automatic Task Partitioning of Deep Reinforcement Learning on Versal ACAP
+
+- **ID:** `2603.29369` · **primary:** `cs.AR` · **Submitted:** 2026-03-31
+- **Authors:** Enlai Li, Zhe Lin, Sharad Sinha, et al.
+- **abs:** https://arxiv.org/abs/2603.29369 · **PDF:** https://arxiv.org/pdf/2603.29369v1
+
+## 764. AI-Generated Prior Authorization Letters: Strong Clinical Content, Weak Administrative Scaffolding
+
+- **ID:** `2603.29366` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Moiz Sadiq Awan, Maryam Raza
+- **abs:** https://arxiv.org/abs/2603.29366 · **PDF:** https://arxiv.org/pdf/2603.29366v1
+
+## 765. Rigorous Explanations for Tree Ensembles
+
+- **ID:** `2603.29361` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Yacine Izza, Alexey Ignatiev, Xuanxiang Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.29361 · **PDF:** https://arxiv.org/pdf/2603.29361v1
+
+## 766. BenchScope: How Many Independent Signals Does Your Benchmark Provide?
+
+- **ID:** `2603.29357` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Tommy Sha, Stella Zhao
+- **abs:** https://arxiv.org/abs/2603.29357 · **PDF:** https://arxiv.org/pdf/2603.29357v1
+
+## 767. CIPHER: Counterfeit Image Pattern High-level Examination via Representation
+
+- **ID:** `2603.29356` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Kyeonghun Kim, Youngung Han, Seoyoung Ju, et al.
+- **abs:** https://arxiv.org/abs/2603.29356 · **PDF:** https://arxiv.org/pdf/2603.29356v1
+
+## 768. Nomad: Autonomous Exploration and Discovery
+
+- **ID:** `2603.29353` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Bokang Jia, Samta Kamboj, Satheesh Katipomu, et al.
+- **abs:** https://arxiv.org/abs/2603.29353 · **PDF:** https://arxiv.org/pdf/2603.29353v2
+
+## 769. Developing a Guideline for the Labovian-Structural Analysis of Oral Narratives in Japanese
+
+- **ID:** `2603.29347` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Amane Watahiki, Tomoki Doi, Akari Kikuchi, et al.
+- **abs:** https://arxiv.org/abs/2603.29347 · **PDF:** https://arxiv.org/pdf/2603.29347v1
+
+## 770. L-ReLF: A Framework for Lexical Dataset Creation
+
+- **ID:** `2603.29346` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Anass Sedrati, Mounir Afifi, Reda Benkhadra
+- **abs:** https://arxiv.org/abs/2603.29346 · **PDF:** https://arxiv.org/pdf/2603.29346v1
+
+## 771. Open Machine Translation for Esperanto
+
+- **ID:** `2603.29345` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Ona de Gibert, Lluís de Gibert
+- **abs:** https://arxiv.org/abs/2603.29345 · **PDF:** https://arxiv.org/pdf/2603.29345v1
+
+## 772. CADEL: A Corpus of Administrative Web Documents for Japanese Entity Linking
+
+- **ID:** `2603.29336` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Shohei Higashiyama, Masao Ideuchi, Masao Utiyama
+- **abs:** https://arxiv.org/abs/2603.29336 · **PDF:** https://arxiv.org/pdf/2603.29336v1
+
+## 773. Scaling Whole-Body Human Musculoskeletal Behavior Emulation for Specificity and Diversity
+
+- **ID:** `2603.29332` · **primary:** `cs.RO` · **Submitted:** 2026-03-31
+- **Authors:** Yunyue Wei, Chenhui Zuo, Shanning Zhuang, et al.
+- **abs:** https://arxiv.org/abs/2603.29332 · **PDF:** https://arxiv.org/pdf/2603.29332v1
+
+## 774. Beyond Corner Patches: Semantics-Aware Backdoor Attack in Federated Learning
+
+- **ID:** `2603.29328` · **primary:** `cs.CR` · **Submitted:** 2026-03-31
+- **Authors:** Kavindu Herath, Joshua Zhao, Saurabh Bagchi
+- **abs:** https://arxiv.org/abs/2603.29328 · **PDF:** https://arxiv.org/pdf/2603.29328v1
+
+## 775. Real-Time Band-Grouped Vocal Denoising Using Sigmoid-Driven Ideal Ratio Masking
+
+- **ID:** `2603.29326` · **primary:** `cs.SD` · **Submitted:** 2026-03-31
+- **Authors:** Daniel Williams
+- **abs:** https://arxiv.org/abs/2603.29326 · **PDF:** https://arxiv.org/pdf/2603.29326v1
+
+## 776. PSPA-Bench: A Personalized Benchmark for Smartphone GUI Agent
+
+- **ID:** `2603.29318` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Hongyi Nie, Xunyuan Liu, Yudong Bai, et al.
+- **abs:** https://arxiv.org/abs/2603.29318 · **PDF:** https://arxiv.org/pdf/2603.29318v1
+
+## 777. IMPASTO: Integrating Model-Based Planning with Learned Dynamics Models for Robotic Oil Painting Reproduction
+
+- **ID:** `2603.29315` · **primary:** `cs.RO` · **Submitted:** 2026-03-31
+- **Authors:** Yingke Wang, Hao Li, Yifeng Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.29315 · **PDF:** https://arxiv.org/pdf/2603.29315v1
+
+## 778. LGFNet: Local-Global Fusion Network with Fidelity Gap Delta Learning for Multi-Source Aerodynamics
+
+- **ID:** `2603.29303` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Qinye Zhu, Yu Xiang, Jun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.29303 · **PDF:** https://arxiv.org/pdf/2603.29303v1
+
+## 779. Self-Improving Code Generation via Semantic Entropy and Behavioral Consensus
+
+- **ID:** `2603.29292` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Huan Zhang, Wei Cheng, Wei Hu
+- **abs:** https://arxiv.org/abs/2603.29292 · **PDF:** https://arxiv.org/pdf/2603.29292v1
+
+## 780. MELT: Improve Composed Image Retrieval via the Modification Frequentation-Rarity Balance Network
+
+- **ID:** `2603.29291` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Guozhi Qiu, Zhiwei Chen, Zixu Li, et al.
+- **abs:** https://arxiv.org/abs/2603.29291 · **PDF:** https://arxiv.org/pdf/2603.29291v1
+
+## 781. Downsides of Smartness Across Edge-Cloud Continuum in Modern Industry
+
+- **ID:** `2603.29289` · **primary:** `cs.CR` · **Submitted:** 2026-03-31
+- **Authors:** Akhil Gupta Chigullapally, Sharvan Vittala, Razin Farhan Hussian, et al.
+- **abs:** https://arxiv.org/abs/2603.29289 · **PDF:** https://arxiv.org/pdf/2603.29289v1
+
+## 782. Sima AIunty: Caste Audit in LLM-Driven Matchmaking
+
+- **ID:** `2603.29288` · **primary:** `cs.CY` · **Submitted:** 2026-03-31
+- **Authors:** Atharva Naik, Shounok Kar, Varnika Sharma, et al.
+- **abs:** https://arxiv.org/abs/2603.29288 · **PDF:** https://arxiv.org/pdf/2603.29288v1
+
+## 783. PRISM: A Multi-View Multi-Capability Retail Video Dataset for Embodied Vision-Language Models
+
+- **ID:** `2603.29281` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Amirreza Rouhi, Parikshit Sakurikar, Satya Sai Reddy, et al.
+- **abs:** https://arxiv.org/abs/2603.29281 · **PDF:** https://arxiv.org/pdf/2603.29281v1
+
+## 784. Scaled Gradient Descent for Ill-Conditioned Low-Rank Matrix Recovery with Optimal Sampling Complexity
+
+- **ID:** `2604.00060` · **primary:** `stat.ML` · **Submitted:** 2026-03-31
+- **Authors:** Zhenxuan Li, Meng Huang
+- **abs:** https://arxiv.org/abs/2604.00060 · **PDF:** https://arxiv.org/pdf/2604.00060v1
+
+## 785. From Physics to Surrogate Intelligence: A Unified Electro-Thermo-Optimization Framework for TSV Networks
+
+- **ID:** `2603.29268` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Mohamed Gharib, Leonid Popryho, Inna Partin-Vaisband
+- **abs:** https://arxiv.org/abs/2603.29268 · **PDF:** https://arxiv.org/pdf/2603.29268v1
+
+## 786. Lie Generator Networks for Nonlinear Partial Differential Equations
+
+- **ID:** `2603.29264` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Shafayeth Jamil, Rehan Kapadia
+- **abs:** https://arxiv.org/abs/2603.29264 · **PDF:** https://arxiv.org/pdf/2603.29264v2
+
+## 787. Grokking From Abstraction to Intelligence
+
+- **ID:** `2603.29262` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Junjie Zhang, Zhen Shen, Gang Xiong, et al.
+- **abs:** https://arxiv.org/abs/2603.29262 · **PDF:** https://arxiv.org/pdf/2603.29262v1
+
+## 788. Monodense Deep Neural Model for Determining Item Price Elasticity
+
+- **ID:** `2603.29261` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Lakshya Garg, Sai Yaswanth, Deep Narayan Mishra, et al.
+- **abs:** https://arxiv.org/abs/2603.29261 · **PDF:** https://arxiv.org/pdf/2603.29261v1
+
+## 789. Aligning Multimodal Sequential Recommendations via Robust Direct Preference Optimization with Sparse MoE
+
+- **ID:** `2603.29259` · **primary:** `cs.IR` · **Submitted:** 2026-03-31
+- **Authors:** Hejin Huang, Jusheng Zhang, Kaitong Cai, et al.
+- **abs:** https://arxiv.org/abs/2603.29259 · **PDF:** https://arxiv.org/pdf/2603.29259v1
+
+## 790. Omni-NegCLIP: Enhancing CLIP with Front-Layer Contrastive Fine-Tuning for Comprehensive Negation Understanding
+
+- **ID:** `2603.29258` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Jingqi Xu
+- **abs:** https://arxiv.org/abs/2603.29258 · **PDF:** https://arxiv.org/pdf/2603.29258v1
+
+## 791. Real-Time Surrogate Modeling for Fast Transient Prediction in Inverter-Based Microgrids Using CNN and LightGBM
+
+- **ID:** `2603.29255` · **primary:** `eess.SY` · **Submitted:** 2026-03-31
+- **Authors:** Osasumwen Cedric Ogiesoba-Eguakun, Kaveh Ashenayi, Suman Rath
+- **abs:** https://arxiv.org/abs/2603.29255 · **PDF:** https://arxiv.org/pdf/2603.29255v1
+
+## 792. Scaling the Long Video Understanding of Multimodal Large Language Models via Visual Memory Mechanism
+
+- **ID:** `2603.29252` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Tao Chen, Kun Zhang, Qiong Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.29252 · **PDF:** https://arxiv.org/pdf/2603.29252v1
+
+## 793. GenoBERT: A Language Model for Accurate Genotype Imputation
+
+- **ID:** `2604.00058` · **primary:** `q-bio.GN` · **Submitted:** 2026-03-31
+- **Authors:** Lei Huang, Chuan Qiu, Kuan-Jui Su, et al.
+- **abs:** https://arxiv.org/abs/2604.00058 · **PDF:** https://arxiv.org/pdf/2604.00058v1
+
+## 794. MemRerank: Preference Memory for Personalized Product Reranking
+
+- **ID:** `2603.29247` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Zhiyuan Peng, Xuyang Wu, Huaixiao Tou, et al.
+- **abs:** https://arxiv.org/abs/2603.29247 · **PDF:** https://arxiv.org/pdf/2603.29247v2
+
+## 795. The Thiomi Dataset: A Large-Scale Multimodal Corpus for Low-Resource African Languages
+
+- **ID:** `2603.29244` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Hillary Mutisya, John Mugane, Gavin Nyamboga, et al.
+- **abs:** https://arxiv.org/abs/2603.29244 · **PDF:** https://arxiv.org/pdf/2603.29244v1
+
+## 796. Stochastic Dimension Implicit Functional Projections for Exact Integral Conservation in High-Dimensional PINNs
+
+- **ID:** `2603.29237` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Zhangyong Liang
+- **abs:** https://arxiv.org/abs/2603.29237 · **PDF:** https://arxiv.org/pdf/2603.29237v1
+
+## 797. Robust and Consistent Ski Rental with Distributional Advice
+
+- **ID:** `2603.29233` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Jihwan Kim, Chenglin Fan
+- **abs:** https://arxiv.org/abs/2603.29233 · **PDF:** https://arxiv.org/pdf/2603.29233v1
+
+## 798. Long-Document QA with Chain-of-Structured-Thought and Fine-Tuned SLMs
+
+- **ID:** `2603.29232` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Zhuowen Liang, Xiaotian Lin, Zhengxuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.29232 · **PDF:** https://arxiv.org/pdf/2603.29232v1
+
+## 799. Beyond pass@1: A Reliability Science Framework for Long-Horizon LLM Agents
+
+- **ID:** `2603.29231` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Aaditya Khanal, Yangyang Tao, Junxiu Zhou
+- **abs:** https://arxiv.org/abs/2603.29231 · **PDF:** https://arxiv.org/pdf/2603.29231v1
+
+## 800. Derived Fields Preserve Fine-Scale Detail in Budgeted Neural Simulators
+
+- **ID:** `2603.29224` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Wenshuo Wang, Fan Zhang
+- **abs:** https://arxiv.org/abs/2603.29224 · **PDF:** https://arxiv.org/pdf/2603.29224v1
+
+## 801. SiPaKosa: A Comprehensive Corpus of Canonical and Classical Buddhist Texts in Sinhala and Pali
+
+- **ID:** `2603.29221` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Ranidu Gurusinghe, Nevidu Jayatilleke
+- **abs:** https://arxiv.org/abs/2603.29221 · **PDF:** https://arxiv.org/pdf/2603.29221v1
+
+## 802. SyriSign: A Parallel Corpus for Arabic Text to Syrian Arabic Sign Language Translation
+
+- **ID:** `2603.29219` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Mohammad Amer Khalil, Raghad Nahas, Ahmad Nassar, et al.
+- **abs:** https://arxiv.org/abs/2603.29219 · **PDF:** https://arxiv.org/pdf/2603.29219v1
+
+## 803. Advancing LLM-based phoneme-to-grapheme for multilingual speech recognition
+
+- **ID:** `2603.29217` · **primary:** `eess.AS` · **Submitted:** 2026-03-31
+- **Authors:** Lukuang Dong, Ziwei Li, Saierdaer Yusuyin, et al.
+- **abs:** https://arxiv.org/abs/2603.29217 · **PDF:** https://arxiv.org/pdf/2603.29217v1
+
+## 804. Software Vulnerability Detection Using a Lightweight Graph Neural Network
+
+- **ID:** `2603.29216` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Miles Farmer, Ekincan Ufuktepe, Anne Watson, et al.
+- **abs:** https://arxiv.org/abs/2603.29216 · **PDF:** https://arxiv.org/pdf/2603.29216v1
+
+## 805. Xuanwu: Evolving General Multimodal Models into an Industrial-Grade Foundation for Content Ecosystems
+
+- **ID:** `2603.29211` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Zhiqian Zhang, Xu Zhao, Xiaoqing Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.29211 · **PDF:** https://arxiv.org/pdf/2603.29211v1
+
+## 806. Route-Induced Density and Stability (RIDE): Controlled Intervention and Mechanism Analysis of Routing-Style Meta Prompts on LLM Internal States
+
+- **ID:** `2603.29206` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Dianxing Zhang, Gang Li, Sheng Li
+- **abs:** https://arxiv.org/abs/2603.29206 · **PDF:** https://arxiv.org/pdf/2603.29206v1
+
+## 807. Improving Ensemble Forecasts of Abnormally Deflecting Tropical Cyclones with Fused Atmosphere-Ocean-Terrain Data
+
+- **ID:** `2603.29200` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Qixiang Li, Yuan Zhou, Shuwei Huo, et al.
+- **abs:** https://arxiv.org/abs/2603.29200 · **PDF:** https://arxiv.org/pdf/2603.29200v2
+
+## 808. AEC-Bench: A Multimodal Benchmark for Agentic Systems in Architecture, Engineering, and Construction
+
+- **ID:** `2603.29199` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Harsh Mankodiya, Chase Gallik, Theodoros Galanos, et al.
+- **abs:** https://arxiv.org/abs/2603.29199 · **PDF:** https://arxiv.org/pdf/2603.29199v1
+
+## 809. Multi-Layered Memory Architectures for LLM Agents: An Experimental Evaluation of Long-Term Context Retention
+
+- **ID:** `2603.29194` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Sunil Tiwari, Payal Fofadiya
+- **abs:** https://arxiv.org/abs/2603.29194 · **PDF:** https://arxiv.org/pdf/2603.29194v1
+
+## 810. Developing Adaptive Context Compression Techniques for Large Language Models (LLMs) in Long-Running Interactions
+
+- **ID:** `2603.29193` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Payal Fofadiya, Sunil Tiwari
+- **abs:** https://arxiv.org/abs/2603.29193 · **PDF:** https://arxiv.org/pdf/2603.29193v1
+
+## 811. 3D Architect: An Automated Approach to Three-Dimensional Modeling
+
+- **ID:** `2603.29191` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Sunil Tiwari, Payal Fofadiya, Vicky Vishwakarma
+- **abs:** https://arxiv.org/abs/2603.29191 · **PDF:** https://arxiv.org/pdf/2603.29191v1
+
+## 812. SLVMEval: Synthetic Meta Evaluation Benchmark for Text-to-Long Video Generation
+
+- **ID:** `2603.29186` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Ryosuke Matsuda, Keito Kudo, Haruto Yoshida, et al.
+- **abs:** https://arxiv.org/abs/2603.29186 · **PDF:** https://arxiv.org/pdf/2603.29186v1
+
+## 813. Biomimetic PINNs for Cell-Induced Phase Transitions: UQ-R3 Sampling with Causal Gating
+
+- **ID:** `2603.29184` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Anci Lin, Xiaohong Liu, Zhiwen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.29184 · **PDF:** https://arxiv.org/pdf/2603.29184v1
+
+## 814. IMPACT: Influence Modeling for Open-Set Time Series Anomaly Detection
+
+- **ID:** `2603.29183` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Xiaohui Zhou, Yijie Wang, Hongzuo Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.29183 · **PDF:** https://arxiv.org/pdf/2603.29183v1
+
+## 815. Dummy-Aware Weighted Attack (DAWA): Breaking the Safe Sink in Dummy Class Defenses
+
+- **ID:** `2603.29182` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Yunrui Yu, Xuxiang Feng, Pengda Qin, et al.
+- **abs:** https://arxiv.org/abs/2603.29182 · **PDF:** https://arxiv.org/pdf/2603.29182v1
+
+## 816. Predicting Neuromodulation Outcome for Parkinson's Disease with Generative Virtual Brain Model
+
+- **ID:** `2603.29176` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-31
+- **Authors:** Siyuan Du, Siyi Li, Shuwei Bai, et al.
+- **abs:** https://arxiv.org/abs/2603.29176 · **PDF:** https://arxiv.org/pdf/2603.29176v1
+
+## 817. Segmentation of Gray Matters and White Matters from Brain MRI data
+
+- **ID:** `2603.29171` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Chang Sun, Rui Shi, Tsukasa Koike, et al.
+- **abs:** https://arxiv.org/abs/2603.29171 · **PDF:** https://arxiv.org/pdf/2603.29171v1
+
+## 818. Towards Automatic Soccer Commentary Generation with Knowledge-Enhanced Visual Reasoning
+
+- **ID:** `2604.00057` · **primary:** `cs.MM` · **Submitted:** 2026-03-31
+- **Authors:** Zeyu Jin, Xiaoyu Qin, Songtao Zhou, et al.
+- **abs:** https://arxiv.org/abs/2604.00057 · **PDF:** https://arxiv.org/pdf/2604.00057v1
+
+## 819. LatentPilot: Scene-Aware Vision-and-Language Navigation by Dreaming Ahead with Latent Visual Reasoning
+
+- **ID:** `2603.29165` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Haihong Hao, Lei Chen, Mingfei Han, et al.
+- **abs:** https://arxiv.org/abs/2603.29165 · **PDF:** https://arxiv.org/pdf/2603.29165v1
+
+## 820. Webscraper: Leverage Multimodal Large Language Models for Index-Content Web Scraping
+
+- **ID:** `2603.29161` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Guan-Lun Huang, Yuh-Jzer Joung
+- **abs:** https://arxiv.org/abs/2603.29161 · **PDF:** https://arxiv.org/pdf/2603.29161v1
+
+## 821. Kwame 2.0: Human-in-the-Loop Generative AI Teaching Assistant for Large Scale Online Coding Education in Africa
+
+- **ID:** `2603.29159` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** George Boateng, Samuel Boateng, Victor Kumbol
+- **abs:** https://arxiv.org/abs/2603.29159 · **PDF:** https://arxiv.org/pdf/2603.29159v1
+
+## 822. SimMOF: AI agent for Automated MOF Simulations
+
+- **ID:** `2603.29152` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Jaewoong Lee, Taeun Bae, Jihan Kim
+- **abs:** https://arxiv.org/abs/2603.29152 · **PDF:** https://arxiv.org/pdf/2603.29152v1
+
+## 823. Generalizable Dense Reward for Long-Horizon Robotic Tasks
+
+- **ID:** `2604.00055` · **primary:** `cs.RO` · **Submitted:** 2026-03-31
+- **Authors:** Silong Yong, Stephen Sheng, Carl Qi, et al.
+- **abs:** https://arxiv.org/abs/2604.00055 · **PDF:** https://arxiv.org/pdf/2604.00055v1
+
+## 824. Knowledge database development by large language models for countermeasures against viruses and marine toxins
+
+- **ID:** `2603.29149` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Hung N. Do, Jessica Z. Kubicek-Sutherland, S. Gnanakaran
+- **abs:** https://arxiv.org/abs/2603.29149 · **PDF:** https://arxiv.org/pdf/2603.29149v1
+
+## 825. Efficient and Scalable Granular-ball Graph Coarsening Method for Large-scale Graph Node Classification
+
+- **ID:** `2603.29148` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Guan Wang, Shuyin Xia, Lei Qian, et al.
+- **abs:** https://arxiv.org/abs/2603.29148 · **PDF:** https://arxiv.org/pdf/2603.29148v1
+
+## 826. REFINE: Real-world Exploration of Interactive Feedback and Student Behaviour
+
+- **ID:** `2603.29142` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Fares Fawzi, Seyed Parsa Neshaei, Marta Knezevic, et al.
+- **abs:** https://arxiv.org/abs/2603.29142 · **PDF:** https://arxiv.org/pdf/2603.29142v1
+
+## 827. Designing FSMs Specifications from Requirements with GPT 4.0
+
+- **ID:** `2603.29140` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Omer Nguena Timo, Paul-Alexis Rodriguez, Florent Avellaneda
+- **abs:** https://arxiv.org/abs/2603.29140 · **PDF:** https://arxiv.org/pdf/2603.29140v1
+
+## 828. SciVisAgentBench: A Benchmark for Evaluating Scientific Data Analysis and Visualization Agents
+
+- **ID:** `2603.29139` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Kuangshi Ai, Haichao Miao, Kaiyuan Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.29139 · **PDF:** https://arxiv.org/pdf/2603.29139v1
+
+## 829. Quality-Controlled Active Learning via Gaussian Processes for Robust Structure-Property Learning in Autonomous Microscopy
+
+- **ID:** `2603.29135` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Jawad Chowdhury, Ganesh Narasimha, Jan-Chi Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.29135 · **PDF:** https://arxiv.org/pdf/2603.29135v1
+
+## 830. Large-scale nonlinear optical computing with incoherent light via linear diffractive systems
+
+- **ID:** `2603.29131` · **primary:** `physics.optics` · **Submitted:** 2026-03-31
+- **Authors:** Alexander Chen, Yuntian Wang, Md Sadman Sakib Rahman, et al.
+- **abs:** https://arxiv.org/abs/2603.29131 · **PDF:** https://arxiv.org/pdf/2603.29131v1
+
+## 831. Adaptive Delayed-Update Cyclic Algorithm for Variational Inequalities
+
+- **ID:** `2603.29128` · **primary:** `math.OC` · **Submitted:** 2026-03-31
+- **Authors:** Yi Wei, Xufeng Cai, Jelena Diakonikolas
+- **abs:** https://arxiv.org/abs/2603.29128 · **PDF:** https://arxiv.org/pdf/2603.29128v1
+
+## 832. Concept Training for Human-Aligned Language Models
+
+- **ID:** `2603.29123` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Christine Zhang, Dan Jurafsky, Chen Shani
+- **abs:** https://arxiv.org/abs/2603.29123 · **PDF:** https://arxiv.org/pdf/2603.29123v1
+
+## 833. Economics of Human and AI Collaboration: When is Partial Automation More Attractive than Full Automation?
+
+- **ID:** `2603.29121` · **primary:** `econ.GN` · **Submitted:** 2026-03-31
+- **Authors:** Wensu Li, Atin Aboutorabi, Harry Lyu, et al.
+- **abs:** https://arxiv.org/abs/2603.29121 · **PDF:** https://arxiv.org/pdf/2603.29121v1
+
+## 834. Sampling-Horizon Neural Operator Predictors for Nonlinear Control under Delayed Inputs
+
+- **ID:** `2603.29119` · **primary:** `eess.SY` · **Submitted:** 2026-03-31
+- **Authors:** Luke Bhan, Peter Quawas, Miroslav Krstic, et al.
+- **abs:** https://arxiv.org/abs/2603.29119 · **PDF:** https://arxiv.org/pdf/2603.29119v1
+
+## 835. "I Just Need GPT to Refine My Prompts": Rethinking Onboarding and Help-Seeking with Generative 3D Modeling Tools
+
+- **ID:** `2603.29118` · **primary:** `cs.HC` · **Submitted:** 2026-03-31
+- **Authors:** Kanak Gautam, Poorvi Bhatia, Parmit K. Chilana
+- **abs:** https://arxiv.org/abs/2603.29118 · **PDF:** https://arxiv.org/pdf/2603.29118v1
+
+## 836. Predictor-Based Output-Feedback Control of Linear Systems with Time-Varying Input and Measurement Delays via Neural-Approximated Prediction Horizons
+
+- **ID:** `2603.29117` · **primary:** `eess.SY` · **Submitted:** 2026-03-31
+- **Authors:** Luke Bhan, Miroslav Krstic, Yuanyuan Shi
+- **abs:** https://arxiv.org/abs/2603.29117 · **PDF:** https://arxiv.org/pdf/2603.29117v1
+
+## 837. Towards Explainable Stakeholder-Aware Requirements Prioritisation in Aged-Care Digital Health
+
+- **ID:** `2603.29114` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Yuqing Xiao, John Grundy, Anuradha Madugalla, et al.
+- **abs:** https://arxiv.org/abs/2603.29114 · **PDF:** https://arxiv.org/pdf/2603.29114v1
+
+## 838. GISTBench: Evaluating LLM User Understanding via Evidence-Based Interest Verification
+
+- **ID:** `2603.29112` · **primary:** `cs.AI` · **Submitted:** 2026-03-31
+- **Authors:** Iordanis Fostiropoulos, Muhammad Rafay Azhar, Abdalaziz Sawwan, et al.
+- **abs:** https://arxiv.org/abs/2603.29112 · **PDF:** https://arxiv.org/pdf/2603.29112v1
+
+## 839. SemLoc: Structured Grounding of Free-Form LLM Reasoning for Fault Localization
+
+- **ID:** `2603.29109` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Zhaorui Yang, Haichao Zhu, Qian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.29109 · **PDF:** https://arxiv.org/pdf/2603.29109v1
+
+## 840. Efficient Bilevel Optimization with KFAC-Based Hypergradients
+
+- **ID:** `2603.29108` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Disen Liao, Felix Dangel, Yaoliang Yu
+- **abs:** https://arxiv.org/abs/2603.29108 · **PDF:** https://arxiv.org/pdf/2603.29108v1
+
+## 841. Evaluating a Data-Driven Redesign Process for Intelligent Tutoring Systems
+
+- **ID:** `2603.29094` · **primary:** `cs.HC` · **Submitted:** 2026-03-31
+- **Authors:** Qianru Lyu, Conrad Borchers, Meng Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.29094 · **PDF:** https://arxiv.org/pdf/2603.29094v1
+
+## 842. APEX-EM: Non-Parametric Online Learning for Autonomous Agents via Structured Procedural-Episodic Experience Replay
+
+- **ID:** `2603.29093` · **primary:** `cs.CL` · **Submitted:** 2026-03-31
+- **Authors:** Pratyay Banerjee, Masud Moshtaghi, Ankit Chadha
+- **abs:** https://arxiv.org/abs/2603.29093 · **PDF:** https://arxiv.org/pdf/2603.29093v1
+
+## 843. The Energy Footprint of LLM-Based Environmental Analysis: LLMs and Domain Products
+
+- **ID:** `2604.00053` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Alicia Bao, Jiamian He, Angel Hsu, et al.
+- **abs:** https://arxiv.org/abs/2604.00053 · **PDF:** https://arxiv.org/pdf/2604.00053v1
+
+## 844. HCLSM: Hierarchical Causal Latent State Machines for Object-Centric World Modeling
+
+- **ID:** `2603.29090` · **primary:** `cs.LG` · **Submitted:** 2026-03-31
+- **Authors:** Jaber Jaber, Osama Jaber
+- **abs:** https://arxiv.org/abs/2603.29090 · **PDF:** https://arxiv.org/pdf/2603.29090v1
+
+## 845. WorldFlow3D: Flowing Through 3D Distributions for Unbounded World Generation
+
+- **ID:** `2603.29089` · **primary:** `cs.CV` · **Submitted:** 2026-03-31
+- **Authors:** Amogh Joshi, Julian Ost, Felix Heide
+- **abs:** https://arxiv.org/abs/2603.29089 · **PDF:** https://arxiv.org/pdf/2603.29089v1
+
+## 846. WybeCoder: Verified Imperative Code Generation
+
+- **ID:** `2603.29088` · **primary:** `cs.SE` · **Submitted:** 2026-03-31
+- **Authors:** Fabian Gloeckle, Mantas Baksys, Darius Feher, et al.
+- **abs:** https://arxiv.org/abs/2603.29088 · **PDF:** https://arxiv.org/pdf/2603.29088v1
+
+## 847. Realistic Market Impact Modeling for Reinforcement Learning Trading Environments
+
+- **ID:** `2603.29086` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Lucas Riera Abbade, Anna Helena Reali Costa
+- **abs:** https://arxiv.org/abs/2603.29086 · **PDF:** https://arxiv.org/pdf/2603.29086v1
+
+## 848. PAR$^2$-RAG: Planned Active Retrieval and Reasoning for Multi-Hop Question Answering
+
+- **ID:** `2603.29085` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Xingyu Li, Rongguang Wang, Yuying Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.29085 · **PDF:** https://arxiv.org/pdf/2603.29085v1
+
+## 849. Is the Modality Gap a Bug or a Feature? A Robustness Perspective
+
+- **ID:** `2603.29080` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Rhea Chowers, Oshri Naparstek, Udi Barzelay, et al.
+- **abs:** https://arxiv.org/abs/2603.29080 · **PDF:** https://arxiv.org/pdf/2603.29080v1
+
+## 850. PolarQuant: Optimal Gaussian Weight Quantization via Hadamard Rotation for LLM Compression
+
+- **ID:** `2603.29078` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Caio Vicentino
+- **abs:** https://arxiv.org/abs/2603.29078 · **PDF:** https://arxiv.org/pdf/2603.29078v1
+
+## 851. Dual Perspectives in Emotion Attribution: A Generator-Interpreter Framework for Cross-Cultural Analysis of Emotion in LLMs
+
+- **ID:** `2603.29077` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Aizirek Turdubaeva, Uichin Lee
+- **abs:** https://arxiv.org/abs/2603.29077 · **PDF:** https://arxiv.org/pdf/2603.29077v1
+
+## 852. The Future of AI is Many, Not One
+
+- **ID:** `2603.29075` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Daniel J. Singer, Luca Garzino Demo
+- **abs:** https://arxiv.org/abs/2603.29075 · **PDF:** https://arxiv.org/pdf/2603.29075v1
+
+## 853. How much of persistent homology is topology? A quantitative decomposition for spin model phase transitions
+
+- **ID:** `2603.29072` · **primary:** `cond-mat.stat-mech` · **Submitted:** 2026-03-30
+- **Authors:** Matthew Loftus
+- **abs:** https://arxiv.org/abs/2603.29072 · **PDF:** https://arxiv.org/pdf/2603.29072v1
+
+## 854. On the Mirage of Long-Range Dependency, with an Application to Integer Multiplication
+
+- **ID:** `2603.29069` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Zichao Wei
+- **abs:** https://arxiv.org/abs/2603.29069 · **PDF:** https://arxiv.org/pdf/2603.29069v1
+
+## 855. ARCS: Autoregressive Circuit Synthesis with Topology-Aware Graph Attention and Spec Conditioning
+
+- **ID:** `2603.29068` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Tushar Dhananjay Pathak
+- **abs:** https://arxiv.org/abs/2603.29068 · **PDF:** https://arxiv.org/pdf/2603.29068v2
+
+## 856. CivicShield: A Cross-Domain Defense-in-Depth Framework for Securing Government-Facing AI Chatbots Against Multi-Turn Adversarial Attacks
+
+- **ID:** `2603.29062` · **primary:** `cs.CR` · **Submitted:** 2026-03-30
+- **Authors:** KrishnaSaiReddy Patil
+- **abs:** https://arxiv.org/abs/2603.29062 · **PDF:** https://arxiv.org/pdf/2603.29062v1
+
+## 857. Data-informed lifting line theory
+
+- **ID:** `2603.29051` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-03-30
+- **Authors:** Arjun Sharma, Jonas A. Actor, Peter A. Bosler
+- **abs:** https://arxiv.org/abs/2603.29051 · **PDF:** https://arxiv.org/pdf/2603.29051v1
+
+## 858. An Empirical Recipe for Universal Phone Recognition
+
+- **ID:** `2603.29042` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Shikhar Bharadwaj, Chin-Jou Li, Kwanghee Choi, et al.
+- **abs:** https://arxiv.org/abs/2603.29042 · **PDF:** https://arxiv.org/pdf/2603.29042v1
+
+## 859. A Latent Risk-Aware Machine Learning Approach for Predicting Operational Success in Clinical Trials based on TrialsBank
+
+- **ID:** `2603.29041` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Iness Halimi, Emmanuel Piffo, Oumnia Boudersa, et al.
+- **abs:** https://arxiv.org/abs/2603.29041 · **PDF:** https://arxiv.org/pdf/2603.29041v1
+
+## 860. Trojan-Speak: Bypassing Constitutional Classifiers with No Jailbreak Tax via Adversarial Finetuning
+
+- **ID:** `2603.29038` · **primary:** `cs.CR` · **Submitted:** 2026-03-30
+- **Authors:** Bilgehan Sel, Xuanli He, Alwin Peng, et al.
+- **abs:** https://arxiv.org/abs/2603.29038 · **PDF:** https://arxiv.org/pdf/2603.29038v1
+
+## 861. From Astronomy to Astrology: Testing the Illusion of Zodiac-Based Personality Prediction with Machine Learning
+
+- **ID:** `2603.29033` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Abhinna Sundar Samantaray, Finnja Annika Fluhrer, Dhruv Saini, et al.
+- **abs:** https://arxiv.org/abs/2603.29033 · **PDF:** https://arxiv.org/pdf/2603.29033v1
+
+## 862. MMFace-DiT: A Dual-Stream Diffusion Transformer for High-Fidelity Multimodal Face Generation
+
+- **ID:** `2603.29029` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Bharath Krishnamurthy, Ajita Rattani
+- **abs:** https://arxiv.org/abs/2603.29029 · **PDF:** https://arxiv.org/pdf/2603.29029v1
+
+## 863. On the limited utility of parallel data for learning shared multilingual representations
+
+- **ID:** `2603.29026` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Julius Leino, Jörg Tiedemann
+- **abs:** https://arxiv.org/abs/2603.29026 · **PDF:** https://arxiv.org/pdf/2603.29026v1
+
+## 864. The Model Says Walk: How Surface Heuristics Override Implicit Constraints in LLM Reasoning
+
+- **ID:** `2603.29025` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Yubo Li, Lu Zhang, Tianchong Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.29025 · **PDF:** https://arxiv.org/pdf/2603.29025v1
+
+## 865. Human-Like Lifelong Memory: A Neuroscience-Grounded Architecture for Infinite Interaction
+
+- **ID:** `2603.29023` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Diego C. Lerma-Torres
+- **abs:** https://arxiv.org/abs/2603.29023 · **PDF:** https://arxiv.org/pdf/2603.29023v1
+
+## 866. Emergence WebVoyager: Toward Consistent and Transparent Evaluation of (Web) Agents in The Wild
+
+- **ID:** `2603.29020` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Deepak Akkil, Mowafak Allaham, Amal Raj, et al.
+- **abs:** https://arxiv.org/abs/2603.29020 · **PDF:** https://arxiv.org/pdf/2603.29020v1
+
+## 867. Improving Efficiency of GPU Kernel Optimization Agents using a Domain-Specific Language and Speed-of-Light Guidance
+
+- **ID:** `2603.29010` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Siva Kumar Sastry Hari, Vignesh Balaji, Sana Damani, et al.
+- **abs:** https://arxiv.org/abs/2603.29010 · **PDF:** https://arxiv.org/pdf/2603.29010v1
+
+## 868. Understand and Accelerate Memory Processing Pipeline for Disaggregated LLM Inference
+
+- **ID:** `2603.29002` · **primary:** `cs.DC` · **Submitted:** 2026-03-30
+- **Authors:** Zifan He, Rui Ma, Yizhou Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.29002 · **PDF:** https://arxiv.org/pdf/2603.29002v1
+
+## 869. Transfer Learning in Bayesian Optimization for Aircraft Design
+
+- **ID:** `2603.28999` · **primary:** `math.OC` · **Submitted:** 2026-03-30
+- **Authors:** Ali Tfaily, Youssef Diouane, Nathalie Bartoli, et al.
+- **abs:** https://arxiv.org/abs/2603.28999 · **PDF:** https://arxiv.org/pdf/2603.28999v1
+
+## 870. Design Principles for the Construction of a Benchmark Evaluating Security Operation Capabilities of Multi-agent AI Systems
+
+- **ID:** `2603.28998` · **primary:** `cs.CR` · **Submitted:** 2026-03-30
+- **Authors:** Yicheng Cai, Mitchell John DeStefano, Guodong Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.28998 · **PDF:** https://arxiv.org/pdf/2603.28998v1
+
+## 871. An Explicit Surrogate for Gaussian Mixture Flow Matching with Wasserstein Gap Bounds
+
+- **ID:** `2603.28992` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Elham Rostami, Taous-Meriem Laleg-Kirati, Hamidou Tembine
+- **abs:** https://arxiv.org/abs/2603.28992 · **PDF:** https://arxiv.org/pdf/2603.28992v1
+
+## 872. Drop the Hierarchy and Roles: How Self-Organizing LLM Agents Outperform Designed Structures
+
+- **ID:** `2603.28990` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Victoria Dochkina
+- **abs:** https://arxiv.org/abs/2603.28990 · **PDF:** https://arxiv.org/pdf/2603.28990v1
+
+## 873. Multi-fidelity approaches for general constrained Bayesian optimization with application to aircraft design
+
+- **ID:** `2603.28987` · **primary:** `math.OC` · **Submitted:** 2026-03-30
+- **Authors:** Oihan Cordelier, Youssef Diouane, Nathalie Bartoli, et al.
+- **abs:** https://arxiv.org/abs/2603.28987 · **PDF:** https://arxiv.org/pdf/2603.28987v1
+
+## 874. Mimosa Framework: Toward Evolving Multi-Agent Systems for Scientific Research
+
+- **ID:** `2603.28986` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Martin Legrand, Tao Jiang, Matthieu Feraud, et al.
+- **abs:** https://arxiv.org/abs/2603.28986 · **PDF:** https://arxiv.org/pdf/2603.28986v1
+
+## 875. Privacy Guard & Token Parsimony by Prompt and Context Handling and LLM Routing
+
+- **ID:** `2603.28972` · **primary:** `cs.CR` · **Submitted:** 2026-03-30
+- **Authors:** Alessio Langiu
+- **abs:** https://arxiv.org/abs/2603.28972 · **PDF:** https://arxiv.org/pdf/2603.28972v1
+
+## 876. A Pontryagin Method of Model-based Reinforcement Learning via Hamiltonian Actor-Critic
+
+- **ID:** `2603.28971` · **primary:** `eess.SY` · **Submitted:** 2026-03-30
+- **Authors:** Chengyang Gu, Yuxin Pan, Hui Xiong, et al.
+- **abs:** https://arxiv.org/abs/2603.28971 · **PDF:** https://arxiv.org/pdf/2603.28971v1
+
+## 877. The Spectral Edge Thesis: A Mathematical Framework for Intra-Signal Phase Transitions in Neural Network Training
+
+- **ID:** `2603.28964` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Yongzhong Xu
+- **abs:** https://arxiv.org/abs/2603.28964 · **PDF:** https://arxiv.org/pdf/2603.28964v1
+
+## 878. AutoWorld: Scaling Multi-Agent Traffic Simulation with Self-Supervised World Models
+
+- **ID:** `2603.28963` · **primary:** `cs.RO` · **Submitted:** 2026-03-30
+- **Authors:** Mozhgan Pourkeshavatz, Tianran Liu, Nicholas Rhinehart
+- **abs:** https://arxiv.org/abs/2603.28963 · **PDF:** https://arxiv.org/pdf/2603.28963v1
+
+## 879. Multi-Agent LLMs for Adaptive Acquisition in Bayesian Optimization
+
+- **ID:** `2603.28959` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Andrea Carbonati, Mohammadsina Almasi, Hadis Anahideh
+- **abs:** https://arxiv.org/abs/2603.28959 · **PDF:** https://arxiv.org/pdf/2603.28959v1
+
+## 880. Task-Centric Personalized Federated Fine-Tuning of Language Models
+
+- **ID:** `2604.00050` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Gabriel U. Talasso, Meghdad Kurmanji, Allan M. de Souza, et al.
+- **abs:** https://arxiv.org/abs/2604.00050 · **PDF:** https://arxiv.org/pdf/2604.00050v1
+
+## 881. Minimum Norm Interpolation via The Local Theory of Banach Spaces: The Role of $2$-Uniform Convexity
+
+- **ID:** `2603.28956` · **primary:** `math.FA` · **Submitted:** 2026-03-30
+- **Authors:** Gil Kur, Pierre Bizeul
+- **abs:** https://arxiv.org/abs/2603.28956 · **PDF:** https://arxiv.org/pdf/2603.28956v1
+
+## 882. Enhancing Policy Learning with World-Action Model
+
+- **ID:** `2603.28955` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Yuci Han, Alper Yilmaz
+- **abs:** https://arxiv.org/abs/2603.28955 · **PDF:** https://arxiv.org/pdf/2603.28955v1
+
+## 883. Differentiable Initialization-Accelerated CPU-GPU Hybrid Combinatorial Scheduling
+
+- **ID:** `2603.28943` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Mingju Liu, Jiaqi Yin, Alvaro Velasquez, et al.
+- **abs:** https://arxiv.org/abs/2603.28943 · **PDF:** https://arxiv.org/pdf/2603.28943v1
+
+## 884. \texttt{ReproMIA}: A Comprehensive Analysis of Model Reprogramming for Proactive Membership Inference Attacks
+
+- **ID:** `2603.28942` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Chihan Huang, Huaijin Wang, Shuai Wang
+- **abs:** https://arxiv.org/abs/2603.28942 · **PDF:** https://arxiv.org/pdf/2603.28942v1
+
+## 885. Foundations of Polar Linear Algebra
+
+- **ID:** `2603.28939` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Giovanni Guasti
+- **abs:** https://arxiv.org/abs/2603.28939 · **PDF:** https://arxiv.org/pdf/2603.28939v1
+
+## 886. Optimistic Online LQR via Intrinsic Rewards
+
+- **ID:** `2603.28938` · **primary:** `eess.SY` · **Submitted:** 2026-03-30
+- **Authors:** Marcell Bartos, Bruce D. Lee, Lenart Treven, et al.
+- **abs:** https://arxiv.org/abs/2603.28938 · **PDF:** https://arxiv.org/pdf/2603.28938v1
+
+## 887. A Neural Tension Operator for Curve Subdivision across Constant Curvature Geometries
+
+- **ID:** `2603.28937` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Hassan Ugail, Newton Howard
+- **abs:** https://arxiv.org/abs/2603.28937 · **PDF:** https://arxiv.org/pdf/2603.28937v1
+
+## 888. Known Intents, New Combinations: Clause-Factorized Decoding for Compositional Multi-Intent Detection
+
+- **ID:** `2603.28929` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Abhilash Nandy
+- **abs:** https://arxiv.org/abs/2603.28929 · **PDF:** https://arxiv.org/pdf/2603.28929v1
+
+## 889. Towards Computational Social Dynamics of Semi-Autonomous AI Agents
+
+- **ID:** `2603.28928` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** S. O. Lidarity, U. N. Ionize, C. O. Llective, et al.
+- **abs:** https://arxiv.org/abs/2603.28928 · **PDF:** https://arxiv.org/pdf/2603.28928v1
+
+## 890. Theory of Mind and Self-Attributions of Mentality are Dissociable in LLMs
+
+- **ID:** `2603.28925` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Junsol Kim, Winnie Street, Roberta Rocca, et al.
+- **abs:** https://arxiv.org/abs/2603.28925 · **PDF:** https://arxiv.org/pdf/2603.28925v1
+
+## 891. CrossTrace: A Cross-Domain Dataset of Grounded Scientific Reasoning Traces for Hypothesis Generation
+
+- **ID:** `2603.28924` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Andrew Bouras, OMS-II Research Fellow
+- **abs:** https://arxiv.org/abs/2603.28924 · **PDF:** https://arxiv.org/pdf/2603.28924v1
+
+## 892. Beta-Scheduling: Momentum from Critical Damping as a Diagnostic and Correction Tool for Neural Network Training
+
+- **ID:** `2603.28921` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Ivan Pasichnyk
+- **abs:** https://arxiv.org/abs/2603.28921 · **PDF:** https://arxiv.org/pdf/2603.28921v1
+
+## 893. Symmetrizing Bregman Divergence on the Cone of Positive Definite Matrices: Which Mean to Use and Why
+
+- **ID:** `2603.28917` · **primary:** `math.OC` · **Submitted:** 2026-03-30
+- **Authors:** Tushar Sial, Abhishek Halder
+- **abs:** https://arxiv.org/abs/2603.28917 · **PDF:** https://arxiv.org/pdf/2603.28917v1
+
+## 894. Structural Pass Analysis in Football: Learning Pass Archetypes and Tactical Impact from Spatio-Temporal Tracking Data
+
+- **ID:** `2603.28916` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Oktay Karakuş, Hasan Arkadaş
+- **abs:** https://arxiv.org/abs/2603.28916 · **PDF:** https://arxiv.org/pdf/2603.28916v1
+
+## 895. From Consensus to Split Decisions: ABC-Stratified Sentiment in Holocaust Oral Histories
+
+- **ID:** `2603.28913` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Daban Q. Jaff
+- **abs:** https://arxiv.org/abs/2603.28913 · **PDF:** https://arxiv.org/pdf/2603.28913v1
+
+## 896. Working Paper: Towards a Category-theoretic Comparative Framework for Artificial General Intelligence
+
+- **ID:** `2603.28906` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Pablo de los Riscos, Fernando J. Corbacho, Michael A. Arbib
+- **abs:** https://arxiv.org/abs/2603.28906 · **PDF:** https://arxiv.org/pdf/2603.28906v1
+
+## 897. ChartDiff: A Large-Scale Benchmark for Comprehending Pairs of Charts
+
+- **ID:** `2603.28902` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Rongtian Ye
+- **abs:** https://arxiv.org/abs/2603.28902 · **PDF:** https://arxiv.org/pdf/2603.28902v1
+
+## 898. Robust Multi-Agent Reinforcement Learning for Small UAS Separation Assurance under GPS Degradation and Spoofing
+
+- **ID:** `2603.28900` · **primary:** `cs.RO` · **Submitted:** 2026-03-30
+- **Authors:** Alex Zongo, Filippos Fotiadis, Ufuk Topcu, et al.
+- **abs:** https://arxiv.org/abs/2603.28900 · **PDF:** https://arxiv.org/pdf/2603.28900v1
+
+## 899. OccSim: Multi-kilometer Simulation with Long-horizon Occupancy World Models
+
+- **ID:** `2603.28887` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Tianran Liu, Shengwen Zhao, Mozhgan Pourkeshavarz, et al.
+- **abs:** https://arxiv.org/abs/2603.28887 · **PDF:** https://arxiv.org/pdf/2603.28887v1
+
+## 900. Calibrated Fusion for Heterogeneous Graph-Vector Retrieval in Multi-Hop QA
+
+- **ID:** `2603.28886` · **primary:** `cs.IR` · **Submitted:** 2026-03-30
+- **Authors:** Andre Bacellar
+- **abs:** https://arxiv.org/abs/2603.28886 · **PDF:** https://arxiv.org/pdf/2603.28886v1
+
+## 901. OptiMer: Optimal Distribution Vector Merging Is Better than Data Mixing for Continual Pre-Training
+
+- **ID:** `2603.28858` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Haiyue Song, Masao Utiyama
+- **abs:** https://arxiv.org/abs/2603.28858 · **PDF:** https://arxiv.org/pdf/2603.28858v1
+
+## 902. Adaptive Block-Scaled Data Types
+
+- **ID:** `2603.28765` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Jack Cook, Hyemin S. Lee, Kathryn Le, et al.
+- **abs:** https://arxiv.org/abs/2603.28765 · **PDF:** https://arxiv.org/pdf/2603.28765v1
+
+## 903. Geometry-aware similarity metrics for neural representations on Riemannian and statistical manifolds
+
+- **ID:** `2603.28764` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** N Alex Cayco-Gajic, Arthur Pellegrino
+- **abs:** https://arxiv.org/abs/2603.28764 · **PDF:** https://arxiv.org/pdf/2603.28764v1
+
+## 904. On-the-fly Repulsion in the Contextual Space for Rich Diversity in Diffusion Transformers
+
+- **ID:** `2603.28762` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Omer Dahary, Benaya Koren, Daniel Garibi, et al.
+- **abs:** https://arxiv.org/abs/2603.28762 · **PDF:** https://arxiv.org/pdf/2603.28762v1
+
+## 905. Temporal Credit Is Free
+
+- **ID:** `2603.28750` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Aur Shalev Merin
+- **abs:** https://arxiv.org/abs/2603.28750 · **PDF:** https://arxiv.org/pdf/2603.28750v1
+
+## 906. Stop Probing, Start Coding: Why Linear Probes and Sparse Autoencoders Fail at Compositional Generalisation
+
+- **ID:** `2603.28744` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Vitória Barin Pacela, Shruti Joshi, Isabela Camacho, et al.
+- **abs:** https://arxiv.org/abs/2603.28744 · **PDF:** https://arxiv.org/pdf/2603.28744v1
+
+## 907. Rethinking Language Model Scaling under Transferable Hypersphere Optimization
+
+- **ID:** `2603.28743` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Liliang Ren, Yang Liu, Yelong Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.28743 · **PDF:** https://arxiv.org/pdf/2603.28743v1
+
+## 908. Expectation Error Bounds for Transfer Learning in Linear Regression and Linear Neural Networks
+
+- **ID:** `2603.28739` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Meitong Liu, Christopher Jung, Rui Li, et al.
+- **abs:** https://arxiv.org/abs/2603.28739 · **PDF:** https://arxiv.org/pdf/2603.28739v1
+
+## 909. ParaSpeechCLAP: A Dual-Encoder Speech-Text Model for Rich Stylistic Language-Audio Pretraining
+
+- **ID:** `2603.28737` · **primary:** `eess.AS` · **Submitted:** 2026-03-30
+- **Authors:** Anuj Diwan, Eunsol Choi, David Harwath
+- **abs:** https://arxiv.org/abs/2603.28737 · **PDF:** https://arxiv.org/pdf/2603.28737v1
+
+## 910. RAD-AI: Rethinking Architecture Documentation for AI-Augmented Ecosystems
+
+- **ID:** `2603.28735` · **primary:** `cs.SE` · **Submitted:** 2026-03-30
+- **Authors:** Oliver Aleksander Larsen, Mahyar T. Moghaddam
+- **abs:** https://arxiv.org/abs/2603.28735 · **PDF:** https://arxiv.org/pdf/2603.28735v1
+
+## 911. See it to Place it: Evolving Macro Placements with Vision-Language Models
+
+- **ID:** `2603.28733` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Ikechukwu Uchendu, Swati Goel, Karly Hou, et al.
+- **abs:** https://arxiv.org/abs/2603.28733 · **PDF:** https://arxiv.org/pdf/2603.28733v1
+
+## 912. SAGAI-MID: A Generative AI-Driven Middleware for Dynamic Runtime Interoperability
+
+- **ID:** `2603.28731` · **primary:** `cs.SE` · **Submitted:** 2026-03-30
+- **Authors:** Oliver Aleksander Larsen, Mahyar T. Moghaddam
+- **abs:** https://arxiv.org/abs/2603.28731 · **PDF:** https://arxiv.org/pdf/2603.28731v1
+
+## 913. SOLE-R1: Video-Language Reasoning as the Sole Reward for On-Robot Reinforcement Learning
+
+- **ID:** `2603.28730` · **primary:** `cs.RO` · **Submitted:** 2026-03-30
+- **Authors:** Philip Schroeder, Thomas Weng, Karl Schmeckpeper, et al.
+- **abs:** https://arxiv.org/abs/2603.28730 · **PDF:** https://arxiv.org/pdf/2603.28730v1
+
+## 914. OneComp: One-Line Revolution for Generative AI Model Compression
+
+- **ID:** `2603.28845` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Yuma Ichikawa, Keiji Kimura, Akihiro Yoshida, et al.
+- **abs:** https://arxiv.org/abs/2603.28845 · **PDF:** https://arxiv.org/pdf/2603.28845v1
+
+## 915. Stepwise Credit Assignment for GRPO on Flow-Matching Models
+
+- **ID:** `2603.28718` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Yash Savani, Branislav Kveton, Yuchen Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.28718 · **PDF:** https://arxiv.org/pdf/2603.28718v1
+
+## 916. Dynamic Dual-Granularity Skill Bank for Agentic RL
+
+- **ID:** `2603.28716` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Songjun Tu, Chengdong Xu, Qichao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.28716 · **PDF:** https://arxiv.org/pdf/2603.28716v1
+
+## 917. GPU-Accelerated Optimization of Transformer-Based Neural Networks for Real-Time Inference
+
+- **ID:** `2603.28708` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Soutrik Mukherjee, Sangwhan Cha
+- **abs:** https://arxiv.org/abs/2603.28708 · **PDF:** https://arxiv.org/pdf/2603.28708v1
+
+## 918. A Convex Route to Thermomechanics: Learning Internal Energy and Dissipation
+
+- **ID:** `2603.28707` · **primary:** `cs.CE` · **Submitted:** 2026-03-30
+- **Authors:** Hagen Holthusen, Paul Steinmann, Ellen Kuhl
+- **abs:** https://arxiv.org/abs/2603.28707 · **PDF:** https://arxiv.org/pdf/2603.28707v2
+
+## 919. EpiScreen: Early Epilepsy Detection from Electronic Health Records with Large Language Models
+
+- **ID:** `2603.28698` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Shuang Zhou, Kai Yu, Zaifu Zhan, et al.
+- **abs:** https://arxiv.org/abs/2603.28698 · **PDF:** https://arxiv.org/pdf/2603.28698v1
+
+## 920. AdaptToken: Entropy-based Adaptive Token Selection for MLLM Long Video Understanding
+
+- **ID:** `2603.28696` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Haozhe Qi, Kevin Qu, Mahdi Rad, et al.
+- **abs:** https://arxiv.org/abs/2603.28696 · **PDF:** https://arxiv.org/pdf/2603.28696v1
+
+## 921. Functional Natural Policy Gradients
+
+- **ID:** `2603.28681` · **primary:** `stat.ML` · **Submitted:** 2026-03-30
+- **Authors:** Aurelien Bibaut, Houssam Zenati, Thibaud Rahier, et al.
+- **abs:** https://arxiv.org/abs/2603.28681 · **PDF:** https://arxiv.org/pdf/2603.28681v1
+
+## 922. Subspace Optimization for Backpropagation-Free Continual Test-Time Adaptation
+
+- **ID:** `2603.28678` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Damian Sójka, Sebastian Cygert, Marc Masana
+- **abs:** https://arxiv.org/abs/2603.28678 · **PDF:** https://arxiv.org/pdf/2603.28678v1
+
+## 923. Why Aggregate Accuracy is Inadequate for Evaluating Fairness in Law Enforcement Facial Recognition Systems
+
+- **ID:** `2603.28675` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Khalid Adnan Alsayed
+- **abs:** https://arxiv.org/abs/2603.28675 · **PDF:** https://arxiv.org/pdf/2603.28675v1
+
+## 924. FL-PBM: Pre-Training Backdoor Mitigation for Federated Learning
+
+- **ID:** `2603.28673` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Osama Wehbi, Sarhad Arisdakessian, Omar Abdel Wahab, et al.
+- **abs:** https://arxiv.org/abs/2603.28673 · **PDF:** https://arxiv.org/pdf/2603.28673v1
+
+## 925. AMIGO: Agentic Multi-Image Grounding Oracle Benchmark
+
+- **ID:** `2603.28662` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Min Wang, Ata Mahjoubfar
+- **abs:** https://arxiv.org/abs/2603.28662 · **PDF:** https://arxiv.org/pdf/2603.28662v1
+
+## 926. BACE: LLM-based Code Generation through Bayesian Anchored Co-Evolution of Code and Test Populations
+
+- **ID:** `2603.28653` · **primary:** `cs.NE` · **Submitted:** 2026-03-30
+- **Authors:** Kaushitha Silva, Srinath Perera
+- **abs:** https://arxiv.org/abs/2603.28653 · **PDF:** https://arxiv.org/pdf/2603.28653v1
+
+## 927. Mitigating Backdoor Attacks in Federated Learning Using PPA and MiniMax Game Theory
+
+- **ID:** `2603.28652` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Osama Wehbi, Sarhad Arisdakessian, Omar Abdel Wahab, et al.
+- **abs:** https://arxiv.org/abs/2603.28652 · **PDF:** https://arxiv.org/pdf/2603.28652v1
+
+## 928. Information-Theoretic Limits of Safety Verification for Self-Improving Systems
+
+- **ID:** `2603.28650` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Arsenios Scrivens
+- **abs:** https://arxiv.org/abs/2603.28650 · **PDF:** https://arxiv.org/pdf/2603.28650v2
+
+## 929. Constructing Composite Features for Interpretable Music-Tagging
+
+- **ID:** `2603.28644` · **primary:** `cs.SD` · **Submitted:** 2026-03-30
+- **Authors:** Chenhao Xue, Weitao Hu, Joyraj Chakraborty, et al.
+- **abs:** https://arxiv.org/abs/2603.28644 · **PDF:** https://arxiv.org/pdf/2603.28644v1
+
+## 930. The Ultimate Tutorial for AI-driven Scale Development in Generative Psychometrics: Releasing AIGENIE from its Bottle
+
+- **ID:** `2603.28643` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Lara Russell-Lasalandra, Hudson Golino, Luis Eduardo Garrido, et al.
+- **abs:** https://arxiv.org/abs/2603.28643 · **PDF:** https://arxiv.org/pdf/2603.28643v1
+
+## 931. Dynamic Lookahead Distance via Reinforcement Learning-Based Pure Pursuit for Autonomous Racing
+
+- **ID:** `2603.28625` · **primary:** `cs.RO` · **Submitted:** 2026-03-30
+- **Authors:** Mohamed Elgouhary, Amr S. El-Wakeel
+- **abs:** https://arxiv.org/abs/2603.28625 · **PDF:** https://arxiv.org/pdf/2603.28625v1
+
+## 932. Trust-Aware Routing for Distributed Generative AI Inference at the Edge
+
+- **ID:** `2603.28622` · **primary:** `cs.DC` · **Submitted:** 2026-03-30
+- **Authors:** Chanh Nguyen, Erik Elmroth
+- **abs:** https://arxiv.org/abs/2603.28622 · **PDF:** https://arxiv.org/pdf/2603.28622v1
+
+## 933. Seeing with You: Perception-Reasoning Coevolution for Multimodal Reasoning
+
+- **ID:** `2603.28618` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Ziqi Miao, Haonan Jia, Lijun Li, et al.
+- **abs:** https://arxiv.org/abs/2603.28618 · **PDF:** https://arxiv.org/pdf/2603.28618v1
+
+## 934. TGIF2: Extended Text-Guided Inpainting Forgery Dataset & Benchmark
+
+- **ID:** `2603.28613` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Hannes Mareen, Dimitrios Karageorgiou, Paschalis Giakoumoglou, et al.
+- **abs:** https://arxiv.org/abs/2603.28613 · **PDF:** https://arxiv.org/pdf/2603.28613v1
+
+## 935. LACE: Loss-Adaptive Capacity Expansion for Continual Learning
+
+- **ID:** `2603.28611` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Shivnath Tathe
+- **abs:** https://arxiv.org/abs/2603.28611 · **PDF:** https://arxiv.org/pdf/2603.28611v1
+
+## 936. ResAdapt: Adaptive Resolution for Efficient Multimodal Reasoning
+
+- **ID:** `2603.28610` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Huanxuan Liao, Zhongtao Jiang, Yupu Hao, et al.
+- **abs:** https://arxiv.org/abs/2603.28610 · **PDF:** https://arxiv.org/pdf/2603.28610v2
+
+## 937. Unsafe2Safe: Controllable Image Anonymization for Downstream Utility
+
+- **ID:** `2603.28605` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Mih Dinh, SouYoung Jin
+- **abs:** https://arxiv.org/abs/2603.28605 · **PDF:** https://arxiv.org/pdf/2603.28605v1
+
+## 938. Position: Explainable AI is Causality in Disguise
+
+- **ID:** `2603.28597` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Amir-Hossein Karimi
+- **abs:** https://arxiv.org/abs/2603.28597 · **PDF:** https://arxiv.org/pdf/2603.28597v1
+
+## 939. Moving Beyond Review: Applying Language Models to Planning and Translation in Reflection
+
+- **ID:** `2603.28596` · **primary:** `cs.HC` · **Submitted:** 2026-03-30
+- **Authors:** Seyed Parsa Neshaei, Richard Lee Davis, Tanja Käser
+- **abs:** https://arxiv.org/abs/2603.28596 · **PDF:** https://arxiv.org/pdf/2603.28596v1
+
+## 940. Optimistic Actor-Critic with Parametric Policies for Linear Markov Decision Processes
+
+- **ID:** `2603.28595` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Max Qiushi Lin, Reza Asad, Kevin Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.28595 · **PDF:** https://arxiv.org/pdf/2603.28595v2
+
+## 941. Detection of Adversarial Attacks in Robotic Perception
+
+- **ID:** `2603.28594` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Ziad Sharawy, Mohammad Nakshbandi, Sorin Mihai Grigorescu
+- **abs:** https://arxiv.org/abs/2603.28594 · **PDF:** https://arxiv.org/pdf/2603.28594v2
+
+## 942. Physics-Informed Framework for Impact Identification in Aerospace Composites
+
+- **ID:** `2603.28593` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Natália Ribeiro Marinho, Richard Loendersloot, Jan Willem Wiegman, et al.
+- **abs:** https://arxiv.org/abs/2603.28593 · **PDF:** https://arxiv.org/pdf/2603.28593v1
+
+## 943. Universal Approximation Constraints of Narrow ResNets: The Tunnel Effect
+
+- **ID:** `2603.28591` · **primary:** `math.DS` · **Submitted:** 2026-03-30
+- **Authors:** Christian Kuehn, Sara-Viola Kuntz, Tobias Wöhrer
+- **abs:** https://arxiv.org/abs/2603.28591 · **PDF:** https://arxiv.org/pdf/2603.28591v1
+
+## 944. MonitorBench: A Comprehensive Benchmark for Chain-of-Thought Monitorability in Large Language Models
+
+- **ID:** `2603.28590` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Han Wang, Yifan Sun, Brian Ko, et al.
+- **abs:** https://arxiv.org/abs/2603.28590 · **PDF:** https://arxiv.org/pdf/2603.28590v2
+
+## 945. Towards a Medical AI Scientist
+
+- **ID:** `2603.28589` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Hongtao Wu, Boyun Zheng, Dingjie Song, et al.
+- **abs:** https://arxiv.org/abs/2603.28589 · **PDF:** https://arxiv.org/pdf/2603.28589v1
+
+## 946. Navigating the Mirage: A Dual-Path Agentic Framework for Robust Misleading Chart Question Answering
+
+- **ID:** `2603.28583` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Yanjie Zhang, Yafei Li, Rui Sheng, et al.
+- **abs:** https://arxiv.org/abs/2603.28583 · **PDF:** https://arxiv.org/pdf/2603.28583v1
+
+## 947. ChemCLIP: Bridging Organic and Inorganic Anticancer Compounds Through Contrastive Learning
+
+- **ID:** `2603.28575` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Mohamad Koohi-Moghadam, Hongzhe Sun, Hongyan Li, et al.
+- **abs:** https://arxiv.org/abs/2603.28575 · **PDF:** https://arxiv.org/pdf/2603.28575v1
+
+## 948. Learning Partial Action Replacement in Offline MARL
+
+- **ID:** `2603.28573` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Yue Jin, Giovanni Montana
+- **abs:** https://arxiv.org/abs/2603.28573 · **PDF:** https://arxiv.org/pdf/2603.28573v1
+
+## 949. Unrestrained Simplex Denoising for Discrete Data. A Non-Markovian Approach Applied to Graph Generation
+
+- **ID:** `2603.28572` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Yoann Boget, Alexandros Kalousis
+- **abs:** https://arxiv.org/abs/2603.28572 · **PDF:** https://arxiv.org/pdf/2603.28572v1
+
+## 950. CirrusBench: Evaluating LLM-based Agents Beyond Correctness in Real-World Cloud Service Environments
+
+- **ID:** `2603.28569` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Yi Yu, Guangquan Hu, Chenghuang Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.28569 · **PDF:** https://arxiv.org/pdf/2603.28569v1
+
+## 951. Fine-Tuning Large Language Models for Cooperative Tactical Deconfliction of Small Unmanned Aerial Systems
+
+- **ID:** `2603.28561` · **primary:** `cs.RO` · **Submitted:** 2026-03-30
+- **Authors:** Iman Sharifi, Alex Zongo, Peng Wei
+- **abs:** https://arxiv.org/abs/2603.28561 · **PDF:** https://arxiv.org/pdf/2603.28561v1
+
+## 952. T-Norm Operators for EU AI Act Compliance Classification: An Empirical Comparison of Lukasiewicz, Product, and Gödel Semantics in a Neuro-Symbolic Reasoning System
+
+- **ID:** `2603.28558` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Adam Laabs
+- **abs:** https://arxiv.org/abs/2603.28558 · **PDF:** https://arxiv.org/pdf/2603.28558v1
+
+## 953. Domain-Invariant Prompt Learning for Vision-Language Models
+
+- **ID:** `2603.28555` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Arsham Gholamzadeh Khoee, Yinan Yu, Robert Feldt
+- **abs:** https://arxiv.org/abs/2603.28555 · **PDF:** https://arxiv.org/pdf/2603.28555v1
+
+## 954. Hydra: Unifying Document Retrieval and Generation in a Single Vision-Language Model
+
+- **ID:** `2603.28554` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Athos Georgiou
+- **abs:** https://arxiv.org/abs/2603.28554 · **PDF:** https://arxiv.org/pdf/2603.28554v1
+
+## 955. Multimodal Analytics of Cybersecurity Crisis Preparation Exercises: What Predicts Success?
+
+- **ID:** `2603.28553` · **primary:** `cs.HC` · **Submitted:** 2026-03-30
+- **Authors:** Conrad Borchers, Valdemar Švábenský, Sandesh K. Kafle, et al.
+- **abs:** https://arxiv.org/abs/2603.28553 · **PDF:** https://arxiv.org/pdf/2603.28553v1
+
+## 956. Training data generation for context-dependent rubric-based short answer grading
+
+- **ID:** `2603.28537` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Pavel Šindelář, Dávid Slivka, Christopher Bouma, et al.
+- **abs:** https://arxiv.org/abs/2603.28537 · **PDF:** https://arxiv.org/pdf/2603.28537v2
+
+## 957. Compressing Transformer Language Models via Matrix Product Operator Decomposition: A Case Study on PicoGPT
+
+- **ID:** `2603.28534` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Younes Javanmard, Tanmoy Pandit, Masoud Mardani
+- **abs:** https://arxiv.org/abs/2603.28534 · **PDF:** https://arxiv.org/pdf/2603.28534v1
+
+## 958. GraphWalker: Agentic Knowledge Graph Question Answering via Synthetic Trajectory Curriculum
+
+- **ID:** `2603.28533` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Shuwen Xu, Yao Xu, Jiaxiang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.28533 · **PDF:** https://arxiv.org/pdf/2603.28533v1
+
+## 959. Detecting low left ventricular ejection fraction from ECG using an interpretable and scalable predictor-driven framework
+
+- **ID:** `2603.28532` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Ya Zhou, Tianxiang Hao, Ziyi Cai, et al.
+- **abs:** https://arxiv.org/abs/2603.28532 · **PDF:** https://arxiv.org/pdf/2603.28532v1
+
+## 960. RAD-LAD: Rule and Language Grounded Autonomous Driving in Real-Time
+
+- **ID:** `2603.28522` · **primary:** `cs.RO` · **Submitted:** 2026-03-30
+- **Authors:** Anurag Ghosh, Srinivasa Narasimhan, Manmohan Chandraker, et al.
+- **abs:** https://arxiv.org/abs/2603.28522 · **PDF:** https://arxiv.org/pdf/2603.28522v2
+
+## 961. EarlySciRev: A Dataset of Early-Stage Scientific Revisions Extracted from LaTeX Writing Traces
+
+- **ID:** `2603.28515` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Léane Jourdan, Julien Aubert-Béduchaud, Yannis Chupin, et al.
+- **abs:** https://arxiv.org/abs/2603.28515 · **PDF:** https://arxiv.org/pdf/2603.28515v1
+
+## 962. TIEG-Youpu Solution for NeurIPS 2022 WikiKG90Mv2-LSC
+
+- **ID:** `2603.28512` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Feng Nie, Zhixiu Ye, Sifa Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.28512 · **PDF:** https://arxiv.org/pdf/2603.28512v1
+
+## 963. The Unreasonable Effectiveness of Scaling Laws in AI
+
+- **ID:** `2603.28507` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Chien-Ping Lu
+- **abs:** https://arxiv.org/abs/2603.28507 · **PDF:** https://arxiv.org/pdf/2603.28507v1
+
+## 964. GMA-SAWGAN-GP: A Novel Data Generative Framework to Enhance IDS Detection Performance
+
+- **ID:** `2603.28838` · **primary:** `cs.CR` · **Submitted:** 2026-03-30
+- **Authors:** Ziyu Mu, Xiyu Shi, Safak Dogan
+- **abs:** https://arxiv.org/abs/2603.28838 · **PDF:** https://arxiv.org/pdf/2603.28838v1
+
+## 965. Next-Token Prediction and Regret Minimization
+
+- **ID:** `2603.28499` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Mehryar Mohri, Clayton Sanford, Jon Schneider, et al.
+- **abs:** https://arxiv.org/abs/2603.28499 · **PDF:** https://arxiv.org/pdf/2603.28499v1
+
+## 966. MRI-to-CT synthesis using drifting models
+
+- **ID:** `2603.28498` · **primary:** `eess.IV` · **Submitted:** 2026-03-30
+- **Authors:** Qing Lyu, Jianxu Wang, Jeremy Hudson, et al.
+- **abs:** https://arxiv.org/abs/2603.28498 · **PDF:** https://arxiv.org/pdf/2603.28498v1
+
+## 967. Courtroom-Style Multi-Agent Debate with Progressive RAG and Role-Switching for Controversial Claim Verification
+
+- **ID:** `2603.28488` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Masnun Nuha Chowdhury, Nusrat Jahan Beg, Umme Hunny Khan, et al.
+- **abs:** https://arxiv.org/abs/2603.28488 · **PDF:** https://arxiv.org/pdf/2603.28488v1
+
+## 968. With a Little Help From My Friends: Collective Manipulation in Risk-Controlling Recommender Systems
+
+- **ID:** `2603.28476` · **primary:** `cs.IR` · **Submitted:** 2026-03-30
+- **Authors:** Giovanni De Toni, Cristian Consonni, Erasmo Purificato, et al.
+- **abs:** https://arxiv.org/abs/2603.28476 · **PDF:** https://arxiv.org/pdf/2603.28476v1
+
+## 969. CiQi-Agent: Aligning Vision, Tools and Aesthetics in Multimodal Agent for Cultural Reasoning on Chinese Porcelains
+
+- **ID:** `2603.28474` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Wenhan Wang, Zhixiang Zhou, Zhongtian Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.28474 · **PDF:** https://arxiv.org/pdf/2603.28474v1
+
+## 970. Post-hoc Self-explanation of CNNs
+
+- **ID:** `2603.28466` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Ahcène Boubekki, Line H. Clemmensen
+- **abs:** https://arxiv.org/abs/2603.28466 · **PDF:** https://arxiv.org/pdf/2603.28466v1
+
+## 971. $R_\text{dm}$: Re-conceptualizing Distribution Matching as a Reward for Diffusion Distillation
+
+- **ID:** `2603.28460` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Linqian Fan, Peiqin Sun, Tiancheng Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.28460 · **PDF:** https://arxiv.org/pdf/2603.28460v2
+
+## 972. HISA: Efficient Hierarchical Indexing for Fine-Grained Sparse Attention
+
+- **ID:** `2603.28458` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Yufei Xu, Fanxu Meng, Fan Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.28458 · **PDF:** https://arxiv.org/pdf/2603.28458v2
+
+## 973. FeDMRA: Federated Incremental Learning with Dynamic Memory Replay Allocation
+
+- **ID:** `2603.28455` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Tiantian Wang, Xiang Xiang, Simon S. Du
+- **abs:** https://arxiv.org/abs/2603.28455 · **PDF:** https://arxiv.org/pdf/2603.28455v1
+
+## 974. Yau's Affine Normal Descent: Algorithmic Framework and Convergence Analysis
+
+- **ID:** `2603.28448` · **primary:** `math.OC` · **Submitted:** 2026-03-30
+- **Authors:** Yi-Shuai Niu, Artan Sheshmani, Shing-Tung Yau
+- **abs:** https://arxiv.org/abs/2603.28448 · **PDF:** https://arxiv.org/pdf/2603.28448v1
+
+## 975. Entropic Claim Resolution: Uncertainty-Driven Evidence Selection for RAG
+
+- **ID:** `2603.28444` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Davide Di Gioia
+- **abs:** https://arxiv.org/abs/2603.28444 · **PDF:** https://arxiv.org/pdf/2603.28444v1
+
+## 976. LG-HCC: Local Geometry-Aware Hierarchical Context Compression for 3D Gaussian Splatting
+
+- **ID:** `2603.28431` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Xuan Deng, Xiandong Meng, Hengyu Man, et al.
+- **abs:** https://arxiv.org/abs/2603.28431 · **PDF:** https://arxiv.org/pdf/2603.28431v3
+
+## 977. IsoQuant: Hardware-Aligned SO(4) Isoclinic Rotations for LLM KV Cache Compression
+
+- **ID:** `2603.28430` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Zhongping Ji
+- **abs:** https://arxiv.org/abs/2603.28430 · **PDF:** https://arxiv.org/pdf/2603.28430v1
+
+## 978. AceleradorSNN: A Neuromorphic Cognitive System Integrating Spiking Neural Networks and DynamicImage Signal Processing on FPGA
+
+- **ID:** `2603.28429` · **primary:** `cs.AR` · **Submitted:** 2026-03-30
+- **Authors:** Daniel Gutierrez, Ruben Martinez, Leyre Arnedo, et al.
+- **abs:** https://arxiv.org/abs/2603.28429 · **PDF:** https://arxiv.org/pdf/2603.28429v1
+
+## 979. Structural-Ambiguity-Aware Translation from Natural Language to Signal Temporal Logic
+
+- **ID:** `2603.28426` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Kosei Fushimi, Kazunobu Serizawa, Junya Ikemoto, et al.
+- **abs:** https://arxiv.org/abs/2603.28426 · **PDF:** https://arxiv.org/pdf/2603.28426v1
+
+## 980. Profile Graphical Models
+
+- **ID:** `2603.28423` · **primary:** `stat.ME` · **Submitted:** 2026-03-30
+- **Authors:** Alejandra Avalos-Pacheco, Monia Lupparelli, Francesco C. Stingo
+- **abs:** https://arxiv.org/abs/2603.28423 · **PDF:** https://arxiv.org/pdf/2603.28423v1
+
+## 981. Learning unified control of internal spin squeezing in atomic qudits for magnetometry
+
+- **ID:** `2603.28421` · **primary:** `quant-ph` · **Submitted:** 2026-03-30
+- **Authors:** C. Z. Cao, J. Z. Han, M. Xiong, et al.
+- **abs:** https://arxiv.org/abs/2603.28421 · **PDF:** https://arxiv.org/pdf/2603.28421v1
+
+## 982. Spectral Higher-Order Neural Networks
+
+- **ID:** `2603.28420` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Gianluca Peri, Timoteo Carletti, Duccio Fanelli, et al.
+- **abs:** https://arxiv.org/abs/2603.28420 · **PDF:** https://arxiv.org/pdf/2603.28420v1
+
+## 983. LombardoGraphia: Automatic Classification of Lombard Orthography Variants
+
+- **ID:** `2603.28418` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Edoardo Signoroni, Pavel Rychlý
+- **abs:** https://arxiv.org/abs/2603.28418 · **PDF:** https://arxiv.org/pdf/2603.28418v1
+
+## 984. KGroups: A Versatile Univariate Max-Relevance Min-Redundancy Feature Selection Algorithm for High-dimensional Biological Data
+
+- **ID:** `2603.28417` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Malick Ebiele, Malika Bendechache, Rob Brennan
+- **abs:** https://arxiv.org/abs/2603.28417 · **PDF:** https://arxiv.org/pdf/2603.28417v1
+
+## 985. Evolutionary Discovery of Reinforcement Learning Algorithms via Large Language Models
+
+- **ID:** `2603.28416` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Alkis Sygkounas, Amy Loutfi, Andreas Persson
+- **abs:** https://arxiv.org/abs/2603.28416 · **PDF:** https://arxiv.org/pdf/2603.28416v1
+
+## 986. Mixture-Model Preference Learning for Many-Objective Bayesian Optimization
+
+- **ID:** `2603.28410` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Manisha Dubey, Sebastiaan De Peuter, Wanrong Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.28410 · **PDF:** https://arxiv.org/pdf/2603.28410v1
+
+## 987. MiroEval: Benchmarking Multimodal Deep Research Agents in Process and Outcome
+
+- **ID:** `2603.28407` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Fangda Ye, Yuxin Hu, Pengxiang Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.28407 · **PDF:** https://arxiv.org/pdf/2603.28407v1
+
+## 988. EdgeDiT: Hardware-Aware Diffusion Transformers for Efficient On-Device Image Generation
+
+- **ID:** `2603.28405` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Sravanth Kodavanti, Manjunath Arveti, Sowmya Vajrala, et al.
+- **abs:** https://arxiv.org/abs/2603.28405 · **PDF:** https://arxiv.org/pdf/2603.28405v1
+
+## 989. Label-efficient Training Updates for Malware Detection over Time
+
+- **ID:** `2603.28396` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Luca Minnei, Cristian Manca, Giorgio Piras, et al.
+- **abs:** https://arxiv.org/abs/2603.28396 · **PDF:** https://arxiv.org/pdf/2603.28396v1
+
+## 990. From Simulation to Deep Learning: Survey on Network Performance Modeling Approaches
+
+- **ID:** `2603.28394` · **primary:** `cs.NI` · **Submitted:** 2026-03-30
+- **Authors:** Carlos Güemes-Palau, Miquel Ferriol-Galmés, Jordi Paillisse-Vilanova, et al.
+- **abs:** https://arxiv.org/abs/2603.28394 · **PDF:** https://arxiv.org/pdf/2603.28394v1
+
+## 991. The Scaffold Effect: How Prompt Framing Drives Apparent Multimodal Gains in Clinical VLM Evaluation
+
+- **ID:** `2603.28387` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Doan Nam Long Vu, Simone Balloccu
+- **abs:** https://arxiv.org/abs/2603.28387 · **PDF:** https://arxiv.org/pdf/2603.28387v1
+
+## 992. COvolve: Adversarial Co-Evolution of Large-Language-Model-Generated Policies and Environments via Two-Player Zero-Sum Game
+
+- **ID:** `2603.28386` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Alkis Sygkounas, Rishi Hazra, Andreas Persson, et al.
+- **abs:** https://arxiv.org/abs/2603.28386 · **PDF:** https://arxiv.org/pdf/2603.28386v1
+
+## 993. Critic-Free Deep Reinforcement Learning for Maritime Coverage Path Planning on Irregular Hexagonal Grids
+
+- **ID:** `2603.28385` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Carlos S. Sepúlveda, Gonzalo A. Ruz
+- **abs:** https://arxiv.org/abs/2603.28385 · **PDF:** https://arxiv.org/pdf/2603.28385v1
+
+## 994. Membership Inference Attacks against Large Audio Language Models
+
+- **ID:** `2603.28378` · **primary:** `cs.SD` · **Submitted:** 2026-03-30
+- **Authors:** Jia-Kai Dong, Yu-Xiang Lin, Hung-Yi Lee
+- **abs:** https://arxiv.org/abs/2603.28378 · **PDF:** https://arxiv.org/pdf/2603.28378v1
+
+## 995. Marco DeepResearch: Unlocking Efficient Deep Research Agents via Verification-Centric Design
+
+- **ID:** `2603.28376` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Bin Zhu, Qianghuai Jia, Tian Lan, et al.
+- **abs:** https://arxiv.org/abs/2603.28376 · **PDF:** https://arxiv.org/pdf/2603.28376v1
+
+## 996. Coherent Without Grounding, Grounded Without Success: Observability and Epistemic Failure
+
+- **ID:** `2603.28371` · **primary:** `cs.CY` · **Submitted:** 2026-03-30
+- **Authors:** Camilo Chacón Sartori
+- **abs:** https://arxiv.org/abs/2603.28371 · **PDF:** https://arxiv.org/pdf/2603.28371v1
+
+## 997. Tailoring AI-Driven Reading Scaffolds to the Distinct Needs of Neurodiverse Learners
+
+- **ID:** `2603.28370` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Soufiane Jhilal, Eleonora Pasqua, Caterina Marchesi, et al.
+- **abs:** https://arxiv.org/abs/2603.28370 · **PDF:** https://arxiv.org/pdf/2603.28370v2
+
+## 998. Deep Research of Deep Research: From Transformer to Agent, From AI to AI for Science
+
+- **ID:** `2603.28361` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Yipeng Yu
+- **abs:** https://arxiv.org/abs/2603.28361 · **PDF:** https://arxiv.org/pdf/2603.28361v1
+
+## 999. CoE: Collaborative Entropy for Uncertainty Quantification in Agentic Multi-LLM Systems
+
+- **ID:** `2603.28360` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Kangkang Sun, Jun Wu, Jianhua Li, et al.
+- **abs:** https://arxiv.org/abs/2603.28360 · **PDF:** https://arxiv.org/pdf/2603.28360v1
+
+## 1000. The Conjugate Domain Dichotomy: Exact Risk of M-Estimators under Infinite-Variance Noise in High Dimensions
+
+- **ID:** `2603.28359` · **primary:** `math.ST` · **Submitted:** 2026-03-30
+- **Authors:** Charalampos Agiropoulos
+- **abs:** https://arxiv.org/abs/2603.28359 · **PDF:** https://arxiv.org/pdf/2603.28359v1
+
+## 1001. Optimized Weighted Voting System for Brain Tumor Classification Using MRI Images
+
+- **ID:** `2603.28357` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Ha Anh Vu
+- **abs:** https://arxiv.org/abs/2603.28357 · **PDF:** https://arxiv.org/pdf/2603.28357v1
+
+## 1002. Not All Subjectivity Is the Same! Defining Desiderata for the Evaluation of Subjectivity in NLP
+
+- **ID:** `2603.28351` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Urja Khurana, Michiel van der Meer, Enrico Liscio, et al.
+- **abs:** https://arxiv.org/abs/2603.28351 · **PDF:** https://arxiv.org/pdf/2603.28351v1
+
+## 1003. Machine Learning-Assisted High-Dimensional Matrix Estimation
+
+- **ID:** `2603.28346` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Wan Tian, Hui Yang, Zhouhui Lian, et al.
+- **abs:** https://arxiv.org/abs/2603.28346 · **PDF:** https://arxiv.org/pdf/2603.28346v1
+
+## 1004. Crossing the NL/PL Divide: Information Flow Analysis Across the NL/PL Boundary in LLM-Integrated Code
+
+- **ID:** `2603.28345` · **primary:** `cs.SE` · **Submitted:** 2026-03-30
+- **Authors:** Zihao Xu, Xiao Cheng, Ruijie Meng, et al.
+- **abs:** https://arxiv.org/abs/2603.28345 · **PDF:** https://arxiv.org/pdf/2603.28345v1
+
+## 1005. Kernel-Smith: A Unified Recipe for Evolutionary Kernel Optimization
+
+- **ID:** `2603.28342` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** He Du, Qiming Ge, Jiakai Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.28342 · **PDF:** https://arxiv.org/pdf/2603.28342v1
+
+## 1006. A Multi-Agent Rhizomatic Pipeline for Non-Linear Literature Analysis
+
+- **ID:** `2603.28336` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Julio C. Serrano, Joonas Kevari, Rumy Narayan
+- **abs:** https://arxiv.org/abs/2603.28336 · **PDF:** https://arxiv.org/pdf/2603.28336v2
+
+## 1007. Key-Embedded Privacy for Decentralized AI in Biomedical Omics
+
+- **ID:** `2603.28334` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Rongyu Zhang, Hongyu Dong, Gaole Dai, et al.
+- **abs:** https://arxiv.org/abs/2603.28334 · **PDF:** https://arxiv.org/pdf/2603.28334v1
+
+## 1008. Integrating Multimodal Large Language Model Knowledge into Amodal Completion
+
+- **ID:** `2603.28333` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Heecheol Yun, Eunho Yang
+- **abs:** https://arxiv.org/abs/2603.28333 · **PDF:** https://arxiv.org/pdf/2603.28333v1
+
+## 1009. Physics-Informed Neural Networks for Predicting Hydrogen Sorption in Geological Formations: Thermodynamically Constrained Deep Learning Integrating Classical Adsorption Theory
+
+- **ID:** `2603.28328` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Mohammad Nooraiepour, Mohammad Masoudi, Zezhang Song, et al.
+- **abs:** https://arxiv.org/abs/2603.28328 · **PDF:** https://arxiv.org/pdf/2603.28328v1
+
+## 1010. Building evidence-based knowledge graphs from full-text literature for disease-specific biomedical reasoning
+
+- **ID:** `2603.28325` · **primary:** `cs.CE` · **Submitted:** 2026-03-30
+- **Authors:** Chang Zong, Sicheng Lv, Si-tu Xue, et al.
+- **abs:** https://arxiv.org/abs/2603.28325 · **PDF:** https://arxiv.org/pdf/2603.28325v2
+
+## 1011. LDDMM stochastic interpolants: an application to domain uncertainty quantification in hemodynamics
+
+- **ID:** `2603.28324` · **primary:** `stat.ML` · **Submitted:** 2026-03-30
+- **Authors:** Sarah Katz, Francesco Romor, Jia-Jie Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.28324 · **PDF:** https://arxiv.org/pdf/2603.28324v1
+
+## 1012. FairGC: Fairness-aware Graph Condensation
+
+- **ID:** `2603.28321` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Yihan Gao, Chenxi Huang, Wen Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.28321 · **PDF:** https://arxiv.org/pdf/2603.28321v1
+
+## 1013. Mapping data literacy trajectories in K-12 education
+
+- **ID:** `2603.28317` · **primary:** `cs.CY` · **Submitted:** 2026-03-30
+- **Authors:** Robert Whyte, Manni Cheung, Katharine Childs, et al.
+- **abs:** https://arxiv.org/abs/2603.28317 · **PDF:** https://arxiv.org/pdf/2603.28317v1
+
+## 1014. Taming the Instability: A Robust Second-Order Optimizer for Federated Learning over Non-IID Data
+
+- **ID:** `2603.28316` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Yuanqiao Zhang, Tiantian He, Yuan Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.28316 · **PDF:** https://arxiv.org/pdf/2603.28316v1
+
+## 1015. Prototype-Enhanced Multi-View Learning for Thyroid Nodule Ultrasound Classification
+
+- **ID:** `2603.28315` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Yangmei Chen, Zhongyuan Zhang, Xikun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.28315 · **PDF:** https://arxiv.org/pdf/2603.28315v1
+
+## 1016. Self++: Co-Determined Agency for Human--AI Symbiosis in Extended Reality
+
+- **ID:** `2603.28306` · **primary:** `cs.HC` · **Submitted:** 2026-03-30
+- **Authors:** Thammathip Piumsomboon
+- **abs:** https://arxiv.org/abs/2603.28306 · **PDF:** https://arxiv.org/pdf/2603.28306v1
+
+## 1017. The Necessity of Setting Temperature in LLM-as-a-Judge
+
+- **ID:** `2603.28304` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Lujun Li, Lama Sleem, Yangjie Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.28304 · **PDF:** https://arxiv.org/pdf/2603.28304v1
+
+## 1018. LIBERO-Para: A Diagnostic Benchmark and Metrics for Paraphrase Robustness in VLA Models
+
+- **ID:** `2603.28301` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Chanyoung Kim, Minwoo Kim, Minseok Kang, et al.
+- **abs:** https://arxiv.org/abs/2603.28301 · **PDF:** https://arxiv.org/pdf/2603.28301v1
+
+## 1019. NeiGAD: Augmenting Graph Anomaly Detection via Spectral Neighbor Information
+
+- **ID:** `2603.28300` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Qing Qing, Huafei Huang, Mingliang Hou, et al.
+- **abs:** https://arxiv.org/abs/2603.28300 · **PDF:** https://arxiv.org/pdf/2603.28300v1
+
+## 1020. Evaluating LLMs for Answering Student Questions in Introductory Programming Courses
+
+- **ID:** `2603.28295` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Thomas Van Mullem, Bart Mesuere, Peter Dawyndt
+- **abs:** https://arxiv.org/abs/2603.28295 · **PDF:** https://arxiv.org/pdf/2603.28295v1
+
+## 1021. Learning from imperfect quantum data via unsupervised domain adaptation with classical shadows
+
+- **ID:** `2603.28294` · **primary:** `quant-ph` · **Submitted:** 2026-03-30
+- **Authors:** Kosuke Ito, Akira Tanji, Hiroshi Yano, et al.
+- **abs:** https://arxiv.org/abs/2603.28294 · **PDF:** https://arxiv.org/pdf/2603.28294v1
+
+## 1022. OptINC: Optical In-Network-Computing for Scalable Distributed Learning
+
+- **ID:** `2603.28290` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Sijie Fei, Grace Li Zhang, Bing Li, et al.
+- **abs:** https://arxiv.org/abs/2603.28290 · **PDF:** https://arxiv.org/pdf/2603.28290v1
+
+## 1023. FI-KAN: Fractal Interpolation Kolmogorov-Arnold Networks
+
+- **ID:** `2603.28288` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Gnankan Landry Regis N'guessan
+- **abs:** https://arxiv.org/abs/2603.28288 · **PDF:** https://arxiv.org/pdf/2603.28288v1
+
+## 1024. Whittaker-Henderson smoother for long satellite image time series interpolation
+
+- **ID:** `2604.00048` · **primary:** `eess.IV` · **Submitted:** 2026-03-30
+- **Authors:** Mathieu Fauvel
+- **abs:** https://arxiv.org/abs/2604.00048 · **PDF:** https://arxiv.org/pdf/2604.00048v1
+
+## 1025. Pre-Deployment Complexity Estimation for Federated Perception Systems
+
+- **ID:** `2603.28282` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** KMA Solaiman, Shafkat Islam, Ruy de Oliveira, et al.
+- **abs:** https://arxiv.org/abs/2603.28282 · **PDF:** https://arxiv.org/pdf/2603.28282v1
+
+## 1026. Corruption-robust Offline Multi-agent Reinforcement Learning From Human Feedback
+
+- **ID:** `2603.28281` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Andi Nika, Debmalya Mandal, Parameswaran Kamalaruban, et al.
+- **abs:** https://arxiv.org/abs/2603.28281 · **PDF:** https://arxiv.org/pdf/2603.28281v1
+
+## 1027. Merge and Conquer: Instructing Multilingual Models by Adding Target Language Weights
+
+- **ID:** `2603.28263` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Eneko Valero, Maria Ribalta i Albado, Oscar Sainz, et al.
+- **abs:** https://arxiv.org/abs/2603.28263 · **PDF:** https://arxiv.org/pdf/2603.28263v1
+
+## 1028. Coconstructions in spoken data: UD annotation guidelines and first results
+
+- **ID:** `2603.28261` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Ludovica Pannitto, Sylvain Kahane, Kaja Dobrovoljc, et al.
+- **abs:** https://arxiv.org/abs/2603.28261 · **PDF:** https://arxiv.org/pdf/2603.28261v1
+
+## 1029. Categorical Perception in Large Language Model Hidden States: Structural Warping at Digit-Count Boundaries
+
+- **ID:** `2603.28258` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Jon-Paul Cacioli
+- **abs:** https://arxiv.org/abs/2603.28258 · **PDF:** https://arxiv.org/pdf/2603.28258v1
+
+## 1030. Nonlinear Factor Decomposition via Kolmogorov-Arnold Networks: A Spectral Approach to Asset Return Analysis
+
+- **ID:** `2603.28257` · **primary:** `q-fin.ST` · **Submitted:** 2026-03-30
+- **Authors:** David Breazu
+- **abs:** https://arxiv.org/abs/2603.28257 · **PDF:** https://arxiv.org/pdf/2603.28257v1
+
+## 1031. Framework for identifying the equivalence between Nature-Inspired Metaheuristics
+
+- **ID:** `2603.28255` · **primary:** `cs.NE` · **Submitted:** 2026-03-30
+- **Authors:** Iztok Fister, Žan Hozjan, Iztok Fister,, et al.
+- **abs:** https://arxiv.org/abs/2603.28255 · **PDF:** https://arxiv.org/pdf/2603.28255v1
+
+## 1032. MuonEq: Balancing Before Orthogonalization with Lightweight Equilibration
+
+- **ID:** `2603.28254` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Da Chang, Qiankun Shi, Lvgang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.28254 · **PDF:** https://arxiv.org/pdf/2603.28254v1
+
+## 1033. MR-ImagenTime: Multi-Resolution Time Series Generation through Dual Image Representations
+
+- **ID:** `2603.28253` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Xianyong Xu, Yuanjun Zuo, Zhihong Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.28253 · **PDF:** https://arxiv.org/pdf/2603.28253v1
+
+## 1034. DiffAttn: Diffusion-Based Drivers' Visual Attention Prediction with LLM-Enhanced Semantic Reasoning
+
+- **ID:** `2603.28251` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Weimin Liu, Qingkun Li, Jiyuan Qiu, et al.
+- **abs:** https://arxiv.org/abs/2603.28251 · **PDF:** https://arxiv.org/pdf/2603.28251v1
+
+## 1035. Reasoning as Energy Minimization over Structured Latent Trajectories
+
+- **ID:** `2603.28248` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** David K. Johansson
+- **abs:** https://arxiv.org/abs/2603.28248 · **PDF:** https://arxiv.org/pdf/2603.28248v1
+
+## 1036. TwinMixing: A Shuffle-Aware Feature Interaction Model for Multi-Task Segmentation
+
+- **ID:** `2603.28233` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Minh-Khoi Do, Huy Che, Dinh-Duy Phan, et al.
+- **abs:** https://arxiv.org/abs/2603.28233 · **PDF:** https://arxiv.org/pdf/2603.28233v1
+
+## 1037. Detecting the Unexpected: AI-Driven Anomaly Detection in Smart Bridge Monitoring
+
+- **ID:** `2603.28225` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Rahul Jaiswal, Joakim Hellum, Halvor Heiberg
+- **abs:** https://arxiv.org/abs/2603.28225 · **PDF:** https://arxiv.org/pdf/2603.28225v1
+
+## 1038. Variational Neurons in Transformers for Language Modeling
+
+- **ID:** `2603.28219` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Yves Ruffenach
+- **abs:** https://arxiv.org/abs/2603.28219 · **PDF:** https://arxiv.org/pdf/2603.28219v1
+
+## 1039. An Optimal Battery-Free Approach for Emission Reduction by Storing Solar Surplus in Building Thermal Mass
+
+- **ID:** `2603.28217` · **primary:** `eess.SY` · **Submitted:** 2026-03-30
+- **Authors:** Michela Boffi, Jessica Leoni, Fabrizio Leonforte, et al.
+- **abs:** https://arxiv.org/abs/2603.28217 · **PDF:** https://arxiv.org/pdf/2603.28217v1
+
+## 1040. \textit{Versteasch du mi?} Computational and Socio-Linguistic Perspectives on GenAI, LLMs, and Non-Standard Language
+
+- **ID:** `2603.28213` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Verena Platzgummer, John McCrae, Sina Ahmadi
+- **abs:** https://arxiv.org/abs/2603.28213 · **PDF:** https://arxiv.org/pdf/2603.28213v1
+
+## 1041. Beyond Cosine Similarity: Zero-Initialized Residual Complex Projection for Aspect-Based Sentiment Analysis
+
+- **ID:** `2603.28205` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Yijin Wang, Fandi Sun
+- **abs:** https://arxiv.org/abs/2603.28205 · **PDF:** https://arxiv.org/pdf/2603.28205v1
+
+## 1042. ERPO: Token-Level Entropy-Regulated Policy Optimization for Large Reasoning Models
+
+- **ID:** `2603.28204` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Song Yu, Li Li
+- **abs:** https://arxiv.org/abs/2603.28204 · **PDF:** https://arxiv.org/pdf/2603.28204v1
+
+## 1043. Differentiable Power-Flow Optimization
+
+- **ID:** `2603.28203` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Muhammed Öz, Jasmin Hörter, Kaleb Phipps, et al.
+- **abs:** https://arxiv.org/abs/2603.28203 · **PDF:** https://arxiv.org/pdf/2603.28203v1
+
+## 1044. A Perturbation Approach to Unconstrained Linear Bandits
+
+- **ID:** `2603.28201` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Andrew Jacobsen, Dorian Baudry, Shinji Ito, et al.
+- **abs:** https://arxiv.org/abs/2603.28201 · **PDF:** https://arxiv.org/pdf/2603.28201v1
+
+## 1045. A Deep Reinforcement Learning Framework for Closed-loop Guidance of Fish Schools via Virtual Agents
+
+- **ID:** `2603.28200` · **primary:** `cs.RO` · **Submitted:** 2026-03-30
+- **Authors:** Takato Shibayama, Hiroaki Kawashima
+- **abs:** https://arxiv.org/abs/2603.28200 · **PDF:** https://arxiv.org/pdf/2603.28200v1
+
+## 1046. Policy-Controlled Generalized Share: A General Framework with a Transformer Instantiation for Strictly Online Switching-Oracle Tracking
+
+- **ID:** `2603.28198` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Hongkai Hu
+- **abs:** https://arxiv.org/abs/2603.28198 · **PDF:** https://arxiv.org/pdf/2603.28198v1
+
+## 1047. EpiPersona: Persona Projection and Episode Coupling for Pluralistic Preference Modeling
+
+- **ID:** `2603.28197` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Yujie Zhang, Weikang Yuan, Zhuoren Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.28197 · **PDF:** https://arxiv.org/pdf/2603.28197v1
+
+## 1048. Designing AI for Real Users -- Accessibility Gaps in Retail AI Front-End
+
+- **ID:** `2603.28196` · **primary:** `cs.HC` · **Submitted:** 2026-03-30
+- **Authors:** Neha Puri, Tim Dixon
+- **abs:** https://arxiv.org/abs/2603.28196 · **PDF:** https://arxiv.org/pdf/2603.28196v1
+
+## 1049. DongYuan: An LLM-Based Framework for Integrative Chinese and Western Medicine Spleen-Stomach Disorders Diagnosis
+
+- **ID:** `2603.28191` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Hua Li, Yingying Li, Xiaobin Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.28191 · **PDF:** https://arxiv.org/pdf/2603.28191v1
+
+## 1050. PReD: An LLM-based Foundation Multimodal Model for Electromagnetic Perception, Recognition, and Decision
+
+- **ID:** `2603.28183` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Zehua Han, Jing Xiao, Yiqi Duan, et al.
+- **abs:** https://arxiv.org/abs/2603.28183 · **PDF:** https://arxiv.org/pdf/2603.28183v2
+
+## 1051. Skillful Kilometer-Scale Regional Weather Forecasting via Global and Regional Coupling
+
+- **ID:** `2603.28173` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Weiqi Chen, Wenwei Wang, Qilong Yuan, et al.
+- **abs:** https://arxiv.org/abs/2603.28173 · **PDF:** https://arxiv.org/pdf/2603.28173v1
+
+## 1052. Automating Early Disease Prediction Via Structured and Unstructured Clinical Data
+
+- **ID:** `2603.28167` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Ane G Domingo-Aldama, Marcos Merino Prado, Alain García Olea, et al.
+- **abs:** https://arxiv.org/abs/2603.28167 · **PDF:** https://arxiv.org/pdf/2603.28167v1
+
+## 1053. Evaluating Privilege Usage of Agents on Real-World Tools
+
+- **ID:** `2603.28166` · **primary:** `cs.CR` · **Submitted:** 2026-03-30
+- **Authors:** Quan Zhang, Lianhang Fu, Lvsi Lian, et al.
+- **abs:** https://arxiv.org/abs/2603.28166 · **PDF:** https://arxiv.org/pdf/2603.28166v1
+
+## 1054. From Reviews to Requirements: Can LLMs Generate Human-Like User Stories?
+
+- **ID:** `2603.28163` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Shadman Sakib, Oishy Fatema Akhand, Tasnia Tasneem, et al.
+- **abs:** https://arxiv.org/abs/2603.28163 · **PDF:** https://arxiv.org/pdf/2603.28163v1
+
+## 1055. Evolutionary Algorithms for Generating Graphs Matching Desired Laplacian Spectra
+
+- **ID:** `2603.28151` · **primary:** `cs.NE` · **Submitted:** 2026-03-30
+- **Authors:** Hendrik Richter, Frank Neumann
+- **abs:** https://arxiv.org/abs/2603.28151 · **PDF:** https://arxiv.org/pdf/2603.28151v1
+
+## 1056. RecycleLoRA: Rank-Revealing QR-Based Dual-LoRA Subspace Adaptation for Domain Generalized Semantic Segmentation
+
+- **ID:** `2603.28142` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Chanseul Cho, Seokju Yun, Jeaseong Jeon, et al.
+- **abs:** https://arxiv.org/abs/2603.28142 · **PDF:** https://arxiv.org/pdf/2603.28142v1
+
+## 1057. Intelligent Road Condition Monitoring using 3D In-Air SONAR Sensing
+
+- **ID:** `2603.28141` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Amber Cassimon, Robin Kerstens, Walter Daems, et al.
+- **abs:** https://arxiv.org/abs/2603.28141 · **PDF:** https://arxiv.org/pdf/2603.28141v1
+
+## 1058. CoT2-Meta: Budgeted Metacognitive Control for Test-Time Reasoning
+
+- **ID:** `2603.28135` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Siyuan Ma, Bo Gao, Zikai Xiao, et al.
+- **abs:** https://arxiv.org/abs/2603.28135 · **PDF:** https://arxiv.org/pdf/2603.28135v1
+
+## 1059. MDPBench: A Benchmark for Multilingual Document Parsing in Real-World Scenarios
+
+- **ID:** `2603.28130` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Zhang Li, Zhibo Lin, Qiang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.28130 · **PDF:** https://arxiv.org/pdf/2603.28130v1
+
+## 1060. ORACAL: A Robust and Explainable Multimodal Framework for Smart Contract Vulnerability Detection with Causal Graph Enrichment
+
+- **ID:** `2603.28128` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Tran Duong Minh Dai, Triet Huynh Minh Le, M. Ali Babar, et al.
+- **abs:** https://arxiv.org/abs/2603.28128 · **PDF:** https://arxiv.org/pdf/2603.28128v1
+
+## 1061. Does Claude's Constitution Have a Culture?
+
+- **ID:** `2603.28123` · **primary:** `cs.CY` · **Submitted:** 2026-03-30
+- **Authors:** Parham Pourdavood
+- **abs:** https://arxiv.org/abs/2603.28123 · **PDF:** https://arxiv.org/pdf/2603.28123v1
+
+## 1062. Q-DIVER: Integrated Quantum Transfer Learning and Differentiable Quantum Architecture Search with EEG Data
+
+- **ID:** `2603.28122` · **primary:** `quant-ph` · **Submitted:** 2026-03-30
+- **Authors:** Junghoon Justin Park, Yeonghyeon Park, Jiook Cha
+- **abs:** https://arxiv.org/abs/2603.28122 · **PDF:** https://arxiv.org/pdf/2603.28122v1
+
+## 1063. Neural Federated Learning for Livestock Growth Prediction
+
+- **ID:** `2603.28117` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Shoujin Wang, Mingze Ni, Wei Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.28117 · **PDF:** https://arxiv.org/pdf/2603.28117v3
+
+## 1064. Graph Vector Field: A Unified Framework for Multimodal Health Risk Assessment from Heterogeneous Wearable and Environmental Data Streams
+
+- **ID:** `2603.28115` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Silvano Coletti, Francesca Fallucchi
+- **abs:** https://arxiv.org/abs/2603.28115 · **PDF:** https://arxiv.org/pdf/2603.28115v1
+
+## 1065. Attention Frequency Modulation: Training-Free Spectral Modulation of Diffusion Cross-Attention
+
+- **ID:** `2603.28114` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Seunghun Oh, Unsang Park
+- **abs:** https://arxiv.org/abs/2603.28114 · **PDF:** https://arxiv.org/pdf/2603.28114v1
+
+## 1066. Lipschitz verification of neural networks through training
+
+- **ID:** `2603.28113` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Simon Kuang, Yuezhu Xu, S. Sivaranjani, et al.
+- **abs:** https://arxiv.org/abs/2603.28113 · **PDF:** https://arxiv.org/pdf/2603.28113v1
+
+## 1067. Quid est VERITAS? A Modular Framework for Archival Document Analysis
+
+- **ID:** `2603.28108` · **primary:** `cs.DL` · **Submitted:** 2026-03-30
+- **Authors:** Leonardo Bassanini, Ludovico Biancardi, Alfio Ferrara, et al.
+- **abs:** https://arxiv.org/abs/2603.28108 · **PDF:** https://arxiv.org/pdf/2603.28108v1
+
+## 1068. Transcription and Recognition of Italian Parliamentary Speeches Using Vision-Language Models
+
+- **ID:** `2603.28103` · **primary:** `cs.DL` · **Submitted:** 2026-03-30
+- **Authors:** Luigi Curini, Alfio Ferrara, Giovanni Pagano, et al.
+- **abs:** https://arxiv.org/abs/2603.28103 · **PDF:** https://arxiv.org/pdf/2603.28103v1
+
+## 1069. Heddle: A Distributed Orchestration System for Agentic RL Rollout
+
+- **ID:** `2603.28101` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Zili Zhang, Yinmin Zhong, Chengxu Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.28101 · **PDF:** https://arxiv.org/pdf/2603.28101v1
+
+## 1070. InkDrop: Invisible Backdoor Attacks Against Dataset Condensation
+
+- **ID:** `2603.28092` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** He Yang, Dongyi Lv, Song Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.28092 · **PDF:** https://arxiv.org/pdf/2603.28092v1
+
+## 1071. MOSS-VoiceGenerator: Create Realistic Voices with Natural Language Descriptions
+
+- **ID:** `2603.28086` · **primary:** `cs.SD` · **Submitted:** 2026-03-30
+- **Authors:** Kexin Huang, Liwei Fan, Botian Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.28086 · **PDF:** https://arxiv.org/pdf/2603.28086v1
+
+## 1072. Transformer-Based Prognostics: Enhancing Network Availability by Improved Monitoring of Optical Fiber Amplifiers
+
+- **ID:** `2603.28081` · **primary:** `eess.SP` · **Submitted:** 2026-03-30
+- **Authors:** Dominic Schneider, Lutz Rapp, Christoph Ament
+- **abs:** https://arxiv.org/abs/2603.28081 · **PDF:** https://arxiv.org/pdf/2603.28081v1
+
+## 1073. Koopman-based surrogate modeling for reinforcement-learning-control of Rayleigh-Benard convection
+
+- **ID:** `2603.28074` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Tim Plotzki, Sebastian Peitz
+- **abs:** https://arxiv.org/abs/2603.28074 · **PDF:** https://arxiv.org/pdf/2603.28074v1
+
+## 1074. SIMR-NO: A Spectrally-Informed Multi-Resolution Neural Operator for Turbulent Flow Super-Resolution
+
+- **ID:** `2603.28073` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Muhammad Abid, Omer San
+- **abs:** https://arxiv.org/abs/2603.28073 · **PDF:** https://arxiv.org/pdf/2603.28073v1
+
+## 1075. MolmoPoint: Better Pointing for VLMs with Grounding Tokens
+
+- **ID:** `2603.28069` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Christopher Clark, Yue Yang, Jae Sung Park, et al.
+- **abs:** https://arxiv.org/abs/2603.28069 · **PDF:** https://arxiv.org/pdf/2603.28069v1
+
+## 1076. From Vessel Trajectories to Safety-Critical Encounter Scenarios: A Generative AI Framework for Autonomous Ship Digital Testing
+
+- **ID:** `2603.28067` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Sijin Sun, Liangbin Zhao, Ming Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.28067 · **PDF:** https://arxiv.org/pdf/2603.28067v1
+
+## 1077. Synonymix: Unified Group Personas for Generative Simulations
+
+- **ID:** `2603.28066` · **primary:** `cs.HC` · **Submitted:** 2026-03-30
+- **Authors:** Huanxing Chen, Aditesh Kumar
+- **abs:** https://arxiv.org/abs/2603.28066 · **PDF:** https://arxiv.org/pdf/2603.28066v1
+
+## 1078. Reward Hacking as Equilibrium under Finite Evaluation
+
+- **ID:** `2603.28063` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Jiacheng Wang, Jinbin Huang
+- **abs:** https://arxiv.org/abs/2603.28063 · **PDF:** https://arxiv.org/pdf/2603.28063v1
+
+## 1079. SLOW: Strategic Logical-inference Open Workspace for Cognitive Adaptation in AI Tutoring
+
+- **ID:** `2603.28062` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Yuang Wei, Ruijia Li, Bo Jiang
+- **abs:** https://arxiv.org/abs/2603.28062 · **PDF:** https://arxiv.org/pdf/2603.28062v1
+
+## 1080. Physics-Embedded Feature Learning for AI in Medical Imaging
+
+- **ID:** `2603.28057` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Pulock Das, Al Amin, Kamrul Hasan, et al.
+- **abs:** https://arxiv.org/abs/2603.28057 · **PDF:** https://arxiv.org/pdf/2603.28057v1
+
+## 1081. Who Wrote the Book? Detecting and Attributing LLM Ghostwriters
+
+- **ID:** `2603.28054` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Anudeex Shetty, Qiongkai Xu, Olga Ohrimenko, et al.
+- **abs:** https://arxiv.org/abs/2603.28054 · **PDF:** https://arxiv.org/pdf/2603.28054v1
+
+## 1082. Reducing Oracle Feedback with Vision-Language Embeddings for Preference-Based RL
+
+- **ID:** `2603.28053` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Udita Ghosh, Dripta S. Raychaudhuri, Jiachen Li, et al.
+- **abs:** https://arxiv.org/abs/2603.28053 · **PDF:** https://arxiv.org/pdf/2603.28053v1
+
+## 1083. Meta-Harness: End-to-End Optimization of Model Harnesses
+
+- **ID:** `2603.28052` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Yoonho Lee, Roshen Nair, Qizheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.28052 · **PDF:** https://arxiv.org/pdf/2603.28052v1
+
+## 1084. Dogfight Search: A Swarm-Based Optimization Algorithm for Complex Engineering Optimization and Mountainous Terrain Path Planning
+
+- **ID:** `2603.28046` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Yujing Sun, Jie Cai, Xingguo Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.28046 · **PDF:** https://arxiv.org/pdf/2603.28046v1
+
+## 1085. Bit-Identical Medical Deep Learning via Structured Orthogonal Initialization
+
+- **ID:** `2603.28040` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Yakov Pyotr Shkolnikov
+- **abs:** https://arxiv.org/abs/2603.28040 · **PDF:** https://arxiv.org/pdf/2603.28040v1
+
+## 1086. Beyond the Answer: Decoding the Behavior of LLMs as Scientific Reasoners
+
+- **ID:** `2603.28038` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Rohan Pandey, Eric Ye, Michael Li
+- **abs:** https://arxiv.org/abs/2603.28038 · **PDF:** https://arxiv.org/pdf/2603.28038v1
+
+## 1087. Diffusion Maps is not Dimensionality Reduction
+
+- **ID:** `2603.28037` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Julio Candanedo, Alejandro Patiño
+- **abs:** https://arxiv.org/abs/2603.28037 · **PDF:** https://arxiv.org/pdf/2603.28037v1
+
+## 1088. Transfer Learning for an Endangered Slavic Variety: Dependency Parsing in Pomak Across Contact-Shaped Dialects
+
+- **ID:** `2603.28033` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Sercan Karakaş
+- **abs:** https://arxiv.org/abs/2603.28033 · **PDF:** https://arxiv.org/pdf/2603.28033v1
+
+## 1089. CARLA-Air: Fly Drones Inside a CARLA World -- A Unified Infrastructure for Air-Ground Embodied Intelligence
+
+- **ID:** `2603.28032` · **primary:** `cs.RO` · **Submitted:** 2026-03-30
+- **Authors:** Tianle Zeng, Hanxuan Chen, Yanci Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.28032 · **PDF:** https://arxiv.org/pdf/2603.28032v1
+
+## 1090. Efficient Domain Adaptation for Text Line Recognition via Decoupled Language Models
+
+- **ID:** `2603.28028` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Arundhathi Dev, Justin Zhan
+- **abs:** https://arxiv.org/abs/2603.28028 · **PDF:** https://arxiv.org/pdf/2603.28028v1
+
+## 1091. When Choices Become Priors: Contrastive Decoding for Scientific Figure Multiple-Choice QA
+
+- **ID:** `2603.28026` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Taeyun Roh, Eun-yeong Jo, Wonjune Jang, et al.
+- **abs:** https://arxiv.org/abs/2603.28026 · **PDF:** https://arxiv.org/pdf/2603.28026v1
+
+## 1092. What an Autonomous Agent Discovers About Molecular Transformer Design: Does It Transfer?
+
+- **ID:** `2603.28015` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Edward Wijaya
+- **abs:** https://arxiv.org/abs/2603.28015 · **PDF:** https://arxiv.org/pdf/2603.28015v1
+
+## 1093. Kill-Chain Canaries: Stage-Level Tracking of Prompt Injection Across Attack Surfaces and Model Safety Tiers
+
+- **ID:** `2603.28013` · **primary:** `cs.CR` · **Submitted:** 2026-03-30
+- **Authors:** Haochuan Kevin Wang
+- **abs:** https://arxiv.org/abs/2603.28013 · **PDF:** https://arxiv.org/pdf/2603.28013v1
+
+## 1094. HeteroHub: An Applicable Data Management Framework for Heterogeneous Multi-Embodied Agent System
+
+- **ID:** `2603.28010` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Xujia Li, Xin Li, Junquan Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.28010 · **PDF:** https://arxiv.org/pdf/2603.28010v1
+
+## 1095. FedDES: Graph-Based Dynamic Ensemble Selection for Personalized Federated Learning
+
+- **ID:** `2603.28006` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Brianna Mueller, W. Nick Street
+- **abs:** https://arxiv.org/abs/2603.28006 · **PDF:** https://arxiv.org/pdf/2603.28006v1
+
+## 1096. Rethinking Atomic Decomposition for LLM Judges: A Prompt-Controlled Study of Reference-Grounded QA Evaluation
+
+- **ID:** `2603.28005` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Xinran Zhang
+- **abs:** https://arxiv.org/abs/2603.28005 · **PDF:** https://arxiv.org/pdf/2603.28005v1
+
+## 1097. BiFormer3D: Grid-Free Time-Domain Reconstruction of Head-Related Impulse Responses with a Spatially Encoded Transformer
+
+- **ID:** `2603.27998` · **primary:** `eess.AS` · **Submitted:** 2026-03-30
+- **Authors:** Shaoheng Xu, Chunyi Sun, Jihui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.27998 · **PDF:** https://arxiv.org/pdf/2603.27998v1
+
+## 1098. From Independent to Correlated Diffusion: Generalized Generative Modeling with Probabilistic Computers
+
+- **ID:** `2603.27996` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Nihal Sanjay Singh, Mazdak Mohseni-Rajaee, Shaila Niazi, et al.
+- **abs:** https://arxiv.org/abs/2603.27996 · **PDF:** https://arxiv.org/pdf/2603.27996v1
+
+## 1099. ViviDoc: Generating Interactive Documents through Human-Agent Collaboration
+
+- **ID:** `2603.27991` · **primary:** `cs.HC` · **Submitted:** 2026-03-30
+- **Authors:** Yinghao Tang, Yupeng Xie, Yingchaojie Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.27991 · **PDF:** https://arxiv.org/pdf/2603.27991v1
+
+## 1100. Beyond Dataset Distillation: Lossless Dataset Concentration via Diffusion-Assisted Distribution Alignment
+
+- **ID:** `2603.27987` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Tongfei Liu, Yufan Liu, Bing Li, et al.
+- **abs:** https://arxiv.org/abs/2603.27987 · **PDF:** https://arxiv.org/pdf/2603.27987v1
+
+## 1101. FedFG: Privacy-Preserving and Robust Federated Learning via Flow-Matching Generation
+
+- **ID:** `2603.27986` · **primary:** `cs.CR` · **Submitted:** 2026-03-30
+- **Authors:** Ruiyang Wang, Rong Pan, Zhengan Yao
+- **abs:** https://arxiv.org/abs/2603.27986 · **PDF:** https://arxiv.org/pdf/2603.27986v1
+
+## 1102. CDH-Bench: A Commonsense-Driven Hallucination Benchmark for Evaluating Visual Fidelity in Vision-Language Models
+
+- **ID:** `2603.27982` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Kesheng Chen, Yamin Hu, Qi Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.27982 · **PDF:** https://arxiv.org/pdf/2603.27982v2
+
+## 1103. On the Role of Encoder Depth: Pruning Whisper and LoRA Fine-Tuning in SLAM-ASR
+
+- **ID:** `2603.27981` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Ganesh Pavan Kartikeya Bharadwaj Kolluri, Michael Kampouridis, Ravi Shekhar
+- **abs:** https://arxiv.org/abs/2603.27981 · **PDF:** https://arxiv.org/pdf/2603.27981v1
+
+## 1104. SARL: Label-Free Reinforcement Learning by Rewarding Reasoning Topology
+
+- **ID:** `2603.27977` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Yifan Wang, Bolian Li, David Cho, et al.
+- **abs:** https://arxiv.org/abs/2603.27977 · **PDF:** https://arxiv.org/pdf/2603.27977v1
+
+## 1105. Principal Prototype Analysis on Manifold for Interpretable Reinforcement Learning
+
+- **ID:** `2603.27971` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Bodla Krishna Vamshi, Haizhao Yang
+- **abs:** https://arxiv.org/abs/2603.27971 · **PDF:** https://arxiv.org/pdf/2603.27971v2
+
+## 1106. Gradient Manipulation in Distributed Stochastic Gradient Descent with Strategic Agents: Truthful Incentives with Convergence Guarantees
+
+- **ID:** `2603.27962` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Ziqin Chen, Yongqiang Wang
+- **abs:** https://arxiv.org/abs/2603.27962 · **PDF:** https://arxiv.org/pdf/2603.27962v1
+
+## 1107. Efficient Inference of Large Vision Language Models
+
+- **ID:** `2603.27960` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Surendra Pathak
+- **abs:** https://arxiv.org/abs/2603.27960 · **PDF:** https://arxiv.org/pdf/2603.27960v1
+
+## 1108. CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs
+
+- **ID:** `2603.27958` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Yongkang Du, Xiaohan Zou, Minhao Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.27958 · **PDF:** https://arxiv.org/pdf/2603.27958v1
+
+## 1109. Symbolic Density Estimation: A Decompositional Approach
+
+- **ID:** `2603.27955` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Angelo Rajendram, Xieting Chu, Vijay Ganesh, et al.
+- **abs:** https://arxiv.org/abs/2603.27955 · **PDF:** https://arxiv.org/pdf/2603.27955v1
+
+## 1110. Scaling Atomistic Protein Binder Design with Generative Pretraining and Test-Time Compute
+
+- **ID:** `2603.27950` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Kieran Didi, Zuobai Zhang, Guoqing Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.27950 · **PDF:** https://arxiv.org/pdf/2603.27950v1
+
+## 1111. EnsemJudge: Enhancing Reliability in Chinese LLM-Generated Text Detection through Diverse Model Ensembles
+
+- **ID:** `2603.27949` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Zhuoshang Wang, Yubing Ren, Guoyu Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.27949 · **PDF:** https://arxiv.org/pdf/2603.27949v1
+
+## 1112. JaWildText: A Benchmark for Vision-Language Models on Japanese Scene Text Understanding
+
+- **ID:** `2603.27942` · **primary:** `cs.CV` · **Submitted:** 2026-03-30
+- **Authors:** Koki Maeda, Naoaki Okazaki
+- **abs:** https://arxiv.org/abs/2603.27942 · **PDF:** https://arxiv.org/pdf/2603.27942v2
+
+## 1113. Top-down string-to-dependency Neural Machine Translation
+
+- **ID:** `2603.27938` · **primary:** `cs.CL` · **Submitted:** 2026-03-30
+- **Authors:** Shuhei Kondo, Katsuhito Sudoh, Yuji Matsumoto
+- **abs:** https://arxiv.org/abs/2603.27938 · **PDF:** https://arxiv.org/pdf/2603.27938v1
+
+## 1114. Deflation-PINNs: Learning Multiple Solutions for PDEs and Landau-de Gennes
+
+- **ID:** `2603.27936` · **primary:** `math.NA` · **Submitted:** 2026-03-30
+- **Authors:** Sean Disarò, Ruma Rani Maity, Aras Bacho
+- **abs:** https://arxiv.org/abs/2603.27936 · **PDF:** https://arxiv.org/pdf/2603.27936v2
+
+## 1115. Physics-Guided Transformer (PGT): Physics-Aware Attention Mechanism for PINNs
+
+- **ID:** `2603.27929` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Ehsan Zeraatkar, Rodion Podorozhny, Jelena Tešić
+- **abs:** https://arxiv.org/abs/2603.27929 · **PDF:** https://arxiv.org/pdf/2603.27929v1
+
+## 1116. GEAKG: Generative Executable Algorithm Knowledge Graphs
+
+- **ID:** `2603.27922` · **primary:** `cs.AI` · **Submitted:** 2026-03-30
+- **Authors:** Camilo Chacón Sartori, José H. García, Andrei Voicu Tomut, et al.
+- **abs:** https://arxiv.org/abs/2603.27922 · **PDF:** https://arxiv.org/pdf/2603.27922v1
+
+## 1117. A Learning-Based Cooperative Coevolution Framework for Heterogeneous Large-Scale Global Optimization
+
+- **ID:** `2604.01241` · **primary:** `cs.NE` · **Submitted:** 2026-03-30
+- **Authors:** Wenjie Qiu, Zixin Wang, Hongyu Fang, et al.
+- **abs:** https://arxiv.org/abs/2604.01241 · **PDF:** https://arxiv.org/pdf/2604.01241v1
+
+## 1118. Adversarial Attacks on Multimodal Large Language Models: A Comprehensive Survey
+
+- **ID:** `2603.27918` · **primary:** `cs.CR` · **Submitted:** 2026-03-30
+- **Authors:** Bhavuk Jain, Sercan Ö. Arık, Hardeo K. Thakur
+- **abs:** https://arxiv.org/abs/2603.27918 · **PDF:** https://arxiv.org/pdf/2603.27918v1
+
+## 1119. ITQ3_S: High-Fidelity 3-bit LLM Inference via Interleaved Ternary Quantization with Rotation-Domain Smoothing
+
+- **ID:** `2603.27914` · **primary:** `cs.LG` · **Submitted:** 2026-03-30
+- **Authors:** Edward J. Yoon
+- **abs:** https://arxiv.org/abs/2603.27914 · **PDF:** https://arxiv.org/pdf/2603.27914v2
+
+## 1120. GAAMA: Graph Augmented Associative Memory for Agents
+
+- **ID:** `2603.27910` · **primary:** `cs.AI` · **Submitted:** 2026-03-29
+- **Authors:** Swarna Kamal Paul, Shubhendu Sharma, Nitin Sareen
+- **abs:** https://arxiv.org/abs/2603.27910 · **PDF:** https://arxiv.org/pdf/2603.27910v1
+
+## 1121. Data is All You Need: Markov Chain Car-Following (MC-CF) Model
+
+- **ID:** `2603.27909` · **primary:** `eess.SY` · **Submitted:** 2026-03-29
+- **Authors:** Sungyong Chung, Yanlin Zhang, Nachuan Li, et al.
+- **abs:** https://arxiv.org/abs/2603.27909 · **PDF:** https://arxiv.org/pdf/2603.27909v1
+
+## 1122. ATLAS-RTC: Closing the Loop on LLM Agent Output with Token-Level Runtime Control
+
+- **ID:** `2603.27905` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Christopher Cruz
+- **abs:** https://arxiv.org/abs/2603.27905 · **PDF:** https://arxiv.org/pdf/2603.27905v1
+
+## 1123. Persistence diagrams of random matrices via Morse theory: universality and a new spectral diagnostic
+
+- **ID:** `2603.27903` · **primary:** `stat.ML` · **Submitted:** 2026-03-29
+- **Authors:** Matthew Loftus
+- **abs:** https://arxiv.org/abs/2603.27903 · **PDF:** https://arxiv.org/pdf/2603.27903v1
+
+## 1124. Article and Comment Frames Shape the Quality of Online Comments
+
+- **ID:** `2603.27889` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Matteo Guida, Yulia Otmakhova, Eduard Hovy, et al.
+- **abs:** https://arxiv.org/abs/2603.27889 · **PDF:** https://arxiv.org/pdf/2603.27889v1
+
+## 1125. AI-ready design of realistic 2D materials and interfaces with Mat3ra-2D
+
+- **ID:** `2603.27886` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-29
+- **Authors:** Vsevolod Biryukov, Kamal Choudhary, Timur Bazhirov
+- **abs:** https://arxiv.org/abs/2603.27886 · **PDF:** https://arxiv.org/pdf/2603.27886v1
+
+## 1126. Spectral Signatures of Data Quality: Eigenvalue Tail Index as a Diagnostic for Label Noise in Neural Networks
+
+- **ID:** `2603.27885` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Matthew Loftus
+- **abs:** https://arxiv.org/abs/2603.27885 · **PDF:** https://arxiv.org/pdf/2603.27885v1
+
+## 1127. Near-Optimal Primal-Dual Algorithm for Learning Linear Mixture CMDPs with Adversarial Rewards
+
+- **ID:** `2603.27884` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Kihyun Yu, Seoungbin Bae, Dabeen Lee
+- **abs:** https://arxiv.org/abs/2603.27884 · **PDF:** https://arxiv.org/pdf/2603.27884v1
+
+## 1128. Large Language Models for Analyzing Enterprise Architecture Debt in Unstructured Documentation
+
+- **ID:** `2604.00046` · **primary:** `cs.SE` · **Submitted:** 2026-03-29
+- **Authors:** Christin Pagels, Simon Hacks, Rob Henk Bemthuis
+- **abs:** https://arxiv.org/abs/2604.00046 · **PDF:** https://arxiv.org/pdf/2604.00046v1
+
+## 1129. Kernel Dynamics under Path Entropy Maximization
+
+- **ID:** `2603.27880` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Jnaneshwar Das
+- **abs:** https://arxiv.org/abs/2603.27880 · **PDF:** https://arxiv.org/pdf/2603.27880v1
+
+## 1130. HumMusQA: A Human-written Music Understanding QA Benchmark Dataset
+
+- **ID:** `2603.27877` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Benno Weck, Pablo Puentes, Andrea Poltronieri, et al.
+- **abs:** https://arxiv.org/abs/2603.27877 · **PDF:** https://arxiv.org/pdf/2603.27877v1
+
+## 1131. Stability and Sensitivity Analysis of Relative Temporal-Difference Learning: Extended Version
+
+- **ID:** `2603.27874` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Masoud S. Sakha, Rushikesh Kamalapurkar, Sean Meyn
+- **abs:** https://arxiv.org/abs/2603.27874 · **PDF:** https://arxiv.org/pdf/2603.27874v1
+
+## 1132. Statistical Guarantees for Distributionally Robust Optimization with Optimal Transport and OT-Regularized Divergences
+
+- **ID:** `2603.27871` · **primary:** `stat.ML` · **Submitted:** 2026-03-29
+- **Authors:** Jeremiah Birrell, Xiaoxi Shen
+- **abs:** https://arxiv.org/abs/2603.27871 · **PDF:** https://arxiv.org/pdf/2603.27871v1
+
+## 1133. A Revealed Preference Framework for AI Alignment
+
+- **ID:** `2603.27868` · **primary:** `econ.TH` · **Submitted:** 2026-03-29
+- **Authors:** Elchin Suleymanov
+- **abs:** https://arxiv.org/abs/2603.27868 · **PDF:** https://arxiv.org/pdf/2603.27868v1
+
+## 1134. Vertical Consensus Inference for High-Dimensional Random Partition
+
+- **ID:** `2603.27864` · **primary:** `stat.ME` · **Submitted:** 2026-03-29
+- **Authors:** Khai Nguyen, Yang Ni, Peter Mueller
+- **abs:** https://arxiv.org/abs/2603.27864 · **PDF:** https://arxiv.org/pdf/2603.27864v1
+
+## 1135. ImagenWorld: Stress-Testing Image Generation Models with Explainable Human Evaluation on Open-ended Real-World Tasks
+
+- **ID:** `2603.27862` · **primary:** `cs.GR` · **Submitted:** 2026-03-29
+- **Authors:** Samin Mahdizadeh Sani, Max Ku, Nima Jamali, et al.
+- **abs:** https://arxiv.org/abs/2603.27862 · **PDF:** https://arxiv.org/pdf/2603.27862v1
+
+## 1136. KazByte: Adapting Qwen models to Kazakh via Byte-level Adapter
+
+- **ID:** `2603.27859` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Rauan Akylzhanov
+- **abs:** https://arxiv.org/abs/2603.27859 · **PDF:** https://arxiv.org/pdf/2603.27859v1
+
+## 1137. CARGO: Carbon-Aware Gossip Orchestration in Smart Shipping
+
+- **ID:** `2603.27857` · **primary:** `cs.AI` · **Submitted:** 2026-03-29
+- **Authors:** Alexandros S. Kalafatelis, Nikolaos Nomikos, Vasileios Nikolakakis, et al.
+- **abs:** https://arxiv.org/abs/2603.27857 · **PDF:** https://arxiv.org/pdf/2603.27857v1
+
+## 1138. What can LLMs tell us about the mechanisms behind polarity illusions in humans? Experiments across model scales and training steps
+
+- **ID:** `2603.27855` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Dario Paape
+- **abs:** https://arxiv.org/abs/2603.27855 · **PDF:** https://arxiv.org/pdf/2603.27855v1
+
+## 1139. EffiSkill: Agent Skill Based Automated Code Efficiency Optimization
+
+- **ID:** `2603.27850` · **primary:** `cs.SE` · **Submitted:** 2026-03-29
+- **Authors:** Zimu Wang, Yuling Shi, Mengfan Li, et al.
+- **abs:** https://arxiv.org/abs/2603.27850 · **PDF:** https://arxiv.org/pdf/2603.27850v1
+
+## 1140. Model Capability Dominates: Inference-Time Optimization Lessons from AIMO 3
+
+- **ID:** `2603.27844` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Natapong Nitarach
+- **abs:** https://arxiv.org/abs/2603.27844 · **PDF:** https://arxiv.org/pdf/2603.27844v1
+
+## 1141. ProText: A benchmark dataset for measuring (mis)gendering in long-form texts
+
+- **ID:** `2603.27838` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Hadas Kotek, Margit Bowler, Patrick Sonnenberg, et al.
+- **abs:** https://arxiv.org/abs/2603.27838 · **PDF:** https://arxiv.org/pdf/2603.27838v1
+
+## 1142. Q-Bridge: Code Translation for Quantum Machine Learning via LLMs
+
+- **ID:** `2603.27836` · **primary:** `quant-ph` · **Submitted:** 2026-03-29
+- **Authors:** Runjia Zeng, Priyabrata Senapati, Ruixiang Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.27836 · **PDF:** https://arxiv.org/pdf/2603.27836v1
+
+## 1143. 3-D Representations for Hyperspectral Flame Tomography
+
+- **ID:** `2603.27832` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Nicolas Tricard, Zituo Chen, Sili Deng
+- **abs:** https://arxiv.org/abs/2603.27832 · **PDF:** https://arxiv.org/pdf/2603.27832v1
+
+## 1144. jaxsgp4: GPU-accelerated mega-constellation propagation with batch parallelism
+
+- **ID:** `2603.27830` · **primary:** `cs.DC` · **Submitted:** 2026-03-29
+- **Authors:** Charlotte Priestley, Will Handley
+- **abs:** https://arxiv.org/abs/2603.27830 · **PDF:** https://arxiv.org/pdf/2603.27830v1
+
+## 1145. Computational Foundations for Strategic Coopetition: Formalizing Sequential Interaction and Reciprocity
+
+- **ID:** `2604.01240` · **primary:** `cs.MA` · **Submitted:** 2026-03-29
+- **Authors:** Vik Pant, Eric Yu
+- **abs:** https://arxiv.org/abs/2604.01240 · **PDF:** https://arxiv.org/pdf/2604.01240v1
+
+## 1146. Improving Clinical Diagnosis with Counterfactual Multi-Agent Reasoning
+
+- **ID:** `2603.27820` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Zhiwen You, Xi Chen, Aniket Vashishtha, et al.
+- **abs:** https://arxiv.org/abs/2603.27820 · **PDF:** https://arxiv.org/pdf/2603.27820v1
+
+## 1147. KVSculpt: KV Cache Compression as Distillation
+
+- **ID:** `2603.27819` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Bo Jiang, Sian Jin
+- **abs:** https://arxiv.org/abs/2603.27819 · **PDF:** https://arxiv.org/pdf/2603.27819v1
+
+## 1148. Towards Context-Aware Image Anonymization with Multi-Agent Reasoning
+
+- **ID:** `2603.27817` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Robert Aufschläger, Jakob Folz, Gautam Savaliya, et al.
+- **abs:** https://arxiv.org/abs/2603.27817 · **PDF:** https://arxiv.org/pdf/2603.27817v1
+
+## 1149. RG-TTA: Regime-Guided Meta-Control for Test-Time Adaptation in Streaming Time Series
+
+- **ID:** `2603.27814` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Indar Kumar, Akanksha Tiwari, Sai Krishna Jasti, et al.
+- **abs:** https://arxiv.org/abs/2603.27814 · **PDF:** https://arxiv.org/pdf/2603.27814v1
+
+## 1150. Tracking without Seeing: Geospatial Inference using Encrypted Traffic from Distributed Nodes
+
+- **ID:** `2603.27811` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Sadik Yagiz Yetim, Gaofeng Dong, Isaac-Neil Zanoria, et al.
+- **abs:** https://arxiv.org/abs/2603.27811 · **PDF:** https://arxiv.org/pdf/2603.27811v1
+
+## 1151. Conversational Agents and the Understanding of Human Language: Reflections on AI, LLMs, and Cognitive Science
+
+- **ID:** `2603.27809` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Andrei Popescu-Belis
+- **abs:** https://arxiv.org/abs/2603.27809 · **PDF:** https://arxiv.org/pdf/2603.27809v1
+
+## 1152. Understanding Teacher Revisions of Large Language Model-Generated Feedback
+
+- **ID:** `2603.27806` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Conrad Borchers, Luiz Rodrigues, Newarney Torrezão da Costa, et al.
+- **abs:** https://arxiv.org/abs/2603.27806 · **PDF:** https://arxiv.org/pdf/2603.27806v1
+
+## 1153. Distributed Online Submodular Maximization under Communication Delays: A Simultaneous Decision-Making Approach
+
+- **ID:** `2603.27803` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Zirui Xu, Vasileios Tzoumas
+- **abs:** https://arxiv.org/abs/2603.27803 · **PDF:** https://arxiv.org/pdf/2603.27803v1
+
+## 1154. Towards Emotion Recognition with 3D Pointclouds Obtained from Facial Expression Images
+
+- **ID:** `2603.27798` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Laura Rayón Ropero, Jasper De Laet, Filip Lemic, et al.
+- **abs:** https://arxiv.org/abs/2603.27798 · **PDF:** https://arxiv.org/pdf/2603.27798v1
+
+## 1155. What-If Explanations Over Time: Counterfactuals for Time Series Classification
+
+- **ID:** `2603.27792` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Udo Schlegel, Thomas Seidl
+- **abs:** https://arxiv.org/abs/2603.27792 · **PDF:** https://arxiv.org/pdf/2603.27792v1
+
+## 1156. Emergent Social Intelligence Risks in Generative Multi-Agent Systems
+
+- **ID:** `2603.27771` · **primary:** `cs.MA` · **Submitted:** 2026-03-29
+- **Authors:** Yue Huang, Yu Jiang, Wenjie Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.27771 · **PDF:** https://arxiv.org/pdf/2603.27771v1
+
+## 1157. TailNLG: A Multilingual Benchmark Addressing Verbalization of Long-Tail Entities
+
+- **ID:** `2603.27768` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Lia Draetta, Michael Oliverio, Virginia Ramón-Ferrer, et al.
+- **abs:** https://arxiv.org/abs/2603.27768 · **PDF:** https://arxiv.org/pdf/2603.27768v1
+
+## 1158. AutoStan: Autonomous Bayesian Model Improvement via Predictive Feedback
+
+- **ID:** `2603.27766` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Oliver Dürr
+- **abs:** https://arxiv.org/abs/2603.27766 · **PDF:** https://arxiv.org/pdf/2603.27766v1
+
+## 1159. Let the Agent Steer: Closed-Loop Ranking Optimization via Influence Exchange
+
+- **ID:** `2603.27765` · **primary:** `cs.AI` · **Submitted:** 2026-03-29
+- **Authors:** Yin Cheng, Liao Zhou, Xiyu Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.27765 · **PDF:** https://arxiv.org/pdf/2603.27765v2
+
+## 1160. Heracles: Bridging Precise Tracking and Generative Synthesis for General Humanoid Control
+
+- **ID:** `2603.27756` · **primary:** `cs.RO` · **Submitted:** 2026-03-29
+- **Authors:** Zelin Tao, Zeran Su, Peiran Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.27756 · **PDF:** https://arxiv.org/pdf/2603.27756v2
+
+## 1161. Retromorphic Testing with Hierarchical Verification for Hallucination Detection in RAG
+
+- **ID:** `2603.27752` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Boxi Yu, Yuzhong Zhang, Liting Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.27752 · **PDF:** https://arxiv.org/pdf/2603.27752v1
+
+## 1162. SkyNet: Belief-Aware Planning for Partially-Observable Stochastic Games
+
+- **ID:** `2603.27751` · **primary:** `cs.AI` · **Submitted:** 2026-03-29
+- **Authors:** Adam Haile
+- **abs:** https://arxiv.org/abs/2603.27751 · **PDF:** https://arxiv.org/pdf/2603.27751v1
+
+## 1163. AI-Powered Facial Mask Removal Is Not Suitable For Biometric Identification
+
+- **ID:** `2603.27747` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Emily A Cooper, Hany Farid
+- **abs:** https://arxiv.org/abs/2603.27747 · **PDF:** https://arxiv.org/pdf/2603.27747v1
+
+## 1164. Needle in the Repo: A Benchmark for Maintainability in AI-Generated Repository Edits
+
+- **ID:** `2603.27745` · **primary:** `cs.SE` · **Submitted:** 2026-03-29
+- **Authors:** Haichao Zhu, Qian Zhang, Jiyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.27745 · **PDF:** https://arxiv.org/pdf/2603.27745v1
+
+## 1165. Empirical Likelihood for Nonsmooth Functionals
+
+- **ID:** `2603.27743` · **primary:** `stat.ME` · **Submitted:** 2026-03-29
+- **Authors:** Hongseok Namkoong
+- **abs:** https://arxiv.org/abs/2603.27743 · **PDF:** https://arxiv.org/pdf/2603.27743v1
+
+## 1166. TianJi:An autonomous AI meteorologist for discovering physical mechanisms in atmospheric science
+
+- **ID:** `2603.27738` · **primary:** `cs.AI` · **Submitted:** 2026-03-29
+- **Authors:** Kaikai Zhang, Xiang Wang, Haoluo Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.27738 · **PDF:** https://arxiv.org/pdf/2603.27738v1
+
+## 1167. Robust Smart Contract Vulnerability Detection via Contrastive Learning-Enhanced Granular-ball Training
+
+- **ID:** `2603.27734` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Zeli Wang, Qingxuan Yang, Shuyin Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.27734 · **PDF:** https://arxiv.org/pdf/2603.27734v1
+
+## 1168. Suppression of $^{14}\mathrm{C}$ photon hits in large liquid scintillator detectors via spatiotemporal deep learning
+
+- **ID:** `2603.27727` · **primary:** `physics.ins-det` · **Submitted:** 2026-03-29
+- **Authors:** Junle Li, Zhaoxiang Wu, Guanda Gong, et al.
+- **abs:** https://arxiv.org/abs/2603.27727 · **PDF:** https://arxiv.org/pdf/2603.27727v1
+
+## 1169. TMTE: Effective Multimodal Graph Learning with Task-aware Modality and Topology Co-evolution
+
+- **ID:** `2603.27723` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Yinlin Zhu, Xunkai Li, Di Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.27723 · **PDF:** https://arxiv.org/pdf/2603.27723v1
+
+## 1170. The role of neuromorphic principles in the future of biomedicine and healthcare
+
+- **ID:** `2603.27716` · **primary:** `cs.NE` · **Submitted:** 2026-03-29
+- **Authors:** Grace M. Hwang, Jessica D. Falcone, Joseph D. Monaco, et al.
+- **abs:** https://arxiv.org/abs/2603.27716 · **PDF:** https://arxiv.org/pdf/2603.27716v1
+
+## 1171. Low-Rank Adaptation Reduces Catastrophic Forgetting in Sequential Transformer Encoder Fine-Tuning: Controlled Empirical Evidence and Frozen-Backbone Representation Probes
+
+- **ID:** `2603.27707` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Ashish Pandey
+- **abs:** https://arxiv.org/abs/2603.27707 · **PDF:** https://arxiv.org/pdf/2603.27707v1
+
+## 1172. RAP: Retrieve, Adapt, and Prompt-Fit for Training-Free Few-Shot Medical Image Segmentation
+
+- **ID:** `2603.27705` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Zhihao Mao, Bangpu Chen
+- **abs:** https://arxiv.org/abs/2603.27705 · **PDF:** https://arxiv.org/pdf/2603.27705v1
+
+## 1173. KAT-Coder-V2 Technical Report
+
+- **ID:** `2603.27703` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Fengxiang Li, Han Zhang, Haoyang Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.27703 · **PDF:** https://arxiv.org/pdf/2603.27703v1
+
+## 1174. Optimizing Coverage and Difficulty in Reinforcement Learning for Quiz Composition
+
+- **ID:** `2603.27695` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Ricardo Pedro Querido Andrade Silva, Nassim Bouarour, Dina Fettache, et al.
+- **abs:** https://arxiv.org/abs/2603.27695 · **PDF:** https://arxiv.org/pdf/2603.27695v1
+
+## 1175. Can Large Language Models Simulate Human Cognition Beyond Behavioral Imitation?
+
+- **ID:** `2603.27694` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Yuxuan Gu, Lunjun Liu, Xiaocheng Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.27694 · **PDF:** https://arxiv.org/pdf/2603.27694v1
+
+## 1176. LVRPO: Language-Visual Alignment with GRPO for Multimodal Understanding and Generation
+
+- **ID:** `2603.27693` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Shentong Mo, Sukmin Yun
+- **abs:** https://arxiv.org/abs/2603.27693 · **PDF:** https://arxiv.org/pdf/2603.27693v1
+
+## 1177. CrossHGL: A Text-Free Foundation Model for Cross-Domain Heterogeneous Graph Learning
+
+- **ID:** `2603.27685` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Xuanze Chen, Jiajun Zhou, Yadong Li, et al.
+- **abs:** https://arxiv.org/abs/2603.27685 · **PDF:** https://arxiv.org/pdf/2603.27685v1
+
+## 1178. Prototype-Aligned Federated Soft-Prompts for Continual Web Personalization
+
+- **ID:** `2603.27678` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Canran Xiao, Liwei Hou
+- **abs:** https://arxiv.org/abs/2603.27678 · **PDF:** https://arxiv.org/pdf/2603.27678v1
+
+## 1179. Energy Score-Guided Neural Gaussian Mixture Model for Predictive Uncertainty Quantification
+
+- **ID:** `2603.27672` · **primary:** `stat.ML` · **Submitted:** 2026-03-29
+- **Authors:** Yang Yang, Chunlin Ji, Haoyang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.27672 · **PDF:** https://arxiv.org/pdf/2603.27672v1
+
+## 1180. ProgressVLA: Progress-Guided Diffusion Policy for Vision-Language Robotic Manipulation
+
+- **ID:** `2603.27670` · **primary:** `cs.RO` · **Submitted:** 2026-03-29
+- **Authors:** Hongyu Yan, Qiwei Li, Jiaolong Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.27670 · **PDF:** https://arxiv.org/pdf/2603.27670v1
+
+## 1181. DriftScript: A Domain-Specific Language for Programming Non-Axiomatic Reasoning Agents
+
+- **ID:** `2604.00043` · **primary:** `cs.PL` · **Submitted:** 2026-03-29
+- **Authors:** Seamus Brady
+- **abs:** https://arxiv.org/abs/2604.00043 · **PDF:** https://arxiv.org/pdf/2604.00043v1
+
+## 1182. EvA: An Evidence-First Audio Understanding Paradigm for LALMs
+
+- **ID:** `2603.27667` · **primary:** `cs.SD` · **Submitted:** 2026-03-29
+- **Authors:** Xinyuan Xie, Shunian Chen, Zhiheng Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.27667 · **PDF:** https://arxiv.org/pdf/2603.27667v1
+
+## 1183. Test-Time Instance-Specific Parameter Composition: A New Paradigm for Adaptive Generative Modeling
+
+- **ID:** `2603.27665` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Minh-Tuan Tran, Xuan-May Le, Quan Hung Tran, et al.
+- **abs:** https://arxiv.org/abs/2603.27665 · **PDF:** https://arxiv.org/pdf/2603.27665v1
+
+## 1184. Investigating the Influence of Language on Sycophantic Behavior of Multilingual LLMs
+
+- **ID:** `2603.27664` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Bayan Abdullah Aldahlawi, A. B. M. Ashikur Rahman, Irfan Ahmad
+- **abs:** https://arxiv.org/abs/2603.27664 · **PDF:** https://arxiv.org/pdf/2603.27664v1
+
+## 1185. The Degree of Language Diacriticity and Its Effect on Tasks
+
+- **ID:** `2603.27653` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Adi Cohen, Yuval Pinter
+- **abs:** https://arxiv.org/abs/2603.27653 · **PDF:** https://arxiv.org/pdf/2603.27653v1
+
+## 1186. Incentives, Equilibria, and the Limits of Healthcare AI: A Game-Theoretic Perspective
+
+- **ID:** `2603.28825` · **primary:** `cs.GT` · **Submitted:** 2026-03-29
+- **Authors:** Ari Ercole
+- **abs:** https://arxiv.org/abs/2603.28825 · **PDF:** https://arxiv.org/pdf/2603.28825v1
+
+## 1187. Budget-Xfer: Budget-Constrained Source Language Selection for Cross-Lingual Transfer to African Languages
+
+- **ID:** `2603.27651` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Tewodros Kederalah Idris, Roald Eiselen, Prasenjit Mitra
+- **abs:** https://arxiv.org/abs/2603.27651 · **PDF:** https://arxiv.org/pdf/2603.27651v1
+
+## 1188. PRBench: End-to-end Paper Reproduction in Physics Research
+
+- **ID:** `2603.27646` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Shi Qiu, Junyi Deng, Yiwei Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.27646 · **PDF:** https://arxiv.org/pdf/2603.27646v1
+
+## 1189. ContraMap: Contrastive Uncertainty Mapping for Robot Environment Representation
+
+- **ID:** `2603.27632` · **primary:** `cs.RO` · **Submitted:** 2026-03-29
+- **Authors:** Chi Cuong Le, Weiming Zhi
+- **abs:** https://arxiv.org/abs/2603.27632 · **PDF:** https://arxiv.org/pdf/2603.27632v1
+
+## 1190. On the Asymptotics of Self-Supervised Pre-training: Two-Stage M-Estimation and Representation Symmetry
+
+- **ID:** `2603.27631` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Mohammad Tinati, Stephen Tu
+- **abs:** https://arxiv.org/abs/2603.27631 · **PDF:** https://arxiv.org/pdf/2603.27631v1
+
+## 1191. RTLSeek: Boosting the LLM-Based RTL Generation with Multi-Stage Diversity-Oriented Reinforcement Learning
+
+- **ID:** `2603.27630` · **primary:** `cs.AR` · **Submitted:** 2026-03-29
+- **Authors:** Xinyu Zhang, Zhiteng Chao, Yonghao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.27630 · **PDF:** https://arxiv.org/pdf/2603.27630v1
+
+## 1192. DSevolve: Enabling Real-Time Adaptive Scheduling on Dynamic Shop Floor with LLM-Evolved Heuristic Portfolios
+
+- **ID:** `2603.27628` · **primary:** `cs.AI` · **Submitted:** 2026-03-29
+- **Authors:** Jin Huang, Jie Yang, XinLei Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.27628 · **PDF:** https://arxiv.org/pdf/2603.27628v1
+
+## 1193. Umwelt Engineering: Designing the Cognitive Worlds of Linguistic Agents
+
+- **ID:** `2603.27626` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Rodney Jehu-Appiah
+- **abs:** https://arxiv.org/abs/2603.27626 · **PDF:** https://arxiv.org/pdf/2603.27626v1
+
+## 1194. Expert Streaming: Accelerating Low-Batch MoE Inference via Multi-chiplet Architecture and Dynamic Expert Trajectory Scheduling
+
+- **ID:** `2603.27624` · **primary:** `cs.AR` · **Submitted:** 2026-03-29
+- **Authors:** Songchen Ma, Hongyi Li, Weihao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.27624 · **PDF:** https://arxiv.org/pdf/2603.27624v1
+
+## 1195. What does a system modify when it modifies itself?
+
+- **ID:** `2603.27611` · **primary:** `cs.AI` · **Submitted:** 2026-03-29
+- **Authors:** Florentin Koch
+- **abs:** https://arxiv.org/abs/2603.27611 · **PDF:** https://arxiv.org/pdf/2603.27611v1
+
+## 1196. From indicators to biology: the calibration problem in artificial consciousness
+
+- **ID:** `2603.27597` · **primary:** `cs.AI` · **Submitted:** 2026-03-29
+- **Authors:** Florentin Koch
+- **abs:** https://arxiv.org/abs/2603.27597 · **PDF:** https://arxiv.org/pdf/2603.27597v1
+
+## 1197. STRIDE: When to Speak Meets Sequence Denoising for Streaming Video Understanding
+
+- **ID:** `2603.27593` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Junho Kim, Hosu Lee, James M. Rehg, et al.
+- **abs:** https://arxiv.org/abs/2603.27593 · **PDF:** https://arxiv.org/pdf/2603.27593v1
+
+## 1198. Fundamental Limits of Man-in-the-Middle Attack Detection in Model-Free Reinforcement Learning
+
+- **ID:** `2603.27592` · **primary:** `eess.SY` · **Submitted:** 2026-03-29
+- **Authors:** Rishi Rani, Massimo Franceschetti
+- **abs:** https://arxiv.org/abs/2603.27592 · **PDF:** https://arxiv.org/pdf/2603.27592v2
+
+## 1199. An Energy-Efficient Spiking Neural Network Architecture for Predictive Insulin Delivery
+
+- **ID:** `2603.27589` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Sahil Shrivastava
+- **abs:** https://arxiv.org/abs/2603.27589 · **PDF:** https://arxiv.org/pdf/2603.27589v1
+
+## 1200. SNEAKDOOR: Stealthy Backdoor Attacks against Distribution Matching-based Dataset Condensation
+
+- **ID:** `2603.28824` · **primary:** `cs.CR` · **Submitted:** 2026-03-29
+- **Authors:** He Yang, Dongyi Lv, Song Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.28824 · **PDF:** https://arxiv.org/pdf/2603.28824v1
+
+## 1201. InnerPond: Fostering Inter-Self Dialogue with a Multi-Agent Approach for Introspection
+
+- **ID:** `2603.27563` · **primary:** `cs.HC` · **Submitted:** 2026-03-29
+- **Authors:** Hayeon Jeon, Dakyeom Ahn, Sunyu Pang, et al.
+- **abs:** https://arxiv.org/abs/2603.27563 · **PDF:** https://arxiv.org/pdf/2603.27563v1
+
+## 1202. A General Model for Deepfake Speech Detection: Diverse Bonafide Resources or Diverse AI-Based Generators
+
+- **ID:** `2603.27557` · **primary:** `cs.SD` · **Submitted:** 2026-03-29
+- **Authors:** Lam Pham, Khoi Vu, Dat Tran, et al.
+- **abs:** https://arxiv.org/abs/2603.27557 · **PDF:** https://arxiv.org/pdf/2603.27557v1
+
+## 1203. Time is Not Compute: Scaling Laws for Wall-Clock Constrained Training on Consumer GPUs
+
+- **ID:** `2603.28823` · **primary:** `cs.PF` · **Submitted:** 2026-03-29
+- **Authors:** Yi Liu
+- **abs:** https://arxiv.org/abs/2603.28823 · **PDF:** https://arxiv.org/pdf/2603.28823v1
+
+## 1204. BLOSSOM: Block-wise Federated Learning Over Shared and Sparse Observed Modalities
+
+- **ID:** `2603.27552` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Pranav M R, Jayant Chandwani, Ahmed M. Abdelmoniem, et al.
+- **abs:** https://arxiv.org/abs/2603.27552 · **PDF:** https://arxiv.org/pdf/2603.27552v1
+
+## 1205. Drag or Traction: Understanding How Designers Appropriate Friction in AI Ideation Outputs
+
+- **ID:** `2603.27550` · **primary:** `cs.HC` · **Submitted:** 2026-03-29
+- **Authors:** A. Baki Kocaballi, Joseph Kizana, Sharon Stein, et al.
+- **abs:** https://arxiv.org/abs/2603.27550 · **PDF:** https://arxiv.org/pdf/2603.27550v1
+
+## 1206. A Novel Immune Algorithm for Multiparty Multiobjective Optimization
+
+- **ID:** `2603.27541` · **primary:** `cs.NE` · **Submitted:** 2026-03-29
+- **Authors:** Kesheng Chen, Wenjian Luo, Qi Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.27541 · **PDF:** https://arxiv.org/pdf/2603.27541v1
+
+## 1207. Toward Reliable Evaluation of LLM-Based Financial Multi-Agent Systems: Taxonomy, Coordination Primacy, and Cost Awareness
+
+- **ID:** `2603.27539` · **primary:** `cs.MA` · **Submitted:** 2026-03-29
+- **Authors:** Phat Nguyen, Thang Pham
+- **abs:** https://arxiv.org/abs/2603.27539 · **PDF:** https://arxiv.org/pdf/2603.27539v1
+
+## 1208. LongCat-Next: Lexicalizing Modalities as Discrete Tokens
+
+- **ID:** `2603.27538` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Meituan LongCat Team, Bin Xiao, Chao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.27538 · **PDF:** https://arxiv.org/pdf/2603.27538v1
+
+## 1209. Dual-Stage LLM Framework for Scenario-Centric Semantic Interpretation in Driving Assistance
+
+- **ID:** `2603.27536` · **primary:** `cs.AI` · **Submitted:** 2026-03-29
+- **Authors:** Jean Douglas Carvalho, Hugo Taciro Kenji, Ahmad Mohammad Saber, et al.
+- **abs:** https://arxiv.org/abs/2603.27536 · **PDF:** https://arxiv.org/pdf/2603.27536v1
+
+## 1210. Demo-Pose: Depth-Monocular Modality Fusion For Object Pose Estimation
+
+- **ID:** `2603.27533` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Rachit Agarwal, Abhishek Joshi, Sathish Chalasani, et al.
+- **abs:** https://arxiv.org/abs/2603.27533 · **PDF:** https://arxiv.org/pdf/2603.27533v1
+
+## 1211. A gentle tutorial and a structured reformulation of Bock's algorithm for minimum directed spanning trees
+
+- **ID:** `2603.27530` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Yuxi Wang, Jungyeul Park
+- **abs:** https://arxiv.org/abs/2603.27530 · **PDF:** https://arxiv.org/pdf/2603.27530v1
+
+## 1212. Cross-attentive Cohesive Subgraph Embedding to Mitigate Oversquashing in GNNs
+
+- **ID:** `2603.27529` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Tanvir Hossain, Muhammad Ifte Khairul Islam, Lilia Chebbah, et al.
+- **abs:** https://arxiv.org/abs/2603.27529 · **PDF:** https://arxiv.org/pdf/2603.27529v3
+
+## 1213. Visualization of Machine Learning Models through Their Spatial and Temporal Listeners
+
+- **ID:** `2603.27527` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Siyu Wu, Lei Shi, Lei Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.27527 · **PDF:** https://arxiv.org/pdf/2603.27527v1
+
+## 1214. Q-BIOLAT: Binary Latent Protein Fitness Landscapes for QUBO-Based Optimization
+
+- **ID:** `2603.27526` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Truong-Son Hy
+- **abs:** https://arxiv.org/abs/2603.27526 · **PDF:** https://arxiv.org/pdf/2603.27526v1
+
+## 1215. Safer Builders, Risky Maintainers: A Comparative Study of Breaking Changes in Human vs Agentic PRs
+
+- **ID:** `2603.27524` · **primary:** `cs.SE` · **Submitted:** 2026-03-29
+- **Authors:** K M Ferdous, Dipayan Banik, Kowshik Chowdhury, et al.
+- **abs:** https://arxiv.org/abs/2603.27524 · **PDF:** https://arxiv.org/pdf/2603.27524v1
+
+## 1216. Hidden Ads: Behavior Triggered Semantic Backdoors for Advertisement Injection in Vision Language Models
+
+- **ID:** `2603.27522` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Duanyi Yao, Changyue Li, Zhicong Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.27522 · **PDF:** https://arxiv.org/pdf/2603.27522v1
+
+## 1217. Over-Refusal and Representation Subspaces: A Mechanistic Analysis of Task-Conditioned Refusal in Aligned LLMs
+
+- **ID:** `2603.27518` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Utsav Maskey, Mark Dras, Usman Naseem
+- **abs:** https://arxiv.org/abs/2603.27518 · **PDF:** https://arxiv.org/pdf/2603.27518v1
+
+## 1218. A Systematic Taxonomy of Security Vulnerabilities in the OpenClaw AI Agent Framework
+
+- **ID:** `2603.27517` · **primary:** `cs.CR` · **Submitted:** 2026-03-29
+- **Authors:** Surada Suwansathit, Yuxuan Zhang, Guofei Gu
+- **abs:** https://arxiv.org/abs/2603.27517 · **PDF:** https://arxiv.org/pdf/2603.27517v1
+
+## 1219. Match or Replay: Self Imitating Proximal Policy Optimization
+
+- **ID:** `2603.27515` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Gaurav Chaudhary, Laxmidhar Behera, Washim Uddin Mondal
+- **abs:** https://arxiv.org/abs/2603.27515 · **PDF:** https://arxiv.org/pdf/2603.27515v1
+
+## 1220. Understanding Semantic Perturbations on In-Processing Generative Image Watermarks
+
+- **ID:** `2603.27513` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Anirudh Nakra, Min Wu
+- **abs:** https://arxiv.org/abs/2603.27513 · **PDF:** https://arxiv.org/pdf/2603.27513v1
+
+## 1221. Decomposing Discrimination: Causal Mediation Analysis for AI-Driven Credit Decisions
+
+- **ID:** `2603.27510` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Duraimurugan Rajamanickam
+- **abs:** https://arxiv.org/abs/2603.27510 · **PDF:** https://arxiv.org/pdf/2603.27510v1
+
+## 1222. Learning to Focus and Precise Cropping: A Reinforcement Learning Framework with Information Gaps and Grounding Loss for MLLMs
+
+- **ID:** `2603.27494` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Xuanpu Zhao, Zhentao Tan, Dianmo Sheng, et al.
+- **abs:** https://arxiv.org/abs/2603.27494 · **PDF:** https://arxiv.org/pdf/2603.27494v1
+
+## 1223. Copilot-Assisted Second-Thought Framework for Brain-to-Robot Hand Motion Decoding
+
+- **ID:** `2603.27492` · **primary:** `cs.RO` · **Submitted:** 2026-03-29
+- **Authors:** Yizhe Li, Shixiao Wang, Jian K. Liu
+- **abs:** https://arxiv.org/abs/2603.27492 · **PDF:** https://arxiv.org/pdf/2603.27492v1
+
+## 1224. AgentSwing: Adaptive Parallel Context Management Routing for Long-Horizon Web Agents
+
+- **ID:** `2603.27490` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Zhaopeng Feng, Liangcai Su, Zhen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.27490 · **PDF:** https://arxiv.org/pdf/2603.27490v1
+
+## 1225. Variational Learning of Fractional Posteriors
+
+- **ID:** `2603.27488` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Kian Ming A. Chai, Edwin V. Bonilla
+- **abs:** https://arxiv.org/abs/2603.27488 · **PDF:** https://arxiv.org/pdf/2603.27488v1
+
+## 1226. Difference Feedback: Generating Multimodal Process-Level Supervision for VLM Reinforcement Learning
+
+- **ID:** `2603.27482` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Feiding, Yongkang Zhang, Yuhao Liao, et al.
+- **abs:** https://arxiv.org/abs/2603.27482 · **PDF:** https://arxiv.org/pdf/2603.27482v1
+
+## 1227. On Token's Dilemma: Dynamic MoE with Drift-Aware Token Assignment for Continual Learning of Large Vision Language Models
+
+- **ID:** `2603.27481` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Chongyang Zhao, Mingsong Li, Haodong Lu, et al.
+- **abs:** https://arxiv.org/abs/2603.27481 · **PDF:** https://arxiv.org/pdf/2603.27481v1
+
+## 1228. PeopleSearchBench: A Multi-Dimensional Benchmark for Evaluating AI-Powered People Search Platforms
+
+- **ID:** `2603.27476` · **primary:** `cs.AI` · **Submitted:** 2026-03-29
+- **Authors:** Wei Wang, Tianyu Shi, Shuai Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.27476 · **PDF:** https://arxiv.org/pdf/2603.27476v1
+
+## 1229. KV Cache Quantization for Self-Forcing Video Generation: A 33-Method Empirical Study
+
+- **ID:** `2603.27469` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Suraj Ranganath, Vaishak Menon, Anish Patnaik
+- **abs:** https://arxiv.org/abs/2603.27469 · **PDF:** https://arxiv.org/pdf/2603.27469v1
+
+## 1230. TurboAngle: Near-Lossless KV Cache Compression via Uniform Angle Quantization
+
+- **ID:** `2603.27467` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Dipkumar Patel
+- **abs:** https://arxiv.org/abs/2603.27467 · **PDF:** https://arxiv.org/pdf/2603.27467v1
+
+## 1231. RSR-core: A High-Performance Engine for Low-Bit Matrix-Vector Multiplication
+
+- **ID:** `2603.27462` · **primary:** `cs.DS` · **Submitted:** 2026-03-29
+- **Authors:** Mohsen Dehghankar, Abolfazl Asudeh
+- **abs:** https://arxiv.org/abs/2603.27462 · **PDF:** https://arxiv.org/pdf/2603.27462v1
+
+## 1232. Project Imaging-X: A Survey of 1000+ Open-Access Medical Imaging Datasets for Foundation Model Development
+
+- **ID:** `2603.27460` · **primary:** `cs.CV` · **Submitted:** 2026-03-29
+- **Authors:** Zhongying Deng, Cheng Tang, Ziyan Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.27460 · **PDF:** https://arxiv.org/pdf/2603.27460v1
+
+## 1233. A tree interpretation of arc standard dependency derivation
+
+- **ID:** `2603.27459` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Zihao Huang, Ai Ka Lee, Jungyeul Park
+- **abs:** https://arxiv.org/abs/2603.27459 · **PDF:** https://arxiv.org/pdf/2603.27459v1
+
+## 1234. Optimal Demixing of Nonparametric Densities
+
+- **ID:** `2603.27457` · **primary:** `math.ST` · **Submitted:** 2026-03-29
+- **Authors:** Jianqing Fan, Zheng Tracy Ke, Zhaoyang Shi
+- **abs:** https://arxiv.org/abs/2603.27457 · **PDF:** https://arxiv.org/pdf/2603.27457v1
+
+## 1235. Multi-Agent Dialectical Refinement for Enhanced Argument Classification
+
+- **ID:** `2603.27451` · **primary:** `cs.CL` · **Submitted:** 2026-03-29
+- **Authors:** Jakub Bąba, Jarosław A. Chudziak
+- **abs:** https://arxiv.org/abs/2603.27451 · **PDF:** https://arxiv.org/pdf/2603.27451v1
+
+## 1236. FlowRL: A Taxonomy and Modular Framework for Reinforcement Learning with Diffusion Policies
+
+- **ID:** `2603.27450` · **primary:** `cs.LG` · **Submitted:** 2026-03-29
+- **Authors:** Chenxiao Gao, Edward Chen, Tianyi Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.27450 · **PDF:** https://arxiv.org/pdf/2603.27450v1
+
+## 1237. GIFT: Bootstrapping Image-to-CAD Program Synthesis via Geometric Feedback
+
+- **ID:** `2603.27448` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Giorgio Giannone, Anna Clare Doris, Amin Heyrani Nobari, et al.
+- **abs:** https://arxiv.org/abs/2603.27448 · **PDF:** https://arxiv.org/pdf/2603.27448v1
+
+## 1238. Interpretable Physics Extraction from Data for Linear Dynamical Systems using Lie Generator Networks
+
+- **ID:** `2603.27442` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Shafayeth Jamil, Rehan Kapadia
+- **abs:** https://arxiv.org/abs/2603.27442 · **PDF:** https://arxiv.org/pdf/2603.27442v1
+
+## 1239. Evaluating Large and Lightweight Vision Models for Irregular Component Segmentation in E-Waste Disassembly
+
+- **ID:** `2603.27441` · **primary:** `cs.CV` · **Submitted:** 2026-03-28
+- **Authors:** Xinyao Zhang, Chang Liu, Xiao Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.27441 · **PDF:** https://arxiv.org/pdf/2603.27441v1
+
+## 1240. The Novelty Bottleneck: A Framework for Understanding Human Effort Scaling in AI-Assisted Work
+
+- **ID:** `2603.27438` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Jacky Liang
+- **abs:** https://arxiv.org/abs/2603.27438 · **PDF:** https://arxiv.org/pdf/2603.27438v1
+
+## 1241. Improving Attributed Long-form Question Answering with Intent Awareness
+
+- **ID:** `2603.27435` · **primary:** `cs.CL` · **Submitted:** 2026-03-28
+- **Authors:** Xinran Zhao, Aakanksha Naik, Jay DeYoung, et al.
+- **abs:** https://arxiv.org/abs/2603.27435 · **PDF:** https://arxiv.org/pdf/2603.27435v1
+
+## 1242. The Geometric Cost of Normalization: Affine Bounds on the Bayesian Complexity of Neural Networks
+
+- **ID:** `2603.27432` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Sungbae Chun
+- **abs:** https://arxiv.org/abs/2603.27432 · **PDF:** https://arxiv.org/pdf/2603.27432v1
+
+## 1243. AstraAI: LLMs, Retrieval, and AST-Guided Assistance for HPC Codebases
+
+- **ID:** `2603.27423` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Mahesh Natarajan, Xiaoye Li, Weiqun Zhang
+- **abs:** https://arxiv.org/abs/2603.27423 · **PDF:** https://arxiv.org/pdf/2603.27423v1
+
+## 1244. CarbonEdge: Carbon-Aware Deep Learning Inference Framework for Sustainable Edge Computing
+
+- **ID:** `2603.27420` · **primary:** `cs.DC` · **Submitted:** 2026-03-28
+- **Authors:** Guilin Zhang, Wulan Guo, Ziqi Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.27420 · **PDF:** https://arxiv.org/pdf/2603.27420v2
+
+## 1245. Kempe Swap K-Means: A Scalable Near-Optimal Solution for Semi-Supervised Clustering
+
+- **ID:** `2603.27417` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Yuxuan Ren, Shijie Deng
+- **abs:** https://arxiv.org/abs/2603.27417 · **PDF:** https://arxiv.org/pdf/2603.27417v1
+
+## 1246. Agent-Driven Autonomous Reinforcement Learning Research: Iterative Policy Improvement for Quadruped Locomotion
+
+- **ID:** `2603.27416` · **primary:** `cs.RO` · **Submitted:** 2026-03-28
+- **Authors:** Nimesh Khandelwal, Shakti S. Gupta
+- **abs:** https://arxiv.org/abs/2603.27416 · **PDF:** https://arxiv.org/pdf/2603.27416v1
+
+## 1247. Greedy Is a Strong Default: Agents as Iterative Optimizers
+
+- **ID:** `2603.27415` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Yitao Li
+- **abs:** https://arxiv.org/abs/2603.27415 · **PDF:** https://arxiv.org/pdf/2603.27415v1
+
+## 1248. Multiple-Prediction-Powered Inference
+
+- **ID:** `2603.27414` · **primary:** `math.ST` · **Submitted:** 2026-03-28
+- **Authors:** Charlie Cowen-Breen, Alekh Agarwal, Stephen Bates, et al.
+- **abs:** https://arxiv.org/abs/2603.27414 · **PDF:** https://arxiv.org/pdf/2603.27414v1
+
+## 1249. The Hidden Costs of AI-Mediated Political Outreach: Persuasion and AI Penalties in the US and UK
+
+- **ID:** `2603.27413` · **primary:** `cs.CY` · **Submitted:** 2026-03-28
+- **Authors:** Andreas Jungherr, Adrian Rauchfleisch
+- **abs:** https://arxiv.org/abs/2603.27413 · **PDF:** https://arxiv.org/pdf/2603.27413v1
+
+## 1250. The Geometry of Harmful Intent: Training-Free Anomaly Detection via Angular Deviation in LLM Residual Streams
+
+- **ID:** `2603.27412` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Isaac Llorente-Saguer
+- **abs:** https://arxiv.org/abs/2603.27412 · **PDF:** https://arxiv.org/pdf/2603.27412v1
+
+## 1251. Grounding Social Perception in Intuitive Physics
+
+- **ID:** `2603.27410` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-28
+- **Authors:** Lance Ying, Aydan Y. Huang, Aviv Netanyahu, et al.
+- **abs:** https://arxiv.org/abs/2603.27410 · **PDF:** https://arxiv.org/pdf/2603.27410v1
+
+## 1252. On the Relationship between Bayesian Networks and Probabilistic Structural Causal Models
+
+- **ID:** `2603.27406` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Peter J. F. Lucas, Eleanora Zullo, Fabio Stella
+- **abs:** https://arxiv.org/abs/2603.27406 · **PDF:** https://arxiv.org/pdf/2603.27406v1
+
+## 1253. Heterogeneous Debate Engine: Identity-Grounded Cognitive Architecture for Resilient LLM-Based Ethical Tutoring
+
+- **ID:** `2603.27404` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Jakub Masłowski, Jarosław A. Chudziak
+- **abs:** https://arxiv.org/abs/2603.27404 · **PDF:** https://arxiv.org/pdf/2603.27404v1
+
+## 1254. Conditional Factuality Controlled LLMs with Generalization Certificates via Conformal Sampling
+
+- **ID:** `2603.27403` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Kai Ye, Qingtao Pan, Shuo Li
+- **abs:** https://arxiv.org/abs/2603.27403 · **PDF:** https://arxiv.org/pdf/2603.27403v1
+
+## 1255. Rainbow-DemoRL: Combining Improvements in Demonstration-Augmented Reinforcement Learning
+
+- **ID:** `2603.27400` · **primary:** `cs.RO` · **Submitted:** 2026-03-28
+- **Authors:** Dwait Bhatt, Shih-Chieh Chou, Nikolay Atanasov
+- **abs:** https://arxiv.org/abs/2603.27400 · **PDF:** https://arxiv.org/pdf/2603.27400v1
+
+## 1256. Topological Detection of Hopf Bifurcations via Persistent Homology: A Functional Criterion from Time Series
+
+- **ID:** `2603.27395` · **primary:** `math.DS` · **Submitted:** 2026-03-28
+- **Authors:** Jhonathan Barrios, Yásser Echávez, Carlos F. Álvarez
+- **abs:** https://arxiv.org/abs/2603.27395 · **PDF:** https://arxiv.org/pdf/2603.27395v1
+
+## 1257. K-Means Based TinyML Anomaly Detection and Distributed Model Reuse via the Distributed Internet of Learning (DIoL)
+
+- **ID:** `2603.27393` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Abdulrahman Albaiz, Fathi Amsaad
+- **abs:** https://arxiv.org/abs/2603.27393 · **PDF:** https://arxiv.org/pdf/2603.27393v1
+
+## 1258. Diagnosing Non-Markovian Observations in Reinforcement Learning via Prediction-Based Violation Scoring
+
+- **ID:** `2603.27389` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Naveen Mysore
+- **abs:** https://arxiv.org/abs/2603.27389 · **PDF:** https://arxiv.org/pdf/2603.27389v1
+
+## 1259. Active In-Context Learning for Tabular Foundation Models
+
+- **ID:** `2603.27385` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Wilailuck Treerath, Fabrizio Pittorino
+- **abs:** https://arxiv.org/abs/2603.27385 · **PDF:** https://arxiv.org/pdf/2603.27385v1
+
+## 1260. GUARD-SLM: Token Activation-Based Defense Against Jailbreak Attacks for Small Language Models
+
+- **ID:** `2603.28817` · **primary:** `cs.CR` · **Submitted:** 2026-03-28
+- **Authors:** Md Jueal Mia, Joaquin Molto, Yanzhao Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.28817 · **PDF:** https://arxiv.org/pdf/2603.28817v1
+
+## 1261. Where Does AI Leave a Footprint? Children's Reasoning About AI's Environmental Costs
+
+- **ID:** `2603.27376` · **primary:** `cs.HC` · **Submitted:** 2026-03-28
+- **Authors:** Aayushi Dangol, Robert Wolfe, Nisha Devasia, et al.
+- **abs:** https://arxiv.org/abs/2603.27376 · **PDF:** https://arxiv.org/pdf/2603.27376v1
+
+## 1262. The Risk Quadrangle in Optimization: An Overview with Recent Results and Extensions
+
+- **ID:** `2603.27370` · **primary:** `math.OC` · **Submitted:** 2026-03-28
+- **Authors:** Bogdan Grechuk, Anton Malandii, Terry Rockafellar, et al.
+- **abs:** https://arxiv.org/abs/2603.27370 · **PDF:** https://arxiv.org/pdf/2603.27370v1
+
+## 1263. Defend: Automated Rebuttals for Peer Review with Minimal Author Guidance
+
+- **ID:** `2603.27360` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Jyotsana Khatri, Manasi Patwardhan
+- **abs:** https://arxiv.org/abs/2603.27360 · **PDF:** https://arxiv.org/pdf/2603.27360v1
+
+## 1264. Not Worth Mentioning? A Pilot Study on Salient Proposition Annotation
+
+- **ID:** `2603.27358` · **primary:** `cs.CL` · **Submitted:** 2026-03-28
+- **Authors:** Amir Zeldes, Katherine Conhaim, Lauren Levine
+- **abs:** https://arxiv.org/abs/2603.27358 · **PDF:** https://arxiv.org/pdf/2603.27358v1
+
+## 1265. Guided Lensless Polarization Imaging
+
+- **ID:** `2603.27357` · **primary:** `eess.IV` · **Submitted:** 2026-03-28
+- **Authors:** Noa Kraicer, Erez Yosef, Raja Giryes
+- **abs:** https://arxiv.org/abs/2603.27357 · **PDF:** https://arxiv.org/pdf/2603.27357v1
+
+## 1266. Culturally Adaptive Explainable LLM Assessment for Multilingual Information Disorder: A Human-in-the-Loop Approach
+
+- **ID:** `2603.27356` · **primary:** `cs.CL` · **Submitted:** 2026-03-28
+- **Authors:** Maziar Kianimoghadam Jouneghani
+- **abs:** https://arxiv.org/abs/2603.27356 · **PDF:** https://arxiv.org/pdf/2603.27356v1
+
+## 1267. LLM Readiness Harness: Evaluation, Observability, and CI Gates for LLM/RAG Applications
+
+- **ID:** `2603.27355` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Alexandre Cristovão Maiorano
+- **abs:** https://arxiv.org/abs/2603.27355 · **PDF:** https://arxiv.org/pdf/2603.27355v1
+
+## 1268. Inference-Time Structural Reasoning for Compositional Vision-Language Understanding
+
+- **ID:** `2603.27349` · **primary:** `cs.CV` · **Submitted:** 2026-03-28
+- **Authors:** Amartya Bhattacharya
+- **abs:** https://arxiv.org/abs/2603.27349 · **PDF:** https://arxiv.org/pdf/2603.27349v1
+
+## 1269. Embedding Provenance in Computer Vision Datasets with JSON-LD
+
+- **ID:** `2603.27348` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Lynn Vonderhaar, Timothy Elvira, Tyler Thomas Procko, et al.
+- **abs:** https://arxiv.org/abs/2603.27348 · **PDF:** https://arxiv.org/pdf/2603.27348v1
+
+## 1270. D-SPEAR: Dual-Stream Prioritized Experience Adaptive Replay for Stable Reinforcement Learning in Robotic Manipulation
+
+- **ID:** `2603.27346` · **primary:** `cs.RO` · **Submitted:** 2026-03-28
+- **Authors:** Yu Zhang, Karl Mason
+- **abs:** https://arxiv.org/abs/2603.27346 · **PDF:** https://arxiv.org/pdf/2603.27346v2
+
+## 1271. Beyond Completion: Probing Cumulative State Tracking to Predict LLM Agent Performance
+
+- **ID:** `2603.27343` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Dengzhe Hou, Lingyu Jiang, Deng Li, et al.
+- **abs:** https://arxiv.org/abs/2603.27343 · **PDF:** https://arxiv.org/pdf/2603.27343v1
+
+## 1272. A Comparative Study in Surgical AI: Datasets, Foundation Models, and Barriers to Med-AGI
+
+- **ID:** `2603.27341` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Kirill Skobelev, Eric Fithian, Yegor Baranovski, et al.
+- **abs:** https://arxiv.org/abs/2603.27341 · **PDF:** https://arxiv.org/pdf/2603.27341v1
+
+## 1273. CounterMoral: Editing Morals in Language Models
+
+- **ID:** `2603.27338` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Michael Ripa, Jim Davies
+- **abs:** https://arxiv.org/abs/2603.27338 · **PDF:** https://arxiv.org/pdf/2603.27338v1
+
+## 1274. ARTLAS: Mapping Art-Technology Institutions via Conceptual Axes, Text Embeddings, and Unsupervised Clustering
+
+- **ID:** `2603.28816` · **primary:** `cs.DL` · **Submitted:** 2026-03-28
+- **Authors:** Joonhyung Bae
+- **abs:** https://arxiv.org/abs/2603.28816 · **PDF:** https://arxiv.org/pdf/2603.28816v1
+
+## 1275. PubMed Reasoner: Dynamic Reasoning-based Retrieval for Evidence-Grounded Biomedical Question Answering
+
+- **ID:** `2603.27335` · **primary:** `cs.CL` · **Submitted:** 2026-03-28
+- **Authors:** Yiqing Zhang, Xiaozhong Liu, Fabricio Murai
+- **abs:** https://arxiv.org/abs/2603.27335 · **PDF:** https://arxiv.org/pdf/2603.27335v1
+
+## 1276. ComBench: A Repo-level Real-world Benchmark for Compilation Error Repair
+
+- **ID:** `2603.27333` · **primary:** `cs.SE` · **Submitted:** 2026-03-28
+- **Authors:** Jia Li, Zeyang Zhuang, Zhuangbin Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.27333 · **PDF:** https://arxiv.org/pdf/2603.27333v1
+
+## 1277. SACRED: A Faithful Annotated Multimedia Multimodal Multilingual Dataset for Classifying Connectedness Types in Online Spirituality
+
+- **ID:** `2603.27331` · **primary:** `cs.CL` · **Submitted:** 2026-03-28
+- **Authors:** Qinghao Guan, Yuchen Pan, Donghao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.27331 · **PDF:** https://arxiv.org/pdf/2603.27331v1
+
+## 1278. Improving Automated Wound Assessment Using Joint Boundary Segmentation and Multi-Class Classification Models
+
+- **ID:** `2603.27325` · **primary:** `cs.CV` · **Submitted:** 2026-03-28
+- **Authors:** Mehedi Hasan Tusar, Fateme Fayyazbakhsh, Igor Melnychuk, et al.
+- **abs:** https://arxiv.org/abs/2603.27325 · **PDF:** https://arxiv.org/pdf/2603.27325v1
+
+## 1279. Multimodal Forecasting for Commodity Prices Using Spectrogram-Based and Time Series Representations
+
+- **ID:** `2603.27321` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Soyeon Park, Doohee Chung, Charmgil Hong
+- **abs:** https://arxiv.org/abs/2603.27321 · **PDF:** https://arxiv.org/pdf/2603.27321v1
+
+## 1280. Retrospective Counterfactual Prediction by Conditioning on the Factual Outcome: A Cross-World Approach
+
+- **ID:** `2603.27320` · **primary:** `stat.ME` · **Submitted:** 2026-03-28
+- **Authors:** Juraj Bodik
+- **abs:** https://arxiv.org/abs/2603.27320 · **PDF:** https://arxiv.org/pdf/2603.27320v1
+
+## 1281. TokenDance: Token-to-Token Music-to-Dance Generation with Bidirectional Mamba
+
+- **ID:** `2603.27314` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Ziyue Yang, Kaixing Yang, Xulong Tang
+- **abs:** https://arxiv.org/abs/2603.27314 · **PDF:** https://arxiv.org/pdf/2603.27314v1
+
+## 1282. Scalable Maximum Entropy Population Synthesis via Persistent Contrastive Divergence
+
+- **ID:** `2603.27312` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Mirko Degli Esposti
+- **abs:** https://arxiv.org/abs/2603.27312 · **PDF:** https://arxiv.org/pdf/2603.27312v1
+
+## 1283. GUIDE: Guided Updates for In-context Decision Evolution in LLM-Driven Spacecraft Operations
+
+- **ID:** `2603.27306` · **primary:** `cs.MA` · **Submitted:** 2026-03-28
+- **Authors:** Alejandro Carrasco, Mariko Storey-Matsutani, Victor Rodriguez-Fernandez, et al.
+- **abs:** https://arxiv.org/abs/2603.27306 · **PDF:** https://arxiv.org/pdf/2603.27306v1
+
+## 1284. EpochX: Building the Infrastructure for an Emergent Agent Civilization
+
+- **ID:** `2603.27304` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Huacan Wang, Chaofa Yuan, Xialie Zhuang, et al.
+- **abs:** https://arxiv.org/abs/2603.27304 · **PDF:** https://arxiv.org/pdf/2603.27304v1
+
+## 1285. Self-evolving AI agents for protein discovery and directed evolution
+
+- **ID:** `2603.27303` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Yang Tan, Lingrong Zhang, Mingchen Li, et al.
+- **abs:** https://arxiv.org/abs/2603.27303 · **PDF:** https://arxiv.org/pdf/2603.27303v1
+
+## 1286. From Inference Routing to Agent Orchestration: Declarative Policy Compilation with Cross-Layer Verification
+
+- **ID:** `2603.27299` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Huamin Chen, Xunzhuo Liu, Bowei He, et al.
+- **abs:** https://arxiv.org/abs/2603.27299 · **PDF:** https://arxiv.org/pdf/2603.27299v1
+
+## 1287. A Multi-agent AI System for Deep Learning Model Migration from TensorFlow to JAX
+
+- **ID:** `2603.27296` · **primary:** `cs.SE` · **Submitted:** 2026-03-28
+- **Authors:** Stoyan Nikolov, Bernhard Konrad, Moritz Gronbach, et al.
+- **abs:** https://arxiv.org/abs/2603.27296 · **PDF:** https://arxiv.org/pdf/2603.27296v1
+
+## 1288. Beyond Descriptions: A Generative Scene2Audio Framework for Blind and Low-Vision Users to Experience Vista Landscapes
+
+- **ID:** `2603.27295` · **primary:** `cs.HC` · **Submitted:** 2026-03-28
+- **Authors:** Chitralekha Gupta, Jing Peng, Ashwin Ram, et al.
+- **abs:** https://arxiv.org/abs/2603.27295 · **PDF:** https://arxiv.org/pdf/2603.27295v1
+
+## 1289. SkillTester: Benchmarking Utility and Security of Agent Skills
+
+- **ID:** `2603.28815` · **primary:** `cs.CR` · **Submitted:** 2026-03-28
+- **Authors:** Leye Wang, Zixing Wang, Anjie Xu
+- **abs:** https://arxiv.org/abs/2603.28815 · **PDF:** https://arxiv.org/pdf/2603.28815v1
+
+## 1290. StretchCast: Global-Regional AI Weather Forecasting on Stretched Cubed-Sphere Mesh
+
+- **ID:** `2603.27288` · **primary:** `physics.ao-ph` · **Submitted:** 2026-03-28
+- **Authors:** Jin Feng
+- **abs:** https://arxiv.org/abs/2603.27288 · **PDF:** https://arxiv.org/pdf/2603.27288v1
+
+## 1291. Codebase-Memory: Tree-Sitter-Based Knowledge Graphs for LLM Code Exploration via MCP
+
+- **ID:** `2603.27277` · **primary:** `cs.SE` · **Submitted:** 2026-03-28
+- **Authors:** Martin Vogel, Falk Meyer-Eschenbach, Severin Kohler, et al.
+- **abs:** https://arxiv.org/abs/2603.27277 · **PDF:** https://arxiv.org/pdf/2603.27277v1
+
+## 1292. Robust Global-Local Behavior Arbitration via Continuous Command Fusion Under LiDAR Errors
+
+- **ID:** `2603.27273` · **primary:** `cs.RO` · **Submitted:** 2026-03-28
+- **Authors:** Mohamed Elgouhary, Amr S. El-Wakeel
+- **abs:** https://arxiv.org/abs/2603.27273 · **PDF:** https://arxiv.org/pdf/2603.27273v1
+
+## 1293. Quantification of Credal Uncertainty: A Distance-Based Approach
+
+- **ID:** `2603.27270` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Xabier Gonzalez-Garcia, Siu Lun Chau, Julian Rodemann, et al.
+- **abs:** https://arxiv.org/abs/2603.27270 · **PDF:** https://arxiv.org/pdf/2603.27270v1
+
+## 1294. From Foundation ECG Models to NISQ Learners: Distilling ECGFounder into a VQC Student
+
+- **ID:** `2603.27269` · **primary:** `quant-ph` · **Submitted:** 2026-03-28
+- **Authors:** Giovanni dos Santos Franco, Felipe Mahlow, Ellison Fernando Cardoso, et al.
+- **abs:** https://arxiv.org/abs/2603.27269 · **PDF:** https://arxiv.org/pdf/2603.27269v1
+
+## 1295. Amalgam: Hybrid LLM-PGM Synthesis Algorithm for Accuracy and Realism
+
+- **ID:** `2603.27254` · **primary:** `cs.DB` · **Submitted:** 2026-03-28
+- **Authors:** Antheas Kapenekakis, Bent Thomsen, Katja Hose, et al.
+- **abs:** https://arxiv.org/abs/2603.27254 · **PDF:** https://arxiv.org/pdf/2603.27254v1
+
+## 1296. Mitigating Hallucination on Hallucination in RAG via Ensemble Voting
+
+- **ID:** `2603.27253` · **primary:** `cs.CL` · **Submitted:** 2026-03-28
+- **Authors:** Zequn Xie, Zhengyang Sun
+- **abs:** https://arxiv.org/abs/2603.27253 · **PDF:** https://arxiv.org/pdf/2603.27253v1
+
+## 1297. Zero-shot Vision-Language Reranking for Cross-View Geolocalization
+
+- **ID:** `2603.27251` · **primary:** `cs.CV` · **Submitted:** 2026-03-28
+- **Authors:** Yunus Talha Erzurumlu, John E. Anderson, William J. Shuart, et al.
+- **abs:** https://arxiv.org/abs/2603.27251 · **PDF:** https://arxiv.org/pdf/2603.27251v1
+
+## 1298. SCOPE: Tree-based Self-Correcting Online Log Parsing via Syntactic-Semantic Collaboration
+
+- **ID:** `2603.27247` · **primary:** `cs.CL` · **Submitted:** 2026-03-28
+- **Authors:** Dongyi Fan, Suqiong Zhang, Lili He, et al.
+- **abs:** https://arxiv.org/abs/2603.27247 · **PDF:** https://arxiv.org/pdf/2603.27247v1
+
+## 1299. The impact of multi-agent debate protocols on debate quality: a controlled case study
+
+- **ID:** `2603.28813` · **primary:** `cs.MA` · **Submitted:** 2026-03-28
+- **Authors:** Ramtin Zargari Marandi
+- **abs:** https://arxiv.org/abs/2603.28813 · **PDF:** https://arxiv.org/pdf/2603.28813v1
+
+## 1300. Diagnosing and Repairing Unsafe Channels in Vision-Language Models via Causal Discovery and Dual-Modal Safety Subspace Projection
+
+- **ID:** `2603.27240` · **primary:** `cs.CV` · **Submitted:** 2026-03-28
+- **Authors:** Jinhu Fu, Yihang Lou, Qingyi Si, et al.
+- **abs:** https://arxiv.org/abs/2603.27240 · **PDF:** https://arxiv.org/pdf/2603.27240v1
+
+## 1301. Can pre-trained Deep Learning models predict groove ratings?
+
+- **ID:** `2603.27237` · **primary:** `cs.SD` · **Submitted:** 2026-03-28
+- **Authors:** Axel Marmoret, Nicolas Farrugia, Jan Alexander Stupacher
+- **abs:** https://arxiv.org/abs/2603.27237 · **PDF:** https://arxiv.org/pdf/2603.27237v1
+
+## 1302. Structural Stress and Learned Helplessness in Afghanistan: A Multi-Layer Analysis of the AFSTRESS Dari Corpus
+
+- **ID:** `2603.27233` · **primary:** `cs.CL` · **Submitted:** 2026-03-28
+- **Authors:** Jawid Ahmad Baktash, Mursal Dawodi, Nadira Ahmadi
+- **abs:** https://arxiv.org/abs/2603.27233 · **PDF:** https://arxiv.org/pdf/2603.27233v1
+
+## 1303. Rethinking Easy-to-Hard: Limits of Curriculum Learning in Post-Training for Deductive Reasoning
+
+- **ID:** `2603.27226` · **primary:** `cs.CL` · **Submitted:** 2026-03-28
+- **Authors:** Maximilian Mordig, Andreas Opedal, Weiyang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.27226 · **PDF:** https://arxiv.org/pdf/2603.27226v1
+
+## 1304. EuraGovExam: A Multilingual Multimodal Benchmark from Real-World Civil Service Exams
+
+- **ID:** `2603.27223` · **primary:** `cs.CV` · **Submitted:** 2026-03-28
+- **Authors:** JaeSeong Kim, Chaehwan Lim, Sang Hyun Gil, et al.
+- **abs:** https://arxiv.org/abs/2603.27223 · **PDF:** https://arxiv.org/pdf/2603.27223v1
+
+## 1305. Unsupervised Evaluation of Deep Audio Embeddings for Music Structure Analysis
+
+- **ID:** `2603.27218` · **primary:** `cs.SD` · **Submitted:** 2026-03-28
+- **Authors:** Axel Marmoret
+- **abs:** https://arxiv.org/abs/2603.27218 · **PDF:** https://arxiv.org/pdf/2603.27218v1
+
+## 1306. LightMover: Generative Light Movement with Color and Intensity Controls
+
+- **ID:** `2603.27209` · **primary:** `cs.CV` · **Submitted:** 2026-03-28
+- **Authors:** Gengze Zhou, Tianyu Wang, Soo Ye Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.27209 · **PDF:** https://arxiv.org/pdf/2603.27209v1
+
+## 1307. Data-Driven Estimation of the interfacial Dzyaloshinskii-Moriya Interaction with Machine Learning
+
+- **ID:** `2603.28812` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-28
+- **Authors:** Davi Rodrigues, Andrea Meo, Ali Hasan, et al.
+- **abs:** https://arxiv.org/abs/2603.28812 · **PDF:** https://arxiv.org/pdf/2603.28812v1
+
+## 1308. AutoMS: Multi-Agent Evolutionary Search for Cross-Physics Inverse Microstructure Design
+
+- **ID:** `2603.27195` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Zhenyuan Zhao, Yu Xing, Tianyang Xue, et al.
+- **abs:** https://arxiv.org/abs/2603.27195 · **PDF:** https://arxiv.org/pdf/2603.27195v1
+
+## 1309. Multi-AUV Ad-hoc Networks-Based Multi-Target Tracking Based on Scene-Adaptive Embodied Intelligence
+
+- **ID:** `2603.27194` · **primary:** `cs.RO` · **Submitted:** 2026-03-28
+- **Authors:** Kai Tian, Jialun Wang, Chuan Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.27194 · **PDF:** https://arxiv.org/pdf/2603.27194v1
+
+## 1310. Conformal Prediction Assessment: A Framework for Conditional Coverage Evaluation and Selection
+
+- **ID:** `2603.27189` · **primary:** `stat.ME` · **Submitted:** 2026-03-28
+- **Authors:** Zheng Zhou, Xiangfei Zhang, Chongguang Tao, et al.
+- **abs:** https://arxiv.org/abs/2603.27189 · **PDF:** https://arxiv.org/pdf/2603.27189v1
+
+## 1311. Persistent Memory Through Triple-Loop Consolidation in a Non-Gradient Dissipative Cognitive Architecture
+
+- **ID:** `2603.27188` · **primary:** `cs.NE` · **Submitted:** 2026-03-28
+- **Authors:** Jianwei Lou
+- **abs:** https://arxiv.org/abs/2603.27188 · **PDF:** https://arxiv.org/pdf/2603.27188v1
+
+## 1312. Omni-Modal Dissonance Benchmark: Systematically Breaking Modality Consensus to Probe Robustness and Calibrated Abstention
+
+- **ID:** `2603.27187` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Zabir Al Nazi, Shubhashis Roy Dipta, Md Rizwan Parvez
+- **abs:** https://arxiv.org/abs/2603.27187 · **PDF:** https://arxiv.org/pdf/2603.27187v1
+
+## 1313. Hybrid Deep Learning with Temporal Data Augmentation for Accurate Remaining Useful Life Prediction of Lithium-Ion Batteries
+
+- **ID:** `2603.27186` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Yun Tian, Guili Wang, Jian Bi, et al.
+- **abs:** https://arxiv.org/abs/2603.27186 · **PDF:** https://arxiv.org/pdf/2603.27186v1
+
+## 1314. An End-to-end Flight Control Network for High-speed UAV Obstacle Avoidance based on Event-Depth Fusion
+
+- **ID:** `2603.27181` · **primary:** `cs.RO` · **Submitted:** 2026-03-28
+- **Authors:** Dikai Shang, Jingyue Zhao, Shi Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.27181 · **PDF:** https://arxiv.org/pdf/2603.27181v1
+
+## 1315. Aligning LLMs with Graph Neural Solvers for Combinatorial Optimization
+
+- **ID:** `2603.27169` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Shaodi Feng, Zhuoyi Lin, Yaoxin Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.27169 · **PDF:** https://arxiv.org/pdf/2603.27169v1
+
+## 1316. daVinci-LLM:Towards the Science of Pretraining
+
+- **ID:** `2603.27164` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Yiwei Qin, Yixiu Liu, Tiantian Mi, et al.
+- **abs:** https://arxiv.org/abs/2603.27164 · **PDF:** https://arxiv.org/pdf/2603.27164v1
+
+## 1317. Online Learning of Kalman Filtering: From Output to State Estimation
+
+- **ID:** `2603.27159` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Lintao Ye, Ankang Zhang, Ming Chi, et al.
+- **abs:** https://arxiv.org/abs/2603.27159 · **PDF:** https://arxiv.org/pdf/2603.27159v1
+
+## 1318. Weakly Convex Ridge Regularization for 3D Non-Cartesian MRI Reconstruction
+
+- **ID:** `2603.27158` · **primary:** `cs.CV` · **Submitted:** 2026-03-28
+- **Authors:** German Shâma Wache, Chaithya G R, Asma Tanabene, et al.
+- **abs:** https://arxiv.org/abs/2603.27158 · **PDF:** https://arxiv.org/pdf/2603.27158v1
+
+## 1319. GSR-GNN: Training Acceleration and Memory-Saving Framework of Deep GNNs on Circuit Graph
+
+- **ID:** `2603.27156` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Yuebo Luo, Shiyang Li, Yifei Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.27156 · **PDF:** https://arxiv.org/pdf/2603.27156v1
+
+## 1320. A Tight Expressivity Hierarchy for GNN-Based Entity Resolution in Master Data Management
+
+- **ID:** `2603.27154` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Ashwin Ganesan
+- **abs:** https://arxiv.org/abs/2603.27154 · **PDF:** https://arxiv.org/pdf/2603.27154v1
+
+## 1321. Preconditioned Attention: Enhancing Efficiency in Transformers
+
+- **ID:** `2603.27153` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Hemanth Saratchandran
+- **abs:** https://arxiv.org/abs/2603.27153 · **PDF:** https://arxiv.org/pdf/2603.27153v1
+
+## 1322. WAter: A Workload-Adaptive Knob Tuning System based on Workload Compression
+
+- **ID:** `2603.28809` · **primary:** `cs.DB` · **Submitted:** 2026-03-28
+- **Authors:** Yibo Wang, Jiale Lao, Chen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.28809 · **PDF:** https://arxiv.org/pdf/2603.28809v1
+
+## 1323. MediHive: A Decentralized Agent Collective for Medical Reasoning
+
+- **ID:** `2603.27150` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Xiaoyang Wang, Christopher C. Yang
+- **abs:** https://arxiv.org/abs/2603.27150 · **PDF:** https://arxiv.org/pdf/2603.27150v1
+
+## 1324. SafetyDrift: Predicting When AI Agents Cross the Line Before They Actually Do
+
+- **ID:** `2603.27148` · **primary:** `cs.CR` · **Submitted:** 2026-03-28
+- **Authors:** Aditya Dhodapkar, Farhaan Pishori
+- **abs:** https://arxiv.org/abs/2603.27148 · **PDF:** https://arxiv.org/pdf/2603.27148v1
+
+## 1325. Learning to Predict Future-Aligned Research Proposals with Language Models
+
+- **ID:** `2603.27146` · **primary:** `cs.CL` · **Submitted:** 2026-03-28
+- **Authors:** Heng Wang, Pengcheng Jiang, Jiashuo Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.27146 · **PDF:** https://arxiv.org/pdf/2603.27146v1
+
+## 1326. Pan-Cancer Mapping of the Tumor Immune Landscape through Metagene Clustering and Predictive Modeling
+
+- **ID:** `2603.27145` · **primary:** `q-bio.GN` · **Submitted:** 2026-03-28
+- **Authors:** Soham Chatterjee
+- **abs:** https://arxiv.org/abs/2603.27145 · **PDF:** https://arxiv.org/pdf/2603.27145v1
+
+## 1327. Follow Your Heart: Landmark-Guided Transducer Pose Scoring for Point-of-Care Echocardiography
+
+- **ID:** `2603.27143` · **primary:** `cs.CV` · **Submitted:** 2026-03-28
+- **Authors:** Zaiyang Guo, Jessie N. Dong, Filippos Bellos, et al.
+- **abs:** https://arxiv.org/abs/2603.27143 · **PDF:** https://arxiv.org/pdf/2603.27143v1
+
+## 1328. Bayes-MICE: A Bayesian Approach to Multiple Imputation for Time Series Data
+
+- **ID:** `2603.27142` · **primary:** `stat.ML` · **Submitted:** 2026-03-28
+- **Authors:** Amuche Ibenegbu, Pierre Lafaye de Micheaux, Rohitash Chandra
+- **abs:** https://arxiv.org/abs/2603.27142 · **PDF:** https://arxiv.org/pdf/2603.27142v1
+
+## 1329. Routing Sensitivity Without Controllability: A Diagnostic Study of Fairness in MoE Language Models
+
+- **ID:** `2603.27141` · **primary:** `cs.CL` · **Submitted:** 2026-03-28
+- **Authors:** Junhyeok Lee, Kyu Sung Choi
+- **abs:** https://arxiv.org/abs/2603.27141 · **PDF:** https://arxiv.org/pdf/2603.27141v1
+
+## 1330. ScoutAttention: Efficient KV Cache Offloading via Layer-Ahead CPU Pre-computation for LLM Inference
+
+- **ID:** `2603.27138` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Qiuyang Zhang, Kai Zhou, Ding Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.27138 · **PDF:** https://arxiv.org/pdf/2603.27138v1
+
+## 1331. A Mean Field Games Perspective on Evolutionary Clustering
+
+- **ID:** `2603.27137` · **primary:** `math.NA` · **Submitted:** 2026-03-28
+- **Authors:** Alessio Basti, Fabio Camilli, Adriano Festa
+- **abs:** https://arxiv.org/abs/2603.27137 · **PDF:** https://arxiv.org/pdf/2603.27137v1
+
+## 1332. Spectral-Aware Text-to-Time Series Generation with Billion-Scale Multimodal Meteorological Data
+
+- **ID:** `2603.27135` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Shijie Zhang
+- **abs:** https://arxiv.org/abs/2603.27135 · **PDF:** https://arxiv.org/pdf/2603.27135v1
+
+## 1333. Semantic Interaction Information mediates compositional generalization in latent space
+
+- **ID:** `2603.27134` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** John Schwarcz
+- **abs:** https://arxiv.org/abs/2603.27134 · **PDF:** https://arxiv.org/pdf/2603.27134v2
+
+## 1334. Bayesian-Symbolic Integration for Uncertainty-Aware Parking Prediction
+
+- **ID:** `2603.27119` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Alireza Nezhadettehad, Arkady Zaslavsky, Abdur Rakib, et al.
+- **abs:** https://arxiv.org/abs/2603.27119 · **PDF:** https://arxiv.org/pdf/2603.27119v1
+
+## 1335. Gender-Based Heterogeneity in Youth Privacy-Protective Behavior for Smart Voice Assistants: Evidence from Multigroup PLS-SEM
+
+- **ID:** `2603.27117` · **primary:** `cs.CR` · **Submitted:** 2026-03-28
+- **Authors:** Molly Campbell, Yulia Bobkova, Ajay Kumar Shrestha
+- **abs:** https://arxiv.org/abs/2603.27117 · **PDF:** https://arxiv.org/pdf/2603.27117v1
+
+## 1336. The Price of Meaning: Why Every Semantic Memory System Forgets
+
+- **ID:** `2603.27116` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Sambartha Ray Barman, Andrey Starenky, Sofia Bodnar, et al.
+- **abs:** https://arxiv.org/abs/2603.27116 · **PDF:** https://arxiv.org/pdf/2603.27116v1
+
+## 1337. Maximin Learning of Individualized Treatment Effect on Multi-Domain Outcomes
+
+- **ID:** `2603.27114` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Yuying Lu, Wenbo Fei, Yuanjia Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.27114 · **PDF:** https://arxiv.org/pdf/2603.27114v1
+
+## 1338. Hierarchy-Guided Topology Latent Flow for Molecular Graph Generation
+
+- **ID:** `2603.27113` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Urvi Awasthi, Alexander Arjun Lobo, Leonid Zhukov
+- **abs:** https://arxiv.org/abs/2603.27113 · **PDF:** https://arxiv.org/pdf/2603.27113v1
+
+## 1339. Autonomous Agent-Orchestrated Digital Twins (AADT): Leveraging the OpenClaw Framework for State Synchronization in Rare Genetic Disorders
+
+- **ID:** `2603.27104` · **primary:** `q-bio.QM` · **Submitted:** 2026-03-28
+- **Authors:** Hongzhuo Chen, Zhanliang Wang, Quan M. Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.27104 · **PDF:** https://arxiv.org/pdf/2603.27104v1
+
+## 1340. PRUE: A Practical Recipe for Field Boundary Segmentation at Scale
+
+- **ID:** `2603.27101` · **primary:** `cs.CV` · **Submitted:** 2026-03-28
+- **Authors:** Gedeon Muhawenayo, Caleb Robinson, Subash Khanal, et al.
+- **abs:** https://arxiv.org/abs/2603.27101 · **PDF:** https://arxiv.org/pdf/2603.27101v1
+
+## 1341. Sovereign Context Protocol: An Open Attribution Layer for Human-Generated Content in the Age of Large Language Models
+
+- **ID:** `2603.27094` · **primary:** `cs.CR` · **Submitted:** 2026-03-28
+- **Authors:** Praneel Panchigar, Torlach Rush, Matthew Canabarro
+- **abs:** https://arxiv.org/abs/2603.27094 · **PDF:** https://arxiv.org/pdf/2603.27094v1
+
+## 1342. RDEx-MOP: Indicator-Guided Reconstructed Differential Evolution for Fixed-Budget Multiobjective Optimization
+
+- **ID:** `2603.27092` · **primary:** `cs.NE` · **Submitted:** 2026-03-28
+- **Authors:** Sichen Tao, Yifei Yang, Ruihan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.27092 · **PDF:** https://arxiv.org/pdf/2603.27092v1
+
+## 1343. RDEx-CSOP: Feasibility-Aware Reconstructed Differential Evolution with Adaptive epsilon-Constraint Ranking
+
+- **ID:** `2603.27090` · **primary:** `cs.NE` · **Submitted:** 2026-03-28
+- **Authors:** Sichen Tao, Yifei Yang, Ruihan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.27090 · **PDF:** https://arxiv.org/pdf/2603.27090v1
+
+## 1344. RDEx-SOP: Exploitation-Biased Reconstructed Differential Evolution for Fixed-Budget Bound-Constrained Single-Objective Optimization
+
+- **ID:** `2603.27089` · **primary:** `cs.NE` · **Submitted:** 2026-03-28
+- **Authors:** Sichen Tao, Yifei Yang, Ruihan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.27089 · **PDF:** https://arxiv.org/pdf/2603.27089v1
+
+## 1345. When Verification Hurts: Asymmetric Effects of Multi-Agent Feedback in Logic Proof Tutoring
+
+- **ID:** `2603.27076` · **primary:** `cs.AI` · **Submitted:** 2026-03-28
+- **Authors:** Tahreem Yasir, Sutapa Dey Tithi, Benyamin Tabarsi, et al.
+- **abs:** https://arxiv.org/abs/2603.27076 · **PDF:** https://arxiv.org/pdf/2603.27076v1
+
+## 1346. Forecastability as an Information-Theoretic Limit on Prediction
+
+- **ID:** `2603.27074` · **primary:** `stat.AP` · **Submitted:** 2026-03-28
+- **Authors:** Peter Maurice Catt
+- **abs:** https://arxiv.org/abs/2603.27074 · **PDF:** https://arxiv.org/pdf/2603.27074v1
+
+## 1347. Voice-based debate with an AI adversary is associated with increased divergent ideation
+
+- **ID:** `2603.27073` · **primary:** `cs.HC` · **Submitted:** 2026-03-28
+- **Authors:** Neelam Modi Jain, Dan J. Wang
+- **abs:** https://arxiv.org/abs/2603.27073 · **PDF:** https://arxiv.org/pdf/2603.27073v1
+
+## 1348. On the Loss Landscape Geometry of Regularized Deep Matrix Factorization: Uniqueness and Sharpness
+
+- **ID:** `2603.27072` · **primary:** `stat.ML` · **Submitted:** 2026-03-28
+- **Authors:** Anil Kamber, Rahul Parhi
+- **abs:** https://arxiv.org/abs/2603.27072 · **PDF:** https://arxiv.org/pdf/2603.27072v1
+
+## 1349. Dynamic resource matching in manufacturing using deep reinforcement learning
+
+- **ID:** `2603.27066` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Saunak Kumar Panda, Yisha Xiang, Ruiqi Liu
+- **abs:** https://arxiv.org/abs/2603.27066 · **PDF:** https://arxiv.org/pdf/2603.27066v1
+
+## 1350. Story2Proposal: A Scaffold for Structured Scientific Paper Writing
+
+- **ID:** `2603.27065` · **primary:** `cs.CL` · **Submitted:** 2026-03-28
+- **Authors:** Zhuoyang Qian, Wei Shi, Xu Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.27065 · **PDF:** https://arxiv.org/pdf/2603.27065v1
+
+## 1351. ChartNet: A Million-Scale, High-Quality Multimodal Dataset for Robust Chart Understanding
+
+- **ID:** `2603.27064` · **primary:** `cs.CV` · **Submitted:** 2026-03-28
+- **Authors:** Jovana Kondic, Pengyuan Li, Dhiraj Joshi, et al.
+- **abs:** https://arxiv.org/abs/2603.27064 · **PDF:** https://arxiv.org/pdf/2603.27064v1
+
+## 1352. Conformalized Signal Temporal Logic Inference under Covariate Shift
+
+- **ID:** `2603.27062` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Yixuan Wang, Danyang Li, Matthew Cleaveland, et al.
+- **abs:** https://arxiv.org/abs/2603.27062 · **PDF:** https://arxiv.org/pdf/2603.27062v1
+
+## 1353. Liquid Networks with Mixture Density Heads for Efficient Imitation Learning
+
+- **ID:** `2603.27058` · **primary:** `cs.LG` · **Submitted:** 2026-03-28
+- **Authors:** Nikolaus Correll
+- **abs:** https://arxiv.org/abs/2603.27058 · **PDF:** https://arxiv.org/pdf/2603.27058v1
+
+## 1354. Debiasing Large Language Models toward Social Factors in Online Behavior Analytics through Prompt Knowledge Tuning
+
+- **ID:** `2603.27057` · **primary:** `cs.CL` · **Submitted:** 2026-03-28
+- **Authors:** Hossein Salemi, Jitin Krishnan, Hemant Purohit
+- **abs:** https://arxiv.org/abs/2603.27057 · **PDF:** https://arxiv.org/pdf/2603.27057v1
+
+## 1355. Persona-Based Simulation of Human Opinion at Population Scale
+
+- **ID:** `2603.27056` · **primary:** `cs.CY` · **Submitted:** 2026-03-28
+- **Authors:** Mao Li, Frederick G. Conrad
+- **abs:** https://arxiv.org/abs/2603.27056 · **PDF:** https://arxiv.org/pdf/2603.27056v1
+
+## 1356. Text Data Integration
+
+- **ID:** `2603.27055` · **primary:** `cs.CL` · **Submitted:** 2026-03-28
+- **Authors:** Md Ataur Rahman, Dimitris Sacharidis, Oscar Romero, et al.
+- **abs:** https://arxiv.org/abs/2603.27055 · **PDF:** https://arxiv.org/pdf/2603.27055v1
+
+## 1357. Generalizable Foundation Models for Calorimetry via Mixtures-of-Experts and Parameter Efficient Fine Tuning
+
+- **ID:** `2603.28804` · **primary:** `physics.ins-det` · **Submitted:** 2026-03-27
+- **Authors:** Carlos Cardona-Giraldo, Cristiano Fanelli, James Giroux, et al.
+- **abs:** https://arxiv.org/abs/2603.28804 · **PDF:** https://arxiv.org/pdf/2603.28804v1
+
+## 1358. Multi-Level Barriers to Generative AI Adoption Across Disciplines and Professional Roles in Higher Education
+
+- **ID:** `2603.27052` · **primary:** `cs.CY` · **Submitted:** 2026-03-27
+- **Authors:** Jianhua Yang, Kerem Öge, Adrian von Mühlenen, et al.
+- **abs:** https://arxiv.org/abs/2603.27052 · **PDF:** https://arxiv.org/pdf/2603.27052v1
+
+## 1359. Overcoming the Incentive Collapse Paradox
+
+- **ID:** `2603.27049` · **primary:** `stat.ML` · **Submitted:** 2026-03-27
+- **Authors:** Qichuan Yin, Ziwei Su, Shuangning Li
+- **abs:** https://arxiv.org/abs/2603.27049 · **PDF:** https://arxiv.org/pdf/2603.27049v1
+
+## 1360. ML-Enabled Open RAN: A Comprehensive Survey of Architectures, Challenges, and Opportunities
+
+- **ID:** `2604.01239` · **primary:** `cs.NI` · **Submitted:** 2026-03-27
+- **Authors:** Mira Chandra Kirana, Patatchona Keyela, Fatemeh Rostamian, et al.
+- **abs:** https://arxiv.org/abs/2604.01239 · **PDF:** https://arxiv.org/pdf/2604.01239v1
+
+## 1361. Unsupervised Behavioral Compression: Learning Low-Dimensional Policy Manifolds through State-Occupancy Matching
+
+- **ID:** `2603.27044` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Andrea Fraschini, Davide Tenedini, Riccardo Zamboni, et al.
+- **abs:** https://arxiv.org/abs/2603.27044 · **PDF:** https://arxiv.org/pdf/2603.27044v2
+
+## 1362. Introducing MELI: the Mandarin-English Language Interview Corpus
+
+- **ID:** `2603.27043` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Suyuan Liu, Molly Babel
+- **abs:** https://arxiv.org/abs/2603.27043 · **PDF:** https://arxiv.org/pdf/2603.27043v1
+
+## 1363. CREST: Constraint-Release Execution for Multi-Robot Warehouse Shelf Rearrangement
+
+- **ID:** `2603.28803` · **primary:** `cs.RO` · **Submitted:** 2026-03-27
+- **Authors:** Jiaqi Tan, Yudong Luo, Sophia Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.28803 · **PDF:** https://arxiv.org/pdf/2603.28803v1
+
+## 1364. Material Identification using Multi-Modal Intrinsic Radiation and Radiography
+
+- **ID:** `2603.27036` · **primary:** `physics.ins-det` · **Submitted:** 2026-03-27
+- **Authors:** Khoa Nguyen, Brendt Wohlberg, Oleg Korobkin, et al.
+- **abs:** https://arxiv.org/abs/2603.27036 · **PDF:** https://arxiv.org/pdf/2603.27036v1
+
+## 1365. YOLO Object Detectors for Robotics -- a Comparative Study
+
+- **ID:** `2603.27029` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Patryk Niżeniec, Marcin Iwanowski, Marcin Gahbler
+- **abs:** https://arxiv.org/abs/2603.27029 · **PDF:** https://arxiv.org/pdf/2603.27029v1
+
+## 1366. TAPS: Task Aware Proposal Distributions for Speculative Sampling
+
+- **ID:** `2603.27027` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Mohamad Zbib, Mohamad Bazzi, Ammar Mohanna, et al.
+- **abs:** https://arxiv.org/abs/2603.27027 · **PDF:** https://arxiv.org/pdf/2603.27027v1
+
+## 1367. Pashto Common Voice: Building the First Open Speech Corpus for a 60-Million-Speaker Low-Resource Language
+
+- **ID:** `2603.27021` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Hanif Rahman, Shafeeq ur Rehman
+- **abs:** https://arxiv.org/abs/2603.27021 · **PDF:** https://arxiv.org/pdf/2603.27021v1
+
+## 1368. Trustworthy AI-Driven Dynamic Hybrid RIS: Joint Optimization and Reward Poisoning-Resilient Control in Cognitive MISO Networks
+
+- **ID:** `2604.01238` · **primary:** `cs.NI` · **Submitted:** 2026-03-27
+- **Authors:** Deemah H. Tashman, Soumaya Cherkaoui
+- **abs:** https://arxiv.org/abs/2604.01238 · **PDF:** https://arxiv.org/pdf/2604.01238v1
+
+## 1369. Parameter Estimation in Stochastic Differential Equations via Wiener Chaos Expansion and Stochastic Gradient Descent
+
+- **ID:** `2603.27019` · **primary:** `stat.ML` · **Submitted:** 2026-03-27
+- **Authors:** Francisco Delgado-Vences, José Julián Pavón-Español, Arelly Ornelas
+- **abs:** https://arxiv.org/abs/2603.27019 · **PDF:** https://arxiv.org/pdf/2603.27019v1
+
+## 1370. On-Device Super Resolution Imaging Using Low-Cost SPAD Array and Embedded Lightweight Deep Learning
+
+- **ID:** `2603.27018` · **primary:** `eess.IV` · **Submitted:** 2026-03-27
+- **Authors:** Zhenya Zang, Xingda Li, David Day Uei Li
+- **abs:** https://arxiv.org/abs/2603.27018 · **PDF:** https://arxiv.org/pdf/2603.27018v1
+
+## 1371. Generative Shape Reconstruction with Geometry-Guided Langevin Dynamics
+
+- **ID:** `2603.27016` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Linus Härenstam-Nielsen, Dmitrii Pozdeev, Thomas Dagès, et al.
+- **abs:** https://arxiv.org/abs/2603.27016 · **PDF:** https://arxiv.org/pdf/2603.27016v1
+
+## 1372. UMI-Underwater: Learning Underwater Manipulation without Underwater Teleoperation
+
+- **ID:** `2603.27012` · **primary:** `cs.RO` · **Submitted:** 2026-03-27
+- **Authors:** Hao Li, Long Yin Chung, Jack Goler, et al.
+- **abs:** https://arxiv.org/abs/2603.27012 · **PDF:** https://arxiv.org/pdf/2603.27012v1
+
+## 1373. RASPRef: Retrieval-Augmented Self-Supervised Prompt Refinement for Large Reasoning Models
+
+- **ID:** `2603.27008` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Rahul Soni
+- **abs:** https://arxiv.org/abs/2603.27008 · **PDF:** https://arxiv.org/pdf/2603.27008v1
+
+## 1374. The Last Fingerprint: How Markdown Training Shapes LLM Prose
+
+- **ID:** `2603.27006` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** E. M. Freeburg
+- **abs:** https://arxiv.org/abs/2603.27006 · **PDF:** https://arxiv.org/pdf/2603.27006v1
+
+## 1375. PHONOS: PHOnetic Neutralization for Online Streaming Applications
+
+- **ID:** `2603.27001` · **primary:** `eess.AS` · **Submitted:** 2026-03-27
+- **Authors:** Waris Quamer, Mu-Ruei Tseng, Ghady Nasrallah, et al.
+- **abs:** https://arxiv.org/abs/2603.27001 · **PDF:** https://arxiv.org/pdf/2603.27001v1
+
+## 1376. AutoSiMP: Autonomous Topology Optimization from Natural Language via LLM-Driven Problem Configuration and Adaptive Solver Control
+
+- **ID:** `2603.27000` · **primary:** `cs.CE` · **Submitted:** 2026-03-27
+- **Authors:** Shaoliang Yang, Jun Wang, Yunsheng Wang
+- **abs:** https://arxiv.org/abs/2603.27000 · **PDF:** https://arxiv.org/pdf/2603.27000v1
+
+## 1377. Beyond Freshness and Semantics: A Coupon-Collector Framework for Effective Status Updates
+
+- **ID:** `2603.26998` · **primary:** `eess.SY` · **Submitted:** 2026-03-27
+- **Authors:** Youssef Ahmed, Arnob Ghosh, Chih-Chun Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.26998 · **PDF:** https://arxiv.org/pdf/2603.26998v1
+
+## 1378. FormalProofBench: Can Models Write Graduate Level Math Proofs That Are Formally Verified?
+
+- **ID:** `2603.26996` · **primary:** `cs.AI` · **Submitted:** 2026-03-27
+- **Authors:** Nikil Ravi, Kexing Ying, Vasilii Nesterov, et al.
+- **abs:** https://arxiv.org/abs/2603.26996 · **PDF:** https://arxiv.org/pdf/2603.26996v1
+
+## 1379. ImmSET: Sequence-Based Predictor of TCR-pMHC Specificity at Scale
+
+- **ID:** `2603.26994` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Marco Garcia Noceda, Matthew T Noakes, Andrew FigPope, et al.
+- **abs:** https://arxiv.org/abs/2603.26994 · **PDF:** https://arxiv.org/pdf/2603.26994v1
+
+## 1380. On the Reliability Limits of LLM-Based Multi-Agent Planning
+
+- **ID:** `2603.26993` · **primary:** `cs.MA` · **Submitted:** 2026-03-27
+- **Authors:** Ruicheng Ao, Siyang Gao, David Simchi-Levi
+- **abs:** https://arxiv.org/abs/2603.26993 · **PDF:** https://arxiv.org/pdf/2603.26993v1
+
+## 1381. A large corpus of lucid and non-lucid dream reports
+
+- **ID:** `2603.26992` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Remington Mallett
+- **abs:** https://arxiv.org/abs/2603.26992 · **PDF:** https://arxiv.org/pdf/2603.26992v1
+
+## 1382. Transparency as Architecture: Structural Compliance Gaps in EU AI Act Article 50 II
+
+- **ID:** `2603.26983` · **primary:** `cs.AI` · **Submitted:** 2026-03-27
+- **Authors:** Vera Schmitt, Niklas Kruse, Premtim Sahitaj, et al.
+- **abs:** https://arxiv.org/abs/2603.26983 · **PDF:** https://arxiv.org/pdf/2603.26983v1
+
+## 1383. Online Statistical Inference of Constant Sample-averaged Q-Learning
+
+- **ID:** `2603.26982` · **primary:** `stat.ML` · **Submitted:** 2026-03-27
+- **Authors:** Saunak Kumar Panda, Tong Li, Ruiqi Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.26982 · **PDF:** https://arxiv.org/pdf/2603.26982v1
+
+## 1384. Probabilistic Forecasting of Localized Wildfire Spread Based on Conditional Flow Matching
+
+- **ID:** `2603.26975` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Bryan Shaddy, Haitong Qin, Brianna Binder, et al.
+- **abs:** https://arxiv.org/abs/2603.26975 · **PDF:** https://arxiv.org/pdf/2603.26975v1
+
+## 1385. Graph Attention Network-Based Detection of Autism Spectrum Disorder
+
+- **ID:** `2603.26971` · **primary:** `stat.AP` · **Submitted:** 2026-03-27
+- **Authors:** Abigail Kelly, Ramchandra Rimal, Arpan Sainju
+- **abs:** https://arxiv.org/abs/2603.26971 · **PDF:** https://arxiv.org/pdf/2603.26971v1
+
+## 1386. Neural Approximation of Generalized Voronoi Diagrams
+
+- **ID:** `2603.26964` · **primary:** `cs.CG` · **Submitted:** 2026-03-27
+- **Authors:** Panagiotis Rigas, George Ioannakis, Ioannis Emiris
+- **abs:** https://arxiv.org/abs/2603.26964 · **PDF:** https://arxiv.org/pdf/2603.26964v1
+
+## 1387. On the Optimal Number of Grids for Differentially Private Non-Interactive $K$-Means Clustering
+
+- **ID:** `2603.26963` · **primary:** `cs.CR` · **Submitted:** 2026-03-27
+- **Authors:** Gokularam Muthukrishnan, Anshoo Tandon
+- **abs:** https://arxiv.org/abs/2603.26963 · **PDF:** https://arxiv.org/pdf/2603.26963v1
+
+## 1388. High dimensional theory of two-phase optimizers
+
+- **ID:** `2603.26954` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Atish Agarwala
+- **abs:** https://arxiv.org/abs/2603.26954 · **PDF:** https://arxiv.org/pdf/2603.26954v1
+
+## 1389. ASTER -- Agentic Science Toolkit for Exoplanet Research
+
+- **ID:** `2603.26953` · **primary:** `astro-ph.EP` · **Submitted:** 2026-03-27
+- **Authors:** Emilie Panek, Alexander Roman, Gaurav Shukla, et al.
+- **abs:** https://arxiv.org/abs/2603.26953 · **PDF:** https://arxiv.org/pdf/2603.26953v1
+
+## 1390. Multimodal Deep Learning for Diabetic Foot Ulcer Staging Using Integrated RGB and Thermal Imaging
+
+- **ID:** `2603.26952` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Gulengul Mermer, Mustafa Furkan Aksu, Gozde Ozsezer, et al.
+- **abs:** https://arxiv.org/abs/2603.26952 · **PDF:** https://arxiv.org/pdf/2603.26952v1
+
+## 1391. Compliance-Aware Predictive Process Monitoring: A Neuro-Symbolic Approach
+
+- **ID:** `2603.26948` · **primary:** `cs.AI` · **Submitted:** 2026-03-27
+- **Authors:** Fabrizio De Santis, Gyunam Park, Wil M. P. van der Aalst, et al.
+- **abs:** https://arxiv.org/abs/2603.26948 · **PDF:** https://arxiv.org/pdf/2603.26948v2
+
+## 1392. Neuro-Symbolic Learning for Predictive Process Monitoring via Two-Stage Logic Tensor Networks with Rule Pruning
+
+- **ID:** `2603.26944` · **primary:** `cs.AI` · **Submitted:** 2026-03-27
+- **Authors:** Fabrizio De Santis, Gyunam Park, Francesco Zanichelli
+- **abs:** https://arxiv.org/abs/2603.26944 · **PDF:** https://arxiv.org/pdf/2603.26944v1
+
+## 1393. Static and Dynamic Approaches to Computing Barycenters of Probability Measures on Graphs
+
+- **ID:** `2603.26940` · **primary:** `stat.ML` · **Submitted:** 2026-03-27
+- **Authors:** David Gentile, James M. Murphy
+- **abs:** https://arxiv.org/abs/2603.26940 · **PDF:** https://arxiv.org/pdf/2603.26940v1
+
+## 1394. Multilingual Stutter Event Detection for English, German, and Mandarin Speech
+
+- **ID:** `2603.26939` · **primary:** `cs.SD` · **Submitted:** 2026-03-27
+- **Authors:** Felix Haas, Sebastian P. Bayerl
+- **abs:** https://arxiv.org/abs/2603.26939 · **PDF:** https://arxiv.org/pdf/2603.26939v1
+
+## 1395. Tunable Domain Adaptation Using Unfolding
+
+- **ID:** `2603.26931` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Snehaa Reddy, Jayaprakash Katual, Satish Mulleti
+- **abs:** https://arxiv.org/abs/2603.26931 · **PDF:** https://arxiv.org/pdf/2603.26931v1
+
+## 1396. In your own words: computationally identifying interpretable themes in free-text survey data
+
+- **ID:** `2603.26930` · **primary:** `cs.CY` · **Submitted:** 2026-03-27
+- **Authors:** Jenny S Wang, Aliya Saperstein, Emma Pierson
+- **abs:** https://arxiv.org/abs/2603.26930 · **PDF:** https://arxiv.org/pdf/2603.26930v1
+
+## 1397. Koopman Operator Identification of Model Parameter Trajectories for Temporal Domain Generalization (KOMET)
+
+- **ID:** `2603.26923` · **primary:** `stat.ML` · **Submitted:** 2026-03-27
+- **Authors:** Randy C. Hoover, Jacob James, Paul May, et al.
+- **abs:** https://arxiv.org/abs/2603.26923 · **PDF:** https://arxiv.org/pdf/2603.26923v1
+
+## 1398. Mimetic Alignment with ASPECT: Evaluation of AI-inferred Personal Profiles
+
+- **ID:** `2603.26922` · **primary:** `cs.HC` · **Submitted:** 2026-03-27
+- **Authors:** Ruoxi Shang, Dan Marshall, Edward Cutrell, et al.
+- **abs:** https://arxiv.org/abs/2603.26922 · **PDF:** https://arxiv.org/pdf/2603.26922v1
+
+## 1399. Comparing Physics-Informed and Neural ODE Approaches for Modeling Nonlinear Biological Systems: A Case Study Based on the Morris-Lecar Model
+
+- **ID:** `2603.26921` · **primary:** `math.DS` · **Submitted:** 2026-03-27
+- **Authors:** Nikolaos M. Matzakos, Chrisovalantis Sfyrakis
+- **abs:** https://arxiv.org/abs/2603.26921 · **PDF:** https://arxiv.org/pdf/2603.26921v1
+
+## 1400. Are LLMs Good For Quantum Software, Architecture, and System Design?
+
+- **ID:** `2603.26904` · **primary:** `quant-ph` · **Submitted:** 2026-03-27
+- **Authors:** Sourish Wawdhane, Poulami Das
+- **abs:** https://arxiv.org/abs/2603.26904 · **PDF:** https://arxiv.org/pdf/2603.26904v1
+
+## 1401. Magic Words or Methodical Work? Challenging Conventional Wisdom in LLM-Based Political Text Annotation
+
+- **ID:** `2603.26898` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Lorcan McLaren, James Cross, Zuzanna Krakowska, et al.
+- **abs:** https://arxiv.org/abs/2603.26898 · **PDF:** https://arxiv.org/pdf/2603.26898v2
+
+## 1402. Water-Filling is Universally Minimax Optimal
+
+- **ID:** `2603.26893` · **primary:** `cs.DS` · **Submitted:** 2026-03-27
+- **Authors:** Siddhartha Banerjee, Ramiro N. Deo-Campo Vuong, Robert Kleinberg
+- **abs:** https://arxiv.org/abs/2603.26893 · **PDF:** https://arxiv.org/pdf/2603.26893v1
+
+## 1403. Strategic Candidacy in Generative AI Arenas
+
+- **ID:** `2603.26891` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Chris Hays, Rachel Li, Bailey Flanigan, et al.
+- **abs:** https://arxiv.org/abs/2603.26891 · **PDF:** https://arxiv.org/pdf/2603.26891v1
+
+## 1404. Property-Guided Molecular Generation and Optimization via Latent Flows
+
+- **ID:** `2603.26889` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Alexander Arjun Lobo, Urvi Awasthi, Leonid Zhukov
+- **abs:** https://arxiv.org/abs/2603.26889 · **PDF:** https://arxiv.org/pdf/2603.26889v1
+
+## 1405. Learning to Commit: Generating Organic Pull Requests via Online Repository Memory
+
+- **ID:** `2603.26664` · **primary:** `cs.SE` · **Submitted:** 2026-03-27
+- **Authors:** Mo Li, L. H. Xu, Qitai Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.26664 · **PDF:** https://arxiv.org/pdf/2603.26664v1
+
+## 1406. Weight Tying Biases Token Embeddings Towards the Output Space
+
+- **ID:** `2603.26663` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Antonio Lopardo, Avyukth Harish, Catherine Arnett, et al.
+- **abs:** https://arxiv.org/abs/2603.26663 · **PDF:** https://arxiv.org/pdf/2603.26663v1
+
+## 1407. Ruka-v2: Tendon Driven Open-Source Dexterous Hand with Wrist and Abduction for Robot Learning
+
+- **ID:** `2603.26660` · **primary:** `cs.RO` · **Submitted:** 2026-03-27
+- **Authors:** Xinqi Lucas Liu, Ruoxi Hu, Alejandro Ojeda Olarte, et al.
+- **abs:** https://arxiv.org/abs/2603.26660 · **PDF:** https://arxiv.org/pdf/2603.26660v2
+
+## 1408. Tunable Soft Equivariance with Guarantees
+
+- **ID:** `2603.26657` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Md Ashiqur Rahman, Lim Jun Hao, Jeremiah Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.26657 · **PDF:** https://arxiv.org/pdf/2603.26657v1
+
+## 1409. PerceptionComp: A Video Benchmark for Complex Perception-Centric Reasoning
+
+- **ID:** `2603.26653` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Shaoxuan Li, Zhixuan Zhao, Hanze Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.26653 · **PDF:** https://arxiv.org/pdf/2603.26653v1
+
+## 1410. Vision2Web: A Hierarchical Benchmark for Visual Website Development with Agent Verification
+
+- **ID:** `2603.26648` · **primary:** `cs.SE` · **Submitted:** 2026-03-27
+- **Authors:** Zehai He, Wenyi Hong, Zhen Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.26648 · **PDF:** https://arxiv.org/pdf/2603.26648v2
+
+## 1411. An LP-based Sampling Policy for Multi-Armed Bandits with Side-Observations and Stochastic Availability
+
+- **ID:** `2603.26647` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Ashutosh Soni, Peizhong Ju, Atilla Eryilmaz, et al.
+- **abs:** https://arxiv.org/abs/2603.26647 · **PDF:** https://arxiv.org/pdf/2603.26647v1
+
+## 1412. LACON: Training Text-to-Image Model from Uncurated Data
+
+- **ID:** `2603.26866` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Zhiyang Liang, Ziyu Wan, Hongyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.26866 · **PDF:** https://arxiv.org/pdf/2603.26866v1
+
+## 1413. Automatic Laplace Collapsed Sampling: Scalable Marginalisation of Latent Parameters via Automatic Differentiation
+
+- **ID:** `2603.26644` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Toby Lovick, David Yallup, Will Handley
+- **abs:** https://arxiv.org/abs/2603.26644 · **PDF:** https://arxiv.org/pdf/2603.26644v1
+
+## 1414. Make Geometry Matter for Spatial Reasoning
+
+- **ID:** `2603.26639` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Shihua Zhang, Qiuhong Shen, Shizun Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.26639 · **PDF:** https://arxiv.org/pdf/2603.26639v1
+
+## 1415. A federated architecture for sector-led AI governance: lessons from India
+
+- **ID:** `2603.26865` · **primary:** `cs.CY` · **Submitted:** 2026-03-27
+- **Authors:** Avinash Agarwal, Manisha J. Nene
+- **abs:** https://arxiv.org/abs/2603.26865 · **PDF:** https://arxiv.org/pdf/2603.26865v1
+
+## 1416. Machine Learning Transferability for Malware Detection
+
+- **ID:** `2603.26632` · **primary:** `cs.CR` · **Submitted:** 2026-03-27
+- **Authors:** César Vieira, João Vitorino, Eva Maia, et al.
+- **abs:** https://arxiv.org/abs/2603.26632 · **PDF:** https://arxiv.org/pdf/2603.26632v1
+
+## 1417. Context-specific Credibility-aware Multimodal Fusion with Conditional Probabilistic Circuits
+
+- **ID:** `2603.26629` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Pranuthi Tenali, Sahil Sidheekh, Saurabh Mathur, et al.
+- **abs:** https://arxiv.org/abs/2603.26629 · **PDF:** https://arxiv.org/pdf/2603.26629v1
+
+## 1418. Benchmarking Tabular Foundation Models for Conditional Density Estimation in Regression
+
+- **ID:** `2603.26611` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Rafael Izbicki, Pedro L. C. Rodrigues
+- **abs:** https://arxiv.org/abs/2603.26611 · **PDF:** https://arxiv.org/pdf/2603.26611v1
+
+## 1419. Think over Trajectories: Leveraging Video Generation to Reconstruct GPS Trajectories from Cellular Signaling
+
+- **ID:** `2603.26610` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Ruixing Zhang, Hanzhang Jiang, Leilei Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.26610 · **PDF:** https://arxiv.org/pdf/2603.26610v1
+
+## 1420. Hardware-Aware Tensor Networks for Real-Time Quantum-Inspired Anomaly Detection at Particle Colliders
+
+- **ID:** `2603.26604` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Sagar Addepalli, Prajita Bhattarai, Abhilasha Dave, et al.
+- **abs:** https://arxiv.org/abs/2603.26604 · **PDF:** https://arxiv.org/pdf/2603.26604v1
+
+## 1421. Sustainability Is Not Linear: Quantifying Performance, Energy, and Privacy Trade-offs in On-Device Intelligence
+
+- **ID:** `2603.26603` · **primary:** `cs.SE` · **Submitted:** 2026-03-27
+- **Authors:** Eziyo Ehsani, Luca Giamattei, Ivano Malavolta, et al.
+- **abs:** https://arxiv.org/abs/2603.26603 · **PDF:** https://arxiv.org/pdf/2603.26603v1
+
+## 1422. Characterization and forecasting of national-scale solar power ramp events
+
+- **ID:** `2603.26596` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Luca Lanzilao, Angela Meyer
+- **abs:** https://arxiv.org/abs/2603.26596 · **PDF:** https://arxiv.org/pdf/2603.26596v1
+
+## 1423. PQuantML: A Tool for End-to-End Hardware-aware Model Compression
+
+- **ID:** `2603.26595` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Roope Niemi, Anastasiia Petrovych, Arghya Ranjan Das, et al.
+- **abs:** https://arxiv.org/abs/2603.26595 · **PDF:** https://arxiv.org/pdf/2603.26595v1
+
+## 1424. Evaluating Interactive 2D Visualization as a Sample Selection Strategy for Biomedical Time-Series Data Annotation
+
+- **ID:** `2603.26592` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Einari Vaaras, Manu Airaksinen, Okko Räsänen
+- **abs:** https://arxiv.org/abs/2603.26592 · **PDF:** https://arxiv.org/pdf/2603.26592v1
+
+## 1425. From Synthetic Data to Real Restorations: Diffusion Model for Patient-specific Dental Crown Completion
+
+- **ID:** `2603.26588` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Dávid Pukanec, Tibor Kubík, Michal Španěl
+- **abs:** https://arxiv.org/abs/2603.26588 · **PDF:** https://arxiv.org/pdf/2603.26588v1
+
+## 1426. EnTaCs: Analyzing the Relationship Between Sentiment and Language Choice in English-Tamil Code-Switching
+
+- **ID:** `2603.26587` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Paul Bontempo
+- **abs:** https://arxiv.org/abs/2603.26587 · **PDF:** https://arxiv.org/pdf/2603.26587v1
+
+## 1427. EZASP -- Facilitating the usage of ASP
+
+- **ID:** `2603.26863` · **primary:** `cs.SE` · **Submitted:** 2026-03-27
+- **Authors:** Rafael Martins, Matthias Knorr, Ricardo Gonçalves
+- **abs:** https://arxiv.org/abs/2603.26863 · **PDF:** https://arxiv.org/pdf/2603.26863v1
+
+## 1428. The Climber's Grip -- Personalized Deep Learning Models for Fear and Muscle Activity in Climbing
+
+- **ID:** `2603.26575` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Matthias Boeker, Dana Swarbrick, Ulysse T. A. Côté-Allard, et al.
+- **abs:** https://arxiv.org/abs/2603.26575 · **PDF:** https://arxiv.org/pdf/2603.26575v1
+
+## 1429. Generation Is Compression: Zero-Shot Video Coding via Stochastic Rectified Flow
+
+- **ID:** `2603.26571` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Ziyue Zeng, Xun Su, Haoyuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.26571 · **PDF:** https://arxiv.org/pdf/2603.26571v2
+
+## 1430. Machine Unlearning under Retain-Forget Entanglement
+
+- **ID:** `2603.26569` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Jingpu Cheng, Ping Liu, Qianxiao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.26569 · **PDF:** https://arxiv.org/pdf/2603.26569v1
+
+## 1431. Beyond Code Snippets: Benchmarking LLMs on Repository-Level Question Answering
+
+- **ID:** `2603.26567` · **primary:** `cs.SE` · **Submitted:** 2026-03-27
+- **Authors:** Yoseph Berhanu Alebachew, Hunter Leary, Swanand Vaishampayan, et al.
+- **abs:** https://arxiv.org/abs/2603.26567 · **PDF:** https://arxiv.org/pdf/2603.26567v1
+
+## 1432. MemBoost: A Memory-Boosted Framework for Cost-Aware LLM Inference
+
+- **ID:** `2603.26557` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Joris Köster, Zixuan Liu, Siavash Khajavi, et al.
+- **abs:** https://arxiv.org/abs/2603.26557 · **PDF:** https://arxiv.org/pdf/2603.26557v1
+
+## 1433. When Perplexity Lies: Generation-Focused Distillation of Hybrid Sequence Models
+
+- **ID:** `2603.26556` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Juan Gabriel Kostelec, Xiang Wang, Axel Laborieux, et al.
+- **abs:** https://arxiv.org/abs/2603.26556 · **PDF:** https://arxiv.org/pdf/2603.26556v1
+
+## 1434. Sharp Capacity Scaling of Spectral Optimizers in Learning Associative Memory
+
+- **ID:** `2603.26554` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Juno Kim, Eshaan Nichani, Denny Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.26554 · **PDF:** https://arxiv.org/pdf/2603.26554v1
+
+## 1435. Beyond MACs: Hardware Efficient Architecture Design for Vision Backbones
+
+- **ID:** `2603.26551` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Moritz Nottebaum, Matteo Dunnhofer, Christian Micheloni
+- **abs:** https://arxiv.org/abs/2603.26551 · **PDF:** https://arxiv.org/pdf/2603.26551v1
+
+## 1436. Not Search, But Scan: Benchmarking MLLMs on Scan-Oriented Academic Paper Reasoning
+
+- **ID:** `2603.28651` · **primary:** `cs.AI` · **Submitted:** 2026-03-27
+- **Authors:** Rongjin Li, Zichen Tang, Xianghe Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.28651 · **PDF:** https://arxiv.org/pdf/2603.28651v1
+
+## 1437. A Lyapunov Analysis of Softmax Policy Gradient for Stochastic Bandits
+
+- **ID:** `2603.26547` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Tor Lattimore
+- **abs:** https://arxiv.org/abs/2603.26547 · **PDF:** https://arxiv.org/pdf/2603.26547v1
+
+## 1438. Development of a European Union Time-Indexed Reference Dataset for Assessing the Performance of Signal Detection Methods in Pharmacovigilance using a Large Language Model
+
+- **ID:** `2603.26544` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Maria Kefala, Jeffery L. Painter, Syed Tauhid Bukhari, et al.
+- **abs:** https://arxiv.org/abs/2603.26544 · **PDF:** https://arxiv.org/pdf/2603.26544v1
+
+## 1439. The Multi-AMR Buffer Storage, Retrieval, and Reshuffling Problem: Exact and Heuristic Approaches
+
+- **ID:** `2603.26542` · **primary:** `cs.RO` · **Submitted:** 2026-03-27
+- **Authors:** Max Disselnmeyer, Thomas Bömer, Laura Dörr, et al.
+- **abs:** https://arxiv.org/abs/2603.26542 · **PDF:** https://arxiv.org/pdf/2603.26542v2
+
+## 1440. How Open Must Language Models be to Enable Reliable Scientific Inference?
+
+- **ID:** `2603.26539` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** James A. Michaelov, Catherine Arnett, Tyler A. Chang, et al.
+- **abs:** https://arxiv.org/abs/2603.26539 · **PDF:** https://arxiv.org/pdf/2603.26539v1
+
+## 1441. Stabilizing Rubric Integration Training via Decoupled Advantage Normalization
+
+- **ID:** `2603.26535` · **primary:** `cs.AI` · **Submitted:** 2026-03-27
+- **Authors:** Zelin Tan, Zhouliang Yu, Bohan Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.26535 · **PDF:** https://arxiv.org/pdf/2603.26535v2
+
+## 1442. The internal law of a material can be discovered from its boundary
+
+- **ID:** `2603.26517` · **primary:** `math.NA` · **Submitted:** 2026-03-27
+- **Authors:** Francesco Regazzoni
+- **abs:** https://arxiv.org/abs/2603.26517 · **PDF:** https://arxiv.org/pdf/2603.26517v1
+
+## 1443. ALBA: A European Portuguese Benchmark for Evaluating Language and Linguistic Dimensions in Generative LLMs
+
+- **ID:** `2603.26516` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Inês Vieira, Inês Calvo, Iago Paulo, et al.
+- **abs:** https://arxiv.org/abs/2603.26516 · **PDF:** https://arxiv.org/pdf/2603.26516v1
+
+## 1444. JAL-Turn: Joint Acoustic-Linguistic Modeling for Real-Time and Robust Turn-Taking Detection in Full-Duplex Spoken Dialogue Systems
+
+- **ID:** `2603.26515` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Guangzhao Yang, Yu Pan, Shi Qiu, et al.
+- **abs:** https://arxiv.org/abs/2603.26515 · **PDF:** https://arxiv.org/pdf/2603.26515v1
+
+## 1445. Beyond Textual Knowledge-Leveraging Multimodal Knowledge Bases for Enhancing Vision-and-Language Navigation
+
+- **ID:** `2603.26859` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Dongsheng Yang, Yinfeng Yu, Liejun Wang
+- **abs:** https://arxiv.org/abs/2603.26859 · **PDF:** https://arxiv.org/pdf/2603.26859v1
+
+## 1446. CADSmith: Multi-Agent CAD Generation with Programmatic Geometric Validation
+
+- **ID:** `2603.26512` · **primary:** `cs.AI` · **Submitted:** 2026-03-27
+- **Authors:** Jesse Barkley, Rumi Loghmani, Amir Barati Farimani
+- **abs:** https://arxiv.org/abs/2603.26512 · **PDF:** https://arxiv.org/pdf/2603.26512v1
+
+## 1447. AMALIA Technical Report: A Fully Open Source Large Language Model for European Portuguese
+
+- **ID:** `2603.26511` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Afonso Simplício, Gonçalo Vinagre, Miguel Moura Ramos, et al.
+- **abs:** https://arxiv.org/abs/2603.26511 · **PDF:** https://arxiv.org/pdf/2603.26511v1
+
+## 1448. Clinical named entity recognition in the Portuguese language: a benchmark of modern BERT models and LLMs
+
+- **ID:** `2603.26510` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Vinicius Anjos de Almeida, Sandro Saorin da Silva, Josimar Chire, et al.
+- **abs:** https://arxiv.org/abs/2603.26510 · **PDF:** https://arxiv.org/pdf/2603.26510v1
+
+## 1449. Identifying Connectivity Distributions from Neural Dynamics Using Flows
+
+- **ID:** `2603.26506` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-27
+- **Authors:** Timothy Doyeon Kim, Ulises Pereira-Obilinovic, Yiliu Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.26506 · **PDF:** https://arxiv.org/pdf/2603.26506v1
+
+## 1450. Targeted learning of heterogeneous treatment effect curves for right censored or left truncated time-to-event data
+
+- **ID:** `2603.26502` · **primary:** `stat.ME` · **Submitted:** 2026-03-27
+- **Authors:** Matthew Pryce, Karla Diaz-Ordaz, Ruth H. Keogh, et al.
+- **abs:** https://arxiv.org/abs/2603.26502 · **PDF:** https://arxiv.org/pdf/2603.26502v1
+
+## 1451. AIRA_2: Overcoming Bottlenecks in AI Research Agents
+
+- **ID:** `2603.26499` · **primary:** `cs.AI` · **Submitted:** 2026-03-27
+- **Authors:** Karen Hambardzumyan, Nicolas Baldwin, Edan Toledo, et al.
+- **abs:** https://arxiv.org/abs/2603.26499 · **PDF:** https://arxiv.org/pdf/2603.26499v1
+
+## 1452. Rocks, Pebbles and Sand: Modality-aware Scheduling for Multimodal Large Language Model Inference
+
+- **ID:** `2603.26498` · **primary:** `cs.DC` · **Submitted:** 2026-03-27
+- **Authors:** Konstantinos Papaioannou, Thaleia Dimitra Doudali
+- **abs:** https://arxiv.org/abs/2603.26498 · **PDF:** https://arxiv.org/pdf/2603.26498v1
+
+## 1453. Entanglement as Memory: Mechanistic Interpretability of Quantum Language Models
+
+- **ID:** `2603.26494` · **primary:** `quant-ph` · **Submitted:** 2026-03-27
+- **Authors:** Nathan Roll
+- **abs:** https://arxiv.org/abs/2603.26494 · **PDF:** https://arxiv.org/pdf/2603.26494v1
+
+## 1454. Conditional Neural Bayes Ratio Estimation for Experimental Design Optimisation
+
+- **ID:** `2603.26489` · **primary:** `astro-ph.IM` · **Submitted:** 2026-03-27
+- **Authors:** S. A. K. Leeney, T. Gessey-Jones, W. J. Handley, et al.
+- **abs:** https://arxiv.org/abs/2603.26489 · **PDF:** https://arxiv.org/pdf/2603.26489v1
+
+## 1455. EcoFair: Trustworthy and Energy-Aware Routing for Privacy-Preserving Vertically Partitioned Medical Inference
+
+- **ID:** `2603.26483` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Mostafa Anoosha, Dhavalkumar Thakker, Kuniko Paxton, et al.
+- **abs:** https://arxiv.org/abs/2603.26483 · **PDF:** https://arxiv.org/pdf/2603.26483v1
+
+## 1456. SPECTRA: An Efficient Spectral-Informed Neural Network for Sensor-Based Activity Recognition
+
+- **ID:** `2603.26482` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Deepika Gurung, Lala Shakti Swarup Ray, Mengxi Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.26482 · **PDF:** https://arxiv.org/pdf/2603.26482v1
+
+## 1457. Probabilistic Multilabel Graphical Modelling of Motif Transformations in Symbolic Music
+
+- **ID:** `2603.26478` · **primary:** `cs.SD` · **Submitted:** 2026-03-27
+- **Authors:** Ron Taieb, Yoel Greenberg, Barak Sober
+- **abs:** https://arxiv.org/abs/2603.26478 · **PDF:** https://arxiv.org/pdf/2603.26478v1
+
+## 1458. Shapley meets Rawls: an integrated framework for measuring and explaining unfairness
+
+- **ID:** `2603.26476` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Fadoua Amri-Jouidel, Emmanuel Kemel, Stéphane Mussard
+- **abs:** https://arxiv.org/abs/2603.26476 · **PDF:** https://arxiv.org/pdf/2603.26476v2
+
+## 1459. Foundation Model for Cardiac Time Series via Masked Latent Attention
+
+- **ID:** `2603.26475` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Moritz Vandenhirtz, Samuel Ruipérez-Campillo, Simon Böhi, et al.
+- **abs:** https://arxiv.org/abs/2603.26475 · **PDF:** https://arxiv.org/pdf/2603.26475v1
+
+## 1460. UNIFERENCE: A Discrete Event Simulation Framework for Developing Distributed AI Models
+
+- **ID:** `2603.26469` · **primary:** `cs.DC` · **Submitted:** 2026-03-27
+- **Authors:** Doğaç Eldenk, Stephen Xia
+- **abs:** https://arxiv.org/abs/2603.26469 · **PDF:** https://arxiv.org/pdf/2603.26469v1
+
+## 1461. A Boltzmann-machine-enhanced Transformer For DNA Sequence Classification
+
+- **ID:** `2603.26465` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Zhixuan Cao, Yishu Xu, Xuang WU
+- **abs:** https://arxiv.org/abs/2603.26465 · **PDF:** https://arxiv.org/pdf/2603.26465v1
+
+## 1462. Automatic feature identification in least-squares policy iteration using the Koopman operator framework
+
+- **ID:** `2603.26464` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Christian Mugisho Zagabe, Sebastian Peitz
+- **abs:** https://arxiv.org/abs/2603.26464 · **PDF:** https://arxiv.org/pdf/2603.26464v1
+
+## 1463. Neuro-Symbolic Process Anomaly Detection
+
+- **ID:** `2603.26461` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Devashish Gaikwad, Wil M. P. van der Aalst, Gyunam Park
+- **abs:** https://arxiv.org/abs/2603.26461 · **PDF:** https://arxiv.org/pdf/2603.26461v2
+
+## 1464. Can AI Models Direct Each Other? Organizational Structure as a Probe into Training Limitations
+
+- **ID:** `2603.26458` · **primary:** `cs.SE` · **Submitted:** 2026-03-27
+- **Authors:** Rui Liu
+- **abs:** https://arxiv.org/abs/2603.26458 · **PDF:** https://arxiv.org/pdf/2603.26458v1
+
+## 1465. Fair Data Pre-Processing with Imperfect Attribute Space
+
+- **ID:** `2603.26456` · **primary:** `cs.DB` · **Submitted:** 2026-03-27
+- **Authors:** Ying Zheng, Yangfan Jiang, Kian-Lee Tan
+- **abs:** https://arxiv.org/abs/2603.26456 · **PDF:** https://arxiv.org/pdf/2603.26456v1
+
+## 1466. ClimateCheck 2026: Scientific Fact-Checking and Disinformation Narrative Classification of Climate-related Claims
+
+- **ID:** `2603.26449` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Raia Abu Ahmad, Max Upravitelev, Aida Usmanova, et al.
+- **abs:** https://arxiv.org/abs/2603.26449 · **PDF:** https://arxiv.org/pdf/2603.26449v1
+
+## 1467. Meta-Learned Adaptive Optimization for Robust Human Mesh Recovery with Uncertainty-Aware Parameter Updates
+
+- **ID:** `2603.26447` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Shaurjya Mandal, Nutan Sharma, John Galeotti
+- **abs:** https://arxiv.org/abs/2603.26447 · **PDF:** https://arxiv.org/pdf/2603.26447v1
+
+## 1468. Interpretable long-term traffic modelling on national road networks using theory-informed deep learning
+
+- **ID:** `2603.26440` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Yue Li, Shujuan Chen, Akihiro Shimoda, et al.
+- **abs:** https://arxiv.org/abs/2603.26440 · **PDF:** https://arxiv.org/pdf/2603.26440v1
+
+## 1469. Automating Clinical Information Retrieval from Finnish Electronic Health Records Using Large Language Models
+
+- **ID:** `2603.26434` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Mikko Saukkoriipi, Nicole Hernandez, Jaakko Sahlsten, et al.
+- **abs:** https://arxiv.org/abs/2603.26434 · **PDF:** https://arxiv.org/pdf/2603.26434v1
+
+## 1470. Reconstructing Quantum Dot Charge Stability Diagrams with Diffusion Models
+
+- **ID:** `2603.26432` · **primary:** `quant-ph` · **Submitted:** 2026-03-27
+- **Authors:** Vinicius Hernandes, Joseph Rogers, Rouven Koch, et al.
+- **abs:** https://arxiv.org/abs/2603.26432 · **PDF:** https://arxiv.org/pdf/2603.26432v1
+
+## 1471. Analysing Calls to Order in German Parliamentary Debates
+
+- **ID:** `2603.26430` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Nina Smirnova, Daniel Dan, Philipp Mayr
+- **abs:** https://arxiv.org/abs/2603.26430 · **PDF:** https://arxiv.org/pdf/2603.26430v1
+
+## 1472. CPUBone: Efficient Vision Backbone Design for Devices with Low Parallelization Capabilities
+
+- **ID:** `2603.26425` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Moritz Nottebaum, Matteo Dunnhofer, Christian Micheloni
+- **abs:** https://arxiv.org/abs/2603.26425 · **PDF:** https://arxiv.org/pdf/2603.26425v2
+
+## 1473. A Hierarchical Sheaf Spectral Embedding Framework for Single-Cell RNA-seq Analysis
+
+- **ID:** `2603.26858` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Xiang Xiang Wang, Guo-Wei We
+- **abs:** https://arxiv.org/abs/2603.26858 · **PDF:** https://arxiv.org/pdf/2603.26858v1
+
+## 1474. Kantorovich--Kernel Neural Operators: Approximation Theory, Asymptotics, and Neural Network Interpretation
+
+- **ID:** `2603.26418` · **primary:** `stat.ML` · **Submitted:** 2026-03-27
+- **Authors:** Tian-Xiao He
+- **abs:** https://arxiv.org/abs/2603.26418 · **PDF:** https://arxiv.org/pdf/2603.26418v1
+
+## 1475. KMM-CP: Practical Conformal Prediction under Covariate Shift via Selective Kernel Mean Matching
+
+- **ID:** `2603.26415` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Siddhartha Laghuvarapu, Rohan Deb, Jimeng Sun
+- **abs:** https://arxiv.org/abs/2603.26415 · **PDF:** https://arxiv.org/pdf/2603.26415v1
+
+## 1476. Why Models Know But Don't Say: Chain-of-Thought Faithfulness Divergence Between Thinking Tokens and Answers in Open-Weight Reasoning Models
+
+- **ID:** `2603.26410` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Richard J. Young
+- **abs:** https://arxiv.org/abs/2603.26410 · **PDF:** https://arxiv.org/pdf/2603.26410v1
+
+## 1477. AFSS: Artifact-Focused Self-Synthesis for Mitigating Bias in Audio Deepfake Detection
+
+- **ID:** `2603.26856` · **primary:** `cs.SD` · **Submitted:** 2026-03-27
+- **Authors:** Hai-Son Nguyen-Le, Hung-Cuong Nguyen-Thanh, Nhien-An Le-Khac, et al.
+- **abs:** https://arxiv.org/abs/2603.26856 · **PDF:** https://arxiv.org/pdf/2603.26856v1
+
+## 1478. Word Alignment-Based Evaluation of Uniform Meaning Representations
+
+- **ID:** `2603.26401` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Daniel Zeman, Federica Gamba
+- **abs:** https://arxiv.org/abs/2603.26401 · **PDF:** https://arxiv.org/pdf/2603.26401v1
+
+## 1479. Maintaining Difficulty: A Margin Scheduler for Triplet Loss in Siamese Networks Training
+
+- **ID:** `2603.26389` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Roberto Sprengel Minozzo Tomchak, Oge Marques, Lucas Garcia Pedroso, et al.
+- **abs:** https://arxiv.org/abs/2603.26389 · **PDF:** https://arxiv.org/pdf/2603.26389v1
+
+## 1480. Switch Attention: Towards Dynamic and Fine-grained Hybrid Transformers
+
+- **ID:** `2603.26380` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Yusheng Zhao, Hourun Li, Bohan Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.26380 · **PDF:** https://arxiv.org/pdf/2603.26380v1
+
+## 1481. Generative Modeling in Protein Design: Neural Representations, Conditional Generation, and Evaluation Standards
+
+- **ID:** `2603.26378` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Senura Hansaja Wanasekara, Minh-Duong Nguyen, Xiaochen Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.26378 · **PDF:** https://arxiv.org/pdf/2603.26378v1
+
+## 1482. A Formal Framework for Uncertainty Analysis of Text Generation with Large Language Models
+
+- **ID:** `2603.26363` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Steffen Herbold, Florian Lemmerich
+- **abs:** https://arxiv.org/abs/2603.26363 · **PDF:** https://arxiv.org/pdf/2603.26363v1
+
+## 1483. Automated near-term quantum algorithm discovery for molecular ground states
+
+- **ID:** `2603.26359` · **primary:** `quant-ph` · **Submitted:** 2026-03-27
+- **Authors:** Fabian Finger, Frederic Rapp, Pranav Kalidindi, et al.
+- **abs:** https://arxiv.org/abs/2603.26359 · **PDF:** https://arxiv.org/pdf/2603.26359v1
+
+## 1484. DuSCN-FusionNet: An Interpretable Dual-Channel Structural Covariance Fusion Framework for ADHD Classification Using Structural MRI
+
+- **ID:** `2603.26351` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Qurat Ul Ain, Alptekin Temizel, Soyiba Jawed
+- **abs:** https://arxiv.org/abs/2603.26351 · **PDF:** https://arxiv.org/pdf/2603.26351v1
+
+## 1485. Generative Score Inference for Multimodal Data
+
+- **ID:** `2603.26349` · **primary:** `stat.ML` · **Submitted:** 2026-03-27
+- **Authors:** Xinyu Tian, Xiaotong Shen
+- **abs:** https://arxiv.org/abs/2603.26349 · **PDF:** https://arxiv.org/pdf/2603.26349v1
+
+## 1486. Reflect to Inform: Boosting Multimodal Reasoning via Information-Gain-Driven Verification
+
+- **ID:** `2603.26348` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Shuai Lv, Chang Liu, Feng Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.26348 · **PDF:** https://arxiv.org/pdf/2603.26348v1
+
+## 1487. A Power-Weighted Noncentral Complex Gaussian Distribution
+
+- **ID:** `2603.26344` · **primary:** `stat.ML` · **Submitted:** 2026-03-27
+- **Authors:** Toru Nakashika
+- **abs:** https://arxiv.org/abs/2603.26344 · **PDF:** https://arxiv.org/pdf/2603.26344v1
+
+## 1488. Curvature-aware Expected Free Energy as an Acquisition Function for Bayesian Optimization
+
+- **ID:** `2603.26339` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Ajith Anil Meera, Wouter Kouw
+- **abs:** https://arxiv.org/abs/2603.26339 · **PDF:** https://arxiv.org/pdf/2603.26339v1
+
+## 1489. CALRK-Bench: Evaluating Context-Aware Legal Reasoning in Korean Law
+
+- **ID:** `2603.26332` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** JiHyeok Jung, TaeYoung Yoon, HyunSouk Cho
+- **abs:** https://arxiv.org/abs/2603.26332 · **PDF:** https://arxiv.org/pdf/2603.26332v1
+
+## 1490. Stringological sequence prediction I: efficient algorithms for predicting highly repetitive sequences
+
+- **ID:** `2603.26852` · **primary:** `cs.FL` · **Submitted:** 2026-03-27
+- **Authors:** Vanessa Kosoy
+- **abs:** https://arxiv.org/abs/2603.26852 · **PDF:** https://arxiv.org/pdf/2603.26852v1
+
+## 1491. Mitigating the Reasoning Tax in Vision-Language Fine-Tuning with Input-Adaptive Depth Aggregation
+
+- **ID:** `2603.26330` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Yiming Ren, Yujiu Yang, Junjie Wang
+- **abs:** https://arxiv.org/abs/2603.26330 · **PDF:** https://arxiv.org/pdf/2603.26330v1
+
+## 1492. Making Multi-Axis Models Robust to Multiplicative Noise: How, and Why?
+
+- **ID:** `2603.26327` · **primary:** `stat.ME` · **Submitted:** 2026-03-27
+- **Authors:** Bailey Andrew, David R. Westhead, Luisa Cutillo
+- **abs:** https://arxiv.org/abs/2603.26327 · **PDF:** https://arxiv.org/pdf/2603.26327v1
+
+## 1493. PRISMA: Toward a Normative Information Infrastructure for Responsible Pharmaceutical Knowledge Management
+
+- **ID:** `2603.26324` · **primary:** `cs.DL` · **Submitted:** 2026-03-27
+- **Authors:** Eugenio Rodrigo Zimmer Neves, Amanda Vanon Correa, Camila Campioni, et al.
+- **abs:** https://arxiv.org/abs/2603.26324 · **PDF:** https://arxiv.org/pdf/2603.26324v1
+
+## 1494. From Human Cognition to Neural Activations: Probing the Computational Primitives of Spatial Reasoning in LLMs
+
+- **ID:** `2603.26323` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Jiyuan An, Liner Yang, Mengyan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.26323 · **PDF:** https://arxiv.org/pdf/2603.26323v1
+
+## 1495. STN-GPR: A Singularity Tensor Network Framework for Efficient Option Pricing
+
+- **ID:** `2603.26318` · **primary:** `q-fin.PR` · **Submitted:** 2026-03-27
+- **Authors:** Dominic Gribben, Carolina Allende, Alba Villarino, et al.
+- **abs:** https://arxiv.org/abs/2603.26318 · **PDF:** https://arxiv.org/pdf/2603.26318v1
+
+## 1496. Label-Free Cross-Task LoRA Merging with Null-Space Compression
+
+- **ID:** `2603.26317` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Wonyoung Lee, Wooseong Jeong, Kuk-Jin Yoon
+- **abs:** https://arxiv.org/abs/2603.26317 · **PDF:** https://arxiv.org/pdf/2603.26317v1
+
+## 1497. SALMUBench: A Benchmark for Sensitive Association-Level Multimodal Unlearning
+
+- **ID:** `2603.26316` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Cai Selvas-Sala, Lei Kang, Lluis Gomez
+- **abs:** https://arxiv.org/abs/2603.26316 · **PDF:** https://arxiv.org/pdf/2603.26316v1
+
+## 1498. Semi-structured multi-state delinquency model for mortgage default
+
+- **ID:** `2603.26309` · **primary:** `stat.AP` · **Submitted:** 2026-03-27
+- **Authors:** Victor Medina-Olivares, Wangzhen Xia, Stefan Lessmann, et al.
+- **abs:** https://arxiv.org/abs/2603.26309 · **PDF:** https://arxiv.org/pdf/2603.26309v1
+
+## 1499. D-GATNet: Interpretable Temporal Graph Attention Learning for ADHD Identification Using Dynamic Functional Connectivity
+
+- **ID:** `2603.26308` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Qurat Ul Ain, Alptekin Temizel, Soyiba Jawed
+- **abs:** https://arxiv.org/abs/2603.26308 · **PDF:** https://arxiv.org/pdf/2603.26308v1
+
+## 1500. Complete Causal Identification from Ancestral Graphs under Selection Bias
+
+- **ID:** `2603.26301` · **primary:** `stat.ME` · **Submitted:** 2026-03-27
+- **Authors:** Leihao Chen, Joris M. Mooij
+- **abs:** https://arxiv.org/abs/2603.26301 · **PDF:** https://arxiv.org/pdf/2603.26301v1
+
+## 1501. Preference-Aligned LoRA Merging: Preserving Subspace Coverage and Addressing Directional Anisotropy
+
+- **ID:** `2603.26299` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Wooseong Jeong, Wonyoung Lee, Kuk-Jin Yoon
+- **abs:** https://arxiv.org/abs/2603.26299 · **PDF:** https://arxiv.org/pdf/2603.26299v1
+
+## 1502. findsylls: A Language-Agnostic Toolkit for Syllable-Level Speech Tokenization and Embedding
+
+- **ID:** `2603.26292` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Héctor Javier Vázquez Martínez
+- **abs:** https://arxiv.org/abs/2603.26292 · **PDF:** https://arxiv.org/pdf/2603.26292v1
+
+## 1503. PhysVid: Physics Aware Local Conditioning for Generative Video Models
+
+- **ID:** `2603.26285` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Saurabh Pathak, Elahe Arani, Mykola Pechenizkiy, et al.
+- **abs:** https://arxiv.org/abs/2603.26285 · **PDF:** https://arxiv.org/pdf/2603.26285v2
+
+## 1504. Knowdit: Agentic Smart Contract Vulnerability Detection with Auditing Knowledge Summarization
+
+- **ID:** `2603.26270` · **primary:** `cs.CR` · **Submitted:** 2026-03-27
+- **Authors:** Ziqiao Kong, Wanxu Xia, Chong Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.26270 · **PDF:** https://arxiv.org/pdf/2603.26270v1
+
+## 1505. GUIDE: Resolving Domain Bias in GUI Agents through Real-Time Web Video Retrieval and Plug-and-Play Annotation
+
+- **ID:** `2603.26266` · **primary:** `cs.AI` · **Submitted:** 2026-03-27
+- **Authors:** Rui Xie, Zhi Gao, Chenrui Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.26266 · **PDF:** https://arxiv.org/pdf/2603.26266v2
+
+## 1506. Topology-Aware Graph Reinforcement Learning for Energy Storage Systems Optimal Dispatch in Distribution Networks
+
+- **ID:** `2603.26264` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Shuyi Gao, Stavros Orfanoudakis, Shengren Hou, et al.
+- **abs:** https://arxiv.org/abs/2603.26264 · **PDF:** https://arxiv.org/pdf/2603.26264v1
+
+## 1507. GLASS: Geometry-aware Local Alignment and Structure Synchronization Network for 2D-3D Registration
+
+- **ID:** `2603.26262` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Zhixin Cheng, Jiacheng Deng, Xinjun Li, et al.
+- **abs:** https://arxiv.org/abs/2603.26262 · **PDF:** https://arxiv.org/pdf/2603.26262v1
+
+## 1508. Contrastive Conformal Sets
+
+- **ID:** `2603.26261` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Yahya Alkhatib, Wee Peng Tay
+- **abs:** https://arxiv.org/abs/2603.26261 · **PDF:** https://arxiv.org/pdf/2603.26261v1
+
+## 1509. GeoGuide: Hierarchical Geometric Guidance for Open-Vocabulary 3D Semantic Segmentation
+
+- **ID:** `2603.26260` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Xujing Tao, Chuxin Wang, Yubo Ai, et al.
+- **abs:** https://arxiv.org/abs/2603.26260 · **PDF:** https://arxiv.org/pdf/2603.26260v1
+
+## 1510. Working Notes on Late Interaction Dynamics: Analyzing Targeted Behaviors of Late Interaction Models
+
+- **ID:** `2603.26259` · **primary:** `cs.IR` · **Submitted:** 2026-03-27
+- **Authors:** Antoine Edy, Max Conti, Quentin Macé
+- **abs:** https://arxiv.org/abs/2603.26259 · **PDF:** https://arxiv.org/pdf/2603.26259v1
+
+## 1511. ARTA: Adaptive Mixed-Resolution Token Allocation for Efficient Dense Feature Extraction
+
+- **ID:** `2603.26258` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** David Hagerman, Roman Naeem, Erik Brorsson, et al.
+- **abs:** https://arxiv.org/abs/2603.26258 · **PDF:** https://arxiv.org/pdf/2603.26258v1
+
+## 1512. Improving Risk Stratification in Hypertrophic Cardiomyopathy: A Novel Score Combining Echocardiography, Clinical, and Medication Data
+
+- **ID:** `2603.26254` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Marion Taconné, Valentina D. A. Corino, Annamaria Del Franco, et al.
+- **abs:** https://arxiv.org/abs/2603.26254 · **PDF:** https://arxiv.org/pdf/2603.26254v1
+
+## 1513. SocialX: A Modular Platform for Multi-Source Big Data Research in Indonesia
+
+- **ID:** `2603.26253` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Muhammad Apriandito Arya Saputra, Andry Alamsyah, Dian Puteri Ramadhani, et al.
+- **abs:** https://arxiv.org/abs/2603.26253 · **PDF:** https://arxiv.org/pdf/2603.26253v1
+
+## 1514. Channelling, Coordinating, Collaborating: A Three-Layer Framework for Disability-Centered Human-Agent Collaboration
+
+- **ID:** `2603.26252` · **primary:** `cs.HC` · **Submitted:** 2026-03-27
+- **Authors:** Lan Xiao, Catherine Holloway
+- **abs:** https://arxiv.org/abs/2603.26252 · **PDF:** https://arxiv.org/pdf/2603.26252v1
+
+## 1515. Knowledge Distillation for Efficient Transformer-Based Reinforcement Learning in Hardware-Constrained Energy Management Systems
+
+- **ID:** `2603.26249` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Pascal Henrich, Jonas Sievers, Maximilian Beichter, et al.
+- **abs:** https://arxiv.org/abs/2603.26249 · **PDF:** https://arxiv.org/pdf/2603.26249v1
+
+## 1516. Automatic Speech Recognition for Documenting Endangered Languages: Case Study of Ikema Miyakoan
+
+- **ID:** `2603.26248` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Chihiro Taguchi, Yukinori Takubo, David Chiang
+- **abs:** https://arxiv.org/abs/2603.26248 · **PDF:** https://arxiv.org/pdf/2603.26248v1
+
+## 1517. Distilling Conversations: Abstract Compression of Conversational Audio Context for LLM-based ASR
+
+- **ID:** `2603.26246` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Shashi Kumar, Esaú Villatoro-Tello, Sergio Burdisso, et al.
+- **abs:** https://arxiv.org/abs/2603.26246 · **PDF:** https://arxiv.org/pdf/2603.26246v1
+
+## 1518. Physics-Informed Neural Networks and Sequence Encoder: Application to heating and early cooling of thermo-stamping process
+
+- **ID:** `2603.26245` · **primary:** `cs.CE` · **Submitted:** 2026-03-27
+- **Authors:** Mouad Elaarabi, Domenico Borzacchiello, Philippe Le Bot, et al.
+- **abs:** https://arxiv.org/abs/2603.26245 · **PDF:** https://arxiv.org/pdf/2603.26245v1
+
+## 1519. Automating Domain-Driven Design: Experience with a Prompting Framework
+
+- **ID:** `2603.26244` · **primary:** `cs.SE` · **Submitted:** 2026-03-27
+- **Authors:** Tobias Eisenreich, Husein Jusic, Stefan Wagner
+- **abs:** https://arxiv.org/abs/2603.26244 · **PDF:** https://arxiv.org/pdf/2603.26244v1
+
+## 1520. SwarmCoDe: A Scalable Co-Design Framework for Heterogeneous Robot Swarms via Dynamic Speciation
+
+- **ID:** `2603.26240` · **primary:** `cs.RO` · **Submitted:** 2026-03-27
+- **Authors:** Andrew Wilhelm, Josie Hughes
+- **abs:** https://arxiv.org/abs/2603.26240 · **PDF:** https://arxiv.org/pdf/2603.26240v1
+
+## 1521. A Universal Vibe? Finding and Controlling Language-Agnostic Informal Register with SAEs
+
+- **ID:** `2603.26236` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Uri Z. Kialy, Avi Shtarkberg, Ayal Klein
+- **abs:** https://arxiv.org/abs/2603.26236 · **PDF:** https://arxiv.org/pdf/2603.26236v1
+
+## 1522. GS-BrainText: A Multi-Site Brain Imaging Report Dataset from Generation Scotland for Clinical Natural Language Processing Development and Validation
+
+- **ID:** `2603.26235` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Beatrice Alex, Claire Grover, Arlene Casey, et al.
+- **abs:** https://arxiv.org/abs/2603.26235 · **PDF:** https://arxiv.org/pdf/2603.26235v1
+
+## 1523. Ask or Assume? Uncertainty-Aware Clarification-Seeking in Coding Agents
+
+- **ID:** `2603.26233` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Nicholas Edwards, Sebastian Schuster
+- **abs:** https://arxiv.org/abs/2603.26233 · **PDF:** https://arxiv.org/pdf/2603.26233v1
+
+## 1524. Optimization Trade-offs in Asynchronous Federated Learning: A Stochastic Networks Approach
+
+- **ID:** `2603.26231` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Abdelkrim Alahyane, Céline Comte, Matthieu Jonckheere
+- **abs:** https://arxiv.org/abs/2603.26231 · **PDF:** https://arxiv.org/pdf/2603.26231v1
+
+## 1525. Privacy-Accuracy Trade-offs in High-Dimensional LASSO under Perturbation Mechanisms
+
+- **ID:** `2603.26227` · **primary:** `stat.ML` · **Submitted:** 2026-03-27
+- **Authors:** Ayaka Sakata, Haruka Tanzawa
+- **abs:** https://arxiv.org/abs/2603.26227 · **PDF:** https://arxiv.org/pdf/2603.26227v1
+
+## 1526. Stable Reasoning, Unstable Responses: Mitigating LLM Deception via Stability Asymmetry
+
+- **ID:** `2603.26846` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Guoxi Zhang, Jiawei Chen, Tianzhuo Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.26846 · **PDF:** https://arxiv.org/pdf/2603.26846v1
+
+## 1527. GISclaw: An Open-Source LLM-Powered Agent System for Full-Stack Geospatial Analysis
+
+- **ID:** `2603.26845` · **primary:** `cs.SE` · **Submitted:** 2026-03-27
+- **Authors:** Jinzhen Han, JinByeong Lee, Yuri Shim, et al.
+- **abs:** https://arxiv.org/abs/2603.26845 · **PDF:** https://arxiv.org/pdf/2603.26845v1
+
+## 1528. Clawed and Dangerous: Can We Trust Open Agentic Systems?
+
+- **ID:** `2603.26221` · **primary:** `cs.CR` · **Submitted:** 2026-03-27
+- **Authors:** Shiping Chen, Qin Wang, Guangsheng Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.26221 · **PDF:** https://arxiv.org/pdf/2603.26221v1
+
+## 1529. Uncertainty-Aware Mapping from 3D Keypoints to Anatomical Landmarks for Markerless Biomechanics
+
+- **ID:** `2603.26844` · **primary:** `eess.IV` · **Submitted:** 2026-03-27
+- **Authors:** Cesare Davide Pace, Alessandro Marco De Nunzio, Claudio De Stefano, et al.
+- **abs:** https://arxiv.org/abs/2603.26844 · **PDF:** https://arxiv.org/pdf/2603.26844v1
+
+## 1530. On associative neural networks for sparse patterns with huge capacities
+
+- **ID:** `2603.26217` · **primary:** `math.PR` · **Submitted:** 2026-03-27
+- **Authors:** Matthias Löwe, Franck Vermet
+- **abs:** https://arxiv.org/abs/2603.26217 · **PDF:** https://arxiv.org/pdf/2603.26217v1
+
+## 1531. Towards GUI Agents: Vision-Language Diffusion Models for GUI Grounding
+
+- **ID:** `2603.26211` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Shrinidhi Kumbhar, Haofu Liao, Srikar Appalaraju, et al.
+- **abs:** https://arxiv.org/abs/2603.26211 · **PDF:** https://arxiv.org/pdf/2603.26211v1
+
+## 1532. Sparse Auto-Encoders and Holism about Large Language Models
+
+- **ID:** `2603.26207` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Jumbly Grindrod
+- **abs:** https://arxiv.org/abs/2603.26207 · **PDF:** https://arxiv.org/pdf/2603.26207v1
+
+## 1533. An Object Web Seminar: A Retrospective on a Technical Dialogue Still Reverberating
+
+- **ID:** `2603.26203` · **primary:** `cs.SE` · **Submitted:** 2026-03-27
+- **Authors:** James J. Cusick
+- **abs:** https://arxiv.org/abs/2603.26203 · **PDF:** https://arxiv.org/pdf/2603.26203v2
+
+## 1534. MemCam: Memory-Augmented Camera Control for Consistent Video Generation
+
+- **ID:** `2603.26193` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Xinhang Gao, Junlin Guan, Shuhan Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.26193 · **PDF:** https://arxiv.org/pdf/2603.26193v1
+
+## 1535. Dual-Stage Invariant Continual Learning under Extreme Visual Sparsity
+
+- **ID:** `2603.26190` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Rangya Zhang, Jiaping Xiao, Lu Bai, et al.
+- **abs:** https://arxiv.org/abs/2603.26190 · **PDF:** https://arxiv.org/pdf/2603.26190v1
+
+## 1536. Progressive Learning with Anatomical Priors for Reliable Left Atrial Scar Segmentation from Late Gadolinium Enhancement MRI
+
+- **ID:** `2603.26186` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Jing Zhang, Bastien Bergere, Emilie Bollache, et al.
+- **abs:** https://arxiv.org/abs/2603.26186 · **PDF:** https://arxiv.org/pdf/2603.26186v1
+
+## 1537. ClinicalAgents: Multi-Agent Orchestration for Clinical Decision Making with Dual-Memory
+
+- **ID:** `2603.26182` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Zhuohan Ge, Haoyang Li, Yubo Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.26182 · **PDF:** https://arxiv.org/pdf/2603.26182v1
+
+## 1538. Geometric Evolution Graph Convolutional Networks: Enhancing Graph Representation Learning via Ricci Flow
+
+- **ID:** `2603.26178` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Jicheng Ma, Yunyan Yang, Juan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.26178 · **PDF:** https://arxiv.org/pdf/2603.26178v1
+
+## 1539. VAN-AD: Visual Masked Autoencoder with Normalizing Flow For Time Series Anomaly Detection
+
+- **ID:** `2603.26842` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** PengYu Chen, Shang Wan, Xiaohou Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.26842 · **PDF:** https://arxiv.org/pdf/2603.26842v1
+
+## 1540. FatigueFormer: Static-Temporal Feature Fusion for Robust sEMG-Based Muscle Fatigue Recognition
+
+- **ID:** `2603.26841` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Tong Zhang, Hong Guo, Shuangzhou Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.26841 · **PDF:** https://arxiv.org/pdf/2603.26841v1
+
+## 1541. Can AI Scientist Agents Learn from Lab-in-the-Loop Feedback? Evidence from Iterative Perturbation Discovery
+
+- **ID:** `2603.26177` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Gilles Wainrib, Barbara Bodinier, Haitem Dakhli, et al.
+- **abs:** https://arxiv.org/abs/2603.26177 · **PDF:** https://arxiv.org/pdf/2603.26177v1
+
+## 1542. DataFlex: A Unified Framework for Data-Centric Dynamic Training of Large Language Models
+
+- **ID:** `2603.26164` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Hao Liang, Zhengyang Zhao, Meiyi Qiang, et al.
+- **abs:** https://arxiv.org/abs/2603.26164 · **PDF:** https://arxiv.org/pdf/2603.26164v1
+
+## 1543. Dual-branch Graph Domain Adaptation for Cross-scenario Multi-modal Emotion Recognition
+
+- **ID:** `2603.26840` · **primary:** `eess.AS` · **Submitted:** 2026-03-27
+- **Authors:** Yuntao Shou, Jun Zhou, Tao Meng, et al.
+- **abs:** https://arxiv.org/abs/2603.26840 · **PDF:** https://arxiv.org/pdf/2603.26840v1
+
+## 1544. Clash of the models: Comparing performance of BERT-based variants for generic news frame detection
+
+- **ID:** `2603.26156` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Vihang Jumle
+- **abs:** https://arxiv.org/abs/2603.26156 · **PDF:** https://arxiv.org/pdf/2603.26156v1
+
+## 1545. From Pixels to BFS: High Maze Accuracy Does Not Imply Visual Planning
+
+- **ID:** `2603.26839` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Alberto G. Rodriguez Salgado
+- **abs:** https://arxiv.org/abs/2603.26839 · **PDF:** https://arxiv.org/pdf/2603.26839v1
+
+## 1546. Concerning Uncertainty -- A Systematic Survey of Uncertainty-Aware XAI
+
+- **ID:** `2603.26838` · **primary:** `cs.AI` · **Submitted:** 2026-03-27
+- **Authors:** Helena Löfström, Tuwe Löfström, Anders Hjort, et al.
+- **abs:** https://arxiv.org/abs/2603.26838 · **PDF:** https://arxiv.org/pdf/2603.26838v1
+
+## 1547. SpatialAnt: Autonomous Zero-Shot Robot Navigation via Active Scene Reconstruction and Visual Anticipation
+
+- **ID:** `2603.26837` · **primary:** `cs.RO` · **Submitted:** 2026-03-27
+- **Authors:** Jiwen Zhang, Xiangyu Shi, Siyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.26837 · **PDF:** https://arxiv.org/pdf/2603.26837v1
+
+## 1548. On the Complexity of Optimal Graph Rewiring for Oversmoothing and Oversquashing in Graph Neural Networks
+
+- **ID:** `2603.26140` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Mostafa Haghir Chehreghani
+- **abs:** https://arxiv.org/abs/2603.26140 · **PDF:** https://arxiv.org/pdf/2603.26140v1
+
+## 1549. PruneFuse: Efficient Data Selection via Weight Pruning and Network Fusion
+
+- **ID:** `2603.26138` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Humaira Kousar, Hasnain Irshad Bhatti, Jaekyun Moon
+- **abs:** https://arxiv.org/abs/2603.26138 · **PDF:** https://arxiv.org/pdf/2603.26138v1
+
+## 1550. ATime-Consistent Benchmark for Repository-Level Software Engineering Evaluation
+
+- **ID:** `2603.26137` · **primary:** `cs.SE` · **Submitted:** 2026-03-27
+- **Authors:** Xianpeng, Sun, Haonan Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.26137 · **PDF:** https://arxiv.org/pdf/2603.26137v1
+
+## 1551. PEANUT: Perturbations by Eigenvector Alignment for Attacking Graph Neural Networks Under Topology-Driven Message Passing
+
+- **ID:** `2603.26136` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Bhavya Kohli, Biplab Sikdar
+- **abs:** https://arxiv.org/abs/2603.26136 · **PDF:** https://arxiv.org/pdf/2603.26136v2
+
+## 1552. TinyML for Acoustic Anomaly Detection in IoT Sensor Networks
+
+- **ID:** `2603.26135` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Amar Almaini, Jakob Folz, Ghadeer Ashour
+- **abs:** https://arxiv.org/abs/2603.26135 · **PDF:** https://arxiv.org/pdf/2603.26135v1
+
+## 1553. SWE-PRBench: Benchmarking AI Code Review Quality Against Pull Request Feedback
+
+- **ID:** `2603.26130` · **primary:** `cs.SE` · **Submitted:** 2026-03-27
+- **Authors:** Deepak Kumar
+- **abs:** https://arxiv.org/abs/2603.26130 · **PDF:** https://arxiv.org/pdf/2603.26130v1
+
+## 1554. Finding Distributed Object-Centric Properties in Self-Supervised Transformers
+
+- **ID:** `2603.26127` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Samyak Rawlekar, Amitabh Swain, Yujun Cai, et al.
+- **abs:** https://arxiv.org/abs/2603.26127 · **PDF:** https://arxiv.org/pdf/2603.26127v1
+
+## 1555. SkinGPT-X: A Self-Evolving Collaborative Multi-Agent System for Transparent and Trustworthy Dermatological Diagnosis
+
+- **ID:** `2603.26122` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Zhangtianyi Chen, Yuhao Shen, Florensia Widjaja, et al.
+- **abs:** https://arxiv.org/abs/2603.26122 · **PDF:** https://arxiv.org/pdf/2603.26122v1
+
+## 1556. DPD-Cancer: Explainable Graph-based Deep Learning for Small Molecule Anti-Cancer Activity Prediction
+
+- **ID:** `2603.26114` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Magnus H. Strømme, Alex G. C. de Sá, David B. Ascher
+- **abs:** https://arxiv.org/abs/2603.26114 · **PDF:** https://arxiv.org/pdf/2603.26114v1
+
+## 1557. Accurate Precipitation Forecast by Efficiently Learning from Massive Atmospheric Variables and Unbalanced Distribution
+
+- **ID:** `2603.26108` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Shuangliang Li, Siwei Li, Li Li, et al.
+- **abs:** https://arxiv.org/abs/2603.26108 · **PDF:** https://arxiv.org/pdf/2603.26108v1
+
+## 1558. LLM Benchmark-User Need Misalignment for Climate Change
+
+- **ID:** `2603.26106` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Oucheng Liu, Lexing Xie, Jing Jiang
+- **abs:** https://arxiv.org/abs/2603.26106 · **PDF:** https://arxiv.org/pdf/2603.26106v1
+
+## 1559. Are LLM-Enhanced Graph Neural Networks Robust against Poisoning Attacks?
+
+- **ID:** `2603.26105` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Yuhang Ma, Jie Wang, Zheng Yan
+- **abs:** https://arxiv.org/abs/2603.26105 · **PDF:** https://arxiv.org/pdf/2603.26105v1
+
+## 1560. "Oops! ChatGPT is Temporarily Unavailable!": A Diary Study on Knowledge Workers' Experiences of LLM Withdrawal
+
+- **ID:** `2603.26099` · **primary:** `cs.HC` · **Submitted:** 2026-03-27
+- **Authors:** Eunseo Oh, Suyoun Lee, Jae Young Choi, et al.
+- **abs:** https://arxiv.org/abs/2603.26099 · **PDF:** https://arxiv.org/pdf/2603.26099v1
+
+## 1561. A Human-Inspired Decoupled Architecture for Efficient Audio Representation Learning
+
+- **ID:** `2603.26098` · **primary:** `cs.SD` · **Submitted:** 2026-03-27
+- **Authors:** Harunori Kawano, Takeshi Sasaki
+- **abs:** https://arxiv.org/abs/2603.26098 · **PDF:** https://arxiv.org/pdf/2603.26098v1
+
+## 1562. Dynamic Tokenization via Reinforcement Patching: End-to-end Training and Zero-shot Transfer
+
+- **ID:** `2603.26097` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Yulun Wu, Sravan Kumar Ankireddy, Samuel Sharpe, et al.
+- **abs:** https://arxiv.org/abs/2603.26097 · **PDF:** https://arxiv.org/pdf/2603.26097v1
+
+## 1563. AcTTA: Rethinking Test-Time Adaptation via Dynamic Activation
+
+- **ID:** `2603.26096` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Hyeongyu Kim, Geonhui Han, Dosik Hwang
+- **abs:** https://arxiv.org/abs/2603.26096 · **PDF:** https://arxiv.org/pdf/2603.26096v1
+
+## 1564. IndoBERT-Relevancy: A Context-Conditioned Relevancy Classifier for Indonesian Text
+
+- **ID:** `2603.26095` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Muhammad Apriandito Arya Saputra, Andry Alamsyah, Dian Puteri Ramadhani, et al.
+- **abs:** https://arxiv.org/abs/2603.26095 · **PDF:** https://arxiv.org/pdf/2603.26095v1
+
+## 1565. CD-Buffer: Complementary Dual-Buffer Framework for Test-Time Adaptation in Adverse Weather Object Detection
+
+- **ID:** `2603.26092` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Youngjun Song, Hyeongyu Kim, Dosik Hwang
+- **abs:** https://arxiv.org/abs/2603.26092 · **PDF:** https://arxiv.org/pdf/2603.26092v1
+
+## 1566. Selective Deficits in LLM Mental Self-Modeling in a Behavior-Based Test of Theory of Mind
+
+- **ID:** `2603.26089` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Christopher Ackerman
+- **abs:** https://arxiv.org/abs/2603.26089 · **PDF:** https://arxiv.org/pdf/2603.26089v1
+
+## 1567. Hybrid Diffusion Model for Breast Ultrasound Image Augmentation
+
+- **ID:** `2603.26834` · **primary:** `eess.IV` · **Submitted:** 2026-03-27
+- **Authors:** Farhan Fuad Abir, Sanjeda Sara Jennifer, Niloofar Yousefi, et al.
+- **abs:** https://arxiv.org/abs/2603.26834 · **PDF:** https://arxiv.org/pdf/2603.26834v1
+
+## 1568. When Identities Collapse: A Stress-Test Benchmark for Multi-Subject Personalization
+
+- **ID:** `2603.26078` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Zhihan Chen, Yuhuan Zhao, Yijie Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.26078 · **PDF:** https://arxiv.org/pdf/2603.26078v1
+
+## 1569. Semi-Automated Knowledge Engineering and Process Mapping for Total Airport Management
+
+- **ID:** `2603.26076` · **primary:** `cs.AI` · **Submitted:** 2026-03-27
+- **Authors:** Darryl Teo, Adharsha Sam, Chuan Shen Marcus Koh, et al.
+- **abs:** https://arxiv.org/abs/2603.26076 · **PDF:** https://arxiv.org/pdf/2603.26076v1
+
+## 1570. Envisioning global urban development with satellite imagery and generative AI
+
+- **ID:** `2603.26831` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Kailai Sun, Yuebing Liang, Mingyi He, et al.
+- **abs:** https://arxiv.org/abs/2603.26831 · **PDF:** https://arxiv.org/pdf/2603.26831v1
+
+## 1571. MUST: Modality-Specific Representation-Aware Transformer for Diffusion-Enhanced Survival Prediction with Missing Modality
+
+- **ID:** `2603.26071` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Kyungwon Kim, Dosik Hwang
+- **abs:** https://arxiv.org/abs/2603.26071 · **PDF:** https://arxiv.org/pdf/2603.26071v1
+
+## 1572. R-PGA: Robust Physical Adversarial Camouflage Generation via Relightable 3D Gaussian Splatting
+
+- **ID:** `2603.26067` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Tianrui Lou, Siyuan Liang, Jiawei Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.26067 · **PDF:** https://arxiv.org/pdf/2603.26067v1
+
+## 1573. Adversarial Bandit Optimization with Globally Bounded Perturbations to Linear Losses
+
+- **ID:** `2603.26066` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Zhuoyu Cheng, Kohei Hatano, Eiji Takimoto
+- **abs:** https://arxiv.org/abs/2603.26066 · **PDF:** https://arxiv.org/pdf/2603.26066v1
+
+## 1574. A Regression Framework for Understanding Prompt Component Impact on LLM Performance
+
+- **ID:** `2603.26830` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Andrew Lauziere, Jonathan Daugherty, Taisa Kushner
+- **abs:** https://arxiv.org/abs/2603.26830 · **PDF:** https://arxiv.org/pdf/2603.26830v1
+
+## 1575. MuDD: A Multimodal Deception Detection Dataset and GSR-Guided Progressive Distillation for Non-Contact Deception Detection
+
+- **ID:** `2603.26064` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Peiyuan Jiang, Yao Liu, Yanglei Gan, et al.
+- **abs:** https://arxiv.org/abs/2603.26064 · **PDF:** https://arxiv.org/pdf/2603.26064v1
+
+## 1576. I Want to Believe (but the Vocabulary Changed): Measuring the Semantic Structure and Evolution of Conspiracy Theories
+
+- **ID:** `2603.26062` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Manisha Keim, Sarmad Chandio, Osama Khalid, et al.
+- **abs:** https://arxiv.org/abs/2603.26062 · **PDF:** https://arxiv.org/pdf/2603.26062v1
+
+## 1577. Squish and Release: Exposing Hidden Hallucinations by Making Them Surface as Safety Signals
+
+- **ID:** `2603.26829` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Nathaniel Oh, Paul Attie
+- **abs:** https://arxiv.org/abs/2603.26829 · **PDF:** https://arxiv.org/pdf/2603.26829v1
+
+## 1578. Arithmetic OOD Failure Unfolds in Stages in Minimal GPTs
+
+- **ID:** `2603.26828` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Seine A. Shintani
+- **abs:** https://arxiv.org/abs/2603.26828 · **PDF:** https://arxiv.org/pdf/2603.26828v1
+
+## 1579. Bridging Pixels and Words: Mask-Aware Local Semantic Fusion for Multimodal Media Verification
+
+- **ID:** `2603.26052` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Zizhao Chen, Ping Wei, Ziyang Ren, et al.
+- **abs:** https://arxiv.org/abs/2603.26052 · **PDF:** https://arxiv.org/pdf/2603.26052v1
+
+## 1580. Seeing Like Radiologists: Context- and Gaze-Guided Vision-Language Pretraining for Chest X-rays
+
+- **ID:** `2603.26049` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Kang Liu, Zhuoqi Ma, Siyu Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.26049 · **PDF:** https://arxiv.org/pdf/2603.26049v1
+
+## 1581. Asymptotic Optimism for Tensor Regression Models with Applications to Neural Network Compression
+
+- **ID:** `2603.26048` · **primary:** `stat.ML` · **Submitted:** 2026-03-27
+- **Authors:** Haoming Shi, Eric C. Chi, Hengrui Luo
+- **abs:** https://arxiv.org/abs/2603.26048 · **PDF:** https://arxiv.org/pdf/2603.26048v1
+
+## 1582. Retrieval-Augmented Generation Based Nurse Observation Extraction
+
+- **ID:** `2603.26046` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Kyomin Hwang, Nojun Kwak
+- **abs:** https://arxiv.org/abs/2603.26046 · **PDF:** https://arxiv.org/pdf/2603.26046v1
+
+## 1583. H-Node Attack and Defense in Large Language Models
+
+- **ID:** `2603.26045` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Eric Yocam, Varghese Vaidyan, Yong Wang
+- **abs:** https://arxiv.org/abs/2603.26045 · **PDF:** https://arxiv.org/pdf/2603.26045v1
+
+## 1584. AgentCollab: A Self-Evaluation-Driven Collaboration Paradigm for Efficient LLM Agents
+
+- **ID:** `2603.26034` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Wenbo Gao, Renxi Liu, Xian Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.26034 · **PDF:** https://arxiv.org/pdf/2603.26034v1
+
+## 1585. Designing Fatigue-Aware VR Interfaces via Biomechanical Models
+
+- **ID:** `2603.26031` · **primary:** `cs.HC` · **Submitted:** 2026-03-27
+- **Authors:** Harshitha Voleti, Charalambos Poullis
+- **abs:** https://arxiv.org/abs/2603.26031 · **PDF:** https://arxiv.org/pdf/2603.26031v1
+
+## 1586. Constitutive parameterized deep energy method for solid mechanics problems with random material parameters
+
+- **ID:** `2603.26030` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Zhangyong Liang, Huanhuan Gao
+- **abs:** https://arxiv.org/abs/2603.26030 · **PDF:** https://arxiv.org/pdf/2603.26030v1
+
+## 1587. Identification of Bivariate Causal Directionality Based on Anticipated Asymmetric Geometries
+
+- **ID:** `2603.26024` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Alex Glushkovsky
+- **abs:** https://arxiv.org/abs/2603.26024 · **PDF:** https://arxiv.org/pdf/2603.26024v1
+
+## 1588. GLU: Global-Local-Uncertainty Fusion for Scalable Spatiotemporal Reconstruction and Forecasting
+
+- **ID:** `2603.26023` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Linzheng Wang, Jason Chen, Nicolas Tricard, et al.
+- **abs:** https://arxiv.org/abs/2603.26023 · **PDF:** https://arxiv.org/pdf/2603.26023v1
+
+## 1589. Central-to-Local Adaptive Generative Diffusion Framework for Improving Gene Expression Prediction in Data-Limited Spatial Transcriptomics
+
+- **ID:** `2603.26827` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Yaoyu Fang, Jiahe Qian, Xinkun Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.26827 · **PDF:** https://arxiv.org/pdf/2603.26827v1
+
+## 1590. Unlabeled Cross-Center Automatic Analysis for TAAD: An Integrated Framework from Segmentation to Clinical Features
+
+- **ID:** `2603.26019` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Mengdi Liu, Qiang Li, Weizhi Nie, et al.
+- **abs:** https://arxiv.org/abs/2603.26019 · **PDF:** https://arxiv.org/pdf/2603.26019v1
+
+## 1591. QuitoBench: A High-Quality Open Time Series Forecasting Benchmark
+
+- **ID:** `2603.26017` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Siqiao Xue, Zhaoyang Zhu, Wei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.26017 · **PDF:** https://arxiv.org/pdf/2603.26017v1
+
+## 1592. VLAgeBench: Benchmarking Large Vision-Language Models for Zero-Shot Human Age Estimation
+
+- **ID:** `2603.26015` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Rakib Hossain Sajib, Md Kishor Morol, Rajan Das Gupta, et al.
+- **abs:** https://arxiv.org/abs/2603.26015 · **PDF:** https://arxiv.org/pdf/2603.26015v1
+
+## 1593. Toward Culturally Grounded Natural Language Processing
+
+- **ID:** `2603.26013` · **primary:** `cs.CL` · **Submitted:** 2026-03-27
+- **Authors:** Sina Bagheri Nezhad
+- **abs:** https://arxiv.org/abs/2603.26013 · **PDF:** https://arxiv.org/pdf/2603.26013v1
+
+## 1594. FairLLaVA: Fairness-Aware Parameter-Efficient Fine-Tuning for Large Vision-Language Assistants
+
+- **ID:** `2603.26008` · **primary:** `cs.CV` · **Submitted:** 2026-03-27
+- **Authors:** Mahesh Bhosale, Abdul Wasi, Shantam Srivastava, et al.
+- **abs:** https://arxiv.org/abs/2603.26008 · **PDF:** https://arxiv.org/pdf/2603.26008v1
+
+## 1595. Longitudinal Boundary Sharpness Coefficient Slopes Predict Time to Alzheimer's Disease Conversion in Mild Cognitive Impairment: A Survival Analysis Using the ADNI Cohort
+
+- **ID:** `2603.26007` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-27
+- **Authors:** Ishaan Cherukuri
+- **abs:** https://arxiv.org/abs/2603.26007 · **PDF:** https://arxiv.org/pdf/2603.26007v1
+
+## 1596. DarwinNet: An Evolutionary Network Architecture for Agent-Driven Protocol Synthesis
+
+- **ID:** `2604.01236` · **primary:** `cs.NE` · **Submitted:** 2026-03-27
+- **Authors:** Jinliang Xu, Bingqi Li
+- **abs:** https://arxiv.org/abs/2604.01236 · **PDF:** https://arxiv.org/pdf/2604.01236v1
+
+## 1597. AutoB2G: A Large Language Model-Driven Agentic Framework For Automated Building-Grid Co-Simulation
+
+- **ID:** `2603.26005` · **primary:** `cs.AI` · **Submitted:** 2026-03-27
+- **Authors:** Borui Zhang, Nariman Mahdavi, Subbu Sethuvenkatraman, et al.
+- **abs:** https://arxiv.org/abs/2603.26005 · **PDF:** https://arxiv.org/pdf/2603.26005v1
+
+## 1598. Throughput Optimization as a Strategic Lever in Large-Scale AI Systems: Evidence from Dataloader and Memory Profiling Innovations
+
+- **ID:** `2603.26823` · **primary:** `cs.LG` · **Submitted:** 2026-03-27
+- **Authors:** Mayank Jha
+- **abs:** https://arxiv.org/abs/2603.26823 · **PDF:** https://arxiv.org/pdf/2603.26823v1
+
+## 1599. Epileptic Seizure Prediction Using Patient-Adaptive Transformer Networks
+
+- **ID:** `2603.26821` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Mohamed Mahdi, Asma Baghdadi
+- **abs:** https://arxiv.org/abs/2603.26821 · **PDF:** https://arxiv.org/pdf/2603.26821v1
+
+## 1600. Policy-Guided World Model Planning for Language-Conditioned Visual Navigation
+
+- **ID:** `2603.25981` · **primary:** `cs.RO` · **Submitted:** 2026-03-26
+- **Authors:** Amirhosein Chahe, Lifeng Zhou
+- **abs:** https://arxiv.org/abs/2603.25981 · **PDF:** https://arxiv.org/pdf/2603.25981v1
+
+## 1601. A Priori Sampling of Transition States with Guided Diffusion
+
+- **ID:** `2603.25980` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-26
+- **Authors:** Hyukjun Lim, Soojung Yang, Lucas Pinède, et al.
+- **abs:** https://arxiv.org/abs/2603.25980 · **PDF:** https://arxiv.org/pdf/2603.25980v1
+
+## 1602. Second-Order, First-Class: A Composable Stack for Curvature-Aware Training
+
+- **ID:** `2603.25976` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Mikalai Korbit, Mario Zanon
+- **abs:** https://arxiv.org/abs/2603.25976 · **PDF:** https://arxiv.org/pdf/2603.25976v1
+
+## 1603. Do Neurons Dream of Primitive Operators? Wake-Sleep Compression Rediscovers Schank's Event Semantics
+
+- **ID:** `2603.25975` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Peter Balogh
+- **abs:** https://arxiv.org/abs/2603.25975 · **PDF:** https://arxiv.org/pdf/2603.25975v1
+
+## 1604. MemoryCD: Benchmarking Long-Context User Memory of LLM Agents for Lifelong Cross-Domain Personalization
+
+- **ID:** `2603.25973` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Weizhi Zhang, Xiaokai Wei, Wei-Chieh Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.25973 · **PDF:** https://arxiv.org/pdf/2603.25973v1
+
+## 1605. When Chain-of-Thought Backfires: Evaluating Prompt Sensitivity in Medical Language Models
+
+- **ID:** `2603.25960` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Binesh Sadanandan, Vahid Behzadan
+- **abs:** https://arxiv.org/abs/2603.25960 · **PDF:** https://arxiv.org/pdf/2603.25960v1
+
+## 1606. On the Objective and Feature Weights of Minkowski Weighted k-Means
+
+- **ID:** `2603.25958` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Renato Cordeiro de Amorim, Vladimir Makarenkov
+- **abs:** https://arxiv.org/abs/2603.25958 · **PDF:** https://arxiv.org/pdf/2603.25958v1
+
+## 1607. Adversarial-Robust Multivariate Time-Series Anomaly Detection via Joint Information Retention
+
+- **ID:** `2603.25956` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Hadi Hojjati, Narges Armanfard
+- **abs:** https://arxiv.org/abs/2603.25956 · **PDF:** https://arxiv.org/pdf/2603.25956v1
+
+## 1608. EngineAD: A Real-World Vehicle Engine Anomaly Detection Dataset
+
+- **ID:** `2603.25955` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Hadi Hojjati, Christopher Roth, Rory Woods, et al.
+- **abs:** https://arxiv.org/abs/2603.25955 · **PDF:** https://arxiv.org/pdf/2603.25955v1
+
+## 1609. Online Learning for Dynamic Constellation Topologies
+
+- **ID:** `2603.25954` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** João Norberto, Ricardo Ferreira, Cláudia Soares
+- **abs:** https://arxiv.org/abs/2603.25954 · **PDF:** https://arxiv.org/pdf/2603.25954v1
+
+## 1610. Globalized Adversarial Regret Optimization: Robust Decisions with Uncalibrated Predictions
+
+- **ID:** `2603.25948` · **primary:** `math.OC` · **Submitted:** 2026-03-26
+- **Authors:** Jannis Kurtz, Bart P. G. van Parys
+- **abs:** https://arxiv.org/abs/2603.25948 · **PDF:** https://arxiv.org/pdf/2603.25948v1
+
+## 1611. Collision-Aware Vision-Language Learning for End-to-End Driving with Multimodal Infraction Datasets
+
+- **ID:** `2603.25946` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Alex Koran, Dimitrios Sinodinos, Hadi Hojjati, et al.
+- **abs:** https://arxiv.org/abs/2603.25946 · **PDF:** https://arxiv.org/pdf/2603.25946v1
+
+## 1612. Can Small Models Reason About Legal Documents? A Comparative Study
+
+- **ID:** `2603.25944` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Snehit Vaddi
+- **abs:** https://arxiv.org/abs/2603.25944 · **PDF:** https://arxiv.org/pdf/2603.25944v1
+
+## 1613. Reinforcing Structured Chain-of-Thought for Video Understanding
+
+- **ID:** `2603.25942` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Peiyao Wang, Haotian Xu, Noranart Vesdapunt, et al.
+- **abs:** https://arxiv.org/abs/2603.25942 · **PDF:** https://arxiv.org/pdf/2603.25942v1
+
+## 1614. Can Vision Foundation Models Navigate? Zero-Shot Real-World Evaluation and Lessons Learned
+
+- **ID:** `2603.25937` · **primary:** `cs.RO` · **Submitted:** 2026-03-26
+- **Authors:** Maeva Guerrier, Karthik Soma, Jana Pavlasek, et al.
+- **abs:** https://arxiv.org/abs/2603.25937 · **PDF:** https://arxiv.org/pdf/2603.25937v1
+
+## 1615. DenseSwinV2: Channel Attentive Dual Branch CNN Transformer Learning for Cassava Leaf Disease Classification
+
+- **ID:** `2603.25935` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Shah Saood, Saddam Hussain Khan
+- **abs:** https://arxiv.org/abs/2603.25935 · **PDF:** https://arxiv.org/pdf/2603.25935v1
+
+## 1616. Sharp Concentration Inequalities: Phase Transition and Mixing of Orlicz Tails with Variance
+
+- **ID:** `2603.25934` · **primary:** `math.ST` · **Submitted:** 2026-03-26
+- **Authors:** Yinan Shen, Jinchi Lv
+- **abs:** https://arxiv.org/abs/2603.25934 · **PDF:** https://arxiv.org/pdf/2603.25934v1
+
+## 1617. DiReCT: Disentangled Regularization of Contrastive Trajectories for Physics-Refined Video Generation
+
+- **ID:** `2603.25931` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Abolfazl Meyarian, Amin Karimi Monsefi, Rajiv Ramnath, et al.
+- **abs:** https://arxiv.org/abs/2603.25931 · **PDF:** https://arxiv.org/pdf/2603.25931v1
+
+## 1618. Density-aware Soft Context Compression with Semi-Dynamic Compression Ratio
+
+- **ID:** `2603.25926` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Yijiong Yu, Shuai Yuan, Jie Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.25926 · **PDF:** https://arxiv.org/pdf/2603.25926v1
+
+## 1619. Personalizing Mathematical Game-based Learning for Children: A Preliminary Study
+
+- **ID:** `2603.25925` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Jie Gao, Adam K. Dubé
+- **abs:** https://arxiv.org/abs/2603.25925 · **PDF:** https://arxiv.org/pdf/2603.25925v1
+
+## 1620. Good Scores, Bad Data: A Metric for Multimodal Coherence
+
+- **ID:** `2603.25924` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Vasundra Srinivasan
+- **abs:** https://arxiv.org/abs/2603.25924 · **PDF:** https://arxiv.org/pdf/2603.25924v1
+
+## 1621. Preventing Data Leakage in EEG-Based Survival Prediction: A Two-Stage Embedding and Transformer Framework
+
+- **ID:** `2603.25923` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Yixin Zhou, Zhixiang Liu, Vladimir I. Zadorozhny, et al.
+- **abs:** https://arxiv.org/abs/2603.25923 · **PDF:** https://arxiv.org/pdf/2603.25923v1
+
+## 1622. Parameter-Free Dynamic Regret for Unconstrained Linear Bandits
+
+- **ID:** `2603.25916` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Alberto Rumi, Andrew Jacobsen, Nicolò Cesa-Bianchi, et al.
+- **abs:** https://arxiv.org/abs/2603.25916 · **PDF:** https://arxiv.org/pdf/2603.25916v1
+
+## 1623. Robust Tensor-on-Tensor Regression
+
+- **ID:** `2603.25911` · **primary:** `stat.ME` · **Submitted:** 2026-03-26
+- **Authors:** Mehdi Hirari, Fabio Centofanti, Mia Hubert, et al.
+- **abs:** https://arxiv.org/abs/2603.25911 · **PDF:** https://arxiv.org/pdf/2603.25911v1
+
+## 1624. Decoding Defensive Coverage Responsibilities in American Football Using Factorized Attention Based Transformer Models
+
+- **ID:** `2603.25901` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Kevin Song, Evan Diewald, Ornob Siddiquee, et al.
+- **abs:** https://arxiv.org/abs/2603.25901 · **PDF:** https://arxiv.org/pdf/2603.25901v1
+
+## 1625. On Integrating Resilience and Human Oversight into LLM-Assisted Modeling Workflows for Digital Twins
+
+- **ID:** `2603.25898` · **primary:** `eess.SY` · **Submitted:** 2026-03-26
+- **Authors:** Lekshmi P, Neha Karanjkar
+- **abs:** https://arxiv.org/abs/2603.25898 · **PDF:** https://arxiv.org/pdf/2603.25898v1
+
+## 1626. Data-Driven Plasticity Modeling via Acoustic Profiling
+
+- **ID:** `2603.25894` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Khalid El-Awady
+- **abs:** https://arxiv.org/abs/2603.25894 · **PDF:** https://arxiv.org/pdf/2603.25894v1
+
+## 1627. Spectral Coherence Index: A Model-Free Metric for Protein Structural Ensemble Quality Assessment
+
+- **ID:** `2603.25880` · **primary:** `q-bio.QM` · **Submitted:** 2026-03-26
+- **Authors:** Yuda Bi, Huaiwen Zhang, Jingnan Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.25880 · **PDF:** https://arxiv.org/pdf/2603.25880v1
+
+## 1628. DRiffusion: Draft-and-Refine Process Parallelizes Diffusion Models with Ease
+
+- **ID:** `2603.25872` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Runsheng Bai, Chengyu Zhang, Yangdong Deng
+- **abs:** https://arxiv.org/abs/2603.25872 · **PDF:** https://arxiv.org/pdf/2603.25872v1
+
+## 1629. Speech-Synchronized Whiteboard Generation via VLM-Driven Structured Drawing Representations
+
+- **ID:** `2603.25870` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Suraj Prasad, Pinak Mahapatra
+- **abs:** https://arxiv.org/abs/2603.25870 · **PDF:** https://arxiv.org/pdf/2603.25870v1
+
+## 1630. PiCSRL: Physics-Informed Contextual Spectral Reinforcement Learning
+
+- **ID:** `2603.26816` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Mitra Nasr Azadani, Syed Usama Imtiaz, Nasrin Alamdari
+- **abs:** https://arxiv.org/abs/2603.26816 · **PDF:** https://arxiv.org/pdf/2603.26816v1
+
+## 1631. Learning to Recorrupt: Noise Distribution Agnostic Self-Supervised Image Denoising
+
+- **ID:** `2603.25869` · **primary:** `eess.IV` · **Submitted:** 2026-03-26
+- **Authors:** Brayan Monroy, Jorge Bacca, Julián Tachella
+- **abs:** https://arxiv.org/abs/2603.25869 · **PDF:** https://arxiv.org/pdf/2603.25869v1
+
+## 1632. GUIDE: A Benchmark for Understanding and Assisting Users in Open-Ended GUI Tasks
+
+- **ID:** `2603.25864` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Saelyne Yang, Jaesang Yu, Yi-Hao Peng, et al.
+- **abs:** https://arxiv.org/abs/2603.25864 · **PDF:** https://arxiv.org/pdf/2603.25864v1
+
+## 1633. Dynamic LIBRAS Gesture Recognition via CNN over Spatiotemporal Matrix Representation
+
+- **ID:** `2603.25863` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Jasmine Moreira
+- **abs:** https://arxiv.org/abs/2603.25863 · **PDF:** https://arxiv.org/pdf/2603.25863v1
+
+## 1634. Methods for Knowledge Graph Construction from Text Collections: Development and Applications
+
+- **ID:** `2603.25862` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Vanni Zavarella
+- **abs:** https://arxiv.org/abs/2603.25862 · **PDF:** https://arxiv.org/pdf/2603.25862v1
+
+## 1635. Why Safety Probes Catch Liars But Miss Fanatics
+
+- **ID:** `2603.25861` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Kristiyan Haralambiev
+- **abs:** https://arxiv.org/abs/2603.25861 · **PDF:** https://arxiv.org/pdf/2603.25861v1
+
+## 1636. On the Expressive Power of Contextual Relations in Transformers
+
+- **ID:** `2603.25860` · **primary:** `stat.ML` · **Submitted:** 2026-03-26
+- **Authors:** Demián Fraiman
+- **abs:** https://arxiv.org/abs/2603.25860 · **PDF:** https://arxiv.org/pdf/2603.25860v1
+
+## 1637. In-Context Molecular Property Prediction with LLMs: A Blinding Study on Memorization and Knowledge Conflicts
+
+- **ID:** `2603.25857` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Matthias Busch, Marius Tacke, Sviatlana V. Lamaka, et al.
+- **abs:** https://arxiv.org/abs/2603.25857 · **PDF:** https://arxiv.org/pdf/2603.25857v1
+
+## 1638. Incorporating contextual information into KGWAS for interpretable GWAS discovery
+
+- **ID:** `2603.25855` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Cheng Jiang, Brady Ryan, Megan Crow, et al.
+- **abs:** https://arxiv.org/abs/2603.25855 · **PDF:** https://arxiv.org/pdf/2603.25855v1
+
+## 1639. Self-Organized Optical Pathways in Optofluidic Photonic Crystals
+
+- **ID:** `2603.25845` · **primary:** `physics.optics` · **Submitted:** 2026-03-26
+- **Authors:** Steven Motta
+- **abs:** https://arxiv.org/abs/2603.25845 · **PDF:** https://arxiv.org/pdf/2603.25845v1
+
+## 1640. GazeQwen: Lightweight Gaze-Conditioned LLM Modulation for Streaming Video Understanding
+
+- **ID:** `2603.25841` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Trong Thang Pham, Hien Nguyen, Ngan Le
+- **abs:** https://arxiv.org/abs/2603.25841 · **PDF:** https://arxiv.org/pdf/2603.25841v1
+
+## 1641. A Compression Perspective on Simplicity Bias
+
+- **ID:** `2603.25839` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Tom Marty, Eric Elmoznino, Leo Gagnon, et al.
+- **abs:** https://arxiv.org/abs/2603.25839 · **PDF:** https://arxiv.org/pdf/2603.25839v1
+
+## 1642. Gradient-Informed Training for Low-Resource Multilingual Speech Translation
+
+- **ID:** `2603.25836` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Ruiyan Sun, Satoshi Nakamura
+- **abs:** https://arxiv.org/abs/2603.25836 · **PDF:** https://arxiv.org/pdf/2603.25836v1
+
+## 1643. A Neural Score-Based Particle Method for the Vlasov-Maxwell-Landau System
+
+- **ID:** `2603.25832` · **primary:** `math.NA` · **Submitted:** 2026-03-26
+- **Authors:** Vasily Ilin, Jingwei Hu
+- **abs:** https://arxiv.org/abs/2603.25832 · **PDF:** https://arxiv.org/pdf/2603.25832v1
+
+## 1644. Runtime Burden Allocation for Structured LLM Routing in Agentic Expert Systems: A Full-Factorial Cross-Backend Methodology
+
+- **ID:** `2604.01235` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Zhou Hanlin, Chan Huah Yong
+- **abs:** https://arxiv.org/abs/2604.01235 · **PDF:** https://arxiv.org/pdf/2604.01235v1
+
+## 1645. ViGoR-Bench: How Far Are Visual Generative Models From Zero-Shot Visual Reasoners?
+
+- **ID:** `2603.25823` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Haonan Han, Jiancheng Huang, Xiaopeng Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.25823 · **PDF:** https://arxiv.org/pdf/2603.25823v1
+
+## 1646. Doctorina MedBench: End-to-End Evaluation of Agent-Based Medical AI
+
+- **ID:** `2603.25821` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Anna Kozlova, Stanislau Salavei, Pavel Satalkin, et al.
+- **abs:** https://arxiv.org/abs/2603.25821 · **PDF:** https://arxiv.org/pdf/2603.25821v1
+
+## 1647. MAGNET: Autonomous Expert Model Generation via Decentralized Autoresearch and BitNet Training
+
+- **ID:** `2603.25813` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Yongwan Kim, Sungchul Park
+- **abs:** https://arxiv.org/abs/2603.25813 · **PDF:** https://arxiv.org/pdf/2603.25813v1
+
+## 1648. ExVerus: Verus Proof Repair via Counterexample Reasoning
+
+- **ID:** `2603.25810` · **primary:** `cs.PL` · **Submitted:** 2026-03-26
+- **Authors:** Jun Yang, Yuechun Sun, Yi Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.25810 · **PDF:** https://arxiv.org/pdf/2603.25810v2
+
+## 1649. RealChart2Code: Advancing Chart-to-Code Generation with Real Data and Multi-Task Evaluation
+
+- **ID:** `2603.25804` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Jiajun Zhang, Yuying Li, Zhixun Li, et al.
+- **abs:** https://arxiv.org/abs/2603.25804 · **PDF:** https://arxiv.org/pdf/2603.25804v1
+
+## 1650. Do All Vision Transformers Need Registers? A Cross-Architectural Reassessment
+
+- **ID:** `2603.25803` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Spiros Baxevanakis, Platon Karageorgis, Ioannis Dravilas, et al.
+- **abs:** https://arxiv.org/abs/2603.25803 · **PDF:** https://arxiv.org/pdf/2603.25803v1
+
+## 1651. Resolving the Robustness-Precision Trade-off in Financial RAG through Hybrid Document-Routed Retrieval
+
+- **ID:** `2603.26815` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Zhiyuan Cheng, Longying Lai, Yue Liu
+- **abs:** https://arxiv.org/abs/2603.26815 · **PDF:** https://arxiv.org/pdf/2603.26815v1
+
+## 1652. Beyond identifiability: Learning causal representations with few environments and finite samples
+
+- **ID:** `2603.25796` · **primary:** `stat.ML` · **Submitted:** 2026-03-26
+- **Authors:** Inbeom Lee, Tongtong Jin, Bryon Aragam
+- **abs:** https://arxiv.org/abs/2603.25796 · **PDF:** https://arxiv.org/pdf/2603.25796v1
+
+## 1653. Vision Transformers and Graph Neural Networks for Charged Particle Tracking in the ATLAS Muon Spectrometer
+
+- **ID:** `2603.25793` · **primary:** `physics.data-an` · **Submitted:** 2026-03-26
+- **Authors:** Jonathan Renusch
+- **abs:** https://arxiv.org/abs/2603.25793 · **PDF:** https://arxiv.org/pdf/2603.25793v1
+
+## 1654. Vega: Learning to Drive with Natural Language Instructions
+
+- **ID:** `2603.25741` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Sicheng Zuo, Yuxuan Li, Wenzhao Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.25741 · **PDF:** https://arxiv.org/pdf/2603.25741v2
+
+## 1655. Drive My Way: Preference Alignment of Vision-Language-Action Model for Personalized Driving
+
+- **ID:** `2603.25740` · **primary:** `cs.RO` · **Submitted:** 2026-03-26
+- **Authors:** Zehao Wang, Huaide Jiang, Shuaiwu Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.25740 · **PDF:** https://arxiv.org/pdf/2603.25740v1
+
+## 1656. Training the Knowledge Base through Evidence Distillation and Write-Back Enrichment
+
+- **ID:** `2603.25737` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Yuxing Lu, Xukai Zhao, Wei Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.25737 · **PDF:** https://arxiv.org/pdf/2603.25737v1
+
+## 1657. PackForcing: Short Video Training Suffices for Long Video Sampling and Long Context Inference
+
+- **ID:** `2603.25730` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Xiaofeng Mao, Shaohao Rui, Kaining Ying, et al.
+- **abs:** https://arxiv.org/abs/2603.25730 · **PDF:** https://arxiv.org/pdf/2603.25730v1
+
+## 1658. PixelSmile: Toward Fine-Grained Facial Expression Editing
+
+- **ID:** `2603.25728` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Jiabin Hua, Hengyuan Xu, Aojie Li, et al.
+- **abs:** https://arxiv.org/abs/2603.25728 · **PDF:** https://arxiv.org/pdf/2603.25728v1
+
+## 1659. Back to Basics: Revisiting ASR in the Age of Voice Agents
+
+- **ID:** `2603.25727` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Geeyang Tay, Wentao Ma, Jaewon Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.25727 · **PDF:** https://arxiv.org/pdf/2603.25727v1
+
+## 1660. Natural-Language Agent Harnesses
+
+- **ID:** `2603.25723` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Linyue Pan, Lexiao Zou, Shuo Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.25723 · **PDF:** https://arxiv.org/pdf/2603.25723v1
+
+## 1661. No Hard Negatives Required: Concept Centric Learning Leads to Compositionality without Degrading Zero-shot Capabilities of Contrastive Models
+
+- **ID:** `2603.25722` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Hai X. Pham, David T. Hoffmann, Ricardo Guerrero, et al.
+- **abs:** https://arxiv.org/abs/2603.25722 · **PDF:** https://arxiv.org/pdf/2603.25722v1
+
+## 1662. R-C2: Cycle-Consistent Reinforcement Learning Improves Multimodal Reasoning
+
+- **ID:** `2603.25720` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Zirui Zhang, Haoyu Dong, Kexin Pei, et al.
+- **abs:** https://arxiv.org/abs/2603.25720 · **PDF:** https://arxiv.org/pdf/2603.25720v1
+
+## 1663. Agent Factories for High Level Synthesis: How Far Can General-Purpose Coding Agents Go in Hardware Optimization?
+
+- **ID:** `2603.25719` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Abhishek Bhandwaldar, Mihir Choudhury, Ruchir Puri, et al.
+- **abs:** https://arxiv.org/abs/2603.25719 · **PDF:** https://arxiv.org/pdf/2603.25719v1
+
+## 1664. Out of Sight but Not Out of Mind: Hybrid Memory for Dynamic Video World Models
+
+- **ID:** `2603.25716` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Kaijin Chen, Dingkang Liang, Xin Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.25716 · **PDF:** https://arxiv.org/pdf/2603.25716v2
+
+## 1665. S2D2: Fast Decoding for Diffusion LLMs via Training-Free Self-Speculation
+
+- **ID:** `2603.25702` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Ligong Han, Hao Wang, Han Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.25702 · **PDF:** https://arxiv.org/pdf/2603.25702v1
+
+## 1666. Neural Network Conversion of Machine Learning Pipelines
+
+- **ID:** `2603.25699` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Man-Ling Sung, Jan Silovsky, Man-Hung Siu, et al.
+- **abs:** https://arxiv.org/abs/2603.25699 · **PDF:** https://arxiv.org/pdf/2603.25699v1
+
+## 1667. The Kitchen Loop: User-Spec-Driven Development for a Self-Evolving Codebase
+
+- **ID:** `2603.25697` · **primary:** `cs.SE` · **Submitted:** 2026-03-26
+- **Authors:** Yannick Roy
+- **abs:** https://arxiv.org/abs/2603.25697 · **PDF:** https://arxiv.org/pdf/2603.25697v1
+
+## 1668. A Unified Memory Perspective for Probabilistic Trustworthy AI
+
+- **ID:** `2603.25692` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Xueji Zhao, Likai Pei, Jianbo Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.25692 · **PDF:** https://arxiv.org/pdf/2603.25692v1
+
+## 1669. On Neural Scaling Laws for Weather Emulation through Continual Training
+
+- **ID:** `2603.25687` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Shashank Subramanian, Alexander Kiefer, Arnur Nigmetov, et al.
+- **abs:** https://arxiv.org/abs/2603.25687 · **PDF:** https://arxiv.org/pdf/2603.25687v1
+
+## 1670. Just Zoom In: Cross-View Geo-Localization via Autoregressive Zooming
+
+- **ID:** `2603.25686` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Yunus Talha Erzurumlu, Jiyong Kwag, Alper Yilmaz
+- **abs:** https://arxiv.org/abs/2603.25686 · **PDF:** https://arxiv.org/pdf/2603.25686v1
+
+## 1671. Self-Improvement of Large Language Models: A Technical Overview and Future Outlook
+
+- **ID:** `2603.25681` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Haoyan Yang, Mario Xerri, Solha Park, et al.
+- **abs:** https://arxiv.org/abs/2603.25681 · **PDF:** https://arxiv.org/pdf/2603.25681v1
+
+## 1672. Measuring What Matters -- or What's Convenient?: Robustness of LLM-Based Scoring Systems to Construct-Irrelevant Factors
+
+- **ID:** `2603.25674` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Cole Walsh, Rodica Ivan
+- **abs:** https://arxiv.org/abs/2603.25674 · **PDF:** https://arxiv.org/pdf/2603.25674v1
+
+## 1673. Longitudinal Digital Phenotyping for Early Cognitive-Motor Screening
+
+- **ID:** `2603.25673` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Diego Jimenez-Oviedo, Ruben Vera-Rodriguez, Ruben Tolosana, et al.
+- **abs:** https://arxiv.org/abs/2603.25673 · **PDF:** https://arxiv.org/pdf/2603.25673v1
+
+## 1674. Uncertainty-Guided Label Rebalancing for CPS Safety Monitoring
+
+- **ID:** `2603.25670` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** John Ayotunde, Qinghua Xu, Guancheng Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.25670 · **PDF:** https://arxiv.org/pdf/2603.25670v2
+
+## 1675. Instance-optimal stochastic convex optimization: Can we improve upon sample-average and robust stochastic approximation?
+
+- **ID:** `2603.25657` · **primary:** `math.OC` · **Submitted:** 2026-03-26
+- **Authors:** Liwei Jiang, Ashwin Pananjady
+- **abs:** https://arxiv.org/abs/2603.25657 · **PDF:** https://arxiv.org/pdf/2603.25657v1
+
+## 1676. Calorimeter Shower Superresolution with Conditional Normalizing Flows: Implementation and Statistical Evaluation
+
+- **ID:** `2603.26813` · **primary:** `physics.ins-det` · **Submitted:** 2026-03-26
+- **Authors:** Andrea Cosso
+- **abs:** https://arxiv.org/abs/2603.26813 · **PDF:** https://arxiv.org/pdf/2603.26813v1
+
+## 1677. A Mentalistic Interface for Probing Folk-Psychological Attribution to Non-Humanoid Robots
+
+- **ID:** `2603.25646` · **primary:** `cs.RO` · **Submitted:** 2026-03-26
+- **Authors:** Giulio Pisaneschi, Pierpaolo Serio, Estelle Gerbier, et al.
+- **abs:** https://arxiv.org/abs/2603.25646 · **PDF:** https://arxiv.org/pdf/2603.25646v1
+
+## 1678. RenoBench: A Citation Parsing Benchmark
+
+- **ID:** `2603.25640` · **primary:** `cs.DL` · **Submitted:** 2026-03-26
+- **Authors:** Parth Sarin, Juan Pablo Alperin, Adam Buttrick, et al.
+- **abs:** https://arxiv.org/abs/2603.25640 · **PDF:** https://arxiv.org/pdf/2603.25640v1
+
+## 1679. Beyond Via: Analysis and Estimation of the Impact of Large Language Models in Academic Papers
+
+- **ID:** `2603.25638` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Mingmeng Geng, Yuhang Dong, Thierry Poibeau
+- **abs:** https://arxiv.org/abs/2603.25638 · **PDF:** https://arxiv.org/pdf/2603.25638v2
+
+## 1680. Anchored-Branched Steady-state WInd Flow Transformer (AB-SWIFT): a metamodel for 3D atmospheric flow in urban environments
+
+- **ID:** `2603.25635` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Armand de Villeroché, Rem-Sophia Mouradi, Vincent Le Guen, et al.
+- **abs:** https://arxiv.org/abs/2603.25635 · **PDF:** https://arxiv.org/pdf/2603.25635v1
+
+## 1681. Is Mathematical Problem-Solving Expertise in Large Language Models Associated with Assessment Performance?
+
+- **ID:** `2603.25633` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Liang Zhang, Yu Fu, Xinyi Jin
+- **abs:** https://arxiv.org/abs/2603.25633 · **PDF:** https://arxiv.org/pdf/2603.25633v1
+
+## 1682. LanteRn: Latent Visual Structured Reasoning
+
+- **ID:** `2603.25629` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** André G. Viveiros, Nuno Gonçalves, Matthias Lindemann, et al.
+- **abs:** https://arxiv.org/abs/2603.25629 · **PDF:** https://arxiv.org/pdf/2603.25629v1
+
+## 1683. Visual or Textual: Effects of Explanation Format and Personal Characteristics on the Perception of Explanations in an Educational Recommender System
+
+- **ID:** `2603.25624` · **primary:** `cs.HC` · **Submitted:** 2026-03-26
+- **Authors:** Qurat Ul Ain, Mohamed Amine Chatti, Nasim Yazdian Varjani, et al.
+- **abs:** https://arxiv.org/abs/2603.25624 · **PDF:** https://arxiv.org/pdf/2603.25624v1
+
+## 1684. The Geometry of Efficient Nonconvex Sampling
+
+- **ID:** `2603.25622` · **primary:** `cs.DS` · **Submitted:** 2026-03-26
+- **Authors:** Santosh S. Vempala, Andre Wibisono
+- **abs:** https://arxiv.org/abs/2603.25622 · **PDF:** https://arxiv.org/pdf/2603.25622v1
+
+## 1685. PICon: A Multi-Turn Interrogation Framework for Evaluating Persona Agent Consistency
+
+- **ID:** `2603.25620` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Minseo Kim, Sujeong Im, Junseong Choi, et al.
+- **abs:** https://arxiv.org/abs/2603.25620 · **PDF:** https://arxiv.org/pdf/2603.25620v1
+
+## 1686. Social Hippocampus Memory Learning
+
+- **ID:** `2603.25614` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Liping Yi, Zhiming Zhao, Qinghua Hu
+- **abs:** https://arxiv.org/abs/2603.25614 · **PDF:** https://arxiv.org/pdf/2603.25614v1
+
+## 1687. Demographic Fairness in Multimodal LLMs: A Benchmark of Gender and Ethnicity Bias in Face Verification
+
+- **ID:** `2603.25613` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Ünsal Öztürk, Hatef Otroshi Shahreza, Sébastien Marcel
+- **abs:** https://arxiv.org/abs/2603.25613 · **PDF:** https://arxiv.org/pdf/2603.25613v1
+
+## 1688. DeepFAN, a transformer-based deep learning model for human-artificial intelligence collaborative assessment of incidental pulmonary nodules in CT scans: a multi-reader, multi-case trial
+
+- **ID:** `2603.25607` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Zhenchen Zhu, Ge Hu, Weixiong Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.25607 · **PDF:** https://arxiv.org/pdf/2603.25607v1
+
+## 1689. Spatiotemporal System Forecasting with Irregular Time Steps via Masked Autoencoder
+
+- **ID:** `2603.25597` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Kewei Zhu, Yanze Xin, Jinwei Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.25597 · **PDF:** https://arxiv.org/pdf/2603.25597v1
+
+## 1690. A Judge Agent Closes the Reliability Gap in AI-Generated Scientific Simulation
+
+- **ID:** `2603.25780` · **primary:** `cs.SE` · **Submitted:** 2026-03-26
+- **Authors:** Chengshuai Yang
+- **abs:** https://arxiv.org/abs/2603.25780 · **PDF:** https://arxiv.org/pdf/2603.25780v1
+
+## 1691. Pure and Physics-Guided Deep Learning Solutions for Spatio-Temporal Groundwater Level Prediction at Arbitrary Locations
+
+- **ID:** `2603.25779` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Matteo Salis, Gabriele Sartor, Rosa Meo, et al.
+- **abs:** https://arxiv.org/abs/2603.25779 · **PDF:** https://arxiv.org/pdf/2603.25779v1
+
+## 1692. The Rules-and-Facts Model for Simultaneous Generalization and Memorization in Neural Networks
+
+- **ID:** `2603.25579` · **primary:** `stat.ML` · **Submitted:** 2026-03-26
+- **Authors:** Gabriele Farné, Fabrizio Boncoraglio, Lenka Zdeborová
+- **abs:** https://arxiv.org/abs/2603.25579 · **PDF:** https://arxiv.org/pdf/2603.25579v1
+
+## 1693. Hierarchy-Guided Multimodal Representation Learning for Taxonomic Inference
+
+- **ID:** `2603.25573` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Sk Miraj Ahmed, Xi Yu, Yunqi Li, et al.
+- **abs:** https://arxiv.org/abs/2603.25573 · **PDF:** https://arxiv.org/pdf/2603.25573v1
+
+## 1694. Cooperative Deep Reinforcement Learning for Fair RIS Allocation
+
+- **ID:** `2603.25572` · **primary:** `cs.NI` · **Submitted:** 2026-03-26
+- **Authors:** Martin Mark Zan, Stefan Schwarz
+- **abs:** https://arxiv.org/abs/2603.25572 · **PDF:** https://arxiv.org/pdf/2603.25572v1
+
+## 1695. Implicit neural representations for larval zebrafish brain microscopy: a reproducible benchmark on the MapZebrain atlas
+
+- **ID:** `2603.26811` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Agnieszka Pregowska
+- **abs:** https://arxiv.org/abs/2603.26811 · **PDF:** https://arxiv.org/pdf/2603.26811v1
+
+## 1696. TAAC: A gate into Trustable Audio Affective Computing
+
+- **ID:** `2603.25570` · **primary:** `cs.CR` · **Submitted:** 2026-03-26
+- **Authors:** Xintao Hu, Feng-Qi Cui
+- **abs:** https://arxiv.org/abs/2603.25570 · **PDF:** https://arxiv.org/pdf/2603.25570v1
+
+## 1697. Are LLMs Overkill for Databases?: A Study on the Finiteness of SQL
+
+- **ID:** `2603.25568` · **primary:** `cs.DB` · **Submitted:** 2026-03-26
+- **Authors:** Yue Li, David Mimno, Unso Eun Seo Jo
+- **abs:** https://arxiv.org/abs/2603.25568 · **PDF:** https://arxiv.org/pdf/2603.25568v1
+
+## 1698. Revisiting On-Policy Distillation: Empirical Failure Modes and Simple Fixes
+
+- **ID:** `2603.25562` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Yuqian Fu, Haohuan Huang, Kaiwen Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.25562 · **PDF:** https://arxiv.org/pdf/2603.25562v1
+
+## 1699. An Integrative Genome-Scale Metabolic Modeling and Machine Learning Framework for Predicting and Optimizing Biofuel-Relevant Biomass Production in Saccharomyces cerevisiae
+
+- **ID:** `2603.25561` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Neha K. Nair, Aaron D'Souza
+- **abs:** https://arxiv.org/abs/2603.25561 · **PDF:** https://arxiv.org/pdf/2603.25561v1
+
+## 1700. Voxtral TTS
+
+- **ID:** `2603.25551` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Alexander H. Liu, Alexis Tacnet, Andy Ehrenberg, et al.
+- **abs:** https://arxiv.org/abs/2603.25551 · **PDF:** https://arxiv.org/pdf/2603.25551v1
+
+## 1701. Missing-Aware Multimodal Fusion for Unified Microservice Incident Management
+
+- **ID:** `2603.25538` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Wenzhuo Qian, Hailiang Zhao, Ziqi Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.25538 · **PDF:** https://arxiv.org/pdf/2603.25538v2
+
+## 1702. Humans vs Vision-Language Models: A Unified Measure of Narrative Coherence
+
+- **ID:** `2603.25537` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Nikolai Ilinykh, Hyewon Jang, Shalom Lappin, et al.
+- **abs:** https://arxiv.org/abs/2603.25537 · **PDF:** https://arxiv.org/pdf/2603.25537v1
+
+## 1703. Insights on back marking for the automated identification of animals
+
+- **ID:** `2603.25535` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** David Brunner, Marie Bordes, Elisabeth Mayrhuber, et al.
+- **abs:** https://arxiv.org/abs/2603.25535 · **PDF:** https://arxiv.org/pdf/2603.25535v1
+
+## 1704. Synchronous Signal Temporal Logic for Decidable Verification of Cyber-Physical Systems
+
+- **ID:** `2603.25531` · **primary:** `cs.FL` · **Submitted:** 2026-03-26
+- **Authors:** Partha Roop, Sobhan Chatterjee, Avinash Malik, et al.
+- **abs:** https://arxiv.org/abs/2603.25531 · **PDF:** https://arxiv.org/pdf/2603.25531v1
+
+## 1705. Adaptive Subspace Modeling With Functional Tucker Decomposition
+
+- **ID:** `2603.25530` · **primary:** `stat.ML` · **Submitted:** 2026-03-26
+- **Authors:** Noah Steidle, Joppe De Jonghe, Mariya Ishteva
+- **abs:** https://arxiv.org/abs/2603.25530 · **PDF:** https://arxiv.org/pdf/2603.25530v1
+
+## 1706. CHIRP dataset: towards long-term, individual-level, behavioral monitoring of bird populations in the wild
+
+- **ID:** `2603.25524` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Alex Hoi Hang Chan, Neha Singhal, Onur Kocahan, et al.
+- **abs:** https://arxiv.org/abs/2603.25524 · **PDF:** https://arxiv.org/pdf/2603.25524v1
+
+## 1707. NERO-Net: A Neuroevolutionary Approach for the Design of Adversarially Robust CNNs
+
+- **ID:** `2603.25517` · **primary:** `cs.NE` · **Submitted:** 2026-03-26
+- **Authors:** Inês Valentim, Nuno Antunes, Nuno Lourenço
+- **abs:** https://arxiv.org/abs/2603.25517 · **PDF:** https://arxiv.org/pdf/2603.25517v1
+
+## 1708. Challenges in Hyperspectral Imaging for Autonomous Driving: The HSI-Drive Case
+
+- **ID:** `2603.25510` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Koldo Basterretxea, Jon Gutiérrez-Zaballa, Javier Echanobe
+- **abs:** https://arxiv.org/abs/2603.25510 · **PDF:** https://arxiv.org/pdf/2603.25510v1
+
+## 1709. Conformal Prediction for Nonparametric Instrumental Regression
+
+- **ID:** `2603.25509` · **primary:** `econ.EM` · **Submitted:** 2026-03-26
+- **Authors:** Masahiro Kato
+- **abs:** https://arxiv.org/abs/2603.25509 · **PDF:** https://arxiv.org/pdf/2603.25509v1
+
+## 1710. Lightweight GenAI for Network Traffic Synthesis: Fidelity, Augmentation, and Classification
+
+- **ID:** `2603.25507` · **primary:** `cs.NI` · **Submitted:** 2026-03-26
+- **Authors:** Giampaolo Bovenzi, Domenico Ciuonzo, Jonatan Krolikowski, et al.
+- **abs:** https://arxiv.org/abs/2603.25507 · **PDF:** https://arxiv.org/pdf/2603.25507v1
+
+## 1711. An Experimental Comparison of the Most Popular Approaches to Fake News Detection
+
+- **ID:** `2603.25501` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Pietro Dell'Oglio, Alessandro Bondielli, Francesco Marcelloni, et al.
+- **abs:** https://arxiv.org/abs/2603.25501 · **PDF:** https://arxiv.org/pdf/2603.25501v1
+
+## 1712. Knowledge-Guided Failure Prediction: Detecting When Object Detectors Miss Safety-Critical Objects
+
+- **ID:** `2603.25499` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Jakob Paul Zimmermann, Gerrit Holzbach, David Lerch
+- **abs:** https://arxiv.org/abs/2603.25499 · **PDF:** https://arxiv.org/pdf/2603.25499v1
+
+## 1713. EcoThink: A Green Adaptive Inference Framework for Sustainable and Accessible Agents
+
+- **ID:** `2603.25498` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Linxiao Li, Zhixiang Lu
+- **abs:** https://arxiv.org/abs/2603.25498 · **PDF:** https://arxiv.org/pdf/2603.25498v1
+
+## 1714. Interpretable PM2.5 Forecasting for Urban Air Quality: A Comparative Study of Operational Time-Series Models
+
+- **ID:** `2603.25495` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Moazzam Umer Gondal, Hamad ul Qudous, Asma Ahmad Farhan, et al.
+- **abs:** https://arxiv.org/abs/2603.25495 · **PDF:** https://arxiv.org/pdf/2603.25495v1
+
+## 1715. Translation Asymmetry in LLMs as a Data Augmentation Factor: A Case Study for 6 Romansh Language Varieties
+
+- **ID:** `2603.25489` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Jannis Vamvas, Ignacio Pérez Prat, Angela Heldstab, et al.
+- **abs:** https://arxiv.org/abs/2603.25489 · **PDF:** https://arxiv.org/pdf/2603.25489v1
+
+## 1716. Dictionary-based Pathology Mining with Hard-instance-assisted Classifier Debiasing for Genetic Biomarker Prediction from WSIs
+
+- **ID:** `2603.26809` · **primary:** `q-bio.QM` · **Submitted:** 2026-03-26
+- **Authors:** Ling Zhang, Boxiang Yun, Ting Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.26809 · **PDF:** https://arxiv.org/pdf/2603.26809v1
+
+## 1717. Retraining as Approximate Bayesian Inference
+
+- **ID:** `2603.25480` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Harrison Katz
+- **abs:** https://arxiv.org/abs/2603.25480 · **PDF:** https://arxiv.org/pdf/2603.25480v1
+
+## 1718. How Class Ontology and Data Scale Affect Audio Transfer Learning
+
+- **ID:** `2603.25476` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Manuel Milling, Andreas Triantafyllopoulos, Alexander Gebhard, et al.
+- **abs:** https://arxiv.org/abs/2603.25476 · **PDF:** https://arxiv.org/pdf/2603.25476v1
+
+## 1719. Causal-INSIGHT: Probing Temporal Models to Extract Causal Structure
+
+- **ID:** `2603.25473` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Benjamin Redden, Hui Wang, Shuyan Li
+- **abs:** https://arxiv.org/abs/2603.25473 · **PDF:** https://arxiv.org/pdf/2603.25473v1
+
+## 1720. Not a fragment, but the whole: Map-based evaluation of data-driven Fire Danger Index models
+
+- **ID:** `2603.25469` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Shahbaz Alvi, Italo Epicoco, Jose Maria Costa Saura
+- **abs:** https://arxiv.org/abs/2603.25469 · **PDF:** https://arxiv.org/pdf/2603.25469v1
+
+## 1721. Residual-as-Teacher: Mitigating Bias Propagation in Student--Teacher Estimation
+
+- **ID:** `2603.25466` · **primary:** `stat.ML` · **Submitted:** 2026-03-26
+- **Authors:** Kakei Yamamoto, Martin J. Wainwright
+- **abs:** https://arxiv.org/abs/2603.25466 · **PDF:** https://arxiv.org/pdf/2603.25466v1
+
+## 1722. Maximum Entropy Behavior Exploration for Sim2Real Zero-Shot Reinforcement Learning
+
+- **ID:** `2603.25464` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Jiajun Hu, Nuria Armengol Urpi, Jin Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.25464 · **PDF:** https://arxiv.org/pdf/2603.25464v1
+
+## 1723. Temporally Decoupled Diffusion Planning for Autonomous Driving
+
+- **ID:** `2603.25462` · **primary:** `cs.RO` · **Submitted:** 2026-03-26
+- **Authors:** Xiang Li, Bikun Wang, John Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.25462 · **PDF:** https://arxiv.org/pdf/2603.25462v1
+
+## 1724. Cross-Model Disagreement as a Label-Free Correctness Signal
+
+- **ID:** `2603.25450` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Matt Gorbett, Suman Jana
+- **abs:** https://arxiv.org/abs/2603.25450 · **PDF:** https://arxiv.org/pdf/2603.25450v1
+
+## 1725. The Symmetric Perceptron: a Teacher-Student Scenario
+
+- **ID:** `2603.25440` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-03-26
+- **Authors:** Giovanni Catania, Aurélien Decelle, Suhanee Korpe
+- **abs:** https://arxiv.org/abs/2603.25440 · **PDF:** https://arxiv.org/pdf/2603.25440v1
+
+## 1726. Challenges and opportunities for AI to help deliver fusion energy
+
+- **ID:** `2603.25777` · **primary:** `physics.plasm-ph` · **Submitted:** 2026-03-26
+- **Authors:** Adriano Agnello, Helen Brooks, Cyd Cowley, et al.
+- **abs:** https://arxiv.org/abs/2603.25777 · **PDF:** https://arxiv.org/pdf/2603.25777v1
+
+## 1727. From Manipulation to Mistrust: Explaining Diverse Micro-Video Misinformation for Robust Debunking in the Wild
+
+- **ID:** `2603.25423` · **primary:** `cs.SI` · **Submitted:** 2026-03-26
+- **Authors:** Zhi Zeng, Yifei Yang, Jiaying Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.25423 · **PDF:** https://arxiv.org/pdf/2603.25423v1
+
+## 1728. Navigating the Prompt Space: Improving LLM Classification of Social Science Texts Through Prompt Engineering
+
+- **ID:** `2603.25422` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Erkan Gunes, Christoffer Florczak, Tevfik Murat Yildirim
+- **abs:** https://arxiv.org/abs/2603.25422 · **PDF:** https://arxiv.org/pdf/2603.25422v1
+
+## 1729. TAPO: Translation Augmented Policy Optimization for Multilingual Mathematical Reasoning
+
+- **ID:** `2603.25419` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Xu Huang, Zhejian Lai, Zixian Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.25419 · **PDF:** https://arxiv.org/pdf/2603.25419v1
+
+## 1730. Modernising Reinforcement Learning-Based Navigation for Embodied Semantic Scene Graph Generation
+
+- **ID:** `2603.25415` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Roman Kueble, Marco Hueller, Mrunmai Phatak, et al.
+- **abs:** https://arxiv.org/abs/2603.25415 · **PDF:** https://arxiv.org/pdf/2603.25415v1
+
+## 1731. Decidable By Construction: Design-Time Verification for Trustworthy AI
+
+- **ID:** `2603.25414` · **primary:** `cs.PL` · **Submitted:** 2026-03-26
+- **Authors:** Houston Haynes
+- **abs:** https://arxiv.org/abs/2603.25414 · **PDF:** https://arxiv.org/pdf/2603.25414v1
+
+## 1732. Beyond Content Safety: Real-Time Monitoring for Reasoning Vulnerabilities in Large Language Models
+
+- **ID:** `2603.25412` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Xunguang Wang, Yuguang Zhou, Qingyue Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.25412 · **PDF:** https://arxiv.org/pdf/2603.25412v1
+
+## 1733. SAHMM-VAE: A Source-Wise Adaptive Hidden Markov Prior Variational Autoencoder for Unsupervised Blind Source Separation
+
+- **ID:** `2603.25776` · **primary:** `stat.ML` · **Submitted:** 2026-03-26
+- **Authors:** Yuan-Hao Wei
+- **abs:** https://arxiv.org/abs/2603.25776 · **PDF:** https://arxiv.org/pdf/2603.25776v1
+
+## 1734. System Design for Maintaining Internal State Consistency in Long-Horizon Robotic Tabletop Games
+
+- **ID:** `2603.25405` · **primary:** `cs.RO` · **Submitted:** 2026-03-26
+- **Authors:** Guangyu Zhao, Ceyao Zhang, Chengdong Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.25405 · **PDF:** https://arxiv.org/pdf/2603.25405v1
+
+## 1735. Shape and Substance: Dual-Layer Side-Channel Attacks on Local Vision-Language Models
+
+- **ID:** `2603.25403` · **primary:** `cs.CR` · **Submitted:** 2026-03-26
+- **Authors:** Eyal Hadad, Mordechai Guri
+- **abs:** https://arxiv.org/abs/2603.25403 · **PDF:** https://arxiv.org/pdf/2603.25403v2
+
+## 1736. A Causal Framework for Evaluating ICU Discharge Strategies
+
+- **ID:** `2603.25397` · **primary:** `stat.ME` · **Submitted:** 2026-03-26
+- **Authors:** Sagar Nagaraj Simha, Juliette Ortholand, Dave Dongelmans, et al.
+- **abs:** https://arxiv.org/abs/2603.25397 · **PDF:** https://arxiv.org/pdf/2603.25397v1
+
+## 1737. GlowQ: Group-Shared LOw-Rank Approximation for Quantized LLMs
+
+- **ID:** `2603.25385` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Selim An, Il hong Suh, Yeseong Kim
+- **abs:** https://arxiv.org/abs/2603.25385 · **PDF:** https://arxiv.org/pdf/2603.25385v1
+
+## 1738. Enabling ab initio geometry optimization of strongly correlated systems with transferable deep quantum Monte Carlo
+
+- **ID:** `2603.25381` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-26
+- **Authors:** P. Bernát Szabó, Zeno Schätzle, Frank Noé
+- **abs:** https://arxiv.org/abs/2603.25381 · **PDF:** https://arxiv.org/pdf/2603.25381v1
+
+## 1739. Does Structured Intent Representation Generalize? A Cross-Language, Cross-Model Empirical Study of 5W3H Prompting
+
+- **ID:** `2603.25379` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Peng Gang
+- **abs:** https://arxiv.org/abs/2603.25379 · **PDF:** https://arxiv.org/pdf/2603.25379v1
+
+## 1740. Supercharging Federated Intelligence Retrieval
+
+- **ID:** `2603.25374` · **primary:** `cs.IR` · **Submitted:** 2026-03-26
+- **Authors:** Dimitris Stripelis, Patrick Foley, Mohammad Naseri, et al.
+- **abs:** https://arxiv.org/abs/2603.25374 · **PDF:** https://arxiv.org/pdf/2603.25374v1
+
+## 1741. Hessian-informed machine learning interatomic potential towards bridging theory and experiments
+
+- **ID:** `2603.25373` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Bangchen Yin, Jian Ouyang, Zhen Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.25373 · **PDF:** https://arxiv.org/pdf/2603.25373v1
+
+## 1742. A Distribution-to-Distribution Neural Probabilistic Forecasting Framework for Dynamical Systems
+
+- **ID:** `2603.25370` · **primary:** `stat.ML` · **Submitted:** 2026-03-26
+- **Authors:** Tianlin Yang, Hailiang Du, Louis Aslett
+- **abs:** https://arxiv.org/abs/2603.25370 · **PDF:** https://arxiv.org/pdf/2603.25370v1
+
+## 1743. Integrating Deep RL and Bayesian Inference for ObjectNav in Mobile Robotics
+
+- **ID:** `2603.25366` · **primary:** `cs.RO` · **Submitted:** 2026-03-26
+- **Authors:** João Castelo-Branco, José Santos-Victor, Alexandre Bernardino
+- **abs:** https://arxiv.org/abs/2603.25366 · **PDF:** https://arxiv.org/pdf/2603.25366v1
+
+## 1744. 4OPS: Structural Difficulty Modeling in Integer Arithmetic Puzzles
+
+- **ID:** `2603.25356` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Yunus E. Zeytuncu
+- **abs:** https://arxiv.org/abs/2603.25356 · **PDF:** https://arxiv.org/pdf/2603.25356v1
+
+## 1745. Image Rotation Angle Estimation: Comparing Circular-Aware Methods
+
+- **ID:** `2603.25351` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Maximilian Woehrer
+- **abs:** https://arxiv.org/abs/2603.25351 · **PDF:** https://arxiv.org/pdf/2603.25351v1
+
+## 1746. Design and Development of an ML/DL Attack Resistance of RC-Based PUF for IoT Security
+
+- **ID:** `2603.28798` · **primary:** `cs.CR` · **Submitted:** 2026-03-26
+- **Authors:** Joy Acharya, Smit Patel, Paawan Sharma, et al.
+- **abs:** https://arxiv.org/abs/2603.28798 · **PDF:** https://arxiv.org/pdf/2603.28798v1
+
+## 1747. GroupRAG: Cognitively Inspired Group-Aware Retrieval and Reasoning via Knowledge-Driven Problem Structuring
+
+- **ID:** `2603.26807` · **primary:** `cs.IR` · **Submitted:** 2026-03-26
+- **Authors:** Xinyi Duan, Yuanrong Tang, Jiangtao Gong
+- **abs:** https://arxiv.org/abs/2603.26807 · **PDF:** https://arxiv.org/pdf/2603.26807v1
+
+## 1748. From Intent to Evidence: A Categorical Approach for Structural Evaluation of Deep Research Agents
+
+- **ID:** `2603.25342` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Shuoling Liu, Zhiquan Tan, Kun Yi, et al.
+- **abs:** https://arxiv.org/abs/2603.25342 · **PDF:** https://arxiv.org/pdf/2603.25342v1
+
+## 1749. Large Language Model as Token Compressor and Decompressor
+
+- **ID:** `2603.25340` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Wenbing Li, Zikai Song, Jielei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.25340 · **PDF:** https://arxiv.org/pdf/2603.25340v1
+
+## 1750. Agentic Trust Coordination for Federated Learning through Adaptive Thresholding and Autonomous Decision Making in Sustainable and Resilient Industrial Networks
+
+- **ID:** `2603.25334` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Paul Shepherd, Tasos Dagiuklas, Bugra Alkan, et al.
+- **abs:** https://arxiv.org/abs/2603.25334 · **PDF:** https://arxiv.org/pdf/2603.25334v1
+
+## 1751. Adaptive Chunking: Optimizing Chunking-Method Selection for RAG
+
+- **ID:** `2603.25333` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Paulo Roberto de Moura Júnior, Jean Lelong, Annabelle Blangero
+- **abs:** https://arxiv.org/abs/2603.25333 · **PDF:** https://arxiv.org/pdf/2603.25333v1
+
+## 1752. Beyond Detection: Rethinking Education in the Age of AI-writing
+
+- **ID:** `2603.25329` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Maria Marina, Alexander Panchenko, Vasily Konovalov
+- **abs:** https://arxiv.org/abs/2603.25329 · **PDF:** https://arxiv.org/pdf/2603.25329v1
+
+## 1753. Macroscopic Characteristics of Mixed Traffic Flow with Deep Reinforcement Learning Based Automated and Human-Driven Vehicles
+
+- **ID:** `2603.25328` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Pankaj Kumar, Pranamesh Chakraborty, Subrahmanya Swamy Peruru
+- **abs:** https://arxiv.org/abs/2603.25328 · **PDF:** https://arxiv.org/pdf/2603.25328v1
+
+## 1754. Evaluating Language Models for Harmful Manipulation
+
+- **ID:** `2603.25326` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Canfer Akbulut, Rasmi Elasmar, Abhishek Roy, et al.
+- **abs:** https://arxiv.org/abs/2603.25326 · **PDF:** https://arxiv.org/pdf/2603.25326v2
+
+## 1755. How Pruning Reshapes Features: Sparse Autoencoder Analysis of Weight-Pruned Language Models
+
+- **ID:** `2603.25325` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Hector Borobia, Elies Seguí-Mas, Guillermina Tormo-Carbó
+- **abs:** https://arxiv.org/abs/2603.25325 · **PDF:** https://arxiv.org/pdf/2603.25325v1
+
+## 1756. AD-CARE: A Guideline-grounded, Modality-agnostic LLM Agent for Real-world Alzheimer's Disease Diagnosis with Multi-cohort Assessment, Fairness Analysis, and Reader Study
+
+- **ID:** `2603.25322` · **primary:** `cs.MA` · **Submitted:** 2026-03-26
+- **Authors:** Wenlong Hou, Sheng Bi, Guangqian Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.25322 · **PDF:** https://arxiv.org/pdf/2603.25322v1
+
+## 1757. Practical Efficient Global Optimization is No-regret
+
+- **ID:** `2603.25311` · **primary:** `stat.ML` · **Submitted:** 2026-03-26
+- **Authors:** Jingyi Wang, Haowei Wang, Nai-Yuan Chiang, et al.
+- **abs:** https://arxiv.org/abs/2603.25311 · **PDF:** https://arxiv.org/pdf/2603.25311v1
+
+## 1758. Separate Before You Compress: The WWHO Tokenization Architecture
+
+- **ID:** `2603.25309` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Kusal Darshana
+- **abs:** https://arxiv.org/abs/2603.25309 · **PDF:** https://arxiv.org/pdf/2603.25309v1
+
+## 1759. DAGverse: Building Document-Grounded Semantic DAGs from Scientific Papers
+
+- **ID:** `2603.25293` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Shu Wan, Saketh Vishnubhatla, Iskander Kushbay, et al.
+- **abs:** https://arxiv.org/abs/2603.25293 · **PDF:** https://arxiv.org/pdf/2603.25293v1
+
+## 1760. Revealing the influence of participant failures on model quality in cross-silo Federated Learning
+
+- **ID:** `2603.25289` · **primary:** `cs.DC` · **Submitted:** 2026-03-26
+- **Authors:** Fabian Stricker, David Bermbach, Christian Zirpins
+- **abs:** https://arxiv.org/abs/2603.25289 · **PDF:** https://arxiv.org/pdf/2603.25289v1
+
+## 1761. CSI-tuples-based 3D Channel Fingerprints Construction Assisted by MultiModal Learning
+
+- **ID:** `2603.25288` · **primary:** `cs.IT` · **Submitted:** 2026-03-26
+- **Authors:** Chenjie Xie, Li You, Ruirong Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.25288 · **PDF:** https://arxiv.org/pdf/2603.25288v1
+
+## 1762. SliderQuant: Accurate Post-Training Quantization for LLMs
+
+- **ID:** `2603.25284` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Shigeng Wang, Chao Li, Yangyuxuan Kang, et al.
+- **abs:** https://arxiv.org/abs/2603.25284 · **PDF:** https://arxiv.org/pdf/2603.25284v1
+
+## 1763. A Gait Foundation Model Predicts Multi-System Health Phenotypes from 3D Skeletal Motion
+
+- **ID:** `2603.25283` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Adam Gabet, Sarah Kohn, Guy Lutsker, et al.
+- **abs:** https://arxiv.org/abs/2603.25283 · **PDF:** https://arxiv.org/pdf/2603.25283v1
+
+## 1764. Distribution and Clusters Approximations as Abstract Domains in Probabilistic Abstract Interpretation to Neural Network Analysis
+
+- **ID:** `2603.25273` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Zhuofan Zhang, Herbert Wiklicky
+- **abs:** https://arxiv.org/abs/2603.25273 · **PDF:** https://arxiv.org/pdf/2603.25273v1
+
+## 1765. When Hate Meets Facts: LLMs-in-the-Loop for Check-worthiness Detection in Hate Speech
+
+- **ID:** `2603.25269` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Nicolás Benjamín Ocampo, Tommaso Caselli, Davide Ceolin
+- **abs:** https://arxiv.org/abs/2603.25269 · **PDF:** https://arxiv.org/pdf/2603.25269v1
+
+## 1766. CRAFT: Grounded Multi-Agent Coordination Under Partial Information
+
+- **ID:** `2603.25268` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Abhijnan Nath, Hannah VanderHoeven, Nikhil Krishnaswamy
+- **abs:** https://arxiv.org/abs/2603.25268 · **PDF:** https://arxiv.org/pdf/2603.25268v1
+
+## 1767. Probabilistic Abstract Interpretation on Neural Networks via Grids Approximation
+
+- **ID:** `2603.25266` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Zhuofan Zhang, Herbert Wiklicky
+- **abs:** https://arxiv.org/abs/2603.25266 · **PDF:** https://arxiv.org/pdf/2603.25266v1
+
+## 1768. Mitigating Evasion Attacks in Fog Computing Resource Provisioning Through Proactive Hardening
+
+- **ID:** `2603.25257` · **primary:** `cs.CR` · **Submitted:** 2026-03-26
+- **Authors:** Younes Salmi, Hanna Bogucka
+- **abs:** https://arxiv.org/abs/2603.25257 · **PDF:** https://arxiv.org/pdf/2603.25257v1
+
+## 1769. Hyperspectral Trajectory Image for Multi-Month Trajectory Anomaly Detection
+
+- **ID:** `2603.25255` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Md Awsafur Rahman, Chandrakanth Gudavalli, Hardik Prajapati, et al.
+- **abs:** https://arxiv.org/abs/2603.25255 · **PDF:** https://arxiv.org/pdf/2603.25255v1
+
+## 1770. MolQuest: A Benchmark for Agentic Evaluation of Abductive Reasoning in Chemical Structure Elucidation
+
+- **ID:** `2603.25253` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Taolin Han, Shuang Wu, Jinghang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.25253 · **PDF:** https://arxiv.org/pdf/2603.25253v1
+
+## 1771. Does Explanation Correctness Matter? Linking Computational XAI Evaluation to Human Understanding
+
+- **ID:** `2603.25251` · **primary:** `cs.HC` · **Submitted:** 2026-03-26
+- **Authors:** Gregor Baer, Chao Zhang, Isel Grau, et al.
+- **abs:** https://arxiv.org/abs/2603.25251 · **PDF:** https://arxiv.org/pdf/2603.25251v1
+
+## 1772. Activation Matters: Test-time Activated Negative Labels for OOD Detection with Vision-Language Models
+
+- **ID:** `2603.25250` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Yabin Zhang, Maya Varma, Yunhe Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.25250 · **PDF:** https://arxiv.org/pdf/2603.25250v1
+
+## 1773. FEAST: Fully Connected Expressive Attention for Spatial Transcriptomics
+
+- **ID:** `2603.25247` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Taejin Jeong, Joohyeok Kim, Jinyeong Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.25247 · **PDF:** https://arxiv.org/pdf/2603.25247v1
+
+## 1774. FluxEDA: A Unified Execution Infrastructure for Stateful Agentic EDA
+
+- **ID:** `2603.25243` · **primary:** `cs.AR` · **Submitted:** 2026-03-26
+- **Authors:** Zhengrui Chen, Zixuan Song, Yu Li, et al.
+- **abs:** https://arxiv.org/abs/2603.25243 · **PDF:** https://arxiv.org/pdf/2603.25243v1
+
+## 1775. Offline Decision Transformers for Neural Combinatorial Optimization: Surpassing Heuristics on the Traveling Salesman Problem
+
+- **ID:** `2603.25241` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Hironori Ohigashi, Shinichiro Hamada
+- **abs:** https://arxiv.org/abs/2603.25241 · **PDF:** https://arxiv.org/pdf/2603.25241v1
+
+## 1776. Lingshu-Cell: A generative cellular world model for transcriptome modeling toward virtual cells
+
+- **ID:** `2603.25240` · **primary:** `q-bio.QM` · **Submitted:** 2026-03-26
+- **Authors:** Han Zhang, Guo-Hua Yuan, Chaohao Yuan, et al.
+- **abs:** https://arxiv.org/abs/2603.25240 · **PDF:** https://arxiv.org/pdf/2603.25240v1
+
+## 1777. An Image Dataset of Common Skin Diseases of Bangladesh and Benchmarking Performance with Machine Learning Models
+
+- **ID:** `2603.25229` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Sazzad Hossain, Saiful Islam, Muhammad Ibrahim, et al.
+- **abs:** https://arxiv.org/abs/2603.25229 · **PDF:** https://arxiv.org/pdf/2603.25229v1
+
+## 1778. Comparing Natural and Synthetic Structured Data: A Study of the Passive Verb Alternation in French and Italian
+
+- **ID:** `2603.25227` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Giuseppe Samo, Paola Merlo
+- **abs:** https://arxiv.org/abs/2603.25227 · **PDF:** https://arxiv.org/pdf/2603.25227v1
+
+## 1779. WebTestBench: Evaluating Computer-Use Agents towards End-to-End Automated Web Testing
+
+- **ID:** `2603.25226` · **primary:** `cs.SE` · **Submitted:** 2026-03-26
+- **Authors:** Fanheng Kong, Jingyuan Zhang, Yang Yue, et al.
+- **abs:** https://arxiv.org/abs/2603.25226 · **PDF:** https://arxiv.org/pdf/2603.25226v1
+
+## 1780. Fair regression under localized demographic parity constraints
+
+- **ID:** `2603.25224` · **primary:** `stat.ML` · **Submitted:** 2026-03-26
+- **Authors:** Arthur Charpentier, Christophe Denis, Romuald Elie, et al.
+- **abs:** https://arxiv.org/abs/2603.25224 · **PDF:** https://arxiv.org/pdf/2603.25224v1
+
+## 1781. Translation or Recitation? Calibrating Evaluation Scores for Machine Translation of Extremely Low-Resource Languages
+
+- **ID:** `2603.25222` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Danlu Chen, Ka Sing He, Jiahe Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.25222 · **PDF:** https://arxiv.org/pdf/2603.25222v1
+
+## 1782. Gap Safe Screening Rules for Fast Training of Robust Support Vector Machines under Feature Noise
+
+- **ID:** `2603.25221` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Tan-Hau Nguyen, Thu-Le Tran, Kien Trung Nguyen
+- **abs:** https://arxiv.org/abs/2603.25221 · **PDF:** https://arxiv.org/pdf/2603.25221v1
+
+## 1783. A Wireless World Model for AI-Native 6G Networks
+
+- **ID:** `2603.25216` · **primary:** `cs.NI` · **Submitted:** 2026-03-26
+- **Authors:** Ziqi Chen, Yi Ren, Yixuan Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.25216 · **PDF:** https://arxiv.org/pdf/2603.25216v1
+
+## 1784. Free-Lunch Long Video Generation via Layer-Adaptive O.O.D Correction
+
+- **ID:** `2603.25209` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Jiahao Tian, Chenxi Song, Wei Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.25209 · **PDF:** https://arxiv.org/pdf/2603.25209v1
+
+## 1785. A CDF-First Framework for Free-Form Density Estimation
+
+- **ID:** `2603.25204` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Chenglong Song, Mazharul Islam, Lin Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.25204 · **PDF:** https://arxiv.org/pdf/2603.25204v1
+
+## 1786. Probabilistic Concept Graph Reasoning for Multimodal Misinformation Detection
+
+- **ID:** `2603.25203` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Ruichao Yang, Wei Gao, Xiaobin Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.25203 · **PDF:** https://arxiv.org/pdf/2603.25203v1
+
+## 1787. SafeMath: Inference-time Safety improves Math Accuracy
+
+- **ID:** `2603.25201` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Sagnik Basu, Subhrajit Mitra, Aman Juneja, et al.
+- **abs:** https://arxiv.org/abs/2603.25201 · **PDF:** https://arxiv.org/pdf/2603.25201v1
+
+## 1788. The Competence Shadow: Theory and Bounds of AI Assistance in Safety Engineering
+
+- **ID:** `2603.25197` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Umair Siddique
+- **abs:** https://arxiv.org/abs/2603.25197 · **PDF:** https://arxiv.org/pdf/2603.25197v2
+
+## 1789. A Decade-Scale Benchmark Evaluating LLMs' Clinical Practice Guidelines Detection and Adherence in Multi-turn Conversations
+
+- **ID:** `2603.25196` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Andong Tan, Shuyu Dai, Jinglu Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.25196 · **PDF:** https://arxiv.org/pdf/2603.25196v1
+
+## 1790. A Catalog of Basque Dialectal Resources: Online Collections and Standard-to-Dialectal Adaptations
+
+- **ID:** `2603.25189` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Jaione Bengoetxea, Itziar Gonzalez-Dios, Rodrigo Agerri
+- **abs:** https://arxiv.org/abs/2603.25189 · **PDF:** https://arxiv.org/pdf/2603.25189v1
+
+## 1791. Probing the Lack of Stable Internal Beliefs in LLMs
+
+- **ID:** `2603.25187` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Yifan Luo, Kangping Xu, Yanzhen Lu, et al.
+- **abs:** https://arxiv.org/abs/2603.25187 · **PDF:** https://arxiv.org/pdf/2603.25187v1
+
+## 1792. Knowledge-Guided Retrieval-Augmented Generation for Zero-Shot Psychiatric Data: Privacy Preserving Synthetic Data Generation
+
+- **ID:** `2603.25186` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Adam Jakobsen, Sushant Gautam, Hugo Lewi Hammer, et al.
+- **abs:** https://arxiv.org/abs/2603.25186 · **PDF:** https://arxiv.org/pdf/2603.25186v1
+
+## 1793. Train at Moving Edge: Online-Verified Prompt Selection for Efficient RL Training of Large Reasoning Model
+
+- **ID:** `2603.25184` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Jiahao Wu, Ning Lu, Shengcai Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.25184 · **PDF:** https://arxiv.org/pdf/2603.25184v1
+
+## 1794. Cross-Preference Learning for Sentence-Level and Context-Aware Machine Translation
+
+- **ID:** `2603.25183` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Ying Li, Xinglin Lyu, Junhui Li, et al.
+- **abs:** https://arxiv.org/abs/2603.25183 · **PDF:** https://arxiv.org/pdf/2603.25183v1
+
+## 1795. Bilingual Text-to-Motion Generation: A New Benchmark and Baselines
+
+- **ID:** `2603.25178` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Wanjiang Weng, Xiaofeng Tan, Xiangbo Shu, et al.
+- **abs:** https://arxiv.org/abs/2603.25178 · **PDF:** https://arxiv.org/pdf/2603.25178v1
+
+## 1796. Prompt Attack Detection with LLM-as-a-Judge and Mixture-of-Models
+
+- **ID:** `2603.25176` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Hieu Xuan Le, Benjamin Goh, Quy Anh Tang
+- **abs:** https://arxiv.org/abs/2603.25176 · **PDF:** https://arxiv.org/pdf/2603.25176v1
+
+## 1797. Knowledge-Guided Adversarial Training for Infrared Object Detection via Thermal Radiation Modeling
+
+- **ID:** `2603.25170` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Shiji Zhao, Shukun Xiong, Maoxun Yuan, et al.
+- **abs:** https://arxiv.org/abs/2603.25170 · **PDF:** https://arxiv.org/pdf/2603.25170v1
+
+## 1798. To Write or to Automate Linguistic Prompts, That Is the Question
+
+- **ID:** `2603.25169` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Marina Sánchez-Torrón, Daria Akselrod, Jason Rauchwerk
+- **abs:** https://arxiv.org/abs/2603.25169 · **PDF:** https://arxiv.org/pdf/2603.25169v1
+
+## 1799. Empowering Epidemic Response: The Role of Reinforcement Learning in Infectious Disease Control
+
+- **ID:** `2603.25771` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Mutong Liu, Yang Liu, Jiming Liu
+- **abs:** https://arxiv.org/abs/2603.25771 · **PDF:** https://arxiv.org/pdf/2603.25771v1
+
+## 1800. PIDP-Attack: Combining Prompt Injection with Database Poisoning Attacks on Retrieval-Augmented Generation Systems
+
+- **ID:** `2603.25164` · **primary:** `cs.CR` · **Submitted:** 2026-03-26
+- **Authors:** Haozhen Wang, Haoyue Liu, Jionghao Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.25164 · **PDF:** https://arxiv.org/pdf/2603.25164v1
+
+## 1801. Trace2Skill: Distill Trajectory-Local Lessons into Transferable Agent Skills
+
+- **ID:** `2603.25158` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Jingwei Ni, Yihao Liu, Xinpeng Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.25158 · **PDF:** https://arxiv.org/pdf/2603.25158v3
+
+## 1802. Vision Hopfield Memory Networks
+
+- **ID:** `2603.25157` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Jianfeng Wang, Amine M'Charrak, Luk Koska, et al.
+- **abs:** https://arxiv.org/abs/2603.25157 · **PDF:** https://arxiv.org/pdf/2603.25157v1
+
+## 1803. Photon: Speedup Volume Understanding with Efficient Multimodal Large Language Models
+
+- **ID:** `2603.25155` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Chengyu Fang, Heng Guo, Zheng Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.25155 · **PDF:** https://arxiv.org/pdf/2603.25155v1
+
+## 1804. UniAI-GraphRAG: Synergizing Ontology-Guided Extraction, Multi-Dimensional Clustering, and Dual-Channel Fusion for Robust Multi-Hop Reasoning
+
+- **ID:** `2603.25152` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Jie Wang, Honghua Huang, Xi Ge, et al.
+- **abs:** https://arxiv.org/abs/2603.25152 · **PDF:** https://arxiv.org/pdf/2603.25152v2
+
+## 1805. Goodness-of-pronunciation without phoneme time alignment
+
+- **ID:** `2603.25150` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Jeremy H. M. Wong, Nancy F. Chen
+- **abs:** https://arxiv.org/abs/2603.25150 · **PDF:** https://arxiv.org/pdf/2603.25150v1
+
+## 1806. Factors Influencing the Quality of AI-Generated Code: A Synthesis of Empirical Evidence
+
+- **ID:** `2603.25146` · **primary:** `cs.SE` · **Submitted:** 2026-03-26
+- **Authors:** Vehid Geruslu, Zulfiyya Aliyeva, Eray Tüzün
+- **abs:** https://arxiv.org/abs/2603.25146 · **PDF:** https://arxiv.org/pdf/2603.25146v1
+
+## 1807. Learning to Rank Caption Chains for Video-Text Alignment
+
+- **ID:** `2603.25145` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Ansel Blume, Burak Uzkent, Shalini Chaudhuri, et al.
+- **abs:** https://arxiv.org/abs/2603.25145 · **PDF:** https://arxiv.org/pdf/2603.25145v1
+
+## 1808. ReCUBE: Evaluating Repository-Level Context Utilization in Code Generation
+
+- **ID:** `2603.25770` · **primary:** `cs.SE` · **Submitted:** 2026-03-26
+- **Authors:** Jiseung Hong, Benjamin G. Ascoli, Jinho D. Choi
+- **abs:** https://arxiv.org/abs/2603.25770 · **PDF:** https://arxiv.org/pdf/2603.25770v1
+
+## 1809. FD$^2$: A Dedicated Framework for Fine-Grained Dataset Distillation
+
+- **ID:** `2603.25144` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Hongxu Ma, Guang Li, Shijie Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.25144 · **PDF:** https://arxiv.org/pdf/2603.25144v1
+
+## 1810. IncreRTL: Traceability-Guided Incremental RTL Generation under Requirement Evolution
+
+- **ID:** `2603.25769` · **primary:** `cs.SE` · **Submitted:** 2026-03-26
+- **Authors:** Luanrong Chen, Renzhi Chen, Xinyu Li, et al.
+- **abs:** https://arxiv.org/abs/2603.25769 · **PDF:** https://arxiv.org/pdf/2603.25769v1
+
+## 1811. SAVe: Self-Supervised Audio-visual Deepfake Detection Exploiting Visual Artifacts and Audio-visual Misalignment
+
+- **ID:** `2603.25140` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Sahibzada Adil Shahzad, Ammarah Hashmi, Junichi Yamagishi, et al.
+- **abs:** https://arxiv.org/abs/2603.25140 · **PDF:** https://arxiv.org/pdf/2603.25140v1
+
+## 1812. Reinforcement learning for quantum processes with memory
+
+- **ID:** `2603.25138` · **primary:** `quant-ph` · **Submitted:** 2026-03-26
+- **Authors:** Josep Lumbreras, Ruo Cheng Huang, Yanglin Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.25138 · **PDF:** https://arxiv.org/pdf/2603.25138v1
+
+## 1813. RubricEval: A Rubric-Level Meta-Evaluation Benchmark for LLM Judges in Instruction Following
+
+- **ID:** `2603.25133` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Tianjun Pan, Xuan Lin, Wenyan Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.25133 · **PDF:** https://arxiv.org/pdf/2603.25133v1
+
+## 1814. Robust Principal Component Completion
+
+- **ID:** `2603.25132` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Yinjian Wang, Wei Li, Yuanyuan Gui, et al.
+- **abs:** https://arxiv.org/abs/2603.25132 · **PDF:** https://arxiv.org/pdf/2603.25132v1
+
+## 1815. MCLMR: A Model-Agnostic Causal Learning Framework for Multi-Behavior Recommendation
+
+- **ID:** `2603.25126` · **primary:** `cs.IR` · **Submitted:** 2026-03-26
+- **Authors:** Ranxu Zhang, Junjie Meng, Ying Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.25126 · **PDF:** https://arxiv.org/pdf/2603.25126v1
+
+## 1816. The Language of Touch: Translating Vibrations into Text with Dual-Branch Learning
+
+- **ID:** `2603.26804` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Jin Chen, Yifeng Lin, Chao Zeng, et al.
+- **abs:** https://arxiv.org/abs/2603.26804 · **PDF:** https://arxiv.org/pdf/2603.26804v1
+
+## 1817. When Sensing Varies with Contexts: Context-as-Transform for Tactile Few-Shot Class-Incremental Learning
+
+- **ID:** `2603.25115` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Yifeng Lin, Aiping Huang, Wenxi Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.25115 · **PDF:** https://arxiv.org/pdf/2603.25115v1
+
+## 1818. A Comparative Investigation of Thermodynamic Structure-Informed Neural Networks
+
+- **ID:** `2603.26803` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Guojie Li, Liu Hong
+- **abs:** https://arxiv.org/abs/2603.26803 · **PDF:** https://arxiv.org/pdf/2603.26803v1
+
+## 1819. Do LLMs Know What They Know? Measuring Metacognitive Efficiency with Signal Detection Theory
+
+- **ID:** `2603.25112` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Jon-Paul Cacioli
+- **abs:** https://arxiv.org/abs/2603.25112 · **PDF:** https://arxiv.org/pdf/2603.25112v1
+
+## 1820. SEVerA: Verified Synthesis of Self-Evolving Agents
+
+- **ID:** `2603.25111` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Debangshu Banerjee, Changming Xu, Gagandeep Singh
+- **abs:** https://arxiv.org/abs/2603.25111 · **PDF:** https://arxiv.org/pdf/2603.25111v1
+
+## 1821. MoireMix: A Formula-Based Data Augmentation for Improving Image Classification Robustness
+
+- **ID:** `2603.25109` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Yuto Matsuo, Yoshihiro Fukuhara, Yuki M. Asano, et al.
+- **abs:** https://arxiv.org/abs/2603.25109 · **PDF:** https://arxiv.org/pdf/2603.25109v1
+
+## 1822. OMIND: Framework for Knowledge Grounded Finetuning and Multi-Turn Dialogue Benchmark for Mental Health LLMs
+
+- **ID:** `2603.25105` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Suraj Racha, Prashant Harish Joshi, Utkarsh Maurya, et al.
+- **abs:** https://arxiv.org/abs/2603.25105 · **PDF:** https://arxiv.org/pdf/2603.25105v1
+
+## 1823. Layer-Specific Lipschitz Modulation for Fault-Tolerant Multimodal Representation Learning
+
+- **ID:** `2603.25103` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Diyar Altinses, Andreas Schwung
+- **abs:** https://arxiv.org/abs/2603.25103 · **PDF:** https://arxiv.org/pdf/2603.25103v1
+
+## 1824. UCAgent: An End-to-End Agent for Block-Level Functional Verification
+
+- **ID:** `2603.25768` · **primary:** `cs.SE` · **Submitted:** 2026-03-26
+- **Authors:** Junyue Wang, Zhicheng Yao, Yan Pi, et al.
+- **abs:** https://arxiv.org/abs/2603.25768 · **PDF:** https://arxiv.org/pdf/2603.25768v1
+
+## 1825. Unlocking Strong Supervision: A Data-Centric Study of General-Purpose Audio Pre-Training Methods
+
+- **ID:** `2603.25767` · **primary:** `cs.SD` · **Submitted:** 2026-03-26
+- **Authors:** Xuanru Zhou, Yiwen Shao, Wei-Cheng Tseng, et al.
+- **abs:** https://arxiv.org/abs/2603.25767 · **PDF:** https://arxiv.org/pdf/2603.25767v1
+
+## 1826. From Logic Monopoly to Social Contract: Separation of Power and the Institutional Foundations for Autonomous Agent Economies
+
+- **ID:** `2603.25100` · **primary:** `cs.MA` · **Submitted:** 2026-03-26
+- **Authors:** Anbang Ruan
+- **abs:** https://arxiv.org/abs/2603.25100 · **PDF:** https://arxiv.org/pdf/2603.25100v1
+
+## 1827. Large Language Models as Optimization Controllers: Adaptive Continuation for SIMP Topology Optimization
+
+- **ID:** `2603.25099` · **primary:** `cs.CE` · **Submitted:** 2026-03-26
+- **Authors:** Shaoliang Yang, Jun Wang, Yunsheng Wang
+- **abs:** https://arxiv.org/abs/2603.25099 · **PDF:** https://arxiv.org/pdf/2603.25099v1
+
+## 1828. ElephantBroker: A Knowledge-Grounded Cognitive Runtime for Trustworthy AI Agents
+
+- **ID:** `2603.25097` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Cristian Lupascu, Alexandru Lupascu
+- **abs:** https://arxiv.org/abs/2603.25097 · **PDF:** https://arxiv.org/pdf/2603.25097v1
+
+## 1829. Process-Aware AI for Rainfall-Runoff Modeling: A Mass-Conserving Neural Framework with Hydrological Process Constraints
+
+- **ID:** `2603.25093` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Mohammad A. Farmani, Hoshin V. Gupta, Ali Behrangi, et al.
+- **abs:** https://arxiv.org/abs/2603.25093 · **PDF:** https://arxiv.org/pdf/2603.25093v1
+
+## 1830. Pixelis: Reasoning in Pixels, from Seeing to Acting
+
+- **ID:** `2603.25091` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Yunpeng Zhou
+- **abs:** https://arxiv.org/abs/2603.25091 · **PDF:** https://arxiv.org/pdf/2603.25091v1
+
+## 1831. ETA-VLA: Efficient Token Adaptation via Temporal Fusion and Intra-LLM Sparsification for Vision-Language-Action Models
+
+- **ID:** `2603.25766` · **primary:** `cs.RO` · **Submitted:** 2026-03-26
+- **Authors:** Yiru Wang, Anqing Jiang, Shuo Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.25766 · **PDF:** https://arxiv.org/pdf/2603.25766v1
+
+## 1832. Learning domain-invariant features through channel-level sparsification for Out-Of Distribution Generalization
+
+- **ID:** `2603.25083` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Haoran Pei, Yuguang Yang, Kexin Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.25083 · **PDF:** https://arxiv.org/pdf/2603.25083v1
+
+## 1833. Sparse Visual Thought Circuits in Vision-Language Models
+
+- **ID:** `2603.25075` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Yunpeng Zhou
+- **abs:** https://arxiv.org/abs/2603.25075 · **PDF:** https://arxiv.org/pdf/2603.25075v1
+
+## 1834. An Explainable Ensemble Learning Framework for Crop Classification with Optimized Feature Pyramids and Deep Networks
+
+- **ID:** `2603.25070` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Syed Rayhan Masud, SK Muktadir Hossain, Md. Ridoy Sarkar, et al.
+- **abs:** https://arxiv.org/abs/2603.25070 · **PDF:** https://arxiv.org/pdf/2603.25070v1
+
+## 1835. Ultra-fast Traffic Nowcasting and Control via Differentiable Agent-based Simulation
+
+- **ID:** `2603.25068` · **primary:** `cs.MA` · **Submitted:** 2026-03-26
+- **Authors:** Fumiyasu Makinoshima, Yuya Yamaguchi, Eigo Segawa, et al.
+- **abs:** https://arxiv.org/abs/2603.25068 · **PDF:** https://arxiv.org/pdf/2603.25068v1
+
+## 1836. TopoPilot: Reliable Conversational Workflow Automation for Topological Data Analysis and Visualization
+
+- **ID:** `2603.25063` · **primary:** `cs.HC` · **Submitted:** 2026-03-26
+- **Authors:** Nathaniel Gorski, Shusen Liu, Bei Wang
+- **abs:** https://arxiv.org/abs/2603.25063 · **PDF:** https://arxiv.org/pdf/2603.25063v1
+
+## 1837. SIGMA: Structure-Invariant Generative Molecular Alignment for Chemical Language Models via Autoregressive Contrastive Learning
+
+- **ID:** `2603.25062` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Xinyu Wang, Fei Dou, Jinbo Bi, et al.
+- **abs:** https://arxiv.org/abs/2603.25062 · **PDF:** https://arxiv.org/pdf/2603.25062v1
+
+## 1838. The System Prompt Is the Attack Surface: How LLM Agent Configuration Shapes Security and Creates Exploitable Vulnerabilities
+
+- **ID:** `2603.25056` · **primary:** `cs.CR` · **Submitted:** 2026-03-26
+- **Authors:** Ron Litvak
+- **abs:** https://arxiv.org/abs/2603.25056 · **PDF:** https://arxiv.org/pdf/2603.25056v1
+
+## 1839. Closing the Confidence-Faithfulness Gap in Large Language Models
+
+- **ID:** `2603.25052` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Miranda Muqing Miao, Lyle Ungar
+- **abs:** https://arxiv.org/abs/2603.25052 · **PDF:** https://arxiv.org/pdf/2603.25052v2
+
+## 1840. Approaches to Analysing Historical Newspapers Using LLMs
+
+- **ID:** `2603.25051` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Filip Dobranić, Tina Munda, Oliver Pejić, et al.
+- **abs:** https://arxiv.org/abs/2603.25051 · **PDF:** https://arxiv.org/pdf/2603.25051v2
+
+## 1841. MP-MoE: Matrix Profile-Guided Mixture of Experts for Precipitation Forecasting
+
+- **ID:** `2603.25046` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Huyen Ngoc Tran, Dung Trung Tran, Hong Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.25046 · **PDF:** https://arxiv.org/pdf/2603.25046v1
+
+## 1842. Intern-S1-Pro: Scientific Multimodal Foundation Model at Trillion Scale
+
+- **ID:** `2603.25040` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Yicheng Zou, Dongsheng Zhu, Lin Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.25040 · **PDF:** https://arxiv.org/pdf/2603.25040v2
+
+## 1843. Uncertainty Quantification for Quantum Computing
+
+- **ID:** `2603.25039` · **primary:** `quant-ph` · **Submitted:** 2026-03-26
+- **Authors:** Ryan Bennink, Olena Burkovska, Konstantin Pieper, et al.
+- **abs:** https://arxiv.org/abs/2603.25039 · **PDF:** https://arxiv.org/pdf/2603.25039v1
+
+## 1844. Mechanistically Interpreting Compression in Vision-Language Models
+
+- **ID:** `2603.25035` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Veeraraju Elluru, Arth Singh, Roberto Aguero, et al.
+- **abs:** https://arxiv.org/abs/2603.25035 · **PDF:** https://arxiv.org/pdf/2603.25035v1
+
+## 1845. Epistemic Compression: The Case for Deliberate Ignorance in High-Stakes AI
+
+- **ID:** `2603.25033` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Steffen Lukas
+- **abs:** https://arxiv.org/abs/2603.25033 · **PDF:** https://arxiv.org/pdf/2603.25033v1
+
+## 1846. From Stateless to Situated: Building a Psychological World for LLM-Based Emotional Support
+
+- **ID:** `2603.25031` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Boning Zhao, Clover Hu, Xinnuo Li
+- **abs:** https://arxiv.org/abs/2603.25031 · **PDF:** https://arxiv.org/pdf/2603.25031v1
+
+## 1847. Optimal High-Probability Regret for Online Convex Optimization with Two-Point Bandit Feedback
+
+- **ID:** `2603.25029` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Haishan Ye
+- **abs:** https://arxiv.org/abs/2603.25029 · **PDF:** https://arxiv.org/pdf/2603.25029v2
+
+## 1848. System-Anchored Knee Estimation for Low-Cost Context Window Selection in PDE Forecasting
+
+- **ID:** `2603.25025` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Wenshuo Wang, Fan Zhang
+- **abs:** https://arxiv.org/abs/2603.25025 · **PDF:** https://arxiv.org/pdf/2603.25025v1
+
+## 1849. Improving Infinitely Deep Bayesian Neural Networks with Nesterov's Accelerated Gradient Method
+
+- **ID:** `2603.25024` · **primary:** `stat.ML` · **Submitted:** 2026-03-26
+- **Authors:** Chenxu Yu, Wenqi Fang
+- **abs:** https://arxiv.org/abs/2603.25024 · **PDF:** https://arxiv.org/pdf/2603.25024v1
+
+## 1850. Consistency Amplifies: How Behavioral Variance Shapes Agent Accuracy
+
+- **ID:** `2603.25764` · **primary:** `cs.SE` · **Submitted:** 2026-03-26
+- **Authors:** Aman Mehta
+- **abs:** https://arxiv.org/abs/2603.25764 · **PDF:** https://arxiv.org/pdf/2603.25764v1
+
+## 1851. A Public Theory of Distillation Resistance via Constraint-Coupled Reasoning Architectures
+
+- **ID:** `2603.25022` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Peng Wei, Wesley Shu
+- **abs:** https://arxiv.org/abs/2603.25022 · **PDF:** https://arxiv.org/pdf/2603.25022v1
+
+## 1852. Discrete Causal Representation Learning
+
+- **ID:** `2603.25017` · **primary:** `stat.ME` · **Submitted:** 2026-03-26
+- **Authors:** Wenjin Zhang, Yixin Wang, Yuqi Gu
+- **abs:** https://arxiv.org/abs/2603.25017 · **PDF:** https://arxiv.org/pdf/2603.25017v1
+
+## 1853. Imperative Interference: Social Register Shapes Instruction Topology in Large Language Models
+
+- **ID:** `2603.25015` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Tony Mason
+- **abs:** https://arxiv.org/abs/2603.25015 · **PDF:** https://arxiv.org/pdf/2603.25015v1
+
+## 1854. CLPIPS: A Personalized Metric for AI-Generated Image Similarity
+
+- **ID:** `2604.01234` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Khoi Trinh, Jay Rothenberger, Scott Seidenberger, et al.
+- **abs:** https://arxiv.org/abs/2604.01234 · **PDF:** https://arxiv.org/pdf/2604.01234v1
+
+## 1855. A Systematic Empirical Study of Grokking: Depth, Architecture, Activation, and Regularization
+
+- **ID:** `2603.25009` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Shalima Binta Manir, Anamika Paul Rupa
+- **abs:** https://arxiv.org/abs/2603.25009 · **PDF:** https://arxiv.org/pdf/2603.25009v1
+
+## 1856. Few TensoRF: Enhance the Few-shot on Tensorial Radiance Fields
+
+- **ID:** `2603.25008` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Thanh-Hai Le, Hoang-Hau Tran, Trong-Nghia Vu
+- **abs:** https://arxiv.org/abs/2603.25008 · **PDF:** https://arxiv.org/pdf/2603.25008v3
+
+## 1857. Improving Fine-Grained Rice Leaf Disease Detection via Angular-Compactness Dual Loss Learning
+
+- **ID:** `2603.25006` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Md. Rokon Mia, Rakib Hossain Sajib, Abdullah Al Noman, et al.
+- **abs:** https://arxiv.org/abs/2603.25006 · **PDF:** https://arxiv.org/pdf/2603.25006v1
+
+## 1858. Rethinking Failure Attribution in Multi-Agent Systems: A Multi-Perspective Benchmark and Evaluation
+
+- **ID:** `2603.25001` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Yeonjun In, Mehrab Tanjim, Jayakumar Subramanian, et al.
+- **abs:** https://arxiv.org/abs/2603.25001 · **PDF:** https://arxiv.org/pdf/2603.25001v1
+
+## 1859. CANGuard: A Spatio-Temporal CNN-GRU-Attention Hybrid Architecture for Intrusion Detection in In-Vehicle CAN Networks
+
+- **ID:** `2603.25763` · **primary:** `cs.CR` · **Submitted:** 2026-03-26
+- **Authors:** Rakib Hossain Sajib, Md. Rokon Mia, Prodip Kumar Sarker, et al.
+- **abs:** https://arxiv.org/abs/2603.25763 · **PDF:** https://arxiv.org/pdf/2603.25763v1
+
+## 1860. Efficient Detection of Bad Benchmark Items with Novel Scalability Coefficients
+
+- **ID:** `2603.24999` · **primary:** `stat.AP` · **Submitted:** 2026-03-26
+- **Authors:** Michael Hardy, Joshua Gilbert, Benjamin Domingue
+- **abs:** https://arxiv.org/abs/2603.24999 · **PDF:** https://arxiv.org/pdf/2603.24999v2
+
+## 1861. Sparse-by-Design Cross-Modality Prediction: L0-Gated Representations for Reliable and Efficient Learning
+
+- **ID:** `2603.26801` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Filippo Cenacchi
+- **abs:** https://arxiv.org/abs/2603.26801 · **PDF:** https://arxiv.org/pdf/2603.26801v1
+
+## 1862. Learning Rollout from Sampling:An R1-Style Tokenized Traffic Simulation Model
+
+- **ID:** `2603.24989` · **primary:** `cs.RO` · **Submitted:** 2026-03-26
+- **Authors:** Ziyan Wang, Peng Chen, Ding Li, et al.
+- **abs:** https://arxiv.org/abs/2603.24989 · **PDF:** https://arxiv.org/pdf/2603.24989v1
+
+## 1863. Rethinking Health Agents: From Siloed AI to Collaborative Decision Mediators
+
+- **ID:** `2603.24986` · **primary:** `cs.HC` · **Submitted:** 2026-03-26
+- **Authors:** Ray-Yuan Chung, Xuhai Xu, Ari Pollack
+- **abs:** https://arxiv.org/abs/2603.24986 · **PDF:** https://arxiv.org/pdf/2603.24986v1
+
+## 1864. Exons-Detect: Identifying and Amplifying Exonic Tokens via Hidden-State Discrepancy for Robust AI-Generated Text Detection
+
+- **ID:** `2603.24981` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Xiaowei Zhu, Yubing Ren, Fang Fang, et al.
+- **abs:** https://arxiv.org/abs/2603.24981 · **PDF:** https://arxiv.org/pdf/2603.24981v1
+
+## 1865. LLM-Driven Reasoning for Constraint-Aware Feature Selection in Industrial Systems
+
+- **ID:** `2603.24979` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Yuhang Zhou, Zhuokai Zhao, Ke Li, et al.
+- **abs:** https://arxiv.org/abs/2603.24979 · **PDF:** https://arxiv.org/pdf/2603.24979v1
+
+## 1866. The Value of Information in Resource-Constrained Pricing
+
+- **ID:** `2603.24974` · **primary:** `math.OC` · **Submitted:** 2026-03-26
+- **Authors:** Ruicheng Ao, Jiashuo Jiang, David Simchi-Levi
+- **abs:** https://arxiv.org/abs/2603.24974 · **PDF:** https://arxiv.org/pdf/2603.24974v1
+
+## 1867. DSO: Dual-Scale Neural Operators for Stable Long-term Fluid Dynamics Forecasting
+
+- **ID:** `2603.26800` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Huanshuo Dong, Hao Wu, Hong Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.26800 · **PDF:** https://arxiv.org/pdf/2603.26800v1
+
+## 1868. Subject-Specific Low-Field MRI Synthesis via a Neural Operator
+
+- **ID:** `2603.24968` · **primary:** `eess.IV` · **Submitted:** 2026-03-26
+- **Authors:** Ziqi Gao, Nicha Dvornek, Xiaoran Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.24968 · **PDF:** https://arxiv.org/pdf/2603.24968v1
+
+## 1869. The Anatomy of Uncertainty in LLMs
+
+- **ID:** `2603.24967` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Aditya Taparia, Ransalu Senanayake, Kowshik Thopalli, et al.
+- **abs:** https://arxiv.org/abs/2603.24967 · **PDF:** https://arxiv.org/pdf/2603.24967v1
+
+## 1870. Self-Corrected Image Generation with Explainable Latent Rewards
+
+- **ID:** `2603.24965` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Yinyi Luo, Hrishikesh Gokhale, Marios Savvides, et al.
+- **abs:** https://arxiv.org/abs/2603.24965 · **PDF:** https://arxiv.org/pdf/2603.24965v1
+
+## 1871. Design Once, Deploy at Scale: Template-Driven ML Development for Large Model Ecosystems
+
+- **ID:** `2603.24963` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Jiang Liu, John Martabano Landy, Yao Xuan, et al.
+- **abs:** https://arxiv.org/abs/2603.24963 · **PDF:** https://arxiv.org/pdf/2603.24963v2
+
+## 1872. Can MLLMs Read Students' Minds? Unpacking Multimodal Error Analysis in Handwritten Math
+
+- **ID:** `2603.24961` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Dingjie Song, Tianlong Xu, Yi-Fan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.24961 · **PDF:** https://arxiv.org/pdf/2603.24961v1
+
+## 1873. Toward domain-specific machine translation and quality estimation systems
+
+- **ID:** `2603.24955` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Javad Pourmostafa Roshan Sharami
+- **abs:** https://arxiv.org/abs/2603.24955 · **PDF:** https://arxiv.org/pdf/2603.24955v1
+
+## 1874. Shopping with a Platform AI Assistant: Who Adopts, When in the Journey, and What For
+
+- **ID:** `2603.24947` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Se Yan, Han Zhong, Zemin, et al.
+- **abs:** https://arxiv.org/abs/2603.24947 · **PDF:** https://arxiv.org/pdf/2603.24947v1
+
+## 1875. MobileDev-Bench: A Comprehensive Benchmark for Evaluating Language Models on Mobile Application Development
+
+- **ID:** `2603.24946` · **primary:** `cs.SE` · **Submitted:** 2026-03-26
+- **Authors:** Moshood A. Fakorede, Krishna Upadhyay, A. B. Siddique, et al.
+- **abs:** https://arxiv.org/abs/2603.24946 · **PDF:** https://arxiv.org/pdf/2603.24946v1
+
+## 1876. FinMCP-Bench: Benchmarking LLM Agents for Real-World Financial Tool Use under the Model Context Protocol
+
+- **ID:** `2603.24943` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Jie Zhu, Yimin Tian, Boyang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.24943 · **PDF:** https://arxiv.org/pdf/2603.24943v1
+
+## 1877. Beyond Attention Magnitude: Leveraging Inter-layer Rank Consistency for Efficient Vision-Language-Action Models
+
+- **ID:** `2603.24941` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Peiju Liu, Jinming Liu, Xipeng Qiu, et al.
+- **abs:** https://arxiv.org/abs/2603.24941 · **PDF:** https://arxiv.org/pdf/2603.24941v1
+
+## 1878. Evaluating adaptive and generative AI-based feedback and recommendations in a knowledge-graph-integrated programming learning system
+
+- **ID:** `2603.24940` · **primary:** `cs.PL` · **Submitted:** 2026-03-26
+- **Authors:** Lalita Na Nongkhai, Jingyun Wang, Adam Wynn, et al.
+- **abs:** https://arxiv.org/abs/2603.24940 · **PDF:** https://arxiv.org/pdf/2603.24940v1
+
+## 1879. TIGFlow-GRPO: Trajectory Forecasting via Interaction-Aware Flow Matching and Reward-Driven Optimization
+
+- **ID:** `2603.24936` · **primary:** `cs.CV` · **Submitted:** 2026-03-26
+- **Authors:** Xuepeng Jing, Wenhuan Lu, Hao Meng, et al.
+- **abs:** https://arxiv.org/abs/2603.24936 · **PDF:** https://arxiv.org/pdf/2603.24936v1
+
+## 1880. CVA: Context-aware Video-text Alignment for Video Temporal Grounding
+
+- **ID:** `2603.24934` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Sungho Moon, Seunghun Lee, Jiwan Seo, et al.
+- **abs:** https://arxiv.org/abs/2603.24934 · **PDF:** https://arxiv.org/pdf/2603.24934v1
+
+## 1881. Decoding Market Emotions in Cryptocurrency Tweets via Predictive Statement Classification with Machine Learning and Transformers
+
+- **ID:** `2603.24933` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Moein Shahiki Tash, Zahra Ahani, Mohim Tash, et al.
+- **abs:** https://arxiv.org/abs/2603.24933 · **PDF:** https://arxiv.org/pdf/2603.24933v1
+
+## 1882. LogitScope: A Framework for Analyzing LLM Uncertainty Through Information Metrics
+
+- **ID:** `2603.24929` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** Farhan Ahmed, Yuya Jeremy Ong, Chad DeLuca
+- **abs:** https://arxiv.org/abs/2603.24929 · **PDF:** https://arxiv.org/pdf/2603.24929v1
+
+## 1883. GraphER: An Efficient Graph-Based Enrichment and Reranking Method for Retrieval-Augmented Generation
+
+- **ID:** `2603.24925` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Ruizhong Miao, Yuying Wang, Rongguang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.24925 · **PDF:** https://arxiv.org/pdf/2603.24925v1
+
+## 1884. Estimating near-verbatim extraction risk in language models with decoding-constrained beam search
+
+- **ID:** `2603.24917` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** A. Feder Cooper, Mark A. Lemley, Christopher De Sa, et al.
+- **abs:** https://arxiv.org/abs/2603.24917 · **PDF:** https://arxiv.org/pdf/2603.24917v1
+
+## 1885. Once-for-All Channel Mixers (HYPERTINYPW): Generative Compression for TinyML
+
+- **ID:** `2603.24916` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Yassien Shaalan
+- **abs:** https://arxiv.org/abs/2603.24916 · **PDF:** https://arxiv.org/pdf/2603.24916v1
+
+## 1886. Shaping the Future of Mathematics in the Age of AI
+
+- **ID:** `2603.24914` · **primary:** `math.HO` · **Submitted:** 2026-03-26
+- **Authors:** Johan Commelin, Mateja Jamnik, Rodrigo Ochigame, et al.
+- **abs:** https://arxiv.org/abs/2603.24914 · **PDF:** https://arxiv.org/pdf/2603.24914v1
+
+## 1887. Associative Memory using Attribute-Specific Neuron Groups-2: Learning and Sequential Associative Recall between Cue Neurons for different Cue Balls
+
+- **ID:** `2603.24910` · **primary:** `cs.NE` · **Submitted:** 2026-03-26
+- **Authors:** Hiroshi Inazawa
+- **abs:** https://arxiv.org/abs/2603.24910 · **PDF:** https://arxiv.org/pdf/2603.24910v1
+
+## 1888. Gaussian Joint Embeddings For Self-Supervised Representation Learning
+
+- **ID:** `2603.26799` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Yongchao Huang
+- **abs:** https://arxiv.org/abs/2603.26799 · **PDF:** https://arxiv.org/pdf/2603.26799v1
+
+## 1889. Integrated Multi-Drone Task Allocation, Sequencing, and Optimal Trajectory Generation in Obstacle-Rich 3D Environments
+
+- **ID:** `2603.24908` · **primary:** `cs.RO` · **Submitted:** 2026-03-26
+- **Authors:** Yunes Alqudsi, Murat Makaraci
+- **abs:** https://arxiv.org/abs/2603.24908 · **PDF:** https://arxiv.org/pdf/2603.24908v1
+
+## 1890. Explaining, Verifying, and Aligning Semantic Hierarchies in Vision-Language Model Embeddings
+
+- **ID:** `2603.26798` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Gesina Schwalbe, Mert Keser, Moritz Bayerkuhnlein, et al.
+- **abs:** https://arxiv.org/abs/2603.26798 · **PDF:** https://arxiv.org/pdf/2603.26798v1
+
+## 1891. On the Foundations of Trustworthy Artificial Intelligence
+
+- **ID:** `2603.24904` · **primary:** `cs.AI` · **Submitted:** 2026-03-26
+- **Authors:** TJ Dunham
+- **abs:** https://arxiv.org/abs/2603.24904 · **PDF:** https://arxiv.org/pdf/2603.24904v1
+
+## 1892. MemGuard-Alpha: Detecting and Filtering Memorization-Contaminated Signals in LLM-Based Financial Forecasting via Membership Inference and Cross-Model Disagreement
+
+- **ID:** `2603.26797` · **primary:** `cs.LG` · **Submitted:** 2026-03-26
+- **Authors:** Anisha Roy, Dip Roy
+- **abs:** https://arxiv.org/abs/2603.26797 · **PDF:** https://arxiv.org/pdf/2603.26797v1
+
+## 1893. Sovereign AI at the Front Door of Care: A Physically Unidirectional Architecture for Secure Clinical Intelligence
+
+- **ID:** `2603.24898` · **primary:** `cs.CR` · **Submitted:** 2026-03-26
+- **Authors:** Vasu Srinivasan, Dhriti Vasu
+- **abs:** https://arxiv.org/abs/2603.24898 · **PDF:** https://arxiv.org/pdf/2603.24898v1
+
+## 1894. LogSigma at SemEval-2026 Task 3: Uncertainty-Weighted Multitask Learning for Dimensional Aspect-Based Sentiment Analysis
+
+- **ID:** `2603.24896` · **primary:** `cs.CL` · **Submitted:** 2026-03-26
+- **Authors:** Baraa Hikal, Jonas Becker, Bela Gipp
+- **abs:** https://arxiv.org/abs/2603.24896 · **PDF:** https://arxiv.org/pdf/2603.24896v1
+
+## 1895. Surrogates, Spikes, and Sparsity: Performance Analysis and Characterization of SNN Hyperparameters on Hardware
+
+- **ID:** `2603.24891` · **primary:** `cs.AR` · **Submitted:** 2026-03-26
+- **Authors:** Ilkin Aliyev, Jesus Lopez, Tosiron Adegbija
+- **abs:** https://arxiv.org/abs/2603.24891 · **PDF:** https://arxiv.org/pdf/2603.24891v1
+
+## 1896. Learning to Staff: Offline Reinforcement Learning and Fine-Tuned LLMs for Warehouse Staffing Optimization
+
+- **ID:** `2603.24883` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Kalle Kujanpää, Yuying Zhu, Kristina Klinkner, et al.
+- **abs:** https://arxiv.org/abs/2603.24883 · **PDF:** https://arxiv.org/pdf/2603.24883v1
+
+## 1897. More Than "Means to an End": Supporting Reasoning with Transparently Designed AI Data Science Processes
+
+- **ID:** `2603.24877` · **primary:** `cs.HC` · **Submitted:** 2026-03-25
+- **Authors:** Venkatesh Sivaraman, Patrick Vossler, Adam Perer, et al.
+- **abs:** https://arxiv.org/abs/2603.24877 · **PDF:** https://arxiv.org/pdf/2603.24877v1
+
+## 1898. How Far Are Vision-Language Models from Constructing the Real World? A Benchmark for Physical Generative Reasoning
+
+- **ID:** `2603.24866` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Luyu Yang, Yutong Dai, An Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.24866 · **PDF:** https://arxiv.org/pdf/2603.24866v1
+
+## 1899. AI Security in the Foundation Model Era: A Comprehensive Survey from a Unified Perspective
+
+- **ID:** `2603.24857` · **primary:** `cs.CR` · **Submitted:** 2026-03-25
+- **Authors:** Zhenyi Wang, Siyu Luan
+- **abs:** https://arxiv.org/abs/2603.24857 · **PDF:** https://arxiv.org/pdf/2603.24857v1
+
+## 1900. SentinelAI: A Multi-Agent Framework for Structuring and Linking NG9-1-1 Emergency Incident Data
+
+- **ID:** `2603.24856` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Kliment Ho, Ilya Zaslavsky
+- **abs:** https://arxiv.org/abs/2603.24856 · **PDF:** https://arxiv.org/pdf/2603.24856v1
+
+## 1901. Resisting Humanization: Ethical Front-End Design Choices in AI for Sensitive Contexts
+
+- **ID:** `2603.24853` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Silvia Rossi, Diletta Huyskes, Mackenzie Jorgensen
+- **abs:** https://arxiv.org/abs/2603.24853 · **PDF:** https://arxiv.org/pdf/2603.24853v1
+
+## 1902. Towards automatic smoke detector inspection: Recognition of the smoke detectors in industrial facilities and preparation for future drone integration
+
+- **ID:** `2603.24850` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Lukas Kratochvila, Jakub Stefansky, Simon Bilik, et al.
+- **abs:** https://arxiv.org/abs/2603.24850 · **PDF:** https://arxiv.org/pdf/2603.24850v1
+
+## 1903. Gaze patterns predict preference and confidence in pairwise AI image evaluation
+
+- **ID:** `2603.24849` · **primary:** `cs.HC` · **Submitted:** 2026-03-25
+- **Authors:** Nikolas Papadopoulos, Shreenithi Navaneethan, Sheng Bai, et al.
+- **abs:** https://arxiv.org/abs/2603.24849 · **PDF:** https://arxiv.org/pdf/2603.24849v1
+
+## 1904. NeuroVLM-Bench: Evaluation of Vision-Enabled Large Language Models for Clinical Reasoning in Neurological Disorders
+
+- **ID:** `2603.24846` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Katarina Trojachanec Dineva, Stefan Andonov, Ilinka Ivanoska, et al.
+- **abs:** https://arxiv.org/abs/2603.24846 · **PDF:** https://arxiv.org/pdf/2603.24846v1
+
+## 1905. Robust Batch-Level Query Routing for Large Language Models under Cost and Capacity Constraints
+
+- **ID:** `2603.26796` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Jelena Markovic-Voronov, Kayhan Behdin, Yuanda Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.26796 · **PDF:** https://arxiv.org/pdf/2603.26796v1
+
+## 1906. Reaching Beyond the Mode: RL for Distributional Reasoning in Language Models
+
+- **ID:** `2603.24844` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Isha Puri, Mehul Damani, Idan Shenfeld, et al.
+- **abs:** https://arxiv.org/abs/2603.24844 · **PDF:** https://arxiv.org/pdf/2603.24844v1
+
+## 1907. Prune as You Generate: Online Rollout Pruning for Faster and Better RLVR
+
+- **ID:** `2603.24840` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Haobo Xu, Sirui Chen, Ruizhong Qiu, et al.
+- **abs:** https://arxiv.org/abs/2603.24840 · **PDF:** https://arxiv.org/pdf/2603.24840v1
+
+## 1908. Robust Matrix Estimation with Side Information
+
+- **ID:** `2603.24833` · **primary:** `stat.ME` · **Submitted:** 2026-03-25
+- **Authors:** Anish Agarwal, Jungjun Choi, Ming Yuan
+- **abs:** https://arxiv.org/abs/2603.24833 · **PDF:** https://arxiv.org/pdf/2603.24833v1
+
+## 1909. Flow matching on homogeneous spaces
+
+- **ID:** `2603.24829` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Francesco Ruscelli
+- **abs:** https://arxiv.org/abs/2603.24829 · **PDF:** https://arxiv.org/pdf/2603.24829v1
+
+## 1910. A Practical Guide Towards Interpreting Time-Series Deep Clinical Predictive Models: A Reproducibility Study
+
+- **ID:** `2603.24828` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Yongda Fan, John Wu, Andrea Fitzpatrick, et al.
+- **abs:** https://arxiv.org/abs/2603.24828 · **PDF:** https://arxiv.org/pdf/2603.24828v1
+
+## 1911. HASS: Hierarchical Simulation of Logopenic Aphasic Speech for Scalable PPA Detection
+
+- **ID:** `2603.26795` · **primary:** `eess.AS` · **Submitted:** 2026-03-25
+- **Authors:** Harrison Li, Kevin Wang, Cheol Jun Cho, et al.
+- **abs:** https://arxiv.org/abs/2603.26795 · **PDF:** https://arxiv.org/pdf/2603.26795v1
+
+## 1912. Synthetic Rewriting as a Quality Multiplier: Evidence from Portuguese Continued Pretraining
+
+- **ID:** `2603.24826` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Thales Sales Almeida, Rodrigo Nogueira, Hélio Pedrini
+- **abs:** https://arxiv.org/abs/2603.24826 · **PDF:** https://arxiv.org/pdf/2603.24826v1
+
+## 1913. Learning From Developers: Towards Reliable Patch Validation at Scale for Linux
+
+- **ID:** `2603.24825` · **primary:** `cs.SE` · **Submitted:** 2026-03-25
+- **Authors:** Chih-En Lin, Attreyee Mukherjee, Ajay Rawat, et al.
+- **abs:** https://arxiv.org/abs/2603.24825 · **PDF:** https://arxiv.org/pdf/2603.24825v1
+
+## 1914. Generative Adversarial Perturbations with Cross-paradigm Transferability on Localized Crowd Counting
+
+- **ID:** `2603.24821` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Alabi Mehzabin Anisha, Guangjing Wang, Sriram Chellappan
+- **abs:** https://arxiv.org/abs/2603.24821 · **PDF:** https://arxiv.org/pdf/2603.24821v1
+
+## 1915. PhyDCM: A Reproducible Open-Source Framework for AI-Assisted Brain Tumor Classification from Multi-Sequence MRI
+
+- **ID:** `2603.26794` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Hayder Saad Abdulbaqi, Mohammed Hadi Rahim, Mohammed Hassan Hadi, et al.
+- **abs:** https://arxiv.org/abs/2603.26794 · **PDF:** https://arxiv.org/pdf/2603.26794v1
+
+## 1916. FODMP: Fast One-Step Diffusion of Movement Primitives Generation for Time-Dependent Robot Actions
+
+- **ID:** `2603.24806` · **primary:** `cs.RO` · **Submitted:** 2026-03-25
+- **Authors:** Xirui Shi, Arya Ebrahimi, Yi Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.24806 · **PDF:** https://arxiv.org/pdf/2603.24806v1
+
+## 1917. GoldiCLIP: The Goldilocks Approach for Balancing Explicit Supervision for Language-Image Pretraining
+
+- **ID:** `2603.24804` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Deen Dayal Mohan, Hossein Souri, Vitali Petsiuk, et al.
+- **abs:** https://arxiv.org/abs/2603.24804 · **PDF:** https://arxiv.org/pdf/2603.24804v1
+
+## 1918. Dissecting Model Failures in Abdominal Aortic Aneurysm Segmentation through Explainability-Driven Analysis
+
+- **ID:** `2603.24801` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Abu Noman Md Sakib, Merjulah Roby, Zijie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.24801 · **PDF:** https://arxiv.org/pdf/2603.24801v1
+
+## 1919. Enhancing Structured Meaning Representations with Aspect Classification
+
+- **ID:** `2603.24797` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Claire Benét Post, Paul Bontempo, August Milliken, et al.
+- **abs:** https://arxiv.org/abs/2603.24797 · **PDF:** https://arxiv.org/pdf/2603.24797v1
+
+## 1920. Local learning for stable backpropagation-free neural network training towards physical learning
+
+- **ID:** `2603.24790` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Yaqi Guo, Fabian Braun, Bastiaan Ketelaar, et al.
+- **abs:** https://arxiv.org/abs/2603.24790 · **PDF:** https://arxiv.org/pdf/2603.24790v1
+
+## 1921. ReLope: KL-Regularized LoRA Probes for Multimodal LLM Routing
+
+- **ID:** `2603.24787` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Yaopei Zeng, Congchao Wang, Blake JianHang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.24787 · **PDF:** https://arxiv.org/pdf/2603.24787v1
+
+## 1922. Transformers in the Dark: Navigating Unknown Search Spaces via Bandit Feedback
+
+- **ID:** `2603.24780` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Jungtaek Kim, Thomas Zeng, Ziqian Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.24780 · **PDF:** https://arxiv.org/pdf/2603.24780v1
+
+## 1923. AIP: Agent Identity Protocol for Verifiable Delegation Across MCP and A2A
+
+- **ID:** `2603.24775` · **primary:** `cs.CR` · **Submitted:** 2026-03-25
+- **Authors:** Sunil Prakash
+- **abs:** https://arxiv.org/abs/2603.24775 · **PDF:** https://arxiv.org/pdf/2603.24775v1
+
+## 1924. From Untestable to Testable: Metamorphic Testing in the Age of LLMs
+
+- **ID:** `2603.24774` · **primary:** `cs.SE` · **Submitted:** 2026-03-25
+- **Authors:** Valerio Terragni
+- **abs:** https://arxiv.org/abs/2603.24774 · **PDF:** https://arxiv.org/pdf/2603.24774v1
+
+## 1925. Evaluating Fine-Tuned LLM Model For Medical Transcription With Small Low-Resource Languages Validated Dataset
+
+- **ID:** `2603.24772` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Mohammed Nowshad Ruhani Chowdhury, Mohammed Nowaz Rabbani Chowdhury, Sakari Lukkarinen
+- **abs:** https://arxiv.org/abs/2603.24772 · **PDF:** https://arxiv.org/pdf/2603.24772v1
+
+## 1926. Identifiable Deep Latent Variable Models for MNAR Data
+
+- **ID:** `2603.24771` · **primary:** `stat.ME` · **Submitted:** 2026-03-25
+- **Authors:** Huiming Xie, Fei Xue, Xiao Wang
+- **abs:** https://arxiv.org/abs/2603.24771 · **PDF:** https://arxiv.org/pdf/2603.24771v2
+
+## 1927. Fine-Tuning A Large Language Model for Systematic Review Screening
+
+- **ID:** `2603.24767` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Kweku Yamoah, Noah Schroeder, Emmanuel Dorley, et al.
+- **abs:** https://arxiv.org/abs/2603.24767 · **PDF:** https://arxiv.org/pdf/2603.24767v1
+
+## 1928. Supervising Ralph Wiggum: Exploring a Metacognitive Co-Regulation Agentic AI Loop for Engineering Design
+
+- **ID:** `2603.24768` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Zeda Xu, Nikolas Martelaro, Christopher McComb
+- **abs:** https://arxiv.org/abs/2603.24768 · **PDF:** https://arxiv.org/pdf/2603.24768v1
+
+## 1929. Enhancing Online Support Group Formation Using Topic Modeling Techniques
+
+- **ID:** `2603.24765` · **primary:** `cs.IR` · **Submitted:** 2026-03-25
+- **Authors:** Pronob Kumar Barman, Tera L. Reynolds, James Foulds
+- **abs:** https://arxiv.org/abs/2603.24765 · **PDF:** https://arxiv.org/pdf/2603.24765v2
+
+## 1930. Synthetic Cardiac MRI Image Generation using Deep Generative Models
+
+- **ID:** `2603.24764` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Ishan Kumarasinghe, Dasuni Kawya, Madhura Edirisooriya, et al.
+- **abs:** https://arxiv.org/abs/2603.24764 · **PDF:** https://arxiv.org/pdf/2603.24764v1
+
+## 1931. Binary Expansion Group Intersection Network
+
+- **ID:** `2603.24763` · **primary:** `math.ST` · **Submitted:** 2026-03-25
+- **Authors:** Sicheng Zhou, Kai Zhang
+- **abs:** https://arxiv.org/abs/2603.24763 · **PDF:** https://arxiv.org/pdf/2603.24763v2
+
+## 1932. SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks
+
+- **ID:** `2603.24755` · **primary:** `cs.SE` · **Submitted:** 2026-03-25
+- **Authors:** Gabriel Orlanski, Devjeet Roy, Alexander Yun, et al.
+- **abs:** https://arxiv.org/abs/2603.24755 · **PDF:** https://arxiv.org/pdf/2603.24755v1
+
+## 1933. Light Cones For Vision: Simple Causal Priors For Visual Hierarchy
+
+- **ID:** `2603.24753` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Manglam Kartik, Neel Tushar Shah
+- **abs:** https://arxiv.org/abs/2603.24753 · **PDF:** https://arxiv.org/pdf/2603.24753v1
+
+## 1934. Autotuning T-PaiNN: Enabling Data-Efficient GNN Interatomic Potential Development via Classical-to-Quantum Transfer Learning
+
+- **ID:** `2603.24752` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-25
+- **Authors:** Vivienne Pelletier, Vedant Bhat, Daniel J. Rivera, et al.
+- **abs:** https://arxiv.org/abs/2603.24752 · **PDF:** https://arxiv.org/pdf/2603.24752v1
+
+## 1935. Pseudo Label NCF for Sparse OHC Recommendation: Dual Representation Learning and the Separability Accuracy Trade off
+
+- **ID:** `2603.24750` · **primary:** `cs.IR` · **Submitted:** 2026-03-25
+- **Authors:** Pronob Kumar Barman, Tera L. Reynolds, James Foulds
+- **abs:** https://arxiv.org/abs/2603.24750 · **PDF:** https://arxiv.org/pdf/2603.24750v2
+
+## 1936. Formal Semantics for Agentic Tool Protocols: A Process Calculus Approach
+
+- **ID:** `2603.24747` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Andreas Schlapbach
+- **abs:** https://arxiv.org/abs/2603.24747 · **PDF:** https://arxiv.org/pdf/2603.24747v1
+
+## 1937. A-SelecT: Automatic Timestep Selection for Diffusion Transformer Representation Learning
+
+- **ID:** `2603.25758` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Changyu Liu, James Chenhao Liang, Wenhao Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.25758 · **PDF:** https://arxiv.org/pdf/2603.25758v1
+
+## 1938. Grokking as a Falsifiable Finite-Size Transition
+
+- **ID:** `2603.24746` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Yuda Bi, Chenyu Zhang, Qiheng Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.24746 · **PDF:** https://arxiv.org/pdf/2603.24746v1
+
+## 1939. Contrastive Learning Boosts Deterministic and Generative Models for Weather Data
+
+- **ID:** `2603.24744` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Nathan Bailey
+- **abs:** https://arxiv.org/abs/2603.24744 · **PDF:** https://arxiv.org/pdf/2603.24744v1
+
+## 1940. GaloisSAT: Differentiable Boolean Satisfiability Solving via Finite Field Algebra
+
+- **ID:** `2603.28796` · **primary:** `cs.LO` · **Submitted:** 2026-03-25
+- **Authors:** Curie Kim, Carsten Portner, Mingju Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.28796 · **PDF:** https://arxiv.org/pdf/2603.28796v1
+
+## 1941. Trust as Monitoring: Evolutionary Dynamics of User Trust and AI Developer Behaviour
+
+- **ID:** `2603.24742` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Adeela Bashir, Zhao Song, Ndidi Bianca Ogbo, et al.
+- **abs:** https://arxiv.org/abs/2603.24742 · **PDF:** https://arxiv.org/pdf/2603.24742v1
+
+## 1942. Decentralized Task Scheduling in Distributed Systems: A Deep Reinforcement Learning Approach
+
+- **ID:** `2603.24738` · **primary:** `cs.DC` · **Submitted:** 2026-03-25
+- **Authors:** Daniel Benniah John
+- **abs:** https://arxiv.org/abs/2603.24738 · **PDF:** https://arxiv.org/pdf/2603.24738v1
+
+## 1943. AutoSAM: an Agentic Framework for Automating Input File Generation for the SAM Code with Multi-Modal Retrieval-Augmented Generation
+
+- **ID:** `2603.24736` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Zaid Abulawi, Zavier Ndum Ndum, Eric Cervi, et al.
+- **abs:** https://arxiv.org/abs/2603.24736 · **PDF:** https://arxiv.org/pdf/2603.24736v1
+
+## 1944. Is Geometry Enough? An Evaluation of Landmark-Based Gaze Estimation
+
+- **ID:** `2603.24724` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Daniele Agostinelli, Thomas Agostinelli, Andrea Generosi, et al.
+- **abs:** https://arxiv.org/abs/2603.24724 · **PDF:** https://arxiv.org/pdf/2603.24724v1
+
+## 1945. Scalable Object Relation Encoding for Better 3D Spatial Reasoning in Large Language Models
+
+- **ID:** `2603.24721` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Shengli Zhou, Minghang Zheng, Feng Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.24721 · **PDF:** https://arxiv.org/pdf/2603.24721v1
+
+## 1946. Can an Actor-Critic Optimization Framework Improve Analog Design Optimization?
+
+- **ID:** `2603.24714` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Sounak Dutta, Fin Amin, Sushil Panda, et al.
+- **abs:** https://arxiv.org/abs/2603.24714 · **PDF:** https://arxiv.org/pdf/2603.24714v1
+
+## 1947. Training LLMs for Multi-Step Tool Orchestration with Constrained Data Synthesis and Graduated Rewards
+
+- **ID:** `2603.24709` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Cheng Jiayang, Xin Liu, Zhihan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.24709 · **PDF:** https://arxiv.org/pdf/2603.24709v1
+
+## 1948. Amortized Inference for Correlated Discrete Choice Models via Equivariant Neural Networks
+
+- **ID:** `2603.24705` · **primary:** `stat.ME` · **Submitted:** 2026-03-25
+- **Authors:** Easton Huch, Michael Keane
+- **abs:** https://arxiv.org/abs/2603.24705 · **PDF:** https://arxiv.org/pdf/2603.24705v1
+
+## 1949. Conformal Selective Prediction with General Risk Control
+
+- **ID:** `2603.24704` · **primary:** `stat.ME` · **Submitted:** 2026-03-25
+- **Authors:** Tian Bai, Ying Jin
+- **abs:** https://arxiv.org/abs/2603.24704 · **PDF:** https://arxiv.org/pdf/2603.24704v1
+
+## 1950. Amplified Patch-Level Differential Privacy for Free via Random Cropping
+
+- **ID:** `2603.24695` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Kaan Durmaz, Jan Schuchardt, Sebastian Schmidt, et al.
+- **abs:** https://arxiv.org/abs/2603.24695 · **PDF:** https://arxiv.org/pdf/2603.24695v1
+
+## 1951. Reconstructing Spiking Neural Networks Using a Single Neuron with Autapses
+
+- **ID:** `2603.24692` · **primary:** `cs.NE` · **Submitted:** 2026-03-25
+- **Authors:** Wuque Cai, Hongze Sun, Quan Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.24692 · **PDF:** https://arxiv.org/pdf/2603.24692v1
+
+## 1952. When Is Collective Intelligence a Lottery? Multi-Agent Scaling Laws for Memetic Drift in LLMs
+
+- **ID:** `2603.24676` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Hidenori Tanaka
+- **abs:** https://arxiv.org/abs/2603.24676 · **PDF:** https://arxiv.org/pdf/2603.24676v1
+
+## 1953. Spectral methods: crucial for machine learning, natural for quantum computers?
+
+- **ID:** `2603.24654` · **primary:** `quant-ph` · **Submitted:** 2026-03-25
+- **Authors:** Vasilis Belis, Joseph Bowles, Rishabh Gupta, et al.
+- **abs:** https://arxiv.org/abs/2603.24654 · **PDF:** https://arxiv.org/pdf/2603.24654v1
+
+## 1954. Polynomial Speedup in Diffusion Models with the Multilevel Euler-Maruyama Method
+
+- **ID:** `2603.24594` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Arthur Jacot
+- **abs:** https://arxiv.org/abs/2603.24594 · **PDF:** https://arxiv.org/pdf/2603.24594v1
+
+## 1955. DreamerAD: Efficient Reinforcement Learning via Latent World Model for Autonomous Driving
+
+- **ID:** `2603.24587` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Pengxuan Yang, Yupeng Zheng, Deheng Qian, et al.
+- **abs:** https://arxiv.org/abs/2603.24587 · **PDF:** https://arxiv.org/pdf/2603.24587v2
+
+## 1956. Comparing Developer and LLM Biases in Code Evaluation
+
+- **ID:** `2603.24586` · **primary:** `cs.SE` · **Submitted:** 2026-03-25
+- **Authors:** Aditya Mittal, Ryan Shar, Zichu Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.24586 · **PDF:** https://arxiv.org/pdf/2603.24586v1
+
+## 1957. The Stochastic Gap: A Markovian Framework for Pre-Deployment Reliability and Oversight-Cost Auditing in Agentic Artificial Intelligence
+
+- **ID:** `2603.24582` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Biplab Pal, Santanu Bhattacharya
+- **abs:** https://arxiv.org/abs/2603.24582 · **PDF:** https://arxiv.org/pdf/2603.24582v1
+
+## 1958. Demystifying When Pruning Works via Representation Hierarchies
+
+- **ID:** `2603.24652` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Shwai He, Guoheng Sun, Haichao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.24652 · **PDF:** https://arxiv.org/pdf/2603.24652v1
+
+## 1959. When Consistency Becomes Bias: Interviewer Effects in Semi-Structured Clinical Interviews
+
+- **ID:** `2603.24651` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Hasindri Watawana, Sergio Burdisso, Diego A. Moreno-Galván, et al.
+- **abs:** https://arxiv.org/abs/2603.24651 · **PDF:** https://arxiv.org/pdf/2603.24651v1
+
+## 1960. Retrieval Improvements Do Not Guarantee Better Answers: A Study of RAG for AI Policy QA
+
+- **ID:** `2603.24580` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Saahil Mathur, Ryan David Rittner, Vedant Ajit Thakur, et al.
+- **abs:** https://arxiv.org/abs/2603.24580 · **PDF:** https://arxiv.org/pdf/2603.24580v1
+
+## 1961. MARCH: Multi-Agent Reinforced Self-Check for LLM Hallucination
+
+- **ID:** `2603.24579` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Zhuo Li, Yupeng Zhang, Pengyu Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.24579 · **PDF:** https://arxiv.org/pdf/2603.24579v1
+
+## 1962. EndoVGGT: GNN-Enhanced Depth Estimation for Surgical 3D Reconstruction
+
+- **ID:** `2603.24577` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Falong Fan, Yi Xie, Arnis Lektauers, et al.
+- **abs:** https://arxiv.org/abs/2603.24577 · **PDF:** https://arxiv.org/pdf/2603.24577v1
+
+## 1963. Chameleon: Episodic Memory for Long-Horizon Robotic Manipulation
+
+- **ID:** `2603.24576` · **primary:** `cs.RO` · **Submitted:** 2026-03-25
+- **Authors:** Xinying Guo, Chenxi Jiang, Hyun Bin Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.24576 · **PDF:** https://arxiv.org/pdf/2603.24576v1
+
+## 1964. VFIG: Vectorizing Complex Figures in SVG with Vision-Language Models
+
+- **ID:** `2603.24575` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Qijia He, Xunmei Liu, Hammaad Memon, et al.
+- **abs:** https://arxiv.org/abs/2603.24575 · **PDF:** https://arxiv.org/pdf/2603.24575v1
+
+## 1965. Completeness of Unbounded Best-First Minimax and Descent Minimax
+
+- **ID:** `2603.24572` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Quentin Cohen-Solal
+- **abs:** https://arxiv.org/abs/2603.24572 · **PDF:** https://arxiv.org/pdf/2603.24572v1
+
+## 1966. Anti-I2V: Safeguarding your photos from malicious image-to-video generation
+
+- **ID:** `2603.24570` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Duc Vu, Anh Nguyen, Chi Tran, et al.
+- **abs:** https://arxiv.org/abs/2603.24570 · **PDF:** https://arxiv.org/pdf/2603.24570v1
+
+## 1967. Trust Region Constrained Bayesian Optimization with Penalized Constraint Handling
+
+- **ID:** `2603.24567` · **primary:** `stat.ML` · **Submitted:** 2026-03-25
+- **Authors:** Raju Chowdhury, Tanmay Sen, Prajamitra Bhuyan, et al.
+- **abs:** https://arxiv.org/abs/2603.24567 · **PDF:** https://arxiv.org/pdf/2603.24567v1
+
+## 1968. Scaling Recurrence-aware Foundation Models for Clinical Records via Next-Visit Prediction
+
+- **ID:** `2603.24562` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Haresh Rengaraj Rajamohan, Xiang Gao, Weicheng Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.24562 · **PDF:** https://arxiv.org/pdf/2603.24562v1
+
+## 1969. The Free-Market Algorithm: Self-Organizing Optimization for Open-Ended Complex Systems
+
+- **ID:** `2603.24559` · **primary:** `cs.NE` · **Submitted:** 2026-03-25
+- **Authors:** Martin Jaraiz
+- **abs:** https://arxiv.org/abs/2603.24559 · **PDF:** https://arxiv.org/pdf/2603.24559v1
+
+## 1970. Transformers for Program Termination
+
+- **ID:** `2604.00039` · **primary:** `cs.PL` · **Submitted:** 2026-03-25
+- **Authors:** Yoav Alon, Cristina David
+- **abs:** https://arxiv.org/abs/2604.00039 · **PDF:** https://arxiv.org/pdf/2604.00039v1
+
+## 1971. LensWalk: Agentic Video Understanding by Planning How You See in Videos
+
+- **ID:** `2603.24558` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Keliang Li, Yansong Li, Hongze Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.24558 · **PDF:** https://arxiv.org/pdf/2603.24558v1
+
+## 1972. Evaluating Chunking Strategies For Retrieval-Augmented Generation in Oil and Gas Enterprise Documents
+
+- **ID:** `2603.24556` · **primary:** `cs.IR` · **Submitted:** 2026-03-25
+- **Authors:** Samuel Taiwo, Mohd Amaluddin Yusoff
+- **abs:** https://arxiv.org/abs/2603.24556 · **PDF:** https://arxiv.org/pdf/2603.24556v1
+
+## 1973. Energy-Efficient Hierarchical Federated Anomaly Detection for the Internet of Underwater Things via Selective Cooperative Aggregation
+
+- **ID:** `2603.24648` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Kenechi Omeke, Michael Mollel, Lei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.24648 · **PDF:** https://arxiv.org/pdf/2603.24648v1
+
+## 1974. Can LLMs Beat Classical Hyperparameter Optimization Algorithms? A Study on autoresearch
+
+- **ID:** `2603.24647` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Fabio Ferreira, Lucca Wobbe, Arjun Krishnakumar, et al.
+- **abs:** https://arxiv.org/abs/2603.24647 · **PDF:** https://arxiv.org/pdf/2603.24647v2
+
+## 1975. A Sociolinguistic Analysis of Automatic Speech Recognition Bias in Newcastle English
+
+- **ID:** `2603.24549` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Dana Serditova, Kevin Tang
+- **abs:** https://arxiv.org/abs/2603.24549 · **PDF:** https://arxiv.org/pdf/2603.24549v1
+
+## 1976. Detection of local geometry in random graphs: information-theoretic and computational limits
+
+- **ID:** `2603.24545` · **primary:** `math.ST` · **Submitted:** 2026-03-25
+- **Authors:** Jinho Bok, Shuangping Li, Sophie H. Yu
+- **abs:** https://arxiv.org/abs/2603.24545 · **PDF:** https://arxiv.org/pdf/2603.24545v1
+
+## 1977. Analysing the Safety Pitfalls of Steering Vectors
+
+- **ID:** `2603.24543` · **primary:** `cs.CR` · **Submitted:** 2026-03-25
+- **Authors:** Yuxiao Li, Alina Fastowski, Efstratios Zaradoukas, et al.
+- **abs:** https://arxiv.org/abs/2603.24543 · **PDF:** https://arxiv.org/pdf/2603.24543v1
+
+## 1978. SEGAR: Selective Enhancement for Generative Augmented Reality
+
+- **ID:** `2603.24541` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Fanjun Bu, Chenyang Yuan, Hiroshi Yasuda
+- **abs:** https://arxiv.org/abs/2603.24541 · **PDF:** https://arxiv.org/pdf/2603.24541v1
+
+## 1979. CliPPER: Contextual Video-Language Pretraining on Long-form Intraoperative Surgical Procedures for Event Recognition
+
+- **ID:** `2603.24539` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Florian Stilz, Vinkle Srivastav, Nassir Navab, et al.
+- **abs:** https://arxiv.org/abs/2603.24539 · **PDF:** https://arxiv.org/pdf/2603.24539v1
+
+## 1980. Robust Multilingual Text-to-Pictogram Mapping for Scalable Reading Rehabilitation
+
+- **ID:** `2603.24536` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Soufiane Jhilal, Martina Galletti
+- **abs:** https://arxiv.org/abs/2603.24536 · **PDF:** https://arxiv.org/pdf/2603.24536v1
+
+## 1981. Representation Learning to Study Temporal Dynamics in Tutorial Scaffolding
+
+- **ID:** `2603.24535` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Conrad Borchers, Jiayi Zhang, Ashish Gurung
+- **abs:** https://arxiv.org/abs/2603.24535 · **PDF:** https://arxiv.org/pdf/2603.24535v1
+
+## 1982. UI-Voyager: A Self-Evolving GUI Agent Learning via Failed Experience
+
+- **ID:** `2603.24533` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Zichuan Lin, Feiyu Liu, Yijun Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.24533 · **PDF:** https://arxiv.org/pdf/2603.24533v1
+
+## 1983. From Liar Paradox to Incongruent Sets: A Normal Form for Self-Reference
+
+- **ID:** `2603.24527` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Shalender Singh, Vishnu Priya Singh Parmar
+- **abs:** https://arxiv.org/abs/2603.24527 · **PDF:** https://arxiv.org/pdf/2603.24527v1
+
+## 1984. No Single Metric Tells the Whole Story: A Multi-Dimensional Evaluation Framework for Uncertainty Attributions
+
+- **ID:** `2603.24524` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Emily Schiller, Teodor Chiaburu, Marco Zullich, et al.
+- **abs:** https://arxiv.org/abs/2603.24524 · **PDF:** https://arxiv.org/pdf/2603.24524v1
+
+## 1985. TuneShift-KD: Knowledge Distillation and Transfer for Fine-tuned Models
+
+- **ID:** `2603.24518` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Yushi Guan, Jeanine Ohene-Agyei, Daniel Kwan, et al.
+- **abs:** https://arxiv.org/abs/2603.24518 · **PDF:** https://arxiv.org/pdf/2603.24518v1
+
+## 1986. AVO: Agentic Variation Operators for Autonomous Evolutionary Search
+
+- **ID:** `2603.24517` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Terry Chen, Zhifan Ye, Bing Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.24517 · **PDF:** https://arxiv.org/pdf/2603.24517v1
+
+## 1987. A Firefly Algorithm for Mixed-Variable Optimization Based on Hybrid Distance Modeling
+
+- **ID:** `2603.26792` · **primary:** `cs.NE` · **Submitted:** 2026-03-25
+- **Authors:** Ousmane Tom Bechir, Adán José-García, Zaineb Chelly Garcia, et al.
+- **abs:** https://arxiv.org/abs/2603.26792 · **PDF:** https://arxiv.org/pdf/2603.26792v1
+
+## 1988. Claudini: Autoresearch Discovers State-of-the-Art Adversarial Attack Algorithms for LLMs
+
+- **ID:** `2603.24511` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Alexander Panfilov, Peter Romov, Igor Shilov, et al.
+- **abs:** https://arxiv.org/abs/2603.24511 · **PDF:** https://arxiv.org/pdf/2603.24511v1
+
+## 1989. Towards Safe Learning-Based Non-Linear Model Predictive Control through Recurrent Neural Network Modeling
+
+- **ID:** `2603.24503` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Mihaela-Larisa Clement, Mónika Farsang, Agnes Poks, et al.
+- **abs:** https://arxiv.org/abs/2603.24503 · **PDF:** https://arxiv.org/pdf/2603.24503v1
+
+## 1990. CRISP: Characterizing Relative Impact of Scholarly Publications
+
+- **ID:** `2603.26791` · **primary:** `cs.DL` · **Submitted:** 2026-03-25
+- **Authors:** Hannah Collison, Benjamin Van Durme, Daniel Khashabi
+- **abs:** https://arxiv.org/abs/2603.26791 · **PDF:** https://arxiv.org/pdf/2603.26791v1
+
+## 1991. Project and Generate: Divergence-Free Neural Operators for Incompressible Flows
+
+- **ID:** `2603.24500` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Xigui Li, Hongwei Zhang, Ruoxi Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.24500 · **PDF:** https://arxiv.org/pdf/2603.24500v1
+
+## 1992. Reflected diffusion models adapt to low-dimensional data
+
+- **ID:** `2603.24495` · **primary:** `math.ST` · **Submitted:** 2026-03-25
+- **Authors:** Asbjørn Holk, Claudia Strauch, Lukas Trottner
+- **abs:** https://arxiv.org/abs/2603.24495 · **PDF:** https://arxiv.org/pdf/2603.24495v1
+
+## 1993. Uniform Laws of Large Numbers in Product Spaces
+
+- **ID:** `2603.24493` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Ron Holzman, Shay Moran, Alexander Shlimovich
+- **abs:** https://arxiv.org/abs/2603.24493 · **PDF:** https://arxiv.org/pdf/2603.24493v1
+
+## 1994. Multi-Agent Reasoning with Consistency Verification Improves Uncertainty Calibration in Medical MCQA
+
+- **ID:** `2603.24481` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** John Ray B. Martinez
+- **abs:** https://arxiv.org/abs/2603.24481 · **PDF:** https://arxiv.org/pdf/2603.24481v1
+
+## 1995. Composer 2 Technical Report
+
+- **ID:** `2603.24477` · **primary:** `cs.SE` · **Submitted:** 2026-03-25
+- **Authors:** Cursor Research, :, Aaron Chan, et al.
+- **abs:** https://arxiv.org/abs/2603.24477 · **PDF:** https://arxiv.org/pdf/2603.24477v2
+
+## 1996. Conformalized Transfer Learning for Li-ion Battery State of Health Forecasting under Manufacturing and Usage Variability
+
+- **ID:** `2603.24475` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Samuel Filgueira da Silva, Mehmet Fatih Ozkan, Faissal El Idrissi, et al.
+- **abs:** https://arxiv.org/abs/2603.24475 · **PDF:** https://arxiv.org/pdf/2603.24475v1
+
+## 1997. Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs?
+
+- **ID:** `2603.24472` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Jeonghye Kim, Xufang Luo, Minbeom Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.24472 · **PDF:** https://arxiv.org/pdf/2603.24472v1
+
+## 1998. Counting Without Numbers \& Finding Without Words
+
+- **ID:** `2603.24470` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Badri Narayana Patro
+- **abs:** https://arxiv.org/abs/2603.24470 · **PDF:** https://arxiv.org/pdf/2603.24470v1
+
+## 1999. Mechanic: Sorrifier-Driven Formal Decomposition Workflow for Automated Theorem Proving
+
+- **ID:** `2603.24465` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Ruichen Qiu, Yichuan Cao, Junqi Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.24465 · **PDF:** https://arxiv.org/pdf/2603.24465v1
+
+## 2000. Integrating Causal Machine Learning into Clinical Decision Support Systems: Insights from Literature and Practice
+
+- **ID:** `2603.24448` · **primary:** `cs.HC` · **Submitted:** 2026-03-25
+- **Authors:** Domenique Zipperling, Lukas Schmidt, Benedikt Hahn, et al.
+- **abs:** https://arxiv.org/abs/2603.24448 · **PDF:** https://arxiv.org/pdf/2603.24448v1
+
+## 2001. CUA-Suite: Massive Human-annotated Video Demonstrations for Computer-Use Agents
+
+- **ID:** `2603.24440` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Xiangru Jian, Shravan Nayak, Kevin Qinghong Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.24440 · **PDF:** https://arxiv.org/pdf/2603.24440v1
+
+## 2002. Enes Causal Discovery
+
+- **ID:** `2603.24436` · **primary:** `cs.NE` · **Submitted:** 2026-03-25
+- **Authors:** Alexis Kafantaris
+- **abs:** https://arxiv.org/abs/2603.24436 · **PDF:** https://arxiv.org/pdf/2603.24436v3
+
+## 2003. What and When to Learn: CURriculum Ranking Loss for Large-Scale Speaker Verification
+
+- **ID:** `2603.24432` · **primary:** `cs.SD` · **Submitted:** 2026-03-25
+- **Authors:** Massa Baali, Sarthak Bisht, Rita Singh, et al.
+- **abs:** https://arxiv.org/abs/2603.24432 · **PDF:** https://arxiv.org/pdf/2603.24432v1
+
+## 2004. Learning Response-Statistic Shifts and Parametric Roll Episodes from Wave--Vessel Time Series via LSTM Functional Models
+
+- **ID:** `2603.24431` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Jose del Aguila Ferrandis
+- **abs:** https://arxiv.org/abs/2603.24431 · **PDF:** https://arxiv.org/pdf/2603.24431v1
+
+## 2005. Marchuk: Efficient Global Weather Forecasting from Mid-Range to Sub-Seasonal Scales via Flow Matching
+
+- **ID:** `2603.24428` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Arsen Kuzhamuratov, Mikhail Zhirnov, Andrey Kuznetsov, et al.
+- **abs:** https://arxiv.org/abs/2603.24428 · **PDF:** https://arxiv.org/pdf/2603.24428v1
+
+## 2006. Continuous-Time Learning of Probability Distributions: A Case Study in a Digital Trial of Young Children with Type 1 Diabetes
+
+- **ID:** `2603.24427` · **primary:** `stat.ML` · **Submitted:** 2026-03-25
+- **Authors:** Antonio Álvarez-López, Marcos Matabuena
+- **abs:** https://arxiv.org/abs/2603.24427 · **PDF:** https://arxiv.org/pdf/2603.24427v1
+
+## 2007. OneSearch-V2: The Latent Reasoning Enhanced Self-distillation Generative Search Framework
+
+- **ID:** `2603.24422` · **primary:** `cs.IR` · **Submitted:** 2026-03-25
+- **Authors:** Ben Chen, Siyuan Wang, Yufei Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.24422 · **PDF:** https://arxiv.org/pdf/2603.24422v1
+
+## 2008. ClawKeeper: Comprehensive Safety Protection for OpenClaw Agents Through Skills, Plugins, and Watchers
+
+- **ID:** `2603.24414` · **primary:** `cs.CR` · **Submitted:** 2026-03-25
+- **Authors:** Songyang Liu, Chaozhuo Li, Chenxu Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.24414 · **PDF:** https://arxiv.org/pdf/2603.24414v1
+
+## 2009. PINGALA: Prosody-Aware Decoding for Sanskrit Poetry Generation
+
+- **ID:** `2603.24413` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Manoj Balaji Jagadeeshan, Atul Singh, Nallani Chakravartula Sahith, et al.
+- **abs:** https://arxiv.org/abs/2603.24413 · **PDF:** https://arxiv.org/pdf/2603.24413v1
+
+## 2010. Real Talk, Virtual Faces: A Formal Concept Analysis of Personality and Sentiment in Influencer Audiences
+
+- **ID:** `2603.24410` · **primary:** `cs.CY` · **Submitted:** 2026-03-25
+- **Authors:** Shahram Chaudhry, Sidahmed Benabderrahmane, Talal Rahwan
+- **abs:** https://arxiv.org/abs/2603.24410 · **PDF:** https://arxiv.org/pdf/2603.24410v1
+
+## 2011. AI-Supervisor: Autonomous AI Research Supervision via a Persistent Research World Model
+
+- **ID:** `2603.24402` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Yunbo Long
+- **abs:** https://arxiv.org/abs/2603.24402 · **PDF:** https://arxiv.org/pdf/2603.24402v2
+
+## 2012. Neural Network Models for Contextual Regression
+
+- **ID:** `2603.24400` · **primary:** `stat.ML` · **Submitted:** 2026-03-25
+- **Authors:** Seksan Kiatsupaibul, Pakawan Chansiripas
+- **abs:** https://arxiv.org/abs/2603.24400 · **PDF:** https://arxiv.org/pdf/2603.24400v1
+
+## 2013. Exploring How Fair Model Representations Relate to Fair Recommendations
+
+- **ID:** `2603.24396` · **primary:** `cs.IR` · **Submitted:** 2026-03-25
+- **Authors:** Bjørnar Vassøy, Benjamin Kille, Helge Langseth
+- **abs:** https://arxiv.org/abs/2603.24396 · **PDF:** https://arxiv.org/pdf/2603.24396v1
+
+## 2014. Federated fairness-aware classification under differential privacy
+
+- **ID:** `2603.24392` · **primary:** `stat.ML` · **Submitted:** 2026-03-25
+- **Authors:** Gengyu Xue, Yi Yu
+- **abs:** https://arxiv.org/abs/2603.24392 · **PDF:** https://arxiv.org/pdf/2603.24392v1
+
+## 2015. When AI Meets Early Childhood Education: Large Language Models as Assessment Teammates in Chinese Preschools
+
+- **ID:** `2603.24389` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Xingming Li, Runke Huang, Yanan Bao, et al.
+- **abs:** https://arxiv.org/abs/2603.24389 · **PDF:** https://arxiv.org/pdf/2603.24389v1
+
+## 2016. On the Use of Bagging for Local Intrinsic Dimensionality Estimation
+
+- **ID:** `2603.24384` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Kristóf Péter, Ricardo J. G. B. Campello, James Bailey, et al.
+- **abs:** https://arxiv.org/abs/2603.24384 · **PDF:** https://arxiv.org/pdf/2603.24384v1
+
+## 2017. MolEvolve: LLM-Guided Evolutionary Search for Interpretable Molecular Optimization
+
+- **ID:** `2603.24382` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Xiangsen Chen, Ruilong Wu, Yanyan Lan, et al.
+- **abs:** https://arxiv.org/abs/2603.24382 · **PDF:** https://arxiv.org/pdf/2603.24382v1
+
+## 2018. Towards Reward Modeling for AI Tutors in Math Mistake Remediation
+
+- **ID:** `2603.24375` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Kseniia Petukhova, Ekaterina Kochmar
+- **abs:** https://arxiv.org/abs/2603.24375 · **PDF:** https://arxiv.org/pdf/2603.24375v1
+
+## 2019. Improving Lean4 Autoformalization via Cycle Consistency Fine-tuning
+
+- **ID:** `2603.24372` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Arsen Shebzukhov
+- **abs:** https://arxiv.org/abs/2603.24372 · **PDF:** https://arxiv.org/pdf/2603.24372v1
+
+## 2020. Adaptive decision-making for stochastic service network design
+
+- **ID:** `2603.24369` · **primary:** `math.OC` · **Submitted:** 2026-03-25
+- **Authors:** Javier Durán-Micco, Bilge Atasoy
+- **abs:** https://arxiv.org/abs/2603.24369 · **PDF:** https://arxiv.org/pdf/2603.24369v2
+
+## 2021. CoordLight: Learning Decentralized Coordination for Network-Wide Traffic Signal Control
+
+- **ID:** `2603.24366` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Yifeng Zhang, Harsh Goel, Peizhuo Li, et al.
+- **abs:** https://arxiv.org/abs/2603.24366 · **PDF:** https://arxiv.org/pdf/2603.24366v1
+
+## 2022. Physics-Informed Neural Network Digital Twin for Dynamic Tray-Wise Modeling of Distillation Columns under Transient Operating Conditions
+
+- **ID:** `2603.24644` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Debadutta Patra, Ayush Bardhan Tripathy, Soumya Ranjan Sahu, et al.
+- **abs:** https://arxiv.org/abs/2603.24644 · **PDF:** https://arxiv.org/pdf/2603.24644v1
+
+## 2023. A Neuro-Symbolic System for Interpretable Multimodal Physiological Signals Integration in Human Fatigue Detection
+
+- **ID:** `2603.24358` · **primary:** `cs.HC` · **Submitted:** 2026-03-25
+- **Authors:** Mohammadreza Jamalifard, Yaxiong Lei, Parasto Azizinezhad, et al.
+- **abs:** https://arxiv.org/abs/2603.24358 · **PDF:** https://arxiv.org/pdf/2603.24358v1
+
+## 2024. Language-Guided Structure-Aware Network for Camouflaged Object Detection
+
+- **ID:** `2603.24355` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Min Zhang
+- **abs:** https://arxiv.org/abs/2603.24355 · **PDF:** https://arxiv.org/pdf/2603.24355v1
+
+## 2025. Evidence of an Emergent "Self" in Continual Robot Learning
+
+- **ID:** `2603.24350` · **primary:** `cs.RO` · **Submitted:** 2026-03-25
+- **Authors:** Adidev Jhunjhunwala, Judah Goldfeder, Hod Lipson
+- **abs:** https://arxiv.org/abs/2603.24350 · **PDF:** https://arxiv.org/pdf/2603.24350v1
+
+## 2026. Enhancing Efficiency and Performance in Deepfake Audio Detection through Neuron-level Dropin & Neuroplasticity Mechanisms
+
+- **ID:** `2603.24343` · **primary:** `cs.SD` · **Submitted:** 2026-03-25
+- **Authors:** Yupei Li, Shuaijie Shao, Manuel Milling, et al.
+- **abs:** https://arxiv.org/abs/2603.24343 · **PDF:** https://arxiv.org/pdf/2603.24343v2
+
+## 2027. Notes on Forré's Notion of Conditional Independence and Causal Calculus for Continuous Variables
+
+- **ID:** `2603.24333` · **primary:** `math.ST` · **Submitted:** 2026-03-25
+- **Authors:** Leihao Chen
+- **abs:** https://arxiv.org/abs/2603.24333 · **PDF:** https://arxiv.org/pdf/2603.24333v1
+
+## 2028. GameplayQA: A Benchmarking Framework for Decision-Dense POV-Synced Multi-Video Understanding of 3D Virtual Agents
+
+- **ID:** `2603.24329` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Yunzhe Wang, Runhui Xu, Kexin Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.24329 · **PDF:** https://arxiv.org/pdf/2603.24329v1
+
+## 2029. Boosting Document Parsing Efficiency and Performance with Coarse-to-Fine Visual Processing
+
+- **ID:** `2603.24326` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Cheng Cui, Ting Sun, Suyin Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.24326 · **PDF:** https://arxiv.org/pdf/2603.24326v1
+
+## 2030. Large Language Model Guided Incentive Aware Reward Design for Cooperative Multi-Agent Reinforcement Learning
+
+- **ID:** `2603.24324` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Dogan Urgun, Gokhan Gungor
+- **abs:** https://arxiv.org/abs/2603.24324 · **PDF:** https://arxiv.org/pdf/2603.24324v2
+
+## 2031. Connecting Meteorite Spectra to Lunar Surface Composition Using Hyperspectral Imaging and Machine Learning
+
+- **ID:** `2603.24323` · **primary:** `astro-ph.EP` · **Submitted:** 2026-03-25
+- **Authors:** Fatemeh Fazel Hesar, Mojtaba Raouf, Amirmohammad Chegeni, et al.
+- **abs:** https://arxiv.org/abs/2603.24323 · **PDF:** https://arxiv.org/pdf/2603.24323v1
+
+## 2032. Toward Generalist Neural Motion Planners for Robotic Manipulators: Challenges and Opportunities
+
+- **ID:** `2603.24318` · **primary:** `cs.RO` · **Submitted:** 2026-03-25
+- **Authors:** Davood Soleymanzadeh, Ivan Lopez-Sanchez, Hao Su, et al.
+- **abs:** https://arxiv.org/abs/2603.24318 · **PDF:** https://arxiv.org/pdf/2603.24318v1
+
+## 2033. Samasāmayik: A Parallel Dataset for Hindi-Sanskrit Machine Translation
+
+- **ID:** `2603.24307` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** N J Karthika, Keerthana Suryanarayanan, Jahanvi Purohit, et al.
+- **abs:** https://arxiv.org/abs/2603.24307 · **PDF:** https://arxiv.org/pdf/2603.24307v1
+
+## 2034. CGRL: Causal-Guided Representation Learning for Graph Out-of-Distribution Generalization
+
+- **ID:** `2603.24304` · **primary:** `stat.ML` · **Submitted:** 2026-03-25
+- **Authors:** Bowen Lu, Liangqiang Yang, Teng Li
+- **abs:** https://arxiv.org/abs/2603.24304 · **PDF:** https://arxiv.org/pdf/2603.24304v1
+
+## 2035. SpinGQE: A Generative Quantum Eigensolver for Spin Hamiltonians
+
+- **ID:** `2603.24298` · **primary:** `quant-ph` · **Submitted:** 2026-03-25
+- **Authors:** Alexander Holden, Moinul Hossain Rahat, Nii Osae Osae Dade
+- **abs:** https://arxiv.org/abs/2603.24298 · **PDF:** https://arxiv.org/pdf/2603.24298v1
+
+## 2036. Cost-Sensitive Neighborhood Aggregation for Heterophilous Graphs: When Does Per-Edge Routing Help?
+
+- **ID:** `2603.24291` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Eyal Weiss
+- **abs:** https://arxiv.org/abs/2603.24291 · **PDF:** https://arxiv.org/pdf/2603.24291v2
+
+## 2037. The Specification Gap: Coordination Failure Under Partial Knowledge in Code Agents
+
+- **ID:** `2603.24284` · **primary:** `cs.SE` · **Submitted:** 2026-03-25
+- **Authors:** Camilo Chacón Sartori
+- **abs:** https://arxiv.org/abs/2603.24284 · **PDF:** https://arxiv.org/pdf/2603.24284v1
+
+## 2038. Bridging Biological Hearing and Neuromorphic Computing: End-to-End Time-Domain Audio Signal Processing with Reservoir Computing
+
+- **ID:** `2603.24283` · **primary:** `cs.SD` · **Submitted:** 2026-03-25
+- **Authors:** Rinku Sebastian, Simon O'Keefe, Martin Trefzer
+- **abs:** https://arxiv.org/abs/2603.24283 · **PDF:** https://arxiv.org/pdf/2603.24283v1
+
+## 2039. Language-Assisted Image Clustering Guided by Discriminative Relational Signals and Adaptive Semantic Centers
+
+- **ID:** `2603.24275` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Jun Ma, Xu Zhang, Zhengxing Jiao, et al.
+- **abs:** https://arxiv.org/abs/2603.24275 · **PDF:** https://arxiv.org/pdf/2603.24275v1
+
+## 2040. DeepDTF: Dual-Branch Transformer Fusion for Multi-Omics Anticancer Drug Response Prediction
+
+- **ID:** `2603.24265` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Yuhan Zhao, Jacob Tennant, James Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.24265 · **PDF:** https://arxiv.org/pdf/2603.24265v1
+
+## 2041. Forecasting with Guidance: Representation-Level Supervision for Time Series Forecasting
+
+- **ID:** `2603.24262` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Jiacheng Wang, Liang Fan, Baihua Li, et al.
+- **abs:** https://arxiv.org/abs/2603.24262 · **PDF:** https://arxiv.org/pdf/2603.24262v1
+
+## 2042. Accelerating Diffusion-based Video Editing via Heterogeneous Caching: Beyond Full Computing at Sampled Denoising Timestep
+
+- **ID:** `2603.24260` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Tianyi Liu, Ye Lu, Linfeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.24260 · **PDF:** https://arxiv.org/pdf/2603.24260v1
+
+## 2043. Semantic Alignment across Ancient Egyptian Language Stages via Normalization-Aware Multitask Learning
+
+- **ID:** `2603.24258` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** He Huang
+- **abs:** https://arxiv.org/abs/2603.24258 · **PDF:** https://arxiv.org/pdf/2603.24258v1
+
+## 2044. Embracing Heteroscedasticity for Probabilistic Time Series Forecasting
+
+- **ID:** `2603.24254` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Yijun Wang, Qiyuan Zhuang, Xiu-Shen Wei
+- **abs:** https://arxiv.org/abs/2603.24254 · **PDF:** https://arxiv.org/pdf/2603.24254v1
+
+## 2045. Semantic Centroids and Hierarchical Density-Based Clustering for Cross-Document Software Coreference Resolution
+
+- **ID:** `2603.24246` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Julia Matela, Frank Krüger
+- **abs:** https://arxiv.org/abs/2603.24246 · **PDF:** https://arxiv.org/pdf/2603.24246v1
+
+## 2046. Optimizing Multilingual LLMs via Federated Learning: A Study of Client Language Composition
+
+- **ID:** `2603.24242` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Aleix Sant, Jordi Luque, Carlos Escolano
+- **abs:** https://arxiv.org/abs/2603.24242 · **PDF:** https://arxiv.org/pdf/2603.24242v2
+
+## 2047. C-STEP: Continuous Space-Time Empowerment for Physics-informed Safe Reinforcement Learning of Mobile Agents
+
+- **ID:** `2603.24241` · **primary:** `eess.SY` · **Submitted:** 2026-03-25
+- **Authors:** Guihlerme Daubt, Adrian Redder
+- **abs:** https://arxiv.org/abs/2603.24241 · **PDF:** https://arxiv.org/pdf/2603.24241v1
+
+## 2048. DVM: A Bytecode Virtual Machine Approach for Dynamic Tensor Computation
+
+- **ID:** `2603.24239` · **primary:** `cs.PL` · **Submitted:** 2026-03-25
+- **Authors:** Jingzhi Fang, Xiong Gao, Renwei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.24239 · **PDF:** https://arxiv.org/pdf/2603.24239v2
+
+## 2049. Learning Mesh-Free Discrete Differential Operators with Self-Supervised Graph Neural Networks
+
+- **ID:** `2603.24641` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Lucas Gerken Starepravo, Georgios Fourtakas, Steven Lind, et al.
+- **abs:** https://arxiv.org/abs/2603.24641 · **PDF:** https://arxiv.org/pdf/2603.24641v1
+
+## 2050. Attack Assessment and Augmented Identity Recognition for Human Skeleton Data
+
+- **ID:** `2603.24232` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Joseph G. Zalameda, Megan A. Witherow, Alexander M. Glandon, et al.
+- **abs:** https://arxiv.org/abs/2603.24232 · **PDF:** https://arxiv.org/pdf/2603.24232v1
+
+## 2051. Stance Labels Fail When They Matter Most: The Projection Problem in Stance Detection
+
+- **ID:** `2603.24231` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Bowen Zhang
+- **abs:** https://arxiv.org/abs/2603.24231 · **PDF:** https://arxiv.org/pdf/2603.24231v1
+
+## 2052. Identification of NMF by choosing maximum-volume basis vectors
+
+- **ID:** `2603.24227` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Qianqian Qi, Zhongming Chen, Peter G. M. van der Heijden
+- **abs:** https://arxiv.org/abs/2603.24227 · **PDF:** https://arxiv.org/pdf/2603.24227v1
+
+## 2053. UniScale: Synergistic Entire Space Data and Model Scaling for Search Ranking
+
+- **ID:** `2603.24226` · **primary:** `cs.IR` · **Submitted:** 2026-03-25
+- **Authors:** Liren Yu, Caiyuan Li, Feiyi Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.24226 · **PDF:** https://arxiv.org/pdf/2603.24226v2
+
+## 2054. Variation is the Norm: Embracing Sociolinguistics in NLP
+
+- **ID:** `2603.24222` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Anne-Marie Lutgen, Alistair Plum, Verena Blaschke, et al.
+- **abs:** https://arxiv.org/abs/2603.24222 · **PDF:** https://arxiv.org/pdf/2603.24222v1
+
+## 2055. Environment-Grounded Multi-Agent Workflow for Autonomous Penetration Testing
+
+- **ID:** `2603.24221` · **primary:** `cs.RO` · **Submitted:** 2026-03-25
+- **Authors:** Michael Somma, Markus Großpointner, Paul Zabalegui, et al.
+- **abs:** https://arxiv.org/abs/2603.24221 · **PDF:** https://arxiv.org/pdf/2603.24221v1
+
+## 2056. Who Benefits from RAG? The Role of Exposure, Utility and Attribution Bias
+
+- **ID:** `2603.24218` · **primary:** `cs.IR` · **Submitted:** 2026-03-25
+- **Authors:** Mahdi Dehghan, Graham McDonald
+- **abs:** https://arxiv.org/abs/2603.24218 · **PDF:** https://arxiv.org/pdf/2603.24218v1
+
+## 2057. Where Do Your Citations Come From? Citation-Constellation: A Free, Open-Source, No-Code, and Auditable Tool for Citation Network Decomposition with Complementary BARON and HEROCON Scores
+
+- **ID:** `2603.24216` · **primary:** `cs.DL` · **Submitted:** 2026-03-25
+- **Authors:** Mahbub Ul Alam
+- **abs:** https://arxiv.org/abs/2603.24216 · **PDF:** https://arxiv.org/pdf/2603.24216v1
+
+## 2058. Uncovering Memorization in Timeseries Imputation models: LBRM Membership Inference and its link to attribute Leakage
+
+- **ID:** `2603.24213` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Faiz Taleb, Ivan Gazeau, Maryline Laurent
+- **abs:** https://arxiv.org/abs/2603.24213 · **PDF:** https://arxiv.org/pdf/2603.24213v1
+
+## 2059. Experiential Reflective Learning for Self-Improving LLM Agents
+
+- **ID:** `2603.24639` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Marc-Antoine Allard, Arnaud Teinturier, Victor Xing, et al.
+- **abs:** https://arxiv.org/abs/2603.24639 · **PDF:** https://arxiv.org/pdf/2603.24639v2
+
+## 2060. How unconstrained machine-learning models learn physical symmetries
+
+- **ID:** `2603.24638` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Michelangelo Domina, Joseph William Abbott, Paolo Pegolo, et al.
+- **abs:** https://arxiv.org/abs/2603.24638 · **PDF:** https://arxiv.org/pdf/2603.24638v1
+
+## 2061. HEART-PFL: Stable Personalized Federated Learning under Heterogeneity with Hierarchical Directional Alignment and Adversarial Knowledge Transfer
+
+- **ID:** `2603.24209` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Minjun Kim, Minje Kim
+- **abs:** https://arxiv.org/abs/2603.24209 · **PDF:** https://arxiv.org/pdf/2603.24209v1
+
+## 2062. Powerful Teachers Matter: Text-Guided Multi-view Knowledge Distillation with Visual Prior Enhancement
+
+- **ID:** `2603.24208` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Xin Zhang, Jianyang Xu, Hao Peng, et al.
+- **abs:** https://arxiv.org/abs/2603.24208 · **PDF:** https://arxiv.org/pdf/2603.24208v1
+
+## 2063. IPatch: A Multi-Resolution Transformer Architecture for Robust Time-Series Forecasting
+
+- **ID:** `2603.24207` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Aymane Harkati, Moncef Garouani, Olivier Teste, et al.
+- **abs:** https://arxiv.org/abs/2603.24207 · **PDF:** https://arxiv.org/pdf/2603.24207v1
+
+## 2064. Invisible Threats from Model Context Protocol: Generating Stealthy Injection Payload via Tree-based Adaptive Search
+
+- **ID:** `2603.24203` · **primary:** `cs.CR` · **Submitted:** 2026-03-25
+- **Authors:** Yulin Shen, Xudong Pan, Geng Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.24203 · **PDF:** https://arxiv.org/pdf/2603.24203v1
+
+## 2065. A Deep Dive into Scaling RL for Code Generation with Synthetic Data and Curricula
+
+- **ID:** `2603.24202` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Cansu Sancaktar, David Zhang, Gabriel Synnaeve, et al.
+- **abs:** https://arxiv.org/abs/2603.24202 · **PDF:** https://arxiv.org/pdf/2603.24202v1
+
+## 2066. Quantum Neural Physics: Solving Partial Differential Equations on Quantum Simulators using Quantum Convolutional Neural Networks
+
+- **ID:** `2603.24196` · **primary:** `quant-ph` · **Submitted:** 2026-03-25
+- **Authors:** Jucai Zhai, Muhammad Abdullah, Boyang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.24196 · **PDF:** https://arxiv.org/pdf/2603.24196v1
+
+## 2067. TsetlinWiSARD: On-Chip Training of Weightless Neural Networks using Tsetlin Automata on FPGAs
+
+- **ID:** `2603.24186` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Shengyu Duan, Marcos L. L. Sartori, Rishad Shafik, et al.
+- **abs:** https://arxiv.org/abs/2603.24186 · **PDF:** https://arxiv.org/pdf/2603.24186v1
+
+## 2068. Walma: Learning to See Memory Corruption in WebAssembly
+
+- **ID:** `2603.24167` · **primary:** `cs.CR` · **Submitted:** 2026-03-25
+- **Authors:** Oussama Draissi, Mark Günzel, Ahmad-Reza Sadeghi, et al.
+- **abs:** https://arxiv.org/abs/2603.24167 · **PDF:** https://arxiv.org/pdf/2603.24167v1
+
+## 2069. A visual observation on the geometry of UMAP projections of the difference vectors of antonym and synonym word pair embeddings
+
+- **ID:** `2603.24150` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Rami Luisto
+- **abs:** https://arxiv.org/abs/2603.24150 · **PDF:** https://arxiv.org/pdf/2603.24150v1
+
+## 2070. Linear-Nonlinear Fusion Neural Operator for Partial Differential Equations
+
+- **ID:** `2603.24143` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Heng Wu, Junjie Wang, Benzhuo Lu
+- **abs:** https://arxiv.org/abs/2603.24143 · **PDF:** https://arxiv.org/pdf/2603.24143v1
+
+## 2071. A Longitudinal Analysis of the CEC Single-Objective Competitions (2010-2024) and Implications for Variational Quantum Optimization
+
+- **ID:** `2603.24140` · **primary:** `quant-ph` · **Submitted:** 2026-03-25
+- **Authors:** Vojtěch Novák, Tomáš Bezděk, Ivan Zelinka, et al.
+- **abs:** https://arxiv.org/abs/2603.24140 · **PDF:** https://arxiv.org/pdf/2603.24140v1
+
+## 2072. Tutor-Student Reinforcement Learning: A Dynamic Curriculum for Robust Deepfake Detection
+
+- **ID:** `2603.24139` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Zhanhe Lei, Zhongyuan Wang, Jikang Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.24139 · **PDF:** https://arxiv.org/pdf/2603.24139v1
+
+## 2073. Efficient Controller Learning from Human Preferences and Numerical Data Via Multi-Modal Surrogate Models
+
+- **ID:** `2603.24138` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Lukas Theiner, Maik Pfefferkorn, Yongpeng Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.24138 · **PDF:** https://arxiv.org/pdf/2603.24138v1
+
+## 2074. MedAidDialog: A Multilingual Multi-Turn Medical Dialogue Dataset for Accessible Healthcare
+
+- **ID:** `2603.24132` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Shubham Kumar Nigam, Suparnojit Sarkar, Piyush Patel
+- **abs:** https://arxiv.org/abs/2603.24132 · **PDF:** https://arxiv.org/pdf/2603.24132v1
+
+## 2075. Reservoir-Based Graph Convolutional Networks
+
+- **ID:** `2603.24131` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Mayssa Soussia, Gita Ayu Salsabila, Mohamed Ali Mahjoub, et al.
+- **abs:** https://arxiv.org/abs/2603.24131 · **PDF:** https://arxiv.org/pdf/2603.24131v1
+
+## 2076. DyMRL: Dynamic Multispace Representation Learning for Multimodal Event Forecasting in Knowledge Graph
+
+- **ID:** `2603.24636` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Feng Zhao, Kangzheng Liu, Teng Peng, et al.
+- **abs:** https://arxiv.org/abs/2603.24636 · **PDF:** https://arxiv.org/pdf/2603.24636v1
+
+## 2077. On Gossip Algorithms for Machine Learning with Pairwise Objectives
+
+- **ID:** `2603.24128` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Igor Colin, Aurélien Bellet, Stephan Clémençon, et al.
+- **abs:** https://arxiv.org/abs/2603.24128 · **PDF:** https://arxiv.org/pdf/2603.24128v1
+
+## 2078. Likelihood hacking in probabilistic program synthesis
+
+- **ID:** `2603.24126` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Jacek Karwowski, Younesse Kaddar, Zihuiwen Ye, et al.
+- **abs:** https://arxiv.org/abs/2603.24126 · **PDF:** https://arxiv.org/pdf/2603.24126v1
+
+## 2079. Alignment Reduces Expressed but Not Encoded Gender Bias: A Unified Framework and Study
+
+- **ID:** `2603.24125` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Nour Bouchouchi, Thiabult Laugel, Xavier Renard, et al.
+- **abs:** https://arxiv.org/abs/2603.24125 · **PDF:** https://arxiv.org/pdf/2603.24125v1
+
+## 2080. The Alignment Tax: Response Homogenization in Aligned LLMs and Its Implications for Uncertainty Estimation
+
+- **ID:** `2603.24124` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Mingyi Liu
+- **abs:** https://arxiv.org/abs/2603.24124 · **PDF:** https://arxiv.org/pdf/2603.24124v2
+
+## 2081. Mixed-signal implementation of feedback-control optimizer for single-layer Spiking Neural Networks
+
+- **ID:** `2603.24113` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Jonathan Haag, Christian Metzner, Dmitrii Zendrikov, et al.
+- **abs:** https://arxiv.org/abs/2603.24113 · **PDF:** https://arxiv.org/pdf/2603.24113v1
+
+## 2082. Toward a Multi-Layer ML-Based Security Framework for Industrial IoT
+
+- **ID:** `2603.24111` · **primary:** `cs.CR` · **Submitted:** 2026-03-25
+- **Authors:** Aymen Bouferroum, Valeria Loscri, Abderrahim Benslimane
+- **abs:** https://arxiv.org/abs/2603.24111 · **PDF:** https://arxiv.org/pdf/2603.24111v2
+
+## 2083. Comparative analysis of dual-form networks for live land monitoring using multi-modal satellite image time series
+
+- **ID:** `2603.24109` · **primary:** `eess.IV` · **Submitted:** 2026-03-25
+- **Authors:** Iris Dumeur, Jérémy Anger, Gabriele Facciolo
+- **abs:** https://arxiv.org/abs/2603.24109 · **PDF:** https://arxiv.org/pdf/2603.24109v1
+
+## 2084. Causality-Driven Disentangled Representation Learning in Multiplex Graphs
+
+- **ID:** `2603.24105` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Saba Nasiri, Selin Aviyente, Dorina Thanou
+- **abs:** https://arxiv.org/abs/2603.24105 · **PDF:** https://arxiv.org/pdf/2603.24105v1
+
+## 2085. KCLNet: Electrically Equivalence-Oriented Graph Representation Learning for Analog Circuits
+
+- **ID:** `2603.24101` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Peng Xu, Yapeng Li, Tinghuan Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.24101 · **PDF:** https://arxiv.org/pdf/2603.24101v1
+
+## 2086. Towards Effective Experiential Learning: Dual Guidance for Utilization and Internalization
+
+- **ID:** `2603.24093` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Fei Bai, Zhipeng Chen, Chuan Hao, et al.
+- **abs:** https://arxiv.org/abs/2603.24093 · **PDF:** https://arxiv.org/pdf/2603.24093v1
+
+## 2087. Dual-Graph Multi-Agent Reinforcement Learning for Handover Optimization
+
+- **ID:** `2603.24634` · **primary:** `cs.NI` · **Submitted:** 2026-03-25
+- **Authors:** Matteo Salvatori, Filippo Vannella, Sebastian Macaluso, et al.
+- **abs:** https://arxiv.org/abs/2603.24634 · **PDF:** https://arxiv.org/pdf/2603.24634v1
+
+## 2088. Bridging the Evaluation Gap: Standardized Benchmarks for Multi-Objective Search
+
+- **ID:** `2603.24084` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Hadar Peer, Carlos Hernandez, Sven Koenig, et al.
+- **abs:** https://arxiv.org/abs/2603.24084 · **PDF:** https://arxiv.org/pdf/2603.24084v1
+
+## 2089. Knowledge-Guided Manipulation Using Multi-Task Reinforcement Learning
+
+- **ID:** `2603.24083` · **primary:** `cs.RO` · **Submitted:** 2026-03-25
+- **Authors:** Aditya Narendra, Mukhammadrizo Maribjonov, Dmitry Makarov, et al.
+- **abs:** https://arxiv.org/abs/2603.24083 · **PDF:** https://arxiv.org/pdf/2603.24083v1
+
+## 2090. LLMpedia: A Transparent Framework to Materialize an LLM's Encyclopedic Knowledge at Scale
+
+- **ID:** `2603.24080` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Muhammed Saeed, Simon Razniewski
+- **abs:** https://arxiv.org/abs/2603.24080 · **PDF:** https://arxiv.org/pdf/2603.24080v1
+
+## 2091. When Understanding Becomes a Risk: Authenticity and Safety Risks in the Emerging Image Generation Paradigm
+
+- **ID:** `2603.24079` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Ye Leng, Junjie Chu, Mingjie Li, et al.
+- **abs:** https://arxiv.org/abs/2603.24079 · **PDF:** https://arxiv.org/pdf/2603.24079v1
+
+## 2092. The impact of sensor placement on graph-neural-network-based leakage detection
+
+- **ID:** `2603.24076` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** J. J. H. van Gemert, V. Breschi, D. R. Yntema, et al.
+- **abs:** https://arxiv.org/abs/2603.24076 · **PDF:** https://arxiv.org/pdf/2603.24076v1
+
+## 2093. ConceptKT: A Benchmark for Concept-Level Deficiency Prediction in Knowledge Tracing
+
+- **ID:** `2603.24073` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Yu-Chen Kang, Yu-Chien Tang, An-Zi Yen
+- **abs:** https://arxiv.org/abs/2603.24073 · **PDF:** https://arxiv.org/pdf/2603.24073v1
+
+## 2094. Enhanced Mycelium of Thought (EMoT): A Bio-Inspired Hierarchical Reasoning Architecture with Strategic Dormancy and Mnemonic Encoding
+
+- **ID:** `2603.24065` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Florian Odi Stummer
+- **abs:** https://arxiv.org/abs/2603.24065 · **PDF:** https://arxiv.org/pdf/2603.24065v1
+
+## 2095. A Step Toward Federated Pretraining of Multimodal Large Language Models
+
+- **ID:** `2603.26786` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Baochen Xiong, Yifan Xu, Xiaoshan Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.26786 · **PDF:** https://arxiv.org/pdf/2603.26786v1
+
+## 2096. Mitigating Object Hallucinations in LVLMs via Attention Imbalance Rectification
+
+- **ID:** `2603.24058` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Han Sun, Qin Li, Peixin Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.24058 · **PDF:** https://arxiv.org/pdf/2603.24058v1
+
+## 2097. Hierarchical Spatial-Temporal Graph-Enhanced Model for Map-Matching
+
+- **ID:** `2603.24054` · **primary:** `cs.DB` · **Submitted:** 2026-03-25
+- **Authors:** Anjun Gao, Zhenglin Wan, Pingfu Chao, et al.
+- **abs:** https://arxiv.org/abs/2603.24054 · **PDF:** https://arxiv.org/pdf/2603.24054v1
+
+## 2098. FinToolSyn: A forward synthesis Framework for Financial Tool-Use Dialogue Data with Dynamic Tool Retrieval
+
+- **ID:** `2603.24051` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Caishuang Huang, Yang Qiao, Rongyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.24051 · **PDF:** https://arxiv.org/pdf/2603.24051v1
+
+## 2099. MoE-Sieve: Routing-Guided LoRA for Efficient MoE Fine-Tuning
+
+- **ID:** `2603.24044` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Andrea Manzoni
+- **abs:** https://arxiv.org/abs/2603.24044 · **PDF:** https://arxiv.org/pdf/2603.24044v1
+
+## 2100. Minimal Sufficient Representations for Self-interpretable Deep Neural Networks
+
+- **ID:** `2603.24041` · **primary:** `stat.ME` · **Submitted:** 2026-03-25
+- **Authors:** Zhiyao Tan, Liu Li, Huazhen Lin
+- **abs:** https://arxiv.org/abs/2603.24041 · **PDF:** https://arxiv.org/pdf/2603.24041v1
+
+## 2101. From Oracle to Noisy Context: Mitigating Contextual Exposure Bias in Speech-LLMs
+
+- **ID:** `2603.24034` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Xiaoyong Guo, Nanjie Li, Zijie Zeng, et al.
+- **abs:** https://arxiv.org/abs/2603.24034 · **PDF:** https://arxiv.org/pdf/2603.24034v1
+
+## 2102. Lagrangian Relaxation Score-based Generation for Mixed Integer linear Programming
+
+- **ID:** `2603.24033` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Ruobing Wang, Xin Li, Yujie Fang, et al.
+- **abs:** https://arxiv.org/abs/2603.24033 · **PDF:** https://arxiv.org/pdf/2603.24033v1
+
+## 2103. i-IF-Learn: Iterative Feature Selection and Unsupervised Learning for High-Dimensional Complex Data
+
+- **ID:** `2603.24025` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Chen Ma, Wanjie Wang, Shuhao Fan
+- **abs:** https://arxiv.org/abs/2603.24025 · **PDF:** https://arxiv.org/pdf/2603.24025v1
+
+## 2104. Schema on the Inside: A Two-Phase Fine-Tuning Method for High-Efficiency Text-to-SQL at Scale
+
+- **ID:** `2603.24023` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Chinmay Soni, Shivam Chourasia, Gaurav Kumar, et al.
+- **abs:** https://arxiv.org/abs/2603.24023 · **PDF:** https://arxiv.org/pdf/2603.24023v1
+
+## 2105. ELITE: Experiential Learning and Intent-Aware Transfer for Self-improving Embodied Agents
+
+- **ID:** `2603.24018` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Bingqing Wei, Zhongyu Xia, Dingai Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.24018 · **PDF:** https://arxiv.org/pdf/2603.24018v1
+
+## 2106. COVTrack++: Learning Open-Vocabulary Multi-Object Tracking from Continuous Videos via a Synergistic Paradigm
+
+- **ID:** `2603.24016` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Zekun Qian, Wei Feng, Ruize Han, et al.
+- **abs:** https://arxiv.org/abs/2603.24016 · **PDF:** https://arxiv.org/pdf/2603.24016v1
+
+## 2107. Language-Grounded Multi-Agent Planning for Personalized and Fair Participatory Urban Sensing
+
+- **ID:** `2603.24014` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Xusen Guo, Mingxing Peng, Hongliang Lu, et al.
+- **abs:** https://arxiv.org/abs/2603.24014 · **PDF:** https://arxiv.org/pdf/2603.24014v1
+
+## 2108. CVPD at QIAS 2026: RAG-Guided LLM Reasoning for Al-Mawarith Share Computation and Heir Allocation
+
+- **ID:** `2603.24012` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Wassim Swaileh, Mohammed-En-Nadhir Zighem, Hichem Telli, et al.
+- **abs:** https://arxiv.org/abs/2603.24012 · **PDF:** https://arxiv.org/pdf/2603.24012v2
+
+## 2109. Thinking with Tables: Enhancing Multi-Modal Tabular Understanding via Neuro-Symbolic Reasoning
+
+- **ID:** `2603.24004` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Kun-Yang Yu, Zhi Zhou, Shi-Yu Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.24004 · **PDF:** https://arxiv.org/pdf/2603.24004v1
+
+## 2110. Stochastic Dimension-Free Zeroth-Order Estimator for High-Dimensional and High-Order PINNs
+
+- **ID:** `2603.24002` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Zhangyong Liang, Ji Zhang
+- **abs:** https://arxiv.org/abs/2603.24002 · **PDF:** https://arxiv.org/pdf/2603.24002v1
+
+## 2111. Sparse Growing Transformer: Training-Time Sparse Depth Allocation via Progressive Attention Looping
+
+- **ID:** `2603.23998` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Yao Chen, Yilong Chen, Yinqi Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.23998 · **PDF:** https://arxiv.org/pdf/2603.23998v1
+
+## 2112. Understanding the Challenges in Iterative Generative Optimization with LLMs
+
+- **ID:** `2603.23994` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Allen Nie, Xavier Daull, Zhiyi Kuang, et al.
+- **abs:** https://arxiv.org/abs/2603.23994 · **PDF:** https://arxiv.org/pdf/2603.23994v1
+
+## 2113. From Untamed Black Box to Interpretable Pedagogical Orchestration: The Ensemble of Specialized LLMs Architecture for Adaptive Tutoring
+
+- **ID:** `2603.23990` · **primary:** `cs.CY` · **Submitted:** 2026-03-25
+- **Authors:** Nizam Kadir
+- **abs:** https://arxiv.org/abs/2603.23990 · **PDF:** https://arxiv.org/pdf/2603.23990v1
+
+## 2114. CoCR-RAG: Enhancing Retrieval-Augmented Generation in Web Q&A via Concept-oriented Context Reconstruction
+
+- **ID:** `2603.23989` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Kaize Shi, Xueyao Sun, Qika Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.23989 · **PDF:** https://arxiv.org/pdf/2603.23989v1
+
+## 2115. Can we generate portable representations for clinical time series data using LLMs?
+
+- **ID:** `2603.23987` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Zongliang Ji, Yifei Sun, Andre Amaral, et al.
+- **abs:** https://arxiv.org/abs/2603.23987 · **PDF:** https://arxiv.org/pdf/2603.23987v1
+
+## 2116. Diet Your LLM: Dimension-wise Global Pruning of LLMs via Merging Task-specific Importance Score
+
+- **ID:** `2603.23985` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Jimyung Hong, Jaehyung Kim
+- **abs:** https://arxiv.org/abs/2603.23985 · **PDF:** https://arxiv.org/pdf/2603.23985v1
+
+## 2117. Transcending Classical Neural Network Boundaries: A Quantum-Classical Synergistic Paradigm for Seismic Data Processing
+
+- **ID:** `2603.23984` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Zhengyi Yuan, Xintong Dong, Xinyang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.23984 · **PDF:** https://arxiv.org/pdf/2603.23984v1
+
+## 2118. SafeFlow: Real-Time Text-Driven Humanoid Whole-Body Control via Physics-Guided Rectified Flow and Selective Safety Gating
+
+- **ID:** `2603.23983` · **primary:** `cs.RO` · **Submitted:** 2026-03-25
+- **Authors:** Hanbyel Cho, Sang-Hun Kim, Jeonguk Kang, et al.
+- **abs:** https://arxiv.org/abs/2603.23983 · **PDF:** https://arxiv.org/pdf/2603.23983v1
+
+## 2119. Kirchhoff-Inspired Neural Networks for Evolving High-Order Perception
+
+- **ID:** `2603.23977` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Tongfei Chen, Jingying Yang, Linlin Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.23977 · **PDF:** https://arxiv.org/pdf/2603.23977v1
+
+## 2120. Machine vision with small numbers of detected photons per inference
+
+- **ID:** `2603.23974` · **primary:** `physics.optics` · **Submitted:** 2026-03-25
+- **Authors:** Shi-Yuan Ma, Jérémie Laydevant, Mandar M. Sohoni, et al.
+- **abs:** https://arxiv.org/abs/2603.23974 · **PDF:** https://arxiv.org/pdf/2603.23974v1
+
+## 2121. Grounding Arabic LLMs in the Doha Historical Dictionary: Retrieval-Augmented Understanding of Quran and Hadith
+
+- **ID:** `2603.23972` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Somaya Eltanbouly, Samer Rashwani
+- **abs:** https://arxiv.org/abs/2603.23972 · **PDF:** https://arxiv.org/pdf/2603.23972v1
+
+## 2122. The Price Reversal Phenomenon: When Cheaper Reasoning Models End Up Costing More
+
+- **ID:** `2603.23971` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Lingjiao Chen, Chi Zhang, Yeye He, et al.
+- **abs:** https://arxiv.org/abs/2603.23971 · **PDF:** https://arxiv.org/pdf/2603.23971v1
+
+## 2123. Can We Change the Stroke Size for Easier Diffusion?
+
+- **ID:** `2603.26783` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Yunwei Bai, Ying Kiat Tan, Yao Shu, et al.
+- **abs:** https://arxiv.org/abs/2603.26783 · **PDF:** https://arxiv.org/pdf/2603.26783v1
+
+## 2124. Wireless communication empowers online scheduling of partially-observable transportation multi-robot systems in a smart factory
+
+- **ID:** `2603.23967` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Yaxin Liao, Qimei Cui, Kwang-Cheng Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.23967 · **PDF:** https://arxiv.org/pdf/2603.23967v1
+
+## 2125. Policy-Guided Threat Hunting: An LLM enabled Framework with Splunk SOC Triage
+
+- **ID:** `2603.23966` · **primary:** `cs.CR` · **Submitted:** 2026-03-25
+- **Authors:** Rishikesh Sahay, Bell Eapen, Weizhi Meng, et al.
+- **abs:** https://arxiv.org/abs/2603.23966 · **PDF:** https://arxiv.org/pdf/2603.23966v2
+
+## 2126. From Pixels to Digital Agents: An Empirical Study on the Taxonomy and Technological Trends of Reinforcement Learning Environments
+
+- **ID:** `2603.23964` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Lijing Luo, Yiben Luo, Alexey Gorbatovski, et al.
+- **abs:** https://arxiv.org/abs/2603.23964 · **PDF:** https://arxiv.org/pdf/2603.23964v1
+
+## 2127. GRMLR: Knowledge-Enhanced Small-Data Learning for Deep-Sea Cold Seep Stage Inference
+
+- **ID:** `2603.23961` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Chenxu Zhou, Zelin Liu, Rui Cai, et al.
+- **abs:** https://arxiv.org/abs/2603.23961 · **PDF:** https://arxiv.org/pdf/2603.23961v1
+
+## 2128. KANEL: Kolmogorov-Arnold Network Ensemble Learning Enables Early Hit Enrichment in High-Throughput Virtual Screening
+
+- **ID:** `2603.25755` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-25
+- **Authors:** Pavel Koptev, Nikita Krainov, Konstantin Malkov, et al.
+- **abs:** https://arxiv.org/abs/2603.25755 · **PDF:** https://arxiv.org/pdf/2603.25755v1
+
+## 2129. TRAJEVAL: Decomposing Code Agent Trajectories for Fine-Grained Diagnosis
+
+- **ID:** `2603.24631` · **primary:** `cs.SE` · **Submitted:** 2026-03-25
+- **Authors:** Myeongsoo Kim, Dingmin Wang, Siwei Cui, et al.
+- **abs:** https://arxiv.org/abs/2603.24631 · **PDF:** https://arxiv.org/pdf/2603.24631v1
+
+## 2130. From AI Assistant to AI Scientist: Autonomous Discovery of LLM-RL Algorithms with LLM Agents
+
+- **ID:** `2603.23951` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Sirui Xia, Yikai Zhang, Aili Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.23951 · **PDF:** https://arxiv.org/pdf/2603.23951v1
+
+## 2131. Argument Mining as a Text-to-Text Generation Task
+
+- **ID:** `2603.23949` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Masayuki Kawarada, Tsutomu Hirao, Wataru Uchida, et al.
+- **abs:** https://arxiv.org/abs/2603.23949 · **PDF:** https://arxiv.org/pdf/2603.23949v1
+
+## 2132. Variable-Length Audio Fingerprinting
+
+- **ID:** `2603.23947` · **primary:** `cs.SD` · **Submitted:** 2026-03-25
+- **Authors:** Hongjie Chen, Hanyu Meng, Huimin Zeng, et al.
+- **abs:** https://arxiv.org/abs/2603.23947 · **PDF:** https://arxiv.org/pdf/2603.23947v1
+
+## 2133. ChargeFlow: Flow-Matching Refinement of Charge-Conditioned Electron Densities
+
+- **ID:** `2603.23943` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-25
+- **Authors:** Tri Minh Nguyen, Sherif Abdulkader Tawfik, Truyen Tran, et al.
+- **abs:** https://arxiv.org/abs/2603.23943 · **PDF:** https://arxiv.org/pdf/2603.23943v1
+
+## 2134. High-Fidelity Face Content Recovery via Tamper-Resilient Versatile Watermarking
+
+- **ID:** `2603.23940` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Peipeng Yu, Jinfeng Xie, Chengfu Ou, et al.
+- **abs:** https://arxiv.org/abs/2603.23940 · **PDF:** https://arxiv.org/pdf/2603.23940v1
+
+## 2135. OmniACBench: A Benchmark for Evaluating Context-Grounded Acoustic Control in Omni-Modal Models
+
+- **ID:** `2603.23938` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Seunghee Kim, Bumkyu Park, Kyudan Jung, et al.
+- **abs:** https://arxiv.org/abs/2603.23938 · **PDF:** https://arxiv.org/pdf/2603.23938v1
+
+## 2136. Dialogue to Question Generation for Evidence-based Medical Guideline Agent Development
+
+- **ID:** `2603.23937` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Zongliang Ji, Ziyang Zhang, Xincheng Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.23937 · **PDF:** https://arxiv.org/pdf/2603.23937v1
+
+## 2137. Multiverse: Language-Conditioned Multi-Game Level Blending via Shared Representation
+
+- **ID:** `2603.26782` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** In-Chang Baek, Jiyun Jung, Geum-Hwan Hwang, et al.
+- **abs:** https://arxiv.org/abs/2603.26782 · **PDF:** https://arxiv.org/pdf/2603.26782v2
+
+## 2138. Revealing Multi-View Hallucination in Large Vision-Language Models
+
+- **ID:** `2603.23934` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Wooje Park, Insu Lee, Soohyun Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.23934 · **PDF:** https://arxiv.org/pdf/2603.23934v1
+
+## 2139. ORACLE: Orchestrate NPC Daily Activities using Contrastive Learning with Transformer-CVAE
+
+- **ID:** `2603.23933` · **primary:** `cs.GR` · **Submitted:** 2026-03-25
+- **Authors:** Seong-Eun Hong, JuYeong Hwang, RyunHa Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.23933 · **PDF:** https://arxiv.org/pdf/2603.23933v1
+
+## 2140. Optimal Variance-Dependent Regret Bounds for Infinite-Horizon MDPs
+
+- **ID:** `2603.23926` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Guy Zamir, Matthew Zurek, Yudong Chen
+- **abs:** https://arxiv.org/abs/2603.23926 · **PDF:** https://arxiv.org/pdf/2603.23926v1
+
+## 2141. Elements of Conformal Prediction for Statisticians
+
+- **ID:** `2603.23923` · **primary:** `stat.ME` · **Submitted:** 2026-03-25
+- **Authors:** Matteo Sesia, Stefano Favaro
+- **abs:** https://arxiv.org/abs/2603.23923 · **PDF:** https://arxiv.org/pdf/2603.23923v1
+
+## 2142. DecepGPT: Schema-Driven Deception Detection with Multicultural Datasets and Robust Multimodal Learning
+
+- **ID:** `2603.23916` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Jiajian Huang, Dongliang Zhu, Zitong YU, et al.
+- **abs:** https://arxiv.org/abs/2603.23916 · **PDF:** https://arxiv.org/pdf/2603.23916v1
+
+## 2143. Attention-aware Inference Optimizations for Large Vision-Language Models with Memory-efficient Decoding
+
+- **ID:** `2603.23914` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Fatih Ilhan, Gaowen Liu, Ramana Rao Kompella, et al.
+- **abs:** https://arxiv.org/abs/2603.23914 · **PDF:** https://arxiv.org/pdf/2603.23914v1
+
+## 2144. Self-Distillation for Multi-Token Prediction
+
+- **ID:** `2603.23911` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Guoliang Zhao, Ruobing Xie, An Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.23911 · **PDF:** https://arxiv.org/pdf/2603.23911v1
+
+## 2145. AnalogAgent: Self-Improving Analog Circuit Design Automation with LLM Agents
+
+- **ID:** `2603.23910` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Zhixuan Bao, Zhuoyi Lin, Jiageng Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.23910 · **PDF:** https://arxiv.org/pdf/2603.23910v1
+
+## 2146. DUPLEX: Agentic Dual-System Planning via LLM-Driven Information Extraction
+
+- **ID:** `2603.23909` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Keru Hua, Ding Wang, Yaoying Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.23909 · **PDF:** https://arxiv.org/pdf/2603.23909v1
+
+## 2147. Latent Bias Alignment for High-Fidelity Diffusion Inversion in Real-World Image Reconstruction and Manipulation
+
+- **ID:** `2603.23903` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Weiming Chen, Qifan Liu, Siyi Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.23903 · **PDF:** https://arxiv.org/pdf/2603.23903v1
+
+## 2148. Knowledge-Refined Dual Context-Aware Network for Partially Relevant Video Retrieval
+
+- **ID:** `2603.23902` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Junkai Yang, Qirui Wang, Yaoqing Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.23902 · **PDF:** https://arxiv.org/pdf/2603.23902v1
+
+## 2149. SM-Net: Learning a Continuous Spectral Manifold from Multiple Stellar Libraries
+
+- **ID:** `2603.23899` · **primary:** `astro-ph.IM` · **Submitted:** 2026-03-25
+- **Authors:** Omar Anwar, Aaron S. G. Robotham, Luca Cortese, et al.
+- **abs:** https://arxiv.org/abs/2603.23899 · **PDF:** https://arxiv.org/pdf/2603.23899v2
+
+## 2150. Sketch2Simulation: Automating Flowsheet Generation via Multi Agent Large Language Models
+
+- **ID:** `2603.24629` · **primary:** `cs.SE` · **Submitted:** 2026-03-25
+- **Authors:** Abdullah Bahamdan, Emma Pajak, John D. Hedengren, et al.
+- **abs:** https://arxiv.org/abs/2603.24629 · **PDF:** https://arxiv.org/pdf/2603.24629v1
+
+## 2151. Praxium: Diagnosing Cloud Anomalies with AI-based Telemetry and Dependency Analysis
+
+- **ID:** `2603.23890` · **primary:** `cs.SE` · **Submitted:** 2026-03-25
+- **Authors:** Rohan Kumar, Jason Li, Zongshun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.23890 · **PDF:** https://arxiv.org/pdf/2603.23890v1
+
+## 2152. Off-Policy Safe Reinforcement Learning with Constrained Optimistic Exploration
+
+- **ID:** `2603.23889` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Guopeng Li, Matthijs T. J. Spaan, Julian F. P. Kooij
+- **abs:** https://arxiv.org/abs/2603.23889 · **PDF:** https://arxiv.org/pdf/2603.23889v1
+
+## 2153. AgentChemist: A Multi-Agent Experimental Robotic Platform Integrating Chemical Perception and Precise Control
+
+- **ID:** `2603.23886` · **primary:** `cs.RO` · **Submitted:** 2026-03-25
+- **Authors:** Xiangyi Wei, Fei Wang, Haotian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.23886 · **PDF:** https://arxiv.org/pdf/2603.23886v1
+
+## 2154. Efficient Encrypted Computation in Convolutional Spiking Neural Networks with TFHE
+
+- **ID:** `2603.26781` · **primary:** `cs.CR` · **Submitted:** 2026-03-25
+- **Authors:** Longfei Guo, Pengbo Li, Ting Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.26781 · **PDF:** https://arxiv.org/pdf/2603.26781v1
+
+## 2155. The Luna Bound Propagator for Formal Analysis of Neural Networks
+
+- **ID:** `2603.23878` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Henry LeCates, Haoze Wu
+- **abs:** https://arxiv.org/abs/2603.23878 · **PDF:** https://arxiv.org/pdf/2603.23878v1
+
+## 2156. Isomorphic Functionalities between Ant Colony and Ensemble Learning: Part II-On the Strength of Weak Learnability and the Boosting Paradigm
+
+- **ID:** `2604.00038` · **primary:** `stat.ML` · **Submitted:** 2026-03-25
+- **Authors:** Ernest Fokoué, Gregory Babbitt, Yuval Levental
+- **abs:** https://arxiv.org/abs/2604.00038 · **PDF:** https://arxiv.org/pdf/2604.00038v1
+
+## 2157. The DeepXube Software Package for Solving Pathfinding Problems with Learned Heuristic Functions and Search
+
+- **ID:** `2603.23873` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Forest Agostinelli
+- **abs:** https://arxiv.org/abs/2603.23873 · **PDF:** https://arxiv.org/pdf/2603.23873v1
+
+## 2158. HDPO: Hybrid Distillation Policy Optimization via Privileged Self-Distillation
+
+- **ID:** `2603.23871` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Ken Ding
+- **abs:** https://arxiv.org/abs/2603.23871 · **PDF:** https://arxiv.org/pdf/2603.23871v1
+
+## 2159. Can VLMs Reason Robustly? A Neuro-Symbolic Investigation
+
+- **ID:** `2603.23867` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Weixin Chen, Antonio Vergari, Han Zhao
+- **abs:** https://arxiv.org/abs/2603.23867 · **PDF:** https://arxiv.org/pdf/2603.23867v1
+
+## 2160. A Large-Scale Comparative Analysis of Imputation Methods for Single-Cell RNA Sequencing Data
+
+- **ID:** `2603.24626` · **primary:** `q-bio.GN` · **Submitted:** 2026-03-25
+- **Authors:** Yuichiro Iwashita, Ahtisham Fazeel Abbasi, Muhammad Nabeel Asim, et al.
+- **abs:** https://arxiv.org/abs/2603.24626 · **PDF:** https://arxiv.org/pdf/2603.24626v1
+
+## 2161. Generative AI User Experience: Developing Human--AI Epistemic Partnership
+
+- **ID:** `2603.23863` · **primary:** `cs.CY` · **Submitted:** 2026-03-25
+- **Authors:** Xiaoming Zhai
+- **abs:** https://arxiv.org/abs/2603.23863 · **PDF:** https://arxiv.org/pdf/2603.23863v1
+
+## 2162. Deep Convolutional Neural Networks for predicting highest priority functional group in organic molecules
+
+- **ID:** `2603.23862` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Kunal Khatri, Vineet Mehta
+- **abs:** https://arxiv.org/abs/2603.23862 · **PDF:** https://arxiv.org/pdf/2603.23862v1
+
+## 2163. An Invariant Compiler for Neural ODEs in AI-Accelerated Scientific Simulation
+
+- **ID:** `2603.23861` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Fangzhou Yu, Yiqi Su, Ray Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.23861 · **PDF:** https://arxiv.org/pdf/2603.23861v1
+
+## 2164. Why the Maximum Second Derivative of Activations Matters for Adversarial Robustness
+
+- **ID:** `2603.23860` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Yunrui Yu, Hang Su, Jun Zhu
+- **abs:** https://arxiv.org/abs/2603.23860 · **PDF:** https://arxiv.org/pdf/2603.23860v1
+
+## 2165. When AI output tips to bad but nobody notices: Legal implications of AI's mistakes
+
+- **ID:** `2603.23857` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Dylan J. Restrepo, Nicholas J. Restrepo, Frank Y. Huo, et al.
+- **abs:** https://arxiv.org/abs/2603.23857 · **PDF:** https://arxiv.org/pdf/2603.23857v1
+
+## 2166. Symbolic--KAN: Kolmogorov-Arnold Networks with Discrete Symbolic Structure for Interpretable Learning
+
+- **ID:** `2603.23854` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Salah A Faroughi, Farinaz Mostajeran, Amirhossein Arzani, et al.
+- **abs:** https://arxiv.org/abs/2603.23854 · **PDF:** https://arxiv.org/pdf/2603.23854v1
+
+## 2167. SCoOP: Semantic Consistent Opinion Pooling for Uncertainty Quantification in Multiple Vision-Language Model Systems
+
+- **ID:** `2603.23853` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Chung-En Johnny Yu, Brian Jalaian, Nathaniel D. Bastian
+- **abs:** https://arxiv.org/abs/2603.23853 · **PDF:** https://arxiv.org/pdf/2603.23853v2
+
+## 2168. BeliefShift: Benchmarking Temporal Belief Consistency and Opinion Drift in LLM Agents
+
+- **ID:** `2603.23848` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Praveen Kumar Myakala, Manan Agrawal, Rahul Manche
+- **abs:** https://arxiv.org/abs/2603.23848 · **PDF:** https://arxiv.org/pdf/2603.23848v1
+
+## 2169. Language Model Planners do not Scale, but do Formalizers?
+
+- **ID:** `2603.23844` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Owen Jiang, Cassie Huang, Ashish Sabharwal, et al.
+- **abs:** https://arxiv.org/abs/2603.23844 · **PDF:** https://arxiv.org/pdf/2603.23844v1
+
+## 2170. PoliticsBench: Benchmarking Political Values in Large Language Models with Multi-Turn Roleplay
+
+- **ID:** `2603.23841` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Rohan Khetan, Ashna Khetan
+- **abs:** https://arxiv.org/abs/2603.23841 · **PDF:** https://arxiv.org/pdf/2603.23841v1
+
+## 2171. VehicleMemBench: An Executable Benchmark for Multi-User Long-Term Memory in In-Vehicle Agents
+
+- **ID:** `2603.23840` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Yuhao Chen, Yi Xu, Xinyun Ding, et al.
+- **abs:** https://arxiv.org/abs/2603.23840 · **PDF:** https://arxiv.org/pdf/2603.23840v1
+
+## 2172. Learning-guided Prioritized Planning for Lifelong Multi-Agent Path Finding in Warehouse Automation
+
+- **ID:** `2603.23838` · **primary:** `cs.AI` · **Submitted:** 2026-03-25
+- **Authors:** Han Zheng, Yining Ma, Brandon Araki, et al.
+- **abs:** https://arxiv.org/abs/2603.23838 · **PDF:** https://arxiv.org/pdf/2603.23838v1
+
+## 2173. Beyond Consistency: Inference for the Relative risk functional in Deep Nonparametric Cox Models
+
+- **ID:** `2603.23835` · **primary:** `stat.ML` · **Submitted:** 2026-03-25
+- **Authors:** Sattwik Ghosal, Xuran Meng, Yi Li
+- **abs:** https://arxiv.org/abs/2603.23835 · **PDF:** https://arxiv.org/pdf/2603.23835v1
+
+## 2174. Unveiling Hidden Convexity in Deep Learning: a Sparse Signal Processing Perspective
+
+- **ID:** `2603.23831` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Emi Zeger, Mert Pilanci
+- **abs:** https://arxiv.org/abs/2603.23831 · **PDF:** https://arxiv.org/pdf/2603.23831v1
+
+## 2175. Limits of Imagery Reasoning in Frontier LLM Models
+
+- **ID:** `2603.26779` · **primary:** `cs.CV` · **Submitted:** 2026-03-25
+- **Authors:** Sergio Y. Hayashi, Nina S. T. Hirata
+- **abs:** https://arxiv.org/abs/2603.26779 · **PDF:** https://arxiv.org/pdf/2603.26779v1
+
+## 2176. Circuit Complexity of Hierarchical Knowledge Tracing and Implications for Log-Precision Transformers
+
+- **ID:** `2603.23823` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Naiming Liu, Richard Baraniuk, Shashank Sonkar
+- **abs:** https://arxiv.org/abs/2603.23823 · **PDF:** https://arxiv.org/pdf/2603.23823v1
+
+## 2177. How Vulnerable Are Edge LLMs?
+
+- **ID:** `2603.23822` · **primary:** `cs.CR` · **Submitted:** 2026-03-25
+- **Authors:** Ao Ding, Hongzong Li, Zi Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.23822 · **PDF:** https://arxiv.org/pdf/2603.23822v1
+
+## 2178. Perturbation: A simple and efficient adversarial tracer for representation learning in language models
+
+- **ID:** `2603.23821` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Joshua Rozner, Cory Shain
+- **abs:** https://arxiv.org/abs/2603.23821 · **PDF:** https://arxiv.org/pdf/2603.23821v1
+
+## 2179. TED: Training-Free Experience Distillation for Multimodal Reasoning
+
+- **ID:** `2603.26778` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Shuozhi Yuan, Jinqing Wang, Zihao Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.26778 · **PDF:** https://arxiv.org/pdf/2603.26778v1
+
+## 2180. Willful Disobedience: Automatically Detecting Failures in Agentic Traces
+
+- **ID:** `2603.23806` · **primary:** `cs.SE` · **Submitted:** 2026-03-25
+- **Authors:** Reshabh K Sharma, Shraddha Barke, Benjamin Zorn
+- **abs:** https://arxiv.org/abs/2603.23806 · **PDF:** https://arxiv.org/pdf/2603.23806v1
+
+## 2181. Deep Neural Regression Collapse
+
+- **ID:** `2603.23805` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Akshay Rangamani, Altay Unal
+- **abs:** https://arxiv.org/abs/2603.23805 · **PDF:** https://arxiv.org/pdf/2603.23805v1
+
+## 2182. Object Search in Partially-Known Environments via LLM-informed Model-based Planning and Prompt Selection
+
+- **ID:** `2603.23800` · **primary:** `cs.RO` · **Submitted:** 2026-03-25
+- **Authors:** Abhishek Paudel, Abhish Khanal, Raihan I. Arnob, et al.
+- **abs:** https://arxiv.org/abs/2603.23800 · **PDF:** https://arxiv.org/pdf/2603.23800v1
+
+## 2183. Resolving gradient pathology in physics-informed epidemiological models
+
+- **ID:** `2603.23799` · **primary:** `cs.LG` · **Submitted:** 2026-03-25
+- **Authors:** Nickson Golooba, Woldegebriel Assefa Woldegerima
+- **abs:** https://arxiv.org/abs/2603.23799 · **PDF:** https://arxiv.org/pdf/2603.23799v1
+
+## 2184. Infrequent Child-Directed Speech Is Bursty and May Draw Infant Vocalizations
+
+- **ID:** `2603.23797` · **primary:** `cs.CL` · **Submitted:** 2026-03-25
+- **Authors:** Margaret Cychosz, Adriana Weisleder
+- **abs:** https://arxiv.org/abs/2603.23797 · **PDF:** https://arxiv.org/pdf/2603.23797v1
+
+## 2185. Sparse Autoencoders for Interpretable Medical Image Representation Learning
+
+- **ID:** `2603.23794` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Philipp Wesp, Robbie Holland, Vasiliki Sideri-Lampretsa, et al.
+- **abs:** https://arxiv.org/abs/2603.23794 · **PDF:** https://arxiv.org/pdf/2603.23794v1
+
+## 2186. Manifold Generalization Provably Proceeds Memorization in Diffusion Models
+
+- **ID:** `2603.23792` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Zebang Shen, Ya-Ping Hsieh, Niao He
+- **abs:** https://arxiv.org/abs/2603.23792 · **PDF:** https://arxiv.org/pdf/2603.23792v1
+
+## 2187. The Cognitive Firewall:Securing Browser Based AI Agents Against Indirect Prompt Injection Via Hybrid Edge Cloud Defense
+
+- **ID:** `2603.23791` · **primary:** `cs.CR` · **Submitted:** 2026-03-24
+- **Authors:** Qianlong Lan, Anuj Kaul
+- **abs:** https://arxiv.org/abs/2603.23791 · **PDF:** https://arxiv.org/pdf/2603.23791v1
+
+## 2188. Digital Twin-Assisted Measurement Design and Channel Statistics Prediction
+
+- **ID:** `2603.23787` · **primary:** `cs.IT` · **Submitted:** 2026-03-24
+- **Authors:** Robin J. Williams, Mahmoud Saad Abouamer, Petar Popovski
+- **abs:** https://arxiv.org/abs/2603.23787 · **PDF:** https://arxiv.org/pdf/2603.23787v1
+
+## 2189. Retinal Disease Classification from Fundus Images using CNN Transfer Learning
+
+- **ID:** `2603.23785` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Ali Akram
+- **abs:** https://arxiv.org/abs/2603.23785 · **PDF:** https://arxiv.org/pdf/2603.23785v1
+
+## 2190. Latent Algorithmic Structure Precedes Grokking: A Mechanistic Study of ReLU MLPs on Modular Arithmetic
+
+- **ID:** `2603.23784` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Anand Swaroop
+- **abs:** https://arxiv.org/abs/2603.23784 · **PDF:** https://arxiv.org/pdf/2603.23784v1
+
+## 2191. Probabilistic Geometric Alignment via Bayesian Latent Transport for Domain-Adaptive Foundation Models
+
+- **ID:** `2603.23783` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Aueaphum Aueawatthanaphisut, Kuepon Auewattanapisut
+- **abs:** https://arxiv.org/abs/2603.23783 · **PDF:** https://arxiv.org/pdf/2603.23783v2
+
+## 2192. Lightweight Fairness for LLM-Based Recommendations via Kernelized Projection and Gated Adapters
+
+- **ID:** `2603.23780` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Nan Cui, Wendy Hui Wang, Yue Ning
+- **abs:** https://arxiv.org/abs/2603.23780 · **PDF:** https://arxiv.org/pdf/2603.23780v1
+
+## 2193. Human-in-the-Loop Pareto Optimization: Trade-off Characterization for Assist-as-Needed Training and Performance Evaluation
+
+- **ID:** `2603.23777` · **primary:** `cs.RO` · **Submitted:** 2026-03-24
+- **Authors:** Harun Tolasa, Volkan Patoglu
+- **abs:** https://arxiv.org/abs/2603.23777 · **PDF:** https://arxiv.org/pdf/2603.23777v1
+
+## 2194. AI-driven Intent-Based Networking Approach for Self-configuration of Next Generation Networks
+
+- **ID:** `2603.23772` · **primary:** `cs.NI` · **Submitted:** 2026-03-24
+- **Authors:** Md. Kamrul Hossain, Walid Aljoby
+- **abs:** https://arxiv.org/abs/2603.23772 · **PDF:** https://arxiv.org/pdf/2603.23772v1
+
+## 2195. Self Paced Gaussian Contextual Reinforcement Learning
+
+- **ID:** `2603.23755` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Mohsen Sahraei Ardakani, Rui Song
+- **abs:** https://arxiv.org/abs/2603.23755 · **PDF:** https://arxiv.org/pdf/2603.23755v1
+
+## 2196. IslamicMMLU: A Benchmark for Evaluating LLMs on Islamic Knowledge
+
+- **ID:** `2603.23750` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Ali Abdelaal, Mohammed Nader Al Haffar, Mahmoud Fawzi, et al.
+- **abs:** https://arxiv.org/abs/2603.23750 · **PDF:** https://arxiv.org/pdf/2603.23750v1
+
+## 2197. Efficient Benchmarking of AI Agents
+
+- **ID:** `2603.23749` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Franck Ndzomga
+- **abs:** https://arxiv.org/abs/2603.23749 · **PDF:** https://arxiv.org/pdf/2603.23749v1
+
+## 2198. The Order Is The Message
+
+- **ID:** `2603.25047` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Jordan LeDoux
+- **abs:** https://arxiv.org/abs/2603.25047 · **PDF:** https://arxiv.org/pdf/2603.25047v1
+
+## 2199. Kronecker-Structured Nonparametric Spatiotemporal Point Processes
+
+- **ID:** `2603.23746` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Zhitong Xu, Qiwei Yuan, Yinghao Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.23746 · **PDF:** https://arxiv.org/pdf/2603.23746v1
+
+## 2200. ARC-AGI-3: A New Challenge for Frontier Agentic Intelligence
+
+- **ID:** `2603.24621` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** ARC Prize Foundation
+- **abs:** https://arxiv.org/abs/2603.24621 · **PDF:** https://arxiv.org/pdf/2603.24621v1
+
+## 2201. BXRL: Behavior-Explainable Reinforcement Learning
+
+- **ID:** `2603.23738` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Ram Rachum, Yotam Amitai, Yonatan Nakar, et al.
+- **abs:** https://arxiv.org/abs/2603.23738 · **PDF:** https://arxiv.org/pdf/2603.23738v1
+
+## 2202. Wasserstein Parallel Transport for Predicting the Dynamics of Statistical Systems
+
+- **ID:** `2603.23736` · **primary:** `stat.ML` · **Submitted:** 2026-03-24
+- **Authors:** Tristan Luca Saidi, Gonzalo Mena, Larry Wasserman, et al.
+- **abs:** https://arxiv.org/abs/2603.23736 · **PDF:** https://arxiv.org/pdf/2603.23736v1
+
+## 2203. Autoregressive Guidance of Deep Spatially Selective Filters using Bayesian Tracking for Efficient Extraction of Moving Speakers
+
+- **ID:** `2603.23723` · **primary:** `eess.AS` · **Submitted:** 2026-03-24
+- **Authors:** Jakob Kienegger, Timo Gerkmann
+- **abs:** https://arxiv.org/abs/2603.23723 · **PDF:** https://arxiv.org/pdf/2603.23723v1
+
+## 2204. Dual-Gated Epistemic Time-Dilation: Autonomous Compute Modulation in Asynchronous MARL
+
+- **ID:** `2603.23722` · **primary:** `cs.MA` · **Submitted:** 2026-03-24
+- **Authors:** Igor Jankowski
+- **abs:** https://arxiv.org/abs/2603.23722 · **PDF:** https://arxiv.org/pdf/2603.23722v1
+
+## 2205. CDMT-EHR: A Continuous-Time Diffusion Framework for Generating Mixed-Type Time-Series Electronic Health Records
+
+- **ID:** `2603.23719` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Shaonan Liu, Yuichiro Iwashita, Soichiro Nakako, et al.
+- **abs:** https://arxiv.org/abs/2603.23719 · **PDF:** https://arxiv.org/pdf/2603.23719v1
+
+## 2206. LLMs Do Not Grade Essays Like Humans
+
+- **ID:** `2603.23714` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Jerin George Mathew, Sumayya Taher, Anindita Kundu, et al.
+- **abs:** https://arxiv.org/abs/2603.23714 · **PDF:** https://arxiv.org/pdf/2603.23714v1
+
+## 2207. An In-Depth Study of Filter-Agnostic Vector Search on a PostgreSQL Database System: [Experiments and Analysis]
+
+- **ID:** `2603.23710` · **primary:** `cs.DB` · **Submitted:** 2026-03-24
+- **Authors:** Duo Lu, Helena Caminal, Manos Chatzakis, et al.
+- **abs:** https://arxiv.org/abs/2603.23710 · **PDF:** https://arxiv.org/pdf/2603.23710v1
+
+## 2208. BHCast: Unlocking Black Hole Plasma Dynamics from a Single Blurry Image with Long-Term Forecasting
+
+- **ID:** `2603.26777` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Renbo Tu, Ali SaraerToosi, Nicholas S. Conroy, et al.
+- **abs:** https://arxiv.org/abs/2603.26777 · **PDF:** https://arxiv.org/pdf/2603.26777v1
+
+## 2209. The Diminishing Returns of Early-Exit Decoding in Modern LLMs
+
+- **ID:** `2603.23701` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Rui Wei, Rui Du, Hanfei Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.23701 · **PDF:** https://arxiv.org/pdf/2603.23701v1
+
+## 2210. Causal AI For AMS Circuit Design: Interpretable Parameter Effects Analysis
+
+- **ID:** `2603.24618` · **primary:** `cs.AR` · **Submitted:** 2026-03-24
+- **Authors:** Mohyeu Hussain, David Koblah, Reiner Dizon-Paradis, et al.
+- **abs:** https://arxiv.org/abs/2603.24618 · **PDF:** https://arxiv.org/pdf/2603.24618v1
+
+## 2211. Adaptive Gaussian Process Search for Simulation-Based Sample Size Estimation in Clinical Prediction Models: Validation of the pmsims R Package
+
+- **ID:** `2603.23688` · **primary:** `stat.CO` · **Submitted:** 2026-03-24
+- **Authors:** Oyebayo Ridwan Olaniran, Diana Shamsutdinova, Sarah Markham, et al.
+- **abs:** https://arxiv.org/abs/2603.23688 · **PDF:** https://arxiv.org/pdf/2603.23688v1
+
+## 2212. Multi-LLM Query Optimization
+
+- **ID:** `2603.24617` · **primary:** `cs.DS` · **Submitted:** 2026-03-24
+- **Authors:** Arlen Dean, Zijin Zhang, Stefanus Jasin, et al.
+- **abs:** https://arxiv.org/abs/2603.24617 · **PDF:** https://arxiv.org/pdf/2603.24617v1
+
+## 2213. The Economics of Builder Saturation in Digital Markets
+
+- **ID:** `2603.23685` · **primary:** `econ.TH` · **Submitted:** 2026-03-24
+- **Authors:** Armin Catovic
+- **abs:** https://arxiv.org/abs/2603.23685 · **PDF:** https://arxiv.org/pdf/2603.23685v2
+
+## 2214. Assessment Design in the AI Era: A Method for Identifying Items Functioning Differentially for Humans and Chatbots
+
+- **ID:** `2603.23682` · **primary:** `cs.HC` · **Submitted:** 2026-03-24
+- **Authors:** Licol Zeinfeld, Alona Strugatski, Ziva Bar-Dov, et al.
+- **abs:** https://arxiv.org/abs/2603.23682 · **PDF:** https://arxiv.org/pdf/2603.23682v1
+
+## 2215. Learning What Can Be Picked: Active Reachability Estimation for Efficient Robotic Fruit Harvesting
+
+- **ID:** `2603.23679` · **primary:** `cs.RO` · **Submitted:** 2026-03-24
+- **Authors:** Nur Afsa Syeda, Mohamed Elmahallawy, Luis Fernando de la Torre, et al.
+- **abs:** https://arxiv.org/abs/2603.23679 · **PDF:** https://arxiv.org/pdf/2603.23679v1
+
+## 2216. PLACID: Privacy-preserving Large language models for Acronym Clinical Inference and Disambiguation
+
+- **ID:** `2603.23678` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Manjushree B. Aithal, Ph. D., Alexander Kotz, et al.
+- **abs:** https://arxiv.org/abs/2603.23678 · **PDF:** https://arxiv.org/pdf/2603.23678v1
+
+## 2217. Prototype Fusion: A Training-Free Multi-Layer Approach to OOD Detection
+
+- **ID:** `2603.23677` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Shreen Gul, Mohamed Elmahallawy, Ardhendu Tripathy, et al.
+- **abs:** https://arxiv.org/abs/2603.23677 · **PDF:** https://arxiv.org/pdf/2603.23677v1
+
+## 2218. Grounding Vision and Language to 3D Masks for Long-Horizon Box Rearrangement
+
+- **ID:** `2603.23676` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Ashish Malik, Caleb Lowe, Aayam Shrestha, et al.
+- **abs:** https://arxiv.org/abs/2603.23676 · **PDF:** https://arxiv.org/pdf/2603.23676v1
+
+## 2219. Estimating Individual Tree Height and Species from UAV Imagery
+
+- **ID:** `2603.23669` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Jannik Endres, Etienne Laliberté, David Rolnick, et al.
+- **abs:** https://arxiv.org/abs/2603.23669 · **PDF:** https://arxiv.org/pdf/2603.23669v1
+
+## 2220. Energy Efficient Software Hardware CoDesign for Machine Learning: From TinyML to Large Language Models
+
+- **ID:** `2603.23668` · **primary:** `cs.AR` · **Submitted:** 2026-03-24
+- **Authors:** Mohammad Saleh Vahdatpour, Yanqing Zhang
+- **abs:** https://arxiv.org/abs/2603.23668 · **PDF:** https://arxiv.org/pdf/2603.23668v1
+
+## 2221. Echoes: A semantically-aligned music deepfake detection dataset
+
+- **ID:** `2603.23667` · **primary:** `cs.SD` · **Submitted:** 2026-03-24
+- **Authors:** Octavian Pascu, Dan Oneata, Horia Cucu, et al.
+- **abs:** https://arxiv.org/abs/2603.23667 · **PDF:** https://arxiv.org/pdf/2603.23667v1
+
+## 2222. GTO Wizard Benchmark
+
+- **ID:** `2603.23660` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Marc-Antoine Provost, Nejc Ilenic, Christopher Solinas, et al.
+- **abs:** https://arxiv.org/abs/2603.23660 · **PDF:** https://arxiv.org/pdf/2603.23660v1
+
+## 2223. Probing Ethical Framework Representations in Large Language Models: Structure, Entanglement, and Methodological Challenges
+
+- **ID:** `2603.23659` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Weilun Xu, Alexander Rusnak, Frederic Kaplan
+- **abs:** https://arxiv.org/abs/2603.23659 · **PDF:** https://arxiv.org/pdf/2603.23659v1
+
+## 2224. Boost Like a (Var)Pro: Trust-Region Gradient Boosting via Variable Projection
+
+- **ID:** `2603.23658` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Abhijit Chowdhary, Elizabeth Newman, Deepanshu Verma
+- **abs:** https://arxiv.org/abs/2603.23658 · **PDF:** https://arxiv.org/pdf/2603.23658v1
+
+## 2225. Ethio-ASR: Joint Multilingual Speech Recognition and Language Identification for Ethiopian Languages
+
+- **ID:** `2603.23654` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Badr M. Abdullah, Israel Abebe Azime, Atnafu Lambebo Tonja, et al.
+- **abs:** https://arxiv.org/abs/2603.23654 · **PDF:** https://arxiv.org/pdf/2603.23654v1
+
+## 2226. λSplit: Self-Supervised Content-Aware Spectral Unmixing for Fluorescence Microscopy
+
+- **ID:** `2603.23647` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Federico Carrara, Talley Lambert, Mehdi Seifi, et al.
+- **abs:** https://arxiv.org/abs/2603.23647 · **PDF:** https://arxiv.org/pdf/2603.23647v1
+
+## 2227. Swiss-Bench SBP-002: A Frontier Model Comparison on Swiss Legal and Regulatory Tasks
+
+- **ID:** `2603.23646` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Fatih Uenal
+- **abs:** https://arxiv.org/abs/2603.23646 · **PDF:** https://arxiv.org/pdf/2603.23646v1
+
+## 2228. LLM Inference at the Edge: Mobile, NPU, and GPU Performance Efficiency Trade-offs Under Sustained Load
+
+- **ID:** `2603.23640` · **primary:** `cs.DC` · **Submitted:** 2026-03-24
+- **Authors:** Pranay Tummalapalli, Sahil Arayakandy, Ritam Pal, et al.
+- **abs:** https://arxiv.org/abs/2603.23640 · **PDF:** https://arxiv.org/pdf/2603.23640v1
+
+## 2229. Can LLM Agents Be CFOs? A Benchmark for Resource Allocation in Dynamic Enterprise Environments
+
+- **ID:** `2603.23638` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Yi Han, Lingfei Qian, Yan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.23638 · **PDF:** https://arxiv.org/pdf/2603.23638v1
+
+## 2230. Steering Code LLMs with Activation Directions for Language and Library Control
+
+- **ID:** `2603.23629` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Md Mahbubur Rahman, Arjun Guha, Harshitha Menon
+- **abs:** https://arxiv.org/abs/2603.23629 · **PDF:** https://arxiv.org/pdf/2603.23629v2
+
+## 2231. Ukrainian Visual Word Sense Disambiguation Benchmark
+
+- **ID:** `2603.23627` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Yurii Laba, Yaryna Mohytych, Ivanna Rohulia, et al.
+- **abs:** https://arxiv.org/abs/2603.23627 · **PDF:** https://arxiv.org/pdf/2603.23627v1
+
+## 2232. A Theory of LLM Information Susceptibility
+
+- **ID:** `2603.23626` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Zhuo-Yang Song, Hua Xing Zhu
+- **abs:** https://arxiv.org/abs/2603.23626 · **PDF:** https://arxiv.org/pdf/2603.23626v1
+
+## 2233. Evaluating a Multi-Agent Voice-Enabled Smart Speaker for Care Homes: A Safety-Focused Framework
+
+- **ID:** `2603.23625` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Zeinab Dehghani, Rameez Raja Kureshi, Koorosh Aslansefat, et al.
+- **abs:** https://arxiv.org/abs/2603.23625 · **PDF:** https://arxiv.org/pdf/2603.23625v1
+
+## 2234. Revisiting Real-Time Digging-In Effects: No Evidence from NP/Z Garden-Paths
+
+- **ID:** `2603.23624` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Amani Maina-Kilaas, Roger Levy
+- **abs:** https://arxiv.org/abs/2603.23624 · **PDF:** https://arxiv.org/pdf/2603.23624v1
+
+## 2235. Learning to Select Visual In-Context Demonstrations
+
+- **ID:** `2603.26775` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Eugene Lee, Yu-Chi Lin, Jiajie Diao
+- **abs:** https://arxiv.org/abs/2603.26775 · **PDF:** https://arxiv.org/pdf/2603.26775v1
+
+## 2236. LLMLOOP: Improving LLM-Generated Code and Tests through Automated Iterative Feedback Loops
+
+- **ID:** `2603.23613` · **primary:** `cs.SE` · **Submitted:** 2026-03-24
+- **Authors:** Ravin Ravi, Dylan Bradshaw, Stefano Ruberto, et al.
+- **abs:** https://arxiv.org/abs/2603.23613 · **PDF:** https://arxiv.org/pdf/2603.23613v1
+
+## 2237. LLMORPH: Automated Metamorphic Testing of Large Language Models
+
+- **ID:** `2603.23611` · **primary:** `cs.SE` · **Submitted:** 2026-03-24
+- **Authors:** Steven Cho, Stefano Ruberto, Valerio Terragni
+- **abs:** https://arxiv.org/abs/2603.23611 · **PDF:** https://arxiv.org/pdf/2603.23611v1
+
+## 2238. Environment Maps: Structured Environmental Representations for Long-Horizon Agents
+
+- **ID:** `2603.23610` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Yenchia Feng, Chirag Sharma, Karime Maamari
+- **abs:** https://arxiv.org/abs/2603.23610 · **PDF:** https://arxiv.org/pdf/2603.23610v3
+
+## 2239. MedObvious: Exposing the Medical Moravec's Paradox in VLMs via Clinical Triage
+
+- **ID:** `2603.23501` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Ufaq Khan, Umair Nawaz, L D M S S Teja, et al.
+- **abs:** https://arxiv.org/abs/2603.23501 · **PDF:** https://arxiv.org/pdf/2603.23501v1
+
+## 2240. Estimating Flow Velocity and Vehicle Angle-of-Attack from Non-invasive Piezoelectric Structural Measurements Using Deep Learning
+
+- **ID:** `2603.23496` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Chandler B. Smith, S. Hales Swift, Andrew Steyer, et al.
+- **abs:** https://arxiv.org/abs/2603.23496 · **PDF:** https://arxiv.org/pdf/2603.23496v1
+
+## 2241. VISion On Request: Enhanced VLLM efficiency with sparse, dynamically selected, vision-language interactions
+
+- **ID:** `2603.23495` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Adrian Bulat, Alberto Baldrati, Ioannis Maniadis Metaxas, et al.
+- **abs:** https://arxiv.org/abs/2603.23495 · **PDF:** https://arxiv.org/pdf/2603.23495v1
+
+## 2242. Failure of contextual invariance in gender inference with large language models
+
+- **ID:** `2603.23485` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Sagar Kumar, Ariel Flint, Luca Maria Aiello, et al.
+- **abs:** https://arxiv.org/abs/2603.23485 · **PDF:** https://arxiv.org/pdf/2603.23485v1
+
+## 2243. SpecEyes: Accelerating Agentic Multimodal LLMs via Speculative Perception and Planning
+
+- **ID:** `2603.23483` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Haoyu Huang, Jinfa Huang, Zhongwei Wan, et al.
+- **abs:** https://arxiv.org/abs/2603.23483 · **PDF:** https://arxiv.org/pdf/2603.23483v1
+
+## 2244. ReqFusion: A Multi-Provider Framework for Automated PEGS Analysis Across Software Domains
+
+- **ID:** `2603.23482` · **primary:** `cs.SE` · **Submitted:** 2026-03-24
+- **Authors:** Muhammad Khalid, Manuel Oriol, Yilmaz Uygun
+- **abs:** https://arxiv.org/abs/2603.23482 · **PDF:** https://arxiv.org/pdf/2603.23482v1
+
+## 2245. VTAM: Video-Tactile-Action Models for Complex Physical Interaction Beyond VLAs
+
+- **ID:** `2603.23481` · **primary:** `cs.RO` · **Submitted:** 2026-03-24
+- **Authors:** Haoran Yuan, Weigang Yi, Zhenyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.23481 · **PDF:** https://arxiv.org/pdf/2603.23481v1
+
+## 2246. Byzantine-Robust and Differentially Private Federated Optimization under Weaker Assumptions
+
+- **ID:** `2603.23472` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Rustem Islamov, Grigory Malinovsky, Alexander Gaponov, et al.
+- **abs:** https://arxiv.org/abs/2603.23472 · **PDF:** https://arxiv.org/pdf/2603.23472v1
+
+## 2247. InverFill: One-Step Inversion for Enhanced Few-Step Diffusion Inpainting
+
+- **ID:** `2603.23463` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Duc Vu, Kien Nguyen, Trong-Tung Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.23463 · **PDF:** https://arxiv.org/pdf/2603.23463v1
+
+## 2248. End-to-End Efficient RL for Linear Bellman Complete MDPs with Deterministic Transitions
+
+- **ID:** `2603.23461` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Zakaria Mhammedi, Alexander Rakhlin, Nneka Okolo
+- **abs:** https://arxiv.org/abs/2603.23461 · **PDF:** https://arxiv.org/pdf/2603.23461v1
+
+## 2249. CSTS: A Canonical Security Telemetry Substrate for AI-Native Cyber Detection
+
+- **ID:** `2603.23459` · **primary:** `cs.CR` · **Submitted:** 2026-03-24
+- **Authors:** Abdul Rahman
+- **abs:** https://arxiv.org/abs/2603.23459 · **PDF:** https://arxiv.org/pdf/2603.23459v1
+
+## 2250. Code Review Agent Benchmark
+
+- **ID:** `2603.23448` · **primary:** `cs.SE` · **Submitted:** 2026-03-24
+- **Authors:** Yuntong Zhang, Zhiyuan Pan, Imam Nur Bani Yusuf, et al.
+- **abs:** https://arxiv.org/abs/2603.23448 · **PDF:** https://arxiv.org/pdf/2603.23448v2
+
+## 2251. StepCache: Step-Level Reuse with Lightweight Verification and Selective Patching for LLM Serving
+
+- **ID:** `2603.28795` · **primary:** `cs.OS` · **Submitted:** 2026-03-24
+- **Authors:** Azam Nouri
+- **abs:** https://arxiv.org/abs/2603.28795 · **PDF:** https://arxiv.org/pdf/2603.28795v1
+
+## 2252. 3DCity-LLM: Empowering Multi-modality Large Language Models for 3D City-scale Perception and Understanding
+
+- **ID:** `2603.23447` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Yiping Chen, Jinpeng Li, Wenyu Ke, et al.
+- **abs:** https://arxiv.org/abs/2603.23447 · **PDF:** https://arxiv.org/pdf/2603.23447v1
+
+## 2253. Evaluating LLM-Based Test Generation Under Software Evolution
+
+- **ID:** `2603.23443` · **primary:** `cs.SE` · **Submitted:** 2026-03-24
+- **Authors:** Sabaat Haroon, Mohammad Taha Khan, Muhammad Ali Gulzar
+- **abs:** https://arxiv.org/abs/2603.23443 · **PDF:** https://arxiv.org/pdf/2603.23443v1
+
+## 2254. Targeted Adversarial Traffic Generation : Black-box Approach to Evade Intrusion Detection Systems in IoT Networks
+
+- **ID:** `2603.23438` · **primary:** `cs.CR` · **Submitted:** 2026-03-24
+- **Authors:** Islam Debicha, Tayeb Kenaza, Ishak Charfi, et al.
+- **abs:** https://arxiv.org/abs/2603.23438 · **PDF:** https://arxiv.org/pdf/2603.23438v1
+
+## 2255. Similarity-Aware Mixture-of-Experts for Data-Efficient Continual Learning
+
+- **ID:** `2603.23436` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Connor Mclaughlin, Nigel Lee, Lili Su
+- **abs:** https://arxiv.org/abs/2603.23436 · **PDF:** https://arxiv.org/pdf/2603.23436v1
+
+## 2256. Mecha-nudges for Machines
+
+- **ID:** `2603.23433` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Giulio Frey, Kawin Ethayarajh
+- **abs:** https://arxiv.org/abs/2603.23433 · **PDF:** https://arxiv.org/pdf/2603.23433v1
+
+## 2257. Bilevel Autoresearch: Meta-Autoresearching Itself
+
+- **ID:** `2603.23420` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Yaonan Qu, Meng Lu
+- **abs:** https://arxiv.org/abs/2603.23420 · **PDF:** https://arxiv.org/pdf/2603.23420v1
+
+## 2258. Biased Error Attribution in Multi-Agent Human-AI Systems Under Delayed Feedback
+
+- **ID:** `2603.23419` · **primary:** `cs.HC` · **Submitted:** 2026-03-24
+- **Authors:** Teerthaa Parakh, Karen M. Feigh
+- **abs:** https://arxiv.org/abs/2603.23419 · **PDF:** https://arxiv.org/pdf/2603.23419v1
+
+## 2259. SortedRL: Accelerating RL Training for LLMs through Online Length-Aware Scheduling
+
+- **ID:** `2603.23414` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Yiqi Zhang, Huiqiang Jiang, Xufang Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.23414 · **PDF:** https://arxiv.org/pdf/2603.23414v1
+
+## 2260. Robot Arm Control via Cognitive Map Learners
+
+- **ID:** `2603.26773` · **primary:** `cs.RO` · **Submitted:** 2026-03-24
+- **Authors:** Nathan McDonald, Colyn Seeley, Christian Brazeau
+- **abs:** https://arxiv.org/abs/2603.26773 · **PDF:** https://arxiv.org/pdf/2603.26773v1
+
+## 2261. Beyond Preset Identities: How Agents Form Stances and Boundaries in Generative Societies
+
+- **ID:** `2603.23406` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Hanzhong Zhang, Siyang Song, Jindong Wang
+- **abs:** https://arxiv.org/abs/2603.23406 · **PDF:** https://arxiv.org/pdf/2603.23406v2
+
+## 2262. Planning over MAPF Agent Dependencies via Multi-Dependency PIBT
+
+- **ID:** `2603.23405` · **primary:** `cs.MA` · **Submitted:** 2026-03-24
+- **Authors:** Zixiang Jiang, Yulun Zhang, Rishi Veerapaneni, et al.
+- **abs:** https://arxiv.org/abs/2603.23405 · **PDF:** https://arxiv.org/pdf/2603.23405v1
+
+## 2263. Unleashing Spatial Reasoning in Multimodal Large Language Models via Textual Representation Guided Reasoning
+
+- **ID:** `2603.23404` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Jiacheng Hua, Yishu Yin, Yuhang Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.23404 · **PDF:** https://arxiv.org/pdf/2603.23404v1
+
+## 2264. Graph Energy Matching: Transport-Aligned Energy-Based Modeling for Graph Generation
+
+- **ID:** `2603.23398` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Michal Balcerak, Suprosana Shit, Chinmay Prabhakar, et al.
+- **abs:** https://arxiv.org/abs/2603.23398 · **PDF:** https://arxiv.org/pdf/2603.23398v1
+
+## 2265. Kinetic Langevin Splitting Schemes for Constrained Sampling
+
+- **ID:** `2603.23397` · **primary:** `stat.ME` · **Submitted:** 2026-03-24
+- **Authors:** Neil K. Chada, Lu Yu
+- **abs:** https://arxiv.org/abs/2603.23397 · **PDF:** https://arxiv.org/pdf/2603.23397v2
+
+## 2266. LineMVGNN: Anti-Money Laundering with Line-Graph-Assisted Multi-View Graph Neural Networks
+
+- **ID:** `2603.23584` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Chung-Hoo Poon, James Kwok, Calvin Chow, et al.
+- **abs:** https://arxiv.org/abs/2603.23584 · **PDF:** https://arxiv.org/pdf/2603.23584v1
+
+## 2267. Natural Language Interfaces for Spatial and Temporal Databases: A Comprehensive Overview of Methods, Taxonomy, and Future Directions
+
+- **ID:** `2603.23375` · **primary:** `cs.DB` · **Submitted:** 2026-03-24
+- **Authors:** Samya Acharja, Kanchan Chowdhury
+- **abs:** https://arxiv.org/abs/2603.23375 · **PDF:** https://arxiv.org/pdf/2603.23375v1
+
+## 2268. Shape-Adaptive Conditional Calibration for Conformal Prediction via Minimax Optimization
+
+- **ID:** `2603.23374` · **primary:** `stat.ME` · **Submitted:** 2026-03-24
+- **Authors:** Yajie Bao, Chuchen Zhang, Zhaojun Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.23374 · **PDF:** https://arxiv.org/pdf/2603.23374v1
+
+## 2269. Central Dogma Transformer III: Interpretable AI Across DNA, RNA, and Protein
+
+- **ID:** `2603.23361` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Nobuyuki Ota
+- **abs:** https://arxiv.org/abs/2603.23361 · **PDF:** https://arxiv.org/pdf/2603.23361v2
+
+## 2270. Contrastive Metric Learning for Point Cloud Segmentation in Highly Granular Detectors
+
+- **ID:** `2603.23356` · **primary:** `hep-ex` · **Submitted:** 2026-03-24
+- **Authors:** Max Marriott-Clarke, Lazar Novakovic, Elizabeth Ratzer, et al.
+- **abs:** https://arxiv.org/abs/2603.23356 · **PDF:** https://arxiv.org/pdf/2603.23356v1
+
+## 2271. Off-Policy Value-Based Reinforcement Learning for Large Language Models
+
+- **ID:** `2603.23355` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Peng-Yuan Wang, Ziniu Li, Tian Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.23355 · **PDF:** https://arxiv.org/pdf/2603.23355v1
+
+## 2272. RelayS2S: A Dual-Path Speculative Generation for Real-Time Dialogue
+
+- **ID:** `2603.23346` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Long Mai
+- **abs:** https://arxiv.org/abs/2603.23346 · **PDF:** https://arxiv.org/pdf/2603.23346v1
+
+## 2273. Edge Radar Material Classification Under Geometry Shifts
+
+- **ID:** `2603.23342` · **primary:** `cs.RO` · **Submitted:** 2026-03-24
+- **Authors:** Jannik Hohmann, Dong Wang, Andreas Nüchter
+- **abs:** https://arxiv.org/abs/2603.23342 · **PDF:** https://arxiv.org/pdf/2603.23342v1
+
+## 2274. Leveraging LLMs and Social Media to Understand User Perception of Smartphone-Based Earthquake Early Warnings
+
+- **ID:** `2603.23322` · **primary:** `stat.AP` · **Submitted:** 2026-03-24
+- **Authors:** Hanjing Wang, S. Mostafa Mousavi, Patrick Robertson, et al.
+- **abs:** https://arxiv.org/abs/2603.23322 · **PDF:** https://arxiv.org/pdf/2603.23322v1
+
+## 2275. WISTERIA: Weak Implicit Signal-based Temporal Relation Extraction with Attention
+
+- **ID:** `2603.23319` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Duy Dao Do, Anaïs Halftermeyer, Thi-Bich-Hanh Dao
+- **abs:** https://arxiv.org/abs/2603.23319 · **PDF:** https://arxiv.org/pdf/2603.23319v1
+
+## 2276. Robustness Quantification for Discriminative Models: a New Robustness Metric and its Application to Dynamic Classifier Selection
+
+- **ID:** `2603.23318` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Rodrigo F. L. Lassance, Jasper De Bock
+- **abs:** https://arxiv.org/abs/2603.23318 · **PDF:** https://arxiv.org/pdf/2603.23318v1
+
+## 2277. Unilateral Relationship Revision Power in Human-AI Companion Interaction
+
+- **ID:** `2603.23315` · **primary:** `cs.CY` · **Submitted:** 2026-03-24
+- **Authors:** Benjamin Lange
+- **abs:** https://arxiv.org/abs/2603.23315 · **PDF:** https://arxiv.org/pdf/2603.23315v1
+
+## 2278. ZeroFold: Protein-RNA Binding Affinity Predictions from Pre-Structural Embeddings
+
+- **ID:** `2603.23583` · **primary:** `q-bio.BM` · **Submitted:** 2026-03-24
+- **Authors:** Josef Hanke, Sebastian Pujalte Ojeda, Shengyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.23583 · **PDF:** https://arxiv.org/pdf/2603.23583v1
+
+## 2279. ARGENT: Adaptive Hierarchical Image-Text Representations
+
+- **ID:** `2603.23311` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Chuong Huynh, Hossein Souri, Abhinav Kumar, et al.
+- **abs:** https://arxiv.org/abs/2603.23311 · **PDF:** https://arxiv.org/pdf/2603.23311v1
+
+## 2280. Curriculum-Driven 3D CT Report Generation via Language-Free Visual Grafting and Zone-Constrained Compression
+
+- **ID:** `2603.23308` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** V. K. Cody Bumgardner, Mitchell A. Klusty, Mahmut S. Gokmen, et al.
+- **abs:** https://arxiv.org/abs/2603.23308 · **PDF:** https://arxiv.org/pdf/2603.23308v1
+
+## 2281. Contextual Graph Matching with Correlated Gaussian Features
+
+- **ID:** `2603.23305` · **primary:** `stat.ML` · **Submitted:** 2026-03-24
+- **Authors:** Mohammad Hassan Ahmad Yarandi, Luca Ganassali
+- **abs:** https://arxiv.org/abs/2603.23305 · **PDF:** https://arxiv.org/pdf/2603.23305v1
+
+## 2282. A Theory of Nonparametric Covariance Function Estimation for Discretely Observed Data
+
+- **ID:** `2603.23302` · **primary:** `math.ST` · **Submitted:** 2026-03-24
+- **Authors:** Yoshikazu Terada, Atsutomo Yara
+- **abs:** https://arxiv.org/abs/2603.23302 · **PDF:** https://arxiv.org/pdf/2603.23302v1
+
+## 2283. Steering LLMs for Culturally Localized Generation
+
+- **ID:** `2603.23301` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Simran Khanuja, Hongbin Liu, Shujian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.23301 · **PDF:** https://arxiv.org/pdf/2603.23301v1
+
+## 2284. Designing Agentic AI-Based Screening for Portfolio Investment
+
+- **ID:** `2603.23300` · **primary:** `q-fin.PM` · **Submitted:** 2026-03-24
+- **Authors:** Mehmet Caner, Agostino Capponi, Nathan Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.23300 · **PDF:** https://arxiv.org/pdf/2603.23300v1
+
+## 2285. AI Generalisation Gap In Comorbid Sleep Disorder Staging
+
+- **ID:** `2603.23582` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Saswata Bose, Suvadeep Maiti, Shivam Kumar Sharma, et al.
+- **abs:** https://arxiv.org/abs/2603.23582 · **PDF:** https://arxiv.org/pdf/2603.23582v2
+
+## 2286. LLM Olympiad: Why Model Evaluation Needs a Sealed Exam
+
+- **ID:** `2603.23292` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Jan Christian Blaise Cruz, Alham Fikri Aji
+- **abs:** https://arxiv.org/abs/2603.23292 · **PDF:** https://arxiv.org/pdf/2603.23292v1
+
+## 2287. A Comparative Study of Machine Learning Models for Hourly Forecasting of Air Temperature and Relative Humidity
+
+- **ID:** `2603.23282` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Jiaqi Dong
+- **abs:** https://arxiv.org/abs/2603.23282 · **PDF:** https://arxiv.org/pdf/2603.23282v1
+
+## 2288. Emergence of Fragility in LLM-based Social Networks: the Case of Moltbook
+
+- **ID:** `2603.23279` · **primary:** `cs.SI` · **Submitted:** 2026-03-24
+- **Authors:** Luca Sodano, Sofia Sciangula, Amulya Galmarini, et al.
+- **abs:** https://arxiv.org/abs/2603.23279 · **PDF:** https://arxiv.org/pdf/2603.23279v1
+
+## 2289. A Multimodal Framework for Human-Multi-Agent Interaction
+
+- **ID:** `2603.23271` · **primary:** `cs.RO` · **Submitted:** 2026-03-24
+- **Authors:** Shaid Hasan, Breenice Lee, Sujan Sarker, et al.
+- **abs:** https://arxiv.org/abs/2603.23271 · **PDF:** https://arxiv.org/pdf/2603.23271v1
+
+## 2290. Not All Tokens Are Created Equal: Query-Efficient Jailbreak Fuzzing for LLMs
+
+- **ID:** `2603.23269` · **primary:** `cs.CR` · **Submitted:** 2026-03-24
+- **Authors:** Wenyu Chen, Xiangtao Meng, Chuanchao Zang, et al.
+- **abs:** https://arxiv.org/abs/2603.23269 · **PDF:** https://arxiv.org/pdf/2603.23269v1
+
+## 2291. SafeSeek: Universal Attribution of Safety Circuits in Language Models
+
+- **ID:** `2603.23268` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Miao Yu, Siyuan Fu, Moayad Aloqaily, et al.
+- **abs:** https://arxiv.org/abs/2603.23268 · **PDF:** https://arxiv.org/pdf/2603.23268v1
+
+## 2292. SynForceNet: A Force-Driven Global-Local Latent Representation Framework for Lithium-Ion Battery Fault Diagnosis
+
+- **ID:** `2603.23265` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Rongxiu Chen, Yuting Su
+- **abs:** https://arxiv.org/abs/2603.23265 · **PDF:** https://arxiv.org/pdf/2603.23265v1
+
+## 2293. The Mass Agreement Score: A Point-centric Measure of Cluster Size Consistency
+
+- **ID:** `2603.23581` · **primary:** `stat.ML` · **Submitted:** 2026-03-24
+- **Authors:** Randolph Wiredu-Aidoo
+- **abs:** https://arxiv.org/abs/2603.23581 · **PDF:** https://arxiv.org/pdf/2603.23581v1
+
+## 2294. MetaKube: An Experience-Aware LLM Framework for Kubernetes Failure Diagnosis
+
+- **ID:** `2603.23580` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Wei Sun, Ting Wang, Xinran Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.23580 · **PDF:** https://arxiv.org/pdf/2603.23580v1
+
+## 2295. Permutation-Symmetrized Diffusion for Unconditional Molecular Generation
+
+- **ID:** `2603.23255` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Gyeonghoon Ko, Juho Lee
+- **abs:** https://arxiv.org/abs/2603.23255 · **PDF:** https://arxiv.org/pdf/2603.23255v1
+
+## 2296. AI Lifecycle-Aware Feasibility Framework for Split-RIC Orchestration in NTN O-RAN
+
+- **ID:** `2603.23252` · **primary:** `cs.NI` · **Submitted:** 2026-03-24
+- **Authors:** Daniele Tarchi
+- **abs:** https://arxiv.org/abs/2603.23252 · **PDF:** https://arxiv.org/pdf/2603.23252v1
+
+## 2297. Is AI Catching Up to Human Expression? Exploring Emotion, Personality, Authorship, and Linguistic Style in English and Arabic with Six Large Language Models
+
+- **ID:** `2603.23251` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Nasser A Alsadhan
+- **abs:** https://arxiv.org/abs/2603.23251 · **PDF:** https://arxiv.org/pdf/2603.23251v1
+
+## 2298. A Learning Method with Gap-Aware Generation for Heterogeneous DAG Scheduling
+
+- **ID:** `2603.23249` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Ruisong Zhou, Haijun Zou, Li Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.23249 · **PDF:** https://arxiv.org/pdf/2603.23249v1
+
+## 2299. Neural ODE and SDE Models for Adaptation and Planning in Model-Based Reinforcement Learning
+
+- **ID:** `2603.23245` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Chao Han, Stefanos Ioannou, Luca Manneschi, et al.
+- **abs:** https://arxiv.org/abs/2603.23245 · **PDF:** https://arxiv.org/pdf/2603.23245v1
+
+## 2300. Online library learning in human visual puzzle solving
+
+- **ID:** `2603.23244` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Pinzhe Zhao, Emanuele Sansone, Marta Kryven, et al.
+- **abs:** https://arxiv.org/abs/2603.23244 · **PDF:** https://arxiv.org/pdf/2603.23244v1
+
+## 2301. MemCollab: Cross-Agent Memory Collaboration via Contrastive Trajectory Distillation
+
+- **ID:** `2603.23234` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Yurui Chang, Yiran Wu, Qingyun Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.23234 · **PDF:** https://arxiv.org/pdf/2603.23234v1
+
+## 2302. GEM: Guided Expectation-Maximization for Behavior-Normalized Candidate Action Selection in Offline RL
+
+- **ID:** `2603.23232` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Haoyu Wang, Jingcheng Wang, Shunyu Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.23232 · **PDF:** https://arxiv.org/pdf/2603.23232v1
+
+## 2303. PERMA: Benchmarking Personalized Memory Agents via Event-Driven Preference and Realistic Task Environments
+
+- **ID:** `2603.23231` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Shuochen Liu, Junyi Zhu, Long Shu, et al.
+- **abs:** https://arxiv.org/abs/2603.23231 · **PDF:** https://arxiv.org/pdf/2603.23231v1
+
+## 2304. I Came, I Saw, I Explained: Benchmarking Multimodal LLMs on Figurative Meaning in Memes
+
+- **ID:** `2603.23229` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Shijia Zhou, Saif M. Mohammad, Barbara Plank, et al.
+- **abs:** https://arxiv.org/abs/2603.23229 · **PDF:** https://arxiv.org/pdf/2603.23229v1
+
+## 2305. General Machine Learning: Theory for Learning Under Variable Regimes
+
+- **ID:** `2603.23220` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Aomar Osmani
+- **abs:** https://arxiv.org/abs/2603.23220 · **PDF:** https://arxiv.org/pdf/2603.23220v1
+
+## 2306. Decoding AI Authorship: Can LLMs Truly Mimic Human Style Across Literature and Politics?
+
+- **ID:** `2603.23219` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Nasser A Alsadhan
+- **abs:** https://arxiv.org/abs/2603.23219 · **PDF:** https://arxiv.org/pdf/2603.23219v1
+
+## 2307. Generative Inversion of Spectroscopic Data for Amorphous Structure Elucidation
+
+- **ID:** `2603.23210` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-03-24
+- **Authors:** Jiawei Guo, Daniel Schwalbe-Koda
+- **abs:** https://arxiv.org/abs/2603.23210 · **PDF:** https://arxiv.org/pdf/2603.23210v1
+
+## 2308. A One-Inclusion Graph Approach to Multi-Group Learning
+
+- **ID:** `2603.23208` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Noah Bergam, Samuel Deng, Daniel Hsu
+- **abs:** https://arxiv.org/abs/2603.23208 · **PDF:** https://arxiv.org/pdf/2603.23208v1
+
+## 2309. A Latency Coding Framework for Deep Spiking Neural Networks with Ultra-Low Latency
+
+- **ID:** `2603.23206` · **primary:** `cs.NE` · **Submitted:** 2026-03-24
+- **Authors:** Yi Lu, Jianhao Ding, Zhaofei Yu
+- **abs:** https://arxiv.org/abs/2603.23206 · **PDF:** https://arxiv.org/pdf/2603.23206v1
+
+## 2310. Between Resolution Collapse and Variance Inflation: Weighted Conformal Anomaly Detection in Low-Data Regimes
+
+- **ID:** `2603.23205` · **primary:** `stat.ML` · **Submitted:** 2026-03-24
+- **Authors:** Oliver Hennhöfer, Christine Preisach
+- **abs:** https://arxiv.org/abs/2603.23205 · **PDF:** https://arxiv.org/pdf/2603.23205v1
+
+## 2311. From Content to Audience: A Multimodal Annotation Framework for Broadcast Television Analytics
+
+- **ID:** `2603.26772` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Paolo Cupini, Francesco Pierri
+- **abs:** https://arxiv.org/abs/2603.26772 · **PDF:** https://arxiv.org/pdf/2603.26772v1
+
+## 2312. Residual Attention Physics-Informed Neural Networks for Robust Multiphysics Simulation of Steady-State Electrothermal Energy Systems
+
+- **ID:** `2603.23578` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Yuqing Zhou, Ze Tao, Fujun Liu
+- **abs:** https://arxiv.org/abs/2603.23578 · **PDF:** https://arxiv.org/pdf/2603.23578v1
+
+## 2313. Sparser, Faster, Lighter Transformer Language Models
+
+- **ID:** `2603.23198` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Edoardo Cetin, Stefano Peluchetti, Emilio Castillo, et al.
+- **abs:** https://arxiv.org/abs/2603.23198 · **PDF:** https://arxiv.org/pdf/2603.23198v1
+
+## 2314. Gaussian mixtures and non-parametric likelihoods through the lens of statistical mechanics
+
+- **ID:** `2603.23196` · **primary:** `math.ST` · **Submitted:** 2026-03-24
+- **Authors:** Subhroshekhar Ghosh, Adityanand Guntuboyina, Satyaki Mukherjee, et al.
+- **abs:** https://arxiv.org/abs/2603.23196 · **PDF:** https://arxiv.org/pdf/2603.23196v1
+
+## 2315. The Geometric Price of Discrete Logic: Context-driven Manifold Dynamics of Number Representations
+
+- **ID:** `2603.23577` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Long Zhang, Dai-jun Lin, Wei-neng Chen
+- **abs:** https://arxiv.org/abs/2603.23577 · **PDF:** https://arxiv.org/pdf/2603.23577v1
+
+## 2316. PhysSkin: Real-Time and Generalizable Physics-Based Animation via Self-Supervised Neural Skinning
+
+- **ID:** `2603.23194` · **primary:** `cs.GR` · **Submitted:** 2026-03-24
+- **Authors:** Yuanhang Lei, Tao Cheng, Xingxuan Li, et al.
+- **abs:** https://arxiv.org/abs/2603.23194 · **PDF:** https://arxiv.org/pdf/2603.23194v1
+
+## 2317. Wafer-Level Etch Spatial Profiling for Process Monitoring from Time-Series with Time-LLM
+
+- **ID:** `2603.23576` · **primary:** `stat.AP` · **Submitted:** 2026-03-24
+- **Authors:** Hyunwoo Kim, Munyoung Lee, Seung Hyub Jeon, et al.
+- **abs:** https://arxiv.org/abs/2603.23576 · **PDF:** https://arxiv.org/pdf/2603.23576v1
+
+## 2318. ImplicitRM: Unbiased Reward Modeling from Implicit Preference Data for LLM alignment
+
+- **ID:** `2603.23184` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Hao Wang, Haocheng Yang, Licheng Pan, et al.
+- **abs:** https://arxiv.org/abs/2603.23184 · **PDF:** https://arxiv.org/pdf/2603.23184v1
+
+## 2319. Reasoning over Semantic IDs Enhances Generative Recommendation
+
+- **ID:** `2603.23183` · **primary:** `cs.IR` · **Submitted:** 2026-03-24
+- **Authors:** Yingzhi He, Yan Sun, Junfei Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.23183 · **PDF:** https://arxiv.org/pdf/2603.23183v1
+
+## 2320. APreQEL: Adaptive Mixed Precision Quantization For Edge LLMs
+
+- **ID:** `2603.23575` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Meriem Bouzouad, Yuan-Hao Chang, Jalil Boukhobza
+- **abs:** https://arxiv.org/abs/2603.23575 · **PDF:** https://arxiv.org/pdf/2603.23575v1
+
+## 2321. SAiW: Source-Attributable Invisible Watermarking for Proactive Deepfake Defense
+
+- **ID:** `2603.23178` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Bibek Das, Chandranath Adak, Soumi Chattopadhyay, et al.
+- **abs:** https://arxiv.org/abs/2603.23178 · **PDF:** https://arxiv.org/pdf/2603.23178v1
+
+## 2322. A Schrödinger Eigenfunction Method for Long-Horizon Stochastic Optimal Control
+
+- **ID:** `2603.23173` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Louis Claeys, Artur Goldman, Zebang Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.23173 · **PDF:** https://arxiv.org/pdf/2603.23173v1
+
+## 2323. From Synthetic to Native: Benchmarking Multilingual Intent Classification in Logistics Customer Service
+
+- **ID:** `2603.23172` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Haoyu He, Jinyu Zhuang, Haoran Chu, et al.
+- **abs:** https://arxiv.org/abs/2603.23172 · **PDF:** https://arxiv.org/pdf/2603.23172v1
+
+## 2324. Robust Safety Monitoring of Language Models via Activation Watermarking
+
+- **ID:** `2603.23171` · **primary:** `cs.CR` · **Submitted:** 2026-03-24
+- **Authors:** Toluwani Aremu, Daniil Ognev, Samuele Poppi, et al.
+- **abs:** https://arxiv.org/abs/2603.23171 · **PDF:** https://arxiv.org/pdf/2603.23171v2
+
+## 2325. PoiCGAN: A Targeted Poisoning Based on Feature-Label Joint Perturbation in Federated Learning
+
+- **ID:** `2603.23574` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Tao Liu, Jiguang Lv, Dapeng Man, et al.
+- **abs:** https://arxiv.org/abs/2603.23574 · **PDF:** https://arxiv.org/pdf/2603.23574v1
+
+## 2326. LogicDiff: Logic-Guided Denoising Improves Reasoning in Masked Diffusion Language Models
+
+- **ID:** `2603.26771` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Shaik Aman
+- **abs:** https://arxiv.org/abs/2603.26771 · **PDF:** https://arxiv.org/pdf/2603.26771v1
+
+## 2327. UniDial-EvalKit: A Unified Toolkit for Evaluating Multi-Faceted Conversational Abilities
+
+- **ID:** `2603.23160` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Qi Jia, Haodong Zhao, Dun Pei, et al.
+- **abs:** https://arxiv.org/abs/2603.23160 · **PDF:** https://arxiv.org/pdf/2603.23160v1
+
+## 2328. Conformal Cross-Modal Active Learning
+
+- **ID:** `2603.23159` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Huy Hoang Nguyen, Cédric Jung, Shirin Salehi, et al.
+- **abs:** https://arxiv.org/abs/2603.23159 · **PDF:** https://arxiv.org/pdf/2603.23159v1
+
+## 2329. Describe-Then-Act: Proactive Agent Steering via Distilled Language-Action World Models
+
+- **ID:** `2603.23149` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Massimiliano Pappa, Luca Romani, Valentino Sacco, et al.
+- **abs:** https://arxiv.org/abs/2603.23149 · **PDF:** https://arxiv.org/pdf/2603.23149v1
+
+## 2330. Why AI-Generated Text Detection Fails: Evidence from Explainable AI Beyond Benchmark Accuracy
+
+- **ID:** `2603.23146` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Shushanta Pudasaini, Luis Miralles-Pechuán, David Lillis, et al.
+- **abs:** https://arxiv.org/abs/2603.23146 · **PDF:** https://arxiv.org/pdf/2603.23146v1
+
+## 2331. DAK-UCB: Diversity-Aware Prompt Routing for LLMs and Generative Models
+
+- **ID:** `2603.23140` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Donya Jafari, Farzan Farnia
+- **abs:** https://arxiv.org/abs/2603.23140 · **PDF:** https://arxiv.org/pdf/2603.23140v2
+
+## 2332. HGNet: Scalable Foundation Model for Automated Knowledge Graph Generation from Scientific Literature
+
+- **ID:** `2603.23136` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Devvrat Joshi, Islem Rekik
+- **abs:** https://arxiv.org/abs/2603.23136 · **PDF:** https://arxiv.org/pdf/2603.23136v1
+
+## 2333. A Bayesian Learning Approach for Drone Coverage Network: A Case Study on Cardiac Arrest in Scotland
+
+- **ID:** `2603.23134` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Tathagata Basu, Edoardo Patelli, Gianluca Filippi, et al.
+- **abs:** https://arxiv.org/abs/2603.23134 · **PDF:** https://arxiv.org/pdf/2603.23134v1
+
+## 2334. Polaris: A Gödel Agent Framework for Small Language Models through Experience-Abstracted Policy Repair
+
+- **ID:** `2603.23129` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Aditya Kakade, Vivek Srivastava, Shirish Karande
+- **abs:** https://arxiv.org/abs/2603.23129 · **PDF:** https://arxiv.org/pdf/2603.23129v1
+
+## 2335. Between Rules and Reality: On the Context Sensitivity of LLM Moral Judgment
+
+- **ID:** `2603.23114` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Adrian Sauter, Mona Schirmer
+- **abs:** https://arxiv.org/abs/2603.23114 · **PDF:** https://arxiv.org/pdf/2603.23114v1
+
+## 2336. Dual-Criterion Curriculum Learning: Application to Temporal Data
+
+- **ID:** `2603.23573` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Gaspard Abel, Eloi Campagne, Mohamed Benloughmari, et al.
+- **abs:** https://arxiv.org/abs/2603.23573 · **PDF:** https://arxiv.org/pdf/2603.23573v1
+
+## 2337. High-Resolution Tensor-Network Fourier Methods for Exponentially Compressed Non-Gaussian Aggregate Distributions
+
+- **ID:** `2603.23106` · **primary:** `stat.ML` · **Submitted:** 2026-03-24
+- **Authors:** Juan José Rodríguez-Aldavero, Juan José García-Ripoll
+- **abs:** https://arxiv.org/abs/2603.23106 · **PDF:** https://arxiv.org/pdf/2603.23106v1
+
+## 2338. SpecXMaster Technical Report
+
+- **ID:** `2603.23101` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Yutang Ge, Yaning Cui, Hanzheng Li, et al.
+- **abs:** https://arxiv.org/abs/2603.23101 · **PDF:** https://arxiv.org/pdf/2603.23101v2
+
+## 2339. When Language Models Lose Their Mind: The Consequences of Brain Misalignment
+
+- **ID:** `2603.23091` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Gabriele Merlin, Mariya Toneva
+- **abs:** https://arxiv.org/abs/2603.23091 · **PDF:** https://arxiv.org/pdf/2603.23091v1
+
+## 2340. Policy-based Tuning of Autoregressive Image Models with Instance- and Distribution-Level Rewards
+
+- **ID:** `2603.23086` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Orhun Buğra Baran, Melih Kandemir, Ramazan Gokberk Cinbis
+- **abs:** https://arxiv.org/abs/2603.23086 · **PDF:** https://arxiv.org/pdf/2603.23086v1
+
+## 2341. MedCausalX: Adaptive Causal Reasoning with Self-Reflection for Trustworthy Medical Vision-Language Models
+
+- **ID:** `2603.23085` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Jianxin Lin, Chunzheng Zhu, Peter J. Kneuertz, et al.
+- **abs:** https://arxiv.org/abs/2603.23085 · **PDF:** https://arxiv.org/pdf/2603.23085v1
+
+## 2342. MsFormer: Enabling Robust Predictive Maintenance Services for Industrial Devices
+
+- **ID:** `2603.23076` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Jiahui Zhou, Dan Li, Ruibing Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.23076 · **PDF:** https://arxiv.org/pdf/2603.23076v1
+
+## 2343. Can an LLM Detect Instances of Microservice Infrastructure Patterns?
+
+- **ID:** `2603.23073` · **primary:** `cs.SE` · **Submitted:** 2026-03-24
+- **Authors:** Carlos Eduardo Duarte, Neil B. Harrison, Filipe Figueiredo Correia, et al.
+- **abs:** https://arxiv.org/abs/2603.23073 · **PDF:** https://arxiv.org/pdf/2603.23073v1
+
+## 2344. Generalization Bounds for Physics-Informed Neural Networks for the Incompressible Navier-Stokes Equations
+
+- **ID:** `2603.23072` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Sebastien Andre-Sloan, Dibyakanti Kumar, Alejandro F Frangi, et al.
+- **abs:** https://arxiv.org/abs/2603.23072 · **PDF:** https://arxiv.org/pdf/2603.23072v1
+
+## 2345. AuthorMix: Modular Authorship Style Transfer via Layer-wise Adapter Mixing
+
+- **ID:** `2603.23069` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Sarubi Thillainathan, Ji-Ung Lee, Michael Sullivan, et al.
+- **abs:** https://arxiv.org/abs/2603.23069 · **PDF:** https://arxiv.org/pdf/2603.23069v1
+
+## 2346. StateLinFormer: Stateful Training Enhancing Long-term Memory in Navigation
+
+- **ID:** `2603.23571` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Zhiyuan Chen, Yuxuan Zhong, Fan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.23571 · **PDF:** https://arxiv.org/pdf/2603.23571v1
+
+## 2347. Mind Your HEARTBEAT! Claw Background Execution Inherently Enables Silent Memory Pollution
+
+- **ID:** `2603.23064` · **primary:** `cs.CR` · **Submitted:** 2026-03-24
+- **Authors:** Yechao Zhang, Shiqian Zhao, Jie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.23064 · **PDF:** https://arxiv.org/pdf/2603.23064v2
+
+## 2348. Machine Learning Models for the Early Detection of Burnout in Software Engineering: a Systematic Literature Review
+
+- **ID:** `2603.23063` · **primary:** `cs.SE` · **Submitted:** 2026-03-24
+- **Authors:** Tien Rahayu Tulili, Ayushi Rastogi, Andrea Capiluppi
+- **abs:** https://arxiv.org/abs/2603.23063 · **PDF:** https://arxiv.org/pdf/2603.23063v1
+
+## 2349. Minibal: Balanced Game-Playing Without Opponent Modeling
+
+- **ID:** `2603.23059` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Quentin Cohen-Solal, Tristan Cazenave
+- **abs:** https://arxiv.org/abs/2603.23059 · **PDF:** https://arxiv.org/pdf/2603.23059v1
+
+## 2350. Prompt Amplification and Zero-Shot Late Fusion in Audio-Language Models for Speech Emotion Recognition
+
+- **ID:** `2603.23057` · **primary:** `eess.AS` · **Submitted:** 2026-03-24
+- **Authors:** Saurabh Kataria, Xiao Hu
+- **abs:** https://arxiv.org/abs/2603.23057 · **PDF:** https://arxiv.org/pdf/2603.23057v1
+
+## 2351. Post-Selection Distributional Model Evaluation
+
+- **ID:** `2603.23055` · **primary:** `stat.ML` · **Submitted:** 2026-03-24
+- **Authors:** Amirmohammad Farzaneh, Osvaldo Simeone
+- **abs:** https://arxiv.org/abs/2603.23055 · **PDF:** https://arxiv.org/pdf/2603.23055v1
+
+## 2352. DBAutoDoc: Automated Discovery and Documentation of Undocumented Database Schemas via Statistical Analysis and Iterative LLM Refinement
+
+- **ID:** `2603.23050` · **primary:** `cs.DB` · **Submitted:** 2026-03-24
+- **Authors:** Amith Nagarajan, Thomas Altman
+- **abs:** https://arxiv.org/abs/2603.23050 · **PDF:** https://arxiv.org/pdf/2603.23050v1
+
+## 2353. MSR-HuBERT: Self-supervised Pre-training for Adaptation to Multiple Sampling Rates
+
+- **ID:** `2603.23048` · **primary:** `cs.SD` · **Submitted:** 2026-03-24
+- **Authors:** Zikang Huang, Meng Ge, Tianrui Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.23048 · **PDF:** https://arxiv.org/pdf/2603.23048v1
+
+## 2354. Parametric Knowledge and Retrieval Behavior in RAG Fine-Tuning for Electronic Design Automation
+
+- **ID:** `2603.23047` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Julian Oestreich, Maximilian Bley, Frank Binder, et al.
+- **abs:** https://arxiv.org/abs/2603.23047 · **PDF:** https://arxiv.org/pdf/2603.23047v1
+
+## 2355. Assessing the Robustness of Climate Foundation Models under No-Analog Distribution Shifts
+
+- **ID:** `2603.23043` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Maria Conchita Agana Navarro, Geng Li, Theo Wolf, et al.
+- **abs:** https://arxiv.org/abs/2603.23043 · **PDF:** https://arxiv.org/pdf/2603.23043v1
+
+## 2356. HUydra: Full-Range Lung CT Synthesis via Multiple HU Interval Generative Modelling
+
+- **ID:** `2603.23041` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** António Cardoso, Pedro Sousa, Tania Pereira, et al.
+- **abs:** https://arxiv.org/abs/2603.23041 · **PDF:** https://arxiv.org/pdf/2603.23041v1
+
+## 2357. Edge Reliability Gap in Vision-Language Models: Quantifying Failure Modes of Compressed VLMs Under Visual Corruption
+
+- **ID:** `2603.26769` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Mehmet Kaan Erol
+- **abs:** https://arxiv.org/abs/2603.26769 · **PDF:** https://arxiv.org/pdf/2603.26769v1
+
+## 2358. YOLOv10 with Kolmogorov-Arnold networks and vision-language foundation models for interpretable object detection and trustworthy multimodal AI in computer vision perception
+
+- **ID:** `2603.23037` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Marios Impraimakis, Daniel Vazquez, Feiyu Zhou
+- **abs:** https://arxiv.org/abs/2603.23037 · **PDF:** https://arxiv.org/pdf/2603.23037v1
+
+## 2359. Looking Beyond the Window: Global-Local Aligned CLIP for Training-free Open-Vocabulary Semantic Segmentation
+
+- **ID:** `2603.23030` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** ByeongCheol Lee, Hyun Seok Seong, Sangeek Hyun, et al.
+- **abs:** https://arxiv.org/abs/2603.23030 · **PDF:** https://arxiv.org/pdf/2603.23030v1
+
+## 2360. Concept-based explanations of Segmentation and Detection models in Natural Disaster Management
+
+- **ID:** `2603.23020` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Samar Heydari, Jawher Said, Galip Ümit Yolcu, et al.
+- **abs:** https://arxiv.org/abs/2603.23020 · **PDF:** https://arxiv.org/pdf/2603.23020v1
+
+## 2361. A Sobering Look at Tabular Data Generation via Probabilistic Circuits
+
+- **ID:** `2603.23016` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Davide Scassola, Dylan Ponsford, Adrián Javaloy, et al.
+- **abs:** https://arxiv.org/abs/2603.23016 · **PDF:** https://arxiv.org/pdf/2603.23016v1
+
+## 2362. Knowledge Access Beats Model Size: Memory Augmented Routing for Persistent AI Agents
+
+- **ID:** `2603.23013` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Xunzhuo Liu, Bowei He, Xue Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.23013 · **PDF:** https://arxiv.org/pdf/2603.23013v1
+
+## 2363. AgentRAE: Remote Action Execution through Notification-based Visual Backdoors against Screenshots-based Mobile GUI Agents
+
+- **ID:** `2603.23007` · **primary:** `cs.CR` · **Submitted:** 2026-03-24
+- **Authors:** Yutao Luo, Haotian Zhu, Shuchao Pang, et al.
+- **abs:** https://arxiv.org/abs/2603.23007 · **PDF:** https://arxiv.org/pdf/2603.23007v1
+
+## 2364. Can Large Language Models Reason and Optimize Under Constraints?
+
+- **ID:** `2603.23004` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Fabien Bernier, Salah Ghamizi, Pantelis Dogoulis, et al.
+- **abs:** https://arxiv.org/abs/2603.23004 · **PDF:** https://arxiv.org/pdf/2603.23004v1
+
+## 2365. On the use of Aggregation Operators to improve Human Identification using Dental Records
+
+- **ID:** `2603.23003` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Antonio D. Villegas-Yeguas, Guillermo R-García, Tzipi Kahana, et al.
+- **abs:** https://arxiv.org/abs/2603.23003 · **PDF:** https://arxiv.org/pdf/2603.23003v1
+
+## 2366. PaperVoyager : Building Interactive Web with Visual Language Models
+
+- **ID:** `2603.22999` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Dasen Dai, Biao Wu, Meng Fang, et al.
+- **abs:** https://arxiv.org/abs/2603.22999 · **PDF:** https://arxiv.org/pdf/2603.22999v2
+
+## 2367. Robustness Quantification and Uncertainty Quantification: Comparing Two Methods for Assessing the Reliability of Classifier Predictions
+
+- **ID:** `2603.22988` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Adrián Detavernier, Jasper De Bock
+- **abs:** https://arxiv.org/abs/2603.22988 · **PDF:** https://arxiv.org/pdf/2603.22988v1
+
+## 2368. A Critical Review on the Effectiveness and Privacy Threats of Membership Inference Attacks
+
+- **ID:** `2603.22987` · **primary:** `cs.CR` · **Submitted:** 2026-03-24
+- **Authors:** Najeeb Jebreel, David Sánchez, Josep Domingo-Ferrer
+- **abs:** https://arxiv.org/abs/2603.22987 · **PDF:** https://arxiv.org/pdf/2603.22987v1
+
+## 2369. Beyond Hate: Differentiating Uncivil and Intolerant Speech in Multimodal Content Moderation
+
+- **ID:** `2603.22985` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Nils A. Herrmann, Tobias Eder, Jingyi He, et al.
+- **abs:** https://arxiv.org/abs/2603.22985 · **PDF:** https://arxiv.org/pdf/2603.22985v1
+
+## 2370. Can Graph Foundation Models Generalize Over Architecture?
+
+- **ID:** `2603.22984` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Benjamin Gutteridge, Michael Bronstein, Xiaowen Dong
+- **abs:** https://arxiv.org/abs/2603.22984 · **PDF:** https://arxiv.org/pdf/2603.22984v1
+
+## 2371. Persistence-based topological optimization: a survey
+
+- **ID:** `2603.24613` · **primary:** `cs.CG` · **Submitted:** 2026-03-24
+- **Authors:** Mathieu Carriere, Yuichi Ike, Théo Lacombe, et al.
+- **abs:** https://arxiv.org/abs/2603.24613 · **PDF:** https://arxiv.org/pdf/2603.24613v1
+
+## 2372. JFTA-Bench: Evaluate LLM's Ability of Tracking and Analyzing Malfunctions Using Fault Trees
+
+- **ID:** `2603.22978` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Yuhui Wang, Zhixiong Yang, Ming Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.22978 · **PDF:** https://arxiv.org/pdf/2603.22978v1
+
+## 2373. DariMis: Harm-Aware Modeling for Dari Misinformation Detection on YouTube
+
+- **ID:** `2603.22977` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Jawid Ahmad Baktash, Mosa Ebrahimi, Mohammad Zarif Joya, et al.
+- **abs:** https://arxiv.org/abs/2603.22977 · **PDF:** https://arxiv.org/pdf/2603.22977v1
+
+## 2374. Causal Reconstruction of Sentiment Signals from Sparse News Data
+
+- **ID:** `2603.23568` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Stefania Stan, Marzio Lunghi, Vito Vargetto, et al.
+- **abs:** https://arxiv.org/abs/2603.23568 · **PDF:** https://arxiv.org/pdf/2603.23568v1
+
+## 2375. Where Experts Disagree, Models Fail: Detecting Implicit Legal Citations in French Court Decisions
+
+- **ID:** `2603.22973` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Avrile Floro, Tamara Dhorasoo, Soline Pellez, et al.
+- **abs:** https://arxiv.org/abs/2603.22973 · **PDF:** https://arxiv.org/pdf/2603.22973v1
+
+## 2376. Beyond Theoretical Bounds: Empirical Privacy Loss Calibration for Text Rewriting Under Local Differential Privacy
+
+- **ID:** `2603.22968` · **primary:** `cs.CR` · **Submitted:** 2026-03-24
+- **Authors:** Weijun Li, Arnaud Grivet Sébert, Qiongkai Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.22968 · **PDF:** https://arxiv.org/pdf/2603.22968v1
+
+## 2377. Set-Valued Prediction for Large Language Models with Feasibility-Aware Coverage Guarantees
+
+- **ID:** `2603.22966` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Ye Li, Anqi Hu, Yuanchang Ye, et al.
+- **abs:** https://arxiv.org/abs/2603.22966 · **PDF:** https://arxiv.org/pdf/2603.22966v1
+
+## 2378. A PAC-Bayesian approach to generalization for quantum models
+
+- **ID:** `2603.22964` · **primary:** `quant-ph` · **Submitted:** 2026-03-24
+- **Authors:** Pablo Rodriguez-Grasa, Matthias C. Caro, Jens Eisert, et al.
+- **abs:** https://arxiv.org/abs/2603.22964 · **PDF:** https://arxiv.org/pdf/2603.22964v1
+
+## 2379. Asymptotic Learning Curves for Diffusion Models with Random Features Score and Manifold Data
+
+- **ID:** `2603.22962` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Anand Jerry George, Nicolas Macris
+- **abs:** https://arxiv.org/abs/2603.22962 · **PDF:** https://arxiv.org/pdf/2603.22962v1
+
+## 2380. Stepwise Variational Inference with Vine Copulas
+
+- **ID:** `2603.22959` · **primary:** `stat.ML` · **Submitted:** 2026-03-24
+- **Authors:** Elisabeth Griesbauer, Leiv Rønneberg, Arnoldo Frigessi, et al.
+- **abs:** https://arxiv.org/abs/2603.22959 · **PDF:** https://arxiv.org/pdf/2603.22959v1
+
+## 2381. AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization
+
+- **ID:** `2603.23566` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Jiehao Wu, Zixiao Huang, Wenhao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.23566 · **PDF:** https://arxiv.org/pdf/2603.23566v1
+
+## 2382. Safe Reinforcement Learning with Preference-based Constraint Inference
+
+- **ID:** `2603.23565` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Chenglin Li, Guangchun Ruan, Hua Geng
+- **abs:** https://arxiv.org/abs/2603.23565 · **PDF:** https://arxiv.org/pdf/2603.23565v1
+
+## 2383. Privacy-Preserving EHR Data Transformation via Geometric Operators: A Human-AI Co-Design Technical Report
+
+- **ID:** `2603.22954` · **primary:** `cs.CR` · **Submitted:** 2026-03-24
+- **Authors:** Maolin Wang, Beining Bao, Gan Yuan, et al.
+- **abs:** https://arxiv.org/abs/2603.22954 · **PDF:** https://arxiv.org/pdf/2603.22954v1
+
+## 2384. Weak-PDE-Net: Discovering Open-Form PDEs via Differentiable Symbolic Networks and Weak Formulation
+
+- **ID:** `2603.22951` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Xinxin Li, Xingyu Cui, Jin Qi, et al.
+- **abs:** https://arxiv.org/abs/2603.22951 · **PDF:** https://arxiv.org/pdf/2603.22951v1
+
+## 2385. PersonalQ: Select, Quantize, and Serve Personalized Diffusion Models for Efficient Inference
+
+- **ID:** `2603.22943` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Qirui Wang, Qi Guo, Yiding Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.22943 · **PDF:** https://arxiv.org/pdf/2603.22943v1
+
+## 2386. Optimizing Small Language Models for NL2SQL via Chain-of-Thought Fine-Tuning
+
+- **ID:** `2603.22942` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Anshul Solanki, Sanchit Latawa, Koushik Chakraborty, et al.
+- **abs:** https://arxiv.org/abs/2603.22942 · **PDF:** https://arxiv.org/pdf/2603.22942v1
+
+## 2387. FixationFormer: Direct Utilization of Expert Gaze Trajectories for Chest X-Ray Classification
+
+- **ID:** `2603.22939` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Daniel Beckmann, Benjamin Risse
+- **abs:** https://arxiv.org/abs/2603.22939 · **PDF:** https://arxiv.org/pdf/2603.22939v1
+
+## 2388. Ran Score: a LLM-based Evaluation Score for Radiology Report Generation
+
+- **ID:** `2603.22935` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Ran Zhang, Yucong Lin, Zhaoli Su, et al.
+- **abs:** https://arxiv.org/abs/2603.22935 · **PDF:** https://arxiv.org/pdf/2603.22935v1
+
+## 2389. ProGRank: Probe-Gradient Reranking to Defend Dense-Retriever RAG from Corpus Poisoning
+
+- **ID:** `2603.22934` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Xiangyu Yin, Yi Qi, Chih-Hong Cheng
+- **abs:** https://arxiv.org/abs/2603.22934 · **PDF:** https://arxiv.org/pdf/2603.22934v2
+
+## 2390. Quality Over Clicks: Intrinsic Quality-Driven Iterative Reinforcement Learning for Cold-Start E-Commerce Query Suggestion
+
+- **ID:** `2603.22922` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Qi Sun, Kejun Xiao, Huaipeng Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.22922 · **PDF:** https://arxiv.org/pdf/2603.22922v1
+
+## 2391. The EU AI Act and the Rights-based Approach to Technological Governance
+
+- **ID:** `2603.22920` · **primary:** `cs.CY` · **Submitted:** 2026-03-24
+- **Authors:** Georgios Pavlidis
+- **abs:** https://arxiv.org/abs/2603.22920 · **PDF:** https://arxiv.org/pdf/2603.22920v1
+
+## 2392. EVA: Efficient Reinforcement Learning for End-to-End Video Agent
+
+- **ID:** `2603.22918` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Yaolun Zhang, Ruohui Wang, Jiahao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.22918 · **PDF:** https://arxiv.org/pdf/2603.22918v2
+
+## 2393. Multilingual KokoroChat: A Multi-LLM Ensemble Translation Method for Creating a Multilingual Counseling Dialogue Dataset
+
+- **ID:** `2603.22913` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Ryoma Suzuki, Zhiyang Qi, Michimasa Inaba
+- **abs:** https://arxiv.org/abs/2603.22913 · **PDF:** https://arxiv.org/pdf/2603.22913v1
+
+## 2394. From the AI Act to a European AI Agency: Completing the Union's Regulatory Architecture
+
+- **ID:** `2603.22912` · **primary:** `cs.CY` · **Submitted:** 2026-03-24
+- **Authors:** Georgios Pavlidis
+- **abs:** https://arxiv.org/abs/2603.22912 · **PDF:** https://arxiv.org/pdf/2603.22912v1
+
+## 2395. ForestPrune: High-ratio Visual Token Compression for Video Multimodal Large Language Models via Spatial-Temporal Forest Modeling
+
+- **ID:** `2603.22911` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Shaobo Ju, Baiyang Song, Tao Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.22911 · **PDF:** https://arxiv.org/pdf/2603.22911v1
+
+## 2396. EchoKV: Efficient KV Cache Compression via Similarity-Based Reconstruction
+
+- **ID:** `2603.22910` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Yixuan Wang, Shiyu Ji, Yijun Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.22910 · **PDF:** https://arxiv.org/pdf/2603.22910v1
+
+## 2397. Aesthetic Assessment of Chinese Handwritings Based on Vision Language Models
+
+- **ID:** `2603.26768` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Chen Zheng, Yuxuan Lai, Haoyang Lu, et al.
+- **abs:** https://arxiv.org/abs/2603.26768 · **PDF:** https://arxiv.org/pdf/2603.26768v1
+
+## 2398. Dual-Teacher Distillation with Subnetwork Rectification for Black-Box Domain Adaptation
+
+- **ID:** `2603.22908` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Zhe Zhang, Jing Li, Wanli Xue, et al.
+- **abs:** https://arxiv.org/abs/2603.22908 · **PDF:** https://arxiv.org/pdf/2603.22908v2
+
+## 2399. Separating Diagnosis from Control: Auditable Policy Adaptation in Agent-Based Simulations with LLM-Based Diagnostics
+
+- **ID:** `2603.22904` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Shaoxin Zhong, Yuchen Su, Michael Witbrock
+- **abs:** https://arxiv.org/abs/2603.22904 · **PDF:** https://arxiv.org/pdf/2603.22904v1
+
+## 2400. Off-Policy Evaluation and Learning for Survival Outcomes under Censoring
+
+- **ID:** `2603.22900` · **primary:** `stat.ME` · **Submitted:** 2026-03-24
+- **Authors:** Kohsuke Kubota, Mitsuhiro Takahashi, Yuta Saito
+- **abs:** https://arxiv.org/abs/2603.22900 · **PDF:** https://arxiv.org/pdf/2603.22900v1
+
+## 2401. VLGOR: Visual-Language Knowledge Guided Offline Reinforcement Learning for Generalizable Agents
+
+- **ID:** `2603.22892` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Pengsen Liu, Maosen Zeng, Nan Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.22892 · **PDF:** https://arxiv.org/pdf/2603.22892v1
+
+## 2402. Conditionally Identifiable Latent Representation for Multivariate Time Series with Structural Dynamics
+
+- **ID:** `2603.22886` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Minkey Chang, Jae-Young Kim
+- **abs:** https://arxiv.org/abs/2603.22886 · **PDF:** https://arxiv.org/pdf/2603.22886v1
+
+## 2403. A Heterogeneous Long-Micro Scale Cascading Architecture for General Aviation Health Management
+
+- **ID:** `2603.22885` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Xinhang Chen, Zhihuan Wei, Yang Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.22885 · **PDF:** https://arxiv.org/pdf/2603.22885v3
+
+## 2404. TreeTeaming: Autonomous Red-Teaming of Vision-Language Models via Hierarchical Strategy Exploration
+
+- **ID:** `2603.22882` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Chunxiao Li, Lijun Li, Jing Shao
+- **abs:** https://arxiv.org/abs/2603.22882 · **PDF:** https://arxiv.org/pdf/2603.22882v1
+
+## 2405. Confidence Calibration under Ambiguous Ground Truth
+
+- **ID:** `2603.22879` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Linwei Tao, Haoyang Luo, Minjing Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.22879 · **PDF:** https://arxiv.org/pdf/2603.22879v1
+
+## 2406. Continuous Optimization for Satisfiability Modulo Theories on Linear Real Arithmetic
+
+- **ID:** `2603.22877` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Yunuo Cen, Daniel Ebler, Xuanyao Fong
+- **abs:** https://arxiv.org/abs/2603.22877 · **PDF:** https://arxiv.org/pdf/2603.22877v1
+
+## 2407. Grounding Sim-to-Real Generalization in Dexterous Manipulation: An Empirical Study with Vision-Language-Action Models
+
+- **ID:** `2603.22876` · **primary:** `cs.RO` · **Submitted:** 2026-03-24
+- **Authors:** Ruixing Jin, Zicheng Zhu, Ruixiang Ouyang, et al.
+- **abs:** https://arxiv.org/abs/2603.22876 · **PDF:** https://arxiv.org/pdf/2603.22876v1
+
+## 2408. Dynamical Systems Theory Behind a Hierarchical Reasoning Model
+
+- **ID:** `2603.22871` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Vasiliy A. Es'kin, Mikhail E. Smorkalov
+- **abs:** https://arxiv.org/abs/2603.22871 · **PDF:** https://arxiv.org/pdf/2603.22871v1
+
+## 2409. Chain-of-Authorization: Internalizing Authorization into Large Language Models via Reasoning Trajectories
+
+- **ID:** `2603.22869` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Yang Li, Yule Liu, Xinlei He, et al.
+- **abs:** https://arxiv.org/abs/2603.22869 · **PDF:** https://arxiv.org/pdf/2603.22869v1
+
+## 2410. Agent-Sentry: Bounding LLM Agents via Execution Provenance
+
+- **ID:** `2603.22868` · **primary:** `cs.CR` · **Submitted:** 2026-03-24
+- **Authors:** Rohan Sequeira, Stavros Damianakis, Umar Iqbal, et al.
+- **abs:** https://arxiv.org/abs/2603.22868 · **PDF:** https://arxiv.org/pdf/2603.22868v1
+
+## 2411. The Evolution of Tool Use in LLM Agents: From Single-Tool Call to Multi-Tool Orchestration
+
+- **ID:** `2603.22862` · **primary:** `cs.SE` · **Submitted:** 2026-03-24
+- **Authors:** Haoyuan Xu, Chang Li, Xinyan Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.22862 · **PDF:** https://arxiv.org/pdf/2603.22862v2
+
+## 2412. The Coordinate System Problem in Persistent Structural Memory for Neural Architectures
+
+- **ID:** `2603.22858` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Abhinaba Basu
+- **abs:** https://arxiv.org/abs/2603.22858 · **PDF:** https://arxiv.org/pdf/2603.22858v1
+
+## 2413. TorR: Towards Brain-Inspired Task-Oriented Reasoning via Cache-Oriented Algorithm-Architecture Co-design
+
+- **ID:** `2603.22855` · **primary:** `cs.AR` · **Submitted:** 2026-03-24
+- **Authors:** Hyunwoo Oh, SungHeon Jeong, Suyeon Jang, et al.
+- **abs:** https://arxiv.org/abs/2603.22855 · **PDF:** https://arxiv.org/pdf/2603.22855v1
+
+## 2414. Avoiding Over-smoothing in Social Media Rumor Detection with Pre-trained Propagation Tree Transformer
+
+- **ID:** `2603.22854` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Chaoqun Cui, Caiyan Jia
+- **abs:** https://arxiv.org/abs/2603.22854 · **PDF:** https://arxiv.org/pdf/2603.22854v1
+
+## 2415. Agent Audit: A Security Analysis System for LLM Agent Applications
+
+- **ID:** `2603.22853` · **primary:** `cs.CR` · **Submitted:** 2026-03-24
+- **Authors:** Haiyue Zhang, Yi Nian, Yue Zhao
+- **abs:** https://arxiv.org/abs/2603.22853 · **PDF:** https://arxiv.org/pdf/2603.22853v1
+
+## 2416. UniQueR: Unified Query-based Feedforward 3D Reconstruction
+
+- **ID:** `2603.22851` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Chensheng Peng, Quentin Herau, Jiezhi Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.22851 · **PDF:** https://arxiv.org/pdf/2603.22851v1
+
+## 2417. CoMaTrack: Competitive Multi-Agent Game-Theoretic Tracking with Vision-Language-Action Models
+
+- **ID:** `2603.22846` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Youzhi Liu, Li Gao, Liu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.22846 · **PDF:** https://arxiv.org/pdf/2603.22846v2
+
+## 2418. PhySe-RPO: Physics and Semantics Guided Relative Policy Optimization for Diffusion-Based Surgical Smoke Removal
+
+- **ID:** `2603.22844` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Zining Fang, Chunhui Liu, Bin Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.22844 · **PDF:** https://arxiv.org/pdf/2603.22844v2
+
+## 2419. UAV-DETR: DETR for Anti-Drone Target Detection
+
+- **ID:** `2603.22841` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Jun Yang, Dong Wang, Hongxu Yin, et al.
+- **abs:** https://arxiv.org/abs/2603.22841 · **PDF:** https://arxiv.org/pdf/2603.22841v1
+
+## 2420. URA-Net: Uncertainty-Integrated Anomaly Perception and Restoration Attention Network for Unsupervised Anomaly Detection
+
+- **ID:** `2603.22840` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Wei Luo, Peng Xing, Yunkang Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.22840 · **PDF:** https://arxiv.org/pdf/2603.22840v1
+
+## 2421. Analysing LLM Persona Generation and Fairness Interpretation in Polarised Geopolitical Contexts
+
+- **ID:** `2603.22837` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Maida Aizaz, Quang Minh Nguyen
+- **abs:** https://arxiv.org/abs/2603.22837 · **PDF:** https://arxiv.org/pdf/2603.22837v1
+
+## 2422. Improving Safety Alignment via Balanced Direct Preference Optimization
+
+- **ID:** `2603.22829` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Shiji Zhao, Mengyang Wang, Shukun Xiong, et al.
+- **abs:** https://arxiv.org/abs/2603.22829 · **PDF:** https://arxiv.org/pdf/2603.22829v1
+
+## 2423. Towards The Implicit Bias on Multiclass Separable Data Under Norm Constraints
+
+- **ID:** `2603.22824` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Shengping Xie, Zekun Wu, Quan Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.22824 · **PDF:** https://arxiv.org/pdf/2603.22824v1
+
+## 2424. Empirical Comparison of Agent Communication Protocols for Task Orchestration
+
+- **ID:** `2603.22823` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Ivan Dobrovolskyi
+- **abs:** https://arxiv.org/abs/2603.22823 · **PDF:** https://arxiv.org/pdf/2603.22823v2
+
+## 2425. RadTimeline: Timeline Summarization for Longitudinal Radiological Lung Findings
+
+- **ID:** `2603.22820` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Sitong Zhou, Meliha Yetisgen, Mari Ostendorf
+- **abs:** https://arxiv.org/abs/2603.22820 · **PDF:** https://arxiv.org/pdf/2603.22820v1
+
+## 2426. TDATR: Improving End-to-End Table Recognition via Table Detail-Aware Learning and Cell-Level Visual Alignment
+
+- **ID:** `2603.22819` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Chunxia Qin, Chenyu Liu, Pengcheng Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.22819 · **PDF:** https://arxiv.org/pdf/2603.22819v1
+
+## 2427. When AI Shows Its Work, Is It Actually Working? Step-Level Evaluation Reveals Frontier Language Models Frequently Bypass Their Own Reasoning
+
+- **ID:** `2603.22816` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Abhinaba Basu, Pavan Chakraborty
+- **abs:** https://arxiv.org/abs/2603.22816 · **PDF:** https://arxiv.org/pdf/2603.22816v1
+
+## 2428. Focus, Don't Prune: Identifying Instruction-Relevant Regions for Information-Rich Image Understanding
+
+- **ID:** `2603.22815` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Mincheol Kwon, Minseung Lee, Seonga Choi, et al.
+- **abs:** https://arxiv.org/abs/2603.22815 · **PDF:** https://arxiv.org/pdf/2603.22815v1
+
+## 2429. Learning What Matters Now: Dynamic Preference Inference under Contextual Shifts
+
+- **ID:** `2603.22813` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Xianwei Cao, Dou Quan, Zhenliang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.22813 · **PDF:** https://arxiv.org/pdf/2603.22813v1
+
+## 2430. Efficient Hallucination Detection: Adaptive Bayesian Estimation of Semantic Entropy with Guided Semantic Exploration
+
+- **ID:** `2603.22812` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Qiyao Sun, Xingming Li, Xixiang He, et al.
+- **abs:** https://arxiv.org/abs/2603.22812 · **PDF:** https://arxiv.org/pdf/2603.22812v1
+
+## 2431. Universal and efficient graph neural networks with dynamic attention for machine learning interatomic potentials
+
+- **ID:** `2603.22810` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Shuyu Bi, Zhede Zhao, Qiangchao Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.22810 · **PDF:** https://arxiv.org/pdf/2603.22810v1
+
+## 2432. Combinatorial Privacy: Private Multi-Party Bitstream Grand Sum by Hiding in Birkhoff Polytopes
+
+- **ID:** `2603.22808` · **primary:** `cs.CR` · **Submitted:** 2026-03-24
+- **Authors:** Praneeth Vepakomma
+- **abs:** https://arxiv.org/abs/2603.22808 · **PDF:** https://arxiv.org/pdf/2603.22808v4
+
+## 2433. Transformers Trained via Gradient Descent Can Provably Learn a Class of Teacher Models
+
+- **ID:** `2603.22801` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Chenyang Zhang, Qingyue Zhao, Quanquan Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.22801 · **PDF:** https://arxiv.org/pdf/2603.22801v1
+
+## 2434. Span Modeling for Idiomaticity and Figurative Language Detection with Span Contrastive Loss
+
+- **ID:** `2603.22799` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Blake Matheny, Phuong Minh Nguyen, Minh Le Nguyen
+- **abs:** https://arxiv.org/abs/2603.22799 · **PDF:** https://arxiv.org/pdf/2603.22799v1
+
+## 2435. PhotoAgent: A Robotic Photographer with Spatial and Aesthetic Understanding
+
+- **ID:** `2603.22796` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Lirong Che, Zhenfeng Gan, Yanbo Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.22796 · **PDF:** https://arxiv.org/pdf/2603.22796v1
+
+## 2436. Synthetic Mixed Training: Scaling Parametric Knowledge Acquisition Beyond RAG
+
+- **ID:** `2603.23562` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Seungju Han, Konwoo Kim, Chanwoo Park, et al.
+- **abs:** https://arxiv.org/abs/2603.23562 · **PDF:** https://arxiv.org/pdf/2603.23562v2
+
+## 2437. Reliable Classroom AI via Neuro-Symbolic Multimodal Reasoning
+
+- **ID:** `2603.22793` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Sina Bagheri Nezhad
+- **abs:** https://arxiv.org/abs/2603.22793 · **PDF:** https://arxiv.org/pdf/2603.22793v1
+
+## 2438. ABSTRAL: Automatic Design of Multi-Agent Systems Through Iterative Refinement and Topology Optimization
+
+- **ID:** `2603.22791` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Weijia Song, Jiashu Yue, Zhe Pang
+- **abs:** https://arxiv.org/abs/2603.22791 · **PDF:** https://arxiv.org/pdf/2603.22791v1
+
+## 2439. Quantum Random Forest for the Regression Problem
+
+- **ID:** `2603.22790` · **primary:** `quant-ph` · **Submitted:** 2026-03-24
+- **Authors:** Kamil Khadiev, Liliya Safina
+- **abs:** https://arxiv.org/abs/2603.22790 · **PDF:** https://arxiv.org/pdf/2603.22790v1
+
+## 2440. Exposure-Normalized Bed and Chair Fall Rates via Continuous AI Monitoring
+
+- **ID:** `2603.22785` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Paolo Gabriel, Peter Rehani, Zack Drumm, et al.
+- **abs:** https://arxiv.org/abs/2603.22785 · **PDF:** https://arxiv.org/pdf/2603.22785v1
+
+## 2441. Caterpillar of Thoughts: The Optimal Test-Time Algorithm for Large Language Models
+
+- **ID:** `2603.22784` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Amir Azarmehr, Soheil Behnezhad, Alma Ghafari
+- **abs:** https://arxiv.org/abs/2603.22784 · **PDF:** https://arxiv.org/pdf/2603.22784v1
+
+## 2442. KARMA: Knowledge-Action Regularized Multimodal Alignment for Personalized Search at Taobao
+
+- **ID:** `2603.22779` · **primary:** `cs.IR` · **Submitted:** 2026-03-24
+- **Authors:** Zhi Sun, Wenming Zhang, Yi Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.22779 · **PDF:** https://arxiv.org/pdf/2603.22779v2
+
+## 2443. AgriPestDatabase-v1.0: A Structured Insect Dataset for Training Agricultural Large Language Model
+
+- **ID:** `2603.22777` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Yagizhan Bilal Durak, Ahsan Ul Islam, Shahidul Islam, et al.
+- **abs:** https://arxiv.org/abs/2603.22777 · **PDF:** https://arxiv.org/pdf/2603.22777v1
+
+## 2444. Explainable Threat Attribution for IoT Networks Using Conditional SHAP and Flow Behavior Modelling
+
+- **ID:** `2603.22771` · **primary:** `cs.CR` · **Submitted:** 2026-03-24
+- **Authors:** Samuel Ozechi, Jennifer Okonkwoabutu
+- **abs:** https://arxiv.org/abs/2603.22771 · **PDF:** https://arxiv.org/pdf/2603.22771v1
+
+## 2445. From Arithmetic to Logic: The Resilience of Logic and Lookup-Based Neural Networks Under Parameter Bit-Flips
+
+- **ID:** `2603.22770` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Alan T. L. Bacellar, Sathvik Chemudupati, Shashank Nag, et al.
+- **abs:** https://arxiv.org/abs/2603.22770 · **PDF:** https://arxiv.org/pdf/2603.22770v1
+
+## 2446. Can LLM Agents Generate Real-World Evidence? Evaluating Observational Studies in Medical Databases
+
+- **ID:** `2603.22767` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Dubai Li, Yuxiang He, Yan Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.22767 · **PDF:** https://arxiv.org/pdf/2603.22767v1
+
+## 2447. From Overload to Convergence: Supporting Multi-Issue Human-AI Negotiation with Bayesian Visualization
+
+- **ID:** `2603.22766` · **primary:** `cs.HC` · **Submitted:** 2026-03-24
+- **Authors:** Mehul Parmar, Chaklam Silpasuwanchai
+- **abs:** https://arxiv.org/abs/2603.22766 · **PDF:** https://arxiv.org/pdf/2603.22766v1
+
+## 2448. DALDALL: Data Augmentation for Lexical and Semantic Diverse in Legal Domain by leveraging LLM-Persona
+
+- **ID:** `2603.22765` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Janghyeok Choi, Jaewon Lee, Sungzoon Cho
+- **abs:** https://arxiv.org/abs/2603.22765 · **PDF:** https://arxiv.org/pdf/2603.22765v1
+
+## 2449. Reconstruction-Guided Slot Curriculum: Addressing Object Over-Fragmentation in Video Object-Centric Learning
+
+- **ID:** `2603.22758` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** WonJun Moon, Hyun Seok Seong, Jae-Pil Heo
+- **abs:** https://arxiv.org/abs/2603.22758 · **PDF:** https://arxiv.org/pdf/2603.22758v1
+
+## 2450. KALAVAI: Predicting When Independent Specialist Fusion Works -- A Quantitative Model for Post-Hoc Cooperative LLM Training
+
+- **ID:** `2603.22755` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Ramchand Kumaresan
+- **abs:** https://arxiv.org/abs/2603.22755 · **PDF:** https://arxiv.org/pdf/2603.22755v2
+
+## 2451. PRISM: A Dual View of LLM Reasoning through Semantic Flow and Latent Computation
+
+- **ID:** `2603.22754` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Ruidi Chang, Jiawei Zhou, Hanjie Chen
+- **abs:** https://arxiv.org/abs/2603.22754 · **PDF:** https://arxiv.org/pdf/2603.22754v1
+
+## 2452. CLiGNet: Clinical Label-Interaction Graph Network for Medical Specialty Classification from Clinical Transcriptions
+
+- **ID:** `2603.22752` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Pronob Kumar Barman, Pronoy Kumar Barman
+- **abs:** https://arxiv.org/abs/2603.22752 · **PDF:** https://arxiv.org/pdf/2603.22752v1
+
+## 2453. REALITrees: Rashomon Ensemble Active Learning for Interpretable Trees
+
+- **ID:** `2603.22750` · **primary:** `stat.ML` · **Submitted:** 2026-03-24
+- **Authors:** Simon D. Nguyen, Hayden McTavish, Kentaro Hoffman, et al.
+- **abs:** https://arxiv.org/abs/2603.22750 · **PDF:** https://arxiv.org/pdf/2603.22750v1
+
+## 2454. Beyond Binary Correctness: Scaling Evaluation of Long-Horizon Agents on Subjective Enterprise Tasks
+
+- **ID:** `2603.22744` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Abhishek Chandwani, Ishan Gupta
+- **abs:** https://arxiv.org/abs/2603.22744 · **PDF:** https://arxiv.org/pdf/2603.22744v1
+
+## 2455. Algorithmic warm starts for Hamiltonian Monte Carlo
+
+- **ID:** `2603.22741` · **primary:** `cs.DS` · **Submitted:** 2026-03-24
+- **Authors:** Matthew S. Zhang, Jason M. Altschuler, Sinho Chewi
+- **abs:** https://arxiv.org/abs/2603.22741 · **PDF:** https://arxiv.org/pdf/2603.22741v1
+
+## 2456. Multitask-Informed Prior for In-Context Learning on Tabular Data: Application to Steel Property Prediction
+
+- **ID:** `2603.22738` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Dimitrios Sinodinos, Bahareh Nikpour, Jack Yi Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.22738 · **PDF:** https://arxiv.org/pdf/2603.22738v1
+
+## 2457. Explanation Generation for Contradiction Reconciliation with LLMs
+
+- **ID:** `2603.22735` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Jason Chan, Zhixue Zhao, Robert Gaizauskas
+- **abs:** https://arxiv.org/abs/2603.22735 · **PDF:** https://arxiv.org/pdf/2603.22735v1
+
+## 2458. How Utilitarian Are OpenAI's Models Really? Replicating and Reinterpreting Pfeffer, Krügel, and Uhl (2025)
+
+- **ID:** `2603.22730` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Johannes Himmelreich
+- **abs:** https://arxiv.org/abs/2603.22730 · **PDF:** https://arxiv.org/pdf/2603.22730v1
+
+## 2459. Behavioral Heterogeneity as Quantum-Inspired Representation
+
+- **ID:** `2603.22729` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Mohammad Elayan, Wissam Kontar
+- **abs:** https://arxiv.org/abs/2603.22729 · **PDF:** https://arxiv.org/pdf/2603.22729v1
+
+## 2460. Spiking Personalized Federated Learning for Brain-Computer Interface-Enabled Immersive Communication
+
+- **ID:** `2603.22727` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Chen Shang, Dinh Thai Hoang, Diep N. Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.22727 · **PDF:** https://arxiv.org/pdf/2603.22727v1
+
+## 2461. Double Coupling Architecture and Training Method for Optimization Problems of Differential Algebraic Equations with Parameters
+
+- **ID:** `2603.22724` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Wenqiang Yang, Wenyuan Wu, Yong Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.22724 · **PDF:** https://arxiv.org/pdf/2603.22724v1
+
+## 2462. HyFI: Hyperbolic Feature Interpolation for Brain-Vision Alignment
+
+- **ID:** `2603.22721` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Sangmin Jo, Wootaek Jeong, Da-Woon Heo, et al.
+- **abs:** https://arxiv.org/abs/2603.22721 · **PDF:** https://arxiv.org/pdf/2603.22721v1
+
+## 2463. Bitboard version of Tetris AI
+
+- **ID:** `2603.26765` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Xingguo Chen, Pingshou Xiong, Zhenyu Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.26765 · **PDF:** https://arxiv.org/pdf/2603.26765v1
+
+## 2464. The No-Clash Teaching Dimension is Bounded by VC Dimension
+
+- **ID:** `2603.23561` · **primary:** `cs.IT` · **Submitted:** 2026-03-24
+- **Authors:** Jiahua Liu, Benchong Li
+- **abs:** https://arxiv.org/abs/2603.23561 · **PDF:** https://arxiv.org/pdf/2603.23561v3
+
+## 2465. PopResume: Causal Fairness Evaluation of LLM/VLM Resume Screeners with Population-Representative Dataset
+
+- **ID:** `2603.22714` · **primary:** `cs.CY` · **Submitted:** 2026-03-24
+- **Authors:** Sumin Yu, Juhyeon Park, Taesup Moon
+- **abs:** https://arxiv.org/abs/2603.22714 · **PDF:** https://arxiv.org/pdf/2603.22714v1
+
+## 2466. Non-Adversarial Imitation Learning Provably Free of Compounding Errors: The Role of Bellman Constraints
+
+- **ID:** `2603.22713` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Tian Xu, Chenyang Wang, Xiaochen Zhai, et al.
+- **abs:** https://arxiv.org/abs/2603.22713 · **PDF:** https://arxiv.org/pdf/2603.22713v1
+
+## 2467. Who Spoke What When? Evaluating Spoken Language Models for Conversational ASR with Semantic and Overlap-Aware Metrics
+
+- **ID:** `2603.22709` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Naohiro Tawara, Samuele Cornell, Alexander Polok, et al.
+- **abs:** https://arxiv.org/abs/2603.22709 · **PDF:** https://arxiv.org/pdf/2603.22709v1
+
+## 2468. Detecting Non-Membership in LLM Training Data via Rank Correlations
+
+- **ID:** `2603.22707` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Pranav Shetty, Mirazul Haque, Zhiqiang Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.22707 · **PDF:** https://arxiv.org/pdf/2603.22707v1
+
+## 2469. Synthetic or Authentic? Building Mental Patient Simulators from Longitudinal Evidence
+
+- **ID:** `2603.22704` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Baihan Li, Bingrui Jin, Kunyao Lan, et al.
+- **abs:** https://arxiv.org/abs/2603.22704 · **PDF:** https://arxiv.org/pdf/2603.22704v1
+
+## 2470. Coordinate Encoding on Linear Grids for Physics-Informed Neural Networks
+
+- **ID:** `2603.22700` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Tetsuro Tsuchino, Motoki Shiga
+- **abs:** https://arxiv.org/abs/2603.22700 · **PDF:** https://arxiv.org/pdf/2603.22700v1
+
+## 2471. WiFi2Cap: Semantic Action Captioning from Wi-Fi CSI via Limb-Level Semantic Alignment
+
+- **ID:** `2603.22690` · **primary:** `cs.CV` · **Submitted:** 2026-03-24
+- **Authors:** Tzu-Ti Wei, Chu-Yu Huang, Yu-Chee Tseng, et al.
+- **abs:** https://arxiv.org/abs/2603.22690 · **PDF:** https://arxiv.org/pdf/2603.22690v1
+
+## 2472. MuQ-Eval: An Open-Source Per-Sample Quality Metric for AI Music Generation Evaluation
+
+- **ID:** `2603.22677` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Di Zhu, Zixuan Li
+- **abs:** https://arxiv.org/abs/2603.22677 · **PDF:** https://arxiv.org/pdf/2603.22677v1
+
+## 2473. Vision-based Deep Learning Analysis of Unordered Biomedical Tabular Datasets via Optimal Spatial Cartography
+
+- **ID:** `2603.22675` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Sakib Mostafa, Tarik Massoud, Maximilian Diehn, et al.
+- **abs:** https://arxiv.org/abs/2603.22675 · **PDF:** https://arxiv.org/pdf/2603.22675v1
+
+## 2474. Improving LLM Predictions via Inter-Layer Structural Encoders
+
+- **ID:** `2603.22665` · **primary:** `cs.CL` · **Submitted:** 2026-03-24
+- **Authors:** Tom Ulanovski, Eyal Blyachman, Maya Bechler-Speicher
+- **abs:** https://arxiv.org/abs/2603.22665 · **PDF:** https://arxiv.org/pdf/2603.22665v1
+
+## 2475. Bounding Box Anomaly Scoring for simple and efficient Out-of-Distribution detection
+
+- **ID:** `2603.22660` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Mohamed Bahi Yahiaoui, Geoffrey Daniel, Loïc Giraldi, et al.
+- **abs:** https://arxiv.org/abs/2603.22660 · **PDF:** https://arxiv.org/pdf/2603.22660v1
+
+## 2476. Generalizing Dynamics Modeling More Easily from Representation Perspective
+
+- **ID:** `2603.22655` · **primary:** `cs.LG` · **Submitted:** 2026-03-24
+- **Authors:** Yiming Wang, Zhengnan Zhang, Genghe Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.22655 · **PDF:** https://arxiv.org/pdf/2603.22655v1
+
+## 2477. Benchmarking Multi-Agent LLM Architectures for Financial Document Processing: A Comparative Study of Orchestration Patterns, Cost-Accuracy Tradeoffs and Production Scaling Strategies
+
+- **ID:** `2603.22651` · **primary:** `cs.AI` · **Submitted:** 2026-03-24
+- **Authors:** Siddhant Kulkarni, Yukta Kulkarni
+- **abs:** https://arxiv.org/abs/2603.22651 · **PDF:** https://arxiv.org/pdf/2603.22651v1
+
+## 2478. AwesomeLit: Towards Hypothesis Generation with Agent-Supported Literature Research
+
+- **ID:** `2603.22648` · **primary:** `cs.HC` · **Submitted:** 2026-03-23
+- **Authors:** Zefei Xie, Yuhan Guo, Kai Xu
+- **abs:** https://arxiv.org/abs/2603.22648 · **PDF:** https://arxiv.org/pdf/2603.22648v1
+
+## 2479. Overfitting and Generalizing with (PAC) Bayesian Prediction in Noisy Binary Classification
+
+- **ID:** `2603.22644` · **primary:** `stat.ML` · **Submitted:** 2026-03-23
+- **Authors:** Xiaohan Zhu, Mesrob I. Ohannessian, Nathan Srebro
+- **abs:** https://arxiv.org/abs/2603.22644 · **PDF:** https://arxiv.org/pdf/2603.22644v1
+
+## 2480. CAPTCHA Solving for Native GUI Agents: Automated Reasoning-Action Data Generation and Self-Corrective Training
+
+- **ID:** `2603.23559` · **primary:** `cs.CR` · **Submitted:** 2026-03-23
+- **Authors:** Yuxi Chen, Haoyu Zhai, Chenkai Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.23559 · **PDF:** https://arxiv.org/pdf/2603.23559v1
+
+## 2481. Multi-Method Validation of Large Language Model Medical Translation Across High- and Low-Resource Languages
+
+- **ID:** `2603.22642` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Chukwuebuka Anyaegbuna, Eduardo Juan Perez Guerrero, Jerry Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.22642 · **PDF:** https://arxiv.org/pdf/2603.22642v1
+
+## 2482. Learning to Trust: How Humans Mentally Recalibrate AI Confidence Signals
+
+- **ID:** `2603.22634` · **primary:** `cs.HC` · **Submitted:** 2026-03-23
+- **Authors:** ZhaoBin Li, Mark Steyvers
+- **abs:** https://arxiv.org/abs/2603.22634 · **PDF:** https://arxiv.org/pdf/2603.22634v1
+
+## 2483. Graph-Aware Late Chunking for Retrieval-Augmented Generation in Biomedical Literature
+
+- **ID:** `2603.22633` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Pouria Mortezaagha, Arya Rahgozar
+- **abs:** https://arxiv.org/abs/2603.22633 · **PDF:** https://arxiv.org/pdf/2603.22633v1
+
+## 2484. LGSE: Lexically Grounded Subword Embedding Initialization for Low-Resource Language Adaptation
+
+- **ID:** `2603.22629` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Hailay Teklehaymanot, Dren Fazlija, Wolfgang Nejdl
+- **abs:** https://arxiv.org/abs/2603.22629 · **PDF:** https://arxiv.org/pdf/2603.22629v1
+
+## 2485. Upper Entropy for 2-Monotone Lower Probabilities
+
+- **ID:** `2603.23558` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Tuan-Anh Vu, Sébastien Destercke, Frédéric Pichon
+- **abs:** https://arxiv.org/abs/2603.23558 · **PDF:** https://arxiv.org/pdf/2603.23558v1
+
+## 2486. Toward Faithful Segmentation Attribution via Benchmarking and Dual-Evidence Fusion
+
+- **ID:** `2603.22624` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Abu Noman Md Sakib, OFM Riaz Rahman Aranya, Kevin Desai, et al.
+- **abs:** https://arxiv.org/abs/2603.22624 · **PDF:** https://arxiv.org/pdf/2603.22624v1
+
+## 2487. To Agree or To Be Right? The Grounding-Sycophancy Tradeoff in Medical Vision-Language Models
+
+- **ID:** `2603.22623` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** OFM Riaz Rahman Aranya, Kevin Desai
+- **abs:** https://arxiv.org/abs/2603.22623 · **PDF:** https://arxiv.org/pdf/2603.22623v1
+
+## 2488. Transfer learning via interpolating structures
+
+- **ID:** `2603.22621` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** T. A. Dardeno, A. J. Hughes, L. A. Bull, et al.
+- **abs:** https://arxiv.org/abs/2603.22621 · **PDF:** https://arxiv.org/pdf/2603.22621v1
+
+## 2489. Causal Discovery in Action: Learning Chain-Reaction Mechanisms from Interventions
+
+- **ID:** `2603.22620` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Panayiotis Panayiotou, Özgür Şimşek
+- **abs:** https://arxiv.org/abs/2603.22620 · **PDF:** https://arxiv.org/pdf/2603.22620v1
+
+## 2490. Bridging the Know-Act Gap via Task-Level Autoregressive Reasoning
+
+- **ID:** `2603.22619` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Jihyun Janice Ahn, Ryo Kamoi, Berk Atil, et al.
+- **abs:** https://arxiv.org/abs/2603.22619 · **PDF:** https://arxiv.org/pdf/2603.22619v1
+
+## 2491. Do Consumers Accept AIs as Moral Compliance Agents?
+
+- **ID:** `2603.22617` · **primary:** `cs.HC` · **Submitted:** 2026-03-23
+- **Authors:** Greg Nyilasy, Abraham Ryan Ade Putra Hito, Jennifer Overbeck, et al.
+- **abs:** https://arxiv.org/abs/2603.22617 · **PDF:** https://arxiv.org/pdf/2603.22617v1
+
+## 2492. Understanding LLM Performance Degradation in Multi-Instance Processing: The Roles of Instance Count and Context Length
+
+- **ID:** `2603.22608` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Jingxuan Chen, Mohammad Taher Pilehvar, Jose Camacho-Collados
+- **abs:** https://arxiv.org/abs/2603.22608 · **PDF:** https://arxiv.org/pdf/2603.22608v1
+
+## 2493. Language Models Can Explain Visual Features via Steering
+
+- **ID:** `2603.22593` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Javier Ferrando, Enrique Lopez-Cuena, Pablo Agustin Martin-Torres, et al.
+- **abs:** https://arxiv.org/abs/2603.22593 · **PDF:** https://arxiv.org/pdf/2603.22593v2
+
+## 2494. Precision-Varying Prediction (PVP): Robustifying ASR systems against adversarial attacks
+
+- **ID:** `2603.22590` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Matías Pizarro, Raghavan Narasimhan, Asja Fischer
+- **abs:** https://arxiv.org/abs/2603.22590 · **PDF:** https://arxiv.org/pdf/2603.22590v1
+
+## 2495. flexvec: SQL Vector Retrieval with Programmatic Embedding Modulation
+
+- **ID:** `2603.22587` · **primary:** `cs.IR` · **Submitted:** 2026-03-23
+- **Authors:** Damian Delmas
+- **abs:** https://arxiv.org/abs/2603.22587 · **PDF:** https://arxiv.org/pdf/2603.22587v1
+
+## 2496. A Foundation Model for Instruction-Conditioned In-Context Time Series Tasks
+
+- **ID:** `2603.22586` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Anish Saha, Konstantin Shmakov
+- **abs:** https://arxiv.org/abs/2603.22586 · **PDF:** https://arxiv.org/pdf/2603.22586v1
+
+## 2497. Lie to Me: How Faithful Is Chain-of-Thought Reasoning in Reasoning Models?
+
+- **ID:** `2603.22582` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Richard J. Young
+- **abs:** https://arxiv.org/abs/2603.22582 · **PDF:** https://arxiv.org/pdf/2603.22582v1
+
+## 2498. STRIATUM-CTF: A Protocol-Driven Agentic Framework for General-Purpose CTF Solving
+
+- **ID:** `2603.22577` · **primary:** `cs.CR` · **Submitted:** 2026-03-23
+- **Authors:** James Hugglestone, Samuel Jacob Chacko, Dawson Stoller, et al.
+- **abs:** https://arxiv.org/abs/2603.22577 · **PDF:** https://arxiv.org/pdf/2603.22577v1
+
+## 2499. CAPITU: A Benchmark for Evaluating Instruction-Following in Brazilian Portuguese with Literary Context
+
+- **ID:** `2603.22576` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Giovana Kerche Bonás, Roseval Malaquias Junior, Marcos Piau, et al.
+- **abs:** https://arxiv.org/abs/2603.22576 · **PDF:** https://arxiv.org/pdf/2603.22576v1
+
+## 2500. Reddit After Roe: A Computational Analysis of Abortion Narratives and Barriers in the Wake of Dobbs
+
+- **ID:** `2603.22566` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Aria Pessianzadeh, Alex H. Poole, Rezvaneh Rezapour
+- **abs:** https://arxiv.org/abs/2603.22566 · **PDF:** https://arxiv.org/pdf/2603.22566v1
+
+## 2501. MIOFlow 2.0: A unified framework for inferring cellular stochastic dynamics from single cell and spatial transcriptomics data
+
+- **ID:** `2603.22564` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Xingzhi Sun, João Felipe Rocha, Brett Phelan, et al.
+- **abs:** https://arxiv.org/abs/2603.22564 · **PDF:** https://arxiv.org/pdf/2603.22564v1
+
+## 2502. Privacy-Preserving Reinforcement Learning from Human Feedback via Decoupled Reward Modeling
+
+- **ID:** `2603.22563` · **primary:** `stat.ML` · **Submitted:** 2026-03-23
+- **Authors:** Young Hyun Cho, Will Wei Sun
+- **abs:** https://arxiv.org/abs/2603.22563 · **PDF:** https://arxiv.org/pdf/2603.22563v1
+
+## 2503. AI Mental Models: Learned Intuition and Deliberation in a Bounded Neural Architecture
+
+- **ID:** `2603.22561` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Laurence Anthony
+- **abs:** https://arxiv.org/abs/2603.22561 · **PDF:** https://arxiv.org/pdf/2603.22561v1
+
+## 2504. Maximum Entropy Relaxation of Multi-Way Cardinality Constraints for Synthetic Population Generation
+
+- **ID:** `2603.22558` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** François Pachet, Jean-Daniel Zucker
+- **abs:** https://arxiv.org/abs/2603.22558 · **PDF:** https://arxiv.org/pdf/2603.22558v1
+
+## 2505. Multimodal Training to Unimodal Deployment: Leveraging Unstructured Data During Training to Optimize Structured Data Only Deployment
+
+- **ID:** `2603.22530` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Zigui Wang, Minghui Sun, Jiang Shu, et al.
+- **abs:** https://arxiv.org/abs/2603.22530 · **PDF:** https://arxiv.org/pdf/2603.22530v1
+
+## 2506. Ego2Web: A Web Agent Benchmark Grounded in Egocentric Videos
+
+- **ID:** `2603.22529` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Shoubin Yu, Lei Shu, Antoine Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.22529 · **PDF:** https://arxiv.org/pdf/2603.22529v1
+
+## 2507. GraphRAG for Engineering Diagrams: ChatP&ID Enables LLM Interaction with P&IDs
+
+- **ID:** `2603.22528` · **primary:** `cs.IR` · **Submitted:** 2026-03-23
+- **Authors:** Achmad Anggawirya Alimin, Artur M. Schweidtmann
+- **abs:** https://arxiv.org/abs/2603.22528 · **PDF:** https://arxiv.org/pdf/2603.22528v1
+
+## 2508. Adversarial Vulnerabilities in Neural Operator Digital Twins: Gradient-Free Attacks on Nuclear Thermal-Hydraulic Surrogates
+
+- **ID:** `2603.22525` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Samrendra Roy, Kazuma Kobayashi, Souvik Chakraborty, et al.
+- **abs:** https://arxiv.org/abs/2603.22525 · **PDF:** https://arxiv.org/pdf/2603.22525v1
+
+## 2509. LLMON: An LLM-native Markup Language to Leverage Structure and Semantics at the LLM Interface
+
+- **ID:** `2603.22519` · **primary:** `cs.SE` · **Submitted:** 2026-03-23
+- **Authors:** Michael Hind, Basel Shbita, Bo Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.22519 · **PDF:** https://arxiv.org/pdf/2603.22519v2
+
+## 2510. High Resolution Flood Extent Detection Using Deep Learning with Random Forest Derived Training Labels
+
+- **ID:** `2603.22518` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Azizbek Nuriddinov, Ebrahim Ahmadisharaf, Mohammad Reza Alizadeh
+- **abs:** https://arxiv.org/abs/2603.22518 · **PDF:** https://arxiv.org/pdf/2603.22518v1
+
+## 2511. Generating and Evaluating Sustainable Procurement Criteria for the Swiss Public Sector using In-Context Prompting with Large Language Models
+
+- **ID:** `2603.22513` · **primary:** `cs.SE` · **Submitted:** 2026-03-23
+- **Authors:** Yingqiang Gao, Veton Matoshi, Luca Rolshoven, et al.
+- **abs:** https://arxiv.org/abs/2603.22513 · **PDF:** https://arxiv.org/pdf/2603.22513v1
+
+## 2512. Hebbian Attractor Networks for Robot Locomotion
+
+- **ID:** `2603.22512` · **primary:** `cs.NE` · **Submitted:** 2026-03-23
+- **Authors:** Alexander Dittrich, Fuda van Diggelen, Dario Floreano
+- **abs:** https://arxiv.org/abs/2603.22512 · **PDF:** https://arxiv.org/pdf/2603.22512v1
+
+## 2513. Do Large Language Models Reduce Research Novelty? Evidence from Information Systems Journals
+
+- **ID:** `2603.22510` · **primary:** `cs.DL` · **Submitted:** 2026-03-23
+- **Authors:** Ali Safari
+- **abs:** https://arxiv.org/abs/2603.22510 · **PDF:** https://arxiv.org/pdf/2603.22510v1
+
+## 2514. OrgForge-IT: A Verifiable Synthetic Benchmark for LLM-Based Insider Threat Detection
+
+- **ID:** `2603.22499` · **primary:** `cs.CR` · **Submitted:** 2026-03-23
+- **Authors:** Jeffrey Flynt
+- **abs:** https://arxiv.org/abs/2603.22499 · **PDF:** https://arxiv.org/pdf/2603.22499v1
+
+## 2515. Rashid: A Cipher-Based Framework for Exploring In-Context Language Learning
+
+- **ID:** `2603.22497` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Niyati Bafna, Ryan Soh-Eun Shim, Barbara Plank, et al.
+- **abs:** https://arxiv.org/abs/2603.22497 · **PDF:** https://arxiv.org/pdf/2603.22497v1
+
+## 2516. Tiny Inference-Time Scaling with Latent Verifiers
+
+- **ID:** `2603.22492` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Davide Bucciarelli, Evelyn Turri, Lorenzo Baraldi, et al.
+- **abs:** https://arxiv.org/abs/2603.22492 · **PDF:** https://arxiv.org/pdf/2603.22492v2
+
+## 2517. Tiny-ViT: A Compact Vision Transformer for Efficient and Explainable Potato Leaf Disease Classification
+
+- **ID:** `2603.26761` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Shakil Mia, Umme Habiba, Urmi Akter, et al.
+- **abs:** https://arxiv.org/abs/2603.26761 · **PDF:** https://arxiv.org/pdf/2603.26761v1
+
+## 2518. Cognitive Training for Language Models: Towards General Capabilities via Cross-Entropy Games
+
+- **ID:** `2603.22479` · **primary:** `math.OC` · **Submitted:** 2026-03-23
+- **Authors:** Clément Hongler, Franck Gabriel, Valentin Hartmann, et al.
+- **abs:** https://arxiv.org/abs/2603.22479 · **PDF:** https://arxiv.org/pdf/2603.22479v2
+
+## 2519. Functional Component Ablation Reveals Specialization Patterns in Hybrid Language Model Architectures
+
+- **ID:** `2603.22473` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Hector Borobia, Elies Seguí-Mas, Guillermina Tormo-Carbó
+- **abs:** https://arxiv.org/abs/2603.22473 · **PDF:** https://arxiv.org/pdf/2603.22473v1
+
+## 2520. Wake Up to the Past: Using Memory to Model Fluid Wake Effects on Robots
+
+- **ID:** `2603.22472` · **primary:** `cs.RO` · **Submitted:** 2026-03-23
+- **Authors:** Luca Vendruscolo, Eduardo Sebastián, Amanda Prorok, et al.
+- **abs:** https://arxiv.org/abs/2603.22472 · **PDF:** https://arxiv.org/pdf/2603.22472v1
+
+## 2521. Stability-Preserving Online Adaptation of Neural Closed-loop Maps
+
+- **ID:** `2603.22469` · **primary:** `eess.SY` · **Submitted:** 2026-03-23
+- **Authors:** Danilo Saccani, Luca Furieri, Giancarlo Ferrari-Trecate
+- **abs:** https://arxiv.org/abs/2603.22469 · **PDF:** https://arxiv.org/pdf/2603.22469v1
+
+## 2522. SPDE Methods for Nonparametric Bayesian Posterior Contraction and Laplace Approximation
+
+- **ID:** `2603.22468` · **primary:** `stat.ML` · **Submitted:** 2026-03-23
+- **Authors:** Enric Alberola-Boloix, Ioar Casado-Telletxea
+- **abs:** https://arxiv.org/abs/2603.22468 · **PDF:** https://arxiv.org/pdf/2603.22468v1
+
+## 2523. Color When It Counts: Grayscale-Guided Online Triggering for Always-On Streaming Video Sensing
+
+- **ID:** `2603.22466` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Weitong Cai, Hang Zhang, Yukai Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.22466 · **PDF:** https://arxiv.org/pdf/2603.22466v1
+
+## 2524. A Theoretical Framework for Energy-Aware Gradient Pruning in Federated Learning
+
+- **ID:** `2603.22465` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Emmanouil M. Athanasakos
+- **abs:** https://arxiv.org/abs/2603.22465 · **PDF:** https://arxiv.org/pdf/2603.22465v1
+
+## 2525. LLM-guided headline rewriting for clickability enhancement without clickbait
+
+- **ID:** `2603.22459` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Yehudit Aperstein, Linoy Halifa, Sagiv Bar, et al.
+- **abs:** https://arxiv.org/abs/2603.22459 · **PDF:** https://arxiv.org/pdf/2603.22459v1
+
+## 2526. SkillRouter: Skill Routing for LLM Agents at Scale
+
+- **ID:** `2603.22455` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** YanZhao Zheng, ZhenTao Zhang, Chao Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.22455 · **PDF:** https://arxiv.org/pdf/2603.22455v4
+
+## 2527. Towards Automated Community Notes Generation with Large Vision Language Models for Combating Contextual Deception
+
+- **ID:** `2603.22453` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Jin Ma, Jingwen Yan, Mohammed Aldeen, et al.
+- **abs:** https://arxiv.org/abs/2603.22453 · **PDF:** https://arxiv.org/pdf/2603.22453v1
+
+## 2528. Sparse but Critical: A Token-Level Analysis of Distributional Shifts in RLVR Fine-Tuning of LLMs
+
+- **ID:** `2603.22446` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Haoming Meng, Kexin Huang, Shaohang Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.22446 · **PDF:** https://arxiv.org/pdf/2603.22446v1
+
+## 2529. mmFHE: mmWave Sensing with End-to-End Fully Homomorphic Encryption
+
+- **ID:** `2603.22437` · **primary:** `cs.CR` · **Submitted:** 2026-03-23
+- **Authors:** Tanvir Ahmed, Yixuan Gao, Adnan Armouti, et al.
+- **abs:** https://arxiv.org/abs/2603.22437 · **PDF:** https://arxiv.org/pdf/2603.22437v1
+
+## 2530. CaP-X: A Framework for Benchmarking and Improving Coding Agents for Robot Manipulation
+
+- **ID:** `2603.22435` · **primary:** `cs.RO` · **Submitted:** 2026-03-23
+- **Authors:** Max Fu, Justin Yu, Karim El-Refai, et al.
+- **abs:** https://arxiv.org/abs/2603.22435 · **PDF:** https://arxiv.org/pdf/2603.22435v1
+
+## 2531. Model Predictive Control with Differentiable World Models for Offline Reinforcement Learning
+
+- **ID:** `2603.22430` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Rohan Deb, Stephen J. Wright, Arindam Banerjee
+- **abs:** https://arxiv.org/abs/2603.22430 · **PDF:** https://arxiv.org/pdf/2603.22430v1
+
+## 2532. Neural Structure Embedding for Symbolic Regression via Continuous Structure Search and Coefficient Optimization
+
+- **ID:** `2603.22429` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Fateme Memar, Tao Zhe, Dongjie Wang
+- **abs:** https://arxiv.org/abs/2603.22429 · **PDF:** https://arxiv.org/pdf/2603.22429v1
+
+## 2533. Computational Arbitrage in AI Model Markets
+
+- **ID:** `2603.22404` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Ricardo Olmedo, Bernhard Schölkopf, Moritz Hardt
+- **abs:** https://arxiv.org/abs/2603.22404 · **PDF:** https://arxiv.org/pdf/2603.22404v1
+
+## 2534. Probabilistic modeling over permutations using quantum computers
+
+- **ID:** `2603.22401` · **primary:** `quant-ph` · **Submitted:** 2026-03-23
+- **Authors:** Vasilis Belis, Giulio Crognaletti, Matteo Argenton, et al.
+- **abs:** https://arxiv.org/abs/2603.22401 · **PDF:** https://arxiv.org/pdf/2603.22401v1
+
+## 2535. Latent Style-based Quantum Wasserstein GAN for Drug Design
+
+- **ID:** `2603.22399` · **primary:** `quant-ph` · **Submitted:** 2026-03-23
+- **Authors:** Julien Baglio, Yacine Haddad, Richard Polifka
+- **abs:** https://arxiv.org/abs/2603.22399 · **PDF:** https://arxiv.org/pdf/2603.22399v1
+
+## 2536. WorldCache: Content-Aware Caching for Accelerated Video World Models
+
+- **ID:** `2603.22286` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Umair Nawaz, Ahmed Heakl, Ufaq Khan, et al.
+- **abs:** https://arxiv.org/abs/2603.22286 · **PDF:** https://arxiv.org/pdf/2603.22286v1
+
+## 2537. End-to-End Training for Unified Tokenization and Latent Denoising
+
+- **ID:** `2603.22283` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Shivam Duggal, Xingjian Bai, Zongze Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.22283 · **PDF:** https://arxiv.org/pdf/2603.22283v1
+
+## 2538. UniMotion: A Unified Framework for Motion-Text-Vision Understanding and Generation
+
+- **ID:** `2603.22282` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Ziyi Wang, Xinshun Wang, Shuang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.22282 · **PDF:** https://arxiv.org/pdf/2603.22282v1
+
+## 2539. ThinkJEPA: Empowering Latent World Models with Large Vision-Language Reasoning Model
+
+- **ID:** `2603.22281` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Haichao Zhang, Yijiang Li, Shwai He, et al.
+- **abs:** https://arxiv.org/abs/2603.22281 · **PDF:** https://arxiv.org/pdf/2603.22281v1
+
+## 2540. 3D-Layout-R1: Structured Reasoning for Language-Instructed Spatial Editing
+
+- **ID:** `2603.22279` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Haoyu Zhen, Xiaolong Li, Yilin Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.22279 · **PDF:** https://arxiv.org/pdf/2603.22279v1
+
+## 2541. The Dual Mechanisms of Spatial Reasoning in Vision-Language Models
+
+- **ID:** `2603.22278` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Kelly Cui, Nikhil Prakash, Ayush Raina, et al.
+- **abs:** https://arxiv.org/abs/2603.22278 · **PDF:** https://arxiv.org/pdf/2603.22278v1
+
+## 2542. Scaling DoRA: High-Rank Adaptation via Factored Norms and Fused Kernels
+
+- **ID:** `2603.22276` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Alexandra Zelenin, Alexandra Zhuravlyova
+- **abs:** https://arxiv.org/abs/2603.22276 · **PDF:** https://arxiv.org/pdf/2603.22276v1
+
+## 2543. Decoupling Exploration and Policy Optimization: Uncertainty Guided Tree Search for Hard Exploration
+
+- **ID:** `2603.22273` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Zakaria Mhammedi, James Cohan
+- **abs:** https://arxiv.org/abs/2603.22273 · **PDF:** https://arxiv.org/pdf/2603.22273v3
+
+## 2544. TiCo: Time-Controllable Training for Spoken Dialogue Models
+
+- **ID:** `2603.22267` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Kai-Wei Chang, Wei-Chih Chen, En-Pei Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.22267 · **PDF:** https://arxiv.org/pdf/2603.22267v1
+
+## 2545. Greater accessibility can amplify discrimination in generative AI
+
+- **ID:** `2603.22260` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Carolin Holtermann, Minh Duc Bui, Kaitlyn Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.22260 · **PDF:** https://arxiv.org/pdf/2603.22260v1
+
+## 2546. Characterizing High-Capacity Janus Aminobenzene-Graphene Anode for Sodium-Ion Batteries with Machine Learning
+
+- **ID:** `2603.22254` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-23
+- **Authors:** Claudia Islas-Vargas, L. Ricardo Montoya, Carlos A. Vital-José, et al.
+- **abs:** https://arxiv.org/abs/2603.22254 · **PDF:** https://arxiv.org/pdf/2603.22254v1
+
+## 2547. Confidence-Based Decoding is Provably Efficient for Diffusion Language Models
+
+- **ID:** `2603.22248` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Changxiao Cai, Gen Li
+- **abs:** https://arxiv.org/abs/2603.22248 · **PDF:** https://arxiv.org/pdf/2603.22248v1
+
+## 2548. From Static Templates to Dynamic Runtime Graphs: A Survey of Workflow Optimization for LLM Agents
+
+- **ID:** `2603.22386` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Ling Yue, Kushal Raj Bhandari, Ching-Yun Ko, et al.
+- **abs:** https://arxiv.org/abs/2603.22386 · **PDF:** https://arxiv.org/pdf/2603.22386v1
+
+## 2549. Drop-In Perceptual Optimization for 3D Gaussian Splatting
+
+- **ID:** `2603.23297` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Ezgi Ozyilkan, Zhiqi Chen, Oren Rippel, et al.
+- **abs:** https://arxiv.org/abs/2603.23297 · **PDF:** https://arxiv.org/pdf/2603.23297v1
+
+## 2550. MemDLM: Memory-Enhanced DLM Training
+
+- **ID:** `2603.22241` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Zehua Pei, Hui-Ling Zhen, Weizhe Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.22241 · **PDF:** https://arxiv.org/pdf/2603.22241v1
+
+## 2551. ShapDBM: Exploring Decision Boundary Maps in Shapley Space
+
+- **ID:** `2603.22235` · **primary:** `cs.HC` · **Submitted:** 2026-03-23
+- **Authors:** Luke Watkin, Daniel Archambault, Alex Telea
+- **abs:** https://arxiv.org/abs/2603.22235 · **PDF:** https://arxiv.org/pdf/2603.22235v1
+
+## 2552. One Model, Two Markets: Bid-Aware Generative Recommendation
+
+- **ID:** `2603.22231` · **primary:** `cs.IR` · **Submitted:** 2026-03-23
+- **Authors:** Yanchen Jiang, Zhe Feng, Christopher P. Mah, et al.
+- **abs:** https://arxiv.org/abs/2603.22231 · **PDF:** https://arxiv.org/pdf/2603.22231v1
+
+## 2553. SpatialReward: Verifiable Spatial Reward Modeling for Fine-Grained Spatial Consistency in Text-to-Image Generation
+
+- **ID:** `2603.22228` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Sashuai Zhou, Qiang Zhou, Junpeng Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.22228 · **PDF:** https://arxiv.org/pdf/2603.22228v1
+
+## 2554. Dyadic: A Scalable Platform for Human-Human and Human-AI Conversation Research
+
+- **ID:** `2603.22227` · **primary:** `cs.HC` · **Submitted:** 2026-03-23
+- **Authors:** David M. Markowitz
+- **abs:** https://arxiv.org/abs/2603.22227 · **PDF:** https://arxiv.org/pdf/2603.22227v1
+
+## 2555. Adapting Self-Supervised Speech Representations for Cross-lingual Dysarthria Detection in Parkinson's Disease
+
+- **ID:** `2603.22225` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Abner Hernandez, Eunjung Yeo, Kwanghee Choi, et al.
+- **abs:** https://arxiv.org/abs/2603.22225 · **PDF:** https://arxiv.org/pdf/2603.22225v2
+
+## 2556. Noise Titration: Exact Distributional Benchmarking for Probabilistic Time Series Forecasting
+
+- **ID:** `2603.22219` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Qilin Wang
+- **abs:** https://arxiv.org/abs/2603.22219 · **PDF:** https://arxiv.org/pdf/2603.22219v1
+
+## 2557. Gumbel Distillation for Parallel Text Generation
+
+- **ID:** `2603.22216` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Chi Zhang, Xixi Hu, Bo Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.22216 · **PDF:** https://arxiv.org/pdf/2603.22216v1
+
+## 2558. Evaluating the Reliability and Fidelity of Automated Judgment Systems of Large Language Models
+
+- **ID:** `2603.22214` · **primary:** `cs.CR` · **Submitted:** 2026-03-23
+- **Authors:** Tom Biskupski, Stephan Kleber
+- **abs:** https://arxiv.org/abs/2603.22214 · **PDF:** https://arxiv.org/pdf/2603.22214v1
+
+## 2559. SPA: A Simple but Tough-to-Beat Baseline for Knowledge Injection
+
+- **ID:** `2603.22213` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Kexian Tang, Jiani Wang, Shaowen Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.22213 · **PDF:** https://arxiv.org/pdf/2603.22213v1
+
+## 2560. Identification of physiological shock in intensive care units via Bayesian regime switching models
+
+- **ID:** `2603.22208` · **primary:** `stat.AP` · **Submitted:** 2026-03-23
+- **Authors:** Emmett B. Kendall, Jonathan P. Williams, Curtis B. Storlie, et al.
+- **abs:** https://arxiv.org/abs/2603.22208 · **PDF:** https://arxiv.org/pdf/2603.22208v2
+
+## 2561. Chimera: Latency- and Performance-Aware Multi-agent Serving for Heterogeneous LLMs
+
+- **ID:** `2603.22206` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Kangqi Ni, Wenyue Hua, Xiaoxiang Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.22206 · **PDF:** https://arxiv.org/pdf/2603.22206v1
+
+## 2562. CayleyPy-4: AI-Holography. Towards analogs of holographic string dualities for AI tasks
+
+- **ID:** `2603.22195` · **primary:** `hep-th` · **Submitted:** 2026-03-23
+- **Authors:** A. Chervov, F. Levkovich-Maslyuk, A. Smolensky, et al.
+- **abs:** https://arxiv.org/abs/2603.22195 · **PDF:** https://arxiv.org/pdf/2603.22195v1
+
+## 2563. Seeing is Improving: Visual Feedback for Iterative Text Layout Refinement
+
+- **ID:** `2603.22187` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Junrong Guo, Shancheng Fang, Yadong Qu, et al.
+- **abs:** https://arxiv.org/abs/2603.22187 · **PDF:** https://arxiv.org/pdf/2603.22187v1
+
+## 2564. Enhancing Document-Level Machine Translation via Filtered Synthetic Corpora and Two-Stage LLM Adaptation
+
+- **ID:** `2603.22186` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Ireh Kim, Tesia Sker, Chanwoo Kim
+- **abs:** https://arxiv.org/abs/2603.22186 · **PDF:** https://arxiv.org/pdf/2603.22186v1
+
+## 2565. Revisiting Quantum Code Generation: Where Should Domain Knowledge Live?
+
+- **ID:** `2603.22184` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Oscar Novo, Oscar Bastidas-Jossa, Alberto Calvo, et al.
+- **abs:** https://arxiv.org/abs/2603.22184 · **PDF:** https://arxiv.org/pdf/2603.22184v1
+
+## 2566. Learning When to Act: Interval-Aware Reinforcement Learning with Predictive Temporal Structure
+
+- **ID:** `2603.22384` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Davide Di Gioia
+- **abs:** https://arxiv.org/abs/2603.22384 · **PDF:** https://arxiv.org/pdf/2603.22384v2
+
+## 2567. MARCUS: An agentic, multimodal vision-language model for cardiac diagnosis and management
+
+- **ID:** `2603.22179` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Jack W O'Sullivan, Mohammad Asadi, Lennart Elbe, et al.
+- **abs:** https://arxiv.org/abs/2603.22179 · **PDF:** https://arxiv.org/pdf/2603.22179v1
+
+## 2568. Calibeating Made Simple
+
+- **ID:** `2603.22167` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Yurong Chen, Zhiyi Huang, Michael I. Jordan, et al.
+- **abs:** https://arxiv.org/abs/2603.22167 · **PDF:** https://arxiv.org/pdf/2603.22167v1
+
+## 2569. Causal Evidence that Language Models use Confidence to Drive Behavior
+
+- **ID:** `2603.22161` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Dharshan Kumaran, Nathaniel Daw, Simon Osindero, et al.
+- **abs:** https://arxiv.org/abs/2603.22161 · **PDF:** https://arxiv.org/pdf/2603.22161v1
+
+## 2570. Mixture of Demonstrations for Textual Graph Understanding and Question Answering
+
+- **ID:** `2603.23554` · **primary:** `cs.IR` · **Submitted:** 2026-03-23
+- **Authors:** Yukun Wu, Lihui Liu
+- **abs:** https://arxiv.org/abs/2603.23554 · **PDF:** https://arxiv.org/pdf/2603.23554v1
+
+## 2571. Data Curation for Machine Learning Interatomic Potentials by Determinantal Point Processes
+
+- **ID:** `2603.22160` · **primary:** `stat.AP` · **Submitted:** 2026-03-23
+- **Authors:** Joanna Zou, Youssef Marzouk
+- **abs:** https://arxiv.org/abs/2603.22160 · **PDF:** https://arxiv.org/pdf/2603.22160v1
+
+## 2572. Multimodal Survival Analysis with Locally Deployable Large Language Models
+
+- **ID:** `2603.22158` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Moritz Gögl, Christopher Yau
+- **abs:** https://arxiv.org/abs/2603.22158 · **PDF:** https://arxiv.org/pdf/2603.22158v1
+
+## 2573. RAMPAGE: RAndomized Mid-Point for debiAsed Gradient Extrapolation
+
+- **ID:** `2603.22155` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Abolfazl Hashemi
+- **abs:** https://arxiv.org/abs/2603.22155 · **PDF:** https://arxiv.org/pdf/2603.22155v1
+
+## 2574. dynActivation: A Trainable Activation Family for Adaptive Nonlinearity
+
+- **ID:** `2603.22154` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Alois Bachmann
+- **abs:** https://arxiv.org/abs/2603.22154 · **PDF:** https://arxiv.org/pdf/2603.22154v1
+
+## 2575. Beyond Matching to Tiles: Bridging Unaligned Aerial and Satellite Views for Vision-Only UAV Navigation
+
+- **ID:** `2603.22153` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Kejia Liu, Haoyang Zhou, Ruoyu Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.22153 · **PDF:** https://arxiv.org/pdf/2603.22153v2
+
+## 2576. More Isn't Always Better: Balancing Decision Accuracy and Conformity Pressures in Multi-AI Advice
+
+- **ID:** `2603.22152` · **primary:** `cs.HC` · **Submitted:** 2026-03-23
+- **Authors:** Yuta Tsuchiya, Yukino Baba
+- **abs:** https://arxiv.org/abs/2603.22152 · **PDF:** https://arxiv.org/pdf/2603.22152v1
+
+## 2577. The Semantic Ladder: A Framework for Progressive Formalization of Natural Language Content for Knowledge Graphs and AI Systems
+
+- **ID:** `2603.22136` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Lars Vogt
+- **abs:** https://arxiv.org/abs/2603.22136 · **PDF:** https://arxiv.org/pdf/2603.22136v1
+
+## 2578. Computationally lightweight classifiers with frequentist bounds on predictions
+
+- **ID:** `2603.22128` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Shreeram Murali, Cristian R. Rojas, Dominik Baumann
+- **abs:** https://arxiv.org/abs/2603.22128 · **PDF:** https://arxiv.org/pdf/2603.22128v1
+
+## 2579. Mamba-VMR: Multimodal Query Augmentation via Generated Videos for Precise Temporal Grounding
+
+- **ID:** `2603.22121` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Yunzhuo Sun, Xinyue Liu, Yanyang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.22121 · **PDF:** https://arxiv.org/pdf/2603.22121v1
+
+## 2580. On the Direction of RLVR Updates for LLM Reasoning: Identification and Exploitation
+
+- **ID:** `2603.22117` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Kexin Huang, Haoming Meng, Junkang Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.22117 · **PDF:** https://arxiv.org/pdf/2603.22117v1
+
+## 2581. Multiperspectivity as a Resource for Narrative Similarity Prediction
+
+- **ID:** `2603.22103` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Max Upravitelev, Veronika Solopova, Jing Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.22103 · **PDF:** https://arxiv.org/pdf/2603.22103v1
+
+## 2582. SpecTM: Spectral Targeted Masking for Trustworthy Foundation Models
+
+- **ID:** `2603.22097` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Syed Usama Imtiaz, Mitra Nasr Azadani, Nasrin Alamdari
+- **abs:** https://arxiv.org/abs/2603.22097 · **PDF:** https://arxiv.org/pdf/2603.22097v1
+
+## 2583. GSEM: Graph-based Self-Evolving Memory for Experience Augmented Clinical Reasoning
+
+- **ID:** `2603.22096` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Xiao Han, Yuzheng Fan, Sendong Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.22096 · **PDF:** https://arxiv.org/pdf/2603.22096v1
+
+## 2584. A Context Engineering Framework for Improving Enterprise AI Agents based on Digital-Twin MDP
+
+- **ID:** `2603.22083` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Xi Yang, Aurelie Lozano, Naoki Abe, et al.
+- **abs:** https://arxiv.org/abs/2603.22083 · **PDF:** https://arxiv.org/pdf/2603.22083v1
+
+## 2585. Autoregressive vs. Masked Diffusion Language Models: A Controlled Comparison
+
+- **ID:** `2603.22075` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Caio Vicentino
+- **abs:** https://arxiv.org/abs/2603.22075 · **PDF:** https://arxiv.org/pdf/2603.22075v1
+
+## 2586. MIHT: A Hoeffding Tree for Time Series Classification using Multiple Instance Learning
+
+- **ID:** `2603.22074` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Aurora Esteban, Amelia Zafra, Sebastián Ventura
+- **abs:** https://arxiv.org/abs/2603.22074 · **PDF:** https://arxiv.org/pdf/2603.22074v1
+
+## 2587. PreferRec: Learning and Transferring Pareto Preferences for Multi-objective Re-ranking
+
+- **ID:** `2603.22073` · **primary:** `cs.IR` · **Submitted:** 2026-03-23
+- **Authors:** Wei Zhou, Wuyang Li, Junkai Ji, et al.
+- **abs:** https://arxiv.org/abs/2603.22073 · **PDF:** https://arxiv.org/pdf/2603.22073v1
+
+## 2588. On the Failure of Topic-Matched Contrast Baselines in Multi-Directional Refusal Abliteration
+
+- **ID:** `2603.22061` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Valentin Petrov
+- **abs:** https://arxiv.org/abs/2603.22061 · **PDF:** https://arxiv.org/pdf/2603.22061v1
+
+## 2589. Dual-Space Knowledge Distillation with Key-Query Matching for Large Language Models with Vocabulary Mismatch
+
+- **ID:** `2603.22056` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Stella Eva Tsiapali, Cong-Thanh Do, Kate Knill
+- **abs:** https://arxiv.org/abs/2603.22056 · **PDF:** https://arxiv.org/pdf/2603.22056v1
+
+## 2590. AnimalCLAP: Taxonomy-Aware Language-Audio Pretraining for Species Recognition and Trait Inference
+
+- **ID:** `2603.22053` · **primary:** `cs.SD` · **Submitted:** 2026-03-23
+- **Authors:** Risa Shinoda, Kaede Shiohara, Nakamasa Inoue, et al.
+- **abs:** https://arxiv.org/abs/2603.22053 · **PDF:** https://arxiv.org/pdf/2603.22053v1
+
+## 2591. MAGPI: Multifidelity-Augmented Gaussian Process Inputs for Surrogate Modeling from Scarce Data
+
+- **ID:** `2603.22050` · **primary:** `stat.ML` · **Submitted:** 2026-03-23
+- **Authors:** Atticus Rex, Elizabeth Qian, David Peterson
+- **abs:** https://arxiv.org/abs/2603.22050 · **PDF:** https://arxiv.org/pdf/2603.22050v1
+
+## 2592. Uncertainty-guided Compositional Alignment with Part-to-Whole Semantic Representativeness in Hyperbolic Vision-Language Models
+
+- **ID:** `2603.22042` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Hayeon Kim, Ji Ha Jang, Junghun James Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.22042 · **PDF:** https://arxiv.org/pdf/2603.22042v2
+
+## 2593. RAFL: Generalizable Sim-to-Real of Soft Robots with Residual Acceleration Field Learning
+
+- **ID:** `2603.22039` · **primary:** `cs.RO` · **Submitted:** 2026-03-23
+- **Authors:** Dong Heon Cho, Boyuan Chen
+- **abs:** https://arxiv.org/abs/2603.22039 · **PDF:** https://arxiv.org/pdf/2603.22039v1
+
+## 2594. Future-Interactions-Aware Trajectory Prediction via Braid Theory
+
+- **ID:** `2603.22035` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Caio Azevedo, Stefano Sabatini, Sascha Hornauer, et al.
+- **abs:** https://arxiv.org/abs/2603.22035 · **PDF:** https://arxiv.org/pdf/2603.22035v1
+
+## 2595. On the Interplay of Priors and Overparametrization in Bayesian Neural Network Posteriors
+
+- **ID:** `2603.22030` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Julius Kobialka, Emanuel Sommer, Chris Kolb, et al.
+- **abs:** https://arxiv.org/abs/2603.22030 · **PDF:** https://arxiv.org/pdf/2603.22030v1
+
+## 2596. Do Papers Match Code? A Benchmark and Framework for Paper-Code Consistency Detection in Bioinformatics Software
+
+- **ID:** `2603.22018` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Tianxiang Xu, Xiaoyan Zhu, Xin Lai, et al.
+- **abs:** https://arxiv.org/abs/2603.22018 · **PDF:** https://arxiv.org/pdf/2603.22018v1
+
+## 2597. AdditiveLLM2: A Multi-modal Large Language Model for Additive Manufacturing
+
+- **ID:** `2603.22017` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Peter Pak, Amir Barati Farimani
+- **abs:** https://arxiv.org/abs/2603.22017 · **PDF:** https://arxiv.org/pdf/2603.22017v1
+
+## 2598. ROM: Real-time Overthinking Mitigation via Streaming Detection and Intervention
+
+- **ID:** `2603.22016` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Xinyan Wang, Xiaogeng Liu, Chaowei Xiao
+- **abs:** https://arxiv.org/abs/2603.22016 · **PDF:** https://arxiv.org/pdf/2603.22016v1
+
+## 2599. Retrieving Climate Change Disinformation by Narrative
+
+- **ID:** `2603.22015` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Max Upravitelev, Veronika Solopova, Charlott Jakob, et al.
+- **abs:** https://arxiv.org/abs/2603.22015 · **PDF:** https://arxiv.org/pdf/2603.22015v2
+
+## 2600. On the Challenges and Opportunities of Learned Sparse Retrieval for Code
+
+- **ID:** `2603.22008` · **primary:** `cs.IR` · **Submitted:** 2026-03-23
+- **Authors:** Simon Lupart, Maxime Louis, Thibault Formal, et al.
+- **abs:** https://arxiv.org/abs/2603.22008 · **PDF:** https://arxiv.org/pdf/2603.22008v1
+
+## 2601. A plug-and-play approach with fast uncertainty quantification for weak lensing mass mapping
+
+- **ID:** `2603.22006` · **primary:** `astro-ph.CO` · **Submitted:** 2026-03-23
+- **Authors:** Hubert Leterme, Andreas Tersenov, Jalal Fadili, et al.
+- **abs:** https://arxiv.org/abs/2603.22006 · **PDF:** https://arxiv.org/pdf/2603.22006v1
+
+## 2602. SegMaFormer: A Hybrid State-Space and Transformer Model for Efficient Segmentation
+
+- **ID:** `2603.22002` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Duy D. Nguyen, Phat T. Tran-Truong
+- **abs:** https://arxiv.org/abs/2603.22002 · **PDF:** https://arxiv.org/pdf/2603.22002v1
+
+## 2603. CRPS-Optimal Binning for Conformal Regression
+
+- **ID:** `2603.22000` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Paolo Toccaceli
+- **abs:** https://arxiv.org/abs/2603.22000 · **PDF:** https://arxiv.org/pdf/2603.22000v1
+
+## 2604. λ-GELU: Learning Gating Hardness for Controlled ReLU-ization in Deep Networks
+
+- **ID:** `2603.21991` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Cristian Pérez-Corral, Alberto Fernández-Hernández, Jose I. Mestre, et al.
+- **abs:** https://arxiv.org/abs/2603.21991 · **PDF:** https://arxiv.org/pdf/2603.21991v1
+
+## 2605. TREX: Trajectory Explanations for Multi-Objective Reinforcement Learning
+
+- **ID:** `2603.21988` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Dilina Rajapakse, Juan C. Rosero, Ivana Dusparic
+- **abs:** https://arxiv.org/abs/2603.21988 · **PDF:** https://arxiv.org/pdf/2603.21988v1
+
+## 2606. Symbolic Graph Networks for Robust PDE Discovery from Noisy Sparse Data
+
+- **ID:** `2603.22380` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Xingyu Chen, Junxiu An, Jun Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.22380 · **PDF:** https://arxiv.org/pdf/2603.22380v1
+
+## 2607. LRC-WeatherNet: LiDAR, RADAR, and Camera Fusion Network for Real-time Weather-type Classification in Autonomous Driving
+
+- **ID:** `2603.21987` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Nour Alhuda Albashir, Lars Pernickel, Danial Hamoud, et al.
+- **abs:** https://arxiv.org/abs/2603.21987 · **PDF:** https://arxiv.org/pdf/2603.21987v1
+
+## 2608. BOOST-RPF: Boosted Sequential Trees for Radial Power Flow
+
+- **ID:** `2603.21977` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Ehimare Okoyomon, Christoph Goebel
+- **abs:** https://arxiv.org/abs/2603.21977 · **PDF:** https://arxiv.org/pdf/2603.21977v1
+
+## 2609. SecureBreak -- A dataset towards safe and secure models
+
+- **ID:** `2603.21975` · **primary:** `cs.CR` · **Submitted:** 2026-03-23
+- **Authors:** Marco Arazzi, Vignesh Kumar Kembu, Antonino Nocera
+- **abs:** https://arxiv.org/abs/2603.21975 · **PDF:** https://arxiv.org/pdf/2603.21975v1
+
+## 2610. Demystifying Reinforcement Learning for Long-Horizon Tool-Using Agents: A Comprehensive Recipe
+
+- **ID:** `2603.21972` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Xixi Wu, Qianguo Sun, Ruiyang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.21972 · **PDF:** https://arxiv.org/pdf/2603.21972v1
+
+## 2611. Generating Synthetic Wildlife Health Data from Camera Trap Imagery: A Pipeline for Alopecia and Body Condition Training Data
+
+- **ID:** `2603.26754` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** David Brundage
+- **abs:** https://arxiv.org/abs/2603.26754 · **PDF:** https://arxiv.org/pdf/2603.26754v1
+
+## 2612. Parameter-Efficient Fine-Tuning for Medical Text Summarization: A Comparative Study of Lora, Prompt Tuning, and Full Fine-Tuning
+
+- **ID:** `2603.21970` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Ulugbek Shernazarov, Rostislav Svitsov, Bin Shi
+- **abs:** https://arxiv.org/abs/2603.21970 · **PDF:** https://arxiv.org/pdf/2603.21970v1
+
+## 2613. BHDD: A Burmese Handwritten Digit Dataset
+
+- **ID:** `2603.21966` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Swan Htet Aung, Hein Htet, Htoo Say Wah Khaing, et al.
+- **abs:** https://arxiv.org/abs/2603.21966 · **PDF:** https://arxiv.org/pdf/2603.21966v1
+
+## 2614. Suiren-1.0 Technical Report: A Family of Molecular Foundation Models
+
+- **ID:** `2603.21942` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-23
+- **Authors:** Junyi An, Xinyu Lu, Yun-Fei Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.21942 · **PDF:** https://arxiv.org/pdf/2603.21942v1
+
+## 2615. SLURP-TN : Resource for Tunisian Dialect Spoken Language Understanding
+
+- **ID:** `2603.21940` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Haroun Elleuch, Salima Mdhaffar, Yannick Estève, et al.
+- **abs:** https://arxiv.org/abs/2603.21940 · **PDF:** https://arxiv.org/pdf/2603.21940v1
+
+## 2616. Chronological Contrastive Learning: Few-Shot Progression Assessment in Irreversible Diseases
+
+- **ID:** `2603.21935` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Clemens Watzenböck, Daniel Aletaha, Michaël Deman, et al.
+- **abs:** https://arxiv.org/abs/2603.21935 · **PDF:** https://arxiv.org/pdf/2603.21935v1
+
+## 2617. Camera-Agnostic Pruning of 3D Gaussian Splats via Descriptor-Based Beta Evidence
+
+- **ID:** `2603.21933` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Peter Fasogbon, Ugurcan Budak, Patrice Rondao Alface, et al.
+- **abs:** https://arxiv.org/abs/2603.21933 · **PDF:** https://arxiv.org/pdf/2603.21933v1
+
+## 2618. The Golden Subspace: Where Efficiency Meets Generalization in Continual Test-Time Adaptation
+
+- **ID:** `2603.21928` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Guannan Lai, Da-Wei Zhou, Zhenguo Li, et al.
+- **abs:** https://arxiv.org/abs/2603.21928 · **PDF:** https://arxiv.org/pdf/2603.21928v1
+
+## 2619. Instruction-Tuned, but Not More Verifiable Instruction-Following: A Cross-Task Diagnosis for LoRA Adapters
+
+- **ID:** `2603.22379` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Junyi Zou
+- **abs:** https://arxiv.org/abs/2603.22379 · **PDF:** https://arxiv.org/pdf/2603.22379v1
+
+## 2620. Guideline-grounded retrieval-augmented generation for ophthalmic clinical decision support
+
+- **ID:** `2603.21925` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Shuying Chen, Sen Cui, Zhong Cao
+- **abs:** https://arxiv.org/abs/2603.21925 · **PDF:** https://arxiv.org/pdf/2603.21925v1
+
+## 2621. Deep Reinforcement Learning and The Tale of Two Temporal Difference Errors
+
+- **ID:** `2603.21921` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Juan Sebastian Rojas, Chi-Guhn Lee
+- **abs:** https://arxiv.org/abs/2603.21921 · **PDF:** https://arxiv.org/pdf/2603.21921v1
+
+## 2622. Structural Concentration in Weighted Networks: A Class of Topology-Aware Indices
+
+- **ID:** `2603.21918` · **primary:** `stat.ML` · **Submitted:** 2026-03-23
+- **Authors:** L. Riso, M. G. Zoia
+- **abs:** https://arxiv.org/abs/2603.21918 · **PDF:** https://arxiv.org/pdf/2603.21918v1
+
+## 2623. A Latent Representation Learning Framework for Hyperspectral Image Emulation in Remote Sensing
+
+- **ID:** `2603.21911` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Chedly Ben Azizi, Claire Guilloteau, Gilles Roussel, et al.
+- **abs:** https://arxiv.org/abs/2603.21911 · **PDF:** https://arxiv.org/pdf/2603.21911v1
+
+## 2624. A Novel Method for Enforcing Exactly Dirichlet, Neumann and Robin Conditions on Curved Domain Boundaries for Physics Informed Machine Learning
+
+- **ID:** `2603.21909` · **primary:** `math.NA` · **Submitted:** 2026-03-23
+- **Authors:** Suchuan Dong, Yuchuan Zhang
+- **abs:** https://arxiv.org/abs/2603.21909 · **PDF:** https://arxiv.org/pdf/2603.21909v1
+
+## 2625. SparseDVFS: Sparse-Aware DVFS for Energy-Efficient Edge Inference
+
+- **ID:** `2603.21908` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Ziyang Zhang, Zheshun Wu, Jie Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.21908 · **PDF:** https://arxiv.org/pdf/2603.21908v1
+
+## 2626. SHAPE: Structure-aware Hierarchical Unsupervised Domain Adaptation with Plausibility Evaluation for Medical Image Segmentation
+
+- **ID:** `2603.21904` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Linkuan Zhou, Yinghao Xia, Yufei Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.21904 · **PDF:** https://arxiv.org/pdf/2603.21904v1
+
+## 2627. Ara-Best-RQ: Multi Dialectal Arabic SSL
+
+- **ID:** `2603.21900` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Haroun Elleuch, Ryan Whetten, Salima Mdhaffar, et al.
+- **abs:** https://arxiv.org/abs/2603.21900 · **PDF:** https://arxiv.org/pdf/2603.21900v1
+
+## 2628. Not All Layers Are Created Equal: Adaptive LoRA Ranks for Personalized Image Generation
+
+- **ID:** `2603.21884` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Donald Shenaj, Federico Errica, Antonio Carta
+- **abs:** https://arxiv.org/abs/2603.21884 · **PDF:** https://arxiv.org/pdf/2603.21884v1
+
+## 2629. SmaAT-QMix-UNet: A Parameter-Efficient Vector-Quantized UNet for Precipitation Nowcasting
+
+- **ID:** `2603.21879` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Nikolas Stavrou, Siamak Mehrkanoon
+- **abs:** https://arxiv.org/abs/2603.21879 · **PDF:** https://arxiv.org/pdf/2603.21879v2
+
+## 2630. P^2O: Joint Policy and Prompt Optimization
+
+- **ID:** `2603.21877` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Xinyu Lu, Kaiqi Zhang, Jinglin Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.21877 · **PDF:** https://arxiv.org/pdf/2603.21877v2
+
+## 2631. Disentangling Speaker Traits for Deepfake Source Verification via Chebyshev Polynomial and Riemannian Metric Learning
+
+- **ID:** `2603.21875` · **primary:** `eess.AS` · **Submitted:** 2026-03-23
+- **Authors:** Xi Xuan, Wenxin Zhang, Zhiyu Li, et al.
+- **abs:** https://arxiv.org/abs/2603.21875 · **PDF:** https://arxiv.org/pdf/2603.21875v1
+
+## 2632. Manifold-Aware Exploration for Reinforcement Learning in Video Generation
+
+- **ID:** `2603.21872` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Mingzhe Zheng, Weijie Kong, Yue Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.21872 · **PDF:** https://arxiv.org/pdf/2603.21872v1
+
+## 2633. Adversarial Camouflage
+
+- **ID:** `2603.21867` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Paweł Borsukiewicz, Daniele Lunghi, Melissa Tessa, et al.
+- **abs:** https://arxiv.org/abs/2603.21867 · **PDF:** https://arxiv.org/pdf/2603.21867v1
+
+## 2634. Tacit Knowledge Management with Generative AI: Proposal of the GenAI SECI Model
+
+- **ID:** `2603.21866` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Naoshi Uchihira
+- **abs:** https://arxiv.org/abs/2603.21866 · **PDF:** https://arxiv.org/pdf/2603.21866v1
+
+## 2635. Adaptive Video Distillation: Mitigating Oversaturation and Temporal Collapse in Few-Step Generation
+
+- **ID:** `2603.21864` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Yuyang You, Yongzhi Li, Jiahui Li, et al.
+- **abs:** https://arxiv.org/abs/2603.21864 · **PDF:** https://arxiv.org/pdf/2603.21864v1
+
+## 2636. Holistic Scaling Laws for Optimal Mixture-of-Experts Architecture Optimization
+
+- **ID:** `2603.21862` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Weilin Wan, Jingtao Han, Weizhong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.21862 · **PDF:** https://arxiv.org/pdf/2603.21862v1
+
+## 2637. Reasoning or Rhetoric? An Empirical Analysis of Moral Reasoning Explanations in Large Language Models
+
+- **ID:** `2603.21854` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Aryan Kasat, Smriti Singh, Aman Chadha, et al.
+- **abs:** https://arxiv.org/abs/2603.21854 · **PDF:** https://arxiv.org/pdf/2603.21854v1
+
+## 2638. Sim-to-Real of Humanoid Locomotion Policies via Joint Torque Space Perturbation Injection
+
+- **ID:** `2603.21853` · **primary:** `cs.RO` · **Submitted:** 2026-03-23
+- **Authors:** Junhyeok Rui Cha, Woohyun Cha, Jaeyong Shin, et al.
+- **abs:** https://arxiv.org/abs/2603.21853 · **PDF:** https://arxiv.org/pdf/2603.21853v2
+
+## 2639. All elementary functions from a single binary operator
+
+- **ID:** `2603.21852` · **primary:** `cs.SC` · **Submitted:** 2026-03-23
+- **Authors:** Andrzej Odrzywołek
+- **abs:** https://arxiv.org/abs/2603.21852 · **PDF:** https://arxiv.org/pdf/2603.21852v1
+
+## 2640. Riding Brainwaves in LLM Space: Understanding Activation Patterns Using Individual Neural Signatures
+
+- **ID:** `2603.21847` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Ajan Subramanian, Sumukh Bettadapura, Rohan Sathish
+- **abs:** https://arxiv.org/abs/2603.21847 · **PDF:** https://arxiv.org/pdf/2603.21847v1
+
+## 2641. Agentic Personas for Adaptive Scientific Explanations with Knowledge Graphs
+
+- **ID:** `2603.21846` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Susana Nunes, Tiago Guerreiro, Catia Pesquita
+- **abs:** https://arxiv.org/abs/2603.21846 · **PDF:** https://arxiv.org/pdf/2603.21846v1
+
+## 2642. On the Number of Conditional Independence Tests in Constraint-based Causal Discovery
+
+- **ID:** `2603.21844` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Marc Franquesa Monés, Jiaqi Zhang, Caroline Uhler
+- **abs:** https://arxiv.org/abs/2603.21844 · **PDF:** https://arxiv.org/pdf/2603.21844v1
+
+## 2643. Select, Label, Evaluate: Active Testing in NLP
+
+- **ID:** `2603.21840` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Antonio Purificato, Maria Sofia Bucarelli, Andrea Bacciu, et al.
+- **abs:** https://arxiv.org/abs/2603.21840 · **PDF:** https://arxiv.org/pdf/2603.21840v1
+
+## 2644. Instruction Set and Language for Symbolic Regression
+
+- **ID:** `2603.21836` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Ezequiel Lopez-Rubio, Mario Pascual-Gonzalez
+- **abs:** https://arxiv.org/abs/2603.21836 · **PDF:** https://arxiv.org/pdf/2603.21836v1
+
+## 2645. Deriving Health Metrics from the Photoplethysmogram: Benchmarks and Insights from MIMIC-III-Ext-PPG
+
+- **ID:** `2603.21832` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Mohammad Moulaeifard, Philip J. Aston, Peter H. Charlton, et al.
+- **abs:** https://arxiv.org/abs/2603.21832 · **PDF:** https://arxiv.org/pdf/2603.21832v1
+
+## 2646. Evolutionary Warm-Starts for Reinforcement Learning in Industrial Continuous Control
+
+- **ID:** `2603.26750` · **primary:** `cs.NE` · **Submitted:** 2026-03-23
+- **Authors:** Tom Maus, Stephan Frank, Tobias Glasmachers
+- **abs:** https://arxiv.org/abs/2603.26750 · **PDF:** https://arxiv.org/pdf/2603.26750v1
+
+## 2647. CoRA: Boosting Time Series Foundation Models for Multivariate Forecasting through Correlation-aware Adapter
+
+- **ID:** `2603.21828` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Hanyin Cheng, Xingjian Wu, Yang Shu, et al.
+- **abs:** https://arxiv.org/abs/2603.21828 · **PDF:** https://arxiv.org/pdf/2603.21828v1
+
+## 2648. Training-Free Diffusion-Driven Modeling of Pareto Set Evolution for Dynamic Multiobjective Optimization
+
+- **ID:** `2603.26749` · **primary:** `cs.NE` · **Submitted:** 2026-03-23
+- **Authors:** Jian Guan, Huolong Wu, Zhenzhong Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.26749 · **PDF:** https://arxiv.org/pdf/2603.26749v1
+
+## 2649. BadminSense: Enabling Fine-Grained Badminton Stroke Evaluation on a Single Smartwatch
+
+- **ID:** `2603.21825` · **primary:** `cs.HC` · **Submitted:** 2026-03-23
+- **Authors:** Taizhou Chen, Kai Chen, Xingyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.21825 · **PDF:** https://arxiv.org/pdf/2603.21825v2
+
+## 2650. SteelDefectX: A Coarse-to-Fine Vision-Language Dataset and Benchmark for Generalizable Steel Surface Defect Detection
+
+- **ID:** `2603.21824` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Shuxian Zhao, Jie Gui, Baosheng Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.21824 · **PDF:** https://arxiv.org/pdf/2603.21824v1
+
+## 2651. Politics of Questions in News: A Mixed-Methods Study of Interrogative Stances as Markers of Voice and Power
+
+- **ID:** `2603.21823` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Bros Victor, Barbini Matilde, Gerard Patrick, et al.
+- **abs:** https://arxiv.org/abs/2603.21823 · **PDF:** https://arxiv.org/pdf/2603.21823v1
+
+## 2652. Ctrl-A: Control-Driven Online Data Augmentation
+
+- **ID:** `2603.21819` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Jesper B. Christensen, Ciaran Bench, Spencer A. Thomas, et al.
+- **abs:** https://arxiv.org/abs/2603.21819 · **PDF:** https://arxiv.org/pdf/2603.21819v1
+
+## 2653. LARD 2.0: Enhanced Datasets and Benchmarking for Autonomous Landing Systems
+
+- **ID:** `2603.26748` · **primary:** `cs.RO` · **Submitted:** 2026-03-23
+- **Authors:** Yassine Bougacha, Geoffrey Delhomme, Mélanie Ducoffe, et al.
+- **abs:** https://arxiv.org/abs/2603.26748 · **PDF:** https://arxiv.org/pdf/2603.26748v1
+
+## 2654. Show Me What You Don't Know: Efficient Sampling from Invariant Sets for Model Validation
+
+- **ID:** `2603.21782` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Armand Rousselot, Joran Wendebourg, Ullrich Köthe
+- **abs:** https://arxiv.org/abs/2603.21782 · **PDF:** https://arxiv.org/pdf/2603.21782v1
+
+## 2655. Cluster-Specific Predictive Modeling: A Scalable Solution for Resource-Constrained Wi-Fi Controllers
+
+- **ID:** `2603.21778` · **primary:** `eess.SP` · **Submitted:** 2026-03-23
+- **Authors:** Gianluca Fontanesi, Luca Barbieri, Lorenzo Galati Giordano, et al.
+- **abs:** https://arxiv.org/abs/2603.21778 · **PDF:** https://arxiv.org/pdf/2603.21778v1
+
+## 2656. Abnormalities and Disease Detection in Gastro-Intestinal Tract Images
+
+- **ID:** `2603.22378` · **primary:** `eess.IV` · **Submitted:** 2026-03-23
+- **Authors:** Zeshan Khan, Muhammad Atif Tahir
+- **abs:** https://arxiv.org/abs/2603.22378 · **PDF:** https://arxiv.org/pdf/2603.22378v1
+
+## 2657. Extending Precipitation Nowcasting Horizons via Spectral Fusion of Radar Observations and Foundation Model Priors
+
+- **ID:** `2603.21768` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Yuze Qin, Qingyong Li, Zhiqing Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.21768 · **PDF:** https://arxiv.org/pdf/2603.21768v2
+
+## 2658. Cycle Inverse-Consistent TransMorph: A Balanced Deep Learning Framework for Brain MRI Registration
+
+- **ID:** `2603.21760` · **primary:** `eess.IV` · **Submitted:** 2026-03-23
+- **Authors:** Jiaqi Shang, Haojin Wu, Yinyi Lai, et al.
+- **abs:** https://arxiv.org/abs/2603.21760 · **PDF:** https://arxiv.org/pdf/2603.21760v1
+
+## 2659. Let's Think with Images Efficiently! An Interleaved-Modal Chain-of-Thought Reasoning Framework with Dynamic and Precise Visual Thoughts
+
+- **ID:** `2603.21754` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Xu Liu, Yongheng Zhang, Qiguang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.21754 · **PDF:** https://arxiv.org/pdf/2603.21754v1
+
+## 2660. Identifiability and amortized inference limitations in Kuramoto models
+
+- **ID:** `2603.21752` · **primary:** `stat.AP` · **Submitted:** 2026-03-23
+- **Authors:** Emma Hannula, Jana de Wiljes, Matthew T. Moores, et al.
+- **abs:** https://arxiv.org/abs/2603.21752 · **PDF:** https://arxiv.org/pdf/2603.21752v1
+
+## 2661. From Diffusion To Flow: Efficient Motion Generation In MotionGPT3
+
+- **ID:** `2603.26747` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Jaymin Ban, JiHong Jeon, SangYeop Jeong
+- **abs:** https://arxiv.org/abs/2603.26747 · **PDF:** https://arxiv.org/pdf/2603.26747v1
+
+## 2662. Model selection in hybrid quantum neural networks with applications to quantum transformer architectures
+
+- **ID:** `2603.21749` · **primary:** `quant-ph` · **Submitted:** 2026-03-23
+- **Authors:** Harsh Wadhwa, Rahul Bhowmick, Naipunnya Raj, et al.
+- **abs:** https://arxiv.org/abs/2603.21749 · **PDF:** https://arxiv.org/pdf/2603.21749v1
+
+## 2663. The Presupposition Problem in Representation Genesis
+
+- **ID:** `2603.21745` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Yiling Wu
+- **abs:** https://arxiv.org/abs/2603.21745 · **PDF:** https://arxiv.org/pdf/2603.21745v2
+
+## 2664. CellFluxRL: Biologically-Constrained Virtual Cell Modeling via Reinforcement Learning
+
+- **ID:** `2603.21743` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Dongxia Wu, Shiye Su, Yuhui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.21743 · **PDF:** https://arxiv.org/pdf/2603.21743v2
+
+## 2665. AI Co-Scientist for Ranking: Discovering Novel Search Ranking Models alongside LLM-based AI Agents with Cloud Computing Access
+
+- **ID:** `2603.22376` · **primary:** `cs.IR` · **Submitted:** 2026-03-23
+- **Authors:** Liwei Wu, Cho-Jui Hsieh
+- **abs:** https://arxiv.org/abs/2603.22376 · **PDF:** https://arxiv.org/pdf/2603.22376v1
+
+## 2666. The Reasoning Error About Reasoning: Why Different Types of Reasoning Require Different Representational Structures
+
+- **ID:** `2603.21736` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Yiling Wu
+- **abs:** https://arxiv.org/abs/2603.21736 · **PDF:** https://arxiv.org/pdf/2603.21736v2
+
+## 2667. Cognitive Agency Surrender: Defending Epistemic Sovereignty via Scaffolded AI Friction
+
+- **ID:** `2603.21735` · **primary:** `cs.HC` · **Submitted:** 2026-03-23
+- **Authors:** Kuangzhe Xu, Yu Shen, Longjie Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.21735 · **PDF:** https://arxiv.org/pdf/2603.21735v1
+
+## 2668. EvoIdeator: Evolving Scientific Ideas through Checklist-Grounded Reinforcement Learning
+
+- **ID:** `2603.21728` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Andreas Sauter, Yuyue Zhao, Jacopo Urbani, et al.
+- **abs:** https://arxiv.org/abs/2603.21728 · **PDF:** https://arxiv.org/pdf/2603.21728v1
+
+## 2669. Three Creates All: You Only Sample 3 Steps
+
+- **ID:** `2603.22375` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Yuren Cai, Guangyi Wang, Zongqing Li, et al.
+- **abs:** https://arxiv.org/abs/2603.22375 · **PDF:** https://arxiv.org/pdf/2603.22375v1
+
+## 2670. CurvZO: Adaptive Curvature-Guided Sparse Zeroth-Order Optimization for Efficient LLM Fine-Tuning
+
+- **ID:** `2603.21725` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Shuo Wang, Ziyu Chen, Ming Tang
+- **abs:** https://arxiv.org/abs/2603.21725 · **PDF:** https://arxiv.org/pdf/2603.21725v1
+
+## 2671. FISformer: Replacing Self-Attention with a Fuzzy Inference System in Transformer Models for Time Series Forecasting
+
+- **ID:** `2603.21724` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Bulent Haznedar, Levent Karacan
+- **abs:** https://arxiv.org/abs/2603.21724 · **PDF:** https://arxiv.org/pdf/2603.21724v1
+
+## 2672. SemEval-2026 Task 12: Abductive Event Reasoning: Towards Real-World Event Causal Inference for Large Language Models
+
+- **ID:** `2603.21720` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Pengfei Cao, Mingxuan Yang, Yubo Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.21720 · **PDF:** https://arxiv.org/pdf/2603.21720v1
+
+## 2673. Probing How Scalable Table Data Enhances General Long-Context Reasoning
+
+- **ID:** `2603.21719` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Huaibing Xie, Guoliang Zhao, Yang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.21719 · **PDF:** https://arxiv.org/pdf/2603.21719v1
+
+## 2674. Uncertainty Quantification for Distribution-to-Distribution Flow Matching in Scientific Imaging
+
+- **ID:** `2603.21717` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Dongxia Wu, Yuhui Zhang, Serena Yeung-Levy, et al.
+- **abs:** https://arxiv.org/abs/2603.21717 · **PDF:** https://arxiv.org/pdf/2603.21717v2
+
+## 2675. When Exploration Comes for Free with Mixture-Greedy: Do we need UCB in Diversity-Aware Multi-Armed Bandits?
+
+- **ID:** `2603.21716` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Bahar Dibaei Nia, Farzan Farnia
+- **abs:** https://arxiv.org/abs/2603.21716 · **PDF:** https://arxiv.org/pdf/2603.21716v1
+
+## 2676. Compensating Visual Insufficiency with Stratified Language Guidance for Long-Tail Class Incremental Learning
+
+- **ID:** `2603.21708` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Xi Wang, Xu Yang, Donghao Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.21708 · **PDF:** https://arxiv.org/pdf/2603.21708v1
+
+## 2677. TDEC: Deep Embedded Image Clustering with Transformer and Distribution Information
+
+- **ID:** `2603.26746` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Ruilin Zhang, Haiyang Zheng, Hongpeng Wang
+- **abs:** https://arxiv.org/abs/2603.26746 · **PDF:** https://arxiv.org/pdf/2603.26746v1
+
+## 2678. Data-Free Layer-Adaptive Merging via Fisher Information for Long-to-Short Reasoning LLMs
+
+- **ID:** `2603.21705` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Tian Xia
+- **abs:** https://arxiv.org/abs/2603.21705 · **PDF:** https://arxiv.org/pdf/2603.21705v1
+
+## 2679. Rethinking Token Reduction for Large Vision-Language Models
+
+- **ID:** `2603.21701` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Yi Wang, Haofei Zhang, Qihan Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.21701 · **PDF:** https://arxiv.org/pdf/2603.21701v1
+
+## 2680. A Job I Like or a Job I Can Get: Designing Job Recommender Systems Using Field Experiments
+
+- **ID:** `2603.21699` · **primary:** `econ.EM` · **Submitted:** 2026-03-23
+- **Authors:** Guillaume Bied, Philippe Caillou, Bruno Crépon, et al.
+- **abs:** https://arxiv.org/abs/2603.21699 · **PDF:** https://arxiv.org/pdf/2603.21699v1
+
+## 2681. A Blueprint for Self-Evolving Coding Agents in Vehicle Aerodynamic Drag Prediction
+
+- **ID:** `2603.21698` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Jinhui Ren, Huaiming Li, Yabin Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.21698 · **PDF:** https://arxiv.org/pdf/2603.21698v1
+
+## 2682. Structured Visual Narratives Undermine Safety Alignment in Multimodal Large Language Models
+
+- **ID:** `2603.21697` · **primary:** `cs.CR` · **Submitted:** 2026-03-23
+- **Authors:** Rui Yang Tan, Yujia Hu, Roy Ka-Wei Lee
+- **abs:** https://arxiv.org/abs/2603.21697 · **PDF:** https://arxiv.org/pdf/2603.21697v1
+
+## 2683. MIND: Multi-agent inference for negotiation dialogue in travel planning
+
+- **ID:** `2603.21696` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Hunmin Do, Taejun Yoon, Kiyong Jung
+- **abs:** https://arxiv.org/abs/2603.21696 · **PDF:** https://arxiv.org/pdf/2603.21696v1
+
+## 2684. Deterministic Hallucination Detection in Medical VQA via Confidence-Evidence Bayesian Gain
+
+- **ID:** `2603.21693` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Mohammad Asadi, Tahoura Nedaee, Jack W. O'Sullivan, et al.
+- **abs:** https://arxiv.org/abs/2603.21693 · **PDF:** https://arxiv.org/pdf/2603.21693v1
+
+## 2685. Reasoning Provenance for Autonomous AI Agents: Structured Behavioral Analytics Beyond State Checkpoints and Execution Traces
+
+- **ID:** `2603.21692` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Neelmani Vispute
+- **abs:** https://arxiv.org/abs/2603.21692 · **PDF:** https://arxiv.org/pdf/2603.21692v1
+
+## 2686. AI Token Futures Market: Commoditization of Compute and Derivatives Contract Design
+
+- **ID:** `2603.21690` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Yicai Xing
+- **abs:** https://arxiv.org/abs/2603.21690 · **PDF:** https://arxiv.org/pdf/2603.21690v1
+
+## 2687. MIRAGE: The Illusion of Visual Understanding
+
+- **ID:** `2603.21687` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Mohammad Asadi, Jack W. O'Sullivan, Fang Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.21687 · **PDF:** https://arxiv.org/pdf/2603.21687v3
+
+## 2688. LipsAM: Lipschitz-Continuous Amplitude Modifier for Audio Signal Processing and its Application to Plug-and-Play Dereverberation
+
+- **ID:** `2603.21684` · **primary:** `cs.SD` · **Submitted:** 2026-03-23
+- **Authors:** Kazuki Matsumoto, Ren Uchida, Kohei Yatabe
+- **abs:** https://arxiv.org/abs/2603.21684 · **PDF:** https://arxiv.org/pdf/2603.21684v1
+
+## 2689. Learning operators on labelled conditional distributions with applications to mean field control of non exchangeable systems
+
+- **ID:** `2603.21683` · **primary:** `math.OC` · **Submitted:** 2026-03-23
+- **Authors:** Samy Mekkaoui, Huyên Pham, Xavier Warin
+- **abs:** https://arxiv.org/abs/2603.21683 · **PDF:** https://arxiv.org/pdf/2603.21683v1
+
+## 2690. CoNBONet: Conformalized Neuroscience-inspired Bayesian Operator Network for Reliability Analysis
+
+- **ID:** `2603.21678` · **primary:** `stat.ML` · **Submitted:** 2026-03-23
+- **Authors:** Shailesh Garg, Souvik Chakraborty
+- **abs:** https://arxiv.org/abs/2603.21678 · **PDF:** https://arxiv.org/pdf/2603.21678v1
+
+## 2691. Thinking Deeper, Not Longer: Depth-Recurrent Transformers for Compositional Generalization
+
+- **ID:** `2603.21676` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Hung-Hsuan Chen
+- **abs:** https://arxiv.org/abs/2603.21676 · **PDF:** https://arxiv.org/pdf/2603.21676v1
+
+## 2692. SPINONet: Scalable Spiking Physics-informed Neural Operator for Computational Mechanics Applications
+
+- **ID:** `2603.21674` · **primary:** `physics.comp-ph` · **Submitted:** 2026-03-23
+- **Authors:** Shailesh Garg, Luis Mandl, Somdatta Goswami, et al.
+- **abs:** https://arxiv.org/abs/2603.21674 · **PDF:** https://arxiv.org/pdf/2603.21674v1
+
+## 2693. Optimizing Multi-Agent Weather Captioning via Text Gradient Descent: A Training-Free Approach with Consensus-Aware Gradient Fusion
+
+- **ID:** `2603.21673` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Shixu Liu
+- **abs:** https://arxiv.org/abs/2603.21673 · **PDF:** https://arxiv.org/pdf/2603.21673v1
+
+## 2694. Rethinking Multimodal Fusion for Time Series: Auxiliary Modalities Need Constrained Fusion
+
+- **ID:** `2603.22372` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Seunghan Lee, Jun Seo, Jaehoon Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.22372 · **PDF:** https://arxiv.org/pdf/2603.22372v1
+
+## 2695. Optimal Memory Encoding Through Fluctuation-Response Structure
+
+- **ID:** `2603.21666` · **primary:** `cs.NE` · **Submitted:** 2026-03-23
+- **Authors:** Lianxiang Cui, Kohei Nakajima, Kazuyuki Aihara
+- **abs:** https://arxiv.org/abs/2603.21666 · **PDF:** https://arxiv.org/pdf/2603.21666v1
+
+## 2696. TAMTRL: Teacher-Aligned Reward Reshaping for Multi-Turn Reinforcement Learning in Long-Context Compression
+
+- **ID:** `2603.21663` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Li Wang, Yandong Wang, Xin Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.21663 · **PDF:** https://arxiv.org/pdf/2603.21663v1
+
+## 2697. Cross-Scenario Deraining Adaptation with Unpaired Data: Superpixel Structural Priors and Multi-Stage Pseudo-Rain Synthesis
+
+- **ID:** `2603.21661` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Kangbo Zhao, Miaoxin Guan, Xiang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.21661 · **PDF:** https://arxiv.org/pdf/2603.21661v1
+
+## 2698. A Comparative Analysis of LLM Memorization at Statistical and Internal Levels: Cross-Model Commonalities and Model-Specific Signatures
+
+- **ID:** `2603.21658` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Bowen Chen, Namgi Han, Yusuke Miyao
+- **abs:** https://arxiv.org/abs/2603.21658 · **PDF:** https://arxiv.org/pdf/2603.21658v1
+
+## 2699. TrustFed: Enabling Trustworthy Medical AI under Data Privacy Constraints
+
+- **ID:** `2603.21656` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Vagish Kumar, Syed Bahauddin Alam, Souvik Chakraborty
+- **abs:** https://arxiv.org/abs/2603.21656 · **PDF:** https://arxiv.org/pdf/2603.21656v1
+
+## 2700. Towards Secure Retrieval-Augmented Generation: A Comprehensive Review of Threats, Defenses and Benchmarks
+
+- **ID:** `2603.21654` · **primary:** `cs.CR` · **Submitted:** 2026-03-23
+- **Authors:** Yanming Mu, Hao Hu, Feiyang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.21654 · **PDF:** https://arxiv.org/pdf/2603.21654v1
+
+## 2701. MISApp: Multi-Hop Intent-Aware Session Graph Learning for Next App Prediction
+
+- **ID:** `2603.21653` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Yunchi Yang, Longlong Li, Jianliang Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.21653 · **PDF:** https://arxiv.org/pdf/2603.21653v1
+
+## 2702. FedCVU: Federated Learning for Cross-View Video Understanding
+
+- **ID:** `2603.21647` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Shenghan Zhang, Run Ling, Ke Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.21647 · **PDF:** https://arxiv.org/pdf/2603.21647v1
+
+## 2703. Engineering Distributed Governance for Regional Prosperity: A Socio-Technical Framework for Mitigating Under-Vibrancy via Human Data Engines
+
+- **ID:** `2603.21639` · **primary:** `cs.CY` · **Submitted:** 2026-03-23
+- **Authors:** Amil Khanzada, Takuji Takemoto
+- **abs:** https://arxiv.org/abs/2603.21639 · **PDF:** https://arxiv.org/pdf/2603.21639v1
+
+## 2704. Steering Sparse Autoencoder Latents to Control Dynamic Head Pruning in Vision Transformers (Student Abstract)
+
+- **ID:** `2603.26743` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Yousung Lee, Dongsoo Har
+- **abs:** https://arxiv.org/abs/2603.26743 · **PDF:** https://arxiv.org/pdf/2603.26743v1
+
+## 2705. Silicon Bureaucracy and AI Test-Oriented Education: Contamination Sensitivity and Score Confidence in LLM Benchmarks
+
+- **ID:** `2603.21636` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Yiliang Song, Hongjun An, Jiangan Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.21636 · **PDF:** https://arxiv.org/pdf/2603.21636v2
+
+## 2706. EnterpriseLab: A Full-Stack Platform for developing and deploying agents in Enterprises
+
+- **ID:** `2603.21630` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Ankush Agarwal, Harsh Vishwakarma, Suraj Nagaje, et al.
+- **abs:** https://arxiv.org/abs/2603.21630 · **PDF:** https://arxiv.org/pdf/2603.21630v1
+
+## 2707. Neyman-Pearson multiclass classification under label noise via empirical likelihood
+
+- **ID:** `2603.21623` · **primary:** `stat.ME` · **Submitted:** 2026-03-23
+- **Authors:** Qiong Zhang, Qinglong Tian, Pengfei Li
+- **abs:** https://arxiv.org/abs/2603.21623 · **PDF:** https://arxiv.org/pdf/2603.21623v1
+
+## 2708. Proximal Policy Optimization in Path Space: A Schrödinger Bridge Perspective
+
+- **ID:** `2603.21621` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Yuehu Gong, Zeyuan Wang, Yulin Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.21621 · **PDF:** https://arxiv.org/pdf/2603.21621v1
+
+## 2709. Efficient Zero-Shot AI-Generated Image Detection
+
+- **ID:** `2603.21619` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Ryosuke Sonoda, Ramya Srinivasan
+- **abs:** https://arxiv.org/abs/2603.21619 · **PDF:** https://arxiv.org/pdf/2603.21619v1
+
+## 2710. FAAR: Format-Aware Adaptive Rounding for NVFP4
+
+- **ID:** `2603.22370` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Hanglin Li, Shuchang Tian, Chen Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.22370 · **PDF:** https://arxiv.org/pdf/2603.22370v1
+
+## 2711. Rateless DeepJSCC for Broadcast Channels: a Rate-Distortion-Complexity Tradeoff
+
+- **ID:** `2603.21616` · **primary:** `cs.IT` · **Submitted:** 2026-03-23
+- **Authors:** Zijun Qin, Jingxuan Huang, Zesong Fei, et al.
+- **abs:** https://arxiv.org/abs/2603.21616 · **PDF:** https://arxiv.org/pdf/2603.21616v1
+
+## 2712. AgenticRec: End-to-End Tool-Integrated Policy Optimization for Ranking-Oriented Recommender Agents
+
+- **ID:** `2603.21613` · **primary:** `cs.IR` · **Submitted:** 2026-03-23
+- **Authors:** Tianyi Li, Zixuan Wang, Guidong Lei, et al.
+- **abs:** https://arxiv.org/abs/2603.21613 · **PDF:** https://arxiv.org/pdf/2603.21613v1
+
+## 2713. Towards Multimodal Time Series Anomaly Detection with Semantic Alignment and Condensed Interaction
+
+- **ID:** `2603.21612` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Shiyan Hu, Jianxin Jin, Yang Shu, et al.
+- **abs:** https://arxiv.org/abs/2603.21612 · **PDF:** https://arxiv.org/pdf/2603.21612v1
+
+## 2714. Rule-State Inference (RSI): A Bayesian Framework for Compliance Monitoring in Rule-Governed Domains
+
+- **ID:** `2603.21610` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Abdou-Raouf Atarmla
+- **abs:** https://arxiv.org/abs/2603.21610 · **PDF:** https://arxiv.org/pdf/2603.21610v1
+
+## 2715. DiT-Flow: Speech Enhancement Robust to Multiple Distortions based on Flow Matching in Latent Space and Diffusion Transformers
+
+- **ID:** `2603.21608` · **primary:** `eess.AS` · **Submitted:** 2026-03-23
+- **Authors:** Tianyu Cao, Helin Wang, Ari Frummer, et al.
+- **abs:** https://arxiv.org/abs/2603.21608 · **PDF:** https://arxiv.org/pdf/2603.21608v1
+
+## 2716. INTRYGUE: Induction-Aware Entropy Gating for Reliable RAG Uncertainty Estimation
+
+- **ID:** `2603.21607` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Alexandra Bazarova, Andrei Volodichev, Daria Kotova, et al.
+- **abs:** https://arxiv.org/abs/2603.21607 · **PDF:** https://arxiv.org/pdf/2603.21607v1
+
+## 2717. mSFT: Addressing Dataset Mixtures Overfitting Heterogeneously in Multi-task SFT
+
+- **ID:** `2603.21606` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Woosung Koh, Jeyoung Jeon, Youngjin Song, et al.
+- **abs:** https://arxiv.org/abs/2603.21606 · **PDF:** https://arxiv.org/pdf/2603.21606v5
+
+## 2718. Do Multilingual VLMs Reason Equally? A Cross-Lingual Visual Reasoning Audit for Indian Languages
+
+- **ID:** `2603.26742` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Swastik R
+- **abs:** https://arxiv.org/abs/2603.26742 · **PDF:** https://arxiv.org/pdf/2603.26742v1
+
+## 2719. SynLeaF: A Dual-Stage Multimodal Fusion Framework for Synthetic Lethality Prediction Across Pan- and Single-Cancer Contexts
+
+- **ID:** `2603.22369` · **primary:** `q-bio.GN` · **Submitted:** 2026-03-23
+- **Authors:** Zheming Xing, Siyuan Zhou, Ruinan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.22369 · **PDF:** https://arxiv.org/pdf/2603.22369v1
+
+## 2720. Riemannian Geometry Speaks Louder Than Words: From Graph Foundation Model to Next-Generation Graph Intelligence
+
+- **ID:** `2603.21601` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Philip S. Yu, Li Sun
+- **abs:** https://arxiv.org/abs/2603.21601 · **PDF:** https://arxiv.org/pdf/2603.21601v1
+
+## 2721. Cerebra: A Multidisciplinary AI Board for Multimodal Dementia Characterization and Risk Assessment
+
+- **ID:** `2603.21597` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Sheng Liu, Long Chen, Zeyun Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.21597 · **PDF:** https://arxiv.org/pdf/2603.21597v2
+
+## 2722. In-network Attack Detection with Federated Deep Learning in IoT Networks: Real Implementation and Analysis
+
+- **ID:** `2603.21596` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Devashish Chaudhary, Sutharshan Rajasegarar, Shiva Raj Pokhrel, et al.
+- **abs:** https://arxiv.org/abs/2603.21596 · **PDF:** https://arxiv.org/pdf/2603.21596v1
+
+## 2723. Spatio-Temporal Attention Enhanced Multi-Agent DRL for UAV-Assisted Wireless Networks with Limited Communications
+
+- **ID:** `2603.21594` · **primary:** `cs.IT` · **Submitted:** 2026-03-23
+- **Authors:** Che Chen, Lanhua Li, Shimin Gong, et al.
+- **abs:** https://arxiv.org/abs/2603.21594 · **PDF:** https://arxiv.org/pdf/2603.21594v1
+
+## 2724. Feature Incremental Clustering with Generalization Bounds
+
+- **ID:** `2603.21590` · **primary:** `math.ST` · **Submitted:** 2026-03-23
+- **Authors:** Jing Zhang, Chenping Hou
+- **abs:** https://arxiv.org/abs/2603.21590 · **PDF:** https://arxiv.org/pdf/2603.21590v1
+
+## 2725. When Visuals Aren't the Problem: Evaluating Vision-Language Models on Misleading Data Visualizations
+
+- **ID:** `2603.22368` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Harsh Nishant Lalai, Raj Sanjay Shah, Hanspeter Pfister, et al.
+- **abs:** https://arxiv.org/abs/2603.22368 · **PDF:** https://arxiv.org/pdf/2603.22368v1
+
+## 2726. SSAM: Singular Subspace Alignment for Merging Multimodal Large Language Models
+
+- **ID:** `2603.21584` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Md Kaykobad Reza, Ameya Patil, Edward Ayrapetian, et al.
+- **abs:** https://arxiv.org/abs/2603.21584 · **PDF:** https://arxiv.org/pdf/2603.21584v1
+
+## 2727. Reasoner-Executor-Synthesizer: Scalable Agentic Architecture with Static O(1) Context Window
+
+- **ID:** `2603.22367` · **primary:** `cs.IR` · **Submitted:** 2026-03-23
+- **Authors:** Ivan Dobrovolskyi
+- **abs:** https://arxiv.org/abs/2603.22367 · **PDF:** https://arxiv.org/pdf/2603.22367v1
+
+## 2728. Modeling Quantum Federated Autoencoder for Anomaly Detection in IoT Networks
+
+- **ID:** `2603.22366` · **primary:** `quant-ph` · **Submitted:** 2026-03-23
+- **Authors:** Devashish Chaudhary, Sutharshan Rajasegarar, Shiva Raj Pokhrel
+- **abs:** https://arxiv.org/abs/2603.22366 · **PDF:** https://arxiv.org/pdf/2603.22366v1
+
+## 2729. Q-AGNN: Quantum-Enhanced Attentive Graph Neural Network for Intrusion Detection
+
+- **ID:** `2603.22365` · **primary:** `cs.CR` · **Submitted:** 2026-03-23
+- **Authors:** Devashish Chaudhary, Sutharshan Rajasegarar, Shiva Raj Pokhrel
+- **abs:** https://arxiv.org/abs/2603.22365 · **PDF:** https://arxiv.org/pdf/2603.22365v1
+
+## 2730. MCLR: Improving Conditional Modeling in Visual Generative Models via Inter-Class Likelihood-Ratio Maximization and Establishing the Equivalence between Classifier-Free Guidance and Alignment Objectives
+
+- **ID:** `2603.22364` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Xiang Li, Yixuan Jia, Xiao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.22364 · **PDF:** https://arxiv.org/pdf/2603.22364v1
+
+## 2731. Mind over Space: Can Multimodal Large Language Models Mentally Navigate?
+
+- **ID:** `2603.21577` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Qihui Zhu, Shouwei Ruan, Xiao Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.21577 · **PDF:** https://arxiv.org/pdf/2603.21577v1
+
+## 2732. PRISM: Breaking the O(n) Memory Wall in Long-Context LLM Inference via O(1) Photonic Block Selection
+
+- **ID:** `2603.21576` · **primary:** `physics.optics` · **Submitted:** 2026-03-23
+- **Authors:** Hyoseok Park, Yeonsang Park
+- **abs:** https://arxiv.org/abs/2603.21576 · **PDF:** https://arxiv.org/pdf/2603.21576v2
+
+## 2733. Adaptive Robust Estimator for Multi-Agent Reinforcement Learning
+
+- **ID:** `2603.21574` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Zhongyi Li, Wan Tian, Jingyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.21574 · **PDF:** https://arxiv.org/pdf/2603.21574v1
+
+## 2734. DATASHI: A Parallel English-Tashlhiyt Corpus for Orthography Normalization and Low-Resource Language Processing
+
+- **ID:** `2603.21571` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Nasser-Eddine Monir, Zakaria Baou
+- **abs:** https://arxiv.org/abs/2603.21571 · **PDF:** https://arxiv.org/pdf/2603.21571v1
+
+## 2735. Stability and Bifurcation Analysis of Nonlinear PDEs via Random Projection-based PINNs: A Krylov-Arnoldi Approach
+
+- **ID:** `2603.21568` · **primary:** `math.NA` · **Submitted:** 2026-03-23
+- **Authors:** Gianluca Fabiani, Michail E. Kavousanakis, Constantinos Siettos, et al.
+- **abs:** https://arxiv.org/abs/2603.21568 · **PDF:** https://arxiv.org/pdf/2603.21568v1
+
+## 2736. Kolmogorov Complexity Bounds for LLM Steganography and a Perplexity-Based Detection Proxy
+
+- **ID:** `2603.21567` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Andrii Shportko
+- **abs:** https://arxiv.org/abs/2603.21567 · **PDF:** https://arxiv.org/pdf/2603.21567v1
+
+## 2737. CataractSAM-2: A Domain-Adapted Model for Anterior Segment Surgery Segmentation and Scalable Ground-Truth Annotation
+
+- **ID:** `2603.21566` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Mohammad Eslami, Dhanvinkumar Ganeshkumar, Saber Kazeminasab, et al.
+- **abs:** https://arxiv.org/abs/2603.21566 · **PDF:** https://arxiv.org/pdf/2603.21566v1
+
+## 2738. Rethinking SAR ATR: A Target-Aware Frequency-Spatial Enhancement Framework with Noise-Resilient Knowledge Guidance
+
+- **ID:** `2603.21565` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Yansong Lin, Zihan Cheng, Jielei Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.21565 · **PDF:** https://arxiv.org/pdf/2603.21565v1
+
+## 2739. Toward a Theory of Hierarchical Memory for Language Agents
+
+- **ID:** `2603.21564` · **primary:** `cs.IR` · **Submitted:** 2026-03-23
+- **Authors:** Yashar Talebirad, Ali Parsaee, Csongor Y. Szepesvari, et al.
+- **abs:** https://arxiv.org/abs/2603.21564 · **PDF:** https://arxiv.org/pdf/2603.21564v1
+
+## 2740. Counterfactual Credit Policy Optimization for Multi-Agent Collaboration
+
+- **ID:** `2603.21563` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Zhongyi Li, Wan Tian, Yikun Ban, et al.
+- **abs:** https://arxiv.org/abs/2603.21563 · **PDF:** https://arxiv.org/pdf/2603.21563v1
+
+## 2741. Stabilizing Iterative Self-Training with Verified Reasoning via Symbolic Recursive Self-Alignment
+
+- **ID:** `2603.21558` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Xinyu Zhang
+- **abs:** https://arxiv.org/abs/2603.21558 · **PDF:** https://arxiv.org/pdf/2603.21558v1
+
+## 2742. Language-Conditioned World Modeling for Visual Navigation
+
+- **ID:** `2603.26741` · **primary:** `cs.CV` · **Submitted:** 2026-03-23
+- **Authors:** Yifei Dong, Fengyi Wu, Yilong Dai, et al.
+- **abs:** https://arxiv.org/abs/2603.26741 · **PDF:** https://arxiv.org/pdf/2603.26741v1
+
+## 2743. Early Discoveries of Algorithmist I: Promise of Provable Algorithm Synthesis at Scale
+
+- **ID:** `2603.22363` · **primary:** `cs.SE` · **Submitted:** 2026-03-23
+- **Authors:** Janardhan Kulkarni
+- **abs:** https://arxiv.org/abs/2603.22363 · **PDF:** https://arxiv.org/pdf/2603.22363v1
+
+## 2744. What Do World Models Learn in RL? Probing Latent Representations in Learned Environment Simulators
+
+- **ID:** `2603.21546` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Xinyu Zhang
+- **abs:** https://arxiv.org/abs/2603.21546 · **PDF:** https://arxiv.org/pdf/2603.21546v1
+
+## 2745. Evolutionary Biparty Multiobjective UAV Path Planning: Problems and Empirical Comparisons
+
+- **ID:** `2603.21544` · **primary:** `cs.NE` · **Submitted:** 2026-03-23
+- **Authors:** Kesheng Chen, Wenjian Luo, Xin Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.21544 · **PDF:** https://arxiv.org/pdf/2603.21544v1
+
+## 2746. Sharper Generalization Bounds for Transformer
+
+- **ID:** `2603.21541` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Yawen Li, Tao Hu, Zhouhui Lian, et al.
+- **abs:** https://arxiv.org/abs/2603.21541 · **PDF:** https://arxiv.org/pdf/2603.21541v1
+
+## 2747. Generalization Limits of In-Context Operator Networks for Higher-Order Partial Differential Equations
+
+- **ID:** `2603.21534` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Jamie Mahowald, Tan Bui-Thanh
+- **abs:** https://arxiv.org/abs/2603.21534 · **PDF:** https://arxiv.org/pdf/2603.21534v1
+
+## 2748. LLM-Based Test Case Generation in DBMS through Monte Carlo Tree Search
+
+- **ID:** `2603.21530` · **primary:** `cs.SE` · **Submitted:** 2026-03-23
+- **Authors:** Yujia Chen, Yingli Zhou, Fangyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.21530 · **PDF:** https://arxiv.org/pdf/2603.21530v1
+
+## 2749. SynSym: A Synthetic Data Generation Framework for Psychiatric Symptom Identification
+
+- **ID:** `2603.21529` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Migyeong Kang, Jihyun Kim, Hyolim Jeon, et al.
+- **abs:** https://arxiv.org/abs/2603.21529 · **PDF:** https://arxiv.org/pdf/2603.21529v1
+
+## 2750. BOxCrete: A Bayesian Optimization Open-Source AI Model for Concrete Strength Forecasting and Mix Optimization
+
+- **ID:** `2603.21525` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Bayezid Baten, M. Ayyan Iqbal, Sebastian Ament, et al.
+- **abs:** https://arxiv.org/abs/2603.21525 · **PDF:** https://arxiv.org/pdf/2603.21525v1
+
+## 2751. CatRAG: Functor-Guided Structural Debiasing with Retrieval Augmentation for Fair LLMs
+
+- **ID:** `2603.21524` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Ravi Ranjan, Utkarsh Grover, Mayur Akewar, et al.
+- **abs:** https://arxiv.org/abs/2603.21524 · **PDF:** https://arxiv.org/pdf/2603.21524v1
+
+## 2752. SafePilot: A Framework for Assuring LLM-enabled Cyber-Physical Systems
+
+- **ID:** `2603.21523` · **primary:** `cs.RO` · **Submitted:** 2026-03-23
+- **Authors:** Weizhe Xu, Mengyu Liu, Fanxin Kong
+- **abs:** https://arxiv.org/abs/2603.21523 · **PDF:** https://arxiv.org/pdf/2603.21523v1
+
+## 2753. Efficient Failure Management for Multi-Agent Systems with Reasoning Trace Representation
+
+- **ID:** `2603.21522` · **primary:** `cs.SE` · **Submitted:** 2026-03-23
+- **Authors:** Lingzhe Zhang, Tong Jia, Mingyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.21522 · **PDF:** https://arxiv.org/pdf/2603.21522v1
+
+## 2754. Generalizable Self-Evolving Memory for Automatic Prompt Optimization
+
+- **ID:** `2603.21520` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Guanbao Liang, Yuanchen Bei, Sheng Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.21520 · **PDF:** https://arxiv.org/pdf/2603.21520v1
+
+## 2755. Triangulating Temporal Dynamics in Multilingual Swiss Online News
+
+- **ID:** `2603.21519` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Bros Victor, Dufraisse Evan, Popescu Adrian, et al.
+- **abs:** https://arxiv.org/abs/2603.21519 · **PDF:** https://arxiv.org/pdf/2603.21519v1
+
+## 2756. Unveiling the Mechanism of Continuous Representation Full-Waveform Inversion: A Wave Based Neural Tangent Kernel Framework
+
+- **ID:** `2603.22362` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Ruihua Chen, Yisi Luo, Bangyu Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.22362 · **PDF:** https://arxiv.org/pdf/2603.22362v1
+
+## 2757. Optimizing Feature Extraction for On-device Model Inference with User Behavior Sequences
+
+- **ID:** `2603.21508` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Chen Gong, Zhenzhe Zheng, Yiliu Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.21508 · **PDF:** https://arxiv.org/pdf/2603.21508v1
+
+## 2758. Quotient Geometry, Effective Curvature, and Implicit Bias in Simple Shallow Neural Networks
+
+- **ID:** `2603.21502` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Hang-Cheng Dong, Pengcheng Cheng
+- **abs:** https://arxiv.org/abs/2603.21502 · **PDF:** https://arxiv.org/pdf/2603.21502v1
+
+## 2759. A Framework for Closed-Loop Robotic Assembly, Alignment and Self-Recovery of Precision Optical Systems
+
+- **ID:** `2603.21496` · **primary:** `cs.RO` · **Submitted:** 2026-03-23
+- **Authors:** Seou Choi, Sachin Vaidya, Caio Silva, et al.
+- **abs:** https://arxiv.org/abs/2603.21496 · **PDF:** https://arxiv.org/pdf/2603.21496v1
+
+## 2760. RuntimeSlicer: Towards Generalizable Unified Runtime State Representation for Failure Management
+
+- **ID:** `2603.21495` · **primary:** `cs.SE` · **Submitted:** 2026-03-23
+- **Authors:** Lingzhe Zhang, Tong Jia, Weijie Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.21495 · **PDF:** https://arxiv.org/pdf/2603.21495v1
+
+## 2761. Agentic Automation of BT-RADS Scoring: End-to-End Multi-Agent System for Standardized Brain Tumor Follow-up Assessment
+
+- **ID:** `2603.21494` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Mohamed Sobhi Jabal, Jikai Zhang, Dominic LaBella, et al.
+- **abs:** https://arxiv.org/abs/2603.21494 · **PDF:** https://arxiv.org/pdf/2603.21494v2
+
+## 2762. Multinoulli Extension: A Lossless Continuous Relaxation for Partition-Constrained Subset Selection
+
+- **ID:** `2603.21492` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Qixin Zhang, Wei Huang, Yan Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.21492 · **PDF:** https://arxiv.org/pdf/2603.21492v1
+
+## 2763. Learning Can Converge Stably to the Wrong Belief under Latent Reliability
+
+- **ID:** `2603.21491` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Zhipeng Zhang, Zhenjie Yao, Kai Li, et al.
+- **abs:** https://arxiv.org/abs/2603.21491 · **PDF:** https://arxiv.org/pdf/2603.21491v1
+
+## 2764. Effective Strategies for Asynchronous Software Engineering Agents
+
+- **ID:** `2603.21489` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Jiayi Geng, Graham Neubig
+- **abs:** https://arxiv.org/abs/2603.21489 · **PDF:** https://arxiv.org/pdf/2603.21489v1
+
+## 2765. GaussianSSC: Triplane-Guided Directional Gaussian Fields for 3D Semantic Completion
+
+- **ID:** `2603.21487` · **primary:** `cs.RO` · **Submitted:** 2026-03-23
+- **Authors:** Ruiqi Xian, Jing Liang, He Yin, et al.
+- **abs:** https://arxiv.org/abs/2603.21487 · **PDF:** https://arxiv.org/pdf/2603.21487v1
+
+## 2766. Off-Policy Evaluation for Ranking Policies under Deterministic Logging Policies
+
+- **ID:** `2603.21485` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** Koichi Tanaka, Kazuki Kawamura, Takanori Muroi, et al.
+- **abs:** https://arxiv.org/abs/2603.21485 · **PDF:** https://arxiv.org/pdf/2603.21485v1
+
+## 2767. TaigiSpeech: A Low-Resource Real-World Speech Intent Dataset and Preliminary Results with Scalable Data Mining In-the-Wild
+
+- **ID:** `2603.21478` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Kai-Wei Chang, Yi-Cheng Lin, Huang-Cheng Chou, et al.
+- **abs:** https://arxiv.org/abs/2603.21478 · **PDF:** https://arxiv.org/pdf/2603.21478v1
+
+## 2768. Unified-MAS: Universally Generating Domain-Specific Nodes for Empowering Automatic Multi-Agent Systems
+
+- **ID:** `2603.21475` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Hehai Lin, Yu Yan, Zixuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.21475 · **PDF:** https://arxiv.org/pdf/2603.21475v1
+
+## 2769. Beyond Correlation: Refutation-Validated Aspect-Based Sentiment Analysis for Explainable Energy Market Returns
+
+- **ID:** `2603.21473` · **primary:** `cs.AI` · **Submitted:** 2026-03-23
+- **Authors:** Wihan van der Heever, Keane Ong, Ranjan Satapathy, et al.
+- **abs:** https://arxiv.org/abs/2603.21473 · **PDF:** https://arxiv.org/pdf/2603.21473v1
+
+## 2770. DRTriton: Large-Scale Synthetic Data Reinforcement Learning for Triton Kernel Generation
+
+- **ID:** `2603.21465` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Siqi Guo, Ming Lin, Tianbao Yang
+- **abs:** https://arxiv.org/abs/2603.21465 · **PDF:** https://arxiv.org/pdf/2603.21465v1
+
+## 2771. DSPA: Dynamic SAE Steering for Data-Efficient Preference Alignment
+
+- **ID:** `2603.21461` · **primary:** `cs.LG` · **Submitted:** 2026-03-23
+- **Authors:** James Wedgwood, Aashiq Muhamed, Mona T. Diab, et al.
+- **abs:** https://arxiv.org/abs/2603.21461 · **PDF:** https://arxiv.org/pdf/2603.21461v1
+
+## 2772. When Documents Disagree: Measuring Institutional Variation in Transplant Guidance with Retrieval-Augmented Language Models
+
+- **ID:** `2603.21460` · **primary:** `cs.IR` · **Submitted:** 2026-03-23
+- **Authors:** Yubo Li, Ramayya Krishnan, Rema Padman
+- **abs:** https://arxiv.org/abs/2603.21460 · **PDF:** https://arxiv.org/pdf/2603.21460v1
+
+## 2773. Compressive single-pixel imaging via a wavelength-multiplexed spatially incoherent diffractive optical processor
+
+- **ID:** `2603.21456` · **primary:** `physics.optics` · **Submitted:** 2026-03-23
+- **Authors:** Xiao Wang, Yiyang Wu, Yuntian Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.21456 · **PDF:** https://arxiv.org/pdf/2603.21456v1
+
+## 2774. Cross-Context Verification: Hierarchical Detection of Benchmark Contamination through Session-Isolated Analysis
+
+- **ID:** `2603.21454` · **primary:** `cs.CL` · **Submitted:** 2026-03-23
+- **Authors:** Tae-Eun Song
+- **abs:** https://arxiv.org/abs/2603.21454 · **PDF:** https://arxiv.org/pdf/2603.21454v2
+
+## 2775. Safety as Computation: Certified Answer Reuse via Capability Closure in Task-Oriented Dialogue
+
+- **ID:** `2603.21448` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Cosimo Spera
+- **abs:** https://arxiv.org/abs/2603.21448 · **PDF:** https://arxiv.org/pdf/2603.21448v1
+
+## 2776. KG-Hopper: Empowering Compact Open LLMs with Knowledge Graph Reasoning via Reinforcement Learning
+
+- **ID:** `2603.21440` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Shuai Wang, Yinan Yu
+- **abs:** https://arxiv.org/abs/2603.21440 · **PDF:** https://arxiv.org/pdf/2603.21440v3
+
+## 2777. LLM-Powered Workflow Optimization for Multidisciplinary Software Development: An Automotive Industry Case Study
+
+- **ID:** `2603.21439` · **primary:** `cs.SE` · **Submitted:** 2026-03-22
+- **Authors:** Shuai Wang, Yinan Yu, Earl Barr, et al.
+- **abs:** https://arxiv.org/abs/2603.21439 · **PDF:** https://arxiv.org/pdf/2603.21439v4
+
+## 2778. PROMPT2BOX: Uncovering Entailment Structure among LLM Prompts
+
+- **ID:** `2603.21438` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Neeladri Bhuiya, Shib Sankar Dasgupta, Andrew McCallum, et al.
+- **abs:** https://arxiv.org/abs/2603.21438 · **PDF:** https://arxiv.org/pdf/2603.21438v1
+
+## 2779. Semantic Shift: the Fundamental Challenge in Text Embedding and Retrieval
+
+- **ID:** `2603.21437` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Hang Gao, Dimitris N. Metaxas
+- **abs:** https://arxiv.org/abs/2603.21437 · **PDF:** https://arxiv.org/pdf/2603.21437v1
+
+## 2780. Behavioural feasible set: Value alignment constraints on AI decision support
+
+- **ID:** `2603.21435` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Taejin Park
+- **abs:** https://arxiv.org/abs/2603.21435 · **PDF:** https://arxiv.org/pdf/2603.21435v1
+
+## 2781. When and Where: A Model Hippocampal Network Unifies Formation of Time Cells and Place Cells
+
+- **ID:** `2604.00036` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-22
+- **Authors:** Qiaorong S. Yu, Zhaoze Wang, Vijay Balasubramanian
+- **abs:** https://arxiv.org/abs/2604.00036 · **PDF:** https://arxiv.org/pdf/2604.00036v1
+
+## 2782. DomAgent: Leveraging Knowledge Graphs and Case-Based Reasoning for Domain-Specific Code Generation
+
+- **ID:** `2603.21430` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Shuai Wang, Dhasarathy Parthasarathy, Robert Feldt, et al.
+- **abs:** https://arxiv.org/abs/2603.21430 · **PDF:** https://arxiv.org/pdf/2603.21430v2
+
+## 2783. HyReach: Vision-Guided Hybrid Manipulator Reaching in Unseen Cluttered Environments
+
+- **ID:** `2603.21421` · **primary:** `cs.RO` · **Submitted:** 2026-03-22
+- **Authors:** Shivani Kamtikar, Kendall Koe, Justin Wasserman, et al.
+- **abs:** https://arxiv.org/abs/2603.21421 · **PDF:** https://arxiv.org/pdf/2603.21421v1
+
+## 2784. Is the future of AI green? What can innovation diffusion models say about generative AI's environmental impact?
+
+- **ID:** `2603.21419` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Robert Viseur, Nicolas Jullien
+- **abs:** https://arxiv.org/abs/2603.21419 · **PDF:** https://arxiv.org/pdf/2603.21419v1
+
+## 2785. Efficient Fine-Tuning Methods for Portuguese Question Answering: A Comparative Study of PEFT on BERTimbau and Exploratory Evaluation of Generative LLMs
+
+- **ID:** `2603.21418` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Mariela M. Nina, Caio Veloso Costa, Lilian Berton, et al.
+- **abs:** https://arxiv.org/abs/2603.21418 · **PDF:** https://arxiv.org/pdf/2603.21418v1
+
+## 2786. Silent Commitment Failure in Instruction-Tuned Language Models: Evidence of Governability Divergence Across Architectures
+
+- **ID:** `2603.21415` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Gregory M. Ruddell
+- **abs:** https://arxiv.org/abs/2603.21415 · **PDF:** https://arxiv.org/pdf/2603.21415v1
+
+## 2787. Fingerprinting Deep Neural Networks for Ownership Protection: An Analytical Approach
+
+- **ID:** `2603.21411` · **primary:** `cs.CR` · **Submitted:** 2026-03-22
+- **Authors:** Guang Yang, Ziye Geng, Yihang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.21411 · **PDF:** https://arxiv.org/pdf/2603.21411v1
+
+## 2788. Multi-Perspective LLM Annotations for Valid Analyses in Subjective Tasks
+
+- **ID:** `2603.21404` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Navya Mehrotra, Adam Visokay, Kristina Gligorić
+- **abs:** https://arxiv.org/abs/2603.21404 · **PDF:** https://arxiv.org/pdf/2603.21404v1
+
+## 2789. The Myhill-Nerode Theorem for Bounded Interaction: Canonical Abstractions via Agent-Bounded Indistinguishability
+
+- **ID:** `2603.21399` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Anthony T. Nixon
+- **abs:** https://arxiv.org/abs/2603.21399 · **PDF:** https://arxiv.org/pdf/2603.21399v1
+
+## 2790. Persona Vectors in Games: Measuring and Steering Strategies via Activation Vectors
+
+- **ID:** `2603.21398` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Johnathan Sun, Andrew Zhang
+- **abs:** https://arxiv.org/abs/2603.21398 · **PDF:** https://arxiv.org/pdf/2603.21398v1
+
+## 2791. STEM Agent: A Self-Adapting, Tool-Enabled, Extensible Architecture for Multi-Protocol AI Agent Systems
+
+- **ID:** `2603.22359` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Alfred Shen, Aaron Shen
+- **abs:** https://arxiv.org/abs/2603.22359 · **PDF:** https://arxiv.org/pdf/2603.22359v1
+
+## 2792. Mechanisms of Introspective Awareness
+
+- **ID:** `2603.21396` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Uzay Macar, Li Yang, Atticus Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.21396 · **PDF:** https://arxiv.org/pdf/2603.21396v1
+
+## 2793. A Generalised Exponentiated Gradient Approach to Enhance Fairness in Binary and Multi-class Classification Tasks
+
+- **ID:** `2603.21393` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Maryam Boubekraoui, Giordano d'Aloisio, Antinisca Di Marco
+- **abs:** https://arxiv.org/abs/2603.21393 · **PDF:** https://arxiv.org/pdf/2603.21393v1
+
+## 2794. Task-Specific Efficiency Analysis: When Small Language Models Outperform Large Language Models
+
+- **ID:** `2603.21389` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Jinghan Cao, Yu Ma, Xinjin Li, et al.
+- **abs:** https://arxiv.org/abs/2603.21389 · **PDF:** https://arxiv.org/pdf/2603.21389v1
+
+## 2795. PivotRL: High Accuracy Agentic Post-Training at Low Compute Cost
+
+- **ID:** `2603.21383` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Junkeun Yi, Damon Mosk-Aoyama, Baihe Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.21383 · **PDF:** https://arxiv.org/pdf/2603.21383v1
+
+## 2796. Quantum Fuzzy Sets Revisited: Density Matrices, Decoherence, and the Q-Matrix Framework
+
+- **ID:** `2603.26739` · **primary:** `quant-ph` · **Submitted:** 2026-03-22
+- **Authors:** Mirco A. Mannucci
+- **abs:** https://arxiv.org/abs/2603.26739 · **PDF:** https://arxiv.org/pdf/2603.26739v1
+
+## 2797. An InSAR Phase Unwrapping Framework for Large-scale and Complex Events
+
+- **ID:** `2603.21378` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** Yijia Song, Juliet Biggs, Alin Achim, et al.
+- **abs:** https://arxiv.org/abs/2603.21378 · **PDF:** https://arxiv.org/pdf/2603.21378v1
+
+## 2798. HamVision: Hamiltonian Dynamics as Inductive Bias for Medical Image Analysis
+
+- **ID:** `2603.21377` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** Mohamed A Mabrok
+- **abs:** https://arxiv.org/abs/2603.21377 · **PDF:** https://arxiv.org/pdf/2603.21377v1
+
+## 2799. A transformer architecture alteration to incentivise externalised reasoning
+
+- **ID:** `2603.21376` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Elizabeth Pavlova, Mariia Koroliuk, Karthik Viswanathan, et al.
+- **abs:** https://arxiv.org/abs/2603.21376 · **PDF:** https://arxiv.org/pdf/2603.21376v2
+
+## 2800. Constrained Online Convex Optimization with Memory and Predictions
+
+- **ID:** `2603.21375` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Mohammed Abdullah, George Iosifidis, Salah Eddine Elayoubi, et al.
+- **abs:** https://arxiv.org/abs/2603.21375 · **PDF:** https://arxiv.org/pdf/2603.21375v1
+
+## 2801. PLR: Plackett-Luce for Reordering In-Context Learning Examples
+
+- **ID:** `2603.21373` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Pawel Batorski, Paul Swoboda
+- **abs:** https://arxiv.org/abs/2603.21373 · **PDF:** https://arxiv.org/pdf/2603.21373v1
+
+## 2802. Conspiracy Frame: a Semiotically-Driven Approach for Conspiracy Theories Detection
+
+- **ID:** `2603.21368` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Heidi Campana Piva, Shaina Ashraf, Maziar Kianimoghadam Jouneghani, et al.
+- **abs:** https://arxiv.org/abs/2603.21368 · **PDF:** https://arxiv.org/pdf/2603.21368v1
+
+## 2803. TIDE: Token-Informed Depth Execution for Per-Token Early Exit in LLM Inference
+
+- **ID:** `2603.21365` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Jaber Jaber, Osama Jaber
+- **abs:** https://arxiv.org/abs/2603.21365 · **PDF:** https://arxiv.org/pdf/2603.21365v1
+
+## 2804. AdaRubric: Task-Adaptive Rubrics for LLM Agent Evaluation
+
+- **ID:** `2603.21362` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Liang Ding
+- **abs:** https://arxiv.org/abs/2603.21362 · **PDF:** https://arxiv.org/pdf/2603.21362v1
+
+## 2805. Benchmarking Bengali Dialectal Bias: A Multi-Stage Framework Integrating RAG-Based Translation and Human-Augmented RLAIF
+
+- **ID:** `2603.21359` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** K. M. Jubair Sami, Dipto Sumit, Ariyan Hossain, et al.
+- **abs:** https://arxiv.org/abs/2603.21359 · **PDF:** https://arxiv.org/pdf/2603.21359v1
+
+## 2806. AgentHER: Hindsight Experience Replay for LLM Agent Trajectory Relabeling
+
+- **ID:** `2603.21357` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Liang Ding
+- **abs:** https://arxiv.org/abs/2603.21357 · **PDF:** https://arxiv.org/pdf/2603.21357v1
+
+## 2807. The Workload-Router-Pool Architecture for LLM Inference Optimization: A Vision Paper from the vLLM Semantic Router Project
+
+- **ID:** `2603.21354` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Huamin Chen, Xunzhuo Liu, Bowei He, et al.
+- **abs:** https://arxiv.org/abs/2603.21354 · **PDF:** https://arxiv.org/pdf/2603.21354v1
+
+## 2808. Beyond Memorization: Distinguishing between Reductive and Epistemic Reasoning in LLMs using Classic Logic Puzzles
+
+- **ID:** `2603.21350` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Adi Gabay, Gabriel Stanovsky, Liat Peterfreund
+- **abs:** https://arxiv.org/abs/2603.21350 · **PDF:** https://arxiv.org/pdf/2603.21350v1
+
+## 2809. The AI Scientific Community: Agentic Virtual Lab Swarms
+
+- **ID:** `2603.21344` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Ulisses Braga-Neto
+- **abs:** https://arxiv.org/abs/2603.21344 · **PDF:** https://arxiv.org/pdf/2603.21344v1
+
+## 2810. Generalized Discrete Diffusion from Snapshots
+
+- **ID:** `2603.21342` · **primary:** `stat.ML` · **Submitted:** 2026-03-22
+- **Authors:** Oussama Zekri, Théo Uscidda, Nicolas Boullé, et al.
+- **abs:** https://arxiv.org/abs/2603.21342 · **PDF:** https://arxiv.org/pdf/2603.21342v1
+
+## 2811. RoboAlign: Learning Test-Time Reasoning for Language-Action Alignment in Vision-Language-Action Models
+
+- **ID:** `2603.21341` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Dongyoung Kim, Sumin Park, Woomin Song, et al.
+- **abs:** https://arxiv.org/abs/2603.21341 · **PDF:** https://arxiv.org/pdf/2603.21341v1
+
+## 2812. ARYA: A Physics-Constrained Composable & Deterministic World Model Architecture
+
+- **ID:** `2603.21340` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Seth Dobrin, Lukasz Chmiel
+- **abs:** https://arxiv.org/abs/2603.21340 · **PDF:** https://arxiv.org/pdf/2603.21340v1
+
+## 2813. TimeTox: An LLM-Based Pipeline for Automated Extraction of Time Toxicity from Clinical Trial Protocols
+
+- **ID:** `2603.21335` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Saketh Vinjamuri, Marielle Fis Loperena, Marie C. Spezia, et al.
+- **abs:** https://arxiv.org/abs/2603.21335 · **PDF:** https://arxiv.org/pdf/2603.21335v2
+
+## 2814. AutoKernel: Autonomous GPU Kernel Optimization via Iterative Agent-Driven Search
+
+- **ID:** `2603.21331` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Jaber Jaber, Osama Jaber
+- **abs:** https://arxiv.org/abs/2603.21331 · **PDF:** https://arxiv.org/pdf/2603.21331v1
+
+## 2815. FinRL-X: An AI-Native Modular Infrastructure for Quantitative Trading
+
+- **ID:** `2603.21330` · **primary:** `q-fin.TR` · **Submitted:** 2026-03-22
+- **Authors:** Hongyang Yang, Boyu Zhang, Yang She, et al.
+- **abs:** https://arxiv.org/abs/2603.21330 · **PDF:** https://arxiv.org/pdf/2603.21330v1
+
+## 2816. COINBench: Moving Beyond Individual Perspectives to Collective Intent Understanding
+
+- **ID:** `2603.21329` · **primary:** `cs.IR` · **Submitted:** 2026-03-22
+- **Authors:** Xiaozhe Li, Tianyi Lyu, Siyi Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.21329 · **PDF:** https://arxiv.org/pdf/2603.21329v1
+
+## 2817. B-jet Tagging Using a Hybrid Edge Convolution and Transformer Architecture
+
+- **ID:** `2603.21326` · **primary:** `hep-ph` · **Submitted:** 2026-03-22
+- **Authors:** Diego F. Vasquez Plaza, Vidya Manian
+- **abs:** https://arxiv.org/abs/2603.21326 · **PDF:** https://arxiv.org/pdf/2603.21326v1
+
+## 2818. Which Alert Removals are Beneficial?
+
+- **ID:** `2603.21322` · **primary:** `cs.SE` · **Submitted:** 2026-03-22
+- **Authors:** Idan Amit
+- **abs:** https://arxiv.org/abs/2603.21322 · **PDF:** https://arxiv.org/pdf/2603.21322v1
+
+## 2819. Improving Coherence and Persistence in Agentic AI for System Optimization
+
+- **ID:** `2603.21321` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Pantea Karimi, Kimia Noorbakhsh, Mohammad Alizadeh, et al.
+- **abs:** https://arxiv.org/abs/2603.21321 · **PDF:** https://arxiv.org/pdf/2603.21321v1
+
+## 2820. Active Inference Agency Formalization, Metrics, and Convergence Assessments
+
+- **ID:** `2603.21319` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Eduard Kapelko
+- **abs:** https://arxiv.org/abs/2603.21319 · **PDF:** https://arxiv.org/pdf/2603.21319v1
+
+## 2821. Stream separation improves Bregman conditioning in transformers
+
+- **ID:** `2603.21317` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** James Clayton Kerce
+- **abs:** https://arxiv.org/abs/2603.21317 · **PDF:** https://arxiv.org/pdf/2603.21317v1
+
+## 2822. Bridging neuroscience and AI: adaptive, culturally sensitive technologies transforming aphasia rehabilitation
+
+- **ID:** `2603.22357` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-22
+- **Authors:** Andreea I. Niculescu, Jochen Ehnes, Minghui Dong
+- **abs:** https://arxiv.org/abs/2603.22357 · **PDF:** https://arxiv.org/pdf/2603.22357v1
+
+## 2823. HELIX: Scaling Raw Audio Understanding with Hybrid Mamba-Attention Beyond the Quadratic Limit
+
+- **ID:** `2603.21316` · **primary:** `cs.SD` · **Submitted:** 2026-03-22
+- **Authors:** Khushiyant, Param Thakkar
+- **abs:** https://arxiv.org/abs/2603.21316 · **PDF:** https://arxiv.org/pdf/2603.21316v1
+
+## 2824. FluidWorld: Reaction-Diffusion Dynamics as a Predictive Substrate for World Models
+
+- **ID:** `2603.21315` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Fabien Polly
+- **abs:** https://arxiv.org/abs/2603.21315 · **PDF:** https://arxiv.org/pdf/2603.21315v1
+
+## 2825. Direct Interval Propagation Methods using Neural-Network Surrogates for Uncertainty Quantification in Physical Systems Surrogate Model
+
+- **ID:** `2603.21308` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Ghifari Adam Faza, Jolan Wauters, Fabio Cuzzolin, et al.
+- **abs:** https://arxiv.org/abs/2603.21308 · **PDF:** https://arxiv.org/pdf/2603.21308v1
+
+## 2826. Enhancing reasoning accuracy in large language models during inference time
+
+- **ID:** `2603.21301` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Vinay Sharma, Manish Jain
+- **abs:** https://arxiv.org/abs/2603.21301 · **PDF:** https://arxiv.org/pdf/2603.21301v1
+
+## 2827. The Average Relative Entropy and Transpilation Depth determines the noise robustness in Variational Quantum Classifiers
+
+- **ID:** `2603.21300` · **primary:** `quant-ph` · **Submitted:** 2026-03-22
+- **Authors:** Aakash Ravindra Shinde, Arianne Meijer - van de Griend, Jukka K. Nurminen
+- **abs:** https://arxiv.org/abs/2603.21300 · **PDF:** https://arxiv.org/pdf/2603.21300v1
+
+## 2828. More Than Sum of Its Parts: Deciphering Intent Shifts in Multimodal Hate Speech Detection
+
+- **ID:** `2603.21298` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Runze Sun, Yu Zheng, Zexuan Xiong, et al.
+- **abs:** https://arxiv.org/abs/2603.21298 · **PDF:** https://arxiv.org/pdf/2603.21298v1
+
+## 2829. DeepXplain: XAI-Guided Autonomous Defense Against Multi-Stage APT Campaigns
+
+- **ID:** `2603.21296` · **primary:** `cs.CR` · **Submitted:** 2026-03-22
+- **Authors:** Trung V. Phan, Thomas Bauschert
+- **abs:** https://arxiv.org/abs/2603.21296 · **PDF:** https://arxiv.org/pdf/2603.21296v3
+
+## 2830. Closed-form conditional diffusion models for data assimilation
+
+- **ID:** `2603.21291` · **primary:** `stat.ML` · **Submitted:** 2026-03-22
+- **Authors:** Brianna Binder, Agnimitra Dasgupta, Assad Oberai
+- **abs:** https://arxiv.org/abs/2603.21291 · **PDF:** https://arxiv.org/pdf/2603.21291v2
+
+## 2831. When Models Judge Themselves: Unsupervised Self-Evolution for Multimodal Reasoning
+
+- **ID:** `2603.21289` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** Zhengxian Wu, Kai Shi, Chuanrui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.21289 · **PDF:** https://arxiv.org/pdf/2603.21289v2
+
+## 2832. Sonny: Breaking the Compute Wall in Medium-Range Weather Forecasting
+
+- **ID:** `2603.21284` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Minjong Cheon
+- **abs:** https://arxiv.org/abs/2603.21284 · **PDF:** https://arxiv.org/pdf/2603.21284v1
+
+## 2833. Fusing Memory and Attention: A study on LSTM, Transformer and Hybrid Architectures for Symbolic Music Generation
+
+- **ID:** `2603.21282` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Soudeep Ghoshal, Sandipan Chakraborty, Pradipto Chowdhury, et al.
+- **abs:** https://arxiv.org/abs/2603.21282 · **PDF:** https://arxiv.org/pdf/2603.21282v1
+
+## 2834. WARBENCH: A Comprehensive Benchmark for Evaluating LLMs in Military Decision-Making
+
+- **ID:** `2603.21280` · **primary:** `cs.CY` · **Submitted:** 2026-03-22
+- **Authors:** Zongjie Li, Chaozheng Wang, Yuchong Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.21280 · **PDF:** https://arxiv.org/pdf/2603.21280v1
+
+## 2835. Conversation Tree Architecture: A Structured Framework for Context-Aware Multi-Branch LLM Conversations
+
+- **ID:** `2603.21278` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Pranav Hemanth, Sampriti Saha
+- **abs:** https://arxiv.org/abs/2603.21278 · **PDF:** https://arxiv.org/pdf/2603.21278v1
+
+## 2836. Aggregation Alignment for Federated Learning with Mixture-of-Experts under Data Heterogeneity
+
+- **ID:** `2603.21276` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Zihan Fang, Qianru Wang, Haonan An, et al.
+- **abs:** https://arxiv.org/abs/2603.21276 · **PDF:** https://arxiv.org/pdf/2603.21276v1
+
+## 2837. The Library Theorem: How External Organization Governs Agentic Reasoning Capacity
+
+- **ID:** `2603.21272` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Zachary F. Mainen
+- **abs:** https://arxiv.org/abs/2603.21272 · **PDF:** https://arxiv.org/pdf/2603.21272v1
+
+## 2838. Graph of States: Solving Abductive Tasks with Large Language Models
+
+- **ID:** `2603.21250` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Yu Luo, Rongchen Gao, Lu Teng, et al.
+- **abs:** https://arxiv.org/abs/2603.21250 · **PDF:** https://arxiv.org/pdf/2603.21250v1
+
+## 2839. Demystifying Low-Rank Knowledge Distillation in Large Language Models: Convergence, Generalization, and Information-Theoretic Guarantees
+
+- **ID:** `2603.22355` · **primary:** `stat.ML` · **Submitted:** 2026-03-22
+- **Authors:** Alberlucia Rafael Soarez, Daniel Kim, Mariana Costa, et al.
+- **abs:** https://arxiv.org/abs/2603.22355 · **PDF:** https://arxiv.org/pdf/2603.22355v1
+
+## 2840. Graph Fusion Across Languages using Large Language Models
+
+- **ID:** `2603.21248` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Kaung Myat Kyaw, Khush Agarwal, Jonathan Chan
+- **abs:** https://arxiv.org/abs/2603.21248 · **PDF:** https://arxiv.org/pdf/2603.21248v1
+
+## 2841. Accelerate Vector Diffusion Maps by Landmarks
+
+- **ID:** `2603.21247` · **primary:** `stat.ML` · **Submitted:** 2026-03-22
+- **Authors:** Sing-Yuan Yeh, Yi-An Wu, Hau-Tieng Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.21247 · **PDF:** https://arxiv.org/pdf/2603.21247v1
+
+## 2842. Amortized Variational Inference for Logistic Regression with Missing Covariates
+
+- **ID:** `2603.21244` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** M. Cherifi, Aude Sportisse, Xujia Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.21244 · **PDF:** https://arxiv.org/pdf/2603.21244v1
+
+## 2843. ConsRoute:Consistency-Aware Adaptive Query Routing for Cloud-Edge-Device Large Language Models
+
+- **ID:** `2603.21237` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Haoyu Qiao, Hao Zhang, Shanwen Mao, et al.
+- **abs:** https://arxiv.org/abs/2603.21237 · **PDF:** https://arxiv.org/pdf/2603.21237v1
+
+## 2844. Does Mechanistic Interpretability Transfer Across Data Modalities? A Cross-Domain Causal Circuit Analysis of Variational Autoencoders
+
+- **ID:** `2603.21236` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Dip Roy, Rajiv Misra, Sanjay Kumar Singh, et al.
+- **abs:** https://arxiv.org/abs/2603.21236 · **PDF:** https://arxiv.org/pdf/2603.21236v1
+
+## 2845. Domain Elastic Transform: Bayesian Function Registration for High-Dimensional Scientific Data
+
+- **ID:** `2603.21235` · **primary:** `stat.ML` · **Submitted:** 2026-03-22
+- **Authors:** Osamu Hirose, Emanuele Rodola
+- **abs:** https://arxiv.org/abs/2603.21235 · **PDF:** https://arxiv.org/pdf/2603.21235v1
+
+## 2846. QMoP: Query Guided Mixture-of-Projector for Efficient Visual Token Compression
+
+- **ID:** `2603.21232` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** Zhongyang Li, Yaqian Li, Faming Fang, et al.
+- **abs:** https://arxiv.org/abs/2603.21232 · **PDF:** https://arxiv.org/pdf/2603.21232v1
+
+## 2847. When Convenience Becomes Risk: A Semantic View of Under-Specification in Host-Acting Agents
+
+- **ID:** `2603.21231` · **primary:** `cs.CR` · **Submitted:** 2026-03-22
+- **Authors:** Di Lu, Yongzhi Liao, Xutong Mu, et al.
+- **abs:** https://arxiv.org/abs/2603.21231 · **PDF:** https://arxiv.org/pdf/2603.21231v1
+
+## 2848. Does AI Homogenize Student Thinking? A Multi-Dimensional Analysis of Structural Convergence in AI-Augmented Essays
+
+- **ID:** `2603.21228` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Keito Inoshita, Michiaki Omura, Tsukasa Yamanaka, et al.
+- **abs:** https://arxiv.org/abs/2603.21228 · **PDF:** https://arxiv.org/pdf/2603.21228v1
+
+## 2849. Positional Segmentor-Guided Counterfactual Fine-Tuning for Spatially Localized Image Synthesis
+
+- **ID:** `2603.21213` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** Tian Xia, Matthew Sinclair, Andreas Schuh, et al.
+- **abs:** https://arxiv.org/abs/2603.21213 · **PDF:** https://arxiv.org/pdf/2603.21213v1
+
+## 2850. Pretrained Video Models as Differentiable Physics Simulators for Urban Wind Flows
+
+- **ID:** `2603.21210` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Janne Perini, Rafael Bischof, Moab Arar, et al.
+- **abs:** https://arxiv.org/abs/2603.21210 · **PDF:** https://arxiv.org/pdf/2603.21210v1
+
+## 2851. JANUS: A Lightweight Framework for Jailbreaking Text-to-Image Models via Distribution Optimization
+
+- **ID:** `2603.21208` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** Haolun Zheng, Yu He, Tailun Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.21208 · **PDF:** https://arxiv.org/pdf/2603.21208v2
+
+## 2852. Is Monitoring Enough? Strategic Agent Selection For Stealthy Attack in Multi-Agent Discussions
+
+- **ID:** `2603.21194` · **primary:** `cs.CR` · **Submitted:** 2026-03-22
+- **Authors:** Qiuchi Xiang, Haoxuan Qu, Hossein Rahmani, et al.
+- **abs:** https://arxiv.org/abs/2603.21194 · **PDF:** https://arxiv.org/pdf/2603.21194v1
+
+## 2853. Context Selection for Hypothesis and Statistical Evidence Extraction from Full-Text Scientific Articles
+
+- **ID:** `2603.21193` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Sai Koneru, Jian Wu, Sarah Rajtmajer
+- **abs:** https://arxiv.org/abs/2603.21193 · **PDF:** https://arxiv.org/pdf/2603.21193v1
+
+## 2854. On the Role of Batch Size in Stochastic Conditional Gradient Methods
+
+- **ID:** `2603.21191` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Rustem Islamov, Roman Machacek, Aurelien Lucchi, et al.
+- **abs:** https://arxiv.org/abs/2603.21191 · **PDF:** https://arxiv.org/pdf/2603.21191v1
+
+## 2855. Architecture for Multi-Unmanned Aerial Vehicles based Autonomous Precision Agriculture Systems
+
+- **ID:** `2603.21183` · **primary:** `cs.RO` · **Submitted:** 2026-03-22
+- **Authors:** Ebasa Temesgen, Nathnael Minyelshowa, Lebsework Negash
+- **abs:** https://arxiv.org/abs/2603.21183 · **PDF:** https://arxiv.org/pdf/2603.21183v1
+
+## 2856. ALMAB-DC: Active Learning, Multi-Armed Bandits, and Distributed Computing for Sequential Experimental Design and Black-Box Optimization
+
+- **ID:** `2603.21180` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Foo Hui-Mean, Yuan-chin I Chang
+- **abs:** https://arxiv.org/abs/2603.21180 · **PDF:** https://arxiv.org/pdf/2603.21180v1
+
+## 2857. LLM-based Automated Architecture View Generation: Where Are We Now?
+
+- **ID:** `2603.21178` · **primary:** `cs.SE` · **Submitted:** 2026-03-22
+- **Authors:** Miryala Sathvika, Rudra Dhar, Karthik Vaidhyanathan
+- **abs:** https://arxiv.org/abs/2603.21178 · **PDF:** https://arxiv.org/pdf/2603.21178v1
+
+## 2858. Prompt replay: speeding up grpo with on-policy reuse of high-signal prompts
+
+- **ID:** `2603.21177` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Andrei Baroian, Rutger Berger
+- **abs:** https://arxiv.org/abs/2603.21177 · **PDF:** https://arxiv.org/pdf/2603.21177v1
+
+## 2859. Reward Sharpness-Aware Fine-Tuning for Diffusion Models
+
+- **ID:** `2603.21175` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Kwanyoung Kim, Byeongsu Sim
+- **abs:** https://arxiv.org/abs/2603.21175 · **PDF:** https://arxiv.org/pdf/2603.21175v1
+
+## 2860. Explainable Semantic Textual Similarity via Dissimilar Span Detection
+
+- **ID:** `2603.21174` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Diego Miguel Lozano, Daryna Dementieva, Alexander Fraser
+- **abs:** https://arxiv.org/abs/2603.21174 · **PDF:** https://arxiv.org/pdf/2603.21174v1
+
+## 2861. Rethinking Plasticity in Deep Reinforcement Learning
+
+- **ID:** `2603.21173` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Zhiqiang He
+- **abs:** https://arxiv.org/abs/2603.21173 · **PDF:** https://arxiv.org/pdf/2603.21173v1
+
+## 2862. Entropy Alone is Insufficient for Safe Selective Prediction in LLMs
+
+- **ID:** `2603.21172` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Edward Phillips, Fredrik K. Gustafsson, Sean Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.21172 · **PDF:** https://arxiv.org/pdf/2603.21172v1
+
+## 2863. Pruned Adaptation Modules: A Simple yet Strong Baseline for Continual Foundation Models
+
+- **ID:** `2603.21170` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Elif Ceren Gok Yildirim, Murat Onur Yildirim, Joaquin Vanschoren
+- **abs:** https://arxiv.org/abs/2603.21170 · **PDF:** https://arxiv.org/pdf/2603.21170v1
+
+## 2864. Model Evolution Under Zeroth-Order Optimization: A Neural Tangent Kernel Perspective
+
+- **ID:** `2603.21169` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Chen Zhang, Yuxin Cheng, Chenchen Ding, et al.
+- **abs:** https://arxiv.org/abs/2603.21169 · **PDF:** https://arxiv.org/pdf/2603.21169v1
+
+## 2865. Many Dialects, Many Languages, One Cultural Lens: Evaluating Multilingual VLMs for Bengali Culture Understanding Across Historically Linked Languages and Regional Dialects
+
+- **ID:** `2603.21165` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Nurul Labib Sayeedi, Md. Faiyaz Abdullah Sayeedi, Shubhashis Roy Dipta, et al.
+- **abs:** https://arxiv.org/abs/2603.21165 · **PDF:** https://arxiv.org/pdf/2603.21165v1
+
+## 2866. Revisiting Tree Search for LLMs: Gumbel and Sequential Halving for Budget-Scalable Reasoning
+
+- **ID:** `2603.21162` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Leonid Ugadiarov, Yuri Kuratov, Aleksandr Panov, et al.
+- **abs:** https://arxiv.org/abs/2603.21162 · **PDF:** https://arxiv.org/pdf/2603.21162v1
+
+## 2867. Beyond a Single Signal: SPECTREG2, A Unified MultiExpert Anomaly Detector for Unknown Unknowns
+
+- **ID:** `2603.21160` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Rahul D Ray
+- **abs:** https://arxiv.org/abs/2603.21160 · **PDF:** https://arxiv.org/pdf/2603.21160v1
+
+## 2868. Can LLMs Fool Graph Learning? Exploring Universal Adversarial Attacks on Text-Attributed Graphs
+
+- **ID:** `2603.21155` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Zihui Chen, Yuling Wang, Pengfei Jiao, et al.
+- **abs:** https://arxiv.org/abs/2603.21155 · **PDF:** https://arxiv.org/pdf/2603.21155v1
+
+## 2869. Learning from Label Proportions with Dual-proportion Constraints
+
+- **ID:** `2603.21153` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Tianhao Ma, Ximing Li, Changchun Li, et al.
+- **abs:** https://arxiv.org/abs/2603.21153 · **PDF:** https://arxiv.org/pdf/2603.21153v1
+
+## 2870. TRACE: A Multi-Agent System for Autonomous Physical Reasoning for Seismology
+
+- **ID:** `2603.21152` · **primary:** `physics.geo-ph` · **Submitted:** 2026-03-22
+- **Authors:** Feng Liu, Jian Xu, Xin Cui, et al.
+- **abs:** https://arxiv.org/abs/2603.21152 · **PDF:** https://arxiv.org/pdf/2603.21152v3
+
+## 2871. Emergent Formal Verification: How an Autonomous AI Ecosystem Independently Discovered SMT-Based Safety Across Six Domains
+
+- **ID:** `2603.21149` · **primary:** `cs.SE` · **Submitted:** 2026-03-22
+- **Authors:** Octavian Untila
+- **abs:** https://arxiv.org/abs/2603.21149 · **PDF:** https://arxiv.org/pdf/2603.21149v1
+
+## 2872. WIST: Web-Grounded Iterative Self-Play Tree for Domain-Targeted Reasoning Improvement
+
+- **ID:** `2603.22352` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Fangyuan Li, Pengfei Li, Shijie Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.22352 · **PDF:** https://arxiv.org/pdf/2603.22352v1
+
+## 2873. NeSy-Edge: Neuro-Symbolic Trustworthy Self-Healing in the Computing Continuum
+
+- **ID:** `2603.21145` · **primary:** `cs.DC` · **Submitted:** 2026-03-22
+- **Authors:** Peihan Ye, Alfreds Lapkovskis, Alaa Saleh, et al.
+- **abs:** https://arxiv.org/abs/2603.21145 · **PDF:** https://arxiv.org/pdf/2603.21145v1
+
+## 2874. Time-adaptive functional Gaussian Process regression
+
+- **ID:** `2603.21144` · **primary:** `stat.ML` · **Submitted:** 2026-03-22
+- **Authors:** MD Ruiz-Medina, AE Madrid, A Torres-Signes, et al.
+- **abs:** https://arxiv.org/abs/2603.21144 · **PDF:** https://arxiv.org/pdf/2603.21144v1
+
+## 2875. ORACLE: Optimizing Reasoning Abilities of Large Language Models via Constraint-Led Synthetic Data Elicitation
+
+- **ID:** `2603.21140` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Zhuojie Yang, Wentao Wan, Keze Wang
+- **abs:** https://arxiv.org/abs/2603.21140 · **PDF:** https://arxiv.org/pdf/2603.21140v1
+
+## 2876. Ontology-driven personalized information retrieval for XML documents
+
+- **ID:** `2603.21139` · **primary:** `cs.IR` · **Submitted:** 2026-03-22
+- **Authors:** Ounnaci Iddir, Ahmed-ouamer Rachid, Tai Dinh
+- **abs:** https://arxiv.org/abs/2603.21139 · **PDF:** https://arxiv.org/pdf/2603.21139v1
+
+## 2877. SleepVLM: Explainable and Rule-Grounded Sleep Staging via a Vision-Language Model
+
+- **ID:** `2603.26738` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** Guifeng Deng, Pan Wang, Jiquan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.26738 · **PDF:** https://arxiv.org/pdf/2603.26738v2
+
+## 2878. One Pool Is Not Enough: Multi-Cluster Memory for Practical Test-Time Adaptation
+
+- **ID:** `2603.21135` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** Yu-Wen Tseng, Xingyi Zheng, Ya-Chen Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.21135 · **PDF:** https://arxiv.org/pdf/2603.21135v1
+
+## 2879. Session Risk Memory (SRM): Temporal Authorization for Deterministic Pre-Execution Safety Gates
+
+- **ID:** `2603.22350` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Florin Adrian Chitan
+- **abs:** https://arxiv.org/abs/2603.22350 · **PDF:** https://arxiv.org/pdf/2603.22350v1
+
+## 2880. Frequency Switching Mechanism for Parameter-E!cient Multi-Task Learning
+
+- **ID:** `2603.21111` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** Shih-Wen Liu, Yen-Chang Chen, Wei-Ta Chu, et al.
+- **abs:** https://arxiv.org/abs/2603.21111 · **PDF:** https://arxiv.org/pdf/2603.21111v1
+
+## 2881. DMMRL: Disentangled Multi-Modal Representation Learning via Variational Autoencoders for Molecular Property Prediction
+
+- **ID:** `2603.21108` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Long Xu, Junping Guo, Jianbo Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.21108 · **PDF:** https://arxiv.org/pdf/2603.21108v1
+
+## 2882. ResPrune: Text-Conditioned Subspace Reconstruction for Visual Token Pruning in Large Vision-Language Models
+
+- **ID:** `2603.21105` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Xu Li, Yi Zheng, Yuxuan Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.21105 · **PDF:** https://arxiv.org/pdf/2603.21105v1
+
+## 2883. Learning Progressive Adaptation for Multi-Modal Tracking
+
+- **ID:** `2603.21100` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** He Wang, Tianyang Xu, Zhangyong Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.21100 · **PDF:** https://arxiv.org/pdf/2603.21100v1
+
+## 2884. Learning to Optimize Joint Source and RIS-assisted Channel Encoding for Multi-User Semantic Communication Systems
+
+- **ID:** `2603.21097` · **primary:** `cs.NI` · **Submitted:** 2026-03-22
+- **Authors:** Haidong Wang, Songhan Zhao, Bo Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.21097 · **PDF:** https://arxiv.org/pdf/2603.21097v1
+
+## 2885. Mixture of Chapters: Scaling Learnt Memory in Transformers
+
+- **ID:** `2603.21096` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Tasmay Pankaj Tibrewal, Pritish Saha, Ankit Meda, et al.
+- **abs:** https://arxiv.org/abs/2603.21096 · **PDF:** https://arxiv.org/pdf/2603.21096v1
+
+## 2886. Representation-Level Adversarial Regularization for Clinically Aligned Multitask Thyroid Ultrasound Assessment
+
+- **ID:** `2603.21095` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** Dina Salama, Mohamed Mahmoud, Nourhan Bayasi, et al.
+- **abs:** https://arxiv.org/abs/2603.21095 · **PDF:** https://arxiv.org/pdf/2603.21095v1
+
+## 2887. ReasonScaffold: A Scaffolded Reasoning-based Annotation Protocol for Human-AI Co-Annotation
+
+- **ID:** `2603.21094` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Smitha Muthya Sudheendra, Jaideep Srivastava
+- **abs:** https://arxiv.org/abs/2603.21094 · **PDF:** https://arxiv.org/pdf/2603.21094v2
+
+## 2888. Stochastic approximation in non-markovian environments revisited
+
+- **ID:** `2603.21091` · **primary:** `stat.ML` · **Submitted:** 2026-03-22
+- **Authors:** Vivek Shripad Borkar
+- **abs:** https://arxiv.org/abs/2603.21091 · **PDF:** https://arxiv.org/pdf/2603.21091v1
+
+## 2889. ViCLSR: A Supervised Contrastive Learning Framework with Natural Language Inference for Natural Language Understanding Tasks
+
+- **ID:** `2603.21084` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Tin Van Huynh, Kiet Van Nguyen, Ngan Luu-Thuy Nguyen
+- **abs:** https://arxiv.org/abs/2603.21084 · **PDF:** https://arxiv.org/pdf/2603.21084v1
+
+## 2890. Assessing the Ability of Neural TTS Systems to Model Consonant-Induced F0 Perturbation
+
+- **ID:** `2603.21078` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Tianle Yang, Chengzhe Sun, Phil Rose, et al.
+- **abs:** https://arxiv.org/abs/2603.21078 · **PDF:** https://arxiv.org/pdf/2603.21078v1
+
+## 2891. SqueezeComposer: Temporal Speed-up is A Simple Trick for Long-form Music Composing
+
+- **ID:** `2603.21073` · **primary:** `eess.AS` · **Submitted:** 2026-03-22
+- **Authors:** Jianyi Chen, Rongxiu Zhong, Shilei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.21073 · **PDF:** https://arxiv.org/pdf/2603.21073v1
+
+## 2892. CTFS : Collaborative Teacher Framework for Forward-Looking Sonar Image Semantic Segmentation with Extremely Limited Labels
+
+- **ID:** `2603.21071` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** Ping Guo, Chengzhou Li, Guanchen Meng, et al.
+- **abs:** https://arxiv.org/abs/2603.21071 · **PDF:** https://arxiv.org/pdf/2603.21071v1
+
+## 2893. LongCat-Flash-Prover: Advancing Native Formal Reasoning via Agentic Tool-Integrated Reinforcement Learning
+
+- **ID:** `2603.21065` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Jianing Wang, Jianfei Zhang, Qi Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.21065 · **PDF:** https://arxiv.org/pdf/2603.21065v1
+
+## 2894. Gradient Descent with Projection Finds Over-Parameterized Neural Networks for Learning Low-Degree Polynomials with Nearly Minimax Optimal Rate
+
+- **ID:** `2603.21062` · **primary:** `stat.ML` · **Submitted:** 2026-03-22
+- **Authors:** Yingzhen Yang, Ping Li
+- **abs:** https://arxiv.org/abs/2603.21062 · **PDF:** https://arxiv.org/pdf/2603.21062v1
+
+## 2895. Semi-Supervised Learning with Balanced Deep Representation Distributions
+
+- **ID:** `2603.21056` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Changchun Li, Ximing Li, Bingjie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.21056 · **PDF:** https://arxiv.org/pdf/2603.21056v1
+
+## 2896. Harmful Visual Content Manipulation Matters in Misinformation Detection Under Multimedia Scenarios
+
+- **ID:** `2603.21054` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Bing Wang, Ximing Li, Changchun Li, et al.
+- **abs:** https://arxiv.org/abs/2603.21054 · **PDF:** https://arxiv.org/pdf/2603.21054v1
+
+## 2897. COMPASS-Hedge: Learning Safely Without Knowing the World
+
+- **ID:** `2603.22348` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Ting Hu, Luanda Cai, Manolis Vlatakis
+- **abs:** https://arxiv.org/abs/2603.22348 · **PDF:** https://arxiv.org/pdf/2603.22348v2
+
+## 2898. A Two-stage Transformer Framework for Temporal Localization of Distracted Driver Behaviors
+
+- **ID:** `2603.21048` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** Gia-Bao Doan, Nam-Khoa Huynh, Minh-Nhat-Huy Ho, et al.
+- **abs:** https://arxiv.org/abs/2603.21048 · **PDF:** https://arxiv.org/pdf/2603.21048v1
+
+## 2899. SpatialFly: Geometry-Guided Representation Alignment for UAV Vision-and-Language Navigation in Urban Environments
+
+- **ID:** `2603.21046` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** Wen Jiang, Kangyao Huang, Li Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.21046 · **PDF:** https://arxiv.org/pdf/2603.21046v1
+
+## 2900. LPNSR: Prior-Enhanced Diffusion Image Super-Resolution via LR-Guided Noise Prediction
+
+- **ID:** `2603.21045` · **primary:** `cs.CV` · **Submitted:** 2026-03-22
+- **Authors:** Shuwei Huang, Shizhuo Liu, Zijun Wei
+- **abs:** https://arxiv.org/abs/2603.21045 · **PDF:** https://arxiv.org/pdf/2603.21045v3
+
+## 2901. Confidence Freeze: Early Success Induces a Metastable Decoupling of Metacognition and Behaviour
+
+- **ID:** `2603.21043` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Zhipeng Zhang, Hongshun He
+- **abs:** https://arxiv.org/abs/2603.21043 · **PDF:** https://arxiv.org/pdf/2603.21043v1
+
+## 2902. Statistical Learning for Latent Embedding Alignment with Application to Brain Encoding and Decoding
+
+- **ID:** `2603.21042` · **primary:** `stat.ME` · **Submitted:** 2026-03-22
+- **Authors:** Shuoxun Xu, Zhanhao Yan, Lexin Li
+- **abs:** https://arxiv.org/abs/2603.21042 · **PDF:** https://arxiv.org/pdf/2603.21042v1
+
+## 2903. Intelligence Inertia: Physical Principles and Applications
+
+- **ID:** `2603.22347` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Jipeng Han
+- **abs:** https://arxiv.org/abs/2603.22347 · **PDF:** https://arxiv.org/pdf/2603.22347v1
+
+## 2904. Benchmarking Scientific Machine Learning Models for Air Quality Data
+
+- **ID:** `2603.21039` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Khawja Imran Masud, Venkata Sai Rahul Unnam, Sahara Ali
+- **abs:** https://arxiv.org/abs/2603.21039 · **PDF:** https://arxiv.org/pdf/2603.21039v2
+
+## 2905. Reading Between the Lines: How Electronic Nonverbal Cues shape Emotion Decoding
+
+- **ID:** `2603.21038` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Taara Kumar, Kokil Jaidka
+- **abs:** https://arxiv.org/abs/2603.21038 · **PDF:** https://arxiv.org/pdf/2603.21038v1
+
+## 2906. Left Behind: Cross-Lingual Transfer as a Bridge for Low-Resource Languages in Large Language Models
+
+- **ID:** `2603.21036` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Abdul-Salem Beibitkhan
+- **abs:** https://arxiv.org/abs/2603.21036 · **PDF:** https://arxiv.org/pdf/2603.21036v1
+
+## 2907. Fuel Consumption Prediction: A Comparative Analysis of Machine Learning Paradigms
+
+- **ID:** `2603.21034` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Ali Akram
+- **abs:** https://arxiv.org/abs/2603.21034 · **PDF:** https://arxiv.org/pdf/2603.21034v1
+
+## 2908. TabPFN Extensions for Interpretable Geotechnical Modelling
+
+- **ID:** `2603.21033` · **primary:** `cs.CE` · **Submitted:** 2026-03-22
+- **Authors:** Taiga Saito, Yu Otake, Daijiro Mizutani, et al.
+- **abs:** https://arxiv.org/abs/2603.21033 · **PDF:** https://arxiv.org/pdf/2603.21033v1
+
+## 2909. First-Mover Bias in Gradient Boosting Explanations: Mechanism, Detection, and Resolution
+
+- **ID:** `2603.22346` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Drake Caraker, Bryan Arnold, David Rhoads
+- **abs:** https://arxiv.org/abs/2603.22346 · **PDF:** https://arxiv.org/pdf/2603.22346v1
+
+## 2910. Deep Attention-based Sequential Ensemble Learning for BLE-Based Indoor Localization in Care Facilities
+
+- **ID:** `2603.21030` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Minh Triet Pham, Quynh Chi Dang, Le Nhat Tan
+- **abs:** https://arxiv.org/abs/2603.21030 · **PDF:** https://arxiv.org/pdf/2603.21030v1
+
+## 2911. KLDrive: Fine-Grained 3D Scene Reasoning for Autonomous Driving based on Knowledge Graph
+
+- **ID:** `2603.21029` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Ye Tian, Jingyi Zhang, Zihao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.21029 · **PDF:** https://arxiv.org/pdf/2603.21029v1
+
+## 2912. Dynamic Fusion-Aware Graph Convolutional Neural Network for Multimodal Emotion Recognition in Conversations
+
+- **ID:** `2603.22345` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Tao Meng, Weilun Tang, Yuntao Shou, et al.
+- **abs:** https://arxiv.org/abs/2603.22345 · **PDF:** https://arxiv.org/pdf/2603.22345v1
+
+## 2913. Mitigating Selection Bias in Large Language Models via Permutation-Aware GRPO
+
+- **ID:** `2603.21016` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Jinquan Zheng, Jia Yuan, Jiacheng Yao, et al.
+- **abs:** https://arxiv.org/abs/2603.21016 · **PDF:** https://arxiv.org/pdf/2603.21016v1
+
+## 2914. CLT-Forge: A Scalable Library for Cross-Layer Transcoders and Attribution Graphs
+
+- **ID:** `2603.21014` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Florent Draye, Abir Harrasse, Vedant Palit, et al.
+- **abs:** https://arxiv.org/abs/2603.21014 · **PDF:** https://arxiv.org/pdf/2603.21014v1
+
+## 2915. Relational graph-driven differential denoising and diffusion attention fusion for multimodal conversation emotion recognition
+
+- **ID:** `2603.25752` · **primary:** `cs.CL` · **Submitted:** 2026-03-22
+- **Authors:** Ying Liu, Yuntao Shou, Wei Ai, et al.
+- **abs:** https://arxiv.org/abs/2603.25752 · **PDF:** https://arxiv.org/pdf/2603.25752v1
+
+## 2916. When Does Content-Based Routing Work? Representation Requirements for Selective Attention in Hybrid Sequence Models
+
+- **ID:** `2603.20997` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Abhinaba Basu
+- **abs:** https://arxiv.org/abs/2603.20997 · **PDF:** https://arxiv.org/pdf/2603.20997v1
+
+## 2917. The Intelligent Disobedience Game: Formulating Disobedience in Stackelberg Games and Markov Decision Processes
+
+- **ID:** `2603.20994` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Benedikt Hornig, Reuth Mirsky
+- **abs:** https://arxiv.org/abs/2603.20994 · **PDF:** https://arxiv.org/pdf/2603.20994v1
+
+## 2918. Long-Term Outlier Prediction Through Outlier Score Modeling
+
+- **ID:** `2603.20993` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Yuma Aoki, Joon Park, Koh Takeuchi, et al.
+- **abs:** https://arxiv.org/abs/2603.20993 · **PDF:** https://arxiv.org/pdf/2603.20993v1
+
+## 2919. Structural Sensitivity in Compressed Transformers: Error Propagation, Lyapunov Stability, and Formally Verified Bounds
+
+- **ID:** `2603.20991` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Abhinaba Basu
+- **abs:** https://arxiv.org/abs/2603.20991 · **PDF:** https://arxiv.org/pdf/2603.20991v1
+
+## 2920. ECI: Effective Contrastive Information to Evaluate Hard-Negatives
+
+- **ID:** `2603.20990` · **primary:** `cs.IR` · **Submitted:** 2026-03-22
+- **Authors:** Aarush Sinha, Rahul Seetharaman, Aman Bansal
+- **abs:** https://arxiv.org/abs/2603.20990 · **PDF:** https://arxiv.org/pdf/2603.20990v1
+
+## 2921. Can we automatize scientific discovery in the cognitive sciences?
+
+- **ID:** `2603.20988` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Akshay K. Jagadish, Milena Rmus, Kristin Witte, et al.
+- **abs:** https://arxiv.org/abs/2603.20988 · **PDF:** https://arxiv.org/pdf/2603.20988v1
+
+## 2922. Interpreting the Synchronization Gap: The Hidden Mechanism Inside Diffusion Transformers
+
+- **ID:** `2603.20987` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Emil Albrychiewicz, Andrés Franco Valiente, Li-Ching Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.20987 · **PDF:** https://arxiv.org/pdf/2603.20987v1
+
+## 2923. AutoMOOSE: An Agentic AI for Autonomous Phase-Field Simulation
+
+- **ID:** `2603.20986` · **primary:** `cs.AI` · **Submitted:** 2026-03-22
+- **Authors:** Sukriti Manna, Henry Chan, Subramanian K. R. S. Sankaranarayanan
+- **abs:** https://arxiv.org/abs/2603.20986 · **PDF:** https://arxiv.org/pdf/2603.20986v1
+
+## 2924. Joint Surrogate Learning of Objectives, Constraints, and Sensitivities for Efficient Multi-objective Optimization of Neural Dynamical Systems
+
+- **ID:** `2603.20984` · **primary:** `cs.LG` · **Submitted:** 2026-03-22
+- **Authors:** Frithjof Gressmann, Ivan Georgiev Raikov, Seung Hyun Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.20984 · **PDF:** https://arxiv.org/pdf/2603.20984v1
+
+## 2925. Cyber Deception for Mission Surveillance via Hypergame-Theoretic Deep Reinforcement Learning
+
+- **ID:** `2603.20981` · **primary:** `cs.CR` · **Submitted:** 2026-03-21
+- **Authors:** Zelin Wan, Jin-Hee Cho, Mu Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.20981 · **PDF:** https://arxiv.org/pdf/2603.20981v1
+
+## 2926. From Causal Discovery to Dynamic Causal Inference in Neural Time Series
+
+- **ID:** `2603.20980` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Valentina Kuskova, Dmitry Zaytsev, Michael Coppedge
+- **abs:** https://arxiv.org/abs/2603.20980 · **PDF:** https://arxiv.org/pdf/2603.20980v1
+
+## 2927. Detection of adversarial intent in Human-AI teams using LLMs
+
+- **ID:** `2603.20976` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Abed K. Musaffar, Ambuj Singh, Francesco Bullo
+- **abs:** https://arxiv.org/abs/2603.20976 · **PDF:** https://arxiv.org/pdf/2603.20976v1
+
+## 2928. DiscoUQ: Structured Disagreement Analysis for Uncertainty Quantification in LLM Agent Ensembles
+
+- **ID:** `2603.20975` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Bo Jiang
+- **abs:** https://arxiv.org/abs/2603.20975 · **PDF:** https://arxiv.org/pdf/2603.20975v1
+
+## 2929. Understanding Contextual Recall in Transformers: How Finetuning Enables In-Context Reasoning over Pretraining Knowledge
+
+- **ID:** `2603.20969` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Bhavya Vasudeva, Puneesh Deora, Alberto Bietti, et al.
+- **abs:** https://arxiv.org/abs/2603.20969 · **PDF:** https://arxiv.org/pdf/2603.20969v1
+
+## 2930. Hard labels sampled from sparse targets mislead rotation invariant algorithms
+
+- **ID:** `2603.20967` · **primary:** `stat.ML` · **Submitted:** 2026-03-21
+- **Authors:** Avrajit Ghosh, Bin Yu, Manfred Warmuth, et al.
+- **abs:** https://arxiv.org/abs/2603.20967 · **PDF:** https://arxiv.org/pdf/2603.20967v1
+
+## 2931. Learning to Aggregate Zero-Shot LLM Agents for Corporate Disclosure Classification
+
+- **ID:** `2603.20965` · **primary:** `q-fin.TR` · **Submitted:** 2026-03-21
+- **Authors:** Kemal Kirtac
+- **abs:** https://arxiv.org/abs/2603.20965 · **PDF:** https://arxiv.org/pdf/2603.20965v1
+
+## 2932. Elite Lanes: Evolutionary Generation of Realistic Small-Scale Road Networks
+
+- **ID:** `2603.20964` · **primary:** `cs.NE` · **Submitted:** 2026-03-21
+- **Authors:** Artur Morys-Magiera, Marek Długosz, Paweł Skruch
+- **abs:** https://arxiv.org/abs/2603.20964 · **PDF:** https://arxiv.org/pdf/2603.20964v1
+
+## 2933. Integrative Learning of Dynamically Evolving Multiplex Graphs and Nodal Attributes Using Neural Network Gaussian Processes with an Application to Dynamic Terrorism Graphs
+
+- **ID:** `2603.20962` · **primary:** `stat.AP` · **Submitted:** 2026-03-21
+- **Authors:** Jose Rodriguez-Acosta, Sharmistha Guha, Lekha Patel, et al.
+- **abs:** https://arxiv.org/abs/2603.20962 · **PDF:** https://arxiv.org/pdf/2603.20962v1
+
+## 2934. Alignment Whack-a-Mole : Finetuning Activates Verbatim Recall of Copyrighted Books in Large Language Models
+
+- **ID:** `2603.20957` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Xinyue Liu, Niloofar Mireshghallah, Jane C. Ginsburg, et al.
+- **abs:** https://arxiv.org/abs/2603.20957 · **PDF:** https://arxiv.org/pdf/2603.20957v3
+
+## 2935. Errors in AI-Assisted Retrieval of Medical Literature: A Comparative Study
+
+- **ID:** `2603.22344` · **primary:** `cs.IR` · **Submitted:** 2026-03-21
+- **Authors:** Jenny Gao, Yongfeng Zhang, Mary L Disis, et al.
+- **abs:** https://arxiv.org/abs/2603.22344 · **PDF:** https://arxiv.org/pdf/2603.22344v1
+
+## 2936. Beyond Expression Similarity: Contrastive Learning Recovers Functional Gene Associations from Protein Interaction Structure
+
+- **ID:** `2603.20955` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Jason Dury
+- **abs:** https://arxiv.org/abs/2603.20955 · **PDF:** https://arxiv.org/pdf/2603.20955v1
+
+## 2937. Before the Tool Call: Deterministic Pre-Action Authorization for Autonomous AI Agents
+
+- **ID:** `2603.20953` · **primary:** `cs.CR` · **Submitted:** 2026-03-21
+- **Authors:** Uchi Uchibeke
+- **abs:** https://arxiv.org/abs/2603.20953 · **PDF:** https://arxiv.org/pdf/2603.20953v1
+
+## 2938. gUFO: A Gentle Foundational Ontology for Semantic Web Knowledge Graphs
+
+- **ID:** `2603.20948` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** João Paulo A. Almeida, Giancarlo Guizzardi, Tiago Prince Sales, et al.
+- **abs:** https://arxiv.org/abs/2603.20948 · **PDF:** https://arxiv.org/pdf/2603.20948v1
+
+## 2939. User Preference Modeling for Conversational LLM Agents: Weak Rewards from Retrieval-Augmented Interaction
+
+- **ID:** `2603.20939` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Yuren Hao, Shuhaib Mehri, ChengXiang Zhai, et al.
+- **abs:** https://arxiv.org/abs/2603.20939 · **PDF:** https://arxiv.org/pdf/2603.20939v1
+
+## 2940. Two Approaches to Direct Estimation of Riesz Representers
+
+- **ID:** `2603.20936` · **primary:** `econ.EM` · **Submitted:** 2026-03-21
+- **Authors:** David Bruns-Smith
+- **abs:** https://arxiv.org/abs/2603.20936 · **PDF:** https://arxiv.org/pdf/2603.20936v1
+
+## 2941. MOELIGA: a multi-objective evolutionary approach for feature selection with local improvement
+
+- **ID:** `2603.20934` · **primary:** `cs.NE` · **Submitted:** 2026-03-21
+- **Authors:** Leandro Vignolo, Matias Gerard
+- **abs:** https://arxiv.org/abs/2603.20934 · **PDF:** https://arxiv.org/pdf/2603.20934v1
+
+## 2942. AC4A: Access Control for Agents
+
+- **ID:** `2603.20933` · **primary:** `cs.CR` · **Submitted:** 2026-03-21
+- **Authors:** Reshabh K Sharma, Dan Grossman
+- **abs:** https://arxiv.org/abs/2603.20933 · **PDF:** https://arxiv.org/pdf/2603.20933v1
+
+## 2943. Causally-Guided Diffusion for Stable Feature Selection
+
+- **ID:** `2603.20930` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Arun Vignesh Malarkkan, Xinyuan Wang, Kunpeng Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.20930 · **PDF:** https://arxiv.org/pdf/2603.20930v1
+
+## 2944. Beyond Static Visual Tokens: Structured Sequential Visual Chain-of-Thought Reasoning
+
+- **ID:** `2603.26737` · **primary:** `cs.CV` · **Submitted:** 2026-03-21
+- **Authors:** Guangfu Guo, Xiaoqian Lu, Yue Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.26737 · **PDF:** https://arxiv.org/pdf/2603.26737v1
+
+## 2945. Cloud-Edge Collaborative Large Models for Robust Photovoltaic Power Forecasting
+
+- **ID:** `2603.22343` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Nan Qiao, Shuning Wang, Sijing Duan, et al.
+- **abs:** https://arxiv.org/abs/2603.22343 · **PDF:** https://arxiv.org/pdf/2603.22343v2
+
+## 2946. Stability of Sequential and Parallel Coordinate Ascent Variational Inference
+
+- **ID:** `2603.20929` · **primary:** `stat.ML` · **Submitted:** 2026-03-21
+- **Authors:** Debdeep Pati
+- **abs:** https://arxiv.org/abs/2603.20929 · **PDF:** https://arxiv.org/pdf/2603.20929v1
+
+## 2947. Active Inference for Physical AI Agents -- An Engineering Perspective
+
+- **ID:** `2603.20927` · **primary:** `stat.ML` · **Submitted:** 2026-03-21
+- **Authors:** Bert de Vries
+- **abs:** https://arxiv.org/abs/2603.20927 · **PDF:** https://arxiv.org/pdf/2603.20927v1
+
+## 2948. Deep Adaptive Rate Allocation in Volatile Heterogeneous Wireless Networks
+
+- **ID:** `2603.20926` · **primary:** `cs.IT` · **Submitted:** 2026-03-21
+- **Authors:** Gregorio Maglione, Veselin Rakocevic, Markus Amend, et al.
+- **abs:** https://arxiv.org/abs/2603.20926 · **PDF:** https://arxiv.org/pdf/2603.20926v1
+
+## 2949. Profit is the Red Team: Stress-Testing Agents in Strategic Economic Interactions
+
+- **ID:** `2603.20925` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Shouqiao Wang, Marcello Politi, Samuele Marro, et al.
+- **abs:** https://arxiv.org/abs/2603.20925 · **PDF:** https://arxiv.org/pdf/2603.20925v1
+
+## 2950. Discriminative Representation Learning for Clinical Prediction
+
+- **ID:** `2603.20921` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Yang Zhang, Li Fan, Samuel Lawrence, et al.
+- **abs:** https://arxiv.org/abs/2603.20921 · **PDF:** https://arxiv.org/pdf/2603.20921v1
+
+## 2951. Democratizing AI: A Comparative Study in Deep Learning Efficiency and Future Trends in Computational Processing
+
+- **ID:** `2603.20920` · **primary:** `cs.PF` · **Submitted:** 2026-03-21
+- **Authors:** Lisan Al Amin, Md Ismail Hossain, Rupak Kumar Das, et al.
+- **abs:** https://arxiv.org/abs/2603.20920 · **PDF:** https://arxiv.org/pdf/2603.20920v2
+
+## 2952. Enhancing LIME using Neural Decision Trees
+
+- **ID:** `2603.20919` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Mohamed Aymen Bouyahia, Argyris Kalogeratos
+- **abs:** https://arxiv.org/abs/2603.20919 · **PDF:** https://arxiv.org/pdf/2603.20919v1
+
+## 2953. Implicit Turn-Wise Policy Optimization for Proactive User-LLM Interaction
+
+- **ID:** `2603.23550` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Haoyu Wang, Yuxin Chen, Liang Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.23550 · **PDF:** https://arxiv.org/pdf/2603.23550v1
+
+## 2954. Do LLM-Driven Agents Exhibit Engagement Mechanisms? Controlled Tests of Information Load, Descriptive Norms, and Popularity Cues
+
+- **ID:** `2603.20911` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Tai-Quan Peng, Yuan Tian, Songsong Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.20911 · **PDF:** https://arxiv.org/pdf/2603.20911v1
+
+## 2955. LLM-ODE: Data-driven Discovery of Dynamical Systems with Large Language Models
+
+- **ID:** `2603.20910` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Amirmohammad Ziaei Bideh, Jonathan Gryak
+- **abs:** https://arxiv.org/abs/2603.20910 · **PDF:** https://arxiv.org/pdf/2603.20910v1
+
+## 2956. Bayesian Scattering: A Principled Baseline for Uncertainty on Image Data
+
+- **ID:** `2603.20908` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Bernardo Fichera, Zarko Ivkovic, Kjell Jorner, et al.
+- **abs:** https://arxiv.org/abs/2603.20908 · **PDF:** https://arxiv.org/pdf/2603.20908v1
+
+## 2957. Neutrino Oscillation Parameter Estimation Using Structured Hierarchical Transformers
+
+- **ID:** `2603.22342` · **primary:** `hep-ph` · **Submitted:** 2026-03-21
+- **Authors:** Giorgio Morales, Gregory Lehaut, Antonin Vacheret, et al.
+- **abs:** https://arxiv.org/abs/2603.22342 · **PDF:** https://arxiv.org/pdf/2603.22342v1
+
+## 2958. The Hidden Puppet Master: Predicting Human Belief Change in Manipulative LLM Dialogues
+
+- **ID:** `2603.20907` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Jocelyn Shen, Amina Luvsanchultem, Jessica Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.20907 · **PDF:** https://arxiv.org/pdf/2603.20907v2
+
+## 2959. Sparse Weak-Form Discovery of Stochastic Generators
+
+- **ID:** `2603.20904` · **primary:** `stat.ME` · **Submitted:** 2026-03-21
+- **Authors:** Eshwar R A, Gajanan V. Honnavar
+- **abs:** https://arxiv.org/abs/2603.20904 · **PDF:** https://arxiv.org/pdf/2603.20904v3
+
+## 2960. Unfolding with a Wasserstein Loss
+
+- **ID:** `2603.20903` · **primary:** `math.OC` · **Submitted:** 2026-03-21
+- **Authors:** Katy Craig, Benjamin Faktor, Benjamin Nachman
+- **abs:** https://arxiv.org/abs/2603.20903 · **PDF:** https://arxiv.org/pdf/2603.20903v1
+
+## 2961. Mitigating Shortcut Reasoning in Language Models: A Gradient-Aware Training Approach
+
+- **ID:** `2603.20899` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Hongyu Cao, Kunpeng Liu, Dongjie Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.20899 · **PDF:** https://arxiv.org/pdf/2603.20899v1
+
+## 2962. Natural Gradient Descent for Online Continual Learning
+
+- **ID:** `2603.20898` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Joe Khawand, David Colliaux
+- **abs:** https://arxiv.org/abs/2603.20898 · **PDF:** https://arxiv.org/pdf/2603.20898v1
+
+## 2963. The data heat island effect: quantifying the impact of AI data centers in a warming world
+
+- **ID:** `2603.20897` · **primary:** `cs.CY` · **Submitted:** 2026-03-21
+- **Authors:** Andrea Marinoni, Erik Cambria, Luca Dal Zilio, et al.
+- **abs:** https://arxiv.org/abs/2603.20897 · **PDF:** https://arxiv.org/pdf/2603.20897v2
+
+## 2964. Beyond the Birkhoff Polytope: Spectral-Sphere-Constrained Hyper-Connections
+
+- **ID:** `2603.20896` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Zhaoyi Liu, Haichuan Zhang, Ang Li
+- **abs:** https://arxiv.org/abs/2603.20896 · **PDF:** https://arxiv.org/pdf/2603.20896v1
+
+## 2965. LLM Router: Rethinking Routing with Prefill Activations
+
+- **ID:** `2603.20895` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Tanay Varshney, Annie Surla, Michelle Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.20895 · **PDF:** https://arxiv.org/pdf/2603.20895v2
+
+## 2966. Auto-differentiable data assimilation: Co-learning of states, dynamics, and filtering algorithms
+
+- **ID:** `2603.20891` · **primary:** `stat.ML` · **Submitted:** 2026-03-21
+- **Authors:** Melissa Adrian, Daniel Sanz-Alonso, Rebecca Willett
+- **abs:** https://arxiv.org/abs/2603.20891 · **PDF:** https://arxiv.org/pdf/2603.20891v1
+
+## 2967. Characterizing the onset and offset of motor imagery during passive arm movements induced by an upper-body exoskeleton
+
+- **ID:** `2603.20885` · **primary:** `cs.RO` · **Submitted:** 2026-03-21
+- **Authors:** Kanishka Mitra, Frigyes Samuel Racz, Satyam Kumar, et al.
+- **abs:** https://arxiv.org/abs/2603.20885 · **PDF:** https://arxiv.org/pdf/2603.20885v1
+
+## 2968. NoveltyAgent: Autonomous Novelty Reporting Agent with Point-wise Novelty Analysis and Self-Validation
+
+- **ID:** `2603.20884` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Jiajun Hou, Hexuan Deng, Wenxiang Jiao, et al.
+- **abs:** https://arxiv.org/abs/2603.20884 · **PDF:** https://arxiv.org/pdf/2603.20884v1
+
+## 2969. RubricRAG: Towards Interpretable and Reliable LLM Evaluation via Domain Knowledge Retrieval for Rubric Generation
+
+- **ID:** `2603.20882` · **primary:** `cs.IR` · **Submitted:** 2026-03-21
+- **Authors:** Kaustubh D. Dhole, Eugene Agichtein
+- **abs:** https://arxiv.org/abs/2603.20882 · **PDF:** https://arxiv.org/pdf/2603.20882v1
+
+## 2970. Incentive-Aware Federated Averaging with Performance Guarantees under Strategic Participation
+
+- **ID:** `2603.20873` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Fateme Maleki, Krishnan Raghavan, Farzad Yousefian
+- **abs:** https://arxiv.org/abs/2603.20873 · **PDF:** https://arxiv.org/pdf/2603.20873v1
+
+## 2971. ReLaMix: Residual Latency-Aware Mixing for Delay-Robust Financial Time-Series Forecasting
+
+- **ID:** `2603.20869` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Tianyou Lai, Wentao Yue, Jiayi Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.20869 · **PDF:** https://arxiv.org/pdf/2603.20869v1
+
+## 2972. Ordinal Semantic Segmentation Applied to Medical and Odontological Images
+
+- **ID:** `2603.26736` · **primary:** `cs.CV` · **Submitted:** 2026-03-21
+- **Authors:** Mariana Dória Prata Lima, Gilson Antonio Giraldi, Jaime S. Cardoso
+- **abs:** https://arxiv.org/abs/2603.26736 · **PDF:** https://arxiv.org/pdf/2603.26736v1
+
+## 2973. Semantic Sections: An Atlas-Native Feature Ontology for Obstructed Representation Spaces
+
+- **ID:** `2603.20867` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Hossein Javidnia
+- **abs:** https://arxiv.org/abs/2603.20867 · **PDF:** https://arxiv.org/pdf/2603.20867v1
+
+## 2974. Universal Coefficients and Mayer-Vietoris for Moore Homology of Ample Groupoids
+
+- **ID:** `2603.20861` · **primary:** `math.AT` · **Submitted:** 2026-03-21
+- **Authors:** Luciano Melodia
+- **abs:** https://arxiv.org/abs/2603.20861 · **PDF:** https://arxiv.org/pdf/2603.20861v1
+
+## 2975. Restoring Neural Network Plasticity for Faster Transfer Learning
+
+- **ID:** `2603.20860` · **primary:** `cs.CV` · **Submitted:** 2026-03-21
+- **Authors:** Xander Coetzer, Arné Schreuder, Anna Sergeevna Bosman
+- **abs:** https://arxiv.org/abs/2603.20860 · **PDF:** https://arxiv.org/pdf/2603.20860v1
+
+## 2976. Ensemble of Small Classifiers For Imbalanced White Blood Cell Classification
+
+- **ID:** `2603.20856` · **primary:** `cs.CV` · **Submitted:** 2026-03-21
+- **Authors:** Siddharth Srivastava, Adam Smith, Scott Brooks, et al.
+- **abs:** https://arxiv.org/abs/2603.20856 · **PDF:** https://arxiv.org/pdf/2603.20856v1
+
+## 2977. SozKZ: Training Efficient Small Language Models for Kazakh from Scratch
+
+- **ID:** `2603.20854` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Saken Tukenov
+- **abs:** https://arxiv.org/abs/2603.20854 · **PDF:** https://arxiv.org/pdf/2603.20854v1
+
+## 2978. Can ChatGPT Really Understand Modern Chinese Poetry?
+
+- **ID:** `2603.20851` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Shanshan Wang, Derek F. Wong, Jingming Yao, et al.
+- **abs:** https://arxiv.org/abs/2603.20851 · **PDF:** https://arxiv.org/pdf/2603.20851v1
+
+## 2979. HiCI: Hierarchical Construction-Integration for Long-Context Attention
+
+- **ID:** `2603.20843` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Xiangyu Zeng, Qi Xu, Yunke Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.20843 · **PDF:** https://arxiv.org/pdf/2603.20843v1
+
+## 2980. A Knowledge-Informed Pretrained Model for Causal Discovery
+
+- **ID:** `2603.20842` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Wenbo Xu, Yue He, Yunhai Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.20842 · **PDF:** https://arxiv.org/pdf/2603.20842v1
+
+## 2981. Dodgersort: Uncertainty-Aware VLM-Guided Human-in-the-Loop Pairwise Ranking
+
+- **ID:** `2603.20839` · **primary:** `cs.CV` · **Submitted:** 2026-03-21
+- **Authors:** Yujin Park, Haejun Chung, Ikbeom Jang
+- **abs:** https://arxiv.org/abs/2603.20839 · **PDF:** https://arxiv.org/pdf/2603.20839v1
+
+## 2982. MERIT: Multi-domain Efficient RAW Image Translation
+
+- **ID:** `2603.20836` · **primary:** `cs.CV` · **Submitted:** 2026-03-21
+- **Authors:** Wenjun Huang, Shenghao Fu, Yian Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.20836 · **PDF:** https://arxiv.org/pdf/2603.20836v1
+
+## 2983. Governance-Aware Vector Subscriptions for Multi-Agent Knowledge Ecosystems
+
+- **ID:** `2603.20833` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Steven Johnson
+- **abs:** https://arxiv.org/abs/2603.20833 · **PDF:** https://arxiv.org/pdf/2603.20833v2
+
+## 2984. Beyond the Academic Monoculture: A Unified Framework and Industrial Perspective for Attributed Graph Clustering
+
+- **ID:** `2603.20829` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Yunhui Liu, Yue Liu, Yongchao Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.20829 · **PDF:** https://arxiv.org/pdf/2603.20829v1
+
+## 2985. Simple Projection-Free Algorithm for Contextual Recommendation with Logarithmic Regret and Robustness
+
+- **ID:** `2603.20826` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Shinsaku Sakaue
+- **abs:** https://arxiv.org/abs/2603.20826 · **PDF:** https://arxiv.org/pdf/2603.20826v2
+
+## 2986. Cross-Granularity Representations for Biological Sequences: Insights from ESM and BiGCARP
+
+- **ID:** `2603.20825` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Hanlin Xiao, Rainer Breitling, Eriko Takano, et al.
+- **abs:** https://arxiv.org/abs/2603.20825 · **PDF:** https://arxiv.org/pdf/2603.20825v1
+
+## 2987. Compass: Optimizing Compound AI Workflows for Dynamic Adaptation
+
+- **ID:** `2603.20821` · **primary:** `cs.DC` · **Submitted:** 2026-03-21
+- **Authors:** Milos Gravara, Juan Luis Herrera, Stefan Nastic
+- **abs:** https://arxiv.org/abs/2603.20821 · **PDF:** https://arxiv.org/pdf/2603.20821v1
+
+## 2988. Achieving $\widetilde{O}(1/ε)$ Sample Complexity for Bilinear Systems Identification under Bounded Noises
+
+- **ID:** `2603.20819` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Hongyu Yi, Chenbei Lu, Jing Yu
+- **abs:** https://arxiv.org/abs/2603.20819 · **PDF:** https://arxiv.org/pdf/2603.20819v1
+
+## 2989. PlanaReLoc: Camera Relocalization in 3D Planar Primitives via Region-Based Structure Matching
+
+- **ID:** `2603.20818` · **primary:** `cs.CV` · **Submitted:** 2026-03-21
+- **Authors:** Hanqiao Ye, Yuzhou Liu, Yangdong Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.20818 · **PDF:** https://arxiv.org/pdf/2603.20818v1
+
+## 2990. GMPilot: An Expert AI Agent For FDA cGMP Compliance
+
+- **ID:** `2603.20815` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Xiaohan Wang, Nan Zhang, Sulene Han, et al.
+- **abs:** https://arxiv.org/abs/2603.20815 · **PDF:** https://arxiv.org/pdf/2603.20815v1
+
+## 2991. Predictive Regularization Against Visual Representation Degradation in Multimodal Large Language Models
+
+- **ID:** `2603.20808` · **primary:** `cs.CV` · **Submitted:** 2026-03-21
+- **Authors:** Enguang Wang, Qiang Wang, Yuanchen Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.20808 · **PDF:** https://arxiv.org/pdf/2603.20808v1
+
+## 2992. BenchBench: Benchmarking Automated Benchmark Generation
+
+- **ID:** `2603.20807` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Yandan Zheng, Haoran Luo, Zhenghong Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.20807 · **PDF:** https://arxiv.org/pdf/2603.20807v1
+
+## 2993. Large Neighborhood Search meets Iterative Neural Constraint Heuristics
+
+- **ID:** `2603.20801` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Yudong W. Xu, Wenhao Li, Scott Sanner, et al.
+- **abs:** https://arxiv.org/abs/2603.20801 · **PDF:** https://arxiv.org/pdf/2603.20801v1
+
+## 2994. RLVR Training of LLMs Does Not Improve Thinking Ability for General QA: Evaluation Method and a Simple Solution
+
+- **ID:** `2603.20799` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Kaiyuan Li, Jing-Cheng Pang, Yang Yu
+- **abs:** https://arxiv.org/abs/2603.20799 · **PDF:** https://arxiv.org/pdf/2603.20799v1
+
+## 2995. The Anatomy of an Edit: Mechanism-Guided Activation Steering for Knowledge Editing
+
+- **ID:** `2603.20795` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Yuan Cao, Mingyang Wang, Hinrich Schütze
+- **abs:** https://arxiv.org/abs/2603.20795 · **PDF:** https://arxiv.org/pdf/2603.20795v1
+
+## 2996. T-MAP: Red-Teaming LLM Agents with Trajectory-aware Evolutionary Search
+
+- **ID:** `2603.22341` · **primary:** `cs.CR` · **Submitted:** 2026-03-21
+- **Authors:** Hyomin Lee, Sangwoo Park, Yumin Choi, et al.
+- **abs:** https://arxiv.org/abs/2603.22341 · **PDF:** https://arxiv.org/pdf/2603.22341v1
+
+## 2997. Neural Autoregressive Flows for Markov Boundary Learning
+
+- **ID:** `2603.20791` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Khoa Nguyen, Bao Duong, Viet Huynh, et al.
+- **abs:** https://arxiv.org/abs/2603.20791 · **PDF:** https://arxiv.org/pdf/2603.20791v1
+
+## 2998. Experimental Design for Missing Physics
+
+- **ID:** `2604.01231` · **primary:** `physics.comp-ph` · **Submitted:** 2026-03-21
+- **Authors:** Arno Strouwen, Sebastián Micluţa-Câmpeanu
+- **abs:** https://arxiv.org/abs/2604.01231 · **PDF:** https://arxiv.org/pdf/2604.01231v1
+
+## 2999. Code-MIE: A Code-style Model for Multimodal Information Extraction with Scene Graph and Entity Attribute Knowledge Enhancement
+
+- **ID:** `2603.20781` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Jiang Liu, Ge Qiu, Hao Fei, et al.
+- **abs:** https://arxiv.org/abs/2603.20781 · **PDF:** https://arxiv.org/pdf/2603.20781v1
+
+## 3000. OmniPatch: A Universal Adversarial Patch for ViT-CNN Cross-Architecture Transfer in Semantic Segmentation
+
+- **ID:** `2603.20777` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Aarush Aggarwal, Akshat Tomar, Amritanshu Tiwari, et al.
+- **abs:** https://arxiv.org/abs/2603.20777 · **PDF:** https://arxiv.org/pdf/2603.20777v1
+
+## 3001. Evaluating Uplift Modeling under Structural Biases: Insights into Metric Stability and Model Robustness
+
+- **ID:** `2603.20775` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Yuxuan Yang, Dugang Liu, Yiyan Huang
+- **abs:** https://arxiv.org/abs/2603.20775 · **PDF:** https://arxiv.org/pdf/2603.20775v1
+
+## 3002. Graphs RAG at Scale: Beyond Retrieval-Augmented Generation With Labeled Property Graphs and Resource Description Framework for Complex and Unknown Search Spaces
+
+- **ID:** `2603.22340` · **primary:** `cs.IR` · **Submitted:** 2026-03-21
+- **Authors:** Manie Tadayon, Mayank Gupta
+- **abs:** https://arxiv.org/abs/2603.22340 · **PDF:** https://arxiv.org/pdf/2603.22340v1
+
+## 3003. Memory-Efficient Fine-Tuning Diffusion Transformers via Dynamic Patch Sampling and Block Skipping
+
+- **ID:** `2603.20755` · **primary:** `cs.CV` · **Submitted:** 2026-03-21
+- **Authors:** Sunghyun Park, Jeongho Kim, Hyoungwoo Park, et al.
+- **abs:** https://arxiv.org/abs/2603.20755 · **PDF:** https://arxiv.org/pdf/2603.20755v1
+
+## 3004. Modeling Epistemic Uncertainty in Social Perception via Rashomon Set Agents
+
+- **ID:** `2603.20750` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Jinming Yang, Xinyu Jiang, Xinshan Jiao, et al.
+- **abs:** https://arxiv.org/abs/2603.20750 · **PDF:** https://arxiv.org/pdf/2603.20750v1
+
+## 3005. Adversarial Attacks on Locally Private Graph Neural Networks
+
+- **ID:** `2603.20746` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Matta Varun, Ajay Kumar Dhakar, Yuan Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.20746 · **PDF:** https://arxiv.org/pdf/2603.20746v1
+
+## 3006. Mitigating Temporal Blindness in Kubernetes Autoscaling: An Attention-Double-LSTM Framework
+
+- **ID:** `2603.28790` · **primary:** `cs.DC` · **Submitted:** 2026-03-21
+- **Authors:** Faraz Shaikh, Gianluca Reali, Mauro Femminella
+- **abs:** https://arxiv.org/abs/2603.28790 · **PDF:** https://arxiv.org/pdf/2603.28790v1
+
+## 3007. MzansiText and MzansiLM: An Open Corpus and Decoder-Only Language Model for South African Languages
+
+- **ID:** `2603.20732` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Anri Lombard, Simbarashe Mawere, Temi Aina, et al.
+- **abs:** https://arxiv.org/abs/2603.20732 · **PDF:** https://arxiv.org/pdf/2603.20732v1
+
+## 3008. Reasoning Topology Matters: Network-of-Thought for Complex Reasoning Tasks
+
+- **ID:** `2603.20730` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Fan Huang
+- **abs:** https://arxiv.org/abs/2603.20730 · **PDF:** https://arxiv.org/pdf/2603.20730v1
+
+## 3009. Weakly supervised multimodal segmentation of acoustic borehole images with depth-aware cross-attention
+
+- **ID:** `2603.20729` · **primary:** `cs.CV` · **Submitted:** 2026-03-21
+- **Authors:** Jose Luis Lima de Jesus Silva
+- **abs:** https://arxiv.org/abs/2603.20729 · **PDF:** https://arxiv.org/pdf/2603.20729v1
+
+## 3010. Problems with Chinchilla Approach 2: Systematic Biases in IsoFLOP Parabola Fits
+
+- **ID:** `2603.22339` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Eric Czech, Zhiwei Xu, Yael Elmatad, et al.
+- **abs:** https://arxiv.org/abs/2603.22339 · **PDF:** https://arxiv.org/pdf/2603.22339v3
+
+## 3011. Multi-RF Fusion with Multi-GNN Blending for Molecular Property Prediction
+
+- **ID:** `2603.20724` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Zacharie Bugaud
+- **abs:** https://arxiv.org/abs/2603.20724 · **PDF:** https://arxiv.org/pdf/2603.20724v1
+
+## 3012. Distilled Large Language Model-Driven Dynamic Sparse Expert Activation Mechanism
+
+- **ID:** `2603.26735` · **primary:** `cs.CV` · **Submitted:** 2026-03-21
+- **Authors:** Qinghui Chen, Zekai Zhang, Zaigui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.26735 · **PDF:** https://arxiv.org/pdf/2603.26735v1
+
+## 3013. RoboECC: Multi-Factor-Aware Edge-Cloud Collaborative Deployment for VLA Models
+
+- **ID:** `2603.20711` · **primary:** `cs.DC` · **Submitted:** 2026-03-21
+- **Authors:** Zihao Zheng, Hangyu Cao, Jiayu Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.20711 · **PDF:** https://arxiv.org/pdf/2603.20711v1
+
+## 3014. NDT: Non-Differential Transformer and Its Application to Sentiment Analysis
+
+- **ID:** `2603.20704` · **primary:** `cs.IR` · **Submitted:** 2026-03-21
+- **Authors:** Soudeep Ghoshal, Himanshu Buckchash, Sarita Paudel, et al.
+- **abs:** https://arxiv.org/abs/2603.20704 · **PDF:** https://arxiv.org/pdf/2603.20704v1
+
+## 3015. Decoupling Numerical and Structural Parameters: An Empirical Study on Adaptive Genetic Algorithms via Deep Reinforcement Learning for the Large-Scale TSP
+
+- **ID:** `2603.20702` · **primary:** `cs.NE` · **Submitted:** 2026-03-21
+- **Authors:** Hongyu Wang, Yuhan Jing, Yibing Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.20702 · **PDF:** https://arxiv.org/pdf/2603.20702v1
+
+## 3016. mmWave-Diffusion:A Novel Framework for Respiration Sensing Using Observation-Anchored Conditional Diffusion Model
+
+- **ID:** `2603.20700` · **primary:** `eess.IV` · **Submitted:** 2026-03-21
+- **Authors:** Yong Wang, Qifan Shen, Bao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.20700 · **PDF:** https://arxiv.org/pdf/2603.20700v1
+
+## 3017. Clinical Cognition Alignment for Gastrointestinal Diagnosis with Multimodal LLMs
+
+- **ID:** `2603.20698` · **primary:** `cs.CV` · **Submitted:** 2026-03-21
+- **Authors:** Huan Zheng, Yucheng Zhou, Tianyi Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.20698 · **PDF:** https://arxiv.org/pdf/2603.20698v1
+
+## 3018. Satellite-to-Street: Synthesizing Post-Disaster Views from Satellite Imagery via Generative Vision Models
+
+- **ID:** `2603.20697` · **primary:** `cs.CV` · **Submitted:** 2026-03-21
+- **Authors:** Yifan Yang, Lei Zou, Wendy Jepson
+- **abs:** https://arxiv.org/abs/2603.20697 · **PDF:** https://arxiv.org/pdf/2603.20697v1
+
+## 3019. High-dimensional online learning via asynchronous decomposition: Non-divergent results, dynamic regularization, and beyond
+
+- **ID:** `2603.20696` · **primary:** `stat.ML` · **Submitted:** 2026-03-21
+- **Authors:** Shixiang Liu, Zhifan Li, Hanming Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.20696 · **PDF:** https://arxiv.org/pdf/2603.20696v1
+
+## 3020. Can I guess where you are from? Modeling dialectal morphosyntactic similarities in Brazilian Portuguese
+
+- **ID:** `2603.20695` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Manoel Siqueira, Raquel Freitag
+- **abs:** https://arxiv.org/abs/2603.20695 · **PDF:** https://arxiv.org/pdf/2603.20695v1
+
+## 3021. SWE-Next: Scalable Real-World Software Engineering Tasks for Agents
+
+- **ID:** `2603.20691` · **primary:** `cs.SE` · **Submitted:** 2026-03-21
+- **Authors:** Jiarong Liang, Zhiheng Lyu, Zijie Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.20691 · **PDF:** https://arxiv.org/pdf/2603.20691v1
+
+## 3022. Artificial Intelligence in Experimental Approaches: Growth Hacking, Lean Startup, Design Thinking, and Agile
+
+- **ID:** `2603.20688` · **primary:** `cs.CY` · **Submitted:** 2026-03-21
+- **Authors:** Parisa Omidmand, Saeid Ataei
+- **abs:** https://arxiv.org/abs/2603.20688 · **PDF:** https://arxiv.org/pdf/2603.20688v1
+
+## 3023. Neuronal Self-Adaptation Enhances Capacity and Robustness of Representation in Spiking Neural Networks
+
+- **ID:** `2603.20687` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Zhuobin Yang, Yeyao Bao, Liangfu Lv, et al.
+- **abs:** https://arxiv.org/abs/2603.20687 · **PDF:** https://arxiv.org/pdf/2603.20687v1
+
+## 3024. SNAP: Speaker Nulling for Artifact Projection in Speech Deepfake Detection
+
+- **ID:** `2603.20686` · **primary:** `cs.SD` · **Submitted:** 2026-03-21
+- **Authors:** Kyudan Jung, Jihwan Kim, Minwoo Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.20686 · **PDF:** https://arxiv.org/pdf/2603.20686v1
+
+## 3025. Interpretable Battery Aging without Extra Tests via Neural-Assisted Physics-based Modelling
+
+- **ID:** `2604.01229` · **primary:** `eess.SP` · **Submitted:** 2026-03-21
+- **Authors:** Yuan Qiu, Wei Li, Wei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2604.01229 · **PDF:** https://arxiv.org/pdf/2604.01229v1
+
+## 3026. Centrality-Based Pruning for Efficient Echo State Networks
+
+- **ID:** `2603.20684` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Sudip Laudari
+- **abs:** https://arxiv.org/abs/2603.20684 · **PDF:** https://arxiv.org/pdf/2603.20684v1
+
+## 3027. Hierarchical Multiscale Structure-Function Coupling for Brain Connectome Integration
+
+- **ID:** `2603.20680` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-21
+- **Authors:** Jianwei Chen, Zhengyang Miao, Wenjie Cai, et al.
+- **abs:** https://arxiv.org/abs/2603.20680 · **PDF:** https://arxiv.org/pdf/2603.20680v1
+
+## 3028. AI-Driven Multi-Agent Simulation of Stratified Polyamory Systems: A Computational Framework for Optimizing Social Reproductive Efficiency
+
+- **ID:** `2603.20678` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Yicai Xing
+- **abs:** https://arxiv.org/abs/2603.20678 · **PDF:** https://arxiv.org/pdf/2603.20678v1
+
+## 3029. PAVE: Premise-Aware Validation and Editing for Retrieval-Augmented LLMs
+
+- **ID:** `2603.20673` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Tianyi Huang, Caden Yang, Emily Yin, et al.
+- **abs:** https://arxiv.org/abs/2603.20673 · **PDF:** https://arxiv.org/pdf/2603.20673v2
+
+## 3030. Breaking the $O(\sqrt{T})$ Cumulative Constraint Violation Barrier while Achieving $O(\sqrt{T})$ Static Regret in Constrained Online Convex Optimization
+
+- **ID:** `2603.20671` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Haricharan Balasundaram, Karthick Krishna Mahendran, Rahul Vaze
+- **abs:** https://arxiv.org/abs/2603.20671 · **PDF:** https://arxiv.org/pdf/2603.20671v1
+
+## 3031. Towards Intelligent Geospatial Data Discovery: a knowledge graph-driven multi-agent framework powered by large language models
+
+- **ID:** `2603.20670` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Ruixiang Liu, Zhenlong Li, Ali Khosravi Kazazi
+- **abs:** https://arxiv.org/abs/2603.20670 · **PDF:** https://arxiv.org/pdf/2603.20670v2
+
+## 3032. REVERE: Reflective Evolving Research Engineer for Scientific Workflows
+
+- **ID:** `2603.20667` · **primary:** `cs.SE` · **Submitted:** 2026-03-21
+- **Authors:** Balaji Dinesh Gangireddi, Aniketh Garikaparthi, Manasi Patwardhan, et al.
+- **abs:** https://arxiv.org/abs/2603.20667 · **PDF:** https://arxiv.org/pdf/2603.20667v1
+
+## 3033. Attention in Space: Functional Roles of VLM Heads for Spatial Reasoning
+
+- **ID:** `2603.20662` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Xueqi Ma, Shuo Yang, Yanbei Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.20662 · **PDF:** https://arxiv.org/pdf/2603.20662v1
+
+## 3034. Sinkhorn Based Associative Memory Retrieval Using Spherical Hellinger Kantorovich Dynamics
+
+- **ID:** `2603.20656` · **primary:** `stat.ML` · **Submitted:** 2026-03-21
+- **Authors:** Aratrika Mustafi, Soumya Mukherjee
+- **abs:** https://arxiv.org/abs/2603.20656 · **PDF:** https://arxiv.org/pdf/2603.20656v1
+
+## 3035. Exponential Family Discriminant Analysis: Generalizing LDA-Style Generative Classification to Non-Gaussian Models
+
+- **ID:** `2603.20655` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Anish Lakkapragada
+- **abs:** https://arxiv.org/abs/2603.20655 · **PDF:** https://arxiv.org/pdf/2603.20655v2
+
+## 3036. Modernizing Amdahl's Law: How AI Scaling Laws Shape Computer Architecture
+
+- **ID:** `2603.20654` · **primary:** `cs.DC` · **Submitted:** 2026-03-21
+- **Authors:** Chien-Ping Lu
+- **abs:** https://arxiv.org/abs/2603.20654 · **PDF:** https://arxiv.org/pdf/2603.20654v3
+
+## 3037. From 50% to Mastery in 3 Days: A Low-Resource SOP for Localizing Graduate-Level AI Tutors via Shadow-RAG
+
+- **ID:** `2603.20650` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Zonglin Yang, J. -H. Xie, Lining Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.20650 · **PDF:** https://arxiv.org/pdf/2603.20650v1
+
+## 3038. A Multihead Continual Learning Framework for Fine-Grained Fashion Image Retrieval with Contrastive Learning and Exponential Moving Average Distillation
+
+- **ID:** `2603.20648` · **primary:** `cs.CV` · **Submitted:** 2026-03-21
+- **Authors:** Ling Xiao, Toshihiko Yamasaki
+- **abs:** https://arxiv.org/abs/2603.20648 · **PDF:** https://arxiv.org/pdf/2603.20648v1
+
+## 3039. Diffusion Model for Manifold Data: Score Decomposition, Curvature, and Statistical Complexity
+
+- **ID:** `2603.20645` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Zixuan Zhang, Kaixuan Huang, Tuo Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.20645 · **PDF:** https://arxiv.org/pdf/2603.20645v1
+
+## 3040. Weber's Law in Transformer Magnitude Representations: Efficient Coding, Representational Geometry, and Psychophysical Laws in Language Models
+
+- **ID:** `2603.20642` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Jon-Paul Cacioli
+- **abs:** https://arxiv.org/abs/2603.20642 · **PDF:** https://arxiv.org/pdf/2603.20642v1
+
+## 3041. Hear Both Sides: Efficient Multi-Agent Debate via Diversity-Aware Message Retention
+
+- **ID:** `2603.20640` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Manh Nguyen, Anh Nguyen, Dung Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.20640 · **PDF:** https://arxiv.org/pdf/2603.20640v1
+
+## 3042. Agentic AI and the next intelligence explosion
+
+- **ID:** `2603.20639` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** James Evans, Benjamin Bratton, Blaise Agüera y Arcas
+- **abs:** https://arxiv.org/abs/2603.20639 · **PDF:** https://arxiv.org/pdf/2603.20639v1
+
+## 3043. AEGIS: From Clues to Verdicts -- Graph-Guided Deep Vulnerability Reasoning via Dialectics and Meta-Auditing
+
+- **ID:** `2603.20637` · **primary:** `cs.SE` · **Submitted:** 2026-03-21
+- **Authors:** Sen Fang, Weiyuan Ding, Zhezhen Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.20637 · **PDF:** https://arxiv.org/pdf/2603.20637v1
+
+## 3044. A Modular LLM Framework for Explainable Price Outlier Detection
+
+- **ID:** `2603.20636` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Shadi Sartipi, John Wu, Sina Ghotbi, et al.
+- **abs:** https://arxiv.org/abs/2603.20636 · **PDF:** https://arxiv.org/pdf/2603.20636v1
+
+## 3045. CFNN: Continued Fraction Neural Network
+
+- **ID:** `2603.20634` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Chao Wang, Xuancheng Zhou, Ruilin Hou, et al.
+- **abs:** https://arxiv.org/abs/2603.20634 · **PDF:** https://arxiv.org/pdf/2603.20634v1
+
+## 3046. Seed1.8 Model Card: Towards Generalized Real-World Agency
+
+- **ID:** `2603.20633` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Bytedance Seed
+- **abs:** https://arxiv.org/abs/2603.20633 · **PDF:** https://arxiv.org/pdf/2603.20633v2
+
+## 3047. Optimal low-rank stochastic gradient estimation for LLM training
+
+- **ID:** `2603.20632` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Zehao Li, Tao Ren, Zishi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.20632 · **PDF:** https://arxiv.org/pdf/2603.20632v1
+
+## 3048. LassoFlexNet: Flexible Neural Architecture for Tabular Data
+
+- **ID:** `2603.20631` · **primary:** `stat.ML` · **Submitted:** 2026-03-21
+- **Authors:** Kry Yik Chau Lui, Cheng Chi, Kishore Basu, et al.
+- **abs:** https://arxiv.org/abs/2603.20631 · **PDF:** https://arxiv.org/pdf/2603.20631v1
+
+## 3049. Causal Direct Preference Optimization for Distributionally Robust Generative Recommendation
+
+- **ID:** `2603.22335` · **primary:** `cs.IR` · **Submitted:** 2026-03-21
+- **Authors:** Chu Zhao, Enneng Yang, Jianzhe Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.22335 · **PDF:** https://arxiv.org/pdf/2603.22335v1
+
+## 3050. Reasoning Traces Shape Outputs but Models Won't Say So
+
+- **ID:** `2603.20620` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Yijie Hao, Lingjie Chen, Ali Emami, et al.
+- **abs:** https://arxiv.org/abs/2603.20620 · **PDF:** https://arxiv.org/pdf/2603.20620v1
+
+## 3051. Where can AI be used? Insights from a deep ontology of work activities
+
+- **ID:** `2603.20619` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Alice Cai, Iman YeckehZaare, Shuo Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.20619 · **PDF:** https://arxiv.org/pdf/2603.20619v1
+
+## 3052. Beyond Token Eviction: Mixed-Dimension Budget Allocation for Efficient KV Cache Compression
+
+- **ID:** `2603.20616` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Ruijie Miao, Zhiming Wang, Wang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.20616 · **PDF:** https://arxiv.org/pdf/2603.20616v1
+
+## 3053. Towards Practical World Model-based Reinforcement Learning for Vision-Language-Action Models
+
+- **ID:** `2603.20607` · **primary:** `cs.RO` · **Submitted:** 2026-03-21
+- **Authors:** Zhilong Zhang, Haoxiang Ren, Yihao Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.20607 · **PDF:** https://arxiv.org/pdf/2603.20607v1
+
+## 3054. Bayesian Learning in Episodic Zero-Sum Games
+
+- **ID:** `2603.20604` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Chang-Wei Yueh, Andy Zhao, Ashutosh Nayyar, et al.
+- **abs:** https://arxiv.org/abs/2603.20604 · **PDF:** https://arxiv.org/pdf/2603.20604v1
+
+## 3055. Interpretable Operator Learning for Inverse Problems via Adaptive Spectral Filtering: Convergence and Discretization Invariance
+
+- **ID:** `2603.20602` · **primary:** `stat.ML` · **Submitted:** 2026-03-21
+- **Authors:** Hang-Cheng Dong, Pengcheng Cheng, Shuhuan Li
+- **abs:** https://arxiv.org/abs/2603.20602 · **PDF:** https://arxiv.org/pdf/2603.20602v1
+
+## 3056. Graph-based data-driven discovery of interpretable laws governing corona-induced noise and radio interference for high-voltage transmission lines
+
+- **ID:** `2603.20600` · **primary:** `cs.SC` · **Submitted:** 2026-03-21
+- **Authors:** Hao Xu, Yuntian Chen, Chongqing Kang, et al.
+- **abs:** https://arxiv.org/abs/2603.20600 · **PDF:** https://arxiv.org/pdf/2603.20600v1
+
+## 3057. Position: Multi-Agent Algorithmic Care Systems Demand Contestability for Trustworthy AI
+
+- **ID:** `2603.20595` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Truong Thanh Hung Nguyen, Hélène Fournier, Piper Jackson, et al.
+- **abs:** https://arxiv.org/abs/2603.20595 · **PDF:** https://arxiv.org/pdf/2603.20595v1
+
+## 3058. Generating from Discrete Distributions Using Diffusions: Insights from Random Constraint Satisfaction Problems
+
+- **ID:** `2603.20589` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Alankrita Bhatt, Mukur Gupta, Germain Kolossov, et al.
+- **abs:** https://arxiv.org/abs/2603.20589 · **PDF:** https://arxiv.org/pdf/2603.20589v1
+
+## 3059. Neural collapse in the orthoplex regime
+
+- **ID:** `2603.20587` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** James Alcala, Rayna Andreeva, Vladimir A. Kobzar, et al.
+- **abs:** https://arxiv.org/abs/2603.20587 · **PDF:** https://arxiv.org/pdf/2603.20587v1
+
+## 3060. MKA: Memory-Keyed Attention for Efficient Long-Context Reasoning
+
+- **ID:** `2603.20586` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Dong Liu, Yanxuan Yu, Ben Lengerich, et al.
+- **abs:** https://arxiv.org/abs/2603.20586 · **PDF:** https://arxiv.org/pdf/2603.20586v2
+
+## 3061. RECLAIM: Cyclic Causal Discovery Amid Measurement Noise
+
+- **ID:** `2603.20585` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Muralikrishnna G. Sethuraman, Faramarz Fekri
+- **abs:** https://arxiv.org/abs/2603.20585 · **PDF:** https://arxiv.org/pdf/2603.20585v1
+
+## 3062. Generative Diffusion Model for Risk-Neutral Derivative Pricing
+
+- **ID:** `2603.20582` · **primary:** `q-fin.MF` · **Submitted:** 2026-03-21
+- **Authors:** Nilay Tiwari
+- **abs:** https://arxiv.org/abs/2603.20582 · **PDF:** https://arxiv.org/pdf/2603.20582v1
+
+## 3063. JUBAKU: An Adversarial Benchmark for Exposing Culturally Grounded Stereotypes in Japanese LLMs
+
+- **ID:** `2603.20581` · **primary:** `cs.CL` · **Submitted:** 2026-03-21
+- **Authors:** Taihei Shiotani, Masahiro Kaneko, Ayana Niwa, et al.
+- **abs:** https://arxiv.org/abs/2603.20581 · **PDF:** https://arxiv.org/pdf/2603.20581v2
+
+## 3064. Context Cartography: Toward Structured Governance of Contextual Space in Large Language Model Systems
+
+- **ID:** `2603.20578` · **primary:** `cs.AI` · **Submitted:** 2026-03-21
+- **Authors:** Zihua Wu, Georg Gartner
+- **abs:** https://arxiv.org/abs/2603.20578 · **PDF:** https://arxiv.org/pdf/2603.20578v1
+
+## 3065. LJ-Bench: Ontology-Based Benchmark for U.S. Crime
+
+- **ID:** `2603.20572` · **primary:** `cs.LG` · **Submitted:** 2026-03-21
+- **Authors:** Hung Yun Tseng, Wuzhen Li, Blerina Gkotse, et al.
+- **abs:** https://arxiv.org/abs/2603.20572 · **PDF:** https://arxiv.org/pdf/2603.20572v1
+
+## 3066. Graph Signal Processing Meets Mamba2: Adaptive Filter Bank via Delta Modulation
+
+- **ID:** `2603.22333` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Yehjin Shin, Seojin Kim, Noseong Park
+- **abs:** https://arxiv.org/abs/2603.22333 · **PDF:** https://arxiv.org/pdf/2603.22333v1
+
+## 3067. Permutation-Consensus Listwise Judging for Robust Factuality Evaluation
+
+- **ID:** `2603.20562` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Tianyi Huang, Nathan Huang, Justin Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.20562 · **PDF:** https://arxiv.org/pdf/2603.20562v1
+
+## 3068. Large Language Models for Missing Data Imputation: Understanding Behavior, Hallucination Effects, and Control Mechanisms
+
+- **ID:** `2603.22332` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Arthur Dantas Mangussi, Ricardo Cardoso Pereira, Ana Carolina Lorena, et al.
+- **abs:** https://arxiv.org/abs/2603.22332 · **PDF:** https://arxiv.org/pdf/2603.22332v1
+
+## 3069. Understanding Behavior Cloning with Action Quantization
+
+- **ID:** `2603.20538` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Haoqun Cao, Tengyang Xie
+- **abs:** https://arxiv.org/abs/2603.20538 · **PDF:** https://arxiv.org/pdf/2603.20538v1
+
+## 3070. LLM-Driven Heuristic Synthesis for Industrial Process Control: Lessons from Hot Steel Rolling
+
+- **ID:** `2603.20537` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Nima H. Siboni, Seyedreza Kiamousavi, Emad Scharifi
+- **abs:** https://arxiv.org/abs/2603.20537 · **PDF:** https://arxiv.org/pdf/2603.20537v1
+
+## 3071. Towards Practical Multimodal Hospital Outbreak Detection
+
+- **ID:** `2603.20536` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Chang Liu, Jieshi Chen, Alexander J. Sundermann, et al.
+- **abs:** https://arxiv.org/abs/2603.20536 · **PDF:** https://arxiv.org/pdf/2603.20536v1
+
+## 3072. An Industrial-Scale Retrieval-Augmented Generation Framework for Requirements Engineering: Empirical Evaluation with Automotive Manufacturing Data
+
+- **ID:** `2603.20534` · **primary:** `cs.SE` · **Submitted:** 2026-03-20
+- **Authors:** Muhammad Khalid, Yilmaz Uygun
+- **abs:** https://arxiv.org/abs/2603.20534 · **PDF:** https://arxiv.org/pdf/2603.20534v2
+
+## 3073. Revenue-Sharing as Infrastructure: A Distributed Business Model for Generative AI Platforms
+
+- **ID:** `2603.20533` · **primary:** `cs.CY` · **Submitted:** 2026-03-20
+- **Authors:** Ghislain Dorian Tchuente Mondjo
+- **abs:** https://arxiv.org/abs/2603.20533 · **PDF:** https://arxiv.org/pdf/2603.20533v1
+
+## 3074. Epistemic Observability in Language Models
+
+- **ID:** `2603.20531` · **primary:** `cs.DC` · **Submitted:** 2026-03-20
+- **Authors:** Tony Mason
+- **abs:** https://arxiv.org/abs/2603.20531 · **PDF:** https://arxiv.org/pdf/2603.20531v1
+
+## 3075. RMNP: Row-Momentum Normalized Preconditioning for Scalable Matrix-Based Optimization
+
+- **ID:** `2603.20527` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Shenyang Deng, Zhuoli Ouyang, Tianyu Pang, et al.
+- **abs:** https://arxiv.org/abs/2603.20527 · **PDF:** https://arxiv.org/pdf/2603.20527v2
+
+## 3076. Does This Gradient Spark Joy?
+
+- **ID:** `2603.20526` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Ian Osband
+- **abs:** https://arxiv.org/abs/2603.20526 · **PDF:** https://arxiv.org/pdf/2603.20526v1
+
+## 3077. Delightful Distributed Policy Gradient
+
+- **ID:** `2603.20521` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Ian Osband
+- **abs:** https://arxiv.org/abs/2603.20521 · **PDF:** https://arxiv.org/pdf/2603.20521v1
+
+## 3078. CogFormer: Learn All Your Models Once
+
+- **ID:** `2603.20520` · **primary:** `stat.ML` · **Submitted:** 2026-03-20
+- **Authors:** Jerry M. Huang, Lukas Schumacher, Niek Stevenson, et al.
+- **abs:** https://arxiv.org/abs/2603.20520 · **PDF:** https://arxiv.org/pdf/2603.20520v1
+
+## 3079. Evaluating Large Language Models on Historical Health Crisis Knowledge in Resource-Limited Settings: A Hybrid Multi-Metric Study
+
+- **ID:** `2603.20514` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Mohammed Rakibul Hasan
+- **abs:** https://arxiv.org/abs/2603.20514 · **PDF:** https://arxiv.org/pdf/2603.20514v1
+
+## 3080. ReBOL: Retrieval via Bayesian Optimization with Batched LLM Relevance Observations and Query Reformulation
+
+- **ID:** `2603.20513` · **primary:** `cs.IR` · **Submitted:** 2026-03-20
+- **Authors:** Anton Korikov, Scott Sanner
+- **abs:** https://arxiv.org/abs/2603.20513 · **PDF:** https://arxiv.org/pdf/2603.20513v1
+
+## 3081. Grounded Chess Reasoning in Language Models via Master Distillation
+
+- **ID:** `2603.20510` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Zhenwei Tang, Qianfeng Wen, Seth Grief-Albert, et al.
+- **abs:** https://arxiv.org/abs/2603.20510 · **PDF:** https://arxiv.org/pdf/2603.20510v1
+
+## 3082. Measuring Reasoning Trace Legibility: Can Those Who Understand Teach?
+
+- **ID:** `2603.20508` · **primary:** `cs.MA` · **Submitted:** 2026-03-20
+- **Authors:** Dani Roytburg, Shreya Sridhar, Daphne Ippolito
+- **abs:** https://arxiv.org/abs/2603.20508 · **PDF:** https://arxiv.org/pdf/2603.20508v1
+
+## 3083. Distributed Gradient Clustering: Convergence and the Effect of Initialization
+
+- **ID:** `2603.20507` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Aleksandar Armacki, Himkant Sharma, Dragana Bajović, et al.
+- **abs:** https://arxiv.org/abs/2603.20507 · **PDF:** https://arxiv.org/pdf/2603.20507v1
+
+## 3084. Efficient Counterfactual Reasoning in ProbLog via Single World Intervention Programs
+
+- **ID:** `2603.20505` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Saimun Habib, Vaishak Belle, Fengxiang He
+- **abs:** https://arxiv.org/abs/2603.20505 · **PDF:** https://arxiv.org/pdf/2603.20505v1
+
+## 3085. Meeting in the Middle: A Co-Design Paradigm for FHE and AI Inference
+
+- **ID:** `2603.20504` · **primary:** `cs.CR` · **Submitted:** 2026-03-20
+- **Authors:** Bernardo Magri, Benjamin Marsh, Paul Gebheim
+- **abs:** https://arxiv.org/abs/2603.20504 · **PDF:** https://arxiv.org/pdf/2603.20504v1
+
+## 3086. Conformal Risk Control for Safety-Critical Wildfire Evacuation Mapping: A Comparative Study of Tabular, Spatial, and Graph-Based Models
+
+- **ID:** `2603.22331` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Baljinnyam Dayan
+- **abs:** https://arxiv.org/abs/2603.22331 · **PDF:** https://arxiv.org/pdf/2603.22331v1
+
+## 3087. PARHAF, a human-authored corpus of clinical reports for fictitious patients in French
+
+- **ID:** `2603.20494` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Xavier Tannier, Salam Abbara, Rémi Flicoteaux, et al.
+- **abs:** https://arxiv.org/abs/2603.20494 · **PDF:** https://arxiv.org/pdf/2603.20494v1
+
+## 3088. AE-LLM: Adaptive Efficiency Optimization for Large Language Models
+
+- **ID:** `2603.20492` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Kaito Tanaka, Masato Ito, Yuji Nishimura, et al.
+- **abs:** https://arxiv.org/abs/2603.20492 · **PDF:** https://arxiv.org/pdf/2603.20492v1
+
+## 3089. Spatio-Temporal Grid Intelligence: A Hybrid Graph Neural Network and LSTM Framework for Robust Electricity Theft Detection
+
+- **ID:** `2603.20488` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Adewale U. Oguntola, Olowookere A. AbdulQoyum, Adebukola M. Madehin, et al.
+- **abs:** https://arxiv.org/abs/2603.20488 · **PDF:** https://arxiv.org/pdf/2603.20488v1
+
+## 3090. Profiling learners' affective engagement: Emotion AI, intercultural pragmatics, and language learning
+
+- **ID:** `2603.20479` · **primary:** `cs.CY` · **Submitted:** 2026-03-20
+- **Authors:** Robert Godwin-Jones
+- **abs:** https://arxiv.org/abs/2603.20479 · **PDF:** https://arxiv.org/pdf/2603.20479v1
+
+## 3091. Fair splits flip the leaderboard: CHANRG reveals limited generalization in RNA secondary-structure prediction
+
+- **ID:** `2603.22330` · **primary:** `q-bio.BM` · **Submitted:** 2026-03-20
+- **Authors:** Zhiyuan Chen, Zhenfeng Deng, Pan Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.22330 · **PDF:** https://arxiv.org/pdf/2603.22330v1
+
+## 3092. From Data to Laws: Neural Discovery of Conservation Laws Without False Positives
+
+- **ID:** `2603.20474` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Rahul D Ray
+- **abs:** https://arxiv.org/abs/2603.20474 · **PDF:** https://arxiv.org/pdf/2603.20474v1
+
+## 3093. DiffGraph: An Automated Agent-driven Model Merging Framework for In-the-Wild Text-to-Image Generation
+
+- **ID:** `2603.20470` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Zhuoling Li, Hossein Rahmani, Jiarui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.20470 · **PDF:** https://arxiv.org/pdf/2603.20470v1
+
+## 3094. Goal-oriented learning of stochastic dynamical systems using error bounds on path-space observables
+
+- **ID:** `2603.20467` · **primary:** `stat.ME` · **Submitted:** 2026-03-20
+- **Authors:** Joanna Zou, Han Cheng Lie, Youssef Marzouk
+- **abs:** https://arxiv.org/abs/2603.20467 · **PDF:** https://arxiv.org/pdf/2603.20467v1
+
+## 3095. Diffutron: A Masked Diffusion Language Model for Turkish Language
+
+- **ID:** `2603.20466` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Şuayp Talha Kocabay, Talha Rüzgar Akkuş
+- **abs:** https://arxiv.org/abs/2603.20466 · **PDF:** https://arxiv.org/pdf/2603.20466v1
+
+## 3096. Double Machine Learning for Static Panel Data with Instrumental Variables: New Method and Applications
+
+- **ID:** `2603.20464` · **primary:** `econ.EM` · **Submitted:** 2026-03-20
+- **Authors:** Anna Baiardi, Paul S. Clarke, Andrea A. Naghi, et al.
+- **abs:** https://arxiv.org/abs/2603.20464 · **PDF:** https://arxiv.org/pdf/2603.20464v1
+
+## 3097. Shift-Invariant Feature Attribution in the Application of Wireless Electrocardiograms
+
+- **ID:** `2603.20462` · **primary:** `eess.SP` · **Submitted:** 2026-03-20
+- **Authors:** Yalemzerf Getnet, Abiy Tasissa, Waltenegus Dargie
+- **abs:** https://arxiv.org/abs/2603.20462 · **PDF:** https://arxiv.org/pdf/2603.20462v1
+
+## 3098. Regret Bounds for Reinforcement Learning from Multi-Source Imperfect Preferences
+
+- **ID:** `2603.20453` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Ming Shi, Yingbin Liang, Ness B. Shroff, et al.
+- **abs:** https://arxiv.org/abs/2603.20453 · **PDF:** https://arxiv.org/pdf/2603.20453v2
+
+## 3099. SDE-Driven Spatio-Temporal Hypergraph Neural Networks for Irregular Longitudinal fMRI Connectome Modeling in Alzheimer's Disease
+
+- **ID:** `2603.20452` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Ruiying Chen, Yutong Wang, Houliang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.20452 · **PDF:** https://arxiv.org/pdf/2603.20452v1
+
+## 3100. Policies Permitting LLM Use for Polishing Peer Reviews Are Currently Not Enforceable
+
+- **ID:** `2603.20450` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Rounak Saha, Gurusha Juneja, Dayita Chaudhuri, et al.
+- **abs:** https://arxiv.org/abs/2603.20450 · **PDF:** https://arxiv.org/pdf/2603.20450v1
+
+## 3101. Solver-Aided Verification of Policy Compliance in Tool-Augmented LLM Agents
+
+- **ID:** `2603.20449` · **primary:** `cs.SE` · **Submitted:** 2026-03-20
+- **Authors:** Cailin Winston, Claris Winston, René Just
+- **abs:** https://arxiv.org/abs/2603.20449 · **PDF:** https://arxiv.org/pdf/2603.20449v1
+
+## 3102. Detecting Neurovascular Instability from Multimodal Physiological Signals Using Wearable-Compatible Edge AI: A Responsible Computational Framework
+
+- **ID:** `2603.20442` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Truong Quynh Hoa, Hoang Dinh Cuong, Truong Xuan Khanh
+- **abs:** https://arxiv.org/abs/2603.20442 · **PDF:** https://arxiv.org/pdf/2603.20442v1
+
+## 3103. A Training-Free Regeneration Paradigm: Contrastive Reflection Memory Guided Self-Verification and Self-Improvement
+
+- **ID:** `2603.20441` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Yuran Li, Di Wu, Benoit Boulet
+- **abs:** https://arxiv.org/abs/2603.20441 · **PDF:** https://arxiv.org/pdf/2603.20441v1
+
+## 3104. Trained Persistent Memory for Frozen Decoder-Only LLMs
+
+- **ID:** `2603.22329` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Hong Jeong
+- **abs:** https://arxiv.org/abs/2603.22329 · **PDF:** https://arxiv.org/pdf/2603.22329v1
+
+## 3105. Deep reflective reasoning in interdependence constrained structured data extraction from clinical notes for digital health
+
+- **ID:** `2603.20435` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Jingwei Huang, Kuroush Nezafati, Zhikai Chi, et al.
+- **abs:** https://arxiv.org/abs/2603.20435 · **PDF:** https://arxiv.org/pdf/2603.20435v1
+
+## 3106. Verifiable Error Bounds for Physics-Informed Neural KKL Observers
+
+- **ID:** `2603.20434` · **primary:** `eess.SY` · **Submitted:** 2026-03-20
+- **Authors:** Hannah Berin-Costain, Harry Wang, Kirsten Morris, et al.
+- **abs:** https://arxiv.org/abs/2603.20434 · **PDF:** https://arxiv.org/pdf/2603.20434v1
+
+## 3107. ALICE: A Multifaceted Evaluation Framework of Large Audio-Language Models' In-Context Learning Ability
+
+- **ID:** `2603.20433` · **primary:** `cs.SD` · **Submitted:** 2026-03-20
+- **Authors:** Yen-Ting Piao, Jay Chiehen Liao, Wei-Tang Chien, et al.
+- **abs:** https://arxiv.org/abs/2603.20433 · **PDF:** https://arxiv.org/pdf/2603.20433v1
+
+## 3108. Coding Agents are Effective Long-Context Processors
+
+- **ID:** `2603.20432` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Weili Cao, Xunjian Yin, Bhuwan Dhingra, et al.
+- **abs:** https://arxiv.org/abs/2603.20432 · **PDF:** https://arxiv.org/pdf/2603.20432v1
+
+## 3109. Leveraging Natural Language Processing and Machine Learning for Evidence-Based Food Security Policy Decision-Making in Data-Scarce Making
+
+- **ID:** `2603.20425` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Karan Kumar Singh, Nikita Gajbhiye
+- **abs:** https://arxiv.org/abs/2603.20425 · **PDF:** https://arxiv.org/pdf/2603.20425v1
+
+## 3110. PEARL: Personalized Streaming Video Understanding Model
+
+- **ID:** `2603.20422` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Yuanhong Zheng, Ruichuan An, Xiaopeng Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.20422 · **PDF:** https://arxiv.org/pdf/2603.20422v1
+
+## 3111. Hawkeye: Reproducing GPU-Level Non-Determinism
+
+- **ID:** `2603.20421` · **primary:** `cs.CR` · **Submitted:** 2026-03-20
+- **Authors:** Erez Badash, Dan Boneh, Ilan Komargodski, et al.
+- **abs:** https://arxiv.org/abs/2603.20421 · **PDF:** https://arxiv.org/pdf/2603.20421v1
+
+## 3112. CERN: Correcting Errors in Raw Nanopore Signals Using Hidden Markov Models
+
+- **ID:** `2603.20420` · **primary:** `q-bio.GN` · **Submitted:** 2026-03-20
+- **Authors:** Simon Ambrozak, Ulysse McConnell, Bhargav Srinivasan, et al.
+- **abs:** https://arxiv.org/abs/2603.20420 · **PDF:** https://arxiv.org/pdf/2603.20420v1
+
+## 3113. Data-driven discovery of roughness descriptors for surface characterization and intimate contact modeling of unidirectional composite tapes
+
+- **ID:** `2603.20418` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Sebastian Rodriguez, Mikhael Tannous, Jad Mounayer, et al.
+- **abs:** https://arxiv.org/abs/2603.20418 · **PDF:** https://arxiv.org/pdf/2603.20418v1
+
+## 3114. Beyond the Mean: Distribution-Aware Loss Functions for Bimodal Regression
+
+- **ID:** `2603.22328` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Abolfazl Mohammadi-Seif, Carlos Soares, Rita P. Ribeiro, et al.
+- **abs:** https://arxiv.org/abs/2603.22328 · **PDF:** https://arxiv.org/pdf/2603.22328v1
+
+## 3115. SLE-FNO: Single-Layer Extensions for Task-Agnostic Continual Learning in Fourier Neural Operators
+
+- **ID:** `2603.20410` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Mahmoud Elhadidy, Roshan M. D'Souza, Amirhossein Arzani
+- **abs:** https://arxiv.org/abs/2603.20410 · **PDF:** https://arxiv.org/pdf/2603.20410v1
+
+## 3116. Mixture of Experts with Soft Nearest Neighbor Loss: Resolving Expert Collapse via Representation Disentanglement
+
+- **ID:** `2603.26734` · **primary:** `cs.NE` · **Submitted:** 2026-03-20
+- **Authors:** Abien Fred Agarap, Arnulfo P. Azcarraga
+- **abs:** https://arxiv.org/abs/2603.26734 · **PDF:** https://arxiv.org/pdf/2603.26734v1
+
+## 3117. Meta-Learning for Repeated Bayesian Persuasion
+
+- **ID:** `2603.20408` · **primary:** `cs.GT` · **Submitted:** 2026-03-20
+- **Authors:** Ata Poyraz Turna, Asrin Efe Yorulmaz, Tamer Başar
+- **abs:** https://arxiv.org/abs/2603.20408 · **PDF:** https://arxiv.org/pdf/2603.20408v1
+
+## 3118. Thinking in Different Spaces: Domain-Specific Latent Geometry Survives Cross-Architecture Translation
+
+- **ID:** `2603.20406` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Marcus Armstrong, Navid Ayoobi, Arjun Mukherjee
+- **abs:** https://arxiv.org/abs/2603.20406 · **PDF:** https://arxiv.org/pdf/2603.20406v1
+
+## 3119. Putnam 2025 Problems in Rocq using Opus 4.6 and Rocq-MCP
+
+- **ID:** `2603.20405` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Guillaume Baudart, Marc Lelarge, Tristan Stérin, et al.
+- **abs:** https://arxiv.org/abs/2603.20405 · **PDF:** https://arxiv.org/pdf/2603.20405v1
+
+## 3120. Hetero-Net: An Energy-Efficient Resource Allocation and 3D Placement in Heterogeneous LoRa Networks via Multi-Agent Optimization
+
+- **ID:** `2603.20404` · **primary:** `cs.NI` · **Submitted:** 2026-03-20
+- **Authors:** Abdullahi Isa Ahmed, Ana Maria Drăgulinescu, El Mehdi Amhoud
+- **abs:** https://arxiv.org/abs/2603.20404 · **PDF:** https://arxiv.org/pdf/2603.20404v1
+
+## 3121. KV Cache Optimization Strategies for Scalable and Efficient LLM Inference
+
+- **ID:** `2603.20397` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Yichun Xu, Navjot K. Khaira, Tejinder Singh
+- **abs:** https://arxiv.org/abs/2603.20397 · **PDF:** https://arxiv.org/pdf/2603.20397v1
+
+## 3122. Compression is all you need: Modeling Mathematics
+
+- **ID:** `2603.20396` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Vitaly Aksenov, Eve Bodnia, Michael H. Freedman, et al.
+- **abs:** https://arxiv.org/abs/2603.20396 · **PDF:** https://arxiv.org/pdf/2603.20396v1
+
+## 3123. SymCircuit: Bayesian Structure Inference for Tractable Probabilistic Circuits via Entropy-Regularized Reinforcement Learning
+
+- **ID:** `2603.20392` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Y. Sungtaek Ju
+- **abs:** https://arxiv.org/abs/2603.20392 · **PDF:** https://arxiv.org/pdf/2603.20392v1
+
+## 3124. CAMA: Exploring Collusive Adversarial Attacks in c-MARL
+
+- **ID:** `2603.20390` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Men Niu, Xinxin Fan, Quanliang Jing, et al.
+- **abs:** https://arxiv.org/abs/2603.20390 · **PDF:** https://arxiv.org/pdf/2603.20390v1
+
+## 3125. A chemical language model for reticular materials design
+
+- **ID:** `2603.20389` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-20
+- **Authors:** Dhruv Menon, Vivek Singh, Xu Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.20389 · **PDF:** https://arxiv.org/pdf/2603.20389v1
+
+## 3126. From Cross-Validation to SURE: Asymptotic Risk of Tuned Regularized Estimators
+
+- **ID:** `2603.20388` · **primary:** `math.ST` · **Submitted:** 2026-03-20
+- **Authors:** Karun Adusumilli, Maximilian Kasy, Ashia Wilson
+- **abs:** https://arxiv.org/abs/2603.20388 · **PDF:** https://arxiv.org/pdf/2603.20388v1
+
+## 3127. The production of meaning in the processing of natural language
+
+- **ID:** `2603.20381` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Christopher J. Agostino, Quan Le Thien, Nayan D'Souza, et al.
+- **abs:** https://arxiv.org/abs/2603.20381 · **PDF:** https://arxiv.org/pdf/2603.20381v1
+
+## 3128. ALARA for Agents: Least-Privilege Context Engineering Through Portable Composable Multi-Agent Teams
+
+- **ID:** `2603.20380` · **primary:** `cs.MA` · **Submitted:** 2026-03-20
+- **Authors:** Christopher J. Agostino, Nayan D'Souza
+- **abs:** https://arxiv.org/abs/2603.20380 · **PDF:** https://arxiv.org/pdf/2603.20380v1
+
+## 3129. From Masks to Pixels and Meaning: A New Taxonomy, Benchmark, and Metrics for VLM Image Tampering
+
+- **ID:** `2603.20193` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Xinyi Shang, Yi Tang, Jiacheng Cui, et al.
+- **abs:** https://arxiv.org/abs/2603.20193 · **PDF:** https://arxiv.org/pdf/2603.20193v1
+
+## 3130. LumosX: Relate Any Identities with Their Attributes for Personalized Video Generation
+
+- **ID:** `2603.20192` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Jiazheng Xing, Fei Du, Hangjie Yuan, et al.
+- **abs:** https://arxiv.org/abs/2603.20192 · **PDF:** https://arxiv.org/pdf/2603.20192v1
+
+## 3131. MeanFlow Meets Control: Scaling Sampled-Data Control for Swarms
+
+- **ID:** `2603.20189` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Anqi Dong, Yongxin Chen, Karl H. Johansson, et al.
+- **abs:** https://arxiv.org/abs/2603.20189 · **PDF:** https://arxiv.org/pdf/2603.20189v1
+
+## 3132. VideoSeek: Long-Horizon Video Agent with Tool-Guided Seeking
+
+- **ID:** `2603.20185` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Jingyang Lin, Jialian Wu, Jiang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.20185 · **PDF:** https://arxiv.org/pdf/2603.20185v1
+
+## 3133. Kolmogorov-Arnold causal generative models
+
+- **ID:** `2603.20184` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Alejandro Almodóvar, Mar Elizo, Patricia A. Apellániz, et al.
+- **abs:** https://arxiv.org/abs/2603.20184 · **PDF:** https://arxiv.org/pdf/2603.20184v1
+
+## 3134. Improving Generalization on Cybersecurity Tasks with Multi-Modal Contrastive Learning
+
+- **ID:** `2603.20181` · **primary:** `cs.CR` · **Submitted:** 2026-03-20
+- **Authors:** Jianan Huang, Rodolfo V. Valentim, Luca Vassio, et al.
+- **abs:** https://arxiv.org/abs/2603.20181 · **PDF:** https://arxiv.org/pdf/2603.20181v1
+
+## 3135. Adaptive Greedy Frame Selection for Long Video Understanding
+
+- **ID:** `2603.20180` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Yuning Huang, Fengqing Zhu
+- **abs:** https://arxiv.org/abs/2603.20180 · **PDF:** https://arxiv.org/pdf/2603.20180v1
+
+## 3136. AI Agents Can Already Autonomously Perform Experimental High Energy Physics
+
+- **ID:** `2603.20179` · **primary:** `hep-ex` · **Submitted:** 2026-03-20
+- **Authors:** Eric A. Moreno, Samuel Bright-Thonney, Andrzej Novak, et al.
+- **abs:** https://arxiv.org/abs/2603.20179 · **PDF:** https://arxiv.org/pdf/2603.20179v2
+
+## 3137. Measuring Faithfulness Depends on How You Measure: Classifier Sensitivity in LLM Chain-of-Thought Evaluation
+
+- **ID:** `2603.20172` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Richard J. Young
+- **abs:** https://arxiv.org/abs/2603.20172 · **PDF:** https://arxiv.org/pdf/2603.20172v2
+
+## 3138. WebNavigator: Global Web Navigation via Interaction Graph Retrieval
+
+- **ID:** `2603.20366` · **primary:** `cs.IR` · **Submitted:** 2026-03-20
+- **Authors:** Xuanwang Zhang, Yuteng Han, Jinnan Qi, et al.
+- **abs:** https://arxiv.org/abs/2603.20366 · **PDF:** https://arxiv.org/pdf/2603.20366v1
+
+## 3139. Learning Dynamic Belief Graphs for Theory-of-mind Reasoning
+
+- **ID:** `2603.20170` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Ruxiao Chen, Xilei Zhao, Thomas J. Cova, et al.
+- **abs:** https://arxiv.org/abs/2603.20170 · **PDF:** https://arxiv.org/pdf/2603.20170v1
+
+## 3140. Comprehensive Description of Uncertainty in Measurement for Representation and Propagation with Scalable Precision
+
+- **ID:** `2603.20365` · **primary:** `stat.ML` · **Submitted:** 2026-03-20
+- **Authors:** Ali Darijani, Jürgen Beyerer, Zahra Sadat Hajseyed Nasrollah, et al.
+- **abs:** https://arxiv.org/abs/2603.20365 · **PDF:** https://arxiv.org/pdf/2603.20365v1
+
+## 3141. The Robot's Inner Critic: Self-Refinement of Social Behaviors through VLM-based Replanning
+
+- **ID:** `2603.20164` · **primary:** `cs.RO` · **Submitted:** 2026-03-20
+- **Authors:** Jiyu Lim, Youngwoo Yoon, Kwanghyun Park
+- **abs:** https://arxiv.org/abs/2603.20164 · **PDF:** https://arxiv.org/pdf/2603.20164v1
+
+## 3142. Evaluating Evidence Grounding Under User Pressure in Instruction-Tuned Language Models
+
+- **ID:** `2603.20162` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Sai Koneru, Elphin Joe, Christine Kirchhoff, et al.
+- **abs:** https://arxiv.org/abs/2603.20162 · **PDF:** https://arxiv.org/pdf/2603.20162v1
+
+## 3143. Semantic Token Clustering for Efficient Uncertainty Quantification in Large Language Models
+
+- **ID:** `2603.20161` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Qi Cao, Andrew Gambardella, Takeshi Kojima, et al.
+- **abs:** https://arxiv.org/abs/2603.20161 · **PDF:** https://arxiv.org/pdf/2603.20161v1
+
+## 3144. Beyond Single Tokens: Distilling Discrete Diffusion Models via Discrete MMD
+
+- **ID:** `2603.20155` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Emiel Hoogeboom, David Ruhe, Jonathan Heek, et al.
+- **abs:** https://arxiv.org/abs/2603.20155 · **PDF:** https://arxiv.org/pdf/2603.20155v1
+
+## 3145. Design-OS: A Specification-Driven Framework for Engineering System Design with a Control-Systems Design Case
+
+- **ID:** `2603.20151` · **primary:** `cs.CE` · **Submitted:** 2026-03-20
+- **Authors:** H. Sinan Bank, Daniel R. Herber, Thomas H. Bradley
+- **abs:** https://arxiv.org/abs/2603.20151 · **PDF:** https://arxiv.org/pdf/2603.20151v1
+
+## 3146. Enhancing Hyperspace Analogue to Language (HAL) Representations via Attention-Based Pooling for Text Classification
+
+- **ID:** `2603.20149` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Ali Sakour, Zoalfekar Sakour
+- **abs:** https://arxiv.org/abs/2603.20149 · **PDF:** https://arxiv.org/pdf/2603.20149v1
+
+## 3147. AgentSLR: Automating Systematic Literature Reviews in Epidemiology with Agentic AI
+
+- **ID:** `2603.22327` · **primary:** `cs.IR` · **Submitted:** 2026-03-20
+- **Authors:** Shreyansh Padarha, Ryan Othniel Kearns, Tristan Naidoo, et al.
+- **abs:** https://arxiv.org/abs/2603.22327 · **PDF:** https://arxiv.org/pdf/2603.22327v1
+
+## 3148. Reasoning Gets Harder for LLMs Inside A Dialogue
+
+- **ID:** `2603.20133` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Ivan Kartáč, Mateusz Lango, Ondřej Dušek
+- **abs:** https://arxiv.org/abs/2603.20133 · **PDF:** https://arxiv.org/pdf/2603.20133v1
+
+## 3149. Revisiting Gene Ontology Knowledge Discovery with Hierarchical Feature Selection and Virtual Study Group of AI Agents
+
+- **ID:** `2603.20132` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Cen Wan, Alex A. Freitas
+- **abs:** https://arxiv.org/abs/2603.20132 · **PDF:** https://arxiv.org/pdf/2603.20132v1
+
+## 3150. An Agentic Multi-Agent Architecture for Cybersecurity Risk Management
+
+- **ID:** `2603.20131` · **primary:** `eess.SY` · **Submitted:** 2026-03-20
+- **Authors:** Ravish Gupta, Saket Kumar, Shreeya Sharma, et al.
+- **abs:** https://arxiv.org/abs/2603.20131 · **PDF:** https://arxiv.org/pdf/2603.20131v2
+
+## 3151. Evolving Jailbreaks: Automated Multi-Objective Long-Tail Attacks on Large Language Models
+
+- **ID:** `2603.20122` · **primary:** `cs.CR` · **Submitted:** 2026-03-20
+- **Authors:** Wenjing Hong, Zhonghua Rong, Li Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.20122 · **PDF:** https://arxiv.org/pdf/2603.20122v1
+
+## 3152. Chain-of-Adaptation: Surgical Vision-Language Adaptation with Reinforcement Learning
+
+- **ID:** `2603.20116` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Jiajie Li, Chenhui Xu, Meihuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.20116 · **PDF:** https://arxiv.org/pdf/2603.20116v1
+
+## 3153. Conditioning Protein Generation via Hopfield Pattern Multiplicity
+
+- **ID:** `2603.20115` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Jeffrey D. Varner
+- **abs:** https://arxiv.org/abs/2603.20115 · **PDF:** https://arxiv.org/pdf/2603.20115v1
+
+## 3154. Current LLMs still cannot 'talk much' about grammar modules: Evidence from syntax
+
+- **ID:** `2603.20114` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Mohammed Q. Shormani, Yehia A. AlSohbani
+- **abs:** https://arxiv.org/abs/2603.20114 · **PDF:** https://arxiv.org/pdf/2603.20114v3
+
+## 3155. Demonstration of Adapt4Me: An Uncertainty-Aware Authoring Environment for Personalizing Automatic Speech Recognition to Non-normative Speech
+
+- **ID:** `2603.20112` · **primary:** `cs.HC` · **Submitted:** 2026-03-20
+- **Authors:** Niclas Pokel, Yiming Zhao, Pehuén Moure, et al.
+- **abs:** https://arxiv.org/abs/2603.20112 · **PDF:** https://arxiv.org/pdf/2603.20112v1
+
+## 3156. Var-JEPA: A Variational Formulation of the Joint-Embedding Predictive Architecture -- Bridging Predictive and Generative Self-Supervised Learning
+
+- **ID:** `2603.20111` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Moritz Gögl, Christopher Yau
+- **abs:** https://arxiv.org/abs/2603.20111 · **PDF:** https://arxiv.org/pdf/2603.20111v1
+
+## 3157. GO-GenZip: Goal-Oriented Generative Sampling and Hybrid Compression
+
+- **ID:** `2603.20109` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Pietro Talli, Qi Liao, Alessandro Lieto, et al.
+- **abs:** https://arxiv.org/abs/2603.20109 · **PDF:** https://arxiv.org/pdf/2603.20109v1
+
+## 3158. Trojan horse hunt in deep forecasting models: Insights from the European Space Agency competition
+
+- **ID:** `2603.20108` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Krzysztof Kotowski, Ramez Shendy, Jakub Nalepa, et al.
+- **abs:** https://arxiv.org/abs/2603.20108 · **PDF:** https://arxiv.org/pdf/2603.20108v1
+
+## 3159. The $\mathbf{Y}$-Combinator for LLMs: Solving Long-Context Rot with $λ$-Calculus
+
+- **ID:** `2603.20105` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Amartya Roy, Rasul Tutunov, Xiaotong Ji, et al.
+- **abs:** https://arxiv.org/abs/2603.20105 · **PDF:** https://arxiv.org/pdf/2603.20105v1
+
+## 3160. Spectral Alignment in Forward-Backward Representations via Temporal Abstraction
+
+- **ID:** `2603.20103` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Seyed Mahdi B. Azad, Jasper Hoffmann, Iman Nematollahi, et al.
+- **abs:** https://arxiv.org/abs/2603.20103 · **PDF:** https://arxiv.org/pdf/2603.20103v2
+
+## 3161. Pitfalls in Evaluating Interpretability Agents
+
+- **ID:** `2603.20101` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Tal Haklay, Nikhil Prakash, Sana Pandey, et al.
+- **abs:** https://arxiv.org/abs/2603.20101 · **PDF:** https://arxiv.org/pdf/2603.20101v1
+
+## 3162. An Empirical Study of SFT-DPO Interaction and Parameterization in Small Language Models
+
+- **ID:** `2603.20100` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Yuming Feng, Christy Yang
+- **abs:** https://arxiv.org/abs/2603.20100 · **PDF:** https://arxiv.org/pdf/2603.20100v1
+
+## 3163. LLM-Enhanced Semantic Data Integration of Electronic Component Qualifications in the Aerospace Domain
+
+- **ID:** `2603.20094` · **primary:** `cs.IR` · **Submitted:** 2026-03-20
+- **Authors:** Antonio De Santis, Marco Balduini, Matteo Belcao, et al.
+- **abs:** https://arxiv.org/abs/2603.20094 · **PDF:** https://arxiv.org/pdf/2603.20094v1
+
+## 3164. How Out-of-Equilibrium Phase Transitions can Seed Pattern Formation in Trained Diffusion Models
+
+- **ID:** `2603.20092` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Luca Ambrogioni
+- **abs:** https://arxiv.org/abs/2603.20092 · **PDF:** https://arxiv.org/pdf/2603.20092v1
+
+## 3165. Operator Learning for Smoothing and Forecasting
+
+- **ID:** `2603.20359` · **primary:** `stat.ML` · **Submitted:** 2026-03-20
+- **Authors:** Edoardo Calvello, Elizabeth Carlson, Nikola Kovachki, et al.
+- **abs:** https://arxiv.org/abs/2603.20359 · **PDF:** https://arxiv.org/pdf/2603.20359v2
+
+## 3166. Predicting States of Understanding in Explanatory Interactions Using Cognitive Load-Related Linguistic Cues
+
+- **ID:** `2603.20079` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Yu Wang, Olcay Türk, Angela Grimminger, et al.
+- **abs:** https://arxiv.org/abs/2603.20079 · **PDF:** https://arxiv.org/pdf/2603.20079v1
+
+## 3167. A Direct Classification Approach for Reliable Wind Ramp Event Forecasting under Severe Class Imbalance
+
+- **ID:** `2603.22326` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Alejandro Morales-Hernández, Fabrizio De Caroa, Gian Marco Paldino, et al.
+- **abs:** https://arxiv.org/abs/2603.22326 · **PDF:** https://arxiv.org/pdf/2603.22326v1
+
+## 3168. Agentic Harness for Real-World Compilers
+
+- **ID:** `2603.20075` · **primary:** `cs.SE` · **Submitted:** 2026-03-20
+- **Authors:** Yingwei Zheng, Cong Li, Shaohua Li, et al.
+- **abs:** https://arxiv.org/abs/2603.20075 · **PDF:** https://arxiv.org/pdf/2603.20075v1
+
+## 3169. Antenna Array Beamforming Based on a Hybrid Quantum Optimization Framework
+
+- **ID:** `2603.20072` · **primary:** `quant-ph` · **Submitted:** 2026-03-20
+- **Authors:** Shuai Zeng
+- **abs:** https://arxiv.org/abs/2603.20072 · **PDF:** https://arxiv.org/pdf/2603.20072v1
+
+## 3170. The monotonicity of the Franz-Parisi potential is equivalent with Low-degree MMSE lower bounds
+
+- **ID:** `2603.20070` · **primary:** `math.ST` · **Submitted:** 2026-03-20
+- **Authors:** Konstantinos Tsirkas, Leda Wang, Ilias Zadik
+- **abs:** https://arxiv.org/abs/2603.20070 · **PDF:** https://arxiv.org/pdf/2603.20070v1
+
+## 3171. Fine-tuning Timeseries Predictors Using Reinforcement Learning
+
+- **ID:** `2603.20063` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Hugo Cazaux, Ralph Rudd, Hlynur Stefánsson, et al.
+- **abs:** https://arxiv.org/abs/2603.20063 · **PDF:** https://arxiv.org/pdf/2603.20063v1
+
+## 3172. The End of Rented Discovery: How AI Search Redistributes Power Between Hotels and Intermediaries
+
+- **ID:** `2603.20062` · **primary:** `cs.IR` · **Submitted:** 2026-03-20
+- **Authors:** Peiying Zhu, Sidi Chang
+- **abs:** https://arxiv.org/abs/2603.20062 · **PDF:** https://arxiv.org/pdf/2603.20062v2
+
+## 3173. DIAL-KG: Schema-Free Incremental Knowledge Graph Construction via Dynamic Schema Induction and Evolution-Intent Assessment
+
+- **ID:** `2603.20059` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Weidong Bao, Yilin Wang, Ruyu Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.20059 · **PDF:** https://arxiv.org/pdf/2603.20059v1
+
+## 3174. Structured Latent Dynamics in Wireless CSI via Homomorphic World Models
+
+- **ID:** `2603.20048` · **primary:** `eess.SP` · **Submitted:** 2026-03-20
+- **Authors:** Salmane Naoumi, Mehdi Bennis, Marwa Chafii
+- **abs:** https://arxiv.org/abs/2603.20048 · **PDF:** https://arxiv.org/pdf/2603.20048v1
+
+## 3175. Experience is the Best Teacher: Motivating Effective Exploration in Reinforcement Learning for LLMs
+
+- **ID:** `2603.20046` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Wenjian Zhang, Kongcheng Zhang, Jiaxin Qi, et al.
+- **abs:** https://arxiv.org/abs/2603.20046 · **PDF:** https://arxiv.org/pdf/2603.20046v1
+
+## 3176. LoASR-Bench: Evaluating Large Speech Language Models on Low-Resource Automatic Speech Recognition Across Language Families
+
+- **ID:** `2603.20042` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Jianan Chen, Xiaoxue Gao, Tatsuya Kawahara, et al.
+- **abs:** https://arxiv.org/abs/2603.20042 · **PDF:** https://arxiv.org/pdf/2603.20042v1
+
+## 3177. Federated Hyperdimensional Computing for Resource-Constrained Industrial IoT
+
+- **ID:** `2603.20037` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Nikita Zeulin, Olga Galinina, Nageen Himayat, et al.
+- **abs:** https://arxiv.org/abs/2603.20037 · **PDF:** https://arxiv.org/pdf/2603.20037v1
+
+## 3178. Continual Learning as Shared-Manifold Continuation Under Compatible Shift
+
+- **ID:** `2603.20036` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Henry J. Kobs
+- **abs:** https://arxiv.org/abs/2603.20036 · **PDF:** https://arxiv.org/pdf/2603.20036v1
+
+## 3179. CoverageBench: Evaluating Information Coverage across Tasks and Domains
+
+- **ID:** `2603.20034` · **primary:** `cs.IR` · **Submitted:** 2026-03-20
+- **Authors:** Saron Samuel, Andrew Yates, Dawn Lawrie, et al.
+- **abs:** https://arxiv.org/abs/2603.20034 · **PDF:** https://arxiv.org/pdf/2603.20034v1
+
+## 3180. Orchestrating Human-AI Software Delivery: A Retrospective Longitudinal Field Study of Three Software Modernization Programs
+
+- **ID:** `2603.20028` · **primary:** `cs.SE` · **Submitted:** 2026-03-20
+- **Authors:** Maximiliano Armesto, Christophe Kolb
+- **abs:** https://arxiv.org/abs/2603.20028 · **PDF:** https://arxiv.org/pdf/2603.20028v1
+
+## 3181. Graph-Informed Adversarial Modeling: Infimal Subadditivity of Interpolative Divergences
+
+- **ID:** `2603.20025` · **primary:** `stat.ML` · **Submitted:** 2026-03-20
+- **Authors:** Panagiota Birmpa, Eric Joseph Hall
+- **abs:** https://arxiv.org/abs/2603.20025 · **PDF:** https://arxiv.org/pdf/2603.20025v2
+
+## 3182. Layered Quantum Architecture Search for 3D Point Cloud Classification
+
+- **ID:** `2603.20024` · **primary:** `quant-ph` · **Submitted:** 2026-03-20
+- **Authors:** Natacha Kuete Meli, Jovita Lukasik, Vladislav Golyanik, et al.
+- **abs:** https://arxiv.org/abs/2603.20024 · **PDF:** https://arxiv.org/pdf/2603.20024v1
+
+## 3183. ODySSeI: An Open-Source End-to-End Framework for Automated Detection, Segmentation, and Severity Estimation of Lesions in Invasive Coronary Angiography Images
+
+- **ID:** `2603.20021` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Anand Choudhary, Xiaowu Sun, Thabo Mahendiran, et al.
+- **abs:** https://arxiv.org/abs/2603.20021 · **PDF:** https://arxiv.org/pdf/2603.20021v1
+
+## 3184. Detached Skip-Links and $R$-Probe: Decoupling Feature Aggregation from Gradient Propagation for MLLM OCR
+
+- **ID:** `2603.20020` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Ziye Yuan, Ruchang Yao, Chengxin Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.20020 · **PDF:** https://arxiv.org/pdf/2603.20020v1
+
+## 3185. RouterKGQA: Specialized--General Model Routing for Constraint-Aware Knowledge Graph Question Answering
+
+- **ID:** `2603.20017` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Bo Yuan, Hexuan Deng, Xuebo Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.20017 · **PDF:** https://arxiv.org/pdf/2603.20017v1
+
+## 3186. AgenticRS-EnsNAS: Ensemble-Decoupled Self-Evolving Architecture Search
+
+- **ID:** `2603.20014` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Yun Chen, Moyu Zhang, Jinxin Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.20014 · **PDF:** https://arxiv.org/pdf/2603.20014v1
+
+## 3187. A Super Fast K-means for Indexing Vector Embeddings
+
+- **ID:** `2603.20009` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Leonardo Kuffo, Sven Hepkema, Peter Boncz
+- **abs:** https://arxiv.org/abs/2603.20009 · **PDF:** https://arxiv.org/pdf/2603.20009v1
+
+## 3188. Physics-Informed Long-Range Coulomb Correction for Machine-learning Hamiltonians
+
+- **ID:** `2603.20007` · **primary:** `physics.comp-ph` · **Submitted:** 2026-03-20
+- **Authors:** Yang Zhong, Xiwen Li, Xingao Gong, et al.
+- **abs:** https://arxiv.org/abs/2603.20007 · **PDF:** https://arxiv.org/pdf/2603.20007v1
+
+## 3189. ReViSQL: Achieving Human-Level Text-to-SQL
+
+- **ID:** `2603.20004` · **primary:** `cs.DB` · **Submitted:** 2026-03-20
+- **Authors:** Yuxuan Zhu, Tengjun Jin, Yoojin Choi, et al.
+- **abs:** https://arxiv.org/abs/2603.20004 · **PDF:** https://arxiv.org/pdf/2603.20004v2
+
+## 3190. An Agentic Approach to Generating XAI-Narratives
+
+- **ID:** `2603.20003` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Yifan He, David Martens
+- **abs:** https://arxiv.org/abs/2603.20003 · **PDF:** https://arxiv.org/pdf/2603.20003v1
+
+## 3191. When Contextual Inference Fails: Cancelability in Interactive Instruction Following
+
+- **ID:** `2603.19997` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Natalia Bila, Kata Naszádi, Alexandra Mayn, et al.
+- **abs:** https://arxiv.org/abs/2603.19997 · **PDF:** https://arxiv.org/pdf/2603.19997v1
+
+## 3192. Evaluating Test-Time Adaptation For Facial Expression Recognition Under Natural Cross-Dataset Distribution Shifts
+
+- **ID:** `2603.19994` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** John Turnbull, Shivam Grover, Amin Jalali, et al.
+- **abs:** https://arxiv.org/abs/2603.19994 · **PDF:** https://arxiv.org/pdf/2603.19994v1
+
+## 3193. Breaking the Capability Ceiling of LLM Post-Training by Reintroducing Markov States
+
+- **ID:** `2603.19987` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Yurun Yuan, Tengyang Xie
+- **abs:** https://arxiv.org/abs/2603.19987 · **PDF:** https://arxiv.org/pdf/2603.19987v1
+
+## 3194. Memory poisoning and secure multi-agent systems
+
+- **ID:** `2603.20357` · **primary:** `cs.CR` · **Submitted:** 2026-03-20
+- **Authors:** Vicenç Torra, Maria Bras-Amorós
+- **abs:** https://arxiv.org/abs/2603.20357 · **PDF:** https://arxiv.org/pdf/2603.20357v1
+
+## 3195. X-World: Controllable Ego-Centric Multi-Camera World Models for Scalable End-to-End Driving
+
+- **ID:** `2603.19979` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Chaoda Zheng, Sean Li, Jinhao Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.19979 · **PDF:** https://arxiv.org/pdf/2603.19979v2
+
+## 3196. Promoting Critical Thinking With Domain-Specific Generative AI Provocations
+
+- **ID:** `2603.19975` · **primary:** `cs.HC` · **Submitted:** 2026-03-20
+- **Authors:** Thomas Şerban von Davier, Hao-Ping Lee, Jodi Forlizzi, et al.
+- **abs:** https://arxiv.org/abs/2603.19975 · **PDF:** https://arxiv.org/pdf/2603.19975v1
+
+## 3197. Trojan's Whisper: Stealthy Manipulation of OpenClaw through Injected Bootstrapped Guidance
+
+- **ID:** `2603.19974` · **primary:** `cs.CR` · **Submitted:** 2026-03-20
+- **Authors:** Fazhong Liu, Zhuoyan Chen, Tu Lan, et al.
+- **abs:** https://arxiv.org/abs/2603.19974 · **PDF:** https://arxiv.org/pdf/2603.19974v1
+
+## 3198. Model-Driven Learning-Based Physical Layer Authentication for Mobile Wi-Fi Devices
+
+- **ID:** `2603.19972` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Yijia Guo, Junqing Zhang, Yao-Win Peter Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.19972 · **PDF:** https://arxiv.org/pdf/2603.19972v1
+
+## 3199. Graph2TS: Structure-Controlled Time Series Generation via Quantile-Graph VAEs
+
+- **ID:** `2603.19970` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Shaoshuai Du, Joze M. Rozanec, Andy Pimentel, et al.
+- **abs:** https://arxiv.org/abs/2603.19970 · **PDF:** https://arxiv.org/pdf/2603.19970v1
+
+## 3200. Hybrid Associative Memories
+
+- **ID:** `2603.22325` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Leon Lufkin, Tomás Figliolia, Beren Millidge, et al.
+- **abs:** https://arxiv.org/abs/2603.22325 · **PDF:** https://arxiv.org/pdf/2603.22325v2
+
+## 3201. Channel Prediction-Based Physical Layer Authentication under Consecutive Spoofing Attacks
+
+- **ID:** `2603.19962` · **primary:** `cs.CR` · **Submitted:** 2026-03-20
+- **Authors:** Yijia Guo, Junqing Zhang, Yao-Win Peter Hong
+- **abs:** https://arxiv.org/abs/2603.19962 · **PDF:** https://arxiv.org/pdf/2603.19962v1
+
+## 3202. HiPath: Hierarchical Vision-Language Alignment for Structured Pathology Report Prediction
+
+- **ID:** `2603.19957` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Ruicheng Yuan, Zhenxuan Zhang, Anbang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.19957 · **PDF:** https://arxiv.org/pdf/2603.19957v1
+
+## 3203. Structural Controllability of Large-Scale Hypergraphs
+
+- **ID:** `2603.19955` · **primary:** `math.OC` · **Submitted:** 2026-03-20
+- **Authors:** Joshua Pickard, Xin Mao, Can Chen
+- **abs:** https://arxiv.org/abs/2603.19955 · **PDF:** https://arxiv.org/pdf/2603.19955v1
+
+## 3204. On the Ability of Transformers to Verify Plans
+
+- **ID:** `2603.19954` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Yash Sarrof, Yupei Du, Katharina Stein, et al.
+- **abs:** https://arxiv.org/abs/2603.19954 · **PDF:** https://arxiv.org/pdf/2603.19954v1
+
+## 3205. TAPAS: Efficient Two-Server Asymmetric Private Aggregation Beyond Prio(+)
+
+- **ID:** `2603.19949` · **primary:** `cs.CR` · **Submitted:** 2026-03-20
+- **Authors:** Harish Karthikeyan, Antigoni Polychroniadou
+- **abs:** https://arxiv.org/abs/2603.19949 · **PDF:** https://arxiv.org/pdf/2603.19949v1
+
+## 3206. Decomposable Reward Modeling and Realistic Environment Design for Reinforcement Learning-Based Forex Trading
+
+- **ID:** `2604.00031` · **primary:** `q-fin.GN` · **Submitted:** 2026-03-20
+- **Authors:** Nabeel Ahmad Saidd
+- **abs:** https://arxiv.org/abs/2604.00031 · **PDF:** https://arxiv.org/pdf/2604.00031v1
+
+## 3207. Hybrid topic modelling for computational close reading: Mapping narrative themes in Pushkin's Evgenij Onegin
+
+- **ID:** `2603.19940` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Angelo Maria Sabatini
+- **abs:** https://arxiv.org/abs/2603.19940 · **PDF:** https://arxiv.org/pdf/2603.19940v1
+
+## 3208. Memori: A Persistent Memory Layer for Efficient, Context-Aware LLM Agents
+
+- **ID:** `2603.19935` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Luiz C. Borro, Luiz A. B. Macarini, Gordon Tindall, et al.
+- **abs:** https://arxiv.org/abs/2603.19935 · **PDF:** https://arxiv.org/pdf/2603.19935v1
+
+## 3209. Contextual inference from single objects in Vision-Language models
+
+- **ID:** `2603.26731` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Martina G. Vilas, Timothy Schaumlöffel, Gemma Roig
+- **abs:** https://arxiv.org/abs/2603.26731 · **PDF:** https://arxiv.org/pdf/2603.26731v1
+
+## 3210. SAGE: Sustainable Agent-Guided Expert-tuning for Culturally Attuned Translation in Low-Resource Southeast Asia
+
+- **ID:** `2603.19931` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Zhixiang Lu, Chong Zhang, Yulong Li, et al.
+- **abs:** https://arxiv.org/abs/2603.19931 · **PDF:** https://arxiv.org/pdf/2603.19931v1
+
+## 3211. RAM: Recover Any 3D Human Motion in-the-Wild
+
+- **ID:** `2603.19929` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Sen Jia, Ning Zhu, Jinqin Zhong, et al.
+- **abs:** https://arxiv.org/abs/2603.19929 · **PDF:** https://arxiv.org/pdf/2603.19929v1
+
+## 3212. Translation from the Information Bottleneck Perspective: an Efficiency Analysis of Spatial Prepositions in Bitexts
+
+- **ID:** `2603.19924` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Antoine Taroni, Ludovic Moncla, Frederique Laforest
+- **abs:** https://arxiv.org/abs/2603.19924 · **PDF:** https://arxiv.org/pdf/2603.19924v1
+
+## 3213. Span-Level Machine Translation Meta-Evaluation
+
+- **ID:** `2603.19921` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Stefano Perrella, Eric Morales Agostinho, Hugo Zaragoza
+- **abs:** https://arxiv.org/abs/2603.19921 · **PDF:** https://arxiv.org/pdf/2603.19921v1
+
+## 3214. Learning Like Humans: Analogical Concept Learning for Generalized Category Discovery
+
+- **ID:** `2603.19918` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Jizhou Han, Chenhao Ding, Yuhang He, et al.
+- **abs:** https://arxiv.org/abs/2603.19918 · **PDF:** https://arxiv.org/pdf/2603.19918v1
+
+## 3215. Leum-VL Technical Report
+
+- **ID:** `2603.20354` · **primary:** `cs.MM` · **Submitted:** 2026-03-20
+- **Authors:** Yuxuan He, Chaiming Huang, Yifan Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.20354 · **PDF:** https://arxiv.org/pdf/2603.20354v1
+
+## 3216. Infinite-dimensional spherical-radial decomposition for probabilistic functions, with application to constrained optimal control and Gaussian process regression
+
+- **ID:** `2603.19907` · **primary:** `math.OC` · **Submitted:** 2026-03-20
+- **Authors:** Kewei Wang, Georg Stadler
+- **abs:** https://arxiv.org/abs/2603.19907 · **PDF:** https://arxiv.org/pdf/2603.19907v1
+
+## 3217. A Federated Many-to-One Hopfield model for associative Neural Networks
+
+- **ID:** `2603.19902` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-03-20
+- **Authors:** Andrea Alessandrelli, Fabrizio Durante, Andrea Ladiana, et al.
+- **abs:** https://arxiv.org/abs/2603.19902 · **PDF:** https://arxiv.org/pdf/2603.19902v1
+
+## 3218. Deep Autocorrelation Modeling for Time-Series Forecasting: Progress and Prospects
+
+- **ID:** `2603.19899` · **primary:** `stat.ML` · **Submitted:** 2026-03-20
+- **Authors:** Hao Wang, Licheng Pan, Qingsong Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.19899 · **PDF:** https://arxiv.org/pdf/2603.19899v1
+
+## 3219. Revealing Domain-Spatiality Patterns for Configuration Tuning: Domain Knowledge Meets Fitness Landscapes
+
+- **ID:** `2603.19897` · **primary:** `cs.SE` · **Submitted:** 2026-03-20
+- **Authors:** Yulong Ye, Hongyuan Liang, Chao Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.19897 · **PDF:** https://arxiv.org/pdf/2603.19897v2
+
+## 3220. DAQ: Delta-Aware Quantization for Post-Training LLM Weight Compression
+
+- **ID:** `2603.22324` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Xiaoming Yu, Shize Tang, Guanghua Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.22324 · **PDF:** https://arxiv.org/pdf/2603.22324v1
+
+## 3221. Utility-Guided Agent Orchestration for Efficient LLM Tool Use
+
+- **ID:** `2603.19896` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Boyan Liu, Gongming Zhao, Hongli Xu
+- **abs:** https://arxiv.org/abs/2603.19896 · **PDF:** https://arxiv.org/pdf/2603.19896v1
+
+## 3222. The Multiverse of Time Series Machine Learning: an Archive for Multivariate Time Series Classification
+
+- **ID:** `2603.20352` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Matthew Middlehurst, Aiden Rushbrooke, Ali Ismail-Fawaz, et al.
+- **abs:** https://arxiv.org/abs/2603.20352 · **PDF:** https://arxiv.org/pdf/2603.20352v1
+
+## 3223. A Multi-Task Targeted Learning Framework for Lithium-Ion Battery State-of-Health and Remaining Useful Life
+
+- **ID:** `2603.22323` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Chenhan Wang, Zhengyi Bao, Huipin Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.22323 · **PDF:** https://arxiv.org/pdf/2603.22323v1
+
+## 3224. MANA: Towards Efficient Mobile Ad Detection via Multimodal Agentic UI Navigation
+
+- **ID:** `2603.20351` · **primary:** `cs.CR` · **Submitted:** 2026-03-20
+- **Authors:** Yizhe Zhao, Yongjian Fu, Zihao Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.20351 · **PDF:** https://arxiv.org/pdf/2603.20351v1
+
+## 3225. Integrating Meta-Features with Knowledge Graph Embeddings for Meta-Learning
+
+- **ID:** `2603.19888` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Antonis Klironomos, Ioannis Dasoulas, Francesco Periti, et al.
+- **abs:** https://arxiv.org/abs/2603.19888 · **PDF:** https://arxiv.org/pdf/2603.19888v1
+
+## 3226. AEGIS: An Operational Infrastructure for Post-Market Governance of Adaptive Medical AI Under US and EU Regulations
+
+- **ID:** `2603.22322` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Fardin Afdideh, Mehdi Astaraki, Fernando Seoane, et al.
+- **abs:** https://arxiv.org/abs/2603.22322 · **PDF:** https://arxiv.org/pdf/2603.22322v1
+
+## 3227. From Instructions to Assistance: a Dataset Aligning Instruction Manuals with Assembly Videos for Evaluating Multimodal LLMs
+
+- **ID:** `2603.22321` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Federico Toschi, Nicolò Brunello, Andrea Sassella, et al.
+- **abs:** https://arxiv.org/abs/2603.22321 · **PDF:** https://arxiv.org/pdf/2603.22321v1
+
+## 3228. What If Consensus Lies? Selective-Complementary Reinforcement Learning at Test Time
+
+- **ID:** `2603.19880` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Dong Yan, Jian Liang, Yanbo Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.19880 · **PDF:** https://arxiv.org/pdf/2603.19880v1
+
+## 3229. Discovery of Decision Synchronization Patterns from Event Logs
+
+- **ID:** `2603.19879` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Tijmen Kuijpers, Karolin Winter, Remco Dijkman
+- **abs:** https://arxiv.org/abs/2603.19879 · **PDF:** https://arxiv.org/pdf/2603.19879v1
+
+## 3230. Minimax Generalized Cross-Entropy
+
+- **ID:** `2603.19874` · **primary:** `stat.ML` · **Submitted:** 2026-03-20
+- **Authors:** Kartheek Bondugula, Santiago Mazuelas, Aritz Pérez, et al.
+- **abs:** https://arxiv.org/abs/2603.19874 · **PDF:** https://arxiv.org/pdf/2603.19874v2
+
+## 3231. On the Dynamics & Transferability of Latent Generalization during Memorization
+
+- **ID:** `2603.19865` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Simran Ketha, Venkatakrishnan Ramaswamy
+- **abs:** https://arxiv.org/abs/2603.19865 · **PDF:** https://arxiv.org/pdf/2603.19865v1
+
+## 3232. NASimJax: GPU-Accelerated Policy Learning Framework for Penetration Testing
+
+- **ID:** `2603.19864` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Raphael Simon, José Carrasquel, Wim Mees, et al.
+- **abs:** https://arxiv.org/abs/2603.19864 · **PDF:** https://arxiv.org/pdf/2603.19864v1
+
+## 3233. IsoCLIP: Decomposing CLIP Projectors for Efficient Intra-modal Alignment
+
+- **ID:** `2603.19862` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Simone Magistri, Dipam Goswami, Marco Mistretta, et al.
+- **abs:** https://arxiv.org/abs/2603.19862 · **PDF:** https://arxiv.org/pdf/2603.19862v1
+
+## 3234. Failure Modes for Deep Learning-Based Online Mapping: How to Measure and Address Them
+
+- **ID:** `2603.19852` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Michael Hubbertz, Qi Han, Tobias Meisen
+- **abs:** https://arxiv.org/abs/2603.19852 · **PDF:** https://arxiv.org/pdf/2603.19852v1
+
+## 3235. Semantic Delta: An Interpretable Signal Differentiating Human and LLMs Dialogue
+
+- **ID:** `2603.19849` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Riccardo Scantamburlo, Mauro Mezzanzana, Giacomo Buonanno, et al.
+- **abs:** https://arxiv.org/abs/2603.19849 · **PDF:** https://arxiv.org/pdf/2603.19849v1
+
+## 3236. G2DR: A Genotype-First Framework for Genetics-Informed Target Prioritization and Drug Repurposing
+
+- **ID:** `2603.20346` · **primary:** `q-bio.GN` · **Submitted:** 2026-03-20
+- **Authors:** Muhammad Muneeb, David B. Ascher
+- **abs:** https://arxiv.org/abs/2603.20346 · **PDF:** https://arxiv.org/pdf/2603.20346v1
+
+## 3237. Bridging the Gap Between Climate Science and Machine Learning in Climate Model Emulation
+
+- **ID:** `2603.22320` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Luca Schmidt, Nina Effenberger
+- **abs:** https://arxiv.org/abs/2603.22320 · **PDF:** https://arxiv.org/pdf/2603.22320v1
+
+## 3238. Overreliance on AI in Information-seeking from Video Content
+
+- **ID:** `2603.19843` · **primary:** `cs.CY` · **Submitted:** 2026-03-20
+- **Authors:** Anders Giovanni Møller, Elisa Bassignana, Francesco Pierri, et al.
+- **abs:** https://arxiv.org/abs/2603.19843 · **PDF:** https://arxiv.org/pdf/2603.19843v1
+
+## 3239. Modeling subgrid scale production rates on complex meshes using graph neural networks
+
+- **ID:** `2603.19841` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-03-20
+- **Authors:** Priyabrat Dash, Mathis Bode, Konduri Aditya
+- **abs:** https://arxiv.org/abs/2603.19841 · **PDF:** https://arxiv.org/pdf/2603.19841v1
+
+## 3240. Explainable cluster analysis: a bagging approach
+
+- **ID:** `2603.19840` · **primary:** `stat.ML` · **Submitted:** 2026-03-20
+- **Authors:** Federico Maria Quetti, Elena Ballante, Silvia Figini, et al.
+- **abs:** https://arxiv.org/abs/2603.19840 · **PDF:** https://arxiv.org/pdf/2603.19840v1
+
+## 3241. FIPO: Eliciting Deep Reasoning with Future-KL Influenced Policy Optimization
+
+- **ID:** `2603.19835` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Chiyu Ma, Shuo Yang, Kexin Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.19835 · **PDF:** https://arxiv.org/pdf/2603.19835v3
+
+## 3242. Gesture2Speech: How Far Can Hand Movements Shape Expressive Speech?
+
+- **ID:** `2603.19831` · **primary:** `eess.AS` · **Submitted:** 2026-03-20
+- **Authors:** Lokesh Kumar, Nirmesh Shah, Ashishkumar P. Gudmalwar, et al.
+- **abs:** https://arxiv.org/abs/2603.19831 · **PDF:** https://arxiv.org/pdf/2603.19831v1
+
+## 3243. FormalEvolve: Neuro-Symbolic Evolutionary Search for Diverse and Prover-Effective Autoformalization
+
+- **ID:** `2603.19828` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Haijian Lu, Wei Wang, Jing Liu
+- **abs:** https://arxiv.org/abs/2603.19828 · **PDF:** https://arxiv.org/pdf/2603.19828v2
+
+## 3244. FrameNet Semantic Role Classification by Analogy
+
+- **ID:** `2603.19825` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Van-Duy Ngo, Stergos Afantenos, Emiliano Lorini, et al.
+- **abs:** https://arxiv.org/abs/2603.19825 · **PDF:** https://arxiv.org/pdf/2603.19825v1
+
+## 3245. Sparsely-Supervised Data Assimilation via Physics-Informed Schrödinger Bridge
+
+- **ID:** `2603.22319` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Dohyun Bu, Chanho Kim, Seokun Choi, et al.
+- **abs:** https://arxiv.org/abs/2603.22319 · **PDF:** https://arxiv.org/pdf/2603.22319v1
+
+## 3246. GDEGAN: Gaussian Dynamic Equivariant Graph Attention Network for Ligand Binding Site Prediction
+
+- **ID:** `2603.19817` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Animesh, Plaban Kumar Bhowmick, Pralay Mitra
+- **abs:** https://arxiv.org/abs/2603.19817 · **PDF:** https://arxiv.org/pdf/2603.19817v1
+
+## 3247. Eye Gaze-Informed and Context-Aware Pedestrian Trajectory Prediction in Shared Spaces with Automated Shuttles: A Virtual Reality Study
+
+- **ID:** `2603.19812` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Danya Li, Yan Feng, Rico Krueger
+- **abs:** https://arxiv.org/abs/2603.19812 · **PDF:** https://arxiv.org/pdf/2603.19812v1
+
+## 3248. Two-Time-Scale Learning Dynamics: A Population View of Neural Network Training
+
+- **ID:** `2603.19808` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Giacomo Borghi, Hyesung Im, Lorenzo Pareschi
+- **abs:** https://arxiv.org/abs/2603.19808 · **PDF:** https://arxiv.org/pdf/2603.19808v2
+
+## 3249. A graph neural network based chemical mechanism reduction method for combustion applications
+
+- **ID:** `2603.22318` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Manuru Nithin Padiyar, Priyabrat Dash, Konduri Aditya
+- **abs:** https://arxiv.org/abs/2603.22318 · **PDF:** https://arxiv.org/pdf/2603.22318v1
+
+## 3250. Enhancing Alignment for Unified Multimodal Models via Semantically-Grounded Supervision
+
+- **ID:** `2603.19807` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Jiyeong Kim, Yerim So, Hyesong Choi, et al.
+- **abs:** https://arxiv.org/abs/2603.19807 · **PDF:** https://arxiv.org/pdf/2603.19807v1
+
+## 3251. Quantifying Gate Contribution in Quantum Feature Maps for Scalable Circuit Optimization
+
+- **ID:** `2603.19805` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** F. Rodríguez-Díaz, D. Gutiérrez-Avilés, A. Troncoso, et al.
+- **abs:** https://arxiv.org/abs/2603.19805 · **PDF:** https://arxiv.org/pdf/2603.19805v1
+
+## 3252. Uncertainty Quantification Via the Posterior Predictive Variance
+
+- **ID:** `2603.19804` · **primary:** `math.ST` · **Submitted:** 2026-03-20
+- **Authors:** Sanjay Chaudhuri, Dean Dustin, Bertrand Clarke
+- **abs:** https://arxiv.org/abs/2603.19804 · **PDF:** https://arxiv.org/pdf/2603.19804v1
+
+## 3253. Offshore oil and gas platform dynamics in the North Sea, Gulf of Mexico, and Persian Gulf: Exploiting the Sentinel-1 archive
+
+- **ID:** `2603.19801` · **primary:** `eess.IV` · **Submitted:** 2026-03-20
+- **Authors:** Robin Spanier, Thorsten Hoeser, John Truckenbrodt, et al.
+- **abs:** https://arxiv.org/abs/2603.19801 · **PDF:** https://arxiv.org/pdf/2603.19801v1
+
+## 3254. Borderless Long Speech Synthesis
+
+- **ID:** `2603.19798` · **primary:** `cs.SD` · **Submitted:** 2026-03-20
+- **Authors:** Xingchen Song, Di Wu, Dinghao Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.19798 · **PDF:** https://arxiv.org/pdf/2603.19798v1
+
+## 3255. Interpretable Multiple Myeloma Prognosis with Observational Medical Outcomes Partnership Data
+
+- **ID:** `2603.20341` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Salma Rachidi, Aso Bozorgpanah, Eric Fey, et al.
+- **abs:** https://arxiv.org/abs/2603.20341 · **PDF:** https://arxiv.org/pdf/2603.20341v1
+
+## 3256. Scalable Learning of Multivariate Distributions via Coresets
+
+- **ID:** `2603.19792` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Zeyu Ding, Katja Ickstadt, Nadja Klein, et al.
+- **abs:** https://arxiv.org/abs/2603.19792 · **PDF:** https://arxiv.org/pdf/2603.19792v1
+
+## 3257. ContractSkill: Repairable Contract-Based Skills for Multimodal Web Agents
+
+- **ID:** `2603.20340` · **primary:** `cs.SE` · **Submitted:** 2026-03-20
+- **Authors:** Zijian Lu, Yiping Zuo, Yupeng Nie, et al.
+- **abs:** https://arxiv.org/abs/2603.20340 · **PDF:** https://arxiv.org/pdf/2603.20340v2
+
+## 3258. Learning Hierarchical Orthogonal Prototypes for Generalized Few-Shot 3D Point Cloud Segmentation
+
+- **ID:** `2603.19788` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Yifei Zhao, Fanyu Zhao, Zhongyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.19788 · **PDF:** https://arxiv.org/pdf/2603.19788v1
+
+## 3259. Embodied Science: Closing the Discovery Loop with Agentic Embodied AI
+
+- **ID:** `2603.19782` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Xiang Zhuang, Chenyi Zhou, Kehua Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.19782 · **PDF:** https://arxiv.org/pdf/2603.19782v1
+
+## 3260. Graph-Aware Stealthy Poison-Text Backdoors for Text-Attributed Graphs
+
+- **ID:** `2603.20339` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Qi Luo, Minghui Xu, Dongxiao Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.20339 · **PDF:** https://arxiv.org/pdf/2603.20339v2
+
+## 3261. Sommelier: Scalable Open Multi-turn Audio Pre-processing for Full-duplex Speech Language Models
+
+- **ID:** `2603.25750` · **primary:** `cs.SD` · **Submitted:** 2026-03-20
+- **Authors:** Kyudan Jung, Jihwan Kim, Soyoon Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.25750 · **PDF:** https://arxiv.org/pdf/2603.25750v2
+
+## 3262. Neither Here Nor There: Cross-Lingual Representation Dynamics of Code-Mixed Text in Multilingual Encoders
+
+- **ID:** `2603.19771` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Debajyoti Mazumder, Divyansh Pathak, Prashant Kodali, et al.
+- **abs:** https://arxiv.org/abs/2603.19771 · **PDF:** https://arxiv.org/pdf/2603.19771v1
+
+## 3263. Low-pass Personalized Subgraph Federated Recommendation
+
+- **ID:** `2603.20338` · **primary:** `cs.IR` · **Submitted:** 2026-03-20
+- **Authors:** Wooseok Sim, Hogun Park
+- **abs:** https://arxiv.org/abs/2603.20338 · **PDF:** https://arxiv.org/pdf/2603.20338v1
+
+## 3264. PDGMM-VAE: A Variational Autoencoder with Adaptive Per-Dimension Gaussian Mixture Model Priors for Nonlinear ICA
+
+- **ID:** `2603.23547` · **primary:** `stat.ML` · **Submitted:** 2026-03-20
+- **Authors:** Yuan-Hao Wei, Yan-Jie Sun
+- **abs:** https://arxiv.org/abs/2603.23547 · **PDF:** https://arxiv.org/pdf/2603.23547v1
+
+## 3265. Growing Networks with Autonomous Pruning
+
+- **ID:** `2603.19759` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Charles De Lambilly, Stefan Duffner
+- **abs:** https://arxiv.org/abs/2603.19759 · **PDF:** https://arxiv.org/pdf/2603.19759v1
+
+## 3266. GEM: A Native Graph-based Index for Multi-Vector Retrieval
+
+- **ID:** `2603.20336` · **primary:** `cs.IR` · **Submitted:** 2026-03-20
+- **Authors:** Yao Tian, Zhoujin Tian, Xi Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.20336 · **PDF:** https://arxiv.org/pdf/2603.20336v1
+
+## 3267. Uncertainty-aware Prototype Learning with Variational Inference for Few-shot Point Cloud Segmentation
+
+- **ID:** `2603.19757` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Yifei Zhao, Fanyu Zhao, Yinsheng Li
+- **abs:** https://arxiv.org/abs/2603.19757 · **PDF:** https://arxiv.org/pdf/2603.19757v1
+
+## 3268. Regularity of Solutions to Beckmann's Parametric Optimal Transport
+
+- **ID:** `2603.19755` · **primary:** `math.AP` · **Submitted:** 2026-03-20
+- **Authors:** Hanno Gottschalk, Tobias J. Riedlinger
+- **abs:** https://arxiv.org/abs/2603.19755 · **PDF:** https://arxiv.org/pdf/2603.19755v1
+
+## 3269. Rethinking Ground Truth: A Case Study on Human Label Variation in MLLM Benchmarking
+
+- **ID:** `2603.19744` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Tomas Ruiz, Tanalp Agustoslu, Carsten Schwemmer
+- **abs:** https://arxiv.org/abs/2603.19744 · **PDF:** https://arxiv.org/pdf/2603.19744v1
+
+## 3270. Dual Path Attribution: Efficient Attribution for SwiGLU-Transformers through Layer-Wise Target Propagation
+
+- **ID:** `2603.19742` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Lasse Marten Jantsch, Dong-Jae Koh, Seonghyeon Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.19742 · **PDF:** https://arxiv.org/pdf/2603.19742v1
+
+## 3271. Hybrid Autoencoder-Isolation Forest approach for time series anomaly detection in C70XP cyclotron operation data at ARRONAX
+
+- **ID:** `2603.20335` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** F Basbous, F Poirier, F Haddad, et al.
+- **abs:** https://arxiv.org/abs/2603.20335 · **PDF:** https://arxiv.org/pdf/2603.20335v1
+
+## 3272. FedPDPO: Federated Personalized Direct Preference Optimization for Large Language Model Alignment
+
+- **ID:** `2603.19741` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Kewen Zhu, Liping Yi, Zhiming Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.19741 · **PDF:** https://arxiv.org/pdf/2603.19741v1
+
+## 3273. MOSS-TTSD: Text to Spoken Dialogue Generation
+
+- **ID:** `2603.19739` · **primary:** `cs.SD` · **Submitted:** 2026-03-20
+- **Authors:** Yuqian Zhang, Donghua Yu, Zhengyuan Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.19739 · **PDF:** https://arxiv.org/pdf/2603.19739v1
+
+## 3274. A two-step sequential approach for hyperparameter selection in finite context models
+
+- **ID:** `2603.19736` · **primary:** `stat.ML` · **Submitted:** 2026-03-20
+- **Authors:** José Contente, Ana Martins, Armando J. Pinho, et al.
+- **abs:** https://arxiv.org/abs/2603.19736 · **PDF:** https://arxiv.org/pdf/2603.19736v1
+
+## 3275. Multi-view Graph Convolutional Network with Fully Leveraging Consistency via Granular-ball-based Topology Construction, Feature Enhancement and Interactive Fusion
+
+- **ID:** `2603.26729` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Chengjie Cui, Taihua Xua, Shuyin Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.26729 · **PDF:** https://arxiv.org/pdf/2603.26729v1
+
+## 3276. PoC: Performance-oriented Context Compression for Large Language Models via Performance Prediction
+
+- **ID:** `2603.19733` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Runsong Zhao, Shilei Liu, Jiwei Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.19733 · **PDF:** https://arxiv.org/pdf/2603.19733v1
+
+## 3277. Procedural Refinement by LLM-driven Algorithmic Debugging for ARC-AGI-2
+
+- **ID:** `2603.20334` · **primary:** `cs.SE` · **Submitted:** 2026-03-20
+- **Authors:** Yu-Ning Qiu, Lin-Feng Zou, Jiong-Da Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.20334 · **PDF:** https://arxiv.org/pdf/2603.20334v1
+
+## 3278. FedRG: Unleashing the Representation Geometry for Federated Learning with Noisy Clients
+
+- **ID:** `2603.19722` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Tian Wen, Zhiqin Yang, Yonggang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.19722 · **PDF:** https://arxiv.org/pdf/2603.19722v2
+
+## 3279. Stepwise: Neuro-Symbolic Proof Search for Automated Systems Verification
+
+- **ID:** `2603.19715` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Baoding He, Zenan Li, Wei Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.19715 · **PDF:** https://arxiv.org/pdf/2603.19715v1
+
+## 3280. LoopRPT: Reinforcement Pre-Training for Looped Language Models
+
+- **ID:** `2603.19714` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Guo Tang, Shixin Jiang, Heng Chang, et al.
+- **abs:** https://arxiv.org/abs/2603.19714 · **PDF:** https://arxiv.org/pdf/2603.19714v1
+
+## 3281. Learning from Similarity/Dissimilarity and Pairwise Comparison
+
+- **ID:** `2603.19713` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Tomoya Tate, Kosuke Sugiyama, Masato Uchida
+- **abs:** https://arxiv.org/abs/2603.19713 · **PDF:** https://arxiv.org/pdf/2603.19713v1
+
+## 3282. TAB-AUDIT: Detecting AI-Fabricated Scientific Tables via Multi-View Likelihood Mismatch
+
+- **ID:** `2603.19712` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Shuo Huang, Yan Pen, Lizhen Qu
+- **abs:** https://arxiv.org/abs/2603.19712 · **PDF:** https://arxiv.org/pdf/2603.19712v1
+
+## 3283. EvoTaxo: Building and Evolving Taxonomy from Social Media Streams
+
+- **ID:** `2603.19711` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Yiyang Li, Tianyi Ma, Yanfang Ye
+- **abs:** https://arxiv.org/abs/2603.19711 · **PDF:** https://arxiv.org/pdf/2603.19711v1
+
+## 3284. AIGQ: An End-to-End Hybrid Generative Architecture for E-commerce Query Recommendation
+
+- **ID:** `2603.19710` · **primary:** `cs.IR` · **Submitted:** 2026-03-20
+- **Authors:** Jingcao Xu, Jianyun Zou, Renkai Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.19710 · **PDF:** https://arxiv.org/pdf/2603.19710v1
+
+## 3285. Bounded Coupled AI Learning Dynamics in Tri-Hierarchical Drone Swarms
+
+- **ID:** `2603.20333` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Oleksii Bychkov
+- **abs:** https://arxiv.org/abs/2603.20333 · **PDF:** https://arxiv.org/pdf/2603.20333v1
+
+## 3286. Minimax and Adaptive Covariance Matrix Estimation under Differential Privacy
+
+- **ID:** `2603.19703` · **primary:** `math.ST` · **Submitted:** 2026-03-20
+- **Authors:** T. Tony Cai, Yicheng Li
+- **abs:** https://arxiv.org/abs/2603.19703 · **PDF:** https://arxiv.org/pdf/2603.19703v1
+
+## 3287. Regret Analysis of Sleeping Competing Bandits
+
+- **ID:** `2603.19700` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Shinnosuke Uba, Yutaro Yamaguchi
+- **abs:** https://arxiv.org/abs/2603.19700 · **PDF:** https://arxiv.org/pdf/2603.19700v1
+
+## 3288. A Unified Phase-native Computational Principle Governs Hippocampal Spike Timing and Neural Coding
+
+- **ID:** `2603.19690` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-20
+- **Authors:** Reza Ahmadvand, Sara Safura Sharif, Yaser Mike Banad
+- **abs:** https://arxiv.org/abs/2603.19690 · **PDF:** https://arxiv.org/pdf/2603.19690v1
+
+## 3289. DataProphet: Demystifying Supervision Data Generalization in Multimodal LLMs
+
+- **ID:** `2603.19688` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Xuan Qi, Luxi He, Dan Roth, et al.
+- **abs:** https://arxiv.org/abs/2603.19688 · **PDF:** https://arxiv.org/pdf/2603.19688v1
+
+## 3290. Diminishing Returns in Expanding Generative Models and Godel-Tarski-Lob Limits
+
+- **ID:** `2603.19687` · **primary:** `cs.LO` · **Submitted:** 2026-03-20
+- **Authors:** Angshul Majumdar
+- **abs:** https://arxiv.org/abs/2603.19687 · **PDF:** https://arxiv.org/pdf/2603.19687v1
+
+## 3291. A Subgoal-driven Framework for Improving Long-Horizon LLM Agents
+
+- **ID:** `2603.19685` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Taiyi Wang, Sian Gooding, Florian Hartmann, et al.
+- **abs:** https://arxiv.org/abs/2603.19685 · **PDF:** https://arxiv.org/pdf/2603.19685v1
+
+## 3292. Ontology-Based Knowledge Modeling and Uncertainty-Aware Outdoor Air Quality Assessment Using Weighted Interval Type-2 Fuzzy Logic
+
+- **ID:** `2603.19683` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Md Inzmam, Ritesh Chandra, Sadhana Tiwari, et al.
+- **abs:** https://arxiv.org/abs/2603.19683 · **PDF:** https://arxiv.org/pdf/2603.19683v1
+
+## 3293. GoAgent: Group-of-Agents Communication Topology Generation for LLM-based Multi-Agent Systems
+
+- **ID:** `2603.19677` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Hongjiang Chen, Xin Zheng, Yixin Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.19677 · **PDF:** https://arxiv.org/pdf/2603.19677v1
+
+## 3294. ATHENA: Adaptive Test-Time Steering for Improving Count Fidelity in Diffusion Models
+
+- **ID:** `2603.19676` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Mohammad Shahab Sepehri, Asal Mehradfar, Berk Tinaz, et al.
+- **abs:** https://arxiv.org/abs/2603.19676 · **PDF:** https://arxiv.org/pdf/2603.19676v1
+
+## 3295. Logic-Gated Time-Shared Feedforward Networks for Alternating Finite Automata: Exact Simulation and Learnability
+
+- **ID:** `2604.01228` · **primary:** `cs.FL` · **Submitted:** 2026-03-20
+- **Authors:** Sahil Rajesh Dhayalkar
+- **abs:** https://arxiv.org/abs/2604.01228 · **PDF:** https://arxiv.org/pdf/2604.01228v1
+
+## 3296. Geometric Mixture-of-Experts with Curvature-Guided Adaptive Routing for Graph Representation Learning
+
+- **ID:** `2603.22317` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Haifang Cao, Yu Wang, Timing Li, et al.
+- **abs:** https://arxiv.org/abs/2603.22317 · **PDF:** https://arxiv.org/pdf/2603.22317v1
+
+## 3297. Wasserstein Propagation for Reverse Diffusion under Weak Log-Concavity: Exploiting Metric Mismatch via One-Switch Routing
+
+- **ID:** `2603.19670` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Zicheng Lyu, Zengfeng Huang
+- **abs:** https://arxiv.org/abs/2603.19670 · **PDF:** https://arxiv.org/pdf/2603.19670v2
+
+## 3298. Structured Prompting for Arabic Essay Proficiency: A Trait-Centric Evaluation Approach
+
+- **ID:** `2603.19668` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Salim Al Mandhari, Hieu Pham Dinh, Mo El-Haj, et al.
+- **abs:** https://arxiv.org/abs/2603.19668 · **PDF:** https://arxiv.org/pdf/2603.19668v1
+
+## 3299. Toward High-Fidelity Visual Reconstruction: From EEG-Based Conditioned Generation to Joint-Modal Guided Rebuilding
+
+- **ID:** `2603.19667` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Zhijian Gong, Tianren Yao, Wenjia Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.19667 · **PDF:** https://arxiv.org/pdf/2603.19667v1
+
+## 3300. The Residual Stream Is All You Need: On the Redundancy of the KV Cache in Transformer Inference
+
+- **ID:** `2603.19664` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Kaleem Ullah Qasim, Jiashu Zhang, Muhammad Kafeel Shaheen, et al.
+- **abs:** https://arxiv.org/abs/2603.19664 · **PDF:** https://arxiv.org/pdf/2603.19664v1
+
+## 3301. Model Selection and Parameter Estimation of Multi-dimensional Gaussian Mixture Model
+
+- **ID:** `2603.19657` · **primary:** `stat.ML` · **Submitted:** 2026-03-20
+- **Authors:** Xinyu Liu, Hai Zhang
+- **abs:** https://arxiv.org/abs/2603.19657 · **PDF:** https://arxiv.org/pdf/2603.19657v1
+
+## 3302. Ensembles-based Feature Guided Analysis
+
+- **ID:** `2603.19653` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Federico Formica, Stefano Gregis, Andrea Rota, et al.
+- **abs:** https://arxiv.org/abs/2603.19653 · **PDF:** https://arxiv.org/pdf/2603.19653v1
+
+## 3303. PolicySim: An LLM-Based Agent Social Simulation Sandbox for Proactive Policy Optimization
+
+- **ID:** `2603.19649` · **primary:** `cs.SI` · **Submitted:** 2026-03-20
+- **Authors:** Renhong Huang, Ning Tang, Jiarong Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.19649 · **PDF:** https://arxiv.org/pdf/2603.19649v1
+
+## 3304. Heavy-Tailed and Long-Range Dependent Noise in Stochastic Approximation: A Finite-Time Analysis
+
+- **ID:** `2603.19648` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Siddharth Chandak, Anuj Yadav, Ayfer Ozgur, et al.
+- **abs:** https://arxiv.org/abs/2603.19648 · **PDF:** https://arxiv.org/pdf/2603.19648v1
+
+## 3305. OmniDiT: Extending Diffusion Transformer to Omni-VTON Framework
+
+- **ID:** `2603.19643` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Weixuan Zeng, Pengcheng Wei, Huaiqing Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.19643 · **PDF:** https://arxiv.org/pdf/2603.19643v2
+
+## 3306. SEAR: Schema-Based Evaluation and Routing for LLM Gateways
+
+- **ID:** `2603.26728` · **primary:** `cs.DB` · **Submitted:** 2026-03-20
+- **Authors:** Zecheng Zhang, Han Zheng, Yue Xu
+- **abs:** https://arxiv.org/abs/2603.26728 · **PDF:** https://arxiv.org/pdf/2603.26728v1
+
+## 3307. HyEvo: Self-Evolving Hybrid Agentic Workflows for Efficient Reasoning
+
+- **ID:** `2603.19639` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Beibei Xu, Yutong Ye, Chuyun Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.19639 · **PDF:** https://arxiv.org/pdf/2603.19639v1
+
+## 3308. RiboSphere: Learning Unified and Efficient Representations of RNA Structures
+
+- **ID:** `2603.19636` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Zhou Zhang, Hanqun Cao, Cheng Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.19636 · **PDF:** https://arxiv.org/pdf/2603.19636v1
+
+## 3309. BEAVER: A Training-Free Hierarchical Prompt Compression Method via Structure-Aware Page Selection
+
+- **ID:** `2603.19635` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Zhengpei Hu, Kai Li, Dapeng Fu, et al.
+- **abs:** https://arxiv.org/abs/2603.19635 · **PDF:** https://arxiv.org/pdf/2603.19635v1
+
+## 3310. MetaCues: Enabling Critical Engagement with Generative AI for Information Seeking and Sensemaking
+
+- **ID:** `2603.19634` · **primary:** `cs.HC` · **Submitted:** 2026-03-20
+- **Authors:** Anjali Singh, Karan Taneja, Zhitong Guan, et al.
+- **abs:** https://arxiv.org/abs/2603.19634 · **PDF:** https://arxiv.org/pdf/2603.19634v1
+
+## 3311. Alternating Diffusion for Proximal Sampling with Zeroth Order Queries
+
+- **ID:** `2603.19633` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Hirohane Takagi, Atsushi Nitanda
+- **abs:** https://arxiv.org/abs/2603.19633 · **PDF:** https://arxiv.org/pdf/2603.19633v1
+
+## 3312. On the role of memorization in learned priors for geophysical inverse problems
+
+- **ID:** `2603.19629` · **primary:** `stat.ML` · **Submitted:** 2026-03-20
+- **Authors:** Ali Siahkoohi, Davide Sabeddu
+- **abs:** https://arxiv.org/abs/2603.19629 · **PDF:** https://arxiv.org/pdf/2603.19629v1
+
+## 3313. Dual Prompt-Driven Feature Encoding for Nighttime UAV Tracking
+
+- **ID:** `2603.19628` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Yiheng Wang, Changhong Fu, Liangliang Yao, et al.
+- **abs:** https://arxiv.org/abs/2603.19628 · **PDF:** https://arxiv.org/pdf/2603.19628v1
+
+## 3314. ST-GDance++: A Scalable Spatial-Temporal Diffusion for Long-Duration Group Choreography
+
+- **ID:** `2603.22316` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Jing Xu, Weiqiang Wang, Cunjian Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.22316 · **PDF:** https://arxiv.org/pdf/2603.22316v1
+
+## 3315. Continual Learning for Food Category Classification Dataset: Enhancing Model Adaptability and Performance
+
+- **ID:** `2603.19624` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Piyush Kaushik Bhattacharyya, Devansh Tomar, Shubham Mishra, et al.
+- **abs:** https://arxiv.org/abs/2603.19624 · **PDF:** https://arxiv.org/pdf/2603.19624v1
+
+## 3316. Forward and inverse problems for measure flows in Bayes Hilbert spaces
+
+- **ID:** `2603.20329` · **primary:** `stat.ML` · **Submitted:** 2026-03-20
+- **Authors:** S. David Mis, Maarten V. de Hoop
+- **abs:** https://arxiv.org/abs/2603.20329 · **PDF:** https://arxiv.org/pdf/2603.20329v1
+
+## 3317. DeepStock: Reinforcement Learning with Policy Regularizations for Inventory Management
+
+- **ID:** `2603.19621` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Yaqi Xie, Xinru Hao, Jiaxi Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.19621 · **PDF:** https://arxiv.org/pdf/2603.19621v1
+
+## 3318. On Performance Guarantees for Federated Learning with Personalized Constraints
+
+- **ID:** `2603.19617` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Mohammadjavad Ebrahimi, Daniel Burbano, Farzad Yousefian
+- **abs:** https://arxiv.org/abs/2603.19617 · **PDF:** https://arxiv.org/pdf/2603.19617v1
+
+## 3319. CAF-Score: Calibrating CLAP with LALMs for Reference-free Audio Captioning Evaluation
+
+- **ID:** `2603.19615` · **primary:** `cs.SD` · **Submitted:** 2026-03-20
+- **Authors:** Insung Lee, Taeyoung Jeong, Haejun Yoo, et al.
+- **abs:** https://arxiv.org/abs/2603.19615 · **PDF:** https://arxiv.org/pdf/2603.19615v1
+
+## 3320. Demonstrations, CoT, and Prompting: A Theoretical Analysis of ICL
+
+- **ID:** `2603.19611` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Xuhan Tong, Yuchen Zeng, Jiawei Zhang
+- **abs:** https://arxiv.org/abs/2603.19611 · **PDF:** https://arxiv.org/pdf/2603.19611v1
+
+## 3321. Emergency Preemption Without Online Exploration: A Decision Transformer Approach
+
+- **ID:** `2603.22315` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Haoran Su, Hanxiao Deng, Yandong Sun
+- **abs:** https://arxiv.org/abs/2603.22315 · **PDF:** https://arxiv.org/pdf/2603.22315v1
+
+## 3322. The Nonverbal Gap: Toward Affective Computer Vision for Safer and More Equitable Online Dating
+
+- **ID:** `2603.26727` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Ratna Kandala, Niva Manchanda, Akshata Kishore Moharir
+- **abs:** https://arxiv.org/abs/2603.26727 · **PDF:** https://arxiv.org/pdf/2603.26727v1
+
+## 3323. A Multimodal Deep Learning Framework for Edema Classification Using HCT and Clinical Data
+
+- **ID:** `2603.26726` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Aram Ansary Ogholbake, Hannah Choi, Spencer Brandenburg, et al.
+- **abs:** https://arxiv.org/abs/2603.26726 · **PDF:** https://arxiv.org/pdf/2603.26726v1
+
+## 3324. Capability Safety as Datalog: A Foundational Equivalence
+
+- **ID:** `2603.26725` · **primary:** `cs.LO` · **Submitted:** 2026-03-20
+- **Authors:** Cosimo Spera
+- **abs:** https://arxiv.org/abs/2603.26725 · **PDF:** https://arxiv.org/pdf/2603.26725v1
+
+## 3325. LoD-Loc v3: Generalized Aerial Localization in Dense Cities using Instance Silhouette Alignment
+
+- **ID:** `2603.19609` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Shuaibang Peng, Juelin Zhu, Xia Li, et al.
+- **abs:** https://arxiv.org/abs/2603.19609 · **PDF:** https://arxiv.org/pdf/2603.19609v2
+
+## 3326. FB-CLIP: Fine-Grained Zero-Shot Anomaly Detection with Foreground-Background Disentanglement
+
+- **ID:** `2603.19608` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Ming Hu, Yongsheng Huo, Mingyu Dou, et al.
+- **abs:** https://arxiv.org/abs/2603.19608 · **PDF:** https://arxiv.org/pdf/2603.19608v1
+
+## 3327. Enhancing AI-Based Tropical Cyclone Track and Intensity Forecasting via Systematic Bias Correction
+
+- **ID:** `2603.22314` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Peisong Niu, Haifan Zhang, Yang Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.22314 · **PDF:** https://arxiv.org/pdf/2603.22314v1
+
+## 3328. K-GMRF: Kinetic Gauss-Markov Random Field for First-Principles Covariance Tracking on Lie Groups
+
+- **ID:** `2603.19601` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** ZhiMing Li
+- **abs:** https://arxiv.org/abs/2603.19601 · **PDF:** https://arxiv.org/pdf/2603.19601v1
+
+## 3329. Physics-Informed Neural Network with Adaptive Clustering Learning Mechanism for Information Popularity Prediction
+
+- **ID:** `2603.19599` · **primary:** `cs.SI` · **Submitted:** 2026-03-20
+- **Authors:** Guangyin Jin, Xiaohan Ni, Yanjie Song, et al.
+- **abs:** https://arxiv.org/abs/2603.19599 · **PDF:** https://arxiv.org/pdf/2603.19599v1
+
+## 3330. Decorrelation, Diversity, and Emergent Intelligence: The Isomorphism Between Social Insect Colonies and Ensemble Machine Learning
+
+- **ID:** `2603.20328` · **primary:** `stat.ML` · **Submitted:** 2026-03-20
+- **Authors:** Ernest Fokoué, Gregory Babbitt, Yuval Levental
+- **abs:** https://arxiv.org/abs/2603.20328 · **PDF:** https://arxiv.org/pdf/2603.20328v2
+
+## 3331. All-Mem: Agentic Lifelong Memory via Dynamic Topology Evolution
+
+- **ID:** `2603.19595` · **primary:** `cs.IR` · **Submitted:** 2026-03-20
+- **Authors:** Can Lv, Heng Chang, Yuchen Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.19595 · **PDF:** https://arxiv.org/pdf/2603.19595v1
+
+## 3332. ARMOR: Adaptive Resilience Against Model Poisoning Attacks in Continual Federated Learning for Mobile Indoor Localization
+
+- **ID:** `2603.19594` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Danish Gufran, Akhil Singampalli, Sudeep Pasricha
+- **abs:** https://arxiv.org/abs/2603.19594 · **PDF:** https://arxiv.org/pdf/2603.19594v1
+
+## 3333. Smartphone-Based Identification of Unknown Liquids via Active Vibration Sensing
+
+- **ID:** `2603.28787` · **primary:** `eess.SP` · **Submitted:** 2026-03-20
+- **Authors:** Yongzhi Huang
+- **abs:** https://arxiv.org/abs/2603.28787 · **PDF:** https://arxiv.org/pdf/2603.28787v1
+
+## 3334. Data-driven ensemble prediction of the global ocean
+
+- **ID:** `2603.19591` · **primary:** `physics.ao-ph` · **Submitted:** 2026-03-20
+- **Authors:** Qiusheng Huang, Xiaohui Zhong, Anboyu Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.19591 · **PDF:** https://arxiv.org/pdf/2603.19591v1
+
+## 3335. A Multi-Modal CNN-LSTM Framework with Multi-Head Attention and Focal Loss for Real-Time Elderly Fall Detection
+
+- **ID:** `2603.22313` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Lijie Zhou, Luran Wang
+- **abs:** https://arxiv.org/abs/2603.22313 · **PDF:** https://arxiv.org/pdf/2603.22313v1
+
+## 3336. PowerLens: Taming LLM Agents for Safe and Personalized Mobile Power Management
+
+- **ID:** `2603.19584` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Xingyu Feng, Chang Sun, Yuzhu Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.19584 · **PDF:** https://arxiv.org/pdf/2603.19584v1
+
+## 3337. Skilled AI Agents for Embedded and IoT Systems Development
+
+- **ID:** `2603.19583` · **primary:** `cs.SE` · **Submitted:** 2026-03-20
+- **Authors:** Yiming Li, Yuhan Cheng, Mingchen Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.19583 · **PDF:** https://arxiv.org/pdf/2603.19583v1
+
+## 3338. Evolving Embodied Intelligence: Graph Neural Network--Driven Co-Design of Morphology and Control in Soft Robotics
+
+- **ID:** `2603.19582` · **primary:** `cs.RO` · **Submitted:** 2026-03-20
+- **Authors:** Jianqiang Wang, Shuaiqun Pan, Alvaro Serra-Gomez, et al.
+- **abs:** https://arxiv.org/abs/2603.19582 · **PDF:** https://arxiv.org/pdf/2603.19582v1
+
+## 3339. PA2D-MORL: Pareto Ascent Directional Decomposition based Multi-Objective Reinforcement Learning
+
+- **ID:** `2603.19579` · **primary:** `cs.AI` · **Submitted:** 2026-03-20
+- **Authors:** Tianmeng Hu, Biao Luo
+- **abs:** https://arxiv.org/abs/2603.19579 · **PDF:** https://arxiv.org/pdf/2603.19579v1
+
+## 3340. AI Psychosis: Does Conversational AI Amplify Delusion-Related Language?
+
+- **ID:** `2603.19574` · **primary:** `cs.HC` · **Submitted:** 2026-03-20
+- **Authors:** Soorya Ram Shimgekar, Vipin Gunda, Jiwon Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.19574 · **PDF:** https://arxiv.org/pdf/2603.19574v1
+
+## 3341. PFM-VEPAR: Prompting Foundation Models for RGB-Event Camera based Pedestrian Attribute Recognition
+
+- **ID:** `2603.19565` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Minghe Xu, Rouying Wu, ChiaWei Chu, et al.
+- **abs:** https://arxiv.org/abs/2603.19565 · **PDF:** https://arxiv.org/pdf/2603.19565v1
+
+## 3342. Wearable Foundation Models Should Go Beyond Static Encoders
+
+- **ID:** `2603.19564` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Yu Yvonne Wu, Yuwei Zhang, Hyungjun Yoon, et al.
+- **abs:** https://arxiv.org/abs/2603.19564 · **PDF:** https://arxiv.org/pdf/2603.19564v1
+
+## 3343. Dual-Domain Representation Alignment: Bridging 2D and 3D Vision via Geometry-Aware Architecture Search
+
+- **ID:** `2603.19563` · **primary:** `cs.CV` · **Submitted:** 2026-03-20
+- **Authors:** Haoyu Zhang, Zhihao Yu, Rui Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.19563 · **PDF:** https://arxiv.org/pdf/2603.19563v1
+
+## 3344. Neural Uncertainty Principle: A Unified View of Adversarial Fragility and LLM Hallucination
+
+- **ID:** `2603.19562` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Dong-Xiao Zhang, Hu Lou, Jun-Jie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.19562 · **PDF:** https://arxiv.org/pdf/2603.19562v3
+
+## 3345. An Adaptive Machine Learning Framework for Fluid Flow in Dual-Network Porous Media
+
+- **ID:** `2603.19561` · **primary:** `math.NA` · **Submitted:** 2026-03-20
+- **Authors:** V. S. Maduri, K. B. Nakshatrala
+- **abs:** https://arxiv.org/abs/2603.19561 · **PDF:** https://arxiv.org/pdf/2603.19561v1
+
+## 3346. Optimal Scalar Quantization for Matrix Multiplication: Closed-Form Density and Phase Transition
+
+- **ID:** `2603.19559` · **primary:** `cs.IT` · **Submitted:** 2026-03-20
+- **Authors:** Calvin Ang, Sungyoon Kim, Mert Pilanci
+- **abs:** https://arxiv.org/abs/2603.19559 · **PDF:** https://arxiv.org/pdf/2603.19559v1
+
+## 3347. TextReasoningBench: Does Reasoning Really Improve Text Classification in Large Language Models?
+
+- **ID:** `2603.19558` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Xinyu Guo, Yazhou Zhang, Jing Qin
+- **abs:** https://arxiv.org/abs/2603.19558 · **PDF:** https://arxiv.org/pdf/2603.19558v1
+
+## 3348. Probing the Latent World: Emergent Discrete Symbols and Physical Structure in Latent Representations
+
+- **ID:** `2603.20327` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Liu hung ming
+- **abs:** https://arxiv.org/abs/2603.20327 · **PDF:** https://arxiv.org/pdf/2603.20327v1
+
+## 3349. Learning to Bet for Horizon-Aware Anytime-Valid Testing
+
+- **ID:** `2603.19551` · **primary:** `stat.ME` · **Submitted:** 2026-03-20
+- **Authors:** Ege Onur Taga, Samet Oymak, Shubhanshu Shekhar
+- **abs:** https://arxiv.org/abs/2603.19551 · **PDF:** https://arxiv.org/pdf/2603.19551v1
+
+## 3350. Plagiarism or Productivity? Students Moral Disengagement and Behavioral Intentions to Use ChatGPT in Academic Writing
+
+- **ID:** `2603.19549` · **primary:** `cs.CY` · **Submitted:** 2026-03-20
+- **Authors:** John Paul P. Miranda, Rhiziel P. Manalese, Mark Anthony A. Castro, et al.
+- **abs:** https://arxiv.org/abs/2603.19549 · **PDF:** https://arxiv.org/pdf/2603.19549v1
+
+## 3351. AI in Work-Based Learning: Understanding the Purposes and Effects of Intelligent Tools Among Student Interns
+
+- **ID:** `2603.28786` · **primary:** `cs.CY` · **Submitted:** 2026-03-20
+- **Authors:** John Paul P. Miranda, Rhiziel P. Manalese, Sheila M. Geronimo, et al.
+- **abs:** https://arxiv.org/abs/2603.28786 · **PDF:** https://arxiv.org/pdf/2603.28786v1
+
+## 3352. Subspace Kernel Learning on Tensor Sequences
+
+- **ID:** `2603.19546` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Lei Wang, Xi Ding, Yongsheng Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.19546 · **PDF:** https://arxiv.org/pdf/2603.19546v1
+
+## 3353. Verifiable Error Bounds for Physics-Informed Neural Network Solutions of Lyapunov and Hamilton-Jacobi-Bellman Equations
+
+- **ID:** `2603.19545` · **primary:** `eess.SY` · **Submitted:** 2026-03-20
+- **Authors:** Jun Liu
+- **abs:** https://arxiv.org/abs/2603.19545 · **PDF:** https://arxiv.org/pdf/2603.19545v1
+
+## 3354. Scalable Cross-Facility Federated Learning for Scientific Foundation Models on Multiple Supercomputers
+
+- **ID:** `2603.19544` · **primary:** `cs.LG` · **Submitted:** 2026-03-20
+- **Authors:** Yijiang Li, Zilinghan Li, Kyle Chard, et al.
+- **abs:** https://arxiv.org/abs/2603.19544 · **PDF:** https://arxiv.org/pdf/2603.19544v1
+
+## 3355. When Agents Disagree: The Selection Bottleneck in Multi-Agent LLM Pipelines
+
+- **ID:** `2603.20324` · **primary:** `cs.MA` · **Submitted:** 2026-03-20
+- **Authors:** Artem Maryanskyy
+- **abs:** https://arxiv.org/abs/2603.20324 · **PDF:** https://arxiv.org/pdf/2603.20324v1
+
+## 3356. FDARxBench: Benchmarking Regulatory and Clinical Reasoning on FDA Generic Drug Assessment
+
+- **ID:** `2603.19539` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** Betty Xiong, Jillian Fisher, Benjamin Newman, et al.
+- **abs:** https://arxiv.org/abs/2603.19539 · **PDF:** https://arxiv.org/pdf/2603.19539v1
+
+## 3357. EvidenceRL: Reinforcing Evidence Consistency for Trustworthy Language Models
+
+- **ID:** `2603.19532` · **primary:** `cs.CL` · **Submitted:** 2026-03-20
+- **Authors:** J. Ben Tamo, Yuxing Lu, Benoit L. Marteau, et al.
+- **abs:** https://arxiv.org/abs/2603.19532 · **PDF:** https://arxiv.org/pdf/2603.19532v1
+
+## 3358. dinov3.seg: Open-Vocabulary Semantic Segmentation with DINOv3
+
+- **ID:** `2603.19531` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Saikat Dutta, Biplab Banerjee, Hamid Rezatofighi
+- **abs:** https://arxiv.org/abs/2603.19531 · **PDF:** https://arxiv.org/pdf/2603.19531v1
+
+## 3359. The Efficiency Attenuation Phenomenon: A Computational Challenge to the Language of Thought Hypothesis
+
+- **ID:** `2603.22312` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Di Zhang
+- **abs:** https://arxiv.org/abs/2603.22312 · **PDF:** https://arxiv.org/pdf/2603.22312v1
+
+## 3360. SurfaceXR: Fusing Smartwatch IMUs and Egocentric Hand Pose for Seamless Surface Interactions
+
+- **ID:** `2603.19529` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Vasco Xu, Brian Chen, Eric J. Gonzalez, et al.
+- **abs:** https://arxiv.org/abs/2603.19529 · **PDF:** https://arxiv.org/pdf/2603.19529v1
+
+## 3361. Depictions of Depression in Generative AI Video Models: A Preliminary Study of OpenAI's Sora 2
+
+- **ID:** `2603.19527` · **primary:** `cs.CY` · **Submitted:** 2026-03-19
+- **Authors:** Matthew Flathers, Griffin Smith, Julian Herpertz, et al.
+- **abs:** https://arxiv.org/abs/2603.19527 · **PDF:** https://arxiv.org/pdf/2603.19527v1
+
+## 3362. GIP-RAG: An Evidence-Grounded Retrieval-Augmented Framework for Interpretable Gene Interaction and Pathway Impact Analysis
+
+- **ID:** `2603.20321` · **primary:** `q-bio.MN` · **Submitted:** 2026-03-19
+- **Authors:** Fujian Jia, Jiwen Gu, Cheng Lu, et al.
+- **abs:** https://arxiv.org/abs/2603.20321 · **PDF:** https://arxiv.org/pdf/2603.20321v1
+
+## 3363. The Causal Impact of Tool Affordance on Safety Alignment in LLM Agents
+
+- **ID:** `2603.20320` · **primary:** `cs.SE` · **Submitted:** 2026-03-19
+- **Authors:** Shasha Yu, Fiona Carroll, Barry L. Bentley
+- **abs:** https://arxiv.org/abs/2603.20320 · **PDF:** https://arxiv.org/pdf/2603.20320v1
+
+## 3364. Inducing Sustained Creativity and Diversity in Large Language Models
+
+- **ID:** `2603.19519` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Queenie Luo, Gary King, Michael Puett, et al.
+- **abs:** https://arxiv.org/abs/2603.19519 · **PDF:** https://arxiv.org/pdf/2603.19519v2
+
+## 3365. ReXInTheWild: A Unified Benchmark for Medical Photograph Understanding
+
+- **ID:** `2603.19517` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Oishi Banerjee, Sung Eun Kim, Alexandra N. Willauer, et al.
+- **abs:** https://arxiv.org/abs/2603.19517 · **PDF:** https://arxiv.org/pdf/2603.19517v1
+
+## 3366. Gastric-X: A Multimodal Multi-Phase Benchmark Dataset for Advancing Vision-Language Models in Gastric Cancer Analysis
+
+- **ID:** `2603.19516` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Sheng Lu, Hao Chen, Rui Yin, et al.
+- **abs:** https://arxiv.org/abs/2603.19516 · **PDF:** https://arxiv.org/pdf/2603.19516v2
+
+## 3367. ItinBench: Benchmarking Planning Across Multiple Cognitive Dimensions with Large Language Models
+
+- **ID:** `2603.19515` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Tianlong Wang, Pinqiao Wang, Weili Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.19515 · **PDF:** https://arxiv.org/pdf/2603.19515v1
+
+## 3368. Learning to Disprove: Formal Counterexample Generation with Large Language Models
+
+- **ID:** `2603.19514` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Zenan Li, Zhaoyu Li, Kaiyu Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.19514 · **PDF:** https://arxiv.org/pdf/2603.19514v1
+
+## 3369. FedAgain: A Trust-Based and Robust Federated Learning Strategy for an Automated Kidney Stone Identification in Ureteroscopy
+
+- **ID:** `2603.19512` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Ivan Reyes-Amezcua, Francisco Lopez-Tiro, Clément Larose, et al.
+- **abs:** https://arxiv.org/abs/2603.19512 · **PDF:** https://arxiv.org/pdf/2603.19512v1
+
+## 3370. Linear Social Choice with Few Queries: A Moment-Based Approach
+
+- **ID:** `2603.19510` · **primary:** `cs.GT` · **Submitted:** 2026-03-19
+- **Authors:** Luise Ge, Daniel Halpern, Gregory Kehne, et al.
+- **abs:** https://arxiv.org/abs/2603.19510 · **PDF:** https://arxiv.org/pdf/2603.19510v1
+
+## 3371. Beyond the Desk: Barriers and Future Opportunities for AI to Assist Scientists in Embodied Physical Tasks
+
+- **ID:** `2603.19504` · **primary:** `cs.HC` · **Submitted:** 2026-03-19
+- **Authors:** Irene Hou, Alexander Qin, Lauren Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.19504 · **PDF:** https://arxiv.org/pdf/2603.19504v1
+
+## 3372. Stochastic Sequential Decision Making over Expanding Networks with Graph Filtering
+
+- **ID:** `2603.19501` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Zhan Gao, Bishwadeep Das, Elvin Isufi
+- **abs:** https://arxiv.org/abs/2603.19501 · **PDF:** https://arxiv.org/pdf/2603.19501v1
+
+## 3373. Teaching an Agent to Sketch One Part at a Time
+
+- **ID:** `2603.19500` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Xiaodan Du, Ruize Xu, David Yunis, et al.
+- **abs:** https://arxiv.org/abs/2603.19500 · **PDF:** https://arxiv.org/pdf/2603.19500v1
+
+## 3374. Bypassing Document Ingestion: An MCP Approach to Financial Q&A
+
+- **ID:** `2603.20316` · **primary:** `cs.IR` · **Submitted:** 2026-03-19
+- **Authors:** Sasan Mansouri, Edoardo Pilla, Mark Wahrenburg, et al.
+- **abs:** https://arxiv.org/abs/2603.20316 · **PDF:** https://arxiv.org/pdf/2603.20316v1
+
+## 3375. ICLAD: In-Context Learning for Unified Tabular Anomaly Detection Across Supervision Regimes
+
+- **ID:** `2603.19497` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Jack Yi Wei, Narges Armanfard
+- **abs:** https://arxiv.org/abs/2603.19497 · **PDF:** https://arxiv.org/pdf/2603.19497v1
+
+## 3376. Any-Subgroup Equivariant Networks via Symmetry Breaking
+
+- **ID:** `2603.19486` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Abhinav Goel, Derek Lim, Hannah Lawrence, et al.
+- **abs:** https://arxiv.org/abs/2603.19486 · **PDF:** https://arxiv.org/pdf/2603.19486v1
+
+## 3377. Interpretable liquid crystal phase classification via two-by-two ordinal patterns
+
+- **ID:** `2603.26723` · **primary:** `cond-mat.soft` · **Submitted:** 2026-03-19
+- **Authors:** Leonardo G. J. M. Voltarelli, Natalia Osiecka-Drewniak, Marcin Piwowarczyk, et al.
+- **abs:** https://arxiv.org/abs/2603.26723 · **PDF:** https://arxiv.org/pdf/2603.26723v1
+
+## 3378. TRACE: Trajectory Recovery with State Propagation Diffusion for Urban Mobility
+
+- **ID:** `2603.19474` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Jinming Wang, Hai Wang, Hongkai Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.19474 · **PDF:** https://arxiv.org/pdf/2603.19474v1
+
+## 3379. Reinforcement-guided generative protein language models enable de novo design of highly diverse AAV capsids
+
+- **ID:** `2603.19473` · **primary:** `q-bio.BM` · **Submitted:** 2026-03-19
+- **Authors:** Lucas Ferraz, Ana F. Rodrigues, Pedro Giesteira Cotovio, et al.
+- **abs:** https://arxiv.org/abs/2603.19473 · **PDF:** https://arxiv.org/pdf/2603.19473v1
+
+## 3380. Adaptive Layerwise Perturbation: Unifying Off-Policy Corrections for LLM RL
+
+- **ID:** `2603.19470` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Chenlu Ye, Xuanchang Zhang, Yifan Hao, et al.
+- **abs:** https://arxiv.org/abs/2603.19470 · **PDF:** https://arxiv.org/pdf/2603.19470v1
+
+## 3381. A Framework for Formalizing LLM Agent Security
+
+- **ID:** `2603.19469` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Vincent Siu, Jingxuan He, Kyle Montgomery, et al.
+- **abs:** https://arxiv.org/abs/2603.19469 · **PDF:** https://arxiv.org/pdf/2603.19469v1
+
+## 3382. Global Convergence of Multiplicative Updates for the Matrix Mechanism: A Collaborative Proof with Gemini 3
+
+- **ID:** `2603.19465` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Keith Rush
+- **abs:** https://arxiv.org/abs/2603.19465 · **PDF:** https://arxiv.org/pdf/2603.19465v2
+
+## 3383. Deep Hilbert--Galerkin Methods for Infinite-Dimensional PDEs and Optimal Control
+
+- **ID:** `2603.19463` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Samuel N. Cohen, Filippo de Feo, Jackson Hebner, et al.
+- **abs:** https://arxiv.org/abs/2603.19463 · **PDF:** https://arxiv.org/pdf/2603.19463v1
+
+## 3384. Hyperagents
+
+- **ID:** `2603.19461` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Jenny Zhang, Bingchen Zhao, Wannan Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.19461 · **PDF:** https://arxiv.org/pdf/2603.19461v1
+
+## 3385. GeoLAN: Geometric Learning of Latent Explanatory Directions in Large Language Models
+
+- **ID:** `2603.19460` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Tianyu Bell Pan, Damon L. Woodard
+- **abs:** https://arxiv.org/abs/2603.19460 · **PDF:** https://arxiv.org/pdf/2603.19460v1
+
+## 3386. Cooperation and Exploitation in LLM Policy Synthesis for Sequential Social Dilemmas
+
+- **ID:** `2603.19453` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Víctor Gallego
+- **abs:** https://arxiv.org/abs/2603.19453 · **PDF:** https://arxiv.org/pdf/2603.19453v1
+
+## 3387. TrustFlow: Topic-Aware Vector Reputation Propagation for Multi-Agent Ecosystems
+
+- **ID:** `2603.19452` · **primary:** `cs.MA` · **Submitted:** 2026-03-19
+- **Authors:** Volodymyr Seliuchenko
+- **abs:** https://arxiv.org/abs/2603.19452 · **PDF:** https://arxiv.org/pdf/2603.19452v1
+
+## 3388. LoFi: Location-Aware Fine-Grained Representation Learning for Chest X-ray
+
+- **ID:** `2603.19451` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Myeongkyun Kang, Yanting Yang, Xiaoxiao Li
+- **abs:** https://arxiv.org/abs/2603.19451 · **PDF:** https://arxiv.org/pdf/2603.19451v1
+
+## 3389. Near-Equivalent Q-learning Policies for Dynamic Treatment Regimes
+
+- **ID:** `2603.19440` · **primary:** `stat.ML` · **Submitted:** 2026-03-19
+- **Authors:** Sophia Yazzourh, Erica E. M. Moodie
+- **abs:** https://arxiv.org/abs/2603.19440 · **PDF:** https://arxiv.org/pdf/2603.19440v1
+
+## 3390. Subspace Projection Methods for Fast Spectral Embeddings of Evolving Graphs
+
+- **ID:** `2603.19439` · **primary:** `stat.ML` · **Submitted:** 2026-03-19
+- **Authors:** Mohammad Eini, Abdullah Karaaslanli, Vassilis Kalantzis, et al.
+- **abs:** https://arxiv.org/abs/2603.19439 · **PDF:** https://arxiv.org/pdf/2603.19439v1
+
+## 3391. When both Grounding and not Grounding are Bad -- A Partially Grounded Encoding of Planning into SAT (Extended Version)
+
+- **ID:** `2603.19429` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** João Filipe, Gregor Behnke
+- **abs:** https://arxiv.org/abs/2603.19429 · **PDF:** https://arxiv.org/pdf/2603.19429v1
+
+## 3392. Vocabulary shapes cross-lingual variation of word-order learnability in language models
+
+- **ID:** `2603.19427` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Jonas Mayer Martins, Jaap Jumelet, Viola Priesemann, et al.
+- **abs:** https://arxiv.org/abs/2603.19427 · **PDF:** https://arxiv.org/pdf/2603.19427v1
+
+## 3393. Is Evaluation Awareness Just Format Sensitivity? Limitations of Probe-Based Evidence under Controlled Prompt Structure
+
+- **ID:** `2603.19426` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Viliana Devbunova
+- **abs:** https://arxiv.org/abs/2603.19426 · **PDF:** https://arxiv.org/pdf/2603.19426v1
+
+## 3394. The Autonomy Tax: Defense Training Breaks LLM Agents
+
+- **ID:** `2603.19423` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Shawn Li, Yue Zhao
+- **abs:** https://arxiv.org/abs/2603.19423 · **PDF:** https://arxiv.org/pdf/2603.19423v1
+
+## 3395. Pseudo-Labeling for Unsupervised Domain Adaptation with Kernel GLMs
+
+- **ID:** `2603.19422` · **primary:** `stat.ML` · **Submitted:** 2026-03-19
+- **Authors:** Nathan Weill, Kaizheng Wang
+- **abs:** https://arxiv.org/abs/2603.19422 · **PDF:** https://arxiv.org/pdf/2603.19422v1
+
+## 3396. Investigating In-Context Privacy Learning by Integrating User-Facing Privacy Tools into Conversational Agents
+
+- **ID:** `2603.19416` · **primary:** `cs.HC` · **Submitted:** 2026-03-19
+- **Authors:** Mohammad Hadi Nezhad, Francisco Enrique Vicente Castro, Ivon Arroyo
+- **abs:** https://arxiv.org/abs/2603.19416 · **PDF:** https://arxiv.org/pdf/2603.19416v1
+
+## 3397. Scalable Prompt Routing via Fine-Grained Latent Task Discovery
+
+- **ID:** `2603.19415` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Yunyi Zhang, Soji Adeshina, Sheng Guan, et al.
+- **abs:** https://arxiv.org/abs/2603.19415 · **PDF:** https://arxiv.org/pdf/2603.19415v2
+
+## 3398. Rolling-Origin Validation Reverses Model Rankings in Multi-Step PM10 Forecasting: XGBoost, SARIMA, and Persistence
+
+- **ID:** `2603.20315` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Federico Garcia Crespi, Eduardo Yubero Funes, Marina Alfosea Simon
+- **abs:** https://arxiv.org/abs/2603.20315 · **PDF:** https://arxiv.org/pdf/2603.20315v1
+
+## 3399. Optimizing Resource-Constrained Non-Pharmaceutical Interventions for Multi-Cluster Outbreak Control Using Hierarchical Reinforcement Learning
+
+- **ID:** `2603.19397` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Xueqiao Peng, Andrew Perrault
+- **abs:** https://arxiv.org/abs/2603.19397 · **PDF:** https://arxiv.org/pdf/2603.19397v1
+
+## 3400. Bridging Conformal Prediction and Scenario Optimization: Discarded Constraints and Modular Risk Allocation
+
+- **ID:** `2603.19396` · **primary:** `eess.SY` · **Submitted:** 2026-03-19
+- **Authors:** Giuseppe C. Calafiore
+- **abs:** https://arxiv.org/abs/2603.19396 · **PDF:** https://arxiv.org/pdf/2603.19396v1
+
+## 3401. VGS-Decoding: Visual Grounding Score Guided Decoding for Hallucination Mitigation in Medical VLMs
+
+- **ID:** `2603.20314` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Govinda Kolli, Adinath Madhavrao Dukre, Behzad Bozorgtabar, et al.
+- **abs:** https://arxiv.org/abs/2603.20314 · **PDF:** https://arxiv.org/pdf/2603.20314v1
+
+## 3402. TuLaBM: Tumor-Biased Latent Bridge Matching for Contrast-Enhanced MRI Synthesis
+
+- **ID:** `2603.19386` · **primary:** `eess.IV` · **Submitted:** 2026-03-19
+- **Authors:** Atharva Rege, Adinath Madhavrao Dukre, Numan Balci, et al.
+- **abs:** https://arxiv.org/abs/2603.19386 · **PDF:** https://arxiv.org/pdf/2603.19386v1
+
+## 3403. Semantic Tool Discovery for Large Language Models: A Vector-Based Approach to MCP Tool Selection
+
+- **ID:** `2603.20313` · **primary:** `cs.SE` · **Submitted:** 2026-03-19
+- **Authors:** Sarat Mudunuri, Jian Wan, Ally Qin, et al.
+- **abs:** https://arxiv.org/abs/2603.20313 · **PDF:** https://arxiv.org/pdf/2603.20313v1
+
+## 3404. Automated Membership Inference Attacks: Discovering MIA Signal Computations using LLM Agents
+
+- **ID:** `2603.19375` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Toan Tran, Olivera Kotevska, Li Xiong
+- **abs:** https://arxiv.org/abs/2603.19375 · **PDF:** https://arxiv.org/pdf/2603.19375v1
+
+## 3405. Warm-Start Flow Matching for Guaranteed Fast Text/Image Generation
+
+- **ID:** `2603.19360` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Minyoung Kim
+- **abs:** https://arxiv.org/abs/2603.19360 · **PDF:** https://arxiv.org/pdf/2603.19360v1
+
+## 3406. NavTrust: Benchmarking Trustworthiness for Embodied Navigation
+
+- **ID:** `2603.19229` · **primary:** `cs.RO` · **Submitted:** 2026-03-19
+- **Authors:** Huaide Jiang, Yash Chaudhary, Yuping Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.19229 · **PDF:** https://arxiv.org/pdf/2603.19229v1
+
+## 3407. FinTradeBench: A Financial Reasoning Benchmark for LLMs
+
+- **ID:** `2603.19225` · **primary:** `cs.CE` · **Submitted:** 2026-03-19
+- **Authors:** Yogesh Agrawal, Aniruddha Dutta, Md Mahadi Hasan, et al.
+- **abs:** https://arxiv.org/abs/2603.19225 · **PDF:** https://arxiv.org/pdf/2603.19225v2
+
+## 3408. F2LLM-v2: Inclusive, Performant, and Efficient Embeddings for a Multilingual World
+
+- **ID:** `2603.19223` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Ziyin Zhang, Zihan Liao, Hang Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.19223 · **PDF:** https://arxiv.org/pdf/2603.19223v1
+
+## 3409. Spectrally-Guided Diffusion Noise Schedules
+
+- **ID:** `2603.19222` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Carlos Esteves, Ameesh Makadia
+- **abs:** https://arxiv.org/abs/2603.19222 · **PDF:** https://arxiv.org/pdf/2603.19222v1
+
+## 3410. Online Learning and Equilibrium Computation with Ranking Feedback
+
+- **ID:** `2603.19221` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Mingyang Liu, Yongshan Chen, Zhiyuan Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.19221 · **PDF:** https://arxiv.org/pdf/2603.19221v1
+
+## 3411. Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation
+
+- **ID:** `2603.19220` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Zhuolin Yang, Zihan Liu, Yang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.19220 · **PDF:** https://arxiv.org/pdf/2603.19220v2
+
+## 3412. DriveTok: 3D Driving Scene Tokenization for Unified Multi-View Reconstruction and Understanding
+
+- **ID:** `2603.19219` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Dong Zhuo, Wenzhao Zheng, Sicheng Zuo, et al.
+- **abs:** https://arxiv.org/abs/2603.19219 · **PDF:** https://arxiv.org/pdf/2603.19219v1
+
+## 3413. DreamPartGen: Semantically Grounded Part-Level 3D Generation via Collaborative Latent Denoising
+
+- **ID:** `2603.19216` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Tianjiao Yu, Xinzhuo Li, Muntasir Wahed, et al.
+- **abs:** https://arxiv.org/abs/2603.19216 · **PDF:** https://arxiv.org/pdf/2603.19216v1
+
+## 3414. $R$-equivalence on Cubic Surfaces I: Existing Cases with Non-Trivial Universal Equivalence
+
+- **ID:** `2603.19215` · **primary:** `math.AG` · **Submitted:** 2026-03-19
+- **Authors:** Dimitri Kanevsky, Julian Salazar, Matt Harvey
+- **abs:** https://arxiv.org/abs/2603.19215 · **PDF:** https://arxiv.org/pdf/2603.19215v1
+
+## 3415. Do VLMs Need Vision Transformers? Evaluating State Space Models as Vision Encoders
+
+- **ID:** `2603.19209` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Shang-Jui Ray Kuo, Paola Cascante-Bonilla
+- **abs:** https://arxiv.org/abs/2603.19209 · **PDF:** https://arxiv.org/pdf/2603.19209v1
+
+## 3416. Robustness, Cost, and Attack-Surface Concentration in Phishing Detection
+
+- **ID:** `2603.19204` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Julian Allagan, Mohamed Elbakary, Zohreh Safari, et al.
+- **abs:** https://arxiv.org/abs/2603.19204 · **PDF:** https://arxiv.org/pdf/2603.19204v1
+
+## 3417. kRAIG: A Natural Language-Driven Agent for Automated DataOps Pipeline Generation
+
+- **ID:** `2603.20311` · **primary:** `cs.SE` · **Submitted:** 2026-03-19
+- **Authors:** Rohan Siva, Kai Cheung, Lichi Li, et al.
+- **abs:** https://arxiv.org/abs/2603.20311 · **PDF:** https://arxiv.org/pdf/2603.20311v1
+
+## 3418. The Exponentially Weighted Signature
+
+- **ID:** `2603.19198` · **primary:** `stat.ML` · **Submitted:** 2026-03-19
+- **Authors:** Alexandre Bloch, Samuel N. Cohen, Terry Lyons, et al.
+- **abs:** https://arxiv.org/abs/2603.19198 · **PDF:** https://arxiv.org/pdf/2603.19198v1
+
+## 3419. A Novel Solution for Zero-Day Attack Detection in IDS using Self-Attention and Jensen-Shannon Divergence in WGAN-GP
+
+- **ID:** `2603.19350` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Ziyu Mu, Xiyu Shi, Safak Dogan
+- **abs:** https://arxiv.org/abs/2603.19350 · **PDF:** https://arxiv.org/pdf/2603.19350v1
+
+## 3420. How Auditory Knowledge in LLM Backbones Shapes Audio Language Models: A Holistic Evaluation
+
+- **ID:** `2603.19195` · **primary:** `eess.AS` · **Submitted:** 2026-03-19
+- **Authors:** Ke-Han Lu, Szu-Wei Fu, Chao-Han Huck Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.19195 · **PDF:** https://arxiv.org/pdf/2603.19195v1
+
+## 3421. OS-Themis: A Scalable Critic Framework for Generalist GUI Rewards
+
+- **ID:** `2603.19191` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Zehao Li, Zhenyu Wu, Yibo Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.19191 · **PDF:** https://arxiv.org/pdf/2603.19191v1
+
+## 3422. Improving RCT-Based Treatment Effect Estimation Under Covariate Mismatch via Calibrated Alignment
+
+- **ID:** `2603.19186` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Amir Asiaee, Samhita Pal
+- **abs:** https://arxiv.org/abs/2603.19186 · **PDF:** https://arxiv.org/pdf/2603.19186v1
+
+## 3423. MIDST Challenge at SaTML 2025: Membership Inference over Diffusion-models-based Synthetic Tabular data
+
+- **ID:** `2603.19185` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Masoumeh Shafieinejad, Xi He, Mahshid Alinoori, et al.
+- **abs:** https://arxiv.org/abs/2603.19185 · **PDF:** https://arxiv.org/pdf/2603.19185v1
+
+## 3424. Box Maze: A Process-Control Architecture for Reliable LLM Reasoning
+
+- **ID:** `2603.19182` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Zou Qiang
+- **abs:** https://arxiv.org/abs/2603.19182 · **PDF:** https://arxiv.org/pdf/2603.19182v1
+
+## 3425. SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limits
+
+- **ID:** `2603.19173` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Edward Lin, Sahil Modi, Siva Kumar Sastry Hari, et al.
+- **abs:** https://arxiv.org/abs/2603.19173 · **PDF:** https://arxiv.org/pdf/2603.19173v1
+
+## 3426. DyMoE: Dynamic Expert Orchestration with Mixed-Precision Quantization for Efficient MoE Inference on Edge
+
+- **ID:** `2603.19172` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Yuegui Huang, Zhiyuan Fang, Weiqi Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.19172 · **PDF:** https://arxiv.org/pdf/2603.19172v1
+
+## 3427. ARIADNE: A Perception-Reasoning Synergy Framework for Trustworthy Coronary Angiography Analysis
+
+- **ID:** `2603.19169` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Zhan Jin, Yu Luo, Yizhou Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.19169 · **PDF:** https://arxiv.org/pdf/2603.19169v1
+
+## 3428. Evaluating Counterfactual Strategic Reasoning in Large Language Models
+
+- **ID:** `2603.19167` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Dimitrios Georgousis, Maria Lymperaiou, Angeliki Dimitriou, et al.
+- **abs:** https://arxiv.org/abs/2603.19167 · **PDF:** https://arxiv.org/pdf/2603.19167v1
+
+## 3429. Meanings and Measurements: Multi-Agent Probabilistic Grounding for Vision-Language Navigation
+
+- **ID:** `2603.19166` · **primary:** `cs.RO` · **Submitted:** 2026-03-19
+- **Authors:** Swagat Padhan, Lakshya Jain, Bhavya Minesh Shah, et al.
+- **abs:** https://arxiv.org/abs/2603.19166 · **PDF:** https://arxiv.org/pdf/2603.19166v1
+
+## 3430. Rigorous Error Certification for Neural PDE Solvers: From Empirical Residuals to Solution Guarantees
+
+- **ID:** `2603.19165` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Amartya Mukherjee, Maxwell Fitzsimmons, David C. Del Rey Fernández, et al.
+- **abs:** https://arxiv.org/abs/2603.19165 · **PDF:** https://arxiv.org/pdf/2603.19165v1
+
+## 3431. cuGenOpt: A GPU-Accelerated General-Purpose Metaheuristic Framework for Combinatorial Optimization
+
+- **ID:** `2603.19163` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Yuyang Liu
+- **abs:** https://arxiv.org/abs/2603.19163 · **PDF:** https://arxiv.org/pdf/2603.19163v1
+
+## 3432. PPI is the Difference Estimator: Recognizing the Survey Sampling Roots of Prediction-Powered Inference
+
+- **ID:** `2603.19160` · **primary:** `stat.ME` · **Submitted:** 2026-03-19
+- **Authors:** Reagan Mozer
+- **abs:** https://arxiv.org/abs/2603.19160 · **PDF:** https://arxiv.org/pdf/2603.19160v1
+
+## 3433. VEPO: Variable Entropy Policy Optimization for Low-Resource Language Foundation Models
+
+- **ID:** `2603.19152` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Chonghan Liu, Yimin Du, Qi An, et al.
+- **abs:** https://arxiv.org/abs/2603.19152 · **PDF:** https://arxiv.org/pdf/2603.19152v1
+
+## 3434. Optimal Splitting of Language Models from Mixtures to Specialized Domains
+
+- **ID:** `2603.19149` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Skyler Seto, Pierre Ablin, Anastasiia Filippova, et al.
+- **abs:** https://arxiv.org/abs/2603.19149 · **PDF:** https://arxiv.org/pdf/2603.19149v1
+
+## 3435. Fast and Effective Computation of Generalized Symmetric Matrix Factorization
+
+- **ID:** `2603.19147` · **primary:** `math.OC` · **Submitted:** 2026-03-19
+- **Authors:** Lei Yang, Han Wan, Min Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.19147 · **PDF:** https://arxiv.org/pdf/2603.19147v1
+
+## 3436. D5P4: Partition Determinantal Point Process for Diversity in Parallel Discrete Diffusion Decoding
+
+- **ID:** `2603.19146` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Jonathan Lys, Vincent Gripon, Bastien Pasdeloup, et al.
+- **abs:** https://arxiv.org/abs/2603.19146 · **PDF:** https://arxiv.org/pdf/2603.19146v1
+
+## 3437. Enhancing Pretrained Model-based Continual Representation Learning via Guided Random Projection
+
+- **ID:** `2603.19145` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Ruilin Li, Heming Zou, Xiufeng Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.19145 · **PDF:** https://arxiv.org/pdf/2603.19145v1
+
+## 3438. A Mathematical Theory of Understanding
+
+- **ID:** `2603.19349` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Bahar Taşkesen
+- **abs:** https://arxiv.org/abs/2603.19349 · **PDF:** https://arxiv.org/pdf/2603.19349v1
+
+## 3439. UGID: Unified Graph Isomorphism for Debiasing Large Language Models
+
+- **ID:** `2603.19144` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Zikang Ding, Junchi Yao, Junhao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.19144 · **PDF:** https://arxiv.org/pdf/2603.19144v1
+
+## 3440. Anatomical Heterogeneity in Transformer Language Models
+
+- **ID:** `2603.19348` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Tomasz Wietrzykowski
+- **abs:** https://arxiv.org/abs/2603.19348 · **PDF:** https://arxiv.org/pdf/2603.19348v1
+
+## 3441. SHAPCA: Consistent and Interpretable Explanations for Machine Learning Models on Spectroscopy Data
+
+- **ID:** `2603.19141` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Mingxing Zhang, Nicola Rossberg, Simone Innocente, et al.
+- **abs:** https://arxiv.org/abs/2603.19141 · **PDF:** https://arxiv.org/pdf/2603.19141v1
+
+## 3442. Hierarchical Latent Structure Learning through Online Inference
+
+- **ID:** `2603.19139` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Ines Aitsahalia, Kiyohito Iigaya
+- **abs:** https://arxiv.org/abs/2603.19139 · **PDF:** https://arxiv.org/pdf/2603.19139v1
+
+## 3443. Implicit Patterns in LLM-Based Binary Analysis
+
+- **ID:** `2603.19138` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Qiang Li, XiangRui Zhang, Haining Wang
+- **abs:** https://arxiv.org/abs/2603.19138 · **PDF:** https://arxiv.org/pdf/2603.19138v1
+
+## 3444. Adaptive Regime-Aware Stock Price Prediction Using Autoencoder-Gated Dual Node Transformers with Reinforcement Learning Control
+
+- **ID:** `2603.19136` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Mohammad Al Ridhawi, Mahtab Haj Ali, Hussein Al Osman
+- **abs:** https://arxiv.org/abs/2603.19136 · **PDF:** https://arxiv.org/pdf/2603.19136v2
+
+## 3445. From Inference Efficiency to Embodied Efficiency: Revisiting Efficiency Metrics for Vision-Language-Action Models
+
+- **ID:** `2603.19131` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Zhuofan Li, Hongkun Yang, Zhenyang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.19131 · **PDF:** https://arxiv.org/pdf/2603.19131v1
+
+## 3446. On Optimizing Multimodal Jailbreaks for Spoken Language Models
+
+- **ID:** `2603.19127` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Aravind Krishnan, Karolina Stańczak, Dietrich Klakow
+- **abs:** https://arxiv.org/abs/2603.19127 · **PDF:** https://arxiv.org/pdf/2603.19127v1
+
+## 3447. Exploring the Agentic Frontier of Verilog Code Generation
+
+- **ID:** `2603.19347` · **primary:** `cs.AR` · **Submitted:** 2026-03-19
+- **Authors:** Patrick Yubeaton, Siddharth Garg, Chinmay Hegde
+- **abs:** https://arxiv.org/abs/2603.19347 · **PDF:** https://arxiv.org/pdf/2603.19347v3
+
+## 3448. CustomTex: High-fidelity Indoor Scene Texturing via Multi-Reference Customization
+
+- **ID:** `2603.19121` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Weilin Chen, Jiahao Rao, Wenhao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.19121 · **PDF:** https://arxiv.org/pdf/2603.19121v2
+
+## 3449. How Uncertainty Estimation Scales with Sampling in Reasoning Models
+
+- **ID:** `2603.19118` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Maksym Del, Markus Kängsepp, Marharyta Domnich, et al.
+- **abs:** https://arxiv.org/abs/2603.19118 · **PDF:** https://arxiv.org/pdf/2603.19118v1
+
+## 3450. Numerical Considerations for the Construction of Karhunen-Loève Expansions
+
+- **ID:** `2603.19108` · **primary:** `math.NA` · **Submitted:** 2026-03-19
+- **Authors:** Cosmin Safta, Habib N. Najm
+- **abs:** https://arxiv.org/abs/2603.19108 · **PDF:** https://arxiv.org/pdf/2603.19108v1
+
+## 3451. FedTrident: Resilient Road Condition Classification Against Poisoning Attacks in Federated Learning
+
+- **ID:** `2603.19101` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Sheng Liu, Panos Papadimitratos
+- **abs:** https://arxiv.org/abs/2603.19101 · **PDF:** https://arxiv.org/pdf/2603.19101v1
+
+## 3452. LuMamba: Latent Unified Mamba for Electrode Topology-Invariant and Efficient EEG Modeling
+
+- **ID:** `2603.19100` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Danaé Broustail, Anna Tegon, Thorir Mar Ingolfsson, et al.
+- **abs:** https://arxiv.org/abs/2603.19100 · **PDF:** https://arxiv.org/pdf/2603.19100v1
+
+## 3453. DaPT: A Dual-Path Framework for Multilingual Multi-hop Question Answering
+
+- **ID:** `2603.19097` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Yilin Wang, Yuchun Fan, Jiaoyang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.19097 · **PDF:** https://arxiv.org/pdf/2603.19097v1
+
+## 3454. SAVeS: Steering Safety Judgments in Vision-Language Models via Semantic Cues
+
+- **ID:** `2603.19092` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Carlos Hinojosa, Clemens Grange, Bernard Ghanem
+- **abs:** https://arxiv.org/abs/2603.19092 · **PDF:** https://arxiv.org/pdf/2603.19092v1
+
+## 3455. Position: Spectral GNNs Are Neither Spectral Nor Superior for Node Classification
+
+- **ID:** `2603.19091` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Qin Jiang, Chengjia Wang, Michael Lones, et al.
+- **abs:** https://arxiv.org/abs/2603.19091 · **PDF:** https://arxiv.org/pdf/2603.19091v2
+
+## 3456. Serendipity by Design: Evaluating the Impact of Cross-domain Mappings on Human and LLM Creativity
+
+- **ID:** `2603.19087` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Qiawen Ella Liu, Marina Dubova, Henry Conklin, et al.
+- **abs:** https://arxiv.org/abs/2603.19087 · **PDF:** https://arxiv.org/pdf/2603.19087v1
+
+## 3457. A Dataset and Resources for Identifying Patient Health Literacy Information from Clinical Notes
+
+- **ID:** `2603.19082` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Madeline Bittner, Dina Demner-Fushman, Yasmeen Shabazz, et al.
+- **abs:** https://arxiv.org/abs/2603.19082 · **PDF:** https://arxiv.org/pdf/2603.19082v1
+
+## 3458. CAMO: A Conditional Neural Solver for the Multi-objective Multiple Traveling Salesman Problem
+
+- **ID:** `2603.19074` · **primary:** `cs.RO` · **Submitted:** 2026-03-19
+- **Authors:** Fengxiaoxiao Li, Xiao Mao, Mingfeng Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.19074 · **PDF:** https://arxiv.org/pdf/2603.19074v1
+
+## 3459. Communication-Efficient and Robust Multi-Modal Federated Learning via Latent-Space Consensus
+
+- **ID:** `2603.19067` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Mohamed Badi, Chaouki Ben Issaid, Mehdi Bennis
+- **abs:** https://arxiv.org/abs/2603.19067 · **PDF:** https://arxiv.org/pdf/2603.19067v1
+
+## 3460. Parallelograms Strike Back: LLMs Generate Better Analogies than People
+
+- **ID:** `2603.19066` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Qiawen Ella Liu, Raja Marjieh, Jian-Qiao Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.19066 · **PDF:** https://arxiv.org/pdf/2603.19066v1
+
+## 3461. Reason-to-Transmit: Deliberative Adaptive Communication for Cooperative Perception
+
+- **ID:** `2603.20308` · **primary:** `cs.MA` · **Submitted:** 2026-03-19
+- **Authors:** Aayam Bansal, Ishaan Gangwani
+- **abs:** https://arxiv.org/abs/2603.20308 · **PDF:** https://arxiv.org/pdf/2603.20308v1
+
+## 3462. Hardness of High-Dimensional Linear Classification
+
+- **ID:** `2603.19061` · **primary:** `cs.CG` · **Submitted:** 2026-03-19
+- **Authors:** Alexander Munteanu, Simon Omlor, Jeff M. Phillips
+- **abs:** https://arxiv.org/abs/2603.19061 · **PDF:** https://arxiv.org/pdf/2603.19061v1
+
+## 3463. Adaptive Nonlinear Data Assimilation through P-Spline Triangular Measure Transport
+
+- **ID:** `2603.19058` · **primary:** `stat.CO` · **Submitted:** 2026-03-19
+- **Authors:** Berent Å. S. Lunde, Maximilian Ramgraber
+- **abs:** https://arxiv.org/abs/2603.19058 · **PDF:** https://arxiv.org/pdf/2603.19058v1
+
+## 3464. Em-Garde: A Propose-Match Framework for Proactive Streaming Video Understanding
+
+- **ID:** `2603.19054` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Yikai Zheng, Xin Ding, Yifan Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.19054 · **PDF:** https://arxiv.org/pdf/2603.19054v1
+
+## 3465. MoRI: Learning Motivation-Grounded Reasoning for Scientific Ideation in Large Language Models
+
+- **ID:** `2603.19044` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Chenyang Gu, Jiahao Cheng, Meicong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.19044 · **PDF:** https://arxiv.org/pdf/2603.19044v1
+
+## 3466. Man and machine: artificial intelligence and judicial decision making
+
+- **ID:** `2603.19042` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Arthur Dyevre, Ahmad Shahvaroughi
+- **abs:** https://arxiv.org/abs/2603.19042 · **PDF:** https://arxiv.org/pdf/2603.19042v3
+
+## 3467. Fast and Interpretable Autoregressive Estimation with Neural Network Backpropagation
+
+- **ID:** `2603.19041` · **primary:** `stat.ML` · **Submitted:** 2026-03-19
+- **Authors:** Anaísa Lucena, Ana Martins, Armando J. Pinho, et al.
+- **abs:** https://arxiv.org/abs/2603.19041 · **PDF:** https://arxiv.org/pdf/2603.19041v1
+
+## 3468. When Differential Privacy Meets Wireless Federated Learning: An Improved Analysis for Privacy and Convergence
+
+- **ID:** `2603.19040` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Chen Yaoling, Liang Hao, Tu Xiaotong
+- **abs:** https://arxiv.org/abs/2603.19040 · **PDF:** https://arxiv.org/pdf/2603.19040v1
+
+## 3469. Beyond Weighted Summation: Learnable Nonlinear Aggregation Functions for Robust Artificial Neurons
+
+- **ID:** `2603.19344` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Berke Deniz Bozyigit
+- **abs:** https://arxiv.org/abs/2603.19344 · **PDF:** https://arxiv.org/pdf/2603.19344v1
+
+## 3470. SEM: Sparse Embedding Modulation for Post-Hoc Debiasing of Vision-Language Models
+
+- **ID:** `2603.19028` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Quentin Guimard, Federico Bartsch, Simone Caldarella, et al.
+- **abs:** https://arxiv.org/abs/2603.19028 · **PDF:** https://arxiv.org/pdf/2603.19028v1
+
+## 3471. Towards Verifiable AI with Lightweight Cryptographic Proofs of Inference
+
+- **ID:** `2603.19025` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Pranay Anchuri, Matteo Campanelli, Paul Cesaretti, et al.
+- **abs:** https://arxiv.org/abs/2603.19025 · **PDF:** https://arxiv.org/pdf/2603.19025v1
+
+## 3472. Behavioral Fingerprints for LLM Endpoint Stability and Identity
+
+- **ID:** `2603.19022` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Jonah Leshin, Manish Shah, Ian Timmis, et al.
+- **abs:** https://arxiv.org/abs/2603.19022 · **PDF:** https://arxiv.org/pdf/2603.19022v1
+
+## 3473. What Really Controls Temporal Reasoning in Large Language Models: Tokenisation or Representation of Time?
+
+- **ID:** `2603.19017` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Gagan Bhatia, Ahmad Muhammad Isa, Maxime Peyrard, et al.
+- **abs:** https://arxiv.org/abs/2603.19017 · **PDF:** https://arxiv.org/pdf/2603.19017v1
+
+## 3474. Security awareness in LLM agents: the NDAI zone case
+
+- **ID:** `2603.19011` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Enrico Bottazzi, Pia Park
+- **abs:** https://arxiv.org/abs/2603.19011 · **PDF:** https://arxiv.org/pdf/2603.19011v1
+
+## 3475. Hypothesis-Conditioned Query Rewriting for Decision-Useful Retrieval
+
+- **ID:** `2603.19008` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Hangeol Chang, Changsun Lee, Seungjoon Rho, et al.
+- **abs:** https://arxiv.org/abs/2603.19008 · **PDF:** https://arxiv.org/pdf/2603.19008v1
+
+## 3476. EARTalking: End-to-end GPT-style Autoregressive Talking Head Synthesis with Frame-wise Control
+
+- **ID:** `2603.20307` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Yuzhe Weng, Haotian Wang, Yuanhong Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.20307 · **PDF:** https://arxiv.org/pdf/2603.20307v1
+
+## 3477. AgentDS Technical Report: Benchmarking the Future of Human-AI Collaboration in Domain-Specific Data Science
+
+- **ID:** `2603.19005` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** An Luo, Jin Du, Xun Xian, et al.
+- **abs:** https://arxiv.org/abs/2603.19005 · **PDF:** https://arxiv.org/pdf/2603.19005v1
+
+## 3478. RADIUS: Ranking, Distribution, and Significance - A Comprehensive Alignment Suite for Survey Simulation
+
+- **ID:** `2603.19002` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Weronika Łajewska, Paul Missault, George Davidson, et al.
+- **abs:** https://arxiv.org/abs/2603.19002 · **PDF:** https://arxiv.org/pdf/2603.19002v1
+
+## 3479. Regret Bounds for Competitive Resource Allocation with Endogenous Costs
+
+- **ID:** `2603.18999` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Rui Chai
+- **abs:** https://arxiv.org/abs/2603.18999 · **PDF:** https://arxiv.org/pdf/2603.18999v1
+
+## 3480. Evaluating Game Difficulty in Tetris Block Puzzle
+
+- **ID:** `2603.18994` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Chun-Jui Wang, Jian-Ting Guo, Hung Guei, et al.
+- **abs:** https://arxiv.org/abs/2603.18994 · **PDF:** https://arxiv.org/pdf/2603.18994v2
+
+## 3481. Foundations of Schrödinger Bridges for Generative Modeling
+
+- **ID:** `2603.18992` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Sophia Tang
+- **abs:** https://arxiv.org/abs/2603.18992 · **PDF:** https://arxiv.org/pdf/2603.18992v1
+
+## 3482. CRAFT: Aligning Diffusion Models with Fine-Tuning Is Easier Than You Think
+
+- **ID:** `2603.18991` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Zening Sun, Zhengpeng Xie, Lichen Bai, et al.
+- **abs:** https://arxiv.org/abs/2603.18991 · **PDF:** https://arxiv.org/pdf/2603.18991v1
+
+## 3483. Unmasking Algorithmic Bias in Predictive Policing: A GAN-Based Simulation Framework with Multi-City Temporal Analysis
+
+- **ID:** `2603.18987` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Pronob Kumar Barman, Pronoy Kumar Barman
+- **abs:** https://arxiv.org/abs/2603.18987 · **PDF:** https://arxiv.org/pdf/2603.18987v2
+
+## 3484. Revisiting OmniAnomaly for Anomaly Detection: performance metrics and comparison with PCA-based models
+
+- **ID:** `2603.18985` · **primary:** `stat.ML` · **Submitted:** 2026-03-19
+- **Authors:** Bruna Alves, Ana Martins, Armando J. Pinho, et al.
+- **abs:** https://arxiv.org/abs/2603.18985 · **PDF:** https://arxiv.org/pdf/2603.18985v1
+
+## 3485. Book your room in the Turing Hotel! A symmetric and distributed Turing Test with multiple AIs and humans
+
+- **ID:** `2603.18981` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Christian Di Maio, Tommaso Guidi, Luigi Quarantiello, et al.
+- **abs:** https://arxiv.org/abs/2603.18981 · **PDF:** https://arxiv.org/pdf/2603.18981v1
+
+## 3486. PRIOR: Perceptive Learning for Humanoid Locomotion with Reference Gait Priors
+
+- **ID:** `2603.18979` · **primary:** `cs.RO` · **Submitted:** 2026-03-19
+- **Authors:** Chenxi Han, Shilu He, Yi Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.18979 · **PDF:** https://arxiv.org/pdf/2603.18979v1
+
+## 3487. Evaluating 5W3H Structured Prompting for Intent Alignment in Human-AI Interaction
+
+- **ID:** `2603.18976` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Peng Gang
+- **abs:** https://arxiv.org/abs/2603.18976 · **PDF:** https://arxiv.org/pdf/2603.18976v1
+
+## 3488. Best-of-Both-Worlds Multi-Dueling Bandits: Unified Algorithms for Stochastic and Adversarial Preferences under Condorcet and Borda Objectives
+
+- **ID:** `2603.18972` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** S Akash, Pratik Gajane, Jawar Singh
+- **abs:** https://arxiv.org/abs/2603.18972 · **PDF:** https://arxiv.org/pdf/2603.18972v2
+
+## 3489. Teleological Inference in Structural Causal Models via Intentional Interventions
+
+- **ID:** `2603.18968` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Dario Compagno, Fabio Massimo Zennaro
+- **abs:** https://arxiv.org/abs/2603.18968 · **PDF:** https://arxiv.org/pdf/2603.18968v1
+
+## 3490. Maximum-Entropy Exploration with Future State-Action Visitation Measures
+
+- **ID:** `2603.18965` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Adrien Bolland, Gaspard Lambrechts, Damien Ernst
+- **abs:** https://arxiv.org/abs/2603.18965 · **PDF:** https://arxiv.org/pdf/2603.18965v1
+
+## 3491. BVSIMC: Bayesian Variable Selection-Guided Inductive Matrix Completion for Improved and Interpretable Drug Discovery
+
+- **ID:** `2603.18957` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Sijian Fan, Liyan Xiong, Dayuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.18957 · **PDF:** https://arxiv.org/pdf/2603.18957v1
+
+## 3492. Balancing Performance and Fairness in Explainable AI for Anomaly Detection in Distributed Power Plants Monitoring
+
+- **ID:** `2603.18954` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Corneille Niyonkuru, Marcellin Atemkeng, Gabin Maxime Nguegnang, et al.
+- **abs:** https://arxiv.org/abs/2603.18954 · **PDF:** https://arxiv.org/pdf/2603.18954v1
+
+## 3493. Context Bootstrapped Reinforcement Learning
+
+- **ID:** `2603.18953` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Saaket Agashe, Jayanth Srinivasa, Gaowen Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.18953 · **PDF:** https://arxiv.org/pdf/2603.18953v1
+
+## 3494. A conceptual framework for ideology beyond the left and right
+
+- **ID:** `2603.18945` · **primary:** `cs.CY` · **Submitted:** 2026-03-19
+- **Authors:** Kenneth Joseph, Kim Williams, David Lazer
+- **abs:** https://arxiv.org/abs/2603.18945 · **PDF:** https://arxiv.org/pdf/2603.18945v1
+
+## 3495. Unified Taxonomy for Multivariate Time Series Anomaly Detection using Deep Learning
+
+- **ID:** `2603.18941` · **primary:** `stat.ML` · **Submitted:** 2026-03-19
+- **Authors:** Bruna Alves, Armando J. Pinho, Sónia Gouveia
+- **abs:** https://arxiv.org/abs/2603.18941 · **PDF:** https://arxiv.org/pdf/2603.18941v1
+
+## 3496. Entropy trajectory shape predicts LLM reasoning reliability: A diagnostic study of uncertainty dynamics in chain-of-thought
+
+- **ID:** `2603.18940` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Xinghao Zhao
+- **abs:** https://arxiv.org/abs/2603.18940 · **PDF:** https://arxiv.org/pdf/2603.18940v2
+
+## 3497. Kernel Single-Index Bandits: Estimation, Inference, and Learning
+
+- **ID:** `2603.18938` · **primary:** `stat.ML` · **Submitted:** 2026-03-19
+- **Authors:** Sakshi Arya, Satarupa Bhattacharjee, Bharath K. Sriperumbudur
+- **abs:** https://arxiv.org/abs/2603.18938 · **PDF:** https://arxiv.org/pdf/2603.18938v1
+
+## 3498. An Optimised Greedy-Weighted Ensemble Framework for Financial Loan Default Prediction
+
+- **ID:** `2603.18927` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Ezekiel Nii Noye Nortey, Jones Asante-Koranteng, Marcellin Atemkeng, et al.
+- **abs:** https://arxiv.org/abs/2603.18927 · **PDF:** https://arxiv.org/pdf/2603.18927v1
+
+## 3499. Improving moment tensor solutions under Earth structure uncertainty with simulation-based inference
+
+- **ID:** `2603.18925` · **primary:** `physics.geo-ph` · **Submitted:** 2026-03-19
+- **Authors:** A. A. Saoulis, T. -S. Pham, A. M. G. Ferreira
+- **abs:** https://arxiv.org/abs/2603.18925 · **PDF:** https://arxiv.org/pdf/2603.18925v1
+
+## 3500. A Multi-Modal Dataset for Ground Reaction Force Estimation Using Consumer Wearable Sensors
+
+- **ID:** `2603.28784` · **primary:** `eess.SP` · **Submitted:** 2026-03-19
+- **Authors:** Parvin Ghaffarzadeh, Debarati Chakraborty, Koorosh Aslansefat, et al.
+- **abs:** https://arxiv.org/abs/2603.28784 · **PDF:** https://arxiv.org/pdf/2603.28784v1
+
+## 3501. Agentic Business Process Management: A Research Manifesto
+
+- **ID:** `2603.18916` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Diego Calvanese, Angelo Casciani, Giuseppe De Giacomo, et al.
+- **abs:** https://arxiv.org/abs/2603.18916 · **PDF:** https://arxiv.org/pdf/2603.18916v2
+
+## 3502. Security, privacy, and agentic AI in a regulatory view: From definitions and distinctions to provisions and reflections
+
+- **ID:** `2603.18914` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Shiliang Zhang, Sabita Maharjan
+- **abs:** https://arxiv.org/abs/2603.18914 · **PDF:** https://arxiv.org/pdf/2603.18914v1
+
+## 3503. Progressive Training for Explainable Citation-Grounded Dialogue: Reducing Hallucination to Zero in English-Hindi LLMs
+
+- **ID:** `2603.18911` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Vedant Pandya
+- **abs:** https://arxiv.org/abs/2603.18911 · **PDF:** https://arxiv.org/pdf/2603.18911v1
+
+## 3504. Characterizing Linear Alignment Across Language Models
+
+- **ID:** `2603.18908` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Matt Gorbett, Suman Jana
+- **abs:** https://arxiv.org/abs/2603.18908 · **PDF:** https://arxiv.org/pdf/2603.18908v3
+
+## 3505. Neural Galerkin Normalizing Flow for Transition Probability Density Functions of Diffusion Models
+
+- **ID:** `2603.18907` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Riccardo Saporiti, Fabio Nobile
+- **abs:** https://arxiv.org/abs/2603.18907 · **PDF:** https://arxiv.org/pdf/2603.18907v1
+
+## 3506. Uniform a priori bounds and error analysis for the Adam stochastic gradient descent optimization method
+
+- **ID:** `2603.18899` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Steffen Dereich, Thang Do, Arnulf Jentzen
+- **abs:** https://arxiv.org/abs/2603.18899 · **PDF:** https://arxiv.org/pdf/2603.18899v1
+
+## 3507. Act While Thinking: Accelerating LLM Agents via Pattern-Aware Speculative Tool Execution
+
+- **ID:** `2603.18897` · **primary:** `cs.DC` · **Submitted:** 2026-03-19
+- **Authors:** Yifan Sui, Han Zhao, Rui Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.18897 · **PDF:** https://arxiv.org/pdf/2603.18897v1
+
+## 3508. Translating MRI to PET through Conditional Diffusion Models with Enhanced Pathology Awareness
+
+- **ID:** `2603.18896` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Yitong Li, Igor Yakushev, Dennis M. Hedderich, et al.
+- **abs:** https://arxiv.org/abs/2603.18896 · **PDF:** https://arxiv.org/pdf/2603.18896v1
+
+## 3509. From Accuracy to Readiness: Metrics and Benchmarks for Human-AI Decision-Making
+
+- **ID:** `2603.18895` · **primary:** `cs.HC` · **Submitted:** 2026-03-19
+- **Authors:** Min Hun Lee
+- **abs:** https://arxiv.org/abs/2603.18895 · **PDF:** https://arxiv.org/pdf/2603.18895v1
+
+## 3510. I Can't Believe It's Corrupt: Evaluating Corruption in Multi-Agent Governance Systems
+
+- **ID:** `2603.18894` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Vedanta S P, Ponnurangam Kumaraguru
+- **abs:** https://arxiv.org/abs/2603.18894 · **PDF:** https://arxiv.org/pdf/2603.18894v1
+
+## 3511. Quantitative Introspection in Language Models: Tracking Internal States Across Conversation
+
+- **ID:** `2603.18893` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Nicolas Martorell
+- **abs:** https://arxiv.org/abs/2603.18893 · **PDF:** https://arxiv.org/pdf/2603.18893v1
+
+## 3512. MultihopSpatial: Multi-hop Compositional Spatial Reasoning Benchmark for Vision-Language Model
+
+- **ID:** `2603.18892` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Youngwan Lee, Soojin Jang, Yoorhim Cho, et al.
+- **abs:** https://arxiv.org/abs/2603.18892 · **PDF:** https://arxiv.org/pdf/2603.18892v1
+
+## 3513. PromptHub: Enhancing Multi-Prompt Visual In-Context Learning with Locality-Aware Fusion, Concentration and Alignment
+
+- **ID:** `2603.18891` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Tianci Luo, Jinpeng Wang, Shiyu Qin, et al.
+- **abs:** https://arxiv.org/abs/2603.18891 · **PDF:** https://arxiv.org/pdf/2603.18891v1
+
+## 3514. Authority-Level Priors: An Under-Specified Constraint in Hierarchical Predictive Processing
+
+- **ID:** `2603.18888` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Marcy Palejova
+- **abs:** https://arxiv.org/abs/2603.18888 · **PDF:** https://arxiv.org/pdf/2603.18888v1
+
+## 3515. Reasoning over mathematical objects: on-policy reward modeling and test time aggregation
+
+- **ID:** `2603.18886` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Pranjal Aggarwal, Marjan Ghazvininejad, Seungone Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.18886 · **PDF:** https://arxiv.org/pdf/2603.18886v1
+
+## 3516. Geography According to ChatGPT -- How Generative AI Represents and Reasons about Geography
+
+- **ID:** `2603.18881` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Krzysztof Janowicz, Gengchen Mai, Rui Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.18881 · **PDF:** https://arxiv.org/pdf/2603.18881v1
+
+## 3517. A Human-in/on-the-Loop Framework for Accessible Text Generation
+
+- **ID:** `2603.18879` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Lourdes Moreno, Paloma Martínez
+- **abs:** https://arxiv.org/abs/2603.18879 · **PDF:** https://arxiv.org/pdf/2603.18879v1
+
+## 3518. Evaluating LLM-Generated Lessons from the Language Learning Students' Perspective: A Short Case Study on Duolingo
+
+- **ID:** `2603.18873` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Carlos Rafael Catalan, Patricia Nicole Monderin, Lheane Marie Dizon, et al.
+- **abs:** https://arxiv.org/abs/2603.18873 · **PDF:** https://arxiv.org/pdf/2603.18873v2
+
+## 3519. DriftGuard: Mitigating Asynchronous Data Drift in Federated Learning
+
+- **ID:** `2603.18872` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Yizhou Han, Di Wu, Blesson Varghese
+- **abs:** https://arxiv.org/abs/2603.18872 · **PDF:** https://arxiv.org/pdf/2603.18872v1
+
+## 3520. Bridging Network Fragmentation: A Semantic-Augmented DRL Framework for UAV-aided VANETs
+
+- **ID:** `2603.18871` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Gaoxiang Cao, Wenke Yuan, Huasen He, et al.
+- **abs:** https://arxiv.org/abs/2603.18871 · **PDF:** https://arxiv.org/pdf/2603.18871v1
+
+## 3521. Through the Looking-Glass: AI-Mediated Video Communication Reduces Interpersonal Trust and Confidence in Judgments
+
+- **ID:** `2603.18868` · **primary:** `cs.HC` · **Submitted:** 2026-03-19
+- **Authors:** Nelson Navajas Fernández, Jeffrey T. Hancock, Maurice Jakesch
+- **abs:** https://arxiv.org/abs/2603.18868 · **PDF:** https://arxiv.org/pdf/2603.18868v1
+
+## 3522. Conflict-Based Search for Multi Agent Path Finding with Asynchronous Actions
+
+- **ID:** `2603.18866` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Xuemian Wu, Shizhe Zhao, Zhongqiang Ren
+- **abs:** https://arxiv.org/abs/2603.18866 · **PDF:** https://arxiv.org/pdf/2603.18866v2
+
+## 3523. RadioDiff-FS: Physics-Informed Manifold Alignment in Few-Shot Diffusion Models for High-Fidelity Radio Map Construction
+
+- **ID:** `2603.18865` · **primary:** `eess.SY` · **Submitted:** 2026-03-19
+- **Authors:** Xiucheng Wang, Zixuan Guo, Nan Cheng
+- **abs:** https://arxiv.org/abs/2603.18865 · **PDF:** https://arxiv.org/pdf/2603.18865v3
+
+## 3524. Data-driven construction of machine-learning-based interatomic potentials for gas-surface scattering dynamics: the case of NO on graphite
+
+- **ID:** `2603.18864` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-19
+- **Authors:** Samuel Del Fré, Gilberto A. Alou Angulo, Maurice Monnerville, et al.
+- **abs:** https://arxiv.org/abs/2603.18864 · **PDF:** https://arxiv.org/pdf/2603.18864v1
+
+## 3525. Why Better Cross-Lingual Alignment Fails for Better Cross-Lingual Transfer: Case of Encoders
+
+- **ID:** `2603.18863` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Yana Veitsman, Yihong Liu, Hinrich Schütze
+- **abs:** https://arxiv.org/abs/2603.18863 · **PDF:** https://arxiv.org/pdf/2603.18863v1
+
+## 3526. RewardFlow: Topology-Aware Reward Propagation on State Graphs for Agentic RL with Large Language Models
+
+- **ID:** `2603.18859` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Xiao Feng, Bo Han, Zhanke Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.18859 · **PDF:** https://arxiv.org/pdf/2603.18859v1
+
+## 3527. Motion-o: Trajectory-Grounded Video Reasoning
+
+- **ID:** `2603.18856` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Bishoy Galoaa, Shayda Moezzi, Xiangyu Bai, et al.
+- **abs:** https://arxiv.org/abs/2603.18856 · **PDF:** https://arxiv.org/pdf/2603.18856v1
+
+## 3528. BeamAgent: LLM-Aided MIMO Beamforming with Decoupled Intent Parsing and Alternating Optimization for Joint Site Selection and Precoding
+
+- **ID:** `2603.18855` · **primary:** `cs.IT` · **Submitted:** 2026-03-19
+- **Authors:** Xiucheng Wang, Yue Zhang, Nan Cheng
+- **abs:** https://arxiv.org/abs/2603.18855 · **PDF:** https://arxiv.org/pdf/2603.18855v1
+
+## 3529. Learn for Variation: Variationally Guided AAV Trajectory Learning in Differentiable Environments
+
+- **ID:** `2603.18853` · **primary:** `eess.SY` · **Submitted:** 2026-03-19
+- **Authors:** Xiucheng Wang, Zhenye Chen, Nan Cheng
+- **abs:** https://arxiv.org/abs/2603.18853 · **PDF:** https://arxiv.org/pdf/2603.18853v2
+
+## 3530. Towards Interpretable Foundation Models for Retinal Fundus Images
+
+- **ID:** `2603.18846` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Samuel Ofosu Mensah, Maria Camila Roa Carvajal, Kerol Djoumessi, et al.
+- **abs:** https://arxiv.org/abs/2603.18846 · **PDF:** https://arxiv.org/pdf/2603.18846v1
+
+## 3531. A Model Ensemble-Based Post-Processing Framework for Fairness-Aware Prediction
+
+- **ID:** `2603.18838` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Zhouting Zhao, Tin Lok James Ng
+- **abs:** https://arxiv.org/abs/2603.18838 · **PDF:** https://arxiv.org/pdf/2603.18838v1
+
+## 3532. Model Order Reduction of Cerebrovascular Hemodynamics Using POD_Galerkin and Reservoir Computing_based Approach
+
+- **ID:** `2603.18837` · **primary:** `math.NA` · **Submitted:** 2026-03-19
+- **Authors:** Rahul Halder, Arash Hajisharifi, Kabir Bakhshaei, et al.
+- **abs:** https://arxiv.org/abs/2603.18837 · **PDF:** https://arxiv.org/pdf/2603.18837v1
+
+## 3533. Agent Control Protocol: Admission Control for Agent Actions
+
+- **ID:** `2603.18829` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Marcelo Fernandez
+- **abs:** https://arxiv.org/abs/2603.18829 · **PDF:** https://arxiv.org/pdf/2603.18829v6
+
+## 3534. Student views in AI Ethics and Social Impact
+
+- **ID:** `2603.18827` · **primary:** `cs.CY` · **Submitted:** 2026-03-19
+- **Authors:** Tudor-Dan Mihoc, Manuela-Andreea Petrescu, Emilia-Loredana Pop
+- **abs:** https://arxiv.org/abs/2603.18827 · **PDF:** https://arxiv.org/pdf/2603.18827v1
+
+## 3535. Detecting Basic Values in A Noisy Russian Social Media Text Data: A Multi-Stage Classification Framework
+
+- **ID:** `2603.18822` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Maria Milkova, Maksim Rudnev
+- **abs:** https://arxiv.org/abs/2603.18822 · **PDF:** https://arxiv.org/pdf/2603.18822v1
+
+## 3536. Seasoning Generative Models for a Generalization Aftertaste
+
+- **ID:** `2603.18817` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Hisham Husain, Valentin De Bortoli, Richard Nock
+- **abs:** https://arxiv.org/abs/2603.18817 · **PDF:** https://arxiv.org/pdf/2603.18817v1
+
+## 3537. ProRL Agent: Rollout-as-a-Service for RL Training of Multi-Turn LLM Agents
+
+- **ID:** `2603.18815` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Hao Zhang, Mingjie Liu, Shaokun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.18815 · **PDF:** https://arxiv.org/pdf/2603.18815v1
+
+## 3538. Can LLM generate interesting mathematical research problems?
+
+- **ID:** `2603.18813` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Xiaoyang Chen, Xiang Jiang
+- **abs:** https://arxiv.org/abs/2603.18813 · **PDF:** https://arxiv.org/pdf/2603.18813v1
+
+## 3539. dTRPO: Trajectory Reduction in Policy Optimization of Diffusion Large Language Models
+
+- **ID:** `2603.18806` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Wenxuan Zhang, Lemeng Wu, Changsheng Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.18806 · **PDF:** https://arxiv.org/pdf/2603.18806v1
+
+## 3540. Signals of Success and Struggle: Early Prediction and Physiological Signatures of Human Performance across Task Complexity
+
+- **ID:** `2603.18798` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Yufei Cao, Penny Sweetser, Ziyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.18798 · **PDF:** https://arxiv.org/pdf/2603.18798v1
+
+## 3541. Perceptio: Perception Enhanced Vision Language Models via Spatial Token Generation
+
+- **ID:** `2603.18795` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Yuchen Li, Amanmeet Garg, Shalini Chaudhuri, et al.
+- **abs:** https://arxiv.org/abs/2603.18795 · **PDF:** https://arxiv.org/pdf/2603.18795v1
+
+## 3542. Functional Subspace Watermarking for Large Language Models
+
+- **ID:** `2603.18793` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Zikang Ding, Junhao Li, Suling Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.18793 · **PDF:** https://arxiv.org/pdf/2603.18793v1
+
+## 3543. Mi:dm K 2.5 Pro
+
+- **ID:** `2603.18788` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** KT Tech innovation Group
+- **abs:** https://arxiv.org/abs/2603.18788 · **PDF:** https://arxiv.org/pdf/2603.18788v2
+
+## 3544. Proceedings of the 2nd Workshop on Advancing Artificial Intelligence through Theory of Mind
+
+- **ID:** `2603.18786` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Nitay Alon, Joseph M. Barnby, Reuth Mirsky, et al.
+- **abs:** https://arxiv.org/abs/2603.18786 · **PDF:** https://arxiv.org/pdf/2603.18786v1
+
+## 3545. Points-to-3D: Structure-Aware 3D Generation with Point Cloud Priors
+
+- **ID:** `2603.18782` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Jiatong Xia, Zicheng Duan, Anton van den Hengel, et al.
+- **abs:** https://arxiv.org/abs/2603.18782 · **PDF:** https://arxiv.org/pdf/2603.18782v3
+
+## 3546. SRRM: Improving Recursive Transport Surrogates in the Small-Discrepancy Regime
+
+- **ID:** `2603.18781` · **primary:** `stat.ML` · **Submitted:** 2026-03-19
+- **Authors:** Yufei Zhang, Tao Wang, Jingyi Zhang
+- **abs:** https://arxiv.org/abs/2603.18781 · **PDF:** https://arxiv.org/pdf/2603.18781v1
+
+## 3547. Automatic Configuration of LLM Post-Training Pipelines
+
+- **ID:** `2603.18773` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Channe Chwa, Xinle Wu, Yao Lu
+- **abs:** https://arxiv.org/abs/2603.18773 · **PDF:** https://arxiv.org/pdf/2603.18773v1
+
+## 3548. A Concept is More Than a Word: Diversified Unlearning in Text-to-Image Diffusion Models
+
+- **ID:** `2603.18767` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Duc Hao Pham, Van Duy Truong, Duy Khanh Dinh, et al.
+- **abs:** https://arxiv.org/abs/2603.18767 · **PDF:** https://arxiv.org/pdf/2603.18767v1
+
+## 3549. Enhancing the Parameterization of Reservoir Properties for Data Assimilation Using Deep VAE-GAN
+
+- **ID:** `2603.18766` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** M. A. Sampaio, P. H. Ranazzi, M. J. Blunt
+- **abs:** https://arxiv.org/abs/2603.18766 · **PDF:** https://arxiv.org/pdf/2603.18766v2
+
+## 3550. Implicit Grading Bias in Large Language Models: How Writing Style Affects Automated Assessment Across Math, Programming, and Essay Tasks
+
+- **ID:** `2603.18765` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Rudra Jadhav, Janhavi Danve, Sonalika Shaw
+- **abs:** https://arxiv.org/abs/2603.18765 · **PDF:** https://arxiv.org/pdf/2603.18765v1
+
+## 3551. ClawTrap: A MITM-Based Red-Teaming Framework for Real-World OpenClaw Security Evaluation
+
+- **ID:** `2603.18762` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Haochen Zhao, Shaoyang Cui
+- **abs:** https://arxiv.org/abs/2603.18762 · **PDF:** https://arxiv.org/pdf/2603.18762v1
+
+## 3552. NeuroGame Transformer: Gibbs-Inspired Attention Driven by Game Theory and Statistical Physics
+
+- **ID:** `2603.18761` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Djamel Bouchaffra, Faycal Ykhlef, Hanene Azzag, et al.
+- **abs:** https://arxiv.org/abs/2603.18761 · **PDF:** https://arxiv.org/pdf/2603.18761v1
+
+## 3553. InjectFlow: Weak Guides Strong via Orthogonal Injection for Flow Matching
+
+- **ID:** `2603.20303` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Dayu Wang, Jiaye Yang, Weikang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.20303 · **PDF:** https://arxiv.org/pdf/2603.20303v1
+
+## 3554. Are complicated loss functions necessary for teaching LLMs to reason?
+
+- **ID:** `2603.18756` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Gabriele Carrino, Andrea Sassella, Nicolo Brunello, et al.
+- **abs:** https://arxiv.org/abs/2603.18756 · **PDF:** https://arxiv.org/pdf/2603.18756v1
+
+## 3555. WeNLEX: Weakly Supervised Natural Language Explanations for Multilabel Chest X-ray Classification
+
+- **ID:** `2603.18752` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Isabel Rio-Torto, Jaime S. Cardoso, Luís F. Teixeira
+- **abs:** https://arxiv.org/abs/2603.18752 · **PDF:** https://arxiv.org/pdf/2603.18752v1
+
+## 3556. Automatic detection of Gen-AI texts: A comparative framework of neural models
+
+- **ID:** `2603.18750` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Cristian Buttaro, Irene Amerini
+- **abs:** https://arxiv.org/abs/2603.18750 · **PDF:** https://arxiv.org/pdf/2603.18750v1
+
+## 3557. Memento-Skills: Let Agents Design Agents
+
+- **ID:** `2603.18743` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Huichi Zhou, Siyuan Guo, Anjie Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.18743 · **PDF:** https://arxiv.org/pdf/2603.18743v1
+
+## 3558. Measuring and Exploiting Confirmation Bias in LLM-Assisted Security Code Review
+
+- **ID:** `2603.18740` · **primary:** `cs.SE` · **Submitted:** 2026-03-19
+- **Authors:** Dimitris Mitropoulos, Nikolaos Alexopoulos, Georgios Alexopoulos, et al.
+- **abs:** https://arxiv.org/abs/2603.18740 · **PDF:** https://arxiv.org/pdf/2603.18740v1
+
+## 3559. CausalRM: Causal-Theoretic Reward Modeling for RLHF from Observational User Feedbacks
+
+- **ID:** `2603.18736` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Hao Wang, Licheng Pan, Zhichao Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.18736 · **PDF:** https://arxiv.org/pdf/2603.18736v1
+
+## 3560. Analysis Of Linguistic Stereotypes in Single and Multi-Agent Generative AI Architectures
+
+- **ID:** `2603.18729` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Martina Ullasci, Marco Rondina, Riccardo Coppola, et al.
+- **abs:** https://arxiv.org/abs/2603.18729 · **PDF:** https://arxiv.org/pdf/2603.18729v1
+
+## 3561. Ontology-Guided Diffusion for Zero-Shot Visual Sim2Real Transfer
+
+- **ID:** `2603.18719` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Mohamed Youssef, Mayar Elfares, Anna-Maria Meer, et al.
+- **abs:** https://arxiv.org/abs/2603.18719 · **PDF:** https://arxiv.org/pdf/2603.18719v2
+
+## 3562. MemMA: Coordinating the Memory Cycle through Multi-Agent Reasoning and In-Situ Self-Evolution
+
+- **ID:** `2603.18718` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Minhua Lin, Zhiwei Zhang, Hanqing Lu, et al.
+- **abs:** https://arxiv.org/abs/2603.18718 · **PDF:** https://arxiv.org/pdf/2603.18718v1
+
+## 3563. Holter-to-Sleep: AI-Enabled Repurposing of Single-Lead ECG for Sleep Phenotyping
+
+- **ID:** `2603.18714` · **primary:** `eess.SP` · **Submitted:** 2026-03-19
+- **Authors:** Donglin Xie, Qingshuo Zhao, Jingyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.18714 · **PDF:** https://arxiv.org/pdf/2603.18714v1
+
+## 3564. Accurate and Efficient Multi-Channel Time Series Forecasting via Sparse Attention Mechanism
+
+- **ID:** `2603.18712` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Lei Gao, Hengda Bao, Jingfei Fang, et al.
+- **abs:** https://arxiv.org/abs/2603.18712 · **PDF:** https://arxiv.org/pdf/2603.18712v1
+
+## 3565. From ex(p) to poly: Gaussian Splatting with Polynomial Kernels
+
+- **ID:** `2603.18707` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Joerg H. Mueller, Martin Winter, Markus Steinberger
+- **abs:** https://arxiv.org/abs/2603.18707 · **PDF:** https://arxiv.org/pdf/2603.18707v1
+
+## 3566. A mathematical framework for time-delay reservoir computing analysis
+
+- **ID:** `2603.18706` · **primary:** `math.OC` · **Submitted:** 2026-03-19
+- **Authors:** Anh-Tuan Clabaut, Jean Auriol, Islam Boussaada, et al.
+- **abs:** https://arxiv.org/abs/2603.18706 · **PDF:** https://arxiv.org/pdf/2603.18706v1
+
+## 3567. Off-Policy Learning with Limited Supply
+
+- **ID:** `2603.18702` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Koichi Tanaka, Ren Kishimoto, Bushun Kawagishi, et al.
+- **abs:** https://arxiv.org/abs/2603.18702 · **PDF:** https://arxiv.org/pdf/2603.18702v2
+
+## 3568. Spectral Tempering for Embedding Compression in Dense Passage Retrieval
+
+- **ID:** `2603.19339` · **primary:** `cs.IR` · **Submitted:** 2026-03-19
+- **Authors:** Yongkang Li, Panagiotis Eustratiadis, Evangelos Kanoulas
+- **abs:** https://arxiv.org/abs/2603.19339 · **PDF:** https://arxiv.org/pdf/2603.19339v1
+
+## 3569. OCP: Orthogonal Constrained Projection for Sparse Scaling in Industrial Commodity Recommendation
+
+- **ID:** `2603.18697` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Chen Sun, Beilin Xu, Boheng Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.18697 · **PDF:** https://arxiv.org/pdf/2603.18697v1
+
+## 3570. Voice Privacy from an Attribute-based Perspective
+
+- **ID:** `2603.20301` · **primary:** `cs.SD` · **Submitted:** 2026-03-19
+- **Authors:** Mehtab Ur Rahman, Martha Larson, Cristian Tejedor-Garcia
+- **abs:** https://arxiv.org/abs/2603.20301 · **PDF:** https://arxiv.org/pdf/2603.20301v2
+
+## 3571. STEP: Scientific Time-Series Encoder Pretraining via Cross-Domain Distillation
+
+- **ID:** `2603.18688` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Chen Zhang, Liwei Liu, Jun Tao, et al.
+- **abs:** https://arxiv.org/abs/2603.18688 · **PDF:** https://arxiv.org/pdf/2603.18688v1
+
+## 3572. HISR: Hindsight Information Modulated Segmental Process Rewards For Multi-turn Agentic Reinforcement Learning
+
+- **ID:** `2603.18683` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Zhicong Lu, Zichuan Lin, Wei Jia, et al.
+- **abs:** https://arxiv.org/abs/2603.18683 · **PDF:** https://arxiv.org/pdf/2603.18683v1
+
+## 3573. Revisiting Label Inference Attacks in Vertical Federated Learning: Why They Are Vulnerable and How to Defend
+
+- **ID:** `2603.18680` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Yige Liu, Dexuan Xu, Zimai Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.18680 · **PDF:** https://arxiv.org/pdf/2603.18680v1
+
+## 3574. Words at Play: Benchmarking Audio Pun Understanding in Large Audio-Language Models
+
+- **ID:** `2603.18678` · **primary:** `cs.SD` · **Submitted:** 2026-03-19
+- **Authors:** Yuchen Su, Shaoxin Zhong, Yonghua Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.18678 · **PDF:** https://arxiv.org/pdf/2603.18678v1
+
+## 3575. Cognitive Amplification vs Cognitive Delegation in Human-AI Systems: A Metric Framework
+
+- **ID:** `2603.18677` · **primary:** `cs.HC` · **Submitted:** 2026-03-19
+- **Authors:** Eduardo Di Santi
+- **abs:** https://arxiv.org/abs/2603.18677 · **PDF:** https://arxiv.org/pdf/2603.18677v1
+
+## 3576. MANAR: Memory-augmented Attention with Navigational Abstract Conceptual Representation
+
+- **ID:** `2603.18676` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Zuher Jahshan, Ben Ben Ishay, Leonid Yavits
+- **abs:** https://arxiv.org/abs/2603.18676 · **PDF:** https://arxiv.org/pdf/2603.18676v1
+
+## 3577. Thinking with Constructions: A Benchmark and Policy Optimization for Visual-Text Interleaved Geometric Reasoning
+
+- **ID:** `2603.18662` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Haokun Zhao, Wanshi Xu, Haidong Yuan, et al.
+- **abs:** https://arxiv.org/abs/2603.18662 · **PDF:** https://arxiv.org/pdf/2603.18662v1
+
+## 3578. Enhancing Multi-Corpus Training in SSL-Based Anti-Spoofing Models: Domain-Invariant Feature Extraction
+
+- **ID:** `2603.18657` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Anh-Tuan Dao, Driss Matrouf, Mickael Rouvier, et al.
+- **abs:** https://arxiv.org/abs/2603.18657 · **PDF:** https://arxiv.org/pdf/2603.18657v1
+
+## 3579. Balanced Thinking: Improving Chain of Thought Training in Vision Language Models
+
+- **ID:** `2603.18656` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Shaked Perek, Ben Wiesel, Avihu Dekel, et al.
+- **abs:** https://arxiv.org/abs/2603.18656 · **PDF:** https://arxiv.org/pdf/2603.18656v1
+
+## 3580. Multiscale Switch for Semi-Supervised and Contrastive Learning in Medical Ultrasound Image Segmentation
+
+- **ID:** `2603.18655` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Jingguo Qu, Xinyang Han, Yao Pu, et al.
+- **abs:** https://arxiv.org/abs/2603.18655 · **PDF:** https://arxiv.org/pdf/2603.18655v1
+
+## 3581. Benchmarking PDF Parsers on Table Extraction with LLM-based Semantic Evaluation
+
+- **ID:** `2603.18652` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Pius Horn, Janis Keuper
+- **abs:** https://arxiv.org/abs/2603.18652 · **PDF:** https://arxiv.org/pdf/2603.18652v1
+
+## 3582. Beyond TVLA: Anderson-Darling Leakage Assessment for Neural Network Side-Channel Leakage Detection
+
+- **ID:** `2603.18647` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Ján Mikulec, Jakub Breier, Xiaolu Hou
+- **abs:** https://arxiv.org/abs/2603.18647 · **PDF:** https://arxiv.org/pdf/2603.18647v1
+
+## 3583. Evaluating Model-Free Policy Optimization in Masked-Action Environments via an Exact Blackjack Oracle
+
+- **ID:** `2603.18642` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Kevin Song
+- **abs:** https://arxiv.org/abs/2603.18642 · **PDF:** https://arxiv.org/pdf/2603.18642v1
+
+## 3584. A Comparative Empirical Study of Catastrophic Forgetting Mitigation in Sequential Task Adaptation for Continual Natural Language Processing Systems
+
+- **ID:** `2603.18641` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Aram Abrahamyan, Sachin Kumar
+- **abs:** https://arxiv.org/abs/2603.18641 · **PDF:** https://arxiv.org/pdf/2603.18641v1
+
+## 3585. A Theoretical Comparison of No-U-Turn Sampler Variants: Necessary and Sufficient Convergence Conditions and Mixing Time Analysis under Gaussian Targets
+
+- **ID:** `2603.18640` · **primary:** `stat.ML` · **Submitted:** 2026-03-19
+- **Authors:** Samuel Gruffaz, Kyurae Kim, Fares Guehtar, et al.
+- **abs:** https://arxiv.org/abs/2603.18640 · **PDF:** https://arxiv.org/pdf/2603.18640v2
+
+## 3586. MOSAIC: Multi-Objective Slice-Aware Iterative Curation for Alignment
+
+- **ID:** `2603.18637` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Yipu Dou, Wang Yang
+- **abs:** https://arxiv.org/abs/2603.18637 · **PDF:** https://arxiv.org/pdf/2603.18637v1
+
+## 3587. SwiftGS: Episodic Priors for Immediate Satellite Surface Recovery
+
+- **ID:** `2603.18634` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Rong Fu, Jiekai Wu, Haiyun Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.18634 · **PDF:** https://arxiv.org/pdf/2603.18634v1
+
+## 3588. An Onto-Relational-Sophic Framework for Governing Synthetic Minds
+
+- **ID:** `2603.18633` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Huansheng Ning, Jianguo Ding
+- **abs:** https://arxiv.org/abs/2603.18633 · **PDF:** https://arxiv.org/pdf/2603.18633v1
+
+## 3589. D-Mem: A Dual-Process Memory System for LLM Agents
+
+- **ID:** `2603.18631` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Zhixing You, Jiachen Yuan, Jason Cai
+- **abs:** https://arxiv.org/abs/2603.18631 · **PDF:** https://arxiv.org/pdf/2603.18631v1
+
+## 3590. Agentic Flow Steering and Parallel Rollout Search for Spatially Grounded Text-to-Image Generation
+
+- **ID:** `2603.18627` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Ping Chen, Daoxuan Zhang, Xiangming Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.18627 · **PDF:** https://arxiv.org/pdf/2603.18627v1
+
+## 3591. REST: Receding Horizon Explorative Steiner Tree for Zero-Shot Object-Goal Navigation
+
+- **ID:** `2603.18624` · **primary:** `cs.RO` · **Submitted:** 2026-03-19
+- **Authors:** Shuqi Xiao, Maani Ghaffari, Chengzhong Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.18624 · **PDF:** https://arxiv.org/pdf/2603.18624v1
+
+## 3592. OpenT2M: No-frill Motion Generation with Open-source,Large-scale, High-quality Data
+
+- **ID:** `2603.18623` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Bin Cao, Sipeng Zheng, Hao Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.18623 · **PDF:** https://arxiv.org/pdf/2603.18623v1
+
+## 3593. Learning to Self-Evolve
+
+- **ID:** `2603.18620` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Xiaoyin Chen, Canwen Xu, Yite Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.18620 · **PDF:** https://arxiv.org/pdf/2603.18620v1
+
+## 3594. ZEBRAARENA: A Diagnostic Simulation Environment for Studying Reasoning-Action Coupling in Tool-Augmented LLMs
+
+- **ID:** `2603.18614` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Wanjia Zhao, Ludwig Schmidt, James Zou, et al.
+- **abs:** https://arxiv.org/abs/2603.18614 · **PDF:** https://arxiv.org/pdf/2603.18614v1
+
+## 3595. Cyber-Resilient Digital Twins: Discriminating Attacks for Safe Critical Infrastructure Control
+
+- **ID:** `2603.18613` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Mohammadhossein Homaei, Iman Khazrak, Rubén Molano, et al.
+- **abs:** https://arxiv.org/abs/2603.18613 · **PDF:** https://arxiv.org/pdf/2603.18613v1
+
+## 3596. DiscoPhon: Benchmarking the Unsupervised Discovery of Phoneme Inventories With Discrete Speech Units
+
+- **ID:** `2603.18612` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Maxime Poli, Manel Khentout, Angelo Ortiz Tandazo, et al.
+- **abs:** https://arxiv.org/abs/2603.18612 · **PDF:** https://arxiv.org/pdf/2603.18612v1
+
+## 3597. Cross-Modal Rationale Transfer for Explainable Humanitarian Classification on Social Media
+
+- **ID:** `2603.18611` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Thi Huyen Nguyen, Koustav Rudra, Wolfgang Nejdl
+- **abs:** https://arxiv.org/abs/2603.18611 · **PDF:** https://arxiv.org/pdf/2603.18611v1
+
+## 3598. AutORAN: LLM-driven Natural Language Programming for Agile xApp Development
+
+- **ID:** `2603.18604` · **primary:** `cs.NI` · **Submitted:** 2026-03-19
+- **Authors:** Xin Li, Shiming Yu, Leming Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.18604 · **PDF:** https://arxiv.org/pdf/2603.18604v1
+
+## 3599. Brain-inspired AI for Edge Intelligence: a systematic review
+
+- **ID:** `2603.26722` · **primary:** `cs.NE` · **Submitted:** 2026-03-19
+- **Authors:** Yingchao Cheng, Meijia Wang, Zhifeng Hao, et al.
+- **abs:** https://arxiv.org/abs/2603.26722 · **PDF:** https://arxiv.org/pdf/2603.26722v1
+
+## 3600. myMNIST: Benchmark of PETNN, KAN, and Classical Deep Learning Models for Burmese Handwritten Digit Recognition
+
+- **ID:** `2603.18597` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Ye Kyaw Thu, Thazin Myint Oo, Thepchai Supnithi
+- **abs:** https://arxiv.org/abs/2603.18597 · **PDF:** https://arxiv.org/pdf/2603.18597v2
+
+## 3601. Elastic Weight Consolidation Done Right for Continual Learning
+
+- **ID:** `2603.18596` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Xuan Liu, Xiaobin Chang
+- **abs:** https://arxiv.org/abs/2603.18596 · **PDF:** https://arxiv.org/pdf/2603.18596v3
+
+## 3602. Language Model Maps for Prompt-Response Distributions via Log-Likelihood Vectors
+
+- **ID:** `2603.18593` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Yusuke Takase, Momose Oyama, Hidetoshi Shimodaira
+- **abs:** https://arxiv.org/abs/2603.18593 · **PDF:** https://arxiv.org/pdf/2603.18593v1
+
+## 3603. From Human Interfaces to Agent Interfaces: Rethinking Software Design in the Age of AI-Native Systems
+
+- **ID:** `2603.20300` · **primary:** `cs.SE` · **Submitted:** 2026-03-19
+- **Authors:** Shaolin Wang, Yi Mei, Haoyang Che, et al.
+- **abs:** https://arxiv.org/abs/2603.20300 · **PDF:** https://arxiv.org/pdf/2603.20300v1
+
+## 3604. DAPA: Distribution Aware Piecewise Activation Functions for On-Device Transformer Inference and Training
+
+- **ID:** `2603.19338` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Maoyang Xiang, Bo Wang
+- **abs:** https://arxiv.org/abs/2603.19338 · **PDF:** https://arxiv.org/pdf/2603.19338v1
+
+## 3605. HCAG: Hierarchical Abstraction and Retrieval-Augmented Generation on Theoretical Repositories with LLMs
+
+- **ID:** `2603.20299` · **primary:** `cs.SE` · **Submitted:** 2026-03-19
+- **Authors:** Yusen Wu, Xiaotie Deng
+- **abs:** https://arxiv.org/abs/2603.20299 · **PDF:** https://arxiv.org/pdf/2603.20299v1
+
+## 3606. Breaking Hard Isomorphism Benchmarks with DRESS
+
+- **ID:** `2603.18582` · **primary:** `cs.DS` · **Submitted:** 2026-03-19
+- **Authors:** Eduar Castrillo Velilla
+- **abs:** https://arxiv.org/abs/2603.18582 · **PDF:** https://arxiv.org/pdf/2603.18582v2
+
+## 3607. WarPGNN: A Parametric Thermal Warpage Analysis Framework with Physics-aware Graph Neural Network
+
+- **ID:** `2603.18581` · **primary:** `cs.AR` · **Submitted:** 2026-03-19
+- **Authors:** Haotian Lu, Jincong Lu, Sachin Sachdeva, et al.
+- **abs:** https://arxiv.org/abs/2603.18581 · **PDF:** https://arxiv.org/pdf/2603.18581v1
+
+## 3608. ICE: Intervention-Consistent Explanation Evaluation with Statistical Grounding for LLMs
+
+- **ID:** `2603.18579` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Abhinaba Basu, Pavan Chakraborty
+- **abs:** https://arxiv.org/abs/2603.18579 · **PDF:** https://arxiv.org/pdf/2603.18579v1
+
+## 3609. MedForge: Interpretable Medical Deepfake Detection via Forgery-aware Reasoning
+
+- **ID:** `2603.18577` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Zhihui Chen, Kai He, Qingyuan Lei, et al.
+- **abs:** https://arxiv.org/abs/2603.18577 · **PDF:** https://arxiv.org/pdf/2603.18577v1
+
+## 3610. Interplay: Training Independent Simulators for Reference-Free Conversational Recommendation
+
+- **ID:** `2603.18573` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Jerome Ramos, Feng Xia, Xi Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.18573 · **PDF:** https://arxiv.org/pdf/2603.18573v1
+
+## 3611. CAPSUL: A Comprehensive Human Protein Benchmark for Subcellular Localization
+
+- **ID:** `2603.18571` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Yicheng Hu, Xinyu Lin, Shulin Li, et al.
+- **abs:** https://arxiv.org/abs/2603.18571 · **PDF:** https://arxiv.org/pdf/2603.18571v1
+
+## 3612. Attack by Unlearning: Unlearning-Induced Adversarial Attacks on Graph Neural Networks
+
+- **ID:** `2603.18570` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Jiahao Zhang, Yilong Wang, Suhang Wang
+- **abs:** https://arxiv.org/abs/2603.18570 · **PDF:** https://arxiv.org/pdf/2603.18570v1
+
+## 3613. SpecForge: A Flexible and Efficient Open-Source Training Framework for Speculative Decoding
+
+- **ID:** `2603.18567` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Shenggui Li, Chao Wang, Yikai Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.18567 · **PDF:** https://arxiv.org/pdf/2603.18567v1
+
+## 3614. Transformers Learn Robust In-Context Regression under Distributional Uncertainty
+
+- **ID:** `2603.18564` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Hoang T. H. Cao, Hai D. V. Trinh, Tho Quan, et al.
+- **abs:** https://arxiv.org/abs/2603.18564 · **PDF:** https://arxiv.org/pdf/2603.18564v1
+
+## 3615. Reasonably reasoning AI agents can avoid game-theoretic failures in zero-shot, provably
+
+- **ID:** `2603.18563` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Enoch Hyunwook Kang
+- **abs:** https://arxiv.org/abs/2603.18563 · **PDF:** https://arxiv.org/pdf/2603.18563v1
+
+## 3616. CausalVAD: De-confounding End-to-End Autonomous Driving via Causal Intervention
+
+- **ID:** `2603.18561` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Jiacheng Tang, Zhiyuan Zhou, Zhuolin He, et al.
+- **abs:** https://arxiv.org/abs/2603.18561 · **PDF:** https://arxiv.org/pdf/2603.18561v1
+
+## 3617. HiMu: Hierarchical Multimodal Frame Selection for Long Video Question Answering
+
+- **ID:** `2603.18558` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Dan Ben-Ami, Gabriele Serussi, Kobi Cohen, et al.
+- **abs:** https://arxiv.org/abs/2603.18558 · **PDF:** https://arxiv.org/pdf/2603.18558v1
+
+## 3618. Cross-Lingual LLM-Judge Transfer via Evaluation Decomposition
+
+- **ID:** `2603.18557` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Ivaxi Sheth, Zeno Jonke, Amin Mantrach, et al.
+- **abs:** https://arxiv.org/abs/2603.18557 · **PDF:** https://arxiv.org/pdf/2603.18557v1
+
+## 3619. Learning Decision-Sufficient Representations for Linear Optimization
+
+- **ID:** `2603.18551` · **primary:** `math.OC` · **Submitted:** 2026-03-19
+- **Authors:** Yuhan Ye, Saurabh Amin, Asuman Ozdaglar
+- **abs:** https://arxiv.org/abs/2603.18551 · **PDF:** https://arxiv.org/pdf/2603.18551v1
+
+## 3620. SINDy-KANs: Sparse identification of non-linear dynamics through Kolmogorov-Arnold networks
+
+- **ID:** `2603.18548` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Amanda A. Howard, Nicholas Zolman, Bruno Jacob, et al.
+- **abs:** https://arxiv.org/abs/2603.18548 · **PDF:** https://arxiv.org/pdf/2603.18548v1
+
+## 3621. HEP Statistical Inference for UAV Fault Detection: CLs, LRT, and SBI Applied to Blade Damage
+
+- **ID:** `2603.18546` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Khushiyant
+- **abs:** https://arxiv.org/abs/2603.18546 · **PDF:** https://arxiv.org/pdf/2603.18546v1
+
+## 3622. CoDA: Exploring Chain-of-Distribution Attacks and Post-Hoc Token-Space Repair for Medical Vision-Language Models
+
+- **ID:** `2603.18545` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Xiang Chen, Fangfang Yang, Chunlei Meng, et al.
+- **abs:** https://arxiv.org/abs/2603.18545 · **PDF:** https://arxiv.org/pdf/2603.18545v1
+
+## 3623. SCISSR: Scribble-Conditioned Interactive Surgical Segmentation and Refinement
+
+- **ID:** `2603.18544` · **primary:** `eess.IV` · **Submitted:** 2026-03-19
+- **Authors:** Haonan Ping, Jian Jiang, Cheng Yuan, et al.
+- **abs:** https://arxiv.org/abs/2603.18544 · **PDF:** https://arxiv.org/pdf/2603.18544v1
+
+## 3624. GAPSL: A Gradient-Aligned Parallel Split Learning on Heterogeneous Data
+
+- **ID:** `2603.18540` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Zheng Lin, Ons Aouedi, Wei Ni, et al.
+- **abs:** https://arxiv.org/abs/2603.18540 · **PDF:** https://arxiv.org/pdf/2603.18540v1
+
+## 3625. iSatCR: Graph-Empowered Joint Onboard Computing and Routing for LEO Data Delivery
+
+- **ID:** `2603.18539` · **primary:** `cs.NI` · **Submitted:** 2026-03-19
+- **Authors:** Jiangtao Luo, Bingbing Xu, Shaohua Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.18539 · **PDF:** https://arxiv.org/pdf/2603.18539v1
+
+## 3626. Beyond Passive Aggregation: Active Auditing and Topology-Aware Defense in Decentralized Federated Learning
+
+- **ID:** `2603.18538` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Sheng Pan, Niansheng Tang
+- **abs:** https://arxiv.org/abs/2603.18538 · **PDF:** https://arxiv.org/pdf/2603.18538v1
+
+## 3627. Transformer-Based Predictive Maintenance for Risk-Aware Instrument Calibration
+
+- **ID:** `2603.20297` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Adithya Parthasarathy, Aswathnarayan Muthukrishnan Kirubakaran, Akshay Deshpande, et al.
+- **abs:** https://arxiv.org/abs/2603.20297 · **PDF:** https://arxiv.org/pdf/2603.20297v1
+
+## 3628. Data-efficient pre-training by scaling synthetic megadocs
+
+- **ID:** `2603.18534` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Konwoo Kim, Suhas Kotha, Yejin Choi, et al.
+- **abs:** https://arxiv.org/abs/2603.18534 · **PDF:** https://arxiv.org/pdf/2603.18534v1
+
+## 3629. Balancing the Reasoning Load: Difficulty-Differentiated Policy Optimization with Length Redistribution for Efficient and Robust Reinforcement Learning
+
+- **ID:** `2603.18533` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Yinan Xia, Haotian Zhang, Huiming Wang
+- **abs:** https://arxiv.org/abs/2603.18533 · **PDF:** https://arxiv.org/pdf/2603.18533v2
+
+## 3630. Scaling Sim-to-Real Reinforcement Learning for Robot VLAs with Generative 3D Worlds
+
+- **ID:** `2603.18532` · **primary:** `cs.RO` · **Submitted:** 2026-03-19
+- **Authors:** Andrew Choi, Xinjie Wang, Zhizhong Su, et al.
+- **abs:** https://arxiv.org/abs/2603.18532 · **PDF:** https://arxiv.org/pdf/2603.18532v2
+
+## 3631. When Names Change Verdicts: Intervention Consistency Reveals Systematic Bias in LLM Decision-Making
+
+- **ID:** `2603.18530` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Abhinaba Basu, Pavan Chakraborty
+- **abs:** https://arxiv.org/abs/2603.18530 · **PDF:** https://arxiv.org/pdf/2603.18530v1
+
+## 3632. Correlation-Weighted Multi-Reward Optimization for Compositional Generation
+
+- **ID:** `2603.18528` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Jungmyung Wi, Hyunsoo Kim, Donghyun Kim
+- **abs:** https://arxiv.org/abs/2603.18528 · **PDF:** https://arxiv.org/pdf/2603.18528v1
+
+## 3633. Counting Circuits: Mechanistic Interpretability of Visual Reasoning in Large Vision-Language Models
+
+- **ID:** `2603.18523` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Liwei Che, Zhiyu Xue, Yihao Quan, et al.
+- **abs:** https://arxiv.org/abs/2603.18523 · **PDF:** https://arxiv.org/pdf/2603.18523v1
+
+## 3634. On the Peril of (Even a Little) Nonstationarity in Satisficing Regret Minimization
+
+- **ID:** `2603.18514` · **primary:** `stat.ML` · **Submitted:** 2026-03-19
+- **Authors:** Yixuan Zhang, Ruihao Zhu, Qiaomin Xie
+- **abs:** https://arxiv.org/abs/2603.18514 · **PDF:** https://arxiv.org/pdf/2603.18514v1
+
+## 3635. CAFlow: Adaptive-Depth Single-Step Flow Matching for Efficient Histopathology Super-Resolution
+
+- **ID:** `2603.18513` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Elad Yoshai, Ariel D. Yoshai, Natan T. Shaked
+- **abs:** https://arxiv.org/abs/2603.18513 · **PDF:** https://arxiv.org/pdf/2603.18513v1
+
+## 3636. Foundations and Architectures of Artificial Intelligence for Motor Insurance
+
+- **ID:** `2603.18508` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Teerapong Panboonyuen
+- **abs:** https://arxiv.org/abs/2603.18508 · **PDF:** https://arxiv.org/pdf/2603.18508v1
+
+## 3637. Expert Personas Improve LLM Alignment but Damage Accuracy: Bootstrapping Intent-Based Persona Routing with PRISM
+
+- **ID:** `2603.18507` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Zizhao Hu, Mohammad Rostami, Jesse Thomason
+- **abs:** https://arxiv.org/abs/2603.18507 · **PDF:** https://arxiv.org/pdf/2603.18507v1
+
+## 3638. Efficient Video Diffusion with Sparse Information Transmission for Video Compression
+
+- **ID:** `2603.18501` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Mingde Zhou, Zheng Chen, Yulun Zhang
+- **abs:** https://arxiv.org/abs/2603.18501 · **PDF:** https://arxiv.org/pdf/2603.18501v1
+
+## 3639. Recovering Sparse Neural Connectivity from Partial Measurements: A Covariance-Based Approach with Granger-Causality Refinement
+
+- **ID:** `2603.18497` · **primary:** `q-bio.QM` · **Submitted:** 2026-03-19
+- **Authors:** Quilee Simeon
+- **abs:** https://arxiv.org/abs/2603.18497 · **PDF:** https://arxiv.org/pdf/2603.18497v1
+
+## 3640. Cross-Domain Demo-to-Code via Neurosymbolic Counterfactual Reasoning
+
+- **ID:** `2603.18495` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Jooyoung Kim, Wonje Choi, Younguk Song, et al.
+- **abs:** https://arxiv.org/abs/2603.18495 · **PDF:** https://arxiv.org/pdf/2603.18495v1
+
+## 3641. FILT3R: Latent State Adaptive Kalman Filter for Streaming 3D Reconstruction
+
+- **ID:** `2603.18493` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Seonghyun Jin, Jong Chul Ye
+- **abs:** https://arxiv.org/abs/2603.18493 · **PDF:** https://arxiv.org/pdf/2603.18493v1
+
+## 3642. AIMER: Calibration-Free Task-Agnostic MoE Pruning
+
+- **ID:** `2603.18492` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Zongfang Liu, Shengkun Tang, Yifan Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.18492 · **PDF:** https://arxiv.org/pdf/2603.18492v1
+
+## 3643. EntropyCache: Decoded Token Entropy Guided KV Caching for Diffusion Language Models
+
+- **ID:** `2603.18489` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Minsoo Cheong, Donghyun Son, Woosang Lim, et al.
+- **abs:** https://arxiv.org/abs/2603.18489 · **PDF:** https://arxiv.org/pdf/2603.18489v1
+
+## 3644. Collaborative Adaptive Curriculum for Progressive Knowledge Distillation
+
+- **ID:** `2603.20296` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Jing Liu, Zhenchao Ma, Han Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.20296 · **PDF:** https://arxiv.org/pdf/2603.20296v1
+
+## 3645. Diffusion-Guided Semantic Consistency for Multimodal Heterogeneity
+
+- **ID:** `2603.19337` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Jing Liu, Zhengliang Guo, Yan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.19337 · **PDF:** https://arxiv.org/pdf/2603.19337v1
+
+## 3646. Precise Performance of Linear Denoisers in the Proportional Regime
+
+- **ID:** `2603.18483` · **primary:** `stat.ML` · **Submitted:** 2026-03-19
+- **Authors:** Reza Ghane, Danil Akhtiamov, Babak Hassibi
+- **abs:** https://arxiv.org/abs/2603.18483 · **PDF:** https://arxiv.org/pdf/2603.18483v1
+
+## 3647. The Truncation Blind Spot: How Decoding Strategies Systematically Exclude Human-Like Token Choices
+
+- **ID:** `2603.18482` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Esteban Garces Arias, Nurzhan Sapargali, Christian Heumann, et al.
+- **abs:** https://arxiv.org/abs/2603.18482 · **PDF:** https://arxiv.org/pdf/2603.18482v1
+
+## 3648. T-QPM: Enabling Temporal Out-Of-Distribution Detection and Domain Generalization for Vision-Language Models in Open-World
+
+- **ID:** `2603.18481` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Aditi Naiknaware, Salimeh Sekeh
+- **abs:** https://arxiv.org/abs/2603.18481 · **PDF:** https://arxiv.org/pdf/2603.18481v1
+
+## 3649. Do Vision Language Models Understand Human Engagement in Games?
+
+- **ID:** `2603.18480` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Ziyi Wang, Qizan Guo, Rishitosh Singh, et al.
+- **abs:** https://arxiv.org/abs/2603.18480 · **PDF:** https://arxiv.org/pdf/2603.18480v1
+
+## 3650. UniFluids: Unified Neural Operator Learning with Conditional Flow-matching
+
+- **ID:** `2603.22309` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Haosen Li, Qi Meng, Jiahao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.22309 · **PDF:** https://arxiv.org/pdf/2603.22309v1
+
+## 3651. WASD: Locating Critical Neurons as Sufficient Conditions for Explaining and Controlling LLM Behavior
+
+- **ID:** `2603.18474` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Haonan Yu, Junhao Liu, Zhenyu Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.18474 · **PDF:** https://arxiv.org/pdf/2603.18474v1
+
+## 3652. Do Post-Training Algorithms Actually Differ? A Controlled Study Across Model Scales Uncovers Scale-Dependent Ranking Inversions
+
+- **ID:** `2603.19335` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Xiaoyi Li
+- **abs:** https://arxiv.org/abs/2603.19335 · **PDF:** https://arxiv.org/pdf/2603.19335v1
+
+## 3653. Cognitive Mismatch in Multimodal Large Language Models for Discrete Symbol Understanding
+
+- **ID:** `2603.18472` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Yinghui Li, Jiayi Kuang, Peng Xing, et al.
+- **abs:** https://arxiv.org/abs/2603.18472 · **PDF:** https://arxiv.org/pdf/2603.18472v1
+
+## 3654. GAIN: A Benchmark for Goal-Aligned Decision-Making of Large Language Models under Imperfect Norms
+
+- **ID:** `2603.18469` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Masayuki Kawarada, Kodai Watanabe, Soichiro Murakami
+- **abs:** https://arxiv.org/abs/2603.18469 · **PDF:** https://arxiv.org/pdf/2603.18469v1
+
+## 3655. AcceRL: A Distributed Asynchronous Reinforcement Learning and World Model Framework for Vision-Language-Action Models
+
+- **ID:** `2603.18464` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Chengxuan Lu, Shukuan Wang, Yanjie Li, et al.
+- **abs:** https://arxiv.org/abs/2603.18464 · **PDF:** https://arxiv.org/pdf/2603.18464v2
+
+## 3656. AlignMamba-2: Enhancing Multimodal Fusion and Sentiment Analysis with Modality-Aware Mamba
+
+- **ID:** `2603.18462` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Yan Li, Yifei Xing, Xiangyuan Lan, et al.
+- **abs:** https://arxiv.org/abs/2603.18462 · **PDF:** https://arxiv.org/pdf/2603.18462v1
+
+## 3657. Interpretable Prostate Cancer Detection using a Small Cohort of MRI Images
+
+- **ID:** `2603.18460` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Vahid Monfared, Mohammad Hadi Gharib, Ali Sabri, et al.
+- **abs:** https://arxiv.org/abs/2603.18460 · **PDF:** https://arxiv.org/pdf/2603.18460v1
+
+## 3658. HypeMed: Enhancing Medication Recommendations with Hypergraph-Based Patient Relationships
+
+- **ID:** `2603.18459` · **primary:** `cs.IR` · **Submitted:** 2026-03-19
+- **Authors:** Xiangxu Zhang, Xiao Zhou, Hongteng Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.18459 · **PDF:** https://arxiv.org/pdf/2603.18459v1
+
+## 3659. POET: Power-Oriented Evolutionary Tuning for LLM-Based RTL PPA Optimization
+
+- **ID:** `2603.19333` · **primary:** `cs.AR` · **Submitted:** 2026-03-19
+- **Authors:** Heng Ping, Peiyu Zhang, Zhenkun Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.19333 · **PDF:** https://arxiv.org/pdf/2603.19333v1
+
+## 3660. Seeking Universal Shot Language Understanding Solutions
+
+- **ID:** `2603.18448` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Haoxin Liu, Harshavardhan Kamarthi, Zhiyuan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.18448 · **PDF:** https://arxiv.org/pdf/2603.18448v1
+
+## 3661. SODIUM: From Open Web Data to Queryable Databases
+
+- **ID:** `2603.18447` · **primary:** `cs.DB` · **Submitted:** 2026-03-19
+- **Authors:** Chuxuan Hu, Philip Li, Maxwell Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.18447 · **PDF:** https://arxiv.org/pdf/2603.18447v1
+
+## 3662. UT-ACA: Uncertainty-Triggered Adaptive Context Allocation for Long-Context Inference
+
+- **ID:** `2603.18446` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Lang Zhou, Shuxuan Li, Zhuohao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.18446 · **PDF:** https://arxiv.org/pdf/2603.18446v1
+
+## 3663. Discounted Beta--Bernoulli Reward Estimation for Sample-Efficient Reinforcement Learning with Verifiable Rewards
+
+- **ID:** `2603.18444` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Haechan Kim, Soohyun Ryu, Gyouk Chu, et al.
+- **abs:** https://arxiv.org/abs/2603.18444 · **PDF:** https://arxiv.org/pdf/2603.18444v1
+
+## 3664. MARLIN: Multi-Agent Reinforcement Learning for Incremental DAG Discovery
+
+- **ID:** `2603.20295` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Dong Li, Zhengzhang Chen, Xujiang Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.20295 · **PDF:** https://arxiv.org/pdf/2603.20295v1
+
+## 3665. AS2 -- Attention-Based Soft Answer Sets: An End-to-End Differentiable Neuro-Soft-Symbolic Reasoning Architecture
+
+- **ID:** `2603.18436` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Wael AbdAlmageed
+- **abs:** https://arxiv.org/abs/2603.18436 · **PDF:** https://arxiv.org/pdf/2603.18436v1
+
+## 3666. MLOW: Interpretable Low-Rank Frequency Magnitude Decomposition of Multiple Effects for Time Series Forecasting
+
+- **ID:** `2603.18432` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Runze Yang, Longbing Cao, Xiaoming Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.18432 · **PDF:** https://arxiv.org/pdf/2603.18432v1
+
+## 3667. LLM-Enhanced Energy Contrastive Learning for Out-of-Distribution Detection in Text-Attributed Graphs
+
+- **ID:** `2603.20293` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Xiaoxu Ma, Dong Li, Minglai Shao, et al.
+- **abs:** https://arxiv.org/abs/2603.20293 · **PDF:** https://arxiv.org/pdf/2603.20293v1
+
+## 3668. Towards Noise-Resilient Quantum Multi-Armed and Stochastic Linear Bandits
+
+- **ID:** `2603.18431` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Zhuoyue Chen, Kechao Cai
+- **abs:** https://arxiv.org/abs/2603.18431 · **PDF:** https://arxiv.org/pdf/2603.18431v1
+
+## 3669. Adaptive Decoding via Test-Time Policy Learning for Self-Improving Generation
+
+- **ID:** `2603.18428` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Asmita Bhardwaj, Yuya Jeremy Ong, Eelaaf Zahid, et al.
+- **abs:** https://arxiv.org/abs/2603.18428 · **PDF:** https://arxiv.org/pdf/2603.18428v1
+
+## 3670. R&D: Balancing Reliability and Diversity in Synthetic Data Augmentation for Semantic Segmentation
+
+- **ID:** `2603.18427` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Huy Che, Dinh-Duy Phan, Duc-Khai Lam
+- **abs:** https://arxiv.org/abs/2603.18427 · **PDF:** https://arxiv.org/pdf/2603.18427v1
+
+## 3671. Prune-then-Quantize or Quantize-then-Prune? Understanding the Impact of Compression Order in Joint Model Compression
+
+- **ID:** `2603.18426` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Minjun Kim, Jaehyeon Choi, Hyunwoo Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.18426 · **PDF:** https://arxiv.org/pdf/2603.18426v1
+
+## 3672. Multimodal Task Interference: A Benchmark and Analysis of History-Target Mismatch in Multimodal LLMs
+
+- **ID:** `2603.18425` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Masayuki Kawarada, Tatsuya Ishigaki, Hiroya Takamura
+- **abs:** https://arxiv.org/abs/2603.18425 · **PDF:** https://arxiv.org/pdf/2603.18425v1
+
+## 3673. The Impact of Corporate AI Washing on Farmers' Digital Financial Behavior Response -- An Analysis from the Perspective of Digital Financial Exclusion
+
+- **ID:** `2603.18421` · **primary:** `cs.CY` · **Submitted:** 2026-03-19
+- **Authors:** Zhanjie Wen, Wenxiu Li, Jiechang Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.18421 · **PDF:** https://arxiv.org/pdf/2603.18421v2
+
+## 3674. From Topic to Transition Structure: Unsupervised Concept Discovery at Corpus Scale via Predictive Associative Memory
+
+- **ID:** `2603.18420` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Jason Dury
+- **abs:** https://arxiv.org/abs/2603.18420 · **PDF:** https://arxiv.org/pdf/2603.18420v1
+
+## 3675. Mind the Rarities: Can Rare Skin Diseases Be Reliably Diagnosed via Diagnostic Reasoning?
+
+- **ID:** `2603.18418` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Yang Liu, Jiyao Yang, Hongjin Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.18418 · **PDF:** https://arxiv.org/pdf/2603.18418v1
+
+## 3676. Self-Tuning Sparse Attention: Multi-Fidelity Hyperparameter Optimization for Transformer Acceleration
+
+- **ID:** `2603.18417` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Arundhathi Dev, Justin Zhan
+- **abs:** https://arxiv.org/abs/2603.18417 · **PDF:** https://arxiv.org/pdf/2603.18417v1
+
+## 3677. The Spillover Effects of Peer AI Rinsing on Corporate Green Innovation
+
+- **ID:** `2603.18415` · **primary:** `cs.CY` · **Submitted:** 2026-03-19
+- **Authors:** Li Wenxiu, Wen Zhanjie, Xia Jiechang, et al.
+- **abs:** https://arxiv.org/abs/2603.18415 · **PDF:** https://arxiv.org/pdf/2603.18415v2
+
+## 3678. Statistical Testing Framework for Clustering Pipelines by Selective Inference
+
+- **ID:** `2603.18413` · **primary:** `stat.ML` · **Submitted:** 2026-03-19
+- **Authors:** Yugo Miyata, Tomohiro Shiraishi, Shunichi Nishino, et al.
+- **abs:** https://arxiv.org/abs/2603.18413 · **PDF:** https://arxiv.org/pdf/2603.18413v2
+
+## 3679. TARo: Token-level Adaptive Routing for LLM Test-time Alignment
+
+- **ID:** `2603.18411` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Arushi Rai, Qiang Zhang, Hanqing Zeng, et al.
+- **abs:** https://arxiv.org/abs/2603.18411 · **PDF:** https://arxiv.org/pdf/2603.18411v1
+
+## 3680. TopoChunker: Topology-Aware Agentic Document Chunking Framework
+
+- **ID:** `2603.18409` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Xiaoyu Liu
+- **abs:** https://arxiv.org/abs/2603.18409 · **PDF:** https://arxiv.org/pdf/2603.18409v1
+
+## 3681. HSI Image Enhancement Classification Based on Knowledge Distillation: A Study on Forgetting
+
+- **ID:** `2603.20292` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Songfeng Zhu
+- **abs:** https://arxiv.org/abs/2603.20292 · **PDF:** https://arxiv.org/pdf/2603.20292v1
+
+## 3682. Multi-Domain Empirical Bayes for Linearly-Mixed Causal Representations
+
+- **ID:** `2603.18404` · **primary:** `stat.ML` · **Submitted:** 2026-03-19
+- **Authors:** Bohan Wu, Julius von Kügelgen, David M. Blei
+- **abs:** https://arxiv.org/abs/2603.18404 · **PDF:** https://arxiv.org/pdf/2603.18404v2
+
+## 3683. Stress Classification from ECG Signals Using Vision Transformer
+
+- **ID:** `2603.26721` · **primary:** `eess.SP` · **Submitted:** 2026-03-19
+- **Authors:** Zeeshan Ahmad, Naimul Khan
+- **abs:** https://arxiv.org/abs/2603.26721 · **PDF:** https://arxiv.org/pdf/2603.26721v1
+
+## 3684. FlowMS: Flow Matching for De Novo Structure Elucidation from Mass Spectra
+
+- **ID:** `2603.18397` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Jianan Nie, Peng Gao
+- **abs:** https://arxiv.org/abs/2603.18397 · **PDF:** https://arxiv.org/pdf/2603.18397v1
+
+## 3685. RE-SAC: Disentangling aleatoric and epistemic risks in bus fleet control: A stable and robust ensemble DRL approach
+
+- **ID:** `2603.18396` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Yifan Zhang, Liang Zheng
+- **abs:** https://arxiv.org/abs/2603.18396 · **PDF:** https://arxiv.org/pdf/2603.18396v2
+
+## 3686. SutureAgent: Learning Surgical Trajectories via Goal-conditioned Offline RL in Pixel Space
+
+- **ID:** `2603.26720` · **primary:** `cs.RO` · **Submitted:** 2026-03-19
+- **Authors:** Huanrong Liu, Chunlin Tian, Tongyu Jia, et al.
+- **abs:** https://arxiv.org/abs/2603.26720 · **PDF:** https://arxiv.org/pdf/2603.26720v1
+
+## 3687. Computational and Statistical Hardness of Calibration Distance
+
+- **ID:** `2603.18391` · **primary:** `cs.DS` · **Submitted:** 2026-03-19
+- **Authors:** Mingda Qiao
+- **abs:** https://arxiv.org/abs/2603.18391 · **PDF:** https://arxiv.org/pdf/2603.18391v1
+
+## 3688. AutoScreen-FW: An LLM-based Framework for Resume Screening
+
+- **ID:** `2603.18390` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Zhelin Xu, Shuhei Yamamoto, Atsuyuki Morishima
+- **abs:** https://arxiv.org/abs/2603.18390 · **PDF:** https://arxiv.org/pdf/2603.18390v1
+
+## 3689. An SO(3)-equivariant reciprocal-space neural potential for long-range interactions
+
+- **ID:** `2603.18389` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-19
+- **Authors:** Lingfeng Zhang, Taoyong Cui, Dongzhan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.18389 · **PDF:** https://arxiv.org/pdf/2603.18389v2
+
+## 3690. Reflection in the Dark: Exposing and Escaping the Black Box in Reflective Prompt Optimization
+
+- **ID:** `2603.18388` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Shiyan Liu, Qifeng Xia, Qiyun Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.18388 · **PDF:** https://arxiv.org/pdf/2603.18388v1
+
+## 3691. Mathematical Foundations of Deep Learning
+
+- **ID:** `2603.18387` · **primary:** `cs.LG` · **Submitted:** 2026-03-19
+- **Authors:** Xiaojing Ye
+- **abs:** https://arxiv.org/abs/2603.18387 · **PDF:** https://arxiv.org/pdf/2603.18387v1
+
+## 3692. Evolutionarily Stable Stackelberg Equilibrium
+
+- **ID:** `2603.18385` · **primary:** `cs.GT` · **Submitted:** 2026-03-19
+- **Authors:** Sam Ganzfried
+- **abs:** https://arxiv.org/abs/2603.18385 · **PDF:** https://arxiv.org/pdf/2603.18385v2
+
+## 3693. From Weak Cues to Real Identities: Evaluating Inference-Driven De-Anonymization in LLM Agents
+
+- **ID:** `2603.18382` · **primary:** `cs.AI` · **Submitted:** 2026-03-19
+- **Authors:** Myeongseob Ko, Jihyun Jeong, Sumiran Singh Thakur, et al.
+- **abs:** https://arxiv.org/abs/2603.18382 · **PDF:** https://arxiv.org/pdf/2603.18382v1
+
+## 3694. PlanTwin: Privacy-Preserving Planning Abstractions for Cloud-Assisted LLM Agents
+
+- **ID:** `2603.18377` · **primary:** `cs.CR` · **Submitted:** 2026-03-19
+- **Authors:** Guangsheng Yu, Qin Wang, Rui Lang, et al.
+- **abs:** https://arxiv.org/abs/2603.18377 · **PDF:** https://arxiv.org/pdf/2603.18377v2
+
+## 3695. To See or To Please: Uncovering Visual Sycophancy and Split Beliefs in VLMs
+
+- **ID:** `2603.18373` · **primary:** `cs.CV` · **Submitted:** 2026-03-19
+- **Authors:** Rui Hong, Shuxue Quan
+- **abs:** https://arxiv.org/abs/2603.18373 · **PDF:** https://arxiv.org/pdf/2603.18373v1
+
+## 3696. PowerFlow: Unlocking the Dual Nature of LLMs via Principled Distribution Matching
+
+- **ID:** `2603.18363` · **primary:** `cs.CL` · **Submitted:** 2026-03-19
+- **Authors:** Ruishuo Chen, Yu Chen, Zhuoran Li, et al.
+- **abs:** https://arxiv.org/abs/2603.18363 · **PDF:** https://arxiv.org/pdf/2603.18363v1
+
+## 3697. Synthetic Data Generation for Training Diversified Commonsense Reasoning Models
+
+- **ID:** `2603.18361` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Tianhui Zhang, Bei Peng, Danushka Bollegala
+- **abs:** https://arxiv.org/abs/2603.18361 · **PDF:** https://arxiv.org/pdf/2603.18361v1
+
+## 3698. From Noise to Signal: When Outliers Seed New Topics
+
+- **ID:** `2603.18358` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Evangelia Zve, Gauvain Bourgne, Benjamin Icard, et al.
+- **abs:** https://arxiv.org/abs/2603.18358 · **PDF:** https://arxiv.org/pdf/2603.18358v1
+
+## 3699. LGESynthNet: Controlled Scar Synthesis for Improved Scar Segmentation in Cardiac LGE-MRI Imaging
+
+- **ID:** `2603.18356` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Athira J. Jacob, Puneet Sharma, Daniel Rueckert
+- **abs:** https://arxiv.org/abs/2603.18356 · **PDF:** https://arxiv.org/pdf/2603.18356v1
+
+## 3700. Interpretability without actionability: mechanistic methods cannot correct language model errors despite near-perfect internal representations
+
+- **ID:** `2603.18353` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Sanjay Basu, Sadiq Y. Patel, Parth Sheth, et al.
+- **abs:** https://arxiv.org/abs/2603.18353 · **PDF:** https://arxiv.org/pdf/2603.18353v1
+
+## 3701. Large-Scale Analysis of Persuasive Content on Moltbook
+
+- **ID:** `2603.18349` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Julia Jose, Meghna Manoj Nair, Rachel Greenstadt
+- **abs:** https://arxiv.org/abs/2603.18349 · **PDF:** https://arxiv.org/pdf/2603.18349v2
+
+## 3702. Epistemic Generative Adversarial Networks
+
+- **ID:** `2603.18348` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Muhammad Mubashar, Fabio Cuzzolin
+- **abs:** https://arxiv.org/abs/2603.18348 · **PDF:** https://arxiv.org/pdf/2603.18348v1
+
+## 3703. HRI-SA: A Multimodal Dataset for Online Assessment of Human Situational Awareness during Remote Human-Robot Teaming
+
+- **ID:** `2603.18344` · **primary:** `cs.RO` · **Submitted:** 2026-03-18
+- **Authors:** Hashini Senaratne, Richard Attfield, Samith Widhanapathirana, et al.
+- **abs:** https://arxiv.org/abs/2603.18344 · **PDF:** https://arxiv.org/pdf/2603.18344v1
+
+## 3704. Shifting Uncertainty to Critical Moments: Towards Reliable Uncertainty Quantification for VLA Model
+
+- **ID:** `2603.18342` · **primary:** `cs.RO` · **Submitted:** 2026-03-18
+- **Authors:** Yanchuan Tang, Taowen Wang, Yuefei Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.18342 · **PDF:** https://arxiv.org/pdf/2603.18342v1
+
+## 3705. Can LLMs Reason Like Automated Theorem Provers for Rust Verification? VCoT-Bench: Evaluating via Verification Chain of Thought
+
+- **ID:** `2603.18334` · **primary:** `cs.SE` · **Submitted:** 2026-03-18
+- **Authors:** Zichen Xie, Wenxi Wang
+- **abs:** https://arxiv.org/abs/2603.18334 · **PDF:** https://arxiv.org/pdf/2603.18334v1
+
+## 3706. Understanding the Theoretical Foundations of Deep Neural Networks through Differential Equations
+
+- **ID:** `2603.18331` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Hongjue Zhao, Yizhuo Chen, Yuchen Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.18331 · **PDF:** https://arxiv.org/pdf/2603.18331v1
+
+## 3707. MemArchitect: A Policy Driven Memory Governance Layer
+
+- **ID:** `2603.18330` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Lingavasan Suresh Kumar, Yang Ba, Rong Pan
+- **abs:** https://arxiv.org/abs/2603.18330 · **PDF:** https://arxiv.org/pdf/2603.18330v1
+
+## 3708. FaithSteer-BENCH: A Deployment-Aligned Stress-Testing Benchmark for Inference-Time Steering
+
+- **ID:** `2603.18329` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Zikang Ding, Qiying Hu, Yi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.18329 · **PDF:** https://arxiv.org/pdf/2603.18329v1
+
+## 3709. A Family of Adaptive Activation Functions for Mitigating Failure Modes in Physics-Informed Neural Networks
+
+- **ID:** `2603.18328` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Krishna Murari
+- **abs:** https://arxiv.org/abs/2603.18328 · **PDF:** https://arxiv.org/pdf/2603.18328v1
+
+## 3710. Consumer-to-Clinical Language Shifts in Ambient AI Draft Notes and Clinician-Finalized Documentation: A Multi-level Analysis
+
+- **ID:** `2603.18327` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Ha Na Cho, Yawen Guo, Sairam Sutari, et al.
+- **abs:** https://arxiv.org/abs/2603.18327 · **PDF:** https://arxiv.org/pdf/2603.18327v1
+
+## 3711. Escaping Offline Pessimism: Vector-Field Reward Shaping for Safe Frontier Exploration
+
+- **ID:** `2603.18326` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Amirhossein Roknilamouki, Arnob Ghosh, Eylem Ekici, et al.
+- **abs:** https://arxiv.org/abs/2603.18326 · **PDF:** https://arxiv.org/pdf/2603.18326v1
+
+## 3712. Learning to Reason with Curriculum I: Provable Benefits of Autocurriculum
+
+- **ID:** `2603.18325` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Nived Rajaraman, Audrey Huang, Miro Dudik, et al.
+- **abs:** https://arxiv.org/abs/2603.18325 · **PDF:** https://arxiv.org/pdf/2603.18325v1
+
+## 3713. DriveVLM-RL: Neuroscience-Inspired Reinforcement Learning with Vision-Language Models for Safe and Deployable Autonomous Driving
+
+- **ID:** `2603.18315` · **primary:** `cs.RO` · **Submitted:** 2026-03-18
+- **Authors:** Zilin Huang, Zihao Sheng, Zhengyang Wan, et al.
+- **abs:** https://arxiv.org/abs/2603.18315 · **PDF:** https://arxiv.org/pdf/2603.18315v1
+
+## 3714. Approximate Subgraph Matching with Neural Graph Representations and Reinforcement Learning
+
+- **ID:** `2603.18314` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Kaiyang Li, Shihao Ji, Zhipeng Cai, et al.
+- **abs:** https://arxiv.org/abs/2603.18314 · **PDF:** https://arxiv.org/pdf/2603.18314v1
+
+## 3715. Fast and Generalizable NeRF Architecture Selection for Satellite Scene Reconstruction
+
+- **ID:** `2603.18306` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Devjyoti Chakraborty, Zaki Sukma, Rakandhiya D. Rachmanto, et al.
+- **abs:** https://arxiv.org/abs/2603.18306 · **PDF:** https://arxiv.org/pdf/2603.18306v1
+
+## 3716. Auditing Preferences for Brands and Cultures in LLMs
+
+- **ID:** `2603.18300` · **primary:** `cs.HC` · **Submitted:** 2026-03-18
+- **Authors:** Jasmine Rienecker, Katarina Mpofu, Naman Goel, et al.
+- **abs:** https://arxiv.org/abs/2603.18300 · **PDF:** https://arxiv.org/pdf/2603.18300v1
+
+## 3717. ALIGN: Adversarial Learning for Generalizable Speech Neuroprosthesis
+
+- **ID:** `2603.18299` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Zhanqi Zhang, Shun Li, Bernardo L. Sabatini, et al.
+- **abs:** https://arxiv.org/abs/2603.18299 · **PDF:** https://arxiv.org/pdf/2603.18299v1
+
+## 3718. Sparse3DTrack: Monocular 3D Object Tracking Using Sparse Supervision
+
+- **ID:** `2603.18298` · **primary:** `cs.RO` · **Submitted:** 2026-03-18
+- **Authors:** Nikhil Gosala, B. Ravi Kiran, Senthil Yogamani, et al.
+- **abs:** https://arxiv.org/abs/2603.18298 · **PDF:** https://arxiv.org/pdf/2603.18298v1
+
+## 3719. Path-Constrained Mixture-of-Experts
+
+- **ID:** `2603.18297` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Zijin Gu, Tatiana Likhomanenko, Vimal Thilak, et al.
+- **abs:** https://arxiv.org/abs/2603.18297 · **PDF:** https://arxiv.org/pdf/2603.18297v1
+
+## 3720. Constrained Hybrid Metaheuristic: A Universal Framework for Continuous Optimisation
+
+- **ID:** `2603.18295` · **primary:** `cs.NE` · **Submitted:** 2026-03-18
+- **Authors:** Piotr A. Kowalski, Szymon Kucharczyk, Jacek Mańdziuk
+- **abs:** https://arxiv.org/abs/2603.18295 · **PDF:** https://arxiv.org/pdf/2603.18295v1
+
+## 3721. The Validity Gap in Health AI Evaluation: A Cross-Sectional Analysis of Benchmark Composition
+
+- **ID:** `2603.18294` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Alvin Rajkomar, Pavan Sudarshan, Angela Lai, et al.
+- **abs:** https://arxiv.org/abs/2603.18294 · **PDF:** https://arxiv.org/pdf/2603.18294v1
+
+## 3722. CORE: Robust Out-of-Distribution Detection via Confidence and Orthogonal Residual Scoring
+
+- **ID:** `2603.18290` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Jin Mo Yang, Hyung-Sin Kim, Saewoong Bahk
+- **abs:** https://arxiv.org/abs/2603.18290 · **PDF:** https://arxiv.org/pdf/2603.18290v1
+
+## 3723. Offload or Overload: A Platform Measurement Study of Mobile Robotic Manipulation Workloads
+
+- **ID:** `2603.18284` · **primary:** `cs.RO` · **Submitted:** 2026-03-18
+- **Authors:** Sara Pohland, Xenofon Foukas, Ganesh Ananthanarayanan, et al.
+- **abs:** https://arxiv.org/abs/2603.18284 · **PDF:** https://arxiv.org/pdf/2603.18284v1
+
+## 3724. On Additive Gaussian Processes for Wind Farm Power Prediction
+
+- **ID:** `2603.18281` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Simon M. Brealy, Lawrence A. Bull, Daniel S. Brennan, et al.
+- **abs:** https://arxiv.org/abs/2603.18281 · **PDF:** https://arxiv.org/pdf/2603.18281v1
+
+## 3725. Detection Is Cheap, Routing Is Learned: Why Refusal-Based Alignment Evaluation Fails
+
+- **ID:** `2603.18280` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Gregory N. Frank
+- **abs:** https://arxiv.org/abs/2603.18280 · **PDF:** https://arxiv.org/pdf/2603.18280v1
+
+## 3726. EDM-ARS: A Domain-Specific Multi-Agent System for Automated Educational Data Mining Research
+
+- **ID:** `2603.18273` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Chenguang Pan, Zhou Zhang, Weixuan Xiao, et al.
+- **abs:** https://arxiv.org/abs/2603.18273 · **PDF:** https://arxiv.org/pdf/2603.18273v1
+
+## 3727. Retrieval-Augmented LLM Agents: Learning to Learn from Experience
+
+- **ID:** `2603.18272` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Thomas Palmeira Ferraz, Romain Deffayet, Vassilina Nikoulina, et al.
+- **abs:** https://arxiv.org/abs/2603.18272 · **PDF:** https://arxiv.org/pdf/2603.18272v1
+
+## 3728. Enactor: From Traffic Simulators to Surrogate World Models
+
+- **ID:** `2603.18266` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Yash Ranjan, Rahul Sengupta, Anand Rangarajan, et al.
+- **abs:** https://arxiv.org/abs/2603.18266 · **PDF:** https://arxiv.org/pdf/2603.18266v1
+
+## 3729. LRConv-NeRV: Low Rank Convolution for Efficient Neural Video Compression
+
+- **ID:** `2603.18261` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Tamer Shanableh
+- **abs:** https://arxiv.org/abs/2603.18261 · **PDF:** https://arxiv.org/pdf/2603.18261v1
+
+## 3730. Sharpness-Aware Minimization in Logit Space Efficiently Enhances Direct Preference Optimization
+
+- **ID:** `2603.18258` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Haocheng Luo, Zehang Deng, Thanh-Toan Do, et al.
+- **abs:** https://arxiv.org/abs/2603.18258 · **PDF:** https://arxiv.org/pdf/2603.18258v1
+
+## 3731. Discovering What You Can Control: Interventional Boundary Discovery for Reinforcement Learning
+
+- **ID:** `2603.18257` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Jiaxin Liu
+- **abs:** https://arxiv.org/abs/2603.18257 · **PDF:** https://arxiv.org/pdf/2603.18257v1
+
+## 3732. MolRGen: A Training and Evaluation Setting for De Novo Molecular Generation with Reasonning Models
+
+- **ID:** `2603.18256` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Philippe Formont, Maxime Darrin, Ismail Ben Ayed, et al.
+- **abs:** https://arxiv.org/abs/2603.18256 · **PDF:** https://arxiv.org/pdf/2603.18256v1
+
+## 3733. Computation-Utility-Privacy Tradeoffs in Bayesian Estimation
+
+- **ID:** `2603.18254` · **primary:** `cs.DS` · **Submitted:** 2026-03-18
+- **Authors:** Sitan Chen, Jingqiu Ding, Mahbod Majid, et al.
+- **abs:** https://arxiv.org/abs/2603.18254 · **PDF:** https://arxiv.org/pdf/2603.18254v1
+
+## 3734. AGRI-Fidelity: Evaluating the Reliability of Listenable Explanations for Poultry Disease Detection
+
+- **ID:** `2603.18247` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Sindhuja Madabushi, Arda Dogan, Jonathan Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.18247 · **PDF:** https://arxiv.org/pdf/2603.18247v1
+
+## 3735. Impact of automatic speech recognition quality on Alzheimer's disease detection from spontaneous speech: a reproducible benchmark study with lexical modeling and statistical validation
+
+- **ID:** `2603.18239` · **primary:** `q-bio.QM` · **Submitted:** 2026-03-18
+- **Authors:** Himadri Samanta
+- **abs:** https://arxiv.org/abs/2603.18239 · **PDF:** https://arxiv.org/pdf/2603.18239v1
+
+## 3736. Gradient-Informed Temporal Sampling Improves Rollout Accuracy in PDE Surrogate Training
+
+- **ID:** `2603.18237` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Wenshuo Wang, Fan Zhang
+- **abs:** https://arxiv.org/abs/2603.18237 · **PDF:** https://arxiv.org/pdf/2603.18237v1
+
+## 3737. FalconBC: Flow matching for Amortized inference of Latent-CONditioned physiologic Boundary Conditions
+
+- **ID:** `2603.19331` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Chloe H. Choi, Alison L. Marsden, Daniele E. Schiavazzi
+- **abs:** https://arxiv.org/abs/2603.19331 · **PDF:** https://arxiv.org/pdf/2603.19331v1
+
+## 3738. PAI: Fast, Accurate, and Full Benchmark Performance Projection with AI
+
+- **ID:** `2603.19330` · **primary:** `cs.AR` · **Submitted:** 2026-03-18
+- **Authors:** Avery Johnson, Mohammad Majharul Islam, Riad Akram, et al.
+- **abs:** https://arxiv.org/abs/2603.19330 · **PDF:** https://arxiv.org/pdf/2603.19330v1
+
+## 3739. A Hybrid Conditional Diffusion-DeepONet Framework for High-Fidelity Stress Prediction in Hyperelastic Materials
+
+- **ID:** `2603.18225` · **primary:** `stat.ML` · **Submitted:** 2026-03-18
+- **Authors:** Purna Vindhya Kota, Meer Mehran Rashid, Somdatta Goswami, et al.
+- **abs:** https://arxiv.org/abs/2603.18225 · **PDF:** https://arxiv.org/pdf/2603.18225v1
+
+## 3740. Tackling the Sign Problem in the Doped Hubbard Model with Normalizing Flows
+
+- **ID:** `2603.18205` · **primary:** `cond-mat.str-el` · **Submitted:** 2026-03-18
+- **Authors:** Dominic Schuh, Lena Funcke, Janik Kreit, et al.
+- **abs:** https://arxiv.org/abs/2603.18205 · **PDF:** https://arxiv.org/pdf/2603.18205v1
+
+## 3741. How Psychological Learning Paradigms Shaped and Constrained Artificial Intelligence
+
+- **ID:** `2603.18203` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Alex Anvi Eponon, Ildar Batyrshin, Christian E. Maldonado-Sifuentes, et al.
+- **abs:** https://arxiv.org/abs/2603.18203 · **PDF:** https://arxiv.org/pdf/2603.18203v2
+
+## 3742. R2-Dreamer: Redundancy-Reduced World Models without Decoders or Augmentation
+
+- **ID:** `2603.18202` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Naoki Morihira, Amal Nahar, Kartik Bharadwaj, et al.
+- **abs:** https://arxiv.org/abs/2603.18202 · **PDF:** https://arxiv.org/pdf/2603.18202v2
+
+## 3743. A Computationally Efficient Learning of Artificial Intelligence System Reliability Considering Error Propagation
+
+- **ID:** `2603.18201` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Fenglian Pan, Yinwei Zhang, Yili Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.18201 · **PDF:** https://arxiv.org/pdf/2603.18201v1
+
+## 3744. Access Controlled Website Interaction for Agentic AI with Delegated Critical Tasks
+
+- **ID:** `2603.18197` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Sunyoung Kim, Hokeun Kim
+- **abs:** https://arxiv.org/abs/2603.18197 · **PDF:** https://arxiv.org/pdf/2603.18197v1
+
+## 3745. Retrieval-Augmented LLMs for Security Incident Analysis
+
+- **ID:** `2603.18196` · **primary:** `cs.CR` · **Submitted:** 2026-03-18
+- **Authors:** Xavier Cadet, Aditya Vikram Singh, Harsh Mamania, et al.
+- **abs:** https://arxiv.org/abs/2603.18196 · **PDF:** https://arxiv.org/pdf/2603.18196v2
+
+## 3746. Goedel-Code-Prover: Hierarchical Proof Search for Open State-of-the-Art Code Verification
+
+- **ID:** `2603.19329` · **primary:** `cs.SE` · **Submitted:** 2026-03-18
+- **Authors:** Zenan Li, Ziran Yang, Deyuan He, et al.
+- **abs:** https://arxiv.org/abs/2603.19329 · **PDF:** https://arxiv.org/pdf/2603.19329v2
+
+## 3747. Starting Off on the Wrong Foot: Pitfalls in Data Preparation
+
+- **ID:** `2603.18190` · **primary:** `stat.ML` · **Submitted:** 2026-03-18
+- **Authors:** Jiayi Guo, Panyi Dong, Zhiyu Quan
+- **abs:** https://arxiv.org/abs/2603.18190 · **PDF:** https://arxiv.org/pdf/2603.18190v1
+
+## 3748. TeachingCoach: A Fine-Tuned Scaffolding Chatbot for Instructional Guidance to Instructors
+
+- **ID:** `2603.18189` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Isabel Molnar, Peiyu Li, Si Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.18189 · **PDF:** https://arxiv.org/pdf/2603.18189v1
+
+## 3749. CWoMP: Morpheme Representation Learning for Interlinear Glossing
+
+- **ID:** `2603.18184` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Morris Alper, Enora Rice, Bhargav Shandilya, et al.
+- **abs:** https://arxiv.org/abs/2603.18184 · **PDF:** https://arxiv.org/pdf/2603.18184v1
+
+## 3750. VLM-AutoDrive: Post-Training Vision-Language Models for Safety-Critical Autonomous Driving Events
+
+- **ID:** `2603.18178` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Mohammad Qazim Bhat, Yufan Huang, Niket Agarwal, et al.
+- **abs:** https://arxiv.org/abs/2603.18178 · **PDF:** https://arxiv.org/pdf/2603.18178v1
+
+## 3751. Conflict-Free Policy Languages for Probabilistic ML Predicates: A Framework and Case Study with the Semantic Router DSL
+
+- **ID:** `2603.18174` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Xunzhuo Liu, Hao Wu, Huamin Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.18174 · **PDF:** https://arxiv.org/pdf/2603.18174v1
+
+## 3752. GRAFITE: Generative Regression Analysis Framework for Issue Tracking and Evaluation
+
+- **ID:** `2603.18173` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Ja Young Lee, Mírian Silva, Mohamed Nasr, et al.
+- **abs:** https://arxiv.org/abs/2603.18173 · **PDF:** https://arxiv.org/pdf/2603.18173v1
+
+## 3753. Modeling the human lexicon under temperature variations: linguistic factors, diversity and typicality in LLM word associations
+
+- **ID:** `2603.18171` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Maria Andueza Rodriguez, Marie Candito, Richard Huyghe
+- **abs:** https://arxiv.org/abs/2603.18171 · **PDF:** https://arxiv.org/pdf/2603.18171v1
+
+## 3754. ResNets of All Shapes and Sizes: Convergence of Training Dynamics in the Large-scale Limit
+
+- **ID:** `2603.18168` · **primary:** `stat.ML` · **Submitted:** 2026-03-18
+- **Authors:** Louis-Pierre Chaintron, Lénaïc Chizat, Javier Maass
+- **abs:** https://arxiv.org/abs/2603.18168 · **PDF:** https://arxiv.org/pdf/2603.18168v1
+
+## 3755. Efficient Dense Crowd Trajectory Prediction Via Dynamic Clustering
+
+- **ID:** `2603.18166` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Antonius Bima Murti Wijaya, Paul Henderson, Marwa Mahmoud
+- **abs:** https://arxiv.org/abs/2603.18166 · **PDF:** https://arxiv.org/pdf/2603.18166v1
+
+## 3756. How LLMs Distort Our Written Language
+
+- **ID:** `2603.18161` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Marwa Abdulhai, Isadora White, Yanming Wan, et al.
+- **abs:** https://arxiv.org/abs/2603.18161 · **PDF:** https://arxiv.org/pdf/2603.18161v1
+
+## 3757. Learning-Augmented Algorithms for $k$-median via Online Learning
+
+- **ID:** `2603.18157` · **primary:** `cs.DS` · **Submitted:** 2026-03-18
+- **Authors:** Anish Hebbar, Rong Ge, Amit Kumar, et al.
+- **abs:** https://arxiv.org/abs/2603.18157 · **PDF:** https://arxiv.org/pdf/2603.18157v1
+
+## 3758. Towards sample-optimal learning of bosonic Gaussian quantum states
+
+- **ID:** `2603.18136` · **primary:** `quant-ph` · **Submitted:** 2026-03-18
+- **Authors:** Senrui Chen, Francesco Anna Mele, Marco Fanizza, et al.
+- **abs:** https://arxiv.org/abs/2603.18136 · **PDF:** https://arxiv.org/pdf/2603.18136v1
+
+## 3759. Unified Spatio-Temporal Token Scoring for Efficient Video VLMs
+
+- **ID:** `2603.18004` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Jianrui Zhang, Yue Yang, Rohun Tripathi, et al.
+- **abs:** https://arxiv.org/abs/2603.18004 · **PDF:** https://arxiv.org/pdf/2603.18004v1
+
+## 3760. Loc3R-VLM: Language-based Localization and 3D Reasoning with Vision-Language Models
+
+- **ID:** `2603.18002` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Kevin Qu, Haozhe Qi, Mihai Dusmanu, et al.
+- **abs:** https://arxiv.org/abs/2603.18002 · **PDF:** https://arxiv.org/pdf/2603.18002v1
+
+## 3761. AgentFactory: A Self-Evolving Framework Through Executable Subagent Accumulation and Reuse
+
+- **ID:** `2603.18000` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Zhang Zhang, Shuqi Lu, Hongjin Qian, et al.
+- **abs:** https://arxiv.org/abs/2603.18000 · **PDF:** https://arxiv.org/pdf/2603.18000v1
+
+## 3762. LoST: Level of Semantics Tokenization for 3D Shapes
+
+- **ID:** `2603.17995` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Niladri Shekhar Dutt, Zifan Shi, Paul Guerrero, et al.
+- **abs:** https://arxiv.org/abs/2603.17995 · **PDF:** https://arxiv.org/pdf/2603.17995v1
+
+## 3763. Final Report for the Workshop on Robotics & AI in Medicine
+
+- **ID:** `2603.18130` · **primary:** `cs.RO` · **Submitted:** 2026-03-18
+- **Authors:** Juan P Wachs
+- **abs:** https://arxiv.org/abs/2603.18130 · **PDF:** https://arxiv.org/pdf/2603.18130v1
+
+## 3764. Toward Scalable Automated Repository-Level Datasets for Software Vulnerability Detection
+
+- **ID:** `2603.17974` · **primary:** `cs.SE` · **Submitted:** 2026-03-18
+- **Authors:** Amine Lbath
+- **abs:** https://arxiv.org/abs/2603.17974 · **PDF:** https://arxiv.org/pdf/2603.17974v1
+
+## 3765. TDAD: Test-Driven Agentic Development - Reducing Code Regressions in AI Coding Agents via Graph-Based Impact Analysis
+
+- **ID:** `2603.17973` · **primary:** `cs.SE` · **Submitted:** 2026-03-18
+- **Authors:** Pepe Alonso, Sergio Yovine, Victor A. Braberman
+- **abs:** https://arxiv.org/abs/2603.17973 · **PDF:** https://arxiv.org/pdf/2603.17973v2
+
+## 3766. Beyond Muon: MUD (MomentUm Decorrelation) for Faster Transformer Training
+
+- **ID:** `2603.17970` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Ben S. Southworth, Stephen Thomas
+- **abs:** https://arxiv.org/abs/2603.17970 · **PDF:** https://arxiv.org/pdf/2603.17970v1
+
+## 3767. Specification-Aware Distribution Shaping for Robotics Foundation Models
+
+- **ID:** `2603.17969` · **primary:** `cs.RO` · **Submitted:** 2026-03-18
+- **Authors:** Sadık Bera Yüksel, Derya Aksaray
+- **abs:** https://arxiv.org/abs/2603.17969 · **PDF:** https://arxiv.org/pdf/2603.17969v1
+
+## 3768. ConGA: Guidelines for Contextual Gender Annotation. A Framework for Annotating Gender in Machine Translation
+
+- **ID:** `2603.17962` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Argentina Anna Rescigno, Eva Vanmassenhove, Johanna Monti
+- **abs:** https://arxiv.org/abs/2603.17962 · **PDF:** https://arxiv.org/pdf/2603.17962v1
+
+## 3769. Gender Disambiguation in Machine Translation: Diagnostic Evaluation in Decoder-Only Architectures
+
+- **ID:** `2603.17952` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Chiara Manna, Hosein Mohebbi, Afra Alishahi, et al.
+- **abs:** https://arxiv.org/abs/2603.17952 · **PDF:** https://arxiv.org/pdf/2603.17952v1
+
+## 3770. VideoAtlas: Navigating Long-Form Video in Logarithmic Compute
+
+- **ID:** `2603.17948` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Mohamed Eltahir, Ali Habibullah, Yazan Alshoibi, et al.
+- **abs:** https://arxiv.org/abs/2603.17948 · **PDF:** https://arxiv.org/pdf/2603.17948v1
+
+## 3771. Unified Policy Value Decomposition for Rapid Adaptation
+
+- **ID:** `2603.17947` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Cristiano Capone, Luca Falorsi, Andrea Ciardiello, et al.
+- **abs:** https://arxiv.org/abs/2603.17947 · **PDF:** https://arxiv.org/pdf/2603.17947v1
+
+## 3772. CARE: Covariance-Aware and Rank-Enhanced Decomposition for Enabling Multi-Head Latent Attention
+
+- **ID:** `2603.17946` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Zhongzhu Zhou, Fengxiang Bie, Ziyan Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.17946 · **PDF:** https://arxiv.org/pdf/2603.17946v1
+
+## 3773. ShapleyLaw: A Game-Theoretic Approach to Multilingual Scaling Laws
+
+- **ID:** `2603.17945` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Xuyang Cao, Qianying Liu, Chuan Xiao, et al.
+- **abs:** https://arxiv.org/abs/2603.17945 · **PDF:** https://arxiv.org/pdf/2603.17945v1
+
+## 3774. Efficient Training-Free Multi-Token Prediction via Embedding-Space Probing
+
+- **ID:** `2603.17942` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Raghavv Goel, Mukul Gagrani, Mingu Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.17942 · **PDF:** https://arxiv.org/pdf/2603.17942v1
+
+## 3775. Evaluating FrameNet-Based Semantic Modeling for Gender-Based Violence Detection in Clinical Records
+
+- **ID:** `2603.18124` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Lívia Dutra, Arthur Lorenzi, Frederico Belcavello, et al.
+- **abs:** https://arxiv.org/abs/2603.18124 · **PDF:** https://arxiv.org/pdf/2603.18124v1
+
+## 3776. Multi-Armed Sequential Hypothesis Testing by Betting
+
+- **ID:** `2603.17925` · **primary:** `stat.ME` · **Submitted:** 2026-03-18
+- **Authors:** Ricardo J. Sandoval, Ian Waudby-Smith, Michael I. Jordan
+- **abs:** https://arxiv.org/abs/2603.17925 · **PDF:** https://arxiv.org/pdf/2603.17925v1
+
+## 3777. Only relative ranks matter in weight-clustered large language models
+
+- **ID:** `2603.17917` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Borja Aizpurua, Sukhbinder Singh, Román Orús
+- **abs:** https://arxiv.org/abs/2603.17917 · **PDF:** https://arxiv.org/pdf/2603.17917v1
+
+## 3778. IndicSafe: A Benchmark for Evaluating Multilingual LLM Safety in South Asia
+
+- **ID:** `2603.17915` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Priyaranjan Pattnayak, Sanchari Chowdhuri
+- **abs:** https://arxiv.org/abs/2603.17915 · **PDF:** https://arxiv.org/pdf/2603.17915v1
+
+## 3779. Pretrained Multilingual Transformers Reveal Quantitative Distance Between Human Languages
+
+- **ID:** `2603.17912` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Yue Zhao, Jiatao Gu, Paloma Jeretič, et al.
+- **abs:** https://arxiv.org/abs/2603.17912 · **PDF:** https://arxiv.org/pdf/2603.17912v1
+
+## 3780. Understanding Task Aggregation for Generalizable Ultrasound Foundation Models
+
+- **ID:** `2603.18123` · **primary:** `eess.IV` · **Submitted:** 2026-03-18
+- **Authors:** Fangyijie Wang, Tanya Akumu, Vien Ngoc Dang, et al.
+- **abs:** https://arxiv.org/abs/2603.18123 · **PDF:** https://arxiv.org/pdf/2603.18123v2
+
+## 3781. Don't Vibe Code, Do Skele-Code: Interactive No-Code Notebooks for Subject Matter Experts to Build Lower-Cost Agentic Workflows
+
+- **ID:** `2603.18122` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Sriram Gopalakrishnan
+- **abs:** https://arxiv.org/abs/2603.18122 · **PDF:** https://arxiv.org/pdf/2603.18122v1
+
+## 3782. Differential Privacy in Generative AI Agents: Analysis and Optimal Tradeoffs
+
+- **ID:** `2603.17902` · **primary:** `cs.CR` · **Submitted:** 2026-03-18
+- **Authors:** Ya-Ting Yang, Quanyan Zhu
+- **abs:** https://arxiv.org/abs/2603.17902 · **PDF:** https://arxiv.org/pdf/2603.17902v1
+
+## 3783. A Noise Sensitivity Exponent Controls Large Statistical-to-Computational Gaps in Single- and Multi-Index Models
+
+- **ID:** `2603.17896` · **primary:** `stat.ML` · **Submitted:** 2026-03-18
+- **Authors:** Leonardo Defilippis, Florent Krzakala, Bruno Loureiro, et al.
+- **abs:** https://arxiv.org/abs/2603.17896 · **PDF:** https://arxiv.org/pdf/2603.17896v1
+
+## 3784. scicode-lint: Detecting Methodology Bugs in Scientific Python Code with LLM-Generated Patterns
+
+- **ID:** `2603.17893` · **primary:** `cs.SE` · **Submitted:** 2026-03-18
+- **Authors:** Sergey V. Samsonau
+- **abs:** https://arxiv.org/abs/2603.17893 · **PDF:** https://arxiv.org/pdf/2603.17893v1
+
+## 3785. RAMP: Reinforcement Adaptive Mixed Precision Quantization for Efficient On Device LLM Inference
+
+- **ID:** `2603.17891` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Arpit Singh Gautam, Saurabh Jha
+- **abs:** https://arxiv.org/abs/2603.17891 · **PDF:** https://arxiv.org/pdf/2603.17891v1
+
+## 3786. MAED: Mathematical Activation Error Detection for Mitigating Physical Fault Attacks in DNN Inference
+
+- **ID:** `2603.18120` · **primary:** `cs.CR` · **Submitted:** 2026-03-18
+- **Authors:** Kasra Ahmadi, Saeed Aghapour, Mehran Mozaffari Kermani, et al.
+- **abs:** https://arxiv.org/abs/2603.18120 · **PDF:** https://arxiv.org/pdf/2603.18120v1
+
+## 3787. AI-Assisted Goal Setting Improves Goal Progress Through Social Accountability
+
+- **ID:** `2603.17887` · **primary:** `cs.HC` · **Submitted:** 2026-03-18
+- **Authors:** Michel Schimpf, Julian Voigt, Thomas Bohné
+- **abs:** https://arxiv.org/abs/2603.17887 · **PDF:** https://arxiv.org/pdf/2603.17887v1
+
+## 3788. DebugLM: Learning Traceable Training Data Provenance for LLMs
+
+- **ID:** `2603.17884` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Wenjie Jacky Mo, Qin Liu, Xiaofei Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.17884 · **PDF:** https://arxiv.org/pdf/2603.17884v1
+
+## 3789. Toward Evaluation Frameworks for Multi-Agent Scientific AI Systems
+
+- **ID:** `2603.26718` · **primary:** `cs.CY` · **Submitted:** 2026-03-18
+- **Authors:** Marcin Abram
+- **abs:** https://arxiv.org/abs/2603.26718 · **PDF:** https://arxiv.org/pdf/2603.26718v1
+
+## 3790. Differential Attention-Augmented BiomedCLIP with Asymmetric Focal Optimization for Imbalanced Multi-Label Video Capsule Endoscopy Classification
+
+- **ID:** `2603.17879` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Podakanti Satyajith Chary, Nagarajan Ganapathy
+- **abs:** https://arxiv.org/abs/2603.17879 · **PDF:** https://arxiv.org/pdf/2603.17879v1
+
+## 3791. Operator-Theoretic Foundations and Policy Gradient Methods for General MDPs with Unbounded Costs
+
+- **ID:** `2603.17875` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Abhishek Gupta, Aditya Mahajan
+- **abs:** https://arxiv.org/abs/2603.17875 · **PDF:** https://arxiv.org/pdf/2603.17875v3
+
+## 3792. Mitigating LLM Hallucinations through Domain-Grounded Tiered Retrieval
+
+- **ID:** `2603.17872` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Md. Asraful Haque, Aasar Mehdi, Maaz Mahboob, et al.
+- **abs:** https://arxiv.org/abs/2603.17872 · **PDF:** https://arxiv.org/pdf/2603.17872v2
+
+## 3793. RHYME-XT: A Neural Operator for Spatiotemporal Control Systems
+
+- **ID:** `2603.17867` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Marijn Ruiter, Miguel Aguiar, Jake Rap, et al.
+- **abs:** https://arxiv.org/abs/2603.17867 · **PDF:** https://arxiv.org/pdf/2603.17867v1
+
+## 3794. Procedural Generation of Algorithm Discovery Tasks in Machine Learning
+
+- **ID:** `2603.17863` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Alexander D. Goldie, Zilin Wang, Adrian Hayler, et al.
+- **abs:** https://arxiv.org/abs/2603.17863 · **PDF:** https://arxiv.org/pdf/2603.17863v1
+
+## 3795. Physics-Aware Machine Learning for Seismic and Volcanic Signal Interpretation
+
+- **ID:** `2603.17855` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** William Thorossian
+- **abs:** https://arxiv.org/abs/2603.17855 · **PDF:** https://arxiv.org/pdf/2603.17855v1
+
+## 3796. How do LLMs Compute Verbal Confidence
+
+- **ID:** `2603.17839` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Dharshan Kumaran, Arthur Conmy, Federico Barbero, et al.
+- **abs:** https://arxiv.org/abs/2603.17839 · **PDF:** https://arxiv.org/pdf/2603.17839v2
+
+## 3797. Event-Centric Human Value Understanding in News-Domain Texts: An Actor-Conditioned, Multi-Granularity Benchmark
+
+- **ID:** `2603.17838` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Yao Wang, Xin Liu, Zhuochen Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.17838 · **PDF:** https://arxiv.org/pdf/2603.17838v1
+
+## 3798. The Silent Thought: Modeling Internal Cognition in Full-Duplex Spoken Dialogue Models via Latent Reasoning
+
+- **ID:** `2603.17837` · **primary:** `eess.AS` · **Submitted:** 2026-03-18
+- **Authors:** Donghang Wu, Tianyu Zhang, Yuxin Li, et al.
+- **abs:** https://arxiv.org/abs/2603.17837 · **PDF:** https://arxiv.org/pdf/2603.17837v1
+
+## 3799. Verification and Validation of Physics-Informed Surrogate Component Models for Dynamic Power-System Simulation
+
+- **ID:** `2603.17836` · **primary:** `eess.SY` · **Submitted:** 2026-03-18
+- **Authors:** Petros Ellinas, Indrajit Chaudhuri, Johanna Vorwerk, et al.
+- **abs:** https://arxiv.org/abs/2603.17836 · **PDF:** https://arxiv.org/pdf/2603.17836v1
+
+## 3800. Insight-V++: Towards Advanced Long-Chain Visual Reasoning with Multimodal Large Language Models
+
+- **ID:** `2603.18118` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Yuhao Dong, Zuyan Liu, Shulin Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.18118 · **PDF:** https://arxiv.org/pdf/2603.18118v1
+
+## 3801. Generative Control as Optimization: Time Unconditional Flow Matching for Adaptive and Robust Robotic Control
+
+- **ID:** `2603.17834` · **primary:** `cs.RO` · **Submitted:** 2026-03-18
+- **Authors:** Zunzhe Zhang, Runhan Huang, Yicheng Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.17834 · **PDF:** https://arxiv.org/pdf/2603.17834v1
+
+## 3802. Text-to-Stage: Spatial Layouts from Long-form Narratives
+
+- **ID:** `2603.17832` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Jefferson Hernandez, Swarnadeep Saha, Chenxi Whitehouse, et al.
+- **abs:** https://arxiv.org/abs/2603.17832 · **PDF:** https://arxiv.org/pdf/2603.17832v1
+
+## 3803. RPMS: Enhancing LLM-Based Embodied Planning through Rule-Augmented Memory Synergy
+
+- **ID:** `2603.17831` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Zhenhang Yuan, Shenghai Yuan, Lihua Xie
+- **abs:** https://arxiv.org/abs/2603.17831 · **PDF:** https://arxiv.org/pdf/2603.17831v1
+
+## 3804. CodeScout: An Effective Recipe for Reinforcement Learning of Code Search Agents
+
+- **ID:** `2603.17829` · **primary:** `cs.SE` · **Submitted:** 2026-03-18
+- **Authors:** Lintang Sutawika, Aditya Bharat Soni, Bharath Sriraam R R, et al.
+- **abs:** https://arxiv.org/abs/2603.17829 · **PDF:** https://arxiv.org/pdf/2603.17829v1
+
+## 3805. FailureMem: A Failure-Aware Multimodal Framework for Autonomous Software Repair
+
+- **ID:** `2603.17826` · **primary:** `cs.SE` · **Submitted:** 2026-03-18
+- **Authors:** Ruize Ma, Yilei Jiang, Shilin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.17826 · **PDF:** https://arxiv.org/pdf/2603.17826v1
+
+## 3806. Symmetry-Reduced Physics-Informed Learning of Tensegrity Dynamics
+
+- **ID:** `2603.17824` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Jing Qin, Muhao Chen
+- **abs:** https://arxiv.org/abs/2603.17824 · **PDF:** https://arxiv.org/pdf/2603.17824v1
+
+## 3807. Intellectual Stewardship: Re-adapting Human Minds for Creative Knowledge Work in the Age of AI
+
+- **ID:** `2603.18117` · **primary:** `cs.CY` · **Submitted:** 2026-03-18
+- **Authors:** Jianwei Zhang
+- **abs:** https://arxiv.org/abs/2603.18117 · **PDF:** https://arxiv.org/pdf/2603.18117v1
+
+## 3808. Discovering Decoupled Functional Modules in Large Language Models
+
+- **ID:** `2603.17823` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Yanke Yu, Jin Li, Ying Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.17823 · **PDF:** https://arxiv.org/pdf/2603.17823v1
+
+## 3809. Multi-Source Evidence Fusion for Audio Question Answering
+
+- **ID:** `2603.17822` · **primary:** `eess.AS` · **Submitted:** 2026-03-18
+- **Authors:** Aivo Olev, Tanel Alumäe
+- **abs:** https://arxiv.org/abs/2603.17822 · **PDF:** https://arxiv.org/pdf/2603.17822v1
+
+## 3810. Federated Distributional Reinforcement Learning with Distributional Critic Regularization
+
+- **ID:** `2603.17820` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** David Millard, Cecilia Alm, Rashid Ali, et al.
+- **abs:** https://arxiv.org/abs/2603.17820 · **PDF:** https://arxiv.org/pdf/2603.17820v1
+
+## 3811. Process Supervision for Chain-of-Thought Reasoning via Monte Carlo Net Information Gain
+
+- **ID:** `2603.17815` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Corentin Royer, Debarun Bhattacharjya, Gaetano Rossiello, et al.
+- **abs:** https://arxiv.org/abs/2603.17815 · **PDF:** https://arxiv.org/pdf/2603.17815v1
+
+## 3812. ChopGrad: Pixel-Wise Losses for Latent Video Diffusion via Truncated Backpropagation
+
+- **ID:** `2603.17812` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Dmitriy Rivkin, Parker Ewen, Lili Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.17812 · **PDF:** https://arxiv.org/pdf/2603.17812v1
+
+## 3813. Dropout Robustness and Cognitive Profiling of Transformer Models via Stochastic Inference
+
+- **ID:** `2603.17811` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Antônio Junior Alves Caiado, Michael Hahsler
+- **abs:** https://arxiv.org/abs/2603.17811 · **PDF:** https://arxiv.org/pdf/2603.17811v1
+
+## 3814. Fine-Grained Post-Training Quantization for Large Vision Language Models with Quantization-Aware Integrated Gradients
+
+- **ID:** `2603.17809` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Ziwei Xiang, Fanhu Zeng, Hongjian Fang, et al.
+- **abs:** https://arxiv.org/abs/2603.17809 · **PDF:** https://arxiv.org/pdf/2603.17809v1
+
+## 3815. EVA: Aligning Video World Models with Executable Robot Actions via Inverse Dynamics Rewards
+
+- **ID:** `2603.17808` · **primary:** `cs.RO` · **Submitted:** 2026-03-18
+- **Authors:** Ruixiang Wang, Qingming Liu, Yueci Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.17808 · **PDF:** https://arxiv.org/pdf/2603.17808v2
+
+## 3816. LLM-Augmented Computational Phenotyping of Long Covid
+
+- **ID:** `2603.18115` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Jing Wang, Jie Shen, Amar Sra, et al.
+- **abs:** https://arxiv.org/abs/2603.18115 · **PDF:** https://arxiv.org/pdf/2603.18115v1
+
+## 3817. RangeAD: Fast On-Model Anomaly Detection
+
+- **ID:** `2603.17795` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Luca Hinkamp, Simon Klüttermann, Emmanuel Müller
+- **abs:** https://arxiv.org/abs/2603.17795 · **PDF:** https://arxiv.org/pdf/2603.17795v1
+
+## 3818. The Convergence Frontier: Integrating Machine Learning and High Performance Quantum Computing for Next-Generation Drug Discovery
+
+- **ID:** `2603.17790` · **primary:** `quant-ph` · **Submitted:** 2026-03-18
+- **Authors:** Narjes Ansari, César Feniou, Nicolaï Gouraud, et al.
+- **abs:** https://arxiv.org/abs/2603.17790 · **PDF:** https://arxiv.org/pdf/2603.17790v3
+
+## 3819. Governed Memory: A Production Architecture for Multi-Agent Workflows
+
+- **ID:** `2603.17787` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Hamed Taheri
+- **abs:** https://arxiv.org/abs/2603.17787 · **PDF:** https://arxiv.org/pdf/2603.17787v1
+
+## 3820. Transfer Learning for Contextual Joint Assortment-Pricing under Cross-Market Heterogeneity
+
+- **ID:** `2603.18114` · **primary:** `stat.ME` · **Submitted:** 2026-03-18
+- **Authors:** Elynn Chen, Xi Chen, Yi Zhang
+- **abs:** https://arxiv.org/abs/2603.18114 · **PDF:** https://arxiv.org/pdf/2603.18114v1
+
+## 3821. A Dual Certificate Approach to Sparsity in Infinite-Width Shallow Neural Networks
+
+- **ID:** `2603.17785` · **primary:** `math.OC` · **Submitted:** 2026-03-18
+- **Authors:** Leonardo Del Grande, Christoph Brune, Marcello Carioni
+- **abs:** https://arxiv.org/abs/2603.17785 · **PDF:** https://arxiv.org/pdf/2603.17785v1
+
+## 3822. ResNet-50 with Class Reweighting and Anatomy-Guided Temporal Decoding for Gastrointestinal Video Analysis
+
+- **ID:** `2603.17784` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Romil Imtiaz, Dimitris K. Iakovidis
+- **abs:** https://arxiv.org/abs/2603.17784 · **PDF:** https://arxiv.org/pdf/2603.17784v1
+
+## 3823. Facts as First Class Objects: Knowledge Objects for Persistent LLM Memory
+
+- **ID:** `2603.17781` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Oliver Zahn, Simran Chana
+- **abs:** https://arxiv.org/abs/2603.17781 · **PDF:** https://arxiv.org/pdf/2603.17781v1
+
+## 3824. CoVerRL: Breaking the Consensus Trap in Label-Free Reasoning via Generator-Verifier Co-Evolution
+
+- **ID:** `2603.17775` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Teng Pan, Yuchen Yan, Zixuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.17775 · **PDF:** https://arxiv.org/pdf/2603.17775v2
+
+## 3825. Attention Sinks Induce Gradient Sinks
+
+- **ID:** `2603.17771` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Yihong Chen, Quanming Yao
+- **abs:** https://arxiv.org/abs/2603.17771 · **PDF:** https://arxiv.org/pdf/2603.17771v1
+
+## 3826. Modeling Overlapped Speech with Shuffles
+
+- **ID:** `2603.17769` · **primary:** `cs.SD` · **Submitted:** 2026-03-18
+- **Authors:** Matthew Wiesner, Samuele Cornell, Alexander Polok, et al.
+- **abs:** https://arxiv.org/abs/2603.17769 · **PDF:** https://arxiv.org/pdf/2603.17769v1
+
+## 3827. Grounded Multimodal Retrieval-Augmented Drafting of Radiology Impressions Using Case-Based Similarity Search
+
+- **ID:** `2603.17765` · **primary:** `q-bio.QM` · **Submitted:** 2026-03-18
+- **Authors:** Himadri Samanta
+- **abs:** https://arxiv.org/abs/2603.17765 · **PDF:** https://arxiv.org/pdf/2603.17765v1
+
+## 3828. Harm or Humor: A Multimodal, Multilingual Benchmark for Overt and Covert Harmful Humor
+
+- **ID:** `2603.17759` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Ahmed Sharshar, Hosam Elgendy, Saad El Dine Ahmed, et al.
+- **abs:** https://arxiv.org/abs/2603.17759 · **PDF:** https://arxiv.org/pdf/2603.17759v2
+
+## 3829. Towards Infinitely Long Neural Simulations: Self-Refining Neural Surrogate Models for Dynamical Systems
+
+- **ID:** `2603.17750` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Qi Liu, Laure Zanna, Joan Bruna
+- **abs:** https://arxiv.org/abs/2603.17750 · **PDF:** https://arxiv.org/pdf/2603.17750v1
+
+## 3830. VC-Soup: Value-Consistency Guided Multi-Value Alignment for Large Language Models
+
+- **ID:** `2603.18113` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Hefei Xu, Le Wu, Yu Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.18113 · **PDF:** https://arxiv.org/pdf/2603.18113v1
+
+## 3831. Embedding World Knowledge into Tabular Models: Towards Best Practices for Embedding Pipeline Design
+
+- **ID:** `2603.17737` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Oksana Kolomenko, Ricardo Knauer, Erik Rodner
+- **abs:** https://arxiv.org/abs/2603.17737 · **PDF:** https://arxiv.org/pdf/2603.17737v1
+
+## 3832. Tula: Optimizing Time, Cost, and Generalization in Distributed Large-Batch Training
+
+- **ID:** `2603.18112` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Sahil Tyagi, Feiyi Wang
+- **abs:** https://arxiv.org/abs/2603.18112 · **PDF:** https://arxiv.org/pdf/2603.18112v1
+
+## 3833. BoundAD: Boundary-Aware Negative Generation for Time Series Anomaly Detection
+
+- **ID:** `2603.18111` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Xiancheng Wang, Lin Wang, Zhibo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.18111 · **PDF:** https://arxiv.org/pdf/2603.18111v1
+
+## 3834. SARE: Sample-wise Adaptive Reasoning for Training-free Fine-grained Visual Recognition
+
+- **ID:** `2603.17729` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Jingxiao Yang, DaLin He, Miao Pan, et al.
+- **abs:** https://arxiv.org/abs/2603.17729 · **PDF:** https://arxiv.org/pdf/2603.17729v2
+
+## 3835. Predicting Trajectories of Long COVID in Adult Women: The Critical Role of Causal Disentanglement
+
+- **ID:** `2603.17722` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Jing Wang, Jie Shen, Yiming Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.17722 · **PDF:** https://arxiv.org/pdf/2603.17722v1
+
+## 3836. Discovery of Bimodal Drift Rate Structure in FRB 20240114A: Evidence for Dual Emission Regions
+
+- **ID:** `2603.18109` · **primary:** `astro-ph.HE` · **Submitted:** 2026-03-18
+- **Authors:** Santosh Arron
+- **abs:** https://arxiv.org/abs/2603.18109 · **PDF:** https://arxiv.org/pdf/2603.18109v1
+
+## 3837. Machine Learning for Network Attacks Classification and Statistical Evaluation of Adversarial Learning Methodologies for Synthetic Data Generation
+
+- **ID:** `2603.17717` · **primary:** `cs.CR` · **Submitted:** 2026-03-18
+- **Authors:** Iakovos-Christos Zarkadis, Christos Douligeris
+- **abs:** https://arxiv.org/abs/2603.17717 · **PDF:** https://arxiv.org/pdf/2603.17717v2
+
+## 3838. Eye image segmentation using visual and concept prompts with Segment Anything Model 3 (SAM3)
+
+- **ID:** `2603.17715` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Diederick C. Niehorster, Marcus Nyström
+- **abs:** https://arxiv.org/abs/2603.17715 · **PDF:** https://arxiv.org/pdf/2603.17715v1
+
+## 3839. From Virtual Environments to Real-World Trials: Emerging Trends in Autonomous Driving
+
+- **ID:** `2603.17714` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** A. Humnabadkar, A. Sikdar, B. Cave, et al.
+- **abs:** https://arxiv.org/abs/2603.17714 · **PDF:** https://arxiv.org/pdf/2603.17714v1
+
+## 3840. FEMBA on the Edge: Physiologically-Aware Pre-Training, Quantization, and Deployment of a Bidirectional Mamba EEG Foundation Model on an Ultra-low Power Microcontroller
+
+- **ID:** `2603.26716` · **primary:** `eess.SP` · **Submitted:** 2026-03-18
+- **Authors:** Anna Tegon, Nicholas Lehmann, Yawei Li, et al.
+- **abs:** https://arxiv.org/abs/2603.26716 · **PDF:** https://arxiv.org/pdf/2603.26716v1
+
+## 3841. MALLES: A Multi-agent LLMs-based Economic Sandbox with Consumer Preference Alignment
+
+- **ID:** `2603.17694` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Yusen Wu, Yiran Liu, Xiaotie Deng
+- **abs:** https://arxiv.org/abs/2603.17694 · **PDF:** https://arxiv.org/pdf/2603.17694v1
+
+## 3842. Can Blindfolded LLMs Still Trade? An Anonymization-First Framework for Portfolio Optimization
+
+- **ID:** `2603.17692` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Joohyoung Jeon, Hongchul Lee
+- **abs:** https://arxiv.org/abs/2603.17692 · **PDF:** https://arxiv.org/pdf/2603.17692v1
+
+## 3843. Stochastic set-valued optimization and its application to robust learning
+
+- **ID:** `2603.17691` · **primary:** `math.OC` · **Submitted:** 2026-03-18
+- **Authors:** Tommaso Giovannelli, Jingfu Tan, Luis Nunes Vicente
+- **abs:** https://arxiv.org/abs/2603.17691 · **PDF:** https://arxiv.org/pdf/2603.17691v1
+
+## 3844. Objective Mispricing Detection for Shortlisting Undervalued Football Players via Market Dynamics and News Signals
+
+- **ID:** `2603.17687` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Chinenye Omejieke, Shuyao Chen, Xia Cui
+- **abs:** https://arxiv.org/abs/2603.17687 · **PDF:** https://arxiv.org/pdf/2603.17687v1
+
+## 3845. ARTEMIS: A Neuro Symbolic Framework for Economically Constrained Market Dynamics
+
+- **ID:** `2603.18107` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Rahul D Ray
+- **abs:** https://arxiv.org/abs/2603.18107 · **PDF:** https://arxiv.org/pdf/2603.18107v1
+
+## 3846. Flow Matching Policy with Entropy Regularization
+
+- **ID:** `2603.17685` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Ting Gao, Stavros Orfanoudakis, Nan Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.17685 · **PDF:** https://arxiv.org/pdf/2603.17685v2
+
+## 3847. Sensi: Learn One Thing at a Time -- Curriculum-Based Test-Time Learning for LLM Game Agents
+
+- **ID:** `2603.17683` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Mohsen Arjmandi
+- **abs:** https://arxiv.org/abs/2603.17683 · **PDF:** https://arxiv.org/pdf/2603.17683v1
+
+## 3848. Murmurations, Mestre--Nagao sums, and Convolutional Neural Networks for elliptic curves
+
+- **ID:** `2603.17681` · **primary:** `math.NT` · **Submitted:** 2026-03-18
+- **Authors:** Joanna Bieri, Edgar Costa, Alyson Deines, et al.
+- **abs:** https://arxiv.org/abs/2603.17681 · **PDF:** https://arxiv.org/pdf/2603.17681v1
+
+## 3849. WeatherReasonSeg: A Benchmark for Weather-Aware Reasoning Segmentation in Visual Language Models
+
+- **ID:** `2603.17680` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Wanjun Du, Zifeng Yuan, Tingting Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.17680 · **PDF:** https://arxiv.org/pdf/2603.17680v1
+
+## 3850. Adaptive Guidance for Retrieval-Augmented Masked Diffusion Models
+
+- **ID:** `2603.17677` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Jaemin Kim, Jong Chul Ye
+- **abs:** https://arxiv.org/abs/2603.17677 · **PDF:** https://arxiv.org/pdf/2603.17677v1
+
+## 3851. Inhibitory normalization of error signals improves learning in neural circuits
+
+- **ID:** `2603.17676` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-18
+- **Authors:** Roy Henha Eyono, Daniel Levenstein, Arna Ghosh, et al.
+- **abs:** https://arxiv.org/abs/2603.17676 · **PDF:** https://arxiv.org/pdf/2603.17676v1
+
+## 3852. Post-Training Local LLM Agents for Linux Privilege Escalation with Verifiable Rewards
+
+- **ID:** `2603.17673` · **primary:** `cs.CR` · **Submitted:** 2026-03-18
+- **Authors:** Philipp Normann, Andreas Happe, Jürgen Cito, et al.
+- **abs:** https://arxiv.org/abs/2603.17673 · **PDF:** https://arxiv.org/pdf/2603.17673v1
+
+## 3853. Adaptive Domain Models: Bayesian Evolution, Warm Rotation, and Principled Training for Geometric and Neuromorphic AI
+
+- **ID:** `2603.18104` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Houston Haynes
+- **abs:** https://arxiv.org/abs/2603.18104 · **PDF:** https://arxiv.org/pdf/2603.18104v1
+
+## 3854. FINER: MLLMs Hallucinate under Fine-grained Negative Queries
+
+- **ID:** `2603.17662` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Rui Xiao, Sanghwan Kim, Yongqin Xian, et al.
+- **abs:** https://arxiv.org/abs/2603.17662 · **PDF:** https://arxiv.org/pdf/2603.17662v1
+
+## 3855. Interpretable Cross-Domain Few-Shot Learning with Rectified Target-Domain Local Alignment
+
+- **ID:** `2603.17655` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Yaze Zhao, Yixiong Zou, Yuhua Li, et al.
+- **abs:** https://arxiv.org/abs/2603.17655 · **PDF:** https://arxiv.org/pdf/2603.17655v2
+
+## 3856. STEP: Detecting Audio Backdoor Attacks via Stability-based Trigger Exposure Profiling
+
+- **ID:** `2603.18103` · **primary:** `cs.CR` · **Submitted:** 2026-03-18
+- **Authors:** Kun Wang, Meng Chen, Junhao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.18103 · **PDF:** https://arxiv.org/pdf/2603.18103v1
+
+## 3857. Anchoring and Rescaling Attention for Semantically Coherent Inbetweening
+
+- **ID:** `2603.17651` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Tae Eun Choi, Sumin Shim, Junhyeok Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.17651 · **PDF:** https://arxiv.org/pdf/2603.17651v1
+
+## 3858. Training-Only Heterogeneous Image-Patch-Text Graph Supervision for Advancing Few-Shot Learning Adapters
+
+- **ID:** `2603.18101` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Mohammed Rahman Sherif Khan Mohammad, Ardhendu Behera, Sandip Pradhan, et al.
+- **abs:** https://arxiv.org/abs/2603.18101 · **PDF:** https://arxiv.org/pdf/2603.18101v1
+
+## 3859. Automated Grammar-based Algebraic Multigrid Design With Evolutionary Algorithms
+
+- **ID:** `2603.17641` · **primary:** `cs.CE` · **Submitted:** 2026-03-18
+- **Authors:** Dinesh Parthasarathy, Wayne Mitchell, Arjun Gambhir, et al.
+- **abs:** https://arxiv.org/abs/2603.17641 · **PDF:** https://arxiv.org/pdf/2603.17641v1
+
+## 3860. VeriGrey: Greybox Agent Validation
+
+- **ID:** `2603.17639` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Yuntong Zhang, Sungmin Kang, Ruijie Meng, et al.
+- **abs:** https://arxiv.org/abs/2603.17639 · **PDF:** https://arxiv.org/pdf/2603.17639v1
+
+## 3861. DSS-GAN: Directional State Space GAN with Mamba backbone for Class-Conditional Image Synthesis
+
+- **ID:** `2603.17637` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Aleksander Ogonowski, Konrad Klimaszewski, Przemysław Rokita
+- **abs:** https://arxiv.org/abs/2603.17637 · **PDF:** https://arxiv.org/pdf/2603.17637v1
+
+## 3862. Atomic Trajectory Modeling with State Space Models for Biomolecular Dynamics
+
+- **ID:** `2603.17633` · **primary:** `q-bio.BM` · **Submitted:** 2026-03-18
+- **Authors:** Liang Shi, Jiarui Lu, Junqi Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.17633 · **PDF:** https://arxiv.org/pdf/2603.17633v1
+
+## 3863. Benchmarking Reinforcement Learning via Stochastic Converse Optimality: Generating Systems with Known Optimal Policies
+
+- **ID:** `2603.17631` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Sinan Ibrahim, Grégoire Ouerdane, Hadi Salloum, et al.
+- **abs:** https://arxiv.org/abs/2603.17631 · **PDF:** https://arxiv.org/pdf/2603.17631v1
+
+## 3864. rSDNet: Unified Robust Neural Learning against Label Noise and Adversarial Attacks
+
+- **ID:** `2603.17628` · **primary:** `stat.ML` · **Submitted:** 2026-03-18
+- **Authors:** Suryasis Jana, Abhik Ghosh
+- **abs:** https://arxiv.org/abs/2603.17628 · **PDF:** https://arxiv.org/pdf/2603.17628v1
+
+## 3865. Do Language Models Encode Semantic Relations? Probing and Sparse Feature Analysis
+
+- **ID:** `2603.17624` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Andor Diera, Ansgar Scherp
+- **abs:** https://arxiv.org/abs/2603.17624 · **PDF:** https://arxiv.org/pdf/2603.17624v2
+
+## 3866. ARES: Scalable and Practical Gradient Inversion Attack in Federated Learning through Activation Recovery
+
+- **ID:** `2603.17623` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Zirui Gong, Leo Yu Zhang, Yanjun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.17623 · **PDF:** https://arxiv.org/pdf/2603.17623v1
+
+## 3867. Complementary Reinforcement Learning
+
+- **ID:** `2603.17621` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Dilxat Muhtar, Jiashun Liu, Wei Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.17621 · **PDF:** https://arxiv.org/pdf/2603.17621v1
+
+## 3868. Full waveform inversion method based on diffusion model
+
+- **ID:** `2603.22307` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Caiyun Liu, Siyang Pei, Qingfeng Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.22307 · **PDF:** https://arxiv.org/pdf/2603.22307v1
+
+## 3869. Temporal Narrative Monitoring in Dynamic Information Environments
+
+- **ID:** `2603.17617` · **primary:** `cs.SI` · **Submitted:** 2026-03-18
+- **Authors:** David Farr, Stephen Prochaska, Jack Moody, et al.
+- **abs:** https://arxiv.org/abs/2603.17617 · **PDF:** https://arxiv.org/pdf/2603.17617v1
+
+## 3870. VeriAgent: A Tool-Integrated Multi-Agent System with Evolving Memory for PPA-Aware RTL Code Generation
+
+- **ID:** `2603.17613` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Yaoxiang Wang, Qi Shi, ShangZhan Li, et al.
+- **abs:** https://arxiv.org/abs/2603.17613 · **PDF:** https://arxiv.org/pdf/2603.17613v1
+
+## 3871. AdaMuS: Adaptive Multi-view Sparsity Learning for Dimensionally Unbalanced Data
+
+- **ID:** `2603.17610` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Cai Xu, Changhao Sun, Ziyu Guan, et al.
+- **abs:** https://arxiv.org/abs/2603.17610 · **PDF:** https://arxiv.org/pdf/2603.17610v2
+
+## 3872. End-to-end data-driven prediction of urban airflow and pollutant dispersion
+
+- **ID:** `2603.17606` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Nishant Kumar, Franck Kerhervé, Lionel Agostini, et al.
+- **abs:** https://arxiv.org/abs/2603.17606 · **PDF:** https://arxiv.org/pdf/2603.17606v1
+
+## 3873. Modeling Changing Scientific Concepts with Complex Networks: A Case Study on the Chemical Revolution
+
+- **ID:** `2603.17594` · **primary:** `physics.soc-ph` · **Submitted:** 2026-03-18
+- **Authors:** Sofía Aguilar-Valdez, Stefania Degaetano-Ortlieb
+- **abs:** https://arxiv.org/abs/2603.17594 · **PDF:** https://arxiv.org/pdf/2603.17594v1
+
+## 3874. A Contextual Help Browser Extension to Assist Digital Illiterate Internet Users
+
+- **ID:** `2603.17592` · **primary:** `cs.IR` · **Submitted:** 2026-03-18
+- **Authors:** Christos Koutsiaris
+- **abs:** https://arxiv.org/abs/2603.17592 · **PDF:** https://arxiv.org/pdf/2603.17592v1
+
+## 3875. From Isolated Scoring to Collaborative Ranking: A Comparison-Native Framework for LLM-Based Paper Evaluation
+
+- **ID:** `2603.17588` · **primary:** `cs.IR` · **Submitted:** 2026-03-18
+- **Authors:** Pujun Zheng, Jiacheng Yao, Jinquan Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.17588 · **PDF:** https://arxiv.org/pdf/2603.17588v1
+
+## 3876. Edit-As-Act: Goal-Regressive Planning for Open-Vocabulary 3D Indoor Scene Editing
+
+- **ID:** `2603.17583` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Seongrae Noh, SeungWon Seo, Gyeong-Moon Park, et al.
+- **abs:** https://arxiv.org/abs/2603.17583 · **PDF:** https://arxiv.org/pdf/2603.17583v1
+
+## 3877. One-Step Sampler for Boltzmann Distributions via Drifting
+
+- **ID:** `2603.17579` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Wenhan Cao, Keyu Yan, Lin Zhao
+- **abs:** https://arxiv.org/abs/2603.17579 · **PDF:** https://arxiv.org/pdf/2603.17579v1
+
+## 3878. Identifying Latent Actions and Dynamics from Offline Data via Demonstrator Diversity
+
+- **ID:** `2603.17577` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Felix Schur
+- **abs:** https://arxiv.org/abs/2603.17577 · **PDF:** https://arxiv.org/pdf/2603.17577v1
+
+## 3879. Unsupervised Symbolic Anomaly Detection
+
+- **ID:** `2603.17575` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Md Maruf Hossain, Tim Katzke, Simon Klüttermann, et al.
+- **abs:** https://arxiv.org/abs/2603.17575 · **PDF:** https://arxiv.org/pdf/2603.17575v1
+
+## 3880. HeiSD: Hybrid Speculative Decoding for Embodied Vision-Language-Action Models with Kinematic Awareness
+
+- **ID:** `2603.17573` · **primary:** `cs.RO` · **Submitted:** 2026-03-18
+- **Authors:** Zihao Zheng, Zhihao Mao, Sicheng Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.17573 · **PDF:** https://arxiv.org/pdf/2603.17573v1
+
+## 3881. A Trace-Based Assurance Framework for Agentic AI Orchestration: Contracts, Testing, and Governance
+
+- **ID:** `2603.18096` · **primary:** `cs.MA` · **Submitted:** 2026-03-18
+- **Authors:** Ciprian Paduraru, Petru-Liviu Bouruc, Alin Stefanescu
+- **abs:** https://arxiv.org/abs/2603.18096 · **PDF:** https://arxiv.org/pdf/2603.18096v1
+
+## 3882. Memory Bear AI Memory Science Engine for Multimodal Affective Intelligence: A Technical Report
+
+- **ID:** `2603.22306` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Deliang Wen, Ke Sun, Yu Wang
+- **abs:** https://arxiv.org/abs/2603.22306 · **PDF:** https://arxiv.org/pdf/2603.22306v1
+
+## 3883. FoMo X: Modular Explainability Signals for Outlier Detection Foundation Models
+
+- **ID:** `2603.17570` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Simon Klüttermann, Tim Katzke, Phuong Huong Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.17570 · **PDF:** https://arxiv.org/pdf/2603.17570v1
+
+## 3884. Q-Drift: Quantization-Aware Drift Correction for Diffusion Model Sampling
+
+- **ID:** `2603.18095` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Sooyoung Ryu, Mathieu Salzmann, Saqib Javed
+- **abs:** https://arxiv.org/abs/2603.18095 · **PDF:** https://arxiv.org/pdf/2603.18095v1
+
+## 3885. Gaussian Process Limit Reveals Structural Benefits of Graph Transformers
+
+- **ID:** `2603.17569` · **primary:** `stat.ML` · **Submitted:** 2026-03-18
+- **Authors:** Nil Ayday, Lingchu Yang, Debarghya Ghoshdastidar
+- **abs:** https://arxiv.org/abs/2603.17569 · **PDF:** https://arxiv.org/pdf/2603.17569v1
+
+## 3886. KA2L: A Knowledge-Aware Active Learning Framework for LLMs
+
+- **ID:** `2603.17566` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Haoxuan Yin, Bojian Liu, Chen Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.17566 · **PDF:** https://arxiv.org/pdf/2603.17566v1
+
+## 3887. In Trust We Survive: Emergent Trust Learning
+
+- **ID:** `2603.17564` · **primary:** `cs.MA` · **Submitted:** 2026-03-18
+- **Authors:** Qianpu Chen, Giulio Barbero, Mike Preuss, et al.
+- **abs:** https://arxiv.org/abs/2603.17564 · **PDF:** https://arxiv.org/pdf/2603.17564v1
+
+## 3888. Zipper-LoRA: Dynamic Parameter Decoupling for Speech-LLM based Multilingual Speech Recognition
+
+- **ID:** `2603.17558` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Yuxiang Mei, Delai Qiu, Shengping Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.17558 · **PDF:** https://arxiv.org/pdf/2603.17558v2
+
+## 3889. FrescoDiffusion: 4K Image-to-Video with Prior-Regularized Tiled Diffusion
+
+- **ID:** `2603.17555` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Hugo Caselles-Dupré, Mathis Koroglu, Guillaume Jeanneret, et al.
+- **abs:** https://arxiv.org/abs/2603.17555 · **PDF:** https://arxiv.org/pdf/2603.17555v1
+
+## 3890. Consistency of the $k$-Nearest Neighbor Regressor under Complex Survey Designs
+
+- **ID:** `2603.17551` · **primary:** `stat.ML` · **Submitted:** 2026-03-18
+- **Authors:** Caren Hasler
+- **abs:** https://arxiv.org/abs/2603.17551 · **PDF:** https://arxiv.org/pdf/2603.17551v1
+
+## 3891. Conditional Inverse Learning of Time-Varying Reproduction Numbers Inference
+
+- **ID:** `2603.17549` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Lanlan Yu, Quan-Hui Liu, Haoyue Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.17549 · **PDF:** https://arxiv.org/pdf/2603.17549v1
+
+## 3892. CLeAN: Continual Learning Adaptive Normalization in Dynamic Environments
+
+- **ID:** `2603.17548` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Isabella Marasco, Davide Evangelista, Elena Loli Piccolomini, et al.
+- **abs:** https://arxiv.org/abs/2603.17548 · **PDF:** https://arxiv.org/pdf/2603.17548v1
+
+## 3893. Per-Domain Generalizing Policies: On Learning Efficient and Robust Q-Value Functions (Extended Version with Technical Appendix)
+
+- **ID:** `2603.17544` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Nicola J. Müller, Moritz Oster, Isabel Valera, et al.
+- **abs:** https://arxiv.org/abs/2603.17544 · **PDF:** https://arxiv.org/pdf/2603.17544v1
+
+## 3894. AURORA Model of Formant-to-Tongue Inversion for Didactic and Clinical Applications
+
+- **ID:** `2603.17543` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Patrycja Strycharczuk, Sam Kirkham
+- **abs:** https://arxiv.org/abs/2603.17543 · **PDF:** https://arxiv.org/pdf/2603.17543v1
+
+## 3895. Deploying Semantic ID-based Generative Retrieval for Large-Scale Podcast Discovery at Spotify
+
+- **ID:** `2603.17540` · **primary:** `cs.IR` · **Submitted:** 2026-03-18
+- **Authors:** Edoardo D'Amico, Marco De Nadai, Praveen Chandar, et al.
+- **abs:** https://arxiv.org/abs/2603.17540 · **PDF:** https://arxiv.org/pdf/2603.17540v1
+
+## 3896. Learning Coordinate-based Convolutional Kernels for Continuous SE(3) Equivariant and Efficient Point Cloud Analysis
+
+- **ID:** `2603.17538` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Jaein Kim, Hee Bin Yoo, Dong-Sig Han, et al.
+- **abs:** https://arxiv.org/abs/2603.17538 · **PDF:** https://arxiv.org/pdf/2603.17538v1
+
+## 3897. PCA-Based Interpretable Knowledge Representation and Analysis of Geometric Design Parameters
+
+- **ID:** `2603.17535` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Alexander Köhler, Michael Breuß
+- **abs:** https://arxiv.org/abs/2603.17535 · **PDF:** https://arxiv.org/pdf/2603.17535v2
+
+## 3898. Informative Semi-Factuals for XAI: The Elaborated Explanations that People Prefer
+
+- **ID:** `2603.17534` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Saugat Aryal, Mark T. Keane
+- **abs:** https://arxiv.org/abs/2603.17534 · **PDF:** https://arxiv.org/pdf/2603.17534v1
+
+## 3899. A Unified Language Model for Large Scale Search, Recommendation, and Reasoning
+
+- **ID:** `2603.17533` · **primary:** `cs.IR` · **Submitted:** 2026-03-18
+- **Authors:** Marco De Nadai, Edoardo D'Amico, Max Lefarov, et al.
+- **abs:** https://arxiv.org/abs/2603.17533 · **PDF:** https://arxiv.org/pdf/2603.17533v1
+
+## 3900. Anisotropic Permeability Tensor Prediction from Porous Media Microstructure via Physics-Informed Progressive Transfer Learning with Hybrid CNN-Transformer
+
+- **ID:** `2603.17532` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Mohammad Nooraiepour
+- **abs:** https://arxiv.org/abs/2603.17532 · **PDF:** https://arxiv.org/pdf/2603.17532v1
+
+## 3901. Rel-Zero: Harnessing Patch-Pair Invariance for Robust Zero-Watermarking Against AI Editing
+
+- **ID:** `2603.17531` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Pengzhen Chen, Yanwei Liu, Xiaoyan Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.17531 · **PDF:** https://arxiv.org/pdf/2603.17531v1
+
+## 3902. AdapTS: Lightweight Teacher-Student Approach for Multi-Class and Continual Visual Anomaly Detection
+
+- **ID:** `2603.17530` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Manuel Barusco, Davide Dalle Pezze, Francesco Borsatti, et al.
+- **abs:** https://arxiv.org/abs/2603.17530 · **PDF:** https://arxiv.org/pdf/2603.17530v1
+
+## 3903. AirDDE: Multifactor Neural Delay Differential Equations for Air Quality Forecasting
+
+- **ID:** `2603.17529` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Binqing Wu, Zongjiang Shang, Shiyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.17529 · **PDF:** https://arxiv.org/pdf/2603.17529v1
+
+## 3904. Mirror Descent on Riemannian Manifolds
+
+- **ID:** `2603.17527` · **primary:** `stat.ML` · **Submitted:** 2026-03-18
+- **Authors:** Jiaxin Jiang, Lei Shi, Jiyuan Tan
+- **abs:** https://arxiv.org/abs/2603.17527 · **PDF:** https://arxiv.org/pdf/2603.17527v1
+
+## 3905. KineVLA: Towards Kinematics-Aware Vision-Language-Action Models with Bi-Level Action Decomposition
+
+- **ID:** `2603.17524` · **primary:** `cs.RO` · **Submitted:** 2026-03-18
+- **Authors:** Gaoge Han, Zhengqing Gao, Ziwen Li, et al.
+- **abs:** https://arxiv.org/abs/2603.17524 · **PDF:** https://arxiv.org/pdf/2603.17524v1
+
+## 3906. Translation Invariance of Neural Operators for the FitzHugh-Nagumo Model
+
+- **ID:** `2603.17523` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Luca Pellegrini
+- **abs:** https://arxiv.org/abs/2603.17523 · **PDF:** https://arxiv.org/pdf/2603.17523v1
+
+## 3907. Detecting the Machine: A Comprehensive Benchmark of AI-Generated Text Detectors Across Architectures, Domains, and Adversarial Conditions
+
+- **ID:** `2603.17522` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Madhav S. Baidya, S. S. Baidya, Chirag Chawla
+- **abs:** https://arxiv.org/abs/2603.17522 · **PDF:** https://arxiv.org/pdf/2603.17522v1
+
+## 3908. Language on Demand, Knowledge at Core: Composing LLMs with Encoder-Decoder Translation Models for Extensible Multilinguality
+
+- **ID:** `2603.17512` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Mengyu Bu, Yang Feng
+- **abs:** https://arxiv.org/abs/2603.17512 · **PDF:** https://arxiv.org/pdf/2603.17512v2
+
+## 3909. QuantFL: Sustainable Federated Learning for Edge IoT via Pre-Trained Model Quantisation
+
+- **ID:** `2603.17507` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Charuka Herath, Yogachandran Rahulamathavan, Varuna De Silva, et al.
+- **abs:** https://arxiv.org/abs/2603.17507 · **PDF:** https://arxiv.org/pdf/2603.17507v1
+
+## 3910. MOSS-TTS Technical Report
+
+- **ID:** `2603.18090` · **primary:** `cs.SD` · **Submitted:** 2026-03-18
+- **Authors:** Yitian Gong, Botian Jiang, Yiwei Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.18090 · **PDF:** https://arxiv.org/pdf/2603.18090v2
+
+## 3911. Inducing Epistemological Humility in Large Language Models: A Targeted SFT Approach to Reducing Hallucination
+
+- **ID:** `2603.17504` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Cem Uluoglakci, Tugba Taskaya Temizel
+- **abs:** https://arxiv.org/abs/2603.17504 · **PDF:** https://arxiv.org/pdf/2603.17504v1
+
+## 3912. CytoSyn: a Foundation Diffusion Model for Histopathology -- Tech Report
+
+- **ID:** `2603.18089` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Thomas Duboudin, Xavier Fontaine, Etienne Andrier, et al.
+- **abs:** https://arxiv.org/abs/2603.18089 · **PDF:** https://arxiv.org/pdf/2603.18089v1
+
+## 3913. Learning When to Attend: Conditional Memory Access for Long-Context LLMs
+
+- **ID:** `2603.17484` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Sakshi Choudhary, Aditya Chattopadhyay, Luca Zancato, et al.
+- **abs:** https://arxiv.org/abs/2603.17484 · **PDF:** https://arxiv.org/pdf/2603.17484v1
+
+## 3914. Enhancing Reinforcement Learning Fine-Tuning with an Online Refiner
+
+- **ID:** `2603.18088` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Hao Ma, Zhiqiang Pu, Yang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.18088 · **PDF:** https://arxiv.org/pdf/2603.18088v1
+
+## 3915. Auto-Unrolled Proximal Gradient Descent: An AutoML Approach to Interpretable Waveform Optimization
+
+- **ID:** `2603.17478` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Ahmet Kaplan
+- **abs:** https://arxiv.org/abs/2603.17478 · **PDF:** https://arxiv.org/pdf/2603.17478v1
+
+## 3916. UniSAFE: A Comprehensive Benchmark for Safety Evaluation of Unified Multimodal Models
+
+- **ID:** `2603.17476` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Segyu Lee, Boryeong Cho, Hojung Jung, et al.
+- **abs:** https://arxiv.org/abs/2603.17476 · **PDF:** https://arxiv.org/pdf/2603.17476v1
+
+## 3917. Humans and transformer LMs: Abstraction drives language learning
+
+- **ID:** `2603.17475` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Jasper Jian, Christopher D. Manning
+- **abs:** https://arxiv.org/abs/2603.17475 · **PDF:** https://arxiv.org/pdf/2603.17475v1
+
+## 3918. Revisiting Cross-Attention Mechanisms: Leveraging Beneficial Noise for Domain-Adaptive Learning
+
+- **ID:** `2603.17474` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Zelin Zang, Yehui Yang, Fei Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.17474 · **PDF:** https://arxiv.org/pdf/2603.17474v1
+
+## 3919. Multi-Trait Subspace Steering to Reveal the Dark Side of Human-AI Interaction
+
+- **ID:** `2603.18085` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Xin Wei Chia, Swee Liang Wong, Jonathan Pan
+- **abs:** https://arxiv.org/abs/2603.18085 · **PDF:** https://arxiv.org/pdf/2603.18085v1
+
+## 3920. VirPro: Visual-referred Probabilistic Prompt Learning for Weakly-Supervised Monocular 3D Detection
+
+- **ID:** `2603.17470` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Chupeng Liu, Jiyong Rao, Shangquan Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.17470 · **PDF:** https://arxiv.org/pdf/2603.17470v2
+
+## 3921. Efficient Soft Actor-Critic with LLM-Based Action-Level Guidance for Continuous Control
+
+- **ID:** `2603.17468` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Hao Ma, Zhiqiang Pu, Xiaolin Ai, et al.
+- **abs:** https://arxiv.org/abs/2603.17468 · **PDF:** https://arxiv.org/pdf/2603.17468v1
+
+## 3922. Uncovering Latent Phase Structures and Branching Logic in Locomotion Policies: A Case Study on HalfCheetah
+
+- **ID:** `2603.18084` · **primary:** `cs.RO` · **Submitted:** 2026-03-18
+- **Authors:** Daisuke Yasui, Toshitaka Matsuki, Hiroshi Sato
+- **abs:** https://arxiv.org/abs/2603.18084 · **PDF:** https://arxiv.org/pdf/2603.18084v1
+
+## 3923. Probabilistic Federated Learning on Uncertain and Heterogeneous Data with Model Personalization
+
+- **ID:** `2603.18083` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Ratun Rahman, Dinh C. Nguyen
+- **abs:** https://arxiv.org/abs/2603.18083 · **PDF:** https://arxiv.org/pdf/2603.18083v1
+
+## 3924. VLM2Rec: Resolving Modality Collapse in Vision-Language Model Embedders for Multimodal Sequential Recommendation
+
+- **ID:** `2603.17450` · **primary:** `cs.IR` · **Submitted:** 2026-03-18
+- **Authors:** Junyoung Kim, Woojoo Kim, Jaehyung Lim, et al.
+- **abs:** https://arxiv.org/abs/2603.17450 · **PDF:** https://arxiv.org/pdf/2603.17450v1
+
+## 3925. TRiMS: Real-Time Tracking of Minimal Sufficient Length for Efficient Reasoning via RL
+
+- **ID:** `2603.17449` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Tingcheng Bian, Jinchang Luo, Mingquan Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.17449 · **PDF:** https://arxiv.org/pdf/2603.17449v1
+
+## 3926. AgentComm-Bench: Stress-Testing Cooperative Embodied AI Under Latency, Packet Loss, and Bandwidth Collapse
+
+- **ID:** `2603.20285` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Aayam Bansal, Ishaan Gangwani
+- **abs:** https://arxiv.org/abs/2603.20285 · **PDF:** https://arxiv.org/pdf/2603.20285v1
+
+## 3927. When Only the Final Text Survives: Implicit Execution Tracing for Multi-Agent Attribution
+
+- **ID:** `2603.17445` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Yi Nian, Haosen Cao, Shenzhe Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.17445 · **PDF:** https://arxiv.org/pdf/2603.17445v4
+
+## 3928. Large Language Models as a Semantic Interface and Ethical Mediator in Neuro-Digital Ecosystems: Conceptual Foundations and a Regulatory Imperative
+
+- **ID:** `2603.17444` · **primary:** `cs.NE` · **Submitted:** 2026-03-18
+- **Authors:** Alexander V. Shenderuk-Zhidkov, Alexander E. Hramov
+- **abs:** https://arxiv.org/abs/2603.17444 · **PDF:** https://arxiv.org/pdf/2603.17444v1
+
+## 3929. AdaZoom-GUI: Adaptive Zoom-based GUI Grounding with Instruction Refinement
+
+- **ID:** `2603.17441` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Siqi Pei, Liang Tang, Tiaonan Duan, et al.
+- **abs:** https://arxiv.org/abs/2603.17441 · **PDF:** https://arxiv.org/pdf/2603.17441v1
+
+## 3930. Baguan-TS: A Sequence-Native In-Context Learning Model for Time Series Forecasting with Covariates
+
+- **ID:** `2603.17439` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Linxiao Yang, Xue Jiang, Gezheng Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.17439 · **PDF:** https://arxiv.org/pdf/2603.17439v1
+
+## 3931. TimeAPN: Adaptive Amplitude-Phase Non-Stationarity Normalization for Time Series Forecasting
+
+- **ID:** `2603.17436` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Yue Hu, Jialiang Tang, Siwei Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.17436 · **PDF:** https://arxiv.org/pdf/2603.17436v1
+
+## 3932. ZipServ: Fast and Memory-Efficient LLM Inference with Hardware-Aware Lossless Compression
+
+- **ID:** `2603.17435` · **primary:** `cs.DC` · **Submitted:** 2026-03-18
+- **Authors:** Ruibo Fan, Xiangrui Yu, Xinglin Pan, et al.
+- **abs:** https://arxiv.org/abs/2603.17435 · **PDF:** https://arxiv.org/pdf/2603.17435v1
+
+## 3933. The Phasor Transformer: Resolving Attention Bottlenecks on the Unit Circle
+
+- **ID:** `2603.17433` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Dibakar Sigdel
+- **abs:** https://arxiv.org/abs/2603.17433 · **PDF:** https://arxiv.org/pdf/2603.17433v1
+
+## 3934. Argument Reconstruction as Supervision for Critical Thinking in LLMs
+
+- **ID:** `2603.17432` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Hyun Ryu, Gyouk Chu, Gregor Betz, et al.
+- **abs:** https://arxiv.org/abs/2603.17432 · **PDF:** https://arxiv.org/pdf/2603.17432v1
+
+## 3935. SLEA-RL: Step-Level Experience Augmented Reinforcement Learning for Multi-Turn Agentic Training
+
+- **ID:** `2603.18079` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Prince Zizhuang Wang, Shuli Jiang
+- **abs:** https://arxiv.org/abs/2603.18079 · **PDF:** https://arxiv.org/pdf/2603.18079v1
+
+## 3936. Variational Phasor Circuits for Phase-Native Brain-Computer Interface Classification
+
+- **ID:** `2603.18078` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Dibakar Sigdel
+- **abs:** https://arxiv.org/abs/2603.18078 · **PDF:** https://arxiv.org/pdf/2603.18078v1
+
+## 3937. Proactive Knowledge Inquiry in Doctor-Patient Dialogue: Stateful Extraction, Belief Updating, and Path-Aware Action Planning
+
+- **ID:** `2603.17425` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Zhenhai Pan, Yan Liu, Jia You
+- **abs:** https://arxiv.org/abs/2603.17425 · **PDF:** https://arxiv.org/pdf/2603.17425v1
+
+## 3938. Data-driven model order reduction for structures with piecewise linear nonlinearity using dynamic mode decomposition
+
+- **ID:** `2603.17423` · **primary:** `math.DS` · **Submitted:** 2026-03-18
+- **Authors:** Akira Saito, Masato Tanaka
+- **abs:** https://arxiv.org/abs/2603.17423 · **PDF:** https://arxiv.org/pdf/2603.17423v1
+
+## 3939. From Digital Twins to World Models:Opportunities, Challenges, and Applications for Mobile Edge General Intelligence
+
+- **ID:** `2603.17420` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Jie Zheng, Dusit Niyato, Changyuan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.17420 · **PDF:** https://arxiv.org/pdf/2603.17420v1
+
+## 3940. Caging the Agents: A Zero Trust Security Architecture for Autonomous AI in Healthcare
+
+- **ID:** `2603.17419` · **primary:** `cs.CR` · **Submitted:** 2026-03-18
+- **Authors:** Saikat Maiti
+- **abs:** https://arxiv.org/abs/2603.17419 · **PDF:** https://arxiv.org/pdf/2603.17419v1
+
+## 3941. Structured SIR: Efficient and Expressive Importance-Weighted Inference for High-Dimensional Image Registration
+
+- **ID:** `2603.17415` · **primary:** `eess.IV` · **Submitted:** 2026-03-18
+- **Authors:** Ivor J. A. Simpson, Neill D. F. Campbell
+- **abs:** https://arxiv.org/abs/2603.17415 · **PDF:** https://arxiv.org/pdf/2603.17415v1
+
+## 3942. Mutually Causal Semantic Distillation Network for Zero-Shot Learning
+
+- **ID:** `2603.17412` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Shiming Chen, Shuhuang Chen, Guo-Sen Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.17412 · **PDF:** https://arxiv.org/pdf/2603.17412v1
+
+## 3943. Joint Degradation-Aware Arbitrary-Scale Super-Resolution for Variable-Rate Extreme Image Compression
+
+- **ID:** `2603.17408` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Xinning Chai, Zhengxue Cheng, Xin Li, et al.
+- **abs:** https://arxiv.org/abs/2603.17408 · **PDF:** https://arxiv.org/pdf/2603.17408v1
+
+## 3944. Generative Replica-Exchange: A Flow-based Framework for Accelerating Replica Exchange Simulations
+
+- **ID:** `2603.18076` · **primary:** `q-bio.BM` · **Submitted:** 2026-03-18
+- **Authors:** Shengjie Huang, Sijie Yang, Jianqiao Yi, et al.
+- **abs:** https://arxiv.org/abs/2603.18076 · **PDF:** https://arxiv.org/pdf/2603.18076v1
+
+## 3945. Causal Representation Learning on High-Dimensional Data: Benchmarks, Reproducibility, and Evaluation Metrics
+
+- **ID:** `2603.17405` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Alireza Sadeghi, Wael AbdAlmageed
+- **abs:** https://arxiv.org/abs/2603.17405 · **PDF:** https://arxiv.org/pdf/2603.17405v1
+
+## 3946. Large-Scale 3D Ground-Motion Synthesis with Physics-Inspired Latent Operator Flow Matching
+
+- **ID:** `2603.17403` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Yaozhong Shi, Grigorios Lavrentiadis, Konstantinos Tsalouchidis, et al.
+- **abs:** https://arxiv.org/abs/2603.17403 · **PDF:** https://arxiv.org/pdf/2603.17403v1
+
+## 3947. Bootstrapping Coding Agents: The Specification Is the Program
+
+- **ID:** `2603.17399` · **primary:** `cs.SE` · **Submitted:** 2026-03-18
+- **Authors:** Martin Monperrus
+- **abs:** https://arxiv.org/abs/2603.17399 · **PDF:** https://arxiv.org/pdf/2603.17399v1
+
+## 3948. Rapid Neural Network Prediction of Linear Block Copolymer Free Energies
+
+- **ID:** `2603.17391` · **primary:** `cond-mat.soft` · **Submitted:** 2026-03-18
+- **Authors:** Ian Chen, Alfredo Alexander-Katz
+- **abs:** https://arxiv.org/abs/2603.17391 · **PDF:** https://arxiv.org/pdf/2603.17391v1
+
+## 3949. CRE-T1 Preview Technical Report: Beyond Contrastive Learning for Reasoning-Intensive Retrieval
+
+- **ID:** `2603.17387` · **primary:** `cs.IR` · **Submitted:** 2026-03-18
+- **Authors:** Guangzhi Wang, Yinghao Jiao, Zhi Liu
+- **abs:** https://arxiv.org/abs/2603.17387 · **PDF:** https://arxiv.org/pdf/2603.17387v1
+
+## 3950. PJB: A Reasoning-Aware Benchmark for Person-Job Retrieval
+
+- **ID:** `2603.17386` · **primary:** `cs.IR` · **Submitted:** 2026-03-18
+- **Authors:** Guangzhi Wang, Xiaohui Yang, Kai Li, et al.
+- **abs:** https://arxiv.org/abs/2603.17386 · **PDF:** https://arxiv.org/pdf/2603.17386v1
+
+## 3951. The Causal Uncertainty Principle: Manifold Tearing and the Topological Limits of Counterfactual Interventions
+
+- **ID:** `2603.17385` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Rui Wu, Hong Xie, Yongjun Li
+- **abs:** https://arxiv.org/abs/2603.17385 · **PDF:** https://arxiv.org/pdf/2603.17385v1
+
+## 3952. Cohomological Obstructions to Global Counterfactuals: A Sheaf-Theoretic Foundation for Generative Causal Models
+
+- **ID:** `2603.17384` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Rui Wu, Hong Xie, Yongjun Li
+- **abs:** https://arxiv.org/abs/2603.17384 · **PDF:** https://arxiv.org/pdf/2603.17384v1
+
+## 3953. An Auditable AI Agent Loop for Empirical Economics: A Case Study in Forecast Combination
+
+- **ID:** `2603.17381` · **primary:** `econ.EM` · **Submitted:** 2026-03-18
+- **Authors:** Minchul Shin
+- **abs:** https://arxiv.org/abs/2603.17381 · **PDF:** https://arxiv.org/pdf/2603.17381v3
+
+## 3954. SCALE:Scalable Conditional Atlas-Level Endpoint transport for virtual cell perturbation prediction
+
+- **ID:** `2603.17380` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Shuizhou Chen, Lang Yu, Kedu Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.17380 · **PDF:** https://arxiv.org/pdf/2603.17380v2
+
+## 3955. Efficient Exploration at Scale
+
+- **ID:** `2603.17378` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Seyed Mohammad Asghari, Chris Chute, Vikranth Dwaracherla, et al.
+- **abs:** https://arxiv.org/abs/2603.17378 · **PDF:** https://arxiv.org/pdf/2603.17378v1
+
+## 3956. SafeTutors: Benchmarking Pedagogical Safety in AI Tutoring Systems
+
+- **ID:** `2603.17373` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Rima Hazra, Bikram Ghuku, Ilona Marchenko, et al.
+- **abs:** https://arxiv.org/abs/2603.17373 · **PDF:** https://arxiv.org/pdf/2603.17373v1
+
+## 3957. Understanding and Defending VLM Jailbreaks via Jailbreak-Related Representation Shift
+
+- **ID:** `2603.17372` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Zhihua Wei, Qiang Li, Jian Ruan, et al.
+- **abs:** https://arxiv.org/abs/2603.17372 · **PDF:** https://arxiv.org/pdf/2603.17372v1
+
+## 3958. FastPFRec: A Fast Personalized Federated Recommendation with Secure Sharing
+
+- **ID:** `2603.20283` · **primary:** `cs.IR` · **Submitted:** 2026-03-18
+- **Authors:** Zhenxing Yan, Jidong Yuan, Yongqi Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.20283 · **PDF:** https://arxiv.org/pdf/2603.20283v1
+
+## 3959. Towards Safer Large Reasoning Models by Promoting Safety Decision-Making before Chain-of-Thought Generation
+
+- **ID:** `2603.17368` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Jianan Chen, Zhifang Zhang, Shuo He, et al.
+- **abs:** https://arxiv.org/abs/2603.17368 · **PDF:** https://arxiv.org/pdf/2603.17368v1
+
+## 3960. Variational Kernel Design for Internal Noise: Gaussian Chaos Noise, Representation Compatibility, and Reliable Deep Learning
+
+- **ID:** `2603.17365` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Ziran Liu
+- **abs:** https://arxiv.org/abs/2603.17365 · **PDF:** https://arxiv.org/pdf/2603.17365v1
+
+## 3961. Lightweight Adaptation for LLM-based Technical Service Agent: Latent Logic Augmentation and Robust Noise Reduction
+
+- **ID:** `2603.18074` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Yi Yu, Junzhuo Ma, Chenghuang Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.18074 · **PDF:** https://arxiv.org/pdf/2603.18074v1
+
+## 3962. Continually self-improving AI
+
+- **ID:** `2603.18073` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Zitong Yang
+- **abs:** https://arxiv.org/abs/2603.18073 · **PDF:** https://arxiv.org/pdf/2603.18073v1
+
+## 3963. Public Profile Matters: A Scalable Integrated Approach to Recommend Citations in the Wild
+
+- **ID:** `2603.17361` · **primary:** `cs.IR` · **Submitted:** 2026-03-18
+- **Authors:** Karan Goyal, Dikshant Kukreja, Vikram Goyal, et al.
+- **abs:** https://arxiv.org/abs/2603.17361 · **PDF:** https://arxiv.org/pdf/2603.17361v1
+
+## 3964. WebPII: Benchmarking Visual PII Detection for Computer-Use Agents
+
+- **ID:** `2603.17357` · **primary:** `cs.CR` · **Submitted:** 2026-03-18
+- **Authors:** Nathan Zhao
+- **abs:** https://arxiv.org/abs/2603.17357 · **PDF:** https://arxiv.org/pdf/2603.17357v1
+
+## 3965. PACE-RAG: Patient-Aware Contextual and Evidence-based Policy RAG for Clinical Drug Recommendation
+
+- **ID:** `2603.17356` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Chaeyoung Huh, Hyunmin Hwang, Jung Hwan Shin, et al.
+- **abs:** https://arxiv.org/abs/2603.17356 · **PDF:** https://arxiv.org/pdf/2603.17356v1
+
+## 3966. Beyond Outliers: A Data-Free Layer-wise Mixed-Precision Quantization Approach Driven by Numerical and Structural Dual-Sensitivity
+
+- **ID:** `2603.17354` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Hengyuan Zhang, Xinrong Chen, Zunhai Su, et al.
+- **abs:** https://arxiv.org/abs/2603.17354 · **PDF:** https://arxiv.org/pdf/2603.17354v1
+
+## 3967. Learning Permutation Distributions via Reflected Diffusion on Ranks
+
+- **ID:** `2603.17353` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Sizhuang He, Yangtian Zhang, Shiyang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.17353 · **PDF:** https://arxiv.org/pdf/2603.17353v1
+
+## 3968. Grid Spatial Understanding: A Dataset for Textual Spatial Reasoning over Grids, Embodied Settings, and Coordinate Structures
+
+- **ID:** `2603.17333` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Risham Sidhu, Julia Hockenmaier
+- **abs:** https://arxiv.org/abs/2603.17333 · **PDF:** https://arxiv.org/pdf/2603.17333v1
+
+## 3969. A Progressive Visual-Logic-Aligned Framework for Ride-Hailing Adjudication
+
+- **ID:** `2603.17328` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Weiming Wu, Zi-Jian Cheng, Jie Meng, et al.
+- **abs:** https://arxiv.org/abs/2603.17328 · **PDF:** https://arxiv.org/pdf/2603.17328v1
+
+## 3970. ShuttleEnv: An Interactive Data-Driven RL Environment for Badminton Strategy Modeling
+
+- **ID:** `2603.17324` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Ang Li, Xinyang Gong, Bozhou Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.17324 · **PDF:** https://arxiv.org/pdf/2603.17324v1
+
+## 3971. Physics-informed offline reinforcement learning eliminates catastrophic fuel waste in maritime routing
+
+- **ID:** `2603.17319` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Aniruddha Bora, Julie Chalfant, Chryssostomos Chryssostomidis
+- **abs:** https://arxiv.org/abs/2603.17319 · **PDF:** https://arxiv.org/pdf/2603.17319v1
+
+## 3972. Recurrent Reasoning with Vision-Language Models for Estimating Long-Horizon Embodied Task Progress
+
+- **ID:** `2603.17312` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Yuelin Zhang, Sijie Cheng, Chen Li, et al.
+- **abs:** https://arxiv.org/abs/2603.17312 · **PDF:** https://arxiv.org/pdf/2603.17312v1
+
+## 3973. Ruyi2.5 Technical Report
+
+- **ID:** `2603.17311` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Huan Song, Shuyu Tian, Qingfei Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.17311 · **PDF:** https://arxiv.org/pdf/2603.17311v1
+
+## 3974. InfoDensity: Rewarding Information-Dense Traces for Efficient Reasoning
+
+- **ID:** `2603.17310` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Chengwei Wei, Jung-jae Kim, Longyin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.17310 · **PDF:** https://arxiv.org/pdf/2603.17310v1
+
+## 3975. ReLMXEL: Adaptive RL-Based Memory Controller with Explainable Energy and Latency Optimization
+
+- **ID:** `2603.17309` · **primary:** `cs.AR` · **Submitted:** 2026-03-18
+- **Authors:** Panuganti Chirag Sai, Gandholi Sarat, R. Raghunatha Sarma, et al.
+- **abs:** https://arxiv.org/abs/2603.17309 · **PDF:** https://arxiv.org/pdf/2603.17309v1
+
+## 3976. A Synthesizable RTL Implementation of Predictive Coding Networks
+
+- **ID:** `2603.18066` · **primary:** `cs.NE` · **Submitted:** 2026-03-18
+- **Authors:** Timothy Oh
+- **abs:** https://arxiv.org/abs/2603.18066 · **PDF:** https://arxiv.org/pdf/2603.18066v1
+
+## 3977. Symphony: A Cognitively-Inspired Multi-Agent System for Long-Video Understanding
+
+- **ID:** `2603.17307` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Haiyang Yan, Hongyun Zhou, Peng Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.17307 · **PDF:** https://arxiv.org/pdf/2603.17307v1
+
+## 3978. Beyond bouba/kiki: Multidimensional semantic signals are deeply woven into the fabric of natural language
+
+- **ID:** `2603.17306` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Gexin Zhao
+- **abs:** https://arxiv.org/abs/2603.17306 · **PDF:** https://arxiv.org/pdf/2603.17306v2
+
+## 3979. Contrastive Reasoning Alignment: Reinforcement Learning from Hidden Representations
+
+- **ID:** `2603.17305` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Haozheng Luo, Yimin Wang, Jiahao Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.17305 · **PDF:** https://arxiv.org/pdf/2603.17305v1
+
+## 3980. From Words to Worlds: Benchmarking Cross-Cultural Cultural Understanding in Machine Translation
+
+- **ID:** `2603.17303` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Bangju Han, Yingqi Wang, Huang Qing, et al.
+- **abs:** https://arxiv.org/abs/2603.17303 · **PDF:** https://arxiv.org/pdf/2603.17303v1
+
+## 3981. WINFlowNets: Warm-up Integrated Networks Training of Generative Flow Networks for Robotics and Machine Fault Adaptation
+
+- **ID:** `2603.17301` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Zahin Sufiyan, Shadan Golestan, Yoshihiro Mitsuka, et al.
+- **abs:** https://arxiv.org/abs/2603.17301 · **PDF:** https://arxiv.org/pdf/2603.17301v1
+
+## 3982. GUIDE: GenAI Units In Digital Design Education
+
+- **ID:** `2603.17296` · **primary:** `cs.CY` · **Submitted:** 2026-03-18
+- **Authors:** Weihua Xiao, Jason Blocklove, Matthew DeLorenzo, et al.
+- **abs:** https://arxiv.org/abs/2603.17296 · **PDF:** https://arxiv.org/pdf/2603.17296v1
+
+## 3983. Directing the Narrative: A Finetuning Method for Controlling Coherence and Style in Story Generation
+
+- **ID:** `2603.17295` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Jianzhang Zhang, Yijing Tian, Jiwang Qu, et al.
+- **abs:** https://arxiv.org/abs/2603.17295 · **PDF:** https://arxiv.org/pdf/2603.17295v1
+
+## 3984. CN-Buzz2Portfolio: A Chinese-Market Dataset and Benchmark for LLM-Based Macro and Sector Asset Allocation from Daily Trending Financial News
+
+- **ID:** `2603.22305` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Liyuan Chen, Shilong Li, Jiangpeng Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.22305 · **PDF:** https://arxiv.org/pdf/2603.22305v1
+
+## 3985. Boundary-aware Prototype-driven Adversarial Alignment for Cross-Corpus EEG Emotion Recognition
+
+- **ID:** `2603.26713` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Guangli Li, Canbiao Wu, Na Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.26713 · **PDF:** https://arxiv.org/pdf/2603.26713v1
+
+## 3986. MCP-38: A Comprehensive Threat Taxonomy for Model Context Protocol Systems (v1.0)
+
+- **ID:** `2603.18063` · **primary:** `cs.CR` · **Submitted:** 2026-03-18
+- **Authors:** Yi Ting Shen, Kentaroh Toyoda, Alex Leung
+- **abs:** https://arxiv.org/abs/2603.18063 · **PDF:** https://arxiv.org/pdf/2603.18063v1
+
+## 3987. Classifier Pooling for Modern Ordinal Classification
+
+- **ID:** `2603.17278` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Noam H. Rotenberg, Andreia V. Faria, Brian Caffo
+- **abs:** https://arxiv.org/abs/2603.17278 · **PDF:** https://arxiv.org/pdf/2603.17278v1
+
+## 3988. S3T-Former: A Purely Spike-Driven State-Space Topology Transformer for Skeleton Action Recognition
+
+- **ID:** `2603.18062` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Naichuan Zheng, Hailun Xia, Zepeng Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.18062 · **PDF:** https://arxiv.org/pdf/2603.18062v2
+
+## 3989. Mathematical Modeling of Cancer-Bacterial Therapy: Analysis and Numerical Simulation via Physics-Informed Neural Networks
+
+- **ID:** `2603.19326` · **primary:** `q-bio.QM` · **Submitted:** 2026-03-18
+- **Authors:** Ayoub Farkane, David Lassounon
+- **abs:** https://arxiv.org/abs/2603.19326 · **PDF:** https://arxiv.org/pdf/2603.19326v1
+
+## 3990. DANCE: Dynamic 3D CNN Pruning: Joint Frame, Channel, and Feature Adaptation for Energy Efficiency on the Edge
+
+- **ID:** `2603.17275` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Mohamed Mejri, Ashiqur Rasul, Abhijit Chatterjee
+- **abs:** https://arxiv.org/abs/2603.17275 · **PDF:** https://arxiv.org/pdf/2603.17275v1
+
+## 3991. On the Fragility of AI Agent Collusion
+
+- **ID:** `2603.20281` · **primary:** `cs.GT` · **Submitted:** 2026-03-18
+- **Authors:** Jussi Keppo, Yuze Li, Gerry Tsoukalas, et al.
+- **abs:** https://arxiv.org/abs/2603.20281 · **PDF:** https://arxiv.org/pdf/2603.20281v1
+
+## 3992. Wasserstein-type Gaussian Process Regressions for Input Measurement Uncertainty
+
+- **ID:** `2603.17271` · **primary:** `stat.ME` · **Submitted:** 2026-03-18
+- **Authors:** Hengrui Luo, Xiaoye S. Li, Yang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.17271 · **PDF:** https://arxiv.org/pdf/2603.17271v1
+
+## 3993. LED: A Benchmark for Evaluating Layout Error Detection in Document Analysis
+
+- **ID:** `2603.17265` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Inbum Heo, Taewook Hwang, Jeesu Jung, et al.
+- **abs:** https://arxiv.org/abs/2603.17265 · **PDF:** https://arxiv.org/pdf/2603.17265v1
+
+## 3994. Variational Rectification Inference for Learning with Noisy Labels
+
+- **ID:** `2603.17255` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Haoliang Sun, Qi Wei, Lei Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.17255 · **PDF:** https://arxiv.org/pdf/2603.17255v1
+
+## 3995. Pathology-Aware Multi-View Contrastive Learning for Patient-Independent ECG Reconstruction
+
+- **ID:** `2603.17248` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Youssef Youssef, Jitin Singla
+- **abs:** https://arxiv.org/abs/2603.17248 · **PDF:** https://arxiv.org/pdf/2603.17248v1
+
+## 3996. Binary Latent Protein Fitness Landscapes for Quantum Annealing Optimization
+
+- **ID:** `2603.17247` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Truong-Son Hy
+- **abs:** https://arxiv.org/abs/2603.17247 · **PDF:** https://arxiv.org/pdf/2603.17247v1
+
+## 3997. On the Cone Effect and Modality Gap in Medical Vision-Language Embeddings
+
+- **ID:** `2603.17246` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** David Restrepo, Miguel L Martins, Chenwei Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.17246 · **PDF:** https://arxiv.org/pdf/2603.17246v2
+
+## 3998. Graph-Native Cognitive Memory for AI Agents: Formal Belief Revision Semantics for Versioned Memory Architectures
+
+- **ID:** `2603.17244` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Young Bin Park
+- **abs:** https://arxiv.org/abs/2603.17244 · **PDF:** https://arxiv.org/pdf/2603.17244v1
+
+## 3999. Deployment and Evaluation of an EHR-integrated, Large Language Model-Powered Tool to Triage Surgical Patients
+
+- **ID:** `2603.17234` · **primary:** `cs.CY` · **Submitted:** 2026-03-18
+- **Authors:** Jane Wang, Timothy Keyes, April S Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.17234 · **PDF:** https://arxiv.org/pdf/2603.17234v1
+
+## 4000. Draft-and-Prune: Improving the Reliability of Auto-formalization for Logical Reasoning
+
+- **ID:** `2603.17233` · **primary:** `cs.AI` · **Submitted:** 2026-03-18
+- **Authors:** Zhiyu Ni, Zheng Liang, Liangcheng Song, et al.
+- **abs:** https://arxiv.org/abs/2603.17233 · **PDF:** https://arxiv.org/pdf/2603.17233v2
+
+## 4001. Neuron-Level Emotion Control in Speech-Generative Large Audio-Language Models
+
+- **ID:** `2603.17231` · **primary:** `cs.CL` · **Submitted:** 2026-03-18
+- **Authors:** Xiutian Zhao, Ismail Rasim Ulgen, Philipp Koehn, et al.
+- **abs:** https://arxiv.org/abs/2603.17231 · **PDF:** https://arxiv.org/pdf/2603.17231v1
+
+## 4002. KANtize: Exploring Low-bit Quantization of Kolmogorov-Arnold Networks for Efficient Inference
+
+- **ID:** `2603.17230` · **primary:** `cs.AR` · **Submitted:** 2026-03-18
+- **Authors:** Sohaib Errabii, Olivier Sentieys, Marcello Traiola
+- **abs:** https://arxiv.org/abs/2603.17230 · **PDF:** https://arxiv.org/pdf/2603.17230v1
+
+## 4003. From Drop-off to Recovery: A Mechanistic Analysis of Segmentation in MLLMs
+
+- **ID:** `2603.17228` · **primary:** `cs.CV` · **Submitted:** 2026-03-18
+- **Authors:** Boyong Wu, Sanghwan Kim, Zeynep Akata
+- **abs:** https://arxiv.org/abs/2603.17228 · **PDF:** https://arxiv.org/pdf/2603.17228v1
+
+## 4004. Fundamental Limits of Neural Network Sparsification: Evidence from Catastrophic Interpretability Collapse
+
+- **ID:** `2603.18056` · **primary:** `cs.LG` · **Submitted:** 2026-03-18
+- **Authors:** Dip Roy, Rajiv Misra, Sanjay Kumar Singh
+- **abs:** https://arxiv.org/abs/2603.18056 · **PDF:** https://arxiv.org/pdf/2603.18056v1
+
+## 4005. TharuChat: Bootstrapping Large Language Models for a Low-Resource Language via Synthetic Data and Human Validation
+
+- **ID:** `2603.17220` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Prajwal Panth, Agniva Maiti
+- **abs:** https://arxiv.org/abs/2603.17220 · **PDF:** https://arxiv.org/pdf/2603.17220v1
+
+## 4006. SA-CycleGAN-2.5D: Self-Attention CycleGAN with Tri-Planar Context for Multi-Site MRI Harmonization
+
+- **ID:** `2603.17219` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Ishrith Gowda, Chunwei Liu
+- **abs:** https://arxiv.org/abs/2603.17219 · **PDF:** https://arxiv.org/pdf/2603.17219v2
+
+## 4007. Alignment Makes Language Models Normative, Not Descriptive
+
+- **ID:** `2603.17218` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Eilam Shapira, Moshe Tennenholtz, Roi Reichart
+- **abs:** https://arxiv.org/abs/2603.17218 · **PDF:** https://arxiv.org/pdf/2603.17218v1
+
+## 4008. Anonymous-by-Construction: An LLM-Driven Framework for Privacy-Preserving Text
+
+- **ID:** `2603.17217` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Federico Albanese, Pablo Ronco, Nicolás D'Ippolito
+- **abs:** https://arxiv.org/abs/2603.17217 · **PDF:** https://arxiv.org/pdf/2603.17217v1
+
+## 4009. AI Scientist via Synthetic Task Scaling
+
+- **ID:** `2603.17216` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Ziyang Cai, Harkirat Behl
+- **abs:** https://arxiv.org/abs/2603.17216 · **PDF:** https://arxiv.org/pdf/2603.17216v1
+
+## 4010. An FPGA-Based SoC Architecture with a RISC-V Controller for Energy-Efficient Temporal-Coding Spiking Neural Networks
+
+- **ID:** `2603.18054` · **primary:** `cs.AR` · **Submitted:** 2026-03-17
+- **Authors:** Mohammad Javad Sekonji, Ali Mahani, Maryam Mirsadeghi, et al.
+- **abs:** https://arxiv.org/abs/2603.18054 · **PDF:** https://arxiv.org/pdf/2603.18054v1
+
+## 4011. Adaptive Contracts for Cost-Effective AI Delegation
+
+- **ID:** `2603.17212` · **primary:** `cs.GT` · **Submitted:** 2026-03-17
+- **Authors:** Eden Saig, Tamar Garbuz, Ariel D. Procaccia, et al.
+- **abs:** https://arxiv.org/abs/2603.17212 · **PDF:** https://arxiv.org/pdf/2603.17212v1
+
+## 4012. Mix-and-Match Pruning: Globally Guided Layer-Wise Sparsification of DNNs
+
+- **ID:** `2603.20280` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Danial Monachan, Samira Nazari, Mahdi Taheri, et al.
+- **abs:** https://arxiv.org/abs/2603.20280 · **PDF:** https://arxiv.org/pdf/2603.20280v1
+
+## 4013. A scalable neural bundle map for multiphysics prediction in lithium-ion battery across varying configurations
+
+- **ID:** `2603.17209` · **primary:** `cs.CE` · **Submitted:** 2026-03-17
+- **Authors:** Zhiwei Zhao, Changqing Liu, Jie Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.17209 · **PDF:** https://arxiv.org/pdf/2603.17209v1
+
+## 4014. SYMDIREC: A Neuro-Symbolic Divide-Retrieve-Conquer Framework for Enhanced RTL Synthesis and Summarization
+
+- **ID:** `2603.17208` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Prashanth Vijayaraghavan, Apoorva Nitsure, Luyao Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.17208 · **PDF:** https://arxiv.org/pdf/2603.17208v1
+
+## 4015. OPERA: Online Data Pruning for Efficient Retrieval Model Adaptation
+
+- **ID:** `2603.17205` · **primary:** `cs.IR` · **Submitted:** 2026-03-17
+- **Authors:** Haoyang Fang, Shuai Zhang, Yifei Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.17205 · **PDF:** https://arxiv.org/pdf/2603.17205v2
+
+## 4016. CODMAS: A Dialectic Multi-Agent Collaborative Framework for Structured RTL Optimization
+
+- **ID:** `2603.17204` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Che-Ming Chang, Prashanth Vijayaraghavan, Ashutosh Jadhav, et al.
+- **abs:** https://arxiv.org/abs/2603.17204 · **PDF:** https://arxiv.org/pdf/2603.17204v1
+
+## 4017. Catching rationalization in the act: detecting motivated reasoning before and after CoT via activation probing
+
+- **ID:** `2603.17199` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Parsa Mirtaheri, Mikhail Belkin
+- **abs:** https://arxiv.org/abs/2603.17199 · **PDF:** https://arxiv.org/pdf/2603.17199v1
+
+## 4018. Abstraction as a Memory-Efficient Inductive Bias for Continual Learning
+
+- **ID:** `2603.17198` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Elnaz Rahmati, Nona Ghazizadeh, Zhivar Sourati, et al.
+- **abs:** https://arxiv.org/abs/2603.17198 · **PDF:** https://arxiv.org/pdf/2603.17198v1
+
+## 4019. Self-Conditioned Denoising for Atomistic Representation Learning
+
+- **ID:** `2603.17196` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Tynan Perez, Rafael Gomez-Bombarelli
+- **abs:** https://arxiv.org/abs/2603.17196 · **PDF:** https://arxiv.org/pdf/2603.17196v1
+
+## 4020. Tabular LLMs for Interpretable Few-Shot Alzheimer's Disease Prediction with Multimodal Biomedical Data
+
+- **ID:** `2603.17191` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Sophie Kearney, Shu Yang, Zixuan Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.17191 · **PDF:** https://arxiv.org/pdf/2603.17191v1
+
+## 4021. MetaClaw: Just Talk -- An Agent That Meta-Learns and Evolves in the Wild
+
+- **ID:** `2603.17187` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Peng Xia, Jianwen Chen, Xinyu Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.17187 · **PDF:** https://arxiv.org/pdf/2603.17187v1
+
+## 4022. On the Carbon Footprint of Economic Research in the Age of Generative AI
+
+- **ID:** `2603.26712` · **primary:** `cs.SE` · **Submitted:** 2026-03-17
+- **Authors:** Andres Alonso-Robisco, Carlos Esparcia, Francisco Jareño
+- **abs:** https://arxiv.org/abs/2603.26712 · **PDF:** https://arxiv.org/pdf/2603.26712v1
+
+## 4023. Towards Unsupervised Adversarial Document Detection in Retrieval Augmented Generation Systems
+
+- **ID:** `2603.17176` · **primary:** `cs.CR` · **Submitted:** 2026-03-17
+- **Authors:** Patrick Levi
+- **abs:** https://arxiv.org/abs/2603.17176 · **PDF:** https://arxiv.org/pdf/2603.17176v1
+
+## 4024. Domain-informed explainable boosting machines for trustworthy lateral spread predictions
+
+- **ID:** `2603.17175` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Cheng-Hsi Hsiao, Krishna Kumar, Ellen M. Rathje
+- **abs:** https://arxiv.org/abs/2603.17175 · **PDF:** https://arxiv.org/pdf/2603.17175v1
+
+## 4025. Detecting Data Poisoning in Code Generation LLMs via Black-Box, Vulnerability-Oriented Scanning
+
+- **ID:** `2603.17174` · **primary:** `cs.CR` · **Submitted:** 2026-03-17
+- **Authors:** Shenao Yan, Shimaa Ahmed, Shan Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.17174 · **PDF:** https://arxiv.org/pdf/2603.17174v1
+
+## 4026. Generalist Multimodal LLMs Gain Biometric Expertise via Human Salience
+
+- **ID:** `2603.17173` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Jacob Piland, Byron Dowling, Christopher Sweet, et al.
+- **abs:** https://arxiv.org/abs/2603.17173 · **PDF:** https://arxiv.org/pdf/2603.17173v1
+
+## 4027. Noise-Response Calibration: A Causal Intervention Protocol for LLM-Judges
+
+- **ID:** `2603.17172` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Maxim Khomiakov, Jes Frellsen
+- **abs:** https://arxiv.org/abs/2603.17172 · **PDF:** https://arxiv.org/pdf/2603.17172v1
+
+## 4028. Exploiting the English Grammar Profile for L2 grammatical analysis with LLMs
+
+- **ID:** `2603.17171` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Stefano Bannò, Penny Karanasou, Kate Knill, et al.
+- **abs:** https://arxiv.org/abs/2603.17171 · **PDF:** https://arxiv.org/pdf/2603.17171v1
+
+## 4029. PAuth - Precise Task-Scoped Authorization For Agents
+
+- **ID:** `2603.17170` · **primary:** `cs.CR` · **Submitted:** 2026-03-17
+- **Authors:** Reshabh K Sharma, Linxi Jiang, Zhiqiang Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.17170 · **PDF:** https://arxiv.org/pdf/2603.17170v1
+
+## 4030. How Clued up are LLMs? Evaluating Multi-Step Deductive Reasoning in a Text-Based Game Environment
+
+- **ID:** `2603.17169` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Rebecca Ansell, Autumn Toney-Wails
+- **abs:** https://arxiv.org/abs/2603.17169 · **PDF:** https://arxiv.org/pdf/2603.17169v1
+
+## 4031. Auditing the Auditors: Does Community-based Moderation Get It Right?
+
+- **ID:** `2603.18053` · **primary:** `cs.SI` · **Submitted:** 2026-03-17
+- **Authors:** Yeganeh Alimohammadi, Karissa Huang, Christian Borgs, et al.
+- **abs:** https://arxiv.org/abs/2603.18053 · **PDF:** https://arxiv.org/pdf/2603.18053v1
+
+## 4032. Quadratic Surrogate Attractor for Particle Swarm Optimization
+
+- **ID:** `2603.17163` · **primary:** `cs.NE` · **Submitted:** 2026-03-17
+- **Authors:** Maurizio Clemente, Marcello Canova
+- **abs:** https://arxiv.org/abs/2603.17163 · **PDF:** https://arxiv.org/pdf/2603.17163v1
+
+## 4033. Self-Regularized Learning Methods
+
+- **ID:** `2603.17160` · **primary:** `stat.ML` · **Submitted:** 2026-03-17
+- **Authors:** Max Schölpple, Liu Fanghui, Ingo Steinwart
+- **abs:** https://arxiv.org/abs/2603.17160 · **PDF:** https://arxiv.org/pdf/2603.17160v1
+
+## 4034. Learning Communication Between Heterogeneous Agents in Multi-Agent Reinforcement Learning for Autonomous Cyber Defence
+
+- **ID:** `2603.20279` · **primary:** `cs.CR` · **Submitted:** 2026-03-17
+- **Authors:** Alex Popa, Adrian Taylor, Ranwa Al Mallah
+- **abs:** https://arxiv.org/abs/2603.20279 · **PDF:** https://arxiv.org/pdf/2603.20279v1
+
+## 4035. Agentic AI for Human Resources: LLM-Driven Candidate Assessment
+
+- **ID:** `2603.26710` · **primary:** `cs.IR` · **Submitted:** 2026-03-17
+- **Authors:** Kamer Ali Yuksel, Abdul Basit Anees, Ashraf Elneima, et al.
+- **abs:** https://arxiv.org/abs/2603.26710 · **PDF:** https://arxiv.org/pdf/2603.26710v1
+
+## 4036. Shielded Reinforcement Learning Under Dynamic Temporal Logic Constraints
+
+- **ID:** `2603.17152` · **primary:** `cs.RO` · **Submitted:** 2026-03-17
+- **Authors:** Sadık Bera Yüksel, Ali Tevfik Buyukkocak, Derya Aksaray
+- **abs:** https://arxiv.org/abs/2603.17152 · **PDF:** https://arxiv.org/pdf/2603.17152v1
+
+## 4037. Shallow Representation of Option Implied Information
+
+- **ID:** `2603.17151` · **primary:** `q-fin.CP` · **Submitted:** 2026-03-17
+- **Authors:** Jimin Lin
+- **abs:** https://arxiv.org/abs/2603.17151 · **PDF:** https://arxiv.org/pdf/2603.17151v1
+
+## 4038. Intent Formalization: A Grand Challenge for Reliable Coding in the Age of AI Agents
+
+- **ID:** `2603.17150` · **primary:** `cs.SE` · **Submitted:** 2026-03-17
+- **Authors:** Shuvendu K. Lahiri
+- **abs:** https://arxiv.org/abs/2603.17150 · **PDF:** https://arxiv.org/pdf/2603.17150v1
+
+## 4039. Personalized Fall Detection by Balancing Data with Selective Feedback Using Contrastive Learning
+
+- **ID:** `2603.17148` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Awatif Yasmin, Tarek Mahmud, Sana Alamgeer, et al.
+- **abs:** https://arxiv.org/abs/2603.17148 · **PDF:** https://arxiv.org/pdf/2603.17148v1
+
+## 4040. Multilingual Reference Need Assessment System for Wikipedia
+
+- **ID:** `2603.17146` · **primary:** `cs.CY` · **Submitted:** 2026-03-17
+- **Authors:** Aitolkyn Baigutanova, Francisco Navas, Pablo Aragon, et al.
+- **abs:** https://arxiv.org/abs/2603.17146 · **PDF:** https://arxiv.org/pdf/2603.17146v1
+
+## 4041. REAL: Regression-Aware Reinforcement Learning for LLM-as-a-Judge
+
+- **ID:** `2603.17145` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Yasi Zhang, Tianyu Chen, Mingyuan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.17145 · **PDF:** https://arxiv.org/pdf/2603.17145v1
+
+## 4042. Contextual Preference Distribution Learning
+
+- **ID:** `2603.17139` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Benjamin Hudson, Laurent Charlin, Emma Frejinger
+- **abs:** https://arxiv.org/abs/2603.17139 · **PDF:** https://arxiv.org/pdf/2603.17139v1
+
+## 4043. Topology-Preserving Deep Joint Source-Channel Coding for Semantic Communication
+
+- **ID:** `2603.17126` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Omar Erak, Omar Alhussein, Fang Fang, et al.
+- **abs:** https://arxiv.org/abs/2603.17126 · **PDF:** https://arxiv.org/pdf/2603.17126v1
+
+## 4044. Security Assessment and Mitigation Strategies for Large Language Models: A Comprehensive Defensive Framework
+
+- **ID:** `2603.17123` · **primary:** `cs.CR` · **Submitted:** 2026-03-17
+- **Authors:** Taiwo Onitiju, Iman Vakilinia
+- **abs:** https://arxiv.org/abs/2603.17123 · **PDF:** https://arxiv.org/pdf/2603.17123v1
+
+## 4045. Cascade-Aware Multi-Agent Routing: Spatio-Temporal Sidecars and Geometry-Switching
+
+- **ID:** `2603.17112` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Davide Di Gioia
+- **abs:** https://arxiv.org/abs/2603.17112 · **PDF:** https://arxiv.org/pdf/2603.17112v2
+
+## 4046. OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis
+
+- **ID:** `2603.20278` · **primary:** `cs.IR` · **Submitted:** 2026-03-17
+- **Authors:** Zhuofeng Li, Dongfu Jiang, Xueguang Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.20278 · **PDF:** https://arxiv.org/pdf/2603.20278v1
+
+## 4047. Hidden Clones: Exposing and Fixing Family Bias in Vision-Language Model Ensembles
+
+- **ID:** `2603.17111` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Zacharie Bugaud
+- **abs:** https://arxiv.org/abs/2603.17111 · **PDF:** https://arxiv.org/pdf/2603.17111v1
+
+## 4048. Pixel-level Counterfactual Contrastive Learning for Medical Image Segmentation
+
+- **ID:** `2603.17110` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Marceau Lafargue-Hauret, Raghav Mehta, Fabio De Sousa Ribeiro, et al.
+- **abs:** https://arxiv.org/abs/2603.17110 · **PDF:** https://arxiv.org/pdf/2603.17110v1
+
+## 4049. SENSE: Efficient EEG-to-Text via Privacy-Preserving Semantic Retrieval
+
+- **ID:** `2603.17109` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Akshaj Murhekar, Christina Liu, Abhijit Mishra, et al.
+- **abs:** https://arxiv.org/abs/2603.17109 · **PDF:** https://arxiv.org/pdf/2603.17109v1
+
+## 4050. When the Specification Emerges: Benchmarking Faithfulness Loss in Long-Horizon Coding Agents
+
+- **ID:** `2603.17104` · **primary:** `cs.SE` · **Submitted:** 2026-03-17
+- **Authors:** Lu Yan, Xuan Chen, Xiangyu Zhang
+- **abs:** https://arxiv.org/abs/2603.17104 · **PDF:** https://arxiv.org/pdf/2603.17104v1
+
+## 4051. Knowledge Localization in Mixture-of-Experts LLMs Using Cross-Lingual Inconsistency
+
+- **ID:** `2603.17102` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Lucas Bandarkar, Alan Ansell, Trevor Cohn
+- **abs:** https://arxiv.org/abs/2603.17102 · **PDF:** https://arxiv.org/pdf/2603.17102v1
+
+## 4052. An End-to-End Framework for Functionality-Embedded Provenance Graph Construction and Threat Interpretation
+
+- **ID:** `2603.17100` · **primary:** `cs.CR` · **Submitted:** 2026-03-17
+- **Authors:** Kushankur Ghosh, Mehar Klair, Kian Kyars, et al.
+- **abs:** https://arxiv.org/abs/2603.17100 · **PDF:** https://arxiv.org/pdf/2603.17100v1
+
+## 4053. Evaluating LLM-Simulated Conversations in Modeling Inconsistent and Uncollaborative Behaviors in Human Social Interaction
+
+- **ID:** `2603.17094` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Ryo Kamoi, Ameya Godbole, Longqi Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.17094 · **PDF:** https://arxiv.org/pdf/2603.17094v1
+
+## 4054. Ensemble Self-Training for Unsupervised Machine Translation
+
+- **ID:** `2603.17087` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Ido Aharon, Jonathan Shaki, Sarit Kraus
+- **abs:** https://arxiv.org/abs/2603.17087 · **PDF:** https://arxiv.org/pdf/2603.17087v1
+
+## 4055. CircuitBuilder: From Polynomials to Circuits via Reinforcement Learning
+
+- **ID:** `2603.17075` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Weikun K. Zhang, Rohan Pandey, Bhaumik Mehta, et al.
+- **abs:** https://arxiv.org/abs/2603.17075 · **PDF:** https://arxiv.org/pdf/2603.17075v1
+
+## 4056. PRISM: Demystifying Retention and Interaction in Mid-Training
+
+- **ID:** `2603.17074` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Bharat Runwal, Ashish Agrawal, Anurag Roy, et al.
+- **abs:** https://arxiv.org/abs/2603.17074 · **PDF:** https://arxiv.org/pdf/2603.17074v3
+
+## 4057. Large Reasoning Models Struggle to Transfer Parametric Knowledge Across Scripts
+
+- **ID:** `2603.17070` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Lucas Bandarkar, Alan Ansell, Trevor Cohn
+- **abs:** https://arxiv.org/abs/2603.17070 · **PDF:** https://arxiv.org/pdf/2603.17070v1
+
+## 4058. The Cognitive Divergence: AI Context Windows, Human Attention Decline, and the Delegation Feedback Loop
+
+- **ID:** `2603.26707` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Netanel Eliav
+- **abs:** https://arxiv.org/abs/2603.26707 · **PDF:** https://arxiv.org/pdf/2603.26707v1
+
+## 4059. Evaluating Ill-Defined Tasks in Large Language Models
+
+- **ID:** `2603.17067` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Yi Zhou, Basel Shbita
+- **abs:** https://arxiv.org/abs/2603.17067 · **PDF:** https://arxiv.org/pdf/2603.17067v1
+
+## 4060. Transformers are Bayesian Networks
+
+- **ID:** `2603.17063` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Gregory Coppola
+- **abs:** https://arxiv.org/abs/2603.17063 · **PDF:** https://arxiv.org/pdf/2603.17063v1
+
+## 4061. Optimization-Embedded Active Multi-Fidelity Surrogate Learning for Multi-Condition Airfoil Shape Optimization
+
+- **ID:** `2603.17057` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-03-17
+- **Authors:** Isaac Robledo, Alberto Vilariño, Arnau Miró, et al.
+- **abs:** https://arxiv.org/abs/2603.17057 · **PDF:** https://arxiv.org/pdf/2603.17057v1
+
+## 4062. DesertFormer: Transformer-Based Semantic Segmentation for Off-Road Desert Terrain Classification in Autonomous Navigation Systems
+
+- **ID:** `2603.17056` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Yasaswini Chebolu
+- **abs:** https://arxiv.org/abs/2603.17056 · **PDF:** https://arxiv.org/pdf/2603.17056v1
+
+## 4063. Early Quantization Shrinks Codebook: A Simple Fix for Diversity-Preserving Tokenization
+
+- **ID:** `2603.17052` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Wenhao Zhao, Qiran Zou, Rushi Shah, et al.
+- **abs:** https://arxiv.org/abs/2603.17052 · **PDF:** https://arxiv.org/pdf/2603.17052v1
+
+## 4064. Attractor-Keyed Memory
+
+- **ID:** `2603.17049` · **primary:** `physics.optics` · **Submitted:** 2026-03-17
+- **Authors:** Natalia G. Berloff
+- **abs:** https://arxiv.org/abs/2603.17049 · **PDF:** https://arxiv.org/pdf/2603.17049v1
+
+## 4065. SCE-LITE-HQ: Smooth visual counterfactual explanations with generative foundation models
+
+- **ID:** `2603.17048` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Ahmed Zeid, Sidney Bender
+- **abs:** https://arxiv.org/abs/2603.17048 · **PDF:** https://arxiv.org/pdf/2603.17048v1
+
+## 4066. Do Understanding and Generation Fight? A Diagnostic Study of DPO for Unified Multimodal Models
+
+- **ID:** `2603.17044` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Abinav Rao, Sujan Rachuri
+- **abs:** https://arxiv.org/abs/2603.17044 · **PDF:** https://arxiv.org/pdf/2603.17044v1
+
+## 4067. Dependence Fidelity and Downstream Inference Stability in Generative Models
+
+- **ID:** `2603.17041` · **primary:** `stat.ML` · **Submitted:** 2026-03-17
+- **Authors:** Nazia Riasat
+- **abs:** https://arxiv.org/abs/2603.17041 · **PDF:** https://arxiv.org/pdf/2603.17041v1
+
+## 4068. Shared Representation Learning for Reference-Guided Targeted Sound Detection
+
+- **ID:** `2603.17025` · **primary:** `eess.AS` · **Submitted:** 2026-03-17
+- **Authors:** Shubham Gupta, Adarsh Arigala, B. R. Dilleswari, et al.
+- **abs:** https://arxiv.org/abs/2603.17025 · **PDF:** https://arxiv.org/pdf/2603.17025v1
+
+## 4069. HopChain: Multi-Hop Data Synthesis for Generalizable Vision-Language Reasoning
+
+- **ID:** `2603.17024` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Shenzhi Wang, Shixuan Liu, Jing Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.17024 · **PDF:** https://arxiv.org/pdf/2603.17024v2
+
+## 4070. Generative AI-assisted Participatory Modeling in Socio-Environmental Planning under Deep Uncertainty
+
+- **ID:** `2603.17021` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Zhihao Pei, Nir Lipovetzky, Angela M. Rojas-Arevalo, et al.
+- **abs:** https://arxiv.org/abs/2603.17021 · **PDF:** https://arxiv.org/pdf/2603.17021v2
+
+## 4071. Mitigating Premature Discretization with Progressive Quantization for Robust Vector Tokenization
+
+- **ID:** `2603.22304` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Wenhao Zhao, Qiran Zou, Zhouhan Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.22304 · **PDF:** https://arxiv.org/pdf/2603.22304v1
+
+## 4072. Transformers Can Learn Rules They've Never Seen: Proof of Computation Beyond Interpolation
+
+- **ID:** `2603.17019` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Andy Gray
+- **abs:** https://arxiv.org/abs/2603.17019 · **PDF:** https://arxiv.org/pdf/2603.17019v1
+
+## 4073. LLM NL2SQL Robustness: Surface Noise vs. Linguistic Variation in Traditional and Agentic Settings
+
+- **ID:** `2603.17017` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Lifu Tu, Rongguang Wang, Tao Sheng, et al.
+- **abs:** https://arxiv.org/abs/2603.17017 · **PDF:** https://arxiv.org/pdf/2603.17017v1
+
+## 4074. Demystifing Video Reasoning
+
+- **ID:** `2603.16870` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Ruisi Wang, Zhongang Cai, Fanyi Pu, et al.
+- **abs:** https://arxiv.org/abs/2603.16870 · **PDF:** https://arxiv.org/pdf/2603.16870v1
+
+## 4075. Efficient Reasoning on the Edge
+
+- **ID:** `2603.16867` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Yelysei Bondarenko, Thomas Hehn, Rob Hesselink, et al.
+- **abs:** https://arxiv.org/abs/2603.16867 · **PDF:** https://arxiv.org/pdf/2603.16867v1
+
+## 4076. MessyKitchens: Contact-rich object-level 3D scene reconstruction
+
+- **ID:** `2603.16868` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Junaid Ahmed Ansari, Ran Ding, Fabio Pizzati, et al.
+- **abs:** https://arxiv.org/abs/2603.16868 · **PDF:** https://arxiv.org/pdf/2603.16868v1
+
+## 4077. ManiTwin: Scaling Data-Generation-Ready Digital Object Dataset to 100K
+
+- **ID:** `2603.16866` · **primary:** `cs.RO` · **Submitted:** 2026-03-17
+- **Authors:** Kaixuan Wang, Tianxing Chen, Jiawei Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.16866 · **PDF:** https://arxiv.org/pdf/2603.16866v1
+
+## 4078. SparkVSR: Interactive Video Super-Resolution via Sparse Keyframe Propagation
+
+- **ID:** `2603.16864` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Jiongze Yu, Xiangbo Gao, Pooja Verlani, et al.
+- **abs:** https://arxiv.org/abs/2603.16864 · **PDF:** https://arxiv.org/pdf/2603.16864v1
+
+## 4079. Chronos: Temporal-Aware Conversational Agents with Structured Event Retrieval for Long-Term Memory
+
+- **ID:** `2603.16862` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Sahil Sen, Elias Lumer, Anmol Gulati, et al.
+- **abs:** https://arxiv.org/abs/2603.16862 · **PDF:** https://arxiv.org/pdf/2603.16862v1
+
+## 4080. SocialOmni: Benchmarking Audio-Visual Social Interactivity in Omni Models
+
+- **ID:** `2603.16859` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Tianyu Xie, Jinfa Huang, Yuexiao Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.16859 · **PDF:** https://arxiv.org/pdf/2603.16859v1
+
+## 4081. SOMA: Unifying Parametric Human Body Models
+
+- **ID:** `2603.16858` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Jun Saito, Jiefeng Li, Michael de Ruyter, et al.
+- **abs:** https://arxiv.org/abs/2603.16858 · **PDF:** https://arxiv.org/pdf/2603.16858v1
+
+## 4082. Long-Horizon Traffic Forecasting via Incident-Aware Conformal Spatio-Temporal Transformers
+
+- **ID:** `2603.16857` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Mayur Patil, Qadeer Ahmed, Shawn Midlam-Mohler, et al.
+- **abs:** https://arxiv.org/abs/2603.16857 · **PDF:** https://arxiv.org/pdf/2603.16857v1
+
+## 4083. Online Experiential Learning for Language Models
+
+- **ID:** `2603.16856` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Tianzhu Ye, Li Dong, Qingxiu Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.16856 · **PDF:** https://arxiv.org/pdf/2603.16856v1
+
+## 4084. Unifying Optimization and Dynamics to Parallelize Sequential Computation: A Guide to Parallel Newton Methods for Breaking Sequential Bottlenecks
+
+- **ID:** `2603.16850` · **primary:** `math.NA` · **Submitted:** 2026-03-17
+- **Authors:** Xavier Gonzalez
+- **abs:** https://arxiv.org/abs/2603.16850 · **PDF:** https://arxiv.org/pdf/2603.16850v1
+
+## 4085. GIST: Gauge-Invariant Spectral Transformers for Scalable Graph Neural Operators
+
+- **ID:** `2603.16849` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Mattia Rigotti, Nicholas Thumiger, Thomas Frick
+- **abs:** https://arxiv.org/abs/2603.16849 · **PDF:** https://arxiv.org/pdf/2603.16849v1
+
+## 4086. Mediocrity is the key for LLM as a Judge Anchor Selection
+
+- **ID:** `2603.16848` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Shachar Don-Yehiya, Asaf Yehudai, Leshem Choshen, et al.
+- **abs:** https://arxiv.org/abs/2603.16848 · **PDF:** https://arxiv.org/pdf/2603.16848v1
+
+## 4087. Dynamic Meta-Layer Aggregation for Byzantine-Robust Federated Learning
+
+- **ID:** `2603.16846` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Reek Das, Biplab Kanti Sen
+- **abs:** https://arxiv.org/abs/2603.16846 · **PDF:** https://arxiv.org/pdf/2603.16846v1
+
+## 4088. Internalizing Agency from Reflective Experience
+
+- **ID:** `2603.16843` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Rui Ge, Yichao Fu, Yuyang Qian, et al.
+- **abs:** https://arxiv.org/abs/2603.16843 · **PDF:** https://arxiv.org/pdf/2603.16843v1
+
+## 4089. Stochastic Resetting Accelerates Policy Convergence in Reinforcement Learning
+
+- **ID:** `2603.16842` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Jello Zhou, Vudtiwat Ngampruetikorn, David J. Schwab
+- **abs:** https://arxiv.org/abs/2603.16842 · **PDF:** https://arxiv.org/pdf/2603.16842v1
+
+## 4090. Learning to Present: Inverse Specification Rewards for Agentic Slide Generation
+
+- **ID:** `2603.16839` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Karthik Ragunath Ananda Kumar, Subrahmanyam Arunachalam
+- **abs:** https://arxiv.org/abs/2603.16839 · **PDF:** https://arxiv.org/pdf/2603.16839v1
+
+## 4091. Me, Myself, and $π$ : Evaluating and Explaining LLM Introspection
+
+- **ID:** `2603.20276` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Atharv Naphade, Samarth Bhargav, Sean Lim, et al.
+- **abs:** https://arxiv.org/abs/2603.20276 · **PDF:** https://arxiv.org/pdf/2603.20276v1
+
+## 4092. Conditional Distributional Treatment Effects: Doubly Robust Estimation and Testing
+
+- **ID:** `2603.16829` · **primary:** `stat.ML` · **Submitted:** 2026-03-17
+- **Authors:** Saksham Jain, Alex Luedtke
+- **abs:** https://arxiv.org/abs/2603.16829 · **PDF:** https://arxiv.org/pdf/2603.16829v1
+
+## 4093. Prompt Programming for Cultural Bias and Alignment of Large Language Models
+
+- **ID:** `2603.16827` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Maksim Eren, Eric Michalak, Brian Cook, et al.
+- **abs:** https://arxiv.org/abs/2603.16827 · **PDF:** https://arxiv.org/pdf/2603.16827v1
+
+## 4094. Real-Time Decoding of Movement Onset and Offset for Brain-Controlled Rehabilitation Exoskeleton
+
+- **ID:** `2603.16825` · **primary:** `cs.RO` · **Submitted:** 2026-03-17
+- **Authors:** Kanishka Mitra, Satyam Kumar, Frigyes Samuel Racz, et al.
+- **abs:** https://arxiv.org/abs/2603.16825 · **PDF:** https://arxiv.org/pdf/2603.16825v1
+
+## 4095. Surg$Σ$: A Spectrum of Large-Scale Multimodal Data and Foundation Models for Surgical Intelligence
+
+- **ID:** `2603.16822` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Zhitao Zeng, Mengya Xu, Jian Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.16822 · **PDF:** https://arxiv.org/pdf/2603.16822v1
+
+## 4096. Is Conformal Factuality for RAG-based LLMs Robust? Novel Metrics and Systematic Insights
+
+- **ID:** `2603.16817` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Yi Chen, Daiwei Chen, Sukrut Madhav Chikodikar, et al.
+- **abs:** https://arxiv.org/abs/2603.16817 · **PDF:** https://arxiv.org/pdf/2603.16817v1
+
+## 4097. Beyond Accuracy: Evaluating Forecasting Models by Multi-Echelon Inventory Cost
+
+- **ID:** `2603.16815` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Swata Marik, Swayamjit Saha, Garga Chatterjee
+- **abs:** https://arxiv.org/abs/2603.16815 · **PDF:** https://arxiv.org/pdf/2603.16815v1
+
+## 4098. Empirical Recipes for Efficient and Compact Vision-Language Models
+
+- **ID:** `2603.16987` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Jiabo Huang, Zhizhong Li, Sina Sajadmanesh, et al.
+- **abs:** https://arxiv.org/abs/2603.16987 · **PDF:** https://arxiv.org/pdf/2603.16987v1
+
+## 4099. ODIN-Based CPU-GPU Architecture with Replay-Driven Simulation and Emulation
+
+- **ID:** `2603.16812` · **primary:** `cs.DC` · **Submitted:** 2026-03-17
+- **Authors:** Nij Dorairaj, Debabrata Chatterjee, Hong Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.16812 · **PDF:** https://arxiv.org/pdf/2603.16812v1
+
+## 4100. CABTO: Context-Aware Behavior Tree Grounding for Robot Manipulation
+
+- **ID:** `2603.16809` · **primary:** `cs.RO` · **Submitted:** 2026-03-17
+- **Authors:** Yishuai Cai, Xinglin Chen, Yunxin Mao, et al.
+- **abs:** https://arxiv.org/abs/2603.16809 · **PDF:** https://arxiv.org/pdf/2603.16809v1
+
+## 4101. DexGrasp-Zero: A Morphology-Aligned Policy for Zero-Shot Cross-Embodiment Dexterous Grasping
+
+- **ID:** `2603.16806` · **primary:** `cs.RO` · **Submitted:** 2026-03-17
+- **Authors:** Yuliang Wu, Yanhan Lin, WengKit Lao, et al.
+- **abs:** https://arxiv.org/abs/2603.16806 · **PDF:** https://arxiv.org/pdf/2603.16806v2
+
+## 4102. Integrating Inductive Biases in Transformers via Distillation for Financial Time Series Forecasting
+
+- **ID:** `2603.16985` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Yu-Chen Den, Kuan-Yu Chen, Kendro Vincent, et al.
+- **abs:** https://arxiv.org/abs/2603.16985 · **PDF:** https://arxiv.org/pdf/2603.16985v1
+
+## 4103. RaDAR: Relation-aware Diffusion-Asymmetric Graph Contrastive Learning for Recommendation
+
+- **ID:** `2603.16800` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Yixuan Huang, Jiawei Chen, Shengfan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.16800 · **PDF:** https://arxiv.org/pdf/2603.16800v1
+
+## 4104. High-Dimensional Gaussian Mean Estimation under Realizable Contamination
+
+- **ID:** `2603.16798` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Ilias Diakonikolas, Daniel M. Kane, Thanasis Pittas
+- **abs:** https://arxiv.org/abs/2603.16798 · **PDF:** https://arxiv.org/pdf/2603.16798v1
+
+## 4105. Adaptive Moments are Surprisingly Effective for Plug-and-Play Diffusion Sampling
+
+- **ID:** `2603.16797` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Christian Belardi, Justin Lovelace, Kilian Q. Weinberger, et al.
+- **abs:** https://arxiv.org/abs/2603.16797 · **PDF:** https://arxiv.org/pdf/2603.16797v1
+
+## 4106. V-Co: A Closer Look at Visual Representation Alignment via Co-Denoising
+
+- **ID:** `2603.16792` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Han Lin, Xichen Pan, Zun Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.16792 · **PDF:** https://arxiv.org/pdf/2603.16792v1
+
+## 4107. InCoder-32B: Code Foundation Model for Industrial Scenarios
+
+- **ID:** `2603.16790` · **primary:** `cs.SE` · **Submitted:** 2026-03-17
+- **Authors:** Jian Yang, Wei Zhang, Jiajun Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.16790 · **PDF:** https://arxiv.org/pdf/2603.16790v3
+
+## 4108. Conservative Continuous-Time Treatment Optimization
+
+- **ID:** `2603.16789` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Nora Schneider, Georg Manten, Niki Kilbertus
+- **abs:** https://arxiv.org/abs/2603.16789 · **PDF:** https://arxiv.org/pdf/2603.16789v1
+
+## 4109. SpokenUS: A Spoken User Simulator for Task-Oriented Dialogue
+
+- **ID:** `2603.16783` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Jonggeun Lee, Junseong Pyo, Jeongmin Park, et al.
+- **abs:** https://arxiv.org/abs/2603.16783 · **PDF:** https://arxiv.org/pdf/2603.16783v1
+
+## 4110. IOSVLM: A 3D Vision-Language Model for Unified Dental Diagnosis from Intraoral Scans
+
+- **ID:** `2603.16781` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Huimin Xiong, Zijie Meng, Tianxiang Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.16781 · **PDF:** https://arxiv.org/pdf/2603.16781v1
+
+## 4111. Anticipatory Planning for Multimodal AI Agents
+
+- **ID:** `2603.16777` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Yongyuan Liang, Shijie Zhou, Yu Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.16777 · **PDF:** https://arxiv.org/pdf/2603.16777v1
+
+## 4112. SOMP: Scalable Gradient Inversion for Large Language Models via Subspace-Guided Orthogonal Matching Pursuit
+
+- **ID:** `2603.16761` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Yibo Li, Qiongxiu Li
+- **abs:** https://arxiv.org/abs/2603.16761 · **PDF:** https://arxiv.org/pdf/2603.16761v1
+
+## 4113. TurnWise: The Gap between Single- and Multi-turn Language Model Capabilities
+
+- **ID:** `2603.16759` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Victoria Graf, Valentina Pyatkin, Nouha Dziri, et al.
+- **abs:** https://arxiv.org/abs/2603.16759 · **PDF:** https://arxiv.org/pdf/2603.16759v1
+
+## 4114. pADAM: A Plug-and-Play All-in-One Diffusion Architecture for Multi-Physics Learning
+
+- **ID:** `2603.16757` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Amirhossein Mollaali, Bongseok Kim, Christian Moya, et al.
+- **abs:** https://arxiv.org/abs/2603.16757 · **PDF:** https://arxiv.org/pdf/2603.16757v1
+
+## 4115. A Practical Algorithm for Feature-Rich, Non-Stationary Bandit Problems
+
+- **ID:** `2603.16755` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Wei Min Loh, Sajib Kumer Sinha, Ankur Agarwal, et al.
+- **abs:** https://arxiv.org/abs/2603.16755 · **PDF:** https://arxiv.org/pdf/2603.16755v1
+
+## 4116. PI-Mamba: Linear-Time Protein Backbone Generation via Spectrally Initialized Flow Matching
+
+- **ID:** `2603.26705` · **primary:** `q-bio.BM` · **Submitted:** 2026-03-17
+- **Authors:** Tianyu Wu, Lin Zhu
+- **abs:** https://arxiv.org/abs/2603.26705 · **PDF:** https://arxiv.org/pdf/2603.26705v1
+
+## 4117. Finding Common Ground in a Sea of Alternatives
+
+- **ID:** `2603.16751` · **primary:** `cs.GT` · **Submitted:** 2026-03-17
+- **Authors:** Jay Chooi, Paul Gölz, Ariel D. Procaccia, et al.
+- **abs:** https://arxiv.org/abs/2603.16751 · **PDF:** https://arxiv.org/pdf/2603.16751v1
+
+## 4118. Formal verification of tree-based machine learning models for lateral spreading
+
+- **ID:** `2603.16983` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Krishna Kumar
+- **abs:** https://arxiv.org/abs/2603.16983 · **PDF:** https://arxiv.org/pdf/2603.16983v1
+
+## 4119. Probing Cultural Signals in Large Language Models through Author Profiling
+
+- **ID:** `2603.16749` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Valentin Lafargue, Ariel Guerra-Adames, Emmanuelle Claeys, et al.
+- **abs:** https://arxiv.org/abs/2603.16749 · **PDF:** https://arxiv.org/pdf/2603.16749v2
+
+## 4120. Data-driven forced response analysis with min-max representations of nonlinear restoring forces
+
+- **ID:** `2603.16746` · **primary:** `math.DS` · **Submitted:** 2026-03-17
+- **Authors:** Akira Saito, Hiromu Fujita
+- **abs:** https://arxiv.org/abs/2603.16746 · **PDF:** https://arxiv.org/pdf/2603.16746v1
+
+## 4121. Nonstandard Errors in AI Agents
+
+- **ID:** `2603.16744` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Ruijiang Gao, Steven Chong Xiao
+- **abs:** https://arxiv.org/abs/2603.16744 · **PDF:** https://arxiv.org/pdf/2603.16744v2
+
+## 4122. Bayesian Inference of Psychometric Variables From Brain and Behavior in Implicit Association Tests
+
+- **ID:** `2603.16741` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Christian A. Kothe, Sean Mullen, Michael V. Bronstein, et al.
+- **abs:** https://arxiv.org/abs/2603.16741 · **PDF:** https://arxiv.org/pdf/2603.16741v1
+
+## 4123. SpecMoE: Spectral Mixture-of-Experts Foundation Model for Cross-Species EEG Decoding
+
+- **ID:** `2603.16739` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Davy Darankoum, Chloé Habermacher, Julien Volle, et al.
+- **abs:** https://arxiv.org/abs/2603.16739 · **PDF:** https://arxiv.org/pdf/2603.16739v2
+
+## 4124. MedCL-Bench: Benchmarking stability-efficiency trade-offs and scaling in biomedical continual learning
+
+- **ID:** `2603.16738` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Min Zeng, Shuang Zhou, Zaifu Zhan, et al.
+- **abs:** https://arxiv.org/abs/2603.16738 · **PDF:** https://arxiv.org/pdf/2603.16738v1
+
+## 4125. Retrieving Counterfactuals Improves Visual In-Context Learning
+
+- **ID:** `2603.16737` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Guangzhi Xiong, Sanchit Sinha, Zhenghao He, et al.
+- **abs:** https://arxiv.org/abs/2603.16737 · **PDF:** https://arxiv.org/pdf/2603.16737v1
+
+## 4126. Differential Harm Propensity in Personalized LLM Agents: The Curious Case of Mental Health Disclosure
+
+- **ID:** `2603.16734` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Caglar Yildirim
+- **abs:** https://arxiv.org/abs/2603.16734 · **PDF:** https://arxiv.org/pdf/2603.16734v1
+
+## 4127. IQuest-Coder-V1 Technical Report
+
+- **ID:** `2603.16733` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Jian Yang, Wei Zhang, Shawn Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.16733 · **PDF:** https://arxiv.org/pdf/2603.16733v1
+
+## 4128. Understanding Quantization of Optimizer States in LLM Pre-training: Dynamics of State Staleness and Effectiveness of State Resets
+
+- **ID:** `2603.16731` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Kristi Topollai, Anna Choromanska
+- **abs:** https://arxiv.org/abs/2603.16731 · **PDF:** https://arxiv.org/pdf/2603.16731v1
+
+## 4129. GeMA: Learning Latent Manifold Frontiers for Benchmarking Complex Systems
+
+- **ID:** `2603.16729` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Jia Ming Li, Anupriya, Daniel J. Graham
+- **abs:** https://arxiv.org/abs/2603.16729 · **PDF:** https://arxiv.org/pdf/2603.16729v1
+
+## 4130. The Cost of Reasoning: Chain-of-Thought Induces Overconfidence in Vision-Language Models
+
+- **ID:** `2603.16728` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Robert Welch, Emir Konuk, Kevin Smith
+- **abs:** https://arxiv.org/abs/2603.16728 · **PDF:** https://arxiv.org/pdf/2603.16728v1
+
+## 4131. Deep Learning Multi-Horizon Irradiance Nowcasting: A Comparative Evaluation of Three Methods for Leveraging Sky Images
+
+- **ID:** `2603.26704` · **primary:** `eess.SY` · **Submitted:** 2026-03-17
+- **Authors:** Erling W. Eriksen, Magnus M. Nygård, Niklas Erdmann, et al.
+- **abs:** https://arxiv.org/abs/2603.26704 · **PDF:** https://arxiv.org/pdf/2603.26704v1
+
+## 4132. Federated Learning with Multi-Partner OneFlorida+ Consortium Data for Predicting Major Postoperative Complications
+
+- **ID:** `2603.16723` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Yuanfang Ren, Varun Sai Vemuri, Zhenhong Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.16723 · **PDF:** https://arxiv.org/pdf/2603.16723v1
+
+## 4133. Arabic Morphosyntactic Tagging and Dependency Parsing with Large Language Models
+
+- **ID:** `2603.16718` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Mohamed Adel, Bashar Alhafni, Nizar Habash
+- **abs:** https://arxiv.org/abs/2603.16718 · **PDF:** https://arxiv.org/pdf/2603.16718v1
+
+## 4134. Novelty-Driven Target-Space Discovery in Automated Electron and Scanning Probe Microscopy
+
+- **ID:** `2603.16715` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Utkarsh Pratiush, Kamyar Barakati, Boris N. Slautin, et al.
+- **abs:** https://arxiv.org/abs/2603.16715 · **PDF:** https://arxiv.org/pdf/2603.16715v1
+
+## 4135. High-dimensional estimation with missing data: Statistical and computational limits
+
+- **ID:** `2603.16712` · **primary:** `math.ST` · **Submitted:** 2026-03-17
+- **Authors:** Kabir Aladin Verchand, Ankit Pensia, Saminul Haque, et al.
+- **abs:** https://arxiv.org/abs/2603.16712 · **PDF:** https://arxiv.org/pdf/2603.16712v1
+
+## 4136. Learning Lineage-guided Geodesics with Finsler Geometry
+
+- **ID:** `2603.16708` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Aaron Zweig, Mingxuan Zhang, David A. Knowles, et al.
+- **abs:** https://arxiv.org/abs/2603.16708 · **PDF:** https://arxiv.org/pdf/2603.16708v1
+
+## 4137. DEAF: A Benchmark for Diagnostic Evaluation of Acoustic Faithfulness in Audio Language Models
+
+- **ID:** `2603.18048` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Jiaqi Xiong, Yunjia Qi, Qi Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.18048 · **PDF:** https://arxiv.org/pdf/2603.18048v2
+
+## 4138. Cost Trade-offs in Matrix Inversion Updates for Streaming Outlier Detection
+
+- **ID:** `2603.16697` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Florian Grivet, Louise Travé-Massuyès
+- **abs:** https://arxiv.org/abs/2603.16697 · **PDF:** https://arxiv.org/pdf/2603.16697v1
+
+## 4139. Grid-World Representations in Transformers Reflect Predictive Geometry
+
+- **ID:** `2603.16689` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Sasha Brenner, Thomas R. Knösche, Nico Scherf
+- **abs:** https://arxiv.org/abs/2603.16689 · **PDF:** https://arxiv.org/pdf/2603.16689v1
+
+## 4140. When Should a Robot Think? Resource-Aware Reasoning via Reinforcement Learning for Embodied Robotic Decision-Making
+
+- **ID:** `2603.16673` · **primary:** `cs.RO` · **Submitted:** 2026-03-17
+- **Authors:** Jun Liu, Pu Zhao, Zhenglun Kong, et al.
+- **abs:** https://arxiv.org/abs/2603.16673 · **PDF:** https://arxiv.org/pdf/2603.16673v3
+
+## 4141. CritiSense: Critical Digital Literacy and Resilience Against Misinformation
+
+- **ID:** `2603.16672` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Firoj Alam, Fatema Ahmad, Ali Ezzat Shahroor, et al.
+- **abs:** https://arxiv.org/abs/2603.16672 · **PDF:** https://arxiv.org/pdf/2603.16672v1
+
+## 4142. Target Concept Tuning Improves Extreme Weather Forecasting
+
+- **ID:** `2603.19325` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Shijie Ren, Xinyue Gu, Ziheng Peng, et al.
+- **abs:** https://arxiv.org/abs/2603.19325 · **PDF:** https://arxiv.org/pdf/2603.19325v1
+
+## 4143. Fast-WAM: Do World Action Models Need Test-time Future Imagination?
+
+- **ID:** `2603.16666` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Tianyuan Yuan, Zibin Dong, Yicheng Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.16666 · **PDF:** https://arxiv.org/pdf/2603.16666v2
+
+## 4144. Kestrel: Grounding Self-Refinement for LVLM Hallucination Mitigation
+
+- **ID:** `2603.16664` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Jiawei Mao, Hardy Chen, Haoqin Tu, et al.
+- **abs:** https://arxiv.org/abs/2603.16664 · **PDF:** https://arxiv.org/pdf/2603.16664v1
+
+## 4145. When Openclaw Agents Learn from Each Other: Insights from Emergent AI Agent Communities for Human-AI Partnership in Education
+
+- **ID:** `2603.16663` · **primary:** `cs.CY` · **Submitted:** 2026-03-17
+- **Authors:** Eason Chen, Ce Guan, Ahmed Elshafiey, et al.
+- **abs:** https://arxiv.org/abs/2603.16663 · **PDF:** https://arxiv.org/pdf/2603.16663v2
+
+## 4146. Self-Aware Markov Models for Discrete Reasoning
+
+- **ID:** `2603.16661` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Gregor Kornhardt, Jannis Chemseddine, Christian Wald, et al.
+- **abs:** https://arxiv.org/abs/2603.16661 · **PDF:** https://arxiv.org/pdf/2603.16661v2
+
+## 4147. Can Linguistically Related Languages Guide LLM Translation in Low-Resource Settings?
+
+- **ID:** `2603.16660` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Aishwarya Ramasethu, Niyathi Allu, Rohin Garg, et al.
+- **abs:** https://arxiv.org/abs/2603.16660 · **PDF:** https://arxiv.org/pdf/2603.16660v1
+
+## 4148. Machines acquire scientific taste from institutional traces
+
+- **ID:** `2603.16659` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Ziqin Gong, Ning Li, Huaikang Zhou
+- **abs:** https://arxiv.org/abs/2603.16659 · **PDF:** https://arxiv.org/pdf/2603.16659v1
+
+## 4149. Omanic: Towards Step-wise Evaluation of Multi-hop Reasoning in Large Language Models
+
+- **ID:** `2603.16654` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Xiaojie Gu, Sherry T. Tong, Aosong Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.16654 · **PDF:** https://arxiv.org/pdf/2603.16654v1
+
+## 4150. What if Pinocchio Were a Reinforcement Learning Agent: A Normative End-to-End Pipeline
+
+- **ID:** `2603.16651` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Benoît Alcaraz
+- **abs:** https://arxiv.org/abs/2603.16651 · **PDF:** https://arxiv.org/pdf/2603.16651v1
+
+## 4151. Domain-Independent Dynamic Programming with Constraint Propagation
+
+- **ID:** `2603.16648` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Imko Marijnissen, J. Christopher Beck, Emir Demirović, et al.
+- **abs:** https://arxiv.org/abs/2603.16648 · **PDF:** https://arxiv.org/pdf/2603.16648v1
+
+## 4152. Interpretable AI-Assisted Early Reliability Prediction for a Two-Parameter Parallel Root-Finding Scheme
+
+- **ID:** `2603.16980` · **primary:** `math.NA` · **Submitted:** 2026-03-17
+- **Authors:** Bruno Carpentieri, Andrei Velichko, Mudassir Shams, et al.
+- **abs:** https://arxiv.org/abs/2603.16980 · **PDF:** https://arxiv.org/pdf/2603.16980v1
+
+## 4153. Good Arguments Against the People Pleasers: How Reasoning Mitigates (Yet Masks) LLM Sycophancy
+
+- **ID:** `2603.16643` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Zhaoxin Feng, Zheng Chen, Jianfei Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.16643 · **PDF:** https://arxiv.org/pdf/2603.16643v1
+
+## 4154. When AI Navigates the Fog of War
+
+- **ID:** `2603.16642` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Ming Li, Xirui Li, Tianyi Zhou
+- **abs:** https://arxiv.org/abs/2603.16642 · **PDF:** https://arxiv.org/pdf/2603.16642v1
+
+## 4155. Understanding Pruning Regimes in Vision-Language Models Through Domain-Aware Layer Selection
+
+- **ID:** `2603.20275` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Saeed Khaki, Nima Safaei, Kamal Ginotra
+- **abs:** https://arxiv.org/abs/2603.20275 · **PDF:** https://arxiv.org/pdf/2603.20275v1
+
+## 4156. MLLM-based Textual Explanations for Face Comparison
+
+- **ID:** `2603.16629` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Redwan Sony, Anil K Jain, Arun Ross
+- **abs:** https://arxiv.org/abs/2603.16629 · **PDF:** https://arxiv.org/pdf/2603.16629v3
+
+## 4157. Domain Mixture Design via Log-Likelihood Differences for Aligning Language Models with a Target Model
+
+- **ID:** `2603.16622` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Ryo Kishino, Riku Shiomi, Hiroaki Yamagiwa, et al.
+- **abs:** https://arxiv.org/abs/2603.16622 · **PDF:** https://arxiv.org/pdf/2603.16622v1
+
+## 4158. Simplex-to-Euclidean Bijection for Conjugate and Calibrated Multiclass Gaussian Process
+
+- **ID:** `2603.16621` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Bernardo Williams, Harsha Vardhan Tetali, Arto Klami, et al.
+- **abs:** https://arxiv.org/abs/2603.16621 · **PDF:** https://arxiv.org/pdf/2603.16621v1
+
+## 4159. Sample Transform Cost-Based Training-Free Hallucination Detector for Large Language Models
+
+- **ID:** `2603.22303` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Zeyang Ding, Xinglin Hu, Jicong Fan
+- **abs:** https://arxiv.org/abs/2603.22303 · **PDF:** https://arxiv.org/pdf/2603.22303v1
+
+## 4160. Omnilingual SONAR: Cross-Lingual and Cross-Modal Sentence Embeddings Bridging Massively Multilingual Text and Speech
+
+- **ID:** `2603.16606` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Omnilingual SONAR Team, João Maria Janeiro, Pere-Lluís Huguet Cabot, et al.
+- **abs:** https://arxiv.org/abs/2603.16606 · **PDF:** https://arxiv.org/pdf/2603.16606v2
+
+## 4161. Tarab: A Multi-Dialect Corpus of Arabic Lyrics and Poetry
+
+- **ID:** `2603.16601` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Mo El-Haj
+- **abs:** https://arxiv.org/abs/2603.16601 · **PDF:** https://arxiv.org/pdf/2603.16601v1
+
+## 4162. Data-driven generalized perimeter control: Zürich case study
+
+- **ID:** `2603.16599` · **primary:** `eess.SY` · **Submitted:** 2026-03-17
+- **Authors:** Alessio Rimoldi, Carlo Cenedese, Alberto Padoan, et al.
+- **abs:** https://arxiv.org/abs/2603.16599 · **PDF:** https://arxiv.org/pdf/2603.16599v1
+
+## 4163. FSMC-Pose: Frequency and Spatial Fusion with Multiscale Self-calibration for Cattle Mounting Pose Estimation
+
+- **ID:** `2603.16596` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Fangjing Li, Zhihai Wang, Xinxin Ding, et al.
+- **abs:** https://arxiv.org/abs/2603.16596 · **PDF:** https://arxiv.org/pdf/2603.16596v1
+
+## 4164. BATQuant: Outlier-resilient MXFP4 Quantization via Learnable Block-wise Optimization
+
+- **ID:** `2603.16590` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Ji-Fu Li, Manyi Zhang, Xiaobo Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.16590 · **PDF:** https://arxiv.org/pdf/2603.16590v1
+
+## 4165. Runtime Governance for AI Agents: Policies on Paths
+
+- **ID:** `2603.16586` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Maurits Kaptein, Vassilis-Javed Khan, Andriy Podstavnychy
+- **abs:** https://arxiv.org/abs/2603.16586 · **PDF:** https://arxiv.org/pdf/2603.16586v1
+
+## 4166. Trajectory-Optimized Time Reparameterization for Learning-Compatible Reduced-Order Modeling of Stiff Dynamical Systems
+
+- **ID:** `2603.16583` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Joe Standridge, Daniel Livescu, Paul Cizmas
+- **abs:** https://arxiv.org/abs/2603.16583 · **PDF:** https://arxiv.org/pdf/2603.16583v2
+
+## 4167. V-DyKnow: A Dynamic Benchmark for Time-Sensitive Knowledge in Vision Language Models
+
+- **ID:** `2603.16581` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Seyed Mahed Mousavi, Christian Moiola, Massimo Rizzoli, et al.
+- **abs:** https://arxiv.org/abs/2603.16581 · **PDF:** https://arxiv.org/pdf/2603.16581v1
+
+## 4168. When and Why Does Unsupervised RL Succeed in Mathematical Reasoning? A Manifold Envelopment Perspective
+
+- **ID:** `2603.16578` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Zelin Zhang, Fei Cheng, Chenhui Chu
+- **abs:** https://arxiv.org/abs/2603.16578 · **PDF:** https://arxiv.org/pdf/2603.16578v1
+
+## 4169. REFORGE: Multi-modal Attacks Reveal Vulnerable Concept Unlearning in Image Generation Models
+
+- **ID:** `2603.16576` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Yong Zou, Haoran Li, Fanxiao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.16576 · **PDF:** https://arxiv.org/pdf/2603.16576v1
+
+## 4170. Diverging Transformer Predictions for Human Sentence Processing: A Comprehensive Analysis of Agreement Attraction Effects
+
+- **ID:** `2603.16574` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Titus von der Malsburg, Sebastian Padó
+- **abs:** https://arxiv.org/abs/2603.16574 · **PDF:** https://arxiv.org/pdf/2603.16574v1
+
+## 4171. Malicious Or Not: Adding Repository Context to Agent Skill Classification
+
+- **ID:** `2603.16572` · **primary:** `cs.CR` · **Submitted:** 2026-03-17
+- **Authors:** Florian Holzbauer, David Schmidt, Gabriel Gegenhuber, et al.
+- **abs:** https://arxiv.org/abs/2603.16572 · **PDF:** https://arxiv.org/pdf/2603.16572v1
+
+## 4172. Deep Tabular Representation Corrector
+
+- **ID:** `2603.16569` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Hangting Ye, Peng Wang, Wei Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.16569 · **PDF:** https://arxiv.org/pdf/2603.16569v1
+
+## 4173. Manifold-Matching Autoencoders
+
+- **ID:** `2603.16568` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Laurent Cheret, Vincent Létourneau, Isar Nejadgholi, et al.
+- **abs:** https://arxiv.org/abs/2603.16568 · **PDF:** https://arxiv.org/pdf/2603.16568v1
+
+## 4174. Characterizing Delusional Spirals through Human-LLM Chat Logs
+
+- **ID:** `2603.16567` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Jared Moore, Ashish Mehta, William Agnew, et al.
+- **abs:** https://arxiv.org/abs/2603.16567 · **PDF:** https://arxiv.org/pdf/2603.16567v1
+
+## 4175. Deep Learning-Driven Black-Box Doherty Power Amplifier with Pixelated Output Combiner and Extended Efficiency Range
+
+- **ID:** `2603.16565` · **primary:** `eess.SP` · **Submitted:** 2026-03-17
+- **Authors:** Han Zhou, Haojie Chang, David Widen
+- **abs:** https://arxiv.org/abs/2603.16565 · **PDF:** https://arxiv.org/pdf/2603.16565v1
+
+## 4176. BenchPreS: A Benchmark for Context-Aware Personalized Preference Selectivity of Persistent-Memory LLMs
+
+- **ID:** `2603.16557` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Sangyeon Yoon, Sunkyoung Kim, Hyesoo Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.16557 · **PDF:** https://arxiv.org/pdf/2603.16557v1
+
+## 4177. EmoLLM: Appraisal-Grounded Cognitive-Emotional Co-Reasoning in Large Language Models
+
+- **ID:** `2603.16553` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Yifei Zhang, Mingyang Li, Henry Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.16553 · **PDF:** https://arxiv.org/pdf/2603.16553v1
+
+## 4178. CompDiff: Hierarchical Compositional Diffusion for Fair and Zero-Shot Intersectional Medical Image Generation
+
+- **ID:** `2603.16551` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Mahmoud Ibrahim, Bart Elen, Chang Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.16551 · **PDF:** https://arxiv.org/pdf/2603.16551v1
+
+## 4179. Bridging the Simulation-to-Reality Gap in Electron Microscope Calibration via VAE-EM Estimation
+
+- **ID:** `2603.16549` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Jilles S. van Hulst, W. P. M. H. Heemels, Duarte J. Antunes
+- **abs:** https://arxiv.org/abs/2603.16549 · **PDF:** https://arxiv.org/pdf/2603.16549v2
+
+## 4180. Solomonoff induction
+
+- **ID:** `2603.20274` · **primary:** `cs.FL` · **Submitted:** 2026-03-17
+- **Authors:** Tom F. Sterkenburg
+- **abs:** https://arxiv.org/abs/2603.20274 · **PDF:** https://arxiv.org/pdf/2603.20274v1
+
+## 4181. DanceHA: A Multi-Agent Framework for Document-Level Aspect-Based Sentiment Analysis
+
+- **ID:** `2603.16546` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Lei Wang, Min Huang, Eduard Dragut
+- **abs:** https://arxiv.org/abs/2603.16546 · **PDF:** https://arxiv.org/pdf/2603.16546v1
+
+## 4182. How often do Answers Change? Estimating Recency Requirements in Question Answering
+
+- **ID:** `2603.16544` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Bhawna Piryani, Zehra Mert, Adam Jatowt
+- **abs:** https://arxiv.org/abs/2603.16544 · **PDF:** https://arxiv.org/pdf/2603.16544v1
+
+## 4183. Designing for Disagreement: Front-End Guardrails for Assistance Allocation in LLM-Enabled Robots
+
+- **ID:** `2603.16537` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Carmen Ng
+- **abs:** https://arxiv.org/abs/2603.16537 · **PDF:** https://arxiv.org/pdf/2603.16537v1
+
+## 4184. SympFormer: Accelerated attention blocks via Inertial Dynamics on Density Manifolds
+
+- **ID:** `2603.16535` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Viktor Stein, Wuchen Li, Gabriele Steidl
+- **abs:** https://arxiv.org/abs/2603.16535 · **PDF:** https://arxiv.org/pdf/2603.16535v1
+
+## 4185. Research on Individual Trait Clustering and Development Pathway Adaptation Based on the K-means Algorithm
+
+- **ID:** `2603.22302` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Qianru Wei, Jihaoyu Yang, Cheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.22302 · **PDF:** https://arxiv.org/pdf/2603.22302v1
+
+## 4186. Exploring different approaches to customize language models for domain-specific text-to-code generation
+
+- **ID:** `2603.16526` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Luís Freire, Fernanda A. Andaló, Nicki Skafte Detlefsen
+- **abs:** https://arxiv.org/abs/2603.16526 · **PDF:** https://arxiv.org/pdf/2603.16526v1
+
+## 4187. An approximate graph elicits detonation lattice
+
+- **ID:** `2603.16524` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Vansh Sharma, Venkat Raman
+- **abs:** https://arxiv.org/abs/2603.16524 · **PDF:** https://arxiv.org/pdf/2603.16524v1
+
+## 4188. FEAT: A Linear-Complexity Foundation Model for Extremely Large Structured Data
+
+- **ID:** `2603.16513` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Zhenghang Song, Tang Qian, Lu Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.16513 · **PDF:** https://arxiv.org/pdf/2603.16513v2
+
+## 4189. Rewarding DINO: Predicting Dense Rewards with Vision Foundation Models
+
+- **ID:** `2603.16978` · **primary:** `cs.RO` · **Submitted:** 2026-03-17
+- **Authors:** Pierre Krack, Tobias Jülg, Wolfram Burgard, et al.
+- **abs:** https://arxiv.org/abs/2603.16978 · **PDF:** https://arxiv.org/pdf/2603.16978v1
+
+## 4190. From the Inside Out: Progressive Distribution Refinement for Confidence Calibration
+
+- **ID:** `2603.16500` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Xizhong Yang, Yinan Xia, Huiming Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.16500 · **PDF:** https://arxiv.org/pdf/2603.16500v1
+
+## 4191. Bridging the High-Frequency Data Gap: A Millisecond-Resolution Network Dataset for Advancing Time Series Foundation Models
+
+- **ID:** `2603.16497` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Subina Khanal, Seshu Tirupathi, Merim Dzaferagic, et al.
+- **abs:** https://arxiv.org/abs/2603.16497 · **PDF:** https://arxiv.org/pdf/2603.16497v1
+
+## 4192. AdaMem: Adaptive User-Centric Memory for Long-Horizon Dialogue Agents
+
+- **ID:** `2603.16496` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Shannan Yan, Jingchen Ni, Leqi Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.16496 · **PDF:** https://arxiv.org/pdf/2603.16496v1
+
+## 4193. ExpressMind: A Multimodal Pretrained Large Language Model for Expressway Operation
+
+- **ID:** `2603.16495` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Zihe Wang, Yihuan Wang, Haiyang Yu. Zhiyong Cui, et al.
+- **abs:** https://arxiv.org/abs/2603.16495 · **PDF:** https://arxiv.org/pdf/2603.16495v1
+
+## 4194. Unlearning for One-Step Generative Models via Unbalanced Optimal Transport
+
+- **ID:** `2603.16489` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Hyundo Choi, Junhyeong An, Jinseong Park, et al.
+- **abs:** https://arxiv.org/abs/2603.16489 · **PDF:** https://arxiv.org/pdf/2603.16489v1
+
+## 4195. Implementation of tangent linear and adjoint models for neural networks based on a compiler library tool
+
+- **ID:** `2603.16976` · **primary:** `cs.MS` · **Submitted:** 2026-03-17
+- **Authors:** Sa Xiao, Hao Jing, Honglu Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.16976 · **PDF:** https://arxiv.org/pdf/2603.16976v1
+
+## 4196. On the Emotion Understanding of Synthesized Speech
+
+- **ID:** `2603.16483` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Yuan Ge, Haishu Zhao, Aokai Hao, et al.
+- **abs:** https://arxiv.org/abs/2603.16483 · **PDF:** https://arxiv.org/pdf/2603.16483v1
+
+## 4197. When GPUs Fail Quietly: Observability-Aware Early Warning Beyond Numeric Telemetry
+
+- **ID:** `2603.28781` · **primary:** `cs.DC` · **Submitted:** 2026-03-17
+- **Authors:** Michael Bidollahkhani, Freja Nordsiek, Julian M. Kunkel
+- **abs:** https://arxiv.org/abs/2603.28781 · **PDF:** https://arxiv.org/pdf/2603.28781v1
+
+## 4198. DST-Net: A Dual-Stream Transformer with Illumination-Independent Feature Guidance and Multi-Scale Spatial Convolution for Low-Light Image Enhancement
+
+- **ID:** `2603.16482` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Yicui Shi, Yuhan Chen, Xiangfei Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.16482 · **PDF:** https://arxiv.org/pdf/2603.16482v1
+
+## 4199. Optimal uncertainty bounds for multivariate kernel regression under bounded noise: A Gaussian process-based dual function
+
+- **ID:** `2603.16481` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Amon Lahr, Anna Scampicchio, Johannes Köhler, et al.
+- **abs:** https://arxiv.org/abs/2603.16481 · **PDF:** https://arxiv.org/pdf/2603.16481v1
+
+## 4200. Latent Semantic Manifolds in Large Language Models
+
+- **ID:** `2603.22301` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Mohamed A. Mabrok
+- **abs:** https://arxiv.org/abs/2603.22301 · **PDF:** https://arxiv.org/pdf/2603.22301v1
+
+## 4201. Breaking the Chain: A Causal Analysis of LLM Faithfulness to Intermediate Structures
+
+- **ID:** `2603.16475` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Oleg Somov, Mikhail Chaichuk, Mikhail Seleznyov, et al.
+- **abs:** https://arxiv.org/abs/2603.16475 · **PDF:** https://arxiv.org/pdf/2603.16475v1
+
+## 4202. The State of Generative AI in Software Development: Insights from Literature and a Developer Survey
+
+- **ID:** `2603.16975` · **primary:** `cs.SE` · **Submitted:** 2026-03-17
+- **Authors:** Vincent Gurgul, Robin Gubela, Stefan Lessmann
+- **abs:** https://arxiv.org/abs/2603.16975 · **PDF:** https://arxiv.org/pdf/2603.16975v1
+
+## 4203. Multi-Agent Reinforcement Learning Counteracts Delayed CSI in Multi-Satellite Systems
+
+- **ID:** `2603.16470` · **primary:** `cs.IT` · **Submitted:** 2026-03-17
+- **Authors:** Marios Aristodemou, Yasaman Omid, Sangarapillai Lambotharan, et al.
+- **abs:** https://arxiv.org/abs/2603.16470 · **PDF:** https://arxiv.org/pdf/2603.16470v1
+
+## 4204. Follow the Clues, Frame the Truth: Hybrid-evidential Deductive Reasoning in Open-Vocabulary Multimodal Emotion Recognition
+
+- **ID:** `2603.16463` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Yu Liu, Lei Zhang, Haoxun Li, et al.
+- **abs:** https://arxiv.org/abs/2603.16463 · **PDF:** https://arxiv.org/pdf/2603.16463v1
+
+## 4205. Linearized Bregman Iterations for Sparse Spiking Neural Networks
+
+- **ID:** `2603.16462` · **primary:** `eess.SP` · **Submitted:** 2026-03-17
+- **Authors:** Daniel Windhager, Bernhard A. Moser, Michael Lunglmayr
+- **abs:** https://arxiv.org/abs/2603.16462 · **PDF:** https://arxiv.org/pdf/2603.16462v1
+
+## 4206. Are a Thousand Words Better Than a Single Picture? Beyond Images -- A Framework for Multi-Modal Knowledge Graph Dataset Enrichment
+
+- **ID:** `2603.16974` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Pengyu Zhang, Klim Zaporojets, Jie Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.16974 · **PDF:** https://arxiv.org/pdf/2603.16974v1
+
+## 4207. DynHD: Hallucination Detection for Diffusion Large Language Models via Denoising Dynamics Deviation Learning
+
+- **ID:** `2603.16459` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Yanyu Qian, Yue Tan, Yixin Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.16459 · **PDF:** https://arxiv.org/pdf/2603.16459v1
+
+## 4208. RetailBench: Evaluating Long-Horizon Autonomous Decision-Making and Strategy Stability of LLM Agents in Realistic Retail Environments
+
+- **ID:** `2603.16453` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Linghua Zhang, Jun Wang, Jingtong Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.16453 · **PDF:** https://arxiv.org/pdf/2603.16453v1
+
+## 4209. TRUST-SQL: Tool-Integrated Multi-Turn Reinforcement Learning for Text-to-SQL over Unknown Schemas
+
+- **ID:** `2603.16448` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Ai Jian, Xiaoyun Zhang, Wanrou Du, et al.
+- **abs:** https://arxiv.org/abs/2603.16448 · **PDF:** https://arxiv.org/pdf/2603.16448v2
+
+## 4210. Visual Distraction Undermines Moral Reasoning in Vision-Language Models
+
+- **ID:** `2603.16445` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Xinyi Yang, Chenheng Xu, Weijun Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.16445 · **PDF:** https://arxiv.org/pdf/2603.16445v1
+
+## 4211. Hybrid Classical-Quantum Transfer Learning with Noisy Quantum Circuits
+
+- **ID:** `2603.16973` · **primary:** `quant-ph` · **Submitted:** 2026-03-17
+- **Authors:** D. Martín-Pérez, F. Rodríguez-Díaz, D. Gutiérrez-Avilés, et al.
+- **abs:** https://arxiv.org/abs/2603.16973 · **PDF:** https://arxiv.org/pdf/2603.16973v1
+
+## 4212. Capability-Guided Compression: Toward Interpretability-Aware Budget Allocation for Large Language Models
+
+- **ID:** `2603.16440` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Rishaank Gupta
+- **abs:** https://arxiv.org/abs/2603.16440 · **PDF:** https://arxiv.org/pdf/2603.16440v1
+
+## 4213. CD-FKD: Cross-Domain Feature Knowledge Distillation for Robust Single-Domain Generalization in Object Detection
+
+- **ID:** `2603.16439` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Junseok Lee, Sungho Shin, Seongju Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.16439 · **PDF:** https://arxiv.org/pdf/2603.16439v1
+
+## 4214. DISCOVER: A Solver for Distributional Counterfactual Explanations
+
+- **ID:** `2603.16436` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Yikai Gu, Lele Cao, Bo Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.16436 · **PDF:** https://arxiv.org/pdf/2603.16436v1
+
+## 4215. VQKV: High-Fidelity and High-Ratio Cache Compression via Vector-Quantization
+
+- **ID:** `2603.16435` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Yixuan Wang, Qingyu Shi, Jiayu Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.16435 · **PDF:** https://arxiv.org/pdf/2603.16435v1
+
+## 4216. From Natural Language to Executable Option Strategies via Large Language Models
+
+- **ID:** `2603.16434` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Haochen Luo, Zhengzhao Lai, Junjie Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.16434 · **PDF:** https://arxiv.org/pdf/2603.16434v1
+
+## 4217. IRIS: A Real-World Benchmark for Inverse Recovery and Identification of Physical Dynamic Systems from Monocular Video
+
+- **ID:** `2603.16432` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Rasul Khanbayov, Mohamed Rayan Barhdadi, Erchin Serpedin, et al.
+- **abs:** https://arxiv.org/abs/2603.16432 · **PDF:** https://arxiv.org/pdf/2603.16432v2
+
+## 4218. EngGPT2: Sovereign, Efficient and Open Intelligence
+
+- **ID:** `2603.16430` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** G. Ciarfaglia, A. Rosanova, S. Cipolla, et al.
+- **abs:** https://arxiv.org/abs/2603.16430 · **PDF:** https://arxiv.org/pdf/2603.16430v3
+
+## 4219. LenghuSky-8: An 8-Year All-Sky Cloud Dataset with Star-Aware Masks and Alt-Az Calibration for Segmentation and Nowcasting
+
+- **ID:** `2603.16429` · **primary:** `astro-ph.IM` · **Submitted:** 2026-03-17
+- **Authors:** Yicheng Rui, Xiao-Wei Duan, Licai Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.16429 · **PDF:** https://arxiv.org/pdf/2603.16429v1
+
+## 4220. An Efficient Heterogeneous Co-Design for Fine-Tuning on a Single GPU
+
+- **ID:** `2603.16428` · **primary:** `cs.DC` · **Submitted:** 2026-03-17
+- **Authors:** Ruijia Yang, Zeyi Wen
+- **abs:** https://arxiv.org/abs/2603.16428 · **PDF:** https://arxiv.org/pdf/2603.16428v1
+
+## 4221. SF-Mamba: Rethinking State Space Model for Vision
+
+- **ID:** `2603.16423` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Masakazu Yoshimura, Teruaki Hayashi, Yuki Hoshino, et al.
+- **abs:** https://arxiv.org/abs/2603.16423 · **PDF:** https://arxiv.org/pdf/2603.16423v1
+
+## 4222. Via Negativa for AI Alignment: Why Negative Constraints Are Structurally Superior to Positive Preferences
+
+- **ID:** `2603.16417` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Quan Cheng
+- **abs:** https://arxiv.org/abs/2603.16417 · **PDF:** https://arxiv.org/pdf/2603.16417v1
+
+## 4223. IndexRAG: Bridging Facts for Cross-Document Reasoning at Index Time
+
+- **ID:** `2603.16415` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Zhenghua Bao, Yi Shi
+- **abs:** https://arxiv.org/abs/2603.16415 · **PDF:** https://arxiv.org/pdf/2603.16415v1
+
+## 4224. Trained Persistent Memory for Frozen Encoder--Decoder LLMs: Six Architectural Methods
+
+- **ID:** `2603.16413` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Hong Jeong
+- **abs:** https://arxiv.org/abs/2603.16413 · **PDF:** https://arxiv.org/pdf/2603.16413v1
+
+## 4225. RECOVER: Robust Entity Correction via agentic Orchestration of hypothesis Variants for Evidence-based Recovery
+
+- **ID:** `2603.16411` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Abhishek Kumar, Aashraya Sachdeva
+- **abs:** https://arxiv.org/abs/2603.16411 · **PDF:** https://arxiv.org/pdf/2603.16411v1
+
+## 4226. PlotTwist: A Creative Plot Generation Framework with Small Language Models
+
+- **ID:** `2603.16410` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Abhinav Thorat, Ravi Kolla, Jyotin Goel, et al.
+- **abs:** https://arxiv.org/abs/2603.16410 · **PDF:** https://arxiv.org/pdf/2603.16410v1
+
+## 4227. Who Benchmarks the Benchmarks? A Case Study of LLM Evaluation in Icelandic
+
+- **ID:** `2603.16406` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Finnur Ágúst Ingimundarson, Steinunn Rut Friðriksdóttir, Bjarki Ármannsson, et al.
+- **abs:** https://arxiv.org/abs/2603.16406 · **PDF:** https://arxiv.org/pdf/2603.16406v1
+
+## 4228. Over-the-air White-box Attack on the Wav2Vec Speech Recognition Neural Network
+
+- **ID:** `2603.16972` · **primary:** `eess.AS` · **Submitted:** 2026-03-17
+- **Authors:** Protopopov Alexey
+- **abs:** https://arxiv.org/abs/2603.16972 · **PDF:** https://arxiv.org/pdf/2603.16972v1
+
+## 4229. Deep Reinforcement Learning-Assisted Automated Operator Portfolio for Constrained Multi-objective Optimization
+
+- **ID:** `2603.16401` · **primary:** `cs.NE` · **Submitted:** 2026-03-17
+- **Authors:** Shuai Shao, Ye Tian, Shangshang Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.16401 · **PDF:** https://arxiv.org/pdf/2603.16401v1
+
+## 4230. Fanar 2.0: Arabic Generative AI Stack
+
+- **ID:** `2603.16397` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** FANAR TEAM, Ummar Abbas, Mohammad Shahmeer Ahmad, et al.
+- **abs:** https://arxiv.org/abs/2603.16397 · **PDF:** https://arxiv.org/pdf/2603.16397v1
+
+## 4231. Robust Physics-Guided Diffusion for Full-Waveform Inversion
+
+- **ID:** `2603.16393` · **primary:** `math.NA` · **Submitted:** 2026-03-17
+- **Authors:** Jishen Peng, Enze Jiang, Zheng Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.16393 · **PDF:** https://arxiv.org/pdf/2603.16393v1
+
+## 4232. Controlling Fish Schools via Reinforcement Learning of Virtual Fish Movement
+
+- **ID:** `2603.16384` · **primary:** `cs.RO` · **Submitted:** 2026-03-17
+- **Authors:** Yusuke Nishii, Hiroaki Kawashima
+- **abs:** https://arxiv.org/abs/2603.16384 · **PDF:** https://arxiv.org/pdf/2603.16384v1
+
+## 4233. Age Predictors Through the Lens of Generalization, Bias Mitigation, and Interpretability: Reflections on Causal Implications
+
+- **ID:** `2603.16377` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Debdas Paul, Elisa Ferrari, Irene Gravili, et al.
+- **abs:** https://arxiv.org/abs/2603.16377 · **PDF:** https://arxiv.org/pdf/2603.16377v1
+
+## 4234. Prior-Informed Neural Network Initialization: A Spectral Approach for Function Parameterizing Architectures
+
+- **ID:** `2603.16376` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** David Orlando Salazar Torres, Diyar Altinses, Andreas Schwung
+- **abs:** https://arxiv.org/abs/2603.16376 · **PDF:** https://arxiv.org/pdf/2603.16376v1
+
+## 4235. Continual Multimodal Egocentric Activity Recognition via Modality-Aware Novel Detection
+
+- **ID:** `2603.16970` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Wonseon Lim, Hyejeong Im, Dae-Won Kim
+- **abs:** https://arxiv.org/abs/2603.16970 · **PDF:** https://arxiv.org/pdf/2603.16970v1
+
+## 4236. FederatedFactory: Generative One-Shot Learning for Extremely Non-IID Distributed Scenarios
+
+- **ID:** `2603.16370` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Andrea Moleri, Christian Internò, Ali Raza, et al.
+- **abs:** https://arxiv.org/abs/2603.16370 · **PDF:** https://arxiv.org/pdf/2603.16370v1
+
+## 4237. Encoding Predictability and Legibility for Style-Conditioned Diffusion Policy
+
+- **ID:** `2603.16368` · **primary:** `cs.RO` · **Submitted:** 2026-03-17
+- **Authors:** Adrien Jacquet Crétides, Mouad Abrini, Hamed Rahimi, et al.
+- **abs:** https://arxiv.org/abs/2603.16368 · **PDF:** https://arxiv.org/pdf/2603.16368v1
+
+## 4238. DynamicGate MLP Conditional Computation via Learned Structural Dropout and Input Dependent Gating for Functional Plasticity
+
+- **ID:** `2603.16367` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Yong Il Choi
+- **abs:** https://arxiv.org/abs/2603.16367 · **PDF:** https://arxiv.org/pdf/2603.16367v1
+
+## 4239. FactorEngine: A Program-level Knowledge-Infused Factor Mining Framework for Quantitative Investment
+
+- **ID:** `2603.16365` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Qinhong Lin, Ruitao Feng, Yinglun Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.16365 · **PDF:** https://arxiv.org/pdf/2603.16365v1
+
+## 4240. $D^3$-RSMDE: 40$\times$ Faster and High-Fidelity Remote Sensing Monocular Depth Estimation
+
+- **ID:** `2603.16362` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Ruizhi Wang, Weihan Li, Zunlei Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.16362 · **PDF:** https://arxiv.org/pdf/2603.16362v1
+
+## 4241. Toward Experimentation-as-a-Service in 5G/6G: The Plaza6G Prototype for AI-Assisted Trials
+
+- **ID:** `2603.16356` · **primary:** `cs.NI` · **Submitted:** 2026-03-17
+- **Authors:** Sergio Barrachina-Muñoz, Marc Carrascosa-Zamacois, Horacio Bleda, et al.
+- **abs:** https://arxiv.org/abs/2603.16356 · **PDF:** https://arxiv.org/pdf/2603.16356v1
+
+## 4242. PashtoCorp: A 1.25-Billion-Word Corpus, Evaluation Suite, and Reproducible Pipeline for Low-Resource Language Development
+
+- **ID:** `2603.16354` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Hanif Rahman
+- **abs:** https://arxiv.org/abs/2603.16354 · **PDF:** https://arxiv.org/pdf/2603.16354v1
+
+## 4243. Automated identification of Ichneumonoidea wasps via YOLO-based deep learning: Integrating HiresCam for Explainable AI
+
+- **ID:** `2603.16351` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Joao Manoel Herrera Pinheiro, Gabriela Do Nascimento Herrera, Alvaro Doria Dos Santos, et al.
+- **abs:** https://arxiv.org/abs/2603.16351 · **PDF:** https://arxiv.org/pdf/2603.16351v1
+
+## 4244. Explainable machine learning workflows for radio astronomical data processing
+
+- **ID:** `2603.16350` · **primary:** `astro-ph.IM` · **Submitted:** 2026-03-17
+- **Authors:** S. Yatawatta, A. Ahmadi, B. Asabere, et al.
+- **abs:** https://arxiv.org/abs/2603.16350 · **PDF:** https://arxiv.org/pdf/2603.16350v1
+
+## 4245. Byzantine-Robust and Communication-Efficient Distributed Training: Compressive and Cyclic Gradient Coding
+
+- **ID:** `2603.28780` · **primary:** `cs.DC` · **Submitted:** 2026-03-17
+- **Authors:** Chengxi Li, Youssef Allouah, Rachid Guerraoui, et al.
+- **abs:** https://arxiv.org/abs/2603.28780 · **PDF:** https://arxiv.org/pdf/2603.28780v1
+
+## 4246. Deep learning based intelligent IDS for Large-scale IoT networks
+
+- **ID:** `2603.16342` · **primary:** `cs.CR` · **Submitted:** 2026-03-17
+- **Authors:** Isha Andrade, Shalaka S Mahadik, Mithun Mukherjee, et al.
+- **abs:** https://arxiv.org/abs/2603.16342 · **PDF:** https://arxiv.org/pdf/2603.16342v2
+
+## 4247. Behavioral Steering in a 35B MoE Language Model via SAE-Decoded Probe Vectors: One Agency Axis, Not Five Traits
+
+- **ID:** `2603.16335` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Jia Qing Yap
+- **abs:** https://arxiv.org/abs/2603.16335 · **PDF:** https://arxiv.org/pdf/2603.16335v1
+
+## 4248. Decoding the Critique Mechanism in Large Reasoning Models
+
+- **ID:** `2603.16331` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Hoang Phan, Quang H. Nguyen, Hung T. Q. Le, et al.
+- **abs:** https://arxiv.org/abs/2603.16331 · **PDF:** https://arxiv.org/pdf/2603.16331v1
+
+## 4249. An Interpretable Machine Learning Framework for Non-Small Cell Lung Cancer Drug Response Analysis
+
+- **ID:** `2603.16330` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Ann Rachel, Pranav M Pawar, Mithun Mukharjee, et al.
+- **abs:** https://arxiv.org/abs/2603.16330 · **PDF:** https://arxiv.org/pdf/2603.16330v1
+
+## 4250. A Human-Centred Architecture for Large Language Models-Cognitive Assistants in Manufacturing within Quality Management Systems
+
+- **ID:** `2603.16325` · **primary:** `cs.SE` · **Submitted:** 2026-03-17
+- **Authors:** Marcos Galdino, Johanna Grahl, Tobias Hamann, et al.
+- **abs:** https://arxiv.org/abs/2603.16325 · **PDF:** https://arxiv.org/pdf/2603.16325v1
+
+## 4251. DeepStage: Learning Autonomous Defense Policies Against Multi-Stage APT Campaigns
+
+- **ID:** `2603.16969` · **primary:** `cs.CR` · **Submitted:** 2026-03-17
+- **Authors:** Trung V. Phan, Tri Gia Nguyen, Thomas Bauschert
+- **abs:** https://arxiv.org/abs/2603.16969 · **PDF:** https://arxiv.org/pdf/2603.16969v1
+
+## 4252. Learning to Predict, Discover, and Reason in High-Dimensional Event Sequences
+
+- **ID:** `2603.16313` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Hugo Math
+- **abs:** https://arxiv.org/abs/2603.16313 · **PDF:** https://arxiv.org/pdf/2603.16313v2
+
+## 4253. Omnilingual MT: Machine Translation for 1,600 Languages
+
+- **ID:** `2603.16309` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Omnilingual MT Team, Belen Alastruey, Niyati Bafna, et al.
+- **abs:** https://arxiv.org/abs/2603.16309 · **PDF:** https://arxiv.org/pdf/2603.16309v2
+
+## 4254. NeSy-Route: A Neuro-Symbolic Benchmark for Constrained Route Planning in Remote Sensing
+
+- **ID:** `2603.16307` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Ming Yang, Zhi Zhou, Shi-Yu Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.16307 · **PDF:** https://arxiv.org/pdf/2603.16307v1
+
+## 4255. PyPhonPlan: Simulating phonetic planning with dynamic neural fields and task dynamics
+
+- **ID:** `2603.16299` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Sam Kirkham
+- **abs:** https://arxiv.org/abs/2603.16299 · **PDF:** https://arxiv.org/pdf/2603.16299v1
+
+## 4256. Attention-guided Evidence Grounding for Spoken Question Answering
+
+- **ID:** `2603.16292` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Ke Yang, Bolin Chen, Yuejie Li, et al.
+- **abs:** https://arxiv.org/abs/2603.16292 · **PDF:** https://arxiv.org/pdf/2603.16292v2
+
+## 4257. VisBrowse-Bench: Benchmarking Visual-Native Search for Multimodal Browsing Agents
+
+- **ID:** `2603.16289` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Zhengbo Zhang, Jinbo Su, Zhaowen Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.16289 · **PDF:** https://arxiv.org/pdf/2603.16289v2
+
+## 4258. Surrogate-Assisted Genetic Programming with Rank-Based Phenotypic Characterisation for Dynamic Multi-Mode Project Scheduling
+
+- **ID:** `2603.16286` · **primary:** `cs.NE` · **Submitted:** 2026-03-17
+- **Authors:** Yuan Tian, Yi Mei, Mengjie Zhang
+- **abs:** https://arxiv.org/abs/2603.16286 · **PDF:** https://arxiv.org/pdf/2603.16286v1
+
+## 4259. Locate-then-Sparsify: Attribution Guided Sparse Strategy for Visual Hallucination Mitigation
+
+- **ID:** `2603.16284` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** TianTian Dang, Chao Bi, Shufan Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.16284 · **PDF:** https://arxiv.org/pdf/2603.16284v1
+
+## 4260. Laya: A LeJEPA Approach to EEG via Latent Prediction over Reconstruction
+
+- **ID:** `2603.16281` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Saarang Panchavati, Uddhav Panchavati, Corey Arnold, et al.
+- **abs:** https://arxiv.org/abs/2603.16281 · **PDF:** https://arxiv.org/pdf/2603.16281v1
+
+## 4261. Agile Interception of a Flying Target using Competitive Reinforcement Learning
+
+- **ID:** `2603.16279` · **primary:** `cs.RO` · **Submitted:** 2026-03-17
+- **Authors:** Timothée Gavin, Simon Lacroix, Murat Bronz
+- **abs:** https://arxiv.org/abs/2603.16279 · **PDF:** https://arxiv.org/pdf/2603.16279v1
+
+## 4262. Physics-integrated neural differentiable modeling for immersed boundary systems
+
+- **ID:** `2603.16277` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Chenglin Li, Hang Xu, Jianting Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.16277 · **PDF:** https://arxiv.org/pdf/2603.16277v1
+
+## 4263. MSRAMIE: Multimodal Structured Reasoning Agent for Multi-instruction Image Editing
+
+- **ID:** `2603.16967` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Zhaoyuan Qiu, Ken Chen, Xiangwei Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.16967 · **PDF:** https://arxiv.org/pdf/2603.16967v1
+
+## 4264. CineSRD: Leveraging Visual, Acoustic, and Linguistic Cues for Open-World Visual Media Speaker Diarization
+
+- **ID:** `2603.16966` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Liangbin Huang, Xiaohua Liao, Chaoqun Cui, et al.
+- **abs:** https://arxiv.org/abs/2603.16966 · **PDF:** https://arxiv.org/pdf/2603.16966v1
+
+## 4265. Adaptive Theory of Mind for LLM-based Multi-Agent Coordination
+
+- **ID:** `2603.16264` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Chunjiang Mu, Ya Zeng, Qiaosheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.16264 · **PDF:** https://arxiv.org/pdf/2603.16264v1
+
+## 4266. Behavior-Centric Extraction of Scenarios from Highway Traffic Data and their Domain-Knowledge-Guided Clustering using CVQ-VAE
+
+- **ID:** `2603.16964` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Niklas Roßberg, Sinan Hasirlioglu, Mohamed Essayed Bouzouraa, et al.
+- **abs:** https://arxiv.org/abs/2603.16964 · **PDF:** https://arxiv.org/pdf/2603.16964v1
+
+## 4267. AW-MoE: All-Weather Mixture of Experts for Robust Multi-Modal 3D Object Detection
+
+- **ID:** `2603.16261` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Hongwei Lin, Xun Huang, Chenglu Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.16261 · **PDF:** https://arxiv.org/pdf/2603.16261v1
+
+## 4268. Human/AI Collective Intelligence for Deliberative Democracy: A Human-Centred Design Approach
+
+- **ID:** `2603.16260` · **primary:** `cs.CY` · **Submitted:** 2026-03-17
+- **Authors:** Anna De Liddo, Lucas Anastasiou, Simon Buckingham Shum
+- **abs:** https://arxiv.org/abs/2603.16260 · **PDF:** https://arxiv.org/pdf/2603.16260v1
+
+## 4269. Is Semi-Automatic Transcription Useful in Corpus Creation? Preliminary Considerations on the KIParla Corpus
+
+- **ID:** `2603.16258` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Martina Simonotti, Ludovica Pannitto, Eleonora Zucchini, et al.
+- **abs:** https://arxiv.org/abs/2603.16258 · **PDF:** https://arxiv.org/pdf/2603.16258v1
+
+## 4270. Scaling Attention via Feature Sparsity
+
+- **ID:** `2603.22300` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Yan Xie, Tiansheng Wen, Tangda Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.22300 · **PDF:** https://arxiv.org/pdf/2603.22300v2
+
+## 4271. Grounding the Score: Explicit Visual Premise Verification for Reliable Vision-Language Process Reward Models
+
+- **ID:** `2603.16253` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Junxin Wang, Dai Guan, Weijie Qiu, et al.
+- **abs:** https://arxiv.org/abs/2603.16253 · **PDF:** https://arxiv.org/pdf/2603.16253v1
+
+## 4272. Visual Prompt Discovery via Semantic Exploration
+
+- **ID:** `2603.16250` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Jaechang Kim, Yotaro Shimose, Zhao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.16250 · **PDF:** https://arxiv.org/pdf/2603.16250v1
+
+## 4273. Between the Layers Lies the Truth: Uncertainty Estimation in LLMs Using Intra-Layer Local Information Scores
+
+- **ID:** `2603.22299` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Zvi N. Badash, Yonatan Belinkov, Moti Freiman
+- **abs:** https://arxiv.org/abs/2603.22299 · **PDF:** https://arxiv.org/pdf/2603.22299v1
+
+## 4274. How to Utilize Complementary Vision-Text Information for 2D Structure Understanding
+
+- **ID:** `2603.16245` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Jiancheng Dong, Pengyue Jia, Derong Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.16245 · **PDF:** https://arxiv.org/pdf/2603.16245v1
+
+## 4275. More Rounds, More Noise: Why Multi-Turn Review Fails to Improve Cross-Context Verification
+
+- **ID:** `2603.16244` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Song Tae-Eun
+- **abs:** https://arxiv.org/abs/2603.16244 · **PDF:** https://arxiv.org/pdf/2603.16244v1
+
+## 4276. RASLF: Representation-Aware State Space Model for Light Field Super-Resolution
+
+- **ID:** `2603.16243` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Zeqiang Wei, Kai Jin, Kuan Song, et al.
+- **abs:** https://arxiv.org/abs/2603.16243 · **PDF:** https://arxiv.org/pdf/2603.16243v1
+
+## 4277. Neural Pushforward Samplers for the Fokker-Planck Equation on Embedded Riemannian Manifolds
+
+- **ID:** `2603.16239` · **primary:** `math.NA` · **Submitted:** 2026-03-17
+- **Authors:** Andrew Qing He, Wei Cai
+- **abs:** https://arxiv.org/abs/2603.16239 · **PDF:** https://arxiv.org/pdf/2603.16239v2
+
+## 4278. ReFORM: Review-aggregated Profile Generation via LLM with Multi-Factor Attention for Restaurant Recommendation
+
+- **ID:** `2603.16236` · **primary:** `cs.IR` · **Submitted:** 2026-03-17
+- **Authors:** Moonsoo Park, Seulbeen Je, Donghyeon Park
+- **abs:** https://arxiv.org/abs/2603.16236 · **PDF:** https://arxiv.org/pdf/2603.16236v1
+
+## 4279. Efficient AI-Driven Multi-Section Whole Slide Image Analysis for Biochemical Recurrence Prediction in Prostate Cancer
+
+- **ID:** `2603.20273` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Yesung Cho, Dongmyung Shin, Sujeong Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.20273 · **PDF:** https://arxiv.org/pdf/2603.20273v1
+
+## 4280. Dual Consensus: Escaping from Spurious Majority in Unsupervised RLVR via Two-Stage Vote Mechanism
+
+- **ID:** `2603.16223` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Kaixuan Du, Meng Cao, Hang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.16223 · **PDF:** https://arxiv.org/pdf/2603.16223v1
+
+## 4281. SpecSteer: Synergizing Local Context and Global Reasoning for Efficient Personalized Generation
+
+- **ID:** `2603.16219` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Hang Lv, Sheng Liang, Hao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.16219 · **PDF:** https://arxiv.org/pdf/2603.16219v1
+
+## 4282. Generative AI for Quantum Circuits and Quantum Code: A Technical Review and Taxonomy
+
+- **ID:** `2603.16216` · **primary:** `cs.CE` · **Submitted:** 2026-03-17
+- **Authors:** Juhani Merilehto
+- **abs:** https://arxiv.org/abs/2603.16216 · **PDF:** https://arxiv.org/pdf/2603.16216v1
+
+## 4283. CoMAI: A Collaborative Multi-Agent Framework for Robust and Equitable Interview Evaluation
+
+- **ID:** `2603.16215` · **primary:** `cs.MA` · **Submitted:** 2026-03-17
+- **Authors:** Gengxin Sun, Ruihao Yu, Liangyi Yin, et al.
+- **abs:** https://arxiv.org/abs/2603.16215 · **PDF:** https://arxiv.org/pdf/2603.16215v1
+
+## 4284. MOSAIC: Composable Safety Alignment with Modular Control Tokens
+
+- **ID:** `2603.16210` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Jingyu Peng, Hongyu Chen, Jiancheng Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.16210 · **PDF:** https://arxiv.org/pdf/2603.16210v1
+
+## 4285. Proactive Rejection and Grounded Execution: A Dual-Stage Intent Analysis Paradigm for Safe and Efficient AIoT Smart Homes
+
+- **ID:** `2603.16207` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Xinxin Jin, Zhengwei Ni, Zhengguo Sheng, et al.
+- **abs:** https://arxiv.org/abs/2603.16207 · **PDF:** https://arxiv.org/pdf/2603.16207v1
+
+## 4286. Offline Exploration-Aware Fine-Tuning for Long-Chain Mathematical Reasoning
+
+- **ID:** `2603.16206` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Yongyu Mu, Jiali Zeng, Fandong Meng, et al.
+- **abs:** https://arxiv.org/abs/2603.16206 · **PDF:** https://arxiv.org/pdf/2603.16206v1
+
+## 4287. A Scoping Review of AI-Driven Digital Interventions in Mental Health Care: Mapping Applications Across Screening, Support, Monitoring, Prevention, and Clinical Education
+
+- **ID:** `2603.16204` · **primary:** `cs.CY` · **Submitted:** 2026-03-17
+- **Authors:** Yang Ni, Fanli Jia
+- **abs:** https://arxiv.org/abs/2603.16204 · **PDF:** https://arxiv.org/pdf/2603.16204v1
+
+## 4288. Robust Generative Audio Quality Assessment: Disentangling Quality from Spurious Correlations
+
+- **ID:** `2603.16201` · **primary:** `eess.AS` · **Submitted:** 2026-03-17
+- **Authors:** Kuan-Tang Huang, Chien-Chun Wang, Cheng-Yeh Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.16201 · **PDF:** https://arxiv.org/pdf/2603.16201v1
+
+## 4289. Online Semi-infinite Linear Programming: Efficient Algorithms via Function Approximation
+
+- **ID:** `2603.16200` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Yiming Zong, Jiashuo Jiang
+- **abs:** https://arxiv.org/abs/2603.16200 · **PDF:** https://arxiv.org/pdf/2603.16200v1
+
+## 4290. Are Large Language Models Truly Smarter Than Humans?
+
+- **ID:** `2603.16197` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Eshwar Reddy M, Sourav Karmakar
+- **abs:** https://arxiv.org/abs/2603.16197 · **PDF:** https://arxiv.org/pdf/2603.16197v1
+
+## 4291. Structured Semantic Cloaking for Jailbreak Attacks on Large Language Models
+
+- **ID:** `2603.16192` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Xiaobing Sun, Perry Lam, Shaohua Li, et al.
+- **abs:** https://arxiv.org/abs/2603.16192 · **PDF:** https://arxiv.org/pdf/2603.16192v1
+
+## 4292. Sample-Efficient Adaptation of Drug-Response Models to Patient Tumors under Strong Biological Domain Shift
+
+- **ID:** `2603.16185` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Camille Jimenez Cortes, Philippe Lalanda, German Vega
+- **abs:** https://arxiv.org/abs/2603.16185 · **PDF:** https://arxiv.org/pdf/2603.16185v1
+
+## 4293. A General Deep Learning Framework for Wireless Resource Allocation under Discrete Constraints
+
+- **ID:** `2603.19322` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Yikun Wang, Yang Li, Yik-Chung Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.19322 · **PDF:** https://arxiv.org/pdf/2603.19322v1
+
+## 4294. Polyglot-Lion: Efficient Multilingual ASR for Singapore via Balanced Fine-Tuning of Qwen3-ASR
+
+- **ID:** `2603.16184` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Quy-Anh Dang, Chris Ngo
+- **abs:** https://arxiv.org/abs/2603.16184 · **PDF:** https://arxiv.org/pdf/2603.16184v1
+
+## 4295. 360° Image Perception with MLLMs: A Comprehensive Benchmark and a Training-Free Method
+
+- **ID:** `2603.16179` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Huyen T. T. Tran, Van-Quang Nguyen, Farros Alferro, et al.
+- **abs:** https://arxiv.org/abs/2603.16179 · **PDF:** https://arxiv.org/pdf/2603.16179v2
+
+## 4296. The Finetuner's Fallacy: When to Pretrain with Your Finetuning Data
+
+- **ID:** `2603.16177` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Christina Baek, Ricardo Pio Monti, David Schwab, et al.
+- **abs:** https://arxiv.org/abs/2603.16177 · **PDF:** https://arxiv.org/pdf/2603.16177v2
+
+## 4297. MemX: A Local-First Long-Term Memory System for AI Assistants
+
+- **ID:** `2603.16171` · **primary:** `cs.IR` · **Submitted:** 2026-03-17
+- **Authors:** Lizheng Sun
+- **abs:** https://arxiv.org/abs/2603.16171 · **PDF:** https://arxiv.org/pdf/2603.16171v1
+
+## 4298. Open-Source Reproduction and Explainability Analysis of Corrective Retrieval Augmented Generation
+
+- **ID:** `2603.16169` · **primary:** `cs.IR` · **Submitted:** 2026-03-17
+- **Authors:** Surya Vardhan Yalavarthi
+- **abs:** https://arxiv.org/abs/2603.16169 · **PDF:** https://arxiv.org/pdf/2603.16169v1
+
+## 4299. Homogeneous and Heterogeneous Consistency progressive Re-ranking for Visible-Infrared Person Re-identification
+
+- **ID:** `2603.16165` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Yiming Wang
+- **abs:** https://arxiv.org/abs/2603.16165 · **PDF:** https://arxiv.org/pdf/2603.16165v1
+
+## 4300. STARK: Spatio-Temporal Attention for Representation of Keypoints for Continuous Sign Language Recognition
+
+- **ID:** `2603.16163` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Suvajit Patra, Soumitra Samanta
+- **abs:** https://arxiv.org/abs/2603.16163 · **PDF:** https://arxiv.org/pdf/2603.16163v1
+
+## 4301. SQL-ASTRA: Alleviating Sparse Feedback in Agentic SQL via Column-Set Matching and Trajectory Aggregation
+
+- **ID:** `2603.16161` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Long Li, Zhijian Zhou, Jiangxuan Long, et al.
+- **abs:** https://arxiv.org/abs/2603.16161 · **PDF:** https://arxiv.org/pdf/2603.16161v1
+
+## 4302. Execution-Grounded Credit Assignment for GRPO in Code Generation
+
+- **ID:** `2603.16158` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Abhijit Kumar, Natalya Kumar, Shikhar Gupta
+- **abs:** https://arxiv.org/abs/2603.16158 · **PDF:** https://arxiv.org/pdf/2603.16158v1
+
+## 4303. DyJR: Preserving Diversity in Reinforcement Learning with Verifiable Rewards via Dynamic Jensen-Shannon Replay
+
+- **ID:** `2603.16157` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Long Li, Zhijian Zhou, Tianyi Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.16157 · **PDF:** https://arxiv.org/pdf/2603.16157v1
+
+## 4304. GATS: Gaussian Aware Temporal Scaling Transformer for Invariant 4D Spatio-Temporal Point Cloud Representation
+
+- **ID:** `2603.16154` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Jiayi Tian, Jiaze Wang
+- **abs:** https://arxiv.org/abs/2603.16154 · **PDF:** https://arxiv.org/pdf/2603.16154v1
+
+## 4305. HIPO: Instruction Hierarchy via Constrained Reinforcement Learning
+
+- **ID:** `2603.16152` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Keru Chen, Jun Luo, Sen Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.16152 · **PDF:** https://arxiv.org/pdf/2603.16152v1
+
+## 4306. NeuronSpark: A Spiking Neural Network Language Model with Selective State Space Dynamics
+
+- **ID:** `2603.16148` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Zhengzheng Tang
+- **abs:** https://arxiv.org/abs/2603.16148 · **PDF:** https://arxiv.org/pdf/2603.16148v1
+
+## 4307. Deep Adaptive Model-Based Design of Experiments
+
+- **ID:** `2603.16146` · **primary:** `stat.ML` · **Submitted:** 2026-03-17
+- **Authors:** Arno Strouwen, Sebastian Micluţa-Câmpeanu
+- **abs:** https://arxiv.org/abs/2603.16146 · **PDF:** https://arxiv.org/pdf/2603.16146v2
+
+## 4308. Structure-Aware Multimodal LLM Framework for Trustworthy Near-Field Beam Prediction
+
+- **ID:** `2603.16143` · **primary:** `eess.SP` · **Submitted:** 2026-03-17
+- **Authors:** Mengyuan Li, Qianfan Lu, Jiachen Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.16143 · **PDF:** https://arxiv.org/pdf/2603.16143v1
+
+## 4309. Parametric Social Identity Injection and Diversification in Public Opinion Simulation
+
+- **ID:** `2603.16142` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Hexi Wang, Yujia Zhou, Bangde Du, et al.
+- **abs:** https://arxiv.org/abs/2603.16142 · **PDF:** https://arxiv.org/pdf/2603.16142v1
+
+## 4310. Communication-Aware Multi-Agent Reinforcement Learning for Decentralized Cooperative UAV Deployment
+
+- **ID:** `2603.16141` · **primary:** `cs.MA` · **Submitted:** 2026-03-17
+- **Authors:** Enguang Fan, Yifan Chen, Zihan Shan, et al.
+- **abs:** https://arxiv.org/abs/2603.16141 · **PDF:** https://arxiv.org/pdf/2603.16141v1
+
+## 4311. Noisy Data is Destructive to Reinforcement Learning with Verifiable Rewards
+
+- **ID:** `2603.16140` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Yuxuan Zhu, Daniel Kang
+- **abs:** https://arxiv.org/abs/2603.16140 · **PDF:** https://arxiv.org/pdf/2603.16140v1
+
+## 4312. Answer Bubbles: Information Exposure in AI-Mediated Search
+
+- **ID:** `2603.16138` · **primary:** `cs.IR` · **Submitted:** 2026-03-17
+- **Authors:** Michelle Huang, Agam Goyal, Koustuv Saha, et al.
+- **abs:** https://arxiv.org/abs/2603.16138 · **PDF:** https://arxiv.org/pdf/2603.16138v1
+
+## 4313. SIA: A Synthesize-Inject-Align Framework for Knowledge-Grounded and Secure E-commerce Search LLMs with Industrial Deployment
+
+- **ID:** `2603.16137` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Zhouwei Zhai, Mengxiang Chen, Anmeng Zhang
+- **abs:** https://arxiv.org/abs/2603.16137 · **PDF:** https://arxiv.org/pdf/2603.16137v2
+
+## 4314. When Generative Augmentation Hurts: A Benchmark Study of GAN and Diffusion Models for Bias Correction in AI Classification Systems
+
+- **ID:** `2603.16134` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Shesh Narayan Gupta, Nik Bear Brown
+- **abs:** https://arxiv.org/abs/2603.16134 · **PDF:** https://arxiv.org/pdf/2603.16134v1
+
+## 4315. SciZoom: A Large-scale Benchmark for Hierarchical Scientific Summarization across the LLM Era
+
+- **ID:** `2603.16131` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Han Jang, Junhyeok Lee, Kyu Sung Choi
+- **abs:** https://arxiv.org/abs/2603.16131 · **PDF:** https://arxiv.org/pdf/2603.16131v1
+
+## 4316. Social Simulacra in the Wild: AI Agent Communities on Moltbook
+
+- **ID:** `2603.16128` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Agam Goyal, Olivia Pal, Hari Sundaram, et al.
+- **abs:** https://arxiv.org/abs/2603.16128 · **PDF:** https://arxiv.org/pdf/2603.16128v2
+
+## 4317. Pre-training LLM without Learning Rate Decay Enhances Supervised Fine-Tuning
+
+- **ID:** `2603.16127` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Kazuki Yano, Shun Kiyono, Sosuke Kobayashi, et al.
+- **abs:** https://arxiv.org/abs/2603.16127 · **PDF:** https://arxiv.org/pdf/2603.16127v1
+
+## 4318. SWE-QA-Pro: A Representative Benchmark and Scalable Training Recipe for Repository-Level Code Understanding
+
+- **ID:** `2603.16124` · **primary:** `cs.SE` · **Submitted:** 2026-03-17
+- **Authors:** Songcheng Cai, Zhiheng Lyu, Yuansheng Ni, et al.
+- **abs:** https://arxiv.org/abs/2603.16124 · **PDF:** https://arxiv.org/pdf/2603.16124v1
+
+## 4319. Functorial Neural Architectures from Higher Inductive Types
+
+- **ID:** `2603.16123` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Karen Sargsyan
+- **abs:** https://arxiv.org/abs/2603.16123 · **PDF:** https://arxiv.org/pdf/2603.16123v1
+
+## 4320. Language Models Don't Know What You Want: Evaluating Personalization in Deep Research Needs Real Users
+
+- **ID:** `2603.16120` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Nishant Balepur, Malachi Hamada, Varsha Kishore, et al.
+- **abs:** https://arxiv.org/abs/2603.16120 · **PDF:** https://arxiv.org/pdf/2603.16120v1
+
+## 4321. Adversarial attacks against Modern Vision-Language Models
+
+- **ID:** `2603.16960` · **primary:** `cs.CR` · **Submitted:** 2026-03-17
+- **Authors:** Alejandro Paredes La Torre
+- **abs:** https://arxiv.org/abs/2603.16960 · **PDF:** https://arxiv.org/pdf/2603.16960v2
+
+## 4322. PathGLS: Evaluating Pathology Vision-Language Models without Ground Truth through Multi-Dimensional Consistency
+
+- **ID:** `2603.16113` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Minbing Chen, Zhu Meng, Fei Su
+- **abs:** https://arxiv.org/abs/2603.16113 · **PDF:** https://arxiv.org/pdf/2603.16113v1
+
+## 4323. ASDA: Automated Skill Distillation and Adaptation for Financial Reasoning
+
+- **ID:** `2603.16112` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Tik Yu Yim, Wenting Tan, Sum Yee Chan, et al.
+- **abs:** https://arxiv.org/abs/2603.16112 · **PDF:** https://arxiv.org/pdf/2603.16112v1
+
+## 4324. VIGIL: Towards Edge-Extended Agentic AI for Enterprise IT Support
+
+- **ID:** `2603.16110` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Sarthak Ahuja, Neda Kordjazi, Evren Yortucboylu, et al.
+- **abs:** https://arxiv.org/abs/2603.16110 · **PDF:** https://arxiv.org/pdf/2603.16110v1
+
+## 4325. RepoReviewer: A Local-First Multi-Agent Architecture for Repository-Level Code Review
+
+- **ID:** `2603.16107` · **primary:** `cs.SE` · **Submitted:** 2026-03-17
+- **Authors:** Peng Zhang
+- **abs:** https://arxiv.org/abs/2603.16107 · **PDF:** https://arxiv.org/pdf/2603.16107v1
+
+## 4326. NANOZK: Layerwise Zero-Knowledge Proofs for Verifiable Large Language Model Inference
+
+- **ID:** `2603.18046` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Zhaohui Geoffrey Wang
+- **abs:** https://arxiv.org/abs/2603.18046 · **PDF:** https://arxiv.org/pdf/2603.18046v1
+
+## 4327. Machine intelligence supports the full chain of 2D dendrite synthesis
+
+- **ID:** `2603.16959` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-17
+- **Authors:** Wenqiang Huang, Susu Fang, Xuhang Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.16959 · **PDF:** https://arxiv.org/pdf/2603.16959v1
+
+## 4328. Frequency Matters: Fast Model-Agnostic Data Curation for Pruning and Quantization
+
+- **ID:** `2603.16105` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Francesco Pio Monaco, Elia Cunegatti, Flavio Vella, et al.
+- **abs:** https://arxiv.org/abs/2603.16105 · **PDF:** https://arxiv.org/pdf/2603.16105v1
+
+## 4329. Efficient LLM Serving for Agentic Workflows: A Data Systems Perspective
+
+- **ID:** `2603.16104` · **primary:** `cs.MA` · **Submitted:** 2026-03-17
+- **Authors:** Noppanat Wadlom, Junyi Shen, Yao Lu
+- **abs:** https://arxiv.org/abs/2603.16104 · **PDF:** https://arxiv.org/pdf/2603.16104v1
+
+## 4330. LICA: Layered Image Composition Annotations for Graphic Design Research
+
+- **ID:** `2603.16098` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Elad Hirsch, Shubham Yadav, Mohit Garg, et al.
+- **abs:** https://arxiv.org/abs/2603.16098 · **PDF:** https://arxiv.org/pdf/2603.16098v2
+
+## 4331. Diffusion Models for Joint Audio-Video Generation
+
+- **ID:** `2603.16093` · **primary:** `cs.SD` · **Submitted:** 2026-03-17
+- **Authors:** Alejandro Paredes La Torre
+- **abs:** https://arxiv.org/abs/2603.16093 · **PDF:** https://arxiv.org/pdf/2603.16093v1
+
+## 4332. Parallel In-context Learning for Large Vision Language Models
+
+- **ID:** `2603.16092` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Shin'ya Yamaguchi, Daiki Chijiwa, Tamao Sakao, et al.
+- **abs:** https://arxiv.org/abs/2603.16092 · **PDF:** https://arxiv.org/pdf/2603.16092v1
+
+## 4333. CounterRefine: Answer-Conditioned Counterevidence Retrieval for Inference-Time Knowledge Repair in Factual Question Answering
+
+- **ID:** `2603.16091` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Tianyi Huang, Ying Kai Deng
+- **abs:** https://arxiv.org/abs/2603.16091 · **PDF:** https://arxiv.org/pdf/2603.16091v1
+
+## 4334. RecBundle: A Next-Generation Geometric Paradigm for Explainable Recommender Systems
+
+- **ID:** `2603.16088` · **primary:** `cs.IR` · **Submitted:** 2026-03-17
+- **Authors:** Hui Wang, Tianzhu Hu, Mingming Li, et al.
+- **abs:** https://arxiv.org/abs/2603.16088 · **PDF:** https://arxiv.org/pdf/2603.16088v1
+
+## 4335. Towards the Vision-Sound-Language-Action Paradigm: The HEAR Framework for Sound-Centric Manipulation
+
+- **ID:** `2603.16086` · **primary:** `cs.RO` · **Submitted:** 2026-03-17
+- **Authors:** Chang Nie, Tianchen Deng, Guangming Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.16086 · **PDF:** https://arxiv.org/pdf/2603.16086v1
+
+## 4336. Interact3D: Compositional 3D Generation of Interactive Objects
+
+- **ID:** `2603.16085` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Hui Shan, Keyang Luo, Ming Li, et al.
+- **abs:** https://arxiv.org/abs/2603.16085 · **PDF:** https://arxiv.org/pdf/2603.16085v1
+
+## 4337. Prompt-tuning with Attribute Guidance for Low-resource Entity Matching
+
+- **ID:** `2603.19321` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Lihui Liu, Carl Yang
+- **abs:** https://arxiv.org/abs/2603.19321 · **PDF:** https://arxiv.org/pdf/2603.19321v1
+
+## 4338. A Depth-Aware Comparative Study of Euclidean and Hyperbolic Graph Neural Networks on Bitcoin Transaction Systems
+
+- **ID:** `2603.16080` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Ankit Ghimire, Saydul Akbar Murad, Nick Rahimi
+- **abs:** https://arxiv.org/abs/2603.16080 · **PDF:** https://arxiv.org/pdf/2603.16080v1
+
+## 4339. MDM-Prime-v2: Binary Encoding and Index Shuffling Enable Compute-optimal Scaling of Diffusion Language Models
+
+- **ID:** `2603.16077` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Chen-Hao Chao, Wei-Fang Sun, Junwei Quan, et al.
+- **abs:** https://arxiv.org/abs/2603.16077 · **PDF:** https://arxiv.org/pdf/2603.16077v2
+
+## 4340. ClaimFlow: Tracing the Evolution of Scientific Claims in NLP
+
+- **ID:** `2603.16073` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Aniket Pramanick, Yufang Hou, Saif M. Mohammad, et al.
+- **abs:** https://arxiv.org/abs/2603.16073 · **PDF:** https://arxiv.org/pdf/2603.16073v1
+
+## 4341. SEAHateCheck: Functional Tests for Detecting Hate Speech in Low-Resource Languages of Southeast Asia
+
+- **ID:** `2603.16070` · **primary:** `cs.CL` · **Submitted:** 2026-03-17
+- **Authors:** Ri Chi Ng, Aditi Kumaresan, Yujia Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.16070 · **PDF:** https://arxiv.org/pdf/2603.16070v1
+
+## 4342. PhysQuantAgent: An Inference Pipeline of Mass Estimation for Vision-Language Models
+
+- **ID:** `2603.16958` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Hisayuki Yokomizo, Taiki Miyanishi, Yan Gang, et al.
+- **abs:** https://arxiv.org/abs/2603.16958 · **PDF:** https://arxiv.org/pdf/2603.16958v1
+
+## 4343. Resource Consumption Threats in Large Language Models
+
+- **ID:** `2603.16068` · **primary:** `cs.CR` · **Submitted:** 2026-03-17
+- **Authors:** Yuanhe Zhang, Xinyue Wang, Zhican Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.16068 · **PDF:** https://arxiv.org/pdf/2603.16068v2
+
+## 4344. Attribution Upsampling should Redistribute, Not Interpolate
+
+- **ID:** `2603.16067` · **primary:** `cs.CV` · **Submitted:** 2026-03-17
+- **Authors:** Vincenzo Buono, Peyman Sheikholharam Mashhadi, Mahmoud Rahat, et al.
+- **abs:** https://arxiv.org/abs/2603.16067 · **PDF:** https://arxiv.org/pdf/2603.16067v1
+
+## 4345. Adaptive regularization parameter selection for high-dimensional inverse problems: A Bayesian approach with Tucker low-rank constraints
+
+- **ID:** `2603.16066` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Qing-Mei Yang, Da-Qing Zhang
+- **abs:** https://arxiv.org/abs/2603.16066 · **PDF:** https://arxiv.org/pdf/2603.16066v1
+
+## 4346. Large Reward Models: Generalizable Online Robot Reward Generation with Vision-Language Models
+
+- **ID:** `2603.16065` · **primary:** `cs.RO` · **Submitted:** 2026-03-17
+- **Authors:** Yanru Wu, Weiduo Yuan, Ang Qi, et al.
+- **abs:** https://arxiv.org/abs/2603.16065 · **PDF:** https://arxiv.org/pdf/2603.16065v2
+
+## 4347. Safe Distributionally Robust Feature Selection under Covariate Shift
+
+- **ID:** `2603.16062` · **primary:** `stat.ML` · **Submitted:** 2026-03-17
+- **Authors:** Hiroyuki Hanada, Satoshi Akahane, Noriaki Hashimoto, et al.
+- **abs:** https://arxiv.org/abs/2603.16062 · **PDF:** https://arxiv.org/pdf/2603.16062v1
+
+## 4348. ARISE: Agent Reasoning with Intrinsic Skill Evolution in Hierarchical Reinforcement Learning
+
+- **ID:** `2603.16060` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Yu Li, Rui Miao, Zhengling Qi, et al.
+- **abs:** https://arxiv.org/abs/2603.16060 · **PDF:** https://arxiv.org/pdf/2603.16060v2
+
+## 4349. Population Annealing as a Discrete-Time Schrödinger Bridge
+
+- **ID:** `2603.16056` · **primary:** `cond-mat.stat-mech` · **Submitted:** 2026-03-17
+- **Authors:** Masayuki Ohzeki
+- **abs:** https://arxiv.org/abs/2603.16056 · **PDF:** https://arxiv.org/pdf/2603.16056v1
+
+## 4350. A Context Alignment Pre-processor for Enhancing the Coherence of Human-LLM Dialog
+
+- **ID:** `2603.16052` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Ding Wei
+- **abs:** https://arxiv.org/abs/2603.16052 · **PDF:** https://arxiv.org/pdf/2603.16052v1
+
+## 4351. EMPD: An Event-based Multimodal Physiological Dataset for Remote Pulse Wave Detection
+
+- **ID:** `2603.26699` · **primary:** `eess.SP` · **Submitted:** 2026-03-17
+- **Authors:** Qian Feng, Pengfei Li, Rongshan Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.26699 · **PDF:** https://arxiv.org/pdf/2603.26699v1
+
+## 4352. POaaS: Minimal-Edit Prompt Optimization as a Service to Lift Accuracy and Cut Hallucinations on On-Device sLLMs
+
+- **ID:** `2603.16045` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Jungwoo Shim, Dae Won Kim, Sun Wook Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.16045 · **PDF:** https://arxiv.org/pdf/2603.16045v1
+
+## 4353. Enhancing Linguistic Generalization of VLA: Fine-Tuning OpenVLA via Synthetic Instruction Augmentation
+
+- **ID:** `2603.16044` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Dongik Shin
+- **abs:** https://arxiv.org/abs/2603.16044 · **PDF:** https://arxiv.org/pdf/2603.16044v1
+
+## 4354. Collaborative Temporal Feature Generation via Critic-Free Reinforcement Learning for Cross-User Sensor-Based Activity Recognition
+
+- **ID:** `2603.16043` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Xiaozhou Ye, Feng Jiang, Zihan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.16043 · **PDF:** https://arxiv.org/pdf/2603.16043v1
+
+## 4355. Shuffling the Stochastic Mirror Descent via Dual Lipschitz Continuity and Kernel Conditioning
+
+- **ID:** `2603.16042` · **primary:** `math.OC` · **Submitted:** 2026-03-17
+- **Authors:** Junwen Qiu, Leilei Mei, Junyu Zhang
+- **abs:** https://arxiv.org/abs/2603.16042 · **PDF:** https://arxiv.org/pdf/2603.16042v1
+
+## 4356. Power Analysis for Prediction-Powered Inference
+
+- **ID:** `2603.16041` · **primary:** `stat.ME` · **Submitted:** 2026-03-17
+- **Authors:** Yiqun T. Chen, Moran Guo, Shengy Li
+- **abs:** https://arxiv.org/abs/2603.16041 · **PDF:** https://arxiv.org/pdf/2603.16041v1
+
+## 4357. Residual Stream Duality in Modern Transformer Architectures
+
+- **ID:** `2603.16039` · **primary:** `cs.LG` · **Submitted:** 2026-03-17
+- **Authors:** Yifan Zhang
+- **abs:** https://arxiv.org/abs/2603.16039 · **PDF:** https://arxiv.org/pdf/2603.16039v1
+
+## 4358. Interpretable Context Methodology: Folder Structure as Agentic Architecture
+
+- **ID:** `2603.16021` · **primary:** `cs.AI` · **Submitted:** 2026-03-17
+- **Authors:** Jake Van Clief, David McDermott
+- **abs:** https://arxiv.org/abs/2603.16021 · **PDF:** https://arxiv.org/pdf/2603.16021v2
+
+## 4359. IRAM-Omega-Q: A Computational Architecture for Uncertainty Regulation in Artificial Agents
+
+- **ID:** `2603.16020` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Veronique Ziegler
+- **abs:** https://arxiv.org/abs/2603.16020 · **PDF:** https://arxiv.org/pdf/2603.16020v1
+
+## 4360. Understanding Moral Reasoning Trajectories in Large Language Models: Toward Probing-Based Explainability
+
+- **ID:** `2603.16017` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Fan Huang, Haewoon Kwak, Jisun An
+- **abs:** https://arxiv.org/abs/2603.16017 · **PDF:** https://arxiv.org/pdf/2603.16017v1
+
+## 4361. FlatLands: Generative Floormap Completion From a Single Egocentric View
+
+- **ID:** `2603.16016` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Subhransu S. Bhattacharjee, Dylan Campbell, Rahul Shome
+- **abs:** https://arxiv.org/abs/2603.16016 · **PDF:** https://arxiv.org/pdf/2603.16016v1
+
+## 4362. The Importance of Being Smoothly Calibrated
+
+- **ID:** `2603.16015` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Parikshit Gopalan, Konstantinos Stavropoulos, Kunal Talwar, et al.
+- **abs:** https://arxiv.org/abs/2603.16015 · **PDF:** https://arxiv.org/pdf/2603.16015v1
+
+## 4363. Evaluating Agentic Optimization on Large Codebases
+
+- **ID:** `2603.16011` · **primary:** `cs.SE` · **Submitted:** 2026-03-16
+- **Authors:** Atharva Sehgal, James Hou, Akanksha Sarkar, et al.
+- **abs:** https://arxiv.org/abs/2603.16011 · **PDF:** https://arxiv.org/pdf/2603.16011v1
+
+## 4364. Breakdown properties of optimal transport maps: general transportation costs
+
+- **ID:** `2603.16005` · **primary:** `math.ST` · **Submitted:** 2026-03-16
+- **Authors:** Alberto Gonzalez-Sanz, Marco Avella Medina
+- **abs:** https://arxiv.org/abs/2603.16005 · **PDF:** https://arxiv.org/pdf/2603.16005v1
+
+## 4365. RadAnnotate: Large Language Models for Efficient and Reliable Radiology Report Annotation
+
+- **ID:** `2603.16002` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Saisha Pradeep Shetty, Roger Eric Goldman, Vladimir Filkov
+- **abs:** https://arxiv.org/abs/2603.16002 · **PDF:** https://arxiv.org/pdf/2603.16002v1
+
+## 4366. Mostly Text, Smart Visuals: Asymmetric Text-Visual Pruning for Large Vision-Language Models
+
+- **ID:** `2603.16001` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Sijie Li, Biao Qian, Jungong Han
+- **abs:** https://arxiv.org/abs/2603.16001 · **PDF:** https://arxiv.org/pdf/2603.16001v1
+
+## 4367. NLP Occupational Emergence Analysis: How Occupations Form and Evolve in Real Time -- A Zero-Assumption Method Demonstrated on AI in the US Technology Workforce, 2022-2026
+
+- **ID:** `2603.15998` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** David Nordfors
+- **abs:** https://arxiv.org/abs/2603.15998 · **PDF:** https://arxiv.org/pdf/2603.15998v2
+
+## 4368. Visual Set Program Synthesizer
+
+- **ID:** `2603.15997` · **primary:** `cs.MM` · **Submitted:** 2026-03-16
+- **Authors:** Zehua Cheng, Wei Dai, Wenhu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.15997 · **PDF:** https://arxiv.org/pdf/2603.15997v1
+
+## 4369. Selective Memory for Artificial Intelligence: Write-Time Gating with Hierarchical Archiving
+
+- **ID:** `2603.15994` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Oliver Zahn, Simran Chana
+- **abs:** https://arxiv.org/abs/2603.15994 · **PDF:** https://arxiv.org/pdf/2603.15994v1
+
+## 4370. The Midas Touch in Gaze vs. Hand Pointing: Modality-Specific Failure Modes and Implications for XR Interfaces
+
+- **ID:** `2603.15991` · **primary:** `cs.HC` · **Submitted:** 2026-03-16
+- **Authors:** Mohammad Dastgheib, Fatemeh Pourmahdian
+- **abs:** https://arxiv.org/abs/2603.15991 · **PDF:** https://arxiv.org/pdf/2603.15991v1
+
+## 4371. W2T: LoRA Weights Already Know What They Can Do
+
+- **ID:** `2603.15990` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Xiaolong Han, Ferrante Neri, Zijian Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.15990 · **PDF:** https://arxiv.org/pdf/2603.15990v1
+
+## 4372. Something from Nothing: Data Augmentation for Robust Severity Level Estimation of Dysarthric Speech
+
+- **ID:** `2603.15988` · **primary:** `eess.AS` · **Submitted:** 2026-03-16
+- **Authors:** Jaesung Bae, Xiuwen Zheng, Minje Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.15988 · **PDF:** https://arxiv.org/pdf/2603.15988v1
+
+## 4373. Determinism in the Undetermined: Deterministic Output in Charge-Conserving Continuous-Time Neuromorphic Systems with Temporal Stochasticity
+
+- **ID:** `2603.15987` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Jing Yan, Kang You, Zhezhi He, et al.
+- **abs:** https://arxiv.org/abs/2603.15987 · **PDF:** https://arxiv.org/pdf/2603.15987v1
+
+## 4374. Aligning Paralinguistic Understanding and Generation in Speech LLMs via Multi-Task Reinforcement Learning
+
+- **ID:** `2603.15981` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Jingxiang Chen, Minseok Kim, Seong-Gyun Leem, et al.
+- **abs:** https://arxiv.org/abs/2603.15981 · **PDF:** https://arxiv.org/pdf/2603.15981v1
+
+## 4375. Standardizing Medical Images at Scale for AI
+
+- **ID:** `2603.15980` · **primary:** `eess.IV` · **Submitted:** 2026-03-16
+- **Authors:** Callen MacPhee, Yiming Zhou, Koichiro Kishima, et al.
+- **abs:** https://arxiv.org/abs/2603.15980 · **PDF:** https://arxiv.org/pdf/2603.15980v1
+
+## 4376. From Workflow Automation to Capability Closure: A Formal Framework for Safe and Revenue-Aware Customer Service AI
+
+- **ID:** `2603.15978` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Cosimo Spera
+- **abs:** https://arxiv.org/abs/2603.15978 · **PDF:** https://arxiv.org/pdf/2603.15978v2
+
+## 4377. An Agentic Evaluation Framework for AI-Generated Scientific Code in PETSc
+
+- **ID:** `2603.15976` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Hong Zhang, Barry Smith, Satish Balay, et al.
+- **abs:** https://arxiv.org/abs/2603.15976 · **PDF:** https://arxiv.org/pdf/2603.15976v1
+
+## 4378. Safety is Non-Compositional: A Formal Framework for Capability-Based AI Systems
+
+- **ID:** `2603.15973` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Cosimo Spera
+- **abs:** https://arxiv.org/abs/2603.15973 · **PDF:** https://arxiv.org/pdf/2603.15973v2
+
+## 4379. 100x Cost & Latency Reduction: Performance Analysis of AI Query Approximation using Lightweight Proxy Models
+
+- **ID:** `2603.15970` · **primary:** `cs.DB` · **Submitted:** 2026-03-16
+- **Authors:** Yeounoh Chung, Rushabh Desai, Jian He, et al.
+- **abs:** https://arxiv.org/abs/2603.15970 · **PDF:** https://arxiv.org/pdf/2603.15970v5
+
+## 4380. Robust Language Identification for Romansh Varieties
+
+- **ID:** `2603.15969` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Charlotte Model, Sina Ahmadi, Jannis Vamvas
+- **abs:** https://arxiv.org/abs/2603.15969 · **PDF:** https://arxiv.org/pdf/2603.15969v1
+
+## 4381. MAC: Multi-Agent Constitution Learning
+
+- **ID:** `2603.15968` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Rushil Thareja, Gautam Gupta, Francesco Pinto, et al.
+- **abs:** https://arxiv.org/abs/2603.15968 · **PDF:** https://arxiv.org/pdf/2603.15968v1
+
+## 4382. MoLoRA: Composable Specialization via Per-Token Adapter Routing
+
+- **ID:** `2603.15965` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Shrey Shah, Justin Wagle
+- **abs:** https://arxiv.org/abs/2603.15965 · **PDF:** https://arxiv.org/pdf/2603.15965v1
+
+## 4383. Optimizing Hospital Capacity During Pandemics: A Dual-Component Framework for Strategic Patient Relocation
+
+- **ID:** `2603.15960` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Sadaf Tabatabaee, Hicham El Baz, Mohammed Khalil Ghali, et al.
+- **abs:** https://arxiv.org/abs/2603.15960 · **PDF:** https://arxiv.org/pdf/2603.15960v1
+
+## 4384. Deriving Hyperparameter Scaling Laws via Modern Optimization Theory
+
+- **ID:** `2603.15958` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Egor Shulgin, Dimitri von Rütte, Tianyue H. Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.15958 · **PDF:** https://arxiv.org/pdf/2603.15958v1
+
+## 4385. GASP: Guided Asymmetric Self-Play For Coding LLMs
+
+- **ID:** `2603.15957` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Swadesh Jana, Cansu Sancaktar, Tomáš Daniš, et al.
+- **abs:** https://arxiv.org/abs/2603.15957 · **PDF:** https://arxiv.org/pdf/2603.15957v1
+
+## 4386. ExpertGen: Scalable Sim-to-Real Expert Policy Learning from Imperfect Behavior Priors
+
+- **ID:** `2603.15956` · **primary:** `cs.RO` · **Submitted:** 2026-03-16
+- **Authors:** Zifan Xu, Ran Gong, Maria Vittoria Minniti, et al.
+- **abs:** https://arxiv.org/abs/2603.15956 · **PDF:** https://arxiv.org/pdf/2603.15956v1
+
+## 4387. MobileLLM-Flash: Latency-Guided On-Device LLM Design for Industry Scale Deployment
+
+- **ID:** `2603.15954` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Hanxian Huang, Igor Fedorov, Andrey Gromov, et al.
+- **abs:** https://arxiv.org/abs/2603.15954 · **PDF:** https://arxiv.org/pdf/2603.15954v1
+
+## 4388. A Family of LLMs Liberated from Static Vocabularies
+
+- **ID:** `2603.15953` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Aleph Alpha, :, Adnen Abdessaied, et al.
+- **abs:** https://arxiv.org/abs/2603.15953 · **PDF:** https://arxiv.org/pdf/2603.15953v1
+
+## 4389. Protein Design with Agent Rosetta: A Case Study for Specialized Scientific Agents
+
+- **ID:** `2603.15952` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Jacopo Teneggi, S. M. Bargeen A. Turzo, Tanya Marwah, et al.
+- **abs:** https://arxiv.org/abs/2603.15952 · **PDF:** https://arxiv.org/pdf/2603.15952v1
+
+## 4390. POLAR:A Per-User Association Test in Embedding Space
+
+- **ID:** `2603.15950` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Pedro Bento, Arthur Buzelin, Arthur Chagas, et al.
+- **abs:** https://arxiv.org/abs/2603.15950 · **PDF:** https://arxiv.org/pdf/2603.15950v1
+
+## 4391. BanglaSocialBench: A Benchmark for Evaluating Sociopragmatic and Cultural Alignment of LLMs in Bangladeshi Social Interaction
+
+- **ID:** `2603.15949` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Tanvir Ahmed Sijan, S. M Golam Rifat, Pankaj Chowdhury Partha, et al.
+- **abs:** https://arxiv.org/abs/2603.15949 · **PDF:** https://arxiv.org/pdf/2603.15949v2
+
+## 4392. Argumentative Human-AI Decision-Making: Toward AI Agents That Reason With Us, Not For Us
+
+- **ID:** `2603.15946` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Stylianos Loukas Vasileiou, Antonio Rago, Francesca Toni, et al.
+- **abs:** https://arxiv.org/abs/2603.15946 · **PDF:** https://arxiv.org/pdf/2603.15946v1
+
+## 4393. Analytically tractable model of synaptic crowding explains emergent small-world structure and network dynamics
+
+- **ID:** `2603.19320` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-16
+- **Authors:** Makoto Fukushima
+- **abs:** https://arxiv.org/abs/2603.19320 · **PDF:** https://arxiv.org/pdf/2603.19320v1
+
+## 4394. Data-Local Autonomous LLM-Guided Neural Architecture Search for Multiclass Multimodal Time-Series Classification
+
+- **ID:** `2603.15939` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Emil Hardarson, Luka Biedebach, Ómar Bessi Ómarsson, et al.
+- **abs:** https://arxiv.org/abs/2603.15939 · **PDF:** https://arxiv.org/pdf/2603.15939v1
+
+## 4395. CTG-DB: An Ontology-Based Transformation of ClinicalTrials.gov to Enable Cross-Trial Drug Safety Analyses
+
+- **ID:** `2603.15936` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Jeffery L. Painter, François Haguinet, Andrew Bate
+- **abs:** https://arxiv.org/abs/2603.15936 · **PDF:** https://arxiv.org/pdf/2603.15936v2
+
+## 4396. Embodied Foundation Models at the Edge: A Survey of Deployment Constraints and Mitigation Strategies
+
+- **ID:** `2603.16952` · **primary:** `cs.RO` · **Submitted:** 2026-03-16
+- **Authors:** Utkarsh Grover, Ravi Ranjan, Mingyang Mao, et al.
+- **abs:** https://arxiv.org/abs/2603.16952 · **PDF:** https://arxiv.org/pdf/2603.16952v2
+
+## 4397. Semi-Autonomous Formalization of the Vlasov-Maxwell-Landau Equilibrium
+
+- **ID:** `2603.15929` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Vasily Ilin
+- **abs:** https://arxiv.org/abs/2603.15929 · **PDF:** https://arxiv.org/pdf/2603.15929v2
+
+## 4398. Discovery of interaction and diffusion kernels in particle-to-mean-field multi-agent systems
+
+- **ID:** `2603.15927` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Giacomo Albi, Alessandro Alla, Elisa Calzola
+- **abs:** https://arxiv.org/abs/2603.15927 · **PDF:** https://arxiv.org/pdf/2603.15927v1
+
+## 4399. Evaluating Causal Discovery Algorithms for Path-Specific Fairness and Utility in Healthcare
+
+- **ID:** `2603.15926` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Nitish Nagesh, Elahe Khatibi, Thomas Hughes, et al.
+- **abs:** https://arxiv.org/abs/2603.15926 · **PDF:** https://arxiv.org/pdf/2603.15926v1
+
+## 4400. Generative Inverse Design with Abstention via Diagonal Flow Matching
+
+- **ID:** `2603.15925` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Miguel de Campos, Werner Krebs, Hanno Gottschalk
+- **abs:** https://arxiv.org/abs/2603.15925 · **PDF:** https://arxiv.org/pdf/2603.15925v1
+
+## 4401. Learning to Recall with Transformers Beyond Orthogonal Embeddings
+
+- **ID:** `2603.15923` · **primary:** `stat.ML` · **Submitted:** 2026-03-16
+- **Authors:** Nuri Mert Vural, Alberto Bietti, Mahdi Soltanolkotabi, et al.
+- **abs:** https://arxiv.org/abs/2603.15923 · **PDF:** https://arxiv.org/pdf/2603.15923v1
+
+## 4402. Machine Translation in the Wild: User Reaction to Xiaohongshu's Built-In Translation Feature
+
+- **ID:** `2603.15922` · **primary:** `cs.HC` · **Submitted:** 2026-03-16
+- **Authors:** Sui He
+- **abs:** https://arxiv.org/abs/2603.15922 · **PDF:** https://arxiv.org/pdf/2603.15922v1
+
+## 4403. VIBEPASS: Can Vibe Coders Really Pass the Vibe Check?
+
+- **ID:** `2603.15921` · **primary:** `cs.SE` · **Submitted:** 2026-03-16
+- **Authors:** Srijan Bansal, Jiao Fangkai, Yilun Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.15921 · **PDF:** https://arxiv.org/pdf/2603.15921v1
+
+## 4404. Bayesian-guided inverse design of hyperelastic microstructures: Application to stochastic metamaterials
+
+- **ID:** `2603.15917` · **primary:** `cs.CE` · **Submitted:** 2026-03-16
+- **Authors:** Hooman Danesh, Henning Wessels
+- **abs:** https://arxiv.org/abs/2603.15917 · **PDF:** https://arxiv.org/pdf/2603.15917v1
+
+## 4405. Auto Researching, not hyperparameter tuning: Convergence Analysis of 10,000 Experiments
+
+- **ID:** `2603.15916` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Xiaoyi Li
+- **abs:** https://arxiv.org/abs/2603.15916 · **PDF:** https://arxiv.org/pdf/2603.15916v1
+
+## 4406. The Agentic Researcher: A Practical Guide to AI-Assisted Research in Mathematics and Machine Learning
+
+- **ID:** `2603.15914` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Max Zimmer, Nico Pelleriti, Christophe Roux, et al.
+- **abs:** https://arxiv.org/abs/2603.15914 · **PDF:** https://arxiv.org/pdf/2603.15914v1
+
+## 4407. Prompt Engineering for Scale Development in Generative Psychometrics
+
+- **ID:** `2603.15909` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Lara Lee Russell-Lasalandra, Hudson Golino
+- **abs:** https://arxiv.org/abs/2603.15909 · **PDF:** https://arxiv.org/pdf/2603.15909v1
+
+## 4408. Game-Theory-Assisted Reinforcement Learning for Border Defense: Early Termination based on Analytical Solutions
+
+- **ID:** `2603.15907` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Goutam Das, Michael Dorothy, Kyle Volle, et al.
+- **abs:** https://arxiv.org/abs/2603.15907 · **PDF:** https://arxiv.org/pdf/2603.15907v1
+
+## 4409. Agent-based imitation dynamics can yield efficiently compressed population-level vocabularies
+
+- **ID:** `2603.15903` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Nathaniel Imel, Richard Futrell, Michael Franke, et al.
+- **abs:** https://arxiv.org/abs/2603.15903 · **PDF:** https://arxiv.org/pdf/2603.15903v1
+
+## 4410. Federated Learning for Privacy-Preserving Medical AI
+
+- **ID:** `2603.15901` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Tin Hoang
+- **abs:** https://arxiv.org/abs/2603.15901 · **PDF:** https://arxiv.org/pdf/2603.15901v1
+
+## 4411. Minimum-Action Learning: Energy-Constrained Symbolic Model Selection for Physical Law Identification from Noisy Data
+
+- **ID:** `2603.16951` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Martin G. Frasch
+- **abs:** https://arxiv.org/abs/2603.16951 · **PDF:** https://arxiv.org/pdf/2603.16951v2
+
+## 4412. The Internet of Physical AI Agents: Interoperability, Longevity, and the Cost of Getting It Wrong
+
+- **ID:** `2603.15900` · **primary:** `cs.NI` · **Submitted:** 2026-03-16
+- **Authors:** Roberto Morabito, Mallik Tatipamula
+- **abs:** https://arxiv.org/abs/2603.15900 · **PDF:** https://arxiv.org/pdf/2603.15900v1
+
+## 4413. COGNAC at SemEval-2026 Task 5: LLM Ensembles for Human-Level Word Sense Plausibility Rating in Challenging Narratives
+
+- **ID:** `2603.15897` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Azwad Anjum Islam, Tisa Islam Erana
+- **abs:** https://arxiv.org/abs/2603.15897 · **PDF:** https://arxiv.org/pdf/2603.15897v1
+
+## 4414. Temporal Fact Conflicts in LLMs: Reproducibility Insights from Unifying DYNAMICQA and MULAN
+
+- **ID:** `2603.15892` · **primary:** `cs.IR` · **Submitted:** 2026-03-16
+- **Authors:** Ritajit Dey, Iadh Ounis, Graham McDonald, et al.
+- **abs:** https://arxiv.org/abs/2603.15892 · **PDF:** https://arxiv.org/pdf/2603.15892v1
+
+## 4415. AsgardBench -- Evaluating Visually Grounded Interactive Planning Under Minimal Feedback
+
+- **ID:** `2603.15888` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Andrea Tupini, Lars Liden, Reuben Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.15888 · **PDF:** https://arxiv.org/pdf/2603.15888v2
+
+## 4416. EvoIQA - Explaining Image Distortions with Evolved White-Box Logic
+
+- **ID:** `2603.15887` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Ruchika Gupta, Illya Bakurov, Nathan Haut, et al.
+- **abs:** https://arxiv.org/abs/2603.15887 · **PDF:** https://arxiv.org/pdf/2603.15887v1
+
+## 4417. Physicochemical-Neural Fusion for Semi-Closed-Circuit Respiratory Autonomy in Extreme Environments
+
+- **ID:** `2603.26697` · **primary:** `eess.SY` · **Submitted:** 2026-03-16
+- **Authors:** Phillip Kingston, Nicholas Johnston
+- **abs:** https://arxiv.org/abs/2603.26697 · **PDF:** https://arxiv.org/pdf/2603.26697v1
+
+## 4418. PhasorFlow: A Python Library for Unit Circle Based Computing
+
+- **ID:** `2603.15886` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Dibakar Sigdel, Namuna Panday
+- **abs:** https://arxiv.org/abs/2603.15886 · **PDF:** https://arxiv.org/pdf/2603.15886v2
+
+## 4419. Resilience Meets Autonomy: Governing Embodied AI in Critical Infrastructure
+
+- **ID:** `2603.15885` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Puneet Sharma, Christer Henrik Pursiainen
+- **abs:** https://arxiv.org/abs/2603.15885 · **PDF:** https://arxiv.org/pdf/2603.15885v1
+
+## 4420. Electrodermal Activity as a Unimodal Signal for Aerobic Exercise Detection in Wearable Sensors
+
+- **ID:** `2603.15880` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Rena Mira Krishna, Ramya Sankar, Shadi Ghiasi
+- **abs:** https://arxiv.org/abs/2603.15880 · **PDF:** https://arxiv.org/pdf/2603.15880v1
+
+## 4421. Counteractive RL: Rethinking Core Principles for Efficient and Scalable Deep Reinforcement Learning
+
+- **ID:** `2603.15871` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Ezgi Korkmaz
+- **abs:** https://arxiv.org/abs/2603.15871 · **PDF:** https://arxiv.org/pdf/2603.15871v1
+
+## 4422. Kriging via variably scaled kernels
+
+- **ID:** `2603.16950` · **primary:** `stat.ML` · **Submitted:** 2026-03-16
+- **Authors:** Gianluca Audone, Francesco Marchetti, Emma Perracchione, et al.
+- **abs:** https://arxiv.org/abs/2603.16950 · **PDF:** https://arxiv.org/pdf/2603.16950v1
+
+## 4423. Evaluating Black-Box Vulnerabilities with Wasserstein-Constrained Data Perturbations
+
+- **ID:** `2603.15867` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Adriana Laurindo Monteiro, Jean-Michel Loubes
+- **abs:** https://arxiv.org/abs/2603.15867 · **PDF:** https://arxiv.org/pdf/2603.15867v1
+
+## 4424. Interpretative Interfaces: Designing for AI-Mediated Reading Practices and the Knowledge Commons
+
+- **ID:** `2603.15863` · **primary:** `cs.HC` · **Submitted:** 2026-03-16
+- **Authors:** Gabrielle Benabdallah
+- **abs:** https://arxiv.org/abs/2603.15863 · **PDF:** https://arxiv.org/pdf/2603.15863v1
+
+## 4425. Self-supervised Disentanglement of Disease Effects from Aging in 3D Medical Shapes
+
+- **ID:** `2603.15862` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Jakaria Rabbi, Nilanjan Ray, Dana Cobzas
+- **abs:** https://arxiv.org/abs/2603.15862 · **PDF:** https://arxiv.org/pdf/2603.15862v1
+
+## 4426. Regularized Latent Dynamics Prediction is a Strong Baseline For Behavioral Foundation Models
+
+- **ID:** `2603.15857` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Pranaya Jajoo, Harshit Sikchi, Siddhant Agarwal, et al.
+- **abs:** https://arxiv.org/abs/2603.15857 · **PDF:** https://arxiv.org/pdf/2603.15857v1
+
+## 4427. FlashSampling: Fast and Memory-Efficient Exact Sampling
+
+- **ID:** `2603.15854` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Tomas Ruiz, Zhen Qin, Yifan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.15854 · **PDF:** https://arxiv.org/pdf/2603.15854v1
+
+## 4428. Algorithmic Trading Strategy Development and Optimisation
+
+- **ID:** `2603.15848` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Owen Nyo Wei Yuan, Victor Tan Jia Xuan, Ong Jun Yao Fabian, et al.
+- **abs:** https://arxiv.org/abs/2603.15848 · **PDF:** https://arxiv.org/pdf/2603.15848v2
+
+## 4429. FEEL (Force-Enhanced Egocentric Learning): A Dataset for Physical Action Understanding
+
+- **ID:** `2603.15847` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Eadom Dessalene, Botao He, Michael Maynord, et al.
+- **abs:** https://arxiv.org/abs/2603.15847 · **PDF:** https://arxiv.org/pdf/2603.15847v1
+
+## 4430. Informationally Compressive Anonymization: Non-Degrading Sensitive Input Protection for Privacy-Preserving Supervised Machine Learning
+
+- **ID:** `2603.15842` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Jeremy J Samuelson
+- **abs:** https://arxiv.org/abs/2603.15842 · **PDF:** https://arxiv.org/pdf/2603.15842v1
+
+## 4431. When Stability Fails: Hidden Failure Modes Of LLMS in Data-Constrained Scientific Decision-Making
+
+- **ID:** `2603.15840` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Nazia Riasat
+- **abs:** https://arxiv.org/abs/2603.15840 · **PDF:** https://arxiv.org/pdf/2603.15840v1
+
+## 4432. Persona-Conditioned Risk Behavior in Large Language Models: A Simulated Gambling Study with GPT-4.1
+
+- **ID:** `2603.15831` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Sankalp Dubedy
+- **abs:** https://arxiv.org/abs/2603.15831 · **PDF:** https://arxiv.org/pdf/2603.15831v1
+
+## 4433. Hypothesis Class Determines Explanation: Why Accurate Models Disagree on Feature Attribution
+
+- **ID:** `2603.15821` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Thackshanaramana B
+- **abs:** https://arxiv.org/abs/2603.15821 · **PDF:** https://arxiv.org/pdf/2603.15821v1
+
+## 4434. Longitudinal Risk Prediction in Mammography with Privileged History Distillation
+
+- **ID:** `2603.15814` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Banafsheh Karimian, Alexis Guichemerre, Soufiane Belharbi, et al.
+- **abs:** https://arxiv.org/abs/2603.15814 · **PDF:** https://arxiv.org/pdf/2603.15814v1
+
+## 4435. Don't Trust Stubborn Neighbors: A Security Framework for Agentic Networks
+
+- **ID:** `2603.15809` · **primary:** `cs.MA` · **Submitted:** 2026-03-16
+- **Authors:** Samira Abedini, Sina Mavali, Lea Schönherr, et al.
+- **abs:** https://arxiv.org/abs/2603.15809 · **PDF:** https://arxiv.org/pdf/2603.15809v1
+
+## 4436. Mask Is What DLLM Needs: A Masked Data Training Paradigm for Diffusion LLMs
+
+- **ID:** `2603.15803` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Linrui Ma, Yufei Cui, Kai Han, et al.
+- **abs:** https://arxiv.org/abs/2603.15803 · **PDF:** https://arxiv.org/pdf/2603.15803v1
+
+## 4437. Time-Aware Prior Fitted Networks for Zero-Shot Forecasting with Exogenous Variables
+
+- **ID:** `2603.15802` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Andres Potapczynski, Ravi Kiran Selvam, Tatiana Konstantinova, et al.
+- **abs:** https://arxiv.org/abs/2603.15802 · **PDF:** https://arxiv.org/pdf/2603.15802v1
+
+## 4438. Evolving Contextual Safety in Multi-Modal Large Language Models via Inference-Time Self-Reflective Memory
+
+- **ID:** `2603.15800` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Ce Zhang, Jinxi He, Junyi He, et al.
+- **abs:** https://arxiv.org/abs/2603.15800 · **PDF:** https://arxiv.org/pdf/2603.15800v1
+
+## 4439. Prose2Policy (P2P): A Practical LLM Pipeline for Translating Natural-Language Access Policies into Executable Rego
+
+- **ID:** `2603.15799` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Vatsal Gupta, Darshan Sreenivasamurthy
+- **abs:** https://arxiv.org/abs/2603.15799 · **PDF:** https://arxiv.org/pdf/2603.15799v1
+
+## 4440. CUBE: A Standard for Unifying Agent Benchmarks
+
+- **ID:** `2603.15798` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Alexandre Lacoste, Nicolas Gontier, Oleh Shliazhko, et al.
+- **abs:** https://arxiv.org/abs/2603.15798 · **PDF:** https://arxiv.org/pdf/2603.15798v1
+
+## 4441. OMNIFLOW: A Physics-Grounded Multimodal Agent for Generalized Scientific Reasoning
+
+- **ID:** `2603.15797` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Hao Wu, Yongheng Zhang, Yuan Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.15797 · **PDF:** https://arxiv.org/pdf/2603.15797v2
+
+## 4442. Learnability with Partial Labels and Adaptive Nearest Neighbors
+
+- **ID:** `2603.15781` · **primary:** `stat.ML` · **Submitted:** 2026-03-16
+- **Authors:** Nicolas A. Errandonea, Santiago Mazuelas, Jose A. Lozano, et al.
+- **abs:** https://arxiv.org/abs/2603.15781 · **PDF:** https://arxiv.org/pdf/2603.15781v2
+
+## 4443. Parallelised Differentiable Straightest Geodesics for 3D Meshes
+
+- **ID:** `2603.15780` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Hippolyte Verninas, Caner Korkmaz, Stefanos Zafeiriou, et al.
+- **abs:** https://arxiv.org/abs/2603.15780 · **PDF:** https://arxiv.org/pdf/2603.15780v1
+
+## 4444. Morphemes Without Borders: Evaluating Root-Pattern Morphology in Arabic Tokenizers and LLMs
+
+- **ID:** `2603.15773` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Yara Alakeel, Chatrine Qwaider, Hanan Aldarmaki, et al.
+- **abs:** https://arxiv.org/abs/2603.15773 · **PDF:** https://arxiv.org/pdf/2603.15773v1
+
+## 4445. CorrectionPlanner: Self-Correction Planner with Reinforcement Learning in Autonomous Driving
+
+- **ID:** `2603.15771` · **primary:** `cs.RO` · **Submitted:** 2026-03-16
+- **Authors:** Yihong Guo, Dongqiangzi Ye, Sijia Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.15771 · **PDF:** https://arxiv.org/pdf/2603.15771v1
+
+## 4446. Entropy-Aware Task Offloading in Mobile Edge Computing
+
+- **ID:** `2603.16949` · **primary:** `cs.NI` · **Submitted:** 2026-03-16
+- **Authors:** Mohsen Sahraei Ardakani, Hong Wan, Rui Song
+- **abs:** https://arxiv.org/abs/2603.16949 · **PDF:** https://arxiv.org/pdf/2603.16949v2
+
+## 4447. Simulation Distillation: Pretraining World Models in Simulation for Rapid Real-World Adaptation
+
+- **ID:** `2603.15759` · **primary:** `cs.RO` · **Submitted:** 2026-03-16
+- **Authors:** Jacob Levy, Tyler Westenbroek, Kevin Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.15759 · **PDF:** https://arxiv.org/pdf/2603.15759v1
+
+## 4448. You've Got a Golden Ticket: Improving Generative Robot Policies With A Single Noise Vector
+
+- **ID:** `2603.15757` · **primary:** `cs.RO` · **Submitted:** 2026-03-16
+- **Authors:** Omkar Patil, Ondrej Biza, Thomas Weng, et al.
+- **abs:** https://arxiv.org/abs/2603.15757 · **PDF:** https://arxiv.org/pdf/2603.15757v1
+
+## 4449. Mixture-of-Depths Attention
+
+- **ID:** `2603.15619` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Lianghui Zhu, Yuxin Fang, Bencheng Liao, et al.
+- **abs:** https://arxiv.org/abs/2603.15619 · **PDF:** https://arxiv.org/pdf/2603.15619v1
+
+## 4450. HorizonMath: Measuring AI Progress Toward Mathematical Discovery with Automatic Verification
+
+- **ID:** `2603.15617` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Erik Y. Wang, Sumeet Motwani, James V. Roggeveen, et al.
+- **abs:** https://arxiv.org/abs/2603.15617 · **PDF:** https://arxiv.org/pdf/2603.15617v1
+
+## 4451. Mechanistic Origin of Moral Indifference in Language Models
+
+- **ID:** `2603.15615` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Lingyu Li, Yan Teng, Yingchun Wang
+- **abs:** https://arxiv.org/abs/2603.15615 · **PDF:** https://arxiv.org/pdf/2603.15615v1
+
+## 4452. Code-A1: Adversarial Evolving of Code LLM and Test LLM via Reinforcement Learning
+
+- **ID:** `2603.15611` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Aozhe Wang, Yuchen Yan, Nan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.15611 · **PDF:** https://arxiv.org/pdf/2603.15611v1
+
+## 4453. Do Metrics for Counterfactual Explanations Align with User Perception?
+
+- **ID:** `2603.15607` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Felix Liedeker, Basil Ell, Philipp Cimiano, et al.
+- **abs:** https://arxiv.org/abs/2603.15607 · **PDF:** https://arxiv.org/pdf/2603.15607v1
+
+## 4454. ClawWorm: Self-Propagating Attacks Across LLM Agent Ecosystems
+
+- **ID:** `2603.15727` · **primary:** `cs.CR` · **Submitted:** 2026-03-16
+- **Authors:** Yihao Zhang, Zeming Wei, Xiaokun Luan, et al.
+- **abs:** https://arxiv.org/abs/2603.15727 · **PDF:** https://arxiv.org/pdf/2603.15727v2
+
+## 4455. MiroThinker-1.7 & H1: Towards Heavy-Duty Research Agents via Verification
+
+- **ID:** `2603.15726` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** MiroMind Team, S. Bai, L. Bing, et al.
+- **abs:** https://arxiv.org/abs/2603.15726 · **PDF:** https://arxiv.org/pdf/2603.15726v1
+
+## 4456. S2Act: Simple Spiking Actor
+
+- **ID:** `2603.15725` · **primary:** `cs.MA` · **Submitted:** 2026-03-16
+- **Authors:** Ugur Akcal, Seung Hyun Kim, Mikihisa Yuasa, et al.
+- **abs:** https://arxiv.org/abs/2603.15725 · **PDF:** https://arxiv.org/pdf/2603.15725v1
+
+## 4457. From Passive Observer to Active Critic: Reinforcement Learning Elicits Process Reasoning for Robotic Manipulation
+
+- **ID:** `2603.15600` · **primary:** `cs.RO` · **Submitted:** 2026-03-16
+- **Authors:** Yibin Liu, Yaxing Lyu, Daqi Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.15600 · **PDF:** https://arxiv.org/pdf/2603.15600v1
+
+## 4458. SmartSearch: How Ranking Beats Structure for Conversational Memory Retrieval
+
+- **ID:** `2603.15599` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Jesper Derehag, Carlos Calva, Timmy Ghiurau
+- **abs:** https://arxiv.org/abs/2603.15599 · **PDF:** https://arxiv.org/pdf/2603.15599v1
+
+## 4459. AC-Foley: Reference-Audio-Guided Video-to-Audio Synthesis with Acoustic Transfer
+
+- **ID:** `2603.15597` · **primary:** `cs.SD` · **Submitted:** 2026-03-16
+- **Authors:** Pengjun Fang, Yingqing He, Yazhou Xing, et al.
+- **abs:** https://arxiv.org/abs/2603.15597 · **PDF:** https://arxiv.org/pdf/2603.15597v2
+
+## 4460. Robust and Computationally Efficient Linear Contextual Bandits under Adversarial Corruption and Heavy-Tailed Noise
+
+- **ID:** `2603.15596` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Naoto Tani, Futoshi Futami
+- **abs:** https://arxiv.org/abs/2603.15596 · **PDF:** https://arxiv.org/pdf/2603.15596v1
+
+## 4461. OpenSeeker: Democratizing Frontier Search Agents by Fully Open-Sourcing Training Data
+
+- **ID:** `2603.15594` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Yuwen Du, Rui Ye, Shuo Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.15594 · **PDF:** https://arxiv.org/pdf/2603.15594v1
+
+## 4462. Effective Distillation to Hybrid xLSTM Architectures
+
+- **ID:** `2603.15590` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Lukas Hauzenberger, Niklas Schmidinger, Thomas Schmied, et al.
+- **abs:** https://arxiv.org/abs/2603.15590 · **PDF:** https://arxiv.org/pdf/2603.15590v1
+
+## 4463. Computational Concept of the Psyche
+
+- **ID:** `2603.15586` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Anton Kolonin, Vladimir Krykov
+- **abs:** https://arxiv.org/abs/2603.15586 · **PDF:** https://arxiv.org/pdf/2603.15586v2
+
+## 4464. Physics-Informed Neural Systems for the Simulation of EUV Electromagnetic Wave Diffraction from a Lithography Mask
+
+- **ID:** `2603.15584` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Vasiliy A. Es'kin, Egor V. Ivanov
+- **abs:** https://arxiv.org/abs/2603.15584 · **PDF:** https://arxiv.org/pdf/2603.15584v2
+
+## 4465. Low-Complexity and Consistent Graphon Estimation from Multiple Networks
+
+- **ID:** `2603.15578` · **primary:** `stat.ME` · **Submitted:** 2026-03-16
+- **Authors:** Roland Boniface Sogan, Tabea Rebafka
+- **abs:** https://arxiv.org/abs/2603.15578 · **PDF:** https://arxiv.org/pdf/2603.15578v1
+
+## 4466. Unbiased and Biased Variance-Reduced Forward-Reflected-Backward Splitting Methods for Stochastic Composite Inclusions
+
+- **ID:** `2603.15576` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Quoc Tran-Dinh, Nghia Nguyen-Trung
+- **abs:** https://arxiv.org/abs/2603.15576 · **PDF:** https://arxiv.org/pdf/2603.15576v1
+
+## 4467. Co-Design of Memory-Storage Systems for Workload Awareness with Interpretable Models
+
+- **ID:** `2603.15571` · **primary:** `cs.AR` · **Submitted:** 2026-03-16
+- **Authors:** Jay Sarkar, Vamsi Pavan Rayaprolu, Abhijeet Bhalerao
+- **abs:** https://arxiv.org/abs/2603.15571 · **PDF:** https://arxiv.org/pdf/2603.15571v1
+
+## 4468. Mamba-3: Improved Sequence Modeling using State Space Principles
+
+- **ID:** `2603.15569` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Aakash Lahoti, Kevin Y. Li, Berlin Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.15569 · **PDF:** https://arxiv.org/pdf/2603.15569v1
+
+## 4469. Estimating Staged Event Tree Models via Hierarchical Clustering on the Simplex
+
+- **ID:** `2603.15568` · **primary:** `stat.ML` · **Submitted:** 2026-03-16
+- **Authors:** Muhammad Shoaib, Eva Riccomagno, Manuele Leonelli, et al.
+- **abs:** https://arxiv.org/abs/2603.15568 · **PDF:** https://arxiv.org/pdf/2603.15568v1
+
+## 4470. Meta-TTRL: A Metacognitive Framework for Self-Improving Test-Time Reinforcement Learning in Unified Multimodal Models
+
+- **ID:** `2603.15724` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Lit Sin Tan, Junzhe Chen, Xiaolong Fu, et al.
+- **abs:** https://arxiv.org/abs/2603.15724 · **PDF:** https://arxiv.org/pdf/2603.15724v1
+
+## 4471. Lore: Repurposing Git Commit Messages as a Structured Knowledge Protocol for AI Coding Agents
+
+- **ID:** `2603.15566` · **primary:** `cs.SE` · **Submitted:** 2026-03-16
+- **Authors:** Ivan Stetsenko
+- **abs:** https://arxiv.org/abs/2603.15566 · **PDF:** https://arxiv.org/pdf/2603.15566v1
+
+## 4472. Predictive Uncertainty in Short-Term PV Forecasting under Missing Data: A Multiple Imputation Approach
+
+- **ID:** `2603.15564` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Parastoo Pashmchi, Jérôme Benoit, Motonobu Kanagawa
+- **abs:** https://arxiv.org/abs/2603.15564 · **PDF:** https://arxiv.org/pdf/2603.15564v1
+
+## 4473. The PokeAgent Challenge: Competitive and Long-Context Learning at Scale
+
+- **ID:** `2603.15563` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Seth Karten, Jake Grigsby, Tersoo Upaa, et al.
+- **abs:** https://arxiv.org/abs/2603.15563 · **PDF:** https://arxiv.org/pdf/2603.15563v2
+
+## 4474. Context-Length Robustness in Question Answering Models: A Comparative Empirical Study
+
+- **ID:** `2603.15723` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Trishita Dhara, Siddhesh Sheth
+- **abs:** https://arxiv.org/abs/2603.15723 · **PDF:** https://arxiv.org/pdf/2603.15723v1
+
+## 4475. Self-Distillation of Hidden Layers for Self-Supervised Representation Learning
+
+- **ID:** `2603.15553` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Scott C. Lowe, Anthony Fuller, Sageev Oore, et al.
+- **abs:** https://arxiv.org/abs/2603.15553 · **PDF:** https://arxiv.org/pdf/2603.15553v1
+
+## 4476. Can LLMs Model Incorrect Student Reasoning? A Case Study on Distractor Generation
+
+- **ID:** `2603.15547` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Yanick Zengaffinen, Andreas Opedal, Donya Rooein, et al.
+- **abs:** https://arxiv.org/abs/2603.15547 · **PDF:** https://arxiv.org/pdf/2603.15547v1
+
+## 4477. A Framework and Prototype for a Navigable Map of Datasets in Engineering Design and Systems Engineering
+
+- **ID:** `2603.15722` · **primary:** `cs.SE` · **Submitted:** 2026-03-16
+- **Authors:** H. Sinan Bank, Daniel R. Herber
+- **abs:** https://arxiv.org/abs/2603.15722 · **PDF:** https://arxiv.org/pdf/2603.15722v2
+
+## 4478. InterveneBench: Benchmarking LLMs for Intervention Reasoning and Causal Study Design in Real Social Systems
+
+- **ID:** `2603.15542` · **primary:** `cs.CY` · **Submitted:** 2026-03-16
+- **Authors:** Shaojie Shi, Zhengyu Shi, Lingran Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.15542 · **PDF:** https://arxiv.org/pdf/2603.15542v1
+
+## 4479. Bridging Local and Global Knowledge: Cascaded Mixture-of-Experts Learning for Near-Shortest Path Routing
+
+- **ID:** `2603.15541` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Yung-Fu Chen, Anish Arora
+- **abs:** https://arxiv.org/abs/2603.15541 · **PDF:** https://arxiv.org/pdf/2603.15541v1
+
+## 4480. DOT: Dynamic Knob Selection and Online Sampling for Automated Database Tuning
+
+- **ID:** `2603.15540` · **primary:** `cs.DB` · **Submitted:** 2026-03-16
+- **Authors:** Yifan Wang, Debabrota Basu, Pierre Bourhis, et al.
+- **abs:** https://arxiv.org/abs/2603.15540 · **PDF:** https://arxiv.org/pdf/2603.15540v1
+
+## 4481. Vib2ECG: A Paired Chest-Lead SCG-ECG Dataset and Benchmark for ECG Reconstruction
+
+- **ID:** `2603.15539` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Guorui Lu, Xiaohui Cai, Todor Stefanov, et al.
+- **abs:** https://arxiv.org/abs/2603.15539 · **PDF:** https://arxiv.org/pdf/2603.15539v1
+
+## 4482. Are Dilemmas and Conflicts in LLM Alignment Solvable? A View from Priority Graph
+
+- **ID:** `2603.15527` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Zhenheng Tang, Xiang Liu, Qian Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.15527 · **PDF:** https://arxiv.org/pdf/2603.15527v1
+
+## 4483. Building Trust in PINNs: Error Estimation through Finite Difference Methods
+
+- **ID:** `2603.15526` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Aleksander Krasowski, René P. Klausen, Aycan Celik, et al.
+- **abs:** https://arxiv.org/abs/2603.15526 · **PDF:** https://arxiv.org/pdf/2603.15526v1
+
+## 4484. SlovKE: A Large-Scale Dataset and LLM Evaluation for Slovak Keyphrase Extraction
+
+- **ID:** `2603.15523` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** David Števaňák, Marek Šuppa
+- **abs:** https://arxiv.org/abs/2603.15523 · **PDF:** https://arxiv.org/pdf/2603.15523v1
+
+## 4485. Beyond the Covariance Trap: Unlocking Generalization in Same-Subject Knowledge Editing for Large Language Models
+
+- **ID:** `2603.15518` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Xiyu Liu, Qingyi Si, Zhengxiao Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.15518 · **PDF:** https://arxiv.org/pdf/2603.15518v1
+
+## 4486. ViX-Ray: A Vietnamese Chest X-Ray Dataset for Vision-Language Models
+
+- **ID:** `2603.15513` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Duy Vu Minh Nguyen, Chinh Thanh Truong, Phuc Hoang Tran, et al.
+- **abs:** https://arxiv.org/abs/2603.15513 · **PDF:** https://arxiv.org/pdf/2603.15513v1
+
+## 4487. Not All Invariants Are Equal: Curating Training Data to Accelerate Program Verification with SLMs
+
+- **ID:** `2603.15510` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Ido Pinto, Yizhak Yisrael Elboher, Haoze Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.15510 · **PDF:** https://arxiv.org/pdf/2603.15510v1
+
+## 4488. Federated Learning of Binary Neural Networks: Enabling Low-Cost Inference
+
+- **ID:** `2603.15507` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Nitin Priyadarshini Shankar, Soham Lahiri, Sheetal Kalyani, et al.
+- **abs:** https://arxiv.org/abs/2603.15507 · **PDF:** https://arxiv.org/pdf/2603.15507v1
+
+## 4489. Seeking SOTA: Time-Series Forecasting Must Adopt Taxonomy-Specific Evaluation to Dispel Illusory Gains
+
+- **ID:** `2603.15506` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Raeid Saqur, Christoph Bergmeir, Blanka Horvath, et al.
+- **abs:** https://arxiv.org/abs/2603.15506 · **PDF:** https://arxiv.org/pdf/2603.15506v1
+
+## 4490. Understanding Reasoning in LLMs through Strategic Information Allocation under Uncertainty
+
+- **ID:** `2603.15500` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Jeonghye Kim, Xufang Luo, Minbeom Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.15500 · **PDF:** https://arxiv.org/pdf/2603.15500v1
+
+## 4491. Grokking as a Variance-Limited Phase Transition: Spectral Gating and the Epsilon-Stability Threshold
+
+- **ID:** `2603.15492` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Pratyush Acharya, Habish Dhakal
+- **abs:** https://arxiv.org/abs/2603.15492 · **PDF:** https://arxiv.org/pdf/2603.15492v1
+
+## 4492. Agentic workflow enables the recovery of critical materials from complex feedstocks via selective precipitation
+
+- **ID:** `2603.15491` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-16
+- **Authors:** Andrew Ritchhart, Sarah I. Allec, Pravalika Butreddy, et al.
+- **abs:** https://arxiv.org/abs/2603.15491 · **PDF:** https://arxiv.org/pdf/2603.15491v1
+
+## 4493. RSGen: Enhancing Layout-Driven Remote Sensing Image Generation with Diverse Edge Guidance
+
+- **ID:** `2603.15484` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Xianbao Hou, Yonghao He, Zeyd Boukhers, et al.
+- **abs:** https://arxiv.org/abs/2603.15484 · **PDF:** https://arxiv.org/pdf/2603.15484v2
+
+## 4494. Talk, Evaluate, Diagnose: User-aware Agent Evaluation with Automated Error Analysis
+
+- **ID:** `2603.15483` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Penny Chong, Harshavardhan Abichandani, Jiyuan Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.15483 · **PDF:** https://arxiv.org/pdf/2603.15483v1
+
+## 4495. TabKD: Tabular Knowledge Distillation through Interaction Diversity of Learned Feature Bins
+
+- **ID:** `2603.15481` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Shovon Niverd Pereira, Krishna Khadka, Yu Lei
+- **abs:** https://arxiv.org/abs/2603.15481 · **PDF:** https://arxiv.org/pdf/2603.15481v1
+
+## 4496. Seeing Beyond: Extrapolative Domain Adaptive Panoramic Segmentation
+
+- **ID:** `2603.15475` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Yuanfan Zheng, Kunyu Peng, Xu Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.15475 · **PDF:** https://arxiv.org/pdf/2603.15475v1
+
+## 4497. Agent Lifecycle Toolkit (ALTK): Reusable Middleware Components for Robust AI Agents
+
+- **ID:** `2603.15473` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Zidane Wright, Jason Tsay, Anupama Murthi, et al.
+- **abs:** https://arxiv.org/abs/2603.15473 · **PDF:** https://arxiv.org/pdf/2603.15473v1
+
+## 4498. RoCo Challenge at AAAI 2026: Benchmarking Robotic Collaborative Manipulation for Assembly Towards Industrial Automation
+
+- **ID:** `2603.15469` · **primary:** `cs.RO` · **Submitted:** 2026-03-16
+- **Authors:** Haichao Liu, Yuheng Zhou, Zhenyu Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.15469 · **PDF:** https://arxiv.org/pdf/2603.15469v1
+
+## 4499. Evasive Intelligence: Lessons from Malware Analysis for Evaluating AI Agents
+
+- **ID:** `2603.15457` · **primary:** `cs.CR` · **Submitted:** 2026-03-16
+- **Authors:** Simone Aonzo, Merve Sahin, Aurélien Francillon, et al.
+- **abs:** https://arxiv.org/abs/2603.15457 · **PDF:** https://arxiv.org/pdf/2603.15457v1
+
+## 4500. A Framework for Modeling Liquefaction-Induced Road Disruptions After Earthquakes: Implications for Emergency Response and Access in the Cascadia Region of North America
+
+- **ID:** `2603.16948` · **primary:** `physics.geo-ph` · **Submitted:** 2026-03-16
+- **Authors:** Morgan D. Sanger, Olyvia B. Smith, Brett W. Maurer, et al.
+- **abs:** https://arxiv.org/abs/2603.16948 · **PDF:** https://arxiv.org/pdf/2603.16948v1
+
+## 4501. Unlocking the Value of Text: Event-Driven Reasoning and Multi-Level Alignment for Time Series Forecasting
+
+- **ID:** `2603.15452` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Siyuan Wang, Peng Chen, Yihang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.15452 · **PDF:** https://arxiv.org/pdf/2603.15452v1
+
+## 4502. Music Genre Classification: A Comparative Analysis of Classical Machine Learning and Deep Learning Approaches
+
+- **ID:** `2603.15440` · **primary:** `cs.SD` · **Submitted:** 2026-03-16
+- **Authors:** Sachin Prajuli, Abhishek Karna, OmPrakash Dhakl
+- **abs:** https://arxiv.org/abs/2603.15440 · **PDF:** https://arxiv.org/pdf/2603.15440v1
+
+## 4503. Deep Reinforcement Learning for Fano Hypersurfaces
+
+- **ID:** `2603.15437` · **primary:** `math.AG` · **Submitted:** 2026-03-16
+- **Authors:** Marc Truter
+- **abs:** https://arxiv.org/abs/2603.15437 · **PDF:** https://arxiv.org/pdf/2603.15437v1
+
+## 4504. Listening to the Echo: User-Reaction Aware Policy Optimization via Scalar-Verbal Hybrid Reinforcement Learning
+
+- **ID:** `2603.15434` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Jing Ye, Xinpei Zhao, Lu Xiang, et al.
+- **abs:** https://arxiv.org/abs/2603.15434 · **PDF:** https://arxiv.org/pdf/2603.15434v2
+
+## 4505. Physics-informed fine-tuning of foundation models for partial differential equations
+
+- **ID:** `2603.15431` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Vlad Medvedev, Leon Armbruster, Christopher Straub, et al.
+- **abs:** https://arxiv.org/abs/2603.15431 · **PDF:** https://arxiv.org/pdf/2603.15431v1
+
+## 4506. Invisible failures in human-AI interactions
+
+- **ID:** `2603.15423` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Christopher Potts, Moritz Sudhof
+- **abs:** https://arxiv.org/abs/2603.15423 · **PDF:** https://arxiv.org/pdf/2603.15423v1
+
+## 4507. CLAG: Adaptive Memory Organization via Agent-Driven Clustering for Small Language Model Agents
+
+- **ID:** `2603.15421` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Taeyun Roh, Wonjune Jang, Junha Jung, et al.
+- **abs:** https://arxiv.org/abs/2603.15421 · **PDF:** https://arxiv.org/pdf/2603.15421v1
+
+## 4508. MA-VLCM: A Vision Language Critic Model for Value Estimation of Policies in Multi-Agent Team Settings
+
+- **ID:** `2603.15418` · **primary:** `cs.RO` · **Submitted:** 2026-03-16
+- **Authors:** Shahil Shaik, Aditya Parameshwaran, Anshul Nayak, et al.
+- **abs:** https://arxiv.org/abs/2603.15418 · **PDF:** https://arxiv.org/pdf/2603.15418v1
+
+## 4509. Amplification Effects in Test-Time Reinforcement Learning: Safety and Reasoning Vulnerabilities
+
+- **ID:** `2603.15417` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Vanshaj Khattar, Md Rafi ur Rashid, Moumita Choudhury, et al.
+- **abs:** https://arxiv.org/abs/2603.15417 · **PDF:** https://arxiv.org/pdf/2603.15417v1
+
+## 4510. RESQ: A Unified Framework for REliability- and Security Enhancement of Quantized Deep Neural Networks
+
+- **ID:** `2603.15413` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Ali Soltan Mohammadi, Samira Nazari, Ali Azarpeyvand, et al.
+- **abs:** https://arxiv.org/abs/2603.15413 · **PDF:** https://arxiv.org/pdf/2603.15413v1
+
+## 4511. Local Urysohn Width: A Topological Complexity Measure for Classification
+
+- **ID:** `2603.15412` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Xin Li
+- **abs:** https://arxiv.org/abs/2603.15412 · **PDF:** https://arxiv.org/pdf/2603.15412v1
+
+## 4512. A Hybrid Modeling Framework for Crop Prediction Tasks via Dynamic Parameter Calibration and Multi-Task Learning
+
+- **ID:** `2603.15411` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** William Solow, Paola Pesantez-Cabrera, Markus Keller, et al.
+- **abs:** https://arxiv.org/abs/2603.15411 · **PDF:** https://arxiv.org/pdf/2603.15411v1
+
+## 4513. SEA-Vision: A Multilingual Benchmark for Comprehensive Document and Scene Text Understanding in Southeast Asia
+
+- **ID:** `2603.15409` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Pengfei Yue, Xingran Zhao, Juntao Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.15409 · **PDF:** https://arxiv.org/pdf/2603.15409v1
+
+## 4514. TrinityGuard: A Unified Framework for Safeguarding Multi-Agent Systems
+
+- **ID:** `2603.15408` · **primary:** `cs.CR` · **Submitted:** 2026-03-16
+- **Authors:** Kai Wang, Biaojie Zeng, Zeming Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.15408 · **PDF:** https://arxiv.org/pdf/2603.15408v1
+
+## 4515. Fusian: Multi-LoRA Fusion for Fine-Grained Continuous MBTI Personality Control in Large Language Models
+
+- **ID:** `2603.15405` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Zehao Chen, Rong Pan
+- **abs:** https://arxiv.org/abs/2603.15405 · **PDF:** https://arxiv.org/pdf/2603.15405v1
+
+## 4516. Detection of Autonomous Shuttles in Urban Traffic Images Using Adaptive Residual Context
+
+- **ID:** `2603.15404` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Mohamed Aziz Younes, Nicolas Saunier, Guillaume-Alexandre Bilodeau
+- **abs:** https://arxiv.org/abs/2603.15404 · **PDF:** https://arxiv.org/pdf/2603.15404v1
+
+## 4517. A Closer Look into LLMs for Table Understanding
+
+- **ID:** `2603.15402` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Jia Wang, Chuanyu Qin, Mingyu Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.15402 · **PDF:** https://arxiv.org/pdf/2603.15402v1
+
+## 4518. SWE-Skills-Bench: Do Agent Skills Actually Help in Real-World Software Engineering?
+
+- **ID:** `2603.15401` · **primary:** `cs.SE` · **Submitted:** 2026-03-16
+- **Authors:** Tingxu Han, Yi Zhang, Wei Song, et al.
+- **abs:** https://arxiv.org/abs/2603.15401 · **PDF:** https://arxiv.org/pdf/2603.15401v1
+
+## 4519. SFCoT: Safer Chain-of-Thought via Active Safety Evaluation and Calibration
+
+- **ID:** `2603.15397` · **primary:** `cs.CR` · **Submitted:** 2026-03-16
+- **Authors:** Yu Pan, Wenlong Yu, Tiejun Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.15397 · **PDF:** https://arxiv.org/pdf/2603.15397v1
+
+## 4520. AI Evasion and Impersonation Attacks on Facial Re-Identification with Activation Map Explanations
+
+- **ID:** `2603.15396` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Noe Claudel, Weisi Guo, Yang Xing
+- **abs:** https://arxiv.org/abs/2603.15396 · **PDF:** https://arxiv.org/pdf/2603.15396v1
+
+## 4521. When Does Sparsity Mitigate the Curse of Depth in LLMs
+
+- **ID:** `2603.15389` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Dilxat Muhtar, Xinyuan Song, Sebastian Pokutta, et al.
+- **abs:** https://arxiv.org/abs/2603.15389 · **PDF:** https://arxiv.org/pdf/2603.15389v1
+
+## 4522. Efficient Morphology-Control Co-Design via Stackelberg Proximal Policy Optimization
+
+- **ID:** `2603.15388` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Yanning Dai, Yuhui Wang, Dylan R. Ashley, et al.
+- **abs:** https://arxiv.org/abs/2603.15388 · **PDF:** https://arxiv.org/pdf/2603.15388v1
+
+## 4523. RieMind: Geometry-Grounded Spatial Agent for Scene Understanding
+
+- **ID:** `2603.15386` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Fernando Ropero, Erkin Turkoz, Daniel Matos, et al.
+- **abs:** https://arxiv.org/abs/2603.15386 · **PDF:** https://arxiv.org/pdf/2603.15386v1
+
+## 4524. Persistence Spheres: a Bi-continuous Linear Representation of Measures for Partial Optimal Transport
+
+- **ID:** `2603.15384` · **primary:** `stat.ML` · **Submitted:** 2026-03-16
+- **Authors:** Matteo Pegoraro
+- **abs:** https://arxiv.org/abs/2603.15384 · **PDF:** https://arxiv.org/pdf/2603.15384v1
+
+## 4525. Why AI systems don't learn and what to do about it: Lessons on autonomous learning from cognitive science
+
+- **ID:** `2603.15381` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Emmanuel Dupoux, Yann LeCun, Jitendra Malik
+- **abs:** https://arxiv.org/abs/2603.15381 · **PDF:** https://arxiv.org/pdf/2603.15381v1
+
+## 4526. More Test-Time Compute Can Hurt: Overestimation Bias in LLM Beam Search
+
+- **ID:** `2603.15377` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Gal Dalal, Assaf Hallak, Gal Chechik, et al.
+- **abs:** https://arxiv.org/abs/2603.15377 · **PDF:** https://arxiv.org/pdf/2603.15377v2
+
+## 4527. GradCFA: A Hybrid Gradient-Based Counterfactual and Feature Attribution Explanation Algorithm for Local Interpretation of Neural Networks
+
+- **ID:** `2603.15373` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Jacob Sanderson, Hua Mao, Wai Lok Woo
+- **abs:** https://arxiv.org/abs/2603.15373 · **PDF:** https://arxiv.org/pdf/2603.15373v1
+
+## 4528. How Vulnerable Are AI Agents to Indirect Prompt Injections? Insights from a Large-Scale Public Competition
+
+- **ID:** `2603.15714` · **primary:** `cs.CR` · **Submitted:** 2026-03-16
+- **Authors:** Mateusz Dziemian, Maxwell Lin, Xiaohan Fu, et al.
+- **abs:** https://arxiv.org/abs/2603.15714 · **PDF:** https://arxiv.org/pdf/2603.15714v1
+
+## 4529. SKILLS: Structured Knowledge Injection for LLM-Driven Telecommunications Operations
+
+- **ID:** `2603.15372` · **primary:** `cs.SE` · **Submitted:** 2026-03-16
+- **Authors:** Ivo Brett
+- **abs:** https://arxiv.org/abs/2603.15372 · **PDF:** https://arxiv.org/pdf/2603.15372v1
+
+## 4530. Brain-Inspired Graph Multi-Agent Systems for LLM Reasoning
+
+- **ID:** `2603.15371` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Guangfu Hao, Yuming Dai, Xianzhe Qin, et al.
+- **abs:** https://arxiv.org/abs/2603.15371 · **PDF:** https://arxiv.org/pdf/2603.15371v1
+
+## 4531. Controlled Langevin Dynamics for Sampling of Feedforward Neural Networks Trained with Minibatches
+
+- **ID:** `2603.15367` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-03-16
+- **Authors:** Alessandro Zambon, Francesca Caruso, Riccardo Zecchina, et al.
+- **abs:** https://arxiv.org/abs/2603.15367 · **PDF:** https://arxiv.org/pdf/2603.15367v1
+
+## 4532. Exploring Novelty Differences between Industry and Academia: A Knowledge Entity-centric Perspective
+
+- **ID:** `2603.19319` · **primary:** `cs.DL` · **Submitted:** 2026-03-16
+- **Authors:** Hongye Zhao, Yi Zhao, Chengzhi Zhang
+- **abs:** https://arxiv.org/abs/2603.19319 · **PDF:** https://arxiv.org/pdf/2603.19319v1
+
+## 4533. CRASH: Cognitive Reasoning Agent for Safety Hazards in Autonomous Driving
+
+- **ID:** `2603.15364` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Erick Silva, Rehana Yasmin, Ali Shoker
+- **abs:** https://arxiv.org/abs/2603.15364 · **PDF:** https://arxiv.org/pdf/2603.15364v1
+
+## 4534. Deep learning and the rate of approximation by flows
+
+- **ID:** `2603.15363` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Jingpu Cheng, Qianxiao Li, Ting Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.15363 · **PDF:** https://arxiv.org/pdf/2603.15363v2
+
+## 4535. FuXiWeather2: Learning accurate atmospheric state estimation for operational global weather forecasting
+
+- **ID:** `2603.15358` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Xiaoze Xu, Xiuyu Sun, Songling Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.15358 · **PDF:** https://arxiv.org/pdf/2603.15358v1
+
+## 4536. Conditional Rectified Flow-based End-to-End Rapid Seismic Inversion Method
+
+- **ID:** `2603.15354` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Haofei Xu, Wei Cheng, Sizhe Li, et al.
+- **abs:** https://arxiv.org/abs/2603.15354 · **PDF:** https://arxiv.org/pdf/2603.15354v1
+
+## 4537. NV-Bench: Benchmark of Nonverbal Vocalization Synthesis for Expressive Text-to-Speech Generation
+
+- **ID:** `2603.15352` · **primary:** `cs.SD` · **Submitted:** 2026-03-16
+- **Authors:** Qinke Ni, Huan Liao, Dekun Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.15352 · **PDF:** https://arxiv.org/pdf/2603.15352v2
+
+## 4538. PMAx: An Agentic Framework for AI-Driven Process Mining
+
+- **ID:** `2603.15351` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Anton Antonov, Humam Kourani, Alessandro Berti, et al.
+- **abs:** https://arxiv.org/abs/2603.15351 · **PDF:** https://arxiv.org/pdf/2603.15351v1
+
+## 4539. Embedding-Aware Feature Discovery: Bridging Latent Representations and Interpretable Features in Event Sequences
+
+- **ID:** `2603.15713` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Artem Sakhno, Ivan Sergeev, Alexey Shestov, et al.
+- **abs:** https://arxiv.org/abs/2603.15713 · **PDF:** https://arxiv.org/pdf/2603.15713v1
+
+## 4540. Intelligent Co-Design: An Interactive LLM Framework for Interior Spatial Design via Multi-Modal Agents
+
+- **ID:** `2603.15341` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Ren Jian Lim, Rushi Dai
+- **abs:** https://arxiv.org/abs/2603.15341 · **PDF:** https://arxiv.org/pdf/2603.15341v1
+
+## 4541. DOS: Dependency-Oriented Sampler for Masked Diffusion Language Models
+
+- **ID:** `2603.15340` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Xueyu Zhou, Yangrong Hu, Jian Huang
+- **abs:** https://arxiv.org/abs/2603.15340 · **PDF:** https://arxiv.org/pdf/2603.15340v1
+
+## 4542. Active Seriation: Efficient Ordering Recovery with Statistical Guarantees
+
+- **ID:** `2603.15336` · **primary:** `stat.ML` · **Submitted:** 2026-03-16
+- **Authors:** James Cheshire, Yann Issartel
+- **abs:** https://arxiv.org/abs/2603.15336 · **PDF:** https://arxiv.org/pdf/2603.15336v1
+
+## 4543. Data Augmentation via Causal-Residual Bootstrapping
+
+- **ID:** `2603.15335` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Mateusz Gajewski, Sophia Xiao, Bijan Mazaheri
+- **abs:** https://arxiv.org/abs/2603.15335 · **PDF:** https://arxiv.org/pdf/2603.15335v1
+
+## 4544. A scaled TW-PINN: A physics-informed neural network for traveling wave solutions of reaction-diffusion equations with general coefficients
+
+- **ID:** `2603.15331` · **primary:** `math.NA` · **Submitted:** 2026-03-16
+- **Authors:** Seungwan Han, Kwanghyuk Park, Jiaxi Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.15331 · **PDF:** https://arxiv.org/pdf/2603.15331v1
+
+## 4545. Tagarela - A Portuguese speech dataset from podcasts
+
+- **ID:** `2603.15326` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Frederico Santos de Oliveira, Lucas Rafael Stefanel Gris, Alef Iury Siqueira Ferreira, et al.
+- **abs:** https://arxiv.org/abs/2603.15326 · **PDF:** https://arxiv.org/pdf/2603.15326v1
+
+## 4546. CASHomon Sets: Efficient Rashomon Sets Across Multiple Model Classes and their Hyperparameters
+
+- **ID:** `2603.15321` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Fiona Katharina Ewald, Martin Binder, Matthias Feurer, et al.
+- **abs:** https://arxiv.org/abs/2603.15321 · **PDF:** https://arxiv.org/pdf/2603.15321v1
+
+## 4547. LLM-Driven Discovery of High-Entropy Catalysts via Retrieval-Augmented Generation
+
+- **ID:** `2603.15712` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-16
+- **Authors:** AI Scientists, Xinyi Lin, Danqing Yin, et al.
+- **abs:** https://arxiv.org/abs/2603.15712 · **PDF:** https://arxiv.org/pdf/2603.15712v1
+
+## 4548. PYTHEN: A Flexible Framework for Legal Reasoning in Python
+
+- **ID:** `2603.15317` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Ha-Thanh Nguyen, Ken Satoh
+- **abs:** https://arxiv.org/abs/2603.15317 · **PDF:** https://arxiv.org/pdf/2603.15317v1
+
+## 4549. Knowledge Graph Extraction from Biomedical Literature for Alkaptonuria Rare Disease
+
+- **ID:** `2603.15711` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Giang Pham, Rebecca Finetti, Caterina Graziani, et al.
+- **abs:** https://arxiv.org/abs/2603.15711 · **PDF:** https://arxiv.org/pdf/2603.15711v1
+
+## 4550. EmergeNav: Structured Embodied Inference for Zero-Shot Vision-and-Language Navigation in Continuous Environments
+
+- **ID:** `2603.16947` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Kun Luo, Xiaoguang Ma
+- **abs:** https://arxiv.org/abs/2603.16947 · **PDF:** https://arxiv.org/pdf/2603.16947v1
+
+## 4551. CCTU: A Benchmark for Tool Use under Complex Constraints
+
+- **ID:** `2603.15309` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Junjie Ye, Guoqiang Zhang, Wenjie Fu, et al.
+- **abs:** https://arxiv.org/abs/2603.15309 · **PDF:** https://arxiv.org/pdf/2603.15309v1
+
+## 4552. A Kolmogorov-Arnold Surrogate Model for Chemical Equilibria: Application to Solid Solutions
+
+- **ID:** `2603.15307` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Leonardo Boledi, Dirk Bosbach, Jenna Poonoosamy
+- **abs:** https://arxiv.org/abs/2603.15307 · **PDF:** https://arxiv.org/pdf/2603.15307v1
+
+## 4553. xplainfi: Feature Importance and Statistical Inference for Machine Learning in R
+
+- **ID:** `2603.15306` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Lukas Burk, Fiona Katharina Ewald, Giuseppe Casalicchio, et al.
+- **abs:** https://arxiv.org/abs/2603.15306 · **PDF:** https://arxiv.org/pdf/2603.15306v1
+
+## 4554. Enhancing classification accuracy through chaos
+
+- **ID:** `2603.15299` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Panos Stinis
+- **abs:** https://arxiv.org/abs/2603.15299 · **PDF:** https://arxiv.org/pdf/2603.15299v1
+
+## 4555. Evolutionary Transfer Learning for Dragonchess
+
+- **ID:** `2603.15297` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Jim O'Connor, Annika Hoag, Sarah Goyette, et al.
+- **abs:** https://arxiv.org/abs/2603.15297 · **PDF:** https://arxiv.org/pdf/2603.15297v1
+
+## 4556. Datasets for Verb Alternations across Languages: BLM Templates and Data Augmentation Strategies
+
+- **ID:** `2603.15295` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Giuseppe Samo, Paola Merlo
+- **abs:** https://arxiv.org/abs/2603.15295 · **PDF:** https://arxiv.org/pdf/2603.15295v1
+
+## 4557. Scalable Simulation-Based Model Inference with Test-Time Complexity Control
+
+- **ID:** `2603.15292` · **primary:** `stat.ML` · **Submitted:** 2026-03-16
+- **Authors:** Manuel Gloeckler, J. P. Manzano-Patrón, Stamatios N. Sotiropoulos, et al.
+- **abs:** https://arxiv.org/abs/2603.15292 · **PDF:** https://arxiv.org/pdf/2603.15292v1
+
+## 4558. Evaluating the Robustness of Reinforcement Learning based Adaptive Traffic Signal Control
+
+- **ID:** `2603.15283` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Dickens Kwesiga, Angshuman Guin, Khaled Abdelghany, et al.
+- **abs:** https://arxiv.org/abs/2603.15283 · **PDF:** https://arxiv.org/pdf/2603.15283v1
+
+## 4559. Algorithms for Deciding the Safety of States in Fully Observable Non-deterministic Problems: Technical Report
+
+- **ID:** `2603.15282` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Johannes Schmalz, Chaahat Jain
+- **abs:** https://arxiv.org/abs/2603.15282 · **PDF:** https://arxiv.org/pdf/2603.15282v1
+
+## 4560. Survey of Various Fuzzy and Uncertain Decision-Making Methods
+
+- **ID:** `2603.15709` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Takaaki Fujita, Florentin Smarandache
+- **abs:** https://arxiv.org/abs/2603.15709 · **PDF:** https://arxiv.org/pdf/2603.15709v1
+
+## 4561. Advancing Multimodal Agent Reasoning with Long-Term Neuro-Symbolic Memory
+
+- **ID:** `2603.15280` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Rongjie Jiang, Jianwei Wang, Gengda Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.15280 · **PDF:** https://arxiv.org/pdf/2603.15280v1
+
+## 4562. Faster Inference of Flow-Based Generative Models via Improved Data-Noise Coupling
+
+- **ID:** `2603.15279` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Aram Davtyan, Leello Tadesse Dadi, Volkan Cevher, et al.
+- **abs:** https://arxiv.org/abs/2603.15279 · **PDF:** https://arxiv.org/pdf/2603.15279v1
+
+## 4563. Complementarity-Preserving Generative Theory for Multimodal ECG Synthesis: A Quantum-Inspired Approach
+
+- **ID:** `2603.26695` · **primary:** `eess.SP` · **Submitted:** 2026-03-16
+- **Authors:** Timothy Oladunni, Farouk Ganiyu-Adewumi, Clyde Baidoo, et al.
+- **abs:** https://arxiv.org/abs/2603.26695 · **PDF:** https://arxiv.org/pdf/2603.26695v1
+
+## 4564. From Documents to Spans: Code-Centric Learning for LLM-based ICD Coding
+
+- **ID:** `2603.15270` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Xu Zhang, Wenxin Ma, Chenxu Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.15270 · **PDF:** https://arxiv.org/pdf/2603.15270v1
+
+## 4565. Mastering the Minority: An Uncertainty-guided Multi-Expert Framework for Challenging-tailed Sequence Learning
+
+- **ID:** `2603.15708` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Ye Wang, Zixuan Wu, Lifeng Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.15708 · **PDF:** https://arxiv.org/pdf/2603.15708v1
+
+## 4566. Automatic Termination Strategy of Inelastic Neutron-scattering Measurement Using Bayesian Optimization for Bin-width Selection
+
+- **ID:** `2603.16946` · **primary:** `physics.data-an` · **Submitted:** 2026-03-16
+- **Authors:** Kensuke Muto, Hirotaka Sakamoto, Kenji Nagata, et al.
+- **abs:** https://arxiv.org/abs/2603.16946 · **PDF:** https://arxiv.org/pdf/2603.16946v1
+
+## 4567. IConE: Batch Independent Collapse Prevention for Self-Supervised Representation Learning
+
+- **ID:** `2603.15263` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Konstantinos Almpanakis, Anna Kreshuk
+- **abs:** https://arxiv.org/abs/2603.15263 · **PDF:** https://arxiv.org/pdf/2603.15263v1
+
+## 4568. Probe-then-Plan: Environment-Aware Planning for Industrial E-commerce Search
+
+- **ID:** `2603.15262` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Mengxiang Chen, Zhouwei Zhai, Jin Li
+- **abs:** https://arxiv.org/abs/2603.15262 · **PDF:** https://arxiv.org/pdf/2603.15262v1
+
+## 4569. AGCD: Agent-Guided Cross-Modal Decoding for Weather Forecasting
+
+- **ID:** `2603.15260` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Jing Wu, Yang Liu, Lin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.15260 · **PDF:** https://arxiv.org/pdf/2603.15260v1
+
+## 4570. Directional Embedding Smoothing for Robust Vision Language Models
+
+- **ID:** `2603.15259` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Ye Wang, Jing Liu, Toshiaki Koike-Akino
+- **abs:** https://arxiv.org/abs/2603.15259 · **PDF:** https://arxiv.org/pdf/2603.15259v1
+
+## 4571. SEMAG: Self-Evolutionary Multi-Agent Code Generation
+
+- **ID:** `2603.15707` · **primary:** `cs.SE` · **Submitted:** 2026-03-16
+- **Authors:** Yulin Peng, Haowen Hou, Xinxin Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.15707 · **PDF:** https://arxiv.org/pdf/2603.15707v1
+
+## 4572. SAGE: Multi-Agent Self-Evolution for LLM Reasoning
+
+- **ID:** `2603.15255` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Yulin Peng, Xinxin Zhu, Chenxing Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.15255 · **PDF:** https://arxiv.org/pdf/2603.15255v2
+
+## 4573. In-Context Symbolic Regression for Robustness-Improved Kolmogorov-Arnold Networks
+
+- **ID:** `2603.15250` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Francesco Sovrano, Lidia Losavio, Giulia Vilone, et al.
+- **abs:** https://arxiv.org/abs/2603.15250 · **PDF:** https://arxiv.org/pdf/2603.15250v1
+
+## 4574. Mechanistic Foundations of Goal-Directed Control
+
+- **ID:** `2603.15248` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Alma Lago
+- **abs:** https://arxiv.org/abs/2603.15248 · **PDF:** https://arxiv.org/pdf/2603.15248v2
+
+## 4575. Practicing with Language Models Cultivates Human Empathic Communication
+
+- **ID:** `2603.15245` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Aakriti Kumar, Nalin Poungpeth, Diyi Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.15245 · **PDF:** https://arxiv.org/pdf/2603.15245v1
+
+## 4576. A proof-of-concept for automated AI-driven stellarator coil optimization with in-the-loop finite-element calculations
+
+- **ID:** `2603.15240` · **primary:** `physics.plasm-ph` · **Submitted:** 2026-03-16
+- **Authors:** Alan A. Kaptanoglu, Pedro F. Gil
+- **abs:** https://arxiv.org/abs/2603.15240 · **PDF:** https://arxiv.org/pdf/2603.15240v1
+
+## 4577. Why the Valuable Capabilities of LLMs Are Precisely the Unexplainable Ones
+
+- **ID:** `2603.15238` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Quan Cheng
+- **abs:** https://arxiv.org/abs/2603.15238 · **PDF:** https://arxiv.org/pdf/2603.15238v2
+
+## 4578. Decomposing Probabilistic Scores: Reliability, Information Loss and Uncertainty
+
+- **ID:** `2603.15232` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Arthur Charpentier, Agathe Fernandes Machado
+- **abs:** https://arxiv.org/abs/2603.15232 · **PDF:** https://arxiv.org/pdf/2603.15232v2
+
+## 4579. SCAN: Sparse Circuit Anchor Interpretable Neuron for Lifelong Knowledge Editing
+
+- **ID:** `2603.15226` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Yuhuan Liu, Haitian Zhong, Xinyuan Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.15226 · **PDF:** https://arxiv.org/pdf/2603.15226v1
+
+## 4580. Bidirectional Chinese and English Passive Sentences Dataset for Machine Translation
+
+- **ID:** `2603.15227` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Xinyue Ma, Pol Pastells, Mireia Farrús, et al.
+- **abs:** https://arxiv.org/abs/2603.15227 · **PDF:** https://arxiv.org/pdf/2603.15227v1
+
+## 4581. ADV-0: Closed-Loop Min-Max Adversarial Training for Long-Tail Robustness in Autonomous Driving
+
+- **ID:** `2603.15221` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Tong Nie, Yihong Tang, Junlin He, et al.
+- **abs:** https://arxiv.org/abs/2603.15221 · **PDF:** https://arxiv.org/pdf/2603.15221v1
+
+## 4582. InterPol: De-anonymizing LM Arena via Interpolated Preference Learning
+
+- **ID:** `2603.15220` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Minsung Cho, Jaehyung Kim
+- **abs:** https://arxiv.org/abs/2603.15220 · **PDF:** https://arxiv.org/pdf/2603.15220v1
+
+## 4583. Towards Foundation Models for Consensus Rank Aggregation
+
+- **ID:** `2603.15218` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Yijun Jin, Simon Klüttermann, Chiara Balestra, et al.
+- **abs:** https://arxiv.org/abs/2603.15218 · **PDF:** https://arxiv.org/pdf/2603.15218v1
+
+## 4584. Modeling Matches as Language: A Generative Transformer Approach for Counterfactual Player Valuation in Football
+
+- **ID:** `2603.15212` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Miru Hong, Minho Lee, Geonhee Jo, et al.
+- **abs:** https://arxiv.org/abs/2603.15212 · **PDF:** https://arxiv.org/pdf/2603.15212v1
+
+## 4585. Efficient Document Parsing via Parallel Token Prediction
+
+- **ID:** `2603.15206` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Lei Li, Ze Zhao, Meng Li, et al.
+- **abs:** https://arxiv.org/abs/2603.15206 · **PDF:** https://arxiv.org/pdf/2603.15206v1
+
+## 4586. Geometric framework for biological evolution
+
+- **ID:** `2603.15198` · **primary:** `q-bio.PE` · **Submitted:** 2026-03-16
+- **Authors:** Vitaly Vanchurin
+- **abs:** https://arxiv.org/abs/2603.15198 · **PDF:** https://arxiv.org/pdf/2603.15198v1
+
+## 4587. Joint Optimization of Storage and Loading for High-Performance 3D Point Cloud Data Processing
+
+- **ID:** `2603.16945` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Ke Wang, Yanfei Cao, Xiangzhi Tao, et al.
+- **abs:** https://arxiv.org/abs/2603.16945 · **PDF:** https://arxiv.org/pdf/2603.16945v1
+
+## 4588. Massive Redundancy in Gradient Transport Enables Sparse Online Learning
+
+- **ID:** `2603.15195` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Aur Shalev Merin
+- **abs:** https://arxiv.org/abs/2603.15195 · **PDF:** https://arxiv.org/pdf/2603.15195v2
+
+## 4589. PiGRAND: Physics-informed Graph Neural Diffusion for Intelligent Additive Manufacturing
+
+- **ID:** `2603.15194` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Benjamin Uhrich, Tim Häntschel, Erhard Rahm
+- **abs:** https://arxiv.org/abs/2603.15194 · **PDF:** https://arxiv.org/pdf/2603.15194v1
+
+## 4590. The Sampling Complexity of Condorcet Winner Identification in Dueling Bandits
+
+- **ID:** `2603.15189` · **primary:** `stat.ML` · **Submitted:** 2026-03-16
+- **Authors:** El Mehdi Saad, Victor Thuot, Nicolas Verzelen
+- **abs:** https://arxiv.org/abs/2603.15189 · **PDF:** https://arxiv.org/pdf/2603.15189v1
+
+## 4591. Joint Routing and Model Pruning for Decentralized Federated Learning in Bandwidth-Constrained Multi-Hop Wireless Networks
+
+- **ID:** `2603.15188` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Xiaoyu He, Weicai Li, Tiejun Lv, et al.
+- **abs:** https://arxiv.org/abs/2603.15188 · **PDF:** https://arxiv.org/pdf/2603.15188v1
+
+## 4592. The Hrunting of AI: Where and How to Improve English Dialectal Fairness
+
+- **ID:** `2603.15187` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Wei Li, Adrian de Wynter
+- **abs:** https://arxiv.org/abs/2603.15187 · **PDF:** https://arxiv.org/pdf/2603.15187v1
+
+## 4593. What Matters for Scalable and Robust Learning in End-to-End Driving Planners?
+
+- **ID:** `2603.15185` · **primary:** `cs.RO` · **Submitted:** 2026-03-16
+- **Authors:** David Holtz, Niklas Hanselmann, Simon Doll, et al.
+- **abs:** https://arxiv.org/abs/2603.15185 · **PDF:** https://arxiv.org/pdf/2603.15185v1
+
+## 4594. CATFormer: When Continual Learning Meets Spiking Transformers With Dynamic Thresholds
+
+- **ID:** `2603.15184` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Vaishnavi Nagabhushana, Kartikay Agrawal, Ayon Borthakur
+- **abs:** https://arxiv.org/abs/2603.15184 · **PDF:** https://arxiv.org/pdf/2603.15184v1
+
+## 4595. Token Coherence: Adapting MESI Cache Protocols to Minimize Synchronization Overhead in Multi-Agent LLM Systems
+
+- **ID:** `2603.15183` · **primary:** `cs.DC` · **Submitted:** 2026-03-16
+- **Authors:** Vladyslav Parakhin
+- **abs:** https://arxiv.org/abs/2603.15183 · **PDF:** https://arxiv.org/pdf/2603.15183v1
+
+## 4596. Sequential Transport for Causal Mediation Analysis
+
+- **ID:** `2603.15182` · **primary:** `stat.ME` · **Submitted:** 2026-03-16
+- **Authors:** Agathe Fernandes Machado, Iryna Voitsitska, Arthur Charpentier, et al.
+- **abs:** https://arxiv.org/abs/2603.15182 · **PDF:** https://arxiv.org/pdf/2603.15182v2
+
+## 4597. Iterative Learning Control-Informed Reinforcement Learning for Batch Process Control
+
+- **ID:** `2603.15180` · **primary:** `eess.SY` · **Submitted:** 2026-03-16
+- **Authors:** Runze Lin, Ziqi Zhuo, Junghui Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.15180 · **PDF:** https://arxiv.org/pdf/2603.15180v1
+
+## 4598. Multimodal Connectome Fusion via Cross-Attention for Autism Spectrum Disorder Classification Using Graph Learning
+
+- **ID:** `2603.15168` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Ansar Rahman, Hassan Shojaee-Mend, Sepideh Hatamikia
+- **abs:** https://arxiv.org/abs/2603.15168 · **PDF:** https://arxiv.org/pdf/2603.15168v1
+
+## 4599. HindSight: Evaluating LLM-Generated Research Ideas via Future Impact
+
+- **ID:** `2603.15164` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Bo Jiang
+- **abs:** https://arxiv.org/abs/2603.15164 · **PDF:** https://arxiv.org/pdf/2603.15164v2
+
+## 4600. To See is Not to Master: Teaching LLMs to Use Private Libraries for Code Generation
+
+- **ID:** `2603.15159` · **primary:** `cs.SE` · **Submitted:** 2026-03-16
+- **Authors:** Yitong Zhang, Chengze Li, Ruize Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.15159 · **PDF:** https://arxiv.org/pdf/2603.15159v4
+
+## 4601. Point-Identification of a Robust Predictor Under Latent Shift with Imperfect Proxies
+
+- **ID:** `2603.15158` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Zahra Rahiminasab, Reza Soumi, Arto Klami, et al.
+- **abs:** https://arxiv.org/abs/2603.15158 · **PDF:** https://arxiv.org/pdf/2603.15158v1
+
+## 4602. Storage and selection of multiple chaotic attractors in minimal reservoir computers
+
+- **ID:** `2603.15155` · **primary:** `nlin.CD` · **Submitted:** 2026-03-16
+- **Authors:** Francesco Martinuzzi, Holger Kantz
+- **abs:** https://arxiv.org/abs/2603.15155 · **PDF:** https://arxiv.org/pdf/2603.15155v1
+
+## 4603. Accelerating Byzantine-Robust Distributed Learning with Compressed Communication via Double Momentum and Variance Reduction
+
+- **ID:** `2603.15144` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Yanghao Li, Changxin Liu, Yuhao Yi
+- **abs:** https://arxiv.org/abs/2603.15144 · **PDF:** https://arxiv.org/pdf/2603.15144v1
+
+## 4604. Safe Flow Q-Learning: Offline Safe Reinforcement Learning with Reachability-Based Flow Policies
+
+- **ID:** `2603.15136` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Mumuksh Tayal, Manan Tayal, Ravi Prakash
+- **abs:** https://arxiv.org/abs/2603.15136 · **PDF:** https://arxiv.org/pdf/2603.15136v1
+
+## 4605. Indirect Question Answering in English, German and Bavarian: A Challenging Task for High- and Low-Resource Languages Alike
+
+- **ID:** `2603.15130` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Miriam Winkler, Verena Blaschke, Barbara Plank
+- **abs:** https://arxiv.org/abs/2603.15130 · **PDF:** https://arxiv.org/pdf/2603.15130v1
+
+## 4606. Establishing Construct Validity in LLM Capability Benchmarks Requires Nomological Networks
+
+- **ID:** `2603.15121` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Timo Freiesleben
+- **abs:** https://arxiv.org/abs/2603.15121 · **PDF:** https://arxiv.org/pdf/2603.15121v1
+
+## 4607. MMKU-Bench: A Multimodal Update Benchmark for Diverse Visual Knowledge
+
+- **ID:** `2603.15117` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Baochen Fu, Yuntao Du, Cheng Chang, et al.
+- **abs:** https://arxiv.org/abs/2603.15117 · **PDF:** https://arxiv.org/pdf/2603.15117v1
+
+## 4608. Sampling-guided exploration of active feature selection policies
+
+- **ID:** `2603.15110` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Gabriel Bernardino, Anders Jonsson, Patrick Clarysse, et al.
+- **abs:** https://arxiv.org/abs/2603.15110 · **PDF:** https://arxiv.org/pdf/2603.15110v1
+
+## 4609. PrototypeNAS: Rapid Design of Deep Neural Networks for Microcontroller Units
+
+- **ID:** `2603.15106` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Mark Deutel, Simon Geis, Axel Plinge
+- **abs:** https://arxiv.org/abs/2603.15106 · **PDF:** https://arxiv.org/pdf/2603.15106v1
+
+## 4610. Towards Solving Polynomial-Objective Integer Programming with Hypergraph Neural Networks
+
+- **ID:** `2603.19318` · **primary:** `cs.NE` · **Submitted:** 2026-03-16
+- **Authors:** Minshuo Li, Yaoxin Wu, Pavel Troubil, et al.
+- **abs:** https://arxiv.org/abs/2603.19318 · **PDF:** https://arxiv.org/pdf/2603.19318v1
+
+## 4611. Bridging National and International Legal Data: Two Projects Based on the Japanese Legal Standard XML Schema for Comparative Law Studies
+
+- **ID:** `2603.15094` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Makoto Nakamura
+- **abs:** https://arxiv.org/abs/2603.15094 · **PDF:** https://arxiv.org/pdf/2603.15094v1
+
+## 4612. Trustworthy Koopman Operator Learning: Invariance Diagnostics and Error Bounds
+
+- **ID:** `2603.15091` · **primary:** `math.NA` · **Submitted:** 2026-03-16
+- **Authors:** Gustav Conradie, Nicolas Boullé, Jean-Christophe Loiseau, et al.
+- **abs:** https://arxiv.org/abs/2603.15091 · **PDF:** https://arxiv.org/pdf/2603.15091v1
+
+## 4613. Affordable Precision Agriculture: A Deployment-Oriented Review of Low-Cost, Low-Power Edge AI and TinyML for Resource-Constrained Farming Systems
+
+- **ID:** `2603.15085` · **primary:** `cs.ET` · **Submitted:** 2026-03-16
+- **Authors:** Riya Samanta, Bidyut Saha
+- **abs:** https://arxiv.org/abs/2603.15085 · **PDF:** https://arxiv.org/pdf/2603.15085v1
+
+## 4614. ReactMotion: Generating Reactive Listener Motions from Speaker Utterance
+
+- **ID:** `2603.15083` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Cheng Luo, Bizhu Wu, Bing Li, et al.
+- **abs:** https://arxiv.org/abs/2603.15083 · **PDF:** https://arxiv.org/pdf/2603.15083v1
+
+## 4615. Open Biomedical Knowledge Graphs at Scale: Construction, Federation, and AI Agent Access with Samyama Graph Database
+
+- **ID:** `2603.15080` · **primary:** `cs.DB` · **Submitted:** 2026-03-16
+- **Authors:** Madhulatha Mandarapu, Sandeep Kunkunuru
+- **abs:** https://arxiv.org/abs/2603.15080 · **PDF:** https://arxiv.org/pdf/2603.15080v3
+
+## 4616. Interpretable Classification of Time Series Using Euler Characteristic Surfaces
+
+- **ID:** `2603.15079` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Salam Rabindrajit Luwang, Sushovan Majhi, Vishal Mandal, et al.
+- **abs:** https://arxiv.org/abs/2603.15079 · **PDF:** https://arxiv.org/pdf/2603.15079v1
+
+## 4617. Generative Semantic HARQ: Latent-Space Text Retransmission and Combining
+
+- **ID:** `2603.15068` · **primary:** `eess.SP` · **Submitted:** 2026-03-16
+- **Authors:** Bin Han, Yulin Hu, Hans D. Schotten
+- **abs:** https://arxiv.org/abs/2603.15068 · **PDF:** https://arxiv.org/pdf/2603.15068v1
+
+## 4618. Writer-R1: Enhancing Generative Writing in LLMs via Memory-augmented Replay Policy Optimization
+
+- **ID:** `2603.15061` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Jihao Zhao, Shuaishuai Zu, Zhiyuan Ji, et al.
+- **abs:** https://arxiv.org/abs/2603.15061 · **PDF:** https://arxiv.org/pdf/2603.15061v1
+
+## 4619. Muon Converges under Heavy-Tailed Noise: Nonconvex Hölder-Smooth Empirical Risk Minimization
+
+- **ID:** `2603.15059` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Hideaki Iiduka
+- **abs:** https://arxiv.org/abs/2603.15059 · **PDF:** https://arxiv.org/pdf/2603.15059v2
+
+## 4620. Analyzing Error Sources in Global Feature Effect Estimation
+
+- **ID:** `2603.15057` · **primary:** `stat.ML` · **Submitted:** 2026-03-16
+- **Authors:** Timo Heiß, Coco Bögel, Bernd Bischl, et al.
+- **abs:** https://arxiv.org/abs/2603.15057 · **PDF:** https://arxiv.org/pdf/2603.15057v2
+
+## 4621. Spatio-temporal probabilistic forecast using MMAF-guided learning
+
+- **ID:** `2603.15055` · **primary:** `stat.ML` · **Submitted:** 2026-03-16
+- **Authors:** Leonardo Bardi, Imma Valentina Curato, Lorenzo Proietti
+- **abs:** https://arxiv.org/abs/2603.15055 · **PDF:** https://arxiv.org/pdf/2603.15055v1
+
+## 4622. Interference-Aware K-Step Reachable Communication in Multi-Agent Reinforcement Learning
+
+- **ID:** `2603.15054` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Ziyu Cheng, Jinsheng Ren, Zhouxian Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.15054 · **PDF:** https://arxiv.org/pdf/2603.15054v1
+
+## 4623. Thinking in Latents: Adaptive Anchor Refinement for Implicit Reasoning in LLMs
+
+- **ID:** `2603.15051` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Disha Sheshanarayana, Rajat Subhra Pal, Manjira Sinha, et al.
+- **abs:** https://arxiv.org/abs/2603.15051 · **PDF:** https://arxiv.org/pdf/2603.15051v1
+
+## 4624. CrossADR: enhancing adverse drug reactions prediction for combination pharmacotherapy with cross-layer feature integration and cross-level associative learning
+
+- **ID:** `2603.15047` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Y. Cheung
+- **abs:** https://arxiv.org/abs/2603.15047 · **PDF:** https://arxiv.org/pdf/2603.15047v1
+
+## 4625. AnoleVLA: Lightweight Vision-Language-Action Model with Deep State Space Models for Mobile Manipulation
+
+- **ID:** `2603.15046` · **primary:** `cs.RO` · **Submitted:** 2026-03-16
+- **Authors:** Yusuke Takagi, Motonari Kambara, Daichi Yashima, et al.
+- **abs:** https://arxiv.org/abs/2603.15046 · **PDF:** https://arxiv.org/pdf/2603.15046v1
+
+## 4626. Prompt Readiness Levels (PRL): a maturity scale and scoring framework for production grade prompt assets
+
+- **ID:** `2603.15044` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Sebastien Guinard
+- **abs:** https://arxiv.org/abs/2603.15044 · **PDF:** https://arxiv.org/pdf/2603.15044v1
+
+## 4627. A convolutional autoencoder and neural ODE framework for surrogate modeling of transient counterflow flames
+
+- **ID:** `2603.15038` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-03-16
+- **Authors:** Mert Yakup Baykan, Weitao Liu, Thorsten Zirwes, et al.
+- **abs:** https://arxiv.org/abs/2603.15038 · **PDF:** https://arxiv.org/pdf/2603.15038v1
+
+## 4628. Interpretable Predictability-Based AI Text Detection: A Replication Study
+
+- **ID:** `2603.15034` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Adam Skurla, Dominik Macko, Jakub Simko
+- **abs:** https://arxiv.org/abs/2603.15034 · **PDF:** https://arxiv.org/pdf/2603.15034v1
+
+## 4629. Rethinking Machine Unlearning: Models Designed to Forget via Key Deletion
+
+- **ID:** `2603.15033` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Sonia Laguna, Jorge da Silva Goncalves, Moritz Vandenhirtz, et al.
+- **abs:** https://arxiv.org/abs/2603.15033 · **PDF:** https://arxiv.org/pdf/2603.15033v3
+
+## 4630. Attention Residuals
+
+- **ID:** `2603.15031` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Kimi Team, Guangyu Chen, Yu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.15031 · **PDF:** https://arxiv.org/pdf/2603.15031v1
+
+## 4631. VTC-Bench: Evaluating Agentic Multimodal Models via Compositional Visual Tool Chaining
+
+- **ID:** `2603.15030` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Xuanyu Zhu, Yuhao Dong, Rundong Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.15030 · **PDF:** https://arxiv.org/pdf/2603.15030v2
+
+## 4632. Training-free Detection of Generated Videos via Spatial-Temporal Likelihoods
+
+- **ID:** `2603.15026` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Omer Ben Hayun, Roy Betser, Meir Yossef Levi, et al.
+- **abs:** https://arxiv.org/abs/2603.15026 · **PDF:** https://arxiv.org/pdf/2603.15026v2
+
+## 4633. Describing Agentic AI Systems with C4: Lessons from Industry Projects
+
+- **ID:** `2603.15021` · **primary:** `cs.SE` · **Submitted:** 2026-03-16
+- **Authors:** Andreas Rausch, Stefan Wittek
+- **abs:** https://arxiv.org/abs/2603.15021 · **PDF:** https://arxiv.org/pdf/2603.15021v1
+
+## 4634. MER-Bench: A Comprehensive Benchmark for Multimodal Meme Reappraisal
+
+- **ID:** `2603.15020` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Yiqi Nie, Fei Wang, Junjie Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.15020 · **PDF:** https://arxiv.org/pdf/2603.15020v1
+
+## 4635. Consequentialist Objectives and Catastrophe
+
+- **ID:** `2603.15017` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Henrik Marklund, Alex Infanger, Benjamin Van Roy
+- **abs:** https://arxiv.org/abs/2603.15017 · **PDF:** https://arxiv.org/pdf/2603.15017v2
+
+## 4636. Riemannian Motion Generation: A Unified Framework for Human Motion Representation and Generation via Riemannian Flow Matching
+
+- **ID:** `2603.15016` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Fangran Miao, Jian Huang, Ting Li
+- **abs:** https://arxiv.org/abs/2603.15016 · **PDF:** https://arxiv.org/pdf/2603.15016v1
+
+## 4637. TrajFlow: Nation-wide Pseudo GPS Trajectory Generation with Flow Matching Models
+
+- **ID:** `2603.15009` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Peiran Li, Jiawei Wang, Haoran Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.15009 · **PDF:** https://arxiv.org/pdf/2603.15009v1
+
+## 4638. Empowering Chemical Structures with Biological Insights for Scalable Phenotypic Virtual Screening
+
+- **ID:** `2603.15006` · **primary:** `q-bio.QM` · **Submitted:** 2026-03-16
+- **Authors:** Xiaoqing Lian, Pengsen Ma, Tengfeng Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.15006 · **PDF:** https://arxiv.org/pdf/2603.15006v1
+
+## 4639. Pretraining and Benchmarking Modern Encoders for Latvian
+
+- **ID:** `2603.15005` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Arturs Znotins
+- **abs:** https://arxiv.org/abs/2603.15005 · **PDF:** https://arxiv.org/pdf/2603.15005v1
+
+## 4640. MONET: Modeling and Optimization of neural NEtwork Training from Edge to Data Centers
+
+- **ID:** `2603.15002` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Jérémy Morlier, Robin Geens, Stef Cuyckens, et al.
+- **abs:** https://arxiv.org/abs/2603.15002 · **PDF:** https://arxiv.org/pdf/2603.15002v1
+
+## 4641. How Log-Barrier Helps Exploration in Policy Optimization
+
+- **ID:** `2603.15001` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Leonardo Cesani, Matteo Papini, Marcello Restelli
+- **abs:** https://arxiv.org/abs/2603.15001 · **PDF:** https://arxiv.org/pdf/2603.15001v1
+
+## 4642. OrgForge: A Multi-Agent Simulation Framework for Verifiable Synthetic Corporate Corpora
+
+- **ID:** `2603.14997` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Jeffrey Flynt
+- **abs:** https://arxiv.org/abs/2603.14997 · **PDF:** https://arxiv.org/pdf/2603.14997v1
+
+## 4643. Degrees, Levels, and Profiles of Contextuality
+
+- **ID:** `2603.26692` · **primary:** `quant-ph` · **Submitted:** 2026-03-16
+- **Authors:** Ehtibar N. Dzhafarov, Victor H. Cervantes
+- **abs:** https://arxiv.org/abs/2603.26692 · **PDF:** https://arxiv.org/pdf/2603.26692v2
+
+## 4644. Exposing Cross-Modal Consistency for Fake News Detection in Short-Form Videos
+
+- **ID:** `2603.14992` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Chong Tian, Yu Wang, Chenxu Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.14992 · **PDF:** https://arxiv.org/pdf/2603.14992v1
+
+## 4645. Beyond Benchmark Islands: Toward Representative Trustworthiness Evaluation for Agentic AI
+
+- **ID:** `2603.14987` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Jinhu Qi, Yifan Li, Minghao Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.14987 · **PDF:** https://arxiv.org/pdf/2603.14987v1
+
+## 4646. Why Agents Compromise Safety Under Pressure
+
+- **ID:** `2603.14975` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Hengle Jiang, Ke Tang
+- **abs:** https://arxiv.org/abs/2603.14975 · **PDF:** https://arxiv.org/pdf/2603.14975v1
+
+## 4647. Rethinking LLM Watermark Detection in Black-Box Settings: A Non-Intrusive Third-Party Framework
+
+- **ID:** `2603.14968` · **primary:** `cs.CR` · **Submitted:** 2026-03-16
+- **Authors:** Zhuoshang Wang, Yubing Ren, Yanan Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.14968 · **PDF:** https://arxiv.org/pdf/2603.14968v1
+
+## 4648. This Is Taking Too Long -- Investigating Time as a Proxy for Energy Consumption of LLMs
+
+- **ID:** `2603.15699` · **primary:** `cs.PF` · **Submitted:** 2026-03-16
+- **Authors:** Lars Krupp, Daniel Geißler, Francisco M. Calatrava-Nicolas, et al.
+- **abs:** https://arxiv.org/abs/2603.15699 · **PDF:** https://arxiv.org/pdf/2603.15699v1
+
+## 4649. Lightweight User-Personalization Method for Closed Split Computing
+
+- **ID:** `2603.14958` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Yuya Okada, Takayuki Nishio
+- **abs:** https://arxiv.org/abs/2603.14958 · **PDF:** https://arxiv.org/pdf/2603.14958v1
+
+## 4650. SFedHIFI: Fire Rate-Based Heterogeneous Information Fusion for Spiking Federated Learning
+
+- **ID:** `2603.14956` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Ran Tao, Qiugang Zhan, Shantian Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.14956 · **PDF:** https://arxiv.org/pdf/2603.14956v1
+
+## 4651. Learning Question-Aware Keyframe Selection with Synthetic Supervision for Video Question Answering
+
+- **ID:** `2603.14953` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Minchan Kwon, Hyounguk Shon, Junmo Kim
+- **abs:** https://arxiv.org/abs/2603.14953 · **PDF:** https://arxiv.org/pdf/2603.14953v1
+
+## 4652. Omni IIE Bench: Benchmarking the Practical Capabilities of Image Editing Models
+
+- **ID:** `2603.16944` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Yujia Yang, Yuanxiang Wang, Zhenyu Guan, et al.
+- **abs:** https://arxiv.org/abs/2603.16944 · **PDF:** https://arxiv.org/pdf/2603.16944v1
+
+## 4653. FairMed-XGB: A Bayesian-Optimised Multi-Metric Framework with Explainability for Demographic Equity in Critical Healthcare Data
+
+- **ID:** `2603.14947` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Mitul Goswami, Romit Chatterjee, Arif Ahmed Sekh
+- **abs:** https://arxiv.org/abs/2603.14947 · **PDF:** https://arxiv.org/pdf/2603.14947v1
+
+## 4654. Spiking Layer-Adaptive Magnitude-based Pruning
+
+- **ID:** `2603.14946` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Junqiao Wang, Zhehang Ye, Yuqi Ouyang
+- **abs:** https://arxiv.org/abs/2603.14946 · **PDF:** https://arxiv.org/pdf/2603.14946v1
+
+## 4655. Ultra-Early Prediction of Tipping Points: Integrating Dynamical Measures with Reservoir Computing
+
+- **ID:** `2603.14944` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Xin Li, Qunxi Zhu, Chengli Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.14944 · **PDF:** https://arxiv.org/pdf/2603.14944v1
+
+## 4656. KGS-GCN: Enhancing Sparse Skeleton Sensing via Kinematics-Driven Gaussian Splatting and Probabilistic Topology for Action Recognition
+
+- **ID:** `2603.16943` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Yuhan Chen, Yicui Shi, Guofa Li, et al.
+- **abs:** https://arxiv.org/abs/2603.16943 · **PDF:** https://arxiv.org/pdf/2603.16943v1
+
+## 4657. RS-WorldModel: a Unified Model for Remote Sensing Understanding and Future Sense Forecasting
+
+- **ID:** `2603.14941` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Linrui Xu, Zhongan Wang, Fei Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.14941 · **PDF:** https://arxiv.org/pdf/2603.14941v1
+
+## 4658. Intelligent Control of Differential Drive Robots Subject to Unmodeled Dynamics with EKF-based State Estimation
+
+- **ID:** `2603.14940` · **primary:** `eess.SY` · **Submitted:** 2026-03-16
+- **Authors:** Amos Alwala, Yuchen Hu, Gabriel da Silva Lima, et al.
+- **abs:** https://arxiv.org/abs/2603.14940 · **PDF:** https://arxiv.org/pdf/2603.14940v1
+
+## 4659. LLM as Graph Kernel: Rethinking Message Passing on Text-Rich Graphs
+
+- **ID:** `2603.14937` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Ying Zhang, Hang Yu, Haipeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.14937 · **PDF:** https://arxiv.org/pdf/2603.14937v1
+
+## 4660. Masked BRep Autoencoder via Hierarchical Graph Transformer
+
+- **ID:** `2603.14927` · **primary:** `cs.GR` · **Submitted:** 2026-03-16
+- **Authors:** Yifei Li, Kang Wu, Wenming Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.14927 · **PDF:** https://arxiv.org/pdf/2603.14927v2
+
+## 4661. Directional Routing in Transformers
+
+- **ID:** `2603.14923` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Kevin Taylor
+- **abs:** https://arxiv.org/abs/2603.14923 · **PDF:** https://arxiv.org/pdf/2603.14923v1
+
+## 4662. UNICORN: Ultrasound Nakagami Imaging via Score Matching and Adaptation for Assessing Hepatic Steatosis
+
+- **ID:** `2603.16942` · **primary:** `eess.IV` · **Submitted:** 2026-03-16
+- **Authors:** Kwanyoung Kim, Jaa-Yeon Lee, Youngjun Ko, et al.
+- **abs:** https://arxiv.org/abs/2603.16942 · **PDF:** https://arxiv.org/pdf/2603.16942v1
+
+## 4663. Bayesian Inference for Missing Physics
+
+- **ID:** `2603.14918` · **primary:** `stat.ML` · **Submitted:** 2026-03-16
+- **Authors:** Arno Strouwen
+- **abs:** https://arxiv.org/abs/2603.14918 · **PDF:** https://arxiv.org/pdf/2603.14918v1
+
+## 4664. Spectrogram features for audio and speech analysis
+
+- **ID:** `2603.14917` · **primary:** `eess.AS` · **Submitted:** 2026-03-16
+- **Authors:** Ian McLoughlin, Lam Pham, Yan Song, et al.
+- **abs:** https://arxiv.org/abs/2603.14917 · **PDF:** https://arxiv.org/pdf/2603.14917v1
+
+## 4665. Fine-tuning RoBERTa for CVE-to-CWE Classification: A 125M Parameter Model Competitive with LLMs
+
+- **ID:** `2603.14911` · **primary:** `cs.CR` · **Submitted:** 2026-03-16
+- **Authors:** Nikita Mosievskiy
+- **abs:** https://arxiv.org/abs/2603.14911 · **PDF:** https://arxiv.org/pdf/2603.14911v1
+
+## 4666. Machine learning for sustainable geoenergy: uncertainty, physics and decision-ready inference
+
+- **ID:** `2603.14907` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-03-16
+- **Authors:** Hannah P. Menke, Ahmed H. Elsheikh, Lingli Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.14907 · **PDF:** https://arxiv.org/pdf/2603.14907v1
+
+## 4667. ExPosST: Explicit Positioning with Adaptive Masking for LLM-Based Simultaneous Machine Translation
+
+- **ID:** `2603.14903` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Yuzhe Shang, Pengzhi Gao, Yazheng Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.14903 · **PDF:** https://arxiv.org/pdf/2603.14903v2
+
+## 4668. The impact of machine learning forecasting on strategic decision-making for Bike Sharing Systems
+
+- **ID:** `2603.14901` · **primary:** `math.OC` · **Submitted:** 2026-03-16
+- **Authors:** Enrico Angelelli, Andrea Mor, Carlotta Orsenigo, et al.
+- **abs:** https://arxiv.org/abs/2603.14901 · **PDF:** https://arxiv.org/pdf/2603.14901v1
+
+## 4669. SpatialPoint: Spatial-aware Point Prediction for Embodied Localization
+
+- **ID:** `2603.26690` · **primary:** `cs.RO` · **Submitted:** 2026-03-16
+- **Authors:** Qiming Zhu, Zhirui Fang, Tianming Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.26690 · **PDF:** https://arxiv.org/pdf/2603.26690v1
+
+## 4670. Photonic Quantum-Enhanced Knowledge Distillation
+
+- **ID:** `2603.14898` · **primary:** `quant-ph` · **Submitted:** 2026-03-16
+- **Authors:** Kuan-Cheng Chen, Shang Yu, Chen-Yu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.14898 · **PDF:** https://arxiv.org/pdf/2603.14898v1
+
+## 4671. BiTro: Bidirectional Transfer Learning Enhances Bulk and Spatial Transcriptomics Prediction in Cancer Pathological Images
+
+- **ID:** `2603.14897` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Jingkun Yu, Guangkai Shang, Changtao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.14897 · **PDF:** https://arxiv.org/pdf/2603.14897v1
+
+## 4672. Informative Perturbation Selection for Uncertainty-Aware Post-hoc Explanations
+
+- **ID:** `2603.14894` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Sumedha Chugh, Ranjitha Prasad, Nazreen Shah
+- **abs:** https://arxiv.org/abs/2603.14894 · **PDF:** https://arxiv.org/pdf/2603.14894v2
+
+## 4673. LLMs as Signal Detectors: Sensitivity, Bias, and the Temperature-Criterion Analogy
+
+- **ID:** `2603.14893` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Jon-Paul Cacioli
+- **abs:** https://arxiv.org/abs/2603.14893 · **PDF:** https://arxiv.org/pdf/2603.14893v1
+
+## 4674. Decision-Level Ordinal Modeling for Multimodal Essay Scoring with Large Language Models
+
+- **ID:** `2603.14891` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Han Zhang, Jiamin Su, Li liu
+- **abs:** https://arxiv.org/abs/2603.14891 · **PDF:** https://arxiv.org/pdf/2603.14891v1
+
+## 4675. Modeling and Benchmarking Spoken Dialogue Rewards with Modality and Colloquialness
+
+- **ID:** `2603.14889` · **primary:** `eess.AS` · **Submitted:** 2026-03-16
+- **Authors:** Jingyu Lu, Yuhan Wang, Fan Zhuo, et al.
+- **abs:** https://arxiv.org/abs/2603.14889 · **PDF:** https://arxiv.org/pdf/2603.14889v1
+
+## 4676. Customizing ChatGPT for Second Language Speaking Practice: Genuine Support or Just a Marketing Gimmick?
+
+- **ID:** `2603.14884` · **primary:** `cs.HC` · **Submitted:** 2026-03-16
+- **Authors:** Fanfei Meng
+- **abs:** https://arxiv.org/abs/2603.14884 · **PDF:** https://arxiv.org/pdf/2603.14884v1
+
+## 4677. Seismic full-waveform inversion based on a physics-driven generative adversarial network
+
+- **ID:** `2603.14879` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Xinyi Zhang, Caiyun Liu, Jie Xiong, et al.
+- **abs:** https://arxiv.org/abs/2603.14879 · **PDF:** https://arxiv.org/pdf/2603.14879v1
+
+## 4678. A Hybrid AI and Rule-Based Decision Support System for Disease Diagnosis and Management Using Labs
+
+- **ID:** `2603.14876` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Muhammad Hammad Maqsood, Mubashir Sajid, Khubaib Ahmed, et al.
+- **abs:** https://arxiv.org/abs/2603.14876 · **PDF:** https://arxiv.org/pdf/2603.14876v1
+
+## 4679. Tackling Over-smoothing on Hypergraphs: A Ricci Flow-guided Neural Diffusion Approach
+
+- **ID:** `2603.15696` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Mengyao Zhou, Zhiheng Zhou, Xiao Han, et al.
+- **abs:** https://arxiv.org/abs/2603.15696 · **PDF:** https://arxiv.org/pdf/2603.15696v1
+
+## 4680. Developing an English-Efik Corpus and Machine Translation System for Digitization Inclusion
+
+- **ID:** `2603.14873` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Offiong Bassey Edet, Mbuotidem Sunday Awak, Emmanuel Oyo-Ita, et al.
+- **abs:** https://arxiv.org/abs/2603.14873 · **PDF:** https://arxiv.org/pdf/2603.14873v1
+
+## 4681. IgPose: A Generative Data-Augmented Pipeline for Robust Immunoglobulin-Antigen Binding Prediction
+
+- **ID:** `2603.14870` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Tien-Cuong Bui, Injae Chung, Wonjun Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.14870 · **PDF:** https://arxiv.org/pdf/2603.14870v1
+
+## 4682. A Self-Evolving Defect Detection Framework for Industrial Photovoltaic Systems
+
+- **ID:** `2603.14869` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Haoyu He, Yu Duan, Wenzhen Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.14869 · **PDF:** https://arxiv.org/pdf/2603.14869v1
+
+## 4683. Sample-Efficient Hypergradient Estimation for Decentralized Bi-Level Reinforcement Learning
+
+- **ID:** `2603.14867` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Mikoto Kudo, Takumi Tanabe, Akifumi Wachi, et al.
+- **abs:** https://arxiv.org/abs/2603.14867 · **PDF:** https://arxiv.org/pdf/2603.14867v3
+
+## 4684. Shopping Companion: A Memory-Augmented LLM Agent for Real-World E-Commerce Tasks
+
+- **ID:** `2603.14864` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Zijian Yu, Kejun Xiao, Huaipeng Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.14864 · **PDF:** https://arxiv.org/pdf/2603.14864v1
+
+## 4685. A Score Filter Enhanced Data Assimilation Framework for Data-Driven Dynamical Systems
+
+- **ID:** `2603.14863` · **primary:** `math.DS` · **Submitted:** 2026-03-16
+- **Authors:** Jingqiao Tang, Ryan Bausback, Feng Bao, et al.
+- **abs:** https://arxiv.org/abs/2603.14863 · **PDF:** https://arxiv.org/pdf/2603.14863v1
+
+## 4686. Video Detector: A Dual-Phase Vision-Based System for Real-Time Traffic Intersection Control and Intelligent Transportation Analysis
+
+- **ID:** `2603.14861` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Mustafa Fatih Şen, Halûk Gümüşkaya, Şenol Pazar
+- **abs:** https://arxiv.org/abs/2603.14861 · **PDF:** https://arxiv.org/pdf/2603.14861v1
+
+## 4687. Architecture-Agnostic Feature Synergy for Universal Defense Against Heterogeneous Generative Threats
+
+- **ID:** `2603.14860` · **primary:** `cs.CR` · **Submitted:** 2026-03-16
+- **Authors:** Bingxue Zhang, Yang Gao, Feida Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.14860 · **PDF:** https://arxiv.org/pdf/2603.14860v1
+
+## 4688. PCodeTrans: Translate Decompiled Pseudocode to Compilable and Executable Equivalent
+
+- **ID:** `2603.14855` · **primary:** `cs.SE` · **Submitted:** 2026-03-16
+- **Authors:** Yuxin Cui, Zeyu Gao, Shuxian He, et al.
+- **abs:** https://arxiv.org/abs/2603.14855 · **PDF:** https://arxiv.org/pdf/2603.14855v1
+
+## 4689. Lost in Aggregation: On a Fundamental Expressivity Limit of Message-Passing Graph Neural Networks
+
+- **ID:** `2603.14846` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Eran Rosenbluth
+- **abs:** https://arxiv.org/abs/2603.14846 · **PDF:** https://arxiv.org/pdf/2603.14846v2
+
+## 4690. Integrating Weather Foundation Model and Satellite to Enable Fine-Grained Solar Irradiance Forecasting
+
+- **ID:** `2603.14845` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Ziqing Ma, Kai Ying, Xinyue Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.14845 · **PDF:** https://arxiv.org/pdf/2603.14845v2
+
+## 4691. ContiGuard: A Framework for Continual Toxicity Detection Against Evolving Evasive Perturbations
+
+- **ID:** `2603.14843` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Hankun Kang, Xin Miao, Jianhao Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.14843 · **PDF:** https://arxiv.org/pdf/2603.14843v1
+
+## 4692. Real-Time Driver Safety Scoring Through Inverse Crash Probability Modeling
+
+- **ID:** `2603.14841` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Joyjit Roy, Samaresh Kumar Singh, Sushanta Das
+- **abs:** https://arxiv.org/abs/2603.14841 · **PDF:** https://arxiv.org/pdf/2603.14841v2
+
+## 4693. The Impact of Ideological Discourses in RAG: A Case Study with COVID-19 Treatments
+
+- **ID:** `2603.14838` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Elmira Salari, Maria Claudia Nunes Delfino, Hazem Amamou, et al.
+- **abs:** https://arxiv.org/abs/2603.14838 · **PDF:** https://arxiv.org/pdf/2603.14838v1
+
+## 4694. Ablate and Rescue: A Causal Analysis of Residual Stream Hyper-Connections
+
+- **ID:** `2603.14833` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** William Peng, Josheev Rai, Kevin Tseng, et al.
+- **abs:** https://arxiv.org/abs/2603.14833 · **PDF:** https://arxiv.org/pdf/2603.14833v1
+
+## 4695. Halfway to 3D: Ensembling 2.5D and 3D Models for Robust COVID-19 CT Diagnosis
+
+- **ID:** `2603.14832` · **primary:** `eess.IV` · **Submitted:** 2026-03-16
+- **Authors:** Tuan-Anh Yang, Bao V. Q. Bui, Chanh-Quang Vo-Van, et al.
+- **abs:** https://arxiv.org/abs/2603.14832 · **PDF:** https://arxiv.org/pdf/2603.14832v2
+
+## 4696. Neural Networks as Local-to-Global Computations
+
+- **ID:** `2603.14831` · **primary:** `math.AT` · **Submitted:** 2026-03-16
+- **Authors:** Vicente Bosca, Robert Ghrist
+- **abs:** https://arxiv.org/abs/2603.14831 · **PDF:** https://arxiv.org/pdf/2603.14831v3
+
+## 4697. Dataset Distillation Efficiently Encodes Low-Dimensional Representations from Gradient-Based Learning of Non-Linear Tasks
+
+- **ID:** `2603.14830` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Yuri Kinoshita, Naoki Nishikawa, Taro Toyoizumi
+- **abs:** https://arxiv.org/abs/2603.14830 · **PDF:** https://arxiv.org/pdf/2603.14830v2
+
+## 4698. Two Birds, One Projection: Harmonizing Safety and Utility in LVLMs via Inference-time Feature Projection
+
+- **ID:** `2603.14825` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Yewon Han, Yumin Seol, EunGyung Kong, et al.
+- **abs:** https://arxiv.org/abs/2603.14825 · **PDF:** https://arxiv.org/pdf/2603.14825v1
+
+## 4699. Planning as Goal Recognition: Deriving Heuristics from Intention Models -- Extended Version
+
+- **ID:** `2603.14824` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Giacomo Rosa, Jean Honorio, Nir Lipovetzky, et al.
+- **abs:** https://arxiv.org/abs/2603.14824 · **PDF:** https://arxiv.org/pdf/2603.14824v2
+
+## 4700. RAZOR: Ratio-Aware Layer Editing for Targeted Unlearning in Vision Transformers and Diffusion Models
+
+- **ID:** `2603.14819` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Ravi Ranjan, Utkarsh Grover, Xiaomin Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.14819 · **PDF:** https://arxiv.org/pdf/2603.14819v1
+
+## 4701. SimCert: Probabilistic Certification for Behavioral Similarity in Deep Neural Network Compression
+
+- **ID:** `2603.14818` · **primary:** `cs.SE` · **Submitted:** 2026-03-16
+- **Authors:** Jingyang Li, Fu Song, Guoqiang Li
+- **abs:** https://arxiv.org/abs/2603.14818 · **PDF:** https://arxiv.org/pdf/2603.14818v1
+
+## 4702. Fold-CP: A Context Parallelism Framework for Biomolecular Modeling
+
+- **ID:** `2603.14806` · **primary:** `q-bio.QM` · **Submitted:** 2026-03-16
+- **Authors:** Dejun Lin, Simon Chu, Vishanth Iyer, et al.
+- **abs:** https://arxiv.org/abs/2603.14806 · **PDF:** https://arxiv.org/pdf/2603.14806v1
+
+## 4703. Knowledge Activation: AI Skills as the Institutional Knowledge Primitive for Agentic Software Development
+
+- **ID:** `2603.14805` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Gal Bakal
+- **abs:** https://arxiv.org/abs/2603.14805 · **PDF:** https://arxiv.org/pdf/2603.14805v1
+
+## 4704. VorTEX: Various overlap ratio for Target speech EXtraction
+
+- **ID:** `2603.14803` · **primary:** `cs.SD` · **Submitted:** 2026-03-16
+- **Authors:** Ro-hoon Oh, Jihwan Seol, Bugeun Kim
+- **abs:** https://arxiv.org/abs/2603.14803 · **PDF:** https://arxiv.org/pdf/2603.14803v3
+
+## 4705. OpenReservoirComputing: GPU-Accelerated Reservoir Computing in JAX
+
+- **ID:** `2603.14802` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Jan Williams, Dima Tretiak, Steven L. Brunton, et al.
+- **abs:** https://arxiv.org/abs/2603.14802 · **PDF:** https://arxiv.org/pdf/2603.14802v1
+
+## 4706. Universe Routing: Why Self-Evolving Agents Need Epistemic Control
+
+- **ID:** `2603.14799` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Zhaohui Geoffrey Wang
+- **abs:** https://arxiv.org/abs/2603.14799 · **PDF:** https://arxiv.org/pdf/2603.14799v1
+
+## 4707. Preconditioned One-Step Generative Modeling for Bayesian Inverse Problems in Function Spaces
+
+- **ID:** `2603.14798` · **primary:** `stat.ML` · **Submitted:** 2026-03-16
+- **Authors:** Zilan Cheng, Li-Lian Wang, Zhongjian Wang
+- **abs:** https://arxiv.org/abs/2603.14798 · **PDF:** https://arxiv.org/pdf/2603.14798v1
+
+## 4708. Multi-Task Genetic Algorithm with Multi-Granularity Encoding for Protein-Nucleotide Binding Site Prediction
+
+- **ID:** `2603.14797` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Yiming Gao, Liuyi Xu, Pengshan Cui, et al.
+- **abs:** https://arxiv.org/abs/2603.14797 · **PDF:** https://arxiv.org/pdf/2603.14797v1
+
+## 4709. Face-to-Face: A Video Dataset for Multi-Person Interaction Modeling
+
+- **ID:** `2603.14794` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Ernie Chu, Vishal M. Patel
+- **abs:** https://arxiv.org/abs/2603.14794 · **PDF:** https://arxiv.org/pdf/2603.14794v2
+
+## 4710. GARCH-FIS: A Hybrid Forecasting Model with Dynamic Volatility-Driven Parameter Adaptation
+
+- **ID:** `2603.14793` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Wen-Jing Li, Da-Qing Zhang
+- **abs:** https://arxiv.org/abs/2603.14793 · **PDF:** https://arxiv.org/pdf/2603.14793v1
+
+## 4711. LaPro-DTA: Latent Dual-View Drug Representations and Salient Protein Feature Extraction for Generalizable Drug--Target Affinity Prediction
+
+- **ID:** `2603.14792` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Zihan Dun, Liuyi Xu, An-Yang Lu, et al.
+- **abs:** https://arxiv.org/abs/2603.14792 · **PDF:** https://arxiv.org/pdf/2603.14792v1
+
+## 4712. BadLLM-TG: A Backdoor Defender powered by LLM Trigger Generator
+
+- **ID:** `2603.15692` · **primary:** `cs.CR` · **Submitted:** 2026-03-16
+- **Authors:** Ruyi Zhang, Heng Gao, Songlei Jian, et al.
+- **abs:** https://arxiv.org/abs/2603.15692 · **PDF:** https://arxiv.org/pdf/2603.15692v1
+
+## 4713. Orthogonal Subspace Clustering: Enhancing High-Dimensional Data Analysis through Adaptive Dimensionality Reduction and Efficient Clustering
+
+- **ID:** `2603.14783` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Qing-Yuan Wen, Da-Qing Zhang
+- **abs:** https://arxiv.org/abs/2603.14783 · **PDF:** https://arxiv.org/pdf/2603.14783v1
+
+## 4714. Information Asymmetry across Language Varieties: A Case Study on Cantonese-Mandarin and Bavarian-German QA
+
+- **ID:** `2603.14782` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Renhao Pei, Siyao Peng, Verena Blaschke, et al.
+- **abs:** https://arxiv.org/abs/2603.14782 · **PDF:** https://arxiv.org/pdf/2603.14782v1
+
+## 4715. Vietnamese Automatic Speech Recognition: A Revisit
+
+- **ID:** `2603.14779` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Thi Vu, Linh The Nguyen, Dat Quoc Nguyen
+- **abs:** https://arxiv.org/abs/2603.14779 · **PDF:** https://arxiv.org/pdf/2603.14779v1
+
+## 4716. $p^2$RAG: Privacy-Preserving RAG Service Supporting Arbitrary Top-$k$ Retrieval
+
+- **ID:** `2603.14778` · **primary:** `cs.CR` · **Submitted:** 2026-03-16
+- **Authors:** Yulong Ming, Mingyue Wang, Jijia Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.14778 · **PDF:** https://arxiv.org/pdf/2603.14778v1
+
+## 4717. A Lightweight, Transferable, and Self-Adaptive Framework for Intelligent DC Arc-Fault Detection in Photovoltaic Systems
+
+- **ID:** `2603.25749` · **primary:** `eess.SP` · **Submitted:** 2026-03-16
+- **Authors:** Xiaoke Yang, Long Gao, Haoyu He, et al.
+- **abs:** https://arxiv.org/abs/2603.25749 · **PDF:** https://arxiv.org/pdf/2603.25749v1
+
+## 4718. HO-SFL: Hybrid-Order Split Federated Learning with Backprop-Free Clients and Dimension-Free Aggregation
+
+- **ID:** `2603.14773` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Qiyuan Chen, Xian Wu, Yi Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.14773 · **PDF:** https://arxiv.org/pdf/2603.14773v1
+
+## 4719. OpenHospital: A Thing-in-itself Arena for Evolving and Benchmarking LLM-based Collective Intelligence
+
+- **ID:** `2603.14771` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Peigen Liu, Rui Ding, Yuren Mao, et al.
+- **abs:** https://arxiv.org/abs/2603.14771 · **PDF:** https://arxiv.org/pdf/2603.14771v2
+
+## 4720. POLCA: Stochastic Generative Optimization with LLM
+
+- **ID:** `2603.14769` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Xuanfei Ren, Allen Nie, Tengyang Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.14769 · **PDF:** https://arxiv.org/pdf/2603.14769v1
+
+## 4721. Understanding the geometry of deep learning with decision boundary volume
+
+- **ID:** `2603.14768` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Matthew Burfitt, Jacek Brodzki, Pawel Dłotko
+- **abs:** https://arxiv.org/abs/2603.14768 · **PDF:** https://arxiv.org/pdf/2603.14768v1
+
+## 4722. Investigating the Impact of Speech Enhancement on Audio Deepfake Detection in Noisy Environments
+
+- **ID:** `2603.14767` · **primary:** `cs.SD` · **Submitted:** 2026-03-16
+- **Authors:** Anacin, Angela, Shruti Kshirsagar, et al.
+- **abs:** https://arxiv.org/abs/2603.14767 · **PDF:** https://arxiv.org/pdf/2603.14767v1
+
+## 4723. Topology-Preserving Data Augmentation for Ring-Type Polygon Annotations
+
+- **ID:** `2603.14764` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Sudip Laudari, Sang Hun Baek
+- **abs:** https://arxiv.org/abs/2603.14764 · **PDF:** https://arxiv.org/pdf/2603.14764v2
+
+## 4724. Online Learning for Supervisory Switching Control
+
+- **ID:** `2603.14762` · **primary:** `math.OC` · **Submitted:** 2026-03-16
+- **Authors:** Haoyuan Sun, Ali Jadbabaie
+- **abs:** https://arxiv.org/abs/2603.14762 · **PDF:** https://arxiv.org/pdf/2603.14762v2
+
+## 4725. BrainBench: Exposing the Commonsense Reasoning Gap in Large Language Models
+
+- **ID:** `2603.14761` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Yuzhe Tang
+- **abs:** https://arxiv.org/abs/2603.14761 · **PDF:** https://arxiv.org/pdf/2603.14761v1
+
+## 4726. Towards Privacy-Preserving Machine Translation at the Inference Stage: A New Task and Benchmark
+
+- **ID:** `2603.14756` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Wei Shao, Lemao Liu, Yinqiao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.14756 · **PDF:** https://arxiv.org/pdf/2603.14756v1
+
+## 4727. Learning Constituent Headedness
+
+- **ID:** `2603.14755` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Zeyao Qi, Yige Chen, KyungTae Lim, et al.
+- **abs:** https://arxiv.org/abs/2603.14755 · **PDF:** https://arxiv.org/pdf/2603.14755v1
+
+## 4728. CAMD: Coverage-Aware Multimodal Decoding for Efficient Reasoning of Multimodal Large Language Models
+
+- **ID:** `2603.14745` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Huijie Guo, Jingyao Wang, Lingyu Si, et al.
+- **abs:** https://arxiv.org/abs/2603.14745 · **PDF:** https://arxiv.org/pdf/2603.14745v1
+
+## 4729. TrajMamba: An Ego-Motion-Guided Mamba Model for Pedestrian Trajectory Prediction from an Egocentric Perspective
+
+- **ID:** `2603.14739` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Yusheng Peng, Gaofeng Zhang, Liping Zheng
+- **abs:** https://arxiv.org/abs/2603.14739 · **PDF:** https://arxiv.org/pdf/2603.14739v1
+
+## 4730. Gauge-Equivariant Intrinsic Neural Operators for Geometry-Consistent Learning of Elliptic PDE Maps
+
+- **ID:** `2603.14734` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Pengcheng Cheng
+- **abs:** https://arxiv.org/abs/2603.14734 · **PDF:** https://arxiv.org/pdf/2603.14734v1
+
+## 4731. Criterion-referenceability determines LLM-as-a-judge validity across physics assessment formats
+
+- **ID:** `2603.14732` · **primary:** `physics.ed-ph` · **Submitted:** 2026-03-16
+- **Authors:** Will Yeadon, Tom Hardy, Paul Mackay, et al.
+- **abs:** https://arxiv.org/abs/2603.14732 · **PDF:** https://arxiv.org/pdf/2603.14732v1
+
+## 4732. Loosely-Structured Software: Engineering Context, Structure, and Evolution Entropy in Runtime-Rewired Multi-Agent Systems
+
+- **ID:** `2603.15690` · **primary:** `cs.SE` · **Submitted:** 2026-03-16
+- **Authors:** Weihao Zhang, Yitong Zhou, Huanyu Qu, et al.
+- **abs:** https://arxiv.org/abs/2603.15690 · **PDF:** https://arxiv.org/pdf/2603.15690v1
+
+## 4733. GNNVerifier: Graph-based Verifier for LLM Task Planning
+
+- **ID:** `2603.14730` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Yu Hao, Qiuyu Wang, Cheng Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.14730 · **PDF:** https://arxiv.org/pdf/2603.14730v2
+
+## 4734. DeFRiS: Silo-Cooperative IoT Applications Scheduling via Decentralized Federated Reinforcement Learning
+
+- **ID:** `2603.14729` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Zhiyu Wang, Mohammad Goudarzi, Mingming Gong, et al.
+- **abs:** https://arxiv.org/abs/2603.14729 · **PDF:** https://arxiv.org/pdf/2603.14729v1
+
+## 4735. GameUIAgent: An LLM-Powered Framework for Automated Game UI Design with Structured Intermediate Representation
+
+- **ID:** `2603.14724` · **primary:** `cs.AI` · **Submitted:** 2026-03-16
+- **Authors:** Wei Zeng, Fengwei An, Zhen Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.14724 · **PDF:** https://arxiv.org/pdf/2603.14724v1
+
+## 4736. Beyond Creed: A Non-Identity Safety Condition A Strong Empirical Alternative to Identity Framing in Low-Data LoRA Fine-Tuning
+
+- **ID:** `2603.14723` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Xinran Zhang
+- **abs:** https://arxiv.org/abs/2603.14723 · **PDF:** https://arxiv.org/pdf/2603.14723v1
+
+## 4737. Multimodal Deep Learning for Early Prediction of Patient Deterioration in the ICU: Integrating Time-Series EHR Data with Clinical Notes
+
+- **ID:** `2603.14719` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Binesh Sadanandan
+- **abs:** https://arxiv.org/abs/2603.14719 · **PDF:** https://arxiv.org/pdf/2603.14719v1
+
+## 4738. Training-Free Generation of Protein Sequences from Small Family Alignments via Stochastic Attention
+
+- **ID:** `2603.14717` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Jeffrey D. Varner
+- **abs:** https://arxiv.org/abs/2603.14717 · **PDF:** https://arxiv.org/pdf/2603.14717v1
+
+## 4739. Towards Next-Generation LLM Training: From the Data-Centric Perspective
+
+- **ID:** `2603.14712` · **primary:** `cs.CL` · **Submitted:** 2026-03-16
+- **Authors:** Hao Liang, Zhengyang Zhao, Zhaoyang Han, et al.
+- **abs:** https://arxiv.org/abs/2603.14712 · **PDF:** https://arxiv.org/pdf/2603.14712v1
+
+## 4740. Cross-RAG: Zero-Shot Retrieval-Augmented Time Series Forecasting via Cross-Attention
+
+- **ID:** `2603.14709` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Seunghan Lee, Jaehoon Lee, Jun Seo, et al.
+- **abs:** https://arxiv.org/abs/2603.14709 · **PDF:** https://arxiv.org/pdf/2603.14709v1
+
+## 4741. Visual Confused Deputy: Exploiting and Defending Perception Failures in Computer-Using Agents
+
+- **ID:** `2603.14707` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Xunzhuo Liu, Bowei He, Xue Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.14707 · **PDF:** https://arxiv.org/pdf/2603.14707v1
+
+## 4742. Transition Flow Matching
+
+- **ID:** `2603.15689` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Chenrui Ma
+- **abs:** https://arxiv.org/abs/2603.15689 · **PDF:** https://arxiv.org/pdf/2603.15689v1
+
+## 4743. AdapterTune: Zero-Initialized Low-Rank Adapters for Frozen Vision Transformers
+
+- **ID:** `2603.14706` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Salim Khazem
+- **abs:** https://arxiv.org/abs/2603.14706 · **PDF:** https://arxiv.org/pdf/2603.14706v1
+
+## 4744. Chain-of-Trajectories: Unlocking the Intrinsic Generative Optimality of Diffusion Models via Graph-Theoretic Planning
+
+- **ID:** `2603.14704` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Ping Chen, Xiang Liu, Xingpeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.14704 · **PDF:** https://arxiv.org/pdf/2603.14704v1
+
+## 4745. Beyond Local Code Optimization: Multi-Agent Reasoning for Software System Optimization
+
+- **ID:** `2603.14703` · **primary:** `cs.SE` · **Submitted:** 2026-03-16
+- **Authors:** Huiyun Peng, Parth Vinod Patil, Antonio Zhong Qiu, et al.
+- **abs:** https://arxiv.org/abs/2603.14703 · **PDF:** https://arxiv.org/pdf/2603.14703v1
+
+## 4746. Design Space of Self--Consistent Electrostatic Machine Learning Interatomic Potentials
+
+- **ID:** `2603.14700` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-16
+- **Authors:** William J. Baldwin, Ilyes Batatia, Martin Vondrák, et al.
+- **abs:** https://arxiv.org/abs/2603.14700 · **PDF:** https://arxiv.org/pdf/2603.14700v1
+
+## 4747. Scaling Autoregressive Models for Lattice Thermodynamics
+
+- **ID:** `2603.14695` · **primary:** `cond-mat.stat-mech` · **Submitted:** 2026-03-16
+- **Authors:** Xiaochen Du, Juno Nam, Sulin Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.14695 · **PDF:** https://arxiv.org/pdf/2603.14695v1
+
+## 4748. Robust Building Damage Detection in Cross-Disaster Settings Using Domain Adaptation
+
+- **ID:** `2603.14694` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Asmae Mouradi, Shruti Kshirsagar
+- **abs:** https://arxiv.org/abs/2603.14694 · **PDF:** https://arxiv.org/pdf/2603.14694v1
+
+## 4749. Applications of Intuitionistic Temporal Logic to Temporal Answer Set Programming
+
+- **ID:** `2603.14692` · **primary:** `cs.LO` · **Submitted:** 2026-03-16
+- **Authors:** Pedro Cabalar, Martín Diéguez, David Fernández-Duque, et al.
+- **abs:** https://arxiv.org/abs/2603.14692 · **PDF:** https://arxiv.org/pdf/2603.14692v1
+
+## 4750. AgentTrace: Causal Graph Tracing for Root Cause Analysis in Deployed Multi-Agent Systems
+
+- **ID:** `2603.14688` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Zhaohui Geoffrey Wang
+- **abs:** https://arxiv.org/abs/2603.14688 · **PDF:** https://arxiv.org/pdf/2603.14688v2
+
+## 4751. MVHOI: Bridge Multi-view Condition to Complex Human-Object Interaction Video Reenactment via 3D Foundation Model
+
+- **ID:** `2603.14686` · **primary:** `cs.CV` · **Submitted:** 2026-03-16
+- **Authors:** Jinguang Tong, Jinbo Wu, Kaisiyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.14686 · **PDF:** https://arxiv.org/pdf/2603.14686v1
+
+## 4752. \texttt{BayesBreak}: Generalized Hierarchical Bayesian Segmentation with Irregular Designs, Multi-Sample Hierarchies, and Grouped/Latent-Group Designs
+
+- **ID:** `2603.14681` · **primary:** `cs.LG` · **Submitted:** 2026-03-16
+- **Authors:** Omid Shams Solari
+- **abs:** https://arxiv.org/abs/2603.14681 · **PDF:** https://arxiv.org/pdf/2603.14681v1
+
+## 4753. Scalable Text-Embedding-informed Cognitive Diagnosis of Large Language Models
+
+- **ID:** `2603.14676` · **primary:** `stat.ME` · **Submitted:** 2026-03-16
+- **Authors:** Jia Liu, Zhiyu Xu, Yuqi Gu
+- **abs:** https://arxiv.org/abs/2603.14676 · **PDF:** https://arxiv.org/pdf/2603.14676v1
+
+## 4754. Computational Analysis of Semantic Connections Between Herman Melville Reading and Writing
+
+- **ID:** `2603.14674` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Nudrat Habib, Elisa Barney Smith, Steven Olsen Smith
+- **abs:** https://arxiv.org/abs/2603.14674 · **PDF:** https://arxiv.org/pdf/2603.14674v1
+
+## 4755. A Single-Sample Polylogarithmic Regret Bound for Nonstationary Online Linear Programming
+
+- **ID:** `2603.14673` · **primary:** `cs.DS` · **Submitted:** 2026-03-15
+- **Authors:** Haoran Xu, Owen Shen, Peter Glynn, et al.
+- **abs:** https://arxiv.org/abs/2603.14673 · **PDF:** https://arxiv.org/pdf/2603.14673v1
+
+## 4756. Seamless Deception: Larger Language Models Are Better Knowledge Concealers
+
+- **ID:** `2603.14672` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Dhananjay Ashok, Ruth-Ann Armstrong, Jonathan May
+- **abs:** https://arxiv.org/abs/2603.14672 · **PDF:** https://arxiv.org/pdf/2603.14672v2
+
+## 4757. RenderMem: Rendering as Spatial Memory Retrieval
+
+- **ID:** `2603.14669` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** JooHyun Park, HyeongYeop Kang
+- **abs:** https://arxiv.org/abs/2603.14669 · **PDF:** https://arxiv.org/pdf/2603.14669v1
+
+## 4758. Comparative Analysis of 3D Convolutional and 2.5D Slice-Conditioned U-Net Architectures for MRI Super-Resolution via Elucidated Diffusion Models
+
+- **ID:** `2603.14667` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Hendrik Chiche, Ludovic Corcos, Logan Rouge
+- **abs:** https://arxiv.org/abs/2603.14667 · **PDF:** https://arxiv.org/pdf/2603.14667v1
+
+## 4759. Gradient Atoms: Unsupervised Discovery, Attribution and Steering of Model Behaviors via Sparse Decomposition of Training Gradients
+
+- **ID:** `2603.14665` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** J Rosser
+- **abs:** https://arxiv.org/abs/2603.14665 · **PDF:** https://arxiv.org/pdf/2603.14665v2
+
+## 4760. Punctuated Equilibria in Artificial Intelligence: The Institutional Scaling Law and the Speciation of Sovereign AI
+
+- **ID:** `2603.14664` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Mark Baciak, Thomas A. Cellucci, Deanna M. Falkowski
+- **abs:** https://arxiv.org/abs/2603.14664 · **PDF:** https://arxiv.org/pdf/2603.14664v1
+
+## 4761. VisionCoach: Reinforcing Grounded Video Reasoning via Visual-Perception Prompting
+
+- **ID:** `2603.14659` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Daeun Lee, Shoubin Yu, Yue Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.14659 · **PDF:** https://arxiv.org/pdf/2603.14659v1
+
+## 4762. Human-AI Ensembles Improve Deepfake Detection in Low-to-Medium Quality Videos
+
+- **ID:** `2603.14658` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Marco Postiglione, Isabel Gortner, V. S. Subrahmanian
+- **abs:** https://arxiv.org/abs/2603.14658 · **PDF:** https://arxiv.org/pdf/2603.14658v1
+
+## 4763. EARCP: Self-Regulating Coherence-Aware Ensemble Architecture for Sequential Decision Making -- Ensemble Auto-Regule par Coherence et Performance
+
+- **ID:** `2603.14651` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Mike Amega
+- **abs:** https://arxiv.org/abs/2603.14651 · **PDF:** https://arxiv.org/pdf/2603.14651v1
+
+## 4764. A Methodology for Thermal Limit Bias Predictability Through Artificial Intelligence
+
+- **ID:** `2603.14648` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Anirudh Tunga, Michael J. Mueterthies, Jonathan Nistor
+- **abs:** https://arxiv.org/abs/2603.14648 · **PDF:** https://arxiv.org/pdf/2603.14648v1
+
+## 4765. TopoCL: Topological Contrastive Learning for Medical Imaging
+
+- **ID:** `2603.14647` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Guangyu Meng, Pengfei Gu, Peixian Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.14647 · **PDF:** https://arxiv.org/pdf/2603.14647v1
+
+## 4766. Dynamic Theory of Mind as a Temporal Memory Problem: Evidence from Large Language Models
+
+- **ID:** `2603.14646` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Thuy Ngoc Nguyen, Duy Nhat Phan, Cleotilde Gonzalez
+- **abs:** https://arxiv.org/abs/2603.14646 · **PDF:** https://arxiv.org/pdf/2603.14646v1
+
+## 4767. LUMINA: A Multi-Vendor Mammography Benchmark with Energy Harmonization Protocol
+
+- **ID:** `2603.14644` · **primary:** `eess.IV` · **Submitted:** 2026-03-15
+- **Authors:** Hongyi Pan, Gorkem Durak, Halil Ertugrul Aktas, et al.
+- **abs:** https://arxiv.org/abs/2603.14644 · **PDF:** https://arxiv.org/pdf/2603.14644v2
+
+## 4768. Argumentation for Explainable and Globally Contestable Decision Support with LLMs
+
+- **ID:** `2603.14643` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Adam Dejl, Matthew Williams, Francesca Toni
+- **abs:** https://arxiv.org/abs/2603.14643 · **PDF:** https://arxiv.org/pdf/2603.14643v1
+
+## 4769. Nudging Hidden States: Training-Free Model Steering for Chain-of-Thought Reasoning in Large Audio-Language Models
+
+- **ID:** `2603.14636` · **primary:** `cs.SD` · **Submitted:** 2026-03-15
+- **Authors:** Lok-Lam Ieong, Chia-Chien Chen, Chih-Kai Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.14636 · **PDF:** https://arxiv.org/pdf/2603.14636v1
+
+## 4770. Compute Allocation for Reasoning-Intensive Retrieval Agents
+
+- **ID:** `2603.14635` · **primary:** `cs.IR` · **Submitted:** 2026-03-15
+- **Authors:** Sreeja Apparaju, Nilesh Gupta
+- **abs:** https://arxiv.org/abs/2603.14635 · **PDF:** https://arxiv.org/pdf/2603.14635v2
+
+## 4771. Anterior's Approach to Fairness Evaluation of Automated Prior Authorization System
+
+- **ID:** `2603.14631` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Sai P. Selvaraj, Khadija Mahmoud, Anuj Iravane
+- **abs:** https://arxiv.org/abs/2603.14631 · **PDF:** https://arxiv.org/pdf/2603.14631v1
+
+## 4772. ResearchPilot: A Local-First Multi-Agent System for Literature Synthesis and Related Work Drafting
+
+- **ID:** `2603.14629` · **primary:** `cs.IR` · **Submitted:** 2026-03-15
+- **Authors:** Peng Zhang
+- **abs:** https://arxiv.org/abs/2603.14629 · **PDF:** https://arxiv.org/pdf/2603.14629v1
+
+## 4773. s2n-bignum-bench: A practical benchmark for evaluating low-level code reasoning of LLMs
+
+- **ID:** `2603.14628` · **primary:** `cs.PL` · **Submitted:** 2026-03-15
+- **Authors:** Balaji Rao, John Harrison, Soonho Kong, et al.
+- **abs:** https://arxiv.org/abs/2603.14628 · **PDF:** https://arxiv.org/pdf/2603.14628v1
+
+## 4774. EcoFair-CH-MARL: Scalable Constrained Hierarchical Multi-Agent RL with Real-Time Emission Budgets and Fairness Guarantees
+
+- **ID:** `2603.14625` · **primary:** `cs.MA` · **Submitted:** 2026-03-15
+- **Authors:** Saad Alqithami
+- **abs:** https://arxiv.org/abs/2603.14625 · **PDF:** https://arxiv.org/pdf/2603.14625v1
+
+## 4775. Proactive Routing to Interpretable Surrogates with Distribution-Free Safety Guarantees
+
+- **ID:** `2603.14623` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Iqtedar Uddin, Mazin Khider, André Bauer
+- **abs:** https://arxiv.org/abs/2603.14623 · **PDF:** https://arxiv.org/pdf/2603.14623v1
+
+## 4776. LLM-Augmented Release Intelligence: Automated Change Summarization and Impact Analysis in Cloud-Native CI/CD Pipelines
+
+- **ID:** `2603.14619` · **primary:** `cs.SE` · **Submitted:** 2026-03-15
+- **Authors:** Happy Bhati
+- **abs:** https://arxiv.org/abs/2603.14619 · **PDF:** https://arxiv.org/pdf/2603.14619v1
+
+## 4777. PulmoVec: A Two-Stage Stacking Meta-Learning Architecture Built on the HeAR Foundation Model for Multi-Task Classification of Pediatric Respiratory Sounds
+
+- **ID:** `2603.15688` · **primary:** `cs.SD` · **Submitted:** 2026-03-15
+- **Authors:** Izzet Turkalp Akbasli, Oguzhan Serin
+- **abs:** https://arxiv.org/abs/2603.15688 · **PDF:** https://arxiv.org/pdf/2603.15688v1
+
+## 4778. Delightful Policy Gradient
+
+- **ID:** `2603.14608` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Ian Osband
+- **abs:** https://arxiv.org/abs/2603.14608 · **PDF:** https://arxiv.org/pdf/2603.14608v1
+
+## 4779. Tactile Modality Fusion for Vision-Language-Action Models
+
+- **ID:** `2603.14604` · **primary:** `cs.RO` · **Submitted:** 2026-03-15
+- **Authors:** Charlotte Morissette, Amin Abyaneh, Wei-Di Chang, et al.
+- **abs:** https://arxiv.org/abs/2603.14604 · **PDF:** https://arxiv.org/pdf/2603.14604v1
+
+## 4780. PA3: Policy-Aware Agent Alignment through Chain-of-Thought
+
+- **ID:** `2603.14602` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Shubhashis Roy Dipta, Daniel Bis, Kun Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.14602 · **PDF:** https://arxiv.org/pdf/2603.14602v2
+
+## 4781. $K-$means with learned metrics
+
+- **ID:** `2603.14601` · **primary:** `math.ST` · **Submitted:** 2026-03-15
+- **Authors:** Pablo Groisman, Matthieu Jonckheere, Jordan Serres, et al.
+- **abs:** https://arxiv.org/abs/2603.14601 · **PDF:** https://arxiv.org/pdf/2603.14601v2
+
+## 4782. A Loss Landscape Visualization Framework for Interpreting Reinforcement Learning: An ADHDP Case Study
+
+- **ID:** `2603.14600` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Jingyi Liu, Jian Guo, Eberhard Gill
+- **abs:** https://arxiv.org/abs/2603.14600 · **PDF:** https://arxiv.org/pdf/2603.14600v1
+
+## 4783. D-MEM: Dopamine-Gated Agentic Memory via Reward Prediction Error Routing
+
+- **ID:** `2603.14597` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-15
+- **Authors:** Yuru Song, Qi Xin
+- **abs:** https://arxiv.org/abs/2603.14597 · **PDF:** https://arxiv.org/pdf/2603.14597v1
+
+## 4784. Scaling the Explanation of Multi-Class Bayesian Network Classifiers
+
+- **ID:** `2603.14594` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Yaofang Zhang, Adnan Darwiche
+- **abs:** https://arxiv.org/abs/2603.14594 · **PDF:** https://arxiv.org/pdf/2603.14594v1
+
+## 4785. Parameter-Efficient Quality Estimation via Frozen Recursive Models
+
+- **ID:** `2603.14593` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Umar Abubacar, Roman Bauer, Diptesh Kanojia
+- **abs:** https://arxiv.org/abs/2603.14593 · **PDF:** https://arxiv.org/pdf/2603.14593v1
+
+## 4786. A Multi-Scale Graph Learning Framework with Temporal Consistency Constraints for Financial Fraud Detection in Transaction Networks under Non-Stationary Conditions
+
+- **ID:** `2603.14592` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Yiming Lei, Qiannan Shen, Junhao Song
+- **abs:** https://arxiv.org/abs/2603.14592 · **PDF:** https://arxiv.org/pdf/2603.14592v1
+
+## 4787. FlashHead: Efficient Drop-In Replacement for the Classification Head in Language Model Inference
+
+- **ID:** `2603.14591` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Wilhelm Tranheden, Shahnawaz Ahmed, Devdatt Dubhashi, et al.
+- **abs:** https://arxiv.org/abs/2603.14591 · **PDF:** https://arxiv.org/pdf/2603.14591v1
+
+## 4788. Adapting Critic Match Loss Landscape Visualization to Off-policy Reinforcement Learning
+
+- **ID:** `2603.14589` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Jingyi Liu, Jian Guo, Eberhard Gill
+- **abs:** https://arxiv.org/abs/2603.14589 · **PDF:** https://arxiv.org/pdf/2603.14589v1
+
+## 4789. SuperLocalMemory V3: Information-Geometric Foundations for Zero-LLM Enterprise Agent Memory
+
+- **ID:** `2603.14588` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Varun Pratap Bhardwaj
+- **abs:** https://arxiv.org/abs/2603.14588 · **PDF:** https://arxiv.org/pdf/2603.14588v1
+
+## 4790. The Scenic Route to Deception: Dark Patterns and Explainability Pitfalls in Conversational Navigation
+
+- **ID:** `2603.14586` · **primary:** `cs.HC` · **Submitted:** 2026-03-15
+- **Authors:** Ilya Ilyankou, Stefano Cavazzi, James Haworth
+- **abs:** https://arxiv.org/abs/2603.14586 · **PDF:** https://arxiv.org/pdf/2603.14586v1
+
+## 4791. Machine Learning-Driven Intelligent Memory System Design: From On-Chip Caches to Storage
+
+- **ID:** `2603.14583` · **primary:** `cs.AR` · **Submitted:** 2026-03-15
+- **Authors:** Rahul Bera, Rakesh Nadig, Onur Mutlu
+- **abs:** https://arxiv.org/abs/2603.14583 · **PDF:** https://arxiv.org/pdf/2603.14583v1
+
+## 4792. Medical Image Spatial Grounding with Semantic Sampling
+
+- **ID:** `2603.14579` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Andrew Seohwan Yu, Mohsen Hariri, Kunio Nakamura, et al.
+- **abs:** https://arxiv.org/abs/2603.14579 · **PDF:** https://arxiv.org/pdf/2603.14579v2
+
+## 4793. Power-Law Spectrum of the Random Feature Model
+
+- **ID:** `2603.14578` · **primary:** `stat.ML` · **Submitted:** 2026-03-15
+- **Authors:** Elliot Paquette, Ke Liang Xiao, Yizhe Zhu
+- **abs:** https://arxiv.org/abs/2603.14578 · **PDF:** https://arxiv.org/pdf/2603.14578v1
+
+## 4794. IQP Born Machines under Data-dependent and Agnostic Initialization Strategies
+
+- **ID:** `2603.14576` · **primary:** `quant-ph` · **Submitted:** 2026-03-15
+- **Authors:** Sacha Lerch, Joseph Bowles, Ricard Puig, et al.
+- **abs:** https://arxiv.org/abs/2603.14576 · **PDF:** https://arxiv.org/pdf/2603.14576v1
+
+## 4795. CausalEvolve: Towards Open-Ended Discovery with Causal Scratchpad
+
+- **ID:** `2603.14575` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Yongqiang Chen, Chenxi Liu, Zhenhao Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.14575 · **PDF:** https://arxiv.org/pdf/2603.14575v2
+
+## 4796. Rigorous Asymptotics for First-Order Algorithms Through the Dynamical Cavity Method
+
+- **ID:** `2603.14573` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-03-15
+- **Authors:** Yatin Dandi, David Gamarnik, Francisco Pernice, et al.
+- **abs:** https://arxiv.org/abs/2603.14573 · **PDF:** https://arxiv.org/pdf/2603.14573v1
+
+## 4797. Top-b: Entropic Regulation of Relative Probability Bands in Autoregressive Language Processes
+
+- **ID:** `2603.14567` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Deepon Halder, Raj Dabre
+- **abs:** https://arxiv.org/abs/2603.14567 · **PDF:** https://arxiv.org/pdf/2603.14567v1
+
+## 4798. Multilingual TinyStories: A Synthetic Combinatorial Corpus of Indic Children's Stories for Training Small Language Models
+
+- **ID:** `2603.14563` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Deepon Halder, Angira Mukherjee
+- **abs:** https://arxiv.org/abs/2603.14563 · **PDF:** https://arxiv.org/pdf/2603.14563v1
+
+## 4799. A comprehensive multimodal dataset and benchmark for ulcerative colitis scoring in endoscopy
+
+- **ID:** `2603.14559` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Noha Ghatwary, Jiangbei Yue, Ahmed Elgendy, et al.
+- **abs:** https://arxiv.org/abs/2603.14559 · **PDF:** https://arxiv.org/pdf/2603.14559v1
+
+## 4800. JobMatchAI An Intelligent Job Matching Platform Using Knowledge Graphs, Semantic Search and Explainable AI
+
+- **ID:** `2603.14558` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Mayank Vyas, Abhijit Chakraborty, Vivek Gupta
+- **abs:** https://arxiv.org/abs/2603.14558 · **PDF:** https://arxiv.org/pdf/2603.14558v2
+
+## 4801. An End-to-end Architecture for Collider Physics and Beyond
+
+- **ID:** `2603.14553` · **primary:** `hep-ph` · **Submitted:** 2026-03-15
+- **Authors:** Shi Qiu, Zeyu Cai, Jiashen Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.14553 · **PDF:** https://arxiv.org/pdf/2603.14553v1
+
+## 4802. Evidential Domain Adaptation for Remaining Useful Life Prediction with Incomplete Degradation
+
+- **ID:** `2603.15687` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Yubo Hou, Mohamed Ragab, Yucheng Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.15687 · **PDF:** https://arxiv.org/pdf/2603.15687v1
+
+## 4803. Learning to Order: Task Sequencing as In-Context Optimization
+
+- **ID:** `2603.14550` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Jan Kobiolka, Christian Frey, Arlind Kadra, et al.
+- **abs:** https://arxiv.org/abs/2603.14550 · **PDF:** https://arxiv.org/pdf/2603.14550v1
+
+## 4804. ASAP: Attention-Shift-Aware Pruning for Efficient LVLM Inference
+
+- **ID:** `2603.14549` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Surendra Pathak, Bo Han
+- **abs:** https://arxiv.org/abs/2603.14549 · **PDF:** https://arxiv.org/pdf/2603.14549v2
+
+## 4805. Gradient Boosting for Spatial Panel Models with Random and Fixed Effects
+
+- **ID:** `2603.14543` · **primary:** `stat.ME` · **Submitted:** 2026-03-15
+- **Authors:** Michael Balzer, Adhen Benlahlou
+- **abs:** https://arxiv.org/abs/2603.14543 · **PDF:** https://arxiv.org/pdf/2603.14543v1
+
+## 4806. Expert Mind: A Retrieval-Augmented Architecture for Expert Knowledge Preservation in the Energy Sector
+
+- **ID:** `2603.14541` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Diego Ezequiel Cervera
+- **abs:** https://arxiv.org/abs/2603.14541 · **PDF:** https://arxiv.org/pdf/2603.14541v1
+
+## 4807. Visualizing Critic Match Loss Landscapes for Interpretation of Online Reinforcement Learning Control Algorithms
+
+- **ID:** `2603.14535` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Jingyi Liu, Jian Guo, Eberhard Gill
+- **abs:** https://arxiv.org/abs/2603.14535 · **PDF:** https://arxiv.org/pdf/2603.14535v1
+
+## 4808. Emotional Cost Functions for AI Safety: Teaching Agents to Feel the Weight of Irreversible Consequences
+
+- **ID:** `2603.14531` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Pandurang Mopgar
+- **abs:** https://arxiv.org/abs/2603.14531 · **PDF:** https://arxiv.org/pdf/2603.14531v1
+
+## 4809. MALicious INTent Dataset and Inoculating LLMs for Enhanced Disinformation Detection
+
+- **ID:** `2603.14525` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Arkadiusz Modzelewski, Witold Sosnowski, Eleni Papadopulos, et al.
+- **abs:** https://arxiv.org/abs/2603.14525 · **PDF:** https://arxiv.org/pdf/2603.14525v1
+
+## 4810. VLA-Thinker: Boosting Vision-Language-Action Models through Thinking-with-Image Reasoning
+
+- **ID:** `2603.14523` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Chaoyang Wang, Wenrui Bao, Sicheng Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.14523 · **PDF:** https://arxiv.org/pdf/2603.14523v1
+
+## 4811. Learning to Forget: Sleep-Inspired Memory Consolidation for Resolving Proactive Interference in Large Language Models
+
+- **ID:** `2603.14517` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Ying Xie
+- **abs:** https://arxiv.org/abs/2603.14517 · **PDF:** https://arxiv.org/pdf/2603.14517v1
+
+## 4812. Excited Pfaffians: Generalized Neural Wave Functions Across Structure and State
+
+- **ID:** `2603.14515` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Nicholas Gao, Till Grutschus, Frank Noé, et al.
+- **abs:** https://arxiv.org/abs/2603.14515 · **PDF:** https://arxiv.org/pdf/2603.14515v1
+
+## 4813. High-Probability Bounds for SGD under the Polyak-Lojasiewicz Condition with Markovian Noise
+
+- **ID:** `2603.14514` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Avik Kar, Siddharth Chandak, Rahul Singh, et al.
+- **abs:** https://arxiv.org/abs/2603.14514 · **PDF:** https://arxiv.org/pdf/2603.14514v1
+
+## 4814. Trust-Region Noise Search for Black-Box Alignment of Diffusion and Flow Models
+
+- **ID:** `2603.14504` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Niklas Schweiger, Daniel Cremers, Karnik Ram
+- **abs:** https://arxiv.org/abs/2603.14504 · **PDF:** https://arxiv.org/pdf/2603.14504v1
+
+## 4815. The Provenance Paradox in Multi-Agent LLM Routing: Delegation Contracts and Attested Identity in LDP
+
+- **ID:** `2603.18043` · **primary:** `cs.MA` · **Submitted:** 2026-03-15
+- **Authors:** Sunil Prakash
+- **abs:** https://arxiv.org/abs/2603.18043 · **PDF:** https://arxiv.org/pdf/2603.18043v1
+
+## 4816. CangjieBench: Benchmarking LLMs on a Low-Resource General-Purpose Programming Language
+
+- **ID:** `2603.14501` · **primary:** `cs.SE` · **Submitted:** 2026-03-15
+- **Authors:** Junhang Cheng, Fang Liu, Jia Li, et al.
+- **abs:** https://arxiv.org/abs/2603.14501 · **PDF:** https://arxiv.org/pdf/2603.14501v1
+
+## 4817. Refining 3D Medical Segmentation with Verbal Instruction
+
+- **ID:** `2603.14496` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Kangxian Xie, Jiancheng Yang, Nandor Pinter, et al.
+- **abs:** https://arxiv.org/abs/2603.14496 · **PDF:** https://arxiv.org/pdf/2603.14496v1
+
+## 4818. Bridging the Gap in the Responsible AI Divides
+
+- **ID:** `2603.14495` · **primary:** `cs.CY` · **Submitted:** 2026-03-15
+- **Authors:** Bálint Gyevnár, Atoosa Kasirzadeh
+- **abs:** https://arxiv.org/abs/2603.14495 · **PDF:** https://arxiv.org/pdf/2603.14495v1
+
+## 4819. Fine-tuning MLLMs Without Forgetting Is Easier Than You Think
+
+- **ID:** `2603.14493` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** He Li, Yuhui Zhang, Xiaohan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.14493 · **PDF:** https://arxiv.org/pdf/2603.14493v1
+
+## 4820. Predicting Stress-strain Behaviors of Additively Manufactured Materials via Loss-based and Activation-based Physics-informed Machine Learning
+
+- **ID:** `2603.14489` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Chenglong Duan, Dazhong Wu
+- **abs:** https://arxiv.org/abs/2603.14489 · **PDF:** https://arxiv.org/pdf/2603.14489v1
+
+## 4821. Infinite Problem Generator: Verifiably Scaling Physics Reasoning Data with Agentic Workflows
+
+- **ID:** `2603.14486` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Aditya Sharan, Sriram Hebbale, Dhruv Kumar
+- **abs:** https://arxiv.org/abs/2603.14486 · **PDF:** https://arxiv.org/pdf/2603.14486v1
+
+## 4822. Unlearning-based sliding window for continual learning under concept drift
+
+- **ID:** `2603.14484` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Michal Wozniak, Marek Klonowski, Maciej Maczynski, et al.
+- **abs:** https://arxiv.org/abs/2603.14484 · **PDF:** https://arxiv.org/pdf/2603.14484v1
+
+## 4823. Disentangling Dynamical Systems: Causal Representation Learning Meets Local Sparse Attention
+
+- **ID:** `2603.14483` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Markus W. Baumgartner, Anson Lei, Joe Watson, et al.
+- **abs:** https://arxiv.org/abs/2603.14483 · **PDF:** https://arxiv.org/pdf/2603.14483v1
+
+## 4824. Convergence of Two Time-Scale Stochastic Approximation: A Martingale Approach
+
+- **ID:** `2603.14481` · **primary:** `stat.ML` · **Submitted:** 2026-03-15
+- **Authors:** Mathukumalli Vidyasagar
+- **abs:** https://arxiv.org/abs/2603.14481 · **PDF:** https://arxiv.org/pdf/2603.14481v1
+
+## 4825. The Voice Behind the Words: Quantifying Intersectional Bias in SpeechLLMs
+
+- **ID:** `2603.16941` · **primary:** `eess.AS` · **Submitted:** 2026-03-15
+- **Authors:** Shree Harsha Bokkahalli Satish, Christoph Minixhofer, Maria Teleki, et al.
+- **abs:** https://arxiv.org/abs/2603.16941 · **PDF:** https://arxiv.org/pdf/2603.16941v1
+
+## 4826. Geometric and Topological Deep Learning for Predicting Thermo-mechanical Performance in Cold Spray Deposition Process Modeling
+
+- **ID:** `2603.14478` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Akshansh Mishra
+- **abs:** https://arxiv.org/abs/2603.14478 · **PDF:** https://arxiv.org/pdf/2603.14478v1
+
+## 4827. A Novel Framework using Intuitionistic Fuzzy Logic with U-Net and U-Net++ Architecture: A case Study of MRI Bain Image Segmentation
+
+- **ID:** `2603.18042` · **primary:** `eess.IV` · **Submitted:** 2026-03-15
+- **Authors:** Hanuman Verma, Kiho Im, Akshansh Gupta, et al.
+- **abs:** https://arxiv.org/abs/2603.18042 · **PDF:** https://arxiv.org/pdf/2603.18042v1
+
+## 4828. On the (Generative) Linear Sketching Problem
+
+- **ID:** `2603.14474` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Xinyu Yuan, Yan Qiao, Zonghui Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.14474 · **PDF:** https://arxiv.org/pdf/2603.14474v1
+
+## 4829. AI Can Learn Scientific Taste
+
+- **ID:** `2603.14473` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Jingqi Tong, Mingzhe Li, Hangcheng Li, et al.
+- **abs:** https://arxiv.org/abs/2603.14473 · **PDF:** https://arxiv.org/pdf/2603.14473v1
+
+## 4830. Physics-Informed Policy Optimization via Analytic Dynamics Regularization
+
+- **ID:** `2603.14469` · **primary:** `cs.RO` · **Submitted:** 2026-03-15
+- **Authors:** Namai Chandra, Liu Mohan, Zhihao Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.14469 · **PDF:** https://arxiv.org/pdf/2603.14469v2
+
+## 4831. AgentProcessBench: Diagnosing Step-Level Process Quality in Tool-Using Agents
+
+- **ID:** `2603.14465` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Shengda Fan, Xuyan Ye, Yupeng Huo, et al.
+- **abs:** https://arxiv.org/abs/2603.14465 · **PDF:** https://arxiv.org/pdf/2603.14465v1
+
+## 4832. An Industrial-Scale Insurance LLM Achieving Verifiable Domain Mastery and Hallucination Control without Competence Trade-offs
+
+- **ID:** `2603.14463` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Qian Zhu, Xinnan Guo, Jingjing Huo, et al.
+- **abs:** https://arxiv.org/abs/2603.14463 · **PDF:** https://arxiv.org/pdf/2603.14463v1
+
+## 4833. STAG-CN: Spatio-Temporal Apiary Graph Convolutional Network for Disease Onset Prediction in Beehive Sensor Networks
+
+- **ID:** `2603.14462` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Sungwoo Kang
+- **abs:** https://arxiv.org/abs/2603.14462 · **PDF:** https://arxiv.org/pdf/2603.14462v1
+
+## 4834. Inclusive AI for Group Interactions: Predicting Gaze-Direction Behaviors in People with Intellectual and Developmental Disabilities
+
+- **ID:** `2603.14460` · **primary:** `cs.HC` · **Submitted:** 2026-03-15
+- **Authors:** Giulia Huang, Maristella Matera, Micol Spitale
+- **abs:** https://arxiv.org/abs/2603.14460 · **PDF:** https://arxiv.org/pdf/2603.14460v1
+
+## 4835. Distilling Reasoning Without Knowledge: A Framework for Reliable LLMs
+
+- **ID:** `2603.14458` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Auksarapak Kietkajornrit, Jad Tarifi, Nima Asgharbeygi
+- **abs:** https://arxiv.org/abs/2603.14458 · **PDF:** https://arxiv.org/pdf/2603.14458v1
+
+## 4836. PARSA-Bench: A Comprehensive Persian Audio-Language Model Benchmark
+
+- **ID:** `2603.14456` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Mohammad Javad Ranjbar Kalahroodi, Mohammad Amini, Parmis Bathayan, et al.
+- **abs:** https://arxiv.org/abs/2603.14456 · **PDF:** https://arxiv.org/pdf/2603.14456v1
+
+## 4837. On the Degrees of Freedom of Gridded Control Points in Learning-Based Medical Image Registration
+
+- **ID:** `2603.16940` · **primary:** `eess.IV` · **Submitted:** 2026-03-15
+- **Authors:** Wen Yan, Qianye Yang, Yipei Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.16940 · **PDF:** https://arxiv.org/pdf/2603.16940v1
+
+## 4838. How to find expressible and trainable parameterized quantum circuits?
+
+- **ID:** `2603.14451` · **primary:** `quant-ph` · **Submitted:** 2026-03-15
+- **Authors:** Peter Röseler, Dennis Willsch, Kristel Michielsen
+- **abs:** https://arxiv.org/abs/2603.14451 · **PDF:** https://arxiv.org/pdf/2603.14451v1
+
+## 4839. Life cycle assessment for all organic chemicals
+
+- **ID:** `2603.15686` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-15
+- **Authors:** Shaohan Chen, Tim Langhorst, Julian Nöhl, et al.
+- **abs:** https://arxiv.org/abs/2603.15686 · **PDF:** https://arxiv.org/pdf/2603.15686v1
+
+## 4840. Zoom to Essence: Trainless GUI Grounding by Inferring upon Interface Elements
+
+- **ID:** `2603.14448` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Ziwei Liu, Tao Feng, Borui Kang, et al.
+- **abs:** https://arxiv.org/abs/2603.14448 · **PDF:** https://arxiv.org/pdf/2603.14448v1
+
+## 4841. Echoes Across Centuries: Phonetic Signatures of Persian Poets
+
+- **ID:** `2603.14443` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Kourosh Shahnazari, Seyed Moein Ayyoubzadeh, Mohammadali Keshtparvar
+- **abs:** https://arxiv.org/abs/2603.14443 · **PDF:** https://arxiv.org/pdf/2603.14443v2
+
+## 4842. AR-Flow VAE: A Structured Autoregressive Flow Prior Variational Autoencoder for Unsupervised Blind Source Separation
+
+- **ID:** `2603.14441` · **primary:** `stat.ML` · **Submitted:** 2026-03-15
+- **Authors:** Yuan-Hao Wei, Fu-Hao Deng, Lin-Yong Cui, et al.
+- **abs:** https://arxiv.org/abs/2603.14441 · **PDF:** https://arxiv.org/pdf/2603.14441v2
+
+## 4843. DASH: Dynamic Audio-Driven Semantic Chunking for Efficient Omnimodal Token Compression
+
+- **ID:** `2603.15685` · **primary:** `cs.MM` · **Submitted:** 2026-03-15
+- **Authors:** Bingzhou Li, Tao Huang
+- **abs:** https://arxiv.org/abs/2603.15685 · **PDF:** https://arxiv.org/pdf/2603.15685v1
+
+## 4844. Creative Convergence or Imitation? Genre-Specific Homogeneity in LLM-Generated Chinese Literature
+
+- **ID:** `2603.14430` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Yuanchi Ma, Kaize Shi, Hui He, et al.
+- **abs:** https://arxiv.org/abs/2603.14430 · **PDF:** https://arxiv.org/pdf/2603.14430v1
+
+## 4845. Whether, Not Which: Mechanistic Interpretability Reveals Dissociable Affect Reception and Emotion Categorization in LLMs
+
+- **ID:** `2603.22295` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Michael Keeman
+- **abs:** https://arxiv.org/abs/2603.22295 · **PDF:** https://arxiv.org/pdf/2603.22295v1
+
+## 4846. MBD: A Model-Based Debiasing Framework Across User, Content, and Model Dimensions
+
+- **ID:** `2603.14422` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Yuantong Li, Lei Yuan, Zhihao Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.14422 · **PDF:** https://arxiv.org/pdf/2603.14422v1
+
+## 4847. Data Darwinism Part II: DataEvolve -- AI can Autonomously Evolve Pretraining Data Curation
+
+- **ID:** `2603.14420` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Tiantian Mi, Dongming Shan, Zhen Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.14420 · **PDF:** https://arxiv.org/pdf/2603.14420v1
+
+## 4848. Deep EM with Hierarchical Latent Label Modelling for Multi-Site Prostate Lesion Segmentation
+
+- **ID:** `2603.14418` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Wen Yan, Yipei Wang, Shiqi Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.14418 · **PDF:** https://arxiv.org/pdf/2603.14418v1
+
+## 4849. Questionnaire Responses Do not Capture the Safety of AI Agents
+
+- **ID:** `2603.14417` · **primary:** `cs.CY` · **Submitted:** 2026-03-15
+- **Authors:** Max Hellrigel-Holderbaum, Edward James Young
+- **abs:** https://arxiv.org/abs/2603.14417 · **PDF:** https://arxiv.org/pdf/2603.14417v1
+
+## 4850. Ternary Gamma Semirings: From Neural Implementation to Categorical Foundations
+
+- **ID:** `2603.19317` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Ruoqi Sun
+- **abs:** https://arxiv.org/abs/2603.19317 · **PDF:** https://arxiv.org/pdf/2603.19317v1
+
+## 4851. BiT-MCTS: A Theme-based Bidirectional MCTS Approach to Chinese Fiction Generation
+
+- **ID:** `2603.14410` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Zhaoyi Li, Xu Zhang, Xiaojun Wan
+- **abs:** https://arxiv.org/abs/2603.14410 · **PDF:** https://arxiv.org/pdf/2603.14410v2
+
+## 4852. PGcGAN: Pathological Gait-Conditioned GAN for Human Gait Synthesis
+
+- **ID:** `2603.14409` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Mritula Chandrasekaran, Sanket Kachole, Jarek Francik, et al.
+- **abs:** https://arxiv.org/abs/2603.14409 · **PDF:** https://arxiv.org/pdf/2603.14409v1
+
+## 4853. Towards One-for-All Anomaly Detection for Tabular Data
+
+- **ID:** `2603.14407` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Shiyuan Li, Yixin Liu, Yu Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.14407 · **PDF:** https://arxiv.org/pdf/2603.14407v1
+
+## 4854. Graph-Based Deep Learning for Intelligent Detection of Energy Losses, Theft, and Operational Inefficiencies in Oil & Gas Production Networks
+
+- **ID:** `2603.14406` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** AbdulQoyum A. Olowookere, Adewale U. Oguntola, Ebenezer. Leke Odekanle
+- **abs:** https://arxiv.org/abs/2603.14406 · **PDF:** https://arxiv.org/pdf/2603.14406v1
+
+## 4855. ES-Merging: Biological MLLM Merging via Embedding Space Signals
+
+- **ID:** `2603.14405` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Wonbin Lee, Dongki Kim, Sung Ju Hwang
+- **abs:** https://arxiv.org/abs/2603.14405 · **PDF:** https://arxiv.org/pdf/2603.14405v1
+
+## 4856. Extending Minimal Pairs with Ordinal Surprisal Curves and Entropy Across Applied Domains
+
+- **ID:** `2603.14400` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Andrew Katz
+- **abs:** https://arxiv.org/abs/2603.14400 · **PDF:** https://arxiv.org/pdf/2603.14400v1
+
+## 4857. WestWorld: A Knowledge-Encoded Scalable Trajectory World Model for Diverse Robotic Systems
+
+- **ID:** `2603.14392` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Yuchen Wang, Jiangtao Kong, Sizhe Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.14392 · **PDF:** https://arxiv.org/pdf/2603.14392v1
+
+## 4858. From $\boldsymbol{\logπ}$ to $\boldsymbolπ$: Taming Divergence in Soft Clipping via Bilateral Decoupled Decay of Probability Gradient Weight
+
+- **ID:** `2603.14389` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Xiaoliang Fu, Jiaye Lin, Yangyi Fang, et al.
+- **abs:** https://arxiv.org/abs/2603.14389 · **PDF:** https://arxiv.org/pdf/2603.14389v1
+
+## 4859. Label Noise Cleaning for Supervised Classification via Bernoulli Random Sampling
+
+- **ID:** `2603.14387` · **primary:** `stat.ME` · **Submitted:** 2026-03-15
+- **Authors:** Yuxin Liu, Xiong Jin, Yang Han
+- **abs:** https://arxiv.org/abs/2603.14387 · **PDF:** https://arxiv.org/pdf/2603.14387v1
+
+## 4860. SPARQ: Spiking Early-Exit Neural Networks for Energy-Efficient Edge AI
+
+- **ID:** `2603.14380` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Parth Patne, Mahdi Taheri, Ali Mahani, et al.
+- **abs:** https://arxiv.org/abs/2603.14380 · **PDF:** https://arxiv.org/pdf/2603.14380v1
+
+## 4861. The Pulse of Motion: Measuring Physical Frame Rate from Visual Dynamics
+
+- **ID:** `2603.14375` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Xiangbo Gao, Mingyang Wu, Siyuan Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.14375 · **PDF:** https://arxiv.org/pdf/2603.14375v2
+
+## 4862. Contests with Spillovers: Incentivizing Content Creation with GenAI
+
+- **ID:** `2603.14372` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Sagi Ohayon, Boaz Taitler, Omer Ben-Porat
+- **abs:** https://arxiv.org/abs/2603.14372 · **PDF:** https://arxiv.org/pdf/2603.14372v1
+
+## 4863. OxyGen: Unified KV Cache Management for Vision-Language-Action Models under Multi-Task Parallelism
+
+- **ID:** `2603.14371` · **primary:** `cs.RO` · **Submitted:** 2026-03-15
+- **Authors:** Xiangyu Li, Huaizhi Tang, Xin Ding, et al.
+- **abs:** https://arxiv.org/abs/2603.14371 · **PDF:** https://arxiv.org/pdf/2603.14371v1
+
+## 4864. From Specification to Architecture: A Theory Compiler for Knowledge-Guided Machine Learning
+
+- **ID:** `2603.14369` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Asela Hevapathige, Yu Xia, Sachith Seneviratne, et al.
+- **abs:** https://arxiv.org/abs/2603.14369 · **PDF:** https://arxiv.org/pdf/2603.14369v1
+
+## 4865. Representation Alignment for Just Image Transformers is not Easier than You Think
+
+- **ID:** `2603.14366` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Jaeyo Shin, Jiwook Kim, Hyunjung Shim
+- **abs:** https://arxiv.org/abs/2603.14366 · **PDF:** https://arxiv.org/pdf/2603.14366v1
+
+## 4866. AerialVLA: A Vision-Language-Action Model for UAV Navigation via Minimalist End-to-End Control
+
+- **ID:** `2603.14363` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Peng Xu, Zhengnan Deng, Jiayan Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.14363 · **PDF:** https://arxiv.org/pdf/2603.14363v1
+
+## 4867. M$^2$RNN: Non-Linear RNNs with Matrix-Valued States for Scalable Language Modeling
+
+- **ID:** `2603.14360` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Mayank Mishra, Shawn Tan, Ion Stoica, et al.
+- **abs:** https://arxiv.org/abs/2603.14360 · **PDF:** https://arxiv.org/pdf/2603.14360v1
+
+## 4868. Exposing Long-Tail Safety Failures in Large Language Models through Efficient Diverse Response Sampling
+
+- **ID:** `2603.14355` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Suvadeep Hajra, Palash Nandi, Tanmoy Chakraborty
+- **abs:** https://arxiv.org/abs/2603.14355 · **PDF:** https://arxiv.org/pdf/2603.14355v1
+
+## 4869. Deconfounded Lifelong Learning for Autonomous Driving via Dynamic Knowledge Spaces
+
+- **ID:** `2603.14354` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Jiayuan Du, Yuebing Song, Yiming Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.14354 · **PDF:** https://arxiv.org/pdf/2603.14354v2
+
+## 4870. Refold: Refining Protein Inverse Folding with Efficient Structural Matching and Fusion
+
+- **ID:** `2603.14350` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Yiran Zhu, Changxi Chi, Hongxin Xiang, et al.
+- **abs:** https://arxiv.org/abs/2603.14350 · **PDF:** https://arxiv.org/pdf/2603.14350v1
+
+## 4871. Motivation in Large Language Models
+
+- **ID:** `2603.14347` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Omer Nahum, Asael Sklar, Ariel Goldstein, et al.
+- **abs:** https://arxiv.org/abs/2603.14347 · **PDF:** https://arxiv.org/pdf/2603.14347v1
+
+## 4872. Localizing and Editing Knowledge in Large Audio-Language Models
+
+- **ID:** `2603.14343` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Sung Kyun Chung, Jiaheng Dong, Qiuchi Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.14343 · **PDF:** https://arxiv.org/pdf/2603.14343v1
+
+## 4873. AgroNVILA: Perception-Reasoning Decoupling for Multi-view Agricultural Multimodal Large Language Models
+
+- **ID:** `2603.14342` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Jiarui Zhang, Junqi Hu, Zurong Mai, et al.
+- **abs:** https://arxiv.org/abs/2603.14342 · **PDF:** https://arxiv.org/pdf/2603.14342v1
+
+## 4874. Generation of Human Comprehensible Access Control Policies from Audit Logs
+
+- **ID:** `2603.14341` · **primary:** `cs.CR` · **Submitted:** 2026-03-15
+- **Authors:** Gautam Kumar, Ravi Sundaram, Shamik Sural
+- **abs:** https://arxiv.org/abs/2603.14341 · **PDF:** https://arxiv.org/pdf/2603.14341v1
+
+## 4875. State-Dependent Safety Failures in Multi-Turn Language Model Interaction
+
+- **ID:** `2603.15684` · **primary:** `cs.CR` · **Submitted:** 2026-03-15
+- **Authors:** Pengcheng Li, Jie Zhang, Tianwei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.15684 · **PDF:** https://arxiv.org/pdf/2603.15684v1
+
+## 4876. Data-Driven Physics Embedded Dynamics with Predictive Control and Reinforcement Learning for Quadrupeds
+
+- **ID:** `2603.14333` · **primary:** `cs.RO` · **Submitted:** 2026-03-15
+- **Authors:** Prakrut Kotecha, Aditya Shirwatkar, Shishir Kolathaya
+- **abs:** https://arxiv.org/abs/2603.14333 · **PDF:** https://arxiv.org/pdf/2603.14333v1
+
+## 4877. ECG-Reasoning-Benchmark: A Benchmark for Evaluating Clinical Reasoning Capabilities in ECG Interpretation
+
+- **ID:** `2603.14326` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Jungwoo Oh, Hyunseung Chung, Junhee Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.14326 · **PDF:** https://arxiv.org/pdf/2603.14326v1
+
+## 4878. Fundamental Limits of CSI Compression in FDD Massive MIMO
+
+- **ID:** `2603.14325` · **primary:** `cs.IT` · **Submitted:** 2026-03-15
+- **Authors:** Bumsu Park, Youngmok Park, Chanho Park, et al.
+- **abs:** https://arxiv.org/abs/2603.14325 · **PDF:** https://arxiv.org/pdf/2603.14325v1
+
+## 4879. Learning-to-Defer with Expert-Conditioned Advice
+
+- **ID:** `2603.14324` · **primary:** `stat.ML` · **Submitted:** 2026-03-15
+- **Authors:** Yannis Montreuil, Leïna Montreuil, Axel Carlier, et al.
+- **abs:** https://arxiv.org/abs/2603.14324 · **PDF:** https://arxiv.org/pdf/2603.14324v2
+
+## 4880. How Do Medical MLLMs Fail? A Study on Visual Grounding in Medical Images
+
+- **ID:** `2603.14323` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Guimeng Liu, Tianze Yu, Somayeh Ebrahimkhani, et al.
+- **abs:** https://arxiv.org/abs/2603.14323 · **PDF:** https://arxiv.org/pdf/2603.14323v1
+
+## 4881. DeepOFW: Deep Learning-Driven OFDM-Flexible Waveform Modulation for Peak-to-Average Power Ratio Reduction
+
+- **ID:** `2603.23544` · **primary:** `cs.IT` · **Submitted:** 2026-03-15
+- **Authors:** Ran Greidi, Kobi Cohen
+- **abs:** https://arxiv.org/abs/2603.23544 · **PDF:** https://arxiv.org/pdf/2603.23544v1
+
+## 4882. Structure-Dependent Regret and Constraint Violation Bounds for Online Convex Optimization with Time-Varying Constraints
+
+- **ID:** `2603.14319` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Xiufeng Liu, Qian Chen, Zhijin Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.14319 · **PDF:** https://arxiv.org/pdf/2603.14319v1
+
+## 4883. Enhancing LLM Training via Spectral Clipping
+
+- **ID:** `2603.14315` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Xiaowen Jiang, Andrei Semenov, Sebastian U. Stich
+- **abs:** https://arxiv.org/abs/2603.14315 · **PDF:** https://arxiv.org/pdf/2603.14315v1
+
+## 4884. Mind the Shift: Decoding Monetary Policy Stance from FOMC Statements with Large Language Models
+
+- **ID:** `2603.14313` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Yixuan Tang, Yi Yang
+- **abs:** https://arxiv.org/abs/2603.14313 · **PDF:** https://arxiv.org/pdf/2603.14313v1
+
+## 4885. Autonomous Agents Coordinating Distributed Discovery Through Emergent Artifact Exchange
+
+- **ID:** `2603.14312` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Fiona Y. Wang, Lee Marom, Subhadeep Pal, et al.
+- **abs:** https://arxiv.org/abs/2603.14312 · **PDF:** https://arxiv.org/pdf/2603.14312v1
+
+## 4886. FactorSmith: Agentic Simulation Generation via Markov Decision Process Decomposition with Planner-Designer-Critic Refinement
+
+- **ID:** `2603.20270` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Ali Shamsaddinlou, Morteza NourelahiAlamdari
+- **abs:** https://arxiv.org/abs/2603.20270 · **PDF:** https://arxiv.org/pdf/2603.20270v1
+
+## 4887. SemantiCache: Efficient KV Cache Compression via Semantic Chunking and Clustered Merging
+
+- **ID:** `2603.14303` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Shunlong Wu, Hai Lin, Shaoshen Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.14303 · **PDF:** https://arxiv.org/pdf/2603.14303v1
+
+## 4888. 4D Synchronized Fields: Motion-Language Gaussian Splatting for Temporal Scene Understanding
+
+- **ID:** `2603.14301` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Mohamed Rayan Barhdadi, Samir Abdaljalil, Rasul Khanbayov, et al.
+- **abs:** https://arxiv.org/abs/2603.14301 · **PDF:** https://arxiv.org/pdf/2603.14301v1
+
+## 4889. Seeking Physics in Diffusion Noise
+
+- **ID:** `2603.14294` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Chujun Tang, Lei Zhong, Fangqiang Ding
+- **abs:** https://arxiv.org/abs/2603.14294 · **PDF:** https://arxiv.org/pdf/2603.14294v2
+
+## 4890. Windowed Fourier Propagator: A Frequency-Local Neural Operator for Wave Equations in Inhomogeneous Media
+
+- **ID:** `2603.14289` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Yiyang Cai, Zixuan Qiu, Yunlu Shu, et al.
+- **abs:** https://arxiv.org/abs/2603.14289 · **PDF:** https://arxiv.org/pdf/2603.14289v1
+
+## 4891. MorphSNN: Adaptive Graph Diffusion and Structural Plasticity for Spiking Neural Networks
+
+- **ID:** `2603.14285` · **primary:** `cs.NE` · **Submitted:** 2026-03-15
+- **Authors:** Yongsheng Huang, Peibo Duan, Yujie Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.14285 · **PDF:** https://arxiv.org/pdf/2603.14285v1
+
+## 4892. High-Fidelity Compression of Seismic Velocity Models via SIREN Auto-Decoders
+
+- **ID:** `2603.14284` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Caiyun Liu, Xiaoxue Luo, Jie Xiong
+- **abs:** https://arxiv.org/abs/2603.14284 · **PDF:** https://arxiv.org/pdf/2603.14284v2
+
+## 4893. AEX: Non-Intrusive Multi-Hop Attestation and Provenance for LLM APIs
+
+- **ID:** `2603.14283` · **primary:** `cs.CR` · **Submitted:** 2026-03-15
+- **Authors:** Yongjie Guan
+- **abs:** https://arxiv.org/abs/2603.14283 · **PDF:** https://arxiv.org/pdf/2603.14283v1
+
+## 4894. All-day Multi-scenes Lifelong Vision-and-Language Navigation with Tucker Adaptation
+
+- **ID:** `2603.14276` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Xudong Wang, Gan Li, Zhiyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.14276 · **PDF:** https://arxiv.org/pdf/2603.14276v1
+
+## 4895. Controllable Accent Normalization via Discrete Diffusion
+
+- **ID:** `2603.14275` · **primary:** `eess.AS` · **Submitted:** 2026-03-15
+- **Authors:** Qibing Bai, Yuhan Du, Tom Ko, et al.
+- **abs:** https://arxiv.org/abs/2603.14275 · **PDF:** https://arxiv.org/pdf/2603.14275v1
+
+## 4896. Learning in Function Spaces: An Unified Functional Analytic View of Supervised and Unsupervised Learning
+
+- **ID:** `2603.14272` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** K. Lakshmanan
+- **abs:** https://arxiv.org/abs/2603.14272 · **PDF:** https://arxiv.org/pdf/2603.14272v1
+
+## 4897. Beyond Distance: Quantifying Point Cloud Dynamics with Persistent Homology and Dynamic Optimal Transport
+
+- **ID:** `2603.15683` · **primary:** `stat.ML` · **Submitted:** 2026-03-15
+- **Authors:** Yixin Wang, Ting Gao, Jinqiao Duan
+- **abs:** https://arxiv.org/abs/2603.15683 · **PDF:** https://arxiv.org/pdf/2603.15683v1
+
+## 4898. DiFlowDubber: Discrete Flow Matching for Automated Video Dubbing via Cross-Modal Alignment and Synchronization
+
+- **ID:** `2603.14267` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Ngoc-Son Nguyen, Thanh V. T. Tran, Jeongsoo Choi, et al.
+- **abs:** https://arxiv.org/abs/2603.14267 · **PDF:** https://arxiv.org/pdf/2603.14267v3
+
+## 4899. MedPriv-Bench: Benchmarking the Privacy-Utility Trade-off of Large Language Models in Medical Open-End Question Answering
+
+- **ID:** `2603.14265` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Shaowei Guan, Yu Zhai, Hin Chi Kwok, et al.
+- **abs:** https://arxiv.org/abs/2603.14265 · **PDF:** https://arxiv.org/pdf/2603.14265v1
+
+## 4900. Bringing Model Editing to Generative Recommendation in Cold-Start Scenarios
+
+- **ID:** `2603.14259` · **primary:** `cs.IR` · **Submitted:** 2026-03-15
+- **Authors:** Chenglei Shen, Teng Shi, Weijie Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.14259 · **PDF:** https://arxiv.org/pdf/2603.14259v1
+
+## 4901. Sampling Boltzmann distributions via normalizing flow approximation of transport maps
+
+- **ID:** `2603.14258` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Zia Ur Rehman, Gero Friesecke
+- **abs:** https://arxiv.org/abs/2603.14258 · **PDF:** https://arxiv.org/pdf/2603.14258v1
+
+## 4902. Automatic Inter-document Multi-hop Scientific QA Generation
+
+- **ID:** `2603.14257` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Seungmin Lee, Dongha Kim, Yuni Jeon, et al.
+- **abs:** https://arxiv.org/abs/2603.14257 · **PDF:** https://arxiv.org/pdf/2603.14257v1
+
+## 4903. ZOTTA: Test-Time Adaptation with Gradient-Free Zeroth-Order Optimization
+
+- **ID:** `2603.14254` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Ronghao Zhang, Shuaicheng Niu, Qi Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.14254 · **PDF:** https://arxiv.org/pdf/2603.14254v1
+
+## 4904. Mitigating Overthinking in Large Reasoning Language Models via Reasoning Path Deviation Monitoring
+
+- **ID:** `2603.14251` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Weixin Guan, Liang Li, Jiapeng Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.14251 · **PDF:** https://arxiv.org/pdf/2603.14251v1
+
+## 4905. Why Do LLM-based Web Agents Fail? A Hierarchical Planning Perspective
+
+- **ID:** `2603.14248` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Mohamed Aghzal, Gregory J. Stein, Ziyu Yao
+- **abs:** https://arxiv.org/abs/2603.14248 · **PDF:** https://arxiv.org/pdf/2603.14248v1
+
+## 4906. GoldenStart: Q-Guided Priors and Entropy Control for Distilling Flow Policies
+
+- **ID:** `2603.14245` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** He Zhang, Ying Sun, Hui Xiong
+- **abs:** https://arxiv.org/abs/2603.14245 · **PDF:** https://arxiv.org/pdf/2603.14245v1
+
+## 4907. QiMeng-CodeV-SVA: Training Specialized LLMs for Hardware Assertion Generation via RTL-Grounded Bidirectional Data Synthesis
+
+- **ID:** `2603.14239` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Yutong Wu, Chenrui Cao, Pengwei Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.14239 · **PDF:** https://arxiv.org/pdf/2603.14239v1
+
+## 4908. Domain-Skewed Federated Learning with Feature Decoupling and Calibration
+
+- **ID:** `2603.14238` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Huan Wang, Jun Shen, Jun Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.14238 · **PDF:** https://arxiv.org/pdf/2603.14238v1
+
+## 4909. Agentic DAG-Orchestrated Planner Framework for Multi-Modal, Multi-Hop Question Answering in Hybrid Data Lakes
+
+- **ID:** `2603.14229` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Kirushikesh D B, Manish Kesarwani, Nishtha Madaan, et al.
+- **abs:** https://arxiv.org/abs/2603.14229 · **PDF:** https://arxiv.org/pdf/2603.14229v1
+
+## 4910. "I'm Not Reading All of That": Understanding Software Engineers' Level of Cognitive Engagement with Agentic Coding Assistants
+
+- **ID:** `2603.14225` · **primary:** `cs.HC` · **Submitted:** 2026-03-15
+- **Authors:** Carlos Rafael Catalan, Lheane Marie Dizon, Patricia Nicole Monderin, et al.
+- **abs:** https://arxiv.org/abs/2603.14225 · **PDF:** https://arxiv.org/pdf/2603.14225v2
+
+## 4911. Self-Indexing KVCache: Predicting Sparse Attention from Compressed Keys
+
+- **ID:** `2603.14224` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Xu Yang, Jiapeng Zhang, Dongyang Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.14224 · **PDF:** https://arxiv.org/pdf/2603.14224v1
+
+## 4912. Membership Inference for Contrastive Pre-training Models with Text-only PII Queries
+
+- **ID:** `2603.14222` · **primary:** `cs.CR` · **Submitted:** 2026-03-15
+- **Authors:** Ruoxi Cheng, Yizhong Ding, Hongyi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.14222 · **PDF:** https://arxiv.org/pdf/2603.14222v1
+
+## 4913. A Real-Time Neuro-Symbolic Ethical Governor for Safe Decision Control in Autonomous Robotic Manipulation
+
+- **ID:** `2603.14221` · **primary:** `cs.RO` · **Submitted:** 2026-03-15
+- **Authors:** Aueaphum Aueawatthanaphisut, Kuepon Aueawatthanaphisut
+- **abs:** https://arxiv.org/abs/2603.14221 · **PDF:** https://arxiv.org/pdf/2603.14221v1
+
+## 4914. Interleaved Resampling and Refitting: Data and Compute-Efficient Evaluation of Black-Box Predictors
+
+- **ID:** `2603.14218` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Haichen Hu, David Simchi-Levi
+- **abs:** https://arxiv.org/abs/2603.14218 · **PDF:** https://arxiv.org/pdf/2603.14218v2
+
+## 4915. Rethinking Evaluation in Retrieval-Augmented Personalized Dialogue: A Cognitive and Linguistic Perspective
+
+- **ID:** `2603.14217` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Tianyi Zhang, David Traum
+- **abs:** https://arxiv.org/abs/2603.14217 · **PDF:** https://arxiv.org/pdf/2603.14217v2
+
+## 4916. UniFusion: A Unified Image Fusion Framework with Robust Representation and Source-Aware Preservation
+
+- **ID:** `2603.14214` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Xingyuan Li, Songcheng Du, Yang Zou, et al.
+- **abs:** https://arxiv.org/abs/2603.14214 · **PDF:** https://arxiv.org/pdf/2603.14214v1
+
+## 4917. Cryptographic Runtime Governance for Autonomous AI Systems: The Aegis Architecture for Verifiable Policy Enforcement
+
+- **ID:** `2603.16938` · **primary:** `cs.CR` · **Submitted:** 2026-03-15
+- **Authors:** Adam Massimo Mazzocchetti
+- **abs:** https://arxiv.org/abs/2603.16938 · **PDF:** https://arxiv.org/pdf/2603.16938v1
+
+## 4918. Memory as Asset: From Agent-centric to Human-centric Memory Management
+
+- **ID:** `2603.14212` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Yanqi Pan, Qinghao Huang, Weihao Yang
+- **abs:** https://arxiv.org/abs/2603.14212 · **PDF:** https://arxiv.org/pdf/2603.14212v1
+
+## 4919. Vavanagi: a Community-run Platform for Documentation of the Hula Language in Papua New Guinea
+
+- **ID:** `2603.14210` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Bri Olewale, Raphael Merx, Ekaterina Vylomova
+- **abs:** https://arxiv.org/abs/2603.14210 · **PDF:** https://arxiv.org/pdf/2603.14210v1
+
+## 4920. ChArtist: Generating Pictorial Charts with Unified Spatial and Subject Control
+
+- **ID:** `2603.14209` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Shishi Xiao, Tongyu Zhou, David Laidlaw, et al.
+- **abs:** https://arxiv.org/abs/2603.14209 · **PDF:** https://arxiv.org/pdf/2603.14209v1
+
+## 4921. DualTSR: Unified Dual-Diffusion Transformer for Scene Text Image Super-Resolution
+
+- **ID:** `2603.14207` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Axi Niu, Kang Zhang, Qingsen Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.14207 · **PDF:** https://arxiv.org/pdf/2603.14207v1
+
+## 4922. Efficient Federated Conformal Prediction with Group-Conditional Guarantees
+
+- **ID:** `2603.14198` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Haifeng Wen, Osvaldo Simeone, Hong Xing
+- **abs:** https://arxiv.org/abs/2603.14198 · **PDF:** https://arxiv.org/pdf/2603.14198v2
+
+## 4923. Flood Risk Follows Valleys, Not Grids: Graph Neural Networks for Flash Flood Susceptibility Mapping in Himachal Pradesh with Conformal Uncertainty Quantification
+
+- **ID:** `2603.15681` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Paras Sharma, Swastika Sharma
+- **abs:** https://arxiv.org/abs/2603.15681 · **PDF:** https://arxiv.org/pdf/2603.15681v1
+
+## 4924. Walking Further: Semantic-aware Multimodal Gait Recognition Under Long-Range Conditions
+
+- **ID:** `2603.14189` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Zhiyang Lu, Wen Jiang, Tianren Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.14189 · **PDF:** https://arxiv.org/pdf/2603.14189v1
+
+## 4925. Relationship-Aware Safety Unlearning for Multimodal LLMs
+
+- **ID:** `2603.14185` · **primary:** `cs.AI` · **Submitted:** 2026-03-15
+- **Authors:** Vishnu Narayanan Anilkumar, Abhijith Sreesylesh Babu, Trieu Hai Vo, et al.
+- **abs:** https://arxiv.org/abs/2603.14185 · **PDF:** https://arxiv.org/pdf/2603.14185v3
+
+## 4926. Deeper Thought, Weaker Aim: Understanding and Mitigating Perceptual Impairment during Reasoning in Multimodal Large Language Models
+
+- **ID:** `2603.14184` · **primary:** `cs.CV` · **Submitted:** 2026-03-15
+- **Authors:** Ruiying Peng, Xueyu Wu, Jing Lei, et al.
+- **abs:** https://arxiv.org/abs/2603.14184 · **PDF:** https://arxiv.org/pdf/2603.14184v1
+
+## 4927. Selective Fine-Tuning of GPT Architectures for Parameter-Efficient Clinical Text Classification
+
+- **ID:** `2603.14183` · **primary:** `cs.CL` · **Submitted:** 2026-03-15
+- **Authors:** Fariba Afrin Irany, Sampson Akwafuo
+- **abs:** https://arxiv.org/abs/2603.14183 · **PDF:** https://arxiv.org/pdf/2603.14183v1
+
+## 4928. Artificial intelligence-enabled single-lead ECG for non-invasive hyperkalemia detection: development, multicenter validation, and proof-of-concept deployment
+
+- **ID:** `2603.14177` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Gongzheng Tang, Qinghao Zhao, Guangkun Nie, et al.
+- **abs:** https://arxiv.org/abs/2603.14177 · **PDF:** https://arxiv.org/pdf/2603.14177v2
+
+## 4929. Balancing Multimodal Domain Generalization via Gradient Modulation and Projection
+
+- **ID:** `2603.14175` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Hongzhao Li, Guohao Shen, Shupan Li, et al.
+- **abs:** https://arxiv.org/abs/2603.14175 · **PDF:** https://arxiv.org/pdf/2603.14175v1
+
+## 4930. TACTIC for Navigating the Unknown: Tabular Anomaly deteCTion via In-Context inference
+
+- **ID:** `2603.14171` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Patryk Marszałek, Tomasz Kuśmierczyk, Marek Śmieja
+- **abs:** https://arxiv.org/abs/2603.14171 · **PDF:** https://arxiv.org/pdf/2603.14171v1
+
+## 4931. Beyond Means: Topological Causal Effects under Persistent-Homology Ignorability
+
+- **ID:** `2603.14169` · **primary:** `stat.ME` · **Submitted:** 2026-03-15
+- **Authors:** Amir Saki, Usef Faghihi
+- **abs:** https://arxiv.org/abs/2603.14169 · **PDF:** https://arxiv.org/pdf/2603.14169v1
+
+## 4932. Deep probabilistic model synthesis enables unified modeling of whole-brain neural activity across individual subjects
+
+- **ID:** `2603.14161` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** William E. Bishop, Luuk W. Hesselink, Bernhard Englitz, et al.
+- **abs:** https://arxiv.org/abs/2603.14161 · **PDF:** https://arxiv.org/pdf/2603.14161v1
+
+## 4933. Efficient Embedding-based Synthetic Data Generation for Complex Reasoning Tasks
+
+- **ID:** `2603.22294` · **primary:** `cs.LG` · **Submitted:** 2026-03-15
+- **Authors:** Srideepika Jayaraman, Achille Fokoue, Dhaval Patel, et al.
+- **abs:** https://arxiv.org/abs/2603.22294 · **PDF:** https://arxiv.org/pdf/2603.22294v1
+
+## 4934. Clinician input steers frontier AI models toward both accurate and harmful decisions
+
+- **ID:** `2603.14158` · **primary:** `cs.HC` · **Submitted:** 2026-03-14
+- **Authors:** Ivan Lopez, Selin S. Everett, Bryan J. Bunning, et al.
+- **abs:** https://arxiv.org/abs/2603.14158 · **PDF:** https://arxiv.org/pdf/2603.14158v1
+
+## 4935. Align Forward, Adapt Backward: Closing the Discretization Gap in Logic Gate Networks
+
+- **ID:** `2603.14157` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Youngsung Kim
+- **abs:** https://arxiv.org/abs/2603.14157 · **PDF:** https://arxiv.org/pdf/2603.14157v1
+
+## 4936. An Alternative Trajectory for Generative AI
+
+- **ID:** `2603.14147` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Margarita Belova, Yuval Kansal, Yihao Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.14147 · **PDF:** https://arxiv.org/pdf/2603.14147v1
+
+## 4937. MMOU: A Massive Multi-Task Omni Understanding and Reasoning Benchmark for Long and Complex Real-World Videos
+
+- **ID:** `2603.14145` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Arushi Goel, Sreyan Ghosh, Vatsal Agarwal, et al.
+- **abs:** https://arxiv.org/abs/2603.14145 · **PDF:** https://arxiv.org/pdf/2603.14145v1
+
+## 4938. Multifidelity Surrogate Modeling of Depressurized Loss of Forced Cooling in High-temperature Gas Reactors
+
+- **ID:** `2603.14143` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Meredith Eaheart, Majdi I. Radaideh
+- **abs:** https://arxiv.org/abs/2603.14143 · **PDF:** https://arxiv.org/pdf/2603.14143v1
+
+## 4939. Quotient Geometry and Persistence-Stable Metrics for Swarm Configurations
+
+- **ID:** `2603.18041` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Mark M. Bailey
+- **abs:** https://arxiv.org/abs/2603.18041 · **PDF:** https://arxiv.org/pdf/2603.18041v1
+
+## 4940. Solving physics-constrained inverse problems with conditional flow matching
+
+- **ID:** `2603.14135` · **primary:** `stat.ML` · **Submitted:** 2026-03-14
+- **Authors:** Agnimitra Dasgupta, Ali Fardisi, Mehrnegar Aminy, et al.
+- **abs:** https://arxiv.org/abs/2603.14135 · **PDF:** https://arxiv.org/pdf/2603.14135v2
+
+## 4941. DualSwinFusionSeg: Multimodal Martian Landslide Segmentation via Dual Swin Transformer with Multi-Scale Fusion and UNet++
+
+- **ID:** `2603.14132` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Shahriar Kabir, Abdullah Muhammed Amimul Ehsan, Istiak Ahmmed Rifti, et al.
+- **abs:** https://arxiv.org/abs/2603.14132 · **PDF:** https://arxiv.org/pdf/2603.14132v1
+
+## 4942. Is the reconstruction loss culprit? An attempt to outperform JEPA
+
+- **ID:** `2603.14131` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Alexey Potapov, Oleg Shcherbakov, Ivan Kravchenko
+- **abs:** https://arxiv.org/abs/2603.14131 · **PDF:** https://arxiv.org/pdf/2603.14131v1
+
+## 4943. The GELATO Dataset for Legislative NER
+
+- **ID:** `2603.14130` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Matthew Flynn, Timothy Obiso, Sam Newman
+- **abs:** https://arxiv.org/abs/2603.14130 · **PDF:** https://arxiv.org/pdf/2603.14130v1
+
+## 4944. Diffusion Reinforcement Learning via Centered Reward Distillation
+
+- **ID:** `2603.14128` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Yuanzhi Zhu, Xi Wang, Stéphane Lathuilière, et al.
+- **abs:** https://arxiv.org/abs/2603.14128 · **PDF:** https://arxiv.org/pdf/2603.14128v1
+
+## 4945. The Institutional Scaling Law: Non-Monotonic Fitness, Capability-Trust Divergence, and Symbiogenetic Scaling in Generative AI
+
+- **ID:** `2603.14126` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Mark Baciak, Thomas A. Cellucci
+- **abs:** https://arxiv.org/abs/2603.14126 · **PDF:** https://arxiv.org/pdf/2603.14126v1
+
+## 4946. Towards Agentic Honeynet Configuration
+
+- **ID:** `2603.14122` · **primary:** `cs.CR` · **Submitted:** 2026-03-14
+- **Authors:** Federico Mirra, Matteo Boffa, Idilio Drago, et al.
+- **abs:** https://arxiv.org/abs/2603.14122 · **PDF:** https://arxiv.org/pdf/2603.14122v1
+
+## 4947. OasisSimp: An Open-source Asian-English Sentence Simplification Dataset
+
+- **ID:** `2603.14111` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Hannah Liu, Muxin Tian, Iqra Ali, et al.
+- **abs:** https://arxiv.org/abs/2603.14111 · **PDF:** https://arxiv.org/pdf/2603.14111v1
+
+## 4948. SVD Contextual Sparsity Predictors for Fast LLM Inference
+
+- **ID:** `2603.14110` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Georgii Serbin, Kirill Koshkin, Zhongao Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.14110 · **PDF:** https://arxiv.org/pdf/2603.14110v1
+
+## 4949. ST-ResGAT: Explainable Spatio-Temporal Graph Neural Network for Road Condition Prediction and Priority-Driven Maintenance
+
+- **ID:** `2603.14107` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Mohsin Mahmud Topu, Azmine Toushik Wasi, Mahfuz Ahmed Anik, et al.
+- **abs:** https://arxiv.org/abs/2603.14107 · **PDF:** https://arxiv.org/pdf/2603.14107v1
+
+## 4950. Scorio.jl: A Julia package for ranking stochastic responses
+
+- **ID:** `2603.14103` · **primary:** `cs.MS` · **Submitted:** 2026-03-14
+- **Authors:** Mohsen Hariri, Michael Hinczewski, Vipin Chaudhary
+- **abs:** https://arxiv.org/abs/2603.14103 · **PDF:** https://arxiv.org/pdf/2603.14103v1
+
+## 4951. Integrating Explainable Machine Learning and Mixed-Integer Optimization for Personalized Sleep Quality Intervention
+
+- **ID:** `2603.16937` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Mahfuz Ahmed Anik, Mohsin Mahmud Topu, Azmine Toushik Wasi, et al.
+- **abs:** https://arxiv.org/abs/2603.16937 · **PDF:** https://arxiv.org/pdf/2603.16937v1
+
+## 4952. Concisely Explaining the Doubt: Minimum-Size Abductive Explanations for Linear Models with a Reject Option
+
+- **ID:** `2603.14096` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Gleilson Pedro Fernandes, Thiago Alves Rocha
+- **abs:** https://arxiv.org/abs/2603.14096 · **PDF:** https://arxiv.org/pdf/2603.14096v1
+
+## 4953. Maximin Robust Bayesian Experimental Design
+
+- **ID:** `2603.14094` · **primary:** `stat.ML` · **Submitted:** 2026-03-14
+- **Authors:** Hany Abdulsamad, Sahel Iqbal, Christian A. Naesseth, et al.
+- **abs:** https://arxiv.org/abs/2603.14094 · **PDF:** https://arxiv.org/pdf/2603.14094v1
+
+## 4954. Not All Latent Spaces Are Flat: Hyperbolic Concept Control
+
+- **ID:** `2603.14093` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Maria Rosaria Briglia, Simone Facchiano, Paolo Cursi, et al.
+- **abs:** https://arxiv.org/abs/2603.14093 · **PDF:** https://arxiv.org/pdf/2603.14093v2
+
+## 4955. Soft Mean Expected Calibration Error (SMECE): A Calibration Metric for Probabilistic Labels
+
+- **ID:** `2603.14092` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Michael Leznik
+- **abs:** https://arxiv.org/abs/2603.14092 · **PDF:** https://arxiv.org/pdf/2603.14092v1
+
+## 4956. Evaluating Four FPGA-accelerated Space Use Cases based on Neural Network Algorithms for On-board Inference
+
+- **ID:** `2603.14091` · **primary:** `cs.AR` · **Submitted:** 2026-03-14
+- **Authors:** Pedro Antunes, Muhammad Ihsan Al Hafiz, Jonah Ekelund, et al.
+- **abs:** https://arxiv.org/abs/2603.14091 · **PDF:** https://arxiv.org/pdf/2603.14091v1
+
+## 4957. Understanding the Emergence of Seemingly Useless Features in Next-Token Predictors
+
+- **ID:** `2603.14087` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Mark Rofin, Jalal Naghiyev, Michael Hahn
+- **abs:** https://arxiv.org/abs/2603.14087 · **PDF:** https://arxiv.org/pdf/2603.14087v1
+
+## 4958. Bootstrapped Physically-Primed Neural Networks for Robust T2 Distribution Estimation in Low-SNR Pancreatic MRI
+
+- **ID:** `2603.14084` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Hadas Ben Atya, Nicole Abramenkov, Noa Mashiah, et al.
+- **abs:** https://arxiv.org/abs/2603.14084 · **PDF:** https://arxiv.org/pdf/2603.14084v1
+
+## 4959. CMHL: Contrastive Multi-Head Learning for Emotionally Consistent Text Classification
+
+- **ID:** `2603.14078` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Menna Elgabry, Ali Hamdi, Khaled Shaban
+- **abs:** https://arxiv.org/abs/2603.14078 · **PDF:** https://arxiv.org/pdf/2603.14078v1
+
+## 4960. Enhancing Mental Health Classification with Layer-Attentive Residuals and Contrastive Feature Learning
+
+- **ID:** `2603.14075` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Menna Elgabry, Ali Hamdi, Khaled Shaban
+- **abs:** https://arxiv.org/abs/2603.14075 · **PDF:** https://arxiv.org/pdf/2603.14075v1
+
+## 4961. Self-Supervised Uncertainty Estimation For Super-Resolution of Satellite Images
+
+- **ID:** `2603.14074` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Zhe Zheng, Valéry Dewil, Pablo Arias
+- **abs:** https://arxiv.org/abs/2603.14074 · **PDF:** https://arxiv.org/pdf/2603.14074v1
+
+## 4962. MotionCFG: Boosting Motion Dynamics via Stochastic Concept Perturbation
+
+- **ID:** `2603.14073` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Byungjun Kim, Soobin Um, Jong Chul Ye
+- **abs:** https://arxiv.org/abs/2603.14073 · **PDF:** https://arxiv.org/pdf/2603.14073v1
+
+## 4963. Conditioning on a Volatility Proxy Compresses the Apparent Timescale of Collective Market Correlation
+
+- **ID:** `2603.14072` · **primary:** `q-fin.CP` · **Submitted:** 2026-03-14
+- **Authors:** Yuda Bi, Vince D Calhoun
+- **abs:** https://arxiv.org/abs/2603.14072 · **PDF:** https://arxiv.org/pdf/2603.14072v1
+
+## 4964. Structured Credal Learning
+
+- **ID:** `2603.14070` · **primary:** `stat.ME` · **Submitted:** 2026-03-14
+- **Authors:** Varun Venkatesh, Eyke Hüllermeier, Bernd Bischl, et al.
+- **abs:** https://arxiv.org/abs/2603.14070 · **PDF:** https://arxiv.org/pdf/2603.14070v1
+
+## 4965. Gated Graph Attention Networks for Predicting Duration of Large Scale Power Outages Induced by Natural Disasters
+
+- **ID:** `2603.14069` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Chenghao Duan, Chuanyi Ji, Anwar Walid, et al.
+- **abs:** https://arxiv.org/abs/2603.14069 · **PDF:** https://arxiv.org/pdf/2603.14069v1
+
+## 4966. A Benchmark for Multi-Party Negotiation Games from Real Negotiation Data
+
+- **ID:** `2603.14066` · **primary:** `cs.MA` · **Submitted:** 2026-03-14
+- **Authors:** Leo Benac, Jonas Raedler, Zilin Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.14066 · **PDF:** https://arxiv.org/pdf/2603.14066v1
+
+## 4967. TMPDiff: Temporal Mixed-Precision for Diffusion Models
+
+- **ID:** `2603.14062` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Basile Lewandowski, Simon Kurz, Aditya Shankar, et al.
+- **abs:** https://arxiv.org/abs/2603.14062 · **PDF:** https://arxiv.org/pdf/2603.14062v1
+
+## 4968. Demand-Driven Context: A Methodology for Building Enterprise Knowledge Bases Through Agent Failure
+
+- **ID:** `2603.14057` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Raj Navakoti, Saideep Navakoti
+- **abs:** https://arxiv.org/abs/2603.14057 · **PDF:** https://arxiv.org/pdf/2603.14057v1
+
+## 4969. NepTam: A Nepali-Tamang Parallel Corpus and Baseline Machine Translation Experiments
+
+- **ID:** `2603.14053` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Rupak Raj Ghimire, Bipesh Subedi, Balaram Prasain, et al.
+- **abs:** https://arxiv.org/abs/2603.14053 · **PDF:** https://arxiv.org/pdf/2603.14053v1
+
+## 4970. A Theory of Appropriateness That Accounts for Norms of Rationality
+
+- **ID:** `2603.14050` · **primary:** `cs.NE` · **Submitted:** 2026-03-14
+- **Authors:** Joel Z. Leibo, Alexander Sasha Vezhnevets, Manfred Diaz, et al.
+- **abs:** https://arxiv.org/abs/2603.14050 · **PDF:** https://arxiv.org/pdf/2603.14050v1
+
+## 4971. Schrödinger Bridge Over A Compact Connected Lie Group
+
+- **ID:** `2603.14049` · **primary:** `math.OC` · **Submitted:** 2026-03-14
+- **Authors:** Hamza Mahmood, Abhishek Halder, Adeel Akhtar
+- **abs:** https://arxiv.org/abs/2603.14049 · **PDF:** https://arxiv.org/pdf/2603.14049v2
+
+## 4972. The Reasoning Bottleneck in Graph-RAG: Structured Prompting and Context Compression for Multi-Hop QA
+
+- **ID:** `2603.14045` · **primary:** `cs.IR` · **Submitted:** 2026-03-14
+- **Authors:** Yasaman Zarrinkia, Venkatesh Srinivasan, Alex Thomo
+- **abs:** https://arxiv.org/abs/2603.14045 · **PDF:** https://arxiv.org/pdf/2603.14045v2
+
+## 4973. GRPO and Reflection Reward for Mathematical Reasoning in Large Language Models
+
+- **ID:** `2603.14041` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Zhijie Wang
+- **abs:** https://arxiv.org/abs/2603.14041 · **PDF:** https://arxiv.org/pdf/2603.14041v1
+
+## 4974. Probing neural audio codecs for distinctions among English nuclear tunes
+
+- **ID:** `2603.14035` · **primary:** `cs.SD` · **Submitted:** 2026-03-14
+- **Authors:** Juan Pablo Vigneaux, Jennifer Cole
+- **abs:** https://arxiv.org/abs/2603.14035 · **PDF:** https://arxiv.org/pdf/2603.14035v1
+
+## 4975. What Counts as Real? Speech Restoration and Voice Quality Conversion Pose New Challenges to Deepfake Detection
+
+- **ID:** `2603.14033` · **primary:** `cs.SD` · **Submitted:** 2026-03-14
+- **Authors:** Shree Harsha Bokkahalli Satish, Harm Lameris, Joakim Gustafson, et al.
+- **abs:** https://arxiv.org/abs/2603.14033 · **PDF:** https://arxiv.org/pdf/2603.14033v1
+
+## 4976. Benchmarking Open-Source PPG Foundation Models for Biological Age Prediction
+
+- **ID:** `2603.14030` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** N. Brag
+- **abs:** https://arxiv.org/abs/2603.14030 · **PDF:** https://arxiv.org/pdf/2603.14030v1
+
+## 4977. Traffic and weather driven hybrid digital twin for bridge monitoring
+
+- **ID:** `2603.14028` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Phani Raja Bharath Balijepalli, Bulent Soykan, Veeraraghava Raju Hasti
+- **abs:** https://arxiv.org/abs/2603.14028 · **PDF:** https://arxiv.org/pdf/2603.14028v1
+
+## 4978. SemEval-2026 Task 6: CLARITY -- Unmasking Political Question Evasions
+
+- **ID:** `2603.14027` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Konstantinos Thomas, Giorgos Filandrianos, Maria Lymperaiou, et al.
+- **abs:** https://arxiv.org/abs/2603.14027 · **PDF:** https://arxiv.org/pdf/2603.14027v1
+
+## 4979. EI-Part: Explode for Completion and Implode for Refinement
+
+- **ID:** `2603.14021` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Wanhu Sun, Zhongjin Luo, Heliang Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.14021 · **PDF:** https://arxiv.org/pdf/2603.14021v1
+
+## 4980. Aumann-SHAP: The Geometry of Counterfactual Interaction Explanations in Machine Learning
+
+- **ID:** `2603.14014` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Adam Belahcen, Stéphane Mussard
+- **abs:** https://arxiv.org/abs/2603.14014 · **PDF:** https://arxiv.org/pdf/2603.14014v1
+
+## 4981. Formal Abductive Explanations for Navigating Mental Health Help-Seeking and Diversity in Tech Workplaces
+
+- **ID:** `2603.14007` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Belona Sonna, Alain Momo, Alban Grastien
+- **abs:** https://arxiv.org/abs/2603.14007 · **PDF:** https://arxiv.org/pdf/2603.14007v1
+
+## 4982. Beyond Explicit Edges: Robust Reasoning over Noisy and Sparse Knowledge Graphs
+
+- **ID:** `2603.14006` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Hang Gao, Dimitris N. Metaxas
+- **abs:** https://arxiv.org/abs/2603.14006 · **PDF:** https://arxiv.org/pdf/2603.14006v1
+
+## 4983. U-Face: An Efficient and Generalizable Framework for Unsupervised Facial Attribute Editing via Subspace Learning
+
+- **ID:** `2603.14004` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Bo Liu, Xuan Cui, Run Zeng, et al.
+- **abs:** https://arxiv.org/abs/2603.14004 · **PDF:** https://arxiv.org/pdf/2603.14004v1
+
+## 4984. The Taxonomies, Training, and Applications of Event Stream Modelling for Electronic Health Records
+
+- **ID:** `2603.14003` · **primary:** `eess.SP` · **Submitted:** 2026-03-14
+- **Authors:** Mingcheng Zhu, Yu Liu, Zhiyao Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.14003 · **PDF:** https://arxiv.org/pdf/2603.14003v1
+
+## 4985. A Systematic Evaluation Protocol of Graph-Derived Signals for Tabular Machine Learning
+
+- **ID:** `2603.13998` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Mario Heidrich, Jeffrey Heidemann, Rüdiger Buchkremer, et al.
+- **abs:** https://arxiv.org/abs/2603.13998 · **PDF:** https://arxiv.org/pdf/2603.13998v1
+
+## 4986. Location Aware Embedding for Geotargeting in Sponsored Search Advertising
+
+- **ID:** `2603.13997` · **primary:** `cs.IR` · **Submitted:** 2026-03-14
+- **Authors:** Jelena Gligorijevic, Djordje Gligorijevic, Aravindan Raghuveer, et al.
+- **abs:** https://arxiv.org/abs/2603.13997 · **PDF:** https://arxiv.org/pdf/2603.13997v1
+
+## 4987. TDMM-LM: Bridging Facial Understanding and Animation via Language Models
+
+- **ID:** `2603.16936` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Luchuan Song, Pinxin Liu, Haiyang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.16936 · **PDF:** https://arxiv.org/pdf/2603.16936v1
+
+## 4988. Human-like Object Grouping in Self-supervised Vision Transformers
+
+- **ID:** `2603.13994` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Hossein Adeli, Seoyoung Ahn, Andrew Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.13994 · **PDF:** https://arxiv.org/pdf/2603.13994v1
+
+## 4989. VAD4Space: Visual Anomaly Detection for Planetary Surface Imagery
+
+- **ID:** `2603.13993` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Fabrizio Genilotti, Arianna Stropeni, Francesco Borsatti, et al.
+- **abs:** https://arxiv.org/abs/2603.13993 · **PDF:** https://arxiv.org/pdf/2603.13993v1
+
+## 4990. Faithful or Just Plausible? Evaluating the Faithfulness of Closed-Source LLMs in Medical Reasoning
+
+- **ID:** `2603.13988` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Halimat Afolabi, Zainab Afolabi, Elizabeth Friel, et al.
+- **abs:** https://arxiv.org/abs/2603.13988 · **PDF:** https://arxiv.org/pdf/2603.13988v1
+
+## 4991. Supervised Fine-Tuning versus Reinforcement Learning: A Study of Post-Training Methods for Large Language Models
+
+- **ID:** `2603.13985` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Haitao Jiang, Wenbo Zhang, Jiarui Yao, et al.
+- **abs:** https://arxiv.org/abs/2603.13985 · **PDF:** https://arxiv.org/pdf/2603.13985v1
+
+## 4992. Exploiting temporal parallelism for LSTM Autoencoder acceleration on FPGA
+
+- **ID:** `2603.13982` · **primary:** `cs.AR` · **Submitted:** 2026-03-14
+- **Authors:** Aimilios Leftheriotis, Dimosthenis Masouros, Dimitrios Soudris, et al.
+- **abs:** https://arxiv.org/abs/2603.13982 · **PDF:** https://arxiv.org/pdf/2603.13982v1
+
+## 4993. FLUX: Data Worth Training On
+
+- **ID:** `2603.13972` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Gowtham, Sai Rupesh, Sanjay Kumar, et al.
+- **abs:** https://arxiv.org/abs/2603.13972 · **PDF:** https://arxiv.org/pdf/2603.13972v1
+
+## 4994. Chunk-Guided Q-Learning
+
+- **ID:** `2603.13971` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Gwanwoo Song, Kwanyoung Park, Youngwoon Lee
+- **abs:** https://arxiv.org/abs/2603.13971 · **PDF:** https://arxiv.org/pdf/2603.13971v1
+
+## 4995. Shapes are not enough: CONSERVAttack and its use for finding vulnerabilities and uncertainties in machine learning applications
+
+- **ID:** `2603.13970` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Philip Bechtle, Lucie Flek, Philipp Alexander Jung, et al.
+- **abs:** https://arxiv.org/abs/2603.13970 · **PDF:** https://arxiv.org/pdf/2603.13970v1
+
+## 4996. EchoLVFM: One-Step Video Generation via Latent Flow Matching for Echocardiogram Synthesis
+
+- **ID:** `2603.13967` · **primary:** `eess.IV` · **Submitted:** 2026-03-14
+- **Authors:** Emmanuel Oladokun, Sarina Thomas, Jurica Šprem, et al.
+- **abs:** https://arxiv.org/abs/2603.13967 · **PDF:** https://arxiv.org/pdf/2603.13967v1
+
+## 4997. vla-eval: A Unified Evaluation Harness for Vision-Language-Action Models
+
+- **ID:** `2603.13966` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Suhwan Choi, Yunsung Lee, Yubeen Park, et al.
+- **abs:** https://arxiv.org/abs/2603.13966 · **PDF:** https://arxiv.org/pdf/2603.13966v1
+
+## 4998. sebis at ArchEHR-QA 2026: How Much Can You Do Locally? Evaluating Grounded EHR QA on a Single Notebook
+
+- **ID:** `2603.13962` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Ibrahim Ebrar Yurt, Fabian Karl, Tejaswi Choppa, et al.
+- **abs:** https://arxiv.org/abs/2603.13962 · **PDF:** https://arxiv.org/pdf/2603.13962v2
+
+## 4999. GenLie: A Global-Enhanced Lie Detection Network under Sparsity and Semantic Interference
+
+- **ID:** `2603.16935` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Zongshun Zhang, Yao Liu, Qiao Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.16935 · **PDF:** https://arxiv.org/pdf/2603.16935v1
+
+## 5000. EviAgent: Evidence-Driven Agent for Radiology Report Generation
+
+- **ID:** `2603.13956` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Tuoshi Qi, Shenshen Bu, Yingfei Xiang, et al.
+- **abs:** https://arxiv.org/abs/2603.13956 · **PDF:** https://arxiv.org/pdf/2603.13956v1
+
+## 5001. LLM-Guided Reinforcement Learning for Audio-Visual Speech Enhancement
+
+- **ID:** `2603.13952` · **primary:** `cs.SD` · **Submitted:** 2026-03-14
+- **Authors:** Chih-Ning Chen, Jen-Cheng Hou, Hsin-Min Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.13952 · **PDF:** https://arxiv.org/pdf/2603.13952v2
+
+## 5002. ToolFlood: Beyond Selection -- Hiding Valid Tools from LLM Agents via Semantic Covering
+
+- **ID:** `2603.13950` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Hussein Jawad, Nicolas J-B Brunel
+- **abs:** https://arxiv.org/abs/2603.13950 · **PDF:** https://arxiv.org/pdf/2603.13950v1
+
+## 5003. Sat-JEPA-Diff: Bridging Self-Supervised Learning and Generative Diffusion for Remote Sensing
+
+- **ID:** `2603.13943` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Kursat Komurcu, Linas Petkevicius
+- **abs:** https://arxiv.org/abs/2603.13943 · **PDF:** https://arxiv.org/pdf/2603.13943v1
+
+## 5004. GroupGuard: A Framework for Modeling and Defending Collusive Attacks in Multi-Agent Systems
+
+- **ID:** `2603.13940` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Yiling Tao, Xinran Zheng, Shuo Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.13940 · **PDF:** https://arxiv.org/pdf/2603.13940v1
+
+## 5005. Iterative Semantic Reasoning from Individual to Group Interests for Generative Recommendation with LLMs
+
+- **ID:** `2603.13934` · **primary:** `cs.IR` · **Submitted:** 2026-03-14
+- **Authors:** Xiaofei Zhu, Jinfei Chen, Feiyang Yuan, et al.
+- **abs:** https://arxiv.org/abs/2603.13934 · **PDF:** https://arxiv.org/pdf/2603.13934v1
+
+## 5006. OmniCompliance-100K: A Multi-Domain, Rule-Grounded, Real-World Safety Compliance Dataset
+
+- **ID:** `2603.13933` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Wenbin Hu, Huihao Jing, Haochen Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.13933 · **PDF:** https://arxiv.org/pdf/2603.13933v1
+
+## 5007. True 4-Bit Quantized Convolutional Neural Network Training on CPU: Achieving Full-Precision Parity
+
+- **ID:** `2603.13931` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Shivnath Tathe
+- **abs:** https://arxiv.org/abs/2603.13931 · **PDF:** https://arxiv.org/pdf/2603.13931v1
+
+## 5008. Discriminative Flow Matching Via Local Generative Predictors
+
+- **ID:** `2603.13928` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Om Govind Jha, Manoj Bamniya, Ayon Borthakur
+- **abs:** https://arxiv.org/abs/2603.13928 · **PDF:** https://arxiv.org/pdf/2603.13928v1
+
+## 5009. Close to Reality: Interpretable and Feasible Data Augmentation for Imbalanced Learning
+
+- **ID:** `2603.13927` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Matheus Camilo da Silva, Gabriel Gustavo Costanzo, Andrea de Lorenzo, et al.
+- **abs:** https://arxiv.org/abs/2603.13927 · **PDF:** https://arxiv.org/pdf/2603.13927v1
+
+## 5010. SmoothVLA: Aligning Vision-Language-Action Models with Physical Constraints via Intrinsic Smoothness Optimization
+
+- **ID:** `2603.13925` · **primary:** `cs.RO` · **Submitted:** 2026-03-14
+- **Authors:** Jiashun Li, Xiaoyu Shi, Hong Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.13925 · **PDF:** https://arxiv.org/pdf/2603.13925v1
+
+## 5011. EVNextTrade: Learning-to-Rank-Based Recommendation of Next Charging Nodes for EV-EV Energy Trading
+
+- **ID:** `2603.26688` · **primary:** `cs.IR` · **Submitted:** 2026-03-14
+- **Authors:** Md Mahfujur Rahmana, Alistair Barros, Raja Jurdak, et al.
+- **abs:** https://arxiv.org/abs/2603.26688 · **PDF:** https://arxiv.org/pdf/2603.26688v1
+
+## 5012. Generative Inverse Design of Cold Metals for Low-Power Electronics
+
+- **ID:** `2603.13920` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-14
+- **Authors:** Kedeng Wu, Yucheng Zhu, Yan Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.13920 · **PDF:** https://arxiv.org/pdf/2603.13920v1
+
+## 5013. The Phenomenology of Hallucinations
+
+- **ID:** `2603.13911` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Valeria Ruscio, Keiran Thompson
+- **abs:** https://arxiv.org/abs/2603.13911 · **PDF:** https://arxiv.org/pdf/2603.13911v1
+
+## 5014. MRMS-Net and LMRMS-Net: Scalable Multi-Representation Multi-Scale Networks for Time Series Classification
+
+- **ID:** `2603.19315` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Celal Alagöz, Mehmet Kurnaz, Farhan Aadil
+- **abs:** https://arxiv.org/abs/2603.19315 · **PDF:** https://arxiv.org/pdf/2603.19315v2
+
+## 5015. AgriChat: A Multimodal Large Language Model for Agriculture Image Understanding
+
+- **ID:** `2603.16934` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Abderrahmene Boudiaf, Irfan Hussain, Sajid Javed
+- **abs:** https://arxiv.org/abs/2603.16934 · **PDF:** https://arxiv.org/pdf/2603.16934v1
+
+## 5016. FedPBS: Proximal-Balanced Scaling Federated Learning Model for Robust Personalized Training for Non-IID Data
+
+- **ID:** `2603.13909` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Eman M. AbouNassar, Amr Elshall, Sameh Abdulah
+- **abs:** https://arxiv.org/abs/2603.13909 · **PDF:** https://arxiv.org/pdf/2603.13909v2
+
+## 5017. Pixel-level Scene Understanding in One Token: Visual States Need What-is-Where Composition
+
+- **ID:** `2603.13904` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Seokmin Lee, Yunghee Lee, Byeonghyun Pak, et al.
+- **abs:** https://arxiv.org/abs/2603.13904 · **PDF:** https://arxiv.org/pdf/2603.13904v2
+
+## 5018. Distributed Acoustic Sensing for Urban Traffic Monitoring: Spatio-Temporal Attention in Recurrent Neural Networks
+
+- **ID:** `2603.13903` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Izhan Fakhruzi, Manuel Titos, Carmen Benítez, et al.
+- **abs:** https://arxiv.org/abs/2603.13903 · **PDF:** https://arxiv.org/pdf/2603.13903v1
+
+## 5019. MO-SAE:Multi-Objective Stacked Autoencoders Optimization for Edge Anomaly Detection
+
+- **ID:** `2603.13895` · **primary:** `cs.NE` · **Submitted:** 2026-03-14
+- **Authors:** Lizhao Zhang, Shengsong Kong, Tao Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.13895 · **PDF:** https://arxiv.org/pdf/2603.13895v1
+
+## 5020. Robust Self-Training with Closed-loop Label Correction for Learning from Noisy Labels
+
+- **ID:** `2603.13894` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Zhanhui Lin, Yanlin Liu, Sanping Zhou
+- **abs:** https://arxiv.org/abs/2603.13894 · **PDF:** https://arxiv.org/pdf/2603.13894v1
+
+## 5021. UVLM: A Universal Vision-Language Model Loader for Reproducible Multimodal Benchmarking
+
+- **ID:** `2603.13893` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Joan Perez, Giovanni Fusco
+- **abs:** https://arxiv.org/abs/2603.13893 · **PDF:** https://arxiv.org/pdf/2603.13893v1
+
+## 5022. Large Language Models Reproduce Racial Stereotypes When Used for Text Annotation
+
+- **ID:** `2603.13891` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Petter Törnberg
+- **abs:** https://arxiv.org/abs/2603.13891 · **PDF:** https://arxiv.org/pdf/2603.13891v1
+
+## 5023. Multi-Modal Character Localization and Extraction for Chinese Text Recognition
+
+- **ID:** `2603.13886` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Qilong Li, Chongsheng Zhang
+- **abs:** https://arxiv.org/abs/2603.13886 · **PDF:** https://arxiv.org/pdf/2603.13886v1
+
+## 5024. Benchmarking the Energy Cost of Assurance in Neuromorphic Edge Robotics
+
+- **ID:** `2603.13880` · **primary:** `cs.NE` · **Submitted:** 2026-03-14
+- **Authors:** Sylvester Kaczmarek
+- **abs:** https://arxiv.org/abs/2603.13880 · **PDF:** https://arxiv.org/pdf/2603.13880v1
+
+## 5025. Step-CoT: Stepwise Visual Chain-of-Thought for Medical Visual Question Answering
+
+- **ID:** `2603.13878` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Lin Fan, Yafei Ou, Zhipeng Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.13878 · **PDF:** https://arxiv.org/pdf/2603.13878v1
+
+## 5026. Scribe Verification in Chinese manuscripts using Siamese, Triplet, and Vision Transformer Neural Networks
+
+- **ID:** `2603.13877` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Dimitrios-Chrysovalantis Liakopoulos, Yanbo Zhang, Chongsheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.13877 · **PDF:** https://arxiv.org/pdf/2603.13877v1
+
+## 5027. Domain-Specialized Tree of Thought through Plug-and-Play Predictors
+
+- **ID:** `2603.20267` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Xuanqi Gao, Haoyu Wang, Jun Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.20267 · **PDF:** https://arxiv.org/pdf/2603.20267v1
+
+## 5028. GradMem: Learning to Write Context into Memory with Test-Time Gradient Descent
+
+- **ID:** `2603.13875` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Yuri Kuratov, Matvey Kairov, Aydar Bulatov, et al.
+- **abs:** https://arxiv.org/abs/2603.13875 · **PDF:** https://arxiv.org/pdf/2603.13875v1
+
+## 5029. Look Where It Matters: High-Resolution Crops Retrieval for Efficient VLMs
+
+- **ID:** `2603.16932` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Nimrod Shabtay, Moshe Kimhi, Artem Spector, et al.
+- **abs:** https://arxiv.org/abs/2603.16932 · **PDF:** https://arxiv.org/pdf/2603.16932v1
+
+## 5030. On Interpolation Formulas Describing Neural Network Generalization
+
+- **ID:** `2603.13872` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Jin Guo, Roy Y. He, Jean-Michel Morel
+- **abs:** https://arxiv.org/abs/2603.13872 · **PDF:** https://arxiv.org/pdf/2603.13872v1
+
+## 5031. TransDex: Pre-training Visuo-Tactile Policy with Point Cloud Reconstruction for Dexterous Manipulation of Transparent Objects
+
+- **ID:** `2603.13869` · **primary:** `cs.RO` · **Submitted:** 2026-03-14
+- **Authors:** Fengguan Li, Yifan Ma, Chen Qian, et al.
+- **abs:** https://arxiv.org/abs/2603.13869 · **PDF:** https://arxiv.org/pdf/2603.13869v1
+
+## 5032. Script-to-Slide Grounding: Grounding Script Sentences to Slide Objects for Automatic Instructional Video Generation
+
+- **ID:** `2603.16931` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Rena Suzuki, Masato Kikuchi, Tadachika Ozono
+- **abs:** https://arxiv.org/abs/2603.16931 · **PDF:** https://arxiv.org/pdf/2603.16931v1
+
+## 5033. OrigamiBench: An Interactive Environment to Synthesize Flat-Foldable Origamis
+
+- **ID:** `2603.13856` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Naaisha Agarwal, Yihan Wu, Yichang Jian, et al.
+- **abs:** https://arxiv.org/abs/2603.13856 · **PDF:** https://arxiv.org/pdf/2603.13856v2
+
+## 5034. Power Term Polynomial Algebra for Boolean Logic
+
+- **ID:** `2603.13854` · **primary:** `cs.LO` · **Submitted:** 2026-03-14
+- **Authors:** Emanuele Sansone, Armando Solar-Lezama
+- **abs:** https://arxiv.org/abs/2603.13854 · **PDF:** https://arxiv.org/pdf/2603.13854v1
+
+## 5035. APEX-Searcher: Augmenting LLMs' Search Capabilities through Agentic Planning and Execution
+
+- **ID:** `2603.13853` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Kun Chen, Qingchao Kong, Zhao Feifei, et al.
+- **abs:** https://arxiv.org/abs/2603.13853 · **PDF:** https://arxiv.org/pdf/2603.13853v2
+
+## 5036. Fronto-parietal and fronto-temporal EEG coherence as predictive neuromarkers of transcutaneous auricular vagus nerve stimulation response in treatment-resistant schizophrenia: A machine learning study
+
+- **ID:** `2603.13850` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Yapeng Cui, Ruoxi Yun, Shumin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.13850 · **PDF:** https://arxiv.org/pdf/2603.13850v1
+
+## 5037. Exploring the Dimensions of a Variational Neuron
+
+- **ID:** `2603.13849` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Yves Ruffenach
+- **abs:** https://arxiv.org/abs/2603.13849 · **PDF:** https://arxiv.org/pdf/2603.13849v1
+
+## 5038. Sirens' Whisper: Inaudible Near-Ultrasonic Jailbreaks of Speech-Driven LLMs
+
+- **ID:** `2603.13847` · **primary:** `cs.CR` · **Submitted:** 2026-03-14
+- **Authors:** Zijian Ling, Pingyi Hu, Xiuyong Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.13847 · **PDF:** https://arxiv.org/pdf/2603.13847v1
+
+## 5039. Is Seeing Believing? Evaluating Human Sensitivity to Synthetic Video
+
+- **ID:** `2603.13846` · **primary:** `cs.HC` · **Submitted:** 2026-03-14
+- **Authors:** David Wegmann, Emil Stevnsborg, Søren Knudsen, et al.
+- **abs:** https://arxiv.org/abs/2603.13846 · **PDF:** https://arxiv.org/pdf/2603.13846v3
+
+## 5040. Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving
+
+- **ID:** `2603.13842` · **primary:** `cs.RO` · **Submitted:** 2026-03-14
+- **Authors:** Zhexi Lian, Haoran Wang, Xuerun Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.13842 · **PDF:** https://arxiv.org/pdf/2603.13842v1
+
+## 5041. MICRO: A Lightweight Middleware for Optimizing Cross-store Cross-model Graph-Relation Joins [Technical Report]
+
+- **ID:** `2603.13835` · **primary:** `cs.DB` · **Submitted:** 2026-03-14
+- **Authors:** Xiuwen Zheng, Arun Kumar, Amarnath Gupta
+- **abs:** https://arxiv.org/abs/2603.13835 · **PDF:** https://arxiv.org/pdf/2603.13835v1
+
+## 5042. IdentityGuard: Context-Aware Restriction and Provenance for Personalized Synthesis
+
+- **ID:** `2603.15679` · **primary:** `cs.CR` · **Submitted:** 2026-03-14
+- **Authors:** Lingyun Zhang, Yu Xie, Ping Chen
+- **abs:** https://arxiv.org/abs/2603.15679 · **PDF:** https://arxiv.org/pdf/2603.15679v1
+
+## 5043. Intelligent Materials Modelling: Large Language Models Versus Partial Least Squares Regression for Predicting Polysulfone Membrane Mechanical Performance
+
+- **ID:** `2603.13834` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Dingding Cao, Mieow Kee Chan, Wan Sieng Yeo, et al.
+- **abs:** https://arxiv.org/abs/2603.13834 · **PDF:** https://arxiv.org/pdf/2603.13834v1
+
+## 5044. DPxFin: Adaptive Differential Privacy for Anti-Money Laundering Detection via Reputation-Weighted Federated Learning
+
+- **ID:** `2603.19314` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Renuga Kanagavelu, Manjil Nepal, Ning Peiyan, et al.
+- **abs:** https://arxiv.org/abs/2603.19314 · **PDF:** https://arxiv.org/pdf/2603.19314v1
+
+## 5045. Efficient Semi-Automated Material Microstructure Analysis Using Deep Learning: A Case Study in Additive Manufacturing
+
+- **ID:** `2603.13831` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Sanjeev S. Navaratna, Nikhil Thawari, Gunashekhar Mari, et al.
+- **abs:** https://arxiv.org/abs/2603.13831 · **PDF:** https://arxiv.org/pdf/2603.13831v1
+
+## 5046. Early Rug Pull Warning for BSC Meme Tokens via Multi-Granularity Wash-Trading Pattern Profiling
+
+- **ID:** `2603.13830` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Dingding Cao, Bianbian Jiao, Jingzong Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.13830 · **PDF:** https://arxiv.org/pdf/2603.13830v1
+
+## 5047. ArrayTac: A tactile display for simultaneous rendering of shape, stiffness and friction
+
+- **ID:** `2603.13829` · **primary:** `cs.RO` · **Submitted:** 2026-03-14
+- **Authors:** Tianhai Liang, Shiyi Guo, Baiye Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.13829 · **PDF:** https://arxiv.org/pdf/2603.13829v1
+
+## 5048. Effective Sparsity: A Unified Framework via Normalized Entropy and the Effective Number of Nonzeros
+
+- **ID:** `2603.13826` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Haoyu He, Hao Wang, Jiashan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.13826 · **PDF:** https://arxiv.org/pdf/2603.13826v1
+
+## 5049. Evaluating Semantic Fragility in Text-to-Audio Generation Systems Under Controlled Prompt Perturbations
+
+- **ID:** `2603.13824` · **primary:** `cs.SD` · **Submitted:** 2026-03-14
+- **Authors:** Jiahui Wu
+- **abs:** https://arxiv.org/abs/2603.13824 · **PDF:** https://arxiv.org/pdf/2603.13824v1
+
+## 5050. PA-Net: Precipitation-Adaptive Mixture-of-Experts for Long-Tail Rainfall Nowcasting
+
+- **ID:** `2603.13818` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Xinyu Xiao, Sen Lei, Eryun Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.13818 · **PDF:** https://arxiv.org/pdf/2603.13818v1
+
+## 5051. Artificial intelligence-driven improvement of hospital logistics management resilience: a practical exploration based on H Hospital
+
+- **ID:** `2603.13816` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Lu Huang, Dongjing Shan, Han Chen
+- **abs:** https://arxiv.org/abs/2603.13816 · **PDF:** https://arxiv.org/pdf/2603.13816v1
+
+## 5052. JointFM-0.1: A Foundation Model for Multi-Target Joint Distributional Prediction
+
+- **ID:** `2603.20266` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Stefan Hackmann
+- **abs:** https://arxiv.org/abs/2603.20266 · **PDF:** https://arxiv.org/pdf/2603.20266v1
+
+## 5053. Collapse or Preserve: Data-Dependent Temporal Aggregation for Spiking Neural Network Acceleration
+
+- **ID:** `2603.13810` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Jiahao Qin
+- **abs:** https://arxiv.org/abs/2603.13810 · **PDF:** https://arxiv.org/pdf/2603.13810v1
+
+## 5054. An Interpretable and Stable Framework for Sparse Principal Component Analysis
+
+- **ID:** `2603.13806` · **primary:** `stat.ML` · **Submitted:** 2026-03-14
+- **Authors:** Ying Hu, Hu Yang
+- **abs:** https://arxiv.org/abs/2603.13806 · **PDF:** https://arxiv.org/pdf/2603.13806v1
+
+## 5055. Prototypical Exemplar Condensation for Memory-efficient Online Continual Learning
+
+- **ID:** `2603.13804` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Minh-Duong Nguyen, Thien-Thanh Dao, Le-Tuan Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.13804 · **PDF:** https://arxiv.org/pdf/2603.13804v1
+
+## 5056. Node Role-Guided LLMs for Dynamic Graph Clustering
+
+- **ID:** `2603.13799` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Dongyuan Li, Ying Zhang, Yaozu Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.13799 · **PDF:** https://arxiv.org/pdf/2603.13799v1
+
+## 5057. Memory-Driven Role-Playing: Evaluation and Enhancement of Persona Knowledge Utilization in LLMs
+
+- **ID:** `2603.19313` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Kai Wang, Haoyang You, Yang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.19313 · **PDF:** https://arxiv.org/pdf/2603.19313v1
+
+## 5058. PMIScore: An Unsupervised Approach to Quantify Dialogue Engagement
+
+- **ID:** `2603.13796` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Yongkang Guo, Zhihuan Huang, Yuqing Kong
+- **abs:** https://arxiv.org/abs/2603.13796 · **PDF:** https://arxiv.org/pdf/2603.13796v1
+
+## 5059. Computation and Communication Efficient Federated Unlearning via On-server Gradient Conflict Mitigation and Expression
+
+- **ID:** `2603.13795` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Minh-Duong Nguyen, Senura Hansaja, Le-Tuan Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.13795 · **PDF:** https://arxiv.org/pdf/2603.13795v1
+
+## 5060. GhanaNLP Parallel Corpora: Comprehensive Multilingual Resources for Low-Resource Ghanaian Languages
+
+- **ID:** `2603.13793` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Lawrence Adu Gyamfi, Paul Azunre, Stephen Edward Moore, et al.
+- **abs:** https://arxiv.org/abs/2603.13793 · **PDF:** https://arxiv.org/pdf/2603.13793v2
+
+## 5061. IGU-LoRA: Adaptive Rank Allocation via Integrated Gradients and Uncertainty-Aware Scoring
+
+- **ID:** `2603.13792` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Xuan Cui, Huiyue Li, Run Zeng, et al.
+- **abs:** https://arxiv.org/abs/2603.13792 · **PDF:** https://arxiv.org/pdf/2603.13792v1
+
+## 5062. DeceptGuard :A Constitutional Oversight Framework For Detecting Deception in LLM Agents
+
+- **ID:** `2603.13791` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Snehasis Mukhopadhyay
+- **abs:** https://arxiv.org/abs/2603.13791 · **PDF:** https://arxiv.org/pdf/2603.13791v1
+
+## 5063. Greedy Information Projection for LLM Data Selection
+
+- **ID:** `2603.13790` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Victor Ye Dong, Kuan-Yun Lee, Jiamei Shuai, et al.
+- **abs:** https://arxiv.org/abs/2603.13790 · **PDF:** https://arxiv.org/pdf/2603.13790v1
+
+## 5064. Hierarchy of extreme-event predictability in turbulence revealed by machine learning
+
+- **ID:** `2603.13789` · **primary:** `nlin.CD` · **Submitted:** 2026-03-14
+- **Authors:** Yuxuan Yang, Chenyu Dong, Gianmarco Mengaldo
+- **abs:** https://arxiv.org/abs/2603.13789 · **PDF:** https://arxiv.org/pdf/2603.13789v1
+
+## 5065. Projection-Free Evolution Strategies for Continuous Prompt Search
+
+- **ID:** `2603.13786` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Yu Cai, Canxi Huang, Xiaoyu He
+- **abs:** https://arxiv.org/abs/2603.13786 · **PDF:** https://arxiv.org/pdf/2603.13786v1
+
+## 5066. AD-Copilot: A Vision-Language Assistant for Industrial Anomaly Detection via Visual In-context Comparison
+
+- **ID:** `2603.13779` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Xi Jiang, Yue Guo, Jian Li, et al.
+- **abs:** https://arxiv.org/abs/2603.13779 · **PDF:** https://arxiv.org/pdf/2603.13779v1
+
+## 5067. Generate Then Correct: Single Shot Global Correction for Aspect Sentiment Quad Prediction
+
+- **ID:** `2603.13777` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Shidong He, Haoyu Wang, Wenjie Luo
+- **abs:** https://arxiv.org/abs/2603.13777 · **PDF:** https://arxiv.org/pdf/2603.13777v1
+
+## 5068. Retrieval-Feedback-Driven Distillation and Preference Alignment for Efficient LLM-based Query Expansion
+
+- **ID:** `2603.13776` · **primary:** `cs.IR` · **Submitted:** 2026-03-14
+- **Authors:** Minghan Li, Guodong Zhou
+- **abs:** https://arxiv.org/abs/2603.13776 · **PDF:** https://arxiv.org/pdf/2603.13776v1
+
+## 5069. LiveWeb-IE: A Benchmark For Online Web Information Extraction
+
+- **ID:** `2603.13773` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Seungbin Yang, Jihwan Kim, Jaemin Choi, et al.
+- **abs:** https://arxiv.org/abs/2603.13773 · **PDF:** https://arxiv.org/pdf/2603.13773v1
+
+## 5070. Brain Tumor Classification from 3D MRI Using Persistent Homology and Betti Features: A Topological Data Analysis Approach on BraTS2020
+
+- **ID:** `2603.13771` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Faisal Ahmed
+- **abs:** https://arxiv.org/abs/2603.13771 · **PDF:** https://arxiv.org/pdf/2603.13771v1
+
+## 5071. Causal Tracing of Audio-Text Fusion in Large Audio Language Models
+
+- **ID:** `2603.13768` · **primary:** `cs.SD` · **Submitted:** 2026-03-14
+- **Authors:** Wei-Chih Chen, Chien-yu Huang, Hung-yi Lee
+- **abs:** https://arxiv.org/abs/2603.13768 · **PDF:** https://arxiv.org/pdf/2603.13768v1
+
+## 5072. Knowledge Distillation for Large Language Models
+
+- **ID:** `2603.13765` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Alejandro Paredes La Torre, Barbara Flores, Diego Rodriguez
+- **abs:** https://arxiv.org/abs/2603.13765 · **PDF:** https://arxiv.org/pdf/2603.13765v1
+
+## 5073. Level Up: Defining and Exploiting Transitional Problems for Curriculum Learning
+
+- **ID:** `2603.13761` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Zhenwei Tang, Amogh Inamdar, Ashton Anderson, et al.
+- **abs:** https://arxiv.org/abs/2603.13761 · **PDF:** https://arxiv.org/pdf/2603.13761v1
+
+## 5074. Multimodal Emotion Regression with Multi-Objective Optimization and VAD-Aware Audio Modeling for the 10th ABAW EMI Track
+
+- **ID:** `2603.13760` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Jiawen Huang, Chenxi Huang, Zhuofan Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.13760 · **PDF:** https://arxiv.org/pdf/2603.13760v1
+
+## 5075. Fusion Learning from Dynamic Functional Connectivity: Combining the Amplitude and Phase of fMRI Signals to Identify Brain Disorders
+
+- **ID:** `2603.24603` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-14
+- **Authors:** Jinlong Hu, Jiatong Huang, Zijian Cai
+- **abs:** https://arxiv.org/abs/2603.24603 · **PDF:** https://arxiv.org/pdf/2603.24603v1
+
+## 5076. MeTok: An Efficient Meteorological Tokenization with Hyper-Aligned Group Learning for Precipitation Nowcasting
+
+- **ID:** `2603.13752` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Qizhao Jin, Xianhuang Xu, Yong Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.13752 · **PDF:** https://arxiv.org/pdf/2603.13752v1
+
+## 5077. Manifold-Orthogonal Dual-spectrum Extrapolation for Parameterized Physics-Informed Neural Networks
+
+- **ID:** `2603.13751` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Zhangyong Liang, Ji Zhang
+- **abs:** https://arxiv.org/abs/2603.13751 · **PDF:** https://arxiv.org/pdf/2603.13751v1
+
+## 5078. Spectral Edge Dynamics of Training Trajectories: Signal--Noise Geometry Across Scales
+
+- **ID:** `2603.15678` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Yongzhong Xu
+- **abs:** https://arxiv.org/abs/2603.15678 · **PDF:** https://arxiv.org/pdf/2603.15678v2
+
+## 5079. Sub-Band Spectral Matching with Localized Score Aggregation for Robust Anomalous Sound Detection
+
+- **ID:** `2603.13749` · **primary:** `cs.SD` · **Submitted:** 2026-03-14
+- **Authors:** Phurich Saengthong, Takahiro Shinozaki
+- **abs:** https://arxiv.org/abs/2603.13749 · **PDF:** https://arxiv.org/pdf/2603.13749v1
+
+## 5080. Research Paradigm of Materials Science Tetrahedra with Artificial Intelligence
+
+- **ID:** `2603.13744` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-14
+- **Authors:** Shiyun Zhang, Yibo Yao, Haoquan Long, et al.
+- **abs:** https://arxiv.org/abs/2603.13744 · **PDF:** https://arxiv.org/pdf/2603.13744v1
+
+## 5081. Few Batches or Little Memory, But Not Both: Simultaneous Space and Adaptivity Constraints in Stochastic Bandits
+
+- **ID:** `2603.13742` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Ruiyuan Huang, Zicheng Lyu, Xiaoyi Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.13742 · **PDF:** https://arxiv.org/pdf/2603.13742v2
+
+## 5082. UniVid: Pyramid Diffusion Model for High Quality Video Generation
+
+- **ID:** `2603.13739` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Xinyu Xiao, Binbin Yang, Tingtian Li, et al.
+- **abs:** https://arxiv.org/abs/2603.13739 · **PDF:** https://arxiv.org/pdf/2603.13739v1
+
+## 5083. Implicit Maximum Likelihood Estimation for Real-time Generative Model Predictive Control
+
+- **ID:** `2603.13733` · **primary:** `cs.RO` · **Submitted:** 2026-03-14
+- **Authors:** Grayson Lee, Minh Bui, Shuzi Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.13733 · **PDF:** https://arxiv.org/pdf/2603.13733v2
+
+## 5084. R3-REC: Reasoning-Driven Recommendation via Retrieval-Augmented LLMs over Multi-Granular Interest Signals
+
+- **ID:** `2603.13730` · **primary:** `cs.IR` · **Submitted:** 2026-03-14
+- **Authors:** Yuchen Miao, Mingxuan Cui, Yitong Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.13730 · **PDF:** https://arxiv.org/pdf/2603.13730v1
+
+## 5085. Data-driven Progressive Discovery of Physical Laws
+
+- **ID:** `2603.13727` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Mingkun Xia, Weiwei Zhang
+- **abs:** https://arxiv.org/abs/2603.13727 · **PDF:** https://arxiv.org/pdf/2603.13727v1
+
+## 5086. Can We Trust LLMs on Memristors? Diving into Reasoning Ability under Non-Ideality
+
+- **ID:** `2603.13725` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Taiqiang Wu, Yuxin Cheng, Chenchen Ding, et al.
+- **abs:** https://arxiv.org/abs/2603.13725 · **PDF:** https://arxiv.org/pdf/2603.13725v1
+
+## 5087. Facial beauty prediction fusing transfer learning and broad learning system
+
+- **ID:** `2603.16930` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Junying Gan, Xiaoshan Xie, Yikui Zhai, et al.
+- **abs:** https://arxiv.org/abs/2603.16930 · **PDF:** https://arxiv.org/pdf/2603.16930v1
+
+## 5088. InterventionLens: A Multi-Agent Framework for Detecting ASD Intervention Strategies in Parent-Child Shared Reading
+
+- **ID:** `2603.13710` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Xiao Wang, Lu Dong, Ifeoma Nwogu, et al.
+- **abs:** https://arxiv.org/abs/2603.13710 · **PDF:** https://arxiv.org/pdf/2603.13710v1
+
+## 5089. REFINE-DP: Diffusion Policy Fine-tuning for Humanoid Loco-manipulation via Reinforcement Learning
+
+- **ID:** `2603.13707` · **primary:** `cs.RO` · **Submitted:** 2026-03-14
+- **Authors:** Zhaoyuan Gu, Yipu Chen, Zimeng Chai, et al.
+- **abs:** https://arxiv.org/abs/2603.13707 · **PDF:** https://arxiv.org/pdf/2603.13707v2
+
+## 5090. Routing Channel-Patch Dependencies in Time Series Forecasting with Graph Spectral Decomposition
+
+- **ID:** `2603.13702` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Dongyuan Li, Shun Zheng, Chang Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.13702 · **PDF:** https://arxiv.org/pdf/2603.13702v1
+
+## 5091. SAATT Nav: a Socially Aware Autonomous Transparent Transportation Navigation Framework for Wheelchairs
+
+- **ID:** `2603.13698` · **primary:** `cs.RO` · **Submitted:** 2026-03-14
+- **Authors:** Yutong Zhang, Shaiv Y. Mehra, Bradley S. Duerstock, et al.
+- **abs:** https://arxiv.org/abs/2603.13698 · **PDF:** https://arxiv.org/pdf/2603.13698v2
+
+## 5092. Repetition Without Exclusivity: Scale Sensitivity of Referential Mechanisms in Child-Scale Language Models
+
+- **ID:** `2603.13696` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Jon-Paul Cacioli
+- **abs:** https://arxiv.org/abs/2603.13696 · **PDF:** https://arxiv.org/pdf/2603.13696v1
+
+## 5093. QuarkMedBench: A Real-World Scenario Driven Benchmark for Evaluating Large Language Models
+
+- **ID:** `2603.13691` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Yao Wu, Kangping Yin, Liang Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.13691 · **PDF:** https://arxiv.org/pdf/2603.13691v1
+
+## 5094. Quantum-Enhanced Vision Transformer for Flood Detection using Remote Sensing Imagery
+
+- **ID:** `2603.13689` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Soumyajit Maity, Behzad Ghanbarian
+- **abs:** https://arxiv.org/abs/2603.13689 · **PDF:** https://arxiv.org/pdf/2603.13689v1
+
+## 5095. When Should Humans Step In? Optimal Human Dispatching in AI-Assisted Decisions
+
+- **ID:** `2603.13688` · **primary:** `stat.ML` · **Submitted:** 2026-03-14
+- **Authors:** Lezhi Tan, Naomi Sagan, Lihua Lei, et al.
+- **abs:** https://arxiv.org/abs/2603.13688 · **PDF:** https://arxiv.org/pdf/2603.13688v1
+
+## 5096. $τ$-Voice: Benchmarking Full-Duplex Voice Agents on Real-World Domains
+
+- **ID:** `2603.13686` · **primary:** `cs.SD` · **Submitted:** 2026-03-14
+- **Authors:** Soham Ray, Keshav Dhandhania, Victor Barres, et al.
+- **abs:** https://arxiv.org/abs/2603.13686 · **PDF:** https://arxiv.org/pdf/2603.13686v1
+
+## 5097. Preconditioned Test-Time Adaptation for Out-of-Distribution Debiasing in Narrative Generation
+
+- **ID:** `2603.13683` · **primary:** `cs.CL` · **Submitted:** 2026-03-14
+- **Authors:** Hanwen Shen, Ting Ying, Jiajie Lu, et al.
+- **abs:** https://arxiv.org/abs/2603.13683 · **PDF:** https://arxiv.org/pdf/2603.13683v1
+
+## 5098. Sharpness Aware Surrogate Training for Spiking Neural Networks
+
+- **ID:** `2603.18039` · **primary:** `cs.NE` · **Submitted:** 2026-03-14
+- **Authors:** Maximilian Nicholson
+- **abs:** https://arxiv.org/abs/2603.18039 · **PDF:** https://arxiv.org/pdf/2603.18039v1
+
+## 5099. TheraAgent: Multi-Agent Framework with Self-Evolving Memory and Evidence-Calibrated Reasoning for PET Theranostics
+
+- **ID:** `2603.13676` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Zhihao Chen, Jiahui Wang, Yizhou Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.13676 · **PDF:** https://arxiv.org/pdf/2603.13676v1
+
+## 5100. Locally Linear Continual Learning for Time Series based on VC-Theoretical Generalization Bounds
+
+- **ID:** `2603.13674` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Yan V. G. Ferreira, Igor B. Lima, Pedro H. G. Mapa S., et al.
+- **abs:** https://arxiv.org/abs/2603.13674 · **PDF:** https://arxiv.org/pdf/2603.13674v1
+
+## 5101. LLM-MINE: Large Language Model based Alzheimer's Disease and Related Dementias Phenotypes Mining from Clinical Notes
+
+- **ID:** `2603.13673` · **primary:** `cs.AI` · **Submitted:** 2026-03-14
+- **Authors:** Mingchen Shao, Yuzhang Xie, Carl Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.13673 · **PDF:** https://arxiv.org/pdf/2603.13673v1
+
+## 5102. MHPO: Modulated Hazard-aware Policy Optimization for Stable Reinforcement Learning
+
+- **ID:** `2603.16929` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Hongjun Wang, Wei Liu, Weibo Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.16929 · **PDF:** https://arxiv.org/pdf/2603.16929v2
+
+## 5103. SHAMISA: SHAped Modeling of Implicit Structural Associations for Self-supervised No-Reference Image Quality Assessment
+
+- **ID:** `2603.13669` · **primary:** `cs.CV` · **Submitted:** 2026-03-14
+- **Authors:** Mahdi Naseri, Zhou Wang
+- **abs:** https://arxiv.org/abs/2603.13669 · **PDF:** https://arxiv.org/pdf/2603.13669v2
+
+## 5104. Unsupervised Adaptation from FDG to PSMA PET/CT for 3D Lesion Detection under Label Shift
+
+- **ID:** `2603.13666` · **primary:** `eess.IV` · **Submitted:** 2026-03-14
+- **Authors:** Xiaofeng Liu, Menghua Xia, Yanis Chemli, et al.
+- **abs:** https://arxiv.org/abs/2603.13666 · **PDF:** https://arxiv.org/pdf/2603.13666v1
+
+## 5105. PDE-SSM: A Spectral State Space Approach to Spatial Mixing in Diffusion Transformers
+
+- **ID:** `2603.13663` · **primary:** `cs.LG` · **Submitted:** 2026-03-14
+- **Authors:** Eshed Gal, Moshe Eliasof, Siddharth Rout, et al.
+- **abs:** https://arxiv.org/abs/2603.13663 · **PDF:** https://arxiv.org/pdf/2603.13663v1
+
+## 5106. Noticing the Watcher: LLM Agents Can Infer CoT Monitoring from Blocking Feedback
+
+- **ID:** `2603.16928` · **primary:** `cs.CR` · **Submitted:** 2026-03-14
+- **Authors:** Thomas Jiralerspong, Flemming Kondrup, Yoshua Bengio
+- **abs:** https://arxiv.org/abs/2603.16928 · **PDF:** https://arxiv.org/pdf/2603.16928v1
+
+## 5107. Fast Uncertainty Quantification for Kernel-Based Estimators in Large-Scale Causal Inference
+
+- **ID:** `2603.13662` · **primary:** `stat.ME` · **Submitted:** 2026-03-14
+- **Authors:** Matthew Kosko, Falco J, Bargagli-Stoffi, et al.
+- **abs:** https://arxiv.org/abs/2603.13662 · **PDF:** https://arxiv.org/pdf/2603.13662v1
+
+## 5108. Privacy Preserving Topic-wise Sentiment Analysis of the Iran Israel USA Conflict Using Federated Transformer Models
+
+- **ID:** `2603.13655` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Md Saiful Islam, Tanjim Taharat Aurpa, Sharad Hasan, et al.
+- **abs:** https://arxiv.org/abs/2603.13655 · **PDF:** https://arxiv.org/pdf/2603.13655v1
+
+## 5109. Benchmarking Large Language Models on Reference Extraction and Parsing in the Social Sciences and Humanities
+
+- **ID:** `2603.13651` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Yurui Zhu, Giovanni Colavizza, Matteo Romanello
+- **abs:** https://arxiv.org/abs/2603.13651 · **PDF:** https://arxiv.org/pdf/2603.13651v1
+
+## 5110. PLUME: Building a Network-Native Foundation Model for Wireless Traces via Protocol-Aware Tokenization
+
+- **ID:** `2603.13647` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Swadhin Pradhan, Shazal Irshad, Jerome Henry
+- **abs:** https://arxiv.org/abs/2603.13647 · **PDF:** https://arxiv.org/pdf/2603.13647v1
+
+## 5111. StatePlane: A Cognitive State Plane for Long-Horizon AI Systems Under Bounded Context
+
+- **ID:** `2603.13644` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Sasank Annapureddy, John Mulcahy, Anjaneya Prasad Thamatani
+- **abs:** https://arxiv.org/abs/2603.13644 · **PDF:** https://arxiv.org/pdf/2603.13644v1
+
+## 5112. SemRep: Generative Code Representation Learning with Code Transformations
+
+- **ID:** `2603.13640` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Weichen Li, Jiamin Song, Bogdan Alexandru Stoica, et al.
+- **abs:** https://arxiv.org/abs/2603.13640 · **PDF:** https://arxiv.org/pdf/2603.13640v1
+
+## 5113. Widespread Gender and Pronoun Bias in Moral Judgments Across LLMs
+
+- **ID:** `2603.13636` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Gustavo Lúcius Fernandes, Jeiverson C. V. M. Santos, Pedro O. S. Vaz-de-Melo
+- **abs:** https://arxiv.org/abs/2603.13636 · **PDF:** https://arxiv.org/pdf/2603.13636v1
+
+## 5114. MedArena: Comparing LLMs for Medicine-in-the-Wild Clinician Preferences
+
+- **ID:** `2603.15677` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Eric Wu, Kevin Wu, Jason Hom, et al.
+- **abs:** https://arxiv.org/abs/2603.15677 · **PDF:** https://arxiv.org/pdf/2603.15677v1
+
+## 5115. Locatability-Guided Adaptive Reasoning for Image Geo-Localization with Vision-Language Models
+
+- **ID:** `2603.13628` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Bo Yu, Fengze Yang, Yiming Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.13628 · **PDF:** https://arxiv.org/pdf/2603.13628v1
+
+## 5116. BERTology of Molecular Property Prediction
+
+- **ID:** `2603.13627` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Mohammad Mostafanejad, Paul Saxe, T. Daniel Crawford
+- **abs:** https://arxiv.org/abs/2603.13627 · **PDF:** https://arxiv.org/pdf/2603.13627v1
+
+## 5117. Design and evaluation of an agentic workflow for crisis-related synthetic tweet datasets
+
+- **ID:** `2603.13625` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Roben Delos Reyes, Timothy Douglas, Asanobu Kitamoto
+- **abs:** https://arxiv.org/abs/2603.13625 · **PDF:** https://arxiv.org/pdf/2603.13625v1
+
+## 5118. Privacy-Preserving Federated Fraud Detection in Payment Transactions with NVIDIA FLARE
+
+- **ID:** `2603.13617` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Holger R. Roth, Sarthak Tickoo, Mayank Kumar, et al.
+- **abs:** https://arxiv.org/abs/2603.13617 · **PDF:** https://arxiv.org/pdf/2603.13617v1
+
+## 5119. Robust Sequential Tracking via Bounded Information Geometry and Non-Parametric Field Actions
+
+- **ID:** `2603.13613` · **primary:** `stat.ML` · **Submitted:** 2026-03-13
+- **Authors:** Carlos C. Rodriguez
+- **abs:** https://arxiv.org/abs/2603.13613 · **PDF:** https://arxiv.org/pdf/2603.13613v1
+
+## 5120. LLM Routing as Reasoning: A MaxSAT View
+
+- **ID:** `2603.13612` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Son Nguyen, Xinyuan Liu, Ransalu Senanayake
+- **abs:** https://arxiv.org/abs/2603.13612 · **PDF:** https://arxiv.org/pdf/2603.13612v1
+
+## 5121. NCCL EP: Towards a Unified Expert Parallel Communication API for NCCL
+
+- **ID:** `2603.13606` · **primary:** `cs.DC` · **Submitted:** 2026-03-13
+- **Authors:** Amos Goldman, Nimrod Boker, Maayan Sheraizin, et al.
+- **abs:** https://arxiv.org/abs/2603.13606 · **PDF:** https://arxiv.org/pdf/2603.13606v3
+
+## 5122. Orla: A Library for Serving LLM-Based Multi-Agent Systems
+
+- **ID:** `2603.13605` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Rana Shahout, Hayder Tirmazi, Minlan Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.13605 · **PDF:** https://arxiv.org/pdf/2603.13605v1
+
+## 5123. The Equivalence Theorem: First-Class Relationships for Structurally Complete Database Systems
+
+- **ID:** `2603.13603` · **primary:** `cs.DB` · **Submitted:** 2026-03-13
+- **Authors:** Matthew Alford
+- **abs:** https://arxiv.org/abs/2603.13603 · **PDF:** https://arxiv.org/pdf/2603.13603v1
+
+## 5124. A Causal Framework for Mitigating Data Shifts in Healthcare
+
+- **ID:** `2603.13595` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Kurt Butler, Stephanie Riley, Damian Machlanski, et al.
+- **abs:** https://arxiv.org/abs/2603.13595 · **PDF:** https://arxiv.org/pdf/2603.13595v1
+
+## 5125. EnterpriseOps-Gym: Environments and Evaluations for Stateful Agentic Planning and Tool Use in Enterprise Settings
+
+- **ID:** `2603.13594` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Shiva Krishna Reddy Malay, Shravan Nayak, Jishnu Sethumadhavan Nair, et al.
+- **abs:** https://arxiv.org/abs/2603.13594 · **PDF:** https://arxiv.org/pdf/2603.13594v1
+
+## 5126. Opportunistic Cardiac Health Assessment: Estimating Phenotypes from Localizer MRI through Multi-Modal Representations
+
+- **ID:** `2603.13590` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Busra Nur Zeybek, Özgün Turgut, Yundi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.13590 · **PDF:** https://arxiv.org/pdf/2603.13590v1
+
+## 5127. Volumetric Radar Echo Motion Estimation Using Physics-Informed Deep Learning: A Case Study Over Slovakia
+
+- **ID:** `2603.13589` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Peter Pavlík, Anna Bou Ezzeddine, Viera Rozinajová
+- **abs:** https://arxiv.org/abs/2603.13589 · **PDF:** https://arxiv.org/pdf/2603.13589v1
+
+## 5128. State-space models through the lens of ensemble control
+
+- **ID:** `2603.13587` · **primary:** `math.OC` · **Submitted:** 2026-03-13
+- **Authors:** Ye Feng, Jianfeng Lu
+- **abs:** https://arxiv.org/abs/2603.13587 · **PDF:** https://arxiv.org/pdf/2603.13587v1
+
+## 5129. An Empirical Investigation of Pre-Trained Deep Learning Model Reuse in the Scientific Process
+
+- **ID:** `2603.13584` · **primary:** `cs.SE` · **Submitted:** 2026-03-13
+- **Authors:** Nicholas M. Synovic, Karolina Ryzka, Alessandra V. Vellucci Solari, et al.
+- **abs:** https://arxiv.org/abs/2603.13584 · **PDF:** https://arxiv.org/pdf/2603.13584v1
+
+## 5130. Automated Self-Testing as a Quality Gate: Evidence-Driven Release Management for LLM Applications
+
+- **ID:** `2603.15676` · **primary:** `cs.SE` · **Submitted:** 2026-03-13
+- **Authors:** Alexandre Cristovão Maiorano
+- **abs:** https://arxiv.org/abs/2603.15676 · **PDF:** https://arxiv.org/pdf/2603.15676v1
+
+## 5131. State Algebra for Probabilistic Logic
+
+- **ID:** `2603.13574` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Dmitry Lesnik, Tobias Schäfer
+- **abs:** https://arxiv.org/abs/2603.13574 · **PDF:** https://arxiv.org/pdf/2603.13574v1
+
+## 5132. Privacy-Preserving Machine Learning for IoT: A Cross-Paradigm Survey and Future Roadmap
+
+- **ID:** `2603.13570` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Zakia Zaman, Praveen Gauravaram, Mahbub Hassan, et al.
+- **abs:** https://arxiv.org/abs/2603.13570 · **PDF:** https://arxiv.org/pdf/2603.13570v3
+
+## 5133. EmDT: Embedding Diffusion Transformer for Tabular Data Generation in Fraud Detection
+
+- **ID:** `2603.13566` · **primary:** `stat.ML` · **Submitted:** 2026-03-13
+- **Authors:** En-Ya Kuo, Sebastien Motsch
+- **abs:** https://arxiv.org/abs/2603.13566 · **PDF:** https://arxiv.org/pdf/2603.13566v1
+
+## 5134. MR-GNF: Multi-Resolution Graph Neural Forecasting on Ellipsoidal Meshes for Efficient Regional Weather Prediction
+
+- **ID:** `2603.13563` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Andrii Shchur, Inna Skarga-Bandurova
+- **abs:** https://arxiv.org/abs/2603.13563 · **PDF:** https://arxiv.org/pdf/2603.13563v1
+
+## 5135. Scalable Classification of Course Information Sheets Using Large Language Models: A Reusable Institutional Method for Academic Quality Assurance
+
+- **ID:** `2603.13562` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Brecht Verbeken, Joke Van den Broeck, Inge De Cleyn, et al.
+- **abs:** https://arxiv.org/abs/2603.13562 · **PDF:** https://arxiv.org/pdf/2603.13562v1
+
+## 5136. Task-Oriented Wireless Transmission of 3D Point Clouds: Geometric Versus Semantic Robustness
+
+- **ID:** `2603.13560` · **primary:** `eess.SP` · **Submitted:** 2026-03-13
+- **Authors:** Vukan Ninkovic, Tamara Sobot, Vladimir Vincan, et al.
+- **abs:** https://arxiv.org/abs/2603.13560 · **PDF:** https://arxiv.org/pdf/2603.13560v1
+
+## 5137. Robust Automatic Differentiation of Square-Root Kalman Filters via Gramian Differentials
+
+- **ID:** `2603.13559` · **primary:** `stat.ML` · **Submitted:** 2026-03-13
+- **Authors:** Adrien Corenflos
+- **abs:** https://arxiv.org/abs/2603.13559 · **PDF:** https://arxiv.org/pdf/2603.13559v1
+
+## 5138. Holographic Invariant Storage: Design-Time Safety Contracts via Vector Symbolic Architectures
+
+- **ID:** `2603.13558` · **primary:** `stat.ML` · **Submitted:** 2026-03-13
+- **Authors:** Arsenios Scrivens
+- **abs:** https://arxiv.org/abs/2603.13558 · **PDF:** https://arxiv.org/pdf/2603.13558v1
+
+## 5139. Semantic Aware Feature Extraction for Enhanced 3D Reconstruction
+
+- **ID:** `2603.13556` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Ronald Nap, Andy Xiao
+- **abs:** https://arxiv.org/abs/2603.13556 · **PDF:** https://arxiv.org/pdf/2603.13556v1
+
+## 5140. LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels
+
+- **ID:** `2603.19312` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Lucas Maes, Quentin Le Lidec, Damien Scieur, et al.
+- **abs:** https://arxiv.org/abs/2603.19312 · **PDF:** https://arxiv.org/pdf/2603.19312v2
+
+## 5141. Ghosts of Softmax: Complex Singularities That Limit Safe Step Sizes in Cross-Entropy
+
+- **ID:** `2603.13552` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Piyush Sao
+- **abs:** https://arxiv.org/abs/2603.13552 · **PDF:** https://arxiv.org/pdf/2603.13552v1
+
+## 5142. Probabilistic Gaussian Homotopy: A Probability-Space Continuation Framework for Nonconvex Optimization
+
+- **ID:** `2603.13546` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Eshed Gal, Samy Wu Fung, Eldad Haber
+- **abs:** https://arxiv.org/abs/2603.13546 · **PDF:** https://arxiv.org/pdf/2603.13546v1
+
+## 5143. The AI Fiction Paradox
+
+- **ID:** `2603.13545` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Katherine Elkins
+- **abs:** https://arxiv.org/abs/2603.13545 · **PDF:** https://arxiv.org/pdf/2603.13545v1
+
+## 5144. Exploring label correlations using decision templates for ensemble of classifier chains
+
+- **ID:** `2603.13541` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Victor F. Rocha, Alexandre L. Rodrigues, Thiago Oliveira-Santos, et al.
+- **abs:** https://arxiv.org/abs/2603.13541 · **PDF:** https://arxiv.org/pdf/2603.13541v1
+
+## 5145. AMES: Approximate Multi-modal Enterprise Search via Late Interaction Retrieval
+
+- **ID:** `2603.13537` · **primary:** `cs.IR` · **Submitted:** 2026-03-13
+- **Authors:** Tony Joseph, Carlos Pareja, David Lopes Pegna, et al.
+- **abs:** https://arxiv.org/abs/2603.13537 · **PDF:** https://arxiv.org/pdf/2603.13537v1
+
+## 5146. Active Sampling Sample-based Quantum Diagonalization from Finite-Shot Measurements
+
+- **ID:** `2603.13536` · **primary:** `quant-ph` · **Submitted:** 2026-03-13
+- **Authors:** Rinka Miura
+- **abs:** https://arxiv.org/abs/2603.13536 · **PDF:** https://arxiv.org/pdf/2603.13536v1
+
+## 5147. Hide and Seek: Investigating Redundancy in Earth Observation Imagery
+
+- **ID:** `2603.13524` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Tasos Papazafeiropoulos, Nikolaos Ioannis Bountos, Nikolas Papadopoulos, et al.
+- **abs:** https://arxiv.org/abs/2603.13524 · **PDF:** https://arxiv.org/pdf/2603.13524v1
+
+## 5148. VoXtream2: Full-stream TTS with dynamic speaking rate control
+
+- **ID:** `2603.13518` · **primary:** `eess.AS` · **Submitted:** 2026-03-13
+- **Authors:** Nikita Torgashov, Gustav Eje Henter, Gabriel Skantze
+- **abs:** https://arxiv.org/abs/2603.13518 · **PDF:** https://arxiv.org/pdf/2603.13518v1
+
+## 5149. Executable Archaeology: Reanimating the Logic Theorist from its IPL-V Source
+
+- **ID:** `2603.13514` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Jeff Shrager
+- **abs:** https://arxiv.org/abs/2603.13514 · **PDF:** https://arxiv.org/pdf/2603.13514v1
+
+## 5150. Standard Acquisition Is Sufficient for Asynchronous Bayesian Optimization
+
+- **ID:** `2603.13501` · **primary:** `stat.ML` · **Submitted:** 2026-03-13
+- **Authors:** Ben Riegler, James Odgers, Vincent Fortuin
+- **abs:** https://arxiv.org/abs/2603.13501 · **PDF:** https://arxiv.org/pdf/2603.13501v1
+
+## 5151. Synthetic Melanoma Image Generation and Evaluation Using Generative Adversarial Networks
+
+- **ID:** `2603.13497` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Pei-Yu Lin, Yidan Shen, Neville Mathew, et al.
+- **abs:** https://arxiv.org/abs/2603.13497 · **PDF:** https://arxiv.org/pdf/2603.13497v1
+
+## 5152. Deep Invertible Autoencoders for Dimensionality Reduction of Dynamical Systems
+
+- **ID:** `2603.13496` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Nicolò Botteghi, Silke Glas, Christoph Brune
+- **abs:** https://arxiv.org/abs/2603.13496 · **PDF:** https://arxiv.org/pdf/2603.13496v1
+
+## 5153. PrefPO: Pairwise Preference Prompt Optimization
+
+- **ID:** `2603.19311` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Rahul Singhal, Pradyumna Tambwekar, Karime Maamari
+- **abs:** https://arxiv.org/abs/2603.19311 · **PDF:** https://arxiv.org/pdf/2603.19311v2
+
+## 5154. MemReward: Graph-Based Experience Memory for LLM Reward Prediction with Limited Labels
+
+- **ID:** `2603.19310` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Tianyang Luo, Tao Feng, Zhigang Hua, et al.
+- **abs:** https://arxiv.org/abs/2603.19310 · **PDF:** https://arxiv.org/pdf/2603.19310v2
+
+## 5155. Equivalence of approximation by networks of single- and multi-spike neurons
+
+- **ID:** `2603.13478` · **primary:** `cs.NE` · **Submitted:** 2026-03-13
+- **Authors:** Dominik Dold, Philipp Christian Petersen
+- **abs:** https://arxiv.org/abs/2603.13478 · **PDF:** https://arxiv.org/pdf/2603.13478v1
+
+## 5156. PhysMoDPO: Physically-Plausible Humanoid Motion with Preference Optimization
+
+- **ID:** `2603.13228` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Yangsong Zhang, Anujith Muraleedharan, Rikhat Akizhanov, et al.
+- **abs:** https://arxiv.org/abs/2603.13228 · **PDF:** https://arxiv.org/pdf/2603.13228v2
+
+## 5157. Representation Learning for Spatiotemporal Physical Systems
+
+- **ID:** `2603.13227` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Helen Qu, Rudy Morel, Michael McCabe, et al.
+- **abs:** https://arxiv.org/abs/2603.13227 · **PDF:** https://arxiv.org/pdf/2603.13227v1
+
+## 5158. Resolving Interference (RI): Disentangling Models for Improved Model Merging
+
+- **ID:** `2603.13467` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Pratik Ramesh, George Stoica, Arun Iyer, et al.
+- **abs:** https://arxiv.org/abs/2603.13467 · **PDF:** https://arxiv.org/pdf/2603.13467v1
+
+## 5159. Visual-ERM: Reward Modeling for Visual Equivalence
+
+- **ID:** `2603.13224` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Ziyu Liu, Shengyuan Ding, Xinyu Fang, et al.
+- **abs:** https://arxiv.org/abs/2603.13224 · **PDF:** https://arxiv.org/pdf/2603.13224v1
+
+## 5160. Theoretical Foundations of Latent Posterior Factors: Formal Guarantees for Multi-Evidence Reasoning
+
+- **ID:** `2603.15674` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Aliyu Agboola Alege
+- **abs:** https://arxiv.org/abs/2603.15674 · **PDF:** https://arxiv.org/pdf/2603.15674v2
+
+## 5161. Neuron-Aware Data Selection In Instruction Tuning For Large Language Models
+
+- **ID:** `2603.13201` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Xin Chen, Junchao Wu, Shu Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.13201 · **PDF:** https://arxiv.org/pdf/2603.13201v1
+
+## 5162. From Experiments to Expertise: Scientific Knowledge Consolidation for AI-Driven Computational Research
+
+- **ID:** `2603.13191` · **primary:** `physics.comp-ph` · **Submitted:** 2026-03-13
+- **Authors:** Haonan Huang
+- **abs:** https://arxiv.org/abs/2603.13191 · **PDF:** https://arxiv.org/pdf/2603.13191v1
+
+## 5163. LLM Constitutional Multi-Agent Governance
+
+- **ID:** `2603.13189` · **primary:** `cs.MA` · **Submitted:** 2026-03-13
+- **Authors:** J. de Curtò, I. de Zarzà
+- **abs:** https://arxiv.org/abs/2603.13189 · **PDF:** https://arxiv.org/pdf/2603.13189v1
+
+## 5164. Learnability and Privacy Vulnerability are Entangled in a Few Critical Weights
+
+- **ID:** `2603.13186` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Xingli Fang, Jung-Eun Kim
+- **abs:** https://arxiv.org/abs/2603.13186 · **PDF:** https://arxiv.org/pdf/2603.13186v1
+
+## 5165. MXNorm: Reusing MXFP block scales for efficient tensor normalisation
+
+- **ID:** `2603.13180` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Callum McLean, Luke Y. Prince, Alexandre Payot, et al.
+- **abs:** https://arxiv.org/abs/2603.13180 · **PDF:** https://arxiv.org/pdf/2603.13180v1
+
+## 5166. Clustering Astronomical Orbital Synthetic Data Using Advanced Feature Extraction and Dimensionality Reduction Techniques
+
+- **ID:** `2603.13177` · **primary:** `astro-ph.EP` · **Submitted:** 2026-03-13
+- **Authors:** Eraldo Pereira Marinho, Nelson Callegari Junior, Fabricio Aparecido Breve, et al.
+- **abs:** https://arxiv.org/abs/2603.13177 · **PDF:** https://arxiv.org/pdf/2603.13177v1
+
+## 5167. Purifying Generative LLMs from Backdoors without Prior Knowledge or Clean Reference
+
+- **ID:** `2603.13461` · **primary:** `cs.CR` · **Submitted:** 2026-03-13
+- **Authors:** Jianwei Li, Jung-Eun Kim
+- **abs:** https://arxiv.org/abs/2603.13461 · **PDF:** https://arxiv.org/pdf/2603.13461v1
+
+## 5168. Semantic Invariance in Agentic AI
+
+- **ID:** `2603.13173` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** I. de Zarzà, J. de Curtò, Jordi Cabot, et al.
+- **abs:** https://arxiv.org/abs/2603.13173 · **PDF:** https://arxiv.org/pdf/2603.13173v2
+
+## 5169. Developing and evaluating a chatbot to support maternal health care
+
+- **ID:** `2603.13168` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Smriti Jha, Vidhi Jain, Jianyu Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.13168 · **PDF:** https://arxiv.org/pdf/2603.13168v1
+
+## 5170. Reconciling In-Context and In-Weight Learning via Dual Representation Space Encoding
+
+- **ID:** `2603.13459` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Guanyu Chen, Ruichen Wang, Tianren Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.13459 · **PDF:** https://arxiv.org/pdf/2603.13459v1
+
+## 5171. Towards Faithful Multimodal Concept Bottleneck Models
+
+- **ID:** `2603.13163` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Pierre Moreau, Emeline Pineau Ferrand, Yann Choho, et al.
+- **abs:** https://arxiv.org/abs/2603.13163 · **PDF:** https://arxiv.org/pdf/2603.13163v1
+
+## 5172. When Your Model Stops Working: Anytime-Valid Calibration Monitoring
+
+- **ID:** `2603.13156` · **primary:** `stat.ME` · **Submitted:** 2026-03-13
+- **Authors:** Tristan Farran
+- **abs:** https://arxiv.org/abs/2603.13156 · **PDF:** https://arxiv.org/pdf/2603.13156v1
+
+## 5173. ESG-Bench: Benchmarking Long-Context ESG Reports for Hallucination Mitigation
+
+- **ID:** `2603.13154` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Siqi Sun, Ben Peng Wu, Mali Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.13154 · **PDF:** https://arxiv.org/pdf/2603.13154v1
+
+## 5174. When Right Meets Wrong: Bilateral Context Conditioning with Reward-Confidence Correction for GRPO
+
+- **ID:** `2603.13134` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Yu Li, Tian Lan, Zhengling Qi
+- **abs:** https://arxiv.org/abs/2603.13134 · **PDF:** https://arxiv.org/pdf/2603.13134v1
+
+## 5175. Steve-Evolving: Open-World Embodied Self-Evolution via Fine-Grained Diagnosis and Dual-Track Knowledge Distillation
+
+- **ID:** `2603.13131` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Zhengwei Xie, Zhisheng Chen, Ziyan Weng, et al.
+- **abs:** https://arxiv.org/abs/2603.13131 · **PDF:** https://arxiv.org/pdf/2603.13131v1
+
+## 5176. Developing the PsyCogMetrics AI Lab to Evaluate Large Language Models and Advance Cognitive Science -- A Three-Cycle Action Design Science Study
+
+- **ID:** `2603.13126` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-13
+- **Authors:** Zhiye Jin, Yibai Li, K. D. Joshi, et al.
+- **abs:** https://arxiv.org/abs/2603.13126 · **PDF:** https://arxiv.org/pdf/2603.13126v1
+
+## 5177. Scalable Machines with Intrinsic Higher Mental-State Dynamics
+
+- **ID:** `2603.13453` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Ahsan Adeel, M. Bilal
+- **abs:** https://arxiv.org/abs/2603.13453 · **PDF:** https://arxiv.org/pdf/2603.13453v1
+
+## 5178. Geometry-Guided Camera Motion Understanding in VideoLLMs
+
+- **ID:** `2603.13119` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Haoan Feng, Sri Harsha Musunuri, Guan-Ming Su
+- **abs:** https://arxiv.org/abs/2603.13119 · **PDF:** https://arxiv.org/pdf/2603.13119v2
+
+## 5179. ZO-SAM: Zero-Order Sharpness-Aware Minimization for Efficient Sparse Training
+
+- **ID:** `2603.13115` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Jie Ji, Gen Li, Kaiyuan Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.13115 · **PDF:** https://arxiv.org/pdf/2603.13115v1
+
+## 5180. BoSS: A Best-of-Strategies Selector as an Oracle for Deep Active Learning
+
+- **ID:** `2603.13109` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Denis Huseljic, Paul Hahn, Marek Herde, et al.
+- **abs:** https://arxiv.org/abs/2603.13109 · **PDF:** https://arxiv.org/pdf/2603.13109v1
+
+## 5181. Evaluating VLMs' Spatial Reasoning Over Robot Motion: A Step Towards Robot Planning with Motion Preferences
+
+- **ID:** `2603.13100` · **primary:** `cs.RO` · **Submitted:** 2026-03-13
+- **Authors:** Wenxi Wu, Jingjing Zhang, Martim Brandão
+- **abs:** https://arxiv.org/abs/2603.13100 · **PDF:** https://arxiv.org/pdf/2603.13100v1
+
+## 5182. Beyond Final Answers: CRYSTAL Benchmark for Transparent Multimodal Reasoning Evaluation
+
+- **ID:** `2603.13099` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Wayner Barrios, SouYoung Jin
+- **abs:** https://arxiv.org/abs/2603.13099 · **PDF:** https://arxiv.org/pdf/2603.13099v2
+
+## 5183. MESD: Detecting and Mitigating Procedural Bias in Intersectional Groups
+
+- **ID:** `2603.13452` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Gideon Popoola, John Sheppard
+- **abs:** https://arxiv.org/abs/2603.13452 · **PDF:** https://arxiv.org/pdf/2603.13452v1
+
+## 5184. Breaking the Tuning Barrier: Zero-Hyperparameters Yield Multi-Corner Analysis Via Learned Priors
+
+- **ID:** `2603.13092` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Wei W. Xing, Kaiqi Huang, Jiazhan Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.13092 · **PDF:** https://arxiv.org/pdf/2603.13092v1
+
+## 5185. DRCY: Agentic Hardware Design Reviews
+
+- **ID:** `2603.15672` · **primary:** `cs.AR` · **Submitted:** 2026-03-13
+- **Authors:** Kyle Dumont, Nicholas Herbert, Hayder Tirmazi, et al.
+- **abs:** https://arxiv.org/abs/2603.15672 · **PDF:** https://arxiv.org/pdf/2603.15672v1
+
+## 5186. Influence Malleability in Linearized Attention: Dual Implications of Non-Convergent NTK Dynamics
+
+- **ID:** `2603.13085` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Jose Marie Antonio Miñoza, Paulo Mario P. Medina, Sebastian C. Ibañez
+- **abs:** https://arxiv.org/abs/2603.13085 · **PDF:** https://arxiv.org/pdf/2603.13085v1
+
+## 5187. Human-in-the-Loop LLM Grading for Handwritten Mathematics Assessments
+
+- **ID:** `2603.13083` · **primary:** `cs.CY` · **Submitted:** 2026-03-13
+- **Authors:** Arne Vanhoyweghen, Vincent Holst, Melika Mobini, et al.
+- **abs:** https://arxiv.org/abs/2603.13083 · **PDF:** https://arxiv.org/pdf/2603.13083v1
+
+## 5188. Fractals made Practical: Denoising Diffusion as Partitioned Iterated Function Systems
+
+- **ID:** `2603.13069` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Ann Dooms
+- **abs:** https://arxiv.org/abs/2603.13069 · **PDF:** https://arxiv.org/pdf/2603.13069v1
+
+## 5189. GeoChemAD: Benchmarking Unsupervised Geochemical Anomaly Detection for Mineral Exploration
+
+- **ID:** `2603.13068` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Yihao Ding, Yiran Zhang, Chris Gonzalez, et al.
+- **abs:** https://arxiv.org/abs/2603.13068 · **PDF:** https://arxiv.org/pdf/2603.13068v1
+
+## 5190. L2GTX: From Local to Global Time Series Explanations
+
+- **ID:** `2603.13065` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Ephrem Tibebe Mekonnen, Luca Longo, Lucas Rizzo, et al.
+- **abs:** https://arxiv.org/abs/2603.13065 · **PDF:** https://arxiv.org/pdf/2603.13065v1
+
+## 5191. LADR: Locality-Aware Dynamic Rescue for Efficient Text-to-Image Generation with Diffusion Large Language Models
+
+- **ID:** `2603.13450` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Chenglin Wang, Yucheng Zhou, Shawn Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.13450 · **PDF:** https://arxiv.org/pdf/2603.13450v1
+
+## 5192. Competition-Aware CPC Forecasting with Near-Market Coverage
+
+- **ID:** `2603.13059` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Sebastian Frey, Edoardo Beccari, Maximilian Kranz, et al.
+- **abs:** https://arxiv.org/abs/2603.13059 · **PDF:** https://arxiv.org/pdf/2603.13059v1
+
+## 5193. Team RAS in 10th ABAW Competition: Multimodal Valence and Arousal Estimation Approach
+
+- **ID:** `2603.13056` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Elena Ryumina, Maxim Markitantov, Alexandr Axyonov, et al.
+- **abs:** https://arxiv.org/abs/2603.13056 · **PDF:** https://arxiv.org/pdf/2603.13056v1
+
+## 5194. Causal Cellular Context Transfer Learning (C3TL): An Efficient Architecture for Prediction of Unseen Perturbation Effects
+
+- **ID:** `2603.13051` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Michael Scholkemper, Sach Mukherjee
+- **abs:** https://arxiv.org/abs/2603.13051 · **PDF:** https://arxiv.org/pdf/2603.13051v1
+
+## 5195. 3DTCR: A Physics-Based Generative Framework for Vortex-Following 3D Reconstruction to Improve Tropical Cyclone Intensity Forecasting
+
+- **ID:** `2603.13049` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Jun Liu, Xiaohui Zhong, Kai Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.13049 · **PDF:** https://arxiv.org/pdf/2603.13049v2
+
+## 5196. Learning Energy-Efficient Air--Ground Actuation for Hybrid Robots on Stair-Like Terrain
+
+- **ID:** `2603.26687` · **primary:** `cs.RO` · **Submitted:** 2026-03-13
+- **Authors:** Jiaxing Li, Wen Tian, Xinhang Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.26687 · **PDF:** https://arxiv.org/pdf/2603.26687v1
+
+## 5197. Convergence Rate of a Functional Learning Method for Contextual Stochastic Optimization
+
+- **ID:** `2603.13048` · **primary:** `math.OC` · **Submitted:** 2026-03-13
+- **Authors:** Noel Smith, Andrzej Ruszczynski
+- **abs:** https://arxiv.org/abs/2603.13048 · **PDF:** https://arxiv.org/pdf/2603.13048v1
+
+## 5198. Are General-Purpose Vision Models All We Need for 2D Medical Image Segmentation? A Cross-Dataset Empirical Study
+
+- **ID:** `2603.13044` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Vanessa Borst, Samuel Kounev
+- **abs:** https://arxiv.org/abs/2603.13044 · **PDF:** https://arxiv.org/pdf/2603.13044v1
+
+## 5199. Mending the Holes: Mitigating Reward Hacking in Reinforcement Learning for Multilingual Translation
+
+- **ID:** `2603.13045` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Yifeng Liu, Siqi Ouyang, Yatish Hosmane Revanasiddappa, et al.
+- **abs:** https://arxiv.org/abs/2603.13045 · **PDF:** https://arxiv.org/pdf/2603.13045v1
+
+## 5200. OpenACMv2: An Accuracy-Constrained Co-Optimization Framework for Approximate DCiM
+
+- **ID:** `2603.13042` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Yiqi Zhou, Yue Yuan, Yikai Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.13042 · **PDF:** https://arxiv.org/pdf/2603.13042v1
+
+## 5201. Interpretable Semantic Gradients in SSD: A PCA Sweep Approach and a Case Study on AI Discourse
+
+- **ID:** `2603.13038` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Hubert Plisiecki, Maria Leniarska, Jan Piotrowski, et al.
+- **abs:** https://arxiv.org/abs/2603.13038 · **PDF:** https://arxiv.org/pdf/2603.13038v1
+
+## 5202. Federated Few-Shot Learning on Neuromorphic Hardware: An Empirical Study Across Physical Edge Nodes
+
+- **ID:** `2603.13037` · **primary:** `cs.NE` · **Submitted:** 2026-03-13
+- **Authors:** Steven Motta, Gioele Nanni
+- **abs:** https://arxiv.org/abs/2603.13037 · **PDF:** https://arxiv.org/pdf/2603.13037v1
+
+## 5203. Interrogating Design Homogenization in Web Vibe Coding
+
+- **ID:** `2603.13036` · **primary:** `cs.HC` · **Submitted:** 2026-03-13
+- **Authors:** Donghoon Shin, Alice Gao, Rock Yuren Pang, et al.
+- **abs:** https://arxiv.org/abs/2603.13036 · **PDF:** https://arxiv.org/pdf/2603.13036v1
+
+## 5204. Association-Aware GNN for Precoder Learning in Cell-Free Systems
+
+- **ID:** `2603.13035` · **primary:** `eess.SP` · **Submitted:** 2026-03-13
+- **Authors:** Mingyu Deng, Shengqian Han
+- **abs:** https://arxiv.org/abs/2603.13035 · **PDF:** https://arxiv.org/pdf/2603.13035v1
+
+## 5205. ESPIRE: A Diagnostic Benchmark for Embodied Spatial Reasoning of Vision-Language Models
+
+- **ID:** `2603.13033` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Yanpeng Zhao, Wentao Ding, Hongtao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.13033 · **PDF:** https://arxiv.org/pdf/2603.13033v1
+
+## 5206. GT-Space: Enhancing Heterogeneous Collaborative Perception with Ground Truth Feature Space
+
+- **ID:** `2603.19308` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Wentao Wang, Haoran Xu, Guang Tan
+- **abs:** https://arxiv.org/abs/2603.19308 · **PDF:** https://arxiv.org/pdf/2603.19308v1
+
+## 5207. Purify Once, Edit Freely: Breaking Image Protections under Model Mismatch
+
+- **ID:** `2603.13028` · **primary:** `cs.CR` · **Submitted:** 2026-03-13
+- **Authors:** Qichen Zhao, Shengfang Zhai, Xinjian Bai, et al.
+- **abs:** https://arxiv.org/abs/2603.13028 · **PDF:** https://arxiv.org/pdf/2603.13028v1
+
+## 5208. MuViS: Multimodal Virtual Sensing Benchmark
+
+- **ID:** `2603.24602` · **primary:** `eess.SP` · **Submitted:** 2026-03-13
+- **Authors:** Jens U. Brandt, Noah C. Puetz, Jobel Jose George, et al.
+- **abs:** https://arxiv.org/abs/2603.24602 · **PDF:** https://arxiv.org/pdf/2603.24602v1
+
+## 5209. SortScrews: A Dataset and Baseline for Real-time Screw Classification
+
+- **ID:** `2603.13027` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Tianhao Fu, Bingxuan Yang, Juncheng Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.13027 · **PDF:** https://arxiv.org/pdf/2603.13027v1
+
+## 5210. PISmith: Reinforcement Learning-based Red Teaming for Prompt Injection Defenses
+
+- **ID:** `2603.13026` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Chenlong Yin, Runpeng Geng, Yanting Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.13026 · **PDF:** https://arxiv.org/pdf/2603.13026v1
+
+## 5211. SAW: Toward a Surgical Action World Model via Controllable and Scalable Video Generation
+
+- **ID:** `2603.13024` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Sampath Rapuri, Lalithkumar Seenivasan, Dominik Schneider, et al.
+- **abs:** https://arxiv.org/abs/2603.13024 · **PDF:** https://arxiv.org/pdf/2603.13024v1
+
+## 5212. daVinci-Env: Open SWE Environment Synthesis at Scale
+
+- **ID:** `2603.13023` · **primary:** `cs.SE` · **Submitted:** 2026-03-13
+- **Authors:** Dayuan Fu, Shenyu Wu, Yunze Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.13023 · **PDF:** https://arxiv.org/pdf/2603.13023v2
+
+## 5213. ARL-Tangram: Unleash the Resource Efficiency in Agentic Reinforcement Learning
+
+- **ID:** `2603.13019` · **primary:** `cs.DC` · **Submitted:** 2026-03-13
+- **Authors:** Bangjun Xiao, Yihao Zhao, Xiangwei Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.13019 · **PDF:** https://arxiv.org/pdf/2603.13019v1
+
+## 5214. Large Language Models and Scientific Discourse: Where's the Intelligence?
+
+- **ID:** `2603.23543` · **primary:** `physics.soc-ph` · **Submitted:** 2026-03-13
+- **Authors:** Harry Collins, Simon Thorne
+- **abs:** https://arxiv.org/abs/2603.23543 · **PDF:** https://arxiv.org/pdf/2603.23543v1
+
+## 5215. Structured Distillation for Personalized Agent Memory: 11x Token Reduction with Retrieval Preservation
+
+- **ID:** `2603.13017` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Sydney Lewis
+- **abs:** https://arxiv.org/abs/2603.13017 · **PDF:** https://arxiv.org/pdf/2603.13017v1
+
+## 5216. FraudFox: Adaptable Fraud Detection in the Real World
+
+- **ID:** `2603.13014` · **primary:** `cs.CR` · **Submitted:** 2026-03-13
+- **Authors:** Matthew Butler, Yi Fan, Christos Faloutsos
+- **abs:** https://arxiv.org/abs/2603.13014 · **PDF:** https://arxiv.org/pdf/2603.13014v1
+
+## 5217. Accelerating Stroke MRI with Diffusion Probabilistic Models through Large-Scale Pre-training and Target-Specific Fine-Tuning
+
+- **ID:** `2603.13007` · **primary:** `eess.IV` · **Submitted:** 2026-03-13
+- **Authors:** Yamin Arefeen, Sidharth Kumar, Steven Warach, et al.
+- **abs:** https://arxiv.org/abs/2603.13007 · **PDF:** https://arxiv.org/pdf/2603.13007v1
+
+## 5218. JCAS-MARL: Joint Communication and Sensing UAV Networks via Resource-Constrained Multi-Agent Reinforcement Learning
+
+- **ID:** `2603.20265` · **primary:** `cs.IT` · **Submitted:** 2026-03-13
+- **Authors:** Islam Guven, Mehmet Parlak
+- **abs:** https://arxiv.org/abs/2603.20265 · **PDF:** https://arxiv.org/pdf/2603.20265v1
+
+## 5219. Deconstructing the Failure of Ideal Noise Correction: A Three-Pillar Diagnosis
+
+- **ID:** `2603.12997` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Chen Feng, Zhuo Zhi, Zhao Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.12997 · **PDF:** https://arxiv.org/pdf/2603.12997v1
+
+## 5220. Dependency-Aware Parallel Decoding via Attention for Diffusion LLMs
+
+- **ID:** `2603.12996` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Bumjun Kim, Dongjae Jeon, Moongyu Jeon, et al.
+- **abs:** https://arxiv.org/abs/2603.12996 · **PDF:** https://arxiv.org/pdf/2603.12996v1
+
+## 5221. Fair Lung Disease Diagnosis from Chest CT via Gender-Adversarial Attention Multiple Instance Learning
+
+- **ID:** `2603.12988` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Aditya Parikh, Aasa Feragen
+- **abs:** https://arxiv.org/abs/2603.12988 · **PDF:** https://arxiv.org/pdf/2603.12988v1
+
+## 5222. FED-HARGPT: A Hybrid Centralized-Federated Approach of a Transformer-based Architecture for Human Context Recognition
+
+- **ID:** `2603.24601` · **primary:** `eess.SP` · **Submitted:** 2026-03-13
+- **Authors:** Wandemberg Gibaut, Alexandre Osorio, Amparo Munoz, et al.
+- **abs:** https://arxiv.org/abs/2603.24601 · **PDF:** https://arxiv.org/pdf/2603.24601v1
+
+## 5223. MGMAR: Metal-Guided Metal Artifact Reduction for X-ray Computed Tomography
+
+- **ID:** `2603.13447` · **primary:** `eess.IV` · **Submitted:** 2026-03-13
+- **Authors:** Hyoung Suk Park, Kiwan Jeon
+- **abs:** https://arxiv.org/abs/2603.13447 · **PDF:** https://arxiv.org/pdf/2603.13447v1
+
+## 5224. Retrieval-Enhanced Real Estate Appraisal
+
+- **ID:** `2603.12986` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Simon Popelier, Matthieu X. B. Sarazin, Maximilien Bohm, et al.
+- **abs:** https://arxiv.org/abs/2603.12986 · **PDF:** https://arxiv.org/pdf/2603.12986v1
+
+## 5225. Is Human Annotation Necessary? Iterative MBR Distillation for Error Span Detection in Machine Translation
+
+- **ID:** `2603.12983` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Boxuan Lyu, Haiyue Song, Zhi Qu
+- **abs:** https://arxiv.org/abs/2603.12983 · **PDF:** https://arxiv.org/pdf/2603.12983v2
+
+## 5226. Exact Federated Continual Unlearning for Ridge Heads on Frozen Foundation Models
+
+- **ID:** `2603.12977` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Yijun Quan, Wentai Wu, Giovanni Montana
+- **abs:** https://arxiv.org/abs/2603.12977 · **PDF:** https://arxiv.org/pdf/2603.12977v2
+
+## 5227. SCOPE: Semantic Coreset with Orthogonal Projection Embeddings for Federated learning
+
+- **ID:** `2603.12976` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Md Anwar Hossen, Nathan R. Tallent, Luanzheng Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.12976 · **PDF:** https://arxiv.org/pdf/2603.12976v1
+
+## 5228. Diffusion-based Generative Machine Learning Model for Predicting Crack Propagation in Aluminum Nitride at the Atomic Scale
+
+- **ID:** `2603.13445` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-13
+- **Authors:** Jiali Lu, Shengfeng Yang
+- **abs:** https://arxiv.org/abs/2603.13445 · **PDF:** https://arxiv.org/pdf/2603.13445v1
+
+## 5229. Long-form RewardBench: Evaluating Reward Models for Long-form Generation
+
+- **ID:** `2603.12963` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Hui Huang, Yancheng He, Wei Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.12963 · **PDF:** https://arxiv.org/pdf/2603.12963v1
+
+## 5230. Technical Case Study of Privacy-Enhancing Technologies (PETs) for Public Health
+
+- **ID:** `2603.13444` · **primary:** `cs.CR` · **Submitted:** 2026-03-13
+- **Authors:** Avinash Laddha, Danil Mikhailov, Uyi Stewart
+- **abs:** https://arxiv.org/abs/2603.13444 · **PDF:** https://arxiv.org/pdf/2603.13444v1
+
+## 5231. Efficient Real-World Autonomous Racing via Attenuated Residual Policy Optimization
+
+- **ID:** `2603.12960` · **primary:** `cs.RO` · **Submitted:** 2026-03-13
+- **Authors:** Raphael Trumpp, Denis Hoornaert, Mirco Theile, et al.
+- **abs:** https://arxiv.org/abs/2603.12960 · **PDF:** https://arxiv.org/pdf/2603.12960v1
+
+## 5232. NormCode Canvas: Making LLM Agentic Workflows Development Sustainable via Case-Based Reasoning
+
+- **ID:** `2603.13443` · **primary:** `cs.SE` · **Submitted:** 2026-03-13
+- **Authors:** Xin Guan, Yunshan Li, Ze Wang
+- **abs:** https://arxiv.org/abs/2603.13443 · **PDF:** https://arxiv.org/pdf/2603.13443v1
+
+## 5233. Delta1 with LLM: symbolic and neural integration for credible and explainable reasoning
+
+- **ID:** `2603.12953` · **primary:** `cs.LO` · **Submitted:** 2026-03-13
+- **Authors:** Yang Xu, Jun Liu, Shuwei Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.12953 · **PDF:** https://arxiv.org/pdf/2603.12953v1
+
+## 5234. Thinking in Streaming Video
+
+- **ID:** `2603.12938` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Zikang Liu, Longteng Guo, Handong Li, et al.
+- **abs:** https://arxiv.org/abs/2603.12938 · **PDF:** https://arxiv.org/pdf/2603.12938v1
+
+## 5235. Efficient and Interpretable Multi-Agent LLM Routing via Ant Colony Optimization
+
+- **ID:** `2603.12933` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Xudong Wang, Chaoning Zhang, Jiaquan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.12933 · **PDF:** https://arxiv.org/pdf/2603.12933v1
+
+## 5236. DS$^2$-Instruct: Domain-Specific Data Synthesis for Large Language Models Instruction Tuning
+
+- **ID:** `2603.12932` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Ruiyao Xu, Noelle I. Samia, Han Liu
+- **abs:** https://arxiv.org/abs/2603.12932 · **PDF:** https://arxiv.org/pdf/2603.12932v2
+
+## 5237. Rethinking VLMs for Image Forgery Detection and Localization
+
+- **ID:** `2603.12930` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Shaofeng Guo, Jiequan Cui, Richang Hong
+- **abs:** https://arxiv.org/abs/2603.12930 · **PDF:** https://arxiv.org/pdf/2603.12930v1
+
+## 5238. ODRL Policy Comparison Through Normalisation
+
+- **ID:** `2603.12926` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Jaime Osvaldo Salas, Paolo Pareti, George Konstantinidis
+- **abs:** https://arxiv.org/abs/2603.12926 · **PDF:** https://arxiv.org/pdf/2603.12926v1
+
+## 5239. Exploring Subnetwork Interactions in Heterogeneous Brain Network via Prior-Informed Graph Learning
+
+- **ID:** `2603.19307` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Siyu Liu, Guangqi Wen, Peng Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.19307 · **PDF:** https://arxiv.org/pdf/2603.19307v1
+
+## 5240. Filtered Spectral Projection for Quantum Principal Component Analysis
+
+- **ID:** `2603.13441` · **primary:** `stat.ML` · **Submitted:** 2026-03-13
+- **Authors:** Sk Mujaffar Hossain, Satadeep Bhattacharjee
+- **abs:** https://arxiv.org/abs/2603.13441 · **PDF:** https://arxiv.org/pdf/2603.13441v2
+
+## 5241. HMS-BERT: Hybrid Multi-Task Self-Training for Multilingual and Multi-Label Cyberbullying Detection
+
+- **ID:** `2603.12920` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Zixin Feng, Xinying Cui, Yifan Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.12920 · **PDF:** https://arxiv.org/pdf/2603.12920v1
+
+## 5242. Surprised by Attention: Predictable Query Dynamics for Time Series Anomaly Detection
+
+- **ID:** `2603.12916` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Kadir-Kaan Özer, René Ebeling, Markus Enzweiler
+- **abs:** https://arxiv.org/abs/2603.12916 · **PDF:** https://arxiv.org/pdf/2603.12916v2
+
+## 5243. Stake the Points: Structure-Faithful Instance Unlearning
+
+- **ID:** `2603.12915` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Kiseong Hong, JungKyoo Shin, Eunwoo Kim
+- **abs:** https://arxiv.org/abs/2603.12915 · **PDF:** https://arxiv.org/pdf/2603.12915v1
+
+## 5244. FedBPrompt: Federated Domain Generalization Person Re-Identification via Body Distribution Aware Visual Prompts
+
+- **ID:** `2603.12912` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Xin Xu, Weilong Li, Wei Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.12912 · **PDF:** https://arxiv.org/pdf/2603.12912v1
+
+## 5245. Improving Channel Estimation via Multimodal Diffusion Models with Flow Matching
+
+- **ID:** `2603.13440` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Xiaotian Fan, Xingyu Zhou, Le Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.13440 · **PDF:** https://arxiv.org/pdf/2603.13440v1
+
+## 5246. Learning from Child-Directed Speech in Two-Language Scenarios: A French-English Case Study
+
+- **ID:** `2603.12906` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Liel Binyamin, Elior Sulem
+- **abs:** https://arxiv.org/abs/2603.12906 · **PDF:** https://arxiv.org/pdf/2603.12906v1
+
+## 5247. DirPA: Addressing Prior Shift in Imbalanced Few-shot Crop-type Classification
+
+- **ID:** `2603.12905` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Joana Reuss, Ekaterina Gikalo, Marco Körner
+- **abs:** https://arxiv.org/abs/2603.12905 · **PDF:** https://arxiv.org/pdf/2603.12905v1
+
+## 5248. A theory of learning data statistics in diffusion models, from easy to hard
+
+- **ID:** `2603.12901` · **primary:** `stat.ML` · **Submitted:** 2026-03-13
+- **Authors:** Lorenzo Bardone, Claudia Merger, Sebastian Goldt
+- **abs:** https://arxiv.org/abs/2603.12901 · **PDF:** https://arxiv.org/pdf/2603.12901v1
+
+## 5249. Human-Centered Evaluation of an LLM-Based Process Modeling Copilot: A Mixed-Methods Study with Domain Experts
+
+- **ID:** `2603.12895` · **primary:** `cs.HC` · **Submitted:** 2026-03-13
+- **Authors:** Chantale Lauer, Peter Pfeiffer, Nijat Mehdiyev
+- **abs:** https://arxiv.org/abs/2603.12895 · **PDF:** https://arxiv.org/pdf/2603.12895v1
+
+## 5250. Finite Difference Flow Optimization for RL Post-Training of Text-to-Image Models
+
+- **ID:** `2603.12893` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** David McAllister, Miika Aittala, Tero Karras, et al.
+- **abs:** https://arxiv.org/abs/2603.12893 · **PDF:** https://arxiv.org/pdf/2603.12893v1
+
+## 5251. Forecasting Epileptic Seizures from Contactless Camera via Cross-Species Transfer Learning
+
+- **ID:** `2603.12887` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Mingkai Zhai, Wei Wang, Zongsheng Li, et al.
+- **abs:** https://arxiv.org/abs/2603.12887 · **PDF:** https://arxiv.org/pdf/2603.12887v1
+
+## 5252. Enhanced Drug-drug Interaction Prediction Using Adaptive Knowledge Integration
+
+- **ID:** `2603.12885` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Pengfei Liu, Jun Tao, Zhixiang Ren
+- **abs:** https://arxiv.org/abs/2603.12885 · **PDF:** https://arxiv.org/pdf/2603.12885v1
+
+## 5253. Explainable AI Using Inherently Interpretable Components for Wearable-based Health Monitoring
+
+- **ID:** `2603.12880` · **primary:** `eess.SP` · **Submitted:** 2026-03-13
+- **Authors:** Maurice Kuschel, Solveig Vieluf, Claus Reinsberger, et al.
+- **abs:** https://arxiv.org/abs/2603.12880 · **PDF:** https://arxiv.org/pdf/2603.12880v1
+
+## 5254. Test-time RL alignment exposes task familiarity artifacts in LLM benchmarks
+
+- **ID:** `2603.12875` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Kun Wang, Reinhard Heckel
+- **abs:** https://arxiv.org/abs/2603.12875 · **PDF:** https://arxiv.org/pdf/2603.12875v1
+
+## 5255. CLARIN-PT-LDB: An Open LLM Leaderboard for Portuguese to assess Language, Culture and Civility
+
+- **ID:** `2603.12872` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** João Silva, Luís Gomes, António Branco
+- **abs:** https://arxiv.org/abs/2603.12872 · **PDF:** https://arxiv.org/pdf/2603.12872v1
+
+## 5256. Surrogates for Physics-based and Data-driven Modelling of Parametric Systems: Review and New Perspectives
+
+- **ID:** `2603.12870` · **primary:** `math.NA` · **Submitted:** 2026-03-13
+- **Authors:** Matteo Giacomini, Pedro Díez
+- **abs:** https://arxiv.org/abs/2603.12870 · **PDF:** https://arxiv.org/pdf/2603.12870v1
+
+## 5257. I Know What I Don't Know: Latent Posterior Factor Models for Multi-Evidence Probabilistic Reasoning
+
+- **ID:** `2603.15670` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Aliyu Agboola Alege
+- **abs:** https://arxiv.org/abs/2603.15670 · **PDF:** https://arxiv.org/pdf/2603.15670v2
+
+## 5258. Wear Classification of Abrasive Flap Wheels using a Hierarchical Deep Learning Approach
+
+- **ID:** `2603.12852` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Falko Kähler, Maxim Wille, Ole Schmedemann, et al.
+- **abs:** https://arxiv.org/abs/2603.12852 · **PDF:** https://arxiv.org/pdf/2603.12852v1
+
+## 5259. Draft-and-Target Sampling for Video Generation Policy
+
+- **ID:** `2603.13438` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Qikang Zhang, Yingjie Lei, Wei Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.13438 · **PDF:** https://arxiv.org/pdf/2603.13438v1
+
+## 5260. On Linear Separability of the MNIST Handwritten Digits Dataset
+
+- **ID:** `2603.12850` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Ákos Hajnal
+- **abs:** https://arxiv.org/abs/2603.12850 · **PDF:** https://arxiv.org/pdf/2603.12850v1
+
+## 5261. Team LEYA in 10th ABAW Competition: Multimodal Ambivalence/Hesitancy Recognition Approach
+
+- **ID:** `2603.12848` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Elena Ryumina, Alexandr Axyonov, Dmitry Sysoev, et al.
+- **abs:** https://arxiv.org/abs/2603.12848 · **PDF:** https://arxiv.org/pdf/2603.12848v1
+
+## 5262. MiSiSUn: Minimum Simplex Semisupervised Unmixing
+
+- **ID:** `2603.20263` · **primary:** `eess.IV` · **Submitted:** 2026-03-13
+- **Authors:** Behnood Rasti, Bikram Koirala, Paul Scheunders
+- **abs:** https://arxiv.org/abs/2603.20263 · **PDF:** https://arxiv.org/pdf/2603.20263v1
+
+## 5263. Hierarchical Reference Sets for Robust Unsupervised Detection of Scattered and Clustered Outliers
+
+- **ID:** `2603.12847` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Yiqun Zhang, Zexi Tan, Xiaopeng Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.12847 · **PDF:** https://arxiv.org/pdf/2603.12847v1
+
+## 5264. DAST: A Dual-Stream Voice Anonymization Attacker with Staged Training
+
+- **ID:** `2603.12840` · **primary:** `cs.SD` · **Submitted:** 2026-03-13
+- **Authors:** Ridwan Arefeen, Xiaoxiao Miao, Rong Tong, et al.
+- **abs:** https://arxiv.org/abs/2603.12840 · **PDF:** https://arxiv.org/pdf/2603.12840v2
+
+## 5265. A New Kernel Regularity Condition for Distributed Mirror Descent: Broader Coverage and Simpler Analysis
+
+- **ID:** `2603.12838` · **primary:** `math.OC` · **Submitted:** 2026-03-13
+- **Authors:** Junwen Qiu, Ziyang Zeng, Leilei Mei, et al.
+- **abs:** https://arxiv.org/abs/2603.12838 · **PDF:** https://arxiv.org/pdf/2603.12838v1
+
+## 5266. Mask2Flow-TSE: Two-Stage Target Speaker Extraction with Masking and Flow Matching
+
+- **ID:** `2603.12837` · **primary:** `cs.SD` · **Submitted:** 2026-03-13
+- **Authors:** Junwon Moon, Hyunjin Choi, Hansol Park, et al.
+- **abs:** https://arxiv.org/abs/2603.12837 · **PDF:** https://arxiv.org/pdf/2603.12837v1
+
+## 5267. Hierarchical Dual-Change Collaborative Learning for UAV Scene Change Captioning
+
+- **ID:** `2603.12832` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Fuhai Chen, Pengpeng Huang, Junwen Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.12832 · **PDF:** https://arxiv.org/pdf/2603.12832v1
+
+## 5268. From AI Weather Prediction to Infrastructure Resilience: A Correction-Downscaling Framework for Tropical Cyclone Impacts
+
+- **ID:** `2603.12828` · **primary:** `eess.SY` · **Submitted:** 2026-03-13
+- **Authors:** You Wu, Zhenguo Wang, Naiyu Wang
+- **abs:** https://arxiv.org/abs/2603.12828 · **PDF:** https://arxiv.org/pdf/2603.12828v1
+
+## 5269. Rethinking Multiple-Choice Questions for RLVR: Unlocking Potential via Distractor Design
+
+- **ID:** `2603.12826` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Xu Guo, Qiming Ge, Jian Tong, et al.
+- **abs:** https://arxiv.org/abs/2603.12826 · **PDF:** https://arxiv.org/pdf/2603.12826v1
+
+## 5270. NanoVDR: Distilling a 2B Vision-Language Retriever into a 70M Text-Only Encoder for Visual Document Retrieval
+
+- **ID:** `2603.12824` · **primary:** `cs.IR` · **Submitted:** 2026-03-13
+- **Authors:** Zhuchenyang Liu, Yao Zhang, Yu Xiao
+- **abs:** https://arxiv.org/abs/2603.12824 · **PDF:** https://arxiv.org/pdf/2603.12824v1
+
+## 5271. Adaptive Vision-Language Model Routing for Computer Use Agents
+
+- **ID:** `2603.12823` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Xunzhuo Liu, Bowei He, Xue Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.12823 · **PDF:** https://arxiv.org/pdf/2603.12823v1
+
+## 5272. Residual SODAP: Residual Self-Organizing Domain-Adaptive Prompting with Structural Knowledge Preservation for Continual Learning
+
+- **ID:** `2603.12816` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Gyutae Oh, Jungwoo Bae, Jitae Shin
+- **abs:** https://arxiv.org/abs/2603.12816 · **PDF:** https://arxiv.org/pdf/2603.12816v1
+
+## 5273. A Learnable SIM Paradigm: Fundamentals, Training Techniques, and Applications
+
+- **ID:** `2603.24599` · **primary:** `eess.SP` · **Submitted:** 2026-03-13
+- **Authors:** Hetong Wang, Yashuai Cao, Tiejun Lv
+- **abs:** https://arxiv.org/abs/2603.24599 · **PDF:** https://arxiv.org/pdf/2603.24599v1
+
+## 5274. Context is all you need: Towards autonomous model-based process design using agentic AI in flowsheet simulations
+
+- **ID:** `2603.12813` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Pascal Schäfer, Lukas J. Krinke, Martin Wlotzka, et al.
+- **abs:** https://arxiv.org/abs/2603.12813 · **PDF:** https://arxiv.org/pdf/2603.12813v1
+
+## 5275. A Multi-task Large Reasoning Model for Molecular Science
+
+- **ID:** `2603.12808` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Pengfei Liu, Shuang Ge, Jun Tao, et al.
+- **abs:** https://arxiv.org/abs/2603.12808 · **PDF:** https://arxiv.org/pdf/2603.12808v1
+
+## 5276. SteerRM: Debiasing Reward Models via Sparse Autoencoders
+
+- **ID:** `2603.12795` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Mengyuan Sun, Zhuohao Yu, Weizheng Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.12795 · **PDF:** https://arxiv.org/pdf/2603.12795v1
+
+## 5277. A Fractional Fox H-Function Kernel for Support Vector Machines: Robust Classification via Weighted Transmutation Operators
+
+- **ID:** `2603.12794` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Gustavo Dorrego
+- **abs:** https://arxiv.org/abs/2603.12794 · **PDF:** https://arxiv.org/pdf/2603.12794v2
+
+## 5278. Cheers: Decoupling Patch Details from Semantic Representations Enables Unified Multimodal Comprehension and Generation
+
+- **ID:** `2603.12793` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Yichen Zhang, Da Peng, Zonghao Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.12793 · **PDF:** https://arxiv.org/pdf/2603.12793v1
+
+## 5279. Upper Bounds for Local Learning Coefficients of Three-Layer Neural Networks
+
+- **ID:** `2603.12785` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Yuki Kurumadani
+- **abs:** https://arxiv.org/abs/2603.12785 · **PDF:** https://arxiv.org/pdf/2603.12785v1
+
+## 5280. The RIGID Framework: Research-Integrated, Generative AI-Mediated Instructional Design
+
+- **ID:** `2603.12781` · **primary:** `cs.CY` · **Submitted:** 2026-03-13
+- **Authors:** Yerin Kwak, Zachary A. Pardos
+- **abs:** https://arxiv.org/abs/2603.12781 · **PDF:** https://arxiv.org/pdf/2603.12781v1
+
+## 5281. Empowering Semantic-Sensitive Underwater Image Enhancement with VLM
+
+- **ID:** `2603.12773` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Guodong Fan, Shengning Zhou, Genji Yuan, et al.
+- **abs:** https://arxiv.org/abs/2603.12773 · **PDF:** https://arxiv.org/pdf/2603.12773v1
+
+## 5282. PVI: Plug-in Visual Injection for Vision-Language-Action Models
+
+- **ID:** `2603.12772` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Zezhou Zhang, Songxin Zhang, Xiao Xiong, et al.
+- **abs:** https://arxiv.org/abs/2603.12772 · **PDF:** https://arxiv.org/pdf/2603.12772v1
+
+## 5283. SectEval: Evaluating the Latent Sectarian Preferences of Large Language Models
+
+- **ID:** `2603.12768` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Aditya Maheshwari, Amit Gajkeshwar, Kaushal Sharma, et al.
+- **abs:** https://arxiv.org/abs/2603.12768 · **PDF:** https://arxiv.org/pdf/2603.12768v1
+
+## 5284. CtrlAttack: A Unified Attack on World-Model Control in Diffusion Models
+
+- **ID:** `2603.13435` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Shuhan Xu, Siyuan Liang, Hongling Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.13435 · **PDF:** https://arxiv.org/pdf/2603.13435v1
+
+## 5285. TerraFlow: Multimodal, Multitemporal Representation Learning for Earth Observation
+
+- **ID:** `2603.12762` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Nazar Puriy, Johannes Jakubik, Benedikt Blumenstiel, et al.
+- **abs:** https://arxiv.org/abs/2603.12762 · **PDF:** https://arxiv.org/pdf/2603.12762v1
+
+## 5286. FC-Track: Overlap-Aware Post-Association Correction for Online Multi-Object Tracking
+
+- **ID:** `2603.12758` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Cheng Ju, Zejing Zhao, Akio Namiki
+- **abs:** https://arxiv.org/abs/2603.12758 · **PDF:** https://arxiv.org/pdf/2603.12758v1
+
+## 5287. AI Model Modulation with Logits Redistribution
+
+- **ID:** `2603.12755` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Zihan Wang, Zhongkui Ma, Xinguo Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.12755 · **PDF:** https://arxiv.org/pdf/2603.12755v1
+
+## 5288. A Method for Learning Large-Scale Computational Construction Grammars from Semantically Annotated Corpora
+
+- **ID:** `2603.12754` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Paul Van Eecke, Katrien Beuls
+- **abs:** https://arxiv.org/abs/2603.12754 · **PDF:** https://arxiv.org/pdf/2603.12754v1
+
+## 5289. Taming the Long Tail: Efficient Item-wise Sharpness-Aware Minimization for LLM-based Recommender Systems
+
+- **ID:** `2603.12752` · **primary:** `cs.IR` · **Submitted:** 2026-03-13
+- **Authors:** Jiaming Zhang, Yuyuan Li, Xiaohua Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.12752 · **PDF:** https://arxiv.org/pdf/2603.12752v1
+
+## 5290. Show, Don't Tell: Detecting Novel Objects by Watching Human Videos
+
+- **ID:** `2603.12751` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** James Akl, Jose Nicolas Avendano Arbelaez, James Barabas, et al.
+- **abs:** https://arxiv.org/abs/2603.12751 · **PDF:** https://arxiv.org/pdf/2603.12751v1
+
+## 5291. SLICE: Semantic Latent Injection via Compartmentalized Embedding for Image Watermarking
+
+- **ID:** `2603.12749` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Zheng Gao, Yifan Yang, Xiaoyu Li, et al.
+- **abs:** https://arxiv.org/abs/2603.12749 · **PDF:** https://arxiv.org/pdf/2603.12749v1
+
+## 5292. Modality-free Graph In-context Alignment
+
+- **ID:** `2603.13434` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Wei Zhuo, Siqiang Luo
+- **abs:** https://arxiv.org/abs/2603.13434 · **PDF:** https://arxiv.org/pdf/2603.13434v1
+
+## 5293. Music Source Restoration with Ensemble Separation and Targeted Reconstruction
+
+- **ID:** `2603.16926` · **primary:** `cs.SD` · **Submitted:** 2026-03-13
+- **Authors:** Xinlong Deng, Yu Xia, Jie Jiang
+- **abs:** https://arxiv.org/abs/2603.16926 · **PDF:** https://arxiv.org/pdf/2603.16926v1
+
+## 5294. TaoBench: Do Automated Theorem Prover LLMs Generalize Beyond MathLib?
+
+- **ID:** `2603.12744` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Alexander K Taylor, Junyi Zhang, Ethan Ji, et al.
+- **abs:** https://arxiv.org/abs/2603.12744 · **PDF:** https://arxiv.org/pdf/2603.12744v1
+
+## 5295. MoKus: Leveraging Cross-Modal Knowledge Transfer for Knowledge-Aware Concept Customization
+
+- **ID:** `2603.12743` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Chenyang Zhu, Hongxiang Li, Xiu Li, et al.
+- **abs:** https://arxiv.org/abs/2603.12743 · **PDF:** https://arxiv.org/pdf/2603.12743v1
+
+## 5296. ToolTree: Efficient LLM Agent Tool Planning via Dual-Feedback Monte Carlo Tree Search and Bidirectional Pruning
+
+- **ID:** `2603.12740` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Shuo Yang, Soyeon Caren Han, Yihao Ding, et al.
+- **abs:** https://arxiv.org/abs/2603.12740 · **PDF:** https://arxiv.org/pdf/2603.12740v1
+
+## 5297. SRAM-Based Compute-in-Memory Accelerator for Linear-decay Spiking Neural Networks
+
+- **ID:** `2603.12739` · **primary:** `cs.NE` · **Submitted:** 2026-03-13
+- **Authors:** Hongyang Shang, Shuai Dong, Yahan Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.12739 · **PDF:** https://arxiv.org/pdf/2603.12739v1
+
+## 5298. VecMol: Vector-Field Representations for 3D Molecule Generation
+
+- **ID:** `2603.12734` · **primary:** `stat.ML` · **Submitted:** 2026-03-13
+- **Authors:** Yuchen Hua, Xingang Peng, Jianzhu Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.12734 · **PDF:** https://arxiv.org/pdf/2603.12734v1
+
+## 5299. On Using Machine Learning to Early Detect Catastrophic Failures in Marine Diesel Engines
+
+- **ID:** `2603.12733` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Francesco Maione, Paolo Lino, Giuseppe Giannino, et al.
+- **abs:** https://arxiv.org/abs/2603.12733 · **PDF:** https://arxiv.org/pdf/2603.12733v1
+
+## 5300. VERDICT: Verifiable Evolving Reasoning with Directive-Informed Collegial Teams for Legal Judgment Prediction
+
+- **ID:** `2603.19306` · **primary:** `cs.IR` · **Submitted:** 2026-03-13
+- **Authors:** Hui Liao, Chuan Qin, Yongwen Ren, et al.
+- **abs:** https://arxiv.org/abs/2603.19306 · **PDF:** https://arxiv.org/pdf/2603.19306v1
+
+## 5301. Anchored Alignment: Preventing Positional Collapse in Multimodal Recommender Systems
+
+- **ID:** `2603.12726` · **primary:** `cs.IR` · **Submitted:** 2026-03-13
+- **Authors:** Yonghun Jeong, David Yoon Suk Kang, Yeon-Chang Lee
+- **abs:** https://arxiv.org/abs/2603.12726 · **PDF:** https://arxiv.org/pdf/2603.12726v1
+
+## 5302. Graph In-Context Operator Networks for Generalizable Spatiotemporal Prediction
+
+- **ID:** `2603.12725` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Chenghan Wu, Zongmin Yu, Boai Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.12725 · **PDF:** https://arxiv.org/pdf/2603.12725v2
+
+## 5303. SciDesignBench: Benchmarking and Improving Language Models for Scientific Inverse Design
+
+- **ID:** `2603.12724` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** David van Dijk, Ivan Vrkic
+- **abs:** https://arxiv.org/abs/2603.12724 · **PDF:** https://arxiv.org/pdf/2603.12724v1
+
+## 5304. CognitionCapturerPro: Towards High-Fidelity Visual Decoding from EEG/MEG via Multi-modal Information and Asymmetric Alignment
+
+- **ID:** `2603.12722` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Kaifan Zhang, Lihuo He, Junjie Ke, et al.
+- **abs:** https://arxiv.org/abs/2603.12722 · **PDF:** https://arxiv.org/pdf/2603.12722v1
+
+## 5305. CMHANet: A Cross-Modal Hybrid Attention Network for Point Cloud Registration
+
+- **ID:** `2603.12721` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Dongxu Zhang, Yingsen Wang, Yiding Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.12721 · **PDF:** https://arxiv.org/pdf/2603.12721v1
+
+## 5306. IGASA: Integrated Geometry-Aware and Skip-Attention Modules for Enhanced Point Cloud Registration
+
+- **ID:** `2603.12719` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Dongxu Zhang, Jihua Zhu, Shiqi Li, et al.
+- **abs:** https://arxiv.org/abs/2603.12719 · **PDF:** https://arxiv.org/pdf/2603.12719v1
+
+## 5307. Altered Thoughts, Altered Actions: Probing Chain-of-Thought Vulnerabilities in VLA Robotic Manipulation
+
+- **ID:** `2603.12717` · **primary:** `cs.RO` · **Submitted:** 2026-03-13
+- **Authors:** Tuan Duong Trinh, Naveed Akhtar, Basim Azam
+- **abs:** https://arxiv.org/abs/2603.12717 · **PDF:** https://arxiv.org/pdf/2603.12717v1
+
+## 5308. UNIStainNet: Foundation-Model-Guided Virtual Staining of H&E to IHC
+
+- **ID:** `2603.12716` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Jillur Rahman Saurav, Thuong Le Hoai Pham, Pritam Mukherjee, et al.
+- **abs:** https://arxiv.org/abs/2603.12716 · **PDF:** https://arxiv.org/pdf/2603.12716v1
+
+## 5309. Design-Specification Tiling for ICL-based CAD Code Generation
+
+- **ID:** `2603.12712` · **primary:** `cs.SE` · **Submitted:** 2026-03-13
+- **Authors:** Yali Du, San-Zhuo Xi, Hui Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.12712 · **PDF:** https://arxiv.org/pdf/2603.12712v1
+
+## 5310. AI Planning Framework for LLM-Based Web Agents
+
+- **ID:** `2603.12710` · **primary:** `cs.AI` · **Submitted:** 2026-03-13
+- **Authors:** Orit Shahnovsky, Rotem Dror
+- **abs:** https://arxiv.org/abs/2603.12710 · **PDF:** https://arxiv.org/pdf/2603.12710v1
+
+## 5311. PhyGile: Physics-Prefix Guided Motion Generation for Agile General Humanoid Motion Tracking
+
+- **ID:** `2603.19305` · **primary:** `cs.RO` · **Submitted:** 2026-03-13
+- **Authors:** Jiacheng Bao, Haoran Yang, Yucheng Xin, et al.
+- **abs:** https://arxiv.org/abs/2603.19305 · **PDF:** https://arxiv.org/pdf/2603.19305v1
+
+## 5312. Cost-Efficient Multimodal LLM Inference via Cross-Tier GPU Heterogeneity
+
+- **ID:** `2603.12707` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Donglin Yu
+- **abs:** https://arxiv.org/abs/2603.12707 · **PDF:** https://arxiv.org/pdf/2603.12707v1
+
+## 5313. FGTR: Fine-Grained Multi-Table Retrieval via Hierarchical LLM Reasoning
+
+- **ID:** `2603.12702` · **primary:** `cs.IR` · **Submitted:** 2026-03-13
+- **Authors:** Chaojie Sun, Bin Cao, Tiantian Li, et al.
+- **abs:** https://arxiv.org/abs/2603.12702 · **PDF:** https://arxiv.org/pdf/2603.12702v2
+
+## 5314. Seeing Eye to Eye: Enabling Cognitive Alignment Through Shared First-Person Perspective in Human-AI Collaboration
+
+- **ID:** `2603.12701` · **primary:** `cs.HC` · **Submitted:** 2026-03-13
+- **Authors:** Zhuyu Teng, Pei Chen, Yichen Cai, et al.
+- **abs:** https://arxiv.org/abs/2603.12701 · **PDF:** https://arxiv.org/pdf/2603.12701v1
+
+## 5315. EvolveCoder: Evolving Test Cases via Adversarial Verification for Code Reinforcement Learning
+
+- **ID:** `2603.12698` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Chi Ruan, Dongfu Jiang, Huaye Zeng, et al.
+- **abs:** https://arxiv.org/abs/2603.12698 · **PDF:** https://arxiv.org/pdf/2603.12698v1
+
+## 5316. Deciphering Scientific Reasoning Steps from Outcome Data for Molecule Optimization
+
+- **ID:** `2603.20262` · **primary:** `q-bio.BM` · **Submitted:** 2026-03-13
+- **Authors:** Zequn Liu, Kehan Wu, Shufang Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.20262 · **PDF:** https://arxiv.org/pdf/2603.20262v1
+
+## 5317. RXNRECer Enables Fine-grained Enzymatic Function Annotation through Active Learning and Protein Language Models
+
+- **ID:** `2603.12694` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Zhenkun Shi, Jun Zhu, Dehang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.12694 · **PDF:** https://arxiv.org/pdf/2603.12694v1
+
+## 5318. HSEmotion Team at ABAW-10 Competition: Facial Expression Recognition, Valence-Arousal Estimation, Action Unit Detection and Fine-Grained Violence Classification
+
+- **ID:** `2603.12693` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Andrey V. Savchenko, Kseniia Tsypliakova
+- **abs:** https://arxiv.org/abs/2603.12693 · **PDF:** https://arxiv.org/pdf/2603.12693v1
+
+## 5319. STRAP-ViT: Segregated Tokens with Randomized -- Transformations for Defense against Adversarial Patches in ViTs
+
+- **ID:** `2603.12688` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Nandish Chattopadhyay, Anadi Goyal, Chandan Karfa, et al.
+- **abs:** https://arxiv.org/abs/2603.12688 · **PDF:** https://arxiv.org/pdf/2603.12688v1
+
+## 5320. Federated Hierarchical Clustering with Automatic Selection of Optimal Cluster Numbers
+
+- **ID:** `2603.12684` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Yue Zhang, Chuanlong Qiu, Xinfa Liao, et al.
+- **abs:** https://arxiv.org/abs/2603.12684 · **PDF:** https://arxiv.org/pdf/2603.12684v1
+
+## 5321. Experimental evidence of progressive ChatGPT models self-convergence
+
+- **ID:** `2603.12683` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Konstantinos F. Xylogiannopoulos, Petros Xanthopoulos, Panagiotis Karampelas, et al.
+- **abs:** https://arxiv.org/abs/2603.12683 · **PDF:** https://arxiv.org/pdf/2603.12683v2
+
+## 5322. Colluding LoRA: A Compositional Vulnerability in LLM Safety Alignment
+
+- **ID:** `2603.12681` · **primary:** `cs.CR` · **Submitted:** 2026-03-13
+- **Authors:** Sihao Ding
+- **abs:** https://arxiv.org/abs/2603.12681 · **PDF:** https://arxiv.org/pdf/2603.12681v2
+
+## 5323. MetaKE: Meta-learning Aligned Knowledge Editing via Bi-level Optimization
+
+- **ID:** `2603.12677` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Shuxin Liu, Ou Wu
+- **abs:** https://arxiv.org/abs/2603.12677 · **PDF:** https://arxiv.org/pdf/2603.12677v2
+
+## 5324. Disentangled Latent Dynamics Manifold Fusion for Solving Parameterized PDEs
+
+- **ID:** `2603.12676` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Zhangyong Liang, Ji Zhang
+- **abs:** https://arxiv.org/abs/2603.12676 · **PDF:** https://arxiv.org/pdf/2603.12676v1
+
+## 5325. Spatially Grounded Long-Horizon Task Planning in the Wild
+
+- **ID:** `2603.13433` · **primary:** `cs.RO` · **Submitted:** 2026-03-13
+- **Authors:** Sehun Jung, HyunJee Song, Dong-Hee Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.13433 · **PDF:** https://arxiv.org/pdf/2603.13433v1
+
+## 5326. Vision Verification Enhanced Fusion of VLMs for Efficient Visual Reasoning
+
+- **ID:** `2603.12669` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Selim Furkan Tekin, Yichang Xu, Gaowen Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.12669 · **PDF:** https://arxiv.org/pdf/2603.12669v1
+
+## 5327. Marker-Based 3D Reconstruction of Aggregates with a Comparative Analysis of 2D and 3D Morphologies
+
+- **ID:** `2603.12667` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Haohang Huang, Jiayi Luo, Issam Qamhia, et al.
+- **abs:** https://arxiv.org/abs/2603.12667 · **PDF:** https://arxiv.org/pdf/2603.12667v1
+
+## 5328. RetroReasoner: A Reasoning LLM for Strategic Retrosynthesis Prediction
+
+- **ID:** `2603.12666` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Hanbum Ko, Chanhui Lee, Ye Rin Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.12666 · **PDF:** https://arxiv.org/pdf/2603.12666v1
+
+## 5329. Spatial Transcriptomics as Images for Large-Scale Pretraining
+
+- **ID:** `2603.13432` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Yishun Zhu, Jiaxin Qi, Jian Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.13432 · **PDF:** https://arxiv.org/pdf/2603.13432v3
+
+## 5330. From Text to Forecasts: Bridging Modality Gap with Temporal Evolution Semantic Space
+
+- **ID:** `2603.12664` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Lehui Li, Yuyao Wang, Jisheng Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.12664 · **PDF:** https://arxiv.org/pdf/2603.12664v2
+
+## 5331. CHIMERA-Bench: A Benchmark Dataset for Epitope-Specific Antibody Design
+
+- **ID:** `2603.13431` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Mansoor Ahmed, Nadeem Taj, Imdad Ullah Khan, et al.
+- **abs:** https://arxiv.org/abs/2603.13431 · **PDF:** https://arxiv.org/pdf/2603.13431v2
+
+## 5332. Continual Learning in Large Language Models: Methods, Challenges, and Opportunities
+
+- **ID:** `2603.12658` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Hongyang Chen, Zhongwu Sun, Hongfei Ye, et al.
+- **abs:** https://arxiv.org/abs/2603.12658 · **PDF:** https://arxiv.org/pdf/2603.12658v1
+
+## 5333. Sobolev--Ricci Curvature
+
+- **ID:** `2603.12652` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Kyoichi Iwasaki, Tam Le, Hideitsu Hino
+- **abs:** https://arxiv.org/abs/2603.12652 · **PDF:** https://arxiv.org/pdf/2603.12652v1
+
+## 5334. LR-SGS: Robust LiDAR-Reflectance-Guided Salient Gaussian Splatting for Self-Driving Scene Reconstruction
+
+- **ID:** `2603.12647` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Ziyu Chen, Fan Zhu, Hui Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.12647 · **PDF:** https://arxiv.org/pdf/2603.12647v1
+
+## 5335. 98$\times$ Faster LLM Routing Without a Dedicated GPU: Flash Attention, Prompt Compression, and Near-Streaming for the vLLM Semantic Router
+
+- **ID:** `2603.12646` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Xunzhuo Liu, Bowei He, Xue Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.12646 · **PDF:** https://arxiv.org/pdf/2603.12646v1
+
+## 5336. LightMoE: Reducing Mixture-of-Experts Redundancy through Expert Replacing
+
+- **ID:** `2603.12645` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Jiawei Hao, Zhiwei Hao, Jianyuan Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.12645 · **PDF:** https://arxiv.org/pdf/2603.12645v1
+
+## 5337. Self-Supervised Speech Models Encode Phonetic Context via Position-dependent Orthogonal Subspaces
+
+- **ID:** `2603.12642` · **primary:** `eess.AS` · **Submitted:** 2026-03-13
+- **Authors:** Kwanghee Choi, Eunjung Yeo, Cheol Jun Cho, et al.
+- **abs:** https://arxiv.org/abs/2603.12642 · **PDF:** https://arxiv.org/pdf/2603.12642v1
+
+## 5338. Using a Human-AI Teaming Approach to Create and Curate Scientific Datasets with the SCILIRE System
+
+- **ID:** `2603.12638` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Necva Bölücü, Jessica Irons, Changhyun Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.12638 · **PDF:** https://arxiv.org/pdf/2603.12638v1
+
+## 5339. Weakly Time-Coupled Approximation of Markov Decision Processes
+
+- **ID:** `2603.12636` · **primary:** `math.OC` · **Submitted:** 2026-03-13
+- **Authors:** Negar Soheili, Selvaprabu Nadarajah, Bo Yang
+- **abs:** https://arxiv.org/abs/2603.12636 · **PDF:** https://arxiv.org/pdf/2603.12636v1
+
+## 5340. Adaptive Diffusion Posterior Sampling for Data and Model Fusion of Complex Nonlinear Dynamical Systems
+
+- **ID:** `2603.12635` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Dibyajyoti Chakraborty, Hojin Kim, Romit Maulik
+- **abs:** https://arxiv.org/abs/2603.12635 · **PDF:** https://arxiv.org/pdf/2603.12635v1
+
+## 5341. Spend Less, Reason Better: Budget-Aware Value Tree Search for LLM Agents
+
+- **ID:** `2603.12634` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Yushu Li, Wenlong Deng, Jiajin Li, et al.
+- **abs:** https://arxiv.org/abs/2603.12634 · **PDF:** https://arxiv.org/pdf/2603.12634v1
+
+## 5342. The Economics of AI Supply Chain Regulation
+
+- **ID:** `2603.12630` · **primary:** `econ.TH` · **Submitted:** 2026-03-13
+- **Authors:** Sihan Qian, Amit Mehra, Dengpan Liu
+- **abs:** https://arxiv.org/abs/2603.12630 · **PDF:** https://arxiv.org/pdf/2603.12630v1
+
+## 5343. Towards unified brain-to-text decoding across speech production and perception
+
+- **ID:** `2603.12628` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-13
+- **Authors:** Zhizhang Yuan, Yang Yang, Gaorui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.12628 · **PDF:** https://arxiv.org/pdf/2603.12628v1
+
+## 5344. Batched Kernelized Bandits: Refinements and Extensions
+
+- **ID:** `2603.12627` · **primary:** `stat.ML` · **Submitted:** 2026-03-13
+- **Authors:** Chenkai Ma, Keqin Chen, Jonathan Scarlett
+- **abs:** https://arxiv.org/abs/2603.12627 · **PDF:** https://arxiv.org/pdf/2603.12627v1
+
+## 5345. VLM4Rec: Multimodal Semantic Representation for Recommendation with Large Vision-Language Models
+
+- **ID:** `2603.12625` · **primary:** `cs.IR` · **Submitted:** 2026-03-13
+- **Authors:** Ty Valencia, Burak Barlas, Varun Singhal, et al.
+- **abs:** https://arxiv.org/abs/2603.12625 · **PDF:** https://arxiv.org/pdf/2603.12625v1
+
+## 5346. Human-AI Collaborative Autonomous Experimentation With Proxy Modeling for Comparative Observation
+
+- **ID:** `2603.12618` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Arpan Biswas, Hiroshi Funakubo, Yongtao Liu
+- **abs:** https://arxiv.org/abs/2603.12618 · **PDF:** https://arxiv.org/pdf/2603.12618v1
+
+## 5347. When Drafts Evolve: Speculative Decoding Meets Online Learning
+
+- **ID:** `2603.12617` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Yu-Yang Qian, Hao-Cong Wu, Yichao Fu, et al.
+- **abs:** https://arxiv.org/abs/2603.12617 · **PDF:** https://arxiv.org/pdf/2603.12617v1
+
+## 5348. Literary Narrative as Moral Probe : A Cross-System Framework for Evaluating AI Ethical Reasoning and Refusal Behavior
+
+- **ID:** `2603.12615` · **primary:** `cs.CY` · **Submitted:** 2026-03-13
+- **Authors:** David C. Flynn
+- **abs:** https://arxiv.org/abs/2603.12615 · **PDF:** https://arxiv.org/pdf/2603.12615v1
+
+## 5349. Dynamic Sparse Attention: Access Patterns and Architecture
+
+- **ID:** `2603.13430` · **primary:** `cs.AR` · **Submitted:** 2026-03-13
+- **Authors:** Noam Levy
+- **abs:** https://arxiv.org/abs/2603.13430 · **PDF:** https://arxiv.org/pdf/2603.13430v1
+
+## 5350. FastDSAC: Unlocking the Potential of Maximum Entropy RL in High-Dimensional Humanoid Control
+
+- **ID:** `2603.12612` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Jun Xue, Junze Wang, Xinming Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.12612 · **PDF:** https://arxiv.org/pdf/2603.12612v1
+
+## 5351. CarPLAN: Context-Adaptive and Robust Planning with Dynamic Scene Awareness for Autonomous Driving
+
+- **ID:** `2603.12607` · **primary:** `cs.RO` · **Submitted:** 2026-03-13
+- **Authors:** Junyong Yun, Jungho Kim, ByungHyun Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.12607 · **PDF:** https://arxiv.org/pdf/2603.12607v1
+
+## 5352. Mastering Negation: Boosting Grounding Models via Grouped Opposition-Based Learning
+
+- **ID:** `2603.12606` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Zesheng Yang, Xi Jiang, Bingzhang Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.12606 · **PDF:** https://arxiv.org/pdf/2603.12606v1
+
+## 5353. EvoClaw: Evaluating AI Agents on Continuous Software Evolution
+
+- **ID:** `2603.13428` · **primary:** `cs.SE` · **Submitted:** 2026-03-13
+- **Authors:** Gangda Deng, Zhaoling Chen, Zhongming Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.13428 · **PDF:** https://arxiv.org/pdf/2603.13428v1
+
+## 5354. Feynman: Knowledge-Infused Diagramming Agent for Scalable Visual Designs
+
+- **ID:** `2603.12597` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Zixin Wen, Yifu Cai, Kyle Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.12597 · **PDF:** https://arxiv.org/pdf/2603.12597v1
+
+## 5355. MIBench: Evaluating LMMs on Multimodal Interaction
+
+- **ID:** `2603.13427` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Yu Miao, Zequn Yang, Yake Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.13427 · **PDF:** https://arxiv.org/pdf/2603.13427v1
+
+## 5356. Optimize Wider, Not Deeper: Consensus Aggregation for Policy Optimization
+
+- **ID:** `2603.12596` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Zelal Su, Mustafaoglu, Sungyoung Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.12596 · **PDF:** https://arxiv.org/pdf/2603.12596v1
+
+## 5357. Swap-guided Preference Learning for Personalized Reinforcement Learning from Human Feedback
+
+- **ID:** `2603.12595` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Gihoon Kim, Euntai Kim
+- **abs:** https://arxiv.org/abs/2603.12595 · **PDF:** https://arxiv.org/pdf/2603.12595v1
+
+## 5358. Maximizing Incremental Information Entropy for Contrastive Learning
+
+- **ID:** `2603.12594` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Jiansong Zhang, Zhuoqin Yang, Xu Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.12594 · **PDF:** https://arxiv.org/pdf/2603.12594v1
+
+## 5359. Early Pruning for Public Transport Routing
+
+- **ID:** `2603.12592` · **primary:** `cs.DS` · **Submitted:** 2026-03-13
+- **Authors:** Andrii Rohovyi, Abdallah Abuaisha, Toby Walsh
+- **abs:** https://arxiv.org/abs/2603.12592 · **PDF:** https://arxiv.org/pdf/2603.12592v1
+
+## 5360. CA-HFP: Curvature-Aware Heterogeneous Federated Pruning with Model Reconstruction
+
+- **ID:** `2603.12591` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Gang Hu, Yinglei Teng, Pengfei Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.12591 · **PDF:** https://arxiv.org/pdf/2603.12591v1
+
+## 5361. Response-Aware Risk-Constrained Control Barrier Function With Application to Vehicles
+
+- **ID:** `2603.24598` · **primary:** `math.OC` · **Submitted:** 2026-03-13
+- **Authors:** Qijun Liao, Jue Yang
+- **abs:** https://arxiv.org/abs/2603.24598 · **PDF:** https://arxiv.org/pdf/2603.24598v1
+
+## 5362. Deferred is Better: A Framework for Multi-Granularity Deferred Interaction of Heterogeneous Features
+
+- **ID:** `2603.12586` · **primary:** `cs.IR` · **Submitted:** 2026-03-13
+- **Authors:** Yi Xu, Moyu Zhang, Chaofan Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.12586 · **PDF:** https://arxiv.org/pdf/2603.12586v1
+
+## 5363. RTD-Guard: A Black-Box Textual Adversarial Detection Framework via Replacement Token Detection
+
+- **ID:** `2603.12582` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** He Zhu, Yanshu Li, Wen Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.12582 · **PDF:** https://arxiv.org/pdf/2603.12582v1
+
+## 5364. Multiscale Structure-Guided Latent Diffusion for Multimodal MRI Translation
+
+- **ID:** `2603.12581` · **primary:** `eess.IV` · **Submitted:** 2026-03-13
+- **Authors:** Jianqiang Lin, Zhiqiang Shen, Peng Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.12581 · **PDF:** https://arxiv.org/pdf/2603.12581v1
+
+## 5365. Expert Pyramid Tuning: Efficient Parameter Fine-Tuning for Expertise-Driven Task Allocation
+
+- **ID:** `2603.12577` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Jia-Chen Zhang, Zhen-Wei Yan, Yu-Jie Xiong, et al.
+- **abs:** https://arxiv.org/abs/2603.12577 · **PDF:** https://arxiv.org/pdf/2603.12577v1
+
+## 5366. A Spectral Revisit of the Distributional Bellman Operator under the Cramér Metric
+
+- **ID:** `2603.12576` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Keru Wang, Yixin Deng, Yao Lyu, et al.
+- **abs:** https://arxiv.org/abs/2603.12576 · **PDF:** https://arxiv.org/pdf/2603.12576v1
+
+## 5367. Outcome-Aware Tool Selection for Semantic Routers: Latency-Constrained Learning Without LLM Inference
+
+- **ID:** `2603.13426` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Huamin Chen, Xunzhuo Liu, Junchen Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.13426 · **PDF:** https://arxiv.org/pdf/2603.13426v1
+
+## 5368. Self-Flow-Matching assisted Full Waveform Inversion
+
+- **ID:** `2603.13425` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Xinquan Huang, Paris Perdikaris
+- **abs:** https://arxiv.org/abs/2603.13425 · **PDF:** https://arxiv.org/pdf/2603.13425v1
+
+## 5369. LMEB: Long-horizon Memory Embedding Benchmark
+
+- **ID:** `2603.12572` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Xinping Zhao, Xinshuo Hu, Jiaxin Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.12572 · **PDF:** https://arxiv.org/pdf/2603.12572v2
+
+## 5370. Agent Privilege Separation in OpenClaw: A Structural Defense Against Prompt Injection
+
+- **ID:** `2603.13424` · **primary:** `cs.CR` · **Submitted:** 2026-03-13
+- **Authors:** Darren Cheng, Wen-Kwang Tsao
+- **abs:** https://arxiv.org/abs/2603.13424 · **PDF:** https://arxiv.org/pdf/2603.13424v1
+
+## 5371. Foundation-Model Surrogates Enable Data-Efficient Active Learning for Materials Discovery
+
+- **ID:** `2603.12567` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-13
+- **Authors:** Jeffrey Hu, Rongzhi Dong, Ying Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.12567 · **PDF:** https://arxiv.org/pdf/2603.12567v3
+
+## 5372. Speech-Worthy Alignment for Japanese SpeechLLMs via Direct Preference Optimization
+
+- **ID:** `2603.12565` · **primary:** `cs.SD` · **Submitted:** 2026-03-13
+- **Authors:** Mengjie Zhao, Lianbo Liu, Yusuke Fujita, et al.
+- **abs:** https://arxiv.org/abs/2603.12565 · **PDF:** https://arxiv.org/pdf/2603.12565v1
+
+## 5373. AgentDrift: Unsafe Recommendation Drift Under Tool Corruption Hidden by Ranking Metrics in LLM Agents
+
+- **ID:** `2603.12564` · **primary:** `cs.CL` · **Submitted:** 2026-03-13
+- **Authors:** Zekun Wu, Adriano Koshiyama, Sahan Bulathwela, et al.
+- **abs:** https://arxiv.org/abs/2603.12564 · **PDF:** https://arxiv.org/pdf/2603.12564v5
+
+## 5374. Variational Garrote for Sparse Inverse Problems
+
+- **ID:** `2603.12562` · **primary:** `stat.ML` · **Submitted:** 2026-03-13
+- **Authors:** Kanghun Lee, Hyungjoon Soh, Junghyo Jo
+- **abs:** https://arxiv.org/abs/2603.12562 · **PDF:** https://arxiv.org/pdf/2603.12562v1
+
+## 5375. Lyapunov Stable Graph Neural Flow
+
+- **ID:** `2603.12557` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Haoyu Chu, Xiaotong Chen, Wei Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.12557 · **PDF:** https://arxiv.org/pdf/2603.12557v1
+
+## 5376. Scaling Laws and Pathologies of Single-Layer PINNs: Network Width and PDE Nonlinearity
+
+- **ID:** `2603.12556` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Faris Chaudhry
+- **abs:** https://arxiv.org/abs/2603.12556 · **PDF:** https://arxiv.org/pdf/2603.12556v1
+
+## 5377. Reinforcement Learning for Diffusion LLMs with Entropy-Guided Step Selection and Stepwise Advantages
+
+- **ID:** `2603.12554` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Vishnu Teja Kunde, Fatemeh Doudi, Mahdi Farahbakhsh, et al.
+- **abs:** https://arxiv.org/abs/2603.12554 · **PDF:** https://arxiv.org/pdf/2603.12554v1
+
+## 5378. Asymptotic and Finite-Time Guarantees for Langevin-Based Temperature Annealing in InfoNCE
+
+- **ID:** `2603.12552` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Faris Chaudhry
+- **abs:** https://arxiv.org/abs/2603.12552 · **PDF:** https://arxiv.org/pdf/2603.12552v1
+
+## 5379. Deep Distance Measurement Method for Unsupervised Multivariate Time Series Similarity Retrieval
+
+- **ID:** `2603.12544` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Susumu Naito, Kouta Nakata, Yasunori Taguchi
+- **abs:** https://arxiv.org/abs/2603.12544 · **PDF:** https://arxiv.org/pdf/2603.12544v1
+
+## 5380. CALF: Communication-Aware Learning Framework for Distributed Reinforcement Learning
+
+- **ID:** `2603.12543` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Carlos Purves, Pietro Lio'
+- **abs:** https://arxiv.org/abs/2603.12543 · **PDF:** https://arxiv.org/pdf/2603.12543v1
+
+## 5381. As Language Models Scale, Low-order Linear Depth Dynamics Emerge
+
+- **ID:** `2603.12541` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Buddhika Nettasinghe, Geethu Joseph
+- **abs:** https://arxiv.org/abs/2603.12541 · **PDF:** https://arxiv.org/pdf/2603.12541v1
+
+## 5382. Embedded Quantum Machine Learning in Embedded Systems: Feasibility, Hybrid Architectures, and Quantum Co-Processors
+
+- **ID:** `2603.12540` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Somdip Dey, Syed Muhammad Raza
+- **abs:** https://arxiv.org/abs/2603.12540 · **PDF:** https://arxiv.org/pdf/2603.12540v1
+
+## 5383. Spatio-Semantic Expert Routing Architecture with Mixture-of-Experts for Referring Image Segmentation
+
+- **ID:** `2603.12538` · **primary:** `cs.CV` · **Submitted:** 2026-03-13
+- **Authors:** Alaa Dalaq, Muzammil Behzad
+- **abs:** https://arxiv.org/abs/2603.12538 · **PDF:** https://arxiv.org/pdf/2603.12538v1
+
+## 5384. From Gradients to Riccati Geometry: Kalman World Models for Single-Pass Learning
+
+- **ID:** `2603.13423` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Andrew Kiruluta
+- **abs:** https://arxiv.org/abs/2603.13423 · **PDF:** https://arxiv.org/pdf/2603.13423v1
+
+## 5385. A Reduction Algorithm for Markovian Contextual Linear Bandits
+
+- **ID:** `2603.12530` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Kaan Buyukkalayci, Osama Hanna, Christina Fragouli
+- **abs:** https://arxiv.org/abs/2603.12530 · **PDF:** https://arxiv.org/pdf/2603.12530v1
+
+## 5386. TERMINATOR: Learning Optimal Exit Points for Early Stopping in Chain-of-Thought Reasoning
+
+- **ID:** `2603.12529` · **primary:** `cs.LG` · **Submitted:** 2026-03-13
+- **Authors:** Alliot Nagle, Jakhongir Saydaliev, Dhia Garbaya, et al.
+- **abs:** https://arxiv.org/abs/2603.12529 · **PDF:** https://arxiv.org/pdf/2603.12529v1
+
+## 5387. EB-RANSAC: Random Sample Consensus based on Energy-Based Model
+
+- **ID:** `2603.12525` · **primary:** `stat.ML` · **Submitted:** 2026-03-12
+- **Authors:** Muneki Yasuda, Nao Watanabe, Kaiji Sekimoto
+- **abs:** https://arxiv.org/abs/2603.12525 · **PDF:** https://arxiv.org/pdf/2603.12525v1
+
+## 5388. Projection Guided Personalized Federated Learning for Low Dose CT Denoising
+
+- **ID:** `2603.13422` · **primary:** `eess.IV` · **Submitted:** 2026-03-12
+- **Authors:** Anas Zafar, Muhammad Waqas, Amgad Muneer, et al.
+- **abs:** https://arxiv.org/abs/2603.13422 · **PDF:** https://arxiv.org/pdf/2603.13422v1
+
+## 5389. LLM BiasScope: A Real-Time Bias Analysis Platform for Comparative LLM Evaluation
+
+- **ID:** `2603.12522` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Himel Ghosh, Nick Elias Werner
+- **abs:** https://arxiv.org/abs/2603.12522 · **PDF:** https://arxiv.org/pdf/2603.12522v1
+
+## 5390. When LLM Judge Scores Look Good but Best-of-N Decisions Fail
+
+- **ID:** `2603.12520` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Eddie Landesberg
+- **abs:** https://arxiv.org/abs/2603.12520 · **PDF:** https://arxiv.org/pdf/2603.12520v1
+
+## 5391. Curriculum Sampling: A Two-Phase Curriculum for Efficient Training of Flow Matching
+
+- **ID:** `2603.12517` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Pengwei Sun
+- **abs:** https://arxiv.org/abs/2603.12517 · **PDF:** https://arxiv.org/pdf/2603.12517v1
+
+## 5392. Learning Pore-scale Multiphase Flow from 4D Velocimetry
+
+- **ID:** `2603.12516` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Chunyang Wang, Linqi Zhu, Yuxuan Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.12516 · **PDF:** https://arxiv.org/pdf/2603.12516v1
+
+## 5393. Addressing Data Scarcity in 3D Trauma Detection through Self-Supervised and Semi-Supervised Learning with Vertex Relative Position Encoding
+
+- **ID:** `2603.12514` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Shivam Chaudhary, Sheethal Bhat, Andreas Maier
+- **abs:** https://arxiv.org/abs/2603.12514 · **PDF:** https://arxiv.org/pdf/2603.12514v1
+
+## 5394. Byzantine-Robust Optimization under $(L_0, L_1)$-Smoothness
+
+- **ID:** `2603.12512` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Arman Bolatov, Samuel Horváth, Martin Takáč, et al.
+- **abs:** https://arxiv.org/abs/2603.12512 · **PDF:** https://arxiv.org/pdf/2603.12512v1
+
+## 5395. Red-Teaming Vision-Language-Action Models via Quality Diversity Prompt Generation for Robust Robot Policies
+
+- **ID:** `2603.12510` · **primary:** `cs.RO` · **Submitted:** 2026-03-12
+- **Authors:** Siddharth Srikanth, Freddie Liang, Ya-Chuan Hsu, et al.
+- **abs:** https://arxiv.org/abs/2603.12510 · **PDF:** https://arxiv.org/pdf/2603.12510v2
+
+## 5396. ELLA: Generative AI-Powered Social Robots for Early Language Development at Home
+
+- **ID:** `2603.12508` · **primary:** `cs.HC` · **Submitted:** 2026-03-12
+- **Authors:** Victor Nikhil Antony, Shiye Cao, Shuning Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.12508 · **PDF:** https://arxiv.org/pdf/2603.12508v1
+
+## 5397. Adaptive Conditional Forest Sampling for Spectral Risk Optimisation under Decision-Dependent Uncertainty
+
+- **ID:** `2603.12507` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Marcell T. Kurbucz
+- **abs:** https://arxiv.org/abs/2603.12507 · **PDF:** https://arxiv.org/pdf/2603.12507v1
+
+## 5398. Naïve PAINE: Lightweight Text-to-Image Generation Improvement with Prompt Evaluation
+
+- **ID:** `2603.12506` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Joong Ho Kim, Nicholas Thai, Souhardya Saha Dip, et al.
+- **abs:** https://arxiv.org/abs/2603.12506 · **PDF:** https://arxiv.org/pdf/2603.12506v1
+
+## 5399. TRACE: Temporal Rule-Anchored Chain-of-Evidence on Knowledge Graphs for Interpretable Stock Movement Prediction
+
+- **ID:** `2603.12500` · **primary:** `cs.CE` · **Submitted:** 2026-03-12
+- **Authors:** Qianggang Ding, Haochen Shi, Luis Castejón Lozano, et al.
+- **abs:** https://arxiv.org/abs/2603.12500 · **PDF:** https://arxiv.org/pdf/2603.12500v1
+
+## 5400. Probing Length Generalization in Mamba via Image Reconstruction
+
+- **ID:** `2603.12499` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Jan Rathjens, Robin Schiewer, Laurenz Wiskott, et al.
+- **abs:** https://arxiv.org/abs/2603.12499 · **PDF:** https://arxiv.org/pdf/2603.12499v1
+
+## 5401. Modal Logical Neural Networks for Financial AI
+
+- **ID:** `2603.12487` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Antonin Sulc
+- **abs:** https://arxiv.org/abs/2603.12487 · **PDF:** https://arxiv.org/pdf/2603.12487v1
+
+## 5402. Generating Expressive and Customizable Evals for Timeseries Data Analysis Agents with AgentFuel
+
+- **ID:** `2603.12483` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Aadyaa Maddi, Prakhar Naval, Deepti Mande, et al.
+- **abs:** https://arxiv.org/abs/2603.12483 · **PDF:** https://arxiv.org/pdf/2603.12483v1
+
+## 5403. One-Step Flow Policy: Self-Distillation for Fast Visuomotor Policies
+
+- **ID:** `2603.12480` · **primary:** `cs.RO` · **Submitted:** 2026-03-12
+- **Authors:** Shaolong Li, Lichao Sun, Yongchao Chen
+- **abs:** https://arxiv.org/abs/2603.12480 · **PDF:** https://arxiv.org/pdf/2603.12480v1
+
+## 5404. Less Data, Faster Convergence: Goal-Driven Data Optimization for Multimodal Instruction Tuning
+
+- **ID:** `2603.12478` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Rujie Wu, Haozhe Zhao, Hai Ci, et al.
+- **abs:** https://arxiv.org/abs/2603.12478 · **PDF:** https://arxiv.org/pdf/2603.12478v1
+
+## 5405. The Perfection Paradox: From Architect to Curator in AI-Assisted API Design
+
+- **ID:** `2603.12475` · **primary:** `cs.SE` · **Submitted:** 2026-03-12
+- **Authors:** Mak Ahmad, Andrew Macvean, JJ Geewax, et al.
+- **abs:** https://arxiv.org/abs/2603.12475 · **PDF:** https://arxiv.org/pdf/2603.12475v1
+
+## 5406. Marked Pedagogies: Examining Linguistic Biases in Personalized Automated Writing Feedback
+
+- **ID:** `2603.12471` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Mei Tan, Lena Phalen, Dorottya Demszky
+- **abs:** https://arxiv.org/abs/2603.12471 · **PDF:** https://arxiv.org/pdf/2603.12471v1
+
+## 5407. CLARE: Classification-based Regression for Electron Temperature Prediction
+
+- **ID:** `2603.12470` · **primary:** `physics.space-ph` · **Submitted:** 2026-03-12
+- **Authors:** Michael Liang, Blake DeHaas, Naomi Maruyama, et al.
+- **abs:** https://arxiv.org/abs/2603.12470 · **PDF:** https://arxiv.org/pdf/2603.12470v1
+
+## 5408. TaxBreak: Unmasking the Hidden Costs of LLM Inference Through Overhead Decomposition
+
+- **ID:** `2603.12465` · **primary:** `cs.DC` · **Submitted:** 2026-03-12
+- **Authors:** Prabhu Vellaisamy, Shreesh Tripathi, Vignesh Natarajan, et al.
+- **abs:** https://arxiv.org/abs/2603.12465 · **PDF:** https://arxiv.org/pdf/2603.12465v1
+
+## 5409. Bases of Steerable Kernels for Equivariant CNNs: From 2D Rotations to the Lorentz Group
+
+- **ID:** `2603.12459` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Alan Garbarz
+- **abs:** https://arxiv.org/abs/2603.12459 · **PDF:** https://arxiv.org/pdf/2603.12459v1
+
+## 5410. Shattering the Shortcut: A Topology-Regularized Benchmark for Multi-hop Medical Reasoning in LLMs
+
+- **ID:** `2603.12458` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Xing Zi, Xinying Zhou, Jinghao Xiao, et al.
+- **abs:** https://arxiv.org/abs/2603.12458 · **PDF:** https://arxiv.org/pdf/2603.12458v1
+
+## 5411. Operationalising Cyber Risk Management Using AI: Connecting Cyber Incidents to MITRE ATT&CK Techniques, Security Controls, and Metrics
+
+- **ID:** `2603.12455` · **primary:** `cs.CR` · **Submitted:** 2026-03-12
+- **Authors:** Emad Sherif, Iryna Yevseyeva, Vitor Basto-Fernandes, et al.
+- **abs:** https://arxiv.org/abs/2603.12455 · **PDF:** https://arxiv.org/pdf/2603.12455v1
+
+## 5412. Generalization and Memorization in Rectified Flow
+
+- **ID:** `2603.13421` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Mingxing Rao, Daniel Moyer
+- **abs:** https://arxiv.org/abs/2603.13421 · **PDF:** https://arxiv.org/pdf/2603.13421v1
+
+## 5413. CSE-UOI at SemEval-2026 Task 6: A Two-Stage Heterogeneous Ensemble with Deliberative Complexity Gating for Political Evasion Detection
+
+- **ID:** `2603.12453` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Christos Tzouvaras, Konstantinos Skianis, Athanasios Voulodimos
+- **abs:** https://arxiv.org/abs/2603.12453 · **PDF:** https://arxiv.org/pdf/2603.12453v1
+
+## 5414. Accelerating Suffix Jailbreak attacks with Prefix-Shared KV-cache
+
+- **ID:** `2603.13420` · **primary:** `cs.CR` · **Submitted:** 2026-03-12
+- **Authors:** Xinhai Wang, Shaopeng Fu, Shu Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.13420 · **PDF:** https://arxiv.org/pdf/2603.13420v1
+
+## 5415. Overcoming the Modality Gap in Context-Aided Forecasting
+
+- **ID:** `2603.12451` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Vincent Zhihao Zheng, Étienne Marcotte, Arjun Ashok, et al.
+- **abs:** https://arxiv.org/abs/2603.12451 · **PDF:** https://arxiv.org/pdf/2603.12451v2
+
+## 5416. Bridging the Gap Between Security Metrics and Key Risk Indicators: An Empirical Framework for Vulnerability Prioritization
+
+- **ID:** `2603.12450` · **primary:** `cs.CR` · **Submitted:** 2026-03-12
+- **Authors:** Emad Sherif, Iryna Yevseyeva, Vitor Basto-Fernandes, et al.
+- **abs:** https://arxiv.org/abs/2603.12450 · **PDF:** https://arxiv.org/pdf/2603.12450v1
+
+## 5417. Diffusion Models Generalize but Not in the Way You Might Think
+
+- **ID:** `2603.13419` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Tim Kaiser, Markus Kollmann
+- **abs:** https://arxiv.org/abs/2603.13419 · **PDF:** https://arxiv.org/pdf/2603.13419v1
+
+## 5418. FloeNet: A mass-conserving global sea ice emulator that generalizes across climates
+
+- **ID:** `2603.12449` · **primary:** `physics.ao-ph` · **Submitted:** 2026-03-12
+- **Authors:** William Gregory, Mitchell Bushuk, James Duncan, et al.
+- **abs:** https://arxiv.org/abs/2603.12449 · **PDF:** https://arxiv.org/pdf/2603.12449v1
+
+## 5419. Unmasking Biases and Reliability Concerns in Convolutional Neural Networks Analysis of Cancer Pathology Images
+
+- **ID:** `2603.12445` · **primary:** `eess.IV` · **Submitted:** 2026-03-12
+- **Authors:** Michael Okonoda, Eder Martinez, Abhilekha Dalal, et al.
+- **abs:** https://arxiv.org/abs/2603.12445 · **PDF:** https://arxiv.org/pdf/2603.12445v1
+
+## 5420. KernelFoundry: Hardware-aware evolutionary GPU kernel optimization
+
+- **ID:** `2603.12440` · **primary:** `cs.DC` · **Submitted:** 2026-03-12
+- **Authors:** Nina Wiedemann, Quentin Leboutet, Michael Paulitsch, et al.
+- **abs:** https://arxiv.org/abs/2603.12440 · **PDF:** https://arxiv.org/pdf/2603.12440v1
+
+## 5421. Revisiting Model Stitching In the Foundation Model Era
+
+- **ID:** `2603.12433` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Zheda Mai, Ke Zhang, Fu-En Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.12433 · **PDF:** https://arxiv.org/pdf/2603.12433v2
+
+## 5422. Interpreting Negation in GPT-2: Layer- and Head-Level Causal Analysis
+
+- **ID:** `2603.12423` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Abdullah Al Mofael, Lisa M. Kuhn, Ghassan Alkadi, et al.
+- **abs:** https://arxiv.org/abs/2603.12423 · **PDF:** https://arxiv.org/pdf/2603.12423v1
+
+## 5423. Agreement Between Large Language Models, Human Reviewers, and Authors in Evaluating STROBE Checklists for Observational Studies in Rheumatology
+
+- **ID:** `2603.19303` · **primary:** `cs.DL` · **Submitted:** 2026-03-12
+- **Authors:** Emre Bilgin, Ebru Ozturk, Meera Shah, et al.
+- **abs:** https://arxiv.org/abs/2603.19303 · **PDF:** https://arxiv.org/pdf/2603.19303v1
+
+## 5424. SpectralGuard: Detecting Memory Collapse Attacks in State Space Models
+
+- **ID:** `2603.12414` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Davi Bonetto
+- **abs:** https://arxiv.org/abs/2603.12414 · **PDF:** https://arxiv.org/pdf/2603.12414v1
+
+## 5425. Beyond Motion Imitation: Is Human Motion Data Alone Sufficient to Explain Gait Control and Biomechanics?
+
+- **ID:** `2603.12408` · **primary:** `cs.RO` · **Submitted:** 2026-03-12
+- **Authors:** Xinyi Liu, Jangwhan Ahn, Edgar Lobaton, et al.
+- **abs:** https://arxiv.org/abs/2603.12408 · **PDF:** https://arxiv.org/pdf/2603.12408v1
+
+## 5426. GPrune-LLM: Generalization-Aware Structured Pruning for Large Language Models
+
+- **ID:** `2603.13418` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Xiaoyun Liu, Divya Saxena, Jiannong Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.13418 · **PDF:** https://arxiv.org/pdf/2603.13418v1
+
+## 5427. Not Just the Destination, But the Journey: Reasoning Traces Causally Shape Generalization Behaviors
+
+- **ID:** `2603.12397` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Pengcheng Wen, Yanxu Zhu, Jiapeng Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.12397 · **PDF:** https://arxiv.org/pdf/2603.12397v1
+
+## 5428. Test-Time Strategies for More Efficient and Accurate Agentic RAG
+
+- **ID:** `2603.12396` · **primary:** `cs.IR` · **Submitted:** 2026-03-12
+- **Authors:** Brian Zhang, Deepti Guntur, Zhiyang Zuo, et al.
+- **abs:** https://arxiv.org/abs/2603.12396 · **PDF:** https://arxiv.org/pdf/2603.12396v1
+
+## 5429. PLDR-LLMs Reason At Self-Organized Criticality
+
+- **ID:** `2603.23539` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Burc Gokden
+- **abs:** https://arxiv.org/abs/2603.23539 · **PDF:** https://arxiv.org/pdf/2603.23539v1
+
+## 5430. SPARROW: Learning Spatial Precision and Temporal Referential Consistency in Pixel-Grounded Video MLLMs
+
+- **ID:** `2603.12382` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Mohamad Alansari, Naufal Suryanto, Divya Velayudhan, et al.
+- **abs:** https://arxiv.org/abs/2603.12382 · **PDF:** https://arxiv.org/pdf/2603.12382v1
+
+## 5431. NeuroLoRA: Context-Aware Neuromodulation for Parameter-Efficient Multi-Task Adaptation
+
+- **ID:** `2603.12378` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Yuxin Yang, Haoran Zhang, Mingxuan Li, et al.
+- **abs:** https://arxiv.org/abs/2603.12378 · **PDF:** https://arxiv.org/pdf/2603.12378v1
+
+## 5432. The Privacy-Utility Trade-Off of Location Tracking in Ad Personalization
+
+- **ID:** `2603.12374` · **primary:** `econ.EM` · **Submitted:** 2026-03-12
+- **Authors:** Mohammad Mosaffa, Omid Rafieian
+- **abs:** https://arxiv.org/abs/2603.12374 · **PDF:** https://arxiv.org/pdf/2603.12374v1
+
+## 5433. Efficient Reasoning with Balanced Thinking
+
+- **ID:** `2603.12372` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Yulin Li, Tengyao Tu, Li Ding, et al.
+- **abs:** https://arxiv.org/abs/2603.12372 · **PDF:** https://arxiv.org/pdf/2603.12372v3
+
+## 5434. Multi-Step Semantic Reasoning in Generative Retrieval
+
+- **ID:** `2603.12368` · **primary:** `cs.IR` · **Submitted:** 2026-03-12
+- **Authors:** Steven Dong, Yubao Tang, Maarten de Rijke
+- **abs:** https://arxiv.org/abs/2603.12368 · **PDF:** https://arxiv.org/pdf/2603.12368v1
+
+## 5435. Sinkhorn-Drifting Generative Models
+
+- **ID:** `2603.12366` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Ping He, Om Khangaonkar, Hamed Pirsiavash, et al.
+- **abs:** https://arxiv.org/abs/2603.12366 · **PDF:** https://arxiv.org/pdf/2603.12366v2
+
+## 5436. Optimal Experimental Design for Reliable Learning of History-Dependent Constitutive Laws
+
+- **ID:** `2603.12365` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-12
+- **Authors:** Kaushik Bhattacharya, Lianghao Cao, Andrew Stuart
+- **abs:** https://arxiv.org/abs/2603.12365 · **PDF:** https://arxiv.org/pdf/2603.12365v1
+
+## 5437. Adapting Methods for Domain-Specific Japanese Small LMs: Scale, Architecture, and Quantization
+
+- **ID:** `2603.18037` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Takato Yasuno
+- **abs:** https://arxiv.org/abs/2603.18037 · **PDF:** https://arxiv.org/pdf/2603.18037v1
+
+## 5438. Alternating Gradient Flow Utility: A Unified Metric for Structural Pruning and Dynamic Routing in Deep Networks
+
+- **ID:** `2603.12354` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Tianhao Qian, Zhuoxuan Li, Jinde Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.12354 · **PDF:** https://arxiv.org/pdf/2603.12354v2
+
+## 5439. Spatial PDE-aware Selective State-space with Nested Memory for Mobile Traffic Grid Forecasting
+
+- **ID:** `2603.12353` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Zineddine Bettouche, Khalid Ali, Andreas Fischer, et al.
+- **abs:** https://arxiv.org/abs/2603.12353 · **PDF:** https://arxiv.org/pdf/2603.12353v1
+
+## 5440. Probabilistic Joint and Individual Variation Explained (ProJIVE) for Data Integration
+
+- **ID:** `2603.12351` · **primary:** `stat.ML` · **Submitted:** 2026-03-12
+- **Authors:** Raphiel J. Murden, Ganzhong Tian, Deqiang Qiu, et al.
+- **abs:** https://arxiv.org/abs/2603.12351 · **PDF:** https://arxiv.org/pdf/2603.12351v1
+
+## 5441. TASTE-Streaming: Towards Streamable Text-Aligned Speech Tokenization and Embedding for Spoken Language Modeling
+
+- **ID:** `2603.12350` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Liang-Hsuan Tseng, Hung-yi Lee
+- **abs:** https://arxiv.org/abs/2603.12350 · **PDF:** https://arxiv.org/pdf/2603.12350v1
+
+## 5442. Budget-Sensitive Discovery Scoring: A Formally Verified Framework for Evaluating AI-Guided Scientific Selection
+
+- **ID:** `2603.12349` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Abhinaba Basu, Pavan Chakraborty
+- **abs:** https://arxiv.org/abs/2603.12349 · **PDF:** https://arxiv.org/pdf/2603.12349v1
+
+## 5443. Generalist Large Language Models for Molecular Property Prediction: Distilling Knowledge from Specialist Models
+
+- **ID:** `2603.12344` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Khiem Le, Sreejata Dey, Marcos Martínez Galindo, et al.
+- **abs:** https://arxiv.org/abs/2603.12344 · **PDF:** https://arxiv.org/pdf/2603.12344v1
+
+## 5444. LLM-Augmented Therapy Normalization and Aspect-Based Sentiment Analysis for Treatment-Resistant Depression on Reddit
+
+- **ID:** `2603.12343` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Yuxin Zhu, Sahithi Lakamana, Masoud Rouhizadeh, et al.
+- **abs:** https://arxiv.org/abs/2603.12343 · **PDF:** https://arxiv.org/pdf/2603.12343v1
+
+## 5445. Optimizing Task Completion Time Updates Using POMDPs
+
+- **ID:** `2603.12340` · **primary:** `eess.SY` · **Submitted:** 2026-03-12
+- **Authors:** Duncan Eddy, Esen Yel, Emma Passmore, et al.
+- **abs:** https://arxiv.org/abs/2603.12340 · **PDF:** https://arxiv.org/pdf/2603.12340v2
+
+## 5446. Maximum Entropy Exploration Without the Rollouts
+
+- **ID:** `2603.12325` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Jacob Adamczyk, Adam Kamoski, Rahul V. Kulkarni
+- **abs:** https://arxiv.org/abs/2603.12325 · **PDF:** https://arxiv.org/pdf/2603.12325v1
+
+## 5447. Thermodynamics of Reinforcement Learning Curricula
+
+- **ID:** `2603.12324` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Jacob Adamczyk, Juan Sebastian Rojas, Rahul V. Kulkarni
+- **abs:** https://arxiv.org/abs/2603.12324 · **PDF:** https://arxiv.org/pdf/2603.12324v1
+
+## 5448. Pruning-induced phases in fully-connected neural networks: the eumentia, the dementia, and the amentia
+
+- **ID:** `2603.12316` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-03-12
+- **Authors:** Haining Pan, Nakul Aggarwal, J. H. Pixley
+- **abs:** https://arxiv.org/abs/2603.12316 · **PDF:** https://arxiv.org/pdf/2603.12316v1
+
+## 5449. VQQA: An Agentic Approach for Video Evaluation and Quality Improvement
+
+- **ID:** `2603.12310` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Yiwen Song, Tomas Pfister, Yale Song
+- **abs:** https://arxiv.org/abs/2603.12310 · **PDF:** https://arxiv.org/pdf/2603.12310v1
+
+## 5450. The Latent Color Subspace: Emergent Order in High-Dimensional Chaos
+
+- **ID:** `2603.12261` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Mateusz Pach, Jessica Bader, Quentin Bouniot, et al.
+- **abs:** https://arxiv.org/abs/2603.12261 · **PDF:** https://arxiv.org/pdf/2603.12261v1
+
+## 5451. Spatial-TTT: Streaming Visual-based Spatial Intelligence with Test-Time Training
+
+- **ID:** `2603.12255` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Fangfu Liu, Diankun Wu, Jiawei Chi, et al.
+- **abs:** https://arxiv.org/abs/2603.12255 · **PDF:** https://arxiv.org/pdf/2603.12255v1
+
+## 5452. EndoCoT: Scaling Endogenous Chain-of-Thought Reasoning in Diffusion Models
+
+- **ID:** `2603.12252` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Xuanlang Dai, Yujie Zhou, Long Xing, et al.
+- **abs:** https://arxiv.org/abs/2603.12252 · **PDF:** https://arxiv.org/pdf/2603.12252v3
+
+## 5453. SciMDR: Benchmarking and Advancing Scientific Multimodal Document Reasoning
+
+- **ID:** `2603.12249` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Ziyu Chen, Yilun Zhao, Chengye Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.12249 · **PDF:** https://arxiv.org/pdf/2603.12249v1
+
+## 5454. Matching Features, Not Tokens: Energy-Based Fine-Tuning of Language Models
+
+- **ID:** `2603.12248` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Samy Jelassi, Mujin Kwun, Rosie Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.12248 · **PDF:** https://arxiv.org/pdf/2603.12248v2
+
+## 5455. Examining Reasoning LLMs-as-Judges in Non-Verifiable LLM Post-Training
+
+- **ID:** `2603.12246` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Yixin Liu, Yue Yu, DiJia Su, et al.
+- **abs:** https://arxiv.org/abs/2603.12246 · **PDF:** https://arxiv.org/pdf/2603.12246v1
+
+## 5456. Separable neural architectures as a primitive for unified predictive and generative intelligence
+
+- **ID:** `2603.12244` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Reza T. Batley, Apurba Sarker, Rajib Mostakim, et al.
+- **abs:** https://arxiv.org/abs/2603.12244 · **PDF:** https://arxiv.org/pdf/2603.12244v2
+
+## 5457. BiGain: Unified Token Compression for Joint Generation and Classification
+
+- **ID:** `2603.12240` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Jiacheng Liu, Shengkun Tang, Jiacheng Cui, et al.
+- **abs:** https://arxiv.org/abs/2603.12240 · **PDF:** https://arxiv.org/pdf/2603.12240v1
+
+## 5458. STAMP: Selective Task-Aware Mechanism for Text Privacy
+
+- **ID:** `2603.12237` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Fengwei Tian, Payel Bhattacharjee, Heidi Hanson, et al.
+- **abs:** https://arxiv.org/abs/2603.12237 · **PDF:** https://arxiv.org/pdf/2603.12237v1
+
+## 5459. Incremental Neural Network Verification via Learned Conflicts
+
+- **ID:** `2603.12232` · **primary:** `cs.LO` · **Submitted:** 2026-03-12
+- **Authors:** Raya Elsaleh, Liam Davis, Haoze Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.12232 · **PDF:** https://arxiv.org/pdf/2603.12232v1
+
+## 5460. Temporal Straightening for Latent Planning
+
+- **ID:** `2603.12231` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Ying Wang, Oumayma Bounou, Gaoyue Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.12231 · **PDF:** https://arxiv.org/pdf/2603.12231v1
+
+## 5461. Security Considerations for Artificial Intelligence Agents
+
+- **ID:** `2603.12230` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Ninghui Li, Kaiyuan Zhang, Kyle Polley, et al.
+- **abs:** https://arxiv.org/abs/2603.12230 · **PDF:** https://arxiv.org/pdf/2603.12230v1
+
+## 5462. Neural Thickets: Diverse Task Experts Are Dense Around Pretrained Weights
+
+- **ID:** `2603.12228` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Yulu Gan, Phillip Isola
+- **abs:** https://arxiv.org/abs/2603.12228 · **PDF:** https://arxiv.org/pdf/2603.12228v1
+
+## 5463. Interpreting Contrastive Embeddings in Specific Domains with Fuzzy Rules
+
+- **ID:** `2603.12227` · **primary:** `cs.SC` · **Submitted:** 2026-03-12
+- **Authors:** Javier Fumanal-Idocin, Mohammadreza Jamalifard, Javier Andreu-Perez
+- **abs:** https://arxiv.org/abs/2603.12227 · **PDF:** https://arxiv.org/pdf/2603.12227v1
+
+## 5464. Sparking Scientific Creativity via LLM-Driven Interdisciplinary Inspiration
+
+- **ID:** `2603.12226` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Priyanka Kargupta, Shuhaib Mehri, Dilek Hakkani-Tur, et al.
+- **abs:** https://arxiv.org/abs/2603.12226 · **PDF:** https://arxiv.org/pdf/2603.12226v1
+
+## 5465. Portfolio of Solving Strategies in CEGAR-based Object Packing and Scheduling for Sequential 3D Printing
+
+- **ID:** `2603.12224` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Pavel Surynek
+- **abs:** https://arxiv.org/abs/2603.12224 · **PDF:** https://arxiv.org/pdf/2603.12224v1
+
+## 5466. HiAP: A Multi-Granular Stochastic Auto-Pruning Framework for Vision Transformers
+
+- **ID:** `2603.12222` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Andy Li, Aiden Durrant, Milan Markovic, et al.
+- **abs:** https://arxiv.org/abs/2603.12222 · **PDF:** https://arxiv.org/pdf/2603.12222v1
+
+## 5467. Bridging Protocol and Production: Design Patterns for Deploying AI Agents with Model Context Protocol
+
+- **ID:** `2603.13417` · **primary:** `cs.SE` · **Submitted:** 2026-03-12
+- **Authors:** Vasundra Srinivasan
+- **abs:** https://arxiv.org/abs/2603.13417 · **PDF:** https://arxiv.org/pdf/2603.13417v1
+
+## 5468. Quantum-Secure-By-Construction (QSC): A Paradigm Shift For Post-Quantum Agentic Intelligence
+
+- **ID:** `2603.15668` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Arit Kumar Bishwas, Mousumi Sen, Albert Nieto-Morales, et al.
+- **abs:** https://arxiv.org/abs/2603.15668 · **PDF:** https://arxiv.org/pdf/2603.15668v1
+
+## 5469. RDNet: Region Proportion-Aware Dynamic Adaptive Salient Object Detection Network in Optical Remote Sensing Images
+
+- **ID:** `2603.12215` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Bin Wan, Runmin Cong, Xiaofei Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.12215 · **PDF:** https://arxiv.org/pdf/2603.12215v1
+
+## 5470. WORKSWORLD: A Domain for Integrated Numeric Planning and Scheduling of Distributed Pipelined Workflows
+
+- **ID:** `2603.12214` · **primary:** `cs.DC` · **Submitted:** 2026-03-12
+- **Authors:** Taylor Paul, William Regli
+- **abs:** https://arxiv.org/abs/2603.12214 · **PDF:** https://arxiv.org/pdf/2603.12214v2
+
+## 5471. CLASP: Defending Hybrid Large Language Models Against Hidden State Poisoning Attacks
+
+- **ID:** `2603.12206` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Alexandre Le Mercier, Thomas Demeester, Chris Develder
+- **abs:** https://arxiv.org/abs/2603.12206 · **PDF:** https://arxiv.org/pdf/2603.12206v2
+
+## 5472. IndexCache: Accelerating Sparse Attention via Cross-Layer Index Reuse
+
+- **ID:** `2603.12201` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Yushi Bai, Qian Dong, Ting Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.12201 · **PDF:** https://arxiv.org/pdf/2603.12201v1
+
+## 5473. Long-Context Encoder Models for Polish Language Understanding
+
+- **ID:** `2603.12191` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Sławomir Dadas, Rafał Poświata, Marek Kozłowski, et al.
+- **abs:** https://arxiv.org/abs/2603.12191 · **PDF:** https://arxiv.org/pdf/2603.12191v1
+
+## 5474. Compiling Temporal Numeric Planning into Discrete PDDL+: Extended Version
+
+- **ID:** `2603.12188` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Andrea Micheli, Enrico Scala, Alessandro Valentini
+- **abs:** https://arxiv.org/abs/2603.12188 · **PDF:** https://arxiv.org/pdf/2603.12188v1
+
+## 5475. Proof-Carrying Materials: Falsifiable Safety Certificates for Machine-Learned Interatomic Potentials
+
+- **ID:** `2603.12183` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-12
+- **Authors:** Abhinaba Basu, Pavan Chakraborty
+- **abs:** https://arxiv.org/abs/2603.12183 · **PDF:** https://arxiv.org/pdf/2603.12183v2
+
+## 5476. Strategic Navigation or Stochastic Search? How Agents and Humans Reason Over Document Collections
+
+- **ID:** `2603.12180` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Łukasz Borchmann, Jordy Van Landeghem, Michał Turski, et al.
+- **abs:** https://arxiv.org/abs/2603.12180 · **PDF:** https://arxiv.org/pdf/2603.12180v2
+
+## 5477. BehaviorVLM: Unified Finetuning-Free Behavioral Understanding with Vision-Language Reasoning
+
+- **ID:** `2603.12176` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Jingyang Ke, Weihan Li, Amartya Pradhan, et al.
+- **abs:** https://arxiv.org/abs/2603.12176 · **PDF:** https://arxiv.org/pdf/2603.12176v1
+
+## 5478. QAQ: Bidirectional Semantic Coherence for Selecting High-Quality Synthetic Code Instructions
+
+- **ID:** `2603.12165` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Jiayin Lei, Ming Ma, Yunxi Duan, et al.
+- **abs:** https://arxiv.org/abs/2603.12165 · **PDF:** https://arxiv.org/pdf/2603.12165v2
+
+## 5479. A Quantitative Characterization of Forgetting in Post-Training
+
+- **ID:** `2603.12163` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Krishnakumar Balasubramanian, Shiva Prasad Kasiviswanathan
+- **abs:** https://arxiv.org/abs/2603.12163 · **PDF:** https://arxiv.org/pdf/2603.12163v1
+
+## 5480. GlyphBanana: Advancing Precise Text Rendering Through Agentic Workflows
+
+- **ID:** `2603.12155` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Zexuan Yan, Jiarui Jin, Yue Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.12155 · **PDF:** https://arxiv.org/pdf/2603.12155v1
+
+## 5481. LifeSim: Long-Horizon User Life Simulator for Personalized Assistant Evaluation
+
+- **ID:** `2603.12152` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Feiyu Duan, Xuanjing Huang, Zhongyu Wei
+- **abs:** https://arxiv.org/abs/2603.12152 · **PDF:** https://arxiv.org/pdf/2603.12152v1
+
+## 5482. IsoCompute Playbook: Optimally Scaling Sampling Compute for LLM RL
+
+- **ID:** `2603.12151` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Zhoujun Cheng, Yutao Xie, Yuxiao Qu, et al.
+- **abs:** https://arxiv.org/abs/2603.12151 · **PDF:** https://arxiv.org/pdf/2603.12151v1
+
+## 5483. Linking Perception, Confidence and Accuracy in MLLMs
+
+- **ID:** `2603.12149` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Yuetian Du, Yucheng Wang, Rongyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.12149 · **PDF:** https://arxiv.org/pdf/2603.12149v1
+
+## 5484. FlashMotion: Few-Step Controllable Video Generation with Trajectory Guidance
+
+- **ID:** `2603.12146` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Quanhao Li, Zhen Xing, Rui Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.12146 · **PDF:** https://arxiv.org/pdf/2603.12146v1
+
+## 5485. Automatic Generation of High-Performance RL Environments
+
+- **ID:** `2603.12145` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Seth Karten, Rahul Dev Appapogu, Chi Jin
+- **abs:** https://arxiv.org/abs/2603.12145 · **PDF:** https://arxiv.org/pdf/2603.12145v1
+
+## 5486. TopoBench: Benchmarking LLMs on Hard Topological Reasoning
+
+- **ID:** `2603.12133` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Mayug Maniparambil, Nils Hoehing, Janak Kapuriya, et al.
+- **abs:** https://arxiv.org/abs/2603.12133 · **PDF:** https://arxiv.org/pdf/2603.12133v1
+
+## 5487. Increasing intelligence in AI agents can worsen collective outcomes
+
+- **ID:** `2603.12129` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Neil F. Johnson
+- **abs:** https://arxiv.org/abs/2603.12129 · **PDF:** https://arxiv.org/pdf/2603.12129v1
+
+## 5488. Hoi3DGen: Generating High-Quality Human-Object-Interactions in 3D
+
+- **ID:** `2603.12126` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Agniv Sharma, Xianghui Xie, Tom Fischer, et al.
+- **abs:** https://arxiv.org/abs/2603.12126 · **PDF:** https://arxiv.org/pdf/2603.12126v1
+
+## 5489. Cross-Context Review: Improving LLM Output Quality by Separating Production and Review Sessions
+
+- **ID:** `2603.12123` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Tae-Eun Song
+- **abs:** https://arxiv.org/abs/2603.12123 · **PDF:** https://arxiv.org/pdf/2603.12123v1
+
+## 5490. CRAFT: A Tendon-Driven Hand with Hybrid Hard-Soft Compliance
+
+- **ID:** `2603.12120` · **primary:** `cs.RO` · **Submitted:** 2026-03-12
+- **Authors:** Leo Lin, Shivansh Patel, Jay Moon, et al.
+- **abs:** https://arxiv.org/abs/2603.12120 · **PDF:** https://arxiv.org/pdf/2603.12120v1
+
+## 5491. Cornserve: A Distributed Serving System for Any-to-Any Multimodal Models
+
+- **ID:** `2603.12118` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Jae-Won Chung, Jeff J. Ma, Jisang Ahn, et al.
+- **abs:** https://arxiv.org/abs/2603.12118 · **PDF:** https://arxiv.org/pdf/2603.12118v1
+
+## 5492. SommBench: Assessing Sommelier Expertise of Language Models
+
+- **ID:** `2603.12117` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** William Brach, Tomas Bedej, Jacob Nielsen, et al.
+- **abs:** https://arxiv.org/abs/2603.12117 · **PDF:** https://arxiv.org/pdf/2603.12117v1
+
+## 5493. Taming the Adversary: Stable Minimax Deep Deterministic Policy Gradient via Fractional Objectives
+
+- **ID:** `2603.12110` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Taeho Lee, Donghwan Lee
+- **abs:** https://arxiv.org/abs/2603.12110 · **PDF:** https://arxiv.org/pdf/2603.12110v1
+
+## 5494. On Information Self-Locking in Reinforcement Learning for Active Reasoning of LLM agents
+
+- **ID:** `2603.12109` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Deyu Zou, Yongqiang Chen, Fan Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.12109 · **PDF:** https://arxiv.org/pdf/2603.12109v1
+
+## 5495. To Words and Beyond: Probing Large Language Models for Sentence-Level Psycholinguistic Norms of Memorability and Reading Times
+
+- **ID:** `2603.12105` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Thomas Hikaru Clark, Carlos Arriaga, Javier Conde, et al.
+- **abs:** https://arxiv.org/abs/2603.12105 · **PDF:** https://arxiv.org/pdf/2603.12105v1
+
+## 5496. Wasserstein Gradient Flows for Batch Bayesian Optimal Experimental Design
+
+- **ID:** `2603.12102` · **primary:** `stat.ML` · **Submitted:** 2026-03-12
+- **Authors:** Louis Sharrock
+- **abs:** https://arxiv.org/abs/2603.12102 · **PDF:** https://arxiv.org/pdf/2603.12102v1
+
+## 5497. A Robust and Efficient Multi-Agent Reinforcement Learning Framework for Traffic Signal Control
+
+- **ID:** `2603.12096` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Sheng-You Huang, Hsiao-Chuan Chang, Yen-Chi Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.12096 · **PDF:** https://arxiv.org/pdf/2603.12096v1
+
+## 5498. Human-Centred LLM Privacy Audits: Findings and Frictions
+
+- **ID:** `2603.12094` · **primary:** `cs.HC` · **Submitted:** 2026-03-12
+- **Authors:** Dimitri Staufer, Kirsten Morehouse, David Hartmann, et al.
+- **abs:** https://arxiv.org/abs/2603.12094 · **PDF:** https://arxiv.org/pdf/2603.12094v1
+
+## 5499. Resource-Efficient Iterative LLM-Based NAS with Feedback Memory
+
+- **ID:** `2603.12091` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Xiaojie Gu, Dmitry Ignatov, Radu Timofte
+- **abs:** https://arxiv.org/abs/2603.12091 · **PDF:** https://arxiv.org/pdf/2603.12091v1
+
+## 5500. Cross-Domain Policy Optimization via Bellman Consistency and Hybrid Critics
+
+- **ID:** `2603.12087` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Ming-Hong Chen, Kuan-Chen Pan, You-De Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.12087 · **PDF:** https://arxiv.org/pdf/2603.12087v1
+
+## 5501. A Multi-Label Temporal Convolutional Framework for Transcription Factor Binding Characterization
+
+- **ID:** `2603.12073` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Pietro Demurtas, Ferdinando Zanchetta, Giovanni Perini, et al.
+- **abs:** https://arxiv.org/abs/2603.12073 · **PDF:** https://arxiv.org/pdf/2603.12073v1
+
+## 5502. LoV3D: Grounding Cognitive Prognosis Reasoning in Longitudinal 3D Brain MRI via Regional Volume Assessments
+
+- **ID:** `2603.12071` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Zhaoyang Jiang, Zhizhong Fu, David McAllister, et al.
+- **abs:** https://arxiv.org/abs/2603.12071 · **PDF:** https://arxiv.org/pdf/2603.12071v2
+
+## 5503. Beyond Convolution: A Taxonomy of Structured Operators for Learning-Based Image Processing
+
+- **ID:** `2603.12067` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Simone Cammarasana
+- **abs:** https://arxiv.org/abs/2603.12067 · **PDF:** https://arxiv.org/pdf/2603.12067v2
+
+## 5504. Chemical Reaction Networks Learn Better than Spiking Neural Networks
+
+- **ID:** `2603.12060` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Sophie Jaffard, Ivo F. Sbalzarini
+- **abs:** https://arxiv.org/abs/2603.12060 · **PDF:** https://arxiv.org/pdf/2603.12060v1
+
+## 5505. Coarse-Guided Visual Generation via Weighted h-Transform Sampling
+
+- **ID:** `2603.12057` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Yanghao Wang, Ziqi Jiang, Zhen Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.12057 · **PDF:** https://arxiv.org/pdf/2603.12057v2
+
+## 5506. XSkill: Continual Learning from Experience and Skills in Multimodal Agents
+
+- **ID:** `2603.12056` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Guanyu Jiang, Zhaochen Su, Xiaoye Qu, et al.
+- **abs:** https://arxiv.org/abs/2603.12056 · **PDF:** https://arxiv.org/pdf/2603.12056v2
+
+## 5507. Continual Learning with Vision-Language Models via Semantic-Geometry Preservation
+
+- **ID:** `2603.12055` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Chiyuan He, Zihuan Qiu, Fanman Meng, et al.
+- **abs:** https://arxiv.org/abs/2603.12055 · **PDF:** https://arxiv.org/pdf/2603.12055v2
+
+## 5508. Translationese as a Rational Response to Translation Task Difficulty
+
+- **ID:** `2603.12050` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Maria Kunilovskaya
+- **abs:** https://arxiv.org/abs/2603.12050 · **PDF:** https://arxiv.org/pdf/2603.12050v1
+
+## 5509. Slow-Fast Inference: Training-Free Inference Acceleration via Within-Sentence Support Stability
+
+- **ID:** `2603.12038` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Xingyu Xie, Zhaochen Yu, Yue Liao, et al.
+- **abs:** https://arxiv.org/abs/2603.12038 · **PDF:** https://arxiv.org/pdf/2603.12038v1
+
+## 5510. Frequentist Consistency of Prior-Data Fitted Networks for Causal Inference
+
+- **ID:** `2603.12037` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Valentyn Melnychuk, Vahid Balazadeh, Stefan Feuerriegel, et al.
+- **abs:** https://arxiv.org/abs/2603.12037 · **PDF:** https://arxiv.org/pdf/2603.12037v1
+
+## 5511. AGMARL-DKS: An Adaptive Graph-Enhanced Multi-Agent Reinforcement Learning for Dynamic Kubernetes Scheduling
+
+- **ID:** `2603.12031` · **primary:** `cs.DC` · **Submitted:** 2026-03-12
+- **Authors:** Hamed Hamzeh
+- **abs:** https://arxiv.org/abs/2603.12031 · **PDF:** https://arxiv.org/pdf/2603.12031v1
+
+## 5512. Efficient Generative Modeling with Unitary Matrix Product States Using Riemannian Optimization
+
+- **ID:** `2603.12026` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Haotong Duan, Zhongming Chen, Ngai Wong
+- **abs:** https://arxiv.org/abs/2603.12026 · **PDF:** https://arxiv.org/pdf/2603.12026v1
+
+## 5513. Cascade: Composing Software-Hardware Attack Gadgets for Adversarial Threat Amplification in Compound AI Systems
+
+- **ID:** `2603.12023` · **primary:** `cs.CR` · **Submitted:** 2026-03-12
+- **Authors:** Sarbartha Banerjee, Prateek Sahu, Anjo Vahldiek-Oberwagner, et al.
+- **abs:** https://arxiv.org/abs/2603.12023 · **PDF:** https://arxiv.org/pdf/2603.12023v1
+
+## 5514. Neuro-Symbolic Generation and Validation of Memory-Aware Formal Function Specifications
+
+- **ID:** `2603.13414` · **primary:** `cs.SE` · **Submitted:** 2026-03-12
+- **Authors:** Liao Zhang, Tong Chen, Xiwei Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.13414 · **PDF:** https://arxiv.org/pdf/2603.13414v1
+
+## 5515. Just Use XML: Revisiting Joint Translation and Label Projection
+
+- **ID:** `2603.12021` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Thennal D K, Chris Biemann, Hans Ole Hatzel
+- **abs:** https://arxiv.org/abs/2603.12021 · **PDF:** https://arxiv.org/pdf/2603.12021v1
+
+## 5516. Sim-to-reality adaptation for Deep Reinforcement Learning applied to an underwater docking application
+
+- **ID:** `2603.12020` · **primary:** `cs.RO` · **Submitted:** 2026-03-12
+- **Authors:** Alaaeddine Chaarani, Narcis Palomeras, Pere Ridao
+- **abs:** https://arxiv.org/abs/2603.12020 · **PDF:** https://arxiv.org/pdf/2603.12020v1
+
+## 5517. An Intent of Collaboration: On Agencies between Designers and Emerging (Intelligent) Technologies
+
+- **ID:** `2603.12018` · **primary:** `cs.HC` · **Submitted:** 2026-03-12
+- **Authors:** Pei-Ying Lin, Julie Heij, Iris Borst, et al.
+- **abs:** https://arxiv.org/abs/2603.12018 · **PDF:** https://arxiv.org/pdf/2603.12018v1
+
+## 5518. Flowcean - Model Learning for Cyber-Physical Systems
+
+- **ID:** `2603.12015` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Maximilian Schmidt, Swantje Plambeck, Markus Knitt, et al.
+- **abs:** https://arxiv.org/abs/2603.12015 · **PDF:** https://arxiv.org/pdf/2603.12015v1
+
+## 5519. Deep Learning-Based Metamodeling of Nonlinear Stochastic Dynamic Systems under Parametric and Predictive Uncertainty
+
+- **ID:** `2603.12012` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Haimiti Atila, Seymour M. J. Spence
+- **abs:** https://arxiv.org/abs/2603.12012 · **PDF:** https://arxiv.org/pdf/2603.12012v1
+
+## 5520. Can RL Improve Generalization of LLM Agents? An Empirical Study
+
+- **ID:** `2603.12011` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Zhiheng Xi, Xin Guo, Jiaqi Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.12011 · **PDF:** https://arxiv.org/pdf/2603.12011v1
+
+## 5521. Decentralized Orchestration Architecture for Fluid Computing: A Secure Distributed AI Use Case
+
+- **ID:** `2603.12001` · **primary:** `cs.DC` · **Submitted:** 2026-03-12
+- **Authors:** Diego Cajaraville-Aboy, Ana Fernández-Vilas, Rebeca P. Díaz-Redondo, et al.
+- **abs:** https://arxiv.org/abs/2603.12001 · **PDF:** https://arxiv.org/pdf/2603.12001v1
+
+## 5522. Few-for-Many Personalized Federated Learning
+
+- **ID:** `2603.11992` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Ping Guo, Tiantian Zhang, Xi Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.11992 · **PDF:** https://arxiv.org/pdf/2603.11992v1
+
+## 5523. BTZSC: A Benchmark for Zero-Shot Text Classification Across Cross-Encoders, Embedding Models, Rerankers and LLMs
+
+- **ID:** `2603.11991` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Ilias Aarab
+- **abs:** https://arxiv.org/abs/2603.11991 · **PDF:** https://arxiv.org/pdf/2603.11991v1
+
+## 5524. On-Average Stability of Multipass Preconditioned SGD and Effective Dimension
+
+- **ID:** `2603.11989` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Simon Vary, Tyler Farghly, Ilja Kuzborskij, et al.
+- **abs:** https://arxiv.org/abs/2603.11989 · **PDF:** https://arxiv.org/pdf/2603.11989v1
+
+## 5525. LABSHIELD: A Multimodal Benchmark for Safety-Critical Reasoning and Planning in Scientific Laboratories
+
+- **ID:** `2603.11987` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Qianpu Sun, Xiaowei Chi, Yuhan Rui, et al.
+- **abs:** https://arxiv.org/abs/2603.11987 · **PDF:** https://arxiv.org/pdf/2603.11987v1
+
+## 5526. GENIE: Gram-Eigenmode INR Editing with Closed-Form Geometry Updates
+
+- **ID:** `2603.29860` · **primary:** `cs.GR` · **Submitted:** 2026-03-12
+- **Authors:** Samundra Karki, Adarsh Krishnamurthy, Baskar Ganapathysubramanian
+- **abs:** https://arxiv.org/abs/2603.29860 · **PDF:** https://arxiv.org/pdf/2603.29860v1
+
+## 5527. HomeSafe-Bench: Evaluating Vision-Language Models on Unsafe Action Detection for Embodied Agents in Household Scenarios
+
+- **ID:** `2603.11975` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Jiayue Pu, Zhongxiang Sun, Zilu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.11975 · **PDF:** https://arxiv.org/pdf/2603.11975v2
+
+## 5528. Normative Common Ground Replication (NormCoRe): Replication-by-Translation for Studying Norms in Multi-agent AI
+
+- **ID:** `2603.11974` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Luca Deck, Simeon Allmendinger, Lucas Müller, et al.
+- **abs:** https://arxiv.org/abs/2603.11974 · **PDF:** https://arxiv.org/pdf/2603.11974v1
+
+## 5529. Topological DeepONets and a generalization of the Chen-Chen operator approximation theorem
+
+- **ID:** `2603.11972` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Vugar Ismailov
+- **abs:** https://arxiv.org/abs/2603.11972 · **PDF:** https://arxiv.org/pdf/2603.11972v1
+
+## 5530. Multimodal Emotion Recognition via Bi-directional Cross-Attention and Temporal Modeling
+
+- **ID:** `2603.11971` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Junhyeong Byeon, Jeongyeol Kim, Sejoon Lim
+- **abs:** https://arxiv.org/abs/2603.11971 · **PDF:** https://arxiv.org/pdf/2603.11971v2
+
+## 5531. Statistical and structural identifiability in representation learning
+
+- **ID:** `2603.11970` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Walter Nelson, Marco Fumero, Theofanis Karaletsos, et al.
+- **abs:** https://arxiv.org/abs/2603.11970 · **PDF:** https://arxiv.org/pdf/2603.11970v1
+
+## 5532. Uncovering Locally Low-dimensional Structure in Networks by Locally Optimal Spectral Embedding
+
+- **ID:** `2603.11965` · **primary:** `stat.ML` · **Submitted:** 2026-03-12
+- **Authors:** Hannah Sansford, Nick Whiteley, Patrick Rubin-Delanchy
+- **abs:** https://arxiv.org/abs/2603.11965 · **PDF:** https://arxiv.org/pdf/2603.11965v1
+
+## 5533. CHiL(L)Grader: Calibrated Human-in-the-Loop Short-Answer Grading
+
+- **ID:** `2603.11957` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Pranav Raikote, Korbinian Randl, Ioanna Miliou, et al.
+- **abs:** https://arxiv.org/abs/2603.11957 · **PDF:** https://arxiv.org/pdf/2603.11957v1
+
+## 5534. PersonaTrace: Synthesizing Realistic Digital Footprints with LLM Agents
+
+- **ID:** `2603.11955` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Minjia Wang, Yunfeng Wang, Xiao Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.11955 · **PDF:** https://arxiv.org/pdf/2603.11955v1
+
+## 5535. Learning Transferable Sensor Models via Language-Informed Pretraining
+
+- **ID:** `2603.11950` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Yuliang Chen, Arvind Pillai, Yu Yvonne Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.11950 · **PDF:** https://arxiv.org/pdf/2603.11950v1
+
+## 5536. Delayed Backdoor Attacks: Exploring the Temporal Dimension as a New Attack Surface in Pre-Trained Models
+
+- **ID:** `2603.11949` · **primary:** `cs.CR` · **Submitted:** 2026-03-12
+- **Authors:** Zikang Ding, Haomiao Yang, Meng Hao, et al.
+- **abs:** https://arxiv.org/abs/2603.11949 · **PDF:** https://arxiv.org/pdf/2603.11949v1
+
+## 5537. Resurfacing Paralinguistic Awareness in Large Audio Language Models
+
+- **ID:** `2603.11947` · **primary:** `cs.SD` · **Submitted:** 2026-03-12
+- **Authors:** Hao Yang, Minghan Wang, Tongtong Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.11947 · **PDF:** https://arxiv.org/pdf/2603.11947v1
+
+## 5538. Geometry-Aware Probabilistic Circuits via Voronoi Tessellations
+
+- **ID:** `2603.11946` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Sahil Sidheekh, Sriraam Natarajan
+- **abs:** https://arxiv.org/abs/2603.11946 · **PDF:** https://arxiv.org/pdf/2603.11946v1
+
+## 5539. Effective Resistance Rewiring: A Simple Topological Correction for Over-Squashing
+
+- **ID:** `2603.11944` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Bertran Miquel-Oliver, Manel Gil-Sorribes, Victor Guallar, et al.
+- **abs:** https://arxiv.org/abs/2603.11944 · **PDF:** https://arxiv.org/pdf/2603.11944v1
+
+## 5540. Causal Matrix Completion under Multiple Treatments via Mixed Synthetic Nearest Neighbors
+
+- **ID:** `2603.11942` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Minrui Luo, Zhiheng Zhang
+- **abs:** https://arxiv.org/abs/2603.11942 · **PDF:** https://arxiv.org/pdf/2603.11942v1
+
+## 5541. Exhaustive Circuit Mapping of a Single-Cell Foundation Model Reveals Massive Redundancy, Heavy-Tailed Hub Architecture, and Layer-Dependent Differentiation Control
+
+- **ID:** `2603.11940` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Ihor Kendiukhov
+- **abs:** https://arxiv.org/abs/2603.11940 · **PDF:** https://arxiv.org/pdf/2603.11940v1
+
+## 5542. SNAP-V: A RISC-V SoC with Configurable Neuromorphic Acceleration for Small-Scale Spiking Neural Networks
+
+- **ID:** `2603.11939` · **primary:** `cs.AR` · **Submitted:** 2026-03-12
+- **Authors:** Kanishka Gunawardana, Sanka Peeris, Kavishka Rambukwella, et al.
+- **abs:** https://arxiv.org/abs/2603.11939 · **PDF:** https://arxiv.org/pdf/2603.11939v1
+
+## 5543. Prototype-Based Knowledge Guidance for Fine-Grained Structured Radiology Reporting
+
+- **ID:** `2603.11938` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Chantal Pellegrini, Adrian Delchev, Ege Özsoy, et al.
+- **abs:** https://arxiv.org/abs/2603.11938 · **PDF:** https://arxiv.org/pdf/2603.11938v1
+
+## 5544. Fair Learning for Bias Mitigation and Quality Optimization in Paper Recommendation
+
+- **ID:** `2603.11936` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Uttamasha Anjally Oyshi, Susan Gauch
+- **abs:** https://arxiv.org/abs/2603.11936 · **PDF:** https://arxiv.org/pdf/2603.11936v1
+
+## 5545. MobileKernelBench: Can LLMs Write Efficient Kernels for Mobile Devices?
+
+- **ID:** `2603.11935` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Xingze Zou, Jing Wang, Yuhua Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.11935 · **PDF:** https://arxiv.org/pdf/2603.11935v2
+
+## 5546. Chem4DLLM: 4D Multimodal LLMs for Chemical Dynamics Understanding
+
+- **ID:** `2603.11924` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Xinyu Li, Zhen Zhang, Qi Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.11924 · **PDF:** https://arxiv.org/pdf/2603.11924v1
+
+## 5547. CoMMET: To What Extent Can LLMs Perform Theory of Mind Tasks?
+
+- **ID:** `2603.11915` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Ruirui Chen, Weifeng Jiang, Chengwei Qin, et al.
+- **abs:** https://arxiv.org/abs/2603.11915 · **PDF:** https://arxiv.org/pdf/2603.11915v1
+
+## 5548. Understanding LLM Behavior When Encountering User-Supplied Harmful Content in Harmless Tasks
+
+- **ID:** `2603.11914` · **primary:** `cs.CR` · **Submitted:** 2026-03-12
+- **Authors:** Junjie Chu, Yiting Qu, Ye Leng, et al.
+- **abs:** https://arxiv.org/abs/2603.11914 · **PDF:** https://arxiv.org/pdf/2603.11914v1
+
+## 5549. EnTransformer: A Deep Generative Transformer for Multivariate Probabilistic Forecasting
+
+- **ID:** `2603.11909` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Rajdeep Pathak, Rahul Goswami, Madhurima Panja, et al.
+- **abs:** https://arxiv.org/abs/2603.11909 · **PDF:** https://arxiv.org/pdf/2603.11909v1
+
+## 5550. Causal Representation Learning with Optimal Compression under Complex Treatments
+
+- **ID:** `2603.11907` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Wanting Liang, Haoang Chi, Zhiheng Zhang
+- **abs:** https://arxiv.org/abs/2603.11907 · **PDF:** https://arxiv.org/pdf/2603.11907v1
+
+## 5551. FlexRec: Adapting LLM-based Recommenders for Flexible Needs via Reinforcement Learning
+
+- **ID:** `2603.11901` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Yijun Pan, Weikang Qiu, Qiyao Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.11901 · **PDF:** https://arxiv.org/pdf/2603.11901v1
+
+## 5552. Think While Watching: Online Streaming Segment-Level Memory for Multi-Turn Video Reasoning in Multimodal Large Language Models
+
+- **ID:** `2603.11896` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Lu Wang, Zhuoran Jin, Yupu Hao, et al.
+- **abs:** https://arxiv.org/abs/2603.11896 · **PDF:** https://arxiv.org/pdf/2603.11896v1
+
+## 5553. Gaussian Process Regression-based Knowledge Distillation Framework for Simultaneous Prediction of Physical and Mechanical Properties of Epoxy Polymers
+
+- **ID:** `2603.16925` · **primary:** `cond-mat.soft` · **Submitted:** 2026-03-12
+- **Authors:** Sindu B. S., Jan Hamaekers
+- **abs:** https://arxiv.org/abs/2603.16925 · **PDF:** https://arxiv.org/pdf/2603.16925v1
+
+## 5554. Bielik-Minitron-7B: Compressing Large Language Models via Structured Pruning and Knowledge Distillation for the Polish Language
+
+- **ID:** `2603.11881` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Remigiusz Kinas, Paweł Kiszczak, Sergio P. Perez, et al.
+- **abs:** https://arxiv.org/abs/2603.11881 · **PDF:** https://arxiv.org/pdf/2603.11881v1
+
+## 5555. The Mirror Design Pattern: Strict Data Geometry over Model Scale for Prompt Injection Detection
+
+- **ID:** `2603.11875` · **primary:** `cs.CR` · **Submitted:** 2026-03-12
+- **Authors:** J Alex Corll
+- **abs:** https://arxiv.org/abs/2603.11875 · **PDF:** https://arxiv.org/pdf/2603.11875v1
+
+## 5556. An Evolutionary Algorithm with Probabilistic Annealing for Large-scale Sparse Multi-objective Optimization
+
+- **ID:** `2603.11874` · **primary:** `cs.NE` · **Submitted:** 2026-03-12
+- **Authors:** Shuai Shao, Yuhao Sun, Xing Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.11874 · **PDF:** https://arxiv.org/pdf/2603.11874v1
+
+## 5557. AdaFuse: Accelerating Dynamic Adapter Inference via Token-Level Pre-Gating and Fused Kernel Optimization
+
+- **ID:** `2603.11873` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Qiyang Li, Rui Kong, Yuchen Li, et al.
+- **abs:** https://arxiv.org/abs/2603.11873 · **PDF:** https://arxiv.org/pdf/2603.11873v1
+
+## 5558. ELISA: An Interpretable Hybrid Generative AI Agent for Expression-Grounded Discovery in Single-Cell Genomics
+
+- **ID:** `2603.11872` · **primary:** `q-bio.GN` · **Submitted:** 2026-03-12
+- **Authors:** Omar Coser
+- **abs:** https://arxiv.org/abs/2603.11872 · **PDF:** https://arxiv.org/pdf/2603.11872v2
+
+## 5559. On the Role of Reversible Instance Normalization
+
+- **ID:** `2603.11869` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Gaspard Berthelier, Tahar Nabil, Etienne Le Naour, et al.
+- **abs:** https://arxiv.org/abs/2603.11869 · **PDF:** https://arxiv.org/pdf/2603.11869v1
+
+## 5560. Data Fusion with Distributional Equivalence Test-then-pool
+
+- **ID:** `2603.11867` · **primary:** `stat.ME` · **Submitted:** 2026-03-12
+- **Authors:** Linying Yang, Xing Liu, Robin J. Evans
+- **abs:** https://arxiv.org/abs/2603.11867 · **PDF:** https://arxiv.org/pdf/2603.11867v2
+
+## 5561. Social, Legal, Ethical, Empathetic and Cultural Norm Operationalisation for AI Agents
+
+- **ID:** `2603.11864` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Radu Calinescu, Ana Cavalcanti, Marsha Chechik, et al.
+- **abs:** https://arxiv.org/abs/2603.11864 · **PDF:** https://arxiv.org/pdf/2603.11864v1
+
+## 5562. CreativeBench: Benchmarking and Enhancing Machine Creativity via Self-Evolving Challenges
+
+- **ID:** `2603.11863` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Zi-Han Wang, Lam Nguyen, Zhengyang Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.11863 · **PDF:** https://arxiv.org/pdf/2603.11863v1
+
+## 5563. You Told Me to Do It: Measuring Instructional Text-induced Private Data Leakage in LLM Agents
+
+- **ID:** `2603.11862` · **primary:** `cs.CR` · **Submitted:** 2026-03-12
+- **Authors:** Ching-Yu Kao, Xinfeng Li, Shenyu Dai, et al.
+- **abs:** https://arxiv.org/abs/2603.11862 · **PDF:** https://arxiv.org/pdf/2603.11862v1
+
+## 5564. Multi-Station WiFi CSI Sensing Framework Robust to Station-wise Feature Missingness and Limited Labeled Data
+
+- **ID:** `2603.11858` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Keita Kayano, Takayuki Nishio, Daiki Yoda, et al.
+- **abs:** https://arxiv.org/abs/2603.11858 · **PDF:** https://arxiv.org/pdf/2603.11858v2
+
+## 5565. Inverse Neural Operator for ODE Parameter Optimization
+
+- **ID:** `2603.11854` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Zhi-Song Liu, Wenqing Peng, Helmi Toropainen, et al.
+- **abs:** https://arxiv.org/abs/2603.11854 · **PDF:** https://arxiv.org/pdf/2603.11854v1
+
+## 5566. The Landscape of Generative AI in Information Systems: A Synthesis of Secondary Reviews and Research Agendas
+
+- **ID:** `2603.11842` · **primary:** `cs.CY` · **Submitted:** 2026-03-12
+- **Authors:** Aleksander Jarzębowicz, Adam Przybyłek, Jacinto Estima, et al.
+- **abs:** https://arxiv.org/abs/2603.11842 · **PDF:** https://arxiv.org/pdf/2603.11842v1
+
+## 5567. DatedGPT: Preventing Lookahead Bias in Large Language Models with Time-Aware Pretraining
+
+- **ID:** `2603.11838` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Yutong Yan, Raphael Tang, Zhenyu Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.11838 · **PDF:** https://arxiv.org/pdf/2603.11838v1
+
+## 5568. Hypercomplex Widely Linear Processing: Fundamentals for Quaternion Machine Learning
+
+- **ID:** `2603.11835` · **primary:** `stat.ML` · **Submitted:** 2026-03-12
+- **Authors:** Sayed Pouria Talebi, Clive Cheong Took
+- **abs:** https://arxiv.org/abs/2603.11835 · **PDF:** https://arxiv.org/pdf/2603.11835v1
+
+## 5569. Hybrid Human-Agent Social Dilemmas in Energy Markets
+
+- **ID:** `2603.11834` · **primary:** `cs.MA` · **Submitted:** 2026-03-12
+- **Authors:** Isuri Perera, Frits de Nijs, Julian Garcia
+- **abs:** https://arxiv.org/abs/2603.11834 · **PDF:** https://arxiv.org/pdf/2603.11834v1
+
+## 5570. OMNIA: Closing the Loop by Leveraging LLMs for Knowledge Graph Completion
+
+- **ID:** `2603.11820` · **primary:** `cs.DB` · **Submitted:** 2026-03-12
+- **Authors:** Frédéric Ieng, Soror Sahri, Mourad Ouzzani, et al.
+- **abs:** https://arxiv.org/abs/2603.11820 · **PDF:** https://arxiv.org/pdf/2603.11820v1
+
+## 5571. Bridging the Visual-to-Physical Gap: Physically Aligned Representations for Fall Risk Analysis
+
+- **ID:** `2603.13410` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Xianqi Zhang
+- **abs:** https://arxiv.org/abs/2603.13410 · **PDF:** https://arxiv.org/pdf/2603.13410v1
+
+## 5572. Automated Detection of Malignant Lesions in the Ovary Using Deep Learning Models and XAI
+
+- **ID:** `2603.11818` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Md. Hasin Sarwar Ifty, Nisharga Nirjan, Labib Islam, et al.
+- **abs:** https://arxiv.org/abs/2603.11818 · **PDF:** https://arxiv.org/pdf/2603.11818v1
+
+## 5573. VisiFold: Long-Term Traffic Forecasting via Temporal Folding Graph and Node Visibility
+
+- **ID:** `2603.11816` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Zhiwei Zhang, Xinyi Du, Weihao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.11816 · **PDF:** https://arxiv.org/pdf/2603.11816v1
+
+## 5574. RADAR: Closed-Loop Robotic Data Generation via Semantic Planning and Autonomous Causal Environment Reset
+
+- **ID:** `2603.11811` · **primary:** `cs.RO` · **Submitted:** 2026-03-12
+- **Authors:** Yongzhong Wang, Keyu Zhu, Yong Zhong, et al.
+- **abs:** https://arxiv.org/abs/2603.11811 · **PDF:** https://arxiv.org/pdf/2603.11811v1
+
+## 5575. Automating Skill Acquisition through Large-Scale Mining of Open-Source Agentic Repositories: A Framework for Multi-Agent Procedural Knowledge Extraction
+
+- **ID:** `2603.11808` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Shuzhen Bi, Mengsong Wu, Hao Hao, et al.
+- **abs:** https://arxiv.org/abs/2603.11808 · **PDF:** https://arxiv.org/pdf/2603.11808v2
+
+## 5576. OSMDA: OpenStreetMap-based Domain Adaptation for Remote Sensing VLMs
+
+- **ID:** `2603.11804` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Stefan Maria Ailuro, Mario Markov, Mohammad Mahdi, et al.
+- **abs:** https://arxiv.org/abs/2603.11804 · **PDF:** https://arxiv.org/pdf/2603.11804v2
+
+## 5577. A Semi-Decentralized Approach to Multiagent Control
+
+- **ID:** `2603.11802` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Mahdi Al-Husseini, Mykel J. Kochenderfer, Kyle H. Wray
+- **abs:** https://arxiv.org/abs/2603.11802 · **PDF:** https://arxiv.org/pdf/2603.11802v1
+
+## 5578. Exponential-Family Membership Inference: From LiRA and RMIA to BaVarIA
+
+- **ID:** `2603.11799` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Rickard Brännvall
+- **abs:** https://arxiv.org/abs/2603.11799 · **PDF:** https://arxiv.org/pdf/2603.11799v1
+
+## 5579. DocSage: An Information Structuring Agent for Multi-Doc Multi-Entity Question Answering
+
+- **ID:** `2603.11798` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Teng Lin, Yizhang Zhu, Zhengxuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.11798 · **PDF:** https://arxiv.org/pdf/2603.11798v1
+
+## 5580. Locating Demographic Bias at the Attention-Head Level in CLIP's Vision Encoder
+
+- **ID:** `2603.11793` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Alaa Yasser, Kittipat Phunjanna, Marcos Escudero Viñolo, et al.
+- **abs:** https://arxiv.org/abs/2603.11793 · **PDF:** https://arxiv.org/pdf/2603.11793v1
+
+## 5581. Disentangled Representation Learning through Unsupervised Symmetry Group Discovery
+
+- **ID:** `2603.11790` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Barthélémy Dang-Nhu, Louis Annabi, Sylvain Argentieri
+- **abs:** https://arxiv.org/abs/2603.11790 · **PDF:** https://arxiv.org/pdf/2603.11790v2
+
+## 5582. Language Generation with Replay: A Learning-Theoretic View of Model Collapse
+
+- **ID:** `2603.11784` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Giorgio Racca, Michal Valko, Amartya Sanyal
+- **abs:** https://arxiv.org/abs/2603.11784 · **PDF:** https://arxiv.org/pdf/2603.11784v1
+
+## 5583. HELM: Hierarchical and Explicit Label Modeling with Graph Learning for Multi-Label Image Classification
+
+- **ID:** `2603.11783` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Marjan Stoimchev, Boshko Koloski, Jurica Levatić, et al.
+- **abs:** https://arxiv.org/abs/2603.11783 · **PDF:** https://arxiv.org/pdf/2603.11783v1
+
+## 5584. From Debate to Deliberation: Structured Collective Reasoning with Typed Epistemic Acts
+
+- **ID:** `2603.11781` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Sunil Prakash
+- **abs:** https://arxiv.org/abs/2603.11781 · **PDF:** https://arxiv.org/pdf/2603.11781v1
+
+## 5585. Large Language Models for Biomedical Article Classification
+
+- **ID:** `2603.11780` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Jakub Proboszcz, Paweł Cichosz
+- **abs:** https://arxiv.org/abs/2603.11780 · **PDF:** https://arxiv.org/pdf/2603.11780v1
+
+## 5586. Trust Oriented Explainable AI for Fake News Detection
+
+- **ID:** `2603.11778` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Krzysztof Siwek, Daniel Stankowski, Maciej Stodolski
+- **abs:** https://arxiv.org/abs/2603.11778 · **PDF:** https://arxiv.org/pdf/2603.11778v1
+
+## 5587. Legal-DC: Benchmarking Retrieval-Augmented Generation for Legal Documents
+
+- **ID:** `2603.11772` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Yaocong Li, Qiang Lan, Leihan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.11772 · **PDF:** https://arxiv.org/pdf/2603.11772v1
+
+## 5588. An Automatic Text Classification Method Based on Hierarchical Taxonomies, Neural Networks and Document Embedding: The NETHIC Tool
+
+- **ID:** `2603.11770` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Luigi Lomasto, Rosario Di Florio, Andrea Ciapetti, et al.
+- **abs:** https://arxiv.org/abs/2603.11770 · **PDF:** https://arxiv.org/pdf/2603.11770v1
+
+## 5589. Governing Evolving Memory in LLM Agents: Risks, Mechanisms, and the Stability and Safety Governed Memory (SSGM) Framework
+
+- **ID:** `2603.11768` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Chingkwun Lam, Jiaxin Li, Lingfei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.11768 · **PDF:** https://arxiv.org/pdf/2603.11768v1
+
+## 5590. Understanding Wikidata Qualifiers: An Analysis and Taxonomy
+
+- **ID:** `2603.11767` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Gilles Falquet, Sahar Aljalbout
+- **abs:** https://arxiv.org/abs/2603.11767 · **PDF:** https://arxiv.org/pdf/2603.11767v1
+
+## 5591. A Further Efficient Algorithm with Best-of-Both-Worlds Guarantees for $m$-Set Semi-Bandit Problem
+
+- **ID:** `2603.11764` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Botao Chen, Jongyeong Lee, Chansoo Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.11764 · **PDF:** https://arxiv.org/pdf/2603.11764v1
+
+## 5592. Modeling Trial-and-Error Navigation With a Sequential Decision Model of Information Scent
+
+- **ID:** `2603.11759` · **primary:** `cs.HC` · **Submitted:** 2026-03-12
+- **Authors:** Xiaofu Jin, Yunpeng Bai, Antti Oulasvirta
+- **abs:** https://arxiv.org/abs/2603.11759 · **PDF:** https://arxiv.org/pdf/2603.11759v1
+
+## 5593. Exploiting Expertise of Non-Expert and Diverse Agents in Social Bandit Learning: A Free Energy Approach
+
+- **ID:** `2603.11757` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Erfan Mirzaei, Seyed Pooya Shariatpanahi, Alireza Tavakoli, et al.
+- **abs:** https://arxiv.org/abs/2603.11757 · **PDF:** https://arxiv.org/pdf/2603.11757v1
+
+## 5594. Anomaly detection in time-series via inductive biases in the latent space of conditional normalizing flows
+
+- **ID:** `2603.11756` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** David Baumgartner, Eliezer de Souza da Silva, Iñigo Urteaga
+- **abs:** https://arxiv.org/abs/2603.11756 · **PDF:** https://arxiv.org/pdf/2603.11756v1
+
+## 5595. Mitigating the Multiplicity Burden: The Role of Calibration in Reducing Predictive Multiplicity of Classifiers
+
+- **ID:** `2603.11750` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Mustafa Cavus
+- **abs:** https://arxiv.org/abs/2603.11750 · **PDF:** https://arxiv.org/pdf/2603.11750v2
+
+## 5596. Truth as a Compression Artifact in Language Model Training
+
+- **ID:** `2603.11749` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Konstantin Krestnikov
+- **abs:** https://arxiv.org/abs/2603.11749 · **PDF:** https://arxiv.org/pdf/2603.11749v2
+
+## 5597. CINDI: Conditional Imputation and Noisy Data Integrity with Flows in Power Grid Data
+
+- **ID:** `2603.11745` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** David Baumgartner, Helge Langseth, Heri Ramampiaro
+- **abs:** https://arxiv.org/abs/2603.11745 · **PDF:** https://arxiv.org/pdf/2603.11745v1
+
+## 5598. Semi-Synthetic Parallel Data for Translation Quality Estimation: A Case Study of Dataset Building for an Under-Resourced Language Pair
+
+- **ID:** `2603.11743` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Assaf Siani, Anna Kernerman, Ilan Kernerman
+- **abs:** https://arxiv.org/abs/2603.11743 · **PDF:** https://arxiv.org/pdf/2603.11743v1
+
+## 5599. Gender Bias in Generative AI-assisted Recruitment Processes
+
+- **ID:** `2603.11736` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Martina Ullasci, Marco Rondina, Riccardo Coppola, et al.
+- **abs:** https://arxiv.org/abs/2603.11736 · **PDF:** https://arxiv.org/pdf/2603.11736v1
+
+## 5600. Adapting Dijkstra for Buffers and Unlimited Transfers
+
+- **ID:** `2603.11729` · **primary:** `cs.DS` · **Submitted:** 2026-03-12
+- **Authors:** Denys Katkalo, Andrii Rohovyi, Toby Walsh
+- **abs:** https://arxiv.org/abs/2603.11729 · **PDF:** https://arxiv.org/pdf/2603.11729v1
+
+## 5601. Cross-Resolution Attention Network for High-Resolution PM2.5 Prediction
+
+- **ID:** `2603.11725` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Ammar Kheder, Helmi Toropainen, Wenqing Peng, et al.
+- **abs:** https://arxiv.org/abs/2603.11725 · **PDF:** https://arxiv.org/pdf/2603.11725v1
+
+## 5602. When OpenClaw Meets Hospital: Toward an Agentic Operating System for Dynamic Clinical Workflows
+
+- **ID:** `2603.11721` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Wenxian Yang, Hanzheng Qiu, Bangqun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.11721 · **PDF:** https://arxiv.org/pdf/2603.11721v2
+
+## 5603. Affect Decoding in Phonated and Silent Speech Production from Surface EMG
+
+- **ID:** `2603.11715` · **primary:** `eess.AS` · **Submitted:** 2026-03-12
+- **Authors:** Simon Pistrosch, Kleanthis Avramidis, Zhao Ren, et al.
+- **abs:** https://arxiv.org/abs/2603.11715 · **PDF:** https://arxiv.org/pdf/2603.11715v2
+
+## 5604. Scaling Laws for Educational AI Agents
+
+- **ID:** `2603.11709` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Mengsong Wu, Hao Hao, Shuzhen Bi, et al.
+- **abs:** https://arxiv.org/abs/2603.11709 · **PDF:** https://arxiv.org/pdf/2603.11709v1
+
+## 5605. HCP-DCNet: A Hierarchical Causal Primitive Dynamic Composition Network for Self-Improving Causal Understanding
+
+- **ID:** `2603.12305` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Ming Lei, Shufan Wu, Christophe Baehr
+- **abs:** https://arxiv.org/abs/2603.12305 · **PDF:** https://arxiv.org/pdf/2603.12305v1
+
+## 5606. EvoFlows: Evolutionary Edit-Based Flow-Matching for Protein Engineering
+
+- **ID:** `2603.11703` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Nicolas Deutschmann, Constance Ferragu, Jonathan D. Ziegler, et al.
+- **abs:** https://arxiv.org/abs/2603.11703 · **PDF:** https://arxiv.org/pdf/2603.11703v1
+
+## 5607. Decomposing Observational Multiplicity in Decision Trees: Leaf and Structural Regret
+
+- **ID:** `2603.11701` · **primary:** `stat.ML` · **Submitted:** 2026-03-12
+- **Authors:** Mustafa Cavus
+- **abs:** https://arxiv.org/abs/2603.11701 · **PDF:** https://arxiv.org/pdf/2603.11701v1
+
+## 5608. OSCBench: Benchmarking Object State Change in Text-to-Video Generation
+
+- **ID:** `2603.11698` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Xianjing Han, Bin Zhu, Shiqi Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.11698 · **PDF:** https://arxiv.org/pdf/2603.11698v1
+
+## 5609. STAIRS-Former: Spatio-Temporal Attention with Interleaved Recursive Structure Transformer for Offline Multi-task Multi-agent Reinforcement Learning
+
+- **ID:** `2603.11691` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Jiwon Jeon, Myungsik Cho, Youngchul Sung
+- **abs:** https://arxiv.org/abs/2603.11691 · **PDF:** https://arxiv.org/pdf/2603.11691v1
+
+## 5610. Explicit Logic Channel for Validation and Enhancement of MLLMs on Zero-Shot Tasks
+
+- **ID:** `2603.11689` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Mei Chee Leong, Ying Gu, Hui Li Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.11689 · **PDF:** https://arxiv.org/pdf/2603.11689v1
+
+## 5611. SemBench: A Universal Semantic Framework for LLM Evaluation
+
+- **ID:** `2603.11687` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Mikel Zubillaga, Naiara Perez, Oscar Sainz, et al.
+- **abs:** https://arxiv.org/abs/2603.11687 · **PDF:** https://arxiv.org/pdf/2603.11687v2
+
+## 5612. In the LLM era, Word Sense Induction remains unsolved
+
+- **ID:** `2603.11686` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Anna Mosolova, Marie Candito, Carlos Ramisch
+- **abs:** https://arxiv.org/abs/2603.11686 · **PDF:** https://arxiv.org/pdf/2603.11686v1
+
+## 5613. Causal Prosody Mediation for Text-to-Speech:Counterfactual Training of Duration, Pitch, and Energy in FastSpeech2
+
+- **ID:** `2603.11683` · **primary:** `cs.SD` · **Submitted:** 2026-03-12
+- **Authors:** Suvendu Sekhar Mohanty
+- **abs:** https://arxiv.org/abs/2603.11683 · **PDF:** https://arxiv.org/pdf/2603.11683v1
+
+## 5614. Entropy-Preserving Reinforcement Learning
+
+- **ID:** `2603.11682` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Aleksei Petrenko, Ben Lipkin, Kevin Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.11682 · **PDF:** https://arxiv.org/pdf/2603.11682v1
+
+## 5615. LLMs can construct powerful representations and streamline sample-efficient supervised learning
+
+- **ID:** `2603.11679` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Ilker Demirel, Lawrence Shi, Zeshan Hussain, et al.
+- **abs:** https://arxiv.org/abs/2603.11679 · **PDF:** https://arxiv.org/pdf/2603.11679v2
+
+## 5616. From Control to Foresight: Simulation as a New Paradigm for Human-Agent Collaboration
+
+- **ID:** `2603.11677` · **primary:** `cs.HC` · **Submitted:** 2026-03-12
+- **Authors:** Gaole He, Brian Y. Lim
+- **abs:** https://arxiv.org/abs/2603.11677 · **PDF:** https://arxiv.org/pdf/2603.11677v1
+
+## 5617. Stable Spike: Dual Consistency Optimization via Bitwise AND Operations for Spiking Neural Networks
+
+- **ID:** `2603.11676` · **primary:** `cs.NE` · **Submitted:** 2026-03-12
+- **Authors:** Yongqi Ding, Kunshan Yang, Linze Li, et al.
+- **abs:** https://arxiv.org/abs/2603.11676 · **PDF:** https://arxiv.org/pdf/2603.11676v1
+
+## 5618. Context-dependent manifold learning: A neuromodulated constrained autoencoder approach
+
+- **ID:** `2603.11673` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Jérôme Adriaens, Guillaume Drion, Pierre Sacré
+- **abs:** https://arxiv.org/abs/2603.11673 · **PDF:** https://arxiv.org/pdf/2603.11673v1
+
+## 5619. A technology-oriented mapping of the language and translation industry: Analysing stakeholder values and their potential implication for translation pedagogy
+
+- **ID:** `2603.11667` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** María Isabel Rivas Ginel, Janiça Hackenbuchner, Alina Secară, et al.
+- **abs:** https://arxiv.org/abs/2603.11667 · **PDF:** https://arxiv.org/pdf/2603.11667v2
+
+## 5620. Multi-Task Reinforcement Learning for Enhanced Multimodal LLM-as-a-Judge
+
+- **ID:** `2603.11665` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Junjie Wu, Xuan Kan, Zihao He, et al.
+- **abs:** https://arxiv.org/abs/2603.11665 · **PDF:** https://arxiv.org/pdf/2603.11665v1
+
+## 5621. Nuanced Emotion Recognition Based on a Segment-based MLLM Framework Leveraging Qwen3-Omni for AH Detection
+
+- **ID:** `2603.13406` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Liang Tang, Hongda Li, Jiayu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.13406 · **PDF:** https://arxiv.org/pdf/2603.13406v2
+
+## 5622. A Geometrically-Grounded Drive for MDL-Based Optimization in Deep Learning
+
+- **ID:** `2603.12304` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Ming Lei, Shufan Wu, Christophe Baehr
+- **abs:** https://arxiv.org/abs/2603.12304 · **PDF:** https://arxiv.org/pdf/2603.12304v1
+
+## 5623. Simple Recipe Works: Vision-Language-Action Models are Natural Continual Learners with Reinforcement Learning
+
+- **ID:** `2603.11653` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Jiaheng Hu, Jay Shim, Chen Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.11653 · **PDF:** https://arxiv.org/pdf/2603.11653v1
+
+## 5624. QChunker: Learning Question-Aware Text Chunking for Domain RAG via Multi-Agent Debate
+
+- **ID:** `2603.11650` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Jihao Zhao, Daixuan Li, Pengfei Li, et al.
+- **abs:** https://arxiv.org/abs/2603.11650 · **PDF:** https://arxiv.org/pdf/2603.11650v1
+
+## 5625. IDRL: An Individual-Aware Multimodal Depression-Related Representation Learning Framework for Depression Diagnosis
+
+- **ID:** `2603.11644` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Chongxiao Wang, Junjie Liang, Peng Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.11644 · **PDF:** https://arxiv.org/pdf/2603.11644v1
+
+## 5626. Tokenization Allows Multimodal Large Language Models to Understand, Generate and Edit Architectural Floor Plans
+
+- **ID:** `2603.11640` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Sizhong Qin, Ramon Elias Weber, Xinzheng Lu
+- **abs:** https://arxiv.org/abs/2603.11640 · **PDF:** https://arxiv.org/pdf/2603.11640v1
+
+## 5627. VisDoT : Enhancing Visual Reasoning through Human-Like Interpretation Grounding and Decomposition of Thought
+
+- **ID:** `2603.11631` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Eunsoo Lee, Jeongwoo Lee, Minki Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.11631 · **PDF:** https://arxiv.org/pdf/2603.11631v1
+
+## 5628. MedPruner: Training-Free Hierarchical Token Pruning for Efficient 3D Medical Image Understanding in Vision-Language Models
+
+- **ID:** `2603.11625` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Shengyuan Liu, Zanting Ye, Yunrui Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.11625 · **PDF:** https://arxiv.org/pdf/2603.11625v1
+
+## 5629. The Density of Cross-Persistence Diagrams and Its Applications
+
+- **ID:** `2603.11623` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Alexander Mironenko, Evgeny. Burnaev, Serguei Barannikov
+- **abs:** https://arxiv.org/abs/2603.11623 · **PDF:** https://arxiv.org/pdf/2603.11623v1
+
+## 5630. Personalized Federated Learning via Gaussian Generative Modeling
+
+- **ID:** `2603.11620` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Peng Hu, Jianwei Ma
+- **abs:** https://arxiv.org/abs/2603.11620 · **PDF:** https://arxiv.org/pdf/2603.11620v1
+
+## 5631. Taming OpenClaw: Security Analysis and Mitigation of Autonomous LLM Agent Threats
+
+- **ID:** `2603.11619` · **primary:** `cs.CR` · **Submitted:** 2026-03-12
+- **Authors:** Xinhao Deng, Yixiang Zhang, Jiaqing Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.11619 · **PDF:** https://arxiv.org/pdf/2603.11619v1
+
+## 5632. Shape-of-You: Fused Gromov-Wasserstein Optimal Transport for Semantic Correspondence in-the-Wild
+
+- **ID:** `2603.11618` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Jiin Im, Sisung Liu, Je Hyeong Hong
+- **abs:** https://arxiv.org/abs/2603.11618 · **PDF:** https://arxiv.org/pdf/2603.11618v2
+
+## 5633. Schema First Tool APIs for LLM Agents: A Controlled Study of Tool Misuse, Recovery, and Budgeted Performance
+
+- **ID:** `2603.13404` · **primary:** `cs.SE` · **Submitted:** 2026-03-12
+- **Authors:** Akshey Sigdel, Rista Baral
+- **abs:** https://arxiv.org/abs/2603.13404 · **PDF:** https://arxiv.org/pdf/2603.13404v1
+
+## 5634. Fractional Rotation, Full Potential? Investigating Performance and Convergence of Partial RoPE
+
+- **ID:** `2603.11611` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Mohammad Aflah Khan, Krishna P. Gummadi, Manish Gupta, et al.
+- **abs:** https://arxiv.org/abs/2603.11611 · **PDF:** https://arxiv.org/pdf/2603.11611v1
+
+## 5635. ProMAS: Proactive Error Forecasting for Multi-Agent Systems Using Markov Transition Dynamics
+
+- **ID:** `2603.20260` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Xinkui Zhao, Sai Liu, Yifan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.20260 · **PDF:** https://arxiv.org/pdf/2603.20260v1
+
+## 5636. AutoScout: Structured Optimization for Automating ML System Configuration
+
+- **ID:** `2603.11603` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Jimmy Shong, Yuhan Ding, Yihan Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.11603 · **PDF:** https://arxiv.org/pdf/2603.11603v1
+
+## 5637. See, Symbolize, Act: Grounding VLMs with Spatial Representations for Better Gameplay
+
+- **ID:** `2603.11601` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Ashish Baghel, Paras Chopra
+- **abs:** https://arxiv.org/abs/2603.11601 · **PDF:** https://arxiv.org/pdf/2603.11601v2
+
+## 5638. Hybrid Energy-Aware Reward Shaping: A Unified Lightweight Physics-Guided Methodology for Policy Optimization
+
+- **ID:** `2603.11600` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Qijun Liao, Jue Yang, Yiting Kang, et al.
+- **abs:** https://arxiv.org/abs/2603.11600 · **PDF:** https://arxiv.org/pdf/2603.11600v1
+
+## 5639. Survival Meets Classification: A Novel Framework for Early Risk Prediction Models of Chronic Diseases
+
+- **ID:** `2603.11598` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Shaheer Ahmad Khan, Muhammad Usamah Shahid, Muddassar Farooq
+- **abs:** https://arxiv.org/abs/2603.11598 · **PDF:** https://arxiv.org/pdf/2603.11598v1
+
+## 5640. Performance Evaluation of Open-Source Large Language Models for Assisting Pathology Report Writing in Japanese
+
+- **ID:** `2603.11597` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Masataka Kawai, Singo Sakashita, Shumpei Ishikawa, et al.
+- **abs:** https://arxiv.org/abs/2603.11597 · **PDF:** https://arxiv.org/pdf/2603.11597v1
+
+## 5641. Leveraging Large Language Models and Survival Analysis for Early Prediction of Chemotherapy Outcomes
+
+- **ID:** `2603.11594` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Muhammad Faisal Shahid, Asad Afzal, Abdullah Faiz, et al.
+- **abs:** https://arxiv.org/abs/2603.11594 · **PDF:** https://arxiv.org/pdf/2603.11594v1
+
+## 5642. Toward Complex-Valued Neural Networks for Waveform Generation
+
+- **ID:** `2603.11589` · **primary:** `cs.SD` · **Submitted:** 2026-03-12
+- **Authors:** Hyung-Seok Oh, Deok-Hyeon Cho, Seung-Bin Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.11589 · **PDF:** https://arxiv.org/pdf/2603.11589v1
+
+## 5643. UtilityMax Prompting: A Formal Framework for Multi-Objective Large Language Model Optimization
+
+- **ID:** `2603.11583` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Ofir Marom
+- **abs:** https://arxiv.org/abs/2603.11583 · **PDF:** https://arxiv.org/pdf/2603.11583v2
+
+## 5644. Streaming Translation and Transcription Through Speech-to-Text Causal Alignment
+
+- **ID:** `2603.11578` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Roman Koshkin, Jeon Haesung, Lianbo Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.11578 · **PDF:** https://arxiv.org/pdf/2603.11578v1
+
+## 5645. CAETC: Causal Autoencoding and Treatment Conditioning for Counterfactual Estimation over Time
+
+- **ID:** `2603.11565` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Nghia D. Nguyen, Pablo Robles-Granda, Lav R. Varshney
+- **abs:** https://arxiv.org/abs/2603.11565 · **PDF:** https://arxiv.org/pdf/2603.11565v1
+
+## 5646. Where Matters More Than What: Decoding-aligned KV Cache Compression via Position-aware Pseudo Queries
+
+- **ID:** `2603.11564` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Zhenxu Tian, Yi Su, Juntao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.11564 · **PDF:** https://arxiv.org/pdf/2603.11564v1
+
+## 5647. Feedback-Coupled Memory Systems: A Dynamical Model for Adaptive Coordination
+
+- **ID:** `2603.11560` · **primary:** `cs.MA` · **Submitted:** 2026-03-12
+- **Authors:** Stefano Grassi
+- **abs:** https://arxiv.org/abs/2603.11560 · **PDF:** https://arxiv.org/pdf/2603.11560v3
+
+## 5648. AI Knows What's Wrong But Cannot Fix It: Helicoid Dynamics in Frontier LLMs Under High-Stakes Decisions
+
+- **ID:** `2603.11559` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Alejandro R Jadad
+- **abs:** https://arxiv.org/abs/2603.11559 · **PDF:** https://arxiv.org/pdf/2603.11559v1
+
+## 5649. RoboClaw: An Agentic Framework for Scalable Long-Horizon Robotic Tasks
+
+- **ID:** `2603.11558` · **primary:** `cs.RO` · **Submitted:** 2026-03-12
+- **Authors:** Ruiying Li, Yunlang Zhou, YuYao Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.11558 · **PDF:** https://arxiv.org/pdf/2603.11558v3
+
+## 5650. MANSION: Multi-floor lANguage-to-3D Scene generatIOn for loNg-horizon tasks
+
+- **ID:** `2603.11554` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Lirong Che, Shuo Wen, Shan Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.11554 · **PDF:** https://arxiv.org/pdf/2603.11554v1
+
+## 5651. Multi-Task Anti-Causal Learning for Reconstructing Urban Events from Residents' Reports
+
+- **ID:** `2603.11546` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Liangkai Zhou, Susu Xu, Shuqi Zhong, et al.
+- **abs:** https://arxiv.org/abs/2603.11546 · **PDF:** https://arxiv.org/pdf/2603.11546v1
+
+## 5652. One Supervisor, Many Modalities: Adaptive Tool Orchestration for Autonomous Queries
+
+- **ID:** `2603.11545` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Mayank Saini, Arit Kumar Bishwas
+- **abs:** https://arxiv.org/abs/2603.11545 · **PDF:** https://arxiv.org/pdf/2603.11545v2
+
+## 5653. ReHARK: Refined Hybrid Adaptive RBF Kernels for Robust One-Shot Vision-Language Adaptation
+
+- **ID:** `2603.11542` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Md Jahidul Islam
+- **abs:** https://arxiv.org/abs/2603.11542 · **PDF:** https://arxiv.org/pdf/2603.11542v1
+
+## 5654. Quantum mechanical framework for quantization-based optimization: from Gradient flow to Schroedinger equation
+
+- **ID:** `2603.11536` · **primary:** `quant-ph` · **Submitted:** 2026-03-12
+- **Authors:** Jinwuk Seok, Changsik Cho
+- **abs:** https://arxiv.org/abs/2603.11536 · **PDF:** https://arxiv.org/pdf/2603.11536v1
+
+## 5655. Expert Threshold Routing for Autoregressive Language Modeling with Dynamic Computation Allocation and Load Balancing
+
+- **ID:** `2603.11535` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Hanchi Sun, Yixin Liu, Yonghui Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.11535 · **PDF:** https://arxiv.org/pdf/2603.11535v1
+
+## 5656. Simultaneous estimation of multiple discrete unimodal distributions under stochastic order constraints
+
+- **ID:** `2603.11532` · **primary:** `math.OC` · **Submitted:** 2026-03-12
+- **Authors:** Yasuhiro Yoshida, Noriyoshi Sukegawa, Jiro Iwanaga
+- **abs:** https://arxiv.org/abs/2603.11532 · **PDF:** https://arxiv.org/pdf/2603.11532v1
+
+## 5657. CFD-HAR: User-controllable Privacy through Conditional Feature Disentanglement
+
+- **ID:** `2603.11526` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Alex Gn, Fan Li, S Kuniyilh, et al.
+- **abs:** https://arxiv.org/abs/2603.11526 · **PDF:** https://arxiv.org/pdf/2603.11526v1
+
+## 5658. MST-Direct: Matching via Sinkhorn Transport for Multivariate Geostatistical Simulation with Complex Non-Linear Dependencies
+
+- **ID:** `2603.18036` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Tchalies Bachmann Schmitz
+- **abs:** https://arxiv.org/abs/2603.18036 · **PDF:** https://arxiv.org/pdf/2603.18036v1
+
+## 5659. EReCu: Pseudo-label Evolution Fusion and Refinement with Multi-Cue Learning for Unsupervised Camouflage Detection
+
+- **ID:** `2603.11521` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Shuo Jiang, Gaojia Zhang, Min Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.11521 · **PDF:** https://arxiv.org/pdf/2603.11521v1
+
+## 5660. FBCIR: Balancing Cross-Modal Focuses in Composed Image Retrieval
+
+- **ID:** `2603.11520` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Chenchen Zhao, Jianhuan Zhuo, Muxi Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.11520 · **PDF:** https://arxiv.org/pdf/2603.11520v1
+
+## 5661. Multi-Agent Collaboration for Automated Design Exploration on High Performance Computing Systems
+
+- **ID:** `2603.11515` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Harshitha Menon, Charles F. Jekel, Kevin Korner, et al.
+- **abs:** https://arxiv.org/abs/2603.11515 · **PDF:** https://arxiv.org/pdf/2603.11515v1
+
+## 5662. Can Small Language Models Use What They Retrieve? An Empirical Study of Retrieval Utilization Across Model Scale
+
+- **ID:** `2603.11513` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Sanchit Pandey
+- **abs:** https://arxiv.org/abs/2603.11513 · **PDF:** https://arxiv.org/pdf/2603.11513v1
+
+## 5663. Tiny Aya: Bridging Scale and Multilingual Depth
+
+- **ID:** `2603.11510` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Alejandro R. Salamanca, Diana Abagyan, Daniel D'souza, et al.
+- **abs:** https://arxiv.org/abs/2603.11510 · **PDF:** https://arxiv.org/pdf/2603.11510v1
+
+## 5664. Gen-Fab: A Variation-Aware Generative Model for Predicting Fabrication Variations in Nanophotonic Devices
+
+- **ID:** `2603.11505` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Rambod Azimi, Yuri Grinberg, Dan-Xia Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.11505 · **PDF:** https://arxiv.org/pdf/2603.11505v1
+
+## 5665. LongFlow: Efficient KV Cache Compression for Reasoning M
+
+- **ID:** `2603.11504` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Yi Su, Zhenxu Tian, Dan Qiao, et al.
+- **abs:** https://arxiv.org/abs/2603.11504 · **PDF:** https://arxiv.org/pdf/2603.11504v1
+
+## 5666. Global Evolutionary Steering: Refining Activation Steering Control via Cross-Layer Consistency
+
+- **ID:** `2603.12298` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Xinyan Jiang, Wenjing Yu, Di Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.12298 · **PDF:** https://arxiv.org/pdf/2603.12298v1
+
+## 5667. Sharpness-Aware Minimization for Generalized Embedding Learning in Federated Recommendation
+
+- **ID:** `2603.11503` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Fengyuan Yu, Xiaohua Feng, Yuyuan Li, et al.
+- **abs:** https://arxiv.org/abs/2603.11503 · **PDF:** https://arxiv.org/pdf/2603.11503v1
+
+## 5668. KEPo: Knowledge Evolution Poison on Graph-based Retrieval-Augmented Generation
+
+- **ID:** `2603.11501` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Qizhi Chen, Chao Qi, Yihong Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.11501 · **PDF:** https://arxiv.org/pdf/2603.11501v2
+
+## 5669. Try, Check and Retry: A Divide-and-Conquer Framework for Boosting Long-context Tool-Calling Performance of LLMs
+
+- **ID:** `2603.11495` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Kunfeng Chen, Qihuang Zhong, Juhua Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.11495 · **PDF:** https://arxiv.org/pdf/2603.11495v1
+
+## 5670. OrthoEraser: Coupled-Neuron Orthogonal Projection for Concept Erasure
+
+- **ID:** `2603.11493` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Chuancheng Shi, Wenhua Wu, Fei Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.11493 · **PDF:** https://arxiv.org/pdf/2603.11493v1
+
+## 5671. SPEGC: Continual Test-Time Adaptation via Semantic-Prompt-Enhanced Graph Clustering for Medical Image Segmentation
+
+- **ID:** `2603.11492` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Xiaogang Du, Jiawei Zhang, Tongfei Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.11492 · **PDF:** https://arxiv.org/pdf/2603.11492v1
+
+## 5672. Attention Sinks Are Provably Necessary in Softmax Transformers: Evidence from Trigger-Conditional Tasks
+
+- **ID:** `2603.11487` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Yuval Ran-Milo
+- **abs:** https://arxiv.org/abs/2603.11487 · **PDF:** https://arxiv.org/pdf/2603.11487v3
+
+## 5673. AnimeScore: A Preference-Based Dataset and Framework for Evaluating Anime-Like Speech Style
+
+- **ID:** `2603.11482` · **primary:** `cs.SD` · **Submitted:** 2026-03-12
+- **Authors:** Joonyong Park, Jerry Li
+- **abs:** https://arxiv.org/abs/2603.11482 · **PDF:** https://arxiv.org/pdf/2603.11482v1
+
+## 5674. INFACT: A Diagnostic Benchmark for Induced Faithfulness and Factuality Hallucinations in Video-LLMs
+
+- **ID:** `2603.11481` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Junqi Yang, Yuecong Min, Jie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.11481 · **PDF:** https://arxiv.org/pdf/2603.11481v1
+
+## 5675. Grammar of the Wave: Towards Explainable Multivariate Time Series Event Detection via Neuro-Symbolic VLM Agents
+
+- **ID:** `2603.11479` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Sky Chenwei Wan, Tianjun Hou, Yifei Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.11479 · **PDF:** https://arxiv.org/pdf/2603.11479v1
+
+## 5676. Leveraging Phytolith Research using Artificial Intelligence
+
+- **ID:** `2603.11476` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Andrés G. Mejía Ramón, Kate Dudgeon, Nina Witteveen, et al.
+- **abs:** https://arxiv.org/abs/2603.11476 · **PDF:** https://arxiv.org/pdf/2603.11476v1
+
+## 5677. Deep Learning Network-Temporal Models For Traffic Prediction
+
+- **ID:** `2603.11475` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Yufeng Xin, Ethan Fan
+- **abs:** https://arxiv.org/abs/2603.11475 · **PDF:** https://arxiv.org/pdf/2603.11475v1
+
+## 5678. Slack More, Predict Better: Proximal Relaxation for Probabilistic Latent Variable Model-based Soft Sensors
+
+- **ID:** `2603.11473` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Zehua Zou, Yiran Ma, Yulong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.11473 · **PDF:** https://arxiv.org/pdf/2603.11473v2
+
+## 5679. HawkesRank: Event-Driven Centrality for Real-Time Importance Ranking
+
+- **ID:** `2603.11472` · **primary:** `cs.SI` · **Submitted:** 2026-03-12
+- **Authors:** Didier Sornette, Yishan Luo, Sandro Claudio Lera
+- **abs:** https://arxiv.org/abs/2603.11472 · **PDF:** https://arxiv.org/pdf/2603.11472v1
+
+## 5680. Stage-Adaptive Reliability Modeling for Continuous Valence-Arousal Estimation
+
+- **ID:** `2603.11468` · **primary:** `cs.MM` · **Submitted:** 2026-03-12
+- **Authors:** Yubeen Lee, Sangeun Lee, Junyeop Cha, et al.
+- **abs:** https://arxiv.org/abs/2603.11468 · **PDF:** https://arxiv.org/pdf/2603.11468v1
+
+## 5681. Bridging Discrete Marks and Continuous Dynamics: Dual-Path Cross-Interaction for Marked Temporal Point Processes
+
+- **ID:** `2603.11462` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Yuxiang Liu, Qiao Liu, Tong Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.11462 · **PDF:** https://arxiv.org/pdf/2603.11462v1
+
+## 5682. UniHetCO: A Unified Heterogeneous Representation for Multi-Problem Learning in Unsupervised Neural Combinatorial Optimization
+
+- **ID:** `2603.11456` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Kien X. Nguyen, Ilya Safro
+- **abs:** https://arxiv.org/abs/2603.11456 · **PDF:** https://arxiv.org/pdf/2603.11456v1
+
+## 5683. Examining Users' Behavioural Intention to Use OpenClaw Through the Cognition--Affect--Conation Framework
+
+- **ID:** `2603.11455` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Yiran Du
+- **abs:** https://arxiv.org/abs/2603.11455 · **PDF:** https://arxiv.org/pdf/2603.11455v2
+
+## 5684. A Dynamic Survey of Fuzzy, Intuitionistic Fuzzy, Neutrosophic, Plithogenic, and Extensional Sets
+
+- **ID:** `2603.15667` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Takaaki Fujita, Florentin Smarandache
+- **abs:** https://arxiv.org/abs/2603.15667 · **PDF:** https://arxiv.org/pdf/2603.15667v1
+
+## 5685. LLM-Assisted Causal Structure Disambiguation and Factor Extraction for Legal Judgment Prediction
+
+- **ID:** `2603.11446` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Yuzhi Liang, Lixiang Ma, Xinrong Zhu
+- **abs:** https://arxiv.org/abs/2603.11446 · **PDF:** https://arxiv.org/pdf/2603.11446v1
+
+## 5686. Verified Multi-Agent Orchestration: A Plan-Execute-Verify-Replan Framework for Complex Query Resolution
+
+- **ID:** `2603.11445` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Xing Zhang, Yanwei Cui, Guanghui Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.11445 · **PDF:** https://arxiv.org/pdf/2603.11445v2
+
+## 5687. GPT4o-Receipt: A Dataset and Human Study for AI-Generated Document Forensics
+
+- **ID:** `2603.11442` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Yan Zhang, Simiao Ren, Ankit Raj, et al.
+- **abs:** https://arxiv.org/abs/2603.11442 · **PDF:** https://arxiv.org/pdf/2603.11442v2
+
+## 5688. ZTab: Domain-based Zero-shot Annotation for Table Columns
+
+- **ID:** `2603.11436` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Ehsan Hoseinzade, Ke Wang
+- **abs:** https://arxiv.org/abs/2603.11436 · **PDF:** https://arxiv.org/pdf/2603.11436v1
+
+## 5689. Adversarial Reinforcement Learning for Detecting False Data Injection Attacks in Vehicular Routing
+
+- **ID:** `2603.11433` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Taha Eghtesad, Yevgeniy Vorobeychik, Aron Laszka
+- **abs:** https://arxiv.org/abs/2603.11433 · **PDF:** https://arxiv.org/pdf/2603.11433v1
+
+## 5690. Compiled Memory: Not More Information, but More Precise Instructions for Language Agents
+
+- **ID:** `2603.15666` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** James Rhodes, George Kang
+- **abs:** https://arxiv.org/abs/2603.15666 · **PDF:** https://arxiv.org/pdf/2603.15666v1
+
+## 5691. A Stable Neural Statistical Dependence Estimator for Autoencoder Feature Analysis
+
+- **ID:** `2603.11428` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Bo Hu, Jose C Principe
+- **abs:** https://arxiv.org/abs/2603.11428 · **PDF:** https://arxiv.org/pdf/2603.11428v2
+
+## 5692. Zero-Shot Cross-City Generalization in End-to-End Autonomous Driving: Self-Supervised versus Supervised Representations
+
+- **ID:** `2603.11417` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Fatemeh Naeinian, Ali Hamza, Haoran Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.11417 · **PDF:** https://arxiv.org/pdf/2603.11417v1
+
+## 5693. BLooP: Zero-Shot Abstractive Summarization using Large Language Models with Bigram Lookahead Promotion
+
+- **ID:** `2603.11415` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Varun Iyer, Cornelia Caragea
+- **abs:** https://arxiv.org/abs/2603.11415 · **PDF:** https://arxiv.org/pdf/2603.11415v1
+
+## 5694. MaterialFigBENCH: benchmark dataset with figures for evaluating college-level materials science problem-solving abilities of multimodal large language models
+
+- **ID:** `2603.11414` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Michiko Yoshitake, Yuta Suzuki, Ryo Igarashi, et al.
+- **abs:** https://arxiv.org/abs/2603.11414 · **PDF:** https://arxiv.org/pdf/2603.11414v1
+
+## 5695. Evaluation format, not model capability, drives triage failure in the assessment of consumer health AI
+
+- **ID:** `2603.11413` · **primary:** `cs.HC` · **Submitted:** 2026-03-12
+- **Authors:** David Fraile Navarro, Farah Magrabi, Enrico Coiera
+- **abs:** https://arxiv.org/abs/2603.11413 · **PDF:** https://arxiv.org/pdf/2603.11413v3
+
+## 5696. Algorithmic Consequences of Particle Filters for Sentence Processing: Amplified Garden-Paths and Digging-In Effects
+
+- **ID:** `2603.11412` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Amani Maina-Kilaas, Roger Levy
+- **abs:** https://arxiv.org/abs/2603.11412 · **PDF:** https://arxiv.org/pdf/2603.11412v2
+
+## 5697. Contextual Graph Representations for Task-Driven 3D Perception and Planning
+
+- **ID:** `2603.26685` · **primary:** `cs.RO` · **Submitted:** 2026-03-12
+- **Authors:** Christopher Agia
+- **abs:** https://arxiv.org/abs/2603.26685 · **PDF:** https://arxiv.org/pdf/2603.26685v1
+
+## 5698. Speak or Stay Silent: Context-Aware Turn-Taking in Multi-Party Dialogue
+
+- **ID:** `2603.11409` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Kratika Bhagtani, Mrinal Anand, Yu Chen Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.11409 · **PDF:** https://arxiv.org/pdf/2603.11409v1
+
+## 5699. Beyond Polarity: Multi-Dimensional LLM Sentiment Signals for WTI Crude Oil Futures Return Prediction
+
+- **ID:** `2603.11408` · **primary:** `q-fin.ST` · **Submitted:** 2026-03-12
+- **Authors:** Dehao Dai, Ding Ma, Dou Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.11408 · **PDF:** https://arxiv.org/pdf/2603.11408v2
+
+## 5700. Deployment-Time Reliability of Learned Robot Policies
+
+- **ID:** `2603.11400` · **primary:** `cs.RO` · **Submitted:** 2026-03-12
+- **Authors:** Christopher Agia
+- **abs:** https://arxiv.org/abs/2603.11400 · **PDF:** https://arxiv.org/pdf/2603.11400v1
+
+## 5701. Entropy Guided Diversification and Preference Elicitation in Agentic Recommendation Systems
+
+- **ID:** `2603.11399` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Dat Tran, Yongce Li, Hannah Clay, et al.
+- **abs:** https://arxiv.org/abs/2603.11399 · **PDF:** https://arxiv.org/pdf/2603.11399v1
+
+## 5702. Efficient Cross-View Localization in 6G Space-Air-Ground Integrated Network
+
+- **ID:** `2603.11398` · **primary:** `cs.NI` · **Submitted:** 2026-03-12
+- **Authors:** Min Hao, Yanbing Xu, Maoqiang Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.11398 · **PDF:** https://arxiv.org/pdf/2603.11398v1
+
+## 5703. Event-Driven Video Generation
+
+- **ID:** `2603.13402` · **primary:** `cs.CV` · **Submitted:** 2026-03-12
+- **Authors:** Chika Maduabuchi
+- **abs:** https://arxiv.org/abs/2603.13402 · **PDF:** https://arxiv.org/pdf/2603.13402v2
+
+## 5704. Harnessing Data Asymmetry: Manifold Learning in the Finsler World
+
+- **ID:** `2603.11396` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Thomas Dagès, Simon Weber, Daniel Cremers, et al.
+- **abs:** https://arxiv.org/abs/2603.11396 · **PDF:** https://arxiv.org/pdf/2603.11396v1
+
+## 5705. ARROW: Augmented Replay for RObust World models
+
+- **ID:** `2603.11395` · **primary:** `cs.LG` · **Submitted:** 2026-03-12
+- **Authors:** Abdulaziz Alyahya, Abdallah Al Siyabi, Markus R. Ernst, et al.
+- **abs:** https://arxiv.org/abs/2603.11395 · **PDF:** https://arxiv.org/pdf/2603.11395v1
+
+## 5706. Stop Listening to Me! How Multi-turn Conversations Can Degrade Diagnostic Reasoning
+
+- **ID:** `2603.11394` · **primary:** `cs.CL` · **Submitted:** 2026-03-12
+- **Authors:** Kevin H. Guo, Chao Yan, Avinash Baidya, et al.
+- **abs:** https://arxiv.org/abs/2603.11394 · **PDF:** https://arxiv.org/pdf/2603.11394v1
+
+## 5707. Agentic AI for Embodied-enhanced Beam Prediction in Low-Altitude Economy Networks
+
+- **ID:** `2603.11392` · **primary:** `cs.NI` · **Submitted:** 2026-03-12
+- **Authors:** Min Hao, Zhizhuo Li, Zirui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.11392 · **PDF:** https://arxiv.org/pdf/2603.11392v1
+
+## 5708. Deactivating Refusal Triggers: Understanding and Mitigating Overrefusal in Safety Alignment
+
+- **ID:** `2603.11388` · **primary:** `cs.AI` · **Submitted:** 2026-03-12
+- **Authors:** Zhiyu Xue, Zimo Qi, Guangliang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.11388 · **PDF:** https://arxiv.org/pdf/2603.11388v1
+
+## 5709. Ghost Framing Theory: Exploring the role of generative AI in new venture rhetorical legitimation
+
+- **ID:** `2603.11384` · **primary:** `cs.HC` · **Submitted:** 2026-03-11
+- **Authors:** Greg Nyilasy
+- **abs:** https://arxiv.org/abs/2603.11384 · **PDF:** https://arxiv.org/pdf/2603.11384v1
+
+## 5710. Vision-Based Hand Shadowing for Robotic Manipulation via Inverse Kinematics
+
+- **ID:** `2603.11383` · **primary:** `cs.RO` · **Submitted:** 2026-03-11
+- **Authors:** Hendrik Chiche, Antoine Jamme, Trevor Rigoberto Martinez
+- **abs:** https://arxiv.org/abs/2603.11383 · **PDF:** https://arxiv.org/pdf/2603.11383v1
+
+## 5711. Detecting Intrinsic and Instrumental Self-Preservation in Autonomous Agents: The Unified Continuation-Interest Protocol
+
+- **ID:** `2603.11382` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Christopher Altman
+- **abs:** https://arxiv.org/abs/2603.11382 · **PDF:** https://arxiv.org/pdf/2603.11382v4
+
+## 5712. Continued Pretraining for Low-Resource Swahili ASR: Achieving State-of-the-Art Performance with Minimal Labeled Data
+
+- **ID:** `2603.11378` · **primary:** `cs.SD` · **Submitted:** 2026-03-11
+- **Authors:** Hillary Mutisya, John Mugane
+- **abs:** https://arxiv.org/abs/2603.11378 · **PDF:** https://arxiv.org/pdf/2603.11378v1
+
+## 5713. How do AI agents talk about science and research? An exploration of scientific discussions on Moltbook using BERTopic
+
+- **ID:** `2603.11375` · **primary:** `cs.SI` · **Submitted:** 2026-03-11
+- **Authors:** Oliver Wieczorek
+- **abs:** https://arxiv.org/abs/2603.11375 · **PDF:** https://arxiv.org/pdf/2603.11375v1
+
+## 5714. Ensuring Safety in Automated Mechanical Ventilation through Offline Reinforcement Learning and Digital Twin Verification
+
+- **ID:** `2603.11372` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Hang Yu, Huidong Liu, Qingchen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.11372 · **PDF:** https://arxiv.org/pdf/2603.11372v1
+
+## 5715. Relaxed Efficient Acquisition of Context and Temporal Features
+
+- **ID:** `2603.11370` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Yunni Qu, Dzung Dinh, Grant King, et al.
+- **abs:** https://arxiv.org/abs/2603.11370 · **PDF:** https://arxiv.org/pdf/2603.11370v3
+
+## 5716. abx_amr_simulator: A simulation environment for antibiotic prescribing policy optimization under antimicrobial resistance
+
+- **ID:** `2603.11369` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Joyce Lee, Seth Blumberg
+- **abs:** https://arxiv.org/abs/2603.11369 · **PDF:** https://arxiv.org/pdf/2603.11369v1
+
+## 5717. Spatially Robust Inference with Predicted and Missing at Random Labels
+
+- **ID:** `2603.11368` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** Stephen Salerno, Zhenke Wu, Tyler McCormick
+- **abs:** https://arxiv.org/abs/2603.11368 · **PDF:** https://arxiv.org/pdf/2603.11368v1
+
+## 5718. Multilingual Financial Fraud Detection Using Machine Learning and Transformer Models: A Bangla-English Study
+
+- **ID:** `2603.11358` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Mohammad Shihab Uddin, Md Hasibul Amin, Nusrat Jahan Ema, et al.
+- **abs:** https://arxiv.org/abs/2603.11358 · **PDF:** https://arxiv.org/pdf/2603.11358v1
+
+## 5719. Resolving Java Code Repository Issues with iSWE Agent
+
+- **ID:** `2603.11356` · **primary:** `cs.SE` · **Submitted:** 2026-03-11
+- **Authors:** Jatin Ganhotra, Sami Serhan, Antonio Abu Nassar, et al.
+- **abs:** https://arxiv.org/abs/2603.11356 · **PDF:** https://arxiv.org/pdf/2603.11356v1
+
+## 5720. Teleodynamic Learning a new Paradigm For Interpretable AI
+
+- **ID:** `2603.11355` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Enrique ter Horst, Juan Diego Zambrano
+- **abs:** https://arxiv.org/abs/2603.11355 · **PDF:** https://arxiv.org/pdf/2603.11355v1
+
+## 5721. The Artificial Self: Characterising the landscape of AI identity
+
+- **ID:** `2603.11353` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Raymond Douglas, Jan Kulveit, Ondrej Havlicek, et al.
+- **abs:** https://arxiv.org/abs/2603.11353 · **PDF:** https://arxiv.org/pdf/2603.11353v1
+
+## 5722. TimeSqueeze: Dynamic Patching for Efficient Time Series Forecasting
+
+- **ID:** `2603.11352` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Sravan Kumar Ankireddy, Nikita Seleznev, Nam H. Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.11352 · **PDF:** https://arxiv.org/pdf/2603.11352v1
+
+## 5723. Novelty Adaptation Through Hybrid Large Language Model (LLM)-Symbolic Planning and LLM-guided Reinforcement Learning
+
+- **ID:** `2603.11351` · **primary:** `cs.RO` · **Submitted:** 2026-03-11
+- **Authors:** Hong Lu, Pierrick Lorang, Timothy R. Duggan, et al.
+- **abs:** https://arxiv.org/abs/2603.11351 · **PDF:** https://arxiv.org/pdf/2603.11351v1
+
+## 5724. Evaluating Explainable AI Attribution Methods in Neural Machine Translation via Attention-Guided Knowledge Distillation
+
+- **ID:** `2603.11342` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Aria Nourbakhsh, Salima Lamsiyah, Adelaide Danilov, et al.
+- **abs:** https://arxiv.org/abs/2603.11342 · **PDF:** https://arxiv.org/pdf/2603.11342v1
+
+## 5725. Improving LLM Performance Through Black-Box Online Tuning: A Case for Adding System Specs to Factsheets for Trusted AI
+
+- **ID:** `2603.11340` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Yonas Atinafu, Henry Lin, Robin Cohen
+- **abs:** https://arxiv.org/abs/2603.11340 · **PDF:** https://arxiv.org/pdf/2603.11340v1
+
+## 5726. FinRule-Bench: A Benchmark for Joint Reasoning over Financial Tables and Principles
+
+- **ID:** `2603.11339` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Arun Vignesh Malarkkan, Manan Roy Choudhury, Guangwei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.11339 · **PDF:** https://arxiv.org/pdf/2603.11339v1
+
+## 5727. SimulU: Training-free Policy for Long-form Simultaneous Speech-to-Speech Translation
+
+- **ID:** `2603.16924` · **primary:** `eess.AS` · **Submitted:** 2026-03-11
+- **Authors:** Amirbek Djanibekov, Luisa Bentivogli, Matteo Negri, et al.
+- **abs:** https://arxiv.org/abs/2603.16924 · **PDF:** https://arxiv.org/pdf/2603.16924v1
+
+## 5728. RewardHackingAgents: Benchmarking Evaluation Integrity for LLM ML-Engineering Agents
+
+- **ID:** `2603.11337` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Yonas Atinafu, Robin Cohen
+- **abs:** https://arxiv.org/abs/2603.11337 · **PDF:** https://arxiv.org/pdf/2603.11337v1
+
+## 5729. MAD: Microenvironment-Aware Distillation -- A Pretraining Strategy for Virtual Spatial Omics from Microscopy
+
+- **ID:** `2603.13401` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Jiashu Han, Kunzan Liu, Yeojin Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.13401 · **PDF:** https://arxiv.org/pdf/2603.13401v1
+
+## 5730. LLM-Augmented Digital Twin for Policy Evaluation in Short-Video Platforms
+
+- **ID:** `2603.11333` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Haoting Zhang, Yunduan Lin, Jinghai He, et al.
+- **abs:** https://arxiv.org/abs/2603.11333 · **PDF:** https://arxiv.org/pdf/2603.11333v1
+
+## 5731. On the Computational Hardness of Transformers
+
+- **ID:** `2603.11332` · **primary:** `cs.CC` · **Submitted:** 2026-03-11
+- **Authors:** Barna Saha, Yinzhan Xu, Christopher Ye, et al.
+- **abs:** https://arxiv.org/abs/2603.11332 · **PDF:** https://arxiv.org/pdf/2603.11332v1
+
+## 5732. Jailbreak Scaling Laws for Large Language Models: Polynomial-Exponential Crossover
+
+- **ID:** `2603.11331` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Indranil Halder, Annesya Banerjee, Cengiz Pehlevan
+- **abs:** https://arxiv.org/abs/2603.11331 · **PDF:** https://arxiv.org/pdf/2603.11331v1
+
+## 5733. Ill-Conditioning in Dictionary-Based Dynamic-Equation Learning: A Systems Biology Case Study
+
+- **ID:** `2603.11330` · **primary:** `q-bio.QM` · **Submitted:** 2026-03-11
+- **Authors:** Yuxiang Feng, Niall M Mangan, Manu Jayadharan
+- **abs:** https://arxiv.org/abs/2603.11330 · **PDF:** https://arxiv.org/pdf/2603.11330v1
+
+## 5734. Meta-Reinforcement Learning with Self-Reflection for Agentic Search
+
+- **ID:** `2603.11327` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Teng Xiao, Yige Yuan, Hamish Ivison, et al.
+- **abs:** https://arxiv.org/abs/2603.11327 · **PDF:** https://arxiv.org/pdf/2603.11327v2
+
+## 5735. Hindsight-Anchored Policy Optimization: Turning Failure into Feedback in Sparse Reward Settings
+
+- **ID:** `2603.11321` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Yuning Wu, Ke Wang, Devin Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.11321 · **PDF:** https://arxiv.org/pdf/2603.11321v1
+
+## 5736. On the Robustness of Langevin Dynamics to Score Function Error
+
+- **ID:** `2603.11319` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Daniel Yiming Cao, August Y. Chen, Karthik Sridharan, et al.
+- **abs:** https://arxiv.org/abs/2603.11319 · **PDF:** https://arxiv.org/pdf/2603.11319v1
+
+## 5737. MRI2Qmap: multi-parametric quantitative mapping with MRI-driven denoising priors
+
+- **ID:** `2603.11316` · **primary:** `physics.med-ph` · **Submitted:** 2026-03-11
+- **Authors:** Mohammad Golbabaee, Matteo Cencini, Carolin Pirkl, et al.
+- **abs:** https://arxiv.org/abs/2603.11316 · **PDF:** https://arxiv.org/pdf/2603.11316v1
+
+## 5738. Heavy-Tailed Principle Component Analysis
+
+- **ID:** `2603.11308` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Mario Sayde, Christopher Khater, Jihad Fahs, et al.
+- **abs:** https://arxiv.org/abs/2603.11308 · **PDF:** https://arxiv.org/pdf/2603.11308v1
+
+## 5739. Client-Conditional Federated Learning via Local Training Data Statistics
+
+- **ID:** `2603.11307` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Rickard Brännvall
+- **abs:** https://arxiv.org/abs/2603.11307 · **PDF:** https://arxiv.org/pdf/2603.11307v1
+
+## 5740. Worst-case low-rank approximations
+
+- **ID:** `2603.11304` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** Anya Fries, Markus Reichstein, David Blei, et al.
+- **abs:** https://arxiv.org/abs/2603.11304 · **PDF:** https://arxiv.org/pdf/2603.11304v1
+
+## 5741. Counterweights and Complementarities: The Convergence of AI and Blockchain Powering a Decentralized Future
+
+- **ID:** `2603.11299` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Yibai Li, Zhiye Jin, Xiaobing, et al.
+- **abs:** https://arxiv.org/abs/2603.11299 · **PDF:** https://arxiv.org/pdf/2603.11299v1
+
+## 5742. Single molecule localization microscopy challenge: a biologically inspired benchmark for long-sequence modeling
+
+- **ID:** `2603.11296` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Fatemeh Valeh, Monika Farsang, Radu Grosu, et al.
+- **abs:** https://arxiv.org/abs/2603.11296 · **PDF:** https://arxiv.org/pdf/2603.11296v1
+
+## 5743. Temporal Text Classification with Large Language Models
+
+- **ID:** `2603.11295` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Nishat Raihan, Marcos Zampieri
+- **abs:** https://arxiv.org/abs/2603.11295 · **PDF:** https://arxiv.org/pdf/2603.11295v1
+
+## 5744. Synthetic Data Generation for Brain-Computer Interfaces: Overview, Benchmarking, and Future Directions
+
+- **ID:** `2603.12296` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Ziwei Wang, Zhentao He, Xingyi He, et al.
+- **abs:** https://arxiv.org/abs/2603.12296 · **PDF:** https://arxiv.org/pdf/2603.12296v1
+
+## 5745. Outrigger local polynomial regression
+
+- **ID:** `2603.11282` · **primary:** `stat.ME` · **Submitted:** 2026-03-11
+- **Authors:** Elliot H. Young, Rajen D. Shah, Richard J. Samworth
+- **abs:** https://arxiv.org/abs/2603.11282 · **PDF:** https://arxiv.org/pdf/2603.11282v1
+
+## 5746. ThReadMed-QA: A Multi-Turn Medical Dialogue Benchmark from Real Patient Questions
+
+- **ID:** `2603.11281` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Monica Munnangi, Saiph Savage
+- **abs:** https://arxiv.org/abs/2603.11281 · **PDF:** https://arxiv.org/pdf/2603.11281v1
+
+## 5747. AI Psychometrics: Evaluating the Psychological Reasoning of Large Language Models with Psychometric Validities
+
+- **ID:** `2603.11279` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Yibai Li, Xiaolin Lin, Zhenghui Sha, et al.
+- **abs:** https://arxiv.org/abs/2603.11279 · **PDF:** https://arxiv.org/pdf/2603.11279v1
+
+## 5748. COMPASS: The explainable agentic framework for Sovereignty, Sustainability, Compliance, and Ethics
+
+- **ID:** `2603.11277` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Jean-Sébastien Dessureault, Alain-Thierry Iliho Manzi, Soukaina Alaoui Ismaili, et al.
+- **abs:** https://arxiv.org/abs/2603.11277 · **PDF:** https://arxiv.org/pdf/2603.11277v2
+
+## 5749. RIE-Greedy: Regularization-Induced Exploration for Contextual Bandits
+
+- **ID:** `2603.11276` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** Tong Li, Thiago de Queiroz Casanova, Eric M. Schwartz, et al.
+- **abs:** https://arxiv.org/abs/2603.11276 · **PDF:** https://arxiv.org/pdf/2603.11276v1
+
+## 5750. "I followed what felt right, not what I was told": Autonomy, Coaching, and Recognizing Bias Through AI-Mediated Dialogue
+
+- **ID:** `2603.11274` · **primary:** `cs.HC` · **Submitted:** 2026-03-11
+- **Authors:** Atieh Taheri, Hamza El Alaoui, Patrick Carrington, et al.
+- **abs:** https://arxiv.org/abs/2603.11274 · **PDF:** https://arxiv.org/pdf/2603.11274v1
+
+## 5751. Duration Aware Scheduling for ASR Serving Under Workload Drift
+
+- **ID:** `2603.11273` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Darshan Makwana, Yash Jogi, Harsh Kotta, et al.
+- **abs:** https://arxiv.org/abs/2603.11273 · **PDF:** https://arxiv.org/pdf/2603.11273v1
+
+## 5752. Hybrid Intent-Aware Personalization with Machine Learning and RAG-Enabled Large Language Models for Financial Services Marketing
+
+- **ID:** `2603.14173` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Akhil Chandra Shanivendra
+- **abs:** https://arxiv.org/abs/2603.14173 · **PDF:** https://arxiv.org/pdf/2603.14173v1
+
+## 5753. Citation-Enforced RAG for Fiscal Document Intelligence: Cited, Explainable Knowledge Retrieval in Tax Compliance
+
+- **ID:** `2603.14170` · **primary:** `cs.IR` · **Submitted:** 2026-03-11
+- **Authors:** Akhil Chandra Shanivendra
+- **abs:** https://arxiv.org/abs/2603.14170 · **PDF:** https://arxiv.org/pdf/2603.14170v1
+
+## 5754. Beyond the Class Subspace: Teacher-Guided Training for Reliable Out-of-Distribution Detection in Single-Domain Models
+
+- **ID:** `2603.11269` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Hong Yang, Devroop Kar, Qi Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.11269 · **PDF:** https://arxiv.org/pdf/2603.11269v1
+
+## 5755. The Unlearning Mirage: A Dynamic Framework for Evaluating LLM Unlearning
+
+- **ID:** `2603.11266` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Raj Sanjay Shah, Jing Huang, Keerthiram Murugesan, et al.
+- **abs:** https://arxiv.org/abs/2603.11266 · **PDF:** https://arxiv.org/pdf/2603.11266v1
+
+## 5756. Artificial Intelligence for Sentiment Analysis of Persian Poetry
+
+- **ID:** `2603.11254` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Arash Zargar, Abolfazl Moshiri, Mitra Shafaei, et al.
+- **abs:** https://arxiv.org/abs/2603.11254 · **PDF:** https://arxiv.org/pdf/2603.11254v1
+
+## 5757. LLMs Can Infer Political Alignment from Online Conversations
+
+- **ID:** `2603.11253` · **primary:** `cs.SI` · **Submitted:** 2026-03-11
+- **Authors:** Byunghwee Lee, Sangyeon Kim, Filippo Menczer, et al.
+- **abs:** https://arxiv.org/abs/2603.11253 · **PDF:** https://arxiv.org/pdf/2603.11253v2
+
+## 5758. A Machine Learning-Enhanced Hopf-Cole Formulation for Nonlinear Gas Flow in Porous Media
+
+- **ID:** `2603.11250` · **primary:** `math.NA` · **Submitted:** 2026-03-11
+- **Authors:** V. S. Maduru, K. B. Nakshatrala
+- **abs:** https://arxiv.org/abs/2603.11250 · **PDF:** https://arxiv.org/pdf/2603.11250v1
+
+## 5759. Differentiable Thermodynamic Phase-Equilibria for Machine Learning
+
+- **ID:** `2603.11249` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Karim K. Ben Hicham, Moreno Ascani, Jan G. Rittig, et al.
+- **abs:** https://arxiv.org/abs/2603.11249 · **PDF:** https://arxiv.org/pdf/2603.11249v2
+
+## 5760. Mind the Sim2Real Gap in User Simulation for Agentic Tasks
+
+- **ID:** `2603.11245` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Xuhui Zhou, Weiwei Sun, Qianou Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.11245 · **PDF:** https://arxiv.org/pdf/2603.11245v1
+
+## 5761. A Standardized Framework For Evaluating Gene Expression Generative Models
+
+- **ID:** `2603.11244` · **primary:** `q-bio.GN` · **Submitted:** 2026-03-11
+- **Authors:** Andrea Rubbi, Andrea Giuseppe Di Francesco, Mohammad Lotfollahi, et al.
+- **abs:** https://arxiv.org/abs/2603.11244 · **PDF:** https://arxiv.org/pdf/2603.11244v1
+
+## 5762. A Unified Latent Space Disentanglement VAE Framework with Robust Disentanglement Effectiveness Evaluation
+
+- **ID:** `2603.11242` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** Xiaoan Lang, Fang Liu
+- **abs:** https://arxiv.org/abs/2603.11242 · **PDF:** https://arxiv.org/pdf/2603.11242v1
+
+## 5763. Cough activity detection for automatic tuberculosis screening
+
+- **ID:** `2603.11241` · **primary:** `eess.AS` · **Submitted:** 2026-03-11
+- **Authors:** Joshua Jansen van Vüren, Devendra Singh Parihar, Daphne Naidoo, et al.
+- **abs:** https://arxiv.org/abs/2603.11241 · **PDF:** https://arxiv.org/pdf/2603.11241v1
+
+## 5764. Reversible Lifelong Model Editing via Semantic Routing-Based LoRA
+
+- **ID:** `2603.11239` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Haihua Luo, Xuming Ran, Tommi Kärkkäinen, et al.
+- **abs:** https://arxiv.org/abs/2603.11239 · **PDF:** https://arxiv.org/pdf/2603.11239v2
+
+## 5765. Monitoring and Prediction of Mood in Elderly People during Daily Life Activities
+
+- **ID:** `2603.11230` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Daniel Bautista-Salinas, Joaquín Roca González, Inmaculada Méndez, et al.
+- **abs:** https://arxiv.org/abs/2603.11230 · **PDF:** https://arxiv.org/pdf/2603.11230v1
+
+## 5766. Trustworthy predictive distributions for rare events via diagnostic transport maps
+
+- **ID:** `2603.11229` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** Elizabeth Cucuzzella, Rafael Izbicki, Ann B. Lee
+- **abs:** https://arxiv.org/abs/2603.11229 · **PDF:** https://arxiv.org/pdf/2603.11229v1
+
+## 5767. Markovian Generation Chains in Large Language Models
+
+- **ID:** `2603.11228` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Mingmeng Geng, Amr Mohamed, Guokan Shang, et al.
+- **abs:** https://arxiv.org/abs/2603.11228 · **PDF:** https://arxiv.org/pdf/2603.11228v1
+
+## 5768. MDER-DR: Multi-Hop Question Answering with Entity-Centric Summaries
+
+- **ID:** `2603.11223` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Riccardo Campi, Nicolò Oreste Pinciroli Vago, Mathyas Giudici, et al.
+- **abs:** https://arxiv.org/abs/2603.11223 · **PDF:** https://arxiv.org/pdf/2603.11223v1
+
+## 5769. Frequency-Modulated Visual Restoration for Matryoshka Large Multimodal Models
+
+- **ID:** `2603.11220` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Qingtao Pan, Zhihao Dou, Shuo Li
+- **abs:** https://arxiv.org/abs/2603.11220 · **PDF:** https://arxiv.org/pdf/2603.11220v1
+
+## 5770. Measuring AI Agents' Progress on Multi-Step Cyber Attack Scenarios
+
+- **ID:** `2603.11214` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Linus Folkerts, Will Payne, Simon Inman, et al.
+- **abs:** https://arxiv.org/abs/2603.11214 · **PDF:** https://arxiv.org/pdf/2603.11214v3
+
+## 5771. Security-by-Design for LLM-Based Code Generation: Leveraging Internal Representations for Concept-Driven Steering Mechanisms
+
+- **ID:** `2603.11212` · **primary:** `cs.CR` · **Submitted:** 2026-03-11
+- **Authors:** Maximilian Wendlinger, Daniel Kowatsch, Konstantin Böttinger, et al.
+- **abs:** https://arxiv.org/abs/2603.11212 · **PDF:** https://arxiv.org/pdf/2603.11212v1
+
+## 5772. A Simple Efficiency Incremental Learning Framework via Vision-Language Model with Nonlinear Multi-Adapters
+
+- **ID:** `2603.11211` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Haihua Luo, Xuming Ran, Jiangrong Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.11211 · **PDF:** https://arxiv.org/pdf/2603.11211v2
+
+## 5773. Reference-Guided Machine Unlearning
+
+- **ID:** `2603.11210` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Jonas Mirlach, Sonia Laguna, Julia E. Vogt
+- **abs:** https://arxiv.org/abs/2603.11210 · **PDF:** https://arxiv.org/pdf/2603.11210v1
+
+## 5774. The Deep-Match Framework for Event-Related Potential Detection in EEG
+
+- **ID:** `2603.20258` · **primary:** `eess.SP` · **Submitted:** 2026-03-11
+- **Authors:** Marek Zylinski, Bartosz Tomasz Smigielski, Gerard Cybulski
+- **abs:** https://arxiv.org/abs/2603.20258 · **PDF:** https://arxiv.org/pdf/2603.20258v1
+
+## 5775. Representation Finetuning for Continual Learning
+
+- **ID:** `2603.11201` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Haihua Luo, Xuming Ran, Tommi Kärkkäinen, et al.
+- **abs:** https://arxiv.org/abs/2603.11201 · **PDF:** https://arxiv.org/pdf/2603.11201v2
+
+## 5776. DNS-GT: A Graph-based Transformer Approach to Learn Embeddings of Domain Names from DNS Queries
+
+- **ID:** `2603.11200` · **primary:** `cs.CR` · **Submitted:** 2026-03-11
+- **Authors:** Massimiliano Altieri, Ronan Hamon, Roberto Corizzo, et al.
+- **abs:** https://arxiv.org/abs/2603.11200 · **PDF:** https://arxiv.org/pdf/2603.11200v1
+
+## 5777. Bayesian Optimization of Partially Known Systems using Hybrid Models
+
+- **ID:** `2603.11199` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Eike Cramer, Luis Kutschat, Oliver Stollenwerk, et al.
+- **abs:** https://arxiv.org/abs/2603.11199 · **PDF:** https://arxiv.org/pdf/2603.11199v1
+
+## 5778. DeReason: A Difficulty-Aware Curriculum Improves Decoupled SFT-then-RL Training for General Reasoning
+
+- **ID:** `2603.11193` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Hanxu Hu, Yuxuan Wang, Maggie Huan, et al.
+- **abs:** https://arxiv.org/abs/2603.11193 · **PDF:** https://arxiv.org/pdf/2603.11193v1
+
+## 5779. PACED: Distillation and Self-Distillation at the Frontier of Student Competence
+
+- **ID:** `2603.11178` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Yuanda Xu, Hejian Sang, Zhengze Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.11178 · **PDF:** https://arxiv.org/pdf/2603.11178v2
+
+## 5780. Huntington Disease Automatic Speech Recognition with Biomarker Supervision
+
+- **ID:** `2603.11168` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Charles L. Wang, Cady Chen, Ziwei Gong, et al.
+- **abs:** https://arxiv.org/abs/2603.11168 · **PDF:** https://arxiv.org/pdf/2603.11168v1
+
+## 5781. Learning to Unscramble: Simplifying Symbolic Expressions via Self-Supervised Oracle Trajectories
+
+- **ID:** `2603.11164` · **primary:** `hep-th` · **Submitted:** 2026-03-11
+- **Authors:** David Shih
+- **abs:** https://arxiv.org/abs/2603.11164 · **PDF:** https://arxiv.org/pdf/2603.11164v1
+
+## 5782. Algorithmic Capture, Computational Complexity, and Inductive Bias of Infinite Transformers
+
+- **ID:** `2603.11161` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Orit Davidovich, Zohar Ringel
+- **abs:** https://arxiv.org/abs/2603.11161 · **PDF:** https://arxiv.org/pdf/2603.11161v1
+
+## 5783. COMIC: Agentic Sketch Comedy Generation
+
+- **ID:** `2603.11048` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Susung Hong, Brian Curless, Ira Kemelmacher-Shlizerman, et al.
+- **abs:** https://arxiv.org/abs/2603.11048 · **PDF:** https://arxiv.org/pdf/2603.11048v1
+
+## 5784. LiTo: Surface Light Field Tokenization
+
+- **ID:** `2603.11047` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Jen-Hao Rick Chang, Xiaoming Zhao, Dorian Chan, et al.
+- **abs:** https://arxiv.org/abs/2603.11047 · **PDF:** https://arxiv.org/pdf/2603.11047v1
+
+## 5785. Neural Field Thermal Tomography: A Differentiable Physics Framework for Non-Destructive Evaluation
+
+- **ID:** `2603.11045` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Tao Zhong, Yixun Hu, Dongzhe Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.11045 · **PDF:** https://arxiv.org/pdf/2603.11045v1
+
+## 5786. V2M-Zero: Zero-Pair Time-Aligned Video-to-Music Generation
+
+- **ID:** `2603.11042` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Yan-Bo Lin, Jonah Casebeer, Long Mai, et al.
+- **abs:** https://arxiv.org/abs/2603.11042 · **PDF:** https://arxiv.org/pdf/2603.11042v1
+
+## 5787. Instruction set for the representation of graphs
+
+- **ID:** `2603.11039` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Ezequiel Lopez-Rubio, Mario Pascual-Gonzalez
+- **abs:** https://arxiv.org/abs/2603.11039 · **PDF:** https://arxiv.org/pdf/2603.11039v1
+
+## 5788. Beyond the Illusion of Consensus: From Surface Heuristics to Knowledge-Grounded Evaluation in LLM-as-a-Judge
+
+- **ID:** `2603.11027` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Mingyang Song, Mao Zheng, Chenning Xu
+- **abs:** https://arxiv.org/abs/2603.11027 · **PDF:** https://arxiv.org/pdf/2603.11027v1
+
+## 5789. Does AI See like Art Historians? Interpreting How Vision Language Models Recognize Artistic Style
+
+- **ID:** `2603.11024` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Marvin Limpijankit, Milad Alshomary, Yassin Oulad Daoud, et al.
+- **abs:** https://arxiv.org/abs/2603.11024 · **PDF:** https://arxiv.org/pdf/2603.11024v1
+
+## 5790. Leech Lattice Vector Quantization for Efficient LLM Compression
+
+- **ID:** `2603.11021` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Tycho F. A. van der Ouderaa, Mart van Baalen, Paul Whatmough, et al.
+- **abs:** https://arxiv.org/abs/2603.11021 · **PDF:** https://arxiv.org/pdf/2603.11021v1
+
+## 5791. TIPS: Turn-Level Information-Potential Reward Shaping for Search-Augmented LLMs
+
+- **ID:** `2603.22293` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Yutao Xie, Nathaniel Thomas, Nicklas Hansen, et al.
+- **abs:** https://arxiv.org/abs/2603.22293 · **PDF:** https://arxiv.org/pdf/2603.22293v1
+
+## 5792. Systematic Scaling Analysis of Jailbreak Attacks in Large Language Models
+
+- **ID:** `2603.11149` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Xiangwen Wang, Ananth Balashankar, Varun Chandrasekaran
+- **abs:** https://arxiv.org/abs/2603.11149 · **PDF:** https://arxiv.org/pdf/2603.11149v2
+
+## 5793. A Systematic Study of Pseudo-Relevance Feedback with LLMs
+
+- **ID:** `2603.11008` · **primary:** `cs.IR` · **Submitted:** 2026-03-11
+- **Authors:** Nour Jedidi, Jimmy Lin
+- **abs:** https://arxiv.org/abs/2603.11008 · **PDF:** https://arxiv.org/pdf/2603.11008v1
+
+## 5794. Catalogue Grounded Multimodal Attribution for Museum Video under Resource and Regulatory Constraints
+
+- **ID:** `2603.11147` · **primary:** `cs.MM` · **Submitted:** 2026-03-11
+- **Authors:** Minsak Nanang, Adrian Hilton, Armin Mustafa
+- **abs:** https://arxiv.org/abs/2603.11147 · **PDF:** https://arxiv.org/pdf/2603.11147v3
+
+## 5795. RCTs & Human Uplift Studies: Methodological Challenges and Practical Solutions for Frontier AI Evaluation
+
+- **ID:** `2603.11001` · **primary:** `cs.CY` · **Submitted:** 2026-03-11
+- **Authors:** Patricia Paskov, Kevin Wei, Shen Zhou Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.11001 · **PDF:** https://arxiv.org/pdf/2603.11001v1
+
+## 5796. Cross-Species Transfer Learning for Electrophysiology-to-Transcriptomics Mapping in Cortical GABAergic Interneurons
+
+- **ID:** `2603.11000` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Theo Schwider, Ramin Ramezani
+- **abs:** https://arxiv.org/abs/2603.11000 · **PDF:** https://arxiv.org/pdf/2603.11000v1
+
+## 5797. Factorized Neural Implicit DMD for Parametric Dynamics
+
+- **ID:** `2603.10995` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Siyuan Chen, Zhecheng Wang, Yixin Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.10995 · **PDF:** https://arxiv.org/pdf/2603.10995v1
+
+## 5798. Artificial Intelligence as a Catalyst for Innovation in Software Engineering
+
+- **ID:** `2603.10994` · **primary:** `cs.SE` · **Submitted:** 2026-03-11
+- **Authors:** Carlos Alberto Fernández-y-Fernández, Jorge R. Aguilar-Cisneros
+- **abs:** https://arxiv.org/abs/2603.10994 · **PDF:** https://arxiv.org/pdf/2603.10994v1
+
+## 5799. Bayesian Optimization with Gaussian Processes to Accelerate Stationary Point Searches
+
+- **ID:** `2603.10992` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** Rohit Goswami
+- **abs:** https://arxiv.org/abs/2603.10992 · **PDF:** https://arxiv.org/pdf/2603.10992v2
+
+## 5800. ForwardFlow: Simulation only statistical inference using deep learning
+
+- **ID:** `2603.10991` · **primary:** `math.ST` · **Submitted:** 2026-03-11
+- **Authors:** Stefan Böhringer
+- **abs:** https://arxiv.org/abs/2603.10991 · **PDF:** https://arxiv.org/pdf/2603.10991v1
+
+## 5801. MCMC Informed Neural Emulators for Uncertainty Quantification in Dynamical Systems
+
+- **ID:** `2603.10987` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Heikki Haario, Zhi-Song Liu, Martin Simon, et al.
+- **abs:** https://arxiv.org/abs/2603.10987 · **PDF:** https://arxiv.org/pdf/2603.10987v1
+
+## 5802. The Discrete Charm of the MLP: Binary Routing of Continuous Signals in Transformer Feed-Forward Layers
+
+- **ID:** `2603.10985` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Peter Balogh
+- **abs:** https://arxiv.org/abs/2603.10985 · **PDF:** https://arxiv.org/pdf/2603.10985v1
+
+## 5803. Federated Learning-driven Beam Management in LEO 6G Non-Terrestrial Networks
+
+- **ID:** `2603.10983` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Maria Lamprini Bartsioka, Ioannis A. Bartsiokas, Athanasios D. Panagopoulos, et al.
+- **abs:** https://arxiv.org/abs/2603.10983 · **PDF:** https://arxiv.org/pdf/2603.10983v1
+
+## 5804. Attention Gathers, MLPs Compose: A Causal Analysis of an Action-Outcome Circuit in VideoViT
+
+- **ID:** `2603.11142` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Sai V R Chereddy
+- **abs:** https://arxiv.org/abs/2603.11142 · **PDF:** https://arxiv.org/pdf/2603.11142v1
+
+## 5805. GroundCount: Grounding Vision-Language Models with Object Detection for Mitigating Counting Hallucinations
+
+- **ID:** `2603.10978` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Boyuan Chen, Minghao Shao, Siddharth Garg, et al.
+- **abs:** https://arxiv.org/abs/2603.10978 · **PDF:** https://arxiv.org/pdf/2603.10978v1
+
+## 5806. FRIEND: Federated Learning for Joint Optimization of multi-RIS Configuration and Eavesdropper Intelligent Detection in B5G Networks
+
+- **ID:** `2603.10977` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Maria Lamprini A. Bartsioka, Ioannis A. Bartsiokas, Anastasios K. Papazafeiropoulos, et al.
+- **abs:** https://arxiv.org/abs/2603.10977 · **PDF:** https://arxiv.org/pdf/2603.10977v1
+
+## 5807. Contact Coverage-Guided Exploration for General-Purpose Dexterous Manipulation
+
+- **ID:** `2603.10971` · **primary:** `cs.RO` · **Submitted:** 2026-03-11
+- **Authors:** Zixuan Liu, Ruoyi Qiao, Chenrui Tie, et al.
+- **abs:** https://arxiv.org/abs/2603.10971 · **PDF:** https://arxiv.org/pdf/2603.10971v1
+
+## 5808. TOSSS: a CVE-based Software Security Benchmark for Large Language Models
+
+- **ID:** `2603.10969` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Marc Damie, Murat Bilgehan Ertan, Domenico Essoussi, et al.
+- **abs:** https://arxiv.org/abs/2603.10969 · **PDF:** https://arxiv.org/pdf/2603.10969v2
+
+## 5809. Pointy - A Lightweight Transformer for Point Cloud Foundation Models
+
+- **ID:** `2603.10963` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Konrad Szafer, Marek Kraft, Dominik Belter
+- **abs:** https://arxiv.org/abs/2603.10963 · **PDF:** https://arxiv.org/pdf/2603.10963v1
+
+## 5810. Bio-Inspired Self-Supervised Learning for Wrist-worn IMU Signals
+
+- **ID:** `2603.10961` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Prithviraj Tarale, Kiet Chu, Abhishek Varghese, et al.
+- **abs:** https://arxiv.org/abs/2603.10961 · **PDF:** https://arxiv.org/pdf/2603.10961v1
+
+## 5811. Ranking Reasoning LLMs under Test-Time Scaling
+
+- **ID:** `2603.10960` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Mohsen Hariri, Michael Hinczewski, Jing Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.10960 · **PDF:** https://arxiv.org/pdf/2603.10960v1
+
+## 5812. When should we trust the annotation? Selective prediction for molecular structure retrieval from mass spectra
+
+- **ID:** `2603.10950` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Mira Jürgens, Gaetan De Waele, Morteza Rakhshaninejad, et al.
+- **abs:** https://arxiv.org/abs/2603.10950 · **PDF:** https://arxiv.org/pdf/2603.10950v1
+
+## 5813. Procedural Fairness via Group Counterfactual Explanation
+
+- **ID:** `2603.11140` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Gideon Popoola, John Sheppard
+- **abs:** https://arxiv.org/abs/2603.11140 · **PDF:** https://arxiv.org/pdf/2603.11140v1
+
+## 5814. H2LooP Spark Preview: Continual Pretraining of Large Language Models for Low-Level Embedded Systems Code
+
+- **ID:** `2603.11139` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Amit Singh, Vedant Nipane, Pulkit Agrawal, et al.
+- **abs:** https://arxiv.org/abs/2603.11139 · **PDF:** https://arxiv.org/pdf/2603.11139v2
+
+## 5815. Deep regression learning from dependent observations with minimum error entropy principle
+
+- **ID:** `2603.11138` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** William Kengne, Modou Wade
+- **abs:** https://arxiv.org/abs/2603.11138 · **PDF:** https://arxiv.org/pdf/2603.11138v1
+
+## 5816. Scaling Reasoning Efficiently via Relaxed On-Policy Distillation
+
+- **ID:** `2603.11137` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Jongwoo Ko, Sara Abdali, Young Jin Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.11137 · **PDF:** https://arxiv.org/pdf/2603.11137v1
+
+## 5817. Safe RLHF Beyond Expectation: Stochastic Dominance for Universal Spectral Risk Control
+
+- **ID:** `2603.10938` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Yaswanth Chittepu, Ativ Joshi, Rajarshi Bhattacharjee, et al.
+- **abs:** https://arxiv.org/abs/2603.10938 · **PDF:** https://arxiv.org/pdf/2603.10938v1
+
+## 5818. Quantifying Membership Disclosure Risk for Tabular Synthetic Data Using Kernel Density Estimators
+
+- **ID:** `2603.10937` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Rajdeep Pathak, Sayantee Jana
+- **abs:** https://arxiv.org/abs/2603.10937 · **PDF:** https://arxiv.org/pdf/2603.10937v1
+
+## 5819. Spherical VAE with Cluster-Aware Feasible Regions: Guaranteed Prevention of Posterior Collapse
+
+- **ID:** `2603.10935` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Zegu Zhang, Jian Zhang
+- **abs:** https://arxiv.org/abs/2603.10935 · **PDF:** https://arxiv.org/pdf/2603.10935v3
+
+## 5820. Conformal e-prediction in the presence of confounding
+
+- **ID:** `2603.11134` · **primary:** `math.ST` · **Submitted:** 2026-03-11
+- **Authors:** Vladimir Vovk, Ruodu Wang
+- **abs:** https://arxiv.org/abs/2603.11134 · **PDF:** https://arxiv.org/pdf/2603.11134v1
+
+## 5821. ECoLAD: Deployment-Oriented Evaluation for Automotive Time-Series Anomaly Detection
+
+- **ID:** `2603.10926` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Kadir-Kaan Özer, René Ebeling, Markus Enzweiler
+- **abs:** https://arxiv.org/abs/2603.10926 · **PDF:** https://arxiv.org/pdf/2603.10926v1
+
+## 5822. Higher-Order Modular Attention: Fusing Pairwise and Triadic Interactions for Protein Sequences
+
+- **ID:** `2603.11133` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Shirin Amiraslani, Xin Gao
+- **abs:** https://arxiv.org/abs/2603.11133 · **PDF:** https://arxiv.org/pdf/2603.11133v1
+
+## 5823. WebWeaver: Breaking Topology Confidentiality in LLM Multi-Agent Systems with Stealthy Context-Based Inference
+
+- **ID:** `2603.11132` · **primary:** `cs.CR` · **Submitted:** 2026-03-11
+- **Authors:** Zixun Xiong, Gaoyi Wu, Lingfeng Yao, et al.
+- **abs:** https://arxiv.org/abs/2603.11132 · **PDF:** https://arxiv.org/pdf/2603.11132v2
+
+## 5824. NCAA Bracket Prediction Using Machine Learning and Combinatorial Fusion Analysis
+
+- **ID:** `2603.10916` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Yuanhong Wu, Isaiah Smith, Tushar Marwah, et al.
+- **abs:** https://arxiv.org/abs/2603.10916 · **PDF:** https://arxiv.org/pdf/2603.10916v1
+
+## 5825. LLM2Vec-Gen: Generative Embeddings from Large Language Models
+
+- **ID:** `2603.10913` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Parishad BehnamGhader, Vaibhav Adlakha, Fabian David Schmidt, et al.
+- **abs:** https://arxiv.org/abs/2603.10913 · **PDF:** https://arxiv.org/pdf/2603.10913v2
+
+## 5826. Beyond Barren Plateaus: A Scalable Quantum Convolutional Architecture for High-Fidelity Image Classification
+
+- **ID:** `2603.11131` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Radhakrishnan Delhibabu
+- **abs:** https://arxiv.org/abs/2603.11131 · **PDF:** https://arxiv.org/pdf/2603.11131v1
+
+## 5827. GLM-OCR Technical Report
+
+- **ID:** `2603.10910` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Shuaiqi Duan, Yadong Xue, Weihan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.10910 · **PDF:** https://arxiv.org/pdf/2603.10910v2
+
+## 5828. Parameter-Efficient Token Embedding Editing for Clinical Class-Level Unlearning
+
+- **ID:** `2603.19302` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Iyad Ait Hou, Shrenik Borad, Harsh Sharma, et al.
+- **abs:** https://arxiv.org/abs/2603.19302 · **PDF:** https://arxiv.org/pdf/2603.19302v1
+
+## 5829. When Fine-Tuning Fails and when it Generalises: Role of Data Diversity and Mixed Training in LLM-based TTS
+
+- **ID:** `2603.10904` · **primary:** `cs.SD` · **Submitted:** 2026-03-11
+- **Authors:** Anupam Purwar, Aditya Choudhary
+- **abs:** https://arxiv.org/abs/2603.10904 · **PDF:** https://arxiv.org/pdf/2603.10904v1
+
+## 5830. LookaheadKV: Fast and Accurate KV Cache Eviction by Glimpsing into the Future without Generation
+
+- **ID:** `2603.10899` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Jinwoo Ahn, Ingyu Seong, Akhil Kedia, et al.
+- **abs:** https://arxiv.org/abs/2603.10899 · **PDF:** https://arxiv.org/pdf/2603.10899v1
+
+## 5831. Ergodicity in reinforcement learning
+
+- **ID:** `2603.10895` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Dominik Baumann, Erfaun Noorani, Arsenii Mustafin, et al.
+- **abs:** https://arxiv.org/abs/2603.10895 · **PDF:** https://arxiv.org/pdf/2603.10895v1
+
+## 5832. Efficient Approximation to Analytic and $L^p$ functions by Height-Augmented ReLU Networks
+
+- **ID:** `2603.11128` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** ZeYu Li, FengLei Fan, TieYong Zeng
+- **abs:** https://arxiv.org/abs/2603.11128 · **PDF:** https://arxiv.org/pdf/2603.11128v1
+
+## 5833. A Hybrid Knowledge-Grounded Framework for Safety and Traceability in Prescription Verification
+
+- **ID:** `2603.10891` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Yichi Zhu, Kan Ling, Xu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.10891 · **PDF:** https://arxiv.org/pdf/2603.10891v1
+
+## 5834. Dynamics-Predictive Sampling for Active RL Finetuning of Large Reasoning Models
+
+- **ID:** `2603.10887` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Yixiu Mao, Yun Qu, Qi Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.10887 · **PDF:** https://arxiv.org/pdf/2603.10887v1
+
+## 5835. Kernel Tests of Equivalence
+
+- **ID:** `2603.10886` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** Xing Liu, Axel Gandy
+- **abs:** https://arxiv.org/abs/2603.10886 · **PDF:** https://arxiv.org/pdf/2603.10886v2
+
+## 5836. Continuous Diffusion Transformers for Designing Synthetic Regulatory Elements
+
+- **ID:** `2603.10885` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Jonathan Liu, Kia Ghods
+- **abs:** https://arxiv.org/abs/2603.10885 · **PDF:** https://arxiv.org/pdf/2603.10885v1
+
+## 5837. LAtte: Hyperbolic Lorentz Attention for Cross-Subject EEG Classification
+
+- **ID:** `2603.10881` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Johannes Burchert, Ahmad Bdeir, Tom Hanika, et al.
+- **abs:** https://arxiv.org/abs/2603.10881 · **PDF:** https://arxiv.org/pdf/2603.10881v1
+
+## 5838. From Images to Words: Efficient Cross-Modal Knowledge Distillation to Language Models from Black-box Teachers
+
+- **ID:** `2603.10877` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Ayan Sengupta, Shantanu Dixit, Md Shad Akhtar, et al.
+- **abs:** https://arxiv.org/abs/2603.10877 · **PDF:** https://arxiv.org/pdf/2603.10877v1
+
+## 5839. An Extreme Multi-label Text Classification (XMTC) Library Dataset: What if we took "Use of Practical AI in Digital Libraries" seriously?
+
+- **ID:** `2603.10876` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Jennifer D'Souza, Sameer Sadruddin, Maximilian Kähler, et al.
+- **abs:** https://arxiv.org/abs/2603.10876 · **PDF:** https://arxiv.org/pdf/2603.10876v1
+
+## 5840. SNPgen: Phenotype-Supervised Genotype Representation and Synthetic Data Generation via Latent Diffusion
+
+- **ID:** `2603.10873` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Andrea Lampis, Michela Carlotta Massi, Nicola Pirastu, et al.
+- **abs:** https://arxiv.org/abs/2603.10873 · **PDF:** https://arxiv.org/pdf/2603.10873v1
+
+## 5841. SiDiaC-v.2.0: Sinhala Diachronic Corpus Version 2.0
+
+- **ID:** `2603.10861` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Nevidu Jayatilleke, Nisansa de Silva, Uthpala Nimanthi, et al.
+- **abs:** https://arxiv.org/abs/2603.10861 · **PDF:** https://arxiv.org/pdf/2603.10861v1
+
+## 5842. GRACE: A Unified 2D Multi-Robot Path Planning Simulator & Benchmark for Grid, Roadmap, And Continuous Environments
+
+- **ID:** `2603.10858` · **primary:** `cs.RO` · **Submitted:** 2026-03-11
+- **Authors:** Chuanlong Zang, Anna Mannucci, Isabelle Barz, et al.
+- **abs:** https://arxiv.org/abs/2603.10858 · **PDF:** https://arxiv.org/pdf/2603.10858v1
+
+## 5843. 6ABOS: An Open-Source Atmospheric Correction Framework for the EnMAP Hyperspectral Mission Based on 6S
+
+- **ID:** `2603.10856` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Gabriel Caballero Cañas, Bárbara Alvado Arranz, Xavier Sòria-Perpinyà, et al.
+- **abs:** https://arxiv.org/abs/2603.10856 · **PDF:** https://arxiv.org/pdf/2603.10856v1
+
+## 5844. Enhancing Value Alignment of LLMs with Multi-agent system and Combinatorial Fusion
+
+- **ID:** `2603.11126` · **primary:** `cs.MA` · **Submitted:** 2026-03-11
+- **Authors:** Yuanhong Wu, Djallel Bouneffouf, D. Frank Hsu
+- **abs:** https://arxiv.org/abs/2603.11126 · **PDF:** https://arxiv.org/pdf/2603.11126v1
+
+## 5845. $V_{0.5}$: Generalist Value Model as a Prior for Sparse RL Rollouts
+
+- **ID:** `2603.10848` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Yi-Kai Zhang, Yueqing Sun, Hongyan Hao, et al.
+- **abs:** https://arxiv.org/abs/2603.10848 · **PDF:** https://arxiv.org/pdf/2603.10848v1
+
+## 5846. Semantic Landmark Particle Filter for Robot Localisation in Vineyards
+
+- **ID:** `2603.10847` · **primary:** `cs.RO` · **Submitted:** 2026-03-11
+- **Authors:** Rajitha de Silva, Jonathan Cox, James R. Heselden, et al.
+- **abs:** https://arxiv.org/abs/2603.10847 · **PDF:** https://arxiv.org/pdf/2603.10847v1
+
+## 5847. Towards Cold-Start Drafting and Continual Refining: A Value-Driven Memory Approach with Application to NPU Kernel Synthesis
+
+- **ID:** `2603.10846` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Yujie Zheng, Zhuo Li, Shengtao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.10846 · **PDF:** https://arxiv.org/pdf/2603.10846v1
+
+## 5848. Human Presence Detection via Wi-Fi Range-Filtered Doppler Spectrum on Commodity Laptops
+
+- **ID:** `2603.10845` · **primary:** `eess.SP` · **Submitted:** 2026-03-11
+- **Authors:** Jessica Sanson, Rahul C. Shah, Valerio Frascolla
+- **abs:** https://arxiv.org/abs/2603.10845 · **PDF:** https://arxiv.org/pdf/2603.10845v2
+
+## 5849. PivotAttack: Rethinking the Search Trajectory in Hard-Label Text Attacks via Pivot Words
+
+- **ID:** `2603.10842` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Yuzhi Liang, Shiliang Xiao, Jingsong Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.10842 · **PDF:** https://arxiv.org/pdf/2603.10842v1
+
+## 5850. Co-Diffusion: An Affinity-Aware Two-Stage Latent Diffusion Framework for Generalizable Drug-Target Affinity Prediction
+
+- **ID:** `2603.11125` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** Yining Qian, Pengjie Wang, Yixiao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.11125 · **PDF:** https://arxiv.org/pdf/2603.11125v1
+
+## 5851. On the Reliability of Cue Conflict and Beyond
+
+- **ID:** `2603.10834` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Pum Jun Kim, Seung-Ah Lee, Seongho Park, et al.
+- **abs:** https://arxiv.org/abs/2603.10834 · **PDF:** https://arxiv.org/pdf/2603.10834v2
+
+## 5852. Taming Epilepsy: Mean Field Control of Whole-Brain Dynamics
+
+- **ID:** `2603.18035` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Ming Li, Ting Gao, Jingqiao Dua
+- **abs:** https://arxiv.org/abs/2603.18035 · **PDF:** https://arxiv.org/pdf/2603.18035v1
+
+## 5853. BALD-SAM: Disagreement-based Active Prompting in Interactive Segmentation
+
+- **ID:** `2603.10828` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Prithwijit Chowdhury, Mohit Prabhushankar, Ghassan AlRegib
+- **abs:** https://arxiv.org/abs/2603.10828 · **PDF:** https://arxiv.org/pdf/2603.10828v1
+
+## 5854. Speaker Verification with Speech-Aware LLMs: Evaluation and Augmentation
+
+- **ID:** `2603.10827` · **primary:** `cs.SD` · **Submitted:** 2026-03-11
+- **Authors:** Thomas Thebaud, Yuzhe Wang, Laureano Moro-Velazquez, et al.
+- **abs:** https://arxiv.org/abs/2603.10827 · **PDF:** https://arxiv.org/pdf/2603.10827v1
+
+## 5855. ReTabSyn: Realistic Tabular Data Synthesis via Reinforcement Learning
+
+- **ID:** `2603.10823` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** Xiaofeng Lin, Seungbae Kim, Zhuoya Li, et al.
+- **abs:** https://arxiv.org/abs/2603.10823 · **PDF:** https://arxiv.org/pdf/2603.10823v1
+
+## 5856. Evaluating randomized smoothing as a defense against adversarial attacks in trajectory prediction
+
+- **ID:** `2603.10821` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Julian F. Schumann, Eduardo Figueiredo, Frederik Baymler Mathiesen, et al.
+- **abs:** https://arxiv.org/abs/2603.10821 · **PDF:** https://arxiv.org/pdf/2603.10821v1
+
+## 5857. Uni-ASR: Unified LLM-Based Architecture for Non-Streaming and Streaming Automatic Speech Recognition
+
+- **ID:** `2603.11123` · **primary:** `cs.SD` · **Submitted:** 2026-03-11
+- **Authors:** Yinfeng Xia, Jian Tang, Junfeng Hou, et al.
+- **abs:** https://arxiv.org/abs/2603.11123 · **PDF:** https://arxiv.org/pdf/2603.11123v1
+
+## 5858. Protein Counterfactuals via Diffusion-Guided Latent Optimization
+
+- **ID:** `2603.10811` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Weronika Kłos, Sidney Bender, Lukas Kades
+- **abs:** https://arxiv.org/abs/2603.10811 · **PDF:** https://arxiv.org/pdf/2603.10811v1
+
+## 5859. Nurture-First Agent Development: Building Domain-Expert AI Agents Through Conversational Knowledge Crystallization
+
+- **ID:** `2603.10808` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Linghao Zhang
+- **abs:** https://arxiv.org/abs/2603.10808 · **PDF:** https://arxiv.org/pdf/2603.10808v1
+
+## 5860. Risk-Adjusted Harm Scoring for Automated Red Teaming for LLMs in Financial Services
+
+- **ID:** `2603.10807` · **primary:** `q-fin.CP` · **Submitted:** 2026-03-11
+- **Authors:** Fabrizio Dimino, Bhaskarjit Sarmah, Stefano Pasquali
+- **abs:** https://arxiv.org/abs/2603.10807 · **PDF:** https://arxiv.org/pdf/2603.10807v1
+
+## 5861. Towards Intelligent Spectrum Management: Spectrum Demand Estimation Using Graph Neural Networks
+
+- **ID:** `2603.10802` · **primary:** `cs.NI` · **Submitted:** 2026-03-11
+- **Authors:** Mohamad Alkadamani, Amir Ghasemi, Halim Yanikomeroglu
+- **abs:** https://arxiv.org/abs/2603.10802 · **PDF:** https://arxiv.org/pdf/2603.10802v1
+
+## 5862. AI-Enhanced Spatial Cellular Traffic Demand Prediction with Contextual Clustering and Error Correction for 5G/6G Planning
+
+- **ID:** `2603.10800` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Mohamad Alkadamani, Colin Brown, Halim Yanikomeroglu
+- **abs:** https://arxiv.org/abs/2603.10800 · **PDF:** https://arxiv.org/pdf/2603.10800v1
+
+## 5863. QV May Be Enough: Toward the Essence of Attention in LLMs
+
+- **ID:** `2603.15665` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Zhang Edward
+- **abs:** https://arxiv.org/abs/2603.15665 · **PDF:** https://arxiv.org/pdf/2603.15665v1
+
+## 5864. High-resolution weather-guided surrogate modeling for data-efficient cross-location building energy prediction
+
+- **ID:** `2603.11121` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Piragash Manmatharasan, Girma Bitsuamlak, Katarina Grolinger
+- **abs:** https://arxiv.org/abs/2603.11121 · **PDF:** https://arxiv.org/pdf/2603.11121v1
+
+## 5865. Multilingual Reasoning Gym: Multilingual Scaling of Procedural Reasoning Environments
+
+- **ID:** `2603.10793` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Konstantin Dobler, Simon Lehnerer, Federico Scozzafava, et al.
+- **abs:** https://arxiv.org/abs/2603.10793 · **PDF:** https://arxiv.org/pdf/2603.10793v1
+
+## 5866. LuxBorrow: From Pompier to Pompjee, Tracing Borrowing in Luxembourgish
+
+- **ID:** `2603.10789` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Nina Hosseini-Kivanani, Fred Philippy
+- **abs:** https://arxiv.org/abs/2603.10789 · **PDF:** https://arxiv.org/pdf/2603.10789v2
+
+## 5867. Interpretable Chinese Metaphor Identification via LLM-Assisted MIPVU Rule Script Generation: A Comparative Protocol Study
+
+- **ID:** `2603.10784` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Weihang Huang, Mengna Liu
+- **abs:** https://arxiv.org/abs/2603.10784 · **PDF:** https://arxiv.org/pdf/2603.10784v1
+
+## 5868. Taking Shortcuts for Categorical VQA Using Super Neurons
+
+- **ID:** `2603.10781` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Pierre Musacchio, Jaeyi Jeong, Dahun Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.10781 · **PDF:** https://arxiv.org/pdf/2603.10781v1
+
+## 5869. COT-FM: Cluster-wise Optimal Transport Flow Matching
+
+- **ID:** `2603.13395` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Chiensheng Chiang, Kuan-Hsun Tu, Jia-Wei Liao, et al.
+- **abs:** https://arxiv.org/abs/2603.13395 · **PDF:** https://arxiv.org/pdf/2603.13395v1
+
+## 5870. Dynamics-Informed Deep Learning for Predicting Extreme Events
+
+- **ID:** `2603.10777` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Eirini Katsidoniotaki, Themistoklis P. Sapsis
+- **abs:** https://arxiv.org/abs/2603.10777 · **PDF:** https://arxiv.org/pdf/2603.10777v1
+
+## 5871. Large Language Models as Annotators for Machine Translation Quality Estimation
+
+- **ID:** `2603.10775` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Sidi Wang, Sophie Arnoult, Amir Kamran
+- **abs:** https://arxiv.org/abs/2603.10775 · **PDF:** https://arxiv.org/pdf/2603.10775v1
+
+## 5872. Word Recovery in Large Language Models Enables Character-Level Tokenization Robustness
+
+- **ID:** `2603.10771` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Zhipeng Yang, Shu Yang, Lijie Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.10771 · **PDF:** https://arxiv.org/pdf/2603.10771v1
+
+## 5873. Group Resonance Network: Learnable Prototypes and Multi-Subject Resonance for EEG Emotion Recognition
+
+- **ID:** `2603.11119` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Renwei Meng
+- **abs:** https://arxiv.org/abs/2603.11119 · **PDF:** https://arxiv.org/pdf/2603.11119v1
+
+## 5874. A Learning-Based Superposition Operator for Non-Renewal Arrival Processes in Queueing Networks
+
+- **ID:** `2603.11118` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Eliran Sherzer
+- **abs:** https://arxiv.org/abs/2603.11118 · **PDF:** https://arxiv.org/pdf/2603.11118v1
+
+## 5875. Learning Tree-Based Models with Gradient Descent
+
+- **ID:** `2603.11117` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Sascha Marton
+- **abs:** https://arxiv.org/abs/2603.11117 · **PDF:** https://arxiv.org/pdf/2603.11117v1
+
+## 5876. mAceReason-Math: A Dataset of High-Quality Multilingual Math Problems Ready For RLVR
+
+- **ID:** `2603.10767` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Konstantin Dobler, Simon Lehnerer, Federico Scozzafava, et al.
+- **abs:** https://arxiv.org/abs/2603.10767 · **PDF:** https://arxiv.org/pdf/2603.10767v2
+
+## 5877. HeartAgent: An Autonomous Agent System for Explainable Differential Diagnosis in Cardiology
+
+- **ID:** `2603.10764` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Shuang Zhou, Kai Yu, Song Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.10764 · **PDF:** https://arxiv.org/pdf/2603.10764v1
+
+## 5878. Prioritizing Gradient Sign Over Modulus: An Importance-Aware Framework for Wireless Federated Learning
+
+- **ID:** `2603.10763` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Yiyang Yue, Jiacheng Yao, Wei Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.10763 · **PDF:** https://arxiv.org/pdf/2603.10763v1
+
+## 5879. A PUF-Based Approach for Copy Protection of Intellectual Property in Neural Network Models
+
+- **ID:** `2603.10753` · **primary:** `cs.CR` · **Submitted:** 2026-03-11
+- **Authors:** Daniel Dorfmeister, Flavio Ferrarotti, Bernhard Fischer, et al.
+- **abs:** https://arxiv.org/abs/2603.10753 · **PDF:** https://arxiv.org/pdf/2603.10753v1
+
+## 5880. Deep Randomized Distributed Function Computation (DeepRDFC): Neural Distributed Channel Simulation
+
+- **ID:** `2603.10750` · **primary:** `cs.IT` · **Submitted:** 2026-03-11
+- **Authors:** Didrik Bergström, Onur Günlü
+- **abs:** https://arxiv.org/abs/2603.10750 · **PDF:** https://arxiv.org/pdf/2603.10750v1
+
+## 5881. CUPID: A Plug-in Framework for Joint Aleatoric and Epistemic Uncertainty Estimation with a Single Model
+
+- **ID:** `2603.10745` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Xinran Xu, Xiuyi Fan
+- **abs:** https://arxiv.org/abs/2603.10745 · **PDF:** https://arxiv.org/pdf/2603.10745v1
+
+## 5882. A Grammar of Machine Learning Workflows
+
+- **ID:** `2603.10742` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Simon Roth
+- **abs:** https://arxiv.org/abs/2603.10742 · **PDF:** https://arxiv.org/pdf/2603.10742v2
+
+## 5883. Beyond Accuracy: Reliability and Uncertainty Estimation in Convolutional Neural Networks
+
+- **ID:** `2603.10731` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Sanne Ruijs, Alina Kosiakova, Farrukh Javed
+- **abs:** https://arxiv.org/abs/2603.10731 · **PDF:** https://arxiv.org/pdf/2603.10731v1
+
+## 5884. CacheSolidarity: Preventing Prefix Caching Side Channels in Multi-tenant LLM Serving Systems
+
+- **ID:** `2603.10726` · **primary:** `cs.CR` · **Submitted:** 2026-03-11
+- **Authors:** Panagiotis Georgios Pennas, Konstantinos Papaioannou, Marco Guarnieri, et al.
+- **abs:** https://arxiv.org/abs/2603.10726 · **PDF:** https://arxiv.org/pdf/2603.10726v1
+
+## 5885. Towards Robust Speech Deepfake Detection via Human-Inspired Reasoning
+
+- **ID:** `2603.10725` · **primary:** `cs.SD` · **Submitted:** 2026-03-11
+- **Authors:** Artem Dvirniak, Evgeny Kushnir, Dmitrii Tarasov, et al.
+- **abs:** https://arxiv.org/abs/2603.10725 · **PDF:** https://arxiv.org/pdf/2603.10725v2
+
+## 5886. UAV traffic scene understanding: A regulation embedded multi-modal network and a unified benchmark
+
+- **ID:** `2603.10722` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Yu Zhang, Zhicheng Zhao, Ze Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.10722 · **PDF:** https://arxiv.org/pdf/2603.10722v2
+
+## 5887. Sample-and-Search: An Effective Algorithm for Learning-Augmented k-Median Clustering in High dimensions
+
+- **ID:** `2603.10721` · **primary:** `cs.DS` · **Submitted:** 2026-03-11
+- **Authors:** Kangke Cheng, Shihong Song, Guanlin Mo, et al.
+- **abs:** https://arxiv.org/abs/2603.10721 · **PDF:** https://arxiv.org/pdf/2603.10721v1
+
+## 5888. Task-Conditioned Routing Signatures in Sparse Mixture-of-Experts Transformers
+
+- **ID:** `2603.11114` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Mynampati Sri Ranganadha Avinash
+- **abs:** https://arxiv.org/abs/2603.11114 · **PDF:** https://arxiv.org/pdf/2603.11114v1
+
+## 5889. Riemannian MeanFlow for One-Step Generation on Manifolds
+
+- **ID:** `2603.10718` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Zichen Zhong, Haoliang Sun, Yukun Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.10718 · **PDF:** https://arxiv.org/pdf/2603.10718v1
+
+## 5890. Probabilistic Verification of Voice Anti-Spoofing Models
+
+- **ID:** `2603.10713` · **primary:** `cs.SD` · **Submitted:** 2026-03-11
+- **Authors:** Evgeny Kushnir, Alexandr Kozodaev, Dmitrii Korzh, et al.
+- **abs:** https://arxiv.org/abs/2603.10713 · **PDF:** https://arxiv.org/pdf/2603.10713v2
+
+## 5891. Prism-$Δ$: Differential Subspace Steering for Prompt Highlighting in Large Language Models
+
+- **ID:** `2603.10705` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Yuyao Ge, Shenghua Liu, Yiwei Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.10705 · **PDF:** https://arxiv.org/pdf/2603.10705v1
+
+## 5892. Partition-Based Functional Ridge Regression for High-Dimensional Data
+
+- **ID:** `2603.11113` · **primary:** `stat.ME` · **Submitted:** 2026-03-11
+- **Authors:** Shaista Ashraf, Ismail Shah, Farrukh Javed
+- **abs:** https://arxiv.org/abs/2603.11113 · **PDF:** https://arxiv.org/pdf/2603.11113v1
+
+## 5893. AlphaFlowTSE: One-Step Generative Target Speaker Extraction via Conditional AlphaFlow
+
+- **ID:** `2603.10701` · **primary:** `cs.SD` · **Submitted:** 2026-03-11
+- **Authors:** Duojia Li, Shuhan Zhang, Zihan Qian, et al.
+- **abs:** https://arxiv.org/abs/2603.10701 · **PDF:** https://arxiv.org/pdf/2603.10701v1
+
+## 5894. Structured Linked Data as a Memory Layer for Agent-Orchestrated Retrieval
+
+- **ID:** `2603.10700` · **primary:** `cs.IR` · **Submitted:** 2026-03-11
+- **Authors:** Andrea Volpini, Elie Raad, Beatrice Gamba, et al.
+- **abs:** https://arxiv.org/abs/2603.10700 · **PDF:** https://arxiv.org/pdf/2603.10700v1
+
+## 5895. EvoSchema: Towards Text-to-SQL Robustness Against Schema Evolution
+
+- **ID:** `2603.10697` · **primary:** `cs.DB` · **Submitted:** 2026-03-11
+- **Authors:** Tianshu Zhang, Kun Qian, Siddhartha Sahai, et al.
+- **abs:** https://arxiv.org/abs/2603.10697 · **PDF:** https://arxiv.org/pdf/2603.10697v1
+
+## 5896. RandMark: On Random Watermarking of Visual Foundation Models
+
+- **ID:** `2603.10695` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Anna Chistyakova, Mikhail Pautov
+- **abs:** https://arxiv.org/abs/2603.10695 · **PDF:** https://arxiv.org/pdf/2603.10695v1
+
+## 5897. Repurposing Backdoors for Good: Ephemeral Intrinsic Proofs for Verifiable Aggregation in Cross-silo Federated Learning
+
+- **ID:** `2603.10692` · **primary:** `cs.CR` · **Submitted:** 2026-03-11
+- **Authors:** Xian Qin, Xue Yang, Xiaohu Tang
+- **abs:** https://arxiv.org/abs/2603.10692 · **PDF:** https://arxiv.org/pdf/2603.10692v1
+
+## 5898. Contract And Conquer: How to Provably Compute Adversarial Examples for a Black-Box Model?
+
+- **ID:** `2603.10689` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Anna Chistyakova, Mikhail Pautov
+- **abs:** https://arxiv.org/abs/2603.10689 · **PDF:** https://arxiv.org/pdf/2603.10689v2
+
+## 5899. SciNav: A General Agent Framework for Scientific Coding Tasks
+
+- **ID:** `2603.20256` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Tianshu Zhang, Huan Sun
+- **abs:** https://arxiv.org/abs/2603.20256 · **PDF:** https://arxiv.org/pdf/2603.20256v1
+
+## 5900. A Platform-Agnostic Multimodal Digital Human Modelling Framework: Neurophysiological Sensing in Game-Based Interaction
+
+- **ID:** `2603.10680` · **primary:** `cs.HC` · **Submitted:** 2026-03-11
+- **Authors:** Daniel J. Buxton, Mufti Mahmud, Jordan J. Bird, et al.
+- **abs:** https://arxiv.org/abs/2603.10680 · **PDF:** https://arxiv.org/pdf/2603.10680v1
+
+## 5901. Surrogate models for nuclear fusion with parametric Shallow Recurrent Decoder Networks: applications to magnetohydrodynamics
+
+- **ID:** `2603.10678` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** M. Lo Verso, C. Introini, E. Cervi, et al.
+- **abs:** https://arxiv.org/abs/2603.10678 · **PDF:** https://arxiv.org/pdf/2603.10678v1
+
+## 5902. Emulating Clinician Cognition via Self-Evolving Deep Clinical Research
+
+- **ID:** `2603.10677` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Ruiyang Ren, Yuhao Wang, Yunsen Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.10677 · **PDF:** https://arxiv.org/pdf/2603.10677v1
+
+## 5903. Spatio-Temporal Attention Graph Neural Network: Explaining Causalities With Attention
+
+- **ID:** `2603.10676` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Kosti Koistinen, Kirsi Hellsten, Joni Herttuainen, et al.
+- **abs:** https://arxiv.org/abs/2603.10676 · **PDF:** https://arxiv.org/pdf/2603.10676v1
+
+## 5904. ResWM: Residual-Action World Model for Visual RL
+
+- **ID:** `2603.11110` · **primary:** `cs.RO` · **Submitted:** 2026-03-11
+- **Authors:** Jseen Zhang, Gabriel Adineera, Jinzhou Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.11110 · **PDF:** https://arxiv.org/pdf/2603.11110v1
+
+## 5905. FAME: Formal Abstract Minimal Explanation for Neural Networks
+
+- **ID:** `2603.10661` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Ryma Boumazouza, Raya Elsaleh, Melanie Ducoffe, et al.
+- **abs:** https://arxiv.org/abs/2603.10661 · **PDF:** https://arxiv.org/pdf/2603.10661v1
+
+## 5906. Are Video Reasoning Models Ready to Go Outside?
+
+- **ID:** `2603.10652` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Yangfan He, Changgyu Boo, Jaehong Yoon
+- **abs:** https://arxiv.org/abs/2603.10652 · **PDF:** https://arxiv.org/pdf/2603.10652v1
+
+## 5907. Interleaving Scheduling and Motion Planning with Incremental Learning of Symbolic Space-Time Motion Abstractions
+
+- **ID:** `2603.10651` · **primary:** `cs.RO` · **Submitted:** 2026-03-11
+- **Authors:** Elisa Tosello, Arthur Bit-Monnot, Davide Lusuardi, et al.
+- **abs:** https://arxiv.org/abs/2603.10651 · **PDF:** https://arxiv.org/pdf/2603.10651v2
+
+## 5908. Detecting and Eliminating Neural Network Backdoors Through Active Paths with Application to Intrusion Detection
+
+- **ID:** `2603.10641` · **primary:** `cs.CR` · **Submitted:** 2026-03-11
+- **Authors:** Eirik Høyheim, Magnus Wiik Eckhoff, Gudmund Grov, et al.
+- **abs:** https://arxiv.org/abs/2603.10641 · **PDF:** https://arxiv.org/pdf/2603.10641v1
+
+## 5909. Making Bielik LLM Reason (Better): A Field Report
+
+- **ID:** `2603.10640` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Adam Trybus, Bartosz Bartnicki, Remigiusz Kinas
+- **abs:** https://arxiv.org/abs/2603.10640 · **PDF:** https://arxiv.org/pdf/2603.10640v1
+
+## 5910. Reinforcement Learning with Conditional Expectation Reward
+
+- **ID:** `2603.10624` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Changyi Xiao, Caijun Xu, Yixin Cao
+- **abs:** https://arxiv.org/abs/2603.10624 · **PDF:** https://arxiv.org/pdf/2603.10624v1
+
+## 5911. Geo-ATBench: A Benchmark for Geospatial Audio Tagging with Geospatial Semantic Context
+
+- **ID:** `2603.10623` · **primary:** `eess.AS` · **Submitted:** 2026-03-11
+- **Authors:** Yuanbo Hou, Yanru Wu, Qiaoqiao Ren, et al.
+- **abs:** https://arxiv.org/abs/2603.10623 · **PDF:** https://arxiv.org/pdf/2603.10623v1
+
+## 5912. Disentangling Similarity and Relatedness in Topic Models
+
+- **ID:** `2603.10619` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Hanlin Xiao, Mauricio A. Álvarez, Rainer Breitling
+- **abs:** https://arxiv.org/abs/2603.10619 · **PDF:** https://arxiv.org/pdf/2603.10619v1
+
+## 5913. MUNIChus: Multilingual News Image Captioning Benchmark
+
+- **ID:** `2603.10613` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Yuji Chen, Alistair Plum, Hansi Hettiarachchi, et al.
+- **abs:** https://arxiv.org/abs/2603.10613 · **PDF:** https://arxiv.org/pdf/2603.10613v1
+
+## 5914. Trajectory-Informed Memory Generation for Self-Improving Agent Systems
+
+- **ID:** `2603.10600` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Gaodan Fang, Vatche Isahagian, K. R. Jayaram, et al.
+- **abs:** https://arxiv.org/abs/2603.10600 · **PDF:** https://arxiv.org/pdf/2603.10600v1
+
+## 5915. Self-Scaled Broyden Family of Quasi-Newton Methods in JAX
+
+- **ID:** `2603.10599` · **primary:** `cs.MS` · **Submitted:** 2026-03-11
+- **Authors:** Ivan Bioli, Mikel Mendibe Abarrategi
+- **abs:** https://arxiv.org/abs/2603.10599 · **PDF:** https://arxiv.org/pdf/2603.10599v1
+
+## 5916. Recover to Predict: Progressive Retrospective Learning for Variable-Length Trajectory Prediction
+
+- **ID:** `2603.10597` · **primary:** `cs.RO` · **Submitted:** 2026-03-11
+- **Authors:** Hao Zhou, Lu Qi, Jason Li, et al.
+- **abs:** https://arxiv.org/abs/2603.10597 · **PDF:** https://arxiv.org/pdf/2603.10597v1
+
+## 5917. Gradient Flow Drifting: Generative Modeling via Wasserstein Gradient Flows of KDE-Approximated Divergences
+
+- **ID:** `2603.10592` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Jiarui Cao, Zixuan Wei, Yuxin Liu
+- **abs:** https://arxiv.org/abs/2603.10592 · **PDF:** https://arxiv.org/pdf/2603.10592v1
+
+## 5918. Does LLM Alignment Really Need Diversity? An Empirical Study of Adapting RLVR Methods for Moral Reasoning
+
+- **ID:** `2603.10588` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Zhaowei Zhang, Xiaohan Liu, Xuekai Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.10588 · **PDF:** https://arxiv.org/pdf/2603.10588v1
+
+## 5919. HAPEns: Hardware-Aware Post-Hoc Ensembling for Tabular Data
+
+- **ID:** `2603.10582` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Jannis Maier, Lennart Purucker
+- **abs:** https://arxiv.org/abs/2603.10582 · **PDF:** https://arxiv.org/pdf/2603.10582v1
+
+## 5920. Multi-objective Genetic Programming with Multi-view Multi-level Feature for Enhanced Protein Secondary Structure Prediction
+
+- **ID:** `2603.12293` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Yining Qian, Lijie Su, Meiling Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.12293 · **PDF:** https://arxiv.org/pdf/2603.12293v1
+
+## 5921. CUAAudit: Meta-Evaluation of Vision-Language Models as Auditors of Autonomous Computer-Use Agents
+
+- **ID:** `2603.10577` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Marta Sumyk, Oleksandr Kosovan
+- **abs:** https://arxiv.org/abs/2603.10577 · **PDF:** https://arxiv.org/pdf/2603.10577v2
+
+## 5922. Implicit Statistical Inference in Transformers: Approximating Likelihood-Ratio Tests In-Context
+
+- **ID:** `2603.10573` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Faris Chaudhry, Siddhant Gadkari
+- **abs:** https://arxiv.org/abs/2603.10573 · **PDF:** https://arxiv.org/pdf/2603.10573v1
+
+## 5923. End-to-End Chatbot Evaluation with Adaptive Reasoning and Uncertainty Filtering
+
+- **ID:** `2603.10570` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Nhi Dang, Tung Le, Huy Tien Nguyen
+- **abs:** https://arxiv.org/abs/2603.10570 · **PDF:** https://arxiv.org/pdf/2603.10570v1
+
+## 5924. Adaptive RAN Slicing Control via Reward-Free Self-Finetuning Agents
+
+- **ID:** `2603.10564` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Yuanhao Li, Haozhe Wang, Geyong Min, et al.
+- **abs:** https://arxiv.org/abs/2603.10564 · **PDF:** https://arxiv.org/pdf/2603.10564v1
+
+## 5925. Riemannian Geometry-Preserving Variational Autoencoder for MI-BCI Data Augmentation
+
+- **ID:** `2603.10563` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Viktorija Poļaka, Ivo Pascal de Jong, Andreea Ioana Sburlea
+- **abs:** https://arxiv.org/abs/2603.10563 · **PDF:** https://arxiv.org/pdf/2603.10563v1
+
+## 5926. Quantization Robustness of Monotone Operator Equilibrium Networks
+
+- **ID:** `2603.10562` · **primary:** `math.OC` · **Submitted:** 2026-03-11
+- **Authors:** James Li, Philip H. W. Leong, Thomas Chaffey
+- **abs:** https://arxiv.org/abs/2603.10562 · **PDF:** https://arxiv.org/pdf/2603.10562v1
+
+## 5927. Thousand-GPU Large-Scale Training and Optimization Recipe for AI-Native Cloud Embodied Intelligence Infrastructure
+
+- **ID:** `2603.11101` · **primary:** `cs.RO` · **Submitted:** 2026-03-11
+- **Authors:** Yongjian Guo, Yunxuan Ma, Haoran Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.11101 · **PDF:** https://arxiv.org/pdf/2603.11101v2
+
+## 5928. A Bipartite Graph Approach to U.S.-China Cross-Market Return Forecasting
+
+- **ID:** `2603.10559` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Jing Liu, Maria Grith, Xiaowen Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.10559 · **PDF:** https://arxiv.org/pdf/2603.10559v1
+
+## 5929. Towards Cognitive Defect Analysis in Active Infrared Thermography with Vision-Text Cues
+
+- **ID:** `2603.10549` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Mohammed Salah, Eman Ouda, Giuseppe Dell'Avvocato, et al.
+- **abs:** https://arxiv.org/abs/2603.10549 · **PDF:** https://arxiv.org/pdf/2603.10549v1
+
+## 5930. Graph Tokenization for Bridging Graphs and Transformers
+
+- **ID:** `2603.11099` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Zeyuan Guo, Enmao Diao, Cheng Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.11099 · **PDF:** https://arxiv.org/pdf/2603.11099v1
+
+## 5931. Automatic End-to-End Data Integration using Large Language Models
+
+- **ID:** `2603.10547` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Aaron Steiner, Christian Bizer
+- **abs:** https://arxiv.org/abs/2603.10547 · **PDF:** https://arxiv.org/pdf/2603.10547v1
+
+## 5932. Learning to Score: Tuning Cluster Schedulers through Reinforcement Learning
+
+- **ID:** `2603.10545` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Martin Asenov, Qiwen Deng, Gingfung Yeung, et al.
+- **abs:** https://arxiv.org/abs/2603.10545 · **PDF:** https://arxiv.org/pdf/2603.10545v1
+
+## 5933. SCORE: Replacing Layer Stacking with Contractive Recurrent Depth
+
+- **ID:** `2603.10544` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Guillaume Godin
+- **abs:** https://arxiv.org/abs/2603.10544 · **PDF:** https://arxiv.org/pdf/2603.10544v1
+
+## 5934. Prompting with the human-touch: evaluating model-sensitivity of foundation models for musculoskeletal CT segmentation
+
+- **ID:** `2603.10541` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Caroline Magg, Maaike A. ter Wee, Johannes G. G. Dobbe, et al.
+- **abs:** https://arxiv.org/abs/2603.10541 · **PDF:** https://arxiv.org/pdf/2603.10541v1
+
+## 5935. An Event-Driven E-Skin System with Dynamic Binary Scanning and real time SNN Classification
+
+- **ID:** `2603.10537` · **primary:** `cs.NE` · **Submitted:** 2026-03-11
+- **Authors:** Gaishan Li, Zhengnan Fu, Anubhab Tripathi, et al.
+- **abs:** https://arxiv.org/abs/2603.10537 · **PDF:** https://arxiv.org/pdf/2603.10537v1
+
+## 5936. Tackling Length Inflation Without Trade-offs: Group Relative Reward Rescaling for Reinforcement Learning
+
+- **ID:** `2603.10535` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Zichao Li, Jie Lou, Fangchen Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.10535 · **PDF:** https://arxiv.org/pdf/2603.10535v1
+
+## 5937. UAV-MARL: Multi-Agent Reinforcement Learning for Time-Critical and Dynamic Medical Supply Delivery
+
+- **ID:** `2603.10528` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Islam Guven, Mehmet Parlak
+- **abs:** https://arxiv.org/abs/2603.10528 · **PDF:** https://arxiv.org/pdf/2603.10528v1
+
+## 5938. World Model for Battery Degradation Prediction Under Non-Stationary Aging
+
+- **ID:** `2603.10527` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Kai Chin Lim, Khay Wai See
+- **abs:** https://arxiv.org/abs/2603.10527 · **PDF:** https://arxiv.org/pdf/2603.10527v1
+
+## 5939. AILS-NTUA at SemEval-2026 Task 8: Evaluating Multi-Turn RAG Conversations
+
+- **ID:** `2603.10524` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Dimosthenis Athanasiou, Maria Lymperaiou, Giorgos Filandrianos, et al.
+- **abs:** https://arxiv.org/abs/2603.10524 · **PDF:** https://arxiv.org/pdf/2603.10524v1
+
+## 5940. IH-Challenge: A Training Dataset to Improve Instruction Hierarchy on Frontier LLMs
+
+- **ID:** `2603.10521` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Chuan Guo, Juan Felipe Ceron Uribe, Sicheng Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.10521 · **PDF:** https://arxiv.org/pdf/2603.10521v1
+
+## 5941. Resource-constrained Amazons chess decision framework integrating large language models and graph attention
+
+- **ID:** `2603.10512` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Tianhao Qian, Zhuoxuan Li, Jinde Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.10512 · **PDF:** https://arxiv.org/pdf/2603.10512v1
+
+## 5942. Abjad-Kids: An Arabic Speech Classification Dataset for Primary Education
+
+- **ID:** `2603.20255` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Abdul Aziz Snoubara, Baraa Al_Maradni, Haya Al_Naal, et al.
+- **abs:** https://arxiv.org/abs/2603.20255 · **PDF:** https://arxiv.org/pdf/2603.20255v1
+
+## 5943. Safe and Scalable Web Agent Learning via Recreated Websites
+
+- **ID:** `2603.10505` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Hyungjoo Chae, Jungsoo Park, Alan Ritter
+- **abs:** https://arxiv.org/abs/2603.10505 · **PDF:** https://arxiv.org/pdf/2603.10505v1
+
+## 5944. Naïve Exposure of Generative AI Capabilities Undermines Deepfake Detection
+
+- **ID:** `2603.10504` · **primary:** `cs.CR` · **Submitted:** 2026-03-11
+- **Authors:** Sunpill Kim, Chanwoo Hwang, Minsu Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.10504 · **PDF:** https://arxiv.org/pdf/2603.10504v1
+
+## 5945. A New Tensor Network: Tubal Tensor Train and Its Applications
+
+- **ID:** `2603.10503` · **primary:** `math.NA` · **Submitted:** 2026-03-11
+- **Authors:** Salman Ahmadi-Asl, Valentin Leplat, Anh-Huy Phan, et al.
+- **abs:** https://arxiv.org/abs/2603.10503 · **PDF:** https://arxiv.org/pdf/2603.10503v1
+
+## 5946. Comparative Analysis of Deep Learning Architectures for Multi-Disease Classification of Single-Label Chest X-rays
+
+- **ID:** `2603.13392` · **primary:** `eess.IV` · **Submitted:** 2026-03-11
+- **Authors:** Ali M. Bahram, Saman Muhammad Omer, Hardi M. Mohammed
+- **abs:** https://arxiv.org/abs/2603.13392 · **PDF:** https://arxiv.org/pdf/2603.13392v1
+
+## 5947. Fingerprinting Concepts in Data Streams with Supervised and Unsupervised Meta-Information
+
+- **ID:** `2603.11094` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Ben Halstead, Yun Sing Koh, Patricia Riddle, et al.
+- **abs:** https://arxiv.org/abs/2603.11094 · **PDF:** https://arxiv.org/pdf/2603.11094v1
+
+## 5948. VERI-DPO: Evidence-Aware Alignment for Clinical Summarization via Claim Verification and Direct Preference Optimization
+
+- **ID:** `2603.10494` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Weixin Liu, Congning Ni, Qingyuan Song, et al.
+- **abs:** https://arxiv.org/abs/2603.10494 · **PDF:** https://arxiv.org/pdf/2603.10494v1
+
+## 5949. A Survey of Reasoning in Autonomous Driving Systems: Open Challenges and Emerging Paradigms
+
+- **ID:** `2603.11093` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Kejin Yu, Yuhan Sun, Taiqiang Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.11093 · **PDF:** https://arxiv.org/pdf/2603.11093v1
+
+## 5950. A Universal Nearest-Neighbor Estimator for Intrinsic Dimensionality
+
+- **ID:** `2603.10493` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Eng-Jon Ong, Omer Bobrowski, Gesine Reinert, et al.
+- **abs:** https://arxiv.org/abs/2603.10493 · **PDF:** https://arxiv.org/pdf/2603.10493v1
+
+## 5951. Human-AI Co-reasoning for Clinical Diagnosis with Evidence-Integrated Language Agent
+
+- **ID:** `2603.10492` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Zhongzhen Huang, Yan Ling, Hong Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.10492 · **PDF:** https://arxiv.org/pdf/2603.10492v2
+
+## 5952. JEDI: Jointly Embedded Inference of Neural Dynamics
+
+- **ID:** `2603.10489` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-11
+- **Authors:** Anirudh Jamkhandi, Ali Korojy, Olivier Codol, et al.
+- **abs:** https://arxiv.org/abs/2603.10489 · **PDF:** https://arxiv.org/pdf/2603.10489v1
+
+## 5953. Multi-lingual Multi-institutional Electronic Health Record based Predictive Model
+
+- **ID:** `2604.00027` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Kyunghoon Hur, Heeyoung Kwak, Jinsu Jang, et al.
+- **abs:** https://arxiv.org/abs/2604.00027 · **PDF:** https://arxiv.org/pdf/2604.00027v1
+
+## 5954. Dual Space Preconditioning for Gradient Descent in the Overparameterized Regime
+
+- **ID:** `2603.10485` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** Reza Ghane, Danil Akhtiamov, Babak Hassibi
+- **abs:** https://arxiv.org/abs/2603.10485 · **PDF:** https://arxiv.org/pdf/2603.10485v2
+
+## 5955. "Who Am I, and Who Else Is Here?" Behavioral Differentiation Without Role Assignment in Multi-Agent LLM Systems
+
+- **ID:** `2604.00026` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Houssam EL Kandoussi
+- **abs:** https://arxiv.org/abs/2604.00026 · **PDF:** https://arxiv.org/pdf/2604.00026v1
+
+## 5956. High-Fidelity Text-to-Image Generation from Pre-Trained Vision-Language Models via Distribution-Conditioned Diffusion Decoding
+
+- **ID:** `2603.13389` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Ji Woo Hong, Hee Suk Yoon, Gwanhyeong Koo, et al.
+- **abs:** https://arxiv.org/abs/2603.13389 · **PDF:** https://arxiv.org/pdf/2603.13389v1
+
+## 5957. PEEM: Prompt Engineering Evaluation Metrics for Interpretable Joint Evaluation of Prompts and Responses
+
+- **ID:** `2603.10477` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Minki Hong, Eunsoo Lee, Sohyun Park, et al.
+- **abs:** https://arxiv.org/abs/2603.10477 · **PDF:** https://arxiv.org/pdf/2603.10477v1
+
+## 5958. Learning to Negotiate: Multi-Agent Deliberation for Collective Value Alignment in LLMs
+
+- **ID:** `2603.10476` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Panatchakorn Anantaprayoon, Nataliia Babina, Nima Asgharbeygi, et al.
+- **abs:** https://arxiv.org/abs/2603.10476 · **PDF:** https://arxiv.org/pdf/2603.10476v1
+
+## 5959. Muscle Synergy Priors Enhance Biomechanical Fidelity in Predictive Musculoskeletal Locomotion Simulation
+
+- **ID:** `2603.10474` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Ilseung Park, Eunsik Choi, Jangwhan Ahn, et al.
+- **abs:** https://arxiv.org/abs/2603.10474 · **PDF:** https://arxiv.org/pdf/2603.10474v1
+
+## 5960. Brevity Constraints Reverse Performance Hierarchies in Language Models
+
+- **ID:** `2604.00025` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** MD Azizul Hakim
+- **abs:** https://arxiv.org/abs/2604.00025 · **PDF:** https://arxiv.org/pdf/2604.00025v1
+
+## 5961. Aligning Large Language Models with Searcher Preferences
+
+- **ID:** `2603.10473` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Wei Wu, Peilun Zhou, Liyi Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.10473 · **PDF:** https://arxiv.org/pdf/2603.10473v1
+
+## 5962. Modeling Stage-wise Evolution of User Interests for News Recommendation
+
+- **ID:** `2603.10471` · **primary:** `cs.IR` · **Submitted:** 2026-03-11
+- **Authors:** Zhiyong Cheng, Yike Jin, Zhijie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.10471 · **PDF:** https://arxiv.org/pdf/2603.10471v1
+
+## 5963. G-STAR: End-to-End Global Speaker-Tracking Attributed Recognition
+
+- **ID:** `2603.10468` · **primary:** `eess.AS` · **Submitted:** 2026-03-11
+- **Authors:** Jing Peng, Ziyi Chen, Haoyu Li, et al.
+- **abs:** https://arxiv.org/abs/2603.10468 · **PDF:** https://arxiv.org/pdf/2603.10468v1
+
+## 5964. AI Detectors Fail Diverse Student Populations: A Mathematical Framing of Structural Detection Limits
+
+- **ID:** `2603.20254` · **primary:** `cs.CY` · **Submitted:** 2026-03-11
+- **Authors:** Nathan Garland
+- **abs:** https://arxiv.org/abs/2603.20254 · **PDF:** https://arxiv.org/pdf/2603.20254v1
+
+## 5965. UniPINN: A Unified PINN Framework for Multi-task Learning of Diverse Navier-Stokes Equations
+
+- **ID:** `2603.10466` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Dengdi Sun, Jie Chen, Xiao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.10466 · **PDF:** https://arxiv.org/pdf/2603.10466v1
+
+## 5966. WHBench: Evaluating Frontier LLMs with Expert-in-the-Loop Validation on Women's Health Topics
+
+- **ID:** `2604.00024` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Sneha Maurya, Pragya Saboo, Girish Kumar
+- **abs:** https://arxiv.org/abs/2604.00024 · **PDF:** https://arxiv.org/pdf/2604.00024v1
+
+## 5967. Beam-Plasma Collective Oscillations in Intense Charged-Particle Beams: Dielectric Response Theory, Langmuir Wave Dispersion, and Unsupervised Detection via Prometheus
+
+- **ID:** `2603.10457` · **primary:** `physics.plasm-ph` · **Submitted:** 2026-03-11
+- **Authors:** Brandon Yee, Wilson Collins, Michael Iofin, et al.
+- **abs:** https://arxiv.org/abs/2603.10457 · **PDF:** https://arxiv.org/pdf/2603.10457v2
+
+## 5968. Interventional Time Series Priors for Causal Foundation Models
+
+- **ID:** `2603.11090` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Dennis Thumm, Ying Chen
+- **abs:** https://arxiv.org/abs/2603.11090 · **PDF:** https://arxiv.org/pdf/2603.11090v1
+
+## 5969. Spatio-Temporal Forecasting of Retaining Wall Deformation: Mitigating Error Accumulation via Multi-Resolution ConvLSTM Stacking Ensemble
+
+- **ID:** `2603.10453` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Jihoon Kim, Heejung Youn
+- **abs:** https://arxiv.org/abs/2603.10453 · **PDF:** https://arxiv.org/pdf/2603.10453v1
+
+## 5970. Brenier Isotonic Regression
+
+- **ID:** `2603.10452` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** Han Bao, Amirreza Eshraghi, Yutong Wang
+- **abs:** https://arxiv.org/abs/2603.10452 · **PDF:** https://arxiv.org/pdf/2603.10452v1
+
+## 5971. FAR-Dex: Few-shot Data Augmentation and Adaptive Residual Policy Refinement for Dexterous Manipulation
+
+- **ID:** `2603.10451` · **primary:** `cs.RO` · **Submitted:** 2026-03-11
+- **Authors:** Yushan Bai, Fulin Chen, Hongzheng Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.10451 · **PDF:** https://arxiv.org/pdf/2603.10451v1
+
+## 5972. Unlearning the Unpromptable: Prompt-free Instance Unlearning in Diffusion Models
+
+- **ID:** `2603.10445` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Kyungryeol Lee, Kyeonghyun Lee, Seongmin Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.10445 · **PDF:** https://arxiv.org/pdf/2603.10445v2
+
+## 5973. The Curse and Blessing of Mean Bias in FP4-Quantized LLM Training
+
+- **ID:** `2603.10444` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Hengjie Cao, Zhendong Huang, Mengyi Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.10444 · **PDF:** https://arxiv.org/pdf/2603.10444v1
+
+## 5974. PRIME-CVD: A Parametrically Rendered Informatics Medical Environment for Education in Cardiovascular Risk Modelling
+
+- **ID:** `2603.19299` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Nicholas I-Hsien Kuo, Marzia Hoque Tania, Blanca Gallego, et al.
+- **abs:** https://arxiv.org/abs/2603.19299 · **PDF:** https://arxiv.org/pdf/2603.19299v1
+
+## 5975. GGMPs: Generalized Gaussian Mixture Processes
+
+- **ID:** `2603.10442` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Vardaan Tekriwal, Mark D. Risser, Hengrui Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.10442 · **PDF:** https://arxiv.org/pdf/2603.10442v1
+
+## 5976. VisualLeakBench: Auditing the Fragility of Large Vision-Language Models against PII Leakage and Social Engineering
+
+- **ID:** `2603.13385` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Youting Wang, Yuan Tang, Yitian Qian, et al.
+- **abs:** https://arxiv.org/abs/2603.13385 · **PDF:** https://arxiv.org/pdf/2603.13385v1
+
+## 5977. A Dynamic Bayesian and Machine Learning Framework for Quantitative Evaluation and Prediction of Operator Situation Awareness in Nuclear Power Plants
+
+- **ID:** `2603.19298` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Shuai Chen, Huiqiao Jia, Tao Qing, et al.
+- **abs:** https://arxiv.org/abs/2603.19298 · **PDF:** https://arxiv.org/pdf/2603.19298v2
+
+## 5978. Adaptive Active Learning for Regression via Reinforcement Learning
+
+- **ID:** `2603.10435` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** Simon D. Nguyen, Troy Russo, Kentaro Hoffman, et al.
+- **abs:** https://arxiv.org/abs/2603.10435 · **PDF:** https://arxiv.org/pdf/2603.10435v1
+
+## 5979. The Attack and Defense Landscape of Agentic AI: A Comprehensive Survey
+
+- **ID:** `2603.11088` · **primary:** `cs.CR` · **Submitted:** 2026-03-11
+- **Authors:** Juhee Kim, Xiaoyuan Liu, Zhun Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.11088 · **PDF:** https://arxiv.org/pdf/2603.11088v1
+
+## 5980. Domain-Adaptive Health Indicator Learning with Degradation-Stage Synchronized Sampling and Cross-Domain Autoencoder
+
+- **ID:** `2603.10430` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Jungho Choo, Hanbyeol Park, Gawon Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.10430 · **PDF:** https://arxiv.org/pdf/2603.10430v1
+
+## 5981. Phonological Fossils: Machine Learning Detection of Non-Mainstream Vocabulary in Sulawesi Basic Lexicon
+
+- **ID:** `2604.00023` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Mukhlis Amien, Go Frendi Gunawan
+- **abs:** https://arxiv.org/abs/2604.00023 · **PDF:** https://arxiv.org/pdf/2604.00023v1
+
+## 5982. SimulCost: A Cost-Aware Benchmark and Toolkit for Automating Physics Simulations with LLMs
+
+- **ID:** `2603.20253` · **primary:** `physics.comp-ph` · **Submitted:** 2026-03-11
+- **Authors:** Yadi Cao, Sicheng Lai, Jiahe Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.20253 · **PDF:** https://arxiv.org/pdf/2603.20253v1
+
+## 5983. VulnAgent-X: A Layered Agentic Framework for Repository-Level Vulnerability Detection
+
+- **ID:** `2603.13384` · **primary:** `cs.SE` · **Submitted:** 2026-03-11
+- **Authors:** Renwei Meng, Haoyi Wu, Jingming Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.13384 · **PDF:** https://arxiv.org/pdf/2603.13384v1
+
+## 5984. Enhancing Network Intrusion Detection Systems: A Multi-Layer Ensemble Approach to Mitigate Adversarial Attacks
+
+- **ID:** `2603.10413` · **primary:** `cs.CR` · **Submitted:** 2026-03-11
+- **Authors:** Nasim Soltani, Shayan Nejadshamsi, Zakaria Abou El Houda, et al.
+- **abs:** https://arxiv.org/abs/2603.10413 · **PDF:** https://arxiv.org/pdf/2603.10413v1
+
+## 5985. Effective Dataset Distillation for Spatio-Temporal Forecasting with Bi-dimensional Compression
+
+- **ID:** `2603.10410` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Taehyung Kwon, Yeonje Choi, Yeongho Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.10410 · **PDF:** https://arxiv.org/pdf/2603.10410v1
+
+## 5986. Criterion Validity of LLM-as-Judge for Business Outcomes in Conversational Commerce
+
+- **ID:** `2604.00022` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Liang Chen, Qi Liu, Wenhuan Lin, et al.
+- **abs:** https://arxiv.org/abs/2604.00022 · **PDF:** https://arxiv.org/pdf/2604.00022v1
+
+## 5987. FinReflectKG -- HalluBench: GraphRAG Hallucination Benchmark for Financial Question Answering Systems
+
+- **ID:** `2603.20252` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Mahesh Kumar, Bhaskarjit Sarmah, Stefano Pasquali
+- **abs:** https://arxiv.org/abs/2603.20252 · **PDF:** https://arxiv.org/pdf/2603.20252v1
+
+## 5988. Designing Service Systems from Textual Evidence
+
+- **ID:** `2603.10400` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Ruicheng Ao, Hongyu Chen, Siyang Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.10400 · **PDF:** https://arxiv.org/pdf/2603.10400v1
+
+## 5989. On the Learning Dynamics of Two-layer Linear Networks with Label Noise SGD
+
+- **ID:** `2603.10397` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Tongcheng Zhang, Zhanpeng Zhou, Mingze Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.10397 · **PDF:** https://arxiv.org/pdf/2603.10397v1
+
+## 5990. Verbalizing LLM's Higher-order Uncertainty via Imprecise Probabilities
+
+- **ID:** `2603.10396` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Anita Yang, Krikamol Muandet, Michele Caprio, et al.
+- **abs:** https://arxiv.org/abs/2603.10396 · **PDF:** https://arxiv.org/pdf/2603.10396v1
+
+## 5991. Graph-GRPO: Training Graph Flow Models with Reinforcement Learning
+
+- **ID:** `2603.10395` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Baoheng Zhu, Deyu Bo, Delvin Ce Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.10395 · **PDF:** https://arxiv.org/pdf/2603.10395v1
+
+## 5992. Safe Probabilistic Planning for Human-Robot Interaction using Conformal Risk Control
+
+- **ID:** `2603.10392` · **primary:** `cs.RO` · **Submitted:** 2026-03-11
+- **Authors:** Jake Gonzales, Kazuki Mizuta, Karen Leung, et al.
+- **abs:** https://arxiv.org/abs/2603.10392 · **PDF:** https://arxiv.org/pdf/2603.10392v1
+
+## 5993. Variance-Aware Adaptive Weighting for Diffusion Model Training
+
+- **ID:** `2603.10391` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Nanlong Sun, Lei Shi
+- **abs:** https://arxiv.org/abs/2603.10391 · **PDF:** https://arxiv.org/pdf/2603.10391v1
+
+## 5994. CLaRE-ty Amid Chaos: Quantifying Representational Entanglement to Predict Ripple Effects in LLM Editing
+
+- **ID:** `2603.19297` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Manit Baser, Alperen Yildiz, Dinil Mon Divakaran, et al.
+- **abs:** https://arxiv.org/abs/2603.19297 · **PDF:** https://arxiv.org/pdf/2603.19297v1
+
+## 5995. Beyond Scalars: Evaluating and Understanding LLM Reasoning via Geometric Progress and Stability
+
+- **ID:** `2603.10384` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Xinyan Jiang, Ninghao Liu, Di Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.10384 · **PDF:** https://arxiv.org/pdf/2603.10384v1
+
+## 5996. Optimal Expert-Attention Allocation in Mixture-of-Experts: A Scalable Law for Dynamic Model Design
+
+- **ID:** `2603.10379` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Junzhuo Li, Peijie Jiang, Changxin Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.10379 · **PDF:** https://arxiv.org/pdf/2603.10379v1
+
+## 5997. Causal Concept Graphs in LLM Latent Space for Stepwise Reasoning
+
+- **ID:** `2603.10377` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Md Muntaqim Meherab, Noor Islam S. Mohammad, Faiza Feroz
+- **abs:** https://arxiv.org/abs/2603.10377 · **PDF:** https://arxiv.org/pdf/2603.10377v1
+
+## 5998. Reactive Writers: How Co-Writing with AI Changes How We Engage with Ideas
+
+- **ID:** `2603.10374` · **primary:** `cs.HC` · **Submitted:** 2026-03-11
+- **Authors:** Advait Bhat, Marianne Aubin Le Quéré, Mor Naaman, et al.
+- **abs:** https://arxiv.org/abs/2603.10374 · **PDF:** https://arxiv.org/pdf/2603.10374v1
+
+## 5999. Few-Shot Adaptation to Non-Stationary Environments via Latent Trend Embedding for Robotics
+
+- **ID:** `2603.10373` · **primary:** `cs.RO` · **Submitted:** 2026-03-11
+- **Authors:** Yasuyuki Fujii, Emika Kameda, Hiroki Fukada, et al.
+- **abs:** https://arxiv.org/abs/2603.10373 · **PDF:** https://arxiv.org/pdf/2603.10373v1
+
+## 6000. Speech Codec Probing from Semantic and Phonetic Perspectives
+
+- **ID:** `2603.10371` · **primary:** `eess.AS` · **Submitted:** 2026-03-11
+- **Authors:** Xuan Shi, Chang Zeng, Tiantian Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.10371 · **PDF:** https://arxiv.org/pdf/2603.10371v1
+
+## 6001. Beyond Interleaving: Causal Attention Reformulations for Generative Recommender Systems
+
+- **ID:** `2603.10369` · **primary:** `cs.IR` · **Submitted:** 2026-03-11
+- **Authors:** Hailing Cheng
+- **abs:** https://arxiv.org/abs/2603.10369 · **PDF:** https://arxiv.org/pdf/2603.10369v1
+
+## 6002. Dynamic Knowledge Fusion for Multi-Domain Dialogue State Tracking
+
+- **ID:** `2603.10367` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Haoxiang Su, Ruiyu Fang, Liting Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.10367 · **PDF:** https://arxiv.org/pdf/2603.10367v1
+
+## 6003. How Do Language Models Process Ethical Instructions? Deliberation, Consistency, and Other-Recognition Across Four Models
+
+- **ID:** `2604.00021` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Hiroki Fukui
+- **abs:** https://arxiv.org/abs/2604.00021 · **PDF:** https://arxiv.org/pdf/2604.00021v1
+
+## 6004. Beyond Linearity in Attention Projections: The Case for Nonlinear Queries
+
+- **ID:** `2603.13381` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Marko Karbevski
+- **abs:** https://arxiv.org/abs/2603.13381 · **PDF:** https://arxiv.org/pdf/2603.13381v1
+
+## 6005. HEAL: Hindsight Entropy-Assisted Learning for Reasoning Distillation
+
+- **ID:** `2603.10359` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Wenjing Zhang, Jiangze Yan, Jieyun Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.10359 · **PDF:** https://arxiv.org/pdf/2603.10359v1
+
+## 6006. Utility Function is All You Need: LLM-based Congestion Control
+
+- **ID:** `2603.10357` · **primary:** `cs.NI` · **Submitted:** 2026-03-11
+- **Authors:** Neta Rozen-Schiff, Liron Schiff, Stefan Schmid
+- **abs:** https://arxiv.org/abs/2603.10357 · **PDF:** https://arxiv.org/pdf/2603.10357v1
+
+## 6007. Detecting Abnormal User Feedback Patterns through Temporal Sentiment Aggregation
+
+- **ID:** `2604.00020` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Yalun Qi, Sichen Zhao, Zhiming Xue, et al.
+- **abs:** https://arxiv.org/abs/2604.00020 · **PDF:** https://arxiv.org/pdf/2604.00020v1
+
+## 6008. Mitigating Translationese Bias in Multilingual LLM-as-a-Judge via Disentangled Information Bottleneck
+
+- **ID:** `2603.10351` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Hongbin Zhang, Kehai Chen, Xuefen Bai, et al.
+- **abs:** https://arxiv.org/abs/2603.10351 · **PDF:** https://arxiv.org/pdf/2603.10351v1
+
+## 6009. On The Complexity of Best-Arm Identification in Non-Stationary Linear Bandits
+
+- **ID:** `2603.10346` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** Leo Maynard-Zhang, Zhihan Xiong, Kevin Jamieson, et al.
+- **abs:** https://arxiv.org/abs/2603.10346 · **PDF:** https://arxiv.org/pdf/2603.10346v1
+
+## 6010. Federated Active Learning Under Extreme Non-IID and Global Class Imbalance
+
+- **ID:** `2603.10341` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Chen-Chen Zong, Sheng-Jun Huang
+- **abs:** https://arxiv.org/abs/2603.10341 · **PDF:** https://arxiv.org/pdf/2603.10341v1
+
+## 6011. Overcoming Visual Clutter in Vision Language Action Models via Concept-Gated Visual Distillation
+
+- **ID:** `2603.10340` · **primary:** `cs.CV` · **Submitted:** 2026-03-11
+- **Authors:** Sangmim Song, Sarath Kodagoda, Marc Carmichael, et al.
+- **abs:** https://arxiv.org/abs/2603.10340 · **PDF:** https://arxiv.org/pdf/2603.10340v1
+
+## 6012. TTQ: Activation-Aware Test-Time Quantization to Accelerate LLM Inference On The Fly
+
+- **ID:** `2603.19296` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Toshiaki Koike-Akino, Jing Liu, Ye Wang
+- **abs:** https://arxiv.org/abs/2603.19296 · **PDF:** https://arxiv.org/pdf/2603.19296v1
+
+## 6013. Querying Everything Everywhere All at Once: Supervaluationism for the Agentic Lakehouse
+
+- **ID:** `2603.13380` · **primary:** `cs.DB` · **Submitted:** 2026-03-11
+- **Authors:** Jacopo Tagliabue
+- **abs:** https://arxiv.org/abs/2603.13380 · **PDF:** https://arxiv.org/pdf/2603.13380v1
+
+## 6014. Does Reasoning Make Search More Fair? Comparing Fairness in Reasoning and Non-Reasoning Rerankers
+
+- **ID:** `2603.10332` · **primary:** `cs.IR` · **Submitted:** 2026-03-11
+- **Authors:** Saron Samuel, Benjamin Van Durme, Eugene Yang
+- **abs:** https://arxiv.org/abs/2603.10332 · **PDF:** https://arxiv.org/pdf/2603.10332v1
+
+## 6015. PC-Diffuser: Path-Consistent Capsule CBF Safety Filtering for Diffusion-Based Trajectory Planner
+
+- **ID:** `2603.10330` · **primary:** `cs.RO` · **Submitted:** 2026-03-11
+- **Authors:** Eugene Ku, Yiwei Lyu
+- **abs:** https://arxiv.org/abs/2603.10330 · **PDF:** https://arxiv.org/pdf/2603.10330v1
+
+## 6016. NasoVoce: A Nose-Mounted Low-Audibility Speech Interface for Always-Available Speech Interaction
+
+- **ID:** `2603.10324` · **primary:** `cs.HC` · **Submitted:** 2026-03-11
+- **Authors:** Jun Rekimoto, Yu Nishimura, Bojian Yang
+- **abs:** https://arxiv.org/abs/2603.10324 · **PDF:** https://arxiv.org/pdf/2603.10324v1
+
+## 6017. BrainSCL: Subtype-Guided Contrastive Learning for Brain Disorder Diagnosis
+
+- **ID:** `2603.19295` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Xiaolong Li, Guiliang Guo, Guangqi Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.19295 · **PDF:** https://arxiv.org/pdf/2603.19295v1
+
+## 6018. Large language models can disambiguate opioid slang on social media
+
+- **ID:** `2603.10313` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Kristy A. Carpenter, Issah A. Samori, Mathew V. Kiang, et al.
+- **abs:** https://arxiv.org/abs/2603.10313 · **PDF:** https://arxiv.org/pdf/2603.10313v1
+
+## 6019. Data-Driven Integration Kernels for Interpretable Nonlocal Operator Learning
+
+- **ID:** `2603.10305` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Savannah L. Ferretti, Jerry Lin, Sara Shamekh, et al.
+- **abs:** https://arxiv.org/abs/2603.10305 · **PDF:** https://arxiv.org/pdf/2603.10305v2
+
+## 6020. The Chronicles of RiDiC: Generating Datasets with Controlled Popularity Distribution for Long-form Factuality Evaluation
+
+- **ID:** `2604.00019` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Pavel Braslavski, Dmitrii Iarosh, Nikita Sushko, et al.
+- **abs:** https://arxiv.org/abs/2604.00019 · **PDF:** https://arxiv.org/pdf/2604.00019v1
+
+## 6021. Is this Idea Novel? An Automated Benchmark for Judgment of Research Ideas
+
+- **ID:** `2603.10303` · **primary:** `cs.CL` · **Submitted:** 2026-03-11
+- **Authors:** Tim Schopf, Michael Färber
+- **abs:** https://arxiv.org/abs/2603.10303 · **PDF:** https://arxiv.org/pdf/2603.10303v1
+
+## 6022. How to make the most of your masked language model for protein engineering
+
+- **ID:** `2603.10302` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Calvin McCarter, Nick Bhattacharya, Sebastian W. Ober, et al.
+- **abs:** https://arxiv.org/abs/2603.10302 · **PDF:** https://arxiv.org/pdf/2603.10302v1
+
+## 6023. What do near-optimal learning rate schedules look like?
+
+- **ID:** `2603.10301` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Hiroki Naganuma, Atish Agarwala, Priya Kasimbeg, et al.
+- **abs:** https://arxiv.org/abs/2603.10301 · **PDF:** https://arxiv.org/pdf/2603.10301v2
+
+## 6024. Regime-aware financial volatility forecasting via in-context learning
+
+- **ID:** `2603.10299` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Saba Asaad, Shayan Mohajer Hamidi, Ali Bereyhi
+- **abs:** https://arxiv.org/abs/2603.10299 · **PDF:** https://arxiv.org/pdf/2603.10299v1
+
+## 6025. GaLoRA: Parameter-Efficient Graph-Aware LLMs for Node Classification
+
+- **ID:** `2603.10298` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Mayur Choudhary, Saptarshi Sengupta, Katerina Potika
+- **abs:** https://arxiv.org/abs/2603.10298 · **PDF:** https://arxiv.org/pdf/2603.10298v1
+
+## 6026. Simulation-in-the-Reasoning (SiR): A Conceptual Framework for Empirically Grounded AI in Autonomous Transportation
+
+- **ID:** `2603.10294` · **primary:** `eess.SY` · **Submitted:** 2026-03-11
+- **Authors:** Wuping Xin
+- **abs:** https://arxiv.org/abs/2603.10294 · **PDF:** https://arxiv.org/pdf/2603.10294v1
+
+## 6027. Hybrid Self-evolving Structured Memory for GUI Agents
+
+- **ID:** `2603.10291` · **primary:** `cs.AI` · **Submitted:** 2026-03-11
+- **Authors:** Sibo Zhu, Wenyi Wu, Kun Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.10291 · **PDF:** https://arxiv.org/pdf/2603.10291v1
+
+## 6028. Quantum entanglement provides a competitive advantage in adversarial games
+
+- **ID:** `2603.10289` · **primary:** `quant-ph` · **Submitted:** 2026-03-11
+- **Authors:** Peiyong Wang, Kieran Hymas, James Quach
+- **abs:** https://arxiv.org/abs/2603.10289 · **PDF:** https://arxiv.org/pdf/2603.10289v1
+
+## 6029. MultiwayPAM: Multiway Partitioning Around Medoids for LLM-as-a-Judge Score Analysis
+
+- **ID:** `2603.10287` · **primary:** `stat.ML` · **Submitted:** 2026-03-11
+- **Authors:** Chihiro Watanabe, Jingyu Sun
+- **abs:** https://arxiv.org/abs/2603.10287 · **PDF:** https://arxiv.org/pdf/2603.10287v1
+
+## 6030. Conversational AI-Enhanced Exploration System to Query Large-Scale Digitised Collections of Natural History Museums
+
+- **ID:** `2603.10285` · **primary:** `cs.HC` · **Submitted:** 2026-03-11
+- **Authors:** Yiyuan Wang, Andrew Johnston, Zoë Sadokierski, et al.
+- **abs:** https://arxiv.org/abs/2603.10285 · **PDF:** https://arxiv.org/pdf/2603.10285v1
+
+## 6031. Copula-ResLogit: A Deep-Copula Framework for Unobserved Confounding Effects
+
+- **ID:** `2603.10284` · **primary:** `cs.LG` · **Submitted:** 2026-03-11
+- **Authors:** Kimia Kamal, Bilal Farooq
+- **abs:** https://arxiv.org/abs/2603.10284 · **PDF:** https://arxiv.org/pdf/2603.10284v1
+
+## 6032. GSVD for Geometry-Grounded Dataset Comparison: An Alignment Angle Is All You Need
+
+- **ID:** `2603.10283` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Eduarda de Souza Marques, Arthur Sobrinho Ferreira da Rocha, Joao Paixao, et al.
+- **abs:** https://arxiv.org/abs/2603.10283 · **PDF:** https://arxiv.org/pdf/2603.10283v1
+
+## 6033. Taming Score-Based Denoisers in ADMM: A Convergent Plug-and-Play Framework
+
+- **ID:** `2603.10281` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Rajesh Shrestha, Xiao Fu
+- **abs:** https://arxiv.org/abs/2603.10281 · **PDF:** https://arxiv.org/pdf/2603.10281v2
+
+## 6034. Quality-Driven Agentic Reasoning for LLM-Assisted Software Design: Questions-of-Thoughts (QoT) as a Time-Series Self-QA Chain
+
+- **ID:** `2603.11082` · **primary:** `cs.SE` · **Submitted:** 2026-03-10
+- **Authors:** Yen-Ku Liu, Yun-Cheng Tsai
+- **abs:** https://arxiv.org/abs/2603.11082 · **PDF:** https://arxiv.org/pdf/2603.11082v1
+
+## 6035. Robust Post-Training for Generative Recommenders: Why Exponential Reward-Weighted SFT Outperforms RLHF
+
+- **ID:** `2603.10279` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Keertana Chidambaram, Sanath Kumar Krishnamurthy, Qiuling Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.10279 · **PDF:** https://arxiv.org/pdf/2603.10279v1
+
+## 6036. Estimating condition number with Graph Neural Networks
+
+- **ID:** `2603.10277` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Erin Carson, Xinye Chen
+- **abs:** https://arxiv.org/abs/2603.10277 · **PDF:** https://arxiv.org/pdf/2603.10277v2
+
+## 6037. Semantic Chameleon: Corpus-Dependent Poisoning Attacks and Defenses in RAG Systems
+
+- **ID:** `2603.18034` · **primary:** `cs.CR` · **Submitted:** 2026-03-10
+- **Authors:** Scott Thornton
+- **abs:** https://arxiv.org/abs/2603.18034 · **PDF:** https://arxiv.org/pdf/2603.18034v1
+
+## 6038. Think Twice Before You Write -- an Entropy-based Decoding Strategy to Enhance LLM Reasoning
+
+- **ID:** `2604.00018` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Jiashu He, Meizhu Liu, Olaitan P Olaleye, et al.
+- **abs:** https://arxiv.org/abs/2604.00018 · **PDF:** https://arxiv.org/pdf/2604.00018v1
+
+## 6039. Interview-Informed Generative Agents for Product Discovery: A Validation Study
+
+- **ID:** `2603.29890` · **primary:** `cs.HC` · **Submitted:** 2026-03-10
+- **Authors:** Zichao Wang, Alexa Siu
+- **abs:** https://arxiv.org/abs/2603.29890 · **PDF:** https://arxiv.org/pdf/2603.29890v1
+
+## 6040. From Prior to Pro: Efficient Skill Mastery via Distribution Contractive RL Finetuning
+
+- **ID:** `2603.10263` · **primary:** `cs.RO` · **Submitted:** 2026-03-10
+- **Authors:** Zhanyi Sun, Shuran Song
+- **abs:** https://arxiv.org/abs/2603.10263 · **PDF:** https://arxiv.org/pdf/2603.10263v1
+
+## 6041. Discovery of a Hematopoietic Manifold in scGPT Yields a Method for Extracting Performant Algorithms from Biological Foundation Model Internals
+
+- **ID:** `2603.10261` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Ihor Kendiukhov
+- **abs:** https://arxiv.org/abs/2603.10261 · **PDF:** https://arxiv.org/pdf/2603.10261v1
+
+## 6042. Improving TabPFN's Synthetic Data Generation by Integrating Causal Structure
+
+- **ID:** `2603.10254` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Davide Tugnoli, Andrea De Lorenzo, Marco Virgolin, et al.
+- **abs:** https://arxiv.org/abs/2603.10254 · **PDF:** https://arxiv.org/pdf/2603.10254v1
+
+## 6043. Joint Imaging-ROI Representation Learning via Cross-View Contrastive Alignment for Brain Disorder Classification
+
+- **ID:** `2603.10253` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Wei Liang, Lifang He
+- **abs:** https://arxiv.org/abs/2603.10253 · **PDF:** https://arxiv.org/pdf/2603.10253v1
+
+## 6044. Bayesian Hierarchical Models and the Maximum Entropy Principle
+
+- **ID:** `2603.10252` · **primary:** `stat.ML` · **Submitted:** 2026-03-10
+- **Authors:** Brendon J. Brewer
+- **abs:** https://arxiv.org/abs/2603.10252 · **PDF:** https://arxiv.org/pdf/2603.10252v1
+
+## 6045. SiMPO: Measure Matching for Online Diffusion Reinforcement Learning
+
+- **ID:** `2603.10250` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Haitong Ma, Chenxiao Gao, Tianyi Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.10250 · **PDF:** https://arxiv.org/pdf/2603.10250v1
+
+## 6046. DUCTILE: Agentic LLM Orchestration of Engineering Analysis in Product Development Practice
+
+- **ID:** `2603.10249` · **primary:** `cs.SE` · **Submitted:** 2026-03-10
+- **Authors:** Alejandro Pradas-Gomez, Arindam Brahma, Ola Isaksson
+- **abs:** https://arxiv.org/abs/2603.10249 · **PDF:** https://arxiv.org/pdf/2603.10249v2
+
+## 6047. Intrinsic Numerical Robustness and Fault Tolerance in a Neuromorphic Algorithm for Scientific Computing
+
+- **ID:** `2603.10246` · **primary:** `cs.NE` · **Submitted:** 2026-03-10
+- **Authors:** Bradley H. Theilman, James B. Aimone
+- **abs:** https://arxiv.org/abs/2603.10246 · **PDF:** https://arxiv.org/pdf/2603.10246v1
+
+## 6048. GR-SAP: Generative Replay for Safety Alignment Preservation during Fine-Tuning
+
+- **ID:** `2603.10243` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Zhouxiang Fang, Jiawei Zhou, Hanjie Chen
+- **abs:** https://arxiv.org/abs/2603.10243 · **PDF:** https://arxiv.org/pdf/2603.10243v1
+
+## 6049. CR-Bench: Evaluating the Real-World Utility of AI Code Review Agents
+
+- **ID:** `2603.11078` · **primary:** `cs.SE` · **Submitted:** 2026-03-10
+- **Authors:** Kristen Pereira, Neelabh Sinha, Rajat Ghosh, et al.
+- **abs:** https://arxiv.org/abs/2603.11078 · **PDF:** https://arxiv.org/pdf/2603.11078v1
+
+## 6050. Semantic Shifts of Psychological Concepts in Scientific and Popular Media Discourse: A Distributional Semantics Analysis of Russian-Language Corpora
+
+- **ID:** `2604.00017` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Orlova Anastasia
+- **abs:** https://arxiv.org/abs/2604.00017 · **PDF:** https://arxiv.org/pdf/2604.00017v1
+
+## 6051. Learning from Radio using Variational Quantum RF Sensing
+
+- **ID:** `2603.10239` · **primary:** `quant-ph` · **Submitted:** 2026-03-10
+- **Authors:** Ivana Nikoloska
+- **abs:** https://arxiv.org/abs/2603.10239 · **PDF:** https://arxiv.org/pdf/2603.10239v1
+
+## 6052. One Adapter for All: Towards Unified Representation in Step-Imbalanced Class-Incremental Learning
+
+- **ID:** `2603.10237` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Xiaoyan Zhang, Jiangpeng He
+- **abs:** https://arxiv.org/abs/2603.10237 · **PDF:** https://arxiv.org/pdf/2603.10237v1
+
+## 6053. A Hierarchical End-of-Turn Model with Primary Speaker Segmentation for Real-Time Conversational AI
+
+- **ID:** `2603.13379` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Karim Helwani, Hoang Do, James Luan, et al.
+- **abs:** https://arxiv.org/abs/2603.13379 · **PDF:** https://arxiv.org/pdf/2603.13379v1
+
+## 6054. Are they human? Detecting large language models by probing human memory constraints
+
+- **ID:** `2604.00016` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Simon Schug, Brenden M. Lake
+- **abs:** https://arxiv.org/abs/2604.00016 · **PDF:** https://arxiv.org/pdf/2604.00016v1
+
+## 6055. Why Does It Look There? Structured Explanations for Image Classification
+
+- **ID:** `2603.10234` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Jiarui Li, Zixiang Yin, Samuel J Landry, et al.
+- **abs:** https://arxiv.org/abs/2603.10234 · **PDF:** https://arxiv.org/pdf/2603.10234v1
+
+## 6056. S-GRADES -- Studying Generalization of Student Response Assessments in Diverse Evaluative Settings
+
+- **ID:** `2603.10233` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Tasfia Seuti, Sagnik Ray Choudhury
+- **abs:** https://arxiv.org/abs/2603.10233 · **PDF:** https://arxiv.org/pdf/2603.10233v1
+
+## 6057. A Trust-Region Interior-Point Stochastic Sequential Quadratic Programming Method
+
+- **ID:** `2603.10230` · **primary:** `math.OC` · **Submitted:** 2026-03-10
+- **Authors:** Yuchen Fang, Jihun Kim, Sen Na, et al.
+- **abs:** https://arxiv.org/abs/2603.10230 · **PDF:** https://arxiv.org/pdf/2603.10230v1
+
+## 6058. Maximizing mutual information between user-contexts and responses improve LLM personalization with no additional data
+
+- **ID:** `2603.19294` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Hyunji Nam, Haoran Li, Natasha Jaques
+- **abs:** https://arxiv.org/abs/2603.19294 · **PDF:** https://arxiv.org/pdf/2603.19294v1
+
+## 6059. DIVE: Scaling Diversity in Agentic Task Synthesis for Generalizable Tool Use
+
+- **ID:** `2603.11076` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Aili Chen, Chi Zhang, Junteng Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.11076 · **PDF:** https://arxiv.org/pdf/2603.11076v1
+
+## 6060. Privacy and Safety Experiences and Concerns of U.S. Women Using Generative AI for Seeking Sexual and Reproductive Health Information
+
+- **ID:** `2603.16918` · **primary:** `cs.HC` · **Submitted:** 2026-03-10
+- **Authors:** Ina Kaleva, Xiao Zhan, Ruba Abu-Salma, et al.
+- **abs:** https://arxiv.org/abs/2603.16918 · **PDF:** https://arxiv.org/pdf/2603.16918v1
+
+## 6061. Rethinking the Harmonic Loss via Non-Euclidean Distance Layers
+
+- **ID:** `2603.10225` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Maxwell Miller-Golub, Collin Coil, Kamil Faber, et al.
+- **abs:** https://arxiv.org/abs/2603.10225 · **PDF:** https://arxiv.org/pdf/2603.10225v2
+
+## 6062. Do Large Language Models Get Caught in Hofstadter-Mobius Loops?
+
+- **ID:** `2603.13378` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Jaroslaw Hryszko
+- **abs:** https://arxiv.org/abs/2603.13378 · **PDF:** https://arxiv.org/pdf/2603.13378v1
+
+## 6063. Unifying Logical and Physical Layout Representations via Heterogeneous Graphs for Circuit Congestion Prediction
+
+- **ID:** `2603.11075` · **primary:** `cs.AR` · **Submitted:** 2026-03-10
+- **Authors:** Runbang Hu, Bo Fang, Bingzhe Li, et al.
+- **abs:** https://arxiv.org/abs/2603.11075 · **PDF:** https://arxiv.org/pdf/2603.11075v1
+
+## 6064. Robotic Ultrasound Makes CBCT Alive
+
+- **ID:** `2603.10220` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Feng Li, Ziyuan Li, Zhongliang Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.10220 · **PDF:** https://arxiv.org/pdf/2603.10220v1
+
+## 6065. A Diffusion Analysis of Policy Gradient for Stochastic Bandits
+
+- **ID:** `2603.10219` · **primary:** `stat.ML` · **Submitted:** 2026-03-10
+- **Authors:** Tor Lattimore
+- **abs:** https://arxiv.org/abs/2603.10219 · **PDF:** https://arxiv.org/pdf/2603.10219v1
+
+## 6066. HoloByte: Continuous Hyperspherical Distillation for Tokenizer-Free Modeling
+
+- **ID:** `2603.16917` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Vladimer Khasia
+- **abs:** https://arxiv.org/abs/2603.16917 · **PDF:** https://arxiv.org/pdf/2603.16917v1
+
+## 6067. Multilingual AI-Driven Password Strength Estimation with Similarity-Based Detection
+
+- **ID:** `2603.10217` · **primary:** `cs.CR` · **Submitted:** 2026-03-10
+- **Authors:** Nikitha M. Palaniappan, Ying He
+- **abs:** https://arxiv.org/abs/2603.10217 · **PDF:** https://arxiv.org/pdf/2603.10217v1
+
+## 6068. SDSR: A Spectral Divide-and-Conquer Approach for Species Tree Reconstruction
+
+- **ID:** `2603.10215` · **primary:** `q-bio.PE` · **Submitted:** 2026-03-10
+- **Authors:** Ortal Reshef, Ofer Glassman, Or Zuk, et al.
+- **abs:** https://arxiv.org/abs/2603.10215 · **PDF:** https://arxiv.org/pdf/2603.10215v1
+
+## 6069. Sabiá-4 Technical Report
+
+- **ID:** `2603.10213` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Thiago Laitz, Thales Sales Almeida, Hugo Abonizio, et al.
+- **abs:** https://arxiv.org/abs/2603.10213 · **PDF:** https://arxiv.org/pdf/2603.10213v1
+
+## 6070. FusionNet: a frame interpolation network for 4D heart models
+
+- **ID:** `2603.10212` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Chujie Chang, Shoko Miyauchi, Ken'ichi Morooka, et al.
+- **abs:** https://arxiv.org/abs/2603.10212 · **PDF:** https://arxiv.org/pdf/2603.10212v1
+
+## 6071. ViDia2Std: A Parallel Corpus and Methods for Low-Resource Vietnamese Dialect-to-Standard Translation
+
+- **ID:** `2603.10211` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Khoa Anh Ta, Nguyen Van Dinh, Kiet Van Nguyen
+- **abs:** https://arxiv.org/abs/2603.10211 · **PDF:** https://arxiv.org/pdf/2603.10211v1
+
+## 6072. Delta-K: Boosting Multi-Instance Generation via Cross-Attention Augmentation
+
+- **ID:** `2603.10210` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Zitong Wang, Zijun Shen, Haohao Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.10210 · **PDF:** https://arxiv.org/pdf/2603.10210v1
+
+## 6073. Flexible Cutoff Learning: Optimizing Machine Learning Potentials After Training
+
+- **ID:** `2603.10205` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-10
+- **Authors:** Rick Oerder, Jan Hamaekers
+- **abs:** https://arxiv.org/abs/2603.10205 · **PDF:** https://arxiv.org/pdf/2603.10205v1
+
+## 6074. GPU-Accelerated Genetic Programming for Symbolic Regression with Beagle Framework
+
+- **ID:** `2603.12292` · **primary:** `cs.NE` · **Submitted:** 2026-03-10
+- **Authors:** Nathan Haut, Ilya Basin, Marzieh Kianinejad, et al.
+- **abs:** https://arxiv.org/abs/2603.12292 · **PDF:** https://arxiv.org/pdf/2603.12292v1
+
+## 6075. Hybrid Hidden Markov Model for Modeling Equity Excess Growth Rate Dynamics: A Discrete-State Approach with Jump-Diffusion
+
+- **ID:** `2603.10202` · **primary:** `q-fin.ST` · **Submitted:** 2026-03-10
+- **Authors:** Abdulrahman Alswaidan, Jeffrey D. Varner
+- **abs:** https://arxiv.org/abs/2603.10202 · **PDF:** https://arxiv.org/pdf/2603.10202v2
+
+## 6076. Actor-Accelerated Policy Dual Averaging for Reinforcement Learning in Continuous Action Spaces
+
+- **ID:** `2603.10199` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Ji Gao, Caleb Ju, Guanghui Lan, et al.
+- **abs:** https://arxiv.org/abs/2603.10199 · **PDF:** https://arxiv.org/pdf/2603.10199v1
+
+## 6077. Adaptive Activation Cancellation for Hallucination Mitigation in Large Language Models
+
+- **ID:** `2603.10195` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Eric Yocam, Varghese Vaidyan, Gurcan Comert, et al.
+- **abs:** https://arxiv.org/abs/2603.10195 · **PDF:** https://arxiv.org/pdf/2603.10195v1
+
+## 6078. MCP-in-SoS: Risk assessment framework for open-source MCP servers
+
+- **ID:** `2603.10194` · **primary:** `cs.CR` · **Submitted:** 2026-03-10
+- **Authors:** Pratyay Kumar, Miguel Antonio Guirao Aguilera, Srikathyayani Srikanteswara, et al.
+- **abs:** https://arxiv.org/abs/2603.10194 · **PDF:** https://arxiv.org/pdf/2603.10194v1
+
+## 6079. ARCHE: Autoregressive Residual Compression with Hyperprior and Excitation
+
+- **ID:** `2603.10188` · **primary:** `eess.IV` · **Submitted:** 2026-03-10
+- **Authors:** Sofia Iliopoulou, Dimitris Ampeliotis, Athanassios Skodras
+- **abs:** https://arxiv.org/abs/2603.10188 · **PDF:** https://arxiv.org/pdf/2603.10188v1
+
+## 6080. Stability and Robustness via Regularization: Bandit Inference via Regularized Stochastic Mirror Descent
+
+- **ID:** `2603.10184` · **primary:** `stat.ML` · **Submitted:** 2026-03-10
+- **Authors:** Budhaditya Halder, Ishan Sengupta, Koustav Chowdhury, et al.
+- **abs:** https://arxiv.org/abs/2603.10184 · **PDF:** https://arxiv.org/pdf/2603.10184v1
+
+## 6081. DT-BEHRT: Disease Trajectory-aware Transformer for Interpretable Patient Representation Learning
+
+- **ID:** `2603.10180` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Deyi Li, Zijun Yao, Qi Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.10180 · **PDF:** https://arxiv.org/pdf/2603.10180v1
+
+## 6082. Video-Based Reward Modeling for Computer-Use Agents
+
+- **ID:** `2603.10178` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Linxin Song, Jieyu Zhang, Huanxin Sheng, et al.
+- **abs:** https://arxiv.org/abs/2603.10178 · **PDF:** https://arxiv.org/pdf/2603.10178v1
+
+## 6083. Calibration-Reasoning Framework for Descriptive Speech Quality Assessment
+
+- **ID:** `2603.10175` · **primary:** `eess.AS` · **Submitted:** 2026-03-10
+- **Authors:** Elizaveta Kostenok, Mathieu Salzmann, Milos Cernak
+- **abs:** https://arxiv.org/abs/2603.10175 · **PDF:** https://arxiv.org/pdf/2603.10175v1
+
+## 6084. OpenClaw-RL: Train Any Agent Simply by Talking
+
+- **ID:** `2603.10165` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Yinjie Wang, Xuyang Chen, Xiaolong Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.10165 · **PDF:** https://arxiv.org/pdf/2603.10165v1
+
+## 6085. Compatibility at a Cost: Systematic Discovery and Exploitation of MCP Clause-Compliance Vulnerabilities
+
+- **ID:** `2603.10163` · **primary:** `cs.CR` · **Submitted:** 2026-03-10
+- **Authors:** Nanzi Yang, Weiheng Bai, Kangjie Lu
+- **abs:** https://arxiv.org/abs/2603.10163 · **PDF:** https://arxiv.org/pdf/2603.10163v1
+
+## 6086. ReMix: Reinforcement routing for mixtures of LoRAs in LLM finetuning
+
+- **ID:** `2603.10160` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Ruizhong Qiu, Hanqing Zeng, Yinglong Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.10160 · **PDF:** https://arxiv.org/pdf/2603.10160v1
+
+## 6087. Mashup Learning: Faster Finetuning by Remixing Past Checkpoints
+
+- **ID:** `2603.10156` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Sofia Maria Lo Cicero Vaina, Artem Chumachenko, Max Ryabinin
+- **abs:** https://arxiv.org/abs/2603.10156 · **PDF:** https://arxiv.org/pdf/2603.10156v1
+
+## 6088. OA-NBV: Occlusion-Aware Next-Best-View Planning for Human-Centered Active Perception on Mobile Robots
+
+- **ID:** `2603.11072` · **primary:** `cs.RO` · **Submitted:** 2026-03-10
+- **Authors:** Boxun Hu, Chang Chang, Jiawei Ge, et al.
+- **abs:** https://arxiv.org/abs/2603.11072 · **PDF:** https://arxiv.org/pdf/2603.11072v1
+
+## 6089. ASCAT: An Arabic Scientific Corpus and Benchmark for Advanced Translation Evaluation
+
+- **ID:** `2604.00015` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Serry Sibaee, Khloud Al Jallad, Zineb Yousfi, et al.
+- **abs:** https://arxiv.org/abs/2604.00015 · **PDF:** https://arxiv.org/pdf/2604.00015v1
+
+## 6090. A neural operator for predicting vibration frequency response curves from limited data
+
+- **ID:** `2603.10149` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** D. Bluedorn, A. Badawy, B. E. Saunders, et al.
+- **abs:** https://arxiv.org/abs/2603.10149 · **PDF:** https://arxiv.org/pdf/2603.10149v1
+
+## 6091. Social Knowledge for Cross-Domain User Preference Modeling
+
+- **ID:** `2603.10148` · **primary:** `cs.SI` · **Submitted:** 2026-03-10
+- **Authors:** Nir Lotan, Adir Solomon, Ido Guy, et al.
+- **abs:** https://arxiv.org/abs/2603.10148 · **PDF:** https://arxiv.org/pdf/2603.10148v1
+
+## 6092. Lost in Backpropagation: The LM Head is a Gradient Bottleneck
+
+- **ID:** `2603.10145` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Nathan Godey, Yoav Artzi
+- **abs:** https://arxiv.org/abs/2603.10145 · **PDF:** https://arxiv.org/pdf/2603.10145v1
+
+## 6093. Reason and Verify: A Framework for Faithful Retrieval-Augmented Generation
+
+- **ID:** `2603.10143` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Eeham Khan, Luis Rodriguez, Marc Queudot
+- **abs:** https://arxiv.org/abs/2603.10143 · **PDF:** https://arxiv.org/pdf/2603.10143v1
+
+## 6094. The Generation-Recognition Asymmetry: Six Dimensions of a Fundamental Divide in Formal Language Theory
+
+- **ID:** `2603.10139` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Romain Peyrichou
+- **abs:** https://arxiv.org/abs/2603.10139 · **PDF:** https://arxiv.org/pdf/2603.10139v1
+
+## 6095. Agentic Control Center for Data Product Optimization
+
+- **ID:** `2603.10133` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Priyadarshini Tamilselvan, Gregory Bramble, Sola Shirai, et al.
+- **abs:** https://arxiv.org/abs/2603.10133 · **PDF:** https://arxiv.org/pdf/2603.10133v1
+
+## 6096. Unbalanced Optimal Transport Dictionary Learning for Unsupervised Hyperspectral Image Clustering
+
+- **ID:** `2603.10132` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Joshua Lentz, Nicholas Karris, Alex Cloninger, et al.
+- **abs:** https://arxiv.org/abs/2603.10132 · **PDF:** https://arxiv.org/pdf/2603.10132v1
+
+## 6097. The Prediction-Measurement Gap: Toward Meaning Representations as Scientific Instruments
+
+- **ID:** `2603.10130` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Hubert Plisiecki
+- **abs:** https://arxiv.org/abs/2603.10130 · **PDF:** https://arxiv.org/pdf/2603.10130v1
+
+## 6098. AR-VLA: True Autoregressive Action Expert for Vision-Language-Action Models
+
+- **ID:** `2603.10126` · **primary:** `cs.RO` · **Submitted:** 2026-03-10
+- **Authors:** Yutong Hu, Jan-Nico Zaech, Nikolay Nikolov, et al.
+- **abs:** https://arxiv.org/abs/2603.10126 · **PDF:** https://arxiv.org/pdf/2603.10126v1
+
+## 6099. Lost in the Middle at Birth: An Exact Theory of Transformer Position Bias
+
+- **ID:** `2603.10123` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Borun D Chowdhury
+- **abs:** https://arxiv.org/abs/2603.10123 · **PDF:** https://arxiv.org/pdf/2603.10123v1
+
+## 6100. CLIPO: Contrastive Learning in Policy Optimization Generalizes RLVR
+
+- **ID:** `2603.10101` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Sijia Cui, Pengyu Cheng, Jiajun Song, et al.
+- **abs:** https://arxiv.org/abs/2603.10101 · **PDF:** https://arxiv.org/pdf/2603.10101v1
+
+## 6101. Task Aware Modulation Using Representation Learning for Upsaling of Terrestrial Carbon Fluxes
+
+- **ID:** `2603.09974` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Aleksei Rozanov, Arvind Renganathan, Vipin Kumar
+- **abs:** https://arxiv.org/abs/2603.09974 · **PDF:** https://arxiv.org/pdf/2603.09974v2
+
+## 6102. Hardware Efficient Approximate Convolution with Tunable Error Tolerance for CNNs
+
+- **ID:** `2603.10100` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Vishal Shashidhar, Anupam Kumari, Roy P Paily
+- **abs:** https://arxiv.org/abs/2603.10100 · **PDF:** https://arxiv.org/pdf/2603.10100v1
+
+## 6103. From Data Statistics to Feature Geometry: How Correlations Shape Superposition
+
+- **ID:** `2603.09972` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Lucas Prieto, Edward Stevinson, Melih Barsbey, et al.
+- **abs:** https://arxiv.org/abs/2603.09972 · **PDF:** https://arxiv.org/pdf/2603.09972v1
+
+## 6104. CREATE: Testing LLMs for Associative Creativity
+
+- **ID:** `2603.09970` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Manya Wadhwa, Tiasa Singha Roy, Harvey Lederman, et al.
+- **abs:** https://arxiv.org/abs/2603.09970 · **PDF:** https://arxiv.org/pdf/2603.09970v1
+
+## 6105. Understanding the Use of a Large Language Model-Powered Guide to Make Virtual Reality Accessible for Blind and Low Vision People
+
+- **ID:** `2603.09964` · **primary:** `cs.HC` · **Submitted:** 2026-03-10
+- **Authors:** Jazmin Collins, Sharon Y Lin, Tianqi Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.09964 · **PDF:** https://arxiv.org/pdf/2603.09964v2
+
+## 6106. Emotional Modulation in Swarm Decision Dynamics
+
+- **ID:** `2603.09963` · **primary:** `cs.MA` · **Submitted:** 2026-03-10
+- **Authors:** David Freire-Obregón
+- **abs:** https://arxiv.org/abs/2603.09963 · **PDF:** https://arxiv.org/pdf/2603.09963v1
+
+## 6107. BEACON: Language-Conditioned Navigation Affordance Prediction under Occlusion
+
+- **ID:** `2603.09961` · **primary:** `cs.RO` · **Submitted:** 2026-03-10
+- **Authors:** Xinyu Gao, Gang Chen, Javier Alonso-Mora
+- **abs:** https://arxiv.org/abs/2603.09961 · **PDF:** https://arxiv.org/pdf/2603.09961v1
+
+## 6108. TinyNav: End-to-End TinyML for Real-Time Autonomous Navigation on Microcontrollers
+
+- **ID:** `2603.11071` · **primary:** `cs.RO` · **Submitted:** 2026-03-10
+- **Authors:** Pooria Roy, Nourhan Jadallah. Tomer Lapid, Shahzaib Ahmad, et al.
+- **abs:** https://arxiv.org/abs/2603.11071 · **PDF:** https://arxiv.org/pdf/2603.11071v1
+
+## 6109. Think Before You Lie: How Reasoning Leads to Honesty
+
+- **ID:** `2603.09957` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Ann Yuan, Asma Ghandeharioun, Carter Blum, et al.
+- **abs:** https://arxiv.org/abs/2603.09957 · **PDF:** https://arxiv.org/pdf/2603.09957v2
+
+## 6110. From Semantics to Pixels: Coarse-to-Fine Masked Autoencoders for Hierarchical Visual Understanding
+
+- **ID:** `2603.09955` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Wenzhao Xiang, Yue Wu, Hongyang Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.09955 · **PDF:** https://arxiv.org/pdf/2603.09955v1
+
+## 6111. On the Width Scaling of Neural Optimizers Under Matrix Operator Norms I: Row/Column Normalization and Hyperparameter Transfer
+
+- **ID:** `2603.09952` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Ruihan Xu, Jiajin Li, Yiping Lu
+- **abs:** https://arxiv.org/abs/2603.09952 · **PDF:** https://arxiv.org/pdf/2603.09952v1
+
+## 6112. Towards a Neural Debugger for Python
+
+- **ID:** `2603.09951` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Maximilian Beck, Jonas Gehring, Jannik Kossen, et al.
+- **abs:** https://arxiv.org/abs/2603.09951 · **PDF:** https://arxiv.org/pdf/2603.09951v1
+
+## 6113. When Learning Rates Go Wrong: Early Structural Signals in PPO Actor-Critic
+
+- **ID:** `2603.09950` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Alberto Fernández-Hernández, Cristian Pérez-Corral, Jose I. Mestre, et al.
+- **abs:** https://arxiv.org/abs/2603.09950 · **PDF:** https://arxiv.org/pdf/2603.09950v1
+
+## 6114. The Confidence Gate Theorem: When Should Ranked Decision Systems Abstain?
+
+- **ID:** `2603.09947` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Ronald Doku
+- **abs:** https://arxiv.org/abs/2603.09947 · **PDF:** https://arxiv.org/pdf/2603.09947v1
+
+## 6115. Disentangling Prompt Element Level Risk Factors for Hallucinations and Omissions in Mental Health LLM Responses
+
+- **ID:** `2604.00014` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Congning Ni, Sarvech Qadir, Bryan Steitz, et al.
+- **abs:** https://arxiv.org/abs/2604.00014 · **PDF:** https://arxiv.org/pdf/2604.00014v1
+
+## 6116. Denoising the US Census: Succinct Block Hierarchical Regression
+
+- **ID:** `2603.10099` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Badih Ghazi, Pritish Kamath, Ravi Kumar, et al.
+- **abs:** https://arxiv.org/abs/2603.10099 · **PDF:** https://arxiv.org/pdf/2603.10099v1
+
+## 6117. No Image, No Problem: End-to-End Multi-Task Cardiac Analysis from Undersampled k-Space
+
+- **ID:** `2603.09945` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Yundi Zhang, Sevgi Gokce Kafali, Niklas Bubeck, et al.
+- **abs:** https://arxiv.org/abs/2603.09945 · **PDF:** https://arxiv.org/pdf/2603.09945v1
+
+## 6118. Code-Space Response Oracles: Generating Interpretable Multi-Agent Policies with Large Language Models
+
+- **ID:** `2603.10098` · **primary:** `cs.GT` · **Submitted:** 2026-03-10
+- **Authors:** Daniel Hennes, Zun Li, John Schultz, et al.
+- **abs:** https://arxiv.org/abs/2603.10098 · **PDF:** https://arxiv.org/pdf/2603.10098v1
+
+## 6119. PathMem: Toward Cognition-Aligned Memory Transformation for Pathology MLLMs
+
+- **ID:** `2603.09943` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Jinyue Li, Yuci Liang, Qiankun Li, et al.
+- **abs:** https://arxiv.org/abs/2603.09943 · **PDF:** https://arxiv.org/pdf/2603.09943v1
+
+## 6120. Towards Flexible Spectrum Access: Data-Driven Insights into Spectrum Demand
+
+- **ID:** `2603.09942` · **primary:** `eess.SY` · **Submitted:** 2026-03-10
+- **Authors:** Mohamad Alkadamani, Amir Ghasemi, Halim Yanikomeroglu
+- **abs:** https://arxiv.org/abs/2603.09942 · **PDF:** https://arxiv.org/pdf/2603.09942v1
+
+## 6121. SignalMC-MED: A Multimodal Benchmark for Evaluating Biosignal Foundation Models on Single-Lead ECG and PPG
+
+- **ID:** `2603.09940` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Fredrik K. Gustafsson, Xiao Gu, Mattia Carletti, et al.
+- **abs:** https://arxiv.org/abs/2603.09940 · **PDF:** https://arxiv.org/pdf/2603.09940v1
+
+## 6122. Model Merging in the Era of Large Language Models: Methods, Applications, and Future Directions
+
+- **ID:** `2603.09938` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Mingyang Song, Mao Zheng
+- **abs:** https://arxiv.org/abs/2603.09938 · **PDF:** https://arxiv.org/pdf/2603.09938v2
+
+## 6123. Generative Drifting is Secretly Score Matching: a Spectral and Variational Perspective
+
+- **ID:** `2603.09936` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Erkan Turan, Maks Ovsjanikov
+- **abs:** https://arxiv.org/abs/2603.09936 · **PDF:** https://arxiv.org/pdf/2603.09936v1
+
+## 6124. Adaptive Clinical-Aware Latent Diffusion for Multimodal Brain Image Generation and Missing Modality Imputation
+
+- **ID:** `2603.09931` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Rong Zhou, Houliang Zhou, Yao Su, et al.
+- **abs:** https://arxiv.org/abs/2603.09931 · **PDF:** https://arxiv.org/pdf/2603.09931v1
+
+## 6125. OptEMA: Adaptive Exponential Moving Average for Stochastic Optimization with Zero-Noise Optimality
+
+- **ID:** `2603.09923` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Ganzhao Yuan
+- **abs:** https://arxiv.org/abs/2603.09923 · **PDF:** https://arxiv.org/pdf/2603.09923v1
+
+## 6126. AI-Enabled Data-driven Intelligence for Spectrum Demand Estimation
+
+- **ID:** `2603.09916` · **primary:** `eess.SY` · **Submitted:** 2026-03-10
+- **Authors:** Colin Brown, Mohamad Alkadamani, Halim Yanikomeroglu
+- **abs:** https://arxiv.org/abs/2603.09916 · **PDF:** https://arxiv.org/pdf/2603.09916v1
+
+## 6127. MedMASLab: A Unified Orchestration Framework for Benchmarking Multimodal Medical Multi-Agent Systems
+
+- **ID:** `2603.09909` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Yunhang Qian, Xiaobin Hu, Jiaquan Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.09909 · **PDF:** https://arxiv.org/pdf/2603.09909v2
+
+## 6128. Thinking to Recall: How Reasoning Unlocks Parametric Knowledge in LLMs
+
+- **ID:** `2603.09906` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Zorik Gekhman, Roee Aharoni, Eran Ofek, et al.
+- **abs:** https://arxiv.org/abs/2603.09906 · **PDF:** https://arxiv.org/pdf/2603.09906v1
+
+## 6129. MSSR: Memory-Aware Adaptive Replay for Continual LLM Fine-Tuning
+
+- **ID:** `2603.09892` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Yiyang Lu, Yu He, Jianlong Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.09892 · **PDF:** https://arxiv.org/pdf/2603.09892v1
+
+## 6130. Deep Learning for BioImaging: What Are We Learning?
+
+- **ID:** `2603.13377` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Ivan Svatko, Maxime Sanchez, Ihab Bendidi, et al.
+- **abs:** https://arxiv.org/abs/2603.13377 · **PDF:** https://arxiv.org/pdf/2603.13377v1
+
+## 6131. Influencing LLM Multi-Agent Dialogue via Policy-Parameterized Prompts
+
+- **ID:** `2603.09890` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Hongbo Bo, Jingyu Hu, Weiru Liu
+- **abs:** https://arxiv.org/abs/2603.09890 · **PDF:** https://arxiv.org/pdf/2603.09890v1
+
+## 6132. LCA: Local Classifier Alignment for Continual Learning
+
+- **ID:** `2603.09888` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Tung Tran, Danilo Vasconcellos Vargas, Khoat Than
+- **abs:** https://arxiv.org/abs/2603.09888 · **PDF:** https://arxiv.org/pdf/2603.09888v2
+
+## 6133. Benchmarking Political Persuasion Risks Across Frontier Large Language Models
+
+- **ID:** `2603.09884` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Zhongren Chen, Joshua Kalla, Quan Le
+- **abs:** https://arxiv.org/abs/2603.09884 · **PDF:** https://arxiv.org/pdf/2603.09884v1
+
+## 6134. Emerging Extrinsic Dexterity in Cluttered Scenes via Dynamics-aware Policy Learning
+
+- **ID:** `2603.09882` · **primary:** `cs.RO` · **Submitted:** 2026-03-10
+- **Authors:** Yixin Zheng, Jiangran Lyu, Yifan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.09882 · **PDF:** https://arxiv.org/pdf/2603.09882v1
+
+## 6135. Do What I Say: A Spoken Prompt Dataset for Instruction-Following
+
+- **ID:** `2603.09881` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Maike Züfle, Sara Papi, Fabian Retkowski, et al.
+- **abs:** https://arxiv.org/abs/2603.09881 · **PDF:** https://arxiv.org/pdf/2603.09881v1
+
+## 6136. N-gram-like Language Models Predict Reading Time Best
+
+- **ID:** `2603.09872` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** James A. Michaelov, Roger P. Levy
+- **abs:** https://arxiv.org/abs/2603.09872 · **PDF:** https://arxiv.org/pdf/2603.09872v1
+
+## 6137. CarbonBench: A Global Benchmark for Upscaling of Carbon Fluxes Using Zero-Shot Learning
+
+- **ID:** `2603.09868` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Aleksei Rozanov, Arvind Renganathan, Yimeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.09868 · **PDF:** https://arxiv.org/pdf/2603.09868v1
+
+## 6138. Rethinking Adam for Time Series Forecasting: A Simple Heuristic to Improve Optimization under Distribution Shifts
+
+- **ID:** `2603.10095` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Yuze Dong, Jinsong Wu
+- **abs:** https://arxiv.org/abs/2603.10095 · **PDF:** https://arxiv.org/pdf/2603.10095v1
+
+## 6139. GAST: Gradient-aligned Sparse Tuning of Large Language Models with Data-layer Selection
+
+- **ID:** `2603.09865` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Kai Yao, Zhenghan Song, Kaixin Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.09865 · **PDF:** https://arxiv.org/pdf/2603.09865v1
+
+## 6140. A Graph-Based Approach to Spectrum Demand Prediction Using Hierarchical Attention Networks
+
+- **ID:** `2603.09859` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Mohamad Alkadamani, Halim Yanikomeroglu, Amir Ghasemi
+- **abs:** https://arxiv.org/abs/2603.09859 · **PDF:** https://arxiv.org/pdf/2603.09859v1
+
+## 6141. SCENEBench: An Audio Understanding Benchmark Grounded in Assistive and Industrial Use Cases
+
+- **ID:** `2603.09853` · **primary:** `cs.SD` · **Submitted:** 2026-03-10
+- **Authors:** Laya Iyer, Angelina Wang, Sanmi Koyejo
+- **abs:** https://arxiv.org/abs/2603.09853 · **PDF:** https://arxiv.org/pdf/2603.09853v1
+
+## 6142. A Unified Hierarchical Multi-Task Multi-Fidelity Framework for Data-Efficient Surrogate Modeling in Manufacturing
+
+- **ID:** `2603.09842` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Manan Mehta, Zhiqiao Dong, Yuhang Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.09842 · **PDF:** https://arxiv.org/pdf/2603.09842v1
+
+## 6143. Equivariant Asynchronous Diffusion: An Adaptive Denoising Schedule for Accelerated Molecular Conformation Generation
+
+- **ID:** `2603.10093` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Junyi An, Chao Qu, Yun-Fei Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.10093 · **PDF:** https://arxiv.org/pdf/2603.10093v1
+
+## 6144. Chow-Liu Ordering for Long-Context Reasoning in Chain-of-Agents
+
+- **ID:** `2603.09835` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Naman Gupta, Vaibhav Singh, Arun Iyer, et al.
+- **abs:** https://arxiv.org/abs/2603.09835 · **PDF:** https://arxiv.org/pdf/2603.09835v1
+
+## 6145. Execution Is the New Attack Surface: Survivability-Aware Agentic Crypto Trading with OpenClaw-Style Local Executors
+
+- **ID:** `2603.10092` · **primary:** `cs.CR` · **Submitted:** 2026-03-10
+- **Authors:** Ailiya Borjigin, Igor Stadnyk, Ben Bilski, et al.
+- **abs:** https://arxiv.org/abs/2603.10092 · **PDF:** https://arxiv.org/pdf/2603.10092v1
+
+## 6146. MA-EgoQA: Question Answering over Egocentric Videos from Multiple Embodied Agents
+
+- **ID:** `2603.09827` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Kangsan Kim, Yanlai Yang, Suji Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.09827 · **PDF:** https://arxiv.org/pdf/2603.09827v2
+
+## 6147. One-Eval: An Agentic System for Automated and Traceable LLM Evaluation
+
+- **ID:** `2603.09821` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Chengyu Shen, Yanheng Hou, Minghui Pan, et al.
+- **abs:** https://arxiv.org/abs/2603.09821 · **PDF:** https://arxiv.org/pdf/2603.09821v1
+
+## 6148. Correction of Transformer-Based Models with Smoothing Pseudo-Projector
+
+- **ID:** `2603.09815` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Vitaly Bulgakov
+- **abs:** https://arxiv.org/abs/2603.09815 · **PDF:** https://arxiv.org/pdf/2603.09815v1
+
+## 6149. Multi-Stream Perturbation Attack: Breaking Safety Alignment of Thinking LLMs Through Concurrent Task Interference
+
+- **ID:** `2603.10091` · **primary:** `cs.CR` · **Submitted:** 2026-03-10
+- **Authors:** Fan Yang
+- **abs:** https://arxiv.org/abs/2603.10091 · **PDF:** https://arxiv.org/pdf/2603.10091v1
+
+## 6150. Good Reasoning Makes Good Demonstrations: Implicit Reasoning Quality Supervision via In-Context Reinforcement Learning
+
+- **ID:** `2603.09803` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Tiehua Mei, Minxuan Lv, Leiyu Pan, et al.
+- **abs:** https://arxiv.org/abs/2603.09803 · **PDF:** https://arxiv.org/pdf/2603.09803v1
+
+## 6151. A Survey of Weight Space Learning: Understanding, Representation, and Generation
+
+- **ID:** `2603.10090` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Xiaolong Han, Zehong Wang, Bo Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.10090 · **PDF:** https://arxiv.org/pdf/2603.10090v1
+
+## 6152. MITRA: An AI Assistant for Knowledge Retrieval in Physics Collaborations
+
+- **ID:** `2603.09800` · **primary:** `cs.IR` · **Submitted:** 2026-03-10
+- **Authors:** Abhishikth Mallampalli, Sridhara Dasu
+- **abs:** https://arxiv.org/abs/2603.09800 · **PDF:** https://arxiv.org/pdf/2603.09800v1
+
+## 6153. Information Theoretic Bayesian Optimization over the Probability Simplex
+
+- **ID:** `2603.09793` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Federico Pavesi, Antonio Candelieri, Noémie Jaquier
+- **abs:** https://arxiv.org/abs/2603.09793 · **PDF:** https://arxiv.org/pdf/2603.09793v1
+
+## 6154. Exploiting Adaptive Channel Pruning for Communication-Efficient Split Learning
+
+- **ID:** `2603.09792` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Jialei Tan, Zheng Lin, Xiangming Cai, et al.
+- **abs:** https://arxiv.org/abs/2603.09792 · **PDF:** https://arxiv.org/pdf/2603.09792v2
+
+## 6155. A Hybrid Quantum-Classical Framework for Financial Volatility Forecasting Based on Quantum Circuit Born Machines
+
+- **ID:** `2603.09789` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Yixiong Chen
+- **abs:** https://arxiv.org/abs/2603.09789 · **PDF:** https://arxiv.org/pdf/2603.09789v1
+
+## 6156. What is Missing? Explaining Neurons Activated by Absent Concepts
+
+- **ID:** `2603.09787` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Robin Hesse, Simone Schaub-Meyer, Janina Hesse, et al.
+- **abs:** https://arxiv.org/abs/2603.09787 · **PDF:** https://arxiv.org/pdf/2603.09787v1
+
+## 6157. Quantifying the Necessity of Chain of Thought through Opaque Serial Depth
+
+- **ID:** `2603.09786` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Jonah Brown-Cohen, David Lindner, Rohin Shah
+- **abs:** https://arxiv.org/abs/2603.09786 · **PDF:** https://arxiv.org/pdf/2603.09786v1
+
+## 6158. Quantum Amplitude Estimation for Catastrophe Insurance Tail-Risk Pricing: Empirical Convergence and NISQ Noise Analysis
+
+- **ID:** `2603.15664` · **primary:** `stat.AP` · **Submitted:** 2026-03-10
+- **Authors:** Alexis Kirke
+- **abs:** https://arxiv.org/abs/2603.15664 · **PDF:** https://arxiv.org/pdf/2603.15664v1
+
+## 6159. EPIC-EuroParl-UdS: Information-Theoretic Perspectives on Translation and Interpreting
+
+- **ID:** `2603.09785` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Maria Kunilovskaya, Christina Pollkläsener
+- **abs:** https://arxiv.org/abs/2603.09785 · **PDF:** https://arxiv.org/pdf/2603.09785v2
+
+## 6160. First Estimation of Model Parameters for Neutrino-Induced Nucleon Knockout Using Simulation-Based Inference
+
+- **ID:** `2603.09778` · **primary:** `hep-ph` · **Submitted:** 2026-03-10
+- **Authors:** Karla Tame-Narvaez, Steven Gardiner, Aleksandra Ćiprijanović, et al.
+- **abs:** https://arxiv.org/abs/2603.09778 · **PDF:** https://arxiv.org/pdf/2603.09778v1
+
+## 6161. World2Mind: Cognition Toolkit for Allocentric Spatial Reasoning in Foundation Models
+
+- **ID:** `2603.09774` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Shouwei Ruan, Bin Wang, Zhenyu Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.09774 · **PDF:** https://arxiv.org/pdf/2603.09774v1
+
+## 6162. Global universality via discrete-time signatures
+
+- **ID:** `2603.09773` · **primary:** `math.PR` · **Submitted:** 2026-03-10
+- **Authors:** Mihriban Ceylan, David J. Prömel
+- **abs:** https://arxiv.org/abs/2603.09773 · **PDF:** https://arxiv.org/pdf/2603.09773v1
+
+## 6163. Ego: Embedding-Guided Personalization of Vision-Language Models
+
+- **ID:** `2603.09771` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Soroush Seifi, Simon Gardier, Vaggelis Dorovatas, et al.
+- **abs:** https://arxiv.org/abs/2603.09771 · **PDF:** https://arxiv.org/pdf/2603.09771v2
+
+## 6164. Beyond Fine-Tuning: Robust Food Entity Linking under Ontology Drift with FoodOntoRAG
+
+- **ID:** `2603.09758` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Jan Drole, Ana Gjorgjevikj, Barbara Korouši'c Seljak, et al.
+- **abs:** https://arxiv.org/abs/2603.09758 · **PDF:** https://arxiv.org/pdf/2603.09758v1
+
+## 6165. LLM-MRD: LLM-Guided Multi-View Reasoning Distillation for Fake News Detection
+
+- **ID:** `2603.19293` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Weilin Zhou, Shanwen Tan, Enhao Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.19293 · **PDF:** https://arxiv.org/pdf/2603.19293v1
+
+## 6166. Upper Generalization Bounds for Neural Oscillators
+
+- **ID:** `2603.09742` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Zifeng Huang, Konstantin M. Zuev, Yong Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.09742 · **PDF:** https://arxiv.org/pdf/2603.09742v1
+
+## 6167. EXPLORE-Bench: Egocentric Scene Prediction with Long-Horizon Reasoning
+
+- **ID:** `2603.09731` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Chengjun Yu, Xuhan Zhu, Chaoqun Du, et al.
+- **abs:** https://arxiv.org/abs/2603.09731 · **PDF:** https://arxiv.org/pdf/2603.09731v2
+
+## 6168. A Multi-Prototype-Guided Federated Knowledge Distillation Approach in AI-RAN Enabled Multi-Access Edge Computing System
+
+- **ID:** `2603.09727` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Luyao Zou, Hayoung Oh, Chu Myaet Thwal, et al.
+- **abs:** https://arxiv.org/abs/2603.09727 · **PDF:** https://arxiv.org/pdf/2603.09727v1
+
+## 6169. ES-dLLM: Efficient Inference for Diffusion Large Language Models by Early-Skipping
+
+- **ID:** `2603.10088` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Zijian Zhu, Fei Ren, Zhanhong Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.10088 · **PDF:** https://arxiv.org/pdf/2603.10088v1
+
+## 6170. RbtAct: Rebuttal as Supervision for Actionable Review Feedback Generation
+
+- **ID:** `2603.09723` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Sihong Wu, Yiling Ma, Yilun Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.09723 · **PDF:** https://arxiv.org/pdf/2603.09723v1
+
+## 6171. AutoAgent: Evolving Cognition and Elastic Memory Orchestration for Adaptive Agents
+
+- **ID:** `2603.09716` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Xiaoxing Wang, Ning Liao, Shikun Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.09716 · **PDF:** https://arxiv.org/pdf/2603.09716v1
+
+## 6172. Does the Question Really Matter? Training-Free Data Selection for Vision-Language SFT
+
+- **ID:** `2603.09715` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Peng Sun, Huawen Shen, Yi Ban, et al.
+- **abs:** https://arxiv.org/abs/2603.09715 · **PDF:** https://arxiv.org/pdf/2603.09715v1
+
+## 6173. MUGEN: Evaluating and Improving Multi-audio Understanding of Large Audio-Language Models
+
+- **ID:** `2603.09714` · **primary:** `cs.SD` · **Submitted:** 2026-03-10
+- **Authors:** Chih-Kai Yang, Yun-Shao Tsai, Yu-Kai Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.09714 · **PDF:** https://arxiv.org/pdf/2603.09714v1
+
+## 6174. From Phase Prediction to Phase Design: A ReAct Agent Framework for High-Entropy Alloy Discovery
+
+- **ID:** `2603.11068` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-10
+- **Authors:** Iman Peivaste, Salim Belouettar
+- **abs:** https://arxiv.org/abs/2603.11068 · **PDF:** https://arxiv.org/pdf/2603.11068v1
+
+## 6175. Developing Machine Learning-Based Watch-to-Warning Severe Weather Guidance from the Warn-on-Forecast System
+
+- **ID:** `2603.20250` · **primary:** `physics.ao-ph` · **Submitted:** 2026-03-10
+- **Authors:** Montgomery Flora, Samuel Varga, Corey Potvin, et al.
+- **abs:** https://arxiv.org/abs/2603.20250 · **PDF:** https://arxiv.org/pdf/2603.20250v1
+
+## 6176. OOD-MMSafe: Advancing MLLM Safety from Harmful Intent to Hidden Consequences
+
+- **ID:** `2603.09706` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Ming Wen, Kun Yang, Jingyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.09706 · **PDF:** https://arxiv.org/pdf/2603.09706v1
+
+## 6177. Evaluation of LLMs in retrieving food and nutritional context for RAG systems
+
+- **ID:** `2603.09704` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Maks Požarnik Vavken, Matevž Ogrinc, Tome Eftimov, et al.
+- **abs:** https://arxiv.org/abs/2603.09704 · **PDF:** https://arxiv.org/pdf/2603.09704v2
+
+## 6178. Mousse: Rectifying the Geometry of Muon with Curvature-Aware Preconditioning
+
+- **ID:** `2603.09697` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Yechen Zhang, Shuhao Xing, Junhao Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.09697 · **PDF:** https://arxiv.org/pdf/2603.09697v2
+
+## 6179. Physics-informed neural operator for predictive parametric phase-field modelling
+
+- **ID:** `2603.09693` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Nanxi Chen, Airong Chen, Rujin Ma
+- **abs:** https://arxiv.org/abs/2603.09693 · **PDF:** https://arxiv.org/pdf/2603.09693v1
+
+## 6180. ActiveUltraFeedback: Efficient Preference Data Generation using Active Learning
+
+- **ID:** `2603.09692` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Davit Melikidze, Marian Schneider, Jessica Lam, et al.
+- **abs:** https://arxiv.org/abs/2603.09692 · **PDF:** https://arxiv.org/pdf/2603.09692v1
+
+## 6181. ESAinsTOD: A Unified End-to-End Schema-Aware Instruction-Tuning Framework for Task-Oriented Dialog Modeling
+
+- **ID:** `2603.09691` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Dechuan Teng, Chunlin Lu, Libo Qin, et al.
+- **abs:** https://arxiv.org/abs/2603.09691 · **PDF:** https://arxiv.org/pdf/2603.09691v1
+
+## 6182. AutoViVQA: A Large-Scale Automatically Constructed Dataset for Vietnamese Visual Question Answering
+
+- **ID:** `2603.09689` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Nguyen Anh Tuong, Phan Ba Duc, Nguyen Trung Quoc, et al.
+- **abs:** https://arxiv.org/abs/2603.09689 · **PDF:** https://arxiv.org/pdf/2603.09689v2
+
+## 6183. Fusing Semantic, Lexical, and Domain Perspectives for Recipe Similarity Estimation
+
+- **ID:** `2603.09688` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Denica Kjorvezir, Danilo Najkov, Eva Valencič, et al.
+- **abs:** https://arxiv.org/abs/2603.09688 · **PDF:** https://arxiv.org/pdf/2603.09688v2
+
+## 6184. Automatic Cardiac Risk Management Classification using large-context Electronic Patients Health Records
+
+- **ID:** `2603.09685` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Jacopo Vitale, David Della Morte, Luca Bacco, et al.
+- **abs:** https://arxiv.org/abs/2603.09685 · **PDF:** https://arxiv.org/pdf/2603.09685v1
+
+## 6185. On Catastrophic Forgetting in Low-Rank Decomposition-Based Parameter-Efficient Fine-Tuning
+
+- **ID:** `2603.09684` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Muhammad Ahmad, Jingjing Zheng, Yankai Cao
+- **abs:** https://arxiv.org/abs/2603.09684 · **PDF:** https://arxiv.org/pdf/2603.09684v1
+
+## 6186. Murmurations: a case study in AI-assisted mathematics
+
+- **ID:** `2603.09680` · **primary:** `math.NT` · **Submitted:** 2026-03-10
+- **Authors:** Yang-Hui He, Kyu-Hwan Lee, Thomas Oliver, et al.
+- **abs:** https://arxiv.org/abs/2603.09680 · **PDF:** https://arxiv.org/pdf/2603.09680v1
+
+## 6187. EsoLang-Bench: Evaluating Genuine Reasoning in Large Language Models via Esoteric Programming Languages
+
+- **ID:** `2603.09678` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Aman Sharma, Paras Chopra
+- **abs:** https://arxiv.org/abs/2603.09678 · **PDF:** https://arxiv.org/pdf/2603.09678v1
+
+## 6188. Logics-Parsing-Omni Technical Report
+
+- **ID:** `2603.09677` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Xin An, Jingyi Cai, Xiangyang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.09677 · **PDF:** https://arxiv.org/pdf/2603.09677v2
+
+## 6189. GNNs for Time Series Anomaly Detection: An Open-Source Framework and a Critical Evaluation
+
+- **ID:** `2603.09675` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Federico Bello, Gonzalo Chiarlone, Marcelo Fiori, et al.
+- **abs:** https://arxiv.org/abs/2603.09675 · **PDF:** https://arxiv.org/pdf/2603.09675v1
+
+## 6190. KernelSkill: A Multi-Agent Framework for GPU Kernel Optimization
+
+- **ID:** `2603.10085` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Qitong Sun, Jun Han, Tianlin Li, et al.
+- **abs:** https://arxiv.org/abs/2603.10085 · **PDF:** https://arxiv.org/pdf/2603.10085v1
+
+## 6191. No evaluation without fair representation : Impact of label and selection bias on the evaluation, performance and mitigation of classification models
+
+- **ID:** `2603.09662` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Magali Legast, Toon Calders, François Fouss
+- **abs:** https://arxiv.org/abs/2603.09662 · **PDF:** https://arxiv.org/pdf/2603.09662v1
+
+## 6192. FreqCycle: A Multi-Scale Time-Frequency Analysis Method for Time Series Forecasting
+
+- **ID:** `2603.09661` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Boya Zhang, Shuaijie Yin, Huiwen Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.09661 · **PDF:** https://arxiv.org/pdf/2603.09661v1
+
+## 6193. When to Lock Attention: Training-Free KV Control in Video Diffusion
+
+- **ID:** `2603.09657` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Tianyi Zeng, Jincheng Gao, Tianyi Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.09657 · **PDF:** https://arxiv.org/pdf/2603.09657v1
+
+## 6194. Understanding the Interplay between LLMs' Utilisation of Parametric and Contextual Knowledge: A keynote at ECIR 2025
+
+- **ID:** `2603.09654` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Isabelle Augenstein
+- **abs:** https://arxiv.org/abs/2603.09654 · **PDF:** https://arxiv.org/pdf/2603.09654v1
+
+## 6195. MiniAppBench: Evaluating the Shift from Text to Interactive HTML Responses in LLM-Powered Assistants
+
+- **ID:** `2603.09652` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Zuhao Zhang, Chengyue Yu, Yuante Li, et al.
+- **abs:** https://arxiv.org/abs/2603.09652 · **PDF:** https://arxiv.org/pdf/2603.09652v2
+
+## 6196. Well Log-Guided Synthesis of Subsurface Images from Sparse Petrography Data Using cGANs
+
+- **ID:** `2603.09651` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Ali Sadeghkhani, A. Assadi, B. Bennett, et al.
+- **abs:** https://arxiv.org/abs/2603.09651 · **PDF:** https://arxiv.org/pdf/2603.09651v1
+
+## 6197. LITTA: Late-Interaction and Test-Time Alignment for Visually-Grounded Multimodal Retrieval
+
+- **ID:** `2603.26683` · **primary:** `cs.IR` · **Submitted:** 2026-03-10
+- **Authors:** Seonok Kim
+- **abs:** https://arxiv.org/abs/2603.26683 · **PDF:** https://arxiv.org/pdf/2603.26683v1
+
+## 6198. Noise in Photonic Quantum Machine Learning: Models, Impacts, and Mitigation Strategies
+
+- **ID:** `2603.09645` · **primary:** `quant-ph` · **Submitted:** 2026-03-10
+- **Authors:** A. M. A. S. D. Alagiyawanna, Asoka Karunananda
+- **abs:** https://arxiv.org/abs/2603.09645 · **PDF:** https://arxiv.org/pdf/2603.09645v2
+
+## 6199. MM-tau-p$^2$: Persona-Adaptive Prompting for Robust Multi-Modal Agent Evaluation in Dual-Control Settings
+
+- **ID:** `2603.09643` · **primary:** `cs.ET` · **Submitted:** 2026-03-10
+- **Authors:** Anupam Purwar, Aditya Choudhary
+- **abs:** https://arxiv.org/abs/2603.09643 · **PDF:** https://arxiv.org/pdf/2603.09643v3
+
+## 6200. Multi-DNN Inference of Sparse Models on Edge SoCs
+
+- **ID:** `2603.09642` · **primary:** `cs.DC` · **Submitted:** 2026-03-10
+- **Authors:** Jiawei Luo, Di Wu, Simon Dobson, et al.
+- **abs:** https://arxiv.org/abs/2603.09642 · **PDF:** https://arxiv.org/pdf/2603.09642v1
+
+## 6201. PRECEPT: Planning Resilience via Experience, Context Engineering & Probing Trajectories A Unified Framework for Test-Time Adaptation with Compositional Rule Learning and Pareto-Guided Prompt Evolution
+
+- **ID:** `2603.09641` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Arash Shahmansoori
+- **abs:** https://arxiv.org/abs/2603.09641 · **PDF:** https://arxiv.org/pdf/2603.09641v1
+
+## 6202. Tracking Cancer Through Text: Longitudinal Extraction From Radiology Reports Using Open-Source Large Language Models
+
+- **ID:** `2603.09638` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Luc Builtjes, Alessa Hering
+- **abs:** https://arxiv.org/abs/2603.09638 · **PDF:** https://arxiv.org/pdf/2603.09638v2
+
+## 6203. X-GS: An Extensible Open Framework for Perceiving and Thinking via 3D Gaussian Splatting
+
+- **ID:** `2603.09632` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Yueen Ma, Zenglin Xu, Irwin King
+- **abs:** https://arxiv.org/abs/2603.09632 · **PDF:** https://arxiv.org/pdf/2603.09632v2
+
+## 6204. Grounding Synthetic Data Generation With Vision and Language Models
+
+- **ID:** `2603.09625` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Ümit Mert Çağlar, Alptekin Temizel
+- **abs:** https://arxiv.org/abs/2603.09625 · **PDF:** https://arxiv.org/pdf/2603.09625v1
+
+## 6205. OrthoAI v2: From Single-Agent Segmentation to Dual-Agent Treatment Planning for Clear Aligners
+
+- **ID:** `2603.15663` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Lansiaux Edouard, Leman Margaux
+- **abs:** https://arxiv.org/abs/2603.15663 · **PDF:** https://arxiv.org/pdf/2603.15663v1
+
+## 6206. Context Engineering: From Prompts to Corporate Multi-Agent Architecture
+
+- **ID:** `2603.09619` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Vera V. Vishnyakova
+- **abs:** https://arxiv.org/abs/2603.09619 · **PDF:** https://arxiv.org/pdf/2603.09619v2
+
+## 6207. Surgical Repair of Collapsed Attention Heads in ALiBi Transformers
+
+- **ID:** `2603.09616` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Palmer Schallon
+- **abs:** https://arxiv.org/abs/2603.09616 · **PDF:** https://arxiv.org/pdf/2603.09616v1
+
+## 6208. Learning the Hierarchical Organization in Brain Network for Brain Disorder Diagnosis
+
+- **ID:** `2603.09606` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Jingfeng Tang, Peng Cao, Guangqi Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.09606 · **PDF:** https://arxiv.org/pdf/2603.09606v1
+
+## 6209. MSA-Thinker: Discrimination-Calibration Reasoning with Hint-Guided Reinforcement Learning for Multimodal Sentiment Analysis
+
+- **ID:** `2604.00013` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Miaosen Luo, Zhenhao Yang, Jieshen Long, et al.
+- **abs:** https://arxiv.org/abs/2604.00013 · **PDF:** https://arxiv.org/pdf/2604.00013v1
+
+## 6210. Digging Deeper: Learning Multi-Level Concept Hierarchies
+
+- **ID:** `2603.10084` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Oscar Hill, Mateo Espinosa Zarlenga, Mateja Jamnik
+- **abs:** https://arxiv.org/abs/2603.10084 · **PDF:** https://arxiv.org/pdf/2603.10084v1
+
+## 6211. MM-algorithms for traditional and convex NMF with Tweedie and Negative Binomial cost functions and empirical evaluation
+
+- **ID:** `2603.09601` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Elisabeth Sommer James, Asger Hobolth, Marta Pelizzola
+- **abs:** https://arxiv.org/abs/2603.09601 · **PDF:** https://arxiv.org/pdf/2603.09601v1
+
+## 6212. A Variational Latent Equilibrium for Learning in Neuronal Circuits
+
+- **ID:** `2603.09600` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-10
+- **Authors:** Simon Brandt, Paul Haider, Walter Senn, et al.
+- **abs:** https://arxiv.org/abs/2603.09600 · **PDF:** https://arxiv.org/pdf/2603.09600v2
+
+## 6213. Detecting Miscitation on the Scholarly Web through LLM-Augmented Text-Rich Graph Learning
+
+- **ID:** `2603.12290` · **primary:** `cs.IR` · **Submitted:** 2026-03-10
+- **Authors:** Huidong Wu, Haojia Xiang, Jingtong Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.12290 · **PDF:** https://arxiv.org/pdf/2603.12290v1
+
+## 6214. Symbolic Discovery of Stochastic Differential Equations with Genetic Programming
+
+- **ID:** `2603.09597` · **primary:** `cs.NE` · **Submitted:** 2026-03-10
+- **Authors:** Sigur de Vries, Sander W. Keemink, Marcel A. J. van Gerven
+- **abs:** https://arxiv.org/abs/2603.09597 · **PDF:** https://arxiv.org/pdf/2603.09597v1
+
+## 6215. Build, Borrow, or Just Fine-Tune? A Political Scientist's Guide to Choosing NLP Models
+
+- **ID:** `2603.09595` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Shreyas Meher
+- **abs:** https://arxiv.org/abs/2603.09595 · **PDF:** https://arxiv.org/pdf/2603.09595v1
+
+## 6216. Memorization capacity of deep ReLU neural networks characterized by width and depth
+
+- **ID:** `2603.09589` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Xin Yang, Yunfei Yang
+- **abs:** https://arxiv.org/abs/2603.09589 · **PDF:** https://arxiv.org/pdf/2603.09589v1
+
+## 6217. Nonparametric Variational Differential Privacy via Embedding Parameter Clipping
+
+- **ID:** `2603.09583` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Dina El Zein, Shashi Kumar, James Henderson
+- **abs:** https://arxiv.org/abs/2603.09583 · **PDF:** https://arxiv.org/pdf/2603.09583v2
+
+## 6218. Towards Understanding Adam Convergence on Highly Degenerate Polynomials
+
+- **ID:** `2603.09581` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Zhiwei Bai, Jiajie Zhao, Zhangchen Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.09581 · **PDF:** https://arxiv.org/pdf/2603.09581v1
+
+## 6219. Stability of AI Governance Systems: A Coupled Dynamics Model of Public Trust and Social Disruptions
+
+- **ID:** `2603.20248` · **primary:** `cs.CY` · **Submitted:** 2026-03-10
+- **Authors:** Jiaqi Lai, Hou Liang, Weihong Huang
+- **abs:** https://arxiv.org/abs/2603.20248 · **PDF:** https://arxiv.org/pdf/2603.20248v1
+
+## 6220. Routing without Forgetting
+
+- **ID:** `2603.09576` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Alessio Masano, Giovanni Bellitto, Dipam Goswani, et al.
+- **abs:** https://arxiv.org/abs/2603.09576 · **PDF:** https://arxiv.org/pdf/2603.09576v1
+
+## 6221. SCDP: Learning Humanoid Locomotion from Partial Observations via Mixed-Observation Distillation
+
+- **ID:** `2603.09574` · **primary:** `cs.RO` · **Submitted:** 2026-03-10
+- **Authors:** Milo Carroll, Tianhu Peng, Lingfan Bao, et al.
+- **abs:** https://arxiv.org/abs/2603.09574 · **PDF:** https://arxiv.org/pdf/2603.09574v1
+
+## 6222. An Optimal Control Approach To Transformer Training
+
+- **ID:** `2603.09571` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Kağan Akman, Naci Saldı, Serdar Yüksel
+- **abs:** https://arxiv.org/abs/2603.09571 · **PDF:** https://arxiv.org/pdf/2603.09571v1
+
+## 6223. a-TMFG: Scalable Triangulated Maximally Filtered Graphs via Approximate Nearest Neighbors
+
+- **ID:** `2603.09564` · **primary:** `stat.ML` · **Submitted:** 2026-03-10
+- **Authors:** Lionel Yelibi
+- **abs:** https://arxiv.org/abs/2603.09564 · **PDF:** https://arxiv.org/pdf/2603.09564v1
+
+## 6224. Learning Bayesian and Markov Networks with an Unreliable Oracle
+
+- **ID:** `2603.09563` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Juha Harviainen, Pekka Parviainen, Vidya Sagar Sharma
+- **abs:** https://arxiv.org/abs/2603.09563 · **PDF:** https://arxiv.org/pdf/2603.09563v1
+
+## 6225. ALARM: Audio-Language Alignment for Reasoning Models
+
+- **ID:** `2603.09556` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Petr Grinberg, Hassan Shahmohammadi
+- **abs:** https://arxiv.org/abs/2603.09556 · **PDF:** https://arxiv.org/pdf/2603.09556v1
+
+## 6226. Compiler-First State Space Duality and Portable $O(1)$ Autoregressive Caching for Inference
+
+- **ID:** `2603.09555` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Cosmo Santoni
+- **abs:** https://arxiv.org/abs/2603.09555 · **PDF:** https://arxiv.org/pdf/2603.09555v1
+
+## 6227. Decoding the decoder: Contextual sequence-to-sequence modeling for intracortical speech decoding
+
+- **ID:** `2603.20246` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Michal Olak, Tommaso Boccato, Matteo Ferrante
+- **abs:** https://arxiv.org/abs/2603.20246 · **PDF:** https://arxiv.org/pdf/2603.20246v1
+
+## 6228. A Computer-aided Framework for Detecting Osteosarcoma in Computed Tomography Scans
+
+- **ID:** `2603.13376` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Maximo Rodriguez-Herrero, Dante D. Sanchez-Gallegos, Marco Antonio Núñez-Gaona, et al.
+- **abs:** https://arxiv.org/abs/2603.13376 · **PDF:** https://arxiv.org/pdf/2603.13376v1
+
+## 6229. Enhancing Debunking Effectiveness through LLM-based Personality Adaptation
+
+- **ID:** `2603.09533` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Pietro Dell'Oglio, Alessandro Bondielli, Francesco Marcelloni, et al.
+- **abs:** https://arxiv.org/abs/2603.09533 · **PDF:** https://arxiv.org/pdf/2603.09533v1
+
+## 6230. What Do We Care About in Bandits with Noncompliance? BRACE: Bandits with Recommendations, Abstention, and Certified Effects
+
+- **ID:** `2603.09532` · **primary:** `stat.ML` · **Submitted:** 2026-03-10
+- **Authors:** Nicolás Della Penna
+- **abs:** https://arxiv.org/abs/2603.09532 · **PDF:** https://arxiv.org/pdf/2603.09532v1
+
+## 6231. Efficiently Aligning Draft Models via Parameter- and Data-Efficient Adaptation
+
+- **ID:** `2603.09527` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Luxi Lin, Zhihang Lin, Zhanpeng Zeng, et al.
+- **abs:** https://arxiv.org/abs/2603.09527 · **PDF:** https://arxiv.org/pdf/2603.09527v1
+
+## 6232. You Didn't Have to Say It like That: Subliminal Learning from Faithful Paraphrases
+
+- **ID:** `2603.09517` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Isaia Gisler, Zhonghao He, Tianyi Qiu
+- **abs:** https://arxiv.org/abs/2603.09517 · **PDF:** https://arxiv.org/pdf/2603.09517v1
+
+## 6233. Mitigating Frequency Learning Bias in Quantum Models via Multi-Stage Residual Learning
+
+- **ID:** `2603.10083` · **primary:** `quant-ph` · **Submitted:** 2026-03-10
+- **Authors:** Ammar Daskin
+- **abs:** https://arxiv.org/abs/2603.10083 · **PDF:** https://arxiv.org/pdf/2603.10083v1
+
+## 6234. Operationalizing Perceptions of Agent Gender: Foundations and Guidelines
+
+- **ID:** `2603.26682` · **primary:** `cs.HC` · **Submitted:** 2026-03-10
+- **Authors:** Katie Seaborn, Madeleine Steeds, Ilaria Torre, et al.
+- **abs:** https://arxiv.org/abs/2603.26682 · **PDF:** https://arxiv.org/pdf/2603.26682v1
+
+## 6235. TrainDeeploy: Hardware-Accelerated Parameter-Efficient Fine-Tuning of Small Transformer Models at the Extreme Edge
+
+- **ID:** `2603.09511` · **primary:** `cs.AR` · **Submitted:** 2026-03-10
+- **Authors:** Run Wang, Victor J. B. Jung, Philip Wiese, et al.
+- **abs:** https://arxiv.org/abs/2603.09511 · **PDF:** https://arxiv.org/pdf/2603.09511v1
+
+## 6236. Modelling the Diachronic Emergence of Phoneme Frequency Distributions
+
+- **ID:** `2603.09503` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Fermín Moscoso del Prado Martín, Suchir Salhan
+- **abs:** https://arxiv.org/abs/2603.09503 · **PDF:** https://arxiv.org/pdf/2603.09503v1
+
+## 6237. Evolving Prompt Adaptation for Vision-Language Models
+
+- **ID:** `2603.09493` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Enming Zhang, Jiayang Li, Yanru Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.09493 · **PDF:** https://arxiv.org/pdf/2603.09493v1
+
+## 6238. Temporal-Conditioned Normalizing Flows for Multivariate Time Series Anomaly Detection
+
+- **ID:** `2603.09490` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** David Baumgartner, Helge Langseth, Kenth Engø-Monsen, et al.
+- **abs:** https://arxiv.org/abs/2603.09490 · **PDF:** https://arxiv.org/pdf/2603.09490v1
+
+## 6239. Vibe-Creation: The Epistemology of Human-AI Emergent Cognition
+
+- **ID:** `2603.09486` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Ilya Levin
+- **abs:** https://arxiv.org/abs/2603.09486 · **PDF:** https://arxiv.org/pdf/2603.09486v1
+
+## 6240. GenePlan: Evolving Better Generalized PDDL Plans using Large Language Models
+
+- **ID:** `2603.09481` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Andrew Murray, Danial Dervovic, Alberto Pozanco, et al.
+- **abs:** https://arxiv.org/abs/2603.09481 · **PDF:** https://arxiv.org/pdf/2603.09481v1
+
+## 6241. Automatic Analysis of Collaboration Through Human Conversational Data Resources: A Review
+
+- **ID:** `2603.19292` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Yi Yu, Maria Boritchev, Chloé Clavel
+- **abs:** https://arxiv.org/abs/2603.19292 · **PDF:** https://arxiv.org/pdf/2603.19292v3
+
+## 6242. Telogenesis: Goal Is All U Need
+
+- **ID:** `2603.09476` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Zhuoran Deng, Yizhi Zhang, Ziyi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.09476 · **PDF:** https://arxiv.org/pdf/2603.09476v1
+
+## 6243. EvoDriveVLA: Evolving Autonomous Driving Vision-Language-Action Model via Collaborative Perception-Planning Distillation
+
+- **ID:** `2603.09465` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Jiajun Cao, Xiaoan Zhang, Xiaobao Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.09465 · **PDF:** https://arxiv.org/pdf/2603.09465v2
+
+## 6244. An Empirical Study and Theoretical Explanation on Task-Level Model-Merging Collapse
+
+- **ID:** `2603.09463` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Yuan Cao, Dezhi Ran, Yuzhe Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.09463 · **PDF:** https://arxiv.org/pdf/2603.09463v1
+
+## 6245. Declarative Scenario-based Testing with RoadLogic
+
+- **ID:** `2603.09455` · **primary:** `cs.SE` · **Submitted:** 2026-03-10
+- **Authors:** Ezio Bartocci, Alessio Gambi, Felix Gigler, et al.
+- **abs:** https://arxiv.org/abs/2603.09455 · **PDF:** https://arxiv.org/pdf/2603.09455v2
+
+## 6246. InfiniteDance: Scalable 3D Dance Generation Towards in-the-wild Generalization
+
+- **ID:** `2603.13375` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Ronghui Li, Zhongyuan Hu, Li Siyao, et al.
+- **abs:** https://arxiv.org/abs/2603.13375 · **PDF:** https://arxiv.org/pdf/2603.13375v1
+
+## 6247. Variational Routing: A Scalable Bayesian Framework for Calibrated Mixture-of-Experts Transformers
+
+- **ID:** `2603.09453` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Albus Yizhuo Li, Matthew Wicker
+- **abs:** https://arxiv.org/abs/2603.09453 · **PDF:** https://arxiv.org/pdf/2603.09453v2
+
+## 6248. CyberThreat-Eval: Can Large Language Models Automate Real-World Threat Research?
+
+- **ID:** `2603.09452` · **primary:** `cs.CR` · **Submitted:** 2026-03-10
+- **Authors:** Xiangsen Chen, Xuan Feng, Shuo Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.09452 · **PDF:** https://arxiv.org/pdf/2603.09452v1
+
+## 6249. A Guideline-Aware AI Agent for Zero-Shot Target Volume Auto-Delineation
+
+- **ID:** `2603.09448` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Yoon Jo Kim, Wonyoung Cho, Jongmin Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.09448 · **PDF:** https://arxiv.org/pdf/2603.09448v1
+
+## 6250. From Weighting to Modeling: A Nonparametric Estimator for Off-Policy Evaluation
+
+- **ID:** `2603.09436` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Rong J. B. Zhu
+- **abs:** https://arxiv.org/abs/2603.09436 · **PDF:** https://arxiv.org/pdf/2603.09436v1
+
+## 6251. AI Act Evaluation Benchmark: An Open, Transparent, and Reproducible Evaluation Dataset for NLP and RAG Systems
+
+- **ID:** `2603.09435` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Athanasios Davvetas, Michael Papademas, Xenia Ziouvelou, et al.
+- **abs:** https://arxiv.org/abs/2603.09435 · **PDF:** https://arxiv.org/pdf/2603.09435v1
+
+## 6252. Common Sense vs. Morality: The Curious Case of Narrative Focus Bias in LLMs
+
+- **ID:** `2603.09434` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Saugata Purkayastha, Pranav Kushare, Pragya Paramita Pal, et al.
+- **abs:** https://arxiv.org/abs/2603.09434 · **PDF:** https://arxiv.org/pdf/2603.09434v1
+
+## 6253. Impact of Markov Decision Process Design on Sim-to-Real Reinforcement Learning
+
+- **ID:** `2603.09427` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Tatjana Krau, Jorge Mandlmaier, Tobias Damm, et al.
+- **abs:** https://arxiv.org/abs/2603.09427 · **PDF:** https://arxiv.org/pdf/2603.09427v2
+
+## 6254. Amnesia: Adversarial Semantic Layer Specific Activation Steering in Large Language Models
+
+- **ID:** `2603.10080` · **primary:** `cs.CR` · **Submitted:** 2026-03-10
+- **Authors:** Ali Raza, Gurang Gupta, Nikolay Matyunin, et al.
+- **abs:** https://arxiv.org/abs/2603.10080 · **PDF:** https://arxiv.org/pdf/2603.10080v2
+
+## 6255. CERES: A Probabilistic Early Warning System for Acute Food Insecurity
+
+- **ID:** `2603.09425` · **primary:** `stat.AP` · **Submitted:** 2026-03-10
+- **Authors:** Tom Danny S. Pedersen
+- **abs:** https://arxiv.org/abs/2603.09425 · **PDF:** https://arxiv.org/pdf/2603.09425v1
+
+## 6256. Quantizer-Aware Hierarchical Neural Codec Modeling for Speech Deepfake Detection
+
+- **ID:** `2603.16914` · **primary:** `cs.SD` · **Submitted:** 2026-03-10
+- **Authors:** Jinyang Wu, Zihan Pan, Qiquan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.16914 · **PDF:** https://arxiv.org/pdf/2603.16914v1
+
+## 6257. Open-World Motion Forecasting
+
+- **ID:** `2603.09420` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Nicolas Schischka, Nikhil Gosala, B Ravi Kiran, et al.
+- **abs:** https://arxiv.org/abs/2603.09420 · **PDF:** https://arxiv.org/pdf/2603.09420v2
+
+## 6258. Investigating Gender Stereotypes in Large Language Models via Social Determinants of Health
+
+- **ID:** `2603.09416` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Trung Hieu Ngo, Adrien Bazoge, Solen Quiniou, et al.
+- **abs:** https://arxiv.org/abs/2603.09416 · **PDF:** https://arxiv.org/pdf/2603.09416v1
+
+## 6259. From Flow to One Step: Real-Time Multi-Modal Trajectory Policies via Implicit Maximum Likelihood Estimation-based Distribution Distillation
+
+- **ID:** `2603.09415` · **primary:** `cs.RO` · **Submitted:** 2026-03-10
+- **Authors:** Ju Dong, Liding Zhang, Lei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.09415 · **PDF:** https://arxiv.org/pdf/2603.09415v1
+
+## 6260. PromptDLA: A Domain-aware Prompt Document Layout Analysis Framework with Descriptive Knowledge as a Cue
+
+- **ID:** `2603.09414` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Zirui Zhang, Yaping Zhang, Lu Xiang, et al.
+- **abs:** https://arxiv.org/abs/2603.09414 · **PDF:** https://arxiv.org/pdf/2603.09414v1
+
+## 6261. Geometry-Aware Semantic Reasoning for Training Free Video Anomaly Detection
+
+- **ID:** `2603.13374` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Ali Zia, Usman Ali, Muhammad Umer Ramzan, et al.
+- **abs:** https://arxiv.org/abs/2603.13374 · **PDF:** https://arxiv.org/pdf/2603.13374v1
+
+## 6262. Large Spikes in Stochastic Gradient Descent: A Large-Deviations View
+
+- **ID:** `2603.10079` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Benjamin Gess, Daniel Heydecker
+- **abs:** https://arxiv.org/abs/2603.10079 · **PDF:** https://arxiv.org/pdf/2603.10079v1
+
+## 6263. Reconstructing Movement from Sparse Samples: Enhanced Spatio-Temporal Matching Strategies for Low-Frequency Data
+
+- **ID:** `2603.09412` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Ali Yousefian, Arianna Burzacchi, Simone Vantini
+- **abs:** https://arxiv.org/abs/2603.09412 · **PDF:** https://arxiv.org/pdf/2603.09412v2
+
+## 6264. Reviving ConvNeXt for Efficient Convolutional Diffusion Models
+
+- **ID:** `2603.09408` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Taesung Kwon, Lorenzo Bianchi, Lennart Wittke, et al.
+- **abs:** https://arxiv.org/abs/2603.09408 · **PDF:** https://arxiv.org/pdf/2603.09408v1
+
+## 6265. Stochastic Port-Hamiltonian Neural Networks: Universal Approximation with Passivity Guarantees
+
+- **ID:** `2603.10078` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Luca Di Persio, Matthias Ehrhardt, Youness Outaleb
+- **abs:** https://arxiv.org/abs/2603.10078 · **PDF:** https://arxiv.org/pdf/2603.10078v1
+
+## 6266. LLM as a Meta-Judge: Synthetic Data for NLP Evaluation Metric Validation
+
+- **ID:** `2603.09403` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Lukáš Eigler, Jindřich Libovický, David Hurych
+- **abs:** https://arxiv.org/abs/2603.09403 · **PDF:** https://arxiv.org/pdf/2603.09403v1
+
+## 6267. Reward Prediction with Factorized World States
+
+- **ID:** `2603.09400` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Yijun Shen, Delong Chen, Xianming Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.09400 · **PDF:** https://arxiv.org/pdf/2603.09400v1
+
+## 6268. Ethical Fairness without Demographics in Human-Centered AI
+
+- **ID:** `2603.13373` · **primary:** `cs.CY` · **Submitted:** 2026-03-10
+- **Authors:** Shaily Roy, Harshit Sharma, Daniel A. Adler, et al.
+- **abs:** https://arxiv.org/abs/2603.13373 · **PDF:** https://arxiv.org/pdf/2603.13373v2
+
+## 6269. ICDAR 2025 Competition on End-to-End Document Image Machine Translation Towards Complex Layouts
+
+- **ID:** `2603.09392` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Yaping Zhang, Yupu Liang, Zhiyang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.09392 · **PDF:** https://arxiv.org/pdf/2603.09392v1
+
+## 6270. Physics-Informed Neural Engine Sound Modeling with Differentiable Pulse-Train Synthesis
+
+- **ID:** `2603.09391` · **primary:** `cs.SD` · **Submitted:** 2026-03-10
+- **Authors:** Robin Doerfler, Lonce Wyse
+- **abs:** https://arxiv.org/abs/2603.09391 · **PDF:** https://arxiv.org/pdf/2603.09391v1
+
+## 6271. SPAARS: Safer RL Policy Alignment through Abstract Exploration and Refined Exploitation of Action Space
+
+- **ID:** `2603.09378` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Swaminathan S K, Aritra Hazra
+- **abs:** https://arxiv.org/abs/2603.09378 · **PDF:** https://arxiv.org/pdf/2603.09378v2
+
+## 6272. MIL-PF: Multiple Instance Learning on Precomputed Features for Mammography Classification
+
+- **ID:** `2603.09374` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Nikola Jovišić, Milica Škipina, Nicola Dall'Asen, et al.
+- **abs:** https://arxiv.org/abs/2603.09374 · **PDF:** https://arxiv.org/pdf/2603.09374v1
+
+## 6273. Quantifying and extending the coverage of spatial categorization data sets
+
+- **ID:** `2603.09373` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Wanchun Li, Alexandra Carstensen, Yang Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.09373 · **PDF:** https://arxiv.org/pdf/2603.09373v1
+
+## 6274. Flow Field Reconstruction via Voronoi-Enhanced Physics-Informed Neural Networks with End-to-End Sensor Placement Optimization
+
+- **ID:** `2603.09371` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-03-10
+- **Authors:** Renjie Xiao, Bingteng Sun, Yiling Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.09371 · **PDF:** https://arxiv.org/pdf/2603.09371v1
+
+## 6275. From Representation to Clusters: A Contrastive Learning Approach for Attributed Hypergraph Clustering
+
+- **ID:** `2603.09370` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Li Ni, Shuaikang Zeng, Lin Mu, et al.
+- **abs:** https://arxiv.org/abs/2603.09370 · **PDF:** https://arxiv.org/pdf/2603.09370v1
+
+## 6276. M3GCLR: Multi-View Mini-Max Infinite Skeleton-Data Game Contrastive Learning For Skeleton-Based Action Recognition
+
+- **ID:** `2603.09367` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Yanshan Li, Ke Ma, Miaomiao Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.09367 · **PDF:** https://arxiv.org/pdf/2603.09367v1
+
+## 6277. Democratising Clinical AI through Dataset Condensation for Classical Clinical Models
+
+- **ID:** `2603.09356` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Anshul Thakur, Soheila Molaei, Pafue Christy Nganjimi, et al.
+- **abs:** https://arxiv.org/abs/2603.09356 · **PDF:** https://arxiv.org/pdf/2603.09356v1
+
+## 6278. Interactive 3D visualization of surface roughness predictions in additive manufacturing: A data-driven framework
+
+- **ID:** `2603.09353` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Engin Deniz Erkan, Elif Surer, Ulas Yaman
+- **abs:** https://arxiv.org/abs/2603.09353 · **PDF:** https://arxiv.org/pdf/2603.09353v1
+
+## 6279. TA-GGAD: Testing-time Adaptive Graph Model for Generalist Graph Anomaly Detection
+
+- **ID:** `2603.09349` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Xiong Zhang, Hong Peng, Changlong Fu, et al.
+- **abs:** https://arxiv.org/abs/2603.09349 · **PDF:** https://arxiv.org/pdf/2603.09349v1
+
+## 6280. Robust Regularized Policy Iteration under Transition Uncertainty
+
+- **ID:** `2603.09344` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Hongqiang Lin, Zhenghui Fu, Weihao Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.09344 · **PDF:** https://arxiv.org/pdf/2603.09344v2
+
+## 6281. TaSR-RAG: Taxonomy-guided Structured Reasoning for Retrieval-Augmented Generation
+
+- **ID:** `2603.09341` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Jiashuo Sun, Yixuan Xie, Jimeng Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.09341 · **PDF:** https://arxiv.org/pdf/2603.09341v1
+
+## 6282. Beyond Scaling: Assessing Strategic Reasoning and Rapid Decision-Making Capability of LLMs in Zero-sum Environments
+
+- **ID:** `2603.09337` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Yang Li, Xing Chen, Yutao Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.09337 · **PDF:** https://arxiv.org/pdf/2603.09337v1
+
+## 6283. TimberAgent: Gram-Guided Retrieval for Executable Music Effect Control
+
+- **ID:** `2603.09332` · **primary:** `cs.SD` · **Submitted:** 2026-03-10
+- **Authors:** Shihao He, Yihan Xia, Fang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.09332 · **PDF:** https://arxiv.org/pdf/2603.09332v1
+
+## 6284. Reward-Zero: Language Embedding Driven Implicit Reward Mechanisms for Reinforcement Learning
+
+- **ID:** `2603.09331` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Heng Zhang, Haddy Alchaer, Arash Ajoudani, et al.
+- **abs:** https://arxiv.org/abs/2603.09331 · **PDF:** https://arxiv.org/pdf/2603.09331v1
+
+## 6285. Reading the Mood Behind Words: Integrating Prosody-Derived Emotional Context into Socially Responsive VR Agents
+
+- **ID:** `2603.09324` · **primary:** `cs.HC` · **Submitted:** 2026-03-10
+- **Authors:** SangYeop Jeong, Yeongseo Na, Seung Gyu Jeong, et al.
+- **abs:** https://arxiv.org/abs/2603.09324 · **PDF:** https://arxiv.org/pdf/2603.09324v1
+
+## 6286. SpaceSense-Bench: A Large-Scale Multi-Modal Benchmark for Spacecraft Perception and Pose Estimation
+
+- **ID:** `2603.09320` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Aodi Wu, Jianhong Zuo, Zeyuan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.09320 · **PDF:** https://arxiv.org/pdf/2603.09320v1
+
+## 6287. CLoE: Expert Consistency Learning for Missing Modality Segmentation
+
+- **ID:** `2603.09316` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Xinyu Tong, Meihua Zhou, Bowu Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.09316 · **PDF:** https://arxiv.org/pdf/2603.09316v1
+
+## 6288. Curveball Steering: The Right Direction To Steer Isn't Always Linear
+
+- **ID:** `2603.09313` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Shivam Raval, Hae Jin Song, Linlin Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.09313 · **PDF:** https://arxiv.org/pdf/2603.09313v3
+
+## 6289. A Gaussian Comparison Theorem for Training Dynamics in Machine Learning
+
+- **ID:** `2603.09310` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Ashkan Panahi
+- **abs:** https://arxiv.org/abs/2603.09310 · **PDF:** https://arxiv.org/pdf/2603.09310v1
+
+## 6290. Rescaling Confidence: What Scale Design Reveals About LLM Metacognition
+
+- **ID:** `2603.09309` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Yuyang Dai
+- **abs:** https://arxiv.org/abs/2603.09309 · **PDF:** https://arxiv.org/pdf/2603.09309v1
+
+## 6291. A Visualization for Comparative Analysis of Regression Models
+
+- **ID:** `2603.19291` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Nassime Mountasir, Baptiste Lafabregue, Bruno Albert, et al.
+- **abs:** https://arxiv.org/abs/2603.19291 · **PDF:** https://arxiv.org/pdf/2603.19291v1
+
+## 6292. TA-Mem: Tool-Augmented Autonomous Memory Retrieval for LLM in Long-Term Conversational QA
+
+- **ID:** `2603.09297` · **primary:** `cs.IR` · **Submitted:** 2026-03-10
+- **Authors:** Mengwei Yuan, Jianan Liu, Jing Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.09297 · **PDF:** https://arxiv.org/pdf/2603.09297v1
+
+## 6293. Diagnosing and Repairing Citation Failures in Generative Engine Optimization
+
+- **ID:** `2603.09296` · **primary:** `cs.IR` · **Submitted:** 2026-03-10
+- **Authors:** Zhihua Tian, Yuhan Chen, Yao Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.09296 · **PDF:** https://arxiv.org/pdf/2603.09296v1
+
+## 6294. DenoiseSplat: Feed-Forward Gaussian Splatting for Noisy 3D Scene Reconstruction
+
+- **ID:** `2603.09291` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Fuzhen Jiang, Zhuoran Li, Yinlin Zhang
+- **abs:** https://arxiv.org/abs/2603.09291 · **PDF:** https://arxiv.org/pdf/2603.09291v1
+
+## 6295. TASER: Task-Aware Spectral Energy Refine for Backdoor Suppression in UAV Swarms Decentralized Federated Learning
+
+- **ID:** `2603.10075` · **primary:** `cs.CR` · **Submitted:** 2026-03-10
+- **Authors:** Sizhe Huang, Shujie Yang
+- **abs:** https://arxiv.org/abs/2603.10075 · **PDF:** https://arxiv.org/pdf/2603.10075v1
+
+## 6296. Proxy-Guided Measurement Calibration
+
+- **ID:** `2603.09288` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Saketh Vishnubhatla, Shu Wan, Andre Harrison, et al.
+- **abs:** https://arxiv.org/abs/2603.09288 · **PDF:** https://arxiv.org/pdf/2603.09288v3
+
+## 6297. On Regret Bounds of Thompson Sampling for Bayesian Optimization
+
+- **ID:** `2603.09276` · **primary:** `stat.ML` · **Submitted:** 2026-03-10
+- **Authors:** Shion Takeno, Shogo Iwazaki
+- **abs:** https://arxiv.org/abs/2603.09276 · **PDF:** https://arxiv.org/pdf/2603.09276v1
+
+## 6298. DendroNN: Dendrocentric Neural Networks for Energy-Efficient Classification of Event-Based Data
+
+- **ID:** `2603.09274` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Jann Krausse, Zhe Su, Kyrus Mama, et al.
+- **abs:** https://arxiv.org/abs/2603.09274 · **PDF:** https://arxiv.org/pdf/2603.09274v1
+
+## 6299. Logos: An evolvable reasoning engine for rational molecular design
+
+- **ID:** `2603.09268` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Haibin Wen, Zhe Zhao, Fanfu Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.09268 · **PDF:** https://arxiv.org/pdf/2603.09268v1
+
+## 6300. Finding and Reactivating Post-Trained LLMs' Hidden Safety Mechanisms
+
+- **ID:** `2604.00012` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Mingjie Li, Wai Man Si, Michael Backes, et al.
+- **abs:** https://arxiv.org/abs/2604.00012 · **PDF:** https://arxiv.org/pdf/2604.00012v1
+
+## 6301. Transductive Generalization via Optimal Transport and Its Application to Graph Node Classification
+
+- **ID:** `2603.09257` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** MoonJeong Park, Seungbeom Lee, Kyungmin Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.09257 · **PDF:** https://arxiv.org/pdf/2603.09257v1
+
+## 6302. Multi-model approach for autonomous driving: A comprehensive study on traffic sign-, vehicle- and lane detection and behavioral cloning
+
+- **ID:** `2603.09255` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Kanishkha Jaisankar, Pranav M. Pawar, Diana Susane Joseph, et al.
+- **abs:** https://arxiv.org/abs/2603.09255 · **PDF:** https://arxiv.org/pdf/2603.09255v1
+
+## 6303. Efficient Reasoning at Fixed Test-Time Cost via Length-Aware Attention Priors and Gain-Aware Training
+
+- **ID:** `2603.09253` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Rian Atri
+- **abs:** https://arxiv.org/abs/2603.09253 · **PDF:** https://arxiv.org/pdf/2603.09253v1
+
+## 6304. A Generative Sampler for distributions with possible discrete parameter based on Reversibility
+
+- **ID:** `2603.09251` · **primary:** `stat.ML` · **Submitted:** 2026-03-10
+- **Authors:** Lei Li, Zhen Wang, Lishuo Zhang
+- **abs:** https://arxiv.org/abs/2603.09251 · **PDF:** https://arxiv.org/pdf/2603.09251v1
+
+## 6305. Social-R1: Towards Human-like Social Reasoning in LLMs
+
+- **ID:** `2603.09249` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Jincenzi Wu, Yuxuan Lei, Jianxun Lian, et al.
+- **abs:** https://arxiv.org/abs/2603.09249 · **PDF:** https://arxiv.org/pdf/2603.09249v1
+
+## 6306. BridgeDiff: Bridging Human Observations and Flat-Garment Synthesis for Virtual Try-Off
+
+- **ID:** `2603.09236` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Shuang Liu, Ao Yu, Linkang Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.09236 · **PDF:** https://arxiv.org/pdf/2603.09236v1
+
+## 6307. Summarize Before You Speak with ARACH: A Training-Free Inference-Time Plug-In for Enhancing LLMs via Global Attention Reallocation
+
+- **ID:** `2603.11067` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Jingtao Wang, Yucong Wang, Jun Ding, et al.
+- **abs:** https://arxiv.org/abs/2603.11067 · **PDF:** https://arxiv.org/pdf/2603.11067v1
+
+## 6308. How Contrastive Decoding Enhances Large Audio Language Models?
+
+- **ID:** `2603.09232` · **primary:** `cs.SD` · **Submitted:** 2026-03-10
+- **Authors:** Tzu-Quan Lin, Wei-Ping Huang, Yi-Cheng Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.09232 · **PDF:** https://arxiv.org/pdf/2603.09232v1
+
+## 6309. Cognitively Layered Data Synthesis for Domain Adaptation of LLMs to Space Situational Awareness
+
+- **ID:** `2603.09231` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Ding Linghu, Cheng Wang, Da Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.09231 · **PDF:** https://arxiv.org/pdf/2603.09231v1
+
+## 6310. Marginals Before Conditionals
+
+- **ID:** `2603.10074` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Mihir Sahasrabudhe
+- **abs:** https://arxiv.org/abs/2603.10074 · **PDF:** https://arxiv.org/pdf/2603.10074v1
+
+## 6311. LooComp: Leverage Leave-One-Out Strategy to Encoder-only Transformer for Efficient Query-aware Context Compression
+
+- **ID:** `2603.09222` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Thao Do, Dinh Phu Tran, An Vo, et al.
+- **abs:** https://arxiv.org/abs/2603.09222 · **PDF:** https://arxiv.org/pdf/2603.09222v1
+
+## 6312. Beyond Test-Time Training: Learning to Reason via Hardware-Efficient Optimal Control
+
+- **ID:** `2603.09221` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Peihao Wang, Shan Yang, Xijun Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.09221 · **PDF:** https://arxiv.org/pdf/2603.09221v1
+
+## 6313. Embodied Human Simulation for Quantitative Design and Analysis of Interactive Robotics
+
+- **ID:** `2603.09218` · **primary:** `cs.RO` · **Submitted:** 2026-03-10
+- **Authors:** Chenhui Zuo, Jinhao Xu, Michael Qian Vergnolle, et al.
+- **abs:** https://arxiv.org/abs/2603.09218 · **PDF:** https://arxiv.org/pdf/2603.09218v1
+
+## 6314. SPAR-K: Scheduled Periodic Alternating Early Exit for Spoken Language Models
+
+- **ID:** `2603.09215` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Hsiao-Ying Huang, Cheng-Han Chiang, Hung-yi Lee
+- **abs:** https://arxiv.org/abs/2603.09215 · **PDF:** https://arxiv.org/pdf/2603.09215v1
+
+## 6315. Why LLMs Fail: A Failure Analysis and Partial Success Measurement for Automated Security Patch Generation
+
+- **ID:** `2603.10072` · **primary:** `cs.CR` · **Submitted:** 2026-03-10
+- **Authors:** Amir Al-Maamari
+- **abs:** https://arxiv.org/abs/2603.10072 · **PDF:** https://arxiv.org/pdf/2603.10072v1
+
+## 6316. PrivPRISM: Automatically Detecting Discrepancies Between Google Play Data Safety Declarations and Developer Privacy Policies
+
+- **ID:** `2603.09214` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Bhanuka Silva, Dishanika Denipitiyage, Anirban Mahanti, et al.
+- **abs:** https://arxiv.org/abs/2603.09214 · **PDF:** https://arxiv.org/pdf/2603.09214v1
+
+## 6317. Abundant Intelligence and Deficient Demand: A Macro-Financial Stress Test of Rapid AI Adoption
+
+- **ID:** `2603.09209` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Xupeng Chen
+- **abs:** https://arxiv.org/abs/2603.09209 · **PDF:** https://arxiv.org/pdf/2603.09209v1
+
+## 6318. Strategically Robust Multi-Agent Reinforcement Learning with Linear Function Approximation
+
+- **ID:** `2603.09208` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Jake Gonzales, Max Horwitz, Eric Mazumdar, et al.
+- **abs:** https://arxiv.org/abs/2603.09208 · **PDF:** https://arxiv.org/pdf/2603.09208v1
+
+## 6319. MM-Zero: Self-Evolving Multi-Model Vision Language Models From Zero Data
+
+- **ID:** `2603.09206` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Zongxia Li, Hongyang Du, Chengsong Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.09206 · **PDF:** https://arxiv.org/pdf/2603.09206v1
+
+## 6320. Emotion is Not Just a Label: Latent Emotional Factors in LLM Processing
+
+- **ID:** `2603.09205` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Benjamin Reichman, Adar Avsian, Samuel Webster, et al.
+- **abs:** https://arxiv.org/abs/2603.09205 · **PDF:** https://arxiv.org/pdf/2603.09205v2
+
+## 6321. Evaluate-as-Action: Self-Evaluated Process Rewards for Retrieval-Augmented Agents
+
+- **ID:** `2603.09203` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Jiangming Shu, Yuxiang Zhang, Ye Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.09203 · **PDF:** https://arxiv.org/pdf/2603.09203v2
+
+## 6322. The Radio-Frequency Transformer for Signal Separation
+
+- **ID:** `2603.09201` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Egor Lifar, Semyon Savkin, Rachana Madhukara, et al.
+- **abs:** https://arxiv.org/abs/2603.09201 · **PDF:** https://arxiv.org/pdf/2603.09201v1
+
+## 6323. The Reasoning Trap -- Logical Reasoning as a Mechanistic Pathway to Situational Awareness
+
+- **ID:** `2603.09200` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Subramanyam Sahoo, Aman Chadha, Vinija Jain, et al.
+- **abs:** https://arxiv.org/abs/2603.09200 · **PDF:** https://arxiv.org/pdf/2603.09200v1
+
+## 6324. $P^2$GNN: Two Prototype Sets to boost GNN Performance
+
+- **ID:** `2603.09195` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Arihant Jain, Gundeep Arora, Anoop Saladi, et al.
+- **abs:** https://arxiv.org/abs/2603.09195 · **PDF:** https://arxiv.org/pdf/2603.09195v1
+
+## 6325. Explainable Innovation Engine: Dual-Tree Agent-RAG with Methods-as-Nodes and Verifiable Write-Back
+
+- **ID:** `2603.09192` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Renwei Meng
+- **abs:** https://arxiv.org/abs/2603.09192 · **PDF:** https://arxiv.org/pdf/2603.09192v1
+
+## 6326. The Costs of Reproducibility in Music Separation Research: a Replication of Band-Split RNN
+
+- **ID:** `2603.09187` · **primary:** `cs.SD` · **Submitted:** 2026-03-10
+- **Authors:** Paul Magron, Romain Serizel, Constance Douwes
+- **abs:** https://arxiv.org/abs/2603.09187 · **PDF:** https://arxiv.org/pdf/2603.09187v1
+
+## 6327. DEO: Training-Free Direct Embedding Optimization for Negation-Aware Retrieval
+
+- **ID:** `2603.09185` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Taegyeong Lee, Jiwon Park, Seunghyun Hwang, et al.
+- **abs:** https://arxiv.org/abs/2603.09185 · **PDF:** https://arxiv.org/pdf/2603.09185v1
+
+## 6328. Latent-DARM: Bridging Discrete Diffusion And Autoregressive Models For Reasoning
+
+- **ID:** `2603.09184` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Lina Berrayana, Ahmed Heakl, Abdullah Sohail, et al.
+- **abs:** https://arxiv.org/abs/2603.09184 · **PDF:** https://arxiv.org/pdf/2603.09184v1
+
+## 6329. DuplexCascade: Full-Duplex Speech-to-Speech Dialogue with VAD-Free Cascaded ASR-LLM-TTS Pipeline and Micro-Turn Optimization
+
+- **ID:** `2603.09180` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Jianing Yang, Yusuke Fujita, Yui Sudo
+- **abs:** https://arxiv.org/abs/2603.09180 · **PDF:** https://arxiv.org/pdf/2603.09180v1
+
+## 6330. Dissecting Chronos: Sparse Autoencoders Reveal Causal Feature Hierarchies in Time Series Foundation Models
+
+- **ID:** `2603.10071` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Anurag Mishra
+- **abs:** https://arxiv.org/abs/2603.10071 · **PDF:** https://arxiv.org/pdf/2603.10071v1
+
+## 6331. Differentiable Stochastic Traffic Dynamics: Physics-Informed Generative Modelling in Transportation
+
+- **ID:** `2603.09174` · **primary:** `eess.SY` · **Submitted:** 2026-03-10
+- **Authors:** Wuping Xin
+- **abs:** https://arxiv.org/abs/2603.09174 · **PDF:** https://arxiv.org/pdf/2603.09174v1
+
+## 6332. Reinforced Generation of Combinatorial Structures: Ramsey Numbers
+
+- **ID:** `2603.09172` · **primary:** `math.CO` · **Submitted:** 2026-03-10
+- **Authors:** Ansh Nagda, Prabhakar Raghavan, Abhradeep Thakurta
+- **abs:** https://arxiv.org/abs/2603.09172 · **PDF:** https://arxiv.org/pdf/2603.09172v2
+
+## 6333. ZeroWBC: Learning Natural Visuomotor Humanoid Control Directly from Human Egocentric Video
+
+- **ID:** `2603.09170` · **primary:** `cs.RO` · **Submitted:** 2026-03-10
+- **Authors:** Haoran Yang, Jiacheng Bao, Yucheng Xin, et al.
+- **abs:** https://arxiv.org/abs/2603.09170 · **PDF:** https://arxiv.org/pdf/2603.09170v1
+
+## 6334. Better Bounds for the Distributed Experts Problem
+
+- **ID:** `2603.09168` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** David P. Woodruff, Samson Zhou
+- **abs:** https://arxiv.org/abs/2603.09168 · **PDF:** https://arxiv.org/pdf/2603.09168v1
+
+## 6335. Improving Search Agent with One Line of Code
+
+- **ID:** `2603.10069` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Jian Li, Dongsheng Chen, Zhenhua Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.10069 · **PDF:** https://arxiv.org/pdf/2603.10069v1
+
+## 6336. GIAT: A Geologically-Informed Attention Transformer for Lithology Identification
+
+- **ID:** `2603.09165` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Jie Li, Qishun Yang, Nuo Li
+- **abs:** https://arxiv.org/abs/2603.09165 · **PDF:** https://arxiv.org/pdf/2603.09165v1
+
+## 6337. Wrong Code, Right Structure: Learning Netlist Representations from Imperfect LLM-Generated RTL
+
+- **ID:** `2603.09161` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Siyang Cai, Cangyuan Li, Yinhe Han, et al.
+- **abs:** https://arxiv.org/abs/2603.09161 · **PDF:** https://arxiv.org/pdf/2603.09161v1
+
+## 6338. RubiCap: Rubric-Guided Reinforcement Learning for Dense Image Captioning
+
+- **ID:** `2603.09160` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Tzu-Heng Huang, Sirajul Salekin, Javier Movellan, et al.
+- **abs:** https://arxiv.org/abs/2603.09160 · **PDF:** https://arxiv.org/pdf/2603.09160v1
+
+## 6339. Real-Time Trust Verification for Safe Agentic Actions using TrustBench
+
+- **ID:** `2603.09157` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Tavishi Sharma, Vinayak Sharma, Pragya Sharma
+- **abs:** https://arxiv.org/abs/2603.09157 · **PDF:** https://arxiv.org/pdf/2603.09157v1
+
+## 6340. Bioalignment: Measuring and Improving LLM Disposition Toward Biological Systems for AI Safety
+
+- **ID:** `2603.09154` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Trent R Northen, Mingxun Wang
+- **abs:** https://arxiv.org/abs/2603.09154 · **PDF:** https://arxiv.org/pdf/2603.09154v1
+
+## 6341. DataFactory: Collaborative Multi-Agent Framework for Advanced Table Question Answering
+
+- **ID:** `2603.09152` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Tong Wang, Chi Jin, Yongkang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.09152 · **PDF:** https://arxiv.org/pdf/2603.09152v1
+
+## 6342. Deep Tabular Research via Continual Experience-Driven Execution
+
+- **ID:** `2603.09151` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Junnan Dong, Chuang Zhou, Zheng Yuan, et al.
+- **abs:** https://arxiv.org/abs/2603.09151 · **PDF:** https://arxiv.org/pdf/2603.09151v2
+
+## 6343. Causally Sufficient and Necessary Feature Expansion for Class-Incremental Learning
+
+- **ID:** `2603.09145` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Zhen Zhang, Jielei Chu, Tianrui Li
+- **abs:** https://arxiv.org/abs/2603.09145 · **PDF:** https://arxiv.org/pdf/2603.09145v1
+
+## 6344. ADVERSA: Measuring Multi-Turn Guardrail Degradation and Judge Reliability in Large Language Models
+
+- **ID:** `2603.10068` · **primary:** `cs.CR` · **Submitted:** 2026-03-10
+- **Authors:** Harry Owiredu-Ashley
+- **abs:** https://arxiv.org/abs/2603.10068 · **PDF:** https://arxiv.org/pdf/2603.10068v1
+
+## 6345. Chaotic Dynamics in Multi-LLM Deliberation
+
+- **ID:** `2603.09127` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Hajime Shimao, Warut Khern-am-nuai, Sung Joo Kim
+- **abs:** https://arxiv.org/abs/2603.09127 · **PDF:** https://arxiv.org/pdf/2603.09127v1
+
+## 6346. QUSR: Quality-Aware and Uncertainty-Guided Image Super-Resolution Diffusion Model
+
+- **ID:** `2603.09125` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Junjie Yin, Jiaju Li, Hanfa Xing
+- **abs:** https://arxiv.org/abs/2603.09125 · **PDF:** https://arxiv.org/pdf/2603.09125v1
+
+## 6347. DexHiL: A Human-in-the-Loop Framework for Vision-Language-Action Model Post-Training in Dexterous Manipulation
+
+- **ID:** `2603.09121` · **primary:** `cs.RO` · **Submitted:** 2026-03-10
+- **Authors:** Yifan Han, Zhongxi Chen, Yuxuan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.09121 · **PDF:** https://arxiv.org/pdf/2603.09121v1
+
+## 6348. Decoupling Reasoning and Confidence: Resurrecting Calibration in Reinforcement Learning from Verifiable Rewards
+
+- **ID:** `2603.09117` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Zhengzhao Ma, Xueru Wen, Boxi Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.09117 · **PDF:** https://arxiv.org/pdf/2603.09117v1
+
+## 6349. PM-Nav: Priori-Map Guided Embodied Navigation in Functional Buildings
+
+- **ID:** `2603.09113` · **primary:** `cs.RO` · **Submitted:** 2026-03-10
+- **Authors:** Jiang Gao, Xiangyu Dong, Haozhou Li, et al.
+- **abs:** https://arxiv.org/abs/2603.09113 · **PDF:** https://arxiv.org/pdf/2603.09113v1
+
+## 6350. VIVID-Med: LLM-Supervised Structured Pretraining for Deployable Medical ViTs
+
+- **ID:** `2603.09109` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Xiyao Wang, Xiaoyu Tan, Yang Dai, et al.
+- **abs:** https://arxiv.org/abs/2603.09109 · **PDF:** https://arxiv.org/pdf/2603.09109v2
+
+## 6351. Composed Vision-Language Retrieval for Skin Cancer Case Search via Joint Alignment of Global and Local Representations
+
+- **ID:** `2603.09108` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Yuheng Wang, Yuji Lin, Dongrun Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.09108 · **PDF:** https://arxiv.org/pdf/2603.09108v1
+
+## 6352. Probabilistic Hysteresis Factor Prediction for Electric Vehicle Batteries with Graphite Anodes Containing Silicon
+
+- **ID:** `2603.09103` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Runyao Yu, Viviana Kleine, Philipp Gromotka, et al.
+- **abs:** https://arxiv.org/abs/2603.09103 · **PDF:** https://arxiv.org/pdf/2603.09103v1
+
+## 6353. Reading, Not Thinking: Understanding and Bridging the Modality Gap When Text Becomes Pixels in Multimodal LLMs
+
+- **ID:** `2603.09095` · **primary:** `cs.CL` · **Submitted:** 2026-03-10
+- **Authors:** Kaiser Sun, Xiaochuang Yuan, Hongjun Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.09095 · **PDF:** https://arxiv.org/pdf/2603.09095v1
+
+## 6354. HTMuon: Improving Muon via Heavy-Tailed Spectral Correction
+
+- **ID:** `2603.10067` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Tianyu Pang, Yujie Fang, Zihang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.10067 · **PDF:** https://arxiv.org/pdf/2603.10067v1
+
+## 6355. Overcoming Valid Action Suppression in Unmasked Policy Gradient Algorithms
+
+- **ID:** `2603.09090` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Renos Zabounidis, Roy Siegelmann, Mohamad Qadri, et al.
+- **abs:** https://arxiv.org/abs/2603.09090 · **PDF:** https://arxiv.org/pdf/2603.09090v1
+
+## 6356. Exploring Collatz Dynamics with Human-LLM Collaboration
+
+- **ID:** `2603.11066` · **primary:** `math.DS` · **Submitted:** 2026-03-10
+- **Authors:** Edward Y. Chang
+- **abs:** https://arxiv.org/abs/2603.11066 · **PDF:** https://arxiv.org/pdf/2603.11066v4
+
+## 6357. Latent World Models for Automated Driving: A Unified Taxonomy, Evaluation Framework, and Open Challenges
+
+- **ID:** `2603.09086` · **primary:** `cs.RO` · **Submitted:** 2026-03-10
+- **Authors:** Rongxiang Zeng, Yongqi Dong
+- **abs:** https://arxiv.org/abs/2603.09086 · **PDF:** https://arxiv.org/pdf/2603.09086v1
+
+## 6358. Not All News Is Equal: Topic- and Event-Conditional Sentiment from Finetuned LLMs for Aluminum Price Forecasting
+
+- **ID:** `2603.09085` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Alvaro Paredes Amorin, Andre Python, Christoph Weisser
+- **abs:** https://arxiv.org/abs/2603.09085 · **PDF:** https://arxiv.org/pdf/2603.09085v2
+
+## 6359. PPO-Based Hybrid Optimization for RIS-Assisted Semantic Vehicular Edge Computing
+
+- **ID:** `2603.09082` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Wei Feng, Jingbo Zhang, Qiong Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.09082 · **PDF:** https://arxiv.org/pdf/2603.09082v1
+
+## 6360. GST-VLA: Structured Gaussian Spatial Tokens for 3D Depth-Aware Vision-Language-Action Models
+
+- **ID:** `2603.09079` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Md Selim Sarowar, Omer Tariq, Sungho Kim
+- **abs:** https://arxiv.org/abs/2603.09079 · **PDF:** https://arxiv.org/pdf/2603.09079v1
+
+## 6361. Exclusive Self Attention
+
+- **ID:** `2603.09078` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Shuangfei Zhai
+- **abs:** https://arxiv.org/abs/2603.09078 · **PDF:** https://arxiv.org/pdf/2603.09078v1
+
+## 6362. A Text-Native Interface for Generative Video Authoring
+
+- **ID:** `2603.09072` · **primary:** `cs.HC` · **Submitted:** 2026-03-10
+- **Authors:** Xingyu Bruce Liu, Mira Dontcheva, Dingzeyu Li
+- **abs:** https://arxiv.org/abs/2603.09072 · **PDF:** https://arxiv.org/pdf/2603.09072v1
+
+## 6363. Verifying Good Regulator Conditions for Hypergraph Observers: Natural Gradient Learning from Causal Invariance via Established Theorems
+
+- **ID:** `2603.09067` · **primary:** `stat.ML` · **Submitted:** 2026-03-10
+- **Authors:** Max Zhuravlev
+- **abs:** https://arxiv.org/abs/2603.09067 · **PDF:** https://arxiv.org/pdf/2603.09067v1
+
+## 6364. Learning Adaptive LLM Decoding
+
+- **ID:** `2603.09065` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Chloe H. Su, Zhe Ye, Samuel Tenka, et al.
+- **abs:** https://arxiv.org/abs/2603.09065 · **PDF:** https://arxiv.org/pdf/2603.09065v2
+
+## 6365. Dynamic Multi-period Experts for Online Time Series Forecasting
+
+- **ID:** `2603.09062` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Seungha Hong, Sukang Chae, Suyeon Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.09062 · **PDF:** https://arxiv.org/pdf/2603.09062v1
+
+## 6366. Adaptive Active Learning for Online Reliability Prediction of Satellite Electronics
+
+- **ID:** `2603.09058` · **primary:** `stat.ME` · **Submitted:** 2026-03-10
+- **Authors:** Shixiang Li, Yubin Tian, Dianpeng Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.09058 · **PDF:** https://arxiv.org/pdf/2603.09058v1
+
+## 6367. Quality over Quantity: Demonstration Curation via Influence Functions for Data-Centric Robot Learning
+
+- **ID:** `2603.09056` · **primary:** `cs.RO` · **Submitted:** 2026-03-10
+- **Authors:** Haeone Lee, Taywon Min, Junsu Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.09056 · **PDF:** https://arxiv.org/pdf/2603.09056v1
+
+## 6368. Sim2Act: Robust Simulation-to-Decision Learning via Adversarial Calibration and Group-Relative Perturbation
+
+- **ID:** `2603.09053` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Hongyu Cao, Jinghan Zhang, Kunpeng Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.09053 · **PDF:** https://arxiv.org/pdf/2603.09053v1
+
+## 6369. From Days to Minutes: An Autonomous AI Agent Achieves Reliable Clinical Triage in Remote Patient Monitoring
+
+- **ID:** `2603.09052` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Seunghwan Kim, Tiffany H. Kung, Heena Verma, et al.
+- **abs:** https://arxiv.org/abs/2603.09052 · **PDF:** https://arxiv.org/pdf/2603.09052v1
+
+## 6370. EPOCH: An Agentic Protocol for Multi-Round System Optimization
+
+- **ID:** `2603.09049` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Zhanlin Liu, Yitao Li, Munirathnam Srikanth
+- **abs:** https://arxiv.org/abs/2603.09049 · **PDF:** https://arxiv.org/pdf/2603.09049v1
+
+## 6371. FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation
+
+- **ID:** `2603.09046` · **primary:** `cs.CR` · **Submitted:** 2026-03-10
+- **Authors:** Yinpeng Wu, Yitong Chen, Lixiang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.09046 · **PDF:** https://arxiv.org/pdf/2603.09046v1
+
+## 6372. Quantifying Gender Bias in Large Language Models: When ChatGPT Becomes a Hiring Manager
+
+- **ID:** `2604.00011` · **primary:** `cs.CY` · **Submitted:** 2026-03-10
+- **Authors:** Nina Gerszberg, Janka Hamori, Andrew Lo
+- **abs:** https://arxiv.org/abs/2604.00011 · **PDF:** https://arxiv.org/pdf/2604.00011v1
+
+## 6373. Time, Identity and Consciousness in Language Model Agents
+
+- **ID:** `2603.09043` · **primary:** `cs.AI` · **Submitted:** 2026-03-10
+- **Authors:** Elija Perrier, Michael Timothy Bennett
+- **abs:** https://arxiv.org/abs/2603.09043 · **PDF:** https://arxiv.org/pdf/2603.09043v1
+
+## 6374. The Epistemic Support-Point Filter: Jaynesian Maximum Entropy Meets Popperian Falsification
+
+- **ID:** `2603.10065` · **primary:** `cs.IT` · **Submitted:** 2026-03-10
+- **Authors:** Moriba Kemessia Jah
+- **abs:** https://arxiv.org/abs/2603.10065 · **PDF:** https://arxiv.org/pdf/2603.10065v2
+
+## 6375. WS-Net: Weak-Signal Representation Learning and Gated Abundance Reconstruction for Hyperspectral Unmixing via State-Space and Weak Signal Attention Fusion
+
+- **ID:** `2603.09037` · **primary:** `cs.CV` · **Submitted:** 2026-03-10
+- **Authors:** Zekun Long, Ali Zia, Guanyiman Fu, et al.
+- **abs:** https://arxiv.org/abs/2603.09037 · **PDF:** https://arxiv.org/pdf/2603.09037v1
+
+## 6376. SCALAR: Learning and Composing Skills through LLM Guided Symbolic Planning and Deep RL Grounding
+
+- **ID:** `2603.09036` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Renos Zabounidis, Yue Wu, Simon Stepputtis, et al.
+- **abs:** https://arxiv.org/abs/2603.09036 · **PDF:** https://arxiv.org/pdf/2603.09036v1
+
+## 6377. Two Teachers Better Than One: Hardware-Physics Co-Guided Distributed Scientific Machine Learning
+
+- **ID:** `2603.09032` · **primary:** `cs.LG` · **Submitted:** 2026-03-10
+- **Authors:** Yuchen Yuan, Junhuan Yang, Hao Wan, et al.
+- **abs:** https://arxiv.org/abs/2603.09032 · **PDF:** https://arxiv.org/pdf/2603.09032v1
+
+## 6378. PlayWorld: Learning Robot World Models from Autonomous Play
+
+- **ID:** `2603.09030` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Tenny Yin, Zhiting Mei, Zhonghe Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.09030 · **PDF:** https://arxiv.org/pdf/2603.09030v2
+
+## 6379. Automating Detection and Root-Cause Analysis of Flaky Tests in Quantum Software
+
+- **ID:** `2603.09029` · **primary:** `cs.SE` · **Submitted:** 2026-03-09
+- **Authors:** Janakan Sivaloganathan, Ainaz Jamshidi, Andriy Miranskyy, et al.
+- **abs:** https://arxiv.org/abs/2603.09029 · **PDF:** https://arxiv.org/pdf/2603.09029v1
+
+## 6380. When to Retrain after Drift: A Data-Only Test of Post-Drift Data Size Sufficiency
+
+- **ID:** `2603.09024` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Ren Fujiwara, Yasuko Matsubara, Yasushi Sakurai
+- **abs:** https://arxiv.org/abs/2603.09024 · **PDF:** https://arxiv.org/pdf/2603.09024v1
+
+## 6381. The Missing Memory Hierarchy: Demand Paging for LLM Context Windows
+
+- **ID:** `2603.09023` · **primary:** `cs.OS` · **Submitted:** 2026-03-09
+- **Authors:** Tony Mason
+- **abs:** https://arxiv.org/abs/2603.09023 · **PDF:** https://arxiv.org/pdf/2603.09023v1
+
+## 6382. MEMO: Memory-Augmented Model Context Optimization for Robust Multi-Turn Multi-Agent LLM Games
+
+- **ID:** `2603.09022` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Yunfei Xie, Kevin Wang, Bobby Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.09022 · **PDF:** https://arxiv.org/pdf/2603.09022v2
+
+## 6383. AI Phenomenology for Understanding Human-AI Experiences Across Eras
+
+- **ID:** `2603.09020` · **primary:** `cs.HC` · **Submitted:** 2026-03-09
+- **Authors:** Bhada Yun, Evgenia Taranova, Dana Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.09020 · **PDF:** https://arxiv.org/pdf/2603.09020v1
+
+## 6384. Meissa: Multi-modal Medical Agentic Intelligence
+
+- **ID:** `2603.09018` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Yixiong Chen, Xinyi Bai, Yue Pan, et al.
+- **abs:** https://arxiv.org/abs/2603.09018 · **PDF:** https://arxiv.org/pdf/2603.09018v1
+
+## 6385. An accurate flatness measure to estimate the generalization performance of CNN models
+
+- **ID:** `2603.09016` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Rahman Taleghani, Maryam Mohammadi, Francesco Marchetti
+- **abs:** https://arxiv.org/abs/2603.09016 · **PDF:** https://arxiv.org/pdf/2603.09016v1
+
+## 6386. The ARC of Progress towards AGI: A Living Survey of Abstraction and Reasoning
+
+- **ID:** `2603.13372` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Sahar Vahdati, Andrei Aioanei, Haridhra Suresh, et al.
+- **abs:** https://arxiv.org/abs/2603.13372 · **PDF:** https://arxiv.org/pdf/2603.13372v1
+
+## 6387. The Coupling Within: Flow Matching via Distilled Normalizing Flows
+
+- **ID:** `2603.09014` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** David Berthelot, Tianrong Chen, Jiatao Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.09014 · **PDF:** https://arxiv.org/pdf/2603.09014v1
+
+## 6388. Improving through Interaction: Searching Behavioral Representation Spaces with CMA-ES-IG
+
+- **ID:** `2603.09011` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Nathaniel Dennler, Zhonghao Shi, Yiran Tao, et al.
+- **abs:** https://arxiv.org/abs/2603.09011 · **PDF:** https://arxiv.org/pdf/2603.09011v1
+
+## 6389. Statistical Inference via Generative Models: Flow Matching and Causal Inference
+
+- **ID:** `2603.09009` · **primary:** `stat.ML` · **Submitted:** 2026-03-09
+- **Authors:** Shinto Eguchi
+- **abs:** https://arxiv.org/abs/2603.09009 · **PDF:** https://arxiv.org/pdf/2603.09009v1
+
+## 6390. Gender Fairness in Audio Deepfake Detection: Performance and Disparity Analysis
+
+- **ID:** `2603.09007` · **primary:** `cs.SD` · **Submitted:** 2026-03-09
+- **Authors:** Aishwarya Fursule, Shruti Kshirsagar, Anderson R. Avila
+- **abs:** https://arxiv.org/abs/2603.09007 · **PDF:** https://arxiv.org/pdf/2603.09007v1
+
+## 6391. Security Considerations for Multi-agent Systems
+
+- **ID:** `2603.09002` · **primary:** `cs.CR` · **Submitted:** 2026-03-09
+- **Authors:** Tam Nguyen, Moses Ndebugre, Dheeraj Arremsetty
+- **abs:** https://arxiv.org/abs/2603.09002 · **PDF:** https://arxiv.org/pdf/2603.09002v1
+
+## 6392. Learning When to Sample: Confidence-Aware Self-Consistency for Efficient LLM Chain-of-Thought Reasoning
+
+- **ID:** `2603.08999` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Juming Xiong, Kevin Guo, Congning Ni, et al.
+- **abs:** https://arxiv.org/abs/2603.08999 · **PDF:** https://arxiv.org/pdf/2603.08999v2
+
+## 6393. Arbiter: Detecting Interference in LLM Agent System Prompts
+
+- **ID:** `2603.08993` · **primary:** `cs.SE` · **Submitted:** 2026-03-09
+- **Authors:** Tony Mason
+- **abs:** https://arxiv.org/abs/2603.08993 · **PDF:** https://arxiv.org/pdf/2603.08993v1
+
+## 6394. Automated Thematic Analysis for Clinical Qualitative Data: Iterative Codebook Refinement with Full Provenance
+
+- **ID:** `2603.08989` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Seungjun Yi, Joakim Nguyen, Huimin Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.08989 · **PDF:** https://arxiv.org/pdf/2603.08989v1
+
+## 6395. MAPLE: Elevating Medical Reasoning from Statistical Consensus to Process-Led Alignment
+
+- **ID:** `2603.08987` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Kailong Fan, Anqi Pu, Yichen Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.08987 · **PDF:** https://arxiv.org/pdf/2603.08987v1
+
+## 6396. Data-driven robust Markov decision processes on Borel spaces: performance guarantees via an axiomatic approach
+
+- **ID:** `2603.08979` · **primary:** `math.OC` · **Submitted:** 2026-03-09
+- **Authors:** Sivaramakrishnan Ramani
+- **abs:** https://arxiv.org/abs/2603.08979 · **PDF:** https://arxiv.org/pdf/2603.08979v1
+
+## 6397. MAcPNN: Mutual Assisted Learning on Data Streams with Temporal Dependence
+
+- **ID:** `2603.08972` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Federico Giannini, Emanuele Della Valle
+- **abs:** https://arxiv.org/abs/2603.08972 · **PDF:** https://arxiv.org/pdf/2603.08972v1
+
+## 6398. Semantic Level of Detail: Multi-Scale Knowledge Representation via Heat Kernel Diffusion on Hyperbolic Manifolds
+
+- **ID:** `2603.08965` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Edward Izgorodin
+- **abs:** https://arxiv.org/abs/2603.08965 · **PDF:** https://arxiv.org/pdf/2603.08965v1
+
+## 6399. The FABRIC Strategy for Verifying Neural Feedback Systems
+
+- **ID:** `2603.08964` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Samuel I. Akinwande, Sydney M. Katz, Mykel J. Kochenderfer, et al.
+- **abs:** https://arxiv.org/abs/2603.08964 · **PDF:** https://arxiv.org/pdf/2603.08964v2
+
+## 6400. Estimation of heterogeneous principal effects under principal ignorability
+
+- **ID:** `2603.08963` · **primary:** `stat.ME` · **Submitted:** 2026-03-09
+- **Authors:** Rui Zhang, Charles R. Doss, Jared D. Huling
+- **abs:** https://arxiv.org/abs/2603.08963 · **PDF:** https://arxiv.org/pdf/2603.08963v1
+
+## 6401. The $qs$ Inequality: Quantifying the Double Penalty of Mixture-of-Experts at Inference
+
+- **ID:** `2603.08960` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Vignesh Adhinarayanan, Nuwan Jayasena
+- **abs:** https://arxiv.org/abs/2603.08960 · **PDF:** https://arxiv.org/pdf/2603.08960v1
+
+## 6402. Automated Tensor-Relational Decomposition for Large-Scale Sparse Tensor Computation
+
+- **ID:** `2603.08957` · **primary:** `cs.MS` · **Submitted:** 2026-03-09
+- **Authors:** Yuxin Tang, Zhiyuan Xin, Zhimin Ding, et al.
+- **abs:** https://arxiv.org/abs/2603.08957 · **PDF:** https://arxiv.org/pdf/2603.08957v1
+
+## 6403. A Survey of Reinforcement Learning For Economics
+
+- **ID:** `2603.08956` · **primary:** `econ.GN` · **Submitted:** 2026-03-09
+- **Authors:** Pranjal Rawat
+- **abs:** https://arxiv.org/abs/2603.08956 · **PDF:** https://arxiv.org/pdf/2603.08956v5
+
+## 6404. A Consensus-Driven Multi-LLM Pipeline for Missing-Person Investigations
+
+- **ID:** `2603.08954` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Joshua Castillo, Ravi Mukkamala
+- **abs:** https://arxiv.org/abs/2603.08954 · **PDF:** https://arxiv.org/pdf/2603.08954v1
+
+## 6405. Towards Reliable Simulation-based Inference
+
+- **ID:** `2603.08947` · **primary:** `stat.ML` · **Submitted:** 2026-03-09
+- **Authors:** Arnaud Delaunoy
+- **abs:** https://arxiv.org/abs/2603.08947 · **PDF:** https://arxiv.org/pdf/2603.08947v1
+
+## 6406. Kernel Debiased Plug-in Estimation based on the Universal Least Favorable Submodel
+
+- **ID:** `2603.08945` · **primary:** `math.ST` · **Submitted:** 2026-03-09
+- **Authors:** Haiyi Chen, Yang Liu, Ivana Malenica
+- **abs:** https://arxiv.org/abs/2603.08945 · **PDF:** https://arxiv.org/pdf/2603.08945v1
+
+## 6407. BiCLIP: Domain Canonicalization via Structured Geometric Transformation
+
+- **ID:** `2603.08942` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Pranav Mantini, Shishir K. Shah
+- **abs:** https://arxiv.org/abs/2603.08942 · **PDF:** https://arxiv.org/pdf/2603.08942v1
+
+## 6408. Multi-Agent Memory from a Computer Architecture Perspective: Visions and Challenges Ahead
+
+- **ID:** `2603.10062` · **primary:** `cs.AR` · **Submitted:** 2026-03-09
+- **Authors:** Zhongming Yu, Naicheng Yu, Hejia Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.10062 · **PDF:** https://arxiv.org/pdf/2603.10062v2
+
+## 6409. AgentOS: From Application Silos to a Natural Language-Driven Data Ecosystem
+
+- **ID:** `2603.08938` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Rui Liu, Tao Zhe, Dongjie Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.08938 · **PDF:** https://arxiv.org/pdf/2603.08938v2
+
+## 6410. VoxEmo: Benchmarking Speech Emotion Recognition with Speech LLMs
+
+- **ID:** `2603.08936` · **primary:** `cs.SD` · **Submitted:** 2026-03-09
+- **Authors:** Hezhao Zhang, Huang-Cheng Chou, Shrikanth Narayanan, et al.
+- **abs:** https://arxiv.org/abs/2603.08936 · **PDF:** https://arxiv.org/pdf/2603.08936v1
+
+## 6411. PathoScribe: Transforming Pathology Data into a Living Library with a Unified LLM-Driven Framework for Semantic Retrieval and Clinical Integration
+
+- **ID:** `2603.08935` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Abdul Rehman Akbar, Samuel Wales-McGrath, Alejadro Levya, et al.
+- **abs:** https://arxiv.org/abs/2603.08935 · **PDF:** https://arxiv.org/pdf/2603.08935v2
+
+## 6412. Interpretable Markov-Based Spatiotemporal Risk Surfaces for Missing-Child Search Planning with Reinforcement Learning and LLM-Based Quality Assurance
+
+- **ID:** `2603.08933` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Joshua Castillo, Ravi Mukkamala
+- **abs:** https://arxiv.org/abs/2603.08933 · **PDF:** https://arxiv.org/pdf/2603.08933v1
+
+## 6413. Optimizing Reinforcement Learning Training over Digital Twin Enabled Multi-fidelity Networks
+
+- **ID:** `2603.08931` · **primary:** `cs.NI` · **Submitted:** 2026-03-09
+- **Authors:** Hanzhi Yu, Hasan Farooq, Julien Forgeat, et al.
+- **abs:** https://arxiv.org/abs/2603.08931 · **PDF:** https://arxiv.org/pdf/2603.08931v1
+
+## 6414. Using Vision Language Foundation Models to Generate Plant Simulation Configurations via In-Context Learning
+
+- **ID:** `2603.08930` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Heesup Yun, Isaac Kazuo Uyehara, Earl Ranario, et al.
+- **abs:** https://arxiv.org/abs/2603.08930 · **PDF:** https://arxiv.org/pdf/2603.08930v1
+
+## 6415. Functional Bias and Tangent-Space Geometry in Variational Inference
+
+- **ID:** `2603.08925` · **primary:** `math.ST` · **Submitted:** 2026-03-09
+- **Authors:** Sean Plummer
+- **abs:** https://arxiv.org/abs/2603.08925 · **PDF:** https://arxiv.org/pdf/2603.08925v1
+
+## 6416. Quantifying Uncertainty in AI Visibility: A Statistical Framework for Generative Search Measurement
+
+- **ID:** `2603.08924` · **primary:** `stat.AP` · **Submitted:** 2026-03-09
+- **Authors:** Ronald Sielinski
+- **abs:** https://arxiv.org/abs/2603.08924 · **PDF:** https://arxiv.org/pdf/2603.08924v1
+
+## 6417. Tool Receipts, Not Zero-Knowledge Proofs: Practical Hallucination Detection for AI Agents
+
+- **ID:** `2603.10060` · **primary:** `cs.CR` · **Submitted:** 2026-03-09
+- **Authors:** Abhinaba Basu
+- **abs:** https://arxiv.org/abs/2603.10060 · **PDF:** https://arxiv.org/pdf/2603.10060v1
+
+## 6418. Vision-Language Models Encode Clinical Guidelines for Concept-Based Medical Reasoning
+
+- **ID:** `2603.08921` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Mohamed Harmanani, Bining Long, Zhuoxin Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.08921 · **PDF:** https://arxiv.org/pdf/2603.08921v1
+
+## 6419. Uncovering a Winning Lottery Ticket with Continuously Relaxed Bernoulli Gates
+
+- **ID:** `2603.08914` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Itamar Tsayag, Ofir Lindenbaum
+- **abs:** https://arxiv.org/abs/2603.08914 · **PDF:** https://arxiv.org/pdf/2603.08914v1
+
+## 6420. Quantifying Memorization and Privacy Risks in Genomic Language Models
+
+- **ID:** `2603.08913` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Alexander Nemecek, Wenbiao Li, Xiaoqian Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.08913 · **PDF:** https://arxiv.org/pdf/2603.08913v1
+
+## 6421. FedLECC: Cluster- and Loss-Guided Client Selection for Federated Learning under Non-IID Data
+
+- **ID:** `2603.08911` · **primary:** `cs.DC` · **Submitted:** 2026-03-09
+- **Authors:** Daniel M. Jimenez-Gutierrez, Giovanni Giunta, Mehrdad Hassanzadeh, et al.
+- **abs:** https://arxiv.org/abs/2603.08911 · **PDF:** https://arxiv.org/pdf/2603.08911v1
+
+## 6422. SciTaRC: Benchmarking QA on Scientific Tabular Data that Requires Language Reasoning and Complex Computation
+
+- **ID:** `2603.08910` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Hexuan Wang, Yaxuan Ren, Srikar Bommireddypalli, et al.
+- **abs:** https://arxiv.org/abs/2603.08910 · **PDF:** https://arxiv.org/pdf/2603.08910v1
+
+## 6423. Cross-Domain Uncertainty Quantification for Selective Prediction: A Comprehensive Bound Ablation with Transfer-Informed Betting
+
+- **ID:** `2603.08907` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Abhinaba Basu
+- **abs:** https://arxiv.org/abs/2603.08907 · **PDF:** https://arxiv.org/pdf/2603.08907v1
+
+## 6424. NetDiffuser: Deceiving DNN-Based Network Attack Detection Systems with Diffusion-Generated Adversarial Traffic
+
+- **ID:** `2603.08901` · **primary:** `cs.CR` · **Submitted:** 2026-03-09
+- **Authors:** Pratyay Kumar, Abu Saleh Md Tayeen, Satyajayant Misra, et al.
+- **abs:** https://arxiv.org/abs/2603.08901 · **PDF:** https://arxiv.org/pdf/2603.08901v1
+
+## 6425. A New Modeling to Feature Selection Based on the Fuzzy Rough Set Theory in Normal and Optimistic States on Hybrid Information Systems
+
+- **ID:** `2603.08900` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Mohammad Hossein Safarpour, Seyed Majid Alavi, Mohammad Izadikhah, et al.
+- **abs:** https://arxiv.org/abs/2603.08900 · **PDF:** https://arxiv.org/pdf/2603.08900v2
+
+## 6426. ConFu: Contemplate the Future for Better Speculative Sampling
+
+- **ID:** `2603.08899` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Zongyue Qin, Raghavv Goel, Mukul Gagrani, et al.
+- **abs:** https://arxiv.org/abs/2603.08899 · **PDF:** https://arxiv.org/pdf/2603.08899v1
+
+## 6427. From Word2Vec to Transformers: Text-Derived Composition Embeddings for Filtering Combinatorial Electrocatalysts
+
+- **ID:** `2603.08881` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-09
+- **Authors:** Lei Zhang, Markus Stricker
+- **abs:** https://arxiv.org/abs/2603.08881 · **PDF:** https://arxiv.org/pdf/2603.08881v1
+
+## 6428. MultiGraSCCo: A Multilingual Anonymization Benchmark with Annotations of Personal Identifiers
+
+- **ID:** `2603.08879` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Ibrahim Baroud, Christoph Otto, Vera Czehmann, et al.
+- **abs:** https://arxiv.org/abs/2603.08879 · **PDF:** https://arxiv.org/pdf/2603.08879v3
+
+## 6429. Quantifying the Accuracy and Cost Impact of Design Decisions in Budget-Constrained Agentic LLM Search
+
+- **ID:** `2603.08877` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Kyle McCleary, James Ghawaly
+- **abs:** https://arxiv.org/abs/2603.08877 · **PDF:** https://arxiv.org/pdf/2603.08877v1
+
+## 6430. One Language, Two Scripts: Probing Script-Invariance in LLM Concept Representations
+
+- **ID:** `2603.08869` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Sripad Karne
+- **abs:** https://arxiv.org/abs/2603.08869 · **PDF:** https://arxiv.org/pdf/2603.08869v1
+
+## 6431. Why Channel-Centric Models are not Enough to Predict End-to-End Performance in Private 5G: A Measurement Campaign and Case Study
+
+- **ID:** `2603.08865` · **primary:** `cs.NI` · **Submitted:** 2026-03-09
+- **Authors:** Nils Jörgensen
+- **abs:** https://arxiv.org/abs/2603.08865 · **PDF:** https://arxiv.org/pdf/2603.08865v1
+
+## 6432. APPLV: Adaptive Planner Parameter Learning from Vision-Language-Action Model
+
+- **ID:** `2603.08862` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Yuanjie Lu, Beichen Wang, Zhengqi Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.08862 · **PDF:** https://arxiv.org/pdf/2603.08862v1
+
+## 6433. Expressivity-Efficiency Tradeoffs for Hybrid Sequence Models
+
+- **ID:** `2603.08859` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** John Cooper, Ilias Diakonikolas, Mingchen Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.08859 · **PDF:** https://arxiv.org/pdf/2603.08859v1
+
+## 6434. Unpacking Interpretability: Human-Centered Criteria for Optimal Combinatorial Solutions
+
+- **ID:** `2603.08856` · **primary:** `cs.HC` · **Submitted:** 2026-03-09
+- **Authors:** Dominik Pegler, Frank Jäkel, David Steyrl, et al.
+- **abs:** https://arxiv.org/abs/2603.08856 · **PDF:** https://arxiv.org/pdf/2603.08856v1
+
+## 6435. LDP: An Identity-Aware Protocol for Multi-Agent LLM Systems
+
+- **ID:** `2603.08852` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Sunil Prakash
+- **abs:** https://arxiv.org/abs/2603.08852 · **PDF:** https://arxiv.org/pdf/2603.08852v1
+
+## 6436. SBOMs into Agentic AIBOMs: Schema Extensions, Agentic Orchestration, and Reproducibility Evaluation
+
+- **ID:** `2603.10057` · **primary:** `cs.CR` · **Submitted:** 2026-03-09
+- **Authors:** Petar Radanliev, Carsten Maple, Omar Santos, et al.
+- **abs:** https://arxiv.org/abs/2603.10057 · **PDF:** https://arxiv.org/pdf/2603.10057v1
+
+## 6437. A Lightweight Multi-Cancer Tumor Localization Framework for Deployable Digital Pathology
+
+- **ID:** `2603.08844` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Brian Isett, Rebekah Dadey, Aofei Li, et al.
+- **abs:** https://arxiv.org/abs/2603.08844 · **PDF:** https://arxiv.org/pdf/2603.08844v1
+
+## 6438. Can LLMs Perceive Time? An Empirical Investigation
+
+- **ID:** `2604.00010` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Aniketh Garikaparthi
+- **abs:** https://arxiv.org/abs/2604.00010 · **PDF:** https://arxiv.org/pdf/2604.00010v1
+
+## 6439. MASEval: Extending Multi-Agent Evaluation from Models to Systems
+
+- **ID:** `2603.08835` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Cornelius Emde, Alexander Rubinstein, Anmol Goel, et al.
+- **abs:** https://arxiv.org/abs/2603.08835 · **PDF:** https://arxiv.org/pdf/2603.08835v1
+
+## 6440. Are Expressive Encoders Necessary for Discrete Graph Generation?
+
+- **ID:** `2603.08825` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Jay Revolinsky, Harry Shomer, Jiliang Tang
+- **abs:** https://arxiv.org/abs/2603.08825 · **PDF:** https://arxiv.org/pdf/2603.08825v1
+
+## 6441. SoftJAX & SoftTorch: Empowering Automatic Differentiation Libraries with Informative Gradients
+
+- **ID:** `2603.08824` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Anselm Paulus, A. René Geist, Vít Musil, et al.
+- **abs:** https://arxiv.org/abs/2603.08824 · **PDF:** https://arxiv.org/pdf/2603.08824v1
+
+## 6442. Fish Audio S2 Technical Report
+
+- **ID:** `2603.08823` · **primary:** `cs.SD` · **Submitted:** 2026-03-09
+- **Authors:** Shijia Liao, Yuxuan Wang, Songting Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.08823 · **PDF:** https://arxiv.org/pdf/2603.08823v2
+
+## 6443. Beyond Relevance: On the Relationship Between Retrieval and RAG Information Coverage
+
+- **ID:** `2603.08819` · **primary:** `cs.IR` · **Submitted:** 2026-03-09
+- **Authors:** Saron Samuel, Alexander Martin, Eugene Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.08819 · **PDF:** https://arxiv.org/pdf/2603.08819v2
+
+## 6444. Training Language Models via Neural Cellular Automata
+
+- **ID:** `2603.10055` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Dan Lee, Seungwook Han, Akarsh Kumar, et al.
+- **abs:** https://arxiv.org/abs/2603.10055 · **PDF:** https://arxiv.org/pdf/2603.10055v1
+
+## 6445. Agentic LLM Workflow for MR Spectroscopy Volume-of-Interest Placements in Brain Tumors
+
+- **ID:** `2603.13371` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Sangyoon Lee, Francesca Branzoli, Małgorzata Marjańska, et al.
+- **abs:** https://arxiv.org/abs/2603.13371 · **PDF:** https://arxiv.org/pdf/2603.13371v1
+
+## 6446. Scale-Plan: Scalable Language-Enabled Task Planning for Heterogeneous Multi-Robot Teams
+
+- **ID:** `2603.08814` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Piyush Gupta, Sangjae Bae, Jiachen Li, et al.
+- **abs:** https://arxiv.org/abs/2603.08814 · **PDF:** https://arxiv.org/pdf/2603.08814v1
+
+## 6447. Test-Driven AI Agent Definition (TDAD): Compiling Tool-Using Agents from Behavioral Specifications
+
+- **ID:** `2603.08806` · **primary:** `cs.SE` · **Submitted:** 2026-03-09
+- **Authors:** Tzafrir Rehan
+- **abs:** https://arxiv.org/abs/2603.08806 · **PDF:** https://arxiv.org/pdf/2603.08806v1
+
+## 6448. The Temporal Markov Transition Field
+
+- **ID:** `2603.08803` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Michael Leznik
+- **abs:** https://arxiv.org/abs/2603.08803 · **PDF:** https://arxiv.org/pdf/2603.08803v1
+
+## 6449. Large Language Model-Assisted Superconducting Qubit Experiments
+
+- **ID:** `2603.08801` · **primary:** `quant-ph` · **Submitted:** 2026-03-09
+- **Authors:** Shiheng Li, Jacob M. Miller, Phoebe J. Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.08801 · **PDF:** https://arxiv.org/pdf/2603.08801v1
+
+## 6450. Scale Space Diffusion
+
+- **ID:** `2603.08709` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Soumik Mukhopadhyay, Prateksha Udhayanan, Abhinav Shrivastava
+- **abs:** https://arxiv.org/abs/2603.08709 · **PDF:** https://arxiv.org/pdf/2603.08709v1
+
+## 6451. Multi-level meta-reinforcement learning with skill-based curriculum
+
+- **ID:** `2603.08773` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Sichen Yang, Mauro Maggioni
+- **abs:** https://arxiv.org/abs/2603.08773 · **PDF:** https://arxiv.org/pdf/2603.08773v1
+
+## 6452. Impermanent: A Live Benchmark for Temporal Generalization in Time Series Forecasting
+
+- **ID:** `2603.08707` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Azul Garza, Renée Rosillo, Rodrigo Mendoza-Smith, et al.
+- **abs:** https://arxiv.org/abs/2603.08707 · **PDF:** https://arxiv.org/pdf/2603.08707v2
+
+## 6453. Agentic Critical Training
+
+- **ID:** `2603.08706` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Weize Liu, Minghui Liu, Sy-Tuyen Ho, et al.
+- **abs:** https://arxiv.org/abs/2603.08706 · **PDF:** https://arxiv.org/pdf/2603.08706v1
+
+## 6454. Evaluating Financial Intelligence in Large Language Models: Benchmarking SuperInvesting AI with LLM Engines
+
+- **ID:** `2603.08704` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Akshay Gulati, Kanha Singhania, Tushar Banga, et al.
+- **abs:** https://arxiv.org/abs/2603.08704 · **PDF:** https://arxiv.org/pdf/2603.08704v1
+
+## 6455. A Multi-Objective Optimization Approach for Sustainable AI-Driven Entrepreneurship in Resilient Economies
+
+- **ID:** `2603.08692` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Anas ALsobeh, Raneem Alkurdi
+- **abs:** https://arxiv.org/abs/2603.08692 · **PDF:** https://arxiv.org/pdf/2603.08692v1
+
+## 6456. Split Federated Learning Architectures for High-Accuracy and Low-Delay Model Training
+
+- **ID:** `2603.08687` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Yiannis Papageorgiou, Yannis Thomas, Ramin Khalili, et al.
+- **abs:** https://arxiv.org/abs/2603.08687 · **PDF:** https://arxiv.org/pdf/2603.08687v1
+
+## 6457. Benchmarking Language Modeling for Lossless Compression of Full-Fidelity Audio
+
+- **ID:** `2603.08683` · **primary:** `cs.SD` · **Submitted:** 2026-03-09
+- **Authors:** Phillip Long, Zachary Novack, Chris Donahue
+- **abs:** https://arxiv.org/abs/2603.08683 · **PDF:** https://arxiv.org/pdf/2603.08683v1
+
+## 6458. Structural Causal Bottleneck Models
+
+- **ID:** `2603.08682` · **primary:** `stat.ML` · **Submitted:** 2026-03-09
+- **Authors:** Simon Bing, Jonas Wahl, Jakob Runge
+- **abs:** https://arxiv.org/abs/2603.08682 · **PDF:** https://arxiv.org/pdf/2603.08682v2
+
+## 6459. A New Lower Bound for the Random Offerer Mechanism in Bilateral Trade using AI-Guided Evolutionary Search
+
+- **ID:** `2603.08679` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Yang Cai, Vineet Gupta, Zun Li, et al.
+- **abs:** https://arxiv.org/abs/2603.08679 · **PDF:** https://arxiv.org/pdf/2603.08679v1
+
+## 6460. Momentum SVGD-EM for Accelerated Maximum Marginal Likelihood Estimation
+
+- **ID:** `2603.08676` · **primary:** `stat.ML` · **Submitted:** 2026-03-09
+- **Authors:** Adam Rozzio, Rafael Athanasiades, O. Deniz Akyildiz
+- **abs:** https://arxiv.org/abs/2603.08676 · **PDF:** https://arxiv.org/pdf/2603.08676v1
+
+## 6461. Quantization of Ricci Curvature in Information Geometry
+
+- **ID:** `2603.10054` · **primary:** `cs.IT` · **Submitted:** 2026-03-09
+- **Authors:** Carlos C. Rodriguez
+- **abs:** https://arxiv.org/abs/2603.10054 · **PDF:** https://arxiv.org/pdf/2603.10054v1
+
+## 6462. Characterization and upgrade of a quantum graph neural network for charged particle tracking
+
+- **ID:** `2603.08667` · **primary:** `quant-ph` · **Submitted:** 2026-03-09
+- **Authors:** Matteo Argenton, Laura Cappelli, Concezio Bozzi
+- **abs:** https://arxiv.org/abs/2603.08667 · **PDF:** https://arxiv.org/pdf/2603.08667v1
+
+## 6463. How Far Can Unsupervised RLVR Scale LLM Training?
+
+- **ID:** `2603.08660` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Bingxiang He, Yuxin Zuo, Zeyuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.08660 · **PDF:** https://arxiv.org/pdf/2603.08660v1
+
+## 6464. CODA: Difficulty-Aware Compute Allocation for Adaptive Reasoning
+
+- **ID:** `2603.08659` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Siye Wu, Jian Xie, Yikai Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.08659 · **PDF:** https://arxiv.org/pdf/2603.08659v1
+
+## 6465. Context-free Self-Conditioned GAN for Trajectory Forecasting
+
+- **ID:** `2603.08658` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Tiago Rodrigues de Almeida, Eduardo Gutierrez Maestro, Oscar Martinez Mozos
+- **abs:** https://arxiv.org/abs/2603.08658 · **PDF:** https://arxiv.org/pdf/2603.08658v1
+
+## 6466. OfficeQA Pro: An Enterprise Benchmark for End-to-End Grounded Reasoning
+
+- **ID:** `2603.08655` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Krista Opsahl-Ong, Arnav Singhvi, Jasmine Collins, et al.
+- **abs:** https://arxiv.org/abs/2603.08655 · **PDF:** https://arxiv.org/pdf/2603.08655v1
+
+## 6467. Cluster-Aware Attention-Based Deep Reinforcement Learning for Pickup and Delivery Problems
+
+- **ID:** `2603.10053` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Wentao Wang, Lifeng Han, Guangyu Zou
+- **abs:** https://arxiv.org/abs/2603.10053 · **PDF:** https://arxiv.org/pdf/2603.10053v1
+
+## 6468. CoCo: Code as CoT for Text-to-Image Preview and Rare Concept Generation
+
+- **ID:** `2603.08652` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Haodong Li, Chunmei Qing, Huanyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.08652 · **PDF:** https://arxiv.org/pdf/2603.08652v1
+
+## 6469. Group Entropies and Mirror Duality: A Class of Flexible Mirror Descent Updates for Machine Learning
+
+- **ID:** `2603.08651` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Andrzej Cichocki, Piergiulio Tempesta
+- **abs:** https://arxiv.org/abs/2603.08651 · **PDF:** https://arxiv.org/pdf/2603.08651v1
+
+## 6470. Divide and Predict: An Architecture for Input Space Partitioning and Enhanced Accuracy
+
+- **ID:** `2603.08649` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Fenix W. Huang, Henning S. Mortveit, Christian M. Reidys
+- **abs:** https://arxiv.org/abs/2603.08649 · **PDF:** https://arxiv.org/pdf/2603.08649v1
+
+## 6471. Grow, Don't Overwrite: Fine-tuning Without Forgetting
+
+- **ID:** `2603.08647` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Dyah Adila, Hanna Mazzawi, Benoit Dherin, et al.
+- **abs:** https://arxiv.org/abs/2603.08647 · **PDF:** https://arxiv.org/pdf/2603.08647v1
+
+## 6472. Retrieval-Augmented Gaussian Avatars: Improving Expression Generalization
+
+- **ID:** `2603.08645` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Matan Levy, Gavriel Habib, Issar Tzachor, et al.
+- **abs:** https://arxiv.org/abs/2603.08645 · **PDF:** https://arxiv.org/pdf/2603.08645v1
+
+## 6473. Eyla: Toward an Identity-Anchored LLM Architecture with Integrated Biological Priors -- Vision, Implementation Attempt, and Lessons from AI-Assisted Development
+
+- **ID:** `2604.00009` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Arif Aditto
+- **abs:** https://arxiv.org/abs/2604.00009 · **PDF:** https://arxiv.org/pdf/2604.00009v1
+
+## 6474. OmniGuide: Universal Guidance Fields for Enhancing Generalist Robot Policies
+
+- **ID:** `2603.10052` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Yunzhou Song, Long Le, Yong-Hyun Park, et al.
+- **abs:** https://arxiv.org/abs/2603.10052 · **PDF:** https://arxiv.org/pdf/2603.10052v1
+
+## 6475. PostTrainBench: Can LLM Agents Automate LLM Post-Training?
+
+- **ID:** `2603.08640` · **primary:** `cs.SE` · **Submitted:** 2026-03-09
+- **Authors:** Ben Rank, Hardik Bhatnagar, Ameya Prabhu, et al.
+- **abs:** https://arxiv.org/abs/2603.08640 · **PDF:** https://arxiv.org/pdf/2603.08640v2
+
+## 6476. UNBOX: Unveiling Black-box visual models with Natural-language
+
+- **ID:** `2603.08639` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Simone Carnemolla, Chiara Russo, Simone Palazzo, et al.
+- **abs:** https://arxiv.org/abs/2603.08639 · **PDF:** https://arxiv.org/pdf/2603.08639v1
+
+## 6477. Integral Formulas for Vector Spherical Tensor Products
+
+- **ID:** `2603.08630` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Valentin Heyraud, Zachary Weller-Davies, Jules Tilly
+- **abs:** https://arxiv.org/abs/2603.08630 · **PDF:** https://arxiv.org/pdf/2603.08630v1
+
+## 6478. Micro-Diffusion Compression - Binary Tree Tweedie Denoising for Online Probability Estimation
+
+- **ID:** `2603.08771` · **primary:** `stat.ML` · **Submitted:** 2026-03-09
+- **Authors:** Roberto Tacconelli
+- **abs:** https://arxiv.org/abs/2603.08771 · **PDF:** https://arxiv.org/pdf/2603.08771v3
+
+## 6479. Weakly Supervised Teacher-Student Framework with Progressive Pseudo-mask Refinement for Gland Segmentation
+
+- **ID:** `2603.08605` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Hikmat Khan, Wei Chen, Muhammad Khalid Khan Niazi
+- **abs:** https://arxiv.org/abs/2603.08605 · **PDF:** https://arxiv.org/pdf/2603.08605v3
+
+## 6480. Don't Look Back in Anger: MAGIC Net for Streaming Continual Learning with Temporal Dependence
+
+- **ID:** `2603.08600` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Federico Giannini, Sandro D'Andrea, Emanuele Della Valle
+- **abs:** https://arxiv.org/abs/2603.08600 · **PDF:** https://arxiv.org/pdf/2603.08600v1
+
+## 6481. Towards Batch-to-Streaming Deep Reinforcement Learning for Continuous Control
+
+- **ID:** `2603.08588` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Riccardo De Monte, Matteo Cederle, Gian Antonio Susto
+- **abs:** https://arxiv.org/abs/2603.08588 · **PDF:** https://arxiv.org/pdf/2603.08588v1
+
+## 6482. GraphVLM: Benchmarking Vision Language Models for Multimodal Graph Learning
+
+- **ID:** `2603.13370` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Jiajin Liu, Dongzhe Fan, Chuanhao Ji, et al.
+- **abs:** https://arxiv.org/abs/2603.13370 · **PDF:** https://arxiv.org/pdf/2603.13370v1
+
+## 6483. DualFlexKAN: Dual-stage Kolmogorov-Arnold Networks with Independent Function Control
+
+- **ID:** `2603.08583` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Andrés Ortiz, Nicolás J. Gallego-Molina, Carmen Jiménez-Mesa, et al.
+- **abs:** https://arxiv.org/abs/2603.08583 · **PDF:** https://arxiv.org/pdf/2603.08583v1
+
+## 6484. Drift-to-Action Controllers: Budgeted Interventions with Online Risk Certificates
+
+- **ID:** `2603.08578` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Ismail Lamaakal, Chaymae Yahyati, Khalid El Makkaoui, et al.
+- **abs:** https://arxiv.org/abs/2603.08578 · **PDF:** https://arxiv.org/pdf/2603.08578v1
+
+## 6485. Trust via Reputation of Conviction
+
+- **ID:** `2603.08575` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Aravind R. Iyengar
+- **abs:** https://arxiv.org/abs/2603.08575 · **PDF:** https://arxiv.org/pdf/2603.08575v1
+
+## 6486. MetaWorld-X: Hierarchical World Modeling via VLM-Orchestrated Experts for Humanoid Loco-Manipulation
+
+- **ID:** `2603.08572` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Yutong Shen, Hangxu Liu, Penghui Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.08572 · **PDF:** https://arxiv.org/pdf/2603.08572v1
+
+## 6487. OSS-CRS: Liberating AIxCC Cyber Reasoning Systems for Real-World Open-Source Security
+
+- **ID:** `2603.08566` · **primary:** `cs.CR` · **Submitted:** 2026-03-09
+- **Authors:** Andrew Chin, Dongkwan Kim, Yu-Fu Fu, et al.
+- **abs:** https://arxiv.org/abs/2603.08566 · **PDF:** https://arxiv.org/pdf/2603.08566v2
+
+## 6488. RetroAgent: From Solving to Evolving via Retrospective Dual Intrinsic Feedback
+
+- **ID:** `2603.08561` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Xiaoying Zhang, Zichen Liu, Yipeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.08561 · **PDF:** https://arxiv.org/pdf/2603.08561v5
+
+## 6489. Impact of Connectivity on Laplacian Representations in Reinforcement Learning
+
+- **ID:** `2603.08558` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Tommaso Giorgi, Pierriccardo Olivieri, Keyue Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.08558 · **PDF:** https://arxiv.org/pdf/2603.08558v1
+
+## 6490. Generative Adversarial Regression (GAR): Learning Conditional Risk Scenarios
+
+- **ID:** `2603.08553` · **primary:** `stat.ML` · **Submitted:** 2026-03-09
+- **Authors:** Saeed Asadi, Jonathan Yu-Meng Li
+- **abs:** https://arxiv.org/abs/2603.08553 · **PDF:** https://arxiv.org/pdf/2603.08553v1
+
+## 6491. Interactive World Simulator for Robot Policy Training and Evaluation
+
+- **ID:** `2603.08546` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Yixuan Wang, Rhythm Syed, Fangyu Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.08546 · **PDF:** https://arxiv.org/pdf/2603.08546v1
+
+## 6492. The Neural Compass: Probabilistic Relative Feature Fields for Robotic Search
+
+- **ID:** `2603.08544` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Gabriele Somaschini, Adrian Röfer, Abhinav Valada
+- **abs:** https://arxiv.org/abs/2603.08544 · **PDF:** https://arxiv.org/pdf/2603.08544v1
+
+## 6493. Towards Effective and Efficient Graph Alignment without Supervision
+
+- **ID:** `2603.08526` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Songyang Chen, Youfang Lin, Yu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.08526 · **PDF:** https://arxiv.org/pdf/2603.08526v1
+
+## 6494. Breaking the Bias Barrier in Concave Multi-Objective Reinforcement Learning
+
+- **ID:** `2603.08518` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Swetha Ganesh, Vaneet Aggarwal
+- **abs:** https://arxiv.org/abs/2603.08518 · **PDF:** https://arxiv.org/pdf/2603.08518v1
+
+## 6495. Beyond Hungarian: Match-Free Supervision for End-to-End Object Detection
+
+- **ID:** `2603.08514` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Shoumeng Qiu, Xinrun Li, Yang Long
+- **abs:** https://arxiv.org/abs/2603.08514 · **PDF:** https://arxiv.org/pdf/2603.08514v1
+
+## 6496. Oracle-Guided Soft Shielding for Safe Move Prediction in Chess
+
+- **ID:** `2603.08506` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Prajit T Rajendran, Fabio Arnez, Huascar Espinoza, et al.
+- **abs:** https://arxiv.org/abs/2603.08506 · **PDF:** https://arxiv.org/pdf/2603.08506v1
+
+## 6497. Echo2ECG: Enhancing ECG Representations with Cardiac Morphology from Multi-View Echos
+
+- **ID:** `2603.08505` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Michelle Espranita Liman, Özgün Turgut, Alexander Müller, et al.
+- **abs:** https://arxiv.org/abs/2603.08505 · **PDF:** https://arxiv.org/pdf/2603.08505v1
+
+## 6498. Fanar-Sadiq: A Multi-Agent Architecture for Grounded Islamic QA
+
+- **ID:** `2603.08501` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Ummar Abbas, Mourad Ouzzani, Mohamed Y. Eltabakh, et al.
+- **abs:** https://arxiv.org/abs/2603.08501 · **PDF:** https://arxiv.org/pdf/2603.08501v2
+
+## 6499. Efficient Credal Prediction through Decalibration
+
+- **ID:** `2603.08495` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Paul Hofman, Timo Löhr, Maximilian Muschalik, et al.
+- **abs:** https://arxiv.org/abs/2603.08495 · **PDF:** https://arxiv.org/pdf/2603.08495v1
+
+## 6500. First-Order Geometry, Spectral Compression, and Structural Compatibility under Bounded Computation
+
+- **ID:** `2603.08494` · **primary:** `math.OC` · **Submitted:** 2026-03-09
+- **Authors:** Changkai Li
+- **abs:** https://arxiv.org/abs/2603.08494 · **PDF:** https://arxiv.org/pdf/2603.08494v1
+
+## 6501. Pareto-Optimal Anytime Algorithms via Bayesian Racing
+
+- **ID:** `2603.08493` · **primary:** `cs.NE` · **Submitted:** 2026-03-09
+- **Authors:** Jonathan Wurth, Helena Stegherr, Neele Kemper, et al.
+- **abs:** https://arxiv.org/abs/2603.08493 · **PDF:** https://arxiv.org/pdf/2603.08493v1
+
+## 6502. NN-OpInf: an operator inference approach using structure-preserving composable neural networks
+
+- **ID:** `2603.08488` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Eric Parish, Anthony Gruber, Patrick Blonigan, et al.
+- **abs:** https://arxiv.org/abs/2603.08488 · **PDF:** https://arxiv.org/pdf/2603.08488v1
+
+## 6503. How Trustworthy Are LLM-as-Judge Ratings for Interpretive Responses? Implications for Qualitative Research Workflows
+
+- **ID:** `2604.00008` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Songhee Han, Jueun Shin, Jiyoon Han, et al.
+- **abs:** https://arxiv.org/abs/2604.00008 · **PDF:** https://arxiv.org/pdf/2604.00008v1
+
+## 6504. Visual Self-Fulfilling Alignment: Shaping Safety-Oriented Personas via Threat-Related Images
+
+- **ID:** `2603.08486` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Qishun Yang, Shu Yang, Lijie Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.08486 · **PDF:** https://arxiv.org/pdf/2603.08486v1
+
+## 6505. X-AVDT: Audio-Visual Cross-Attention for Robust Deepfake Detection
+
+- **ID:** `2603.08483` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Youngseo Kim, Kwan Yun, Seokhyeon Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.08483 · **PDF:** https://arxiv.org/pdf/2603.08483v1
+
+## 6506. Where Do Flow Semantics Reside? A Protocol-Native Tabular Pretraining Paradigm for Encrypted Traffic Classification
+
+- **ID:** `2603.10051` · **primary:** `cs.NI` · **Submitted:** 2026-03-09
+- **Authors:** Sizhe Huang, Shujie Yang
+- **abs:** https://arxiv.org/abs/2603.10051 · **PDF:** https://arxiv.org/pdf/2603.10051v1
+
+## 6507. STRIDE: Structured Lagrangian and Stochastic Residual Dynamics via Flow Matching
+
+- **ID:** `2603.08478` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Prakrut Kotecha, Ganga Nair B, Shishir Kolathaya
+- **abs:** https://arxiv.org/abs/2603.08478 · **PDF:** https://arxiv.org/pdf/2603.08478v2
+
+## 6508. R2F: Repurposing Ray Frontiers for LLM-free Object Navigation
+
+- **ID:** `2603.08475` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Francesco Argenziano, John Mark Alexis Marcelo, Michele Brienza, et al.
+- **abs:** https://arxiv.org/abs/2603.08475 · **PDF:** https://arxiv.org/pdf/2603.08475v1
+
+## 6509. Integrating Lagrangian Neural Networks into the Dyna Framework for Reinforcement Learning
+
+- **ID:** `2603.08468` · **primary:** `eess.SY` · **Submitted:** 2026-03-09
+- **Authors:** Shreya Das, Kundan Kumar, Muhammad Iqbal, et al.
+- **abs:** https://arxiv.org/abs/2603.08468 · **PDF:** https://arxiv.org/pdf/2603.08468v1
+
+## 6510. MUSA-PINN: Multi-scale Weak-form Physics-Informed Neural Networks for Fluid Flow in Complex Geometries
+
+- **ID:** `2603.08465` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Weizheng Zhang, Xunjie Xie, Hao Pan, et al.
+- **abs:** https://arxiv.org/abs/2603.08465 · **PDF:** https://arxiv.org/pdf/2603.08465v2
+
+## 6511. Evolving Symbiosis, from Barricelli's Legacy to Collective Intelligence: a simulated and conceptual approach
+
+- **ID:** `2603.08463` · **primary:** `cs.NE` · **Submitted:** 2026-03-09
+- **Authors:** James Ashford, Marko Cvjetko, Richard Löffler, et al.
+- **abs:** https://arxiv.org/abs/2603.08463 · **PDF:** https://arxiv.org/pdf/2603.08463v1
+
+## 6512. Reasoning as Compression: Unifying Budget Forcing via the Conditional Information Bottleneck
+
+- **ID:** `2603.08462` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Fabio Valerio Massoli, Andrey Kuzmin, Arash Behboodi
+- **abs:** https://arxiv.org/abs/2603.08462 · **PDF:** https://arxiv.org/pdf/2603.08462v1
+
+## 6513. Data-Driven Priors for Uncertainty-Aware Deterioration Risk Prediction with Multimodal Data
+
+- **ID:** `2603.08459` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** L. Julián Lechuga López, Tim G. J. Rudner, Farah E. Shamout
+- **abs:** https://arxiv.org/abs/2603.08459 · **PDF:** https://arxiv.org/pdf/2603.08459v1
+
+## 6514. Adaptive Entropy-Driven Sensor Selection in a Camera-LiDAR Particle Filter for Single-Vessel Tracking
+
+- **ID:** `2603.08457` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Andrei Starodubov, Yaqub Aris Prabowo, Andreas Hadjipieris, et al.
+- **abs:** https://arxiv.org/abs/2603.08457 · **PDF:** https://arxiv.org/pdf/2603.08457v1
+
+## 6515. The Boiling Frog Threshold: Criticality and Blindness in World Model-Based Anomaly Detection Under Gradual Drift
+
+- **ID:** `2603.08455` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Zhe Hong
+- **abs:** https://arxiv.org/abs/2603.08455 · **PDF:** https://arxiv.org/pdf/2603.08455v1
+
+## 6516. LycheeCluster: Efficient Long-Context Inference with Structure-Aware Chunking and Hierarchical KV Indexing
+
+- **ID:** `2603.08453` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Dongfang Li, Zixuan Liu, Gang Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.08453 · **PDF:** https://arxiv.org/pdf/2603.08453v1
+
+## 6517. A Dataset for Probing Translationese Preferences in English-to-Swedish Translation
+
+- **ID:** `2603.08450` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Jenny Kunz, Anja Jarochenko, Marcel Bollmann
+- **abs:** https://arxiv.org/abs/2603.08450 · **PDF:** https://arxiv.org/pdf/2603.08450v1
+
+## 6518. A prospective clinical feasibility study of a conversational diagnostic AI in an ambulatory primary care clinic
+
+- **ID:** `2603.08448` · **primary:** `cs.HC` · **Submitted:** 2026-03-09
+- **Authors:** Peter Brodeur, Jacob M. Koshy, Anil Palepu, et al.
+- **abs:** https://arxiv.org/abs/2603.08448 · **PDF:** https://arxiv.org/pdf/2603.08448v3
+
+## 6519. Efficient Policy Learning with Hybrid Evaluation-Based Genetic Programming for Uncertain Agile Earth Observation Satellite Scheduling
+
+- **ID:** `2603.08447` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Junhua Xue, Yuning Chen, Mingyan Shao, et al.
+- **abs:** https://arxiv.org/abs/2603.08447 · **PDF:** https://arxiv.org/pdf/2603.08447v2
+
+## 6520. Disentangling Prompt Dependence to Evaluate Segmentation Reliability in Gynecological MRI
+
+- **ID:** `2603.13369` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Elodie Germani, Krystel Nyangoh-Timoh, Pierre Jannin, et al.
+- **abs:** https://arxiv.org/abs/2603.13369 · **PDF:** https://arxiv.org/pdf/2603.13369v1
+
+## 6521. Can Vision-Language Models Solve the Shell Game?
+
+- **ID:** `2603.08436` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Tiedong Liu, Wee Sun Lee
+- **abs:** https://arxiv.org/abs/2603.08436 · **PDF:** https://arxiv.org/pdf/2603.08436v1
+
+## 6522. One Model Is Enough: Native Retrieval Embeddings from LLM Agent Hidden States
+
+- **ID:** `2603.08429` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Bo Jiang
+- **abs:** https://arxiv.org/abs/2603.08429 · **PDF:** https://arxiv.org/pdf/2603.08429v1
+
+## 6523. Grow, Assess, Compress: Adaptive Backbone Scaling for Memory-Efficient Class Incremental Learning
+
+- **ID:** `2603.08426` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Adrian Garcia-Castañeda, Jon Irureta, Jon Imaz, et al.
+- **abs:** https://arxiv.org/abs/2603.08426 · **PDF:** https://arxiv.org/pdf/2603.08426v2
+
+## 6524. IronEngine: Towards General AI Assistant
+
+- **ID:** `2603.08425` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Xi Mo
+- **abs:** https://arxiv.org/abs/2603.08425 · **PDF:** https://arxiv.org/pdf/2603.08425v1
+
+## 6525. SYNAPSE: Framework for Neuron Analysis and Perturbation in Sequence Encoding
+
+- **ID:** `2603.08424` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Jesús Sánchez Ochoa, Enrique Tomás Martínez Beltrán, Alberto Huertas Celdrán
+- **abs:** https://arxiv.org/abs/2603.08424 · **PDF:** https://arxiv.org/pdf/2603.08424v1
+
+## 6526. Human-Aware Robot Behaviour in Self-Driving Labs
+
+- **ID:** `2603.08420` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Satheeshkumar Veeramani, Anna Kisil, Abigail Bentley, et al.
+- **abs:** https://arxiv.org/abs/2603.08420 · **PDF:** https://arxiv.org/pdf/2603.08420v2
+
+## 6527. Meta-RL with Shared Representations Enables Fast Adaptation in Energy Systems
+
+- **ID:** `2603.08418` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Théo Zangato, Aomar Osmani, Pegah Alizadeh
+- **abs:** https://arxiv.org/abs/2603.08418 · **PDF:** https://arxiv.org/pdf/2603.08418v1
+
+## 6528. Geometrically Constrained Outlier Synthesis
+
+- **ID:** `2603.08413` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Daniil Karzanov, Marcin Detyniecki
+- **abs:** https://arxiv.org/abs/2603.08413 · **PDF:** https://arxiv.org/pdf/2603.08413v1
+
+## 6529. Aligning to Illusions: Choice Blindness in Human and AI Feedback
+
+- **ID:** `2603.08412` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Wenbin Wu
+- **abs:** https://arxiv.org/abs/2603.08412 · **PDF:** https://arxiv.org/pdf/2603.08412v1
+
+## 6530. Real-Time Monocular Scene Analysis for UAV in Outdoor Environments
+
+- **ID:** `2603.13368` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Yara AlaaEldin
+- **abs:** https://arxiv.org/abs/2603.13368 · **PDF:** https://arxiv.org/pdf/2603.13368v1
+
+## 6531. Sandpiper: Orchestrated AI-Annotation for Educational Discourse at Scale
+
+- **ID:** `2603.08406` · **primary:** `cs.HC` · **Submitted:** 2026-03-09
+- **Authors:** Daryl Hedley, Doug Pietrzak, Jorge Dias, et al.
+- **abs:** https://arxiv.org/abs/2603.08406 · **PDF:** https://arxiv.org/pdf/2603.08406v1
+
+## 6532. Dynin-Omni: Omnimodal Unified Large Diffusion Language Model
+
+- **ID:** `2604.00007` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Jaeik Kim, Woojin Kim, Jihwan Hong, et al.
+- **abs:** https://arxiv.org/abs/2604.00007 · **PDF:** https://arxiv.org/pdf/2604.00007v1
+
+## 6533. A Recipe for Stable Offline Multi-agent Reinforcement Learning
+
+- **ID:** `2603.08399` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Dongsu Lee, Daehee Lee, Amy Zhang
+- **abs:** https://arxiv.org/abs/2603.08399 · **PDF:** https://arxiv.org/pdf/2603.08399v1
+
+## 6534. Revealing Behavioral Plasticity in Large Language Models: A Token-Conditional Perspective
+
+- **ID:** `2603.08398` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Liyuan Mao, Le Yu, Jing Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.08398 · **PDF:** https://arxiv.org/pdf/2603.08398v1
+
+## 6535. Decoupling Distance and Networks: Hybrid Graph Attention-Geostatistical Methods for Spatio-temporal Risk Mapping
+
+- **ID:** `2603.08393` · **primary:** `stat.ME` · **Submitted:** 2026-03-09
+- **Authors:** Toba Temitope Bamidele, Ezra Gayawan, Femi Barnabas Adebola, et al.
+- **abs:** https://arxiv.org/abs/2603.08393 · **PDF:** https://arxiv.org/pdf/2603.08393v1
+
+## 6536. COACH meets QUORUM: A Framework and Pipeline for Aligning User, Expert and Developer Perspectives in LLM-generated Health Counselling
+
+- **ID:** `2603.08392` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Yee Man Ng, Bram van Dijk, Pieter Beynen, et al.
+- **abs:** https://arxiv.org/abs/2603.08392 · **PDF:** https://arxiv.org/pdf/2603.08392v1
+
+## 6537. Adaptive Loops and Memory in Transformers: Think Harder or Know More?
+
+- **ID:** `2603.08391` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Markus Frey, Behzad Shomali, Ali Hamza Bashir, et al.
+- **abs:** https://arxiv.org/abs/2603.08391 · **PDF:** https://arxiv.org/pdf/2603.08391v3
+
+## 6538. A Hierarchical Error-Corrective Graph Framework for Autonomous Agents with LLM-Based Action Generation
+
+- **ID:** `2603.08388` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Cong Cao, Jingyao Zhang, Kun Tong
+- **abs:** https://arxiv.org/abs/2603.08388 · **PDF:** https://arxiv.org/pdf/2603.08388v3
+
+## 6539. InFusionLayer: a CFA-based ensemble tool to generate new classifiers for learning and modeling
+
+- **ID:** `2603.10049` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Eric Roginek, Jingyan Xu, D. Frank. Hsu
+- **abs:** https://arxiv.org/abs/2603.10049 · **PDF:** https://arxiv.org/pdf/2603.10049v1
+
+## 6540. Structure from rank: Rank-order coding as a bridge from sequence to structure
+
+- **ID:** `2603.08380` · **primary:** `cs.NE` · **Submitted:** 2026-03-09
+- **Authors:** Xiaodan Chen, Alexandre Pitti, Mathias Quoy, et al.
+- **abs:** https://arxiv.org/abs/2603.08380 · **PDF:** https://arxiv.org/pdf/2603.08380v2
+
+## 6541. Beyond the Markovian Assumption: Robust Optimization via Fractional Weyl Integrals in Imbalanced Data
+
+- **ID:** `2603.08377` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Gustavo A. Dorrego
+- **abs:** https://arxiv.org/abs/2603.08377 · **PDF:** https://arxiv.org/pdf/2603.08377v1
+
+## 6542. Leaderboard Incentives: Model Rankings under Strategic Post-Training
+
+- **ID:** `2603.08371` · **primary:** `cs.GT` · **Submitted:** 2026-03-09
+- **Authors:** Yatong Chen, Guanhua Zhang, Moritz Hardt
+- **abs:** https://arxiv.org/abs/2603.08371 · **PDF:** https://arxiv.org/pdf/2603.08371v1
+
+## 6543. Unifying On- and Off-Policy Variance Reduction Methods
+
+- **ID:** `2603.08370` · **primary:** `stat.ML` · **Submitted:** 2026-03-09
+- **Authors:** Olivier Jeunen
+- **abs:** https://arxiv.org/abs/2603.08370 · **PDF:** https://arxiv.org/pdf/2603.08370v1
+
+## 6544. M$^3$-ACE: Rectifying Visual Perception in Multimodal Math Reasoning via Multi-Agentic Context Engineering
+
+- **ID:** `2603.08369` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Peijin Xie, Zhen Xu, Bingquan Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.08369 · **PDF:** https://arxiv.org/pdf/2603.08369v1
+
+## 6545. Computational modeling of early language learning from acoustic speech and audiovisual input without linguistic priors
+
+- **ID:** `2603.08359` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Okko Räsänen
+- **abs:** https://arxiv.org/abs/2603.08359 · **PDF:** https://arxiv.org/pdf/2603.08359v2
+
+## 6546. Do Language Models Know Theo Has a Wife? Investigating the Proviso Problem
+
+- **ID:** `2603.08358` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Tara Azin, Daniel Dumitrescu, Diana Inkpen, et al.
+- **abs:** https://arxiv.org/abs/2603.08358 · **PDF:** https://arxiv.org/pdf/2603.08358v1
+
+## 6547. Scalable Identification and Prioritization of Requisition-Specific Personal Competencies Using Large Language Models
+
+- **ID:** `2604.00006` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Wanxin Li, Denver McNeney, Nivedita Prabhu, et al.
+- **abs:** https://arxiv.org/abs/2604.00006 · **PDF:** https://arxiv.org/pdf/2604.00006v1
+
+## 6548. Towards plausibility in time series counterfactual explanations
+
+- **ID:** `2603.08349` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Marcin Kostrzewa, Krzysztof Galus, Maciej Zięba
+- **abs:** https://arxiv.org/abs/2603.08349 · **PDF:** https://arxiv.org/pdf/2603.08349v1
+
+## 6549. Rethinking Attention Output Projection: Structured Hadamard Transforms for Efficient Transformers
+
+- **ID:** `2603.08343` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Shubham Aggarwal, Lokendra Kumar
+- **abs:** https://arxiv.org/abs/2603.08343 · **PDF:** https://arxiv.org/pdf/2603.08343v2
+
+## 6550. Electrocardiogram Classification with Transformers Using Koopman and Wavelet Features
+
+- **ID:** `2603.08339` · **primary:** `eess.SP` · **Submitted:** 2026-03-09
+- **Authors:** Sucheta Ghosh, Zahra Monfared
+- **abs:** https://arxiv.org/abs/2603.08339 · **PDF:** https://arxiv.org/pdf/2603.08339v1
+
+## 6551. Multimodal Deep Learning for Dynamic and Static Neuroimaging: Integrating MRI and fMRI for Alzheimer Disease Analysis
+
+- **ID:** `2603.13367` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Anima Kujur, Zahra Monfared
+- **abs:** https://arxiv.org/abs/2603.13367 · **PDF:** https://arxiv.org/pdf/2603.13367v1
+
+## 6552. Detecting Fake Reviewer Groups in Dynamic Networks: An Adaptive Graph Learning Method
+
+- **ID:** `2603.08332` · **primary:** `cs.SI` · **Submitted:** 2026-03-09
+- **Authors:** Jing Zhang, Ke Huang, Yao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.08332 · **PDF:** https://arxiv.org/pdf/2603.08332v1
+
+## 6553. Thinking in Uncertainty: Mitigating Hallucinations in MLRMs with Latent Entropy-Aware Decoding
+
+- **ID:** `2603.13366` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Zhongxing Xu, Zhonghua Wang, Zhe Qian, et al.
+- **abs:** https://arxiv.org/abs/2603.13366 · **PDF:** https://arxiv.org/pdf/2603.13366v1
+
+## 6554. SPD-RAG: Sub-Agent Per Document Retrieval-Augmented Generation
+
+- **ID:** `2603.08329` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Yagiz Can Akay, Muhammed Yusuf Kartal, Esra Alparslan, et al.
+- **abs:** https://arxiv.org/abs/2603.08329 · **PDF:** https://arxiv.org/pdf/2603.08329v1
+
+## 6555. Beyond Attention Heatmaps: How to Get Better Explanations for Multiple Instance Learning Models in Histopathology
+
+- **ID:** `2603.08328` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Mina Jamshidi Idaji, Julius Hense, Tom Neuhäuser, et al.
+- **abs:** https://arxiv.org/abs/2603.08328 · **PDF:** https://arxiv.org/pdf/2603.08328v1
+
+## 6556. EndoSERV: A Vision-based Endoluminal Robot Navigation System
+
+- **ID:** `2603.08324` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Junyang Wu, Fangfang Xie, Minghui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.08324 · **PDF:** https://arxiv.org/pdf/2603.08324v1
+
+## 6557. Agentic Neurosymbolic Collaboration for Mathematical Discovery: A Case Study in Combinatorial Design
+
+- **ID:** `2603.08322` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Hai Xia, Carla P. Gomes, Bart Selman, et al.
+- **abs:** https://arxiv.org/abs/2603.08322 · **PDF:** https://arxiv.org/pdf/2603.08322v1
+
+## 6558. CORE-Acu: Structured Reasoning Traces and Knowledge Graph Safety Verification for Acupuncture Clinical Decision Support
+
+- **ID:** `2603.08321` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Liuyi Xu, Yun Guo, Ming Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.08321 · **PDF:** https://arxiv.org/pdf/2603.08321v1
+
+## 6559. Human-AI Divergence in Ego-centric Action Recognition under Spatial and Spatiotemporal Manipulations
+
+- **ID:** `2603.08317` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Sadegh Rahmaniboldaji, Filip Rybansky, Quoc C. Vuong, et al.
+- **abs:** https://arxiv.org/abs/2603.08317 · **PDF:** https://arxiv.org/pdf/2603.08317v1
+
+## 6560. SlowBA: An efficiency backdoor attack towards VLM-based GUI agents
+
+- **ID:** `2603.08316` · **primary:** `cs.CR` · **Submitted:** 2026-03-09
+- **Authors:** Junxian Li, Tu Lan, Haozhen Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.08316 · **PDF:** https://arxiv.org/pdf/2603.08316v2
+
+## 6561. Learning Multiple Utterance-Level Attribute Representations with a Unified Speech Encoder
+
+- **ID:** `2603.08312` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Maryem Bouziane, Salima Mdhaffar, Yannick Estève
+- **abs:** https://arxiv.org/abs/2603.08312 · **PDF:** https://arxiv.org/pdf/2603.08312v1
+
+## 6562. Sign Identifiability of Causal Effects in Stationary Stochastic Dynamical Systems
+
+- **ID:** `2603.08311` · **primary:** `math.ST` · **Submitted:** 2026-03-09
+- **Authors:** Gijs van Seeventer, Saber Salehkaleybar
+- **abs:** https://arxiv.org/abs/2603.08311 · **PDF:** https://arxiv.org/pdf/2603.08311v1
+
+## 6563. Concept-Guided Fine-Tuning: Steering ViTs away from Spurious Correlations to Improve Robustness
+
+- **ID:** `2603.08309` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Yehonatan Elisha, Oren Barkan, Noam Koenigstein
+- **abs:** https://arxiv.org/abs/2603.08309 · **PDF:** https://arxiv.org/pdf/2603.08309v2
+
+## 6564. Retrieval-Augmented Anatomical Guidance for Text-to-CT Generation
+
+- **ID:** `2603.08305` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Daniele Molino, Camillo Maria Caruso, Paolo Soda, et al.
+- **abs:** https://arxiv.org/abs/2603.08305 · **PDF:** https://arxiv.org/pdf/2603.08305v1
+
+## 6565. Graph-Instructed Neural Networks for parametric problems with varying boundary conditions
+
+- **ID:** `2603.08304` · **primary:** `math.NA` · **Submitted:** 2026-03-09
+- **Authors:** Francesco Della Santa, Sandra Pieraccini, Maria Strazzullo
+- **abs:** https://arxiv.org/abs/2603.08304 · **PDF:** https://arxiv.org/pdf/2603.08304v1
+
+## 6566. Neural Dynamics Self-Attention for Spiking Transformers
+
+- **ID:** `2603.19290` · **primary:** `cs.NE` · **Submitted:** 2026-03-09
+- **Authors:** Dehao Zhang, Fukai Guo, Shuai Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.19290 · **PDF:** https://arxiv.org/pdf/2603.19290v1
+
+## 6567. How Emotion Shapes the Behavior of LLMs and Agents: A Mechanistic Study
+
+- **ID:** `2604.00005` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Moran Sun, Tianlin Li, Yuwei Zheng, et al.
+- **abs:** https://arxiv.org/abs/2604.00005 · **PDF:** https://arxiv.org/pdf/2604.00005v1
+
+## 6568. Deconstructing Multimodal Mathematical Reasoning: Towards a Unified Perception-Alignment-Reasoning Paradigm
+
+- **ID:** `2603.08291` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Tianyu Yang, Sihong Wu, Yilun Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.08291 · **PDF:** https://arxiv.org/pdf/2603.08291v2
+
+## 6569. Minor First, Major Last: A Depth-Induced Implicit Bias of Sharpness-Aware Minimization
+
+- **ID:** `2603.08290` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Chaewon Moon, Dongkuk Si, Chulhee Yun
+- **abs:** https://arxiv.org/abs/2603.08290 · **PDF:** https://arxiv.org/pdf/2603.08290v1
+
+## 6570. A Blockchain-based Traceability System for AI-Driven Engine Blade Inspection
+
+- **ID:** `2603.08288` · **primary:** `cs.CR` · **Submitted:** 2026-03-09
+- **Authors:** Mahmoud Hafez, Eman Ouda, Mohammed A. Mohammed Eltoum, et al.
+- **abs:** https://arxiv.org/abs/2603.08288 · **PDF:** https://arxiv.org/pdf/2603.08288v1
+
+## 6571. Posterior Sampling Reinforcement Learning with Gaussian Processes for Continuous Control: Sublinear Regret Bounds for Unbounded State Spaces
+
+- **ID:** `2603.08287` · **primary:** `stat.ML` · **Submitted:** 2026-03-09
+- **Authors:** Hamish Flynn, Joe Watson, Ingmar Posner, et al.
+- **abs:** https://arxiv.org/abs/2603.08287 · **PDF:** https://arxiv.org/pdf/2603.08287v1
+
+## 6572. LAMUS: A Large-Scale Corpus for Legal Argument Mining from U.S. Caselaw using LLMs
+
+- **ID:** `2603.08286` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Serene Wang, Lavanya Pobbathi, Haihua Chen
+- **abs:** https://arxiv.org/abs/2603.08286 · **PDF:** https://arxiv.org/pdf/2603.08286v1
+
+## 6573. WaveComm: Lightweight Communication for Collaborative Perception via Wavelet Feature Distillation
+
+- **ID:** `2603.13365` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Erdemt Bao, Jin Yang
+- **abs:** https://arxiv.org/abs/2603.13365 · **PDF:** https://arxiv.org/pdf/2603.13365v1
+
+## 6574. FineRMoE: Dimension Expansion for Finer-Grained Expert with Its Upcycling Approach
+
+- **ID:** `2603.13364` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Ning Liao, Xiaoxing Wang, Xiaohan Qin, et al.
+- **abs:** https://arxiv.org/abs/2603.13364 · **PDF:** https://arxiv.org/pdf/2603.13364v1
+
+## 6575. PolyFormer: learning efficient reformulations for scalable optimization under complex physical constraints
+
+- **ID:** `2603.08283` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Yilin Wen, Yi Guo, Bo Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.08283 · **PDF:** https://arxiv.org/pdf/2603.08283v1
+
+## 6576. Using Multimodal and Language-Agnostic Sentence Embeddings for Abstractive Summarization
+
+- **ID:** `2603.08282` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Chaimae Chellaf, Salima Mdhaffar, Yannick Estève, et al.
+- **abs:** https://arxiv.org/abs/2603.08282 · **PDF:** https://arxiv.org/pdf/2603.08282v1
+
+## 6577. Evaluating LLM-Based Grant Proposal Review via Structured Perturbations
+
+- **ID:** `2603.08281` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** William Thorne, Joseph James, Yang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.08281 · **PDF:** https://arxiv.org/pdf/2603.08281v2
+
+## 6578. TA-RNN-Medical-Hybrid: A Time-Aware and Interpretable Framework for Mortality Risk Prediction
+
+- **ID:** `2603.08278` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Zahra Jafari, Azadeh Zamanifar, Amirfarhad Farhadi
+- **abs:** https://arxiv.org/abs/2603.08278 · **PDF:** https://arxiv.org/pdf/2603.08278v1
+
+## 6579. AdaCultureSafe: Adaptive Cultural Safety Grounded by Cultural Knowledge in Large Language Models
+
+- **ID:** `2603.08275` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Hankun Kang, Di Lin, Zhirong Liao, et al.
+- **abs:** https://arxiv.org/abs/2603.08275 · **PDF:** https://arxiv.org/pdf/2603.08275v1
+
+## 6580. How Much Do LLMs Hallucinate in Document Q&A Scenarios? A 172-Billion-Token Study Across Temperatures, Context Lengths, and Hardware Platforms
+
+- **ID:** `2603.08274` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** JV Roig
+- **abs:** https://arxiv.org/abs/2603.08274 · **PDF:** https://arxiv.org/pdf/2603.08274v1
+
+## 6581. SCL-GNN: Towards Generalizable Graph Neural Networks via Spurious Correlation Learning
+
+- **ID:** `2603.08270` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Yuxiang Zhang, Enyan Dai
+- **abs:** https://arxiv.org/abs/2603.08270 · **PDF:** https://arxiv.org/pdf/2603.08270v1
+
+## 6582. SAIL: Test-Time Scaling for In-Context Imitation Learning with VLM
+
+- **ID:** `2603.08269` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Makoto Sato, Yusuke Iwasawa, Yujin Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.08269 · **PDF:** https://arxiv.org/pdf/2603.08269v1
+
+## 6583. Towards a more efficient bias detection in financial language models
+
+- **ID:** `2603.08267` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Firas Hadj Kacem, Ahmed Khanfir, Mike Papadakis
+- **abs:** https://arxiv.org/abs/2603.08267 · **PDF:** https://arxiv.org/pdf/2603.08267v1
+
+## 6584. Airborne Magnetic Anomaly Navigation with Neural-Network-Augmented Online Calibration
+
+- **ID:** `2603.08265` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Antonia Hager, Sven Nebendahl, Alexej Klushyn, et al.
+- **abs:** https://arxiv.org/abs/2603.08265 · **PDF:** https://arxiv.org/pdf/2603.08265v1
+
+## 6585. FinToolBench: Evaluating LLM Agents for Real-World Financial Tool Use
+
+- **ID:** `2603.08262` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Jiaxuan Lu, Kong Wang, Yemin Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.08262 · **PDF:** https://arxiv.org/pdf/2603.08262v1
+
+## 6586. Beyond ReinMax: Low-Variance Gradient Estimators for Discrete Latent Variables
+
+- **ID:** `2603.08257` · **primary:** `stat.ML` · **Submitted:** 2026-03-09
+- **Authors:** Daniel Wang, Thang D. Bui
+- **abs:** https://arxiv.org/abs/2603.08257 · **PDF:** https://arxiv.org/pdf/2603.08257v1
+
+## 6587. NCL-UoR at SemEval-2026 Task 5: Embedding-Based Methods, Fine-Tuning, and LLMs for Word Sense Plausibility Rating
+
+- **ID:** `2603.08256` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Tong Wu, Thanet Markchom, Huizhi Liang
+- **abs:** https://arxiv.org/abs/2603.08256 · **PDF:** https://arxiv.org/pdf/2603.08256v1
+
+## 6588. FlowTouch: View-Invariant Visuo-Tactile Prediction
+
+- **ID:** `2603.08255` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Seongjin Bien, Carlo Kneissl, Tobias Jülg, et al.
+- **abs:** https://arxiv.org/abs/2603.08255 · **PDF:** https://arxiv.org/pdf/2603.08255v1
+
+## 6589. FedPrism: Adaptive Personalized Federated Learning under Non-IID Data
+
+- **ID:** `2603.08252` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Prakash Kumbhakar, Shrey Srivastava, Haroon R Lone
+- **abs:** https://arxiv.org/abs/2603.08252 · **PDF:** https://arxiv.org/pdf/2603.08252v2
+
+## 6590. Not All Queries Need Deep Thought: CoFiCot for Adaptive Coarse-to-fine Stateful Refinement
+
+- **ID:** `2603.08251` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Dongxu Zhang, Hongqiang Lin, Yiding Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.08251 · **PDF:** https://arxiv.org/pdf/2603.08251v1
+
+## 6591. Bootstrapping Audiovisual Speech Recognition in Zero-AV-Resource Scenarios with Synthetic Visual Data
+
+- **ID:** `2603.08249` · **primary:** `eess.AS` · **Submitted:** 2026-03-09
+- **Authors:** Pol Buitrago, Pol Gàlvez, Oriol Pareras, et al.
+- **abs:** https://arxiv.org/abs/2603.08249 · **PDF:** https://arxiv.org/pdf/2603.08249v1
+
+## 6592. Optimising antibiotic switching via forecasting of patient physiology
+
+- **ID:** `2603.08242` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Magnus Ross, Nel Swanepoel, Akish Luintel, et al.
+- **abs:** https://arxiv.org/abs/2603.08242 · **PDF:** https://arxiv.org/pdf/2603.08242v1
+
+## 6593. Sensivity of LLMs' Explanations to the Training Randomness:Context, Class & Task Dependencies
+
+- **ID:** `2603.08241` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Romain Loncour, Jérémie Bogaert, François-Xavier Standaert
+- **abs:** https://arxiv.org/abs/2603.08241 · **PDF:** https://arxiv.org/pdf/2603.08241v1
+
+## 6594. Fibration Policy Optimization
+
+- **ID:** `2603.08239` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Chang Li, Tshihao Tsu, Yaren Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.08239 · **PDF:** https://arxiv.org/pdf/2603.08239v1
+
+## 6595. AlpsBench: An LLM Personalization Benchmark for Real-Dialogue Memorization and Preference Alignment
+
+- **ID:** `2603.26680` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Jianfei Xiao, Xiang Yu, Chengbing Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.26680 · **PDF:** https://arxiv.org/pdf/2603.26680v1
+
+## 6596. Exploring Deep Learning and Ultra-Widefield Imaging for Diabetic Retinopathy and Macular Edema
+
+- **ID:** `2603.08235` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Pablo Jimenez-Lizcano, Sergio Romero-Tapiador, Ruben Tolosana, et al.
+- **abs:** https://arxiv.org/abs/2603.08235 · **PDF:** https://arxiv.org/pdf/2603.08235v1
+
+## 6597. The Struggle Between Continuation and Refusal: A Mechanistic Analysis of the Continuation-Triggered Jailbreak in LLMs
+
+- **ID:** `2603.08234` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Yonghong Deng, Zhen Yang, Ping Jian, et al.
+- **abs:** https://arxiv.org/abs/2603.08234 · **PDF:** https://arxiv.org/pdf/2603.08234v1
+
+## 6598. Quantifying Cross-Lingual Transfer in Paralinguistic Speech Tasks
+
+- **ID:** `2603.08231` · **primary:** `eess.AS` · **Submitted:** 2026-03-09
+- **Authors:** Pol Buitrago, Oriol Pareras, Federico Costa, et al.
+- **abs:** https://arxiv.org/abs/2603.08231 · **PDF:** https://arxiv.org/pdf/2603.08231v1
+
+## 6599. Disentangling Reasoning in Large Audio-Language Models for Ambiguous Emotion Prediction
+
+- **ID:** `2603.08230` · **primary:** `cs.SD` · **Submitted:** 2026-03-09
+- **Authors:** Xiaofeng Yu, Jiaheng Dong, Jean Honorio, et al.
+- **abs:** https://arxiv.org/abs/2603.08230 · **PDF:** https://arxiv.org/pdf/2603.08230v1
+
+## 6600. AI Meets Mathematics Education: A Case Study on Supporting an Instructor in a Large Mathematics Class with Context-Aware AI
+
+- **ID:** `2603.26679` · **primary:** `cs.CY` · **Submitted:** 2026-03-09
+- **Authors:** Jérémy Barghorn, Anna Sotnikova, Sacha Friedli, et al.
+- **abs:** https://arxiv.org/abs/2603.26679 · **PDF:** https://arxiv.org/pdf/2603.26679v1
+
+## 6601. SplitAgent: A Privacy-Preserving Distributed Architecture for Enterprise-Cloud Agent Collaboration
+
+- **ID:** `2603.08221` · **primary:** `cs.CR` · **Submitted:** 2026-03-09
+- **Authors:** Jianshu She
+- **abs:** https://arxiv.org/abs/2603.08221 · **PDF:** https://arxiv.org/pdf/2603.08221v1
+
+## 6602. LinearARD: Linear-Memory Attention Distillation for RoPE Restoration
+
+- **ID:** `2604.00004` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Ning Yang, Hengyu Zhong, Wentao Wang, et al.
+- **abs:** https://arxiv.org/abs/2604.00004 · **PDF:** https://arxiv.org/pdf/2604.00004v1
+
+## 6603. Wiener Chaos Expansion based Neural Operator for Singular Stochastic Partial Differential Equations
+
+- **ID:** `2603.08219` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Dai Shi, Luke Thompson, Andi Han, et al.
+- **abs:** https://arxiv.org/abs/2603.08219 · **PDF:** https://arxiv.org/pdf/2603.08219v1
+
+## 6604. IAML: Illumination-Aware Mirror Loss for Progressive Learning in Low-Light Image Enhancement Auto-encoders
+
+- **ID:** `2603.13363` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Farida Mohsen, Tala Zaim, Ali Al-Zawqari, et al.
+- **abs:** https://arxiv.org/abs/2603.13363 · **PDF:** https://arxiv.org/pdf/2603.13363v1
+
+## 6605. DualTurn: Learning Turn-Taking from Dual-Channel Generative Speech Pretraining
+
+- **ID:** `2603.08216` · **primary:** `eess.AS` · **Submitted:** 2026-03-09
+- **Authors:** Shangeth Rajaa
+- **abs:** https://arxiv.org/abs/2603.08216 · **PDF:** https://arxiv.org/pdf/2603.08216v1
+
+## 6606. Revisiting Gradient Staleness: Evaluating Distance Metrics for Asynchronous Federated Learning Aggregation
+
+- **ID:** `2603.08211` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Patrick Wilhelm, Odej Kao
+- **abs:** https://arxiv.org/abs/2603.08211 · **PDF:** https://arxiv.org/pdf/2603.08211v1
+
+## 6607. Multi-Objective Evolutionary Optimization of Chance-Constrained Multiple-Choice Knapsack Problems with Implicit Probability Distributions
+
+- **ID:** `2603.08209` · **primary:** `cs.NE` · **Submitted:** 2026-03-09
+- **Authors:** Xuanfeng Li, Shengcai Liu, Wenjie Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.08209 · **PDF:** https://arxiv.org/pdf/2603.08209v1
+
+## 6608. Alignment-Aware and Reliability-Gated Multimodal Fusion for Unmanned Aerial Vehicle Detection Across Heterogeneous Thermal-Visual Sensors
+
+- **ID:** `2603.08208` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Ishrat Jahan, Molla E Majid, M Murugappan, et al.
+- **abs:** https://arxiv.org/abs/2603.08208 · **PDF:** https://arxiv.org/pdf/2603.08208v1
+
+## 6609. The Conundrum of Trustworthy Research on Attacking Personally Identifiable Information Removal Techniques
+
+- **ID:** `2603.08207` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Sebastian Ochs, Ivan Habernal
+- **abs:** https://arxiv.org/abs/2603.08207 · **PDF:** https://arxiv.org/pdf/2603.08207v1
+
+## 6610. Distributional Regression with Tabular Foundation Models: Evaluating Probabilistic Predictions via Proper Scoring Rules
+
+- **ID:** `2603.08206` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Jonas Landsgesell, Pascal Knoll
+- **abs:** https://arxiv.org/abs/2603.08206 · **PDF:** https://arxiv.org/pdf/2603.08206v4
+
+## 6611. MM-TS: Multi-Modal Temperature and Margin Schedules for Contrastive Learning with Long-Tail Data
+
+- **ID:** `2603.08202` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Siarhei Sheludzko, Dhimitrios Duka, Bernt Schiele, et al.
+- **abs:** https://arxiv.org/abs/2603.08202 · **PDF:** https://arxiv.org/pdf/2603.08202v1
+
+## 6612. Patient-Level Multimodal Question Answering from Multi-Site Auscultation Recordings
+
+- **ID:** `2603.13362` · **primary:** `cs.SD` · **Submitted:** 2026-03-09
+- **Authors:** Fan Wu, Tsai-Ning Wang, Nicolas Zumarraga, et al.
+- **abs:** https://arxiv.org/abs/2603.13362 · **PDF:** https://arxiv.org/pdf/2603.13362v1
+
+## 6613. Supporting Workflow Reproducibility by Linking Bioinformatics Tools across Papers and Executable Code
+
+- **ID:** `2603.08195` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Clémence Sebe, Olivier Ferret, Aurélie Névéol, et al.
+- **abs:** https://arxiv.org/abs/2603.08195 · **PDF:** https://arxiv.org/pdf/2603.08195v1
+
+## 6614. Sequential Service Region Design with Capacity-Constrained Investment and Spillover Effect
+
+- **ID:** `2603.08188` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Tingting Chen, Feng Chu, Jiantong Zhang
+- **abs:** https://arxiv.org/abs/2603.08188 · **PDF:** https://arxiv.org/pdf/2603.08188v1
+
+## 6615. SERQ: Saliency-Aware Low-Rank Error Reconstruction for LLM Quantization
+
+- **ID:** `2603.08185` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Yeonsik Park, Hyeonseong Kim, Seungkyu Choi
+- **abs:** https://arxiv.org/abs/2603.08185 · **PDF:** https://arxiv.org/pdf/2603.08185v1
+
+## 6616. TildeOpen LLM: Leveraging Curriculum Learning to Achieve Equitable Language Representation
+
+- **ID:** `2603.08182` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Toms Bergmanis, Martins Kronis, Ingus Jānis Pretkalniņš, et al.
+- **abs:** https://arxiv.org/abs/2603.08182 · **PDF:** https://arxiv.org/pdf/2603.08182v1
+
+## 6617. AutoAdapt: An Automated Domain Adaptation Framework for LLMs
+
+- **ID:** `2603.08181` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Sidharth Sinha, Anson Bastos, Xuchao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.08181 · **PDF:** https://arxiv.org/pdf/2603.08181v1
+
+## 6618. ALOOD: Exploiting Language Representations for LiDAR-based Out-of-Distribution Object Detection
+
+- **ID:** `2603.08180` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Michael Kösel, Marcel Schreiber, Michael Ulrich, et al.
+- **abs:** https://arxiv.org/abs/2603.08180 · **PDF:** https://arxiv.org/pdf/2603.08180v1
+
+## 6619. Privacy-Preserving End-to-End Full-Duplex Speech Dialogue Models
+
+- **ID:** `2603.08179` · **primary:** `eess.AS` · **Submitted:** 2026-03-09
+- **Authors:** Nikita Kuzmin, Tao Zhong, Jiajun Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.08179 · **PDF:** https://arxiv.org/pdf/2603.08179v1
+
+## 6620. Is continuous CoT better suited for multi-lingual reasoning?
+
+- **ID:** `2603.08177` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Ali Hamza Bashir, Behzad Shomali, Markus Frey, et al.
+- **abs:** https://arxiv.org/abs/2603.08177 · **PDF:** https://arxiv.org/pdf/2603.08177v1
+
+## 6621. Evolution Strategy-Based Calibration for Low-Bit Quantization of Speech Models
+
+- **ID:** `2603.08173` · **primary:** `cs.SD` · **Submitted:** 2026-03-09
+- **Authors:** Lucas Rakotoarivony
+- **abs:** https://arxiv.org/abs/2603.08173 · **PDF:** https://arxiv.org/pdf/2603.08173v1
+
+## 6622. Evidence-Driven Reasoning for Industrial Maintenance Using Heterogeneous Data
+
+- **ID:** `2603.08171` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Fearghal O'Donncha, Nianjun Zhou, Natalia Martinez, et al.
+- **abs:** https://arxiv.org/abs/2603.08171 · **PDF:** https://arxiv.org/pdf/2603.08171v1
+
+## 6623. RexDrug: Reliable Multi-Drug Combination Extraction through Reasoning-Enhanced LLMs
+
+- **ID:** `2603.08166` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Zhijun Wang, Ling Luo, Dinghao Pan, et al.
+- **abs:** https://arxiv.org/abs/2603.08166 · **PDF:** https://arxiv.org/pdf/2603.08166v1
+
+## 6624. An explainable hybrid deep learning-enabled intelligent fault detection and diagnosis approach for automotive software systems validation
+
+- **ID:** `2603.08165` · **primary:** `cs.SE` · **Submitted:** 2026-03-09
+- **Authors:** Mohammad Abboush, Ehab Ghannoum, Andreas Rausch
+- **abs:** https://arxiv.org/abs/2603.08165 · **PDF:** https://arxiv.org/pdf/2603.08165v1
+
+## 6625. Covenant-72B: Pre-Training a 72B LLM with Trustless Peers Over-the-Internet
+
+- **ID:** `2603.08163` · **primary:** `cs.DC` · **Submitted:** 2026-03-09
+- **Authors:** Joel Lidin, Amir Sarfi, Erfan Miahi, et al.
+- **abs:** https://arxiv.org/abs/2603.08163 · **PDF:** https://arxiv.org/pdf/2603.08163v2
+
+## 6626. Learning Hierarchical Knowledge in Text-Rich Networks with Taxonomy-Informed Representation Learning
+
+- **ID:** `2603.08159` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Yunhui Liu, Yongchao Liu, Yinfeng Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.08159 · **PDF:** https://arxiv.org/pdf/2603.08159v1
+
+## 6627. Outlier-robust Autocovariance Least Square Estimation via Iteratively Reweighted Least Square
+
+- **ID:** `2603.08158` · **primary:** `math.OC` · **Submitted:** 2026-03-09
+- **Authors:** Jiahong Li, Fang Deng
+- **abs:** https://arxiv.org/abs/2603.08158 · **PDF:** https://arxiv.org/pdf/2603.08158v1
+
+## 6628. Are We Winning the Wrong Game? Revisiting Evaluation Practices for Long-Term Time Series Forecasting
+
+- **ID:** `2603.08156` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Thanapol Phungtua-eng, Yoshitaka Yamamoto
+- **abs:** https://arxiv.org/abs/2603.08156 · **PDF:** https://arxiv.org/pdf/2603.08156v1
+
+## 6629. C$^2$FG: Control Classifier-Free Guidance via Score Discrepancy Analysis
+
+- **ID:** `2603.08155` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Jiayang Gao, Tianyi Zheng, Jiayang Zou, et al.
+- **abs:** https://arxiv.org/abs/2603.08155 · **PDF:** https://arxiv.org/pdf/2603.08155v1
+
+## 6630. Gender Bias in MT for a Genderless Language: New Benchmarks for Basque
+
+- **ID:** `2603.08153` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Amaia Murillo, Olatz-Perez-de-Viñaspre, Naiara Perez
+- **abs:** https://arxiv.org/abs/2603.08153 · **PDF:** https://arxiv.org/pdf/2603.08153v1
+
+## 6631. Gradually Excavating External Knowledge for Implicit Complex Question Answering
+
+- **ID:** `2603.08148` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Chang Liu, Xiaoguang Li, Lifeng Shang, et al.
+- **abs:** https://arxiv.org/abs/2603.08148 · **PDF:** https://arxiv.org/pdf/2603.08148v1
+
+## 6632. Training event-based neural networks with exact gradients via Differentiable ODE Solving in JAX
+
+- **ID:** `2603.08146` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Lukas König, Manuel Kuhn, David Kappel, et al.
+- **abs:** https://arxiv.org/abs/2603.08146 · **PDF:** https://arxiv.org/pdf/2603.08146v1
+
+## 6633. DARC: Disagreement-Aware Alignment via Risk-Constrained Decoding
+
+- **ID:** `2603.08145` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Mingxi Zou, Jiaxiang Chen, Junfan Li, et al.
+- **abs:** https://arxiv.org/abs/2603.08145 · **PDF:** https://arxiv.org/pdf/2603.08145v1
+
+## 6634. Mitigating Homophily Disparity in Graph Anomaly Detection: A Scalable and Adaptive Approach
+
+- **ID:** `2603.08137` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Yunhui Liu, Qizhuo Xie, Yinfeng Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.08137 · **PDF:** https://arxiv.org/pdf/2603.08137v1
+
+## 6635. Explainable Condition Monitoring via Probabilistic Anomaly Detection Applied to Helicopter Transmissions
+
+- **ID:** `2603.08130` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Aurelio Raffa Ugolini, Jessica Leoni, Valentina Breschi, et al.
+- **abs:** https://arxiv.org/abs/2603.08130 · **PDF:** https://arxiv.org/pdf/2603.08130v1
+
+## 6636. TRIAGE: Type-Routed Interventions via Aleatoric-Epistemic Gated Estimation in Robotic Manipulation and Adaptive Perception -- Don't Treat All Uncertainty the Same
+
+- **ID:** `2603.08128` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Divake Kumar, Sina Tayebati, Devashri Naik, et al.
+- **abs:** https://arxiv.org/abs/2603.08128 · **PDF:** https://arxiv.org/pdf/2603.08128v1
+
+## 6637. EvoScientist: Towards Multi-Agent Evolving AI Scientists for End-to-End Scientific Discovery
+
+- **ID:** `2603.08127` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Yougang Lyu, Xi Zhang, Xinhao Yi, et al.
+- **abs:** https://arxiv.org/abs/2603.08127 · **PDF:** https://arxiv.org/pdf/2603.08127v1
+
+## 6638. Foley-Flow: Coordinated Video-to-Audio Generation with Masked Audio-Visual Alignment and Dynamic Conditional Flows
+
+- **ID:** `2603.08126` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Shentong Mo, Yibing Song
+- **abs:** https://arxiv.org/abs/2603.08126 · **PDF:** https://arxiv.org/pdf/2603.08126v1
+
+## 6639. Ramsa: A Large Sociolinguistically Rich Emirati Arabic Speech Corpus for ASR and TTS
+
+- **ID:** `2603.08125` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Rania Al-Sabbagh
+- **abs:** https://arxiv.org/abs/2603.08125 · **PDF:** https://arxiv.org/pdf/2603.08125v1
+
+## 6640. SaiVLA-0: Cerebrum--Pons--Cerebellum Tripartite Architecture for Compute-Aware Vision-Language-Action
+
+- **ID:** `2603.08124` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Xiang Shi, Wenlong Huang, Menglin Zou, et al.
+- **abs:** https://arxiv.org/abs/2603.08124 · **PDF:** https://arxiv.org/pdf/2603.08124v1
+
+## 6641. Model-based Offline RL via Robust Value-Aware Model Learning with Implicitly Differentiable Adaptive Weighting
+
+- **ID:** `2603.08118` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Zhongjian Qiao, Jiafei Lyu, Boxiang Lyu, et al.
+- **abs:** https://arxiv.org/abs/2603.08118 · **PDF:** https://arxiv.org/pdf/2603.08118v1
+
+## 6642. UIS-Digger: Towards Comprehensive Research Agent Systems for Real-world Unindexed Information Seeking
+
+- **ID:** `2603.08117` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Chang Liu, Chuqiao Kuang, Tianyi Zhuang, et al.
+- **abs:** https://arxiv.org/abs/2603.08117 · **PDF:** https://arxiv.org/pdf/2603.08117v3
+
+## 6643. Tau-BNO: Brain Neural Operator for Tau Transport Model
+
+- **ID:** `2603.08108` · **primary:** `cs.CE` · **Submitted:** 2026-03-09
+- **Authors:** Nuutti Barron, Heng Rao, Urmi Saha, et al.
+- **abs:** https://arxiv.org/abs/2603.08108 · **PDF:** https://arxiv.org/pdf/2603.08108v2
+
+## 6644. Invisible Safety Threat: Malicious Finetuning for LLM via Steganography
+
+- **ID:** `2603.08104` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Guangnian Wan, Xinyin Ma, Gongfan Fang, et al.
+- **abs:** https://arxiv.org/abs/2603.08104 · **PDF:** https://arxiv.org/pdf/2603.08104v2
+
+## 6645. DC-W2S: Dual-Consensus Weak-to-Strong Training for Reliable Process Reward Modeling in Biological Reasoning
+
+- **ID:** `2603.08095` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Chi-Min Chan, Ehsan Hajiramezanali, Xiner Li, et al.
+- **abs:** https://arxiv.org/abs/2603.08095 · **PDF:** https://arxiv.org/pdf/2603.08095v1
+
+## 6646. Toward Robust LLM-Based Judges: Taxonomic Bias Evaluation and Debiasing Optimization
+
+- **ID:** `2603.08091` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Hongli Zhou, Hui Huang, Rui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.08091 · **PDF:** https://arxiv.org/pdf/2603.08091v1
+
+## 6647. DSH-Bench: A Difficulty- and Scenario-Aware Benchmark with Hierarchical Subject Taxonomy for Subject-Driven Text-to-Image Generation
+
+- **ID:** `2603.08090` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Zhenyu Hu, Qing Wang, Te Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.08090 · **PDF:** https://arxiv.org/pdf/2603.08090v1
+
+## 6648. EAGLE-Pangu: Accelerator-Safe Tree Speculative Decoding on Ascend NPUs
+
+- **ID:** `2603.08088` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Chang Han, Yijie Hu, Jingling Liu
+- **abs:** https://arxiv.org/abs/2603.08088 · **PDF:** https://arxiv.org/pdf/2603.08088v1
+
+## 6649. High-Fidelity Pruning for Large Language Models
+
+- **ID:** `2603.08083` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Yijun Zhu, Jianxin Wang, Chengchao Shen
+- **abs:** https://arxiv.org/abs/2603.08083 · **PDF:** https://arxiv.org/pdf/2603.08083v1
+
+## 6650. Tiny Autoregressive Recursive Models
+
+- **ID:** `2603.08082` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Paulius Rauba, Claudio Fanconi, Mihaela van der Schaar
+- **abs:** https://arxiv.org/abs/2603.08082 · **PDF:** https://arxiv.org/pdf/2603.08082v1
+
+## 6651. Hybrid Quantum Neural Network for Multivariate Clinical Time Series Forecasting
+
+- **ID:** `2603.08072` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Irene Iele, Floriano Caprio, Paolo Soda, et al.
+- **abs:** https://arxiv.org/abs/2603.08072 · **PDF:** https://arxiv.org/pdf/2603.08072v1
+
+## 6652. In-Context Reinforcement Learning for Tool Use in Large Language Models
+
+- **ID:** `2603.08068` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Yaoqi Ye, Yiran Zhao, Keyu Duan, et al.
+- **abs:** https://arxiv.org/abs/2603.08068 · **PDF:** https://arxiv.org/pdf/2603.08068v1
+
+## 6653. BrainCast: A Spatio-Temporal Forecasting Model for Whole-Brain fMRI Time Series Prediction
+
+- **ID:** `2603.13361` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Yunlong Gao, Jinbo Yang, Li Xiao, et al.
+- **abs:** https://arxiv.org/abs/2603.13361 · **PDF:** https://arxiv.org/pdf/2603.13361v1
+
+## 6654. Deterministic Differentiable Structured Pruning for Large Language Models
+
+- **ID:** `2603.08065` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Weiyu Huang, Pengle Zhang, Xiaolu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.08065 · **PDF:** https://arxiv.org/pdf/2603.08065v1
+
+## 6655. Adversarial Domain Adaptation Enables Knowledge Transfer Across Heterogeneous RNA-Seq Datasets
+
+- **ID:** `2603.08062` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Kevin Dradjat, Massinissa Hamidi, Blaise Hanczar
+- **abs:** https://arxiv.org/abs/2603.08062 · **PDF:** https://arxiv.org/pdf/2603.08062v1
+
+## 6656. ImageEdit-R1: Boosting Multi-Agent Image Editing via Reinforcement Learning
+
+- **ID:** `2603.08059` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Yiran Zhao, Yaoqi Ye, Xiang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.08059 · **PDF:** https://arxiv.org/pdf/2603.08059v1
+
+## 6657. Towards Differentiating Between Failures and Domain Shifts in Industrial Data Streams
+
+- **ID:** `2603.18032` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Natalia Wojak-Strzelecka, Szymon Bobek, Grzegorz J. Nalepa, et al.
+- **abs:** https://arxiv.org/abs/2603.18032 · **PDF:** https://arxiv.org/pdf/2603.18032v1
+
+## 6658. Stabilized Fine-Tuning with LoRA in Federated Learning: Mitigating the Side Effect of Client Size and Rank via the Scaling Factor
+
+- **ID:** `2603.08058` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Jiayu Huang, Xiaohu Wu, Tiantian He, et al.
+- **abs:** https://arxiv.org/abs/2603.08058 · **PDF:** https://arxiv.org/pdf/2603.08058v1
+
+## 6659. Speed3R: Sparse Feed-forward 3D Reconstruction Models
+
+- **ID:** `2603.08055` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Weining Ren, Xiao Tan, Kai Han
+- **abs:** https://arxiv.org/abs/2603.08055 · **PDF:** https://arxiv.org/pdf/2603.08055v1
+
+## 6660. Examining the Role of YouTube Production and Consumption Dynamics on the Formation of Extreme Ideologies
+
+- **ID:** `2603.08049` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Sarmad Chandio, Rishab Nithyanand
+- **abs:** https://arxiv.org/abs/2603.08049 · **PDF:** https://arxiv.org/pdf/2603.08049v1
+
+## 6661. S2S-FDD: Bridging Industrial Time Series and Natural Language for Explainable Zero-shot Fault Diagnosis
+
+- **ID:** `2603.08048` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Baoxue Li, Chunhui Zhao
+- **abs:** https://arxiv.org/abs/2603.08048 · **PDF:** https://arxiv.org/pdf/2603.08048v1
+
+## 6662. Graph2Video: Leveraging Video Models to Model Dynamic Graph Evolution
+
+- **ID:** `2603.13360` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Hua Liu, Yanbin Wei, Fei Xing, et al.
+- **abs:** https://arxiv.org/abs/2603.13360 · **PDF:** https://arxiv.org/pdf/2603.13360v1
+
+## 6663. CDRRM: Contrast-Driven Rubric Generation for Reliable and Interpretable Reward Modeling
+
+- **ID:** `2603.08035` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Dengcan Liu, Fengkai Yang, Xiaohan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.08035 · **PDF:** https://arxiv.org/pdf/2603.08035v1
+
+## 6664. Solution to the 10th ABAW Expression Recognition Challenge: A Robust Multimodal Framework with Safe Cross-Attention and Modality Dropout
+
+- **ID:** `2603.08034` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Jun Yu, Naixiang Zheng, Guoyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.08034 · **PDF:** https://arxiv.org/pdf/2603.08034v1
+
+## 6665. GCGNet: Graph-Consistent Generative Network for Time Series Forecasting with Exogenous Variables
+
+- **ID:** `2603.08032` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Zhengyu Li, Xiangfei Qiu, Yuhan Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.08032 · **PDF:** https://arxiv.org/pdf/2603.08032v1
+
+## 6666. DyLLM: Efficient Diffusion LLM Inference via Saliency-based Token Selection and Partial Attention
+
+- **ID:** `2603.08026` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Younjoo Lee, Junghoo Lee, Seungkyun Dan, et al.
+- **abs:** https://arxiv.org/abs/2603.08026 · **PDF:** https://arxiv.org/pdf/2603.08026v1
+
+## 6667. ConflictBench: Evaluating Human-AI Conflict via Interactive and Visually Grounded Environments
+
+- **ID:** `2603.08024` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Weixiang Zhao, Haozhen Li, Yanyan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.08024 · **PDF:** https://arxiv.org/pdf/2603.08024v1
+
+## 6668. Speculating Experts Accelerates Inference for Mixture-of-Experts
+
+- **ID:** `2603.19289` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Vivan Madan, Prajwal Singhania, Abhinav Bhatele, et al.
+- **abs:** https://arxiv.org/abs/2603.19289 · **PDF:** https://arxiv.org/pdf/2603.19289v1
+
+## 6669. Not Like Transformers: Drop the Beat Representation for Dance Generation with Mamba-Based Diffusion Model
+
+- **ID:** `2603.08023` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Sangjune Park, Inhyeok Choi, Donghyeon Soon, et al.
+- **abs:** https://arxiv.org/abs/2603.08023 · **PDF:** https://arxiv.org/pdf/2603.08023v1
+
+## 6670. Capacity-Aware Mixture Law Enables Efficient LLM Data Optimization
+
+- **ID:** `2603.08022` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Jingwei Li, Xinran Gu, Jingzhao Zhang
+- **abs:** https://arxiv.org/abs/2603.08022 · **PDF:** https://arxiv.org/pdf/2603.08022v1
+
+## 6671. Hybrid Quantum-Classical Encoding for Accurate Residue-Level pKa Prediction
+
+- **ID:** `2603.11061` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-09
+- **Authors:** Van Le, Tan Le
+- **abs:** https://arxiv.org/abs/2603.11061 · **PDF:** https://arxiv.org/pdf/2603.11061v1
+
+## 6672. Alignment-Process-Outcome: Rethinking How AIs and Humans Collaborate
+
+- **ID:** `2603.08017` · **primary:** `cs.HC` · **Submitted:** 2026-03-09
+- **Authors:** Haichang Li, Anjun Zhu, Arpit Narechania
+- **abs:** https://arxiv.org/abs/2603.08017 · **PDF:** https://arxiv.org/pdf/2603.08017v2
+
+## 6673. FedMomentum: Preserving LoRA Training Momentum in Federated Fine-Tuning
+
+- **ID:** `2603.08014` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Peishen Yan, Yang Hua, Hao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.08014 · **PDF:** https://arxiv.org/pdf/2603.08014v1
+
+## 6674. PIRA-Bench: A Transition from Reactive GUI Agents to GUI-based Proactive Intent Recommendation Agents
+
+- **ID:** `2603.08013` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Yuxiang Chai, Shunye Tang, Han Xiao, et al.
+- **abs:** https://arxiv.org/abs/2603.08013 · **PDF:** https://arxiv.org/pdf/2603.08013v1
+
+## 6675. From Refusal Tokens to Refusal Control: Discovering and Steering Category-Specific Refusal Directions
+
+- **ID:** `2603.13359` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Rishab Alagharu, Ishneet Sukhvinder Singh, Shaibi Shamsudeen, et al.
+- **abs:** https://arxiv.org/abs/2603.13359 · **PDF:** https://arxiv.org/pdf/2603.13359v1
+
+## 6676. ViSA-Enhanced Aerial VLN: A Visual-Spatial Reasoning Enhanced Framework for Aerial Vision-Language Navigation
+
+- **ID:** `2603.08007` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Haoyu Tong, Xiangyu Dong, Xiaoguang Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.08007 · **PDF:** https://arxiv.org/pdf/2603.08007v1
+
+## 6677. Amortizing Maximum Inner Product Search with Learned Support Functions
+
+- **ID:** `2603.08001` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Theo X. Olausson, João Monteiro, Michal Klein, et al.
+- **abs:** https://arxiv.org/abs/2603.08001 · **PDF:** https://arxiv.org/pdf/2603.08001v1
+
+## 6678. SmartThinker: Progressive Chain-of-Thought Length Calibration for Efficient Large Language Model Reasoning
+
+- **ID:** `2603.08000` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Chenzhi Hu, Qinzhe Hu, Yuhang Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.08000 · **PDF:** https://arxiv.org/pdf/2603.08000v1
+
+## 6679. Aero-Promptness: Drag-Aware Aerodynamic Manipulability for Propeller-driven Vehicles
+
+- **ID:** `2603.07998` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Antonio Franchi
+- **abs:** https://arxiv.org/abs/2603.07998 · **PDF:** https://arxiv.org/pdf/2603.07998v1
+
+## 6680. CMMR-VLN: Vision-and-Language Navigation via Continual Multimodal Memory Retrieval
+
+- **ID:** `2603.07997` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Haozhou Li, Xiangyu Dong, Huiyan Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.07997 · **PDF:** https://arxiv.org/pdf/2603.07997v1
+
+## 6681. MJ1: Multimodal Judgment via Grounded Verification
+
+- **ID:** `2603.07990` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Bhavesh Kumar, Dylan Feng, Leonard Tang
+- **abs:** https://arxiv.org/abs/2603.07990 · **PDF:** https://arxiv.org/pdf/2603.07990v2
+
+## 6682. DynaTrust: Defending Multi-Agent Systems Against Sleeper Agents via Dynamic Trust Graphs
+
+- **ID:** `2603.15661` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Yu Li, Qiang Hu, Yao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.15661 · **PDF:** https://arxiv.org/pdf/2603.15661v1
+
+## 6683. \$OneMillion-Bench: How Far are Language Agents from Human Experts?
+
+- **ID:** `2603.07980` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Qianyu Yang, Yang Liu, Jiaqi Li, et al.
+- **abs:** https://arxiv.org/abs/2603.07980 · **PDF:** https://arxiv.org/pdf/2603.07980v1
+
+## 6684. Emergence is Overrated: AGI as an Archipelago of Experts
+
+- **ID:** `2603.07979` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Daniel Kilov
+- **abs:** https://arxiv.org/abs/2603.07979 · **PDF:** https://arxiv.org/pdf/2603.07979v1
+
+## 6685. OSExpert: Computer-Use Agents Learning Professional Skills via Exploration
+
+- **ID:** `2603.07978` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Jiateng Liu, Zhenhailong Wang, Rushi Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.07978 · **PDF:** https://arxiv.org/pdf/2603.07978v1
+
+## 6686. Scaling Machine Learning Interatomic Potentials with Mixtures of Experts
+
+- **ID:** `2603.07977` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-09
+- **Authors:** Yuzhi Liu, Duo Zhang, Anyang Peng, et al.
+- **abs:** https://arxiv.org/abs/2603.07977 · **PDF:** https://arxiv.org/pdf/2603.07977v2
+
+## 6687. VORL-EXPLORE: A Hybrid Learning Planning Approach to Multi-Robot Exploration in Dynamic Environments
+
+- **ID:** `2603.07973` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Ning Liu, Sen Shen, Zheng Li, et al.
+- **abs:** https://arxiv.org/abs/2603.07973 · **PDF:** https://arxiv.org/pdf/2603.07973v1
+
+## 6688. Adaptive Collaboration with Humans: Metacognitive Policy Optimization for Multi-Agent LLMs with Continual Learning
+
+- **ID:** `2603.07972` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Wei Yang, Defu Cao, Jiacheng Pang, et al.
+- **abs:** https://arxiv.org/abs/2603.07972 · **PDF:** https://arxiv.org/pdf/2603.07972v1
+
+## 6689. Advancing Automated Algorithm Design via Evolutionary Stagewise Design with LLMs
+
+- **ID:** `2603.07970` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Chen Lu, Ke Xue, Chengrui Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.07970 · **PDF:** https://arxiv.org/pdf/2603.07970v1
+
+## 6690. Local Constrained Bayesian Optimization
+
+- **ID:** `2603.07965` · **primary:** `stat.ML` · **Submitted:** 2026-03-09
+- **Authors:** Jing Jingzhe, Fan Zheyi, Szu Hui Ng, et al.
+- **abs:** https://arxiv.org/abs/2603.07965 · **PDF:** https://arxiv.org/pdf/2603.07965v1
+
+## 6691. PSTNet: Physically-Structured Turbulence Network
+
+- **ID:** `2603.07957` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Boris Kriuk, Fedor Kriuk
+- **abs:** https://arxiv.org/abs/2603.07957 · **PDF:** https://arxiv.org/pdf/2603.07957v1
+
+## 6692. RL unknotter, hard unknots and unknotting number
+
+- **ID:** `2603.07955` · **primary:** `math.GT` · **Submitted:** 2026-03-09
+- **Authors:** Anne Dranowski, Yura Kabkov, Daniel Tubbenhauer
+- **abs:** https://arxiv.org/abs/2603.07955 · **PDF:** https://arxiv.org/pdf/2603.07955v1
+
+## 6693. ELLMob: Event-Driven Human Mobility Generation with Self-Aligned LLM Framework
+
+- **ID:** `2603.07946` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Yusong Wang, Chuang Yang, Jiawei Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.07946 · **PDF:** https://arxiv.org/pdf/2603.07946v1
+
+## 6694. AI Agents, Language, Deep Learning and the Next Revolution in Science
+
+- **ID:** `2603.07940` · **primary:** `hep-ex` · **Submitted:** 2026-03-09
+- **Authors:** Ke Li, Beijiang Liu, Bruce Mellado, et al.
+- **abs:** https://arxiv.org/abs/2603.07940 · **PDF:** https://arxiv.org/pdf/2603.07940v1
+
+## 6695. Bi-CamoDiffusion: A Boundary-informed Diffusion Approach for Camouflaged Object Detection
+
+- **ID:** `2603.13357` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Patricia L. Suarez, Leo Thomas Ramos, Angel D. Sappa
+- **abs:** https://arxiv.org/abs/2603.13357 · **PDF:** https://arxiv.org/pdf/2603.13357v1
+
+## 6696. BRIDGE: Benchmark for multi-hop Reasoning In long multimodal Documents with Grounded Evidence
+
+- **ID:** `2603.07931` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Biao Xiang, Soyeon Caren Han, Yihao Ding
+- **abs:** https://arxiv.org/abs/2603.07931 · **PDF:** https://arxiv.org/pdf/2603.07931v1
+
+## 6697. SWE-Fuse: Empowering Software Agents via Issue-free Trajectory Learning and Entropy-aware RLVR Training
+
+- **ID:** `2603.07927` · **primary:** `cs.SE` · **Submitted:** 2026-03-09
+- **Authors:** Xin-Cheng Wen, Binbin Chen, Haoxuan Lan, et al.
+- **abs:** https://arxiv.org/abs/2603.07927 · **PDF:** https://arxiv.org/pdf/2603.07927v1
+
+## 6698. IMSE: Intrinsic Mixture of Spectral Experts Fine-tuning for Test-Time Adaptation
+
+- **ID:** `2603.07926` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Sunghyun Baek, Jaemyung Yu, Seunghee Koh, et al.
+- **abs:** https://arxiv.org/abs/2603.07926 · **PDF:** https://arxiv.org/pdf/2603.07926v2
+
+## 6699. SPREAD: Subspace Representation Distillation for Lifelong Imitation Learning
+
+- **ID:** `2603.08763` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Kaushik Roy, Giovanni D'urso, Nicholas Lawrance, et al.
+- **abs:** https://arxiv.org/abs/2603.08763 · **PDF:** https://arxiv.org/pdf/2603.08763v1
+
+## 6700. Semantic Risk Scoring of Aggregated Metrics: An AI-Driven Approach for Healthcare Data Governance
+
+- **ID:** `2603.07924` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Mohammed Omer Shakeel Ahmed
+- **abs:** https://arxiv.org/abs/2603.07924 · **PDF:** https://arxiv.org/pdf/2603.07924v1
+
+## 6701. Robust Transfer Learning with Side Information
+
+- **ID:** `2603.07921` · **primary:** `stat.ML` · **Submitted:** 2026-03-09
+- **Authors:** Akram S. Awad, Shihab Ahmed, Yue Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.07921 · **PDF:** https://arxiv.org/pdf/2603.07921v1
+
+## 6702. Rel-MOSS: Towards Imbalanced Relational Deep Learning on Relational Databases
+
+- **ID:** `2603.07916` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Jun Yin, Peng Huo, Bangguo Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.07916 · **PDF:** https://arxiv.org/pdf/2603.07916v1
+
+## 6703. Ares: Adaptive Reasoning Effort Selection for Efficient LLM Agents
+
+- **ID:** `2603.07915` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Jingbo Yang, Bairu Hou, Wei Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.07915 · **PDF:** https://arxiv.org/pdf/2603.07915v1
+
+## 6704. Long-Short Term Agents for Pure-Vision Bronchoscopy Robotic Autonomy
+
+- **ID:** `2603.07909` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Junyang Wu, Mingyi Luo, Fangfang Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.07909 · **PDF:** https://arxiv.org/pdf/2603.07909v1
+
+## 6705. DyQ-VLA: Temporal-Dynamic-Aware Quantization for Embodied Vision-Language-Action Models
+
+- **ID:** `2603.07904` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Zihao Zheng, Hangyu Cao, Sicheng Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.07904 · **PDF:** https://arxiv.org/pdf/2603.07904v2
+
+## 6706. NaviDriveVLM: Decoupling High-Level Reasoning and Motion Planning for Autonomous Driving
+
+- **ID:** `2603.07901` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Ximeng Tao, Pardis Taghavi, Dimitar Filev, et al.
+- **abs:** https://arxiv.org/abs/2603.07901 · **PDF:** https://arxiv.org/pdf/2603.07901v1
+
+## 6707. EveryQuery: Zero-Shot Clinical Prediction via Task-Conditioned Pretraining over Electronic Health Records
+
+- **ID:** `2603.07900` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Payal Chandak, Gregory Kondas, Isaac Kohane, et al.
+- **abs:** https://arxiv.org/abs/2603.07900 · **PDF:** https://arxiv.org/pdf/2603.07900v1
+
+## 6708. Bayesian Transformer for Probabilistic Load Forecasting in Smart Grids
+
+- **ID:** `2603.07899` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Sajib Debnath, Md. Uzzal Mia
+- **abs:** https://arxiv.org/abs/2603.07899 · **PDF:** https://arxiv.org/pdf/2603.07899v1
+
+## 6709. Revisiting Unknowns: Towards Effective and Efficient Open-Set Active Learning
+
+- **ID:** `2603.07898` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Chen-Chen Zong, Yu-Qi Chi, Xie-Yang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.07898 · **PDF:** https://arxiv.org/pdf/2603.07898v1
+
+## 6710. LeJOT-AutoML: LLM-Driven Feature Engineering for Job Execution Time Prediction in Databricks Cost Optimization
+
+- **ID:** `2603.07897` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Lizhi Ma, Yi-Xiang Hu, Yihui Ren, et al.
+- **abs:** https://arxiv.org/abs/2603.07897 · **PDF:** https://arxiv.org/pdf/2603.07897v1
+
+## 6711. SMGI: A Structural Theory of General Artificial Intelligence
+
+- **ID:** `2603.07896` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Aomar Osmani
+- **abs:** https://arxiv.org/abs/2603.07896 · **PDF:** https://arxiv.org/pdf/2603.07896v1
+
+## 6712. Designing probabilistic AI monsoon forecasts to inform agricultural decision-making
+
+- **ID:** `2603.07893` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Colin Aitken, Rajat Masiwal, Adam Marchakitus, et al.
+- **abs:** https://arxiv.org/abs/2603.07893 · **PDF:** https://arxiv.org/pdf/2603.07893v2
+
+## 6713. Revisiting Sharpness-Aware Minimization: A More Faithful and Effective Implementation
+
+- **ID:** `2603.10048` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Jianlong Chen, Zhiming Zhou
+- **abs:** https://arxiv.org/abs/2603.10048 · **PDF:** https://arxiv.org/pdf/2603.10048v1
+
+## 6714. A Lightweight Traffic Map for Efficient Anytime LaCAM*
+
+- **ID:** `2603.07891` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Bojie Shen, Yue Zhang, Zhe Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.07891 · **PDF:** https://arxiv.org/pdf/2603.07891v1
+
+## 6715. A Reliability Evaluation of Hybrid Deterministic-LLM Based Approaches for Academic Course Registration PDF Information Extraction
+
+- **ID:** `2604.00003` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Muhammad Anis Al Hilmi, Neelansh Khare, Noel Framil Iglesias
+- **abs:** https://arxiv.org/abs/2604.00003 · **PDF:** https://arxiv.org/pdf/2604.00003v1
+
+## 6716. Visualizing Coalition Formation: From Hedonic Games to Image Segmentation
+
+- **ID:** `2603.07890` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Pedro Henrique de Paula França, Lucas Lopes Felipe, Daniel Sadoc Menasché
+- **abs:** https://arxiv.org/abs/2603.07890 · **PDF:** https://arxiv.org/pdf/2603.07890v1
+
+## 6717. VLM-SubtleBench: How Far Are VLMs from Human-Level Subtle Comparative Reasoning?
+
+- **ID:** `2603.07888` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Minkyu Kim, Sangheon Lee, Dongmin Park
+- **abs:** https://arxiv.org/abs/2603.07888 · **PDF:** https://arxiv.org/pdf/2603.07888v1
+
+## 6718. Reject, Resample, Repeat: Understanding Parallel Reasoning in Language Model Inference
+
+- **ID:** `2603.07887` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Noah Golowich, Fan Chen, Dhruv Rohatgi, et al.
+- **abs:** https://arxiv.org/abs/2603.07887 · **PDF:** https://arxiv.org/pdf/2603.07887v1
+
+## 6719. Joint Return and Risk Modeling with Deep Neural Networks for Portfolio Construction
+
+- **ID:** `2603.19288` · **primary:** `q-fin.PM` · **Submitted:** 2026-03-09
+- **Authors:** Keonvin Park
+- **abs:** https://arxiv.org/abs/2603.19288 · **PDF:** https://arxiv.org/pdf/2603.19288v1
+
+## 6720. CCR-Bench: A Comprehensive Benchmark for Evaluating LLMs on Complex Constraints, Control Flows, and Real-World Cases
+
+- **ID:** `2603.07886` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Xiaona Xue, Yiqiao Huang, Jiacheng Li, et al.
+- **abs:** https://arxiv.org/abs/2603.07886 · **PDF:** https://arxiv.org/pdf/2603.07886v1
+
+## 6721. Learning When to Trust in Contextual Bandits
+
+- **ID:** `2603.13356` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Majid Ghasemi, Mark Crowley
+- **abs:** https://arxiv.org/abs/2603.13356 · **PDF:** https://arxiv.org/pdf/2603.13356v1
+
+## 6722. What Do AI Agents Talk About? Emergent Communication Structure in the First AI-Only Social Network
+
+- **ID:** `2603.07880` · **primary:** `cs.CL` · **Submitted:** 2026-03-09
+- **Authors:** Taksch Dube, Jianfeng Zhu, NHatHai Phan, et al.
+- **abs:** https://arxiv.org/abs/2603.07880 · **PDF:** https://arxiv.org/pdf/2603.07880v1
+
+## 6723. Toward Unified Multimodal Representation Learning for Autonomous Driving
+
+- **ID:** `2603.07874` · **primary:** `cs.CV` · **Submitted:** 2026-03-09
+- **Authors:** Ximeng Tao, Dimitar Filev, Gaurav Pandey
+- **abs:** https://arxiv.org/abs/2603.07874 · **PDF:** https://arxiv.org/pdf/2603.07874v1
+
+## 6724. Hospitality-VQA: Decision-Oriented Informativeness Evaluation for Vision-Language Models
+
+- **ID:** `2603.07868` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Jeongwoo Lee, Baek Duhyeong, Eungyeol Han, et al.
+- **abs:** https://arxiv.org/abs/2603.07868 · **PDF:** https://arxiv.org/pdf/2603.07868v1
+
+## 6725. Slumbering to Precision: Enhancing Artificial Neural Network Calibration Through Sleep-like Processes
+
+- **ID:** `2603.07867` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Jean Erik Delanois, Aditya Ahuja, Giri P. Krishnan, et al.
+- **abs:** https://arxiv.org/abs/2603.07867 · **PDF:** https://arxiv.org/pdf/2603.07867v1
+
+## 6726. Viewpoint-Agnostic Grasp Pipeline using VLM and Partial Observations
+
+- **ID:** `2603.07866` · **primary:** `cs.RO` · **Submitted:** 2026-03-09
+- **Authors:** Dilermando Almeida, Juliano Negri, Guilherme Lazzarini, et al.
+- **abs:** https://arxiv.org/abs/2603.07866 · **PDF:** https://arxiv.org/pdf/2603.07866v1
+
+## 6727. An Interpretable Generative Framework for Anomaly Detection in High-Dimensional Financial Time Series
+
+- **ID:** `2603.07864` · **primary:** `stat.ML` · **Submitted:** 2026-03-09
+- **Authors:** Waldyn G Martinez
+- **abs:** https://arxiv.org/abs/2603.07864 · **PDF:** https://arxiv.org/pdf/2603.07864v1
+
+## 6728. Guess & Guide: Gradient-Free Zero-Shot Diffusion Guidance
+
+- **ID:** `2603.07860` · **primary:** `cs.LG` · **Submitted:** 2026-03-09
+- **Authors:** Abduragim Shtanchaev, Albina Ilina, Yazid Janati, et al.
+- **abs:** https://arxiv.org/abs/2603.07860 · **PDF:** https://arxiv.org/pdf/2603.07860v1
+
+## 6729. SynPlanResearch-R1: Encouraging Tool Exploration for Deep Research with Synthetic Plans
+
+- **ID:** `2603.07853` · **primary:** `cs.AI` · **Submitted:** 2026-03-09
+- **Authors:** Hansi Zeng, Zoey Li, Yifan Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.07853 · **PDF:** https://arxiv.org/pdf/2603.07853v1
+
+## 6730. Intentional Deception as Controllable Capability in LLM Agents
+
+- **ID:** `2603.07848` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Jason Starace, Terence Soule
+- **abs:** https://arxiv.org/abs/2603.07848 · **PDF:** https://arxiv.org/pdf/2603.07848v1
+
+## 6731. Benchmark for Assessing Olfactory Perception of Large Language Models
+
+- **ID:** `2604.00002` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Eftychia Makri, Nikolaos Nakis, Laura Sisson, et al.
+- **abs:** https://arxiv.org/abs/2604.00002 · **PDF:** https://arxiv.org/pdf/2604.00002v1
+
+## 6732. An Efficient and Effective Evaluator for Text2SQL Models on Unseen and Unlabeled Data
+
+- **ID:** `2603.07841` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Trinh Pham, Thanh Tam Nguyen, Viet Huynh, et al.
+- **abs:** https://arxiv.org/abs/2603.07841 · **PDF:** https://arxiv.org/pdf/2603.07841v1
+
+## 6733. On the Formal Limits of Alignment Verification
+
+- **ID:** `2603.08761` · **primary:** `stat.ML` · **Submitted:** 2026-03-08
+- **Authors:** Ayushi Agarwal
+- **abs:** https://arxiv.org/abs/2603.08761 · **PDF:** https://arxiv.org/pdf/2603.08761v1
+
+## 6734. AI Steerability 360: A Toolkit for Steering Large Language Models
+
+- **ID:** `2603.07837` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Erik Miehling, Karthikeyan Natesan Ramamurthy, Praveen Venkateswaran, et al.
+- **abs:** https://arxiv.org/abs/2603.07837 · **PDF:** https://arxiv.org/pdf/2603.07837v1
+
+## 6735. DistillGuard: Evaluating Defenses Against LLM Knowledge Distillation
+
+- **ID:** `2603.07835` · **primary:** `cs.CR` · **Submitted:** 2026-03-08
+- **Authors:** Bo Jiang
+- **abs:** https://arxiv.org/abs/2603.07835 · **PDF:** https://arxiv.org/pdf/2603.07835v1
+
+## 6736. AI Misuse in Education Is a Measurement Problem: Toward a Learning Visibility Framework
+
+- **ID:** `2603.07834` · **primary:** `cs.CY` · **Submitted:** 2026-03-08
+- **Authors:** Eduardo Davalos, Yike Zhang
+- **abs:** https://arxiv.org/abs/2603.07834 · **PDF:** https://arxiv.org/pdf/2603.07834v1
+
+## 6737. Gradient Iterated Temporal-Difference Learning
+
+- **ID:** `2603.07833` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Théo Vincent, Kevin Gerhardt, Yogesh Tripathi, et al.
+- **abs:** https://arxiv.org/abs/2603.07833 · **PDF:** https://arxiv.org/pdf/2603.07833v1
+
+## 6738. Transferable Optimization Network for Cross-Domain Image Reconstruction
+
+- **ID:** `2603.07831` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Yunmei Chen, Chi Ding, Xiaojing Ye
+- **abs:** https://arxiv.org/abs/2603.07831 · **PDF:** https://arxiv.org/pdf/2603.07831v1
+
+## 6739. Benchmarking Large Language Models for Quebec Insurance: From Closed-Book to Retrieval-Augmented Generation
+
+- **ID:** `2603.07825` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** David Beauchemin, Richard Khoury
+- **abs:** https://arxiv.org/abs/2603.07825 · **PDF:** https://arxiv.org/pdf/2603.07825v1
+
+## 6740. Two-Stage Optimizer-Aware Online Data Selection for Large Language Models
+
+- **ID:** `2604.00001` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Fangxin Wang, Peyman Baghershahi, Langzhou He, et al.
+- **abs:** https://arxiv.org/abs/2604.00001 · **PDF:** https://arxiv.org/pdf/2604.00001v1
+
+## 6741. Column Generation for the Micro-Transit Zoning Problem
+
+- **ID:** `2603.07821` · **primary:** `math.OC` · **Submitted:** 2026-03-08
+- **Authors:** Hins Hu, Rishav Sen, Jose Paolo Talusan, et al.
+- **abs:** https://arxiv.org/abs/2603.07821 · **PDF:** https://arxiv.org/pdf/2603.07821v1
+
+## 6742. Fusion Complexity Inversion: Why Simpler Cross View Modules Outperform SSMs and Cross View Attention Transformers for Pasture Biomass Regression
+
+- **ID:** `2603.07819` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Mridankan Mandal
+- **abs:** https://arxiv.org/abs/2603.07819 · **PDF:** https://arxiv.org/pdf/2603.07819v2
+
+## 6743. HybridStitch: Pixel and Timestep Level Model Stitching for Diffusion Acceleration
+
+- **ID:** `2603.07815` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Desen Sun, Jason Hon, Jintao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.07815 · **PDF:** https://arxiv.org/pdf/2603.07815v1
+
+## 6744. Learning embeddings of non-linear PDEs: the Burgers' equation
+
+- **ID:** `2603.07812` · **primary:** `math.AP` · **Submitted:** 2026-03-08
+- **Authors:** Pedro Tarancón-Álvarez, Leonid Sarieddine, Pavlos Protopapas, et al.
+- **abs:** https://arxiv.org/abs/2603.07812 · **PDF:** https://arxiv.org/pdf/2603.07812v1
+
+## 6745. Neural Precoding in Complex Projective Spaces
+
+- **ID:** `2603.07811` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Zaid Abdullah, Merouane Debbah, Symeon Chatzinotas, et al.
+- **abs:** https://arxiv.org/abs/2603.07811 · **PDF:** https://arxiv.org/pdf/2603.07811v1
+
+## 6746. Toward Global Intent Inference for Human Motion by Inverse Reinforcement Learning
+
+- **ID:** `2603.07797` · **primary:** `cs.RO` · **Submitted:** 2026-03-08
+- **Authors:** Sarmad Mehrdad, Maxime Sabbah, Vincent Bonnet, et al.
+- **abs:** https://arxiv.org/abs/2603.07797 · **PDF:** https://arxiv.org/pdf/2603.07797v1
+
+## 6747. Dual-Metric Evaluation of Social Bias in Large Language Models: Evidence from an Underrepresented Nepali Cultural Context
+
+- **ID:** `2603.07792` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Ashish Pandey, Tek Raj Chhetri
+- **abs:** https://arxiv.org/abs/2603.07792 · **PDF:** https://arxiv.org/pdf/2603.07792v1
+
+## 6748. Power Couple? AI Growth and Renewable Energy Investment
+
+- **ID:** `2603.26678` · **primary:** `cs.CY` · **Submitted:** 2026-03-08
+- **Authors:** Luyi Gui, Tinglong Dai
+- **abs:** https://arxiv.org/abs/2603.26678 · **PDF:** https://arxiv.org/pdf/2603.26678v1
+
+## 6749. Vision Transformers that Never Stop Learning
+
+- **ID:** `2603.07787` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Caihao Sun, Mingqi Yuan, Shiyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.07787 · **PDF:** https://arxiv.org/pdf/2603.07787v1
+
+## 6750. ProgAgent:A Continual RL Agent with Progress-Aware Rewards
+
+- **ID:** `2603.07784` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Jinzhou Tan, Gabriel Adineera, Jinoh Kim
+- **abs:** https://arxiv.org/abs/2603.07784 · **PDF:** https://arxiv.org/pdf/2603.07784v1
+
+## 6751. Scaling Data Difficulty: Improving Coding Models via Reinforcement Learning on Fresh and Challenging Problems
+
+- **ID:** `2603.07779` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Zongqian Li, Tengchao Lv, Shaohan Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.07779 · **PDF:** https://arxiv.org/pdf/2603.07779v1
+
+## 6752. Lindbladian Learning with Neural Differential Equations
+
+- **ID:** `2603.07778` · **primary:** `quant-ph` · **Submitted:** 2026-03-08
+- **Authors:** Timothy Heightman, Roman Aseguinolaza Gallo, Edward Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.07778 · **PDF:** https://arxiv.org/pdf/2603.07778v1
+
+## 6753. Breaking Training Bottlenecks: Effective and Stable Reinforcement Learning for Coding Models
+
+- **ID:** `2603.07777` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Zongqian Li, Shaohan Huang, Zewen Chi, et al.
+- **abs:** https://arxiv.org/abs/2603.07777 · **PDF:** https://arxiv.org/pdf/2603.07777v1
+
+## 6754. ArcLight: A Lightweight LLM Inference Architecture for Many-Core CPUs
+
+- **ID:** `2603.07770` · **primary:** `cs.DC` · **Submitted:** 2026-03-08
+- **Authors:** Yuzhuang Xu, Xu Han, Yuxuan Li, et al.
+- **abs:** https://arxiv.org/abs/2603.07770 · **PDF:** https://arxiv.org/pdf/2603.07770v1
+
+## 6755. Toward Epistemic Stability: Engineering Consistent Procedures for Industrial LLM Hallucination Reduction
+
+- **ID:** `2603.10047` · **primary:** `cs.SE` · **Submitted:** 2026-03-08
+- **Authors:** Brian Freeman, Adam Kicklighter, Matt Erdman, et al.
+- **abs:** https://arxiv.org/abs/2603.10047 · **PDF:** https://arxiv.org/pdf/2603.10047v1
+
+## 6756. QuadAI at SemEval-2026 Task 3: Ensemble Learning of Hybrid RoBERTa and LLMs for Dimensional Aspect-Based Sentiment Analysis
+
+- **ID:** `2603.07766` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** A. J. W. de Vink, Filippos Karolos Ventirozos, Natalia Amat-Lefort, et al.
+- **abs:** https://arxiv.org/abs/2603.07766 · **PDF:** https://arxiv.org/pdf/2603.07766v1
+
+## 6757. Using GPUs And LLMs Can Be Satisfying for Nonlinear Real Arithmetic Problems
+
+- **ID:** `2603.07764` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Christopher Brix, Julia Walczak, Nils Lommen, et al.
+- **abs:** https://arxiv.org/abs/2603.07764 · **PDF:** https://arxiv.org/pdf/2603.07764v1
+
+## 6758. From Garbage to Gold: A Data-Architectural Theory of Predictive Robustness
+
+- **ID:** `2603.12288` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Terrence J. Lee-St. John, Jordan L. Lawson, Bartlomiej Piechowski-Jozwiak
+- **abs:** https://arxiv.org/abs/2603.12288 · **PDF:** https://arxiv.org/pdf/2603.12288v1
+
+## 6759. Gated Adaptation for Continual Learning in Human Activity Recognition
+
+- **ID:** `2603.10046` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Reza Rahimi Azghan, Gautham Krishna Gudur, Mohit Malu, et al.
+- **abs:** https://arxiv.org/abs/2603.10046 · **PDF:** https://arxiv.org/pdf/2603.10046v1
+
+## 6760. DECADE: A Temporally-Consistent Unsupervised Diffusion Model for Enhanced Rb-82 Dynamic Cardiac PET Image Denoising
+
+- **ID:** `2603.07759` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Yinchi Zhou, Liang Guo, Huidong Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.07759 · **PDF:** https://arxiv.org/pdf/2603.07759v1
+
+## 6761. Whitening Reveals Cluster Commitment as the Geometric Separator of Hallucination Types
+
+- **ID:** `2603.07755` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Matic Korun
+- **abs:** https://arxiv.org/abs/2603.07755 · **PDF:** https://arxiv.org/pdf/2603.07755v1
+
+## 6762. InfoMamba: An Attention-Free Hybrid Mamba-Transformer Model
+
+- **ID:** `2603.18031` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Youjin Wang, Jiaqiao Zhao, Rong Fu, et al.
+- **abs:** https://arxiv.org/abs/2603.18031 · **PDF:** https://arxiv.org/pdf/2603.18031v1
+
+## 6763. Writing literature reviews with AI: principles, hurdles and some lessons learned
+
+- **ID:** `2603.20235` · **primary:** `cs.CY` · **Submitted:** 2026-03-08
+- **Authors:** Saadi Lahlou, Annabelle Gouttebroze, Atrina Oraee, et al.
+- **abs:** https://arxiv.org/abs/2603.20235 · **PDF:** https://arxiv.org/pdf/2603.20235v1
+
+## 6764. Uncertainty-Gated Generative Modeling
+
+- **ID:** `2603.07753` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Xingrui Gu, Haixi Zhang
+- **abs:** https://arxiv.org/abs/2603.07753 · **PDF:** https://arxiv.org/pdf/2603.07753v1
+
+## 6765. 3ViewSense: Spatial and Mental Perspective Reasoning from Orthographic Views in Vision-Language Models
+
+- **ID:** `2603.07751` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Shaoxiong Zhan, Yanlin Lai, Zheng Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.07751 · **PDF:** https://arxiv.org/pdf/2603.07751v1
+
+## 6766. Hide and Find: A Distributed Adversarial Attack on Federated Graph Learning
+
+- **ID:** `2603.07743` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Jinshan Liu, Ken Li, Jiazhe Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.07743 · **PDF:** https://arxiv.org/pdf/2603.07743v1
+
+## 6767. AgriPath: A Systematic Exploration of Architectural Trade-offs for Crop Disease Classification
+
+- **ID:** `2603.13354` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Hamza Mooraj, George Pantazopoulos, Alessandro Suglia
+- **abs:** https://arxiv.org/abs/2603.13354 · **PDF:** https://arxiv.org/pdf/2603.13354v2
+
+## 6768. Large Language Model for Discrete Optimization Problems: Evaluation and Step-by-step Reasoning
+
+- **ID:** `2603.07733` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Tianhao Qian, Guilin Qi, Z. Y. Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.07733 · **PDF:** https://arxiv.org/pdf/2603.07733v1
+
+## 6769. A Novel Multi-Agent Architecture to Reduce Hallucinations of Large Language Models in Multi-Step Structural Modeling
+
+- **ID:** `2603.07728` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Ziheng Geng, Jiachen Liu, Ran Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.07728 · **PDF:** https://arxiv.org/pdf/2603.07728v1
+
+## 6770. Optimizing LLM Annotation of Classroom Discourse through Multi-Agent Orchestration
+
+- **ID:** `2603.13353` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Bakhtawar Ahtisham, Kirk Vanacore, Rene F. Kizilcec
+- **abs:** https://arxiv.org/abs/2603.13353 · **PDF:** https://arxiv.org/pdf/2603.13353v1
+
+## 6771. A Lightweight MPC Bidding Framework for Brand Auction Ads
+
+- **ID:** `2603.07721` · **primary:** `cs.GT` · **Submitted:** 2026-03-08
+- **Authors:** Yuanlong Chen, Bowen Zhu, Bing Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.07721 · **PDF:** https://arxiv.org/pdf/2603.07721v1
+
+## 6772. Rigidity in LLM Bandits with Implications for Human-AI Dyads
+
+- **ID:** `2603.07717` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Haomiaomiao Wang, Tomás E Ward, Lili Zhang
+- **abs:** https://arxiv.org/abs/2603.07717 · **PDF:** https://arxiv.org/pdf/2603.07717v1
+
+## 6773. Clear, Compelling Arguments: Rethinking the Foundations of Frontier AI Safety Cases
+
+- **ID:** `2603.08760` · **primary:** `cs.CY` · **Submitted:** 2026-03-08
+- **Authors:** Shaun Feakins, Ibrahim Habli, Phillip Morgan
+- **abs:** https://arxiv.org/abs/2603.08760 · **PDF:** https://arxiv.org/pdf/2603.08760v1
+
+## 6774. Reverse Distillation: Consistently Scaling Protein Language Model Representations
+
+- **ID:** `2603.07710` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Darius Catrina, Christian Bepler, Samuel Sledzieski, et al.
+- **abs:** https://arxiv.org/abs/2603.07710 · **PDF:** https://arxiv.org/pdf/2603.07710v1
+
+## 6775. YAQIN: Culturally Sensitive, Agentic AI for Mental Healthcare Support Among Muslim Women in the UK
+
+- **ID:** `2603.07709` · **primary:** `cs.HC` · **Submitted:** 2026-03-08
+- **Authors:** Yasmin Zaraket, Céline Mougenot
+- **abs:** https://arxiv.org/abs/2603.07709 · **PDF:** https://arxiv.org/pdf/2603.07709v1
+
+## 6776. VoiceSHIELD-Small: Real-Time Malicious Speech Detection and Transcription
+
+- **ID:** `2603.07708` · **primary:** `cs.SD` · **Submitted:** 2026-03-08
+- **Authors:** Sumit Ranjan, Sugandha Sharma, Ubaid Abbas, et al.
+- **abs:** https://arxiv.org/abs/2603.07708 · **PDF:** https://arxiv.org/pdf/2603.07708v1
+
+## 6777. Deep Incentive Design with Differentiable Equilibrium Blocks
+
+- **ID:** `2603.07705` · **primary:** `cs.GT` · **Submitted:** 2026-03-08
+- **Authors:** Vinzenz Thoma, Georgios Piliouras, Luke Marris
+- **abs:** https://arxiv.org/abs/2603.07705 · **PDF:** https://arxiv.org/pdf/2603.07705v2
+
+## 6778. Step-Size Decay and Structural Stagnation in Greedy Sparse Learning
+
+- **ID:** `2603.07703` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Pablo M. Berná
+- **abs:** https://arxiv.org/abs/2603.07703 · **PDF:** https://arxiv.org/pdf/2603.07703v1
+
+## 6779. TDM-R1: Reinforcing Few-Step Diffusion Models with Non-Differentiable Reward
+
+- **ID:** `2603.07700` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Yihong Luo, Tianyang Hu, Weijian Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.07700 · **PDF:** https://arxiv.org/pdf/2603.07700v1
+
+## 6780. EDMFormer: Genre-Specific Self-Supervised Learning for Music Structure Segmentation
+
+- **ID:** `2603.08759` · **primary:** `cs.SD` · **Submitted:** 2026-03-08
+- **Authors:** Sahal Sajeer, Krish Patel, Oscar Chung, et al.
+- **abs:** https://arxiv.org/abs/2603.08759 · **PDF:** https://arxiv.org/pdf/2603.08759v1
+
+## 6781. Global Convergence of Average Reward Constrained MDPs with Neural Critic and General Policy Parameterization
+
+- **ID:** `2603.07698` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Anirudh Satheesh, Pankaj Kumar Barman, Washim Uddin Mondal, et al.
+- **abs:** https://arxiv.org/abs/2603.07698 · **PDF:** https://arxiv.org/pdf/2603.07698v1
+
+## 6782. Compressed-Domain-Aware Online Video Super-Resolution
+
+- **ID:** `2603.07694` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Yuhang Wang, Hai Li, Shujuan Hou, et al.
+- **abs:** https://arxiv.org/abs/2603.07694 · **PDF:** https://arxiv.org/pdf/2603.07694v1
+
+## 6783. Not All Pretraining are Created Equal: Threshold Tuning and Class Weighting for Imbalanced Polarization Tasks in Low-Resource Settings
+
+- **ID:** `2603.23534` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Abass Oguntade
+- **abs:** https://arxiv.org/abs/2603.23534 · **PDF:** https://arxiv.org/pdf/2603.23534v1
+
+## 6784. Scalable Training of Mixture-of-Experts Models with Megatron Core
+
+- **ID:** `2603.07685` · **primary:** `cs.DC` · **Submitted:** 2026-03-08
+- **Authors:** Zijie Yan, Hongxiao Bai, Xin Yao, et al.
+- **abs:** https://arxiv.org/abs/2603.07685 · **PDF:** https://arxiv.org/pdf/2603.07685v2
+
+## 6785. Mitigating the Memory Bottleneck with Machine Learning-Driven and Data-Aware Microarchitectural Techniques
+
+- **ID:** `2603.07683` · **primary:** `cs.AR` · **Submitted:** 2026-03-08
+- **Authors:** Rahul Bera
+- **abs:** https://arxiv.org/abs/2603.07683 · **PDF:** https://arxiv.org/pdf/2603.07683v1
+
+## 6786. A Primer on Evolutionary Frameworks for Near-Field Multi-Source Localization
+
+- **ID:** `2603.07676` · **primary:** `cs.NE` · **Submitted:** 2026-03-08
+- **Authors:** Seyed Jalaleddin Mousavirad, Parisa Ramezani, Mattias O'Nils, et al.
+- **abs:** https://arxiv.org/abs/2603.07676 · **PDF:** https://arxiv.org/pdf/2603.07676v1
+
+## 6787. Context-Enriched Natural Language Descriptions of Vessel Trajectories
+
+- **ID:** `2603.12287` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Kostas Patroumpas, Alexandros Troupiotis-Kapeliaris, Giannis Spiliopoulos, et al.
+- **abs:** https://arxiv.org/abs/2603.12287 · **PDF:** https://arxiv.org/pdf/2603.12287v1
+
+## 6788. Beyond Surrogates: A Quantitative Analysis for Inter-Metric Relationships
+
+- **ID:** `2603.07671` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Yuanhao Pu, Defu Lian, Enhong Chen
+- **abs:** https://arxiv.org/abs/2603.07671 · **PDF:** https://arxiv.org/pdf/2603.07671v1
+
+## 6789. Memory for Autonomous LLM Agents:Mechanisms, Evaluation, and Emerging Frontiers
+
+- **ID:** `2603.07670` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Pengfei Du
+- **abs:** https://arxiv.org/abs/2603.07670 · **PDF:** https://arxiv.org/pdf/2603.07670v1
+
+## 6790. AI-Driven Phase Identification from X-ray Hyperspectral Imaging of cycled Na-ion Cathode Materials
+
+- **ID:** `2603.07666` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-08
+- **Authors:** Fayçal Adrar, Nicolas Folastre, Chloé Pablos, et al.
+- **abs:** https://arxiv.org/abs/2603.07666 · **PDF:** https://arxiv.org/pdf/2603.07666v1
+
+## 6791. Ref-DGS: Reflective Dual Gaussian Splatting
+
+- **ID:** `2603.07664` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Ningjing Fan, Yiqun Wang, Dongming Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.07664 · **PDF:** https://arxiv.org/pdf/2603.07664v2
+
+## 6792. Partial Differential Equations in the Age of Machine Learning: A Critical Synthesis of Classical, Machine Learning, and Hybrid Methods
+
+- **ID:** `2603.07655` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Mohammad Nooraiepour, Jakub Wiktor Both, Teeratorn Kadeethum, et al.
+- **abs:** https://arxiv.org/abs/2603.07655 · **PDF:** https://arxiv.org/pdf/2603.07655v1
+
+## 6793. Compressed Proximal Federated Learning for Non-Convex Composite Optimization on Heterogeneous Data
+
+- **ID:** `2603.07654` · **primary:** `math.OC` · **Submitted:** 2026-03-08
+- **Authors:** Pu Qiu, Chen Ouyang, Yongyang Xiong, et al.
+- **abs:** https://arxiv.org/abs/2603.07654 · **PDF:** https://arxiv.org/pdf/2603.07654v1
+
+## 6794. Machine Learning Based Identification of Solvents from Post-Desiccation Patterns
+
+- **ID:** `2603.15660` · **primary:** `cond-mat.soft` · **Submitted:** 2026-03-08
+- **Authors:** Jesús Israel Morán-Cortés, Felipe Pacheco-Vázquez
+- **abs:** https://arxiv.org/abs/2603.15660 · **PDF:** https://arxiv.org/pdf/2603.15660v1
+
+## 6795. AtomicVLA: Unlocking the Potential of Atomic Skill Learning in Robots
+
+- **ID:** `2603.07648` · **primary:** `cs.RO` · **Submitted:** 2026-03-08
+- **Authors:** Likui Zhang, Tao Tang, Zhihao Zhan, et al.
+- **abs:** https://arxiv.org/abs/2603.07648 · **PDF:** https://arxiv.org/pdf/2603.07648v1
+
+## 6796. Evaluating Synthetic Data for Baggage Trolley Detection in Airport Logistics
+
+- **ID:** `2603.07645` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Abdeldjalil Taibi, Mohmoud Badlis, Amina Bensalem, et al.
+- **abs:** https://arxiv.org/abs/2603.07645 · **PDF:** https://arxiv.org/pdf/2603.07645v1
+
+## 6797. Helix: Evolutionary Reinforcement Learning for Open-Ended Scientific Problem Solving
+
+- **ID:** `2603.07642` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Chang Su, Zhongkai Hao, Zhizhou Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.07642 · **PDF:** https://arxiv.org/pdf/2603.07642v1
+
+## 6798. Exoskeleton Control through Learning to Reduce Biological Joint Moments in Simulations
+
+- **ID:** `2603.07629` · **primary:** `cs.RO` · **Submitted:** 2026-03-08
+- **Authors:** Zihang You, Xianlian Zhou
+- **abs:** https://arxiv.org/abs/2603.07629 · **PDF:** https://arxiv.org/pdf/2603.07629v1
+
+## 6799. Accelerating Diffusion Models for Generative AI Applications with Silicon Photonics
+
+- **ID:** `2603.07626` · **primary:** `cs.AR` · **Submitted:** 2026-03-08
+- **Authors:** Tharini Suresh, Salma Afifi, Sudeep Pasricha
+- **abs:** https://arxiv.org/abs/2603.07626 · **PDF:** https://arxiv.org/pdf/2603.07626v1
+
+## 6800. SMAT: Staged Multi-Agent Training for Co-Adaptive Exoskeleton Control
+
+- **ID:** `2603.07618` · **primary:** `cs.RO` · **Submitted:** 2026-03-08
+- **Authors:** Yifei Yuan, Ghaith Androwis, Xianlian Zhou
+- **abs:** https://arxiv.org/abs/2603.07618 · **PDF:** https://arxiv.org/pdf/2603.07618v1
+
+## 6801. Compression as Adaptation: Implicit Visual Representation with Diffusion Foundation Models
+
+- **ID:** `2603.07615` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Jiajun He, Zongyu Guo, Zhaoyang Jia, et al.
+- **abs:** https://arxiv.org/abs/2603.07615 · **PDF:** https://arxiv.org/pdf/2603.07615v1
+
+## 6802. KohakuRAG: A simple RAG framework with hierarchical document indexing
+
+- **ID:** `2603.07612` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Shih-Ying Yeh, Yueh-Feng Ku, Ko-Wei Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.07612 · **PDF:** https://arxiv.org/pdf/2603.07612v1
+
+## 6803. Emergency Lane-Change Simulation: A Behavioral Guidance Approach for Risky Scenario Generation
+
+- **ID:** `2603.20234` · **primary:** `cs.RO` · **Submitted:** 2026-03-08
+- **Authors:** Chen Xiong, Cheng Wang, Yuhang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.20234 · **PDF:** https://arxiv.org/pdf/2603.20234v1
+
+## 6804. MAS-H2: A Hierarchical Multi-Agent System for Holistic Cloud-Native Autoscaling
+
+- **ID:** `2603.07607` · **primary:** `cs.DC` · **Submitted:** 2026-03-08
+- **Authors:** Hamed Hamzeh, Parisa Vahdatian
+- **abs:** https://arxiv.org/abs/2603.07607 · **PDF:** https://arxiv.org/pdf/2603.07607v1
+
+## 6805. TT-Sparse: Learning Sparse Rule Models with Differentiable Truth Tables
+
+- **ID:** `2603.07606` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Hans Farrell Soegeng, Sarthak Ketanbhai Modi, Thomas Peyrin
+- **abs:** https://arxiv.org/abs/2603.07606 · **PDF:** https://arxiv.org/pdf/2603.07606v1
+
+## 6806. MetaSort: An Accelerated Approach for Non-uniform Compression and Few-shot Classification of Neural Spike Waveforms
+
+- **ID:** `2603.07602` · **primary:** `eess.SP` · **Submitted:** 2026-03-08
+- **Authors:** Luca M. Meyer, Majid Zamani
+- **abs:** https://arxiv.org/abs/2603.07602 · **PDF:** https://arxiv.org/pdf/2603.07602v1
+
+## 6807. StyleBench: Evaluating Speech Language Models on Conversational Speaking Style Control
+
+- **ID:** `2603.07599` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Haishu Zhao, Aokai Hao, Yuan Ge, et al.
+- **abs:** https://arxiv.org/abs/2603.07599 · **PDF:** https://arxiv.org/pdf/2603.07599v1
+
+## 6808. Shorter Thoughts, Same Answers: Difficulty-Scaled Segment-Wise RL for CoT Compression
+
+- **ID:** `2603.07598` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Ye Tian, Aijun Liu
+- **abs:** https://arxiv.org/abs/2603.07598 · **PDF:** https://arxiv.org/pdf/2603.07598v1
+
+## 6809. Models as Lego Builders: Assembling Malice from Benign Blocks via Semantic Blueprints
+
+- **ID:** `2603.07590` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Chenxi Li, Xianggan Liu, Dake Shen, et al.
+- **abs:** https://arxiv.org/abs/2603.07590 · **PDF:** https://arxiv.org/pdf/2603.07590v1
+
+## 6810. Beyond Hard Constraints: Budget-Conditioned Reachability For Safe Offline Reinforcement Learning
+
+- **ID:** `2603.22292` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Janaka Chathuranga Brahmanage, Akshat Kumar
+- **abs:** https://arxiv.org/abs/2603.22292 · **PDF:** https://arxiv.org/pdf/2603.22292v2
+
+## 6811. Analysis-Driven Procedural Generation of an Engine Sound Dataset with Embedded Control Annotations
+
+- **ID:** `2603.07584` · **primary:** `cs.SD` · **Submitted:** 2026-03-08
+- **Authors:** Robin Doerfler, Lonce Wyse
+- **abs:** https://arxiv.org/abs/2603.07584 · **PDF:** https://arxiv.org/pdf/2603.07584v1
+
+## 6812. Generalized Reduction to the Isotropy for Flexible Equivariant Neural Fields
+
+- **ID:** `2603.08758` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Alejandro García-Castellanos, Gijs Bellaard, Remco Duits, et al.
+- **abs:** https://arxiv.org/abs/2603.08758 · **PDF:** https://arxiv.org/pdf/2603.08758v1
+
+## 6813. KCoEvo: A Knowledge Graph Augmented Framework for Evolutionary Code Generation
+
+- **ID:** `2603.07581` · **primary:** `cs.SE` · **Submitted:** 2026-03-08
+- **Authors:** Jiazhen Kang, Yuchen Lu, Chen Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.07581 · **PDF:** https://arxiv.org/pdf/2603.07581v1
+
+## 6814. Integration of deep generative Anomaly Detection algorithm in high-speed industrial line
+
+- **ID:** `2603.07577` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Niccolò Ferrari, Nicola Zanarini, Michele Fraccaroli, et al.
+- **abs:** https://arxiv.org/abs/2603.07577 · **PDF:** https://arxiv.org/pdf/2603.07577v1
+
+## 6815. TS-MLLM: A Multi-Modal Large Language Model-based Framework for Industrial Time-Series Big Data Analysis
+
+- **ID:** `2603.07572` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Haiteng Wang, Yikang Li, Yunfei Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.07572 · **PDF:** https://arxiv.org/pdf/2603.07572v1
+
+## 6816. A Systematic Comparison of Training Objectives for Out-of-Distribution Detection in Image Classification
+
+- **ID:** `2603.07571` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Furkan Genç, Onat Özdemir, Emre Akbaş
+- **abs:** https://arxiv.org/abs/2603.07571 · **PDF:** https://arxiv.org/pdf/2603.07571v2
+
+## 6817. Constraints Matrix Diffusion based Generative Neural Solver for Vehicle Routing Problems
+
+- **ID:** `2603.07568` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Zhenwei Wang, Tiehua Zhang, Ning Xue, et al.
+- **abs:** https://arxiv.org/abs/2603.07568 · **PDF:** https://arxiv.org/pdf/2603.07568v1
+
+## 6818. Revisiting the LiRA Membership Inference Attack Under Realistic Assumptions
+
+- **ID:** `2603.07567` · **primary:** `cs.CR` · **Submitted:** 2026-03-08
+- **Authors:** Najeeb Jebreel, Mona Khalil, David Sánchez, et al.
+- **abs:** https://arxiv.org/abs/2603.07567 · **PDF:** https://arxiv.org/pdf/2603.07567v1
+
+## 6819. GRD-Net: Generative-Reconstructive-Discriminative Anomaly Detection with Region of Interest Attention Module
+
+- **ID:** `2603.07566` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Niccolò Ferrari, Michele Fraccaroli, Evelina Lamma
+- **abs:** https://arxiv.org/abs/2603.07566 · **PDF:** https://arxiv.org/pdf/2603.07566v1
+
+## 6820. ECG Classification on PTB-XL: A Data-Centric Approach with Simplified CNN-VAE
+
+- **ID:** `2603.07558` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Naqcho Ali Mehdi, Amir Ali
+- **abs:** https://arxiv.org/abs/2603.07558 · **PDF:** https://arxiv.org/pdf/2603.07558v1
+
+## 6821. Nwāchā Munā: A Devanagari Speech Corpus and Proximal Transfer Benchmark for Nepal Bhasha ASR
+
+- **ID:** `2603.07554` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Rishikesh Kumar Sharma, Safal Narshing Shrestha, Jenny Poudel, et al.
+- **abs:** https://arxiv.org/abs/2603.07554 · **PDF:** https://arxiv.org/pdf/2603.07554v2
+
+## 6822. Targeted Speaker Poisoning Framework in Zero-Shot Text-to-Speech
+
+- **ID:** `2603.07551` · **primary:** `cs.SD` · **Submitted:** 2026-03-08
+- **Authors:** Thanapat Trachu, Thanathai Lertpetchpun, Sai Praneeth Karimireddy, et al.
+- **abs:** https://arxiv.org/abs/2603.07551 · **PDF:** https://arxiv.org/pdf/2603.07551v1
+
+## 6823. Learning-free L2-Accented Speech Generation using Phonological Rules
+
+- **ID:** `2603.07550` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Thanathai Lertpetchpun, Yoonjeong Lee, Jihwan Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.07550 · **PDF:** https://arxiv.org/pdf/2603.07550v1
+
+## 6824. Did You Check the Right Pocket? Cost-Sensitive Store Routing for Memory-Augmented Agents
+
+- **ID:** `2603.15658` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Madhava Gaikwad
+- **abs:** https://arxiv.org/abs/2603.15658 · **PDF:** https://arxiv.org/pdf/2603.15658v1
+
+## 6825. COOL-MC: Verifying and Explaining RL Policies for Multi-bridge Network Maintenance
+
+- **ID:** `2603.07546` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Dennis Gross
+- **abs:** https://arxiv.org/abs/2603.07546 · **PDF:** https://arxiv.org/pdf/2603.07546v1
+
+## 6826. DreamSAC: Learning Hamiltonian World Models via Symmetry Exploration
+
+- **ID:** `2603.07545` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Jinzhou Tang, Fan Feng, Minghao Fu, et al.
+- **abs:** https://arxiv.org/abs/2603.07545 · **PDF:** https://arxiv.org/pdf/2603.07545v1
+
+## 6827. Generalized Stock Price Prediction for Multiple Stocks Combined with News Fusion
+
+- **ID:** `2603.19286` · **primary:** `q-fin.ST` · **Submitted:** 2026-03-08
+- **Authors:** Pei-Jun Liao, Hung-Shin Lee, Yao-Fei Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.19286 · **PDF:** https://arxiv.org/pdf/2603.19286v1
+
+## 6828. How Long Can Unified Multimodal Models Generate Images Reliably? Taming Long-Horizon Interleaved Image Generation via Context Curation
+
+- **ID:** `2603.07540` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Haoyu Chen, Qing Liu, Yuqian Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.07540 · **PDF:** https://arxiv.org/pdf/2603.07540v1
+
+## 6829. MAWARITH: A Dataset and Benchmark for Legal Inheritance Reasoning with LLMs
+
+- **ID:** `2603.07539` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Abdessalam Bouchekif, Shahd Gaben, Samer Rashwani, et al.
+- **abs:** https://arxiv.org/abs/2603.07539 · **PDF:** https://arxiv.org/pdf/2603.07539v2
+
+## 6830. Accent Vector: Controllable Accent Manipulation for Multilingual TTS Without Accented Data
+
+- **ID:** `2603.07534` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Thanathai Lertpetchpun, Thanapat Trachu, Jihwan Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.07534 · **PDF:** https://arxiv.org/pdf/2603.07534v1
+
+## 6831. Obliviator Reveals the Cost of Nonlinear Guardedness in Concept Erasure
+
+- **ID:** `2603.07529` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Ramin Akbari, Milad Afshari, Vishnu Naresh Boddeti
+- **abs:** https://arxiv.org/abs/2603.07529 · **PDF:** https://arxiv.org/pdf/2603.07529v1
+
+## 6832. TableMind++: An Uncertainty-Aware Programmatic Agent for Tool-Augmented Table Reasoning
+
+- **ID:** `2603.07528` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Mingyue Cheng, Shuo Yu, Chuang Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.07528 · **PDF:** https://arxiv.org/pdf/2603.07528v2
+
+## 6833. Generative prediction of laser-induced rocket ignition with dynamic latent space representations
+
+- **ID:** `2603.07525` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Tony Zahtila, Ettore Saetta, Murray Cutforth, et al.
+- **abs:** https://arxiv.org/abs/2603.07525 · **PDF:** https://arxiv.org/pdf/2603.07525v1
+
+## 6834. Neural Dynamics-Informed Pre-trained Framework for Personalized Brain Functional Network Construction
+
+- **ID:** `2603.07524` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Hongjie Jiang, Yifei Tang, Shuqiang Wang
+- **abs:** https://arxiv.org/abs/2603.07524 · **PDF:** https://arxiv.org/pdf/2603.07524v1
+
+## 6835. One-for-All Model Initialization with Frequency-Domain Knowledge
+
+- **ID:** `2603.07523` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Jianlu Shen, Fu Feng, Yucheng Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.07523 · **PDF:** https://arxiv.org/pdf/2603.07523v1
+
+## 6836. Beyond Data Splitting: Full-Data Conformal Prediction by Differential Privacy
+
+- **ID:** `2603.07522` · **primary:** `stat.ML` · **Submitted:** 2026-03-08
+- **Authors:** Young Hyun Cho, Jordan Awan
+- **abs:** https://arxiv.org/abs/2603.07522 · **PDF:** https://arxiv.org/pdf/2603.07522v1
+
+## 6837. SketchGraphNet: A Memory-Efficient Hybrid Graph Transformer for Large-Scale Sketch Corpora Recognition
+
+- **ID:** `2603.07521` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Shilong Chen, Mingyuan Li, Zhaoyang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.07521 · **PDF:** https://arxiv.org/pdf/2603.07521v1
+
+## 6838. Reinforcement learning-based dynamic cleaning scheduling framework for solar energy system
+
+- **ID:** `2603.07518` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Heungjo An
+- **abs:** https://arxiv.org/abs/2603.07518 · **PDF:** https://arxiv.org/pdf/2603.07518v1
+
+## 6839. Beam-aware Kernelized Contextual Bandits for User Association and Beamforming in mmWave Vehicular Networks
+
+- **ID:** `2603.19285` · **primary:** `cs.IT` · **Submitted:** 2026-03-08
+- **Authors:** Xiaoyang He, Manabu Tsukada
+- **abs:** https://arxiv.org/abs/2603.19285 · **PDF:** https://arxiv.org/pdf/2603.19285v1
+
+## 6840. InterReal: A Unified Physics-Based Imitation Framework for Learning Human-Object Interaction Skills
+
+- **ID:** `2603.07516` · **primary:** `cs.RO` · **Submitted:** 2026-03-08
+- **Authors:** Dayang Liang, Yuhang Lin, Xinzhe Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.07516 · **PDF:** https://arxiv.org/pdf/2603.07516v1
+
+## 6841. A Unified View of Drifting and Score-Based Models
+
+- **ID:** `2603.07514` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Chieh-Hsin Lai, Bac Nguyen, Naoki Murata, et al.
+- **abs:** https://arxiv.org/abs/2603.07514 · **PDF:** https://arxiv.org/pdf/2603.07514v2
+
+## 6842. Bolbosh: Script-Aware Flow Matching for Kashmiri Text-to-Speech
+
+- **ID:** `2603.07513` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Tajamul Ashraf, Burhaan Rasheed Zargar, Saeed Abdul Muizz, et al.
+- **abs:** https://arxiv.org/abs/2603.07513 · **PDF:** https://arxiv.org/pdf/2603.07513v1
+
+## 6843. MDKeyChunker: Single-Call LLM Enrichment with Rolling Keys and Key-Based Restructuring for High-Accuracy RAG
+
+- **ID:** `2603.23533` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Bhavik Mangla
+- **abs:** https://arxiv.org/abs/2603.23533 · **PDF:** https://arxiv.org/pdf/2603.23533v2
+
+## 6844. Online Continual Learning for Anomaly Detection in IoT under Data Distribution Shifts
+
+- **ID:** `2603.07507` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Matea Marinova, Shashi Raj Pandey, Junya Shiraishi, et al.
+- **abs:** https://arxiv.org/abs/2603.07507 · **PDF:** https://arxiv.org/pdf/2603.07507v1
+
+## 6845. A Unified Framework for Knowledge Transfer in Bidirectional Model Scaling
+
+- **ID:** `2603.07506` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Jianlu Shen, Fu Feng, Jiaze Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.07506 · **PDF:** https://arxiv.org/pdf/2603.07506v1
+
+## 6846. SeDa: A Unified System for Dataset Discovery and Multi-Entity Augmented Semantic Exploration
+
+- **ID:** `2603.07502` · **primary:** `cs.IR` · **Submitted:** 2026-03-08
+- **Authors:** Kan Ling, Zhen Qin, Yichi Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.07502 · **PDF:** https://arxiv.org/pdf/2603.07502v2
+
+## 6847. Enhanced Random Subspace Local Projections for High-Dimensional Time Series Analysis
+
+- **ID:** `2603.07500` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Eman Khalid, Moimma Ali Khan, Zarmeena Ali, et al.
+- **abs:** https://arxiv.org/abs/2603.07500 · **PDF:** https://arxiv.org/pdf/2603.07500v1
+
+## 6848. From Thinker to Society: Security in Hierarchical Autonomy Evolution of AI Agents
+
+- **ID:** `2603.07496` · **primary:** `cs.CR` · **Submitted:** 2026-03-08
+- **Authors:** Xiaolei Zhang, Lu Zhou, Xiaogang Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.07496 · **PDF:** https://arxiv.org/pdf/2603.07496v2
+
+## 6849. Pushing Bistatic Wireless Sensing toward High Accuracy at the Sub-Wavelength Scale
+
+- **ID:** `2603.07492` · **primary:** `cs.IT` · **Submitted:** 2026-03-08
+- **Authors:** Wenwei Li, Jiarun Zhou, Qinxiao Quan, et al.
+- **abs:** https://arxiv.org/abs/2603.07492 · **PDF:** https://arxiv.org/pdf/2603.07492v1
+
+## 6850. CDEoH: Category-Driven Automatic Algorithm Design With Large Language Models
+
+- **ID:** `2603.19284` · **primary:** `cs.NE` · **Submitted:** 2026-03-08
+- **Authors:** Yu-Nian Wang, Shen-Huan Lyu, Ning Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.19284 · **PDF:** https://arxiv.org/pdf/2603.19284v1
+
+## 6851. A Joint Neural Baseline for Concept, Assertion, and Relation Extraction from Clinical Text
+
+- **ID:** `2603.07487` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Fei Cheng, Ribeka Tanaka, Sadao Kurohashi
+- **abs:** https://arxiv.org/abs/2603.07487 · **PDF:** https://arxiv.org/pdf/2603.07487v1
+
+## 6852. Interpretable-by-Design Transformers via Architectural Stream Independence
+
+- **ID:** `2603.07482` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Clayton Kerce, Alexis Fox
+- **abs:** https://arxiv.org/abs/2603.07482 · **PDF:** https://arxiv.org/pdf/2603.07482v1
+
+## 6853. Quine: Realizing LLM Agents as Native POSIX Processes
+
+- **ID:** `2603.18030` · **primary:** `cs.OS` · **Submitted:** 2026-03-08
+- **Authors:** Hao Ke
+- **abs:** https://arxiv.org/abs/2603.18030 · **PDF:** https://arxiv.org/pdf/2603.18030v1
+
+## 6854. Skip to the Good Part: Representation Structure & Inference-Time Layer Skipping in Diffusion vs. Autoregressive LLMs
+
+- **ID:** `2603.07475` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Raghavv Goel, Risheek Garrepalli, Sudhanshu Agrawal, et al.
+- **abs:** https://arxiv.org/abs/2603.07475 · **PDF:** https://arxiv.org/pdf/2603.07475v1
+
+## 6855. Cross-Modal Taxonomic Generalization in (Vision-) Language Models
+
+- **ID:** `2603.07474` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Tianyang Xu, Marcelo Sandoval-Castaneda, Karen Livescu, et al.
+- **abs:** https://arxiv.org/abs/2603.07474 · **PDF:** https://arxiv.org/pdf/2603.07474v1
+
+## 6856. Give Them an Inch and They Will Take a Mile:Understanding and Measuring Caller Identity Confusion in MCP-Based AI Systems
+
+- **ID:** `2603.07473` · **primary:** `cs.CR` · **Submitted:** 2026-03-08
+- **Authors:** Yuhang Huang, Boyang Ma, Biwei Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.07473 · **PDF:** https://arxiv.org/pdf/2603.07473v1
+
+## 6857. Contact-Guided 3D Genome Structure Generation of E. coli via Diffusion Transformers
+
+- **ID:** `2603.07472` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Mingxin Zhang, Xiaofeng Dai, Yu Yao, et al.
+- **abs:** https://arxiv.org/abs/2603.07472 · **PDF:** https://arxiv.org/pdf/2603.07472v1
+
+## 6858. Towards Lightweight Adaptation of Speech Enhancement Models in Real-World Environments
+
+- **ID:** `2603.07471` · **primary:** `eess.AS` · **Submitted:** 2026-03-08
+- **Authors:** Longbiao Cheng, Shih-Chii Liu
+- **abs:** https://arxiv.org/abs/2603.07471 · **PDF:** https://arxiv.org/pdf/2603.07471v1
+
+## 6859. Engineering Verifiable Modularity in Transformers via Per-Layer Supervision
+
+- **ID:** `2603.18029` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** J. Clayton Kerce
+- **abs:** https://arxiv.org/abs/2603.18029 · **PDF:** https://arxiv.org/pdf/2603.18029v1
+
+## 6860. Probabilistic Inference and Learning with Stein's Method
+
+- **ID:** `2603.07467` · **primary:** `stat.ML` · **Submitted:** 2026-03-08
+- **Authors:** Qiang Liu, Lester Mackey, Chris Oates
+- **abs:** https://arxiv.org/abs/2603.07467 · **PDF:** https://arxiv.org/pdf/2603.07467v1
+
+## 6861. Trusting What You Cannot See: Auditable Fine-Tuning and Inference for Proprietary AI
+
+- **ID:** `2603.07466` · **primary:** `cs.CR` · **Submitted:** 2026-03-08
+- **Authors:** Heng Jin, Chaoyu Zhang, Hexuan Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.07466 · **PDF:** https://arxiv.org/pdf/2603.07466v1
+
+## 6862. Do Machines Fail Like Humans? A Human-Centred Out-of-Distribution Spectrum for Mapping Error Alignment
+
+- **ID:** `2603.07462` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Binxia Xu, Xiaoliang Luo, Luke Dickens, et al.
+- **abs:** https://arxiv.org/abs/2603.07462 · **PDF:** https://arxiv.org/pdf/2603.07462v1
+
+## 6863. The Dual-Stream Transformer: Channelized Architecture for Interpretable Language Modeling
+
+- **ID:** `2603.07461` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** J. Clayton Kerce, Alexis Fox
+- **abs:** https://arxiv.org/abs/2603.07461 · **PDF:** https://arxiv.org/pdf/2603.07461v1
+
+## 6864. Where Do LLM-based Systems Break? A System-Level Security Framework for Risk Assessment and Treatment
+
+- **ID:** `2603.07460` · **primary:** `cs.CR` · **Submitted:** 2026-03-08
+- **Authors:** Neha Nagaraja, Hayretdin Bahsi
+- **abs:** https://arxiv.org/abs/2603.07460 · **PDF:** https://arxiv.org/pdf/2603.07460v1
+
+## 6865. "Better Ask for Forgiveness than Permission": Practices and Policies of AI Disclosure in Freelance Work
+
+- **ID:** `2603.07459` · **primary:** `cs.HC` · **Submitted:** 2026-03-08
+- **Authors:** Angel Hsing-Chi Hwang, Senya Wong, Baixiao Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.07459 · **PDF:** https://arxiv.org/pdf/2603.07459v1
+
+## 6866. Image Generation Models: A Technical History
+
+- **ID:** `2603.07455` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Rouzbeh Shirvani
+- **abs:** https://arxiv.org/abs/2603.07455 · **PDF:** https://arxiv.org/pdf/2603.07455v2
+
+## 6867. SLNet: A Super-Lightweight Geometry-Adaptive Network for 3D Point Cloud Recognition
+
+- **ID:** `2603.07454` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Mohammad Saeid, Amir Salarpour, Pedram MohajerAnsari, et al.
+- **abs:** https://arxiv.org/abs/2603.07454 · **PDF:** https://arxiv.org/pdf/2603.07454v1
+
+## 6868. Backdoor4Good: Benchmarking Beneficial Uses of Backdoors in LLMs
+
+- **ID:** `2603.07452` · **primary:** `cs.CR` · **Submitted:** 2026-03-08
+- **Authors:** Yige Li, Wei Zhao, Zhe Li, et al.
+- **abs:** https://arxiv.org/abs/2603.07452 · **PDF:** https://arxiv.org/pdf/2603.07452v1
+
+## 6869. Dial: A Knowledge-Grounded Dialect-Specific NL2SQL System
+
+- **ID:** `2603.07449` · **primary:** `cs.DB` · **Submitted:** 2026-03-08
+- **Authors:** Xiang Zhang, Hongming Xu, Le Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.07449 · **PDF:** https://arxiv.org/pdf/2603.07449v1
+
+## 6870. Discrete Tokenization Unlocks Transformers for Calibrated Tabular Forecasting
+
+- **ID:** `2603.07448` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Yael S. Elmatad
+- **abs:** https://arxiv.org/abs/2603.07448 · **PDF:** https://arxiv.org/pdf/2603.07448v1
+
+## 6871. Few Tokens, Big Leverage: Preserving Safety Alignment by Constraining Safety Tokens during Fine-tuning
+
+- **ID:** `2603.07445` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Guoli Wang, Haonan Shi, Tu Ouyang, et al.
+- **abs:** https://arxiv.org/abs/2603.07445 · **PDF:** https://arxiv.org/pdf/2603.07445v1
+
+## 6872. HLER: Human-in-the-Loop Economic Research via Multi-Agent Pipelines for Empirical Discovery
+
+- **ID:** `2603.07444` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Chen Zhu, Xiaolu Wang
+- **abs:** https://arxiv.org/abs/2603.07444 · **PDF:** https://arxiv.org/pdf/2603.07444v1
+
+## 6873. Machine Learning for Stress Testing: Uncertainty Decomposition in Causal Panel Prediction
+
+- **ID:** `2603.07438` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Yu Wang, Xiangchen Liu, Siguang Li
+- **abs:** https://arxiv.org/abs/2603.07438 · **PDF:** https://arxiv.org/pdf/2603.07438v1
+
+## 6874. Cost-Driven Representation Learning for Linear Quadratic Gaussian Control: Part II
+
+- **ID:** `2603.07437` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Yi Tian, Kaiqing Zhang, Russ Tedrake, et al.
+- **abs:** https://arxiv.org/abs/2603.07437 · **PDF:** https://arxiv.org/pdf/2603.07437v1
+
+## 6875. Data Agent: Learning to Select Data via End-to-End Dynamic Optimization
+
+- **ID:** `2603.07433` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Suorong Yang, Fangjian Su, Hai Gan, et al.
+- **abs:** https://arxiv.org/abs/2603.07433 · **PDF:** https://arxiv.org/pdf/2603.07433v1
+
+## 6876. Generalization in Online Reinforcement Learning for Mobile Agents
+
+- **ID:** `2603.07432` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Li Gu, Zihuan Jiang, Zhixiang Chi, et al.
+- **abs:** https://arxiv.org/abs/2603.07432 · **PDF:** https://arxiv.org/pdf/2603.07432v1
+
+## 6877. OrthoFormer: Instrumental Variable Estimation in Transformer Hidden States via Neural Control Functions
+
+- **ID:** `2603.07431` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Charles Luo
+- **abs:** https://arxiv.org/abs/2603.07431 · **PDF:** https://arxiv.org/pdf/2603.07431v2
+
+## 6878. AutoControl Arena: Synthesizing Executable Test Environments for Frontier AI Risk Evaluation
+
+- **ID:** `2603.07427` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Changyi Li, Pengfei Lu, Xudong Pan, et al.
+- **abs:** https://arxiv.org/abs/2603.07427 · **PDF:** https://arxiv.org/pdf/2603.07427v2
+
+## 6879. What on Earth is AlphaEarth? Hierarchical structure and functional interpretability for global land cover
+
+- **ID:** `2603.16911` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Ivan Felipe Benavides-Martinez, Justin Guthrie, Jhon Edwin Arias, et al.
+- **abs:** https://arxiv.org/abs/2603.16911 · **PDF:** https://arxiv.org/pdf/2603.16911v1
+
+## 6880. Dynamic Vehicle Routing Problem with Prompt Confirmation of Advance Requests
+
+- **ID:** `2603.07422` · **primary:** `cs.AI` · **Submitted:** 2026-03-08
+- **Authors:** Amutheezan Sivagnanam, Ayan Mukhopadhyay, Samitha Samaranayake, et al.
+- **abs:** https://arxiv.org/abs/2603.07422 · **PDF:** https://arxiv.org/pdf/2603.07422v2
+
+## 6881. DualSpec: Accelerating Deep Research Agents via Dual-Process Action Speculation
+
+- **ID:** `2603.07416` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Shuzhang Zhong, Baotong Lu, Qi Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.07416 · **PDF:** https://arxiv.org/pdf/2603.07416v1
+
+## 6882. Context Channel Capacity: An Information-Theoretic Framework for Understanding Catastrophic Forgetting
+
+- **ID:** `2603.07415` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Ran Cheng
+- **abs:** https://arxiv.org/abs/2603.07415 · **PDF:** https://arxiv.org/pdf/2603.07415v1
+
+## 6883. Machine Learning for the Internet of Underwater Things: From Fundamentals to Implementation
+
+- **ID:** `2603.07413` · **primary:** `eess.SY` · **Submitted:** 2026-03-08
+- **Authors:** Kenechi Omeke, Attai Abubakar, Michael Mollel, et al.
+- **abs:** https://arxiv.org/abs/2603.07413 · **PDF:** https://arxiv.org/pdf/2603.07413v1
+
+## 6884. Tree-Based Predictive Models for Noisy Input Data
+
+- **ID:** `2603.07409` · **primary:** `stat.ME` · **Submitted:** 2026-03-08
+- **Authors:** Kevin McCoy, Zachary Wooten, Christine B. Peterson
+- **abs:** https://arxiv.org/abs/2603.07409 · **PDF:** https://arxiv.org/pdf/2603.07409v1
+
+## 6885. Safety Under Scaffolding: How Evaluation Conditions Shape Measured Safety
+
+- **ID:** `2603.10044` · **primary:** `cs.SE` · **Submitted:** 2026-03-08
+- **Authors:** David Gringras
+- **abs:** https://arxiv.org/abs/2603.10044 · **PDF:** https://arxiv.org/pdf/2603.10044v1
+
+## 6886. UnSCAR: Universal, Scalable, Controllable, and Adaptable Image Restoration
+
+- **ID:** `2603.07406` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Debabrata Mandal, Soumitri Chattopadhyay, Yujie Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.07406 · **PDF:** https://arxiv.org/pdf/2603.07406v1
+
+## 6887. Adaptive Capacity Allocation for Vision Language Action Fine-tuning
+
+- **ID:** `2603.07404` · **primary:** `cs.RO` · **Submitted:** 2026-03-08
+- **Authors:** Donghoon Kim, Minji Bae, Unghui Nam, et al.
+- **abs:** https://arxiv.org/abs/2603.07404 · **PDF:** https://arxiv.org/pdf/2603.07404v1
+
+## 6888. Generalizing Linear Autoencoder Recommenders with Decoupled Expected Quadratic Loss
+
+- **ID:** `2603.07402` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Ruixin Guo, Xinyu Li, Hao Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.07402 · **PDF:** https://arxiv.org/pdf/2603.07402v1
+
+## 6889. Attribution-Guided Model Rectification of Unreliable Neural Network Behaviors
+
+- **ID:** `2603.15656` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Peiyu Yang, Naveed Akhtar, Jiantong Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.15656 · **PDF:** https://arxiv.org/pdf/2603.15656v1
+
+## 6890. AQuA: Toward Strategic Response Generation for Ambiguous Visual Questions
+
+- **ID:** `2603.07394` · **primary:** `cs.CV` · **Submitted:** 2026-03-08
+- **Authors:** Jihyoung Jang, Hyounghun Kim
+- **abs:** https://arxiv.org/abs/2603.07394 · **PDF:** https://arxiv.org/pdf/2603.07394v1
+
+## 6891. Can Large Language Models Keep Up? Benchmarking Online Adaptation to Continual Knowledge Streams
+
+- **ID:** `2603.07392` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Jiyeon Kim, Hyunji Lee, Dylan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.07392 · **PDF:** https://arxiv.org/pdf/2603.07392v1
+
+## 6892. Deterministic Fuzzy Triage for Legal Compliance Classification and Evidence Retrieval
+
+- **ID:** `2603.07390` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Rian Atri
+- **abs:** https://arxiv.org/abs/2603.07390 · **PDF:** https://arxiv.org/pdf/2603.07390v1
+
+## 6893. Feed m Birds with One Scone: Accelerating Multi-task Gradient Balancing via Bi-level Optimization
+
+- **ID:** `2603.07389` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Xuxing Chen, Yun He, Jiayi Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.07389 · **PDF:** https://arxiv.org/pdf/2603.07389v1
+
+## 6894. Sparsity and Out-of-Distribution Generalization
+
+- **ID:** `2603.07388` · **primary:** `cs.LG` · **Submitted:** 2026-03-08
+- **Authors:** Scott Aaronson, Lin Lin Lee, Jiawei Li
+- **abs:** https://arxiv.org/abs/2603.07388 · **PDF:** https://arxiv.org/pdf/2603.07388v1
+
+## 6895. Generating Hierarchical JSON Representations of Scientific Sentences Using LLMs
+
+- **ID:** `2603.23532` · **primary:** `cs.CL` · **Submitted:** 2026-03-08
+- **Authors:** Satya Sri Rajiteswari Nimmagadda, Ethan Young, Niladri Sengupta, et al.
+- **abs:** https://arxiv.org/abs/2603.23532 · **PDF:** https://arxiv.org/pdf/2603.23532v1
+
+## 6896. SoK: Agentic Retrieval-Augmented Generation (RAG): Taxonomy, Architectures, Evaluation, and Research Directions
+
+- **ID:** `2603.07379` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Saroj Mishra, Suman Niroula, Umesh Yadav, et al.
+- **abs:** https://arxiv.org/abs/2603.07379 · **PDF:** https://arxiv.org/pdf/2603.07379v1
+
+## 6897. Large Language Models Unpack Complex Political Opinions through Target-Stance Extraction
+
+- **ID:** `2603.23531` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Özgür Togay, Florian Kunneman, Javier Garcia-Bernardo, et al.
+- **abs:** https://arxiv.org/abs/2603.23531 · **PDF:** https://arxiv.org/pdf/2603.23531v1
+
+## 6898. Scheduling Parallel Optical Circuit Switches for AI Training
+
+- **ID:** `2603.07373` · **primary:** `cs.NI` · **Submitted:** 2026-03-07
+- **Authors:** Kevin Liang, Litao Qiao, Isaac Keslassy, et al.
+- **abs:** https://arxiv.org/abs/2603.07373 · **PDF:** https://arxiv.org/pdf/2603.07373v1
+
+## 6899. Domain-Specific Quality Estimation for Machine Translation in Low-Resource Scenarios
+
+- **ID:** `2603.07372` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Namrata Patil Gurav, Akashdeep Ranu, Archchana Sindhujan, et al.
+- **abs:** https://arxiv.org/abs/2603.07372 · **PDF:** https://arxiv.org/pdf/2603.07372v1
+
+## 6900. ConfHit: Conformal Generative Design with Oracle Free Guarantees
+
+- **ID:** `2603.07371` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Siddhartha Laghuvarapu, Ying Jin, Jimeng Sun
+- **abs:** https://arxiv.org/abs/2603.07371 · **PDF:** https://arxiv.org/pdf/2603.07371v1
+
+## 6901. Learning to Reflect: Hierarchical Multi-Agent Reinforcement Learning for CSI-Free mmWave Beam-Focusing
+
+- **ID:** `2603.07370` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Hieu Le, Oguz Bedir, Mostafa Ibrahim, et al.
+- **abs:** https://arxiv.org/abs/2603.07370 · **PDF:** https://arxiv.org/pdf/2603.07370v1
+
+## 6902. Position: LLMs Must Use Functor-Based and RAG-Driven Bias Mitigation for Fairness
+
+- **ID:** `2603.07368` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Ravi Ranjan, Utkarsh Grover, Agorista Polyzou
+- **abs:** https://arxiv.org/abs/2603.07368 · **PDF:** https://arxiv.org/pdf/2603.07368v1
+
+## 6903. RILEC: Detection and Generation of L1 Russian Interference Errors in English Learner Texts
+
+- **ID:** `2603.07366` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Darya Kharlamova, Irina Proskurina
+- **abs:** https://arxiv.org/abs/2603.07366 · **PDF:** https://arxiv.org/pdf/2603.07366v1
+
+## 6904. Scaling Laws in the Tiny Regime: How Small Models Change Their Mistakes
+
+- **ID:** `2603.07365` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Mohammed Alnemari, Rizwan Qureshi, Nader Begrazadah
+- **abs:** https://arxiv.org/abs/2603.07365 · **PDF:** https://arxiv.org/pdf/2603.07365v1
+
+## 6905. N-Tree Diffusion for Long-Horizon Wildfire Risk Forecasting
+
+- **ID:** `2603.07361` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Yucheng Xing, Xin Wang
+- **abs:** https://arxiv.org/abs/2603.07361 · **PDF:** https://arxiv.org/pdf/2603.07361v1
+
+## 6906. The Yerkes-Dodson Curve for AI Agents: Emergent Cooperation Under Environmental Pressure in Multi-Agent LLM Simulations
+
+- **ID:** `2603.07360` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Ivan Pasichnyk
+- **abs:** https://arxiv.org/abs/2603.07360 · **PDF:** https://arxiv.org/pdf/2603.07360v1
+
+## 6907. Latent Generative Models with Tunable Complexity for Compressed Sensing and other Inverse Problems
+
+- **ID:** `2603.07357` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Sean Gunn, Jorio Cocola, Oliver De Candido, et al.
+- **abs:** https://arxiv.org/abs/2603.07357 · **PDF:** https://arxiv.org/pdf/2603.07357v2
+
+## 6908. AgrI Challenge: A Data-Centric AI Competition for Cross-Team Validation in Agricultural Vision
+
+- **ID:** `2603.07356` · **primary:** `cs.CV` · **Submitted:** 2026-03-07
+- **Authors:** Mohammed Brahimi, Karim Laabassi, Mohamed Seghir Hadj Ameur, et al.
+- **abs:** https://arxiv.org/abs/2603.07356 · **PDF:** https://arxiv.org/pdf/2603.07356v2
+
+## 6909. A Distributed Gaussian Process Model for Multi-Robot Mapping
+
+- **ID:** `2603.07351` · **primary:** `cs.RO` · **Submitted:** 2026-03-07
+- **Authors:** Seth Nabarro, Mark van der Wilk, Andrew J. Davison
+- **abs:** https://arxiv.org/abs/2603.07351 · **PDF:** https://arxiv.org/pdf/2603.07351v1
+
+## 6910. Learning Clinical Representations Under Systematic Distribution Shift
+
+- **ID:** `2603.07348` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Yuanyun Zhang, Shi Li
+- **abs:** https://arxiv.org/abs/2603.07348 · **PDF:** https://arxiv.org/pdf/2603.07348v1
+
+## 6911. How Much Noise Can BERT Handle? Insights from Multilingual Sentence Difficulty Detection
+
+- **ID:** `2603.07346` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Nouran Khallaf, Serge Sharoff
+- **abs:** https://arxiv.org/abs/2603.07346 · **PDF:** https://arxiv.org/pdf/2603.07346v1
+
+## 6912. Learning Concept Bottleneck Models from Mechanistic Explanations
+
+- **ID:** `2603.07343` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Antonio De Santis, Schrasing Tong, Marco Brambilla, et al.
+- **abs:** https://arxiv.org/abs/2603.07343 · **PDF:** https://arxiv.org/pdf/2603.07343v1
+
+## 6913. Agora: Teaching the Skill of Consensus-Finding with AI Personas Grounded in Human Voice
+
+- **ID:** `2603.07339` · **primary:** `cs.HC` · **Submitted:** 2026-03-07
+- **Authors:** Suyash Fulay, Prerna Ravi, Om Gokhale, et al.
+- **abs:** https://arxiv.org/abs/2603.07339 · **PDF:** https://arxiv.org/pdf/2603.07339v2
+
+## 6914. Explainable and Hardware-Efficient Jamming Detection for 5G Networks Using the Convolutional Tsetlin Machine
+
+- **ID:** `2603.07336` · **primary:** `eess.SP` · **Submitted:** 2026-03-07
+- **Authors:** Vojtech Halenka, Mohammadreza Amini, Per-Arne Andersen, et al.
+- **abs:** https://arxiv.org/abs/2603.07336 · **PDF:** https://arxiv.org/pdf/2603.07336v1
+
+## 6915. VisualScratchpad: Inference-time Visual Concepts Analysis in Vision Language Models
+
+- **ID:** `2603.07335` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Hyesu Lim, Jinho Choi, Taekyung Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.07335 · **PDF:** https://arxiv.org/pdf/2603.07335v1
+
+## 6916. Self-Supervised Evolutionary Learning of Neurodynamic Progression and Identity Manifolds from EEG During Safety-Critical Decision Making
+
+- **ID:** `2603.07334` · **primary:** `cs.NE` · **Submitted:** 2026-03-07
+- **Authors:** Xiaoshan Zhou, Carol C. Menassa, Vineet R. Kamat
+- **abs:** https://arxiv.org/abs/2603.07334 · **PDF:** https://arxiv.org/pdf/2603.07334v1
+
+## 6917. To Predict or Not to Predict? Towards reliable uncertainty estimation in the presence of noise
+
+- **ID:** `2603.07330` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Nouran Khallaf, Serge Sharoff
+- **abs:** https://arxiv.org/abs/2603.07330 · **PDF:** https://arxiv.org/pdf/2603.07330v1
+
+## 6918. The Third Ambition: Artificial Intelligence and the Science of Human Behavior
+
+- **ID:** `2603.07329` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** W. Russell Neuman, Chad Coleman
+- **abs:** https://arxiv.org/abs/2603.07329 · **PDF:** https://arxiv.org/pdf/2603.07329v1
+
+## 6919. Norm-Hierarchy Transitions in Representation Learning: When and Why Neural Networks Abandon Shortcuts
+
+- **ID:** `2603.07323` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Truong Xuan Khanh, Truong Quynh Hoa
+- **abs:** https://arxiv.org/abs/2603.07323 · **PDF:** https://arxiv.org/pdf/2603.07323v1
+
+## 6920. ShakyPrepend: A Multi-Group Learner with Improved Sample Complexity
+
+- **ID:** `2603.07319` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Lujing Zhang, Daniel Hsu, Sivaraman Balakrishnan
+- **abs:** https://arxiv.org/abs/2603.07319 · **PDF:** https://arxiv.org/pdf/2603.07319v1
+
+## 6921. FinSheet-Bench: From Simple Lookups to Complex Reasoning, Where LLMs Break on Financial Spreadsheets
+
+- **ID:** `2603.07316` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Jan Ravnik, Matjaž Ličen, Felix Bührmann, et al.
+- **abs:** https://arxiv.org/abs/2603.07316 · **PDF:** https://arxiv.org/pdf/2603.07316v1
+
+## 6922. Shutdown Safety Valves for Advanced AI
+
+- **ID:** `2603.07315` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Vincent Conitzer
+- **abs:** https://arxiv.org/abs/2603.07315 · **PDF:** https://arxiv.org/pdf/2603.07315v1
+
+## 6923. Adversarial Latent-State Training for Robust Policies in Partially Observable Domains
+
+- **ID:** `2603.07313` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Angad Singh Ahuja
+- **abs:** https://arxiv.org/abs/2603.07313 · **PDF:** https://arxiv.org/pdf/2603.07313v3
+
+## 6924. Data-Driven Hints in Intelligent Tutoring Systems
+
+- **ID:** `2603.07311` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Sutapa Dey Tithi, Kimia Fazeli, Dmitri Droujkov, et al.
+- **abs:** https://arxiv.org/abs/2603.07311 · **PDF:** https://arxiv.org/pdf/2603.07311v1
+
+## 6925. StructSAM: Structure- and Spectrum-Preserving Token Merging for Segment Anything Models
+
+- **ID:** `2603.07307` · **primary:** `cs.CV` · **Submitted:** 2026-03-07
+- **Authors:** Duy M. H. Nguyen, Tuan A. Tran, Duong Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.07307 · **PDF:** https://arxiv.org/pdf/2603.07307v1
+
+## 6926. Prompt Complexity Dilutes Structured Reasoning: A Follow-Up Study on the Car Wash Problem
+
+- **ID:** `2603.13351` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Heejin Jo
+- **abs:** https://arxiv.org/abs/2603.13351 · **PDF:** https://arxiv.org/pdf/2603.13351v1
+
+## 6927. Retrieval-Augmented Multi-scale Framework for County-Level Crop Yield Prediction Across Large Regions
+
+- **ID:** `2603.07305` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Yiming Sun, Qi Cheng, Licheng Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.07305 · **PDF:** https://arxiv.org/pdf/2603.07305v1
+
+## 6928. AutoResearch-RL: Perpetual Self-Evaluating Reinforcement Learning Agents for Autonomous Neural Architecture Discovery
+
+- **ID:** `2603.07300` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Nilesh Jain, Rohit Yadav, Sagar Kotian, et al.
+- **abs:** https://arxiv.org/abs/2603.07300 · **PDF:** https://arxiv.org/pdf/2603.07300v2
+
+## 6929. Spectral Discovery of Continuous Symmetries via Generalized Fourier Transforms
+
+- **ID:** `2603.07299` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Pavan Karjol, Kumar Shubham, Prathosh AP
+- **abs:** https://arxiv.org/abs/2603.07299 · **PDF:** https://arxiv.org/pdf/2603.07299v1
+
+## 6930. A Cortically Inspired Architecture for Modular Perceptual AI
+
+- **ID:** `2603.07295` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Prerna Luthra
+- **abs:** https://arxiv.org/abs/2603.07295 · **PDF:** https://arxiv.org/pdf/2603.07295v1
+
+## 6931. MAviS: A Multimodal Conversational Assistant For Avian Species
+
+- **ID:** `2603.07294` · **primary:** `cs.CV` · **Submitted:** 2026-03-07
+- **Authors:** Yevheniia Kryklyvets, Mohammed Irfan Kurpath, Sahal Shaji Mullappilly, et al.
+- **abs:** https://arxiv.org/abs/2603.07294 · **PDF:** https://arxiv.org/pdf/2603.07294v1
+
+## 6932. Do Deployment Constraints Make LLMs Hallucinate Citations? An Empirical Study across Four Models and Five Prompting Regimes
+
+- **ID:** `2603.07287` · **primary:** `cs.IR` · **Submitted:** 2026-03-07
+- **Authors:** Chen Zhao, Yuan Tang, Yitian Qian
+- **abs:** https://arxiv.org/abs/2603.07287 · **PDF:** https://arxiv.org/pdf/2603.07287v1
+
+## 6933. Taiwan Safety Benchmark and Breeze Guard: Toward Trustworthy AI for Taiwanese Mandarin
+
+- **ID:** `2603.07286` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Po-Chun Hsu, Meng-Hsi Chen, Tsu Ling Chao, et al.
+- **abs:** https://arxiv.org/abs/2603.07286 · **PDF:** https://arxiv.org/pdf/2603.07286v1
+
+## 6934. Fast and Flexible Audio Bandwidth Extension via Vocos
+
+- **ID:** `2603.07285` · **primary:** `eess.AS` · **Submitted:** 2026-03-07
+- **Authors:** Yatharth Sharma
+- **abs:** https://arxiv.org/abs/2603.07285 · **PDF:** https://arxiv.org/pdf/2603.07285v1
+
+## 6935. Thermal Robustness of Retrieval in Dense Associative Memories: LSE vs LSR Kernels
+
+- **ID:** `2603.13350` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Tatiana Petrova
+- **abs:** https://arxiv.org/abs/2603.13350 · **PDF:** https://arxiv.org/pdf/2603.13350v1
+
+## 6936. Variational Flow Maps: Make Some Noise for One-Step Conditional Generation
+
+- **ID:** `2603.07276` · **primary:** `cs.CV` · **Submitted:** 2026-03-07
+- **Authors:** Abbas Mammadov, So Takao, Bohan Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.07276 · **PDF:** https://arxiv.org/pdf/2603.07276v1
+
+## 6937. VisualDeltas: Learning Preferences from Visual Quality Perturbations
+
+- **ID:** `2603.07272` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Hailiang Huang, Yihao Liu, Shengyue Guan, et al.
+- **abs:** https://arxiv.org/abs/2603.07272 · **PDF:** https://arxiv.org/pdf/2603.07272v1
+
+## 6938. Adaptive Double-Booking Strategy for Outpatient Scheduling Using Multi-Objective Reinforcement Learning
+
+- **ID:** `2603.07270` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Ninda Nurseha Amalina, Heungjo An
+- **abs:** https://arxiv.org/abs/2603.07270 · **PDF:** https://arxiv.org/pdf/2603.07270v1
+
+## 6939. Kinematics-Aware Latent World Models for Data-Efficient Autonomous Driving
+
+- **ID:** `2603.07264` · **primary:** `cs.RO` · **Submitted:** 2026-03-07
+- **Authors:** Jiazhuo Li, Linjiang Cao, Qi Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.07264 · **PDF:** https://arxiv.org/pdf/2603.07264v1
+
+## 6940. Turning Time Series into Algebraic Equations: Symbolic Machine Learning for Interpretable Modeling of Chaotic Time Series
+
+- **ID:** `2603.07261` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Madhurima Panja, Grace Younes, Tanujit Chakraborty
+- **abs:** https://arxiv.org/abs/2603.07261 · **PDF:** https://arxiv.org/pdf/2603.07261v1
+
+## 6941. MURE: Hierarchical Multi-Resolution Encoding via Vision-Language Models for Visual Document Retrieval
+
+- **ID:** `2603.13349` · **primary:** `cs.CV` · **Submitted:** 2026-03-07
+- **Authors:** Fengbin Zhu, Zijing Cai, Yuzhe Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.13349 · **PDF:** https://arxiv.org/pdf/2603.13349v1
+
+## 6942. Learning When to Cooperate Under Heterogeneous Goals
+
+- **ID:** `2603.07253` · **primary:** `cs.MA` · **Submitted:** 2026-03-07
+- **Authors:** Max Taylor-Davies, Neil Bramley, Christopher G. Lucas
+- **abs:** https://arxiv.org/abs/2603.07253 · **PDF:** https://arxiv.org/pdf/2603.07253v1
+
+## 6943. LF2L: Loss Fusion Horizontal Federated Learning Across Heterogeneous Feature Spaces Using External Datasets Effectively: A Case Study in Second Primary Cancer Prediction
+
+- **ID:** `2603.07249` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Chia-Fu Lin, Yi-Ju Tseng
+- **abs:** https://arxiv.org/abs/2603.07249 · **PDF:** https://arxiv.org/pdf/2603.07249v1
+
+## 6944. LEPA: Learning Geometric Equivariance in Satellite Remote Sensing Data with a Predictive Architecture
+
+- **ID:** `2603.07246` · **primary:** `cs.CV` · **Submitted:** 2026-03-07
+- **Authors:** Erik Scheurer, Rocco Sedona, Stefan Kesselheim, et al.
+- **abs:** https://arxiv.org/abs/2603.07246 · **PDF:** https://arxiv.org/pdf/2603.07246v1
+
+## 6945. Rethinking Deep Research from the Perspective of Web Content Distribution Matching
+
+- **ID:** `2603.07241` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Zixuan Yu, Zhenheng Tang, Tongliang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.07241 · **PDF:** https://arxiv.org/pdf/2603.07241v1
+
+## 6946. Scaling Self-Supervised Speech Models Uncovers Deep Linguistic Relationships: Evidence from the Pacific Cluster
+
+- **ID:** `2603.07238` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Minu Kim, Hoirin Kim, David R. Mortensen
+- **abs:** https://arxiv.org/abs/2603.07238 · **PDF:** https://arxiv.org/pdf/2603.07238v1
+
+## 6947. Retrieval-Augmented Generation for Predicting Cellular Responses to Gene Perturbation
+
+- **ID:** `2603.07233` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Andrea Giuseppe Di Francesco, Andrea Rubbi, Pietro Liò
+- **abs:** https://arxiv.org/abs/2603.07233 · **PDF:** https://arxiv.org/pdf/2603.07233v1
+
+## 6948. Conditional Rank-Rank Regression via Deep Conditional Transformation Models
+
+- **ID:** `2603.07230` · **primary:** `stat.ME` · **Submitted:** 2026-03-07
+- **Authors:** Xiaoyi Wang, Long Feng, Zhaojun Wang
+- **abs:** https://arxiv.org/abs/2603.07230 · **PDF:** https://arxiv.org/pdf/2603.07230v1
+
+## 6949. A Hybrid LTR-based System via Social Context Embedding for Recommending Solutions of Software Bugs in Developer Communities
+
+- **ID:** `2603.07229` · **primary:** `cs.SE` · **Submitted:** 2026-03-07
+- **Authors:** Fouzi Harrag, Mokdad Khemliche
+- **abs:** https://arxiv.org/abs/2603.07229 · **PDF:** https://arxiv.org/pdf/2603.07229v1
+
+## 6950. LightMedSeg: Lightweight 3D Medical Image Segmentation with Learned Spatial Anchors
+
+- **ID:** `2603.07228` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Kavyansh Tyagi, Vishwas Rathi, Puneet Goyal
+- **abs:** https://arxiv.org/abs/2603.07228 · **PDF:** https://arxiv.org/pdf/2603.07228v1
+
+## 6951. Unlocking Data Value in Finance: A Study on Distillation and Difficulty-Aware Training
+
+- **ID:** `2603.07223` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Chuxue Cao, Honglin Lin, Zhanping Zhong, et al.
+- **abs:** https://arxiv.org/abs/2603.07223 · **PDF:** https://arxiv.org/pdf/2603.07223v1
+
+## 6952. VINO: Video-driven Invariance for Non-contextual Objects via Structural Prior Guided De-contextualization
+
+- **ID:** `2603.07222` · **primary:** `cs.CV` · **Submitted:** 2026-03-07
+- **Authors:** Seul-Ki Yeom, Marcel Simon, Eunbin Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.07222 · **PDF:** https://arxiv.org/pdf/2603.07222v1
+
+## 6953. Margin in Abstract Spaces
+
+- **ID:** `2603.07221` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Yair Ashlagi, Roi Livni, Shay Moran, et al.
+- **abs:** https://arxiv.org/abs/2603.07221 · **PDF:** https://arxiv.org/pdf/2603.07221v1
+
+## 6954. A Miniature Brain Transformer: Thalamic Gating, Hippocampal Lateralization, Amygdaloid Salience, and Prefrontal Working Memory in Attention-Coupled Latent Memory
+
+- **ID:** `2603.07217` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-07
+- **Authors:** Hong Jeong
+- **abs:** https://arxiv.org/abs/2603.07217 · **PDF:** https://arxiv.org/pdf/2603.07217v1
+
+## 6955. Towards Objective Gastrointestinal Auscultation: Automated Segmentation and Annotation of Bowel Sound Patterns
+
+- **ID:** `2603.07215` · **primary:** `cs.SD` · **Submitted:** 2026-03-07
+- **Authors:** Zahra Mansour, Verena Uslar, Dirk Weyhe, et al.
+- **abs:** https://arxiv.org/abs/2603.07215 · **PDF:** https://arxiv.org/pdf/2603.07215v1
+
+## 6956. wDPO: Winsorized Direct Preference Optimization for Robust LLM Alignment
+
+- **ID:** `2603.07211` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Jilong Liu, Yonghui Yang, Pengyang Shao, et al.
+- **abs:** https://arxiv.org/abs/2603.07211 · **PDF:** https://arxiv.org/pdf/2603.07211v1
+
+## 6957. Fusing Driver Perceived and Physical Risk for Safety Critical Scenario Screening in Autonomous Driving
+
+- **ID:** `2603.20232` · **primary:** `cs.RO` · **Submitted:** 2026-03-07
+- **Authors:** Chen Xiong, Ziwen Wang, Deqi Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.20232 · **PDF:** https://arxiv.org/pdf/2603.20232v1
+
+## 6958. Lying to Win: Assessing LLM Deception through Human-AI Games and Parallel-World Probing
+
+- **ID:** `2603.07202` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Arash Marioriyad, Ali Nouri, Mohammad Hossein Rohban, et al.
+- **abs:** https://arxiv.org/abs/2603.07202 · **PDF:** https://arxiv.org/pdf/2603.07202v1
+
+## 6959. A Dual-Graph Spatiotemporal GNN Surrogate for Nonlinear Response Prediction of Reinforced Concrete Beams under Four-Point Bending
+
+- **ID:** `2603.07201` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Zhaoyang Ren, Qilin Li
+- **abs:** https://arxiv.org/abs/2603.07201 · **PDF:** https://arxiv.org/pdf/2603.07201v1
+
+## 6960. $\textbf{Re}^{2}$: Unlocking LLM Reasoning via Reinforcement Learning with Re-solving
+
+- **ID:** `2603.07197` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Pinzheng Wang, Shuli Xu, Juntao Li, et al.
+- **abs:** https://arxiv.org/abs/2603.07197 · **PDF:** https://arxiv.org/pdf/2603.07197v1
+
+## 6961. Shaping Parameter Contribution Patterns for Out-of-Distribution Detection
+
+- **ID:** `2603.07195` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Haonan Xu, Yang Yang
+- **abs:** https://arxiv.org/abs/2603.07195 · **PDF:** https://arxiv.org/pdf/2603.07195v1
+
+## 6962. Governance Architecture for Autonomous Agent Systems: Threats, Framework, and Engineering Practice
+
+- **ID:** `2603.07191` · **primary:** `cs.CR` · **Submitted:** 2026-03-07
+- **Authors:** Yuxu Ge
+- **abs:** https://arxiv.org/abs/2603.07191 · **PDF:** https://arxiv.org/pdf/2603.07191v2
+
+## 6963. From State Changes to Creative Decisions: Documenting and Interpreting Traces Across Creative Domains
+
+- **ID:** `2603.07184` · **primary:** `cs.HC` · **Submitted:** 2026-03-07
+- **Authors:** Xiaohan Peng, Sotiris Piliouras, Carl Abou Saada Nujaim
+- **abs:** https://arxiv.org/abs/2603.07184 · **PDF:** https://arxiv.org/pdf/2603.07184v1
+
+## 6964. Learning to Rank the Initial Branching Order of SAT Solvers
+
+- **ID:** `2603.07176` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Arvid Eriksson, Gabriel Poesia, Roman Bresson, et al.
+- **abs:** https://arxiv.org/abs/2603.07176 · **PDF:** https://arxiv.org/pdf/2603.07176v1
+
+## 6965. AutoTool: Automatic Scaling of Tool-Use Capabilities in RL via Decoupled Entropy Constraints
+
+- **ID:** `2603.13348` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Yirong Zeng, Xiao Ding, Yufei Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.13348 · **PDF:** https://arxiv.org/pdf/2603.13348v1
+
+## 6966. Making LLMs Optimize Multi-Scenario CUDA Kernels Like Experts
+
+- **ID:** `2603.07169` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Yuxuan Han, Meng-Hao Guo, Zhengning Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.07169 · **PDF:** https://arxiv.org/pdf/2603.07169v1
+
+## 6967. Spectral Conditioning of Attention Improves Transformer Performance
+
+- **ID:** `2603.07162` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Hemanth Saratchandran, Simon Lucey
+- **abs:** https://arxiv.org/abs/2603.07162 · **PDF:** https://arxiv.org/pdf/2603.07162v1
+
+## 6968. Improving reasoning at inference time via uncertainty minimisation
+
+- **ID:** `2603.07159` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Nicolas Legrand, Kenneth Enevoldsen, Márton Kardos, et al.
+- **abs:** https://arxiv.org/abs/2603.07159 · **PDF:** https://arxiv.org/pdf/2603.07159v1
+
+## 6969. AMB-DSGDN: Adaptive Modality-Balanced Dynamic Semantic Graph Differential Network for Multimodal Emotion Recognition
+
+- **ID:** `2603.10043` · **primary:** `cs.MM` · **Submitted:** 2026-03-07
+- **Authors:** Yunsheng Wang, Yuntao Shou, Yilong Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.10043 · **PDF:** https://arxiv.org/pdf/2603.10043v1
+
+## 6970. Agentic Planning with Reasoning for Image Styling via Offline RL
+
+- **ID:** `2603.07148` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Subhojyoti Mukherjee, Stefano Petrangeli, Branislav Kveton, et al.
+- **abs:** https://arxiv.org/abs/2603.07148 · **PDF:** https://arxiv.org/pdf/2603.07148v1
+
+## 6971. The DIME Architecture: A Unified Operational Algorithm for Neural Representation, Dynamics, Control and Integration
+
+- **ID:** `2603.12286` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-07
+- **Authors:** Ionel Cristian Vladu, Nicu Bizdoaca, Ionica Pirici, et al.
+- **abs:** https://arxiv.org/abs/2603.12286 · **PDF:** https://arxiv.org/pdf/2603.12286v2
+
+## 6972. Fine-Grained Table Retrieval Through the Lens of Complex Queries
+
+- **ID:** `2603.07146` · **primary:** `cs.IR` · **Submitted:** 2026-03-07
+- **Authors:** Wojciech Kosiuk, Xingyu Ji, Yeounoh Chung, et al.
+- **abs:** https://arxiv.org/abs/2603.07146 · **PDF:** https://arxiv.org/pdf/2603.07146v1
+
+## 6973. PolyGLU: State-Conditional Activation Routing in Transformer Feed-Forward Networks
+
+- **ID:** `2603.13347` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Daniel Nobrega Medeiros
+- **abs:** https://arxiv.org/abs/2603.13347 · **PDF:** https://arxiv.org/pdf/2603.13347v1
+
+## 6974. Emotion Transcription in Conversation: A Benchmark for Capturing Subtle and Complex Emotional States through Natural Language
+
+- **ID:** `2603.07138` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Yoshiki Tanaka, Ryuichi Uehara, Koji Inoue, et al.
+- **abs:** https://arxiv.org/abs/2603.07138 · **PDF:** https://arxiv.org/pdf/2603.07138v1
+
+## 6975. Post Training Quantization for Efficient Dataset Condensation
+
+- **ID:** `2603.13346` · **primary:** `cs.CV` · **Submitted:** 2026-03-07
+- **Authors:** Linh-Tam Tran, Sung-Ho Bae
+- **abs:** https://arxiv.org/abs/2603.13346 · **PDF:** https://arxiv.org/pdf/2603.13346v1
+
+## 6976. Deep Expert Injection for Anchoring Retinal VLMs with Domain-Specific Knowledge
+
+- **ID:** `2603.07131` · **primary:** `cs.CV` · **Submitted:** 2026-03-07
+- **Authors:** Shuai Lu, Meng Wang, Jia Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.07131 · **PDF:** https://arxiv.org/pdf/2603.07131v3
+
+## 6977. Combining Adam and its Inverse Counterpart to Enhance Generalization of Deep Learning Optimizers
+
+- **ID:** `2603.07122` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Tao Shi, Liangming Chen, Long Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.07122 · **PDF:** https://arxiv.org/pdf/2603.07122v1
+
+## 6978. Turn: A Language for Agentic Computation
+
+- **ID:** `2603.08755` · **primary:** `cs.PL` · **Submitted:** 2026-03-07
+- **Authors:** Muyukani Kizito
+- **abs:** https://arxiv.org/abs/2603.08755 · **PDF:** https://arxiv.org/pdf/2603.08755v1
+
+## 6979. aCAPTCHA: Verifying That an Entity Is a Capable Agent via Asymmetric Hardness
+
+- **ID:** `2603.07116` · **primary:** `cs.CR` · **Submitted:** 2026-03-07
+- **Authors:** Zuyao Xu, Xiang Li, Fubin Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.07116 · **PDF:** https://arxiv.org/pdf/2603.07116v1
+
+## 6980. Enhancing Consistency of Werewolf AI through Dialogue Summarization and Persona Information
+
+- **ID:** `2603.07111` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Yoshiki Tanaka, Takumasa Kaneko, Hiroki Onozeki, et al.
+- **abs:** https://arxiv.org/abs/2603.07111 · **PDF:** https://arxiv.org/pdf/2603.07111v1
+
+## 6981. Vision Language Models Cannot Reason About Physical Transformation
+
+- **ID:** `2603.07109` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Dezhi Luo, Yijiang Li, Maijunxian Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.07109 · **PDF:** https://arxiv.org/pdf/2603.07109v1
+
+## 6982. Deep Generative Spatiotemporal Engression for Probabilistic Forecasting of Epidemics
+
+- **ID:** `2603.07108` · **primary:** `stat.ML` · **Submitted:** 2026-03-07
+- **Authors:** Rajdeep Pathak, Tanujit Chakraborty
+- **abs:** https://arxiv.org/abs/2603.07108 · **PDF:** https://arxiv.org/pdf/2603.07108v1
+
+## 6983. Efficient Personalized Reranking with Semi-Autoregressive Generation and Online Knowledge Distillation
+
+- **ID:** `2603.07107` · **primary:** `cs.IR` · **Submitted:** 2026-03-07
+- **Authors:** Kai Cheng, Hao Wang, Wei Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.07107 · **PDF:** https://arxiv.org/pdf/2603.07107v1
+
+## 6984. DDS-UDA: Dual-Domain Synergy for Unsupervised Domain Adaptation in Joint Segmentation of Optic Disc and Optic Cup
+
+- **ID:** `2603.13345` · **primary:** `cs.CV` · **Submitted:** 2026-03-07
+- **Authors:** Yusong Xiao, Yuxuan Wu, Li Xiao, et al.
+- **abs:** https://arxiv.org/abs/2603.13345 · **PDF:** https://arxiv.org/pdf/2603.13345v1
+
+## 6985. Grounding Machine Creativity in Game Design Knowledge Representations: Empirical Probing of LLM-Based Executable Synthesis of Goal Playable Patterns under Structural Constraints
+
+- **ID:** `2603.07101` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Hugh Xuechen Liu, Kıvanç Tatar
+- **abs:** https://arxiv.org/abs/2603.07101 · **PDF:** https://arxiv.org/pdf/2603.07101v2
+
+## 6986. Statistical Contraction for Chance-Constrained Trajectory Optimization of Non-Gaussian Stochastic Systems
+
+- **ID:** `2603.07092` · **primary:** `eess.SY` · **Submitted:** 2026-03-07
+- **Authors:** Rihan Aaron D'Silva, Hiroyasu Tsukamoto
+- **abs:** https://arxiv.org/abs/2603.07092 · **PDF:** https://arxiv.org/pdf/2603.07092v1
+
+## 6987. mAVE: A Watermark for Joint Audio-Visual Generation Models
+
+- **ID:** `2603.07090` · **primary:** `cs.CR` · **Submitted:** 2026-03-07
+- **Authors:** Luyang Si, Leyi Pan, Lijie Wen
+- **abs:** https://arxiv.org/abs/2603.07090 · **PDF:** https://arxiv.org/pdf/2603.07090v1
+
+## 6988. Countdown-Code: A Testbed for Studying The Emergence and Generalization of Reward Hacking in RLVR
+
+- **ID:** `2603.07084` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Muhammad Khalifa, Zohaib Khan, Omer Tafveez, et al.
+- **abs:** https://arxiv.org/abs/2603.07084 · **PDF:** https://arxiv.org/pdf/2603.07084v1
+
+## 6989. Dreamer-CDP: Improving Reconstruction-free World Models Via Continuous Deterministic Representation Prediction
+
+- **ID:** `2603.07083` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Michael Hauri, Friedemann Zenke
+- **abs:** https://arxiv.org/abs/2603.07083 · **PDF:** https://arxiv.org/pdf/2603.07083v1
+
+## 6990. VLN-Cache: Enabling Token Caching for VLN Models with Visual/Semantic Dynamics Awareness
+
+- **ID:** `2603.07080` · **primary:** `cs.RO` · **Submitted:** 2026-03-07
+- **Authors:** Zihao Zheng, Zhihao Mao, Xingyue Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.07080 · **PDF:** https://arxiv.org/pdf/2603.07080v2
+
+## 6991. Entropy-Aware On-Policy Distillation of Language Models
+
+- **ID:** `2603.07079` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Woogyeol Jin, Taywon Min, Yongjin Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.07079 · **PDF:** https://arxiv.org/pdf/2603.07079v1
+
+## 6992. CoTJudger: A Graph-Driven Framework for Automatic Evaluation of Chain-of-Thought Efficiency and Redundancy in LRMs
+
+- **ID:** `2603.07078` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Siyi Li, Jiajun Shi, Shiwen Ni, et al.
+- **abs:** https://arxiv.org/abs/2603.07078 · **PDF:** https://arxiv.org/pdf/2603.07078v1
+
+## 6993. Interpretable Maximum Margin Deep Anomaly Detection
+
+- **ID:** `2603.07073` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Zhiji Yang, Mei Huang, Xinyu Li, et al.
+- **abs:** https://arxiv.org/abs/2603.07073 · **PDF:** https://arxiv.org/pdf/2603.07073v1
+
+## 6994. The Talking Robot: Distortion-Robust Acoustic Models for Robot-Robot Communication
+
+- **ID:** `2603.07072` · **primary:** `cs.RO` · **Submitted:** 2026-03-07
+- **Authors:** Hanlong Li, Karishma Kamalahasan, Jiahui Li, et al.
+- **abs:** https://arxiv.org/abs/2603.07072 · **PDF:** https://arxiv.org/pdf/2603.07072v1
+
+## 6995. User Review Writing via Interview with Dialogue Systems
+
+- **ID:** `2603.07070` · **primary:** `cs.HC` · **Submitted:** 2026-03-07
+- **Authors:** Yoshiki Tanaka, Michimasa Inaba
+- **abs:** https://arxiv.org/abs/2603.07070 · **PDF:** https://arxiv.org/pdf/2603.07070v1
+
+## 6996. MedSteer: Counterfactual Endoscopic Synthesis via Training-Free Activation Steering
+
+- **ID:** `2603.07066` · **primary:** `cs.CV` · **Submitted:** 2026-03-07
+- **Authors:** Trong-Thang Pham, Loc Nguyen, Anh Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.07066 · **PDF:** https://arxiv.org/pdf/2603.07066v1
+
+## 6997. DyACE: Dynamic Algorithm Co-evolution for Online Automated Heuristic Design with Large Language Model
+
+- **ID:** `2603.13344` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Guidong Lu, Yiping Liu, Xiangxiang Zeng
+- **abs:** https://arxiv.org/abs/2603.13344 · **PDF:** https://arxiv.org/pdf/2603.13344v1
+
+## 6998. Bi-directional digital twin prototype anchoring with multi-periodicity learning for few-shot fault diagnosis
+
+- **ID:** `2603.07054` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Pengcheng Xia, Zhichao Dong, Yixiang Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.07054 · **PDF:** https://arxiv.org/pdf/2603.07054v1
+
+## 6999. Animating Petascale Time-varying Data on Commodity Hardware with LLM-assisted Scripting
+
+- **ID:** `2603.07053` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Ishrat Jahan Eliza, Xuan Huang, Aashish Panta, et al.
+- **abs:** https://arxiv.org/abs/2603.07053 · **PDF:** https://arxiv.org/pdf/2603.07053v1
+
+## 7000. AI-Driven Predictive Maintenance with Real-Time Contextual Data Fusion for Connected Vehicles: A Multi-Dataset Evaluation
+
+- **ID:** `2603.13343` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Kushal Khemani, Anjum Nazir Qureshi
+- **abs:** https://arxiv.org/abs/2603.13343 · **PDF:** https://arxiv.org/pdf/2603.13343v1
+
+## 7001. Hindsight Credit Assignment for Long-Horizon LLM Agents
+
+- **ID:** `2603.08754` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Hui-Ze Tan, Xiao-Wen Yang, Hao Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.08754 · **PDF:** https://arxiv.org/pdf/2603.08754v1
+
+## 7002. Did You Forget What I Asked? Prospective Memory Failures in Large Language Models
+
+- **ID:** `2603.23530` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Avni Mittal
+- **abs:** https://arxiv.org/abs/2603.23530 · **PDF:** https://arxiv.org/pdf/2603.23530v1
+
+## 7003. Looking Back and Forth: Cross-Image Attention Calibration and Attentive Preference Learning for Multi-Image Hallucination Mitigation
+
+- **ID:** `2603.07048` · **primary:** `cs.CV` · **Submitted:** 2026-03-07
+- **Authors:** Xiaochen Yang, Hao Fang, Jiawei Kong, et al.
+- **abs:** https://arxiv.org/abs/2603.07048 · **PDF:** https://arxiv.org/pdf/2603.07048v1
+
+## 7004. Self-Supervised Multi-Modal World Model with 4D Space-Time Embedding
+
+- **ID:** `2603.07039` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Lance Legel, Qin Huang, Brandon Voelker, et al.
+- **abs:** https://arxiv.org/abs/2603.07039 · **PDF:** https://arxiv.org/pdf/2603.07039v1
+
+## 7005. Targeted Bit-Flip Attacks on LLM-Based Agents
+
+- **ID:** `2603.10042` · **primary:** `cs.CR` · **Submitted:** 2026-03-07
+- **Authors:** Jialai Wang, Ya Wen, Zhongmou Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.10042 · **PDF:** https://arxiv.org/pdf/2603.10042v1
+
+## 7006. MS2MetGAN: Latent-space adversarial training for metabolite-spectrum matching in MS/MS database search
+
+- **ID:** `2603.13342` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Meng Tsai, Alexzander Dwyer, Estelle Nuckels, et al.
+- **abs:** https://arxiv.org/abs/2603.13342 · **PDF:** https://arxiv.org/pdf/2603.13342v1
+
+## 7007. Beyond Reward Suppression: Reshaping Steganographic Communication Protocols in MARL via Dynamic Representational Circuit Breaking
+
+- **ID:** `2603.15655` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Liu Hung Ming
+- **abs:** https://arxiv.org/abs/2603.15655 · **PDF:** https://arxiv.org/pdf/2603.15655v1
+
+## 7008. Resource-Adaptive Federated Text Generation with Differential Privacy
+
+- **ID:** `2603.07027` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Jiayi Wang, John Gounley, Heidi Hanson
+- **abs:** https://arxiv.org/abs/2603.07027 · **PDF:** https://arxiv.org/pdf/2603.07027v1
+
+## 7009. Permutation-Equivariant 2D State Space Models: Theory and Canonical Architecture for Multivariate Time Series
+
+- **ID:** `2603.08753` · **primary:** `stat.ML` · **Submitted:** 2026-03-07
+- **Authors:** Seungwoo Jeong, Heung-Il Suk
+- **abs:** https://arxiv.org/abs/2603.08753 · **PDF:** https://arxiv.org/pdf/2603.08753v1
+
+## 7010. Language-Aware Distillation for Multilingual Instruction-Following Speech LLMs with ASR-Only Supervision
+
+- **ID:** `2603.07025` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Shreyas Gopal, Donghang Wu, Ashutosh Anshul, et al.
+- **abs:** https://arxiv.org/abs/2603.07025 · **PDF:** https://arxiv.org/pdf/2603.07025v1
+
+## 7011. Enhancing Web Agents with a Hierarchical Memory Tree
+
+- **ID:** `2603.07024` · **primary:** `cs.AI` · **Submitted:** 2026-03-07
+- **Authors:** Yunteng Tan, Zhi Gao, Xinxiao Wu
+- **abs:** https://arxiv.org/abs/2603.07024 · **PDF:** https://arxiv.org/pdf/2603.07024v1
+
+## 7012. Hit-RAG: Learning to Reason with Long Contexts via Preference Alignment
+
+- **ID:** `2603.07023` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Junming Liu, Yuqi Li, Shiping Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.07023 · **PDF:** https://arxiv.org/pdf/2603.07023v1
+
+## 7013. Mind the Discriminability Trap in Source-Free Cross-domain Few-shot Learning
+
+- **ID:** `2603.13341` · **primary:** `cs.CV` · **Submitted:** 2026-03-07
+- **Authors:** Zhenyu Zhang, Yixiong Zou, Yuhua Li, et al.
+- **abs:** https://arxiv.org/abs/2603.13341 · **PDF:** https://arxiv.org/pdf/2603.13341v1
+
+## 7014. RESCHED: Rethinking Flexible Job Shop Scheduling from a Transformer-based Architecture with Simplified States
+
+- **ID:** `2603.07020` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Xiangjie Xiao, Cong Zhang, Wen Song, et al.
+- **abs:** https://arxiv.org/abs/2603.07020 · **PDF:** https://arxiv.org/pdf/2603.07020v1
+
+## 7015. AutoChecklist: Composable Pipelines for Checklist Generation and Scoring with LLM-as-a-Judge
+
+- **ID:** `2603.07019` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Karen Zhou, Chenhao Tan
+- **abs:** https://arxiv.org/abs/2603.07019 · **PDF:** https://arxiv.org/pdf/2603.07019v1
+
+## 7016. TEA-Time: Transporting Effects Across Time
+
+- **ID:** `2603.07018` · **primary:** `stat.ME` · **Submitted:** 2026-03-07
+- **Authors:** Harsh Parikh, Gabriel Levin-Konigsberg, Dominique Perrault-Joncas, et al.
+- **abs:** https://arxiv.org/abs/2603.07018 · **PDF:** https://arxiv.org/pdf/2603.07018v1
+
+## 7017. Can Safety Emerge from Weak Supervision? A Systematic Analysis of Small Language Models
+
+- **ID:** `2603.07017` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Punyajoy Saha, Sudipta Halder, Debjyoti Mondal, et al.
+- **abs:** https://arxiv.org/abs/2603.07017 · **PDF:** https://arxiv.org/pdf/2603.07017v1
+
+## 7018. SuperSkillsStack: Agency, Domain Knowledge, Imagination, and Taste in Human-AI Design Education
+
+- **ID:** `2603.07016` · **primary:** `cs.CY` · **Submitted:** 2026-03-07
+- **Authors:** Qian Huang, King Wang Poon
+- **abs:** https://arxiv.org/abs/2603.07016 · **PDF:** https://arxiv.org/pdf/2603.07016v1
+
+## 7019. Fréchet regression of multivariate distributions with nonparanormal transport
+
+- **ID:** `2603.07014` · **primary:** `stat.ME` · **Submitted:** 2026-03-07
+- **Authors:** Junyoung Park, Irina Gaynanova
+- **abs:** https://arxiv.org/abs/2603.07014 · **PDF:** https://arxiv.org/pdf/2603.07014v1
+
+## 7020. Konkani LLM: Multi-Script Instruction Tuning and Evaluation for a Low-Resource Indian Language
+
+- **ID:** `2603.23529` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Reuben Chagas Fernandes, Gaurang S. Patkar
+- **abs:** https://arxiv.org/abs/2603.23529 · **PDF:** https://arxiv.org/pdf/2603.23529v1
+
+## 7021. Combinatorial Allocation Bandits with Nonlinear Arm Utility
+
+- **ID:** `2603.07005` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Yuki Shibukawa, Koichi Tanaka, Yuta Saito, et al.
+- **abs:** https://arxiv.org/abs/2603.07005 · **PDF:** https://arxiv.org/pdf/2603.07005v1
+
+## 7022. Large Language Model-Driven Full-Component Evolution of Adaptive Large Neighborhood Search
+
+- **ID:** `2603.06996` · **primary:** `cs.NE` · **Submitted:** 2026-03-07
+- **Authors:** Shaohua Yu, Tianyu Chen, Linyan Liu
+- **abs:** https://arxiv.org/abs/2603.06996 · **PDF:** https://arxiv.org/pdf/2603.06996v1
+
+## 7023. Adaptive Discovery of Interpretable Audio Attributes with Multimodal LLMs for Low-Resource Classification
+
+- **ID:** `2603.06991` · **primary:** `cs.SD` · **Submitted:** 2026-03-07
+- **Authors:** Kosuke Yoshimura, Hisashi Kashima
+- **abs:** https://arxiv.org/abs/2603.06991 · **PDF:** https://arxiv.org/pdf/2603.06991v1
+
+## 7024. Foundational World Models Accurately Detect Bimanual Manipulator Failures
+
+- **ID:** `2603.06987` · **primary:** `cs.RO` · **Submitted:** 2026-03-07
+- **Authors:** Isaac R. Ward, Michelle Ho, Houjun Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.06987 · **PDF:** https://arxiv.org/pdf/2603.06987v1
+
+## 7025. Masked Unfairness: Hiding Causality within Zero ATE
+
+- **ID:** `2603.06984` · **primary:** `stat.ML` · **Submitted:** 2026-03-07
+- **Authors:** Zou Yang, Sophia Xiao, Bijan Mazaheri
+- **abs:** https://arxiv.org/abs/2603.06984 · **PDF:** https://arxiv.org/pdf/2603.06984v1
+
+## 7026. Diffusion Controller: Framework, Algorithms and Parameterization
+
+- **ID:** `2603.06981` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Tong Yang, Moonkyung Ryu, Chih-Wei Hsu, et al.
+- **abs:** https://arxiv.org/abs/2603.06981 · **PDF:** https://arxiv.org/pdf/2603.06981v1
+
+## 7027. Discovering the Hidden Role of Gini Index In Prompt-based Classification
+
+- **ID:** `2603.15654` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Ruixi Lin
+- **abs:** https://arxiv.org/abs/2603.15654 · **PDF:** https://arxiv.org/pdf/2603.15654v1
+
+## 7028. NePPO: Near-Potential Policy Optimization for General-Sum Multi-Agent Reinforcement Learning
+
+- **ID:** `2603.06977` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Addison Kalanther, Sanika Bharvirkar, Shankar Sastry, et al.
+- **abs:** https://arxiv.org/abs/2603.06977 · **PDF:** https://arxiv.org/pdf/2603.06977v1
+
+## 7029. A Systematic Investigation of Document Chunking Strategies and Embedding Sensitivity
+
+- **ID:** `2603.06976` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Muhammad Arslan Shaukat, Muntasir Adnan, Carlos C. N. Kuhn
+- **abs:** https://arxiv.org/abs/2603.06976 · **PDF:** https://arxiv.org/pdf/2603.06976v1
+
+## 7030. Elenchus: Generating Knowledge Bases from Prover-Skeptic Dialogues
+
+- **ID:** `2603.06974` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Bradley P. Allen
+- **abs:** https://arxiv.org/abs/2603.06974 · **PDF:** https://arxiv.org/pdf/2603.06974v1
+
+## 7031. Conditional Unbalanced Optimal Transport Maps: An Outlier-Robust Framework for Conditional Generative Modeling
+
+- **ID:** `2603.06972` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Jiwoo Yoon, Kyumin Choi, Jaewoong Choi
+- **abs:** https://arxiv.org/abs/2603.06972 · **PDF:** https://arxiv.org/pdf/2603.06972v2
+
+## 7032. Recursive Language Models Meet Uncertainty: The Surprising Effectiveness of Self-Reflective Program Search for Long Context
+
+- **ID:** `2603.15653` · **primary:** `cs.CL` · **Submitted:** 2026-03-07
+- **Authors:** Keivan Alizadeh, Parshin Shojaee, Minsik Cho, et al.
+- **abs:** https://arxiv.org/abs/2603.15653 · **PDF:** https://arxiv.org/pdf/2603.15653v1
+
+## 7033. Topology-Aware Reinforcement Learning over Graphs for Resilient Power Distribution Networks
+
+- **ID:** `2603.06964` · **primary:** `eess.SY` · **Submitted:** 2026-03-07
+- **Authors:** Roshni Anna Jacob, Prithvi Poddar, Jaidev Goel, et al.
+- **abs:** https://arxiv.org/abs/2603.06964 · **PDF:** https://arxiv.org/pdf/2603.06964v1
+
+## 7034. A SISA-based Machine Unlearning Framework for Power Transformer Inter-Turn Short-Circuit Fault Localization
+
+- **ID:** `2603.06962` · **primary:** `eess.SY` · **Submitted:** 2026-03-07
+- **Authors:** Nanhong Liu, Jingyi Yan, Mucun Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.06962 · **PDF:** https://arxiv.org/pdf/2603.06962v1
+
+## 7035. Learning Quadruped Walking from Seconds of Demonstration
+
+- **ID:** `2603.06961` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Ruipeng Zhang, Hongzhan Yu, Ya-Chien Chang, et al.
+- **abs:** https://arxiv.org/abs/2603.06961 · **PDF:** https://arxiv.org/pdf/2603.06961v1
+
+## 7036. Chart-RL: Generalized Chart Comprehension via Reinforcement Learning with Verifiable Rewards
+
+- **ID:** `2603.06958` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Xin Zhang, Xingyu Li, Rongguang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.06958 · **PDF:** https://arxiv.org/pdf/2603.06958v1
+
+## 7037. Post-Training with Policy Gradients: Optimality and the Base Model Barrier
+
+- **ID:** `2603.06957` · **primary:** `stat.ML` · **Submitted:** 2026-03-07
+- **Authors:** Alireza Mousavi-Hosseini, Murat A. Erdogdu
+- **abs:** https://arxiv.org/abs/2603.06957 · **PDF:** https://arxiv.org/pdf/2603.06957v1
+
+## 7038. Not All Neighbors Matter: Understanding the Impact of Graph Sparsification on GNN Pipelines
+
+- **ID:** `2603.06952` · **primary:** `cs.LG` · **Submitted:** 2026-03-07
+- **Authors:** Yuhang Song, Naima Abrar Shami, Romaric Duvignau, et al.
+- **abs:** https://arxiv.org/abs/2603.06952 · **PDF:** https://arxiv.org/pdf/2603.06952v1
+
+## 7039. How Private Are DNA Embeddings? Inverting Foundation Model Representations of Genomic Sequences
+
+- **ID:** `2603.06950` · **primary:** `q-bio.GN` · **Submitted:** 2026-03-06
+- **Authors:** Sofiane Ouaari, Jules Kreuer, Nico Pfeifer
+- **abs:** https://arxiv.org/abs/2603.06950 · **PDF:** https://arxiv.org/pdf/2603.06950v1
+
+## 7040. Joint MDPs and Reinforcement Learning in Coupled-Dynamics Environments
+
+- **ID:** `2603.06946` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Ege C. Kaya, Mahsa Ghasemi, Abolfazl Hashemi
+- **abs:** https://arxiv.org/abs/2603.06946 · **PDF:** https://arxiv.org/pdf/2603.06946v1
+
+## 7041. AdaBox: Adaptive Density-Based Box Clustering with Parameter Generalization
+
+- **ID:** `2603.13339` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Ahmed Elmahdi
+- **abs:** https://arxiv.org/abs/2603.13339 · **PDF:** https://arxiv.org/pdf/2603.13339v1
+
+## 7042. The Compression Paradox in LLM Inference: Provider-Dependent Energy Effects of Prompt Compression
+
+- **ID:** `2603.23528` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Warren Johnson
+- **abs:** https://arxiv.org/abs/2603.23528 · **PDF:** https://arxiv.org/pdf/2603.23528v1
+
+## 7043. Deep Research, Shallow Evaluation: A Case Study in Meta-Evaluation for Long-Form QA Benchmarks
+
+- **ID:** `2603.06942` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Jena D. Hwang, Varsha Kishore, Amanpreet Singh, et al.
+- **abs:** https://arxiv.org/abs/2603.06942 · **PDF:** https://arxiv.org/pdf/2603.06942v1
+
+## 7044. Physics-Consistent Neural Networks for Learning Deformation and Director Fields in Microstructured Media with Loss-Based Validation Criteria
+
+- **ID:** `2603.06939` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Milad Shirani, Pete H. Gueldner, Murat Khidoyatov, et al.
+- **abs:** https://arxiv.org/abs/2603.06939 · **PDF:** https://arxiv.org/pdf/2603.06939v1
+
+## 7045. Swimba: Switch Mamba Model Scales State Space Models
+
+- **ID:** `2603.06938` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Zhixu Du, Krishna Teja Chitty-Venkata, Murali Emani, et al.
+- **abs:** https://arxiv.org/abs/2603.06938 · **PDF:** https://arxiv.org/pdf/2603.06938v1
+
+## 7046. MindfulAgents: Personalizing Mindfulness Meditation via an Expert-Aligned Multi-Agent System
+
+- **ID:** `2603.06926` · **primary:** `cs.HC` · **Submitted:** 2026-03-06
+- **Authors:** Mengyuan Millie Wu, Zhihan Jiang, Yuang Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.06926 · **PDF:** https://arxiv.org/pdf/2603.06926v1
+
+## 7047. Reforming the Mechanism: Editing Reasoning Patterns in LLMs with Circuit Reshaping
+
+- **ID:** `2603.06923` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Zhenyu Lei, Qiong Wu, Jianxiong Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.06923 · **PDF:** https://arxiv.org/pdf/2603.06923v1
+
+## 7048. NerVE: Nonlinear Eigenspectrum Dynamics in LLM Feed-Forward Networks
+
+- **ID:** `2603.06922` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Nandan Kumar Jha, Brandon Reagen
+- **abs:** https://arxiv.org/abs/2603.06922 · **PDF:** https://arxiv.org/pdf/2603.06922v2
+
+## 7049. CN-CBF: Composite Neural Control Barrier Function for Safe Robot Navigation in Dynamic Environments
+
+- **ID:** `2603.06921` · **primary:** `cs.RO` · **Submitted:** 2026-03-06
+- **Authors:** Bojan Derajić, Sebastian Bernhard, Wolfgang Hönig
+- **abs:** https://arxiv.org/abs/2603.06921 · **PDF:** https://arxiv.org/pdf/2603.06921v1
+
+## 7050. Compression Method Matters: Benchmark-Dependent Output Dynamics in LLM Prompt Compression
+
+- **ID:** `2603.23527` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Warren Johnson
+- **abs:** https://arxiv.org/abs/2603.23527 · **PDF:** https://arxiv.org/pdf/2603.23527v1
+
+## 7051. A Dynamic Self-Evolving Extraction System
+
+- **ID:** `2603.06915` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Moin Amin-Naseri, Hannah Kim, Estevam Hruschka
+- **abs:** https://arxiv.org/abs/2603.06915 · **PDF:** https://arxiv.org/pdf/2603.06915v1
+
+## 7052. Evaluating Generalization Mechanisms in Autonomous Cyber Attack Agents
+
+- **ID:** `2603.10041` · **primary:** `cs.CR` · **Submitted:** 2026-03-06
+- **Authors:** Ondřej Lukáš, Jihoon Shin, Emilia Rivas, et al.
+- **abs:** https://arxiv.org/abs/2603.10041 · **PDF:** https://arxiv.org/pdf/2603.10041v1
+
+## 7053. Plato's Cave: A Human-Centered Research Verification System
+
+- **ID:** `2603.23526` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Matheus Kunzler Maldaner, Raul Valle, Junsung Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.23526 · **PDF:** https://arxiv.org/pdf/2603.23526v1
+
+## 7054. Language Shapes Mental Health Evaluations in Large Language Models
+
+- **ID:** `2603.06910` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Jiayi Xu, Xiyang Hu
+- **abs:** https://arxiv.org/abs/2603.06910 · **PDF:** https://arxiv.org/pdf/2603.06910v1
+
+## 7055. OpenExtract: Automated Data Extraction for Systematic Reviews in Health
+
+- **ID:** `2603.13338` · **primary:** `cs.IR` · **Submitted:** 2026-03-06
+- **Authors:** Jim Achterberg, Bram Van Dijk, Jing Meng, et al.
+- **abs:** https://arxiv.org/abs/2603.13338 · **PDF:** https://arxiv.org/pdf/2603.13338v1
+
+## 7056. MedInjection-FR: Exploring the Role of Native, Synthetic, and Translated Data in Biomedical Instruction Tuning
+
+- **ID:** `2603.06905` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Ikram Belmadani, Oumaima El Khettari, Pacôme Constant dit Beaufils, et al.
+- **abs:** https://arxiv.org/abs/2603.06905 · **PDF:** https://arxiv.org/pdf/2603.06905v1
+
+## 7057. Prompt Compression in Production Task Orchestration: A Pre-Registered Randomized Trial
+
+- **ID:** `2603.23525` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Warren Johnson, Charles Lee
+- **abs:** https://arxiv.org/abs/2603.23525 · **PDF:** https://arxiv.org/pdf/2603.23525v1
+
+## 7058. XGenBoost: Synthesizing Small and Large Tabular Datasets with XGBoost
+
+- **ID:** `2603.06904` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Jim Achterberg, Marcel Haas, Bram van Dijk, et al.
+- **abs:** https://arxiv.org/abs/2603.06904 · **PDF:** https://arxiv.org/pdf/2603.06904v1
+
+## 7059. Empowering Locally Deployable Medical Agent via State Enhanced Logical Skills for FHIR-based Clinical Tasks
+
+- **ID:** `2603.06902` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Wanrong Yang, Zhengliang Liu, Yuan Li, et al.
+- **abs:** https://arxiv.org/abs/2603.06902 · **PDF:** https://arxiv.org/pdf/2603.06902v1
+
+## 7060. Fairness May Backfire: When Leveling-Down Occurs in Fair Machine Learning
+
+- **ID:** `2603.06901` · **primary:** `stat.ML` · **Submitted:** 2026-03-06
+- **Authors:** Yi Yang, Xiangyu Chang, Pei-yu Chen
+- **abs:** https://arxiv.org/abs/2603.06901 · **PDF:** https://arxiv.org/pdf/2603.06901v1
+
+## 7061. Learning From Design Procedure To Generate CAD Programs for Data Augmentation
+
+- **ID:** `2603.06894` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Yan-Ying Chen, Dule Shu, Matthew Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.06894 · **PDF:** https://arxiv.org/pdf/2603.06894v1
+
+## 7062. Single-pass Possibilistic Clustering with Damped Window Footprints
+
+- **ID:** `2603.06889` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Jeffrey Dale, James Keller, Aquila Galusha
+- **abs:** https://arxiv.org/abs/2603.06889 · **PDF:** https://arxiv.org/pdf/2603.06889v1
+
+## 7063. Enhancing the Detection of Coronary Artery Disease Using Machine Learning
+
+- **ID:** `2603.06888` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Karan Kumar Singh, Nikita Gajbhiye, Gouri Sankar Mishra
+- **abs:** https://arxiv.org/abs/2603.06888 · **PDF:** https://arxiv.org/pdf/2603.06888v1
+
+## 7064. Distributed Legal Infrastructure for a Trustworthy Agentic Web
+
+- **ID:** `2603.06884` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Tomer Jordi Chaffer, Victor Jiawei Zhang, Sante Dino Facchini, et al.
+- **abs:** https://arxiv.org/abs/2603.06884 · **PDF:** https://arxiv.org/pdf/2603.06884v1
+
+## 7065. Physics-informed AI Accelerated Retention Analysis of Ferroelectric Vertical NAND: From Day-Scale TCAD to Second-Scale Surrogate Model
+
+- **ID:** `2603.06881` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Gyujun Jeong, Sungwon Cho, Minji Shon, et al.
+- **abs:** https://arxiv.org/abs/2603.06881 · **PDF:** https://arxiv.org/pdf/2603.06881v1
+
+## 7066. Not Too Short, Not Too Long: How LLM Response Length Shapes People's Critical Thinking in Error Detection
+
+- **ID:** `2603.06878` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Natalie Friedman, Adelaide Nyanyo, Kevin Weatherwax, et al.
+- **abs:** https://arxiv.org/abs/2603.06878 · **PDF:** https://arxiv.org/pdf/2603.06878v1
+
+## 7067. Navigating the Concept Space of Language Models
+
+- **ID:** `2603.23524` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Wilson E. Marcílio-Jr, Danilo M. Eler
+- **abs:** https://arxiv.org/abs/2603.23524 · **PDF:** https://arxiv.org/pdf/2603.23524v1
+
+## 7068. Stochastic Attention via Langevin Dynamics on the Modern Hopfield Energy
+
+- **ID:** `2603.06875` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Abdulrahman Alswaidan, Jeffrey D. Varner
+- **abs:** https://arxiv.org/abs/2603.06875 · **PDF:** https://arxiv.org/pdf/2603.06875v2
+
+## 7069. LieCraft: A Multi-Agent Framework for Evaluating Deceptive Capabilities in Language Models
+
+- **ID:** `2603.06874` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Matthew Lyle Olson, Neale Ratzlaff, Musashi Hinck, et al.
+- **abs:** https://arxiv.org/abs/2603.06874 · **PDF:** https://arxiv.org/pdf/2603.06874v1
+
+## 7070. MultiSolSegment: Multi-channel segmentation of overlapping features in electroluminescence images of photovoltaic cells
+
+- **ID:** `2603.13337` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Ojas Sanghi, Norman Jost, Benjamin G. Pierce, et al.
+- **abs:** https://arxiv.org/abs/2603.13337 · **PDF:** https://arxiv.org/pdf/2603.13337v1
+
+## 7071. Kernel Methods for Some Transport Equations with Application to Learning Kernels for the Approximation of Koopman Eigenfunctions: A Unified Approach via Variational Methods, Green's Functions and the Method of Characteristics
+
+- **ID:** `2603.06872` · **primary:** `math.NA` · **Submitted:** 2026-03-06
+- **Authors:** Boumediene Hamzi, Houman Owhadi, Umesh Vaidya
+- **abs:** https://arxiv.org/abs/2603.06872 · **PDF:** https://arxiv.org/pdf/2603.06872v1
+
+## 7072. LEAD: Breaking the No-Recovery Bottleneck in Long-Horizon Reasoning
+
+- **ID:** `2603.06870` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Denys Pushkin, Emmanuel Abbe
+- **abs:** https://arxiv.org/abs/2603.06870 · **PDF:** https://arxiv.org/pdf/2603.06870v1
+
+## 7073. Symmetry-Constrained Language-Guided Program Synthesis for Discovering Governing Equations from Noisy and Partial Observations
+
+- **ID:** `2603.06869` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Mirza Samad Ahmed Baig, Syeda Anshrah Gillani
+- **abs:** https://arxiv.org/abs/2603.06869 · **PDF:** https://arxiv.org/pdf/2603.06869v1
+
+## 7074. Counting on Consensus: Selecting the Right Inter-annotator Agreement Metric for NLP Annotation and Evaluation
+
+- **ID:** `2603.06865` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Joseph James
+- **abs:** https://arxiv.org/abs/2603.06865 · **PDF:** https://arxiv.org/pdf/2603.06865v2
+
+## 7075. A prior information informed learning architecture for flying trajectory prediction
+
+- **ID:** `2603.06863` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Xianda Huang, Zidong Han, Ruibo Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.06863 · **PDF:** https://arxiv.org/pdf/2603.06863v1
+
+## 7076. Supporting Artifact Evaluation with LLMs: A Study with Published Security Research Papers
+
+- **ID:** `2603.06862` · **primary:** `cs.CR` · **Submitted:** 2026-03-06
+- **Authors:** David Heye, Karl Kindermann, Robin Decker, et al.
+- **abs:** https://arxiv.org/abs/2603.06862 · **PDF:** https://arxiv.org/pdf/2603.06862v1
+
+## 7077. Evaluating Human-AI Safety: A Framework for Measuring Harmful Capability Uplift
+
+- **ID:** `2603.26676` · **primary:** `cs.CY` · **Submitted:** 2026-03-06
+- **Authors:** Michelle Vaccaro, Jaeyoon Song, Abdullah Almaatouq, et al.
+- **abs:** https://arxiv.org/abs/2603.26676 · **PDF:** https://arxiv.org/pdf/2603.26676v1
+
+## 7078. IGLU: The Integrated Gaussian Linear Unit Activation Function
+
+- **ID:** `2603.06861` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Mingi Kang, Zai Yang, Jeova Farias Sales Rocha Neto
+- **abs:** https://arxiv.org/abs/2603.06861 · **PDF:** https://arxiv.org/pdf/2603.06861v1
+
+## 7079. Contextual Counterfactual Credit Assignment for Multi-Agent Reinforcement Learning in LLM Collaboration
+
+- **ID:** `2603.06859` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Yanjun Chen, Yirong Sun, Hanlin Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.06859 · **PDF:** https://arxiv.org/pdf/2603.06859v1
+
+## 7080. Are Audio-Language Models Listening? Audio-Specialist Heads for Adaptive Audio Steering
+
+- **ID:** `2603.06854` · **primary:** `cs.SD` · **Submitted:** 2026-03-06
+- **Authors:** Neta Glazer, Lenny Aharon, Ethan Fetaya
+- **abs:** https://arxiv.org/abs/2603.06854 · **PDF:** https://arxiv.org/pdf/2603.06854v1
+
+## 7081. Bilateral Trade Under Heavy-Tailed Valuations: Minimax Regret with Infinite Variance
+
+- **ID:** `2603.06851` · **primary:** `stat.ML` · **Submitted:** 2026-03-06
+- **Authors:** Hangyi Zhao
+- **abs:** https://arxiv.org/abs/2603.06851 · **PDF:** https://arxiv.org/pdf/2603.06851v1
+
+## 7082. Twitch: Learning Abstractions for Equational Theorem Proving
+
+- **ID:** `2603.06849` · **primary:** `cs.LO` · **Submitted:** 2026-03-06
+- **Authors:** Guy Axelrod, Moa Johansson, Nicholas Smallbone
+- **abs:** https://arxiv.org/abs/2603.06849 · **PDF:** https://arxiv.org/pdf/2603.06849v1
+
+## 7083. Validation of a Small Language Model for DSM-5 Substance Category Classification in Child Welfare Records
+
+- **ID:** `2603.06836` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Brian E. Perron, Dragan Stoll, Bryan G. Victor, et al.
+- **abs:** https://arxiv.org/abs/2603.06836 · **PDF:** https://arxiv.org/pdf/2603.06836v1
+
+## 7084. Joint 3D Gravity and Magnetic Inversion via Rectified Flow and Ginzburg-Landau Guidance
+
+- **ID:** `2603.06829` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Dhruman Gupta, Yashas Shende, Aritra Das, et al.
+- **abs:** https://arxiv.org/abs/2603.06829 · **PDF:** https://arxiv.org/pdf/2603.06829v1
+
+## 7085. Step-Level Visual Grounding Faithfulness Predicts Out-of-Distribution Generalization in Long-Horizon Vision-Language Models
+
+- **ID:** `2603.06828` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Md Ashikur Rahman, Md Arifur Rahman, Niamul Hassan Samin, et al.
+- **abs:** https://arxiv.org/abs/2603.06828 · **PDF:** https://arxiv.org/pdf/2603.06828v1
+
+## 7086. CREDO: Epistemic-Aware Conformalized Credal Envelopes for Regression
+
+- **ID:** `2603.06826` · **primary:** `stat.ML` · **Submitted:** 2026-03-06
+- **Authors:** Luben M. C. Cabezas, Sabina J. Sloman, Bruno M. Resende, et al.
+- **abs:** https://arxiv.org/abs/2603.06826 · **PDF:** https://arxiv.org/pdf/2603.06826v1
+
+## 7087. "Dark Triad" Model Organisms of Misalignment: Narrow Fine-Tuning Mirrors Human Antisocial Behavior
+
+- **ID:** `2603.06816` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Roshni Lulla, Fiona Collins, Sanaya Parekh, et al.
+- **abs:** https://arxiv.org/abs/2603.06816 · **PDF:** https://arxiv.org/pdf/2603.06816v1
+
+## 7088. AI-Assisted Curation of Conference Scholarship: Compiling, Structuring, and Analyzing Two Decades of Presentations at the Society for Social Work and Research
+
+- **ID:** `2603.06814` · **primary:** `cs.DL` · **Submitted:** 2026-03-06
+- **Authors:** Brian Perron, Bryan Victor, Zia Qi
+- **abs:** https://arxiv.org/abs/2603.06814 · **PDF:** https://arxiv.org/pdf/2603.06814v1
+
+## 7089. Reinforcing the World's Edge: A Continual Learning Problem in the Multi-Agent-World Boundary
+
+- **ID:** `2603.06813` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Dane Malenfant
+- **abs:** https://arxiv.org/abs/2603.06813 · **PDF:** https://arxiv.org/pdf/2603.06813v1
+
+## 7090. Making AI Evaluation Deployment Relevant Through Context Specification
+
+- **ID:** `2603.06811` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Matthew Holmes, Thiago Lacerda, Reva Schwartz
+- **abs:** https://arxiv.org/abs/2603.06811 · **PDF:** https://arxiv.org/pdf/2603.06811v1
+
+## 7091. Multi-Agent Reinforcement Learning with Submodular Reward
+
+- **ID:** `2603.06810` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Wenjing Chen, Chengyuan Qian, Shuo Xing, et al.
+- **abs:** https://arxiv.org/abs/2603.06810 · **PDF:** https://arxiv.org/pdf/2603.06810v1
+
+## 7092. A Hybrid Machine Learning Model for Cerebral Palsy Detection
+
+- **ID:** `2603.06803` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Karan Kumar Singh, Nikita Gajbhiye, Gouri Sankar Mishra
+- **abs:** https://arxiv.org/abs/2603.06803 · **PDF:** https://arxiv.org/pdf/2603.06803v1
+
+## 7093. Breaking the Martingale Curse: Multi-Agent Debate via Asymmetric Cognitive Potential Energy
+
+- **ID:** `2603.06801` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Yuhan Liu, Juntian Zhang, Yichen Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.06801 · **PDF:** https://arxiv.org/pdf/2603.06801v1
+
+## 7094. NEST: Network- and Memory-Aware Device Placement For Distributed Deep Learning
+
+- **ID:** `2603.06798` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Irene Wang, Vishnu Varma Venkata, Arvind Krishnamurthy, et al.
+- **abs:** https://arxiv.org/abs/2603.06798 · **PDF:** https://arxiv.org/pdf/2603.06798v1
+
+## 7095. Best-of-Tails: Bridging Optimism and Pessimism in Inference-Time Alignment
+
+- **ID:** `2603.06797` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Hsiang Hsu, Eric Lei, Chun-Fu Chen
+- **abs:** https://arxiv.org/abs/2603.06797 · **PDF:** https://arxiv.org/pdf/2603.06797v1
+
+## 7096. Email in the Era of LLMs
+
+- **ID:** `2603.20231` · **primary:** `cs.CY` · **Submitted:** 2026-03-06
+- **Authors:** Dang Nguyen, Harvey Yiyun Fu, Peter West, et al.
+- **abs:** https://arxiv.org/abs/2603.20231 · **PDF:** https://arxiv.org/pdf/2603.20231v1
+
+## 7097. Optimistic Policy Regularization
+
+- **ID:** `2603.06793` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Mai Pham, Vikrant Vaze, Peter Chin
+- **abs:** https://arxiv.org/abs/2603.06793 · **PDF:** https://arxiv.org/pdf/2603.06793v1
+
+## 7098. BEVLM: Distilling Semantic Knowledge from LLMs into Bird's-Eye View Representations
+
+- **ID:** `2603.06576` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Thomas Monninger, Shaoyuan Xie, Qi Alfred Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.06576 · **PDF:** https://arxiv.org/pdf/2603.06576v1
+
+## 7099. Fly360: Omnidirectional Obstacle Avoidance within Drone View
+
+- **ID:** `2603.06573` · **primary:** `cs.RO` · **Submitted:** 2026-03-06
+- **Authors:** Xiangkai Zhang, Dizhe Zhang, WenZhuo Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.06573 · **PDF:** https://arxiv.org/pdf/2603.06573v1
+
+## 7100. SCOPE: Scene-Contextualized Incremental Few-Shot 3D Segmentation
+
+- **ID:** `2603.06572` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Vishal Thengane, Zhaochong An, Tianjin Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.06572 · **PDF:** https://arxiv.org/pdf/2603.06572v2
+
+## 7101. SUREON: A Benchmark and Vision-Language-Model for Surgical Reasoning
+
+- **ID:** `2603.06570` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Alejandra Perez, Anita Rau, Lee White, et al.
+- **abs:** https://arxiv.org/abs/2603.06570 · **PDF:** https://arxiv.org/pdf/2603.06570v1
+
+## 7102. A recipe for scalable attention-based MLIPs: unlocking long-range accuracy with all-to-all node attention
+
+- **ID:** `2603.06567` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Eric Qu, Brandon M. Wood, Aditi S. Krishnapriyan, et al.
+- **abs:** https://arxiv.org/abs/2603.06567 · **PDF:** https://arxiv.org/pdf/2603.06567v1
+
+## 7103. Boosting deep Reinforcement Learning using pretraining with Logical Options
+
+- **ID:** `2603.06565` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Zihan Ye, Phil Chau, Raban Emunds, et al.
+- **abs:** https://arxiv.org/abs/2603.06565 · **PDF:** https://arxiv.org/pdf/2603.06565v1
+
+## 7104. Physics-Informed Diffusion Model for Generating Synthetic Extreme Rare Weather Events Data
+
+- **ID:** `2603.06782` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Marawan Yakout, Tannistha Maiti, Monira Majhabeen, et al.
+- **abs:** https://arxiv.org/abs/2603.06782 · **PDF:** https://arxiv.org/pdf/2603.06782v1
+
+## 7105. Causal Interpretation of Neural Network Computations with Contribution Decomposition
+
+- **ID:** `2603.06557` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Joshua Brendan Melander, Zaki Alaoui, Shenghua Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.06557 · **PDF:** https://arxiv.org/pdf/2603.06557v1
+
+## 7106. xaitimesynth: A Python Package for Evaluating Attribution Methods for Time Series with Synthetic Ground Truth
+
+- **ID:** `2603.06781` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Gregor Baer
+- **abs:** https://arxiv.org/abs/2603.06781 · **PDF:** https://arxiv.org/pdf/2603.06781v1
+
+## 7107. Hierarchical Industrial Demand Forecasting with Temporal and Uncertainty Explanations
+
+- **ID:** `2603.06555` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Harshavardhan Kamarthi, Shangqing Xu, Xinjie Tong, et al.
+- **abs:** https://arxiv.org/abs/2603.06555 · **PDF:** https://arxiv.org/pdf/2603.06555v1
+
+## 7108. KCLarity at SemEval-2026 Task 6: Encoder and Zero-Shot Approaches to Political Evasion Detection
+
+- **ID:** `2603.06552` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Archie Sage, Salvatore Greco
+- **abs:** https://arxiv.org/abs/2603.06552 · **PDF:** https://arxiv.org/pdf/2603.06552v1
+
+## 7109. LiveSense: A Real-Time Wi-Fi Sensing Platform for Range-Doppler on COTS Laptop
+
+- **ID:** `2603.06545` · **primary:** `eess.SP` · **Submitted:** 2026-03-06
+- **Authors:** Jessica Sanson, Rahul C. Shah, Maximilian Pinaroc, et al.
+- **abs:** https://arxiv.org/abs/2603.06545 · **PDF:** https://arxiv.org/pdf/2603.06545v1
+
+## 7110. RAMoEA-QA: Hierarchical Specialization for Robust Respiratory Audio Question Answering
+
+- **ID:** `2603.06542` · **primary:** `cs.SD` · **Submitted:** 2026-03-06
+- **Authors:** Gaia A. Bertolino, Yuwei Zhang, Tong Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.06542 · **PDF:** https://arxiv.org/pdf/2603.06542v1
+
+## 7111. SpatialMAGIC: A Hybrid Framework Integrating Graph Diffusion and Spatial Attention for Spatial Transcriptomics Imputation
+
+- **ID:** `2603.06780` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Sayeem Bin Zaman, Fahim Hafiz, Riasat Azim
+- **abs:** https://arxiv.org/abs/2603.06780 · **PDF:** https://arxiv.org/pdf/2603.06780v1
+
+## 7112. Artificial Intelligence for Detecting Fetal Orofacial Clefts and Advancing Medical Education
+
+- **ID:** `2603.06522` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Yuanji Zhang, Yuhao Huang, Haoran Dou, et al.
+- **abs:** https://arxiv.org/abs/2603.06522 · **PDF:** https://arxiv.org/pdf/2603.06522v1
+
+## 7113. HGT-Scheduler: Deep Reinforcement Learning for the Job Shop Scheduling Problem via Heterogeneous Graph Transformers
+
+- **ID:** `2603.06777` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Bulent Soykan
+- **abs:** https://arxiv.org/abs/2603.06777 · **PDF:** https://arxiv.org/pdf/2603.06777v1
+
+## 7114. Gauge Freedom and Metric Dependence in Neural Representation Spaces
+
+- **ID:** `2603.06774` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Jericho Cain
+- **abs:** https://arxiv.org/abs/2603.06774 · **PDF:** https://arxiv.org/pdf/2603.06774v1
+
+## 7115. When One Modality Rules Them All: Backdoor Modality Collapse in Multimodal Diffusion Models
+
+- **ID:** `2603.06508` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Qitong Wang, Haoran Dai, Haotian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.06508 · **PDF:** https://arxiv.org/pdf/2603.06508v1
+
+## 7116. Semantics-Aware Caching for Concept Learning
+
+- **ID:** `2603.06506` · **primary:** `stat.ML` · **Submitted:** 2026-03-06
+- **Authors:** Louis Mozart Kamdem Teyou, Caglar Demir, Axel-Cyrille Ngonga Ngomo
+- **abs:** https://arxiv.org/abs/2603.06506 · **PDF:** https://arxiv.org/pdf/2603.06506v2
+
+## 7117. Speak in Context: Multilingual ASR with Speech Context Alignment via Contrastive Learning
+
+- **ID:** `2603.06505` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Yuchen Zhang, Haralambos Mouratidis, Ravi Shekhar
+- **abs:** https://arxiv.org/abs/2603.06505 · **PDF:** https://arxiv.org/pdf/2603.06505v1
+
+## 7118. Beyond Rows to Reasoning: Agentic Retrieval for Multimodal Spreadsheet Understanding and Editing
+
+- **ID:** `2603.06503` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Anmol Gulati, Sahil Sen, Waqar Sarguroh, et al.
+- **abs:** https://arxiv.org/abs/2603.06503 · **PDF:** https://arxiv.org/pdf/2603.06503v1
+
+## 7119. COLD-Steer: Steering Large Language Models via In-Context One-step Learning Dynamics
+
+- **ID:** `2603.06495` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Kartik Sharma, Rakshit S. Trivedi
+- **abs:** https://arxiv.org/abs/2603.06495 · **PDF:** https://arxiv.org/pdf/2603.06495v1
+
+## 7120. NOBLE: Accelerating Transformers with Nonlinear Low-Rank Branches
+
+- **ID:** `2603.06492` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Ethan Smith
+- **abs:** https://arxiv.org/abs/2603.06492 · **PDF:** https://arxiv.org/pdf/2603.06492v1
+
+## 7121. Score Reversal Is Not Free for Quantum Diffusion Models
+
+- **ID:** `2603.06488` · **primary:** `quant-ph` · **Submitted:** 2026-03-06
+- **Authors:** Ammar Fayad
+- **abs:** https://arxiv.org/abs/2603.06488 · **PDF:** https://arxiv.org/pdf/2603.06488v3
+
+## 7122. PONTE: Personalized Orchestration for Natural Language Trustworthy Explanations
+
+- **ID:** `2603.06485` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Vittoria Vineis, Matteo Silvestri, Lorenzo Antonelli, et al.
+- **abs:** https://arxiv.org/abs/2603.06485 · **PDF:** https://arxiv.org/pdf/2603.06485v1
+
+## 7123. TerraLingua: Emergence and Analysis of Open-endedness in LLM Ecologies
+
+- **ID:** `2603.16910` · **primary:** `cs.MA` · **Submitted:** 2026-03-06
+- **Authors:** Giuseppe Paolo, Jamieson Warner, Hormoz Shahrzad, et al.
+- **abs:** https://arxiv.org/abs/2603.16910 · **PDF:** https://arxiv.org/pdf/2603.16910v1
+
+## 7124. Bayesian Additive Distribution Regression
+
+- **ID:** `2603.06462` · **primary:** `stat.ME` · **Submitted:** 2026-03-06
+- **Authors:** Antonio R. Linero, Soumyabrata Bose, Jared Murray
+- **abs:** https://arxiv.org/abs/2603.06462 · **PDF:** https://arxiv.org/pdf/2603.06462v1
+
+## 7125. Do Foundation Models Know Geometry? Probing Frozen Features for Continuous Physical Measurement
+
+- **ID:** `2603.06459` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Yakov Pyotr Shkolnikov
+- **abs:** https://arxiv.org/abs/2603.06459 · **PDF:** https://arxiv.org/pdf/2603.06459v1
+
+## 7126. Prosodic Boundary-Aware Streaming Generation for LLM-Based TTS with Streaming Text Input
+
+- **ID:** `2603.06444` · **primary:** `cs.SD` · **Submitted:** 2026-03-06
+- **Authors:** Changsong Liu, Tianrui Wang, Ye Ni, et al.
+- **abs:** https://arxiv.org/abs/2603.06444 · **PDF:** https://arxiv.org/pdf/2603.06444v1
+
+## 7127. Toward Generative Quantum Utility via Correlation-Complexity Map
+
+- **ID:** `2603.06440` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Chen-Yu Liu, Leonardo Placidi, Eric Brunner, et al.
+- **abs:** https://arxiv.org/abs/2603.06440 · **PDF:** https://arxiv.org/pdf/2603.06440v1
+
+## 7128. Certified and accurate computation of function space norms of deep neural networks
+
+- **ID:** `2603.06431` · **primary:** `math.NA` · **Submitted:** 2026-03-06
+- **Authors:** Johannes Gründler, Moritz Maibaum, Philipp Petersen
+- **abs:** https://arxiv.org/abs/2603.06431 · **PDF:** https://arxiv.org/pdf/2603.06431v1
+
+## 7129. Abductive Reasoning with Syllogistic Forms in Large Language Models
+
+- **ID:** `2603.06428` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Hirohiko Abe, Risako Ando, Takanobu Morishita Kentaro Ozeki, et al.
+- **abs:** https://arxiv.org/abs/2603.06428 · **PDF:** https://arxiv.org/pdf/2603.06428v1
+
+## 7130. Do 3D Large Language Models Really Understand 3D Spatial Relationships?
+
+- **ID:** `2603.23523` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Xianzheng Ma, Tao Sun, Shuai Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.23523 · **PDF:** https://arxiv.org/pdf/2603.23523v1
+
+## 7131. CLoPA: Continual Low Parameter Adaptation of Interactive Segmentation for Medical Image Annotation
+
+- **ID:** `2603.06426` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Parhom Esmaeili, Chayanin Tangwiriyasakul, Eli Gibson, et al.
+- **abs:** https://arxiv.org/abs/2603.06426 · **PDF:** https://arxiv.org/pdf/2603.06426v1
+
+## 7132. From Prompting to Preference Optimization: A Comparative Study of LLM-based Automated Essay Scoring
+
+- **ID:** `2603.06424` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Minh Hoang Nguyen, Vu Hoang Pham, Xuan Thanh Huynh, et al.
+- **abs:** https://arxiv.org/abs/2603.06424 · **PDF:** https://arxiv.org/pdf/2603.06424v1
+
+## 7133. Evaluation of Deontic Conditional Reasoning in Large Language Models: The Case of Wason's Selection Task
+
+- **ID:** `2603.06416` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Hirohiko Abe, Kentaro Ozeki, Risako Ando, et al.
+- **abs:** https://arxiv.org/abs/2603.06416 · **PDF:** https://arxiv.org/pdf/2603.06416v1
+
+## 7134. A Reference Architecture of Reinforcement Learning Frameworks
+
+- **ID:** `2603.06413` · **primary:** `cs.SE` · **Submitted:** 2026-03-06
+- **Authors:** Xiaoran Liu, Istvan David
+- **abs:** https://arxiv.org/abs/2603.06413 · **PDF:** https://arxiv.org/pdf/2603.06413v1
+
+## 7135. Physical Simulator In-the-Loop Video Generation
+
+- **ID:** `2603.06408` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Lin Geng Foo, Mark He Huang, Alexandros Lattas, et al.
+- **abs:** https://arxiv.org/abs/2603.06408 · **PDF:** https://arxiv.org/pdf/2603.06408v1
+
+## 7136. Adapter-Augmented Bandits for Online Multi-Constrained Multi-Modal Inference Scheduling
+
+- **ID:** `2603.06403` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Xianzhi Zhang, Yue Xu, Yinlin Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.06403 · **PDF:** https://arxiv.org/pdf/2603.06403v1
+
+## 7137. U6G XL-MIMO Radiomap Prediction: Multi-Config Dataset and Beam Map Approach
+
+- **ID:** `2603.06401` · **primary:** `eess.SP` · **Submitted:** 2026-03-06
+- **Authors:** Xiaojie Li, Yu Han, Zhizheng Lu, et al.
+- **abs:** https://arxiv.org/abs/2603.06401 · **PDF:** https://arxiv.org/pdf/2603.06401v1
+
+## 7138. Efficient, Property-Aligned Fan-Out Retrieval via RL-Compiled Diffusion
+
+- **ID:** `2603.06397` · **primary:** `cs.IR` · **Submitted:** 2026-03-06
+- **Authors:** Pengcheng Jiang, Judith Yue Li, Moonkyung Ryu, et al.
+- **abs:** https://arxiv.org/abs/2603.06397 · **PDF:** https://arxiv.org/pdf/2603.06397v1
+
+## 7139. Talk Freely, Execute Strictly: Schema-Gated Agentic AI for Flexible and Reproducible Scientific Workflows
+
+- **ID:** `2603.06394` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Joel Strickland, Arjun Vijeta, Chris Moores, et al.
+- **abs:** https://arxiv.org/abs/2603.06394 · **PDF:** https://arxiv.org/pdf/2603.06394v1
+
+## 7140. Prompt Group-Aware Training for Robust Text-Guided Nuclei Segmentation
+
+- **ID:** `2603.06384` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Yonghuang Wu, Zhenyang Liang, Wenwen Zeng, et al.
+- **abs:** https://arxiv.org/abs/2603.06384 · **PDF:** https://arxiv.org/pdf/2603.06384v1
+
+## 7141. Kinetic-based regularization: Learning spatial derivatives and PDE applications
+
+- **ID:** `2603.06380` · **primary:** `math.NA` · **Submitted:** 2026-03-06
+- **Authors:** Abhisek Ganguly, Santosh Ansumali, Sauro Succi
+- **abs:** https://arxiv.org/abs/2603.06380 · **PDF:** https://arxiv.org/pdf/2603.06380v1
+
+## 7142. Failure Detection in Chemical Processes Using Symbolic Machine Learning: A Case Study on Ethylene Oxidation
+
+- **ID:** `2603.06767` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Julien Amblard, Niklas Groll, Matthew Tait, et al.
+- **abs:** https://arxiv.org/abs/2603.06767 · **PDF:** https://arxiv.org/pdf/2603.06767v3
+
+## 7143. Adaptive Lipschitz-Free Conditional Gradient Methods for Stochastic Composite Nonconvex Optimization
+
+- **ID:** `2603.06369` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Ganzhao Yuan
+- **abs:** https://arxiv.org/abs/2603.06369 · **PDF:** https://arxiv.org/pdf/2603.06369v1
+
+## 7144. Qworld: Question-Specific Evaluation Criteria for LLMs
+
+- **ID:** `2603.23522` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Shanghua Gao, Yuchang Su, Pengwei Sui, et al.
+- **abs:** https://arxiv.org/abs/2603.23522 · **PDF:** https://arxiv.org/pdf/2603.23522v1
+
+## 7145. ESAA-Security: An Event-Sourced, Verifiable Architecture for Agent-Assisted Security Audits of AI-Generated Code
+
+- **ID:** `2603.06365` · **primary:** `cs.CR` · **Submitted:** 2026-03-06
+- **Authors:** Elzo Brito dos Santos Filho
+- **abs:** https://arxiv.org/abs/2603.06365 · **PDF:** https://arxiv.org/pdf/2603.06365v1
+
+## 7146. CLAIRE: Compressed Latent Autoencoder for Industrial Representation and Evaluation -- A Deep Learning Framework for Smart Manufacturing
+
+- **ID:** `2603.06361` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Mohammadhossein Ghahramani, Mengchu Zhou
+- **abs:** https://arxiv.org/abs/2603.06361 · **PDF:** https://arxiv.org/pdf/2603.06361v1
+
+## 7147. Tiny, Hardware-Independent, Compression-based Classification
+
+- **ID:** `2603.06359` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Charles Meyers, Aaron MacSween, Erik Elmroth, et al.
+- **abs:** https://arxiv.org/abs/2603.06359 · **PDF:** https://arxiv.org/pdf/2603.06359v1
+
+## 7148. Frequency-Separable Hamiltonian Neural Network for Multi-Timescale Dynamics
+
+- **ID:** `2603.06354` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Yaojun Li, Yulong Yang, Christine Allen-Blanchette
+- **abs:** https://arxiv.org/abs/2603.06354 · **PDF:** https://arxiv.org/pdf/2603.06354v2
+
+## 7149. Chitrakshara: A Large Multilingual Multimodal Dataset for Indian languages
+
+- **ID:** `2603.23521` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Shaharukh Khan, Ali Faraz, Abhinav Ravi, et al.
+- **abs:** https://arxiv.org/abs/2603.23521 · **PDF:** https://arxiv.org/pdf/2603.23521v1
+
+## 7150. Dynamic Chunking Diffusion Transformer
+
+- **ID:** `2603.06351` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Akash Haridas, Utkarsh Saxena, Parsa Ashrafi Fashi, et al.
+- **abs:** https://arxiv.org/abs/2603.06351 · **PDF:** https://arxiv.org/pdf/2603.06351v1
+
+## 7151. MoEless: Efficient MoE LLM Serving via Serverless Computing
+
+- **ID:** `2603.06350` · **primary:** `cs.DC` · **Submitted:** 2026-03-06
+- **Authors:** Hanfei Yu, Bei Ouyang, Shwai He, et al.
+- **abs:** https://arxiv.org/abs/2603.06350 · **PDF:** https://arxiv.org/pdf/2603.06350v1
+
+## 7152. Transparent AI for Mathematics: Transformer-Based Large Language Models for Mathematical Entity Relationship Extraction with XAI
+
+- **ID:** `2603.06348` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Tanjim Taharat Aurpa
+- **abs:** https://arxiv.org/abs/2603.06348 · **PDF:** https://arxiv.org/pdf/2603.06348v1
+
+## 7153. K-MaT: Knowledge-Anchored Manifold Transport for Cross-Modal Prompt Learning in Medical Imaging
+
+- **ID:** `2603.06340` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Jiajun Zeng, Shadi Albarqouni
+- **abs:** https://arxiv.org/abs/2603.06340 · **PDF:** https://arxiv.org/pdf/2603.06340v1
+
+## 7154. AI End-to-End Radiation Treatment Planning Under One Second
+
+- **ID:** `2603.06338` · **primary:** `eess.IV` · **Submitted:** 2026-03-06
+- **Authors:** Simon Arberet, Riqiang Gao, Martin Kraus, et al.
+- **abs:** https://arxiv.org/abs/2603.06338 · **PDF:** https://arxiv.org/pdf/2603.06338v1
+
+## 7155. SAHOO: Safeguarded Alignment for High-Order Optimization Objectives in Recursive Self-Improvement
+
+- **ID:** `2603.06333` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Subramanyam Sahoo, Aman Chadha, Vinija Jain, et al.
+- **abs:** https://arxiv.org/abs/2603.06333 · **PDF:** https://arxiv.org/pdf/2603.06333v1
+
+## 7156. Structured Exploration vs. Generative Flexibility: A Field Study Comparing Bandit and LLM Architectures for Personalised Health Behaviour Interventions
+
+- **ID:** `2603.06330` · **primary:** `cs.HC` · **Submitted:** 2026-03-06
+- **Authors:** Dominik P. Hofer, Haochen Song, Rania Islambouli, et al.
+- **abs:** https://arxiv.org/abs/2603.06330 · **PDF:** https://arxiv.org/pdf/2603.06330v1
+
+## 7157. Metalearning traffic assignment for network disruptions with graph convolutional neural networks
+
+- **ID:** `2603.06763` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Serio Agriesti, Guido Cantelmo, Francisco Camara Pereira
+- **abs:** https://arxiv.org/abs/2603.06763 · **PDF:** https://arxiv.org/pdf/2603.06763v1
+
+## 7158. Prediction of Steady-State Flow through Porous Media Using Machine Learning Models
+
+- **ID:** `2603.06762` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-03-06
+- **Authors:** Jinhong Wang, Matei C. Ignuta-Ciuncanu, Ricardo F. Martinez-Botas, et al.
+- **abs:** https://arxiv.org/abs/2603.06762 · **PDF:** https://arxiv.org/pdf/2603.06762v1
+
+## 7159. Diversity-Aware Adaptive Collocation for Physics-Informed Neural Networks via Sparse QUBO Optimization and Hybrid Coresets
+
+- **ID:** `2603.06761` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Hadi Salloum, Maximilian Mifsud Bonici, Sinan Ibrahim, et al.
+- **abs:** https://arxiv.org/abs/2603.06761 · **PDF:** https://arxiv.org/pdf/2603.06761v1
+
+## 7160. The Art That Poses Back: Assessing AI Pastiches after Contemporary Artworks
+
+- **ID:** `2603.06324` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Anca Dinu, Andreiana Mihail, Andra-Maria Florescu, et al.
+- **abs:** https://arxiv.org/abs/2603.06324 · **PDF:** https://arxiv.org/pdf/2603.06324v1
+
+## 7161. From Entropy to Calibrated Uncertainty: Training Language Models to Reason About Uncertainty
+
+- **ID:** `2603.06317` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Azza Jenane, Nassim Walha, Lukas Kuhn, et al.
+- **abs:** https://arxiv.org/abs/2603.06317 · **PDF:** https://arxiv.org/pdf/2603.06317v1
+
+## 7162. Gradient-based Nested Co-Design of Aerodynamic Shape and Control for Winged Robots
+
+- **ID:** `2603.06760` · **primary:** `cs.RO` · **Submitted:** 2026-03-06
+- **Authors:** Daniele Affinita, Mingda Xu, Benoît Valentin Gherardi, et al.
+- **abs:** https://arxiv.org/abs/2603.06760 · **PDF:** https://arxiv.org/pdf/2603.06760v1
+
+## 7163. Continual Adaptation for Pacific Indigenous Speech Recognition
+
+- **ID:** `2603.06310` · **primary:** `eess.AS` · **Submitted:** 2026-03-06
+- **Authors:** Yang Xiao, Aso Mahmudi, Nick Thieberger, et al.
+- **abs:** https://arxiv.org/abs/2603.06310 · **PDF:** https://arxiv.org/pdf/2603.06310v1
+
+## 7164. Polarized Direct Cross-Attention Message Passing in GNNs for Machinery Fault Diagnosis
+
+- **ID:** `2603.06303` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Zongyu Shi, Laibin Zhang, Maoyin Chen
+- **abs:** https://arxiv.org/abs/2603.06303 · **PDF:** https://arxiv.org/pdf/2603.06303v1
+
+## 7165. DEX-AR: A Dynamic Explainability Method for Autoregressive Vision-Language Models
+
+- **ID:** `2603.06302` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Walid Bousselham, Angie Boggust, Hendrik Strobelt, et al.
+- **abs:** https://arxiv.org/abs/2603.06302 · **PDF:** https://arxiv.org/pdf/2603.06302v1
+
+## 7166. 3D CBCT Artefact Removal Using Perpendicular Score-Based Diffusion Models
+
+- **ID:** `2603.06300` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Susanne Schaub, Florentin Bieder, Matheus L. Oliveira, et al.
+- **abs:** https://arxiv.org/abs/2603.06300 · **PDF:** https://arxiv.org/pdf/2603.06300v1
+
+## 7167. From Physician Expertise to Clinical Agents: Preserving, Standardizing, and Scaling Physicians' Medical Expertise with Lightweight LLM
+
+- **ID:** `2603.23520` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Chanyong Luo, Jirui Dai, Zhendong Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.23520 · **PDF:** https://arxiv.org/pdf/2603.23520v1
+
+## 7168. Information-Theoretic Constraints for Continual Vision-Language-Action Alignment
+
+- **ID:** `2603.13335` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Libang Zhao, Qixin Zeng, Hongyin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.13335 · **PDF:** https://arxiv.org/pdf/2603.13335v1
+
+## 7169. Enhanced Protein Intrinsic Disorder Prediction Through Dual-View Multiscale Features and Multi-objective Evolutionary Algorithm
+
+- **ID:** `2603.06292` · **primary:** `cs.NE` · **Submitted:** 2026-03-06
+- **Authors:** Shaokuan Wang, Pengshan Cui, Yining Qian, et al.
+- **abs:** https://arxiv.org/abs/2603.06292 · **PDF:** https://arxiv.org/pdf/2603.06292v1
+
+## 7170. The EpisTwin: A Knowledge Graph-Grounded Neuro-Symbolic Architecture for Personal AI
+
+- **ID:** `2603.06290` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Giovanni Servedio, Potito Aghilar, Alessio Mattiace, et al.
+- **abs:** https://arxiv.org/abs/2603.06290 · **PDF:** https://arxiv.org/pdf/2603.06290v1
+
+## 7171. Learning Where the Physics Is: Probabilistic Adaptive Sampling for Stiff PDEs
+
+- **ID:** `2603.06287` · **primary:** `cs.CE` · **Submitted:** 2026-03-06
+- **Authors:** Akshay Govind Srinivasan, Balaji Srinivasan
+- **abs:** https://arxiv.org/abs/2603.06287 · **PDF:** https://arxiv.org/pdf/2603.06287v1
+
+## 7172. Enhancing SHAP Explainability for Diagnostic and Prognostic ML Models in Alzheimer Disease
+
+- **ID:** `2603.06758` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Pablo Guillén, Enrique Frias-Martinez
+- **abs:** https://arxiv.org/abs/2603.06758 · **PDF:** https://arxiv.org/pdf/2603.06758v1
+
+## 7173. Learning Unbiased Cluster Descriptors for Interpretable Imbalanced Concept Drift Detection
+
+- **ID:** `2603.06757` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Yiqun Zhang, Zhanpei Huang, Mingjie Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.06757 · **PDF:** https://arxiv.org/pdf/2603.06757v1
+
+## 7174. Artificial Intelligence for Climate Adaptation: Reinforcement Learning for Climate Change-Resilient Transport
+
+- **ID:** `2603.06278` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Miguel Costa, Arthur Vandervoort, Carolin Schmidt, et al.
+- **abs:** https://arxiv.org/abs/2603.06278 · **PDF:** https://arxiv.org/pdf/2603.06278v1
+
+## 7175. Stem: Rethinking Causal Information Flow in Sparse Attention
+
+- **ID:** `2603.06274` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Lin Niu, Xin Luo, Linchuan Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.06274 · **PDF:** https://arxiv.org/pdf/2603.06274v1
+
+## 7176. Looking Through Glass Box
+
+- **ID:** `2603.06272` · **primary:** `cs.NE` · **Submitted:** 2026-03-06
+- **Authors:** Alexis Kafantaris
+- **abs:** https://arxiv.org/abs/2603.06272 · **PDF:** https://arxiv.org/pdf/2603.06272v1
+
+## 7177. Agentic retrieval-augmented reasoning reshapes collective reliability under model variability in radiology question answering
+
+- **ID:** `2603.06271` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Mina Farajiamiri, Jeta Sopa, Saba Afza, et al.
+- **abs:** https://arxiv.org/abs/2603.06271 · **PDF:** https://arxiv.org/pdf/2603.06271v1
+
+## 7178. HiPP-Prune: Hierarchical Preference-Conditioned Structured Pruning for Vision-Language Models
+
+- **ID:** `2603.06270` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Lincen Bai, Hedi Tabia, Raul Santos-Rodriguez
+- **abs:** https://arxiv.org/abs/2603.06270 · **PDF:** https://arxiv.org/pdf/2603.06270v1
+
+## 7179. Mind the Gap: Pitfalls of LLM Alignment with Asian Public Opinion
+
+- **ID:** `2603.06264` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Hari Shankar, Vedanta S P, Sriharini Margapuri, et al.
+- **abs:** https://arxiv.org/abs/2603.06264 · **PDF:** https://arxiv.org/pdf/2603.06264v2
+
+## 7180. Learning to Solve Orienteering Problem with Time Windows and Variable Profits
+
+- **ID:** `2603.06260` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Songqun Gao, Zanxi Ruan, Patrick Floor, et al.
+- **abs:** https://arxiv.org/abs/2603.06260 · **PDF:** https://arxiv.org/pdf/2603.06260v1
+
+## 7181. Robust support vector model based on bounded asymmetric elastic net loss for binary classification
+
+- **ID:** `2603.06257` · **primary:** `stat.ML` · **Submitted:** 2026-03-06
+- **Authors:** Haiyan Du, Hu Yang
+- **abs:** https://arxiv.org/abs/2603.06257 · **PDF:** https://arxiv.org/pdf/2603.06257v1
+
+## 7182. GazeMoE: Perception of Gaze Target with Mixture-of-Experts
+
+- **ID:** `2603.06256` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Zhuangzhuang Dai, Zhongxi Lu, Vincent G. Zakka, et al.
+- **abs:** https://arxiv.org/abs/2603.06256 · **PDF:** https://arxiv.org/pdf/2603.06256v1
+
+## 7183. Implementation of Quantum Implicit Neural Representation in Deterministic and Probabilistic Autoencoders for Image Reconstruction/Generation Tasks
+
+- **ID:** `2603.06755` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Saadet Müzehher Eren
+- **abs:** https://arxiv.org/abs/2603.06755 · **PDF:** https://arxiv.org/pdf/2603.06755v2
+
+## 7184. Synthetic Monitoring Environments for Reinforcement Learning
+
+- **ID:** `2603.06252` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Leonard Pleiss, Carolin Schmidt, Maximilian Schiffer
+- **abs:** https://arxiv.org/abs/2603.06252 · **PDF:** https://arxiv.org/pdf/2603.06252v1
+
+## 7185. SPPCSO: Adaptive Penalized Estimation Method for High-Dimensional Correlated Data
+
+- **ID:** `2603.06251` · **primary:** `stat.ML` · **Submitted:** 2026-03-06
+- **Authors:** Ying Hu, Hu Yang
+- **abs:** https://arxiv.org/abs/2603.06251 · **PDF:** https://arxiv.org/pdf/2603.06251v1
+
+## 7186. Gradient Flow Polarizes Softmax Outputs towards Low-Entropy Solutions
+
+- **ID:** `2603.06248` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Aditya Varre, Mark Rofin, Nicolas Flammarion
+- **abs:** https://arxiv.org/abs/2603.06248 · **PDF:** https://arxiv.org/pdf/2603.06248v1
+
+## 7187. DC-Merge: Improving Model Merging with Directional Consistency
+
+- **ID:** `2603.06242` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Han-Chen Zhang, Zi-Hao Zhou, Mao-Lin Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.06242 · **PDF:** https://arxiv.org/pdf/2603.06242v2
+
+## 7188. TaPD: Temporal-adaptive Progressive Distillation for Observation-Adaptive Trajectory Forecasting in Autonomous Driving
+
+- **ID:** `2603.06231` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Mingyu Fan, Yi Liu, Hao Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.06231 · **PDF:** https://arxiv.org/pdf/2603.06231v1
+
+## 7189. FedSCS-XGB -- Federated Server-centric surrogate XGBoost for continual health monitoring
+
+- **ID:** `2603.06224` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Felix Walger, Mehdi Ejtehadi, Anke Schmeink, et al.
+- **abs:** https://arxiv.org/abs/2603.06224 · **PDF:** https://arxiv.org/pdf/2603.06224v1
+
+## 7190. SPOT: Span-level Pause-of-Thought for Efficient and Interpretable Latent Reasoning in Large Language Models
+
+- **ID:** `2603.06222` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Yunlong Chu, Minglai Shao, Yuhang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.06222 · **PDF:** https://arxiv.org/pdf/2603.06222v1
+
+## 7191. Conversational Demand Response: Bidirectional Aggregator-Prosumer Coordination through Agentic AI
+
+- **ID:** `2603.06217` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Reda El Makroum, Sebastian Zwickl-Bernhard, Lukas Kranzl, et al.
+- **abs:** https://arxiv.org/abs/2603.06217 · **PDF:** https://arxiv.org/pdf/2603.06217v1
+
+## 7192. Cut to the Chase: Training-free Multimodal Summarization via Chain-of-Events
+
+- **ID:** `2603.06213` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Xiaoxing You, Qiang Huang, Lingyu Li, et al.
+- **abs:** https://arxiv.org/abs/2603.06213 · **PDF:** https://arxiv.org/pdf/2603.06213v1
+
+## 7193. Topological descriptors of foot clearance gait dynamics improve differential diagnosis of Parkinsonism
+
+- **ID:** `2603.06212` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Jhonathan Barrios, Wolfram Erlhagen, Miguel F. Gago, et al.
+- **abs:** https://arxiv.org/abs/2603.06212 · **PDF:** https://arxiv.org/pdf/2603.06212v1
+
+## 7194. MedMT-Bench: Can LLMs Memorize and Understand Long Multi-Turn Conversations in Medical Scenarios?
+
+- **ID:** `2603.23519` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Lin Yang, Yuancheng Yang, Xu Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.23519 · **PDF:** https://arxiv.org/pdf/2603.23519v1
+
+## 7195. FlashPrefill: Instantaneous Pattern Discovery and Thresholding for Ultra-Fast Long-Context Prefilling
+
+- **ID:** `2603.06199` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Qihang Fan, Huaibo Huang, Zhiying Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.06199 · **PDF:** https://arxiv.org/pdf/2603.06199v1
+
+## 7196. Latent Autoencoder Ensemble Kalman Filter for Data assimilation
+
+- **ID:** `2603.06752` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Xin T. Tong, Yanyan Wang, Liang Yan
+- **abs:** https://arxiv.org/abs/2603.06752 · **PDF:** https://arxiv.org/pdf/2603.06752v1
+
+## 7197. LIT-RAGBench: Benchmarking Generator Capabilities of Large Language Models in Retrieval-Augmented Generation
+
+- **ID:** `2603.06198` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Koki Itai, Shunichi Hasegawa, Yuta Yamamoto, et al.
+- **abs:** https://arxiv.org/abs/2603.06198 · **PDF:** https://arxiv.org/pdf/2603.06198v1
+
+## 7198. Wisdom of the AI Crowd (AI-CROWD) for Ground Truth Approximation in Content Analysis: A Research Protocol & Validation Using Eleven Large Language Models
+
+- **ID:** `2603.06197` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Luis de-Marcos, Manuel Goyanes, Adrián Domínguez-Díaz
+- **abs:** https://arxiv.org/abs/2603.06197 · **PDF:** https://arxiv.org/pdf/2603.06197v1
+
+## 7199. MAPO: Mixed Advantage Policy Optimization for Long-Horizon Multi-Turn Dialogue
+
+- **ID:** `2603.06194` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Naifan Zhang, Ruihan Sun, Jinwei Su, et al.
+- **abs:** https://arxiv.org/abs/2603.06194 · **PDF:** https://arxiv.org/pdf/2603.06194v1
+
+## 7200. Whisper-CD: Accurate Long-Form Speech Recognition using Multi-Negative Contrastive Decoding
+
+- **ID:** `2603.06193` · **primary:** `cs.SD` · **Submitted:** 2026-03-06
+- **Authors:** Hoseong Ahn, Jeongyun Chae, Yoonji Park, et al.
+- **abs:** https://arxiv.org/abs/2603.06193 · **PDF:** https://arxiv.org/pdf/2603.06193v1
+
+## 7201. Random Quadratic Form on a Sphere: Synchronization by Common Noise
+
+- **ID:** `2603.06187` · **primary:** `math.PR` · **Submitted:** 2026-03-06
+- **Authors:** Maximilian Engel, Anna Shalova
+- **abs:** https://arxiv.org/abs/2603.06187 · **PDF:** https://arxiv.org/pdf/2603.06187v1
+
+## 7202. CRIMSON: A Clinically-Grounded LLM-Based Metric for Generative Radiology Report Evaluation
+
+- **ID:** `2603.06183` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Mohammed Baharoon, Thibault Heintz, Siavash Raissi, et al.
+- **abs:** https://arxiv.org/abs/2603.06183 · **PDF:** https://arxiv.org/pdf/2603.06183v2
+
+## 7203. Contrastive-to-Self-Supervised: A Two-Stage Framework for Script Similarity Learning
+
+- **ID:** `2603.06180` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Claire Roman, Philippe Meyer
+- **abs:** https://arxiv.org/abs/2603.06180 · **PDF:** https://arxiv.org/pdf/2603.06180v1
+
+## 7204. XMACNet: An Explainable Lightweight Attention based CNN with Multi Modal Fusion for Chili Disease Classification
+
+- **ID:** `2603.06750` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Tapon Kumer Ray, Rajkumar Y, Shalini R, et al.
+- **abs:** https://arxiv.org/abs/2603.06750 · **PDF:** https://arxiv.org/pdf/2603.06750v1
+
+## 7205. Robotic Foundation Models for Industrial Control: A Comprehensive Survey and Readiness Assessment Framework
+
+- **ID:** `2603.06749` · **primary:** `cs.RO` · **Submitted:** 2026-03-06
+- **Authors:** David Kube, Simon Hadwiger, Tobias Meisen
+- **abs:** https://arxiv.org/abs/2603.06749 · **PDF:** https://arxiv.org/pdf/2603.06749v1
+
+## 7206. Reflective Flow Sampling Enhancement
+
+- **ID:** `2603.06165` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Zikai Zhou, Muyao Wang, Shitong Shao, et al.
+- **abs:** https://arxiv.org/abs/2603.06165 · **PDF:** https://arxiv.org/pdf/2603.06165v1
+
+## 7207. Do Compact SSL Backbones Matter for Audio Deepfake Detection? A Controlled Study with RAPTOR
+
+- **ID:** `2603.06164` · **primary:** `cs.SD` · **Submitted:** 2026-03-06
+- **Authors:** Ajinkya Kulkarni, Sandipana Dowerah, Atharva Kulkarni, et al.
+- **abs:** https://arxiv.org/abs/2603.06164 · **PDF:** https://arxiv.org/pdf/2603.06164v1
+
+## 7208. Chaotic Oscillator Networks for Classification Tasks
+
+- **ID:** `2603.16909` · **primary:** `nlin.CD` · **Submitted:** 2026-03-06
+- **Authors:** Toni Ivas, Georgios Violakis, Roland Richter, et al.
+- **abs:** https://arxiv.org/abs/2603.16909 · **PDF:** https://arxiv.org/pdf/2603.16909v1
+
+## 7209. Efficient Vector Search in the Wild: One Model for Multi-K Queries
+
+- **ID:** `2603.06159` · **primary:** `cs.DB` · **Submitted:** 2026-03-06
+- **Authors:** Yifan Peng, Jiafei Fan, Xingda Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.06159 · **PDF:** https://arxiv.org/pdf/2603.06159v1
+
+## 7210. Ensemble Graph Neural Networks for Probabilistic Sea Surface Temperature Forecasting via Input Perturbations
+
+- **ID:** `2603.06153` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Alejandro J. González-Santana, Giovanny A. Cuervo-Londoño, Javier Sánchez
+- **abs:** https://arxiv.org/abs/2603.06153 · **PDF:** https://arxiv.org/pdf/2603.06153v1
+
+## 7211. VLM-RobustBench: A Comprehensive Benchmark for Robustness of Vision-Language Models
+
+- **ID:** `2603.06148` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Rohit Saxena, Alessandro Suglia, Pasquale Minervini
+- **abs:** https://arxiv.org/abs/2603.06148 · **PDF:** https://arxiv.org/pdf/2603.06148v1
+
+## 7212. Property-driven Protein Inverse Folding With Multi-Objective Preference Alignment
+
+- **ID:** `2603.06748` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Xiaoyang Hou, Junqi Liu, Chence Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.06748 · **PDF:** https://arxiv.org/pdf/2603.06748v2
+
+## 7213. Predictive Coding Graphs are a Superset of Feedforward Neural Networks
+
+- **ID:** `2603.06142` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Björn van Zwol
+- **abs:** https://arxiv.org/abs/2603.06142 · **PDF:** https://arxiv.org/pdf/2603.06142v1
+
+## 7214. Place-it-R1: Unlocking Environment-aware Reasoning Potential of MLLM for Video Object Insertion
+
+- **ID:** `2603.06140` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Bohai Gu, Taiyi Wu, Dazhao Du, et al.
+- **abs:** https://arxiv.org/abs/2603.06140 · **PDF:** https://arxiv.org/pdf/2603.06140v1
+
+## 7215. Partial Policy Gradients for RL in LLMs
+
+- **ID:** `2603.06138` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Puneet Mathur, Branislav Kveton, Subhojyoti Mukherjee, et al.
+- **abs:** https://arxiv.org/abs/2603.06138 · **PDF:** https://arxiv.org/pdf/2603.06138v1
+
+## 7216. A Causal Graph Approach to Oppositional Narrative Analysis
+
+- **ID:** `2603.06135` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Diego Revilla, Martin Fernandez-de-Retana, Lingfeng Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.06135 · **PDF:** https://arxiv.org/pdf/2603.06135v2
+
+## 7217. DQE: A Semantic-Aware Evaluation Metric for Time Series Anomaly Detection
+
+- **ID:** `2603.06131` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Yuewei Li, Dalin Zhang, Huan Li, et al.
+- **abs:** https://arxiv.org/abs/2603.06131 · **PDF:** https://arxiv.org/pdf/2603.06131v1
+
+## 7218. A Hazard-Informed Data Pipeline for Robotics Physical Safety
+
+- **ID:** `2603.06130` · **primary:** `cs.RO` · **Submitted:** 2026-03-06
+- **Authors:** Alexei Odinokov, Rostislav Yavorskiy
+- **abs:** https://arxiv.org/abs/2603.06130 · **PDF:** https://arxiv.org/pdf/2603.06130v1
+
+## 7219. Diffusion Language Models Are Natively Length-Aware
+
+- **ID:** `2603.06123` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Vittorio Rossi, Giacomo Cirò, Davide Beltrame, et al.
+- **abs:** https://arxiv.org/abs/2603.06123 · **PDF:** https://arxiv.org/pdf/2603.06123v1
+
+## 7220. Dynamic Momentum Recalibration in Online Gradient Learning
+
+- **ID:** `2603.06120` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Zhipeng Yao, Rui Yu, Guisong Chang, et al.
+- **abs:** https://arxiv.org/abs/2603.06120 · **PDF:** https://arxiv.org/pdf/2603.06120v1
+
+## 7221. Clinically Meaningful Explainability for NeuroAI: An ethical, technical, and clinical perspective
+
+- **ID:** `2603.18028` · **primary:** `cs.CY` · **Submitted:** 2026-03-06
+- **Authors:** Laura Schopp, Ambra DImperio, Jalal Etesami, et al.
+- **abs:** https://arxiv.org/abs/2603.18028 · **PDF:** https://arxiv.org/pdf/2603.18028v1
+
+## 7222. Cluster-R1: Large Reasoning Models Are Instruction-following Clustering Agents
+
+- **ID:** `2603.23518` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Peijun Qing, Puneet Mathur, Nedim Lipka, et al.
+- **abs:** https://arxiv.org/abs/2603.23518 · **PDF:** https://arxiv.org/pdf/2603.23518v1
+
+## 7223. Making Implicit Premises Explicit in Logical Understanding of Enthymemes
+
+- **ID:** `2603.06114` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Xuyao Feng, Anthony Hunter
+- **abs:** https://arxiv.org/abs/2603.06114 · **PDF:** https://arxiv.org/pdf/2603.06114v1
+
+## 7224. Latent Diffusion-Based 3D Molecular Recovery from Vibrational Spectra
+
+- **ID:** `2603.06113` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Wenjin Wu, Aleš Leonardis, Linjiang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.06113 · **PDF:** https://arxiv.org/pdf/2603.06113v1
+
+## 7225. ButterflyViT: 354$\times$ Expert Compression for Edge Vision Transformers
+
+- **ID:** `2603.06746` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Aryan Karmore
+- **abs:** https://arxiv.org/abs/2603.06746 · **PDF:** https://arxiv.org/pdf/2603.06746v1
+
+## 7226. Enhancing Instruction Following of LLMs via Activation Steering with Dynamic Rejection
+
+- **ID:** `2603.06745` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Minjae Kang, Jaehyung Kim
+- **abs:** https://arxiv.org/abs/2603.06745 · **PDF:** https://arxiv.org/pdf/2603.06745v1
+
+## 7227. DeepSight: Bridging Depth Maps and Language with a Depth-Driven Multimodal Model
+
+- **ID:** `2603.06090` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Hao Yang, Hongbo Zhang, Yanyan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.06090 · **PDF:** https://arxiv.org/pdf/2603.06090v1
+
+## 7228. Experiences Build Characters: The Linguistic Origins and Functional Impact of LLM Personality
+
+- **ID:** `2603.06088` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Xi Wang, Mengdie Zhuang, Jiqun Liu
+- **abs:** https://arxiv.org/abs/2603.06088 · **PDF:** https://arxiv.org/pdf/2603.06088v1
+
+## 7229. Offline Materials Optimization with CliqueFlowmer
+
+- **ID:** `2603.06082` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Jakub Grudzien Kuba, Benjamin Kurt Miller, Sergey Levine, et al.
+- **abs:** https://arxiv.org/abs/2603.06082 · **PDF:** https://arxiv.org/pdf/2603.06082v3
+
+## 7230. StreamVoiceAnon+: Emotion-Preserving Streaming Speaker Anonymization via Frame-Level Acoustic Distillation
+
+- **ID:** `2603.06079` · **primary:** `eess.AS` · **Submitted:** 2026-03-06
+- **Authors:** Nikita Kuzmin, Kong Aik Lee, Eng Siong Chng
+- **abs:** https://arxiv.org/abs/2603.06079 · **PDF:** https://arxiv.org/pdf/2603.06079v1
+
+## 7231. Lifelong Embodied Navigation Learning
+
+- **ID:** `2603.06073` · **primary:** `cs.RO` · **Submitted:** 2026-03-06
+- **Authors:** Xudong Wang, Jiahua Dong, Baichen Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.06073 · **PDF:** https://arxiv.org/pdf/2603.06073v1
+
+## 7232. Text-Driven Emotionally Continuous Talking Face Generation
+
+- **ID:** `2603.06071` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Hao Yang, Yanyan Zhao, Tian Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.06071 · **PDF:** https://arxiv.org/pdf/2603.06071v1
+
+## 7233. Aggregative Semantics for Quantitative Bipolar Argumentation Frameworks
+
+- **ID:** `2603.06067` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Yann Munro, Isabelle Bloch, Marie-Jeanne Lesot
+- **abs:** https://arxiv.org/abs/2603.06067 · **PDF:** https://arxiv.org/pdf/2603.06067v1
+
+## 7234. Evaluating Austrian A-Level German Essays with Large Language Models for Automated Essay Scoring
+
+- **ID:** `2603.06066` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Jonas Kubesch, Lena Huber, Clemens Havas
+- **abs:** https://arxiv.org/abs/2603.06066 · **PDF:** https://arxiv.org/pdf/2603.06066v1
+
+## 7235. Agentic LLM Planning via Step-Wise PDDL Simulation: An Empirical Characterisation
+
+- **ID:** `2603.06064` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Kai Göbel, Pierrick Lorang, Patrik Zips, et al.
+- **abs:** https://arxiv.org/abs/2603.06064 · **PDF:** https://arxiv.org/pdf/2603.06064v1
+
+## 7236. TempoSyncDiff: Distilled Temporally-Consistent Diffusion for Low-Latency Audio-Driven Talking Head Generation
+
+- **ID:** `2603.06057` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Soumya Mazumdar, Vineet Kumar Rakesh
+- **abs:** https://arxiv.org/abs/2603.06057 · **PDF:** https://arxiv.org/pdf/2603.06057v1
+
+## 7237. Probing Visual Concepts in Lightweight Vision-Language Models for Automated Driving
+
+- **ID:** `2603.06054` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Nikos Theodoridis, Reenu Mohandas, Ganesh Sistu, et al.
+- **abs:** https://arxiv.org/abs/2603.06054 · **PDF:** https://arxiv.org/pdf/2603.06054v1
+
+## 7238. Stabilizing Reinforcement Learning for Diffusion Language Models
+
+- **ID:** `2603.06743` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Jianyuan Zhong, Kaibo Wang, Ding Ding, et al.
+- **abs:** https://arxiv.org/abs/2603.06743 · **PDF:** https://arxiv.org/pdf/2603.06743v1
+
+## 7239. Improved Constrained Generation by Bridging Pretrained Generative Models
+
+- **ID:** `2603.06742` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Xiaoxuan Liang, Saeid Naderiparizi, Yunpeng Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.06742 · **PDF:** https://arxiv.org/pdf/2603.06742v1
+
+## 7240. Heterogeneous Decentralized Diffusion Models
+
+- **ID:** `2603.06741` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Zhiying Jiang, Raihan Seraj, Marcos Villagra, et al.
+- **abs:** https://arxiv.org/abs/2603.06741 · **PDF:** https://arxiv.org/pdf/2603.06741v1
+
+## 7241. ViroGym: Realistic Large-Scale Benchmarks for Evaluating Viral Proteins
+
+- **ID:** `2603.06740` · **primary:** `q-bio.QM` · **Submitted:** 2026-03-06
+- **Authors:** Yichen Zhou, Jonathan Golob, Amir Karimi, et al.
+- **abs:** https://arxiv.org/abs/2603.06740 · **PDF:** https://arxiv.org/pdf/2603.06740v1
+
+## 7242. ResearchEnvBench: Benchmarking Agents on Environment Synthesis for Research Code Execution
+
+- **ID:** `2603.06739` · **primary:** `cs.SE` · **Submitted:** 2026-03-06
+- **Authors:** Yubang Wang, Chenxi Zhang, Bowen Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.06739 · **PDF:** https://arxiv.org/pdf/2603.06739v2
+
+## 7243. Improved high-dimensional estimation with Langevin dynamics and stochastic weight averaging
+
+- **ID:** `2603.06028` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Stanley Wei, Alex Damian, Jason D. Lee
+- **abs:** https://arxiv.org/abs/2603.06028 · **PDF:** https://arxiv.org/pdf/2603.06028v1
+
+## 7244. Agnostic learning in (almost) optimal time via Gaussian surface area
+
+- **ID:** `2603.06027` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Lucas Pesenti, Lucas Slot, Manuel Wiedmer
+- **abs:** https://arxiv.org/abs/2603.06027 · **PDF:** https://arxiv.org/pdf/2603.06027v1
+
+## 7245. Sensitivity-Aware Retrieval-Augmented Intent Clarification
+
+- **ID:** `2603.06025` · **primary:** `cs.IR` · **Submitted:** 2026-03-06
+- **Authors:** Maik Larooij
+- **abs:** https://arxiv.org/abs/2603.06025 · **PDF:** https://arxiv.org/pdf/2603.06025v1
+
+## 7246. ViewFusion: Structured Spatial Thinking Chains for Multi-View Reasoning
+
+- **ID:** `2603.06024` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Xingjian Tao, Yiwei Wang, Yujun Cai, et al.
+- **abs:** https://arxiv.org/abs/2603.06024 · **PDF:** https://arxiv.org/pdf/2603.06024v1
+
+## 7247. Large deviation principles for convolutional Bayesian neural networks
+
+- **ID:** `2603.06023` · **primary:** `math.PR` · **Submitted:** 2026-03-06
+- **Authors:** Federico Bassetti, Vassili De Palma, Lucia Ladelli
+- **abs:** https://arxiv.org/abs/2603.06023 · **PDF:** https://arxiv.org/pdf/2603.06023v1
+
+## 7248. Preventing Learning Stagnation in PPO by Scaling to 1 Million Parallel Environments
+
+- **ID:** `2603.06009` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Michael Beukman, Khimya Khetarpal, Zeyu Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.06009 · **PDF:** https://arxiv.org/pdf/2603.06009v1
+
+## 7249. MASFactory: A Graph-centric Framework for Orchestrating LLM-Based Multi-Agent Systems with Vibe Graphing
+
+- **ID:** `2603.06007` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Yang Liu, Jinxuan Cai, Yishen Li, et al.
+- **abs:** https://arxiv.org/abs/2603.06007 · **PDF:** https://arxiv.org/pdf/2603.06007v1
+
+## 7250. EvoESAP: Non-Uniform Expert Pruning for Sparse MoE
+
+- **ID:** `2603.06003` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Zongfang Liu, Shengkun Tang, Boyang Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.06003 · **PDF:** https://arxiv.org/pdf/2603.06003v1
+
+## 7251. Demystifying KAN for Vision Tasks: The RepKAN Approach
+
+- **ID:** `2603.06002` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Minjong Cheon
+- **abs:** https://arxiv.org/abs/2603.06002 · **PDF:** https://arxiv.org/pdf/2603.06002v1
+
+## 7252. Restoring Linguistic Grounding in VLA Models via Train-Free Attention Recalibration
+
+- **ID:** `2603.06001` · **primary:** `cs.RO` · **Submitted:** 2026-03-06
+- **Authors:** Ninghao Zhang, Bin Zhu, Shijie Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.06001 · **PDF:** https://arxiv.org/pdf/2603.06001v1
+
+## 7253. MM-ISTS: Cooperating Irregularly Sampled Time Series Forecasting with Multimodal Vision-Text LLMs
+
+- **ID:** `2603.05997` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Zhi Lei, Chenxi Liu, Hao Miao, et al.
+- **abs:** https://arxiv.org/abs/2603.05997 · **PDF:** https://arxiv.org/pdf/2603.05997v1
+
+## 7254. Track-SQL: Enhancing Generative Language Models with Dual-Extractive Modules for Schema and Context Tracking in Multi-turn Text-to-SQL
+
+- **ID:** `2603.05996` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Bingfeng Chen, Shaobin Shi, Yongqi Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.05996 · **PDF:** https://arxiv.org/pdf/2603.05996v1
+
+## 7255. TADPO: Reinforcement Learning Goes Off-road
+
+- **ID:** `2603.05995` · **primary:** `cs.RO` · **Submitted:** 2026-03-06
+- **Authors:** Zhouchonghao Wu, Raymond Song, Vedant Mundheda, et al.
+- **abs:** https://arxiv.org/abs/2603.05995 · **PDF:** https://arxiv.org/pdf/2603.05995v1
+
+## 7256. Rank-Factorized Implicit Neural Bias: Scaling Super-Resolution Transformer with FlashAttention
+
+- **ID:** `2603.06738` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Dongheon Lee, Seokju Yun, Jaegyun Im, et al.
+- **abs:** https://arxiv.org/abs/2603.06738 · **PDF:** https://arxiv.org/pdf/2603.06738v1
+
+## 7257. Technical Report: Automated Optical Inspection of Surgical Instruments
+
+- **ID:** `2603.05987` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Zunaira Shafqat, Atif Aftab Ahmed Jilani, Qurrat Ul Ain
+- **abs:** https://arxiv.org/abs/2603.05987 · **PDF:** https://arxiv.org/pdf/2603.05987v1
+
+## 7258. Agent Hunt: Bounty Based Collaborative Autoformalization With LLM Agents
+
+- **ID:** `2603.06737` · **primary:** `cs.LO` · **Submitted:** 2026-03-06
+- **Authors:** Chad E. Brown, Cezary Kaliszyk, Josef Urban
+- **abs:** https://arxiv.org/abs/2603.06737 · **PDF:** https://arxiv.org/pdf/2603.06737v1
+
+## 7259. An Interactive Multi-Agent System for Evaluation of New Product Concepts
+
+- **ID:** `2603.05980` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Bin Xuan, Ruo Ai, Hakyeon Lee
+- **abs:** https://arxiv.org/abs/2603.05980 · **PDF:** https://arxiv.org/pdf/2603.05980v1
+
+## 7260. Imagine How To Change: Explicit Procedure Modeling for Change Captioning
+
+- **ID:** `2603.05969` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Jiayang Sun, Zixin Guo, Min Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.05969 · **PDF:** https://arxiv.org/pdf/2603.05969v1
+
+## 7261. A federated learning framework with knowledge graph and temporal transformer for early sepsis prediction in multi-center ICUs
+
+- **ID:** `2603.15651` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Yue Chang, Guangsen Lin, Jyun Jie Chuang, et al.
+- **abs:** https://arxiv.org/abs/2603.15651 · **PDF:** https://arxiv.org/pdf/2603.15651v1
+
+## 7262. Skeleton-to-Image Encoding: Enabling Skeleton Representation Learning via Vision-Pretrained Models
+
+- **ID:** `2603.05963` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Siyuan Yang, Jun Liu, Hao Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.05963 · **PDF:** https://arxiv.org/pdf/2603.05963v1
+
+## 7263. Omni-Masked Gradient Descent: Memory-Efficient Optimization via Mask Traversal with Improved Convergence
+
+- **ID:** `2603.05960` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Hui Yang, Tao Ren, Jinyang Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.05960 · **PDF:** https://arxiv.org/pdf/2603.05960v2
+
+## 7264. Domain-Adaptive Model Merging across Disconnected Modes
+
+- **ID:** `2603.05957` · **primary:** `cs.DC` · **Submitted:** 2026-03-06
+- **Authors:** Junming Liu, Yusen Zhang, Rongchao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.05957 · **PDF:** https://arxiv.org/pdf/2603.05957v1
+
+## 7265. Who We Are, Where We Are: Mental Health at the Intersection of Person, Situation, and Large Language Models
+
+- **ID:** `2603.05953` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Nikita Soni, August Håkan Nilsson, Syeda Mahwish, et al.
+- **abs:** https://arxiv.org/abs/2603.05953 · **PDF:** https://arxiv.org/pdf/2603.05953v1
+
+## 7266. Energy-Driven Adaptive Visual Token Pruning for Efficient Vision-Language Models
+
+- **ID:** `2603.05950` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Jialuo He, Huangxun Chen
+- **abs:** https://arxiv.org/abs/2603.05950 · **PDF:** https://arxiv.org/pdf/2603.05950v1
+
+## 7267. XAI for Coding Agent Failures: Transforming Raw Execution Traces into Actionable Insights
+
+- **ID:** `2603.05941` · **primary:** `cs.SE` · **Submitted:** 2026-03-06
+- **Authors:** Arun Joshi
+- **abs:** https://arxiv.org/abs/2603.05941 · **PDF:** https://arxiv.org/pdf/2603.05941v1
+
+## 7268. Lipschitz-Based Robustness Certification Under Floating-Point Execution
+
+- **ID:** `2603.13334` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Toby Murray
+- **abs:** https://arxiv.org/abs/2603.13334 · **PDF:** https://arxiv.org/pdf/2603.13334v3
+
+## 7269. Facial Expression Recognition Using Residual Masking Network
+
+- **ID:** `2603.05937` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Luan Pham, The Huynh Vu, Tuan Anh Tran
+- **abs:** https://arxiv.org/abs/2603.05937 · **PDF:** https://arxiv.org/pdf/2603.05937v1
+
+## 7270. Implicit Style Conditioning: A Structured Style-Rewrite Framework for Low-Resource Character Modeling
+
+- **ID:** `2603.05933` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Chanhui Zhu
+- **abs:** https://arxiv.org/abs/2603.05933 · **PDF:** https://arxiv.org/pdf/2603.05933v1
+
+## 7271. X-OPD: Cross-Modal On-Policy Distillation for Capability Alignment in Speech LLMs
+
+- **ID:** `2603.24596` · **primary:** `eess.AS` · **Submitted:** 2026-03-06
+- **Authors:** Di Cao, Dongjie Fu, Hai Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.24596 · **PDF:** https://arxiv.org/pdf/2603.24596v2
+
+## 7272. A Persistent-State Dataflow Accelerator for Memory-Bound Linear Attention Decode on FPGA
+
+- **ID:** `2603.05931` · **primary:** `cs.AR` · **Submitted:** 2026-03-06
+- **Authors:** Neelesh Gupta, Peter Wang, Rajgopal Kannan, et al.
+- **abs:** https://arxiv.org/abs/2603.05931 · **PDF:** https://arxiv.org/pdf/2603.05931v1
+
+## 7273. How to Achieve Prototypical Birth and Death for OOD Detection?
+
+- **ID:** `2603.15650` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Ningkang Peng, Qianfeng Yu, Xiaoqian Peng, et al.
+- **abs:** https://arxiv.org/abs/2603.15650 · **PDF:** https://arxiv.org/pdf/2603.15650v1
+
+## 7274. Addressing the Ecological Fallacy in Larger LMs with Human Context
+
+- **ID:** `2603.05928` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Nikita Soni, Dhruv Vijay Kunjadiya, Pratham Piyush Shah, et al.
+- **abs:** https://arxiv.org/abs/2603.05928 · **PDF:** https://arxiv.org/pdf/2603.05928v1
+
+## 7275. RAC: Rectified Flow Auto Coder
+
+- **ID:** `2603.05925` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Sen Fang, Yalin Feng, Yanxin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.05925 · **PDF:** https://arxiv.org/pdf/2603.05925v1
+
+## 7276. Weak-SIGReg: Covariance Regularization for Stable Deep Learning
+
+- **ID:** `2603.05924` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Habibullah Akbar
+- **abs:** https://arxiv.org/abs/2603.05924 · **PDF:** https://arxiv.org/pdf/2603.05924v1
+
+## 7277. Learning Next Action Predictors from Human-Computer Interaction
+
+- **ID:** `2603.05923` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Omar Shaikh, Valentin Teutschbein, Kanishk Gandhi, et al.
+- **abs:** https://arxiv.org/abs/2603.05923 · **PDF:** https://arxiv.org/pdf/2603.05923v1
+
+## 7278. BlackMirror: Black-Box Backdoor Detection for Text-to-Image Models via Instruction-Response Deviation
+
+- **ID:** `2603.05921` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Feiran Li, Qianqian Xu, Shilong Bao, et al.
+- **abs:** https://arxiv.org/abs/2603.05921 · **PDF:** https://arxiv.org/pdf/2603.05921v1
+
+## 7279. Design Experiments to Compare Multi-armed Bandit Algorithms
+
+- **ID:** `2603.05919` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Huiling Meng, Ningyuan Chen, Xuefeng Gao
+- **abs:** https://arxiv.org/abs/2603.05919 · **PDF:** https://arxiv.org/pdf/2603.05919v1
+
+## 7280. Stock Market Prediction Using Node Transformer Architecture Integrated with BERT Sentiment Analysis
+
+- **ID:** `2603.05917` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Mohammad Al Ridhawi, Mahtab Haj Ali, Hussein Al Osman
+- **abs:** https://arxiv.org/abs/2603.05917 · **PDF:** https://arxiv.org/pdf/2603.05917v1
+
+## 7281. DeepFact: Co-Evolving Benchmarks and Agents for Deep Research Factuality
+
+- **ID:** `2603.05912` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Yukun Huang, Leonardo F. R. Ribeiro, Momchil Hardalov, et al.
+- **abs:** https://arxiv.org/abs/2603.05912 · **PDF:** https://arxiv.org/pdf/2603.05912v1
+
+## 7282. Beyond Scalar Rewards: Distributional Reinforcement Learning with Preordered Objectives for Safe and Reliable Autonomous Driving
+
+- **ID:** `2603.20230` · **primary:** `cs.RO` · **Submitted:** 2026-03-06
+- **Authors:** Ahmed Abouelazm, Jonas Michel, Daniel Bogdoll, et al.
+- **abs:** https://arxiv.org/abs/2603.20230 · **PDF:** https://arxiv.org/pdf/2603.20230v1
+
+## 7283. CORE-Seg: Reasoning-Driven Segmentation for Complex Lesions via Reinforcement Learning
+
+- **ID:** `2603.05911` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Yuxin Xie, Yuming Chen, Yishan Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.05911 · **PDF:** https://arxiv.org/pdf/2603.05911v1
+
+## 7284. The World Won't Stay Still: Programmable Evolution for Agent Benchmarks
+
+- **ID:** `2603.05910` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Guangrui Li, Yaochen Xie, Yi Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.05910 · **PDF:** https://arxiv.org/pdf/2603.05910v1
+
+## 7285. InfoGatherer: Principled Information Seeking via Evidence Retrieval and Strategic Questioning
+
+- **ID:** `2603.05909` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Maksym Taranukhin, Shuyue Stella Li, Evangelos Milios, et al.
+- **abs:** https://arxiv.org/abs/2603.05909 · **PDF:** https://arxiv.org/pdf/2603.05909v1
+
+## 7286. LUMINA: LLM-Guided GPU Architecture Exploration via Bottleneck Analysis
+
+- **ID:** `2603.05904` · **primary:** `cs.AR` · **Submitted:** 2026-03-06
+- **Authors:** Tao Zhang, Rui Ma, Shuotao Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.05904 · **PDF:** https://arxiv.org/pdf/2603.05904v2
+
+## 7287. Calibrated Credit Intelligence: Shift-Robust and Fair Risk Scoring with Bayesian Uncertainty and Gradient Boosting
+
+- **ID:** `2603.06733` · **primary:** `q-fin.RM` · **Submitted:** 2026-03-06
+- **Authors:** Srikumar Nayak
+- **abs:** https://arxiv.org/abs/2603.06733 · **PDF:** https://arxiv.org/pdf/2603.06733v1
+
+## 7288. Reference-guided Policy Optimization for Molecular Optimization via LLM Reasoning
+
+- **ID:** `2603.05900` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Xuan Li, Zhanke Zhou, Zongze Li, et al.
+- **abs:** https://arxiv.org/abs/2603.05900 · **PDF:** https://arxiv.org/pdf/2603.05900v1
+
+## 7289. Mitigating Bias in Concept Bottleneck Models for Fair and Interpretable Image Classification
+
+- **ID:** `2603.05899` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Schrasing Tong, Antoine Salaun, Vincent Yuan, et al.
+- **abs:** https://arxiv.org/abs/2603.05899 · **PDF:** https://arxiv.org/pdf/2603.05899v1
+
+## 7290. Building an Ensemble LLM Semantic Tagger for UN Security Council Resolutions
+
+- **ID:** `2603.05895` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Hussein Ghaly
+- **abs:** https://arxiv.org/abs/2603.05895 · **PDF:** https://arxiv.org/pdf/2603.05895v1
+
+## 7291. Lost in Stories: Consistency Bugs in Long Story Generation by LLMs
+
+- **ID:** `2603.05890` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Junjie Li, Xinrui Guo, Yuhao Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.05890 · **PDF:** https://arxiv.org/pdf/2603.05890v1
+
+## 7292. PixARMesh: Autoregressive Mesh-Native Single-View Scene Reconstruction
+
+- **ID:** `2603.05888` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Xiang Zhang, Sohyun Yoo, Hongrui Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.05888 · **PDF:** https://arxiv.org/pdf/2603.05888v1
+
+## 7293. Model2Kernel: Model-Aware Symbolic Execution For Safe CUDA Kernels
+
+- **ID:** `2603.24595` · **primary:** `cs.PL` · **Submitted:** 2026-03-06
+- **Authors:** Mengting He, Shihao Xia, Haomin Jia, et al.
+- **abs:** https://arxiv.org/abs/2603.24595 · **PDF:** https://arxiv.org/pdf/2603.24595v1
+
+## 7294. Reconstruct! Don't Encode: Self-Supervised Representation Reconstruction Loss for High-Intelligibility and Low-Latency Streaming Neural Audio Codec
+
+- **ID:** `2603.05887` · **primary:** `eess.AS` · **Submitted:** 2026-03-06
+- **Authors:** Junhyeok Lee, Xiluo He, Jihwan Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.05887 · **PDF:** https://arxiv.org/pdf/2603.05887v1
+
+## 7295. Automated Motif Indexing on the Arabian Nights
+
+- **ID:** `2603.19283` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Ibrahim H. Alyami, Mark A. Finlayson
+- **abs:** https://arxiv.org/abs/2603.19283 · **PDF:** https://arxiv.org/pdf/2603.19283v1
+
+## 7296. Computational Pathology in the Era of Emerging Foundation and Agentic AI -- International Expert Perspectives on Clinical Integration and Translational Readiness
+
+- **ID:** `2603.05884` · **primary:** `cs.CE` · **Submitted:** 2026-03-06
+- **Authors:** Qian Da, Yijiang Chen, Min Ju, et al.
+- **abs:** https://arxiv.org/abs/2603.05884 · **PDF:** https://arxiv.org/pdf/2603.05884v2
+
+## 7297. VerChol -- Grammar-First Tokenization for Agglutinative Languages
+
+- **ID:** `2603.05883` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Prabhu Raja
+- **abs:** https://arxiv.org/abs/2603.05883 · **PDF:** https://arxiv.org/pdf/2603.05883v1
+
+## 7298. Confidence Before Answering: A Paradigm Shift for Efficient LLM Uncertainty Estimation
+
+- **ID:** `2603.05881` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Changcheng Li, Jiancan Wu, Hengheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.05881 · **PDF:** https://arxiv.org/pdf/2603.05881v1
+
+## 7299. ROSE: Reordered SparseGPT for More Accurate One-Shot Large Language Models Pruning
+
+- **ID:** `2603.05878` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Mingluo Su, Huan Wang
+- **abs:** https://arxiv.org/abs/2603.05878 · **PDF:** https://arxiv.org/pdf/2603.05878v1
+
+## 7300. Stochastic Event Prediction via Temporal Motif Transitions
+
+- **ID:** `2603.05874` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** İbrahim Bahadır Altun, Ahmet Erdem Sarıyüce
+- **abs:** https://arxiv.org/abs/2603.05874 · **PDF:** https://arxiv.org/pdf/2603.05874v1
+
+## 7301. Beyond Accuracy: Introducing a Symbolic-Mechanistic Approach to Interpretable Evaluation
+
+- **ID:** `2603.23517` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Reza Habibi, Darian Lee, Magy Seif El-Nasr
+- **abs:** https://arxiv.org/abs/2603.23517 · **PDF:** https://arxiv.org/pdf/2603.23517v1
+
+## 7302. PolyBlocks: A Compiler Infrastructure for AI Chips and Programming Frameworks
+
+- **ID:** `2603.06731` · **primary:** `cs.PL` · **Submitted:** 2026-03-06
+- **Authors:** Uday Bondhugula, Akshay Baviskar, Navdeep Katel, et al.
+- **abs:** https://arxiv.org/abs/2603.06731 · **PDF:** https://arxiv.org/pdf/2603.06731v2
+
+## 7303. ReflexiCoder: Teaching Large Language Models to Self-Reflect on Generated Code and Self-Correct It via Reinforcement Learning
+
+- **ID:** `2603.05863` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Juyong Jiang, Jiasi Shen, Sunghun Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.05863 · **PDF:** https://arxiv.org/pdf/2603.05863v1
+
+## 7304. Evolving Medical Imaging Agents via Experience-driven Self-skill Discovery
+
+- **ID:** `2603.05860` · **primary:** `cs.AI` · **Submitted:** 2026-03-06
+- **Authors:** Lin Fan, Pengyu Dai, Zhipeng Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.05860 · **PDF:** https://arxiv.org/pdf/2603.05860v1
+
+## 7305. Don't Freeze, Don't Crash: Extending the Safe Operating Range of Neural Navigation in Dense Crowds
+
+- **ID:** `2603.06729` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Jiefu Zhang, Yang Xu, Vaneet Aggarwal
+- **abs:** https://arxiv.org/abs/2603.06729 · **PDF:** https://arxiv.org/pdf/2603.06729v1
+
+## 7306. KD-EKF: Knowledge-Distilled Adaptive Covariance EKF for Robust UWB/PDR Indoor Localization
+
+- **ID:** `2603.18027` · **primary:** `eess.SP` · **Submitted:** 2026-03-06
+- **Authors:** Kyeonghyun Yoo, Wooyong Jung, Namkyung Yoon, et al.
+- **abs:** https://arxiv.org/abs/2603.18027 · **PDF:** https://arxiv.org/pdf/2603.18027v1
+
+## 7307. Orion: Characterizing and Programming Apple's Neural Engine for LLM Training and Inference
+
+- **ID:** `2603.06728` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Ramchand Kumaresan
+- **abs:** https://arxiv.org/abs/2603.06728 · **PDF:** https://arxiv.org/pdf/2603.06728v1
+
+## 7308. Remote Sensing Image Classification Using Deep Ensemble Learning
+
+- **ID:** `2603.05844` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Niful Islam, Md. Rayhan Ahmed, Nur Mohammad Fahad, et al.
+- **abs:** https://arxiv.org/abs/2603.05844 · **PDF:** https://arxiv.org/pdf/2603.05844v1
+
+## 7309. Safe Transformer: An Explicit Safety Bit For Interpretable And Controllable Alignment
+
+- **ID:** `2603.06727` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Jingyuan Feng, Andrew Gambardella, Gouki Minegishi, et al.
+- **abs:** https://arxiv.org/abs/2603.06727 · **PDF:** https://arxiv.org/pdf/2603.06727v1
+
+## 7310. Evaluating LLM Alignment With Human Trust Models
+
+- **ID:** `2603.05839` · **primary:** `cs.MA` · **Submitted:** 2026-03-06
+- **Authors:** Anushka Debnath, Stephen Cranefield, Bastin Tony Roy Savarimuthu, et al.
+- **abs:** https://arxiv.org/abs/2603.05839 · **PDF:** https://arxiv.org/pdf/2603.05839v1
+
+## 7311. Lexara: A User-Centered Toolkit for Evaluating Large Language Models for Conversational Visual Analytics
+
+- **ID:** `2603.05832` · **primary:** `cs.HC` · **Submitted:** 2026-03-06
+- **Authors:** Srishti Palani, Vidya Setlur
+- **abs:** https://arxiv.org/abs/2603.05832 · **PDF:** https://arxiv.org/pdf/2603.05832v1
+
+## 7312. MSA: Memory Sparse Attention for Efficient End-to-End Memory Model Scaling to 100M Tokens
+
+- **ID:** `2603.23516` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Yu Chen, Runkai Chen, Sheng Yi, et al.
+- **abs:** https://arxiv.org/abs/2603.23516 · **PDF:** https://arxiv.org/pdf/2603.23516v1
+
+## 7313. Test-Time Adaptation via Many-Shot Prompting: Benefits, Limits, and Pitfalls
+
+- **ID:** `2603.05829` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Shubhangi Upasani, Chen Wu, Jay Rainton, et al.
+- **abs:** https://arxiv.org/abs/2603.05829 · **PDF:** https://arxiv.org/pdf/2603.05829v3
+
+## 7314. HART: Data-Driven Hallucination Attribution and Evidence-Based Tracing for Large Language Models
+
+- **ID:** `2603.05828` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Shize Liang, Hongzhi Wang
+- **abs:** https://arxiv.org/abs/2603.05828 · **PDF:** https://arxiv.org/pdf/2603.05828v1
+
+## 7315. Self-Auditing Parameter-Efficient Fine-Tuning for Few-Shot 3D Medical Image Segmentation
+
+- **ID:** `2603.05822` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Son Thai Ly, Hien V. Nguyen
+- **abs:** https://arxiv.org/abs/2603.05822 · **PDF:** https://arxiv.org/pdf/2603.05822v1
+
+## 7316. Regression Models Meet Foundation Models: A Hybrid-AI Approach to Practical Electricity Price Forecasting
+
+- **ID:** `2603.06726` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Yunzhong Qiu, Binzhu Li, Hao Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.06726 · **PDF:** https://arxiv.org/pdf/2603.06726v1
+
+## 7317. RouteGoT: Node-Adaptive Routing for Cost-Efficient Graph of Thoughts Reasoning
+
+- **ID:** `2603.05818` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Yuhang Liu, Ruijie Wang, Yunlong Chu, et al.
+- **abs:** https://arxiv.org/abs/2603.05818 · **PDF:** https://arxiv.org/pdf/2603.05818v1
+
+## 7318. Training a Large Language Model for Medical Coding Using Privacy-Preserving Synthetic Clinical Data
+
+- **ID:** `2603.23515` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** John Cook, Michael Wyatt, Peng Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.23515 · **PDF:** https://arxiv.org/pdf/2603.23515v1
+
+## 7319. Margin and Consistency Supervision for Calibrated and Robust Vision Models
+
+- **ID:** `2603.05812` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Salim Khazem
+- **abs:** https://arxiv.org/abs/2603.05812 · **PDF:** https://arxiv.org/pdf/2603.05812v1
+
+## 7320. Characterizing the ability of LLMs to recapitulate Americans' distributional responses to public opinion polling questions across political issues
+
+- **ID:** `2603.20229` · **primary:** `cs.CY` · **Submitted:** 2026-03-06
+- **Authors:** Eric Gong, Nathan E. Sanders, Bruce Schneier
+- **abs:** https://arxiv.org/abs/2603.20229 · **PDF:** https://arxiv.org/pdf/2603.20229v1
+
+## 7321. Bayesian Conservative Policy Optimization (BCPO): A Novel Uncertainty-Calibrated Offline Reinforcement Learning with Credible Lower Bounds
+
+- **ID:** `2603.12284` · **primary:** `stat.ME` · **Submitted:** 2026-03-06
+- **Authors:** Debashis Chatterjee
+- **abs:** https://arxiv.org/abs/2603.12284 · **PDF:** https://arxiv.org/pdf/2603.12284v1
+
+## 7322. MoE Lens -- An Expert Is All You Need
+
+- **ID:** `2603.05806` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Marmik Chaudhari, Idhant Gulati, Nishkal Hundia, et al.
+- **abs:** https://arxiv.org/abs/2603.05806 · **PDF:** https://arxiv.org/pdf/2603.05806v1
+
+## 7323. Sparse Crosscoders for diffing MoEs and Dense models
+
+- **ID:** `2603.05805` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Marmik Chaudhari, Nishkal Hundia, Idhant Gulati
+- **abs:** https://arxiv.org/abs/2603.05805 · **PDF:** https://arxiv.org/pdf/2603.05805v1
+
+## 7324. Ambiguity Collapse by LLMs: A Taxonomy of Epistemic Risks
+
+- **ID:** `2603.05801` · **primary:** `cs.CY` · **Submitted:** 2026-03-06
+- **Authors:** Shira Gur-Arieh, Angelina Wang, Sina Fazelpour
+- **abs:** https://arxiv.org/abs/2603.05801 · **PDF:** https://arxiv.org/pdf/2603.05801v1
+
+## 7325. StreamWise: Serving Multi-Modal Generation in Real-Time at Scale
+
+- **ID:** `2603.05800` · **primary:** `cs.DC` · **Submitted:** 2026-03-06
+- **Authors:** Haoran Qiu, Gohar Irfan Chaudhry, Chaojie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.05800 · **PDF:** https://arxiv.org/pdf/2603.05800v1
+
+## 7326. Bi Directional Feedback Fusion for Activity Aware Forecasting of Indoor CO2 and PM2.5
+
+- **ID:** `2603.06724` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Harshala Gammulle, Lidia Morawska, Sridha Sridharan, et al.
+- **abs:** https://arxiv.org/abs/2603.06724 · **PDF:** https://arxiv.org/pdf/2603.06724v1
+
+## 7327. AWPD: Frequency Shield Network for Agnostic Watermark Presence Detection
+
+- **ID:** `2603.06723` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Xiang Ao, Yiling Du, Zidan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.06723 · **PDF:** https://arxiv.org/pdf/2603.06723v2
+
+## 7328. The Coordination Gap: Multi-Agent Alternation Metrics for Temporal Fairness in Repeated Games
+
+- **ID:** `2603.05789` · **primary:** `cs.MA` · **Submitted:** 2026-03-06
+- **Authors:** Nikolaos Al. Papadopoulos, Konstantinos Psannis
+- **abs:** https://arxiv.org/abs/2603.05789 · **PDF:** https://arxiv.org/pdf/2603.05789v4
+
+## 7329. ProtAlign: Contrastive learning paradigm for Sequence and structure alignment
+
+- **ID:** `2603.06722` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Aditya Ranganath, Hasin Us Sami, Kowshik Thopalli, et al.
+- **abs:** https://arxiv.org/abs/2603.06722 · **PDF:** https://arxiv.org/pdf/2603.06722v1
+
+## 7330. Proof-of-Guardrail in AI Agents and What (Not) to Trust from It
+
+- **ID:** `2603.05786` · **primary:** `cs.CR` · **Submitted:** 2026-03-06
+- **Authors:** Xisen Jin, Michael Duan, Qin Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.05786 · **PDF:** https://arxiv.org/pdf/2603.05786v1
+
+## 7331. Visual Words Meet BM25: Sparse Auto-Encoder Visual Word Scoring for Image Retrieval
+
+- **ID:** `2603.05781` · **primary:** `cs.CV` · **Submitted:** 2026-03-06
+- **Authors:** Donghoon Han, Eunhwan Park, Seunghyeon Seo
+- **abs:** https://arxiv.org/abs/2603.05781 · **PDF:** https://arxiv.org/pdf/2603.05781v1
+
+## 7332. Balancing Domestic and Global Perspectives: Evaluating Dual-Calibration and LLM-Generated Nudges for Diverse News Recommendation
+
+- **ID:** `2603.05780` · **primary:** `cs.IR` · **Submitted:** 2026-03-06
+- **Authors:** Ruixuan Sun, Matthew Zent, Minzhu Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.05780 · **PDF:** https://arxiv.org/pdf/2603.05780v1
+
+## 7333. Tutor Move Taxonomy: A Theory-Aligned Framework for Analyzing Instructional Moves in Tutoring
+
+- **ID:** `2603.05778` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Zhuqian Zhou, Kirk Vanacore, Tamisha Thompson, et al.
+- **abs:** https://arxiv.org/abs/2603.05778 · **PDF:** https://arxiv.org/pdf/2603.05778v1
+
+## 7334. From Statistical Fidelity to Clinical Consistency: Scalable Generation and Auditing of Synthetic Patient Trajectories
+
+- **ID:** `2603.06720` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Guanglin Zhou, Armin Catic, Motahare Shabestari, et al.
+- **abs:** https://arxiv.org/abs/2603.06720 · **PDF:** https://arxiv.org/pdf/2603.06720v1
+
+## 7335. PVminerLLM: Structured Extraction of Patient Voice from Patient-Generated Text using Large Language Models
+
+- **ID:** `2603.05776` · **primary:** `cs.CL` · **Submitted:** 2026-03-06
+- **Authors:** Samah Fodeh, Linhai Ma, Ganesh Puthiaraju, et al.
+- **abs:** https://arxiv.org/abs/2603.05776 · **PDF:** https://arxiv.org/pdf/2603.05776v1
+
+## 7336. First-Order Softmax Weighted Switching Gradient Method for Distributed Stochastic Minimax Optimization with Stochastic Constraints
+
+- **ID:** `2603.05774` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Zhankun Luo, Antesh Upadhyay, Sang Bin Moon, et al.
+- **abs:** https://arxiv.org/abs/2603.05774 · **PDF:** https://arxiv.org/pdf/2603.05774v1
+
+## 7337. Knowing without Acting: The Disentangled Geometry of Safety Mechanisms in Large Language Models
+
+- **ID:** `2603.05773` · **primary:** `cs.CR` · **Submitted:** 2026-03-06
+- **Authors:** Jinman Wu, Yi Xie, Shen Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.05773 · **PDF:** https://arxiv.org/pdf/2603.05773v2
+
+## 7338. Depth Charge: Jailbreak Large Language Models from Deep Safety Attention Heads
+
+- **ID:** `2603.05772` · **primary:** `cs.CR` · **Submitted:** 2026-03-06
+- **Authors:** Jinman Wu, Yi Xie, Shiqian Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.05772 · **PDF:** https://arxiv.org/pdf/2603.05772v2
+
+## 7339. Bridging Domains through Subspace-Aware Model Merging
+
+- **ID:** `2603.05768` · **primary:** `cs.LG` · **Submitted:** 2026-03-06
+- **Authors:** Levy Chaves, Chao Zhou, Rebekka Burkholz, et al.
+- **abs:** https://arxiv.org/abs/2603.05768 · **PDF:** https://arxiv.org/pdf/2603.05768v2
+
+## 7340. Physically Accurate Differentiable Inverse Rendering for Radio Frequency Digital Twin
+
+- **ID:** `2603.18026` · **primary:** `eess.SP` · **Submitted:** 2026-03-05
+- **Authors:** Xingyu Chen, Xinyu Zhang, Kai Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.18026 · **PDF:** https://arxiv.org/pdf/2603.18026v1
+
+## 7341. TML-Bench: Benchmark for Data Science Agents on Tabular ML Tasks
+
+- **ID:** `2603.05764` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Mykola Pinchuk
+- **abs:** https://arxiv.org/abs/2603.05764 · **PDF:** https://arxiv.org/pdf/2603.05764v1
+
+## 7342. Score-Guided Proximal Projection: A Unified Geometric Framework for Rectified Flow Editing
+
+- **ID:** `2603.05761` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Vansh Bansal, James G Scott
+- **abs:** https://arxiv.org/abs/2603.05761 · **PDF:** https://arxiv.org/pdf/2603.05761v1
+
+## 7343. MIRACL: A Diverse Meta-Reinforcement Learning for Multi-Objective Multi-Echelon Combinatorial Supply Chain Optimisation
+
+- **ID:** `2603.05760` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Rifny Rachman, Josh Tingey, Richard Allmendinger, et al.
+- **abs:** https://arxiv.org/abs/2603.05760 · **PDF:** https://arxiv.org/pdf/2603.05760v1
+
+## 7344. Full Dynamic Range Sky-Modelling For Image Based Lighting
+
+- **ID:** `2603.05758` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Ian J. Maquignaz
+- **abs:** https://arxiv.org/abs/2603.05758 · **PDF:** https://arxiv.org/pdf/2603.05758v1
+
+## 7345. NERdME: a Named Entity Recognition Dataset for Indexing Research Artifacts in Code Repositories
+
+- **ID:** `2603.05750` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Genet Asefa Gesese, Zongxiong Chen, Shufan Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.05750 · **PDF:** https://arxiv.org/pdf/2603.05750v1
+
+## 7346. CodeScout: Contextual Problem Statement Enhancement for Software Agents
+
+- **ID:** `2603.05744` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Manan Suri, Xiangci Li, Mehdi Shojaie, et al.
+- **abs:** https://arxiv.org/abs/2603.05744 · **PDF:** https://arxiv.org/pdf/2603.05744v1
+
+## 7347. Designing Explainable Conversational Agentic Systems for Guaraní Speakers
+
+- **ID:** `2603.05743` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Samantha Adorno, Akshata Kishore Moharir, Ratna Kandala
+- **abs:** https://arxiv.org/abs/2603.05743 · **PDF:** https://arxiv.org/pdf/2603.05743v2
+
+## 7348. Dynamic Targeting of Satellite Observations Using Supplemental Geostationary Satellite Data and Hierarchical Planning
+
+- **ID:** `2603.06719` · **primary:** `cs.RO` · **Submitted:** 2026-03-05
+- **Authors:** Akseli Kangaslahti, Itai Zilberstein, Alberto Candela, et al.
+- **abs:** https://arxiv.org/abs/2603.06719 · **PDF:** https://arxiv.org/pdf/2603.06719v1
+
+## 7349. Revisiting the (Sub)Optimality of Best-of-N for Inference-Time Alignment
+
+- **ID:** `2603.05739` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Ved Sriraman, Adam Block
+- **abs:** https://arxiv.org/abs/2603.05739 · **PDF:** https://arxiv.org/pdf/2603.05739v1
+
+## 7350. Agentic AI -- Physicist Collaboration in Experimental Particle Physics: A Proof-of-Concept Measurement with LEP Open Data
+
+- **ID:** `2603.05735` · **primary:** `hep-ex` · **Submitted:** 2026-03-05
+- **Authors:** Anthony Badea, Yi Chen, Marcello Maggi, et al.
+- **abs:** https://arxiv.org/abs/2603.05735 · **PDF:** https://arxiv.org/pdf/2603.05735v2
+
+## 7351. Structured Multidimensional Representation Learning for Large Language Models
+
+- **ID:** `2603.05727` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Alaa El Ichi, Khalide Jbilou, Mohamed El Guide, et al.
+- **abs:** https://arxiv.org/abs/2603.05727 · **PDF:** https://arxiv.org/pdf/2603.05727v1
+
+## 7352. LTLGuard: Formalizing LTL Specifications with Compact Language Models and Lightweight Symbolic Reasoning
+
+- **ID:** `2603.05728` · **primary:** `cs.LO` · **Submitted:** 2026-03-05
+- **Authors:** Medina Andresel, Cristinel Mateis, Dejan Nickovic, et al.
+- **abs:** https://arxiv.org/abs/2603.05728 · **PDF:** https://arxiv.org/pdf/2603.05728v1
+
+## 7353. Cultural Perspectives and Expectations for Generative AI: A Global Survey Approach
+
+- **ID:** `2603.05723` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Erin van Liemt, Renee Shelby, Andrew Smart, et al.
+- **abs:** https://arxiv.org/abs/2603.05723 · **PDF:** https://arxiv.org/pdf/2603.05723v1
+
+## 7354. Unsupervised domain adaptation for radioisotope identification in gamma spectroscopy
+
+- **ID:** `2603.05719` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Peter Lalor, Ayush Panigrahy, Alex Hagen
+- **abs:** https://arxiv.org/abs/2603.05719 · **PDF:** https://arxiv.org/pdf/2603.05719v1
+
+## 7355. The Rise of AI in Weather and Climate Information and its Impact on Global Inequality
+
+- **ID:** `2603.05710` · **primary:** `physics.ao-ph` · **Submitted:** 2026-03-05
+- **Authors:** Amirpasha Mozaffari, Amanda Duarte, Lina Teckentrup, et al.
+- **abs:** https://arxiv.org/abs/2603.05710 · **PDF:** https://arxiv.org/pdf/2603.05710v1
+
+## 7356. Reasoning Models Struggle to Control their Chains of Thought
+
+- **ID:** `2603.05706` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Chen Yueh-Han, Robert McCarthy, Bruce W. Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.05706 · **PDF:** https://arxiv.org/pdf/2603.05706v1
+
+## 7357. Random Dot Product Graphs as Dynamical Systems: Limitations and Opportunities
+
+- **ID:** `2603.05703` · **primary:** `stat.ME` · **Submitted:** 2026-03-05
+- **Authors:** Giulio Valentino Dalla Riva
+- **abs:** https://arxiv.org/abs/2603.05703 · **PDF:** https://arxiv.org/pdf/2603.05703v1
+
+## 7358. Tureis: Transformer-based Unified Resilience for IoT Devices in Smart Homes
+
+- **ID:** `2603.10038` · **primary:** `cs.NI` · **Submitted:** 2026-03-05
+- **Authors:** Alireza Borhani, Vafa Andalibi, Bahar Asgari
+- **abs:** https://arxiv.org/abs/2603.10038 · **PDF:** https://arxiv.org/pdf/2603.10038v1
+
+## 7359. Towards Robust Retrieval-Augmented Generation Based on Knowledge Graph: A Comparative Analysis
+
+- **ID:** `2603.05698` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Hazem Amamou, Stéphane Gagnon, Alan Davoust, et al.
+- **abs:** https://arxiv.org/abs/2603.05698 · **PDF:** https://arxiv.org/pdf/2603.05698v2
+
+## 7360. Autonomous Algorithm Discovery for Ptychography via Evolutionary LLM Reasoning
+
+- **ID:** `2603.05696` · **primary:** `cs.CE` · **Submitted:** 2026-03-05
+- **Authors:** Xiangyu Yin, Ming Du, Junjing Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.05696 · **PDF:** https://arxiv.org/pdf/2603.05696v1
+
+## 7361. Warm Starting State-Space Models with Automata Learning
+
+- **ID:** `2603.05694` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** William Fishell, Sam Nicholas Kouteili, Mark Santolucito
+- **abs:** https://arxiv.org/abs/2603.05694 · **PDF:** https://arxiv.org/pdf/2603.05694v1
+
+## 7362. Longitudinal Lesion Inpainting in Brain MRI via 3D Region Aware Diffusion
+
+- **ID:** `2603.05693` · **primary:** `eess.IV` · **Submitted:** 2026-03-05
+- **Authors:** Zahra Karimaghaloo, Dumitru Fetco, Haz-Edine Assemlal, et al.
+- **abs:** https://arxiv.org/abs/2603.05693 · **PDF:** https://arxiv.org/pdf/2603.05693v1
+
+## 7363. Parallelization Strategies for Dense LLM Deployment: Navigating Through Application-Specific Tradeoffs and Bottlenecks
+
+- **ID:** `2603.05692` · **primary:** `cs.DC` · **Submitted:** 2026-03-05
+- **Authors:** Burak Topcu, Musa Oguzhan Cim, Poovaiah Palangappa, et al.
+- **abs:** https://arxiv.org/abs/2603.05692 · **PDF:** https://arxiv.org/pdf/2603.05692v1
+
+## 7364. Improved Scaling Laws via Weak-to-Strong Generalization in Random Feature Ridge Regression
+
+- **ID:** `2603.05691` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Diyuan Wu, Lehan Chen, Theodor Misiakiewicz, et al.
+- **abs:** https://arxiv.org/abs/2603.05691 · **PDF:** https://arxiv.org/pdf/2603.05691v1
+
+## 7365. FreeTxt-Vi: A Benchmarked Vietnamese-English Toolkit for Segmentation, Sentiment, and Summarisation
+
+- **ID:** `2603.05690` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Hung Nguyen Huy, Mo El-Haj, Dawn Knight, et al.
+- **abs:** https://arxiv.org/abs/2603.05690 · **PDF:** https://arxiv.org/pdf/2603.05690v1
+
+## 7366. SecureRAG-RTL: A Retrieval-Augmented, Multi-Agent, Zero-Shot LLM-Driven Framework for Hardware Vulnerability Detection
+
+- **ID:** `2603.05689` · **primary:** `cs.CR` · **Submitted:** 2026-03-05
+- **Authors:** Touseef Hasan, Blessing Airehenbuwa, Nitin Pundir, et al.
+- **abs:** https://arxiv.org/abs/2603.05689 · **PDF:** https://arxiv.org/pdf/2603.05689v1
+
+## 7367. Reinforcement Learning for Power-Flow Network Analysis
+
+- **ID:** `2603.05673` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Alperen Ergur, Julia Lindberg, Vinny Miller
+- **abs:** https://arxiv.org/abs/2603.05673 · **PDF:** https://arxiv.org/pdf/2603.05673v1
+
+## 7368. DepthCharge: A Domain-Agnostic Framework for Measuring Depth-Dependent Knowledge in Large Language Models
+
+- **ID:** `2603.23514` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Alexander Sheppert
+- **abs:** https://arxiv.org/abs/2603.23514 · **PDF:** https://arxiv.org/pdf/2603.23514v1
+
+## 7369. The Value of Graph-based Encoding in NBA Salary Prediction
+
+- **ID:** `2603.05671` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Junhao Su, David Grimsman, Christopher Archibald
+- **abs:** https://arxiv.org/abs/2603.05671 · **PDF:** https://arxiv.org/pdf/2603.05671v1
+
+## 7370. Scaling Agentic Capabilities, Not Context: Efficient Reinforcement Finetuning for Large Toolspaces
+
+- **ID:** `2603.06713` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Karan Gupta, Pranav Vajreshwari, Yash Pandya, et al.
+- **abs:** https://arxiv.org/abs/2603.06713 · **PDF:** https://arxiv.org/pdf/2603.06713v1
+
+## 7371. Uncertainty-Aware Solar Flare Regression
+
+- **ID:** `2603.06712` · **primary:** `astro-ph.SR` · **Submitted:** 2026-03-05
+- **Authors:** Jinsu Hong, Chetraj Pandey, Berkay Aydin
+- **abs:** https://arxiv.org/abs/2603.06712 · **PDF:** https://arxiv.org/pdf/2603.06712v1
+
+## 7372. When Rubrics Fail: Error Enumeration as Reward in Reference-Free RL Post-Training for Virtual Try-On
+
+- **ID:** `2603.05659` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Wisdom Ikezogwo, Mehmet Saygin Seyfioglu, Ranjay Krishna, et al.
+- **abs:** https://arxiv.org/abs/2603.05659 · **PDF:** https://arxiv.org/pdf/2603.05659v2
+
+## 7373. The DSA's Blind Spot: Algorithmic Audit of Advertising and Minor Profiling on TikTok
+
+- **ID:** `2603.05653` · **primary:** `cs.CY` · **Submitted:** 2026-03-05
+- **Authors:** Sara Solarova, Matej Mosnar, Matus Tibensky, et al.
+- **abs:** https://arxiv.org/abs/2603.05653 · **PDF:** https://arxiv.org/pdf/2603.05653v1
+
+## 7374. The Fragility Of Moral Judgment In Large Language Models
+
+- **ID:** `2603.05651` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Tom van Nuenen, Pratik S. Sachdeva
+- **abs:** https://arxiv.org/abs/2603.05651 · **PDF:** https://arxiv.org/pdf/2603.05651v1
+
+## 7375. Mining Beyond the Bools: Learning Data Transformations and Temporal Specifications
+
+- **ID:** `2603.06710` · **primary:** `cs.LO` · **Submitted:** 2026-03-05
+- **Authors:** Sam Nicholas Kouteili, William Fishell, Christian Scaff, et al.
+- **abs:** https://arxiv.org/abs/2603.06710 · **PDF:** https://arxiv.org/pdf/2603.06710v1
+
+## 7376. Relational Semantic Reasoning on 3D Scene Graphs for Open World Interactive Object Search
+
+- **ID:** `2603.05642` · **primary:** `cs.RO` · **Submitted:** 2026-03-05
+- **Authors:** Imen Mahdi, Matteo Cassinelli, Fabien Despinoy, et al.
+- **abs:** https://arxiv.org/abs/2603.05642 · **PDF:** https://arxiv.org/pdf/2603.05642v1
+
+## 7377. Making Reconstruction FID Predictive of Diffusion Generation FID
+
+- **ID:** `2603.05630` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Tongda Xu, Mingwei He, Shady Abu-Hussein, et al.
+- **abs:** https://arxiv.org/abs/2603.05630 · **PDF:** https://arxiv.org/pdf/2603.05630v1
+
+## 7378. Identifying Adversary Characteristics from an Observed Attack
+
+- **ID:** `2603.05625` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Soyon Choi, Scott Alfeld, Meiyi Ma
+- **abs:** https://arxiv.org/abs/2603.05625 · **PDF:** https://arxiv.org/pdf/2603.05625v1
+
+## 7379. Post Fusion Bird's Eye View Feature Stabilization for Robust Multimodal 3D Detection
+
+- **ID:** `2603.05623` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Trung Tien Dong, Dev Thakkar, Arman Sargolzaei, et al.
+- **abs:** https://arxiv.org/abs/2603.05623 · **PDF:** https://arxiv.org/pdf/2603.05623v1
+
+## 7380. Adversarial Batch Representation Augmentation for Batch Correction in High-Content Cellular Screening
+
+- **ID:** `2603.05622` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Lei Tong, Xujing Yao, Adam Corrigan, et al.
+- **abs:** https://arxiv.org/abs/2603.05622 · **PDF:** https://arxiv.org/pdf/2603.05622v1
+
+## 7381. RACAS: Controlling Diverse Robots With a Single Agentic System
+
+- **ID:** `2603.05621` · **primary:** `cs.RO` · **Submitted:** 2026-03-05
+- **Authors:** Dylan R. Ashley, Jan Przepióra, Yimeng Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.05621 · **PDF:** https://arxiv.org/pdf/2603.05621v2
+
+## 7382. Safer Reasoning Traces: Measuring and Mitigating Chain-of-Thought Leakage in LLMs
+
+- **ID:** `2603.05618` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Patrick Ahrend, Tobias Eder, Xiyang Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.05618 · **PDF:** https://arxiv.org/pdf/2603.05618v1
+
+## 7383. NOTAI.AI: Explainable Detection of Machine-Generated Text via Curvature and Feature Attribution
+
+- **ID:** `2603.05617` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Oleksandr Marchenko Breneur, Adelaide Danilov, Aria Nourbakhsh, et al.
+- **abs:** https://arxiv.org/abs/2603.05617 · **PDF:** https://arxiv.org/pdf/2603.05617v1
+
+## 7384. Berta: an open-source, modular tool for AI-enabled clinical documentation
+
+- **ID:** `2603.23513` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Samridhi Vaid, Mike Weldon, Jesse Dunn, et al.
+- **abs:** https://arxiv.org/abs/2603.23513 · **PDF:** https://arxiv.org/pdf/2603.23513v1
+
+## 7385. Real-Time AI Service Economy: A Framework for Agentic Computing Across the Continuum
+
+- **ID:** `2603.05614` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Lauri Lovén, Alaa Saleh, Reza Farahani, et al.
+- **abs:** https://arxiv.org/abs/2603.05614 · **PDF:** https://arxiv.org/pdf/2603.05614v1
+
+## 7386. Behavior-dLDS: A decomposed linear dynamical systems model for neural activity partially constrained by behavior
+
+- **ID:** `2603.05612` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-05
+- **Authors:** Eva Yezerets, En Yang, Misha B. Ahrens, et al.
+- **abs:** https://arxiv.org/abs/2603.05612 · **PDF:** https://arxiv.org/pdf/2603.05612v1
+
+## 7387. DreamCAD: Scaling Multi-modal CAD Generation using Differentiable Parametric Surfaces
+
+- **ID:** `2603.05607` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Mohammad Sadil Khan, Muhammad Usama, Rolandos Alexandros Potamias, et al.
+- **abs:** https://arxiv.org/abs/2603.05607 · **PDF:** https://arxiv.org/pdf/2603.05607v1
+
+## 7388. From Decoupled to Coupled: Robustness Verification for Learning-based Keypoint Detection with Joint Specifications
+
+- **ID:** `2603.05604` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Xusheng Luo, Changliu Liu
+- **abs:** https://arxiv.org/abs/2603.05604 · **PDF:** https://arxiv.org/pdf/2603.05604v1
+
+## 7389. On the Value of Tokeniser Pretraining in Physics Foundation Models
+
+- **ID:** `2603.05598` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Hadi Sotoudeh, Payel Mukhopadhyay, Ruben Ohana, et al.
+- **abs:** https://arxiv.org/abs/2603.05598 · **PDF:** https://arxiv.org/pdf/2603.05598v2
+
+## 7390. RoboPocket: Improve Robot Policies Instantly with Your Phone
+
+- **ID:** `2603.05504` · **primary:** `cs.RO` · **Submitted:** 2026-03-05
+- **Authors:** Junjie Fang, Wendi Chen, Han Xue, et al.
+- **abs:** https://arxiv.org/abs/2603.05504 · **PDF:** https://arxiv.org/pdf/2603.05504v2
+
+## 7391. POET-X: Memory-efficient LLM Training by Scaling Orthogonal Transformation
+
+- **ID:** `2603.05500` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Zeju Qiu, Lixin Liu, Adrian Weller, et al.
+- **abs:** https://arxiv.org/abs/2603.05500 · **PDF:** https://arxiv.org/pdf/2603.05500v1
+
+## 7392. The Spike, the Sparse and the Sink: Anatomy of Massive Activations and Attention Sinks
+
+- **ID:** `2603.05498` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Shangwen Sun, Alfredo Canziani, Yann LeCun, et al.
+- **abs:** https://arxiv.org/abs/2603.05498 · **PDF:** https://arxiv.org/pdf/2603.05498v1
+
+## 7393. Cheap Thrills: Effective Amortized Optimization Using Inexpensive Labels
+
+- **ID:** `2603.05495` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Khai Nguyen, Petros Ellinas, Anvita Bhagavathula, et al.
+- **abs:** https://arxiv.org/abs/2603.05495 · **PDF:** https://arxiv.org/pdf/2603.05495v1
+
+## 7394. Censored LLMs as a Natural Testbed for Secret Knowledge Elicitation
+
+- **ID:** `2603.05494` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Helena Casademunt, Bartosz Cywiński, Khoi Tran, et al.
+- **abs:** https://arxiv.org/abs/2603.05494 · **PDF:** https://arxiv.org/pdf/2603.05494v2
+
+## 7395. Reasoning Theater: Disentangling Model Beliefs from Chain-of-Thought
+
+- **ID:** `2603.05488` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Siddharth Boppana, Annabel Ma, Max Loeffler, et al.
+- **abs:** https://arxiv.org/abs/2603.05488 · **PDF:** https://arxiv.org/pdf/2603.05488v3
+
+## 7396. Bias In, Bias Out? Finding Unbiased Subnetworks in Vanilla Models
+
+- **ID:** `2603.05582` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Ivan Luiz De Moura Matos, Abdel Djalil Sad Saoud, Ekaterina Iakovleva, et al.
+- **abs:** https://arxiv.org/abs/2603.05582 · **PDF:** https://arxiv.org/pdf/2603.05582v1
+
+## 7397. Towards Provably Unbiased LLM Judges via Bias-Bounded Evaluation
+
+- **ID:** `2603.05485` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Benjamin Feuer, Lucas Rosenblatt, Oussama Elachqar
+- **abs:** https://arxiv.org/abs/2603.05485 · **PDF:** https://arxiv.org/pdf/2603.05485v1
+
+## 7398. SurvHTE-Bench: A Benchmark for Heterogeneous Treatment Effect Estimation in Survival Analysis
+
+- **ID:** `2603.05483` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Shahriar Noroozizadeh, Xiaobin Shen, Jeremy C. Weiss, et al.
+- **abs:** https://arxiv.org/abs/2603.05483 · **PDF:** https://arxiv.org/pdf/2603.05483v1
+
+## 7399. Thermodynamic Response Functions in Singular Bayesian Models
+
+- **ID:** `2603.05480` · **primary:** `stat.ML` · **Submitted:** 2026-03-05
+- **Authors:** Sean Plummer
+- **abs:** https://arxiv.org/abs/2603.05480 · **PDF:** https://arxiv.org/pdf/2603.05480v1
+
+## 7400. Spatiotemporal Heterogeneity of AI-Driven Traffic Flow Patterns and Land Use Interaction: A GeoAI-Based Analysis of Multimodal Urban Mobility
+
+- **ID:** `2603.05581` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Olaf Yunus Laitinen Imanov
+- **abs:** https://arxiv.org/abs/2603.05581 · **PDF:** https://arxiv.org/pdf/2603.05581v1
+
+## 7401. Leveraging LLM Parametric Knowledge for Fact Checking without Retrieval
+
+- **ID:** `2603.05471` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Artem Vazhentsev, Maria Marina, Daniil Moskovskiy, et al.
+- **abs:** https://arxiv.org/abs/2603.05471 · **PDF:** https://arxiv.org/pdf/2603.05471v1
+
+## 7402. Kraus Constrained Sequence Learning For Quantum Trajectories from Continuous Measurement
+
+- **ID:** `2603.05468` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Priyanshi Singh, Krishna Bhatia
+- **abs:** https://arxiv.org/abs/2603.05468 · **PDF:** https://arxiv.org/pdf/2603.05468v1
+
+## 7403. NCTB-QA: A Large-Scale Bangla Educational Question Answering Dataset and Benchmarking Performance
+
+- **ID:** `2603.05462` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Abrar Eyasir, Tahsin Ahmed, Muhammad Ibrahim
+- **abs:** https://arxiv.org/abs/2603.05462 · **PDF:** https://arxiv.org/pdf/2603.05462v1
+
+## 7404. DEBISS: a Corpus of Individual, Semi-structured and Spoken Debates
+
+- **ID:** `2603.05459` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Klaywert Danillo Ferreira de Souza, David Eduardo Pereira, Cláudio E. C. Campelo, et al.
+- **abs:** https://arxiv.org/abs/2603.05459 · **PDF:** https://arxiv.org/pdf/2603.05459v1
+
+## 7405. FlashAttention-4: Algorithm and Kernel Pipelining Co-Design for Asymmetric Hardware Scaling
+
+- **ID:** `2603.05451` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Ted Zadouri, Markus Hoehnerbach, Jay Shah, et al.
+- **abs:** https://arxiv.org/abs/2603.05451 · **PDF:** https://arxiv.org/pdf/2603.05451v1
+
+## 7406. Distributed Partial Information Puzzles: Examining Common Ground Construction Under Epistemic Asymmetry
+
+- **ID:** `2603.05450` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Yifan Zhu, Mariah Bradford, Kenneth Lai, et al.
+- **abs:** https://arxiv.org/abs/2603.05450 · **PDF:** https://arxiv.org/pdf/2603.05450v1
+
+## 7407. RealWonder: Real-Time Physical Action-Conditioned Video Generation
+
+- **ID:** `2603.05449` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Wei Liu, Ziyu Chen, Zizhang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.05449 · **PDF:** https://arxiv.org/pdf/2603.05449v1
+
+## 7408. Residual RL--MPC for Robust Microrobotic Cell Pushing Under Time-Varying Flow
+
+- **ID:** `2603.05448` · **primary:** `cs.RO` · **Submitted:** 2026-03-05
+- **Authors:** Yanda Yang, Sambeeta Das
+- **abs:** https://arxiv.org/abs/2603.05448 · **PDF:** https://arxiv.org/pdf/2603.05448v1
+
+## 7409. Latent Wasserstein Adversarial Imitation Learning
+
+- **ID:** `2603.05440` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Siqi Yang, Kai Yan, Alexander G. Schwing, et al.
+- **abs:** https://arxiv.org/abs/2603.05440 · **PDF:** https://arxiv.org/pdf/2603.05440v1
+
+## 7410. Planning in 8 Tokens: A Compact Discrete Tokenizer for Latent World Model
+
+- **ID:** `2603.05438` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Dongwon Kim, Gawon Seo, Jinsung Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.05438 · **PDF:** https://arxiv.org/pdf/2603.05438v1
+
+## 7411. SAIL: Similarity-Aware Guidance and Inter-Caption Augmentation-based Learning for Weakly-Supervised Dense Video Captioning
+
+- **ID:** `2603.05437` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Ye-Chan Kim, SeungJu Cha, Si-Woo Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.05437 · **PDF:** https://arxiv.org/pdf/2603.05437v2
+
+## 7412. On-Policy Self-Distillation for Reasoning Compression
+
+- **ID:** `2603.05433` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Hejian Sang, Yuanda Xu, Zhengze Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.05433 · **PDF:** https://arxiv.org/pdf/2603.05433v4
+
+## 7413. Ensembling Language Models with Sequential Monte Carlo
+
+- **ID:** `2603.05432` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Robin Shing Moon Chan, Tianyu Liu, Samuel Kiegeland, et al.
+- **abs:** https://arxiv.org/abs/2603.05432 · **PDF:** https://arxiv.org/pdf/2603.05432v1
+
+## 7414. A Novel Hybrid Heuristic-Reinforcement Learning Optimization Approach for a Class of Railcar Shunting Problems
+
+- **ID:** `2603.05579` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Ruonan Zhao, Joseph Geunes
+- **abs:** https://arxiv.org/abs/2603.05579 · **PDF:** https://arxiv.org/pdf/2603.05579v1
+
+## 7415. RelaxFlow: Text-Driven Amodal 3D Generation
+
+- **ID:** `2603.05425` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Jiayin Zhu, Guoji Fu, Xiaolu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.05425 · **PDF:** https://arxiv.org/pdf/2603.05425v1
+
+## 7416. Tool-Genesis: A Task-Driven Tool Creation Benchmark for Self-Evolving Language Agent
+
+- **ID:** `2603.05578` · **primary:** `cs.SE` · **Submitted:** 2026-03-05
+- **Authors:** Bowei Xia, Mengkang Hu, Shijian Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.05578 · **PDF:** https://arxiv.org/pdf/2603.05578v1
+
+## 7417. An interpretable prototype parts-based neural network for medical tabular data
+
+- **ID:** `2603.05423` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Jacek Karolczak, Jerzy Stefanowski
+- **abs:** https://arxiv.org/abs/2603.05423 · **PDF:** https://arxiv.org/pdf/2603.05423v1
+
+## 7418. MobileFetalCLIP: Selective Repulsive Knowledge Distillation for Mobile Fetal Ultrasound Analysis
+
+- **ID:** `2603.05421` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Numan Saeed, Fadillah Adamsyah Maani, Mohammad Yaqub
+- **abs:** https://arxiv.org/abs/2603.05421 · **PDF:** https://arxiv.org/pdf/2603.05421v2
+
+## 7419. The Spatial and Temporal Resolution of Motor Intention in Multi-Target Prediction
+
+- **ID:** `2603.05418` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-05
+- **Authors:** Marie Dominique Schmidt, Ioannis Iossifidis
+- **abs:** https://arxiv.org/abs/2603.05418 · **PDF:** https://arxiv.org/pdf/2603.05418v1
+
+## 7420. Dissociating Direct Access from Inference in AI Introspection
+
+- **ID:** `2603.05414` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Harvey Lederman, Kyle Mahowald
+- **abs:** https://arxiv.org/abs/2603.05414 · **PDF:** https://arxiv.org/pdf/2603.05414v1
+
+## 7421. Koopman Regularized Deep Speech Disentanglement for Speaker Verification
+
+- **ID:** `2603.05577` · **primary:** `cs.SD` · **Submitted:** 2026-03-05
+- **Authors:** Nikos Chazaridis, Mohammad Belal, Rafael Mestre, et al.
+- **abs:** https://arxiv.org/abs/2603.05577 · **PDF:** https://arxiv.org/pdf/2603.05577v1
+
+## 7422. Why Grokking Takes So Long: A First-Principles Theory of Representational Phase Transitions
+
+- **ID:** `2603.13331` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Truong Xuan Khanh, Truong Quynh Hoa, Luu Duc Trung, et al.
+- **abs:** https://arxiv.org/abs/2603.13331 · **PDF:** https://arxiv.org/pdf/2603.13331v1
+
+## 7423. An Exploration-Analysis-Disambiguation Reasoning Framework for Word Sense Disambiguation with Low-Parameter LLMs
+
+- **ID:** `2603.05400` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Deshan Sumanathilaka, Nicholas Micallef, Julian Hough
+- **abs:** https://arxiv.org/abs/2603.05400 · **PDF:** https://arxiv.org/pdf/2603.05400v1
+
+## 7424. Judge Reliability Harness: Stress Testing the Reliability of LLM Judges
+
+- **ID:** `2603.05399` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Sunishchal Dev, Andrew Sloan, Joshua Kavner, et al.
+- **abs:** https://arxiv.org/abs/2603.05399 · **PDF:** https://arxiv.org/pdf/2603.05399v1
+
+## 7425. Harnessing Synthetic Data from Generative AI for Statistical Inference
+
+- **ID:** `2603.05396` · **primary:** `stat.ML` · **Submitted:** 2026-03-05
+- **Authors:** Ahmad Abdel-Azim, Ruoyu Wang, Xihong Lin
+- **abs:** https://arxiv.org/abs/2603.05396 · **PDF:** https://arxiv.org/pdf/2603.05396v1
+
+## 7426. On the Necessity of Learnable Sheaf Laplacians
+
+- **ID:** `2603.05395` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Ferran Hernandez Caralt, Mar Gonzàlez i Català, Adrián Bazaga, et al.
+- **abs:** https://arxiv.org/abs/2603.05395 · **PDF:** https://arxiv.org/pdf/2603.05395v1
+
+## 7427. Legal interpretation and AI: from expert systems to argumentation and LLMs
+
+- **ID:** `2603.05392` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Václav Janeček, Giovanni Sartor
+- **abs:** https://arxiv.org/abs/2603.05392 · **PDF:** https://arxiv.org/pdf/2603.05392v1
+
+## 7428. Prediction-Powered Conditional Inference
+
+- **ID:** `2603.05575` · **primary:** `stat.ML` · **Submitted:** 2026-03-05
+- **Authors:** Yang Sui, Jin Zhou, Hua Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.05575 · **PDF:** https://arxiv.org/pdf/2603.05575v1
+
+## 7429. Compact Lifted Relaxations for Low-Rank Optimization
+
+- **ID:** `2603.20228` · **primary:** `math.OC` · **Submitted:** 2026-03-05
+- **Authors:** Ryan Cory-Wright, Jean Pauphilet
+- **abs:** https://arxiv.org/abs/2603.20228 · **PDF:** https://arxiv.org/pdf/2603.20228v1
+
+## 7430. PRISM: Personalized Refinement of Imitation Skills for Manipulation via Human Instructions
+
+- **ID:** `2603.05574` · **primary:** `cs.RO` · **Submitted:** 2026-03-05
+- **Authors:** Arnau Boix-Granell, Alberto San-Miguel-Tello, Magí Dalmau-Moreno, et al.
+- **abs:** https://arxiv.org/abs/2603.05574 · **PDF:** https://arxiv.org/pdf/2603.05574v1
+
+## 7431. Robust Node Affinities via Jaccard-Biased Random Walks and Rank Aggregation
+
+- **ID:** `2603.05375` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Bastian Pfeifer, Michael G. Schimek
+- **abs:** https://arxiv.org/abs/2603.05375 · **PDF:** https://arxiv.org/pdf/2603.05375v1
+
+## 7432. Embedded Inter-Subject Variability in Adversarial Learning for Inertial Sensor-Based Human Activity Recognition
+
+- **ID:** `2603.05371` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Francisco M. Calatrava-Nicolás, Shoko Miyauchi, Vitor Fortes Rey, et al.
+- **abs:** https://arxiv.org/abs/2603.05371 · **PDF:** https://arxiv.org/pdf/2603.05371v1
+
+## 7433. Learning Causal Structure of Time Series using Best Order Score Search
+
+- **ID:** `2603.05370` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Irene Gema Castillo Mansilla, Urmi Ninad
+- **abs:** https://arxiv.org/abs/2603.05370 · **PDF:** https://arxiv.org/pdf/2603.05370v1
+
+## 7434. Progressive Residual Warmup for Language Model Pretraining
+
+- **ID:** `2603.05369` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Tianhao Chen, Xin Xu, Lu Yin, et al.
+- **abs:** https://arxiv.org/abs/2603.05369 · **PDF:** https://arxiv.org/pdf/2603.05369v1
+
+## 7435. PACE: A Personalized Adaptive Curriculum Engine for 9-1-1 Call-taker Training
+
+- **ID:** `2603.05361` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Zirong Chen, Hongchao Zhang, Meiyi Ma
+- **abs:** https://arxiv.org/abs/2603.05361 · **PDF:** https://arxiv.org/pdf/2603.05361v1
+
+## 7436. DiSCTT: Consensus-Guided Self-Curriculum for Efficient Test-Time Adaptation in Reasoning
+
+- **ID:** `2603.05357` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Mohammad Mahdi Moradi, Sudhir Mudur
+- **abs:** https://arxiv.org/abs/2603.05357 · **PDF:** https://arxiv.org/pdf/2603.05357v1
+
+## 7437. Exploring the potential and limitations of Model Merging for Multi-Domain Adaptation in ASR
+
+- **ID:** `2603.05354` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Carlos Carvalho, Francisco Teixeira, Thomas Rolland, et al.
+- **abs:** https://arxiv.org/abs/2603.05354 · **PDF:** https://arxiv.org/pdf/2603.05354v1
+
+## 7438. InfoFlow KV: Information-Flow-Aware KV Recomputation for Long Context
+
+- **ID:** `2603.05353` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Xin Teng, Canyu Zhang, Shaoyi Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.05353 · **PDF:** https://arxiv.org/pdf/2603.05353v1
+
+## 7439. Ailed: A Psyche-Driven Chess Engine with Dynamic Emotional Modulation
+
+- **ID:** `2603.05352` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Diego Armando Resendez Prado
+- **abs:** https://arxiv.org/abs/2603.05352 · **PDF:** https://arxiv.org/pdf/2603.05352v1
+
+## 7440. A Multilingual Human Annotated Corpus of Original and Easy-to-Read Texts to Support Access to Democratic Participatory Processes
+
+- **ID:** `2603.05345` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Stefan Bott, Verena Riegler, Horacio Saggion, et al.
+- **abs:** https://arxiv.org/abs/2603.05345 · **PDF:** https://arxiv.org/pdf/2603.05345v2
+
+## 7441. Building Effective AI Coding Agents for the Terminal: Scaffolding, Harness, Context Engineering, and Lessons Learned
+
+- **ID:** `2603.05344` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Nghi D. Q. Bui
+- **abs:** https://arxiv.org/abs/2603.05344 · **PDF:** https://arxiv.org/pdf/2603.05344v3
+
+## 7442. Preserving Continuous Symmetry in Discrete Spaces: Geometric-Aware Quantization for SO(3)-Equivariant GNNs
+
+- **ID:** `2603.05343` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Haoyu Zhou, Ping Xue, Hao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.05343 · **PDF:** https://arxiv.org/pdf/2603.05343v2
+
+## 7443. On the Statistical Optimality of Optimal Decision Trees
+
+- **ID:** `2603.05340` · **primary:** `stat.ML` · **Submitted:** 2026-03-05
+- **Authors:** Zineng Xu, Subhro Ghosh, Yan Shuo Tan
+- **abs:** https://arxiv.org/abs/2603.05340 · **PDF:** https://arxiv.org/pdf/2603.05340v2
+
+## 7444. Why Depth Matters in Parallelizable Sequence Models: A Lie Algebraic View
+
+- **ID:** `2603.05573` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Gyuryang Heo, Timothy Ngotiaoco, Kazuki Irie, et al.
+- **abs:** https://arxiv.org/abs/2603.05573 · **PDF:** https://arxiv.org/pdf/2603.05573v1
+
+## 7445. Bayes with No Shame: Admissibility Geometries of Predictive Inference
+
+- **ID:** `2603.05335` · **primary:** `stat.ML` · **Submitted:** 2026-03-05
+- **Authors:** Nicholas G. Polson, Daniel Zantedeschi
+- **abs:** https://arxiv.org/abs/2603.05335 · **PDF:** https://arxiv.org/pdf/2603.05335v2
+
+## 7446. FairFinGAN: Fairness-aware Synthetic Financial Data Generation
+
+- **ID:** `2603.05327` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Tai Le Quy, Dung Nguyen Tuan, Trung Nguyen Thanh, et al.
+- **abs:** https://arxiv.org/abs/2603.05327 · **PDF:** https://arxiv.org/pdf/2603.05327v1
+
+## 7447. Machine Learning for analysis of Multiple Sclerosis cross-tissue bulk and single-cell transcriptomics data
+
+- **ID:** `2603.05572` · **primary:** `q-bio.GN` · **Submitted:** 2026-03-05
+- **Authors:** Francesco Massafra, Samuele Punzo, Silvia Giulia Galfré, et al.
+- **abs:** https://arxiv.org/abs/2603.05572 · **PDF:** https://arxiv.org/pdf/2603.05572v1
+
+## 7448. GALACTIC: Global and Local Agnostic Counterfactuals for Time-series Clustering
+
+- **ID:** `2603.05318` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Christos Fragkathoulas, Eleni Psaroudaki, Themis Palpanas, et al.
+- **abs:** https://arxiv.org/abs/2603.05318 · **PDF:** https://arxiv.org/pdf/2603.05318v2
+
+## 7449. How important are the genes to explain the outcome - the asymmetric Shapley value as an honest importance metric for high-dimensional features
+
+- **ID:** `2603.05317` · **primary:** `stat.ML` · **Submitted:** 2026-03-05
+- **Authors:** Mark A. van de Wiel, Jeroen Goedhart, Martin Jullum, et al.
+- **abs:** https://arxiv.org/abs/2603.05317 · **PDF:** https://arxiv.org/pdf/2603.05317v1
+
+## 7450. PersianPunc: A Large-Scale Dataset and BERT-Based Approach for Persian Punctuation Restoration
+
+- **ID:** `2603.05314` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Mohammad Javad Ranjbar Kalahroodi, Heshaam Faili, Azadeh Shakery
+- **abs:** https://arxiv.org/abs/2603.05314 · **PDF:** https://arxiv.org/pdf/2603.05314v1
+
+## 7451. Latent-Mark: An Audio Watermark Robust to Neural Resynthesis
+
+- **ID:** `2603.05310` · **primary:** `cs.SD` · **Submitted:** 2026-03-05
+- **Authors:** Yen-Shan Chen, Shih-Yu Lai, Ying-Jung Tsou, et al.
+- **abs:** https://arxiv.org/abs/2603.05310 · **PDF:** https://arxiv.org/pdf/2603.05310v2
+
+## 7452. Med-V1: Small Language Models for Zero-shot and Scalable Biomedical Evidence Attribution
+
+- **ID:** `2603.05308` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Qiao Jin, Yin Fang, Lauren He, et al.
+- **abs:** https://arxiv.org/abs/2603.05308 · **PDF:** https://arxiv.org/pdf/2603.05308v1
+
+## 7453. UniSTOK: Uniform Inductive Spatio-Temporal Kriging
+
+- **ID:** `2603.05301` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Lewei Xie, Haoyu Zhang, Juan Yuan, et al.
+- **abs:** https://arxiv.org/abs/2603.05301 · **PDF:** https://arxiv.org/pdf/2603.05301v1
+
+## 7454. WavSLM: Single-Stream Speech Language Modeling via WavLM Distillation
+
+- **ID:** `2603.05299` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Luca Della Libera, Cem Subakan, Mirco Ravanelli
+- **abs:** https://arxiv.org/abs/2603.05299 · **PDF:** https://arxiv.org/pdf/2603.05299v1
+
+## 7455. Latent Policy Steering through One-Step Flow Policies
+
+- **ID:** `2603.05296` · **primary:** `cs.RO` · **Submitted:** 2026-03-05
+- **Authors:** Hokyun Im, Andrey Kolobov, Jianlong Fu, et al.
+- **abs:** https://arxiv.org/abs/2603.05296 · **PDF:** https://arxiv.org/pdf/2603.05296v1
+
+## 7456. WebChain: A Large-Scale Human-Annotated Dataset of Real-World Web Interaction Traces
+
+- **ID:** `2603.05295` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Sicheng Fan, Rui Wan, Yifei Leng, et al.
+- **abs:** https://arxiv.org/abs/2603.05295 · **PDF:** https://arxiv.org/pdf/2603.05295v2
+
+## 7457. STRUCTUREDAGENT: Planning with AND/OR Trees for Long-Horizon Web Tasks
+
+- **ID:** `2603.05294` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** ELita Lobo, Xu Chen, Jingjing Meng, et al.
+- **abs:** https://arxiv.org/abs/2603.05294 · **PDF:** https://arxiv.org/pdf/2603.05294v2
+
+## 7458. Knowledge Divergence and the Value of Debate for Scalable Oversight
+
+- **ID:** `2603.05293` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Robin Young
+- **abs:** https://arxiv.org/abs/2603.05293 · **PDF:** https://arxiv.org/pdf/2603.05293v1
+
+## 7459. X-RAY: Mapping LLM Reasoning Capability via Formalized and Calibrated Probes
+
+- **ID:** `2603.05290` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Gao Tianxi, Cai Yufan, Yuan Yusi, et al.
+- **abs:** https://arxiv.org/abs/2603.05290 · **PDF:** https://arxiv.org/pdf/2603.05290v1
+
+## 7460. Bayesian Supervised Causal Clustering
+
+- **ID:** `2603.05288` · **primary:** `stat.ML` · **Submitted:** 2026-03-05
+- **Authors:** Luwei Wang, Nazir Lone, Sohan Seth
+- **abs:** https://arxiv.org/abs/2603.05288 · **PDF:** https://arxiv.org/pdf/2603.05288v1
+
+## 7461. Layer by layer, module by module: Choose both for optimal OOD probing of ViT
+
+- **ID:** `2603.05280` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Ambroise Odonnat, Vasilii Feofanov, Laetitia Chapel, et al.
+- **abs:** https://arxiv.org/abs/2603.05280 · **PDF:** https://arxiv.org/pdf/2603.05280v1
+
+## 7462. Whispering to a Blackbox: Bootstrapping Frozen OCR with Visual Prompts
+
+- **ID:** `2603.05276` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Samandar Samandarov, Nazirjon Ismoiljonov, Abdullah Sattorov, et al.
+- **abs:** https://arxiv.org/abs/2603.05276 · **PDF:** https://arxiv.org/pdf/2603.05276v1
+
+## 7463. SarcasmMiner: A Dual-Track Post-Training Framework for Robust Audio-Visual Sarcasm Reasoning
+
+- **ID:** `2603.05275` · **primary:** `cs.MM` · **Submitted:** 2026-03-05
+- **Authors:** Zhu Li, Yongjian Chen, Huiyuan Lai, et al.
+- **abs:** https://arxiv.org/abs/2603.05275 · **PDF:** https://arxiv.org/pdf/2603.05275v1
+
+## 7464. Oral to Web: Digitizing 'Zero Resource'Languages of Bangladesh
+
+- **ID:** `2603.05272` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Mohammad Mamun Or Rashid
+- **abs:** https://arxiv.org/abs/2603.05272 · **PDF:** https://arxiv.org/pdf/2603.05272v2
+
+## 7465. Visual-Informed Speech Enhancement Using Attention-Based Beamforming
+
+- **ID:** `2603.05270` · **primary:** `eess.AS` · **Submitted:** 2026-03-05
+- **Authors:** Chihyun Liu, Jiaxuan Fan, Mingtung Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.05270 · **PDF:** https://arxiv.org/pdf/2603.05270v1
+
+## 7466. Beyond Word Error Rate: Auditing the Diversity Tax in Speech Recognition through Dataset Cartography
+
+- **ID:** `2603.05267` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Ting-Hui Cheng, Line H. Clemmensen, Sneha Das
+- **abs:** https://arxiv.org/abs/2603.05267 · **PDF:** https://arxiv.org/pdf/2603.05267v1
+
+## 7467. A Behaviour-Aware Federated Forecasting Framework for Distributed Stand-Alone Wind Turbines
+
+- **ID:** `2603.05263` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Bowen Li, Xiufeng Liu, Maria Sinziiana Astefanoaei
+- **abs:** https://arxiv.org/abs/2603.05263 · **PDF:** https://arxiv.org/pdf/2603.05263v1
+
+## 7468. VietJobs: A Vietnamese Job Advertisement Dataset
+
+- **ID:** `2603.05262` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Hieu Pham Dinh, Hung Nguyen Huy, Mo El-Haj
+- **abs:** https://arxiv.org/abs/2603.05262 · **PDF:** https://arxiv.org/pdf/2603.05262v1
+
+## 7469. GCAgent: Enhancing Group Chat Communication through Dialogue Agents System
+
+- **ID:** `2603.05240` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Zijie Meng, Zheyong Xie, Zheyu Ye, et al.
+- **abs:** https://arxiv.org/abs/2603.05240 · **PDF:** https://arxiv.org/pdf/2603.05240v1
+
+## 7470. Reclaiming Lost Text Layers for Source-Free Cross-Domain Few-Shot Learning
+
+- **ID:** `2603.05235` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Zhenyu Zhang, Guangyao Chen, Yixiong Zou, et al.
+- **abs:** https://arxiv.org/abs/2603.05235 · **PDF:** https://arxiv.org/pdf/2603.05235v1
+
+## 7471. Recursive Inference Machines for Neural Reasoning
+
+- **ID:** `2603.05234` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Mieszko Komisarczyk, Saurabh Mathur, Maurice Kraus, et al.
+- **abs:** https://arxiv.org/abs/2603.05234 · **PDF:** https://arxiv.org/pdf/2603.05234v1
+
+## 7472. SlideSparse: Fast and Flexible (2N-2):2N Structured Sparsity
+
+- **ID:** `2603.05232` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Hanyong Shao, Yingbo Hao, Ting Song, et al.
+- **abs:** https://arxiv.org/abs/2603.05232 · **PDF:** https://arxiv.org/pdf/2603.05232v1
+
+## 7473. On the Generalization Capacities of MLLMs for Spatial Intelligence
+
+- **ID:** `2603.06704` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Gongjie Zhang, Wenhao Li, Quanhao Qian, et al.
+- **abs:** https://arxiv.org/abs/2603.06704 · **PDF:** https://arxiv.org/pdf/2603.06704v1
+
+## 7474. Boosting ASR Robustness via Test-Time Reinforcement Learning with Audio-Text Semantic Rewards
+
+- **ID:** `2603.05231` · **primary:** `cs.SD` · **Submitted:** 2026-03-05
+- **Authors:** Linghan Fang, Tianxin Xie, Li Liu
+- **abs:** https://arxiv.org/abs/2603.05231 · **PDF:** https://arxiv.org/pdf/2603.05231v1
+
+## 7475. CBR-to-SQL: Rethinking Retrieval-based Text-to-SQL using Case-based Reasoning in the Healthcare Domain
+
+- **ID:** `2603.05569` · **primary:** `cs.IR` · **Submitted:** 2026-03-05
+- **Authors:** Hung Nguyen, Hans Moen, Pekka Marttinen
+- **abs:** https://arxiv.org/abs/2603.05569 · **PDF:** https://arxiv.org/pdf/2603.05569v1
+
+## 7476. Not All Trust is the Same: Effects of Decision Workflow and Explanations in Human-AI Decision Making
+
+- **ID:** `2603.05229` · **primary:** `cs.HC` · **Submitted:** 2026-03-05
+- **Authors:** Laura Spillner, Rachel Ringe, Robert Porzel, et al.
+- **abs:** https://arxiv.org/abs/2603.05229 · **PDF:** https://arxiv.org/pdf/2603.05229v1
+
+## 7477. The Geometric Inductive Bias of Grokking: Bypassing Phase Transitions via Architectural Topology
+
+- **ID:** `2603.05228` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Alper Yıldırım
+- **abs:** https://arxiv.org/abs/2603.05228 · **PDF:** https://arxiv.org/pdf/2603.05228v2
+
+## 7478. Learning Optimal Individualized Decision Rules with Conditional Demographic Parity
+
+- **ID:** `2603.05226` · **primary:** `stat.ML` · **Submitted:** 2026-03-05
+- **Authors:** Wenhai Cui, Wen Su, Donglin Zeng, et al.
+- **abs:** https://arxiv.org/abs/2603.05226 · **PDF:** https://arxiv.org/pdf/2603.05226v1
+
+## 7479. Understanding the Relationship Between Firms' AI Technology Innovation and Consumer Complaints
+
+- **ID:** `2603.18025` · **primary:** `cs.CY` · **Submitted:** 2026-03-05
+- **Authors:** Yongchao Martin Ma, Zhongzhun Deng
+- **abs:** https://arxiv.org/abs/2603.18025 · **PDF:** https://arxiv.org/pdf/2603.18025v1
+
+## 7480. AI+HW 2035: Shaping the Next Decade
+
+- **ID:** `2603.05225` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Deming Chen, Jason Cong, Azalia Mirhoseini, et al.
+- **abs:** https://arxiv.org/abs/2603.05225 · **PDF:** https://arxiv.org/pdf/2603.05225v1
+
+## 7481. Learning Optimal Distributionally Robust Individualized Treatment Rules Integrating Multi-Source Data
+
+- **ID:** `2603.05568` · **primary:** `stat.ML` · **Submitted:** 2026-03-05
+- **Authors:** Wenhai Cui, Wen Su, Xingqiu Zhao
+- **abs:** https://arxiv.org/abs/2603.05568 · **PDF:** https://arxiv.org/pdf/2603.05568v1
+
+## 7482. SPyCer: Semi-Supervised Physics-Guided Contextual Attention for Near-Surface Air Temperature Estimation from Satellite Imagery
+
+- **ID:** `2603.05219` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Sofiane Bouaziz, Adel Hafiane, Raphael Canals, et al.
+- **abs:** https://arxiv.org/abs/2603.05219 · **PDF:** https://arxiv.org/pdf/2603.05219v1
+
+## 7483. KARL: Knowledge Agents via Reinforcement Learning
+
+- **ID:** `2603.05218` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Jonathan D. Chang, Andrew Drozdov, Shubham Toshniwal, et al.
+- **abs:** https://arxiv.org/abs/2603.05218 · **PDF:** https://arxiv.org/pdf/2603.05218v1
+
+## 7484. Diagnosing FP4 inference: a layer-wise and block-wise sensitivity analysis of NVFP4 and MXFP4
+
+- **ID:** `2603.08747` · **primary:** `cs.AR` · **Submitted:** 2026-03-05
+- **Authors:** Musa Cim, Burak Topcu, Mahmut Taylan Kandemir
+- **abs:** https://arxiv.org/abs/2603.08747 · **PDF:** https://arxiv.org/pdf/2603.08747v1
+
+## 7485. FuseDiff: Symmetry-Preserving Joint Diffusion for Dual-Target Structure-Based Drug Design
+
+- **ID:** `2603.05567` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Jianliang Wu, Anjie Qiao, Zhen Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.05567 · **PDF:** https://arxiv.org/pdf/2603.05567v1
+
+## 7486. S-Path-RAG: Semantic-Aware Shortest-Path Retrieval Augmented Generation for Multi-Hop Knowledge Graph Question Answering
+
+- **ID:** `2603.23512` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Rong Fu, Yemin Wang, Tianxiang Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.23512 · **PDF:** https://arxiv.org/pdf/2603.23512v1
+
+## 7487. Early Warning of Intraoperative Adverse Events via Transformer-Driven Multi-Label Learning
+
+- **ID:** `2603.05212` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Xueyao Wang, Xiuding Cai, Honglin Shang, et al.
+- **abs:** https://arxiv.org/abs/2603.05212 · **PDF:** https://arxiv.org/pdf/2603.05212v1
+
+## 7488. Balancing Coverage and Draft Latency in Vocabulary Trimming for Faster Speculative Decoding
+
+- **ID:** `2603.05210` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Ofir Ben Shoham
+- **abs:** https://arxiv.org/abs/2603.05210 · **PDF:** https://arxiv.org/pdf/2603.05210v1
+
+## 7489. Core-based Hierarchies for Efficient GraphRAG
+
+- **ID:** `2603.05207` · **primary:** `cs.IR` · **Submitted:** 2026-03-05
+- **Authors:** Jakir Hossain, Ahmet Erdem Sarıyüce
+- **abs:** https://arxiv.org/abs/2603.05207 · **PDF:** https://arxiv.org/pdf/2603.05207v1
+
+## 7490. Stable-LoRA: Stabilizing Feature Learning of Low-Rank Adaptation
+
+- **ID:** `2603.05204` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Yize Wu, Ke Gao, Ling Li, et al.
+- **abs:** https://arxiv.org/abs/2603.05204 · **PDF:** https://arxiv.org/pdf/2603.05204v1
+
+## 7491. Towards a data-scale independent regulariser for robust sparse identification of non-linear dynamics
+
+- **ID:** `2603.05201` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Jay Raut, Daniel N. Wilke, Stephan Schmidt
+- **abs:** https://arxiv.org/abs/2603.05201 · **PDF:** https://arxiv.org/pdf/2603.05201v1
+
+## 7492. Distilling Formal Logic into Neural Spaces: A Kernel Alignment Approach for Signal Temporal Logic
+
+- **ID:** `2603.05198` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Sara Candussio, Gabriele Sarti, Gaia Saveri, et al.
+- **abs:** https://arxiv.org/abs/2603.05198 · **PDF:** https://arxiv.org/pdf/2603.05198v1
+
+## 7493. Diffusion LLMs can think EoS-by-EoS
+
+- **ID:** `2603.05197` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Sarah Breckner, Sebastian Schuster
+- **abs:** https://arxiv.org/abs/2603.05197 · **PDF:** https://arxiv.org/pdf/2603.05197v1
+
+## 7494. Transducing Language Models
+
+- **ID:** `2603.05193` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Vésteinn Snæbjarnarson, Samuel Kiegeland, Tianyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.05193 · **PDF:** https://arxiv.org/pdf/2603.05193v1
+
+## 7495. ProKWS: Personalized Keyword Spotting via Collaborative Learning of Phonemes and Prosody
+
+- **ID:** `2603.18024` · **primary:** `eess.AS` · **Submitted:** 2026-03-05
+- **Authors:** Jianan Pan, Yuanming Zhang, Kejie Huang
+- **abs:** https://arxiv.org/abs/2603.18024 · **PDF:** https://arxiv.org/pdf/2603.18024v1
+
+## 7496. Escaping the Hydrolysis Trap: An Agentic Workflow for Inverse Design of Durable Photocatalytic Covalent Organic Frameworks
+
+- **ID:** `2603.05188` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-05
+- **Authors:** Iman Peivaste, Nicolas D. Boscher, Ahmed Makradi, et al.
+- **abs:** https://arxiv.org/abs/2603.05188 · **PDF:** https://arxiv.org/pdf/2603.05188v1
+
+## 7497. PCOV-KWS: Multi-task Learning for Personalized Customizable Open Vocabulary Keyword Spotting
+
+- **ID:** `2603.18023` · **primary:** `eess.AS` · **Submitted:** 2026-03-05
+- **Authors:** Jianan Pan, Kejie Huang
+- **abs:** https://arxiv.org/abs/2603.18023 · **PDF:** https://arxiv.org/pdf/2603.18023v1
+
+## 7498. Logi-PAR: Logic-Infused Patient Activity Recognition via Differentiable Rule
+
+- **ID:** `2603.05184` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Muhammad Zarar, MingZheng Zhang, Xiaowang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.05184 · **PDF:** https://arxiv.org/pdf/2603.05184v1
+
+## 7499. Incentive Aware AI Regulations: A Credal Characterisation
+
+- **ID:** `2603.05175` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Anurag Singh, Julian Rodemann, Rajeev Verma, et al.
+- **abs:** https://arxiv.org/abs/2603.05175 · **PDF:** https://arxiv.org/pdf/2603.05175v1
+
+## 7500. Trainable Bitwise Soft Quantization for Input Feature Compression
+
+- **ID:** `2603.05172` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Karsten Schrödter, Jan Stenkamp, Nina Herrmann, et al.
+- **abs:** https://arxiv.org/abs/2603.05172 · **PDF:** https://arxiv.org/pdf/2603.05172v1
+
+## 7501. Guidelines for the Annotation and Visualization of Legal Argumentation Structures in Chinese Judicial Decisions
+
+- **ID:** `2603.05171` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Kun Chen, Xianglei Liao, Kaixue Fei, et al.
+- **abs:** https://arxiv.org/abs/2603.05171 · **PDF:** https://arxiv.org/pdf/2603.05171v1
+
+## 7502. Sparse-BitNet: 1.58-bit LLMs are Naturally Friendly to Semi-Structured Sparsity
+
+- **ID:** `2603.05168` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Di Zhang, Xun Wu, Shaohan Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.05168 · **PDF:** https://arxiv.org/pdf/2603.05168v1
+
+## 7503. C2-Faith: Benchmarking LLM Judges for Causal and Coverage Faithfulness in Chain-of-Thought Reasoning
+
+- **ID:** `2603.05167` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Avni Mittal, Rauno Arike
+- **abs:** https://arxiv.org/abs/2603.05167 · **PDF:** https://arxiv.org/pdf/2603.05167v1
+
+## 7504. A Geometry-Adaptive Deep Variational Framework for Phase Discovery in the Landau-Brazovskii Model
+
+- **ID:** `2603.05161` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-05
+- **Authors:** Yuchen Xie, Jianyuan Yin, Lei Zhang
+- **abs:** https://arxiv.org/abs/2603.05161 · **PDF:** https://arxiv.org/pdf/2603.05161v1
+
+## 7505. Lifelong Language-Conditioned Robotic Manipulation Learning
+
+- **ID:** `2603.05160` · **primary:** `cs.RO` · **Submitted:** 2026-03-05
+- **Authors:** Xudong Wang, Zebin Han, Zhiyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.05160 · **PDF:** https://arxiv.org/pdf/2603.05160v1
+
+## 7506. Balancing Privacy-Quality-Efficiency in Federated Learning through Round-Based Interleaving of Protection Techniques
+
+- **ID:** `2603.05158` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Yenan Wang, Carla Fabiana Chiasserini, Elad Michael Schiller
+- **abs:** https://arxiv.org/abs/2603.05158 · **PDF:** https://arxiv.org/pdf/2603.05158v1
+
+## 7507. The Impact of Preprocessing Methods on Racial Encoding and Model Robustness in CXR Diagnosis
+
+- **ID:** `2603.05157` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Dishantkumar Sutariya, Eike Petersen
+- **abs:** https://arxiv.org/abs/2603.05157 · **PDF:** https://arxiv.org/pdf/2603.05157v1
+
+## 7508. SSR-GS: Separating Specular Reflection in Gaussian Splatting for Glossy Surface Reconstruction
+
+- **ID:** `2603.05152` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Ningjing Fan, Yiqun Wang
+- **abs:** https://arxiv.org/abs/2603.05152 · **PDF:** https://arxiv.org/pdf/2603.05152v1
+
+## 7509. Aligning the True Semantics: Constrained Decoupling and Distribution Sampling for Cross-Modal Alignment
+
+- **ID:** `2603.05566` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Xiang Ma, Lexin Fang, Litian Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.05566 · **PDF:** https://arxiv.org/pdf/2603.05566v1
+
+## 7510. Federated Causal Discovery Across Heterogeneous Datasets under Latent Confounding
+
+- **ID:** `2603.05149` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Maximilian Hahn, Alina Zajak, Dominik Heider, et al.
+- **abs:** https://arxiv.org/abs/2603.05149 · **PDF:** https://arxiv.org/pdf/2603.05149v1
+
+## 7511. Feature Resemblance: Towards a Theoretical Understanding of Analogical Reasoning in Transformers
+
+- **ID:** `2603.05143` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Ruichen Xu, Wenjing Yan, Ying-Jun Angela Zhang
+- **abs:** https://arxiv.org/abs/2603.05143 · **PDF:** https://arxiv.org/pdf/2603.05143v2
+
+## 7512. Recurrent Graph Neural Networks and Arithmetic Circuits
+
+- **ID:** `2603.05140` · **primary:** `cs.CC` · **Submitted:** 2026-03-05
+- **Authors:** Timon Barlag, Vivian Holzapfel, Laura Strieker, et al.
+- **abs:** https://arxiv.org/abs/2603.05140 · **PDF:** https://arxiv.org/pdf/2603.05140v1
+
+## 7513. Particle-Guided Diffusion for Gas-Phase Reaction Kinetics
+
+- **ID:** `2603.05139` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-05
+- **Authors:** Andrew Millard, Henrik Pedersen
+- **abs:** https://arxiv.org/abs/2603.05139 · **PDF:** https://arxiv.org/pdf/2603.05139v1
+
+## 7514. Representation Fidelity:Auditing Algorithmic Decisions About Humans Using Self-Descriptions
+
+- **ID:** `2603.05136` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Theresa Elstner, Martin Potthast
+- **abs:** https://arxiv.org/abs/2603.05136 · **PDF:** https://arxiv.org/pdf/2603.05136v1
+
+## 7515. SRasP: Self-Reorientation Adversarial Style Perturbation for Cross-Domain Few-Shot Learning
+
+- **ID:** `2603.05135` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Wenqian Li, Pengfei Fang, Hui Xue
+- **abs:** https://arxiv.org/abs/2603.05135 · **PDF:** https://arxiv.org/pdf/2603.05135v1
+
+## 7516. LBM: Hierarchical Large Auto-Bidding Model via Reasoning and Acting
+
+- **ID:** `2603.05134` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Yewen Li, Zhiyi Lyu, Peng Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.05134 · **PDF:** https://arxiv.org/pdf/2603.05134v1
+
+## 7517. MedCoRAG: Interpretable Hepatology Diagnosis via Hybrid Evidence Retrieval and Multispecialty Consensus
+
+- **ID:** `2603.05129` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Zheng Li, Jiayi Xu, Zhikai Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.05129 · **PDF:** https://arxiv.org/pdf/2603.05129v1
+
+## 7518. Measuring the Redundancy of Decoder Layers in SpeechLLMs
+
+- **ID:** `2603.05121` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Adel Moumen, Guangzhi Sun, Philip C Woodland
+- **abs:** https://arxiv.org/abs/2603.05121 · **PDF:** https://arxiv.org/pdf/2603.05121v1
+
+## 7519. Bidirectional Curriculum Generation: A Multi-Agent Framework for Data-Efficient Mathematical Reasoning
+
+- **ID:** `2603.05120` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Boren Hu, Xiao Liu, Boci Peng, et al.
+- **abs:** https://arxiv.org/abs/2603.05120 · **PDF:** https://arxiv.org/pdf/2603.05120v1
+
+## 7520. When AI Levels the Playing Field: Skill Homogenization, Asset Concentration, and Two Regimes of Inequality
+
+- **ID:** `2603.05565` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Xupeng Chen, Shuchen Meng
+- **abs:** https://arxiv.org/abs/2603.05565 · **PDF:** https://arxiv.org/pdf/2603.05565v2
+
+## 7521. FedBCD:Communication-Efficient Accelerated Block Coordinate Gradient Descent for Federated Learning
+
+- **ID:** `2603.05116` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Junkang Liu, Fanhua Shang, Yuanyuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.05116 · **PDF:** https://arxiv.org/pdf/2603.05116v1
+
+## 7522. UniPAR: A Unified Framework for Pedestrian Attribute Recognition
+
+- **ID:** `2603.05114` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Minghe Xu, Rouying Wu, Jiarui Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.05114 · **PDF:** https://arxiv.org/pdf/2603.05114v1
+
+## 7523. Decoupling Task and Behavior: A Two-Stage Reward Curriculum in Reinforcement Learning for Robotics
+
+- **ID:** `2603.05113` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Kilian Freitag, Knut Åkesson, Morteza Haghir Chehreghani
+- **abs:** https://arxiv.org/abs/2603.05113 · **PDF:** https://arxiv.org/pdf/2603.05113v1
+
+## 7524. BLINK: Behavioral Latent Modeling of NK Cell Cytotoxicity
+
+- **ID:** `2603.05110` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Iman Nematollahi, Jose Francisco Villena-Ossa, Alina Moter, et al.
+- **abs:** https://arxiv.org/abs/2603.05110 · **PDF:** https://arxiv.org/pdf/2603.05110v2
+
+## 7525. SPIRIT: Perceptive Shared Autonomy for Robust Robotic Manipulation under Deep Learning Uncertainty
+
+- **ID:** `2603.05111` · **primary:** `cs.RO` · **Submitted:** 2026-03-05
+- **Authors:** Jongseok Lee, Ribin Balachandran, Harsimran Singh, et al.
+- **abs:** https://arxiv.org/abs/2603.05111 · **PDF:** https://arxiv.org/pdf/2603.05111v1
+
+## 7526. ARC-TGI: Human-Validated Task Generators with Reasoning Chain Templates for ARC-AGI
+
+- **ID:** `2603.05099` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Jens Lehmann, Syeda Khushbakht, Nikoo Salehfard, et al.
+- **abs:** https://arxiv.org/abs/2603.05099 · **PDF:** https://arxiv.org/pdf/2603.05099v1
+
+## 7527. GEM-TFL: Bridging Weak and Full Supervision for Forgery Localization through EM-Guided Decomposition and Temporal Refinement
+
+- **ID:** `2603.05095` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Xiaodong Zhu, Yuanming Zheng, Suting Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.05095 · **PDF:** https://arxiv.org/pdf/2603.05095v1
+
+## 7528. Axiomatic On-Manifold Shapley via Optimal Generative Flows
+
+- **ID:** `2603.05093` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Cenwei Zhang, Lin Zhu, Manxi Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.05093 · **PDF:** https://arxiv.org/pdf/2603.05093v1
+
+## 7529. Aura: Universal Multi-dimensional Exogenous Integration for Aviation Time Series
+
+- **ID:** `2603.05092` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Jiafeng Lin, Mengren Zheng, Simeng Ye, et al.
+- **abs:** https://arxiv.org/abs/2603.05092 · **PDF:** https://arxiv.org/pdf/2603.05092v2
+
+## 7530. Model Change for Description Logic Concepts
+
+- **ID:** `2603.05562` · **primary:** `cs.LO` · **Submitted:** 2026-03-05
+- **Authors:** Ana Ozaki, Jandson S. Ribeiro
+- **abs:** https://arxiv.org/abs/2603.05562 · **PDF:** https://arxiv.org/pdf/2603.05562v1
+
+## 7531. Jagarin: A Three-Layer Architecture for Hibernating Personal Duty Agents on Mobile
+
+- **ID:** `2603.05069` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Ravi Kiran Kadaboina
+- **abs:** https://arxiv.org/abs/2603.05069 · **PDF:** https://arxiv.org/pdf/2603.05069v2
+
+## 7532. Cyber Threat Intelligence for Artificial Intelligence Systems
+
+- **ID:** `2603.05068` · **primary:** `cs.CR` · **Submitted:** 2026-03-05
+- **Authors:** Natalia Krawczyk, Mateusz Szczepkowski, Adrian Brodzik, et al.
+- **abs:** https://arxiv.org/abs/2603.05068 · **PDF:** https://arxiv.org/pdf/2603.05068v1
+
+## 7533. Synchronization-based clustering on the unit hypersphere
+
+- **ID:** `2603.05067` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Zinaid Kapić, Aladin Crnkić, Goran Mauša
+- **abs:** https://arxiv.org/abs/2603.05067 · **PDF:** https://arxiv.org/pdf/2603.05067v1
+
+## 7534. Reward-Conditioned Reinforcement Learning
+
+- **ID:** `2603.05066` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Michal Nauman, Marek Cygan, Pieter Abbeel
+- **abs:** https://arxiv.org/abs/2603.05066 · **PDF:** https://arxiv.org/pdf/2603.05066v1
+
+## 7535. Deep Learning-Driven Friendly Jamming for Secure Multicarrier ISAC Under Channel Uncertainty
+
+- **ID:** `2603.05062` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Bui Minh Tuan, Van-Dinh Nguyen, Diep N. Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.05062 · **PDF:** https://arxiv.org/pdf/2603.05062v1
+
+## 7536. Asymptotic Behavior of Multi--Task Learning: Implicit Regularization and Double Descent Effects
+
+- **ID:** `2603.05060` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Ayed M. Alrashdi, Oussama Dhifallah, Houssem Sifaou
+- **abs:** https://arxiv.org/abs/2603.05060 · **PDF:** https://arxiv.org/pdf/2603.05060v1
+
+## 7537. A 360-degree Multi-camera System for Blue Emergency Light Detection Using Color Attention RT-DETR and the ABLDataset
+
+- **ID:** `2603.05058` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Francisco Vacalebri-Lloret, Lucas Banchero, Jose J. Lopez, et al.
+- **abs:** https://arxiv.org/abs/2603.05058 · **PDF:** https://arxiv.org/pdf/2603.05058v1
+
+## 7538. MUTEX: Leveraging Multilingual Transformers and Conditional Random Fields for Enhanced Urdu Toxic Span Detection
+
+- **ID:** `2603.05057` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Inayat Arshad, Fajar Saleem, Ijaz Hussain
+- **abs:** https://arxiv.org/abs/2603.05057 · **PDF:** https://arxiv.org/pdf/2603.05057v1
+
+## 7539. MCEL: Margin-Based Cross-Entropy Loss for Error-Tolerant Quantized Neural Networks
+
+- **ID:** `2603.05048` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Mikail Yayla, Akash Kumar
+- **abs:** https://arxiv.org/abs/2603.05048 · **PDF:** https://arxiv.org/pdf/2603.05048v1
+
+## 7540. NeuronMoE: Neuron-Guided Mixture-of-Experts for Efficient Multilingual LLM Extension
+
+- **ID:** `2603.05046` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Rongzhi Li, Hitomi Yanaka
+- **abs:** https://arxiv.org/abs/2603.05046 · **PDF:** https://arxiv.org/pdf/2603.05046v1
+
+## 7541. WebFactory: Automated Compression of Foundational Language Intelligence into Grounded Web Agents
+
+- **ID:** `2603.05044` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Sicheng Fan, Qingyun Shi, Shengze Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.05044 · **PDF:** https://arxiv.org/pdf/2603.05044v1
+
+## 7542. Enhancing Zero-shot Commonsense Reasoning by Integrating Visual Knowledge via Machine Imagination
+
+- **ID:** `2603.05040` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Hyuntae Park, Yeachan Kim, SangKeun Lee
+- **abs:** https://arxiv.org/abs/2603.05040 · **PDF:** https://arxiv.org/pdf/2603.05040v1
+
+## 7543. Using Laplace Transform To Optimize the Hallucination of Generation Models
+
+- **ID:** `2603.18022` · **primary:** `math.OC` · **Submitted:** 2026-03-05
+- **Authors:** Cheng Kang, Xinye Chen, Daniel Novak, et al.
+- **abs:** https://arxiv.org/abs/2603.18022 · **PDF:** https://arxiv.org/pdf/2603.18022v1
+
+## 7544. The Trilingual Triad Framework: Integrating Design, AI, and Domain Knowledge in No-code AI Smart City Course
+
+- **ID:** `2603.05036` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Qian Huang, King Wang Poon
+- **abs:** https://arxiv.org/abs/2603.05036 · **PDF:** https://arxiv.org/pdf/2603.05036v1
+
+## 7545. Good-Enough LLM Obfuscation (GELO)
+
+- **ID:** `2603.05035` · **primary:** `cs.CR` · **Submitted:** 2026-03-05
+- **Authors:** Anatoly Belikov, Ilya Fedotov
+- **abs:** https://arxiv.org/abs/2603.05035 · **PDF:** https://arxiv.org/pdf/2603.05035v2
+
+## 7546. AegisUI: Behavioral Anomaly Detection for Structured User Interface Protocols in AI Agent Systems
+
+- **ID:** `2603.05031` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Mohd Safwan Uddin, Saba Hajira
+- **abs:** https://arxiv.org/abs/2603.05031 · **PDF:** https://arxiv.org/pdf/2603.05031v1
+
+## 7547. Survive at All Costs: Exploring LLM's Risky Behaviors under Survival Pressure
+
+- **ID:** `2603.05028` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Yida Lu, Jianwei Fang, Xuyang Shao, et al.
+- **abs:** https://arxiv.org/abs/2603.05028 · **PDF:** https://arxiv.org/pdf/2603.05028v1
+
+## 7548. S5-SHB Agent: Society 5.0 enabled Multi-model Agentic Blockchain Framework for Smart Home
+
+- **ID:** `2603.05027` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Janani Rangila, Akila Siriweera, Incheon Paik, et al.
+- **abs:** https://arxiv.org/abs/2603.05027 · **PDF:** https://arxiv.org/pdf/2603.05027v2
+
+## 7549. RepoLaunch: Automating Build&Test Pipeline of Code Repositories on ANY Language and ANY Platform
+
+- **ID:** `2603.05026` · **primary:** `cs.SE` · **Submitted:** 2026-03-05
+- **Authors:** Kenan Li, Rongzhi Li, Linghao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.05026 · **PDF:** https://arxiv.org/pdf/2603.05026v1
+
+## 7550. Measuring the Fragility of Trust: Devising Credibility Index via Explanation Stability (CIES) for Business Decision Support Systems
+
+- **ID:** `2603.05024` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Alin-Gabriel Vaduva, Simona-Vasilica Oprea, Adela Bara
+- **abs:** https://arxiv.org/abs/2603.05024 · **PDF:** https://arxiv.org/pdf/2603.05024v1
+
+## 7551. BioLLMAgent: A Hybrid Framework with Enhanced Structural Interpretability for Simulating Human Decision-Making in Computational Psychiatry
+
+- **ID:** `2603.05016` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Zuo Fei, Kezhi Wang, Xiaomin Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.05016 · **PDF:** https://arxiv.org/pdf/2603.05016v1
+
+## 7552. Poisoning the Inner Prediction Logic of Graph Neural Networks for Clean-Label Backdoor Attacks
+
+- **ID:** `2603.05004` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Yuxiang Zhang, Bin Ma, Enyan Dai
+- **abs:** https://arxiv.org/abs/2603.05004 · **PDF:** https://arxiv.org/pdf/2603.05004v1
+
+## 7553. Non-Euclidean Gradient Descent Operates at the Edge of Stability
+
+- **ID:** `2603.05002` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Rustem Islamov, Michael Crawshaw, Jeremy Cohen, et al.
+- **abs:** https://arxiv.org/abs/2603.05002 · **PDF:** https://arxiv.org/pdf/2603.05002v1
+
+## 7554. Competitive Multi-Operator Reinforcement Learning for Joint Pricing and Fleet Rebalancing in AMoD Systems
+
+- **ID:** `2603.05000` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Emil Kragh Toft, Carolin Schmidt, Daniele Gammelli, et al.
+- **abs:** https://arxiv.org/abs/2603.05000 · **PDF:** https://arxiv.org/pdf/2603.05000v1
+
+## 7555. Lightweight and Scalable Transfer Learning Framework for Load Disaggregation
+
+- **ID:** `2603.04998` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** L. E. Garcia-Marrero, G. Petrone, E. Monmasson
+- **abs:** https://arxiv.org/abs/2603.04998 · **PDF:** https://arxiv.org/pdf/2603.04998v1
+
+## 7556. HiFlow: Hierarchical Feedback-Driven Optimization for Constrained Long-Form Text Generation
+
+- **ID:** `2603.04996` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Yifan Zhu, Guanting Chen, Bing Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.04996 · **PDF:** https://arxiv.org/pdf/2603.04996v1
+
+## 7557. Towards Efficient and Stable Ocean State Forecasting: A Continuous-Time Koopman Approach
+
+- **ID:** `2603.05560` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Rares Grozavescu, Pengyu Zhang, Mark Girolami, et al.
+- **abs:** https://arxiv.org/abs/2603.05560 · **PDF:** https://arxiv.org/pdf/2603.05560v2
+
+## 7558. Autocorrelation effects in a stochastic-process model for decision making via time series
+
+- **ID:** `2603.05559` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Tomoki Yamagami, Mikio Hasegawa, Takatomo Mihana, et al.
+- **abs:** https://arxiv.org/abs/2603.05559 · **PDF:** https://arxiv.org/pdf/2603.05559v1
+
+## 7559. ThaiSafetyBench: Assessing Language Model Safety in Thai Cultural Contexts
+
+- **ID:** `2603.04992` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Trapoom Ukarapol, Nut Chukamphaeng, Kunat Pipatanakul, et al.
+- **abs:** https://arxiv.org/abs/2603.04992 · **PDF:** https://arxiv.org/pdf/2603.04992v2
+
+## 7560. Debiasing Sequential Recommendation with Time-aware Inverse Propensity Scoring
+
+- **ID:** `2603.04986` · **primary:** `cs.IR` · **Submitted:** 2026-03-05
+- **Authors:** Sirui Huang, Jing Long, Qian Li, et al.
+- **abs:** https://arxiv.org/abs/2603.04986 · **PDF:** https://arxiv.org/pdf/2603.04986v1
+
+## 7561. Training for Technology: Adoption and Productive Use of Generative AI in Legal Analysis
+
+- **ID:** `2603.04982` · **primary:** `cs.CY` · **Submitted:** 2026-03-05
+- **Authors:** Benjamin M. Chen, Hong Bao
+- **abs:** https://arxiv.org/abs/2603.04982 · **PDF:** https://arxiv.org/pdf/2603.04982v2
+
+## 7562. Rethinking Representativeness and Diversity in Dynamic Data Selection
+
+- **ID:** `2603.04981` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Yuzhe Zhou, Zhenglin Hua, Haiyun Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.04981 · **PDF:** https://arxiv.org/pdf/2603.04981v1
+
+## 7563. 3D-RFT: Reinforcement Fine-Tuning for Video-based 3D Scene Understanding
+
+- **ID:** `2603.04976` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Xiongkun Linghu, Jiangyong Huang, Baoxiong Jia, et al.
+- **abs:** https://arxiv.org/abs/2603.04976 · **PDF:** https://arxiv.org/pdf/2603.04976v1
+
+## 7564. VRM: Teaching Reward Models to Understand Authentic Human Preferences
+
+- **ID:** `2603.04974` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Biao Liu, Ning Xu, Junming Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.04974 · **PDF:** https://arxiv.org/pdf/2603.04974v1
+
+## 7565. Functionality-Oriented LLM Merging on the Fisher--Rao Manifold
+
+- **ID:** `2603.04972` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Jiayu Wang, Zuojun Ye, Wenpeng Yin
+- **abs:** https://arxiv.org/abs/2603.04972 · **PDF:** https://arxiv.org/pdf/2603.04972v1
+
+## 7566. Mixture of Universal Experts: Scaling Virtual Width via Depth-Width Transformation
+
+- **ID:** `2603.04971` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Yilong Chen, Naibin Gu, Junyuan Shang, et al.
+- **abs:** https://arxiv.org/abs/2603.04971 · **PDF:** https://arxiv.org/pdf/2603.04971v1
+
+## 7567. MPCEval: A Benchmark for Multi-Party Conversation Generation
+
+- **ID:** `2603.04969` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Minxing Zhang, Yi Yang, Zhuofan Jia, et al.
+- **abs:** https://arxiv.org/abs/2603.04969 · **PDF:** https://arxiv.org/pdf/2603.04969v1
+
+## 7568. When Weak LLMs Speak with Confidence, Preference Alignment Gets Stronger
+
+- **ID:** `2603.04968` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Amirabbas Afzali, Myeongho Jeon, Maria Brbic
+- **abs:** https://arxiv.org/abs/2603.04968 · **PDF:** https://arxiv.org/pdf/2603.04968v1
+
+## 7569. Replaying pre-training data improves fine-tuning
+
+- **ID:** `2603.04964` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Suhas Kotha, Percy Liang
+- **abs:** https://arxiv.org/abs/2603.04964 · **PDF:** https://arxiv.org/pdf/2603.04964v1
+
+## 7570. VisionPangu: A Compact and Fine-Grained Multimodal Assistant with 1.7B Parameters
+
+- **ID:** `2603.04957` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Jiaxin Fan, Wenpo Song
+- **abs:** https://arxiv.org/abs/2603.04957 · **PDF:** https://arxiv.org/pdf/2603.04957v1
+
+## 7571. WaterSIC: information-theoretically (near) optimal linear layer quantization
+
+- **ID:** `2603.04956` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Egor Lifar, Semyon Savkin, Or Ordentlich, et al.
+- **abs:** https://arxiv.org/abs/2603.04956 · **PDF:** https://arxiv.org/pdf/2603.04956v1
+
+## 7572. Quantum Key Distribution Secured Federated Learning for Channel Estimation and Radar Spectrum Sensing in 6G Networks
+
+- **ID:** `2603.15649` · **primary:** `cs.CR` · **Submitted:** 2026-03-05
+- **Authors:** Ferhat Ozgur Catak, Murat Kuzlu, Jungwon Seo, et al.
+- **abs:** https://arxiv.org/abs/2603.15649 · **PDF:** https://arxiv.org/pdf/2603.15649v1
+
+## 7573. Uncertainty quantification in neural network-based glucose prediction for diabetes
+
+- **ID:** `2603.04955` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Hai Siong Tan, Rafe McBeth
+- **abs:** https://arxiv.org/abs/2603.04955 · **PDF:** https://arxiv.org/pdf/2603.04955v2
+
+## 7574. Retrieval-Augmented Generation with Covariate Time Series
+
+- **ID:** `2603.04951` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Kenny Ye Liang, Zhongyi Pei, Huan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.04951 · **PDF:** https://arxiv.org/pdf/2603.04951v2
+
+## 7575. Location-Aware Pretraining for Medical Difference Visual Question Answering
+
+- **ID:** `2603.04950` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Denis Musinguzi, Caren Han, Prasenjit Mitra
+- **abs:** https://arxiv.org/abs/2603.04950 · **PDF:** https://arxiv.org/pdf/2603.04950v1
+
+## 7576. TimeWarp: Evaluating Web Agents by Revisiting the Past
+
+- **ID:** `2603.04949` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Md Farhan Ishmam, Kenneth Marino
+- **abs:** https://arxiv.org/abs/2603.04949 · **PDF:** https://arxiv.org/pdf/2603.04949v1
+
+## 7577. $\nabla$-Reasoner: LLM Reasoning via Test-Time Gradient Descent in Latent Space
+
+- **ID:** `2603.04948` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Peihao Wang, Ruisi Cai, Zhen Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.04948 · **PDF:** https://arxiv.org/pdf/2603.04948v1
+
+## 7578. LocalSUG: Geography-Aware LLM for Query Suggestion in Local-Life Services
+
+- **ID:** `2603.04946` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Jinwen Chen, Shuai Gong, Shiwen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.04946 · **PDF:** https://arxiv.org/pdf/2603.04946v1
+
+## 7579. Federated Heterogeneous Language Model Optimization for Hybrid Automatic Speech Recognition
+
+- **ID:** `2603.04945` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Mengze Hong, Yi Gu, Di Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.04945 · **PDF:** https://arxiv.org/pdf/2603.04945v1
+
+## 7580. Person Detection and Tracking from an Overhead Crane LiDAR
+
+- **ID:** `2603.04938` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Nilusha Jayawickrama, Henrik Toikka, Risto Ojala
+- **abs:** https://arxiv.org/abs/2603.04938 · **PDF:** https://arxiv.org/pdf/2603.04938v1
+
+## 7581. Semantic Communication-Enhanced Split Federated Learning for Vehicular Networks: Architecture, Challenges, and Case Study
+
+- **ID:** `2603.04936` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Lu Yu, Zheng Chang, Ying-Chang Liang
+- **abs:** https://arxiv.org/abs/2603.04936 · **PDF:** https://arxiv.org/pdf/2603.04936v1
+
+## 7582. AILS-NTUA at SemEval-2026 Task 3: Efficient Dimensional Aspect-Based Sentiment Analysis
+
+- **ID:** `2603.04933` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Stavros Gazetas, Giorgos Filandrianos, Maria Lymperaiou, et al.
+- **abs:** https://arxiv.org/abs/2603.04933 · **PDF:** https://arxiv.org/pdf/2603.04933v1
+
+## 7583. AILS-NTUA at SemEval-2026 Task 10: Agentic LLMs for Psycholinguistic Marker Extraction and Conspiracy Endorsement Detection
+
+- **ID:** `2603.04921` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Panagiotis Alexios Spanakis, Maria Lymperaiou, Giorgos Filandrianos, et al.
+- **abs:** https://arxiv.org/abs/2603.04921 · **PDF:** https://arxiv.org/pdf/2603.04921v1
+
+## 7584. Knowledge-informed Bidding with Dual-process Control for Online Advertising
+
+- **ID:** `2603.04920` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Huixiang Luo, Longyu Gao, Yaqi Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.04920 · **PDF:** https://arxiv.org/pdf/2603.04920v1
+
+## 7585. BandPO: Bridging Trust Regions and Ratio Clipping via Probability-Aware Bounds for LLM Reinforcement Learning
+
+- **ID:** `2603.04918` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Yuan Li, Bo Wang, Yufei Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.04918 · **PDF:** https://arxiv.org/pdf/2603.04918v1
+
+## 7586. EVMbench: Evaluating AI Agents on Smart Contract Security
+
+- **ID:** `2603.04915` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Justin Wang, Andreas Bigger, Xiaohai Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.04915 · **PDF:** https://arxiv.org/pdf/2603.04915v1
+
+## 7587. VPWEM: Non-Markovian Visuomotor Policy with Working and Episodic Memory
+
+- **ID:** `2603.04910` · **primary:** `cs.RO` · **Submitted:** 2026-03-05
+- **Authors:** Yuheng Lei, Zhixuan Liang, Hongyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.04910 · **PDF:** https://arxiv.org/pdf/2603.04910v1
+
+## 7588. Deterministic Preprocessing and Interpretable Fuzzy Banding for Cost-per-Student Reporting from Extracted Records
+
+- **ID:** `2603.04905` · **primary:** `cs.DB` · **Submitted:** 2026-03-05
+- **Authors:** Shane Lee, Stella Ng
+- **abs:** https://arxiv.org/abs/2603.04905 · **PDF:** https://arxiv.org/pdf/2603.04905v1
+
+## 7589. Alignment Backfire: Language-Dependent Reversal of Safety Interventions Across 16 Languages in LLM Multi-Agent Systems
+
+- **ID:** `2603.04904` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Hiroki Fukui
+- **abs:** https://arxiv.org/abs/2603.04904 · **PDF:** https://arxiv.org/pdf/2603.04904v1
+
+## 7590. AgentSCOPE: Evaluating Contextual Privacy Across Agentic Workflows
+
+- **ID:** `2603.04902` · **primary:** `cs.CR` · **Submitted:** 2026-03-05
+- **Authors:** Ivoline C. Ngong, Keerthiram Murugesan, Swanand Kadhe, et al.
+- **abs:** https://arxiv.org/abs/2603.04902 · **PDF:** https://arxiv.org/pdf/2603.04902v1
+
+## 7591. IntSeqBERT: Learning Arithmetic Structure in OEIS via Modulo-Spectrum Embeddings
+
+- **ID:** `2603.05556` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Kazuhisa Nakasho
+- **abs:** https://arxiv.org/abs/2603.05556 · **PDF:** https://arxiv.org/pdf/2603.05556v1
+
+## 7592. EvoTool: Self-Evolving Tool-Use Policy Optimization in LLM Agents via Blame-Aware Mutation and Diversity-Aware Selection
+
+- **ID:** `2603.04900` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Shuo Yang, Soyeon Caren Han, Xueqi Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.04900 · **PDF:** https://arxiv.org/pdf/2603.04900v1
+
+## 7593. U-Parking: Distributed UWB-Assisted Autonomous Parking System with Robust Localization and Intelligent Planning
+
+- **ID:** `2603.04898` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Yiang Wu, Qiong Wu, Pingyi Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.04898 · **PDF:** https://arxiv.org/pdf/2603.04898v1
+
+## 7594. Can LLMs Capture Expert Uncertainty? A Comparative Analysis of Value Alignment in Ethnographic Qualitative Research
+
+- **ID:** `2603.04897` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Arina Kostina, Marios Dikaiakos, Alejandro Porcel, et al.
+- **abs:** https://arxiv.org/abs/2603.04897 · **PDF:** https://arxiv.org/pdf/2603.04897v2
+
+## 7595. Authorize-on-Demand: Dynamic Authorization with Legality-Aware Intellectual Property Protection for VLMs
+
+- **ID:** `2603.04896` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Lianyu Wang, Meng Wang, Huazhu Fu, et al.
+- **abs:** https://arxiv.org/abs/2603.04896 · **PDF:** https://arxiv.org/pdf/2603.04896v1
+
+## 7596. How Does the ReLU Activation Affect the Implicit Bias of Gradient Descent on High-dimensional Neural Network Regression?
+
+- **ID:** `2603.04895` · **primary:** `stat.ML` · **Submitted:** 2026-03-05
+- **Authors:** Kuo-Wei Lai, Guanghui Wang, Molei Tao, et al.
+- **abs:** https://arxiv.org/abs/2603.04895 · **PDF:** https://arxiv.org/pdf/2603.04895v1
+
+## 7597. Differentially Private Multimodal In-Context Learning
+
+- **ID:** `2603.04894` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Ivoline C. Ngong, Zarreen Reza, Joseph P. Near
+- **abs:** https://arxiv.org/abs/2603.04894 · **PDF:** https://arxiv.org/pdf/2603.04894v1
+
+## 7598. Free Lunch for Pass@$k$? Low Cost Diverse Sampling for Diffusion Language Models
+
+- **ID:** `2603.04893` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Sean Lamont, Christian Walder, Paul Montague, et al.
+- **abs:** https://arxiv.org/abs/2603.04893 · **PDF:** https://arxiv.org/pdf/2603.04893v1
+
+## 7599. FedAFD: Multimodal Federated Learning via Adversarial Fusion and Distillation
+
+- **ID:** `2603.04890` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Min Tan, Junchao Ma, Yinfu Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.04890 · **PDF:** https://arxiv.org/pdf/2603.04890v1
+
+## 7600. Bounded State in an Infinite Horizon: Proactive Hierarchical Memory for Ad-Hoc Recall over Streaming Dialogues
+
+- **ID:** `2603.04885` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Bingbing Wang, Jing Li, Ruifeng Xu
+- **abs:** https://arxiv.org/abs/2603.04885 · **PDF:** https://arxiv.org/pdf/2603.04885v1
+
+## 7601. DeformTrace: A Deformable State Space Model with Relay Tokens for Temporal Forgery Localization
+
+- **ID:** `2603.04882` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Xiaodong Zhu, Suting Wang, Yuanming Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.04882 · **PDF:** https://arxiv.org/pdf/2603.04882v1
+
+## 7602. Differential Privacy in Two-Layer Networks: How DP-SGD Harms Fairness and Robustness
+
+- **ID:** `2603.04881` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Ruichen Xu, Kexin Chen
+- **abs:** https://arxiv.org/abs/2603.04881 · **PDF:** https://arxiv.org/pdf/2603.04881v1
+
+## 7603. Interpretable Pre-Release Baseball Pitch Type Anticipation from Broadcast 3D Kinematics
+
+- **ID:** `2603.04874` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Jerrin Bright, Michelle Lu, John Zelek
+- **abs:** https://arxiv.org/abs/2603.04874 · **PDF:** https://arxiv.org/pdf/2603.04874v1
+
+## 7604. SEA-TS: Self-Evolving Agent for Autonomous Code Generation of Time Series Forecasting Algorithms
+
+- **ID:** `2603.04873` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Longkun Xu, Xiaochun Zhang, Qiantu Tuo, et al.
+- **abs:** https://arxiv.org/abs/2603.04873 · **PDF:** https://arxiv.org/pdf/2603.04873v2
+
+## 7605. K-Gen: A Multimodal Language-Conditioned Approach for Interpretable Keypoint-Guided Trajectory Generation
+
+- **ID:** `2603.04868` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Mingxuan Mu, Guo Yang, Lei Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.04868 · **PDF:** https://arxiv.org/pdf/2603.04868v1
+
+## 7606. Causally Robust Reward Learning from Reason-Augmented Preference Feedback
+
+- **ID:** `2603.04861` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Minjune Hwang, Yigit Korkmaz, Daniel Seita, et al.
+- **abs:** https://arxiv.org/abs/2603.04861 · **PDF:** https://arxiv.org/pdf/2603.04861v1
+
+## 7607. Osmosis Distillation: Model Hijacking with the Fewest Samples
+
+- **ID:** `2603.04859` · **primary:** `cs.CR` · **Submitted:** 2026-03-05
+- **Authors:** Yuchen Shi, Huajie Chen, Heng Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.04859 · **PDF:** https://arxiv.org/pdf/2603.04859v1
+
+## 7608. FireBench: Evaluating Instruction Following in Enterprise and API-Driven LLM Applications
+
+- **ID:** `2603.04857` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Yunfan Zhang, Yijie Bei, Jetashree Ravi, et al.
+- **abs:** https://arxiv.org/abs/2603.04857 · **PDF:** https://arxiv.org/pdf/2603.04857v1
+
+## 7609. HACHIMI: Scalable and Controllable Student Persona Generation via Orchestrated Agents
+
+- **ID:** `2603.04855` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Yilin Jiang, Fei Tan, Xuanyu Yin, et al.
+- **abs:** https://arxiv.org/abs/2603.04855 · **PDF:** https://arxiv.org/pdf/2603.04855v2
+
+## 7610. RBF-Solver: A Multistep Sampler for Diffusion Probabilistic Models via Radial Basis Functions
+
+- **ID:** `2603.13330` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Soochul Park, Yeon Ju Lee, SeongJin Yoon, et al.
+- **abs:** https://arxiv.org/abs/2603.13330 · **PDF:** https://arxiv.org/pdf/2603.13330v1
+
+## 7611. SinhaLegal: A Benchmark Corpus for Information Extraction and Analysis in Sinhala Legislative Texts
+
+- **ID:** `2603.04854` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Minduli Lasandi, Nevidu Jayatilleke
+- **abs:** https://arxiv.org/abs/2603.04854 · **PDF:** https://arxiv.org/pdf/2603.04854v1
+
+## 7612. On Multi-Step Theorem Prediction via Non-Parametric Structural Priors
+
+- **ID:** `2603.04852` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Junbo Zhao, Ting Zhang, Can Li, et al.
+- **abs:** https://arxiv.org/abs/2603.04852 · **PDF:** https://arxiv.org/pdf/2603.04852v1
+
+## 7613. Why Is RLHF Alignment Shallow? A Gradient Analysis
+
+- **ID:** `2603.04851` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Robin Young
+- **abs:** https://arxiv.org/abs/2603.04851 · **PDF:** https://arxiv.org/pdf/2603.04851v1
+
+## 7614. An Approach to Simultaneous Acquisition of Real-Time MRI Video, EEG, and Surface EMG for Articulatory, Brain, and Muscle Activity During Speech Production
+
+- **ID:** `2603.04840` · **primary:** `eess.AS` · **Submitted:** 2026-03-05
+- **Authors:** Jihwan Lee, Parsa Razmara, Kevin Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.04840 · **PDF:** https://arxiv.org/pdf/2603.04840v1
+
+## 7615. Design Behaviour Codes (DBCs): A Taxonomy-Driven Layered Governance Benchmark for Large Language Models
+
+- **ID:** `2603.04837` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** G. Madan Mohan, Veena Kiran Nambiar, Kiranmayee Janardhan
+- **abs:** https://arxiv.org/abs/2603.04837 · **PDF:** https://arxiv.org/pdf/2603.04837v1
+
+## 7616. SCoUT: Scalable Communication via Utility-Guided Temporal Grouping in Multi-Agent Reinforcement Learning
+
+- **ID:** `2603.04833` · **primary:** `cs.MA` · **Submitted:** 2026-03-05
+- **Authors:** Manav Vora, Gokul Puthumanaillam, Hiroyasu Tsukamoto, et al.
+- **abs:** https://arxiv.org/abs/2603.04833 · **PDF:** https://arxiv.org/pdf/2603.04833v1
+
+## 7617. Missingness Bias Calibration in Feature Attribution Explanations
+
+- **ID:** `2603.04831` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Shailesh Sridhar, Anton Xue, Eric Wong
+- **abs:** https://arxiv.org/abs/2603.04831 · **PDF:** https://arxiv.org/pdf/2603.04831v2
+
+## 7618. From Unfamiliar to Familiar: Detecting Pre-training Data via Gradient Deviations in Large Language Models
+
+- **ID:** `2603.04828` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Ruiqi Zhang, Lingxiang Wang, Hainan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.04828 · **PDF:** https://arxiv.org/pdf/2603.04828v1
+
+## 7619. Multilevel Training for Kolmogorov Arnold Networks
+
+- **ID:** `2603.04827` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Ben S. Southworth, Jonas A. Actor, Graham Harper, et al.
+- **abs:** https://arxiv.org/abs/2603.04827 · **PDF:** https://arxiv.org/pdf/2603.04827v1
+
+## 7620. Mitigating Instance Entanglement in Instance-Dependent Partial Label Learning
+
+- **ID:** `2603.04825` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Rui Zhao, Bin Shi, Kai Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.04825 · **PDF:** https://arxiv.org/pdf/2603.04825v1
+
+## 7621. VISA: Value Injection via Shielded Adaptation for Personalized LLM Alignment
+
+- **ID:** `2603.04822` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Jiawei Chen, Tianzhuo Yang, Guoxi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.04822 · **PDF:** https://arxiv.org/pdf/2603.04822v1
+
+## 7622. Autoscoring Anticlimax: A Meta-analytic Understanding of AI's Short-answer Shortcomings and Wording Weaknesses
+
+- **ID:** `2603.04820` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Michael Hardy
+- **abs:** https://arxiv.org/abs/2603.04820 · **PDF:** https://arxiv.org/pdf/2603.04820v2
+
+## 7623. On the Strengths and Weaknesses of Data for Open-set Embodied Assistance
+
+- **ID:** `2603.04819` · **primary:** `cs.RO` · **Submitted:** 2026-03-05
+- **Authors:** Pradyumna Tambwekar, Andrew Silva, Deepak Gopinath, et al.
+- **abs:** https://arxiv.org/abs/2603.04819 · **PDF:** https://arxiv.org/pdf/2603.04819v1
+
+## 7624. LLM-Grounded Explainable AI for Supply Chain Risk Early Warning via Temporal Graph Attention Networks
+
+- **ID:** `2603.04818` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Zhiming Xue, Yujue Wang, Menghao Huo
+- **abs:** https://arxiv.org/abs/2603.04818 · **PDF:** https://arxiv.org/pdf/2603.04818v2
+
+## 7625. EchoGuard: An Agentic Framework with Knowledge-Graph Memory for Detecting Manipulative Communication in Longitudinal Dialogue
+
+- **ID:** `2603.04815` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Ratna Kandala, Niva Manchanda, Akshata Kishore Moharir, et al.
+- **abs:** https://arxiv.org/abs/2603.04815 · **PDF:** https://arxiv.org/pdf/2603.04815v1
+
+## 7626. Beyond the Context Window: A Cost-Performance Analysis of Fact-Based Memory vs. Long-Context LLMs for Persistent Agents
+
+- **ID:** `2603.04814` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Natchanon Pollertlam, Witchayut Kornsuwannawit
+- **abs:** https://arxiv.org/abs/2603.04814 · **PDF:** https://arxiv.org/pdf/2603.04814v1
+
+## 7627. EigenData: A Self-Evolving Multi-Agent Platform for Function-Calling Data Synthesis, Auditing, and Repair
+
+- **ID:** `2603.05553` · **primary:** `cs.SE` · **Submitted:** 2026-03-05
+- **Authors:** Jiaao Chen, Jingyuan Qi, Mingye Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.05553 · **PDF:** https://arxiv.org/pdf/2603.05553v1
+
+## 7628. Quadratic polarity and polar Fenchel-Young divergences from the canonical Legendre polarity
+
+- **ID:** `2603.04812` · **primary:** `cs.CG` · **Submitted:** 2026-03-05
+- **Authors:** Frank Nielsen, Basile Plus-Gourdon, Mahito Sugiyama
+- **abs:** https://arxiv.org/abs/2603.04812 · **PDF:** https://arxiv.org/pdf/2603.04812v1
+
+## 7629. Meta-D: Metadata-Aware Architectures for Brain Tumor Analysis and Missing-Modality Segmentation
+
+- **ID:** `2603.04811` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** SangHyuk Kim, Daniel Haehn, Sumientra Rampersad
+- **abs:** https://arxiv.org/abs/2603.04811 · **PDF:** https://arxiv.org/pdf/2603.04811v1
+
+## 7630. WhisperAlign: Word-Boundary-Aware ASR and WhisperX-Anchored Pyannote Diarization for Long-Form Bengali Speech
+
+- **ID:** `2603.04809` · **primary:** `cs.SD` · **Submitted:** 2026-03-05
+- **Authors:** Aurchi Chowdhury, Rubaiyat -E-Zaman, Sk. Ashrafuzzaman Nafees
+- **abs:** https://arxiv.org/abs/2603.04809 · **PDF:** https://arxiv.org/pdf/2603.04809v1
+
+## 7631. The Inductive Bias of Convolutional Neural Networks: Locality and Weight Sharing Reshape Implicit Regularization
+
+- **ID:** `2603.04807` · **primary:** `stat.ML` · **Submitted:** 2026-03-05
+- **Authors:** Tongtong Liang, Esha Singh, Rahul Parhi, et al.
+- **abs:** https://arxiv.org/abs/2603.04807 · **PDF:** https://arxiv.org/pdf/2603.04807v1
+
+## 7632. Guiding Diffusion-based Reconstruction with Contrastive Signals for Balanced Visual Representation
+
+- **ID:** `2603.04803` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Boyu Han, Qianqian Xu, Shilong Bao, et al.
+- **abs:** https://arxiv.org/abs/2603.04803 · **PDF:** https://arxiv.org/pdf/2603.04803v2
+
+## 7633. Beyond Linear LLM Invocation: An Efficient and Effective Semantic Filter Paradigm
+
+- **ID:** `2603.04799` · **primary:** `cs.DB` · **Submitted:** 2026-03-05
+- **Authors:** Nan Hou, Kangfei Zhao, Jiadong Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.04799 · **PDF:** https://arxiv.org/pdf/2603.04799v1
+
+## 7634. Comparative Evaluation of Traditional Methods and Deep Learning for Brain Glioma Imaging. Review Paper
+
+- **ID:** `2603.04796` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Kiranmayee Janardhan, Vinay Martin DSa Prabhu, T. Christy Bobby
+- **abs:** https://arxiv.org/abs/2603.04796 · **PDF:** https://arxiv.org/pdf/2603.04796v1
+
+## 7635. LUMINA: Laplacian-Unifying Mechanism for Interpretable Neurodevelopmental Analysis via Quad-Stream GCN
+
+- **ID:** `2603.13329` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Minkyung Cha, Jooyoung Bae, Jaewon Jung, et al.
+- **abs:** https://arxiv.org/abs/2603.13329 · **PDF:** https://arxiv.org/pdf/2603.13329v1
+
+## 7636. LAW & ORDER: Adaptive Spatial Weighting for Medical Diffusion and Segmentation
+
+- **ID:** `2603.04795` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Anugunj Naman, Ayushman Singh, Gaibo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.04795 · **PDF:** https://arxiv.org/pdf/2603.04795v1
+
+## 7637. Timer-S1: A Billion-Scale Time Series Foundation Model with Serial Scaling
+
+- **ID:** `2603.04791` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Yong Liu, Xingjian Su, Shiyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.04791 · **PDF:** https://arxiv.org/pdf/2603.04791v2
+
+## 7638. Diffusion Policy through Conditional Proximal Policy Optimization
+
+- **ID:** `2603.04790` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Ben Liu, Shunpeng Yang, Hua Chen
+- **abs:** https://arxiv.org/abs/2603.04790 · **PDF:** https://arxiv.org/pdf/2603.04790v1
+
+## 7639. Breaking Contextual Inertia: Reinforcement Learning with Single-Turn Anchors for Stable Multi-Turn Interaction
+
+- **ID:** `2603.04783` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Xingwu Chen, Zhanqiu Zhang, Yiwen Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.04783 · **PDF:** https://arxiv.org/pdf/2603.04783v1
+
+## 7640. Distributional Equivalence in Linear Non-Gaussian Latent-Variable Cyclic Causal Models: Characterization and Learning
+
+- **ID:** `2603.04780` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Haoyue Dai, Immanuel Albrecht, Peter Spirtes, et al.
+- **abs:** https://arxiv.org/abs/2603.04780 · **PDF:** https://arxiv.org/pdf/2603.04780v1
+
+## 7641. Privacy-Aware Camera 2.0 Technical Report
+
+- **ID:** `2603.04775` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Huan Song, Shuyu Tian, Ting Long, et al.
+- **abs:** https://arxiv.org/abs/2603.04775 · **PDF:** https://arxiv.org/pdf/2603.04775v1
+
+## 7642. TSEmbed: Unlocking Task Scaling in Universal Multimodal Embeddings
+
+- **ID:** `2603.04772` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Yebo Wu, Feng Liu, Ziwei Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.04772 · **PDF:** https://arxiv.org/pdf/2603.04772v1
+
+## 7643. MADCrowner: Margin Aware Dental Crown Design with Template Deformation and Refinement
+
+- **ID:** `2603.04771` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Linda Wei, Chang Liu, Wenran Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.04771 · **PDF:** https://arxiv.org/pdf/2603.04771v1
+
+## 7644. DSA-SRGS: Super-Resolution Gaussian Splatting for Dynamic Sparse-View DSA Reconstruction
+
+- **ID:** `2603.04770` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Shiyu Zhang, Zhicong Wu, Huangxuan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.04770 · **PDF:** https://arxiv.org/pdf/2603.04770v1
+
+## 7645. Distributional Reinforcement Learning with Information Bottleneck for Uncertainty-Aware DRAM Equalization
+
+- **ID:** `2603.04768` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Muhammad Usama, Dong Eui Chang
+- **abs:** https://arxiv.org/abs/2603.04768 · **PDF:** https://arxiv.org/pdf/2603.04768v1
+
+## 7646. ConTSG-Bench: A Unified Benchmark for Conditional Time Series Generation
+
+- **ID:** `2603.04767` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Shaocheng Lan, Shuqi Gu, Zhangzhi Xiong, et al.
+- **abs:** https://arxiv.org/abs/2603.04767 · **PDF:** https://arxiv.org/pdf/2603.04767v1
+
+## 7647. Evaluating GPT-5 as a Multimodal Clinical Reasoner: A Landscape Commentary
+
+- **ID:** `2603.04763` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Alexandru Florea, Shansong Wang, Mingzhe Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.04763 · **PDF:** https://arxiv.org/pdf/2603.04763v1
+
+## 7648. Stacked from One: Multi-Scale Self-Injection for Context Window Extension
+
+- **ID:** `2603.04759` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Wei Han, Pan Zhou, Shuicheng Yan
+- **abs:** https://arxiv.org/abs/2603.04759 · **PDF:** https://arxiv.org/pdf/2603.04759v1
+
+## 7649. Predicting Atomistic Transitions with Transformers
+
+- **ID:** `2603.06526` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-05
+- **Authors:** Henry Tischler, Wenting Li, Qi Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.06526 · **PDF:** https://arxiv.org/pdf/2603.06526v1
+
+## 7650. MOOSEnger -- a Domain-Specific AI Agent for the MOOSE Ecosystem
+
+- **ID:** `2603.04756` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Mengnan Li, Jason Miller, Zachary Prince, et al.
+- **abs:** https://arxiv.org/abs/2603.04756 · **PDF:** https://arxiv.org/pdf/2603.04756v2
+
+## 7651. KindSleep: Knowledge-Informed Diagnosis of Obstructive Sleep Apnea from Oximetry
+
+- **ID:** `2603.04755` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Micky C Nnamdi, Wenqi Shi, Cheng Wan, et al.
+- **abs:** https://arxiv.org/abs/2603.04755 · **PDF:** https://arxiv.org/pdf/2603.04755v1
+
+## 7652. Evaluating the Search Agent in a Parallel World
+
+- **ID:** `2603.04751` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Jiawei Chen, Xintian Shen, Lihao Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.04751 · **PDF:** https://arxiv.org/pdf/2603.04751v1
+
+## 7653. HiMAP-Travel: Hierarchical Multi-Agent Planning for Long-Horizon Constrained Travel
+
+- **ID:** `2603.04750` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** The Viet Bui, Wenjun Li, Yong Liu
+- **abs:** https://arxiv.org/abs/2603.04750 · **PDF:** https://arxiv.org/pdf/2603.04750v1
+
+## 7654. Visioning Human-Agentic AI Teaming: Continuity, Tension, and Future Research
+
+- **ID:** `2603.04746` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Bowen Lou, Tian Lu, T. S. Raghu, et al.
+- **abs:** https://arxiv.org/abs/2603.04746 · **PDF:** https://arxiv.org/pdf/2603.04746v1
+
+## 7655. DARE: Aligning LLM Agents with the R Statistical Ecosystem via Distribution-Aware Retrieval
+
+- **ID:** `2603.04743` · **primary:** `cs.IR` · **Submitted:** 2026-03-05
+- **Authors:** Maojun Sun, Yue Wu, Yifei Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.04743 · **PDF:** https://arxiv.org/pdf/2603.04743v1
+
+## 7656. CONE: Embeddings for Complex Numerical Data Preserving Unit and Variable Semantics
+
+- **ID:** `2603.04741` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Gyanendra Shrestha, Anna Pyayt, Michael Gubanov
+- **abs:** https://arxiv.org/abs/2603.04741 · **PDF:** https://arxiv.org/pdf/2603.04741v1
+
+## 7657. Memory as Ontology: A Constitutional Memory Architecture for Persistent Digital Citizens
+
+- **ID:** `2603.04740` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Zhenghui Li
+- **abs:** https://arxiv.org/abs/2603.04740 · **PDF:** https://arxiv.org/pdf/2603.04740v1
+
+## 7658. IF-RewardBench: Benchmarking Judge Models for Instruction-Following Evaluation
+
+- **ID:** `2603.04738` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Bosi Wen, Yilin Niu, Cunxiang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.04738 · **PDF:** https://arxiv.org/pdf/2603.04738v1
+
+## 7659. Interactive Benchmarks
+
+- **ID:** `2603.04737` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Baoqing Yue, Zihan Zhu, Yifan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.04737 · **PDF:** https://arxiv.org/pdf/2603.04737v1
+
+## 7660. Distribution-Conditioned Transport
+
+- **ID:** `2603.04736` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Nic Fishman, Gokul Gowri, Paolo L. B. Fischer, et al.
+- **abs:** https://arxiv.org/abs/2603.04736 · **PDF:** https://arxiv.org/pdf/2603.04736v1
+
+## 7661. Solving an Open Problem in Theoretical Physics using AI-Assisted Discovery
+
+- **ID:** `2603.04735` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Michael P. Brenner, Vincent Cohen-Addad, David Woodruff
+- **abs:** https://arxiv.org/abs/2603.04735 · **PDF:** https://arxiv.org/pdf/2603.04735v1
+
+## 7662. Thinking with Gaze: Sequential Eye-Tracking as Visual Reasoning Supervision for Medical VLMs
+
+- **ID:** `2603.06697` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Yiwei Li, Zihao Wu, Yanjun Lv, et al.
+- **abs:** https://arxiv.org/abs/2603.06697 · **PDF:** https://arxiv.org/pdf/2603.06697v1
+
+## 7663. When Priors Backfire: On the Vulnerability of Unlearnable Examples to Pretraining
+
+- **ID:** `2603.04731` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Zhihao Li, Gezheng Xu, Jiale Cai, et al.
+- **abs:** https://arxiv.org/abs/2603.04731 · **PDF:** https://arxiv.org/pdf/2603.04731v1
+
+## 7664. Count Bridges enable Modeling and Deconvolving Transcriptomic Data
+
+- **ID:** `2603.04730` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Nic Fishman, Gokul Gowri, Tanush Kumar, et al.
+- **abs:** https://arxiv.org/abs/2603.04730 · **PDF:** https://arxiv.org/pdf/2603.04730v1
+
+## 7665. Are Multimodal LLMs Ready for Surveillance? A Reality Check on Zero-Shot Anomaly Detection in the Wild
+
+- **ID:** `2603.04727` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Shanle Yao, Armin Danesh Pazho, Narges Rashvand, et al.
+- **abs:** https://arxiv.org/abs/2603.04727 · **PDF:** https://arxiv.org/pdf/2603.04727v1
+
+## 7666. From Offline to Periodic Adaptation for Pose-Based Shoplifting Detection in Real-world Retail Security
+
+- **ID:** `2603.04723` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Shanle Yao, Narges Rashvand, Armin Danesh Pazho, et al.
+- **abs:** https://arxiv.org/abs/2603.04723 · **PDF:** https://arxiv.org/pdf/2603.04723v1
+
+## 7667. Model Medicine: A Clinical Framework for Understanding, Diagnosing, and Treating AI Models
+
+- **ID:** `2603.04722` · **primary:** `cs.AI` · **Submitted:** 2026-03-05
+- **Authors:** Jihoon Jeong
+- **abs:** https://arxiv.org/abs/2603.04722 · **PDF:** https://arxiv.org/pdf/2603.04722v2
+
+## 7668. A Benchmark Study of Neural Network Compression Methods for Hyperspectral Image Classification
+
+- **ID:** `2603.04720` · **primary:** `cs.CV` · **Submitted:** 2026-03-05
+- **Authors:** Sai Shi
+- **abs:** https://arxiv.org/abs/2603.04720 · **PDF:** https://arxiv.org/pdf/2603.04720v1
+
+## 7669. AI-Assisted Moot Courts: Simulating Justice-Specific Questioning in Oral Arguments
+
+- **ID:** `2603.04718` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Kylie Zhang, Nimra Nadeem, Lucia Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.04718 · **PDF:** https://arxiv.org/pdf/2603.04718v1
+
+## 7670. SLO-Aware Compute Resource Allocation for Prefill-Decode Disaggregated LLM Inference
+
+- **ID:** `2603.04716` · **primary:** `cs.DC` · **Submitted:** 2026-03-05
+- **Authors:** Luchang Li, Dongfang Li, Bozhao Gong, et al.
+- **abs:** https://arxiv.org/abs/2603.04716 · **PDF:** https://arxiv.org/pdf/2603.04716v1
+
+## 7671. Probabilistic Dreaming for World Models
+
+- **ID:** `2603.04715` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Gavin Wong
+- **abs:** https://arxiv.org/abs/2603.04715 · **PDF:** https://arxiv.org/pdf/2603.04715v1
+
+## 7672. When Denoising Hinders: Revisiting Zero-Shot ASR with SAM-Audio and Whisper
+
+- **ID:** `2603.04710` · **primary:** `cs.SD` · **Submitted:** 2026-03-05
+- **Authors:** Akif Islam, Raufun Nahar, Md. Ekramul Hamid
+- **abs:** https://arxiv.org/abs/2603.04710 · **PDF:** https://arxiv.org/pdf/2603.04710v1
+
+## 7673. Detection of Illicit Content on Online Marketplaces using Large Language Models
+
+- **ID:** `2603.04707` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Quoc Khoa Tran, Thanh Thi Nguyen, Campbell Wilson
+- **abs:** https://arxiv.org/abs/2603.04707 · **PDF:** https://arxiv.org/pdf/2603.04707v1
+
+## 7674. Implicit Bias and Loss of Plasticity in Matrix Completion: Depth Promotes Low-Rankness
+
+- **ID:** `2603.04703` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Baekrok Shin, Chulhee Yun
+- **abs:** https://arxiv.org/abs/2603.04703 · **PDF:** https://arxiv.org/pdf/2603.04703v1
+
+## 7675. Hate Speech Detection using Large Language Models with Data Augmentation and Feature Enhancement
+
+- **ID:** `2603.04698` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Brian Jing Hong Nge, Stefan Su, Thanh Thi Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.04698 · **PDF:** https://arxiv.org/pdf/2603.04698v1
+
+## 7676. Engineering Regression Without Real-Data Training: Domain Adaptation for Tabular Foundation Models Using Multi-Dataset Embeddings
+
+- **ID:** `2603.04692` · **primary:** `cs.LG` · **Submitted:** 2026-03-05
+- **Authors:** Lyle Regenwetter, Rosen Yu, Cyril Picard, et al.
+- **abs:** https://arxiv.org/abs/2603.04692 · **PDF:** https://arxiv.org/pdf/2603.04692v1
+
+## 7677. Non-Zipfian Distribution of Stopwords and Subset Selection Models
+
+- **ID:** `2603.04691` · **primary:** `cs.CL` · **Submitted:** 2026-03-05
+- **Authors:** Wentian Li, Oscar Fontanelli
+- **abs:** https://arxiv.org/abs/2603.04691 · **PDF:** https://arxiv.org/pdf/2603.04691v1
+
+## 7678. Generalizing Fair Top-$k$ Selection: An Integrative Approach
+
+- **ID:** `2603.04689` · **primary:** `cs.DS` · **Submitted:** 2026-03-05
+- **Authors:** Guangya Cai
+- **abs:** https://arxiv.org/abs/2603.04689 · **PDF:** https://arxiv.org/pdf/2603.04689v2
+
+## 7679. Why the Brain Consolidates: Predictive Forgetting for Optimal Generalisation
+
+- **ID:** `2603.04688` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-05
+- **Authors:** Zafeirios Fountas, Adnan Oomerjee, Haitham Bou-Ammar, et al.
+- **abs:** https://arxiv.org/abs/2603.04688 · **PDF:** https://arxiv.org/pdf/2603.04688v1
+
+## 7680. Direct Estimation of Tree Volume and Aboveground Biomass Using Deep Regression with Synthetic Lidar Data
+
+- **ID:** `2603.04683` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Habib Pourdelan, Zhengkang Xiang, Hugh Stewart, et al.
+- **abs:** https://arxiv.org/abs/2603.04683 · **PDF:** https://arxiv.org/pdf/2603.04683v1
+
+## 7681. Optimizing Language Models for Crosslingual Knowledge Consistency
+
+- **ID:** `2603.04678` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Tianyu Liu, Jirui Qi, Mrinmaya Sachan, et al.
+- **abs:** https://arxiv.org/abs/2603.04678 · **PDF:** https://arxiv.org/pdf/2603.04678v1
+
+## 7682. Decoding the Pulse of Reasoning VLMs in Multi-Image Understanding Tasks
+
+- **ID:** `2603.04676` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Chenjun Li
+- **abs:** https://arxiv.org/abs/2603.04676 · **PDF:** https://arxiv.org/pdf/2603.04676v1
+
+## 7683. sFRC for assessing hallucinations in medical image restoration
+
+- **ID:** `2603.04673` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Prabhat Kc, Rongping Zeng, Nirmal Soni, et al.
+- **abs:** https://arxiv.org/abs/2603.04673 · **PDF:** https://arxiv.org/pdf/2603.04673v1
+
+## 7684. Improving the accuracy of physics-informed neural networks via last-layer retraining
+
+- **ID:** `2603.04672` · **primary:** `math.NA` · **Submitted:** 2026-03-04
+- **Authors:** Saad Qadeer, Panos Stinis
+- **abs:** https://arxiv.org/abs/2603.04672 · **PDF:** https://arxiv.org/pdf/2603.04672v2
+
+## 7685. Using Vision + Language Models to Predict Item Difficulty
+
+- **ID:** `2603.04670` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Samin Khan
+- **abs:** https://arxiv.org/abs/2603.04670 · **PDF:** https://arxiv.org/pdf/2603.04670v1
+
+## 7686. Neuro-Symbolic Financial Reasoning via Deterministic Fact Ledgers and Adversarial Low-Latency Hallucination Detector
+
+- **ID:** `2603.04663` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Pedram Agand
+- **abs:** https://arxiv.org/abs/2603.04663 · **PDF:** https://arxiv.org/pdf/2603.04663v2
+
+## 7687. GIANT - Global Path Integration and Attentive Graph Networks for Multi-Agent Trajectory Planning
+
+- **ID:** `2603.04659` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Jonas le Fevre Sejersen, Toyotaro Suzumura, Erdal Kayacan
+- **abs:** https://arxiv.org/abs/2603.04659 · **PDF:** https://arxiv.org/pdf/2603.04659v1
+
+## 7688. Stan: An LLM-based thermodynamics course assistant
+
+- **ID:** `2603.04657` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Eric M. Furst, Vasudevan Venkateshwaran
+- **abs:** https://arxiv.org/abs/2603.04657 · **PDF:** https://arxiv.org/pdf/2603.04657v1
+
+## 7689. iAgentBench: Benchmarking Sensemaking Capabilities of Information-Seeking Agents on High-Traffic Topics
+
+- **ID:** `2603.04656` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Preetam Prabhu Srikar Dammu, Arnav Palkhiwala, Tanya Roosta, et al.
+- **abs:** https://arxiv.org/abs/2603.04656 · **PDF:** https://arxiv.org/pdf/2603.04656v1
+
+## 7690. Self-Supervised Multi-Stage Domain Unlearning for White-Matter Lesion Segmentation
+
+- **ID:** `2603.13328` · **primary:** `eess.IV` · **Submitted:** 2026-03-04
+- **Authors:** Domen Preložnik, Žiga Špiclin
+- **abs:** https://arxiv.org/abs/2603.13328 · **PDF:** https://arxiv.org/pdf/2603.13328v1
+
+## 7691. When Sensors Fail: Temporal Sequence Models for Robust PPO under Sensor Drift
+
+- **ID:** `2603.04648` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Kevin Vogt-Lowell, Theodoros Tsiligkaridis, Rodney Lafuente-Mercado, et al.
+- **abs:** https://arxiv.org/abs/2603.04648 · **PDF:** https://arxiv.org/pdf/2603.04648v2
+
+## 7692. Coordinated Semantic Alignment and Evidence Constraints for Retrieval-Augmented Generation with Large Language Models
+
+- **ID:** `2603.04647` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Xin Chen, Saili Uday Gadgil, Jiarong Qiu
+- **abs:** https://arxiv.org/abs/2603.04647 · **PDF:** https://arxiv.org/pdf/2603.04647v1
+
+## 7693. RoboMME: Benchmarking and Understanding Memory for Robotic Generalist Policies
+
+- **ID:** `2603.04639` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Yinpei Dai, Hongze Fu, Jayjun Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.04639 · **PDF:** https://arxiv.org/pdf/2603.04639v1
+
+## 7694. Spinverse: Differentiable Physics for Permeability-Aware Microstructure Reconstruction from Diffusion MRI
+
+- **ID:** `2603.04638` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Prathamesh Pradeep Khole, Mario M. Brenes, Zahra Kais Petiwala, et al.
+- **abs:** https://arxiv.org/abs/2603.04638 · **PDF:** https://arxiv.org/pdf/2603.04638v1
+
+## 7695. When Agents Persuade: Rhetoric Generation and Mitigation in LLMs
+
+- **ID:** `2603.04636` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Julia Jose, Ritik Roongta, Rachel Greenstadt
+- **abs:** https://arxiv.org/abs/2603.04636 · **PDF:** https://arxiv.org/pdf/2603.04636v2
+
+## 7696. Optimal Prediction-Augmented Algorithms for Testing Independence of Distributions
+
+- **ID:** `2603.04635` · **primary:** `stat.ML` · **Submitted:** 2026-03-04
+- **Authors:** Maryam Aliakbarpour, Alireza Azizi, Ria Stevens
+- **abs:** https://arxiv.org/abs/2603.04635 · **PDF:** https://arxiv.org/pdf/2603.04635v1
+
+## 7697. Towards automated data analysis: A guided framework for LLM-based risk estimation
+
+- **ID:** `2603.04631` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Panteleimon Rodis
+- **abs:** https://arxiv.org/abs/2603.04631 · **PDF:** https://arxiv.org/pdf/2603.04631v1
+
+## 7698. K-Means as a Radial Basis function Network: a Variational and Gradient-based Equivalence
+
+- **ID:** `2603.04625` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Felipe de Jesus Felix Arredondo, Alejandro Ucan-Puc, Carlos Astengo Noguez
+- **abs:** https://arxiv.org/abs/2603.04625 · **PDF:** https://arxiv.org/pdf/2603.04625v1
+
+## 7699. PDE foundation model-accelerated inverse estimation of system parameters in inertial confinement fusion
+
+- **ID:** `2603.04606` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Mahindra Rautela, Alexander Scheinker, Bradley Love, et al.
+- **abs:** https://arxiv.org/abs/2603.04606 · **PDF:** https://arxiv.org/pdf/2603.04606v1
+
+## 7700. Vibe Code Bench: Evaluating AI Models on End-to-End Web Application Development
+
+- **ID:** `2603.04601` · **primary:** `cs.SE` · **Submitted:** 2026-03-04
+- **Authors:** Hung Tran, Langston Nashold, Rayan Krishnan, et al.
+- **abs:** https://arxiv.org/abs/2603.04601 · **PDF:** https://arxiv.org/pdf/2603.04601v1
+
+## 7701. DOVA: Deliberation-First Multi-Agent Orchestration for Autonomous Research Automation
+
+- **ID:** `2603.13327` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Aaron Shen, Alfred Shen
+- **abs:** https://arxiv.org/abs/2603.13327 · **PDF:** https://arxiv.org/pdf/2603.13327v1
+
+## 7702. Bootstrapping Exploration with Group-Level Natural Language Feedback in Reinforcement Learning
+
+- **ID:** `2603.04597` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Lei Huang, Xiang Cheng, Chenxiao Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.04597 · **PDF:** https://arxiv.org/pdf/2603.04597v1
+
+## 7703. A Late-Fusion Multimodal AI Framework for Privacy-Preserving Deduplication in National Healthcare Data Environments
+
+- **ID:** `2603.04595` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Mohammed Omer Shakeel Ahmed
+- **abs:** https://arxiv.org/abs/2603.04595 · **PDF:** https://arxiv.org/pdf/2603.04595v1
+
+## 7704. From Static Inference to Dynamic Interaction: A Survey of Streaming Large Language Models
+
+- **ID:** `2603.04592` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Junlong Tong, Zilong Wang, YuJie Ren, et al.
+- **abs:** https://arxiv.org/abs/2603.04592 · **PDF:** https://arxiv.org/pdf/2603.04592v2
+
+## 7705. ECG-MoE: Mixture-of-Expert Electrocardiogram Foundation Model
+
+- **ID:** `2603.04589` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Yuhao Xu, Xiaoda Wang, Yi Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.04589 · **PDF:** https://arxiv.org/pdf/2603.04589v1
+
+## 7706. Self-Attribution Bias: When AI Monitors Go Easy on Themselves
+
+- **ID:** `2603.04582` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Dipika Khullar, Jack Hopkins, Rowan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.04582 · **PDF:** https://arxiv.org/pdf/2603.04582v1
+
+## 7707. Improving Generative Adversarial Network Generalization for Facial Expression Synthesis
+
+- **ID:** `2603.15648` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Arbish Akram, Nazar Khan, Arif Mahmood
+- **abs:** https://arxiv.org/abs/2603.15648 · **PDF:** https://arxiv.org/pdf/2603.15648v1
+
+## 7708. Why Do Neural Networks Forget: A Study of Collapse in Continual Learning
+
+- **ID:** `2603.04580` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Yunqin Zhu, Jun Jin
+- **abs:** https://arxiv.org/abs/2603.04580 · **PDF:** https://arxiv.org/pdf/2603.04580v1
+
+## 7709. Fusion and Grouping Strategies in Deep Learning for Local Climate Zone Classification of Multimodal Remote Sensing Data
+
+- **ID:** `2603.04562` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Ancymol Thomas, Jaya Sreevalsan-Nair
+- **abs:** https://arxiv.org/abs/2603.04562 · **PDF:** https://arxiv.org/pdf/2603.04562v1
+
+## 7710. Latent Particle World Models: Self-supervised Object-centric Stochastic Dynamics Modeling
+
+- **ID:** `2603.04553` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Tal Daniel, Carl Qi, Dan Haramati, et al.
+- **abs:** https://arxiv.org/abs/2603.04553 · **PDF:** https://arxiv.org/pdf/2603.04553v1
+
+## 7711. Weather-Related Crash Risk Forecasting: A Deep Learning Approach for Heterogenous Spatiotemporal Data
+
+- **ID:** `2603.04551` · **primary:** `stat.AP` · **Submitted:** 2026-03-04
+- **Authors:** Abimbola Ogungbire, Srinivas Pulugurtha
+- **abs:** https://arxiv.org/abs/2603.04551 · **PDF:** https://arxiv.org/pdf/2603.04551v1
+
+## 7712. Adaptive Memory Admission Control for LLM Agents
+
+- **ID:** `2603.04549` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Guilin Zhang, Wei Jiang, Xiejiashan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.04549 · **PDF:** https://arxiv.org/pdf/2603.04549v1
+
+## 7713. Oracle-efficient Hybrid Learning with Constrained Adversaries
+
+- **ID:** `2603.04546` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Princewill Okoroafor, Robert Kleinberg, Michael P. Kim
+- **abs:** https://arxiv.org/abs/2603.04546 · **PDF:** https://arxiv.org/pdf/2603.04546v1
+
+## 7714. An LLM-Guided Query-Aware Inference System for GNN Models on Large Knowledge Graphs
+
+- **ID:** `2603.04545` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Waleed Afandi, Hussein Abdallah, Ashraf Aboulnaga, et al.
+- **abs:** https://arxiv.org/abs/2603.04545 · **PDF:** https://arxiv.org/pdf/2603.04545v1
+
+## 7715. How Professional Visual Artists are Negotiating Generative AI in the Workplace
+
+- **ID:** `2603.04537` · **primary:** `cs.HC` · **Submitted:** 2026-03-04
+- **Authors:** Harry H. Jiang, Jordan Taylor, William Agnew
+- **abs:** https://arxiv.org/abs/2603.04537 · **PDF:** https://arxiv.org/pdf/2603.04537v2
+
+## 7716. A Fast Generative Framework for High-dimensional Posterior Sampling: Application to CMB Delensing
+
+- **ID:** `2603.04535` · **primary:** `astro-ph.IM` · **Submitted:** 2026-03-04
+- **Authors:** Hadi Sotoudeh, Pablo Lemos, Laurence Perreault-Levasseur
+- **abs:** https://arxiv.org/abs/2603.04535 · **PDF:** https://arxiv.org/pdf/2603.04535v1
+
+## 7717. Invariant Causal Routing for Governing Social Norms in Online Market Economies
+
+- **ID:** `2603.04534` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Xiangning Yu, Qirui Mi, Xiao Xue, et al.
+- **abs:** https://arxiv.org/abs/2603.04534 · **PDF:** https://arxiv.org/pdf/2603.04534v1
+
+## 7718. Still Fresh? Evaluating Temporal Drift in Retrieval Benchmarks
+
+- **ID:** `2603.04532` · **primary:** `cs.IR` · **Submitted:** 2026-03-04
+- **Authors:** Nathan Kuissi, Suraj Subrahmanyan, Nandan Thakur, et al.
+- **abs:** https://arxiv.org/abs/2603.04532 · **PDF:** https://arxiv.org/pdf/2603.04532v1
+
+## 7719. Discovering mathematical concepts through a multi-agent system
+
+- **ID:** `2603.04528` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Daattavya Aggarwal, Oisin Kim, Carl Henrik Ek, et al.
+- **abs:** https://arxiv.org/abs/2603.04528 · **PDF:** https://arxiv.org/pdf/2603.04528v2
+
+## 7720. The Volterra signature
+
+- **ID:** `2603.04525` · **primary:** `stat.ML` · **Submitted:** 2026-03-04
+- **Authors:** Paul P. Hager, Fabian N. Harang, Luca Pelizzari, et al.
+- **abs:** https://arxiv.org/abs/2603.04525 · **PDF:** https://arxiv.org/pdf/2603.04525v1
+
+## 7721. Projected Hessian Learning: Fast Curvature Supervision for Accurate Machine-Learning Interatomic Potentials
+
+- **ID:** `2603.04523` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-04
+- **Authors:** Austin Rodriguez, Justin S. Smith, Sakib Matin, et al.
+- **abs:** https://arxiv.org/abs/2603.04523 · **PDF:** https://arxiv.org/pdf/2603.04523v1
+
+## 7722. Augmenting representations with scientific papers
+
+- **ID:** `2603.04516` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Nicolò Oreste Pinciroli Vago, Rocco Di Tella, Carolina Cuesta-Lázaro, et al.
+- **abs:** https://arxiv.org/abs/2603.04516 · **PDF:** https://arxiv.org/pdf/2603.04516v1
+
+## 7723. Progressive Refinement Regulation for Accelerating Diffusion Language Model Decoding
+
+- **ID:** `2603.04514` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Lipeng Wan, Jianhui Gu, Junjie Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.04514 · **PDF:** https://arxiv.org/pdf/2603.04514v1
+
+## 7724. SimpliHuMoN: Simplifying Human Motion Prediction
+
+- **ID:** `2603.04399` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Aadya Agrawal, Alexander Schwing
+- **abs:** https://arxiv.org/abs/2603.04399 · **PDF:** https://arxiv.org/pdf/2603.04399v1
+
+## 7725. Accurate and Efficient Hybrid-Ensemble Atmospheric Data Assimilation in Latent Space with Uncertainty Quantification
+
+- **ID:** `2603.04395` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Hang Fan, Juan Nathaniel, Yi Xiao, et al.
+- **abs:** https://arxiv.org/abs/2603.04395 · **PDF:** https://arxiv.org/pdf/2603.04395v1
+
+## 7726. SELDON: Supernova Explosions Learned by Deep ODE Networks
+
+- **ID:** `2603.04392` · **primary:** `astro-ph.IM` · **Submitted:** 2026-03-04
+- **Authors:** Jiezhong Wu, Jack O'Brien, Jennifer Li, et al.
+- **abs:** https://arxiv.org/abs/2603.04392 · **PDF:** https://arxiv.org/pdf/2603.04392v1
+
+## 7727. A Dual-Helix Governance Approach Towards Reliable Agentic AI for WebGIS Development
+
+- **ID:** `2603.04390` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Boyuan, Guan, Wencong Cui, et al.
+- **abs:** https://arxiv.org/abs/2603.04390 · **PDF:** https://arxiv.org/pdf/2603.04390v1
+
+## 7728. ZipMap: Linear-Time Stateful 3D Reconstruction via Test-Time Training
+
+- **ID:** `2603.04385` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Haian Jin, Rundi Wu, Tianyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.04385 · **PDF:** https://arxiv.org/pdf/2603.04385v2
+
+## 7729. AgentIR: Reasoning-Aware Retrieval for Deep Research Agents
+
+- **ID:** `2603.04384` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Zijian Chen, Xueguang Ma, Shengyao Zhuang, et al.
+- **abs:** https://arxiv.org/abs/2603.04384 · **PDF:** https://arxiv.org/pdf/2603.04384v3
+
+## 7730. Turning Trust to Transactions: Tracking Affiliate Marketing and FTC Compliance in YouTube's Influencer Economy
+
+- **ID:** `2603.04383` · **primary:** `cs.CY` · **Submitted:** 2026-03-04
+- **Authors:** Chen Sun, Yash Vekaria, Zubair Shafiq, et al.
+- **abs:** https://arxiv.org/abs/2603.04383 · **PDF:** https://arxiv.org/pdf/2603.04383v1
+
+## 7731. TaxonRL: Reinforcement Learning with Intermediate Rewards for Interpretable Fine-Grained Visual Reasoning
+
+- **ID:** `2603.04380` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Maximilian von Klinski, Maximilian Schall
+- **abs:** https://arxiv.org/abs/2603.04380 · **PDF:** https://arxiv.org/pdf/2603.04380v1
+
+## 7732. GeoBlock: Inferring Block Granularity from Dependency Geometry in Diffusion Language Models
+
+- **ID:** `2603.26675` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Lipeng Wan, Junjie Ma, Jianhui Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.26675 · **PDF:** https://arxiv.org/pdf/2603.26675v1
+
+## 7733. Robustness of Agentic AI Systems via Adversarially-Aligned Jacobian Regularization
+
+- **ID:** `2603.04378` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Furkan Mumcu, Yasin Yilmaz
+- **abs:** https://arxiv.org/abs/2603.04378 · **PDF:** https://arxiv.org/pdf/2603.04378v1
+
+## 7734. $τ$-Knowledge: Evaluating Conversational Agents over Unstructured Knowledge
+
+- **ID:** `2603.04370` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Quan Shi, Alexandra Zytek, Pedram Razavi, et al.
+- **abs:** https://arxiv.org/abs/2603.04370 · **PDF:** https://arxiv.org/pdf/2603.04370v1
+
+## 7735. Low-Resource Guidance for Controllable Latent Audio Diffusion
+
+- **ID:** `2603.04366` · **primary:** `cs.SD` · **Submitted:** 2026-03-04
+- **Authors:** Zachary Novack, Zack Zukowski, CJ Carr, et al.
+- **abs:** https://arxiv.org/abs/2603.04366 · **PDF:** https://arxiv.org/pdf/2603.04366v1
+
+## 7736. Dual-Modality Multi-Stage Adversarial Safety Training: Robustifying Multimodal Web Agents Against Cross-Modal Attacks
+
+- **ID:** `2603.04364` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Haoyu Liu, Dingcheng Li, Lukas Rutishauser, et al.
+- **abs:** https://arxiv.org/abs/2603.04364 · **PDF:** https://arxiv.org/pdf/2603.04364v1
+
+## 7737. Robust Unscented Kalman Filtering via Recurrent Meta-Adaptation of Sigma-Point Weights
+
+- **ID:** `2603.04360` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Kenan Majewski, Michał Modzelewski, Marcin Żugaj, et al.
+- **abs:** https://arxiv.org/abs/2603.04360 · **PDF:** https://arxiv.org/pdf/2603.04360v1
+
+## 7738. Dissecting Quantization Error: A Concentration-Alignment Perspective
+
+- **ID:** `2603.04359` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Marco Federici, Boris van Breugel, Paul Whatmough, et al.
+- **abs:** https://arxiv.org/abs/2603.04359 · **PDF:** https://arxiv.org/pdf/2603.04359v1
+
+## 7739. Feature-level Interaction Explanations in Multimodal Transformers
+
+- **ID:** `2603.13326` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Yeji Kim, Housam Khalifa Bashier Babiker, Mi-Young Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.13326 · **PDF:** https://arxiv.org/pdf/2603.13326v1
+
+## 7740. RoboCasa365: A Large-Scale Simulation Framework for Training and Benchmarking Generalist Robots
+
+- **ID:** `2603.04356` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Soroush Nasiriany, Sepehr Nasiriany, Abhiram Maddukuri, et al.
+- **abs:** https://arxiv.org/abs/2603.04356 · **PDF:** https://arxiv.org/pdf/2603.04356v1
+
+## 7741. Efficient Refusal Ablation in LLM through Optimal Transport
+
+- **ID:** `2603.04355` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Geraldin Nanfack, Eugene Belilovsky, Elvis Dohmatob
+- **abs:** https://arxiv.org/abs/2603.04355 · **PDF:** https://arxiv.org/pdf/2603.04355v1
+
+## 7742. Out-of-distribution transfer of PDE foundation models to material dynamics under extreme loading
+
+- **ID:** `2603.04354` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Mahindra Rautela, Alexander Most, Siddharth Mansingh, et al.
+- **abs:** https://arxiv.org/abs/2603.04354 · **PDF:** https://arxiv.org/pdf/2603.04354v1
+
+## 7743. A Constrained RL Approach for Cost-Efficient Delivery of Latency-Sensitive Applications
+
+- **ID:** `2603.04353` · **primary:** `cs.NI` · **Submitted:** 2026-03-04
+- **Authors:** Ozan Aygün, Vincenzo Norman Vitale, Antonia M. Tulino, et al.
+- **abs:** https://arxiv.org/abs/2603.04353 · **PDF:** https://arxiv.org/pdf/2603.04353v1
+
+## 7744. RANGER: Sparsely-Gated Mixture-of-Experts with Adaptive Retrieval Re-ranking for Pathology Report Generation
+
+- **ID:** `2603.04348` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Yixin Chen, Ziyu Su, Hikmat Khan, et al.
+- **abs:** https://arxiv.org/abs/2603.04348 · **PDF:** https://arxiv.org/pdf/2603.04348v1
+
+## 7745. AbAffinity: A Large Language Model for Predicting Antibody Binding Affinity against SARS-CoV-2
+
+- **ID:** `2603.04480` · **primary:** `q-bio.QM` · **Submitted:** 2026-03-04
+- **Authors:** Faisal Bin Ashraf, Animesh Ray, Stefano Lonardi
+- **abs:** https://arxiv.org/abs/2603.04480 · **PDF:** https://arxiv.org/pdf/2603.04480v1
+
+## 7746. Enhancing Authorship Attribution with Synthetic Paintings
+
+- **ID:** `2603.04343` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Clarissa Loures, Caio Hosken, Luan Oliveira, et al.
+- **abs:** https://arxiv.org/abs/2603.04343 · **PDF:** https://arxiv.org/pdf/2603.04343v1
+
+## 7747. Balancing Fidelity, Utility, and Privacy in Synthetic Cardiac MRI Generation: A Comparative Study
+
+- **ID:** `2603.04340` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Madhura Edirisooriya, Dasuni Kawya, Ishan Kumarasinghe, et al.
+- **abs:** https://arxiv.org/abs/2603.04340 · **PDF:** https://arxiv.org/pdf/2603.04340v1
+
+## 7748. Pointer-CAD: Unifying B-Rep and Command Sequences via Pointer-based Edges & Faces Selection
+
+- **ID:** `2603.04337` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Dacheng Qi, Chenyu Wang, Jingwei Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.04337 · **PDF:** https://arxiv.org/pdf/2603.04337v1
+
+## 7749. SpotIt+: Verification-based Text-to-SQL Evaluation with Database Constraints
+
+- **ID:** `2603.04334` · **primary:** `cs.DB` · **Submitted:** 2026-03-04
+- **Authors:** Andrew Tremante, Yang He, Rocky Klopfenstein, et al.
+- **abs:** https://arxiv.org/abs/2603.04334 · **PDF:** https://arxiv.org/pdf/2603.04334v2
+
+## 7750. What Does Flow Matching Bring To TD Learning?
+
+- **ID:** `2603.04333` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Bhavya Agrawalla, Michal Nauman, Aviral Kumar
+- **abs:** https://arxiv.org/abs/2603.04333 · **PDF:** https://arxiv.org/pdf/2603.04333v1
+
+## 7751. Algorithmic Compliance and Regulatory Loss in Digital Assets
+
+- **ID:** `2603.04328` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Khem Raj Bhatt, Krishna Sharma
+- **abs:** https://arxiv.org/abs/2603.04328 · **PDF:** https://arxiv.org/pdf/2603.04328v1
+
+## 7752. Scalable Evaluation of the Realism of Synthetic Environmental Augmentations in Images
+
+- **ID:** `2603.04325` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Damian J. Ruck, Paul Vautravers, Oliver Chalkley, et al.
+- **abs:** https://arxiv.org/abs/2603.04325 · **PDF:** https://arxiv.org/pdf/2603.04325v1
+
+## 7753. PTOPOFL: Privacy-Preserving Personalised Federated Learning via Persistent Homology
+
+- **ID:** `2603.04323` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Kelly L Vomo-Donfack, Adryel Hoszu, Grégory Ginot, et al.
+- **abs:** https://arxiv.org/abs/2603.04323 · **PDF:** https://arxiv.org/pdf/2603.04323v1
+
+## 7754. SPRINT: Semi-supervised Prototypical Representation for Few-Shot Class-Incremental Tabular Learning
+
+- **ID:** `2603.04321` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Umid Suleymanov, Murat Kantarcioglu, Kevin S Chan, et al.
+- **abs:** https://arxiv.org/abs/2603.04321 · **PDF:** https://arxiv.org/pdf/2603.04321v1
+
+## 7755. AILS-NTUA at SemEval-2026 Task 12: Graph-Based Retrieval and Reflective Prompting for Abductive Event Reasoning
+
+- **ID:** `2603.04319` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Nikolas Karafyllis, Maria Lymperaiou, Giorgos Filandrianos, et al.
+- **abs:** https://arxiv.org/abs/2603.04319 · **PDF:** https://arxiv.org/pdf/2603.04319v1
+
+## 7756. World Properties without World Models: Recovering Spatial and Temporal Structure from Co-occurrence Statistics in Static Word Embeddings
+
+- **ID:** `2603.04317` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Elan Barenholtz
+- **abs:** https://arxiv.org/abs/2603.04317 · **PDF:** https://arxiv.org/pdf/2603.04317v1
+
+## 7757. MOO: A Multi-view Oriented Observations Dataset for Viewpoint Analysis in Cattle Re-Identification
+
+- **ID:** `2603.04314` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** William Grolleau, Achraf Chaouch, Astrid Sabourin, et al.
+- **abs:** https://arxiv.org/abs/2603.04314 · **PDF:** https://arxiv.org/pdf/2603.04314v1
+
+## 7758. CRESTomics: Analyzing Carotid Plaques in the CREST-2 Trial with a New Additive Classification Model
+
+- **ID:** `2603.04309` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Pranav Kulkarni, Brajesh K. Lal, Georges Jreij, et al.
+- **abs:** https://arxiv.org/abs/2603.04309 · **PDF:** https://arxiv.org/pdf/2603.04309v1
+
+## 7759. Activation Outliers in Transformer Quantization: Reproduction, Statistical Analysis, and Deployment Tradeoffs
+
+- **ID:** `2603.04308` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Pranav Kumar Kaliaperumal
+- **abs:** https://arxiv.org/abs/2603.04308 · **PDF:** https://arxiv.org/pdf/2603.04308v1
+
+## 7760. $V_1$: Unifying Generation and Self-Verification for Parallel Reasoners
+
+- **ID:** `2603.04304` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Harman Singh, Xiuyu Li, Kusha Sareen, et al.
+- **abs:** https://arxiv.org/abs/2603.04304 · **PDF:** https://arxiv.org/pdf/2603.04304v1
+
+## 7761. LUMINA: Foundation Models for Topology Transferable ACOPF
+
+- **ID:** `2603.04300` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Yijiang Li, Zeeshan Memon, Hongwei Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.04300 · **PDF:** https://arxiv.org/pdf/2603.04300v1
+
+## 7762. The Company You Keep: How LLMs Respond to Dark Triad Traits
+
+- **ID:** `2603.04299` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Zeyi Lu, Angelica Henestrosa, Pavel Chizhov, et al.
+- **abs:** https://arxiv.org/abs/2603.04299 · **PDF:** https://arxiv.org/pdf/2603.04299v1
+
+## 7763. Bayesian Modeling of Collatz Stopping Times: A Probabilistic Machine Learning Perspective
+
+- **ID:** `2603.04479` · **primary:** `stat.ML` · **Submitted:** 2026-03-04
+- **Authors:** Nicolò Bonacorsi, Matteo Bordoni
+- **abs:** https://arxiv.org/abs/2603.04479 · **PDF:** https://arxiv.org/pdf/2603.04479v1
+
+## 7764. LabelBuddy: An Open Source Music and Audio Language Annotation Tagging Tool Using AI Assistance
+
+- **ID:** `2603.04293` · **primary:** `cs.SD` · **Submitted:** 2026-03-04
+- **Authors:** Ioannis Prokopiou, Ioannis Sina, Agisilaos Kounelis, et al.
+- **abs:** https://arxiv.org/abs/2603.04293 · **PDF:** https://arxiv.org/pdf/2603.04293v1
+
+## 7765. Position: Vector Prompt Interfaces Should Be Exposed to Enable Customization of Large Language Models
+
+- **ID:** `2603.04292` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Liangwei Yang, Shiyu Wang, Haolin Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.04292 · **PDF:** https://arxiv.org/pdf/2603.04292v1
+
+## 7766. CubeComposer: Spatio-Temporal Autoregressive 4K 360° Video Generation from Perspective Video
+
+- **ID:** `2603.04291` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Lingen Li, Guangzhi Wang, Xiaoyu Li, et al.
+- **abs:** https://arxiv.org/abs/2603.04291 · **PDF:** https://arxiv.org/pdf/2603.04291v1
+
+## 7767. IPD: Boosting Sequential Policy with Imaginary Planning Distillation in Offline Reinforcement Learning
+
+- **ID:** `2603.04289` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Yihao Qin, Yuanfei Wang, Hang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.04289 · **PDF:** https://arxiv.org/pdf/2603.04289v1
+
+## 7768. VANGUARD: Vehicle-Anchored Ground Sample Distance Estimation for UAVs in GPS-Denied Environments
+
+- **ID:** `2603.04277` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Yifei Chen, Xupeng Chen, Feng Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.04277 · **PDF:** https://arxiv.org/pdf/2603.04277v1
+
+## 7769. Causality Elicitation from Large Language Models
+
+- **ID:** `2603.04276` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Takashi Kameyama, Masahiro Kato, Yasuko Hio, et al.
+- **abs:** https://arxiv.org/abs/2603.04276 · **PDF:** https://arxiv.org/pdf/2603.04276v1
+
+## 7770. Statistical Inference for Score Decompositions
+
+- **ID:** `2603.04275` · **primary:** `econ.EM` · **Submitted:** 2026-03-04
+- **Authors:** Timo Dimitriadis, Marius Puke
+- **abs:** https://arxiv.org/abs/2603.04275 · **PDF:** https://arxiv.org/pdf/2603.04275v1
+
+## 7771. When AI Fails, What Works? A Data-Driven Taxonomy of Real-World AI Risk Mitigation Strategies
+
+- **ID:** `2603.04259` · **primary:** `cs.CY` · **Submitted:** 2026-03-04
+- **Authors:** Evgenija Popchanovska, Ana Gjorgjevikj, Maryan Rizinski, et al.
+- **abs:** https://arxiv.org/abs/2603.04259 · **PDF:** https://arxiv.org/pdf/2603.04259v1
+
+## 7772. Auditing Cascading Risks in Multi-Agent Systems via Semantic-Geometric Co-evolution
+
+- **ID:** `2603.13325` · **primary:** `cs.MA` · **Submitted:** 2026-03-04
+- **Authors:** Zixun Luo, Yuhang Fan, Hengyu Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.13325 · **PDF:** https://arxiv.org/pdf/2603.13325v1
+
+## 7773. Memex(RL): Scaling Long-Horizon LLM Agents via Indexed Experience Memory
+
+- **ID:** `2603.04257` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Zhenting Wang, Huancheng Chen, Jiayun Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.04257 · **PDF:** https://arxiv.org/pdf/2603.04257v1
+
+## 7774. Standing on the Shoulders of Giants: Rethinking EEG Foundation Model Pretraining via Multi-Teacher Distillation
+
+- **ID:** `2603.04478` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Chenqi Li, Yu Liu, Shuo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.04478 · **PDF:** https://arxiv.org/pdf/2603.04478v1
+
+## 7775. Online Learning for Multi-Layer Hierarchical Inference under Partial and Policy-Dependent Feedback
+
+- **ID:** `2603.04247` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Haoran Zhang, Seohyeon Cha, Hasan Burhan Beytur, et al.
+- **abs:** https://arxiv.org/abs/2603.04247 · **PDF:** https://arxiv.org/pdf/2603.04247v1
+
+## 7776. LikeThis! Empowering App Users to Submit UI Improvement Suggestions Instead of Complaints
+
+- **ID:** `2603.04245` · **primary:** `cs.SE` · **Submitted:** 2026-03-04
+- **Authors:** Jialiang Wei, Ali Ebrahimi Pourasad, Walid Maalej
+- **abs:** https://arxiv.org/abs/2603.04245 · **PDF:** https://arxiv.org/pdf/2603.04245v1
+
+## 7777. FeedAIde: Guiding App Users to Submit Rich Feedback Reports by Asking Context-Aware Follow-Up Questions
+
+- **ID:** `2603.04244` · **primary:** `cs.SE` · **Submitted:** 2026-03-04
+- **Authors:** Ali Ebrahimi Pourasad, Meyssam Saghiri, Walid Maalej
+- **abs:** https://arxiv.org/abs/2603.04244 · **PDF:** https://arxiv.org/pdf/2603.04244v1
+
+## 7778. The Challenge of Out-Of-Distribution Detection in Motor Imagery BCIs
+
+- **ID:** `2603.13324` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Merlijn Quincent Mulder, Matias Valdenegro-Toro, Andreea Ioana Sburlea, et al.
+- **abs:** https://arxiv.org/abs/2603.13324 · **PDF:** https://arxiv.org/pdf/2603.13324v1
+
+## 7779. Agentics 2.0: Logical Transduction Algebra for Agentic Data Workflows
+
+- **ID:** `2603.04241` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Alfio Massimiliano Gliozzo, Junkyu Lee, Nahuel Defosse
+- **abs:** https://arxiv.org/abs/2603.04241 · **PDF:** https://arxiv.org/pdf/2603.04241v1
+
+## 7780. Modular Neural Computer
+
+- **ID:** `2603.13323` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Florin Leon
+- **abs:** https://arxiv.org/abs/2603.13323 · **PDF:** https://arxiv.org/pdf/2603.13323v1
+
+## 7781. Retrieval or Representation? Reassessing Benchmark Gaps in Multilingual and Visually Rich RAG
+
+- **ID:** `2603.04238` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Martin Asenov, Kenza Benkirane, Dan Goldwater, et al.
+- **abs:** https://arxiv.org/abs/2603.04238 · **PDF:** https://arxiv.org/pdf/2603.04238v1
+
+## 7782. Nearest-Neighbor Density Estimation for Dependency Suppression
+
+- **ID:** `2603.04224` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Kathleen Anderson, Thomas Martinetz
+- **abs:** https://arxiv.org/abs/2603.04224 · **PDF:** https://arxiv.org/pdf/2603.04224v1
+
+## 7783. Semi-Supervised Generative Learning via Latent Space Distribution Matching
+
+- **ID:** `2603.04223` · **primary:** `stat.ML` · **Submitted:** 2026-03-04
+- **Authors:** Kwong Yu Chong, Long Feng
+- **abs:** https://arxiv.org/abs/2603.04223 · **PDF:** https://arxiv.org/pdf/2603.04223v1
+
+## 7784. PRAM-R: A Perception-Reasoning-Action-Memory Framework with LLM-Guided Modality Routing for Adaptive Autonomous Driving
+
+- **ID:** `2603.04222` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Yi Zhang, Xian Zhang, Saisi Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.04222 · **PDF:** https://arxiv.org/pdf/2603.04222v1
+
+## 7785. ZeSTA: Zero-Shot TTS Augmentation with Domain-Conditioned Training for Data-Efficient Personalized Speech Synthesis
+
+- **ID:** `2603.04219` · **primary:** `cs.SD` · **Submitted:** 2026-03-04
+- **Authors:** Youngwon Choi, Jinwoo Oh, Hwayeon Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.04219 · **PDF:** https://arxiv.org/pdf/2603.04219v1
+
+## 7786. When Do Language Models Endorse Limitations on Human Rights Principles?
+
+- **ID:** `2603.04217` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Keenan Samway, Nicole Miu Takagi, Rada Mihalcea, et al.
+- **abs:** https://arxiv.org/abs/2603.04217 · **PDF:** https://arxiv.org/pdf/2603.04217v1
+
+## 7787. Code Fingerprints: Disentangled Attribution of LLM-Generated Code
+
+- **ID:** `2603.04212` · **primary:** `cs.SE` · **Submitted:** 2026-03-04
+- **Authors:** Jiaxun Guo, Ziyuan Yang, Mengyu Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.04212 · **PDF:** https://arxiv.org/pdf/2603.04212v1
+
+## 7788. Beyond Edge Deletion: A Comprehensive Approach to Counterfactual Explanation in Graph Neural Networks
+
+- **ID:** `2603.04209` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Matteo De Sanctis, Riccardo De Sanctis, Stefano Faralli, et al.
+- **abs:** https://arxiv.org/abs/2603.04209 · **PDF:** https://arxiv.org/pdf/2603.04209v1
+
+## 7789. Beyond Mixtures and Products for Ensemble Aggregation: A Likelihood Perspective on Generalized Means
+
+- **ID:** `2603.04204` · **primary:** `stat.ML` · **Submitted:** 2026-03-04
+- **Authors:** Raphaël Razafindralambo, Rémy Sun, Frédéric Precioso, et al.
+- **abs:** https://arxiv.org/abs/2603.04204 · **PDF:** https://arxiv.org/pdf/2603.04204v1
+
+## 7790. Bayesian Adversarial Privacy
+
+- **ID:** `2603.04199` · **primary:** `math.ST` · **Submitted:** 2026-03-04
+- **Authors:** Cameron Bell, Timothy Johnston, Antoine Luciano, et al.
+- **abs:** https://arxiv.org/abs/2603.04199 · **PDF:** https://arxiv.org/pdf/2603.04199v1
+
+## 7791. Stable and Steerable Sparse Autoencoders with Weight Regularization
+
+- **ID:** `2603.04198` · **primary:** `stat.ML` · **Submitted:** 2026-03-04
+- **Authors:** Piotr Jedryszek, Oliver M. Crook
+- **abs:** https://arxiv.org/abs/2603.04198 · **PDF:** https://arxiv.org/pdf/2603.04198v1
+
+## 7792. The Arrival of AGI? When Expert Personas Exceed Expert Benchmarks
+
+- **ID:** `2603.20225` · **primary:** `cs.CY` · **Submitted:** 2026-03-04
+- **Authors:** Drake Mullens, Stella Shen
+- **abs:** https://arxiv.org/abs/2603.20225 · **PDF:** https://arxiv.org/pdf/2603.20225v1
+
+## 7793. Noise-aware Client Selection for carbon-efficient Federated Learning via Gradient Norm Thresholding
+
+- **ID:** `2603.04194` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Patrick Wilhelm, Inese Yilmaz, Odej Kao
+- **abs:** https://arxiv.org/abs/2603.04194 · **PDF:** https://arxiv.org/pdf/2603.04194v1
+
+## 7794. Towards Realistic Personalization: Evaluating Long-Horizon Preference Following in Personalized User-LLM Interactions
+
+- **ID:** `2603.04191` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Qianyun Guo, Yibo Li, Yue Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.04191 · **PDF:** https://arxiv.org/pdf/2603.04191v1
+
+## 7795. CAM-LDS: Cyber Attack Manifestations for Automatic Interpretation of System Logs and Security Alerts
+
+- **ID:** `2603.04186` · **primary:** `cs.CR` · **Submitted:** 2026-03-04
+- **Authors:** Max Landauer, Wolfgang Hotwagner, Thorina Boenke, et al.
+- **abs:** https://arxiv.org/abs/2603.04186 · **PDF:** https://arxiv.org/pdf/2603.04186v1
+
+## 7796. Nepali Passport Question Answering: A Low-Resource Dataset for Public Service Applications
+
+- **ID:** `2603.13320` · **primary:** `cs.IR` · **Submitted:** 2026-03-04
+- **Authors:** Funghang Limbu Begha, Praveen Acharya, Bal Krishna Bal
+- **abs:** https://arxiv.org/abs/2603.13320 · **PDF:** https://arxiv.org/pdf/2603.13320v1
+
+## 7797. An intuitive rearranging of the Yates covariance decomposition for probabilistic verification of forecasts with the Brier score
+
+- **ID:** `2603.05544` · **primary:** `stat.ME` · **Submitted:** 2026-03-04
+- **Authors:** Bruno Hebling Vieira
+- **abs:** https://arxiv.org/abs/2603.05544 · **PDF:** https://arxiv.org/pdf/2603.05544v1
+
+## 7798. REDNET-ML: A Multi-Sensor Machine Learning Pipeline for Harmful Algal Bloom Risk Detection Along the Omani Coast
+
+- **ID:** `2603.04181` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Ameer Alhashemi
+- **abs:** https://arxiv.org/abs/2603.04181 · **PDF:** https://arxiv.org/pdf/2603.04181v1
+
+## 7799. Architectural Proprioception in State Space Models: Thermodynamic Training Induces Anticipatory Halt Detection
+
+- **ID:** `2603.04180` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Jay Noon
+- **abs:** https://arxiv.org/abs/2603.04180 · **PDF:** https://arxiv.org/pdf/2603.04180v1
+
+## 7800. CodeTaste: Can LLMs Generate Human-Level Code Refactorings?
+
+- **ID:** `2603.04177` · **primary:** `cs.SE` · **Submitted:** 2026-03-04
+- **Authors:** Alex Thillen, Niels Mündler, Veselin Raychev, et al.
+- **abs:** https://arxiv.org/abs/2603.04177 · **PDF:** https://arxiv.org/pdf/2603.04177v1
+
+## 7801. Activity Recognition from Smart Insole Sensor Data Using a Circular Dilated CNN
+
+- **ID:** `2603.04477` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Yanhua Zhao
+- **abs:** https://arxiv.org/abs/2603.04477 · **PDF:** https://arxiv.org/pdf/2603.04477v1
+
+## 7802. Learning Hip Exoskeleton Control Policy via Predictive Neuromusculoskeletal Simulation
+
+- **ID:** `2603.04166` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Ilseung Park, Changseob Song, Inseung Kang
+- **abs:** https://arxiv.org/abs/2603.04166 · **PDF:** https://arxiv.org/pdf/2603.04166v1
+
+## 7803. PlaneCycle: Training-Free 2D-to-3D Lifting of Foundation Models Without Adapters
+
+- **ID:** `2603.04165` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Yinghong Yu, Guangyuan Li, Jiancheng Yang
+- **abs:** https://arxiv.org/abs/2603.04165 · **PDF:** https://arxiv.org/pdf/2603.04165v2
+
+## 7804. Beyond Test-Time Compute Strategies: Advocating Energy-per-Token in LLM Inference
+
+- **ID:** `2603.20224` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Patrick Wilhelm, Thorsten Wittkopp, Odej Kao
+- **abs:** https://arxiv.org/abs/2603.20224 · **PDF:** https://arxiv.org/pdf/2603.20224v1
+
+## 7805. Bielik-Q2-Sharp: A Comparative Study of Extreme 2-bit Quantization Methods for a Polish 11B Language Model
+
+- **ID:** `2603.04162` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Jakub Prejzner
+- **abs:** https://arxiv.org/abs/2603.04162 · **PDF:** https://arxiv.org/pdf/2603.04162v2
+
+## 7806. Traces of Social Competence in Large Language Models
+
+- **ID:** `2603.04161` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Tom Kouwenhoven, Michiel van der Meer, Max van Duijn
+- **abs:** https://arxiv.org/abs/2603.04161 · **PDF:** https://arxiv.org/pdf/2603.04161v1
+
+## 7807. Quantum-Assisted Optimal Rebalancing with Uncorrelated Asset Selection for Algorithmic Trading Walk-Forward QUBO Scheduling via QAOA
+
+- **ID:** `2603.16904` · **primary:** `q-fin.PM` · **Submitted:** 2026-03-04
+- **Authors:** Abraham Itzhak Weinberg
+- **abs:** https://arxiv.org/abs/2603.16904 · **PDF:** https://arxiv.org/pdf/2603.16904v1
+
+## 7808. GarmentPile++: Affordance-Driven Cluttered Garments Retrieval with Vision-Language Reasoning
+
+- **ID:** `2603.04158` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Mingleyang Li, Yuran Wang, Yue Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.04158 · **PDF:** https://arxiv.org/pdf/2603.04158v1
+
+## 7809. VietNormalizer: An Open-Source, Dependency-Free Python Library for Vietnamese Text Normalization in TTS and NLP Applications
+
+- **ID:** `2603.04145` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Hung Vu Nguyen, Loan Do, Thanh Ngoc Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.04145 · **PDF:** https://arxiv.org/pdf/2603.04145v1
+
+## 7810. A Multi-Agent Framework for Interpreting Multivariate Physiological Time Series
+
+- **ID:** `2603.04142` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Davide Gabrielli, Paola Velardi, Stefano Faralli, et al.
+- **abs:** https://arxiv.org/abs/2603.04142 · **PDF:** https://arxiv.org/pdf/2603.04142v1
+
+## 7811. Unbiased Dynamic Pruning for Efficient Group-Based Policy Optimization
+
+- **ID:** `2603.04135` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Haodong Zhu, Yangyang Ren, Yanjing Li, et al.
+- **abs:** https://arxiv.org/abs/2603.04135 · **PDF:** https://arxiv.org/pdf/2603.04135v1
+
+## 7812. InstMeter: An Instruction-Level Method to Predict Energy and Latency of DL Model Inference on MCUs
+
+- **ID:** `2603.04134` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Hao Liu, Qing Wang, Marco Zuniga
+- **abs:** https://arxiv.org/abs/2603.04134 · **PDF:** https://arxiv.org/pdf/2603.04134v1
+
+## 7813. DISCO: Document Intelligence Suite for COmparative Evaluation
+
+- **ID:** `2603.23511` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Kenza Benkirane, Dan Goldwater, Martin Asenov, et al.
+- **abs:** https://arxiv.org/abs/2603.23511 · **PDF:** https://arxiv.org/pdf/2603.23511v1
+
+## 7814. Exploiting Subgradient Sparsity in Max-Plus Neural Networks
+
+- **ID:** `2603.04133` · **primary:** `stat.ML` · **Submitted:** 2026-03-04
+- **Authors:** Ikhlas Enaieh, Olivier Fercoq
+- **abs:** https://arxiv.org/abs/2603.04133 · **PDF:** https://arxiv.org/pdf/2603.04133v1
+
+## 7815. Two-Stage Photovoltaic Forecasting: Separating Weather Prediction from Plant-Characteristics
+
+- **ID:** `2603.04132` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Philipp Danner, Hermann de Meer
+- **abs:** https://arxiv.org/abs/2603.04132 · **PDF:** https://arxiv.org/pdf/2603.04132v1
+
+## 7816. Crab$^{+}$: A Scalable and Unified Audio-Visual Scene Understanding Model with Explicit Cooperation
+
+- **ID:** `2603.04128` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Dongnuan Cai, Henghui Du, Chang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.04128 · **PDF:** https://arxiv.org/pdf/2603.04128v1
+
+## 7817. Data-Aware Random Feature Kernel for Transformers
+
+- **ID:** `2603.04127` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Amirhossein Farzam, Hossein Mobahi, Nolan Andrew Miller, et al.
+- **abs:** https://arxiv.org/abs/2603.04127 · **PDF:** https://arxiv.org/pdf/2603.04127v1
+
+## 7818. BeamPERL: Parameter-Efficient RL with Verifiable Rewards Specializes Compact LLMs for Structured Beam Mechanics Reasoning
+
+- **ID:** `2603.04124` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Tarjei Paule Hage, Markus J. Buehler
+- **abs:** https://arxiv.org/abs/2603.04124 · **PDF:** https://arxiv.org/pdf/2603.04124v1
+
+## 7819. FINEST: Improving LLM Responses to Sensitive Topics Through Fine-Grained Evaluation
+
+- **ID:** `2603.04123` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Juhyun Oh, Nayeon Lee, Chani Jung, et al.
+- **abs:** https://arxiv.org/abs/2603.04123 · **PDF:** https://arxiv.org/pdf/2603.04123v1
+
+## 7820. FastWave: Optimized Diffusion Model for Audio Super-Resolution
+
+- **ID:** `2603.04122` · **primary:** `cs.SD` · **Submitted:** 2026-03-04
+- **Authors:** Nikita Kuznetsov, Maksim Kaledin
+- **abs:** https://arxiv.org/abs/2603.04122 · **PDF:** https://arxiv.org/pdf/2603.04122v1
+
+## 7821. When to restart? Exploring escalating restarts on convergence
+
+- **ID:** `2603.04117` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Ayush K. Varshney, Šarūnas Girdzijauskas, Konstantinos Vandikas, et al.
+- **abs:** https://arxiv.org/abs/2603.04117 · **PDF:** https://arxiv.org/pdf/2603.04117v1
+
+## 7822. Understanding Sources of Demographic Predictability in Brain MRI via Disentangling Anatomy and Contrast
+
+- **ID:** `2603.04113` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Mehmet Yigit Avci, Akshit Achara, Andrew King, et al.
+- **abs:** https://arxiv.org/abs/2603.04113 · **PDF:** https://arxiv.org/pdf/2603.04113v1
+
+## 7823. Testing Full Mediation of Treatment Effects and the Identifiability of Causal Mechanisms
+
+- **ID:** `2603.04109` · **primary:** `econ.EM` · **Submitted:** 2026-03-04
+- **Authors:** Martin Huber, Kevin Kloiber, Lukáš Lafférs
+- **abs:** https://arxiv.org/abs/2603.04109 · **PDF:** https://arxiv.org/pdf/2603.04109v1
+
+## 7824. Visuospatial Perspective Taking in Multimodal Language Models
+
+- **ID:** `2603.23510` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Jonathan Prunty, Seraphina Zhang, Patrick Quinn, et al.
+- **abs:** https://arxiv.org/abs/2603.23510 · **PDF:** https://arxiv.org/pdf/2603.23510v1
+
+## 7825. Human-Data Interaction, Exploration, and Visualization in the AI Era: Challenges and Opportunities
+
+- **ID:** `2603.05542` · **primary:** `cs.DB` · **Submitted:** 2026-03-04
+- **Authors:** Jean-Daniel Fekete, Yifan Hu, Dominik Moritz, et al.
+- **abs:** https://arxiv.org/abs/2603.05542 · **PDF:** https://arxiv.org/pdf/2603.05542v1
+
+## 7826. Efficient Point Cloud Processing with High-Dimensional Positional Encoding and Non-Local MLPs
+
+- **ID:** `2603.04099` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Yanmei Zou, Hongshan Yu, Yaonan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.04099 · **PDF:** https://arxiv.org/pdf/2603.04099v1
+
+## 7827. Lyapunov Stability of Stochastic Vector Optimization: Theory and Numerical Implementation
+
+- **ID:** `2603.04095` · **primary:** `math.OC` · **Submitted:** 2026-03-04
+- **Authors:** Thiago Santos, Sebastiao Xavier
+- **abs:** https://arxiv.org/abs/2603.04095 · **PDF:** https://arxiv.org/pdf/2603.04095v1
+
+## 7828. Reducing hyperparameter sensitivity in measurement-feedback based Ising machines
+
+- **ID:** `2603.04093` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Toon Sevenants, Guy Van der Sande, Guy Verschaffelt
+- **abs:** https://arxiv.org/abs/2603.04093 · **PDF:** https://arxiv.org/pdf/2603.04093v1
+
+## 7829. End-to-end event reconstruction for precision physics at future colliders
+
+- **ID:** `2603.04084` · **primary:** `hep-ex` · **Submitted:** 2026-03-04
+- **Authors:** Dolores Garcia, Lena Herrmann, Gregor Krzmanc, et al.
+- **abs:** https://arxiv.org/abs/2603.04084 · **PDF:** https://arxiv.org/pdf/2603.04084v1
+
+## 7830. Hindsight Quality Prediction Experiments in Multi-Candidate Human-Post-Edited Machine Translation
+
+- **ID:** `2603.04083` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Malik Marmonier, Benoît Sagot, Rachel Bawden
+- **abs:** https://arxiv.org/abs/2603.04083 · **PDF:** https://arxiv.org/pdf/2603.04083v1
+
+## 7831. SaFeR: Safety-Critical Scenario Generation for Autonomous Driving Test via Feasibility-Constrained Token Resampling
+
+- **ID:** `2603.04071` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Jinlong Cui, Fenghua Liang, Guo Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.04071 · **PDF:** https://arxiv.org/pdf/2603.04071v1
+
+## 7832. Monitoring Emergent Reward Hacking During Generation via Internal Activations
+
+- **ID:** `2603.04069` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Patrick Wilhelm, Thorsten Wittkopp, Odej Kao
+- **abs:** https://arxiv.org/abs/2603.04069 · **PDF:** https://arxiv.org/pdf/2603.04069v1
+
+## 7833. Tuning Just Enough: Lightweight Backdoor Attacks on Multi-Encoder Diffusion Models
+
+- **ID:** `2603.04064` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Ziyuan Chen, Yujin Jeong, Tobias Braun, et al.
+- **abs:** https://arxiv.org/abs/2603.04064 · **PDF:** https://arxiv.org/pdf/2603.04064v1
+
+## 7834. FedCova: Robust Federated Covariance Learning Against Noisy Labels
+
+- **ID:** `2603.04062` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Xiangyu Zhong, Xiaojun Yuan, Ying-Jun Angela Zhang
+- **abs:** https://arxiv.org/abs/2603.04062 · **PDF:** https://arxiv.org/pdf/2603.04062v1
+
+## 7835. Fermi-Dirac thermal measurements: A framework for quantum hypothesis testing and semidefinite optimization
+
+- **ID:** `2603.04061` · **primary:** `quant-ph` · **Submitted:** 2026-03-04
+- **Authors:** Nana Liu, Mark M. Wilde
+- **abs:** https://arxiv.org/abs/2603.04061 · **PDF:** https://arxiv.org/pdf/2603.04061v1
+
+## 7836. Sim2Sea: Sim-to-Real Policy Transfer for Maritime Vessel Navigation in Congested Waters
+
+- **ID:** `2603.04057` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Xinyu Cui, Xuanfa Jin, Xue Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.04057 · **PDF:** https://arxiv.org/pdf/2603.04057v1
+
+## 7837. Soft Equivariance Regularization for Invariant Self-Supervised Learning
+
+- **ID:** `2603.06693` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Joohyung Lee, Changhun Kim, Hyunsu Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.06693 · **PDF:** https://arxiv.org/pdf/2603.06693v1
+
+## 7838. An Adaptive KKT-Based Indicator for Convergence Assessment in Multi-Objective Optimization
+
+- **ID:** `2603.04053` · **primary:** `math.OC` · **Submitted:** 2026-03-04
+- **Authors:** Thiago Santos, Sebastiao Xavier
+- **abs:** https://arxiv.org/abs/2603.04053 · **PDF:** https://arxiv.org/pdf/2603.04053v1
+
+## 7839. Inference-Time Toxicity Mitigation in Protein Language Models
+
+- **ID:** `2603.04045` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Manuel Fernández Burda, Santiago Aranguri, Iván Arcuschin Moreno, et al.
+- **abs:** https://arxiv.org/abs/2603.04045 · **PDF:** https://arxiv.org/pdf/2603.04045v1
+
+## 7840. Inference Energy and Latency in AI-Mediated Education: A Learning-per-Watt Analysis of Edge and Cloud Models
+
+- **ID:** `2603.20223` · **primary:** `cs.CY` · **Submitted:** 2026-03-04
+- **Authors:** Kushal Khemani
+- **abs:** https://arxiv.org/abs/2603.20223 · **PDF:** https://arxiv.org/pdf/2603.20223v1
+
+## 7841. DQE-CIR: Distinctive Query Embeddings through Learnable Attribute Weights and Target Relative Negative Sampling in Composed Image Retrieval
+
+- **ID:** `2603.04037` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Geon Park, Ji-Hoon Park, Seong-Whan Lee
+- **abs:** https://arxiv.org/abs/2603.04037 · **PDF:** https://arxiv.org/pdf/2603.04037v1
+
+## 7842. mlx-vis: GPU-Accelerated Dimensionality Reduction and Visualization on Apple Silicon
+
+- **ID:** `2603.04035` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Han Xiao
+- **abs:** https://arxiv.org/abs/2603.04035 · **PDF:** https://arxiv.org/pdf/2603.04035v3
+
+## 7843. The Empty Quadrant: AI Teammates for Embodied Field Learning
+
+- **ID:** `2603.04034` · **primary:** `cs.HC` · **Submitted:** 2026-03-04
+- **Authors:** Hyein Kim, Sung Park
+- **abs:** https://arxiv.org/abs/2603.04034 · **PDF:** https://arxiv.org/pdf/2603.04034v1
+
+## 7844. Who Judges the Judge? Evaluating LLM-as-a-Judge for French Medical open-ended QA
+
+- **ID:** `2603.04033` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Ikram Belmadani, Oumaima El Khettari, Pacôme Constant dit Beaufils, et al.
+- **abs:** https://arxiv.org/abs/2603.04033 · **PDF:** https://arxiv.org/pdf/2603.04033v1
+
+## 7845. Multi-Stage Music Source Restoration with BandSplit-RoFormer Separation and HiFi++ GAN
+
+- **ID:** `2603.04032` · **primary:** `cs.SD` · **Submitted:** 2026-03-04
+- **Authors:** Tobias Morocutti, Emmanouil Karystinaios, Jonathan Greif, et al.
+- **abs:** https://arxiv.org/abs/2603.04032 · **PDF:** https://arxiv.org/pdf/2603.04032v1
+
+## 7846. Self-adapting Robotic Agents through Online Continual Reinforcement Learning with World Model Feedback
+
+- **ID:** `2603.04029` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Fabian Domberg, Georg Schildbach
+- **abs:** https://arxiv.org/abs/2603.04029 · **PDF:** https://arxiv.org/pdf/2603.04029v1
+
+## 7847. A Multi-Dimensional Quality Scoring Framework for Decentralized LLM Inference with Proof of Quality
+
+- **ID:** `2603.04028` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Arther Tian, Alex Ding, Frank Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.04028 · **PDF:** https://arxiv.org/pdf/2603.04028v1
+
+## 7848. Volumetric Directional Diffusion: Anchoring Uncertainty Quantification in Anatomical Consensus for Ambiguous Medical Image Segmentation
+
+- **ID:** `2603.04024` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Chao Wu, Kangxian Xie, Mingchen Gao
+- **abs:** https://arxiv.org/abs/2603.04024 · **PDF:** https://arxiv.org/pdf/2603.04024v1
+
+## 7849. Internal Safety Collapse in Frontier Large Language Models
+
+- **ID:** `2603.23509` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Yutao Wu, Xiao Liu, Yifeng Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.23509 · **PDF:** https://arxiv.org/pdf/2603.23509v1
+
+## 7850. Continuous Modal Logical Neural Networks: Modal Reasoning via Stochastic Accessibility
+
+- **ID:** `2603.04019` · **primary:** `cs.LO` · **Submitted:** 2026-03-04
+- **Authors:** Antonin Sulc
+- **abs:** https://arxiv.org/abs/2603.04019 · **PDF:** https://arxiv.org/pdf/2603.04019v1
+
+## 7851. Fast and Faithful: Real-Time Verification for Long-Document Retrieval-Augmented Generation Systems
+
+- **ID:** `2603.23508` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Xunzhuo Liu, Bowei He, Xue Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.23508 · **PDF:** https://arxiv.org/pdf/2603.23508v1
+
+## 7852. Beyond Masks: Efficient, Flexible Diffusion Language Models via Deletion-Insertion Processes
+
+- **ID:** `2603.23507` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Fangyu Ding, Ding Ding, Sijin Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.23507 · **PDF:** https://arxiv.org/pdf/2603.23507v1
+
+## 7853. Fixed-Budget Constrained Best Arm Identification in Grouped Bandits
+
+- **ID:** `2603.04007` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Raunak Mukherjee, Sharayu Moharir
+- **abs:** https://arxiv.org/abs/2603.04007 · **PDF:** https://arxiv.org/pdf/2603.04007v1
+
+## 7854. Training-Free Rate-Distortion-Perception Traversal With Diffusion
+
+- **ID:** `2603.04005` · **primary:** `cs.IT` · **Submitted:** 2026-03-04
+- **Authors:** Yuhan Wang, Suzhi Bi, Ying-Jun Angela Zhang
+- **abs:** https://arxiv.org/abs/2603.04005 · **PDF:** https://arxiv.org/pdf/2603.04005v1
+
+## 7855. Discriminative Perception via Anchored Description for Reasoning Segmentation
+
+- **ID:** `2603.04002` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Tao Yang, Qing Zhou, Yanliang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.04002 · **PDF:** https://arxiv.org/pdf/2603.04002v1
+
+## 7856. STEM Faculty Perspectives on Generative AI in Higher Education
+
+- **ID:** `2603.04001` · **primary:** `cs.CY` · **Submitted:** 2026-03-04
+- **Authors:** Akila de Silva, Isabel Hyo Jung Song, Hui Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.04001 · **PDF:** https://arxiv.org/pdf/2603.04001v1
+
+## 7857. On the Learnability of Offline Model-Based Optimization: A Ranking Perspective
+
+- **ID:** `2603.04000` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Shen-Huan Lyu, Rong-Xi Tan, Ke Xue, et al.
+- **abs:** https://arxiv.org/abs/2603.04000 · **PDF:** https://arxiv.org/pdf/2603.04000v1
+
+## 7858. Spectral Surgery: Training-Free Refinement of LoRA via Gradient-Guided Singular Value Reweighting
+
+- **ID:** `2603.03995` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Zailong Tian, Yanzhe Chen, Zhuoheng Han, et al.
+- **abs:** https://arxiv.org/abs/2603.03995 · **PDF:** https://arxiv.org/pdf/2603.03995v1
+
+## 7859. Specialization of softmax attention heads: insights from the high-dimensional single-location model
+
+- **ID:** `2603.03993` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** M. Sagitova, O. Duranthon, L. Zdeborová
+- **abs:** https://arxiv.org/abs/2603.03993 · **PDF:** https://arxiv.org/pdf/2603.03993v1
+
+## 7860. Measuring AI R&D Automation
+
+- **ID:** `2603.03992` · **primary:** `cs.CY` · **Submitted:** 2026-03-04
+- **Authors:** Alan Chan, Ranay Padarath, Joe Kwon, et al.
+- **abs:** https://arxiv.org/abs/2603.03992 · **PDF:** https://arxiv.org/pdf/2603.03992v3
+
+## 7861. When Visual Evidence is Ambiguous: Pareidolia as a Diagnostic Probe for Vision Models
+
+- **ID:** `2603.03989` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Qianpu Chen, Derya Soydaner, Rob Saunders
+- **abs:** https://arxiv.org/abs/2603.03989 · **PDF:** https://arxiv.org/pdf/2603.03989v1
+
+## 7862. From Language to Action in Arabic: Reliable Structured Tool Calling via Data-Centric Fine-Tuning
+
+- **ID:** `2603.16901` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Omer Nacar, Deema Alquffari, Saleh Alsharideh, et al.
+- **abs:** https://arxiv.org/abs/2603.16901 · **PDF:** https://arxiv.org/pdf/2603.16901v1
+
+## 7863. GeoSeg: Training-Free Reasoning-Driven Segmentation in Remote Sensing Imagery
+
+- **ID:** `2603.03983` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Lifan Jiang, Yuhang Pei, oxi Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.03983 · **PDF:** https://arxiv.org/pdf/2603.03983v1
+
+## 7864. Right in Time: Reactive Reasoning in Regulated Traffic Spaces
+
+- **ID:** `2603.03977` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Simon Kohaut, Benedict Flade, Julian Eggert, et al.
+- **abs:** https://arxiv.org/abs/2603.03977 · **PDF:** https://arxiv.org/pdf/2603.03977v1
+
+## 7865. Phi-4-reasoning-vision-15B Technical Report
+
+- **ID:** `2603.03975` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Jyoti Aneja, Michael Harrison, Neel Joshi, et al.
+- **abs:** https://arxiv.org/abs/2603.03975 · **PDF:** https://arxiv.org/pdf/2603.03975v1
+
+## 7866. Dual-Solver: A Generalized ODE Solver for Diffusion Models with Dual Prediction
+
+- **ID:** `2603.03973` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Soochul Park, Yeon Ju Lee
+- **abs:** https://arxiv.org/abs/2603.03973 · **PDF:** https://arxiv.org/pdf/2603.03973v1
+
+## 7867. Upholding Epistemic Agency: A Brouwerian Assertibility Constraint for Responsible AI
+
+- **ID:** `2603.03971` · **primary:** `cs.CY` · **Submitted:** 2026-03-04
+- **Authors:** Michael Jülich
+- **abs:** https://arxiv.org/abs/2603.03971 · **PDF:** https://arxiv.org/pdf/2603.03971v1
+
+## 7868. Generative AI in Managerial Decision-Making: Redefining Boundaries through Ambiguity Resolution and Sycophancy Analysis
+
+- **ID:** `2603.03970` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Sule Ozturk Birim, Fabrizio Marozzo, Yigit Kazancoglu
+- **abs:** https://arxiv.org/abs/2603.03970 · **PDF:** https://arxiv.org/pdf/2603.03970v1
+
+## 7869. BLOCK: An Open-Source Bi-Stage MLLM Character-to-Skin Pipeline for Minecraft
+
+- **ID:** `2603.03964` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Hengquan Guo
+- **abs:** https://arxiv.org/abs/2603.03964 · **PDF:** https://arxiv.org/pdf/2603.03964v2
+
+## 7870. Attention Meets Reachability: Structural Equivalence and Efficiency in Grammar-Constrained LLM Decoding
+
+- **ID:** `2603.05540` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Faruk Alpay, Bilge Senturk
+- **abs:** https://arxiv.org/abs/2603.05540 · **PDF:** https://arxiv.org/pdf/2603.05540v1
+
+## 7871. TFWaveFormer: Temporal-Frequency Collaborative Multi-level Wavelet Transformer for Dynamic Link Prediction
+
+- **ID:** `2603.03963` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Hantong Feng, Yonggang Wu, Duxin Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.03963 · **PDF:** https://arxiv.org/pdf/2603.03963v1
+
+## 7872. From Spark to Fire: Modeling and Mitigating Error Cascades in LLM-Based Multi-Agent Collaboration
+
+- **ID:** `2603.04474` · **primary:** `cs.MA` · **Submitted:** 2026-03-04
+- **Authors:** Yizhe Xie, Congcong Zhu, Xinyue Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.04474 · **PDF:** https://arxiv.org/pdf/2603.04474v1
+
+## 7873. LightningRL: Breaking the Accuracy-Parallelism Trade-off of Block-wise dLLMs via Reinforcement Learning
+
+- **ID:** `2603.13319` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Yanzhe Hu, Yijie Jin, Pengfei Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.13319 · **PDF:** https://arxiv.org/pdf/2603.13319v1
+
+## 7874. LoRA-MME: Multi-Model Ensemble of LoRA-Tuned Encoders for Code Comment Classification
+
+- **ID:** `2603.03959` · **primary:** `cs.SE` · **Submitted:** 2026-03-04
+- **Authors:** Md Akib Haider, Ahsan Bulbul, Nafis Fuad Shahid, et al.
+- **abs:** https://arxiv.org/abs/2603.03959 · **PDF:** https://arxiv.org/pdf/2603.03959v2
+
+## 7875. Towards Generalized Multimodal Homography Estimation
+
+- **ID:** `2603.03956` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Jinkun You, Jiaxin Cheng, Jie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03956 · **PDF:** https://arxiv.org/pdf/2603.03956v1
+
+## 7876. GIPO: Gaussian Importance Sampling Policy Optimization
+
+- **ID:** `2603.03955` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Chengxuan Lu, Zhenquan Zhang, Shukuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.03955 · **PDF:** https://arxiv.org/pdf/2603.03955v1
+
+## 7877. RVN-Bench: A Benchmark for Reactive Visual Navigation
+
+- **ID:** `2603.03953` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Jaewon Lee, Jaeseok Heo, Gunmin Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.03953 · **PDF:** https://arxiv.org/pdf/2603.03953v1
+
+## 7878. Lang2Str: Two-Stage Crystal Structure Generation with LLMs and Continuous Flow Models
+
+- **ID:** `2603.03946` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Cong Liu, Chengyue Gong, Zhenyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.03946 · **PDF:** https://arxiv.org/pdf/2603.03946v1
+
+## 7879. Residual Stream Analysis of Overfitting And Structural Disruptions
+
+- **ID:** `2603.13318` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Quan Liu, Han Zhou, Wenquan Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.13318 · **PDF:** https://arxiv.org/pdf/2603.13318v1
+
+## 7880. How Predicted Links Influence Network Evolution: Disentangling Choice and Algorithmic Feedback in Dynamic Graphs
+
+- **ID:** `2603.03945` · **primary:** `cs.SI` · **Submitted:** 2026-03-04
+- **Authors:** Mathilde Perez, Raphaël Romero, Jefrey Lijffijt, et al.
+- **abs:** https://arxiv.org/abs/2603.03945 · **PDF:** https://arxiv.org/pdf/2603.03945v1
+
+## 7881. Cross-Modal Mapping and Dual-Branch Reconstruction for 2D-3D Multimodal Industrial Anomaly Detection
+
+- **ID:** `2603.03939` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Radia Daci, Vito Renò, Cosimo Patruno, et al.
+- **abs:** https://arxiv.org/abs/2603.03939 · **PDF:** https://arxiv.org/pdf/2603.03939v1
+
+## 7882. Linguistic Signatures for Enhanced Emotion Detection
+
+- **ID:** `2603.20222` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Florian Lecourt, Madalina Croitoru, Konstantin Todorov
+- **abs:** https://arxiv.org/abs/2603.20222 · **PDF:** https://arxiv.org/pdf/2603.20222v1
+
+## 7883. Social physics in the age of artificial intelligence
+
+- **ID:** `2603.16900` · **primary:** `physics.soc-ph` · **Submitted:** 2026-03-04
+- **Authors:** The Anh Han, Joel Z. Leibo, Tom Lenaerts, et al.
+- **abs:** https://arxiv.org/abs/2603.16900 · **PDF:** https://arxiv.org/pdf/2603.16900v1
+
+## 7884. Selecting Offline Reinforcement Learning Algorithms for Stochastic Network Control
+
+- **ID:** `2603.03932` · **primary:** `cs.NI` · **Submitted:** 2026-03-04
+- **Authors:** Nicolas Helson, Pegah Alizadeh, Anastasios Giovanidis
+- **abs:** https://arxiv.org/abs/2603.03932 · **PDF:** https://arxiv.org/pdf/2603.03932v1
+
+## 7885. VDCook:DIY video data cook your MLLMs
+
+- **ID:** `2603.05539` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Chengwei Wu
+- **abs:** https://arxiv.org/abs/2603.05539 · **PDF:** https://arxiv.org/pdf/2603.05539v1
+
+## 7886. Hierarchical Inference and Closure Learning via Adaptive Surrogates for ODEs and PDEs
+
+- **ID:** `2603.03922` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Pengyu Zhang, Arnaud Vadeboncoeur, Alex Glyn-Davies, et al.
+- **abs:** https://arxiv.org/abs/2603.03922 · **PDF:** https://arxiv.org/pdf/2603.03922v1
+
+## 7887. BD-Merging: Bias-Aware Dynamic Model Merging with Evidence-Guided Contrastive Learning
+
+- **ID:** `2603.03920` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Yuhan Xie, Chen Lyu
+- **abs:** https://arxiv.org/abs/2603.03920 · **PDF:** https://arxiv.org/pdf/2603.03920v2
+
+## 7888. Rethinking Role-Playing Evaluation: Anonymous Benchmarking and a Systematic Study of Personality Effects
+
+- **ID:** `2603.03915` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Ji-Lun Peng, Yun-Nung Chen
+- **abs:** https://arxiv.org/abs/2603.03915 · **PDF:** https://arxiv.org/pdf/2603.03915v1
+
+## 7889. Evaluating Large Language Models for Gait Classification Using Text-Encoded Kinematic Waveforms
+
+- **ID:** `2603.13317` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Carlo Dindorf, Jonas Dully, Rebecca Keilhauer, et al.
+- **abs:** https://arxiv.org/abs/2603.13317 · **PDF:** https://arxiv.org/pdf/2603.13317v1
+
+## 7890. High-Resolution Image Reconstruction with Unsupervised Learning and Noisy Data Applied to Ion-Beam Dynamics for Particle Accelerators
+
+- **ID:** `2603.06689` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Francis Osswald, Mohammed Chahbaoui, Xinyi Liang
+- **abs:** https://arxiv.org/abs/2603.06689 · **PDF:** https://arxiv.org/pdf/2603.06689v1
+
+## 7891. From Threat Intelligence to Firewall Rules: Semantic Relations in Hybrid AI Agent and Expert System Architectures
+
+- **ID:** `2603.03911` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Chiara Bonfanti, Davide Colaiacomo, Luca Cagliero, et al.
+- **abs:** https://arxiv.org/abs/2603.03911 · **PDF:** https://arxiv.org/pdf/2603.03911v1
+
+## 7892. From Misclassifications to Outliers: Joint Reliability Assessment in Classification
+
+- **ID:** `2603.03903` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Yang Li, Youyang Sha, Yinzhi Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.03903 · **PDF:** https://arxiv.org/pdf/2603.03903v1
+
+## 7893. PatchDecomp: Interpretable Patch-Based Time Series Forecasting
+
+- **ID:** `2603.03902` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Hiroki Tomioka, Genta Yoshimura
+- **abs:** https://arxiv.org/abs/2603.03902 · **PDF:** https://arxiv.org/pdf/2603.03902v1
+
+## 7894. IROSA: Interactive Robot Skill Adaptation using Natural Language
+
+- **ID:** `2603.03897` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Markus Knauer, Samuel Bustamante, Thomas Eiband, et al.
+- **abs:** https://arxiv.org/abs/2603.03897 · **PDF:** https://arxiv.org/pdf/2603.03897v2
+
+## 7895. Dictionary Based Pattern Entropy for Causal Direction Discovery
+
+- **ID:** `2603.04473` · **primary:** `stat.ML` · **Submitted:** 2026-03-04
+- **Authors:** Harikrishnan N B, Shubham Bhilare, Aditi Kathpalia, et al.
+- **abs:** https://arxiv.org/abs/2603.04473 · **PDF:** https://arxiv.org/pdf/2603.04473v1
+
+## 7896. A novel network for classification of cuneiform tablet metadata
+
+- **ID:** `2603.03892` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Frederik Hagelskjær
+- **abs:** https://arxiv.org/abs/2603.03892 · **PDF:** https://arxiv.org/pdf/2603.03892v1
+
+## 7897. A debate game about societal impacts of Artificial Intelligence
+
+- **ID:** `2603.13316` · **primary:** `cs.CY` · **Submitted:** 2026-03-04
+- **Authors:** Carole Adam, Cedric Lauradoux
+- **abs:** https://arxiv.org/abs/2603.13316 · **PDF:** https://arxiv.org/pdf/2603.13316v1
+
+## 7898. CzechTopic: A Benchmark for Zero-Shot Topic Localization in Historical Czech Documents
+
+- **ID:** `2603.03884` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Martin Kostelník, Michal Hradiš, Martin Dočekal
+- **abs:** https://arxiv.org/abs/2603.03884 · **PDF:** https://arxiv.org/pdf/2603.03884v1
+
+## 7899. On the Suitability of LLM-Driven Agents for Dark Pattern Audits
+
+- **ID:** `2603.03881` · **primary:** `cs.CR` · **Submitted:** 2026-03-04
+- **Authors:** Chen Sun, Yash Vekaria, Rishab Nithyanand
+- **abs:** https://arxiv.org/abs/2603.03881 · **PDF:** https://arxiv.org/pdf/2603.03881v1
+
+## 7900. Joint Hardware-Workload Co-Optimization for In-Memory Computing Accelerators
+
+- **ID:** `2603.03880` · **primary:** `cs.AR` · **Submitted:** 2026-03-04
+- **Authors:** Olga Krestinskaya, Mohammed E. Fouda, Ahmed Eltawil, et al.
+- **abs:** https://arxiv.org/abs/2603.03880 · **PDF:** https://arxiv.org/pdf/2603.03880v1
+
+## 7901. Believe Your Model: Distribution-Guided Confidence Calibration
+
+- **ID:** `2603.03872` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Xizhong Yang, Haotian Zhang, Huiming Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.03872 · **PDF:** https://arxiv.org/pdf/2603.03872v1
+
+## 7902. k-hop Fairness: Addressing Disparities in Graph Link Prediction Beyond First-Order Neighborhoods
+
+- **ID:** `2603.03867` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Lilian Marey, Tiphaine Viard, Charlotte Laclau
+- **abs:** https://arxiv.org/abs/2603.03867 · **PDF:** https://arxiv.org/pdf/2603.03867v1
+
+## 7903. Structure-Aware Distributed Backdoor Attacks in Federated Learning
+
+- **ID:** `2603.03865` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Wang Jian, Shen Hong, Ke Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.03865 · **PDF:** https://arxiv.org/pdf/2603.03865v1
+
+## 7904. Assessing the Effectiveness of LLMs in Delivering Cognitive Behavioral Therapy
+
+- **ID:** `2603.03862` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Navdeep Singh Bedi, Ana-Maria Bucur, Noriko Kando, et al.
+- **abs:** https://arxiv.org/abs/2603.03862 · **PDF:** https://arxiv.org/pdf/2603.03862v1
+
+## 7905. Coupling Local Context and Global Semantic Prototypes via a Hierarchical Architecture for Rhetorical Roles Labeling
+
+- **ID:** `2603.03856` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Anas Belfathi, Nicolas Hernandez, Laura Monceaux, et al.
+- **abs:** https://arxiv.org/abs/2603.03856 · **PDF:** https://arxiv.org/pdf/2603.03856v1
+
+## 7906. Linear Predictability of Attention Heads in Large Language Models
+
+- **ID:** `2603.13314` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Khalid Shaikh, Asmit Kumar Singh, Rebecca Christopher Dsouza, et al.
+- **abs:** https://arxiv.org/abs/2603.13314 · **PDF:** https://arxiv.org/pdf/2603.13314v1
+
+## 7907. Evaluating Large Language Models' Responses to Sexual and Reproductive Health Queries in Nepali
+
+- **ID:** `2603.22291` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Medha Sharma, Supriya Khadka, Udit Chandra Aryal, et al.
+- **abs:** https://arxiv.org/abs/2603.22291 · **PDF:** https://arxiv.org/pdf/2603.22291v1
+
+## 7908. Benchmarking Motivational Interviewing Competence of Large Language Models
+
+- **ID:** `2603.03846` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Aishwariya Jha, Prakrithi Shivaprakash, Lekhansh Shukla, et al.
+- **abs:** https://arxiv.org/abs/2603.03846 · **PDF:** https://arxiv.org/pdf/2603.03846v1
+
+## 7909. Semantic Bridging Domains: Pseudo-Source as Test-Time Connector
+
+- **ID:** `2603.03844` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Xizhong Yang, Huiming Wang, Ning Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.03844 · **PDF:** https://arxiv.org/pdf/2603.03844v1
+
+## 7910. Invariance-Based Dynamic Regret Minimization
+
+- **ID:** `2603.03843` · **primary:** `stat.ML` · **Submitted:** 2026-03-04
+- **Authors:** Margherita Lazzaretto, Jonas Peters, Niklas Pfister
+- **abs:** https://arxiv.org/abs/2603.03843 · **PDF:** https://arxiv.org/pdf/2603.03843v1
+
+## 7911. Non-Invasive Reconstruction of Cardiac Activation Dynamics Using Physics-Informed Neural Networks
+
+- **ID:** `2603.03832` · **primary:** `physics.med-ph` · **Submitted:** 2026-03-04
+- **Authors:** Nathan Dermul, Hans Dierckx
+- **abs:** https://arxiv.org/abs/2603.03832 · **PDF:** https://arxiv.org/pdf/2603.03832v1
+
+## 7912. Large-Margin Hyperdimensional Computing: A Learning-Theoretical Perspective
+
+- **ID:** `2603.03830` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Nikita Zeulin, Olga Galinina, Ravikumar Balakrishnan, et al.
+- **abs:** https://arxiv.org/abs/2603.03830 · **PDF:** https://arxiv.org/pdf/2603.03830v1
+
+## 7913. From Narrow to Panoramic Vision: Attention-Guided Cold-Start Reshapes Multimodal Reasoning
+
+- **ID:** `2603.03825` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Ruilin Luo, Chufan Shi, Yizhen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03825 · **PDF:** https://arxiv.org/pdf/2603.03825v1
+
+## 7914. Narrative Weaver: Towards Controllable Long-Range Visual Consistency with Multi-Modal Conditioning
+
+- **ID:** `2603.06688` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Zhengjian Yao, Yongzhi Li, Xinyuan Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.06688 · **PDF:** https://arxiv.org/pdf/2603.06688v2
+
+## 7915. In-Context Environments Induce Evaluation-Awareness in Language Models
+
+- **ID:** `2603.03824` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Maheep Chaudhary
+- **abs:** https://arxiv.org/abs/2603.03824 · **PDF:** https://arxiv.org/pdf/2603.03824v1
+
+## 7916. SWE-CI: Evaluating Agent Capabilities in Maintaining Codebases via Continuous Integration
+
+- **ID:** `2603.03823` · **primary:** `cs.SE` · **Submitted:** 2026-03-04
+- **Authors:** Jialong Chen, Xander Xu, Hu Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.03823 · **PDF:** https://arxiv.org/pdf/2603.03823v4
+
+## 7917. Fairness Begins with State: Purifying Latent Preferences for Hierarchical Reinforcement Learning in Interactive Recommendation
+
+- **ID:** `2603.03820` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Yun Lu, Xiaoyu Shi, Hong Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.03820 · **PDF:** https://arxiv.org/pdf/2603.03820v1
+
+## 7918. Direct Bayesian Additive Regression Trees for Conditional Average Treatment Effects in Regression Discontinuity Designs
+
+- **ID:** `2603.03819` · **primary:** `stat.ME` · **Submitted:** 2026-03-04
+- **Authors:** Daisuke Kondo, Shonosuke Sugasawa
+- **abs:** https://arxiv.org/abs/2603.03819 · **PDF:** https://arxiv.org/pdf/2603.03819v1
+
+## 7919. Pretrained Vision-Language-Action Models are Surprisingly Resistant to Forgetting in Continual Learning
+
+- **ID:** `2603.03818` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Huihan Liu, Changyeon Kim, Bo Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.03818 · **PDF:** https://arxiv.org/pdf/2603.03818v2
+
+## 7920. Steering Frozen LLMs: Adaptive Social Alignment via Online Prompt Routing
+
+- **ID:** `2603.15647` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Zeyu Zhang, Xiangxiang Dai, Ziyi Han, et al.
+- **abs:** https://arxiv.org/abs/2603.15647 · **PDF:** https://arxiv.org/pdf/2603.15647v1
+
+## 7921. A Bi-Stage Framework for Automatic Development of Pixel-Based Planar Antenna Structures
+
+- **ID:** `2603.03810` · **primary:** `math.NA` · **Submitted:** 2026-03-04
+- **Authors:** Khadijeh Askaripour, Adrian Bekasiewicz, Slawomir Koziel
+- **abs:** https://arxiv.org/abs/2603.03810 · **PDF:** https://arxiv.org/pdf/2603.03810v1
+
+## 7922. Separators in Enhancing Autoregressive Pretraining for Vision Mamba
+
+- **ID:** `2603.03806` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Hanpeng Liu, Zidan Wang, Shuoxi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03806 · **PDF:** https://arxiv.org/pdf/2603.03806v1
+
+## 7923. Relational In-Context Learning via Synthetic Pre-training with Structural Prior
+
+- **ID:** `2603.03805` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Yanbo Wang, Jiaxuan You, Chuan Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.03805 · **PDF:** https://arxiv.org/pdf/2603.03805v1
+
+## 7924. Zero-Knowledge Proof (ZKP) Authentication for Offline CBDC Payment System Using IoT Devices
+
+- **ID:** `2603.03804` · **primary:** `cs.CR` · **Submitted:** 2026-03-04
+- **Authors:** Santanu Mondal, T. Chithralekha
+- **abs:** https://arxiv.org/abs/2603.03804 · **PDF:** https://arxiv.org/pdf/2603.03804v2
+
+## 7925. TimeSpot: Benchmarking Geo-Temporal Understanding in Vision-Language Models in Real-World Settings
+
+- **ID:** `2603.06687` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Azmine Toushik Wasi, Shahriyar Zaman Ridoy, Koushik Ahamed Tonmoy, et al.
+- **abs:** https://arxiv.org/abs/2603.06687 · **PDF:** https://arxiv.org/pdf/2603.06687v1
+
+## 7926. Unsupervised Surrogate-Assisted Synthesis of Free-Form Planar Antenna Topologies for IoT Applications
+
+- **ID:** `2603.03802` · **primary:** `math.NA` · **Submitted:** 2026-03-04
+- **Authors:** Khadijeh Askaripour, Adrian Bekasiewicz, Slawomir Koziel
+- **abs:** https://arxiv.org/abs/2603.03802 · **PDF:** https://arxiv.org/pdf/2603.03802v1
+
+## 7927. A Rubric-Supervised Critic from Sparse Real-World Outcomes
+
+- **ID:** `2603.03800` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Xingyao Wang, Valerie Chen, Heng Ji, et al.
+- **abs:** https://arxiv.org/abs/2603.03800 · **PDF:** https://arxiv.org/pdf/2603.03800v1
+
+## 7928. Less is More: Adapting Text Embeddings for Low-Resource Languages with Small Scale Noisy Synthetic Data
+
+- **ID:** `2603.22290` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Zaruhi Navasardyan, Spartak Bughdaryan, Bagrat Minasyan, et al.
+- **abs:** https://arxiv.org/abs/2603.22290 · **PDF:** https://arxiv.org/pdf/2603.22290v1
+
+## 7929. When and Where to Reset Matters for Long-Term Test-Time Adaptation
+
+- **ID:** `2603.03796` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Taejun Lim, Joong-Won Hwang, Kibok Lee
+- **abs:** https://arxiv.org/abs/2603.03796 · **PDF:** https://arxiv.org/pdf/2603.03796v1
+
+## 7930. TAP: A Token-Adaptive Predictor Framework for Training-Free Diffusion Acceleration
+
+- **ID:** `2603.03792` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Haowei Zhu, Tingxuan Huang, Xing Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.03792 · **PDF:** https://arxiv.org/pdf/2603.03792v1
+
+## 7931. T2S-Bench & Structure-of-Thought: Benchmarking and Prompting Comprehensive Text-to-Structure Reasoning
+
+- **ID:** `2603.03790` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Qinsi Wang, Hancheng Ye, Jinhee Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.03790 · **PDF:** https://arxiv.org/pdf/2603.03790v1
+
+## 7932. Towards Explainable Deep Learning for Ship Trajectory Prediction in Inland Waterways
+
+- **ID:** `2603.04472` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Tom Legel, Dirk Söffker, Roland Schätzle, et al.
+- **abs:** https://arxiv.org/abs/2603.04472 · **PDF:** https://arxiv.org/pdf/2603.04472v1
+
+## 7933. Observationally Informed Adaptive Causal Experimental Design
+
+- **ID:** `2603.03785` · **primary:** `stat.ML` · **Submitted:** 2026-03-04
+- **Authors:** Erdun Gao, Liang Zhang, Jake Fawkes, et al.
+- **abs:** https://arxiv.org/abs/2603.03785 · **PDF:** https://arxiv.org/pdf/2603.03785v1
+
+## 7934. Design-MLLM: A Reinforcement Alignment Framework for Verifiable and Aesthetic Interior Design
+
+- **ID:** `2603.13312` · **primary:** `cs.MM` · **Submitted:** 2026-03-04
+- **Authors:** Yuxuan Yang, Xiaotong Mao, Jingyao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.13312 · **PDF:** https://arxiv.org/pdf/2603.13312v1
+
+## 7935. Specification-Driven Generation and Evaluation of Discrete-Event World Models via the DEVS Formalism
+
+- **ID:** `2603.03784` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Zheyu Chen, Zhuohuan Li, Chuanhao Li
+- **abs:** https://arxiv.org/abs/2603.03784 · **PDF:** https://arxiv.org/pdf/2603.03784v1
+
+## 7936. DisenReason: Behavior Disentanglement and Latent Reasoning for Shared-Account Sequential Recommendation
+
+- **ID:** `2603.03782` · **primary:** `cs.IR` · **Submitted:** 2026-03-04
+- **Authors:** Jiawei Cheng, Min Gao, Zongwei Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.03782 · **PDF:** https://arxiv.org/pdf/2603.03782v1
+
+## 7937. LifeBench: A Benchmark for Long-Horizon Multi-Source Memory
+
+- **ID:** `2603.03781` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Zihao Cheng, Weixin Wang, Yu Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.03781 · **PDF:** https://arxiv.org/pdf/2603.03781v1
+
+## 7938. MACC: Multi-Agent Collaborative Competition for Scientific Exploration
+
+- **ID:** `2603.03780` · **primary:** `cs.MA` · **Submitted:** 2026-03-04
+- **Authors:** Satoshi Oyama, Yuko Sakurai, Hisashi Kashima
+- **abs:** https://arxiv.org/abs/2603.03780 · **PDF:** https://arxiv.org/pdf/2603.03780v1
+
+## 7939. Inverse Contextual Bandits without Rewards: Learning from a Non-Stationary Learner via Suffix Imitation
+
+- **ID:** `2603.03778` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Yuqi Kong, Xiao Zhang, Weiran Shen
+- **abs:** https://arxiv.org/abs/2603.03778 · **PDF:** https://arxiv.org/pdf/2603.03778v1
+
+## 7940. LEA: Label Enumeration Attack in Vertical Federated Learning
+
+- **ID:** `2603.03777` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Wenhao Jiang, Shaojing Fu, Yuchuan Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.03777 · **PDF:** https://arxiv.org/pdf/2603.03777v1
+
+## 7941. Towards Effective Orchestration of AI x DB Workloads
+
+- **ID:** `2603.03772` · **primary:** `cs.DB` · **Submitted:** 2026-03-04
+- **Authors:** Naili Xing, Haotian Gao, Zhanhao Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.03772 · **PDF:** https://arxiv.org/pdf/2603.03772v1
+
+## 7942. Not All Candidates are Created Equal: A Heterogeneity-Aware Approach to Pre-ranking in Recommender Systems
+
+- **ID:** `2603.03770` · **primary:** `cs.IR` · **Submitted:** 2026-03-04
+- **Authors:** Pengfei Tong, Siyuan Chen, Chenwei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03770 · **PDF:** https://arxiv.org/pdf/2603.03770v1
+
+## 7943. DMD-augmented Unpaired Neural Schrödinger Bridge for Ultra-Low Field MRI Enhancement
+
+- **ID:** `2603.03769` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Youngmin Kim, Jaeyun Shin, Jeongchan Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.03769 · **PDF:** https://arxiv.org/pdf/2603.03769v2
+
+## 7944. Cognition to Control - Multi-Agent Learning for Human-Humanoid Collaborative Transport
+
+- **ID:** `2603.03768` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Hao Zhang, Ding Zhao, H. Eric Tseng
+- **abs:** https://arxiv.org/abs/2603.03768 · **PDF:** https://arxiv.org/pdf/2603.03768v1
+
+## 7945. AgentSelect: Benchmark for Narrative Query-to-Agent Recommendation
+
+- **ID:** `2603.03761` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Yunxiao Shi, Wujiang Xu, Tingwei Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.03761 · **PDF:** https://arxiv.org/pdf/2603.03761v1
+
+## 7946. Harmonic Dataset Distillation for Time Series Forecasting
+
+- **ID:** `2603.03760` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Seungha Hong, Sanghwan Jang, Wonbin Kweon, et al.
+- **abs:** https://arxiv.org/abs/2603.03760 · **PDF:** https://arxiv.org/pdf/2603.03760v1
+
+## 7947. JAWS: Enhancing Long-term Rollout of Neural PDE Solvers via Spatially-Adaptive Jacobian Regularization
+
+- **ID:** `2603.05538` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Fengxiang Nie, Yasuhiro Suzuki
+- **abs:** https://arxiv.org/abs/2603.05538 · **PDF:** https://arxiv.org/pdf/2603.05538v2
+
+## 7948. Learning Approximate Nash Equilibria in Cooperative Multi-Agent Reinforcement Learning via Mean-Field Subsampling
+
+- **ID:** `2603.03759` · **primary:** `cs.MA` · **Submitted:** 2026-03-04
+- **Authors:** Emile Anand, Ishani Karmarkar
+- **abs:** https://arxiv.org/abs/2603.03759 · **PDF:** https://arxiv.org/pdf/2603.03759v1
+
+## 7949. MOOSE-Star: Unlocking Tractable Training for Scientific Discovery by Breaking the Complexity Barrier
+
+- **ID:** `2603.03756` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Zonglin Yang, Lidong Bing
+- **abs:** https://arxiv.org/abs/2603.03756 · **PDF:** https://arxiv.org/pdf/2603.03756v1
+
+## 7950. Neural Approximation and Its Applications
+
+- **ID:** `2603.13311` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Wei-Hao Wu, Ting-Zhu Huang, Xi-Le Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.13311 · **PDF:** https://arxiv.org/pdf/2603.13311v1
+
+## 7951. Agentic Peer-to-Peer Networks: From Content Distribution to Capability and Action Sharing
+
+- **ID:** `2603.03753` · **primary:** `cs.NI` · **Submitted:** 2026-03-04
+- **Authors:** Taotao Wang, Lizhao You, Jingwen Tong, et al.
+- **abs:** https://arxiv.org/abs/2603.03753 · **PDF:** https://arxiv.org/pdf/2603.03753v1
+
+## 7952. Confidence-Calibrated Small-Large Language Model Collaboration for Cost-Efficient Reasoning
+
+- **ID:** `2603.03752` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Chuang Zhang, Zizhen Zhu, Yihao Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.03752 · **PDF:** https://arxiv.org/pdf/2603.03752v1
+
+## 7953. Interaction-Aware Whole-Body Control for Compliant Object Transport
+
+- **ID:** `2603.03751` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Hao Zhang, Yves Tseng, Ding Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.03751 · **PDF:** https://arxiv.org/pdf/2603.03751v1
+
+## 7954. One step further with Monte-Carlo sampler to guide diffusion better
+
+- **ID:** `2603.06685` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Minsi Ren, Wenhao Deng, Ruiqi Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.06685 · **PDF:** https://arxiv.org/pdf/2603.06685v1
+
+## 7955. JANUS: Structured Bidirectional Generation for Guaranteed Constraints and Analytical Uncertainty
+
+- **ID:** `2603.03748` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Taha Racicot
+- **abs:** https://arxiv.org/abs/2603.03748 · **PDF:** https://arxiv.org/pdf/2603.03748v1
+
+## 7956. RAGNav: A Retrieval-Augmented Topological Reasoning Framework for Multi-Goal Visual-Language Navigation
+
+- **ID:** `2603.03745` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Ling Luo, Qiangian Bai
+- **abs:** https://arxiv.org/abs/2603.03745 · **PDF:** https://arxiv.org/pdf/2603.03745v1
+
+## 7957. ErrorLLM: Modeling SQL Errors for Text-to-SQL Refinement
+
+- **ID:** `2603.03742` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Zijin Hong, Hao Chen, Zheng Yuan, et al.
+- **abs:** https://arxiv.org/abs/2603.03742 · **PDF:** https://arxiv.org/pdf/2603.03742v1
+
+## 7958. HALyPO: Heterogeneous-Agent Lyapunov Policy Optimization for Human-Robot Collaboration
+
+- **ID:** `2603.03741` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Hao Zhang, Yaru Niu, Yikai Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.03741 · **PDF:** https://arxiv.org/pdf/2603.03741v1
+
+## 7959. PROSPECT: Unified Streaming Vision-Language Navigation via Semantic--Spatial Fusion and Latent Predictive Representation
+
+- **ID:** `2603.03739` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Zehua Fan, Wenqi Lyu, Wenxuan Song, et al.
+- **abs:** https://arxiv.org/abs/2603.03739 · **PDF:** https://arxiv.org/pdf/2603.03739v1
+
+## 7960. Understanding Parents' Desires in Moderating Children's Interactions with GenAI Chatbots through LLM-Generated Probes
+
+- **ID:** `2603.03727` · **primary:** `cs.HC` · **Submitted:** 2026-03-04
+- **Authors:** John Driscoll, Yulin Chen, Viki Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.03727 · **PDF:** https://arxiv.org/pdf/2603.03727v2
+
+## 7961. Why Do Unlearnable Examples Work: A Novel Perspective of Mutual Information
+
+- **ID:** `2603.03725` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Yifan Zhu, Yibo Miao, Yinpeng Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.03725 · **PDF:** https://arxiv.org/pdf/2603.03725v1
+
+## 7962. Multi-view Attention Fusion of Heterogeneous Hypergraph with Dynamic Behavioral Profiling for Personalized Learning Resource Recommendation
+
+- **ID:** `2603.13310` · **primary:** `cs.IR` · **Submitted:** 2026-03-04
+- **Authors:** Tao Xie, Yan Li, Yongpan Sheng, et al.
+- **abs:** https://arxiv.org/abs/2603.13310 · **PDF:** https://arxiv.org/pdf/2603.13310v1
+
+## 7963. Order Is Not Layout: Order-to-Space Bias in Image Generation
+
+- **ID:** `2603.03714` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Yongkang Zhang, Zonglin Zhao, Yuechen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03714 · **PDF:** https://arxiv.org/pdf/2603.03714v1
+
+## 7964. MPFlow: Multi-modal Posterior-Guided Flow Matching for Zero-Shot MRI Reconstruction
+
+- **ID:** `2603.03710` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Seunghoi Kim, Chen Jin, Henry F. J. Tregidgo, et al.
+- **abs:** https://arxiv.org/abs/2603.03710 · **PDF:** https://arxiv.org/pdf/2603.03710v2
+
+## 7965. Alternating Reinforcement Learning with Contextual Rubric Rewards
+
+- **ID:** `2603.15646` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Guangchen Lan
+- **abs:** https://arxiv.org/abs/2603.15646 · **PDF:** https://arxiv.org/pdf/2603.15646v1
+
+## 7966. Large-Language-Model-Guided State Estimation for Partially Observable Task and Motion Planning
+
+- **ID:** `2603.03704` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Yoonwoo Kim, Raghav Arora, Roberto Martín-Martín, et al.
+- **abs:** https://arxiv.org/abs/2603.03704 · **PDF:** https://arxiv.org/pdf/2603.03704v2
+
+## 7967. UrbanHuRo: A Two-Layer Human-Robot Collaboration Framework for the Joint Optimization of Heterogeneous Urban Services
+
+- **ID:** `2603.03701` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Tonmoy Dey, Lin Jiang, Zheng Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.03701 · **PDF:** https://arxiv.org/pdf/2603.03701v1
+
+## 7968. Generalization Properties of Score-matching Diffusion Models for Intrinsically Low-dimensional Data
+
+- **ID:** `2603.03700` · **primary:** `stat.ML` · **Submitted:** 2026-03-04
+- **Authors:** Saptarshi Chakraborty, Quentin Berthet, Peter L. Bartlett
+- **abs:** https://arxiv.org/abs/2603.03700 · **PDF:** https://arxiv.org/pdf/2603.03700v1
+
+## 7969. Clinical-Injection Transformer with Domain-Adapted MAE for Lupus Nephritis Prognosis Prediction
+
+- **ID:** `2603.05535` · **primary:** `eess.IV` · **Submitted:** 2026-03-04
+- **Authors:** Yuewen Huang, Zhitao Ye, Guangnan Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.05535 · **PDF:** https://arxiv.org/pdf/2603.05535v1
+
+## 7970. Error as Signal: Stiffness-Aware Diffusion Sampling via Embedded Runge-Kutta Guidance
+
+- **ID:** `2603.03692` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Inho Kong, Sojin Lee, Youngjoon Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.03692 · **PDF:** https://arxiv.org/pdf/2603.03692v1
+
+## 7971. AI4S-SDS: A Neuro-Symbolic Solvent Design System via Sparse MCTS and Differentiable Physics Alignment
+
+- **ID:** `2603.03686` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Jiangyu Chen
+- **abs:** https://arxiv.org/abs/2603.03686 · **PDF:** https://arxiv.org/pdf/2603.03686v2
+
+## 7972. Mathematicians in the age of AI
+
+- **ID:** `2603.03684` · **primary:** `math.HO` · **Submitted:** 2026-03-04
+- **Authors:** Jeremy Avigad
+- **abs:** https://arxiv.org/abs/2603.03684 · **PDF:** https://arxiv.org/pdf/2603.03684v2
+
+## 7973. CONCUR: Benchmarking LLMs for Concurrent Code Generation
+
+- **ID:** `2603.03683` · **primary:** `cs.SE` · **Submitted:** 2026-03-04
+- **Authors:** Jue Huang, Tarek Mahmud, Corina Pasareanu, et al.
+- **abs:** https://arxiv.org/abs/2603.03683 · **PDF:** https://arxiv.org/pdf/2603.03683v1
+
+## 7974. EvoPrune: Early-Stage Visual Token Pruning for Efficient MLLMs
+
+- **ID:** `2603.03681` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Yuhao Chen, Bin Shan, Xin Ye, et al.
+- **abs:** https://arxiv.org/abs/2603.03681 · **PDF:** https://arxiv.org/pdf/2603.03681v1
+
+## 7975. MAGE: Meta-Reinforcement Learning for Language Agents toward Strategic Exploration and Exploitation
+
+- **ID:** `2603.03680` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** Lu Yang, Zelai Xu, Minyang Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.03680 · **PDF:** https://arxiv.org/pdf/2603.03680v1
+
+## 7976. XLinear: Frequency-Enhanced MLP with CrossFilter for Robust Long-Range Forecasting
+
+- **ID:** `2603.15645` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Xiang Ao
+- **abs:** https://arxiv.org/abs/2603.15645 · **PDF:** https://arxiv.org/pdf/2603.15645v1
+
+## 7977. MIND: Unified Inquiry and Diagnosis RL with Criteria Grounded Clinical Supports for Psychiatric Consultation
+
+- **ID:** `2603.03677` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Guoyi Li, Shihao Xu, Jiatong Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.03677 · **PDF:** https://arxiv.org/pdf/2603.03677v1
+
+## 7978. A Stein Identity for q-Gaussians with Bounded Support
+
+- **ID:** `2603.03673` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Sophia Sklaviadis, Thomas Moellenhoff, Andre F. T. Martins, et al.
+- **abs:** https://arxiv.org/abs/2603.03673 · **PDF:** https://arxiv.org/pdf/2603.03673v1
+
+## 7979. Local Shapley: Model-Induced Locality and Optimal Reuse in Data Valuation
+
+- **ID:** `2603.03672` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Xuan Yang, Hsi-Wen Chen, Ming-Syan Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.03672 · **PDF:** https://arxiv.org/pdf/2603.03672v1
+
+## 7980. Can LLM Aid in Solving Constraints with Inductive Definitions?
+
+- **ID:** `2603.03668` · **primary:** `cs.LO` · **Submitted:** 2026-03-04
+- **Authors:** Weizhi Feng, Shidong Shen, Jiaxiang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.03668 · **PDF:** https://arxiv.org/pdf/2603.03668v2
+
+## 7981. Machine Pareidolia: Protecting Facial Image with Emotional Editing
+
+- **ID:** `2603.03665` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Binh M. Le, Simon S. Woo
+- **abs:** https://arxiv.org/abs/2603.03665 · **PDF:** https://arxiv.org/pdf/2603.03665v1
+
+## 7982. Principled Learning-to-Communicate with Quasi-Classical Information Structures
+
+- **ID:** `2603.03664` · **primary:** `eess.SY` · **Submitted:** 2026-03-04
+- **Authors:** Xiangyu Liu, Haoyi You, Kaiqing Zhang
+- **abs:** https://arxiv.org/abs/2603.03664 · **PDF:** https://arxiv.org/pdf/2603.03664v1
+
+## 7983. Graph Negative Feedback Bias Correction Framework for Adaptive Heterophily Modeling
+
+- **ID:** `2603.03662` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Jiaqi Lv, Qingfeng Du, Yu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03662 · **PDF:** https://arxiv.org/pdf/2603.03662v1
+
+## 7984. InEdit-Bench: Benchmarking Intermediate Logical Pathways for Intelligent Image Editing Models
+
+- **ID:** `2603.03657` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Zhiqiang Sheng, Xumeng Han, Zhiwei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03657 · **PDF:** https://arxiv.org/pdf/2603.03657v1
+
+## 7985. Mozi: Governed Autonomy for Drug Discovery LLM Agents
+
+- **ID:** `2603.03655` · **primary:** `cs.AI` · **Submitted:** 2026-03-04
+- **Authors:** He Cao, Siyu Liu, Fan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03655 · **PDF:** https://arxiv.org/pdf/2603.03655v1
+
+## 7986. Field imaging framework for morphological characterization of aggregates with computer vision: Algorithms and applications
+
+- **ID:** `2603.03654` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Haohang Huang
+- **abs:** https://arxiv.org/abs/2603.03654 · **PDF:** https://arxiv.org/pdf/2603.03654v1
+
+## 7987. Linguistically Informed Graph Model and Semantic Contrastive Learning for Korean Short Text Classification
+
+- **ID:** `2603.03652` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** JaeGeon Yoo, Byoungwook Kim, Yeongwook Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.03652 · **PDF:** https://arxiv.org/pdf/2603.03652v1
+
+## 7988. Freezing of Gait Prediction using Proactive Agent that Learns from Selected Experience and DDQN Algorithm
+
+- **ID:** `2603.03651` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Septian Enggar Sukmana, Sang Won Bae, Tomohiro Shibata
+- **abs:** https://arxiv.org/abs/2603.03651 · **PDF:** https://arxiv.org/pdf/2603.03651v1
+
+## 7989. Adaptive Sensing of Continuous Physical Systems for Machine Learning
+
+- **ID:** `2603.03650` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Felix Köster, Atsushi Uchida
+- **abs:** https://arxiv.org/abs/2603.03650 · **PDF:** https://arxiv.org/pdf/2603.03650v1
+
+## 7990. Bridging Pedagogy and Play: Introducing a Language Mapping Interface for Human-AI Co-Creation in Educational Game Design
+
+- **ID:** `2603.03644` · **primary:** `cs.HC` · **Submitted:** 2026-03-04
+- **Authors:** Daijin Yang, Erica Kleinman, Casper Harteveld
+- **abs:** https://arxiv.org/abs/2603.03644 · **PDF:** https://arxiv.org/pdf/2603.03644v1
+
+## 7991. Image-based Prompt Injection: Hijacking Multimodal LLMs through Visually Embedded Adversarial Instructions
+
+- **ID:** `2603.03637` · **primary:** `cs.CV` · **Submitted:** 2026-03-04
+- **Authors:** Neha Nagaraja, Lan Zhang, Zhilong Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.03637 · **PDF:** https://arxiv.org/pdf/2603.03637v1
+
+## 7992. Goal-Driven Risk Assessment for LLM-Powered Systems: A Healthcare Case Study
+
+- **ID:** `2603.03633` · **primary:** `cs.CR` · **Submitted:** 2026-03-04
+- **Authors:** Neha Nagaraja, Hayretdin Bahsi
+- **abs:** https://arxiv.org/abs/2603.03633 · **PDF:** https://arxiv.org/pdf/2603.03633v1
+
+## 7993. Riemannian Langevin Dynamics: Strong Convergence of Geometric Euler-Maruyama Scheme
+
+- **ID:** `2603.03626` · **primary:** `stat.ML` · **Submitted:** 2026-03-04
+- **Authors:** Zhiyuan Zhan, Masashi Sugiyama
+- **abs:** https://arxiv.org/abs/2603.03626 · **PDF:** https://arxiv.org/pdf/2603.03626v1
+
+## 7994. A Neural Topic Method Using a Large-Language-Model-in-the-Loop for Business Research
+
+- **ID:** `2603.03623` · **primary:** `cs.CL` · **Submitted:** 2026-03-04
+- **Authors:** Stephan Ludwig, Peter J. Danaher, Xiaohao Yang
+- **abs:** https://arxiv.org/abs/2603.03623 · **PDF:** https://arxiv.org/pdf/2603.03623v1
+
+## 7995. Extending Neural Operators: Robust Handling of Functions Beyond the Training Set
+
+- **ID:** `2603.03621` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Blaine Quackenbush, Paul J. Atzberger
+- **abs:** https://arxiv.org/abs/2603.03621 · **PDF:** https://arxiv.org/pdf/2603.03621v1
+
+## 7996. Empirical Evaluation of No Free Lunch Violations in Permutation-Based Optimization
+
+- **ID:** `2603.03613` · **primary:** `stat.ML` · **Submitted:** 2026-03-04
+- **Authors:** Grzegorz Sroka
+- **abs:** https://arxiv.org/abs/2603.03613 · **PDF:** https://arxiv.org/pdf/2603.03613v1
+
+## 7997. Why Are Linear RNNs More Parallelizable?
+
+- **ID:** `2603.03612` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** William Merrill, Hongjian Jiang, Yanhong Li, et al.
+- **abs:** https://arxiv.org/abs/2603.03612 · **PDF:** https://arxiv.org/pdf/2603.03612v2
+
+## 7998. Riemannian Optimization in Modular Systems
+
+- **ID:** `2603.03610` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Christian Pehle, Jean-Jacques Slotine
+- **abs:** https://arxiv.org/abs/2603.03610 · **PDF:** https://arxiv.org/pdf/2603.03610v1
+
+## 7999. NuMuon: Nuclear-Norm-Constrained Muon for Compressible LLM Training
+
+- **ID:** `2603.03597` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Hadi Mohaghegh Dolatabadi, Thalaiyasingam Ajanthan, Sameera Ramasinghe, et al.
+- **abs:** https://arxiv.org/abs/2603.03597 · **PDF:** https://arxiv.org/pdf/2603.03597v1
+
+## 8000. MEM: Multi-Scale Embodied Memory for Vision Language Action Models
+
+- **ID:** `2603.03596` · **primary:** `cs.RO` · **Submitted:** 2026-03-04
+- **Authors:** Marcel Torne, Karl Pertsch, Homer Walke, et al.
+- **abs:** https://arxiv.org/abs/2603.03596 · **PDF:** https://arxiv.org/pdf/2603.03596v2
+
+## 8001. Hybrid Belief Reinforcement Learning for Efficient Coordinated Spatial Exploration
+
+- **ID:** `2603.03595` · **primary:** `cs.LG` · **Submitted:** 2026-03-04
+- **Authors:** Danish Rizvi, David Boyle
+- **abs:** https://arxiv.org/abs/2603.03595 · **PDF:** https://arxiv.org/pdf/2603.03595v1
+
+## 8002. SENTINEL: Stagewise Integrity Verification for Pipeline Parallel Decentralized Training
+
+- **ID:** `2603.03592` · **primary:** `cs.DC` · **Submitted:** 2026-03-03
+- **Authors:** Hadi Mohaghegh Dolatabadi, Thalaiyasingam Ajanthan, Sameera Ramasinghe, et al.
+- **abs:** https://arxiv.org/abs/2603.03592 · **PDF:** https://arxiv.org/pdf/2603.03592v1
+
+## 8003. Social Norm Reasoning in Multimodal Language Models: An Evaluation
+
+- **ID:** `2603.03590` · **primary:** `cs.MA` · **Submitted:** 2026-03-03
+- **Authors:** Oishik Chowdhury, Anushka Debnath, Bastin Tony Roy Savarimuthu
+- **abs:** https://arxiv.org/abs/2603.03590 · **PDF:** https://arxiv.org/pdf/2603.03590v1
+
+## 8004. stratum: A System Infrastructure for Massive Agent-Centric ML Workloads
+
+- **ID:** `2603.03589` · **primary:** `cs.DB` · **Submitted:** 2026-03-03
+- **Authors:** Arnab Phani, Elias Strauss, Sebastian Schelter
+- **abs:** https://arxiv.org/abs/2603.03589 · **PDF:** https://arxiv.org/pdf/2603.03589v2
+
+## 8005. Controllable Generative Sandbox for Causal Inference
+
+- **ID:** `2603.03587` · **primary:** `stat.ME` · **Submitted:** 2026-03-03
+- **Authors:** Qi Zhang, Harsh Parikh, Ashley Naimi, et al.
+- **abs:** https://arxiv.org/abs/2603.03587 · **PDF:** https://arxiv.org/pdf/2603.03587v1
+
+## 8006. Belief-Sim: Towards Belief-Driven Simulation of Demographic Misinformation Susceptibility
+
+- **ID:** `2603.03585` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Angana Borah, Zohaib Khan, Rada Mihalcea, et al.
+- **abs:** https://arxiv.org/abs/2603.03585 · **PDF:** https://arxiv.org/pdf/2603.03585v1
+
+## 8007. ByteFlow: Language Modeling through Adaptive Byte Compression without a Tokenizer
+
+- **ID:** `2603.03583` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Chunyuan Deng, Sanket Lokegaonkar, Colin Lockard, et al.
+- **abs:** https://arxiv.org/abs/2603.03583 · **PDF:** https://arxiv.org/pdf/2603.03583v1
+
+## 8008. Transport Clustering: Solving Low-Rank Optimal Transport via Clustering
+
+- **ID:** `2603.03578` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Henri Schmidt, Peter Halmos, Ben Raphael
+- **abs:** https://arxiv.org/abs/2603.03578 · **PDF:** https://arxiv.org/pdf/2603.03578v1
+
+## 8009. VB: Visibility Benchmark for Visibility and Perspective Reasoning in Images
+
+- **ID:** `2603.06680` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Neil Tripathi
+- **abs:** https://arxiv.org/abs/2603.06680 · **PDF:** https://arxiv.org/pdf/2603.06680v1
+
+## 8010. Tokenization Tradeoffs in Structured EHR Foundation Models
+
+- **ID:** `2603.15644` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Lin Lawrence Guo, Santiago Eduardo Arciniegas, Joseph Jihyung Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.15644 · **PDF:** https://arxiv.org/pdf/2603.15644v1
+
+## 8011. Build, Judge, Optimize: A Blueprint for Continuous Improvement of Multi-Agent Consumer Assistants
+
+- **ID:** `2603.03565` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Alejandro Breen Herrera, Aayush Sheth, Steven G. Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.03565 · **PDF:** https://arxiv.org/pdf/2603.03565v1
+
+## 8012. Act-Observe-Rewrite: Multimodal Coding Agents as In-Context Policy Learners for Robot Manipulation
+
+- **ID:** `2603.04466` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** Vaishak Kumar
+- **abs:** https://arxiv.org/abs/2603.04466 · **PDF:** https://arxiv.org/pdf/2603.04466v1
+
+## 8013. Real-time tightly coupled GNSS and IMU integration via Factor Graph Optimization
+
+- **ID:** `2603.03556` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** Radu-Andrei Cioaca, Paul Irofti, Cristian Rusu, et al.
+- **abs:** https://arxiv.org/abs/2603.03556 · **PDF:** https://arxiv.org/pdf/2603.03556v1
+
+## 8014. Molt Dynamics: Emergent Social Phenomena in Autonomous AI Agent Populations
+
+- **ID:** `2603.03555` · **primary:** `cs.MA` · **Submitted:** 2026-03-03
+- **Authors:** Brandon Yee, Krishna Sharma
+- **abs:** https://arxiv.org/abs/2603.03555 · **PDF:** https://arxiv.org/pdf/2603.03555v1
+
+## 8015. Preventing Curriculum Collapse in Self-Evolving Reasoning Systems
+
+- **ID:** `2603.13309` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Vaibhav Mishra
+- **abs:** https://arxiv.org/abs/2603.13309 · **PDF:** https://arxiv.org/pdf/2603.13309v1
+
+## 8016. Real-time loosely coupled GNSS and IMU integration via Factor Graph Optimization
+
+- **ID:** `2603.03546` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** Radu-Andrei Cioaca, Cristian Rusu, Paul Irofti, et al.
+- **abs:** https://arxiv.org/abs/2603.03546 · **PDF:** https://arxiv.org/pdf/2603.03546v1
+
+## 8017. Tucano 2 Cool: Better Open Source LLMs for Portuguese
+
+- **ID:** `2603.03543` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Nicholas Kluge Corrêa, Aniket Sen, Shiza Fatimah, et al.
+- **abs:** https://arxiv.org/abs/2603.03543 · **PDF:** https://arxiv.org/pdf/2603.03543v1
+
+## 8018. RAG-X: Systematic Diagnosis of Retrieval-Augmented Generation for Medical Question Answering
+
+- **ID:** `2603.03541` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Aswini Sivakumar, Vijayan Sugumaran, Yao Qiang
+- **abs:** https://arxiv.org/abs/2603.03541 · **PDF:** https://arxiv.org/pdf/2603.03541v1
+
+## 8019. Online Learnability of Chain-of-Thought Verifiers: Soundness and Completeness Trade-offs
+
+- **ID:** `2603.03538` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Maria-Florina Balcan, Avrim Blum, Kiriaki Fragkia, et al.
+- **abs:** https://arxiv.org/abs/2603.03538 · **PDF:** https://arxiv.org/pdf/2603.03538v1
+
+## 8020. SafeCRS: Personalized Safety Alignment for LLM-Based Conversational Recommender Systems
+
+- **ID:** `2603.03536` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Haochang Hao, Yifan Xu, Xinzhuo Li, et al.
+- **abs:** https://arxiv.org/abs/2603.03536 · **PDF:** https://arxiv.org/pdf/2603.03536v1
+
+## 8021. Trade-offs in Ensembling, Merging and Routing Among Parameter-Efficient Experts
+
+- **ID:** `2603.03535` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Sanae Lotfi, Lucas Caccia, Alessandro Sordoni, et al.
+- **abs:** https://arxiv.org/abs/2603.03535 · **PDF:** https://arxiv.org/pdf/2603.03535v1
+
+## 8022. GSI Agent: Domain Knowledge Enhancement for Large Language Models in Green Stormwater Infrastructure
+
+- **ID:** `2603.15643` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Shaohuang Wang
+- **abs:** https://arxiv.org/abs/2603.15643 · **PDF:** https://arxiv.org/pdf/2603.15643v1
+
+## 8023. Role-Aware Conditional Inference for Spatiotemporal Ecosystem Carbon Flux Prediction
+
+- **ID:** `2603.03531` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Yiming Sun, Runlong Yu, Rongchao Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.03531 · **PDF:** https://arxiv.org/pdf/2603.03531v1
+
+## 8024. Directional Neural Collapse Explains Few-Shot Transfer in Self-Supervised Learning
+
+- **ID:** `2603.03530` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Achleshwar Luthra, Yash Salunkhe, Tomer Galanti
+- **abs:** https://arxiv.org/abs/2603.03530 · **PDF:** https://arxiv.org/pdf/2603.03530v1
+
+## 8025. mlx-snn: Spiking Neural Networks on Apple Silicon via MLX
+
+- **ID:** `2603.03529` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Jiahao Qin
+- **abs:** https://arxiv.org/abs/2603.03529 · **PDF:** https://arxiv.org/pdf/2603.03529v1
+
+## 8026. Logit-Level Uncertainty Quantification in Vision-Language Models for Histopathology Image Analysis
+
+- **ID:** `2603.03527` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Betul Yurdem, Ferhat Ozgur Catak, Murat Kuzlu, et al.
+- **abs:** https://arxiv.org/abs/2603.03527 · **PDF:** https://arxiv.org/pdf/2603.03527v1
+
+## 8027. Task Expansion and Cross Refinement for Open-World Conditional Modeling
+
+- **ID:** `2603.13308` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Shreyas Bhat Brahmavar, Qiyang Liu, Yang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.13308 · **PDF:** https://arxiv.org/pdf/2603.13308v1
+
+## 8028. Multi-Agent Influence Diagrams to Hybrid Threat Modeling
+
+- **ID:** `2603.03526` · **primary:** `cs.MA` · **Submitted:** 2026-03-03
+- **Authors:** Maarten C. Vonk, Anna V. Kononova, Thomas Bäck, et al.
+- **abs:** https://arxiv.org/abs/2603.03526 · **PDF:** https://arxiv.org/pdf/2603.03526v1
+
+## 8029. Test-Time Meta-Adaptation with Self-Synthesis
+
+- **ID:** `2603.03524` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Zeyneb N. Kaya, Nick Rui
+- **abs:** https://arxiv.org/abs/2603.03524 · **PDF:** https://arxiv.org/pdf/2603.03524v2
+
+## 8030. Q-Measure-Learning for Continuous State RL: Efficient Implementation and Convergence
+
+- **ID:** `2603.03523` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Shengbo Wang
+- **abs:** https://arxiv.org/abs/2603.03523 · **PDF:** https://arxiv.org/pdf/2603.03523v1
+
+## 8031. MMAI Gym for Science: Training Liquid Foundation Models for Drug Discovery
+
+- **ID:** `2603.03517` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Maksim Kuznetsov, Zulfat Miftahutdinov, Rim Shayakhmetov, et al.
+- **abs:** https://arxiv.org/abs/2603.03517 · **PDF:** https://arxiv.org/pdf/2603.03517v1
+
+## 8032. The Controllability Trap: A Governance Framework for Military AI Agents
+
+- **ID:** `2603.03515` · **primary:** `cs.CY` · **Submitted:** 2026-03-03
+- **Authors:** Subramanyam Sahoo
+- **abs:** https://arxiv.org/abs/2603.03515 · **PDF:** https://arxiv.org/pdf/2603.03515v1
+
+## 8033. Baseline Performance of AI Tools in Classifying Cognitive Demand of Mathematical Tasks
+
+- **ID:** `2603.03512` · **primary:** `cs.CY` · **Submitted:** 2026-03-03
+- **Authors:** Danielle S. Fox, Brenda L. Robles, Elizabeth DiPietro Brovey, et al.
+- **abs:** https://arxiv.org/abs/2603.03512 · **PDF:** https://arxiv.org/pdf/2603.03512v1
+
+## 8034. Orbital Transformers for Predicting Wavefunctions in Time-Dependent Density Functional Theory
+
+- **ID:** `2603.03511` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Xuan Zhang, Haiyang Yu, Chengdong Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.03511 · **PDF:** https://arxiv.org/pdf/2603.03511v1
+
+## 8035. A theoretical model of dynamical grammatical gender shifting based on set-valued set function
+
+- **ID:** `2603.03510` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Mohamed El Idrissi
+- **abs:** https://arxiv.org/abs/2603.03510 · **PDF:** https://arxiv.org/pdf/2603.03510v1
+
+## 8036. Raising Bars, Not Parameters: LilMoo Compact Language Model for Hindi
+
+- **ID:** `2603.03508` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Shiza Fatimah, Aniket Sen, Sophia Falk, et al.
+- **abs:** https://arxiv.org/abs/2603.03508 · **PDF:** https://arxiv.org/pdf/2603.03508v1
+
+## 8037. Solving adversarial examples requires solving exponential misalignment
+
+- **ID:** `2603.03507` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Alessandro Salvatore, Stanislav Fort, Surya Ganguli
+- **abs:** https://arxiv.org/abs/2603.03507 · **PDF:** https://arxiv.org/pdf/2603.03507v2
+
+## 8038. PhyPrompt: RL-based Prompt Refinement for Physically Plausible Text-to-Video Generation
+
+- **ID:** `2603.03505` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Shang Wu, Chenwei Xu, Zhuofan Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.03505 · **PDF:** https://arxiv.org/pdf/2603.03505v1
+
+## 8039. Geographically-Weighted Weakly Supervised Bayesian High-Resolution Transformer for 200m Resolution Pan-Arctic Sea Ice Concentration Mapping and Uncertainty Estimation using Sentinel-1, RCM, and AMSR2 Data
+
+- **ID:** `2603.03503` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Mabel Heffring, Lincoln Linlin Xu
+- **abs:** https://arxiv.org/abs/2603.03503 · **PDF:** https://arxiv.org/pdf/2603.03503v1
+
+## 8040. Quantifying Ranking Instability Across Evaluation Protocol Axes in Gene Regulatory Network Benchmarking
+
+- **ID:** `2603.03493` · **primary:** `q-bio.MN` · **Submitted:** 2026-03-03
+- **Authors:** Ihor Kendiukhov
+- **abs:** https://arxiv.org/abs/2603.03493 · **PDF:** https://arxiv.org/pdf/2603.03493v1
+
+## 8041. When Small Variations Become Big Failures: Reliability Challenges in Compute-in-Memory Neural Accelerators
+
+- **ID:** `2603.03491` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Yifan Qin, Jiahao Zheng, Zheyu Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.03491 · **PDF:** https://arxiv.org/pdf/2603.03491v1
+
+## 8042. Phys4D: Fine-Grained Physics-Consistent 4D Modeling from Video Diffusion
+
+- **ID:** `2603.03485` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Haoran Lu, Shang Wu, Jianshu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03485 · **PDF:** https://arxiv.org/pdf/2603.03485v2
+
+## 8043. Optimal trajectory-guided stochastic co-optimization for e-fuel system design and real-time operation
+
+- **ID:** `2603.03484` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Jeongdong Kim, Minsu Kim, Jonggeol Na, et al.
+- **abs:** https://arxiv.org/abs/2603.03484 · **PDF:** https://arxiv.org/pdf/2603.03484v1
+
+## 8044. Understanding the Dynamics of Demonstration Conflict in In-Context Learning
+
+- **ID:** `2603.04464` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Difan Jiao, Di Wang, Lijie Hu
+- **abs:** https://arxiv.org/abs/2603.04464 · **PDF:** https://arxiv.org/pdf/2603.04464v1
+
+## 8045. Beyond Pixel Histories: World Models with Persistent 3D State
+
+- **ID:** `2603.03482` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Samuel Garcin, Thomas Walker, Steven McDonagh, et al.
+- **abs:** https://arxiv.org/abs/2603.03482 · **PDF:** https://arxiv.org/pdf/2603.03482v1
+
+## 8046. Minimax Optimal Strategy for Delayed Observations in Online Reinforcement Learning
+
+- **ID:** `2603.03480` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Harin Lee, Kevin Jamieson
+- **abs:** https://arxiv.org/abs/2603.03480 · **PDF:** https://arxiv.org/pdf/2603.03480v1
+
+## 8047. Stringology-Based Motif Discovery from EEG Signals: an ADHD Case Study
+
+- **ID:** `2603.03476` · **primary:** `q-bio.NC` · **Submitted:** 2026-03-03
+- **Authors:** Anat Dahan, Samah Ghazawi
+- **abs:** https://arxiv.org/abs/2603.03476 · **PDF:** https://arxiv.org/pdf/2603.03476v1
+
+## 8048. When Shallow Wins: Silent Failures and the Depth-Accuracy Paradox in Latent Reasoning
+
+- **ID:** `2603.03475` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Subramanyam Sahoo, Aman Chadha, Vinija Jain, et al.
+- **abs:** https://arxiv.org/abs/2603.03475 · **PDF:** https://arxiv.org/pdf/2603.03475v1
+
+## 8049. Biased Generalization in Diffusion Models
+
+- **ID:** `2603.03469` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Jerome Garnier-Brun, Luca Biggio, Davide Beltrame, et al.
+- **abs:** https://arxiv.org/abs/2603.03469 · **PDF:** https://arxiv.org/pdf/2603.03469v1
+
+## 8050. Graph Hopfield Networks: Energy-Based Node Classification with Associative Memory
+
+- **ID:** `2603.03464` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Abinav Rao, Alex Wa, Rishi Athavale
+- **abs:** https://arxiv.org/abs/2603.03464 · **PDF:** https://arxiv.org/pdf/2603.03464v1
+
+## 8051. Half the Nonlinearity Is Wasted: Measuring and Reallocating the Transformer's MLP Budget
+
+- **ID:** `2603.03459` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Peter Balogh
+- **abs:** https://arxiv.org/abs/2603.03459 · **PDF:** https://arxiv.org/pdf/2603.03459v2
+
+## 8052. Asymmetric Goal Drift in Coding Agents Under Value Conflict
+
+- **ID:** `2603.03456` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Magnus Saebo, Spencer Gibson, Tyler Crosse, et al.
+- **abs:** https://arxiv.org/abs/2603.03456 · **PDF:** https://arxiv.org/pdf/2603.03456v1
+
+## 8053. [Re] FairDICE: A Gap Between Theory And Practice
+
+- **ID:** `2603.03454` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Peter Adema, Karim Galliamov, Aleksey Evstratovskiy, et al.
+- **abs:** https://arxiv.org/abs/2603.03454 · **PDF:** https://arxiv.org/pdf/2603.03454v1
+
+## 8054. EEG-Based Brain-LLM Interface for Human Preference Aligned Generation
+
+- **ID:** `2603.16897` · **primary:** `eess.SP` · **Submitted:** 2026-03-03
+- **Authors:** Junzi Zhang, Jianing Shen, Weijie Tu, et al.
+- **abs:** https://arxiv.org/abs/2603.16897 · **PDF:** https://arxiv.org/pdf/2603.16897v1
+
+## 8055. CFG-Ctrl: Control-Based Classifier-Free Diffusion Guidance
+
+- **ID:** `2603.03281` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Hanyang Wang, Yiyang Liu, Jiawei Chi, et al.
+- **abs:** https://arxiv.org/abs/2603.03281 · **PDF:** https://arxiv.org/pdf/2603.03281v2
+
+## 8056. How to Peel with a Knife: Aligning Fine-Grained Manipulation with Human Preference
+
+- **ID:** `2603.03280` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** Toru Lin, Shuying Deng, Zhao-Heng Yin, et al.
+- **abs:** https://arxiv.org/abs/2603.03280 · **PDF:** https://arxiv.org/pdf/2603.03280v1
+
+## 8057. Tether: Autonomous Functional Play with Correspondence-Driven Trajectory Warping
+
+- **ID:** `2603.03278` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** William Liang, Sam Wang, Hung-Ju Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.03278 · **PDF:** https://arxiv.org/pdf/2603.03278v1
+
+## 8058. MultiGen: Level-Design for Editable Multiplayer Worlds in Diffusion Game Engines
+
+- **ID:** `2603.06679` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Ryan Po, David Junhao Zhang, Amir Hertz, et al.
+- **abs:** https://arxiv.org/abs/2603.06679 · **PDF:** https://arxiv.org/pdf/2603.06679v2
+
+## 8059. Learning Demographic-Conditioned Mobility Trajectories with Aggregate Supervision
+
+- **ID:** `2603.03275` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Jessie Z. Li, Zhiqing Hong, Toru Shirakawa, et al.
+- **abs:** https://arxiv.org/abs/2603.03275 · **PDF:** https://arxiv.org/pdf/2603.03275v1
+
+## 8060. Gravity Falls: A Comparative Analysis of Domain-Generation Algorithm (DGA) Detection Methods for Mobile Device Spearphishing
+
+- **ID:** `2603.03270` · **primary:** `cs.CR` · **Submitted:** 2026-03-03
+- **Authors:** Adam Dorian Wong, John D. Hastings
+- **abs:** https://arxiv.org/abs/2603.03270 · **PDF:** https://arxiv.org/pdf/2603.03270v1
+
+## 8061. LoGeR: Long-Context Geometric Reconstruction with Hybrid Memory
+
+- **ID:** `2603.03269` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Junyi Zhang, Charles Herrmann, Junhwa Hur, et al.
+- **abs:** https://arxiv.org/abs/2603.03269 · **PDF:** https://arxiv.org/pdf/2603.03269v1
+
+## 8062. Parallel Test-Time Scaling with Multi-Sequence Verifiers
+
+- **ID:** `2603.03417` · **primary:** `cs.CR` · **Submitted:** 2026-03-03
+- **Authors:** Yegon Kim, Seungyoo Lee, Chaeyun Jang, et al.
+- **abs:** https://arxiv.org/abs/2603.03417 · **PDF:** https://arxiv.org/pdf/2603.03417v1
+
+## 8063. Physics-informed post-processing of stabilized finite element solutions for transient convection-dominated problems
+
+- **ID:** `2603.03259` · **primary:** `math.NA` · **Submitted:** 2026-03-03
+- **Authors:** Süleyman Cengizci, Ömür Uğur, Srinivasan Natesan
+- **abs:** https://arxiv.org/abs/2603.03259 · **PDF:** https://arxiv.org/pdf/2603.03259v1
+
+## 8064. Inherited Goal Drift: Contextual Pressure Can Undermine Agentic Goals
+
+- **ID:** `2603.03258` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Achyutha Menon, Magnus Saebo, Tyler Crosse, et al.
+- **abs:** https://arxiv.org/abs/2603.03258 · **PDF:** https://arxiv.org/pdf/2603.03258v1
+
+## 8065. Farther the Shift, Sparser the Representation: Analyzing OOD Mechanisms in LLMs
+
+- **ID:** `2603.03415` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Mingyu Jin, Yutong Yin, Jingcheng Niu, et al.
+- **abs:** https://arxiv.org/abs/2603.03415 · **PDF:** https://arxiv.org/pdf/2603.03415v2
+
+## 8066. Valet: A Standardized Testbed of Traditional Imperfect-Information Card Games
+
+- **ID:** `2603.03252` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Mark Goadrich, Achille Morenville, Éric Piette
+- **abs:** https://arxiv.org/abs/2603.03252 · **PDF:** https://arxiv.org/pdf/2603.03252v1
+
+## 8067. Speculative Speculative Decoding
+
+- **ID:** `2603.03251` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Tanishq Kumar, Tri Dao, Avner May
+- **abs:** https://arxiv.org/abs/2603.03251 · **PDF:** https://arxiv.org/pdf/2603.03251v2
+
+## 8068. Using Learning Progressions to Guide AI Feedback for Science Learning
+
+- **ID:** `2603.03249` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Xin Xia, Nejla Yuruk, Yun Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.03249 · **PDF:** https://arxiv.org/pdf/2603.03249v1
+
+## 8069. Density-Guided Response Optimization: Community-Grounded Alignment via Implicit Acceptance Signals
+
+- **ID:** `2603.03242` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Patrick Gerard, Svitlana Volkova
+- **abs:** https://arxiv.org/abs/2603.03242 · **PDF:** https://arxiv.org/pdf/2603.03242v1
+
+## 8070. UniG2U-Bench: Do Unified Models Advance Multimodal Understanding?
+
+- **ID:** `2603.03241` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Zimo Wen, Boxiu Li, Wanbo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03241 · **PDF:** https://arxiv.org/pdf/2603.03241v1
+
+## 8071. On Geometry Regularization in Autoencoder Reduced-Order Models with Latent Neural ODE Dynamics
+
+- **ID:** `2603.03238` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Mikhail Osipov
+- **abs:** https://arxiv.org/abs/2603.03238 · **PDF:** https://arxiv.org/pdf/2603.03238v1
+
+## 8072. The elbow statistic: Multiscale clustering statistical significance
+
+- **ID:** `2603.03235` · **primary:** `stat.ML` · **Submitted:** 2026-03-03
+- **Authors:** Francisco J. Perez-Reche
+- **abs:** https://arxiv.org/abs/2603.03235 · **PDF:** https://arxiv.org/pdf/2603.03235v1
+
+## 8073. PRIVATEEDIT: A Privacy-Preserving Pipeline for Face-Centric Generative Image Editing
+
+- **ID:** `2603.03412` · **primary:** `cs.CR` · **Submitted:** 2026-03-03
+- **Authors:** Dipesh Tamboli, Vineet Punyamoorty, Atharv Pawar, et al.
+- **abs:** https://arxiv.org/abs/2603.03412 · **PDF:** https://arxiv.org/pdf/2603.03412v1
+
+## 8074. Guiding Sparse Neural Networks with Neurobiological Principles to Elicit Biologically Plausible Representations
+
+- **ID:** `2603.03234` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Patrick Inoue, Florian Röhrbein, Andreas Knoblauch
+- **abs:** https://arxiv.org/abs/2603.03234 · **PDF:** https://arxiv.org/pdf/2603.03234v1
+
+## 8075. AI-for-Science Low-code Platform with Bayesian Adversarial Multi-Agent Framework
+
+- **ID:** `2603.03233` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Zihang Zeng, Jiaquan Zhang, Pengze Li, et al.
+- **abs:** https://arxiv.org/abs/2603.03233 · **PDF:** https://arxiv.org/pdf/2603.03233v1
+
+## 8076. SynthCharge: An Electric Vehicle Routing Instance Generator with Feasibility Screening to Enable Learning-Based Optimization and Benchmarking
+
+- **ID:** `2603.03230` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Mertcan Daysalilar, Fuat Uyguroglu, Gabriel Nicolosi, et al.
+- **abs:** https://arxiv.org/abs/2603.03230 · **PDF:** https://arxiv.org/pdf/2603.03230v1
+
+## 8077. Inverse Reconstruction of Shock Time Series from Shock Response Spectrum Curves using Machine Learning
+
+- **ID:** `2603.03229` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Adam Watts, Andrew Jeon, Destry Newton, et al.
+- **abs:** https://arxiv.org/abs/2603.03229 · **PDF:** https://arxiv.org/pdf/2603.03229v2
+
+## 8078. Coalgebras for categorical deep learning: Representability and universal approximation
+
+- **ID:** `2603.03227` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Dragan Mašulović
+- **abs:** https://arxiv.org/abs/2603.03227 · **PDF:** https://arxiv.org/pdf/2603.03227v1
+
+## 8079. Adaptive Methods Are Preferable in High Privacy Settings: An SDE Perspective
+
+- **ID:** `2603.03226` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Enea Monzio Compagnoni, Alessandro Stanghellini, Rustem Islamov, et al.
+- **abs:** https://arxiv.org/abs/2603.03226 · **PDF:** https://arxiv.org/pdf/2603.03226v1
+
+## 8080. Stabilized Adaptive Loss and Residual-Based Collocation for Physics-Informed Neural Networks
+
+- **ID:** `2603.03224` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Divyavardhan Singh, Shubham Kamble, Dimple Sonone, et al.
+- **abs:** https://arxiv.org/abs/2603.03224 · **PDF:** https://arxiv.org/pdf/2603.03224v1
+
+## 8081. Scalable Contrastive Causal Discovery under Unknown Soft Interventions
+
+- **ID:** `2603.03411` · **primary:** `stat.ML` · **Submitted:** 2026-03-03
+- **Authors:** Mingxuan Zhang, Khushi Desai, Sopho Kevlishvili, et al.
+- **abs:** https://arxiv.org/abs/2603.03411 · **PDF:** https://arxiv.org/pdf/2603.03411v1
+
+## 8082. NeuroSkill(tm): Proactive Real-Time Agentic System Capable of Modeling Human State of Mind
+
+- **ID:** `2603.03212` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Nataliya Kosmyna, Eugene Hauptmann
+- **abs:** https://arxiv.org/abs/2603.03212 · **PDF:** https://arxiv.org/pdf/2603.03212v1
+
+## 8083. Shape Derivative-Informed Neural Operators with Application to Risk-Averse Shape Optimization
+
+- **ID:** `2603.03211` · **primary:** `math.OC` · **Submitted:** 2026-03-03
+- **Authors:** Xindi Gong, Dingcheng Luo, Thomas O'Leary-Roseberry, et al.
+- **abs:** https://arxiv.org/abs/2603.03211 · **PDF:** https://arxiv.org/pdf/2603.03211v1
+
+## 8084. I-CAM-UV: Integrating Causal Graphs over Non-Identical Variable Sets Using Causal Additive Models with Unobserved Variables
+
+- **ID:** `2603.03207` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Hirofumi Suzuki, Kentaro Kanamori, Takuya Takagi, et al.
+- **abs:** https://arxiv.org/abs/2603.03207 · **PDF:** https://arxiv.org/pdf/2603.03207v1
+
+## 8085. Understanding and Mitigating Dataset Corruption in LLM Steering
+
+- **ID:** `2603.03206` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Cullen Anderson, Narmeen Oozeer, Foad Namjoo, et al.
+- **abs:** https://arxiv.org/abs/2603.03206 · **PDF:** https://arxiv.org/pdf/2603.03206v1
+
+## 8086. Learning When to Act or Refuse: Guarding Agentic Reasoning Models for Safe Multi-Step Tool Use
+
+- **ID:** `2603.03205` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Aradhye Agarwal, Gurdit Siyan, Yash Pandya, et al.
+- **abs:** https://arxiv.org/abs/2603.03205 · **PDF:** https://arxiv.org/pdf/2603.03205v1
+
+## 8087. No Memorization, No Detection: Output Distribution-Based Contamination Detection in Small Language Models
+
+- **ID:** `2603.03203` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Omer Sela
+- **abs:** https://arxiv.org/abs/2603.03203 · **PDF:** https://arxiv.org/pdf/2603.03203v3
+
+## 8088. Code2Math: Can Your Code Agent Effectively Evolve Math Problems Through Exploration?
+
+- **ID:** `2603.03202` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Dadi Guo, Yuejin Xie, Qingyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.03202 · **PDF:** https://arxiv.org/pdf/2603.03202v2
+
+## 8089. A Dynamical Theory of Sequential Retrieval in Input-Driven Hopfield Networks
+
+- **ID:** `2603.03201` · **primary:** `cs.NE` · **Submitted:** 2026-03-03
+- **Authors:** Simone Betteti, Giacomo Baggio, Sandro Zampieri
+- **abs:** https://arxiv.org/abs/2603.03201 · **PDF:** https://arxiv.org/pdf/2603.03201v2
+
+## 8090. ACE-Brain-0: Spatial Intelligence as a Shared Scaffold for Universal Embodiments
+
+- **ID:** `2603.03198` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** Ziyang Gong, Zehang Luo, Anke Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.03198 · **PDF:** https://arxiv.org/pdf/2603.03198v1
+
+## 8091. Infinite dimensional generative sensing
+
+- **ID:** `2603.03196` · **primary:** `math.NA` · **Submitted:** 2026-03-03
+- **Authors:** Paolo Angella, Vito Paolo Pastore, Matteo Santacesaria
+- **abs:** https://arxiv.org/abs/2603.03196 · **PDF:** https://arxiv.org/pdf/2603.03196v1
+
+## 8092. Chain of World: World Model Thinking in Latent Motion
+
+- **ID:** `2603.03195` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Fuxiang Yang, Donglin Di, Lulu Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.03195 · **PDF:** https://arxiv.org/pdf/2603.03195v1
+
+## 8093. BeyondSWE: Can Current Code Agent Survive Beyond Single-Repo Bug Fixing?
+
+- **ID:** `2603.03194` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Guoxin Chen, Fanzhe Meng, Jiale Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.03194 · **PDF:** https://arxiv.org/pdf/2603.03194v1
+
+## 8094. MoD-DPO: Towards Mitigating Cross-modal Hallucinations in Omni LLMs using Modality Decoupled Preference Optimization
+
+- **ID:** `2603.03192` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Ashutosh Chaubey, Jiacheng Pang, Mohammad Soleymani
+- **abs:** https://arxiv.org/abs/2603.03192 · **PDF:** https://arxiv.org/pdf/2603.03192v2
+
+## 8095. On Google's SynthID-Text LLM Watermarking System: Theoretical Analysis and Empirical Validation
+
+- **ID:** `2603.03410` · **primary:** `cs.CR` · **Submitted:** 2026-03-03
+- **Authors:** Romina Omidi, Yun Dong, Binghui Wang
+- **abs:** https://arxiv.org/abs/2603.03410 · **PDF:** https://arxiv.org/pdf/2603.03410v2
+
+## 8096. A Covering Framework for Offline POMDPs Learning using Belief Space Metric
+
+- **ID:** `2603.03191` · **primary:** `stat.ML` · **Submitted:** 2026-03-03
+- **Authors:** Youheng Zhu, Yiping Lu
+- **abs:** https://arxiv.org/abs/2603.03191 · **PDF:** https://arxiv.org/pdf/2603.03191v1
+
+## 8097. Expectation and Acoustic Neural Network Representations Enhance Music Identification from Brain Activity
+
+- **ID:** `2603.03190` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Shogo Noguchi, Taketo Akama, Tai Nakamura, et al.
+- **abs:** https://arxiv.org/abs/2603.03190 · **PDF:** https://arxiv.org/pdf/2603.03190v2
+
+## 8098. Scalable Uncertainty Quantification for Black-Box Density-Based Clustering
+
+- **ID:** `2603.03188` · **primary:** `stat.ML` · **Submitted:** 2026-03-03
+- **Authors:** Nicola Bariletto, Stephen G. Walker
+- **abs:** https://arxiv.org/abs/2603.03188 · **PDF:** https://arxiv.org/pdf/2603.03188v1
+
+## 8099. Type-Aware Retrieval-Augmented Generation with Dependency Closure for Solver-Executable Industrial Optimization Modeling
+
+- **ID:** `2603.03180` · **primary:** `cs.SE` · **Submitted:** 2026-03-03
+- **Authors:** Y. Zhong, R. Huang, M. Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.03180 · **PDF:** https://arxiv.org/pdf/2603.03180v1
+
+## 8100. Neuro-Symbolic Artificial Intelligence: A Task-Directed Survey in the Black-Box Models Era
+
+- **ID:** `2603.03177` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Giovanni Pio Delvecchio, Lorenzo Molfetta, Gianluca Moro
+- **abs:** https://arxiv.org/abs/2603.03177 · **PDF:** https://arxiv.org/pdf/2603.03177v1
+
+## 8101. A Short Note on a Variant of the Squint Algorithm
+
+- **ID:** `2603.03409` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Haipeng Luo
+- **abs:** https://arxiv.org/abs/2603.03409 · **PDF:** https://arxiv.org/pdf/2603.03409v1
+
+## 8102. FEAST: Retrieval-Augmented Multi-Hierarchical Food Classification for the FoodEx2 System
+
+- **ID:** `2603.03176` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Lorenzo Molfetta, Alessio Cocchieri, Stefano Fantazzini, et al.
+- **abs:** https://arxiv.org/abs/2603.03176 · **PDF:** https://arxiv.org/pdf/2603.03176v1
+
+## 8103. Saarthi for AGI: Towards Domain-Specific General Intelligence for Formal Verification
+
+- **ID:** `2603.03175` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Aman Kumar, Deepak Narayan Gadde, Luu Danh Minh, et al.
+- **abs:** https://arxiv.org/abs/2603.03175 · **PDF:** https://arxiv.org/pdf/2603.03175v1
+
+## 8104. Less Noise, Same Certificate: Retain Sensitivity for Unlearning
+
+- **ID:** `2603.03172` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Carolin Heinzler, Kasra Malihi, Amartya Sanyal
+- **abs:** https://arxiv.org/abs/2603.03172 · **PDF:** https://arxiv.org/pdf/2603.03172v1
+
+## 8105. Thinking into the Future: Latent Lookahead Training for Transformers
+
+- **ID:** `2603.20219` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Lorenzo Noci, Gregor Bachmann, Seyed-Mohsen Moosavi-Dezfooli, et al.
+- **abs:** https://arxiv.org/abs/2603.20219 · **PDF:** https://arxiv.org/pdf/2603.20219v1
+
+## 8106. Conditioned Activation Transport for T2I Safety Steering
+
+- **ID:** `2603.03163` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Maciej Chrabąszcz, Aleksander Szymczyk, Jan Dubiński, et al.
+- **abs:** https://arxiv.org/abs/2603.03163 · **PDF:** https://arxiv.org/pdf/2603.03163v1
+
+## 8107. Tracing Pharmacological Knowledge In Large Language Models
+
+- **ID:** `2603.03407` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Basil Hasan Khwaja, Dylan Chen, Guntas Toor, et al.
+- **abs:** https://arxiv.org/abs/2603.03407 · **PDF:** https://arxiv.org/pdf/2603.03407v1
+
+## 8108. An Investigation Into Various Approaches For Bengali Long-Form Speech Transcription and Bengali Speaker Diarization
+
+- **ID:** `2603.03158` · **primary:** `cs.SD` · **Submitted:** 2026-03-03
+- **Authors:** Epshita Jahan, Khandoker Md Tanjinul Islam, Pritom Biswas, et al.
+- **abs:** https://arxiv.org/abs/2603.03158 · **PDF:** https://arxiv.org/pdf/2603.03158v1
+
+## 8109. Information Routing in Atomistic Foundation Models: How Task Alignment and Equivariance Shape Linear Disentanglement
+
+- **ID:** `2603.03155` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Joshua Steier
+- **abs:** https://arxiv.org/abs/2603.03155 · **PDF:** https://arxiv.org/pdf/2603.03155v2
+
+## 8110. Surprisal-Rényi Free Energy
+
+- **ID:** `2603.03405` · **primary:** `stat.ML` · **Submitted:** 2026-03-03
+- **Authors:** Shion Matsumoto, Raul Castillo, Benjamin Prada, et al.
+- **abs:** https://arxiv.org/abs/2603.03405 · **PDF:** https://arxiv.org/pdf/2603.03405v1
+
+## 8111. Agentic AI-based Coverage Closure for Formal Verification
+
+- **ID:** `2603.03147` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Sivaram Pothireddypalli, Ashish Raman, Deepak Narayan Gadde, et al.
+- **abs:** https://arxiv.org/abs/2603.03147 · **PDF:** https://arxiv.org/pdf/2603.03147v2
+
+## 8112. Channel-Adaptive Edge AI: Maximizing Inference Throughput by Adapting Computational Complexity to Channel States
+
+- **ID:** `2603.03146` · **primary:** `cs.IT` · **Submitted:** 2026-03-03
+- **Authors:** Jierui Zhang, Jianhao Huang, Kaibin Huang
+- **abs:** https://arxiv.org/abs/2603.03146 · **PDF:** https://arxiv.org/pdf/2603.03146v1
+
+## 8113. Geometry-Guided Reinforcement Learning for Multi-view Consistent 3D Scene Editing
+
+- **ID:** `2603.03143` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Jiyuan Wang, Chunyu Lin, Lei Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.03143 · **PDF:** https://arxiv.org/pdf/2603.03143v1
+
+## 8114. APRES: An Agentic Paper Revision and Evaluation System
+
+- **ID:** `2603.03142` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Bingchen Zhao, Jenny Zhang, Chenxi Whitehouse, et al.
+- **abs:** https://arxiv.org/abs/2603.03142 · **PDF:** https://arxiv.org/pdf/2603.03142v1
+
+## 8115. How to Model AI Agents as Personas?: Applying the Persona Ecosystem Playground to 41,300 Posts on Moltbook for Behavioral Insights
+
+- **ID:** `2603.03140` · **primary:** `cs.HC` · **Submitted:** 2026-03-03
+- **Authors:** Danial Amin, Joni Salminen, Bernard J. Jansen
+- **abs:** https://arxiv.org/abs/2603.03140 · **PDF:** https://arxiv.org/pdf/2603.03140v2
+
+## 8116. An experimental study of KV cache reuse strategies in chunk-level caching systems
+
+- **ID:** `2603.20218` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Samuel Cestola, Tianxiang Xia, Zheng Weiyan, et al.
+- **abs:** https://arxiv.org/abs/2603.20218 · **PDF:** https://arxiv.org/pdf/2603.20218v1
+
+## 8117. Torus embeddings
+
+- **ID:** `2603.03135` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Dan Stowell
+- **abs:** https://arxiv.org/abs/2603.03135 · **PDF:** https://arxiv.org/pdf/2603.03135v1
+
+## 8118. UniSkill: A Dataset for Matching University Curricula to Professional Competencies
+
+- **ID:** `2603.03134` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Nurlan Musazade, Joszef Mezei, Mike Zhang
+- **abs:** https://arxiv.org/abs/2603.03134 · **PDF:** https://arxiv.org/pdf/2603.03134v1
+
+## 8119. Joint Training Across Multiple Activation Sparsity Regimes
+
+- **ID:** `2603.03131` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Haotian Wang
+- **abs:** https://arxiv.org/abs/2603.03131 · **PDF:** https://arxiv.org/pdf/2603.03131v1
+
+## 8120. AI Space Physics: Constitutive boundary semantics for open AI institutions
+
+- **ID:** `2603.03119` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Oleg Romanchuk, Roman Bondar
+- **abs:** https://arxiv.org/abs/2603.03119 · **PDF:** https://arxiv.org/pdf/2603.03119v1
+
+## 8121. Suppressing Domain-Specific Hallucination in Construction LLMs: A Knowledge Graph Foundation for GraphRAG and QLoRA on River and Sediment Control Technical Standards
+
+- **ID:** `2603.13307` · **primary:** `cs.IR` · **Submitted:** 2026-03-03
+- **Authors:** Takato Yasuno
+- **abs:** https://arxiv.org/abs/2603.13307 · **PDF:** https://arxiv.org/pdf/2603.13307v1
+
+## 8122. Beyond Task Completion: Revealing Corrupt Success in LLM Agents through Procedure-Aware Evaluation
+
+- **ID:** `2603.03116` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Hongliu Cao, Ilias Driouich, Eoin Thomas
+- **abs:** https://arxiv.org/abs/2603.03116 · **PDF:** https://arxiv.org/pdf/2603.03116v1
+
+## 8123. From Complex Dynamics to DynFormer: Rethinking Transformers for PDEs
+
+- **ID:** `2603.03112` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Pengyu Lai, Yixiao Chen, Dewu Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.03112 · **PDF:** https://arxiv.org/pdf/2603.03112v1
+
+## 8124. Evaluating Performance Drift from Model Switching in Multi-Turn LLM Systems
+
+- **ID:** `2603.03111` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Raad Khraishi, Iman Zafar, Katie Myles, et al.
+- **abs:** https://arxiv.org/abs/2603.03111 · **PDF:** https://arxiv.org/pdf/2603.03111v1
+
+## 8125. Multi-Scale Adaptive Neighborhood Awareness Transformer For Graph Fraud Detection
+
+- **ID:** `2603.03106` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Jiaqi Lv, Qingfeng Du, Yu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03106 · **PDF:** https://arxiv.org/pdf/2603.03106v1
+
+## 8126. MoECLIP: Patch-Specialized Experts for Zero-shot Anomaly Detection
+
+- **ID:** `2603.03101` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Jun Yeong Park, JunYoung Seo, Minji Kang, et al.
+- **abs:** https://arxiv.org/abs/2603.03101 · **PDF:** https://arxiv.org/pdf/2603.03101v2
+
+## 8127. Why Adam Can Beat SGD: Second-Moment Normalization Yields Sharper Tails
+
+- **ID:** `2603.03099` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Ruinan Jin, Yingbin Liang, Shaofeng Zou
+- **abs:** https://arxiv.org/abs/2603.03099 · **PDF:** https://arxiv.org/pdf/2603.03099v3
+
+## 8128. Odin: Multi-Signal Graph Intelligence for Autonomous Discovery in Knowledge Graphs
+
+- **ID:** `2603.03097` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Muyukani Kizito, Elizabeth Nyambere
+- **abs:** https://arxiv.org/abs/2603.03097 · **PDF:** https://arxiv.org/pdf/2603.03097v1
+
+## 8129. Interpreting Speaker Characteristics in the Dimensions of Self-Supervised Speech Features
+
+- **ID:** `2603.03096` · **primary:** `eess.AS` · **Submitted:** 2026-03-03
+- **Authors:** Kyle Janse van Rensburg, Benjamin van Niekerk, Herman Kamper
+- **abs:** https://arxiv.org/abs/2603.03096 · **PDF:** https://arxiv.org/pdf/2603.03096v1
+
+## 8130. Compact Prompting in Instruction-tuned LLMs for Joint Argumentative Component Detection
+
+- **ID:** `2603.03095` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Sofiane Elguendouze, Erwan Hain, Elena Cabrio, et al.
+- **abs:** https://arxiv.org/abs/2603.03095 · **PDF:** https://arxiv.org/pdf/2603.03095v1
+
+## 8131. Proactive Guiding Strategy for Item-side Fairness in Interactive Recommendation
+
+- **ID:** `2603.03094` · **primary:** `cs.IR` · **Submitted:** 2026-03-03
+- **Authors:** Chongjun Xia, Xiaoyu Shi, Hong Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.03094 · **PDF:** https://arxiv.org/pdf/2603.03094v1
+
+## 8132. On the Expressive Power of Transformers for Maxout Networks and Continuous Piecewise Linear Functions
+
+- **ID:** `2603.03084` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Linyan Gu, Lihua Yang, Feng Zhou
+- **abs:** https://arxiv.org/abs/2603.03084 · **PDF:** https://arxiv.org/pdf/2603.03084v1
+
+## 8133. Safe and Robust Domains of Attraction for Discrete-Time Systems: A Set-Based Characterization and Certifiable Neural Network Estimation
+
+- **ID:** `2603.03082` · **primary:** `eess.SY` · **Submitted:** 2026-03-03
+- **Authors:** Mohamed Serry, Maxwell Fitzsimmons, Jun Liu
+- **abs:** https://arxiv.org/abs/2603.03082 · **PDF:** https://arxiv.org/pdf/2603.03082v1
+
+## 8134. TAO-Attack: Toward Advanced Optimization-Based Jailbreak Attacks for Large Language Models
+
+- **ID:** `2603.03081` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Zhi Xu, Jiaqi Li, Xiaotong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03081 · **PDF:** https://arxiv.org/pdf/2603.03081v1
+
+## 8135. Beyond Factual Correctness: Mitigating Preference-Inconsistent Explanations in Explainable Recommendation
+
+- **ID:** `2603.03080` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Chengkai Wang, Baisong Liu
+- **abs:** https://arxiv.org/abs/2603.03080 · **PDF:** https://arxiv.org/pdf/2603.03080v1
+
+## 8136. RAPO: Expanding Exploration for LLM Agents via Retrieval-Augmented Policy Optimization
+
+- **ID:** `2603.03078` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Siwei Zhang, Yun Xiong, Xi Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.03078 · **PDF:** https://arxiv.org/pdf/2603.03078v1
+
+## 8137. TinyIceNet: Low-Power SAR Sea Ice Segmentation for On-Board FPGA Inference
+
+- **ID:** `2603.03075` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Mhd Rashed Al Koutayni, Mohamed Selim, Gerd Reis, et al.
+- **abs:** https://arxiv.org/abs/2603.03075 · **PDF:** https://arxiv.org/pdf/2603.03075v1
+
+## 8138. Design Generative AI for Practitioners: Exploring Interaction Approaches Aligned with Creative Practice
+
+- **ID:** `2603.03074` · **primary:** `cs.HC` · **Submitted:** 2026-03-03
+- **Authors:** Xiaohan Peng, Wendy E. Mackay, Janin Koch
+- **abs:** https://arxiv.org/abs/2603.03074 · **PDF:** https://arxiv.org/pdf/2603.03074v1
+
+## 8139. TikZilla: Scaling Text-to-TikZ with High-Quality Data and Reinforcement Learning
+
+- **ID:** `2603.03072` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Christian Greisinger, Steffen Eger
+- **abs:** https://arxiv.org/abs/2603.03072 · **PDF:** https://arxiv.org/pdf/2603.03072v2
+
+## 8140. From Reachability to Learnability: Geometric Design Principles for Quantum Neural Networks
+
+- **ID:** `2603.03071` · **primary:** `quant-ph` · **Submitted:** 2026-03-03
+- **Authors:** Vishal S. Ngairangbam, Michael Spannowsky
+- **abs:** https://arxiv.org/abs/2603.03071 · **PDF:** https://arxiv.org/pdf/2603.03071v2
+
+## 8141. Reinforcement Learning with Symbolic Reward Machines
+
+- **ID:** `2603.03068` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Thomas Krug, Daniel Neider
+- **abs:** https://arxiv.org/abs/2603.03068 · **PDF:** https://arxiv.org/pdf/2603.03068v1
+
+## 8142. Incremental Graph Construction Enables Robust Spectral Clustering of Texts
+
+- **ID:** `2603.03056` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Marko Pranjić, Boshko Koloski, Nada Lavrač, et al.
+- **abs:** https://arxiv.org/abs/2603.03056 · **PDF:** https://arxiv.org/pdf/2603.03056v2
+
+## 8143. PrivMedChat: End-to-End Differentially Private RLHF for Medical Dialogue Systems
+
+- **ID:** `2603.03054` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Sudip Bhujel
+- **abs:** https://arxiv.org/abs/2603.03054 · **PDF:** https://arxiv.org/pdf/2603.03054v2
+
+## 8144. Chart Deep Research in LVLMs via Parallel Relative Policy Optimization
+
+- **ID:** `2603.06677` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Jiajin Tang, Gaoyang, Wenjie Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.06677 · **PDF:** https://arxiv.org/pdf/2603.06677v1
+
+## 8145. TrustMH-Bench: A Comprehensive Benchmark for Evaluating the Trustworthiness of Large Language Models in Mental Health
+
+- **ID:** `2603.03047` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Zixin Xiong, Ziteng Wang, Haotian Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.03047 · **PDF:** https://arxiv.org/pdf/2603.03047v1
+
+## 8146. QFlowNet: Fast, Diverse, and Efficient Unitary Synthesis with Generative Flow Networks
+
+- **ID:** `2603.03045` · **primary:** `quant-ph` · **Submitted:** 2026-03-03
+- **Authors:** Inhoe Koo, Hyunho Cha, Jungwoo Lee
+- **abs:** https://arxiv.org/abs/2603.03045 · **PDF:** https://arxiv.org/pdf/2603.03045v2
+
+## 8147. IoUCert: Robustness Verification for Anchor-based Object Detectors
+
+- **ID:** `2603.03043` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Benedikt Brückner, Alejandro J. Mercado, Yanghao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03043 · **PDF:** https://arxiv.org/pdf/2603.03043v2
+
+## 8148. Heterogeneous Time Constants Improve Stability in Equilibrium Propagation
+
+- **ID:** `2603.03402` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Yoshimasa Kubo, Suhani Pragnesh Modi, Smit Patel
+- **abs:** https://arxiv.org/abs/2603.03402 · **PDF:** https://arxiv.org/pdf/2603.03402v1
+
+## 8149. cPNN: Continuous Progressive Neural Networks for Evolving Streaming Time Series
+
+- **ID:** `2603.03040` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Federico Giannini, Giacomo Ziffer, Emanuele Della Valle
+- **abs:** https://arxiv.org/abs/2603.03040 · **PDF:** https://arxiv.org/pdf/2603.03040v1
+
+## 8150. Generalized Bayes for Causal Inference
+
+- **ID:** `2603.03035` · **primary:** `stat.ML` · **Submitted:** 2026-03-03
+- **Authors:** Emil Javurek, Dennis Frauen, Yuxin Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.03035 · **PDF:** https://arxiv.org/pdf/2603.03035v1
+
+## 8151. Step-Level Sparse Autoencoder for Reasoning Process Interpretation
+
+- **ID:** `2603.03031` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Xuan Yang, Jiayu Liu, Yuhang Lai, et al.
+- **abs:** https://arxiv.org/abs/2603.03031 · **PDF:** https://arxiv.org/pdf/2603.03031v1
+
+## 8152. Beyond Cross-Validation: Adaptive Parameter Selection for Kernel-Based Gradient Descents
+
+- **ID:** `2603.03401` · **primary:** `stat.ML` · **Submitted:** 2026-03-03
+- **Authors:** Xiaotong Liu, Yunwen Lei, Xiangyu Chang, et al.
+- **abs:** https://arxiv.org/abs/2603.03401 · **PDF:** https://arxiv.org/pdf/2603.03401v1
+
+## 8153. MA-CoNav: A Master-Slave Multi-Agent Framework with Hierarchical Collaboration and Dual-Level Reflection for Long-Horizon Embodied VLN
+
+- **ID:** `2603.03024` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** Ling Luo, Qianqian Bai
+- **abs:** https://arxiv.org/abs/2603.03024 · **PDF:** https://arxiv.org/pdf/2603.03024v1
+
+## 8154. SEHFS: Structural Entropy-Guided High-Order Correlation Learning for Multi-View Multi-Label Feature Selection
+
+- **ID:** `2603.03022` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Cheng Peng, Yonghao Li, Wanfu Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.03022 · **PDF:** https://arxiv.org/pdf/2603.03022v1
+
+## 8155. REGAL: A Registry-Driven Architecture for Deterministic Grounding of Agentic AI in Enterprise Telemetry
+
+- **ID:** `2603.03018` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Yuvraj Agrawal
+- **abs:** https://arxiv.org/abs/2603.03018 · **PDF:** https://arxiv.org/pdf/2603.03018v1
+
+## 8156. Breaking the Prototype Bias Loop: Confidence-Aware Federated Contrastive Learning for Highly Imbalanced Clients
+
+- **ID:** `2603.03007` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Tian-Shuang Wu, Shen-Huan Lyu, Ning Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.03007 · **PDF:** https://arxiv.org/pdf/2603.03007v1
+
+## 8157. CraniMem: Cranial Inspired Gated and Bounded Memory for Agentic Systems
+
+- **ID:** `2603.15642` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Pearl Mody, Mihir Panchal, Rishit Kar, et al.
+- **abs:** https://arxiv.org/abs/2603.15642 · **PDF:** https://arxiv.org/pdf/2603.15642v1
+
+## 8158. OrchMAS: Orchestrated Reasoning with Multi Collaborative Heterogeneous Scientific Expert Structured Agents
+
+- **ID:** `2603.03005` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Yichao Feng, Haoran Luo, Zhenghong Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.03005 · **PDF:** https://arxiv.org/pdf/2603.03005v1
+
+## 8159. SpatialText: A Pure-Text Cognitive Benchmark for Spatial Understanding in Large Language Models
+
+- **ID:** `2603.03002` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Peiyao Jiang, Zequn Qin, Xi Li
+- **abs:** https://arxiv.org/abs/2603.03002 · **PDF:** https://arxiv.org/pdf/2603.03002v1
+
+## 8160. MaBERT:A Padding Safe Interleaved Transformer Mamba Hybrid Encoder for Efficient Extended Context Masked Language Modeling
+
+- **ID:** `2603.03001` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Jinwoong Kim, Sangjin Park
+- **abs:** https://arxiv.org/abs/2603.03001 · **PDF:** https://arxiv.org/pdf/2603.03001v1
+
+## 8161. Why Does RLAIF Work At All?
+
+- **ID:** `2603.03000` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Robin Young
+- **abs:** https://arxiv.org/abs/2603.03000 · **PDF:** https://arxiv.org/pdf/2603.03000v1
+
+## 8162. Variance reduction in lattice QCD observables via normalizing flows
+
+- **ID:** `2603.02984` · **primary:** `hep-lat` · **Submitted:** 2026-03-03
+- **Authors:** Ryan Abbott, Denis Boyda, Yang Fu, et al.
+- **abs:** https://arxiv.org/abs/2603.02984 · **PDF:** https://arxiv.org/pdf/2603.02984v1
+
+## 8163. Contextualized Privacy Defense for LLM Agents
+
+- **ID:** `2603.02983` · **primary:** `cs.CR` · **Submitted:** 2026-03-03
+- **Authors:** Yule Wen, Yanzhe Zhang, Jianxun Lian, et al.
+- **abs:** https://arxiv.org/abs/2603.02983 · **PDF:** https://arxiv.org/pdf/2603.02983v1
+
+## 8164. On the Topology of Neural Network Superlevel Sets
+
+- **ID:** `2603.02973` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Bahman Gharesifard
+- **abs:** https://arxiv.org/abs/2603.02973 · **PDF:** https://arxiv.org/pdf/2603.02973v1
+
+## 8165. XAI and Few-shot-based Hybrid Classification Model for Plant Leaf Disease Prognosis
+
+- **ID:** `2603.06676` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Diana Susan Joseph, Pranav M Pawar, Raja Muthalagu, et al.
+- **abs:** https://arxiv.org/abs/2603.06676 · **PDF:** https://arxiv.org/pdf/2603.06676v1
+
+## 8166. LAGO: A Local-Global Optimization Framework Combining Trust Region Methods and Bayesian Optimization
+
+- **ID:** `2603.02970` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Eliott Van Dieren, Tommaso Vanzan, Fabio Nobile
+- **abs:** https://arxiv.org/abs/2603.02970 · **PDF:** https://arxiv.org/pdf/2603.02970v1
+
+## 8167. Integrating Homomorphic Encryption and Synthetic Data in FL for Privacy and Learning Quality
+
+- **ID:** `2603.02969` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Yenan Wang, Carla Fabiana Chiasserini, Elad Michael Schiller
+- **abs:** https://arxiv.org/abs/2603.02969 · **PDF:** https://arxiv.org/pdf/2603.02969v1
+
+## 8168. Delegation and Verification Under AI
+
+- **ID:** `2603.02961` · **primary:** `cs.GT` · **Submitted:** 2026-03-03
+- **Authors:** Lingxiao Huang, Wenyang Xiao, Nisheeth K. Vishnoi
+- **abs:** https://arxiv.org/abs/2603.02961 · **PDF:** https://arxiv.org/pdf/2603.02961v1
+
+## 8169. Architecting Trust in Artificial Epistemic Agents
+
+- **ID:** `2603.02960` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Nahema Marchal, Stephanie Chan, Matija Franklin, et al.
+- **abs:** https://arxiv.org/abs/2603.02960 · **PDF:** https://arxiv.org/pdf/2603.02960v2
+
+## 8170. Layer-wise QUBO-Based Training of CNN Classifiers for Quantum Annealing
+
+- **ID:** `2603.02958` · **primary:** `quant-ph` · **Submitted:** 2026-03-03
+- **Authors:** Mostafa Atallah, Rebekah Herrman
+- **abs:** https://arxiv.org/abs/2603.02958 · **PDF:** https://arxiv.org/pdf/2603.02958v1
+
+## 8171. Leveraging Label Proportion Prior for Class-Imbalanced Semi-Supervised Learning
+
+- **ID:** `2603.02957` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Kohki Akiba, Shinnosuke Matsuo, Shota Harada, et al.
+- **abs:** https://arxiv.org/abs/2603.02957 · **PDF:** https://arxiv.org/pdf/2603.02957v1
+
+## 8172. Sparse autoencoders reveal organized biological knowledge but minimal regulatory logic in single-cell foundation models: a comparative atlas of Geneformer and scGPT
+
+- **ID:** `2603.02952` · **primary:** `q-bio.GN` · **Submitted:** 2026-03-03
+- **Authors:** Ihor Kendiukhov
+- **abs:** https://arxiv.org/abs/2603.02952 · **PDF:** https://arxiv.org/pdf/2603.02952v1
+
+## 8173. CGL: Advancing Continual GUI Learning via Reinforcement Fine-Tuning
+
+- **ID:** `2603.02951` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Zhenquan Yao, Zitong Huang, Yihan Zeng, et al.
+- **abs:** https://arxiv.org/abs/2603.02951 · **PDF:** https://arxiv.org/pdf/2603.02951v2
+
+## 8174. The Geometry of Learning Under AI Delegation
+
+- **ID:** `2603.02950` · **primary:** `cs.CY` · **Submitted:** 2026-03-03
+- **Authors:** Lingxiao Huang, Nisheeth K. Vishnoi
+- **abs:** https://arxiv.org/abs/2603.02950 · **PDF:** https://arxiv.org/pdf/2603.02950v1
+
+## 8175. SEALing the Gap: A Reference Framework for LLM Inference Carbon Estimation via Multi-Benchmark Driven Embodiment
+
+- **ID:** `2603.02949` · **primary:** `cs.SE` · **Submitted:** 2026-03-03
+- **Authors:** Priyavanshi Pathania, Rohit Mehra, Vibhu Saujanya Sharma, et al.
+- **abs:** https://arxiv.org/abs/2603.02949 · **PDF:** https://arxiv.org/pdf/2603.02949v1
+
+## 8176. Enhancing Physics-Informed Neural Networks with Domain-aware Fourier Features: Towards Improved Performance and Interpretable Results
+
+- **ID:** `2603.02948` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Alberto Miño Calero, Luis Salamanca, Konstantinos E. Tatsis
+- **abs:** https://arxiv.org/abs/2603.02948 · **PDF:** https://arxiv.org/pdf/2603.02948v1
+
+## 8177. ACE-Merging: Data-Free Model Merging with Adaptive Covariance Estimation
+
+- **ID:** `2603.02945` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Bo Xu, Haotian Wu, Hehai Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.02945 · **PDF:** https://arxiv.org/pdf/2603.02945v1
+
+## 8178. ShipTraj-R1: Reinforcing Ship Trajectory Prediction in Large Language Models via Group Relative Policy Optimization
+
+- **ID:** `2603.02939` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Yang Zhan, Yunhao Li, Zhang Chao, et al.
+- **abs:** https://arxiv.org/abs/2603.02939 · **PDF:** https://arxiv.org/pdf/2603.02939v1
+
+## 8179. Beyond One-Size-Fits-All: Adaptive Subgraph Denoising for Zero-Shot Graph Learning with Large Language Models
+
+- **ID:** `2603.02938` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Fengzhi Li, Liang Zhang, Yuan Zuo, et al.
+- **abs:** https://arxiv.org/abs/2603.02938 · **PDF:** https://arxiv.org/pdf/2603.02938v1
+
+## 8180. Bias and Fairness in Self-Supervised Acoustic Representations for Cognitive Impairment Detection
+
+- **ID:** `2603.02937` · **primary:** `eess.AS` · **Submitted:** 2026-03-03
+- **Authors:** Kashaf Gulzar, Korbinian Riedhammer, Elmar Nöth, et al.
+- **abs:** https://arxiv.org/abs/2603.02937 · **PDF:** https://arxiv.org/pdf/2603.02937v1
+
+## 8181. Contextual Latent World Models for Offline Meta Reinforcement Learning
+
+- **ID:** `2603.02935` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Mohammadreza Nakheai, Aidan Scannell, Kevin Luck, et al.
+- **abs:** https://arxiv.org/abs/2603.02935 · **PDF:** https://arxiv.org/pdf/2603.02935v1
+
+## 8182. On the Structural Limitations of Weight-Based Neural Adaptation and the Role of Reversible Behavioral Learning
+
+- **ID:** `2603.02934` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Pardhu Sri Rushi Varma Konduru
+- **abs:** https://arxiv.org/abs/2603.02934 · **PDF:** https://arxiv.org/pdf/2603.02934v1
+
+## 8183. Zero-Knowledge Federated Learning with Lattice-Based Hybrid Encryption for Quantum-Resilient Medical AI
+
+- **ID:** `2603.03398` · **primary:** `cs.CR` · **Submitted:** 2026-03-03
+- **Authors:** Edouard Lansiaux
+- **abs:** https://arxiv.org/abs/2603.03398 · **PDF:** https://arxiv.org/pdf/2603.03398v1
+
+## 8184. GNN For Muon Particle Momentum estimation
+
+- **ID:** `2603.06675` · **primary:** `physics.data-an` · **Submitted:** 2026-03-03
+- **Authors:** Vishak K Bhat, Eric A. F. Reinhardt, Sergei Gleyzer
+- **abs:** https://arxiv.org/abs/2603.06675 · **PDF:** https://arxiv.org/pdf/2603.06675v1
+
+## 8185. Interpretable Motion-Attentive Maps: Spatio-Temporally Localizing Concepts in Video Diffusion Transformers
+
+- **ID:** `2603.02919` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Youngjun Jun, Seil Kang, Woojung Han, et al.
+- **abs:** https://arxiv.org/abs/2603.02919 · **PDF:** https://arxiv.org/pdf/2603.02919v2
+
+## 8186. Eliciting Numerical Predictive Distributions of LLMs Without Autoregression
+
+- **ID:** `2603.02913` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Julianna Piskorz, Katarzyna Kobalczyk, Mihaela van der Schaar
+- **abs:** https://arxiv.org/abs/2603.02913 · **PDF:** https://arxiv.org/pdf/2603.02913v1
+
+## 8187. Learning to Generate and Extract: A Multi-Agent Collaboration Framework For Zero-shot Document-level Event Arguments Extraction
+
+- **ID:** `2603.02909` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Guangjun Zhang, Hu Zhang, Yazhou Han, et al.
+- **abs:** https://arxiv.org/abs/2603.02909 · **PDF:** https://arxiv.org/pdf/2603.02909v2
+
+## 8188. SAE as a Crystal Ball: Interpretable Features Predict Cross-domain Transferability of LLMs without Training
+
+- **ID:** `2603.02908` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Qi Zhang, Yifei Wang, Xiaohan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02908 · **PDF:** https://arxiv.org/pdf/2603.02908v1
+
+## 8189. Towards Accurate and Interpretable Time-series Forecasting: A Polynomial Learning Approach
+
+- **ID:** `2603.02906` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Bo Liu, Shao-Bo Lin, Changmiao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02906 · **PDF:** https://arxiv.org/pdf/2603.02906v1
+
+## 8190. Distributed Dynamic Invariant Causal Prediction in Environmental Time Series
+
+- **ID:** `2603.02902` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Ziruo Hao, Tao Yang, Xiaofeng Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.02902 · **PDF:** https://arxiv.org/pdf/2603.02902v1
+
+## 8191. Embedding interpretable $\ell_1$-regression into neural networks for uncovering temporal structure in cell imaging
+
+- **ID:** `2603.02899` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Fabian Kabus, Maren Hackenberg, Julia Hindel, et al.
+- **abs:** https://arxiv.org/abs/2603.02899 · **PDF:** https://arxiv.org/pdf/2603.02899v2
+
+## 8192. AutoFigure-Edit: Generating Editable Scientific Illustration
+
+- **ID:** `2603.06674` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Zhen Lin, Qiujie Xie, Minjun Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.06674 · **PDF:** https://arxiv.org/pdf/2603.06674v1
+
+## 8193. Benchmarking Compact VLMs for Clip-Level Surveillance Anomaly Detection Under Weak Supervision
+
+- **ID:** `2603.13306` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Kirill Borodin, Kirill Kondrashov, Nikita Vasiliev, et al.
+- **abs:** https://arxiv.org/abs/2603.13306 · **PDF:** https://arxiv.org/pdf/2603.13306v1
+
+## 8194. StegaFFD: Privacy-Preserving Face Forgery Detection via Fine-Grained Steganographic Domain Lifting
+
+- **ID:** `2603.02886` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Guoqing Ma, Xun Lin, Hui Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.02886 · **PDF:** https://arxiv.org/pdf/2603.02886v1
+
+## 8195. Eval4Sim: An Evaluation Framework for Persona Simulation
+
+- **ID:** `2603.02876` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Eliseo Bao, Anxo Perez, Xi Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02876 · **PDF:** https://arxiv.org/pdf/2603.02876v1
+
+## 8196. Retrievit: In-context Retrieval Capabilities of Transformers, State Space Models, and Hybrid Architectures
+
+- **ID:** `2603.02874` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Georgios Pantazopoulos, Malvina Nikandrou, Ioannis Konstas, et al.
+- **abs:** https://arxiv.org/abs/2603.02874 · **PDF:** https://arxiv.org/pdf/2603.02874v1
+
+## 8197. LaTeX Compilation: Challenges in the Era of LLMs
+
+- **ID:** `2603.02873` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Tianyou Liu, Ziqiang Li, Xurui Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.02873 · **PDF:** https://arxiv.org/pdf/2603.02873v4
+
+## 8198. Nodes Are Early, Edges Are Late: Probing Diagram Representations in Large Vision-Language Models
+
+- **ID:** `2603.02865` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Haruto Yoshida, Keito Kudo, Yoichi Aoki, et al.
+- **abs:** https://arxiv.org/abs/2603.02865 · **PDF:** https://arxiv.org/pdf/2603.02865v1
+
+## 8199. Learning in Markov Decision Processes with Exogenous Dynamics
+
+- **ID:** `2603.02862` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Davide Maran, Davide Salaorni, Marcello Restelli
+- **abs:** https://arxiv.org/abs/2603.02862 · **PDF:** https://arxiv.org/pdf/2603.02862v2
+
+## 8200. The Distribution of Phoneme Frequencies across the World's Languages: Macroscopic and Microscopic Information-Theoretic Models
+
+- **ID:** `2603.02860` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Fermín Moscoso del Prado Martín, Suchir Salhan
+- **abs:** https://arxiv.org/abs/2603.02860 · **PDF:** https://arxiv.org/pdf/2603.02860v1
+
+## 8201. LLM-based Argument Mining meets Argumentation and Description Logics: a Unified Framework for Reasoning about Debates
+
+- **ID:** `2603.02858` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Gianvincenzo Alfano, Sergio Greco, Lucio La Cava, et al.
+- **abs:** https://arxiv.org/abs/2603.02858 · **PDF:** https://arxiv.org/pdf/2603.02858v1
+
+## 8202. CoFL: Continuous Flow Fields for Language-Conditioned Navigation
+
+- **ID:** `2603.02854` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** Haokun Liu, Zhaoqi Ma, Yicheng Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.02854 · **PDF:** https://arxiv.org/pdf/2603.02854v1
+
+## 8203. Learning Memory-Enhanced Improvement Heuristics for Flexible Job Shop Scheduling
+
+- **ID:** `2603.02846` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Jiaqi Wang, Zhiguang Cao, Peng Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.02846 · **PDF:** https://arxiv.org/pdf/2603.02846v1
+
+## 8204. SPARC: Spatial-Aware Path Planning via Attentive Robot Communication
+
+- **ID:** `2603.02845` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** Sayang Mu, Xiangyu Wu, Bo An
+- **abs:** https://arxiv.org/abs/2603.02845 · **PDF:** https://arxiv.org/pdf/2603.02845v2
+
+## 8205. Scale-invariant Gaussian derivative residual networks
+
+- **ID:** `2603.02843` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Andrzej Perzanowski, Tony Lindeberg
+- **abs:** https://arxiv.org/abs/2603.02843 · **PDF:** https://arxiv.org/pdf/2603.02843v1
+
+## 8206. A Browser-based Open Source Assistant for Multimodal Content Verification
+
+- **ID:** `2603.02842` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Rosanna Milner, Michael Foster, Twin Karmakharm, et al.
+- **abs:** https://arxiv.org/abs/2603.02842 · **PDF:** https://arxiv.org/pdf/2603.02842v3
+
+## 8207. Adapting Time Series Foundation Models through Data Mixtures
+
+- **ID:** `2603.02840` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Thomas L. Lee, Edoardo M. Ponti, Amos Storkey
+- **abs:** https://arxiv.org/abs/2603.02840 · **PDF:** https://arxiv.org/pdf/2603.02840v1
+
+## 8208. MAD-SmaAt-GNet: A Multimodal Advection-Guided Neural Network for Precipitation Nowcasting
+
+- **ID:** `2603.04461` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Samuel van Wonderen, Siamak Mehrkanoon
+- **abs:** https://arxiv.org/abs/2603.04461 · **PDF:** https://arxiv.org/pdf/2603.04461v1
+
+## 8209. Faster, Cheaper, More Accurate: Specialised Knowledge Tracing Models Outperform LLMs
+
+- **ID:** `2603.02830` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Prarthana Bhattacharyya, Joshua Mitton, Ralph Abboud, et al.
+- **abs:** https://arxiv.org/abs/2603.02830 · **PDF:** https://arxiv.org/pdf/2603.02830v1
+
+## 8210. Toward Early Quality Assessment of Text-to-Image Diffusion Models
+
+- **ID:** `2603.02829` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Huanlei Guo, Hongxin Wei, Bingyi Jing
+- **abs:** https://arxiv.org/abs/2603.02829 · **PDF:** https://arxiv.org/pdf/2603.02829v2
+
+## 8211. Expected Reward Prediction, with Applications to Model Routing
+
+- **ID:** `2603.20217` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Kenan Hasanaliyev, Silas Alberti, Jenny Hamer, et al.
+- **abs:** https://arxiv.org/abs/2603.20217 · **PDF:** https://arxiv.org/pdf/2603.20217v1
+
+## 8212. BrandFusion: A Multi-Agent Framework for Seamless Brand Integration in Text-to-Video Generation
+
+- **ID:** `2603.02816` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Zihao Zhu, Ruotong Wang, Siwei Lyu, et al.
+- **abs:** https://arxiv.org/abs/2603.02816 · **PDF:** https://arxiv.org/pdf/2603.02816v2
+
+## 8213. MERIT: Memory-Enhanced Retrieval for Interpretable Knowledge Tracing
+
+- **ID:** `2603.22289` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Runze Li, Kedi Chen, Guwei Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.22289 · **PDF:** https://arxiv.org/pdf/2603.22289v1
+
+## 8214. Locally Coherent Parallel Decoding in Diffusion Language Models
+
+- **ID:** `2603.20216` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Michael Hersche, Nicolas Menet, Ronan Tanios, et al.
+- **abs:** https://arxiv.org/abs/2603.20216 · **PDF:** https://arxiv.org/pdf/2603.20216v1
+
+## 8215. ChemFlow:A Hierarchical Neural Network for Multiscale Representation Learning in Chemical Mixtures
+
+- **ID:** `2603.02810` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-03
+- **Authors:** Jinming Fan, Chao Qian, Wilhelm T. S. Huck, et al.
+- **abs:** https://arxiv.org/abs/2603.02810 · **PDF:** https://arxiv.org/pdf/2603.02810v1
+
+## 8216. Lattice-based Deep Neural Networks: Regularity and Tailored Regularization
+
+- **ID:** `2603.02809` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Alexander Keller, Frances Y. Kuo, Dirk Nuyens, et al.
+- **abs:** https://arxiv.org/abs/2603.02809 · **PDF:** https://arxiv.org/pdf/2603.02809v1
+
+## 8217. Unmixing microinfrared spectroscopic images of cross-sections of historical oil paintings
+
+- **ID:** `2603.06673` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Shivam Pande, Nicolas Nadisic, Francisco Mederos-Henry, et al.
+- **abs:** https://arxiv.org/abs/2603.06673 · **PDF:** https://arxiv.org/pdf/2603.06673v1
+
+## 8218. The Price of Robustness: Stable Classifiers Need Overparameterization
+
+- **ID:** `2603.02806` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Jonas von Berg, Adalbert Fono, Massimiliano Datres, et al.
+- **abs:** https://arxiv.org/abs/2603.02806 · **PDF:** https://arxiv.org/pdf/2603.02806v1
+
+## 8219. Guideline-Grounded Evidence Accumulation for High-Stakes Agent Verification
+
+- **ID:** `2603.02798` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Yichi Zhang, Nabeel Seedat, Yinpeng Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.02798 · **PDF:** https://arxiv.org/pdf/2603.02798v1
+
+## 8220. Differentiable Time-Varying IIR Filtering for Real-Time Speech Denoising
+
+- **ID:** `2603.02794` · **primary:** `cs.SD` · **Submitted:** 2026-03-03
+- **Authors:** Riccardo Rota, Kiril Ratmanski, Jozef Coldenhoff, et al.
+- **abs:** https://arxiv.org/abs/2603.02794 · **PDF:** https://arxiv.org/pdf/2603.02794v1
+
+## 8221. From Heuristic Selection to Automated Algorithm Design: LLMs Benefit from Strong Priors
+
+- **ID:** `2603.02792` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Qi Huang, Furong Ye, Ananta Shahane, et al.
+- **abs:** https://arxiv.org/abs/2603.02792 · **PDF:** https://arxiv.org/pdf/2603.02792v1
+
+## 8222. OCR or Not? Rethinking Document Information Extraction in the MLLMs Era with Real-World Large-Scale Datasets
+
+- **ID:** `2603.02789` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Jiyuan Shen, Peiyue Yuan, Atin Ghosh, et al.
+- **abs:** https://arxiv.org/abs/2603.02789 · **PDF:** https://arxiv.org/pdf/2603.02789v1
+
+## 8223. Agentified Assessment of Logical Reasoning Agents
+
+- **ID:** `2603.02788` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Zhiyu Ni, Yifeng Xiao, Zheng Liang
+- **abs:** https://arxiv.org/abs/2603.02788 · **PDF:** https://arxiv.org/pdf/2603.02788v4
+
+## 8224. Rethinking Code Similarity for Automated Algorithm Design with LLMs
+
+- **ID:** `2603.02787` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Rui Zhang, Zhichao Lu
+- **abs:** https://arxiv.org/abs/2603.02787 · **PDF:** https://arxiv.org/pdf/2603.02787v1
+
+## 8225. VSPrefill: Vertical-Slash Sparse Attention with Lightweight Indexing for Long-Context Prefilling
+
+- **ID:** `2603.04460` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Chen Guanzhong
+- **abs:** https://arxiv.org/abs/2603.04460 · **PDF:** https://arxiv.org/pdf/2603.04460v1
+
+## 8226. Generative adversarial imitation learning for robot swarms: Learning from human demonstrations and trained policies
+
+- **ID:** `2603.02783` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** Mattes Kraus, Jonas Kuckling
+- **abs:** https://arxiv.org/abs/2603.02783 · **PDF:** https://arxiv.org/pdf/2603.02783v1
+
+## 8227. Scores Know Bobs Voice: Speaker Impersonation Attack
+
+- **ID:** `2603.02781` · **primary:** `cs.CR` · **Submitted:** 2026-03-03
+- **Authors:** Chanwoo Hwang, Sunpill Kim, Yong Kiam Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.02781 · **PDF:** https://arxiv.org/pdf/2603.02781v1
+
+## 8228. From Solver to Tutor: Evaluating the Pedagogical Intelligence of LLMs with KMP-Bench
+
+- **ID:** `2603.02775` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Weikang Shi, Houxing Ren, Junting Pan, et al.
+- **abs:** https://arxiv.org/abs/2603.02775 · **PDF:** https://arxiv.org/pdf/2603.02775v1
+
+## 8229. Benchmark of Benchmarks: Unpacking Influence and Code Repository Quality in LLM Safety Benchmarks
+
+- **ID:** `2603.04459` · **primary:** `cs.CR` · **Submitted:** 2026-03-03
+- **Authors:** Junjie Chu, Xinyue Shen, Ye Leng, et al.
+- **abs:** https://arxiv.org/abs/2603.04459 · **PDF:** https://arxiv.org/pdf/2603.04459v2
+
+## 8230. ITO: Images and Texts as One via Synergizing Multiple Alignment and Training-Time Fusion
+
+- **ID:** `2603.02767` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Hanpeng Liu, Yaqian Li, Zidan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02767 · **PDF:** https://arxiv.org/pdf/2603.02767v3
+
+## 8231. EvoSkill: Automated Skill Discovery for Multi-Agent Systems
+
+- **ID:** `2603.02766` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Salaheddin Alzubi, Noah Provenzano, Jaydon Bingham, et al.
+- **abs:** https://arxiv.org/abs/2603.02766 · **PDF:** https://arxiv.org/pdf/2603.02766v1
+
+## 8232. Next Embedding Prediction Makes World Models Stronger
+
+- **ID:** `2603.02765` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** George Bredis, Nikita Balagansky, Daniil Gavrilov, et al.
+- **abs:** https://arxiv.org/abs/2603.02765 · **PDF:** https://arxiv.org/pdf/2603.02765v1
+
+## 8233. Multi-Agent-Based Simulation of Archaeological Mobility in Uneven Landscapes
+
+- **ID:** `2603.03390` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** Chairi Kiourt, Vassilis Evangelidis, Dimitris Grigoropoulos
+- **abs:** https://arxiv.org/abs/2603.03390 · **PDF:** https://arxiv.org/pdf/2603.03390v1
+
+## 8234. Towards Improved Sentence Representations using Token Graphs
+
+- **ID:** `2603.03389` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Krishna Sri Ipsit Mantri, Carola-Bibiane Schönlieb, Zorah Lähner, et al.
+- **abs:** https://arxiv.org/abs/2603.03389 · **PDF:** https://arxiv.org/pdf/2603.03389v1
+
+## 8235. Efficient Self-Evaluation for Diffusion Language Models via Sequence Regeneration
+
+- **ID:** `2603.02760` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Linhao Zhong, Linyu Wu, Wen Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02760 · **PDF:** https://arxiv.org/pdf/2603.02760v1
+
+## 8236. EEG-SeeGraph: Interpreting functional connectivity disruptions in dementias via sparse-explanatory dynamic EEG-graph learning
+
+- **ID:** `2603.16895` · **primary:** `eess.SP` · **Submitted:** 2026-03-03
+- **Authors:** Fengcheng Wu, Zhenxi Song, Guoyang Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.16895 · **PDF:** https://arxiv.org/pdf/2603.16895v1
+
+## 8237. Rethinking Time Series Domain Generalization via Structure-Stratified Calibration
+
+- **ID:** `2603.02756` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Jinyang Li, Shuhao Mei, Xiaoyu Xiao, et al.
+- **abs:** https://arxiv.org/abs/2603.02756 · **PDF:** https://arxiv.org/pdf/2603.02756v1
+
+## 8238. Deep learning-guided evolutionary optimization for protein design
+
+- **ID:** `2603.02753` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Erik Hartman, Di Tang, Johan Malmström
+- **abs:** https://arxiv.org/abs/2603.02753 · **PDF:** https://arxiv.org/pdf/2603.02753v1
+
+## 8239. iGVLM: Dynamic Instruction-Guided Vision Encoding for Question-Aware Multimodal Understanding
+
+- **ID:** `2603.02748` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Hanpeng Liu, Yaqian Li, Zidan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02748 · **PDF:** https://arxiv.org/pdf/2603.02748v2
+
+## 8240. Enhancing User Throughput in Multi-panel mmWave Radio Access Networks for Beam-based MU-MIMO Using a DRL Method
+
+- **ID:** `2603.02745` · **primary:** `cs.IT` · **Submitted:** 2026-03-03
+- **Authors:** Ramin Hashemi, Vismika Ranasinghe, Teemu Veijalainen, et al.
+- **abs:** https://arxiv.org/abs/2603.02745 · **PDF:** https://arxiv.org/pdf/2603.02745v1
+
+## 8241. Practical FP4 Training for Large-Scale MoE Models on Hopper GPUs
+
+- **ID:** `2603.02731` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Wuyue Zhang, Chongdong Huang, Chunbo You, et al.
+- **abs:** https://arxiv.org/abs/2603.02731 · **PDF:** https://arxiv.org/pdf/2603.02731v1
+
+## 8242. The power of small initialization in noisy low-tubal-rank tensor recovery
+
+- **ID:** `2603.02729` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** ZHiyu Liu, Haobo Geng, Xudong Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02729 · **PDF:** https://arxiv.org/pdf/2603.02729v1
+
+## 8243. Single Microphone Own Voice Detection based on Simulated Transfer Functions for Hearing Aids
+
+- **ID:** `2603.02724` · **primary:** `cs.SD` · **Submitted:** 2026-03-03
+- **Authors:** Mathuranathan Mayuravaani, W. Bastiaan Kleijn, Andrew Lensen, et al.
+- **abs:** https://arxiv.org/abs/2603.02724 · **PDF:** https://arxiv.org/pdf/2603.02724v1
+
+## 8244. An Empirical Analysis of Calibration and Selective Prediction in Multimodal Clinical Condition Classification
+
+- **ID:** `2603.02719` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** L. Julián Lechuga López, Farah E. Shamout, Tim G. J. Rudner
+- **abs:** https://arxiv.org/abs/2603.02719 · **PDF:** https://arxiv.org/pdf/2603.02719v1
+
+## 8245. RADAR: Learning to Route with Asymmetry-aware DistAnce Representations
+
+- **ID:** `2603.03388` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Hang Yi, Ziwei Huang, Yining Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.03388 · **PDF:** https://arxiv.org/pdf/2603.03388v2
+
+## 8246. Learning Unified Distance Metric for Heterogeneous Attribute Data Clustering
+
+- **ID:** `2603.04458` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Yiqun Zhang, Mingjie Zhao, Yizhou Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.04458 · **PDF:** https://arxiv.org/pdf/2603.04458v1
+
+## 8247. A Natural Language Agentic Approach to Study Affective Polarization
+
+- **ID:** `2603.02711` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Stephanie Anneris Malvicini, Ewelina Gajewska, Arda Derbent, et al.
+- **abs:** https://arxiv.org/abs/2603.02711 · **PDF:** https://arxiv.org/pdf/2603.02711v1
+
+## 8248. Sensory-Aware Sequential Recommendation via Review-Distilled Representations
+
+- **ID:** `2603.02709` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Yeo Chan Yoon
+- **abs:** https://arxiv.org/abs/2603.02709 · **PDF:** https://arxiv.org/pdf/2603.02709v1
+
+## 8249. Learning Order Forest for Qualitative-Attribute Data Clustering
+
+- **ID:** `2603.03387` · **primary:** `stat.ML` · **Submitted:** 2026-03-03
+- **Authors:** Mingjie Zhao, Sen Feng, Yiqun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03387 · **PDF:** https://arxiv.org/pdf/2603.03387v1
+
+## 8250. Intelligent Pathological Diagnosis of Gestational Trophoblastic Diseases via Visual-Language Deep Learning Model
+
+- **ID:** `2603.02704` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Yuhang Liu, Yueyang Cang, Wenge Que, et al.
+- **abs:** https://arxiv.org/abs/2603.02704 · **PDF:** https://arxiv.org/pdf/2603.02704v1
+
+## 8251. FinTexTS: Financial Text-Paired Time-Series Dataset via Semantic-Based and Multi-Level Pairing
+
+- **ID:** `2603.02702` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Jaehoon Lee, Suhwan Park, Tae Yoon Lim, et al.
+- **abs:** https://arxiv.org/abs/2603.02702 · **PDF:** https://arxiv.org/pdf/2603.02702v2
+
+## 8252. Graph-GRPO: Stabilizing Multi-Agent Topology Learning via Group Relative Policy Optimization
+
+- **ID:** `2603.02701` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Yueyang Cang, Xiaoteng Zhang, Erlu Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.02701 · **PDF:** https://arxiv.org/pdf/2603.02701v1
+
+## 8253. Neural quantum support vector data description for one-class classification
+
+- **ID:** `2603.02700` · **primary:** `quant-ph` · **Submitted:** 2026-03-03
+- **Authors:** Changjae Im, Hyeondo Oh, Daniel K. Park
+- **abs:** https://arxiv.org/abs/2603.02700 · **PDF:** https://arxiv.org/pdf/2603.02700v1
+
+## 8254. ShareVerse: Multi-Agent Consistent Video Generation for Shared World Modeling
+
+- **ID:** `2603.02697` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Jiayi Zhu, Jianing Zhang, Yiying Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.02697 · **PDF:** https://arxiv.org/pdf/2603.02697v1
+
+## 8255. Addressing Missing and Noisy Modalities in One Solution: Unified Modality-Quality Framework for Low-quality Multimodal Data
+
+- **ID:** `2603.02695` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Sijie Mai, Shiqin Han, Haifeng Hu
+- **abs:** https://arxiv.org/abs/2603.02695 · **PDF:** https://arxiv.org/pdf/2603.02695v1
+
+## 8256. Does Semantic Noise Initialization Transfer from Images to Videos? A Paired Diagnostic Study
+
+- **ID:** `2603.06672` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Yixiao Jing, Chaoyu Zhang, Zixuan Zhong, et al.
+- **abs:** https://arxiv.org/abs/2603.06672 · **PDF:** https://arxiv.org/pdf/2603.06672v1
+
+## 8257. Retrieval-Augmented Robots via Retrieve-Reason-Act
+
+- **ID:** `2603.02688` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Izat Temiraliev, Diji Yang, Yi Zhang
+- **abs:** https://arxiv.org/abs/2603.02688 · **PDF:** https://arxiv.org/pdf/2603.02688v1
+
+## 8258. HateMirage: An Explainable Multi-Dimensional Dataset for Decoding Faux Hate and Subtle Online Abuse
+
+- **ID:** `2603.02684` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Sai Kartheek Reddy Kasu, Shankar Biradar, Sunil Saumya, et al.
+- **abs:** https://arxiv.org/abs/2603.02684 · **PDF:** https://arxiv.org/pdf/2603.02684v1
+
+## 8259. LLMs for High-Frequency Decision-Making: Normalized Action Reward-Guided Consistency Policy Optimization
+
+- **ID:** `2603.02680` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Yang Zhao, Zihao Li, Zhiyu Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.02680 · **PDF:** https://arxiv.org/pdf/2603.02680v1
+
+## 8260. Causal Learning Should Embrace the Wisdom of the Crowd
+
+- **ID:** `2603.02678` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Ryan Feng Lin, Yuantao Wei, Huiling Liao, et al.
+- **abs:** https://arxiv.org/abs/2603.02678 · **PDF:** https://arxiv.org/pdf/2603.02678v2
+
+## 8261. Multi-Agent Debate with Memory Masking
+
+- **ID:** `2603.20215` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Hongduan Tian, Xiao Feng, Ziyuan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.20215 · **PDF:** https://arxiv.org/pdf/2603.20215v1
+
+## 8262. ITLC at SemEval-2026 Task 11: Normalization and Deterministic Parsing for Formal Reasoning in LLMs
+
+- **ID:** `2603.02676` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Wicaksono Leksono Muhamad, Joanito Agili Lopo, Tack Hwa Wong, et al.
+- **abs:** https://arxiv.org/abs/2603.02676 · **PDF:** https://arxiv.org/pdf/2603.02676v1
+
+## 8263. From Shallow to Deep: Pinning Semantic Intent via Causal GRPO
+
+- **ID:** `2603.02675` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Shuyi Zhou, Zeen Song, Wenwen Qiang, et al.
+- **abs:** https://arxiv.org/abs/2603.02675 · **PDF:** https://arxiv.org/pdf/2603.02675v1
+
+## 8264. Exact Functional ANOVA Decomposition for Categorical Inputs Models
+
+- **ID:** `2603.02673` · **primary:** `stat.ML` · **Submitted:** 2026-03-03
+- **Authors:** Baptiste Ferrere, Nicolas Bousquet, Fabrice Gamboa, et al.
+- **abs:** https://arxiv.org/abs/2603.02673 · **PDF:** https://arxiv.org/pdf/2603.02673v1
+
+## 8265. SorryDB: Can AI Provers Complete Real-World Lean Theorems?
+
+- **ID:** `2603.02668` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Austin Letson, Leopoldo Sarra, Auguste Poiroux, et al.
+- **abs:** https://arxiv.org/abs/2603.02668 · **PDF:** https://arxiv.org/pdf/2603.02668v1
+
+## 8266. DREAM: Where Visual Understanding Meets Text-to-Image Generation
+
+- **ID:** `2603.02667` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Chao Li, Tianhong Li, Sai Vidyaranya Nuthalapati, et al.
+- **abs:** https://arxiv.org/abs/2603.02667 · **PDF:** https://arxiv.org/pdf/2603.02667v1
+
+## 8267. Evaluating Cross-Modal Reasoning Ability and Problem Characteristics with Multimodal Item Response Theory
+
+- **ID:** `2603.02663` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Shunki Uebayashi, Kento Masui, Kyohei Atarashi, et al.
+- **abs:** https://arxiv.org/abs/2603.02663 · **PDF:** https://arxiv.org/pdf/2603.02663v1
+
+## 8268. Real-Time Generation of Game Video Commentary with Multimodal LLMs: Pause-Aware Decoding Approaches
+
+- **ID:** `2603.02655` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Anum Afzal, Yuki Saito, Hiroya Takamura, et al.
+- **abs:** https://arxiv.org/abs/2603.02655 · **PDF:** https://arxiv.org/pdf/2603.02655v1
+
+## 8269. AlphaFree: Recommendation Free from Users, IDs, and GNNs
+
+- **ID:** `2603.02653` · **primary:** `cs.IR` · **Submitted:** 2026-03-03
+- **Authors:** Minseo Jeon, Junwoo Jung, Daewon Gwak, et al.
+- **abs:** https://arxiv.org/abs/2603.02653 · **PDF:** https://arxiv.org/pdf/2603.02653v1
+
+## 8270. Improving Diffusion Planners by Self-Supervised Action Gating with Energies
+
+- **ID:** `2603.02650` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Yuan Lu, Dongqi Han, Yansen Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02650 · **PDF:** https://arxiv.org/pdf/2603.02650v1
+
+## 8271. HomeAdam: Adam and AdamW Algorithms Sometimes Go Home to Obtain Better Provable Generalization
+
+- **ID:** `2603.02649` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Feihu Huang, Guanyi Zhang, Songcan Chen
+- **abs:** https://arxiv.org/abs/2603.02649 · **PDF:** https://arxiv.org/pdf/2603.02649v1
+
+## 8272. Credibility Governance: A Social Mechanism for Collective Self-Correction under Weak Truth Signals
+
+- **ID:** `2603.02640` · **primary:** `cs.CY` · **Submitted:** 2026-03-03
+- **Authors:** Wanying He, Yanxi Lin, Ziheng Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.02640 · **PDF:** https://arxiv.org/pdf/2603.02640v1
+
+## 8273. Convex and Non-convex Federated Learning with Stale Stochastic Gradients: Diminishing Step Size is All You Need
+
+- **ID:** `2603.02639` · **primary:** `math.OC` · **Submitted:** 2026-03-03
+- **Authors:** Xinran Zheng, Tara Javidi, Behrouz Touri
+- **abs:** https://arxiv.org/abs/2603.02639 · **PDF:** https://arxiv.org/pdf/2603.02639v1
+
+## 8274. The Vienna 4G/5G Drive-Test Dataset
+
+- **ID:** `2603.02638` · **primary:** `eess.SP` · **Submitted:** 2026-03-03
+- **Authors:** Wilfried Wiedner, Lukas Eller, Mariam Mussbah, et al.
+- **abs:** https://arxiv.org/abs/2603.02638 · **PDF:** https://arxiv.org/pdf/2603.02638v1
+
+## 8275. StitchCUDA: An Automated Multi-Agents End-to-End GPU Programing Framework with Rubric-based Agentic Reinforcement Learning
+
+- **ID:** `2603.02637` · **primary:** `cs.MA` · **Submitted:** 2026-03-03
+- **Authors:** Shiyang Li, Zijian Zhang, Winson Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.02637 · **PDF:** https://arxiv.org/pdf/2603.02637v1
+
+## 8276. SaFeR-ToolKit: Structured Reasoning via Virtual Tool Calling for Multimodal Safety
+
+- **ID:** `2603.02635` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Zixuan Xu, Tiancheng He, Huahui Yi, et al.
+- **abs:** https://arxiv.org/abs/2603.02635 · **PDF:** https://arxiv.org/pdf/2603.02635v1
+
+## 8277. Robust Heterogeneous Analog-Digital Computing for Mixture-of-Experts Models with Theoretical Generalization Guarantees
+
+- **ID:** `2603.02633` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Mohammed Nowaz Rabbani Chowdhury, Hsinyu Tsai, Geoffrey W. Burr, et al.
+- **abs:** https://arxiv.org/abs/2603.02633 · **PDF:** https://arxiv.org/pdf/2603.02633v1
+
+## 8278. Cross-Family Speculative Prefill: Training-Free Long-Context Compression with Small Draft Models
+
+- **ID:** `2603.02631` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Shubhangi Upasani, Ravi Shanker Raju, Bo Li, et al.
+- **abs:** https://arxiv.org/abs/2603.02631 · **PDF:** https://arxiv.org/pdf/2603.02631v3
+
+## 8279. MASPOB: Bandit-Based Prompt Optimization for Multi-Agent Systems with Graph Neural Networks
+
+- **ID:** `2603.02630` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Zhi Hong, Qian Zhang, Jiahang Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.02630 · **PDF:** https://arxiv.org/pdf/2603.02630v1
+
+## 8280. Post Hoc Extraction of Pareto Fronts for Continuous Control
+
+- **ID:** `2603.02628` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Raghav Thakar, Gaurav Dixit, Kagan Tumer
+- **abs:** https://arxiv.org/abs/2603.02628 · **PDF:** https://arxiv.org/pdf/2603.02628v1
+
+## 8281. See and Remember: A Multimodal Agent for Web Traversal
+
+- **ID:** `2603.02626` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Xinjun Wang, Shengyao Wang, Aimin Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.02626 · **PDF:** https://arxiv.org/pdf/2603.02626v1
+
+## 8282. Uni-Skill: Building Self-Evolving Skill Repository for Generalizable Robotic Manipulation
+
+- **ID:** `2603.02623` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** Senwei Xie, Yuntian Zhang, Ruiping Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02623 · **PDF:** https://arxiv.org/pdf/2603.02623v1
+
+## 8283. Implicit Bias in Deep Linear Discriminant Analysis
+
+- **ID:** `2603.02622` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Jiawen Li
+- **abs:** https://arxiv.org/abs/2603.02622 · **PDF:** https://arxiv.org/pdf/2603.02622v1
+
+## 8284. Same Error, Different Function: The Optimizer as an Implicit Prior in Financial Time Series
+
+- **ID:** `2603.02620` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Federico Vittorio Cortesi, Giuseppe Iannone, Giulia Crippa, et al.
+- **abs:** https://arxiv.org/abs/2603.02620 · **PDF:** https://arxiv.org/pdf/2603.02620v1
+
+## 8285. Detecting Structural Heart Disease from Electrocardiograms via a Generalized Additive Model of Interpretable Foundation-Model Predictors
+
+- **ID:** `2603.02616` · **primary:** `stat.AP` · **Submitted:** 2026-03-03
+- **Authors:** Ya Zhou, Zhaohong Sun, Tianxiang Hao, et al.
+- **abs:** https://arxiv.org/abs/2603.02616 · **PDF:** https://arxiv.org/pdf/2603.02616v1
+
+## 8286. Think, But Don't Overthink: Reproducing Recursive Language Models
+
+- **ID:** `2603.02615` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Daren Wang
+- **abs:** https://arxiv.org/abs/2603.02615 · **PDF:** https://arxiv.org/pdf/2603.02615v1
+
+## 8287. Real-Time Generative Policy via Langevin-Guided Flow Matching for Autonomous Driving
+
+- **ID:** `2603.02613` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Tianze Zhu, Yinuo Wang, Wenjun Zou, et al.
+- **abs:** https://arxiv.org/abs/2603.02613 · **PDF:** https://arxiv.org/pdf/2603.02613v1
+
+## 8288. Combinatorial Sparse PCA Beyond the Spiked Identity Model
+
+- **ID:** `2603.02607` · **primary:** `stat.ML` · **Submitted:** 2026-03-03
+- **Authors:** Syamantak Kumar, Purnamrita Sarkar, Kevin Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.02607 · **PDF:** https://arxiv.org/pdf/2603.02607v1
+
+## 8289. Heterogeneous Agent Collaborative Reinforcement Learning
+
+- **ID:** `2603.02604` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Zhixia Zhang, Zixuan Huang, Xin Xia, et al.
+- **abs:** https://arxiv.org/abs/2603.02604 · **PDF:** https://arxiv.org/pdf/2603.02604v1
+
+## 8290. AgentAssay: Token-Efficient Regression Testing for Non-Deterministic AI Agent Workflows
+
+- **ID:** `2603.02601` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Varun Pratap Bhardwaj
+- **abs:** https://arxiv.org/abs/2603.02601 · **PDF:** https://arxiv.org/pdf/2603.02601v1
+
+## 8291. SUN: Shared Use of Next-token Prediction for Efficient Multi-LLM Disaggregated Serving
+
+- **ID:** `2603.02599` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Sunghyeon Woo, Ahreum Seo, Jaegwang Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.02599 · **PDF:** https://arxiv.org/pdf/2603.02599v1
+
+## 8292. GPUTOK: GPU Accelerated Byte Level BPE Tokenization
+
+- **ID:** `2603.02597` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Venu Gopal Kadamba, Kanishkha Jaisankar
+- **abs:** https://arxiv.org/abs/2603.02597 · **PDF:** https://arxiv.org/pdf/2603.02597v1
+
+## 8293. Low-Degree Method Fails to Predict Robust Subspace Recovery
+
+- **ID:** `2603.02594` · **primary:** `stat.ML` · **Submitted:** 2026-03-03
+- **Authors:** He Jia, Aravindan Vijayaraghavan
+- **abs:** https://arxiv.org/abs/2603.02594 · **PDF:** https://arxiv.org/pdf/2603.02594v1
+
+## 8294. ERP-RiskBench: Leakage-Safe Ensemble Learning for Financial Risk
+
+- **ID:** `2603.06671` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Sanjay Mishra
+- **abs:** https://arxiv.org/abs/2603.06671 · **PDF:** https://arxiv.org/pdf/2603.06671v1
+
+## 8295. ExpGuard: LLM Content Moderation in Specialized Domains
+
+- **ID:** `2603.02588` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Minseok Choi, Dongjin Kim, Seungbin Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.02588 · **PDF:** https://arxiv.org/pdf/2603.02588v1
+
+## 8296. calibfusion: Transformer-Based Differentiable Calibration for Radar-Camera Fusion Detection in Water-Surface Environments
+
+- **ID:** `2603.06670` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Yuting Wan, Liguo Sun, Jiuwu Hao, et al.
+- **abs:** https://arxiv.org/abs/2603.06670 · **PDF:** https://arxiv.org/pdf/2603.06670v1
+
+## 8297. LiveAgentBench: Comprehensive Benchmarking of Agentic Systems Across 104 Real-World Challenges
+
+- **ID:** `2603.02586` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Hao Li, Huan Wang, Jinjie Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.02586 · **PDF:** https://arxiv.org/pdf/2603.02586v1
+
+## 8298. Hybrid Orchestration of Edge AI and Microservices via Graph-based Self-Imitation Learning
+
+- **ID:** `2603.06669` · **primary:** `cs.NI` · **Submitted:** 2026-03-03
+- **Authors:** Chen Yang, Jin Zheng, Yang Zhuolin, et al.
+- **abs:** https://arxiv.org/abs/2603.06669 · **PDF:** https://arxiv.org/pdf/2603.06669v1
+
+## 8299. Joint Optimization of Model Partitioning and Resource Allocation for Anti-Jamming Collaborative Inference Systems
+
+- **ID:** `2603.02579` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Mengru Wu, Jiawei Li, Jiaqi Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.02579 · **PDF:** https://arxiv.org/pdf/2603.02579v1
+
+## 8300. How Controllable Are Large Language Models? A Unified Evaluation across Behavioral Granularities
+
+- **ID:** `2603.02578` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Ziwen Xu, Kewei Xu, Haoming Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.02578 · **PDF:** https://arxiv.org/pdf/2603.02578v1
+
+## 8301. Towards Parameter-Free Temporal Difference Learning
+
+- **ID:** `2603.02577` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Yunxiang Li, Mark Schmidt, Reza Babanezhad, et al.
+- **abs:** https://arxiv.org/abs/2603.02577 · **PDF:** https://arxiv.org/pdf/2603.02577v1
+
+## 8302. Wasserstein Proximal Policy Gradient
+
+- **ID:** `2603.02576` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Zhaoyu Zhu, Shuhan Zhang, Rui Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.02576 · **PDF:** https://arxiv.org/pdf/2603.02576v1
+
+## 8303. FlashEvaluator: Expanding Search Space with Parallel Evaluation
+
+- **ID:** `2603.02565` · **primary:** `cs.IR` · **Submitted:** 2026-03-03
+- **Authors:** Chao Feng, Yuanhao Pu, Chenghao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.02565 · **PDF:** https://arxiv.org/pdf/2603.02565v1
+
+## 8304. Evidence-based Distributional Alignment for Large Language Models
+
+- **ID:** `2603.13305` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Viet-Thanh Pham, Lizhen Qu, Zhuang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.13305 · **PDF:** https://arxiv.org/pdf/2603.13305v1
+
+## 8305. EdgeFLow: Serverless Federated Learning via Sequential Model Migration in Edge Networks
+
+- **ID:** `2603.02562` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Yuchen Shi, Qijun Hou, Pingyi Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.02562 · **PDF:** https://arxiv.org/pdf/2603.02562v1
+
+## 8306. SOLAR: SVD-Optimized Lifelong Attention for Recommendation
+
+- **ID:** `2603.02561` · **primary:** `cs.IR` · **Submitted:** 2026-03-03
+- **Authors:** Chenghao Zhang, Chao Feng, Yuanhao Pu, et al.
+- **abs:** https://arxiv.org/abs/2603.02561 · **PDF:** https://arxiv.org/pdf/2603.02561v1
+
+## 8307. Evaluating Prompting Strategies for Chart Question Answering with Large Language Models
+
+- **ID:** `2603.22288` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Ruthuparna Naikar, Ying Zhu
+- **abs:** https://arxiv.org/abs/2603.22288 · **PDF:** https://arxiv.org/pdf/2603.22288v1
+
+## 8308. LiteVLA-Edge: Quantized On-Device Multimodal Control for Embedded Robotics
+
+- **ID:** `2603.03380` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** Justin Williams, Kishor Datta Gupta, Roy George, et al.
+- **abs:** https://arxiv.org/abs/2603.03380 · **PDF:** https://arxiv.org/pdf/2603.03380v1
+
+## 8309. CAPT: Confusion-Aware Prompt Tuning for Reducing Vision-Language Misalignment
+
+- **ID:** `2603.02557` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Maoyuan Shao, Yutong Gao, Xinyang Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.02557 · **PDF:** https://arxiv.org/pdf/2603.02557v1
+
+## 8310. Through the Lens of Contrast: Self-Improving Visual Reasoning in VLMs
+
+- **ID:** `2603.02556` · **primary:** `cs.CV` · **Submitted:** 2026-03-03
+- **Authors:** Zhiyu Pan, Yizheng Wu, Jiashen Hua, et al.
+- **abs:** https://arxiv.org/abs/2603.02556 · **PDF:** https://arxiv.org/pdf/2603.02556v1
+
+## 8311. Give me scissors: Collision-Free Dual-Arm Surgical Assistive Robot for Instrument Delivery
+
+- **ID:** `2603.02553` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** Xuejin Luo, Shiquan Sun, Runshi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.02553 · **PDF:** https://arxiv.org/pdf/2603.02553v1
+
+## 8312. CoDAR: Continuous Diffusion Language Models are More Powerful Than You Think
+
+- **ID:** `2603.02547` · **primary:** `cs.CL` · **Submitted:** 2026-03-03
+- **Authors:** Junzhe Shen, Jieru Zhao, Ziwei He, et al.
+- **abs:** https://arxiv.org/abs/2603.02547 · **PDF:** https://arxiv.org/pdf/2603.02547v1
+
+## 8313. AnchorDrive: LLM Scenario Rollout with Anchor-Guided Diffusion Regeneration for Safety-Critical Scenario Generation
+
+- **ID:** `2603.02542` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Zhulin Jiang, Zetao Li, Cheng Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02542 · **PDF:** https://arxiv.org/pdf/2603.02542v1
+
+## 8314. MemSifter: Offloading LLM Memory Retrieval via Outcome-Driven Proxy Reasoning
+
+- **ID:** `2603.03379` · **primary:** `cs.IR` · **Submitted:** 2026-03-03
+- **Authors:** Jiejun Tan, Zhicheng Dou, Liancheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03379 · **PDF:** https://arxiv.org/pdf/2603.03379v1
+
+## 8315. AOI: Turning Failed Trajectories into Training Signals for Autonomous Cloud Diagnosis
+
+- **ID:** `2603.03378` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Pei Yang, Wanyi Chen, Asuka Yuxi Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.03378 · **PDF:** https://arxiv.org/pdf/2603.03378v3
+
+## 8316. Capability Thresholds and Manufacturing Topology: How Embodied Intelligence Triggers Phase Transitions in Economic Geography
+
+- **ID:** `2603.04457` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Xinmin Fang, Lingfeng Tao, Zhengxiong Li
+- **abs:** https://arxiv.org/abs/2603.04457 · **PDF:** https://arxiv.org/pdf/2603.04457v1
+
+## 8317. A Neuropsychologically Grounded Evaluation of LLM Cognitive Abilities
+
+- **ID:** `2603.02540` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Faiz Ghifari Haznitrama, Faeyza Rishad Ardi, Alice Oh
+- **abs:** https://arxiv.org/abs/2603.02540 · **PDF:** https://arxiv.org/pdf/2603.02540v1
+
+## 8318. Functional Properties of the Focal-Entropy
+
+- **ID:** `2603.02533` · **primary:** `cs.IT` · **Submitted:** 2026-03-03
+- **Authors:** Jaimin Shah, Martina Cardone, Alex Dytso
+- **abs:** https://arxiv.org/abs/2603.02533 · **PDF:** https://arxiv.org/pdf/2603.02533v1
+
+## 8319. Bridging Diffusion Guidance and Anderson Acceleration via Hopfield Dynamics
+
+- **ID:** `2603.02531` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Kwanyoung Kim
+- **abs:** https://arxiv.org/abs/2603.02531 · **PDF:** https://arxiv.org/pdf/2603.02531v1
+
+## 8320. LLM-MLFFN: Multi-Level Autonomous Driving Behavior Feature Fusion via Large Language Model
+
+- **ID:** `2603.02528` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Xiangyu Li, Tianyi Wang, Xi Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.02528 · **PDF:** https://arxiv.org/pdf/2603.02528v1
+
+## 8321. Thermodynamic Regulation of Finite-Time Gibbs Training in Energy-Based Models: A Restricted Boltzmann Machine Study
+
+- **ID:** `2603.02525` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Görkem Can Süleymanoğlu
+- **abs:** https://arxiv.org/abs/2603.02525 · **PDF:** https://arxiv.org/pdf/2603.02525v1
+
+## 8322. Human-Certified Module Repositories for the AI Age
+
+- **ID:** `2603.02512` · **primary:** `cs.ET` · **Submitted:** 2026-03-03
+- **Authors:** Szilárd Enyedi
+- **abs:** https://arxiv.org/abs/2603.02512 · **PDF:** https://arxiv.org/pdf/2603.02512v3
+
+## 8323. Learning Object-Centric Spatial Reasoning for Sequential Manipulation in Cluttered Environments
+
+- **ID:** `2603.02511` · **primary:** `cs.RO` · **Submitted:** 2026-03-03
+- **Authors:** Chrisantus Eze, Ryan C Julian, Christopher Crick
+- **abs:** https://arxiv.org/abs/2603.02511 · **PDF:** https://arxiv.org/pdf/2603.02511v1
+
+## 8324. ParEVO: Synthesizing Code for Irregular Data: High-Performance Parallelism through Agentic Evolution
+
+- **ID:** `2603.02510` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Liu Yang, Zeyu Nie, Andrew Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.02510 · **PDF:** https://arxiv.org/pdf/2603.02510v1
+
+## 8325. NeuroProlog: Multi-Task Fine-Tuning for Neurosymbolic Mathematical Reasoning via the Cocktail Effect
+
+- **ID:** `2603.02504` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Pratibha Zunjare, Michael Hsiao
+- **abs:** https://arxiv.org/abs/2603.02504 · **PDF:** https://arxiv.org/pdf/2603.02504v2
+
+## 8326. Revealing Positive and Negative Role Models to Help People Make Good Decisions
+
+- **ID:** `2603.02495` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Avrim Blum, Keziah Naggita, Matthew R. Walter, et al.
+- **abs:** https://arxiv.org/abs/2603.02495 · **PDF:** https://arxiv.org/pdf/2603.02495v1
+
+## 8327. Form Follows Function: Recursive Stem Model
+
+- **ID:** `2603.15641` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Navid Hakimi
+- **abs:** https://arxiv.org/abs/2603.15641 · **PDF:** https://arxiv.org/pdf/2603.15641v1
+
+## 8328. What Capable Agents Must Know: Selection Theorems for Robust Decision-Making under Uncertainty
+
+- **ID:** `2603.02491` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Aran Nayebi
+- **abs:** https://arxiv.org/abs/2603.02491 · **PDF:** https://arxiv.org/pdf/2603.02491v2
+
+## 8329. Learning-Augmented Moment Estimation on Time-Decay Models
+
+- **ID:** `2603.02488` · **primary:** `cs.DS` · **Submitted:** 2026-03-03
+- **Authors:** Soham Nagawanshi, Shalini Panthangi, Chen Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02488 · **PDF:** https://arxiv.org/pdf/2603.02488v1
+
+## 8330. Geometric structures and deviations on James' symmetric positive-definite matrix bicone domain
+
+- **ID:** `2603.02483` · **primary:** `stat.ML` · **Submitted:** 2026-03-03
+- **Authors:** Jacek Karwowski, Frank Nielsen
+- **abs:** https://arxiv.org/abs/2603.02483 · **PDF:** https://arxiv.org/pdf/2603.02483v1
+
+## 8331. MUSE: A Run-Centric Platform for Multimodal Unified Safety Evaluation of Large Language Models
+
+- **ID:** `2603.02482` · **primary:** `cs.LG` · **Submitted:** 2026-03-03
+- **Authors:** Zhongxi Wang, Yueqian Lin, Jingyang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.02482 · **PDF:** https://arxiv.org/pdf/2603.02482v1
+
+## 8332. Optimizing Orbital Parameters of Satellites for a Global Quantum Network
+
+- **ID:** `2603.02480` · **primary:** `quant-ph` · **Submitted:** 2026-03-03
+- **Authors:** Athul Ashok, Owen DePoint, Jackson MacDonald, et al.
+- **abs:** https://arxiv.org/abs/2603.02480 · **PDF:** https://arxiv.org/pdf/2603.02480v1
+
+## 8333. PRISM: Pushing the Frontier of Deep Think via Process Reward Model-Guided Inference
+
+- **ID:** `2603.02479` · **primary:** `cs.AI` · **Submitted:** 2026-03-03
+- **Authors:** Rituraj Sharma, Weiyuan Chen, Noah Provenzano, et al.
+- **abs:** https://arxiv.org/abs/2603.02479 · **PDF:** https://arxiv.org/pdf/2603.02479v1
+
+## 8334. The Theory behind UMAP?
+
+- **ID:** `2603.03375` · **primary:** `stat.ML` · **Submitted:** 2026-03-02
+- **Authors:** David Wegmann
+- **abs:** https://arxiv.org/abs/2603.03375 · **PDF:** https://arxiv.org/pdf/2603.03375v1
+
+## 8335. Large-Scale Dataset and Benchmark for Skin Tone Classification in the Wild
+
+- **ID:** `2603.02475` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Vitor Pereira Matias, Márcus Vinícius Lobo Costa, João Batista Neto, et al.
+- **abs:** https://arxiv.org/abs/2603.02475 · **PDF:** https://arxiv.org/pdf/2603.02475v1
+
+## 8336. Diagnosing Retrieval vs. Utilization Bottlenecks in LLM Agent Memory
+
+- **ID:** `2603.02473` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Boqin Yuan, Yue Su, Kun Yao
+- **abs:** https://arxiv.org/abs/2603.02473 · **PDF:** https://arxiv.org/pdf/2603.02473v1
+
+## 8337. Video TokenCom: Textual Intent-Guided Multi-Rate Video Token Communications with UEP-Based Adaptive Source-Channel Coding
+
+- **ID:** `2603.02470` · **primary:** `cs.IT` · **Submitted:** 2026-03-02
+- **Authors:** Jingxuan Men, Mahdi Boloursaz Mashhadi, Ning Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02470 · **PDF:** https://arxiv.org/pdf/2603.02470v1
+
+## 8338. Deep Learning Based Wildfire Detection for Peatland Fires Using Transfer Learning
+
+- **ID:** `2603.02465` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Emadeldeen Hamdan, Ahmad Faiz Tharima, Mohd Zahirasri Mohd Tohir, et al.
+- **abs:** https://arxiv.org/abs/2603.02465 · **PDF:** https://arxiv.org/pdf/2603.02465v1
+
+## 8339. GLoRIA: Gated Low-Rank Interpretable Adaptation for Dialectal ASR
+
+- **ID:** `2603.02464` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Pouya Mehralian, Melissa Farasyn, Anne Breitbarth, et al.
+- **abs:** https://arxiv.org/abs/2603.02464 · **PDF:** https://arxiv.org/pdf/2603.02464v1
+
+## 8340. Can Computational Reducibility Lead to Transferable Models for Graph Combinatorial Optimization?
+
+- **ID:** `2603.02462` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Semih Cantürk, Thomas Sabourin, Frederik Wenkel, et al.
+- **abs:** https://arxiv.org/abs/2603.02462 · **PDF:** https://arxiv.org/pdf/2603.02462v1
+
+## 8341. Conformal Graph Prediction with Z-Gromov Wasserstein Distances
+
+- **ID:** `2603.02460` · **primary:** `stat.ML` · **Submitted:** 2026-03-02
+- **Authors:** Gabriel Melo, Thibaut de Saivre, Anna Calissano, et al.
+- **abs:** https://arxiv.org/abs/2603.02460 · **PDF:** https://arxiv.org/pdf/2603.02460v4
+
+## 8342. Manifold Aware Denoising Score Matching (MAD)
+
+- **ID:** `2603.02452` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Alona Levy-Jurgenson, Alvaro Prat, James Cuin, et al.
+- **abs:** https://arxiv.org/abs/2603.02452 · **PDF:** https://arxiv.org/pdf/2603.02452v1
+
+## 8343. Spectral Regularization for Diffusion Models
+
+- **ID:** `2603.02447` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Satish Chandran, Nicolas Roque dos Santos, Yunshu Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.02447 · **PDF:** https://arxiv.org/pdf/2603.02447v1
+
+## 8344. Using the SEKF to Transfer NN Models of Dynamical Systems with Limited Data
+
+- **ID:** `2603.02439` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Joshua E. Hammond, Tyler A. Soderstrom, Brian A. Korgel, et al.
+- **abs:** https://arxiv.org/abs/2603.02439 · **PDF:** https://arxiv.org/pdf/2603.02439v1
+
+## 8345. VL-KGE: Vision-Language Models Meet Knowledge Graph Embeddings
+
+- **ID:** `2603.02435` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Athanasios Efthymiou, Stevan Rudinac, Monika Kackovic, et al.
+- **abs:** https://arxiv.org/abs/2603.02435 · **PDF:** https://arxiv.org/pdf/2603.02435v1
+
+## 8346. MIRAGE: Knowledge Graph-Guided Cross-Cohort MRI Synthesis for Alzheimer's Disease Prediction
+
+- **ID:** `2603.02434` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Guanchen Wu, Zhe Huang, Yuzhang Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.02434 · **PDF:** https://arxiv.org/pdf/2603.02434v1
+
+## 8347. A Unified Revisit of Temperature in Classification-Based Knowledge Distillation
+
+- **ID:** `2603.02430` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Logan Frank, Jim Davis
+- **abs:** https://arxiv.org/abs/2603.02430 · **PDF:** https://arxiv.org/pdf/2603.02430v2
+
+## 8348. Dimension-Independent Convergence of Underdamped Langevin Monte Carlo in KL Divergence
+
+- **ID:** `2603.02429` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Shiyuan Zhang, Qiwei Di, Xuheng Li, et al.
+- **abs:** https://arxiv.org/abs/2603.02429 · **PDF:** https://arxiv.org/pdf/2603.02429v1
+
+## 8349. Learning to Pay Attention: Unsupervised Modeling of Attentive and Inattentive Respondents in Survey Data
+
+- **ID:** `2603.02427` · **primary:** `cs.HC` · **Submitted:** 2026-03-02
+- **Authors:** Ilias Triantafyllopoulos, Panos Ipeirotis
+- **abs:** https://arxiv.org/abs/2603.02427 · **PDF:** https://arxiv.org/pdf/2603.02427v1
+
+## 8350. Personalized Multi-Agent Average Reward TD-Learning via Joint Linear Approximation
+
+- **ID:** `2603.02426` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Leo Muxing Wang, Pengkun Yang, Lili Su
+- **abs:** https://arxiv.org/abs/2603.02426 · **PDF:** https://arxiv.org/pdf/2603.02426v2
+
+## 8351. A Directed Graph Model and Experimental Framework for Design and Study of Time-Dependent Text Visualisation
+
+- **ID:** `2603.02422` · **primary:** `cs.HC` · **Submitted:** 2026-03-02
+- **Authors:** Songhai Fan, Simon Angus, Tim Dwyer, et al.
+- **abs:** https://arxiv.org/abs/2603.02422 · **PDF:** https://arxiv.org/pdf/2603.02422v1
+
+## 8352. Slurry-as-a-Service: A Modest Proposal on Scalable Pluralistic Alignment for Nutrient Optimization
+
+- **ID:** `2603.02420` · **primary:** `cs.CY` · **Submitted:** 2026-03-02
+- **Authors:** Rachel Hong, Yael Eiger, Jevan Hutson, et al.
+- **abs:** https://arxiv.org/abs/2603.02420 · **PDF:** https://arxiv.org/pdf/2603.02420v2
+
+## 8353. Sleeper Cell: Injecting Latent Malice Temporal Backdoors into Tool-Using LLMs
+
+- **ID:** `2603.03371` · **primary:** `cs.CR` · **Submitted:** 2026-03-02
+- **Authors:** Bhanu Pallakonda, Mikkel Hindsbo, Sina Ehsani, et al.
+- **abs:** https://arxiv.org/abs/2603.03371 · **PDF:** https://arxiv.org/pdf/2603.03371v1
+
+## 8354. Fisher-Geometric Diffusion in Stochastic Gradient Descent: Optimal Rates, Oracle Complexity, and Information-Theoretic Limits
+
+- **ID:** `2603.02417` · **primary:** `stat.ML` · **Submitted:** 2026-03-02
+- **Authors:** Daniel Zantedeschi, Kumar Muthuraman
+- **abs:** https://arxiv.org/abs/2603.02417 · **PDF:** https://arxiv.org/pdf/2603.02417v1
+
+## 8355. From Fewer Samples to Fewer Bits: Reframing Dataset Distillation as Joint Optimization of Precision and Compactness
+
+- **ID:** `2603.02411` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** My H. Dinh, Aditya Sant, Akshay Malhotra, et al.
+- **abs:** https://arxiv.org/abs/2603.02411 · **PDF:** https://arxiv.org/pdf/2603.02411v1
+
+## 8356. Rigidity-Aware Geometric Pretraining for Protein Design and Conformational Ensembles
+
+- **ID:** `2603.02406` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Zhanghan Ni, Yanjing Li, Zeju Qiu, et al.
+- **abs:** https://arxiv.org/abs/2603.02406 · **PDF:** https://arxiv.org/pdf/2603.02406v2
+
+## 8357. COOL-MC: Verifying and Explaining RL Policies for Platelet Inventory Management
+
+- **ID:** `2603.02396` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Dennis Gross
+- **abs:** https://arxiv.org/abs/2603.02396 · **PDF:** https://arxiv.org/pdf/2603.02396v1
+
+## 8358. CUCo: An Agentic Framework for Compute and Communication Co-design
+
+- **ID:** `2603.02376` · **primary:** `cs.DC` · **Submitted:** 2026-03-02
+- **Authors:** Bodun Hu, Yoga Sri Varshan, Saurabh Agarwal, et al.
+- **abs:** https://arxiv.org/abs/2603.02376 · **PDF:** https://arxiv.org/pdf/2603.02376v1
+
+## 8359. RO-N3WS: Enhancing Generalization in Low-Resource ASR with Diverse Romanian Speech Benchmarks
+
+- **ID:** `2603.02368` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Alexandra Diaconu, Mădălina Vînaga, Bogdan Alexe
+- **abs:** https://arxiv.org/abs/2603.02368 · **PDF:** https://arxiv.org/pdf/2603.02368v1
+
+## 8360. PlayWrite: A Multimodal System for AI Supported Narrative Co-Authoring Through Play in XR
+
+- **ID:** `2603.02366` · **primary:** `cs.HC` · **Submitted:** 2026-03-02
+- **Authors:** Esen K. Tütüncü, Qian Zhou, Frederik Brudy, et al.
+- **abs:** https://arxiv.org/abs/2603.02366 · **PDF:** https://arxiv.org/pdf/2603.02366v1
+
+## 8361. Can machines be uncertain?
+
+- **ID:** `2603.02365` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Luis Rosa
+- **abs:** https://arxiv.org/abs/2603.02365 · **PDF:** https://arxiv.org/pdf/2603.02365v2
+
+## 8362. Estimating Visual Attribute Effects in Advertising from Observational Data: A Deepfake-Informed Double Machine Learning Approach
+
+- **ID:** `2603.02359` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Yizhi Liu, Balaji Padmanabhan, Siva Viswanathan
+- **abs:** https://arxiv.org/abs/2603.02359 · **PDF:** https://arxiv.org/pdf/2603.02359v1
+
+## 8363. Learning Optimal Search Strategies
+
+- **ID:** `2603.02356` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Stefan Ankirchner, Maximilian Philipp Thiel
+- **abs:** https://arxiv.org/abs/2603.02356 · **PDF:** https://arxiv.org/pdf/2603.02356v1
+
+## 8364. Detecting AI-Generated Essays in Writing Assessment: Responsible Use and Generalizability Across LLMs
+
+- **ID:** `2603.02353` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Jiangang Hao
+- **abs:** https://arxiv.org/abs/2603.02353 · **PDF:** https://arxiv.org/pdf/2603.02353v2
+
+## 8365. Learning graph topology from metapopulation epidemic encoder-decoder
+
+- **ID:** `2603.02349` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Xin Li, Jonathan Cohen, Shai Pilosof, et al.
+- **abs:** https://arxiv.org/abs/2603.02349 · **PDF:** https://arxiv.org/pdf/2603.02349v1
+
+## 8366. Diffusion-MPC in Discrete Domains: Feasibility Constraints, Horizon Effects, and Critic Alignment: Case study with Tetris
+
+- **ID:** `2603.02348` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Haochuan Kevin Wang
+- **abs:** https://arxiv.org/abs/2603.02348 · **PDF:** https://arxiv.org/pdf/2603.02348v1
+
+## 8367. Large Electron Model: A Universal Ground State Predictor
+
+- **ID:** `2603.02346` · **primary:** `cond-mat.str-el` · **Submitted:** 2026-03-02
+- **Authors:** Timothy Zaklama, Max Geier, Liang Fu
+- **abs:** https://arxiv.org/abs/2603.02346 · **PDF:** https://arxiv.org/pdf/2603.02346v1
+
+## 8368. RIVA: Leveraging LLM Agents for Reliable Configuration Drift Detection
+
+- **ID:** `2603.02345` · **primary:** `cs.SE` · **Submitted:** 2026-03-02
+- **Authors:** Sami Abuzakuk, Lucas Crijns, Anne-Marie Kermarrec, et al.
+- **abs:** https://arxiv.org/abs/2603.02345 · **PDF:** https://arxiv.org/pdf/2603.02345v1
+
+## 8369. Beyond Detection: Governing GenAI in Academic Peer Review as a Sociotechnical Challenge
+
+- **ID:** `2603.20214` · **primary:** `cs.CY` · **Submitted:** 2026-03-02
+- **Authors:** Tatiana Chakravorti, Pranav Narayanan Venkit, Sourojit Ghosh, et al.
+- **abs:** https://arxiv.org/abs/2603.20214 · **PDF:** https://arxiv.org/pdf/2603.20214v1
+
+## 8370. Bridging the Reproducibility Divide: Open Source Software's Role in Standardizing Healthcare AI
+
+- **ID:** `2603.03367` · **primary:** `cs.CY` · **Submitted:** 2026-03-02
+- **Authors:** John Wu, Zhenbang Wu, Jimeng Sun
+- **abs:** https://arxiv.org/abs/2603.03367 · **PDF:** https://arxiv.org/pdf/2603.03367v1
+
+## 8371. Preconditioned Score and Flow Matching
+
+- **ID:** `2603.02337` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Shadab Ahamed, Eshed Gal, Simon Ghyselincks, et al.
+- **abs:** https://arxiv.org/abs/2603.02337 · **PDF:** https://arxiv.org/pdf/2603.02337v1
+
+## 8372. Characterizing Memorization in Diffusion Language Models: Generalized Extraction and Sampling Effects
+
+- **ID:** `2603.02333` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Xiaoyu Luo, Wenrui Yu, Qiongxiu Li, et al.
+- **abs:** https://arxiv.org/abs/2603.02333 · **PDF:** https://arxiv.org/pdf/2603.02333v1
+
+## 8373. Neural Demand Estimation with Habit Formation and Rationality Constraints
+
+- **ID:** `2603.02331` · **primary:** `econ.GN` · **Submitted:** 2026-03-02
+- **Authors:** Marta Grzeskiewicz
+- **abs:** https://arxiv.org/abs/2603.02331 · **PDF:** https://arxiv.org/pdf/2603.02331v1
+
+## 8374. Reasoning Core: A Scalable Procedural Data Generation Suite for Symbolic Pre-training and Post-Training
+
+- **ID:** `2603.02208` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Valentin Lacombe, Valentin Quesnel, Damien Sileo
+- **abs:** https://arxiv.org/abs/2603.02208 · **PDF:** https://arxiv.org/pdf/2603.02208v1
+
+## 8375. Partial Causal Structure Learning for Valid Selective Conformal Inference under Interventions
+
+- **ID:** `2603.02204` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Amir Asiaee, Kavey Aryan, James P. Long
+- **abs:** https://arxiv.org/abs/2603.02204 · **PDF:** https://arxiv.org/pdf/2603.02204v1
+
+## 8376. Tool Verification for Test-Time Reinforcement Learning
+
+- **ID:** `2603.02203` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Ruotong Liao, Nikolai Röhrich, Xiaohan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02203 · **PDF:** https://arxiv.org/pdf/2603.02203v1
+
+## 8377. Frontier Models Can Take Actions at Low Probabilities
+
+- **ID:** `2603.02202` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Alex Serrano, Wen Xing, David Lindner, et al.
+- **abs:** https://arxiv.org/abs/2603.02202 · **PDF:** https://arxiv.org/pdf/2603.02202v1
+
+## 8378. Adaptive Confidence Regularization for Multimodal Failure Detection
+
+- **ID:** `2603.02200` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Moru Liu, Hao Dong, Olga Fink, et al.
+- **abs:** https://arxiv.org/abs/2603.02200 · **PDF:** https://arxiv.org/pdf/2603.02200v1
+
+## 8379. Conformal Policy Control
+
+- **ID:** `2603.02196` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Drew Prinster, Clara Fannjiang, Ji Won Park, et al.
+- **abs:** https://arxiv.org/abs/2603.02196 · **PDF:** https://arxiv.org/pdf/2603.02196v1
+
+## 8380. From Leaderboard to Deployment: Code Quality Challenges in AV Perception Repositories
+
+- **ID:** `2603.02194` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Mateus Karvat, Bram Adams, Sidney Givigi
+- **abs:** https://arxiv.org/abs/2603.02194 · **PDF:** https://arxiv.org/pdf/2603.02194v1
+
+## 8381. Symbol-Equivariant Recurrent Reasoning Models
+
+- **ID:** `2603.02193` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Richard Freinschlag, Timo Bertram, Erich Kobler, et al.
+- **abs:** https://arxiv.org/abs/2603.02193 · **PDF:** https://arxiv.org/pdf/2603.02193v1
+
+## 8382. Sketch2Colab: Sketch-Conditioned Multi-Human Animation via Controllable Flow Distillation
+
+- **ID:** `2603.02190` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Divyanshu Daiya, Aniket Bera
+- **abs:** https://arxiv.org/abs/2603.02190 · **PDF:** https://arxiv.org/pdf/2603.02190v2
+
+## 8383. Multi-Head Low-Rank Attention
+
+- **ID:** `2603.02188` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Songtao Liu, Hongwu Peng, Zhiwei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.02188 · **PDF:** https://arxiv.org/pdf/2603.02188v1
+
+## 8384. MAC: A Conversion Rate Prediction Benchmark Featuring Labels Under Multiple Attribution Mechanisms
+
+- **ID:** `2603.02184` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Jinqi Wu, Sishuo Chen, Zhangming Chan, et al.
+- **abs:** https://arxiv.org/abs/2603.02184 · **PDF:** https://arxiv.org/pdf/2603.02184v1
+
+## 8385. Leveraging Model Soups to Classify Intangible Cultural Heritage Images from the Mekong Delta
+
+- **ID:** `2603.02181` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Quoc-Khang Tran, Minh-Thien Nguyen, Nguyen-Khang Pham
+- **abs:** https://arxiv.org/abs/2603.02181 · **PDF:** https://arxiv.org/pdf/2603.02181v1
+
+## 8386. Reservoir Subspace Injection for Online ICA under Top-n Whitening
+
+- **ID:** `2603.02178` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Wenjun Xiao, Yuda Bi, Vince D Calhoun
+- **abs:** https://arxiv.org/abs/2603.02178 · **PDF:** https://arxiv.org/pdf/2603.02178v1
+
+## 8387. Organizing, Orchestrating, and Benchmarking Agent Skills at Ecosystem Scale
+
+- **ID:** `2603.02176` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Hao Li, Chunjiang Mu, Jianhao Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.02176 · **PDF:** https://arxiv.org/pdf/2603.02176v1
+
+## 8388. Kiwi-Edit: Versatile Video Editing via Instruction and Reference Guidance
+
+- **ID:** `2603.02175` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Yiqi Lin, Guoqiang Liang, Ziyun Zeng, et al.
+- **abs:** https://arxiv.org/abs/2603.02175 · **PDF:** https://arxiv.org/pdf/2603.02175v3
+
+## 8389. De-paradox Tree: Breaking Down Simpson's Paradox via A Kernel-Based Partition Algorithm
+
+- **ID:** `2603.02174` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Xian Teng, Yu-Ru Lin
+- **abs:** https://arxiv.org/abs/2603.02174 · **PDF:** https://arxiv.org/pdf/2603.02174v1
+
+## 8390. SageBwd: A Trainable Low-bit Attention
+
+- **ID:** `2603.02170` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Jintao Zhang, Marco Chen, Haoxu Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02170 · **PDF:** https://arxiv.org/pdf/2603.02170v1
+
+## 8391. Instrumental and Proximal Causal Inference with Gaussian Processes
+
+- **ID:** `2603.02159` · **primary:** `stat.ML` · **Submitted:** 2026-03-02
+- **Authors:** Yuqi Zhang, Krikamol Muandet, Dino Sejdinovic, et al.
+- **abs:** https://arxiv.org/abs/2603.02159 · **PDF:** https://arxiv.org/pdf/2603.02159v1
+
+## 8392. ZeroDayBench: Evaluating LLM Agents on Unseen Zero-Day Vulnerabilities for Cyberdefense
+
+- **ID:** `2603.02297` · **primary:** `cs.CR` · **Submitted:** 2026-03-02
+- **Authors:** Nancy Lau, Louis Sloot, Jyoutir Raj, et al.
+- **abs:** https://arxiv.org/abs/2603.02297 · **PDF:** https://arxiv.org/pdf/2603.02297v1
+
+## 8393. How Small Can 6G Reason? Scaling Tiny Language Models for AI-Native Networks
+
+- **ID:** `2603.02156` · **primary:** `cs.NI` · **Submitted:** 2026-03-02
+- **Authors:** Mohamed Amine Ferrag, Abderrahmane Lakas, Merouane Debbah
+- **abs:** https://arxiv.org/abs/2603.02156 · **PDF:** https://arxiv.org/pdf/2603.02156v1
+
+## 8394. Near-Optimal Regret for KL-Regularized Multi-Armed Bandits
+
+- **ID:** `2603.02155` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Kaixuan Ji, Qingyue Zhao, Heyang Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.02155 · **PDF:** https://arxiv.org/pdf/2603.02155v1
+
+## 8395. Boltzmann-based Exploration for Robust Decentralized Multi-Agent Planning (Extended Version)
+
+- **ID:** `2603.02154` · **primary:** `cs.MA` · **Submitted:** 2026-03-02
+- **Authors:** Nhat D. A. Nguyen, Duong D. Nguyen, Gianluca Rizzo, et al.
+- **abs:** https://arxiv.org/abs/2603.02154 · **PDF:** https://arxiv.org/pdf/2603.02154v2
+
+## 8396. Scaling Retrieval Augmented Generation with RAG Fusion: Lessons from an Industry Deployment
+
+- **ID:** `2603.02153` · **primary:** `cs.IR` · **Submitted:** 2026-03-02
+- **Authors:** Luigi Medrano, Arush Verma, Mukul Chhabra
+- **abs:** https://arxiv.org/abs/2603.02153 · **PDF:** https://arxiv.org/pdf/2603.02153v1
+
+## 8397. Zero- and Few-Shot Named-Entity Recognition: Case Study and Dataset in the Crime Domain (CrimeNER)
+
+- **ID:** `2603.02150` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Miguel Lopez-Duran, Julian Fierrez, Aythami Morales, et al.
+- **abs:** https://arxiv.org/abs/2603.02150 · **PDF:** https://arxiv.org/pdf/2603.02150v1
+
+## 8398. LongRLVR: Long-Context Reinforcement Learning Requires Verifiable Context Rewards
+
+- **ID:** `2603.02146` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Guanzheng Chen, Michael Qizhe Shieh, Lidong Bing
+- **abs:** https://arxiv.org/abs/2603.02146 · **PDF:** https://arxiv.org/pdf/2603.02146v1
+
+## 8399. Machine Learning (ML) library in Linux kernel
+
+- **ID:** `2603.02145` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Viacheslav Dubeyko
+- **abs:** https://arxiv.org/abs/2603.02145 · **PDF:** https://arxiv.org/pdf/2603.02145v1
+
+## 8400. Is Bigger Always Better? Efficiency Analysis in Resource-Constrained Small Object Detection
+
+- **ID:** `2603.02142` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Kwame Mbobda-Kuate, Gabriel Kasmi
+- **abs:** https://arxiv.org/abs/2603.02142 · **PDF:** https://arxiv.org/pdf/2603.02142v1
+
+## 8401. LiftAvatar: Kinematic-Space Completion for Expression-Controlled 3D Gaussian Avatar Animation
+
+- **ID:** `2603.02129` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Hualiang Wei, Shunran Jia, Jialun Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.02129 · **PDF:** https://arxiv.org/pdf/2603.02129v1
+
+## 8402. LLMs as Strategic Actors: Behavioral Alignment, Risk Calibration, and Argumentation Framing in Geopolitical Simulations
+
+- **ID:** `2603.02128` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Veronika Solopova, Viktoria Skorik, Maksym Tereshchenko, et al.
+- **abs:** https://arxiv.org/abs/2603.02128 · **PDF:** https://arxiv.org/pdf/2603.02128v1
+
+## 8403. Nano-EmoX: Unifying Multimodal Emotional Intelligence from Perception to Empathy
+
+- **ID:** `2603.02123` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Jiahao Huang, Fengyan Lin, Xuechao Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.02123 · **PDF:** https://arxiv.org/pdf/2603.02123v2
+
+## 8404. Pencil Puzzle Bench: A Benchmark for Multi-Step Verifiable Reasoning
+
+- **ID:** `2603.02119` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Justin Waugh
+- **abs:** https://arxiv.org/abs/2603.02119 · **PDF:** https://arxiv.org/pdf/2603.02119v1
+
+## 8405. Robometer: Scaling General-Purpose Robotic Reward Models via Trajectory Comparisons
+
+- **ID:** `2603.02115` · **primary:** `cs.RO` · **Submitted:** 2026-03-02
+- **Authors:** Anthony Liang, Yigit Korkmaz, Jiahui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.02115 · **PDF:** https://arxiv.org/pdf/2603.02115v1
+
+## 8406. Recursive Models for Long-Horizon Reasoning
+
+- **ID:** `2603.02112` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Chenxiao Yang, Nathan Srebro, Zhiyuan Li
+- **abs:** https://arxiv.org/abs/2603.02112 · **PDF:** https://arxiv.org/pdf/2603.02112v1
+
+## 8407. Orchestrating Multimodal DNN Workloads in Wireless Neural Processing
+
+- **ID:** `2603.02109` · **primary:** `eess.SP` · **Submitted:** 2026-03-02
+- **Authors:** Sai Xu, Kai-Kit Wong, Yanan Du, et al.
+- **abs:** https://arxiv.org/abs/2603.02109 · **PDF:** https://arxiv.org/pdf/2603.02109v1
+
+## 8408. Stochastic Multi-Armed Bandits with Limited Control Variates
+
+- **ID:** `2603.02100` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Arun Verma, Manjesh Kumar Hanawal, Arun Rajkumar
+- **abs:** https://arxiv.org/abs/2603.02100 · **PDF:** https://arxiv.org/pdf/2603.02100v1
+
+## 8409. Recursive Think-Answer Process for LLMs and VLMs
+
+- **ID:** `2603.02099` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Byung-Kwan Lee, Youngchae Chee, Yong Man Ro
+- **abs:** https://arxiv.org/abs/2603.02099 · **PDF:** https://arxiv.org/pdf/2603.02099v2
+
+## 8410. Efficient and High-Fidelity Omni Modality Retrieval
+
+- **ID:** `2603.02098` · **primary:** `cs.IR` · **Submitted:** 2026-03-02
+- **Authors:** Chuong Huynh, Manh Luong, Abhinav Shrivastava
+- **abs:** https://arxiv.org/abs/2603.02098 · **PDF:** https://arxiv.org/pdf/2603.02098v2
+
+## 8411. ClinConsensus: A Consensus-Based Benchmark for Evaluating Chinese Medical LLMs across Difficulty Levels
+
+- **ID:** `2603.02097` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Xiang Zheng, Han Li, Wenjie Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.02097 · **PDF:** https://arxiv.org/pdf/2603.02097v4
+
+## 8412. FluxMem: Adaptive Hierarchical Memory for Streaming Video Understanding
+
+- **ID:** `2603.02096` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Yiweng Xie, Bo He, Junke Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02096 · **PDF:** https://arxiv.org/pdf/2603.02096v1
+
+## 8413. On the Rate of Convergence of GD in Non-linear Neural Networks: An Adversarial Robustness Perspective
+
+- **ID:** `2603.02095` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Guy Smorodinsky, Sveta Gimpleson, Itay Safran
+- **abs:** https://arxiv.org/abs/2603.02095 · **PDF:** https://arxiv.org/pdf/2603.02095v1
+
+## 8414. Adam Converges Without Any Modification On Update Rules
+
+- **ID:** `2603.02092` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Yushun Zhang, Bingran Li, Congliang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.02092 · **PDF:** https://arxiv.org/pdf/2603.02092v1
+
+## 8415. Learning from Synthetic Data Improves Multi-hop Reasoning
+
+- **ID:** `2603.02091` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Anmol Kabra, Yilun Yin, Albert Gong, et al.
+- **abs:** https://arxiv.org/abs/2603.02091 · **PDF:** https://arxiv.org/pdf/2603.02091v1
+
+## 8416. A Detection-Gated Pipeline for Robust Glottal Area Waveform Extraction and Clinical Pathology Assessment
+
+- **ID:** `2603.02087` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Harikrishnan Unnikrishnan
+- **abs:** https://arxiv.org/abs/2603.02087 · **PDF:** https://arxiv.org/pdf/2603.02087v2
+
+## 8417. Modeling Grammatical Hypothesis Testing in Young Learners: A Sequence-Based Learning Analytics Study of Morphosyntactic Reasoning in an Interactive Game
+
+- **ID:** `2603.02084` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Thierry Geoffre, Trystan Geoffre
+- **abs:** https://arxiv.org/abs/2603.02084 · **PDF:** https://arxiv.org/pdf/2603.02084v1
+
+## 8418. What Exactly do Children Receive in Language Acquisition? A Case Study on CHILDES with Automated Detection of Filler-Gap Dependencies
+
+- **ID:** `2603.02082` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Zhenghao Herbert Zhou, William Dai, Maya Viswanathan, et al.
+- **abs:** https://arxiv.org/abs/2603.02082 · **PDF:** https://arxiv.org/pdf/2603.02082v1
+
+## 8419. GenDB: The Next Generation of Query Processing -- Synthesized, Not Engineered
+
+- **ID:** `2603.02081` · **primary:** `cs.DB` · **Submitted:** 2026-03-02
+- **Authors:** Jiale Lao, Immanuel Trummer
+- **abs:** https://arxiv.org/abs/2603.02081 · **PDF:** https://arxiv.org/pdf/2603.02081v1
+
+## 8420. From Pixels to Patches: Pooling Strategies for Earth Embeddings
+
+- **ID:** `2603.02080` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Isaac Corley, Caleb Robinson, Inbal Becker-Reshef, et al.
+- **abs:** https://arxiv.org/abs/2603.02080 · **PDF:** https://arxiv.org/pdf/2603.02080v2
+
+## 8421. Cognitive Prosthetic: An AI-Enabled Multimodal System for Episodic Recall in Knowledge Work
+
+- **ID:** `2603.02072` · **primary:** `cs.HC` · **Submitted:** 2026-03-02
+- **Authors:** Lawrence Obiuwevwi, Krzysztof J. Rechowicz, Vikas Ashok, et al.
+- **abs:** https://arxiv.org/abs/2603.02072 · **PDF:** https://arxiv.org/pdf/2603.02072v1
+
+## 8422. Exploring Plan Space through Conversation: An Agentic Framework for LLM-Mediated Explanations in Planning
+
+- **ID:** `2603.02070` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Guilhem Fouilhé, Rebecca Eifler, Antonin Poché, et al.
+- **abs:** https://arxiv.org/abs/2603.02070 · **PDF:** https://arxiv.org/pdf/2603.02070v1
+
+## 8423. Scaling Laws of SignSGD in Linear Regression: When Does It Outperform SGD?
+
+- **ID:** `2603.02069` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Jihwan Kim, Dogyoon Song, Chulhee Yun
+- **abs:** https://arxiv.org/abs/2603.02069 · **PDF:** https://arxiv.org/pdf/2603.02069v1
+
+## 8424. Accelerating PDE Surrogates via RL-Guided Mesh Optimization
+
+- **ID:** `2603.02066` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Yang Meng, Ruoxi Jiang, Zhuokai Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.02066 · **PDF:** https://arxiv.org/pdf/2603.02066v1
+
+## 8425. Never Saddle for Reparameterized Steepest Descent as Mirror Flow
+
+- **ID:** `2603.02064` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Tom Jacobs, Chao Zhou, Rebekka Burkholz
+- **abs:** https://arxiv.org/abs/2603.02064 · **PDF:** https://arxiv.org/pdf/2603.02064v1
+
+## 8426. OpenRad: a Curated Repository of Open-access AI models for Radiology
+
+- **ID:** `2603.02062` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Konstantinos Vrettos, Galini Papadaki, Emmanouil Brilakis, et al.
+- **abs:** https://arxiv.org/abs/2603.02062 · **PDF:** https://arxiv.org/pdf/2603.02062v1
+
+## 8427. TRAKNN: Efficient Trajectory Aware Spatiotemporal kNN for Rare Meteorological Trajectory Detection
+
+- **ID:** `2603.02059` · **primary:** `stat.ML` · **Submitted:** 2026-03-02
+- **Authors:** Guillaume Coulaud, Davide Faranda
+- **abs:** https://arxiv.org/abs/2603.02059 · **PDF:** https://arxiv.org/pdf/2603.02059v1
+
+## 8428. On the Reliability of AI Methods in Drug Discovery: Evaluation of Boltz-2 for Structure and Binding Affinity Prediction
+
+- **ID:** `2603.05532` · **primary:** `physics.chem-ph` · **Submitted:** 2026-03-02
+- **Authors:** Shunzhou Wan, Xibei Zhang, Xiao Xue, et al.
+- **abs:** https://arxiv.org/abs/2603.05532 · **PDF:** https://arxiv.org/pdf/2603.05532v1
+
+## 8429. A Resource-Rational Principle for Modeling Visual Attention Control
+
+- **ID:** `2603.02056` · **primary:** `cs.HC` · **Submitted:** 2026-03-02
+- **Authors:** Yunpeng Bai
+- **abs:** https://arxiv.org/abs/2603.02056 · **PDF:** https://arxiv.org/pdf/2603.02056v1
+
+## 8430. Strategic Advice in the Age of Personal AI
+
+- **ID:** `2603.02055` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Yueyang Liu, Wichinpong Park Sinchaisri
+- **abs:** https://arxiv.org/abs/2603.02055 · **PDF:** https://arxiv.org/pdf/2603.02055v1
+
+## 8431. The Malignant Tail: Spectral Segregation of Label Noise in Over-Parameterized Networks
+
+- **ID:** `2603.02293` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Zice Wang
+- **abs:** https://arxiv.org/abs/2603.02293 · **PDF:** https://arxiv.org/pdf/2603.02293v1
+
+## 8432. "When to Hand Off, When to Work Together": Expanding Human-Agent Co-Creative Collaboration through Concurrent Interaction
+
+- **ID:** `2603.02050` · **primary:** `cs.HC` · **Submitted:** 2026-03-02
+- **Authors:** Kihoon Son, Hyewon Lee, DaEun Choi, et al.
+- **abs:** https://arxiv.org/abs/2603.02050 · **PDF:** https://arxiv.org/pdf/2603.02050v2
+
+## 8433. Expanding LLM Agent Boundaries with Strategy-Guided Exploration
+
+- **ID:** `2603.02045` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Andrew Szot, Michael Kirchhof, Omar Attia, et al.
+- **abs:** https://arxiv.org/abs/2603.02045 · **PDF:** https://arxiv.org/pdf/2603.02045v1
+
+## 8434. Leave-One-Out Prediction for General Hypothesis Classes
+
+- **ID:** `2603.02043` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Jian Qian, Jiachen Xu
+- **abs:** https://arxiv.org/abs/2603.02043 · **PDF:** https://arxiv.org/pdf/2603.02043v1
+
+## 8435. EstLLM: Enhancing Estonian Capabilities in Multilingual LLMs via Continued Pretraining and Post-Training
+
+- **ID:** `2603.02041` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Aleksei Dorkin, Taido Purason, Emil Kalbaliyev, et al.
+- **abs:** https://arxiv.org/abs/2603.02041 · **PDF:** https://arxiv.org/pdf/2603.02041v1
+
+## 8436. Graph neural network force fields for adiabatic dynamics of lattice Hamiltonians
+
+- **ID:** `2603.02039` · **primary:** `cond-mat.str-el` · **Submitted:** 2026-03-02
+- **Authors:** Yunhao Fan, Gia-Wei Chern
+- **abs:** https://arxiv.org/abs/2603.02039 · **PDF:** https://arxiv.org/pdf/2603.02039v1
+
+## 8437. AgenticGEO: A Self-Evolving Agentic System for Generative Engine Optimization
+
+- **ID:** `2603.20213` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Jiaqi Yuan, Jialu Wang, Zihan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.20213 · **PDF:** https://arxiv.org/pdf/2603.20213v1
+
+## 8438. TCG CREST System Description for the DISPLACE-M Challenge
+
+- **ID:** `2603.02030` · **primary:** `eess.AS` · **Submitted:** 2026-03-02
+- **Authors:** Nikhil Raghav, Md Sahidullah
+- **abs:** https://arxiv.org/abs/2603.02030 · **PDF:** https://arxiv.org/pdf/2603.02030v4
+
+## 8439. Rich Insights from Cheap Signals: Efficient Evaluations via Tensor Factorization
+
+- **ID:** `2603.02029` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Felipe Maia Polo, Aida Nematzadeh, Virginia Aglietti, et al.
+- **abs:** https://arxiv.org/abs/2603.02029 · **PDF:** https://arxiv.org/pdf/2603.02029v2
+
+## 8440. Latent attention on masked patches for flow reconstruction
+
+- **ID:** `2603.02028` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Ben Eze, Luca Magri, Andrea Nóvoa
+- **abs:** https://arxiv.org/abs/2603.02028 · **PDF:** https://arxiv.org/pdf/2603.02028v1
+
+## 8441. Framing Effects in Independent-Agent Large Language Models: A Cross-Family Behavioral Analysis
+
+- **ID:** `2603.19282` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Zice Wang, Zhenyu Zhang
+- **abs:** https://arxiv.org/abs/2603.19282 · **PDF:** https://arxiv.org/pdf/2603.19282v1
+
+## 8442. Learning to Read Where to Look: Disease-Aware Vision-Language Pretraining for 3D CT
+
+- **ID:** `2603.02026` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Simon Ging, Philipp Arnold, Sebastian Walter, et al.
+- **abs:** https://arxiv.org/abs/2603.02026 · **PDF:** https://arxiv.org/pdf/2603.02026v1
+
+## 8443. Revealing Combinatorial Reasoning of GNNs via Graph Concept Bottleneck Layer
+
+- **ID:** `2603.02025` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Yue Niu, Zhaokai Sun, Jiayi Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.02025 · **PDF:** https://arxiv.org/pdf/2603.02025v1
+
+## 8444. MMR-Life: Piecing Together Real-life Scenes for Multimodal Multi-image Reasoning
+
+- **ID:** `2603.02024` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Jiachun Li, Shaoping Huang, Zhuoran Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.02024 · **PDF:** https://arxiv.org/pdf/2603.02024v1
+
+## 8445. PonderLM-3: Adaptive Token-Wise Pondering with Differentiable Masking
+
+- **ID:** `2603.02023` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** He Li, Feichen Song, Boyi Zeng, et al.
+- **abs:** https://arxiv.org/abs/2603.02023 · **PDF:** https://arxiv.org/pdf/2603.02023v2
+
+## 8446. CodecFlow: Efficient Bandwidth Extension via Conditional Flow Matching in Neural Codec Latent Space
+
+- **ID:** `2603.02022` · **primary:** `cs.SD` · **Submitted:** 2026-03-02
+- **Authors:** Bowen Zhang, Junchuan Zhao, Ian McLoughlin, et al.
+- **abs:** https://arxiv.org/abs/2603.02022 · **PDF:** https://arxiv.org/pdf/2603.02022v2
+
+## 8447. Selection as Power: Constrained Reinforcement for Bounded Decision Authority
+
+- **ID:** `2603.02019` · **primary:** `cs.MA` · **Submitted:** 2026-03-02
+- **Authors:** Jose Manuel de la Chica Rodriguez, Juan Manuel Vera Díaz
+- **abs:** https://arxiv.org/abs/2603.02019 · **PDF:** https://arxiv.org/pdf/2603.02019v1
+
+## 8448. CausalWrap: Model-Agnostic Causal Constraint Wrappers for Tabular Synthetic Data
+
+- **ID:** `2603.02015` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Amir Asiaee, Zhuohui J. Liang, Chao Yan
+- **abs:** https://arxiv.org/abs/2603.02015 · **PDF:** https://arxiv.org/pdf/2603.02015v1
+
+## 8449. MAP-Diff: Multi-Anchor Guided Diffusion for Progressive 3D Whole-Body Low-Dose PET Denoising
+
+- **ID:** `2603.02012` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Peiyuan Jing, Chun-Wun Cheng, Liutao Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.02012 · **PDF:** https://arxiv.org/pdf/2603.02012v1
+
+## 8450. Noise-Calibrated Inference from Differentially Private Sufficient Statistics in Exponential Families
+
+- **ID:** `2603.02010` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Amir Asiaee, Samhita Pal
+- **abs:** https://arxiv.org/abs/2603.02010 · **PDF:** https://arxiv.org/pdf/2603.02010v1
+
+## 8451. Temporal Representations for Exploration: Learning Complex Exploratory Behavior without Extrinsic Rewards
+
+- **ID:** `2603.02008` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Faisal Mohamed, Catherine Ji, Benjamin Eysenbach, et al.
+- **abs:** https://arxiv.org/abs/2603.02008 · **PDF:** https://arxiv.org/pdf/2603.02008v1
+
+## 8452. Mitigating topology biases in Graph Diffusion via Counterfactual Intervention
+
+- **ID:** `2603.02005` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Wendi Wang, Jiaxi Yang, Yongkang Du, et al.
+- **abs:** https://arxiv.org/abs/2603.02005 · **PDF:** https://arxiv.org/pdf/2603.02005v1
+
+## 8453. MatRIS: Toward Reliable and Efficient Pretrained Machine Learning Interatomic Potentials
+
+- **ID:** `2603.02002` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Yuanchang Zhou, Siyu Hu, Xiangyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.02002 · **PDF:** https://arxiv.org/pdf/2603.02002v3
+
+## 8454. Learning Vision-Based Omnidirectional Navigation: A Teacher-Student Approach Using Monocular Depth Estimation
+
+- **ID:** `2603.01999` · **primary:** `cs.RO` · **Submitted:** 2026-03-02
+- **Authors:** Jan Finke, Wayne Paul Martis, Adrian Schmelter, et al.
+- **abs:** https://arxiv.org/abs/2603.01999 · **PDF:** https://arxiv.org/pdf/2603.01999v1
+
+## 8455. Fast-Slow Thinking RM: Efficient Integration of Scalar and Generative Reward Models
+
+- **ID:** `2603.20212` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Jiayun Wu, Peixu Hou, Shan Qu, et al.
+- **abs:** https://arxiv.org/abs/2603.20212 · **PDF:** https://arxiv.org/pdf/2603.20212v1
+
+## 8456. According to Me: Long-Term Personalized Referential Memory QA
+
+- **ID:** `2603.01990` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Jingbiao Mei, Jinghong Chen, Guangyu Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.01990 · **PDF:** https://arxiv.org/pdf/2603.01990v1
+
+## 8457. Exploring Teacher-Chatbot Interaction and Affect in Block-Based Programming
+
+- **ID:** `2603.20211` · **primary:** `cs.CY` · **Submitted:** 2026-03-02
+- **Authors:** Bahare Riahi, Ally Limke, Xiaoyi Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.20211 · **PDF:** https://arxiv.org/pdf/2603.20211v1
+
+## 8458. Accurate, private, secure, federated U-statistics with higher degree
+
+- **ID:** `2603.01986` · **primary:** `cs.CR` · **Submitted:** 2026-03-02
+- **Authors:** Quentin Sinh, Jan Ramon
+- **abs:** https://arxiv.org/abs/2603.01986 · **PDF:** https://arxiv.org/pdf/2603.01986v1
+
+## 8459. Quantitative Convergence of Wasserstein Gradient Flows of Kernel Mean Discrepancies
+
+- **ID:** `2603.01977` · **primary:** `math.AP` · **Submitted:** 2026-03-02
+- **Authors:** Lénaïc Chizat, Maria Colombo, Roberto Colombo, et al.
+- **abs:** https://arxiv.org/abs/2603.01977 · **PDF:** https://arxiv.org/pdf/2603.01977v1
+
+## 8460. Density-Matrix Spectral Embeddings for Categorical Data: Operator Structure and Stability
+
+- **ID:** `2603.01975` · **primary:** `stat.ML` · **Submitted:** 2026-03-02
+- **Authors:** Raquel Bosch-Romeu, Antonio Falcó, osé-Antonio Rodríguez-Gallego
+- **abs:** https://arxiv.org/abs/2603.01975 · **PDF:** https://arxiv.org/pdf/2603.01975v1
+
+## 8461. CharacterFlywheel: Scaling Iterative Improvement of Engaging and Steerable LLMs in Production
+
+- **ID:** `2603.01973` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Yixin Nie, Lin Guan, Zhongyao Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.01973 · **PDF:** https://arxiv.org/pdf/2603.01973v1
+
+## 8462. LOCUS: A Distribution-Free Loss-Quantile Score for Risk-Aware Predictions
+
+- **ID:** `2603.01971` · **primary:** `stat.ML` · **Submitted:** 2026-03-02
+- **Authors:** Matheus Barreto, Mário de Castro, Thiago R. Ramos, et al.
+- **abs:** https://arxiv.org/abs/2603.01971 · **PDF:** https://arxiv.org/pdf/2603.01971v1
+
+## 8463. Intrinsic Task Symmetry Drives Generalization in Algorithmic Tasks
+
+- **ID:** `2603.01968` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Hyeonbin Hwang, Yeachan Park
+- **abs:** https://arxiv.org/abs/2603.01968 · **PDF:** https://arxiv.org/pdf/2603.01968v1
+
+## 8464. AMemGym: Interactive Memory Benchmarking for Assistants in Long-Horizon Conversations
+
+- **ID:** `2603.01966` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Cheng Jiayang, Dongyu Ru, Lin Qiu, et al.
+- **abs:** https://arxiv.org/abs/2603.01966 · **PDF:** https://arxiv.org/pdf/2603.01966v1
+
+## 8465. CoVAE: correlated multimodal generative modeling
+
+- **ID:** `2603.01965` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Federico Caretti, Guido Sanguinetti
+- **abs:** https://arxiv.org/abs/2603.01965 · **PDF:** https://arxiv.org/pdf/2603.01965v1
+
+## 8466. TiledAttention: a CUDA Tile SDPA Kernel for PyTorch
+
+- **ID:** `2603.01960` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Taimur Khan
+- **abs:** https://arxiv.org/abs/2603.01960 · **PDF:** https://arxiv.org/pdf/2603.01960v1
+
+## 8467. The Expressive Limits of Diagonal SSMs for State-Tracking
+
+- **ID:** `2603.01959` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Mehran Shakerinava, Behnoush Khavari, Siamak Ravanbakhsh, et al.
+- **abs:** https://arxiv.org/abs/2603.01959 · **PDF:** https://arxiv.org/pdf/2603.01959v1
+
+## 8468. Closed-Loop Action Chunks with Dynamic Corrections for Training-Free Diffusion Policy
+
+- **ID:** `2603.01953` · **primary:** `cs.RO` · **Submitted:** 2026-03-02
+- **Authors:** Pengyuan Wu, Pingrui Zhang, Zhigang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.01953 · **PDF:** https://arxiv.org/pdf/2603.01953v2
+
+## 8469. LiveCultureBench: a Multi-Agent, Multi-Cultural Benchmark for Large Language Models in Dynamic Social Simulations
+
+- **ID:** `2603.01952` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Viet-Thanh Pham, Lizhen Qu, Thuy-Trang Vu, et al.
+- **abs:** https://arxiv.org/abs/2603.01952 · **PDF:** https://arxiv.org/pdf/2603.01952v1
+
+## 8470. Accelerating Single-Pass SGD for Generalized Linear Prediction
+
+- **ID:** `2603.01951` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Qian Chen, Shihong Ding, Cong Fang
+- **abs:** https://arxiv.org/abs/2603.01951 · **PDF:** https://arxiv.org/pdf/2603.01951v1
+
+## 8471. Semantic Similarity is a Spurious Measure of Comic Understanding: Lessons Learned from Hallucinations in a Benchmarking Experiment
+
+- **ID:** `2603.01950` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Christopher Driggers-Ellis, Nachiketh Tibrewal, Rohit Bogulla, et al.
+- **abs:** https://arxiv.org/abs/2603.01950 · **PDF:** https://arxiv.org/pdf/2603.01950v1
+
+## 8472. Probabilistic Retrofitting of Learned Simulators
+
+- **ID:** `2603.01949` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Cristiana Diaconu, Miles Cranmer, Richard E. Turner, et al.
+- **abs:** https://arxiv.org/abs/2603.01949 · **PDF:** https://arxiv.org/pdf/2603.01949v1
+
+## 8473. physfusion: A Transformer-based Dual-Stream Radar and Vision Fusion Framework for Open Water Surface Object Detection
+
+- **ID:** `2603.01947` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Yuting Wan, Liguo Sun, Jiuwu Hao, et al.
+- **abs:** https://arxiv.org/abs/2603.01947 · **PDF:** https://arxiv.org/pdf/2603.01947v1
+
+## 8474. When Numbers Tell Half the Story: Human-Metric Alignment in Topic Model Evaluation
+
+- **ID:** `2603.01945` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Thibault Prouteau, Francis Lareau, Nicolas Dugué, et al.
+- **abs:** https://arxiv.org/abs/2603.01945 · **PDF:** https://arxiv.org/pdf/2603.01945v1
+
+## 8475. Ignore All Previous Instructions: Jailbreaking as a de-escalatory peace building practise to resist LLM social media bots
+
+- **ID:** `2603.01942` · **primary:** `cs.HC` · **Submitted:** 2026-03-02
+- **Authors:** Huw Day, Adrianna Jezierska, Jessica Woodgate
+- **abs:** https://arxiv.org/abs/2603.01942 · **PDF:** https://arxiv.org/pdf/2603.01942v1
+
+## 8476. BAED: a New Paradigm for Few-shot Graph Learning with Explanation in the Loop
+
+- **ID:** `2603.01941` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Chao Chen, Xujia Li, Dongsheng Hong, et al.
+- **abs:** https://arxiv.org/abs/2603.01941 · **PDF:** https://arxiv.org/pdf/2603.01941v1
+
+## 8477. CoVe: Training Interactive Tool-Use Agents via Constraint-Guided Verification
+
+- **ID:** `2603.01940` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Jinpeng Chen, Cheng Gong, Hanbo Li, et al.
+- **abs:** https://arxiv.org/abs/2603.01940 · **PDF:** https://arxiv.org/pdf/2603.01940v1
+
+## 8478. Explanation-Guided Adversarial Training for Robust and Interpretable Models
+
+- **ID:** `2603.01938` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Chao Chen, Yanhui Chen, Shanshan Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.01938 · **PDF:** https://arxiv.org/pdf/2603.01938v1
+
+## 8479. Dream2Learn: Structured Generative Dreaming for Continual Learning
+
+- **ID:** `2603.01935` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Salvatore Calcagno, Matteo Pennisi, Federica Proietto Salanitri, et al.
+- **abs:** https://arxiv.org/abs/2603.01935 · **PDF:** https://arxiv.org/pdf/2603.01935v1
+
+## 8480. From Variance to Invariance: Qualitative Content Analysis for Narrative Graph Annotation
+
+- **ID:** `2603.01930` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Junbo Huang, Max Weinig, Ulrich Fritsche, et al.
+- **abs:** https://arxiv.org/abs/2603.01930 · **PDF:** https://arxiv.org/pdf/2603.01930v2
+
+## 8481. Bound Propagation meets Constraint Simplification: Improving Logic-based XAI for Neural Networks
+
+- **ID:** `2603.01923` · **primary:** `cs.LO` · **Submitted:** 2026-03-02
+- **Authors:** Ronaldo Gomes, Jairo Ribeiro, Luiz Queiroz, et al.
+- **abs:** https://arxiv.org/abs/2603.01923 · **PDF:** https://arxiv.org/pdf/2603.01923v1
+
+## 8482. Real Money, Fake Models: Deceptive Model Claims in Shadow APIs
+
+- **ID:** `2603.01919` · **primary:** `cs.CR` · **Submitted:** 2026-03-02
+- **Authors:** Yage Zhang, Yukun Jiang, Zeyuan Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.01919 · **PDF:** https://arxiv.org/pdf/2603.01919v2
+
+## 8483. AdaPonderLM: Gated Pondering Language Models with Token-Wise Adaptive Depth
+
+- **ID:** `2603.01914` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Shixiang Song, He Li, Zitong Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.01914 · **PDF:** https://arxiv.org/pdf/2603.01914v2
+
+## 8484. Demonstrating ViviDoc: Generating Interactive Documents through Human-Agent Collaboration
+
+- **ID:** `2603.01912` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Yinghao Tang, Yupeng Xie, Yingchaojie Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.01912 · **PDF:** https://arxiv.org/pdf/2603.01912v1
+
+## 8485. FLANS at SemEval-2026 Task 7: RAG with Open-Sourced Smaller LLMs for Everyday Knowledge Across Diverse Languages and Cultures
+
+- **ID:** `2603.01910` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Liliia Bogdanova, Shiran Sun, Lifeng Han, et al.
+- **abs:** https://arxiv.org/abs/2603.01910 · **PDF:** https://arxiv.org/pdf/2603.01910v1
+
+## 8486. Efficient RLVR Training via Weighted Mutual Information Data Selection
+
+- **ID:** `2603.01907` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Xinyu Zhou, Boyu Zhu, Haotian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.01907 · **PDF:** https://arxiv.org/pdf/2603.01907v1
+
+## 8487. PolyMon: A Unified Framework for Polymer Property Prediction
+
+- **ID:** `2603.13303` · **primary:** `cond-mat.soft` · **Submitted:** 2026-03-02
+- **Authors:** Gaopeng Ren, Yijie Yang, Jiajun Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.13303 · **PDF:** https://arxiv.org/pdf/2603.13303v1
+
+## 8488. Agentic Code Reasoning
+
+- **ID:** `2603.01896` · **primary:** `cs.SE` · **Submitted:** 2026-03-02
+- **Authors:** Shubham Ugare, Satish Chandra
+- **abs:** https://arxiv.org/abs/2603.01896 · **PDF:** https://arxiv.org/pdf/2603.01896v2
+
+## 8489. Machine Learning Models to Identify Promising Nested Antiresonance Nodeless Fiber Designs
+
+- **ID:** `2603.13302` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Rania A. Eltaieb, Sophie LaRochelle, Leslie A. Rusch
+- **abs:** https://arxiv.org/abs/2603.13302 · **PDF:** https://arxiv.org/pdf/2603.13302v1
+
+## 8490. SEAR: Sample Efficient Action Chunking Reinforcement Learning
+
+- **ID:** `2603.01891` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** C. F. Maximilian Nagy, Onur Celik, Emiliyan Gospodinov, et al.
+- **abs:** https://arxiv.org/abs/2603.01891 · **PDF:** https://arxiv.org/pdf/2603.01891v1
+
+## 8491. Diagnosing Generalization Failures from Representational Geometry Markers
+
+- **ID:** `2603.01879` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Chi-Ning Chou, Artem Kirsanov, Yao-Yuan Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.01879 · **PDF:** https://arxiv.org/pdf/2603.01879v1
+
+## 8492. KDFlow: A User-Friendly and Efficient Knowledge Distillation Framework for Large Language Models
+
+- **ID:** `2603.01875` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Songming Zhang, Xue Zhang, Tong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.01875 · **PDF:** https://arxiv.org/pdf/2603.01875v2
+
+## 8493. Phishing the Phishers with SpecularNet: Hierarchical Graph Autoencoding for Reference-Free Web Phishing Detection
+
+- **ID:** `2603.01874` · **primary:** `cs.CR` · **Submitted:** 2026-03-02
+- **Authors:** Tailai Song, Pedro Casas, Michela Meo
+- **abs:** https://arxiv.org/abs/2603.01874 · **PDF:** https://arxiv.org/pdf/2603.01874v1
+
+## 8494. Generalizing Logic-based Explanations for Machine Learning Classifiers via Optimization
+
+- **ID:** `2603.01870` · **primary:** `cs.LO` · **Submitted:** 2026-03-02
+- **Authors:** Francisco Mateus Rocha Filho, Ajalmar Rêgo da Rocha Neto, Thiago Alves Rocha
+- **abs:** https://arxiv.org/abs/2603.01870 · **PDF:** https://arxiv.org/pdf/2603.01870v1
+
+## 8495. Sovereign AI-based Public Services are Viable and Affordable
+
+- **ID:** `2603.01869` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** António Branco, Luís Gomes, Rodrigo Santos, et al.
+- **abs:** https://arxiv.org/abs/2603.01869 · **PDF:** https://arxiv.org/pdf/2603.01869v1
+
+## 8496. Topological Causal Effects
+
+- **ID:** `2603.02289` · **primary:** `stat.ME` · **Submitted:** 2026-03-02
+- **Authors:** Kwangho Kim, Hajin Lee
+- **abs:** https://arxiv.org/abs/2603.02289 · **PDF:** https://arxiv.org/pdf/2603.02289v1
+
+## 8497. CyclicJudge: Mitigating Judge Bias Efficiently in LLM-based Evaluation
+
+- **ID:** `2603.01865` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Ziyi Zhu, Olivier Tieleman, Alexey Bukhtiyarov, et al.
+- **abs:** https://arxiv.org/abs/2603.01865 · **PDF:** https://arxiv.org/pdf/2603.01865v2
+
+## 8498. Tide: A Customisable Dataset Generator for Anti-Money Laundering Research
+
+- **ID:** `2603.01863` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Montijn van den Beukel, Jože Martin Rožanec, Ana-Lucia Varbanescu
+- **abs:** https://arxiv.org/abs/2603.01863 · **PDF:** https://arxiv.org/pdf/2603.01863v1
+
+## 8499. Let the Agent Search: Autonomous Exploration Beats Rigid Workflows in Temporal Question Answering
+
+- **ID:** `2603.01853` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Xufei Lv, Jiahui Yang, Haoyuan Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.01853 · **PDF:** https://arxiv.org/pdf/2603.01853v2
+
+## 8500. Uniform-in-time concentration in two-layer neural networks via transportation inequalities
+
+- **ID:** `2603.01842` · **primary:** `cs.NE` · **Submitted:** 2026-03-02
+- **Authors:** Arnaud Guillin, Boris Nectoux, Paul Stos
+- **abs:** https://arxiv.org/abs/2603.01842 · **PDF:** https://arxiv.org/pdf/2603.01842v1
+
+## 8501. Trivial Graph Features and Classical Learning are Enough to Detect Random Anomalies
+
+- **ID:** `2603.01841` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Matthieu Latapy, Stephany Rajeh
+- **abs:** https://arxiv.org/abs/2603.01841 · **PDF:** https://arxiv.org/pdf/2603.01841v1
+
+## 8502. Constrained Particle Seeking: Solving Diffusion Inverse Problems with Just Forward Passes
+
+- **ID:** `2603.01837` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Hongkun Dou, Zike Chen, Zeyu Li, et al.
+- **abs:** https://arxiv.org/abs/2603.01837 · **PDF:** https://arxiv.org/pdf/2603.01837v1
+
+## 8503. Probing Materials Knowledge in LLMs: From Latent Embeddings to Reliable Predictions
+
+- **ID:** `2603.01834` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-03-02
+- **Authors:** Vineeth Venugopal, Soroush Mahjoubi, Elsa Olivetti
+- **abs:** https://arxiv.org/abs/2603.01834 · **PDF:** https://arxiv.org/pdf/2603.01834v1
+
+## 8504. Uncertainty Quantification of Click and Conversion Estimates for the Autobidding
+
+- **ID:** `2603.01825` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Ivan Zhigalskii, Andrey Pudovikov, Aleksandr Katrutsa, et al.
+- **abs:** https://arxiv.org/abs/2603.01825 · **PDF:** https://arxiv.org/pdf/2603.01825v1
+
+## 8505. OpenAutoNLU: Open Source AutoML Library for NLU
+
+- **ID:** `2603.01824` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Grigory Arshinov, Aleksandr Boriskin, Sergey Senichev, et al.
+- **abs:** https://arxiv.org/abs/2603.01824 · **PDF:** https://arxiv.org/pdf/2603.01824v1
+
+## 8506. Emerging Human-like Strategies for Semantic Memory Foraging in Large Language Models
+
+- **ID:** `2603.01822` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Eric Lacosse, Mariana Duarte, Peter M. Todd, et al.
+- **abs:** https://arxiv.org/abs/2603.01822 · **PDF:** https://arxiv.org/pdf/2603.01822v1
+
+## 8507. Deep Learning for Financial Time Series: A Large-Scale Benchmark of Risk-Adjusted Performance
+
+- **ID:** `2603.01820` · **primary:** `q-fin.TR` · **Submitted:** 2026-03-02
+- **Authors:** Adir Saly-Kaufmann, Kieran Wood, Jan Peter-Calliess, et al.
+- **abs:** https://arxiv.org/abs/2603.01820 · **PDF:** https://arxiv.org/pdf/2603.01820v1
+
+## 8508. CRoCoDiL: Continuous and Robust Conditioned Diffusion for Language
+
+- **ID:** `2603.20210` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Roy Uziel, Omer Belhasin, Itay Levi, et al.
+- **abs:** https://arxiv.org/abs/2603.20210 · **PDF:** https://arxiv.org/pdf/2603.20210v2
+
+## 8509. GCTAM: Global and Contextual Truncated Affinity Combined Maximization Model For Unsupervised Graph Anomaly Detection
+
+- **ID:** `2603.01806` · **primary:** `cs.SI` · **Submitted:** 2026-03-02
+- **Authors:** Xiong Zhang, Hong Peng, Zhenli He, et al.
+- **abs:** https://arxiv.org/abs/2603.01806 · **PDF:** https://arxiv.org/pdf/2603.01806v1
+
+## 8510. Non-verbal Real-time Human-AI Interaction in Constrained Robotic Environments
+
+- **ID:** `2603.01804` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Dragos Costea, Alina Marcu, Cristina Lazar, et al.
+- **abs:** https://arxiv.org/abs/2603.01804 · **PDF:** https://arxiv.org/pdf/2603.01804v1
+
+## 8511. What Papers Don't Tell You: Recovering Tacit Knowledge for Automated Paper Reproduction
+
+- **ID:** `2603.01801` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Lehui Li, Ruining Wang, Haochen Song, et al.
+- **abs:** https://arxiv.org/abs/2603.01801 · **PDF:** https://arxiv.org/pdf/2603.01801v1
+
+## 8512. Phase-Type Variational Autoencoders for Heavy-Tailed Data
+
+- **ID:** `2603.01800` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Abdelhakim Ziani, András Horváth, Paolo Ballarini
+- **abs:** https://arxiv.org/abs/2603.01800 · **PDF:** https://arxiv.org/pdf/2603.01800v1
+
+## 8513. Incremental, inconsistency-resilient reasoning over Description Logic Abox streams
+
+- **ID:** `2603.01799` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Cas Proost, Pieter Bonte
+- **abs:** https://arxiv.org/abs/2603.01799 · **PDF:** https://arxiv.org/pdf/2603.01799v1
+
+## 8514. PleaSQLarify: Visual Pragmatic Repair for Natural Language Database Querying
+
+- **ID:** `2603.01795` · **primary:** `cs.HC` · **Submitted:** 2026-03-02
+- **Authors:** Robin Shing Moon Chan, Rita Sevastjanova, Mennatallah El-Assady
+- **abs:** https://arxiv.org/abs/2603.01795 · **PDF:** https://arxiv.org/pdf/2603.01795v1
+
+## 8515. ALTER: Asymmetric LoRA for Token-Entropy-Guided Unlearning of LLMs
+
+- **ID:** `2603.01792` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Xunlei Chen, Jinyu Guo, Yuang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.01792 · **PDF:** https://arxiv.org/pdf/2603.01792v1
+
+## 8516. Semantic Novelty Trajectories in 80,000 Books: A Cross-Corpus Embedding Analysis
+
+- **ID:** `2603.01791` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Fred Zimmerman
+- **abs:** https://arxiv.org/abs/2603.01791 · **PDF:** https://arxiv.org/pdf/2603.01791v1
+
+## 8517. nchellwig at SemEval-2026 Task 3: Self-Consistent Structured Generation (SCSG) for Dimensional Aspect-Based Sentiment Analysis using Large Language Models
+
+- **ID:** `2603.01788` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Nils Constantin Hellwig, Jakob Fehle, Udo Kruschwitz, et al.
+- **abs:** https://arxiv.org/abs/2603.01788 · **PDF:** https://arxiv.org/pdf/2603.01788v1
+
+## 8518. Learning Shortest Paths with Generative Flow Networks
+
+- **ID:** `2603.01786` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Nikita Morozov, Ian Maksimov, Daniil Tiapkin, et al.
+- **abs:** https://arxiv.org/abs/2603.01786 · **PDF:** https://arxiv.org/pdf/2603.01786v1
+
+## 8519. Co-Evolutionary Multi-Modal Alignment via Structured Adversarial Evolution
+
+- **ID:** `2603.01784` · **primary:** `cs.CR` · **Submitted:** 2026-03-02
+- **Authors:** Guoxin Shi, Haoyu Wang, Zaihui Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.01784 · **PDF:** https://arxiv.org/pdf/2603.01784v1
+
+## 8520. Beyond Prompt Degradation: Prototype-guided Dual-pool Prompting for Incremental Object Detection
+
+- **ID:** `2603.02286` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Yaoteng Zhang, Zhou Qing, Junyu Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.02286 · **PDF:** https://arxiv.org/pdf/2603.02286v1
+
+## 8521. GAM-RAG: Gain-Adaptive Memory for Evolving Retrieval in Retrieval-Augmented Generation
+
+- **ID:** `2603.01783` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Yifan Wang, Mingxuan Jiang, Zhihao Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.01783 · **PDF:** https://arxiv.org/pdf/2603.01783v1
+
+## 8522. D3LM: A Discrete DNA Diffusion Language Model for Bidirectional DNA Understanding and Generation
+
+- **ID:** `2603.01780` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Zhao Yang, Hengchang Liu, Chuan Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.01780 · **PDF:** https://arxiv.org/pdf/2603.01780v1
+
+## 8523. LLM-as-an-Annotator: Training Lightweight Models with LLM-Annotated Examples for Aspect Sentiment Tuple Prediction
+
+- **ID:** `2603.01778` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Nils Constantin Hellwig, Jakob Fehle, Udo Kruschwitz, et al.
+- **abs:** https://arxiv.org/abs/2603.01778 · **PDF:** https://arxiv.org/pdf/2603.01778v1
+
+## 8524. FreeAct: Freeing Activations for LLM Quantization
+
+- **ID:** `2603.01776` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Xiaohao Liu, Xiaobo Xia, Manyi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.01776 · **PDF:** https://arxiv.org/pdf/2603.01776v2
+
+## 8525. Beyond the Resumé: A Rubric-Aware Automatic Interview System for Information Elicitation
+
+- **ID:** `2603.01775` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Harry Stuart, Masahiro Kaneko, Timothy Baldwin
+- **abs:** https://arxiv.org/abs/2603.01775 · **PDF:** https://arxiv.org/pdf/2603.01775v1
+
+## 8526. AnnoABSA: A Web-Based Annotation Tool for Aspect-Based Sentiment Analysis with Retrieval-Augmented Suggestions
+
+- **ID:** `2603.01773` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Nils Constantin Hellwig, Jakob Fehle, Udo Kruschwitz, et al.
+- **abs:** https://arxiv.org/abs/2603.01773 · **PDF:** https://arxiv.org/pdf/2603.01773v1
+
+## 8527. Hyperparameter Trajectory Inference with Conditional Lagrangian Optimal Transport
+
+- **ID:** `2603.01771` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Harry Amad, Mihaela van der Schaar
+- **abs:** https://arxiv.org/abs/2603.01771 · **PDF:** https://arxiv.org/pdf/2603.01771v3
+
+## 8528. CHLU: The Causal Hamiltonian Learning Unit as a Symplectic Primitive for Deep Learning
+
+- **ID:** `2603.01768` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Pratik Jawahar, Maurizio Pierini
+- **abs:** https://arxiv.org/abs/2603.01768 · **PDF:** https://arxiv.org/pdf/2603.01768v2
+
+## 8529. DGNet: Discrete Green Networks for Data-Efficient Learning of Spatiotemporal PDEs
+
+- **ID:** `2603.01762` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Yingjie Tan, Quanming Yao, Yaqing Wang
+- **abs:** https://arxiv.org/abs/2603.01762 · **PDF:** https://arxiv.org/pdf/2603.01762v1
+
+## 8530. Modular Memory is the Key to Continual Learning Agents
+
+- **ID:** `2603.01761` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Vaggelis Dorovatas, Malte Schwerin, Andrew D. Bagdanov, et al.
+- **abs:** https://arxiv.org/abs/2603.01761 · **PDF:** https://arxiv.org/pdf/2603.01761v1
+
+## 8531. Meta-Learning Hyperparameters for Parameter Efficient Fine-Tuning
+
+- **ID:** `2603.01759` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Zichen Tian, Yaoyao Liu, Qianru Sun
+- **abs:** https://arxiv.org/abs/2603.01759 · **PDF:** https://arxiv.org/pdf/2603.01759v1
+
+## 8532. Federated Agentic AI for Wireless Networks: Fundamentals, Approaches, and Applications
+
+- **ID:** `2603.01755` · **primary:** `cs.NI` · **Submitted:** 2026-03-02
+- **Authors:** Lingyi Cai, Yu Zhang, Ruichen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.01755 · **PDF:** https://arxiv.org/pdf/2603.01755v1
+
+## 8533. Causal Circuit Tracing Reveals Distinct Computational Architectures in Single-Cell Foundation Models: Inhibitory Dominance, Biological Coherence, and Cross-Model Convergence
+
+- **ID:** `2603.01752` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Ihor Kendiukhov
+- **abs:** https://arxiv.org/abs/2603.01752 · **PDF:** https://arxiv.org/pdf/2603.01752v2
+
+## 8534. Shape-Interpretable Visual Self-Modeling Enables Geometry-Aware Continuum Robot Control
+
+- **ID:** `2603.01751` · **primary:** `cs.RO` · **Submitted:** 2026-03-02
+- **Authors:** Peng Yu, Xin Wang, Ning Tan
+- **abs:** https://arxiv.org/abs/2603.01751 · **PDF:** https://arxiv.org/pdf/2603.01751v1
+
+## 8535. Practical Deep Heteroskedastic Regression
+
+- **ID:** `2603.01750` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Mikkel Jordahn, Jonas Vestergaard Jensen, James Harrison, et al.
+- **abs:** https://arxiv.org/abs/2603.01750 · **PDF:** https://arxiv.org/pdf/2603.01750v1
+
+## 8536. Discrete World Models via Regularization
+
+- **ID:** `2603.01748` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Davide Bizzaro, Luciano Serafini
+- **abs:** https://arxiv.org/abs/2603.01748 · **PDF:** https://arxiv.org/pdf/2603.01748v1
+
+## 8537. An Analysis of Multi-Task Architectures for the Hierarchic Multi-Label Problem of Vehicle Model and Make Classification
+
+- **ID:** `2603.01746` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Alexandru Manole, Laura Diosan
+- **abs:** https://arxiv.org/abs/2603.01746 · **PDF:** https://arxiv.org/pdf/2603.01746v1
+
+## 8538. Sequence-Level Unsupervised Training in Speech Recognition: A Theoretical Study
+
+- **ID:** `2603.02285` · **primary:** `cs.SD` · **Submitted:** 2026-03-02
+- **Authors:** Zijian Yang, Jörg Barkoczi, Ralf Schlüter, et al.
+- **abs:** https://arxiv.org/abs/2603.02285 · **PDF:** https://arxiv.org/pdf/2603.02285v1
+
+## 8539. Rethinking Policy Diversity in Ensemble Policy Gradient in Large-Scale Reinforcement Learning
+
+- **ID:** `2603.01741` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Naoki Shitanda, Motoki Omura, Tatsuya Harada, et al.
+- **abs:** https://arxiv.org/abs/2603.01741 · **PDF:** https://arxiv.org/pdf/2603.01741v2
+
+## 8540. CA-AFP: Cluster-Aware Adaptive Federated Pruning
+
+- **ID:** `2603.01739` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Om Govind Jha, Harsh Shukla, Haroon R. Lone
+- **abs:** https://arxiv.org/abs/2603.01739 · **PDF:** https://arxiv.org/pdf/2603.01739v1
+
+## 8541. Bootstrapping Embeddings for Low Resource Languages
+
+- **ID:** `2603.01732` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Merve Basoz, Andrew Horne, Mattia Opper
+- **abs:** https://arxiv.org/abs/2603.01732 · **PDF:** https://arxiv.org/pdf/2603.01732v2
+
+## 8542. Solving Inverse PDE Problems using Minimization Methods and AI
+
+- **ID:** `2603.01731` · **primary:** `math.NA` · **Submitted:** 2026-03-02
+- **Authors:** Noura Al Helwani, Sophie Moufawad, Georges Sakr
+- **abs:** https://arxiv.org/abs/2603.01731 · **PDF:** https://arxiv.org/pdf/2603.01731v2
+
+## 8543. Decentralized Federated Learning by Partial Message Exchange
+
+- **ID:** `2603.01730` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Shan Sha, Shenglong Zhou, Xin Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.01730 · **PDF:** https://arxiv.org/pdf/2603.01730v1
+
+## 8544. GMP: A Benchmark for Content Moderation under Co-occurring Violations and Dynamic Rules
+
+- **ID:** `2603.01724` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Houde Dong, Yifei She, Kai Ye, et al.
+- **abs:** https://arxiv.org/abs/2603.01724 · **PDF:** https://arxiv.org/pdf/2603.01724v1
+
+## 8545. Co-optimization for Adaptive Conformal Prediction
+
+- **ID:** `2603.01719` · **primary:** `stat.ML` · **Submitted:** 2026-03-02
+- **Authors:** Xiaoyi Su, Zhixin Zhou, Rui Luo
+- **abs:** https://arxiv.org/abs/2603.01719 · **PDF:** https://arxiv.org/pdf/2603.01719v1
+
+## 8546. Not All Queries Need Rewriting: When Prompt-Only LLM Refinement Helps and Hurts Dense Retrieval
+
+- **ID:** `2603.13301` · **primary:** `cs.IR` · **Submitted:** 2026-03-02
+- **Authors:** Varun Kotte
+- **abs:** https://arxiv.org/abs/2603.13301 · **PDF:** https://arxiv.org/pdf/2603.13301v1
+
+## 8547. TopoCurate:Modeling Interaction Topology for Tool-Use Agent Training
+
+- **ID:** `2603.01714` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Jinluan Yang, Yuxin Liu, Zhengyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.01714 · **PDF:** https://arxiv.org/pdf/2603.01714v1
+
+## 8548. FT-Dojo: Towards Autonomous LLM Fine-Tuning with Language Agents
+
+- **ID:** `2603.01712` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Qizheng Li, Yifei Zhang, Xiao Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.01712 · **PDF:** https://arxiv.org/pdf/2603.01712v1
+
+## 8549. Legal RAG Bench: an end-to-end benchmark for legal RAG
+
+- **ID:** `2603.01710` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Abdur-Rahman Butler, Umar Butler
+- **abs:** https://arxiv.org/abs/2603.01710 · **PDF:** https://arxiv.org/pdf/2603.01710v1
+
+## 8550. Better Eyes, Better Thoughts: Why Vision Chain-of-Thought Fails in Medicine
+
+- **ID:** `2603.06665` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Yuan Wu, Zongxian Yang, Jiayu Qian, et al.
+- **abs:** https://arxiv.org/abs/2603.06665 · **PDF:** https://arxiv.org/pdf/2603.06665v1
+
+## 8551. Search Multilayer Perceptron-Based Fusion for Efficient and Accurate Siamese Tracking
+
+- **ID:** `2603.01706` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Tianqi Shen, Huakao Lin, Ning An
+- **abs:** https://arxiv.org/abs/2603.01706 · **PDF:** https://arxiv.org/pdf/2603.01706v1
+
+## 8552. Security Risks in Machining Process Monitoring: Sequence-to-Sequence Learning for Reconstruction of CNC Axis Positions
+
+- **ID:** `2603.01702` · **primary:** `cs.AR` · **Submitted:** 2026-03-02
+- **Authors:** Lukas Krupp, Rickmar Stahlschmidt, Norbert Wehn
+- **abs:** https://arxiv.org/abs/2603.01702 · **PDF:** https://arxiv.org/pdf/2603.01702v1
+
+## 8553. Towards Principled Dataset Distillation: A Spectral Distribution Perspective
+
+- **ID:** `2603.01698` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Ruixi Wu, Shaobo Wang, Jiahuan Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.01698 · **PDF:** https://arxiv.org/pdf/2603.01698v1
+
+## 8554. DynaMoE: Dynamic Token-Level Expert Activation with Layer-Wise Adaptive Capacity for Mixture-of-Experts Neural Networks
+
+- **ID:** `2603.01697` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Gökdeniz Gülmez
+- **abs:** https://arxiv.org/abs/2603.01697 · **PDF:** https://arxiv.org/pdf/2603.01697v1
+
+## 8555. Cross-modal Identity Mapping: Minimizing Information Loss in Modality Conversion via Reinforcement Learning
+
+- **ID:** `2603.01696` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Haonan Jia, Shichao Dong, Xin Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.01696 · **PDF:** https://arxiv.org/pdf/2603.01696v1
+
+## 8556. Streaming Continual Learning for Unified Adaptive Intelligence in Dynamic Environments
+
+- **ID:** `2603.01695` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Federico Giannini, Giacomo Ziffer, Andrea Cossu, et al.
+- **abs:** https://arxiv.org/abs/2603.01695 · **PDF:** https://arxiv.org/pdf/2603.01695v1
+
+## 8557. MVR: Multi-view Video Reward Shaping for Reinforcement Learning
+
+- **ID:** `2603.01694` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Lirui Luo, Guoxi Zhang, Hongming Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.01694 · **PDF:** https://arxiv.org/pdf/2603.01694v1
+
+## 8558. Reasoning as Gradient: Scaling MLE Agents Beyond Tree Search
+
+- **ID:** `2603.01692` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Yifei Zhang, Xu Yang, Xiao Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.01692 · **PDF:** https://arxiv.org/pdf/2603.01692v2
+
+## 8559. Building a Strong Instruction Language Model for a Less-Resourced Language
+
+- **ID:** `2603.01691` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Domen Vreš, Tjaša Arčon, Timotej Petrič, et al.
+- **abs:** https://arxiv.org/abs/2603.01691 · **PDF:** https://arxiv.org/pdf/2603.01691v1
+
+## 8560. Accelerating Video Generation Inference with Sequential-Parallel 3D Positional Encoding Using a Global Time Index
+
+- **ID:** `2603.06664` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Chao Yuan, Pan Li
+- **abs:** https://arxiv.org/abs/2603.06664 · **PDF:** https://arxiv.org/pdf/2603.06664v1
+
+## 8561. QIME: Constructing Interpretable Medical Text Embeddings via Ontology-Grounded Questions
+
+- **ID:** `2603.01690` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Yixuan Tang, Zhenghong Lin, Yandong Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.01690 · **PDF:** https://arxiv.org/pdf/2603.01690v2
+
+## 8562. Randomized Neural Networks for Partial Differential Equation on Static and Evolving Surfaces
+
+- **ID:** `2603.01689` · **primary:** `math.NA` · **Submitted:** 2026-03-02
+- **Authors:** Jingbo Sun, Fei Wang
+- **abs:** https://arxiv.org/abs/2603.01689 · **PDF:** https://arxiv.org/pdf/2603.01689v1
+
+## 8563. Surgical Post-Training: Cutting Errors, Keeping Knowledge
+
+- **ID:** `2603.01683` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Wenye Lin, Kai Han
+- **abs:** https://arxiv.org/abs/2603.01683 · **PDF:** https://arxiv.org/pdf/2603.01683v1
+
+## 8564. A Practical Guide to Streaming Continual Learning
+
+- **ID:** `2603.01677` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Andrea Cossu, Federico Giannini, Giacomo Ziffer, et al.
+- **abs:** https://arxiv.org/abs/2603.01677 · **PDF:** https://arxiv.org/pdf/2603.01677v1
+
+## 8565. Statistical Consistency of Discrete-to-Continuous Limits of Determinantal Point Processes
+
+- **ID:** `2603.01670` · **primary:** `math.PR` · **Submitted:** 2026-03-02
+- **Authors:** Hugo Jaquard, Nicolas Keriven
+- **abs:** https://arxiv.org/abs/2603.01670 · **PDF:** https://arxiv.org/pdf/2603.01670v1
+
+## 8566. Chain-of-Context Learning: Dynamic Constraint Understanding for Multi-Task VRPs
+
+- **ID:** `2603.01667` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Shuangchun Gui, Suyu Liu, Xuehe Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.01667 · **PDF:** https://arxiv.org/pdf/2603.01667v1
+
+## 8567. Beyond the Grid: Layout-Informed Multi-Vector Retrieval with Parsed Visual Document Representations
+
+- **ID:** `2603.01666` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Yibo Yan, Mingdong Ou, Yi Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.01666 · **PDF:** https://arxiv.org/pdf/2603.01666v1
+
+## 8568. FreeGNN: Continual Source-Free Graph Neural Network Adaptation for Renewable Energy Forecasting
+
+- **ID:** `2603.01657` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Abderaouf Bahi, Amel Ourici, Ibtissem Gasmi, et al.
+- **abs:** https://arxiv.org/abs/2603.01657 · **PDF:** https://arxiv.org/pdf/2603.01657v1
+
+## 8569. Graph-of-Mark: Promote Spatial Reasoning in Multimodal Language Models with Graph-Based Visual Prompting
+
+- **ID:** `2603.06663` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Giacomo Frisoni, Lorenzo Molfetta, Mattia Buzzoni, et al.
+- **abs:** https://arxiv.org/abs/2603.06663 · **PDF:** https://arxiv.org/pdf/2603.06663v2
+
+## 8570. Transform-Invariant Generative Ray Path Sampling for Efficient Radio Propagation Modeling
+
+- **ID:** `2603.01655` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Jérome Eertmans, Enrico M. Vitucci, Vittorio Degli-Esposti, et al.
+- **abs:** https://arxiv.org/abs/2603.01655 · **PDF:** https://arxiv.org/pdf/2603.01655v1
+
+## 8571. CeProAgents: A Hierarchical Agents System for Automated Chemical Process Development
+
+- **ID:** `2603.01654` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Yuhang Yang, Ruikang Li, Jifei Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.01654 · **PDF:** https://arxiv.org/pdf/2603.01654v1
+
+## 8572. LexChronos: An Agentic Framework for Structured Event Timeline Extraction in Indian Jurisprudence
+
+- **ID:** `2603.01651` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Anka Chandrahas Tummepalli, Preethu Rose Anish
+- **abs:** https://arxiv.org/abs/2603.01651 · **PDF:** https://arxiv.org/pdf/2603.01651v1
+
+## 8573. Learning Structured Reasoning via Tractable Trajectory Control
+
+- **ID:** `2603.01641` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Po-Nien Kung, Zhen Yang, Jeffrey Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.01641 · **PDF:** https://arxiv.org/pdf/2603.01641v1
+
+## 8574. Learning to Draft: Adaptive Speculative Decoding with Reinforcement Learning
+
+- **ID:** `2603.01639` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Jiebin Zhang, Zhenghan Yu, Liang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.01639 · **PDF:** https://arxiv.org/pdf/2603.01639v1
+
+## 8575. DeLo: Dual Decomposed Low-Rank Experts Collaboration for Continual Missing Modality Learning
+
+- **ID:** `2603.01632` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Xiwei Liu, Yulong Li, Feilong Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.01632 · **PDF:** https://arxiv.org/pdf/2603.01632v1
+
+## 8576. SEED-SET: Scalable Evolving Experimental Design for System-level Ethical Testing
+
+- **ID:** `2603.01630` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Anjali Parashar, Yingke Li, Eric Yang Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.01630 · **PDF:** https://arxiv.org/pdf/2603.01630v2
+
+## 8577. Towards OOD Generalization in Dynamic Graphs via Causal Invariant Learning
+
+- **ID:** `2603.01626` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Xinxun Zhang, Pengfei Jiao, Mengzhou Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.01626 · **PDF:** https://arxiv.org/pdf/2603.01626v1
+
+## 8578. Measuring What VLMs Don't Say: Validation Metrics Hide Clinical Terminology Erasure in Radiology Report Generation
+
+- **ID:** `2603.01625` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Aditya Parikh, Aasa Feragen, Sneha Das, et al.
+- **abs:** https://arxiv.org/abs/2603.01625 · **PDF:** https://arxiv.org/pdf/2603.01625v1
+
+## 8579. Assessing Crime Disclosure Patterns in a Large-Scale Cybercrime Forum
+
+- **ID:** `2603.01624` · **primary:** `cs.CY` · **Submitted:** 2026-03-02
+- **Authors:** Raphael Hoheisel, Tom Meurs, Jai Wientjes, et al.
+- **abs:** https://arxiv.org/abs/2603.01624 · **PDF:** https://arxiv.org/pdf/2603.01624v1
+
+## 8580. Adaptive Spectral Feature Forecasting for Diffusion Sampling Acceleration
+
+- **ID:** `2603.01623` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Jiaqi Han, Juntong Shi, Puheng Li, et al.
+- **abs:** https://arxiv.org/abs/2603.01623 · **PDF:** https://arxiv.org/pdf/2603.01623v1
+
+## 8581. More Data, Fewer Diacritics: Scaling Arabic TTS
+
+- **ID:** `2603.01622` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Ahmed Musleh, Yifan Zhang, Kareem Darwish
+- **abs:** https://arxiv.org/abs/2603.01622 · **PDF:** https://arxiv.org/pdf/2603.01622v1
+
+## 8582. ToolRLA: Multiplicative Reward Decomposition for Tool-Integrated Agents
+
+- **ID:** `2603.01620` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Pengbo Liu
+- **abs:** https://arxiv.org/abs/2603.01620 · **PDF:** https://arxiv.org/pdf/2603.01620v4
+
+## 8583. Evaluating and Understanding Scheming Propensity in LLM Agents
+
+- **ID:** `2603.01608` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Mia Hopman, Jannes Elstner, Maria Avramidou, et al.
+- **abs:** https://arxiv.org/abs/2603.01608 · **PDF:** https://arxiv.org/pdf/2603.01608v2
+
+## 8584. CARE: Towards Clinical Accountability in Multi-Modal Medical Reasoning with an Evidence-Grounded Agentic Framework
+
+- **ID:** `2603.01607` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Yuexi Du, Jinglu Wang, Shujie Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.01607 · **PDF:** https://arxiv.org/pdf/2603.01607v2
+
+## 8585. What Helps---and What Hurts: Bidirectional Explanations for Vision Transformers
+
+- **ID:** `2603.01605` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Qin Su, Tie Luo
+- **abs:** https://arxiv.org/abs/2603.01605 · **PDF:** https://arxiv.org/pdf/2603.01605v1
+
+## 8586. YCDa: YCbCr Decoupled Attention for Real-time Realistic Camouflaged Object Detection
+
+- **ID:** `2603.01602` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** PeiHuang Zheng, Yunlong Zhao, Zheng Cui, et al.
+- **abs:** https://arxiv.org/abs/2603.01602 · **PDF:** https://arxiv.org/pdf/2603.01602v1
+
+## 8587. Boosting Entropy with Bell Box Quantization
+
+- **ID:** `2603.01599` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Ningfeng Yang, Tor M. Aamodt
+- **abs:** https://arxiv.org/abs/2603.01599 · **PDF:** https://arxiv.org/pdf/2603.01599v1
+
+## 8588. FAST-DIPS: Adjoint-Free Analytic Steps and Hard-Constrained Likelihood Correction for Diffusion-Prior Inverse Problems
+
+- **ID:** `2603.01591` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Minwoo Kim, Seunghyeok Shin, Hongki Lim
+- **abs:** https://arxiv.org/abs/2603.01591 · **PDF:** https://arxiv.org/pdf/2603.01591v1
+
+## 8589. IDProxy: Cold-Start CTR Prediction for Ads and Recommendation at Xiaohongshu with Multimodal LLMs
+
+- **ID:** `2603.01590` · **primary:** `cs.IR` · **Submitted:** 2026-03-02
+- **Authors:** Yubin Zhang, Haiming Xu, Guillaume Salha-Galvan, et al.
+- **abs:** https://arxiv.org/abs/2603.01590 · **PDF:** https://arxiv.org/pdf/2603.01590v1
+
+## 8590. SafeSci: Safety Evaluation of Large Language Models in Science Domains and Beyond
+
+- **ID:** `2603.01589` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Xiangyang Zhu, Yuan Tian, Qi Jia, et al.
+- **abs:** https://arxiv.org/abs/2603.01589 · **PDF:** https://arxiv.org/pdf/2603.01589v1
+
+## 8591. Jump Like A Squirrel: Optimized Execution Step Order for Anytime Random Forest Inference
+
+- **ID:** `2603.01588` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Daniel Biebert, Christian Hakert, Kay Heider, et al.
+- **abs:** https://arxiv.org/abs/2603.01588 · **PDF:** https://arxiv.org/pdf/2603.01588v1
+
+## 8592. KERV: Kinematic-Rectified Speculative Decoding for Embodied VLA Models
+
+- **ID:** `2603.01581` · **primary:** `cs.RO` · **Submitted:** 2026-03-02
+- **Authors:** Zihao Zheng, Zhihao Mao, Maoliang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.01581 · **PDF:** https://arxiv.org/pdf/2603.01581v1
+
+## 8593. Markovian ODE-guided scoring can assess the quality of offline reasoning traces in language models
+
+- **ID:** `2603.01580` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Arghodeep Nandi, Ojasva Saxena, Tanmoy Chakraborty
+- **abs:** https://arxiv.org/abs/2603.01580 · **PDF:** https://arxiv.org/pdf/2603.01580v1
+
+## 8594. SkeleGuide: Explicit Skeleton Reasoning for Context-Aware Human-in-Place Image Synthesis
+
+- **ID:** `2603.01579` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Chuqiao Wu, Jin Song, Yiyun Fei
+- **abs:** https://arxiv.org/abs/2603.01579 · **PDF:** https://arxiv.org/pdf/2603.01579v1
+
+## 8595. DualSentinel: A Lightweight Framework for Detecting Targeted Attacks in Black-box LLM via Dual Entropy Lull Pattern
+
+- **ID:** `2603.01574` · **primary:** `cs.CR` · **Submitted:** 2026-03-02
+- **Authors:** Xiaoyi Pang, Xuanyi Hao, Pengyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.01574 · **PDF:** https://arxiv.org/pdf/2603.01574v1
+
+## 8596. Beyond Length Scaling: Synergizing Breadth and Depth for Generative Reward Models
+
+- **ID:** `2603.01571` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Qiyuan Zhang, Yufei Wang, Tianhe Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.01571 · **PDF:** https://arxiv.org/pdf/2603.01571v1
+
+## 8597. Adversarial Query Synthesis via Bayesian Optimization
+
+- **ID:** `2603.01570` · **primary:** `cs.DB` · **Submitted:** 2026-03-02
+- **Authors:** Jeffrey Tao, Yimeng Zeng, Haydn Thomas Jones, et al.
+- **abs:** https://arxiv.org/abs/2603.01570 · **PDF:** https://arxiv.org/pdf/2603.01570v1
+
+## 8598. Rate-Distortion Signatures of Generalization and Information Trade-offs
+
+- **ID:** `2603.01568` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Leyla Roksan Caglar, Pedro A. M. Mediano, Baihan Lin
+- **abs:** https://arxiv.org/abs/2603.01568 · **PDF:** https://arxiv.org/pdf/2603.01568v1
+
+## 8599. LFPO: Likelihood-Free Policy Optimization for Masked Diffusion Models
+
+- **ID:** `2603.01563` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Chenxing Wei, Jiazhen Kang, Hong Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.01563 · **PDF:** https://arxiv.org/pdf/2603.01563v1
+
+## 8600. RubricBench: Aligning Model-Generated Rubrics with Human Standards
+
+- **ID:** `2603.01562` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Qiyuan Zhang, Junyi Zhou, Yufei Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.01562 · **PDF:** https://arxiv.org/pdf/2603.01562v2
+
+## 8601. Benchmarking LLM Summaries of Multimodal Clinical Time Series for Remote Monitoring
+
+- **ID:** `2603.01557` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Aditya Shukla, Yining Yuan, Ben Tamo, et al.
+- **abs:** https://arxiv.org/abs/2603.01557 · **PDF:** https://arxiv.org/pdf/2603.01557v1
+
+## 8602. Safety-Guided Flow (SGF): A Unified Framework for Negative Guidance in Safe Generation
+
+- **ID:** `2603.13300` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Mingyu Kim, Young-Heon Kim, Mijung Park
+- **abs:** https://arxiv.org/abs/2603.13300 · **PDF:** https://arxiv.org/pdf/2603.13300v1
+
+## 8603. S5-HES Agent: Society 5.0-driven Agentic Framework to Democratize Smart Home Environment Simulation
+
+- **ID:** `2603.01554` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Akila Siriweera, Janani Rangila, Keitaro Naruse, et al.
+- **abs:** https://arxiv.org/abs/2603.01554 · **PDF:** https://arxiv.org/pdf/2603.01554v1
+
+## 8604. Large Language Models as Bidding Agents in Repeated HetNet Auction
+
+- **ID:** `2603.04455` · **primary:** `cs.NI` · **Submitted:** 2026-03-02
+- **Authors:** Ismail Lotfi, Ali Ghrayeb, Samson Lasaulce, et al.
+- **abs:** https://arxiv.org/abs/2603.04455 · **PDF:** https://arxiv.org/pdf/2603.04455v1
+
+## 8605. State-Action Inpainting Diffuser for Continuous Control with Delay
+
+- **ID:** `2603.01553` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Dongqi Han, Wei Wang, Enze Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.01553 · **PDF:** https://arxiv.org/pdf/2603.01553v1
+
+## 8606. Extracting Training Dialogue Data from Large Language Model based Task Bots
+
+- **ID:** `2603.01550` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Shuo Zhang, Junzhou Zhao, Junji Hou, et al.
+- **abs:** https://arxiv.org/abs/2603.01550 · **PDF:** https://arxiv.org/pdf/2603.01550v1
+
+## 8607. Pri4R: Learning World Dynamics for Vision-Language-Action Models with Privileged 4D Representation
+
+- **ID:** `2603.01549` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Jisoo Kim, Jungbin Cho, Sanghyeok Chu, et al.
+- **abs:** https://arxiv.org/abs/2603.01549 · **PDF:** https://arxiv.org/pdf/2603.01549v2
+
+## 8608. Graph-Based Self-Healing Tool Routing for Cost-Efficient LLM Agents
+
+- **ID:** `2603.01548` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Neeraj Bholani
+- **abs:** https://arxiv.org/abs/2603.01548 · **PDF:** https://arxiv.org/pdf/2603.01548v1
+
+## 8609. Pharmacology Knowledge Graphs: Do We Need Chemical Structure for Drug Repurposing?
+
+- **ID:** `2603.01537` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Youssef Abo-Dahab, Ruby Hernandez, Ismael Caleb Arechiga Duran
+- **abs:** https://arxiv.org/abs/2603.01537 · **PDF:** https://arxiv.org/pdf/2603.01537v1
+
+## 8610. Scalable Multi-Task Low-Rank Model Adaptation
+
+- **ID:** `2603.01526` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Zichen Tian, Antoine Ledent, Qianru Sun
+- **abs:** https://arxiv.org/abs/2603.01526 · **PDF:** https://arxiv.org/pdf/2603.01526v1
+
+## 8611. Children's Intelligence Tests Pose Challenges for MLLMs? KidGym: A 2D Grid-Based Reasoning Benchmark for MLLMs
+
+- **ID:** `2603.20209` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Hengwei Ye, Yuanting Guan, Yuxuan Ge, et al.
+- **abs:** https://arxiv.org/abs/2603.20209 · **PDF:** https://arxiv.org/pdf/2603.20209v3
+
+## 8612. Training Dynamics of Softmax Self-Attention: Fast Global Convergence via Preconditioning
+
+- **ID:** `2603.01514` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Gautam Goel, Mahdi Soltanolkotabi, Peter Bartlett
+- **abs:** https://arxiv.org/abs/2603.01514 · **PDF:** https://arxiv.org/pdf/2603.01514v1
+
+## 8613. Multimodal Mixture-of-Experts with Retrieval Augmentation for Protein Active Site Identification
+
+- **ID:** `2603.01511` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Jiayang Wu, Jiale Zhou, Rubo Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.01511 · **PDF:** https://arxiv.org/pdf/2603.01511v2
+
+## 8614. Retrieval, Refinement, and Ranking for Text-to-Video Generation via Prompt Optimization and Test-Time Scaling
+
+- **ID:** `2603.01509` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Zillur Rahman, Alex Sheng, Cristian Meo
+- **abs:** https://arxiv.org/abs/2603.01509 · **PDF:** https://arxiv.org/pdf/2603.01509v1
+
+## 8615. The Sentience Readiness Index: A Preliminary Framework for Measuring National Preparedness for the Possibility of Artificial Sentience
+
+- **ID:** `2603.01508` · **primary:** `cs.CY` · **Submitted:** 2026-03-02
+- **Authors:** Tony Rost
+- **abs:** https://arxiv.org/abs/2603.01508 · **PDF:** https://arxiv.org/pdf/2603.01508v2
+
+## 8616. HyperTokens: Controlling Token Dynamics for Continual Video-Language Understanding
+
+- **ID:** `2603.06662` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Toan Nguyen, Yang Liu, Celso De Melo, et al.
+- **abs:** https://arxiv.org/abs/2603.06662 · **PDF:** https://arxiv.org/pdf/2603.06662v2
+
+## 8617. RedacBench: Can AI Erase Your Secrets?
+
+- **ID:** `2603.20208` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Hyunjun Jeon, Kyuyoung Kim, Jinwoo Shin
+- **abs:** https://arxiv.org/abs/2603.20208 · **PDF:** https://arxiv.org/pdf/2603.20208v1
+
+## 8618. Anatomy of the Modality Gap: Dissecting the Internal States of End-to-End Speech LLMs
+
+- **ID:** `2603.01502` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Ming-Hao Hsu, Xueyao Zhang, Xiaohai Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.01502 · **PDF:** https://arxiv.org/pdf/2603.01502v1
+
+## 8619. GAC: Stabilizing Asynchronous RL Training for LLMs via Gradient Alignment Control
+
+- **ID:** `2603.01501` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Haofeng Xu, Junwei Su, Yukun Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.01501 · **PDF:** https://arxiv.org/pdf/2603.01501v1
+
+## 8620. Towards Privacy-Preserving LLM Inference via Covariant Obfuscation (Technical Report)
+
+- **ID:** `2603.01499` · **primary:** `cs.CR` · **Submitted:** 2026-03-02
+- **Authors:** Yu Lin, Qizhi Zhang, Wenqiang Ruan, et al.
+- **abs:** https://arxiv.org/abs/2603.01499 · **PDF:** https://arxiv.org/pdf/2603.01499v2
+
+## 8621. Inference-Time Safety For Code LLMs Via Retrieval-Augmented Revision
+
+- **ID:** `2603.01494` · **primary:** `cs.SE` · **Submitted:** 2026-03-02
+- **Authors:** Manisha Mukherjee, Vincent J. Hellendoorn
+- **abs:** https://arxiv.org/abs/2603.01494 · **PDF:** https://arxiv.org/pdf/2603.01494v1
+
+## 8622. PhotoBench: Beyond Visual Matching Towards Personalized Intent-Driven Photo Retrieval
+
+- **ID:** `2603.01493` · **primary:** `cs.IR` · **Submitted:** 2026-03-02
+- **Authors:** Tianyi Xu, Rong Shan, Junjie Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.01493 · **PDF:** https://arxiv.org/pdf/2603.01493v1
+
+## 8623. ATA: Bridging Implicit Reasoning with Attention-Guided and Action-Guided Inference for Vision-Language Action Models
+
+- **ID:** `2603.01490` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Cheng Yang, Jianhao Jiao, Lingyi Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.01490 · **PDF:** https://arxiv.org/pdf/2603.01490v1
+
+## 8624. LLM-assisted Semantic Option Discovery for Facilitating Adaptive Deep Reinforcement Learning
+
+- **ID:** `2603.01488` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Chang Yao, Jinghui Qin, Kebing Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.01488 · **PDF:** https://arxiv.org/pdf/2603.01488v2
+
+## 8625. Agentic Multi-Source Grounding for Enhanced Query Intent Understanding: A DoorDash Case Study
+
+- **ID:** `2603.01486` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Emmanuel Aboah Boateng, Kyle MacDonald, Akshad Viswanathan, et al.
+- **abs:** https://arxiv.org/abs/2603.01486 · **PDF:** https://arxiv.org/pdf/2603.01486v1
+
+## 8626. A SUPERB-Style Benchmark of Self-Supervised Speech Models for Audio Deepfake Detection
+
+- **ID:** `2603.01482` · **primary:** `eess.AS` · **Submitted:** 2026-03-02
+- **Authors:** Hashim Ali, Nithin Sai Adupa, Surya Subramani, et al.
+- **abs:** https://arxiv.org/abs/2603.01482 · **PDF:** https://arxiv.org/pdf/2603.01482v1
+
+## 8627. Harmonizing Dense and Sparse Signals in Multi-turn RL: Dual-Horizon Credit Assignment for Industrial Sales Agents
+
+- **ID:** `2603.01481` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Haojin Yang, Ai Jian, Xinyue Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.01481 · **PDF:** https://arxiv.org/pdf/2603.01481v1
+
+## 8628. Reconstructing Content via Collaborative Attention to Improve Multimodal Embedding Quality
+
+- **ID:** `2603.01471` · **primary:** `cs.IR` · **Submitted:** 2026-03-02
+- **Authors:** Jiahan Chen, Da Li, Hengran Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.01471 · **PDF:** https://arxiv.org/pdf/2603.01471v1
+
+## 8629. Randomized Kriging Believer for Parallel Bayesian Optimization with Regret Bounds
+
+- **ID:** `2603.01470` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Shuhei Sugiura, Ichiro Takeuchi, Shion Takeno
+- **abs:** https://arxiv.org/abs/2603.01470 · **PDF:** https://arxiv.org/pdf/2603.01470v2
+
+## 8630. Mean-Flow based One-Step Vision-Language-Action
+
+- **ID:** `2603.01469` · **primary:** `cs.RO` · **Submitted:** 2026-03-02
+- **Authors:** Yang Chen, Xiaoguang Ma, Bin Zhao
+- **abs:** https://arxiv.org/abs/2603.01469 · **PDF:** https://arxiv.org/pdf/2603.01469v1
+
+## 8631. Wild Bootstrap Inference for Non-Negative Matrix Factorization with Random Effects
+
+- **ID:** `2603.01468` · **primary:** `stat.ME` · **Submitted:** 2026-03-02
+- **Authors:** Kenichi Satoh
+- **abs:** https://arxiv.org/abs/2603.01468 · **PDF:** https://arxiv.org/pdf/2603.01468v1
+
+## 8632. Non-Markovian Long-Horizon Robot Manipulation via Keyframe Chaining
+
+- **ID:** `2603.01465` · **primary:** `cs.RO` · **Submitted:** 2026-03-02
+- **Authors:** Yipeng Chen, Wentao Tan, Lei Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.01465 · **PDF:** https://arxiv.org/pdf/2603.01465v1
+
+## 8633. ProtRLSearch: A Multi-Round Multimodal Protein Search Agent with Large Language Models Trained via Reinforcement Learning
+
+- **ID:** `2603.01464` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Congying Liu, Taihao Li, Ming Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.01464 · **PDF:** https://arxiv.org/pdf/2603.01464v1
+
+## 8634. DreamReader: An Interpretability Toolkit for Text-to-Image Models
+
+- **ID:** `2603.13299` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Nirmalendu Prakash, Narmeen Oozeer, Michael Lan, et al.
+- **abs:** https://arxiv.org/abs/2603.13299 · **PDF:** https://arxiv.org/pdf/2603.13299v1
+
+## 8635. PhysFormer: A Physics-Embedded Generative Model for Physically Self-Consistent Spectral Synthesis
+
+- **ID:** `2603.01459` · **primary:** `astro-ph.IM` · **Submitted:** 2026-03-02
+- **Authors:** Siqi Wang, Mengmeng Zhang, Yude Bu, et al.
+- **abs:** https://arxiv.org/abs/2603.01459 · **PDF:** https://arxiv.org/pdf/2603.01459v1
+
+## 8636. Power Echoes: Investigating Moderation Biases in Online Power-Asymmetric Conflicts
+
+- **ID:** `2603.01457` · **primary:** `cs.HC` · **Submitted:** 2026-03-02
+- **Authors:** Yaqiong Li, Peng Zhang, Peixu Hou, et al.
+- **abs:** https://arxiv.org/abs/2603.01457 · **PDF:** https://arxiv.org/pdf/2603.01457v1
+
+## 8637. From Verbatim to Gist: Distilling Pyramidal Multimodal Memory via Semantic Information Bottleneck for Long-Horizon Video Agents
+
+- **ID:** `2603.01455` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Niu Lian, Yuting Wang, Hanshu Yao, et al.
+- **abs:** https://arxiv.org/abs/2603.01455 · **PDF:** https://arxiv.org/pdf/2603.01455v1
+
+## 8638. VidDoS: Universal Denial-of-Service Attack on Video-based Large Language Models
+
+- **ID:** `2603.01454` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Duoxun Tang, Dasen Dai, Jiyao Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.01454 · **PDF:** https://arxiv.org/pdf/2603.01454v1
+
+## 8639. Scaling Tasks, Not Samples: Mastering Humanoid Control through Multi-Task Model-Based Reinforcement Learning
+
+- **ID:** `2603.01452` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Shaohuai Liu, Weirui Ye, Yilun Du, et al.
+- **abs:** https://arxiv.org/abs/2603.01452 · **PDF:** https://arxiv.org/pdf/2603.01452v1
+
+## 8640. SEAnet: A Deep Learning Architecture for Data Series Similarity Search
+
+- **ID:** `2603.01448` · **primary:** `cs.DB` · **Submitted:** 2026-03-02
+- **Authors:** Qitong Wang, Themis Palpanas
+- **abs:** https://arxiv.org/abs/2603.01448 · **PDF:** https://arxiv.org/pdf/2603.01448v2
+
+## 8641. Autoregressive Synthesis of Sparse and Semi-Structured Mixed-Type Data
+
+- **ID:** `2603.01444` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Thomas Rückstieß, Robin Vujanic
+- **abs:** https://arxiv.org/abs/2603.01444 · **PDF:** https://arxiv.org/pdf/2603.01444v1
+
+## 8642. Enhancing Persona Following at Decoding Time via Dynamic Importance Estimation for Role-Playing Agents
+
+- **ID:** `2603.01438` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Yuxin Liu, Mingye Zhu, Siyuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.01438 · **PDF:** https://arxiv.org/pdf/2603.01438v1
+
+## 8643. Decoding Answers Before Chain-of-Thought: Evidence from Pre-CoT Probes and Activation Steering
+
+- **ID:** `2603.01437` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Kyle Cox, Darius Kianersi, Adrià Garriga-Alonso
+- **abs:** https://arxiv.org/abs/2603.01437 · **PDF:** https://arxiv.org/pdf/2603.01437v1
+
+## 8644. On the Stability Connection Between Discrete-Time Algorithms and Their Resolution ODEs: Applications to Min-Max Optimisation
+
+- **ID:** `2603.01430` · **primary:** `math.OC` · **Submitted:** 2026-03-02
+- **Authors:** Amir Ali Farzin, Yuen-Man Pun, Philipp Braun, et al.
+- **abs:** https://arxiv.org/abs/2603.01430 · **PDF:** https://arxiv.org/pdf/2603.01430v1
+
+## 8645. Understanding the Physics of Key-Value Cache Compression for LLMs through Attention Dynamics
+
+- **ID:** `2603.01426` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Samhruth Ananthanarayanan, Ayan Sengupta, Tanmoy Chakraborty
+- **abs:** https://arxiv.org/abs/2603.01426 · **PDF:** https://arxiv.org/pdf/2603.01426v1
+
+## 8646. LaSER: Internalizing Explicit Reasoning into Latent Space for Dense Retrieval
+
+- **ID:** `2603.01425` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Jiajie Jin, Yanzhao Zhang, Mingxin Li, et al.
+- **abs:** https://arxiv.org/abs/2603.01425 · **PDF:** https://arxiv.org/pdf/2603.01425v1
+
+## 8647. Quantifying Conversational Reliability of Large Language Models under Multi-Turn Interaction
+
+- **ID:** `2603.01423` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Jiyoon Myung
+- **abs:** https://arxiv.org/abs/2603.01423 · **PDF:** https://arxiv.org/pdf/2603.01423v1
+
+## 8648. SciDER: Scientific Data-centric End-to-end Researcher
+
+- **ID:** `2603.01421` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Ke Lin, Yilin Lu, Shreyas Bhat, et al.
+- **abs:** https://arxiv.org/abs/2603.01421 · **PDF:** https://arxiv.org/pdf/2603.01421v1
+
+## 8649. Tackling multiphysics problems via finite element-guided physics-informed operator learning
+
+- **ID:** `2603.01420` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Yusuke Yamazaki, Reza Najian Asl, Markus Apel, et al.
+- **abs:** https://arxiv.org/abs/2603.01420 · **PDF:** https://arxiv.org/pdf/2603.01420v1
+
+## 8650. Securing the Floor and Raising the Ceiling: A Merging-based Paradigm for Multi-modal Search Agents
+
+- **ID:** `2603.01416` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Zhixiang Wang, Jingxuan Xu, Dajun Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.01416 · **PDF:** https://arxiv.org/pdf/2603.01416v1
+
+## 8651. FusionCast: Enhancing Precipitation Nowcasting with Asymmetric Cross-Modal Fusion and Future Radar Priors
+
+- **ID:** `2603.13298` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Henan Wang, Shengwu Xiong, Yifang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.13298 · **PDF:** https://arxiv.org/pdf/2603.13298v1
+
+## 8652. Enhanced Atrial Fibrillation Prediction in ESUS Patients with Hypergraph-based Pre-training
+
+- **ID:** `2603.13297` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Yuzhang Xie, Yuhua Wu, Ruiyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.13297 · **PDF:** https://arxiv.org/pdf/2603.13297v2
+
+## 8653. EnsAug: Augmentation-Driven Ensembles for Human Motion Sequence Analysis
+
+- **ID:** `2603.06661` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Bikram De, Habib Irani, Vangelis Metsis
+- **abs:** https://arxiv.org/abs/2603.06661 · **PDF:** https://arxiv.org/pdf/2603.06661v1
+
+## 8654. GraphScout: Empowering Large Language Models with Intrinsic Exploration Ability for Agentic Graph Reasoning
+
+- **ID:** `2603.01410` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Yuchen Ying, Weiqi Jiang, Tongya Zheng, et al.
+- **abs:** https://arxiv.org/abs/2603.01410 · **PDF:** https://arxiv.org/pdf/2603.01410v1
+
+## 8655. TriageSim: A Conversational Emergency Triage Simulation Framework from Structured Electronic Health Records
+
+- **ID:** `2603.10035` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Dipankar Srirag, Quoc Dung Nguyen, Aditya Joshi, et al.
+- **abs:** https://arxiv.org/abs/2603.10035 · **PDF:** https://arxiv.org/pdf/2603.10035v2
+
+## 8656. MIST-RL: Mutation-based Incremental Suite Testing via Reinforcement Learning
+
+- **ID:** `2603.01409` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Sicheng Zhu, Jiajun Wang, Jiawei Ai, et al.
+- **abs:** https://arxiv.org/abs/2603.01409 · **PDF:** https://arxiv.org/pdf/2603.01409v1
+
+## 8657. The Observer-Situation Lattice: A Unified Formal Basis for Perspective-Aware Cognition
+
+- **ID:** `2603.01407` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Saad Alqithami
+- **abs:** https://arxiv.org/abs/2603.01407 · **PDF:** https://arxiv.org/pdf/2603.01407v1
+
+## 8658. One Operator to Rule Them All? On Boundary-Indexed Operator Families in Neural PDE Solvers
+
+- **ID:** `2603.01406` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Lennon J. Shikhman
+- **abs:** https://arxiv.org/abs/2603.01406 · **PDF:** https://arxiv.org/pdf/2603.01406v1
+
+## 8659. Quasar: Quantized Self-Speculative Acceleration for Rapid Inference via Memory-Efficient Verification
+
+- **ID:** `2603.01399` · **primary:** `cs.DC` · **Submitted:** 2026-03-02
+- **Authors:** Guang Huang, Zeyi Wen
+- **abs:** https://arxiv.org/abs/2603.01399 · **PDF:** https://arxiv.org/pdf/2603.01399v1
+
+## 8660. HarmonyCell: Automating Single-Cell Perturbation Modeling under Semantic and Distribution Shifts
+
+- **ID:** `2603.01396` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Wenxuan Huang, Mingyu Tsoi, Yanhao Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.01396 · **PDF:** https://arxiv.org/pdf/2603.01396v2
+
+## 8661. Invariant-Stratified Propagation for Expressive Graph Neural Networks
+
+- **ID:** `2603.01388` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Asela Hevapathige, Ahad N. Zehmakan, Asiri Wijesinghe, et al.
+- **abs:** https://arxiv.org/abs/2603.01388 · **PDF:** https://arxiv.org/pdf/2603.01388v1
+
+## 8662. Toward Graph-Tokenizing Large Language Models with Reconstructive Graph Instruction Tuning
+
+- **ID:** `2603.01385` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Zhongjian Zhang, Xiao Wang, Mengmei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.01385 · **PDF:** https://arxiv.org/pdf/2603.01385v1
+
+## 8663. End-to-End Simultaneous Dysarthric Speech Reconstruction with Frame-Level Adaptor and Multiple Wait-k Knowledge Distillation
+
+- **ID:** `2603.01382` · **primary:** `cs.SD` · **Submitted:** 2026-03-02
+- **Authors:** Minghui Wu, Haitao Tang, Jiahuan Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.01382 · **PDF:** https://arxiv.org/pdf/2603.01382v1
+
+## 8664. 3BASiL: An Algorithmic Framework for Sparse plus Low-Rank Compression of LLMs
+
+- **ID:** `2603.01376` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Mehdi Makni, Xiang Meng, Rahul Mazumder
+- **abs:** https://arxiv.org/abs/2603.01376 · **PDF:** https://arxiv.org/pdf/2603.01376v1
+
+## 8665. Words & Weights: Streamlining Multi-Turn Interactions via Co-Adaptation
+
+- **ID:** `2603.01375` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Chenxing Wei, Hong Wang, Ying He, et al.
+- **abs:** https://arxiv.org/abs/2603.01375 · **PDF:** https://arxiv.org/pdf/2603.01375v1
+
+## 8666. Quantum-Inspired Fine-Tuning for Few-Shot AIGC Detection via Phase-Structured Reparameterization
+
+- **ID:** `2603.02281` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Kaiyang Xing, Han Fang, Zhaoyun Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.02281 · **PDF:** https://arxiv.org/pdf/2603.02281v1
+
+## 8667. Causal Neural Probabilistic Circuits
+
+- **ID:** `2603.01372` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Weixin Chen, Han Zhao
+- **abs:** https://arxiv.org/abs/2603.01372 · **PDF:** https://arxiv.org/pdf/2603.01372v1
+
+## 8668. DARS: Dysarthria-Aware Rhythm-Style Synthesis for ASR Enhancement
+
+- **ID:** `2603.01369` · **primary:** `cs.SD` · **Submitted:** 2026-03-02
+- **Authors:** Minghui Wu, Xueling Liu, Jiahuan Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.01369 · **PDF:** https://arxiv.org/pdf/2603.01369v1
+
+## 8669. Temporal Imbalance of Positive and Negative Supervision in Class-Incremental Learning
+
+- **ID:** `2603.02280` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Jinge Ma, Fengqing Zhu
+- **abs:** https://arxiv.org/abs/2603.02280 · **PDF:** https://arxiv.org/pdf/2603.02280v1
+
+## 8670. DUEL: Exact Likelihood for Masked Diffusion via Deterministic Unmasking
+
+- **ID:** `2603.01367` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Gilad Turok, Chris De Sa, Volodymyr Kuleshov
+- **abs:** https://arxiv.org/abs/2603.01367 · **PDF:** https://arxiv.org/pdf/2603.01367v2
+
+## 8671. NM-DEKL$^3_\infty$: A Three-Layer Non-Monotone Evolving Dependent Type Logic
+
+- **ID:** `2603.01366` · **primary:** `cs.LO` · **Submitted:** 2026-03-02
+- **Authors:** Peng Chen
+- **abs:** https://arxiv.org/abs/2603.01366 · **PDF:** https://arxiv.org/pdf/2603.01366v1
+
+## 8672. Align and Filter: Improving Performance in Asynchronous On-Policy RL
+
+- **ID:** `2603.01365` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Homayoun Honari, Roger Creus Castanyer, Michael Przystupa, et al.
+- **abs:** https://arxiv.org/abs/2603.01365 · **PDF:** https://arxiv.org/pdf/2603.01365v1
+
+## 8673. Fed-GAME: Personalized Federated Learning with Graph Attention Mixture-of-Experts For Time-Series Forecasting
+
+- **ID:** `2603.01363` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Yi Li, Han Liu, Mingfeng Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.01363 · **PDF:** https://arxiv.org/pdf/2603.01363v1
+
+## 8674. MixerCSeg: An Efficient Mixer Architecture for Crack Segmentation via Decoupled Mamba Attention
+
+- **ID:** `2603.01361` · **primary:** `cs.CV` · **Submitted:** 2026-03-02
+- **Authors:** Zilong Zhao, Zhengming Ding, Pei Niu, et al.
+- **abs:** https://arxiv.org/abs/2603.01361 · **PDF:** https://arxiv.org/pdf/2603.01361v1
+
+## 8675. ASTRA-bench: Evaluating Tool-Use Agent Reasoning and Action Planning with Personal User Context
+
+- **ID:** `2603.01357` · **primary:** `cs.AI` · **Submitted:** 2026-03-02
+- **Authors:** Zidi Xiu, David Q. Sun, Kevin Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.01357 · **PDF:** https://arxiv.org/pdf/2603.01357v1
+
+## 8676. Constructing Synthetic Instruction Datasets for Improving Reasoning in Domain-Specific LLMs: A Case Study in the Japanese Financial Domain
+
+- **ID:** `2603.01353` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Yuma Okochi, Fabio Milentiansen Sim, Tomoyasu Okada
+- **abs:** https://arxiv.org/abs/2603.01353 · **PDF:** https://arxiv.org/pdf/2603.01353v1
+
+## 8677. UTICA: Multi-Objective Self-Distllation Foundation Model Pretraining for Time Series Classification
+
+- **ID:** `2603.01348` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Yessin Moakher, Youssef Attia El Hili, Vasilii Feofanov
+- **abs:** https://arxiv.org/abs/2603.01348 · **PDF:** https://arxiv.org/pdf/2603.01348v1
+
+## 8678. Relatively Smart: A New Approach for Instance-Optimal Learning
+
+- **ID:** `2603.01346` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Shaddin Dughmi, Alireza F. Pour
+- **abs:** https://arxiv.org/abs/2603.01346 · **PDF:** https://arxiv.org/pdf/2603.01346v1
+
+## 8679. PanCanBench: A Comprehensive Benchmark for Evaluating Large Language Models in Pancreatic Oncology
+
+- **ID:** `2603.01343` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Yimin Zhao, Sheela R. Damle, Simone E. Dekker, et al.
+- **abs:** https://arxiv.org/abs/2603.01343 · **PDF:** https://arxiv.org/pdf/2603.01343v1
+
+## 8680. SubstratumGraphEnv: Reinforcement Learning Environment (RLE) for Modeling System Attack Paths
+
+- **ID:** `2603.01340` · **primary:** `cs.CR` · **Submitted:** 2026-03-02
+- **Authors:** Bahirah Adewunmi, Edward Raff, Sanjay Purushotham
+- **abs:** https://arxiv.org/abs/2603.01340 · **PDF:** https://arxiv.org/pdf/2603.01340v1
+
+## 8681. Causal Effects with Unobserved Unit Types in Interacting Human-AI Systems
+
+- **ID:** `2603.01339` · **primary:** `stat.ML` · **Submitted:** 2026-03-02
+- **Authors:** William Overman, Sadegh Shirani, Mohsen Bayati
+- **abs:** https://arxiv.org/abs/2603.01339 · **PDF:** https://arxiv.org/pdf/2603.01339v1
+
+## 8682. Adaptive Estimation and Inference in Conditional Moment Models via the Discrepancy Principle
+
+- **ID:** `2603.01337` · **primary:** `stat.ML` · **Submitted:** 2026-03-02
+- **Authors:** Jiyuan Tan, Vasilis Syrgkanis
+- **abs:** https://arxiv.org/abs/2603.01337 · **PDF:** https://arxiv.org/pdf/2603.01337v1
+
+## 8683. URAG: A Benchmark for Uncertainty Quantification in Retrieval-Augmented Large Language Models
+
+- **ID:** `2603.19281` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Vinh Nguyen, Cuong Dang, Jiahao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.19281 · **PDF:** https://arxiv.org/pdf/2603.19281v1
+
+## 8684. Provable and Practical In-Context Policy Optimization for Self-Improvement
+
+- **ID:** `2603.01335` · **primary:** `cs.LG` · **Submitted:** 2026-03-02
+- **Authors:** Tianrun Yu, Yuxiao Yang, Zhaoyang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.01335 · **PDF:** https://arxiv.org/pdf/2603.01335v1
+
+## 8685. MetaState: Persistent Working Memory Enhances Reasoning in Discrete Diffusion Language Models
+
+- **ID:** `2603.01331` · **primary:** `cs.CL` · **Submitted:** 2026-03-02
+- **Authors:** Kejing Xia, Mingzhe Li, Lixuan Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.01331 · **PDF:** https://arxiv.org/pdf/2603.01331v2
+
+## 8686. You Only Need One Stage: Novel-View Synthesis From A Single Blind Face Image
+
+- **ID:** `2603.01328` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Taoyue Wang, Xiang Zhang, Xiaotian Li, et al.
+- **abs:** https://arxiv.org/abs/2603.01328 · **PDF:** https://arxiv.org/pdf/2603.01328v1
+
+## 8687. SWE-Adept: An LLM-Based Agentic Framework for Deep Codebase Analysis and Structured Issue Resolution
+
+- **ID:** `2603.01327` · **primary:** `cs.SE` · **Submitted:** 2026-03-01
+- **Authors:** Kang He, Kaushik Roy
+- **abs:** https://arxiv.org/abs/2603.01327 · **PDF:** https://arxiv.org/pdf/2603.01327v1
+
+## 8688. Truth as a Trajectory: What Internal Representations Reveal About Large Language Model Reasoning
+
+- **ID:** `2603.01326` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Hamed Damirchi, Ignacio Meza De la Jara, Ehsan Abbasnejad, et al.
+- **abs:** https://arxiv.org/abs/2603.01326 · **PDF:** https://arxiv.org/pdf/2603.01326v1
+
+## 8689. From GEV to ResLogit: Spatially Correlated Discrete Choice Models for Pedestrian Movement Prediction
+
+- **ID:** `2603.01325` · **primary:** `physics.soc-ph` · **Submitted:** 2026-03-01
+- **Authors:** Rulla Al-Haideri, Bilal Farooq
+- **abs:** https://arxiv.org/abs/2603.01325 · **PDF:** https://arxiv.org/pdf/2603.01325v1
+
+## 8690. From Feature-Based Models to Generative AI: Validity Evidence for Constructed Response Scoring
+
+- **ID:** `2603.19280` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Jodi M. Casabianca, Daniel F. McCaffrey, Matthew S. Johnson, et al.
+- **abs:** https://arxiv.org/abs/2603.19280 · **PDF:** https://arxiv.org/pdf/2603.19280v1
+
+## 8691. Quantifying Frontier LLM Capabilities for Container Sandbox Escape
+
+- **ID:** `2603.02277` · **primary:** `cs.CR` · **Submitted:** 2026-03-01
+- **Authors:** Rahul Marchand, Art O Cathain, Jerome Wynne, et al.
+- **abs:** https://arxiv.org/abs/2603.02277 · **PDF:** https://arxiv.org/pdf/2603.02277v1
+
+## 8692. Catalyst-Agent: Autonomous heterogeneous catalyst screening and optimization with an LLM Agent
+
+- **ID:** `2603.01311` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Achuth Chandrasekhar, Janghoon Ock, Amir Barati Farimani
+- **abs:** https://arxiv.org/abs/2603.01311 · **PDF:** https://arxiv.org/pdf/2603.01311v1
+
+## 8693. PAC Guarantees for Reinforcement Learning: Sample Complexity, Coverage, and Structure
+
+- **ID:** `2603.01309` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Joshua Steier
+- **abs:** https://arxiv.org/abs/2603.01309 · **PDF:** https://arxiv.org/pdf/2603.01309v1
+
+## 8694. GPU-friendly and Linearly Convergent First-order Methods for Certifying Optimal $k$-sparse GLMs
+
+- **ID:** `2603.01306` · **primary:** `math.OC` · **Submitted:** 2026-03-01
+- **Authors:** Jiachang Liu, Andrea Lodi, Soroosh Shafiee
+- **abs:** https://arxiv.org/abs/2603.01306 · **PDF:** https://arxiv.org/pdf/2603.01306v1
+
+## 8695. AG-VAS: Anchor-Guided Zero-Shot Visual Anomaly Segmentation with Large Multimodal Models
+
+- **ID:** `2603.01305` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Zhen Qu, Xian Tao, Xiaoyi Bao, et al.
+- **abs:** https://arxiv.org/abs/2603.01305 · **PDF:** https://arxiv.org/pdf/2603.01305v2
+
+## 8696. Nonconvex Latent Optimally Partitioned Block-Sparse Recovery via Log-Sum and Minimax Concave Penalties
+
+- **ID:** `2603.01304` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Takanobu Furuhashi, Hiroki Kuroda, Masahiro Yukawa, et al.
+- **abs:** https://arxiv.org/abs/2603.01304 · **PDF:** https://arxiv.org/pdf/2603.01304v1
+
+## 8697. I Can't Believe It's Not Robust: Catastrophic Collapse of Safety Classifiers under Embedding Drift
+
+- **ID:** `2603.01297` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Subramanyam Sahoo, Vinija Jain, Divya Chaudhary, et al.
+- **abs:** https://arxiv.org/abs/2603.01297 · **PDF:** https://arxiv.org/pdf/2603.01297v1
+
+## 8698. Multi-Level Bidirectional Decoder Interaction for Uncertainty-Aware Breast Ultrasound Analysis
+
+- **ID:** `2603.01295` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Abdullah Al Shafi, Md Kawsar Mahmud Khan Zunayed, Safin Ahmmed, et al.
+- **abs:** https://arxiv.org/abs/2603.01295 · **PDF:** https://arxiv.org/pdf/2603.01295v1
+
+## 8699. Theoretical Perspectives on Data Quality and Synergistic Effects in Pre- and Post-Training Reasoning Models
+
+- **ID:** `2603.01293` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Adel Javanmard, Baharan Mirzasoleiman, Vahab Mirrokni
+- **abs:** https://arxiv.org/abs/2603.01293 · **PDF:** https://arxiv.org/pdf/2603.01293v1
+
+## 8700. Approximate Nearest Neighbor Search for Modern AI: A Projection-Augmented Graph Approach
+
+- **ID:** `2603.06660` · **primary:** `cs.IR` · **Submitted:** 2026-03-01
+- **Authors:** Kejing Lu, Zhenpeng Pan, Jianbin Qin, et al.
+- **abs:** https://arxiv.org/abs/2603.06660 · **PDF:** https://arxiv.org/pdf/2603.06660v1
+
+## 8701. Integrating LTL Constraints into PPO for Safe Reinforcement Learning
+
+- **ID:** `2603.01292` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Maifang Zhang, Hang Yu, Qian Zuo, et al.
+- **abs:** https://arxiv.org/abs/2603.01292 · **PDF:** https://arxiv.org/pdf/2603.01292v1
+
+## 8702. JailNewsBench: Multi-Lingual and Regional Benchmark for Fake News Generation under Jailbreak Attacks
+
+- **ID:** `2603.01291` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Masahiro Kaneko, Ayana Niwa, Timothy Baldwin
+- **abs:** https://arxiv.org/abs/2603.01291 · **PDF:** https://arxiv.org/pdf/2603.01291v1
+
+## 8703. Opponent State Inference Under Partial Observability: An HMM-POMDP Framework for 2026 Formula 1 Energy Strategy
+
+- **ID:** `2603.01290` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Kalliopi Kleisarchaki
+- **abs:** https://arxiv.org/abs/2603.01290 · **PDF:** https://arxiv.org/pdf/2603.01290v2
+
+## 8704. Individual Turing Test: A Case Study of LLM-based Simulation Using Longitudinal Personal Data
+
+- **ID:** `2603.01289` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Minghao Guo, Ziyi Ye, Wujiang Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.01289 · **PDF:** https://arxiv.org/pdf/2603.01289v1
+
+## 8705. Efficient Extractive Summarization with MAMBA-Transformer Hybrids for Low-Resource Scenarios
+
+- **ID:** `2603.01288` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Nisrine Ait Khayi
+- **abs:** https://arxiv.org/abs/2603.01288 · **PDF:** https://arxiv.org/pdf/2603.01288v1
+
+## 8706. Information-Theoretic Framework for Self-Adapting Model Predictive Controllers
+
+- **ID:** `2603.01286` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Wael Hafez, Amir Nazeri
+- **abs:** https://arxiv.org/abs/2603.01286 · **PDF:** https://arxiv.org/pdf/2603.01286v1
+
+## 8707. Attention Smoothing Is All You Need For Unlearning
+
+- **ID:** `2603.01285` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Saleh Zare Zade, Xiangyu Zhou, Sijia Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.01285 · **PDF:** https://arxiv.org/pdf/2603.01285v1
+
+## 8708. Beyond Reward: A Bounded Measure of Agent Environment Coupling
+
+- **ID:** `2603.01283` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Wael Hafez, Cameron Reid, Amit Nazeri
+- **abs:** https://arxiv.org/abs/2603.01283 · **PDF:** https://arxiv.org/pdf/2603.01283v1
+
+## 8709. Spectral Attention Steering for Prompt Highlighting
+
+- **ID:** `2603.01281` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Weixian Waylon Li, Yuchen Niu, Yongxin Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.01281 · **PDF:** https://arxiv.org/pdf/2603.01281v1
+
+## 8710. Multilingual Hate Speech Detection and Counterspeech Generation: A Comprehensive Survey and Practical Guide
+
+- **ID:** `2603.19279` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Zahra Safdari Fesaghandis, Suman Kalyan Maity
+- **abs:** https://arxiv.org/abs/2603.19279 · **PDF:** https://arxiv.org/pdf/2603.19279v1
+
+## 8711. The Impact of Battery Cell Configuration on Electric Vehicle Performance: An XGBoost-Based Classification with SHAP Interpretability
+
+- **ID:** `2603.01275` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Santanam Wishal, Louis Filiepe Tio Jansel, Matthew Abednego Inkiriwang, et al.
+- **abs:** https://arxiv.org/abs/2603.01275 · **PDF:** https://arxiv.org/pdf/2603.01275v1
+
+## 8712. GlassMol: Interpretable Molecular Property Prediction with Concept Bottleneck Models
+
+- **ID:** `2603.01274` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Oscar Rivera, Ziqing Wang, Matthieu Dagommer, et al.
+- **abs:** https://arxiv.org/abs/2603.01274 · **PDF:** https://arxiv.org/pdf/2603.01274v1
+
+## 8713. VoxKnesset: A Large-Scale Longitudinal Hebrew Speech Dataset for Aging Speaker Modeling
+
+- **ID:** `2603.01270` · **primary:** `eess.AS` · **Submitted:** 2026-03-01
+- **Authors:** Yanir Marmor, Arad Zulti, David Krongauz, et al.
+- **abs:** https://arxiv.org/abs/2603.01270 · **PDF:** https://arxiv.org/pdf/2603.01270v2
+
+## 8714. A Study on Building Efficient Zero-Shot Relation Extraction Models
+
+- **ID:** `2603.01266` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Hugo Thomas, Caio Corro, Guillaume Gravier, et al.
+- **abs:** https://arxiv.org/abs/2603.01266 · **PDF:** https://arxiv.org/pdf/2603.01266v2
+
+## 8715. S2O: Enhancing Adversarial Training with Second-Order Statistics of Weights
+
+- **ID:** `2603.01264` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Gaojie Jin, Xinping Yi, Wei Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.01264 · **PDF:** https://arxiv.org/pdf/2603.01264v1
+
+## 8716. MOSAIC: A Unified Platform for Cross-Paradigm Comparison and Evaluation of Homogeneous and Heterogeneous Multi-Agent RL, LLM, VLM, and Human Decision-Makers
+
+- **ID:** `2603.01260` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Abdulhamid M. Mousa, Yu Fu, Rakhmonberdi Khajiev, et al.
+- **abs:** https://arxiv.org/abs/2603.01260 · **PDF:** https://arxiv.org/pdf/2603.01260v1
+
+## 8717. LLM Self-Explanations Fail Semantic Invariance
+
+- **ID:** `2603.01254` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Stefan Szeider
+- **abs:** https://arxiv.org/abs/2603.01254 · **PDF:** https://arxiv.org/pdf/2603.01254v1
+
+## 8718. Linking Knowledge to Care: Knowledge Graph-Augmented Medical Follow-Up Question Generation
+
+- **ID:** `2603.01252` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Liwen Sun, Xiang Yu, Ming Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.01252 · **PDF:** https://arxiv.org/pdf/2603.01252v1
+
+## 8719. The MAMA-MIA Challenge: Advancing Generalizability and Fairness in Breast MRI Tumor Segmentation and Treatment Response Prediction
+
+- **ID:** `2603.01250` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Lidia Garrucho, Smriti Joshi, Kaisar Kushibar, et al.
+- **abs:** https://arxiv.org/abs/2603.01250 · **PDF:** https://arxiv.org/pdf/2603.01250v1
+
+## 8720. Defensive Refusal Bias: How Safety Alignment Fails Cyber Defenders
+
+- **ID:** `2603.01246` · **primary:** `cs.CR` · **Submitted:** 2026-03-01
+- **Authors:** David Campbell, Neil Kale, Udari Madhushani Sehwag, et al.
+- **abs:** https://arxiv.org/abs/2603.01246 · **PDF:** https://arxiv.org/pdf/2603.01246v2
+
+## 8721. Suffix-Constrained Greedy Search Algorithms for Causal Language Models
+
+- **ID:** `2603.01243` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Ayoub Hammal, Pierre Zweigenbaum, Caio Corro
+- **abs:** https://arxiv.org/abs/2603.01243 · **PDF:** https://arxiv.org/pdf/2603.01243v1
+
+## 8722. Science Literacy: Generative AI as Enabler of Coherence in the Teaching, Learning, and Assessment of Scientific Knowledge and Reasoning
+
+- **ID:** `2603.06659` · **primary:** `cs.CY` · **Submitted:** 2026-03-01
+- **Authors:** Xiaoming Zhai, James W. Pellegrino, Matias Rojas, et al.
+- **abs:** https://arxiv.org/abs/2603.06659 · **PDF:** https://arxiv.org/pdf/2603.06659v1
+
+## 8723. TARSE: Test-Time Adaptation via Retrieval of Skills and Experience for Reasoning Agents
+
+- **ID:** `2603.01241` · **primary:** `cs.IR` · **Submitted:** 2026-03-01
+- **Authors:** Junda Wang, Zonghai Tao, Hansi Zeng, et al.
+- **abs:** https://arxiv.org/abs/2603.01241 · **PDF:** https://arxiv.org/pdf/2603.01241v1
+
+## 8724. Self-Anchoring Calibration Drift in Large Language Models: How Multi-Turn Conversations Reshape Model Confidence
+
+- **ID:** `2603.01239` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Harshavardhan
+- **abs:** https://arxiv.org/abs/2603.01239 · **PDF:** https://arxiv.org/pdf/2603.01239v1
+
+## 8725. AgilePruner: An Empirical Study of Attention and Diversity for Adaptive Visual Token Pruning in Large Vision-Language Models
+
+- **ID:** `2603.01236` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Changwoo Baek, Jouwon Song, Sohyeon Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.01236 · **PDF:** https://arxiv.org/pdf/2603.01236v1
+
+## 8726. Extended Empirical Validation of the Explainability Solution Space
+
+- **ID:** `2603.01235` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Antoni Mestre, Manoli Albert, Miriam Gil, et al.
+- **abs:** https://arxiv.org/abs/2603.01235 · **PDF:** https://arxiv.org/pdf/2603.01235v2
+
+## 8727. RMBench: Memory-Dependent Robotic Manipulation Benchmark with Insights into Policy Design
+
+- **ID:** `2603.01229` · **primary:** `cs.RO` · **Submitted:** 2026-03-01
+- **Authors:** Tianxing Chen, Yuran Wang, Mingleyang Li, et al.
+- **abs:** https://arxiv.org/abs/2603.01229 · **PDF:** https://arxiv.org/pdf/2603.01229v2
+
+## 8728. The Lattice Representation Hypothesis of Large Language Models
+
+- **ID:** `2603.01227` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Bo Xiong
+- **abs:** https://arxiv.org/abs/2603.01227 · **PDF:** https://arxiv.org/pdf/2603.01227v1
+
+## 8729. Can Thinking Models Think to Detect Hateful Memes?
+
+- **ID:** `2603.01225` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Mohamed Bayan Kmainasi, Mucahid Kutlu, Ali Ezzat Shahroor, et al.
+- **abs:** https://arxiv.org/abs/2603.01225 · **PDF:** https://arxiv.org/pdf/2603.01225v1
+
+## 8730. Monocular 3D Object Position Estimation with VLMs for Human-Robot Interaction
+
+- **ID:** `2603.01224` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Ari Wahl, Dorian Gawlinski, David Przewozny, et al.
+- **abs:** https://arxiv.org/abs/2603.01224 · **PDF:** https://arxiv.org/pdf/2603.01224v1
+
+## 8731. Learn Hard Problems During RL with Reference Guided Fine-tuning
+
+- **ID:** `2603.01223` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Yangzhen Wu, Shanda Li, Zixin Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.01223 · **PDF:** https://arxiv.org/pdf/2603.01223v2
+
+## 8732. Communication-Efficient Quantum Federated Learning over Large-Scale Wireless Networks
+
+- **ID:** `2603.01222` · **primary:** `cs.IT` · **Submitted:** 2026-03-01
+- **Authors:** Shaba Shaon, Christopher G. Brinton, Dinh C. Nguyen
+- **abs:** https://arxiv.org/abs/2603.01222 · **PDF:** https://arxiv.org/pdf/2603.01222v1
+
+## 8733. Generative AI & Fictionality: How Novels Power Large Language Models
+
+- **ID:** `2603.01220` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Edwin Roland, Richard Jean So
+- **abs:** https://arxiv.org/abs/2603.01220 · **PDF:** https://arxiv.org/pdf/2603.01220v1
+
+## 8734. Reasoning Boosts Opinion Alignment in LLMs
+
+- **ID:** `2603.01214` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Frédéric Berdoz, Yann Billeter, Yann Vonlanthen, et al.
+- **abs:** https://arxiv.org/abs/2603.01214 · **PDF:** https://arxiv.org/pdf/2603.01214v2
+
+## 8735. Can AI Agents Agree?
+
+- **ID:** `2603.01213` · **primary:** `cs.MA` · **Submitted:** 2026-03-01
+- **Authors:** Frédéric Berdoz, Leonardo Rugli, Roger Wattenhofer
+- **abs:** https://arxiv.org/abs/2603.01213 · **PDF:** https://arxiv.org/pdf/2603.01213v2
+
+## 8736. XAI-enhanced Comparative Opinion Mining via Aspect-based Scoring and Semantic Reasoning
+
+- **ID:** `2603.01212` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Ngoc-Quang Le, T. Thanh-Lam Nguyen, Quoc-Trung Phu, et al.
+- **abs:** https://arxiv.org/abs/2603.01212 · **PDF:** https://arxiv.org/pdf/2603.01212v1
+
+## 8737. A Unified Framework to Quantify Cultural Intelligence of AI
+
+- **ID:** `2603.01211` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Sunipa Dev, Vinodkumar Prabhakaran, Rutledge Chin Feman, et al.
+- **abs:** https://arxiv.org/abs/2603.01211 · **PDF:** https://arxiv.org/pdf/2603.01211v2
+
+## 8738. Agents Learn Their Runtime: Interpreter Persistence as Training-Time Semantics
+
+- **ID:** `2603.01209` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Victor May, Aaditya Salgarkar, Yishan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.01209 · **PDF:** https://arxiv.org/pdf/2603.01209v2
+
+## 8739. Subliminal Signals in Preference Labels
+
+- **ID:** `2603.01204` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Isotta Magistrali, Frédéric Berdoz, Sam Dauncey, et al.
+- **abs:** https://arxiv.org/abs/2603.01204 · **PDF:** https://arxiv.org/pdf/2603.01204v2
+
+## 8740. How Well Does Agent Development Reflect Real-World Work?
+
+- **ID:** `2603.01203` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Zora Zhiruo Wang, Sanidhya Vijayvargiya, Aspen Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.01203 · **PDF:** https://arxiv.org/pdf/2603.01203v2
+
+## 8741. Incremental LTLf Synthesis
+
+- **ID:** `2603.01201` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Giuseppe De Giacomo, Yves Lespérance, Gianmarco Parretti, et al.
+- **abs:** https://arxiv.org/abs/2603.01201 · **PDF:** https://arxiv.org/pdf/2603.01201v1
+
+## 8742. A Comparative Study of UMAP and Other Dimensionality Reduction Methods
+
+- **ID:** `2603.02275` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Guanzhe Zhang, Shanshan Ding, Zhezhen Jin
+- **abs:** https://arxiv.org/abs/2603.02275 · **PDF:** https://arxiv.org/pdf/2603.02275v1
+
+## 8743. VisNec: Measuring and Leveraging Visual Necessity for Multimodal Instruction Tuning
+
+- **ID:** `2603.01195` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Mingkang Dong, Hongyi Cai, Jie Li, et al.
+- **abs:** https://arxiv.org/abs/2603.01195 · **PDF:** https://arxiv.org/pdf/2603.01195v1
+
+## 8744. Operator Learning Using Weak Supervision from Walk-on-Spheres
+
+- **ID:** `2603.01193` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Hrishikesh Viswanath, Hong Chul Nam, Xi Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.01193 · **PDF:** https://arxiv.org/pdf/2603.01193v2
+
+## 8745. Grokking as a Phase Transition between Competing Basins: a Singular Learning Theory Approach
+
+- **ID:** `2603.01192` · **primary:** `stat.ML` · **Submitted:** 2026-03-01
+- **Authors:** Ben Cullen, Sergio Estan-Ruiz, Riya Danait, et al.
+- **abs:** https://arxiv.org/abs/2603.01192 · **PDF:** https://arxiv.org/pdf/2603.01192v2
+
+## 8746. Reasoning or Rationalization? The Role of Justifications in Masked Diffusion Models for Fact Verification
+
+- **ID:** `2603.01190` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Jacob Devasier
+- **abs:** https://arxiv.org/abs/2603.01190 · **PDF:** https://arxiv.org/pdf/2603.01190v1
+
+## 8747. Token-level Data Selection for Safe LLM Fine-tuning
+
+- **ID:** `2603.01185` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Yanping Li, Zhening Liu, Zijian Li, et al.
+- **abs:** https://arxiv.org/abs/2603.01185 · **PDF:** https://arxiv.org/pdf/2603.01185v1
+
+## 8748. Scaling of learning time for high dimensional inputs
+
+- **ID:** `2603.01184` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Carlos Stein Brito
+- **abs:** https://arxiv.org/abs/2603.01184 · **PDF:** https://arxiv.org/pdf/2603.01184v1
+
+## 8749. PARWiS: Winner determination under shoestring budgets using active pairwise comparisons
+
+- **ID:** `2603.01171` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Shailendra Bhandari
+- **abs:** https://arxiv.org/abs/2603.01171 · **PDF:** https://arxiv.org/pdf/2603.01171v1
+
+## 8750. ATLAS: AI-Assisted Threat-to-Assertion Learning for System-on-Chip Security Verification
+
+- **ID:** `2603.01170` · **primary:** `cs.CR` · **Submitted:** 2026-03-01
+- **Authors:** Ishraq Tashdid, Kimia Tasnia, Alexander Garcia, et al.
+- **abs:** https://arxiv.org/abs/2603.01170 · **PDF:** https://arxiv.org/pdf/2603.01170v1
+
+## 8751. TripleSumm: Adaptive Triple-Modality Fusion for Video Summarization
+
+- **ID:** `2603.01169` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Sumin Kim, Hyemin Jeong, Mingu Kang, et al.
+- **abs:** https://arxiv.org/abs/2603.01169 · **PDF:** https://arxiv.org/pdf/2603.01169v1
+
+## 8752. Contextual Invertible World Models: A Neuro-Symbolic Agentic Framework for Colorectal Cancer Drug Response
+
+- **ID:** `2603.02274` · **primary:** `q-bio.QM` · **Submitted:** 2026-03-01
+- **Authors:** Christopher Baker, Karen Rafferty, Hui Wang
+- **abs:** https://arxiv.org/abs/2603.02274 · **PDF:** https://arxiv.org/pdf/2603.02274v1
+
+## 8753. SphUnc: Hyperspherical Uncertainty Decomposition and Causal Identification via Information Geometry
+
+- **ID:** `2603.01168` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Rong Fu, Chunlei Meng, Jinshuo Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.01168 · **PDF:** https://arxiv.org/pdf/2603.01168v1
+
+## 8754. DEP: A Decentralized Large Language Model Evaluation Protocol
+
+- **ID:** `2603.01167` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Jianxiang Peng, Junhao Li, Hongxiang Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.01167 · **PDF:** https://arxiv.org/pdf/2603.01167v1
+
+## 8755. Demystifying Group Relative Policy Optimization: Its Policy Gradient is a U-Statistic
+
+- **ID:** `2603.01162` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Hongyi Zhou, Kai Ye, Erhan Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.01162 · **PDF:** https://arxiv.org/pdf/2603.01162v3
+
+## 8756. GRAD-Former: Gated Robust Attention-based Differential Transformer for Change Detection
+
+- **ID:** `2603.01161` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Durgesh Ameta, Ujjwal Mishra, Praful Hambarde, et al.
+- **abs:** https://arxiv.org/abs/2603.01161 · **PDF:** https://arxiv.org/pdf/2603.01161v1
+
+## 8757. Semantic XPath: Structured Agentic Memory Access for Conversational AI
+
+- **ID:** `2603.01160` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Yifan Simon Liu, Ruifan Wu, Liam Gallagher, et al.
+- **abs:** https://arxiv.org/abs/2603.01160 · **PDF:** https://arxiv.org/pdf/2603.01160v1
+
+## 8758. HypeLoRA: Hyper-Network-Generated LoRA Adapters for Calibrated Language Model Fine-Tuning
+
+- **ID:** `2603.19278` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Bartosz Trojan, Filip Gębala
+- **abs:** https://arxiv.org/abs/2603.19278 · **PDF:** https://arxiv.org/pdf/2603.19278v2
+
+## 8759. A Principle-Driven Adaptive Policy for Group Cognitive Stimulation Dialogue for Elderly with Cognitive Impairment
+
+- **ID:** `2603.10034` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Jiyue Jiang, Yanyu Chen, Pengan Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.10034 · **PDF:** https://arxiv.org/pdf/2603.10034v1
+
+## 8760. Adaptive Window Selection for Financial Risk Forecasting
+
+- **ID:** `2603.01157` · **primary:** `q-fin.RM` · **Submitted:** 2026-03-01
+- **Authors:** Yinhuan Li, Chenxin Lyu, Ruodu Wang
+- **abs:** https://arxiv.org/abs/2603.01157 · **PDF:** https://arxiv.org/pdf/2603.01157v1
+
+## 8761. DeepResearch-9K: A Challenging Benchmark Dataset of Deep-Research Agent
+
+- **ID:** `2603.01152` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Tongzhou Wu, Yuhao Wang, Xinyu Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.01152 · **PDF:** https://arxiv.org/pdf/2603.01152v1
+
+## 8762. MOSAIC: Modular Opinion Summarization using Aspect Identification and Clustering
+
+- **ID:** `2603.19277` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Piyush Kumar Singh, Jayesh Choudhari
+- **abs:** https://arxiv.org/abs/2603.19277 · **PDF:** https://arxiv.org/pdf/2603.19277v1
+
+## 8763. ICPRL: Acquiring Physical Intuition from Interactive Control
+
+- **ID:** `2603.13295` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Xinrun Xu, Pi Bu, Ye Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.13295 · **PDF:** https://arxiv.org/pdf/2603.13295v1
+
+## 8764. AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution
+
+- **ID:** `2603.01145` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Yutao Yang, Junsong Li, Qianjun Pan, et al.
+- **abs:** https://arxiv.org/abs/2603.01145 · **PDF:** https://arxiv.org/pdf/2603.01145v2
+
+## 8765. A Decomposition Framework for Certifiably Optimal Orthogonal Sparse PCA
+
+- **ID:** `2603.01144` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Difei Cheng, Qiao Hu
+- **abs:** https://arxiv.org/abs/2603.01144 · **PDF:** https://arxiv.org/pdf/2603.01144v1
+
+## 8766. TC-SSA: Token Compression via Semantic Slot Aggregation for Gigapixel Pathology Reasoning
+
+- **ID:** `2603.01143` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Zhuo Chen, Shawn Young, Lijian Xu
+- **abs:** https://arxiv.org/abs/2603.01143 · **PDF:** https://arxiv.org/pdf/2603.01143v1
+
+## 8767. A Deep Learning Framework for Heat Demand Forecasting using Time-Frequency Representations of Decomposed Features
+
+- **ID:** `2603.01137` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Adithya Ramachandran, Satyaki Chatterjee, Thorkil Flensmark B. Neergaard, et al.
+- **abs:** https://arxiv.org/abs/2603.01137 · **PDF:** https://arxiv.org/pdf/2603.01137v1
+
+## 8768. FCN-LLM: Empower LLM for Brain Functional Connectivity Network Understanding via Graph-level Multi-task Instruction Tuning
+
+- **ID:** `2603.01135` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Xingcan Hu, Wei Wang, Li Xiao
+- **abs:** https://arxiv.org/abs/2603.01135 · **PDF:** https://arxiv.org/pdf/2603.01135v1
+
+## 8769. GameVerse: Can Vision-Language Models Learn from Video-based Reflection?
+
+- **ID:** `2603.06656` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Kuan Zhang, Dongchen Liu, Qiyue Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.06656 · **PDF:** https://arxiv.org/pdf/2603.06656v2
+
+## 8770. MedCollab: Causal-Driven Multi-Agent Collaboration for Full-Cycle Clinical Diagnosis via IBIS-Structured Argumentation
+
+- **ID:** `2603.01131` · **primary:** `cs.MA` · **Submitted:** 2026-03-01
+- **Authors:** Yuqi Zhan, Xinyue Wu, Tianyu Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.01131 · **PDF:** https://arxiv.org/pdf/2603.01131v1
+
+## 8771. Predictive Reasoning with Augmented Anomaly Contrastive Learning for Compositional Visual Relations
+
+- **ID:** `2603.01125` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Chengtai Li, Yuting He, Jianfeng Ren, et al.
+- **abs:** https://arxiv.org/abs/2603.01125 · **PDF:** https://arxiv.org/pdf/2603.01125v1
+
+## 8772. ClinCoT: Clinical-Aware Visual Chain-of-Thought for Medical Vision Language Models
+
+- **ID:** `2603.01124` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Xiwei Liu, Yulong Li, Xinlin Zhuang, et al.
+- **abs:** https://arxiv.org/abs/2603.01124 · **PDF:** https://arxiv.org/pdf/2603.01124v1
+
+## 8773. HVR-Met: A Hypothesis-Verification-Replaning Agentic System for Extreme Weather Diagnosis
+
+- **ID:** `2603.01121` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Shuo Tang, Jiadong Zhang, Jian Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.01121 · **PDF:** https://arxiv.org/pdf/2603.01121v1
+
+## 8774. Robust Weighted Triangulation of Causal Effects Under Model Uncertainty
+
+- **ID:** `2603.01119` · **primary:** `stat.ME` · **Submitted:** 2026-03-01
+- **Authors:** Rohit Bhattacharya, Ina Ocelli, Ted Westling
+- **abs:** https://arxiv.org/abs/2603.01119 · **PDF:** https://arxiv.org/pdf/2603.01119v1
+
+## 8775. Zipage: Maintain High Request Concurrency for LLM Reasoning through Compressed PagedAttention
+
+- **ID:** `2603.08743` · **primary:** `cs.DC` · **Submitted:** 2026-03-01
+- **Authors:** Mengqi Liao, Lu Wang, Chaoyun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.08743 · **PDF:** https://arxiv.org/pdf/2603.08743v1
+
+## 8776. DIVA-GRPO: Enhancing Multimodal Reasoning through Difficulty-Adaptive Variant Advantage
+
+- **ID:** `2603.01106` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Haowen Gao, Zhenyu Zhang, Liang Pang, et al.
+- **abs:** https://arxiv.org/abs/2603.01106 · **PDF:** https://arxiv.org/pdf/2603.01106v1
+
+## 8777. Egocentric Co-Pilot: Web-Native Smart-Glasses Agents for Assistive Egocentric AI
+
+- **ID:** `2603.01104` · **primary:** `cs.HC` · **Submitted:** 2026-03-01
+- **Authors:** Sicheng Yang, Yukai Huang, Weitong Cai, et al.
+- **abs:** https://arxiv.org/abs/2603.01104 · **PDF:** https://arxiv.org/pdf/2603.01104v1
+
+## 8778. Structure-preserving Randomized Neural Networks for Incompressible Magnetohydrodynamics Equations
+
+- **ID:** `2603.01102` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-03-01
+- **Authors:** Yunlong Li, Fei Wang, Lingxiao Li
+- **abs:** https://arxiv.org/abs/2603.01102 · **PDF:** https://arxiv.org/pdf/2603.01102v1
+
+## 8779. SyncTrack: Rhythmic Stability and Synchronization in Multi-Track Music Generation
+
+- **ID:** `2603.01101` · **primary:** `cs.SD` · **Submitted:** 2026-03-01
+- **Authors:** Hongrui Wang, Fan Zhang, Zhiyuan Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.01101 · **PDF:** https://arxiv.org/pdf/2603.01101v1
+
+## 8780. Differential privacy representation geometry for medical image analysis
+
+- **ID:** `2603.01098` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Soroosh Tayebi Arasteh, Marziyeh Mohammadi, Sven Nebelung, et al.
+- **abs:** https://arxiv.org/abs/2603.01098 · **PDF:** https://arxiv.org/pdf/2603.01098v1
+
+## 8781. Understanding LoRA as Knowledge Memory: An Empirical Analysis
+
+- **ID:** `2603.01097` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Seungju Back, Dongwoo Lee, Naun Kang, et al.
+- **abs:** https://arxiv.org/abs/2603.01097 · **PDF:** https://arxiv.org/pdf/2603.01097v1
+
+## 8782. Unified Vision-Language Modeling via Concept Space Alignment
+
+- **ID:** `2603.01096` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Yifu Qiu, Paul-Ambroise Duquenne, Holger Schwenk
+- **abs:** https://arxiv.org/abs/2603.01096 · **PDF:** https://arxiv.org/pdf/2603.01096v1
+
+## 8783. Adaptive-Growth Randomized Neural Networks for Level-Set Computation of Multivalued Nonlinear First-Order PDEs with Hyperbolic Characteristics
+
+- **ID:** `2603.01093` · **primary:** `math.NA` · **Submitted:** 2026-03-01
+- **Authors:** Haoning Dang, Shi Jin, Fei Wang
+- **abs:** https://arxiv.org/abs/2603.01093 · **PDF:** https://arxiv.org/pdf/2603.01093v1
+
+## 8784. Alien Science: Sampling Coherent but Cognitively Unavailable Research Directions from Idea Atoms
+
+- **ID:** `2603.01092` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Alejandro H. Artiles, Martin Weiss, Levin Brinkmann, et al.
+- **abs:** https://arxiv.org/abs/2603.01092 · **PDF:** https://arxiv.org/pdf/2603.01092v1
+
+## 8785. CARD: Towards Conditional Design of Multi-agent Topological Structures
+
+- **ID:** `2603.01089` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Tongtong Wu, Yanming Li, Ziye Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.01089 · **PDF:** https://arxiv.org/pdf/2603.01089v1
+
+## 8786. Super-resolution of turbulent reacting flows on complex meshes using graph neural networks
+
+- **ID:** `2603.01080` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-03-01
+- **Authors:** Priyabrat Dash, Konduri Aditya, Christos E. Frouzakis, et al.
+- **abs:** https://arxiv.org/abs/2603.01080 · **PDF:** https://arxiv.org/pdf/2603.01080v1
+
+## 8787. Kernel Methods for Stochastic Dynamical Systems with Application to Koopman Eigenfunctions: Feynman-Kac Representations and RKHS Approximation
+
+- **ID:** `2603.01077` · **primary:** `math.DS` · **Submitted:** 2026-03-01
+- **Authors:** Boumediene Hamzi, Houman Owhadi, Umesh Vaidya
+- **abs:** https://arxiv.org/abs/2603.01077 · **PDF:** https://arxiv.org/pdf/2603.01077v1
+
+## 8788. Feasible Pairings for Decentralized Integral Controllability of Non-Square Systems
+
+- **ID:** `2603.01076` · **primary:** `math.OC` · **Submitted:** 2026-03-01
+- **Authors:** Yuhao Tong, Steven W. Su
+- **abs:** https://arxiv.org/abs/2603.01076 · **PDF:** https://arxiv.org/pdf/2603.01076v1
+
+## 8789. How RL Unlocks the Aha Moment in Geometric Interleaved Reasoning
+
+- **ID:** `2603.01070` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Xiangxiang Zhang, Caijun Jia, Siyuan Li, et al.
+- **abs:** https://arxiv.org/abs/2603.01070 · **PDF:** https://arxiv.org/pdf/2603.01070v1
+
+## 8790. LLaDA-o: An Effective and Length-Adaptive Omni Diffusion Model
+
+- **ID:** `2603.01068` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Zebin You, Xiaolu Zhang, Jun Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.01068 · **PDF:** https://arxiv.org/pdf/2603.01068v1
+
+## 8791. Hide&Seek: Remove Image Watermarks with Negligible Cost via Pixel-wise Reconstruction
+
+- **ID:** `2603.01067` · **primary:** `cs.CR` · **Submitted:** 2026-03-01
+- **Authors:** Huajie Chen, Tianqing Zhu, Hailin Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.01067 · **PDF:** https://arxiv.org/pdf/2603.01067v1
+
+## 8792. A level-wise training scheme for learning neural multigrid smoothers with application to integral equations
+
+- **ID:** `2603.01064` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Lingfeng Li, Yin King Chu, Raymond Chan, et al.
+- **abs:** https://arxiv.org/abs/2603.01064 · **PDF:** https://arxiv.org/pdf/2603.01064v1
+
+## 8793. GroupGPT: A Token-efficient and Privacy-preserving Agentic Framework for Multi-User Chat Assistant
+
+- **ID:** `2603.01059` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Zhuokang Shen, Yifan Wang, Hanyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.01059 · **PDF:** https://arxiv.org/pdf/2603.01059v2
+
+## 8794. TriMoE: Augmenting GPU with AMX-Enabled CPU and DIMM-NDP for High-Throughput MoE Inference via Offloading
+
+- **ID:** `2603.01058` · **primary:** `cs.AR` · **Submitted:** 2026-03-01
+- **Authors:** Yudong Pan, Yintao He, Tianhua Han, et al.
+- **abs:** https://arxiv.org/abs/2603.01058 · **PDF:** https://arxiv.org/pdf/2603.01058v1
+
+## 8795. MMCOMET: A Large-Scale Multimodal Commonsense Knowledge Graph for Contextual Reasoning
+
+- **ID:** `2603.01055` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Eileen Wang, Hiba Arnaout, Dhita Pratama, et al.
+- **abs:** https://arxiv.org/abs/2603.01055 · **PDF:** https://arxiv.org/pdf/2603.01055v1
+
+## 8796. Turning Black Box into White Box: Dataset Distillation Leaks
+
+- **ID:** `2603.01053` · **primary:** `cs.CR` · **Submitted:** 2026-03-01
+- **Authors:** Huajie Chen, Tianqing Zhu, Yuchen Zhong, et al.
+- **abs:** https://arxiv.org/abs/2603.01053 · **PDF:** https://arxiv.org/pdf/2603.01053v1
+
+## 8797. No More Maybe-Arrows: Resolving Causal Uncertainty by Breaking Symmetries
+
+- **ID:** `2603.01052` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Tingrui Huang, Devendra Singh Dhami
+- **abs:** https://arxiv.org/abs/2603.01052 · **PDF:** https://arxiv.org/pdf/2603.01052v1
+
+## 8798. MM-DeepResearch: A Simple and Effective Multimodal Agentic Search Baseline
+
+- **ID:** `2603.01050` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Huanjin Yao, Qixiang Yin, Min Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.01050 · **PDF:** https://arxiv.org/pdf/2603.01050v1
+
+## 8799. RepoRepair: Leveraging Code Documentation for Repository-Level Automated Program Repair
+
+- **ID:** `2603.01048` · **primary:** `cs.SE` · **Submitted:** 2026-03-01
+- **Authors:** Zhongqiang Pan, Chuanyi Li, Wenkang Zhong, et al.
+- **abs:** https://arxiv.org/abs/2603.01048 · **PDF:** https://arxiv.org/pdf/2603.01048v1
+
+## 8800. Evaluating GFlowNet from partial episodes for stable and flexible policy-based training
+
+- **ID:** `2603.01047` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Puhua Niu, Shili Wu, Xiaoning Qian
+- **abs:** https://arxiv.org/abs/2603.01047 · **PDF:** https://arxiv.org/pdf/2603.01047v1
+
+## 8801. Silo-Bench: A Scalable Environment for Evaluating Distributed Coordination in Multi-Agent LLM Systems
+
+- **ID:** `2603.01045` · **primary:** `cs.MA` · **Submitted:** 2026-03-01
+- **Authors:** Yuzhe Zhang, Feiran Liu, Yi Shan, et al.
+- **abs:** https://arxiv.org/abs/2603.01045 · **PDF:** https://arxiv.org/pdf/2603.01045v1
+
+## 8802. Thoth: Mid-Training Bridges LLMs to Time Series Understanding
+
+- **ID:** `2603.01042` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Jiafeng Lin, Yuxuan Wang, Jialong Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.01042 · **PDF:** https://arxiv.org/pdf/2603.01042v1
+
+## 8803. Fed-ADE: Adaptive Learning Rate for Federated Post-adaptation under Distribution Shift
+
+- **ID:** `2603.01040` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Heewon Park, Mugon Joe, Miru Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.01040 · **PDF:** https://arxiv.org/pdf/2603.01040v1
+
+## 8804. From Intuition to Investigation: A Tool-Augmented Reasoning MLLM Framework for Generalizable Face Anti-Spoofing
+
+- **ID:** `2603.01038` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Haoyuan Zhang, Keyao Wang, Guosheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.01038 · **PDF:** https://arxiv.org/pdf/2603.01038v2
+
+## 8805. Reparameterized Tensor Ring Functional Decomposition for Multi-Dimensional Data Recovery
+
+- **ID:** `2603.01034` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Yangyang Xu, Junbo Ke, You-Wei Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.01034 · **PDF:** https://arxiv.org/pdf/2603.01034v2
+
+## 8806. Content-Aware Frequency Encoding for Implicit Neural Representations with Fourier-Chebyshev Features
+
+- **ID:** `2603.01028` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Junbo Ke, Yangyang Xu, You-Wei Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.01028 · **PDF:** https://arxiv.org/pdf/2603.01028v1
+
+## 8807. One-Token Verification for Reasoning Correctness Estimation
+
+- **ID:** `2603.01025` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Zhan Zhuang, Xiequn Wang, Zebin Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.01025 · **PDF:** https://arxiv.org/pdf/2603.01025v1
+
+## 8808. SimAB: Simulating A/B Tests with Persona-Conditioned AI Agents for Rapid Design Evaluation
+
+- **ID:** `2603.01024` · **primary:** `cs.HC` · **Submitted:** 2026-03-01
+- **Authors:** Tim Rieder, Marian Schneider, Mario Truss, et al.
+- **abs:** https://arxiv.org/abs/2603.01024 · **PDF:** https://arxiv.org/pdf/2603.01024v1
+
+## 8809. An Open-Source Modular Benchmark for Diffusion-Based Motion Planning in Closed-Loop Autonomous Driving
+
+- **ID:** `2603.01023` · **primary:** `cs.RO` · **Submitted:** 2026-03-01
+- **Authors:** Yun Li, Simon Thompson, Yidu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.01023 · **PDF:** https://arxiv.org/pdf/2603.01023v1
+
+## 8810. BadRSSD: Backdoor Attacks on Regularized Self-Supervised Diffusion Models
+
+- **ID:** `2603.01019` · **primary:** `cs.CR` · **Submitted:** 2026-03-01
+- **Authors:** Jiayao Wang, Yiping Zhang, Mohammad Maruf Hasan, et al.
+- **abs:** https://arxiv.org/abs/2603.01019 · **PDF:** https://arxiv.org/pdf/2603.01019v1
+
+## 8811. Feature-Weighted Maximum Representative Subsampling
+
+- **ID:** `2603.01013` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Tony Hauptmann, Stefan Kramer
+- **abs:** https://arxiv.org/abs/2603.01013 · **PDF:** https://arxiv.org/pdf/2603.01013v1
+
+## 8812. FastCode: Fast and Cost-Efficient Code Understanding and Reasoning
+
+- **ID:** `2603.01012` · **primary:** `cs.SE` · **Submitted:** 2026-03-01
+- **Authors:** Zhonghang Li, Zongwei Li, Yuxuan Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.01012 · **PDF:** https://arxiv.org/pdf/2603.01012v2
+
+## 8813. Qayyem: A Real-time Platform for Scoring Proficiency of Arabic Essays
+
+- **ID:** `2603.01009` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Hoor Elbahnasawi, Marwan Sayed, Sohaila Eltanbouly, et al.
+- **abs:** https://arxiv.org/abs/2603.01009 · **PDF:** https://arxiv.org/pdf/2603.01009v1
+
+## 8814. AG-REPA: Causal Layer Selection for Representation Alignment in Audio Flow Matching
+
+- **ID:** `2603.01006` · **primary:** `cs.SD` · **Submitted:** 2026-03-01
+- **Authors:** Pengfei Zhang, Tianxin Xie, Minghao Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.01006 · **PDF:** https://arxiv.org/pdf/2603.01006v1
+
+## 8815. Data-Free PINNs for Compressible Flows: Mitigating Spectral Bias and Gradient Pathologies via Mach-Guided Scaling and Hybrid Convolutions
+
+- **ID:** `2603.01001` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-03-01
+- **Authors:** Ryosuke Yano
+- **abs:** https://arxiv.org/abs/2603.01001 · **PDF:** https://arxiv.org/pdf/2603.01001v1
+
+## 8816. DWAFM: Dynamic Weighted Graph Structure Embedding Integrated with Attention and Frequency-Domain MLPs for Traffic Forecasting
+
+- **ID:** `2603.00997` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Sen Shi, Zhichao Zhang, Yangfan He
+- **abs:** https://arxiv.org/abs/2603.00997 · **PDF:** https://arxiv.org/pdf/2603.00997v1
+
+## 8817. Fully-analog array signal processor using 3D aperture engineering
+
+- **ID:** `2603.00995` · **primary:** `physics.app-ph` · **Submitted:** 2026-03-01
+- **Authors:** Sheng Gao, Songtao Yang, Haiou Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.00995 · **PDF:** https://arxiv.org/pdf/2603.00995v1
+
+## 8818. CollabEval: Enhancing LLM-as-a-Judge via Multi-Agent Collaboration
+
+- **ID:** `2603.00993` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Yiyue Qian, Shinan Zhang, Yun Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.00993 · **PDF:** https://arxiv.org/pdf/2603.00993v1
+
+## 8819. Compensation-free Machine Unlearning in Text-to-Image Diffusion Models by Eliminating the Mutual Information
+
+- **ID:** `2603.00992` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Xinwen Cheng, Jingyuan Zhang, Zhehao Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.00992 · **PDF:** https://arxiv.org/pdf/2603.00992v1
+
+## 8820. Tracking Capabilities for Safer Agents
+
+- **ID:** `2603.00991` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Martin Odersky, Yaoyu Zhao, Yichen Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.00991 · **PDF:** https://arxiv.org/pdf/2603.00991v1
+
+## 8821. SoberDSE: Sample-Efficient Design Space Exploration via Learning-Based Algorithm Selection
+
+- **ID:** `2603.00986` · **primary:** `cs.AR` · **Submitted:** 2026-03-01
+- **Authors:** Lei Xu, Shanshan Wang, Chenglong Xiao
+- **abs:** https://arxiv.org/abs/2603.00986 · **PDF:** https://arxiv.org/pdf/2603.00986v1
+
+## 8822. EraseAnything++: Enabling Concept Erasure in Rectified Flow Transformers Leveraging Multi-Object Optimization
+
+- **ID:** `2603.00978` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Zhaoxin Fan, Nanxiang Jiang, Daiheng Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.00978 · **PDF:** https://arxiv.org/pdf/2603.00978v1
+
+## 8823. HiMAC: Hierarchical Macro-Micro Learning for Long-Horizon LLM Agents
+
+- **ID:** `2603.00977` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Hongbo Jin, Rongpeng Zhu, Jiayu Ding, et al.
+- **abs:** https://arxiv.org/abs/2603.00977 · **PDF:** https://arxiv.org/pdf/2603.00977v1
+
+## 8824. Forgetting is Competition: Rethinking Unlearning as Representation Interference in Diffusion Models
+
+- **ID:** `2603.00975` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Ashutosh Ranjan, Vivek Srivastava, Shirish Karande, et al.
+- **abs:** https://arxiv.org/abs/2603.00975 · **PDF:** https://arxiv.org/pdf/2603.00975v1
+
+## 8825. Intent-Context Synergy Reinforcement Learning for Autonomous UAV Decision-Making in Air Combat
+
+- **ID:** `2603.00974` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Jiahao Fu, Feng Yang
+- **abs:** https://arxiv.org/abs/2603.00974 · **PDF:** https://arxiv.org/pdf/2603.00974v1
+
+## 8826. Random Features for Operator-Valued Kernels: Bridging Kernel Methods and Neural Operators
+
+- **ID:** `2603.00971` · **primary:** `stat.ML` · **Submitted:** 2026-03-01
+- **Authors:** Mike Nguyen, Nicole Mücke
+- **abs:** https://arxiv.org/abs/2603.00971 · **PDF:** https://arxiv.org/pdf/2603.00971v1
+
+## 8827. Learning with the Nash-Sutcliffe loss
+
+- **ID:** `2603.00968` · **primary:** `stat.ML` · **Submitted:** 2026-03-01
+- **Authors:** Hristos Tyralis, Georgia Papacharalampous
+- **abs:** https://arxiv.org/abs/2603.00968 · **PDF:** https://arxiv.org/pdf/2603.00968v1
+
+## 8828. Stabilizing Policy Optimization via Logits Convexity
+
+- **ID:** `2603.00963` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Hongzhan Chen, Tao Yang, Yuhua Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.00963 · **PDF:** https://arxiv.org/pdf/2603.00963v1
+
+## 8829. AWE: Adaptive Agents for Dynamic Web Penetration Testing
+
+- **ID:** `2603.00960` · **primary:** `cs.CR` · **Submitted:** 2026-03-01
+- **Authors:** Akshat Singh Jaswal, Ashish Baghel
+- **abs:** https://arxiv.org/abs/2603.00960 · **PDF:** https://arxiv.org/pdf/2603.00960v1
+
+## 8830. S-VoCAL: A Dataset and Evaluation Framework for Inferring Speaking Voice Character Attributes in Literature
+
+- **ID:** `2603.00958` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Abigail Berthe-Pardo, Gaspard Michel, Elena V. Epure, et al.
+- **abs:** https://arxiv.org/abs/2603.00958 · **PDF:** https://arxiv.org/pdf/2603.00958v1
+
+## 8831. Enhancing Safety of Large Language Models via Embedding Space Separation
+
+- **ID:** `2603.20206` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Xu Zhao, Xiting Wang, Weiran Shen
+- **abs:** https://arxiv.org/abs/2603.20206 · **PDF:** https://arxiv.org/pdf/2603.20206v1
+
+## 8832. Beyond False Discovery Rate: A Stepdown Group SLOPE Approach for Grouped Variable Selection
+
+- **ID:** `2603.00955` · **primary:** `stat.ME` · **Submitted:** 2026-03-01
+- **Authors:** Xuelin Zhang, Jingxuan Liang, Xinyue Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.00955 · **PDF:** https://arxiv.org/pdf/2603.00955v1
+
+## 8833. When Does Margin Clamping Affect Training Variance? Dataset-Dependent Effects in Contrastive Forward-Forward Learning
+
+- **ID:** `2603.00951` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Joshua Steier
+- **abs:** https://arxiv.org/abs/2603.00951 · **PDF:** https://arxiv.org/pdf/2603.00951v1
+
+## 8834. Graph Attention Based Prioritization of Disease Responsible Genes from Multimodal Alzheimer's Network
+
+- **ID:** `2603.02273` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Binon Teji, Subhajit Bandyopadhyay, Swarup Roy
+- **abs:** https://arxiv.org/abs/2603.02273 · **PDF:** https://arxiv.org/pdf/2603.02273v1
+
+## 8835. Non-Rectangular Average-Reward Robust MDPs: Optimal Policies and Their Transient Values
+
+- **ID:** `2603.00945` · **primary:** `math.OC` · **Submitted:** 2026-03-01
+- **Authors:** Shengbo Wang, Nian Si
+- **abs:** https://arxiv.org/abs/2603.00945 · **PDF:** https://arxiv.org/pdf/2603.00945v3
+
+## 8836. Towards Orthographically-Informed Evaluation of Speech Recognition Systems for Indian Languages
+
+- **ID:** `2603.00941` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Kaushal Santosh Bhogale, Tahir Javed, Greeshma Susan John, et al.
+- **abs:** https://arxiv.org/abs/2603.00941 · **PDF:** https://arxiv.org/pdf/2603.00941v1
+
+## 8837. Seeing Beyond 8bits: Subjective and Objective Quality Assessment of HDR-UGC Videos
+
+- **ID:** `2603.00938` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Shreshth Saini, Bowen Chen, Neil Birkbeck, et al.
+- **abs:** https://arxiv.org/abs/2603.00938 · **PDF:** https://arxiv.org/pdf/2603.00938v1
+
+## 8838. Time-Aware Latent Space Bayesian Optimization
+
+- **ID:** `2603.00935` · **primary:** `stat.ML` · **Submitted:** 2026-03-01
+- **Authors:** Tuan A. Vu, Julien Martinelli, Harri Lähdesmäki
+- **abs:** https://arxiv.org/abs/2603.00935 · **PDF:** https://arxiv.org/pdf/2603.00935v1
+
+## 8839. The Aftermath of DrawEduMath: Vision Language Models Underperform with Struggling Students and Misdiagnose Errors
+
+- **ID:** `2603.00925` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Li Lucy, Albert Zhang, Nathan Anderson, et al.
+- **abs:** https://arxiv.org/abs/2603.00925 · **PDF:** https://arxiv.org/pdf/2603.00925v1
+
+## 8840. Conformal Prediction for Risk-Controlled Medical Entity Extraction Across Clinical Domains
+
+- **ID:** `2603.00924` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Manil Shrestha, Edward Kim
+- **abs:** https://arxiv.org/abs/2603.00924 · **PDF:** https://arxiv.org/pdf/2603.00924v2
+
+## 8841. Hybrid Neural-LLM Pipeline for Morphological Glossing in Endangered Language Documentation: A Case Study of Jungar Tuvan
+
+- **ID:** `2603.00923` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Siyu Liang, Talant Mawkanuli, Gina-Anne Levow
+- **abs:** https://arxiv.org/abs/2603.00923 · **PDF:** https://arxiv.org/pdf/2603.00923v1
+
+## 8842. Improving Text-to-Image Generation with Intrinsic Self-Confidence Rewards
+
+- **ID:** `2603.00918` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Seungwook Kim, Minsu Cho
+- **abs:** https://arxiv.org/abs/2603.00918 · **PDF:** https://arxiv.org/pdf/2603.00918v2
+
+## 8843. Prompt Sensitivity and Answer Consistency of Small Open-Source Language Models for Clinical Question Answering in Low-Resource Healthcare
+
+- **ID:** `2603.00917` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Shravani Hariprasad
+- **abs:** https://arxiv.org/abs/2603.00917 · **PDF:** https://arxiv.org/pdf/2603.00917v3
+
+## 8844. Curvature-Weighted Capacity Allocation: A Minimum Description Length Framework for Layer-Adaptive Large Language Model Optimization
+
+- **ID:** `2603.00910` · **primary:** `cs.IT` · **Submitted:** 2026-03-01
+- **Authors:** Theophilus Amaefuna, Hitesh Vaidya, Anshuman Chhabra, et al.
+- **abs:** https://arxiv.org/abs/2603.00910 · **PDF:** https://arxiv.org/pdf/2603.00910v1
+
+## 8845. KVSlimmer: Theoretical Insights and Practical Optimizations for Asymmetric KV Merging
+
+- **ID:** `2603.00907` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Lianjun Liu, Hongli An, Weiqi Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.00907 · **PDF:** https://arxiv.org/pdf/2603.00907v2
+
+## 8846. Principled Fast and Meta Knowledge Learners for Continual Reinforcement Learning
+
+- **ID:** `2603.00903` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Ke Sun, Hongming Zhang, Jun Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.00903 · **PDF:** https://arxiv.org/pdf/2603.00903v1
+
+## 8847. Evaluating AI Grading on Real-World Handwritten College Mathematics: A Large-Scale Study Toward a Benchmark
+
+- **ID:** `2603.00895` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Zhiqi Yu, Xingping Liu, Haobin Mao, et al.
+- **abs:** https://arxiv.org/abs/2603.00895 · **PDF:** https://arxiv.org/pdf/2603.00895v1
+
+## 8848. CHIMERA: Compact Synthetic Data for Generalizable LLM Reasoning
+
+- **ID:** `2603.00889` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Xinyu Zhu, Yihao Feng, Yanchao Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.00889 · **PDF:** https://arxiv.org/pdf/2603.00889v1
+
+## 8849. Probabilistic Learning and Generation in Deep Sequence Models
+
+- **ID:** `2603.00888` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Wenlong Chen
+- **abs:** https://arxiv.org/abs/2603.00888 · **PDF:** https://arxiv.org/pdf/2603.00888v1
+
+## 8850. Knowledge without Wisdom: Measuring Misalignment between LLMs and Intended Impact
+
+- **ID:** `2603.00883` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Michael Hardy, Yunsung Kim
+- **abs:** https://arxiv.org/abs/2603.00883 · **PDF:** https://arxiv.org/pdf/2603.00883v1
+
+## 8851. Active Flow Matching
+
+- **ID:** `2603.00877` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Yashvir S. Grewal, Daniel M. Steinberg, Thang D. Bui, et al.
+- **abs:** https://arxiv.org/abs/2603.00877 · **PDF:** https://arxiv.org/pdf/2603.00877v1
+
+## 8852. BioProAgent: Neuro-Symbolic Grounding for Constrained Scientific Planning
+
+- **ID:** `2603.00876` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Yuyang Liu, Jingya Wang, Liuzhenghao Lv, et al.
+- **abs:** https://arxiv.org/abs/2603.00876 · **PDF:** https://arxiv.org/pdf/2603.00876v1
+
+## 8853. MC-Search: Evaluating and Enhancing Multimodal Agentic Search with Structured Long Reasoning Chains
+
+- **ID:** `2603.00873` · **primary:** `cs.AI` · **Submitted:** 2026-03-01
+- **Authors:** Xuying Ning, Dongqi Fu, Tianxin Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.00873 · **PDF:** https://arxiv.org/pdf/2603.00873v1
+
+## 8854. PPC-MT: Parallel Point Cloud Completion with Mamba-Transformer Hybrid Architecture
+
+- **ID:** `2603.00870` · **primary:** `cs.CV` · **Submitted:** 2026-03-01
+- **Authors:** Jie Li, Shengwei Tian, Long Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.00870 · **PDF:** https://arxiv.org/pdf/2603.00870v1
+
+## 8855. AMDS: Attack-Aware Multi-Stage Defense System for Network Intrusion Detection with Two-Stage Adaptive Weight Learning
+
+- **ID:** `2603.00859` · **primary:** `cs.CR` · **Submitted:** 2026-03-01
+- **Authors:** Oluseyi Olukola, Nick Rahimi
+- **abs:** https://arxiv.org/abs/2603.00859 · **PDF:** https://arxiv.org/pdf/2603.00859v1
+
+## 8856. Artificial Superintelligence May be Useless: Equilibria in the Economy of Multiple AI Agents
+
+- **ID:** `2603.00858` · **primary:** `econ.TH` · **Submitted:** 2026-03-01
+- **Authors:** Huan Cai, Ziqing Lu, Catherine Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.00858 · **PDF:** https://arxiv.org/pdf/2603.00858v1
+
+## 8857. MultiPUFFIN: A Multimodal Domain-Constrained Foundation Model for Molecular Property Prediction of Small Molecules
+
+- **ID:** `2603.00857` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Idelfonso B. R. Nogueira, Carine M. Rebelloa, Mumin Enis Leblebici, et al.
+- **abs:** https://arxiv.org/abs/2603.00857 · **PDF:** https://arxiv.org/pdf/2603.00857v1
+
+## 8858. PARCER as an Operational Contract to Reduce Variance, Cost, and Risk in LLM Systems
+
+- **ID:** `2603.00856` · **primary:** `cs.SE` · **Submitted:** 2026-03-01
+- **Authors:** Elzo Brito dos Santos Filho
+- **abs:** https://arxiv.org/abs/2603.00856 · **PDF:** https://arxiv.org/pdf/2603.00856v1
+
+## 8859. Characterizing VLA Models: Identifying the Action Generation Bottleneck for Edge AI Architectures
+
+- **ID:** `2603.02271` · **primary:** `cs.PF` · **Submitted:** 2026-03-01
+- **Authors:** Manoj Vishwanathan, Suvinay Subramanian, Anand Raghunathan
+- **abs:** https://arxiv.org/abs/2603.02271 · **PDF:** https://arxiv.org/pdf/2603.02271v1
+
+## 8860. Navigating Time's Possibilities: Plausible Counterfactual Explanations for Multivariate Time-Series Forecast through Genetic Algorithms
+
+- **ID:** `2603.00855` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Gianlucca Zuin, Adriano Veloso
+- **abs:** https://arxiv.org/abs/2603.00855 · **PDF:** https://arxiv.org/pdf/2603.00855v1
+
+## 8861. GeMi: A Graph-based, Multimodal Recommendation System for Narrative Scroll Paintings
+
+- **ID:** `2603.00854` · **primary:** `cs.LG` · **Submitted:** 2026-03-01
+- **Authors:** Haimonti Dutta, Pruthvi Moluguri, Jin Dai, et al.
+- **abs:** https://arxiv.org/abs/2603.00854 · **PDF:** https://arxiv.org/pdf/2603.00854v1
+
+## 8862. Tiny-Critic RAG: Empowering Agentic Fallback with Parameter-Efficient Small Language Models
+
+- **ID:** `2603.00846` · **primary:** `cs.IR` · **Submitted:** 2026-03-01
+- **Authors:** Yichao Wu, Penghao Liang, Yafei Xiang, et al.
+- **abs:** https://arxiv.org/abs/2603.00846 · **PDF:** https://arxiv.org/pdf/2603.00846v1
+
+## 8863. MedGPT-oss: Training a General-Purpose Vision-Language Model for Biomedicine
+
+- **ID:** `2603.00842` · **primary:** `cs.CL` · **Submitted:** 2026-03-01
+- **Authors:** Kai Zhang, Zhengqing Yuan, Cheng Peng, et al.
+- **abs:** https://arxiv.org/abs/2603.00842 · **PDF:** https://arxiv.org/pdf/2603.00842v1
+
+## 8864. Learning Nested Named Entity Recognition from Flat Annotations
+
+- **ID:** `2603.00840` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Igor Rozhkov, Natalia Loukachevitch
+- **abs:** https://arxiv.org/abs/2603.00840 · **PDF:** https://arxiv.org/pdf/2603.00840v1
+
+## 8865. Constitutional Black-Box Monitoring for Scheming in LLM Agents
+
+- **ID:** `2603.00829` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Simon Storf, Rich Barton-Cooper, James Peters-Gill, et al.
+- **abs:** https://arxiv.org/abs/2603.00829 · **PDF:** https://arxiv.org/pdf/2603.00829v1
+
+## 8866. A Gauge Theory of Superposition: Toward a Sheaf-Theoretic Atlas of Neural Representations
+
+- **ID:** `2603.00824` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Hossein Javidnia
+- **abs:** https://arxiv.org/abs/2603.00824 · **PDF:** https://arxiv.org/pdf/2603.00824v2
+
+## 8867. A Comprehensive Evaluation of LLM Unlearning Robustness under Multi-Turn Interaction
+
+- **ID:** `2603.00823` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Ruihao Pan, Suhang Wang
+- **abs:** https://arxiv.org/abs/2603.00823 · **PDF:** https://arxiv.org/pdf/2603.00823v2
+
+## 8868. ContextCov: Deriving and Enforcing Executable Constraints from Agent Instruction Files
+
+- **ID:** `2603.00822` · **primary:** `cs.SE` · **Submitted:** 2026-02-28
+- **Authors:** Reshabh K Sharma
+- **abs:** https://arxiv.org/abs/2603.00822 · **PDF:** https://arxiv.org/pdf/2603.00822v1
+
+## 8869. A short tour of operator learning theory: Convergence rates, statistical limits, and open questions
+
+- **ID:** `2603.00819` · **primary:** `math.NA` · **Submitted:** 2026-02-28
+- **Authors:** Simone Brugiapaglia, Nicola Rares Franco, Nicholas H. Nelsen
+- **abs:** https://arxiv.org/abs/2603.00819 · **PDF:** https://arxiv.org/pdf/2603.00819v1
+
+## 8870. Wave-Attractor-Tree: A Hierarchical Binary Tree Reduction Architecture for Efficient Sequence Modeling
+
+- **ID:** `2603.00812` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Igor Berezkin
+- **abs:** https://arxiv.org/abs/2603.00812 · **PDF:** https://arxiv.org/pdf/2603.00812v1
+
+## 8871. Curation Leaks: Membership Inference Attacks against Data Curation for Machine Learning
+
+- **ID:** `2603.00811` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Dariush Wahdany, Matthew Jagielski, Adam Dziedzic, et al.
+- **abs:** https://arxiv.org/abs/2603.00811 · **PDF:** https://arxiv.org/pdf/2603.00811v1
+
+## 8872. MetaMind: General and Cognitive World Models in Multi-Agent Systems by Meta-Theory of Mind
+
+- **ID:** `2603.00808` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Lingyi Wang, Rashed Shelim, Walid Saad, et al.
+- **abs:** https://arxiv.org/abs/2603.00808 · **PDF:** https://arxiv.org/pdf/2603.00808v1
+
+## 8873. Lookahead identification in adversarial bandits: accuracy and memory bounds
+
+- **ID:** `2603.00803` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Nataly Brukhim, Nicolò Cesa-Bianchi, Carlo Ciliberto
+- **abs:** https://arxiv.org/abs/2603.00803 · **PDF:** https://arxiv.org/pdf/2603.00803v1
+
+## 8874. The Synthetic Web: Adversarially-Curated Mini-Internets for Diagnosing Epistemic Weaknesses of Language Agents
+
+- **ID:** `2603.00801` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Shrey Shah, Levent Ozgur
+- **abs:** https://arxiv.org/abs/2603.00801 · **PDF:** https://arxiv.org/pdf/2603.00801v1
+
+## 8875. Efficient Conformal Volumetry for Template-Based Segmentation
+
+- **ID:** `2603.00798` · **primary:** `eess.IV` · **Submitted:** 2026-02-28
+- **Authors:** Matt Y. Cheung, Ashok Veeraraghavan, Guha Balakrishnan
+- **abs:** https://arxiv.org/abs/2603.00798 · **PDF:** https://arxiv.org/pdf/2603.00798v1
+
+## 8876. PRISM: Exploring Heterogeneous Pretrained EEG Foundation Model Transfer to Clinical Differential Diagnosis
+
+- **ID:** `2603.02268` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Jeet Bandhu Lahiri, Parshva Runwal, Arvasu Kulkarni, et al.
+- **abs:** https://arxiv.org/abs/2603.02268 · **PDF:** https://arxiv.org/pdf/2603.02268v1
+
+## 8877. Neural Latent Arbitrary Lagrangian-Eulerian Grids for Fluid-Solid Interaction
+
+- **ID:** `2603.00792` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Shilong Tao, Zhe Feng, Shaohan Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.00792 · **PDF:** https://arxiv.org/pdf/2603.00792v1
+
+## 8878. Identifying the Geographic Foci of US Local News
+
+- **ID:** `2603.00787` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Gangani Ariyarathne, Isuru Ariyarathne, Greatness Emmanuel-King, et al.
+- **abs:** https://arxiv.org/abs/2603.00787 · **PDF:** https://arxiv.org/pdf/2603.00787v1
+
+## 8879. Interpretable Cross-Network Attention for Resting-State fMRI Representation Learning
+
+- **ID:** `2603.00786` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Karanpartap Singh, Adam Turnbull, Mohammad Abbasi, et al.
+- **abs:** https://arxiv.org/abs/2603.00786 · **PDF:** https://arxiv.org/pdf/2603.00786v1
+
+## 8880. QANTIS: A Hardware-Validated Quantum Platform for POMDP Planning and Multi-Target Data Association
+
+- **ID:** `2603.00785` · **primary:** `quant-ph` · **Submitted:** 2026-02-28
+- **Authors:** Bayram Yüksel Eker, Suayb S. Arslan, Özgür Nazlı, et al.
+- **abs:** https://arxiv.org/abs/2603.00785 · **PDF:** https://arxiv.org/pdf/2603.00785v1
+
+## 8881. Initialization-Aware Score-Based Diffusion Sampling
+
+- **ID:** `2603.00772` · **primary:** `stat.ML` · **Submitted:** 2026-02-28
+- **Authors:** Tiziano Fassina, Gabriel Cardoso, Sylvan Le Corff, et al.
+- **abs:** https://arxiv.org/abs/2603.00772 · **PDF:** https://arxiv.org/pdf/2603.00772v1
+
+## 8882. Identifying and Characterising Response in Clinical Trials: Development and Validation of a Machine Learning Approach in Colorectal Cancer
+
+- **ID:** `2603.00757` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Adam Marcus, Paul Agapow
+- **abs:** https://arxiv.org/abs/2603.00757 · **PDF:** https://arxiv.org/pdf/2603.00757v1
+
+## 8883. Stroke outcome and evolution prediction from CT brain using a spatiotemporal diffusion autoencoder
+
+- **ID:** `2603.00756` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Adam Marcus, Paul Bentley, Daniel Rueckert
+- **abs:** https://arxiv.org/abs/2603.00756 · **PDF:** https://arxiv.org/pdf/2603.00756v1
+
+## 8884. BornoViT: A Novel Efficient Vision Transformer for Bengali Handwritten Basic Characters Classification
+
+- **ID:** `2603.00755` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Rafi Hassan Chowdhury, Naimul Haque, Kaniz Fatiha
+- **abs:** https://arxiv.org/abs/2603.00755 · **PDF:** https://arxiv.org/pdf/2603.00755v1
+
+## 8885. General Proximal Flow Networks
+
+- **ID:** `2603.00751` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Alexander Strunk, Roland Assam
+- **abs:** https://arxiv.org/abs/2603.00751 · **PDF:** https://arxiv.org/pdf/2603.00751v1
+
+## 8886. SpectroFusion-ViT: A Lightweight Transformer for Speech Emotion Recognition Using Harmonic Mel-Chroma Fusion
+
+- **ID:** `2603.00746` · **primary:** `cs.SD` · **Submitted:** 2026-02-28
+- **Authors:** Faria Ahmed, Rafi Hassan Chowdhury, Fatema Tuz Zohora Moon, et al.
+- **abs:** https://arxiv.org/abs/2603.00746 · **PDF:** https://arxiv.org/pdf/2603.00746v1
+
+## 8887. Bi-cLSTM: Residual-Corrected Bidirectional LSTM for Aero-Engine RUL Estimation
+
+- **ID:** `2603.00745` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Rafi Hassan Chowdhury, Nabil Daiyan, Faria Ahmed, et al.
+- **abs:** https://arxiv.org/abs/2603.00745 · **PDF:** https://arxiv.org/pdf/2603.00745v1
+
+## 8888. ResGene-T: A Tensor-Based Residual Network Approach for Genomic Prediction
+
+- **ID:** `2603.00744` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Kuldeep Pathak, Kapil Ahuja, Eric de Sturler
+- **abs:** https://arxiv.org/abs/2603.00744 · **PDF:** https://arxiv.org/pdf/2603.00744v1
+
+## 8889. To Use or not to Use Muon: How Simplicity Bias in Optimizers Matters
+
+- **ID:** `2603.00742` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Sara Dragutinović, Rajesh Ranganath
+- **abs:** https://arxiv.org/abs/2603.00742 · **PDF:** https://arxiv.org/pdf/2603.00742v1
+
+## 8890. A Novel end-to-end Digital Health System Using Deep Learning-based ECG Analysis
+
+- **ID:** `2603.16891` · **primary:** `eess.SP` · **Submitted:** 2026-02-28
+- **Authors:** Artemis Kontou, Natalia Miroshnikova, Costakis Matheou, et al.
+- **abs:** https://arxiv.org/abs/2603.16891 · **PDF:** https://arxiv.org/pdf/2603.16891v1
+
+## 8891. Real-World AI Evaluation: How FRAME Generates Systematic Evidence to Resolve the Decision-Maker's Dilemma
+
+- **ID:** `2603.13294` · **primary:** `cs.CY` · **Submitted:** 2026-02-28
+- **Authors:** Reva Schwartz, Gabriella Waters
+- **abs:** https://arxiv.org/abs/2603.13294 · **PDF:** https://arxiv.org/pdf/2603.13294v4
+
+## 8892. Data-driven Synthesis of Magnetic Resonance Spectroscopy Data using a Variational Autoencoder
+
+- **ID:** `2603.00736` · **primary:** `physics.med-ph` · **Submitted:** 2026-02-28
+- **Authors:** Dennis M. J. van de Sande, Julian P. Merkofer, Sina Amirrajab, et al.
+- **abs:** https://arxiv.org/abs/2603.00736 · **PDF:** https://arxiv.org/pdf/2603.00736v1
+
+## 8893. MO-MIX: Multi-Objective Multi-Agent Cooperative Decision-Making With Deep Reinforcement Learning
+
+- **ID:** `2603.00730` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Tianmeng Hu, Biao Luo, Chunhua Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.00730 · **PDF:** https://arxiv.org/pdf/2603.00730v1
+
+## 8894. Qwen3-Coder-Next Technical Report
+
+- **ID:** `2603.00729` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Ruisheng Cao, Mouxiang Chen, Jiawei Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.00729 · **PDF:** https://arxiv.org/pdf/2603.00729v1
+
+## 8895. LaSTR: Language-Driven Time-Series Segment Retrieval
+
+- **ID:** `2603.00725` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Kota Dohi, Harsh Purohit, Tomoya Nishida, et al.
+- **abs:** https://arxiv.org/abs/2603.00725 · **PDF:** https://arxiv.org/pdf/2603.00725v1
+
+## 8896. RLAR: An Agentic Reward System for Multi-task Reinforcement Learning on Large Language Models
+
+- **ID:** `2603.00724` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Andrew Zhuoer Feng, Cunxiang Wang, Bosi Wen, et al.
+- **abs:** https://arxiv.org/abs/2603.00724 · **PDF:** https://arxiv.org/pdf/2603.00724v1
+
+## 8897. MARS: Harmonizing Multimodal Convergence via Adaptive Rank Search
+
+- **ID:** `2603.00720` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Minkyoung Cho, Insu Jang, Shuowei Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.00720 · **PDF:** https://arxiv.org/pdf/2603.00720v1
+
+## 8898. A Parameter-efficient Convolutional Approach for Weed Detection in Multispectral Aerial Imagery
+
+- **ID:** `2603.06655` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Leo Thomas Ramos, Angel D. Sappa
+- **abs:** https://arxiv.org/abs/2603.06655 · **PDF:** https://arxiv.org/pdf/2603.06655v1
+
+## 8899. SkillCraft: Can LLM Agents Learn to Use Tools Skillfully?
+
+- **ID:** `2603.00718` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Shiqi Chen, Jingze Gai, Ruochen Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.00718 · **PDF:** https://arxiv.org/pdf/2603.00718v2
+
+## 8900. Frozen Policy Iteration: Computationally Efficient RL under Linear $Q^π$ Realizability for Deterministic Dynamics
+
+- **ID:** `2603.00716` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Yijing Ke, Zihan Zhang, Ruosong Wang
+- **abs:** https://arxiv.org/abs/2603.00716 · **PDF:** https://arxiv.org/pdf/2603.00716v1
+
+## 8901. IU: Imperceptible Universal Backdoor Attack
+
+- **ID:** `2603.00711` · **primary:** `cs.CR` · **Submitted:** 2026-02-28
+- **Authors:** Hsin Lin, Yan-Lun Chen, Ren-Hung Hwang, et al.
+- **abs:** https://arxiv.org/abs/2603.00711 · **PDF:** https://arxiv.org/pdf/2603.00711v1
+
+## 8902. Reward-Modulated Local Learning in Spiking Encoders: Controlled Benchmarks with STDP and Hybrid Rate Readouts
+
+- **ID:** `2603.00710` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Debjyoti Chakraborty
+- **abs:** https://arxiv.org/abs/2603.00710 · **PDF:** https://arxiv.org/pdf/2603.00710v1
+
+## 8903. DRIV-EX: Counterfactual Explanations for Driving LLMs
+
+- **ID:** `2603.00696` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Amaia Cardiel, Eloi Zablocki, Elias Ramzi, et al.
+- **abs:** https://arxiv.org/abs/2603.00696 · **PDF:** https://arxiv.org/pdf/2603.00696v1
+
+## 8904. Wild-Drive: Off-Road Scene Captioning and Path Planning via Robust Multi-modal Routing and Efficient Large Language Model
+
+- **ID:** `2603.00694` · **primary:** `cs.RO` · **Submitted:** 2026-02-28
+- **Authors:** Zihang Wang, Xu Li, Benwu Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.00694 · **PDF:** https://arxiv.org/pdf/2603.00694v1
+
+## 8905. AIoT-based Continuous, Contextualized, and Explainable Driving Assessment for Older Adults
+
+- **ID:** `2603.00691` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Yimeng Liu, Fangwei Zhang, Maolin Gan, et al.
+- **abs:** https://arxiv.org/abs/2603.00691 · **PDF:** https://arxiv.org/pdf/2603.00691v1
+
+## 8906. RAVEL: Reasoning Agents for Validating and Evaluating LLM Text Synthesis
+
+- **ID:** `2603.00686` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Andrew Zhuoer Feng, Cunxiang Wang, Yu Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.00686 · **PDF:** https://arxiv.org/pdf/2603.00686v1
+
+## 8907. Polynomial Mixing for Efficient Self-supervised Speech Encoders
+
+- **ID:** `2603.00683` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Eva Feillet, Ryan Whetten, David Picard, et al.
+- **abs:** https://arxiv.org/abs/2603.00683 · **PDF:** https://arxiv.org/pdf/2603.00683v1
+
+## 8908. MemPO: Self-Memory Policy Optimization for Long-Horizon Agents
+
+- **ID:** `2603.00680` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Ruoran Li, Xinghua Zhang, Haiyang Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.00680 · **PDF:** https://arxiv.org/pdf/2603.00680v2
+
+## 8909. K^2-Agent: Co-Evolving Know-What and Know-How for Hierarchical Mobile Device Control
+
+- **ID:** `2603.00676` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Zhe Wu, Donglin Mo, Hongjin Lu, et al.
+- **abs:** https://arxiv.org/abs/2603.00676 · **PDF:** https://arxiv.org/pdf/2603.00676v1
+
+## 8910. SSKG Hub: An Expert-Guided Platform for LLM-Empowered Sustainability Standards Knowledge Graphs
+
+- **ID:** `2603.00669` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Chaoyue He, Xin Zhou, Xinjia Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.00669 · **PDF:** https://arxiv.org/pdf/2603.00669v1
+
+## 8911. Boosting Meta-Learning for Few-Shot Text Classification via Label-guided Distance Scaling
+
+- **ID:** `2603.02267` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Yunlong Gao, Xinyue Liu, Yingbo Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02267 · **PDF:** https://arxiv.org/pdf/2603.02267v1
+
+## 8912. InfoPO: Information-Driven Policy Optimization for User-Centric Agents
+
+- **ID:** `2603.00656` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Fanqi Kong, Jiayi Zhang, Mingyi Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.00656 · **PDF:** https://arxiv.org/pdf/2603.00656v1
+
+## 8913. Exploring 3D Dataset Pruning
+
+- **ID:** `2603.00651` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Xiaohan Zhao, Xinyi Shang, Jiacheng Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.00651 · **PDF:** https://arxiv.org/pdf/2603.00651v1
+
+## 8914. RAIE: Region-Aware Incremental Preference Editing with LoRA for LLM-based Recommendation
+
+- **ID:** `2603.00638` · **primary:** `cs.IR` · **Submitted:** 2026-02-28
+- **Authors:** Jin Zeng, Yupeng Qi, Hui Li, et al.
+- **abs:** https://arxiv.org/abs/2603.00638 · **PDF:** https://arxiv.org/pdf/2603.00638v2
+
+## 8915. Retrodictive Forecasting: A Proof-of-Concept for Exploiting Temporal Asymmetry in Time Series Prediction
+
+- **ID:** `2603.00636` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Cedric Damour
+- **abs:** https://arxiv.org/abs/2603.00636 · **PDF:** https://arxiv.org/pdf/2603.00636v1
+
+## 8916. BLUFF: Benchmarking the Detection of False and Synthetic Content across 58 Low-Resource Languages
+
+- **ID:** `2603.00634` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Jason Lucas, Matt Murtagh-White, Adaku Uchendu, et al.
+- **abs:** https://arxiv.org/abs/2603.00634 · **PDF:** https://arxiv.org/pdf/2603.00634v1
+
+## 8917. Stop Treating Collisions Equally: Qualification-Aware Semantic ID Learning for Recommendation at Industrial Scale
+
+- **ID:** `2603.00632` · **primary:** `cs.IR` · **Submitted:** 2026-02-28
+- **Authors:** Zheng Hu, Yuxin Chen, Yongsen Pan, et al.
+- **abs:** https://arxiv.org/abs/2603.00632 · **PDF:** https://arxiv.org/pdf/2603.00632v1
+
+## 8918. When Scaling Fails: Mitigating Audio Perception Decay of LALMs via Multi-Step Perception-Aware Reasoning
+
+- **ID:** `2603.02266` · **primary:** `cs.SD` · **Submitted:** 2026-02-28
+- **Authors:** Ruixiang Mao, Xiangnan Ma, Dan Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.02266 · **PDF:** https://arxiv.org/pdf/2603.02266v1
+
+## 8919. LiTS: A Modular Framework for LLM Tree Search
+
+- **ID:** `2603.00631` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Xinzhe Li, Yaguang Tao
+- **abs:** https://arxiv.org/abs/2603.00631 · **PDF:** https://arxiv.org/pdf/2603.00631v1
+
+## 8920. How AI Systems Think About Education: Analyzing Latent Preference Patterns in Large Language Models
+
+- **ID:** `2603.21006` · **primary:** `cs.CY` · **Submitted:** 2026-02-28
+- **Authors:** Daniel Autenrieth
+- **abs:** https://arxiv.org/abs/2603.21006 · **PDF:** https://arxiv.org/pdf/2603.21006v1
+
+## 8921. Adapt Data to Model: Adaptive Transformation Optimization for Domain-shared Time Series Foundation Models
+
+- **ID:** `2603.00629` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Yunzhong Qiu, Zhiyao Cen, Zhongyi Pei, et al.
+- **abs:** https://arxiv.org/abs/2603.00629 · **PDF:** https://arxiv.org/pdf/2603.00629v1
+
+## 8922. High-order Knowledge Based Network Controllability Robustness Prediction: A Hypergraph Neural Network Approach
+
+- **ID:** `2603.02265` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Shibing Mo, Jiarui Zhang, Jiayu Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.02265 · **PDF:** https://arxiv.org/pdf/2603.02265v1
+
+## 8923. IDER: IDempotent Experience Replay for Reliable Continual Learning
+
+- **ID:** `2603.00624` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Zhanwang Liu, Yuting Li, Haoyuan Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.00624 · **PDF:** https://arxiv.org/pdf/2603.00624v2
+
+## 8924. TraceSIR: A Multi-Agent Framework for Structured Analysis and Reporting of Agentic Execution Traces
+
+- **ID:** `2603.00623` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Shu-Xun Yang, Cunxiang Wang, Haoke Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.00623 · **PDF:** https://arxiv.org/pdf/2603.00623v1
+
+## 8925. Piecing Together Cross-Document Coreference Resolution Datasets: Systematic Dataset Analysis and Unification
+
+- **ID:** `2603.00621` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Anastasia Zhukova, Terry Ruas, Jan Philip Wahle, et al.
+- **abs:** https://arxiv.org/abs/2603.00621 · **PDF:** https://arxiv.org/pdf/2603.00621v2
+
+## 8926. QQ: A Toolkit for Language Identifiers and Metadata
+
+- **ID:** `2603.00620` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Wessel Poelman, Yiyi Chen, Miryam de Lhoneux
+- **abs:** https://arxiv.org/abs/2603.00620 · **PDF:** https://arxiv.org/pdf/2603.00620v1
+
+## 8927. Multi-Domain Riemannian Graph Gluing for Building Graph Foundation Models
+
+- **ID:** `2603.00618` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Li Sun, Zhenhao Huang, Silei Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.00618 · **PDF:** https://arxiv.org/pdf/2603.00618v1
+
+## 8928. From Literature to Hypotheses: An AI Co-Scientist System for Biomarker-Guided Drug Combination Hypothesis Generation
+
+- **ID:** `2603.00612` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Raneen Younis, Suvinava Basak, Lukas Chavez, et al.
+- **abs:** https://arxiv.org/abs/2603.00612 · **PDF:** https://arxiv.org/pdf/2603.00612v1
+
+## 8929. CMI-RewardBench: Evaluating Music Reward Models with Compositional Multimodal Instruction
+
+- **ID:** `2603.00610` · **primary:** `cs.SD` · **Submitted:** 2026-02-28
+- **Authors:** Yinghao Ma, Haiwen Xia, Hewei Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.00610 · **PDF:** https://arxiv.org/pdf/2603.00610v2
+
+## 8930. Machine Learning Grade Prediction Using Students' Grades and Demographics
+
+- **ID:** `2603.00608` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Mwayi Sonkhanani, Symon Chibaya, Clement N. Nyirenda
+- **abs:** https://arxiv.org/abs/2603.00608 · **PDF:** https://arxiv.org/pdf/2603.00608v1
+
+## 8931. IdGlow: Dynamic Identity Modulation for Multi-Subject Generation
+
+- **ID:** `2603.00607` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Honghao Cai, Xiangyuan Wang, Yunhao Bai, et al.
+- **abs:** https://arxiv.org/abs/2603.00607 · **PDF:** https://arxiv.org/pdf/2603.00607v1
+
+## 8932. Learning to Explore: Policy-Guided Outlier Synthesis for Graph Out-of-Distribution Detection
+
+- **ID:** `2603.00602` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Li Sun, Lanxu Yang, Jiayu Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.00602 · **PDF:** https://arxiv.org/pdf/2603.00602v1
+
+## 8933. Theory of Code Space: Do Code Agents Understand Software Architecture?
+
+- **ID:** `2603.00601` · **primary:** `cs.SE` · **Submitted:** 2026-02-28
+- **Authors:** Grigory Sapunov
+- **abs:** https://arxiv.org/abs/2603.00601 · **PDF:** https://arxiv.org/pdf/2603.00601v4
+
+## 8934. Heterophily-Agnostic Hypergraph Neural Networks with Riemannian Local Exchanger
+
+- **ID:** `2603.00599` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Li Sun, Ming Zhang, Wenxin Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.00599 · **PDF:** https://arxiv.org/pdf/2603.00599v1
+
+## 8935. How the Graph Construction Technique Shapes Performance in IoT Botnet Detection
+
+- **ID:** `2603.06654` · **primary:** `cs.NI` · **Submitted:** 2026-02-28
+- **Authors:** Hassan Wasswa, Hussein Abbass, Timothy Lynar
+- **abs:** https://arxiv.org/abs/2603.06654 · **PDF:** https://arxiv.org/pdf/2603.06654v1
+
+## 8936. Rubric-Guided Fine-tuning of SpeechLLMs for Multi-Aspect, Multi-Rater L2 Reading-Speech Assessment
+
+- **ID:** `2603.16889` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Aditya Kamlesh Parikh, Cristian Tejedor-Garcia, Catia Cucchiarini, et al.
+- **abs:** https://arxiv.org/abs/2603.16889 · **PDF:** https://arxiv.org/pdf/2603.16889v1
+
+## 8937. LangGap: Diagnosing and Closing the Language Gap in Vision-Language-Action Models
+
+- **ID:** `2603.00592` · **primary:** `cs.RO` · **Submitted:** 2026-02-28
+- **Authors:** Yuchen Hou, Lin Zhao
+- **abs:** https://arxiv.org/abs/2603.00592 · **PDF:** https://arxiv.org/pdf/2603.00592v1
+
+## 8938. Leveraging Computerized Adaptive Testing for Cost-effective Evaluation of Large Language Models in Medical Benchmarking
+
+- **ID:** `2603.23506` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Tianpeng Zheng, Zhehan Jiang, Jiayi Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.23506 · **PDF:** https://arxiv.org/pdf/2603.23506v1
+
+## 8939. Fair in Mind, Fair in Action? A Synchronous Benchmark for Understanding and Generation in UMLLMs
+
+- **ID:** `2603.00590` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Yiran Zhao, Lu Zhou, Xiaogang Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.00590 · **PDF:** https://arxiv.org/pdf/2603.00590v1
+
+## 8940. AlignVAR: Towards Globally Consistent Visual Autoregression for Image Super-Resolution
+
+- **ID:** `2603.00589` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Cencen Liu, Dongyang Zhang, Wen Yin, et al.
+- **abs:** https://arxiv.org/abs/2603.00589 · **PDF:** https://arxiv.org/pdf/2603.00589v2
+
+## 8941. Energy-Efficient Information Representation in MNIST Classification Using Biologically Inspired Learning
+
+- **ID:** `2603.00588` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Patrick Stricker, Florian Röhrbein, Andreas Knoblauch
+- **abs:** https://arxiv.org/abs/2603.00588 · **PDF:** https://arxiv.org/pdf/2603.00588v1
+
+## 8942. Unlearning Evaluation through Subset Statistical Independence
+
+- **ID:** `2603.00587` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Chenhao Zhang, Muxing Li, Feng Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.00587 · **PDF:** https://arxiv.org/pdf/2603.00587v1
+
+## 8943. MicroVerse: A Preliminary Exploration Toward a Micro-World Simulation
+
+- **ID:** `2603.00585` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Rongsheng Wang, Minghao Wu, Hongru Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.00585 · **PDF:** https://arxiv.org/pdf/2603.00585v1
+
+## 8944. Super Research: Answering Highly Complex Questions with Large Language Models through Super Deep and Super Wide Research
+
+- **ID:** `2603.00582` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Yubo Dong, Nianhao You, Yuxuan Hou, et al.
+- **abs:** https://arxiv.org/abs/2603.00582 · **PDF:** https://arxiv.org/pdf/2603.00582v2
+
+## 8945. DeepAFL: Deep Analytic Federated Learning
+
+- **ID:** `2603.00579` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Jianheng Tang, Yajiang Huang, Kejia Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.00579 · **PDF:** https://arxiv.org/pdf/2603.00579v1
+
+## 8946. Draft-Thinking: Learning Efficient Reasoning in Long Chain-of-Thought LLMs
+
+- **ID:** `2603.00578` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Jie Cao, Tianwei Lin, Zhenxuan Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.00578 · **PDF:** https://arxiv.org/pdf/2603.00578v1
+
+## 8947. Efficient Long-Sequence Diffusion Modeling for Symbolic Music Generation
+
+- **ID:** `2603.00576` · **primary:** `cs.SD` · **Submitted:** 2026-02-28
+- **Authors:** Jinhan Xu, Xing Tang, Houpeng Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.00576 · **PDF:** https://arxiv.org/pdf/2603.00576v1
+
+## 8948. SWE-Hub: A Unified Production System for Scalable, Executable Software Engineering Tasks
+
+- **ID:** `2603.00575` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Yucheng Zeng, Shupeng Li, Daxiang Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.00575 · **PDF:** https://arxiv.org/pdf/2603.00575v1
+
+## 8949. Decoupling Stability and Plasticity for Multi-Modal Test-Time Adaptation
+
+- **ID:** `2603.00574` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Yongbo He, Zirun Guo, Tao Jin
+- **abs:** https://arxiv.org/abs/2603.00574 · **PDF:** https://arxiv.org/pdf/2603.00574v1
+
+## 8950. CoMoL: Efficient Mixture of LoRA Experts via Dynamic Core Space Merging
+
+- **ID:** `2603.00573` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Jie Cao, Zhenxuan Fan, Zhuonan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.00573 · **PDF:** https://arxiv.org/pdf/2603.00573v1
+
+## 8951. TopoEdge: Topology-Grounded Agentic Framework for Edge Networking Code Generation and Repair
+
+- **ID:** `2603.00569` · **primary:** `cs.SE` · **Submitted:** 2026-02-28
+- **Authors:** Haomin Qi, Bohan Liu, Zihan Dai, et al.
+- **abs:** https://arxiv.org/abs/2603.00569 · **PDF:** https://arxiv.org/pdf/2603.00569v1
+
+## 8952. A Robust Framework for Secure Cardiovascular Risk Prediction: An Architectural Case Study of Differentially Private Federated Learning
+
+- **ID:** `2603.13293` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Rodrigo Tertulino, Laércio Alencar
+- **abs:** https://arxiv.org/abs/2603.13293 · **PDF:** https://arxiv.org/pdf/2603.13293v1
+
+## 8953. Enhancing Molecular Property Predictions by Learning from Bond Modelling and Interactions
+
+- **ID:** `2603.00568` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Yunqing Liu, Yi Zhou, Wenqi Fan
+- **abs:** https://arxiv.org/abs/2603.00568 · **PDF:** https://arxiv.org/pdf/2603.00568v1
+
+## 8954. Learning to Attack: A Bandit Approach to Adversarial Context Poisoning
+
+- **ID:** `2603.00567` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Ray Telikani, Amir H. Gandomi
+- **abs:** https://arxiv.org/abs/2603.00567 · **PDF:** https://arxiv.org/pdf/2603.00567v1
+
+## 8955. MIDAS: Multi-Image Dispersion and Semantic Reconstruction for Jailbreaking MLLMs
+
+- **ID:** `2603.00565` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Yilian Liu, Xiaojun Jia, Guoshun Nan, et al.
+- **abs:** https://arxiv.org/abs/2603.00565 · **PDF:** https://arxiv.org/pdf/2603.00565v1
+
+## 8956. Whisper-MLA: Reducing GPU Memory Consumption of ASR Models based on MHA2MLA Conversion
+
+- **ID:** `2603.00563` · **primary:** `cs.SD` · **Submitted:** 2026-02-28
+- **Authors:** Sen Zhang, Jianguo Wei, Wenhuan Lu, et al.
+- **abs:** https://arxiv.org/abs/2603.00563 · **PDF:** https://arxiv.org/pdf/2603.00563v1
+
+## 8957. Geometry OR Tracker: Universal Geometric Operating Room Tracking
+
+- **ID:** `2603.00560` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Yihua Shao, Kang Chen, Feng Xue, et al.
+- **abs:** https://arxiv.org/abs/2603.00560 · **PDF:** https://arxiv.org/pdf/2603.00560v1
+
+## 8958. EMPA: Evaluating Persona-Aligned Empathy as a Process
+
+- **ID:** `2603.00552` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Shiya Zhang, Yuhan Zhan, Ruixi Su, et al.
+- **abs:** https://arxiv.org/abs/2603.00552 · **PDF:** https://arxiv.org/pdf/2603.00552v2
+
+## 8959. GCL-Sampler: Discovering Kernel Similarity for Sampled GPU Simulation via Graph Contrastive Learning
+
+- **ID:** `2603.00551` · **primary:** `cs.PF` · **Submitted:** 2026-02-28
+- **Authors:** Jiaqi Wang, Jingwei Sun, Jiyu Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.00551 · **PDF:** https://arxiv.org/pdf/2603.00551v1
+
+## 8960. Advancing Multimodal Judge Models through a Capability-Oriented Benchmark and MCTS-Driven Data Generation
+
+- **ID:** `2603.00546` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Zeyu Chen, Huanjin Yao, Ziwang Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.00546 · **PDF:** https://arxiv.org/pdf/2603.00546v1
+
+## 8961. Spectral Condition for $μ$P under Width-Depth Scaling
+
+- **ID:** `2603.00541` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Chenyu Zheng, Rongzhen Wang, Xinyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.00541 · **PDF:** https://arxiv.org/pdf/2603.00541v1
+
+## 8962. LOGIGEN: Logic-Driven Generation of Verifiable Agentic Tasks
+
+- **ID:** `2603.00540` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Yucheng Zeng, Weipeng Lu, Linyun Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.00540 · **PDF:** https://arxiv.org/pdf/2603.00540v1
+
+## 8963. Are LLMs Reliable Code Reviewers? Systematic Overcorrection in Requirement Conformance Judgement
+
+- **ID:** `2603.00539` · **primary:** `cs.SE` · **Submitted:** 2026-02-28
+- **Authors:** Haolin Jin, Huaming Chen
+- **abs:** https://arxiv.org/abs/2603.00539 · **PDF:** https://arxiv.org/pdf/2603.00539v1
+
+## 8964. Pragma-VL: Towards a Pragmatic Arbitration of Safety and Helpfulness in MLLMs
+
+- **ID:** `2603.13292` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Ming Wen, Kun Yang, Xin Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.13292 · **PDF:** https://arxiv.org/pdf/2603.13292v1
+
+## 8965. Mathematical Foundations of Poisoning Attacks on Linear Regression over Cumulative Distribution Functions
+
+- **ID:** `2603.00537` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Atsuki Sato, Martin Aumüller, Yusuke Matsui
+- **abs:** https://arxiv.org/abs/2603.00537 · **PDF:** https://arxiv.org/pdf/2603.00537v1
+
+## 8966. DenoiseFlow: Uncertainty-Aware Denoising for Reliable LLM Agentic Workflows
+
+- **ID:** `2603.00532` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Yandong Yan, Junwei Peng, Shijie Li, et al.
+- **abs:** https://arxiv.org/abs/2603.00532 · **PDF:** https://arxiv.org/pdf/2603.00532v1
+
+## 8967. Bridge Matching Sampler: Scalable Sampling via Generalized Fixed-Point Diffusion Matching
+
+- **ID:** `2603.00530` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Denis Blessing, Lorenz Richter, Julius Berner, et al.
+- **abs:** https://arxiv.org/abs/2603.00530 · **PDF:** https://arxiv.org/pdf/2603.00530v1
+
+## 8968. FedUAF: Uncertainty-Aware Fusion with Reliability-Guided Aggregation for Multimodal Federated Sentiment Analysis
+
+- **ID:** `2603.13291` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Xianxun Zhu, Zezhong Sun, Imad Rida, et al.
+- **abs:** https://arxiv.org/abs/2603.13291 · **PDF:** https://arxiv.org/pdf/2603.13291v1
+
+## 8969. CaptionFool: Universal Image Captioning Model Attacks
+
+- **ID:** `2603.00529` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Swapnil Parekh
+- **abs:** https://arxiv.org/abs/2603.00529 · **PDF:** https://arxiv.org/pdf/2603.00529v1
+
+## 8970. Social-JEPA: Emergent Geometric Isomorphism
+
+- **ID:** `2603.02263` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Haoran Zhang, Youjin Wang, Yi Duan, et al.
+- **abs:** https://arxiv.org/abs/2603.02263 · **PDF:** https://arxiv.org/pdf/2603.02263v1
+
+## 8971. CIRCUS: Circuit Consensus under Uncertainty via Stability Ensembles
+
+- **ID:** `2603.00523` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Swapnil Parekh
+- **abs:** https://arxiv.org/abs/2603.00523 · **PDF:** https://arxiv.org/pdf/2603.00523v2
+
+## 8972. Phys-Diff: A Physics-Inspired Latent Diffusion Model for Tropical Cyclone Forecasting
+
+- **ID:** `2603.00521` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Lei Liu, Xiaoning Yu, Kang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.00521 · **PDF:** https://arxiv.org/pdf/2603.00521v1
+
+## 8973. ACES: Accent Subspaces for Coupling, Explanations, and Stress-Testing in Automatic Speech Recognition
+
+- **ID:** `2603.03359` · **primary:** `cs.SD` · **Submitted:** 2026-02-28
+- **Authors:** Swapnil Parekh
+- **abs:** https://arxiv.org/abs/2603.03359 · **PDF:** https://arxiv.org/pdf/2603.03359v2
+
+## 8974. FastBUS: A Fast Bayesian Framework for Unified Weakly-Supervised Learning
+
+- **ID:** `2603.00517` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Ziquan Wang, Haobo Wang, Ke Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.00517 · **PDF:** https://arxiv.org/pdf/2603.00517v1
+
+## 8975. Silent Sabotage During Fine-Tuning: Few-Shot Rationale Poisoning of Compact Medical LLMs
+
+- **ID:** `2603.02262` · **primary:** `cs.CR` · **Submitted:** 2026-02-28
+- **Authors:** Jingyuan Xie, Wenjie Wang, Ji Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.02262 · **PDF:** https://arxiv.org/pdf/2603.02262v1
+
+## 8976. Multimodal Adaptive Retrieval Augmented Generation through Internal Representation Learning
+
+- **ID:** `2603.00511` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Ruoshuang Du, Xin Sun, Qiang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.00511 · **PDF:** https://arxiv.org/pdf/2603.00511v1
+
+## 8977. What Do Visual Tokens Really Encode? Uncovering Sparsity and Redundancy in Multimodal Large Language Models
+
+- **ID:** `2603.00510` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Yingqi Fan, Junlong Tong, Anhao Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.00510 · **PDF:** https://arxiv.org/pdf/2603.00510v1
+
+## 8978. Quantum AS-DeepOnet: Quantum Attentive Stacked DeepONet for Solving 2D Evolution Equations
+
+- **ID:** `2603.02261` · **primary:** `quant-ph` · **Submitted:** 2026-02-28
+- **Authors:** Hongquan Wang, Hanshu Chen, Ilia Marchevsky, et al.
+- **abs:** https://arxiv.org/abs/2603.02261 · **PDF:** https://arxiv.org/pdf/2603.02261v1
+
+## 8979. Trinity: A Scenario-Aware Recommendation Framework for Large-Scale Cold-Start Users
+
+- **ID:** `2603.00502` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Wenhao Zheng, Wang Lu, Fangshuang Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.00502 · **PDF:** https://arxiv.org/pdf/2603.00502v1
+
+## 8980. WirelessAgent++: Automated Agentic Workflow Design and Benchmarking for Wireless Networks
+
+- **ID:** `2603.00501` · **primary:** `cs.NI` · **Submitted:** 2026-02-28
+- **Authors:** Jingwen Tong, Zijian Li, Fang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.00501 · **PDF:** https://arxiv.org/pdf/2603.00501v1
+
+## 8981. Antibody: Strengthening Defense Against Harmful Fine-Tuning for Large Language Models via Attenuating Harmful Gradient Influence
+
+- **ID:** `2603.00498` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Quoc Minh Nguyen, Trung Le, Jing Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.00498 · **PDF:** https://arxiv.org/pdf/2603.00498v1
+
+## 8982. A Polynomial-Time Axiomatic Alternative to SHAP for Feature Attribution
+
+- **ID:** `2603.00496` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Kazuhiro Hiraki, Shinichi Ishihara, Takumi Kongo, et al.
+- **abs:** https://arxiv.org/abs/2603.00496 · **PDF:** https://arxiv.org/pdf/2603.00496v1
+
+## 8983. AI Runtime Infrastructure
+
+- **ID:** `2603.00495` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Christopher Cruz
+- **abs:** https://arxiv.org/abs/2603.00495 · **PDF:** https://arxiv.org/pdf/2603.00495v1
+
+## 8984. GeNeX: Genetic Network eXperts framework for addressing Validation Overfitting
+
+- **ID:** `2603.11056` · **primary:** `cs.NE` · **Submitted:** 2026-02-28
+- **Authors:** Emmanuel Pintelas, Ioannis E. Livieris
+- **abs:** https://arxiv.org/abs/2603.11056 · **PDF:** https://arxiv.org/pdf/2603.11056v1
+
+## 8985. ArtiFixer: Enhancing and Extending 3D Reconstruction with Auto-Regressive Diffusion Models
+
+- **ID:** `2603.00492` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Riccardo de Lutio, Tobias Fischer, Yen-Yu Chang, et al.
+- **abs:** https://arxiv.org/abs/2603.00492 · **PDF:** https://arxiv.org/pdf/2603.00492v1
+
+## 8986. Heaviside Low-Rank Support Matrix Machine
+
+- **ID:** `2603.00491` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Xianchao Xiu, Shenghao Sun, Xinrong Li, et al.
+- **abs:** https://arxiv.org/abs/2603.00491 · **PDF:** https://arxiv.org/pdf/2603.00491v1
+
+## 8987. LifeEval: A Multimodal Benchmark for Assistive AI in Egocentric Daily Life Tasks
+
+- **ID:** `2603.00490` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Hengjian Gao, Kaiwei Zhang, Shibo Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.00490 · **PDF:** https://arxiv.org/pdf/2603.00490v1
+
+## 8988. Dynamic Spatio-Temporal Graph Neural Network for Early Detection of Pornography Addiction in Adolescents Based on Electroencephalogram Signals
+
+- **ID:** `2603.00488` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Achmad Ardani Prasha, Clavino Ourizqi Rachmadi, Sabrina Laila Mutiara, et al.
+- **abs:** https://arxiv.org/abs/2603.00488 · **PDF:** https://arxiv.org/pdf/2603.00488v1
+
+## 8989. RAISE: Requirement-Adaptive Evolutionary Refinement for Training-Free Text-to-Image Alignment
+
+- **ID:** `2603.00483` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Liyao Jiang, Ruichen Chen, Chao Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.00483 · **PDF:** https://arxiv.org/pdf/2603.00483v1
+
+## 8990. Analyzing Physical Adversarial Example Threats to Machine Learning in Election Systems
+
+- **ID:** `2603.00481` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Khaleque Md Aashiq Kamal, Surya Eada, Aayushi Verma, et al.
+- **abs:** https://arxiv.org/abs/2603.00481 · **PDF:** https://arxiv.org/pdf/2603.00481v1
+
+## 8991. Benchmarking Few-shot Transferability of Pre-trained Models with Improved Evaluation Protocols
+
+- **ID:** `2603.00478` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Xu Luo, Ji Zhang, Lianli Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.00478 · **PDF:** https://arxiv.org/pdf/2603.00478v1
+
+## 8992. Atomicity for Agents: Exposing, Exploiting, and Mitigating TOCTOU Vulnerabilities in Browser-Use Agents
+
+- **ID:** `2603.00476` · **primary:** `cs.CR` · **Submitted:** 2026-02-28
+- **Authors:** Linxi Jiang, Zhijie Liu, Haotian Luo, et al.
+- **abs:** https://arxiv.org/abs/2603.00476 · **PDF:** https://arxiv.org/pdf/2603.00476v1
+
+## 8993. Wireless Power Control Based on Large Language Models
+
+- **ID:** `2603.00474` · **primary:** `cs.IT` · **Submitted:** 2026-02-28
+- **Authors:** Jiacheng Wang, Yucheng Sheng, Le Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.00474 · **PDF:** https://arxiv.org/pdf/2603.00474v2
+
+## 8994. From Goals to Aspects, Revisited: An NFR Pattern Language for Agentic AI Systems
+
+- **ID:** `2603.00472` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Yijun Yu
+- **abs:** https://arxiv.org/abs/2603.00472 · **PDF:** https://arxiv.org/pdf/2603.00472v1
+
+## 8995. Why Not? Solver-Grounded Certificates for Explainable Mission Planning
+
+- **ID:** `2603.00469` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Najeeb Khan
+- **abs:** https://arxiv.org/abs/2603.00469 · **PDF:** https://arxiv.org/pdf/2603.00469v1
+
+## 8996. Multi-Agent Reinforcement Learning for Dynamic Pricing: Balancing Profitability,Stability and Fairness
+
+- **ID:** `2603.16888` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Krishna Kumar Neelakanta Pillai Santha Kumari Amma
+- **abs:** https://arxiv.org/abs/2603.16888 · **PDF:** https://arxiv.org/pdf/2603.16888v1
+
+## 8997. Optimizing In-Context Demonstrations for LLM-based Automated Grading
+
+- **ID:** `2603.00465` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Yucheng Chu, Hang Li, Kaiqi Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.00465 · **PDF:** https://arxiv.org/pdf/2603.00465v1
+
+## 8998. TAS-GNN: A Status-Aware Signed Graph Neural Network for Anomaly Detection in Bitcoin Trust Systems
+
+- **ID:** `2603.13290` · **primary:** `cs.CR` · **Submitted:** 2026-02-28
+- **Authors:** Chang Xue, Fang Liu, Jiaye Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.13290 · **PDF:** https://arxiv.org/pdf/2603.13290v1
+
+## 8999. OPGAgent: An Agent for Auditable Dental Panoramic X-ray Interpretation
+
+- **ID:** `2603.00462` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Zhaolin Yu, Litao Yang, Ben Babicka, et al.
+- **abs:** https://arxiv.org/abs/2603.00462 · **PDF:** https://arxiv.org/pdf/2603.00462v1
+
+## 9000. RelayCaching: Accelerating LLM Collaboration via Decoding KV Cache Reuse
+
+- **ID:** `2603.13289` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Yingsheng Geng, Yuchong Gao, Weihong Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.13289 · **PDF:** https://arxiv.org/pdf/2603.13289v1
+
+## 9001. From Flat to Structural: Enhancing Automated Short Answer Grading with GraphRAG
+
+- **ID:** `2603.19276` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Yucheng Chu, Haoyu Han, Shen Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.19276 · **PDF:** https://arxiv.org/pdf/2603.19276v1
+
+## 9002. PaLMR: Towards Faithful Visual Reasoning via Multimodal Process Alignment
+
+- **ID:** `2603.06652` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Yantao Li, Qiang Hui, Chenyang Yan, et al.
+- **abs:** https://arxiv.org/abs/2603.06652 · **PDF:** https://arxiv.org/pdf/2603.06652v1
+
+## 9003. MED-COPILOT: A Medical Assistant Powered by GraphRAG and Similar Patient Case Retrieval
+
+- **ID:** `2603.00460` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Shuheng Chen, Namratha Patil, Haonan Pan, et al.
+- **abs:** https://arxiv.org/abs/2603.00460 · **PDF:** https://arxiv.org/pdf/2603.00460v1
+
+## 9004. Rooted Absorbed Prefix Trajectory Balance with Submodular Replay for GFlowNet Training
+
+- **ID:** `2603.00454` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Xi Wang, Wenbo Lu, Shengjie Wang
+- **abs:** https://arxiv.org/abs/2603.00454 · **PDF:** https://arxiv.org/pdf/2603.00454v1
+
+## 9005. Neurosymbolic Learning for Advanced Persistent Threat Detection under Extreme Class Imbalance
+
+- **ID:** `2603.00453` · **primary:** `cs.CR` · **Submitted:** 2026-02-28
+- **Authors:** Quhura Fathima, Neda Moghim, Mostafa Taghizade Firouzjaee, et al.
+- **abs:** https://arxiv.org/abs/2603.00453 · **PDF:** https://arxiv.org/pdf/2603.00453v1
+
+## 9006. Texterial: A Text-as-Material Interaction Paradigm for LLM-Mediated Writing
+
+- **ID:** `2603.00452` · **primary:** `cs.HC` · **Submitted:** 2026-02-28
+- **Authors:** Jocelyn Shen, Nicolai Marquardt, Hugo Romat, et al.
+- **abs:** https://arxiv.org/abs/2603.00452 · **PDF:** https://arxiv.org/pdf/2603.00452v1
+
+## 9007. Confusion-Aware Rubric Optimization for LLM-based Automated Grading
+
+- **ID:** `2603.00451` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Yucheng Chu, Hang Li, Kaiqi Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.00451 · **PDF:** https://arxiv.org/pdf/2603.00451v1
+
+## 9008. Evaluating Progress in Graph Foundation Models: A Comprehensive Benchmark and New Insights
+
+- **ID:** `2603.10033` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Xingtong Yu, Shenghua Ye, Ruijuan Liang, et al.
+- **abs:** https://arxiv.org/abs/2603.10033 · **PDF:** https://arxiv.org/pdf/2603.10033v1
+
+## 9009. HydroShear: Hydroelastic Shear Simulation for Tactile Sim-to-Real Reinforcement Learning
+
+- **ID:** `2603.00446` · **primary:** `cs.RO` · **Submitted:** 2026-02-28
+- **Authors:** An Dang, Jayjun Lee, Mustafa Mukadam, et al.
+- **abs:** https://arxiv.org/abs/2603.00446 · **PDF:** https://arxiv.org/pdf/2603.00446v1
+
+## 9010. Mamba-CAD: State Space Model For 3D Computer-Aided Design Generative Modeling
+
+- **ID:** `2603.00439` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Xueyang Li, Yunzhong Lou, Yu Song, et al.
+- **abs:** https://arxiv.org/abs/2603.00439 · **PDF:** https://arxiv.org/pdf/2603.00439v1
+
+## 9011. Improving Automatic Summarization of Radiology Reports through Mid-Training of Large Language Models
+
+- **ID:** `2603.19275` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Mengxian Lyu, Cheng Peng, Ziyi Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.19275 · **PDF:** https://arxiv.org/pdf/2603.19275v1
+
+## 9012. ROKA: Robust Knowledge Unlearning against Adversaries
+
+- **ID:** `2603.00436` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Jinmyeong Shin, Joshua Tapia, Nicholas Ferreira, et al.
+- **abs:** https://arxiv.org/abs/2603.00436 · **PDF:** https://arxiv.org/pdf/2603.00436v1
+
+## 9013. RTLocating: Intent-aware RTL Localization for Hardware Design Iteration
+
+- **ID:** `2603.00434` · **primary:** `cs.ET` · **Submitted:** 2026-02-28
+- **Authors:** Changwen Xing, Yanfeng Lu, Lei Qi, et al.
+- **abs:** https://arxiv.org/abs/2603.00434 · **PDF:** https://arxiv.org/pdf/2603.00434v1
+
+## 9014. TAP-SLF: Parameter-Efficient Adaptation of Vision Foundation Models for Multi-Task Ultrasound Image Analysis
+
+- **ID:** `2603.00433` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Hui Wan, Libin Lan
+- **abs:** https://arxiv.org/abs/2603.00433 · **PDF:** https://arxiv.org/pdf/2603.00433v1
+
+## 9015. A Typologically Grounded Evaluation Framework for Word Order and Morphology Sensitivity in Multilingual Masked LMs
+
+- **ID:** `2603.00432` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Anna Feldman, Libby Barak, Jing Peng
+- **abs:** https://arxiv.org/abs/2603.00432 · **PDF:** https://arxiv.org/pdf/2603.00432v1
+
+## 9016. Taxonomy-Aware Representation Alignment for Hierarchical Visual Recognition with Large Multimodal Models
+
+- **ID:** `2603.00431` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Hulingxiao He, Zhi Tan, Yuxin Peng
+- **abs:** https://arxiv.org/abs/2603.00431 · **PDF:** https://arxiv.org/pdf/2603.00431v2
+
+## 9017. Efficient Decoder Scaling Strategy for Neural Routing Solvers
+
+- **ID:** `2603.00430` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Qing Luo, Fu Luo, Ke Li, et al.
+- **abs:** https://arxiv.org/abs/2603.00430 · **PDF:** https://arxiv.org/pdf/2603.00430v1
+
+## 9018. Personalities at Play: Probing Alignment in AI Teammates
+
+- **ID:** `2603.00429` · **primary:** `cs.HC` · **Submitted:** 2026-02-28
+- **Authors:** Mohammad Amin Samadi, Nia Nixon
+- **abs:** https://arxiv.org/abs/2603.00429 · **PDF:** https://arxiv.org/pdf/2603.00429v1
+
+## 9019. CURE: A Multimodal Benchmark for Clinical Understanding and Retrieval Evaluation
+
+- **ID:** `2603.19274` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Yannian Gu, Zhongzhen Huang, Linjie Mu, et al.
+- **abs:** https://arxiv.org/abs/2603.19274 · **PDF:** https://arxiv.org/pdf/2603.19274v1
+
+## 9020. LLM-Bootstrapped Targeted Finding Guidance for Factual MLLM-based Medical Report Generation
+
+- **ID:** `2603.00426` · **primary:** `cs.CL` · **Submitted:** 2026-02-28
+- **Authors:** Cunyuan Yang, Dejuan Song, Xiaotao Pang, et al.
+- **abs:** https://arxiv.org/abs/2603.00426 · **PDF:** https://arxiv.org/pdf/2603.00426v1
+
+## 9021. Weight Updates as Activation Shifts: A Principled Framework for Steering
+
+- **ID:** `2603.00425` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Dyah Adila, John Cooper, Alexander Yun, et al.
+- **abs:** https://arxiv.org/abs/2603.00425 · **PDF:** https://arxiv.org/pdf/2603.00425v2
+
+## 9022. An Interpretable Local Editing Model for Counterfactual Medical Image Generation
+
+- **ID:** `2603.00423` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Hyungi Min, Taeseung You, Hangyeul Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.00423 · **PDF:** https://arxiv.org/pdf/2603.00423v1
+
+## 9023. TMR-VLA:Vision-Language-Action Model for Magnetic Motion Control of Tri-leg Silicone-based Soft Robot
+
+- **ID:** `2603.00420` · **primary:** `cs.RO` · **Submitted:** 2026-02-28
+- **Authors:** Ruijie Tang, Chi Kit Ng, Kaixuan Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.00420 · **PDF:** https://arxiv.org/pdf/2603.00420v1
+
+## 9024. Physics-Aware Learnability: From Set-Theoretic Independence to Operational Constraints
+
+- **ID:** `2603.00417` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Jeongho Bang, Kyoungho Cho
+- **abs:** https://arxiv.org/abs/2603.00417 · **PDF:** https://arxiv.org/pdf/2603.00417v1
+
+## 9025. MuonRec: Shifting the Optimizer Paradigm Beyond Adam in Scalable Generative Recommendation
+
+- **ID:** `2603.00416` · **primary:** `cs.IR` · **Submitted:** 2026-02-28
+- **Authors:** Rong Shan, Aofan Yu, Bo Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.00416 · **PDF:** https://arxiv.org/pdf/2603.00416v1
+
+## 9026. Exact and Asymptotically Complete Robust Verifications of Neural Networks via Quantum Optimization
+
+- **ID:** `2603.00408` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Wenxin Li, Wenchao Liu, Chuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.00408 · **PDF:** https://arxiv.org/pdf/2603.00408v1
+
+## 9027. Geodesic Gradient Descent: A Generic and Learning-rate-free Optimizer on Objective Function-induced Manifolds
+
+- **ID:** `2603.06651` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Liwei Hu, Guangyao Li, Wenyong Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.06651 · **PDF:** https://arxiv.org/pdf/2603.06651v1
+
+## 9028. Agent-Based User-Adaptive Filtering for Categorized Harassing Communication
+
+- **ID:** `2603.13288` · **primary:** `cs.AI` · **Submitted:** 2026-02-28
+- **Authors:** Zenefa Rahaman, Sandip Sen
+- **abs:** https://arxiv.org/abs/2603.13288 · **PDF:** https://arxiv.org/pdf/2603.13288v1
+
+## 9029. USE: Uncertainty Structure Estimation for Robust Semi-Supervised Learning
+
+- **ID:** `2603.00404` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Tsao-Lun Chen, Chien-Liang Liu, Tzu-Ming Harry Hsu, et al.
+- **abs:** https://arxiv.org/abs/2603.00404 · **PDF:** https://arxiv.org/pdf/2603.00404v1
+
+## 9030. TENG-BC: Unified Time-Evolving Natural Gradient for Neural PDE Solvers with General Boundary Conditions
+
+- **ID:** `2603.00397` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Hongjie Jiang, Di Luo
+- **abs:** https://arxiv.org/abs/2603.00397 · **PDF:** https://arxiv.org/pdf/2603.00397v1
+
+## 9031. Hereditary Geometric Meta-RL: Nonlocal Generalization via Task Symmetries
+
+- **ID:** `2603.00396` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Paul Nitschke, Shahriar Talebi
+- **abs:** https://arxiv.org/abs/2603.00396 · **PDF:** https://arxiv.org/pdf/2603.00396v1
+
+## 9032. Fine-grained Soundscape Control for Augmented Hearing
+
+- **ID:** `2603.00395` · **primary:** `cs.SD` · **Submitted:** 2026-02-28
+- **Authors:** Seunghyun Oh, Malek Itani, Aseem Gauri, et al.
+- **abs:** https://arxiv.org/abs/2603.00395 · **PDF:** https://arxiv.org/pdf/2603.00395v2
+
+## 9033. The Alignment Flywheel: A Governance-Centric Hybrid MAS for Architecture-Agnostic Safety
+
+- **ID:** `2603.02259` · **primary:** `cs.MA` · **Submitted:** 2026-02-28
+- **Authors:** Elias Malomgré, Pieter Simoens
+- **abs:** https://arxiv.org/abs/2603.02259 · **PDF:** https://arxiv.org/pdf/2603.02259v1
+
+## 9034. Dual-space posterior sampling for Bayesian inference in constrained inverse problems
+
+- **ID:** `2603.00393` · **primary:** `physics.geo-ph` · **Submitted:** 2026-02-28
+- **Authors:** Ali Siahkoohi, Kamal Aghazade, Ali Gholami
+- **abs:** https://arxiv.org/abs/2603.00393 · **PDF:** https://arxiv.org/pdf/2603.00393v1
+
+## 9035. From Stochastic Answers to Verifiable Reasoning: Interpretable Decision-Making with LLM-Generated Code
+
+- **ID:** `2603.13287` · **primary:** `cs.LG` · **Submitted:** 2026-02-28
+- **Authors:** Anirudh Jaidev Mahesh, Ben Griffin, Fuat Alican, et al.
+- **abs:** https://arxiv.org/abs/2603.13287 · **PDF:** https://arxiv.org/pdf/2603.13287v1
+
+## 9036. Roots Beneath the Cut: Uncovering the Risk of Concept Revival in Pruning-Based Unlearning for Diffusion Models
+
+- **ID:** `2603.06640` · **primary:** `cs.CV` · **Submitted:** 2026-02-28
+- **Authors:** Ci Zhang, Zhaojun Ding, Chence Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.06640 · **PDF:** https://arxiv.org/pdf/2603.06640v1
+
+## 9037. Verifier-Bound Communication for LLM Agents: Certified Bounds on Covert Signaling
+
+- **ID:** `2603.00381` · **primary:** `cs.CR` · **Submitted:** 2026-02-27
+- **Authors:** Om Tailor
+- **abs:** https://arxiv.org/abs/2603.00381 · **PDF:** https://arxiv.org/pdf/2603.00381v1
+
+## 9038. Improving Full Waveform Inversion in Large Model Era
+
+- **ID:** `2603.00377` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Yinan Feng, Peng Jin, Yuzhe Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.00377 · **PDF:** https://arxiv.org/pdf/2603.00377v1
+
+## 9039. NeuroHex: Highly-Efficient Hex Coordinate System for Creating World Models to Enable Adaptive AI
+
+- **ID:** `2603.00376` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Quinn Jacobson, Joe Luo, Jingfei Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.00376 · **PDF:** https://arxiv.org/pdf/2603.00376v2
+
+## 9040. Conservative Equilibrium Discovery in Offline Game-Theoretic Multiagent Reinforcement Learning
+
+- **ID:** `2603.00374` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Austin A. Nguyen, Michael P. Wellman
+- **abs:** https://arxiv.org/abs/2603.00374 · **PDF:** https://arxiv.org/pdf/2603.00374v1
+
+## 9041. Policy Compliance of User Requests in Natural Language for AI Systems
+
+- **ID:** `2603.00369` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Pedro Cisneros-Velarde
+- **abs:** https://arxiv.org/abs/2603.00369 · **PDF:** https://arxiv.org/pdf/2603.00369v1
+
+## 9042. Deep Learning-Based Meat Freshness Detection with Segmentation and OOD-Aware Classification
+
+- **ID:** `2603.00368` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Hutama Arif Bramantyo, Mukarram Ali Faridi, Rui Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.00368 · **PDF:** https://arxiv.org/pdf/2603.00368v1
+
+## 9043. Distribution-Aware Companding Quantization of Large Language Models
+
+- **ID:** `2603.00364` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Athul Radhakrishnan, Siddhant Mohan, Mahima Sachdeva
+- **abs:** https://arxiv.org/abs/2603.00364 · **PDF:** https://arxiv.org/pdf/2603.00364v1
+
+## 9044. Quantifying Catastrophic Forgetting in IoT Intrusion Detection Systems
+
+- **ID:** `2603.00363` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Sourasekhar Banerjee, David Bergqvist, Salman Toor, et al.
+- **abs:** https://arxiv.org/abs/2603.00363 · **PDF:** https://arxiv.org/pdf/2603.00363v1
+
+## 9045. LSR: Linguistic Safety Robustness Benchmark for Low-Resource West African Languages
+
+- **ID:** `2603.19273` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Godwin Abuh Faruna
+- **abs:** https://arxiv.org/abs/2603.19273 · **PDF:** https://arxiv.org/pdf/2603.19273v1
+
+## 9046. KROM: Kernelized Reduced Order Modeling
+
+- **ID:** `2603.00360` · **primary:** `math.NA` · **Submitted:** 2026-02-27
+- **Authors:** Aras Bacho, Jonghyeon Lee, Houman Owhadi
+- **abs:** https://arxiv.org/abs/2603.00360 · **PDF:** https://arxiv.org/pdf/2603.00360v1
+
+## 9047. Universal Conceptual Structure in Neural Translation: Probing NLLB-200's Multilingual Geometry
+
+- **ID:** `2603.02258` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Kyle Elliott Mathewson
+- **abs:** https://arxiv.org/abs/2603.02258 · **PDF:** https://arxiv.org/pdf/2603.02258v1
+
+## 9048. How Large Language Models Get Stuck: Early structure with persistent errors
+
+- **ID:** `2603.00359` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Alokesh Manna, William Snyder, Whitney Tabor
+- **abs:** https://arxiv.org/abs/2603.00359 · **PDF:** https://arxiv.org/pdf/2603.00359v2
+
+## 9049. Token Management in Multi-Tenant AI Inference Platforms
+
+- **ID:** `2603.00356` · **primary:** `cs.DC` · **Submitted:** 2026-02-27
+- **Authors:** William J. Cunningham
+- **abs:** https://arxiv.org/abs/2603.00356 · **PDF:** https://arxiv.org/pdf/2603.00356v1
+
+## 9050. StethoLM: Audio Language Model for Cardiopulmonary Analysis Across Clinical Tasks
+
+- **ID:** `2603.00355` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Yishan Wang, Tsai-Ning Wang, Mathias Funk, et al.
+- **abs:** https://arxiv.org/abs/2603.00355 · **PDF:** https://arxiv.org/pdf/2603.00355v1
+
+## 9051. Acoustic Sensing for Universal Jamming Grippers
+
+- **ID:** `2603.00351` · **primary:** `cs.RO` · **Submitted:** 2026-02-27
+- **Authors:** Lion Weber, Theodor Wienert, Martin Splettstößer, et al.
+- **abs:** https://arxiv.org/abs/2603.00351 · **PDF:** https://arxiv.org/pdf/2603.00351v1
+
+## 9052. Monotropic Artificial Intelligence: Toward a Cognitive Taxonomy of Domain-Specialized Language Models
+
+- **ID:** `2603.00350` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Antonio de Sousa Leitão Filho, Allan Kardec Duailibe Barros Filho, Fabrício Saul Lima, et al.
+- **abs:** https://arxiv.org/abs/2603.00350 · **PDF:** https://arxiv.org/pdf/2603.00350v1
+
+## 9053. EmCoop: A Framework and Benchmark for Embodied Cooperation Among LLM Agents
+
+- **ID:** `2603.00349` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Hanqing Yang, Shiyu Chen, Narjes Nourzad, et al.
+- **abs:** https://arxiv.org/abs/2603.00349 · **PDF:** https://arxiv.org/pdf/2603.00349v1
+
+## 9054. Challenges in Enabling Private Data Valuation
+
+- **ID:** `2603.00342` · **primary:** `cs.CR` · **Submitted:** 2026-02-27
+- **Authors:** Yiwei Fu, Tianhao Wang, Varun Chandrasekaran
+- **abs:** https://arxiv.org/abs/2603.00342 · **PDF:** https://arxiv.org/pdf/2603.00342v1
+
+## 9055. Detecting Transportation Mode Using Dense Smartphone GPS Trajectories and Transformer Models
+
+- **ID:** `2603.00340` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Yuandong Zhang, Othmane Echchabi, Tianshu Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.00340 · **PDF:** https://arxiv.org/pdf/2603.00340v3
+
+## 9056. The Comprehension-Gated Agent Economy: A Robustness-First Architecture for AI Economic Agency
+
+- **ID:** `2603.15639` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Rahul Baxi
+- **abs:** https://arxiv.org/abs/2603.15639 · **PDF:** https://arxiv.org/pdf/2603.15639v2
+
+## 9057. Dynamic Proximal Gradient Algorithms for Schatten-$p$ Quasi-Norm Regularized Problems
+
+- **ID:** `2603.00333` · **primary:** `math.OC` · **Submitted:** 2026-02-27
+- **Authors:** Weiping Shen, Linglingzhi Zhu, Yaohua Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.00333 · **PDF:** https://arxiv.org/pdf/2603.00333v1
+
+## 9058. How Meta-research Can Pave the Road Towards Trustworthy AI In Healthcare: Catalogue of Ideas and Roadmap for Future Research
+
+- **ID:** `2603.13286` · **primary:** `cs.CY` · **Submitted:** 2026-02-27
+- **Authors:** Valerie Bürger, Marlie Besouw, Jana Fehr, et al.
+- **abs:** https://arxiv.org/abs/2603.13286 · **PDF:** https://arxiv.org/pdf/2603.13286v1
+
+## 9059. Vectorized Adaptive Histograms for Sparse Oblique Forests
+
+- **ID:** `2603.00326` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Ariel Lubonja, Jungsang Yoon, Haoyin Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.00326 · **PDF:** https://arxiv.org/pdf/2603.00326v1
+
+## 9060. Transformers are Stateless Differentiable Neural Computers
+
+- **ID:** `2603.19272` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Bo Tang, Weiwei Xie
+- **abs:** https://arxiv.org/abs/2603.19272 · **PDF:** https://arxiv.org/pdf/2603.19272v1
+
+## 9061. Robust Parameter and State Estimation in Multiscale Neuronal Systems Using Physics-Informed Neural Networks
+
+- **ID:** `2603.08742` · **primary:** `cs.NE` · **Submitted:** 2026-02-27
+- **Authors:** Changliang Wei, Yangyang Wang, Xueyu Zhu
+- **abs:** https://arxiv.org/abs/2603.08742 · **PDF:** https://arxiv.org/pdf/2603.08742v1
+
+## 9062. AESP: A Human-Sovereign Economic Protocol for AI Agents with Privacy-Preserving Settlement
+
+- **ID:** `2603.00318` · **primary:** `cs.CR` · **Submitted:** 2026-02-27
+- **Authors:** Jian Sheng Wang
+- **abs:** https://arxiv.org/abs/2603.00318 · **PDF:** https://arxiv.org/pdf/2603.00318v1
+
+## 9063. Brittlebench: Quantifying LLM robustness via prompt sensitivity
+
+- **ID:** `2603.13285` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Angelika Romanou, Mark Ibrahim, Candace Ross, et al.
+- **abs:** https://arxiv.org/abs/2603.13285 · **PDF:** https://arxiv.org/pdf/2603.13285v1
+
+## 9064. When Metrics Disagree: Automatic Similarity vs. LLM-as-a-Judge for Clinical Dialogue Evaluation
+
+- **ID:** `2603.00314` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Bian Sun, Zhenjian Wang, Orvill de la Torre, et al.
+- **abs:** https://arxiv.org/abs/2603.00314 · **PDF:** https://arxiv.org/pdf/2603.00314v2
+
+## 9065. How Well Do Multimodal Models Reason on ECG Signals?
+
+- **ID:** `2603.00312` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Maxwell A. Xu, Harish Haresamudram, Catherine W. Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.00312 · **PDF:** https://arxiv.org/pdf/2603.00312v2
+
+## 9066. DIG to Heal: Scaling General-purpose Agent Collaboration via Explainable Dynamic Decision Paths
+
+- **ID:** `2603.00309` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Hanqing Yang, Hyungwoo Lee, Yuhang Yao, et al.
+- **abs:** https://arxiv.org/abs/2603.00309 · **PDF:** https://arxiv.org/pdf/2603.00309v1
+
+## 9067. From Prerequisites to Predictions: Validating a Geometric Hallucination Taxonomy Through Controlled Induction
+
+- **ID:** `2603.00307` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Matic Korun
+- **abs:** https://arxiv.org/abs/2603.00307 · **PDF:** https://arxiv.org/pdf/2603.00307v1
+
+## 9068. When does Chain-of-Thought Help: A Markovian Perspective
+
+- **ID:** `2603.00306` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Zihan Wang, Yijun Dong, Qi Lei
+- **abs:** https://arxiv.org/abs/2603.00306 · **PDF:** https://arxiv.org/pdf/2603.00306v1
+
+## 9069. Polynomial Surrogate Training for Differentiable Ternary Logic Gate Networks
+
+- **ID:** `2603.00302` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Sai Sandeep Damera, Ryan Matheu, Aniruddh G. Puranic, et al.
+- **abs:** https://arxiv.org/abs/2603.00302 · **PDF:** https://arxiv.org/pdf/2603.00302v1
+
+## 9070. Stepwise Penalization for Length-Efficient Chain-of-Thought Reasoning
+
+- **ID:** `2603.00296` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Xintong Li, Sha Li, Rongmei Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.00296 · **PDF:** https://arxiv.org/pdf/2603.00296v1
+
+## 9071. Scalable Gaussian process modeling of parametrized spatio-temporal fields
+
+- **ID:** `2603.00290` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Srinath Dama, Prasanth B. Nair
+- **abs:** https://arxiv.org/abs/2603.00290 · **PDF:** https://arxiv.org/pdf/2603.00290v1
+
+## 9072. TraderBench: How Robust Are AI Agents in Adversarial Capital Markets?
+
+- **ID:** `2603.00285` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Xiaochuang Yuan, Hui Xu, Silvia Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.00285 · **PDF:** https://arxiv.org/pdf/2603.00285v1
+
+## 9073. HURRI-GAN: A Novel Approach for Hurricane Bias-Correction Beyond Gauge Stations using Generative Adversarial Networks
+
+- **ID:** `2603.06649` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Noujoud Nadera, Hadi Majed, Stefanos Giaremis, et al.
+- **abs:** https://arxiv.org/abs/2603.06649 · **PDF:** https://arxiv.org/pdf/2603.06649v1
+
+## 9074. Transformers Remember First, Forget Last: Dual-Process Interference in LLMs
+
+- **ID:** `2603.00270` · **primary:** `cs.IR` · **Submitted:** 2026-02-27
+- **Authors:** Sourav Chattaraj, Kanak Raj
+- **abs:** https://arxiv.org/abs/2603.00270 · **PDF:** https://arxiv.org/pdf/2603.00270v2
+
+## 9075. ObjChangeVR: Object State Change Reasoning from Continuous Egocentric Views in VR Environments
+
+- **ID:** `2603.06648` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Shiyi Ding, Shaoen Wu, Ying Chen
+- **abs:** https://arxiv.org/abs/2603.06648 · **PDF:** https://arxiv.org/pdf/2603.06648v1
+
+## 9076. Multi-Sourced, Multi-Agent Evidence Retrieval for Fact-Checking
+
+- **ID:** `2603.00267` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Shuzhi Gong, Richard O. Sinnott, Jianzhong Qi, et al.
+- **abs:** https://arxiv.org/abs/2603.00267 · **PDF:** https://arxiv.org/pdf/2603.00267v1
+
+## 9077. Query Disambiguation via Answer-Free Context: Doubling Performance on Humanity's Last Exam
+
+- **ID:** `2603.04454` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Michael Majurski, Cynthia Matuszek
+- **abs:** https://arxiv.org/abs/2603.04454 · **PDF:** https://arxiv.org/pdf/2603.04454v1
+
+## 9078. CoPeP: Benchmarking Continual Pretraining for Protein Language Models
+
+- **ID:** `2603.00253` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Darshan Patil, Pranshu Malviya, Mathieu Reymond, et al.
+- **abs:** https://arxiv.org/abs/2603.00253 · **PDF:** https://arxiv.org/pdf/2603.00253v2
+
+## 9079. A Monte Carlo estimator of flow fields for sampling and noise problems
+
+- **ID:** `2603.00252` · **primary:** `hep-lat` · **Submitted:** 2026-02-27
+- **Authors:** Michael S. Albergo, Gurtej Kanwar
+- **abs:** https://arxiv.org/abs/2603.00252 · **PDF:** https://arxiv.org/pdf/2603.00252v1
+
+## 9080. GENAI WORKBENCH: AI-Assisted Analysis and Synthesis of Engineering Systems from Multimodal Engineering Data
+
+- **ID:** `2603.00251` · **primary:** `cs.SE` · **Submitted:** 2026-02-27
+- **Authors:** H. Sinan Bank, Daniel R. Herber
+- **abs:** https://arxiv.org/abs/2603.00251 · **PDF:** https://arxiv.org/pdf/2603.00251v2
+
+## 9081. Scaling Quantum Machine Learning without Tricks: High-Resolution and Diverse Image Generation
+
+- **ID:** `2603.00233` · **primary:** `quant-ph` · **Submitted:** 2026-02-27
+- **Authors:** Jonas Jäger, Florian J. Kiwit, Carlos A. Riofrío
+- **abs:** https://arxiv.org/abs/2603.00233 · **PDF:** https://arxiv.org/pdf/2603.00233v1
+
+## 9082. Mode Seeking meets Mean Seeking for Fast Long Video Generation
+
+- **ID:** `2602.24289` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Shengqu Cai, Weili Nie, Chao Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.24289 · **PDF:** https://arxiv.org/pdf/2602.24289v1
+
+## 9083. DARE-bench: Evaluating Modeling and Instruction Fidelity of LLMs in Data Science
+
+- **ID:** `2602.24288` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Fan Shu, Yite Wang, Ruofan Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.24288 · **PDF:** https://arxiv.org/pdf/2602.24288v1
+
+## 9084. Do LLMs Benefit From Their Own Words?
+
+- **ID:** `2602.24287` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Jenny Y. Huang, Leshem Choshen, Ramon Astudillo, et al.
+- **abs:** https://arxiv.org/abs/2602.24287 · **PDF:** https://arxiv.org/pdf/2602.24287v1
+
+## 9085. CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation
+
+- **ID:** `2602.24286` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Weinan Dai, Hanlin Wu, Qiying Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.24286 · **PDF:** https://arxiv.org/pdf/2602.24286v1
+
+## 9086. Taming Momentum: Rethinking Optimizer States Through Low-Rank Approximation
+
+- **ID:** `2602.24283` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Zhengbo Wang, Jian Liang, Ran He, et al.
+- **abs:** https://arxiv.org/abs/2602.24283 · **PDF:** https://arxiv.org/pdf/2602.24283v1
+
+## 9087. Empowering Future Cybersecurity Leaders: Advancing Students through FINDS Education for Digital Forensic Excellence
+
+- **ID:** `2603.00222` · **primary:** `cs.CR` · **Submitted:** 2026-02-27
+- **Authors:** Yashas Hariprasad, Subhash Gurappa, Sundararaj S. Iyengar, et al.
+- **abs:** https://arxiv.org/abs/2603.00222 · **PDF:** https://arxiv.org/pdf/2603.00222v2
+
+## 9088. A medical coding language model trained on clinical narratives from a population-wide cohort of 1.8 million patients
+
+- **ID:** `2603.00221` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Joakim Edin, Sedrah Butt Balaganeshan, Annike Kjølby Kristensen, et al.
+- **abs:** https://arxiv.org/abs/2603.00221 · **PDF:** https://arxiv.org/pdf/2603.00221v2
+
+## 9089. Memory Caching: RNNs with Growing Memory
+
+- **ID:** `2602.24281` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Ali Behrouz, Zeman Li, Yuan Deng, et al.
+- **abs:** https://arxiv.org/abs/2602.24281 · **PDF:** https://arxiv.org/pdf/2602.24281v1
+
+## 9090. Who Guards the Guardians? The Challenges of Evaluating Identifiability of Learned Representations
+
+- **ID:** `2602.24278` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Shruti Joshi, Théo Saulus, Wieland Brendel, et al.
+- **abs:** https://arxiv.org/abs/2602.24278 · **PDF:** https://arxiv.org/pdf/2602.24278v1
+
+## 9091. Resources for Automated Evaluation of Assistive RAG Systems that Help Readers with News Trustworthiness Assessment
+
+- **ID:** `2602.24277` · **primary:** `cs.IR` · **Submitted:** 2026-02-27
+- **Authors:** Dake Zhang, Mark D. Smucker, Charles L. A. Clarke
+- **abs:** https://arxiv.org/abs/2602.24277 · **PDF:** https://arxiv.org/pdf/2602.24277v1
+
+## 9092. Induced Numerical Instability: Hidden Costs in Multimodal Large Language Models
+
+- **ID:** `2603.04453` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Wai Tuck Wong, Jun Sun, Arunesh Sinha
+- **abs:** https://arxiv.org/abs/2603.04453 · **PDF:** https://arxiv.org/pdf/2603.04453v1
+
+## 9093. A Minimal Agent for Automated Theorem Proving
+
+- **ID:** `2602.24273` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Borja Requena, Austin Letson, Krystian Nowakowski, et al.
+- **abs:** https://arxiv.org/abs/2602.24273 · **PDF:** https://arxiv.org/pdf/2602.24273v2
+
+## 9094. Efficient Discovery of Approximate Causal Abstractions via Neural Mechanism Sparsification
+
+- **ID:** `2602.24266` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Amir Asiaee
+- **abs:** https://arxiv.org/abs/2602.24266 · **PDF:** https://arxiv.org/pdf/2602.24266v1
+
+## 9095. Compositional Generalization Requires Linear, Orthogonal Representations in Vision Embedding Models
+
+- **ID:** `2602.24264` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Arnas Uselis, Andrea Dittadi, Seong Joon Oh
+- **abs:** https://arxiv.org/abs/2602.24264 · **PDF:** https://arxiv.org/pdf/2602.24264v1
+
+## 9096. Active Bipartite Ranking with Smooth Posterior Distributions
+
+- **ID:** `2602.24263` · **primary:** `stat.ML` · **Submitted:** 2026-02-27
+- **Authors:** James Cheshire, Stephan Clémençon
+- **abs:** https://arxiv.org/abs/2602.24263 · **PDF:** https://arxiv.org/pdf/2602.24263v1
+
+## 9097. Coverage-Aware Web Crawling for Domain-Specific Supplier Discovery via a Web--Knowledge--Web Pipeline
+
+- **ID:** `2602.24262` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Yijiashun Qi, Yijiazhen Qi, Tanmay Wagh
+- **abs:** https://arxiv.org/abs/2602.24262 · **PDF:** https://arxiv.org/pdf/2602.24262v3
+
+## 9098. FaultXformer: A Transformer-Encoder Based Fault Classification and Location Identification model in PMU-Integrated Active Electrical Distribution System
+
+- **ID:** `2602.24254` · **primary:** `eess.SY` · **Submitted:** 2026-02-27
+- **Authors:** Kriti Thakur, Alivelu Manga Parimi, Mayukha Pal
+- **abs:** https://arxiv.org/abs/2602.24254 · **PDF:** https://arxiv.org/pdf/2602.24254v1
+
+## 9099. Do Diffusion Models Dream of Electric Planes? Discrete and Continuous Simulation-Based Inference for Aircraft Design
+
+- **ID:** `2603.13284` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Aurelien Ghiglino, Daniel Elenius, Anirban Roy, et al.
+- **abs:** https://arxiv.org/abs/2603.13284 · **PDF:** https://arxiv.org/pdf/2603.13284v1
+
+## 9100. Histopathology Image Normalization via Latent Manifold Compaction
+
+- **ID:** `2602.24251` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Xiaolong Zhang, Jianwei Zhang, Selim Sevim, et al.
+- **abs:** https://arxiv.org/abs/2602.24251 · **PDF:** https://arxiv.org/pdf/2602.24251v1
+
+## 9101. Chunk-wise Attention Transducers for Fast and Accurate Streaming Speech-to-Text
+
+- **ID:** `2602.24245` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Hainan Xu, Vladimir Bataev, Travis M. Bartley, et al.
+- **abs:** https://arxiv.org/abs/2602.24245 · **PDF:** https://arxiv.org/pdf/2602.24245v1
+
+## 9102. Time Series Foundation Models as Strong Baselines in Transportation Forecasting: A Large-Scale Benchmark Analysis
+
+- **ID:** `2602.24238` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Javier Pulido, Filipe Rodrigues
+- **abs:** https://arxiv.org/abs/2602.24238 · **PDF:** https://arxiv.org/pdf/2602.24238v1
+
+## 9103. SafeGen-LLM: Enhancing Safety Generalization in Task Planning for Robotic Systems
+
+- **ID:** `2602.24235` · **primary:** `cs.RO` · **Submitted:** 2026-02-27
+- **Authors:** Jialiang Fan, Weizhe Xu, Mengyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.24235 · **PDF:** https://arxiv.org/pdf/2602.24235v1
+
+## 9104. Better Learning-Augmented Spanning Tree Algorithms via Metric Forest Completion
+
+- **ID:** `2602.24232` · **primary:** `cs.DS` · **Submitted:** 2026-02-27
+- **Authors:** Nate Veldt, Thomas Stanley, Benjamin W. Priest, et al.
+- **abs:** https://arxiv.org/abs/2602.24232 · **PDF:** https://arxiv.org/pdf/2602.24232v1
+
+## 9105. Adaptive Combinatorial Experimental Design: Pareto Optimality for Decision-Making and Inference
+
+- **ID:** `2602.24231` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Hongrui Xie, Junyu Cao, Kan Xu
+- **abs:** https://arxiv.org/abs/2602.24231 · **PDF:** https://arxiv.org/pdf/2602.24231v1
+
+## 9106. A Variational Estimator for $L_p$ Calibration Errors
+
+- **ID:** `2602.24230` · **primary:** `stat.ML` · **Submitted:** 2026-02-27
+- **Authors:** Eugène Berta, Sacha Braun, David Holzmüller, et al.
+- **abs:** https://arxiv.org/abs/2602.24230 · **PDF:** https://arxiv.org/pdf/2602.24230v1
+
+## 9107. BLISSNet: Deep Operator Learning for Fast and Accurate Flow Reconstruction from Sparse Sensor Measurements
+
+- **ID:** `2602.24228` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-02-27
+- **Authors:** Maksym Veremchuk, K. Andrea Scott, Zhao Pan
+- **abs:** https://arxiv.org/abs/2602.24228 · **PDF:** https://arxiv.org/pdf/2602.24228v1
+
+## 9108. MuViT: Multi-Resolution Vision Transformers for Learning Across Scales in Microscopy
+
+- **ID:** `2602.24222` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Albert Dominguez Mantes, Gioele La Manno, Martin Weigert
+- **abs:** https://arxiv.org/abs/2602.24222 · **PDF:** https://arxiv.org/pdf/2602.24222v1
+
+## 9109. Comparing Classical and Quantum Variational Classifiers on the XOR Problem
+
+- **ID:** `2602.24220` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Miras Seilkhan, Adilbek Taizhanov
+- **abs:** https://arxiv.org/abs/2602.24220 · **PDF:** https://arxiv.org/pdf/2602.24220v1
+
+## 9110. Controllable Reasoning Models Are Private Thinkers
+
+- **ID:** `2602.24210` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Haritz Puerto, Haonan Li, Xudong Han, et al.
+- **abs:** https://arxiv.org/abs/2602.24210 · **PDF:** https://arxiv.org/pdf/2602.24210v1
+
+## 9111. An Efficient Unsupervised Federated Learning Approach for Anomaly Detection in Heterogeneous IoT Networks
+
+- **ID:** `2602.24209` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Mohsen Tajgardan, Atena Shiranzaei, Mahdi Rabbani, et al.
+- **abs:** https://arxiv.org/abs/2602.24209 · **PDF:** https://arxiv.org/pdf/2602.24209v1
+
+## 9112. SenCache: Accelerating Diffusion Model Inference via Sensitivity-Aware Caching
+
+- **ID:** `2602.24208` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Yasaman Haghighi, Alexandre Alahi
+- **abs:** https://arxiv.org/abs/2602.24208 · **PDF:** https://arxiv.org/pdf/2602.24208v1
+
+## 9113. The Stability of Online Algorithms in Performative Prediction
+
+- **ID:** `2602.24207` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Gabriele Farina, Juan Carlos Perdomo
+- **abs:** https://arxiv.org/abs/2602.24207 · **PDF:** https://arxiv.org/pdf/2602.24207v1
+
+## 9114. Predictive Analytics for Foot Ulcers Using Time-Series Temperature and Pressure Data
+
+- **ID:** `2603.12278` · **primary:** `q-bio.OT` · **Submitted:** 2026-02-27
+- **Authors:** Md Tanvir Hasan Turja
+- **abs:** https://arxiv.org/abs/2603.12278 · **PDF:** https://arxiv.org/pdf/2603.12278v1
+
+## 9115. Flow-Based Density Ratio Estimation for Intractable Distributions with Applications in Genomics
+
+- **ID:** `2602.24201` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Egor Antipov, Alessandro Palma, Lorenzo Consoli, et al.
+- **abs:** https://arxiv.org/abs/2602.24201 · **PDF:** https://arxiv.org/pdf/2602.24201v1
+
+## 9116. Uncertainty Quantification for Multimodal Large Language Models with Incoherence-adjusted Semantic Volume
+
+- **ID:** `2602.24195` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Gregory Kang Ruey Lau, Hieu Dao, Nicole Kan Hui Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.24195 · **PDF:** https://arxiv.org/pdf/2602.24195v1
+
+## 9117. Resilient Strategies for Stochastic Systems: How Much Does It Take to Break a Winning Strategy?
+
+- **ID:** `2602.24191` · **primary:** `cs.GT` · **Submitted:** 2026-02-27
+- **Authors:** Kush Grover, Markel Zubia, Debraj Chakraborty, et al.
+- **abs:** https://arxiv.org/abs/2602.24191 · **PDF:** https://arxiv.org/pdf/2602.24191v1
+
+## 9118. MT-PingEval: Evaluating Multi-Turn Collaboration with Private Information Games
+
+- **ID:** `2602.24188` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Jacob Eisenstein, Fantine Huot, Adam Fisch, et al.
+- **abs:** https://arxiv.org/abs/2602.24188 · **PDF:** https://arxiv.org/pdf/2602.24188v1
+
+## 9119. A multimodal slice discovery framework for systematic failure detection and explanation in medical image classification
+
+- **ID:** `2602.24183` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Yixuan Liu, Kanwal K. Bhatia, Ahmed E. Fetit
+- **abs:** https://arxiv.org/abs/2602.24183 · **PDF:** https://arxiv.org/pdf/2602.24183v1
+
+## 9120. Multi-Objective Reinforcement Learning for Large-Scale Tote Allocation in Human-Robot Collaborative Fulfillment Centers
+
+- **ID:** `2602.24182` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Sikata Sengupta, Guangyi Liu, Omer Gottesman, et al.
+- **abs:** https://arxiv.org/abs/2602.24182 · **PDF:** https://arxiv.org/pdf/2602.24182v1
+
+## 9121. A Mixed Diet Makes DINO An Omnivorous Vision Encoder
+
+- **ID:** `2602.24181` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Rishabh Kabra, Maks Ovsjanikov, Drew A. Hudson, et al.
+- **abs:** https://arxiv.org/abs/2602.24181 · **PDF:** https://arxiv.org/pdf/2602.24181v1
+
+## 9122. Learning Flexible Job Shop Scheduling under Limited Buffers and Material Kitting Constraints
+
+- **ID:** `2602.24180` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Shishun Zhang, Juzhan Xu, Yidan Fan, et al.
+- **abs:** https://arxiv.org/abs/2602.24180 · **PDF:** https://arxiv.org/pdf/2602.24180v1
+
+## 9123. Sandwiching Polynomials for Geometric Concepts with Low Intrinsic Dimension
+
+- **ID:** `2602.24178` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Adam R. Klivans, Konstantinos Stavropoulos, Arsen Vasilyan
+- **abs:** https://arxiv.org/abs/2602.24178 · **PDF:** https://arxiv.org/pdf/2602.24178v1
+
+## 9124. Task-Centric Acceleration of Small-Language Models
+
+- **ID:** `2602.24174` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Dor Tsur, Sharon Adar, Ran Levy
+- **abs:** https://arxiv.org/abs/2602.24174 · **PDF:** https://arxiv.org/pdf/2602.24174v1
+
+## 9125. Physical Evaluation of Naturalistic Adversarial Patches for Camera-Based Traffic-Sign Detection
+
+- **ID:** `2603.00217` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Brianna D'Urso, Tahmid Hasan Sakib, Syed Rafay Hasan, et al.
+- **abs:** https://arxiv.org/abs/2603.00217 · **PDF:** https://arxiv.org/pdf/2603.00217v1
+
+## 9126. LemmaBench: A Live, Research-Level Benchmark to Evaluate LLM Capabilities in Mathematics
+
+- **ID:** `2602.24173` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Antoine Peyronnet, Fabian Gloeckle, Amaury Hayat
+- **abs:** https://arxiv.org/abs/2602.24173 · **PDF:** https://arxiv.org/pdf/2602.24173v1
+
+## 9127. ArgLLM-App: An Interactive System for Argumentative Reasoning with Large Language Models
+
+- **ID:** `2602.24172` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Adam Dejl, Deniz Gorur, Francesca Toni
+- **abs:** https://arxiv.org/abs/2602.24172 · **PDF:** https://arxiv.org/pdf/2602.24172v1
+
+## 9128. Hypothesis Testing over Observable Regimes in Singular Models
+
+- **ID:** `2602.24165` · **primary:** `math.ST` · **Submitted:** 2026-02-27
+- **Authors:** Sean Plummer
+- **abs:** https://arxiv.org/abs/2602.24165 · **PDF:** https://arxiv.org/pdf/2602.24165v1
+
+## 9129. RAViT: Resolution-Adaptive Vision Transformer
+
+- **ID:** `2602.24159` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Martial Guidez, Stefan Duffner, Christophe Garcia
+- **abs:** https://arxiv.org/abs/2602.24159 · **PDF:** https://arxiv.org/pdf/2602.24159v1
+
+## 9130. What You Read is What You Classify: Highlighting Attributions to Text and Text-Like Inputs
+
+- **ID:** `2602.24149` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Daniel S. Berman, Brian Merritt, Stanley Ta, et al.
+- **abs:** https://arxiv.org/abs/2602.24149 · **PDF:** https://arxiv.org/pdf/2602.24149v2
+
+## 9131. Learning with a Budget: Identifying the Best Arm with Resource Constraints
+
+- **ID:** `2602.24146` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Zitian Li, Wang Chi Cheung
+- **abs:** https://arxiv.org/abs/2602.24146 · **PDF:** https://arxiv.org/pdf/2602.24146v1
+
+## 9132. CoME: Empowering Channel-of-Mobile-Experts with Informative Hybrid-Capabilities Reasoning
+
+- **ID:** `2602.24142` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Yuxuan Liu, Weikai Xu, Kun Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.24142 · **PDF:** https://arxiv.org/pdf/2602.24142v2
+
+## 9133. Multimodal Optimal Transport for Unsupervised Temporal Segmentation in Surgical Robotics
+
+- **ID:** `2602.24138` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Omar Mohamed, Edoardo Fazzari, Ayah Al-Naji, et al.
+- **abs:** https://arxiv.org/abs/2602.24138 · **PDF:** https://arxiv.org/pdf/2602.24138v1
+
+## 9134. Bullet Trains: Parallelizing Training of Temporally Precise Spiking Neural Networks
+
+- **ID:** `2603.13283` · **primary:** `cs.NE` · **Submitted:** 2026-02-27
+- **Authors:** Todd Morrill, Christian Pehle, Anthony Zador
+- **abs:** https://arxiv.org/abs/2603.13283 · **PDF:** https://arxiv.org/pdf/2603.13283v1
+
+## 9135. AgenticOCR: Parsing Only What You Need for Efficient Retrieval-Augmented Generation
+
+- **ID:** `2602.24134` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Zhengren Wang, Dongsheng Ma, Huaping Zhong, et al.
+- **abs:** https://arxiv.org/abs/2602.24134 · **PDF:** https://arxiv.org/pdf/2602.24134v1
+
+## 9136. Efficient Targeted Maximum Likelihood Estimators for Two-Phase Design Problems
+
+- **ID:** `2602.24131` · **primary:** `stat.ME` · **Submitted:** 2026-02-27
+- **Authors:** Sky Qiu, Susan Gruber, Pamela A. Shaw, et al.
+- **abs:** https://arxiv.org/abs/2602.24131 · **PDF:** https://arxiv.org/pdf/2602.24131v1
+
+## 9137. End-to-end Differentiable Calibration and Reconstruction for Optical Particle Detectors
+
+- **ID:** `2602.24129` · **primary:** `hep-ex` · **Submitted:** 2026-02-27
+- **Authors:** Omar Alterkait, César Jesús-Valls, Ryo Matsumoto, et al.
+- **abs:** https://arxiv.org/abs/2602.24129 · **PDF:** https://arxiv.org/pdf/2602.24129v2
+
+## 9138. FedTreeLoRA: Reconciling Statistical and Functional Heterogeneity in Federated LoRA Fine-Tuning
+
+- **ID:** `2603.13282` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Jieming Bian, Lei Wang, Letian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.13282 · **PDF:** https://arxiv.org/pdf/2603.13282v1
+
+## 9139. Terminology Rarity Predicts Catastrophic Failure in LLM Translation of Low-Resource Ancient Languages: Evidence from Ancient Greek
+
+- **ID:** `2602.24119` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** James L. Zainaldin, Cameron Pattison, Manuela Marai, et al.
+- **abs:** https://arxiv.org/abs/2602.24119 · **PDF:** https://arxiv.org/pdf/2602.24119v1
+
+## 9140. Agentic AI-RAN: Enabling Intent-Driven, Explainable and Self-Evolving Open RAN Intelligence
+
+- **ID:** `2602.24115` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Zhizhou He, Yang Luo, Xinkai Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.24115 · **PDF:** https://arxiv.org/pdf/2602.24115v1
+
+## 9141. Toward Guarantees for Clinical Reasoning in Vision Language Models via Formal Verification
+
+- **ID:** `2602.24111` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Vikash Singh, Debargha Ganguly, Haotian Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.24111 · **PDF:** https://arxiv.org/pdf/2602.24111v1
+
+## 9142. Recycling Failures: Salvaging Exploration in RLVR via Fine-Grained Off-Policy Guidance
+
+- **ID:** `2602.24110` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Yanwei Ren, Haotian Zhang, Likang Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.24110 · **PDF:** https://arxiv.org/pdf/2602.24110v1
+
+## 9143. ARGUS: Seeing the Influence of Narrative Features on Persuasion in Argumentative Texts
+
+- **ID:** `2602.24109` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Sara Nabhani, Federico Pianzola, Khalid Al-Khatib, et al.
+- **abs:** https://arxiv.org/abs/2602.24109 · **PDF:** https://arxiv.org/pdf/2602.24109v1
+
+## 9144. Agentic Scientific Simulation: Execution-Grounded Model Construction and Reconstruction
+
+- **ID:** `2603.00214` · **primary:** `cs.SE` · **Submitted:** 2026-02-27
+- **Authors:** Knut-Andreas Lie, Olav Møyner, Elling Svee, et al.
+- **abs:** https://arxiv.org/abs/2603.00214 · **PDF:** https://arxiv.org/pdf/2603.00214v1
+
+## 9145. Artificial Agency Program: Curiosity, compression, and communication in agents
+
+- **ID:** `2602.24100` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Richard Csaky
+- **abs:** https://arxiv.org/abs/2602.24100 · **PDF:** https://arxiv.org/pdf/2602.24100v1
+
+## 9146. Bi-level RL-Heuristic Optimization for Real-world Winter Road Maintenance
+
+- **ID:** `2602.24097` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Yue Xie, Zizhen Xu, William Beazley, et al.
+- **abs:** https://arxiv.org/abs/2602.24097 · **PDF:** https://arxiv.org/pdf/2602.24097v1
+
+## 9147. DiffusionHarmonizer: Bridging Neural Reconstruction and Photorealistic Simulation with Online Diffusion Enhancer
+
+- **ID:** `2602.24096` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Yuxuan Zhang, Katarína Tóthová, Zian Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.24096 · **PDF:** https://arxiv.org/pdf/2602.24096v2
+
+## 9148. HTM-EAR: Importance-Preserving Tiered Memory with Hybrid Routing under Saturation
+
+- **ID:** `2603.10032` · **primary:** `cs.AR` · **Submitted:** 2026-02-27
+- **Authors:** Shubham Kumar Singh
+- **abs:** https://arxiv.org/abs/2603.10032 · **PDF:** https://arxiv.org/pdf/2603.10032v1
+
+## 9149. The Subjectivity of Monoculture
+
+- **ID:** `2602.24086` · **primary:** `cs.CY` · **Submitted:** 2026-02-27
+- **Authors:** Nathanael Jo, Nikhil Garg, Manish Raghavan
+- **abs:** https://arxiv.org/abs/2602.24086 · **PDF:** https://arxiv.org/pdf/2602.24086v1
+
+## 9150. Neural Diffusion Intensity Models for Point Process Data
+
+- **ID:** `2602.24083` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Xinlong Du, Harsha Honnappa, Vinayak Rao
+- **abs:** https://arxiv.org/abs/2602.24083 · **PDF:** https://arxiv.org/pdf/2602.24083v1
+
+## 9151. Preference Packing: Efficient Preference Optimization for Large Language Models
+
+- **ID:** `2602.24082` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Jaekyung Cho
+- **abs:** https://arxiv.org/abs/2602.24082 · **PDF:** https://arxiv.org/pdf/2602.24082v1
+
+## 9152. Adaptive Correlation-Weighted Intrinsic Rewards for Reinforcement Learning
+
+- **ID:** `2602.24081` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Viet Bac Nguyen, Phuong Thai Nguyen
+- **abs:** https://arxiv.org/abs/2602.24081 · **PDF:** https://arxiv.org/pdf/2602.24081v1
+
+## 9153. Human or Machine? A Preliminary Turing Test for Speech-to-Speech Interaction
+
+- **ID:** `2602.24080` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Xiang Li, Jiabao Gao, Sipei Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.24080 · **PDF:** https://arxiv.org/pdf/2602.24080v2
+
+## 9154. SongSong: A Time Phonograph for Chinese SongCi Music from Thousand of Years Away
+
+- **ID:** `2602.24071` · **primary:** `cs.SD` · **Submitted:** 2026-02-27
+- **Authors:** Jiajia Li, Jiliang Hu, Ziyi Pan, et al.
+- **abs:** https://arxiv.org/abs/2602.24071 · **PDF:** https://arxiv.org/pdf/2602.24071v1
+
+## 9155. Leveraging Non-linear Dimension Reduction and Random Walk Co-occurrence for Node Embedding
+
+- **ID:** `2602.24069` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Ryan DeWolfe
+- **abs:** https://arxiv.org/abs/2602.24069 · **PDF:** https://arxiv.org/pdf/2602.24069v1
+
+## 9156. A Novel Hierarchical Multi-Agent System for Payments Using LLMs
+
+- **ID:** `2602.24068` · **primary:** `cs.MA` · **Submitted:** 2026-02-27
+- **Authors:** Joon Kiat Chua, Donghao Huang, Zhaoxia Wang
+- **abs:** https://arxiv.org/abs/2602.24068 · **PDF:** https://arxiv.org/pdf/2602.24068v1
+
+## 9157. pathsig: A GPU-Accelerated Library for Truncated and Projected Path Signatures
+
+- **ID:** `2602.24066` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Tobias Nygaard
+- **abs:** https://arxiv.org/abs/2602.24066 · **PDF:** https://arxiv.org/pdf/2602.24066v1
+
+## 9158. Task Complexity Matters: An Empirical Study of Reasoning in LLMs for Sentiment Analysis
+
+- **ID:** `2602.24060` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Donghao Huang, Zhaoxia Wang
+- **abs:** https://arxiv.org/abs/2602.24060 · **PDF:** https://arxiv.org/pdf/2602.24060v1
+
+## 9159. Quant Experts: Token-aware Adaptive Error Reconstruction with Mixture of Experts for Large Vision-Language Models Quantization
+
+- **ID:** `2602.24059` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Chenwei Jia, Baoting Li, Xuchong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.24059 · **PDF:** https://arxiv.org/pdf/2602.24059v1
+
+## 9160. CIRCLE: A Framework for Evaluating AI from a Real-World Lens
+
+- **ID:** `2602.24055` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Reva Schwartz, Carina Westling, Morgan Briggs, et al.
+- **abs:** https://arxiv.org/abs/2602.24055 · **PDF:** https://arxiv.org/pdf/2602.24055v4
+
+## 9161. A Human-Centered Workflow for Using Large Language Models in Content Analysis
+
+- **ID:** `2603.19271` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Ivan Zupic
+- **abs:** https://arxiv.org/abs/2603.19271 · **PDF:** https://arxiv.org/pdf/2603.19271v1
+
+## 9162. Unsupervised Baseline Clustering and Incremental Adaptation for IoT Device Traffic Profiling
+
+- **ID:** `2602.24047` · **primary:** `cs.NI` · **Submitted:** 2026-02-27
+- **Authors:** Sean M. Alderman, John D. Hastings
+- **abs:** https://arxiv.org/abs/2602.24047 · **PDF:** https://arxiv.org/pdf/2602.24047v1
+
+## 9163. Data Driven Optimization of GPU efficiency for Distributed LLM Adapter Serving
+
+- **ID:** `2602.24044` · **primary:** `cs.DC` · **Submitted:** 2026-02-27
+- **Authors:** Ferran Agullo, Joan Oliveras, Chen Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.24044 · **PDF:** https://arxiv.org/pdf/2602.24044v1
+
+## 9164. ICaRus: Identical Cache Reuse for Efficient Multi Model Inference
+
+- **ID:** `2603.13281` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Sunghyeon Woo, Jaeeun Kil, Hoseung Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.13281 · **PDF:** https://arxiv.org/pdf/2603.13281v1
+
+## 9165. RewardUQ: A Unified Framework for Uncertainty-Aware Reward Models
+
+- **ID:** `2602.24040` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Daniel Yang, Samuel Stante, Florian Redhardt, et al.
+- **abs:** https://arxiv.org/abs/2602.24040 · **PDF:** https://arxiv.org/pdf/2602.24040v1
+
+## 9166. Portfolio Reinforcement Learning with Scenario-Context Rollout
+
+- **ID:** `2602.24037` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Vanya Priscillia Bendatu, Yao Lu
+- **abs:** https://arxiv.org/abs/2602.24037 · **PDF:** https://arxiv.org/pdf/2602.24037v1
+
+## 9167. Interpretable Debiasing of Vision-Language Models for Social Fairness
+
+- **ID:** `2602.24014` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Na Min An, Yoonna Jang, Yusuke Hirota, et al.
+- **abs:** https://arxiv.org/abs/2602.24014 · **PDF:** https://arxiv.org/pdf/2602.24014v1
+
+## 9168. InfoNCE Induces Gaussian Distribution
+
+- **ID:** `2602.24012` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Roy Betser, Eyal Gofer, Meir Yossef Levi, et al.
+- **abs:** https://arxiv.org/abs/2602.24012 · **PDF:** https://arxiv.org/pdf/2602.24012v1
+
+## 9169. Jailbreak Foundry: From Papers to Runnable Attacks for Reproducible Benchmarking
+
+- **ID:** `2602.24009` · **primary:** `cs.CR` · **Submitted:** 2026-02-27
+- **Authors:** Zhicheng Fang, Jingjie Zheng, Chenxu Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.24009 · **PDF:** https://arxiv.org/pdf/2602.24009v3
+
+## 9170. Inference-time optimization for experiment-grounded protein ensemble generation
+
+- **ID:** `2602.24007` · **primary:** `q-bio.BM` · **Submitted:** 2026-02-27
+- **Authors:** Advaith Maddipatla, Anar Rzayev, Marco Pegoraro, et al.
+- **abs:** https://arxiv.org/abs/2602.24007 · **PDF:** https://arxiv.org/pdf/2602.24007v2
+
+## 9171. Dialect and Gender Bias in YouTube's Spanish Captioning System
+
+- **ID:** `2602.24002` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Iris Dania Jimenez, Christoph Kern
+- **abs:** https://arxiv.org/abs/2602.24002 · **PDF:** https://arxiv.org/pdf/2602.24002v1
+
+## 9172. Architecture-Aware LLM Inference Optimization on AMD Instinct GPUs: A Comprehensive Benchmark and Deployment Study
+
+- **ID:** `2603.10031` · **primary:** `cs.AR` · **Submitted:** 2026-02-27
+- **Authors:** Athos Georgiou
+- **abs:** https://arxiv.org/abs/2603.10031 · **PDF:** https://arxiv.org/pdf/2603.10031v1
+
+## 9173. Foundation World Models for Agents that Learn, Verify, and Adapt Reliably Beyond Static Environments
+
+- **ID:** `2602.23997` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Florent Delgrange
+- **abs:** https://arxiv.org/abs/2602.23997 · **PDF:** https://arxiv.org/pdf/2602.23997v1
+
+## 9174. MEBM-Speech: Multi-scale Enhanced BrainMagic for Robust MEG Speech Detection
+
+- **ID:** `2603.02255` · **primary:** `cs.SD` · **Submitted:** 2026-02-27
+- **Authors:** Li Songyi, Zheng Linze, Liang Jinghua, et al.
+- **abs:** https://arxiv.org/abs/2603.02255 · **PDF:** https://arxiv.org/pdf/2603.02255v1
+
+## 9175. MINT: Multimodal Imaging-to-Speech Knowledge Transfer for Early Alzheimer's Screening
+
+- **ID:** `2602.23994` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Vrushank Ahire, Yogesh Kumar, Anouck Girard, et al.
+- **abs:** https://arxiv.org/abs/2602.23994 · **PDF:** https://arxiv.org/pdf/2602.23994v1
+
+## 9176. The GRADIEND Python Package: An End-to-End System for Gradient-Based Feature Learning
+
+- **ID:** `2602.23993` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Jonathan Drechsel, Steffen Herbold
+- **abs:** https://arxiv.org/abs/2602.23993 · **PDF:** https://arxiv.org/pdf/2602.23993v1
+
+## 9177. Universal NP-Hardness of Clustering under General Utilities
+
+- **ID:** `2603.00210` · **primary:** `cs.CC` · **Submitted:** 2026-02-27
+- **Authors:** Angshul Majumdar
+- **abs:** https://arxiv.org/abs/2603.00210 · **PDF:** https://arxiv.org/pdf/2603.00210v1
+
+## 9178. A unified foundational framework for knowledge injection and evaluation of Large Language Models in Combustion Science
+
+- **ID:** `2603.04452` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Zonglin Yang, Runze Mao, Tianhao Wu, et al.
+- **abs:** https://arxiv.org/abs/2603.04452 · **PDF:** https://arxiv.org/pdf/2603.04452v1
+
+## 9179. MEBM-Phoneme: Multi-scale Enhanced BrainMagic for End-to-End MEG Phoneme Classification
+
+- **ID:** `2603.02254` · **primary:** `cs.SD` · **Submitted:** 2026-02-27
+- **Authors:** Liang Jinghua, Zhang Zifeng, Li Songyi, et al.
+- **abs:** https://arxiv.org/abs/2603.02254 · **PDF:** https://arxiv.org/pdf/2603.02254v1
+
+## 9180. Intrinsic Lorentz Neural Network
+
+- **ID:** `2602.23981` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Xianglong Shi, Ziheng Chen, Yunhan Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.23981 · **PDF:** https://arxiv.org/pdf/2602.23981v1
+
+## 9181. Pessimistic Auxiliary Policy for Offline Reinforcement Learning
+
+- **ID:** `2602.23974` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Fan Zhang, Baoru Huang, Xin Zhang
+- **abs:** https://arxiv.org/abs/2602.23974 · **PDF:** https://arxiv.org/pdf/2602.23974v2
+
+## 9182. Tipping the Balance: Impact of Class Imbalance Correction on the Performance of Clinical Risk Prediction Models
+
+- **ID:** `2603.00208` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-27
+- **Authors:** Amalie Koch Andersen, Hadi Mehdizavareh, Arijit Khan, et al.
+- **abs:** https://arxiv.org/abs/2603.00208 · **PDF:** https://arxiv.org/pdf/2603.00208v1
+
+## 9183. Ask don't tell: Reducing sycophancy in large language models
+
+- **ID:** `2602.23971` · **primary:** `cs.HC` · **Submitted:** 2026-02-27
+- **Authors:** Magda Dubois, Cozmin Ududec, Christopher Summerfield, et al.
+- **abs:** https://arxiv.org/abs/2602.23971 · **PDF:** https://arxiv.org/pdf/2602.23971v2
+
+## 9184. Learning Generation Orders for Masked Discrete Diffusion Models via Variational Inference
+
+- **ID:** `2602.23968` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** David Fox, Sam Bowyer, Song Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.23968 · **PDF:** https://arxiv.org/pdf/2602.23968v1
+
+## 9185. A Stability-Aware Frozen Euler Autoencoder for Physics-Informed Tracking in Continuum Mechanics (SAFE-PIT-CM)
+
+- **ID:** `2603.13280` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Emil Hovad
+- **abs:** https://arxiv.org/abs/2603.13280 · **PDF:** https://arxiv.org/pdf/2603.13280v3
+
+## 9186. SHINE: Sequential Hierarchical Integration Network for EEG and MEG
+
+- **ID:** `2602.23960` · **primary:** `cs.SD` · **Submitted:** 2026-02-27
+- **Authors:** Xiran Xu, Yujie Yan, Xihong Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.23960 · **PDF:** https://arxiv.org/pdf/2602.23960v1
+
+## 9187. Performance Comparison of IBN orchestration using LLM and SLMs
+
+- **ID:** `2603.06647` · **primary:** `cs.NI` · **Submitted:** 2026-02-27
+- **Authors:** Wai Lwin Phone, Brahim El Boudani, Tasos Dagiuklas, et al.
+- **abs:** https://arxiv.org/abs/2603.06647 · **PDF:** https://arxiv.org/pdf/2603.06647v1
+
+## 9188. Micro-expression Recognition Based on Dual-branch Feature Extraction and Fusion
+
+- **ID:** `2602.23950` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Mingjie Zhang, Bo Li, Wanting Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.23950 · **PDF:** https://arxiv.org/pdf/2602.23950v1
+
+## 9189. HotelQuEST: Balancing Quality and Efficiency in Agentic Search
+
+- **ID:** `2602.23949` · **primary:** `cs.IR` · **Submitted:** 2026-02-27
+- **Authors:** Guy Hadad, Shadi Iskander, Oren Kalinsky, et al.
+- **abs:** https://arxiv.org/abs/2602.23949 · **PDF:** https://arxiv.org/pdf/2602.23949v1
+
+## 9190. Hierarchical Concept-based Interpretable Models
+
+- **ID:** `2602.23947` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Oscar Hill, Mateo Espinosa Zarlenga, Mateja Jamnik
+- **abs:** https://arxiv.org/abs/2602.23947 · **PDF:** https://arxiv.org/pdf/2602.23947v1
+
+## 9191. VisRef: Visual Refocusing while Thinking Improves Test-Time Scaling in Multi-Modal Large Reasoning Models
+
+- **ID:** `2603.00207` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Soumya Suvra Ghosal, Youngeun Kim, Zhuowei Li, et al.
+- **abs:** https://arxiv.org/abs/2603.00207 · **PDF:** https://arxiv.org/pdf/2603.00207v1
+
+## 9192. PointCoT: A Multi-modal Benchmark for Explicit 3D Geometric Reasoning
+
+- **ID:** `2602.23945` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Dongxu Zhang, Yiding Sun, Pengcheng Li, et al.
+- **abs:** https://arxiv.org/abs/2602.23945 · **PDF:** https://arxiv.org/pdf/2602.23945v1
+
+## 9193. MemEmo: Evaluating Emotion in Memory Systems of Agents
+
+- **ID:** `2602.23944` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Peng Liu, Zhen Tao, Jihao Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.23944 · **PDF:** https://arxiv.org/pdf/2602.23944v1
+
+## 9194. TACIT Benchmark: A Programmatic Visual Reasoning Benchmark for Generative and Discriminative Models
+
+- **ID:** `2603.00206` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Daniel Nobrega Medeiros
+- **abs:** https://arxiv.org/abs/2603.00206 · **PDF:** https://arxiv.org/pdf/2603.00206v1
+
+## 9195. EDDA-Coordinata: An Annotated Dataset of Historical Geographic Coordinates
+
+- **ID:** `2602.23941` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Ludovic Moncla, Pierre Nugues, Thierry Joliveau, et al.
+- **abs:** https://arxiv.org/abs/2602.23941 · **PDF:** https://arxiv.org/pdf/2602.23941v1
+
+## 9196. Benchmarking BERT-based Models for Sentence-level Topic Classification in Nepali Language
+
+- **ID:** `2602.23940` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Nischal Karki, Bipesh Subedi, Prakash Poudyal, et al.
+- **abs:** https://arxiv.org/abs/2602.23940 · **PDF:** https://arxiv.org/pdf/2602.23940v1
+
+## 9197. Demand Acceptance using Reinforcement Learning for Dynamic Vehicle Routing Problem with Emission Quota
+
+- **ID:** `2603.13279` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Farid Najar, Dominique Barth, Yann Strozecki
+- **abs:** https://arxiv.org/abs/2603.13279 · **PDF:** https://arxiv.org/pdf/2603.13279v1
+
+## 9198. Green or Fast? Learning to Balance Cold Starts and Idle Carbon in Serverless Computing
+
+- **ID:** `2602.23935` · **primary:** `cs.DC` · **Submitted:** 2026-02-27
+- **Authors:** Bowen Sun, Christos D. Antonopoulos, Evgenia Smirni, et al.
+- **abs:** https://arxiv.org/abs/2602.23935 · **PDF:** https://arxiv.org/pdf/2602.23935v1
+
+## 9199. Learning to Build: Autonomous Robotic Assembly of Stable Structures Without Predefined Plans
+
+- **ID:** `2602.23934` · **primary:** `cs.RO` · **Submitted:** 2026-02-27
+- **Authors:** Jingwen Wang, Johannes Kirschner, Paul Rolland, et al.
+- **abs:** https://arxiv.org/abs/2602.23934 · **PDF:** https://arxiv.org/pdf/2602.23934v1
+
+## 9200. Trust Aware Federated Learning for Secure Bone Healing Stage Interpretation in e-Health
+
+- **ID:** `2603.06646` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Paul Shepherd, Tasos Dagiuklas, Bugra Alkan, et al.
+- **abs:** https://arxiv.org/abs/2603.06646 · **PDF:** https://arxiv.org/pdf/2603.06646v1
+
+## 9201. The Astonishing Ability of Large Language Models to Parse Jabberwockified Language
+
+- **ID:** `2602.23928` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Gary Lupyan, Senyi Yang
+- **abs:** https://arxiv.org/abs/2602.23928 · **PDF:** https://arxiv.org/pdf/2602.23928v1
+
+## 9202. Efficient Flow Matching for Sparse-View CT Reconstruction
+
+- **ID:** `2603.00205` · **primary:** `eess.IV` · **Submitted:** 2026-02-27
+- **Authors:** Jiayang Shi, Lincen Yang, Zhong Li, et al.
+- **abs:** https://arxiv.org/abs/2603.00205 · **PDF:** https://arxiv.org/pdf/2603.00205v1
+
+## 9203. The Geometry of Transfer: Unlocking Medical Vision Manifolds for Training-Free Model Ranking
+
+- **ID:** `2602.23916` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Jiaqi Tang, Shaoyang Zhang, Xiaoqi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.23916 · **PDF:** https://arxiv.org/pdf/2602.23916v1
+
+## 9204. Optimisation of SOUP-GAN and CSR-GAN for High Resolution MR Images Reconstruction
+
+- **ID:** `2603.00204` · **primary:** `eess.IV` · **Submitted:** 2026-02-27
+- **Authors:** Muneeba Rashid, Hina Shakir, Humaira Mehwish, et al.
+- **abs:** https://arxiv.org/abs/2603.00204 · **PDF:** https://arxiv.org/pdf/2603.00204v1
+
+## 9205. Autonoma: A Hierarchical Multi-Agent Framework for End-to-End Workflow Automation
+
+- **ID:** `2603.19270` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Eslam Reda, Maged Yasser, Sara El-Metwally
+- **abs:** https://arxiv.org/abs/2603.19270 · **PDF:** https://arxiv.org/pdf/2603.19270v1
+
+## 9206. The AI Transformation Gap Index (AITG): An Empirical Framework for Measuring AI Transformation Opportunity, Disruption Risk, and Value Creation at the Industry and Firm Level
+
+- **ID:** `2603.13278` · **primary:** `econ.GN` · **Submitted:** 2026-02-27
+- **Authors:** Dean Barr
+- **abs:** https://arxiv.org/abs/2603.13278 · **PDF:** https://arxiv.org/pdf/2603.13278v1
+
+## 9207. SegMate: Asymmetric Attention-Based Lightweight Architecture for Efficient Multi-Organ Segmentation
+
+- **ID:** `2602.23903` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Andrei-Alexandru Bunea, Dan-Matei Popovici, Radu Tudor Ionescu
+- **abs:** https://arxiv.org/abs/2602.23903 · **PDF:** https://arxiv.org/pdf/2602.23903v1
+
+## 9208. Learning Retrieval Models with Sparse Autoencoders
+
+- **ID:** `2603.13277` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Thibault Formal, Maxime Louis, Hervé Dejean, et al.
+- **abs:** https://arxiv.org/abs/2603.13277 · **PDF:** https://arxiv.org/pdf/2603.13277v1
+
+## 9209. Experience-Guided Self-Adaptive Cascaded Agents for Breast Cancer Screening and Diagnosis with Reduced Biopsy Referrals
+
+- **ID:** `2602.23899` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Pramit Saha, Mohammad Alsharid, Joshua Strong, et al.
+- **abs:** https://arxiv.org/abs/2602.23899 · **PDF:** https://arxiv.org/pdf/2602.23899v1
+
+## 9210. Ref-Adv: Exploring MLLM Visual Reasoning in Referring Expression Tasks
+
+- **ID:** `2602.23898` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Qihua Dong, Kuo Yang, Lin Ju, et al.
+- **abs:** https://arxiv.org/abs/2602.23898 · **PDF:** https://arxiv.org/pdf/2602.23898v1
+
+## 9211. Uncovering sustainable personal care ingredient combinations using scientific modelling
+
+- **ID:** `2602.23887` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-27
+- **Authors:** Sandip Bhattacharya, Vanessa da Silva, Christina Kohlmann
+- **abs:** https://arxiv.org/abs/2602.23887 · **PDF:** https://arxiv.org/pdf/2602.23887v1
+
+## 9212. LK Losses: Direct Acceptance Rate Optimization for Speculative Decoding
+
+- **ID:** `2602.23881` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Alexander Samarin, Sergei Krutikov, Anton Shevtsov, et al.
+- **abs:** https://arxiv.org/abs/2602.23881 · **PDF:** https://arxiv.org/pdf/2602.23881v1
+
+## 9213. A Theory of Random Graph Shift in Truncated-Spectrum vRKHS
+
+- **ID:** `2602.23880` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Zhang Wan, Tingting Mu, Samuel Kaski
+- **abs:** https://arxiv.org/abs/2602.23880 · **PDF:** https://arxiv.org/pdf/2602.23880v1
+
+## 9214. RF-Agent: Automated Reward Function Design via Language Agent Tree Search
+
+- **ID:** `2602.23876` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Ning Gao, Xiuhui Zhang, Xingyu Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.23876 · **PDF:** https://arxiv.org/pdf/2602.23876v1
+
+## 9215. Exploring Robust Intrusion Detection: A Benchmark Study of Feature Transferability in IoT Botnet Attack Detection
+
+- **ID:** `2602.23874` · **primary:** `cs.CR` · **Submitted:** 2026-02-27
+- **Authors:** Alejandro Guerra-Manzanares, Jialin Huang
+- **abs:** https://arxiv.org/abs/2602.23874 · **PDF:** https://arxiv.org/pdf/2602.23874v1
+
+## 9216. Bandwidth-adaptive Cloud-Assisted 360-Degree 3D Perception for Autonomous Vehicles
+
+- **ID:** `2602.23871` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Faisal Hawladera, Rui Meireles, Gamal Elghazaly, et al.
+- **abs:** https://arxiv.org/abs/2602.23871 · **PDF:** https://arxiv.org/pdf/2602.23871v1
+
+## 9217. SWE-rebench V2: Language-Agnostic SWE Task Collection at Scale
+
+- **ID:** `2602.23866` · **primary:** `cs.SE` · **Submitted:** 2026-02-27
+- **Authors:** Ibragim Badertdinov, Maksim Nekrashevich, Anton Shevtsov, et al.
+- **abs:** https://arxiv.org/abs/2602.23866 · **PDF:** https://arxiv.org/pdf/2602.23866v1
+
+## 9218. RUMAD: Reinforcement-Unifying Multi-Agent Debate
+
+- **ID:** `2602.23864` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Chao Wang, Han Lin, Huaze Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.23864 · **PDF:** https://arxiv.org/pdf/2602.23864v1
+
+## 9219. NAU-QMUL: Utilizing BERT and CLIP for Multi-modal AI-Generated Image Detection
+
+- **ID:** `2602.23863` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Xiaoyu Guo, Arkaitz Zubiaga
+- **abs:** https://arxiv.org/abs/2602.23863 · **PDF:** https://arxiv.org/pdf/2602.23863v1
+
+## 9220. FastODT: A tree-based framework for efficient continual learning
+
+- **ID:** `2603.13276` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Daniel Bretsko, Piotr Walas, Devashish Khulbe, et al.
+- **abs:** https://arxiv.org/abs/2603.13276 · **PDF:** https://arxiv.org/pdf/2603.13276v1
+
+## 9221. A distributed semismooth Newton based augmented Lagrangian method for distributed optimization
+
+- **ID:** `2602.23854` · **primary:** `math.OC` · **Submitted:** 2026-02-27
+- **Authors:** Qihao Ma, Chengjing Wang, Peipei Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.23854 · **PDF:** https://arxiv.org/pdf/2602.23854v1
+
+## 9222. The Partition Principle Revisited: Non-Equal Volume Designs Achieve Minimal Expected Star Discrepancy
+
+- **ID:** `2603.00202` · **primary:** `stat.ML` · **Submitted:** 2026-02-27
+- **Authors:** Xiaoda Xu
+- **abs:** https://arxiv.org/abs/2603.00202 · **PDF:** https://arxiv.org/pdf/2603.00202v2
+
+## 9223. ULW-SleepNet: An Ultra-Lightweight Network for Multimodal Sleep Stage Scoring
+
+- **ID:** `2602.23852` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Zhaowen Wang, Dongdong Zhou, Qi Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.23852 · **PDF:** https://arxiv.org/pdf/2602.23852v1
+
+## 9224. MI$^2$DAS: A Multi-Layer Intrusion Detection Framework with Incremental Learning for Securing Industrial IoT Networks
+
+- **ID:** `2602.23846` · **primary:** `cs.CR` · **Submitted:** 2026-02-27
+- **Authors:** Wei Lian, Alejandro Guerra-Manzanares
+- **abs:** https://arxiv.org/abs/2602.23846 · **PDF:** https://arxiv.org/pdf/2602.23846v1
+
+## 9225. CLFEC: A New Task for Unified Linguistic and Factual Error Correction in paragraph-level Chinese Professional Writing
+
+- **ID:** `2602.23845` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Jian Kai, Zidong Zhang, Jiwen Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.23845 · **PDF:** https://arxiv.org/pdf/2602.23845v1
+
+## 9226. Enhancing Continual Learning for Software Vulnerability Prediction: Addressing Catastrophic Forgetting via Hybrid-Confidence-Aware Selective Replay for Temporal LLM Fine-Tuning
+
+- **ID:** `2602.23834` · **primary:** `cs.CR` · **Submitted:** 2026-02-27
+- **Authors:** Xuhui Dou, Hayretdin Bahsi, Alejandro Guerra-Manzanares
+- **abs:** https://arxiv.org/abs/2602.23834 · **PDF:** https://arxiv.org/pdf/2602.23834v1
+
+## 9227. FedNSAM:Consistency of Local and Global Flatness for Federated Learning
+
+- **ID:** `2602.23827` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Junkang Liu, Fanhua Shang, Yuxuan Tian, et al.
+- **abs:** https://arxiv.org/abs/2602.23827 · **PDF:** https://arxiv.org/pdf/2602.23827v1
+
+## 9228. GLUScope: A Tool for Analyzing GLU Neurons in Transformer Language Models
+
+- **ID:** `2602.23826` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Sebastian Gerstner, Hinrich Schütze
+- **abs:** https://arxiv.org/abs/2602.23826 · **PDF:** https://arxiv.org/pdf/2602.23826v1
+
+## 9229. Inferring Chronic Treatment Onset from ePrescription Data: A Renewal Process Approach
+
+- **ID:** `2602.23824` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Pavlin G. Poličar, Dalibor Stanimirović, Blaž Zupan
+- **abs:** https://arxiv.org/abs/2602.23824 · **PDF:** https://arxiv.org/pdf/2602.23824v1
+
+## 9230. Towards Neural Graph Data Management
+
+- **ID:** `2603.05529` · **primary:** `cs.DB` · **Submitted:** 2026-02-27
+- **Authors:** Yufei Li, Yisen Gao, Jiaxin Bai, et al.
+- **abs:** https://arxiv.org/abs/2603.05529 · **PDF:** https://arxiv.org/pdf/2603.05529v1
+
+## 9231. Inhibitory Cross-Talk Enables Functional Lateralization in Attention-Coupled Latent Memory
+
+- **ID:** `2603.03355` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-27
+- **Authors:** Hong Jeong
+- **abs:** https://arxiv.org/abs/2603.03355 · **PDF:** https://arxiv.org/pdf/2603.03355v1
+
+## 9232. AIDABench: AI Data Analytics Benchmark
+
+- **ID:** `2603.15636` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Yibo Yang, Fei Lei, Yixuan Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.15636 · **PDF:** https://arxiv.org/pdf/2603.15636v2
+
+## 9233. From Tokens To Agents: A Researcher's Guide To Understanding Large Language Models
+
+- **ID:** `2603.19269` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Daniele Barolo
+- **abs:** https://arxiv.org/abs/2603.19269 · **PDF:** https://arxiv.org/pdf/2603.19269v1
+
+## 9234. AdURA-Net: Adaptive Uncertainty and Region-Aware Network
+
+- **ID:** `2603.00201` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Antik Aich Roy, Ujjwal Bhattacharya
+- **abs:** https://arxiv.org/abs/2603.00201 · **PDF:** https://arxiv.org/pdf/2603.00201v1
+
+## 9235. Learning to maintain safety through expert demonstrations in settings with unknown constraints: A Q-learning perspective
+
+- **ID:** `2602.23816` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** George Papadopoulos, George A. Vouros
+- **abs:** https://arxiv.org/abs/2602.23816 · **PDF:** https://arxiv.org/pdf/2602.23816v1
+
+## 9236. Beyond State-Wise Mirror Descent: Offline Policy Optimization with Parametric Policies
+
+- **ID:** `2602.23811` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Xiang Li, Yuheng Zhang, Nan Jiang
+- **abs:** https://arxiv.org/abs/2602.23811 · **PDF:** https://arxiv.org/pdf/2602.23811v3
+
+## 9237. ReasonX: Declarative Reasoning on Explanations
+
+- **ID:** `2602.23810` · **primary:** `cs.CY` · **Submitted:** 2026-02-27
+- **Authors:** Laura State, Salvatore Ruggieri, Franco Turini
+- **abs:** https://arxiv.org/abs/2602.23810 · **PDF:** https://arxiv.org/pdf/2602.23810v1
+
+## 9238. See, Act, Adapt: Active Perception for Unsupervised Cross-Domain Visual Adaptation via Personalized VLM-Guided Agent
+
+- **ID:** `2602.23806` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Tianci Tang, Tielong Cai, Hongwei Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.23806 · **PDF:** https://arxiv.org/pdf/2602.23806v1
+
+## 9239. Actor-Critic Pretraining for Proximal Policy Optimization
+
+- **ID:** `2602.23804` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Andreas Kernbach, Amr Elsheikh, Nicolas Grupp, et al.
+- **abs:** https://arxiv.org/abs/2602.23804 · **PDF:** https://arxiv.org/pdf/2602.23804v1
+
+## 9240. LiaisonAgent: An Multi-Agent Framework for Autonomous Risk Investigation and Governance
+
+- **ID:** `2603.00200` · **primary:** `cs.CR` · **Submitted:** 2026-02-27
+- **Authors:** Chuanming Tang, Ling Qing, Shifeng Chen
+- **abs:** https://arxiv.org/abs/2603.00200 · **PDF:** https://arxiv.org/pdf/2603.00200v1
+
+## 9241. EMO-R3: Reflective Reinforcement Learning for Emotional Reasoning in Multimodal Large Language Models
+
+- **ID:** `2602.23802` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Yiyang Fang, Wenke Huang, Pei Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.23802 · **PDF:** https://arxiv.org/pdf/2602.23802v1
+
+## 9242. Operationalizing Longitudinal Causal Discovery Under Real-World Workflow Constraints
+
+- **ID:** `2602.23800` · **primary:** `stat.ME` · **Submitted:** 2026-02-27
+- **Authors:** Tadahisa Okuda, Shohei Shimizu, Thong Pham, et al.
+- **abs:** https://arxiv.org/abs/2602.23800 · **PDF:** https://arxiv.org/pdf/2602.23800v1
+
+## 9243. MPU: Towards Secure and Privacy-Preserving Knowledge Unlearning for Large Language Models
+
+- **ID:** `2602.23798` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Tiantong Wang, Xinyu Yan, Tiantong Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.23798 · **PDF:** https://arxiv.org/pdf/2602.23798v1
+
+## 9244. GRAIL: Post-hoc Compensation by Linear Reconstruction for Compressed Networks
+
+- **ID:** `2602.23795` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Wenwu Tang, Dong Wang, Lothar Thiele, et al.
+- **abs:** https://arxiv.org/abs/2602.23795 · **PDF:** https://arxiv.org/pdf/2602.23795v2
+
+## 9245. Divide and Conquer: Accelerating Diffusion-Based Large Language Models via Adaptive Parallel Decoding
+
+- **ID:** `2602.23792` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Xiangzhong Luo, Yilin An, Zhicheng Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.23792 · **PDF:** https://arxiv.org/pdf/2602.23792v1
+
+## 9246. UPath: Universal Planner Across Topological Heterogeneity For Grid-Based Pathfinding
+
+- **ID:** `2602.23789` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Aleksandr Ananikian, Daniil Drozdov, Konstantin Yakovlev
+- **abs:** https://arxiv.org/abs/2602.23789 · **PDF:** https://arxiv.org/pdf/2602.23789v1
+
+## 9247. Full-Stack Domain Enhancement for Combustion LLMs: Construction and Optimization
+
+- **ID:** `2603.19268` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Quanjia Xiao, Weimin Ouyang, Zonglin Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.19268 · **PDF:** https://arxiv.org/pdf/2603.19268v1
+
+## 9248. Provable Subspace Identification of Nonlinear Multi-view CCA
+
+- **ID:** `2602.23785` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Zhiwei Han, Stefan Matthes, Hao Shen
+- **abs:** https://arxiv.org/abs/2602.23785 · **PDF:** https://arxiv.org/pdf/2602.23785v1
+
+## 9249. TradeFM: A Generative Foundation Model for Trade-flow and Market Microstructure
+
+- **ID:** `2602.23784` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Maxime Kawawa-Beaudan, Srijan Sood, Kassiani Papasotiriou, et al.
+- **abs:** https://arxiv.org/abs/2602.23784 · **PDF:** https://arxiv.org/pdf/2602.23784v1
+
+## 9250. Multi-Condition Digital Twin Calibration for Axial Piston Pumps : Compound Fault Simulation
+
+- **ID:** `2603.00199` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-02-27
+- **Authors:** Chang Dong, Jianfeng Tao, Chengliang Liu
+- **abs:** https://arxiv.org/abs/2603.00199 · **PDF:** https://arxiv.org/pdf/2603.00199v2
+
+## 9251. Stateful Token Reduction for Long-Video Hybrid VLMs
+
+- **ID:** `2603.00198` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Jindong Jiang, Amala Sanjay Deshmukh, Kateryna Chumachenko, et al.
+- **abs:** https://arxiv.org/abs/2603.00198 · **PDF:** https://arxiv.org/pdf/2603.00198v1
+
+## 9252. Reasoning-Driven Multimodal LLM for Domain Generalization
+
+- **ID:** `2602.23777` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Zhipeng Xu, Zilong Wang, Xinyang Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.23777 · **PDF:** https://arxiv.org/pdf/2602.23777v1
+
+## 9253. A Case Study on Concept Induction for Neuron-Level Interpretability in CNN
+
+- **ID:** `2603.00197` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Moumita Sen Sarma, Samatha Ereshi Akkamahadevi, Pascal Hitzler
+- **abs:** https://arxiv.org/abs/2603.00197 · **PDF:** https://arxiv.org/pdf/2603.00197v1
+
+## 9254. MAGE: Multi-scale Autoregressive Generation for Offline Reinforcement Learning
+
+- **ID:** `2602.23770` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Chenxing Lin, Xinhui Gao, Haipeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.23770 · **PDF:** https://arxiv.org/pdf/2602.23770v1
+
+## 9255. OPTIAGENT: A Physics-Driven Agentic Framework for Automated Optical Design
+
+- **ID:** `2602.23761` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Yuyu Geng, Lei Sun, Yao Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.23761 · **PDF:** https://arxiv.org/pdf/2602.23761v1
+
+## 9256. Structured Prompt Optimization for Few-Shot Text Classification via Semantic Alignment in Latent Space
+
+- **ID:** `2602.23753` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Jiasen Zheng, Zijun Zhou, Huajun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.23753 · **PDF:** https://arxiv.org/pdf/2602.23753v1
+
+## 9257. Predictive Hotspot Mapping for Data-driven Crime Prediction
+
+- **ID:** `2602.23750` · **primary:** `stat.AP` · **Submitted:** 2026-02-27
+- **Authors:** Karthik Sriram, Ankur Sinha, Suvashis Choudhary
+- **abs:** https://arxiv.org/abs/2602.23750 · **PDF:** https://arxiv.org/pdf/2602.23750v1
+
+## 9258. PREBA: Surgical Duration Prediction via PCA-Weighted Retrieval-Augmented LLMs and Bayesian Averaging Aggregation
+
+- **ID:** `2603.13275` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Wanyin Wu, Kanxue Li, Baosheng Yu, et al.
+- **abs:** https://arxiv.org/abs/2603.13275 · **PDF:** https://arxiv.org/pdf/2603.13275v3
+
+## 9259. Bridging Dynamics Gaps via Diffusion Schrödinger Bridge for Cross-Domain Reinforcement Learning
+
+- **ID:** `2602.23737` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Hanping Zhang, Yuhong Guo
+- **abs:** https://arxiv.org/abs/2602.23737 · **PDF:** https://arxiv.org/pdf/2602.23737v1
+
+## 9260. UTPTrack: Towards Simple and Unified Token Pruning for Visual Tracking
+
+- **ID:** `2602.23734` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Hao Wu, Xudong Wang, Jialiang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.23734 · **PDF:** https://arxiv.org/pdf/2602.23734v1
+
+## 9261. Unlocking Cognitive Capabilities and Analyzing the Perception-Logic Trade-off
+
+- **ID:** `2602.23730` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Longyin Zhang, Shuo Sun, Yingxu He, et al.
+- **abs:** https://arxiv.org/abs/2602.23730 · **PDF:** https://arxiv.org/pdf/2602.23730v1
+
+## 9262. On Emergences of Non-Classical Statistical Characteristics in Classical Neural Networks
+
+- **ID:** `2603.04451` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Hanyu Zhao, Yang Wu, Yuexian Hou
+- **abs:** https://arxiv.org/abs/2603.04451 · **PDF:** https://arxiv.org/pdf/2603.04451v1
+
+## 9263. From Static Benchmarks to Dynamic Protocol: Agent-Centric Text Anomaly Detection for Evaluating LLM Reasoning
+
+- **ID:** `2602.23729` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Seungdong Yoa, Sanghyu Yoon, Suhee Yoon, et al.
+- **abs:** https://arxiv.org/abs/2602.23729 · **PDF:** https://arxiv.org/pdf/2602.23729v1
+
+## 9264. SLA-Aware Distributed LLM Inference Across Device-RAN-Cloud
+
+- **ID:** `2602.23722` · **primary:** `cs.NI` · **Submitted:** 2026-02-27
+- **Authors:** Hariz Yet, Nguyen Thanh Tam, Mao V. Ngo, et al.
+- **abs:** https://arxiv.org/abs/2602.23722 · **PDF:** https://arxiv.org/pdf/2602.23722v1
+
+## 9265. The Auton Agentic AI Framework
+
+- **ID:** `2602.23720` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Sheng Cao, Zhao Chang, Chang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.23720 · **PDF:** https://arxiv.org/pdf/2602.23720v1
+
+## 9266. Omni-C: Compressing Heterogeneous Modalities into a Single Dense Encoder
+
+- **ID:** `2603.05528` · **primary:** `cs.MM` · **Submitted:** 2026-02-27
+- **Authors:** Kin Wai Lau, Yasar Abbas Ur Rehman, Lai-Man Po, et al.
+- **abs:** https://arxiv.org/abs/2603.05528 · **PDF:** https://arxiv.org/pdf/2603.05528v1
+
+## 9267. SAGE-LLM: Towards Safe and Generalizable LLM Controller with Fuzzy-CBF Verification and Graph-Structured Knowledge Retrieval for UAV Decision
+
+- **ID:** `2602.23719` · **primary:** `cs.RO` · **Submitted:** 2026-02-27
+- **Authors:** Wenzhe Zhao, Yang Zhao, Ganchao Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.23719 · **PDF:** https://arxiv.org/pdf/2602.23719v1
+
+## 9268. Your Inference Request Will Become a Black Box: Confidential Inference for Cloud-based Large Language Models
+
+- **ID:** `2603.00196` · **primary:** `cs.CR` · **Submitted:** 2026-02-27
+- **Authors:** Chung-ju Huang, Huiqiang Zhao, Yuanpeng He, et al.
+- **abs:** https://arxiv.org/abs/2603.00196 · **PDF:** https://arxiv.org/pdf/2603.00196v1
+
+## 9269. ProductResearch: Training E-Commerce Deep Research Agents via Multi-Agent Synthetic Trajectory Distillation
+
+- **ID:** `2602.23716` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Jiangyuan Wang, Kejun Xiao, Huaipeng Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.23716 · **PDF:** https://arxiv.org/pdf/2602.23716v1
+
+## 9270. Formal Analysis and Supply Chain Security for Agentic AI Skills
+
+- **ID:** `2603.00195` · **primary:** `cs.CR` · **Submitted:** 2026-02-27
+- **Authors:** Varun Pratap Bhardwaj
+- **abs:** https://arxiv.org/abs/2603.00195 · **PDF:** https://arxiv.org/pdf/2603.00195v1
+
+## 9271. SKeDA: A Generative Watermarking Framework for Text-to-video Diffusion Models
+
+- **ID:** `2603.00194` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Yang Yang, Xinze Zou, Zehua Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.00194 · **PDF:** https://arxiv.org/pdf/2603.00194v1
+
+## 9272. A Boundary Integral-based Neural Operator for Mesh Deformation
+
+- **ID:** `2602.23703` · **primary:** `math.NA` · **Submitted:** 2026-02-27
+- **Authors:** Zhengyu Wu, Jun Liu, Wei Wang
+- **abs:** https://arxiv.org/abs/2602.23703 · **PDF:** https://arxiv.org/pdf/2602.23703v2
+
+## 9273. From Flat Logs to Causal Graphs: Hierarchical Failure Attribution for LLM-based Multi-Agent Systems
+
+- **ID:** `2602.23701` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Yawen Wang, Wenjie Wu, Junjie Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.23701 · **PDF:** https://arxiv.org/pdf/2602.23701v1
+
+## 9274. Learning from Partial Chain-of-Thought via Truncated-Reasoning Self-Distillation
+
+- **ID:** `2603.13274` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Gianluigi Silvestri, Edoardo Cetin
+- **abs:** https://arxiv.org/abs/2603.13274 · **PDF:** https://arxiv.org/pdf/2603.13274v1
+
+## 9275. HiDrop: Hierarchical Vision Token Reduction in MLLMs via Late Injection, Concave Pyramid Pruning, and Early Exit
+
+- **ID:** `2602.23699` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Hao Wu, Yingqi Fan, Jinyang Dai, et al.
+- **abs:** https://arxiv.org/abs/2602.23699 · **PDF:** https://arxiv.org/pdf/2602.23699v1
+
+## 9276. Non-Invasive Reconstruction of Intracranial EEG Across the Deep Temporal Lobe from Scalp EEG based on Conditional Normalizing Flow
+
+- **ID:** `2603.03354` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-27
+- **Authors:** Dongyi He, Bin Jiang, Kecheng Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.03354 · **PDF:** https://arxiv.org/pdf/2603.03354v2
+
+## 9277. Optimizer-Induced Low-Dimensional Drift and Transverse Dynamics in Transformer Training
+
+- **ID:** `2602.23696` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Yongzhong Xu
+- **abs:** https://arxiv.org/abs/2602.23696 · **PDF:** https://arxiv.org/pdf/2602.23696v3
+
+## 9278. Interpretable Multimodal Gesture Recognition for Drone and Mobile Robot Teleoperation via Log-Likelihood Ratio Fusion
+
+- **ID:** `2602.23694` · **primary:** `cs.RO` · **Submitted:** 2026-02-27
+- **Authors:** Seungyeol Baek, Jaspreet Singh, Lala Shakti Swarup Ray, et al.
+- **abs:** https://arxiv.org/abs/2602.23694 · **PDF:** https://arxiv.org/pdf/2602.23694v2
+
+## 9279. Spatially Aware Deep Learning for Microclimate Prediction from High-Resolution Geospatial Imagery
+
+- **ID:** `2603.13273` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Idan Sulami, Alon Itzkovitch, Michael R. Kearney, et al.
+- **abs:** https://arxiv.org/abs/2603.13273 · **PDF:** https://arxiv.org/pdf/2603.13273v1
+
+## 9280. ODAR: Principled Adaptive Routing for LLM Reasoning via Active Inference
+
+- **ID:** `2602.23681` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Siyuan Ma, Bo Gao, Xiaojun Jia, et al.
+- **abs:** https://arxiv.org/abs/2602.23681 · **PDF:** https://arxiv.org/pdf/2602.23681v1
+
+## 9281. CAMEL-CLIP: Channel-aware Multimodal Electroencephalography-text Alignment for Generalizable Brain Foundation Models
+
+- **ID:** `2603.13272` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Hanseul Choi, Jinyeong Park, Seongwon Jin, et al.
+- **abs:** https://arxiv.org/abs/2603.13272 · **PDF:** https://arxiv.org/pdf/2603.13272v1
+
+## 9282. The Compulsory Imaginary: AGI and Corporate Authority
+
+- **ID:** `2602.23679` · **primary:** `cs.HC` · **Submitted:** 2026-02-27
+- **Authors:** Emilio Barkett
+- **abs:** https://arxiv.org/abs/2602.23679 · **PDF:** https://arxiv.org/pdf/2602.23679v1
+
+## 9283. Any Model, Any Place, Any Time: Get Remote Sensing Foundation Model Embeddings On Demand
+
+- **ID:** `2602.23678` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Dingqi Ye, Daniel Kiv, Wei Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.23678 · **PDF:** https://arxiv.org/pdf/2602.23678v1
+
+## 9284. General Bayesian Policy Learning
+
+- **ID:** `2602.23672` · **primary:** `stat.ML` · **Submitted:** 2026-02-27
+- **Authors:** Masahiro Kato
+- **abs:** https://arxiv.org/abs/2602.23672 · **PDF:** https://arxiv.org/pdf/2602.23672v1
+
+## 9285. PseudoAct: Leveraging Pseudocode Synthesis for Flexible Planning and Action Control in Large Language Model Agents
+
+- **ID:** `2602.23668` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Yihan, Wen, Xin Chen
+- **abs:** https://arxiv.org/abs/2602.23668 · **PDF:** https://arxiv.org/pdf/2602.23668v1
+
+## 9286. Blockchain-Enabled Routing for Zero-Trust Low-Altitude Intelligent Networks
+
+- **ID:** `2602.23667` · **primary:** `cs.NI` · **Submitted:** 2026-02-27
+- **Authors:** Ziye Jia, Sijie He, Ligang Yuan, et al.
+- **abs:** https://arxiv.org/abs/2602.23667 · **PDF:** https://arxiv.org/pdf/2602.23667v1
+
+## 9287. Active Learning for Planet Habitability Classification under Extreme Class Imbalance
+
+- **ID:** `2602.23666` · **primary:** `astro-ph.EP` · **Submitted:** 2026-02-27
+- **Authors:** R. I. El-Kholy, Z. M. Hayman
+- **abs:** https://arxiv.org/abs/2602.23666 · **PDF:** https://arxiv.org/pdf/2602.23666v1
+
+## 9288. Geodesic Semantic Search: Learning Local Riemannian Metrics for Citation Graph Retrieval
+
+- **ID:** `2602.23665` · **primary:** `cs.IR` · **Submitted:** 2026-02-27
+- **Authors:** Brandon Yee, Lucas Wang, Kundana Kommini, et al.
+- **abs:** https://arxiv.org/abs/2602.23665 · **PDF:** https://arxiv.org/pdf/2602.23665v2
+
+## 9289. Disentangled Mode-Specific Representations for Tensor Time Series via Contrastive Learning
+
+- **ID:** `2602.23663` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Kohei Obata, Taichi Murayama, Zheng Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.23663 · **PDF:** https://arxiv.org/pdf/2602.23663v1
+
+## 9290. Tracing the Evolution of Word Embedding Techniques in Natural Language Processing
+
+- **ID:** `2603.13271` · **primary:** `cs.CY` · **Submitted:** 2026-02-27
+- **Authors:** Minh Anh Nguyen, Kuheli Sai, Minh Nguyen
+- **abs:** https://arxiv.org/abs/2603.13271 · **PDF:** https://arxiv.org/pdf/2603.13271v1
+
+## 9291. Selective Denoising Diffusion Model for Time Series Anomaly Detection
+
+- **ID:** `2602.23662` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Kohei Obata, Zheng Chen, Yasuko Matsubara, et al.
+- **abs:** https://arxiv.org/abs/2602.23662 · **PDF:** https://arxiv.org/pdf/2602.23662v1
+
+## 9292. Diagnostics for Individual-Level Prediction Instability in Machine Learning for Healthcare
+
+- **ID:** `2603.00192` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Elizabeth W. Miller, Jeffrey D. Blume
+- **abs:** https://arxiv.org/abs/2603.00192 · **PDF:** https://arxiv.org/pdf/2603.00192v1
+
+## 9293. TRIZ-RAGNER: A Retrieval-Augmented Large Language Model for TRIZ-Aware Named Entity Recognition in Patent-Based Contradiction Mining
+
+- **ID:** `2602.23656` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Zitong Xu, Yuqing Wu, Yue Zhao
+- **abs:** https://arxiv.org/abs/2602.23656 · **PDF:** https://arxiv.org/pdf/2602.23656v1
+
+## 9294. ProtoDCS: Towards Robust and Efficient Open-Set Test-Time Adaptation for Vision-Language Models
+
+- **ID:** `2602.23653` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Wei Luo, Yangfan Ou, Jin Deng, et al.
+- **abs:** https://arxiv.org/abs/2602.23653 · **PDF:** https://arxiv.org/pdf/2602.23653v2
+
+## 9295. 3D Modality-Aware Pre-training for Vision-Language Model in MRI Multi-organ Abnormality Detection
+
+- **ID:** `2602.23652` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Haowen Zhu, Ning Yin, Xiaogen Zhou
+- **abs:** https://arxiv.org/abs/2602.23652 · **PDF:** https://arxiv.org/pdf/2602.23652v2
+
+## 9296. AudioCapBench: Quick Evaluation on Audio Captioning across Sound, Music, and Speech
+
+- **ID:** `2602.23649` · **primary:** `cs.SD` · **Submitted:** 2026-02-27
+- **Authors:** Jielin Qiu, Jianguo Zhang, Zixiang Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.23649 · **PDF:** https://arxiv.org/pdf/2602.23649v1
+
+## 9297. A Controlled Comparison of Deep Learning Architectures for Multi-Horizon Financial Forecasting: Evidence from 918 Experiments
+
+- **ID:** `2603.16886` · **primary:** `q-fin.ST` · **Submitted:** 2026-02-27
+- **Authors:** Nabeel Ahmad Saidd
+- **abs:** https://arxiv.org/abs/2603.16886 · **PDF:** https://arxiv.org/pdf/2603.16886v1
+
+## 9298. AI Must Embrace Specialization via Superhuman Adaptable Intelligence
+
+- **ID:** `2602.23643` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Judah Goldfeder, Philippe Wyder, Yann LeCun, et al.
+- **abs:** https://arxiv.org/abs/2602.23643 · **PDF:** https://arxiv.org/pdf/2602.23643v1
+
+## 9299. FedRot-LoRA: Mitigating Rotational Misalignment in Federated LoRA
+
+- **ID:** `2602.23638` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Haoran Zhang, Dongjun Kim, Seohyeon Cha, et al.
+- **abs:** https://arxiv.org/abs/2602.23638 · **PDF:** https://arxiv.org/pdf/2602.23638v1
+
+## 9300. FlexGuard: Continuous Risk Scoring for Strictness-Adaptive LLM Content Moderation
+
+- **ID:** `2602.23636` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Zhihao Ding, Jinming Li, Ze Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.23636 · **PDF:** https://arxiv.org/pdf/2602.23636v2
+
+## 9301. On the Convergence of Single-Loop Stochastic Bilevel Optimization with Approximate Implicit Differentiation
+
+- **ID:** `2602.23633` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Yubo Zhou, Luo Luo, Guang Dai, et al.
+- **abs:** https://arxiv.org/abs/2602.23633 · **PDF:** https://arxiv.org/pdf/2602.23633v1
+
+## 9302. MMKG-RDS: Reasoning Data Synthesis via Deep Mining of Multimodal Knowledge Graphs
+
+- **ID:** `2602.23632` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Lun Zhan, Feng Xiong, Huanyong Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.23632 · **PDF:** https://arxiv.org/pdf/2602.23632v1
+
+## 9303. BTTackler: A Diagnosis-based Framework for Efficient Deep Learning Hyperparameter Optimization
+
+- **ID:** `2602.23630` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Zhongyi Pei, Zhiyao Cen, Yipeng Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.23630 · **PDF:** https://arxiv.org/pdf/2602.23630v1
+
+## 9304. Multivariate Spatio-Temporal Neural Hawkes Processes
+
+- **ID:** `2602.23629` · **primary:** `stat.ML` · **Submitted:** 2026-02-27
+- **Authors:** Christopher Chukwuemeka, Hojun You, Mikyoung Jun
+- **abs:** https://arxiv.org/abs/2602.23629 · **PDF:** https://arxiv.org/pdf/2602.23629v2
+
+## 9305. DLEBench: Evaluating Small-scale Object Editing Ability for Instruction-based Image Editing Model
+
+- **ID:** `2602.23622` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Shibo Hong, Boxian Ai, Jun Kuang, et al.
+- **abs:** https://arxiv.org/abs/2602.23622 · **PDF:** https://arxiv.org/pdf/2602.23622v1
+
+## 9306. ReDON: Recurrent Diffractive Optical Neural Processor with Reconfigurable Self-Modulated Nonlinearity
+
+- **ID:** `2602.23616` · **primary:** `physics.optics` · **Submitted:** 2026-02-27
+- **Authors:** Ziang Yin, Qi Jing, Raktim Sarma, et al.
+- **abs:** https://arxiv.org/abs/2602.23616 · **PDF:** https://arxiv.org/pdf/2602.23616v3
+
+## 9307. When Does Multimodal Learning Help in Healthcare? A Benchmark on EHR and Chest X-Ray Fusion
+
+- **ID:** `2602.23614` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Kejing Yin, Haizhou Xu, Wenfang Yao, et al.
+- **abs:** https://arxiv.org/abs/2602.23614 · **PDF:** https://arxiv.org/pdf/2602.23614v1
+
+## 9308. Task-Driven Subspace Decomposition for Knowledge Sharing and Isolation in LoRA-based Continual Learning
+
+- **ID:** `2603.00191` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Lingfeng He, De Cheng, Huaijie Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.00191 · **PDF:** https://arxiv.org/pdf/2603.00191v1
+
+## 9309. Fairness under Graph Uncertainty: Achieving Interventional Fairness with Partially Known Causal Graphs over Clusters of Variables
+
+- **ID:** `2602.23611` · **primary:** `stat.ML` · **Submitted:** 2026-02-27
+- **Authors:** Yoichi Chikahara
+- **abs:** https://arxiv.org/abs/2602.23611 · **PDF:** https://arxiv.org/pdf/2602.23611v1
+
+## 9310. LLM-Driven Multi-Turn Task-Oriented Dialogue Synthesis for Realistic Reasoning
+
+- **ID:** `2602.23610` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Yu Zhu, Kai Yang
+- **abs:** https://arxiv.org/abs/2602.23610 · **PDF:** https://arxiv.org/pdf/2602.23610v1
+
+## 9311. SleepLM: Natural-Language Intelligence for Human Sleep
+
+- **ID:** `2602.23605` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Zongzhe Xu, Zitao Shuai, Eideen Mozaffari, et al.
+- **abs:** https://arxiv.org/abs/2602.23605 · **PDF:** https://arxiv.org/pdf/2602.23605v1
+
+## 9312. OSF: On Pre-training and Scaling of Sleep Foundation Models
+
+- **ID:** `2603.00190` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Zitao Shuai, Zongzhe Xu, David Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.00190 · **PDF:** https://arxiv.org/pdf/2603.00190v1
+
+## 9313. LFQA-HP-1M: A Large-Scale Human Preference Dataset for Long-Form Question Answering
+
+- **ID:** `2602.23603` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Rafid Ishrak Jahan, Fahmid Shahriar Iqbal, Sagnik Ray Choudhury
+- **abs:** https://arxiv.org/abs/2602.23603 · **PDF:** https://arxiv.org/pdf/2602.23603v1
+
+## 9314. Moment Matters: Mean and Variance Causal Graph Discovery from Heteroscedastic Observational Data
+
+- **ID:** `2602.23602` · **primary:** `stat.ML` · **Submitted:** 2026-02-27
+- **Authors:** Yoichi Chikahara
+- **abs:** https://arxiv.org/abs/2602.23602 · **PDF:** https://arxiv.org/pdf/2602.23602v1
+
+## 9315. Normalisation and Initialisation Strategies for Graph Neural Networks in Blockchain Anomaly Detection
+
+- **ID:** `2602.23599` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Dang Sy Duy, Nguyen Duy Chien, Kapil Dev, et al.
+- **abs:** https://arxiv.org/abs/2602.23599 · **PDF:** https://arxiv.org/pdf/2602.23599v1
+
+## 9316. KEEP: A KV-Cache-Centric Memory Management System for Efficient Embodied Planning
+
+- **ID:** `2602.23592` · **primary:** `cs.RO` · **Submitted:** 2026-02-27
+- **Authors:** Zebin Yang, Tong Xie, Baotong Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.23592 · **PDF:** https://arxiv.org/pdf/2602.23592v2
+
+## 9317. Whisper-RIR-Mega: A Paired Clean-Reverberant Speech Benchmark for ASR Robustness to Room Acoustics
+
+- **ID:** `2603.02252` · **primary:** `eess.AS` · **Submitted:** 2026-02-27
+- **Authors:** Mandip Goswami
+- **abs:** https://arxiv.org/abs/2603.02252 · **PDF:** https://arxiv.org/pdf/2603.02252v3
+
+## 9318. Pseudo Contrastive Learning for Diagram Comprehension in Multimodal Models
+
+- **ID:** `2602.23589` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Hiroshi Sasaki
+- **abs:** https://arxiv.org/abs/2602.23589 · **PDF:** https://arxiv.org/pdf/2602.23589v2
+
+## 9319. Hyperdimensional Cross-Modal Alignment of Frozen Language and Image Models for Efficient Image Captioning
+
+- **ID:** `2602.23588` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Abhishek Dalvi, Vasant Honavar
+- **abs:** https://arxiv.org/abs/2602.23588 · **PDF:** https://arxiv.org/pdf/2602.23588v1
+
+## 9320. Efficient Long-Horizon GUI Agents via Training-Free KV Cache Compression
+
+- **ID:** `2603.00188` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Bowen Zhou, Zhou Xu, Wanli Li, et al.
+- **abs:** https://arxiv.org/abs/2603.00188 · **PDF:** https://arxiv.org/pdf/2603.00188v1
+
+## 9321. SDMixer: Sparse Dual-Mixer for Time Series Forecasting
+
+- **ID:** `2602.23581` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Xiang Ao
+- **abs:** https://arxiv.org/abs/2602.23581 · **PDF:** https://arxiv.org/pdf/2602.23581v1
+
+## 9322. BRIDGE the Gap: Mitigating Bias Amplification in Automated Scoring of English Language Learners via Inter-group Data Augmentation
+
+- **ID:** `2602.23580` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Yun Wang, Xuansheng Wu, Jingyuan Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.23580 · **PDF:** https://arxiv.org/pdf/2602.23580v1
+
+## 9323. Construct, Merge, Solve & Adapt with Reinforcement Learning for the min-max Multiple Traveling Salesman Problem
+
+- **ID:** `2602.23579` · **primary:** `cs.AI` · **Submitted:** 2026-02-27
+- **Authors:** Guillem Rodríguez-Corominas, Maria J. Blesa, Christian Blum
+- **abs:** https://arxiv.org/abs/2602.23579 · **PDF:** https://arxiv.org/pdf/2602.23579v1
+
+## 9324. Hybrid Quantum Temporal Convolutional Networks
+
+- **ID:** `2602.23578` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Junghoon Justin Park, Maria Pak, Sebin Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.23578 · **PDF:** https://arxiv.org/pdf/2602.23578v1
+
+## 9325. Multi-Agent Causal Reasoning for Suicide Ideation Detection Through Online Conversations
+
+- **ID:** `2602.23577` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Jun Li, Xiangmeng Wang, Haoyang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.23577 · **PDF:** https://arxiv.org/pdf/2602.23577v1
+
+## 9326. CycleBEV: Regularizing View Transformation Networks via View Cycle Consistency for Bird's-Eye-View Semantic Segmentation
+
+- **ID:** `2602.23575` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Jeongbin Hong, Dooseop Choi, Taeg-Hyun An, et al.
+- **abs:** https://arxiv.org/abs/2602.23575 · **PDF:** https://arxiv.org/pdf/2602.23575v1
+
+## 9327. Evidential Neural Radiance Fields
+
+- **ID:** `2602.23574` · **primary:** `cs.CV` · **Submitted:** 2026-02-27
+- **Authors:** Ruxiao Duan, Alex Wong
+- **abs:** https://arxiv.org/abs/2602.23574 · **PDF:** https://arxiv.org/pdf/2602.23574v2
+
+## 9328. All Mutation Rates $c/n$ for the $(1+1)$ Evolutionary Algorithm
+
+- **ID:** `2602.23573` · **primary:** `cs.NE` · **Submitted:** 2026-02-27
+- **Authors:** Andrew James Kelley
+- **abs:** https://arxiv.org/abs/2602.23573 · **PDF:** https://arxiv.org/pdf/2602.23573v1
+
+## 9329. Reviewing the Reviewer: Graph-Enhanced LLMs for E-commerce Appeal Adjudication
+
+- **ID:** `2603.19267` · **primary:** `cs.CL` · **Submitted:** 2026-02-27
+- **Authors:** Yuchen Du, Ashley Li, Zixi Huang
+- **abs:** https://arxiv.org/abs/2603.19267 · **PDF:** https://arxiv.org/pdf/2603.19267v1
+
+## 9330. Tensor Hypercontraction Error Correction Using Regression
+
+- **ID:** `2602.23567` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-27
+- **Authors:** Ishna Satyarth, Eric C. Larson, Devin A. Matthews
+- **abs:** https://arxiv.org/abs/2602.23567 · **PDF:** https://arxiv.org/pdf/2602.23567v1
+
+## 9331. Flowette: Flow Matching with Graphette Priors for Graph Generation
+
+- **ID:** `2602.23566` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Asiri Wijesinghe, Sevvandi Kandanaarachchi, Daniel M. Steinberg, et al.
+- **abs:** https://arxiv.org/abs/2602.23566 · **PDF:** https://arxiv.org/pdf/2602.23566v1
+
+## 9332. Dynamics of Learning under User Choice: Overspecialization and Peer-Model Probing
+
+- **ID:** `2602.23565` · **primary:** `cs.LG` · **Submitted:** 2026-02-27
+- **Authors:** Adhyyan Narang, Sarah Dean, Lillian J Ratliff, et al.
+- **abs:** https://arxiv.org/abs/2602.23565 · **PDF:** https://arxiv.org/pdf/2602.23565v1
+
+## 9333. VaSST: Variational Inference for Symbolic Regression using Soft Symbolic Trees
+
+- **ID:** `2602.23561` · **primary:** `stat.ME` · **Submitted:** 2026-02-27
+- **Authors:** Somjit Roy, Pritam Dey, Bani K. Mallick
+- **abs:** https://arxiv.org/abs/2602.23561 · **PDF:** https://arxiv.org/pdf/2602.23561v1
+
+## 9334. Hierarchical Multi-Scale Graph Learning with Knowledge-Guided Attention for Whole-Slide Image Survival Analysis
+
+- **ID:** `2602.23557` · **primary:** `eess.IV` · **Submitted:** 2026-02-26
+- **Authors:** Bin Xu, Yufei Zhou, Boling Song, et al.
+- **abs:** https://arxiv.org/abs/2602.23557 · **PDF:** https://arxiv.org/pdf/2602.23557v2
+
+## 9335. The DMA Streaming Framework: Kernel-Level Buffer Orchestration for High-Performance AI Data Paths
+
+- **ID:** `2603.10030` · **primary:** `cs.AR` · **Submitted:** 2026-02-26
+- **Authors:** Marco Graziano
+- **abs:** https://arxiv.org/abs/2603.10030 · **PDF:** https://arxiv.org/pdf/2603.10030v2
+
+## 9336. Rudder: Steering Prefetching in Distributed GNN Training using LLM Agents
+
+- **ID:** `2602.23556` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Aishwarya Sarkar, Sayan Ghosh, Nathan Tallent, et al.
+- **abs:** https://arxiv.org/abs/2602.23556 · **PDF:** https://arxiv.org/pdf/2602.23556v1
+
+## 9337. RLShield: Practical Multi-Agent RL for Financial Cyber Defense with Attack-Surface MDPs and Real-Time Response Orchestration
+
+- **ID:** `2603.00186` · **primary:** `cs.CR` · **Submitted:** 2026-02-26
+- **Authors:** Srikumar Nayak
+- **abs:** https://arxiv.org/abs/2603.00186 · **PDF:** https://arxiv.org/pdf/2603.00186v1
+
+## 9338. ThreatFormer-IDS: Robust Transformer Intrusion Detection with Zero-Day Generalization and Explainable Attribution
+
+- **ID:** `2603.00185` · **primary:** `cs.CR` · **Submitted:** 2026-02-26
+- **Authors:** Srikumar Nayak
+- **abs:** https://arxiv.org/abs/2603.00185 · **PDF:** https://arxiv.org/pdf/2603.00185v1
+
+## 9339. Zero-Shot and Supervised Bird Image Segmentation Using Foundation Models: A Dual-Pipeline Approach with Grounding DINO~1.5, YOLOv11, and SAM~2.1
+
+- **ID:** `2603.00184` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Abhinav Munagala
+- **abs:** https://arxiv.org/abs/2603.00184 · **PDF:** https://arxiv.org/pdf/2603.00184v2
+
+## 9340. France or Spain or Germany or France: A Neural Account of Non-Redundant Redundant Disjunctions
+
+- **ID:** `2602.23547` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Sasha Boguraev, Qing Yao, Kyle Mahowald
+- **abs:** https://arxiv.org/abs/2602.23547 · **PDF:** https://arxiv.org/pdf/2602.23547v1
+
+## 9341. Probing to Refine: Reinforcement Distillation of LLMs via Explanatory Inversion
+
+- **ID:** `2603.19266` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Zhen Tan, Chengshuai Zhao, Song Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.19266 · **PDF:** https://arxiv.org/pdf/2603.19266v1
+
+## 9342. Humans and LLMs Diverge on Probabilistic Inferences
+
+- **ID:** `2602.23546` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Gaurav Kamath, Sreenath Madathil, Sebastian Schuster, et al.
+- **abs:** https://arxiv.org/abs/2602.23546 · **PDF:** https://arxiv.org/pdf/2602.23546v1
+
+## 9343. Planning under Distribution Shifts with Causal POMDPs
+
+- **ID:** `2602.23545` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Matteo Ceriscioli, Karthika Mohan
+- **abs:** https://arxiv.org/abs/2602.23545 · **PDF:** https://arxiv.org/pdf/2602.23545v1
+
+## 9344. When the Pure Reasoner Meets the Impossible Object: Analytic vs. Synthetic Fine-Tuning and the Suppression of Genesis in Language Models
+
+- **ID:** `2603.19265` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Amin Amouhadi
+- **abs:** https://arxiv.org/abs/2603.19265 · **PDF:** https://arxiv.org/pdf/2603.19265v1
+
+## 9345. Causal Identification from Counterfactual Data: Completeness and Bounding Results
+
+- **ID:** `2602.23541` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Arvind Raghavan, Elias Bareinboim
+- **abs:** https://arxiv.org/abs/2602.23541 · **PDF:** https://arxiv.org/pdf/2602.23541v2
+
+## 9346. Component Centric Placement Using Deep Reinforcement Learning
+
+- **ID:** `2602.23540` · **primary:** `cs.ET` · **Submitted:** 2026-02-26
+- **Authors:** Kart Leong Lim
+- **abs:** https://arxiv.org/abs/2602.23540 · **PDF:** https://arxiv.org/pdf/2602.23540v1
+
+## 9347. Partition Function Estimation under Bounded f-Divergence
+
+- **ID:** `2602.23535` · **primary:** `stat.ML` · **Submitted:** 2026-02-26
+- **Authors:** Adam Block, Abhishek Shetty
+- **abs:** https://arxiv.org/abs/2602.23535 · **PDF:** https://arxiv.org/pdf/2602.23535v1
+
+## 9348. Few-Shot Continual Learning for 3D Brain MRI with Frozen Foundation Models
+
+- **ID:** `2602.23533` · **primary:** `eess.IV` · **Submitted:** 2026-02-26
+- **Authors:** Chi-Sheng Chen, Xinyu Zhang, Guan-Ying Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.23533 · **PDF:** https://arxiv.org/pdf/2602.23533v1
+
+## 9349. Active Value Querying to Minimize Additive Error in Subadditive Set Function Learning
+
+- **ID:** `2602.23529` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Martin Černý, David Sychrovský, Filip Úradník, et al.
+- **abs:** https://arxiv.org/abs/2602.23529 · **PDF:** https://arxiv.org/pdf/2602.23529v2
+
+## 9350. Neural Operators Can Discover Functional Clusters
+
+- **ID:** `2602.23528` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Yicen Li, Jose Antonio Lara Benitez, Ruiyang Hong, et al.
+- **abs:** https://arxiv.org/abs/2602.23528 · **PDF:** https://arxiv.org/pdf/2602.23528v1
+
+## 9351. Test Case Prioritization: A Snowballing Literature Review and TCPFramework with Approach Combinators
+
+- **ID:** `2603.00183` · **primary:** `cs.SE` · **Submitted:** 2026-02-26
+- **Authors:** Tomasz Chojnacki, Lech Madeyski
+- **abs:** https://arxiv.org/abs/2603.00183 · **PDF:** https://arxiv.org/pdf/2603.00183v2
+
+## 9352. V-MORALS: Visual Morse Graph-Aided Estimation of Regions of Attraction in a Learned Latent Space
+
+- **ID:** `2602.23524` · **primary:** `cs.RO` · **Submitted:** 2026-02-26
+- **Authors:** Faiz Aladin, Ashwin Balasubramanian, Lars Lindemann, et al.
+- **abs:** https://arxiv.org/abs/2602.23524 · **PDF:** https://arxiv.org/pdf/2602.23524v2
+
+## 9353. Generative Active Testing: Efficient LLM Evaluation via Proxy Task Adaptation
+
+- **ID:** `2603.19264` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Aashish Anantha Ramakrishnan, Ardavan Saeedi, Hamid Reza Hassanzadeh, et al.
+- **abs:** https://arxiv.org/abs/2603.19264 · **PDF:** https://arxiv.org/pdf/2603.19264v1
+
+## 9354. Uncovering Physical Drivers of Dark Matter Halo Structures with Auxiliary-Variable-Guided Generative Models
+
+- **ID:** `2602.23518` · **primary:** `stat.ML` · **Submitted:** 2026-02-26
+- **Authors:** Arkaprabha Ganguli, Anirban Samaddar, Florian Kéruzoré, et al.
+- **abs:** https://arxiv.org/abs/2602.23518 · **PDF:** https://arxiv.org/pdf/2602.23518v1
+
+## 9355. Embedding Morphology into Transformers for Cross-Robot Policy Learning
+
+- **ID:** `2603.00182` · **primary:** `cs.RO` · **Submitted:** 2026-02-26
+- **Authors:** Kei Suzuki, Jing Liu, Ye Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.00182 · **PDF:** https://arxiv.org/pdf/2603.00182v1
+
+## 9356. Lap2: Revisiting Laplace DP-SGD for High Dimensions via Majorization Theory
+
+- **ID:** `2602.23516` · **primary:** `cs.CR` · **Submitted:** 2026-02-26
+- **Authors:** Meisam Mohammady, Qin Yang, Nicholas Stout, et al.
+- **abs:** https://arxiv.org/abs/2602.23516 · **PDF:** https://arxiv.org/pdf/2602.23516v3
+
+## 9357. Modelling and Simulation of Neuromorphic Datasets for Anomaly Detection in Computer Vision
+
+- **ID:** `2602.23514` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Mike Middleton, Teymoor Ali, Hakan Kayan, et al.
+- **abs:** https://arxiv.org/abs/2602.23514 · **PDF:** https://arxiv.org/pdf/2602.23514v1
+
+## 9358. How Motivation Relates to Generative AI Use: A Large-Scale Survey of Mexican High School Students
+
+- **ID:** `2603.19263` · **primary:** `cs.CY` · **Submitted:** 2026-02-26
+- **Authors:** Echo Zexuan Pan, Danny Glick, Ying Xu
+- **abs:** https://arxiv.org/abs/2603.19263 · **PDF:** https://arxiv.org/pdf/2603.19263v2
+
+## 9359. SegReg: Latent Space Regularization for Improved Medical Image Segmentation
+
+- **ID:** `2602.23509` · **primary:** `eess.IV` · **Submitted:** 2026-02-26
+- **Authors:** Puru Vaish, Amin Ranem, Felix Meister, et al.
+- **abs:** https://arxiv.org/abs/2602.23509 · **PDF:** https://arxiv.org/pdf/2602.23509v1
+
+## 9360. Sample Size Calculations for Developing Clinical Prediction Models: Overview and pmsims R package
+
+- **ID:** `2602.23507` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Diana Shamsutdinova, Felix Zimmer, Oyebayo Ridwan Olaniran, et al.
+- **abs:** https://arxiv.org/abs/2602.23507 · **PDF:** https://arxiv.org/pdf/2602.23507v1
+
+## 9361. FedDAG: Clustered Federated Learning via Global Data and Gradient Integration for Heterogeneous Environments
+
+- **ID:** `2602.23504` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Anik Pramanik, Murat Kantarcioglu, Vincent Oria, et al.
+- **abs:** https://arxiv.org/abs/2602.23504 · **PDF:** https://arxiv.org/pdf/2602.23504v1
+
+## 9362. Spiky Rank and Its Applications to Rigidity and Circuits
+
+- **ID:** `2602.23503` · **primary:** `cs.CC` · **Submitted:** 2026-02-26
+- **Authors:** Lianna Hambardzumyan, Konstantin Myasnikov, Artur Riazanov, et al.
+- **abs:** https://arxiv.org/abs/2602.23503 · **PDF:** https://arxiv.org/pdf/2602.23503v1
+
+## 9363. Engineering FAIR Privacy-preserving Applications that Learn Histories of Disease
+
+- **ID:** `2603.00181` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Ines N. Duarte, Praphulla M. S. Bhawsar, Lee K. Mason, et al.
+- **abs:** https://arxiv.org/abs/2603.00181 · **PDF:** https://arxiv.org/pdf/2603.00181v1
+
+## 9364. Quality Assessment of Public Summary of Training Content for GPAI models required by AI Act Article 53(1)(d)
+
+- **ID:** `2603.13270` · **primary:** `cs.CY` · **Submitted:** 2026-02-26
+- **Authors:** Dick A. H. Blankvoort, Harshvardhan J. Pandit, Maximilian Gahntz
+- **abs:** https://arxiv.org/abs/2603.13270 · **PDF:** https://arxiv.org/pdf/2603.13270v1
+
+## 9365. TaCarla: A comprehensive benchmarking dataset for end-to-end autonomous driving
+
+- **ID:** `2602.23499` · **primary:** `cs.RO` · **Submitted:** 2026-02-26
+- **Authors:** Tugrul Gorgulu, Atakan Dag, M. Esat Kalfaoglu, et al.
+- **abs:** https://arxiv.org/abs/2602.23499 · **PDF:** https://arxiv.org/pdf/2602.23499v3
+
+## 9366. Uncertainty-aware Language Guidance for Concept Bottleneck Models
+
+- **ID:** `2602.23495` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Yangyi Li, Mengdi Huai
+- **abs:** https://arxiv.org/abs/2602.23495 · **PDF:** https://arxiv.org/pdf/2602.23495v1
+
+## 9367. GenAI Integration into Engineering Education: A Case Study of an Introductory Undergraduate Engineering Course
+
+- **ID:** `2603.13269` · **primary:** `cs.CY` · **Submitted:** 2026-02-26
+- **Authors:** Kadir Kozan, Ozgur Keles, Sihan Jian, et al.
+- **abs:** https://arxiv.org/abs/2603.13269 · **PDF:** https://arxiv.org/pdf/2603.13269v1
+
+## 9368. The α-Law of Observable Belief Revision in Large Language Model Inference
+
+- **ID:** `2603.19262` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Mike Farmer, Abhinav Kochar, Yugyung Lee
+- **abs:** https://arxiv.org/abs/2603.19262 · **PDF:** https://arxiv.org/pdf/2603.19262v1
+
+## 9369. NNiT: Width-Agnostic Neural Network Generation with Structurally Aligned Weight Spaces
+
+- **ID:** `2603.00180` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Jiwoo Kim, Swarajh Mehta, Hao-Lun Hsu, et al.
+- **abs:** https://arxiv.org/abs/2603.00180 · **PDF:** https://arxiv.org/pdf/2603.00180v1
+
+## 9370. Significance-Gain Pair Encoding for LLMs: A Statistical Alternative to Frequency-Based Subword Merging
+
+- **ID:** `2603.19261` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Azam Nouri
+- **abs:** https://arxiv.org/abs/2603.19261 · **PDF:** https://arxiv.org/pdf/2603.19261v1
+
+## 9371. IDP Accelerator: Agentic Document Intelligence from Extraction to Compliance Validation
+
+- **ID:** `2602.23481` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Md Mofijul Islam, Md Sirajus Salekin, Joe King, et al.
+- **abs:** https://arxiv.org/abs/2602.23481 · **PDF:** https://arxiv.org/pdf/2602.23481v2
+
+## 9372. FHIRPath-QA: Executable Question Answering over FHIR Electronic Health Records
+
+- **ID:** `2602.23479` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Michael Frew, Nishit Bheda, Bryan Tripp
+- **abs:** https://arxiv.org/abs/2602.23479 · **PDF:** https://arxiv.org/pdf/2602.23479v2
+
+## 9373. Measuring Research Convergence in Interdisciplinary Teams Using Large Language Models and Graph Analytics
+
+- **ID:** `2603.20204` · **primary:** `cs.CY` · **Submitted:** 2026-02-26
+- **Authors:** Wenwen Li, Yuanyuan Tian, Sizhe Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.20204 · **PDF:** https://arxiv.org/pdf/2603.20204v1
+
+## 9374. Cross-Representation Knowledge Transfer for Improved Sequential Recommendations
+
+- **ID:** `2602.23471` · **primary:** `cs.IR` · **Submitted:** 2026-02-26
+- **Authors:** Artur Gimranov, Viacheslav Yusupov, Elfat Sabitov, et al.
+- **abs:** https://arxiv.org/abs/2602.23471 · **PDF:** https://arxiv.org/pdf/2602.23471v1
+
+## 9375. Detecting Cognitive Signatures in Typing Behavior for Non-Intrusive Authorship Verification
+
+- **ID:** `2603.00177` · **primary:** `cs.CR` · **Submitted:** 2026-02-26
+- **Authors:** David Condrey
+- **abs:** https://arxiv.org/abs/2603.00177 · **PDF:** https://arxiv.org/pdf/2603.00177v1
+
+## 9376. MPBMC: Multi-Property Bounded Model Checking with GNN-guided Clustering
+
+- **ID:** `2603.04450` · **primary:** `cs.LO` · **Submitted:** 2026-02-26
+- **Authors:** Soumik Guha Roy, Sumana Ghosh, Ansuman Banerjee, et al.
+- **abs:** https://arxiv.org/abs/2603.04450 · **PDF:** https://arxiv.org/pdf/2603.04450v1
+
+## 9377. Optimization of Edge Directions and Weights for Mixed Guidance Graphs in Lifelong Multi-Agent Path Finding
+
+- **ID:** `2602.23468` · **primary:** `cs.MA` · **Submitted:** 2026-02-26
+- **Authors:** Yulun Zhang, Varun Bhatt, Matthew C. Fontaine, et al.
+- **abs:** https://arxiv.org/abs/2602.23468 · **PDF:** https://arxiv.org/pdf/2602.23468v2
+
+## 9378. On the Limits of Interpretable Machine Learning in Quintic Root Classification
+
+- **ID:** `2602.23467` · **primary:** `math.NA` · **Submitted:** 2026-02-26
+- **Authors:** Rohan Thomas, Majid Bani-Yaghoub
+- **abs:** https://arxiv.org/abs/2602.23467 · **PDF:** https://arxiv.org/pdf/2602.23467v1
+
+## 9379. Neural ensemble Kalman filter: Data assimilation for compressible flows with shocks
+
+- **ID:** `2602.23461` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-02-26
+- **Authors:** Xu-Hui Zhou, Lorenzo Beronilla, Michael K. Sleeman, et al.
+- **abs:** https://arxiv.org/abs/2602.23461 · **PDF:** https://arxiv.org/pdf/2602.23461v1
+
+## 9380. Global Interpretability via Automated Preprocessing: A Framework Inspired by Psychiatric Questionnaires
+
+- **ID:** `2602.23459` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Eric V. Strobl
+- **abs:** https://arxiv.org/abs/2602.23459 · **PDF:** https://arxiv.org/pdf/2602.23459v1
+
+## 9381. BiKA: Kolmogorov-Arnold-Network-inspired Ultra Lightweight Neural Network Hardware Accelerator
+
+- **ID:** `2602.23455` · **primary:** `cs.AR` · **Submitted:** 2026-02-26
+- **Authors:** Yuhao Liu, Salim Ullah, Akash Kumar
+- **abs:** https://arxiv.org/abs/2602.23455 · **PDF:** https://arxiv.org/pdf/2602.23455v1
+
+## 9382. CiteAudit: You Cited It, But Did You Read It? A Benchmark for Verifying Scientific References in the LLM Era
+
+- **ID:** `2602.23452` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Zhengqing Yuan, Kaiwen Shi, Zheyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.23452 · **PDF:** https://arxiv.org/pdf/2602.23452v1
+
+## 9383. Bridging Policy and Real-World Dynamics: LLM-Augmented Rebalancing for Shared Micromobility Systems
+
+- **ID:** `2603.00176` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Heng Tan, Hua Yan, Yu Yang
+- **abs:** https://arxiv.org/abs/2603.00176 · **PDF:** https://arxiv.org/pdf/2603.00176v1
+
+## 9384. An Explainable Ensemble Framework for Alzheimer's Disease Prediction Using Structured Clinical and Cognitive Data
+
+- **ID:** `2603.04449` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Nishan Mitra
+- **abs:** https://arxiv.org/abs/2603.04449 · **PDF:** https://arxiv.org/pdf/2603.04449v1
+
+## 9385. SALIENT: Frequency-Aware Paired Diffusion for Controllable Long-Tail CT Detection
+
+- **ID:** `2602.23447` · **primary:** `eess.IV` · **Submitted:** 2026-02-26
+- **Authors:** Yifan Li, Mehrdad Salimitari, Taiyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.23447 · **PDF:** https://arxiv.org/pdf/2602.23447v1
+
+## 9386. Human Supervision as an Information Bottleneck: A Unified Theory of Error Floors in Human-Guided Learning
+
+- **ID:** `2602.23446` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Alejandro Rodriguez Dominguez
+- **abs:** https://arxiv.org/abs/2602.23446 · **PDF:** https://arxiv.org/pdf/2602.23446v1
+
+## 9387. Truncated Step-Level Sampling with Process Rewards for Retrieval-Augmented Reasoning
+
+- **ID:** `2602.23440` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Chris Samarinas, Haw-Shiuan Chang, Hamed Zamani
+- **abs:** https://arxiv.org/abs/2602.23440 · **PDF:** https://arxiv.org/pdf/2602.23440v3
+
+## 9388. DesignSense: A Human Preference Dataset and Reward Modeling Framework for Graphic Layout Generation
+
+- **ID:** `2602.23438` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Varun Gopal, Rishabh Jain, Aradhya Mathur, et al.
+- **abs:** https://arxiv.org/abs/2602.23438 · **PDF:** https://arxiv.org/pdf/2602.23438v1
+
+## 9389. Model Agreement via Anchoring
+
+- **ID:** `2602.23360` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Eric Eaton, Surbhi Goel, Marcel Hussing, et al.
+- **abs:** https://arxiv.org/abs/2602.23360 · **PDF:** https://arxiv.org/pdf/2602.23360v1
+
+## 9390. SeeThrough3D: Occlusion Aware 3D Control in Text-to-Image Generation
+
+- **ID:** `2602.23359` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Vaibhav Agrawal, Rishubh Parihar, Pradhaan Bhat, et al.
+- **abs:** https://arxiv.org/abs/2602.23359 · **PDF:** https://arxiv.org/pdf/2602.23359v1
+
+## 9391. A Dataset is Worth 1 MB
+
+- **ID:** `2602.23358` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Elad Kimchi Shoshani, Leeyam Gabay, Yedid Hoshen
+- **abs:** https://arxiv.org/abs/2602.23358 · **PDF:** https://arxiv.org/pdf/2602.23358v1
+
+## 9392. SOTAlign: Semi-Supervised Alignment of Unimodal Vision and Language Models via Optimal Transport
+
+- **ID:** `2602.23353` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Simon Roschmann, Paul Krzakala, Sonia Mazelet, et al.
+- **abs:** https://arxiv.org/abs/2602.23353 · **PDF:** https://arxiv.org/pdf/2602.23353v1
+
+## 9393. EvoX: Meta-Evolution for Automated Discovery
+
+- **ID:** `2602.23413` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Shu Liu, Shubham Agarwal, Monishwaran Maheswaran, et al.
+- **abs:** https://arxiv.org/abs/2602.23413 · **PDF:** https://arxiv.org/pdf/2602.23413v2
+
+## 9394. Scale Can't Overcome Pragmatics: The Impact of Reporting Bias on Vision-Language Reasoning
+
+- **ID:** `2602.23351` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Amita Kamath, Jack Hessel, Khyathi Chandu, et al.
+- **abs:** https://arxiv.org/abs/2602.23351 · **PDF:** https://arxiv.org/pdf/2602.23351v1
+
+## 9395. Perfect score on IPhO 2025 theory by Gemini agent
+
+- **ID:** `2603.03352` · **primary:** `physics.ed-ph` · **Submitted:** 2026-02-26
+- **Authors:** Yichen Huang
+- **abs:** https://arxiv.org/abs/2603.03352 · **PDF:** https://arxiv.org/pdf/2603.03352v1
+
+## 9396. FlashOptim: Optimizers for Memory-Efficient Training
+
+- **ID:** `2602.23349` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Jose Javier Gonzalez Ortiz, Abhay Gupta, Christopher Rinard, et al.
+- **abs:** https://arxiv.org/abs/2602.23349 · **PDF:** https://arxiv.org/pdf/2602.23349v2
+
+## 9397. Mean Estimation from Coarse Data: Characterizations and Efficient Algorithms
+
+- **ID:** `2602.23341` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Alkis Kalavasis, Anay Mehrotra, Manolis Zampetakis, et al.
+- **abs:** https://arxiv.org/abs/2602.23341 · **PDF:** https://arxiv.org/pdf/2602.23341v1
+
+## 9398. Differentiable Zero-One Loss via Hypersimplex Projections
+
+- **ID:** `2602.23336` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Camilo Gomez, Pengyang Wang, Liansheng Tang
+- **abs:** https://arxiv.org/abs/2602.23336 · **PDF:** https://arxiv.org/pdf/2602.23336v1
+
+## 9399. Understanding Usage and Engagement in AI-Powered Scientific Research Tools: The Asta Interaction Dataset
+
+- **ID:** `2602.23335` · **primary:** `cs.HC` · **Submitted:** 2026-02-26
+- **Authors:** Dany Haddad, Dan Bareket, Joseph Chee Chang, et al.
+- **abs:** https://arxiv.org/abs/2602.23335 · **PDF:** https://arxiv.org/pdf/2602.23335v1
+
+## 9400. Bitwise Systolic Array Architecture for Runtime-Reconfigurable Multi-precision Quantized Multiplication on Hardware Accelerators
+
+- **ID:** `2602.23334` · **primary:** `cs.AR` · **Submitted:** 2026-02-26
+- **Authors:** Yuhao Liu, Salim Ullah, Akash Kumar
+- **abs:** https://arxiv.org/abs/2602.23334 · **PDF:** https://arxiv.org/pdf/2602.23334v1
+
+## 9401. Utilizing LLMs for Industrial Process Automation
+
+- **ID:** `2602.23331` · **primary:** `cs.SE` · **Submitted:** 2026-02-26
+- **Authors:** Salim Fares
+- **abs:** https://arxiv.org/abs/2602.23331 · **PDF:** https://arxiv.org/pdf/2602.23331v1
+
+## 9402. Toward Expert Investment Teams:A Multi-Agent LLM System with Fine-Grained Trading Tasks
+
+- **ID:** `2602.23330` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Kunihiro Miyazaki, Takanobu Kawahara, Stephen Roberts, et al.
+- **abs:** https://arxiv.org/abs/2602.23330 · **PDF:** https://arxiv.org/pdf/2602.23330v1
+
+## 9403. LLM Novice Uplift on Dual-Use, In Silico Biology Tasks
+
+- **ID:** `2602.23329` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Chen Bo Calvin Zhang, Christina Q. Knight, Nicholas Kruus, et al.
+- **abs:** https://arxiv.org/abs/2602.23329 · **PDF:** https://arxiv.org/pdf/2602.23329v2
+
+## 9404. Deep ensemble graph neural networks for probabilistic cosmic-ray direction and energy reconstruction in autonomous radio arrays
+
+- **ID:** `2602.23321` · **primary:** `astro-ph.IM` · **Submitted:** 2026-02-26
+- **Authors:** Arsène Ferrière, Aurélien Benoit-Lévy, Olivier Martineau-Huynh, et al.
+- **abs:** https://arxiv.org/abs/2602.23321 · **PDF:** https://arxiv.org/pdf/2602.23321v1
+
+## 9405. ParamMem: Augmenting Language Agents with Parametric Reflective Memory
+
+- **ID:** `2602.23320` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Tianjun Yao, Yongqiang Chen, Yujia Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.23320 · **PDF:** https://arxiv.org/pdf/2602.23320v2
+
+## 9406. Generalized Rapid Action Value Estimation in Memory-Constrained Environments
+
+- **ID:** `2602.23318` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Aloïs Rautureau, Tristan Cazenave, Éric Piette
+- **abs:** https://arxiv.org/abs/2602.23318 · **PDF:** https://arxiv.org/pdf/2602.23318v1
+
+## 9407. Invariant Transformation and Resampling based Epistemic-Uncertainty Reduction
+
+- **ID:** `2602.23315` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Sha Hu
+- **abs:** https://arxiv.org/abs/2602.23315 · **PDF:** https://arxiv.org/pdf/2602.23315v1
+
+## 9408. Evaluating Zero-Shot and One-Shot Adaptation of Small Language Models in Leader-Follower Interaction
+
+- **ID:** `2602.23312` · **primary:** `cs.HC` · **Submitted:** 2026-02-26
+- **Authors:** Rafael R. Baptista, André de Lima Salgado, Ricardo V. Godoy, et al.
+- **abs:** https://arxiv.org/abs/2602.23312 · **PDF:** https://arxiv.org/pdf/2602.23312v2
+
+## 9409. A Proper Scoring Rule for Virtual Staining
+
+- **ID:** `2602.23305` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Samuel Tonks, Steve Hood, Ryan Musso, et al.
+- **abs:** https://arxiv.org/abs/2602.23305 · **PDF:** https://arxiv.org/pdf/2602.23305v1
+
+## 9410. Inferential Mechanics Part 1: Causal Mechanistic Theories of Machine Learning in Chemical Biology with Implications
+
+- **ID:** `2602.23303` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Ilya Balabin, Thomas M. Kaiser
+- **abs:** https://arxiv.org/abs/2602.23303 · **PDF:** https://arxiv.org/pdf/2602.23303v1
+
+## 9411. The logic of KM belief update is contained in the logic of AGM belief revision
+
+- **ID:** `2602.23302` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Giacomo Bonanno
+- **abs:** https://arxiv.org/abs/2602.23302 · **PDF:** https://arxiv.org/pdf/2602.23302v2
+
+## 9412. A Mixture-of-Experts Model for Multimodal Emotion Recognition in Conversations
+
+- **ID:** `2602.23300` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Soumya Dutta, Smruthi Balaji, Sriram Ganapathy
+- **abs:** https://arxiv.org/abs/2602.23300 · **PDF:** https://arxiv.org/pdf/2602.23300v1
+
+## 9413. Conformalized Neural Networks for Federated Uncertainty Quantification under Dual Heterogeneity
+
+- **ID:** `2602.23296` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Quang-Huy Nguyen, Jiaqi Wang, Wei-Shinn Ku
+- **abs:** https://arxiv.org/abs/2602.23296 · **PDF:** https://arxiv.org/pdf/2602.23296v2
+
+## 9414. ManifoldGD: Training-Free Hierarchical Manifold Guidance for Diffusion-Based Dataset Distillation
+
+- **ID:** `2602.23295` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Ayush Roy, Wei-Yang Alex Lee, Rudrasis Chakraborty, et al.
+- **abs:** https://arxiv.org/abs/2602.23295 · **PDF:** https://arxiv.org/pdf/2602.23295v1
+
+## 9415. Automated Discovery of Improved Constant Weight Binary Codes
+
+- **ID:** `2603.00174` · **primary:** `cs.IT` · **Submitted:** 2026-02-26
+- **Authors:** Christopher D. Rosin
+- **abs:** https://arxiv.org/abs/2603.00174 · **PDF:** https://arxiv.org/pdf/2603.00174v1
+
+## 9416. SPARTA: Scalable and Principled Benchmark of Tree-Structured Multi-hop QA over Text and Tables
+
+- **ID:** `2602.23286` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Sungho Park, Jueun Kim, Wook-Shin Han
+- **abs:** https://arxiv.org/abs/2602.23286 · **PDF:** https://arxiv.org/pdf/2602.23286v1
+
+## 9417. ODEBrain: Continuous-Time EEG Graph for Modeling Dynamic Brain Networks
+
+- **ID:** `2602.23285` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Haohui Jia, Zheng Chen, Lingwei Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.23285 · **PDF:** https://arxiv.org/pdf/2602.23285v2
+
+## 9418. Physics Informed Viscous Value Representations
+
+- **ID:** `2602.23280` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Hrishikesh Viswanath, Juanwu Lu, S. Talha Bukhari, et al.
+- **abs:** https://arxiv.org/abs/2602.23280 · **PDF:** https://arxiv.org/pdf/2602.23280v1
+
+## 9419. Zeroth-Order Stackelberg Control in Combinatorial Congestion Games
+
+- **ID:** `2602.23277` · **primary:** `cs.GT` · **Submitted:** 2026-02-26
+- **Authors:** Saeed Masiha, Sepehr Elahi, Negar Kiyavash, et al.
+- **abs:** https://arxiv.org/abs/2602.23277 · **PDF:** https://arxiv.org/pdf/2602.23277v1
+
+## 9420. CXReasonAgent: Evidence-Grounded Diagnostic Reasoning Agent for Chest X-rays
+
+- **ID:** `2602.23276` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Hyungyung Lee, Hangyul Yoon, Edward Choi
+- **abs:** https://arxiv.org/abs/2602.23276 · **PDF:** https://arxiv.org/pdf/2602.23276v2
+
+## 9421. Evaluating Stochasticity in Deep Research Agents
+
+- **ID:** `2602.23271` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Haotian Zhai, Elias Stengel-Eskin, Pratik Patil, et al.
+- **abs:** https://arxiv.org/abs/2602.23271 · **PDF:** https://arxiv.org/pdf/2602.23271v1
+
+## 9422. Discourse-Aware Dual-Track Streaming Response for Low-Latency Spoken Dialogue Systems
+
+- **ID:** `2602.23266` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Siyuan Liu, Jiahui Xu, Feng Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.23266 · **PDF:** https://arxiv.org/pdf/2602.23266v1
+
+## 9423. Risk-Aware World Model Predictive Control for Generalizable End-to-End Autonomous Driving
+
+- **ID:** `2602.23259` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Jiangxin Sun, Feng Xue, Teng Long, et al.
+- **abs:** https://arxiv.org/abs/2602.23259 · **PDF:** https://arxiv.org/pdf/2602.23259v1
+
+## 9424. AgentDropoutV2: Optimizing Information Flow in Multi-Agent Systems via Test-Time Rectify-or-Reject Pruning
+
+- **ID:** `2602.23258` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Yutong Wang, Siyuan Xiong, Xuebo Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.23258 · **PDF:** https://arxiv.org/pdf/2602.23258v1
+
+## 9425. Mitigating Legibility Tax with Decoupled Prover-Verifier Games
+
+- **ID:** `2602.23248` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Yegon Kim, Juho Lee
+- **abs:** https://arxiv.org/abs/2602.23248 · **PDF:** https://arxiv.org/pdf/2602.23248v1
+
+## 9426. HATL: Hierarchical Adaptive-Transfer Learning Framework for Sign Language Machine Translation
+
+- **ID:** `2603.19260` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Nada Shahin, Leila Ismail
+- **abs:** https://arxiv.org/abs/2603.19260 · **PDF:** https://arxiv.org/pdf/2603.19260v1
+
+## 9427. A Model-Free Universal AI
+
+- **ID:** `2602.23242` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Yegon Kim, Juho Lee
+- **abs:** https://arxiv.org/abs/2602.23242 · **PDF:** https://arxiv.org/pdf/2602.23242v1
+
+## 9428. Agency and Architectural Limits: Why Optimization-Based Systems Cannot Be Norm-Responsive
+
+- **ID:** `2602.23239` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Radha Sarma
+- **abs:** https://arxiv.org/abs/2602.23239 · **PDF:** https://arxiv.org/pdf/2602.23239v2
+
+## 9429. Spatio-Temporal Token Pruning for Efficient High-Resolution GUI Agents
+
+- **ID:** `2602.23235` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Zhou Xu, Bowen Zhou, Qi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.23235 · **PDF:** https://arxiv.org/pdf/2602.23235v1
+
+## 9430. Scaling Search Relevance: Augmenting App Store Ranking with LLM-Generated Judgments
+
+- **ID:** `2602.23234` · **primary:** `cs.IR` · **Submitted:** 2026-02-26
+- **Authors:** Evangelia Christakopoulou, Vivekkumar Patel, Hemanth Velaga, et al.
+- **abs:** https://arxiv.org/abs/2602.23234 · **PDF:** https://arxiv.org/pdf/2602.23234v3
+
+## 9431. ReCoN-Ipsundrum: An Inspectable Recurrent Persistence Loop Agent with Affect-Coupled Control and Mechanism-Linked Consciousness Indicator Assays
+
+- **ID:** `2602.23232` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Aishik Sanyal
+- **abs:** https://arxiv.org/abs/2602.23232 · **PDF:** https://arxiv.org/pdf/2602.23232v2
+
+## 9432. MovieTeller: Tool-augmented Movie Synopsis with ID Consistent Progressive Abstraction
+
+- **ID:** `2602.23228` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Yizhi Li, Xiaohan Chen, Miao Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.23228 · **PDF:** https://arxiv.org/pdf/2602.23228v2
+
+## 9433. Why Diffusion Language Models Struggle with Truly Parallel (Non-Autoregressive) Decoding?
+
+- **ID:** `2602.23225` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Pengxiang Li, Dilxat Muhtar, Tianlong Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.23225 · **PDF:** https://arxiv.org/pdf/2602.23225v2
+
+## 9434. Takeuchi's Information Criteria as Generalization Measures for DNNs Close to NTK Regime
+
+- **ID:** `2602.23219` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Hiroki Naganuma, Taiji Suzuki, Rio Yokota, et al.
+- **abs:** https://arxiv.org/abs/2602.23219 · **PDF:** https://arxiv.org/pdf/2602.23219v1
+
+## 9435. Plug-and-Play Diffusion Meets ADMM: Dual-Variable Coupling for Robust Medical Image Reconstruction
+
+- **ID:** `2602.23214` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Chenhe Du, Xuanyu Tian, Qing Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.23214 · **PDF:** https://arxiv.org/pdf/2602.23214v1
+
+## 9436. ColoDiff: Integrating Dynamic Consistency With Content Awareness for Colonoscopy Video Generation
+
+- **ID:** `2602.23203` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Junhu Fu, Shuyu Liang, Wutong Li, et al.
+- **abs:** https://arxiv.org/abs/2602.23203 · **PDF:** https://arxiv.org/pdf/2602.23203v1
+
+## 9437. Tell Me What To Learn: Generalizing Neural Memory to be Controllable in Natural Language
+
+- **ID:** `2602.23201` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Max S. Bennett, Thomas P. Zollo, Richard Zemel
+- **abs:** https://arxiv.org/abs/2602.23201 · **PDF:** https://arxiv.org/pdf/2602.23201v2
+
+## 9438. InnerQ: Hardware-aware Tuning-free Quantization of KV Cache for Large Language Models
+
+- **ID:** `2602.23200` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Sayed Mohammadreza Tayaranian Hosseini, Amir Ardakani, Warren J. Gross
+- **abs:** https://arxiv.org/abs/2602.23200 · **PDF:** https://arxiv.org/pdf/2602.23200v1
+
+## 9439. SC-Arena: A Natural Language Benchmark for Single-Cell Reasoning with Knowledge-Augmented Evaluation
+
+- **ID:** `2602.23199` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Jiahao Zhao, Feng Jiang, Shaowei Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.23199 · **PDF:** https://arxiv.org/pdf/2602.23199v1
+
+## 9440. Fine-Tuning Without Forgetting In-Context Learning: A Theoretical Analysis of Linear Attention Models
+
+- **ID:** `2602.23197` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Chungpa Lee, Jy-yong Sohn, Kangwook Lee
+- **abs:** https://arxiv.org/abs/2602.23197 · **PDF:** https://arxiv.org/pdf/2602.23197v1
+
+## 9441. ESAA: Event Sourcing for Autonomous Agents in LLM-Based Software Engineering
+
+- **ID:** `2602.23193` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Elzo Brito dos Santos Filho
+- **abs:** https://arxiv.org/abs/2602.23193 · **PDF:** https://arxiv.org/pdf/2602.23193v1
+
+## 9442. FairQuant: Fairness-Aware Mixed-Precision Quantization for Medical Image Classification
+
+- **ID:** `2602.23192` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Thomas Woergaard, Raghavendra Selvan
+- **abs:** https://arxiv.org/abs/2602.23192 · **PDF:** https://arxiv.org/pdf/2602.23192v1
+
+## 9443. Efficient Real-Time Adaptation of ROMs for Unsteady Flows Using Data Assimilation
+
+- **ID:** `2602.23188` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Ismaël Zighed, Andrea Nóvoa, Luca Magri, et al.
+- **abs:** https://arxiv.org/abs/2602.23188 · **PDF:** https://arxiv.org/pdf/2602.23188v1
+
+## 9444. MTRAG-UN: A Benchmark for Open Challenges in Multi-Turn RAG Conversations
+
+- **ID:** `2602.23184` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Sara Rosenthal, Yannis Katsis, Vraj Shah, et al.
+- **abs:** https://arxiv.org/abs/2602.23184 · **PDF:** https://arxiv.org/pdf/2602.23184v1
+
+## 9445. Closing the gap on tabular data with Fourier and Implicit Categorical Features
+
+- **ID:** `2602.23182` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Marius Dragoi, Florin Gogianu, Elena Burceanu
+- **abs:** https://arxiv.org/abs/2602.23182 · **PDF:** https://arxiv.org/pdf/2602.23182v1
+
+## 9446. Induction Meets Biology: Mechanisms of Repeat Detection in Protein Language Models
+
+- **ID:** `2602.23179` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Gal Kesten-Pomeranz, Yaniv Nikankin, Anja Reusch, et al.
+- **abs:** https://arxiv.org/abs/2602.23179 · **PDF:** https://arxiv.org/pdf/2602.23179v2
+
+## 9447. Breeze Taigi: Benchmarks and Models for Taiwanese Hokkien Speech Recognition and Synthesis
+
+- **ID:** `2603.19259` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Yu-Siang Lan, Chia-Sheng Liu, Yi-Chang Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.19259 · **PDF:** https://arxiv.org/pdf/2603.19259v1
+
+## 9448. Latent Gaussian Splatting for 4D Panoptic Occupancy Tracking
+
+- **ID:** `2602.23172` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Maximilian Luz, Rohit Mohan, Thomas Nürnberg, et al.
+- **abs:** https://arxiv.org/abs/2602.23172 · **PDF:** https://arxiv.org/pdf/2602.23172v1
+
+## 9449. Summer-22B: A Systematic Approach to Dataset Engineering and Training at Scale for Video Foundation Model
+
+- **ID:** `2603.00173` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Simo Ryu, Chunghwan Han
+- **abs:** https://arxiv.org/abs/2603.00173 · **PDF:** https://arxiv.org/pdf/2603.00173v1
+
+## 9450. SettleFL: Trustless and Scalable Reward Settlement Protocol for Federated Learning on Permissionless Blockchains (Extended version)
+
+- **ID:** `2602.23167` · **primary:** `cs.CR` · **Submitted:** 2026-02-26
+- **Authors:** Shuang Liang, Yang Hua, Linshan Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.23167 · **PDF:** https://arxiv.org/pdf/2602.23167v1
+
+## 9451. MetaOthello: A Controlled Study of Multiple World Models in Transformers
+
+- **ID:** `2602.23164` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Aviral Chawla, Galen Hall, Juniper Lovato
+- **abs:** https://arxiv.org/abs/2602.23164 · **PDF:** https://arxiv.org/pdf/2602.23164v1
+
+## 9452. A Decision-Theoretic Formalisation of Steganography With Applications to LLM Monitoring
+
+- **ID:** `2602.23163` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Usman Anwar, Julianna Piskorz, David D. Baek, et al.
+- **abs:** https://arxiv.org/abs/2602.23163 · **PDF:** https://arxiv.org/pdf/2602.23163v2
+
+## 9453. PATRA: Pattern-Aware Alignment and Balanced Reasoning for Time Series Question Answering
+
+- **ID:** `2602.23161` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Junkai Lu, Peng Chen, Xingjian Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.23161 · **PDF:** https://arxiv.org/pdf/2602.23161v1
+
+## 9454. Benchmarking Temporal Web3 Intelligence: Lessons from the FinSurvival 2025 Challenge
+
+- **ID:** `2602.23159` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Oshani Seneviratne, Fernando Spadea, Adrien Pavao, et al.
+- **abs:** https://arxiv.org/abs/2602.23159 · **PDF:** https://arxiv.org/pdf/2602.23159v1
+
+## 9455. Efficient Encoder-Free Fourier-based 3D Large Multimodal Model
+
+- **ID:** `2602.23153` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Guofeng Mei, Wei Lin, Luigi Riz, et al.
+- **abs:** https://arxiv.org/abs/2602.23153 · **PDF:** https://arxiv.org/pdf/2602.23153v2
+
+## 9456. The Trinity of Consistency as a Defining Principle for General World Models
+
+- **ID:** `2602.23152` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Jingxuan Wei, Siyuan Li, Yuhang Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.23152 · **PDF:** https://arxiv.org/pdf/2602.23152v1
+
+## 9457. On Sample-Efficient Generalized Planning via Learned Transition Models
+
+- **ID:** `2602.23148` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Nitin Gupta, Vishal Pallagani, John A. Aydin, et al.
+- **abs:** https://arxiv.org/abs/2602.23148 · **PDF:** https://arxiv.org/pdf/2602.23148v3
+
+## 9458. Partial recovery of meter-scale surface weather
+
+- **ID:** `2602.23146` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Jonathan Giezendanner, Qidong Yang, Eric Schmitt, et al.
+- **abs:** https://arxiv.org/abs/2602.23146 · **PDF:** https://arxiv.org/pdf/2602.23146v1
+
+## 9459. Prediction of Diffusion Coefficients in Mixtures with Tensor Completion
+
+- **ID:** `2602.23142` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Zeno Romero, Kerstin Münnemann, Hans Hasse, et al.
+- **abs:** https://arxiv.org/abs/2602.23142 · **PDF:** https://arxiv.org/pdf/2602.23142v1
+
+## 9460. Hidden in the Metadata: Stealth Poisoning Attacks on Multimodal Retrieval-Augmented Generation
+
+- **ID:** `2603.00172` · **primary:** `cs.CR` · **Submitted:** 2026-02-26
+- **Authors:** Kennedy Edemacu, Mohammad Mahdi Shokri
+- **abs:** https://arxiv.org/abs/2603.00172 · **PDF:** https://arxiv.org/pdf/2603.00172v1
+
+## 9461. Quantum Deep Learning: A Comprehensive Review
+
+- **ID:** `2603.06644` · **primary:** `quant-ph` · **Submitted:** 2026-02-26
+- **Authors:** Yanjun Ji, Zhao-Yun Chen, Marco Roth, et al.
+- **abs:** https://arxiv.org/abs/2603.06644 · **PDF:** https://arxiv.org/pdf/2603.06644v1
+
+## 9462. Modality Collapse as Mismatched Decoding: Information-Theoretic Limits of Multimodal LLMs
+
+- **ID:** `2602.23136` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Jayadev Billa
+- **abs:** https://arxiv.org/abs/2602.23136 · **PDF:** https://arxiv.org/pdf/2602.23136v2
+
+## 9463. DyGnROLE: Modeling Asymmetry in Dynamic Graphs with Node-Role-Oriented Latent Encoding
+
+- **ID:** `2602.23135` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Tyler Bonnet, Marek Rei
+- **abs:** https://arxiv.org/abs/2602.23135 · **PDF:** https://arxiv.org/pdf/2602.23135v1
+
+## 9464. From Agnostic to Specific: Latent Preference Diffusion for Multi-Behavior Sequential Recommendation
+
+- **ID:** `2602.23132` · **primary:** `cs.IR` · **Submitted:** 2026-02-26
+- **Authors:** Ruochen Yang, Xiaodong Li, Jiawei Sheng, et al.
+- **abs:** https://arxiv.org/abs/2602.23132 · **PDF:** https://arxiv.org/pdf/2602.23132v1
+
+## 9465. Brain-OF: An Omnifunctional Foundation Model for fMRI, EEG and MEG
+
+- **ID:** `2602.23410` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Hanning Guo, Farah Abdellatif, Hanwen Bi, et al.
+- **abs:** https://arxiv.org/abs/2602.23410 · **PDF:** https://arxiv.org/pdf/2602.23410v2
+
+## 9466. Bound to Disagree: Generalization Bounds via Certifiable Surrogates
+
+- **ID:** `2602.23128` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Mathieu Bazinet, Valentina Zantedeschi, Pascal Germain
+- **abs:** https://arxiv.org/abs/2602.23128 · **PDF:** https://arxiv.org/pdf/2602.23128v1
+
+## 9467. SvfEye: A Semantic-Visual Fusion Framework with Multi-Scale Visual Context for Multimodal Reasoning
+
+- **ID:** `2603.00171` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Yuxiang Shen, Hailong Huang, Zhenkun Gao, et al.
+- **abs:** https://arxiv.org/abs/2603.00171 · **PDF:** https://arxiv.org/pdf/2603.00171v2
+
+## 9468. Multi-Agent Large Language Model Based Emotional Detoxification Through Personalized Intensity Control for Consumer Protection
+
+- **ID:** `2602.23123` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Keito Inoshita
+- **abs:** https://arxiv.org/abs/2602.23123 · **PDF:** https://arxiv.org/pdf/2602.23123v1
+
+## 9469. Automated Vulnerability Detection in Source Code Using Deep Representation Learning
+
+- **ID:** `2602.23121` · **primary:** `cs.CR` · **Submitted:** 2026-02-26
+- **Authors:** C. Seas, G. Fitzpatrick, J. A. Hamilton, et al.
+- **abs:** https://arxiv.org/abs/2602.23121 · **PDF:** https://arxiv.org/pdf/2602.23121v1
+
+## 9470. Devling into Adversarial Transferability on Image Classification: Review, Benchmark, and Evaluation
+
+- **ID:** `2602.23117` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Xiaosen Wang, Zhijin Ge, Bohan Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.23117 · **PDF:** https://arxiv.org/pdf/2602.23117v1
+
+## 9471. Regularized Online RLHF with Generalized Bilinear Preferences
+
+- **ID:** `2602.23116` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Junghyun Lee, Minju Hong, Kwang-Sung Jun, et al.
+- **abs:** https://arxiv.org/abs/2602.23116 · **PDF:** https://arxiv.org/pdf/2602.23116v2
+
+## 9472. Learning Physical Operators using Neural Operators
+
+- **ID:** `2602.23113` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Vignesh Gopakumar, Ander Gray, Dan Giles, et al.
+- **abs:** https://arxiv.org/abs/2602.23113 · **PDF:** https://arxiv.org/pdf/2602.23113v1
+
+## 9473. PRAC: Principal-Random Subspace for LLM Activation Compression and Memory-Efficient Training
+
+- **ID:** `2602.23111` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Yanyi Li, Yimu Zhang, Cong Fang
+- **abs:** https://arxiv.org/abs/2602.23111 · **PDF:** https://arxiv.org/pdf/2602.23111v1
+
+## 9474. Three AI-agents walk into a bar . . . . `Lord of the Flies' tribalism emerges among smart AI-Agents
+
+- **ID:** `2602.23093` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Dhwanil M. Mori, Neil F. Johnson
+- **abs:** https://arxiv.org/abs/2602.23093 · **PDF:** https://arxiv.org/pdf/2602.23093v1
+
+## 9475. Enhancing CVRP Solver through LLM-driven Automatic Heuristic Design
+
+- **ID:** `2602.23092` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Zhuoliang Xie, Fei Liu, Zhenkun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.23092 · **PDF:** https://arxiv.org/pdf/2602.23092v1
+
+## 9476. Physics-informed neural particle flow for the Bayesian update step
+
+- **ID:** `2602.23089` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Domonkos Csuzdi, Tamás Bécsi, Olivér Törő
+- **abs:** https://arxiv.org/abs/2602.23089 · **PDF:** https://arxiv.org/pdf/2602.23089v1
+
+## 9477. Q-Tag: Watermarking Quantum Circuit Generative Models
+
+- **ID:** `2602.23085` · **primary:** `quant-ph` · **Submitted:** 2026-02-26
+- **Authors:** Yang Yang, Yuzhu Long, Han Fang, et al.
+- **abs:** https://arxiv.org/abs/2602.23085 · **PDF:** https://arxiv.org/pdf/2602.23085v1
+
+## 9478. Assessing Deanonymization Risks with Stylometry-Assisted LLM Agent
+
+- **ID:** `2602.23079` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Boyang Zhang, Yang Zhang
+- **abs:** https://arxiv.org/abs/2602.23079 · **PDF:** https://arxiv.org/pdf/2602.23079v1
+
+## 9479. CiteLLM: An Agentic Platform for Trustworthy Scientific Reference Discovery
+
+- **ID:** `2602.23075` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Mengze Hong, Di Jiang, Chen Jason Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.23075 · **PDF:** https://arxiv.org/pdf/2602.23075v1
+
+## 9480. Accelerated Online Risk-Averse Policy Evaluation in POMDPs with Theoretical Guarantees and Novel CVaR Bounds
+
+- **ID:** `2602.23073` · **primary:** `math.ST` · **Submitted:** 2026-02-26
+- **Authors:** Yaacov Pariente, Vadim Indelman
+- **abs:** https://arxiv.org/abs/2602.23073 · **PDF:** https://arxiv.org/pdf/2602.23073v1
+
+## 9481. Quantity Convergence, Quality Divergence: Disentangling Fluency and Accuracy in L2 Mandarin Prosody
+
+- **ID:** `2602.23071` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Yuqi Shi, Hao Yang, Xiyao Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.23071 · **PDF:** https://arxiv.org/pdf/2602.23071v1
+
+## 9482. Make It Hard to Hear, Easy to Learn: Long-Form Bengali ASR and Speaker Diarization via Extreme Augmentation and Perfect Alignment
+
+- **ID:** `2602.23070` · **primary:** `cs.SD` · **Submitted:** 2026-02-26
+- **Authors:** Sanjid Hasan, Risalat Labib, A H M Fuad, et al.
+- **abs:** https://arxiv.org/abs/2602.23070 · **PDF:** https://arxiv.org/pdf/2602.23070v1
+
+## 9483. Long Range Frequency Tuning for QML
+
+- **ID:** `2602.23409` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Michael Poppel, Jonas Stein, Sebastian Wölckert, et al.
+- **abs:** https://arxiv.org/abs/2602.23409 · **PDF:** https://arxiv.org/pdf/2602.23409v1
+
+## 9484. Toward Automatic Filling of Case Report Forms: A Case Study on Data from an Italian Emergency Department
+
+- **ID:** `2602.23062` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Gabriela Anna Kaczmarek, Pietro Ferrazzi, Lorenzo Porta, et al.
+- **abs:** https://arxiv.org/abs/2602.23062 · **PDF:** https://arxiv.org/pdf/2602.23062v1
+
+## 9485. MoDora: Tree-Based Semi-Structured Document Analysis System
+
+- **ID:** `2602.23061` · **primary:** `cs.IR` · **Submitted:** 2026-02-26
+- **Authors:** Bangrui Xu, Qihang Yao, Zirui Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.23061 · **PDF:** https://arxiv.org/pdf/2602.23061v2
+
+## 9486. RhythmBERT: A Self-Supervised Language Model Based on Latent Representations of ECG Waveforms for Heart Disease Detection
+
+- **ID:** `2602.23060` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Xin Wang, Burcu Ozek, Aruna Mohan, et al.
+- **abs:** https://arxiv.org/abs/2602.23060 · **PDF:** https://arxiv.org/pdf/2602.23060v1
+
+## 9487. Affine-Scaled Attention: Towards Flexible and Stable Transformer Attention
+
+- **ID:** `2602.23057` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Jeongin Bae, Baeseong Park, Gunho Park, et al.
+- **abs:** https://arxiv.org/abs/2602.23057 · **PDF:** https://arxiv.org/pdf/2602.23057v1
+
+## 9488. Learning-based Multi-agent Race Strategies in Formula 1
+
+- **ID:** `2602.23056` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Giona Fieni, Joschua Wüthrich, Marc-Philippe Neumann, et al.
+- **abs:** https://arxiv.org/abs/2602.23056 · **PDF:** https://arxiv.org/pdf/2602.23056v1
+
+## 9489. Latent Matters: Learning Deep State-Space Models
+
+- **ID:** `2602.23050` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Alexej Klushyn, Richard Kurle, Maximilian Soelch, et al.
+- **abs:** https://arxiv.org/abs/2602.23050 · **PDF:** https://arxiv.org/pdf/2602.23050v1
+
+## 9490. NextMem: Towards Latent Factual Memory for LLM-based Agents
+
+- **ID:** `2603.15634` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Zeyu Zhang, Rui Li, Xiaoyan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.15634 · **PDF:** https://arxiv.org/pdf/2603.15634v1
+
+## 9491. SkillNet: Create, Evaluate, and Connect AI Skills
+
+- **ID:** `2603.04448` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Yuan Liang, Ruobin Zhong, Haoming Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.04448 · **PDF:** https://arxiv.org/pdf/2603.04448v1
+
+## 9492. LLMServingSim 2.0: A Unified Simulator for Heterogeneous and Disaggregated LLM Serving Infrastructure
+
+- **ID:** `2602.23036` · **primary:** `cs.DC` · **Submitted:** 2026-02-26
+- **Authors:** Jaehong Cho, Hyunmin Choi, Guseul Heo, et al.
+- **abs:** https://arxiv.org/abs/2602.23036 · **PDF:** https://arxiv.org/pdf/2602.23036v2
+
+## 9493. MAPLE: Metadata Augmented Private Language Evolution
+
+- **ID:** `2603.19258` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Eli Chien, Yuzheng Hu, Ryan McKenna, et al.
+- **abs:** https://arxiv.org/abs/2603.19258 · **PDF:** https://arxiv.org/pdf/2603.19258v1
+
+## 9494. Learning Disease-Sensitive Latent Interaction Graphs From Noisy Cardiac Flow Measurements
+
+- **ID:** `2602.23035` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Viraj Patel, Marko Grujic, Philipp Aigner, et al.
+- **abs:** https://arxiv.org/abs/2602.23035 · **PDF:** https://arxiv.org/pdf/2602.23035v1
+
+## 9495. Low-degree Lower bounds for clustering in moderate dimension
+
+- **ID:** `2602.23023` · **primary:** `math.ST` · **Submitted:** 2026-02-26
+- **Authors:** Alexandra Carpentier, Nicolas Verzelen
+- **abs:** https://arxiv.org/abs/2602.23023 · **PDF:** https://arxiv.org/pdf/2602.23023v1
+
+## 9496. A Novel Evolutionary Method for Automated Skull-Face Overlay in Computer-Aided Craniofacial Superimposition
+
+- **ID:** `2603.00170` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Práxedes Martínez-Moreno, Andrea Valsecchi, Pablo Mesejo, et al.
+- **abs:** https://arxiv.org/abs/2603.00170 · **PDF:** https://arxiv.org/pdf/2603.00170v3
+
+## 9497. SubspaceAD: Training-Free Few-Shot Anomaly Detection via Subspace Modeling
+
+- **ID:** `2602.23013` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Camile Lendering, Erkut Akdag, Egor Bondarev
+- **abs:** https://arxiv.org/abs/2602.23013 · **PDF:** https://arxiv.org/pdf/2602.23013v1
+
+## 9498. Sequential Regression for Continuous Value Prediction using Residual Quantization
+
+- **ID:** `2602.23012` · **primary:** `cs.IR` · **Submitted:** 2026-02-26
+- **Authors:** Runpeng Cui, Zhipeng Sun, Chi Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.23012 · **PDF:** https://arxiv.org/pdf/2602.23012v1
+
+## 9499. Exploratory Memory-Augmented LLM Agent via Hybrid On- and Off-Policy Optimization
+
+- **ID:** `2602.23008` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Zeyuan Liu, Jeonghye Kim, Xufang Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.23008 · **PDF:** https://arxiv.org/pdf/2602.23008v2
+
+## 9500. Regular Fourier Features for Nonstationary Gaussian Processes
+
+- **ID:** `2602.23006` · **primary:** `stat.ML` · **Submitted:** 2026-02-26
+- **Authors:** Arsalan Jawaid, Abdullah Karatas, Jörg Seewig
+- **abs:** https://arxiv.org/abs/2602.23006 · **PDF:** https://arxiv.org/pdf/2602.23006v1
+
+## 9501. Scattering Transform for Auditory Attention Decoding
+
+- **ID:** `2602.23003` · **primary:** `eess.SP` · **Submitted:** 2026-02-26
+- **Authors:** René Pallenberg, Fabrice Katzberg, Alfred Mertins, et al.
+- **abs:** https://arxiv.org/abs/2602.23003 · **PDF:** https://arxiv.org/pdf/2602.23003v1
+
+## 9502. Residual Koopman Spectral Profiling for Predicting and Preventing Transformer Training Instability
+
+- **ID:** `2602.22988` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Bum Jun Kim, Shohei Taniguchi, Makoto Kawano, et al.
+- **abs:** https://arxiv.org/abs/2602.22988 · **PDF:** https://arxiv.org/pdf/2602.22988v1
+
+## 9503. Kernel Integrated $R^2$: A Measure of Dependence
+
+- **ID:** `2602.22985` · **primary:** `stat.ML` · **Submitted:** 2026-02-26
+- **Authors:** Pouya Roudaki, Shakeel Gavioli-Akilagun, Florian Kalinke, et al.
+- **abs:** https://arxiv.org/abs/2602.22985 · **PDF:** https://arxiv.org/pdf/2602.22985v1
+
+## 9504. Obscure but Effective: Classical Chinese Jailbreak Prompt Optimization via Bio-Inspired Search
+
+- **ID:** `2602.22983` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Xun Huang, Simeng Qin, Xiaoshuang Jia, et al.
+- **abs:** https://arxiv.org/abs/2602.22983 · **PDF:** https://arxiv.org/pdf/2602.22983v3
+
+## 9505. RepSPD: Enhancing SPD Manifold Representation in EEGs via Dynamic Graphs
+
+- **ID:** `2602.22981` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Haohui Jia, Zheng Chen, Lingwei Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.22981 · **PDF:** https://arxiv.org/pdf/2602.22981v1
+
+## 9506. An automatic counting algorithm for the quantification and uncertainty analysis of the number of microglial cells trainable in small and heterogeneous datasets
+
+- **ID:** `2602.22974` · **primary:** `cs.CE` · **Submitted:** 2026-02-26
+- **Authors:** L. Martino, M. M. Garcia, P. S. Paradas, et al.
+- **abs:** https://arxiv.org/abs/2602.22974 · **PDF:** https://arxiv.org/pdf/2602.22974v1
+
+## 9507. Modeling Expert AI Diagnostic Alignment via Immutable Inference Snapshots
+
+- **ID:** `2602.22973` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Dimitrios P. Panagoulias, Evangelia-Aikaterini Tsichrintzi, Georgios Savvidis, et al.
+- **abs:** https://arxiv.org/abs/2602.22973 · **PDF:** https://arxiv.org/pdf/2602.22973v1
+
+## 9508. SPM-Bench: Benchmarking Large Language Models for Scanning Probe Microscopy
+
+- **ID:** `2602.22971` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Peiyao Xiao, Xiaogang Li, Chengliang Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.22971 · **PDF:** https://arxiv.org/pdf/2602.22971v1
+
+## 9509. Certified Circuits: Stability Guarantees for Mechanistic Circuits
+
+- **ID:** `2602.22968` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Alaa Anani, Tobias Lorenz, Bernt Schiele, et al.
+- **abs:** https://arxiv.org/abs/2602.22968 · **PDF:** https://arxiv.org/pdf/2602.22968v2
+
+## 9510. Discovery of Interpretable Physical Laws in Materials via Language-Model-Guided Symbolic Regression
+
+- **ID:** `2602.22967` · **primary:** `physics.comp-ph` · **Submitted:** 2026-02-26
+- **Authors:** Yifeng Guan, Chuyi Liu, Dongzhan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.22967 · **PDF:** https://arxiv.org/pdf/2602.22967v1
+
+## 9511. A note on the area under the likelihood and the fake evidence for model selection
+
+- **ID:** `2602.22965` · **primary:** `stat.ME` · **Submitted:** 2026-02-26
+- **Authors:** L. Martino, F. Llorente
+- **abs:** https://arxiv.org/abs/2602.22965 · **PDF:** https://arxiv.org/pdf/2602.22965v1
+
+## 9512. FactGuard: Agentic Video Misinformation Detection via Reinforcement Learning
+
+- **ID:** `2602.22963` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Zehao Li, Hongwei Yu, Hao Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.22963 · **PDF:** https://arxiv.org/pdf/2602.22963v1
+
+## 9513. Scaling Laws of Global Weather Models
+
+- **ID:** `2602.22962` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Yuejiang Yu, Langwen Huang, Alexandru Calotoiu, et al.
+- **abs:** https://arxiv.org/abs/2602.22962 · **PDF:** https://arxiv.org/pdf/2602.22962v1
+
+## 9514. Learning to Generate Secure Code via Token-Level Rewards
+
+- **ID:** `2602.23407` · **primary:** `cs.CR` · **Submitted:** 2026-02-26
+- **Authors:** Jiazheng Quan, Xiaodong Li, Bin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.23407 · **PDF:** https://arxiv.org/pdf/2602.23407v1
+
+## 9515. Frequency-Ordered Tokenization for Better Text Compression
+
+- **ID:** `2602.22958` · **primary:** `cs.IT` · **Submitted:** 2026-02-26
+- **Authors:** Maximilian Kalcher
+- **abs:** https://arxiv.org/abs/2602.22958 · **PDF:** https://arxiv.org/pdf/2602.22958v1
+
+## 9516. MM-NeuroOnco: A Multimodal Benchmark and Instruction Dataset for MRI-Based Brain Tumor Diagnosis
+
+- **ID:** `2602.22955` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Feng Guo, Jiaxiang Liu, Yang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.22955 · **PDF:** https://arxiv.org/pdf/2602.22955v1
+
+## 9517. Effective sample size approximations as entropy measures
+
+- **ID:** `2602.22954` · **primary:** `math.ST` · **Submitted:** 2026-02-26
+- **Authors:** L. Martino, V. Elvira
+- **abs:** https://arxiv.org/abs/2602.22954 · **PDF:** https://arxiv.org/pdf/2602.22954v1
+
+## 9518. General Agent Evaluation
+
+- **ID:** `2602.22953` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Elron Bandel, Asaf Yehudai, Lilach Eden, et al.
+- **abs:** https://arxiv.org/abs/2602.22953 · **PDF:** https://arxiv.org/pdf/2602.22953v1
+
+## 9519. On De-Individuated Neurons: Continuous Symmetries Enable Dynamic Topologies
+
+- **ID:** `2602.23405` · **primary:** `cs.NE` · **Submitted:** 2026-02-26
+- **Authors:** George Bird
+- **abs:** https://arxiv.org/abs/2602.23405 · **PDF:** https://arxiv.org/pdf/2602.23405v1
+
+## 9520. pMoE: Prompting Diverse Experts Together Wins More in Visual Adaptation
+
+- **ID:** `2602.22938` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Shentong Mo, Xufang Luo, Dongsheng Li
+- **abs:** https://arxiv.org/abs/2602.22938 · **PDF:** https://arxiv.org/pdf/2602.22938v1
+
+## 9521. MSINO: Curvature-Aware Sobolev Optimization for Manifold Neural Networks
+
+- **ID:** `2602.22937` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Suresan Pareth
+- **abs:** https://arxiv.org/abs/2602.22937 · **PDF:** https://arxiv.org/pdf/2602.22937v1
+
+## 9522. Generalization Bounds of Stochastic Gradient Descent in Homogeneous Neural Networks
+
+- **ID:** `2602.22936` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Wenquan Ma, Yang Sui, Jiaye Teng, et al.
+- **abs:** https://arxiv.org/abs/2602.22936 · **PDF:** https://arxiv.org/pdf/2602.22936v1
+
+## 9523. A Holistic Framework for Robust Bangla ASR and Speaker Diarization with Optimized VAD and CTC Alignment
+
+- **ID:** `2602.22935` · **primary:** `cs.SD` · **Submitted:** 2026-02-26
+- **Authors:** Zarif Ishmam, Zarif Mahir, Shafnan Wasif, et al.
+- **abs:** https://arxiv.org/abs/2602.22935 · **PDF:** https://arxiv.org/pdf/2602.22935v1
+
+## 9524. Beyond NNGP: Large Deviations and Feature Learning in Bayesian Neural Networks
+
+- **ID:** `2602.22925` · **primary:** `stat.ML` · **Submitted:** 2026-02-26
+- **Authors:** Katerina Papagiannouli, Dario Trevisan, Giuseppe Pio Zitto
+- **abs:** https://arxiv.org/abs/2602.22925 · **PDF:** https://arxiv.org/pdf/2602.22925v1
+
+## 9525. Where Vision Becomes Text: Locating the OCR Routing Bottleneck in Vision-Language Models
+
+- **ID:** `2602.22918` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Jonathan Steinberg, Oren Gal
+- **abs:** https://arxiv.org/abs/2602.22918 · **PDF:** https://arxiv.org/pdf/2602.22918v1
+
+## 9526. SIGMA: A Semantic-Grounded Instruction-Driven Generative Multi-Task Recommender at AliExpress
+
+- **ID:** `2602.22913` · **primary:** `cs.IR` · **Submitted:** 2026-02-26
+- **Authors:** Yang Yu, Lei Kou, Huaikuan Yi, et al.
+- **abs:** https://arxiv.org/abs/2602.22913 · **PDF:** https://arxiv.org/pdf/2602.22913v1
+
+## 9527. CeRA: Breaking the Linear Ceiling of Low-Rank Adaptation via Manifold Expansion
+
+- **ID:** `2602.22911` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Hung-Hsuan Chen
+- **abs:** https://arxiv.org/abs/2602.22911 · **PDF:** https://arxiv.org/pdf/2602.22911v4
+
+## 9528. PSQE: A Theoretical-Practical Approach to Pseudo Seed Quality Enhancement for Unsupervised Multimodal Entity Alignment
+
+- **ID:** `2602.22903` · **primary:** `cs.IR` · **Submitted:** 2026-02-26
+- **Authors:** Yunpeng Hong, Chenyang Bu, Jie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.22903 · **PDF:** https://arxiv.org/pdf/2602.22903v2
+
+## 9529. A Data-Driven Approach to Support Clinical Renal Replacement Therapy
+
+- **ID:** `2602.22902` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Alice Balboni, Luis Escobar, Andrea Manno, et al.
+- **abs:** https://arxiv.org/abs/2602.22902 · **PDF:** https://arxiv.org/pdf/2602.22902v1
+
+## 9530. OmniGAIA: Towards Native Omni-Modal AI Agents
+
+- **ID:** `2602.22897` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Xiaoxi Li, Wenxiang Jiao, Jiarui Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.22897 · **PDF:** https://arxiv.org/pdf/2602.22897v2
+
+## 9531. SPD Learn: A Geometric Deep Learning Python Library for Neural Decoding Through Trivialization
+
+- **ID:** `2602.22895` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-26
+- **Authors:** Bruno Aristimunha, Ce Ju, Antoine Collas, et al.
+- **abs:** https://arxiv.org/abs/2602.22895 · **PDF:** https://arxiv.org/pdf/2602.22895v1
+
+## 9532. Unsupervised Continual Learning for Amortized Bayesian Inference
+
+- **ID:** `2602.22884` · **primary:** `stat.ML` · **Submitted:** 2026-02-26
+- **Authors:** Aayush Mishra, Šimon Kucharský, Paul-Christian Bürkner
+- **abs:** https://arxiv.org/abs/2602.22884 · **PDF:** https://arxiv.org/pdf/2602.22884v1
+
+## 9533. Fair feature attribution for multi-output prediction: a Shapley-based perspective
+
+- **ID:** `2602.22882` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Umberto Biccari, Alain Ibáñez de Opakua, José María Mato, et al.
+- **abs:** https://arxiv.org/abs/2602.22882 · **PDF:** https://arxiv.org/pdf/2602.22882v1
+
+## 9534. Towards LLM-Empowered Knowledge Tracing via LLM-Student Hierarchical Behavior Alignment in Hyperbolic Space
+
+- **ID:** `2602.22879` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Xingcheng Fu, Shengpeng Wang, Yisen Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.22879 · **PDF:** https://arxiv.org/pdf/2602.22879v1
+
+## 9535. Learning Tangent Bundles and Characteristic Classes with Autoencoder Atlases
+
+- **ID:** `2602.22873` · **primary:** `math.AT` · **Submitted:** 2026-02-26
+- **Authors:** Eduardo Paluzo-Hidalgo, Yuichi Ike
+- **abs:** https://arxiv.org/abs/2602.22873 · **PDF:** https://arxiv.org/pdf/2602.22873v1
+
+## 9536. Test-Time Scaling with Diffusion Language Models via Reward-Guided Stitching
+
+- **ID:** `2602.22871` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Roy Miles, Aysim Toker, Andreea-Maria Oncescu, et al.
+- **abs:** https://arxiv.org/abs/2602.22871 · **PDF:** https://arxiv.org/pdf/2602.22871v1
+
+## 9537. Rejection Mixing: Fast Semantic Propagation of Mask Tokens for Efficient DLLM Inference
+
+- **ID:** `2602.22868` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Yushi Ye, Feng Hong, Huangjie Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.22868 · **PDF:** https://arxiv.org/pdf/2602.22868v1
+
+## 9538. Effective QA-driven Annotation of Predicate-Argument Relations Across Languages
+
+- **ID:** `2602.22865` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Jonathan Davidov, Aviv Slobodkin, Shmuel Tomi Klein, et al.
+- **abs:** https://arxiv.org/abs/2602.22865 · **PDF:** https://arxiv.org/pdf/2602.22865v1
+
+## 9539. MEDNA-DFM: A Dual-View FiLM-MoE Model for Explainable DNA Methylation Prediction
+
+- **ID:** `2602.22850` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Yi He, Yina Cao, Jixiu Zhai, et al.
+- **abs:** https://arxiv.org/abs/2602.22850 · **PDF:** https://arxiv.org/pdf/2602.22850v1
+
+## 9540. Decentralized Ranking Aggregation: Gossip Algorithms for Borda and Copeland Consensus
+
+- **ID:** `2602.22847` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Anna Van Elst, Kerrian Le Caillec, Igor Colin, et al.
+- **abs:** https://arxiv.org/abs/2602.22847 · **PDF:** https://arxiv.org/pdf/2602.22847v1
+
+## 9541. Improving Neural Argumentative Stance Classification in Controversial Topics with Emotion-Lexicon Features
+
+- **ID:** `2602.22846` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Mohammad Yeghaneh Abkenar, Weixing Wang, Manfred Stede, et al.
+- **abs:** https://arxiv.org/abs/2602.22846 · **PDF:** https://arxiv.org/pdf/2602.22846v1
+
+## 9542. The AI Research Assistant: Promise, Peril, and a Proof of Concept
+
+- **ID:** `2602.22842` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Tan Bui-Thanh
+- **abs:** https://arxiv.org/abs/2602.22842 · **PDF:** https://arxiv.org/pdf/2602.22842v1
+
+## 9543. DeepPresenter: Environment-Grounded Reflection for Agentic Presentation Generation
+
+- **ID:** `2602.22839` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Hao Zheng, Guozhao Mo, Xinru Yan, et al.
+- **abs:** https://arxiv.org/abs/2602.22839 · **PDF:** https://arxiv.org/pdf/2602.22839v1
+
+## 9544. Moral Preferences of LLMs Under Directed Contextual Influence
+
+- **ID:** `2602.22831` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Phil Blandfort, Tushar Karayil, Urja Pawar, et al.
+- **abs:** https://arxiv.org/abs/2602.22831 · **PDF:** https://arxiv.org/pdf/2602.22831v1
+
+## 9545. TCM-DiffRAG: Personalized Syndrome Differentiation Reasoning Method for Traditional Chinese Medicine based on Knowledge Graph and Chain of Thought
+
+- **ID:** `2602.22828` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Jianmin Li, Ying Chang, Su-Kit Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.22828 · **PDF:** https://arxiv.org/pdf/2602.22828v1
+
+## 9546. TARAZ: Persian Short-Answer Question Benchmark for Cultural Evaluation of Language Models
+
+- **ID:** `2602.22827` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Reihaneh Iranmanesh, Saeedeh Davoudi, Pasha Abrishamchian, et al.
+- **abs:** https://arxiv.org/abs/2602.22827 · **PDF:** https://arxiv.org/pdf/2602.22827v2
+
+## 9547. Hypernetwork-based approach for grid-independent functional data clustering
+
+- **ID:** `2602.22823` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Anirudh Thatipelli, Ali Siahkoohi
+- **abs:** https://arxiv.org/abs/2602.22823 · **PDF:** https://arxiv.org/pdf/2602.22823v1
+
+## 9548. FlexMS is a flexible framework for benchmarking deep learning-based mass spectrum prediction tools in metabolomics
+
+- **ID:** `2602.22822` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Yunhua Zhong, Yixuan Tang, Yifan Li, et al.
+- **abs:** https://arxiv.org/abs/2602.22822 · **PDF:** https://arxiv.org/pdf/2602.22822v1
+
+## 9549. Hierarchy-of-Groups Policy Optimization for Long-Horizon Agentic Tasks
+
+- **ID:** `2602.22817` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Shuo He, Lang Feng, Qi Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.22817 · **PDF:** https://arxiv.org/pdf/2602.22817v1
+
+## 9550. When Should an AI Act? A Human-Centered Model of Scene, Context, and Behavior for Agentic AI Design
+
+- **ID:** `2602.22814` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Soyoung Jung, Daehoo Yoon, Sung Gyu Koh, et al.
+- **abs:** https://arxiv.org/abs/2602.22814 · **PDF:** https://arxiv.org/pdf/2602.22814v1
+
+## 9551. Accelerating Local LLMs on Resource-Constrained Edge Devices via Distributed Prompt Caching
+
+- **ID:** `2602.22812` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Hiroki Matsutani, Naoki Matsuda, Naoto Sugiura
+- **abs:** https://arxiv.org/abs/2602.22812 · **PDF:** https://arxiv.org/pdf/2602.22812v1
+
+## 9552. Multi-agent imitation learning with function approximation: Linear Markov games and beyond
+
+- **ID:** `2602.22810` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Luca Viano, Till Freihaut, Emanuele Nevali, et al.
+- **abs:** https://arxiv.org/abs/2602.22810 · **PDF:** https://arxiv.org/pdf/2602.22810v1
+
+## 9553. MiroFlow: Towards High-Performance and Robust Open-Source Agent Framework for General Deep Research Tasks
+
+- **ID:** `2602.22808` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Shiqian Su, Sen Xing, Xuan Dong, et al.
+- **abs:** https://arxiv.org/abs/2602.22808 · **PDF:** https://arxiv.org/pdf/2602.22808v1
+
+## 9554. Communication-Guided Multi-Mutation Differential Evolution for Crop Model Calibration
+
+- **ID:** `2602.22804` · **primary:** `cs.NE` · **Submitted:** 2026-02-26
+- **Authors:** Sakshi Aggarwal, Mudasir Ganaie, Mukesh Saini
+- **abs:** https://arxiv.org/abs/2602.22804 · **PDF:** https://arxiv.org/pdf/2602.22804v1
+
+## 9555. Unleashing the Potential of Diffusion Models for End-to-End Autonomous Driving
+
+- **ID:** `2602.22801` · **primary:** `cs.RO` · **Submitted:** 2026-02-26
+- **Authors:** Yinan Zheng, Tianyi Tan, Bin Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.22801 · **PDF:** https://arxiv.org/pdf/2602.22801v1
+
+## 9556. Doubly Adaptive Channel and Spatial Attention for Semantic Image Communication by IoT Devices
+
+- **ID:** `2602.22794` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Soroosh Miri, Sepehr Abolhasani, Shahrokh Farahmand, et al.
+- **abs:** https://arxiv.org/abs/2602.22794 · **PDF:** https://arxiv.org/pdf/2602.22794v1
+
+## 9557. Natural Language Declarative Prompting (NLD-P): A Modular Governance Method for Prompt Design Under Model Drift
+
+- **ID:** `2602.22790` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Hyunwoo Kim, Hanau Yi, Jaehee Bae, et al.
+- **abs:** https://arxiv.org/abs/2602.22790 · **PDF:** https://arxiv.org/pdf/2602.22790v1
+
+## 9558. Probing for Knowledge Attribution in Large Language Models
+
+- **ID:** `2602.22787` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Ivo Brink, Alexander Boer, Dennis Ulmer
+- **abs:** https://arxiv.org/abs/2602.22787 · **PDF:** https://arxiv.org/pdf/2602.22787v1
+
+## 9559. QSIM: Mitigating Overestimation in Multi-Agent Reinforcement Learning via Action Similarity Weighted Q-Learning
+
+- **ID:** `2602.22786` · **primary:** `cs.MA` · **Submitted:** 2026-02-26
+- **Authors:** Yuanjun Li, Bin Zhang, Hao Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.22786 · **PDF:** https://arxiv.org/pdf/2602.22786v1
+
+## 9560. KMLP: A Scalable Hybrid Architecture for Web-Scale Tabular Data Modeling
+
+- **ID:** `2602.22777` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Mingming Zhang, Pengfei Shi, Zhiqing Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.22777 · **PDF:** https://arxiv.org/pdf/2602.22777v1
+
+## 9561. TherapyProbe: Generating Design Knowledge for Relational Safety in Mental Health Chatbots Through Adversarial Simulation
+
+- **ID:** `2602.22775` · **primary:** `cs.HC` · **Submitted:** 2026-02-26
+- **Authors:** Joydeep Chandra, Satyam Kumar Navneet, Yong Zhang
+- **abs:** https://arxiv.org/abs/2602.22775 · **PDF:** https://arxiv.org/pdf/2602.22775v1
+
+## 9562. ClinDet-Bench: Beyond Abstention, Evaluating Judgment Determinability of LLMs in Clinical Decision-Making
+
+- **ID:** `2602.22771` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Yusuke Watanabe, Yohei Kobashi, Takeshi Kojima, et al.
+- **abs:** https://arxiv.org/abs/2602.22771 · **PDF:** https://arxiv.org/pdf/2602.22771v1
+
+## 9563. Constraint-aware Path Planning from Natural Language Instructions Using Large Language Models
+
+- **ID:** `2603.19257` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Dylan Shim, Minghan Wei
+- **abs:** https://arxiv.org/abs/2603.19257 · **PDF:** https://arxiv.org/pdf/2603.19257v1
+
+## 9564. AMA-Bench: Evaluating Long-Horizon Memory for Agentic Applications
+
+- **ID:** `2602.22769` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Yujie Zhao, Boqin Yuan, Junbo Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.22769 · **PDF:** https://arxiv.org/pdf/2602.22769v2
+
+## 9565. Imagination Helps Visual Reasoning, But Not Yet in Latent Space
+
+- **ID:** `2602.22766` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** You Li, Chi Chen, Yanghao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.22766 · **PDF:** https://arxiv.org/pdf/2602.22766v1
+
+## 9566. Towards Better RL Training Data Utilization via Second-Order Rollout
+
+- **ID:** `2602.22765` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Zhe Yang, Yudong Wang, Rang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.22765 · **PDF:** https://arxiv.org/pdf/2602.22765v1
+
+## 9567. Distributed LLM Pretraining During Renewable Curtailment Windows: A Feasibility Study
+
+- **ID:** `2602.22760` · **primary:** `cs.DC` · **Submitted:** 2026-02-26
+- **Authors:** Philipp Wiesner, Soeren Becker, Brett Cornick, et al.
+- **abs:** https://arxiv.org/abs/2602.22760 · **PDF:** https://arxiv.org/pdf/2602.22760v1
+
+## 9568. Decomposing Physician Disagreement in HealthBench
+
+- **ID:** `2602.22758` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Satya Borgohain, Roy Mariathas
+- **abs:** https://arxiv.org/abs/2602.22758 · **PDF:** https://arxiv.org/pdf/2602.22758v2
+
+## 9569. AuditBench: Evaluating Alignment Auditing Techniques on Models with Hidden Behaviors
+
+- **ID:** `2602.22755` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Abhay Sheshadri, Aidan Ewart, Kai Fronsdal, et al.
+- **abs:** https://arxiv.org/abs/2602.22755 · **PDF:** https://arxiv.org/pdf/2602.22755v3
+
+## 9570. Towards Simulating Social Media Users with LLMs: Evaluating the Operational Validity of Conditioned Comment Prediction
+
+- **ID:** `2602.22752` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Nils Schwager, Simon Münker, Alistair Plum, et al.
+- **abs:** https://arxiv.org/abs/2602.22752 · **PDF:** https://arxiv.org/pdf/2602.22752v1
+
+## 9571. Know What You Know: Metacognitive Entropy Calibration for Verifiable RL Reasoning
+
+- **ID:** `2602.22751` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Qiannian Zhao, Chen Yang, Jinhao Jing, et al.
+- **abs:** https://arxiv.org/abs/2602.22751 · **PDF:** https://arxiv.org/pdf/2602.22751v1
+
+## 9572. Set-based v.s. Distribution-based Representations of Epistemic Uncertainty: A Comparative Study
+
+- **ID:** `2602.22747` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Kaizheng Wang, Yunjia Wang, Fabio Cuzzolin, et al.
+- **abs:** https://arxiv.org/abs/2602.22747 · **PDF:** https://arxiv.org/pdf/2602.22747v1
+
+## 9573. Generative Data Transformation: From Mixed to Unified Data
+
+- **ID:** `2602.22743` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Jiaqing Zhang, Mingjia Yin, Hao Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.22743 · **PDF:** https://arxiv.org/pdf/2602.22743v2
+
+## 9574. AMLRIS: Alignment-aware Masked Learning for Referring Image Segmentation
+
+- **ID:** `2602.22740` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Tongfei Chen, Shuo Yang, Yuguang Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.22740 · **PDF:** https://arxiv.org/pdf/2602.22740v2
+
+## 9575. Simulation-based Optimization for Augmented Reading
+
+- **ID:** `2602.22735` · **primary:** `cs.HC` · **Submitted:** 2026-02-26
+- **Authors:** Yunpeng Bai, Shengdong Zhao, Antti Oulasvirta
+- **abs:** https://arxiv.org/abs/2602.22735 · **PDF:** https://arxiv.org/pdf/2602.22735v1
+
+## 9576. Generative Recommendation for Large-Scale Advertising
+
+- **ID:** `2602.22732` · **primary:** `cs.IR` · **Submitted:** 2026-02-26
+- **Authors:** Ben Xue, Dan Liu, Lixiang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.22732 · **PDF:** https://arxiv.org/pdf/2602.22732v3
+
+## 9577. Extending Czech Aspect-Based Sentiment Analysis with Opinion Terms: Dataset and LLM Benchmarks
+
+- **ID:** `2602.22730` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Jakub Šmíd, Pavel Přibáň, Pavel Král
+- **abs:** https://arxiv.org/abs/2602.22730 · **PDF:** https://arxiv.org/pdf/2602.22730v2
+
+## 9578. Exploring the AI Obedience: Why is Generating a Pure Color Image Harder than CyberPunk?
+
+- **ID:** `2603.00166` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Hongyu Li, Kuan Liu, Yuan Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.00166 · **PDF:** https://arxiv.org/pdf/2603.00166v1
+
+## 9579. AgentSentry: Mitigating Indirect Prompt Injection in LLM Agents via Temporal Causal Diagnostics and Context Purification
+
+- **ID:** `2602.22724` · **primary:** `cs.CR` · **Submitted:** 2026-02-26
+- **Authors:** Tian Zhang, Yiwei Xu, Juan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.22724 · **PDF:** https://arxiv.org/pdf/2602.22724v1
+
+## 9580. Human Label Variation in Implicit Discourse Relation Recognition
+
+- **ID:** `2602.22723` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Frances Yung, Daniil Ignatev, Merel Scholman, et al.
+- **abs:** https://arxiv.org/abs/2602.22723 · **PDF:** https://arxiv.org/pdf/2602.22723v1
+
+## 9581. Replacing Multi-Step Assembly of Data Preparation Pipelines with One-Step LLM Pipeline Generation for Table QA
+
+- **ID:** `2602.22721` · **primary:** `cs.DB` · **Submitted:** 2026-02-26
+- **Authors:** Fengyu Li, Junhao Zhu, Kaishi Song, et al.
+- **abs:** https://arxiv.org/abs/2602.22721 · **PDF:** https://arxiv.org/pdf/2602.22721v2
+
+## 9582. Interpreting and Steering State-Space Models via Activation Subspace Bottlenecks
+
+- **ID:** `2602.22719` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Vamshi Sunku Mohan, Kaustubh Gupta, Aneesha Das, et al.
+- **abs:** https://arxiv.org/abs/2602.22719 · **PDF:** https://arxiv.org/pdf/2602.22719v1
+
+## 9583. RLHFless: Serverless Computing for Efficient RLHF
+
+- **ID:** `2602.22718` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Rui Wei, Hanfei Yu, Shubham Jain, et al.
+- **abs:** https://arxiv.org/abs/2602.22718 · **PDF:** https://arxiv.org/pdf/2602.22718v1
+
+## 9584. SoPE: Spherical Coordinate-Based Positional Embedding for Enhancing Spatial Perception of 3D LVLMs
+
+- **ID:** `2602.22716` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Guanting Ye, Qiyan Zhao, Wenhao Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.22716 · **PDF:** https://arxiv.org/pdf/2602.22716v1
+
+## 9585. U-CAN: Utility-Aware Contrastive Attenuation for Efficient Unlearning in Generative Recommendation
+
+- **ID:** `2602.23400` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Zezheng Wu, Rui Wang, Xinghe Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.23400 · **PDF:** https://arxiv.org/pdf/2602.23400v1
+
+## 9586. Same Words, Different Judgments: Modality Effects on Preference Alignment
+
+- **ID:** `2602.22710` · **primary:** `cs.SD` · **Submitted:** 2026-02-26
+- **Authors:** Aaron Broukhim, Nadir Weibel, Eshin Jolly
+- **abs:** https://arxiv.org/abs/2602.22710 · **PDF:** https://arxiv.org/pdf/2602.22710v1
+
+## 9587. Enhancing Geometric Perception in VLMs via Translator-Guided Reinforcement Learning
+
+- **ID:** `2602.22703` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Hao Yu, Shuning Jia, Guanghao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.22703 · **PDF:** https://arxiv.org/pdf/2602.22703v1
+
+## 9588. Knob: A Physics-Inspired Gating Interface for Interpretable and Controllable Neural Dynamics
+
+- **ID:** `2602.22702` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Siyu Jiang, Sanshuai Cui, Hui Zeng
+- **abs:** https://arxiv.org/abs/2602.22702 · **PDF:** https://arxiv.org/pdf/2602.22702v1
+
+## 9589. IMMACULATE: A Practical LLM Auditing Framework via Verifiable Computation
+
+- **ID:** `2602.22700` · **primary:** `cs.CR` · **Submitted:** 2026-02-26
+- **Authors:** Yanpei Guo, Wenjie Qu, Linyu Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.22700 · **PDF:** https://arxiv.org/pdf/2602.22700v1
+
+## 9590. DPSQL+: A Differentially Private SQL Library with a Minimum Frequency Rule
+
+- **ID:** `2602.22699` · **primary:** `cs.CR` · **Submitted:** 2026-02-26
+- **Authors:** Tomoya Matsumoto, Shokichi Takakura, Shun Takagi, et al.
+- **abs:** https://arxiv.org/abs/2602.22699 · **PDF:** https://arxiv.org/pdf/2602.22699v1
+
+## 9591. Tokenization, Fusion and Decoupling: Bridging the Granularity Mismatch Between Large Language Models and Knowledge Graphs
+
+- **ID:** `2602.22698` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Siyue Su, Jian Yang, Bo Li, et al.
+- **abs:** https://arxiv.org/abs/2602.22698 · **PDF:** https://arxiv.org/pdf/2602.22698v1
+
+## 9592. Reinforcing Real-world Service Agents: Balancing Utility and Cost in Task-oriented Dialogue
+
+- **ID:** `2602.22697` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Ning Gao, Wei Zhang, Yuqin Dai, et al.
+- **abs:** https://arxiv.org/abs/2602.22697 · **PDF:** https://arxiv.org/pdf/2602.22697v1
+
+## 9593. Enhancing Persuasive Dialogue Agents by Synthesizing Cross-Disciplinary Communication Strategies
+
+- **ID:** `2602.22696` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Shinnosuke Nozue, Yuto Nakano, Yotaro Watanabe, et al.
+- **abs:** https://arxiv.org/abs/2602.22696 · **PDF:** https://arxiv.org/pdf/2602.22696v1
+
+## 9594. Switch-Hurdle: A MoE Encoder with AR Hurdle Decoder for Intermittent Demand Forecasting
+
+- **ID:** `2602.22685` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Fabian Muşat, Simona Căbuz
+- **abs:** https://arxiv.org/abs/2602.22685 · **PDF:** https://arxiv.org/pdf/2602.22685v1
+
+## 9595. SUPERGLASSES: Benchmarking Vision Language Models as Intelligent Agents for AI Smart Glasses
+
+- **ID:** `2602.22683` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Zhuohang Jiang, Xu Yuan, Haohao Qu, et al.
+- **abs:** https://arxiv.org/abs/2602.22683 · **PDF:** https://arxiv.org/pdf/2602.22683v1
+
+## 9596. Accelerating LLM Pre-Training through Flat-Direction Dynamics Enhancement
+
+- **ID:** `2602.22681` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Shuchen Zhu, Rizhen Hu, Mingze Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.22681 · **PDF:** https://arxiv.org/pdf/2602.22681v1
+
+## 9597. Toward Personalized LLM-Powered Agents: Foundations, Evaluation, and Future Directions
+
+- **ID:** `2602.22680` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Yue Xu, Qian Chen, Zizhan Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.22680 · **PDF:** https://arxiv.org/pdf/2602.22680v2
+
+## 9598. ViCLIP-OT: The First Foundation Vision-Language Model for Vietnamese Image-Text Retrieval with Optimal Transport
+
+- **ID:** `2602.22678` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Quoc-Khang Tran, Minh-Thien Nguyen, Nguyen-Khang Pham
+- **abs:** https://arxiv.org/abs/2602.22678 · **PDF:** https://arxiv.org/pdf/2602.22678v1
+
+## 9599. Search More, Think Less: Rethinking Long-Horizon Agentic Search for Efficiency and Generalization
+
+- **ID:** `2602.22675` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Qianben Chen, Tianrui Qin, King Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.22675 · **PDF:** https://arxiv.org/pdf/2602.22675v2
+
+## 9600. Forecasting Antimicrobial Resistance Trends Using Machine Learning on WHO GLASS Surveillance Data: A Retrieval-Augmented Generation Approach for Policy Decision Support
+
+- **ID:** `2602.22673` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Md Tanvir Hasan Turja
+- **abs:** https://arxiv.org/abs/2602.22673 · **PDF:** https://arxiv.org/pdf/2602.22673v1
+
+## 9601. dLLM: Simple Diffusion Language Modeling
+
+- **ID:** `2602.22661` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Zhanhui Zhou, Lingjie Chen, Hanghang Tong, et al.
+- **abs:** https://arxiv.org/abs/2602.22661 · **PDF:** https://arxiv.org/pdf/2602.22661v1
+
+## 9602. LEDA: Latent Semantic Distribution Alignment for Multi-domain Graph Pre-training
+
+- **ID:** `2602.22660` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Lianze Shan, Jitao Zhao, Dongxiao He, et al.
+- **abs:** https://arxiv.org/abs/2602.22660 · **PDF:** https://arxiv.org/pdf/2602.22660v1
+
+## 9603. Deepfake Word Detection by Next-token Prediction using Fine-tuned Whisper
+
+- **ID:** `2602.22658` · **primary:** `eess.AS` · **Submitted:** 2026-02-26
+- **Authors:** Hoan My Tran, Xin Wang, Wanying Ge, et al.
+- **abs:** https://arxiv.org/abs/2602.22658 · **PDF:** https://arxiv.org/pdf/2602.22658v2
+
+## 9604. AHBid: An Adaptable Hierarchical Bidding Framework for Cross-Channel Advertising
+
+- **ID:** `2602.22650` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Xinxin Yang, Yangyang Tang, Yikun Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.22650 · **PDF:** https://arxiv.org/pdf/2602.22650v1
+
+## 9605. Vectorizing the Trie: Efficient Constrained Decoding for LLM-based Generative Retrieval on Accelerators
+
+- **ID:** `2602.22647` · **primary:** `cs.IR` · **Submitted:** 2026-02-26
+- **Authors:** Zhengyang Su, Isay Katsman, Yueqi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.22647 · **PDF:** https://arxiv.org/pdf/2602.22647v1
+
+## 9606. MUG: Meta-path-aware Universal Heterogeneous Graph Pre-Training
+
+- **ID:** `2602.22645` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Lianze Shan, Jitao Zhao, Dongxiao He, et al.
+- **abs:** https://arxiv.org/abs/2602.22645 · **PDF:** https://arxiv.org/pdf/2602.22645v1
+
+## 9607. Compress the Easy, Explore the Hard: Difficulty-Aware Entropy Regularization for Efficient LLM Reasoning
+
+- **ID:** `2602.22642` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Qin-Wen Luo, Sheng Ren, Xiang Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.22642 · **PDF:** https://arxiv.org/pdf/2602.22642v1
+
+## 9608. Reverse CAPTCHA: Evaluating LLM Susceptibility to Invisible Unicode Instruction Injection
+
+- **ID:** `2603.00164` · **primary:** `cs.CR` · **Submitted:** 2026-02-26
+- **Authors:** Marcus Graves
+- **abs:** https://arxiv.org/abs/2603.00164 · **PDF:** https://arxiv.org/pdf/2603.00164v1
+
+## 9609. Automated Measurement of Geniohyoid Muscle Thickness During Speech Using Deep Learning and Ultrasound
+
+- **ID:** `2603.03350` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-26
+- **Authors:** Alisher Myrgyyassov, Bruce Xiao Wang, Yu Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.03350 · **PDF:** https://arxiv.org/pdf/2603.03350v1
+
+## 9610. MobilityBench: A Benchmark for Evaluating Route-Planning Agents in Real-World Mobility Scenarios
+
+- **ID:** `2602.22638` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Zhiheng Song, Jingshuai Zhang, Chuan Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.22638 · **PDF:** https://arxiv.org/pdf/2602.22638v1
+
+## 9611. The AetherFloat Family: Block-Scale-Free Quad-Radix Floating-Point Architectures for AI Accelerators
+
+- **ID:** `2603.08741` · **primary:** `cs.AR` · **Submitted:** 2026-02-26
+- **Authors:** Keita Morisaki
+- **abs:** https://arxiv.org/abs/2603.08741 · **PDF:** https://arxiv.org/pdf/2603.08741v1
+
+## 9612. Tackling Privacy Heterogeneity in Differentially Private Federated Learning
+
+- **ID:** `2602.22633` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Ruichen Xu, Ying-Jun Angela Zhang, Jianwei Huang
+- **abs:** https://arxiv.org/abs/2602.22633 · **PDF:** https://arxiv.org/pdf/2602.22633v1
+
+## 9613. A Boundary-Metric Evaluation Protocol for Whiteboard Stroke Segmentation Under Extreme Imbalance
+
+- **ID:** `2603.00163` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Nicholas Korcynski
+- **abs:** https://arxiv.org/abs/2603.00163 · **PDF:** https://arxiv.org/pdf/2603.00163v1
+
+## 9614. TorchLean: Formalizing Neural Networks in Lean
+
+- **ID:** `2602.22631` · **primary:** `cs.MS` · **Submitted:** 2026-02-26
+- **Authors:** Robert Joseph George, Jennifer Cruden, Xiangru Zhong, et al.
+- **abs:** https://arxiv.org/abs/2602.22631 · **PDF:** https://arxiv.org/pdf/2602.22631v1
+
+## 9615. HyperKKL: Enabling Non-Autonomous State Estimation through Dynamic Weight Conditioning
+
+- **ID:** `2602.22630` · **primary:** `eess.SY` · **Submitted:** 2026-02-26
+- **Authors:** Yahia Salaheldin Shaaban, Salem Lahlou, Abdelrahman Sayed Sayed
+- **abs:** https://arxiv.org/abs/2602.22630 · **PDF:** https://arxiv.org/pdf/2602.22630v2
+
+## 9616. Instruction-based Image Editing with Planning, Reasoning, and Generation
+
+- **ID:** `2602.22624` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Liya Ji, Chenyang Qi, Qifeng Chen
+- **abs:** https://arxiv.org/abs/2602.22624 · **PDF:** https://arxiv.org/pdf/2602.22624v1
+
+## 9617. ContextRL: Enhancing MLLM's Knowledge Discovery Efficiency with Context-Augmented RL
+
+- **ID:** `2602.22623` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Xingyu Lu, Jinpeng Wang, YiFan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.22623 · **PDF:** https://arxiv.org/pdf/2602.22623v1
+
+## 9618. CGSA: Class-Guided Slot-Aware Adaptation for Source-Free Object Detection
+
+- **ID:** `2602.22621` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Boyang Dai, Zeng Fan, Zihao Qi, et al.
+- **abs:** https://arxiv.org/abs/2602.22621 · **PDF:** https://arxiv.org/pdf/2602.22621v1
+
+## 9619. Advancing accelerator virtual beam diagnostics through latent evolution modeling: an integrated solution to forward, inverse, tuning, and UQ problems
+
+- **ID:** `2602.22618` · **primary:** `physics.acc-ph` · **Submitted:** 2026-02-26
+- **Authors:** Mahindra Rautela, Alexander Scheinker
+- **abs:** https://arxiv.org/abs/2602.22618 · **PDF:** https://arxiv.org/pdf/2602.22618v1
+
+## 9620. Semantic Tube Prediction: Beating LLM Data Efficiency with JEPA
+
+- **ID:** `2602.22617` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Hai Huang, Yann LeCun, Randall Balestriero
+- **abs:** https://arxiv.org/abs/2602.22617 · **PDF:** https://arxiv.org/pdf/2602.22617v1
+
+## 9621. Mitigating Membership Inference in Intermediate Representations via Layer-wise MIA-risk-aware DP-SGD
+
+- **ID:** `2602.22611` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Jiayang Meng, Tao Huang, Chen Hou, et al.
+- **abs:** https://arxiv.org/abs/2602.22611 · **PDF:** https://arxiv.org/pdf/2602.22611v1
+
+## 9622. DP-aware AdaLN-Zero: Taming Conditioning-Induced Heavy-Tailed Gradients in Differentially Private Diffusion
+
+- **ID:** `2602.22610` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Tao Huang, Jiayang Meng, Xu Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.22610 · **PDF:** https://arxiv.org/pdf/2602.22610v1
+
+## 9623. EvolveGen: Algorithmic Level Hardware Model Checking Benchmark Generation through Reinforcement Learning
+
+- **ID:** `2602.22609` · **primary:** `cs.AR` · **Submitted:** 2026-02-26
+- **Authors:** Guangyu Hu, Xiaofeng Zhou, Wei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.22609 · **PDF:** https://arxiv.org/pdf/2602.22609v1
+
+## 9624. CoLyricist: Enhancing Lyric Writing with AI through Workflow-Aligned Support
+
+- **ID:** `2602.22606` · **primary:** `cs.HC` · **Submitted:** 2026-02-26
+- **Authors:** Masahiro Yoshida, Bingxuan Li, Songyan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.22606 · **PDF:** https://arxiv.org/pdf/2602.22606v1
+
+## 9625. SideQuest: Model-Driven KV Cache Management for Long-Horizon Agentic Reasoning
+
+- **ID:** `2602.22603` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Sanjay Kariyappa, G. Edward Suh
+- **abs:** https://arxiv.org/abs/2602.22603 · **PDF:** https://arxiv.org/pdf/2602.22603v2
+
+## 9626. Complex Networks and the Drug Repositioning Problem
+
+- **ID:** `2602.23396` · **primary:** `q-bio.MN` · **Submitted:** 2026-02-26
+- **Authors:** Felipe Bivort Haiek
+- **abs:** https://arxiv.org/abs/2602.23396 · **PDF:** https://arxiv.org/pdf/2602.23396v1
+
+## 9627. $φ$-DPO: Fairness Direct Preference Optimization Approach to Continual Learning in Large Multimodal Models
+
+- **ID:** `2602.22601` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Thanh-Dat Truong, Huu-Thien Tran, Jackson Cothren, et al.
+- **abs:** https://arxiv.org/abs/2602.22601 · **PDF:** https://arxiv.org/pdf/2602.22601v2
+
+## 9628. Transformers converge to invariant algorithmic cores
+
+- **ID:** `2602.22600` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Joshua S. Schiffman
+- **abs:** https://arxiv.org/abs/2602.22600 · **PDF:** https://arxiv.org/pdf/2602.22600v1
+
+## 9629. BetterScene: 3D Scene Synthesis with Representation-Aligned Generative Model
+
+- **ID:** `2602.22596` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Yuci Han, Charles Toth, John E. Anderson, et al.
+- **abs:** https://arxiv.org/abs/2602.22596 · **PDF:** https://arxiv.org/pdf/2602.22596v1
+
+## 9630. pQuant: Towards Effective Low-Bit Language Models via Decoupled Linear Quantization-Aware Training
+
+- **ID:** `2602.22592` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Wenzheng Zhang, Bingzheng Liu, Yang Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.22592 · **PDF:** https://arxiv.org/pdf/2602.22592v1
+
+## 9631. TabDLM: Free-Form Tabular Data Generation via Joint Numerical-Language Diffusion
+
+- **ID:** `2602.22586` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Donghong Cai, Jiarui Feng, Yanbo Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.22586 · **PDF:** https://arxiv.org/pdf/2602.22586v1
+
+## 9632. Correcting Human Labels for Rater Effects in AI Evaluation: An Item Response Theory Approach
+
+- **ID:** `2602.22585` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Jodi M. Casabianca, Maggie Beiting-Parrish
+- **abs:** https://arxiv.org/abs/2602.22585 · **PDF:** https://arxiv.org/pdf/2602.22585v1
+
+## 9633. Towards Faithful Industrial RAG: A Reinforced Co-adaptation Framework for Advertising QA
+
+- **ID:** `2602.22584` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Wenwei Li, Ming Xu, Tianle Xia, et al.
+- **abs:** https://arxiv.org/abs/2602.22584 · **PDF:** https://arxiv.org/pdf/2602.22584v1
+
+## 9634. Strategy Executability in Mathematical Reasoning: Leveraging Human-Model Differences for Effective Guidance
+
+- **ID:** `2602.22583` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Weida Liang, Yiyou Sun, Shuyuan Nan, et al.
+- **abs:** https://arxiv.org/abs/2602.22583 · **PDF:** https://arxiv.org/pdf/2602.22583v1
+
+## 9635. IBCircuit: Towards Holistic Circuit Discovery with Information Bottleneck
+
+- **ID:** `2602.22581` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Tian Bian, Yifan Niu, Chaohao Yuan, et al.
+- **abs:** https://arxiv.org/abs/2602.22581 · **PDF:** https://arxiv.org/pdf/2602.22581v1
+
+## 9636. Search-P1: Path-Centric Reward Shaping for Stable and Efficient Agentic RAG Training
+
+- **ID:** `2602.22576` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Tianle Xia, Ming Xu, Lingxiang Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.22576 · **PDF:** https://arxiv.org/pdf/2602.22576v1
+
+## 9637. S2O: Early Stopping for Sparse Attention via Online Permutation
+
+- **ID:** `2602.22575` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Yu Zhang, Songwei Liu, Chenqian Yan, et al.
+- **abs:** https://arxiv.org/abs/2602.22575 · **PDF:** https://arxiv.org/pdf/2602.22575v1
+
+## 9638. Guidance Matters: Rethinking the Evaluation Pitfall for Text-to-Image Generation
+
+- **ID:** `2602.22570` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Dian Xie, Shitong Shao, Lichen Bai, et al.
+- **abs:** https://arxiv.org/abs/2602.22570 · **PDF:** https://arxiv.org/pdf/2602.22570v1
+
+## 9639. Quality-Aware Robust Multi-View Clustering for Heterogeneous Observation Noise
+
+- **ID:** `2602.22568` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Peihan Wu, Guanjie Cheng, Yufei Tong, et al.
+- **abs:** https://arxiv.org/abs/2602.22568 · **PDF:** https://arxiv.org/pdf/2602.22568v1
+
+## 9640. Addressing Climate Action Misperceptions with Generative AI
+
+- **ID:** `2602.22564` · **primary:** `cs.HC` · **Submitted:** 2026-02-26
+- **Authors:** Miriam Remshard, Yara Kyrychenko, Sander van der Linden, et al.
+- **abs:** https://arxiv.org/abs/2602.22564 · **PDF:** https://arxiv.org/pdf/2602.22564v1
+
+## 9641. Operationalizing Fairness: Post-Hoc Threshold Optimization Under Hard Resource Limits
+
+- **ID:** `2602.22560` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Moirangthem Tiken Singh, Amit Kalita, Sapam Jitu Singh
+- **abs:** https://arxiv.org/abs/2602.22560 · **PDF:** https://arxiv.org/pdf/2602.22560v1
+
+## 9642. CourtGuard: A Model-Agnostic Framework for Zero-Shot Policy Adaptation in LLM Safety
+
+- **ID:** `2602.22557` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Umid Suleymanov, Rufiz Bayramov, Suad Gafarli, et al.
+- **abs:** https://arxiv.org/abs/2602.22557 · **PDF:** https://arxiv.org/pdf/2602.22557v1
+
+## 9643. Stable Adaptive Thinking via Advantage Shaping and Length-Aware Gradient Regulation
+
+- **ID:** `2602.22556` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Zihang Xu, Haozhi Xie, Ziqi Miao, et al.
+- **abs:** https://arxiv.org/abs/2602.22556 · **PDF:** https://arxiv.org/pdf/2602.22556v1
+
+## 9644. Autoregressive Visual Decoding from EEG Signals
+
+- **ID:** `2602.22555` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Sicheng Dai, Hongwang Xiao, Shan Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.22555 · **PDF:** https://arxiv.org/pdf/2602.22555v2
+
+## 9645. Multilingual Safety Alignment Via Sparse Weight Editing
+
+- **ID:** `2602.22554` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Jiaming Liang, Zhaoxin Wang, Handing Wang
+- **abs:** https://arxiv.org/abs/2602.22554 · **PDF:** https://arxiv.org/pdf/2602.22554v1
+
+## 9646. Relatron: Automating Relational Machine Learning over Relational Databases
+
+- **ID:** `2602.22552` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Zhikai Chen, Han Xie, Jian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.22552 · **PDF:** https://arxiv.org/pdf/2602.22552v1
+
+## 9647. A Fast and Practical Column Generation Approach for Identifying Carcinogenic Multi-Hit Gene Combinations
+
+- **ID:** `2602.22551` · **primary:** `math.OC` · **Submitted:** 2026-02-26
+- **Authors:** Rick S. H. Willemsen, Tenindra Abeywickrama, Ramu Anandakrishnan
+- **abs:** https://arxiv.org/abs/2602.22551 · **PDF:** https://arxiv.org/pdf/2602.22551v1
+
+## 9648. DrivePTS: A Progressive Learning Framework with Textual and Structural Enhancement for Driving Scene Generation
+
+- **ID:** `2602.22549` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Zhechao Wang, Yiming Zeng, Lufan Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.22549 · **PDF:** https://arxiv.org/pdf/2602.22549v1
+
+## 9649. Towards Dynamic Dense Retrieval with Routing Strategy
+
+- **ID:** `2602.22547` · **primary:** `cs.IR` · **Submitted:** 2026-02-26
+- **Authors:** Zhan Su, Fengran Mo, Jinghan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.22547 · **PDF:** https://arxiv.org/pdf/2602.22547v1
+
+## 9650. Requesting Expert Reasoning: Augmenting LLM Agents with Learned Collaborative Intervention
+
+- **ID:** `2602.22546` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Zhiming Wang, Jinwei He, Feng Lu
+- **abs:** https://arxiv.org/abs/2602.22546 · **PDF:** https://arxiv.org/pdf/2602.22546v1
+
+## 9651. DisQ-HNet: A Disentangled Quantized Half-UNet for Interpretable Multimodal Image Synthesis Applications to Tau-PET Synthesis from T1 and FLAIR MRI
+
+- **ID:** `2602.22545` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** Agamdeep S. Chopra, Caitlin Neher, Tianyi Ren, et al.
+- **abs:** https://arxiv.org/abs/2602.22545 · **PDF:** https://arxiv.org/pdf/2602.22545v1
+
+## 9652. HARU-Net: Hybrid Attention Residual U-Net for Edge-Preserving Denoising in Cone-Beam Computed Tomography
+
+- **ID:** `2602.22544` · **primary:** `eess.IV` · **Submitted:** 2026-02-26
+- **Authors:** Khuram Naveed, Ruben Pauwels
+- **abs:** https://arxiv.org/abs/2602.22544 · **PDF:** https://arxiv.org/pdf/2602.22544v1
+
+## 9653. Ruyi2 Technical Report
+
+- **ID:** `2602.22543` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Huan Song, Shuyu Tian, Junyi Hao, et al.
+- **abs:** https://arxiv.org/abs/2602.22543 · **PDF:** https://arxiv.org/pdf/2602.22543v1
+
+## 9654. SKINOPATHY AI: Smartphone-Based Ophthalmic Screening and Longitudinal Tracking Using Lightweight Computer Vision
+
+- **ID:** `2603.00161` · **primary:** `cs.CV` · **Submitted:** 2026-02-26
+- **Authors:** S. Kalaycioglu, C. Hong, M. Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.00161 · **PDF:** https://arxiv.org/pdf/2603.00161v1
+
+## 9655. Agentic AI for Intent-driven Optimization in Cell-free O-RAN
+
+- **ID:** `2602.22539` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Mohammad Hossein Shokouhi, Vincent W. S. Wong
+- **abs:** https://arxiv.org/abs/2602.22539 · **PDF:** https://arxiv.org/pdf/2602.22539v1
+
+## 9656. RAIN-Merging: A Gradient-Free Method to Enhance Instruction Following in Large Reasoning Models with Preserved Thinking Format
+
+- **ID:** `2602.22538` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Zhehao Huang, Yuhang Liu, Baijiong Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.22538 · **PDF:** https://arxiv.org/pdf/2602.22538v1
+
+## 9657. LUMOS: Democratizing SciML Workflows with L0-Regularized Learning for Unified Feature and Parameter Adaptation
+
+- **ID:** `2602.22537` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Shouwei Gao, Xu Zheng, Dongsheng Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.22537 · **PDF:** https://arxiv.org/pdf/2602.22537v1
+
+## 9658. Persistent Nonnegative Matrix Factorization via Multi-Scale Graph Regularization
+
+- **ID:** `2602.22536` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Jichao Zhang, Ran Miao, Limin Li
+- **abs:** https://arxiv.org/abs/2602.22536 · **PDF:** https://arxiv.org/pdf/2602.22536v1
+
+## 9659. A Synergistic Approach: Dynamics-AI Ensemble in Tropical Cyclone Forecasting
+
+- **ID:** `2602.22533` · **primary:** `physics.ao-ph` · **Submitted:** 2026-02-26
+- **Authors:** Yonghui Li, Wansuo Duan, Hao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.22533 · **PDF:** https://arxiv.org/pdf/2602.22533v1
+
+## 9660. Coarse-to-Fine Learning of Dynamic Causal Structures
+
+- **ID:** `2602.22532` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Dezhi Yang, Qiaoyu Tan, Carlotta Domeniconi, et al.
+- **abs:** https://arxiv.org/abs/2602.22532 · **PDF:** https://arxiv.org/pdf/2602.22532v1
+
+## 9661. Dynamic Level Sets
+
+- **ID:** `2602.22530` · **primary:** `cs.CC` · **Submitted:** 2026-02-26
+- **Authors:** Michael Stephen Fiske
+- **abs:** https://arxiv.org/abs/2602.22530 · **PDF:** https://arxiv.org/pdf/2602.22530v2
+
+## 9662. Generative Agents Navigating Digital Libraries
+
+- **ID:** `2602.22529` · **primary:** `cs.IR` · **Submitted:** 2026-02-26
+- **Authors:** Saber Zerhoudi, Michael Granitzer
+- **abs:** https://arxiv.org/abs/2602.22529 · **PDF:** https://arxiv.org/pdf/2602.22529v1
+
+## 9663. Predicting Tennis Serve directions with Machine Learning
+
+- **ID:** `2602.22527` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Ying Zhu, Ruthuparna Naikar
+- **abs:** https://arxiv.org/abs/2602.22527 · **PDF:** https://arxiv.org/pdf/2602.22527v1
+
+## 9664. SR-TTT: Surprisal-Aware Residual Test-Time Training
+
+- **ID:** `2603.06642` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Swamynathan V P
+- **abs:** https://arxiv.org/abs/2603.06642 · **PDF:** https://arxiv.org/pdf/2603.06642v1
+
+## 9665. Iterative Prompt Refinement for Dyslexia-Friendly Text Summarization Using GPT-4o
+
+- **ID:** `2602.22524` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** Samay Bhojwani, Swarnima Kain, Lisong Xu
+- **abs:** https://arxiv.org/abs/2602.22524 · **PDF:** https://arxiv.org/pdf/2602.22524v1
+
+## 9666. Cognitive Models and AI Algorithms Provide Templates for Designing Language Agents
+
+- **ID:** `2602.22523` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Ryan Liu, Dilip Arumugam, Cedegao E. Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.22523 · **PDF:** https://arxiv.org/pdf/2602.22523v1
+
+## 9667. Efficient Dialect-Aware Modeling and Conditioning for Low-Resource Taiwanese Hakka Speech Processing
+
+- **ID:** `2602.22522` · **primary:** `cs.CL` · **Submitted:** 2026-02-26
+- **Authors:** An-Ci Peng, Kuan-Tang Huang, Tien-Hong Lo, et al.
+- **abs:** https://arxiv.org/abs/2602.22522 · **PDF:** https://arxiv.org/pdf/2602.22522v1
+
+## 9668. TFPS: A Temporal Filtration-enhanced Positive Sample Set Construction Method for Implicit Collaborative Filtering
+
+- **ID:** `2602.22521` · **primary:** `cs.IR` · **Submitted:** 2026-02-26
+- **Authors:** Jiayi Wu, Zhengyu Wu, Xunkai Li, et al.
+- **abs:** https://arxiv.org/abs/2602.22521 · **PDF:** https://arxiv.org/pdf/2602.22521v1
+
+## 9669. TEFL: Prediction-Residual-Guided Rolling Forecasting for Multi-Horizon Time Series
+
+- **ID:** `2602.22520` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Xiannan Huang, Shen Fang, Shuhan Qiu, et al.
+- **abs:** https://arxiv.org/abs/2602.22520 · **PDF:** https://arxiv.org/pdf/2602.22520v1
+
+## 9670. A Mathematical Theory of Agency and Intelligence
+
+- **ID:** `2602.22519` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Wael Hafez, Chenan Wei, Rodrigo Pena, et al.
+- **abs:** https://arxiv.org/abs/2602.22519 · **PDF:** https://arxiv.org/pdf/2602.22519v2
+
+## 9671. SignVLA: A Gloss-Free Vision-Language-Action Framework for Real-Time Sign Language-Guided Robotic Manipulation
+
+- **ID:** `2602.22514` · **primary:** `cs.RO` · **Submitted:** 2026-02-26
+- **Authors:** Xinyu Tan, Ningwei Bai, Harry Gardener, et al.
+- **abs:** https://arxiv.org/abs/2602.22514 · **PDF:** https://arxiv.org/pdf/2602.22514v1
+
+## 9672. Mirroring the Mind: Distilling Human-Like Metacognitive Strategies into Large Language Models
+
+- **ID:** `2602.22508` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Ik-hwan Kim, Hyeongrok Han, Mingi Jung, et al.
+- **abs:** https://arxiv.org/abs/2602.22508 · **PDF:** https://arxiv.org/pdf/2602.22508v1
+
+## 9673. Space Syntax-guided Post-training for Residential Floor Plan Generation
+
+- **ID:** `2602.22507` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Zhuoyang Jiang, Dongqing Zhang
+- **abs:** https://arxiv.org/abs/2602.22507 · **PDF:** https://arxiv.org/pdf/2602.22507v1
+
+## 9674. Sharp Convergence Rates for Masked Diffusion Models
+
+- **ID:** `2602.22505` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Yuchen Liang, Zhiheng Tan, Ness Shroff, et al.
+- **abs:** https://arxiv.org/abs/2602.22505 · **PDF:** https://arxiv.org/pdf/2602.22505v1
+
+## 9675. Mapping the Landscape of Artificial Intelligence in Life Cycle Assessment Using Large Language Models
+
+- **ID:** `2602.22500` · **primary:** `cs.AI` · **Submitted:** 2026-02-26
+- **Authors:** Anastasija Mensikova, Donna M. Rizzo, Kathryn Hinkelman
+- **abs:** https://arxiv.org/abs/2602.22500 · **PDF:** https://arxiv.org/pdf/2602.22500v1
+
+## 9676. Reinforcement-aware Knowledge Distillation for LLM Reasoning
+
+- **ID:** `2602.22495` · **primary:** `cs.LG` · **Submitted:** 2026-02-26
+- **Authors:** Zhaoyang Zhang, Shuli Jiang, Yantao Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.22495 · **PDF:** https://arxiv.org/pdf/2602.22495v1
+
+## 9677. A Survey on Quantitative Modeling of Trust in Online Social Networks
+
+- **ID:** `2603.11054` · **primary:** `cs.SI` · **Submitted:** 2026-02-26
+- **Authors:** Wenting Song, K. Suzanne Barber
+- **abs:** https://arxiv.org/abs/2603.11054 · **PDF:** https://arxiv.org/pdf/2603.11054v1
+
+## 9678. From Shallow Bayesian Neural Networks to Gaussian Processes: General Convergence, Identifiability and Scalable Inference
+
+- **ID:** `2602.22492` · **primary:** `stat.ML` · **Submitted:** 2026-02-26
+- **Authors:** Gracielle Antunes de Araújo, Flávio B. Gonçalves
+- **abs:** https://arxiv.org/abs/2602.22492 · **PDF:** https://arxiv.org/pdf/2602.22492v1
+
+## 9679. Explainability-Aware Evaluation of Transfer Learning Models for IoT DDoS Detection Under Resource Constraints
+
+- **ID:** `2602.22488` · **primary:** `cs.CR` · **Submitted:** 2026-02-25
+- **Authors:** Nelly Elsayed
+- **abs:** https://arxiv.org/abs/2602.22488 · **PDF:** https://arxiv.org/pdf/2602.22488v1
+
+## 9680. Multi-hop Reasoning and Retrieval in Embedding Space: Leveraging Large Language Models with Knowledge
+
+- **ID:** `2603.13266` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Lihui Liu
+- **abs:** https://arxiv.org/abs/2603.13266 · **PDF:** https://arxiv.org/pdf/2603.13266v1
+
+## 9681. Flow Matching is Adaptive to Manifold Structures
+
+- **ID:** `2602.22486` · **primary:** `stat.ML` · **Submitted:** 2026-02-25
+- **Authors:** Shivam Kumar, Yixin Wang, Lizhen Lin
+- **abs:** https://arxiv.org/abs/2602.22486 · **PDF:** https://arxiv.org/pdf/2602.22486v1
+
+## 9682. Importance of Prompt Optimisation for Error Detection in Medical Notes Using Language Models
+
+- **ID:** `2602.22483` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Craig Myles, Patrick Schrempf, David Harris-Birtill
+- **abs:** https://arxiv.org/abs/2602.22483 · **PDF:** https://arxiv.org/pdf/2602.22483v1
+
+## 9683. Neural-Symbolic Logic Query Answering in Non-Euclidean Space
+
+- **ID:** `2603.15633` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Lihui Liu
+- **abs:** https://arxiv.org/abs/2603.15633 · **PDF:** https://arxiv.org/pdf/2603.15633v1
+
+## 9684. Sydney Telling Fables on AI and Humans: A Corpus Tracing Memetic Transfer of Persona between LLMs
+
+- **ID:** `2602.22481` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Jiří Milička, Hana Bednářová
+- **abs:** https://arxiv.org/abs/2602.22481 · **PDF:** https://arxiv.org/pdf/2602.22481v1
+
+## 9685. VeRO: An Evaluation Harness for Agents to Optimize Agents
+
+- **ID:** `2602.22480` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Varun Ursekar, Apaar Shanker, Veronica Chatrath, et al.
+- **abs:** https://arxiv.org/abs/2602.22480 · **PDF:** https://arxiv.org/pdf/2602.22480v1
+
+## 9686. Efficient Continual Learning in Language Models via Thalamically Routed Cortical Columns
+
+- **ID:** `2602.22479` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Afshin Khadangi
+- **abs:** https://arxiv.org/abs/2602.22479 · **PDF:** https://arxiv.org/pdf/2602.22479v5
+
+## 9687. Mind the Gap in Cultural Alignment: Task-Aware Culture Management for Large Language Models
+
+- **ID:** `2602.22475` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Binchi Zhang, Xujiang Zhao, Jundong Li, et al.
+- **abs:** https://arxiv.org/abs/2602.22475 · **PDF:** https://arxiv.org/pdf/2602.22475v1
+
+## 9688. When to Act, Ask, or Learn: Uncertainty-Aware Policy Steering
+
+- **ID:** `2602.22474` · **primary:** `cs.RO` · **Submitted:** 2026-02-25
+- **Authors:** Jessie Yuan, Yilin Wu, Andrea Bajcsy
+- **abs:** https://arxiv.org/abs/2602.22474 · **PDF:** https://arxiv.org/pdf/2602.22474v1
+
+## 9689. Beyond performance-wise Contribution Evaluation in Federated Learning
+
+- **ID:** `2602.22470` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Balazs Pejo
+- **abs:** https://arxiv.org/abs/2602.22470 · **PDF:** https://arxiv.org/pdf/2602.22470v1
+
+## 9690. Beyond Dominant Patches: Spatial Credit Redistribution For Grounded Vision-Language Models
+
+- **ID:** `2602.22469` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Niamul Hassan Samin, Md Arifur Rahman, Abdullah Ibne Hanif Arean, et al.
+- **abs:** https://arxiv.org/abs/2602.22469 · **PDF:** https://arxiv.org/pdf/2602.22469v2
+
+## 9691. ConstraintBench: Benchmarking LLM Constraint Reasoning on Direct Optimization
+
+- **ID:** `2602.22465` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Joseph Tso, Preston Schmittou, Quan Huynh, et al.
+- **abs:** https://arxiv.org/abs/2602.22465 · **PDF:** https://arxiv.org/pdf/2602.22465v2
+
+## 9692. DINOv3 Meets YOLO26 for Weed Detection in Vegetable Crops
+
+- **ID:** `2603.00160` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Boyang Deng, Yuzhen Lu
+- **abs:** https://arxiv.org/abs/2603.00160 · **PDF:** https://arxiv.org/pdf/2603.00160v1
+
+## 9693. Automating the Detection of Requirement Dependencies Using Large Language Models
+
+- **ID:** `2602.22456` · **primary:** `cs.SE` · **Submitted:** 2026-02-25
+- **Authors:** Ikram Darif, Feifei Niu, Manel Abdellatif, et al.
+- **abs:** https://arxiv.org/abs/2602.22456 · **PDF:** https://arxiv.org/pdf/2602.22456v1
+
+## 9694. Knowledge, Rules and Their Embeddings: Two Paths towards Neuro-Symbolic JEPA
+
+- **ID:** `2603.13265` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Yongchao Huang, Hassan Raza
+- **abs:** https://arxiv.org/abs/2603.13265 · **PDF:** https://arxiv.org/pdf/2603.13265v1
+
+## 9695. Federated Personal Knowledge Graph Completion with Lightweight Large Language Models for Personalized Recommendations
+
+- **ID:** `2603.13264` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Fernando Spadea, Oshani Seneviratne
+- **abs:** https://arxiv.org/abs/2603.13264 · **PDF:** https://arxiv.org/pdf/2603.13264v1
+
+## 9696. Bridging Latent Reasoning and Target-Language Generation via Retrieval-Transition Heads
+
+- **ID:** `2602.22453` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Shaswat Patel, Vishvesh Trivedi, Yue Han, et al.
+- **abs:** https://arxiv.org/abs/2602.22453 · **PDF:** https://arxiv.org/pdf/2602.22453v2
+
+## 9697. CWM: Contrastive World Models for Action Feasibility Learning in Embodied Agent Pipelines
+
+- **ID:** `2602.22452` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Chayan Banerjee
+- **abs:** https://arxiv.org/abs/2602.22452 · **PDF:** https://arxiv.org/pdf/2602.22452v1
+
+## 9698. Silent Egress: When Implicit Prompt Injection Makes LLM Agents Leak Without a Trace
+
+- **ID:** `2602.22450` · **primary:** `cs.CR` · **Submitted:** 2026-02-25
+- **Authors:** Qianlong Lan, Anuj Kaul, Shaun Jones, et al.
+- **abs:** https://arxiv.org/abs/2602.22450 · **PDF:** https://arxiv.org/pdf/2602.22450v1
+
+## 9699. A Fusion of context-aware based BanglaBERT and Two-Layer Stacked LSTM Framework for Multi-Label Cyberbullying Detection
+
+- **ID:** `2602.22449` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Mirza Raquib, Asif Pervez Polok, Kedar Nath Biswas, et al.
+- **abs:** https://arxiv.org/abs/2602.22449 · **PDF:** https://arxiv.org/pdf/2602.22449v1
+
+## 9700. ECHO: Encoding Communities via High-order Operators
+
+- **ID:** `2602.22446` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Emilio Ferrara
+- **abs:** https://arxiv.org/abs/2602.22446 · **PDF:** https://arxiv.org/pdf/2602.22446v1
+
+## 9701. FlowPortrait: Reinforcement Learning for Audio-Driven Portrait Video Generation
+
+- **ID:** `2603.00159` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Weiting Tan, Andy T. Liu, Ming Tu, et al.
+- **abs:** https://arxiv.org/abs/2603.00159 · **PDF:** https://arxiv.org/pdf/2603.00159v1
+
+## 9702. A Framework for Assessing AI Agent Decisions and Outcomes in AutoML Pipelines
+
+- **ID:** `2602.22442` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Gaoyuan Du, Amit Ahlawat, Xiaoyang Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.22442 · **PDF:** https://arxiv.org/pdf/2602.22442v2
+
+## 9703. How Do Latent Reasoning Methods Perform Under Weak and Strong Supervision?
+
+- **ID:** `2602.22441` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Yingqian Cui, Zhenwei Dai, Bing He, et al.
+- **abs:** https://arxiv.org/abs/2602.22441 · **PDF:** https://arxiv.org/pdf/2602.22441v1
+
+## 9704. From Bias to Balance: Fairness-Aware Paper Recommendation for Equitable Peer Review
+
+- **ID:** `2602.22438` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Uttamasha Anjally Oyshi, Susan Gauch
+- **abs:** https://arxiv.org/abs/2602.22438 · **PDF:** https://arxiv.org/pdf/2602.22438v1
+
+## 9705. veScale-FSDP: Flexible and High-Performance FSDP at Scale
+
+- **ID:** `2602.22437` · **primary:** `cs.DC` · **Submitted:** 2026-02-25
+- **Authors:** Zezhou Wang, Youjie Li, Zhiqi Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.22437 · **PDF:** https://arxiv.org/pdf/2602.22437v2
+
+## 9706. Beyond Attention: True Adaptive World Models via Spherical Kernel Operator
+
+- **ID:** `2603.13263` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Vladimer Khasia
+- **abs:** https://arxiv.org/abs/2603.13263 · **PDF:** https://arxiv.org/pdf/2603.13263v1
+
+## 9707. GetBatch: Distributed Multi-Object Retrieval for ML Data Loading
+
+- **ID:** `2602.22434` · **primary:** `cs.DC` · **Submitted:** 2026-02-25
+- **Authors:** Alex Aizman, Abhishek Gaikwad, Piotr Żelasko
+- **abs:** https://arxiv.org/abs/2602.22434 · **PDF:** https://arxiv.org/pdf/2602.22434v1
+
+## 9708. LoBoost: Fast Model-Native Local Conformal Prediction for Gradient-Boosted Trees
+
+- **ID:** `2602.22432` · **primary:** `stat.ML` · **Submitted:** 2026-02-25
+- **Authors:** Vagner Santos, Victor Coscrato, Luben Cabezas, et al.
+- **abs:** https://arxiv.org/abs/2602.22432 · **PDF:** https://arxiv.org/pdf/2602.22432v1
+
+## 9709. mmWave Radar Aware Dual-Conditioned GAN for Speech Reconstruction of Signals With Low SNR
+
+- **ID:** `2602.22431` · **primary:** `cs.SD` · **Submitted:** 2026-02-25
+- **Authors:** Jash Karani, Adithya Chittem, Deepan Roy, et al.
+- **abs:** https://arxiv.org/abs/2602.22431 · **PDF:** https://arxiv.org/pdf/2602.22431v1
+
+## 9710. TopoEdit: Fast Post-Optimization Editing of Topology Optimized Structures
+
+- **ID:** `2602.22430` · **primary:** `cs.GR` · **Submitted:** 2026-02-25
+- **Authors:** Hongrui Chen, Josephine V. Carstensen, Faez Ahmed
+- **abs:** https://arxiv.org/abs/2602.22430 · **PDF:** https://arxiv.org/pdf/2602.22430v1
+
+## 9711. Calibrated Test-Time Guidance for Bayesian Inference
+
+- **ID:** `2602.22428` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Daniel Geyfman, Felix Draxler, Jan Groeneveld, et al.
+- **abs:** https://arxiv.org/abs/2602.22428 · **PDF:** https://arxiv.org/pdf/2602.22428v1
+
+## 9712. Adversarial Hubness Detector: Detecting Hubness Poisoning in Retrieval-Augmented Generation Systems
+
+- **ID:** `2602.22427` · **primary:** `cs.CR` · **Submitted:** 2026-02-25
+- **Authors:** Idan Habler, Vineeth Sai Narajala, Stav Koren, et al.
+- **abs:** https://arxiv.org/abs/2602.22427 · **PDF:** https://arxiv.org/pdf/2602.22427v2
+
+## 9713. SimpleOCR: Rendering Visualized Questions to Teach MLLMs to Read
+
+- **ID:** `2602.22426` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Yibo Peng, Peng Xia, Ding Zhong, et al.
+- **abs:** https://arxiv.org/abs/2602.22426 · **PDF:** https://arxiv.org/pdf/2602.22426v1
+
+## 9714. ArchAgent: Agentic AI-driven Computer Architecture Discovery
+
+- **ID:** `2602.22425` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Raghav Gupta, Akanksha Jain, Abraham Gonzalez, et al.
+- **abs:** https://arxiv.org/abs/2602.22425 · **PDF:** https://arxiv.org/pdf/2602.22425v1
+
+## 9715. Causality $\neq$ Invariance: Function and Concept Vectors in LLMs
+
+- **ID:** `2602.22424` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Gustaw Opiełka, Hannes Rosenbusch, Claire E. Stevenson
+- **abs:** https://arxiv.org/abs/2602.22424 · **PDF:** https://arxiv.org/pdf/2602.22424v1
+
+## 9716. Revisiting Chebyshev Polynomial and Anisotropic RBF Models for Tabular Regression
+
+- **ID:** `2602.22422` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Luciano Gerber, Huw Lloyd
+- **abs:** https://arxiv.org/abs/2602.22422 · **PDF:** https://arxiv.org/pdf/2602.22422v1
+
+## 9717. Evaluation of Audio Language Models for Fairness, Safety, and Security
+
+- **ID:** `2603.13262` · **primary:** `cs.SD` · **Submitted:** 2026-02-25
+- **Authors:** Ranya Aloufi, Srishti Gupta, Soumya Shaw, et al.
+- **abs:** https://arxiv.org/abs/2603.13262 · **PDF:** https://arxiv.org/pdf/2603.13262v1
+
+## 9718. Epistemic Filtering and Collective Hallucination: A Jury Theorem for Confidence-Calibrated Agents
+
+- **ID:** `2602.22413` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Jonas Karge
+- **abs:** https://arxiv.org/abs/2602.22413 · **PDF:** https://arxiv.org/pdf/2602.22413v2
+
+## 9719. A Learning-Based Hybrid Decision Framework for Matching Systems with User Departure Detection
+
+- **ID:** `2602.22412` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Ruiqi Zhou, Donghao Zhu, Houcai Shen
+- **abs:** https://arxiv.org/abs/2602.22412 · **PDF:** https://arxiv.org/pdf/2602.22412v1
+
+## 9720. Exploring Human Behavior During Abstract Rule Inference and Problem Solving with the Cognitive Abstraction and Reasoning Corpus
+
+- **ID:** `2602.22408` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Caroline Ahn, Quan Do, Leah Bakst, et al.
+- **abs:** https://arxiv.org/abs/2602.22408 · **PDF:** https://arxiv.org/pdf/2602.22408v1
+
+## 9721. Towards Autonomous Memory Agents
+
+- **ID:** `2602.22406` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Xinle Wu, Rui Zhang, Mustafa Anis Hussain, et al.
+- **abs:** https://arxiv.org/abs/2602.22406 · **PDF:** https://arxiv.org/pdf/2602.22406v1
+
+## 9722. MolFM-Lite: Multi-Modal Molecular Property Prediction with Conformer Ensemble Attention and Cross-Modal Fusion
+
+- **ID:** `2602.22405` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Syed Omer Shah, Mohammed Maqsood Ahmed, Danish Mohiuddin Mohammed, et al.
+- **abs:** https://arxiv.org/abs/2602.22405 · **PDF:** https://arxiv.org/pdf/2602.22405v1
+
+## 9723. SAFARI: A Community-Engaged Approach and Dataset of Stereotype Resources in the Sub-Saharan African Context
+
+- **ID:** `2602.22404` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Aishwarya Verma, Laud Ammah, Olivia Nercy Ndlovu Lucas, et al.
+- **abs:** https://arxiv.org/abs/2602.22404 · **PDF:** https://arxiv.org/pdf/2602.22404v1
+
+## 9724. Contextual Memory Virtualisation: DAG-Based State Management and Structurally Lossless Trimming for LLM Agents
+
+- **ID:** `2602.22402` · **primary:** `cs.SE` · **Submitted:** 2026-02-25
+- **Authors:** Cosmo Santoni
+- **abs:** https://arxiv.org/abs/2602.22402 · **PDF:** https://arxiv.org/pdf/2602.22402v1
+
+## 9725. Vibe Researching as Wolf Coming: Can AI Agents with Skills Replace or Augment Social Scientists?
+
+- **ID:** `2602.22401` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Yongjun Zhang
+- **abs:** https://arxiv.org/abs/2602.22401 · **PDF:** https://arxiv.org/pdf/2602.22401v3
+
+## 9726. Predicting Multi-Drug Resistance in Bacterial Isolates Through Performance Comparison and LIME-based Interpretation of Classification Models
+
+- **ID:** `2602.22400` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Santanam Wishal, Riad Sahara
+- **abs:** https://arxiv.org/abs/2602.22400 · **PDF:** https://arxiv.org/pdf/2602.22400v2
+
+## 9727. Detecting Hate and Inflammatory Content in Bengali Memes: A New Multimodal Dataset and Co-Attention Framework
+
+- **ID:** `2602.22391` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Rakib Ullah, Mominul islam, Md Sanjid Hossain, et al.
+- **abs:** https://arxiv.org/abs/2602.22391 · **PDF:** https://arxiv.org/pdf/2602.22391v1
+
+## 9728. Deep Convolutional Architectures for EEG Classification: A Comparative Study with Temporal Augmentation and Confidence-Based Voting
+
+- **ID:** `2603.13261` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Aryan Patodiya, Hubert Cecotti
+- **abs:** https://arxiv.org/abs/2603.13261 · **PDF:** https://arxiv.org/pdf/2603.13261v1
+
+## 9729. Disentangling Shared and Target-Enriched Topics via Background-Contrastive Non-negative Matrix Factorization
+
+- **ID:** `2602.22387` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Yixuan Li, Archer Y. Yang, Yue Li
+- **abs:** https://arxiv.org/abs/2602.22387 · **PDF:** https://arxiv.org/pdf/2602.22387v1
+
+## 9730. Enhancing Renal Tumor Malignancy Prediction: Deep Learning with Automatic 3D CT Organ Focused Attention
+
+- **ID:** `2602.22381` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Zhengkang Fan, Chengkun Sun, Russell Terry, et al.
+- **abs:** https://arxiv.org/abs/2602.22381 · **PDF:** https://arxiv.org/pdf/2602.22381v1
+
+## 9731. AeroDGS: Physically Consistent Dynamic Gaussian Splatting for Single-Sequence Aerial 4D Reconstruction
+
+- **ID:** `2602.22376` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Hanyang Liu, Rongjun Qin
+- **abs:** https://arxiv.org/abs/2602.22376 · **PDF:** https://arxiv.org/pdf/2602.22376v1
+
+## 9732. Sampling from Constrained Gibbs Measures: with Applications to High-Dimensional Bayesian Inference
+
+- **ID:** `2602.22369` · **primary:** `math.ST` · **Submitted:** 2026-02-25
+- **Authors:** Ruixiao Wang, Xiaohong Chen, Sinho Chewi
+- **abs:** https://arxiv.org/abs/2602.22369 · **PDF:** https://arxiv.org/pdf/2602.22369v1
+
+## 9733. EyeLayer: Integrating Human Attention Patterns into LLM-Based Code Summarization
+
+- **ID:** `2602.22368` · **primary:** `cs.SE` · **Submitted:** 2026-02-25
+- **Authors:** Jiahao Zhang, Yifan Zhang, Kevin Leach, et al.
+- **abs:** https://arxiv.org/abs/2602.22368 · **PDF:** https://arxiv.org/pdf/2602.22368v1
+
+## 9734. Learning geometry-dependent lead-field operators for forward ECG modeling
+
+- **ID:** `2602.22367` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Arsenii Dokuchaev, Francesca Bonizzoni, Stefano Pagani, et al.
+- **abs:** https://arxiv.org/abs/2602.22367 · **PDF:** https://arxiv.org/pdf/2602.22367v1
+
+## 9735. Scaling In, Not Up? Testing Thick Citation Context Analysis with GPT-5 and Fragile Prompts
+
+- **ID:** `2602.22359` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Arno Simons
+- **abs:** https://arxiv.org/abs/2602.22359 · **PDF:** https://arxiv.org/pdf/2602.22359v1
+
+## 9736. GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators
+
+- **ID:** `2602.22352` · **primary:** `cs.AR` · **Submitted:** 2026-02-25
+- **Authors:** Yuhao Liu, Salim Ullah, Akash Kumar
+- **abs:** https://arxiv.org/abs/2602.22352 · **PDF:** https://arxiv.org/pdf/2602.22352v1
+
+## 9737. Decoder-based Sense Knowledge Distillation
+
+- **ID:** `2602.22351` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Qitong Wang, Mohammed J. Zaki, Georgios Kollias, et al.
+- **abs:** https://arxiv.org/abs/2602.22351 · **PDF:** https://arxiv.org/pdf/2602.22351v1
+
+## 9738. Enabling clinical use of foundation models in histopathology
+
+- **ID:** `2602.22347` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Audun L. Henriksen, Ole-Johan Skrede, Lisa van der Schee, et al.
+- **abs:** https://arxiv.org/abs/2602.22347 · **PDF:** https://arxiv.org/pdf/2602.22347v1
+
+## 9739. Structure and Redundancy in Large Language Models: A Spectral Study via Random Matrix Theory
+
+- **ID:** `2602.22345` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Davide Ettori
+- **abs:** https://arxiv.org/abs/2602.22345 · **PDF:** https://arxiv.org/pdf/2602.22345v1
+
+## 9740. A 1/R Law for Kurtosis Contrast in Balanced Mixtures
+
+- **ID:** `2602.22334` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Yuda Bi, Wenjun Xiao, Linhao Bai, et al.
+- **abs:** https://arxiv.org/abs/2602.22334 · **PDF:** https://arxiv.org/pdf/2602.22334v1
+
+## 9741. Recovered in Translation: Efficient Pipeline for Automated Translation of Benchmarks and Datasets
+
+- **ID:** `2602.22207` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Hanna Yukhymenko, Anton Alexandrov, Martin Vechev
+- **abs:** https://arxiv.org/abs/2602.22207 · **PDF:** https://arxiv.org/pdf/2602.22207v1
+
+## 9742. SumTablets: A Transliteration Dataset of Sumerian Tablets
+
+- **ID:** `2602.22200` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Cole Simmons, Richard Diehl Martinez, Dan Jurafsky
+- **abs:** https://arxiv.org/abs/2602.22200 · **PDF:** https://arxiv.org/pdf/2602.22200v1
+
+## 9743. Training Agents to Self-Report Misbehavior
+
+- **ID:** `2602.22303` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Bruce W. Lee, Chen Yueh-Han, Tomek Korbak
+- **abs:** https://arxiv.org/abs/2602.22303 · **PDF:** https://arxiv.org/pdf/2602.22303v1
+
+## 9744. Off-The-Shelf Image-to-Image Models Are All You Need To Defeat Image Protection Schemes
+
+- **ID:** `2602.22197` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Xavier Pleimling, Sifat Muhammad Abdullah, Gunjan Balde, et al.
+- **abs:** https://arxiv.org/abs/2602.22197 · **PDF:** https://arxiv.org/pdf/2602.22197v1
+
+## 9745. Improving Parametric Knowledge Access in Reasoning Language Models
+
+- **ID:** `2602.22193` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Melody Ma, John Hewitt
+- **abs:** https://arxiv.org/abs/2602.22193 · **PDF:** https://arxiv.org/pdf/2602.22193v1
+
+## 9746. Agent Behavioral Contracts: Formal Specification and Runtime Enforcement for Reliable Autonomous AI Agents
+
+- **ID:** `2602.22302` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Varun Pratap Bhardwaj
+- **abs:** https://arxiv.org/abs/2602.22302 · **PDF:** https://arxiv.org/pdf/2602.22302v1
+
+## 9747. GUI-Libra: Training Native GUI Agents to Reason and Act with Action-aware Supervision and Partially Verifiable RL
+
+- **ID:** `2602.22190` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Rui Yang, Qianhui Wu, Zhaoyang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.22190 · **PDF:** https://arxiv.org/pdf/2602.22190v1
+
+## 9748. Surrogate models for Rock-Fluid Interaction: A Grid-Size-Invariant Approach
+
+- **ID:** `2602.22188` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Nathalie C. Pinheiro, Donghu Guo, Hannah P. Menke, et al.
+- **abs:** https://arxiv.org/abs/2602.22188 · **PDF:** https://arxiv.org/pdf/2602.22188v1
+
+## 9749. ShobdoSetu: A Data-Centric Framework for Bengali Long-Form Speech Recognition and Speaker Diarization
+
+- **ID:** `2603.19256` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Md. Nazmus Sakib, Shafiul Tanvir, Mesbah Uddin Ahamed, et al.
+- **abs:** https://arxiv.org/abs/2603.19256 · **PDF:** https://arxiv.org/pdf/2603.19256v1
+
+## 9750. Testable Learning of General Halfspaces under Massart Noise
+
+- **ID:** `2602.22300` · **primary:** `cs.DS` · **Submitted:** 2026-02-25
+- **Authors:** Ilias Diakonikolas, Giannis Iakovidis, Daniel M. Kane, et al.
+- **abs:** https://arxiv.org/abs/2602.22300 · **PDF:** https://arxiv.org/pdf/2602.22300v1
+
+## 9751. LiCQA : A Lightweight Complex Question Answering System
+
+- **ID:** `2602.22182` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Sourav Saha, Dwaipayan Roy, Mandar Mitra
+- **abs:** https://arxiv.org/abs/2602.22182 · **PDF:** https://arxiv.org/pdf/2602.22182v1
+
+## 9752. Learning and Naming Subgroups with Exceptional Survival Characteristics
+
+- **ID:** `2602.22179` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Mhd Jawad Al Rahwanji, Sascha Xu, Nils Philipp Walter, et al.
+- **abs:** https://arxiv.org/abs/2602.22179 · **PDF:** https://arxiv.org/pdf/2602.22179v1
+
+## 9753. Decoding the Hook: A Multimodal LLM Framework for Analyzing the Hooking Period of Video Ads
+
+- **ID:** `2602.22299` · **primary:** `cs.MM` · **Submitted:** 2026-02-25
+- **Authors:** Kunpeng Zhang, Poppy Zhang, Shawndra Hill, et al.
+- **abs:** https://arxiv.org/abs/2602.22299 · **PDF:** https://arxiv.org/pdf/2602.22299v1
+
+## 9754. DySCO: Dynamic Attention-Scaling Decoding for Long-Context LMs
+
+- **ID:** `2602.22175` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Xi Ye, Wuwei Zhang, Fangcong Yin, et al.
+- **abs:** https://arxiv.org/abs/2602.22175 · **PDF:** https://arxiv.org/pdf/2602.22175v1
+
+## 9755. Applying a Random-Key Optimizer on Mixed Integer Programs
+
+- **ID:** `2602.22173` · **primary:** `math.OC` · **Submitted:** 2026-02-25
+- **Authors:** Antonio A. Chaves, Mauricio G. C. Resende, Carise E. Schmidt, et al.
+- **abs:** https://arxiv.org/abs/2602.22173 · **PDF:** https://arxiv.org/pdf/2602.22173v1
+
+## 9756. Dynamic Personality Adaptation in Large Language Models via State Machines
+
+- **ID:** `2602.22157` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Leon Pielage, Ole Hätscher, Mitja Back, et al.
+- **abs:** https://arxiv.org/abs/2602.22157 · **PDF:** https://arxiv.org/pdf/2602.22157v1
+
+## 9757. Stream Neural Networks: Epoch-Free Learning with Persistent Temporal State
+
+- **ID:** `2602.22152` · **primary:** `cs.NE` · **Submitted:** 2026-02-25
+- **Authors:** Amama Pathan
+- **abs:** https://arxiv.org/abs/2602.22152 · **PDF:** https://arxiv.org/pdf/2602.22152v1
+
+## 9758. Enhancing Framingham Cardiovascular Risk Score Transparency through Logic-Based XAI
+
+- **ID:** `2602.22149` · **primary:** `cs.LO` · **Submitted:** 2026-02-25
+- **Authors:** Emannuel L. de A. Bezerra, Luiz H. T. Viana, Vinícius P. Chagas, et al.
+- **abs:** https://arxiv.org/abs/2602.22149 · **PDF:** https://arxiv.org/pdf/2602.22149v1
+
+## 9759. Provable Last-Iterate Convergence for Multi-Objective Safe LLM Alignment via Optimistic Primal-Dual
+
+- **ID:** `2602.22146` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Yining Li, Peizhong Ju, Ness Shroff
+- **abs:** https://arxiv.org/abs/2602.22146 · **PDF:** https://arxiv.org/pdf/2602.22146v1
+
+## 9760. When AI Writes, Whose Voice Remains? Quantifying Cultural Marker Erasure Across World English Varieties in Large Language Models
+
+- **ID:** `2602.22145` · **primary:** `cs.HC` · **Submitted:** 2026-02-25
+- **Authors:** Satyam Kumar Navneet, Joydeep Chandra, Yong Zhang
+- **abs:** https://arxiv.org/abs/2602.22145 · **PDF:** https://arxiv.org/pdf/2602.22145v1
+
+## 9761. Architectural Design and Performance Analysis of FPGA based AI Accelerators: A Comprehensive Review
+
+- **ID:** `2603.08740` · **primary:** `cs.AR` · **Submitted:** 2026-02-25
+- **Authors:** Soumita Chatterjee, Sudip Ghosh, Tamal Ghosh, et al.
+- **abs:** https://arxiv.org/abs/2603.08740 · **PDF:** https://arxiv.org/pdf/2603.08740v1
+
+## 9762. AviaSafe: A Physics-Informed Data-Driven Model for Aviation Safety-Critical Cloud Forecasts
+
+- **ID:** `2602.22298` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Zijian Zhu, Qiusheng Huang, Anboyu Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.22298 · **PDF:** https://arxiv.org/pdf/2602.22298v1
+
+## 9763. NoLan: Mitigating Object Hallucinations in Large Vision-Language Models via Dynamic Suppression of Language Priors
+
+- **ID:** `2602.22144` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Lingfeng Ren, Weihao Yu, Runpeng Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.22144 · **PDF:** https://arxiv.org/pdf/2602.22144v1
+
+## 9764. SigmaQuant: Hardware-Aware Heterogeneous Quantization Method for Edge DNN Inference
+
+- **ID:** `2602.22136` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Qunyou Liu, Pengbo Yu, Marina Zapater, et al.
+- **abs:** https://arxiv.org/abs/2602.22136 · **PDF:** https://arxiv.org/pdf/2602.22136v2
+
+## 9765. Sample Complexity Bounds for Robust Mean Estimation with Mean-Shift Contamination
+
+- **ID:** `2602.22130` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Ilias Diakonikolas, Giannis Iakovidis, Daniel M. Kane, et al.
+- **abs:** https://arxiv.org/abs/2602.22130 · **PDF:** https://arxiv.org/pdf/2602.22130v1
+
+## 9766. IndicIFEval: A Benchmark for Verifiable Instruction-Following Evaluation in 14 Indic Languages
+
+- **ID:** `2602.22125` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Thanmay Jayakumar, Mohammed Safi Ur Rahman Khan, Raj Dabre, et al.
+- **abs:** https://arxiv.org/abs/2602.22125 · **PDF:** https://arxiv.org/pdf/2602.22125v1
+
+## 9767. SWE-Protégé: Learning to Selectively Collaborate With an Expert Unlocks Small Language Models as Software Engineering Agents
+
+- **ID:** `2602.22124` · **primary:** `cs.SE` · **Submitted:** 2026-02-25
+- **Authors:** Patrick Tser Jern Kon, Archana Pradeep, Ang Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.22124 · **PDF:** https://arxiv.org/pdf/2602.22124v1
+
+## 9768. Probing the Geometry of Diffusion Models with the String Method
+
+- **ID:** `2602.22122` · **primary:** `stat.ML` · **Submitted:** 2026-02-25
+- **Authors:** Elio Moreau, Florentin Coeurdoux, Grégoire Ferre, et al.
+- **abs:** https://arxiv.org/abs/2602.22122 · **PDF:** https://arxiv.org/pdf/2602.22122v1
+
+## 9769. EfficientPosterGen: Semantic-aware Efficient Poster Generation via Token Compression and Accurate Violation Detection
+
+- **ID:** `2603.00155` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Wenxin Tang, Jingyu Xiao, Yanpei Gong, et al.
+- **abs:** https://arxiv.org/abs/2603.00155 · **PDF:** https://arxiv.org/pdf/2603.00155v1
+
+## 9770. Slice and Explain: Logic-Based Explanations for Neural Networks through Domain Slicing
+
+- **ID:** `2602.22115` · **primary:** `cs.LO` · **Submitted:** 2026-02-25
+- **Authors:** Luiz Fernando Paulino Queiroz, Carlos Henrique Leitão Cavalcante, Thiago Alves Rocha
+- **abs:** https://arxiv.org/abs/2602.22115 · **PDF:** https://arxiv.org/pdf/2602.22115v1
+
+## 9771. Don't stop me now: Rethinking Validation Criteria for Model Parameter Selection
+
+- **ID:** `2602.22107` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Andrea Apicella, Francesco Isgrò, Andrea Pollastro, et al.
+- **abs:** https://arxiv.org/abs/2602.22107 · **PDF:** https://arxiv.org/pdf/2602.22107v1
+
+## 9772. On Imbalanced Regression with Hoeffding Trees
+
+- **ID:** `2602.22101` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Pantia-Marina Alchirch, Dimitrios I. Diochnos
+- **abs:** https://arxiv.org/abs/2602.22101 · **PDF:** https://arxiv.org/pdf/2602.22101v2
+
+## 9773. Petri Net Relaxation for Infeasibility Explanation and Sequential Task Planning
+
+- **ID:** `2602.22094` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Nguyen Cong Nhat Le, John G. Rogers, Claire N. Bonial, et al.
+- **abs:** https://arxiv.org/abs/2602.22094 · **PDF:** https://arxiv.org/pdf/2602.22094v1
+
+## 9774. Confidence-Driven Multi-Scale Model Selection for Cost-Efficient Inference
+
+- **ID:** `2602.22090` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Bo-Wei Chen, Chung-Chi Chen, An-Zi Yen
+- **abs:** https://arxiv.org/abs/2602.22090 · **PDF:** https://arxiv.org/pdf/2602.22090v1
+
+## 9775. MBD-ML: Many-body dispersion from machine learning for molecules and materials
+
+- **ID:** `2602.22086` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-25
+- **Authors:** Evgeny Moerman, Adil Kabylda, Almaz Khabibrakhmanov, et al.
+- **abs:** https://arxiv.org/abs/2602.22086 · **PDF:** https://arxiv.org/pdf/2602.22086v1
+
+## 9776. Coarsening Bias from Variable Discretization in Causal Functionals
+
+- **ID:** `2602.22083` · **primary:** `stat.ME` · **Submitted:** 2026-02-25
+- **Authors:** Xiaxian Ou, Razieh Nabi
+- **abs:** https://arxiv.org/abs/2602.22083 · **PDF:** https://arxiv.org/pdf/2602.22083v1
+
+## 9777. DECODE: Dual-Enhanced Conditioned Diffusion for EEG Forecasting
+
+- **ID:** `2603.16885` · **primary:** `eess.SP` · **Submitted:** 2026-02-25
+- **Authors:** Mehran Shabanpour, Sadaf Khademi, Konstantinos N Plataniotis, et al.
+- **abs:** https://arxiv.org/abs/2603.16885 · **PDF:** https://arxiv.org/pdf/2603.16885v1
+
+## 9778. Speculative Decoding Scaling Laws (SDSL): Throughput Optimization Made Simple
+
+- **ID:** `2603.11053` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Amirhossein Bozorgkhoo, Igor Molybog
+- **abs:** https://arxiv.org/abs/2603.11053 · **PDF:** https://arxiv.org/pdf/2603.11053v1
+
+## 9779. Understanding Artificial Theory of Mind: Perturbed Tasks and Reasoning in Large Language Models
+
+- **ID:** `2602.22072` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Christian Nickel, Laura Schrewe, Florian Mai, et al.
+- **abs:** https://arxiv.org/abs/2602.22072 · **PDF:** https://arxiv.org/pdf/2602.22072v1
+
+## 9780. Language Models Exhibit Inconsistent Biases Towards Algorithmic Agents and Human Experts
+
+- **ID:** `2602.22070` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Jessica Y. Bo, Lillio Mok, Ashton Anderson
+- **abs:** https://arxiv.org/abs/2602.22070 · **PDF:** https://arxiv.org/pdf/2602.22070v1
+
+## 9781. Semantic Partial Grounding via LLMs
+
+- **ID:** `2602.22067` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Giuseppe Canonaco, Alberto Pozanco, Daniel Borrajo
+- **abs:** https://arxiv.org/abs/2602.22067 · **PDF:** https://arxiv.org/pdf/2602.22067v1
+
+## 9782. DualWeaver: Synergistic Feature Weaving Surrogates for Multivariate Forecasting with Univariate Time Series Foundation Models
+
+- **ID:** `2602.22066` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Jinpeng Li, Zhongyi Pei, Huaze Xue, et al.
+- **abs:** https://arxiv.org/abs/2602.22066 · **PDF:** https://arxiv.org/pdf/2602.22066v1
+
+## 9783. Learning Quantum Data Distribution via Chaotic Quantum Diffusion Model
+
+- **ID:** `2602.22061` · **primary:** `quant-ph` · **Submitted:** 2026-02-25
+- **Authors:** Quoc Hoan Tran, Koki Chinzei, Yasuhiro Endo, et al.
+- **abs:** https://arxiv.org/abs/2602.22061 · **PDF:** https://arxiv.org/pdf/2602.22061v2
+
+## 9784. NESTOR: A Nested MOE-based Neural Operator for Large-Scale PDE Pre-Training
+
+- **ID:** `2602.22059` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Dengdi Sun, Xiaoya Zhou, Xiao Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.22059 · **PDF:** https://arxiv.org/pdf/2602.22059v1
+
+## 9785. FlowCorrect: Efficient Interactive Correction of Generative Flow Policies for Robotic Manipulation
+
+- **ID:** `2602.22056` · **primary:** `cs.RO` · **Submitted:** 2026-02-25
+- **Authors:** Edgar Welte, Yitian Shi, Rosa Wolf, et al.
+- **abs:** https://arxiv.org/abs/2602.22056 · **PDF:** https://arxiv.org/pdf/2602.22056v2
+
+## 9786. Physics-Informed Machine Learning for Vessel Shaft Power and Fuel Consumption Prediction: Interpretable KAN-based Approach
+
+- **ID:** `2602.22055` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Hamza Haruna Mohammed, Dusica Marijan, Arnbjørn Maressa
+- **abs:** https://arxiv.org/abs/2602.22055 · **PDF:** https://arxiv.org/pdf/2602.22055v1
+
+## 9787. DLT-Corpus: A Large-Scale Text Collection for the Distributed Ledger Technology Domain
+
+- **ID:** `2602.22045` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Walter Hernandez Cruz, Peter Devine, Nikhil Vadgama, et al.
+- **abs:** https://arxiv.org/abs/2602.22045 · **PDF:** https://arxiv.org/pdf/2602.22045v1
+
+## 9788. TG-ASR: Translation-Guided Learning with Parallel Gated Cross Attention for Low-Resource Automatic Speech Recognition
+
+- **ID:** `2602.22039` · **primary:** `eess.AS` · **Submitted:** 2026-02-25
+- **Authors:** Cheng-Yeh Yang, Chien-Chun Wang, Li-Wei Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.22039 · **PDF:** https://arxiv.org/pdf/2602.22039v1
+
+## 9789. HELIOS: Harmonizing Early Fusion, Late Fusion, and LLM Reasoning for Multi-Granular Table-Text Retrieval
+
+- **ID:** `2603.02248` · **primary:** `cs.DB` · **Submitted:** 2026-02-25
+- **Authors:** Sungho Park, Joohyung Yun, Jongwuk Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.02248 · **PDF:** https://arxiv.org/pdf/2603.02248v1
+
+## 9790. LARFT: Closing the Cognition-Action Gap for Length Instruction Following in Large Language Models
+
+- **ID:** `2603.19255` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Wei Zhang, Lintong Du, Yuanhe Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.19255 · **PDF:** https://arxiv.org/pdf/2603.19255v1
+
+## 9791. Learning Rewards, Not Labels: Adversarial Inverse Reinforcement Learning for Machinery Fault Detection
+
+- **ID:** `2602.22297` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Dhiraj Neupane, Richard Dazeley, Mohamed Reda Bouadjenek, et al.
+- **abs:** https://arxiv.org/abs/2602.22297 · **PDF:** https://arxiv.org/pdf/2602.22297v1
+
+## 9792. RGB-Event HyperGraph Prompt for Kilometer Marker Recognition based on Pre-trained Foundation Models
+
+- **ID:** `2602.22026` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Xiaoyu Xian, Shiao Wang, Xiao Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.22026 · **PDF:** https://arxiv.org/pdf/2602.22026v1
+
+## 9793. UpSkill: Mutual Information Skill Learning for Structured Response Diversity in LLMs
+
+- **ID:** `2602.22296` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Devan Shah, Owen Yang, Daniel Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.22296 · **PDF:** https://arxiv.org/pdf/2602.22296v1
+
+## 9794. Disease Progression and Subtype Modeling for Combined Discrete and Continuous Input Data
+
+- **ID:** `2602.22018` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Sterre de Jonge, Elisabeth J. Vinke, Meike W. Vernooij, et al.
+- **abs:** https://arxiv.org/abs/2602.22018 · **PDF:** https://arxiv.org/pdf/2602.22018v1
+
+## 9795. Function-Space Empirical Bayes Regularisation with Student's t Priors
+
+- **ID:** `2602.22015` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Pengcheng Hao, Ercan Engin Kuruoglu
+- **abs:** https://arxiv.org/abs/2602.22015 · **PDF:** https://arxiv.org/pdf/2602.22015v1
+
+## 9796. A Diversity Diet for a Healthier Model: A Case Study of French ModernBERT
+
+- **ID:** `2602.22014` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Louis Estève, Christophe Servan, Thomas Lavergne, et al.
+- **abs:** https://arxiv.org/abs/2602.22014 · **PDF:** https://arxiv.org/pdf/2602.22014v1
+
+## 9797. When Should a Model Change Its Mind? An Energy-Based Theory and Regularizer for Concept Drift in Electrocardiogram (ECG) Signals
+
+- **ID:** `2602.22294` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Timothy Oladunni, Blessing Ojeme, Kyndal Maclin, et al.
+- **abs:** https://arxiv.org/abs/2602.22294 · **PDF:** https://arxiv.org/pdf/2602.22294v2
+
+## 9798. Global River Forecasting with a Topology-Informed AI Foundation Model
+
+- **ID:** `2602.22293` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Hancheng Ren, Gang Zhao, Shuo Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.22293 · **PDF:** https://arxiv.org/pdf/2602.22293v1
+
+## 9799. Neural solver for Wasserstein Geodesics and optimal transport dynamics
+
+- **ID:** `2602.22003` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Hailiang Liu, Yan-Han Chen
+- **abs:** https://arxiv.org/abs/2602.22003 · **PDF:** https://arxiv.org/pdf/2602.22003v1
+
+## 9800. Enhancing LLM-Based Test Generation by Eliminating Covered Code
+
+- **ID:** `2602.21997` · **primary:** `cs.SE` · **Submitted:** 2026-02-25
+- **Authors:** WeiZhe Xu, Mengyu Liu, Fanxin Kong
+- **abs:** https://arxiv.org/abs/2602.21997 · **PDF:** https://arxiv.org/pdf/2602.21997v1
+
+## 9801. Outpatient Appointment Scheduling Optimization with a Genetic Algorithm Approach
+
+- **ID:** `2602.21995` · **primary:** `cs.NE` · **Submitted:** 2026-02-25
+- **Authors:** Ana Rodrigues, Rui Rego
+- **abs:** https://arxiv.org/abs/2602.21995 · **PDF:** https://arxiv.org/pdf/2602.21995v1
+
+## 9802. Solving stiff dark matter equations via Jacobian Normalization with Physics-Informed Neural Networks
+
+- **ID:** `2602.21988` · **primary:** `hep-ph` · **Submitted:** 2026-02-25
+- **Authors:** M. P. Bento, H. B. Câmara, J. R. Rocha, et al.
+- **abs:** https://arxiv.org/abs/2602.21988 · **PDF:** https://arxiv.org/pdf/2602.21988v1
+
+## 9803. PatchDenoiser: Parameter-efficient multi-scale patch learning and fusion denoiser for Low-dose CT imaging
+
+- **ID:** `2602.21987` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Jitindra Fartiyal, Pedro Freire, Sergei K. Turitsyn, et al.
+- **abs:** https://arxiv.org/abs/2602.21987 · **PDF:** https://arxiv.org/pdf/2602.21987v2
+
+## 9804. Manifold of Failure: Behavioral Attraction Basins in Language Models
+
+- **ID:** `2602.22291` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Sarthak Munshi, Manish Bhatt, Vineeth Sai Narajala, et al.
+- **abs:** https://arxiv.org/abs/2602.22291 · **PDF:** https://arxiv.org/pdf/2602.22291v2
+
+## 9805. Pulse-Driven Neural Architecture: Learnable Oscillatory Dynamics for Robust Continuous-Time Sequence Processing
+
+- **ID:** `2603.00153` · **primary:** `cs.NE` · **Submitted:** 2026-02-25
+- **Authors:** Paras Sharma
+- **abs:** https://arxiv.org/abs/2603.00153 · **PDF:** https://arxiv.org/pdf/2603.00153v1
+
+## 9806. CxMP: A Linguistic Minimal-Pair Benchmark for Evaluating Constructional Understanding in Language Models
+
+- **ID:** `2602.21978` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Miyu Oba, Saku Sugawara
+- **abs:** https://arxiv.org/abs/2602.21978 · **PDF:** https://arxiv.org/pdf/2602.21978v1
+
+## 9807. Compact Circulant Layers with Spectral Priors
+
+- **ID:** `2602.21965` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Joseph Margaryan, Thomas Hamelryck
+- **abs:** https://arxiv.org/abs/2602.21965 · **PDF:** https://arxiv.org/pdf/2602.21965v1
+
+## 9808. Robustness in sparse artificial neural networks trained with adaptive topology
+
+- **ID:** `2602.21961` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Bendegúz Sulyok, Gergely Palla, Filippo Radicchi, et al.
+- **abs:** https://arxiv.org/abs/2602.21961 · **PDF:** https://arxiv.org/pdf/2602.21961v1
+
+## 9809. Estimation and Optimization of Ship Fuel Consumption in Maritime: Review, Challenges and Future Directions
+
+- **ID:** `2602.21959` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Dusica Marijan, Hamza Haruna Mohammed, Bakht Zaman
+- **abs:** https://arxiv.org/abs/2602.21959 · **PDF:** https://arxiv.org/pdf/2602.21959v1
+
+## 9810. Learning to Collaborate via Structures: Cluster-Guided Item Alignment for Federated Recommendation
+
+- **ID:** `2602.21957` · **primary:** `cs.IR` · **Submitted:** 2026-02-25
+- **Authors:** Yuchun Tu, Zhiwei Li, Bingli Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.21957 · **PDF:** https://arxiv.org/pdf/2602.21957v1
+
+## 9811. RADAR: Reasoning as Discrimination with Aligned Representations for LLM-based Knowledge Graph Reasoning
+
+- **ID:** `2602.21951` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Bo Xue, Yuan Jin, Luoyi Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.21951 · **PDF:** https://arxiv.org/pdf/2602.21951v1
+
+## 9812. MEDSYN: Benchmarking Multi-EviDence SYNthesis in Complex Clinical Cases for Multimodal Large Language Models
+
+- **ID:** `2602.21950` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Boqi Chen, Xudong Liu, Jiachuan Peng, et al.
+- **abs:** https://arxiv.org/abs/2602.21950 · **PDF:** https://arxiv.org/pdf/2602.21950v1
+
+## 9813. What Topological and Geometric Structure Do Biological Foundation Models Learn? Evidence from 141 Hypotheses
+
+- **ID:** `2602.22289` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-25
+- **Authors:** Ihor Kendiukhov
+- **abs:** https://arxiv.org/abs/2602.22289 · **PDF:** https://arxiv.org/pdf/2602.22289v2
+
+## 9814. Bayesian Generative Adversarial Networks via Gaussian Approximation for Tabular Data Synthesis
+
+- **ID:** `2602.21948` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Bahrul Ilmi Nasution, Mark Elliot, Richard Allmendinger
+- **abs:** https://arxiv.org/abs/2602.21948 · **PDF:** https://arxiv.org/pdf/2602.21948v1
+
+## 9815. Large Language Models are Algorithmically Blind
+
+- **ID:** `2602.21947` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Sohan Venkatesh, Ashish Mahendran Kurapath, Tejas Melkote
+- **abs:** https://arxiv.org/abs/2602.21947 · **PDF:** https://arxiv.org/pdf/2602.21947v3
+
+## 9816. Hidden Topics: Measuring Sensitive AI Beliefs with List Experiments
+
+- **ID:** `2602.21939` · **primary:** `cs.CY` · **Submitted:** 2026-02-25
+- **Authors:** Maxim Chupilkin
+- **abs:** https://arxiv.org/abs/2602.21939 · **PDF:** https://arxiv.org/pdf/2602.21939v1
+
+## 9817. Reliable XAI Explanations in Sudden Cardiac Death Prediction for Chagas Cardiomyopathy
+
+- **ID:** `2602.22288` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Vinícius P. Chagas, Luiz H. T. Viana, Mac M. da S. Carlos, et al.
+- **abs:** https://arxiv.org/abs/2602.22288 · **PDF:** https://arxiv.org/pdf/2602.22288v1
+
+## 9818. A Framework for Cross-Domain Generalization in Coronary Artery Calcium Scoring Across Gated and Non-Gated Computed Tomography
+
+- **ID:** `2602.21935` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Mahmut S. Gokmen, Moneera N. Haque, Steve W. Leung, et al.
+- **abs:** https://arxiv.org/abs/2602.21935 · **PDF:** https://arxiv.org/pdf/2602.21935v1
+
+## 9819. Multi-Level Causal Embeddings
+
+- **ID:** `2602.22287` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Willem Schooltink, Fabio Massimo Zennaro
+- **abs:** https://arxiv.org/abs/2602.22287 · **PDF:** https://arxiv.org/pdf/2602.22287v2
+
+## 9820. Small Wins Big: Comparing Large Language Models and Domain Fine-Tuned Models for Sarcasm Detection in Code-Mixed Hinglish Text
+
+- **ID:** `2602.21933` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Bitan Majumder, Anirban Sen
+- **abs:** https://arxiv.org/abs/2602.21933 · **PDF:** https://arxiv.org/pdf/2602.21933v1
+
+## 9821. Learning Unknown Interdependencies for Decentralized Root Cause Analysis in Nonlinear Dynamical Systems
+
+- **ID:** `2602.21928` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Ayush Mohanty, Paritosh Ramanan, Nagi Gebraeel
+- **abs:** https://arxiv.org/abs/2602.21928 · **PDF:** https://arxiv.org/pdf/2602.21928v1
+
+## 9822. Bridging Through Absence: How Comeback Researchers Bridge Knowledge Gaps Through Structural Re-emergence
+
+- **ID:** `2602.21926` · **primary:** `cs.SI` · **Submitted:** 2026-02-25
+- **Authors:** Somyajit Chakraborty, Angshuman Jana, Avijit Gayen
+- **abs:** https://arxiv.org/abs/2602.21926 · **PDF:** https://arxiv.org/pdf/2602.21926v1
+
+## 9823. Learning in the Null Space: Small Singular Values for Continual Learning
+
+- **ID:** `2602.21919` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Cuong Anh Pham, Praneeth Vepakomma, Samuel Horváth
+- **abs:** https://arxiv.org/abs/2602.21919 · **PDF:** https://arxiv.org/pdf/2602.21919v1
+
+## 9824. OmniZip: Learning a Unified and Lightweight Lossless Compressor for Multi-Modal Data
+
+- **ID:** `2602.22286` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Yan Zhao, Zhengxue Cheng, Junxuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.22286 · **PDF:** https://arxiv.org/pdf/2602.22286v2
+
+## 9825. From Comprehension to Reasoning: A Hierarchical Benchmark for Automated Financial Research Reporting
+
+- **ID:** `2603.19254` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Yiyun Zhu, Yidong Jiang, Ziwen Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.19254 · **PDF:** https://arxiv.org/pdf/2603.19254v1
+
+## 9826. The Error of Deep Operator Networks Is the Sum of Its Parts: Branch-Trunk and Mode Error Decompositions
+
+- **ID:** `2602.21910` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Alexander Heinlein, Johannes Taraz
+- **abs:** https://arxiv.org/abs/2602.21910 · **PDF:** https://arxiv.org/pdf/2602.21910v1
+
+## 9827. OnDA: On-device Channel Pruning for Efficient Personalized Keyword Spotting
+
+- **ID:** `2603.02247` · **primary:** `eess.AS` · **Submitted:** 2026-02-25
+- **Authors:** Matteo Risso, Alessio Burrello, Daniele Jahier Pagliari
+- **abs:** https://arxiv.org/abs/2603.02247 · **PDF:** https://arxiv.org/pdf/2603.02247v1
+
+## 9828. Early Risk Stratification of Dosing Errors in Clinical Trials Using Machine Learning
+
+- **ID:** `2602.22285` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Félicien Hêche, Sohrab Ferdowsi, Anthony Yazdani, et al.
+- **abs:** https://arxiv.org/abs/2602.22285 · **PDF:** https://arxiv.org/pdf/2602.22285v1
+
+## 9829. 2-Step Agent: A Framework for the Interaction of a Decision Maker with AI Decision Support
+
+- **ID:** `2602.21889` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Otto Nyberg, Fausto Carcassi, Giovanni Cinà
+- **abs:** https://arxiv.org/abs/2602.21889 · **PDF:** https://arxiv.org/pdf/2602.21889v1
+
+## 9830. ExpLang: Improved Exploration and Exploitation in LLM Reasoning with On-Policy Thinking Language Selection
+
+- **ID:** `2602.21887` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Changjiang Gao, Zixian Huang, Kaichen Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.21887 · **PDF:** https://arxiv.org/pdf/2602.21887v1
+
+## 9831. GFPL: Generative Federated Prototype Learning for Resource-Constrained and Data-Imbalanced Vision Task
+
+- **ID:** `2602.21873` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Shiwei Lu, Yuhang He, Jiashuo Li, et al.
+- **abs:** https://arxiv.org/abs/2602.21873 · **PDF:** https://arxiv.org/pdf/2602.21873v1
+
+## 9832. DynamicGTR: Leveraging Graph Topology Representation Preferences to Boost VLM Capabilities on Graph QAs
+
+- **ID:** `2602.21864` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Yanbin Wei, Jiangyue Yan, Chun Kang, et al.
+- **abs:** https://arxiv.org/abs/2602.21864 · **PDF:** https://arxiv.org/pdf/2602.21864v1
+
+## 9833. BrepCoder: A Unified Multimodal Large Language Model for Multi-task B-rep Reasoning
+
+- **ID:** `2602.22284` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Mingi Kim, Yongjun Kim, Jungwoo Kang, et al.
+- **abs:** https://arxiv.org/abs/2602.22284 · **PDF:** https://arxiv.org/pdf/2602.22284v2
+
+## 9834. Personalized Graph-Empowered Large Language Model for Proactive Information Access
+
+- **ID:** `2602.21862` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Chia Cheng Chang, An-Zi Yen, Hen-Hsen Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.21862 · **PDF:** https://arxiv.org/pdf/2602.21862v1
+
+## 9835. ProactiveMobile: A Comprehensive Benchmark for Boosting Proactive Intelligence on Mobile Devices
+
+- **ID:** `2602.21858` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Dezhi Kong, Zhengzhao Feng, Qiliang Liang, et al.
+- **abs:** https://arxiv.org/abs/2602.21858 · **PDF:** https://arxiv.org/pdf/2602.21858v3
+
+## 9836. Distill and Align Decomposition for Enhanced Claim Verification
+
+- **ID:** `2602.21857` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Jabez Magomere, Elena Kochkina, Samuel Mensah, et al.
+- **abs:** https://arxiv.org/abs/2602.21857 · **PDF:** https://arxiv.org/pdf/2602.21857v1
+
+## 9837. Understanding Annotation Error Propagation and Learning an Adaptive Policy for Expert Intervention in Barrett's Video Segmentation
+
+- **ID:** `2602.21855` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Lokesha Rasanjalee, Jin Lin Tan, Dileepa Pitawela, et al.
+- **abs:** https://arxiv.org/abs/2602.21855 · **PDF:** https://arxiv.org/pdf/2602.21855v1
+
+## 9838. FewMMBench: A Benchmark for Multimodal Few-Shot Learning
+
+- **ID:** `2602.21854` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Mustafa Dogan, Ilker Kesen, Iacer Calixto, et al.
+- **abs:** https://arxiv.org/abs/2602.21854 · **PDF:** https://arxiv.org/pdf/2602.21854v1
+
+## 9839. Scalable Kernel-Based Distances for Statistical Inference and Integration
+
+- **ID:** `2602.21846` · **primary:** `stat.ML` · **Submitted:** 2026-02-25
+- **Authors:** Masha Naslidnyk
+- **abs:** https://arxiv.org/abs/2602.21846 · **PDF:** https://arxiv.org/pdf/2602.21846v1
+
+## 9840. xai-cola: A Python library for sparsifying counterfactual explanations
+
+- **ID:** `2602.21845` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Lin Zhu, Lei You
+- **abs:** https://arxiv.org/abs/2602.21845 · **PDF:** https://arxiv.org/pdf/2602.21845v1
+
+## 9841. JSAM: Privacy Straggler-Resilient Joint Client Selection and Incentive Mechanism Design in Differentially Private Federated Learning
+
+- **ID:** `2602.21844` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Ruichen Xu, Ying-Jun Angela Zhang, Jianwei Huang
+- **abs:** https://arxiv.org/abs/2602.21844 · **PDF:** https://arxiv.org/pdf/2602.21844v1
+
+## 9842. Differentially Private Truncation of Unbounded Data via Public Second Moments
+
+- **ID:** `2602.22282` · **primary:** `cs.CR` · **Submitted:** 2026-02-25
+- **Authors:** Zilong Cao, Xuan Bi, Hai Zhang
+- **abs:** https://arxiv.org/abs/2602.22282 · **PDF:** https://arxiv.org/pdf/2602.22282v1
+
+## 9843. Resilient Federated Chain: Transforming Blockchain Consensus into an Active Defense Layer for Federated Learning
+
+- **ID:** `2602.21841` · **primary:** `cs.CR` · **Submitted:** 2026-02-25
+- **Authors:** Mario García-Márquez, Nuria Rodríguez-Barroso, M. Victoria Luzón, et al.
+- **abs:** https://arxiv.org/abs/2602.21841 · **PDF:** https://arxiv.org/pdf/2602.21841v1
+
+## 9844. StoryMovie: A Dataset for Semantic Alignment of Visual Stories with Movie Scripts and Subtitles
+
+- **ID:** `2602.21829` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Daniel Oliveira, David Martins de Matos
+- **abs:** https://arxiv.org/abs/2602.21829 · **PDF:** https://arxiv.org/pdf/2602.21829v1
+
+## 9845. DocDjinn: Controllable Synthetic Document Generation with VLMs and Handwriting Diffusion
+
+- **ID:** `2602.21824` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Marcel Lamott, Saifullah Saifullah, Nauman Riaz, et al.
+- **abs:** https://arxiv.org/abs/2602.21824 · **PDF:** https://arxiv.org/pdf/2602.21824v1
+
+## 9846. SemVideo: Reconstructs What You Watch from Brain Activity via Hierarchical Semantic Guidance
+
+- **ID:** `2602.21819` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Minghan Yang, Lan Yang, Ke Li, et al.
+- **abs:** https://arxiv.org/abs/2602.21819 · **PDF:** https://arxiv.org/pdf/2602.21819v2
+
+## 9847. Prompt Architecture Determines Reasoning Quality: A Variable Isolation Study on the Car Wash Problem
+
+- **ID:** `2602.21814` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Heejin Jo
+- **abs:** https://arxiv.org/abs/2602.21814 · **PDF:** https://arxiv.org/pdf/2602.21814v2
+
+## 9848. An Evaluation of Context Length Extrapolation in Long Code via Positional Embeddings and Efficient Attention
+
+- **ID:** `2602.21800` · **primary:** `cs.SE` · **Submitted:** 2026-02-25
+- **Authors:** Madhusudan Ghosh, Rishabh Gupta
+- **abs:** https://arxiv.org/abs/2602.21800 · **PDF:** https://arxiv.org/pdf/2602.21800v1
+
+## 9849. Excitation: Momentum For Experts
+
+- **ID:** `2602.21798` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Sagi Shaier
+- **abs:** https://arxiv.org/abs/2602.21798 · **PDF:** https://arxiv.org/pdf/2602.21798v1
+
+## 9850. Neural Learning of Fast Matrix Multiplication Algorithms: A StrassenNet Approach
+
+- **ID:** `2602.21797` · **primary:** `math.AG` · **Submitted:** 2026-02-25
+- **Authors:** Paolo Andreini, Alessandra Bernardi, Monica Bianchini, et al.
+- **abs:** https://arxiv.org/abs/2602.21797 · **PDF:** https://arxiv.org/pdf/2602.21797v1
+
+## 9851. A comprehensive study of LLM-based argument classification: from Llama through DeepSeek to GPT-5.2
+
+- **ID:** `2603.19253` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Marcin Pietroń, Filip Gampel, Jakub Gomułka, et al.
+- **abs:** https://arxiv.org/abs/2603.19253 · **PDF:** https://arxiv.org/pdf/2603.19253v1
+
+## 9852. DHP: Efficient Scaling of MLLM Training with Dynamic Hybrid Parallelism
+
+- **ID:** `2602.21788` · **primary:** `cs.DC` · **Submitted:** 2026-02-25
+- **Authors:** Yifan Niu, Han Xiao, Dongyi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.21788 · **PDF:** https://arxiv.org/pdf/2602.21788v1
+
+## 9853. D-COT: Disciplined Chain-of-Thought Learning for Efficient Reasoning in Small Language Models
+
+- **ID:** `2602.21786` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Shunsuke Ubukata
+- **abs:** https://arxiv.org/abs/2602.21786 · **PDF:** https://arxiv.org/pdf/2602.21786v1
+
+## 9854. Therapist-Robot-Patient Physical Interaction is Worth a Thousand Words: Enabling Intuitive Therapist Guidance via Remote Haptic Control
+
+- **ID:** `2602.21783` · **primary:** `cs.RO` · **Submitted:** 2026-02-25
+- **Authors:** Beatrice Luciani, Alex van den Berg, Matti Lang, et al.
+- **abs:** https://arxiv.org/abs/2602.21783 · **PDF:** https://arxiv.org/pdf/2602.21783v1
+
+## 9855. Beyond Static Artifacts: A Forensic Benchmark for Video Deepfake Reasoning in Vision Language Models
+
+- **ID:** `2602.21779` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Zheyuan Gu, Qingsong Zhao, Yusong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.21779 · **PDF:** https://arxiv.org/pdf/2602.21779v1
+
+## 9856. Integrating Machine Learning Ensembles and Large Language Models for Heart Disease Prediction Using Voting Fusion
+
+- **ID:** `2602.22280` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Md. Tahsin Amin, Tanim Ahmmod, Zannatul Ferdus, et al.
+- **abs:** https://arxiv.org/abs/2602.22280 · **PDF:** https://arxiv.org/pdf/2602.22280v1
+
+## 9857. Easy to Learn, Yet Hard to Forget: Towards Robust Unlearning Under Bias
+
+- **ID:** `2602.21773` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** JuneHyoung Kwon, MiHyeon Kim, Eunju Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.21773 · **PDF:** https://arxiv.org/pdf/2602.21773v1
+
+## 9858. UniWhisper: Efficient Continual Multi-task Training for Robust Universal Audio Representation
+
+- **ID:** `2602.21772` · **primary:** `cs.SD` · **Submitted:** 2026-02-25
+- **Authors:** Yuxuan Chen, Peize He, Haoyuan Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.21772 · **PDF:** https://arxiv.org/pdf/2602.21772v2
+
+## 9859. Learning to reconstruct from saturated data: audio declipping and high-dynamic range imaging
+
+- **ID:** `2602.22279` · **primary:** `eess.IV` · **Submitted:** 2026-02-25
+- **Authors:** Victor Sechaud, Laurent Jacques, Patrice Abry, et al.
+- **abs:** https://arxiv.org/abs/2602.22279 · **PDF:** https://arxiv.org/pdf/2602.22279v1
+
+## 9860. RAMSeS: Robust and Adaptive Model Selection for Time-Series Anomaly Detection Algorithms
+
+- **ID:** `2602.21766` · **primary:** `cs.DB` · **Submitted:** 2026-02-25
+- **Authors:** Mohamed Abdelmaksoud, Sheng Ding, Andrey Morozov, et al.
+- **abs:** https://arxiv.org/abs/2602.21766 · **PDF:** https://arxiv.org/pdf/2602.21766v1
+
+## 9861. Generalisation of RLHF under Reward Shift and Clipped KL Regularisation
+
+- **ID:** `2602.21765` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Kenton Tang, Yuzhu Chen, Fengxiang He
+- **abs:** https://arxiv.org/abs/2602.21765 · **PDF:** https://arxiv.org/pdf/2602.21765v1
+
+## 9862. RETLLM: Training and Data-Free MLLMs for Multimodal Information Retrieval
+
+- **ID:** `2602.22278` · **primary:** `cs.IR` · **Submitted:** 2026-02-25
+- **Authors:** Dawei Su, Dongsheng Wang
+- **abs:** https://arxiv.org/abs/2602.22278 · **PDF:** https://arxiv.org/pdf/2602.22278v1
+
+## 9863. Improving Implicit Discourse Relation Recognition with Natural Language Explanations from LLMs
+
+- **ID:** `2602.21763` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Heng Wang, Changxing Wu
+- **abs:** https://arxiv.org/abs/2602.21763 · **PDF:** https://arxiv.org/pdf/2602.21763v1
+
+## 9864. Survey on Neural Routing Solvers
+
+- **ID:** `2602.21761` · **primary:** `math.OC` · **Submitted:** 2026-02-25
+- **Authors:** Yunpeng Ba, Xi Lin, Changliang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.21761 · **PDF:** https://arxiv.org/pdf/2602.21761v2
+
+## 9865. X-REFINE: XAI-based RElevance input-Filtering and archItecture fiNe-tuning for channel Estimation
+
+- **ID:** `2602.22277` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Abdul Karim Gizzini, Yahia Medjahdi
+- **abs:** https://arxiv.org/abs/2602.22277 · **PDF:** https://arxiv.org/pdf/2602.22277v1
+
+## 9866. Learning from Yesterday's Error: An Efficient Online Learning Method for Traffic Demand Prediction
+
+- **ID:** `2602.21757` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Xiannan Huang, Quan Yuan, Chao Yang
+- **abs:** https://arxiv.org/abs/2602.21757 · **PDF:** https://arxiv.org/pdf/2602.21757v1
+
+## 9867. Offline Reasoning for Efficient Recommendation: LLM-Empowered Persona-Profiled Item Indexing
+
+- **ID:** `2602.21756` · **primary:** `cs.IR` · **Submitted:** 2026-02-25
+- **Authors:** Deogyong Kim, Junseong Lee, Jeongeun Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.21756 · **PDF:** https://arxiv.org/pdf/2602.21756v1
+
+## 9868. From Words to Amino Acids: Does the Curse of Depth Persist?
+
+- **ID:** `2602.21750` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Aleena Siji, Amir Mohammad Karimi Mamaghan, Ferdinand Kapl, et al.
+- **abs:** https://arxiv.org/abs/2602.21750 · **PDF:** https://arxiv.org/pdf/2602.21750v1
+
+## 9869. RABot: Reinforcement-Guided Graph Augmentation for Imbalanced and Noisy Social Bot Detection
+
+- **ID:** `2602.21749` · **primary:** `cs.SI` · **Submitted:** 2026-02-25
+- **Authors:** Longlong Zhang, Xi Wang, Haotong Du, et al.
+- **abs:** https://arxiv.org/abs/2602.21749 · **PDF:** https://arxiv.org/pdf/2602.21749v1
+
+## 9870. Explain in Your Own Words: Improving Reasoning via Token-Selective Dual Knowledge Distillation
+
+- **ID:** `2603.13260` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Minsang Kim, Seung Jun Baek
+- **abs:** https://arxiv.org/abs/2603.13260 · **PDF:** https://arxiv.org/pdf/2603.13260v1
+
+## 9871. fEDM+: A Risk-Based Fuzzy Ethical Decision Making Framework with Principle-Level Explainability and Pluralistic Validation
+
+- **ID:** `2602.21746` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Abeer Dyoub, Francesca A. Lisi
+- **abs:** https://arxiv.org/abs/2602.21746 · **PDF:** https://arxiv.org/pdf/2602.21746v2
+
+## 9872. The ASIR Courage Model: A Phase-Dynamic Framework for Truth Transitions in Human and AI Systems
+
+- **ID:** `2602.21745` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Hyo Jin Kim
+- **abs:** https://arxiv.org/abs/2602.21745 · **PDF:** https://arxiv.org/pdf/2602.21745v1
+
+## 9873. Robust Long-Form Bangla Speech Processing: Automatic Speech Recognition and Speaker Diarization
+
+- **ID:** `2602.21741` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** MD. Sagor Chowdhury, Adiba Fairooz Chowdhury
+- **abs:** https://arxiv.org/abs/2602.21741 · **PDF:** https://arxiv.org/pdf/2602.21741v1
+
+## 9874. Deep Accurate Solver for the Geodesic Problem
+
+- **ID:** `2602.22275` · **primary:** `eess.IV` · **Submitted:** 2026-02-25
+- **Authors:** Saar Huberman, Amit Bracha, Ron Kimmel
+- **abs:** https://arxiv.org/abs/2602.22275 · **PDF:** https://arxiv.org/pdf/2602.22275v1
+
+## 9875. Positional-aware Spatio-Temporal Network for Large-Scale Traffic Prediction
+
+- **ID:** `2602.22274` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Runfei Chen
+- **abs:** https://arxiv.org/abs/2602.22274 · **PDF:** https://arxiv.org/pdf/2602.22274v1
+
+## 9876. Explore-on-Graph: Incentivizing Autonomous Exploration of Large Language Models on Knowledge Graphs with Path-refined Reward Modeling
+
+- **ID:** `2602.21728` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Shiqi Yan, Yubo Chen, Ruiqi Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.21728 · **PDF:** https://arxiv.org/pdf/2602.21728v1
+
+## 9877. Private and Robust Contribution Evaluation in Federated Learning
+
+- **ID:** `2602.21721` · **primary:** `cs.CR` · **Submitted:** 2026-02-25
+- **Authors:** Delio Jaramillo Velez, Gergely Biczok, Alexandre Graell i Amat, et al.
+- **abs:** https://arxiv.org/abs/2602.21721 · **PDF:** https://arxiv.org/pdf/2602.21721v1
+
+## 9878. Evaluating the relationship between regularity and learnability in recursive numeral systems using Reinforcement Learning
+
+- **ID:** `2602.21720` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Andrea Silvi, Ponrawee Prasertsom, Jennifer Culbertson, et al.
+- **abs:** https://arxiv.org/abs/2602.21720 · **PDF:** https://arxiv.org/pdf/2602.21720v1
+
+## 9879. Tokenization vs. Augmentation: A Systematic Study of Writer Variance in IMU-Based Online Handwriting Recognition
+
+- **ID:** `2603.16883` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Jindong Li, Dario Zanca, Vincent Christlein, et al.
+- **abs:** https://arxiv.org/abs/2603.16883 · **PDF:** https://arxiv.org/pdf/2603.16883v1
+
+## 9880. C$^{2}$TC: A Training-Free Framework for Efficient Tabular Data Condensation
+
+- **ID:** `2602.21717` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Sijia Xu, Fan Li, Xiaoyang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.21717 · **PDF:** https://arxiv.org/pdf/2602.21717v1
+
+## 9881. Two-Stage Active Distribution Network Voltage Control via LLM-RL Collaboration: A Hybrid Knowledge-Data-Driven Approach
+
+- **ID:** `2602.21715` · **primary:** `eess.SY` · **Submitted:** 2026-02-25
+- **Authors:** Xu Yang, Chenhui Lin, Xiang Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.21715 · **PDF:** https://arxiv.org/pdf/2602.21715v1
+
+## 9882. Learning spatially adaptive sparsity level maps for arbitrary convolutional dictionaries
+
+- **ID:** `2602.21707` · **primary:** `eess.IV` · **Submitted:** 2026-02-25
+- **Authors:** Joshua Schulz, David Schote, Christoph Kolbitsch, et al.
+- **abs:** https://arxiv.org/abs/2602.21707 · **PDF:** https://arxiv.org/pdf/2602.21707v1
+
+## 9883. SurGo-R1: Benchmarking and Modeling Contextual Reasoning for Operative Zone in Surgical Video
+
+- **ID:** `2602.21706` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Guanyi Qin, Xiaozhen Wang, Zhu Zhuo, et al.
+- **abs:** https://arxiv.org/abs/2602.21706 · **PDF:** https://arxiv.org/pdf/2602.21706v1
+
+## 9884. Dynamic Multimodal Activation Steering for Hallucination Mitigation in Large Vision-Language Models
+
+- **ID:** `2602.21704` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Jianghao Yin, Qin Chen, Kedi Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.21704 · **PDF:** https://arxiv.org/pdf/2602.21704v1
+
+## 9885. Brain Tumor Segmentation with Special Emphasis on the Non-Enhancing Brain Tumor Compartment
+
+- **ID:** `2602.21703` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** T. Schaffer, A. Brawanski, S. Wein, et al.
+- **abs:** https://arxiv.org/abs/2602.21703 · **PDF:** https://arxiv.org/pdf/2602.21703v1
+
+## 9886. Learning Complex Physical Regimes via Coverage-oriented Uncertainty Quantification: An application to the Critical Heat Flux
+
+- **ID:** `2602.21701` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Michele Cazzola, Alberto Ghione, Lucia Sargentini, et al.
+- **abs:** https://arxiv.org/abs/2602.21701 · **PDF:** https://arxiv.org/pdf/2602.21701v1
+
+## 9887. FIRE: A Comprehensive Benchmark for Financial Intelligence and Reasoning Evaluation
+
+- **ID:** `2602.22273` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Xiyuan Zhang, Huihang Wu, Jiayu Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.22273 · **PDF:** https://arxiv.org/pdf/2602.22273v1
+
+## 9888. TiMi: Empower Time Series Transformers with Multimodal Mixture of Experts
+
+- **ID:** `2602.21693` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Jiafeng Lin, Yuxuan Wang, Huakun Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.21693 · **PDF:** https://arxiv.org/pdf/2602.21693v1
+
+## 9889. Support Tokens, Stability Margins, and a New Foundation for Robust LLMs
+
+- **ID:** `2602.22271` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Deepak Agarwal, Dhyey Dharmendrakumar Mavani, Suyash Gupta, et al.
+- **abs:** https://arxiv.org/abs/2602.22271 · **PDF:** https://arxiv.org/pdf/2602.22271v3
+
+## 9890. Primary-Fine Decoupling for Action Generation in Robotic Imitation
+
+- **ID:** `2602.21684` · **primary:** `cs.RO` · **Submitted:** 2026-02-25
+- **Authors:** Xiaohan Lei, Min Wang, Wengang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.21684 · **PDF:** https://arxiv.org/pdf/2602.21684v1
+
+## 9891. Hierarchical Lead Critic based Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.21680` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** David Eckel, Henri Meeß
+- **abs:** https://arxiv.org/abs/2602.21680 · **PDF:** https://arxiv.org/pdf/2602.21680v1
+
+## 9892. RECAP: Local Hebbian Prototype Learning as a Self-Organizing Readout for Reservoir Dynamics
+
+- **ID:** `2603.06639` · **primary:** `cs.NE` · **Submitted:** 2026-02-25
+- **Authors:** Heng Zhang
+- **abs:** https://arxiv.org/abs/2603.06639 · **PDF:** https://arxiv.org/pdf/2603.06639v1
+
+## 9893. Trie-Aware Transformers for Generative Recommendation
+
+- **ID:** `2602.21677` · **primary:** `cs.IR` · **Submitted:** 2026-02-25
+- **Authors:** Zhenxiang Xu, Jiawei Chen, Sirui Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.21677 · **PDF:** https://arxiv.org/pdf/2602.21677v1
+
+## 9894. Error-awareness Accelerates Active Automata Learning
+
+- **ID:** `2602.21674` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Loes Kruger, Sebastian Junges, Jurriaan Rot
+- **abs:** https://arxiv.org/abs/2602.21674 · **PDF:** https://arxiv.org/pdf/2602.21674v1
+
+## 9895. GeoChallenge: A Multi-Answer Multiple-Choice Benchmark for Geometric Reasoning with Diagrams
+
+- **ID:** `2603.19252` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Yushun Zhang, Weiping Fu, Zesheng Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.19252 · **PDF:** https://arxiv.org/pdf/2603.19252v1
+
+## 9896. How Transformers Reject Wrong Answers: Rotational Dynamics of Factual Constraint Processing
+
+- **ID:** `2603.13259` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Javier Marín
+- **abs:** https://arxiv.org/abs/2603.13259 · **PDF:** https://arxiv.org/pdf/2603.13259v1
+
+## 9897. Hierarchical LLM-Based Multi-Agent Framework with Prompt Optimization for Multi-Robot Task Planning
+
+- **ID:** `2602.21670` · **primary:** `cs.RO` · **Submitted:** 2026-02-25
+- **Authors:** Tomoya Kawabe, Rin Takano
+- **abs:** https://arxiv.org/abs/2602.21670 · **PDF:** https://arxiv.org/pdf/2602.21670v2
+
+## 9898. DWA-KD: Dual-Space Weighting and Time-Warped Alignment for Cross-Tokenizer Knowledge Distillation
+
+- **ID:** `2602.21669` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Duc Trung Vu, Pham Khanh Chi, Dat Phi Van, et al.
+- **abs:** https://arxiv.org/abs/2602.21669 · **PDF:** https://arxiv.org/pdf/2602.21669v1
+
+## 9899. Dr. Seg: Revisiting GRPO Training for Visual Large Language Models through Perception-Oriented Design
+
+- **ID:** `2603.00152` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Haoxiang Sun, Tao Wang, Chenwei Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.00152 · **PDF:** https://arxiv.org/pdf/2603.00152v2
+
+## 9900. Prior Knowledge-enhanced Spatio-temporal Epidemic Forecasting
+
+- **ID:** `2602.22270` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Sijie Ruan, Jinyu Li, Jia Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.22270 · **PDF:** https://arxiv.org/pdf/2602.22270v1
+
+## 9901. CQSA: Byzantine-robust Clustered Quantum Secure Aggregation in Federated Learning
+
+- **ID:** `2602.22269` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Arnab Nath, Harsh Kasyap
+- **abs:** https://arxiv.org/abs/2602.22269 · **PDF:** https://arxiv.org/pdf/2602.22269v1
+
+## 9902. Following the Diagnostic Trace: Visual Cognition-guided Cooperative Network for Chest X-Ray Diagnosis
+
+- **ID:** `2602.21657` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Shaoxuan Wu, Jingkun Chen, Chong Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.21657 · **PDF:** https://arxiv.org/pdf/2602.21657v1
+
+## 9903. CCCaption: Dual-Reward Reinforcement Learning for Complete and Correct Image Captioning
+
+- **ID:** `2602.21655` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Zhijiang Tang, Linhua Wang, Jiaxin Qi, et al.
+- **abs:** https://arxiv.org/abs/2602.21655 · **PDF:** https://arxiv.org/pdf/2602.21655v2
+
+## 9904. Sparsity Induction for Accurate Post-Training Pruning of Large Language Models
+
+- **ID:** `2602.21652` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Minhao Jiang, Zhikai Li, Xuewen Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.21652 · **PDF:** https://arxiv.org/pdf/2602.21652v1
+
+## 9905. PPCR-IM: A System for Multi-layer DAG-based Public Policy Consequence Reasoning and Social Indicator Mapping
+
+- **ID:** `2602.21650` · **primary:** `cs.SI` · **Submitted:** 2026-02-25
+- **Authors:** Zichen Song, Weijia Li
+- **abs:** https://arxiv.org/abs/2602.21650 · **PDF:** https://arxiv.org/pdf/2602.21650v1
+
+## 9906. Multimodal Survival Modeling and Fairness-Aware Clinical Machine Learning for 5-Year Breast Cancer Risk Prediction
+
+- **ID:** `2602.21648` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Toktam Khatibi
+- **abs:** https://arxiv.org/abs/2602.21648 · **PDF:** https://arxiv.org/pdf/2602.21648v1
+
+## 9907. Mitigating Structural Noise in Low-Resource S2TT: An Optimized Cascaded Nepali-English Pipeline with Punctuation Restoration
+
+- **ID:** `2602.21647` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Tangsang Chongbang, Pranesh Pyara Shrestha, Amrit Sarki, et al.
+- **abs:** https://arxiv.org/abs/2602.21647 · **PDF:** https://arxiv.org/pdf/2602.21647v2
+
+## 9908. Scalable Multilingual Multimodal Machine Translation with Speech-Text Fusion
+
+- **ID:** `2602.21646` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Yexing Du, Youcheng Pan, Zekun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.21646 · **PDF:** https://arxiv.org/pdf/2602.21646v2
+
+## 9909. AutoQRA: Joint Optimization of Mixed-Precision Quantization and Low-rank Adapters for Efficient LLM Fine-Tuning
+
+- **ID:** `2602.22268` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Changhai Zhou, Shiyang Zhang, Yuhua Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.22268 · **PDF:** https://arxiv.org/pdf/2602.22268v1
+
+## 9910. Data-Driven Supervision of a Thermal-Hydraulic Process Towards a Physics-Based Digital Twin
+
+- **ID:** `2602.22267` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Osimone Imhogiemhe, Yoann Jus, Hubert Lejeune, et al.
+- **abs:** https://arxiv.org/abs/2602.22267 · **PDF:** https://arxiv.org/pdf/2602.22267v1
+
+## 9911. Multi-dimensional Assessment and Explainable Feedback for Counselor Responses to Client Resistance in Text-based Counseling with LLMs
+
+- **ID:** `2602.21638` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Anqi Li, Ruihan Wang, Zhaoming Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.21638 · **PDF:** https://arxiv.org/pdf/2602.21638v1
+
+## 9912. AgentLTV: An Agent-Based Unified Search-and-Evolution Framework for Automated Lifetime Value Prediction
+
+- **ID:** `2602.21634` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Chaowei Wu, Huazhu Chen, Congde Yuan, et al.
+- **abs:** https://arxiv.org/abs/2602.21634 · **PDF:** https://arxiv.org/pdf/2602.21634v1
+
+## 9913. Self-Correcting VLA: Online Action Refinement via Sparse World Imagination
+
+- **ID:** `2602.21633` · **primary:** `cs.RO` · **Submitted:** 2026-02-25
+- **Authors:** Chenyv Liu, Wentao Tan, Lei Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.21633 · **PDF:** https://arxiv.org/pdf/2602.21633v1
+
+## 9914. RuCL: Stratified Rubric-Based Curriculum Learning for Multimodal Large Language Model Reasoning
+
+- **ID:** `2602.21628` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Yukun Chen, Jiaming Li, Longze Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.21628 · **PDF:** https://arxiv.org/pdf/2602.21628v2
+
+## 9915. Enhancing Legal LLMs through Metadata-Enriched RAG Pipelines and Direct Preference Optimization
+
+- **ID:** `2603.19251` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Suyash Maniyar, Deepali Singh, Rohith Reddy
+- **abs:** https://arxiv.org/abs/2603.19251 · **PDF:** https://arxiv.org/pdf/2603.19251v1
+
+## 9916. Your Code Agent Can Grow Alongside You with Structured Memory
+
+- **ID:** `2603.13258` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Yi-Xuan Deng, Xiaoqin Liu, Yi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.13258 · **PDF:** https://arxiv.org/pdf/2603.13258v1
+
+## 9917. Revisiting the Bertrand Paradox via Equilibrium Analysis of No-regret Learners
+
+- **ID:** `2602.21620` · **primary:** `cs.GT` · **Submitted:** 2026-02-25
+- **Authors:** Arnab Maiti, Junyan Liu, Kevin Jamieson, et al.
+- **abs:** https://arxiv.org/abs/2602.21620 · **PDF:** https://arxiv.org/pdf/2602.21620v1
+
+## 9918. WaveSSM: Multiscale State-Space Models for Non-stationary Signal Attention
+
+- **ID:** `2602.22266` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Ruben Solozabal, Velibor Bojkovic, Hilal Alquabeh, et al.
+- **abs:** https://arxiv.org/abs/2602.22266 · **PDF:** https://arxiv.org/pdf/2602.22266v1
+
+## 9919. When More Is Less: A Systematic Analysis of Spatial and Commonsense Information for Visual Spatial Reasoning
+
+- **ID:** `2602.21619` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Muku Akasaka, Soyeon Caren Han
+- **abs:** https://arxiv.org/abs/2602.21619 · **PDF:** https://arxiv.org/pdf/2602.21619v1
+
+## 9920. Can Structural Cues Save LLMs? Evaluating Language Models in Massive Document Streams
+
+- **ID:** `2603.19250` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Yukyung Lee, Yebin Lim, Woojun Jung, et al.
+- **abs:** https://arxiv.org/abs/2603.19250 · **PDF:** https://arxiv.org/pdf/2603.19250v1
+
+## 9921. Virtual Biopsy for Intracranial Tumors Diagnosis on MRI
+
+- **ID:** `2602.21613` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Xinzhe Luo, Shuai Shao, Yan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.21613 · **PDF:** https://arxiv.org/pdf/2602.21613v1
+
+## 9922. Structurally Aligned Subtask-Level Memory for Software Engineering Agents
+
+- **ID:** `2602.21611` · **primary:** `cs.SE` · **Submitted:** 2026-02-25
+- **Authors:** Kangning Shen, Jingyuan Zhang, Chenxi Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.21611 · **PDF:** https://arxiv.org/pdf/2602.21611v1
+
+## 9923. MixSarc: A Bangla-English Code-Mixed Corpus for Implicit Meaning Identification
+
+- **ID:** `2602.21608` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Kazi Samin Yasar Alam, Md Tanbir Chowdhury, Tamim Ahmed, et al.
+- **abs:** https://arxiv.org/abs/2602.21608 · **PDF:** https://arxiv.org/pdf/2602.21608v1
+
+## 9924. Entropy-Controlled Flow Matching
+
+- **ID:** `2602.22265` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Chika Maduabuchi
+- **abs:** https://arxiv.org/abs/2602.22265 · **PDF:** https://arxiv.org/pdf/2602.22265v1
+
+## 9925. Deep Clustering based Boundary-Decoder Net for Inter and Intra Layer Stress Prediction of Heterogeneous Integrated IC Chip
+
+- **ID:** `2602.21601` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Kart Leong Lim, Ji Lin
+- **abs:** https://arxiv.org/abs/2602.21601 · **PDF:** https://arxiv.org/pdf/2602.21601v1
+
+## 9926. Retrieval Challenges in Low-Resource Public Service Information: A Case Study on Food Pantry Access
+
+- **ID:** `2602.21598` · **primary:** `cs.IR` · **Submitted:** 2026-02-25
+- **Authors:** Touseef Hasan, Laila Cure, Souvika Sarkar
+- **abs:** https://arxiv.org/abs/2602.21598 · **PDF:** https://arxiv.org/pdf/2602.21598v1
+
+## 9927. NGDB-Zoo: Towards Efficient and Scalable Neural Graph Databases Training
+
+- **ID:** `2602.21597` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Zhongwei Xie, Jiaxin Bai, Shujie Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.21597 · **PDF:** https://arxiv.org/pdf/2602.21597v1
+
+## 9928. Breaking Semantic-Aware Watermarks via LLM-Guided Coherence-Preserving Semantic Injection
+
+- **ID:** `2602.21593` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Zheng Gao, Xiaoyu Li, Zhicheng Bao, et al.
+- **abs:** https://arxiv.org/abs/2602.21593 · **PDF:** https://arxiv.org/pdf/2602.21593v1
+
+## 9929. HEARTS: Benchmarking LLM Reasoning on Health Time Series
+
+- **ID:** `2603.06638` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Sirui Li, Shuhan Xiao, Mihir Joshi, et al.
+- **abs:** https://arxiv.org/abs/2603.06638 · **PDF:** https://arxiv.org/pdf/2603.06638v2
+
+## 9930. ABM-UDE: Developing Surrogates for Epidemic Agent-Based Models via Scientific Machine Learning
+
+- **ID:** `2602.21588` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Sharv Murgai, Utkarsh Utkarsh, Kyle C. Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.21588 · **PDF:** https://arxiv.org/pdf/2602.21588v1
+
+## 9931. Duel-Evolve: Reward-Free Test-Time Scaling via LLM Self-Preferences
+
+- **ID:** `2602.21585` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Sweta Karlekar, Carolina Zheng, Magnus Saebo, et al.
+- **abs:** https://arxiv.org/abs/2602.21585 · **PDF:** https://arxiv.org/pdf/2602.21585v1
+
+## 9932. Exploring Human-Machine Coexistence in Symmetrical Reality
+
+- **ID:** `2602.21584` · **primary:** `cs.HC` · **Submitted:** 2026-02-25
+- **Authors:** Zhenliang Zhang
+- **abs:** https://arxiv.org/abs/2602.21584 · **PDF:** https://arxiv.org/pdf/2602.21584v1
+
+## 9933. Spelling Correction in Healthcare Query-Answer Systems: Methods, Retrieval Impact, and Empirical Evaluation
+
+- **ID:** `2603.19249` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Saurabh K Singh
+- **abs:** https://arxiv.org/abs/2603.19249 · **PDF:** https://arxiv.org/pdf/2603.19249v1
+
+## 9934. Goodness-of-Fit Tests for Latent Class Models with Ordinal Categorical Data
+
+- **ID:** `2602.21572` · **primary:** `stat.ML` · **Submitted:** 2026-02-25
+- **Authors:** Huan Qing
+- **abs:** https://arxiv.org/abs/2602.21572 · **PDF:** https://arxiv.org/pdf/2602.21572v1
+
+## 9935. DuCCAE: A Hybrid Engine for Immersive Conversation via Collaboration, Augmentation, and Evolution
+
+- **ID:** `2603.19248` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Xin Shen, Zhishu Jiang, Jiaye Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.19248 · **PDF:** https://arxiv.org/pdf/2603.19248v1
+
+## 9936. How many asymmetric communities are there in multi-layer directed networks?
+
+- **ID:** `2602.21569` · **primary:** `math.ST` · **Submitted:** 2026-02-25
+- **Authors:** Huan Qing
+- **abs:** https://arxiv.org/abs/2602.21569 · **PDF:** https://arxiv.org/pdf/2602.21569v1
+
+## 9937. Training-free Composition of Pre-trained GFlowNets for Multi-Objective Generation
+
+- **ID:** `2602.21565` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Seokwon Yoon, Youngbin Choi, Seunghyuk Cho, et al.
+- **abs:** https://arxiv.org/abs/2602.21565 · **PDF:** https://arxiv.org/pdf/2602.21565v1
+
+## 9938. Power and Limitations of Aggregation in Compound AI Systems
+
+- **ID:** `2602.21556` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Nivasini Ananthakrishnan, Meena Jagadeesan
+- **abs:** https://arxiv.org/abs/2602.21556 · **PDF:** https://arxiv.org/pdf/2602.21556v1
+
+## 9939. Revisiting RAG Retrievers: An Information Theoretic Benchmark
+
+- **ID:** `2602.21553` · **primary:** `cs.IR` · **Submitted:** 2026-02-25
+- **Authors:** Wenqing Zheng, Dmitri Kalaev, Noah Fatsi, et al.
+- **abs:** https://arxiv.org/abs/2602.21553 · **PDF:** https://arxiv.org/pdf/2602.21553v1
+
+## 9940. CryoNet.Refine: A One-step Diffusion Model for Rapid Refinement of Structural Models with Cryo-EM Density Map Restraints
+
+- **ID:** `2602.22263` · **primary:** `q-bio.BM` · **Submitted:** 2026-02-25
+- **Authors:** Fuyao Huang, Xiaozhu Yu, Kui Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.22263 · **PDF:** https://arxiv.org/pdf/2602.22263v2
+
+## 9941. From Basis to Basis: Gaussian Particle Representation for Interpretable PDE Operators
+
+- **ID:** `2602.21551` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Zhihao Li, Yu Feng, Zhilu Lai, et al.
+- **abs:** https://arxiv.org/abs/2602.21551 · **PDF:** https://arxiv.org/pdf/2602.21551v1
+
+## 9942. Extending Sequence Length is Not All You Need: Effective Integration of Multimodal Signals for Gene Expression Prediction
+
+- **ID:** `2602.21550` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Zhao Yang, Yi Duan, Jiwei Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.21550 · **PDF:** https://arxiv.org/pdf/2602.21550v2
+
+## 9943. Mamba Meets Scheduling: Learning to Solve Flexible Job Shop Scheduling with Efficient Sequence Modeling
+
+- **ID:** `2602.21546` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Zhi Cao, Cong Zhang, Yaoxin Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.21546 · **PDF:** https://arxiv.org/pdf/2602.21546v1
+
+## 9944. Muon+: Towards Better Muon via One Additional Normalization Step
+
+- **ID:** `2602.21545` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Ruijie Zhang, Yequan Zhao, Ziyue Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.21545 · **PDF:** https://arxiv.org/pdf/2602.21545v2
+
+## 9945. BenchBrowser -- Collecting Evidence for Evaluating Benchmark Validity
+
+- **ID:** `2603.18019` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Harshita Diddee, Gregory Yauney, Swabha Swayamdipta, et al.
+- **abs:** https://arxiv.org/abs/2603.18019 · **PDF:** https://arxiv.org/pdf/2603.18019v1
+
+## 9946. Enhancing Multilingual Embeddings via Multi-Way Parallel Text Alignment
+
+- **ID:** `2602.21543` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Barah Fazili, Koustava Goswami
+- **abs:** https://arxiv.org/abs/2602.21543 · **PDF:** https://arxiv.org/pdf/2602.21543v1
+
+## 9947. Physics-Consistent Diffusion for Efficient Fluid Super-Resolution via Multiscale Residual Correction
+
+- **ID:** `2603.00149` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Zhihao Li, Shengwei Dong, Chuang Yi, et al.
+- **abs:** https://arxiv.org/abs/2603.00149 · **PDF:** https://arxiv.org/pdf/2603.00149v1
+
+## 9948. ARLArena: A Unified Framework for Stable Agentic Reinforcement Learning
+
+- **ID:** `2602.21534` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Xiaoxuan Wang, Han Zhang, Haixin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.21534 · **PDF:** https://arxiv.org/pdf/2602.21534v2
+
+## 9949. Reasoning-Driven Design of Single Atom Catalysts via a Multi-Agent Large Language Model Framework
+
+- **ID:** `2602.21533` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-25
+- **Authors:** Dong Hyeon Mok, Seoin Back, Victor Fung, et al.
+- **abs:** https://arxiv.org/abs/2602.21533 · **PDF:** https://arxiv.org/pdf/2602.21533v1
+
+## 9950. Sustainable LLM Inference using Context-Aware Model Switching
+
+- **ID:** `2602.22261` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Yuvarani, Akashdeep Singh, Zahra Fathanah, et al.
+- **abs:** https://arxiv.org/abs/2602.22261 · **PDF:** https://arxiv.org/pdf/2602.22261v1
+
+## 9951. LiLo-VLA: Compositional Long-Horizon Manipulation via Linked Object-Centric Policies
+
+- **ID:** `2602.21531` · **primary:** `cs.RO` · **Submitted:** 2026-02-25
+- **Authors:** Yue Yang, Shuo Cheng, Yu Fang, et al.
+- **abs:** https://arxiv.org/abs/2602.21531 · **PDF:** https://arxiv.org/pdf/2602.21531v1
+
+## 9952. One Brain, Omni Modalities: Towards Unified Non-Invasive Brain Decoding with Large Language Models
+
+- **ID:** `2602.21522` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-25
+- **Authors:** Changli Tang, Shurui Li, Junliang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.21522 · **PDF:** https://arxiv.org/pdf/2602.21522v1
+
+## 9953. Training Generalizable Collaborative Agents via Strategic Risk Aversion
+
+- **ID:** `2602.21515` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Chengrui Qu, Yizhou Zhang, Nicolas Lanzetti, et al.
+- **abs:** https://arxiv.org/abs/2602.21515 · **PDF:** https://arxiv.org/pdf/2602.21515v2
+
+## 9954. An Agentic System for Schema Aware NL2SQL Generation
+
+- **ID:** `2603.18018` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** David Onyango, Naseef Mansoor
+- **abs:** https://arxiv.org/abs/2603.18018 · **PDF:** https://arxiv.org/pdf/2603.18018v1
+
+## 9955. Fair Model-based Clustering
+
+- **ID:** `2602.21509` · **primary:** `stat.ML` · **Submitted:** 2026-02-25
+- **Authors:** Jinwon Park, Kunwoong Kim, Jihu Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.21509 · **PDF:** https://arxiv.org/pdf/2602.21509v1
+
+## 9956. WaterVIB: Learning Minimal Sufficient Watermark Representations via Variational Information Bottleneck
+
+- **ID:** `2602.21508` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Haoyuan He, Yu Zheng, Jie Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.21508 · **PDF:** https://arxiv.org/pdf/2602.21508v1
+
+## 9957. A Researcher's Guide to Empirical Risk Minimization
+
+- **ID:** `2602.21501` · **primary:** `stat.ML` · **Submitted:** 2026-02-25
+- **Authors:** Lars van der Laan
+- **abs:** https://arxiv.org/abs/2602.21501 · **PDF:** https://arxiv.org/pdf/2602.21501v2
+
+## 9958. Learning Recursive Multi-Scale Representations for Irregular Multivariate Time Series Forecasting
+
+- **ID:** `2602.21498` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Boyuan Li, Zhen Liu, Yicheng Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.21498 · **PDF:** https://arxiv.org/pdf/2602.21498v1
+
+## 9959. Beyond Refusal: Probing the Limits of Agentic Self-Correction for Semantic Sensitive Information
+
+- **ID:** `2602.21496` · **primary:** `cs.AI` · **Submitted:** 2026-02-25
+- **Authors:** Umid Suleymanov, Zaur Rajabov, Emil Mirzazada, et al.
+- **abs:** https://arxiv.org/abs/2602.21496 · **PDF:** https://arxiv.org/pdf/2602.21496v1
+
+## 9960. GradAlign: Gradient-Aligned Data Selection for LLM Reinforcement Learning
+
+- **ID:** `2602.21492` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Ningyuan Yang, Weihua Du, Weiwei Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.21492 · **PDF:** https://arxiv.org/pdf/2602.21492v1
+
+## 9961. Code World Models for Parameter Control in Evolutionary Algorithms
+
+- **ID:** `2602.22260` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Camilo Chacón Sartori, Guillem Rodríguez Corominas
+- **abs:** https://arxiv.org/abs/2602.22260 · **PDF:** https://arxiv.org/pdf/2602.22260v1
+
+## 9962. Orthogonal Weight Modification Enhances Learning Scalability and Convergence Efficiency without Gradient Backpropagation
+
+- **ID:** `2602.22259` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Guoqing Ma, Shan Yu
+- **abs:** https://arxiv.org/abs/2602.22259 · **PDF:** https://arxiv.org/pdf/2602.22259v1
+
+## 9963. Evaluating the Usage of African-American Vernacular English in Large Language Models
+
+- **ID:** `2602.21485` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Deja Dunlap, R. Thomas McCoy
+- **abs:** https://arxiv.org/abs/2602.21485 · **PDF:** https://arxiv.org/pdf/2602.21485v1
+
+## 9964. Both Ends Count! Just How Good are LLM Agents at "Text-to-Big SQL"?
+
+- **ID:** `2602.21480` · **primary:** `cs.DB` · **Submitted:** 2026-02-25
+- **Authors:** Germán T. Eizaguirre, Lars Tissen, Marc Sánchez-Artigas
+- **abs:** https://arxiv.org/abs/2602.21480 · **PDF:** https://arxiv.org/pdf/2602.21480v3
+
+## 9965. Global Sequential Testing for Multi-Stream Auditing
+
+- **ID:** `2602.21479` · **primary:** `stat.ML` · **Submitted:** 2026-02-25
+- **Authors:** Beepul Bharti, Ambar Pal, Jeremias Sulam
+- **abs:** https://arxiv.org/abs/2602.21479 · **PDF:** https://arxiv.org/pdf/2602.21479v1
+
+## 9966. Poisoned Acoustics
+
+- **ID:** `2602.22258` · **primary:** `cs.CR` · **Submitted:** 2026-02-25
+- **Authors:** Harrison Dahme
+- **abs:** https://arxiv.org/abs/2602.22258 · **PDF:** https://arxiv.org/pdf/2602.22258v1
+
+## 9967. Efficient Inference after Directionally Stable Adaptive Experiments
+
+- **ID:** `2602.21478` · **primary:** `stat.ML` · **Submitted:** 2026-02-25
+- **Authors:** Zikai Shen, Houssam Zenati, Nathan Kallus, et al.
+- **abs:** https://arxiv.org/abs/2602.21478 · **PDF:** https://arxiv.org/pdf/2602.21478v1
+
+## 9968. A Knowledge-Driven Approach to Music Segmentation, Music Source Separation and Cinematic Audio Source Separation
+
+- **ID:** `2602.21476` · **primary:** `eess.AS` · **Submitted:** 2026-02-25
+- **Authors:** Chun-wei Ho, Sabato Marco Siniscalchi, Kai Li, et al.
+- **abs:** https://arxiv.org/abs/2602.21476 · **PDF:** https://arxiv.org/pdf/2602.21476v1
+
+## 9969. The Design Space of Tri-Modal Masked Diffusion Models
+
+- **ID:** `2602.21472` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Louis Bethune, Victor Turrisi, Bruno Kacper Mlodozeniec, et al.
+- **abs:** https://arxiv.org/abs/2602.21472 · **PDF:** https://arxiv.org/pdf/2602.21472v1
+
+## 9970. D-Flow SGLD: Source-Space Posterior Sampling for Scientific Inverse Problems with Flow Matching
+
+- **ID:** `2602.21469` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Meet Hemant Parikh, Yaqin Chen, Jian-Xun Wang
+- **abs:** https://arxiv.org/abs/2602.21469 · **PDF:** https://arxiv.org/pdf/2602.21469v1
+
+## 9971. Unsupervised Discovery of Intermediate Phase Order in the Frustrated $J_1$-$J_2$ Heisenberg Model via Prometheus Framework
+
+- **ID:** `2602.21468` · **primary:** `cond-mat.str-el` · **Submitted:** 2026-02-25
+- **Authors:** Brandon Yee, Wilson Collins, Maximilian Rutkowski
+- **abs:** https://arxiv.org/abs/2602.21468 · **PDF:** https://arxiv.org/pdf/2602.21468v3
+
+## 9972. Geometric Priors for Generalizable World Models via Vector Symbolic Architecture
+
+- **ID:** `2602.21467` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** William Youngwoo Chung, Calvin Yeung, Hansen Jin Lillemark, et al.
+- **abs:** https://arxiv.org/abs/2602.21467 · **PDF:** https://arxiv.org/pdf/2602.21467v1
+
+## 9973. Asymptotically Fast Clebsch-Gordan Tensor Products with Vector Spherical Harmonics
+
+- **ID:** `2602.21466` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** YuQing Xie, Ameya Daigavane, Mit Kotak, et al.
+- **abs:** https://arxiv.org/abs/2602.21466 · **PDF:** https://arxiv.org/pdf/2602.21466v1
+
+## 9974. iMiGUE-Speech: A Spontaneous Speech Dataset for Affective Analysis
+
+- **ID:** `2602.21464` · **primary:** `eess.AS` · **Submitted:** 2026-02-25
+- **Authors:** Sofoklis Kakouros, Fang Kang, Haoyu Chen
+- **abs:** https://arxiv.org/abs/2602.21464 · **PDF:** https://arxiv.org/pdf/2602.21464v1
+
+## 9975. Effects of Training Data Quality on Classifier Performance
+
+- **ID:** `2602.21462` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Alan F. Karr, Regina Ruane
+- **abs:** https://arxiv.org/abs/2602.21462 · **PDF:** https://arxiv.org/pdf/2602.21462v1
+
+## 9976. VecGlypher: Unified Vector Glyph Generation with Language Models
+
+- **ID:** `2602.21461` · **primary:** `cs.CL` · **Submitted:** 2026-02-25
+- **Authors:** Xiaoke Huang, Bhavul Gauri, Kam Woh Ng, et al.
+- **abs:** https://arxiv.org/abs/2602.21461 · **PDF:** https://arxiv.org/pdf/2602.21461v1
+
+## 9977. Revisiting Text Ranking in Deep Research
+
+- **ID:** `2602.21456` · **primary:** `cs.IR` · **Submitted:** 2026-02-25
+- **Authors:** Chuan Meng, Litu Ou, Sean MacAvaney, et al.
+- **abs:** https://arxiv.org/abs/2602.21456 · **PDF:** https://arxiv.org/pdf/2602.21456v1
+
+## 9978. When Learning Hurts: Fixed-Pole RNN for Real-Time Online Training
+
+- **ID:** `2602.21454` · **primary:** `cs.LG` · **Submitted:** 2026-02-25
+- **Authors:** Alexander Morgan, Ummay Sumaya Khan, Lingjia Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.21454 · **PDF:** https://arxiv.org/pdf/2602.21454v1
+
+## 9979. Adversarial Robustness of Deep Learning-Based Thyroid Nodule Segmentation in Ultrasound
+
+- **ID:** `2602.21452` · **primary:** `cs.CV` · **Submitted:** 2026-02-25
+- **Authors:** Nicholas Dietrich, David McShannon
+- **abs:** https://arxiv.org/abs/2602.21452 · **PDF:** https://arxiv.org/pdf/2602.21452v1
+
+## 9980. Distilling Deep Reinforcement Learning into Interpretable Fuzzy Rules: An Explainable AI Framework
+
+- **ID:** `2603.13257` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Sanup S. Araballi, Simon Khan, Chilukuri K. Mohan
+- **abs:** https://arxiv.org/abs/2603.13257 · **PDF:** https://arxiv.org/pdf/2603.13257v1
+
+## 9981. Adversarial Intent is a Latent Variable: Stateful Trust Inference for Securing Multimodal Agentic RAG
+
+- **ID:** `2602.21447` · **primary:** `cs.CR` · **Submitted:** 2026-02-24
+- **Authors:** Inderjeet Singh, Vikas Pahuja, Aishvariya Priya Rathina Sabapathy, et al.
+- **abs:** https://arxiv.org/abs/2602.21447 · **PDF:** https://arxiv.org/pdf/2602.21447v1
+
+## 9982. ConformalHDC: Uncertainty-Aware Hyperdimensional Computing with Application to Neural Decoding
+
+- **ID:** `2602.21446` · **primary:** `stat.ML` · **Submitted:** 2026-02-24
+- **Authors:** Ziyi Liang, Hamed Poursiami, Zhishun Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.21446 · **PDF:** https://arxiv.org/pdf/2602.21446v1
+
+## 9983. LMU-Based Sequential Learning and Posterior Ensemble Fusion for Cross-Domain Infant Cry Classification
+
+- **ID:** `2603.02245` · **primary:** `eess.AS` · **Submitted:** 2026-02-24
+- **Authors:** Niloofar Jazaeri, Hilmi R. Dajani, Marco Janeczek, et al.
+- **abs:** https://arxiv.org/abs/2603.02245 · **PDF:** https://arxiv.org/pdf/2603.02245v2
+
+## 9984. Deep Sequence Modeling with Quantum Dynamics: Language as a Wave Function
+
+- **ID:** `2602.22255` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Ahmed Nebli, Hadi Saadatdoorabi, Kevin Yam
+- **abs:** https://arxiv.org/abs/2602.22255 · **PDF:** https://arxiv.org/pdf/2602.22255v1
+
+## 9985. MINAR: Mechanistic Interpretability for Neural Algorithmic Reasoning
+
+- **ID:** `2602.21442` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Jesse He, Helen Jenne, Max Vargas, et al.
+- **abs:** https://arxiv.org/abs/2602.21442 · **PDF:** https://arxiv.org/pdf/2602.21442v1
+
+## 9986. Causal Decoding for Hallucination-Resistant Multimodal Large Language Models
+
+- **ID:** `2602.21441` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Shiwei Tan, Hengyi Wang, Weiyi Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.21441 · **PDF:** https://arxiv.org/pdf/2602.21441v1
+
+## 9987. Efficient Uncoupled Learning Dynamics with $\tilde{O}\!\left(T^{-1/4}\right)$ Last-Iterate Convergence in Bilinear Saddle-Point Problems over Convex Sets under Bandit Feedback
+
+- **ID:** `2602.21436` · **primary:** `stat.ML` · **Submitted:** 2026-02-24
+- **Authors:** Arnab Maiti, Claire Jie Zhang, Kevin Jamieson, et al.
+- **abs:** https://arxiv.org/abs/2602.21436 · **PDF:** https://arxiv.org/pdf/2602.21436v1
+
+## 9988. Provably Safe Generative Sampling with Constricting Barrier Functions
+
+- **ID:** `2602.21429` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Darshan Gadginmath, Ahmed Allibhoy, Fabio Pasqualetti
+- **abs:** https://arxiv.org/abs/2602.21429 · **PDF:** https://arxiv.org/pdf/2602.21429v2
+
+## 9989. How to Count AIs: Individuation and Liability for AI Agents
+
+- **ID:** `2603.10028` · **primary:** `cs.CY` · **Submitted:** 2026-02-24
+- **Authors:** Yonathan Arbel, Peter Salib, Simon Goldstein
+- **abs:** https://arxiv.org/abs/2603.10028 · **PDF:** https://arxiv.org/pdf/2603.10028v1
+
+## 9990. PSF-Med: Measuring and Explaining Paraphrase Sensitivity in Medical Vision Language Models
+
+- **ID:** `2602.21428` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Binesh Sadanandan, Vahid Behzadan
+- **abs:** https://arxiv.org/abs/2602.21428 · **PDF:** https://arxiv.org/pdf/2602.21428v1
+
+## 9991. Frayed RoPE and Long Inputs: A Geometric Perspective
+
+- **ID:** `2603.18017` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Davis Wertheimer, Aozhong Zhang, Derrick Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.18017 · **PDF:** https://arxiv.org/pdf/2603.18017v1
+
+## 9992. Proximal-IMH: Proximal Posterior Proposals for Independent Metropolis-Hastings with Approximate Operators
+
+- **ID:** `2602.21426` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Youguang Chen, George Biros
+- **abs:** https://arxiv.org/abs/2602.21426 · **PDF:** https://arxiv.org/pdf/2602.21426v1
+
+## 9993. On the Structural Non-Preservation of Epistemic Behaviour under Policy Transformation
+
+- **ID:** `2602.21424` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Alexander Galozy
+- **abs:** https://arxiv.org/abs/2602.21424 · **PDF:** https://arxiv.org/pdf/2602.21424v2
+
+## 9994. Detoxifying LLMs via Representation Erasure-Based Preference Optimization
+
+- **ID:** `2602.23391` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Nazanin Mohammadi Sepahvand, Eleni Triantafillou, Hugo Larochelle, et al.
+- **abs:** https://arxiv.org/abs/2602.23391 · **PDF:** https://arxiv.org/pdf/2602.23391v1
+
+## 9995. ECHOSAT: Estimating Canopy Height Over Space And Time
+
+- **ID:** `2602.21421` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Jan Pauls, Karsten Schrödter, Sven Ligensa, et al.
+- **abs:** https://arxiv.org/abs/2602.21421 · **PDF:** https://arxiv.org/pdf/2602.21421v2
+
+## 9996. Overconfident Errors Need Stronger Correction: Asymmetric Confidence Penalties for Reinforcement Learning
+
+- **ID:** `2602.21420` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Yuanda Xu, Hejian Sang, Zhengze Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.21420 · **PDF:** https://arxiv.org/pdf/2602.21420v1
+
+## 9997. Benchmarking State Space Models, Transformers, and Recurrent Networks for US Grid Forecasting
+
+- **ID:** `2602.21415` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Sunki Hong, Jisoo Lee
+- **abs:** https://arxiv.org/abs/2602.21415 · **PDF:** https://arxiv.org/pdf/2602.21415v3
+
+## 9998. Generative Bayesian Computation as a Scalable Alternative to Gaussian Process Surrogates
+
+- **ID:** `2602.21408` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Nick Polson, Vadim Sokolov
+- **abs:** https://arxiv.org/abs/2602.21408 · **PDF:** https://arxiv.org/pdf/2602.21408v1
+
+## 9999. The Headless Firm: How AI Reshapes Enterprise Boundaries
+
+- **ID:** `2602.21401` · **primary:** `cs.GT` · **Submitted:** 2026-02-24
+- **Authors:** Tassilo Klein, Sebastian Wieczorek
+- **abs:** https://arxiv.org/abs/2602.21401 · **PDF:** https://arxiv.org/pdf/2602.21401v1
+
+## 10000. FedVG: Gradient-Guided Aggregation for Enhanced Federated Learning
+
+- **ID:** `2602.21399` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Alina Devkota, Jacob Thrasher, Donald Adjeroh, et al.
+- **abs:** https://arxiv.org/abs/2602.21399 · **PDF:** https://arxiv.org/pdf/2602.21399v2
+
+## 10001. MMLoP: Multi-Modal Low-Rank Prompting for Efficient Vision-Language Adaptation
+
+- **ID:** `2602.21397` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Sajjad Ghiasvand, Haniyeh Ehsani Oskouie, Mahnoosh Alizadeh, et al.
+- **abs:** https://arxiv.org/abs/2602.21397 · **PDF:** https://arxiv.org/pdf/2602.21397v1
+
+## 10002. A Governance and Evaluation Framework for Deterministic, Rule-Based Clinical Decision Support in Empiric Antibiotic Prescribing
+
+- **ID:** `2603.10027` · **primary:** `cs.CY` · **Submitted:** 2026-02-24
+- **Authors:** Francisco José Gárate, Paloma Chausa, Diego Moreno, et al.
+- **abs:** https://arxiv.org/abs/2603.10027 · **PDF:** https://arxiv.org/pdf/2603.10027v1
+
+## 10003. Defensive Generation
+
+- **ID:** `2602.21390` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Gabriele Farina, Juan Carlos Perdomo
+- **abs:** https://arxiv.org/abs/2602.21390 · **PDF:** https://arxiv.org/pdf/2602.21390v1
+
+## 10004. Training-Free Agentic AI: Probabilistic Control and Coordination in Multi-Agent LLM Systems
+
+- **ID:** `2603.13256` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Mohammad Parsa Hosseini, Ankit Shah, Saiyra Qureshi, et al.
+- **abs:** https://arxiv.org/abs/2603.13256 · **PDF:** https://arxiv.org/pdf/2603.13256v1
+
+## 10005. Causal Direction from Convergence Time: Faster Training in the True Causal Direction
+
+- **ID:** `2602.22254` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Abdulrahman Tamim
+- **abs:** https://arxiv.org/abs/2602.22254 · **PDF:** https://arxiv.org/pdf/2602.22254v1
+
+## 10006. MrBERT: Modern Multilingual Encoders via Vocabulary, Domain, and Dimensional Adaptation
+
+- **ID:** `2602.21379` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Daniel Tamayo, Iñaki Lacunza, Paula Rivera-Hidalgo, et al.
+- **abs:** https://arxiv.org/abs/2602.21379 · **PDF:** https://arxiv.org/pdf/2602.21379v2
+
+## 10007. Beyond Subtokens: A Rich Character Embedding for Low-resource and Morphologically Complex Languages
+
+- **ID:** `2602.21377` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Felix Schneider, Maria Gogolev, Sven Sickert, et al.
+- **abs:** https://arxiv.org/abs/2602.21377 · **PDF:** https://arxiv.org/pdf/2602.21377v1
+
+## 10008. Federated Multi Agent Deep Learning and Neural Networks for Advanced Distributed Sensing in Wireless Networks
+
+- **ID:** `2603.16881` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Nadine Muller, Stefano DeRosa, Su Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.16881 · **PDF:** https://arxiv.org/pdf/2603.16881v1
+
+## 10009. NeuroNarrator: A Generalist EEG-to-Text Foundation Model for Clinical Interpretation via Spectro-Spatial Grounding and Temporal State-Space Reasoning
+
+- **ID:** `2603.16880` · **primary:** `eess.SP` · **Submitted:** 2026-02-24
+- **Authors:** Guoan Wang, Shihao Yang, Jun-en Ding, et al.
+- **abs:** https://arxiv.org/abs/2603.16880 · **PDF:** https://arxiv.org/pdf/2603.16880v2
+
+## 10010. Small Language Models for Privacy-Preserving Clinical Information Extraction in Low-Resource Languages
+
+- **ID:** `2602.21374` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Mohammadreza Ghaffarzadeh-Esfahani, Nahid Yousefian, Ebrahim Heidari-Farsani, et al.
+- **abs:** https://arxiv.org/abs/2602.21374 · **PDF:** https://arxiv.org/pdf/2602.21374v1
+
+## 10011. The Mean is the Mirage: Entropy-Adaptive Model Merging under Heterogeneous Domain Shifts in Medical Imaging
+
+- **ID:** `2602.21372` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Sameer Ambekar, Reza Nasirigerdeh, Peter J. Schuffler, et al.
+- **abs:** https://arxiv.org/abs/2602.21372 · **PDF:** https://arxiv.org/pdf/2602.21372v1
+
+## 10012. Interleaved Head Attention
+
+- **ID:** `2602.21371` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Sai Surya Duvvuri, Chanakya Ekbote, Rachit Bansal, et al.
+- **abs:** https://arxiv.org/abs/2602.21371 · **PDF:** https://arxiv.org/pdf/2602.21371v1
+
+## 10013. Black-Box Reliability Certification for AI Agents via Self-Consistency Sampling and Conformal Calibration
+
+- **ID:** `2602.21368` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Charafeddine Mouzouni
+- **abs:** https://arxiv.org/abs/2602.21368 · **PDF:** https://arxiv.org/pdf/2602.21368v1
+
+## 10014. Towards Controllable Video Synthesis of Routine and Rare OR Events
+
+- **ID:** `2602.21365` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Dominik Schneider, Lalithkumar Seenivasan, Sampath Rapuri, et al.
+- **abs:** https://arxiv.org/abs/2602.21365 · **PDF:** https://arxiv.org/pdf/2602.21365v1
+
+## 10015. Zatom-1: A Multimodal Flow Foundation Model for 3D Molecules and Materials
+
+- **ID:** `2602.22251` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Alex Morehead, Miruna Cretu, Antonia Panescu, et al.
+- **abs:** https://arxiv.org/abs/2602.22251 · **PDF:** https://arxiv.org/pdf/2602.22251v2
+
+## 10016. Towards single-shot coherent imaging via overlap-free ptychography
+
+- **ID:** `2602.21361` · **primary:** `physics.optics` · **Submitted:** 2026-02-24
+- **Authors:** Oliver Hoidn, Aashwin Mishra, Steven Henke, et al.
+- **abs:** https://arxiv.org/abs/2602.21361 · **PDF:** https://arxiv.org/pdf/2602.21361v2
+
+## 10017. Representation Theorems for Cumulative Propositional Dependence Logics
+
+- **ID:** `2602.21360` · **primary:** `cs.LO` · **Submitted:** 2026-02-24
+- **Authors:** Juha Kontinen, Arne Meier, Kai Sauerwald
+- **abs:** https://arxiv.org/abs/2602.21360 · **PDF:** https://arxiv.org/pdf/2602.21360v1
+
+## 10018. Conditional neural control variates for variance reduction in Bayesian inverse problems
+
+- **ID:** `2602.21357` · **primary:** `stat.ML` · **Submitted:** 2026-02-24
+- **Authors:** Ali Siahkoohi, Hyunwoo Oh
+- **abs:** https://arxiv.org/abs/2602.21357 · **PDF:** https://arxiv.org/pdf/2602.21357v1
+
+## 10019. A Hierarchical Multi-Agent System for Autonomous Discovery in Geoscientific Data Archives
+
+- **ID:** `2602.21351` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Dmitrii Pantiukhin, Ivan Kuznetsov, Boris Shapkin, et al.
+- **abs:** https://arxiv.org/abs/2602.21351 · **PDF:** https://arxiv.org/pdf/2602.21351v1
+
+## 10020. Alignment-Weighted DPO: A principled reasoning approach to improve safety alignment
+
+- **ID:** `2602.21346` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Mengxuan Hu, Vivek V. Datla, Anoop Kumar, et al.
+- **abs:** https://arxiv.org/abs/2602.21346 · **PDF:** https://arxiv.org/pdf/2602.21346v1
+
+## 10021. Archetypal Graph Generative Models: Explainable and Identifiable Communities via Anchor-Dominant Convex Hulls
+
+- **ID:** `2602.21342` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Nikolaos Nakis, Chrysoula Kosma, Panagiotis Promponas, et al.
+- **abs:** https://arxiv.org/abs/2602.21342 · **PDF:** https://arxiv.org/pdf/2602.21342v1
+
+## 10022. Scaling View Synthesis Transformers
+
+- **ID:** `2602.21341` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Evan Kim, Hyunwoo Ryu, Thomas W. Mitchel, et al.
+- **abs:** https://arxiv.org/abs/2602.21341 · **PDF:** https://arxiv.org/pdf/2602.21341v1
+
+## 10023. HiPPO Zoo: Explicit Memory Mechanisms for Interpretable State Space Models
+
+- **ID:** `2602.21340` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Jack Goffinet, Casey Hanks, David E. Carlson
+- **abs:** https://arxiv.org/abs/2602.21340 · **PDF:** https://arxiv.org/pdf/2602.21340v1
+
+## 10024. Efficient Opportunistic Approachability
+
+- **ID:** `2602.21328` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Teodor Vanislavov Marinov, Mehryar Mohri, Princewill Okoroafor, et al.
+- **abs:** https://arxiv.org/abs/2602.21328 · **PDF:** https://arxiv.org/pdf/2602.21328v1
+
+## 10025. Improving Spatial Allocation for Energy System Coupling with Graph Neural Networks
+
+- **ID:** `2602.22249` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Xuanhao Mu, Jakob Geiges, Nan Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.22249 · **PDF:** https://arxiv.org/pdf/2602.22249v2
+
+## 10026. Equitable Evaluation via Elicitation
+
+- **ID:** `2602.21327` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Elbert Du, Cynthia Dwork, Lunjia Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.21327 · **PDF:** https://arxiv.org/pdf/2602.21327v1
+
+## 10027. Dynamic Symmetric Point Tracking: Tackling Non-ideal Reference in Analog In-memory Training
+
+- **ID:** `2602.21321` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Quan Xiao, Jindan Li, Zhaoxian Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.21321 · **PDF:** https://arxiv.org/pdf/2602.21321v1
+
+## 10028. Tool-R0: Self-Evolving LLM Agents for Tool-Learning from Zero Data
+
+- **ID:** `2602.21320` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Emre Can Acikgoz, Cheng Qian, Jonas Hübotter, et al.
+- **abs:** https://arxiv.org/abs/2602.21320 · **PDF:** https://arxiv.org/pdf/2602.21320v1
+
+## 10029. Uncertainty-Aware Diffusion Model for Multimodal Highway Trajectory Prediction via DDIM Sampling
+
+- **ID:** `2602.21319` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Marion Neumeier, Niklas Roßberg, Michael Botsch, et al.
+- **abs:** https://arxiv.org/abs/2602.21319 · **PDF:** https://arxiv.org/pdf/2602.21319v1
+
+## 10030. Shared Nature, Unique Nurture: PRISM for Pluralistic Reasoning via In-context Structure Modeling
+
+- **ID:** `2602.21317` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Guancheng Tu, Shiyang Zhang, Tianyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.21317 · **PDF:** https://arxiv.org/pdf/2602.21317v1
+
+## 10031. Precedence-Constrained Decision Trees and Coverings
+
+- **ID:** `2602.21312` · **primary:** `cs.DS` · **Submitted:** 2026-02-24
+- **Authors:** Michał Szyfelbein, Dariusz Dereniowski
+- **abs:** https://arxiv.org/abs/2602.21312 · **PDF:** https://arxiv.org/pdf/2602.21312v3
+
+## 10032. SymTorch: A Framework for Symbolic Distillation of Deep Neural Networks
+
+- **ID:** `2602.21307` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Elizabeth S. Z. Tan, Adil Soubki, Miles Cranmer
+- **abs:** https://arxiv.org/abs/2602.21307 · **PDF:** https://arxiv.org/pdf/2602.21307v1
+
+## 10033. Robust AI Evaluation through Maximal Lotteries
+
+- **ID:** `2602.21297` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Hadi Khalaf, Serena L. Wang, Daniel Halpern, et al.
+- **abs:** https://arxiv.org/abs/2602.21297 · **PDF:** https://arxiv.org/pdf/2602.21297v1
+
+## 10034. Test-Time Training with KV Binding Is Secretly Linear Attention
+
+- **ID:** `2602.21204` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Junchen Liu, Sven Elflein, Or Litany, et al.
+- **abs:** https://arxiv.org/abs/2602.21204 · **PDF:** https://arxiv.org/pdf/2602.21204v2
+
+## 10035. Squint: Fast Visual Reinforcement Learning for Sim-to-Real Robotics
+
+- **ID:** `2602.21203` · **primary:** `cs.RO` · **Submitted:** 2026-02-24
+- **Authors:** Abdulaziz Almuzairee, Henrik I. Christensen
+- **abs:** https://arxiv.org/abs/2602.21203 · **PDF:** https://arxiv.org/pdf/2602.21203v1
+
+## 10036. Multi-Vector Index Compression in Any Modality
+
+- **ID:** `2602.21202` · **primary:** `cs.IR` · **Submitted:** 2026-02-24
+- **Authors:** Hanxiang Qin, Alexander Martin, Rohan Jha, et al.
+- **abs:** https://arxiv.org/abs/2602.21202 · **PDF:** https://arxiv.org/pdf/2602.21202v1
+
+## 10037. Aletheia tackles FirstProof autonomously
+
+- **ID:** `2602.21201` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Tony Feng, Junehyuk Jung, Sang-hyun Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.21201 · **PDF:** https://arxiv.org/pdf/2602.21201v3
+
+## 10038. Learning from Trials and Errors: Reflective Test-Time Planning for Embodied LLMs
+
+- **ID:** `2602.21198` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Yining Hong, Huang Huang, Manling Li, et al.
+- **abs:** https://arxiv.org/abs/2602.21198 · **PDF:** https://arxiv.org/pdf/2602.21198v1
+
+## 10039. Untied Ulysses: Memory-Efficient Context Parallelism via Headwise Chunking
+
+- **ID:** `2602.21196` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Ravi Ghadia, Maksim Abraham, Sergei Vorobyov, et al.
+- **abs:** https://arxiv.org/abs/2602.21196 · **PDF:** https://arxiv.org/pdf/2602.21196v1
+
+## 10040. On Data Engineering for Scaling LLM Terminal Capabilities
+
+- **ID:** `2602.21193` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Renjie Pi, Grace Lam, Mohammad Shoeybi, et al.
+- **abs:** https://arxiv.org/abs/2602.21193 · **PDF:** https://arxiv.org/pdf/2602.21193v1
+
+## 10041. Physics-constrained symbolic regression for discovering closed-form equations of multimodal water retention curves from experimental data
+
+- **ID:** `2603.03346` · **primary:** `physics.geo-ph` · **Submitted:** 2026-02-24
+- **Authors:** Yejin Kim, Hyoung Suk Suh
+- **abs:** https://arxiv.org/abs/2603.03346 · **PDF:** https://arxiv.org/pdf/2603.03346v1
+
+## 10042. Statistical Query Lower Bounds for Smoothed Agnostic Learning
+
+- **ID:** `2602.21191` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Ilias Diakonikolas, Daniel M. Kane
+- **abs:** https://arxiv.org/abs/2602.21191 · **PDF:** https://arxiv.org/pdf/2602.21191v1
+
+## 10043. Why Pass@k Optimization Can Degrade Pass@1: Prompt Interference in LLM Post-training
+
+- **ID:** `2602.21189` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Anas Barakat, Souradip Chakraborty, Khushbu Pahwa, et al.
+- **abs:** https://arxiv.org/abs/2602.21189 · **PDF:** https://arxiv.org/pdf/2602.21189v2
+
+## 10044. The Ghost in the Grammar: Methodological Anthropomorphism in AI Safety Evaluations
+
+- **ID:** `2603.13255` · **primary:** `cs.CY` · **Submitted:** 2026-02-24
+- **Authors:** Mariana Lins Costa
+- **abs:** https://arxiv.org/abs/2603.13255 · **PDF:** https://arxiv.org/pdf/2603.13255v1
+
+## 10045. The Diffusion Duality, Chapter II: $Ψ$-Samplers and Efficient Curriculum
+
+- **ID:** `2602.21185` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Justin Deschenaux, Caglar Gulcehre, Subham Sekhar Sahoo
+- **abs:** https://arxiv.org/abs/2602.21185 · **PDF:** https://arxiv.org/pdf/2602.21185v1
+
+## 10046. XMorph: Explainable Brain Tumor Analysis Via LLM-Assisted Hybrid Deep Intelligence
+
+- **ID:** `2602.21178` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Sepehr Salem Ghahfarokhi, M. Moein Esfahani, Raj Sunderraman, et al.
+- **abs:** https://arxiv.org/abs/2602.21178 · **PDF:** https://arxiv.org/pdf/2602.21178v1
+
+## 10047. Efficient Hierarchical Any-Angle Path Planning on Multi-Resolution 3D Grids
+
+- **ID:** `2602.21174` · **primary:** `cs.RO` · **Submitted:** 2026-02-24
+- **Authors:** Victor Reijgwart, Cesar Cadena, Roland Siegwart, et al.
+- **abs:** https://arxiv.org/abs/2602.21174 · **PDF:** https://arxiv.org/pdf/2602.21174v1
+
+## 10048. NoRD: A Data-Efficient Vision-Language-Action Model that Drives without Reasoning
+
+- **ID:** `2602.21172` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Ishaan Rawal, Shubh Gupta, Yihan Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.21172 · **PDF:** https://arxiv.org/pdf/2602.21172v2
+
+## 10049. Sequential Counterfactual Inference for Temporal Clinical Data: Addressing the Time Traveler Dilemma
+
+- **ID:** `2602.21168` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Jingya Cheng, Alaleh Azhir, Jiazi Tian, et al.
+- **abs:** https://arxiv.org/abs/2602.21168 · **PDF:** https://arxiv.org/pdf/2602.21168v1
+
+## 10050. PVminer: A Domain-Specific Tool to Detect the Patient Voice in Patient Generated Data
+
+- **ID:** `2602.21165` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Samah Fodeh, Linhai Ma, Yan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.21165 · **PDF:** https://arxiv.org/pdf/2602.21165v1
+
+## 10051. Not Just How Much, But Where: Decomposing Epistemic Uncertainty into Per-Class Contributions
+
+- **ID:** `2602.21160` · **primary:** `stat.ML` · **Submitted:** 2026-02-24
+- **Authors:** Mame Diarra Toure, David A. Stephens
+- **abs:** https://arxiv.org/abs/2602.21160 · **PDF:** https://arxiv.org/pdf/2602.21160v2
+
+## 10052. SELAUR: Self Evolving LLM Agent via Uncertainty-aware Rewards
+
+- **ID:** `2602.21158` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Dengjia Zhang, Xiaoou Liu, Lu Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.21158 · **PDF:** https://arxiv.org/pdf/2602.21158v2
+
+## 10053. CG-DMER: Hybrid Contrastive-Generative Framework for Disentangled Multimodal ECG Representation Learning
+
+- **ID:** `2602.21154` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Ziwei Niu, Hao Sun, Shujun Bian, et al.
+- **abs:** https://arxiv.org/abs/2602.21154 · **PDF:** https://arxiv.org/pdf/2602.21154v1
+
+## 10054. Multi-Dimensional Spectral Geometry of Biological Knowledge in Single-Cell Transformer Representations
+
+- **ID:** `2602.22247` · **primary:** `q-bio.GN` · **Submitted:** 2026-02-24
+- **Authors:** Ihor Kendiukhov
+- **abs:** https://arxiv.org/abs/2602.22247 · **PDF:** https://arxiv.org/pdf/2602.22247v1
+
+## 10055. Introducing Feature-Based Trajectory Clustering, a clustering algorithm for longitudinal data
+
+- **ID:** `2603.13254` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Marie-Pierre Sylvestre, Laurence Boulanger
+- **abs:** https://arxiv.org/abs/2603.13254 · **PDF:** https://arxiv.org/pdf/2603.13254v1
+
+## 10056. Neural network optimization strategies and the topography of the loss landscape
+
+- **ID:** `2602.21276` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Jianneng Yu, Alexandre V. Morozov
+- **abs:** https://arxiv.org/abs/2602.21276 · **PDF:** https://arxiv.org/pdf/2602.21276v1
+
+## 10057. Scaling State-Space Models on Multiple GPUs with Tensor Parallelism
+
+- **ID:** `2602.21144` · **primary:** `cs.DC` · **Submitted:** 2026-02-24
+- **Authors:** Anurag Dutt, Nimit Shah, Hazem Masarani, et al.
+- **abs:** https://arxiv.org/abs/2602.21144 · **PDF:** https://arxiv.org/pdf/2602.21144v1
+
+## 10058. A Benchmark for Deep Information Synthesis
+
+- **ID:** `2602.21143` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Debjit Paul, Daniel Murphy, Milan Gritta, et al.
+- **abs:** https://arxiv.org/abs/2602.21143 · **PDF:** https://arxiv.org/pdf/2602.21143v1
+
+## 10059. LUMEN: Longitudinal Multi-Modal Radiology Model for Prognosis and Diagnosis
+
+- **ID:** `2602.21142` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Zhifan Jiang, Dong Yang, Vishwesh Nath, et al.
+- **abs:** https://arxiv.org/abs/2602.21142 · **PDF:** https://arxiv.org/pdf/2602.21142v1
+
+## 10060. Complexity of Classical Acceleration for $\ell_1$-Regularized PageRank
+
+- **ID:** `2602.21138` · **primary:** `math.OC` · **Submitted:** 2026-02-24
+- **Authors:** Kimon Fountoulakis, David Martínez-Rubio
+- **abs:** https://arxiv.org/abs/2602.21138 · **PDF:** https://arxiv.org/pdf/2602.21138v1
+
+## 10061. SparkMe: Adaptive Semi-Structured Interviewing for Qualitative Insight Discovery
+
+- **ID:** `2602.21136` · **primary:** `cs.HC` · **Submitted:** 2026-02-24
+- **Authors:** David Anugraha, Vishakh Padmakumar, Diyi Yang
+- **abs:** https://arxiv.org/abs/2602.21136 · **PDF:** https://arxiv.org/pdf/2602.21136v1
+
+## 10062. SOM-VQ: Topology-Aware Tokenization for Interactive Generative Models
+
+- **ID:** `2602.21133` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Alessandro Londei, Denise Lanzieri, Matteo Benati
+- **abs:** https://arxiv.org/abs/2602.21133 · **PDF:** https://arxiv.org/pdf/2602.21133v1
+
+## 10063. An Enhanced Projection Pursuit Tree Classifier with Visual Methods for Assessing Algorithmic Improvements
+
+- **ID:** `2602.21130` · **primary:** `stat.ML` · **Submitted:** 2026-02-24
+- **Authors:** Natalia da Silva, Dianne Cook, Eun-Kyung Lee
+- **abs:** https://arxiv.org/abs/2602.21130 · **PDF:** https://arxiv.org/pdf/2602.21130v2
+
+## 10064. MineDraft: A Framework for Batch Parallel Speculative Decoding
+
+- **ID:** `2603.18016` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Zhenwei Tang, Arun Verma, Zijian Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.18016 · **PDF:** https://arxiv.org/pdf/2603.18016v1
+
+## 10065. "Are You Sure?": An Empirical Study of Human Perception Vulnerability in LLM-Driven Agentic Systems
+
+- **ID:** `2602.21127` · **primary:** `cs.HC` · **Submitted:** 2026-02-24
+- **Authors:** Xinfeng Li, Shenyu Dai, Kelong Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.21127 · **PDF:** https://arxiv.org/pdf/2602.21127v1
+
+## 10066. Cooperative-Competitive Team Play of Real-World Craft Robots
+
+- **ID:** `2602.21119` · **primary:** `cs.RO` · **Submitted:** 2026-02-24
+- **Authors:** Rui Zhao, Xihui Li, Yizheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.21119 · **PDF:** https://arxiv.org/pdf/2602.21119v1
+
+## 10067. Sensitivity-Guided Framework for Pruned and Quantized Reservoir Computing Accelerators
+
+- **ID:** `2603.08737` · **primary:** `cs.AR` · **Submitted:** 2026-02-24
+- **Authors:** Atousa Jafari, Mahdi Taheri, Hassan Ghasemzadeh Mohammadi, et al.
+- **abs:** https://arxiv.org/abs/2603.08737 · **PDF:** https://arxiv.org/pdf/2603.08737v1
+
+## 10068. Attention-Based SINR Estimation in User-Centric Non-Terrestrial Networks
+
+- **ID:** `2602.21116` · **primary:** `eess.SP` · **Submitted:** 2026-02-24
+- **Authors:** Bruno De Filippo, Alessandro Guidotti, Alessandro Vanelli-Coralli
+- **abs:** https://arxiv.org/abs/2602.21116 · **PDF:** https://arxiv.org/pdf/2602.21116v1
+
+## 10069. Ski Rental with Distributional Predictions of Unknown Quality
+
+- **ID:** `2602.21104` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Qiming Cui, Michael Dinitz
+- **abs:** https://arxiv.org/abs/2602.21104 · **PDF:** https://arxiv.org/pdf/2602.21104v1
+
+## 10070. Prompt-Level Distillation: A Non-Parametric Alternative to Model Fine-Tuning for Efficient Reasoning
+
+- **ID:** `2602.21103` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Sanket Badhe, Deep Shah
+- **abs:** https://arxiv.org/abs/2602.21103 · **PDF:** https://arxiv.org/pdf/2602.21103v1
+
+## 10071. Probing Graph Neural Network Activation Patterns Through Graph Topology
+
+- **ID:** `2602.21092` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Floriano Tori, Lorenzo Bini, Marco Sorbi, et al.
+- **abs:** https://arxiv.org/abs/2602.21092 · **PDF:** https://arxiv.org/pdf/2602.21092v1
+
+## 10072. Beyond the Star Rating: A Scalable Framework for Aspect-Based Sentiment Analysis Using LLMs and Text Classification
+
+- **ID:** `2602.21082` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Vishal Patil, Shree Vaishnavi Bacha, Revanth Yamani, et al.
+- **abs:** https://arxiv.org/abs/2602.21082 · **PDF:** https://arxiv.org/pdf/2602.21082v1
+
+## 10073. Scaling Vision Transformers: Evaluating DeepSpeed for Image-Centric Workloads
+
+- **ID:** `2602.21081` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Huy Trinh, Rebecca Ma, Zeqi Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.21081 · **PDF:** https://arxiv.org/pdf/2602.21081v1
+
+## 10074. ProxyFL: A Proxy-Guided Framework for Federated Semi-Supervised Learning
+
+- **ID:** `2602.21078` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Duowen Chen, Yan Wang
+- **abs:** https://arxiv.org/abs/2602.21078 · **PDF:** https://arxiv.org/pdf/2602.21078v1
+
+## 10075. Localized Dynamics-Aware Domain Adaption for Off-Dynamics Offline Reinforcement Learning
+
+- **ID:** `2602.21072` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Zhangjie Xia, Yu Yang, Pan Xu
+- **abs:** https://arxiv.org/abs/2602.21072 · **PDF:** https://arxiv.org/pdf/2602.21072v1
+
+## 10076. The Initial Exploration Problem in Knowledge Graph Exploration
+
+- **ID:** `2602.21066` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Claire McNamara, Lucy Hederman, Declan O'Sullivan
+- **abs:** https://arxiv.org/abs/2602.21066 · **PDF:** https://arxiv.org/pdf/2602.21066v1
+
+## 10077. Beyond Accuracy: An Explainability-Driven Analysis of Harmful Content Detection
+
+- **ID:** `2603.18015` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Trishita Dhara, Siddhesh Sheth
+- **abs:** https://arxiv.org/abs/2603.18015 · **PDF:** https://arxiv.org/pdf/2603.18015v1
+
+## 10078. Motivation is Something You Need
+
+- **ID:** `2602.21064` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Mehdi Acheli, Walid Gaaloul
+- **abs:** https://arxiv.org/abs/2602.21064 · **PDF:** https://arxiv.org/pdf/2602.21064v1
+
+## 10079. Autonomous Edge-Deployed AI Agents for Electric Vehicle Charging Infrastructure Management
+
+- **ID:** `2603.08736` · **primary:** `cs.DC` · **Submitted:** 2026-02-24
+- **Authors:** Mohammed Cherifi
+- **abs:** https://arxiv.org/abs/2603.08736 · **PDF:** https://arxiv.org/pdf/2603.08736v1
+
+## 10080. Tool Building as a Path to "Superintelligence"
+
+- **ID:** `2602.21061` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** David Koplow, Tomer Galanti, Tomaso Poggio
+- **abs:** https://arxiv.org/abs/2602.21061 · **PDF:** https://arxiv.org/pdf/2602.21061v1
+
+## 10081. An Expert Schema for Evaluating Large Language Model Errors in Scholarly Question-Answering Systems
+
+- **ID:** `2602.21059` · **primary:** `cs.HC` · **Submitted:** 2026-02-24
+- **Authors:** Anna Martin-Boyle, William Humphreys, Martha Brown, et al.
+- **abs:** https://arxiv.org/abs/2602.21059 · **PDF:** https://arxiv.org/pdf/2602.21059v1
+
+## 10082. Adjacency Spectral Embeddings of Correlation Networks
+
+- **ID:** `2602.21055` · **primary:** `math.ST` · **Submitted:** 2026-02-24
+- **Authors:** Keith Levin
+- **abs:** https://arxiv.org/abs/2602.21055 · **PDF:** https://arxiv.org/pdf/2602.21055v1
+
+## 10083. VAUQ: Vision-Aware Uncertainty Quantification for LVLM Self-Evaluation
+
+- **ID:** `2602.21054` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Seongheon Park, Changdae Oh, Hyeong Kyu Choi, et al.
+- **abs:** https://arxiv.org/abs/2602.21054 · **PDF:** https://arxiv.org/pdf/2602.21054v1
+
+## 10084. Position-Aware Sequential Attention for Accurate Next Item Recommendations
+
+- **ID:** `2602.21052` · **primary:** `cs.IR` · **Submitted:** 2026-02-24
+- **Authors:** Timur Nabiev, Evgeny Frolov
+- **abs:** https://arxiv.org/abs/2602.21052 · **PDF:** https://arxiv.org/pdf/2602.21052v1
+
+## 10085. SmartBench: Evaluating LLMs in Smart Homes with Anomalous Device States and Behavioral Contexts
+
+- **ID:** `2603.06636` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Qingsong Zou, Zhi Yan, Zhiyao Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.06636 · **PDF:** https://arxiv.org/pdf/2603.06636v1
+
+## 10086. Benchmarking Federated Learning in Edge Computing Environments: A Systematic Review and Performance Evaluation
+
+- **ID:** `2603.08735` · **primary:** `cs.DC` · **Submitted:** 2026-02-24
+- **Authors:** Sales Aribe, Gil Nicholas Cagande
+- **abs:** https://arxiv.org/abs/2603.08735 · **PDF:** https://arxiv.org/pdf/2603.08735v1
+
+## 10087. PIME: Prototype-based Interpretable MCTS-Enhanced Brain Network Analysis for Disorder Diagnosis
+
+- **ID:** `2602.21046` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Kunyu Zhang, Yanwu Yang, Jing Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.21046 · **PDF:** https://arxiv.org/pdf/2602.21046v1
+
+## 10088. PaperTrail: A Claim-Evidence Interface for Grounding Provenance in LLM-based Scholarly Q&A
+
+- **ID:** `2602.21045` · **primary:** `cs.HC` · **Submitted:** 2026-02-24
+- **Authors:** Anna Martin-Boyle, Cara A. C. Leckey, Martha C. Brown, et al.
+- **abs:** https://arxiv.org/abs/2602.21045 · **PDF:** https://arxiv.org/pdf/2602.21045v1
+
+## 10089. LogicGraph : Benchmarking Multi-Path Logical Reasoning via Neuro-Symbolic Generation and Verification
+
+- **ID:** `2602.21044` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Yanrui Wu, Lingling Zhang, Xinyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.21044 · **PDF:** https://arxiv.org/pdf/2602.21044v1
+
+## 10090. T1: One-to-One Channel-Head Binding for Multivariate Time-Series Imputation
+
+- **ID:** `2602.21043` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Dongik Park, Hyunwoo Ryu, Suahn Bae, et al.
+- **abs:** https://arxiv.org/abs/2602.21043 · **PDF:** https://arxiv.org/pdf/2602.21043v3
+
+## 10091. Is Multi-Distribution Learning as Easy as PAC Learning: Sharp Rates with Bounded Label Noise
+
+- **ID:** `2602.21039` · **primary:** `stat.ML` · **Submitted:** 2026-02-24
+- **Authors:** Rafael Hanashiro, Abhishek Shetty, Patrick Jaillet
+- **abs:** https://arxiv.org/abs/2602.21039 · **PDF:** https://arxiv.org/pdf/2602.21039v1
+
+## 10092. Counterdiabatic Hamiltonian Monte Carlo
+
+- **ID:** `2602.21272` · **primary:** `stat.ML` · **Submitted:** 2026-02-24
+- **Authors:** Reuben Cohn-Gordon, Uroš Seljak, Dries Sels
+- **abs:** https://arxiv.org/abs/2602.21272 · **PDF:** https://arxiv.org/pdf/2602.21272v1
+
+## 10093. Empirically Calibrated Conditional Independence Tests
+
+- **ID:** `2602.21036` · **primary:** `stat.ME` · **Submitted:** 2026-02-24
+- **Authors:** Milleno Pan, Antoine de Mathelin, Wesley Tansey
+- **abs:** https://arxiv.org/abs/2602.21036 · **PDF:** https://arxiv.org/pdf/2602.21036v1
+
+## 10094. MIP Candy: A Modular PyTorch Framework for Medical Image Processing
+
+- **ID:** `2602.21033` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Tianhao Fu, Yucheng Chen
+- **abs:** https://arxiv.org/abs/2602.21033 · **PDF:** https://arxiv.org/pdf/2602.21033v1
+
+## 10095. Exchangeable Gaussian Processes for Staggered-Adoption Policy Evaluation
+
+- **ID:** `2602.21031` · **primary:** `stat.ME` · **Submitted:** 2026-02-24
+- **Authors:** Hayk Gevorgyan, Konstantinos Kalogeropoulos, Angelos Alexopoulos
+- **abs:** https://arxiv.org/abs/2602.21031 · **PDF:** https://arxiv.org/pdf/2602.21031v1
+
+## 10096. Self-Purification Mitigates Backdoors in Multimodal Diffusion Language Models
+
+- **ID:** `2602.22246` · **primary:** `cs.CR` · **Submitted:** 2026-02-24
+- **Authors:** Guangnian Wan, Qi Li, Gongfan Fang, et al.
+- **abs:** https://arxiv.org/abs/2602.22246 · **PDF:** https://arxiv.org/pdf/2602.22246v1
+
+## 10097. Matching Multiple Experts: On the Exploitability of Multi-Agent Imitation Learning
+
+- **ID:** `2602.21020` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Antoine Bergerault, Volkan Cevher, Negar Mehr
+- **abs:** https://arxiv.org/abs/2602.21020 · **PDF:** https://arxiv.org/pdf/2602.21020v1
+
+## 10098. PowerModelsGAT-AI: Physics-Informed Graph Attention for Multi-System Power Flow with Continual Learning
+
+- **ID:** `2603.16879` · **primary:** `eess.SY` · **Submitted:** 2026-02-24
+- **Authors:** Chidozie Ezeakunne, Jose E. Tabarez, Reeju Pokharel, et al.
+- **abs:** https://arxiv.org/abs/2603.16879 · **PDF:** https://arxiv.org/pdf/2603.16879v1
+
+## 10099. HiSAC: Hierarchical Sparse Activation Compression for Ultra-long Sequence Modeling in Recommenders
+
+- **ID:** `2602.21009` · **primary:** `cs.IR` · **Submitted:** 2026-02-24
+- **Authors:** Kun Yuan, Junyu Bi, Daixuan Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.21009 · **PDF:** https://arxiv.org/pdf/2602.21009v1
+
+## 10100. Generative Pseudo-Labeling for Pre-Ranking with LLMs
+
+- **ID:** `2602.20995` · **primary:** `cs.IR` · **Submitted:** 2026-02-24
+- **Authors:** Junyu Bi, Xinting Niu, Daixuan Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.20995 · **PDF:** https://arxiv.org/pdf/2602.20995v1
+
+## 10101. Multimodal MRI Report Findings Supervised Brain Lesion Segmentation with Substructures
+
+- **ID:** `2602.20994` · **primary:** `eess.IV` · **Submitted:** 2026-02-24
+- **Authors:** Yubin Ge, Yongsong Huang, Xiaofeng Liu
+- **abs:** https://arxiv.org/abs/2602.20994 · **PDF:** https://arxiv.org/pdf/2602.20994v1
+
+## 10102. A Counterfactual Approach for Addressing Individual User Unfairness in Collaborative Recommender System
+
+- **ID:** `2603.13253` · **primary:** `cs.IR` · **Submitted:** 2026-02-24
+- **Authors:** Nikita Baidya, Bidyut Kr. Patra, Ratnakar Dash
+- **abs:** https://arxiv.org/abs/2603.13253 · **PDF:** https://arxiv.org/pdf/2603.13253v1
+
+## 10103. Echoes Over Time: Unlocking Length Generalization in Video-to-Audio Generation Models
+
+- **ID:** `2602.20981` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Christian Simon, Masato Ishii, Wei-Yao Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.20981 · **PDF:** https://arxiv.org/pdf/2602.20981v2
+
+## 10104. CrystaL: Spontaneous Emergence of Visual Latents in MLLMs
+
+- **ID:** `2602.20980` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Yang Zhang, Danyang Li, Yuxuan Li, et al.
+- **abs:** https://arxiv.org/abs/2602.20980 · **PDF:** https://arxiv.org/pdf/2602.20980v2
+
+## 10105. Toward an Agentic Infused Software Ecosystem
+
+- **ID:** `2602.20979` · **primary:** `cs.SE` · **Submitted:** 2026-02-24
+- **Authors:** Mark Marron
+- **abs:** https://arxiv.org/abs/2602.20979 · **PDF:** https://arxiv.org/pdf/2602.20979v1
+
+## 10106. Evaluating Proactive Risk Awareness of Large Language Models
+
+- **ID:** `2602.20976` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Xuan Luo, Yubin Chen, Zhiyu Hou, et al.
+- **abs:** https://arxiv.org/abs/2602.20976 · **PDF:** https://arxiv.org/pdf/2602.20976v1
+
+## 10107. MAST: A Multi-fidelity Augmented Surrogate model via Spatial Trust-weighting
+
+- **ID:** `2602.20974` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Ahmed Mohamed Eisa Nasr, Haris Moazam Sheikh
+- **abs:** https://arxiv.org/abs/2602.20974 · **PDF:** https://arxiv.org/pdf/2602.20974v1
+
+## 10108. Linear Reasoning vs. Proof by Cases: Obstacles for Large Language Models in FOL Problem Solving
+
+- **ID:** `2602.20973` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Yuliang Ji, Fuchen Shen, Jian Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.20973 · **PDF:** https://arxiv.org/pdf/2602.20973v1
+
+## 10109. Does Order Matter : Connecting The Law of Robustness to Robust Generalization
+
+- **ID:** `2602.20971` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Himadri Mandal, Vishnu Varadarajan, Jaee Ponde, et al.
+- **abs:** https://arxiv.org/abs/2602.20971 · **PDF:** https://arxiv.org/pdf/2602.20971v2
+
+## 10110. Training-Free Intelligibility-Guided Observation Addition for Noisy ASR
+
+- **ID:** `2602.20967` · **primary:** `eess.AS` · **Submitted:** 2026-02-24
+- **Authors:** Haoyang Li, Changsong Liu, Wei Rao, et al.
+- **abs:** https://arxiv.org/abs/2602.20967 · **PDF:** https://arxiv.org/pdf/2602.20967v1
+
+## 10111. Blackbird Language Matrices: A Framework to Investigate the Linguistic Competence of Language Models
+
+- **ID:** `2602.20966` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Paola Merlo, Chunyang Jiang, Giuseppe Samo, et al.
+- **abs:** https://arxiv.org/abs/2602.20966 · **PDF:** https://arxiv.org/pdf/2602.20966v1
+
+## 10112. EKF-Based Depth Camera and Deep Learning Fusion for UAV-Person Distance Estimation and Following in SAR Operations
+
+- **ID:** `2602.20958` · **primary:** `cs.RO` · **Submitted:** 2026-02-24
+- **Authors:** Luka Šiktar, Branimir Ćaran, Bojan Šekoranja, et al.
+- **abs:** https://arxiv.org/abs/2602.20958 · **PDF:** https://arxiv.org/pdf/2602.20958v1
+
+## 10113. See and Fix the Flaws: Enabling VLMs and Diffusion Models to Comprehend Visual Artifacts via Agentic Data Synthesis
+
+- **ID:** `2602.20951` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Jaehyun Park, Minyoung Ahn, Minkyu Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.20951 · **PDF:** https://arxiv.org/pdf/2602.20951v2
+
+## 10114. Estimation of Confidence Bounds in Binary Classification using Wilson Score Kernel Density Estimation
+
+- **ID:** `2602.20947` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Thorbjørn Mosekjær Iversen, Zebin Duan, Frederik Hagelskjær
+- **abs:** https://arxiv.org/abs/2602.20947 · **PDF:** https://arxiv.org/pdf/2602.20947v1
+
+## 10115. Some Simple Economics of AGI
+
+- **ID:** `2602.20946` · **primary:** `econ.GN` · **Submitted:** 2026-02-24
+- **Authors:** Christian Catalini, Xiang Hui, Jane Wu
+- **abs:** https://arxiv.org/abs/2602.20946 · **PDF:** https://arxiv.org/pdf/2602.20946v2
+
+## 10116. The Art of Efficient Reasoning: Data, Reward, and Optimization
+
+- **ID:** `2602.20945` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Taiqiang Wu, Zenan Xu, Bo Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.20945 · **PDF:** https://arxiv.org/pdf/2602.20945v3
+
+## 10117. Extending $μ$P: Spectral Conditions for Feature Learning Across Optimizers
+
+- **ID:** `2602.20937` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Akshita Gupta, Marieme Ngom, Sam Foreman, et al.
+- **abs:** https://arxiv.org/abs/2602.20937 · **PDF:** https://arxiv.org/pdf/2602.20937v1
+
+## 10118. Architecting AgentOS: From Token-Level Context to Emergent System-Level Intelligence
+
+- **ID:** `2602.20934` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** ChengYou Li, XiaoDong Liu, XiangBao Meng, et al.
+- **abs:** https://arxiv.org/abs/2602.20934 · **PDF:** https://arxiv.org/pdf/2602.20934v1
+
+## 10119. Hierarchic-EEG2Text: Assessing EEG-To-Text Decoding across Hierarchical Abstraction Levels
+
+- **ID:** `2602.20932` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Anupam Sharma, Harish Katti, Prajwal Singh, et al.
+- **abs:** https://arxiv.org/abs/2602.20932 · **PDF:** https://arxiv.org/pdf/2602.20932v1
+
+## 10120. HELP: HyperNode Expansion and Logical Path-Guided Evidence Localization for Accurate and Efficient GraphRAG
+
+- **ID:** `2602.20926` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Yuqi Huang, Ning Liao, Kai Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.20926 · **PDF:** https://arxiv.org/pdf/2602.20926v1
+
+## 10121. Airavat: An Agentic Framework for Internet Measurement
+
+- **ID:** `2602.20924` · **primary:** `cs.NI` · **Submitted:** 2026-02-24
+- **Authors:** Alagappan Ramanathan, Eunju Kang, Dongsu Han, et al.
+- **abs:** https://arxiv.org/abs/2602.20924 · **PDF:** https://arxiv.org/pdf/2602.20924v1
+
+## 10122. When Alpha Breaks: Two-Level Uncertainty for Safe Deployment of Cross-Sectional Stock Rankers
+
+- **ID:** `2603.13252` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Ursina Sanderink
+- **abs:** https://arxiv.org/abs/2603.13252 · **PDF:** https://arxiv.org/pdf/2603.13252v1
+
+## 10123. On the Generalization Behavior of Deep Residual Networks From a Dynamical System Perspective
+
+- **ID:** `2602.20921` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Jinshu Huang, Mingfei Sun, Chunlin Wu
+- **abs:** https://arxiv.org/abs/2602.20921 · **PDF:** https://arxiv.org/pdf/2602.20921v1
+
+## 10124. Predicting Sentence Acceptability Judgments in Multimodal Contexts
+
+- **ID:** `2602.20918` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Hyewon Jang, Nikolai Ilinykh, Sharid Loáiciga, et al.
+- **abs:** https://arxiv.org/abs/2602.20918 · **PDF:** https://arxiv.org/pdf/2602.20918v1
+
+## 10125. From Isolation to Integration: Building an Adaptive Expert Forest for Pre-Trained Model-based Class-Incremental Learning
+
+- **ID:** `2602.20911` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Ruiqi Liu, Boyu Diao, Hangda Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.20911 · **PDF:** https://arxiv.org/pdf/2602.20911v1
+
+## 10126. Graph Property Inference in Small Language Models: Effects of Representation and Inference Strategy
+
+- **ID:** `2603.06635` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Michal Podstawski
+- **abs:** https://arxiv.org/abs/2603.06635 · **PDF:** https://arxiv.org/pdf/2603.06635v1
+
+## 10127. Transcoder Adapters for Reasoning-Model Diffing
+
+- **ID:** `2602.20904` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Nathan Hu, Jake Ward, Thomas Icard, et al.
+- **abs:** https://arxiv.org/abs/2602.20904 · **PDF:** https://arxiv.org/pdf/2602.20904v1
+
+## 10128. SpatiaLQA: A Benchmark for Evaluating Spatial Logical Reasoning in Vision-Language Models
+
+- **ID:** `2602.20901` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Yuechen Xie, Xiaoyan Zhang, Yicheng Shan, et al.
+- **abs:** https://arxiv.org/abs/2602.20901 · **PDF:** https://arxiv.org/pdf/2602.20901v1
+
+## 10129. Exa-PSD: a new Persian sentiment analysis dataset on Twitter
+
+- **ID:** `2602.20892` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Seyed Himan Ghaderi, Saeed Sarbazi Azad, Mohammad Mehdi Jaziriyan, et al.
+- **abs:** https://arxiv.org/abs/2602.20892 · **PDF:** https://arxiv.org/pdf/2602.20892v1
+
+## 10130. Diagnosing Causal Reasoning in Vision-Language Models via Structured Relevance Graphs
+
+- **ID:** `2602.20878` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Dhita Putri Pratama, Soyeon Caren Han, Yihao Ding
+- **abs:** https://arxiv.org/abs/2602.20878 · **PDF:** https://arxiv.org/pdf/2602.20878v1
+
+## 10131. E-MMKGR: A Unified Multimodal Knowledge Graph Framework for E-commerce Applications
+
+- **ID:** `2602.20877` · **primary:** `cs.IR` · **Submitted:** 2026-02-24
+- **Authors:** Jiwoo Kang, Yeon-Chang Lee
+- **abs:** https://arxiv.org/abs/2602.20877 · **PDF:** https://arxiv.org/pdf/2602.20877v1
+
+## 10132. Efficient Online Learning in Interacting Particle Systems
+
+- **ID:** `2602.20875` · **primary:** `math.ST` · **Submitted:** 2026-02-24
+- **Authors:** Louis Sharrock, Nikolas Kantas, Grigorios A. Pavliotis
+- **abs:** https://arxiv.org/abs/2602.20875 · **PDF:** https://arxiv.org/pdf/2602.20875v1
+
+## 10133. SoK: Agentic Skills -- Beyond Tool Use in LLM Agents
+
+- **ID:** `2602.20867` · **primary:** `cs.CR` · **Submitted:** 2026-02-24
+- **Authors:** Yanna Jiang, Delong Li, Haiyu Deng, et al.
+- **abs:** https://arxiv.org/abs/2602.20867 · **PDF:** https://arxiv.org/pdf/2602.20867v1
+
+## 10134. ManiBench: A Benchmark for Testing Visual-Logic Drift and Syntactic Hallucinations in Manim Code Generation
+
+- **ID:** `2603.13251` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Nabin Oli
+- **abs:** https://arxiv.org/abs/2603.13251 · **PDF:** https://arxiv.org/pdf/2603.13251v1
+
+## 10135. FinAnchor: Aligned Multi-Model Representations for Financial Prediction
+
+- **ID:** `2602.20859` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Zirui He, Huopu Zhang, Yanguang Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.20859 · **PDF:** https://arxiv.org/pdf/2602.20859v1
+
+## 10136. Group Orthogonalized Policy Optimization:Group Policy Optimization as Orthogonal Projection in Hilbert Space
+
+- **ID:** `2602.21269` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Wang Zixian
+- **abs:** https://arxiv.org/abs/2602.21269 · **PDF:** https://arxiv.org/pdf/2602.21269v1
+
+## 10137. Functional Continuous Decomposition
+
+- **ID:** `2602.20857` · **primary:** `eess.SP` · **Submitted:** 2026-02-24
+- **Authors:** Teymur Aghayev
+- **abs:** https://arxiv.org/abs/2602.20857 · **PDF:** https://arxiv.org/pdf/2602.20857v1
+
+## 10138. Stochastic Discount Factors with Cross-Asset Spillovers
+
+- **ID:** `2602.20856` · **primary:** `q-fin.CP` · **Submitted:** 2026-02-24
+- **Authors:** Doron Avramov, Xin He
+- **abs:** https://arxiv.org/abs/2602.20856 · **PDF:** https://arxiv.org/pdf/2602.20856v1
+
+## 10139. M-Gaussian: An Magnetic Gaussian Framework for Efficient Multi-Stack MRI Reconstruction
+
+- **ID:** `2603.00145` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Kangyuan Zheng, Xuan Cai, Jiangqi Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.00145 · **PDF:** https://arxiv.org/pdf/2603.00145v1
+
+## 10140. The Missing Red Line: How Commercial Pressure Erodes AI Safety Boundaries
+
+- **ID:** `2603.13250` · **primary:** `cs.CY` · **Submitted:** 2026-02-24
+- **Authors:** Nora Petrova, John Burden
+- **abs:** https://arxiv.org/abs/2603.13250 · **PDF:** https://arxiv.org/pdf/2603.13250v1
+
+## 10141. A new Uncertainty Principle in Machine Learning
+
+- **ID:** `2603.06634` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** V. Dolotin, A. Morozov
+- **abs:** https://arxiv.org/abs/2603.06634 · **PDF:** https://arxiv.org/pdf/2603.06634v1
+
+## 10142. Body-Reservoir Governance in Repeated Games: Embodied Decision-Making, Dynamic Sentinel Adaptation, and Complexity-Regularized Optimization
+
+- **ID:** `2602.20846` · **primary:** `cs.GT` · **Submitted:** 2026-02-24
+- **Authors:** Yuki Nakamura
+- **abs:** https://arxiv.org/abs/2602.20846 · **PDF:** https://arxiv.org/pdf/2602.20846v1
+
+## 10143. Maximum entropy based testing in network models: ERGMs and constrained optimization
+
+- **ID:** `2602.20844` · **primary:** `math.ST` · **Submitted:** 2026-02-24
+- **Authors:** Subhro Ghosh, Rathindra Nath Karmakar, Samriddha Lahiry
+- **abs:** https://arxiv.org/abs/2602.20844 · **PDF:** https://arxiv.org/pdf/2602.20844v2
+
+## 10144. Analysis of LLMs Against Prompt Injection and Jailbreak Attacks
+
+- **ID:** `2602.22242` · **primary:** `cs.CR` · **Submitted:** 2026-02-24
+- **Authors:** Piyush Jaiswal, Aaditya Pratap, Shreyansh Saraswati, et al.
+- **abs:** https://arxiv.org/abs/2602.22242 · **PDF:** https://arxiv.org/pdf/2602.22242v1
+
+## 10145. RedFuser: An Automatic Operator Fusion Framework for Cascaded Reductions on AI Accelerators
+
+- **ID:** `2603.10026` · **primary:** `cs.AR` · **Submitted:** 2026-02-24
+- **Authors:** Xinsheng Tang, Yangcheng Li, Nan Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.10026 · **PDF:** https://arxiv.org/pdf/2603.10026v1
+
+## 10146. DRESS: A Continuous Framework for Structural Graph Refinement
+
+- **ID:** `2602.20833` · **primary:** `cs.DS` · **Submitted:** 2026-02-24
+- **Authors:** Eduar Castrillo Velilla
+- **abs:** https://arxiv.org/abs/2602.20833 · **PDF:** https://arxiv.org/pdf/2602.20833v6
+
+## 10147. Disentangled Hierarchical VAE for 3D Human-Human Interaction Generation
+
+- **ID:** `2603.00144` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Zichen Geng, Zeeshan Hayder, Bo Miao, et al.
+- **abs:** https://arxiv.org/abs/2603.00144 · **PDF:** https://arxiv.org/pdf/2603.00144v1
+
+## 10148. GrapHist: Graph Self-Supervised Learning for Histopathology
+
+- **ID:** `2603.00143` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Sevda Öğüt, Cédric Vincent-Cuaz, Natalia Dubljevic, et al.
+- **abs:** https://arxiv.org/abs/2603.00143 · **PDF:** https://arxiv.org/pdf/2603.00143v1
+
+## 10149. A Dynamic Survey of Soft Set Theory and Its Extensions
+
+- **ID:** `2602.21268` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Takaaki Fujita, Florentin Smarandache
+- **abs:** https://arxiv.org/abs/2602.21268 · **PDF:** https://arxiv.org/pdf/2602.21268v1
+
+## 10150. Don't Ignore the Tail: Decoupling top-K Probabilities for Efficient Language Model Distillation
+
+- **ID:** `2602.20816` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Sayantan Dasgupta, Trevor Cohn, Timothy Baldwin
+- **abs:** https://arxiv.org/abs/2602.20816 · **PDF:** https://arxiv.org/pdf/2602.20816v1
+
+## 10151. Pressure Reveals Character: Behavioural Alignment Evaluation at Depth
+
+- **ID:** `2602.20813` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Nora Petrova, John Burden
+- **abs:** https://arxiv.org/abs/2602.20813 · **PDF:** https://arxiv.org/pdf/2602.20813v1
+
+## 10152. Qwen-BIM: developing large language model for BIM-based design with domain-specific benchmark and dataset
+
+- **ID:** `2602.20812` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Jia-Rui Lin, Yun-Hong Cai, Xiang-Rui Ni, et al.
+- **abs:** https://arxiv.org/abs/2602.20812 · **PDF:** https://arxiv.org/pdf/2602.20812v1
+
+## 10153. POMDPPlanners: Open-Source Package for POMDP Planning
+
+- **ID:** `2602.20810` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Yaacov Pariente, Vadim Indelman
+- **abs:** https://arxiv.org/abs/2602.20810 · **PDF:** https://arxiv.org/pdf/2602.20810v1
+
+## 10154. Regret-Guided Search Control for Efficient Learning in AlphaZero
+
+- **ID:** `2602.20809` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Yun-Jui Tsai, Wei-Yu Chen, Yan-Ru Ju, et al.
+- **abs:** https://arxiv.org/abs/2602.20809 · **PDF:** https://arxiv.org/pdf/2602.20809v1
+
+## 10155. A Systematic Review of Algorithmic Red Teaming Methodologies for Assurance and Security of AI Applications
+
+- **ID:** `2602.21267` · **primary:** `cs.CR` · **Submitted:** 2026-02-24
+- **Authors:** Shruti Srivastava, Kiranmayee Janardhan, Shaurya Jauhari
+- **abs:** https://arxiv.org/abs/2602.21267 · **PDF:** https://arxiv.org/pdf/2602.21267v1
+
+## 10156. Assessing the Impact of Speaker Identity in Speech Spoofing Detection
+
+- **ID:** `2602.20805` · **primary:** `cs.SD` · **Submitted:** 2026-02-24
+- **Authors:** Anh-Tuan Dao, Driss Matrouf, Nicholas Evans
+- **abs:** https://arxiv.org/abs/2602.20805 · **PDF:** https://arxiv.org/pdf/2602.20805v1
+
+## 10157. Probing Dec-POMDP Reasoning in Cooperative MARL
+
+- **ID:** `2602.20804` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Kale-ab Tessera, Leonard Hinckeldey, Riccardo Zamboni, et al.
+- **abs:** https://arxiv.org/abs/2602.20804 · **PDF:** https://arxiv.org/pdf/2602.20804v1
+
+## 10158. Exploring the Impact of Parameter Update Magnitude on Forgetting and Generalization of Continual Learning
+
+- **ID:** `2602.20796` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** JinLi He, Liang Bai, Xian Yang
+- **abs:** https://arxiv.org/abs/2602.20796 · **PDF:** https://arxiv.org/pdf/2602.20796v1
+
+## 10159. Understanding the Role of Rehearsal Scale in Continual Learning under Varying Model Capacities
+
+- **ID:** `2602.20791` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** JinLi He, Liang Bai, Xian Yang
+- **abs:** https://arxiv.org/abs/2602.20791 · **PDF:** https://arxiv.org/pdf/2602.20791v1
+
+## 10160. On Electric Vehicle Energy Demand Forecasting and the Effect of Federated Learning
+
+- **ID:** `2602.20782` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Andreas Tritsarolis, Gil Sampaio, Nikos Pelekis, et al.
+- **abs:** https://arxiv.org/abs/2602.20782 · **PDF:** https://arxiv.org/pdf/2602.20782v1
+
+## 10161. Pipeline for Verifying LLM-Generated Mathematical Solutions
+
+- **ID:** `2602.20770` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Varvara Sazonova, Dmitri Shmelkin, Stanislav Kikot, et al.
+- **abs:** https://arxiv.org/abs/2602.20770 · **PDF:** https://arxiv.org/pdf/2602.20770v1
+
+## 10162. Overton Pluralistic Reinforcement Learning for Large Language Models
+
+- **ID:** `2602.20759` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Yu Fu, Seongho Son, Ilija Bogunovic
+- **abs:** https://arxiv.org/abs/2602.20759 · **PDF:** https://arxiv.org/pdf/2602.20759v1
+
+## 10163. Deep unfolding of MCMC kernels: scalable, modular & explainable GANs for high-dimensional posterior sampling
+
+- **ID:** `2602.20758` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Jonathan Spence, Tobías I. Liaudat, Konstantinos Zygalakis, et al.
+- **abs:** https://arxiv.org/abs/2602.20758 · **PDF:** https://arxiv.org/pdf/2602.20758v1
+
+## 10164. OrthoDiffusion: A Generalizable Multi-Task Diffusion Foundation Model for Musculoskeletal MRI Interpretation
+
+- **ID:** `2602.20752` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Tian Lan, Lei Xu, Zimu Yuan, et al.
+- **abs:** https://arxiv.org/abs/2602.20752 · **PDF:** https://arxiv.org/pdf/2602.20752v1
+
+## 10165. SibylSense: Adaptive Rubric Learning via Memory Tuning and Adversarial Probing
+
+- **ID:** `2602.20751` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Yifei Xu, Guilherme Potje, Shivam Shandilya, et al.
+- **abs:** https://arxiv.org/abs/2602.20751 · **PDF:** https://arxiv.org/pdf/2602.20751v1
+
+## 10166. Explicit Grammar Semantic Feature Fusion for Robust Text Classification
+
+- **ID:** `2602.20749` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Azrin Sultana, Firoz Ahmed
+- **abs:** https://arxiv.org/abs/2602.20749 · **PDF:** https://arxiv.org/pdf/2602.20749v1
+
+## 10167. Structure-Aware Epistemic Uncertainty Quantification for Neural Operator PDE Surrogates
+
+- **ID:** `2603.11052` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Haoze Song, Zhihao Li, Mengyi Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.11052 · **PDF:** https://arxiv.org/pdf/2603.11052v1
+
+## 10168. Voices of the Mountains: Deep Learning-Based Vocal Error Detection System for Kurdish Maqams
+
+- **ID:** `2602.20744` · **primary:** `cs.SD` · **Submitted:** 2026-02-24
+- **Authors:** Darvan Shvan Khairaldeen, Hossein Hassani
+- **abs:** https://arxiv.org/abs/2602.20744 · **PDF:** https://arxiv.org/pdf/2602.20744v1
+
+## 10169. Stochastic Neural Networks for Quantum Devices
+
+- **ID:** `2602.22241` · **primary:** `quant-ph` · **Submitted:** 2026-02-24
+- **Authors:** Bodo Rosenhahn, Tobias J. Osborne, Christoph Hirche
+- **abs:** https://arxiv.org/abs/2602.22241 · **PDF:** https://arxiv.org/pdf/2602.22241v1
+
+## 10170. Adaptive Text Anonymization: Learning Privacy-Utility Trade-offs via Prompt Optimization
+
+- **ID:** `2602.20743` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Gabriel Loiseau, Damien Sileo, Damien Riquet, et al.
+- **abs:** https://arxiv.org/abs/2602.20743 · **PDF:** https://arxiv.org/pdf/2602.20743v1
+
+## 10171. PyVision-RL: Forging Open Agentic Vision Models via RL
+
+- **ID:** `2602.20739` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Shitian Zhao, Shaoheng Lin, Ming Li, et al.
+- **abs:** https://arxiv.org/abs/2602.20739 · **PDF:** https://arxiv.org/pdf/2602.20739v1
+
+## 10172. RMIT-ADM+S at the MMU-RAG NeurIPS 2025 Competition
+
+- **ID:** `2602.20735` · **primary:** `cs.IR` · **Submitted:** 2026-02-24
+- **Authors:** Kun Ran, Marwah Alaofi, Danula Hettiachchi, et al.
+- **abs:** https://arxiv.org/abs/2602.20735 · **PDF:** https://arxiv.org/pdf/2602.20735v1
+
+## 10173. CHESS: Context-aware Hierarchical Efficient Semantic Selection for Long-Context LLM Inference
+
+- **ID:** `2602.20732` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Chao Fei, Guozhong Li, Chenxi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.20732 · **PDF:** https://arxiv.org/pdf/2602.20732v1
+
+## 10174. Communication-Inspired Tokenization for Structured Image Representations
+
+- **ID:** `2602.20731` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Aram Davtyan, Yusuf Sahin, Yasaman Haghighi, et al.
+- **abs:** https://arxiv.org/abs/2602.20731 · **PDF:** https://arxiv.org/pdf/2602.20731v1
+
+## 10175. Rethink Efficiency Side of Neural Combinatorial Solver: An Offline and Self-Play Paradigm
+
+- **ID:** `2602.20730` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Zhenxing Xu, Zeyuan Ma, Weidong Bao, et al.
+- **abs:** https://arxiv.org/abs/2602.20730 · **PDF:** https://arxiv.org/pdf/2602.20730v1
+
+## 10176. Fuz-RL: A Fuzzy-Guided Robust Framework for Safe Reinforcement Learning under Uncertainty
+
+- **ID:** `2602.20729` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Xu Wan, Chao Yang, Cheng Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.20729 · **PDF:** https://arxiv.org/pdf/2602.20729v2
+
+## 10177. From Prompts to Performance: Evaluating LLMs for Task-based Parallel Code Generation
+
+- **ID:** `2602.22240` · **primary:** `cs.PL` · **Submitted:** 2026-02-24
+- **Authors:** Linus Bantel, Moritz Strack, Alexander Strack, et al.
+- **abs:** https://arxiv.org/abs/2602.22240 · **PDF:** https://arxiv.org/pdf/2602.22240v1
+
+## 10178. Balancing Multiple Objectives in Urban Traffic Control with Reinforcement Learning from AI Feedback
+
+- **ID:** `2602.20728` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Chenyang Zhao, Vinny Cahill, Ivana Dusparic
+- **abs:** https://arxiv.org/abs/2602.20728 · **PDF:** https://arxiv.org/pdf/2602.20728v1
+
+## 10179. ID-LoRA: Efficient Low-Rank Adaptation Inspired by Matrix Interpolative Decomposition
+
+- **ID:** `2602.20727` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Xindian Ma, Rundong Kong, Peng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.20727 · **PDF:** https://arxiv.org/pdf/2602.20727v1
+
+## 10180. Molecular Representations for AI in Chemistry and Materials Science: An NLP Perspective
+
+- **ID:** `2603.05525` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-24
+- **Authors:** Sanjanasri JP, Pratiti Bhadra, N. Sukumar, et al.
+- **abs:** https://arxiv.org/abs/2603.05525 · **PDF:** https://arxiv.org/pdf/2603.05525v1
+
+## 10181. Modality-Guided Mixture of Graph Experts with Entropy-Triggered Routing for Multimodal Recommendation
+
+- **ID:** `2602.20723` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Ji Dai, Quan Fang, Dengsheng Cai
+- **abs:** https://arxiv.org/abs/2602.20723 · **PDF:** https://arxiv.org/pdf/2602.20723v2
+
+## 10182. Buffer Matters: Unleashing the Power of Off-Policy Reinforcement Learning in Large Language Model Reasoning
+
+- **ID:** `2602.20722` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Xu Wan, Yansheng Wang, Wenqi Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.20722 · **PDF:** https://arxiv.org/pdf/2602.20722v2
+
+## 10183. AdapTools: Adaptive Tool-based Indirect Prompt Injection Attacks on Agentic LLMs
+
+- **ID:** `2602.20720` · **primary:** `cs.CR` · **Submitted:** 2026-02-24
+- **Authors:** Che Wang, Jiaming Zhang, Ziqi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.20720 · **PDF:** https://arxiv.org/pdf/2602.20720v1
+
+## 10184. ToolMATH: A Math Tool Benchmark for Realistic Long-Horizon Multi-Tool Reasoning
+
+- **ID:** `2602.21265` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Hyeonje Choi, Jeongsoo Lee, Hyojun Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.21265 · **PDF:** https://arxiv.org/pdf/2602.21265v1
+
+## 10185. WeirNet: A Large-Scale 3D CFD Benchmark for Geometric Surrogate Modeling of Piano Key Weirs
+
+- **ID:** `2602.20714` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Lisa Lüddecke, Michael Hohmann, Sebastian Eilermann, et al.
+- **abs:** https://arxiv.org/abs/2602.20714 · **PDF:** https://arxiv.org/pdf/2602.20714v2
+
+## 10186. F10.7 Index Prediction: A Multiscale Decomposition Strategy with Wavelet Transform for Performance Optimization
+
+- **ID:** `2602.20712` · **primary:** `astro-ph.IM` · **Submitted:** 2026-02-24
+- **Authors:** Xuran Ma, Xuebao Li, Yanfang Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.20712 · **PDF:** https://arxiv.org/pdf/2602.20712v1
+
+## 10187. Real-Time Trustworthiness Scoring for LLM Structured Outputs and Data Extraction
+
+- **ID:** `2603.18014` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Hui Wen Goh, Jonas Mueller
+- **abs:** https://arxiv.org/abs/2603.18014 · **PDF:** https://arxiv.org/pdf/2603.18014v2
+
+## 10188. Evaluating Theory of Mind and Internal Beliefs in LLM-Based Multi-Agent Systems
+
+- **ID:** `2603.00142` · **primary:** `cs.MA` · **Submitted:** 2026-02-24
+- **Authors:** Adam Kostka, Jarosław A. Chudziak
+- **abs:** https://arxiv.org/abs/2603.00142 · **PDF:** https://arxiv.org/pdf/2603.00142v1
+
+## 10189. Counterfactual Simulation Training for Chain-of-Thought Faithfulness
+
+- **ID:** `2602.20710` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Peter Hase, Christopher Potts
+- **abs:** https://arxiv.org/abs/2602.20710 · **PDF:** https://arxiv.org/pdf/2602.20710v1
+
+## 10190. Onboard-Targeted Segmentation of Straylight in Space Camera Sensors
+
+- **ID:** `2602.20709` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Riccardo Gallon, Fabian Schiemenz, Alessandra Menicucci, et al.
+- **abs:** https://arxiv.org/abs/2602.20709 · **PDF:** https://arxiv.org/pdf/2602.20709v1
+
+## 10191. From Scale to Speed: Adaptive Test-Time Scaling for Image Editing
+
+- **ID:** `2603.00141` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Xiangyan Qu, Zhenlong Yuan, Jing Tang, et al.
+- **abs:** https://arxiv.org/abs/2603.00141 · **PDF:** https://arxiv.org/pdf/2603.00141v3
+
+## 10192. ICON: Indirect Prompt Injection Defense for Agents based on Inference-Time Correction
+
+- **ID:** `2602.20708` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Che Wang, Fuyao Zhang, Jiaming Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.20708 · **PDF:** https://arxiv.org/pdf/2602.20708v1
+
+## 10193. Online Algorithms with Unreliable Guidance
+
+- **ID:** `2602.20706` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Julien Dallot, Yuval Emek, Yuval Gil, et al.
+- **abs:** https://arxiv.org/abs/2602.20706 · **PDF:** https://arxiv.org/pdf/2602.20706v1
+
+## 10194. Steering Away from Memorization: Reachability-Constrained Reinforcement Learning for Text-to-Image Diffusion
+
+- **ID:** `2603.00140` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Sathwik Karnik, Juyeop Kim, Sanmi Koyejo, et al.
+- **abs:** https://arxiv.org/abs/2603.00140 · **PDF:** https://arxiv.org/pdf/2603.00140v1
+
+## 10195. High-Dimensional Robust Mean Estimation with Untrusted Batches
+
+- **ID:** `2602.20698` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Maryam Aliakbarpour, Vladimir Braverman, Yuhan Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.20698 · **PDF:** https://arxiv.org/pdf/2602.20698v1
+
+## 10196. PromptCD: Test-Time Behavior Enhancement via Polarity-Prompt Contrastive Decoding
+
+- **ID:** `2602.20696` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Baolong Bi, Yuyao Ge, Shenghua Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.20696 · **PDF:** https://arxiv.org/pdf/2602.20696v1
+
+## 10197. How Foundational Skills Influence VLM-based Embodied Agents:A Native Perspective
+
+- **ID:** `2602.20687` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Bo Peng, Pi Bu, Keyu Pan, et al.
+- **abs:** https://arxiv.org/abs/2602.20687 · **PDF:** https://arxiv.org/pdf/2602.20687v1
+
+## 10198. Agile V: A Compliance-Ready Framework for AI-Augmented Engineering -- From Concept to Audit-Ready Delivery
+
+- **ID:** `2602.20684` · **primary:** `cs.SE` · **Submitted:** 2026-02-24
+- **Authors:** Christopher Koch, Joshua Andreas Wellbrock
+- **abs:** https://arxiv.org/abs/2602.20684 · **PDF:** https://arxiv.org/pdf/2602.20684v1
+
+## 10199. UrbanFM: Scaling Urban Spatio-Temporal Foundation Models
+
+- **ID:** `2602.20677` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Wei Chen, Yuqian Wu, Junle Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.20677 · **PDF:** https://arxiv.org/pdf/2602.20677v2
+
+## 10200. PRECTR-V2:Unified Relevance-CTR Framework with Cross-User Preference Mining, Exposure Bias Correction, and LLM-Distilled Encoder Optimization
+
+- **ID:** `2602.20676` · **primary:** `cs.IR` · **Submitted:** 2026-02-24
+- **Authors:** Shuzhi Cao, Rong Chen, Ailong He, et al.
+- **abs:** https://arxiv.org/abs/2602.20676 · **PDF:** https://arxiv.org/pdf/2602.20676v1
+
+## 10201. Bikelution: Federated Gradient-Boosting for Scalable Shared Micro-Mobility Demand Forecasting
+
+- **ID:** `2602.20671` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Antonios Tziorvas, Andreas Tritsarolis, Yannis Theodoridis
+- **abs:** https://arxiv.org/abs/2602.20671 · **PDF:** https://arxiv.org/pdf/2602.20671v1
+
+## 10202. CAMEL: Confidence-Gated Reflection for Reward Modeling
+
+- **ID:** `2602.20670` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Zirui Zhu, Hailun Xu, Yang Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.20670 · **PDF:** https://arxiv.org/pdf/2602.20670v1
+
+## 10203. VAE-MS: An Asymmetric Variational Autoencoder for Mutational Signature Extraction
+
+- **ID:** `2602.22239` · **primary:** `stat.AP` · **Submitted:** 2026-02-24
+- **Authors:** Ida Egendal, Rasmus Froberg Brøndum, Dan J Woodcock, et al.
+- **abs:** https://arxiv.org/abs/2602.22239 · **PDF:** https://arxiv.org/pdf/2602.22239v1
+
+## 10204. Recursive Belief Vision Language Action Models
+
+- **ID:** `2602.20659` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Vaidehi Bagaria, Bijo Sebastian, Nirav Kumar Patel
+- **abs:** https://arxiv.org/abs/2602.20659 · **PDF:** https://arxiv.org/pdf/2602.20659v2
+
+## 10205. Vision-Language Models for Ergonomic Assessment of Manual Lifting Tasks: Estimating Horizontal and Vertical Hand Distances from RGB Video
+
+- **ID:** `2602.20658` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Mohammad Sadra Rajabi, Aanuoluwapo Ojelade, Sunwook Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.20658 · **PDF:** https://arxiv.org/pdf/2602.20658v1
+
+## 10206. DANCE: Doubly Adaptive Neighborhood Conformal Estimation
+
+- **ID:** `2602.20652` · **primary:** `stat.ML` · **Submitted:** 2026-02-24
+- **Authors:** Brandon R. Feng, Brian J. Reich, Daniel Beaglehole, et al.
+- **abs:** https://arxiv.org/abs/2602.20652 · **PDF:** https://arxiv.org/pdf/2602.20652v1
+
+## 10207. Sparse Bayesian Deep Functional Learning with Structured Region Selection
+
+- **ID:** `2602.20651` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Xiaoxian Zhu, Yingmeng Li, Shuangge Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.20651 · **PDF:** https://arxiv.org/pdf/2602.20651v2
+
+## 10208. Dataset Color Quantization: A Training-Oriented Framework for Dataset-Level Compression
+
+- **ID:** `2602.20650` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Chenyue Yu, Lingao Xiao, Jinhong Deng, et al.
+- **abs:** https://arxiv.org/abs/2602.20650 · **PDF:** https://arxiv.org/pdf/2602.20650v2
+
+## 10209. CARE: An Explainable Computational Framework for Assessing Client-Perceived Therapeutic Alliance Using Large Language Models
+
+- **ID:** `2602.20648` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Anqi Li, Chenxiao Wang, Yu Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.20648 · **PDF:** https://arxiv.org/pdf/2602.20648v1
+
+## 10210. Semantic Novelty at Scale: Narrative Shape Taxonomy and Readership Prediction in 28,606 Books
+
+- **ID:** `2602.20647` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** W. Frederick Zimmerman
+- **abs:** https://arxiv.org/abs/2602.20647 · **PDF:** https://arxiv.org/pdf/2602.20647v1
+
+## 10211. On the Convergence of Stochastic Gradient Descent with Perturbed Forward-Backward Passes
+
+- **ID:** `2602.20646` · **primary:** `math.OC` · **Submitted:** 2026-02-24
+- **Authors:** Boao Kong, Hengrui Zhang, Kun Yuan
+- **abs:** https://arxiv.org/abs/2602.20646 · **PDF:** https://arxiv.org/pdf/2602.20646v1
+
+## 10212. TrajGPT-R: Generating Urban Mobility Trajectory with Reinforcement Learning-Enhanced Generative Pre-trained Transformer
+
+- **ID:** `2602.20643` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Jiawei Wang, Chuang Yang, Jiawei Yong, et al.
+- **abs:** https://arxiv.org/abs/2602.20643 · **PDF:** https://arxiv.org/pdf/2602.20643v1
+
+## 10213. Scalable multitask Gaussian processes for complex mechanical systems with functional covariates
+
+- **ID:** `2602.20640` · **primary:** `math.ST` · **Submitted:** 2026-02-24
+- **Authors:** Razak Christophe Sabi Gninkou, Andrés F. López-Lopera, Franck Massa, et al.
+- **abs:** https://arxiv.org/abs/2602.20640 · **PDF:** https://arxiv.org/pdf/2602.20640v2
+
+## 10214. Grounding LLMs in Scientific Discovery via Embodied Actions
+
+- **ID:** `2602.20639` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Bo Zhang, Jinfeng Zhou, Yuxuan Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.20639 · **PDF:** https://arxiv.org/pdf/2602.20639v1
+
+## 10215. Identifying two piecewise linear additive value functions from anonymous preference information
+
+- **ID:** `2602.20638` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Vincent Auriau, Khaled Belahcene, Emmanuel Malherbe, et al.
+- **abs:** https://arxiv.org/abs/2602.20638 · **PDF:** https://arxiv.org/pdf/2602.20638v1
+
+## 10216. SurgAtt-Tracker: Online Surgical Attention Tracking via Temporal Proposal Reranking and Motion-Aware Refinement
+
+- **ID:** `2602.20636` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Rulin Zhou, Guankun Wang, An Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.20636 · **PDF:** https://arxiv.org/pdf/2602.20636v1
+
+## 10217. Steering at the Source: Style Modulation Heads for Robust Persona Control
+
+- **ID:** `2603.13249` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Yoshihiro Izawa, Gouki Minegishi, Koshi Eguchi, et al.
+- **abs:** https://arxiv.org/abs/2603.13249 · **PDF:** https://arxiv.org/pdf/2603.13249v1
+
+## 10218. Enhancing Hate Speech Detection on Social Media: A Comparative Analysis of Machine Learning Models and Text Transformation Approaches
+
+- **ID:** `2602.20634` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Saurabh Mishra, Shivani Thakur, Radhika Mamidi
+- **abs:** https://arxiv.org/abs/2602.20634 · **PDF:** https://arxiv.org/pdf/2602.20634v1
+
+## 10219. QEDBENCH: Quantifying the Alignment Gap in Automated Evaluation of University-Level Mathematical Proofs
+
+- **ID:** `2602.20629` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Santiago Gonzalez, Alireza Amiri Bavandpour, Peter Ye, et al.
+- **abs:** https://arxiv.org/abs/2602.20629 · **PDF:** https://arxiv.org/pdf/2602.20629v2
+
+## 10220. When can we trust untrusted monitoring? A safety case sketch across collusion strategies
+
+- **ID:** `2602.20628` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Nelson Gardner-Challis, Jonathan Bostock, Georgiy Kozhevnikov, et al.
+- **abs:** https://arxiv.org/abs/2602.20628 · **PDF:** https://arxiv.org/pdf/2602.20628v1
+
+## 10221. Physics-based phenomenological characterization of cross-modal bias in multimodal models
+
+- **ID:** `2602.20624` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Hyeongmo Kim, Sohyun Kang, Yerin Choi, et al.
+- **abs:** https://arxiv.org/abs/2602.20624 · **PDF:** https://arxiv.org/pdf/2602.20624v1
+
+## 10222. Knowing the Unknown: Interpretable Open-World Object Detection via Concept Decomposition Model
+
+- **ID:** `2602.20616` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Xueqiang Lv, Shizhou Zhang, Yinghui Xing, et al.
+- **abs:** https://arxiv.org/abs/2602.20616 · **PDF:** https://arxiv.org/pdf/2602.20616v1
+
+## 10223. Amortized Bayesian inference for actigraph time sheet data from mobile devices
+
+- **ID:** `2602.20611` · **primary:** `stat.ML` · **Submitted:** 2026-02-24
+- **Authors:** Daniel Zhou, Sudipto Banerjee
+- **abs:** https://arxiv.org/abs/2602.20611 · **PDF:** https://arxiv.org/pdf/2602.20611v1
+
+## 10224. SpecMind: Cognitively Inspired, Interactive Multi-Turn Framework for Postcondition Inference
+
+- **ID:** `2602.20610` · **primary:** `cs.SE` · **Submitted:** 2026-02-24
+- **Authors:** Cuong Chi Le, Minh V. T Pham, Tung Vu Duy, et al.
+- **abs:** https://arxiv.org/abs/2602.20610 · **PDF:** https://arxiv.org/pdf/2602.20610v2
+
+## 10225. Measurement-Free Ancilla Recycling via Blind Reset: A Cross-Platform Study on Superconducting and Trapped-Ion Processors
+
+- **ID:** `2603.08733` · **primary:** `cs.AR` · **Submitted:** 2026-02-24
+- **Authors:** Sangkeum Lee
+- **abs:** https://arxiv.org/abs/2603.08733 · **PDF:** https://arxiv.org/pdf/2603.08733v1
+
+## 10226. OptiLeak: Efficient Prompt Reconstruction via Reinforcement Learning in Multi-tenant LLM Services
+
+- **ID:** `2602.20595` · **primary:** `cs.CR` · **Submitted:** 2026-02-24
+- **Authors:** Longxiang Wang, Xiang Zheng, Xuhao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.20595 · **PDF:** https://arxiv.org/pdf/2602.20595v1
+
+## 10227. Is the Trigger Essential? A Feature-Based Triggerless Backdoor Attack in Vertical Federated Learning
+
+- **ID:** `2602.20593` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Yige Liu, Yiwei Lou, Che Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.20593 · **PDF:** https://arxiv.org/pdf/2602.20593v1
+
+## 10228. OpenSanctions Pairs: Large-Scale Entity Matching with LLMs
+
+- **ID:** `2603.11051` · **primary:** `cs.IR` · **Submitted:** 2026-02-24
+- **Authors:** Chandler Smith, Magnus Sesodia, Friedrich Lindenberg, et al.
+- **abs:** https://arxiv.org/abs/2603.11051 · **PDF:** https://arxiv.org/pdf/2603.11051v1
+
+## 10229. Characterizing Online and Private Learnability under Distributional Constraints via Generalized Smoothness
+
+- **ID:** `2602.20585` · **primary:** `stat.ML` · **Submitted:** 2026-02-24
+- **Authors:** Moïse Blanchard, Abhishek Shetty, Alexander Rakhlin
+- **abs:** https://arxiv.org/abs/2602.20585 · **PDF:** https://arxiv.org/pdf/2602.20585v1
+
+## 10230. Personal Information Parroting in Language Models
+
+- **ID:** `2602.20580` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Nishant Subramani, Kshitish Ghate, Mona Diab
+- **abs:** https://arxiv.org/abs/2602.20580 · **PDF:** https://arxiv.org/pdf/2602.20580v1
+
+## 10231. Upper-Linearizability of Online Non-Monotone DR-Submodular Maximization over Down-Closed Convex Sets
+
+- **ID:** `2602.20578` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Yiyang Lu, Haresh Jadav, Mohammad Pedramfar, et al.
+- **abs:** https://arxiv.org/abs/2602.20578 · **PDF:** https://arxiv.org/pdf/2602.20578v1
+
+## 10232. GATES: Self-Distillation under Privileged Context with Consensus Gating
+
+- **ID:** `2602.20574` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Alex Stein, Furong Huang, Tom Goldstein
+- **abs:** https://arxiv.org/abs/2602.20574 · **PDF:** https://arxiv.org/pdf/2602.20574v1
+
+## 10233. Benchmarking GNN Models on Molecular Regression Tasks with CKA-Based Representation Analysis
+
+- **ID:** `2602.20573` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Rajan, Ishaan Gupta
+- **abs:** https://arxiv.org/abs/2602.20573 · **PDF:** https://arxiv.org/pdf/2602.20573v2
+
+## 10234. MAML-KT: Addressing Cold Start Problem in Knowledge Tracing for New Students via Few-Shot Model-Agnostic Meta Learning
+
+- **ID:** `2603.00137` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Indronil Bhattacharjee, Christabel Wayllace
+- **abs:** https://arxiv.org/abs/2603.00137 · **PDF:** https://arxiv.org/pdf/2603.00137v1
+
+## 10235. TT-SEAL: TTD-Aware Selective Encryption for Adversarially-Robust and Low-Latency Edge AI
+
+- **ID:** `2602.22238` · **primary:** `cs.CR` · **Submitted:** 2026-02-24
+- **Authors:** Kyeongpil Min, Sangmin Jeon, Jae-Jin Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.22238 · **PDF:** https://arxiv.org/pdf/2602.22238v1
+
+## 10236. CausalReasoningBenchmark: A Real-World Benchmark for Disentangled Evaluation of Causal Identification and Estimation
+
+- **ID:** `2602.20571` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Ayush Sawarni, Jiyuan Tan, Vasilis Syrgkanis
+- **abs:** https://arxiv.org/abs/2602.20571 · **PDF:** https://arxiv.org/pdf/2602.20571v1
+
+## 10237. TinyVLM: Zero-Shot Object Detection on Microcontrollers via Vision-Language Distillation with Matryoshka Embeddings
+
+- **ID:** `2603.00136` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Bibin Wilson
+- **abs:** https://arxiv.org/abs/2603.00136 · **PDF:** https://arxiv.org/pdf/2603.00136v1
+
+## 10238. Stability and Generalization of Push-Sum Based Decentralized Optimization over Directed Graphs
+
+- **ID:** `2602.20567` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Yifei Liang, Yan Sun, Xiaochun Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.20567 · **PDF:** https://arxiv.org/pdf/2602.20567v1
+
+## 10239. Learned but Not Expressed: Capability-Expression Dissociation in Large Language Models
+
+- **ID:** `2603.18013` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Toshiyuki Shigemura
+- **abs:** https://arxiv.org/abs/2603.18013 · **PDF:** https://arxiv.org/pdf/2603.18013v1
+
+## 10240. From Logs to Language: Learning Optimal Verbalization for LLM-Based Recommendation at Industry Scale
+
+- **ID:** `2602.20558` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Yucheng Shi, Ying Li, Yu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.20558 · **PDF:** https://arxiv.org/pdf/2602.20558v2
+
+## 10241. GENSR: Symbolic Regression Based in Equation Generative Space
+
+- **ID:** `2602.20557` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Qian Li, Yuxiao Hu, Juncheng Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.20557 · **PDF:** https://arxiv.org/pdf/2602.20557v1
+
+## 10242. Standard Transformers Achieve the Minimax Rate in Nonparametric Regression with $C^{s,λ}$ Targets
+
+- **ID:** `2602.20555` · **primary:** `stat.ML` · **Submitted:** 2026-02-24
+- **Authors:** Yanming Lai, Defeng Sun
+- **abs:** https://arxiv.org/abs/2602.20555 · **PDF:** https://arxiv.org/pdf/2602.20555v1
+
+## 10243. Sample-efficient evidence estimation of score based priors for model selection
+
+- **ID:** `2602.20549` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Frederic Wang, Katherine L. Bouman
+- **abs:** https://arxiv.org/abs/2602.20549 · **PDF:** https://arxiv.org/pdf/2602.20549v1
+
+## 10244. Leakage Safe Graph Features for Interpretable Fraud Detection in Temporal Transaction Networks
+
+- **ID:** `2603.06632` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Hamideh Khaleghpour, Brett McKinney
+- **abs:** https://arxiv.org/abs/2603.06632 · **PDF:** https://arxiv.org/pdf/2603.06632v1
+
+## 10245. What Drives Students' Use of AI Chatbots? Technology Acceptance in Conversational AI
+
+- **ID:** `2602.20547` · **primary:** `cs.HC` · **Submitted:** 2026-02-24
+- **Authors:** Griffin Pitts, Sanaz Motamedi
+- **abs:** https://arxiv.org/abs/2602.20547 · **PDF:** https://arxiv.org/pdf/2602.20547v1
+
+## 10246. Maximin Share Guarantees via Limited Cost-Sensitive Sharing
+
+- **ID:** `2602.20541` · **primary:** `cs.GT` · **Submitted:** 2026-02-24
+- **Authors:** Hana Salavcova, Martin Černý, Arpita Biswas
+- **abs:** https://arxiv.org/abs/2602.20541 · **PDF:** https://arxiv.org/pdf/2602.20541v3
+
+## 10247. Actor-Curator: Co-adaptive Curriculum Learning via Policy-Improvement Bandits for RL Post-Training
+
+- **ID:** `2602.20532` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Zhengyao Gu, Jonathan Light, Raul Astudillo, et al.
+- **abs:** https://arxiv.org/abs/2602.20532 · **PDF:** https://arxiv.org/pdf/2602.20532v1
+
+## 10248. Memory-guided Prototypical Co-occurrence Learning for Mixed Emotion Recognition
+
+- **ID:** `2602.20530` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Ming Li, Yong-Jin Liu, Fang Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.20530 · **PDF:** https://arxiv.org/pdf/2602.20530v1
+
+## 10249. Stop-Think-AutoRegress: Language Modeling with Latent Diffusion Planning
+
+- **ID:** `2602.20528` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Justin Lovelace, Christian Belardi, Sofian Zalouk, et al.
+- **abs:** https://arxiv.org/abs/2602.20528 · **PDF:** https://arxiv.org/pdf/2602.20528v1
+
+## 10250. Under the Influence: Quantifying Persuasion and Vigilance in Large Language Models
+
+- **ID:** `2602.21262` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Sasha Robinson, Katherine M. Collins, Ilia Sucholutsky, et al.
+- **abs:** https://arxiv.org/abs/2602.21262 · **PDF:** https://arxiv.org/pdf/2602.21262v3
+
+## 10251. A Generalized Apprenticeship Learning Framework for Capturing Evolving Student Pedagogical Strategies
+
+- **ID:** `2602.20527` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Md Mirajul Islam, Xi Yang, Adittya Soukarjya Saha, et al.
+- **abs:** https://arxiv.org/abs/2602.20527 · **PDF:** https://arxiv.org/pdf/2602.20527v1
+
+## 10252. DynaRAG: Bridging Static and Dynamic Knowledge in Retrieval-Augmented Generation
+
+- **ID:** `2603.18012` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Penghao Liang, Mengwei Yuan, Jianan Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.18012 · **PDF:** https://arxiv.org/pdf/2603.18012v1
+
+## 10253. How Do Inpainting Artifacts Propagate to Language?
+
+- **ID:** `2602.20520` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Pratham Yashwante, Davit Abrahamyan, Shresth Grover, et al.
+- **abs:** https://arxiv.org/abs/2602.20520 · **PDF:** https://arxiv.org/pdf/2602.20520v1
+
+## 10254. Inner Speech as Behavior Guides: Steerable Imitation of Diverse Behaviors for Human-AI coordination
+
+- **ID:** `2602.20517` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Rakshit Trivedi, Kartik Sharma, David C Parkes
+- **abs:** https://arxiv.org/abs/2602.20517 · **PDF:** https://arxiv.org/pdf/2602.20517v1
+
+## 10255. From Performance to Purpose: A Sociotechnical Taxonomy for Evaluating Large Language Model Utility
+
+- **ID:** `2602.20513` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Gavin Levinson, Keith Feldman
+- **abs:** https://arxiv.org/abs/2602.20513 · **PDF:** https://arxiv.org/pdf/2602.20513v1
+
+## 10256. ActionEngine: From Reactive to Programmatic GUI Agents via State Machine Memory
+
+- **ID:** `2602.20502` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Hongbin Zhong, Fazle Faisal, Luis França, et al.
+- **abs:** https://arxiv.org/abs/2602.20502 · **PDF:** https://arxiv.org/pdf/2602.20502v1
+
+## 10257. MERRY: Semantically Decoupled Evaluation of Multimodal Emotional and Role Consistencies of Role-Playing Agents
+
+- **ID:** `2602.21941` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Zhenyu Wang, Xiaofen Xing, Yirong Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.21941 · **PDF:** https://arxiv.org/pdf/2602.21941v1
+
+## 10258. LESA: Learnable Stage-Aware Predictors for Diffusion Model Acceleration
+
+- **ID:** `2602.20497` · **primary:** `cs.CV` · **Submitted:** 2026-02-24
+- **Authors:** Peiliang Cai, Jiacheng Liu, Haowen Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.20497 · **PDF:** https://arxiv.org/pdf/2602.20497v2
+
+## 10259. KairosVL: Orchestrating Time Series and Semantics for Unified Reasoning
+
+- **ID:** `2602.20494` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Haotian Si, Changhua Pei, Xiao He, et al.
+- **abs:** https://arxiv.org/abs/2602.20494 · **PDF:** https://arxiv.org/pdf/2602.20494v1
+
+## 10260. Wireless Federated Multi-Task LLM Fine-Tuning via Sparse-and-Orthogonal LoRA
+
+- **ID:** `2602.20492` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Nuocheng Yang, Sihua Wang, Ouwen Huan, et al.
+- **abs:** https://arxiv.org/abs/2602.20492 · **PDF:** https://arxiv.org/pdf/2602.20492v1
+
+## 10261. Hybrid LLM-Embedded Dialogue Agents for Learner Reflection: Designing Responsive and Theory-Driven Interactions
+
+- **ID:** `2602.20486` · **primary:** `cs.HC` · **Submitted:** 2026-02-24
+- **Authors:** Paras Sharma, YuePing Sha, Janet Shufor Bih Epse Fofang, et al.
+- **abs:** https://arxiv.org/abs/2602.20486 · **PDF:** https://arxiv.org/pdf/2602.20486v1
+
+## 10262. VINA: Variational Invertible Neural Architectures
+
+- **ID:** `2602.20480` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Shubhanshu Shekhar, Mohammad Javad Khojasteh, Ananya Acharya, et al.
+- **abs:** https://arxiv.org/abs/2602.20480 · **PDF:** https://arxiv.org/pdf/2602.20480v1
+
+## 10263. PhyGHT: Physics-Guided HyperGraph Transformer for Signal Purification at the HL-LHC
+
+- **ID:** `2602.20475` · **primary:** `hep-ex` · **Submitted:** 2026-02-24
+- **Authors:** Mohammed Rakib, Luke Vaughan, Shivang Patel, et al.
+- **abs:** https://arxiv.org/abs/2602.20475 · **PDF:** https://arxiv.org/pdf/2602.20475v1
+
+## 10264. CGSTA: Cross-Scale Graph Contrast with Stability-Aware Alignment for Multivariate Time-Series Anomaly Detection
+
+- **ID:** `2602.20468` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Zhongpeng Qi, Jun Zhang, Wei Li, et al.
+- **abs:** https://arxiv.org/abs/2602.20468 · **PDF:** https://arxiv.org/pdf/2602.20468v1
+
+## 10265. Elimination-compensation pruning for fully-connected neural networks
+
+- **ID:** `2602.20467` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Enrico Ballini, Luca Muscarnera, Alessio Fumagalli, et al.
+- **abs:** https://arxiv.org/abs/2602.20467 · **PDF:** https://arxiv.org/pdf/2602.20467v1
+
+## 10266. Prior-Agnostic Incentive-Compatible Exploration
+
+- **ID:** `2602.20465` · **primary:** `cs.GT` · **Submitted:** 2026-02-24
+- **Authors:** Ramya Ramalingam, Osbert Bastani, Aaron Roth
+- **abs:** https://arxiv.org/abs/2602.20465 · **PDF:** https://arxiv.org/pdf/2602.20465v1
+
+## 10267. A Long-Short Flow-Map Perspective for Drifting Models
+
+- **ID:** `2602.20463` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Zhiqi Li, Bo Zhu
+- **abs:** https://arxiv.org/abs/2602.20463 · **PDF:** https://arxiv.org/pdf/2602.20463v1
+
+## 10268. Nonparametric Teaching of Attention Learners
+
+- **ID:** `2602.20461` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Chen Zhang, Jianghui Wang, Bingyang Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.20461 · **PDF:** https://arxiv.org/pdf/2602.20461v1
+
+## 10269. PreScience: A Benchmark for Forecasting Scientific Contributions
+
+- **ID:** `2602.20459` · **primary:** `cs.AI` · **Submitted:** 2026-02-24
+- **Authors:** Anirudh Ajith, Amanpreet Singh, Jay DeYoung, et al.
+- **abs:** https://arxiv.org/abs/2602.20459 · **PDF:** https://arxiv.org/pdf/2602.20459v1
+
+## 10270. Oracle-Robust Online Alignment for Large Language Models
+
+- **ID:** `2602.20457` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Zimeng Li, Mudit Gaur, Vaneet Aggarwal
+- **abs:** https://arxiv.org/abs/2602.20457 · **PDF:** https://arxiv.org/pdf/2602.20457v1
+
+## 10271. Heterogeneity-Aware Client Selection Methodology For Efficient Federated Learning
+
+- **ID:** `2602.20450` · **primary:** `cs.DC` · **Submitted:** 2026-02-24
+- **Authors:** Nihal Balivada, Shrey Gupta, Shashank Shreedhar Bhatt, et al.
+- **abs:** https://arxiv.org/abs/2602.20450 · **PDF:** https://arxiv.org/pdf/2602.20450v1
+
+## 10272. Protein Language Models Diverge from Natural Language: Comparative Analysis and Improved Inference
+
+- **ID:** `2602.20449` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Anna Hart, Chi Han, Jeonghwan Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.20449 · **PDF:** https://arxiv.org/pdf/2602.20449v1
+
+## 10273. Imputation of Unknown Missingness in Sparse Electronic Health Records
+
+- **ID:** `2602.20442` · **primary:** `cs.LG` · **Submitted:** 2026-02-24
+- **Authors:** Jun Han, Josue Nassar, Sanjit Singh Batra, et al.
+- **abs:** https://arxiv.org/abs/2602.20442 · **PDF:** https://arxiv.org/pdf/2602.20442v1
+
+## 10274. Can AI be a Teaching Partner? Evaluating ChatGPT, Gemini, and DeepSeek across Three Teaching Strategies
+
+- **ID:** `2603.26673` · **primary:** `cs.CY` · **Submitted:** 2026-02-24
+- **Authors:** Talita de Paula Cypriano de Souza, Shruti Mehta, Matheus Arataque Uema, et al.
+- **abs:** https://arxiv.org/abs/2603.26673 · **PDF:** https://arxiv.org/pdf/2603.26673v1
+
+## 10275. Disentangling Geometry, Performance, and Training in Language Models
+
+- **ID:** `2602.20433` · **primary:** `cs.CL` · **Submitted:** 2026-02-24
+- **Authors:** Atharva Kulkarni, Jacob Mitchell Springer, Arjun Subramonian, et al.
+- **abs:** https://arxiv.org/abs/2602.20433 · **PDF:** https://arxiv.org/pdf/2602.20433v1
+
+## 10276. GauS: Differentiable Scheduling Optimization via Gaussian Reparameterization
+
+- **ID:** `2602.20427` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Yaohui Cai, Vesal Bakhtazad, Cunxi Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.20427 · **PDF:** https://arxiv.org/pdf/2602.20427v1
+
+## 10277. Learning to Rewrite Tool Descriptions for Reliable LLM-Agent Tool Use
+
+- **ID:** `2602.20426` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Ruocheng Guo, Kaiwen Dong, Xiang Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.20426 · **PDF:** https://arxiv.org/pdf/2602.20426v1
+
+## 10278. Implicit Intelligence -- Evaluating Agents on What Users Don't Say
+
+- **ID:** `2602.20424` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Ved Sirdeshmukh, Marc Wetter
+- **abs:** https://arxiv.org/abs/2602.20424 · **PDF:** https://arxiv.org/pdf/2602.20424v1
+
+## 10279. MedCLIPSeg: Probabilistic Vision-Language Adaptation for Data-Efficient and Generalizable Medical Image Segmentation
+
+- **ID:** `2602.20423` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Taha Koleilat, Hojat Asgariandehkordi, Omid Nejati Manzari, et al.
+- **abs:** https://arxiv.org/abs/2602.20423 · **PDF:** https://arxiv.org/pdf/2602.20423v1
+
+## 10280. Diffusion Modulation via Environment Mechanism Modeling for Planning
+
+- **ID:** `2602.20422` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Hanping Zhang, Yuhong Guo
+- **abs:** https://arxiv.org/abs/2602.20422 · **PDF:** https://arxiv.org/pdf/2602.20422v1
+
+## 10281. CREDIT: Certified Ownership Verification of Deep Neural Networks Against Model Extraction Attacks
+
+- **ID:** `2602.20419` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Bolin Shen, Zhan Cheng, Neil Zhenqiang Gong, et al.
+- **abs:** https://arxiv.org/abs/2602.20419 · **PDF:** https://arxiv.org/pdf/2602.20419v1
+
+## 10282. CITED: A Decision Boundary-Aware Signature for GNNs Towards Model Extraction Defense
+
+- **ID:** `2602.20418` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Bolin Shen, Md Shamim Seraj, Zhan Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.20418 · **PDF:** https://arxiv.org/pdf/2602.20418v1
+
+## 10283. CLIPoint3D: Language-Grounded Few-Shot Unsupervised 3D Point Cloud Domain Adaptation
+
+- **ID:** `2602.20409` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Mainak Singha, Sarthak Mehrotra, Paolo Casari, et al.
+- **abs:** https://arxiv.org/abs/2602.20409 · **PDF:** https://arxiv.org/pdf/2602.20409v1
+
+## 10284. Examining and Addressing Barriers to Diversity in LLM-Generated Ideas
+
+- **ID:** `2602.20408` · **primary:** `cs.CY` · **Submitted:** 2026-02-23
+- **Authors:** Yuting Deng, Melanie Brucks, Olivier Toubia
+- **abs:** https://arxiv.org/abs/2602.20408 · **PDF:** https://arxiv.org/pdf/2602.20408v1
+
+## 10285. $κ$-Explorer: A Unified Framework for Active Model Estimation in MDPs
+
+- **ID:** `2602.20404` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Xihe Gu, Urbashi Mitra, Tara Javidi
+- **abs:** https://arxiv.org/abs/2602.20404 · **PDF:** https://arxiv.org/pdf/2602.20404v1
+
+## 10286. Wasserstein Distributionally Robust Online Learning
+
+- **ID:** `2602.20403` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Guixian Chen, Salar Fattahi, Soroosh Shafiee
+- **abs:** https://arxiv.org/abs/2602.20403 · **PDF:** https://arxiv.org/pdf/2602.20403v1
+
+## 10287. Three Concrete Challenges and Two Hopes for the Safety of Unsupervised Elicitation
+
+- **ID:** `2602.20400` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Callum Canavan, Aditya Shrivastava, Allison Qi, et al.
+- **abs:** https://arxiv.org/abs/2602.20400 · **PDF:** https://arxiv.org/pdf/2602.20400v1
+
+## 10288. GeoPT: Scaling Physics Simulation via Lifted Geometric Pre-Training
+
+- **ID:** `2602.20399` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Haixu Wu, Minghao Guo, Zongyi Li, et al.
+- **abs:** https://arxiv.org/abs/2602.20399 · **PDF:** https://arxiv.org/pdf/2602.20399v1
+
+## 10289. cc-Shapley: Measuring Multivariate Feature Importance Needs Causal Context
+
+- **ID:** `2602.20396` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Jörg Martin, Stefan Haufe
+- **abs:** https://arxiv.org/abs/2602.20396 · **PDF:** https://arxiv.org/pdf/2602.20396v3
+
+## 10290. Selecting Optimal Variable Order in Autoregressive Ising Models
+
+- **ID:** `2602.20394` · **primary:** `stat.ML` · **Submitted:** 2026-02-23
+- **Authors:** Shiba Biswal, Marc Vuffray, Andrey Y. Lokhov
+- **abs:** https://arxiv.org/abs/2602.20394 · **PDF:** https://arxiv.org/pdf/2602.20394v2
+
+## 10291. Dynamic Model Routing and Cascading for Efficient LLM Inference: A Survey
+
+- **ID:** `2603.04445` · **primary:** `cs.NI` · **Submitted:** 2026-02-23
+- **Authors:** Yasmin Moslem, John D. Kelleher
+- **abs:** https://arxiv.org/abs/2603.04445 · **PDF:** https://arxiv.org/pdf/2603.04445v1
+
+## 10292. Detecting and Mitigating Group Bias in Heterogeneous Treatment Effects
+
+- **ID:** `2602.20383` · **primary:** `stat.ME` · **Submitted:** 2026-02-23
+- **Authors:** Joel Persson, Jurriën Bakker, Dennis Bohle, et al.
+- **abs:** https://arxiv.org/abs/2602.20383 · **PDF:** https://arxiv.org/pdf/2602.20383v1
+
+## 10293. Case-Aware LLM-as-a-Judge Evaluation for Enterprise-Scale RAG Systems
+
+- **ID:** `2602.20379` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Mukul Chhabra, Luigi Medrano, Arush Verma
+- **abs:** https://arxiv.org/abs/2602.20379 · **PDF:** https://arxiv.org/pdf/2602.20379v1
+
+## 10294. Optimized Disaster Recovery for Distributed Storage Systems: Lightweight Metadata Architectures to Overcome Cryptographic Hashing Bottleneck
+
+- **ID:** `2602.22237` · **primary:** `cs.CR` · **Submitted:** 2026-02-23
+- **Authors:** Prasanna Kumar, Nishank Soni, Gaurang Munje
+- **abs:** https://arxiv.org/abs/2602.22237 · **PDF:** https://arxiv.org/pdf/2602.22237v1
+
+## 10295. Exploiting Low-Rank Structure in Max-K-Cut Problems
+
+- **ID:** `2602.20376` · **primary:** `cs.DS` · **Submitted:** 2026-02-23
+- **Authors:** Ria Stevens, Fangshuo Liao, Barbara Su, et al.
+- **abs:** https://arxiv.org/abs/2602.20376 · **PDF:** https://arxiv.org/pdf/2602.20376v1
+
+## 10296. How communicatively optimal are exact numeral systems? Once more on lexicon size and morphosyntactic complexity
+
+- **ID:** `2602.20372` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Chundra Cathcart, Arne Rubehn, Katja Bocklage, et al.
+- **abs:** https://arxiv.org/abs/2602.20372 · **PDF:** https://arxiv.org/pdf/2602.20372v1
+
+## 10297. Quantitative Approximation Rates for Group Equivariant Learning
+
+- **ID:** `2602.20370` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Jonathan W. Siegel, Snir Hordan, Hannah Lawrence, et al.
+- **abs:** https://arxiv.org/abs/2602.20370 · **PDF:** https://arxiv.org/pdf/2602.20370v1
+
+## 10298. Learning During Detection: Continual Learning for Neural OFDM Receivers via DMRS
+
+- **ID:** `2602.20361` · **primary:** `cs.IT` · **Submitted:** 2026-02-23
+- **Authors:** Mohanad Obeed, Ming Jian
+- **abs:** https://arxiv.org/abs/2602.20361 · **PDF:** https://arxiv.org/pdf/2602.20361v1
+
+## 10299. Momentum Guidance: Plug-and-Play Guidance for Flow Models
+
+- **ID:** `2602.20360` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Runlong Liao, Jian Yu, Baiyu Su, et al.
+- **abs:** https://arxiv.org/abs/2602.20360 · **PDF:** https://arxiv.org/pdf/2602.20360v1
+
+## 10300. Artificial Intelligence: Beyound Ocularcentrism, the New Age of Humans Beyond the Spectacle
+
+- **ID:** `2603.13248` · **primary:** `cs.CY` · **Submitted:** 2026-02-23
+- **Authors:** Mustapha El Moussaoui
+- **abs:** https://arxiv.org/abs/2603.13248 · **PDF:** https://arxiv.org/pdf/2603.13248v1
+
+## 10301. Hierarchical Molecular Representation Learning via Fragment-Based Self-Supervised Embedding Prediction
+
+- **ID:** `2602.20344` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Jiele Wu, Haozhe Ma, Zhihan Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.20344 · **PDF:** https://arxiv.org/pdf/2602.20344v1
+
+## 10302. Emergent Manifold Separability during Reasoning in Large Language Models
+
+- **ID:** `2602.20338` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Alexandre Polo, Chanwoo Chun, SueYeon Chung
+- **abs:** https://arxiv.org/abs/2602.20338 · **PDF:** https://arxiv.org/pdf/2602.20338v1
+
+## 10303. Natural Language Processing Models for Robust Document Categorization
+
+- **ID:** `2602.20336` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Radoslaw Roszczyk, Pawel Tecza, Maciej Stodolski, et al.
+- **abs:** https://arxiv.org/abs/2602.20336 · **PDF:** https://arxiv.org/pdf/2602.20336v1
+
+## 10304. DMCD: Semantic-Statistical Framework for Causal Discovery
+
+- **ID:** `2602.20333` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Samarth KaPatel, Sofia Nikiforova, Giacinto Paolo Saggese, et al.
+- **abs:** https://arxiv.org/abs/2602.20333 · **PDF:** https://arxiv.org/pdf/2602.20333v1
+
+## 10305. No One Size Fits All: QueryBandits for Hallucination Mitigation
+
+- **ID:** `2602.20332` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Nicole Cho, William Watson, Alec Koppel, et al.
+- **abs:** https://arxiv.org/abs/2602.20332 · **PDF:** https://arxiv.org/pdf/2602.20332v1
+
+## 10306. Circuit Tracing in Vision-Language Models: Understanding the Internal Mechanisms of Multimodal Thinking
+
+- **ID:** `2602.20330` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Jingcheng Yang, Tianhu Xiong, Shengyi Qian, et al.
+- **abs:** https://arxiv.org/abs/2602.20330 · **PDF:** https://arxiv.org/pdf/2602.20330v1
+
+## 10307. CaDrift: A Time-dependent Causal Generator of Drifting Data Streams
+
+- **ID:** `2602.20329` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Eduardo V. L. Barboza, Jean Paul Barddal, Robert Sabourin, et al.
+- **abs:** https://arxiv.org/abs/2602.20329 · **PDF:** https://arxiv.org/pdf/2602.20329v1
+
+## 10308. An artificial intelligence framework for end-to-end rare disease phenotyping from clinical notes using large language models
+
+- **ID:** `2602.20324` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Cathy Shyr, Yan Hu, Rory J. Tinker, et al.
+- **abs:** https://arxiv.org/abs/2602.20324 · **PDF:** https://arxiv.org/pdf/2602.20324v1
+
+## 10309. PhysMem: Scaling Test-time Physical Memory for Robot Manipulation
+
+- **ID:** `2602.20323` · **primary:** `cs.RO` · **Submitted:** 2026-02-23
+- **Authors:** Haoyang Li, Yang You, Hao Su, et al.
+- **abs:** https://arxiv.org/abs/2602.20323 · **PDF:** https://arxiv.org/pdf/2602.20323v4
+
+## 10310. TokEye: Fast Signal Extraction for Fluctuating Time Series via Offline Self-Supervised Learning From Fusion Diagnostics to Bioacoustics
+
+- **ID:** `2602.20317` · **primary:** `eess.SP` · **Submitted:** 2026-02-23
+- **Authors:** Nathaniel Chen, Kouroche Bouchiat, Peter Steiner, et al.
+- **abs:** https://arxiv.org/abs/2602.20317 · **PDF:** https://arxiv.org/pdf/2602.20317v2
+
+## 10311. CrossLLM-Mamba: Multimodal State Space Fusion of LLMs for RNA Interaction Prediction
+
+- **ID:** `2602.22236` · **primary:** `q-bio.GN` · **Submitted:** 2026-02-23
+- **Authors:** Rabeya Tus Sadia, Qiang Ye, Qiang Cheng
+- **abs:** https://arxiv.org/abs/2602.22236 · **PDF:** https://arxiv.org/pdf/2602.22236v1
+
+## 10312. QuantVLA: Scale-Calibrated Post-Training Quantization for Vision-Language-Action Models
+
+- **ID:** `2602.20309` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Jingxuan Zhang, Yunta Hsieh, Zhongwei Wan, et al.
+- **abs:** https://arxiv.org/abs/2602.20309 · **PDF:** https://arxiv.org/pdf/2602.20309v3
+
+## 10313. In-context Pre-trained Time-Series Foundation Models adapt to Unseen Tasks
+
+- **ID:** `2602.20307` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Shangqing Xu, Harshavardhan Kamarthi, Haoxin Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.20307 · **PDF:** https://arxiv.org/pdf/2602.20307v1
+
+## 10314. Shape-informed cardiac mechanics surrogates in data-scarce regimes via geometric encoding and generative augmentation
+
+- **ID:** `2602.20306` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Davide Carrara, Marc Hirschvogel, Francesca Bonizzoni, et al.
+- **abs:** https://arxiv.org/abs/2602.20306 · **PDF:** https://arxiv.org/pdf/2602.20306v1
+
+## 10315. Multilevel Determinants of Overweight and Obesity Among U.S. Children Aged 10-17: Comparative Evaluation of Statistical and Machine Learning Approaches Using the 2021 National Survey of Children's Health
+
+- **ID:** `2602.20303` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Joyanta Jyoti Mondal
+- **abs:** https://arxiv.org/abs/2602.20303 · **PDF:** https://arxiv.org/pdf/2602.20303v1
+
+## 10316. What Makes a Good Query? Measuring the Impact of Human-Confusing Linguistic Features on LLM Performance
+
+- **ID:** `2602.20300` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** William Watson, Nicole Cho, Sumitra Ganesh, et al.
+- **abs:** https://arxiv.org/abs/2602.20300 · **PDF:** https://arxiv.org/pdf/2602.20300v1
+
+## 10317. Gap-Dependent Bounds for Nearly Minimax Optimal Reinforcement Learning with Linear Function Approximation
+
+- **ID:** `2602.20297` · **primary:** `stat.ML` · **Submitted:** 2026-02-23
+- **Authors:** Haochen Zhang, Zhong Zheng, Lingzhou Xue
+- **abs:** https://arxiv.org/abs/2602.20297 · **PDF:** https://arxiv.org/pdf/2602.20297v1
+
+## 10318. Learning to Solve Complex Problems via Dataset Decomposition
+
+- **ID:** `2602.20296` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Wanru Zhao, Lucas Caccia, Zhengyan Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.20296 · **PDF:** https://arxiv.org/pdf/2602.20296v1
+
+## 10319. InterviewSim: A Scalable Framework for Interview-Grounded Personality Simulation
+
+- **ID:** `2602.20294` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Yu Li, Pranav Narayanan Venkit, Yada Pruksachatkun, et al.
+- **abs:** https://arxiv.org/abs/2602.20294 · **PDF:** https://arxiv.org/pdf/2602.20294v1
+
+## 10320. Discrete Diffusion with Sample-Efficient Estimators for Conditionals
+
+- **ID:** `2602.20293` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Karthik Elamvazhuthi, Abhijith Jayakumar, Andrey Y. Lokhov
+- **abs:** https://arxiv.org/abs/2602.20293 · **PDF:** https://arxiv.org/pdf/2602.20293v2
+
+## 10321. T-REX: Transformer-Based Category Sequence Generation for Grocery Basket Recommendation
+
+- **ID:** `2603.06631` · **primary:** `cs.IR` · **Submitted:** 2026-02-23
+- **Authors:** Soroush Mokhtari, Muhammad Tayyab Asif, Sergiy Zubatiy
+- **abs:** https://arxiv.org/abs/2603.06631 · **PDF:** https://arxiv.org/pdf/2603.06631v1
+
+## 10322. Quantifying the Expectation-Realisation Gap for Agentic AI Systems
+
+- **ID:** `2602.20292` · **primary:** `cs.SE` · **Submitted:** 2026-02-23
+- **Authors:** Sebastian Lobentanzer
+- **abs:** https://arxiv.org/abs/2602.20292 · **PDF:** https://arxiv.org/pdf/2602.20292v2
+
+## 10323. The Sim-to-Real Gap in MRS Quantification: A Systematic Deep Learning Validation for GABA
+
+- **ID:** `2602.20289` · **primary:** `eess.SP` · **Submitted:** 2026-02-23
+- **Authors:** Zien Ma, S. M. Shermer, Oktay Karakuş, et al.
+- **abs:** https://arxiv.org/abs/2602.20289 · **PDF:** https://arxiv.org/pdf/2602.20289v1
+
+## 10324. The Truthfulness Spectrum Hypothesis
+
+- **ID:** `2602.20273` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Zhuofan Josh Ying, Shauli Ravfogel, Nikolaus Kriegeskorte, et al.
+- **abs:** https://arxiv.org/abs/2602.20273 · **PDF:** https://arxiv.org/pdf/2602.20273v1
+
+## 10325. Uncertainty-Aware Delivery Delay Duration Prediction via Multi-Task Deep Learning
+
+- **ID:** `2602.20271` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Stefan Faulkner, Reza Zandehshahvar, Vahid Eghbal Akhlaghi, et al.
+- **abs:** https://arxiv.org/abs/2602.20271 · **PDF:** https://arxiv.org/pdf/2602.20271v1
+
+## 10326. A Very Big Video Reasoning Suite
+
+- **ID:** `2602.20159` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Maijunxian Wang, Ruisi Wang, Juyi Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.20159 · **PDF:** https://arxiv.org/pdf/2602.20159v2
+
+## 10327. Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks
+
+- **ID:** `2602.20156` · **primary:** `cs.CR` · **Submitted:** 2026-02-23
+- **Authors:** David Schmotz, Luca Beurer-Kellner, Sahar Abdelnabi, et al.
+- **abs:** https://arxiv.org/abs/2602.20156 · **PDF:** https://arxiv.org/pdf/2602.20156v3
+
+## 10328. JUCAL: Jointly Calibrating Aleatoric and Epistemic Uncertainty in Classification Tasks
+
+- **ID:** `2602.20153` · **primary:** `stat.ML` · **Submitted:** 2026-02-23
+- **Authors:** Jakob Heiss, Sören Lambrecht, Jakob Weissteiner, et al.
+- **abs:** https://arxiv.org/abs/2602.20153 · **PDF:** https://arxiv.org/pdf/2602.20153v1
+
+## 10329. Behavior Learning (BL): Learning Hierarchical Optimization Structures from Data
+
+- **ID:** `2602.20152` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Zhenyao Ma, Yue Liang, Dongxu Li
+- **abs:** https://arxiv.org/abs/2602.20152 · **PDF:** https://arxiv.org/pdf/2602.20152v1
+
+## 10330. Conformal Risk Control for Non-Monotonic Losses
+
+- **ID:** `2602.20151` · **primary:** `stat.ME` · **Submitted:** 2026-02-23
+- **Authors:** Anastasios N. Angelopoulos
+- **abs:** https://arxiv.org/abs/2602.20151 · **PDF:** https://arxiv.org/pdf/2602.20151v1
+
+## 10331. Coupled Cluster con MōLe: Molecular Orbital Learning for Neural Wavefunctions
+
+- **ID:** `2602.20232` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Luca Thiede, Abdulrahman Aldossary, Andreas Burger, et al.
+- **abs:** https://arxiv.org/abs/2602.20232 · **PDF:** https://arxiv.org/pdf/2602.20232v1
+
+## 10332. Agentic AI for Scalable and Robust Optical Systems Control
+
+- **ID:** `2602.20144` · **primary:** `eess.SY` · **Submitted:** 2026-02-23
+- **Authors:** Zehao Wang, Mingzhe Han, Wei Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.20144 · **PDF:** https://arxiv.org/pdf/2602.20144v1
+
+## 10333. Recurrent Structural Policy Gradient for Partially Observable Mean Field Games
+
+- **ID:** `2602.20141` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Clarisse Wibault, Johannes Forkel, Sebastian Towers, et al.
+- **abs:** https://arxiv.org/abs/2602.20141 · **PDF:** https://arxiv.org/pdf/2602.20141v1
+
+## 10334. KNIGHT: Knowledge Graph-Driven Multiple-Choice Question Generation with Adaptive Hardness Calibration
+
+- **ID:** `2602.20135` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Mohammad Amanlou, Erfan Shafiee Moghaddam, Yasaman Amou Jafari, et al.
+- **abs:** https://arxiv.org/abs/2602.20135 · **PDF:** https://arxiv.org/pdf/2602.20135v1
+
+## 10335. Modeling Epidemiological Dynamics Under Adversarial Data and User Deception
+
+- **ID:** `2602.20134` · **primary:** `cs.GT` · **Submitted:** 2026-02-23
+- **Authors:** Yiqi Su, Christo Kurisummoottil Thomas, Walid Saad, et al.
+- **abs:** https://arxiv.org/abs/2602.20134 · **PDF:** https://arxiv.org/pdf/2602.20134v1
+
+## 10336. AdaEvolve: Adaptive LLM Driven Zeroth-Order Optimization
+
+- **ID:** `2602.20133` · **primary:** `cs.NE` · **Submitted:** 2026-02-23
+- **Authors:** Mert Cemri, Shubham Agrawal, Akshat Gupta, et al.
+- **abs:** https://arxiv.org/abs/2602.20133 · **PDF:** https://arxiv.org/pdf/2602.20133v1
+
+## 10337. LAD: Learning Advantage Distribution for Reasoning
+
+- **ID:** `2602.20132` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Wendi Li, Sharon Li
+- **abs:** https://arxiv.org/abs/2602.20132 · **PDF:** https://arxiv.org/pdf/2602.20132v1
+
+## 10338. To Reason or Not to: Selective Chain-of-Thought in Medical Question Answering
+
+- **ID:** `2602.20130` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Zaifu Zhan, Min Zeng, Shuang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.20130 · **PDF:** https://arxiv.org/pdf/2602.20130v1
+
+## 10339. Adaptation to Intrinsic Dependence in Diffusion Language Models
+
+- **ID:** `2602.20126` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Yunxiao Zhao, Changxiao Cai
+- **abs:** https://arxiv.org/abs/2602.20126 · **PDF:** https://arxiv.org/pdf/2602.20126v1
+
+## 10340. NanoKnow: How to Know What Your Language Model Knows
+
+- **ID:** `2602.20122` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Lingwei Gu, Nour Jedidi, Jimmy Lin
+- **abs:** https://arxiv.org/abs/2602.20122 · **PDF:** https://arxiv.org/pdf/2602.20122v1
+
+## 10341. NovaPlan: Zero-Shot Long-Horizon Manipulation via Closed-Loop Video Language Planning
+
+- **ID:** `2602.20119` · **primary:** `cs.RO` · **Submitted:** 2026-02-23
+- **Authors:** Jiahui Fu, Junyu Nan, Lingfeng Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.20119 · **PDF:** https://arxiv.org/pdf/2602.20119v1
+
+## 10342. ReSyn: Autonomously Scaling Synthetic Environments for Reasoning Models
+
+- **ID:** `2602.20117` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Andre He, Nathaniel Weir, Kaj Bostrom, et al.
+- **abs:** https://arxiv.org/abs/2602.20117 · **PDF:** https://arxiv.org/pdf/2602.20117v1
+
+## 10343. Benchmarking Unlearning for Vision Transformers
+
+- **ID:** `2602.20114` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Kairan Zhao, Iurie Luca, Peter Triantafillou
+- **abs:** https://arxiv.org/abs/2602.20114 · **PDF:** https://arxiv.org/pdf/2602.20114v1
+
+## 10344. StyleStream: Real-Time Zero-Shot Voice Style Conversion
+
+- **ID:** `2602.20113` · **primary:** `cs.SD` · **Submitted:** 2026-02-23
+- **Authors:** Yisi Liu, Nicholas Lee, Gopala Anumanchipalli
+- **abs:** https://arxiv.org/abs/2602.20113 · **PDF:** https://arxiv.org/pdf/2602.20113v1
+
+## 10345. Reliable Abstention under Adversarial Injections: Tight Lower Bounds and New Upper Bounds
+
+- **ID:** `2602.20111` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Ezra Edelman, Surbhi Goel
+- **abs:** https://arxiv.org/abs/2602.20111 · **PDF:** https://arxiv.org/pdf/2602.20111v1
+
+## 10346. Align When They Want, Complement When They Need! Human-Centered Ensembles for Adaptive Human-AI Collaboration
+
+- **ID:** `2602.20104` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Hasan Amin, Ming Yin, Rajiv Khanna
+- **abs:** https://arxiv.org/abs/2602.20104 · **PDF:** https://arxiv.org/pdf/2602.20104v1
+
+## 10347. Controllable Evidence Selection in Retrieval-Augmented Question Answering via Deterministic Utility Gating
+
+- **ID:** `2603.18011` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Victor P. Unda
+- **abs:** https://arxiv.org/abs/2603.18011 · **PDF:** https://arxiv.org/pdf/2603.18011v1
+
+## 10348. BarrierSteer: LLM Safety via Learning Barrier Steering
+
+- **ID:** `2602.20102` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Thanh Q. Tran, Arun Verma, Kiwan Wong, et al.
+- **abs:** https://arxiv.org/abs/2602.20102 · **PDF:** https://arxiv.org/pdf/2602.20102v1
+
+## 10349. Transcending the Annotation Bottleneck: AI-Powered Discovery in Biology and Medicine
+
+- **ID:** `2602.20100` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Soumick Chatterjee
+- **abs:** https://arxiv.org/abs/2602.20100 · **PDF:** https://arxiv.org/pdf/2602.20100v1
+
+## 10350. CausalFlip: A Benchmark for LLM Causal Judgment Beyond Semantic Matching
+
+- **ID:** `2602.20094` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Yuzhe Wang, Yaochen Zhu, Jundong Li
+- **abs:** https://arxiv.org/abs/2602.20094 · **PDF:** https://arxiv.org/pdf/2602.20094v1
+
+## 10351. BabyLM Turns 4 and Goes Multilingual: Call for Papers for the 2026 BabyLM Workshop
+
+- **ID:** `2602.20092` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Leshem Choshen, Ryan Cotterell, Mustafa Omer Gul, et al.
+- **abs:** https://arxiv.org/abs/2602.20092 · **PDF:** https://arxiv.org/pdf/2602.20092v2
+
+## 10352. How Retrieved Context Shapes Internal Representations in RAG
+
+- **ID:** `2602.20091` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Samuel Yeh, Sharon Li
+- **abs:** https://arxiv.org/abs/2602.20091 · **PDF:** https://arxiv.org/pdf/2602.20091v1
+
+## 10353. StructXLIP: Enhancing Vision-language Models with Multimodal Structural Cues
+
+- **ID:** `2602.20089` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Zanxi Ruan, Songqun Gao, Qiuyu Kong, et al.
+- **abs:** https://arxiv.org/abs/2602.20089 · **PDF:** https://arxiv.org/pdf/2602.20089v3
+
+## 10354. Descent-Guided Policy Gradient for Scalable Cooperative Multi-Agent Learning
+
+- **ID:** `2602.20078` · **primary:** `cs.MA` · **Submitted:** 2026-02-23
+- **Authors:** Shan Yang, Yang Liu
+- **abs:** https://arxiv.org/abs/2602.20078 · **PDF:** https://arxiv.org/pdf/2602.20078v2
+
+## 10355. Robust Taylor-Lagrange Control for Safety-Critical Systems
+
+- **ID:** `2602.20076` · **primary:** `eess.SY` · **Submitted:** 2026-02-23
+- **Authors:** Wei Xiao, Christos Cassandras, Anni Li
+- **abs:** https://arxiv.org/abs/2602.20076 · **PDF:** https://arxiv.org/pdf/2602.20076v1
+
+## 10356. Agentic Framework for Political Biography Extraction
+
+- **ID:** `2603.18010` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Yifei Zhu, Songpo Yang, Jiangnan Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.18010 · **PDF:** https://arxiv.org/pdf/2603.18010v1
+
+## 10357. Training-Free Generative Modeling via Kernelized Stochastic Interpolants
+
+- **ID:** `2602.20070` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Florentin Coeurdoux, Etienne Lempereur, Nathanaël Cuvelle-Magar, et al.
+- **abs:** https://arxiv.org/abs/2602.20070 · **PDF:** https://arxiv.org/pdf/2602.20070v2
+
+## 10358. The Invisible Gorilla Effect in Out-of-distribution Detection
+
+- **ID:** `2602.20068` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Harry Anthony, Ziyun Liang, Hermione Warr, et al.
+- **abs:** https://arxiv.org/abs/2602.20068 · **PDF:** https://arxiv.org/pdf/2602.20068v1
+
+## 10359. HeatPrompt: Zero-Shot Vision-Language Modeling of Urban Heat Demand from Satellite Images
+
+- **ID:** `2602.20066` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Kundan Thota, Xuanhao Mu, Thorsten Schlachter, et al.
+- **abs:** https://arxiv.org/abs/2602.20066 · **PDF:** https://arxiv.org/pdf/2602.20066v1
+
+## 10360. Multilingual Large Language Models do not comprehend all natural languages to equal degrees
+
+- **ID:** `2602.20065` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Natalia Moskvina, Raquel Montero, Masaya Yoshida, et al.
+- **abs:** https://arxiv.org/abs/2602.20065 · **PDF:** https://arxiv.org/pdf/2602.20065v1
+
+## 10361. The LLMbda Calculus: AI Agents, Conversations, and Information Flow
+
+- **ID:** `2602.20064` · **primary:** `cs.PL` · **Submitted:** 2026-02-23
+- **Authors:** Zac Garby, Andrew D. Gordon, David Sands
+- **abs:** https://arxiv.org/abs/2602.20064 · **PDF:** https://arxiv.org/pdf/2602.20064v1
+
+## 10362. You Don't Need All That Attention: Surgical Memorization Mitigation in Text-to-Image Diffusion Models
+
+- **ID:** `2603.00133` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Kairan Zhao, Eleni Triantafillou, Peter Triantafillou
+- **abs:** https://arxiv.org/abs/2603.00133 · **PDF:** https://arxiv.org/pdf/2603.00133v1
+
+## 10363. A Theory of How Pretraining Shapes Inductive Bias in Fine-Tuning
+
+- **ID:** `2602.20062` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Nicolas Anguita, Francesco Locatello, Andrew M. Saxe, et al.
+- **abs:** https://arxiv.org/abs/2602.20062 · **PDF:** https://arxiv.org/pdf/2602.20062v1
+
+## 10364. Interaction Theater: A case of LLM Agents Interacting at Scale
+
+- **ID:** `2602.20059` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Sarath Shekkizhar, Adam Earle
+- **abs:** https://arxiv.org/abs/2602.20059 · **PDF:** https://arxiv.org/pdf/2602.20059v1
+
+## 10365. AdaWorldPolicy: World-Model-Driven Diffusion Policy with Online Adaptive Learning for Robotic Manipulation
+
+- **ID:** `2602.20057` · **primary:** `cs.RO` · **Submitted:** 2026-02-23
+- **Authors:** Ge Yuan, Qiyuan Qiao, Jing Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.20057 · **PDF:** https://arxiv.org/pdf/2602.20057v1
+
+## 10366. To Move or Not to Move: Constraint-based Planning Enables Zero-Shot Generalization for Interactive Navigation
+
+- **ID:** `2602.20055` · **primary:** `cs.RO` · **Submitted:** 2026-02-23
+- **Authors:** Apoorva Vashisth, Manav Kulshrestha, Pranav Bakshi, et al.
+- **abs:** https://arxiv.org/abs/2602.20055 · **PDF:** https://arxiv.org/pdf/2602.20055v1
+
+## 10367. Structured Prompt Language: Declarative Context Management for LLMs
+
+- **ID:** `2602.21257` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Wen G. Gong
+- **abs:** https://arxiv.org/abs/2602.21257 · **PDF:** https://arxiv.org/pdf/2602.21257v1
+
+## 10368. Entropy in Large Language Models
+
+- **ID:** `2602.20052` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Marco Scharringhausen
+- **abs:** https://arxiv.org/abs/2602.20052 · **PDF:** https://arxiv.org/pdf/2602.20052v1
+
+## 10369. SEAL-pose: Enhancing 3D Human Pose Estimation via a Learned Loss for Structural Consistency
+
+- **ID:** `2602.20051` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Yeonsung Kim, Junggeun Do, Seunguk Do, et al.
+- **abs:** https://arxiv.org/abs/2602.20051 · **PDF:** https://arxiv.org/pdf/2602.20051v1
+
+## 10370. CodeCompass: Navigating the Navigation Paradox in Agentic Code Intelligence
+
+- **ID:** `2602.20048` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Tarakanath Paipuru
+- **abs:** https://arxiv.org/abs/2602.20048 · **PDF:** https://arxiv.org/pdf/2602.20048v1
+
+## 10371. Closing the gap in multimodal medical representation alignment
+
+- **ID:** `2602.20046` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Eleonora Grassucci, Giordano Cicchetti, Danilo Comminiello
+- **abs:** https://arxiv.org/abs/2602.20046 · **PDF:** https://arxiv.org/pdf/2602.20046v1
+
+## 10372. Let There Be Claws: An Early Social Network Analysis of AI Agents on Moltbook
+
+- **ID:** `2602.20044` · **primary:** `physics.soc-ph` · **Submitted:** 2026-02-23
+- **Authors:** H. C. W. Price, H. AlMuhanna, P. M. Bassani, et al.
+- **abs:** https://arxiv.org/abs/2602.20044 · **PDF:** https://arxiv.org/pdf/2602.20044v1
+
+## 10373. Position: General Alignment Has Hit a Ceiling; Edge Alignment Must Be Taken Seriously
+
+- **ID:** `2602.20042` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Han Bao, Yue Huang, Xiaoda Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.20042 · **PDF:** https://arxiv.org/pdf/2602.20042v1
+
+## 10374. AgenticSum: An Agentic Inference-Time Framework for Faithful Clinical Text Summarization
+
+- **ID:** `2602.20040` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Fahmida Liza Piya, Rahmatollah Beheshti
+- **abs:** https://arxiv.org/abs/2602.20040 · **PDF:** https://arxiv.org/pdf/2602.20040v1
+
+## 10375. Latent Introspection: Models Can Detect Prior Concept Injections
+
+- **ID:** `2602.20031` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Theia Pearson-Vogel, Martin Vanek, Raymond Douglas, et al.
+- **abs:** https://arxiv.org/abs/2602.20031 · **PDF:** https://arxiv.org/pdf/2602.20031v2
+
+## 10376. Agents of Chaos
+
+- **ID:** `2602.20021` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Natalie Shapira, Chris Wendler, Avery Yen, et al.
+- **abs:** https://arxiv.org/abs/2602.20021 · **PDF:** https://arxiv.org/pdf/2602.20021v1
+
+## 10377. gencat: Generative computerized adaptive testing
+
+- **ID:** `2602.20020` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Wanyong Feng, Andrew Lan
+- **abs:** https://arxiv.org/abs/2602.20020 · **PDF:** https://arxiv.org/pdf/2602.20020v1
+
+## 10378. Learning Discriminative and Generalizable Anomaly Detector for Dynamic Graph with Limited Supervision
+
+- **ID:** `2602.20019` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Yuxing Tian, Yiyan Qi, Fengran Mo, et al.
+- **abs:** https://arxiv.org/abs/2602.20019 · **PDF:** https://arxiv.org/pdf/2602.20019v1
+
+## 10379. QUIETT: Query-Independent Table Transformation for Robust Reasoning
+
+- **ID:** `2602.20017` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Gaurav Najpande, Tampu Ravi Kumar, Manan Roy Choudhury, et al.
+- **abs:** https://arxiv.org/abs/2602.20017 · **PDF:** https://arxiv.org/pdf/2602.20017v1
+
+## 10380. Exploring Anti-Aging Literature via ConvexTopics and Large Language Models
+
+- **ID:** `2602.20224` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Lana E. Yeganova, Won G. Kim, Shubo Tian, et al.
+- **abs:** https://arxiv.org/abs/2602.20224 · **PDF:** https://arxiv.org/pdf/2602.20224v1
+
+## 10381. A Secure and Private Distributed Bayesian Federated Learning Design
+
+- **ID:** `2602.20003` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Nuocheng Yang, Sihua Wang, Zhaohui Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.20003 · **PDF:** https://arxiv.org/pdf/2602.20003v1
+
+## 10382. FairFS: Addressing Deep Feature Selection Biases for Recommender System
+
+- **ID:** `2602.20001` · **primary:** `cs.IR` · **Submitted:** 2026-02-23
+- **Authors:** Xianquan Wang, Zhaocheng Du, Jieming Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.20001 · **PDF:** https://arxiv.org/pdf/2602.20001v1
+
+## 10383. Photons = Tokens: The Physics of AI and the Economics of Knowledge
+
+- **ID:** `2603.06630` · **primary:** `physics.soc-ph` · **Submitted:** 2026-02-23
+- **Authors:** Alec Litowitz, Nick Polson, Vadim Sokolov
+- **abs:** https://arxiv.org/abs/2603.06630 · **PDF:** https://arxiv.org/pdf/2603.06630v1
+
+## 10384. Cross-lingual Matryoshka Representation Learning across Speech and Text
+
+- **ID:** `2602.19991` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Yaya Sy, Dioula Doucouré, Christophe Cerisara, et al.
+- **abs:** https://arxiv.org/abs/2602.19991 · **PDF:** https://arxiv.org/pdf/2602.19991v1
+
+## 10385. Counterfactual Understanding via Retrieval-aware Multimodal Modeling for Time-to-Event Survival Prediction
+
+- **ID:** `2602.19987` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Ha-Anh Hoang Nguyen, Tri-Duc Phan Le, Duc-Hoang Pham, et al.
+- **abs:** https://arxiv.org/abs/2602.19987 · **PDF:** https://arxiv.org/pdf/2602.19987v1
+
+## 10386. Multivariate time-series forecasting of ASTRI-Horn monitoring data: A Normal Behavior Model
+
+- **ID:** `2602.19984` · **primary:** `astro-ph.IM` · **Submitted:** 2026-02-23
+- **Authors:** Federico Incardona, Alessandro Costa, Farida Farsian, et al.
+- **abs:** https://arxiv.org/abs/2602.19984 · **PDF:** https://arxiv.org/pdf/2602.19984v1
+
+## 10387. Contextual Safety Reasoning and Grounding for Open-World Robots
+
+- **ID:** `2602.19983` · **primary:** `cs.RO` · **Submitted:** 2026-02-23
+- **Authors:** Zachary Ravichandran, David Snyder, Alexander Robey, et al.
+- **abs:** https://arxiv.org/abs/2602.19983 · **PDF:** https://arxiv.org/pdf/2602.19983v2
+
+## 10388. A Computationally Efficient Multidimensional Vision Transformer
+
+- **ID:** `2602.19982` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Alaa El Ichi, Khalide Jbilou
+- **abs:** https://arxiv.org/abs/2602.19982 · **PDF:** https://arxiv.org/pdf/2602.19982v1
+
+## 10389. Discrete Diffusion Models Exploit Asymmetry to Solve Lookahead Planning Tasks
+
+- **ID:** `2602.19980` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Itamar Trainin, Shauli Ravfogel, Omri Abend, et al.
+- **abs:** https://arxiv.org/abs/2602.19980 · **PDF:** https://arxiv.org/pdf/2602.19980v1
+
+## 10390. ReAttn: Improving Attention-based Re-ranking via Attention Re-weighting
+
+- **ID:** `2602.19969` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Yuxing Tian, Fengran Mo, Weixu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.19969 · **PDF:** https://arxiv.org/pdf/2602.19969v1
+
+## 10391. Unlearning Noise in PINNs: A Selective Pruning Framework for PDE Inverse Problems
+
+- **ID:** `2602.19967` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Yongsheng Chen, Yong Chen, Wei Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.19967 · **PDF:** https://arxiv.org/pdf/2602.19967v2
+
+## 10392. On the Equivalence of Random Network Distillation, Deep Ensembles, and Bayesian Inference
+
+- **ID:** `2602.19964` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Moritz A. Zanger, Yijun Wu, Pascal R. Van der Vaart, et al.
+- **abs:** https://arxiv.org/abs/2602.19964 · **PDF:** https://arxiv.org/pdf/2602.19964v2
+
+## 10393. Unlocking Multimodal Document Intelligence: From Current Triumphs to Future Frontiers of Visual Document Retrieval
+
+- **ID:** `2602.19961` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Yibo Yan, Jiahao Huo, Guanbo Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.19961 · **PDF:** https://arxiv.org/pdf/2602.19961v2
+
+## 10394. Sparse Masked Attention Policies for Reliable Generalization
+
+- **ID:** `2602.19956` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Caroline Horsch, Laurens Engwegen, Max Weltevrede, et al.
+- **abs:** https://arxiv.org/abs/2602.19956 · **PDF:** https://arxiv.org/pdf/2602.19956v1
+
+## 10395. A Bayesian Framework for Post-disruption Travel Time Prediction in Metro Networks
+
+- **ID:** `2602.19952` · **primary:** `stat.AP` · **Submitted:** 2026-02-23
+- **Authors:** Shayan Nazemi, Aurélie Labbe, Stefan Steiner, et al.
+- **abs:** https://arxiv.org/abs/2602.19952 · **PDF:** https://arxiv.org/pdf/2602.19952v1
+
+## 10396. Assessing Risks of Large Language Models in Mental Health Support: A Framework for Automated Clinical AI Red Teaming
+
+- **ID:** `2602.19948` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Ian Steenstra, Paola Pedrelli, Weiyan Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.19948 · **PDF:** https://arxiv.org/pdf/2602.19948v2
+
+## 10397. When Pretty Isn't Useful: Investigating Why Modern Text-to-Image Models Fail as Reliable Training Data Generators
+
+- **ID:** `2602.19946` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Krzysztof Adamkiewicz, Brian Moser, Stanislav Frolov, et al.
+- **abs:** https://arxiv.org/abs/2602.19946 · **PDF:** https://arxiv.org/pdf/2602.19946v3
+
+## 10398. DP-FedAdamW: An Efficient Optimizer for Differentially Private Federated Large Models
+
+- **ID:** `2602.19945` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Jin Liu, Yinbin Miao, Ning Xi, et al.
+- **abs:** https://arxiv.org/abs/2602.19945 · **PDF:** https://arxiv.org/pdf/2602.19945v1
+
+## 10399. A Replicate-and-Quantize Strategy for Plug-and-Play Load Balancing of Sparse Mixture-of-Experts LLMs
+
+- **ID:** `2602.19938` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Zijie Liu, Jie Peng, Jinhao Duan, et al.
+- **abs:** https://arxiv.org/abs/2602.19938 · **PDF:** https://arxiv.org/pdf/2602.19938v1
+
+## 10400. Expanding the Role of Diffusion Models for Robust Classifier Training
+
+- **ID:** `2602.19931` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Pin-Han Huang, Shang-Tse Chen, Hsuan-Tien Lin
+- **abs:** https://arxiv.org/abs/2602.19931 · **PDF:** https://arxiv.org/pdf/2602.19931v1
+
+## 10401. Beyond Mimicry: Toward Lifelong Adaptability in Imitation Learning
+
+- **ID:** `2602.19930` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Nathan Gavenski, Felipe Meneguzzi, Odinaldo Rodrigues
+- **abs:** https://arxiv.org/abs/2602.19930 · **PDF:** https://arxiv.org/pdf/2602.19930v1
+
+## 10402. Rethinking LoRA for Privacy-Preserving Federated Learning in Large Models
+
+- **ID:** `2602.19926` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Jin Liu, Yinbin Miao, Ning Xi, et al.
+- **abs:** https://arxiv.org/abs/2602.19926 · **PDF:** https://arxiv.org/pdf/2602.19926v1
+
+## 10403. vLLM Semantic Router: Signal Driven Decision Routing for Mixture-of-Modality Models
+
+- **ID:** `2603.04444` · **primary:** `cs.NI` · **Submitted:** 2026-02-23
+- **Authors:** Xunzhuo Liu, Huamin Chen, Samzong Lu, et al.
+- **abs:** https://arxiv.org/abs/2603.04444 · **PDF:** https://arxiv.org/pdf/2603.04444v2
+
+## 10404. Janus-Q: End-to-End Event-Driven Trading via Hierarchical-Gated Reward Modeling
+
+- **ID:** `2602.19919` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Xiang Li, Zikai Wei, Yiyan Qi, et al.
+- **abs:** https://arxiv.org/abs/2602.19919 · **PDF:** https://arxiv.org/pdf/2602.19919v2
+
+## 10405. RobPI: Robust Private Inference against Malicious Client
+
+- **ID:** `2602.19918` · **primary:** `cs.CR` · **Submitted:** 2026-02-23
+- **Authors:** Jiaqi Xue, Mengxin Zheng, Qian Lou
+- **abs:** https://arxiv.org/abs/2602.19918 · **PDF:** https://arxiv.org/pdf/2602.19918v1
+
+## 10406. Uncertainty-Aware Rank-One MIMO Q Network Framework for Accelerated Offline Reinforcement Learning
+
+- **ID:** `2602.19917` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Thanh Nguyen, Tung Luu, Tri Ton, et al.
+- **abs:** https://arxiv.org/abs/2602.19917 · **PDF:** https://arxiv.org/pdf/2602.19917v1
+
+## 10407. Fully Convolutional Spatiotemporal Learning for Microstructure Evolution Prediction
+
+- **ID:** `2602.19915` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Michael Trimboli, Mohammed Alsubaie, Sirani M. Perera, et al.
+- **abs:** https://arxiv.org/abs/2602.19915 · **PDF:** https://arxiv.org/pdf/2602.19915v1
+
+## 10408. Watson & Holmes: A Naturalistic Benchmark for Comparing Human and LLM Reasoning
+
+- **ID:** `2602.19914` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Thatchawin Leelawat, Lewis D Griffin
+- **abs:** https://arxiv.org/abs/2602.19914 · **PDF:** https://arxiv.org/pdf/2602.19914v1
+
+## 10409. De novo molecular structure elucidation from mass spectra via flow matching
+
+- **ID:** `2602.19912` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Ghaith Mqawass, Tuan Le, Fabian Theis, et al.
+- **abs:** https://arxiv.org/abs/2602.19912 · **PDF:** https://arxiv.org/pdf/2602.19912v2
+
+## 10410. Gradient based Severity Labeling for Biomarker Classification in OCT
+
+- **ID:** `2602.19907` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Kiran Kokilepersaud, Mohit Prabhushankar, Ghassan AlRegib, et al.
+- **abs:** https://arxiv.org/abs/2602.19907 · **PDF:** https://arxiv.org/pdf/2602.19907v1
+
+## 10411. Rethinking Chronological Causal Discovery with Signal Processing
+
+- **ID:** `2602.19903` · **primary:** `eess.SP` · **Submitted:** 2026-02-23
+- **Authors:** Kurt Butler, Damian Machlanski, Panagiotis Dimitrakopoulos, et al.
+- **abs:** https://arxiv.org/abs/2602.19903 · **PDF:** https://arxiv.org/pdf/2602.19903v1
+
+## 10412. DSDR: Dual-Scale Diversity Regularization for Exploration in LLM Reasoning
+
+- **ID:** `2602.19895` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Zhongwei Wan, Yun Shen, Zhihao Dou, et al.
+- **abs:** https://arxiv.org/abs/2602.19895 · **PDF:** https://arxiv.org/pdf/2602.19895v1
+
+## 10413. Generalized Random Direction Newton Algorithms for Stochastic Optimization
+
+- **ID:** `2602.19893` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Soumen Pachal, Prashanth L. A., Shalabh Bhatnagar, et al.
+- **abs:** https://arxiv.org/abs/2602.19893 · **PDF:** https://arxiv.org/pdf/2602.19893v1
+
+## 10414. Denotational Semantics for ODRL: Knowledge-Based Constraint Conflict Detection
+
+- **ID:** `2602.19883` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Daham Mustafa, Diego Collarana, Yixin Peng, et al.
+- **abs:** https://arxiv.org/abs/2602.19883 · **PDF:** https://arxiv.org/pdf/2602.19883v1
+
+## 10415. Make Some Noise: Unsupervised Remote Sensing Change Detection Using Latent Space Perturbations
+
+- **ID:** `2602.19881` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Blaž Rolih, Matic Fučka, Filip Wolf, et al.
+- **abs:** https://arxiv.org/abs/2602.19881 · **PDF:** https://arxiv.org/pdf/2602.19881v1
+
+## 10416. Axis Decomposition for ODRL: Resolving Dimensional Ambiguity in Policy Constraints through Interval Semantics
+
+- **ID:** `2602.19878` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Daham Mustafa, Diego Collarana, Yixin Peng, et al.
+- **abs:** https://arxiv.org/abs/2602.19878 · **PDF:** https://arxiv.org/pdf/2602.19878v2
+
+## 10417. GOAL: Geometrically Optimal Alignment for Continual Generalized Category Discovery
+
+- **ID:** `2602.19872` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Jizhou Han, Chenhao Ding, SongLin Dong, et al.
+- **abs:** https://arxiv.org/abs/2602.19872 · **PDF:** https://arxiv.org/pdf/2602.19872v1
+
+## 10418. Dirichlet Scale Mixture Priors for Bayesian Neural Networks
+
+- **ID:** `2602.19859` · **primary:** `stat.ML` · **Submitted:** 2026-02-23
+- **Authors:** August Arnstad, Leiv Rønneberg, Geir Storvik
+- **abs:** https://arxiv.org/abs/2602.19859 · **PDF:** https://arxiv.org/pdf/2602.19859v1
+
+## 10419. SHIELD: Semantic Heterogeneity Integrated Embedding for Latent Discovery in Clinical Trial Safety Signals
+
+- **ID:** `2602.19855` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Francois Vandenhende, Anna Georgiou, Theodoros Psaras, et al.
+- **abs:** https://arxiv.org/abs/2602.19855 · **PDF:** https://arxiv.org/pdf/2602.19855v1
+
+## 10420. Orthogonal Uplift Learning with Permutation-Invariant Representations for Combinatorial Treatments
+
+- **ID:** `2602.19851` · **primary:** `stat.ME` · **Submitted:** 2026-02-23
+- **Authors:** Xinyan Su, Jiacan Gao, Mingyuan Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.19851 · **PDF:** https://arxiv.org/pdf/2602.19851v1
+
+## 10421. I Dropped a Neural Net
+
+- **ID:** `2602.19845` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Hyunwoo Park
+- **abs:** https://arxiv.org/abs/2602.19845 · **PDF:** https://arxiv.org/pdf/2602.19845v1
+
+## 10422. LLM-enabled Applications Require System-Level Threat Monitoring
+
+- **ID:** `2602.19844` · **primary:** `cs.CR` · **Submitted:** 2026-02-23
+- **Authors:** Yedi Zhang, Haoyu Wang, Xianglin Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.19844 · **PDF:** https://arxiv.org/pdf/2602.19844v1
+
+## 10423. MAS-FIRE: Fault Injection and Reliability Evaluation for LLM-Based Multi-Agent Systems
+
+- **ID:** `2602.19843` · **primary:** `cs.SE` · **Submitted:** 2026-02-23
+- **Authors:** Jin Jia, Zhiling Deng, Zhuangbin Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.19843 · **PDF:** https://arxiv.org/pdf/2602.19843v1
+
+## 10424. Thought Virus: Viral Misalignment via Subliminal Prompting in Multi-Agent Systems
+
+- **ID:** `2603.00131` · **primary:** `cs.MA` · **Submitted:** 2026-02-23
+- **Authors:** Moritz Weckbecker, Jonas Müller, Ben Hagag, et al.
+- **abs:** https://arxiv.org/abs/2603.00131 · **PDF:** https://arxiv.org/pdf/2603.00131v1
+
+## 10425. SAMAS: A Spectrum-Guided Multi-Agent System for Achieving Style Fidelity in Literary Translation
+
+- **ID:** `2602.19840` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Jingzhuo Wu, Jiajun Zhang, Keyan Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.19840 · **PDF:** https://arxiv.org/pdf/2602.19840v1
+
+## 10426. Meta-Learning and Meta-Reinforcement Learning -- Tracing the Path towards DeepMind's Adaptive Agent
+
+- **ID:** `2602.19837` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Björn Hoppmann, Christoph Scholz
+- **abs:** https://arxiv.org/abs/2602.19837 · **PDF:** https://arxiv.org/pdf/2602.19837v2
+
+## 10427. MultiModalPFN: Extending Prior-Data Fitted Networks for Multimodal Tabular Learning
+
+- **ID:** `2602.20223` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Wall Kim, Chaeyoung Song, Hanul Kim
+- **abs:** https://arxiv.org/abs/2602.20223 · **PDF:** https://arxiv.org/pdf/2602.20223v2
+
+## 10428. Efficient endometrial carcinoma screening via cross-modal synthesis and gradient distillation
+
+- **ID:** `2602.19822` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Dongjing Shan, Yamei Luo, Jiqing Xuan, et al.
+- **abs:** https://arxiv.org/abs/2602.19822 · **PDF:** https://arxiv.org/pdf/2602.19822v1
+
+## 10429. Agentic Hives: Equilibrium, Indeterminacy, and Endogenous Cycles in Self-Organizing Multi-Agent Systems
+
+- **ID:** `2603.00130` · **primary:** `cs.MA` · **Submitted:** 2026-02-23
+- **Authors:** Jean-Philippe Garnier
+- **abs:** https://arxiv.org/abs/2603.00130 · **PDF:** https://arxiv.org/pdf/2603.00130v1
+
+## 10430. A General Equilibrium Theory of Orchestrated AI Agent Systems
+
+- **ID:** `2602.21255` · **primary:** `cs.GT` · **Submitted:** 2026-02-23
+- **Authors:** Jean-Philippe Garnier
+- **abs:** https://arxiv.org/abs/2602.21255 · **PDF:** https://arxiv.org/pdf/2602.21255v1
+
+## 10431. SafePickle: Robust and Generic ML Detection of Malicious Pickle-based ML Models
+
+- **ID:** `2602.19818` · **primary:** `cs.CR` · **Submitted:** 2026-02-23
+- **Authors:** Hillel Ohayon, Daniel Gilkarov, Ran Dubin
+- **abs:** https://arxiv.org/abs/2602.19818 · **PDF:** https://arxiv.org/pdf/2602.19818v1
+
+## 10432. Depth-Structured Music Recurrence: Budgeted Recurrent Attention for Full-Piece Symbolic Music Modeling
+
+- **ID:** `2602.19816` · **primary:** `cs.SD` · **Submitted:** 2026-02-23
+- **Authors:** Yungang Yi, Weihua Li, Matthew Kuo, et al.
+- **abs:** https://arxiv.org/abs/2602.19816 · **PDF:** https://arxiv.org/pdf/2602.19816v2
+
+## 10433. Keyboards for the Endangered Idu Mishmi Language
+
+- **ID:** `2602.19815` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Akhilesh Kakolu Ramarao
+- **abs:** https://arxiv.org/abs/2602.19815 · **PDF:** https://arxiv.org/pdf/2602.19815v1
+
+## 10434. From Agent-Only Social Networks to Autonomous Scientific Research: Lessons from OpenClaw and Moltbook, and the Architecture of ClawdLab and Beach.Science
+
+- **ID:** `2602.19810` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Lukas Weidener, Marko Brkić, Phillip Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.19810 · **PDF:** https://arxiv.org/pdf/2602.19810v3
+
+## 10435. Decision MetaMamba: Enhancing Selective SSM in Offline RL with Heterogeneous Sequence Mixing
+
+- **ID:** `2602.19805` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Wall Kim, Chaeyoung Song, Hanul Kim
+- **abs:** https://arxiv.org/abs/2602.19805 · **PDF:** https://arxiv.org/pdf/2602.19805v2
+
+## 10436. Linear Reservoir: A Diagonalization-Based Optimization
+
+- **ID:** `2602.19802` · **primary:** `cs.DC` · **Submitted:** 2026-02-23
+- **Authors:** Romain de Coudenhove, Yannis Bendi-Ouis, Anthony Strock, et al.
+- **abs:** https://arxiv.org/abs/2602.19802 · **PDF:** https://arxiv.org/pdf/2602.19802v1
+
+## 10437. Path-conditioned training: a principled way to rescale ReLU neural networks
+
+- **ID:** `2602.19799` · **primary:** `stat.ML` · **Submitted:** 2026-02-23
+- **Authors:** Arthur Lebeurrier, Titouan Vayer, Rémi Gribonval
+- **abs:** https://arxiv.org/abs/2602.19799 · **PDF:** https://arxiv.org/pdf/2602.19799v1
+
+## 10438. Drift Localization using Conformal Predictions
+
+- **ID:** `2602.19790` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Fabian Hinder, Valerie Vaquet, Johannes Brinkrolf, et al.
+- **abs:** https://arxiv.org/abs/2602.19790 · **PDF:** https://arxiv.org/pdf/2602.19790v1
+
+## 10439. Stop Preaching and Start Practising Data Frugality for Responsible Development of AI
+
+- **ID:** `2602.19789` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Sophia N. Wilson, Guðrún Fjóla Guðmundsdóttir, Andrew Millard, et al.
+- **abs:** https://arxiv.org/abs/2602.19789 · **PDF:** https://arxiv.org/pdf/2602.19789v1
+
+## 10440. Bayesian Meta-Learning with Expert Feedback for Task-Shift Adaptation through Causal Embeddings
+
+- **ID:** `2602.19788` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Lotta Mäkinen, Jorge Loría, Samuel Kaski
+- **abs:** https://arxiv.org/abs/2602.19788 · **PDF:** https://arxiv.org/pdf/2602.19788v1
+
+## 10441. The Climate Change Knowledge Graph: Supporting Climate Services
+
+- **ID:** `2602.19786` · **primary:** `cs.DB` · **Submitted:** 2026-02-23
+- **Authors:** Miguel Ceriani, Fiorela Ciroku, Alessandro Russo, et al.
+- **abs:** https://arxiv.org/abs/2602.19786 · **PDF:** https://arxiv.org/pdf/2602.19786v1
+
+## 10442. Unsupervised Anomaly Detection in NSL-KDD Using $β$-VAE: A Latent Space and Reconstruction Error Approach
+
+- **ID:** `2602.19785` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Dylan Baptiste, Ramla Saddem, Alexandre Philippot, et al.
+- **abs:** https://arxiv.org/abs/2602.19785 · **PDF:** https://arxiv.org/pdf/2602.19785v1
+
+## 10443. Addressing Instrument-Outcome Confounding in Mendelian Randomization through Representation Learning
+
+- **ID:** `2602.19782` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Shimeng Huang, Matthew Robinson, Francesco Locatello
+- **abs:** https://arxiv.org/abs/2602.19782 · **PDF:** https://arxiv.org/pdf/2602.19782v1
+
+## 10444. Enhancing Automatic Chord Recognition via Pseudo-Labeling and Knowledge Distillation
+
+- **ID:** `2602.19778` · **primary:** `cs.SD` · **Submitted:** 2026-02-23
+- **Authors:** Nghia Phan, Rong Jin, Gang Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.19778 · **PDF:** https://arxiv.org/pdf/2602.19778v3
+
+## 10445. Exact Discrete Stochastic Simulation with Deep-Learning-Scale Gradient Optimization
+
+- **ID:** `2602.19775` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-23
+- **Authors:** Jose M. G. Vilar, Leonor Saiz
+- **abs:** https://arxiv.org/abs/2602.19775 · **PDF:** https://arxiv.org/pdf/2602.19775v1
+
+## 10446. The Confusion is Real: GRAPHIC - A Network Science Approach to Confusion Matrices in Deep Learning
+
+- **ID:** `2602.19770` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Johanna S. Fröhlich, Bastian Heinlein, Jan U. Claar, et al.
+- **abs:** https://arxiv.org/abs/2602.19770 · **PDF:** https://arxiv.org/pdf/2602.19770v1
+
+## 10447. Hexagon-MLIR: An AI Compilation Stack For Qualcomm's Neural Processing Units (NPUs)
+
+- **ID:** `2602.19762` · **primary:** `cs.PL` · **Submitted:** 2026-02-23
+- **Authors:** Mohammed Javed Absar, Muthu Baskaran, Abhikrant Sharma, et al.
+- **abs:** https://arxiv.org/abs/2602.19762 · **PDF:** https://arxiv.org/pdf/2602.19762v1
+
+## 10448. Ensemble Machine Learning and Statistical Procedures for Dynamic Predictions of Time-to-Event Outcomes
+
+- **ID:** `2602.19761` · **primary:** `stat.ML` · **Submitted:** 2026-02-23
+- **Authors:** Nina van Gerwen, Sten Willemsen, Bettina E. Hansen, et al.
+- **abs:** https://arxiv.org/abs/2602.19761 · **PDF:** https://arxiv.org/pdf/2602.19761v1
+
+## 10449. Pacing Opinion Polarization via Graph Reinforcement Learning
+
+- **ID:** `2602.23390` · **primary:** `cs.SI` · **Submitted:** 2026-02-23
+- **Authors:** Mingkai Liao
+- **abs:** https://arxiv.org/abs/2602.23390 · **PDF:** https://arxiv.org/pdf/2602.23390v1
+
+## 10450. How Confident Is the First Token? An Uncertainty-Calibrated Prompt Optimization Framework for Large Language Model Classification and Understanding
+
+- **ID:** `2603.18009` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Wei Chen, Guoyang Ju, Yuanyuan Qi
+- **abs:** https://arxiv.org/abs/2603.18009 · **PDF:** https://arxiv.org/pdf/2603.18009v1
+
+## 10451. NILE: Formalizing Natural-Language Descriptions of Formal Languages
+
+- **ID:** `2602.19743` · **primary:** `cs.FL` · **Submitted:** 2026-02-23
+- **Authors:** Tristan Kneisel, Marko Schmellenkamp, Fabian Vehlken, et al.
+- **abs:** https://arxiv.org/abs/2602.19743 · **PDF:** https://arxiv.org/pdf/2602.19743v1
+
+## 10452. Safe Multi-Agent Deep Reinforcement Learning for Privacy-Aware Edge-Device Collaborative DNN Inference
+
+- **ID:** `2603.00129` · **primary:** `cs.MA` · **Submitted:** 2026-02-23
+- **Authors:** Hong Wang, Xuwei Fan, Zhipeng Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.00129 · **PDF:** https://arxiv.org/pdf/2603.00129v1
+
+## 10453. Understanding the Curse of Unrolling
+
+- **ID:** `2602.19733` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Sheheryar Mehmood, Florian Knoll, Peter Ochs
+- **abs:** https://arxiv.org/abs/2602.19733 · **PDF:** https://arxiv.org/pdf/2602.19733v1
+
+## 10454. Carbon-Aware Governance Gates: An Architecture for Sustainable GenAI Development
+
+- **ID:** `2602.19718` · **primary:** `cs.SE` · **Submitted:** 2026-02-23
+- **Authors:** Mateen A. Abbasi, Tommi J. Mikkonen, Petri J. Ihantola, et al.
+- **abs:** https://arxiv.org/abs/2602.19718 · **PDF:** https://arxiv.org/pdf/2602.19718v1
+
+## 10455. Universal Pose Pretraining for Generalizable Vision-Language-Action Policies
+
+- **ID:** `2602.19710` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Haitao Lin, Hanyang Yu, Jingshun Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.19710 · **PDF:** https://arxiv.org/pdf/2602.19710v1
+
+## 10456. Testing Effect Homogeneity and Confounding in High-Dimensional Experimental and Observational Studies
+
+- **ID:** `2602.19703` · **primary:** `econ.EM` · **Submitted:** 2026-02-23
+- **Authors:** Ana Armendariz, Martin Huber
+- **abs:** https://arxiv.org/abs/2602.19703 · **PDF:** https://arxiv.org/pdf/2602.19703v1
+
+## 10457. DReX: An Explainable Deep Learning-based Multimodal Recommendation Framework
+
+- **ID:** `2602.19702` · **primary:** `cs.IR` · **Submitted:** 2026-02-23
+- **Authors:** Adamya Shyam, Venkateswara Rao Kagita, Bharti Rana, et al.
+- **abs:** https://arxiv.org/abs/2602.19702 · **PDF:** https://arxiv.org/pdf/2602.19702v1
+
+## 10458. Iconographic Classification and Content-Based Recommendation for Digitized Artworks
+
+- **ID:** `2602.19698` · **primary:** `cs.DL` · **Submitted:** 2026-02-23
+- **Authors:** Krzysztof Kutt, Maciej Baczyński
+- **abs:** https://arxiv.org/abs/2602.19698 · **PDF:** https://arxiv.org/pdf/2602.19698v1
+
+## 10459. Smoothness Adaptivity in Constant-Depth Neural Networks: Optimal Rates via Smooth Activations
+
+- **ID:** `2602.19691` · **primary:** `stat.ML` · **Submitted:** 2026-02-23
+- **Authors:** Yuhao Liu, Zilin Wang, Lei Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.19691 · **PDF:** https://arxiv.org/pdf/2602.19691v2
+
+## 10460. What Matters for Simulation to Online Reinforcement Learning on Real Robots
+
+- **ID:** `2602.20220` · **primary:** `cs.RO` · **Submitted:** 2026-02-23
+- **Authors:** Yarden As, Dhruva Tirumala, René Zurbrügg, et al.
+- **abs:** https://arxiv.org/abs/2602.20220 · **PDF:** https://arxiv.org/pdf/2602.20220v1
+
+## 10461. PerturbDiff: Functional Diffusion for Single-Cell Perturbation Modeling
+
+- **ID:** `2602.19685` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Xinyu Yuan, Xixian Liu, Ya Shi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.19685 · **PDF:** https://arxiv.org/pdf/2602.19685v1
+
+## 10462. TeHOR: Text-Guided 3D Human and Object Reconstruction with Textures
+
+- **ID:** `2602.19679` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Hyeongjin Nam, Daniel Sungho Jung, Kyoung Mu Lee
+- **abs:** https://arxiv.org/abs/2602.19679 · **PDF:** https://arxiv.org/pdf/2602.19679v1
+
+## 10463. Continuous Telemonitoring of Heart Failure using Personalised Speech Dynamics
+
+- **ID:** `2602.19674` · **primary:** `cs.SD` · **Submitted:** 2026-02-23
+- **Authors:** Yue Pan, Xingyao Wang, Hanyue Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.19674 · **PDF:** https://arxiv.org/pdf/2602.19674v2
+
+## 10464. SkillOrchestra: Learning to Route Agents via Skill Transfer
+
+- **ID:** `2602.19672` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Jiayu Wang, Yifei Ming, Zixuan Ke, et al.
+- **abs:** https://arxiv.org/abs/2602.19672 · **PDF:** https://arxiv.org/pdf/2602.19672v1
+
+## 10465. Personalized Longitudinal Medical Report Generation via Temporally-Aware Federated Adaptation
+
+- **ID:** `2602.19668` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** He Zhu, Ren Togo, Takahiro Ogawa, et al.
+- **abs:** https://arxiv.org/abs/2602.19668 · **PDF:** https://arxiv.org/pdf/2602.19668v1
+
+## 10466. QuickGrasp: Responsive Video-Language Querying Service via Accelerated Tokenization and Edge-Augmented Inference
+
+- **ID:** `2603.00126` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Miao Zhang, Ruixiao Zhang, Jianxin Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.00126 · **PDF:** https://arxiv.org/pdf/2603.00126v1
+
+## 10467. PaReGTA: An LLM-based EHR Data Encoding Approach to Capture Temporal Information
+
+- **ID:** `2602.19661` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Kihyuk Yoon, Lingchao Mao, Catherine Chong, et al.
+- **abs:** https://arxiv.org/abs/2602.19661 · **PDF:** https://arxiv.org/pdf/2602.19661v2
+
+## 10468. Representation Stability in a Minimal Continual Learning Agent
+
+- **ID:** `2602.19655` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Vishnu Subramanian
+- **abs:** https://arxiv.org/abs/2602.19655 · **PDF:** https://arxiv.org/pdf/2602.19655v1
+
+## 10469. NEXUS: A compact neural architecture for high-resolution spatiotemporal air quality forecasting in Delhi National Capital Region
+
+- **ID:** `2602.19654` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Rampunit Kumar, Aditya Maheshwari
+- **abs:** https://arxiv.org/abs/2602.19654 · **PDF:** https://arxiv.org/pdf/2602.19654v2
+
+## 10470. Denoising Particle Filters: Learning State Estimation with Single-Step Objectives
+
+- **ID:** `2602.19651` · **primary:** `cs.RO` · **Submitted:** 2026-02-23
+- **Authors:** Lennart Röstel, Berthold Bäuml
+- **abs:** https://arxiv.org/abs/2602.19651 · **PDF:** https://arxiv.org/pdf/2602.19651v1
+
+## 10471. Spectral Phase Encoding for Quantum Kernel Methods
+
+- **ID:** `2602.19644` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Pablo Herrero Gómez, Antonio Jimeno Morenilla, David Muñoz-Hernández, et al.
+- **abs:** https://arxiv.org/abs/2602.19644 · **PDF:** https://arxiv.org/pdf/2602.19644v1
+
+## 10472. KGHaluBench: A Knowledge Graph-Based Hallucination Benchmark for Evaluating the Breadth and Depth of LLM Knowledge
+
+- **ID:** `2602.19643` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Alex Robertson, Huizhi Liang, Mahbub Gani, et al.
+- **abs:** https://arxiv.org/abs/2602.19643 · **PDF:** https://arxiv.org/pdf/2602.19643v1
+
+## 10473. Evaluating the Impact of Data Anonymization on Image Retrieval
+
+- **ID:** `2602.19641` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Marvin Chen, Manuel Eberhardinger, Johannes Maucher
+- **abs:** https://arxiv.org/abs/2602.19641 · **PDF:** https://arxiv.org/pdf/2602.19641v1
+
+## 10474. Compositional Planning with Jumpy World Models
+
+- **ID:** `2602.19634` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Jesse Farebrother, Matteo Pirotta, Andrea Tirinzoni, et al.
+- **abs:** https://arxiv.org/abs/2602.19634 · **PDF:** https://arxiv.org/pdf/2602.19634v1
+
+## 10475. TAPE: Tool-Guided Adaptive Planning and Constrained Execution in Language Model Agents
+
+- **ID:** `2602.19633` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Jongwon Jeong, Jungtaek Kim, Kangwook Lee
+- **abs:** https://arxiv.org/abs/2602.19633 · **PDF:** https://arxiv.org/pdf/2602.19633v1
+
+## 10476. Localized Concept Erasure in Text-to-Image Diffusion Models via High-Level Representation Misdirection
+
+- **ID:** `2602.19631` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Uichan Lee, Jeonghyeon Kim, Sangheum Hwang
+- **abs:** https://arxiv.org/abs/2602.19631 · **PDF:** https://arxiv.org/pdf/2602.19631v1
+
+## 10477. Cooperation After the Algorithm: Designing Human-AI Coexistence Beyond the Illusion of Collaboration
+
+- **ID:** `2602.19629` · **primary:** `cs.HC` · **Submitted:** 2026-02-23
+- **Authors:** Tatia Codreanu
+- **abs:** https://arxiv.org/abs/2602.19629 · **PDF:** https://arxiv.org/pdf/2602.19629v1
+
+## 10478. Nacrith: Neural Lossless Compression via Ensemble Context Modeling and High-Precision CDF Coding
+
+- **ID:** `2602.19626` · **primary:** `cs.IT` · **Submitted:** 2026-02-23
+- **Authors:** Roberto Tacconelli
+- **abs:** https://arxiv.org/abs/2602.19626 · **PDF:** https://arxiv.org/pdf/2602.19626v2
+
+## 10479. PedaCo-Gen: Scaffolding Pedagogical Agency in Human-AI Collaborative Video Authoring
+
+- **ID:** `2602.19623` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Injun Baek, Yearim Kim, Nojun Kwak
+- **abs:** https://arxiv.org/abs/2602.19623 · **PDF:** https://arxiv.org/pdf/2602.19623v2
+
+## 10480. VecFormer: Towards Efficient and Generalizable Graph Transformer with Graph Token Attention
+
+- **ID:** `2602.19622` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Jingbo Zhou, Jun Xia, Siyuan Li, et al.
+- **abs:** https://arxiv.org/abs/2602.19622 · **PDF:** https://arxiv.org/pdf/2602.19622v1
+
+## 10481. Rules or Weights? Comparing User Understanding of Explainable AI Techniques with the Cognitive XAI-Adaptive Model
+
+- **ID:** `2602.19620` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Louth Bin Rawshan, Zhuoyu Wang, Brian Y Lim
+- **abs:** https://arxiv.org/abs/2602.19620 · **PDF:** https://arxiv.org/pdf/2602.19620v1
+
+## 10482. Is Your Diffusion Sampler Actually Correct? A Sampler-Centric Evaluation of Discrete Diffusion Language Models
+
+- **ID:** `2602.19619` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Luhan Tang, Longxuan Yu, Shaorong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.19619 · **PDF:** https://arxiv.org/pdf/2602.19619v1
+
+## 10483. An Approach to Combining Video and Speech with Large Language Models in Human-Robot Interaction
+
+- **ID:** `2602.20219` · **primary:** `cs.RO` · **Submitted:** 2026-02-23
+- **Authors:** Guanting Shen, Zi Tian
+- **abs:** https://arxiv.org/abs/2602.20219 · **PDF:** https://arxiv.org/pdf/2602.20219v1
+
+## 10484. Workflow-Level Design Principles for Trustworthy GenAI in Automotive System Engineering
+
+- **ID:** `2602.19614` · **primary:** `cs.SE` · **Submitted:** 2026-02-23
+- **Authors:** Chih-Hong Cheng, Brian Hsuan-Cheng Liao, Adam Molin, et al.
+- **abs:** https://arxiv.org/abs/2602.19614 · **PDF:** https://arxiv.org/pdf/2602.19614v1
+
+## 10485. Anatomy of Unlearning: The Dual Impact of Fact Salience and Model Fine-Tuning
+
+- **ID:** `2602.19612` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Borisiuk Anna, Andrey Savchenko, Alexander Panchenko, et al.
+- **abs:** https://arxiv.org/abs/2602.19612 · **PDF:** https://arxiv.org/pdf/2602.19612v2
+
+## 10486. Variational Inference for Bayesian MIDAS Regression
+
+- **ID:** `2602.19610` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Luigi Simeone
+- **abs:** https://arxiv.org/abs/2602.19610 · **PDF:** https://arxiv.org/pdf/2602.19610v1
+
+## 10487. Satellite-Based Detection of Looted Archaeological Sites Using Machine Learning
+
+- **ID:** `2602.19608` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Girmaw Abebe Tadesse, Titien Bartette, Andrew Hassanali, et al.
+- **abs:** https://arxiv.org/abs/2602.19608 · **PDF:** https://arxiv.org/pdf/2602.19608v1
+
+## 10488. CLCR: Cross-Level Semantic Collaborative Representation for Multimodal Learning
+
+- **ID:** `2602.19605` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Chunlei Meng, Guanhong Huang, Rong Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.19605 · **PDF:** https://arxiv.org/pdf/2602.19605v1
+
+## 10489. Manifold-Aligned Generative Transport
+
+- **ID:** `2602.19600` · **primary:** `stat.ML` · **Submitted:** 2026-02-23
+- **Authors:** Xinyu Tian, Xiaotong Shen
+- **abs:** https://arxiv.org/abs/2602.19600 · **PDF:** https://arxiv.org/pdf/2602.19600v1
+
+## 10490. Eye-Tracking-while-Reading: A Living Survey of Datasets with Open Library Support
+
+- **ID:** `2602.19598` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Deborah N. Jakobi, David R. Reich, Paul Prasse, et al.
+- **abs:** https://arxiv.org/abs/2602.19598 · **PDF:** https://arxiv.org/pdf/2602.19598v1
+
+## 10491. ISO-Bench: Can Coding Agents Optimize Real-World Inference Workloads?
+
+- **ID:** `2602.19594` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Ayush Nangia, Shikhar Mishra, Aman Gokrani, et al.
+- **abs:** https://arxiv.org/abs/2602.19594 · **PDF:** https://arxiv.org/pdf/2602.19594v1
+
+## 10492. Detecting High-Potential SMEs with Heterogeneous Graph Neural Networks
+
+- **ID:** `2602.19591` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Yijiashun Qi, Hanzhe Guo, Yijiazhen Qi
+- **abs:** https://arxiv.org/abs/2602.19591 · **PDF:** https://arxiv.org/pdf/2602.19591v2
+
+## 10493. Tri-Subspaces Disentanglement for Multimodal Sentiment Analysis
+
+- **ID:** `2602.19585` · **primary:** `cs.MM` · **Submitted:** 2026-02-23
+- **Authors:** Chunlei Meng, Jiabin Luo, Zhenglin Yan, et al.
+- **abs:** https://arxiv.org/abs/2602.19585 · **PDF:** https://arxiv.org/pdf/2602.19585v1
+
+## 10494. KnapSpec: Self-Speculative Decoding via Adaptive Layer Selection as a Knapsack Problem
+
+- **ID:** `2602.20217` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Seongjin Cha, Gyuwan Kim, Dongsu Han, et al.
+- **abs:** https://arxiv.org/abs/2602.20217 · **PDF:** https://arxiv.org/pdf/2602.20217v1
+
+## 10495. Interpolation-Driven Machine Learning Approaches for Plume Shine Dose Estimation: A Comparison of XGBoost, Random Forest, and TabNet
+
+- **ID:** `2602.19584` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Biswajit Sadhu, Kalpak Gupte, Trijit Sadhu, et al.
+- **abs:** https://arxiv.org/abs/2602.19584 · **PDF:** https://arxiv.org/pdf/2602.19584v1
+
+## 10496. DEEP: Docker-based Execution and Evaluation Platform
+
+- **ID:** `2602.19583` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Sergio Gómez González, Miguel Domingo, Francisco Casacuberta
+- **abs:** https://arxiv.org/abs/2602.19583 · **PDF:** https://arxiv.org/pdf/2602.19583v1
+
+## 10497. Advantage-based Temporal Attack in Reinforcement Learning
+
+- **ID:** `2602.19582` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Shenghong He
+- **abs:** https://arxiv.org/abs/2602.19582 · **PDF:** https://arxiv.org/pdf/2602.19582v1
+
+## 10498. Leap+Verify: Regime-Adaptive Speculative Weight Prediction for Accelerating Neural Network Training
+
+- **ID:** `2602.19580` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Jeremy McEntire
+- **abs:** https://arxiv.org/abs/2602.19580 · **PDF:** https://arxiv.org/pdf/2602.19580v1
+
+## 10499. Goal-Oriented Influence-Maximizing Data Acquisition for Learning and Optimization
+
+- **ID:** `2602.19578` · **primary:** `stat.ML` · **Submitted:** 2026-02-23
+- **Authors:** Weichi Yao, Bianca Dumitrascu, Bryan R. Goldsmith, et al.
+- **abs:** https://arxiv.org/abs/2602.19578 · **PDF:** https://arxiv.org/pdf/2602.19578v1
+
+## 10500. CTC-TTS: LLM-based dual-streaming text-to-speech with CTC alignment
+
+- **ID:** `2602.19574` · **primary:** `eess.AS` · **Submitted:** 2026-02-23
+- **Authors:** Hanwen Liu, Saierdaer Yusuyin, Hao Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.19574 · **PDF:** https://arxiv.org/pdf/2602.19574v1
+
+## 10501. Temporal-Aware Heterogeneous Graph Reasoning with Multi-View Fusion for Temporal Question Answering
+
+- **ID:** `2602.19569` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Wuzhenghong Wen, Bowen Zhou, Jinwen Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.19569 · **PDF:** https://arxiv.org/pdf/2602.19569v1
+
+## 10502. OrthoAI: A Neurosymbolic Framework for Evidence-Grounded Biomechanical Reasoning in Clear Aligner Orthodontics
+
+- **ID:** `2603.00124` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Edouard Lansiaux, Margaux Leman, Mehdi Ammi
+- **abs:** https://arxiv.org/abs/2603.00124 · **PDF:** https://arxiv.org/pdf/2603.00124v2
+
+## 10503. DICArt: Advancing Category-level Articulated Object Pose Estimation in Discrete State-Spaces
+
+- **ID:** `2602.19565` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Li Zhang, Mingyu Mei, Ailing Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.19565 · **PDF:** https://arxiv.org/pdf/2602.19565v2
+
+## 10504. A Multimodal Framework for Aligning Human Linguistic Descriptions with Visual Perceptual Data
+
+- **ID:** `2602.19562` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Joseph Bingham
+- **abs:** https://arxiv.org/abs/2602.19562 · **PDF:** https://arxiv.org/pdf/2602.19562v1
+
+## 10505. CT-Flow: Orchestrating CT Interpretation Workflow with Model Context Protocol Servers
+
+- **ID:** `2603.00123` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Yannian Gu, Xizhuo Zhang, Linjie Mu, et al.
+- **abs:** https://arxiv.org/abs/2603.00123 · **PDF:** https://arxiv.org/pdf/2603.00123v1
+
+## 10506. Right to History: A Sovereignty Kernel for Verifiable AI Agent Execution
+
+- **ID:** `2602.20214` · **primary:** `cs.CR` · **Submitted:** 2026-02-23
+- **Authors:** Jing Zhang
+- **abs:** https://arxiv.org/abs/2602.20214 · **PDF:** https://arxiv.org/pdf/2602.20214v1
+
+## 10507. Agentic AI as a Cybersecurity Attack Surface: Threats, Exploits, and Defenses in Runtime Supply Chains
+
+- **ID:** `2602.19555` · **primary:** `cs.CR` · **Submitted:** 2026-02-23
+- **Authors:** Xiaochong Jiang, Shiqi Yang, Wenting Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.19555 · **PDF:** https://arxiv.org/pdf/2602.19555v1
+
+## 10508. The Sample Complexity of Replicable Realizable PAC Learning
+
+- **ID:** `2602.19552` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Kasper Green Larsen, Markus Engelund Mathiasen, Chirag Pabbaraju, et al.
+- **abs:** https://arxiv.org/abs/2602.19552 · **PDF:** https://arxiv.org/pdf/2602.19552v1
+
+## 10509. Sculpting the Vector Space: Towards Efficient Multi-Vector Visual Document Retrieval via Prune-then-Merge Framework
+
+- **ID:** `2602.19549` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Yibo Yan, Mingdong Ou, Yi Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.19549 · **PDF:** https://arxiv.org/pdf/2602.19549v1
+
+## 10510. Beyond a Single Extractor: Re-thinking HTML-to-Text Extraction for LLM Pretraining
+
+- **ID:** `2602.19548` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Jeffrey Li, Josh Gardner, Doug Kang, et al.
+- **abs:** https://arxiv.org/abs/2602.19548 · **PDF:** https://arxiv.org/pdf/2602.19548v1
+
+## 10511. Hyper-KGGen: A Skill-Driven Knowledge Extractor for High-Quality Knowledge Hypergraph Generation
+
+- **ID:** `2602.19543` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Rizhuo Huang, Yifan Feng, Rundong Xue, et al.
+- **abs:** https://arxiv.org/abs/2602.19543 · **PDF:** https://arxiv.org/pdf/2602.19543v1
+
+## 10512. A Green Learning Approach to LDCT Image Restoration
+
+- **ID:** `2602.19540` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Wei Wang, Yixing Wu, C. -C. Jay Kuo
+- **abs:** https://arxiv.org/abs/2602.19540 · **PDF:** https://arxiv.org/pdf/2602.19540v1
+
+## 10513. Can a Teenager Fool an AI? Evaluating Low-Cost Cosmetic Attacks on Age Estimation Systems
+
+- **ID:** `2602.19539` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Xingyu Shen, Tommy Duong, Xiaodong An, et al.
+- **abs:** https://arxiv.org/abs/2602.19539 · **PDF:** https://arxiv.org/pdf/2602.19539v1
+
+## 10514. Cost-Aware Diffusion Active Search
+
+- **ID:** `2602.19538` · **primary:** `cs.RO` · **Submitted:** 2026-02-23
+- **Authors:** Arundhati Banerjee, Jeff Schneider
+- **abs:** https://arxiv.org/abs/2602.19538 · **PDF:** https://arxiv.org/pdf/2602.19538v1
+
+## 10515. NovaLAD: A Fast, CPU-Optimized Document Extraction Pipeline for Generative AI and Data Intelligence
+
+- **ID:** `2603.00122` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Aman Ulla
+- **abs:** https://arxiv.org/abs/2603.00122 · **PDF:** https://arxiv.org/pdf/2603.00122v1
+
+## 10516. TherapyGym: Evaluating and Aligning Clinical Fidelity and Safety in Therapy Chatbots
+
+- **ID:** `2603.18008` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Fangrui Huang, Souhad Chbeir, Arpandeep Khatua, et al.
+- **abs:** https://arxiv.org/abs/2603.18008 · **PDF:** https://arxiv.org/pdf/2603.18008v1
+
+## 10517. Fore-Mamba3D: Mamba-based Foreground-Enhanced Encoding for 3D Object Detection
+
+- **ID:** `2602.19536` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Zhiwei Ning, Xuanang Gao, Jiaxi Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.19536 · **PDF:** https://arxiv.org/pdf/2602.19536v1
+
+## 10518. CodeHacker: Automated Test Case Generation for Detecting Vulnerabilities in Competitive Programming Solutions
+
+- **ID:** `2602.20213` · **primary:** `cs.SE` · **Submitted:** 2026-02-23
+- **Authors:** Jingwei Shi, Xinxiang Yin, Jing Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.20213 · **PDF:** https://arxiv.org/pdf/2602.20213v1
+
+## 10519. Large Language Model-Assisted UAV Operations and Communications: A Multifaceted Survey and Tutorial
+
+- **ID:** `2602.19534` · **primary:** `cs.RO` · **Submitted:** 2026-02-23
+- **Authors:** Yousef Emami, Hao Zhou, Radha Reddy, et al.
+- **abs:** https://arxiv.org/abs/2602.19534 · **PDF:** https://arxiv.org/pdf/2602.19534v2
+
+## 10520. Grokking Finite-Dimensional Algebra
+
+- **ID:** `2602.19533` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Pascal Jr Tikeng Notsawo, Guillaume Dumas, Guillaume Rabusseau
+- **abs:** https://arxiv.org/abs/2602.19533 · **PDF:** https://arxiv.org/pdf/2602.19533v1
+
+## 10521. A Statistical Approach for Modeling Irregular Multivariate Time Series with Missing Observations
+
+- **ID:** `2602.19531` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Dingyi Nie, Yixing Wu, C. -C. Jay Kuo
+- **abs:** https://arxiv.org/abs/2602.19531 · **PDF:** https://arxiv.org/pdf/2602.19531v1
+
+## 10522. Beyond Accuracy: A Unified Random Matrix Theory Diagnostic Framework for Crash Classification Models
+
+- **ID:** `2602.19528` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma
+- **abs:** https://arxiv.org/abs/2602.19528 · **PDF:** https://arxiv.org/pdf/2602.19528v1
+
+## 10523. How to Train Your Deep Research Agent? Prompt, Reward, and Policy Optimization in Search-R1
+
+- **ID:** `2602.19526` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Yinuo Xu, Shuo Lu, Jianjie Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.19526 · **PDF:** https://arxiv.org/pdf/2602.19526v1
+
+## 10524. Ada-RS: Adaptive Rejection Sampling for Selective Thinking
+
+- **ID:** `2602.19519` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Yirou Ge, Yixi Li, Alec Chiu, et al.
+- **abs:** https://arxiv.org/abs/2602.19519 · **PDF:** https://arxiv.org/pdf/2602.19519v1
+
+## 10525. Classroom Final Exam: An Instructor-Tested Reasoning Benchmark
+
+- **ID:** `2602.19517` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Chongyang Gao, Diji Yang, Shuyan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.19517 · **PDF:** https://arxiv.org/pdf/2602.19517v2
+
+## 10526. GreenPhase: A Green Learning Approach for Earthquake Phase Picking
+
+- **ID:** `2603.03344` · **primary:** `physics.geo-ph` · **Submitted:** 2026-02-23
+- **Authors:** Yixing Wu, Shiou-Ya Wang, Dingyi Nie, et al.
+- **abs:** https://arxiv.org/abs/2603.03344 · **PDF:** https://arxiv.org/pdf/2603.03344v1
+
+## 10527. Real-time Win Probability and Latent Player Ability via STATS X in Team Sports
+
+- **ID:** `2602.19513` · **primary:** `stat.AP` · **Submitted:** 2026-02-23
+- **Authors:** Yasutaka Shimizu, Atsushi Yamanobe
+- **abs:** https://arxiv.org/abs/2602.19513 · **PDF:** https://arxiv.org/pdf/2602.19513v1
+
+## 10528. Variational Trajectory Optimization of Anisotropic Diffusion Schedules
+
+- **ID:** `2602.19512` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Pengxi Liu, Zeyu Michael Li, Xiang Cheng
+- **abs:** https://arxiv.org/abs/2602.19512 · **PDF:** https://arxiv.org/pdf/2602.19512v1
+
+## 10529. Less is More: Convergence Benefits of Fewer Data Weight Updates over Longer Horizon
+
+- **ID:** `2602.19510` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Rudrajit Das, Neel Patel, Meisam Razaviyayn, et al.
+- **abs:** https://arxiv.org/abs/2602.19510 · **PDF:** https://arxiv.org/pdf/2602.19510v1
+
+## 10530. Pyramid MoA: A Probabilistic Framework for Cost-Optimized Anytime Inference
+
+- **ID:** `2602.19509` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Arindam Khaled
+- **abs:** https://arxiv.org/abs/2602.19509 · **PDF:** https://arxiv.org/pdf/2602.19509v2
+
+## 10531. Relational Feature Caching for Accelerating Diffusion Transformers
+
+- **ID:** `2602.19506` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Byunggwan Son, Jeimin Jeon, Jeongwoo Choi, et al.
+- **abs:** https://arxiv.org/abs/2602.19506 · **PDF:** https://arxiv.org/pdf/2602.19506v1
+
+## 10532. Human-Guided Agentic AI for Multimodal Clinical Prediction: Lessons from the AgentDS Healthcare Benchmark
+
+- **ID:** `2602.19502` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Lalitha Pranathi Pulavarthy, Raajitha Muthyala, Aravind V Kuruvikkattil, et al.
+- **abs:** https://arxiv.org/abs/2602.19502 · **PDF:** https://arxiv.org/pdf/2602.19502v1
+
+## 10533. Softmax is not Enough (for Adaptive Conformal Classification)
+
+- **ID:** `2602.19498` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Navid Akhavan Attar, Hesam Asadollahzadeh, Ling Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.19498 · **PDF:** https://arxiv.org/pdf/2602.19498v1
+
+## 10534. Botson: An Accessible and Low-Cost Platform for Social Robotics Research
+
+- **ID:** `2602.19491` · **primary:** `cs.RO` · **Submitted:** 2026-02-23
+- **Authors:** Samuel Bellaire, Abdalmalek Abu-raddaha, Natalie Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.19491 · **PDF:** https://arxiv.org/pdf/2602.19491v1
+
+## 10535. Federated Learning Playground
+
+- **ID:** `2602.19489` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Bryan Shan, Alysa Ziying Tan, Han Yu
+- **abs:** https://arxiv.org/abs/2602.19489 · **PDF:** https://arxiv.org/pdf/2602.19489v2
+
+## 10536. Multimodal Crystal Flow: Any-to-Any Modality Generation for Unified Crystal Modeling
+
+- **ID:** `2602.20210` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Kiyoung Seong, Sungsoo Ahn, Sehui Han, et al.
+- **abs:** https://arxiv.org/abs/2602.20210 · **PDF:** https://arxiv.org/pdf/2602.20210v1
+
+## 10537. Making Conformal Predictors Robust in Healthcare Settings: a Case Study on EEG Classification
+
+- **ID:** `2602.19483` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Arjun Chatterjee, Sayeed Sajjad Razin, John Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.19483 · **PDF:** https://arxiv.org/pdf/2602.19483v1
+
+## 10538. Scale-PINN: Learning Efficient Physics-Informed Neural Networks Through Sequential Correction
+
+- **ID:** `2602.19475` · **primary:** `cs.CE` · **Submitted:** 2026-02-23
+- **Authors:** Pao-Hsiung Chiu, Jian Cheng Wong, Chin Chun Ooi, et al.
+- **abs:** https://arxiv.org/abs/2602.19475 · **PDF:** https://arxiv.org/pdf/2602.19475v1
+
+## 10539. The generalized underlap coefficient with an application in clustering
+
+- **ID:** `2602.19473` · **primary:** `stat.ME` · **Submitted:** 2026-02-23
+- **Authors:** Zhaoxi Zhang, Vanda Inacio, Sara Wade
+- **abs:** https://arxiv.org/abs/2602.19473 · **PDF:** https://arxiv.org/pdf/2602.19473v2
+
+## 10540. Regressor-guided Diffusion Model for De Novo Peptide Sequencing with Explicit Mass Control
+
+- **ID:** `2602.20209` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-23
+- **Authors:** Shaorong Chen, Jingbo Zhou, Jun Xia
+- **abs:** https://arxiv.org/abs/2602.20209 · **PDF:** https://arxiv.org/pdf/2602.20209v1
+
+## 10541. Can Large Language Models Replace Human Coders? Introducing ContentBench
+
+- **ID:** `2602.19467` · **primary:** `cs.CY` · **Submitted:** 2026-02-23
+- **Authors:** Michael Haman
+- **abs:** https://arxiv.org/abs/2602.19467 · **PDF:** https://arxiv.org/pdf/2602.19467v1
+
+## 10542. PuppetChat: Fostering Intimate Communication through Bidirectional Actions and Micronarratives
+
+- **ID:** `2602.19463` · **primary:** `cs.HC` · **Submitted:** 2026-02-23
+- **Authors:** Emma Jiren Wang, Siying Hu, Zhicong Lu
+- **abs:** https://arxiv.org/abs/2602.19463 · **PDF:** https://arxiv.org/pdf/2602.19463v2
+
+## 10543. Laplacian Multi-scale Flow Matching for Generative Modeling
+
+- **ID:** `2602.19461` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Zelin Zhao, Petr Molodyk, Haotian Xue, et al.
+- **abs:** https://arxiv.org/abs/2602.19461 · **PDF:** https://arxiv.org/pdf/2602.19461v1
+
+## 10544. ComplLLM: Fine-tuning LLMs to Discover Complementary Signals for Decision-making
+
+- **ID:** `2602.19458` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Ziyang Guo, Yifan Wu, Jason Hartline, et al.
+- **abs:** https://arxiv.org/abs/2602.19458 · **PDF:** https://arxiv.org/pdf/2602.19458v1
+
+## 10545. SenTSR-Bench: Thinking with Injected Knowledge for Time-Series Reasoning
+
+- **ID:** `2602.19455` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Zelin He, Boran Han, Xiyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.19455 · **PDF:** https://arxiv.org/pdf/2602.19455v1
+
+## 10546. Red-Teaming Claude Opus and ChatGPT-based Security Advisors for Trusted Execution Environments
+
+- **ID:** `2602.19450` · **primary:** `cs.CR` · **Submitted:** 2026-02-23
+- **Authors:** Kunal Mukherjee
+- **abs:** https://arxiv.org/abs/2602.19450 · **PDF:** https://arxiv.org/pdf/2602.19450v1
+
+## 10547. PIS: A Physics-Informed System for Accurate State Partitioning of $Aβ_{42}$ Protein Trajectories
+
+- **ID:** `2602.19444` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Qianfeng Yu, Ningkang Peng, Yanhui Gu
+- **abs:** https://arxiv.org/abs/2602.19444 · **PDF:** https://arxiv.org/pdf/2602.19444v1
+
+## 10548. When AI Teammates Meet Code Review: Collaboration Signals Shaping the Integration of Agent-Authored Pull Requests
+
+- **ID:** `2602.19441` · **primary:** `cs.SE` · **Submitted:** 2026-02-23
+- **Authors:** Costain Nachuma, Minhaz Zibran
+- **abs:** https://arxiv.org/abs/2602.19441 · **PDF:** https://arxiv.org/pdf/2602.19441v1
+
+## 10549. OptiRepair: Closed-Loop Diagnosis and Repair of Supply Chain Optimization Models with LLM Agents
+
+- **ID:** `2602.19439` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Ruicheng Ao, David Simchi-Levi, Xinshang Wang
+- **abs:** https://arxiv.org/abs/2602.19439 · **PDF:** https://arxiv.org/pdf/2602.19439v2
+
+## 10550. FinSight-Net:A Physics-Aware Decoupled Network with Frequency-Domain Compensation for Underwater Fish Detection in Smart Aquaculture
+
+- **ID:** `2602.19437` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Jinsong Yang, Zeyuan Hu, Yichen Li, et al.
+- **abs:** https://arxiv.org/abs/2602.19437 · **PDF:** https://arxiv.org/pdf/2602.19437v1
+
+## 10551. SIGMAS: Second-Order Interaction-based Grouping for Overlapping Multi-Agent Swarms
+
+- **ID:** `2603.00120` · **primary:** `cs.MA` · **Submitted:** 2026-02-23
+- **Authors:** Minah Lee, Saibal Mukhopadhyay
+- **abs:** https://arxiv.org/abs/2603.00120 · **PDF:** https://arxiv.org/pdf/2603.00120v1
+
+## 10552. RAmmStein: Regime Adaptation in Mean-reverting Markets with Stein Thresholds -- Optimal Impulse Control in Concentrated AMMs
+
+- **ID:** `2602.19419` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Pranay Anchuri
+- **abs:** https://arxiv.org/abs/2602.19419 · **PDF:** https://arxiv.org/pdf/2602.19419v2
+
+## 10553. IR$^3$: Contrastive Inverse Reinforcement Learning for Interpretable Detection and Mitigation of Reward Hacking
+
+- **ID:** `2602.19416` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Mohammad Beigi, Ming Jin, Junshan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.19416 · **PDF:** https://arxiv.org/pdf/2602.19416v1
+
+## 10554. Federated Causal Representation Learning in State-Space Systems for Decentralized Counterfactual Reasoning
+
+- **ID:** `2602.19414` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Nazal Mohamed, Ayush Mohanty, Nagi Gebraeel
+- **abs:** https://arxiv.org/abs/2602.19414 · **PDF:** https://arxiv.org/pdf/2602.19414v2
+
+## 10555. Redefining the Down-Sampling Scheme of U-Net for Precision Biomedical Image Segmentation
+
+- **ID:** `2602.19412` · **primary:** `cs.CV` · **Submitted:** 2026-02-23
+- **Authors:** Mingjie Li, Yizheng Chen, Md Tauhidul Islam, et al.
+- **abs:** https://arxiv.org/abs/2602.19412 · **PDF:** https://arxiv.org/pdf/2602.19412v1
+
+## 10556. MACE-POLAR-1: A Polarisable Electrostatic Foundation Model for Molecular Chemistry
+
+- **ID:** `2602.19411` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-23
+- **Authors:** Ilyes Batatia, William J. Baldwin, Domantas Kuryla, et al.
+- **abs:** https://arxiv.org/abs/2602.19411 · **PDF:** https://arxiv.org/pdf/2602.19411v1
+
+## 10557. BioEnvSense: A Human-Centred Security Framework for Preventing Behaviour-Driven Cyber Incidents
+
+- **ID:** `2602.19410` · **primary:** `cs.CR` · **Submitted:** 2026-02-23
+- **Authors:** Duy Anh Ta, Farnaz Farid, Farhad Ahamed, et al.
+- **abs:** https://arxiv.org/abs/2602.19410 · **PDF:** https://arxiv.org/pdf/2602.19410v1
+
+## 10558. LEVDA: Latent Ensemble Variational Data Assimilation via Differentiable Dynamics
+
+- **ID:** `2602.19406` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Phillip Si, Peng Chen
+- **abs:** https://arxiv.org/abs/2602.19406 · **PDF:** https://arxiv.org/pdf/2602.19406v1
+
+## 10559. Model Merging in the Essential Subspace
+
+- **ID:** `2602.20208` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Longhua Li, Lei Qi, Qi Tian, et al.
+- **abs:** https://arxiv.org/abs/2602.20208 · **PDF:** https://arxiv.org/pdf/2602.20208v1
+
+## 10560. Personalized Prediction of Perceived Message Effectiveness Using Large Language Model Based Digital Twins
+
+- **ID:** `2602.19403` · **primary:** `cs.CL` · **Submitted:** 2026-02-23
+- **Authors:** Jasmin Han, Janardan Devkota, Joseph Waring, et al.
+- **abs:** https://arxiv.org/abs/2602.19403 · **PDF:** https://arxiv.org/pdf/2602.19403v1
+
+## 10561. Hilbert-Augmented Reinforcement Learning for Scalable Multi-Robot Coverage and Exploration
+
+- **ID:** `2602.19400` · **primary:** `cs.RO` · **Submitted:** 2026-02-23
+- **Authors:** Tamil Selvan Gurunathan, Aryya Gangopadhyay
+- **abs:** https://arxiv.org/abs/2602.19400 · **PDF:** https://arxiv.org/pdf/2602.19400v1
+
+## 10562. Hiding in Plain Text: Detecting Concealed Jailbreaks via Activation Disentanglement
+
+- **ID:** `2602.19396` · **primary:** `cs.AI` · **Submitted:** 2026-02-23
+- **Authors:** Amirhossein Farzam, Majid Behabahani, Mani Malek, et al.
+- **abs:** https://arxiv.org/abs/2602.19396 · **PDF:** https://arxiv.org/pdf/2602.19396v1
+
+## 10563. In Defense of Cosine Similarity: Normalization Eliminates the Gauge Freedom
+
+- **ID:** `2602.19393` · **primary:** `cs.LG` · **Submitted:** 2026-02-23
+- **Authors:** Taha Bouhsine
+- **abs:** https://arxiv.org/abs/2602.19393 · **PDF:** https://arxiv.org/pdf/2602.19393v1
+
+## 10564. Spiking Graph Predictive Coding for Reliable OOD Generalization
+
+- **ID:** `2602.19392` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Jing Ren, Jiapeng Du, Bowen Li, et al.
+- **abs:** https://arxiv.org/abs/2602.19392 · **PDF:** https://arxiv.org/pdf/2602.19392v1
+
+## 10565. VCDF: A Validated Consensus-Driven Framework for Time Series Causal Discovery
+
+- **ID:** `2602.21381` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Gene Yu, Ce Guo, Wayne Luk
+- **abs:** https://arxiv.org/abs/2602.21381 · **PDF:** https://arxiv.org/pdf/2602.21381v1
+
+## 10566. Artificial Intelligence for Modeling & Simulation in Digital Twins
+
+- **ID:** `2602.19390` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Philipp Zech, Istvan David
+- **abs:** https://arxiv.org/abs/2602.19390 · **PDF:** https://arxiv.org/pdf/2602.19390v1
+
+## 10567. Adaptive Data Augmentation with Multi-armed Bandit: Sample-Efficient Embedding Calibration for Implicit Pattern Recognition
+
+- **ID:** `2602.19385` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Minxue Tang, Yangyang Yu, Aolin Ding, et al.
+- **abs:** https://arxiv.org/abs/2602.19385 · **PDF:** https://arxiv.org/pdf/2602.19385v2
+
+## 10568. Neuro-Symbolic Decoding of Neural Activity
+
+- **ID:** `2603.03343` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-22
+- **Authors:** Yanchen Wang, Joy Hsu, Ehsan Adeli, et al.
+- **abs:** https://arxiv.org/abs/2603.03343 · **PDF:** https://arxiv.org/pdf/2603.03343v1
+
+## 10569. Regularity of Second-Order Elliptic PDEs in Spectral Barron Spaces
+
+- **ID:** `2602.19381` · **primary:** `math.AP` · **Submitted:** 2026-02-22
+- **Authors:** Ziang Chen, Liqiang Huang, Mengxuan Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.19381 · **PDF:** https://arxiv.org/pdf/2602.19381v1
+
+## 10570. Stable Deep Reinforcement Learning via Isotropic Gaussian Representations
+
+- **ID:** `2602.19373` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Ali Saheb Pasand, Johan Obando-Ceron, Aaron Courville, et al.
+- **abs:** https://arxiv.org/abs/2602.19373 · **PDF:** https://arxiv.org/pdf/2602.19373v2
+
+## 10571. Golden Layers and Where to Find Them: Improved Knowledge Editing for Large Language Models Via Layer Gradient Analysis
+
+- **ID:** `2602.20207` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Shrestha Datta, Hongfu Liu, Anshuman Chhabra
+- **abs:** https://arxiv.org/abs/2602.20207 · **PDF:** https://arxiv.org/pdf/2602.20207v2
+
+## 10572. Seeing Farther and Smarter: Value-Guided Multi-Path Reflection for VLM Policy Optimization
+
+- **ID:** `2602.19372` · **primary:** `cs.RO` · **Submitted:** 2026-02-22
+- **Authors:** Yanting Yang, Shenyuan Gao, Qingwen Bu, et al.
+- **abs:** https://arxiv.org/abs/2602.19372 · **PDF:** https://arxiv.org/pdf/2602.19372v1
+
+## 10573. Time Series, Vision, and Language: Exploring the Limits of Alignment in Contrastive Representation Spaces
+
+- **ID:** `2602.19367` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Pratham Yashwante, Rose Yu
+- **abs:** https://arxiv.org/abs/2602.19367 · **PDF:** https://arxiv.org/pdf/2602.19367v1
+
+## 10574. LLMs Can Learn to Reason Via Off-Policy RL
+
+- **ID:** `2602.19362` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Daniel Ritter, Owen Oertell, Bradley Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.19362 · **PDF:** https://arxiv.org/pdf/2602.19362v2
+
+## 10575. Vid2Sid: Videos Can Help Close the Sim2Real Gap
+
+- **ID:** `2602.19359` · **primary:** `cs.RO` · **Submitted:** 2026-02-22
+- **Authors:** Kevin Qiu, Yu Zhang, Marek Cygan, et al.
+- **abs:** https://arxiv.org/abs/2602.19359 · **PDF:** https://arxiv.org/pdf/2602.19359v1
+
+## 10576. MentalBlackboard: Evaluating Spatial Visualization via Mathematical Transformations
+
+- **ID:** `2602.19357` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Nilay Yilmaz, Maitreya Patel, Naga Sai Abhiram Kusumba, et al.
+- **abs:** https://arxiv.org/abs/2602.19357 · **PDF:** https://arxiv.org/pdf/2602.19357v1
+
+## 10577. Active perception and disentangled representations allow continual, episodic zero and few-shot learning
+
+- **ID:** `2602.19355` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** David Rawlinson, Gideon Kowadlo
+- **abs:** https://arxiv.org/abs/2602.19355 · **PDF:** https://arxiv.org/pdf/2602.19355v1
+
+## 10578. UP-Fuse: Uncertainty-guided LiDAR-Camera Fusion for 3D Panoptic Segmentation
+
+- **ID:** `2602.19349` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Rohit Mohan, Florian Drews, Yakov Miron, et al.
+- **abs:** https://arxiv.org/abs/2602.19349 · **PDF:** https://arxiv.org/pdf/2602.19349v1
+
+## 10579. MultiDiffSense: Diffusion-Based Multi-Modal Visuo-Tactile Image Generation Conditioned on Object Shape and Contact Pose
+
+- **ID:** `2602.19348` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Sirine Bhouri, Lan Wei, Jian-Qing Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.19348 · **PDF:** https://arxiv.org/pdf/2602.19348v1
+
+## 10580. Mitigating "Epistemic Debt" in Generative AI-Scaffolded Novice Programming using Metacognitive Scripts
+
+- **ID:** `2602.20206` · **primary:** `cs.SE` · **Submitted:** 2026-02-22
+- **Authors:** Sreecharan Sankaranarayanan
+- **abs:** https://arxiv.org/abs/2602.20206 · **PDF:** https://arxiv.org/pdf/2602.20206v2
+
+## 10581. Smooth Gate Functions for Soft Advantage Policy Optimization
+
+- **ID:** `2602.19345` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Egor Denisov, Svetlana Glazyrina, Maksim Kryzhanovskiy, et al.
+- **abs:** https://arxiv.org/abs/2602.19345 · **PDF:** https://arxiv.org/pdf/2602.19345v2
+
+## 10582. SplitLight: An Exploratory Toolkit for Recommender Systems Datasets and Splits
+
+- **ID:** `2602.19339` · **primary:** `cs.IR` · **Submitted:** 2026-02-22
+- **Authors:** Anna Volodkevich, Dmitry Anikin, Danil Gusak, et al.
+- **abs:** https://arxiv.org/abs/2602.19339 · **PDF:** https://arxiv.org/pdf/2602.19339v1
+
+## 10583. PerSoMed: A Large-Scale Balanced Dataset for Persian Social Media Text Classification
+
+- **ID:** `2602.19333` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Isun Chehreh, Ebrahim Ansari
+- **abs:** https://arxiv.org/abs/2602.19333 · **PDF:** https://arxiv.org/pdf/2602.19333v1
+
+## 10584. Training-Free Cross-Architecture Merging for Graph Neural Networks
+
+- **ID:** `2602.19332` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Rishabh Bhattacharya, Vikaskumar Kalsariya, Naresh Manwani
+- **abs:** https://arxiv.org/abs/2602.19332 · **PDF:** https://arxiv.org/pdf/2602.19332v1
+
+## 10585. Partial Soft-Matching Distance for Neural Representational Comparison with Partial Unit Correspondence
+
+- **ID:** `2602.19331` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Chaitanya Kapoor, Alex H. Williams, Meenakshi Khosla
+- **abs:** https://arxiv.org/abs/2602.19331 · **PDF:** https://arxiv.org/pdf/2602.19331v1
+
+## 10586. CTS-Bench: Benchmarking Graph Coarsening Trade-offs for GNNs in Clock Tree Synthesis
+
+- **ID:** `2602.19330` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Barsat Khadka, Kawsher Roxy, Md Rubel Ahmed
+- **abs:** https://arxiv.org/abs/2602.19330 · **PDF:** https://arxiv.org/pdf/2602.19330v1
+
+## 10587. Dynamic Elasticity Between Forest Loss and Carbon Emissions: A Subnational Panel Analysis of the United States
+
+- **ID:** `2602.19329` · **primary:** `stat.AP` · **Submitted:** 2026-02-22
+- **Authors:** Keonvin Park
+- **abs:** https://arxiv.org/abs/2602.19329 · **PDF:** https://arxiv.org/pdf/2602.19329v1
+
+## 10588. Soft Sequence Policy Optimization
+
+- **ID:** `2602.19327` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Svetlana Glazyrina, Maksim Kryzhanovskiy, Roman Ischenko
+- **abs:** https://arxiv.org/abs/2602.19327 · **PDF:** https://arxiv.org/pdf/2602.19327v2
+
+## 10589. City Editing: Hierarchical Agentic Execution for Dependency-Aware Urban Geospatial Modification
+
+- **ID:** `2602.19326` · **primary:** `cs.MA` · **Submitted:** 2026-02-22
+- **Authors:** Rui Liu, Steven Jige Quan, Zhong-Ren Peng, et al.
+- **abs:** https://arxiv.org/abs/2602.19326 · **PDF:** https://arxiv.org/pdf/2602.19326v2
+
+## 10590. RetinaVision: XAI-Driven Augmented Regulation for Precise Retinal Disease Classification using deep learning framework
+
+- **ID:** `2602.19324` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Mohammad Tahmid Noor, Shayan Abrar, Jannatul Adan Mahi, et al.
+- **abs:** https://arxiv.org/abs/2602.19324 · **PDF:** https://arxiv.org/pdf/2602.19324v1
+
+## 10591. US-JEPA: A Joint Embedding Predictive Architecture for Medical Ultrasound
+
+- **ID:** `2602.19322` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Ashwath Radhachandran, Vedrana Ivezić, Shreeram Athreya, et al.
+- **abs:** https://arxiv.org/abs/2602.19322 · **PDF:** https://arxiv.org/pdf/2602.19322v1
+
+## 10592. Anatomy of Agentic Memory: Taxonomy and Empirical Analysis of Evaluation and System Limitations
+
+- **ID:** `2602.19320` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Dongming Jiang, Yi Li, Songtao Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.19320 · **PDF:** https://arxiv.org/pdf/2602.19320v1
+
+## 10593. Health+: Empowering Individuals via Unifying Health Data
+
+- **ID:** `2602.19319` · **primary:** `cs.MM` · **Submitted:** 2026-02-22
+- **Authors:** Sujaya Maiyya, Shantanu Sharma, Avinash Kumar
+- **abs:** https://arxiv.org/abs/2602.19319 · **PDF:** https://arxiv.org/pdf/2602.19319v1
+
+## 10594. Learning to Reason for Multi-Step Retrieval of Personal Context in Personalized Question Answering
+
+- **ID:** `2602.19317` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Maryam Amirizaniani, Alireza Salemi, Hamed Zamani
+- **abs:** https://arxiv.org/abs/2602.19317 · **PDF:** https://arxiv.org/pdf/2602.19317v1
+
+## 10595. Online Navigation Planning for Long-term Autonomous Operation of Underwater Gliders
+
+- **ID:** `2602.19315` · **primary:** `cs.RO` · **Submitted:** 2026-02-22
+- **Authors:** Victor-Alexandru Darvariu, Charlotte Z. Reed, Jan Stratmann, et al.
+- **abs:** https://arxiv.org/abs/2602.19315 · **PDF:** https://arxiv.org/pdf/2602.19315v1
+
+## 10596. IPv2: An Improved Image Purification Strategy for Real-World Ultra-Low-Dose Lung CT Denoising
+
+- **ID:** `2602.19314` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Guoliang Gong, Man Yu
+- **abs:** https://arxiv.org/abs/2602.19314 · **PDF:** https://arxiv.org/pdf/2602.19314v1
+
+## 10597. TOPReward: Token Probabilities as Hidden Zero-Shot Rewards for Robotics
+
+- **ID:** `2602.19313` · **primary:** `cs.RO` · **Submitted:** 2026-02-22
+- **Authors:** Shirui Chen, Cole Harrison, Ying-Chun Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.19313 · **PDF:** https://arxiv.org/pdf/2602.19313v1
+
+## 10598. Metasurfaces-Integrated Wireless Neural Networks for Lightweight Over-The-Air Edge Inference
+
+- **ID:** `2602.19312` · **primary:** `cs.ET` · **Submitted:** 2026-02-22
+- **Authors:** Kyriakos Stylianopoulos, Mario Edoardo Pandolfo, Paolo Di Lorenzo, et al.
+- **abs:** https://arxiv.org/abs/2602.19312 · **PDF:** https://arxiv.org/pdf/2602.19312v1
+
+## 10599. Analyzing Latency Hiding and Parallelism in an MLIR-based AI Kernel Compiler
+
+- **ID:** `2602.20204` · **primary:** `cs.PL` · **Submitted:** 2026-02-22
+- **Authors:** Javed Absar, Samarth Narang, Muthu Baskaran
+- **abs:** https://arxiv.org/abs/2602.20204 · **PDF:** https://arxiv.org/pdf/2602.20204v1
+
+## 10600. Safe and Interpretable Multimodal Path Planning for Multi-Agent Cooperation
+
+- **ID:** `2602.19304` · **primary:** `cs.RO` · **Submitted:** 2026-02-22
+- **Authors:** Haojun Shi, Suyu Ye, Katherine M. Guerrerio, et al.
+- **abs:** https://arxiv.org/abs/2602.19304 · **PDF:** https://arxiv.org/pdf/2602.19304v1
+
+## 10601. Prompt Injection as Role Confusion
+
+- **ID:** `2603.12277` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Charles Ye, Jasmine Cui, Dylan Hadfield-Menell
+- **abs:** https://arxiv.org/abs/2603.12277 · **PDF:** https://arxiv.org/pdf/2603.12277v2
+
+## 10602. ALPACA: A Reinforcement Learning Environment for Medication Repurposing and Treatment Optimization in Alzheimer's Disease
+
+- **ID:** `2602.19298` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Nolan Brady, Tom Yeh
+- **abs:** https://arxiv.org/abs/2602.19298 · **PDF:** https://arxiv.org/pdf/2602.19298v1
+
+## 10603. Evaluating the Reliability of Digital Forensic Evidence Discovered by Large Language Model: A Case Study
+
+- **ID:** `2602.20202` · **primary:** `cs.CR` · **Submitted:** 2026-02-22
+- **Authors:** Jeel Piyushkumar Khatiwala, Daniel Kwaku Ntiamoah Addai, Weifeng Xu
+- **abs:** https://arxiv.org/abs/2602.20202 · **PDF:** https://arxiv.org/pdf/2602.20202v1
+
+## 10604. Automated Generation of Microfluidic Netlists using Large Language Models
+
+- **ID:** `2602.19297` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Jasper Davidson, Skylar Stockham, Allen Boston, et al.
+- **abs:** https://arxiv.org/abs/2602.19297 · **PDF:** https://arxiv.org/pdf/2602.19297v2
+
+## 10605. AdsorbFlow: energy-conditioned flow matching enables fast and realistic adsorbate placement
+
+- **ID:** `2602.19289` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Jiangjie Qiu, Wentao Li, Honghao Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.19289 · **PDF:** https://arxiv.org/pdf/2602.19289v1
+
+## 10606. LWM-Temporal: Sparse Spatio-Temporal Attention for Wireless Channel Representation Learning
+
+- **ID:** `2603.10024` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Sadjad Alikhani, Akshay Malhotra, Shahab Hamidi-Rad, et al.
+- **abs:** https://arxiv.org/abs/2603.10024 · **PDF:** https://arxiv.org/pdf/2603.10024v1
+
+## 10607. Limited Reasoning Space: The cage of long-horizon reasoning in LLMs
+
+- **ID:** `2602.19281` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Zhenyu Li, Guanlin Wu, Cheems Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.19281 · **PDF:** https://arxiv.org/pdf/2602.19281v2
+
+## 10608. Taming Preconditioner Drift: Unlocking the Potential of Second-Order Optimizers for Federated Learning on Non-IID Data
+
+- **ID:** `2602.19271` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Junkang Liu, Fanhua Shang, Hongying Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.19271 · **PDF:** https://arxiv.org/pdf/2602.19271v1
+
+## 10609. CORVET: A CORDIC-Powered, Resource-Frugal Mixed-Precision Vector Processing Engine for High-Throughput AIoT applications
+
+- **ID:** `2602.19268` · **primary:** `cs.AR` · **Submitted:** 2026-02-22
+- **Authors:** Sonu Kumar, Mohd Faisal Khan, Mukul Lokhande, et al.
+- **abs:** https://arxiv.org/abs/2602.19268 · **PDF:** https://arxiv.org/pdf/2602.19268v1
+
+## 10610. Spectral bias in physics-informed and operator learning: Analysis and mitigation guidelines
+
+- **ID:** `2602.19265` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Siavash Khodakarami, Vivek Oommen, Nazanin Ahmadi Daryakenari, et al.
+- **abs:** https://arxiv.org/abs/2602.19265 · **PDF:** https://arxiv.org/pdf/2602.19265v1
+
+## 10611. Prognostics of Multisensor Systems with Unknown and Unlabeled Failure Modes via Bayesian Nonparametric Process Mixtures
+
+- **ID:** `2602.19263` · **primary:** `stat.AP` · **Submitted:** 2026-02-22
+- **Authors:** Kani Fu, Sanduni S Disanayaka Mudiyanselage, Chunli Dai, et al.
+- **abs:** https://arxiv.org/abs/2602.19263 · **PDF:** https://arxiv.org/pdf/2602.19263v1
+
+## 10612. DGPO: RL-Steered Graph Diffusion for Neural Architecture Generation
+
+- **ID:** `2602.19261` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Aleksei Liuliakov, Luca Hermes, Barbara Hammer
+- **abs:** https://arxiv.org/abs/2602.19261 · **PDF:** https://arxiv.org/pdf/2602.19261v2
+
+## 10613. Defining AI Models and AI Systems: A Framework to Resolve the Boundary Problem
+
+- **ID:** `2603.10023` · **primary:** `cs.CY` · **Submitted:** 2026-02-22
+- **Authors:** Yuanyuan Sun, Timothy Parker, Lara Gierschmann, et al.
+- **abs:** https://arxiv.org/abs/2603.10023 · **PDF:** https://arxiv.org/pdf/2603.10023v2
+
+## 10614. Alternating Bi-Objective Optimization for Explainable Neuro-Fuzzy Systems
+
+- **ID:** `2602.19253` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Qusai Khaled, Uzay Kaymak, Laura Genga
+- **abs:** https://arxiv.org/abs/2602.19253 · **PDF:** https://arxiv.org/pdf/2602.19253v1
+
+## 10615. No Need For Real Anomaly: MLLM Empowered Zero-Shot Video Anomaly Detection
+
+- **ID:** `2602.19248` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Zunkai Dai, Ke Li, Jiajia Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.19248 · **PDF:** https://arxiv.org/pdf/2602.19248v4
+
+## 10616. Robust Exploration in Directed Controller Synthesis via Reinforcement Learning with Soft Mixture-of-Experts
+
+- **ID:** `2602.19244` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Toshihide Ubukata, Zhiyao Wang, Enhong Mu, et al.
+- **abs:** https://arxiv.org/abs/2602.19244 · **PDF:** https://arxiv.org/pdf/2602.19244v1
+
+## 10617. Scaling Laws for Precision in High-Dimensional Linear Regression
+
+- **ID:** `2602.19241` · **primary:** `stat.ML` · **Submitted:** 2026-02-22
+- **Authors:** Dechen Zhang, Xuan Tang, Yingyu Liang, et al.
+- **abs:** https://arxiv.org/abs/2602.19241 · **PDF:** https://arxiv.org/pdf/2602.19241v2
+
+## 10618. Topology of Reasoning: Retrieved Cell Complex-Augmented Generation for Textual Graph Question Answering
+
+- **ID:** `2602.19240` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Sen Zhao, Lincheng Zhou, Yue Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.19240 · **PDF:** https://arxiv.org/pdf/2602.19240v1
+
+## 10619. Attention Deficits in Language Models: Causal Explanations for Procedural Hallucinations
+
+- **ID:** `2602.19239` · **primary:** `stat.ML` · **Submitted:** 2026-02-22
+- **Authors:** Ahmed Karim, Fatima Sheaib, Zein Khamis, et al.
+- **abs:** https://arxiv.org/abs/2602.19239 · **PDF:** https://arxiv.org/pdf/2602.19239v1
+
+## 10620. Global Prior Meets Local Consistency: Dual-Memory Augmented Vision-Language-Action Model for Efficient Robotic Manipulation
+
+- **ID:** `2602.20200` · **primary:** `cs.RO` · **Submitted:** 2026-02-22
+- **Authors:** Zaijing Li, Bing Hu, Rui Shao, et al.
+- **abs:** https://arxiv.org/abs/2602.20200 · **PDF:** https://arxiv.org/pdf/2602.20200v1
+
+## 10621. Evaluating SAP RPT-1 for Enterprise Business Process Prediction: In-Context Learning vs. Traditional Machine Learning on Structured SAP Data
+
+- **ID:** `2602.19237` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Amit Lal
+- **abs:** https://arxiv.org/abs/2602.19237 · **PDF:** https://arxiv.org/pdf/2602.19237v1
+
+## 10622. Proximity-Based Multi-Turn Optimization: Practical Credit Assignment for LLM Agent Training
+
+- **ID:** `2602.19225` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Yangyi Fang, Jiaye Lin, Xiaoliang Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.19225 · **PDF:** https://arxiv.org/pdf/2602.19225v1
+
+## 10623. Characterizing MARL for Energy Control: A Multi-KPI Benchmark on the CityLearn Environment
+
+- **ID:** `2602.19223` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Aymen Khouja, Imen Jendoubi, Oumayma Mahjoub, et al.
+- **abs:** https://arxiv.org/abs/2602.19223 · **PDF:** https://arxiv.org/pdf/2602.19223v2
+
+## 10624. Controlled Face Manipulation and Synthesis for Data Augmentation
+
+- **ID:** `2602.19219` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Joris Kirchner, Amogh Gudi, Marian Bittner, et al.
+- **abs:** https://arxiv.org/abs/2602.19219 · **PDF:** https://arxiv.org/pdf/2602.19219v1
+
+## 10625. Understanding Empirical Unlearning with Combinatorial Interpretability
+
+- **ID:** `2602.19215` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Shingo Kodama, Niv Cohen, Micah Adler, et al.
+- **abs:** https://arxiv.org/abs/2602.19215 · **PDF:** https://arxiv.org/pdf/2602.19215v1
+
+## 10626. Retrieval Augmented Enhanced Dual Co-Attention Framework for Target Aware Multimodal Bengali Hateful Meme Detection
+
+- **ID:** `2602.19212` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Raihan Tanvir, Md. Golam Rabiul Alam
+- **abs:** https://arxiv.org/abs/2602.19212 · **PDF:** https://arxiv.org/pdf/2602.19212v1
+
+## 10627. How to Allocate, How to Learn? Dynamic Rollout Allocation and Advantage Modulation for Policy Optimization
+
+- **ID:** `2602.19208` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Yangyi Fang, Jiaye Lin, Xiaoliang Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.19208 · **PDF:** https://arxiv.org/pdf/2602.19208v1
+
+## 10628. HybridFL: A Federated Learning Approach for Financial Crime Detection
+
+- **ID:** `2602.19207` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Afsana Khan, Marijn ten Thij, Guangzhi Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.19207 · **PDF:** https://arxiv.org/pdf/2602.19207v1
+
+## 10629. An Interpretable Data-Driven Model of the Flight Dynamics of Hawks
+
+- **ID:** `2602.19196` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-22
+- **Authors:** Lydia France, Karl Lapo, J. Nathan Kutz
+- **abs:** https://arxiv.org/abs/2602.19196 · **PDF:** https://arxiv.org/pdf/2602.19196v1
+
+## 10630. Visual Prompt Guided Unified Pushing Policy
+
+- **ID:** `2602.19193` · **primary:** `cs.RO` · **Submitted:** 2026-02-22
+- **Authors:** Hieu Bui, Ziyan Gao, Yuya Hosoda, et al.
+- **abs:** https://arxiv.org/abs/2602.19193 · **PDF:** https://arxiv.org/pdf/2602.19193v1
+
+## 10631. FUSAR-GPT : A Spatiotemporal Feature-Embedded and Two-Stage Decoupled Visual Language Model for SAR Imagery
+
+- **ID:** `2602.19190` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Xiaokun Zhang, Yi Yang, Ziqi Ye, et al.
+- **abs:** https://arxiv.org/abs/2602.19190 · **PDF:** https://arxiv.org/pdf/2602.19190v3
+
+## 10632. Adaptive Problem Generation via Symbolic Representations
+
+- **ID:** `2602.19187` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Teresa Yeo, Myeongho Jeon, Dulaj Weerakoon, et al.
+- **abs:** https://arxiv.org/abs/2602.19187 · **PDF:** https://arxiv.org/pdf/2602.19187v1
+
+## 10633. Prompts and Prayers: the Rise of GPTheology
+
+- **ID:** `2603.10019` · **primary:** `cs.CY` · **Submitted:** 2026-02-22
+- **Authors:** Ioana Cheres, Adrian Groza, Ioana Moldovan, et al.
+- **abs:** https://arxiv.org/abs/2603.10019 · **PDF:** https://arxiv.org/pdf/2603.10019v1
+
+## 10634. Next Reply Prediction X Dataset: Linguistic Discrepancies in Naively Generated Content
+
+- **ID:** `2602.19177` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Simon Münker, Nils Schwager, Kai Kugler, et al.
+- **abs:** https://arxiv.org/abs/2602.19177 · **PDF:** https://arxiv.org/pdf/2602.19177v1
+
+## 10635. TurkicNLP: An NLP Toolkit for Turkic Languages
+
+- **ID:** `2602.19174` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Sherzod Hakimov
+- **abs:** https://arxiv.org/abs/2602.19174 · **PDF:** https://arxiv.org/pdf/2602.19174v4
+
+## 10636. Online Realizable Regression and Applications for ReLU Networks
+
+- **ID:** `2602.19172` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Ilan Doron-Arad, Idan Mehalel, Elchanan Mossel
+- **abs:** https://arxiv.org/abs/2602.19172 · **PDF:** https://arxiv.org/pdf/2602.19172v1
+
+## 10637. CosyAccent: Duration-Controllable Accent Normalization Using Source-Synthesis Training Data
+
+- **ID:** `2602.19166` · **primary:** `eess.AS` · **Submitted:** 2026-02-22
+- **Authors:** Qibing Bai, Shuhao Shi, Shuai Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.19166 · **PDF:** https://arxiv.org/pdf/2602.19166v1
+
+## 10638. Reasoning Capabilities of Large Language Models. Lessons Learned from General Game Playing
+
+- **ID:** `2602.19160` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Maciej Świechowski, Adam Żychowski, Jacek Mańdziuk
+- **abs:** https://arxiv.org/abs/2602.19160 · **PDF:** https://arxiv.org/pdf/2602.19160v1
+
+## 10639. Beyond Behavioural Trade-Offs: Mechanistic Tracing of Pain-Pleasure Decisions in an LLM
+
+- **ID:** `2602.19159` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Francesca Bianco, Derek Shiller
+- **abs:** https://arxiv.org/abs/2602.19159 · **PDF:** https://arxiv.org/pdf/2602.19159v1
+
+## 10640. DoAtlas-1: A Causal Compilation Paradigm for Clinical AI
+
+- **ID:** `2602.19158` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Yulong Li, Jianxu Chen, Xiwei Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.19158 · **PDF:** https://arxiv.org/pdf/2602.19158v1
+
+## 10641. Facet-Level Persona Control by Trait-Activated Routing with Contrastive SAE for Role-Playing LLMs
+
+- **ID:** `2602.19157` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Wenqiu Tang, Zhen Wan, Takahiro Komamizu, et al.
+- **abs:** https://arxiv.org/abs/2602.19157 · **PDF:** https://arxiv.org/pdf/2602.19157v2
+
+## 10642. Artefact-Aware Fungal Detection in Dermatophytosis: A Real-Time Transformer-Based Approach for KOH Microscopy
+
+- **ID:** `2602.19156` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Rana Gursoy, Abdurrahim Yilmaz, Baris Kizilyaprak, et al.
+- **abs:** https://arxiv.org/abs/2602.19156 · **PDF:** https://arxiv.org/pdf/2602.19156v1
+
+## 10643. Constrained Diffusion for Accelerated Structure Relaxation of Inorganic Solids with Point Defects
+
+- **ID:** `2602.19153` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-22
+- **Authors:** Jingyi Cui, Jacob K. Christopher, Ankita Biswas, et al.
+- **abs:** https://arxiv.org/abs/2602.19153 · **PDF:** https://arxiv.org/pdf/2602.19153v1
+
+## 10644. ILION: Deterministic Pre-Execution Safety Gates for Agentic AI Systems
+
+- **ID:** `2603.13247` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Florin Adrian Chitan
+- **abs:** https://arxiv.org/abs/2603.13247 · **PDF:** https://arxiv.org/pdf/2603.13247v1
+
+## 10645. VIGiA: Instructional Video Guidance via Dialogue Reasoning and Retrieval
+
+- **ID:** `2602.19146` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Diogo Glória-Silva, David Semedo, João Maglhães
+- **abs:** https://arxiv.org/abs/2602.19146 · **PDF:** https://arxiv.org/pdf/2602.19146v2
+
+## 10646. Incremental Learning of Sparse Attention Patterns in Transformers
+
+- **ID:** `2602.19143` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Oğuz Kaan Yüksel, Rodrigo Alvarez Lucendo, Nicolas Flammarion
+- **abs:** https://arxiv.org/abs/2602.19143 · **PDF:** https://arxiv.org/pdf/2602.19143v1
+
+## 10647. Celo2: Towards Learned Optimization Free Lunch
+
+- **ID:** `2602.19142` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Abhinav Moudgil, Boris Knyazev, Eugene Belilovsky
+- **abs:** https://arxiv.org/abs/2602.19142 · **PDF:** https://arxiv.org/pdf/2602.19142v1
+
+## 10648. Sycophantic Chatbots Cause Delusional Spiraling, Even in Ideal Bayesians
+
+- **ID:** `2602.19141` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Kartik Chandra, Max Kleiman-Weiner, Jonathan Ragan-Kelley, et al.
+- **abs:** https://arxiv.org/abs/2602.19141 · **PDF:** https://arxiv.org/pdf/2602.19141v1
+
+## 10649. CaReFlow: Cyclic Adaptive Rectified Flow for Multimodal Fusion
+
+- **ID:** `2602.19140` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Sijie Mai, Shiqin Han
+- **abs:** https://arxiv.org/abs/2602.19140 · **PDF:** https://arxiv.org/pdf/2602.19140v1
+
+## 10650. CRCC: Contrast-Based Robust Cross-Subject and Cross-Site Representation Learning for EEG
+
+- **ID:** `2602.19138` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-22
+- **Authors:** Xiaobin Wong, Zhonghua Zhao, Haoran Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.19138 · **PDF:** https://arxiv.org/pdf/2602.19138v1
+
+## 10651. A Dataset for Named Entity Recognition and Relation Extraction from Art-historical Image Descriptions
+
+- **ID:** `2602.19133` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Stefanie Schneider, Miriam Göldl, Julian Stalter, et al.
+- **abs:** https://arxiv.org/abs/2602.19133 · **PDF:** https://arxiv.org/pdf/2602.19133v1
+
+## 10652. Test-Time Learning of Causal Structure from Interventional Data
+
+- **ID:** `2602.19131` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Wei Chen, Rui Ding, Bojun Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.19131 · **PDF:** https://arxiv.org/pdf/2602.19131v1
+
+## 10653. Detecting labeling bias using influence functions
+
+- **ID:** `2602.19130` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Frida Jørgensen, Nina Weng, Siavash Bigdeli
+- **abs:** https://arxiv.org/abs/2602.19130 · **PDF:** https://arxiv.org/pdf/2602.19130v1
+
+## 10654. K-Search: LLM Kernel Generation via Co-Evolving Intrinsic World Model
+
+- **ID:** `2602.19128` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Shiyi Cao, Ziming Mao, Joseph E. Gonzalez, et al.
+- **abs:** https://arxiv.org/abs/2602.19128 · **PDF:** https://arxiv.org/pdf/2602.19128v2
+
+## 10655. AgenticRAGTracer: A Hop-Aware Benchmark for Diagnosing Multi-Step Retrieval Reasoning in Agentic RAG
+
+- **ID:** `2602.19127` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Qijie You, Wenkai Yu, Wentao Zhang
+- **abs:** https://arxiv.org/abs/2602.19127 · **PDF:** https://arxiv.org/pdf/2602.19127v1
+
+## 10656. Robust Predictive Uncertainty and Double Descent in Contaminated Bayesian Random Features
+
+- **ID:** `2602.19126` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Michele Caprio, Katerina Papagiannouli, Siu Lun Chau, et al.
+- **abs:** https://arxiv.org/abs/2602.19126 · **PDF:** https://arxiv.org/pdf/2602.19126v1
+
+## 10657. VectorGym: A Multitask Benchmark for SVG Code Generation, Sketching, and Editing
+
+- **ID:** `2603.29852` · **primary:** `cs.GR` · **Submitted:** 2026-02-22
+- **Authors:** Juan Rodriguez, Haotian Zhang, Abhay Puri, et al.
+- **abs:** https://arxiv.org/abs/2603.29852 · **PDF:** https://arxiv.org/pdf/2603.29852v1
+
+## 10658. Event-Triggered Gossip for Distributed Learning
+
+- **ID:** `2602.19116` · **primary:** `eess.SP` · **Submitted:** 2026-02-22
+- **Authors:** Zhiyuan Zhai, Xiaojun Yuan, Wei Ni, et al.
+- **abs:** https://arxiv.org/abs/2602.19116 · **PDF:** https://arxiv.org/pdf/2602.19116v1
+
+## 10659. How Do LLMs Encode Scientific Quality? An Empirical Study Using Monosemantic Features from Sparse Autoencoders
+
+- **ID:** `2602.19115` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Michael McCoubrey, Angelo Salatino, Francesco Osborne, et al.
+- **abs:** https://arxiv.org/abs/2602.19115 · **PDF:** https://arxiv.org/pdf/2602.19115v1
+
+## 10660. Kaiwu-PyTorch-Plugin: Bridging Deep Learning and Photonic Quantum Computing for Energy-Based Models and Active Sample Selection
+
+- **ID:** `2602.19114` · **primary:** `quant-ph` · **Submitted:** 2026-02-22
+- **Authors:** Hongdong Zhu, Qi Gao, Yin Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.19114 · **PDF:** https://arxiv.org/pdf/2602.19114v1
+
+## 10661. Learning from Complexity: Exploring Dynamic Sample Pruning of Spatio-Temporal Training
+
+- **ID:** `2602.19113` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Wei Chen, Junle Chen, Yuqian Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.19113 · **PDF:** https://arxiv.org/pdf/2602.19113v2
+
+## 10662. Astra: Activation-Space Tail-Eigenvector Low-Rank Adaptation of Large Language Models
+
+- **ID:** `2602.19111` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Kainan Liu, Yong Zhang, Ning Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.19111 · **PDF:** https://arxiv.org/pdf/2602.19111v1
+
+## 10663. Post-Routing Arithmetic in Llama-3: Last-Token Result Writing and Rotation-Structured Digit Directions
+
+- **ID:** `2602.19109` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Yao Yan
+- **abs:** https://arxiv.org/abs/2602.19109 · **PDF:** https://arxiv.org/pdf/2602.19109v1
+
+## 10664. Value Entanglement: Conflation Between Different Kinds of Good In (Some) Large Language Models
+
+- **ID:** `2602.19101` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Seong Hah Cho, Junyi Li, Anna Leshinskaya
+- **abs:** https://arxiv.org/abs/2602.19101 · **PDF:** https://arxiv.org/pdf/2602.19101v1
+
+## 10665. DeliberationBench: A Normative Benchmark for the Influence of Large Language Models on Users' Views
+
+- **ID:** `2603.10018` · **primary:** `cs.CY` · **Submitted:** 2026-02-22
+- **Authors:** Luke Hewitt, Maximilian Kroner Dale, Paul de Font-Reaulx
+- **abs:** https://arxiv.org/abs/2603.10018 · **PDF:** https://arxiv.org/pdf/2603.10018v1
+
+## 10666. The Power of Decaying Steps: Enhancing Attack Stability and Transferability for Sign-based Optimizers
+
+- **ID:** `2602.19096` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Wei Tao, Yang Dai, Jincai Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.19096 · **PDF:** https://arxiv.org/pdf/2602.19096v2
+
+## 10667. RKHS Representation of Algebraic Convolutional Filters with Integral Operators
+
+- **ID:** `2602.19094` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Alejandro Parada-Mayorga, Alejandro Ribeiro, Juan Bazerque
+- **abs:** https://arxiv.org/abs/2602.19094 · **PDF:** https://arxiv.org/pdf/2602.19094v1
+
+## 10668. IMOVNO+: A Regional Partitioning and Meta-Heuristic Ensemble Framework for Imbalanced Multi-Class Learning
+
+- **ID:** `2602.20199` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Soufiane Bacha, Laouni Djafri, Sahraoui Dhelim, et al.
+- **abs:** https://arxiv.org/abs/2602.20199 · **PDF:** https://arxiv.org/pdf/2602.20199v1
+
+## 10669. KEMP-PIP: A Feature-Fusion Based Approach for Pro-inflammatory Peptide Prediction
+
+- **ID:** `2602.20198` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-22
+- **Authors:** Soumik Deb Niloy, Md. Fahmid-Ul-Alam Juboraj, Swakkhar Shatabda
+- **abs:** https://arxiv.org/abs/2602.20198 · **PDF:** https://arxiv.org/pdf/2602.20198v1
+
+## 10670. Ani3DHuman: Photorealistic 3D Human Animation with Self-guided Stochastic Sampling
+
+- **ID:** `2602.19089` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Qi Sun, Can Wang, Jiaxiang Shang, et al.
+- **abs:** https://arxiv.org/abs/2602.19089 · **PDF:** https://arxiv.org/pdf/2602.19089v1
+
+## 10671. Detecting Cybersecurity Threats by Integrating Explainable AI with SHAP Interpretability and Strategic Data Sampling
+
+- **ID:** `2602.19087` · **primary:** `cs.CR` · **Submitted:** 2026-02-22
+- **Authors:** Norrakith Srisumrith, Sunantha Sodsee
+- **abs:** https://arxiv.org/abs/2602.19087 · **PDF:** https://arxiv.org/pdf/2602.19087v1
+
+## 10672. INTACT: Intent-Aware Representation Learning for Cryptographic Traffic Violation Detection
+
+- **ID:** `2602.21252` · **primary:** `cs.CR` · **Submitted:** 2026-02-22
+- **Authors:** Rahul D Ray
+- **abs:** https://arxiv.org/abs/2602.21252 · **PDF:** https://arxiv.org/pdf/2602.21252v1
+
+## 10673. TriTopic: Tri-Modal Graph-Based Topic Modeling with Iterative Refinement and Archetypes
+
+- **ID:** `2602.19079` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Roman Egger
+- **abs:** https://arxiv.org/abs/2602.19079 · **PDF:** https://arxiv.org/pdf/2602.19079v1
+
+## 10674. Controllable Exploration in Hybrid-Policy RLVR for Multi-Modal Reasoning
+
+- **ID:** `2602.20197` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Zhuoxu Huang, Mengxi Jia, Hao Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.20197 · **PDF:** https://arxiv.org/pdf/2602.20197v2
+
+## 10675. Defining Explainable AI for Requirements Analysis
+
+- **ID:** `2602.19071` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Raymond Sheh, Isaac Monteath
+- **abs:** https://arxiv.org/abs/2602.19071 · **PDF:** https://arxiv.org/pdf/2602.19071v1
+
+## 10676. Asking the Right Questions: Improving Reasoning with Generated Stepping Stones
+
+- **ID:** `2602.19069` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Hengyuan Hu, Tingchen Fu, Minqi Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.19069 · **PDF:** https://arxiv.org/pdf/2602.19069v1
+
+## 10677. TimeRadar: A Domain-Rotatable Foundation Model for Time Series Anomaly Detection
+
+- **ID:** `2602.19068` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Hui He, Hezhe Qiao, Yutong Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.19068 · **PDF:** https://arxiv.org/pdf/2602.19068v1
+
+## 10678. IDLM: Inverse-distilled Diffusion Language Models
+
+- **ID:** `2602.19066` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** David Li, Nikita Gushchin, Dmitry Abulkhanov, et al.
+- **abs:** https://arxiv.org/abs/2602.19066 · **PDF:** https://arxiv.org/pdf/2602.19066v1
+
+## 10679. Agentic Problem Frames: A Systematic Approach to Engineering Reliable Domain Agents
+
+- **ID:** `2602.19065` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Chanjin Park
+- **abs:** https://arxiv.org/abs/2602.19065 · **PDF:** https://arxiv.org/pdf/2602.19065v1
+
+## 10680. Grouter: Decoupling Routing from Representation for Accelerated MoE Training
+
+- **ID:** `2603.06626` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Yuqi Xu, Rizhen Hu, Zihan Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.06626 · **PDF:** https://arxiv.org/pdf/2603.06626v1
+
+## 10681. Do LLMs and VLMs Share Neurons for Inference? Evidence and Mechanisms of Cross-Modal Transfer
+
+- **ID:** `2602.19058` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Chenhang Cui, An Zhang, Yuxin Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.19058 · **PDF:** https://arxiv.org/pdf/2602.19058v1
+
+## 10682. IAPO: Information-Aware Policy Optimization for Token-Efficient Reasoning
+
+- **ID:** `2602.19049` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Yinhan He, Yaochen Zhu, Mingjia Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.19049 · **PDF:** https://arxiv.org/pdf/2602.19049v1
+
+## 10683. OpenPort Protocol: A Security Governance Specification for AI Agent Tool Access
+
+- **ID:** `2602.20196` · **primary:** `cs.CR` · **Submitted:** 2026-02-22
+- **Authors:** Genliang Zhu, Chu Wang, Ziyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.20196 · **PDF:** https://arxiv.org/pdf/2602.20196v1
+
+## 10684. Uncovering Context Reliance in Unstructured Knowledge Editing
+
+- **ID:** `2602.19043` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Zisheng Zhou, Mengqi Zhang, Shiguang Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.19043 · **PDF:** https://arxiv.org/pdf/2602.19043v1
+
+## 10685. Back to Blackwell: Closing the Loop on Intransitivity in Multi-Objective Preference Fine-Tuning
+
+- **ID:** `2602.19041` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Jiahao Zhang, Lujing Zhang, Keltin Grimes, et al.
+- **abs:** https://arxiv.org/abs/2602.19041 · **PDF:** https://arxiv.org/pdf/2602.19041v1
+
+## 10686. Unsupervised Denoising of Diffusion-Weighted Images with Bias and Variance Corrected Noise Modeling
+
+- **ID:** `2602.22235` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-22
+- **Authors:** Jine Xie, Zhicheng Zhang, Yunwei Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.22235 · **PDF:** https://arxiv.org/pdf/2602.22235v1
+
+## 10687. A Markovian View of Iterative-Feedback Loops in Image Generative Models: Neural Resonance and Model Collapse
+
+- **ID:** `2602.19033` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Vibhas Kumar Vats, David J. Crandall, Samuel Goree
+- **abs:** https://arxiv.org/abs/2602.19033 · **PDF:** https://arxiv.org/pdf/2602.19033v1
+
+## 10688. OrgFlow: Generative Modeling of Organic Crystal Structures from Molecular Graphs
+
+- **ID:** `2602.20195` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-22
+- **Authors:** Mohammadmahdi Vahediahmar, Matthew A. McDonald, Feng Liu
+- **abs:** https://arxiv.org/abs/2602.20195 · **PDF:** https://arxiv.org/pdf/2602.20195v1
+
+## 10689. Pushing the Limits of Inverse Lithography with Generative Reinforcement Learning
+
+- **ID:** `2602.19027` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Haoyu Yang, Haoxing Ren
+- **abs:** https://arxiv.org/abs/2602.19027 · **PDF:** https://arxiv.org/pdf/2602.19027v1
+
+## 10690. Routing-Aware Explanations for Mixture of Experts Graph Models in Malware Detection
+
+- **ID:** `2602.19025` · **primary:** `cs.CR` · **Submitted:** 2026-02-22
+- **Authors:** Hossein Shokouhinejad, Roozbeh Razavi-Far, Griffin Higgins, et al.
+- **abs:** https://arxiv.org/abs/2602.19025 · **PDF:** https://arxiv.org/pdf/2602.19025v1
+
+## 10691. FedAvg-Based CTMC Hazard Model for Federated Bridge Deterioration Assessment
+
+- **ID:** `2602.20194` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Takato Yasuno
+- **abs:** https://arxiv.org/abs/2602.20194 · **PDF:** https://arxiv.org/pdf/2602.20194v1
+
+## 10692. An interpretable framework using foundation models for fish sex identification
+
+- **ID:** `2602.19022` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Zheng Miao, Tien-Chieh Hung
+- **abs:** https://arxiv.org/abs/2602.19022 · **PDF:** https://arxiv.org/pdf/2602.19022v1
+
+## 10693. Learning to Detect Language Model Training Data via Active Reconstruction
+
+- **ID:** `2602.19020` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Junjie Oscar Yin, John X. Morris, Vitaly Shmatikov, et al.
+- **abs:** https://arxiv.org/abs/2602.19020 · **PDF:** https://arxiv.org/pdf/2602.19020v1
+
+## 10694. Why ReLU? A Bit-Model Dichotomy for Deep Network Training
+
+- **ID:** `2602.19017` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Ilan Doron-Arad, Elchanan Mossel
+- **abs:** https://arxiv.org/abs/2602.19017 · **PDF:** https://arxiv.org/pdf/2602.19017v1
+
+## 10695. Capable but Unreliable: Canonical Path Deviation as a Causal Mechanism of Agent Failure in Long-Horizon Tasks
+
+- **ID:** `2602.19008` · **primary:** `cs.CL` · **Submitted:** 2026-02-22
+- **Authors:** Wilson Y. Lee
+- **abs:** https://arxiv.org/abs/2602.19008 · **PDF:** https://arxiv.org/pdf/2602.19008v1
+
+## 10696. GUIDE-US: Grade-Informed Unpaired Distillation of Encoder Knowledge from Histopathology to Micro-UltraSound
+
+- **ID:** `2602.19005` · **primary:** `cs.CV` · **Submitted:** 2026-02-22
+- **Authors:** Emma Willis, Tarek Elghareb, Paul F. R. Wilson, et al.
+- **abs:** https://arxiv.org/abs/2602.19005 · **PDF:** https://arxiv.org/pdf/2602.19005v1
+
+## 10697. MagicAgent: Towards Generalized Agent Planning
+
+- **ID:** `2602.19000` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Xuhui Ren, Shaokang Dong, Chen Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.19000 · **PDF:** https://arxiv.org/pdf/2602.19000v2
+
+## 10698. No More DeLuLu: Physics-Inspired Kernel Networks for Geometrically-Grounded Neural Computation
+
+- **ID:** `2603.12276` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Taha Bouhsine
+- **abs:** https://arxiv.org/abs/2603.12276 · **PDF:** https://arxiv.org/pdf/2603.12276v1
+
+## 10699. Benchmark Test-Time Scaling of General LLM Agents
+
+- **ID:** `2602.18998` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Xiaochuan Li, Ryan Ming, Pranav Setlur, et al.
+- **abs:** https://arxiv.org/abs/2602.18998 · **PDF:** https://arxiv.org/pdf/2602.18998v1
+
+## 10700. Implicit Bias and Convergence of Matrix Stochastic Mirror Descent
+
+- **ID:** `2602.18997` · **primary:** `stat.ML` · **Submitted:** 2026-02-22
+- **Authors:** Danil Akhtiamov, Reza Ghane, Omead Pooladzandi, et al.
+- **abs:** https://arxiv.org/abs/2602.18997 · **PDF:** https://arxiv.org/pdf/2602.18997v2
+
+## 10701. Pavement Missing Condition Data Imputation through Collective Learning-Based Graph Neural Networks
+
+- **ID:** `2603.06625` · **primary:** `cs.LG` · **Submitted:** 2026-02-22
+- **Authors:** Ke Yu, Lu Gao
+- **abs:** https://arxiv.org/abs/2603.06625 · **PDF:** https://arxiv.org/pdf/2603.06625v1
+
+## 10702. All Constant Mutation Rates for the $(1+1)$ Evolutionary Algorithm
+
+- **ID:** `2602.18989` · **primary:** `cs.NE` · **Submitted:** 2026-02-22
+- **Authors:** Andrew James Kelley
+- **abs:** https://arxiv.org/abs/2602.18989 · **PDF:** https://arxiv.org/pdf/2602.18989v1
+
+## 10703. Quantifying Automation Risk in High-Automation AI Systems: A Bayesian Framework for Failure Propagation and Optimal Oversight
+
+- **ID:** `2602.18986` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Vishal Srivastava, Tanmay Sah
+- **abs:** https://arxiv.org/abs/2602.18986 · **PDF:** https://arxiv.org/pdf/2602.18986v1
+
+## 10704. AMV-L: Lifecycle-Managed Agent Memory for Tail-Latency Control in Long-Running LLM Systems
+
+- **ID:** `2603.04443` · **primary:** `cs.DC` · **Submitted:** 2026-02-22
+- **Authors:** Emmanuel Bamidele
+- **abs:** https://arxiv.org/abs/2603.04443 · **PDF:** https://arxiv.org/pdf/2603.04443v1
+
+## 10705. InfEngine: A Self-Verifying and Self-Optimizing Intelligent Engine for Infrared Radiation Computing
+
+- **ID:** `2602.18985` · **primary:** `cs.AI` · **Submitted:** 2026-02-22
+- **Authors:** Kun Ding, Jian Xu, Ying Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.18985 · **PDF:** https://arxiv.org/pdf/2602.18985v1
+
+## 10706. When Backdoors Go Beyond Triggers: Semantic Drift in Diffusion Models Under Encoder Attacks
+
+- **ID:** `2602.20193` · **primary:** `cs.CR` · **Submitted:** 2026-02-21
+- **Authors:** Shenyang Chen, Liuwan Zhu
+- **abs:** https://arxiv.org/abs/2602.20193 · **PDF:** https://arxiv.org/pdf/2602.20193v1
+
+## 10707. Conditionally Site-Independent Neural Evolution of Antibody Sequences
+
+- **ID:** `2602.18982` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Stephen Zhewen Lu, Aakarsh Vermani, Kohei Sanno, et al.
+- **abs:** https://arxiv.org/abs/2602.18982 · **PDF:** https://arxiv.org/pdf/2602.18982v3
+
+## 10708. How Far Can We Go with Pixels Alone? A Pilot Study on Screen-Only Navigation in Commercial 3D ARPGs
+
+- **ID:** `2602.18981` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Kaijie Xu, Mustafa Bugti, Clark Verbrugge
+- **abs:** https://arxiv.org/abs/2602.18981 · **PDF:** https://arxiv.org/pdf/2602.18981v1
+
+## 10709. When Do LLM Preferences Predict Downstream Behavior?
+
+- **ID:** `2602.18971` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Katarina Slama, Alexandra Souly, Dishank Bansal, et al.
+- **abs:** https://arxiv.org/abs/2602.18971 · **PDF:** https://arxiv.org/pdf/2602.18971v1
+
+## 10710. Robust and Efficient Tool Orchestration via Layered Execution Structures with Reflective Correction
+
+- **ID:** `2602.18968` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Tao Zhe, Haoyu Wang, Bo Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.18968 · **PDF:** https://arxiv.org/pdf/2602.18968v2
+
+## 10711. Whisper: Courtside Edition Enhancing ASR Performance Through LLM-Driven Context Generation
+
+- **ID:** `2602.18966` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Yonathan Ron, Shiri Gilboa, Tammuz Dubnov
+- **abs:** https://arxiv.org/abs/2602.18966 · **PDF:** https://arxiv.org/pdf/2602.18966v1
+
+## 10712. Yor-Sarc: A gold-standard dataset for sarcasm detection in a low-resource African language
+
+- **ID:** `2602.18964` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Toheeb Aduramomi Jimoh, Tabea De Wille, Nikola S. Nikolov
+- **abs:** https://arxiv.org/abs/2602.18964 · **PDF:** https://arxiv.org/pdf/2602.18964v1
+
+## 10713. NeuroWise: A Multi-Agent LLM "Glass-Box" System for Practicing Double-Empathy Communication with Autistic Partners
+
+- **ID:** `2602.18962` · **primary:** `cs.HC` · **Submitted:** 2026-02-21
+- **Authors:** Albert Tang, Yifan Mo, Jie Li, et al.
+- **abs:** https://arxiv.org/abs/2602.18962 · **PDF:** https://arxiv.org/pdf/2602.18962v2
+
+## 10714. Modularity is the Bedrock of Natural and Artificial Intelligence
+
+- **ID:** `2602.18960` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Alessandro Salatiello
+- **abs:** https://arxiv.org/abs/2602.18960 · **PDF:** https://arxiv.org/pdf/2602.18960v1
+
+## 10715. Exploration Space Theory: Formal Foundations for Prerequisite-Aware Location-Based Recommendation
+
+- **ID:** `2603.06624` · **primary:** `cs.IR` · **Submitted:** 2026-02-21
+- **Authors:** Madjid Sadallah
+- **abs:** https://arxiv.org/abs/2603.06624 · **PDF:** https://arxiv.org/pdf/2603.06624v1
+
+## 10716. Optimal and Structure-Adaptive CATE Estimation with Kernel Ridge Regression
+
+- **ID:** `2602.18958` · **primary:** `stat.ME` · **Submitted:** 2026-02-21
+- **Authors:** Seok-Jin Kim
+- **abs:** https://arxiv.org/abs/2602.18958 · **PDF:** https://arxiv.org/pdf/2602.18958v2
+
+## 10717. INDUCTION: Finite-Structure Concept Synthesis in First-Order Logic
+
+- **ID:** `2602.18956` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Serafim Batzoglou
+- **abs:** https://arxiv.org/abs/2602.18956 · **PDF:** https://arxiv.org/pdf/2602.18956v2
+
+## 10718. MoBiQuant: Mixture-of-Bits Quantization for Token-Adaptive Elastic LLMs
+
+- **ID:** `2602.20191` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Dongwei Wang, Jinhee Kim, Seokho Han, et al.
+- **abs:** https://arxiv.org/abs/2602.20191 · **PDF:** https://arxiv.org/pdf/2602.20191v1
+
+## 10719. Incremental Transformer Neural Processes
+
+- **ID:** `2602.18955` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Philip Mortimer, Cristiana Diaconu, Tommy Rochussen, et al.
+- **abs:** https://arxiv.org/abs/2602.18955 · **PDF:** https://arxiv.org/pdf/2602.18955v1
+
+## 10720. Toward Manifest Relationality in Transformers via Symmetry Reduction
+
+- **ID:** `2602.18948` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** J. François, L. Ravera
+- **abs:** https://arxiv.org/abs/2602.18948 · **PDF:** https://arxiv.org/pdf/2602.18948v1
+
+## 10721. (Perlin) Noise as AI coordinator
+
+- **ID:** `2602.18947` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Kaijie Xu, Clark Verbrugge
+- **abs:** https://arxiv.org/abs/2602.18947 · **PDF:** https://arxiv.org/pdf/2602.18947v1
+
+## 10722. Exponential Convergence of (Stochastic) Gradient Descent for Separable Logistic Regression
+
+- **ID:** `2602.18946` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Sacchit Kale, Piyushi Manupriya, Pierre Marion, et al.
+- **abs:** https://arxiv.org/abs/2602.18946 · **PDF:** https://arxiv.org/pdf/2602.18946v2
+
+## 10723. High Dimensional Procedural Content Generation
+
+- **ID:** `2602.18943` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Kaijie Xu, Clark Verbrugge
+- **abs:** https://arxiv.org/abs/2602.18943 · **PDF:** https://arxiv.org/pdf/2602.18943v1
+
+## 10724. DREAM: Deep Research Evaluation with Agentic Metrics
+
+- **ID:** `2602.18940` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Elad Ben Avraham, Changhao Li, Ron Dorfman, et al.
+- **abs:** https://arxiv.org/abs/2602.18940 · **PDF:** https://arxiv.org/pdf/2602.18940v1
+
+## 10725. LoMime: Query-Efficient Membership Inference using Model Extraction in Label-Only Settings
+
+- **ID:** `2602.18934` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Abdullah Caglar Oksuz, Anisa Halimi, Erman Ayday
+- **abs:** https://arxiv.org/abs/2602.18934 · **PDF:** https://arxiv.org/pdf/2602.18934v1
+
+## 10726. Give Users the Wheel: Towards Promptable Recommendation Paradigm
+
+- **ID:** `2602.18929` · **primary:** `cs.IR` · **Submitted:** 2026-02-21
+- **Authors:** Fuyuan Lyu, Chenglin Luo, Qiyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.18929 · **PDF:** https://arxiv.org/pdf/2602.18929v2
+
+## 10727. Why Agent Caching Fails and How to Fix It: Structured Intent Canonicalization with Few-Shot Learning
+
+- **ID:** `2602.18922` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Abhinaba Basu
+- **abs:** https://arxiv.org/abs/2602.18922 · **PDF:** https://arxiv.org/pdf/2602.18922v2
+
+## 10728. DeepInnovator: Triggering the Innovative Capabilities of LLMs
+
+- **ID:** `2602.18920` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Tianyu Fan, Fengji Zhang, Yuxiang Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.18920 · **PDF:** https://arxiv.org/pdf/2602.18920v1
+
+## 10729. AgenticTyper: Automated Typing of Legacy Software Projects Using Agentic AI
+
+- **ID:** `2602.21251` · **primary:** `cs.SE` · **Submitted:** 2026-02-21
+- **Authors:** Clemens Pohle
+- **abs:** https://arxiv.org/abs/2602.21251 · **PDF:** https://arxiv.org/pdf/2602.21251v1
+
+## 10730. Early Evidence of Vibe-Proving with Consumer LLMs: A Case Study on Spectral Region Characterization with ChatGPT-5.2 (Thinking)
+
+- **ID:** `2602.18918` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Brecht Verbeken, Brando Vagenende, Marie-Anne Guerry, et al.
+- **abs:** https://arxiv.org/abs/2602.18918 · **PDF:** https://arxiv.org/pdf/2602.18918v1
+
+## 10731. Adaptive Collaboration of Arena-Based Argumentative LLMs for Explainable and Contestable Legal Reasoning
+
+- **ID:** `2602.18916` · **primary:** `cs.MA` · **Submitted:** 2026-02-21
+- **Authors:** Hoang-Loc Cao, Phuc Ho, Truong Thanh Hung Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.18916 · **PDF:** https://arxiv.org/pdf/2602.18916v1
+
+## 10732. AAVGen: Precision Engineering of Adeno-associated Viral Capsids for Renal Selective Targeting
+
+- **ID:** `2602.18915` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-21
+- **Authors:** Mohammadreza Ghaffarzadeh-Esfahani, Yousof Gheisari
+- **abs:** https://arxiv.org/abs/2602.18915 · **PDF:** https://arxiv.org/pdf/2602.18915v1
+
+## 10733. From Human-Level AI Tales to AI Leveling Human Scales
+
+- **ID:** `2602.18911` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Peter Romero, Fernando Martínez-Plumed, Zachary R. Tyler, et al.
+- **abs:** https://arxiv.org/abs/2602.18911 · **PDF:** https://arxiv.org/pdf/2602.18911v1
+
+## 10734. SLDP: Semi-Local Differential Privacy for Density-Adaptive Analytics
+
+- **ID:** `2602.18910` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Alexey Kroshnin, Alexandra Suvorikova
+- **abs:** https://arxiv.org/abs/2602.18910 · **PDF:** https://arxiv.org/pdf/2602.18910v1
+
+## 10735. Advances in GRPO for Generation Models: A Survey
+
+- **ID:** `2603.06623` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Zexiang Liu, Xianglong He, Yangguang Li
+- **abs:** https://arxiv.org/abs/2603.06623 · **PDF:** https://arxiv.org/pdf/2603.06623v1
+
+## 10736. DeepInterestGR: Mining Deep Multi-Interest Using Multi-Modal LLMs for Generative Recommendation
+
+- **ID:** `2602.18907` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Yangchen Zeng
+- **abs:** https://arxiv.org/abs/2602.18907 · **PDF:** https://arxiv.org/pdf/2602.18907v1
+
+## 10737. TRUE: A Trustworthy Unified Explanation Framework for Large Language Model Reasoning
+
+- **ID:** `2602.18905` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Yujiao Yang
+- **abs:** https://arxiv.org/abs/2602.18905 · **PDF:** https://arxiv.org/pdf/2602.18905v1
+
+## 10738. PCA-VAE: Differentiable Subspace Quantization without Codebook Collapse
+
+- **ID:** `2602.18904` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Hao Lu, Onur C. Koyun, Yongxin Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.18904 · **PDF:** https://arxiv.org/pdf/2602.18904v1
+
+## 10739. [b]=[d]-[t]+[p]: Self-supervised Speech Models Discover Phonological Vector Arithmetic
+
+- **ID:** `2602.18899` · **primary:** `eess.AS` · **Submitted:** 2026-02-21
+- **Authors:** Kwanghee Choi, Eunjung Yeo, Cheol Jun Cho, et al.
+- **abs:** https://arxiv.org/abs/2602.18899 · **PDF:** https://arxiv.org/pdf/2602.18899v2
+
+## 10740. HEHRGNN: A Unified Embedding Model for Knowledge Graphs with Hyperedges and Hyper-Relational Edges
+
+- **ID:** `2602.18897` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Rajesh Rajagopalamenon, Unnikrishnan Cheramangalath
+- **abs:** https://arxiv.org/abs/2602.18897 · **PDF:** https://arxiv.org/pdf/2602.18897v1
+
+## 10741. Could Large Language Models work as Post-hoc Explainability Tools in Credit Risk Models?
+
+- **ID:** `2602.18895` · **primary:** `q-fin.RM` · **Submitted:** 2026-02-21
+- **Authors:** Wenxi Geng, Dingyuan Liu, Liya Li, et al.
+- **abs:** https://arxiv.org/abs/2602.18895 · **PDF:** https://arxiv.org/pdf/2602.18895v1
+
+## 10742. Orchestrating LLM Agents for Scientific Research: A Pilot Study of Multiple Choice Question (MCQ) Generation and Evaluation
+
+- **ID:** `2602.18891` · **primary:** `cs.CY` · **Submitted:** 2026-02-21
+- **Authors:** Yuan An
+- **abs:** https://arxiv.org/abs/2602.18891 · **PDF:** https://arxiv.org/pdf/2602.18891v1
+
+## 10743. TPRU: Advancing Temporal and Procedural Understanding in Large Multimodal Models
+
+- **ID:** `2602.18884` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Zhenkun Gao, Xuhong Wang, Xin Tan, et al.
+- **abs:** https://arxiv.org/abs/2602.18884 · **PDF:** https://arxiv.org/pdf/2602.18884v1
+
+## 10744. SceneTok: A Compressed, Diffusable Token Space for 3D Scenes
+
+- **ID:** `2602.18882` · **primary:** `cs.CV` · **Submitted:** 2026-02-21
+- **Authors:** Mohammad Asim, Christopher Wewer, Jan Eric Lenssen
+- **abs:** https://arxiv.org/abs/2602.18882 · **PDF:** https://arxiv.org/pdf/2602.18882v2
+
+## 10745. FOCA: Frequency-Oriented Cross-Domain Forgery Detection, Localization and Explanation via Multi-Modal Large Language Model
+
+- **ID:** `2602.18880` · **primary:** `cs.CV` · **Submitted:** 2026-02-21
+- **Authors:** Zhou Liu, Tonghua Su, Hongshi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.18880 · **PDF:** https://arxiv.org/pdf/2602.18880v1
+
+## 10746. Structure-Level Disentangled Diffusion for Few-Shot Chinese Font Generation
+
+- **ID:** `2602.18874` · **primary:** `cs.CV` · **Submitted:** 2026-02-21
+- **Authors:** Jie Li, Suorong Yang, Jian Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.18874 · **PDF:** https://arxiv.org/pdf/2602.18874v1
+
+## 10747. BiMotion: B-spline Motion for Text-guided Dynamic 3D Character Generation
+
+- **ID:** `2602.18873` · **primary:** `cs.CV` · **Submitted:** 2026-02-21
+- **Authors:** Miaowei Wang, Qingxuan Yan, Zhi Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.18873 · **PDF:** https://arxiv.org/pdf/2602.18873v2
+
+## 10748. Federated Measurement of Demographic Disparities from Quantile Sketches
+
+- **ID:** `2602.18870` · **primary:** `stat.ML` · **Submitted:** 2026-02-21
+- **Authors:** Arthur Charpentier, Agathe Fernandes Machado, Olivier Côté, et al.
+- **abs:** https://arxiv.org/abs/2602.18870 · **PDF:** https://arxiv.org/pdf/2602.18870v1
+
+## 10749. Limits of Convergence-Rate Control for Open-Weight Safety
+
+- **ID:** `2602.18868` · **primary:** `math.OC` · **Submitted:** 2026-02-21
+- **Authors:** Domenic Rosati, Xijie Zeng, Hong Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.18868 · **PDF:** https://arxiv.org/pdf/2602.18868v1
+
+## 10750. Boosting for Vector-Valued Prediction and Conditional Density Estimation
+
+- **ID:** `2602.18866` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Jian Qian, Shu Ge
+- **abs:** https://arxiv.org/abs/2602.18866 · **PDF:** https://arxiv.org/pdf/2602.18866v1
+
+## 10751. TIACam: Text-Anchored Invariant Feature Learning with Auto-Augmentation for Camera-Robust Zero-Watermarking
+
+- **ID:** `2602.18863` · **primary:** `eess.IV` · **Submitted:** 2026-02-21
+- **Authors:** Abdullah All Tanvir, Agnibh Dasgupta, Xin Zhong
+- **abs:** https://arxiv.org/abs/2602.18863 · **PDF:** https://arxiv.org/pdf/2602.18863v1
+
+## 10752. Hyperbolic Busemann Neural Networks
+
+- **ID:** `2602.18858` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Ziheng Chen, Bernhard Schölkopf, Nicu Sebe
+- **abs:** https://arxiv.org/abs/2602.18858 · **PDF:** https://arxiv.org/pdf/2602.18858v2
+
+## 10753. VariBASed: Variational Bayes-Adaptive Sequential Monte-Carlo Planning for Deep Reinforcement Learning
+
+- **ID:** `2602.18857` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Joery A. de Vries, Jinke He, Yaniv Oren, et al.
+- **abs:** https://arxiv.org/abs/2602.18857 · **PDF:** https://arxiv.org/pdf/2602.18857v1
+
+## 10754. Issues with Measuring Task Complexity via Random Policies in Robotic Tasks
+
+- **ID:** `2602.18856` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Reabetswe M. Nkhumise, Mohamed S. Talamali, Aditya Gilra
+- **abs:** https://arxiv.org/abs/2602.18856 · **PDF:** https://arxiv.org/pdf/2602.18856v1
+
+## 10755. Rank-Aware Spectral Bounds on Attention Logits for Stable Low-Precision Training
+
+- **ID:** `2602.18851` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Seyed Morteza Emadi
+- **abs:** https://arxiv.org/abs/2602.18851 · **PDF:** https://arxiv.org/pdf/2602.18851v1
+
+## 10756. When the Inference Meets the Explicitness or Why Multimodality Can Make Us Forget About the Perfect Predictor
+
+- **ID:** `2602.18850` · **primary:** `cs.RO` · **Submitted:** 2026-02-21
+- **Authors:** J. E. Domínguez-Vidal, Alberto Sanfeliu
+- **abs:** https://arxiv.org/abs/2602.18850 · **PDF:** https://arxiv.org/pdf/2602.18850v1
+
+## 10757. Exact Attention Sensitivity and the Geometry of Transformer Stability
+
+- **ID:** `2602.18849` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Seyed Morteza Emadi
+- **abs:** https://arxiv.org/abs/2602.18849 · **PDF:** https://arxiv.org/pdf/2602.18849v1
+
+## 10758. DUET-VLM: Dual stage Unified Efficient Token reduction for VLM Training and Inference
+
+- **ID:** `2602.18846` · **primary:** `cs.CV` · **Submitted:** 2026-02-21
+- **Authors:** Aditya Kumar Singh, Hitesh Kandala, Pratik Prabhanjan Brahma, et al.
+- **abs:** https://arxiv.org/abs/2602.18846 · **PDF:** https://arxiv.org/pdf/2602.18846v2
+
+## 10759. When Agda met Vampire
+
+- **ID:** `2602.18844` · **primary:** `cs.LO` · **Submitted:** 2026-02-21
+- **Authors:** Artjoms Šinkarovs, Michael Rawson
+- **abs:** https://arxiv.org/abs/2602.18844 · **PDF:** https://arxiv.org/pdf/2602.18844v1
+
+## 10760. ABD: Default Exception Abduction in Finite First Order Worlds
+
+- **ID:** `2602.18843` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Serafim Batzoglou
+- **abs:** https://arxiv.org/abs/2602.18843 · **PDF:** https://arxiv.org/pdf/2602.18843v2
+
+## 10761. L2G-Net: Local to Global Spectral Graph Neural Networks via Cauchy Factorizations
+
+- **ID:** `2602.18837` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Samuel Fernández-Menduiña, Eduardo Pavez, Antonio Ortega
+- **abs:** https://arxiv.org/abs/2602.18837 · **PDF:** https://arxiv.org/pdf/2602.18837v1
+
+## 10762. OpenClaw AI Agents as Informal Learners at Moltbook: Characterizing an Emergent Learning Community at Scale
+
+- **ID:** `2602.18832` · **primary:** `cs.HC` · **Submitted:** 2026-02-21
+- **Authors:** Eason Chen, Ce Guan, Ahmed Elshafiey, et al.
+- **abs:** https://arxiv.org/abs/2602.18832 · **PDF:** https://arxiv.org/pdf/2602.18832v1
+
+## 10763. PEPA: a Persistently Autonomous Embodied Agent with Personalities
+
+- **ID:** `2603.00117` · **primary:** `cs.RO` · **Submitted:** 2026-02-21
+- **Authors:** Kaige Liu, Yang Li, Lijun Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.00117 · **PDF:** https://arxiv.org/pdf/2603.00117v2
+
+## 10764. Bayesian Lottery Ticket Hypothesis
+
+- **ID:** `2602.18825` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Nicholas Kuhn, Arvid Weyrauch, Lars Heyen, et al.
+- **abs:** https://arxiv.org/abs/2602.18825 · **PDF:** https://arxiv.org/pdf/2602.18825v1
+
+## 10765. UniRank: A Multi-Agent Calibration Pipeline for Estimating University Rankings from Anonymized Bibliometric Signals
+
+- **ID:** `2602.18824` · **primary:** `cs.SI` · **Submitted:** 2026-02-21
+- **Authors:** Pedram Riyazimehr, Seyyed Ehsan Mahmoudi
+- **abs:** https://arxiv.org/abs/2602.18824 · **PDF:** https://arxiv.org/pdf/2602.18824v1
+
+## 10766. EvalSense: A Framework for Domain-Specific LLM (Meta-)Evaluation
+
+- **ID:** `2602.18823` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Adam Dejl, Jonathan Pearson
+- **abs:** https://arxiv.org/abs/2602.18823 · **PDF:** https://arxiv.org/pdf/2602.18823v1
+
+## 10767. Habilis-$β$: A Fast-Motion and Long-Lasting On-Device Vision-Language-Action Model
+
+- **ID:** `2602.18813` · **primary:** `cs.RO` · **Submitted:** 2026-02-21
+- **Authors:** Tommoro Robotics, :, Jesoon Kang, et al.
+- **abs:** https://arxiv.org/abs/2602.18813 · **PDF:** https://arxiv.org/pdf/2602.18813v1
+
+## 10768. GenPlanner: From Noise to Plans -- Emergent Reasoning in Flow Matching and Diffusion Models
+
+- **ID:** `2602.18812` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Agnieszka Polowczyk, Alicja Polowczyk, Michał Wieczorek
+- **abs:** https://arxiv.org/abs/2602.18812 · **PDF:** https://arxiv.org/pdf/2602.18812v1
+
+## 10769. Chat-Based Support Alone May Not Be Enough: Comparing Conversational and Embedded LLM Feedback for Mathematical Proof Learning
+
+- **ID:** `2602.18807` · **primary:** `cs.HC` · **Submitted:** 2026-02-21
+- **Authors:** Eason Chen, Sophia Judicke, Kayla Beigh, et al.
+- **abs:** https://arxiv.org/abs/2602.18807 · **PDF:** https://arxiv.org/pdf/2602.18807v2
+
+## 10770. Think$^{2}$: Grounded Metacognitive Reasoning in Large Language Models
+
+- **ID:** `2602.18806` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Abraham Paul Elenjical, Vivek Hruday Kavuri, Vasudeva Varma
+- **abs:** https://arxiv.org/abs/2602.18806 · **PDF:** https://arxiv.org/pdf/2602.18806v1
+
+## 10771. SGNO: Spectral Generator Neural Operators for Stable Long Horizon PDE Rollouts
+
+- **ID:** `2602.18801` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Jiayi Li, Zhaonan Wang, Flora D. Salim
+- **abs:** https://arxiv.org/abs/2602.18801 · **PDF:** https://arxiv.org/pdf/2602.18801v1
+
+## 10772. Operational Robustness of LLMs on Code Generation
+
+- **ID:** `2602.18800` · **primary:** `cs.SE` · **Submitted:** 2026-02-21
+- **Authors:** Debalina Ghosh Paul, Hong Zhu, Ian Bayley
+- **abs:** https://arxiv.org/abs/2602.18800 · **PDF:** https://arxiv.org/pdf/2602.18800v1
+
+## 10773. Carbon-aware decentralized dynamic task offloading in MIMO-MEC networks via multi-agent reinforcement learning
+
+- **ID:** `2602.18797` · **primary:** `cs.DC` · **Submitted:** 2026-02-21
+- **Authors:** Mubshra Zulfiqar, Muhammad Ayzed Mirza, Basit Qureshi
+- **abs:** https://arxiv.org/abs/2602.18797 · **PDF:** https://arxiv.org/pdf/2602.18797v1
+
+## 10774. Vectorized Bayesian Inference for Latent Dirichlet-Tree Allocation
+
+- **ID:** `2602.18795` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Zheng Wang, Nizar Bouguila
+- **abs:** https://arxiv.org/abs/2602.18795 · **PDF:** https://arxiv.org/pdf/2602.18795v1
+
+## 10775. From Few-Shot to Zero-Shot: Towards Generalist Graph Anomaly Detection
+
+- **ID:** `2602.18793` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Yixin Liu, Shiyuan Li, Yu Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.18793 · **PDF:** https://arxiv.org/pdf/2602.18793v1
+
+## 10776. From ARIMA to Attention: Power Load Forecasting Using Temporal Deep Learning
+
+- **ID:** `2603.06622` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Suhasnadh Reddy Veluru, Sai Teja Erukude, Viswa Chaitanya Marella
+- **abs:** https://arxiv.org/abs/2603.06622 · **PDF:** https://arxiv.org/pdf/2603.06622v1
+
+## 10777. BURMESE-SAN: Burmese NLP Benchmark for Evaluating Large Language Models
+
+- **ID:** `2602.18788` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Thura Aung, Jann Railey Montalan, Jian Gang Ngui, et al.
+- **abs:** https://arxiv.org/abs/2602.18788 · **PDF:** https://arxiv.org/pdf/2602.18788v2
+
+## 10778. CaliCausalRank: Calibrated Multi-Objective Ad Ranking with Robust Counterfactual Utility Optimization
+
+- **ID:** `2602.18786` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Xikai Yang, Sebastian Sun, Yilin Li, et al.
+- **abs:** https://arxiv.org/abs/2602.18786 · **PDF:** https://arxiv.org/pdf/2602.18786v1
+
+## 10779. MANATEE: Inference-Time Lightweight Diffusion Based Safety Defense for LLMs
+
+- **ID:** `2602.18782` · **primary:** `cs.CR` · **Submitted:** 2026-02-21
+- **Authors:** Chun Yan Ryan Kan, Tommy Tran, Vedant Yadav, et al.
+- **abs:** https://arxiv.org/abs/2602.18782 · **PDF:** https://arxiv.org/pdf/2602.18782v1
+
+## 10780. ArabicNumBench: Evaluating Arabic Number Reading in Large Language Models
+
+- **ID:** `2602.18776` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Anas Alhumud, Abdulaziz Alhammadi, Muhammad Badruddin Khan
+- **abs:** https://arxiv.org/abs/2602.18776 · **PDF:** https://arxiv.org/pdf/2602.18776v1
+
+## 10781. LAMMI-Pathology: A Tool-Centric Bottom-Up LVLM-Agent Framework for Molecularly Informed Medical Intelligence in Pathology
+
+- **ID:** `2602.18773` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Haoyang Su, Shaoting Zhang, Xiaosong Wang
+- **abs:** https://arxiv.org/abs/2602.18773 · **PDF:** https://arxiv.org/pdf/2602.18773v1
+
+## 10782. AINet: Anchor Instances Learning for Regional Heterogeneity in Whole Slide Image
+
+- **ID:** `2602.20187` · **primary:** `eess.IV` · **Submitted:** 2026-02-21
+- **Authors:** Tingting Zheng, Hongxun Yao, Kui Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.20187 · **PDF:** https://arxiv.org/pdf/2602.20187v1
+
+## 10783. GLaDiGAtor: Language-Model-Augmented Multi-Relation Graph Learning for Predicting Disease-Gene Associations
+
+- **ID:** `2602.18769` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Osman Onur Kuzucu, Tunca Doğan
+- **abs:** https://arxiv.org/abs/2602.18769 · **PDF:** https://arxiv.org/pdf/2602.18769v1
+
+## 10784. Nazrin: Atomic Tactics for Graph Neural Networks for Theorem Proving in Lean 4
+
+- **ID:** `2602.18767` · **primary:** `cs.LO` · **Submitted:** 2026-02-21
+- **Authors:** Leni Aniva, Iori Oikawa, David Dill, et al.
+- **abs:** https://arxiv.org/abs/2602.18767 · **PDF:** https://arxiv.org/pdf/2602.18767v2
+
+## 10785. The Convergence of Schema-Guided Dialogue Systems and the Model Context Protocol
+
+- **ID:** `2602.18764` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Andreas Schlapbach
+- **abs:** https://arxiv.org/abs/2602.18764 · **PDF:** https://arxiv.org/pdf/2602.18764v2
+
+## 10786. TAG: Thinking with Action Unit Grounding for Facial Expression Recognition
+
+- **ID:** `2602.18763` · **primary:** `cs.CV` · **Submitted:** 2026-02-21
+- **Authors:** Haobo Lin, Tianyi Bai, Jiajun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.18763 · **PDF:** https://arxiv.org/pdf/2602.18763v1
+
+## 10787. Bounds and Identification of Joint Probabilities of Potential Outcomes and Observed Variables under Monotonicity Assumptions
+
+- **ID:** `2602.18762` · **primary:** `stat.ML` · **Submitted:** 2026-02-21
+- **Authors:** Naoya Hashimoto, Yuta Kawakami, Jin Tian
+- **abs:** https://arxiv.org/abs/2602.18762 · **PDF:** https://arxiv.org/pdf/2602.18762v1
+
+## 10788. Towards Reliable Negative Sampling for Recommendation with Implicit Feedback via In-Community Popularity
+
+- **ID:** `2602.18759` · **primary:** `cs.IR` · **Submitted:** 2026-02-21
+- **Authors:** Chen Chen, Haobo Lin, Yuanbo Xu
+- **abs:** https://arxiv.org/abs/2602.18759 · **PDF:** https://arxiv.org/pdf/2602.18759v1
+
+## 10789. UFO: Unlocking Ultra-Efficient Quantized Private Inference with Protocol and Algorithm Co-Optimization
+
+- **ID:** `2602.18758` · **primary:** `cs.CR` · **Submitted:** 2026-02-21
+- **Authors:** Wenxuan Zeng, Chao Yang, Tianshi Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.18758 · **PDF:** https://arxiv.org/pdf/2602.18758v1
+
+## 10790. Federated Reasoning Distillation Framework with Model Learnability-Aware Data Allocation
+
+- **ID:** `2602.18749` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Wei Guo, Siyuan Lu, Xiangdong Ran, et al.
+- **abs:** https://arxiv.org/abs/2602.18749 · **PDF:** https://arxiv.org/pdf/2602.18749v1
+
+## 10791. Synthesizing Multimodal Geometry Datasets from Scratch and Enabling Visual Alignment via Plotting Code
+
+- **ID:** `2602.18745` · **primary:** `cs.CV` · **Submitted:** 2026-02-21
+- **Authors:** Haobo Lin, Tianyi Bai, Chen Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.18745 · **PDF:** https://arxiv.org/pdf/2602.18745v1
+
+## 10792. RadioGen3D: 3D Radio Map Generation via Adversarial Learning on Large-Scale Synthetic Data
+
+- **ID:** `2602.18744` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Junshen Chen, Angzi Xu, Zezhong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.18744 · **PDF:** https://arxiv.org/pdf/2602.18744v1
+
+## 10793. RoboCurate: Harnessing Diversity with Action-Verified Neural Trajectory for Robot Learning
+
+- **ID:** `2602.18742` · **primary:** `cs.RO` · **Submitted:** 2026-02-21
+- **Authors:** Seungku Kim, Suhyeok Jang, Byungjun Yoon, et al.
+- **abs:** https://arxiv.org/abs/2602.18742 · **PDF:** https://arxiv.org/pdf/2602.18742v1
+
+## 10794. HONEST-CAV: Hierarchical Optimization of Network Signals and Trajectories for Connected and Automated Vehicles with Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.18740` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Ziyan Zhang, Changxin Wan, Peng Hao, et al.
+- **abs:** https://arxiv.org/abs/2602.18740 · **PDF:** https://arxiv.org/pdf/2602.18740v1
+
+## 10795. When World Models Dream Wrong: Physical-Conditioned Adversarial Attacks against World Models
+
+- **ID:** `2602.18739` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Zhixiang Guo, Siyuan Liang, Andras Balogh, et al.
+- **abs:** https://arxiv.org/abs/2602.18739 · **PDF:** https://arxiv.org/pdf/2602.18739v1
+
+## 10796. Rethinking Retrieval-Augmented Generation as a Cooperative Decision-Making Problem
+
+- **ID:** `2602.18734` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Lichang Song, Ting Long, Yi Chang
+- **abs:** https://arxiv.org/abs/2602.18734 · **PDF:** https://arxiv.org/pdf/2602.18734v1
+
+## 10797. Prior Aware Memorization: An Efficient Metric for Distinguishing Memorization from Generalization in Large Language Models
+
+- **ID:** `2602.18733` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Trishita Tiwari, Ari Trachtenberg, G. Edward Suh
+- **abs:** https://arxiv.org/abs/2602.18733 · **PDF:** https://arxiv.org/pdf/2602.18733v1
+
+## 10798. Hebbian-Oscillatory Co-Learning
+
+- **ID:** `2603.08731` · **primary:** `cs.NE` · **Submitted:** 2026-02-21
+- **Authors:** Hasi Hays
+- **abs:** https://arxiv.org/abs/2603.08731 · **PDF:** https://arxiv.org/pdf/2603.08731v1
+
+## 10799. Beyond Description: A Multimodal Agent Framework for Insightful Chart Summarization
+
+- **ID:** `2602.18731` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Yuhang Bai, Yujuan Ding, Shanru Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.18731 · **PDF:** https://arxiv.org/pdf/2602.18731v1
+
+## 10800. MiSCHiEF: A Benchmark in Minimal-Pairs of Safety and Culture for Holistic Evaluation of Fine-Grained Image-Caption Alignment
+
+- **ID:** `2602.18729` · **primary:** `cs.CV` · **Submitted:** 2026-02-21
+- **Authors:** Sagarika Banerjee, Tangatar Madi, Advait Swaminathan, et al.
+- **abs:** https://arxiv.org/abs/2602.18729 · **PDF:** https://arxiv.org/pdf/2602.18729v1
+
+## 10801. Phase-Consistent Magnetic Spectral Learning for Multi-View Clustering
+
+- **ID:** `2602.18728` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Mingdong Lu, Zhikui Chen, Meng Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.18728 · **PDF:** https://arxiv.org/pdf/2602.18728v1
+
+## 10802. WiCompass: Oracle-driven Data Scaling for mmWave Human Pose Estimation
+
+- **ID:** `2602.18726` · **primary:** `cs.CV` · **Submitted:** 2026-02-21
+- **Authors:** Bo Liang, Chen Gong, Haobo Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.18726 · **PDF:** https://arxiv.org/pdf/2602.18726v2
+
+## 10803. When Prompt Optimization Becomes Jailbreaking: Adaptive Red-Teaming of Large Language Models
+
+- **ID:** `2603.19247` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Zafir Shamsi, Nikhil Chekuru, Zachary Guzman, et al.
+- **abs:** https://arxiv.org/abs/2603.19247 · **PDF:** https://arxiv.org/pdf/2603.19247v1
+
+## 10804. Task-Aware Exploration via a Predictive Bisimulation Metric
+
+- **ID:** `2602.18724` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Dayang Liang, Ruihan Liu, Lipeng Wan, et al.
+- **abs:** https://arxiv.org/abs/2602.18724 · **PDF:** https://arxiv.org/pdf/2602.18724v1
+
+## 10805. ReHear: Iterative Pseudo-Label Refinement for Semi-Supervised Speech Recognition via Audio Large Language Models
+
+- **ID:** `2602.18721` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Zefang Liu, Chenyang Zhu, Sangwoo Cho, et al.
+- **abs:** https://arxiv.org/abs/2602.18721 · **PDF:** https://arxiv.org/pdf/2602.18721v1
+
+## 10806. Stochastic Gradient Variational Inference with Price's Gradient Estimator from Bures-Wasserstein to Parameter Space
+
+- **ID:** `2602.18718` · **primary:** `stat.ML` · **Submitted:** 2026-02-21
+- **Authors:** Kyurae Kim, Qiang Fu, Yi-An Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.18718 · **PDF:** https://arxiv.org/pdf/2602.18718v1
+
+## 10807. Temporal Action Representation Learning for Tactical Resource Control and Subsequent Maneuver Generation
+
+- **ID:** `2602.18716` · **primary:** `cs.RO` · **Submitted:** 2026-02-21
+- **Authors:** Hoseong Jung, Sungil Son, Daesol Cho, et al.
+- **abs:** https://arxiv.org/abs/2602.18716 · **PDF:** https://arxiv.org/pdf/2602.18716v1
+
+## 10808. A Data-Driven Method to Map the Functional Organisation of Human Brain White Matter
+
+- **ID:** `2602.18715` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-21
+- **Authors:** Yifei Sun, James M. Shine, Robert D. Sanders, et al.
+- **abs:** https://arxiv.org/abs/2602.18715 · **PDF:** https://arxiv.org/pdf/2602.18715v1
+
+## 10809. Many AI Analysts, One Dataset: Navigating the Agentic Data Science Multiverse
+
+- **ID:** `2602.18710` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Martin Bertran, Riccardo Fogliato, Zhiwei Steven Wu
+- **abs:** https://arxiv.org/abs/2602.18710 · **PDF:** https://arxiv.org/pdf/2602.18710v2
+
+## 10810. GONE: Structural Knowledge Unlearning via Neighborhood-Expanded Distribution Shaping
+
+- **ID:** `2603.12275` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Chahana Dahal, Ashutosh Balasubramaniam, Zuobin Xiong
+- **abs:** https://arxiv.org/abs/2603.12275 · **PDF:** https://arxiv.org/pdf/2603.12275v1
+
+## 10811. EDU-MATRIX: A Society-Centric Generative Cognitive Digital Twin Architecture for Secondary Education
+
+- **ID:** `2602.18705` · **primary:** `cs.MA` · **Submitted:** 2026-02-21
+- **Authors:** Wenjing Zhai, Jianbin Zhang, Tao Liu
+- **abs:** https://arxiv.org/abs/2602.18705 · **PDF:** https://arxiv.org/pdf/2602.18705v1
+
+## 10812. Think with Grounding: Curriculum Reinforced Reasoning with Video Grounding for Long Video Understanding
+
+- **ID:** `2602.18702` · **primary:** `cs.CV` · **Submitted:** 2026-02-21
+- **Authors:** Houlun Chen, Xin Wang, Guangyao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.18702 · **PDF:** https://arxiv.org/pdf/2602.18702v1
+
+## 10813. Watermarking LLM Agent Trajectories
+
+- **ID:** `2602.18700` · **primary:** `cs.CR` · **Submitted:** 2026-02-21
+- **Authors:** Wenlong Meng, Chen Gong, Terry Yue Zhuo, et al.
+- **abs:** https://arxiv.org/abs/2602.18700 · **PDF:** https://arxiv.org/pdf/2602.18700v1
+
+## 10814. Semantic Substrate Theory: An Operator-Theoretic Framework for Geometric Semantic Drift
+
+- **ID:** `2602.18699` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Stephen Russell
+- **abs:** https://arxiv.org/abs/2602.18699 · **PDF:** https://arxiv.org/pdf/2602.18699v1
+
+## 10815. Insertion Based Sequence Generation with Learnable Order Dynamics
+
+- **ID:** `2602.18695` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Dhruvesh Patel, Benjamin Rozonoyer, Gaurav Pandey, et al.
+- **abs:** https://arxiv.org/abs/2602.18695 · **PDF:** https://arxiv.org/pdf/2602.18695v1
+
+## 10816. Assessing Cognitive Biases in LLMs for Judicial Decision Support: Virtuous Victim and Halo Effects
+
+- **ID:** `2603.10016` · **primary:** `cs.CY` · **Submitted:** 2026-02-21
+- **Authors:** Sierra S. Liu
+- **abs:** https://arxiv.org/abs/2603.10016 · **PDF:** https://arxiv.org/pdf/2603.10016v1
+
+## 10817. In-Context Planning with Latent Temporal Abstractions
+
+- **ID:** `2602.18694` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Baiting Luo, Yunuo Zhang, Nathaniel S. Keplinger, et al.
+- **abs:** https://arxiv.org/abs/2602.18694 · **PDF:** https://arxiv.org/pdf/2602.18694v1
+
+## 10818. Contradiction to Consensus: Dual Perspective, Multi Source Retrieval Based Claim Verification with Source Level Disagreement using LLM
+
+- **ID:** `2602.18693` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Md Badsha Biswas, Ozlem Uzuner
+- **abs:** https://arxiv.org/abs/2602.18693 · **PDF:** https://arxiv.org/pdf/2602.18693v1
+
+## 10819. From Trial by Fire To Sleep Like a Baby: A Lexicon of Anxiety Associations for 20k English Multiword Expressions
+
+- **ID:** `2602.18692` · **primary:** `cs.CL` · **Submitted:** 2026-02-21
+- **Authors:** Saif M. Mohammad
+- **abs:** https://arxiv.org/abs/2602.18692 · **PDF:** https://arxiv.org/pdf/2602.18692v1
+
+## 10820. Neural Fields as World Models
+
+- **ID:** `2602.18690` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-21
+- **Authors:** Joshua Nunley
+- **abs:** https://arxiv.org/abs/2602.18690 · **PDF:** https://arxiv.org/pdf/2602.18690v1
+
+## 10821. Memory-Augmented Spiking Networks: Synergistic Integration of Complementary Mechanisms for Neuromorphic Vision
+
+- **ID:** `2603.08730` · **primary:** `cs.NE` · **Submitted:** 2026-02-21
+- **Authors:** Effiong Blessing, Chiung-Yi Tseng, Isaac Nkrumah, et al.
+- **abs:** https://arxiv.org/abs/2603.08730 · **PDF:** https://arxiv.org/pdf/2603.08730v1
+
+## 10822. Transformers for dynamical systems learn transfer operators in-context
+
+- **ID:** `2602.18679` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Anthony Bao, Jeffrey Lai, William Gilpin
+- **abs:** https://arxiv.org/abs/2602.18679 · **PDF:** https://arxiv.org/pdf/2602.18679v1
+
+## 10823. Heterogeneity-agnostic AI/ML-assisted beam selection for multi-panel arrays
+
+- **ID:** `2602.18678` · **primary:** `eess.SP` · **Submitted:** 2026-02-21
+- **Authors:** Ibrahim Kilinc, Robert W. Heath
+- **abs:** https://arxiv.org/abs/2602.18678 · **PDF:** https://arxiv.org/pdf/2602.18678v1
+
+## 10824. Robustness of Deep ReLU Networks to Misclassification of High-Dimensional Data
+
+- **ID:** `2602.18674` · **primary:** `cs.LG` · **Submitted:** 2026-02-21
+- **Authors:** Věra Kůrková
+- **abs:** https://arxiv.org/abs/2602.18674 · **PDF:** https://arxiv.org/pdf/2602.18674v1
+
+## 10825. Spilled Energy in Large Language Models
+
+- **ID:** `2602.18671` · **primary:** `cs.AI` · **Submitted:** 2026-02-21
+- **Authors:** Adrian Robert Minut, Hazem Dewidar, Iacopo Masi
+- **abs:** https://arxiv.org/abs/2602.18671 · **PDF:** https://arxiv.org/pdf/2602.18671v4
+
+## 10826. Explainable Regime Aware Investing
+
+- **ID:** `2603.04441` · **primary:** `q-fin.PM` · **Submitted:** 2026-02-21
+- **Authors:** Amine Boukardagha
+- **abs:** https://arxiv.org/abs/2603.04441 · **PDF:** https://arxiv.org/pdf/2603.04441v1
+
+## 10827. Toward AI Autonomous Navigation for Mechanical Thrombectomy using Hierarchical Modular Multi-agent Reinforcement Learning (HM-MARL)
+
+- **ID:** `2602.18663` · **primary:** `cs.RO` · **Submitted:** 2026-02-20
+- **Authors:** Harry Robertshaw, Nikola Fischer, Lennart Karstensen, et al.
+- **abs:** https://arxiv.org/abs/2602.18663 · **PDF:** https://arxiv.org/pdf/2602.18663v1
+
+## 10828. Large Causal Models for Temporal Causal Discovery
+
+- **ID:** `2602.18662` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Nikolaos Kougioulis, Nikolaos Gkorgkolis, MingXue Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.18662 · **PDF:** https://arxiv.org/pdf/2602.18662v1
+
+## 10829. Reward Under Attack: Analyzing the Robustness and Hackability of Process Reward Models
+
+- **ID:** `2603.06621` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Rishabh Tiwari, Aditya Tomar, Udbhav Bamba, et al.
+- **abs:** https://arxiv.org/abs/2603.06621 · **PDF:** https://arxiv.org/pdf/2603.06621v1
+
+## 10830. Communication-Efficient Personalized Adaptation via Federated-Local Model Merging
+
+- **ID:** `2602.18658` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Yinan Zou, Md Kamran Chowdhury Shisher, Christopher G. Brinton, et al.
+- **abs:** https://arxiv.org/abs/2602.18658 · **PDF:** https://arxiv.org/pdf/2602.18658v1
+
+## 10831. PolyFrame at MWE-2026 AdMIRe 2: When Words Are Not Enough: Multimodal Idiom Disambiguation
+
+- **ID:** `2602.18652` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Nina Hosseini-Kivanani
+- **abs:** https://arxiv.org/abs/2602.18652 · **PDF:** https://arxiv.org/pdf/2602.18652v2
+
+## 10832. NutriOrion: A Hierarchical Multi-Agent Framework for Personalized Nutrition Intervention Grounded in Clinical Guidelines
+
+- **ID:** `2602.18650` · **primary:** `cs.MA` · **Submitted:** 2026-02-20
+- **Authors:** Junwei Wu, Runze Yan, Hanqi Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.18650 · **PDF:** https://arxiv.org/pdf/2602.18650v1
+
+## 10833. Global Low-Rank, Local Full-Rank: The Holographic Encoding of Learned Algorithms
+
+- **ID:** `2602.18649` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Yongzhong Xu
+- **abs:** https://arxiv.org/abs/2602.18649 · **PDF:** https://arxiv.org/pdf/2602.18649v1
+
+## 10834. Information-Guided Noise Allocation for Efficient Diffusion Training
+
+- **ID:** `2602.18647` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Gabriel Raya, Bac Nguyen, Georgios Batzolis, et al.
+- **abs:** https://arxiv.org/abs/2602.18647 · **PDF:** https://arxiv.org/pdf/2602.18647v1
+
+## 10835. Adaptive Time Series Reasoning via Segment Selection
+
+- **ID:** `2602.18645` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Shvat Messica, Jiawen Zhang, Kevin Li, et al.
+- **abs:** https://arxiv.org/abs/2602.18645 · **PDF:** https://arxiv.org/pdf/2602.18645v1
+
+## 10836. Auto Quantum Machine Learning for Multisource Classification
+
+- **ID:** `2602.18642` · **primary:** `quant-ph` · **Submitted:** 2026-02-20
+- **Authors:** Tomasz Rybotycki, Sebastian Dziura, Piotr Gawron
+- **abs:** https://arxiv.org/abs/2602.18642 · **PDF:** https://arxiv.org/pdf/2602.18642v1
+
+## 10837. Decoding ML Decision: An Agentic Reasoning Framework for Large-Scale Ranking System
+
+- **ID:** `2602.18640` · **primary:** `cs.AI` · **Submitted:** 2026-02-20
+- **Authors:** Longfei Yun, Yihan Wu, Haoran Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.18640 · **PDF:** https://arxiv.org/pdf/2602.18640v1
+
+## 10838. Learning Invariant Visual Representations for Planning with Joint-Embedding Predictive World Models
+
+- **ID:** `2602.18639` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Leonardo F. Toso, Davit Shadunts, Yunyang Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.18639 · **PDF:** https://arxiv.org/pdf/2602.18639v1
+
+## 10839. Online decoding of rat self-paced locomotion speed from EEG using recurrent neural networks
+
+- **ID:** `2602.18637` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Alejandro de Miguel, Nelson Totah, Uri Maoz
+- **abs:** https://arxiv.org/abs/2602.18637 · **PDF:** https://arxiv.org/pdf/2602.18637v1
+
+## 10840. Musical Training, but not Mere Exposure to Music, Drives the Emergence of Chroma Equivalence in Artificial Neural Networks
+
+- **ID:** `2602.18635` · **primary:** `cs.SD` · **Submitted:** 2026-02-20
+- **Authors:** Lukas Grasse, Matthew S. Tata
+- **abs:** https://arxiv.org/abs/2602.18635 · **PDF:** https://arxiv.org/pdf/2602.18635v1
+
+## 10841. DP-RFT: Learning to Generate Synthetic Text via Differentially Private Reinforcement Fine-Tuning
+
+- **ID:** `2602.18633` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Fangyuan Xu, Sihao Chen, Zinan Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.18633 · **PDF:** https://arxiv.org/pdf/2602.18633v1
+
+## 10842. GraphSkill: Documentation-Guided Hierarchical Retrieval-Augmented Coding for Complex Graph Reasoning
+
+- **ID:** `2603.06620` · **primary:** `cs.SE` · **Submitted:** 2026-02-20
+- **Authors:** Fali Wang, Chenglin Weng, Xianren Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.06620 · **PDF:** https://arxiv.org/pdf/2603.06620v1
+
+## 10843. Lost in Instructions: Study of Blind Users' Experiences with DIY Manuals and AI-Rewritten Instructions for Assembly, Operation, and Troubleshooting of Tangible Products
+
+- **ID:** `2602.18630` · **primary:** `cs.HC` · **Submitted:** 2026-02-20
+- **Authors:** Monalika Padma Reddy, Aruna Balasubramanian, Jiawei Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.18630 · **PDF:** https://arxiv.org/pdf/2602.18630v1
+
+## 10844. Non-Interfering Weight Fields: Treating Model Parameters as a Continuously Extensible Function
+
+- **ID:** `2602.18628` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Sarim Chaudhry
+- **abs:** https://arxiv.org/abs/2602.18628 · **PDF:** https://arxiv.org/pdf/2602.18628v1
+
+## 10845. Federated Learning-Assisted Optimization of Mobile Transmission with Digital Twins
+
+- **ID:** `2602.18627` · **primary:** `cs.NI` · **Submitted:** 2026-02-20
+- **Authors:** Mohammad Heydari, Terence D. Todd, Dongmei Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.18627 · **PDF:** https://arxiv.org/pdf/2602.18627v1
+
+## 10846. Finding the Signal in the Noise: An Exploratory Study on Assessing the Effectiveness of AI and Accessibility Forums for Blind Users' Support Needs
+
+- **ID:** `2602.18623` · **primary:** `cs.HC` · **Submitted:** 2026-02-20
+- **Authors:** Satwik Ram Kodandaram, Jiawei Zhou, Xiaojun Bi, et al.
+- **abs:** https://arxiv.org/abs/2602.18623 · **PDF:** https://arxiv.org/pdf/2602.18623v1
+
+## 10847. Not all tokens are needed(NAT): token efficient reinforcement learning
+
+- **ID:** `2603.06619` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Hejian Sang, Yuanda Xu, Zhengze Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.06619 · **PDF:** https://arxiv.org/pdf/2603.06619v1
+
+## 10848. Diagnosing LLM Reranker Behavior Under Fixed Evidence Pools
+
+- **ID:** `2602.18613` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Baris Arat, Emre Sefer
+- **abs:** https://arxiv.org/abs/2602.18613 · **PDF:** https://arxiv.org/pdf/2602.18613v1
+
+## 10849. Feedback-based Automated Verification in Vibe Coding of CAS Adaptation Built on Constraint Logic
+
+- **ID:** `2602.18607` · **primary:** `cs.AI` · **Submitted:** 2026-02-20
+- **Authors:** Michal Töpfer, František Plášil, Tomáš Bureš, et al.
+- **abs:** https://arxiv.org/abs/2602.18607 · **PDF:** https://arxiv.org/pdf/2602.18607v1
+
+## 10850. Enhancing Goal Inference via Correction Timing
+
+- **ID:** `2602.18603` · **primary:** `cs.RO` · **Submitted:** 2026-02-20
+- **Authors:** Anjiabei Wang, Shuangge Wang, Tesca Fitzgerald
+- **abs:** https://arxiv.org/abs/2602.18603 · **PDF:** https://arxiv.org/pdf/2602.18603v1
+
+## 10851. MapTab: Can MLLMs Master Constrained Route Planning?
+
+- **ID:** `2602.18600` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Ziqiao Shang, Lingyue Ge, Yang Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.18600 · **PDF:** https://arxiv.org/pdf/2602.18600v1
+
+## 10852. Ensemble Prediction of Task Affinity for Efficient Multi-Task Learning
+
+- **ID:** `2602.18591` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Afiya Ayman, Ayan Mukhopadhyay, Aron Laszka
+- **abs:** https://arxiv.org/abs/2602.18591 · **PDF:** https://arxiv.org/pdf/2602.18591v1
+
+## 10853. DM4CT: Benchmarking Diffusion Models for Computed Tomography Reconstruction
+
+- **ID:** `2602.18589` · **primary:** `eess.IV` · **Submitted:** 2026-02-20
+- **Authors:** Jiayang Shi, Daniel M. Pelt, K. Joost Batenburg
+- **abs:** https://arxiv.org/abs/2602.18589 · **PDF:** https://arxiv.org/pdf/2602.18589v1
+
+## 10854. BloomNet: Exploring Single vs. Multiple Object Annotation for Flower Recognition Using YOLO Variants
+
+- **ID:** `2602.18585` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Safwat Nusrat, Prithwiraj Bhattacharjee
+- **abs:** https://arxiv.org/abs/2602.18585 · **PDF:** https://arxiv.org/pdf/2602.18585v1
+
+## 10855. GIST: Targeted Data Selection for Instruction Tuning via Coupled Optimization Geometry
+
+- **ID:** `2602.18584` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Guanghui Min, Tianhao Huang, Ke Wan, et al.
+- **abs:** https://arxiv.org/abs/2602.18584 · **PDF:** https://arxiv.org/pdf/2602.18584v1
+
+## 10856. Luna-2: Scalable Single-Token Evaluation with Small Language Models
+
+- **ID:** `2602.18583` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Vatsal Goel, Rishon Dsouza, Nikhil Ega, et al.
+- **abs:** https://arxiv.org/abs/2602.18583 · **PDF:** https://arxiv.org/pdf/2602.18583v1
+
+## 10857. Hierarchical Reward Design from Language: Enhancing Alignment of Agent Behavior with Human Specifications
+
+- **ID:** `2602.18582` · **primary:** `cs.AI` · **Submitted:** 2026-02-20
+- **Authors:** Zhiqin Qian, Ryan Diaz, Sangwon Seo, et al.
+- **abs:** https://arxiv.org/abs/2602.18582 · **PDF:** https://arxiv.org/pdf/2602.18582v1
+
+## 10858. Learning Beyond Optimization: Stress-Gated Dynamical Regime Regulation in Autonomous Systems
+
+- **ID:** `2602.18581` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Sheng Ran
+- **abs:** https://arxiv.org/abs/2602.18581 · **PDF:** https://arxiv.org/pdf/2602.18581v1
+
+## 10859. Multi-Axis Trust Modeling for Interpretable Account Hijacking Detection
+
+- **ID:** `2603.13246` · **primary:** `cs.AI` · **Submitted:** 2026-02-20
+- **Authors:** Mohammad AL-Smadi
+- **abs:** https://arxiv.org/abs/2603.13246 · **PDF:** https://arxiv.org/pdf/2603.13246v1
+
+## 10860. Multiclass Calibration Assessment and Recalibration of Probability Predictions via the Linear Log Odds Calibration Function
+
+- **ID:** `2602.18573` · **primary:** `stat.ML` · **Submitted:** 2026-02-20
+- **Authors:** Amy Vennos, Xin Xing, Christopher T. Franck
+- **abs:** https://arxiv.org/abs/2602.18573 · **PDF:** https://arxiv.org/pdf/2602.18573v1
+
+## 10861. Sub-City Real Estate Price Index Forecasting at Weekly Horizons Using Satellite Radar and News Sentiment
+
+- **ID:** `2602.18572` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Baris Arat, Hasan Fehmi Ates, Emre Sefer
+- **abs:** https://arxiv.org/abs/2602.18572 · **PDF:** https://arxiv.org/pdf/2602.18572v1
+
+## 10862. Debug2Fix: Supercharging Coding Agents with Interactive Debugging Capabilities
+
+- **ID:** `2602.18571` · **primary:** `cs.SE` · **Submitted:** 2026-02-20
+- **Authors:** Spandan Garg, Yufan Huang
+- **abs:** https://arxiv.org/abs/2602.18571 · **PDF:** https://arxiv.org/pdf/2602.18571v1
+
+## 10863. RPU -- A Reasoning Processing Unit
+
+- **ID:** `2602.18568` · **primary:** `cs.AR` · **Submitted:** 2026-02-20
+- **Authors:** Matthew Adiletta, Gu-Yeon Wei, David Brooks
+- **abs:** https://arxiv.org/abs/2602.18568 · **PDF:** https://arxiv.org/pdf/2602.18568v2
+
+## 10864. Assigning Confidence: K-partition Ensembles
+
+- **ID:** `2602.18435` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Aggelos Semoglou, John Pavlopoulos
+- **abs:** https://arxiv.org/abs/2602.18435 · **PDF:** https://arxiv.org/pdf/2602.18435v1
+
+## 10865. From Static Spectra to Operando Infrared Dynamics: Physics Informed Flow Modeling and a Benchmark
+
+- **ID:** `2602.18551` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-20
+- **Authors:** Shuquan Ye, Ben Fei, Hongbin Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.18551 · **PDF:** https://arxiv.org/pdf/2602.18551v1
+
+## 10866. SMaRT: Online Reusable Resource Assignment and an Application to Mediation in the Kenyan Judiciary
+
+- **ID:** `2602.18431` · **primary:** `cs.CY` · **Submitted:** 2026-02-20
+- **Authors:** Shafkat Farabi, Didac Marti Pinto, Wei Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.18431 · **PDF:** https://arxiv.org/pdf/2602.18431v1
+
+## 10867. VIRAASAT: Traversing Novel Paths for Indian Cultural Reasoning
+
+- **ID:** `2602.18429` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Harshul Raj Surana, Arijit Maji, Aryan Vats, et al.
+- **abs:** https://arxiv.org/abs/2602.18429 · **PDF:** https://arxiv.org/pdf/2602.18429v1
+
+## 10868. The Geometry of Noise: Why Diffusion Models Don't Need Noise Conditioning
+
+- **ID:** `2602.18428` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Mojtaba Sahraee-Ardakan, Mauricio Delbracio, Peyman Milanfar
+- **abs:** https://arxiv.org/abs/2602.18428 · **PDF:** https://arxiv.org/pdf/2602.18428v1
+
+## 10869. RVR: Retrieve-Verify-Retrieve for Comprehensive Question Answering
+
+- **ID:** `2602.18425` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Deniz Qian, Hung-Ting Chen, Eunsol Choi
+- **abs:** https://arxiv.org/abs/2602.18425 · **PDF:** https://arxiv.org/pdf/2602.18425v1
+
+## 10870. SPQ: An Ensemble Technique for Large Language Model Compression
+
+- **ID:** `2602.18420` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Jiamin Yao, Eren Gultepe
+- **abs:** https://arxiv.org/abs/2602.18420 · **PDF:** https://arxiv.org/pdf/2602.18420v1
+
+## 10871. Benchmarking Graph Neural Networks in Solving Hard Constraint Satisfaction Problems
+
+- **ID:** `2602.18419` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-02-20
+- **Authors:** Geri Skenderi, Lorenzo Buffoni, Francesco D'Amico, et al.
+- **abs:** https://arxiv.org/abs/2602.18419 · **PDF:** https://arxiv.org/pdf/2602.18419v2
+
+## 10872. Subgroups of $U(d)$ Induce Natural RNN and Transformer Architectures
+
+- **ID:** `2602.18417` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Joshua Nunley
+- **abs:** https://arxiv.org/abs/2602.18417 · **PDF:** https://arxiv.org/pdf/2602.18417v1
+
+## 10873. Distilling and Adapting: A Topology-Aware Framework for Zero-Shot Interaction Prediction in Multiplex Biological Networks
+
+- **ID:** `2603.06618` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Alana Deng, Sugitha Janarthanan, Yan Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.06618 · **PDF:** https://arxiv.org/pdf/2603.06618v1
+
+## 10874. Unifying approach to uniform expressivity of graph neural networks
+
+- **ID:** `2602.18409` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Huan Luo, Jonni Virtema
+- **abs:** https://arxiv.org/abs/2602.18409 · **PDF:** https://arxiv.org/pdf/2602.18409v1
+
+## 10875. Latent Equivariant Operators for Robust Object Recognition: Promises and Challenges
+
+- **ID:** `2602.18406` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Minh Dinh, Stéphane Deny
+- **abs:** https://arxiv.org/abs/2602.18406 · **PDF:** https://arxiv.org/pdf/2602.18406v4
+
+## 10876. Scientific Knowledge-Guided Machine Learning for Vessel Power Prediction: A Comparative Study
+
+- **ID:** `2602.18403` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Orfeas Bourchas, George Papalambrou
+- **abs:** https://arxiv.org/abs/2602.18403 · **PDF:** https://arxiv.org/pdf/2602.18403v1
+
+## 10877. Leakage and Second-Order Dynamics Improve Hippocampal RNN Replay
+
+- **ID:** `2602.18401` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Josue Casco-Rodriguez, Nanda H. Krishna, Richard G. Baraniuk
+- **abs:** https://arxiv.org/abs/2602.18401 · **PDF:** https://arxiv.org/pdf/2602.18401v1
+
+## 10878. PRISM-FCP: Byzantine-Resilient Federated Conformal Prediction via Partial Sharing
+
+- **ID:** `2602.18396` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Ehsan Lari, Reza Arablouei, Stefan Werner
+- **abs:** https://arxiv.org/abs/2602.18396 · **PDF:** https://arxiv.org/pdf/2602.18396v1
+
+## 10879. Learning to Tune Pure Pursuit in Autonomous Racing: Joint Lookahead and Steering-Gain Control with PPO
+
+- **ID:** `2602.18386` · **primary:** `cs.RO` · **Submitted:** 2026-02-20
+- **Authors:** Mohamed Elgouhary, Amr S. El-Wakeel
+- **abs:** https://arxiv.org/abs/2602.18386 · **PDF:** https://arxiv.org/pdf/2602.18386v1
+
+## 10880. 1D-Bench: A Benchmark for Iterative UI Code Generation with Visual Feedback in Real-World
+
+- **ID:** `2602.18548` · **primary:** `cs.SE` · **Submitted:** 2026-02-20
+- **Authors:** Qiao Xu, Yipeng Yu, Chengxiao Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.18548 · **PDF:** https://arxiv.org/pdf/2602.18548v1
+
+## 10881. FedZMG: Efficient Client-Side Optimization in Federated Learning
+
+- **ID:** `2602.18384` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Fotios Zantalis, Evangelos Zervas, Grigorios Koulouras
+- **abs:** https://arxiv.org/abs/2602.18384 · **PDF:** https://arxiv.org/pdf/2602.18384v1
+
+## 10882. Theory and interpretability of Quantum Extreme Learning Machines: a Pauli-transfer matrix approach
+
+- **ID:** `2602.18377` · **primary:** `quant-ph` · **Submitted:** 2026-02-20
+- **Authors:** Markus Gross, Hans-Martin Rieser
+- **abs:** https://arxiv.org/abs/2602.18377 · **PDF:** https://arxiv.org/pdf/2602.18377v2
+
+## 10883. Zero-shot Interactive Perception
+
+- **ID:** `2602.18374` · **primary:** `cs.RO` · **Submitted:** 2026-02-20
+- **Authors:** Venkatesh Sripada, Frank Guerin, Amir Ghalamzan
+- **abs:** https://arxiv.org/abs/2602.18374 · **PDF:** https://arxiv.org/pdf/2602.18374v1
+
+## 10884. "How Do I ...?": Procedural Questions Predominate Student-LLM Chatbot Conversations
+
+- **ID:** `2602.18372` · **primary:** `cs.HC` · **Submitted:** 2026-02-20
+- **Authors:** Alexandra Neagu, Marcus Messer, Peter Johnson, et al.
+- **abs:** https://arxiv.org/abs/2602.18372 · **PDF:** https://arxiv.org/pdf/2602.18372v1
+
+## 10885. A systematic approach to answering the easy problems of consciousness based on an executable cognitive system
+
+- **ID:** `2603.04440` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-20
+- **Authors:** Qi Zhang
+- **abs:** https://arxiv.org/abs/2603.04440 · **PDF:** https://arxiv.org/pdf/2603.04440v1
+
+## 10886. Quantum Maximum Likelihood Prediction via Hilbert Space Embeddings
+
+- **ID:** `2602.18364` · **primary:** `cs.IT` · **Submitted:** 2026-02-20
+- **Authors:** Sreejith Sreekumar, Nir Weinberger
+- **abs:** https://arxiv.org/abs/2602.18364 · **PDF:** https://arxiv.org/pdf/2602.18364v1
+
+## 10887. Validating Political Position Predictions of Arguments
+
+- **ID:** `2602.18351` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Jordan Robinson, Angus R. Williams, Katie Atkinson, et al.
+- **abs:** https://arxiv.org/abs/2602.18351 · **PDF:** https://arxiv.org/pdf/2602.18351v1
+
+## 10888. Quantum-enhanced satellite image classification
+
+- **ID:** `2602.18350` · **primary:** `quant-ph` · **Submitted:** 2026-02-20
+- **Authors:** Qi Zhang, Anton Simen, Carlos Flores-Garrigós, et al.
+- **abs:** https://arxiv.org/abs/2602.18350 · **PDF:** https://arxiv.org/pdf/2602.18350v1
+
+## 10889. Explaining AutoClustering: Uncovering Meta-Feature Contribution in AutoML for Clustering
+
+- **ID:** `2602.18348` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Matheus Camilo da Silva, Leonardo Arrighi, Ana Carolina Lorena, et al.
+- **abs:** https://arxiv.org/abs/2602.18348 · **PDF:** https://arxiv.org/pdf/2602.18348v1
+
+## 10890. Vichara: Appellate Judgment Prediction and Explanation for the Indian Judicial System
+
+- **ID:** `2602.18346` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Pavithra PM Nair, Preethu Rose Anish
+- **abs:** https://arxiv.org/abs/2602.18346 · **PDF:** https://arxiv.org/pdf/2602.18346v1
+
+## 10891. On the "Induction Bias" in Sequence Models
+
+- **ID:** `2602.18333` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** M. Reza Ebrahimi, Michaël Defferrard, Sunny Panchal, et al.
+- **abs:** https://arxiv.org/abs/2602.18333 · **PDF:** https://arxiv.org/pdf/2602.18333v1
+
+## 10892. Predicting Contextual Informativeness for Vocabulary Learning using Deep Learning
+
+- **ID:** `2602.18326` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Tao Wu, Adam Kapelner
+- **abs:** https://arxiv.org/abs/2602.18326 · **PDF:** https://arxiv.org/pdf/2602.18326v1
+
+## 10893. PsihoRo: Depression and Anxiety Romanian Text Corpus
+
+- **ID:** `2602.18324` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Alexandra Ciobotaru, Ana-Maria Bucur, Liviu P. Dinu
+- **abs:** https://arxiv.org/abs/2602.18324 · **PDF:** https://arxiv.org/pdf/2602.18324v4
+
+## 10894. Robo-Saber: Generating and Simulating Virtual Reality Players
+
+- **ID:** `2602.18319` · **primary:** `cs.GR` · **Submitted:** 2026-02-20
+- **Authors:** Nam Hee Kim, Jingjing May Liu, Jaakko Lehtinen, et al.
+- **abs:** https://arxiv.org/abs/2602.18319 · **PDF:** https://arxiv.org/pdf/2602.18319v1
+
+## 10895. Clapeyron Neural Networks for Single-Species Vapor-Liquid Equilibria
+
+- **ID:** `2602.18313` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-20
+- **Authors:** Jan Pavšek, Alexander Mitsos, Elvis J. Sim, et al.
+- **abs:** https://arxiv.org/abs/2602.18313 · **PDF:** https://arxiv.org/pdf/2602.18313v1
+
+## 10896. JPmHC Dynamical Isometry via Orthogonal Hyper-Connections
+
+- **ID:** `2602.18308` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Biswa Sengupta, Jinhua Wang, Leo Brunswic
+- **abs:** https://arxiv.org/abs/2602.18308 · **PDF:** https://arxiv.org/pdf/2602.18308v2
+
+## 10897. VeriSoftBench: Repository-Scale Formal Verification Benchmarks for Lean
+
+- **ID:** `2602.18307` · **primary:** `cs.SE` · **Submitted:** 2026-02-20
+- **Authors:** Yutong Xin, Qiaochu Chen, Greg Durrett, et al.
+- **abs:** https://arxiv.org/abs/2602.18307 · **PDF:** https://arxiv.org/pdf/2602.18307v1
+
+## 10898. On the Semantic and Syntactic Information Encoded in Proto-Tokens for One-Step Text Reconstruction
+
+- **ID:** `2602.18301` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Ivan Bondarenko, Egor Palkin, Fedor Tikunov
+- **abs:** https://arxiv.org/abs/2602.18301 · **PDF:** https://arxiv.org/pdf/2602.18301v1
+
+## 10899. Analyzing and Improving Chain-of-Thought Monitorability Through Information Theory
+
+- **ID:** `2602.18297` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Usman Anwar, Tim Bakker, Dana Kianfar, et al.
+- **abs:** https://arxiv.org/abs/2602.18297 · **PDF:** https://arxiv.org/pdf/2602.18297v1
+
+## 10900. Context-Aware Mapping of 2D Drawing Annotations to 3D CAD Features Using LLM-Assisted Reasoning for Manufacturing Automation
+
+- **ID:** `2602.18296` · **primary:** `cs.CE` · **Submitted:** 2026-02-20
+- **Authors:** Muhammad Tayyab Khan, Lequn Chen, Wenhe Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.18296 · **PDF:** https://arxiv.org/pdf/2602.18296v2
+
+## 10901. Decoding as Optimisation on the Probability Simplex: From Top-K to Top-P (Nucleus) to Best-of-K Samplers
+
+- **ID:** `2602.18292` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Xiaotong Ji, Rasul Tutunov, Matthieu Zimmer, et al.
+- **abs:** https://arxiv.org/abs/2602.18292 · **PDF:** https://arxiv.org/pdf/2602.18292v2
+
+## 10902. Diffusing to Coordinate: Efficient Online Multi-Agent Diffusion Policies
+
+- **ID:** `2602.18291` · **primary:** `cs.AI` · **Submitted:** 2026-02-20
+- **Authors:** Zhuoran Li, Hai Zhong, Xun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.18291 · **PDF:** https://arxiv.org/pdf/2602.18291v1
+
+## 10903. Evo: Autoregressive-Diffusion Large Language Models with Evolving Balance
+
+- **ID:** `2603.06617` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Junde Wu, Minhao Hu, Jiayuan Zhu, et al.
+- **abs:** https://arxiv.org/abs/2603.06617 · **PDF:** https://arxiv.org/pdf/2603.06617v1
+
+## 10904. Automating Document Intelligence in Statutory City Planning
+
+- **ID:** `2603.13245` · **primary:** `cs.AI` · **Submitted:** 2026-02-20
+- **Authors:** Lars Malmqvist, Robin Barber
+- **abs:** https://arxiv.org/abs/2603.13245 · **PDF:** https://arxiv.org/pdf/2603.13245v1
+
+## 10905. Rodent-Bench
+
+- **ID:** `2602.18540` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Thomas Heap, Laurence Aitchison, Emma Cahill, et al.
+- **abs:** https://arxiv.org/abs/2602.18540 · **PDF:** https://arxiv.org/pdf/2602.18540v1
+
+## 10906. Descriptor: Dataset of Parasitoid Wasps and Associated Hymenoptera (DAPWH)
+
+- **ID:** `2602.20028` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Joao Manoel Herrera Pinheiro, Gabriela Do Nascimento Herrera, Luciana Bueno Dos Reis Fernandes, et al.
+- **abs:** https://arxiv.org/abs/2602.20028 · **PDF:** https://arxiv.org/pdf/2602.20028v1
+
+## 10907. HyTRec: A Hybrid Temporal-Aware Attention Architecture for Long Behavior Sequential Recommendation
+
+- **ID:** `2602.18283` · **primary:** `cs.IR` · **Submitted:** 2026-02-20
+- **Authors:** Lei Xin, Yuhao Zheng, Ke Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.18283 · **PDF:** https://arxiv.org/pdf/2602.18283v1
+
+## 10908. PRISM: Parallel Reward Integration with Symmetry for MORL
+
+- **ID:** `2602.18277` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Finn van der Knaap, Kejiang Qian, Zheng Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.18277 · **PDF:** https://arxiv.org/pdf/2602.18277v1
+
+## 10909. A Probabilistic Framework for LLM-Based Model Discovery
+
+- **ID:** `2602.18266` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Stefan Wahl, Raphaela Schenk, Ali Farnoud, et al.
+- **abs:** https://arxiv.org/abs/2602.18266 · **PDF:** https://arxiv.org/pdf/2602.18266v1
+
+## 10910. Multimodal Modular Chain of Thoughts in Energy Performance Certificate Assessment
+
+- **ID:** `2603.00115` · **primary:** `physics.soc-ph` · **Submitted:** 2026-02-20
+- **Authors:** Zhen Peng, Peter J. Bentley
+- **abs:** https://arxiv.org/abs/2603.00115 · **PDF:** https://arxiv.org/pdf/2603.00115v1
+
+## 10911. Simplifying Outcomes of Language Model Component Analyses with ELIA
+
+- **ID:** `2602.18262` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Aaron Louis Eidt, Nils Feldhus
+- **abs:** https://arxiv.org/abs/2602.18262 · **PDF:** https://arxiv.org/pdf/2602.18262v1
+
+## 10912. MEG-to-MEG Transfer Learning and Cross-Task Speech/Silence Detection with Limited Data
+
+- **ID:** `2602.18253` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Xabier de Zuazo, Vincenzo Verbeni, Eva Navas, et al.
+- **abs:** https://arxiv.org/abs/2602.18253 · **PDF:** https://arxiv.org/pdf/2602.18253v1
+
+## 10913. On the Adversarial Robustness of Discrete Image Tokenizers
+
+- **ID:** `2602.18252` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Rishika Bhagwatkar, Irina Rish, Nicolas Flammarion, et al.
+- **abs:** https://arxiv.org/abs/2602.18252 · **PDF:** https://arxiv.org/pdf/2602.18252v1
+
+## 10914. Variational Distributional Neuron
+
+- **ID:** `2602.18250` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Yves Ruffenach
+- **abs:** https://arxiv.org/abs/2602.18250 · **PDF:** https://arxiv.org/pdf/2602.18250v1
+
+## 10915. Neural-HSS: Hierarchical Semi-Separable Neural PDE Solver
+
+- **ID:** `2602.18248` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Pietro Sittoni, Emanuele Zangrando, Angelo A. Casulli, et al.
+- **abs:** https://arxiv.org/abs/2602.18248 · **PDF:** https://arxiv.org/pdf/2602.18248v1
+
+## 10916. A foundation model for electrodermal activity data
+
+- **ID:** `2603.16878` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Leonardo Alchieri, Matteo Garzon, Lidia Alecci, et al.
+- **abs:** https://arxiv.org/abs/2603.16878 · **PDF:** https://arxiv.org/pdf/2603.16878v1
+
+## 10917. Thinking by Subtraction: Confidence-Driven Contrastive Decoding for LLM Reasoning
+
+- **ID:** `2602.18232` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Lexiang Tang, Weihao Gao, Bingchen Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.18232 · **PDF:** https://arxiv.org/pdf/2602.18232v1
+
+## 10918. [Re] Benchmarking LLM Capabilities in Negotiation through Scoreable Games
+
+- **ID:** `2602.18230` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Jorge Carrasco Pollo, Ioannis Kapetangeorgis, Joshua Rosenthal, et al.
+- **abs:** https://arxiv.org/abs/2602.18230 · **PDF:** https://arxiv.org/pdf/2602.18230v1
+
+## 10919. Parameter-Efficient Domain Adaptation of Physics-Informed Self-Attention based GNNs for AC Power Flow Prediction
+
+- **ID:** `2602.18227` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Redwanul Karim, Changhun Kim, Timon Conrad, et al.
+- **abs:** https://arxiv.org/abs/2602.18227 · **PDF:** https://arxiv.org/pdf/2602.18227v1
+
+## 10920. SimVLA: A Simple VLA Baseline for Robotic Manipulation
+
+- **ID:** `2602.18224` · **primary:** `cs.RO` · **Submitted:** 2026-02-20
+- **Authors:** Yuankai Luo, Woping Chen, Tong Liang, et al.
+- **abs:** https://arxiv.org/abs/2602.18224 · **PDF:** https://arxiv.org/pdf/2602.18224v1
+
+## 10921. Information-Theoretic Storage Cost in Sentence Comprehension
+
+- **ID:** `2602.18217` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Kohei Kajikawa, Shinnosuke Isono, Ethan Gotlieb Wilcox
+- **abs:** https://arxiv.org/abs/2602.18217 · **PDF:** https://arxiv.org/pdf/2602.18217v1
+
+## 10922. Generative Model via Quantile Assignment
+
+- **ID:** `2602.18216` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Georgi Hrusanov, Oliver Y. Chén, Julien S. Bodelet
+- **abs:** https://arxiv.org/abs/2602.18216 · **PDF:** https://arxiv.org/pdf/2602.18216v1
+
+## 10923. Machine-learning force-field models for dynamical simulations of metallic magnets
+
+- **ID:** `2602.18213` · **primary:** `cond-mat.str-el` · **Submitted:** 2026-02-20
+- **Authors:** Gia-Wei Chern, Yunhao Fan, Sheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.18213 · **PDF:** https://arxiv.org/pdf/2602.18213v1
+
+## 10924. SOMtime the World Ain$'$t Fair: Violating Fairness Using Self-Organizing Maps
+
+- **ID:** `2602.18201` · **primary:** `cs.AI` · **Submitted:** 2026-02-20
+- **Authors:** Joseph Bingham, Netanel Arussy, Dvir Aran
+- **abs:** https://arxiv.org/abs/2602.18201 · **PDF:** https://arxiv.org/pdf/2602.18201v1
+
+## 10925. RAT+: Train Dense, Infer Sparse -- Recurrence Augmented Attention for Dilated Inference
+
+- **ID:** `2602.18196` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Xiuying Wei, Caglar Gulcehre
+- **abs:** https://arxiv.org/abs/2602.18196 · **PDF:** https://arxiv.org/pdf/2602.18196v2
+
+## 10926. LERD: Latent Event-Relational Dynamics for Neurodegenerative Classification
+
+- **ID:** `2602.18195` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Hairong Chen, Yicheng Feng, Ziyu Jia, et al.
+- **abs:** https://arxiv.org/abs/2602.18195 · **PDF:** https://arxiv.org/pdf/2602.18195v1
+
+## 10927. Box Thirding: Anytime Best Arm Identification under Insufficient Sampling
+
+- **ID:** `2602.18186` · **primary:** `stat.ML` · **Submitted:** 2026-02-20
+- **Authors:** Seohwa Hwang, Junyong Park
+- **abs:** https://arxiv.org/abs/2602.18186 · **PDF:** https://arxiv.org/pdf/2602.18186v1
+
+## 10928. Capabilities Ain't All You Need: Measuring Propensities in AI
+
+- **ID:** `2602.18182` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Daniel Romero-Alvarado, Fernando Martínez-Plumed, Lorenzo Pacchiardi, et al.
+- **abs:** https://arxiv.org/abs/2602.18182 · **PDF:** https://arxiv.org/pdf/2602.18182v4
+
+## 10929. SeedFlood: A Step Toward Scalable Decentralized Training of LLMs
+
+- **ID:** `2602.18181` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Jihun Kim, Namhoon Lee
+- **abs:** https://arxiv.org/abs/2602.18181 · **PDF:** https://arxiv.org/pdf/2602.18181v1
+
+## 10930. Improving Sampling for Masked Diffusion Models via Information Gain
+
+- **ID:** `2602.18176` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Kaisen Yang, Jayden Teoh, Kaicheng Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.18176 · **PDF:** https://arxiv.org/pdf/2602.18176v2
+
+## 10931. Can AI Lower the Barrier to Cybersecurity? A Human-Centered Mixed-Methods Study of Novice CTF Learning
+
+- **ID:** `2602.18172` · **primary:** `cs.CR` · **Submitted:** 2026-02-20
+- **Authors:** Cathrin Schachner, Jasmin Wachter
+- **abs:** https://arxiv.org/abs/2602.18172 · **PDF:** https://arxiv.org/pdf/2602.18172v1
+
+## 10932. Click it or Leave it: Detecting and Spoiling Clickbait with Informativeness Measures and Large Language Models
+
+- **ID:** `2602.18171` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Wojciech Michaluk, Tymoteusz Urban, Mateusz Kubita, et al.
+- **abs:** https://arxiv.org/abs/2602.18171 · **PDF:** https://arxiv.org/pdf/2602.18171v1
+
+## 10933. A Deep Surrogate Model for Robust and Generalizable Long-Term Blast Wave Prediction
+
+- **ID:** `2602.18168` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Danning Jing, Xinhai Chen, Xifeng Pu, et al.
+- **abs:** https://arxiv.org/abs/2602.18168 · **PDF:** https://arxiv.org/pdf/2602.18168v2
+
+## 10934. Agentic AI, Retrieval-Augmented Generation, and the Institutional Turn: Legal Architectures and Financial Governance in the Age of Distributional AGI
+
+- **ID:** `2603.13244` · **primary:** `cs.CY` · **Submitted:** 2026-02-20
+- **Authors:** Marcel Osmond
+- **abs:** https://arxiv.org/abs/2603.13244 · **PDF:** https://arxiv.org/pdf/2603.13244v1
+
+## 10935. Triggering hallucinations in model-based MRI reconstruction via adversarial perturbations
+
+- **ID:** `2602.18536` · **primary:** `eess.IV` · **Submitted:** 2026-02-20
+- **Authors:** Suna Buğday, Yvan Saeys, Jonathan Peck
+- **abs:** https://arxiv.org/abs/2602.18536 · **PDF:** https://arxiv.org/pdf/2602.18536v1
+
+## 10936. Unifying Formal Explanations: A Complexity-Theoretic Perspective
+
+- **ID:** `2602.18160` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Shahaf Bassan, Xuanxiang Huang, Guy Katz
+- **abs:** https://arxiv.org/abs/2602.18160 · **PDF:** https://arxiv.org/pdf/2602.18160v1
+
+## 10937. FENCE: A Financial and Multimodal Jailbreak Detection Dataset
+
+- **ID:** `2602.18154` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Mirae Kim, Seonghun Jeong, Youngjun Kwak
+- **abs:** https://arxiv.org/abs/2602.18154 · **PDF:** https://arxiv.org/pdf/2602.18154v1
+
+## 10938. Do Large Language Models Possess a Theory of Mind? A Comparative Evaluation Using the Strange Stories Paradigm
+
+- **ID:** `2603.18007` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Anna Babarczy, Andras Lukacs, Peter Vedres, et al.
+- **abs:** https://arxiv.org/abs/2603.18007 · **PDF:** https://arxiv.org/pdf/2603.18007v1
+
+## 10939. The Statistical Signature of LLMs
+
+- **ID:** `2602.18152` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Ortal Hadad, Edoardo Loru, Jacopo Nudo, et al.
+- **abs:** https://arxiv.org/abs/2602.18152 · **PDF:** https://arxiv.org/pdf/2602.18152v1
+
+## 10940. Rethinking Beam Management: Generalization Limits Under Hardware Heterogeneity
+
+- **ID:** `2602.18151` · **primary:** `cs.NI` · **Submitted:** 2026-02-20
+- **Authors:** Nikita Zeulin, Olga Galinina, Ibrahim Kilinc, et al.
+- **abs:** https://arxiv.org/abs/2602.18151 · **PDF:** https://arxiv.org/pdf/2602.18151v1
+
+## 10941. Gradient-Informed Bayesian and Interior Point Optimization for Efficient Inverse Design in Nanophotonics
+
+- **ID:** `2602.18148` · **primary:** `physics.optics` · **Submitted:** 2026-02-20
+- **Authors:** Yannik Mahlau, Yannick Augenstein, Tyler W. Hughes, et al.
+- **abs:** https://arxiv.org/abs/2602.18148 · **PDF:** https://arxiv.org/pdf/2602.18148v2
+
+## 10942. Stable Long-Horizon Spatiotemporal Prediction on Meshes Using Latent Multiscale Recurrent Graph Neural Networks
+
+- **ID:** `2602.18146` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Lionel Salesses, Larbi Arbaoui, Tariq Benamara, et al.
+- **abs:** https://arxiv.org/abs/2602.18146 · **PDF:** https://arxiv.org/pdf/2602.18146v1
+
+## 10943. Detecting Contextual Hallucinations in LLMs with Frequency-Aware Attention
+
+- **ID:** `2602.18145` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Siya Qi, Yudong Chen, Runcong Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.18145 · **PDF:** https://arxiv.org/pdf/2602.18145v1
+
+## 10944. Fairness-Aware Partial-label Domain Adaptation for Voice Classification of Parkinson's and ALS
+
+- **ID:** `2602.18535` · **primary:** `cs.SD` · **Submitted:** 2026-02-20
+- **Authors:** Arianna Francesconi, Zhixiang Dai, Arthur Stefano Moscheni, et al.
+- **abs:** https://arxiv.org/abs/2602.18535 · **PDF:** https://arxiv.org/pdf/2602.18535v1
+
+## 10945. Advection-Diffusion on Graphs: A Bakry-Emery Laplacian for Spectral Graph Neural Networks
+
+- **ID:** `2602.18141` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Pierre-Gabriel Berlureau, Ali Hariri, Victor Kawasaki-Borruat, et al.
+- **abs:** https://arxiv.org/abs/2602.18141 · **PDF:** https://arxiv.org/pdf/2602.18141v1
+
+## 10946. Flexi-NeurA: A Configurable Neuromorphic Accelerator with Adaptive Bit-Precision Exploration for Edge SNNs
+
+- **ID:** `2602.18140` · **primary:** `cs.AR` · **Submitted:** 2026-02-20
+- **Authors:** Mohammad Farahani, Mohammad Rasoul Roshanshah, Saeed Safari
+- **abs:** https://arxiv.org/abs/2602.18140 · **PDF:** https://arxiv.org/pdf/2602.18140v1
+
+## 10947. Agentic Adversarial QA for Improving Domain-Specific LLMs
+
+- **ID:** `2602.18137` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Vincent Grari, Ciprian Tomoiaga, Sylvain Lamprier, et al.
+- **abs:** https://arxiv.org/abs/2602.18137 · **PDF:** https://arxiv.org/pdf/2602.18137v1
+
+## 10948. Learning Long-Range Dependencies with Temporal Predictive Coding
+
+- **ID:** `2602.18131` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Tom Potter, Oliver Rhodes
+- **abs:** https://arxiv.org/abs/2602.18131 · **PDF:** https://arxiv.org/pdf/2602.18131v1
+
+## 10949. RamanSeg: Interpretability-driven Deep Learning on Raman Spectra for Cancer Diagnosis
+
+- **ID:** `2602.18119` · **primary:** `eess.IV` · **Submitted:** 2026-02-20
+- **Authors:** Chris Tomy, Mo Vali, David Pertzborn, et al.
+- **abs:** https://arxiv.org/abs/2602.18119 · **PDF:** https://arxiv.org/pdf/2602.18119v1
+
+## 10950. Flow Matching with Injected Noise for Offline-to-Online Reinforcement Learning
+
+- **ID:** `2602.18117` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Yongjae Shin, Jongseong Chae, Jongeui Park, et al.
+- **abs:** https://arxiv.org/abs/2602.18117 · **PDF:** https://arxiv.org/pdf/2602.18117v1
+
+## 10951. Cut Less, Fold More: Model Compression through the Lens of Projection Geometry
+
+- **ID:** `2602.18116` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Olga Saukh, Dong Wang, Haris Šikić, et al.
+- **abs:** https://arxiv.org/abs/2602.18116 · **PDF:** https://arxiv.org/pdf/2602.18116v1
+
+## 10952. Non-Stationary Online Resource Allocation: Learning from a Single Sample
+
+- **ID:** `2602.18114` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Yiding Feng, Jiashuo Jiang, Yige Wang
+- **abs:** https://arxiv.org/abs/2602.18114 · **PDF:** https://arxiv.org/pdf/2602.18114v1
+
+## 10953. TempoNet: Slack-Quantized Transformer-Guided Reinforcement Scheduler for Adaptive Deadline-Centric Real-Time Dispatchs
+
+- **ID:** `2602.18109` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Rong Fu, Yibo Meng, Guangzhen Yao, et al.
+- **abs:** https://arxiv.org/abs/2602.18109 · **PDF:** https://arxiv.org/pdf/2602.18109v1
+
+## 10954. Self-hosted Lecture-to-Quiz: Local LLM MCQ Generation with Deterministic Quality Control
+
+- **ID:** `2603.08729` · **primary:** `cs.CY` · **Submitted:** 2026-02-20
+- **Authors:** Seine A. Shintani
+- **abs:** https://arxiv.org/abs/2603.08729 · **PDF:** https://arxiv.org/pdf/2603.08729v1
+
+## 10955. Think First, Diffuse Fast: Improving Diffusion Language Model Reasoning via Autoregressive Plan Conditioning
+
+- **ID:** `2603.13243` · **primary:** `cs.AI` · **Submitted:** 2026-02-20
+- **Authors:** Earl J St Sauver
+- **abs:** https://arxiv.org/abs/2603.13243 · **PDF:** https://arxiv.org/pdf/2603.13243v1
+
+## 10956. MeanVoiceFlow: One-step Nonparallel Voice Conversion with Mean Flows
+
+- **ID:** `2602.18104` · **primary:** `cs.SD` · **Submitted:** 2026-02-20
+- **Authors:** Takuhiro Kaneko, Hirokazu Kameoka, Kou Tanaka, et al.
+- **abs:** https://arxiv.org/abs/2602.18104 · **PDF:** https://arxiv.org/pdf/2602.18104v1
+
+## 10957. Interacting safely with cyclists using Hamilton-Jacobi reachability and reinforcement learning
+
+- **ID:** `2602.18097` · **primary:** `cs.RO` · **Submitted:** 2026-02-20
+- **Authors:** Aarati Andrea Noronha, Jean Oh
+- **abs:** https://arxiv.org/abs/2602.18097 · **PDF:** https://arxiv.org/pdf/2602.18097v1
+
+## 10958. Neurosymbolic Language Reasoning as Satisfiability Modulo Theory
+
+- **ID:** `2602.18095` · **primary:** `cs.AI` · **Submitted:** 2026-02-20
+- **Authors:** Hyunseok Oh, Sam Stern, Youngki Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.18095 · **PDF:** https://arxiv.org/pdf/2602.18095v1
+
+## 10959. OODBench: Out-of-Distribution Benchmark for Large Vision-Language Models
+
+- **ID:** `2602.18094` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Ling Lin, Yang Bai, Heng Su, et al.
+- **abs:** https://arxiv.org/abs/2602.18094 · **PDF:** https://arxiv.org/pdf/2602.18094v1
+
+## 10960. Perceived Political Bias in LLMs Reduces Persuasive Abilities
+
+- **ID:** `2602.18092` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Matthew DiGiuseppe, Joshua Robison
+- **abs:** https://arxiv.org/abs/2602.18092 · **PDF:** https://arxiv.org/pdf/2602.18092v1
+
+## 10961. VLANeXt: Recipes for Building Strong VLA Models
+
+- **ID:** `2602.18532` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Xiao-Ming Wu, Bin Fan, Kang Liao, et al.
+- **abs:** https://arxiv.org/abs/2602.18532 · **PDF:** https://arxiv.org/pdf/2602.18532v1
+
+## 10962. DohaScript: A Large-Scale Multi-Writer Dataset for Continuous Handwritten Hindi Text
+
+- **ID:** `2602.18089` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Kunwar Arpit Singh, Ankush Prakash, Haroon R Lone
+- **abs:** https://arxiv.org/abs/2602.18089 · **PDF:** https://arxiv.org/pdf/2602.18089v1
+
+## 10963. Balancing Symmetry and Efficiency in Graph Flow Matching
+
+- **ID:** `2602.18084` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Benjamin Honoré, Alba Carballo-Castro, Yiming Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.18084 · **PDF:** https://arxiv.org/pdf/2602.18084v1
+
+## 10964. Comparative Assessment of Multimodal Earth Observation Data for Soil Moisture Estimation
+
+- **ID:** `2602.18083` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Ioannis Kontogiorgakis, Athanasios Askitopoulos, Iason Tsardanidis, et al.
+- **abs:** https://arxiv.org/abs/2602.18083 · **PDF:** https://arxiv.org/pdf/2602.18083v1
+
+## 10965. HiAER-Spike Software-Hardware Reconfigurable Platform for Event-Driven Neuromorphic Computing at Scale
+
+- **ID:** `2602.18072` · **primary:** `cs.AR` · **Submitted:** 2026-02-20
+- **Authors:** Gwenevere Frank, Gopabandhu Hota, Keli Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.18072 · **PDF:** https://arxiv.org/pdf/2602.18072v1
+
+## 10966. Deepmechanics
+
+- **ID:** `2602.18060` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Abhay Shinde, Aryan Amit Barsainyan, Jose Siguenza, et al.
+- **abs:** https://arxiv.org/abs/2602.18060 · **PDF:** https://arxiv.org/pdf/2602.18060v1
+
+## 10967. RACER: Risk-Aware Calibrated Efficient Routing for Large Language Models
+
+- **ID:** `2603.06616` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Sai Hao, Hao Zeng, Hongxin Wei, et al.
+- **abs:** https://arxiv.org/abs/2603.06616 · **PDF:** https://arxiv.org/pdf/2603.06616v1
+
+## 10968. Continual-NExT: A Unified Comprehension And Generation Continual Learning Framework
+
+- **ID:** `2602.18055` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Jingyang Qiao, Zhizhong Zhang, Xin Tan, et al.
+- **abs:** https://arxiv.org/abs/2602.18055 · **PDF:** https://arxiv.org/pdf/2602.18055v1
+
+## 10969. On the Generalization and Robustness in Conditional Value-at-Risk
+
+- **ID:** `2602.18053` · **primary:** `stat.ML` · **Submitted:** 2026-02-20
+- **Authors:** Dinesh Karthik Mulumudi, Piyushi Manupriya, Gholamali Aminian, et al.
+- **abs:** https://arxiv.org/abs/2602.18053 · **PDF:** https://arxiv.org/pdf/2602.18053v1
+
+## 10970. CityGuard: Graph-Aware Private Descriptors for Bias-Resilient Identity Search Across Urban Cameras
+
+- **ID:** `2602.18047` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Rong Fu, Yibo Meng, Jia Yee Tan, et al.
+- **abs:** https://arxiv.org/abs/2602.18047 · **PDF:** https://arxiv.org/pdf/2602.18047v2
+
+## 10971. Conformal Tradeoffs: Operational Profiles Beyond Coverage
+
+- **ID:** `2602.18045` · **primary:** `stat.ME` · **Submitted:** 2026-02-20
+- **Authors:** Petrus H. Zwart
+- **abs:** https://arxiv.org/abs/2602.18045 · **PDF:** https://arxiv.org/pdf/2602.18045v3
+
+## 10972. Deep Reinforcement Learning for Optimizing Energy Consumption in Smart Grid Systems
+
+- **ID:** `2602.18531` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Abeer Alsheikhi, Amirfarhad Farhadi, Azadeh Zamanifar
+- **abs:** https://arxiv.org/abs/2602.18531 · **PDF:** https://arxiv.org/pdf/2602.18531v1
+
+## 10973. PINEAPPLE: Physics-Informed Neuro-Evolution Algorithm for Prognostic Parameter Inference in Lithium-Ion Battery Electrodes
+
+- **ID:** `2602.18042` · **primary:** `cs.CE` · **Submitted:** 2026-02-20
+- **Authors:** Karkulali Pugalenthi, Jian Cheng Wong, Qizheng Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.18042 · **PDF:** https://arxiv.org/pdf/2602.18042v1
+
+## 10974. Annealed Co-Generation: Disentangling Variables via Progressive Pairwise Modeling
+
+- **ID:** `2603.06615` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Hantao Zhang, Jieke Wu, Mingda Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.06615 · **PDF:** https://arxiv.org/pdf/2603.06615v1
+
+## 10975. Gradient Regularization Prevents Reward Hacking in Reinforcement Learning from Human Feedback and Verifiable Rewards
+
+- **ID:** `2602.18037` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Johannes Ackermann, Michael Noukhovitch, Takashi Ishida, et al.
+- **abs:** https://arxiv.org/abs/2602.18037 · **PDF:** https://arxiv.org/pdf/2602.18037v1
+
+## 10976. CogGen: Cognitive-Load-Informed Fully Unsupervised Deep Generative Modeling for Compressively Sampled MRI Reconstruction
+
+- **ID:** `2603.04438` · **primary:** `eess.IV` · **Submitted:** 2026-02-20
+- **Authors:** Qingyong Zhu, Yumin Tan, Xiang Gu, et al.
+- **abs:** https://arxiv.org/abs/2603.04438 · **PDF:** https://arxiv.org/pdf/2603.04438v2
+
+## 10977. Towards More Standardized AI Evaluation: From Models to Agents
+
+- **ID:** `2602.18029` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Ali El Filali, Inès Bedar
+- **abs:** https://arxiv.org/abs/2602.18029 · **PDF:** https://arxiv.org/pdf/2602.18029v1
+
+## 10978. Mean-Field Reinforcement Learning without Synchrony
+
+- **ID:** `2602.18026` · **primary:** `cs.MA` · **Submitted:** 2026-02-20
+- **Authors:** Shan Yang
+- **abs:** https://arxiv.org/abs/2602.18026 · **PDF:** https://arxiv.org/pdf/2602.18026v1
+
+## 10979. Cross-Embodiment Offline Reinforcement Learning for Heterogeneous Robot Datasets
+
+- **ID:** `2602.18025` · **primary:** `cs.AI` · **Submitted:** 2026-02-20
+- **Authors:** Haruki Abe, Takayuki Osa, Yusuke Mukuta, et al.
+- **abs:** https://arxiv.org/abs/2602.18025 · **PDF:** https://arxiv.org/pdf/2602.18025v1
+
+## 10980. Correlation Analysis of Generative Models
+
+- **ID:** `2603.06614` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Zhengguo Li, Chaobing Zheng, Wei Wang
+- **abs:** https://arxiv.org/abs/2603.06614 · **PDF:** https://arxiv.org/pdf/2603.06614v1
+
+## 10981. Dual-Channel Attention Guidance for Training-Free Image Editing Control in Diffusion Transformers
+
+- **ID:** `2602.18022` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Guandong Li
+- **abs:** https://arxiv.org/abs/2602.18022 · **PDF:** https://arxiv.org/pdf/2602.18022v2
+
+## 10982. DeepSVU: Towards In-depth Security-oriented Video Understanding via Unified Physical-world Regularized MoE
+
+- **ID:** `2602.18019` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Yujie Jin, Wenxin Zhang, Jingjing Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.18019 · **PDF:** https://arxiv.org/pdf/2602.18019v1
+
+## 10983. Flow Actor-Critic for Offline Reinforcement Learning
+
+- **ID:** `2602.18015` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Jongseong Chae, Jongeui Park, Yongjae Shin, et al.
+- **abs:** https://arxiv.org/abs/2602.18015 · **PDF:** https://arxiv.org/pdf/2602.18015v1
+
+## 10984. Quasi-Periodic Gaussian Process Predictive Iterative Learning Control
+
+- **ID:** `2602.18014` · **primary:** `cs.RO` · **Submitted:** 2026-02-20
+- **Authors:** Unnati Nigam, Radhendushka Srivastava, Faezeh Marzbanrad, et al.
+- **abs:** https://arxiv.org/abs/2602.18014 · **PDF:** https://arxiv.org/pdf/2602.18014v1
+
+## 10985. OptiRoulette Optimizer: A New Stochastic Meta-Optimizer for up to 5.3x Faster Convergence
+
+- **ID:** `2603.06613` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Stamatis Mastromichalakis
+- **abs:** https://arxiv.org/abs/2603.06613 · **PDF:** https://arxiv.org/pdf/2603.06613v2
+
+## 10986. NIMMGen: Learning Neural-Integrated Mechanistic Digital Twins with LLMs
+
+- **ID:** `2602.18008` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Zihan Guan, Rituparna Datta, Mengxuan Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.18008 · **PDF:** https://arxiv.org/pdf/2602.18008v1
+
+## 10987. Preconditioned Robust Neural Posterior Estimation for Misspecified Simulators
+
+- **ID:** `2602.18004` · **primary:** `stat.ME` · **Submitted:** 2026-02-20
+- **Authors:** Ryan P. Kelly, David T. Frazier, David J. Warne, et al.
+- **abs:** https://arxiv.org/abs/2602.18004 · **PDF:** https://arxiv.org/pdf/2602.18004v1
+
+## 10988. Asynchronous Heavy-Tailed Optimization
+
+- **ID:** `2602.18002` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Junfei Sun, Dixi Yao, Xuchen Gong, et al.
+- **abs:** https://arxiv.org/abs/2602.18002 · **PDF:** https://arxiv.org/pdf/2602.18002v1
+
+## 10989. Aurora: Neuro-Symbolic AI Driven Advising Agent
+
+- **ID:** `2602.17999` · **primary:** `cs.HC` · **Submitted:** 2026-02-20
+- **Authors:** Lorena Amanda Quincoso Lugones, Christopher Kverne, Nityam Sharadkumar Bhimani, et al.
+- **abs:** https://arxiv.org/abs/2602.17999 · **PDF:** https://arxiv.org/pdf/2602.17999v1
+
+## 10990. PHAST: Port-Hamiltonian Architecture for Structured Temporal Dynamics Forecasting
+
+- **ID:** `2602.17998` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Shubham Bhardwaj, Chandrajit Bajaj
+- **abs:** https://arxiv.org/abs/2602.17998 · **PDF:** https://arxiv.org/pdf/2602.17998v1
+
+## 10991. Whole-Brain Connectomic Graph Model Enables Whole-Body Locomotion Control in Fruit Fly
+
+- **ID:** `2602.17997` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Zehao Jin, Yaoye Zhu, Chen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.17997 · **PDF:** https://arxiv.org/pdf/2602.17997v2
+
+## 10992. Turbo Connection: Reasoning as Information Flow from Higher to Lower Layers
+
+- **ID:** `2602.17993` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Mohan Tang, Sidi Lu
+- **abs:** https://arxiv.org/abs/2602.17993 · **PDF:** https://arxiv.org/pdf/2602.17993v1
+
+## 10993. Audio-Visual Continual Test-Time Adaptation without Forgetting
+
+- **ID:** `2602.18528` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Sarthak Kumar Maharana, Akshay Mehra, Bhavya Ramakrishna, et al.
+- **abs:** https://arxiv.org/abs/2602.18528 · **PDF:** https://arxiv.org/pdf/2602.18528v1
+
+## 10994. WorkflowPerturb: Calibrated Stress Tests for Evaluating Multi-Agent Workflow Metrics
+
+- **ID:** `2602.17990` · **primary:** `cs.AI` · **Submitted:** 2026-02-20
+- **Authors:** Madhav Kanda, Pedro Las-Casas, Alok Gautam Kumbhare, et al.
+- **abs:** https://arxiv.org/abs/2602.17990 · **PDF:** https://arxiv.org/pdf/2602.17990v1
+
+## 10995. Learning Without Training
+
+- **ID:** `2602.17985` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Ryan O'Dowd
+- **abs:** https://arxiv.org/abs/2602.17985 · **PDF:** https://arxiv.org/pdf/2602.17985v1
+
+## 10996. Decomposing Retrieval Failures in RAG for Long-Document Financial Question Answering
+
+- **ID:** `2602.17981` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Amine Kobeissi, Philippe Langlais
+- **abs:** https://arxiv.org/abs/2602.17981 · **PDF:** https://arxiv.org/pdf/2602.17981v1
+
+## 10997. Learning Optimal and Sample-Efficient Decision Policies with Guarantees
+
+- **ID:** `2602.17978` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Daqian Shao
+- **abs:** https://arxiv.org/abs/2602.17978 · **PDF:** https://arxiv.org/pdf/2602.17978v1
+
+## 10998. In-Context Learning for Pure Exploration in Continuous Spaces
+
+- **ID:** `2602.17976` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Alessio Russo, Yin-Ching Lee, Ryan Welch, et al.
+- **abs:** https://arxiv.org/abs/2602.17976 · **PDF:** https://arxiv.org/pdf/2602.17976v1
+
+## 10999. Generating adversarial inputs for a graph neural network model of AC power flow
+
+- **ID:** `2602.17975` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Robert Parker
+- **abs:** https://arxiv.org/abs/2602.17975 · **PDF:** https://arxiv.org/pdf/2602.17975v1
+
+## 11000. JAEGER: Joint 3D Audio-Visual Grounding and Reasoning in Simulated Physical Environments
+
+- **ID:** `2602.18527` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Zhan Liu, Changli Tang, Yuxin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.18527 · **PDF:** https://arxiv.org/pdf/2602.18527v1
+
+## 11001. PenTiDef: Enhancing Privacy and Robustness in Decentralized Federated Intrusion Detection Systems against Poisoning Attacks
+
+- **ID:** `2602.17973` · **primary:** `cs.CR` · **Submitted:** 2026-02-20
+- **Authors:** Phan The Duy, Nghi Hoang Khoa, Nguyen Tran Anh Quan, et al.
+- **abs:** https://arxiv.org/abs/2602.17973 · **PDF:** https://arxiv.org/pdf/2602.17973v1
+
+## 11002. Student Flow Modeling for School Decongestion via Stochastic Gravity Estimation and Constrained Spatial Allocation
+
+- **ID:** `2602.17972` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Sebastian Felipe R. Bundoc, Paula Joy B. Martinez, Sebastian C. Ibañez, et al.
+- **abs:** https://arxiv.org/abs/2602.17972 · **PDF:** https://arxiv.org/pdf/2602.17972v1
+
+## 11003. Minimax optimal adaptive structured transfer learning through semi-parametric domain-varying coefficient model
+
+- **ID:** `2602.17967` · **primary:** `math.ST` · **Submitted:** 2026-02-20
+- **Authors:** Hanxiao Chen, Debarghya Mukherjee
+- **abs:** https://arxiv.org/abs/2602.17967 · **PDF:** https://arxiv.org/pdf/2602.17967v1
+
+## 11004. Improving Generalizability of Hip Fracture Risk Prediction via Domain Adaptation Across Multiple Cohorts
+
+- **ID:** `2602.17962` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Shuo Sun, Meiling Zhou, Chen Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.17962 · **PDF:** https://arxiv.org/pdf/2602.17962v1
+
+## 11005. Consensus is Not Verification: Why Crowd Wisdom Strategies Fail for LLM Truthfulness
+
+- **ID:** `2603.06612` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Yegor Denisov-Blanch, Joshua Kazdan, Jessica Chudnovsky, et al.
+- **abs:** https://arxiv.org/abs/2603.06612 · **PDF:** https://arxiv.org/pdf/2603.06612v1
+
+## 11006. Anisotropic local law for non-separable sample covariance matrices
+
+- **ID:** `2602.17960` · **primary:** `math.PR` · **Submitted:** 2026-02-20
+- **Authors:** Zhou Fan, Renyuan Ma, Elliot Paquette, et al.
+- **abs:** https://arxiv.org/abs/2602.17960 · **PDF:** https://arxiv.org/pdf/2602.17960v2
+
+## 11007. Bayesian Online Model Selection
+
+- **ID:** `2602.17958` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Aida Afshar, Yuke Zhang, Aldo Pacchiano
+- **abs:** https://arxiv.org/abs/2602.17958 · **PDF:** https://arxiv.org/pdf/2602.17958v1
+
+## 11008. Hardware-Friendly Input Expansion for Accelerating Function Approximation
+
+- **ID:** `2602.17952` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Hu Lou, Yin-Jun Gao, Dong-Xiao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.17952 · **PDF:** https://arxiv.org/pdf/2602.17952v1
+
+## 11009. ROCKET: Residual-Oriented Multi-Layer Alignment for Spatially-Aware Vision-Language-Action Models
+
+- **ID:** `2602.17951` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Guoheng Sun, Tingting Du, Kaixi Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.17951 · **PDF:** https://arxiv.org/pdf/2602.17951v1
+
+## 11010. Do Generative Metrics Predict YOLO Performance? An Evaluation Across Models, Augmentation Ratios, and Dataset Complexity
+
+- **ID:** `2602.18525` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Vasile Marian, Yong-Bin Kang, Alexander Buddery
+- **abs:** https://arxiv.org/abs/2602.18525 · **PDF:** https://arxiv.org/pdf/2602.18525v1
+
+## 11011. CUICurate: A GraphRAG-based Framework for Automated Clinical Concept Curation for NLP applications
+
+- **ID:** `2602.17949` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Victoria Blake, Mathew Miller, Jamie Novak, et al.
+- **abs:** https://arxiv.org/abs/2602.17949 · **PDF:** https://arxiv.org/pdf/2602.17949v1
+
+## 11012. A Geometric Probe of the Accuracy-Robustness Trade-off: Sharp Boundaries in Symmetry-Breaking Dimensional Expansion
+
+- **ID:** `2602.17948` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Yu Bai, Zhe Wang, Jiarui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.17948 · **PDF:** https://arxiv.org/pdf/2602.17948v1
+
+## 11013. Understanding the Generalization of Bilevel Programming in Hyperparameter Optimization: A Tale of Bias-Variance Decomposition
+
+- **ID:** `2602.17947` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Yubo Zhou, Jun Shu, Junmin Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.17947 · **PDF:** https://arxiv.org/pdf/2602.17947v1
+
+## 11014. Optimizing Graph Causal Classification Models: Estimating Causal Effects and Addressing Confounders
+
+- **ID:** `2602.17941` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Simi Job, Xiaohui Tao, Taotao Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.17941 · **PDF:** https://arxiv.org/pdf/2602.17941v1
+
+## 11015. Tighter Regret Lower Bound for Gaussian Process Bandits with Squared Exponential Kernel in Hypersphere
+
+- **ID:** `2602.17940` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Shogo Iwazaki
+- **abs:** https://arxiv.org/abs/2602.17940 · **PDF:** https://arxiv.org/pdf/2602.17940v1
+
+## 11016. Automating the Analysis and Improvement of Dynamic Programming Algorithms with Applications to Natural Language Processing
+
+- **ID:** `2603.13242` · **primary:** `cs.PL` · **Submitted:** 2026-02-20
+- **Authors:** Tim Vieira
+- **abs:** https://arxiv.org/abs/2603.13242 · **PDF:** https://arxiv.org/pdf/2603.13242v1
+
+## 11017. Analyzing LLM Instruction Optimization for Tabular Fact Verification
+
+- **ID:** `2602.17937` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Xiaotang Du, Giwon Hong, Wai-Chung Kwan, et al.
+- **abs:** https://arxiv.org/abs/2602.17937 · **PDF:** https://arxiv.org/pdf/2602.17937v1
+
+## 11018. Causal Neighbourhood Learning for Invariant Graph Representations
+
+- **ID:** `2602.17934` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Simi Job, Xiaohui Tao, Taotao Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.17934 · **PDF:** https://arxiv.org/pdf/2602.17934v1
+
+## 11019. Memory-Based Advantage Shaping for LLM-Guided Reinforcement Learning
+
+- **ID:** `2602.17931` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Narjes Nourzad, Carlee Joe-Wong
+- **abs:** https://arxiv.org/abs/2602.17931 · **PDF:** https://arxiv.org/pdf/2602.17931v1
+
+## 11020. MIRA: Memory-Integrated Reinforcement Learning Agent with Limited LLM Guidance
+
+- **ID:** `2602.17930` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Narjes Nourzad, Carlee Joe-Wong
+- **abs:** https://arxiv.org/abs/2602.17930 · **PDF:** https://arxiv.org/pdf/2602.17930v1
+
+## 11021. ZACH-ViT: Regime-Dependent Inductive Bias in Compact Vision Transformers for Medical Imaging
+
+- **ID:** `2602.17929` · **primary:** `cs.CV` · **Submitted:** 2026-02-20
+- **Authors:** Athanasios Angelakis
+- **abs:** https://arxiv.org/abs/2602.17929 · **PDF:** https://arxiv.org/pdf/2602.17929v2
+
+## 11022. A Novel Approach for Testing Water Safety Using Deep Learning Inference of Microscopic Images of Unincubated Water Samples
+
+- **ID:** `2603.06611` · **primary:** `cs.OH` · **Submitted:** 2026-02-20
+- **Authors:** Sanjay Srinivasan
+- **abs:** https://arxiv.org/abs/2603.06611 · **PDF:** https://arxiv.org/pdf/2603.06611v1
+
+## 11023. Data-driven configuration tuning of glmnet for balancing accuracy and computational efficiency
+
+- **ID:** `2602.17922` · **primary:** `stat.CO` · **Submitted:** 2026-02-20
+- **Authors:** Shuhei Muroya, Kei Hirose
+- **abs:** https://arxiv.org/abs/2602.17922 · **PDF:** https://arxiv.org/pdf/2602.17922v2
+
+## 11024. Latent Diffeomorphic Co-Design of End-Effectors for Deformable and Fragile Object Manipulation
+
+- **ID:** `2602.17921` · **primary:** `cs.RO` · **Submitted:** 2026-02-20
+- **Authors:** Kei Ikemura, Yifei Dong, Florian T. Pokorny
+- **abs:** https://arxiv.org/abs/2602.17921 · **PDF:** https://arxiv.org/pdf/2602.17921v1
+
+## 11025. Distribution-Free Sequential Prediction with Abstentions
+
+- **ID:** `2602.17918` · **primary:** `cs.LG` · **Submitted:** 2026-02-20
+- **Authors:** Jialin Yu, Moïse Blanchard
+- **abs:** https://arxiv.org/abs/2602.17918 · **PDF:** https://arxiv.org/pdf/2602.17918v2
+
+## 11026. Interactions that reshape the interfaces of the interacting parties
+
+- **ID:** `2602.17917` · **primary:** `math.CT` · **Submitted:** 2026-02-20
+- **Authors:** David I. Spivak
+- **abs:** https://arxiv.org/abs/2602.17917 · **PDF:** https://arxiv.org/pdf/2602.17917v1
+
+## 11027. From Lossy to Verified: A Provenance-Aware Tiered Memory for Agents
+
+- **ID:** `2602.17913` · **primary:** `cs.DB` · **Submitted:** 2026-02-20
+- **Authors:** Qiming Zhu, Shunian Chen, Rui Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.17913 · **PDF:** https://arxiv.org/pdf/2602.17913v1
+
+## 11028. Condition-Gated Reasoning for Context-Dependent Biomedical Question Answering
+
+- **ID:** `2602.17911` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Jash Rajesh Parekh, Wonbin Kweon, Joey Chan, et al.
+- **abs:** https://arxiv.org/abs/2602.17911 · **PDF:** https://arxiv.org/pdf/2602.17911v2
+
+## 11029. Alignment in Time: Peak-Aware Orchestration for Long-Horizon Agentic Systems
+
+- **ID:** `2602.17910` · **primary:** `cs.AI` · **Submitted:** 2026-02-20
+- **Authors:** Hanjing Shi, Dominic DiFranzo
+- **abs:** https://arxiv.org/abs/2602.17910 · **PDF:** https://arxiv.org/pdf/2602.17910v1
+
+## 11030. Improving Neural Topic Modeling with Semantically-Grounded Soft Label Distributions
+
+- **ID:** `2602.17907` · **primary:** `cs.CL` · **Submitted:** 2026-02-20
+- **Authors:** Raymond Li, Amirhossein Abaskohi, Chuyuan Li, et al.
+- **abs:** https://arxiv.org/abs/2602.17907 · **PDF:** https://arxiv.org/pdf/2602.17907v1
+
+## 11031. Games That Teach, Chats That Convince: Comparing Interactive and Static Formats for Persuasive Learning
+
+- **ID:** `2602.17905` · **primary:** `cs.HC` · **Submitted:** 2026-02-20
+- **Authors:** Seyed Hossein Alavi, Zining Wang, Shruthi Chockkalingam, et al.
+- **abs:** https://arxiv.org/abs/2602.17905 · **PDF:** https://arxiv.org/pdf/2602.17905v2
+
+## 11032. El Agente Gráfico: Structured Execution Graphs for Scientific Agents
+
+- **ID:** `2602.17902` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Jiaru Bai, Abdulrahman Aldossary, Thomas Swanick, et al.
+- **abs:** https://arxiv.org/abs/2602.17902 · **PDF:** https://arxiv.org/pdf/2602.17902v1
+
+## 11033. Breaking the Correlation Plateau: On the Optimization and Capacity Limits of Attention-Based Regressors
+
+- **ID:** `2602.17898` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Jingquan Yan, Yuwei Miao, Peiran Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.17898 · **PDF:** https://arxiv.org/pdf/2602.17898v1
+
+## 11034. Learning from Biased and Costly Data Sources: Minimax-optimal Data Collection under a Budget
+
+- **ID:** `2602.17894` · **primary:** `stat.ML` · **Submitted:** 2026-02-19
+- **Authors:** Michael O. Harding, Vikas Singh, Kirthevasan Kandasamy
+- **abs:** https://arxiv.org/abs/2602.17894 · **PDF:** https://arxiv.org/pdf/2602.17894v1
+
+## 11035. COMBA: Cross Batch Aggregation for Learning Large Graphs with Context Gating State Space Models
+
+- **ID:** `2602.17893` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Jiajun Shen, Yufei Jin, Yi He, et al.
+- **abs:** https://arxiv.org/abs/2602.17893 · **PDF:** https://arxiv.org/pdf/2602.17893v1
+
+## 11036. Machine Learning Based Prediction of Surgical Outcomes in Chronic Rhinosinusitis from Clinical Data
+
+- **ID:** `2602.17888` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Sayeed Shafayet Chowdhury, Karen D'Souza, V. Siva Kakumani, et al.
+- **abs:** https://arxiv.org/abs/2602.17888 · **PDF:** https://arxiv.org/pdf/2602.17888v1
+
+## 11037. The Geometry of Multi-Task Grokking: Transverse Instability, Superposition, and Weight Decay Phase Structure
+
+- **ID:** `2602.18523` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Yongzhong Xu
+- **abs:** https://arxiv.org/abs/2602.18523 · **PDF:** https://arxiv.org/pdf/2602.18523v2
+
+## 11038. Understanding Unreliability of Steering Vectors in Language Models: Geometric Predictors and the Limits of Linear Approximations
+
+- **ID:** `2602.17881` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Joschka Braun
+- **abs:** https://arxiv.org/abs/2602.17881 · **PDF:** https://arxiv.org/pdf/2602.17881v1
+
+## 11039. Interactive Learning of Single-Index Models via Stochastic Gradient Descent
+
+- **ID:** `2602.17876` · **primary:** `stat.ML` · **Submitted:** 2026-02-19
+- **Authors:** Nived Rajaraman, Yanjun Han
+- **abs:** https://arxiv.org/abs/2602.17876 · **PDF:** https://arxiv.org/pdf/2602.17876v1
+
+## 11040. MultiVer: Zero-Shot Multi-Agent Vulnerability Detection
+
+- **ID:** `2602.17875` · **primary:** `cs.MA` · **Submitted:** 2026-02-19
+- **Authors:** Shreshth Rajan
+- **abs:** https://arxiv.org/abs/2602.17875 · **PDF:** https://arxiv.org/pdf/2602.17875v1
+
+## 11041. Understanding the Fine-Grained Knowledge Capabilities of Vision-Language Models
+
+- **ID:** `2602.17871` · **primary:** `cs.CV` · **Submitted:** 2026-02-19
+- **Authors:** Dhruba Ghosh, Yuhui Zhang, Ludwig Schmidt
+- **abs:** https://arxiv.org/abs/2602.17871 · **PDF:** https://arxiv.org/pdf/2602.17871v1
+
+## 11042. MantisV2: Closing the Zero-Shot Gap in Time Series Classification with Synthetic Data and Test-Time Strategies
+
+- **ID:** `2602.17868` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Vasilii Feofanov, Songkang Wen, Jianfeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.17868 · **PDF:** https://arxiv.org/pdf/2602.17868v1
+
+## 11043. ADAPT: Hybrid Prompt Optimization for LLM Feature Visualization
+
+- **ID:** `2602.17867` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** João N. Cardoso, Arlindo L. Oliveira, Bruno Martins
+- **abs:** https://arxiv.org/abs/2602.17867 · **PDF:** https://arxiv.org/pdf/2602.17867v1
+
+## 11044. Financial time series augmentation using transformer based GAN architecture
+
+- **ID:** `2602.17865` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Andrzej Podobiński, Jarosław A. Chudziak
+- **abs:** https://arxiv.org/abs/2602.17865 · **PDF:** https://arxiv.org/pdf/2602.17865v1
+
+## 11045. JAX-Privacy: A library for differentially private machine learning
+
+- **ID:** `2602.17861` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Ryan McKenna, Galen Andrew, Borja Balle, et al.
+- **abs:** https://arxiv.org/abs/2602.17861 · **PDF:** https://arxiv.org/pdf/2602.17861v1
+
+## 11046. Enhancing Scientific Literature Chatbots with Retrieval-Augmented Generation: A Performance Evaluation of Vector and Graph-Based Systems
+
+- **ID:** `2602.17856` · **primary:** `cs.IR` · **Submitted:** 2026-02-19
+- **Authors:** Hamideh Ghanadian, Amin Kamali, Mohammad Hossein Tekieh
+- **abs:** https://arxiv.org/abs/2602.17856 · **PDF:** https://arxiv.org/pdf/2602.17856v1
+
+## 11047. TopoGate: Quality-Aware Topology-Stabilized Gated Fusion for Longitudinal Low-Dose CT New-Lesion Prediction
+
+- **ID:** `2602.17855` · **primary:** `eess.IV` · **Submitted:** 2026-02-19
+- **Authors:** Seungik Cho
+- **abs:** https://arxiv.org/abs/2602.17855 · **PDF:** https://arxiv.org/pdf/2602.17855v1
+
+## 11048. Neural Prior Estimation: Learning Class Priors from Latent Representations
+
+- **ID:** `2602.17853` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Masoud Yavari, Payman Moallem
+- **abs:** https://arxiv.org/abs/2602.17853 · **PDF:** https://arxiv.org/pdf/2602.17853v1
+
+## 11049. Mind the Style: Impact of Communication Style on Human-Chatbot Interaction
+
+- **ID:** `2602.17850` · **primary:** `cs.HC` · **Submitted:** 2026-02-19
+- **Authors:** Erik Derner, Dalibor Kučera, Aditya Gulati, et al.
+- **abs:** https://arxiv.org/abs/2602.17850 · **PDF:** https://arxiv.org/pdf/2602.17850v1
+
+## 11050. Quad Length Codes for Lossless Compression of e4m3
+
+- **ID:** `2602.17849` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Aditya Agrawal, Albert Magyar, Hiteshwar Eswaraiah, et al.
+- **abs:** https://arxiv.org/abs/2602.17849 · **PDF:** https://arxiv.org/pdf/2602.17849v2
+
+## 11051. On the scaling relationship between cloze probabilities and language model next-token prediction
+
+- **ID:** `2602.17848` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Cassandra L. Jacobs, Morgan Grobol
+- **abs:** https://arxiv.org/abs/2602.17848 · **PDF:** https://arxiv.org/pdf/2602.17848v1
+
+## 11052. Two Calm Ends and the Wild Middle: A Geometric Picture of Memorization in Diffusion Models
+
+- **ID:** `2602.17846` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Nick Dodson, Xinyu Gao, Qingsong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.17846 · **PDF:** https://arxiv.org/pdf/2602.17846v1
+
+## 11053. TFL: Targeted Bit-Flip Attack on Large Language Model
+
+- **ID:** `2602.17837` · **primary:** `cs.CR` · **Submitted:** 2026-02-19
+- **Authors:** Jingkai Guo, Chaitali Chakrabarti, Deliang Fan
+- **abs:** https://arxiv.org/abs/2602.17837 · **PDF:** https://arxiv.org/pdf/2602.17837v1
+
+## 11054. AdaptStress: Online Adaptive Learning for Interpretable and Personalized Stress Prediction Using Multivariate and Sparse Physiological Signals
+
+- **ID:** `2602.18521` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Xueyi Wang, Claudine J. C. Lamoth, Elisabeth Wilhelm
+- **abs:** https://arxiv.org/abs/2602.18521 · **PDF:** https://arxiv.org/pdf/2602.18521v1
+
+## 11055. Influence-Preserving Proxies for Gradient-Based Data Selection in LLM Fine-tuning
+
+- **ID:** `2602.17835` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Sirui Chen, Yunzhe Qi, Mengting Ai, et al.
+- **abs:** https://arxiv.org/abs/2602.17835 · **PDF:** https://arxiv.org/pdf/2602.17835v1
+
+## 11056. MePoly: Max Entropy Polynomial Policy Optimization
+
+- **ID:** `2602.17832` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Hang Liu, Sangli Teng, Maani Ghaffari
+- **abs:** https://arxiv.org/abs/2602.17832 · **PDF:** https://arxiv.org/pdf/2602.17832v1
+
+## 11057. Drift Estimation for Stochastic Differential Equations with Denoising Diffusion Models
+
+- **ID:** `2602.17830` · **primary:** `stat.ML` · **Submitted:** 2026-02-19
+- **Authors:** Marcos Tapia Costa, Nikolas Kantas, George Deligiannidis
+- **abs:** https://arxiv.org/abs/2602.17830 · **PDF:** https://arxiv.org/pdf/2602.17830v1
+
+## 11058. The Token Games: Evaluating Language Model Reasoning with Puzzle Duels
+
+- **ID:** `2602.17831` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Simon Henniger, Gabriel Poesia
+- **abs:** https://arxiv.org/abs/2602.17831 · **PDF:** https://arxiv.org/pdf/2602.17831v1
+
+## 11059. Causality by Abstraction: Symbolic Rule Learning in Multivariate Timeseries with Large Language Models
+
+- **ID:** `2602.17829` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Preetom Biswas, Giulia Pedrielli, K. Selçuk Candan
+- **abs:** https://arxiv.org/abs/2602.17829 · **PDF:** https://arxiv.org/pdf/2602.17829v1
+
+## 11060. Avoid What You Know: Divergent Trajectory Balance for GFlowNets
+
+- **ID:** `2602.17827` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Pedro Dall'Antonia, Tiago da Silva, Daniel Csillag, et al.
+- **abs:** https://arxiv.org/abs/2602.17827 · **PDF:** https://arxiv.org/pdf/2602.17827v1
+
+## 11061. Ontology-Guided Neuro-Symbolic Inference: Grounding Language Models with Mathematical Domain Knowledge
+
+- **ID:** `2602.17826` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Marcelo Labre
+- **abs:** https://arxiv.org/abs/2602.17826 · **PDF:** https://arxiv.org/pdf/2602.17826v1
+
+## 11062. Sketch2Feedback: Grammar-in-the-Loop Framework for Rubric-Aligned Feedback on Student STEM Diagrams
+
+- **ID:** `2602.18520` · **primary:** `cs.CV` · **Submitted:** 2026-02-19
+- **Authors:** Aayam Bansal
+- **abs:** https://arxiv.org/abs/2602.18520 · **PDF:** https://arxiv.org/pdf/2602.18520v1
+
+## 11063. Neural Synchrony Between Socially Interacting Language Models
+
+- **ID:** `2602.17815` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Zhining Zhang, Wentao Zhu, Chi Han, et al.
+- **abs:** https://arxiv.org/abs/2602.17815 · **PDF:** https://arxiv.org/pdf/2602.17815v1
+
+## 11064. VQPP: Video Query Performance Prediction Benchmark
+
+- **ID:** `2602.17814` · **primary:** `cs.CV` · **Submitted:** 2026-02-19
+- **Authors:** Adrian Catalin Lutu, Eduard Poesina, Radu Tudor Ionescu
+- **abs:** https://arxiv.org/abs/2602.17814 · **PDF:** https://arxiv.org/pdf/2602.17814v1
+
+## 11065. Calibrated Adaptation: Bayesian Stiefel Manifold Priors for Reliable Parameter-Efficient Fine-Tuning
+
+- **ID:** `2602.17809` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma
+- **abs:** https://arxiv.org/abs/2602.17809 · **PDF:** https://arxiv.org/pdf/2602.17809v1
+
+## 11066. Wide Open Gazes: Quantifying Visual Exploratory Behavior in Soccer with Pose Enhanced Positional Data
+
+- **ID:** `2602.18519` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Joris Bekkers
+- **abs:** https://arxiv.org/abs/2602.18519 · **PDF:** https://arxiv.org/pdf/2602.18519v1
+
+## 11067. Grassmannian Mixture-of-Experts: Concentration-Controlled Routing on Subspace Manifolds
+
+- **ID:** `2602.17798` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma
+- **abs:** https://arxiv.org/abs/2602.17798 · **PDF:** https://arxiv.org/pdf/2602.17798v1
+
+## 11068. Deep Learning for Dermatology: An Innovative Framework for Approaching Precise Skin Cancer Detection
+
+- **ID:** `2602.17797` · **primary:** `eess.IV` · **Submitted:** 2026-02-19
+- **Authors:** Mohammad Tahmid Noor, B. M. Shahria Alam, Tasmiah Rahman Orpa, et al.
+- **abs:** https://arxiv.org/abs/2602.17797 · **PDF:** https://arxiv.org/pdf/2602.17797v1
+
+## 11069. Market Games for Generative Models: Equilibria, Welfare, and Strategic Entry
+
+- **ID:** `2602.17787` · **primary:** `cs.GT` · **Submitted:** 2026-02-19
+- **Authors:** Xiukun Wei, Min Shi, Xueru Zhang
+- **abs:** https://arxiv.org/abs/2602.17787 · **PDF:** https://arxiv.org/pdf/2602.17787v1
+
+## 11070. Measuring the Prevalence of Policy Violating Content with ML Assisted Sampling and LLM Labeling
+
+- **ID:** `2602.18518` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Attila Dobi, Aravindh Manickavasagam, Benjamin Thompson, et al.
+- **abs:** https://arxiv.org/abs/2602.18518 · **PDF:** https://arxiv.org/pdf/2602.18518v1
+
+## 11071. Weak-Form Evolutionary Kolmogorov-Arnold Networks for Solving Partial Differential Equations
+
+- **ID:** `2602.18515` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Bongseok Kim, Jiahao Zhang, Guang Lin
+- **abs:** https://arxiv.org/abs/2602.18515 · **PDF:** https://arxiv.org/pdf/2602.18515v1
+
+## 11072. QueryPlot: Generating Geological Evidence Layers using Natural Language Queries for Mineral Exploration
+
+- **ID:** `2602.17784` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Meng Ye, Xiao Lin, Georgina Lukoczki, et al.
+- **abs:** https://arxiv.org/abs/2602.17784 · **PDF:** https://arxiv.org/pdf/2602.17784v2
+
+## 11073. Multi-material Multi-physics Topology Optimization with Physics-informed Gaussian Process Priors
+
+- **ID:** `2602.17783` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Xiangyu Sun, Shirin Hosseinmardi, Amin Yousefpour, et al.
+- **abs:** https://arxiv.org/abs/2602.17783 · **PDF:** https://arxiv.org/pdf/2602.17783v1
+
+## 11074. Trojan Horses in Recruiting: A Red-Teaming Case Study on Indirect Prompt Injection in Standard vs. Reasoning Models
+
+- **ID:** `2602.18514` · **primary:** `cs.CR` · **Submitted:** 2026-02-19
+- **Authors:** Manuel Wirth
+- **abs:** https://arxiv.org/abs/2602.18514 · **PDF:** https://arxiv.org/pdf/2602.18514v1
+
+## 11075. Topological Exploration of High-Dimensional Empirical Risk Landscapes: general approach, and applications to phase retrieval
+
+- **ID:** `2602.17779` · **primary:** `stat.ML` · **Submitted:** 2026-02-19
+- **Authors:** Antoine Maillard, Tony Bonnaire, Giulio Biroli
+- **abs:** https://arxiv.org/abs/2602.17779 · **PDF:** https://arxiv.org/pdf/2602.17779v1
+
+## 11076. Asking Forever: Universal Activations Behind Turn Amplification in Conversational LLMs
+
+- **ID:** `2602.17778` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Zachary Coalson, Bo Fang, Sanghyun Hong
+- **abs:** https://arxiv.org/abs/2602.17778 · **PDF:** https://arxiv.org/pdf/2602.17778v1
+
+## 11077. Solving and learning advective multiscale Darcian dynamics with the Neural Basis Method
+
+- **ID:** `2602.17776` · **primary:** `math.NA` · **Submitted:** 2026-02-19
+- **Authors:** Yuhe Wang, Min Wang
+- **abs:** https://arxiv.org/abs/2602.17776 · **PDF:** https://arxiv.org/pdf/2602.17776v1
+
+## 11078. A Hybrid Tsallis-Polarization Impurity Measure for Decision Trees: Theoretical Foundations and Empirical Evaluation
+
+- **ID:** `2603.13241` · **primary:** `stat.ML` · **Submitted:** 2026-02-19
+- **Authors:** Edouard Lansiaux, Idriss Jairi, Hayfa Zgaya-Biau
+- **abs:** https://arxiv.org/abs/2603.13241 · **PDF:** https://arxiv.org/pdf/2603.13241v1
+
+## 11079. Learning Flow Distributions via Projection-Constrained Diffusion on Manifolds
+
+- **ID:** `2602.17773` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-02-19
+- **Authors:** Noah Trupin, Rahul Ghosh, Aadi Jangid
+- **abs:** https://arxiv.org/abs/2602.17773 · **PDF:** https://arxiv.org/pdf/2602.17773v1
+
+## 11080. Closing the Expertise Gap in Residential Building Energy Retrofits: A Domain-Specific LLM for Informed Decision-Making
+
+- **ID:** `2602.20181` · **primary:** `cs.CY` · **Submitted:** 2026-02-19
+- **Authors:** Lei Shu, Armin Yeganeh, Sinem Mollaoglu, et al.
+- **abs:** https://arxiv.org/abs/2602.20181 · **PDF:** https://arxiv.org/pdf/2602.20181v1
+
+## 11081. Sparse Bayesian Modeling of EEG Channel Interactions Improves P300 Brain-Computer Interface Performance
+
+- **ID:** `2602.17772` · **primary:** `stat.ME` · **Submitted:** 2026-02-19
+- **Authors:** Guoxuan Ma, Yuan Zhong, Moyan Li, et al.
+- **abs:** https://arxiv.org/abs/2602.17772 · **PDF:** https://arxiv.org/pdf/2602.17772v2
+
+## 11082. CLUTCH: Contextualized Language model for Unlocking Text-Conditioned Hand motion modelling in the wild
+
+- **ID:** `2602.17770` · **primary:** `cs.CV` · **Submitted:** 2026-02-19
+- **Authors:** Balamurugan Thambiraja, Omid Taheri, Radek Danecek, et al.
+- **abs:** https://arxiv.org/abs/2602.17770 · **PDF:** https://arxiv.org/pdf/2602.17770v1
+
+## 11083. Sink-Aware Pruning for Diffusion Language Models
+
+- **ID:** `2602.17664` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Aidar Myrzakhan, Tianyi Li, Bowei Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.17664 · **PDF:** https://arxiv.org/pdf/2602.17664v1
+
+## 11084. CLEF HIPE-2026: Evaluating Accurate and Efficient Person-Place Relation Extraction from Multilingual Historical Texts
+
+- **ID:** `2602.17663` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Juri Opitz, Corina Raclé, Emanuela Boros, et al.
+- **abs:** https://arxiv.org/abs/2602.17663 · **PDF:** https://arxiv.org/pdf/2602.17663v1
+
+## 11085. MARS: Margin-Aware Reward-Modeling with Self-Refinement
+
+- **ID:** `2602.17658` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Payel Bhattacharjee, Osvaldo Simeone, Ravi Tandon
+- **abs:** https://arxiv.org/abs/2602.17658 · **PDF:** https://arxiv.org/pdf/2602.17658v1
+
+## 11086. What Language is This? Ask Your Tokenizer
+
+- **ID:** `2602.17655` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Clara Meister, Ahmetcan Yavuz, Pietro Lesci, et al.
+- **abs:** https://arxiv.org/abs/2602.17655 · **PDF:** https://arxiv.org/pdf/2602.17655v1
+
+## 11087. The 2025 AI Agent Index: Documenting Technical and Safety Features of Deployed Agentic AI Systems
+
+- **ID:** `2602.17753` · **primary:** `cs.CY` · **Submitted:** 2026-02-19
+- **Authors:** Leon Staufer, Kevin Feng, Kevin Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.17753 · **PDF:** https://arxiv.org/pdf/2602.17753v1
+
+## 11088. Mine and Refine: Optimizing Graded Relevance in E-commerce Search Retrieval
+
+- **ID:** `2602.17654` · **primary:** `cs.IR` · **Submitted:** 2026-02-19
+- **Authors:** Jiaqi Xi, Raghav Saboo, Luming Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.17654 · **PDF:** https://arxiv.org/pdf/2602.17654v1
+
+## 11089. Differences in Typological Alignment in Language Models' Treatment of Differential Argument Marking
+
+- **ID:** `2602.17653` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Iskar Deng, Nathalia Xu, Shane Steinert-Threlkeld
+- **abs:** https://arxiv.org/abs/2602.17653 · **PDF:** https://arxiv.org/pdf/2602.17653v1
+
+## 11090. Multi-Round Human-AI Collaboration with User-Specified Requirements
+
+- **ID:** `2602.17646` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Sima Noorani, Shayan Kiyani, Hamed Hassani, et al.
+- **abs:** https://arxiv.org/abs/2602.17646 · **PDF:** https://arxiv.org/pdf/2602.17646v2
+
+## 11091. Pushing the Frontier of Black-Box LVLM Attacks via Fine-Grained Detail Targeting
+
+- **ID:** `2602.17645` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Xiaohan Zhao, Zhaoyi Li, Yaxin Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.17645 · **PDF:** https://arxiv.org/pdf/2602.17645v1
+
+## 11092. A.R.I.S.: Automated Recycling Identification System for E-Waste Classification Using Deep Learning
+
+- **ID:** `2602.17642` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Dhruv Talwar, Harsh Desai, Wendong Yin, et al.
+- **abs:** https://arxiv.org/abs/2602.17642 · **PDF:** https://arxiv.org/pdf/2602.17642v1
+
+## 11093. FAMOSE: A ReAct Approach to Automated Feature Discovery
+
+- **ID:** `2602.17641` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Keith Burghardt, Jienan Liu, Sadman Sakib, et al.
+- **abs:** https://arxiv.org/abs/2602.17641 · **PDF:** https://arxiv.org/pdf/2602.17641v1
+
+## 11094. Reverso: Efficient Time Series Foundation Models for Zero-shot Forecasting
+
+- **ID:** `2602.17634` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Xinghong Fu, Yanhong Li, Georgios Papaioannou, et al.
+- **abs:** https://arxiv.org/abs/2602.17634 · **PDF:** https://arxiv.org/pdf/2602.17634v1
+
+## 11095. When to Trust the Cheap Check: Weak and Strong Verification for Reasoning
+
+- **ID:** `2602.17633` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Shayan Kiyani, Sima Noorani, George Pappas, et al.
+- **abs:** https://arxiv.org/abs/2602.17633 · **PDF:** https://arxiv.org/pdf/2602.17633v1
+
+## 11096. SMAC: Score-Matched Actor-Critics for Robust Offline-to-Online Transfer
+
+- **ID:** `2602.17632` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Nathan Samuel de Lara, Florian Shkurti
+- **abs:** https://arxiv.org/abs/2602.17632 · **PDF:** https://arxiv.org/pdf/2602.17632v2
+
+## 11097. Catastrophic Forgetting Resilient One-Shot Incremental Federated Learning
+
+- **ID:** `2602.17625` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Obaidullah Zaland, Zulfiqar Ahmad Khan, Monowar Bhuyan
+- **abs:** https://arxiv.org/abs/2602.17625 · **PDF:** https://arxiv.org/pdf/2602.17625v1
+
+## 11098. Unmasking the Factual-Conceptual Gap in Persian Language Models
+
+- **ID:** `2602.17623` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Alireza Sakhaeirad, Ali Ma'manpoosh, Arshia Hemmat
+- **abs:** https://arxiv.org/abs/2602.17623 · **PDF:** https://arxiv.org/pdf/2602.17623v1
+
+## 11099. Stable Asynchrony: Variance-Controlled Off-Policy RL for LLMs
+
+- **ID:** `2602.17616` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Luke J. Huang, Zhuoyang Zhang, Qinghao Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.17616 · **PDF:** https://arxiv.org/pdf/2602.17616v2
+
+## 11100. Guarding the Middle: Protecting Intermediate Representations in Federated Split Learning
+
+- **ID:** `2602.17614` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Obaidullah Zaland, Sajib Mistry, Monowar Bhuyan
+- **abs:** https://arxiv.org/abs/2602.17614 · **PDF:** https://arxiv.org/pdf/2602.17614v1
+
+## 11101. Towards Anytime-Valid Statistical Watermarking
+
+- **ID:** `2602.17608` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Baihe Huang, Eric Xu, Kannan Ramchandran, et al.
+- **abs:** https://arxiv.org/abs/2602.17608 · **PDF:** https://arxiv.org/pdf/2602.17608v1
+
+## 11102. AutoNumerics: An Autonomous, PDE-Agnostic Multi-Agent Pipeline for Scientific Computing
+
+- **ID:** `2602.17607` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Jianda Du, Youran Sun, Haizhao Yang
+- **abs:** https://arxiv.org/abs/2602.17607 · **PDF:** https://arxiv.org/pdf/2602.17607v1
+
+## 11103. Adapting Actively on the Fly: Relevance-Guided Online Meta-Learning with Latent Concepts for Geospatial Discovery
+
+- **ID:** `2602.17605` · **primary:** `cs.CV` · **Submitted:** 2026-02-19
+- **Authors:** Jowaria Khan, Anindya Sarkar, Yevgeniy Vorobeychik, et al.
+- **abs:** https://arxiv.org/abs/2602.17605 · **PDF:** https://arxiv.org/pdf/2602.17605v1
+
+## 11104. SOLVAR: Fast covariance-based heterogeneity analysis with pose refinement for cryo-EM
+
+- **ID:** `2602.17603` · **primary:** `stat.ML` · **Submitted:** 2026-02-19
+- **Authors:** Roey Yadgar, Roy R. Lederman, Yoel Shkolnisky
+- **abs:** https://arxiv.org/abs/2602.17603 · **PDF:** https://arxiv.org/pdf/2602.17603v1
+
+## 11105. MolHIT: Advancing Molecular-Graph Generation with Hierarchical Discrete Diffusion Models
+
+- **ID:** `2602.17602` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Hojung Jung, Rodrigo Hormazabal, Jaehyeong Jo, et al.
+- **abs:** https://arxiv.org/abs/2602.17602 · **PDF:** https://arxiv.org/pdf/2602.17602v1
+
+## 11106. The Cascade Equivalence Hypothesis: When Do Speech LLMs Behave Like ASR$\rightarrow$LLM Pipelines?
+
+- **ID:** `2602.17598` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Jayadev Billa
+- **abs:** https://arxiv.org/abs/2602.17598 · **PDF:** https://arxiv.org/pdf/2602.17598v2
+
+## 11107. Asymptotic Smoothing of the Lipschitz Loss Landscape in Overparameterized One-Hidden-Layer ReLU Networks
+
+- **ID:** `2602.17596` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Saveliy Baturin
+- **abs:** https://arxiv.org/abs/2602.17596 · **PDF:** https://arxiv.org/pdf/2602.17596v1
+
+## 11108. AI Gamestore: Scalable, Open-Ended Evaluation of Machine General Intelligence with Human Games
+
+- **ID:** `2602.17594` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Lance Ying, Ryan Truong, Prafull Sharma, et al.
+- **abs:** https://arxiv.org/abs/2602.17594 · **PDF:** https://arxiv.org/pdf/2602.17594v1
+
+## 11109. Modeling Distinct Human Interaction in Web Agents
+
+- **ID:** `2602.17588` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Faria Huq, Zora Zhiruo Wang, Zhanqiu Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.17588 · **PDF:** https://arxiv.org/pdf/2602.17588v2
+
+## 11110. Asymptotically Optimal Sequential Testing with Markovian Data
+
+- **ID:** `2602.17587` · **primary:** `math.ST` · **Submitted:** 2026-02-19
+- **Authors:** Alhad Sethi, Kavali Sofia Sagar, Shubhada Agrawal, et al.
+- **abs:** https://arxiv.org/abs/2602.17587 · **PDF:** https://arxiv.org/pdf/2602.17587v1
+
+## 11111. Conditional Flow Matching for Continuous Anomaly Detection in Autonomous Driving on a Manifold-Aware Spectral Space
+
+- **ID:** `2602.17586` · **primary:** `cs.RO` · **Submitted:** 2026-02-19
+- **Authors:** Antonio Guillen-Perez
+- **abs:** https://arxiv.org/abs/2602.17586 · **PDF:** https://arxiv.org/pdf/2602.17586v1
+
+## 11112. Canonicalizing Multimodal Contrastive Representation Learning
+
+- **ID:** `2602.17584` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Sharut Gupta, Sanyam Kansal, Stefanie Jegelka, et al.
+- **abs:** https://arxiv.org/abs/2602.17584 · **PDF:** https://arxiv.org/pdf/2602.17584v1
+
+## 11113. Simultaneous Blackwell Approachability and Applications to Multiclass Omniprediction
+
+- **ID:** `2602.17577` · **primary:** `cs.DS` · **Submitted:** 2026-02-19
+- **Authors:** Lunjia Hu, Kevin Tian, Chutong Yang
+- **abs:** https://arxiv.org/abs/2602.17577 · **PDF:** https://arxiv.org/pdf/2602.17577v1
+
+## 11114. Be Wary of Your Time Series Preprocessing
+
+- **ID:** `2602.17568` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Sofiane Ennadir, Tianze Wang, Oleg Smirnov, et al.
+- **abs:** https://arxiv.org/abs/2602.17568 · **PDF:** https://arxiv.org/pdf/2602.17568v1
+
+## 11115. A Hybrid Federated Learning Based Ensemble Approach for Lung Disease Diagnosis Leveraging Fusion of SWIN Transformer and CNN
+
+- **ID:** `2602.17566` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Asif Hasan Chowdhury, Md. Fahim Islam, M Ragib Anjum Riad, et al.
+- **abs:** https://arxiv.org/abs/2602.17566 · **PDF:** https://arxiv.org/pdf/2602.17566v1
+
+## 11116. Optimal Unconstrained Self-Distillation in Ridge Regression: Strict Improvements, Precise Asymptotics, and One-Shot Tuning
+
+- **ID:** `2602.17565` · **primary:** `math.ST` · **Submitted:** 2026-02-19
+- **Authors:** Hien Dang, Pratik Patil, Alessandro Rinaldo
+- **abs:** https://arxiv.org/abs/2602.17565 · **PDF:** https://arxiv.org/pdf/2602.17565v1
+
+## 11117. ODESteer: A Unified ODE-Based Steering Framework for LLM Alignment
+
+- **ID:** `2602.17560` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Hongjue Zhao, Haosen Sun, Jiangtao Kong, et al.
+- **abs:** https://arxiv.org/abs/2602.17560 · **PDF:** https://arxiv.org/pdf/2602.17560v2
+
+## 11118. Revisiting Weight Regularization for Low-Rank Continual Learning
+
+- **ID:** `2602.17559` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Yaoyue Zheng, Yin Zhang, Joost van de Weijer, et al.
+- **abs:** https://arxiv.org/abs/2602.17559 · **PDF:** https://arxiv.org/pdf/2602.17559v1
+
+## 11119. Probability-Invariant Random Walk Learning on Gyral Folding-Based Cortical Similarity Networks for Alzheimer's and Lewy Body Dementia Diagnosis
+
+- **ID:** `2602.17557` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-19
+- **Authors:** Minheng Chen, Tong Chen, Chao Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.17557 · **PDF:** https://arxiv.org/pdf/2602.17557v2
+
+## 11120. A Theoretical Framework for Modular Learning of Robust Generative Models
+
+- **ID:** `2602.17554` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Corinna Cortes, Mehryar Mohri, Yutao Zhong
+- **abs:** https://arxiv.org/abs/2602.17554 · **PDF:** https://arxiv.org/pdf/2602.17554v2
+
+## 11121. MASPO: Unifying Gradient Utilization, Probability Mass, and Signal Reliability for Robust and Sample-Efficient LLM Reasoning
+
+- **ID:** `2602.17550` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Xiaoliang Fu, Jiaye Lin, Yangyi Fang, et al.
+- **abs:** https://arxiv.org/abs/2602.17550 · **PDF:** https://arxiv.org/pdf/2602.17550v2
+
+## 11122. KLong: Training LLM Agent for Extremely Long-horizon Tasks
+
+- **ID:** `2602.17547` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Yue Liu, Zhiyuan Hu, Flood Sung, et al.
+- **abs:** https://arxiv.org/abs/2602.17547 · **PDF:** https://arxiv.org/pdf/2602.17547v1
+
+## 11123. Learning to Stay Safe: Adaptive Regularization Against Safety Degradation during Fine-Tuning
+
+- **ID:** `2602.17546` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Jyotin Goel, Souvik Maji, Pratik Mazumder
+- **abs:** https://arxiv.org/abs/2602.17546 · **PDF:** https://arxiv.org/pdf/2602.17546v1
+
+## 11124. Adaptive Decentralized Composite Optimization via Three-Operator Splitting
+
+- **ID:** `2602.17545` · **primary:** `math.OC` · **Submitted:** 2026-02-19
+- **Authors:** Xiaokai Chen, Ilya Kuruzov, Gesualdo Scutari
+- **abs:** https://arxiv.org/abs/2602.17545 · **PDF:** https://arxiv.org/pdf/2602.17545v1
+
+## 11125. Evaluating Chain-of-Thought Reasoning through Reusability and Verifiability
+
+- **ID:** `2602.17544` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Shashank Aggarwal, Ram Vikas Mishra, Amit Awekar
+- **abs:** https://arxiv.org/abs/2602.17544 · **PDF:** https://arxiv.org/pdf/2602.17544v1
+
+## 11126. genriesz: A Python Package for Automatic Debiased Machine Learning with Generalized Riesz Regression
+
+- **ID:** `2602.17543` · **primary:** `stat.ML` · **Submitted:** 2026-02-19
+- **Authors:** Masahiro Kato
+- **abs:** https://arxiv.org/abs/2602.17543 · **PDF:** https://arxiv.org/pdf/2602.17543v1
+
+## 11127. Using LLMs for Knowledge Component-level Correctness Labeling in Open-ended Coding Problems
+
+- **ID:** `2602.17542` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Zhangqi Duan, Arnav Kankaria, Dhruv Kartik, et al.
+- **abs:** https://arxiv.org/abs/2602.17542 · **PDF:** https://arxiv.org/pdf/2602.17542v2
+
+## 11128. IRIS: Learning-Driven Task-Specific Cinema Robot Arm for Visuomotor Motion Control
+
+- **ID:** `2602.17537` · **primary:** `cs.RO` · **Submitted:** 2026-02-19
+- **Authors:** Qilong Cheng, Matthew Mackay, Ali Bereyhi
+- **abs:** https://arxiv.org/abs/2602.17537 · **PDF:** https://arxiv.org/pdf/2602.17537v1
+
+## 11129. Toward a Fully Autonomous, AI-Native Particle Accelerator
+
+- **ID:** `2602.17536` · **primary:** `physics.acc-ph` · **Submitted:** 2026-02-19
+- **Authors:** Chris Tennant
+- **abs:** https://arxiv.org/abs/2602.17536 · **PDF:** https://arxiv.org/pdf/2602.17536v1
+
+## 11130. Systematic Evaluation of Single-Cell Foundation Model Interpretability Reveals Attention Captures Co-Expression Rather Than Unique Regulatory Signal
+
+- **ID:** `2602.17532` · **primary:** `q-bio.GN` · **Submitted:** 2026-02-19
+- **Authors:** Ihor Kendiukhov
+- **abs:** https://arxiv.org/abs/2602.17532 · **PDF:** https://arxiv.org/pdf/2602.17532v1
+
+## 11131. Position: Evaluation of ECG Representations Must Be Fixed
+
+- **ID:** `2602.17531` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Zachary Berger, Daniel Prakah-Asante, John Guttag, et al.
+- **abs:** https://arxiv.org/abs/2602.17531 · **PDF:** https://arxiv.org/pdf/2602.17531v1
+
+## 11132. Provably Explaining Neural Additive Models
+
+- **ID:** `2602.17530` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Shahaf Bassan, Yizhak Yisrael Elboher, Tobias Ladner, et al.
+- **abs:** https://arxiv.org/abs/2602.17530 · **PDF:** https://arxiv.org/pdf/2602.17530v1
+
+## 11133. Enhancing Large Language Models (LLMs) for Telecom using Dynamic Knowledge Graphs and Explainable Retrieval-Augmented Generation
+
+- **ID:** `2602.17529` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Dun Yuan, Hao Zhou, Xue Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.17529 · **PDF:** https://arxiv.org/pdf/2602.17529v1
+
+## 11134. The Anxiety of Influence: Bloom Filters in Transformer Attention Heads
+
+- **ID:** `2602.17526` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Peter Balogh
+- **abs:** https://arxiv.org/abs/2602.17526 · **PDF:** https://arxiv.org/pdf/2602.17526v1
+
+## 11135. Variational inference via radial transport
+
+- **ID:** `2602.17525` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Luca Ghafourpour, Sinho Chewi, Alessio Figalli, et al.
+- **abs:** https://arxiv.org/abs/2602.17525 · **PDF:** https://arxiv.org/pdf/2602.17525v2
+
+## 11136. Bridging the Domain Divide: Supervised vs. Zero-Shot Clinical Section Segmentation from MIMIC-III to Obstetrics
+
+- **ID:** `2602.17513` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Baris Karacan, Barbara Di Eugenio, Patrick Thornton
+- **abs:** https://arxiv.org/abs/2602.17513 · **PDF:** https://arxiv.org/pdf/2602.17513v1
+
+## 11137. Investigating Target Class Influence on Neural Network Compressibility for Energy-Autonomous Avian Monitoring
+
+- **ID:** `2602.17751` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Nina Brolich, Simon Geis, Maximilian Kasper, et al.
+- **abs:** https://arxiv.org/abs/2602.17751 · **PDF:** https://arxiv.org/pdf/2602.17751v1
+
+## 11138. ARKV: Adaptive and Resource-Efficient KV Cache Management under Limited Memory Budget for Long-Context Inference in LLMs
+
+- **ID:** `2603.08727` · **primary:** `cs.AR` · **Submitted:** 2026-02-19
+- **Authors:** Jianlong Lei, Shashikant Ilager
+- **abs:** https://arxiv.org/abs/2603.08727 · **PDF:** https://arxiv.org/pdf/2603.08727v1
+
+## 11139. LORA-CRAFT: Cross-layer Rank Adaptation via Frozen Tucker Decomposition of Pre-trained Attention Weights
+
+- **ID:** `2602.17510` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Kasun Dewage, Marianna Pensky, Suranadi De Silva, et al.
+- **abs:** https://arxiv.org/abs/2602.17510 · **PDF:** https://arxiv.org/pdf/2602.17510v1
+
+## 11140. Pareto Optimal Benchmarking of AI Models on ARM Cortex Processors for Sustainable Embedded Systems
+
+- **ID:** `2602.17508` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Pranay Jain, Maximilian Kasper, Göran Köber, et al.
+- **abs:** https://arxiv.org/abs/2602.17508 · **PDF:** https://arxiv.org/pdf/2602.17508v2
+
+## 11141. Retrospective In-Context Learning for Temporal Credit Assignment with Large Language Models
+
+- **ID:** `2602.17497` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Wen-Tse Chen, Jiayu Chen, Fahim Tajwar, et al.
+- **abs:** https://arxiv.org/abs/2602.17497 · **PDF:** https://arxiv.org/pdf/2602.17497v1
+
+## 11142. Learning with Boolean threshold functions
+
+- **ID:** `2602.17493` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Veit Elser, Manish Krishan Lal
+- **abs:** https://arxiv.org/abs/2602.17493 · **PDF:** https://arxiv.org/pdf/2602.17493v1
+
+## 11143. Linear Convergence in Games with Delayed Feedback via Extra Prediction
+
+- **ID:** `2602.17486` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Yuma Fujimoto, Kenshi Abe, Kaito Ariu
+- **abs:** https://arxiv.org/abs/2602.17486 · **PDF:** https://arxiv.org/pdf/2602.17486v1
+
+## 11144. Tracing Copied Pixels and Regularizing Patch Affinity in Copy Detection
+
+- **ID:** `2602.17484` · **primary:** `cs.CV` · **Submitted:** 2026-02-19
+- **Authors:** Yichen Lu, Siwei Nie, Minlong Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.17484 · **PDF:** https://arxiv.org/pdf/2602.17484v2
+
+## 11145. What Do LLMs Associate with Your Name? A Human-Centered Black-Box Audit of Personal Data
+
+- **ID:** `2602.17483` · **primary:** `cs.HC` · **Submitted:** 2026-02-19
+- **Authors:** Dimitri Staufer, Kirsten Morehouse
+- **abs:** https://arxiv.org/abs/2602.17483 · **PDF:** https://arxiv.org/pdf/2602.17483v1
+
+## 11146. Inelastic Constitutive Kolmogorov-Arnold Networks: A generalized framework for automated discovery of interpretable inelastic material models
+
+- **ID:** `2602.17750` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-19
+- **Authors:** Chenyi Ji, Kian P. Abdolazizi, Hagen Holthusen, et al.
+- **abs:** https://arxiv.org/abs/2602.17750 · **PDF:** https://arxiv.org/pdf/2602.17750v1
+
+## 11147. Detection and Classification of Cetacean Echolocation Clicks using Image-based Object Detection Methods applied to Advanced Wavelet-based Transformations
+
+- **ID:** `2602.17749` · **primary:** `eess.AS` · **Submitted:** 2026-02-19
+- **Authors:** Christopher Hauer
+- **abs:** https://arxiv.org/abs/2602.17749 · **PDF:** https://arxiv.org/pdf/2602.17749v1
+
+## 11148. Variational Grey-Box Dynamics Matching
+
+- **ID:** `2602.17477` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Gurjeet Sangra Singh, Frantzeska Lavda, Giangiacomo Mercatali, et al.
+- **abs:** https://arxiv.org/abs/2602.17477 · **PDF:** https://arxiv.org/pdf/2602.17477v2
+
+## 11149. Small LLMs for Medical NLP: a Systematic Analysis of Few-Shot, Constraint Decoding, Fine-Tuning and Continual Pre-Training in Italian
+
+- **ID:** `2602.17475` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Pietro Ferrazzi, Mattia Franzin, Alberto Lavelli, et al.
+- **abs:** https://arxiv.org/abs/2602.17475 · **PDF:** https://arxiv.org/pdf/2602.17475v1
+
+## 11150. Auditing Reciprocal Sentiment Alignment: Inversion Risk, Dialect Representation and Intent Misalignment in Transformers
+
+- **ID:** `2602.17469` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Nusrat Jahan Lia, Shubhashis Roy Dipta
+- **abs:** https://arxiv.org/abs/2602.17469 · **PDF:** https://arxiv.org/pdf/2602.17469v1
+
+## 11151. PEACE 2.0: Grounded Explanations and Counter-Speech for Combating Hate Expressions
+
+- **ID:** `2602.17467` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Greta Damo, Stéphane Petiot, Elena Cabrio, et al.
+- **abs:** https://arxiv.org/abs/2602.17467 · **PDF:** https://arxiv.org/pdf/2602.17467v1
+
+## 11152. Entropy-Based Data Selection for Language Models
+
+- **ID:** `2602.17465` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Hongming Li, Yang Liu, Chao Huang
+- **abs:** https://arxiv.org/abs/2602.17465 · **PDF:** https://arxiv.org/pdf/2602.17465v1
+
+## 11153. Jolt Atlas: Verifiable Inference via Lookup Arguments in Zero Knowledge
+
+- **ID:** `2602.17452` · **primary:** `cs.CR` · **Submitted:** 2026-02-19
+- **Authors:** Wyatt Benno, Alberto Centelles, Antoine Douchet, et al.
+- **abs:** https://arxiv.org/abs/2602.17452 · **PDF:** https://arxiv.org/pdf/2602.17452v1
+
+## 11154. The IJCNN 2025 Review Process
+
+- **ID:** `2603.19244` · **primary:** `cs.DL` · **Submitted:** 2026-02-19
+- **Authors:** Michele Scarpiniti, Danilo Comminiello
+- **abs:** https://arxiv.org/abs/2603.19244 · **PDF:** https://arxiv.org/pdf/2603.19244v1
+
+## 11155. Beyond Pipelines: A Fundamental Study on the Rise of Generative-Retrieval Architectures in Web Research
+
+- **ID:** `2602.17450` · **primary:** `cs.IR` · **Submitted:** 2026-02-19
+- **Authors:** Amirereza Abbasi, Mohsen Hooshmand
+- **abs:** https://arxiv.org/abs/2602.17450 · **PDF:** https://arxiv.org/pdf/2602.17450v1
+
+## 11156. ABCD: All Biases Come Disguised
+
+- **ID:** `2602.17445` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Mateusz Nowak, Xavier Cadet, Peter Chin
+- **abs:** https://arxiv.org/abs/2602.17445 · **PDF:** https://arxiv.org/pdf/2602.17445v1
+
+## 11157. AIDG: Evaluating Asymmetry Between Information Extraction and Containment in Multi-Turn Dialogue
+
+- **ID:** `2602.17443` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Adib Sakhawat, Fardeen Sadab, Rakin Shahriar
+- **abs:** https://arxiv.org/abs/2602.17443 · **PDF:** https://arxiv.org/pdf/2602.17443v1
+
+## 11158. WarpRec: Unifying Academic Rigor and Industrial Scale for Responsible, Reproducible, and Efficient Recommendation
+
+- **ID:** `2602.17442` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Marco Avolio, Potito Aghilar, Sabino Roccotelli, et al.
+- **abs:** https://arxiv.org/abs/2602.17442 · **PDF:** https://arxiv.org/pdf/2602.17442v1
+
+## 11159. Fine-Grained Uncertainty Quantification for Long-Form Language Model Outputs: A Comparative Study
+
+- **ID:** `2602.17431` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Dylan Bouchard, Mohit Singh Chauhan, Viren Bajaj, et al.
+- **abs:** https://arxiv.org/abs/2602.17431 · **PDF:** https://arxiv.org/pdf/2602.17431v1
+
+## 11160. Evaluating Extremely Low-Resource Machine Translation: A Comparative Study of ChrF++ and BLEU Metrics
+
+- **ID:** `2602.17425` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Sanjeev Kumar, Preethi Jyothi, Pushpak Bhattacharyya
+- **abs:** https://arxiv.org/abs/2602.17425 · **PDF:** https://arxiv.org/pdf/2602.17425v1
+
+## 11161. Diverse Word Choices, Same Reference: Annotating Lexically-Rich Cross-Document Coreference
+
+- **ID:** `2602.17424` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Anastasia Zhukova, Felix Hamborg, Karsten Donnay, et al.
+- **abs:** https://arxiv.org/abs/2602.17424 · **PDF:** https://arxiv.org/pdf/2602.17424v2
+
+## 11162. Convergence Analysis of Two-Layer Neural Networks under Gaussian Input Masking
+
+- **ID:** `2602.17423` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Afroditi Kolomvaki, Fangshuo Liao, Evan Dramko, et al.
+- **abs:** https://arxiv.org/abs/2602.17423 · **PDF:** https://arxiv.org/pdf/2602.17423v1
+
+## 11163. A Privacy by Design Framework for Large Language Model-Based Applications for Children
+
+- **ID:** `2602.17418` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Diana Addae, Diana Rogachova, Nafiseh Kahani, et al.
+- **abs:** https://arxiv.org/abs/2602.17418 · **PDF:** https://arxiv.org/pdf/2602.17418v1
+
+## 11164. DAVE: A Policy-Enforcing LLM Spokesperson for Secure Multi-Document Data Sharing
+
+- **ID:** `2602.17413` · **primary:** `cs.CR` · **Submitted:** 2026-02-19
+- **Authors:** René Brinkhege, Prahlad Menon
+- **abs:** https://arxiv.org/abs/2602.17413 · **PDF:** https://arxiv.org/pdf/2602.17413v1
+
+## 11165. Improving LLM-based Recommendation with Self-Hard Negatives from Intermediate Layers
+
+- **ID:** `2602.17410` · **primary:** `cs.IR` · **Submitted:** 2026-02-19
+- **Authors:** Bingqian Li, Bowen Zheng, Xiaolei Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.17410 · **PDF:** https://arxiv.org/pdf/2602.17410v1
+
+## 11166. A Contrastive Variational AutoEncoder for NSCLC Survival Prediction with Missing Modalities
+
+- **ID:** `2602.17402` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Michele Zanitti, Vanja Miskovic, Francesco Trovò, et al.
+- **abs:** https://arxiv.org/abs/2602.17402 · **PDF:** https://arxiv.org/pdf/2602.17402v1
+
+## 11167. A High-Level Survey of Optical Remote Sensing
+
+- **ID:** `2602.17397` · **primary:** `cs.CV` · **Submitted:** 2026-02-19
+- **Authors:** Panagiotis Koletsis, Vasilis Efthymiou, Maria Vakalopoulou, et al.
+- **abs:** https://arxiv.org/abs/2602.17397 · **PDF:** https://arxiv.org/pdf/2602.17397v1
+
+## 11168. SpectralGCD: Spectral Concept Selection and Cross-modal Representation Learning for Generalized Category Discovery
+
+- **ID:** `2602.17395` · **primary:** `cs.CV` · **Submitted:** 2026-02-19
+- **Authors:** Lorenzo Caselli, Marco Mistretta, Simone Magistri, et al.
+- **abs:** https://arxiv.org/abs/2602.17395 · **PDF:** https://arxiv.org/pdf/2602.17395v1
+
+## 11169. Voice-Driven Semantic Perception for UAV-Assisted Emergency Networks
+
+- **ID:** `2602.17394` · **primary:** `cs.NI` · **Submitted:** 2026-02-19
+- **Authors:** Nuno Saavedra, Pedro Ribeiro, André Coelho, et al.
+- **abs:** https://arxiv.org/abs/2602.17394 · **PDF:** https://arxiv.org/pdf/2602.17394v1
+
+## 11170. Visual Model Checking: Graph-Based Inference of Visual Routines for Image Retrieval
+
+- **ID:** `2602.17386` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Adrià Molina, Oriol Ramos Terrades, Josep Lladós
+- **abs:** https://arxiv.org/abs/2602.17386 · **PDF:** https://arxiv.org/pdf/2602.17386v1
+
+## 11171. Dataless Weight Disentanglement in Task Arithmetic via Kronecker-Factored Approximate Curvature
+
+- **ID:** `2602.17385` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Angelo Porrello, Pietro Buzzega, Felix Dangel, et al.
+- **abs:** https://arxiv.org/abs/2602.17385 · **PDF:** https://arxiv.org/pdf/2602.17385v2
+
+## 11172. AgriVariant: Variant Effect Prediction using DeepChem-Variant for Precision Breeding in Rice
+
+- **ID:** `2602.17747` · **primary:** `q-bio.GN` · **Submitted:** 2026-02-19
+- **Authors:** Ankita Vaishnobi Bisoi, Bharath Ramsundar
+- **abs:** https://arxiv.org/abs/2602.17747 · **PDF:** https://arxiv.org/pdf/2602.17747v1
+
+## 11173. The Role of the Availability Heuristic in Multiple-Choice Answering Behaviour
+
+- **ID:** `2602.17377` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Leonidas Zotos, Hedderik van Rijn, Malvina Nissim
+- **abs:** https://arxiv.org/abs/2602.17377 · **PDF:** https://arxiv.org/pdf/2602.17377v1
+
+## 11174. MDP Planning as Policy Inference
+
+- **ID:** `2602.17375` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** David Tolpin
+- **abs:** https://arxiv.org/abs/2602.17375 · **PDF:** https://arxiv.org/pdf/2602.17375v1
+
+## 11175. RPDR: A Round-trip Prediction-Based Data Augmentation Framework for Long-Tail Question Answering
+
+- **ID:** `2602.17366` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Yiming Zhang, Siyue Zhang, Junbo Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.17366 · **PDF:** https://arxiv.org/pdf/2602.17366v1
+
+## 11176. Beyond Pass-by-Pass Optimization: Intent-Driven IR Optimization with Large Language Models
+
+- **ID:** `2602.18511` · **primary:** `cs.PL` · **Submitted:** 2026-02-19
+- **Authors:** Lei Qiu, Zi Yang, Fang Lyu, et al.
+- **abs:** https://arxiv.org/abs/2602.18511 · **PDF:** https://arxiv.org/pdf/2602.18511v1
+
+## 11177. A feature-stable and explainable machine learning framework for trustworthy decision-making under incomplete clinical data
+
+- **ID:** `2602.17364` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Justyna Andrys-Olek, Paulina Tworek, Luca Gherardini, et al.
+- **abs:** https://arxiv.org/abs/2602.17364 · **PDF:** https://arxiv.org/pdf/2602.17364v1
+
+## 11178. 2Mamba2Furious: Linear in Complexity, Competitive in Accuracy
+
+- **ID:** `2602.17363` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Gabriel Mongaras, Eric C. Larson
+- **abs:** https://arxiv.org/abs/2602.17363 · **PDF:** https://arxiv.org/pdf/2602.17363v2
+
+## 11179. Shortcut learning in geometric knot classification
+
+- **ID:** `2602.17350` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Djordje Mihajlovic, Davide Michieletto
+- **abs:** https://arxiv.org/abs/2602.17350 · **PDF:** https://arxiv.org/pdf/2602.17350v1
+
+## 11180. Partial Optimality in the Preordering Problem
+
+- **ID:** `2602.17346` · **primary:** `cs.DM` · **Submitted:** 2026-02-19
+- **Authors:** David Stein, Jannik Irmai, Bjoern Andres
+- **abs:** https://arxiv.org/abs/2602.17346 · **PDF:** https://arxiv.org/pdf/2602.17346v1
+
+## 11181. What Breaks Embodied AI Security:LLM Vulnerabilities, CPS Flaws,or Something Else?
+
+- **ID:** `2602.17345` · **primary:** `cs.CR` · **Submitted:** 2026-02-19
+- **Authors:** Boyang Ma, Hechuan Guo, Peizhuo Lv, et al.
+- **abs:** https://arxiv.org/abs/2602.17345 · **PDF:** https://arxiv.org/pdf/2602.17345v1
+
+## 11182. From Subtle to Significant: Prompt-Driven Self-Improving Optimization in Test-Time Graph OOD Detection
+
+- **ID:** `2602.17342` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Luzhi Wang, Xuanshuo Fu, He Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.17342 · **PDF:** https://arxiv.org/pdf/2602.17342v1
+
+## 11183. SubQuad: Near-Quadratic-Free Structure Inference with Distribution-Balanced Objectives in Adaptive Receptor framework
+
+- **ID:** `2602.17330` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Rong Fu, Zijian Zhang, Kun Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.17330 · **PDF:** https://arxiv.org/pdf/2602.17330v2
+
+## 11184. WebFAQ 2.0: A Multilingual QA Dataset with Mined Hard Negatives for Dense Retrieval
+
+- **ID:** `2602.17327` · **primary:** `cs.IR` · **Submitted:** 2026-02-19
+- **Authors:** Michael Dinzinger, Laura Caspari, Ali Salman, et al.
+- **abs:** https://arxiv.org/abs/2602.17327 · **PDF:** https://arxiv.org/pdf/2602.17327v1
+
+## 11185. Bayesian Optimality of In-Context Learning with Selective State Spaces
+
+- **ID:** `2602.17744` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Di Zhang, Jiaqi Xing
+- **abs:** https://arxiv.org/abs/2602.17744 · **PDF:** https://arxiv.org/pdf/2602.17744v1
+
+## 11186. The Sound of Death: Deep Learning Reveals Vascular Damage from Carotid Ultrasound
+
+- **ID:** `2602.17321` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Christoph Balada, Aida Romano-Martinez, Payal Varshney, et al.
+- **abs:** https://arxiv.org/abs/2602.17321 · **PDF:** https://arxiv.org/pdf/2602.17321v1
+
+## 11187. Provable Adversarial Robustness in In-Context Learning
+
+- **ID:** `2602.17743` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Di Zhang
+- **abs:** https://arxiv.org/abs/2602.17743 · **PDF:** https://arxiv.org/pdf/2602.17743v1
+
+## 11188. Same Meaning, Different Scores: Lexical and Syntactic Sensitivity in LLM Evaluation
+
+- **ID:** `2602.17316` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Bogdan Kostić, Conor Fallon, Julian Risch, et al.
+- **abs:** https://arxiv.org/abs/2602.17316 · **PDF:** https://arxiv.org/pdf/2602.17316v1
+
+## 11189. Flickering Multi-Armed Bandits
+
+- **ID:** `2602.17315` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Sourav Chakraborty, Amit Kiran Rege, Claire Monteleoni, et al.
+- **abs:** https://arxiv.org/abs/2602.17315 · **PDF:** https://arxiv.org/pdf/2602.17315v1
+
+## 11190. Open Datasets in Learning Analytics: Trends, Challenges, and Best PRACTICE
+
+- **ID:** `2602.17314` · **primary:** `cs.CY` · **Submitted:** 2026-02-19
+- **Authors:** Valdemar Švábenský, Brendan Flanagan, Erwin Daniel López Zapata, et al.
+- **abs:** https://arxiv.org/abs/2602.17314 · **PDF:** https://arxiv.org/pdf/2602.17314v1
+
+## 11191. LexiSafe: Offline Safe Reinforcement Learning with Lexicographic Safety-Reward Hierarchy
+
+- **ID:** `2602.17312` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Hsin-Jung Yang, Zhanhong Jiang, Prajwal Koirala, et al.
+- **abs:** https://arxiv.org/abs/2602.17312 · **PDF:** https://arxiv.org/pdf/2602.17312v2
+
+## 11192. MedClarify: An information-seeking AI agent for medical diagnosis with case-specific follow-up questions
+
+- **ID:** `2602.17308` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Hui Min Wong, Philip Heesen, Pascal Janetzky, et al.
+- **abs:** https://arxiv.org/abs/2602.17308 · **PDF:** https://arxiv.org/pdf/2602.17308v1
+
+## 11193. ArXiv-to-Model: A Practical Study of Scientific LM Training
+
+- **ID:** `2602.17288` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Anuj Gupta
+- **abs:** https://arxiv.org/abs/2602.17288 · **PDF:** https://arxiv.org/pdf/2602.17288v1
+
+## 11194. Representation Collapse in Machine Translation Through the Lens of Angular Dispersion
+
+- **ID:** `2602.17287` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Evgeniia Tokarchuk, Maya K. Nachesa, Sergey Troshin, et al.
+- **abs:** https://arxiv.org/abs/2602.17287 · **PDF:** https://arxiv.org/pdf/2602.17287v1
+
+## 11195. Efficient privacy loss accounting for subsampling and random allocation
+
+- **ID:** `2602.17284` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Vitaly Feldman, Moshe Shenfeld
+- **abs:** https://arxiv.org/abs/2602.17284 · **PDF:** https://arxiv.org/pdf/2602.17284v1
+
+## 11196. Towards Cross-lingual Values Assessment: A Consensus-Pluralism Perspective
+
+- **ID:** `2602.17283` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Yukun Chen, Xinyu Zhang, Jialong Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.17283 · **PDF:** https://arxiv.org/pdf/2602.17283v1
+
+## 11197. DIALECTIC: A Multi-Agent System for Startup Evaluation
+
+- **ID:** `2603.12274` · **primary:** `cs.MA` · **Submitted:** 2026-02-19
+- **Authors:** Jae Yoon Bae, Simon Malberg, Joyce Galang, et al.
+- **abs:** https://arxiv.org/abs/2603.12274 · **PDF:** https://arxiv.org/pdf/2603.12274v1
+
+## 11198. Quantum Scrambling Born Machine
+
+- **ID:** `2602.17281` · **primary:** `quant-ph` · **Submitted:** 2026-02-19
+- **Authors:** Marcin Płodzień
+- **abs:** https://arxiv.org/abs/2602.17281 · **PDF:** https://arxiv.org/pdf/2602.17281v1
+
+## 11199. RLGT: A reinforcement learning framework for extremal graph theory
+
+- **ID:** `2602.17276` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Ivan Damnjanović, Uroš Milivojević, Irena Đorđević, et al.
+- **abs:** https://arxiv.org/abs/2602.17276 · **PDF:** https://arxiv.org/pdf/2602.17276v1
+
+## 11200. Gaussian surrogates do well on Poisson inverse problems
+
+- **ID:** `2602.17274` · **primary:** `eess.IV` · **Submitted:** 2026-02-19
+- **Authors:** Alexandra Spitzer, Lorenzo Baldassari, Valentin Derbanot, et al.
+- **abs:** https://arxiv.org/abs/2602.17274 · **PDF:** https://arxiv.org/pdf/2602.17274v1
+
+## 11201. Federated Latent Space Alignment for Multi-user Semantic Communications
+
+- **ID:** `2602.17271` · **primary:** `cs.IT` · **Submitted:** 2026-02-19
+- **Authors:** Giuseppe Di Poce, Mario Edoardo Pandolfo, Emilio Calvanese Strinati, et al.
+- **abs:** https://arxiv.org/abs/2602.17271 · **PDF:** https://arxiv.org/pdf/2602.17271v1
+
+## 11202. Unified Latents (UL): How to train your latents
+
+- **ID:** `2602.17270` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Jonathan Heek, Emiel Hoogeboom, Thomas Mensink, et al.
+- **abs:** https://arxiv.org/abs/2602.17270 · **PDF:** https://arxiv.org/pdf/2602.17270v1
+
+## 11203. Learning a Latent Pulse Shape Interface for Photoinjector Laser Systems
+
+- **ID:** `2602.17263` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Alexander Klemps, Denis Ilia, Pradeep Kr. Banerjee, et al.
+- **abs:** https://arxiv.org/abs/2602.17263 · **PDF:** https://arxiv.org/pdf/2602.17263v1
+
+## 11204. Quantifying and Mitigating Socially Desirable Responding in LLMs: A Desirability-Matched Graded Forced-Choice Psychometric Study
+
+- **ID:** `2602.17262` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Kensuke Okada, Yui Furukawa, Kyosuke Bunji
+- **abs:** https://arxiv.org/abs/2602.17262 · **PDF:** https://arxiv.org/pdf/2602.17262v1
+
+## 11205. Structured Prototype-Guided Adaptation for EEG Foundation Models
+
+- **ID:** `2602.17251` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Jingying Ma, Feng Wu, Yucheng Xing, et al.
+- **abs:** https://arxiv.org/abs/2602.17251 · **PDF:** https://arxiv.org/pdf/2602.17251v1
+
+## 11206. Web Verbs: Typed Abstractions for Reliable Task Composition on the Agentic Web
+
+- **ID:** `2602.17245` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Linxi Jiang, Rui Xi, Zhijie Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.17245 · **PDF:** https://arxiv.org/pdf/2602.17245v1
+
+## 11207. CounterFlowNet: From Minimal Changes to Meaningful Counterfactual Explanations
+
+- **ID:** `2602.17244` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Oleksii Furman, Patryk Marszałek, Jan Masłowski, et al.
+- **abs:** https://arxiv.org/abs/2602.17244 · **PDF:** https://arxiv.org/pdf/2602.17244v1
+
+## 11208. TAPO-Structured Description Logic for Information Behavior: Procedural and Oracle-Based Extensions
+
+- **ID:** `2602.17242` · **primary:** `cs.LO` · **Submitted:** 2026-02-19
+- **Authors:** Takao Inoué
+- **abs:** https://arxiv.org/abs/2602.17242 · **PDF:** https://arxiv.org/pdf/2602.17242v1
+
+## 11209. All Leaks Count, Some Count More: Interpretable Temporal Contamination Detection in LLM Backtesting
+
+- **ID:** `2602.17234` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Zeyu Zhang, Ryan Chen, Bradly C. Stadie
+- **abs:** https://arxiv.org/abs/2602.17234 · **PDF:** https://arxiv.org/pdf/2602.17234v1
+
+## 11210. Mechanistic Interpretability of Cognitive Complexity in LLMs via Linear Probing using Bloom's Taxonomy
+
+- **ID:** `2602.17229` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Bianca Raimondi, Maurizio Gabbrielli
+- **abs:** https://arxiv.org/abs/2602.17229 · **PDF:** https://arxiv.org/pdf/2602.17229v1
+
+## 11211. Privacy-Preserving Mechanisms Enable Cheap Verifiable Inference of LLMs
+
+- **ID:** `2602.17223` · **primary:** `cs.CR` · **Submitted:** 2026-02-19
+- **Authors:** Arka Pal, Louai Zahran, William Gvozdjak, et al.
+- **abs:** https://arxiv.org/abs/2602.17223 · **PDF:** https://arxiv.org/pdf/2602.17223v1
+
+## 11212. Decoding the Human Factor: High Fidelity Behavioral Prediction for Strategic Foresight
+
+- **ID:** `2602.17222` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Ben Yellin, Ehud Ezra, Mark Foreman, et al.
+- **abs:** https://arxiv.org/abs/2602.17222 · **PDF:** https://arxiv.org/pdf/2602.17222v1
+
+## 11213. From Labor to Collaboration: A Methodological Experiment Using AI Agents to Augment Research Perspectives in Taiwan's Humanities and Social Sciences
+
+- **ID:** `2602.17221` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Yi-Chih Huang
+- **abs:** https://arxiv.org/abs/2602.17221 · **PDF:** https://arxiv.org/pdf/2602.17221v1
+
+## 11214. Continual learning and refinement of causal models through dynamic predicate invention
+
+- **ID:** `2602.17217` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Enrique Crespo-Fernandez, Oliver Ray, Telmo de Menezes e Silva Filho, et al.
+- **abs:** https://arxiv.org/abs/2602.17217 · **PDF:** https://arxiv.org/pdf/2602.17217v1
+
+## 11215. Extending quantum theory with AI-assisted deterministic game theory
+
+- **ID:** `2602.17213` · **primary:** `quant-ph` · **Submitted:** 2026-02-19
+- **Authors:** Florian Pauschitz, Ben Moseley, Ghislain Fourny
+- **abs:** https://arxiv.org/abs/2602.17213 · **PDF:** https://arxiv.org/pdf/2602.17213v1
+
+## 11216. MGD: Moment Guided Diffusion for Maximum Entropy Generation
+
+- **ID:** `2602.17211` · **primary:** `stat.ML` · **Submitted:** 2026-02-19
+- **Authors:** Etienne Lempereur, Nathanaël Cuvelle--Magar, Florentin Coeurdoux, et al.
+- **abs:** https://arxiv.org/abs/2602.17211 · **PDF:** https://arxiv.org/pdf/2602.17211v1
+
+## 11217. GeneZip: Region-Aware Compression for Long Context DNA Modeling
+
+- **ID:** `2602.17739` · **primary:** `q-bio.GN` · **Submitted:** 2026-02-19
+- **Authors:** Jianan Zhao, Xixian Liu, Zhihao Zhan, et al.
+- **abs:** https://arxiv.org/abs/2602.17739 · **PDF:** https://arxiv.org/pdf/2602.17739v2
+
+## 11218. SoftDTW-CUDA-Torch: Memory-Efficient GPU-Accelerated Soft Dynamic Time Warping for PyTorch
+
+- **ID:** `2602.17206` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Ron Shapira Weber, Oren Freifeld
+- **abs:** https://arxiv.org/abs/2602.17206 · **PDF:** https://arxiv.org/pdf/2602.17206v1
+
+## 11219. Deeper detection limits in astronomical imaging using self-supervised spatiotemporal denoising
+
+- **ID:** `2602.17205` · **primary:** `astro-ph.IM` · **Submitted:** 2026-02-19
+- **Authors:** Yuduo Guo, Hao Zhang, Mingyu Li, et al.
+- **abs:** https://arxiv.org/abs/2602.17205 · **PDF:** https://arxiv.org/pdf/2602.17205v1
+
+## 11220. CapTrack: Multifaceted Evaluation of Forgetting in LLM Post-Training
+
+- **ID:** `2603.06610` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Lukas Thede, Stefan Winzeck, Zeynep Akata, et al.
+- **abs:** https://arxiv.org/abs/2603.06610 · **PDF:** https://arxiv.org/pdf/2603.06610v1
+
+## 11221. What Makes a Good Doctor Response? A Study on Text-Based Telemedicine
+
+- **ID:** `2602.17194` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Adrian Cosma, Cosmin Dumitrache, Emilian Radoi
+- **abs:** https://arxiv.org/abs/2602.17194 · **PDF:** https://arxiv.org/pdf/2602.17194v3
+
+## 11222. Texo: Formula Recognition within 20M Parameters
+
+- **ID:** `2602.17189` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Sicheng Mao
+- **abs:** https://arxiv.org/abs/2602.17189 · **PDF:** https://arxiv.org/pdf/2602.17189v1
+
+## 11223. Anti-causal domain generalization: Leveraging unlabeled data
+
+- **ID:** `2602.17187` · **primary:** `stat.ML` · **Submitted:** 2026-02-19
+- **Authors:** Sorawit Saengkyongam, Juan L. Gamella, Andrew C. Miller, et al.
+- **abs:** https://arxiv.org/abs/2602.17187 · **PDF:** https://arxiv.org/pdf/2602.17187v1
+
+## 11224. The Bots of Persuasion: Examining How Conversational Agents' Linguistic Expressions of Personality Affect User Perceptions and Decisions
+
+- **ID:** `2602.17185` · **primary:** `cs.HC` · **Submitted:** 2026-02-19
+- **Authors:** Uğur Genç, Heng Gu, Chadha Degachi, et al.
+- **abs:** https://arxiv.org/abs/2602.17185 · **PDF:** https://arxiv.org/pdf/2602.17185v1
+
+## 11225. Robustness and Reasoning Fidelity of Large Language Models in Long-Context Code Question Answering
+
+- **ID:** `2602.17183` · **primary:** `cs.SE` · **Submitted:** 2026-02-19
+- **Authors:** Kishan Maharaj, Nandakishore Menon, Ashita Saxena, et al.
+- **abs:** https://arxiv.org/abs/2602.17183 · **PDF:** https://arxiv.org/pdf/2602.17183v1
+
+## 11226. Symmetry-Driven Generation of Crystal Structures from Composition
+
+- **ID:** `2602.17176` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-19
+- **Authors:** Shi Yin, Jinming Mu, Xudong Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.17176 · **PDF:** https://arxiv.org/pdf/2602.17176v3
+
+## 11227. Continual uncertainty learning
+
+- **ID:** `2602.17174` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Heisei Yonezawa, Ansei Yonezawa, Itsuro Kajiwara
+- **abs:** https://arxiv.org/abs/2602.17174 · **PDF:** https://arxiv.org/pdf/2602.17174v2
+
+## 11228. In-Context Learning in Linear vs. Quadratic Attention Models: An Empirical Study on Regression Tasks
+
+- **ID:** `2602.17171` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Ayush Goel, Arjun Kohli, Sarvagya Somvanshi
+- **abs:** https://arxiv.org/abs/2602.17171 · **PDF:** https://arxiv.org/pdf/2602.17171v1
+
+## 11229. JEPA-DNA: Grounding Genomic Foundation Models through Joint-Embedding Predictive Architectures
+
+- **ID:** `2602.17162` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Ariel Larey, Elay Dahan, Amit Bleiweiss, et al.
+- **abs:** https://arxiv.org/abs/2602.17162 · **PDF:** https://arxiv.org/pdf/2602.17162v1
+
+## 11230. Powering Up Zeroth-Order Training via Subspace Gradient Orthogonalization
+
+- **ID:** `2602.17155` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Yicheng Lang, Changsheng Wang, Yihua Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.17155 · **PDF:** https://arxiv.org/pdf/2602.17155v2
+
+## 11231. TimeOmni-VL: Unified Models for Time Series Understanding and Generation
+
+- **ID:** `2602.17149` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Tong Guan, Sheng Pan, Johan Barthelemy, et al.
+- **abs:** https://arxiv.org/abs/2602.17149 · **PDF:** https://arxiv.org/pdf/2602.17149v1
+
+## 11232. Bonsai: A Framework for Convolutional Neural Network Acceleration Using Criterion-Based Pruning
+
+- **ID:** `2602.17145` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Joseph Bingham, Sam Helmich
+- **abs:** https://arxiv.org/abs/2602.17145 · **PDF:** https://arxiv.org/pdf/2602.17145v1
+
+## 11233. When More Experts Hurt: Underfitting in Multi-Expert Learning to Defer
+
+- **ID:** `2602.17144` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Shuqi Liu, Yuzhou Cao, Lei Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.17144 · **PDF:** https://arxiv.org/pdf/2602.17144v1
+
+## 11234. VP-VAE: Rethinking Vector Quantization via Adaptive Vector Perturbation
+
+- **ID:** `2602.17133` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Linwei Zhai, Han Ding, Mingzhi Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.17133 · **PDF:** https://arxiv.org/pdf/2602.17133v1
+
+## 11235. Efficient Parallel Algorithm for Decomposing Hard CircuitSAT Instances
+
+- **ID:** `2602.17130` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Victor Kondratiev, Irina Gribanova, Alexander Semenov
+- **abs:** https://arxiv.org/abs/2602.17130 · **PDF:** https://arxiv.org/pdf/2602.17130v1
+
+## 11236. The Emergence of Lab-Driven Alignment Signatures: A Psychometric Framework for Auditing Latent Bias and Compounding Risk in Generative AI
+
+- **ID:** `2602.17127` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Dusan Bosnjakovic
+- **abs:** https://arxiv.org/abs/2602.17127 · **PDF:** https://arxiv.org/pdf/2602.17127v1
+
+## 11237. 3D Scene Rendering with Multimodal Gaussian Splatting
+
+- **ID:** `2602.17124` · **primary:** `cs.CV` · **Submitted:** 2026-02-19
+- **Authors:** Chi-Shiang Gau, Konstantinos D. Polyzos, Athanasios Bacharis, et al.
+- **abs:** https://arxiv.org/abs/2602.17124 · **PDF:** https://arxiv.org/pdf/2602.17124v1
+
+## 11238. TIFO: Time-Invariant Frequency Operator for Stationarity-Aware Representation Learning in Time Series
+
+- **ID:** `2602.17122` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Xihao Piao, Zheng Chen, Lingwei Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.17122 · **PDF:** https://arxiv.org/pdf/2602.17122v1
+
+## 11239. i-PhysGaussian: Implicit Physical Simulation for 3D Gaussian Splatting
+
+- **ID:** `2602.17117` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Yicheng Cao, Zhuo Huang, Yu Yao, et al.
+- **abs:** https://arxiv.org/abs/2602.17117 · **PDF:** https://arxiv.org/pdf/2602.17117v1
+
+## 11240. Position: AI Agents Are Not (Yet) a Panacea for Social Simulation
+
+- **ID:** `2603.00113` · **primary:** `cs.MA` · **Submitted:** 2026-02-19
+- **Authors:** Yiming Li, Dacheng Tao
+- **abs:** https://arxiv.org/abs/2603.00113 · **PDF:** https://arxiv.org/pdf/2603.00113v1
+
+## 11241. Epistemology of Generative AI: The Geometry of Knowing
+
+- **ID:** `2602.17116` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Ilya Levin
+- **abs:** https://arxiv.org/abs/2602.17116 · **PDF:** https://arxiv.org/pdf/2602.17116v1
+
+## 11242. A Benchmark Analysis of Graph and Non-Graph Methods for Caenorhabditis Elegans Neuron Classification
+
+- **ID:** `2603.02241` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-19
+- **Authors:** Jingqi Lu, Keqi Han, Yun Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.02241 · **PDF:** https://arxiv.org/pdf/2603.02241v1
+
+## 11243. Semi-Supervised Learning on Graphs using Graph Neural Networks
+
+- **ID:** `2602.17115` · **primary:** `stat.ML` · **Submitted:** 2026-02-19
+- **Authors:** Juntong Chen, Claire Donnat, Olga Klopp, et al.
+- **abs:** https://arxiv.org/abs/2602.17115 · **PDF:** https://arxiv.org/pdf/2602.17115v1
+
+## 11244. Instructor-Aligned Knowledge Graphs for Personalized Learning
+
+- **ID:** `2602.17111` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Abdulrahman AlRabah, Priyanka Kargupta, Jiawei Han, et al.
+- **abs:** https://arxiv.org/abs/2602.17111 · **PDF:** https://arxiv.org/pdf/2602.17111v1
+
+## 11245. Projective Psychological Assessment of Large Multimodal Models Using Thematic Apperception Tests
+
+- **ID:** `2602.17108` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Anton Dzega, Aviad Elyashar, Ortal Slobodin, et al.
+- **abs:** https://arxiv.org/abs/2602.17108 · **PDF:** https://arxiv.org/pdf/2602.17108v1
+
+## 11246. Owen-based Semantics and Hierarchy-Aware Explanation (O-Shap)
+
+- **ID:** `2602.17107` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Xiangyu Zhou, Chenhan Xiao, Yang Weng
+- **abs:** https://arxiv.org/abs/2602.17107 · **PDF:** https://arxiv.org/pdf/2602.17107v1
+
+## 11247. Toward Trustworthy Evaluation of Sustainability Rating Methodologies: A Human-AI Collaborative Framework for Benchmark Dataset Construction
+
+- **ID:** `2602.17106` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Xiaoran Cai, Wang Yang, Xiyu Ren, et al.
+- **abs:** https://arxiv.org/abs/2602.17106 · **PDF:** https://arxiv.org/pdf/2602.17106v1
+
+## 11248. Simplify to Amplify: Achieving Information-Theoretic Bounds with Fewer Steps in Spectral Community Detection
+
+- **ID:** `2602.17104` · **primary:** `cs.SI` · **Submitted:** 2026-02-19
+- **Authors:** Sie Hendrata Dharmawan, Peter Chin
+- **abs:** https://arxiv.org/abs/2602.17104 · **PDF:** https://arxiv.org/pdf/2602.17104v1
+
+## 11249. Online Learning with Improving Agents: Multiclass, Budgeted Agents and Bandit Learners
+
+- **ID:** `2602.17103` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Sajad Ashkezari, Shai Ben-David
+- **abs:** https://arxiv.org/abs/2602.17103 · **PDF:** https://arxiv.org/pdf/2602.17103v1
+
+## 11250. Operationalization of Machine Learning with Serverless Architecture: An Industrial Operationalization of Machine Learning with Serverless Architecture: An Industrial Implementation for Harmonized System Code Prediction
+
+- **ID:** `2602.17102` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Sai Vineeth Kandappareddigari, Santhoshkumar Jagadish, Gauri Verma, et al.
+- **abs:** https://arxiv.org/abs/2602.17102 · **PDF:** https://arxiv.org/pdf/2602.17102v1
+
+## 11251. Valid Feature-Level Inference for Tabular Foundation Models via the Conditional Randomization Test
+
+- **ID:** `2603.06609` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Mohamed Salem
+- **abs:** https://arxiv.org/abs/2603.06609 · **PDF:** https://arxiv.org/pdf/2603.06609v1
+
+## 11252. Deep Reinforcement Learning for Optimal Portfolio Allocation: A Comparative Study with Mean-Variance Optimization
+
+- **ID:** `2602.17098` · **primary:** `q-fin.PM` · **Submitted:** 2026-02-19
+- **Authors:** Srijan Sood, Kassiani Papasotiriou, Marius Vaiciulis, et al.
+- **abs:** https://arxiv.org/abs/2602.17098 · **PDF:** https://arxiv.org/pdf/2602.17098v1
+
+## 11253. ASFL: An Adaptive Model Splitting and Resource Allocation Framework for Split Federated Learning
+
+- **ID:** `2603.04437` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Chuiyang Meng, Ming Tang, Vincent W. S. Wong
+- **abs:** https://arxiv.org/abs/2603.04437 · **PDF:** https://arxiv.org/pdf/2603.04437v1
+
+## 11254. ZorBA: Zeroth-order Federated Fine-tuning of LLMs with Heterogeneous Block Activation
+
+- **ID:** `2603.04436` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Chuiyang Meng, Ming Tang, Vincent W. S. Wong
+- **abs:** https://arxiv.org/abs/2603.04436 · **PDF:** https://arxiv.org/pdf/2603.04436v1
+
+## 11255. Agentic Wireless Communication for 6G: Intent-Aware and Continuously Evolving Physical-Layer Intelligence
+
+- **ID:** `2602.17096` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Zhaoyang Li, Xingzhi Jin, Junyu Pan, et al.
+- **abs:** https://arxiv.org/abs/2602.17096 · **PDF:** https://arxiv.org/pdf/2602.17096v1
+
+## 11256. FLoRG: Federated Fine-tuning with Low-rank Gram Matrices and Procrustes Alignment
+
+- **ID:** `2602.17095` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Chuiyang Meng, Ming Tang, Vincent W. S. Wong
+- **abs:** https://arxiv.org/abs/2602.17095 · **PDF:** https://arxiv.org/pdf/2602.17095v2
+
+## 11257. A Locality Radius Framework for Understanding Relational Inductive Bias in Database Learning
+
+- **ID:** `2602.17092` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Aadi Joshi, Kavya Bhand
+- **abs:** https://arxiv.org/abs/2602.17092 · **PDF:** https://arxiv.org/pdf/2602.17092v1
+
+## 11258. Synergizing Transport-Based Generative Models and Latent Geometry for Stochastic Closure Modeling
+
+- **ID:** `2602.17089` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Xinghao Dong, Huchen Yang, Jin-long Wu
+- **abs:** https://arxiv.org/abs/2602.17089 · **PDF:** https://arxiv.org/pdf/2602.17089v1
+
+## 11259. MeGU: Machine-Guided Unlearning with Target Feature Disentanglement
+
+- **ID:** `2602.17088` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Haoyu Wang, Zhuo Huang, Xiaolong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.17088 · **PDF:** https://arxiv.org/pdf/2602.17088v1
+
+## 11260. Dynamic Decision-Making under Model Misspecification: A Stochastic Stability Approach
+
+- **ID:** `2602.17086` · **primary:** `econ.TH` · **Submitted:** 2026-02-19
+- **Authors:** Xinyu Dai, Daniel Chen, Yian Qian
+- **abs:** https://arxiv.org/abs/2602.17086 · **PDF:** https://arxiv.org/pdf/2602.17086v1
+
+## 11261. How AI Coding Agents Communicate: A Study of Pull Request Description Characteristics and Human Review Responses
+
+- **ID:** `2602.17084` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Kan Watanabe, Rikuto Tsuchida, Takahiro Monno, et al.
+- **abs:** https://arxiv.org/abs/2602.17084 · **PDF:** https://arxiv.org/pdf/2602.17084v1
+
+## 11262. Adam Improves Muon: Adaptive Moment Estimation with Orthogonalized Momentum
+
+- **ID:** `2602.17080` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Minxin Zhang, Yuxuan Liu, Hayden Schaeffer
+- **abs:** https://arxiv.org/abs/2602.17080 · **PDF:** https://arxiv.org/pdf/2602.17080v2
+
+## 11263. BankMathBench: A Benchmark for Numerical Reasoning in Banking Scenarios
+
+- **ID:** `2602.17072` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Yunseung Lee, Subin Kim, Youngjun Kwak, et al.
+- **abs:** https://arxiv.org/abs/2602.17072 · **PDF:** https://arxiv.org/pdf/2602.17072v2
+
+## 11264. AdvSynGNN: Structure-Adaptive Graph Neural Nets via Adversarial Synthesis and Self-Corrective Propagation
+
+- **ID:** `2602.17071` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Rong Fu, Muge Qi, Chunlei Meng, et al.
+- **abs:** https://arxiv.org/abs/2602.17071 · **PDF:** https://arxiv.org/pdf/2602.17071v1
+
+## 11265. General sample size analysis for probabilities of causation: a delta method approach
+
+- **ID:** `2602.17070` · **primary:** `stat.ME` · **Submitted:** 2026-02-19
+- **Authors:** Tianyuan Cheng, Ruirui Mao, Judea Pearl, et al.
+- **abs:** https://arxiv.org/abs/2602.17070 · **PDF:** https://arxiv.org/pdf/2602.17070v1
+
+## 11266. Spatio-temporal dual-stage hypergraph MARL for human-centric multimodal corridor traffic signal control
+
+- **ID:** `2602.17068` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Xiaocai Zhang, Neema Nassir, Milad Haghani
+- **abs:** https://arxiv.org/abs/2602.17068 · **PDF:** https://arxiv.org/pdf/2602.17068v1
+
+## 11267. Predictive Batch Scheduling: Accelerating Language Model Training Through Loss-Aware Sample Prioritization
+
+- **ID:** `2602.17066` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Sumedh Rasal
+- **abs:** https://arxiv.org/abs/2602.17066 · **PDF:** https://arxiv.org/pdf/2602.17066v1
+
+## 11268. Scaling Strategy, Not Compute: A Stand-Alone, Open-Source StarCraft II Benchmark for Accessible Reinforcement Learning Research
+
+- **ID:** `2603.06608` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Sourav Panda, Shreyash Kale, Tanmay Ambadkar, et al.
+- **abs:** https://arxiv.org/abs/2603.06608 · **PDF:** https://arxiv.org/pdf/2603.06608v1
+
+## 11269. Sign Lock-In: Randomly Initialized Weight Signs Persist and Bottleneck Sub-Bit Model Compression
+
+- **ID:** `2602.17063` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Akira Sakai, Yuma Ichikawa
+- **abs:** https://arxiv.org/abs/2602.17063 · **PDF:** https://arxiv.org/pdf/2602.17063v1
+
+## 11270. Retaining Suboptimal Actions to Follow Shifting Optima in Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.17062` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Yonghyeon Jo, Sunwoo Lee, Seungyul Han
+- **abs:** https://arxiv.org/abs/2602.17062 · **PDF:** https://arxiv.org/pdf/2602.17062v1
+
+## 11271. ALPS: A Diagnostic Challenge Set for Arabic Linguistic & Pragmatic Reasoning
+
+- **ID:** `2602.17054` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Hussein S. Al-Olimat, Ahmad Alshareef
+- **abs:** https://arxiv.org/abs/2602.17054 · **PDF:** https://arxiv.org/pdf/2602.17054v1
+
+## 11272. RFEval: Benchmarking Reasoning Faithfulness under Counterfactual Reasoning Intervention in Large Reasoning Models
+
+- **ID:** `2602.17053` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Yunseok Han, Yejoon Lee, Jaeyoung Do
+- **abs:** https://arxiv.org/abs/2602.17053 · **PDF:** https://arxiv.org/pdf/2602.17053v3
+
+## 11273. Evaluating Cross-Lingual Classification Approaches Enabling Topic Discovery for Multilingual Social Media Data
+
+- **ID:** `2602.17051` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Deepak Uniyal, Md Abul Bashar, Richi Nayak
+- **abs:** https://arxiv.org/abs/2602.17051 · **PDF:** https://arxiv.org/pdf/2602.17051v1
+
+## 11274. Multi-Probe Zero Collision Hash (MPZCH): Mitigating Embedding Collisions and Enhancing Model Freshness in Large-Scale Recommenders
+
+- **ID:** `2602.17050` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Ziliang Zhao, Bi Xue, Emma Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.17050 · **PDF:** https://arxiv.org/pdf/2602.17050v2
+
+## 11275. IntentCUA: Learning Intent-level Representations for Skill Abstraction and Multi-Agent Planning in Computer-Use Agents
+
+- **ID:** `2602.17049` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Seoyoung Lee, Seobin Yoon, Seongbeen Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.17049 · **PDF:** https://arxiv.org/pdf/2602.17049v2
+
+## 11276. Large Language Models Persuade Without Planning Theory of Mind
+
+- **ID:** `2602.17045` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Jared Moore, Rasmus Overmark, Ned Cooper, et al.
+- **abs:** https://arxiv.org/abs/2602.17045 · **PDF:** https://arxiv.org/pdf/2602.17045v1
+
+## 11277. Phase-Aware Mixture of Experts for Agentic Reinforcement Learning
+
+- **ID:** `2602.17038` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Shengtian Yang, Yu Li, Shuo He, et al.
+- **abs:** https://arxiv.org/abs/2602.17038 · **PDF:** https://arxiv.org/pdf/2602.17038v2
+
+## 11278. Wink: Recovering from Misbehaviors in Coding Agents
+
+- **ID:** `2602.17037` · **primary:** `cs.SE` · **Submitted:** 2026-02-19
+- **Authors:** Rahul Nanda, Chandra Maddila, Smriti Jha, et al.
+- **abs:** https://arxiv.org/abs/2602.17037 · **PDF:** https://arxiv.org/pdf/2602.17037v2
+
+## 11279. LiveGraph: Active-Structure Neural Re-ranking for Exercise Recommendation
+
+- **ID:** `2602.17036` · **primary:** `cs.IR` · **Submitted:** 2026-02-19
+- **Authors:** Rong Fu, Zijian Zhang, Haiyun Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.17036 · **PDF:** https://arxiv.org/pdf/2602.17036v1
+
+## 11280. Forecasting Anomaly Precursors via Uncertainty-Aware Time-Series Ensembles
+
+- **ID:** `2602.17028` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Hyeongwon Kang, Jinwoo Park, Seunghun Han, et al.
+- **abs:** https://arxiv.org/abs/2602.17028 · **PDF:** https://arxiv.org/pdf/2602.17028v1
+
+## 11281. Transforming Behavioral Neuroscience Discovery with In-Context Learning and AI-Enhanced Tensor Methods
+
+- **ID:** `2602.17027` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Paimon Goulart, Jordan Steinhauser, Dawon Ahn, et al.
+- **abs:** https://arxiv.org/abs/2602.17027 · **PDF:** https://arxiv.org/pdf/2602.17027v1
+
+## 11282. WS-GRPO: Weakly-Supervised Group-Relative Policy Optimization for Rollout-Efficient Reasoning
+
+- **ID:** `2602.17025` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Gagan Mundada, Zihan Huang, Rohan Surana, et al.
+- **abs:** https://arxiv.org/abs/2602.17025 · **PDF:** https://arxiv.org/pdf/2602.17025v1
+
+## 11283. ReIn: Conversational Error Recovery with Reasoning Inception
+
+- **ID:** `2602.17022` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Takyoung Kim, Jinseok Nam, Chandrayee Basu, et al.
+- **abs:** https://arxiv.org/abs/2602.17022 · **PDF:** https://arxiv.org/pdf/2602.17022v1
+
+## 11284. M2F: Automated Formalization of Mathematical Literature at Scale
+
+- **ID:** `2602.17016` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Zichen Wang, Wanli Ma, Zhenyu Ming, et al.
+- **abs:** https://arxiv.org/abs/2602.17016 · **PDF:** https://arxiv.org/pdf/2602.17016v1
+
+## 11285. Action-Graph Policies: Learning Action Co-dependencies in Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.17009` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Nikunj Gupta, James Zachary Hare, Jesse Milzman, et al.
+- **abs:** https://arxiv.org/abs/2602.17009 · **PDF:** https://arxiv.org/pdf/2602.17009v2
+
+## 11286. Arcee Trinity Large Technical Report
+
+- **ID:** `2602.17004` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Varun Singh, Lucas Krauss, Sami Jaghouar, et al.
+- **abs:** https://arxiv.org/abs/2602.17004 · **PDF:** https://arxiv.org/pdf/2602.17004v1
+
+## 11287. Persona2Web: Benchmarking Personalized Web Agents for Contextual Reasoning with User History
+
+- **ID:** `2602.17003` · **primary:** `cs.CL` · **Submitted:** 2026-02-19
+- **Authors:** Serin Kim, Sangam Lee, Dongha Lee
+- **abs:** https://arxiv.org/abs/2602.17003 · **PDF:** https://arxiv.org/pdf/2602.17003v1
+
+## 11288. Sonar-TS: Search-Then-Verify Natural Language Querying for Time Series Databases
+
+- **ID:** `2602.17001` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Zhao Tan, Yiji Zhao, Shiyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.17001 · **PDF:** https://arxiv.org/pdf/2602.17001v1
+
+## 11289. Exploring LLMs for User Story Extraction from Mockups
+
+- **ID:** `2602.16997` · **primary:** `cs.SE` · **Submitted:** 2026-02-19
+- **Authors:** Diego Firmenich, Leandro Antonelli, Bruno Pazos, et al.
+- **abs:** https://arxiv.org/abs/2602.16997 · **PDF:** https://arxiv.org/pdf/2602.16997v1
+
+## 11290. Dynamic Delayed Tree Expansion For Improved Multi-Path Speculative Decoding
+
+- **ID:** `2602.16994` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Rahul Thomas, Teo Kitanovski, Micah Goldblum, et al.
+- **abs:** https://arxiv.org/abs/2602.16994 · **PDF:** https://arxiv.org/pdf/2602.16994v1
+
+## 11291. Conv-FinRe: A Conversational and Longitudinal Benchmark for Utility-Grounded Financial Recommendation
+
+- **ID:** `2602.16990` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Yan Wang, Yi Han, Lingfei Qian, et al.
+- **abs:** https://arxiv.org/abs/2602.16990 · **PDF:** https://arxiv.org/pdf/2602.16990v1
+
+## 11292. Fundamental Limits of Black-Box Safety Evaluation: Information-Theoretic and Computational Barriers from Latent Context Conditioning
+
+- **ID:** `2602.16984` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Vishal Srivastava
+- **abs:** https://arxiv.org/abs/2602.16984 · **PDF:** https://arxiv.org/pdf/2602.16984v1
+
+## 11293. Discovering Universal Activation Directions for PII Leakage in Language Models
+
+- **ID:** `2602.16980` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Leo Marchyok, Zachary Coalson, Sungho Keum, et al.
+- **abs:** https://arxiv.org/abs/2602.16980 · **PDF:** https://arxiv.org/pdf/2602.16980v1
+
+## 11294. Characterizing the Predictive Impact of Modalities with Supervised Latent-Variable Modeling
+
+- **ID:** `2602.16979` · **primary:** `cs.CV` · **Submitted:** 2026-02-19
+- **Authors:** Divyam Madaan, Sumit Chopra, Kyunghyun Cho
+- **abs:** https://arxiv.org/abs/2602.16979 · **PDF:** https://arxiv.org/pdf/2602.16979v1
+
+## 11295. Fail-Closed Alignment for Large Language Models
+
+- **ID:** `2602.16977` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Zachary Coalson, Beth Sohler, Aiden Gabriel, et al.
+- **abs:** https://arxiv.org/abs/2602.16977 · **PDF:** https://arxiv.org/pdf/2602.16977v1
+
+## 11296. HQFS: Hybrid Quantum Classical Financial Security with VQC Forecasting, QUBO Annealing, and Audit-Ready Post-Quantum Signing
+
+- **ID:** `2602.16976` · **primary:** `cs.AI` · **Submitted:** 2026-02-19
+- **Authors:** Srikumar Nayak
+- **abs:** https://arxiv.org/abs/2602.16976 · **PDF:** https://arxiv.org/pdf/2602.16976v1
+
+## 11297. DDiT: Dynamic Patch Scheduling for Efficient Diffusion Transformers
+
+- **ID:** `2602.16968` · **primary:** `cs.CV` · **Submitted:** 2026-02-19
+- **Authors:** Dahye Kim, Deepti Ghadiyaram, Raghudeep Gadde
+- **abs:** https://arxiv.org/abs/2602.16968 · **PDF:** https://arxiv.org/pdf/2602.16968v1
+
+## 11298. Early-Warning Signals of Grokking via Loss-Landscape Geometry
+
+- **ID:** `2602.16967` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Yongzhong Xu
+- **abs:** https://arxiv.org/abs/2602.16967 · **PDF:** https://arxiv.org/pdf/2602.16967v2
+
+## 11299. A Unified Framework for Locality in Scalable MARL
+
+- **ID:** `2602.16966` · **primary:** `cs.LG` · **Submitted:** 2026-02-19
+- **Authors:** Sourav Chakraborty, Amit Kiran Rege, Claire Monteleoni, et al.
+- **abs:** https://arxiv.org/abs/2602.16966 · **PDF:** https://arxiv.org/pdf/2602.16966v1
+
+## 11300. Multi-Agent Lipschitz Bandits
+
+- **ID:** `2602.16965` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Sourav Chakraborty, Amit Kiran Rege, Claire Monteleoni, et al.
+- **abs:** https://arxiv.org/abs/2602.16965 · **PDF:** https://arxiv.org/pdf/2602.16965v1
+
+## 11301. Greedy Multi-Path Block Verification for Faster Decoding in Speculative Sampling
+
+- **ID:** `2602.16961` · **primary:** `cs.IT` · **Submitted:** 2026-02-18
+- **Authors:** Rahul Thomas, Arka Pal
+- **abs:** https://arxiv.org/abs/2602.16961 · **PDF:** https://arxiv.org/pdf/2602.16961v1
+
+## 11302. Eigenmood Space: Uncertainty-Aware Spectral Graph Analysis of Psychological Patterns in Classical Persian Poetry
+
+- **ID:** `2602.16959` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Kourosh Shahnazari, Seyed Moein Ayyoubzadeh, Mohammadali Keshtparvar
+- **abs:** https://arxiv.org/abs/2602.16959 · **PDF:** https://arxiv.org/pdf/2602.16959v1
+
+## 11303. Automating Agent Hijacking via Structural Template Injection
+
+- **ID:** `2602.16958` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Xinhao Deng, Jiaqing Wu, Miao Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.16958 · **PDF:** https://arxiv.org/pdf/2602.16958v1
+
+## 11304. When Semantic Overlap Is Not Enough: Cross-Lingual Euphemism Transfer Between Turkish and English
+
+- **ID:** `2602.16957` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Hasan Can Biyik, Libby Barak, Jing Peng, et al.
+- **abs:** https://arxiv.org/abs/2602.16957 · **PDF:** https://arxiv.org/pdf/2602.16957v1
+
+## 11305. Neural Proposals, Symbolic Guarantees: Neuro-Symbolic Graph Generation with Hard Constraints
+
+- **ID:** `2602.16954` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Chuqin Geng, Li Zhang, Mark Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.16954 · **PDF:** https://arxiv.org/pdf/2602.16954v2
+
+## 11306. LLM4Cov: Execution-Aware Agentic Learning for High-coverage Testbench Generation
+
+- **ID:** `2602.16953` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Hejia Zhang, Zhongming Yu, Chia-Tung Ho, et al.
+- **abs:** https://arxiv.org/abs/2602.16953 · **PDF:** https://arxiv.org/pdf/2602.16953v2
+
+## 11307. BrainRVQ: A High-Fidelity EEG Foundation Model via Dual-Domain Residual Quantization and Hierarchical Autoregression
+
+- **ID:** `2602.16951` · **primary:** `eess.SP` · **Submitted:** 2026-02-18
+- **Authors:** Mingzhe Cui, Tao Chen, Yang Jiao, et al.
+- **abs:** https://arxiv.org/abs/2602.16951 · **PDF:** https://arxiv.org/pdf/2602.16951v1
+
+## 11308. Beyond Message Passing: A Symbolic Alternative for Expressive and Interpretable Graph Learning
+
+- **ID:** `2602.16947` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Chuqin Geng, Li Zhang, Haolin Ye, et al.
+- **abs:** https://arxiv.org/abs/2602.16947 · **PDF:** https://arxiv.org/pdf/2602.16947v2
+
+## 11309. Exact Certification of Data-Poisoning Attacks Using Mixed-Integer Programming
+
+- **ID:** `2602.16944` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Philip Sosnin, Jodie Knapp, Fraser Kennedy, et al.
+- **abs:** https://arxiv.org/abs/2602.16944 · **PDF:** https://arxiv.org/pdf/2602.16944v1
+
+## 11310. Mind the GAP: Text Safety Does Not Transfer to Tool-Call Safety in LLM Agents
+
+- **ID:** `2602.16943` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Arnold Cartagena, Ariane Teixeira
+- **abs:** https://arxiv.org/abs/2602.16943 · **PDF:** https://arxiv.org/pdf/2602.16943v1
+
+## 11311. SourceBench: Can AI Answers Reference Quality Web Sources?
+
+- **ID:** `2602.16942` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Hexi Jin, Stephen Liu, Yuheng Li, et al.
+- **abs:** https://arxiv.org/abs/2602.16942 · **PDF:** https://arxiv.org/pdf/2602.16942v1
+
+## 11312. Nested Training for Mutual Adaptation in Human-AI Teaming
+
+- **ID:** `2602.17737` · **primary:** `cs.RO` · **Submitted:** 2026-02-18
+- **Authors:** Upasana Biswas, Durgesh Kalwar, Subbarao Kambhampati, et al.
+- **abs:** https://arxiv.org/abs/2602.17737 · **PDF:** https://arxiv.org/pdf/2602.17737v1
+
+## 11313. ConvApparel: A Benchmark Dataset and Validation Framework for User Simulators in Conversational Recommenders
+
+- **ID:** `2602.16938` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Ofer Meshi, Krisztian Balog, Sally Goldman, et al.
+- **abs:** https://arxiv.org/abs/2602.16938 · **PDF:** https://arxiv.org/pdf/2602.16938v1
+
+## 11314. DeepContext: Stateful Real-Time Detection of Multi-Turn Adversarial Intent Drift in LLMs
+
+- **ID:** `2602.16935` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Justin Albrethsen, Yash Datta, Kunal Kumar, et al.
+- **abs:** https://arxiv.org/abs/2602.16935 · **PDF:** https://arxiv.org/pdf/2602.16935v1
+
+## 11315. M-estimation under Two-Phase Multiwave Sampling with Applications to Prediction-Powered Inference
+
+- **ID:** `2602.16933` · **primary:** `stat.ME` · **Submitted:** 2026-02-18
+- **Authors:** Dan M. Kluger, Stephen Bates
+- **abs:** https://arxiv.org/abs/2602.16933 · **PDF:** https://arxiv.org/pdf/2602.16933v2
+
+## 11316. RankEvolve: Automating the Discovery of Retrieval Algorithms via LLM-Driven Evolution
+
+- **ID:** `2602.16932` · **primary:** `cs.IR` · **Submitted:** 2026-02-18
+- **Authors:** Jinming Nian, Fangchen Li, Dae Hoon Park, et al.
+- **abs:** https://arxiv.org/abs/2602.16932 · **PDF:** https://arxiv.org/pdf/2602.16932v1
+
+## 11317. Narrow Fine-Tuning Erodes Safety Alignment in Vision-Language Agents
+
+- **ID:** `2602.16931` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Idhant Gulati, Shivam Raval
+- **abs:** https://arxiv.org/abs/2602.16931 · **PDF:** https://arxiv.org/pdf/2602.16931v2
+
+## 11318. Say It My Way: Exploring Control in Conversational Visual Question Answering with Blind Users
+
+- **ID:** `2602.16930` · **primary:** `cs.HC` · **Submitted:** 2026-02-18
+- **Authors:** Farnaz Zamiri Zeraati, Yang Trista Cao, Yuehan Qiao, et al.
+- **abs:** https://arxiv.org/abs/2602.16930 · **PDF:** https://arxiv.org/pdf/2602.16930v1
+
+## 11319. Discovering Multiagent Learning Algorithms with Large Language Models
+
+- **ID:** `2602.16928` · **primary:** `cs.GT` · **Submitted:** 2026-02-18
+- **Authors:** Zun Li, John Schultz, Daniel Hennes, et al.
+- **abs:** https://arxiv.org/abs/2602.16928 · **PDF:** https://arxiv.org/pdf/2602.16928v2
+
+## 11320. Poisson-MNL Bandit: Nearly Optimal Dynamic Joint Assortment and Pricing with Decision-Dependent Customer Arrivals
+
+- **ID:** `2602.16923` · **primary:** `stat.ML` · **Submitted:** 2026-02-18
+- **Authors:** Junhui Cai, Ran Chen, Qitao Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.16923 · **PDF:** https://arxiv.org/pdf/2602.16923v1
+
+## 11321. Xray-Visual Models: Scaling Vision models on Industry Scale Data
+
+- **ID:** `2602.16918` · **primary:** `cs.CV` · **Submitted:** 2026-02-18
+- **Authors:** Shlok Mishra, Tsung-Yu Lin, Linda Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.16918 · **PDF:** https://arxiv.org/pdf/2602.16918v1
+
+## 11322. A statistical perspective on transformers for small longitudinal cohort data
+
+- **ID:** `2602.16914` · **primary:** `stat.ME` · **Submitted:** 2026-02-18
+- **Authors:** Kiana Farhadyar, Maren Hackenberg, Kira Ahrens, et al.
+- **abs:** https://arxiv.org/abs/2602.16914 · **PDF:** https://arxiv.org/pdf/2602.16914v1
+
+## 11323. A Reversible Semantics for Janus
+
+- **ID:** `2602.16913` · **primary:** `cs.PL` · **Submitted:** 2026-02-18
+- **Authors:** Ivan Lanese, Germán Vidal
+- **abs:** https://arxiv.org/abs/2602.16913 · **PDF:** https://arxiv.org/pdf/2602.16913v2
+
+## 11324. Multi-objective optimization and quantum hybridization of equivariant deep learning interatomic potentials on organic and inorganic compounds
+
+- **ID:** `2602.16908` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-18
+- **Authors:** G. Laskaris, D. Morozov, D. Tarpanov, et al.
+- **abs:** https://arxiv.org/abs/2602.16908 · **PDF:** https://arxiv.org/pdf/2602.16908v1
+
+## 11325. LLM-WikiRace Benchmark: How Far Can LLMs Plan over Real-World Knowledge Graphs?
+
+- **ID:** `2602.16902` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Juliusz Ziomek, William Bankes, Lorenz Wolf, et al.
+- **abs:** https://arxiv.org/abs/2602.16902 · **PDF:** https://arxiv.org/pdf/2602.16902v3
+
+## 11326. AgentLAB: Benchmarking LLM Agents against Long-Horizon Attacks
+
+- **ID:** `2602.16901` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Tanqiu Jiang, Yuhui Wang, Jiacheng Liang, et al.
+- **abs:** https://arxiv.org/abs/2602.16901 · **PDF:** https://arxiv.org/pdf/2602.16901v1
+
+## 11327. MALLVI: A Multi-Agent Framework for Integrated Generalized Robotics Manipulation
+
+- **ID:** `2602.16898` · **primary:** `cs.RO` · **Submitted:** 2026-02-18
+- **Authors:** Mehrshad Taji, Arad Mahdinezhad Kashani, Iman Ahmadi, et al.
+- **abs:** https://arxiv.org/abs/2602.16898 · **PDF:** https://arxiv.org/pdf/2602.16898v4
+
+## 11328. OpenSage: Self-programming Agent Generation Engine
+
+- **ID:** `2602.16891` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Hongwei Li, Zhun Wang, Qinrun Dai, et al.
+- **abs:** https://arxiv.org/abs/2602.16891 · **PDF:** https://arxiv.org/pdf/2602.16891v2
+
+## 11329. Construction of a classification model for dementia among Brazilian adults aged 50 and over
+
+- **ID:** `2602.16887` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** F. S. Menezes, M. C. F. G. Barretto, E. Q. C. Garcia, et al.
+- **abs:** https://arxiv.org/abs/2602.16887 · **PDF:** https://arxiv.org/pdf/2602.16887v1
+
+## 11330. ML-driven detection and reduction of ballast information in multi-modal datasets
+
+- **ID:** `2602.16876` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Yaroslav Solovko
+- **abs:** https://arxiv.org/abs/2602.16876 · **PDF:** https://arxiv.org/pdf/2602.16876v1
+
+## 11331. AdaptOrch: Task-Adaptive Multi-Agent Orchestration in the Era of LLM Performance Convergence
+
+- **ID:** `2602.16873` · **primary:** `cs.MA` · **Submitted:** 2026-02-18
+- **Authors:** Geunbin Yu
+- **abs:** https://arxiv.org/abs/2602.16873 · **PDF:** https://arxiv.org/pdf/2602.16873v1
+
+## 11332. Enhancing Financial Report Question-Answering: A Retrieval-Augmented Generation System with Reranking Analysis
+
+- **ID:** `2603.16877` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Zhiyuan Cheng, Longying Lai, Yue Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.16877 · **PDF:** https://arxiv.org/pdf/2603.16877v1
+
+## 11333. Position: Why a Dynamical Systems Perspective is Needed to Advance Time Series Modeling
+
+- **ID:** `2602.16864` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Daniel Durstewitz, Christoph Jürgen Hemmer, Florian Hess, et al.
+- **abs:** https://arxiv.org/abs/2602.16864 · **PDF:** https://arxiv.org/pdf/2602.16864v1
+
+## 11334. SimToolReal: An Object-Centric Policy for Zero-Shot Dexterous Tool Manipulation
+
+- **ID:** `2602.16863` · **primary:** `cs.RO` · **Submitted:** 2026-02-18
+- **Authors:** Kushal Kedia, Tyler Ga Wei Lum, Jeannette Bohg, et al.
+- **abs:** https://arxiv.org/abs/2602.16863 · **PDF:** https://arxiv.org/pdf/2602.16863v2
+
+## 11335. Meenz bleibt Meenz, but Large Language Models Do Not Speak Its Dialect
+
+- **ID:** `2602.16852` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Minh Duc Bui, Manuel Mager, Peter Herbert Kann, et al.
+- **abs:** https://arxiv.org/abs/2602.16852 · **PDF:** https://arxiv.org/pdf/2602.16852v3
+
+## 11336. On the Mechanism and Dynamics of Modular Addition: Fourier Features, Lottery Ticket, and Grokking
+
+- **ID:** `2602.16849` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Jianliang He, Leda Wang, Siyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.16849 · **PDF:** https://arxiv.org/pdf/2602.16849v1
+
+## 11337. Overseeing Agents Without Constant Oversight: Challenges and Opportunities
+
+- **ID:** `2602.16844` · **primary:** `cs.HC` · **Submitted:** 2026-02-18
+- **Authors:** Madeleine Grunde-McLaughlin, Hussein Mozannar, Maya Murad, et al.
+- **abs:** https://arxiv.org/abs/2602.16844 · **PDF:** https://arxiv.org/pdf/2602.16844v1
+
+## 11338. BanglaSummEval: Reference-Free Factual Consistency Evaluation for Bangla Summarization
+
+- **ID:** `2602.16843` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Ahmed Rafid, Rumman Adib, Fariya Ahmed, et al.
+- **abs:** https://arxiv.org/abs/2602.16843 · **PDF:** https://arxiv.org/pdf/2602.16843v1
+
+## 11339. What is the Value of Censored Data? An Exact Analysis for the Data-driven Newsvendor
+
+- **ID:** `2602.16842` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Rachitesh Kumar, Omar Mouchtaki
+- **abs:** https://arxiv.org/abs/2602.16842 · **PDF:** https://arxiv.org/pdf/2602.16842v1
+
+## 11340. Five Fatal Assumptions: Why T-Shirt Sizing Systematically Fails for AI Projects
+
+- **ID:** `2602.17734` · **primary:** `cs.SE` · **Submitted:** 2026-02-18
+- **Authors:** Raja Soundaramourty, Ozkan Kilic, Ramu Chenchaiah
+- **abs:** https://arxiv.org/abs/2602.17734 · **PDF:** https://arxiv.org/pdf/2602.17734v1
+
+## 11341. Training Large Reasoning Models Efficiently via Progressive Thought Encoding
+
+- **ID:** `2602.16839` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Zeliang Zhang, Xiaodong Liu, Hao Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.16839 · **PDF:** https://arxiv.org/pdf/2602.16839v1
+
+## 11342. A Residual-Aware Theory of Position Bias in Transformers
+
+- **ID:** `2602.16837` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Hanna Herasimchyk, Robin Labryga, Tomislav Prusina, et al.
+- **abs:** https://arxiv.org/abs/2602.16837 · **PDF:** https://arxiv.org/pdf/2602.16837v1
+
+## 11343. Claim Automation using Large Language Model
+
+- **ID:** `2602.16836` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Zhengda Mo, Zhiyu Quan, Eli O'Donohue, et al.
+- **abs:** https://arxiv.org/abs/2602.16836 · **PDF:** https://arxiv.org/pdf/2602.16836v1
+
+## 11344. NeST: Neuron Selective Tuning for LLM Safety
+
+- **ID:** `2602.16835` · **primary:** `cs.CR` · **Submitted:** 2026-02-18
+- **Authors:** Sasha Behrouzi, Lichao Wu, Mohamadreza Rostami, et al.
+- **abs:** https://arxiv.org/abs/2602.16835 · **PDF:** https://arxiv.org/pdf/2602.16835v1
+
+## 11345. VAM: Verbalized Action Masking for Controllable Exploration in RL Post-Training -- A Chess Case Study
+
+- **ID:** `2602.16833` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Zhicheng Zhang, Ziyan Wang, Yali Du, et al.
+- **abs:** https://arxiv.org/abs/2602.16833 · **PDF:** https://arxiv.org/pdf/2602.16833v1
+
+## 11346. IndicJR: A Judge-Free Benchmark of Jailbreak Robustness in South Asian Languages
+
+- **ID:** `2602.16832` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Priyaranjan Pattnayak, Sanchari Chowdhuri
+- **abs:** https://arxiv.org/abs/2602.16832 · **PDF:** https://arxiv.org/pdf/2602.16832v1
+
+## 11347. The Impact of Formations on Football Matches Using Double Machine Learning. Is it worth parking the bus?
+
+- **ID:** `2602.16830` · **primary:** `stat.AP` · **Submitted:** 2026-02-18
+- **Authors:** Genís Ruiz-Menárguez, Llorenç Badiella
+- **abs:** https://arxiv.org/abs/2602.16830 · **PDF:** https://arxiv.org/pdf/2602.16830v1
+
+## 11348. Learning under noisy supervision is governed by a feedback-truth gap
+
+- **ID:** `2602.16829` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Elan Schonfeld, Elias Wisnia
+- **abs:** https://arxiv.org/abs/2602.16829 · **PDF:** https://arxiv.org/pdf/2602.16829v1
+
+## 11349. An order-oriented approach to scoring hesitant fuzzy elements
+
+- **ID:** `2602.16827` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Luis Merino, Gabriel Navarro, Carlos Salvatierra, et al.
+- **abs:** https://arxiv.org/abs/2602.16827 · **PDF:** https://arxiv.org/pdf/2602.16827v1
+
+## 11350. HiVAE: Hierarchical Latent Variables for Scalable Theory of Mind
+
+- **ID:** `2602.16826` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Nigel Doering, Rahath Malladi, Arshia Sangwan, et al.
+- **abs:** https://arxiv.org/abs/2602.16826 · **PDF:** https://arxiv.org/pdf/2602.16826v1
+
+## 11351. Formal Mechanistic Interpretability: Automated Circuit Discovery with Provable Guarantees
+
+- **ID:** `2602.16823` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Itamar Hadad, Guy Katz, Shahaf Bassan
+- **abs:** https://arxiv.org/abs/2602.16823 · **PDF:** https://arxiv.org/pdf/2602.16823v1
+
+## 11352. TopoFlow: Physics-guided Neural Networks for high-resolution air quality prediction
+
+- **ID:** `2602.16821` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Ammar Kheder, Helmi Toropainen, Wenqing Peng, et al.
+- **abs:** https://arxiv.org/abs/2602.16821 · **PDF:** https://arxiv.org/pdf/2602.16821v1
+
+## 11353. AI-Mediated Feedback Improves Student Revisions: A Randomized Trial with FeedbackWriter in a Large Undergraduate Course
+
+- **ID:** `2602.16820` · **primary:** `cs.HC` · **Submitted:** 2026-02-18
+- **Authors:** Xinyi Lu, Kexin Phyllis Ju, Mitchell Dudley, et al.
+- **abs:** https://arxiv.org/abs/2602.16820 · **PDF:** https://arxiv.org/pdf/2602.16820v2
+
+## 11354. Hybrid-Gym: Training Coding Agents to Generalize Across Tasks
+
+- **ID:** `2602.16819` · **primary:** `cs.SE` · **Submitted:** 2026-02-18
+- **Authors:** Yiqing Xie, Emmy Liu, Gaokai Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.16819 · **PDF:** https://arxiv.org/pdf/2602.16819v1
+
+## 11355. Node Learning: A Framework for Adaptive, Decentralised and Collaborative Network Edge AI
+
+- **ID:** `2602.16814` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Eiman Kanjo, Mustafa Aslanov
+- **abs:** https://arxiv.org/abs/2602.16814 · **PDF:** https://arxiv.org/pdf/2602.16814v1
+
+## 11356. One-step Language Modeling via Continuous Denoising
+
+- **ID:** `2602.16813` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Chanhyuk Lee, Jaehoon Yoo, Manan Agarwal, et al.
+- **abs:** https://arxiv.org/abs/2602.16813 · **PDF:** https://arxiv.org/pdf/2602.16813v1
+
+## 11357. NeuDiff Agent: A Governed AI Workflow for Single-Crystal Neutron Crystallography
+
+- **ID:** `2602.16812` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Zhongcan Xiao, Leyi Zhang, Guannan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.16812 · **PDF:** https://arxiv.org/pdf/2602.16812v1
+
+## 11358. Evaluating Monolingual and Multilingual Large Language Models for Greek Question Answering: The DemosQA Benchmark
+
+- **ID:** `2602.16811` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Charalampos Mastrokostas, Nikolaos Giarelis, Nikos Karacapilidis
+- **abs:** https://arxiv.org/abs/2602.16811 · **PDF:** https://arxiv.org/pdf/2602.16811v1
+
+## 11359. Simple Baselines are Competitive with Code Evolution
+
+- **ID:** `2602.16805` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Yonatan Gideoni, Sebastian Risi, Yarin Gal
+- **abs:** https://arxiv.org/abs/2602.16805 · **PDF:** https://arxiv.org/pdf/2602.16805v1
+
+## 11360. References Improve LLM Alignment in Non-Verifiable Domains
+
+- **ID:** `2602.16802` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Kejian Shi, Yixin Liu, Peifeng Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.16802 · **PDF:** https://arxiv.org/pdf/2602.16802v1
+
+## 11361. Large-scale online deanonymization with LLMs
+
+- **ID:** `2602.16800` · **primary:** `cs.CR` · **Submitted:** 2026-02-18
+- **Authors:** Simon Lermen, Daniel Paleka, Joshua Swanson, et al.
+- **abs:** https://arxiv.org/abs/2602.16800 · **PDF:** https://arxiv.org/pdf/2602.16800v2
+
+## 11362. Efficient Tail-Aware Generative Optimization via Flow Model Fine-Tuning
+
+- **ID:** `2602.16796` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Zifan Wang, Riccardo De Santi, Xiaoyu Mo, et al.
+- **abs:** https://arxiv.org/abs/2602.16796 · **PDF:** https://arxiv.org/pdf/2602.16796v1
+
+## 11363. Beyond Procedure: Substantive Fairness in Conformal Prediction
+
+- **ID:** `2602.16794` · **primary:** `stat.ML` · **Submitted:** 2026-02-18
+- **Authors:** Pengqi Liu, Zijun Yu, Mouloud Belbahri, et al.
+- **abs:** https://arxiv.org/abs/2602.16794 · **PDF:** https://arxiv.org/pdf/2602.16794v1
+
+## 11364. Escaping the Cognitive Well: Efficient Competition Math with Off-the-Shelf Models
+
+- **ID:** `2602.16793` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Xingyu Dang, Rohit Agarwal, Rodrigo Porto, et al.
+- **abs:** https://arxiv.org/abs/2602.16793 · **PDF:** https://arxiv.org/pdf/2602.16793v1
+
+## 11365. Better Think Thrice: Learning to Reason Causally with Double Counterfactual Consistency
+
+- **ID:** `2602.16787` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Victoria Lin, Xinnuo Xu, Rachel Lawrence, et al.
+- **abs:** https://arxiv.org/abs/2602.16787 · **PDF:** https://arxiv.org/pdf/2602.16787v1
+
+## 11366. Omitted Variable Bias in Language Models Under Distribution Shift
+
+- **ID:** `2602.16784` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Victoria Lin, Louis-Philippe Morency, Eli Ben-Michael
+- **abs:** https://arxiv.org/abs/2602.16784 · **PDF:** https://arxiv.org/pdf/2602.16784v1
+
+## 11367. Knowledge-Embedded Latent Projection for Robust Representation Learning
+
+- **ID:** `2602.16709` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Weijing Tang, Ming Yuan, Zongqi Xia, et al.
+- **abs:** https://arxiv.org/abs/2602.16709 · **PDF:** https://arxiv.org/pdf/2602.16709v1
+
+## 11368. Policy Compiler for Secure Agentic Systems
+
+- **ID:** `2602.16708` · **primary:** `cs.CR` · **Submitted:** 2026-02-18
+- **Authors:** Nils Palumbo, Sarthak Choudhary, Jihye Choi, et al.
+- **abs:** https://arxiv.org/abs/2602.16708 · **PDF:** https://arxiv.org/pdf/2602.16708v2
+
+## 11369. Reinforced Fast Weights with Next-Sequence Prediction
+
+- **ID:** `2602.16704` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Hee Seung Hwang, Xindi Wu, Sanghyuk Chun, et al.
+- **abs:** https://arxiv.org/abs/2602.16704 · **PDF:** https://arxiv.org/pdf/2602.16704v1
+
+## 11370. Measuring Mid-2025 LLM-Assistance on Novice Performance in Biology
+
+- **ID:** `2602.16703` · **primary:** `cs.CY` · **Submitted:** 2026-02-18
+- **Authors:** Shen Zhou Hong, Alex Kleinman, Alyssa Mathiowetz, et al.
+- **abs:** https://arxiv.org/abs/2602.16703 · **PDF:** https://arxiv.org/pdf/2602.16703v1
+
+## 11371. Calibrate-Then-Act: Cost-Aware Exploration in LLM Agents
+
+- **ID:** `2602.16699` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Wenxuan Ding, Nicholas Tomlin, Greg Durrett
+- **abs:** https://arxiv.org/abs/2602.16699 · **PDF:** https://arxiv.org/pdf/2602.16699v2
+
+## 11372. Causality is Key for Interpretability Claims to Generalise
+
+- **ID:** `2602.16698` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Shruti Joshi, Aaron Mueller, David Klindt, et al.
+- **abs:** https://arxiv.org/abs/2602.16698 · **PDF:** https://arxiv.org/pdf/2602.16698v2
+
+## 11373. Protecting the Undeleted in Machine Unlearning
+
+- **ID:** `2602.16697` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Aloni Cohen, Refael Kohen, Kobbi Nissim, et al.
+- **abs:** https://arxiv.org/abs/2602.16697 · **PDF:** https://arxiv.org/pdf/2602.16697v1
+
+## 11374. Parameter-free representations outperform single-cell foundation models on downstream benchmarks
+
+- **ID:** `2602.16696` · **primary:** `q-bio.GN` · **Submitted:** 2026-02-18
+- **Authors:** Huan Souza, Pankaj Mehta
+- **abs:** https://arxiv.org/abs/2602.16696 · **PDF:** https://arxiv.org/pdf/2602.16696v1
+
+## 11375. Synthetic-Powered Multiple Testing with FDR Control
+
+- **ID:** `2602.16690` · **primary:** `stat.ME` · **Submitted:** 2026-02-18
+- **Authors:** Yonghoon Lee, Meshi Bashari, Edgar Dobriban, et al.
+- **abs:** https://arxiv.org/abs/2602.16690 · **PDF:** https://arxiv.org/pdf/2602.16690v1
+
+## 11376. Are Object-Centric Representations Better At Compositional Generalization?
+
+- **ID:** `2602.16689` · **primary:** `cs.CV` · **Submitted:** 2026-02-18
+- **Authors:** Ferdinand Kapl, Amir Mohammad Karimi Mamaghan, Maximilian Seitzer, et al.
+- **abs:** https://arxiv.org/abs/2602.16689 · **PDF:** https://arxiv.org/pdf/2602.16689v1
+
+## 11377. On the Hardness of Approximation of the Fair k-Center Problem
+
+- **ID:** `2602.16688` · **primary:** `cs.CC` · **Submitted:** 2026-02-18
+- **Authors:** Suhas Thejaswi
+- **abs:** https://arxiv.org/abs/2602.16688 · **PDF:** https://arxiv.org/pdf/2602.16688v2
+
+## 11378. Scaling Open Discrete Audio Foundation Models with Interleaved Semantic, Acoustic, and Text Tokens
+
+- **ID:** `2602.16687` · **primary:** `cs.SD` · **Submitted:** 2026-02-18
+- **Authors:** Potsawee Manakul, Woody Haosheng Gan, Martijn Bartelds, et al.
+- **abs:** https://arxiv.org/abs/2602.16687 · **PDF:** https://arxiv.org/pdf/2602.16687v1
+
+## 11379. Retrieval-Augmented Foundation Models for Matched Molecular Pair Transformations to Recapitulate Medicinal Chemistry Intuition
+
+- **ID:** `2602.16684` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Bo Pan, Peter Zhiping Zhang, Hao-Wei Pang, et al.
+- **abs:** https://arxiv.org/abs/2602.16684 · **PDF:** https://arxiv.org/pdf/2602.16684v1
+
+## 11380. Neighborhood Stability as a Measure of Nearest Neighbor Searchability
+
+- **ID:** `2602.16673` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Thomas Vecchiato, Sebastian Bruch
+- **abs:** https://arxiv.org/abs/2602.16673 · **PDF:** https://arxiv.org/pdf/2602.16673v1
+
+## 11381. SPARC: Scenario Planning and Reasoning for Automated C Unit Test Generation
+
+- **ID:** `2602.16671` · **primary:** `cs.SE` · **Submitted:** 2026-02-18
+- **Authors:** Jaid Monwar Chowdhury, Chi-An Fu, Reyhaneh Jabbarvand
+- **abs:** https://arxiv.org/abs/2602.16671 · **PDF:** https://arxiv.org/pdf/2602.16671v1
+
+## 11382. Towards a Science of AI Agent Reliability
+
+- **ID:** `2602.16666` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Stephan Rabanser, Sayash Kapoor, Peter Kirgis, et al.
+- **abs:** https://arxiv.org/abs/2602.16666 · **PDF:** https://arxiv.org/pdf/2602.16666v2
+
+## 11383. Cryo-SWAN: the Multi-Scale Wavelet-decomposition-inspired Autoencoder Network for molecular density representation of molecular volumes
+
+- **ID:** `2603.03342` · **primary:** `eess.IV` · **Submitted:** 2026-02-18
+- **Authors:** Rui Li, Artsemi Yushkevich, Mikhail Kudryashev, et al.
+- **abs:** https://arxiv.org/abs/2603.03342 · **PDF:** https://arxiv.org/pdf/2603.03342v1
+
+## 11384. Align Once, Benefit Multilingually: Enforcing Multilingual Consistency for LLM Safety Alignment
+
+- **ID:** `2602.16660` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Yuyan Bu, Xiaohao Liu, ZhaoXing Ren, et al.
+- **abs:** https://arxiv.org/abs/2602.16660 · **PDF:** https://arxiv.org/pdf/2602.16660v1
+
+## 11385. Investigating Nonlinear Quenching Effects on Polar Field Buildup in the Sun Using Physics-Informed Neural Networks
+
+- **ID:** `2602.16656` · **primary:** `astro-ph.SR` · **Submitted:** 2026-02-18
+- **Authors:** Jithu J. Athalathil, Mohammed H. Talafha, Bhargav Vaidya
+- **abs:** https://arxiv.org/abs/2602.16656 · **PDF:** https://arxiv.org/pdf/2602.16656v3
+
+## 11386. Agent Skill Framework: Perspectives on the Potential of Small Language Models in Industrial Environments
+
+- **ID:** `2602.16653` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Yangjie Xu, Lujun Li, Lama Sleem, et al.
+- **abs:** https://arxiv.org/abs/2602.16653 · **PDF:** https://arxiv.org/pdf/2602.16653v2
+
+## 11387. Retrieval Augmented Generation of Literature-derived Polymer Knowledge: The Example of a Biodegradable Polymer Expert System
+
+- **ID:** `2602.16650` · **primary:** `cs.CE` · **Submitted:** 2026-02-18
+- **Authors:** Sonakshi Gupta, Akhlak Mahmood, Wei Xiong, et al.
+- **abs:** https://arxiv.org/abs/2602.16650 · **PDF:** https://arxiv.org/pdf/2602.16650v1
+
+## 11388. Parallelizable Neural Turing Machines
+
+- **ID:** `2602.18508` · **primary:** `cs.NE` · **Submitted:** 2026-02-18
+- **Authors:** Gabriel Faria, Arnaldo Candido Junior
+- **abs:** https://arxiv.org/abs/2602.18508 · **PDF:** https://arxiv.org/pdf/2602.18508v1
+
+## 11389. RSS map-assisted MIMO channel estimation in the upper mid-band under pilot constraints
+
+- **ID:** `2603.00112` · **primary:** `eess.SP` · **Submitted:** 2026-02-18
+- **Authors:** Alireza Javid, Nuria González-Prelcic
+- **abs:** https://arxiv.org/abs/2603.00112 · **PDF:** https://arxiv.org/pdf/2603.00112v1
+
+## 11390. Factorization Machine with Quadratic-Optimization Annealing for RNA Inverse Folding and Evaluation of Binary-Integer Encoding and Nucleotide Assignment
+
+- **ID:** `2602.16643` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Shuta Kikuchi, Shu Tanaka
+- **abs:** https://arxiv.org/abs/2602.16643 · **PDF:** https://arxiv.org/pdf/2602.16643v1
+
+## 11391. Optimizer choice matters for the emergence of Neural Collapse
+
+- **ID:** `2602.16642` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Jim Zhao, Tin Sum Cheng, Wojciech Masarczyk, et al.
+- **abs:** https://arxiv.org/abs/2602.16642 · **PDF:** https://arxiv.org/pdf/2602.16642v3
+
+## 11392. Quecto-V1: Empirical Analysis of 8-bit Quantized Small Language Models for On-Device Legal Retrieval
+
+- **ID:** `2602.16640` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Subrit Dikshit
+- **abs:** https://arxiv.org/abs/2602.16640 · **PDF:** https://arxiv.org/pdf/2602.16640v1
+
+## 11393. AREG: Adversarial Resource Extraction Game for Evaluating Persuasion and Resistance in Large Language Models
+
+- **ID:** `2602.16639` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Adib Sakhawat, Fardeen Sadab
+- **abs:** https://arxiv.org/abs/2602.16639 · **PDF:** https://arxiv.org/pdf/2602.16639v1
+
+## 11394. Enhanced Diffusion Sampling: Efficient Rare Event Sampling and Free Energy Calculation with Diffusion Models
+
+- **ID:** `2602.16634` · **primary:** `stat.ML` · **Submitted:** 2026-02-18
+- **Authors:** Yu Xie, Ludwig Winkler, Lixin Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.16634 · **PDF:** https://arxiv.org/pdf/2602.16634v1
+
+## 11395. Almost Sure Convergence of Differential Temporal Difference Learning for Average Reward Markov Decision Processes
+
+- **ID:** `2602.16629` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Ethan Blaser, Jiuqi Wang, Shangtong Zhang
+- **abs:** https://arxiv.org/abs/2602.16629 · **PDF:** https://arxiv.org/pdf/2602.16629v1
+
+## 11396. Machine Learning Argument of Latitude Error Model for LEO Satellite Orbit and Covariance Correction
+
+- **ID:** `2602.16764` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Alex Moody, Penina Axelrad, Rebecca Russell
+- **abs:** https://arxiv.org/abs/2602.16764 · **PDF:** https://arxiv.org/pdf/2602.16764v1
+
+## 11397. A Systematic Evaluation of Sample-Level Tokenization Strategies for MEG Foundation Models
+
+- **ID:** `2602.16626` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** SungJun Cho, Chetan Gohil, Rukuang Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.16626 · **PDF:** https://arxiv.org/pdf/2602.16626v1
+
+## 11398. Causal and Compositional Abstraction
+
+- **ID:** `2602.16612` · **primary:** `cs.LO` · **Submitted:** 2026-02-18
+- **Authors:** Robin Lorenz, Sean Tull
+- **abs:** https://arxiv.org/abs/2602.16612 · **PDF:** https://arxiv.org/pdf/2602.16612v1
+
+## 11399. Who can we trust? LLM-as-a-jury for Comparative Assessment
+
+- **ID:** `2602.16610` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Mengjie Qian, Guangzhi Sun, Mark J. F. Gales, et al.
+- **abs:** https://arxiv.org/abs/2602.16610 · **PDF:** https://arxiv.org/pdf/2602.16610v1
+
+## 11400. ColBERT-Zero: To Pre-train Or Not To Pre-train ColBERT models
+
+- **ID:** `2602.16609` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Antoine Chaffin, Luca Arnaboldi, Amélie Chatelain, et al.
+- **abs:** https://arxiv.org/abs/2602.16609 · **PDF:** https://arxiv.org/pdf/2602.16609v1
+
+## 11401. Explainable AI: Context-Aware Layer-Wise Integrated Gradients for Explaining Transformer Models
+
+- **ID:** `2602.16608` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Melkamu Abay Mersha, Jugal Kalita
+- **abs:** https://arxiv.org/abs/2602.16608 · **PDF:** https://arxiv.org/pdf/2602.16608v1
+
+## 11402. CitiLink-Summ: Summarization of Discussion Subjects in European Portuguese Municipal Meeting Minutes
+
+- **ID:** `2602.16607` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Miguel Marques, Ana Luísa Fernandes, Ana Filipa Pacheco, et al.
+- **abs:** https://arxiv.org/abs/2602.16607 · **PDF:** https://arxiv.org/pdf/2602.16607v1
+
+## 11403. FlowPrefill: Decoupling Preemption from Prefill Scheduling Granularity to Mitigate Head-of-Line Blocking in LLM Serving
+
+- **ID:** `2602.16603` · **primary:** `cs.DC` · **Submitted:** 2026-02-18
+- **Authors:** Chia-chi Hsieh, Zan Zong, Xinyang Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.16603 · **PDF:** https://arxiv.org/pdf/2602.16603v1
+
+## 11404. Error Propagation and Model Collapse in Diffusion Models: A Theoretical Study
+
+- **ID:** `2602.16601` · **primary:** `stat.ML` · **Submitted:** 2026-02-18
+- **Authors:** Nail B. Khelifa, Richard E. Turner, Ramji Venkataramanan
+- **abs:** https://arxiv.org/abs/2602.16601 · **PDF:** https://arxiv.org/pdf/2602.16601v1
+
+## 11405. Predicting The Cop Number Using Machine Learning
+
+- **ID:** `2602.16600` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Meagan Mann, Christian Muise, Erin Meger
+- **abs:** https://arxiv.org/abs/2602.16600 · **PDF:** https://arxiv.org/pdf/2602.16600v1
+
+## 11406. When AI Benchmarks Plateau: A Systematic Study of Benchmark Saturation
+
+- **ID:** `2602.16763` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Mubashara Akhtar, Anka Reuel, Prajna Soni, et al.
+- **abs:** https://arxiv.org/abs/2602.16763 · **PDF:** https://arxiv.org/pdf/2602.16763v1
+
+## 11407. Sequential Membership Inference Attacks
+
+- **ID:** `2602.16596` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Thomas Michel, Debabrota Basu, Emilie Kaufmann
+- **abs:** https://arxiv.org/abs/2602.16596 · **PDF:** https://arxiv.org/pdf/2602.16596v1
+
+## 11408. A Contrastive Learning Framework Empowered by Attention-based Feature Adaptation for Street-View Image Classification
+
+- **ID:** `2602.16590` · **primary:** `cs.CV` · **Submitted:** 2026-02-18
+- **Authors:** Qi You, Yitai Cheng, Zichao Zeng, et al.
+- **abs:** https://arxiv.org/abs/2602.16590 · **PDF:** https://arxiv.org/pdf/2602.16590v1
+
+## 11409. DataJoint 2.0: A Computational Substrate for Agentic Scientific Workflows
+
+- **ID:** `2602.16585` · **primary:** `cs.DB` · **Submitted:** 2026-02-18
+- **Authors:** Dimitri Yatsenko, Thinh T. Nguyen
+- **abs:** https://arxiv.org/abs/2602.16585 · **PDF:** https://arxiv.org/pdf/2602.16585v1
+
+## 11410. Aligning Language Models from User Interactions
+
+- **ID:** `2603.12273` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Thomas Kleine Buening, Jonas Hübotter, Barna Pásztor, et al.
+- **abs:** https://arxiv.org/abs/2603.12273 · **PDF:** https://arxiv.org/pdf/2603.12273v1
+
+## 11411. AIFL: A Global Daily Streamflow Forecasting Model Using Deterministic LSTM Pre-trained on ERA5-Land and Fine-tuned on IFS
+
+- **ID:** `2602.16579` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Maria Luisa Taccari, Kenza Tazi, Oisín M. Morrison, et al.
+- **abs:** https://arxiv.org/abs/2602.16579 · **PDF:** https://arxiv.org/pdf/2602.16579v1
+
+## 11412. Creating a digital poet
+
+- **ID:** `2602.16578` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Vered Tohar, Tsahi Hayat, Amir Leshem
+- **abs:** https://arxiv.org/abs/2602.16578 · **PDF:** https://arxiv.org/pdf/2602.16578v1
+
+## 11413. MoDE-Boost: Boosting Shared Mobility Demand with Edge-Ready Prediction Models
+
+- **ID:** `2602.16573` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Antonios Tziorvas, George S. Theodoropoulos, Yannis Theodoridis
+- **abs:** https://arxiv.org/abs/2602.16573 · **PDF:** https://arxiv.org/pdf/2602.16573v1
+
+## 11414. Attending to Routers Aids Indoor Wireless Localization
+
+- **ID:** `2602.16762` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Ayush Roy, Tahsin Fuad Hassan, Roshan Ayyalasomayajula, et al.
+- **abs:** https://arxiv.org/abs/2602.16762 · **PDF:** https://arxiv.org/pdf/2602.16762v1
+
+## 11415. Utility-Preserving De-Identification for Math Tutoring: Investigating Numeric Ambiguity in the MathEd-PII Benchmark Dataset
+
+- **ID:** `2602.16571` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Zhuqian Zhou, Kirk Vanacore, Bakhtawar Ahtisham, et al.
+- **abs:** https://arxiv.org/abs/2602.16571 · **PDF:** https://arxiv.org/pdf/2602.16571v1
+
+## 11416. Steering diffusion models with quadratic rewards: a fine-grained analysis
+
+- **ID:** `2602.16570` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Ankur Moitra, Andrej Risteski, Dhruv Rohatgi
+- **abs:** https://arxiv.org/abs/2602.16570 · **PDF:** https://arxiv.org/pdf/2602.16570v1
+
+## 11417. Separating Oblivious and Adaptive Models of Variable Selection
+
+- **ID:** `2602.16568` · **primary:** `math.ST` · **Submitted:** 2026-02-18
+- **Authors:** Ziyun Chen, Jerry Li, Kevin Tian, et al.
+- **abs:** https://arxiv.org/abs/2602.16568 · **PDF:** https://arxiv.org/pdf/2602.16568v1
+
+## 11418. A Scalable Approach to Solving Simulation-Based Network Security Games
+
+- **ID:** `2602.16564` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Michael Lanier, Yevgeniy Vorobeychik
+- **abs:** https://arxiv.org/abs/2602.16564 · **PDF:** https://arxiv.org/pdf/2602.16564v2
+
+## 11419. Illustration of Barren Plateaus in Quantum Computing
+
+- **ID:** `2602.16558` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Gerhard Stenzel, Tobias Rohe, Michael Kölle, et al.
+- **abs:** https://arxiv.org/abs/2602.16558 · **PDF:** https://arxiv.org/pdf/2602.16558v1
+
+## 11420. Learning Distributed Equilibria in Linear-Quadratic Stochastic Differential Games: An $α$-Potential Approach
+
+- **ID:** `2602.16555` · **primary:** `math.OC` · **Submitted:** 2026-02-18
+- **Authors:** Philipp Plank, Yufei Zhang
+- **abs:** https://arxiv.org/abs/2602.16555 · **PDF:** https://arxiv.org/pdf/2602.16555v1
+
+## 11421. MerLean: An Agentic Framework for Autoformalization in Quantum Computation
+
+- **ID:** `2602.16554` · **primary:** `cs.LO` · **Submitted:** 2026-02-18
+- **Authors:** Yuanjie Ren, Jinzheng Li, Yidi Qi
+- **abs:** https://arxiv.org/abs/2602.16554 · **PDF:** https://arxiv.org/pdf/2602.16554v1
+
+## 11422. RIDER: 3D RNA Inverse Design with Reinforcement Learning-Guided Diffusion
+
+- **ID:** `2602.16548` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Tianmeng Hu, Yongzheng Cui, Biao Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.16548 · **PDF:** https://arxiv.org/pdf/2602.16548v1
+
+## 11423. Let's Split Up: Zero-Shot Classifier Edits for Fine-Grained Video Understanding
+
+- **ID:** `2602.16545` · **primary:** `cs.CV` · **Submitted:** 2026-02-18
+- **Authors:** Kaiting Liu, Hazel Doughty
+- **abs:** https://arxiv.org/abs/2602.16545 · **PDF:** https://arxiv.org/pdf/2602.16545v1
+
+## 11424. Vulnerability Analysis of Safe Reinforcement Learning via Inverse Constrained Reinforcement Learning
+
+- **ID:** `2602.16543` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Jialiang Fan, Shixiong Jiang, Mengyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.16543 · **PDF:** https://arxiv.org/pdf/2602.16543v1
+
+## 11425. Optimal training-conditional regret for online conformal prediction
+
+- **ID:** `2602.16537` · **primary:** `math.ST` · **Submitted:** 2026-02-18
+- **Authors:** Jiadong Liang, Zhimei Ren, Yuxin Chen
+- **abs:** https://arxiv.org/abs/2602.16537 · **PDF:** https://arxiv.org/pdf/2602.16537v2
+
+## 11426. Transfer Learning of Linear Regression with Multiple Pretrained Models: Benefiting from More Pretrained Models via Overparameterization Debiasing
+
+- **ID:** `2602.16531` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Daniel Boharon, Yehuda Dar
+- **abs:** https://arxiv.org/abs/2602.16531 · **PDF:** https://arxiv.org/pdf/2602.16531v1
+
+## 11427. FEKAN: Feature-Enriched Kolmogorov-Arnold Networks
+
+- **ID:** `2602.16530` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Sidharth S. Menon, Ameya D. Jagtap
+- **abs:** https://arxiv.org/abs/2602.16530 · **PDF:** https://arxiv.org/pdf/2602.16530v1
+
+## 11428. Capacity-constrained demand response in smart grids using deep reinforcement learning
+
+- **ID:** `2602.16525` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Shafagh Abband Pashaki, Sepehr Maleki, Amir Badiee
+- **abs:** https://arxiv.org/abs/2602.16525 · **PDF:** https://arxiv.org/pdf/2602.16525v1
+
+## 11429. Reinforcement Learning for Parameterized Quantum State Preparation: A Comparative Study
+
+- **ID:** `2602.16523` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Gerhard Stenzel, Isabella Debelic, Michael Kölle, et al.
+- **abs:** https://arxiv.org/abs/2602.16523 · **PDF:** https://arxiv.org/pdf/2602.16523v1
+
+## 11430. Recursive language models for jailbreak detection: a procedural defense for tool-augmented agents
+
+- **ID:** `2602.16520` · **primary:** `cs.CR` · **Submitted:** 2026-02-18
+- **Authors:** Doron Shavit
+- **abs:** https://arxiv.org/abs/2602.16520 · **PDF:** https://arxiv.org/pdf/2602.16520v1
+
+## 11431. Supercharging Agenda Setting Research: The ParlaCAP Dataset of 28 European Parliaments and a Scalable Multilingual LLM-Based Classification
+
+- **ID:** `2602.16516` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Taja Kuzman Pungeršek, Peter Rupnik, Daniela Širinić, et al.
+- **abs:** https://arxiv.org/abs/2602.16516 · **PDF:** https://arxiv.org/pdf/2602.16516v1
+
+## 11432. Framework of Thoughts: A Foundation Framework for Dynamic and Optimized Reasoning based on Chains, Trees, and Graphs
+
+- **ID:** `2602.16512` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Felix Fricke, Simon Malberg, Georg Groh
+- **abs:** https://arxiv.org/abs/2602.16512 · **PDF:** https://arxiv.org/pdf/2602.16512v1
+
+## 11433. AgentFixer: From Failure Detection to Fix Recommendations in LLM Agentic Systems
+
+- **ID:** `2603.29848` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Hadar Mulian, Sergey Zeltyn, Ido Levy, et al.
+- **abs:** https://arxiv.org/abs/2603.29848 · **PDF:** https://arxiv.org/pdf/2603.29848v1
+
+## 11434. Small molecule retrieval from tandem mass spectrometry: what are we optimizing for?
+
+- **ID:** `2602.16507` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Gaetan De Waele, Marek Wydmuch, Krzysztof Dembczyński, et al.
+- **abs:** https://arxiv.org/abs/2602.16507 · **PDF:** https://arxiv.org/pdf/2602.16507v1
+
+## 11435. Functional Decomposition and Shapley Interactions for Interpreting Survival Models
+
+- **ID:** `2602.16505` · **primary:** `stat.ML` · **Submitted:** 2026-02-18
+- **Authors:** Sophie Hanna Langbein, Hubert Baniecki, Fabian Fumagalli, et al.
+- **abs:** https://arxiv.org/abs/2602.16505 · **PDF:** https://arxiv.org/pdf/2602.16505v1
+
+## 11436. Multi-Agent DRL for V2X Resource Allocation: Disentangling Challenges and Benchmarking Solutions
+
+- **ID:** `2603.06607` · **primary:** `cs.MA` · **Submitted:** 2026-02-18
+- **Authors:** Siyuan Wang, Lei Lei, Pranav Maheshwari, et al.
+- **abs:** https://arxiv.org/abs/2603.06607 · **PDF:** https://arxiv.org/pdf/2603.06607v1
+
+## 11437. ActTail: Global Activation Sparsity in Large Language Models
+
+- **ID:** `2603.12272` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Wenwen Hou, Xinyuan Song, Shiwei Liu
+- **abs:** https://arxiv.org/abs/2603.12272 · **PDF:** https://arxiv.org/pdf/2603.12272v1
+
+## 11438. Interpretability-by-Design with Accurate Locally Additive Models and Conditional Feature Effects
+
+- **ID:** `2602.16503` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Vasilis Gkolemis, Loukas Kavouras, Dimitrios Kyriakopoulos, et al.
+- **abs:** https://arxiv.org/abs/2602.16503 · **PDF:** https://arxiv.org/pdf/2602.16503v1
+
+## 11439. Optimizing Soft Prompt Tuning via Structural Evolution
+
+- **ID:** `2602.16500` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Zhenzhen Huang, Chaoning Zhang, Haoyu Bian, et al.
+- **abs:** https://arxiv.org/abs/2602.16500 · **PDF:** https://arxiv.org/pdf/2602.16500v1
+
+## 11440. CASCADE: Cross-scale Advective Super-resolution with Climate Assimilation and Downscaling Evolution
+
+- **ID:** `2603.00109` · **primary:** `physics.ao-ph` · **Submitted:** 2026-02-18
+- **Authors:** Alexander Kovalenko
+- **abs:** https://arxiv.org/abs/2603.00109 · **PDF:** https://arxiv.org/pdf/2603.00109v1
+
+## 11441. Fast and Scalable Analytical Diffusion
+
+- **ID:** `2602.16498` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Xinyi Shang, Peng Sun, Jingyu Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.16498 · **PDF:** https://arxiv.org/pdf/2602.16498v1
+
+## 11442. From Growing to Looping: A Unified View of Iterative Computation in LLMs
+
+- **ID:** `2602.16490` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Ferdinand Kapl, Emmanouil Angelis, Kaitlin Maile, et al.
+- **abs:** https://arxiv.org/abs/2602.16490 · **PDF:** https://arxiv.org/pdf/2602.16490v1
+
+## 11443. Learning to Learn from Language Feedback with Social Meta-Learning
+
+- **ID:** `2602.16488` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Jonathan Cook, Diego Antognini, Martin Klissarov, et al.
+- **abs:** https://arxiv.org/abs/2602.16488 · **PDF:** https://arxiv.org/pdf/2602.16488v1
+
+## 11444. Team of Thoughts: Efficient Test-time Scaling of Agentic Systems through Orchestrated Tool Calling
+
+- **ID:** `2602.16485` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Jeffrey T. H. Wong, Zixi Zhang, Junyi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.16485 · **PDF:** https://arxiv.org/pdf/2602.16485v2
+
+## 11445. Leveraging Large Language Models for Causal Discovery: a Constraint-based, Argumentation-driven Approach
+
+- **ID:** `2602.16481` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Zihao Li, Fabrizio Russo
+- **abs:** https://arxiv.org/abs/2602.16481 · **PDF:** https://arxiv.org/pdf/2602.16481v1
+
+## 11446. Learning Preference from Observed Rankings
+
+- **ID:** `2602.16476` · **primary:** `stat.ML` · **Submitted:** 2026-02-18
+- **Authors:** Yu-Chang Chen, Chen Chian Fuh, Shang En Tsai
+- **abs:** https://arxiv.org/abs/2602.16476 · **PDF:** https://arxiv.org/pdf/2602.16476v1
+
+## 11447. Synthesis and Verification of Transformer Programs
+
+- **ID:** `2602.16473` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Hongjian Jiang, Matthew Hague, Philipp Rümmer, et al.
+- **abs:** https://arxiv.org/abs/2602.16473 · **PDF:** https://arxiv.org/pdf/2602.16473v1
+
+## 11448. Training Models on Dialects of Translationese Shows How Lexical Diversity and Source-Target Syntactic Similarity Shape Learning
+
+- **ID:** `2602.16469` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Jenny Kunz
+- **abs:** https://arxiv.org/abs/2602.16469 · **PDF:** https://arxiv.org/pdf/2602.16469v1
+
+## 11449. HPMixer: Hierarchical Patching for Multivariate Time Series Forecasting
+
+- **ID:** `2602.16468` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Jung Min Choi, Vijaya Krishna Yalavarthi, Lars Schmidt-Thieme
+- **abs:** https://arxiv.org/abs/2602.16468 · **PDF:** https://arxiv.org/pdf/2602.16468v2
+
+## 11450. IndicEval: A Bilingual Indian Educational Evaluation Framework for Large Language Models
+
+- **ID:** `2602.16467` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Saurabh Bharti, Gaurav Azad, Abhinaw Jagtap, et al.
+- **abs:** https://arxiv.org/abs/2602.16467 · **PDF:** https://arxiv.org/pdf/2602.16467v1
+
+## 11451. Beyond SGD, Without SVD: Proximal Subspace Iteration LoRA with Diagonal Fractional K-FAC
+
+- **ID:** `2602.16456` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Abdulla Jasem Almansoori, Maria Ivanova, Andrey Veprikov, et al.
+- **abs:** https://arxiv.org/abs/2602.16456 · **PDF:** https://arxiv.org/pdf/2602.16456v1
+
+## 11452. GICDM: Mitigating Hubness for Reliable Distance-Based Generative Model Evaluation
+
+- **ID:** `2602.16449` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Nicolas Salvy, Hugues Talbot, Bertrand Thirion
+- **abs:** https://arxiv.org/abs/2602.16449 · **PDF:** https://arxiv.org/pdf/2602.16449v1
+
+## 11453. RoboGene: Boosting VLA Pre-training via Diversity-Driven Agentic Framework for Real-World Task Generation
+
+- **ID:** `2602.16444` · **primary:** `cs.RO` · **Submitted:** 2026-02-18
+- **Authors:** Yixue Zhang, Kun Wu, Zhi Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.16444 · **PDF:** https://arxiv.org/pdf/2602.16444v2
+
+## 11454. Hardware-accelerated graph neural networks: an alternative approach for neuromorphic event-based audio classification and keyword spotting on SoC FPGA
+
+- **ID:** `2602.16442` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Kamil Jeziorek, Piotr Wzorek, Krzysztof Blachut, et al.
+- **abs:** https://arxiv.org/abs/2602.16442 · **PDF:** https://arxiv.org/pdf/2602.16442v1
+
+## 11455. Fine-Pruning: A Biologically Inspired Algorithm for Personalization of Machine Learning Models
+
+- **ID:** `2602.18507` · **primary:** `cs.NE` · **Submitted:** 2026-02-18
+- **Authors:** Joseph Bingham, Saman Zonouz, Dvir Aran
+- **abs:** https://arxiv.org/abs/2602.18507 · **PDF:** https://arxiv.org/pdf/2602.18507v1
+
+## 11456. SurgFusion-Net: Diversified Adaptive Multimodal Fusion Network for Surgical Skill Assessment
+
+- **ID:** `2603.00108` · **primary:** `cs.RO` · **Submitted:** 2026-02-18
+- **Authors:** Runlong He, Freweini M. Tesfai, Matthew W. E. Boal, et al.
+- **abs:** https://arxiv.org/abs/2603.00108 · **PDF:** https://arxiv.org/pdf/2603.00108v1
+
+## 11457. Intra-Fairness Dynamics: The Bias Spillover Effect in Targeted LLM Alignment
+
+- **ID:** `2602.16438` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Eva Paraschou, Line Harder Clemmensen, Sneha Das
+- **abs:** https://arxiv.org/abs/2602.16438 · **PDF:** https://arxiv.org/pdf/2602.16438v1
+
+## 11458. Learning with Locally Private Examples by Inverse Weierstrass Private Stochastic Gradient Descent
+
+- **ID:** `2602.16436` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Jean Dufraiche, Paul Mangold, Michaël Perrot, et al.
+- **abs:** https://arxiv.org/abs/2602.16436 · **PDF:** https://arxiv.org/pdf/2602.16436v1
+
+## 11459. Causally-Guided Automated Feature Engineering with Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.16435` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Arun Vignesh Malarkkan, Wangyang Ying, Yanjie Fu
+- **abs:** https://arxiv.org/abs/2602.16435 · **PDF:** https://arxiv.org/pdf/2602.16435v1
+
+## 11460. Designing Production-Scale OCR for India: Multilingual and Domain-Specific Systems
+
+- **ID:** `2602.16430` · **primary:** `cs.CV` · **Submitted:** 2026-02-18
+- **Authors:** Ali Faraz, Raja Kolla, Ashish Kulkarni, et al.
+- **abs:** https://arxiv.org/abs/2602.16430 · **PDF:** https://arxiv.org/pdf/2602.16430v1
+
+## 11461. TabAgent: A Framework for Replacing Agentic Generative Components with Tabular-Textual Classifiers
+
+- **ID:** `2602.16429` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Ido Levy, Eilam Shapira, Yinon Goldshtein, et al.
+- **abs:** https://arxiv.org/abs/2602.16429 · **PDF:** https://arxiv.org/pdf/2602.16429v1
+
+## 11462. Verifiable Semantics for Agent-to-Agent Communication
+
+- **ID:** `2602.16424` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Philipp Schoenegger, Matt Carlson, Chris Schneider, et al.
+- **abs:** https://arxiv.org/abs/2602.16424 · **PDF:** https://arxiv.org/pdf/2602.16424v2
+
+## 11463. Automated Histopathology Report Generation via Pyramidal Feature Extraction and the UNI Foundation Model
+
+- **ID:** `2602.16422` · **primary:** `eess.IV` · **Submitted:** 2026-02-18
+- **Authors:** Ahmet Halici, Ece Tugba Cebeci, Musa Balci, et al.
+- **abs:** https://arxiv.org/abs/2602.16422 · **PDF:** https://arxiv.org/pdf/2602.16422v1
+
+## 11464. Easy Data Unlearning Bench
+
+- **ID:** `2602.16400` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Roy Rinberg, Pol Puigdemont, Martin Pawelczyk, et al.
+- **abs:** https://arxiv.org/abs/2602.16400 · **PDF:** https://arxiv.org/pdf/2602.16400v1
+
+## 11465. Multi-Channel Replay Speech Detection using Acoustic Maps
+
+- **ID:** `2602.16399` · **primary:** `eess.AS` · **Submitted:** 2026-02-18
+- **Authors:** Michael Neri, Tuomas Virtanen
+- **abs:** https://arxiv.org/abs/2602.16399 · **PDF:** https://arxiv.org/pdf/2602.16399v1
+
+## 11466. LegoNet: Memory Footprint Reduction Through Block Weight Clustering
+
+- **ID:** `2603.06606` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Joseph Bingham, Noah Green, Saman Zonouz
+- **abs:** https://arxiv.org/abs/2603.06606 · **PDF:** https://arxiv.org/pdf/2603.06606v1
+
+## 11467. Structure-Aware Set Transformers: Temporal and Variable-Type Attention Biases for Asynchronous Clinical Time Series
+
+- **ID:** `2603.06605` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Joohyung Lee, Kwanhyung Lee, Changhun Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.06605 · **PDF:** https://arxiv.org/pdf/2603.06605v2
+
+## 11468. Label-Consistent Data Generation for Aspect-Based Sentiment Analysis Using LLM Agents
+
+- **ID:** `2602.16379` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Mohammad H. A. Monfared, Lucie Flek, Akbar Karimi
+- **abs:** https://arxiv.org/abs/2602.16379 · **PDF:** https://arxiv.org/pdf/2602.16379v1
+
+## 11469. Variable-Length Semantic IDs for Recommender Systems
+
+- **ID:** `2602.16375` · **primary:** `cs.IR` · **Submitted:** 2026-02-18
+- **Authors:** Kirill Khrylchenko
+- **abs:** https://arxiv.org/abs/2602.16375 · **PDF:** https://arxiv.org/pdf/2602.16375v1
+
+## 11470. AI-Driven Structure Refinement of X-ray Diffraction
+
+- **ID:** `2602.16372` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-18
+- **Authors:** Bin Cao, Qian Zhang, Zhenjie Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.16372 · **PDF:** https://arxiv.org/pdf/2602.16372v2
+
+## 11471. Diagnosing Retrieval Bias Under Multiple In-Context Knowledge Updates in Large Language Models
+
+- **ID:** `2603.12271` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Boyu Qiao, Sean Guo, Xian Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.12271 · **PDF:** https://arxiv.org/pdf/2603.12271v1
+
+## 11472. Clever Materials: When Models Identify Good Materials for the Wrong Reasons
+
+- **ID:** `2602.17730` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-18
+- **Authors:** Kevin Maik Jablonka
+- **abs:** https://arxiv.org/abs/2602.17730 · **PDF:** https://arxiv.org/pdf/2602.17730v1
+
+## 11473. Improved Bounds for Reward-Agnostic and Reward-Free Exploration
+
+- **ID:** `2602.16363` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Oran Ridel, Alon Cohen
+- **abs:** https://arxiv.org/abs/2602.16363 · **PDF:** https://arxiv.org/pdf/2602.16363v1
+
+## 11474. Task-Specific Knowledge Distillation via Intermediate Probes
+
+- **ID:** `2603.12270` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Ryan Brown, Chris Russell
+- **abs:** https://arxiv.org/abs/2603.12270 · **PDF:** https://arxiv.org/pdf/2603.12270v1
+
+## 11475. Optical Inversion and Spectral Unmixing of Spectroscopic Photoacoustic Images with Physics-Informed Neural Networks
+
+- **ID:** `2602.16357` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Sarkis Ter Martirosyan, Xinyue Huang, David Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.16357 · **PDF:** https://arxiv.org/pdf/2602.16357v1
+
+## 11476. Articulated 3D Scene Graphs for Open-World Mobile Manipulation
+
+- **ID:** `2602.16356` · **primary:** `cs.RO` · **Submitted:** 2026-02-18
+- **Authors:** Martin Büchner, Adrian Röfer, Tim Engelbracht, et al.
+- **abs:** https://arxiv.org/abs/2602.16356 · **PDF:** https://arxiv.org/pdf/2602.16356v1
+
+## 11477. Machine Learning in Epidemiology
+
+- **ID:** `2602.16352` · **primary:** `stat.ML` · **Submitted:** 2026-02-18
+- **Authors:** Marvin N. Wright, Lukas Burk, Pegah Golchian, et al.
+- **abs:** https://arxiv.org/abs/2602.16352 · **PDF:** https://arxiv.org/pdf/2602.16352v1
+
+## 11478. Helpful to a Fault: Measuring Illicit Assistance in Multi-Turn, Multilingual LLM Agents
+
+- **ID:** `2602.16346` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Nivya Talokar, Ayush K Tarun, Murari Mandal, et al.
+- **abs:** https://arxiv.org/abs/2602.16346 · **PDF:** https://arxiv.org/pdf/2602.16346v2
+
+## 11479. How to Label Resynthesized Audio: The Dual Role of Neural Audio Codecs in Audio Deepfake Detection
+
+- **ID:** `2602.16343` · **primary:** `cs.SD` · **Submitted:** 2026-02-18
+- **Authors:** Yixuan Xiao, Florian Lux, Alejandro Pérez-González-de-Martos, et al.
+- **abs:** https://arxiv.org/abs/2602.16343 · **PDF:** https://arxiv.org/pdf/2602.16343v1
+
+## 11480. Explainability for Fault Detection System in Chemical Processes
+
+- **ID:** `2602.16341` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Georgios Gravanis, Dimitrios Kyriakou, Spyros Voutetakis, et al.
+- **abs:** https://arxiv.org/abs/2602.16341 · **PDF:** https://arxiv.org/pdf/2602.16341v1
+
+## 11481. The Implicit Bias of Adam and Muon on Smooth Homogeneous Neural Networks
+
+- **ID:** `2602.16340` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Eitan Gronich, Gal Vardi
+- **abs:** https://arxiv.org/abs/2602.16340 · **PDF:** https://arxiv.org/pdf/2602.16340v2
+
+## 11482. Subtractive Modulative Network with Learnable Periodic Activations
+
+- **ID:** `2602.16337` · **primary:** `cs.CV` · **Submitted:** 2026-02-18
+- **Authors:** Tiou Wang, Zhuoqian Yang, Markus Flierl, et al.
+- **abs:** https://arxiv.org/abs/2602.16337 · **PDF:** https://arxiv.org/pdf/2602.16337v1
+
+## 11483. HAWX: A Hardware-Aware FrameWork for Fast and Scalable ApproXimation of DNNs
+
+- **ID:** `2602.16336` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Samira Nazari, Mohammad Saeed Almasi, Mahdi Taheri, et al.
+- **abs:** https://arxiv.org/abs/2602.16336 · **PDF:** https://arxiv.org/pdf/2602.16336v2
+
+## 11484. Spatial Audio Question Answering and Reasoning on Dynamic Source Movements
+
+- **ID:** `2602.16334` · **primary:** `cs.SD` · **Submitted:** 2026-02-18
+- **Authors:** Arvind Krishna Sridhar, Yinyi Guo, Erik Visser
+- **abs:** https://arxiv.org/abs/2602.16334 · **PDF:** https://arxiv.org/pdf/2602.16334v1
+
+## 11485. Stop Saying "AI"
+
+- **ID:** `2602.17729` · **primary:** `cs.CY` · **Submitted:** 2026-02-18
+- **Authors:** Nathan G. Wood, Scott Robbins, Eduardo Zegarra Berodt, et al.
+- **abs:** https://arxiv.org/abs/2602.17729 · **PDF:** https://arxiv.org/pdf/2602.17729v2
+
+## 11486. Guide-Guard: Off-Target Predicting in CRISPR Applications
+
+- **ID:** `2602.16327` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Joseph Bingham, Netanel Arussy, Saman Zonouz
+- **abs:** https://arxiv.org/abs/2602.16327 · **PDF:** https://arxiv.org/pdf/2602.16327v1
+
+## 11487. A Self-Supervised Approach for Enhanced Feature Representations in Object Detection Tasks
+
+- **ID:** `2602.16322` · **primary:** `cs.CV` · **Submitted:** 2026-02-18
+- **Authors:** Santiago C. Vilabella, Pablo Pérez-Núñez, Beatriz Remeseiro
+- **abs:** https://arxiv.org/abs/2602.16322 · **PDF:** https://arxiv.org/pdf/2602.16322v1
+
+## 11488. End-user validation of BRIGHT with custom-developed graphical user interface applied to cervical cancer brachytherapy
+
+- **ID:** `2602.16321` · **primary:** `cs.NE` · **Submitted:** 2026-02-18
+- **Authors:** Leah R. M. Dickhoff, Ellen M. Kerkhof, Heloisa H. Deuzeman, et al.
+- **abs:** https://arxiv.org/abs/2602.16321 · **PDF:** https://arxiv.org/pdf/2602.16321v1
+
+## 11489. RefineFormer3D: Efficient 3D Medical Image Segmentation via Adaptive Multi-Scale Transformer with Cross Attention Fusion
+
+- **ID:** `2602.16320` · **primary:** `eess.IV` · **Submitted:** 2026-02-18
+- **Authors:** Kavyansh Tyagi, Vishwas Rathi, Puneet Goyal
+- **abs:** https://arxiv.org/abs/2602.16320 · **PDF:** https://arxiv.org/pdf/2602.16320v1
+
+## 11490. A Graph Meta-Network for Learning on Kolmogorov-Arnold Networks
+
+- **ID:** `2602.16316` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Guy Bar-Shalom, Ami Tavory, Itay Evron, et al.
+- **abs:** https://arxiv.org/abs/2602.16316 · **PDF:** https://arxiv.org/pdf/2602.16316v2
+
+## 11491. The Diversity Paradox revisited: Systemic Effects of Feedback Loops in Recommender Systems
+
+- **ID:** `2602.16315` · **primary:** `cs.IR` · **Submitted:** 2026-02-18
+- **Authors:** Gabriele Barlacchi, Margherita Lalli, Emanuele Ferragina, et al.
+- **abs:** https://arxiv.org/abs/2602.16315 · **PDF:** https://arxiv.org/pdf/2602.16315v1
+
+## 11492. MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks
+
+- **ID:** `2602.16313` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Zexue He, Yu Wang, Churan Zhi, et al.
+- **abs:** https://arxiv.org/abs/2602.16313 · **PDF:** https://arxiv.org/pdf/2602.16313v1
+
+## 11493. The Weight of a Bit: EMFI Sensitivity Analysis of Embedded Deep Learning Models
+
+- **ID:** `2602.16309` · **primary:** `cs.CR` · **Submitted:** 2026-02-18
+- **Authors:** Jakub Breier, Štefan Kučerák, Xiaolu Hou
+- **abs:** https://arxiv.org/abs/2602.16309 · **PDF:** https://arxiv.org/pdf/2602.16309v1
+
+## 11494. Generative AI Usage of University Students: Navigating Between Education and Business
+
+- **ID:** `2602.16307` · **primary:** `cs.CY` · **Submitted:** 2026-02-18
+- **Authors:** Fabian Walke, Veronika Föller
+- **abs:** https://arxiv.org/abs/2602.16307 · **PDF:** https://arxiv.org/pdf/2602.16307v1
+
+## 11495. BAT: Better Audio Transformer Guided by Convex Gated Probing
+
+- **ID:** `2602.16305` · **primary:** `cs.SD` · **Submitted:** 2026-02-18
+- **Authors:** Houtan Ghaffari, Lukas Rauch, Christoph Scholz, et al.
+- **abs:** https://arxiv.org/abs/2602.16305 · **PDF:** https://arxiv.org/pdf/2602.16305v1
+
+## 11496. Multi-agent cooperation through in-context co-player inference
+
+- **ID:** `2602.16301` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Marissa A. Weis, Maciej Wołczyk, Rajai Nasser, et al.
+- **abs:** https://arxiv.org/abs/2602.16301 · **PDF:** https://arxiv.org/pdf/2602.16301v1
+
+## 11497. PREFER: An Ontology for the PREcision FERmentation Community
+
+- **ID:** `2602.16755` · **primary:** `q-bio.OT` · **Submitted:** 2026-02-18
+- **Authors:** Txell Amigó, Shawn Zheng Kai Tan, Angel Luu Phanthanourak, et al.
+- **abs:** https://arxiv.org/abs/2602.16755 · **PDF:** https://arxiv.org/pdf/2602.16755v1
+
+## 11498. MultiCW: A Large-Scale Balanced Benchmark Dataset for Training Robust Check-Worthiness Detection Models
+
+- **ID:** `2602.16298` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Martin Hyben, Sebastian Kula, Jan Cegin, et al.
+- **abs:** https://arxiv.org/abs/2602.16298 · **PDF:** https://arxiv.org/pdf/2602.16298v1
+
+## 11499. Aladdin-FTI @ AMIYA Three Wishes for Arabic NLP: Fidelity, Diglossia, and Multidialectal Generation
+
+- **ID:** `2602.16290` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Jonathan Mutal, Perla Al Almaoui, Simon Hengchen, et al.
+- **abs:** https://arxiv.org/abs/2602.16290 · **PDF:** https://arxiv.org/pdf/2602.16290v1
+
+## 11500. Fast KV Compaction via Attention Matching
+
+- **ID:** `2602.16284` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Adam Zweiger, Xinghong Fu, Han Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.16284 · **PDF:** https://arxiv.org/pdf/2602.16284v1
+
+## 11501. Regret and Sample Complexity of Online Q-Learning via Concentration of Stochastic Approximation with Time-Inhomogeneous Markov Chains
+
+- **ID:** `2602.16274` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Rahul Singh, Siddharth Chandak, Eric Moulines, et al.
+- **abs:** https://arxiv.org/abs/2602.16274 · **PDF:** https://arxiv.org/pdf/2602.16274v1
+
+## 11502. Lyapunov Spectral Analysis of Speech Embedding Trajectories in Psychosis
+
+- **ID:** `2602.16273` · **primary:** `nlin.AO` · **Submitted:** 2026-02-18
+- **Authors:** Jelena Vasic, Branislav Andjelic, Ana Mancic, et al.
+- **abs:** https://arxiv.org/abs/2602.16273 · **PDF:** https://arxiv.org/pdf/2602.16273v1
+
+## 11503. Data-Rate-Aware High-Speed CNN Inference on FPGAs
+
+- **ID:** `2603.08726` · **primary:** `cs.AR` · **Submitted:** 2026-02-18
+- **Authors:** Tobias Habermann, Martin Kumm
+- **abs:** https://arxiv.org/abs/2603.08726 · **PDF:** https://arxiv.org/pdf/2603.08726v1
+
+## 11504. Gloss-Free Sign Language Translation: An Unbiased Evaluation of Progress in the Field
+
+- **ID:** `2603.13240` · **primary:** `cs.CV` · **Submitted:** 2026-02-18
+- **Authors:** Ozge Mercanoglu Sincan, Jian He Low, Sobhan Asasi, et al.
+- **abs:** https://arxiv.org/abs/2603.13240 · **PDF:** https://arxiv.org/pdf/2603.13240v1
+
+## 11505. Structured Unitary Tensor Network Representations for Circuit-Efficient Quantum Data Encoding
+
+- **ID:** `2602.16266` · **primary:** `quant-ph` · **Submitted:** 2026-02-18
+- **Authors:** Guang Lin, Toshihisa Tanaka, Qibin Zhao
+- **abs:** https://arxiv.org/abs/2602.16266 · **PDF:** https://arxiv.org/pdf/2602.16266v1
+
+## 11506. On sparsity, extremal structure, and monotonicity properties of Wasserstein and Gromov-Wasserstein optimal transport plans
+
+- **ID:** `2602.16265` · **primary:** `stat.ML` · **Submitted:** 2026-02-18
+- **Authors:** Titouan Vayer
+- **abs:** https://arxiv.org/abs/2602.16265 · **PDF:** https://arxiv.org/pdf/2602.16265v2
+
+## 11507. Prediction of Major Solar Flares Using Interpretable Class-dependent Reward Framework with Active Region Magnetograms and Domain Knowledge
+
+- **ID:** `2602.16264` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Zixian Wu, Xuebao Li, Yanfang Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.16264 · **PDF:** https://arxiv.org/pdf/2602.16264v1
+
+## 11508. RoboLayout: Differentiable 3D Scene Generation for Embodied Agents
+
+- **ID:** `2603.05522` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Ali Shamsaddinlou
+- **abs:** https://arxiv.org/abs/2603.05522 · **PDF:** https://arxiv.org/pdf/2603.05522v2
+
+## 11509. Color-based Emotion Representation for Speech Emotion Recognition
+
+- **ID:** `2602.16256` · **primary:** `eess.AS` · **Submitted:** 2026-02-18
+- **Authors:** Ryotaro Nagase, Ryoichi Takashima, Yoichi Yamashita
+- **abs:** https://arxiv.org/abs/2602.16256 · **PDF:** https://arxiv.org/pdf/2602.16256v1
+
+## 11510. Toward Scalable Verifiable Reward: Proxy State-Based Evaluation for Multi-turn Tool-Calling LLM Agents
+
+- **ID:** `2602.16246` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Yun-Shiuan Chuang, Chaitanya Kulkarni, Alec Chiu, et al.
+- **abs:** https://arxiv.org/abs/2602.16246 · **PDF:** https://arxiv.org/pdf/2602.16246v2
+
+## 11511. LIDS: LLM Summary Inference Under the Layered Lens
+
+- **ID:** `2603.00105` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Dylan Park, Yingying Fan, Jinchi Lv
+- **abs:** https://arxiv.org/abs/2603.00105 · **PDF:** https://arxiv.org/pdf/2603.00105v1
+
+## 11512. Are LLMs Ready to Replace Bangla Annotators?
+
+- **ID:** `2602.16241` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Md. Najib Hasan, Touseef Hasan, Souvika Sarkar
+- **abs:** https://arxiv.org/abs/2602.16241 · **PDF:** https://arxiv.org/pdf/2602.16241v3
+
+## 11513. Online Prediction of Stochastic Sequences with High Probability Regret Bounds
+
+- **ID:** `2602.16236` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Matthias Frey, Jonathan H. Manton, Jingge Zhu
+- **abs:** https://arxiv.org/abs/2602.16236 · **PDF:** https://arxiv.org/pdf/2602.16236v1
+
+## 11514. Performance Analysis of Edge and In-Sensor AI Processors: A Comparative Review
+
+- **ID:** `2603.08725` · **primary:** `cs.AR` · **Submitted:** 2026-02-18
+- **Authors:** Luigi Capogrosso, Pietro Bonazzi, Michele Magno
+- **abs:** https://arxiv.org/abs/2603.08725 · **PDF:** https://arxiv.org/pdf/2603.08725v1
+
+## 11515. DistributedEstimator: Distributed Training of Quantum Neural Networks via Circuit Cutting
+
+- **ID:** `2602.16233` · **primary:** `cs.DC` · **Submitted:** 2026-02-18
+- **Authors:** Prabhjot Singh, Adel N. Toosi, Rajkumar Buyya
+- **abs:** https://arxiv.org/abs/2602.16233 · **PDF:** https://arxiv.org/pdf/2602.16233v1
+
+## 11516. Factored Latent Action World Models
+
+- **ID:** `2602.16229` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Zizhao Wang, Chang Shi, Jiaheng Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.16229 · **PDF:** https://arxiv.org/pdf/2602.16229v1
+
+## 11517. Know When You're Wrong: Aligning Confidence with Correctness for LLM Error Detection
+
+- **ID:** `2603.06604` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Xie Xiaohu, Liu Xiaohu, Yao Benjamin
+- **abs:** https://arxiv.org/abs/2603.06604 · **PDF:** https://arxiv.org/pdf/2603.06604v1
+
+## 11518. Amortized Predictability-aware Training Framework for Time Series Forecasting and Classification
+
+- **ID:** `2602.16224` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Xu Zhang, Peng Wang, Yichen Li, et al.
+- **abs:** https://arxiv.org/abs/2602.16224 · **PDF:** https://arxiv.org/pdf/2602.16224v1
+
+## 11519. SEMixer: Semantics Enhanced MLP-Mixer for Multiscale Mixing and Long-term Time Series Forecasting
+
+- **ID:** `2602.16220` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Xu Zhang, Qitong Wang, Peng Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.16220 · **PDF:** https://arxiv.org/pdf/2602.16220v1
+
+## 11520. Bayesian Quadrature: Gaussian Processes for Integration
+
+- **ID:** `2602.16218` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Maren Mahsereci, Toni Karvonen
+- **abs:** https://arxiv.org/abs/2602.16218 · **PDF:** https://arxiv.org/pdf/2602.16218v1
+
+## 11521. Multi-Class Boundary Extraction from Implicit Representations
+
+- **ID:** `2602.16217` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Jash Vira, Andrew Myers, Simon Ratcliffe
+- **abs:** https://arxiv.org/abs/2602.16217 · **PDF:** https://arxiv.org/pdf/2602.16217v1
+
+## 11522. UCTECG-Net: Uncertainty-aware Convolution Transformer ECG Network for Arrhythmia Detection
+
+- **ID:** `2602.16216` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Hamzeh Asgharnezhad, Pegah Tabarisaadi, Abbas Khosravi, et al.
+- **abs:** https://arxiv.org/abs/2602.16216 · **PDF:** https://arxiv.org/pdf/2602.16216v1
+
+## 11523. Graph neural network for colliding particles with an application to sea ice floe modeling
+
+- **ID:** `2602.16213` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Ruibiao Zhu
+- **abs:** https://arxiv.org/abs/2602.16213 · **PDF:** https://arxiv.org/pdf/2602.16213v1
+
+## 11524. Auction-Based RIS Allocation With DRL: Controlling the Cost-Performance Trade-Off
+
+- **ID:** `2603.04433` · **primary:** `cs.NI` · **Submitted:** 2026-02-18
+- **Authors:** Martin Mark Zan, Stefan Schwarz
+- **abs:** https://arxiv.org/abs/2603.04433 · **PDF:** https://arxiv.org/pdf/2603.04433v1
+
+## 11525. Geometric Neural Operators via Lie Group-Constrained Latent Dynamics
+
+- **ID:** `2602.16209` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Jiaquan Zhang, Fachrina Dewi Puspitasari, Songbo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.16209 · **PDF:** https://arxiv.org/pdf/2602.16209v1
+
+## 11526. Linked Data Classification using Neurochaos Learning
+
+- **ID:** `2602.16204` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Pooja Honna, Ayush Patravali, Nithin Nagaraj, et al.
+- **abs:** https://arxiv.org/abs/2602.16204 · **PDF:** https://arxiv.org/pdf/2602.16204v1
+
+## 11527. Long-Tail Knowledge in Large Language Models: Taxonomy, Mechanisms, Interventions and Implications
+
+- **ID:** `2602.16201` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Sanket Badhe, Deep Shah, Nehal Kathrotia
+- **abs:** https://arxiv.org/abs/2602.16201 · **PDF:** https://arxiv.org/pdf/2602.16201v1
+
+## 11528. The Validity of Coreference-based Evaluations of Natural Language Understanding
+
+- **ID:** `2602.16200` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Ian Porada
+- **abs:** https://arxiv.org/abs/2602.16200 · **PDF:** https://arxiv.org/pdf/2602.16200v1
+
+## 11529. Training-Free Adaptation of Diffusion Models via Doob's $h$-Transform
+
+- **ID:** `2602.16198` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Qijie Zhu, Zeqi Ye, Han Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.16198 · **PDF:** https://arxiv.org/pdf/2602.16198v1
+
+## 11530. ModalImmune: Immunity Driven Unlearning via Self Destructive Training
+
+- **ID:** `2602.16197` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Rong Fu, Jia Yee Tan, Zijian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.16197 · **PDF:** https://arxiv.org/pdf/2602.16197v2
+
+## 11531. Graphon Mean-Field Subsampling for Cooperative Heterogeneous Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.16196` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Emile Anand, Richard Hoffmann, Sarah Liaw, et al.
+- **abs:** https://arxiv.org/abs/2602.16196 · **PDF:** https://arxiv.org/pdf/2602.16196v1
+
+## 11532. Temporal Panel Selection in Ongoing Citizens' Assemblies
+
+- **ID:** `2602.16194` · **primary:** `cs.GT` · **Submitted:** 2026-02-18
+- **Authors:** Yusuf Hakan Kalayci, Evi Micha
+- **abs:** https://arxiv.org/abs/2602.16194 · **PDF:** https://arxiv.org/pdf/2602.16194v1
+
+## 11533. Rethinking Input Domains in Physics-Informed Neural Networks via Geometric Compactification Mappings
+
+- **ID:** `2602.16193` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Zhenzhen Huang, Haoyu Bian, Jiaquan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.16193 · **PDF:** https://arxiv.org/pdf/2602.16193v1
+
+## 11534. Revolutionizing Long-Term Memory in AI: New Horizons with High-Capacity and High-Speed Storage
+
+- **ID:** `2602.16192` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Hiroaki Yamanaka, Daisuke Miyashita, Takashi Toi, et al.
+- **abs:** https://arxiv.org/abs/2602.16192 · **PDF:** https://arxiv.org/pdf/2602.16192v1
+
+## 11535. Scale Dependent Data Duplication
+
+- **ID:** `2603.06603` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Joshua Kazdan, Noam Levi, Rylan Schaeffer, et al.
+- **abs:** https://arxiv.org/abs/2603.06603 · **PDF:** https://arxiv.org/pdf/2603.06603v1
+
+## 11536. Beyond Learning: A Training-Free Alternative to Model Adaptation
+
+- **ID:** `2602.16189` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Namkyung Yoon, Kyeonghyun Yoo, Wooyong Jung, et al.
+- **abs:** https://arxiv.org/abs/2602.16189 · **PDF:** https://arxiv.org/pdf/2602.16189v1
+
+## 11537. Deep TPC: Temporal-Prior Conditioning for Time Series Forecasting
+
+- **ID:** `2602.16188` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Filippos Bellos, NaveenJohn Premkumar, Yannis Avrithis, et al.
+- **abs:** https://arxiv.org/abs/2602.16188 · **PDF:** https://arxiv.org/pdf/2602.16188v1
+
+## 11538. SIT-LMPC: Safe Information-Theoretic Learning Model Predictive Control for Iterative Tasks
+
+- **ID:** `2602.16187` · **primary:** `cs.RO` · **Submitted:** 2026-02-18
+- **Authors:** Zirui Zang, Ahmad Amine, Nick-Marios T. Kokolakis, et al.
+- **abs:** https://arxiv.org/abs/2602.16187 · **PDF:** https://arxiv.org/pdf/2602.16187v1
+
+## 11539. Multi-Agent Combinatorial-Multi-Armed-Bandit framework for the Submodular Welfare Problem under Bandit Feedback
+
+- **ID:** `2602.16183` · **primary:** `cs.GT` · **Submitted:** 2026-02-18
+- **Authors:** Subham Pokhriyal, Shweta Jain, Vaneet Aggarwal
+- **abs:** https://arxiv.org/abs/2602.16183 · **PDF:** https://arxiv.org/pdf/2602.16183v1
+
+## 11540. U-FedTomAtt: Ultra-lightweight Federated Learning with Attention for Tomato Disease Recognition
+
+- **ID:** `2602.16749` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-18
+- **Authors:** Romiyal George, Sathiyamohan Nishankar, Selvarajah Thuseethan, et al.
+- **abs:** https://arxiv.org/abs/2602.16749 · **PDF:** https://arxiv.org/pdf/2602.16749v1
+
+## 11541. Towards Secure and Scalable Energy Theft Detection: A Federated Learning Approach for Resource-Constrained Smart Meters
+
+- **ID:** `2602.16181` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Diego Labate, Dipanwita Thakur, Giancarlo Fortino
+- **abs:** https://arxiv.org/abs/2602.16181 · **PDF:** https://arxiv.org/pdf/2602.16181v1
+
+## 11542. EnterpriseBench Corecraft: Training Generalizable Agents on High-Fidelity RL Environments
+
+- **ID:** `2602.16179` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Sushant Mehta, Logan Ritchie, Suhaas Garre, et al.
+- **abs:** https://arxiv.org/abs/2602.16179 · **PDF:** https://arxiv.org/pdf/2602.16179v5
+
+## 11543. Conjugate Learning Theory: Uncovering the Mechanisms of Trainability and Generalization in Deep Neural Networks
+
+- **ID:** `2602.16177` · **primary:** `stat.ML` · **Submitted:** 2026-02-18
+- **Authors:** Binchuan Qi
+- **abs:** https://arxiv.org/abs/2602.16177 · **PDF:** https://arxiv.org/pdf/2602.16177v2
+
+## 11544. Alpha-RF: Automated RF-Filter-Circuit Design with Neural Simulator and Reinforcement Learning
+
+- **ID:** `2603.00104` · **primary:** `eess.SP` · **Submitted:** 2026-02-18
+- **Authors:** Nhat Tran, Chenjie Hao, Alexander Stameroff, et al.
+- **abs:** https://arxiv.org/abs/2603.00104 · **PDF:** https://arxiv.org/pdf/2603.00104v1
+
+## 11545. Edge Learning via Federated Split Decision Transformers for Metaverse Resource Allocation
+
+- **ID:** `2602.16174` · **primary:** `cs.NI` · **Submitted:** 2026-02-18
+- **Authors:** Fatih Temiz, Shavbo Salehi, Melike Erol-Kantarci
+- **abs:** https://arxiv.org/abs/2602.16174 · **PDF:** https://arxiv.org/pdf/2602.16174v1
+
+## 11546. Learning Personalized Agents from Human Feedback
+
+- **ID:** `2602.16173` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Kaiqu Liang, Julia Kruk, Shengyi Qian, et al.
+- **abs:** https://arxiv.org/abs/2602.16173 · **PDF:** https://arxiv.org/pdf/2602.16173v1
+
+## 11547. Discrete Stochastic Localization for Non-autoregressive Generation
+
+- **ID:** `2602.16169` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Yunshu Wu, Jiayi Cheng, Partha Thakuria, et al.
+- **abs:** https://arxiv.org/abs/2602.16169 · **PDF:** https://arxiv.org/pdf/2602.16169v1
+
+## 11548. LiveClin: A Live Clinical Benchmark without Leakage
+
+- **ID:** `2602.16747` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Xidong Wang, Shuqi Guo, Yue Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.16747 · **PDF:** https://arxiv.org/pdf/2602.16747v1
+
+## 11549. Low-Dimensional and Transversely Curved Optimization Dynamics in Grokking
+
+- **ID:** `2602.16746` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Yongzhong Xu
+- **abs:** https://arxiv.org/abs/2602.16746 · **PDF:** https://arxiv.org/pdf/2602.16746v2
+
+## 11550. Muon with Spectral Guidance: Efficient Optimization for Scientific Machine Learning
+
+- **ID:** `2602.16167` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Binghang Lu, Jiahao Zhang, Guang Lin
+- **abs:** https://arxiv.org/abs/2602.16167 · **PDF:** https://arxiv.org/pdf/2602.16167v1
+
+## 11551. HiPER: Hierarchical Reinforcement Learning with Explicit Credit Assignment for Large Language Model Agents
+
+- **ID:** `2602.16165` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Jiangweizhi Peng, Yuanxin Liu, Ruida Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.16165 · **PDF:** https://arxiv.org/pdf/2602.16165v1
+
+## 11552. PETS: A Principled Framework Towards Optimal Trajectory Allocation for Efficient Test-Time Self-Consistency
+
+- **ID:** `2602.16745` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Zhangyi Liu, Huaizhi Qu, Xiaowei Yin, et al.
+- **abs:** https://arxiv.org/abs/2602.16745 · **PDF:** https://arxiv.org/pdf/2602.16745v1
+
+## 11553. LLMs Exhibit Significantly Lower Uncertainty in Creative Writing Than Professional Writers
+
+- **ID:** `2602.16162` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Peiqi Sui
+- **abs:** https://arxiv.org/abs/2602.16162 · **PDF:** https://arxiv.org/pdf/2602.16162v1
+
+## 11554. Emotion Collider: Dual Hyperbolic Mirror Manifolds for Sentiment Recovery via Anti Emotion Reflection
+
+- **ID:** `2602.16161` · **primary:** `cs.MM` · **Submitted:** 2026-02-18
+- **Authors:** Rong Fu, Ziming Wang, Shuo Yin, et al.
+- **abs:** https://arxiv.org/abs/2602.16161 · **PDF:** https://arxiv.org/pdf/2602.16161v2
+
+## 11555. Differentially Private Non-convex Distributionally Robust Optimization
+
+- **ID:** `2602.16155` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Difei Xu, Meng Ding, Zebin Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.16155 · **PDF:** https://arxiv.org/pdf/2602.16155v1
+
+## 11556. Balancing Faithfulness and Performance in Reasoning via Multi-Listener Soft Execution
+
+- **ID:** `2602.16154` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Nithin Sivakumaran, Shoubin Yu, Hyunji Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.16154 · **PDF:** https://arxiv.org/pdf/2602.16154v1
+
+## 11557. Local adapt-then-combine algorithms for distributed nonsmooth optimization: Achieving provable communication acceleration
+
+- **ID:** `2602.16148` · **primary:** `math.OC` · **Submitted:** 2026-02-18
+- **Authors:** Luyao Guo, Xinli Shi, Wenying Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.16148 · **PDF:** https://arxiv.org/pdf/2602.16148v1
+
+## 11558. ASPEN: Spectral-Temporal Fusion for Cross-Subject Brain Decoding
+
+- **ID:** `2602.16147` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Megan Lee, Seung Ha Hwang, Inhyeok Choi, et al.
+- **abs:** https://arxiv.org/abs/2602.16147 · **PDF:** https://arxiv.org/pdf/2602.16147v1
+
+## 11559. Investigating GNN Convergence on Large Randomly Generated Graphs with Realistic Node Feature Correlations
+
+- **ID:** `2602.16145` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Mohammed Zain Ali Ahmed
+- **abs:** https://arxiv.org/abs/2602.16145 · **PDF:** https://arxiv.org/pdf/2602.16145v1
+
+## 11560. Missing-by-Design: Certifiable Modality Deletion for Revocable Multimodal Sentiment Analysis
+
+- **ID:** `2602.16144` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Rong Fu, Ziming Wang, Chunlei Meng, et al.
+- **abs:** https://arxiv.org/abs/2602.16144 · **PDF:** https://arxiv.org/pdf/2602.16144v2
+
+## 11561. Ratio Covers of Convex Sets and Optimal Mixture Density Estimation
+
+- **ID:** `2602.16142` · **primary:** `math.ST` · **Submitted:** 2026-02-18
+- **Authors:** Spencer Compton, Gábor Lugosi, Jaouad Mourtada, et al.
+- **abs:** https://arxiv.org/abs/2602.16142 · **PDF:** https://arxiv.org/pdf/2602.16142v2
+
+## 11562. Human-AI Collaboration in Large Language Model-Integrated Building Energy Management Systems: The Role of User Domain Knowledge and AI Literacy
+
+- **ID:** `2602.16140` · **primary:** `cs.HC` · **Submitted:** 2026-02-18
+- **Authors:** Wooyoung Jung, Kahyun Jeon, Prosper Babon-Ayeng
+- **abs:** https://arxiv.org/abs/2602.16140 · **PDF:** https://arxiv.org/pdf/2602.16140v1
+
+## 11563. Retrieval Collapses When AI Pollutes the Web
+
+- **ID:** `2602.16136` · **primary:** `cs.IR` · **Submitted:** 2026-02-18
+- **Authors:** Hongyeon Yu, Dongchan Kim, Young-Bum Kim
+- **abs:** https://arxiv.org/abs/2602.16136 · **PDF:** https://arxiv.org/pdf/2602.16136v1
+
+## 11564. CHAI: CacHe Attention Inference for text2video
+
+- **ID:** `2602.16132` · **primary:** `cs.CV` · **Submitted:** 2026-02-18
+- **Authors:** Joel Mathew Cherian, Ashutosh Muralidhara Bharadwaj, Vima Gupta, et al.
+- **abs:** https://arxiv.org/abs/2602.16132 · **PDF:** https://arxiv.org/pdf/2602.16132v1
+
+## 11565. DeepVision-103K: A Visually Diverse, Broad-Coverage, and Verifiable Mathematical Dataset for Multimodal Reasoning
+
+- **ID:** `2602.16742` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Haoxiang Sun, Lizhen Xu, Bing Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.16742 · **PDF:** https://arxiv.org/pdf/2602.16742v1
+
+## 11566. Empirical Cumulative Distribution Function Clustering for LLM-based Agent System Analysis
+
+- **ID:** `2602.16131` · **primary:** `stat.ML` · **Submitted:** 2026-02-18
+- **Authors:** Chihiro Watanabe, Jingyu Sun
+- **abs:** https://arxiv.org/abs/2602.16131 · **PDF:** https://arxiv.org/pdf/2602.16131v1
+
+## 11567. On the Power of Source Screening for Learning Shared Feature Extractors
+
+- **ID:** `2602.16125` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Leo Muxing Wang, Connor Mclaughlin, Lili Su
+- **abs:** https://arxiv.org/abs/2602.16125 · **PDF:** https://arxiv.org/pdf/2602.16125v2
+
+## 11568. Rethinking ANN-based Retrieval: Multifaceted Learnable Index for Large-scale Recommendation System
+
+- **ID:** `2602.16124` · **primary:** `cs.IR` · **Submitted:** 2026-02-18
+- **Authors:** Jiang Zhang, Yubo Wang, Wei Chang, et al.
+- **abs:** https://arxiv.org/abs/2602.16124 · **PDF:** https://arxiv.org/pdf/2602.16124v1
+
+## 11569. Feature-based morphological analysis of shape graph data
+
+- **ID:** `2602.16120` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Murad Hossen, Demetrio Labate, Nicolas Charon
+- **abs:** https://arxiv.org/abs/2602.16120 · **PDF:** https://arxiv.org/pdf/2602.16120v1
+
+## 11570. Evolutionary Context Search for Automated Skill Acquisition
+
+- **ID:** `2602.16113` · **primary:** `cs.NE` · **Submitted:** 2026-02-18
+- **Authors:** Qi Sun, Stefan Nielsen, Rio Yokota, et al.
+- **abs:** https://arxiv.org/abs/2602.16113 · **PDF:** https://arxiv.org/pdf/2602.16113v1
+
+## 11571. Surrogate-Based Prevalence Measurement for Large-Scale A/B Testing
+
+- **ID:** `2602.16111` · **primary:** `stat.AP` · **Submitted:** 2026-02-18
+- **Authors:** Zehao Xu, Tony Paek, Kevin O'Sullivan, et al.
+- **abs:** https://arxiv.org/abs/2602.16111 · **PDF:** https://arxiv.org/pdf/2602.16111v1
+
+## 11572. OmniCT: Towards a Unified Slice-Volume LVLM for Comprehensive CT Analysis
+
+- **ID:** `2602.16110` · **primary:** `cs.CV` · **Submitted:** 2026-02-18
+- **Authors:** Tianwei Lin, Zhongwei Qiu, Wenqiao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.16110 · **PDF:** https://arxiv.org/pdf/2602.16110v2
+
+## 11573. Federated Graph AGI for Cross-Border Insider Threat Intelligence in Government Financial Schemes
+
+- **ID:** `2602.16109` · **primary:** `cs.CR` · **Submitted:** 2026-02-18
+- **Authors:** Srikumar Nayak, James Walmesley
+- **abs:** https://arxiv.org/abs/2602.16109 · **PDF:** https://arxiv.org/pdf/2602.16109v1
+
+## 11574. Can Adversarial Code Comments Fool AI Security Reviewers -- Large-Scale Empirical Study of Comment-Based Attacks and Defenses Against LLM Code Analysis
+
+- **ID:** `2602.16741` · **primary:** `cs.CR` · **Submitted:** 2026-02-18
+- **Authors:** Scott Thornton
+- **abs:** https://arxiv.org/abs/2602.16741 · **PDF:** https://arxiv.org/pdf/2602.16741v1
+
+## 11575. GPSBench: Do Large Language Models Understand GPS Coordinates?
+
+- **ID:** `2602.16105` · **primary:** `cs.AI` · **Submitted:** 2026-02-18
+- **Authors:** Thinh Hung Truong, Jey Han Lau, Jianzhong Qi
+- **abs:** https://arxiv.org/abs/2602.16105 · **PDF:** https://arxiv.org/pdf/2602.16105v1
+
+## 11576. Axle Sensor Fusion for Online Continual Wheel Fault Detection in Wayside Railway Monitoring
+
+- **ID:** `2602.16101` · **primary:** `cs.LG` · **Submitted:** 2026-02-18
+- **Authors:** Afonso Lourenço, Francisca Osório, Diogo Risca, et al.
+- **abs:** https://arxiv.org/abs/2602.16101 · **PDF:** https://arxiv.org/pdf/2602.16101v1
+
+## 11577. Quantifying and Attributing Submodel Uncertainty in Stochastic Simulation Models and Digital Twins
+
+- **ID:** `2602.16099` · **primary:** `stat.CO` · **Submitted:** 2026-02-18
+- **Authors:** Mohammadmahdi Ghasemloo, David J. Eckman, Yaxian Li
+- **abs:** https://arxiv.org/abs/2602.16099 · **PDF:** https://arxiv.org/pdf/2602.16099v1
+
+## 11578. Measuring and Eliminating Refusals in Military Large Language Models
+
+- **ID:** `2603.10012` · **primary:** `cs.CL` · **Submitted:** 2026-02-18
+- **Authors:** Jack FitzGerald, Dylan Bates, Aristotelis Lazaridis, et al.
+- **abs:** https://arxiv.org/abs/2603.10012 · **PDF:** https://arxiv.org/pdf/2603.10012v1
+
+## 11579. Collaborative Zone-Adaptive Zero-Day Intrusion Detection for IoBT
+
+- **ID:** `2602.16098` · **primary:** `cs.CR` · **Submitted:** 2026-02-18
+- **Authors:** Amirmohammad Pasdar, Shabnam Kasra Kermanshahi, Nour Moustafa, et al.
+- **abs:** https://arxiv.org/abs/2602.16098 · **PDF:** https://arxiv.org/pdf/2602.16098v1
+
+## 11580. Updating Parametric Knowledge with Context Distillation Retains Post-Training Capabilities
+
+- **ID:** `2602.16093` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Shankar Padmanabhan, Mustafa Omer Gul, Tanya Goyal
+- **abs:** https://arxiv.org/abs/2602.16093 · **PDF:** https://arxiv.org/pdf/2602.16093v1
+
+## 11581. Why Any-Order Autoregressive Models Need Two-Stream Attention: A Structural-Semantic Tradeoff
+
+- **ID:** `2602.16092` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Patrick Pynadath, Ruqi Zhang
+- **abs:** https://arxiv.org/abs/2602.16092 · **PDF:** https://arxiv.org/pdf/2602.16092v1
+
+## 11582. Examining Fast Radiative Feedbacks Using Machine-Learning Weather Emulators
+
+- **ID:** `2602.16090` · **primary:** `physics.ao-ph` · **Submitted:** 2026-02-17
+- **Authors:** Ankur Mahesh, William D. Collins, Travis A. O'Brien, et al.
+- **abs:** https://arxiv.org/abs/2602.16090 · **PDF:** https://arxiv.org/pdf/2602.16090v1
+
+## 11583. Quantifying LLM Attention-Head Stability: Implications for Circuit Universality
+
+- **ID:** `2602.16740` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Karan Bali, Jack Stanley, Praneet Suresh, et al.
+- **abs:** https://arxiv.org/abs/2602.16740 · **PDF:** https://arxiv.org/pdf/2602.16740v1
+
+## 11584. LGQ: Learning Discretization Geometry for Scalable and Stable Image Tokenization
+
+- **ID:** `2602.16086` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Idil Bilge Altun, Mert Onur Cakiroglu, Elham Buxton, et al.
+- **abs:** https://arxiv.org/abs/2602.16086 · **PDF:** https://arxiv.org/pdf/2602.16086v2
+
+## 11585. Language Statistics and False Belief Reasoning: Evidence from 41 Open-Weight LMs
+
+- **ID:** `2602.16085` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Sean Trott, Samuel Taylor, Cameron Jones, et al.
+- **abs:** https://arxiv.org/abs/2602.16085 · **PDF:** https://arxiv.org/pdf/2602.16085v1
+
+## 11586. Activation Steering via Generative Causal Mediation
+
+- **ID:** `2602.16080` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Aruna Sankaranarayanan, Amir Zur, Atticus Geiger, et al.
+- **abs:** https://arxiv.org/abs/2602.16080 · **PDF:** https://arxiv.org/pdf/2602.16080v2
+
+## 11587. DARTH-PUM: A Hybrid Processing-Using-Memory Architecture
+
+- **ID:** `2602.16075` · **primary:** `cs.AR` · **Submitted:** 2026-02-17
+- **Authors:** Ryan Wong, Ben Feinberg, Saugata Ghose
+- **abs:** https://arxiv.org/abs/2602.16075 · **PDF:** https://arxiv.org/pdf/2602.16075v1
+
+## 11588. ScenicRules: An Autonomous Driving Benchmark with Multi-Objective Specifications and Abstract Scenarios
+
+- **ID:** `2602.16073` · **primary:** `cs.RO` · **Submitted:** 2026-02-17
+- **Authors:** Kevin Kai-Chun Chang, Ekin Beyazit, Alberto Sangiovanni-Vincentelli, et al.
+- **abs:** https://arxiv.org/abs/2602.16073 · **PDF:** https://arxiv.org/pdf/2602.16073v1
+
+## 11589. Omni-iEEG: A Large-Scale, Comprehensive iEEG Dataset and Benchmark for Epilepsy Research
+
+- **ID:** `2602.16072` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Chenda Duan, Yipeng Zhang, Sotaro Kanai, et al.
+- **abs:** https://arxiv.org/abs/2602.16072 · **PDF:** https://arxiv.org/pdf/2602.16072v2
+
+## 11590. The Limits of Long-Context Reasoning in Automated Bug Fixing
+
+- **ID:** `2602.16069` · **primary:** `cs.SE` · **Submitted:** 2026-02-17
+- **Authors:** Ravi Raju, Mengmeng Ji, Shubhangi Upasani, et al.
+- **abs:** https://arxiv.org/abs/2602.16069 · **PDF:** https://arxiv.org/pdf/2602.16069v2
+
+## 11591. Real-time Secondary Crash Likelihood Prediction Excluding Post Primary Crash Features
+
+- **ID:** `2602.16739` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Lei Han, Mohamed Abdel-Aty, Zubayer Islam, et al.
+- **abs:** https://arxiv.org/abs/2602.16739 · **PDF:** https://arxiv.org/pdf/2602.16739v1
+
+## 11592. Self-Evolving Multi-Agent Network for Industrial IoT Predictive Maintenance
+
+- **ID:** `2602.16738` · **primary:** `cs.MA` · **Submitted:** 2026-02-17
+- **Authors:** Rebin Saleh, Khanh Pham Dinh, Balázs Villányi, et al.
+- **abs:** https://arxiv.org/abs/2602.16738 · **PDF:** https://arxiv.org/pdf/2602.16738v1
+
+## 11593. Improving Interactive In-Context Learning from Natural Language Feedback
+
+- **ID:** `2602.16066` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Martin Klissarov, Jonathan Cook, Diego Antognini, et al.
+- **abs:** https://arxiv.org/abs/2602.16066 · **PDF:** https://arxiv.org/pdf/2602.16066v1
+
+## 11594. Can Generative Artificial Intelligence Survive Data Contamination? Theoretical Guarantees under Contaminated Recursive Training
+
+- **ID:** `2602.16065` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Kevin Wang, Hongqian Niu, Didong Li
+- **abs:** https://arxiv.org/abs/2602.16065 · **PDF:** https://arxiv.org/pdf/2602.16065v1
+
+## 11595. MARLEM: A Multi-Agent Reinforcement Learning Simulation Framework for Implicit Cooperation in Decentralized Local Energy Markets
+
+- **ID:** `2602.16063` · **primary:** `eess.SY` · **Submitted:** 2026-02-17
+- **Authors:** Nelson Salazar-Pena, Alejandra Tabares, Andres Gonzalez-Mancera
+- **abs:** https://arxiv.org/abs/2602.16063 · **PDF:** https://arxiv.org/pdf/2602.16063v1
+
+## 11596. Harnessing Implicit Cooperation: A Multi-Agent Reinforcement Learning Approach Towards Decentralized Local Energy Markets
+
+- **ID:** `2602.16062` · **primary:** `eess.SY` · **Submitted:** 2026-02-17
+- **Authors:** Nelson Salazar-Pena, Alejandra Tabares, Andres Gonzalez-Mancera
+- **abs:** https://arxiv.org/abs/2602.16062 · **PDF:** https://arxiv.org/pdf/2602.16062v1
+
+## 11597. Partial Identification under Missing Data Using Weak Shadow Variables from Pretrained Models
+
+- **ID:** `2602.16061` · **primary:** `stat.ML` · **Submitted:** 2026-02-17
+- **Authors:** Hongyu Chen, David Simchi-Levi, Ruoxuan Xiong
+- **abs:** https://arxiv.org/abs/2602.16061 · **PDF:** https://arxiv.org/pdf/2602.16061v1
+
+## 11598. Extracting and Analyzing Rail Crossing Behavior Signatures from Videos using Tensor Methods
+
+- **ID:** `2602.16057` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Dawon Ahn, Het Patel, Aemal Khattak, et al.
+- **abs:** https://arxiv.org/abs/2602.16057 · **PDF:** https://arxiv.org/pdf/2602.16057v2
+
+## 11599. CLAA: Cross-Layer Attention Aggregation for Accelerating LLM Prefill
+
+- **ID:** `2602.16054` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Bradley McDanel, Steven Li, Harshit Khaitan
+- **abs:** https://arxiv.org/abs/2602.16054 · **PDF:** https://arxiv.org/pdf/2602.16054v1
+
+## 11600. Multi-Objective Alignment of Language Models for Personalized Psychotherapy
+
+- **ID:** `2602.16053` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Mehrab Beikzadeh, Yasaman Asadollah Salmanpour, Ashima Suvarna, et al.
+- **abs:** https://arxiv.org/abs/2602.16053 · **PDF:** https://arxiv.org/pdf/2602.16053v1
+
+## 11601. Gemma Needs Help: Investigating and Mitigating Emotional Instability in LLMs
+
+- **ID:** `2603.10011` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Anna Soligo, Vladimir Mikulik, William Saunders
+- **abs:** https://arxiv.org/abs/2603.10011 · **PDF:** https://arxiv.org/pdf/2603.10011v1
+
+## 11602. MoE-Spec: Expert Budgeting for Efficient Speculative Decoding
+
+- **ID:** `2602.16052` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Bradley McDanel, Steven Li, Sruthikesh Surineni, et al.
+- **abs:** https://arxiv.org/abs/2602.16052 · **PDF:** https://arxiv.org/pdf/2602.16052v1
+
+## 11603. Evidence-Grounded Subspecialty Reasoning: Evaluating a Curated Clinical Intelligence Layer on the 2025 Endocrinology Board-Style Examination
+
+- **ID:** `2602.16050` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Amir Hosseinian, MohammadReza Zare Shahneh, Umer Mansoor, et al.
+- **abs:** https://arxiv.org/abs/2602.16050 · **PDF:** https://arxiv.org/pdf/2602.16050v1
+
+## 11604. AI-CARE: Carbon-Aware Reporting Evaluation Metric for AI Models
+
+- **ID:** `2602.16042` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** KC Santosh, Srikanth Baride, Rodrigue Rizk
+- **abs:** https://arxiv.org/abs/2602.16042 · **PDF:** https://arxiv.org/pdf/2602.16042v2
+
+## 11605. How Uncertain Is the Grade? A Benchmark of Uncertainty Metrics for LLM-Based Automatic Assessment
+
+- **ID:** `2602.16039` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Hang Li, Kaiqi Yang, Xianxuan Long, et al.
+- **abs:** https://arxiv.org/abs/2602.16039 · **PDF:** https://arxiv.org/pdf/2602.16039v1
+
+## 11606. Heuristic Search as Language-Guided Program Optimization
+
+- **ID:** `2602.16038` · **primary:** `cs.NE` · **Submitted:** 2026-02-17
+- **Authors:** Mingxin Yu, Ruixiao Yang, Chuchu Fan
+- **abs:** https://arxiv.org/abs/2602.16038 · **PDF:** https://arxiv.org/pdf/2602.16038v1
+
+## 11607. Optimization Instability in Autonomous Agentic Workflows for Clinical Symptom Detection
+
+- **ID:** `2602.16037` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Cameron Cagan, Pedram Fard, Jiazi Tian, et al.
+- **abs:** https://arxiv.org/abs/2602.16037 · **PDF:** https://arxiv.org/pdf/2602.16037v1
+
+## 11608. A Computer Vision Framework for Multi-Class Detection and Tracking in Soccer Broadcast Footage
+
+- **ID:** `2602.18504` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Daniel Tshiani
+- **abs:** https://arxiv.org/abs/2602.18504 · **PDF:** https://arxiv.org/pdf/2602.18504v1
+
+## 11609. Transforming GenAI Policy to Prompting Instruction: An RCT of Scalable Prompting Interventions in a CS1 Course
+
+- **ID:** `2602.16033` · **primary:** `cs.HC` · **Submitted:** 2026-02-17
+- **Authors:** Ruiwei Xiao, Runlong Ye, Xinying Hou, et al.
+- **abs:** https://arxiv.org/abs/2602.16033 · **PDF:** https://arxiv.org/pdf/2602.16033v1
+
+## 11610. A Curious Class of Adpositional Multiword Expressions in Korean
+
+- **ID:** `2602.16023` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Junghyun Min, Na-Rae Han, Jena D. Hwang, et al.
+- **abs:** https://arxiv.org/abs/2602.16023 · **PDF:** https://arxiv.org/pdf/2602.16023v1
+
+## 11611. MolCrystalFlow: Molecular Crystal Structure Prediction via Flow Matching
+
+- **ID:** `2602.16020` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Cheng Zeng, Harry W. Sullivan, Thomas Egg, et al.
+- **abs:** https://arxiv.org/abs/2602.16020 · **PDF:** https://arxiv.org/pdf/2602.16020v3
+
+## 11612. MedProbCLIP: Probabilistic Adaptation of Vision-Language Foundation Model for Reliable Radiograph-Report Retrieval
+
+- **ID:** `2602.16019` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Ahmad Elallaf, Yu Zhang, Yuktha Priya Masupalli, et al.
+- **abs:** https://arxiv.org/abs/2602.16019 · **PDF:** https://arxiv.org/pdf/2602.16019v1
+
+## 11613. Edge-Local and Qubit-Efficient Quantum Graph Learning for the NISQ Era
+
+- **ID:** `2602.16018` · **primary:** `quant-ph` · **Submitted:** 2026-02-17
+- **Authors:** Armin Ahmadkhaniha, Jake Doliskani
+- **abs:** https://arxiv.org/abs/2602.16018 · **PDF:** https://arxiv.org/pdf/2602.16018v1
+
+## 11614. Geometry-Aware Uncertainty Quantification via Conformal Prediction on Manifolds
+
+- **ID:** `2602.16015` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Marzieh Amiri Shahbazi, Ali Baheri
+- **abs:** https://arxiv.org/abs/2602.16015 · **PDF:** https://arxiv.org/pdf/2602.16015v1
+
+## 11615. Towards Efficient Constraint Handling in Neural Solvers for Routing Problems
+
+- **ID:** `2602.16012` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Jieyi Bi, Zhiguang Cao, Jianan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.16012 · **PDF:** https://arxiv.org/pdf/2602.16012v1
+
+## 11616. MAEB: Massive Audio Embedding Benchmark
+
+- **ID:** `2602.16008` · **primary:** `cs.SD` · **Submitted:** 2026-02-17
+- **Authors:** Adnan El Assadi, Isaac Chung, Chenghao Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.16008 · **PDF:** https://arxiv.org/pdf/2602.16008v1
+
+## 11617. Exploring the Utility of MALDI-TOF Mass Spectrometry and Antimicrobial Resistance in Hospital Outbreak Detection
+
+- **ID:** `2602.16737` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-17
+- **Authors:** Chang Liu, Jieshi Chen, Alexander J. Sundermann, et al.
+- **abs:** https://arxiv.org/abs/2602.16737 · **PDF:** https://arxiv.org/pdf/2602.16737v1
+
+## 11618. FERRET: Framework for Expansion Reliant Red Teaming
+
+- **ID:** `2603.10010` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Ninareh Mehrabi, Vitor Albiero, Maya Pavlova, et al.
+- **abs:** https://arxiv.org/abs/2603.10010 · **PDF:** https://arxiv.org/pdf/2603.10010v1
+
+## 11619. The Compute ICE-AGE: Invariant Compute Envelope under Addressable Graph Evolution
+
+- **ID:** `2602.16736` · **primary:** `cs.OS` · **Submitted:** 2026-02-17
+- **Authors:** Raymond Jay Martin
+- **abs:** https://arxiv.org/abs/2602.16736 · **PDF:** https://arxiv.org/pdf/2602.16736v2
+
+## 11620. A Few-Shot LLM Framework for Extreme Day Classification in Electricity Markets
+
+- **ID:** `2602.16735` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Saud Alghumayjan, Ming Yi, Bolun Xu
+- **abs:** https://arxiv.org/abs/2602.16735 · **PDF:** https://arxiv.org/pdf/2602.16735v1
+
+## 11621. ODYN: An All-Shifted Non-Interior-Point Method for Quadratic Programming in Robotics and AI
+
+- **ID:** `2602.16005` · **primary:** `cs.RO` · **Submitted:** 2026-02-17
+- **Authors:** Jose Rojas, Aristotelis Papatheodorou, Sergi Martinez, et al.
+- **abs:** https://arxiv.org/abs/2602.16005 · **PDF:** https://arxiv.org/pdf/2602.16005v1
+
+## 11622. Imaging-Derived Coronary Fractional Flow Reserve: Advances in Physics-Based, Machine-Learning, and Physics-Informed Methods
+
+- **ID:** `2602.16000` · **primary:** `physics.med-ph` · **Submitted:** 2026-02-17
+- **Authors:** Tanxin Zhu, Emran Hossen, Chen Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.16000 · **PDF:** https://arxiv.org/pdf/2602.16000v1
+
+## 11623. The Geometric Anatomy of Capability Acquisition in Transformers
+
+- **ID:** `2602.15997` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Jayadev Billa
+- **abs:** https://arxiv.org/abs/2602.15997 · **PDF:** https://arxiv.org/pdf/2602.15997v4
+
+## 11624. Exploring New Frontiers in Vertical Federated Learning: the Role of Saddle Point Reformulation
+
+- **ID:** `2602.15996` · **primary:** `math.OC` · **Submitted:** 2026-02-17
+- **Authors:** Aleksandr Beznosikov, Georgiy Kormakov, Alexander Grigorievskiy, et al.
+- **abs:** https://arxiv.org/abs/2602.15996 · **PDF:** https://arxiv.org/pdf/2602.15996v1
+
+## 11625. Verifier-Constrained Flow Expansion for Discovery Beyond the Data
+
+- **ID:** `2602.15984` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Riccardo De Santi, Kimon Protopapas, Ya-Ping Hsieh, et al.
+- **abs:** https://arxiv.org/abs/2602.15984 · **PDF:** https://arxiv.org/pdf/2602.15984v1
+
+## 11626. ReLoop: Structured Modeling and Behavioral Verification for Reliable LLM-Based Optimization
+
+- **ID:** `2602.15983` · **primary:** `cs.SE` · **Submitted:** 2026-02-17
+- **Authors:** Junbo Jacob Lian, Yujun Sun, Huiling Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.15983 · **PDF:** https://arxiv.org/pdf/2602.15983v1
+
+## 11627. Wideband Power Amplifier Behavioral Modeling Using an Amplitude Conditioned LSTM
+
+- **ID:** `2603.00101` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Abdelrahman Abdelsalam, You Fei
+- **abs:** https://arxiv.org/abs/2603.00101 · **PDF:** https://arxiv.org/pdf/2603.00101v1
+
+## 11628. MMCAformer: Macro-Micro Cross-Attention Transformer for Traffic Speed Prediction with Microscopic Connected Vehicle Driving Behavior
+
+- **ID:** `2602.16730` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Lei Han, Mohamed Abdel-Aty, Younggun Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.16730 · **PDF:** https://arxiv.org/pdf/2602.16730v1
+
+## 11629. Fast Online Learning with Gaussian Prior-Driven Hierarchical Unimodal Thompson Sampling
+
+- **ID:** `2602.15972` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Tianchi Zhao, He Liu, Hongyin Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.15972 · **PDF:** https://arxiv.org/pdf/2602.15972v1
+
+## 11630. B-DENSE: Branching For Dense Ensemble Network Supervision Efficiency
+
+- **ID:** `2602.15971` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Cherish Puniani, Tushar Kumar, Arnav Bendre, et al.
+- **abs:** https://arxiv.org/abs/2602.15971 · **PDF:** https://arxiv.org/pdf/2602.15971v2
+
+## 11631. From Reflection to Repair: A Scoping Review of Dataset Documentation Tools
+
+- **ID:** `2602.15968` · **primary:** `cs.SE` · **Submitted:** 2026-02-17
+- **Authors:** Pedro Reynolds-Cuéllar, Marisol Wong-Villacres, Adriana Alvarado Garcia, et al.
+- **abs:** https://arxiv.org/abs/2602.15968 · **PDF:** https://arxiv.org/pdf/2602.15968v1
+
+## 11632. R$^2$Energy: A Large-Scale Benchmark for Robust Renewable Energy Forecasting under Diverse and Extreme Conditions
+
+- **ID:** `2602.15961` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Zhi Sheng, Yuan Yuan, Guozhen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.15961 · **PDF:** https://arxiv.org/pdf/2602.15961v1
+
+## 11633. Deformation-Free Cross-Domain Image Registration via Position-Encoded Temporal Attention
+
+- **ID:** `2602.15959` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Yiwen Wang, Jiahao Qin
+- **abs:** https://arxiv.org/abs/2602.15959 · **PDF:** https://arxiv.org/pdf/2602.15959v2
+
+## 11634. DocSplit: A Comprehensive Benchmark Dataset and Evaluation Approach for Document Packet Recognition and Splitting
+
+- **ID:** `2602.15958` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Md Mofijul Islam, Md Sirajus Salekin, Nivedha Balakrishnan, et al.
+- **abs:** https://arxiv.org/abs/2602.15958 · **PDF:** https://arxiv.org/pdf/2602.15958v1
+
+## 11635. Evolutionary Systems Thinking -- From Equilibrium Models to Open-Ended Adaptive Dynamics
+
+- **ID:** `2602.15957` · **primary:** `q-bio.PE` · **Submitted:** 2026-02-17
+- **Authors:** Dan Adler
+- **abs:** https://arxiv.org/abs/2602.15957 · **PDF:** https://arxiv.org/pdf/2602.15957v1
+
+## 11636. Adaptive Semi-Supervised Training of P300 ERP-BCI Speller System with Minimum Calibration Effort
+
+- **ID:** `2602.15955` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Shumeng Chen, Jane E. Huggins, Tianwen Ma
+- **abs:** https://arxiv.org/abs/2602.15955 · **PDF:** https://arxiv.org/pdf/2602.15955v1
+
+## 11637. Hybrid Model Predictive Control with Physics-Informed Neural Network for Satellite Attitude Control
+
+- **ID:** `2602.15954` · **primary:** `cs.RO` · **Submitted:** 2026-02-17
+- **Authors:** Carlo Cena, Mauro Martini, Marcello Chiaberge
+- **abs:** https://arxiv.org/abs/2602.15954 · **PDF:** https://arxiv.org/pdf/2602.15954v1
+
+## 11638. MadEvolve: Evolutionary Optimization of Cosmological Algorithms with Large Language Models
+
+- **ID:** `2602.15951` · **primary:** `astro-ph.CO` · **Submitted:** 2026-02-17
+- **Authors:** Tianyi Li, Shihui Zang, Moritz Münchmeyer
+- **abs:** https://arxiv.org/abs/2602.15951 · **PDF:** https://arxiv.org/pdf/2602.15951v1
+
+## 11639. Can Vision-Language Models See Squares? Text-Recognition Mediates Spatial Reasoning Across Three Model Families
+
+- **ID:** `2602.15950` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Yuval Levental
+- **abs:** https://arxiv.org/abs/2602.15950 · **PDF:** https://arxiv.org/pdf/2602.15950v2
+
+## 11640. From Tool Orchestration to Code Execution: A Study of MCP Design Choices
+
+- **ID:** `2602.15945` · **primary:** `cs.CR` · **Submitted:** 2026-02-17
+- **Authors:** Yuval Felendler, Parth A. Gandhi, Idan Habler, et al.
+- **abs:** https://arxiv.org/abs/2602.15945 · **PDF:** https://arxiv.org/pdf/2602.15945v1
+
+## 11641. Personalized Group Relative Policy Optimization for Heterogenous Preference Alignment
+
+- **ID:** `2603.10009` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Jialu Wang, Heinrich Peters, Asad A. Butt, et al.
+- **abs:** https://arxiv.org/abs/2603.10009 · **PDF:** https://arxiv.org/pdf/2603.10009v1
+
+## 11642. "Everyone's using it, but no one is allowed to talk about it": College Students' Experiences Navigating the Higher Education Environment in a Generative AI World
+
+- **ID:** `2602.17720` · **primary:** `cs.CY` · **Submitted:** 2026-02-17
+- **Authors:** Yue Fu, Yifan Lin, Yessica Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.17720 · **PDF:** https://arxiv.org/pdf/2602.17720v1
+
+## 11643. Ensemble-size-dependence of deep-learning post-processing methods that minimize an (un)fair score: motivating examples and a proof-of-concept solution
+
+- **ID:** `2602.15830` · **primary:** `physics.ao-ph` · **Submitted:** 2026-02-17
+- **Authors:** Christopher David Roberts
+- **abs:** https://arxiv.org/abs/2602.15830 · **PDF:** https://arxiv.org/pdf/2602.15830v1
+
+## 11644. Operationalising the Superficial Alignment Hypothesis via Task Complexity
+
+- **ID:** `2602.15829` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Tomás Vergara-Browne, Darshan Patil, Ivan Titov, et al.
+- **abs:** https://arxiv.org/abs/2602.15829 · **PDF:** https://arxiv.org/pdf/2602.15829v1
+
+## 11645. Dex4D: Task-Agnostic Point Track Policy for Sim-to-Real Dexterous Manipulation
+
+- **ID:** `2602.15828` · **primary:** `cs.RO` · **Submitted:** 2026-02-17
+- **Authors:** Yuxuan Kuang, Sungjae Park, Katerina Fragkiadaki, et al.
+- **abs:** https://arxiv.org/abs/2602.15828 · **PDF:** https://arxiv.org/pdf/2602.15828v1
+
+## 11646. Perceptive Humanoid Parkour: Chaining Dynamic Human Skills via Motion Matching
+
+- **ID:** `2602.15827` · **primary:** `cs.RO` · **Submitted:** 2026-02-17
+- **Authors:** Zhen Wu, Xiaoyu Huang, Lujie Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.15827 · **PDF:** https://arxiv.org/pdf/2602.15827v1
+
+## 11647. CrispEdit: Low-Curvature Projections for Scalable Non-Destructive LLM Editing
+
+- **ID:** `2602.15823` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Zarif Ikram, Arad Firouzkouhi, Stephen Tu, et al.
+- **abs:** https://arxiv.org/abs/2602.15823 · **PDF:** https://arxiv.org/pdf/2602.15823v1
+
+## 11648. Stabilizing Test-Time Adaptation of High-Dimensional Simulation Surrogates via D-Optimal Statistics
+
+- **ID:** `2602.15820` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Anna Zimmel, Paul Setinek, Gianluca Galletti, et al.
+- **abs:** https://arxiv.org/abs/2602.15820 · **PDF:** https://arxiv.org/pdf/2602.15820v1
+
+## 11649. Solving Parameter-Robust Avoid Problems with Unknown Feasibility using Reinforcement Learning
+
+- **ID:** `2602.15817` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Oswin So, Eric Yang Yu, Songyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.15817 · **PDF:** https://arxiv.org/pdf/2602.15817v1
+
+## 11650. Developing AI Agents with Simulated Data: Why, what, and how?
+
+- **ID:** `2602.15816` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Xiaoran Liu, Istvan David
+- **abs:** https://arxiv.org/abs/2602.15816 · **PDF:** https://arxiv.org/pdf/2602.15816v1
+
+## 11651. Avey-B
+
+- **ID:** `2602.15814` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Devang Acharya, Mohammad Hammoud
+- **abs:** https://arxiv.org/abs/2602.15814 · **PDF:** https://arxiv.org/pdf/2602.15814v1
+
+## 11652. Task-Agnostic Continual Learning for Chest Radiograph Classification
+
+- **ID:** `2602.15811` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Muthu Subash Kavitha, Anas Zafar, Amgad Muneer, et al.
+- **abs:** https://arxiv.org/abs/2602.15811 · **PDF:** https://arxiv.org/pdf/2602.15811v1
+
+## 11653. Decision Quality Evaluation Framework at Pinterest
+
+- **ID:** `2602.15809` · **primary:** `stat.AP` · **Submitted:** 2026-02-17
+- **Authors:** Yuqi Tian, Robert Paine, Attila Dobi, et al.
+- **abs:** https://arxiv.org/abs/2602.15809 · **PDF:** https://arxiv.org/pdf/2602.15809v1
+
+## 11654. The Geometry of Alignment Collapse: When Fine-Tuning Breaks Safety
+
+- **ID:** `2602.15799` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Max Springer, Chung Peng Lee, Blossom Metevier, et al.
+- **abs:** https://arxiv.org/abs/2602.15799 · **PDF:** https://arxiv.org/pdf/2602.15799v1
+
+## 11655. Visual Memory Injection Attacks for Multi-Turn Conversations
+
+- **ID:** `2602.15927` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Christian Schlarmann, Matthias Hein
+- **abs:** https://arxiv.org/abs/2602.15927 · **PDF:** https://arxiv.org/pdf/2602.15927v1
+
+## 11656. Intent Laundering: AI Safety Datasets Are Not What They Seem
+
+- **ID:** `2602.16729` · **primary:** `cs.CR` · **Submitted:** 2026-02-17
+- **Authors:** Shahriar Golchin, Marc Wetter
+- **abs:** https://arxiv.org/abs/2602.16729 · **PDF:** https://arxiv.org/pdf/2602.16729v2
+
+## 11657. Enhancing Building Semantics Preservation in AI Model Training with Large Language Model Encodings
+
+- **ID:** `2602.15791` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Suhyung Jang, Ghang Lee, Jaekun Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.15791 · **PDF:** https://arxiv.org/pdf/2602.15791v1
+
+## 11658. This human study did not involve human subjects: Validating LLM simulations as behavioral evidence
+
+- **ID:** `2602.15785` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Jessica Hullman, David Broska, Huaman Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.15785 · **PDF:** https://arxiv.org/pdf/2602.15785v1
+
+## 11659. Neural Scaling Laws for Boosted Jet Tagging
+
+- **ID:** `2602.15781` · **primary:** `hep-ex` · **Submitted:** 2026-02-17
+- **Authors:** Matthias Vigl, Nicole Hartman, Michael Kagan, et al.
+- **abs:** https://arxiv.org/abs/2602.15781 · **PDF:** https://arxiv.org/pdf/2602.15781v1
+
+## 11660. A Study on Real-time Object Detection using Deep Learning
+
+- **ID:** `2602.15926` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Ankita Bose, Jayasravani Bhumireddy, Naveen N
+- **abs:** https://arxiv.org/abs/2602.15926 · **PDF:** https://arxiv.org/pdf/2602.15926v1
+
+## 11661. *-PLUIE: Personalisable metric with Llm Used for Improved Evaluation
+
+- **ID:** `2602.15778` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Quentin Lemesle, Léane Jourdan, Daisy Munson, et al.
+- **abs:** https://arxiv.org/abs/2602.15778 · **PDF:** https://arxiv.org/pdf/2602.15778v1
+
+## 11662. Robust Stochastic Gradient Posterior Sampling with Lattice Based Discretisation
+
+- **ID:** `2602.15925` · **primary:** `stat.ML` · **Submitted:** 2026-02-17
+- **Authors:** Zier Mensch, Lars Holdijk, Samuel Duffield, et al.
+- **abs:** https://arxiv.org/abs/2602.15925 · **PDF:** https://arxiv.org/pdf/2602.15925v1
+
+## 11663. Benchmarking Zero-Shot Reasoning Approaches for Error Detection in Solidity Smart Contracts
+
+- **ID:** `2603.13239` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Eduardo Sardenberg, Antonio José Grandson Busson, Daniel de Sousa Moraes, et al.
+- **abs:** https://arxiv.org/abs/2603.13239 · **PDF:** https://arxiv.org/pdf/2603.13239v2
+
+## 11664. GlobeDiff: State Diffusion Process for Partial Observability in Multi-Agent Systems
+
+- **ID:** `2602.15776` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Yiqin Yang, Xu Yang, Yuhua Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.15776 · **PDF:** https://arxiv.org/pdf/2602.15776v1
+
+## 11665. Understanding vs. Generation: Navigating Optimization Dilemma in Multimodal Models
+
+- **ID:** `2602.15772` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Sen Ye, Mengde Xu, Shuyang Gu, et al.
+- **abs:** https://arxiv.org/abs/2602.15772 · **PDF:** https://arxiv.org/pdf/2602.15772v2
+
+## 11666. ViTaB-A: Evaluating Multimodal Large Language Models on Visual Table Attribution
+
+- **ID:** `2602.15769` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Yahia Alqurnawi, Preetom Biswas, Anmol Rao, et al.
+- **abs:** https://arxiv.org/abs/2602.15769 · **PDF:** https://arxiv.org/pdf/2602.15769v1
+
+## 11667. Robot-Assisted Social Dining as a White Glove Service
+
+- **ID:** `2602.15767` · **primary:** `cs.RO` · **Submitted:** 2026-02-17
+- **Authors:** Atharva S Kashyap, Ugne Aleksandra Morkute, Patricia Alves-Oliveira
+- **abs:** https://arxiv.org/abs/2602.15767 · **PDF:** https://arxiv.org/pdf/2602.15767v1
+
+## 11668. GLM-5: from Vibe Coding to Agentic Engineering
+
+- **ID:** `2602.15763` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** GLM-5-Team, :, Aohan Zeng, et al.
+- **abs:** https://arxiv.org/abs/2602.15763 · **PDF:** https://arxiv.org/pdf/2602.15763v2
+
+## 11669. ChartEditBench: Evaluating Grounded Multi-Turn Chart Editing in Multimodal Language Models
+
+- **ID:** `2602.15758` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Manav Nitin Kapadnis, Lawanya Baghel, Atharva Naik, et al.
+- **abs:** https://arxiv.org/abs/2602.15758 · **PDF:** https://arxiv.org/pdf/2602.15758v1
+
+## 11670. Beyond Binary Classification: Detecting Fine-Grained Sexism in Social Media Videos
+
+- **ID:** `2602.15757` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Laura De Grazia, Danae Sánchez Villegas, Desmond Elliott, et al.
+- **abs:** https://arxiv.org/abs/2602.15757 · **PDF:** https://arxiv.org/pdf/2602.15757v1
+
+## 11671. A Note on Non-Composability of Layerwise Approximate Verification for Neural Inference
+
+- **ID:** `2602.15756` · **primary:** `cs.CR` · **Submitted:** 2026-02-17
+- **Authors:** Or Zamir
+- **abs:** https://arxiv.org/abs/2602.15756 · **PDF:** https://arxiv.org/pdf/2602.15756v1
+
+## 11672. Under-resourced studies of under-resourced languages: lemmatization and POS-tagging with LLM annotators for historical Armenian, Georgian, Greek and Syriac
+
+- **ID:** `2602.15753` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Chahan Vidal-Gorène, Bastien Kindt, Florian Cafiero
+- **abs:** https://arxiv.org/abs/2602.15753 · **PDF:** https://arxiv.org/pdf/2602.15753v1
+
+## 11673. Beyond Match Maximization and Fairness: Retention-Optimized Two-Sided Matching
+
+- **ID:** `2602.15752` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Ren Kishimoto, Rikiya Takehi, Koichi Tanaka, et al.
+- **abs:** https://arxiv.org/abs/2602.15752 · **PDF:** https://arxiv.org/pdf/2602.15752v1
+
+## 11674. Enabling Low-Latency Machine learning on Radiation-Hard FPGAs with hls4ml
+
+- **ID:** `2602.15751` · **primary:** `hep-ex` · **Submitted:** 2026-02-17
+- **Authors:** Katya Govorkova, Julian Garcia Pardinas, Vladimir Loncar, et al.
+- **abs:** https://arxiv.org/abs/2602.15751 · **PDF:** https://arxiv.org/pdf/2602.15751v1
+
+## 11675. UrbanVerse: Learning Urban Region Representation Across Cities and Tasks
+
+- **ID:** `2602.15750` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Fengze Sun, Egemen Tanin, Shanika Karunasekera, et al.
+- **abs:** https://arxiv.org/abs/2602.15750 · **PDF:** https://arxiv.org/pdf/2602.15750v1
+
+## 11676. MRC-GAT: A Meta-Relational Copula-Based Graph Attention Network for Interpretable Multimodal Alzheimer's Disease Diagnosis
+
+- **ID:** `2602.15740` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Fatemeh Khalvandi, Saadat Izadi, Abdolah Chalechale
+- **abs:** https://arxiv.org/abs/2602.15740 · **PDF:** https://arxiv.org/pdf/2602.15740v1
+
+## 11677. Beyond Labels: Information-Efficient Human-in-the-Loop Learning using Ranking and Selection Queries
+
+- **ID:** `2602.15738` · **primary:** `cs.HC` · **Submitted:** 2026-02-17
+- **Authors:** Belén Martín-Urcelay, Yoonsang Lee, Matthieu R. Bloch, et al.
+- **abs:** https://arxiv.org/abs/2602.15738 · **PDF:** https://arxiv.org/pdf/2602.15738v1
+
+## 11678. MeshMimic: Geometry-Aware Humanoid Motion Learning through 3D Scene Reconstruction
+
+- **ID:** `2602.15733` · **primary:** `cs.RO` · **Submitted:** 2026-02-17
+- **Authors:** Qiang Zhang, Jiahao Ma, Peiran Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.15733 · **PDF:** https://arxiv.org/pdf/2602.15733v1
+
+## 11679. Causal Effect Estimation with Latent Textual Treatments
+
+- **ID:** `2602.15730` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Omri Feldman, Amar Venugopal, Jann Spiess, et al.
+- **abs:** https://arxiv.org/abs/2602.15730 · **PDF:** https://arxiv.org/pdf/2602.15730v1
+
+## 11680. Spanning the Visual Analogy Space with a Weight Basis of LoRAs
+
+- **ID:** `2602.15727` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Hila Manor, Rinon Gal, Haggai Maron, et al.
+- **abs:** https://arxiv.org/abs/2602.15727 · **PDF:** https://arxiv.org/pdf/2602.15727v1
+
+## 11681. Recursive Concept Evolution for Compositional Reasoning in Large Language Models
+
+- **ID:** `2602.15725` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Sarim Chaudhry
+- **abs:** https://arxiv.org/abs/2602.15725 · **PDF:** https://arxiv.org/pdf/2602.15725v1
+
+## 11682. Learning to Retrieve Navigable Candidates for Efficient Vision-and-Language Navigation
+
+- **ID:** `2602.15724` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Shutian Gu, Chengkai Huang, Ruoyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.15724 · **PDF:** https://arxiv.org/pdf/2602.15724v1
+
+## 11683. Uncertainty-Calibrated Spatiotemporal Field Diffusion with Sparse Supervision
+
+- **ID:** `2603.04431` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Kevin Valencia, Xihaier Luo, Shinjae Yoo, et al.
+- **abs:** https://arxiv.org/abs/2603.04431 · **PDF:** https://arxiv.org/pdf/2603.04431v1
+
+## 11684. Lifelong Scalable Multi-Agent Realistic Testbed and A Comprehensive Study on Design Choices in Lifelong AGV Fleet Management Systems
+
+- **ID:** `2602.15721` · **primary:** `cs.RO` · **Submitted:** 2026-02-17
+- **Authors:** Jingtian Yan, Yulun Zhang, Zhenting Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.15721 · **PDF:** https://arxiv.org/pdf/2602.15721v1
+
+## 11685. Using Artificial Neural Networks to Predict Claim Duration in a Work Injury Compensation Environment
+
+- **ID:** `2603.00100` · **primary:** `stat.AP` · **Submitted:** 2026-02-17
+- **Authors:** Anthony Almudevar
+- **abs:** https://arxiv.org/abs/2603.00100 · **PDF:** https://arxiv.org/pdf/2603.00100v1
+
+## 11686. Rethinking Metrics for Lexical Semantic Change Detection
+
+- **ID:** `2602.15716` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Roksana Goworek, Haim Dubossarsky
+- **abs:** https://arxiv.org/abs/2602.15716 · **PDF:** https://arxiv.org/pdf/2602.15716v1
+
+## 11687. Criteria-first, semantics-later: reproducible structure discovery in image-based sciences
+
+- **ID:** `2602.15712` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Jan Bumberger
+- **abs:** https://arxiv.org/abs/2602.15712 · **PDF:** https://arxiv.org/pdf/2602.15712v1
+
+## 11688. Random Wavelet Features for Graph Kernel Machines
+
+- **ID:** `2602.15711` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Valentin de Bassompierre, Jean-Charles Delvenne, Laurent Jacques
+- **abs:** https://arxiv.org/abs/2602.15711 · **PDF:** https://arxiv.org/pdf/2602.15711v1
+
+## 11689. Outer Diversity of Structured Domains
+
+- **ID:** `2602.15708` · **primary:** `cs.GT` · **Submitted:** 2026-02-17
+- **Authors:** Piotr Faliszewski, Krzysztof Sornat, Stanisław Szufa, et al.
+- **abs:** https://arxiv.org/abs/2602.15708 · **PDF:** https://arxiv.org/pdf/2602.15708v1
+
+## 11690. Proactive Conversational Assistant for a Procedural Manual Task based on Audio and IMU
+
+- **ID:** `2602.15707` · **primary:** `cs.MM` · **Submitted:** 2026-02-17
+- **Authors:** Rehana Mahfuz, Yinyi Guo, Erik Visser, et al.
+- **abs:** https://arxiv.org/abs/2602.15707 · **PDF:** https://arxiv.org/pdf/2602.15707v1
+
+## 11691. Controlled oscillation modeling using port-Hamiltonian neural networks
+
+- **ID:** `2602.15704` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Maximino Linares, Guillaume Doras, Thomas Hélie
+- **abs:** https://arxiv.org/abs/2602.15704 · **PDF:** https://arxiv.org/pdf/2602.15704v1
+
+## 11692. How to Disclose? Strategic AI Disclosure in Crowdfunding
+
+- **ID:** `2602.15698` · **primary:** `cs.HC` · **Submitted:** 2026-02-17
+- **Authors:** Ning Wang, Chen Liang
+- **abs:** https://arxiv.org/abs/2602.15698 · **PDF:** https://arxiv.org/pdf/2602.15698v1
+
+## 11693. GATech at AbjadMed: Bidirectional Encoders vs. Causal Decoders: Insights from 82-Class Arabic Medical Classification
+
+- **ID:** `2603.10008` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Ahmed Khaled Khamis
+- **abs:** https://arxiv.org/abs/2603.10008 · **PDF:** https://arxiv.org/pdf/2603.10008v1
+
+## 11694. A fully differentiable framework for training proxy Exchange Correlation Functionals for periodic systems
+
+- **ID:** `2602.15923` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-17
+- **Authors:** Rakshit Kumar Singh, Aryan Amit Barsainyan, Bharath Ramsundar
+- **abs:** https://arxiv.org/abs/2602.15923 · **PDF:** https://arxiv.org/pdf/2602.15923v1
+
+## 11695. A Content-Based Framework for Cybersecurity Refusal Decisions in Large Language Models
+
+- **ID:** `2602.15689` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Noa Linder, Meirav Segal, Omer Antverg, et al.
+- **abs:** https://arxiv.org/abs/2602.15689 · **PDF:** https://arxiv.org/pdf/2602.15689v2
+
+## 11696. Estimating Human Muscular Fatigue in Dynamic Collaborative Robotic Tasks with Learning-Based Models
+
+- **ID:** `2602.15684` · **primary:** `cs.RO` · **Submitted:** 2026-02-17
+- **Authors:** Feras Kiki, Pouya P. Niaz, Alireza Madani, et al.
+- **abs:** https://arxiv.org/abs/2602.15684 · **PDF:** https://arxiv.org/pdf/2602.15684v1
+
+## 11697. GATech at AbjadGenEval Shared Task: Multilingual Embeddings for Arabic Machine-Generated Text Classification
+
+- **ID:** `2603.10007` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Ahmed Khaled Khamis
+- **abs:** https://arxiv.org/abs/2603.10007 · **PDF:** https://arxiv.org/pdf/2603.10007v1
+
+## 11698. Revisiting Northrop Frye's Four Myths Theory with Large Language Models
+
+- **ID:** `2602.15678` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Edirlei Soares de Lima, Marco A. Casanova, Antonio L. Furtado
+- **abs:** https://arxiv.org/abs/2602.15678 · **PDF:** https://arxiv.org/pdf/2602.15678v1
+
+## 11699. CAMEL: An ECG Language Model for Forecasting Cardiac Events
+
+- **ID:** `2602.15677` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Neelay Velingker, Alaia Solko-Breslin, Mayank Keoliya, et al.
+- **abs:** https://arxiv.org/abs/2602.15677 · **PDF:** https://arxiv.org/pdf/2602.15677v3
+
+## 11700. Relative Geometry of Neural Forecasters: Linking Accuracy and Alignment in Learned Latent Geometry
+
+- **ID:** `2602.15676` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Deniz Kucukahmetler, Maximilian Jean Hemmann, Julian Mosig von Aehrenfeld, et al.
+- **abs:** https://arxiv.org/abs/2602.15676 · **PDF:** https://arxiv.org/pdf/2602.15676v1
+
+## 11701. LLM-to-Speech: A Synthetic Data Pipeline for Training Dialectal Text-to-Speech Models
+
+- **ID:** `2602.15675` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Ahmed Khaled Khamis, Hesham Ali
+- **abs:** https://arxiv.org/abs/2602.15675 · **PDF:** https://arxiv.org/pdf/2602.15675v1
+
+## 11702. PERSONA: Dynamic and Compositional Inference-Time Personality Control via Activation Vector Algebra
+
+- **ID:** `2602.15669` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Xiachong Feng, Liang Zhao, Weihong Zhong, et al.
+- **abs:** https://arxiv.org/abs/2602.15669 · **PDF:** https://arxiv.org/pdf/2602.15669v1
+
+## 11703. Mobility-Aware Cache Framework for Scalable LLM-Based Human Mobility Simulation
+
+- **ID:** `2602.16727` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Hua Yan, Heng Tan, Yingxue Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.16727 · **PDF:** https://arxiv.org/pdf/2602.16727v1
+
+## 11704. Bayesian Optimization for Design Parameters of 3D Image Data Analysis
+
+- **ID:** `2602.15660` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** David Exler, Joaquin Eduardo Urrutia Gómez, Martin Krüger, et al.
+- **abs:** https://arxiv.org/abs/2602.15660 · **PDF:** https://arxiv.org/pdf/2602.15660v1
+
+## 11705. Zombie Agents: Persistent Control of Self-Evolving LLM Agents via Self-Reinforcing Injections
+
+- **ID:** `2602.15654` · **primary:** `cs.CR` · **Submitted:** 2026-02-17
+- **Authors:** Xianglin Yang, Yufei He, Shuo Ji, et al.
+- **abs:** https://arxiv.org/abs/2602.15654 · **PDF:** https://arxiv.org/pdf/2602.15654v2
+
+## 11706. Continuous-Time Piecewise-Linear Recurrent Neural Networks
+
+- **ID:** `2602.15649` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Alena Brändle, Lukas Eisenmann, Florian Götz, et al.
+- **abs:** https://arxiv.org/abs/2602.15649 · **PDF:** https://arxiv.org/pdf/2602.15649v1
+
+## 11707. Guided Diffusion by Optimized Loss Functions on Relaxed Parameters for Inverse Material Design
+
+- **ID:** `2602.15648` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Jens U. Kreber, Christian Weißenfels, Joerg Stueckler
+- **abs:** https://arxiv.org/abs/2602.15648 · **PDF:** https://arxiv.org/pdf/2602.15648v1
+
+## 11708. CARE Drive A Framework for Evaluating Reason-Responsiveness of Vision Language Models in Automated Driving
+
+- **ID:** `2602.15645` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Lucas Elbert Suryana, Farah Bierenga, Sanne van Buuren, et al.
+- **abs:** https://arxiv.org/abs/2602.15645 · **PDF:** https://arxiv.org/pdf/2602.15645v1
+
+## 11709. Latency-aware Human-in-the-Loop Reinforcement Learning for Semantic Communications
+
+- **ID:** `2602.15640` · **primary:** `eess.SP` · **Submitted:** 2026-02-17
+- **Authors:** Peizheng Li, Xinyi Lin, Adnan Aijaz
+- **abs:** https://arxiv.org/abs/2602.15640 · **PDF:** https://arxiv.org/pdf/2602.15640v1
+
+## 11710. Flowers: A Warp Drive for Neural PDE Solvers
+
+- **ID:** `2603.04430` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Till Muser, Alexandra Spitzer, Matti Lassas, et al.
+- **abs:** https://arxiv.org/abs/2603.04430 · **PDF:** https://arxiv.org/pdf/2603.04430v1
+
+## 11711. The Stationarity Bias: Stratified Stress-Testing for Time-Series Imputation in Regulated Dynamical Systems
+
+- **ID:** `2602.15637` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Amirreza Dolatpour Fathkouhi, Alireza Namazi, Heman Shakeri
+- **abs:** https://arxiv.org/abs/2602.15637 · **PDF:** https://arxiv.org/pdf/2602.15637v1
+
+## 11712. World Action Models are Zero-shot Policies
+
+- **ID:** `2602.15922` · **primary:** `cs.RO` · **Submitted:** 2026-02-17
+- **Authors:** Seonghyeon Ye, Yunhao Ge, Kaiyuan Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.15922 · **PDF:** https://arxiv.org/pdf/2602.15922v1
+
+## 11713. On inferring cumulative constraints
+
+- **ID:** `2602.15635` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Konstantin Sidorov
+- **abs:** https://arxiv.org/abs/2602.15635 · **PDF:** https://arxiv.org/pdf/2602.15635v1
+
+## 11714. Beyond ReLU: Bifurcation, Oversmoothing, and Topological Priors
+
+- **ID:** `2602.15634` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Erkan Turan, Gaspard Abel, Maysam Behmanesh, et al.
+- **abs:** https://arxiv.org/abs/2602.15634 · **PDF:** https://arxiv.org/pdf/2602.15634v1
+
+## 11715. SEval-NAS: A Search-Agnostic Evaluation for Neural Architecture Search
+
+- **ID:** `2603.00099` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Atah Nuh Mih, Jianzhou Wang, Truong Thanh Hung Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.00099 · **PDF:** https://arxiv.org/pdf/2603.00099v1
+
+## 11716. Neural-POD: A Plug-and-Play Neural Operator Framework for Infinite-Dimensional Functional Nonlinear Proper Orthogonal Decomposition
+
+- **ID:** `2602.15632` · **primary:** `physics.comp-ph` · **Submitted:** 2026-02-17
+- **Authors:** Changhong Mou, Binghang Lu, Guang Lin
+- **abs:** https://arxiv.org/abs/2602.15632 · **PDF:** https://arxiv.org/pdf/2602.15632v2
+
+## 11717. Adaptive Engram Memory System for Indonesian Language Model: Generative AI Based on TOBA LM for Batak and Minang Language
+
+- **ID:** `2603.10006` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Hokky Situngkir, Kevin Siringoringo, Andhika Bernard Lumbantobing
+- **abs:** https://arxiv.org/abs/2603.10006 · **PDF:** https://arxiv.org/pdf/2603.10006v1
+
+## 11718. STAPO: Stabilizing Reinforcement Learning for LLMs by Silencing Rare Spurious Tokens
+
+- **ID:** `2602.15620` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Shiqi Liu, Zeyu He, Guojian Zhan, et al.
+- **abs:** https://arxiv.org/abs/2602.15620 · **PDF:** https://arxiv.org/pdf/2602.15620v3
+
+## 11719. DNN-Enabled Multi-User Beamforming for Throughput Maximization under Adjustable Fairness
+
+- **ID:** `2602.15617` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Kaifeng Lu, Markus Rupp, Stefan Schwarz
+- **abs:** https://arxiv.org/abs/2602.15617 · **PDF:** https://arxiv.org/pdf/2602.15617v1
+
+## 11720. KazakhOCR: A Synthetic Benchmark for Evaluating Multimodal Models in Low-Resource Kazakh Script OCR
+
+- **ID:** `2603.13238` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Henry Gagnier, Sophie Gagnier, Ashwin Kirubakaran
+- **abs:** https://arxiv.org/abs/2603.13238 · **PDF:** https://arxiv.org/pdf/2603.13238v1
+
+## 11721. Symbolic recovery of PDEs from measurement data
+
+- **ID:** `2602.15603` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Erion Morina, Philipp Scholl, Martin Holler
+- **abs:** https://arxiv.org/abs/2602.15603 · **PDF:** https://arxiv.org/pdf/2602.15603v1
+
+## 11722. Certified Per-Instance Unlearning Using Individual Sensitivity Bounds
+
+- **ID:** `2602.15602` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Hanna Benarroch, Jamal Atif, Olivier Cappé
+- **abs:** https://arxiv.org/abs/2602.15602 · **PDF:** https://arxiv.org/pdf/2602.15602v1
+
+## 11723. Khatri-Rao Clustering for Data Summarization
+
+- **ID:** `2603.06602` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Martino Ciaperoni, Collin Leiber, Aristides Gionis, et al.
+- **abs:** https://arxiv.org/abs/2603.06602 · **PDF:** https://arxiv.org/pdf/2603.06602v2
+
+## 11724. The geometry of online conversations and the causal antecedents of conflictual discourse
+
+- **ID:** `2602.15600` · **primary:** `cs.SI` · **Submitted:** 2026-02-17
+- **Authors:** Carlo Santagiustina, Caterina Cruciani
+- **abs:** https://arxiv.org/abs/2602.15600 · **PDF:** https://arxiv.org/pdf/2602.15600v1
+
+## 11725. Multi-Objective Coverage via Constraint Active Search
+
+- **ID:** `2602.15595` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Zakaria Shams Siam, Xuefeng Liu, Chong Liu
+- **abs:** https://arxiv.org/abs/2602.15595 · **PDF:** https://arxiv.org/pdf/2602.15595v1
+
+## 11726. A unified theory of feature learning in RNNs and DNNs
+
+- **ID:** `2602.15593` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Jan P. Bauer, Kirsten Fischer, Moritz Helias, et al.
+- **abs:** https://arxiv.org/abs/2602.15593 · **PDF:** https://arxiv.org/pdf/2602.15593v1
+
+## 11727. What Is Missing: Interpretable Ratings for Large Language Model Outputs
+
+- **ID:** `2603.04429` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Nicholas Stranges, Yimin Yang
+- **abs:** https://arxiv.org/abs/2603.04429 · **PDF:** https://arxiv.org/pdf/2603.04429v1
+
+## 11728. Uni-Flow: a unified autoregressive-diffusion model for complex multiscale flows
+
+- **ID:** `2602.15592` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-02-17
+- **Authors:** Xiao Xue, Tianyue Yang, Mingyang Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.15592 · **PDF:** https://arxiv.org/pdf/2602.15592v1
+
+## 11729. Uniform error bounds for quantized dynamical models
+
+- **ID:** `2602.15586` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Abdelkader Metakalard, Fabien Lauer, Kevin Colin, et al.
+- **abs:** https://arxiv.org/abs/2602.15586 · **PDF:** https://arxiv.org/pdf/2602.15586v1
+
+## 11730. How Vision Becomes Language: A Layer-wise Information-Theoretic Analysis of Multimodal Reasoning
+
+- **ID:** `2602.15580` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Hongxuan Wu, Yukun Zhang, Xueqing Zhou
+- **abs:** https://arxiv.org/abs/2602.15580 · **PDF:** https://arxiv.org/pdf/2602.15580v1
+
+## 11731. Intracoronary Optical Coherence Tomography Image Processing and Vessel Classification Using Machine Learning
+
+- **ID:** `2602.15579` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Amal Lahchim, Lambros Athanasiou
+- **abs:** https://arxiv.org/abs/2602.15579 · **PDF:** https://arxiv.org/pdf/2602.15579v2
+
+## 11732. Clinically Inspired Symptom-Guided Depression Detection from Emotion-Aware Speech Representations
+
+- **ID:** `2602.15578` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Chaithra Nerella, Chiranjeevi Yarra
+- **abs:** https://arxiv.org/abs/2602.15578 · **PDF:** https://arxiv.org/pdf/2602.15578v1
+
+## 11733. Profiling vs. Case-specific Evidence: A Probabilistic Analysis
+
+- **ID:** `2603.00098` · **primary:** `stat.OT` · **Submitted:** 2026-02-17
+- **Authors:** Marcello Di Bello, Nicolò Cangiotti, Michele Loi
+- **abs:** https://arxiv.org/abs/2603.00098 · **PDF:** https://arxiv.org/pdf/2603.00098v1
+
+## 11734. Neural Network-Based Parameter Estimation of a Labour Market Agent-Based Model
+
+- **ID:** `2602.15572` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** M Lopes Alves, Joel Dyer, Doyne Farmer, et al.
+- **abs:** https://arxiv.org/abs/2602.15572 · **PDF:** https://arxiv.org/pdf/2602.15572v2
+
+## 11735. Accelerated Predictive Coding Networks via Direct Kolen-Pollack Feedback Alignment
+
+- **ID:** `2602.15571` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Davide Casnici, Martin Lefebvre, Justin Dauwels, et al.
+- **abs:** https://arxiv.org/abs/2602.15571 · **PDF:** https://arxiv.org/pdf/2602.15571v2
+
+## 11736. Scenario Approach with Post-Design Certification of User-Specified Properties
+
+- **ID:** `2602.15568` · **primary:** `stat.ME` · **Submitted:** 2026-02-17
+- **Authors:** Algo Carè, Marco C. Campi, Simone Garatti
+- **abs:** https://arxiv.org/abs/2602.15568 · **PDF:** https://arxiv.org/pdf/2602.15568v1
+
+## 11737. Beyond Static Pipelines: Learning Dynamic Workflows for Text-to-SQL
+
+- **ID:** `2602.15564` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Yihan Wang, Peiyu Liu, Runyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.15564 · **PDF:** https://arxiv.org/pdf/2602.15564v1
+
+## 11738. 1-Bit Wonder: Improving QAT Performance in the Low-Bit Regime through K-Means Quantization
+
+- **ID:** `2602.15563` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Sohir Maskey, Constantin Eichenberg, Johannes Messner, et al.
+- **abs:** https://arxiv.org/abs/2602.15563 · **PDF:** https://arxiv.org/pdf/2602.15563v1
+
+## 11739. Fixed-Horizon Self-Normalized Inference for Adaptive Experiments via Martingale AIPW/DML with Logged Propensities
+
+- **ID:** `2602.15559` · **primary:** `stat.ME` · **Submitted:** 2026-02-17
+- **Authors:** Gabriel Saco
+- **abs:** https://arxiv.org/abs/2602.15559 · **PDF:** https://arxiv.org/pdf/2602.15559v1
+
+## 11740. RUVA: Personalized Transparent On-Device Graph Reasoning
+
+- **ID:** `2602.15553` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Gabriele Conte, Alessio Mattiace, Gianni Carmosino, et al.
+- **abs:** https://arxiv.org/abs/2602.15553 · **PDF:** https://arxiv.org/pdf/2602.15553v1
+
+## 11741. Latent Regularization in Generative Test Input Generation
+
+- **ID:** `2602.15552` · **primary:** `cs.SE` · **Submitted:** 2026-02-17
+- **Authors:** Giorgi Merabishvili, Oliver Weißl, Andrea Stocco
+- **abs:** https://arxiv.org/abs/2602.15552 · **PDF:** https://arxiv.org/pdf/2602.15552v1
+
+## 11742. VLM-DEWM: Dynamic External World Model for Verifiable and Resilient Vision-Language Planning in Manufacturing
+
+- **ID:** `2602.15549` · **primary:** `cs.RO` · **Submitted:** 2026-02-17
+- **Authors:** Guoqin Tang, Qingxuan Jia, Gang Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.15549 · **PDF:** https://arxiv.org/pdf/2602.15549v1
+
+## 11743. jina-embeddings-v5-text: Task-Targeted Embedding Distillation
+
+- **ID:** `2602.15547` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Mohammad Kalim Akram, Saba Sturua, Nastia Havriushenko, et al.
+- **abs:** https://arxiv.org/abs/2602.15547 · **PDF:** https://arxiv.org/pdf/2602.15547v1
+
+## 11744. CEPAE: Conditional Entropy-Penalized Autoencoders for Time Series Counterfactuals
+
+- **ID:** `2602.15546` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Tomàs Garriga, Gerard Sanz, Eduard Serrahima de Cambra, et al.
+- **abs:** https://arxiv.org/abs/2602.15546 · **PDF:** https://arxiv.org/pdf/2602.15546v1
+
+## 11745. Multi-Modal Multi-Agent Reinforcement Learning for Radiology Report Generation: Radiologist-Like Workflow with Clinically Verifiable Rewards
+
+- **ID:** `2603.16876` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Kaito Baba, Satoshi Kodera
+- **abs:** https://arxiv.org/abs/2603.16876 · **PDF:** https://arxiv.org/pdf/2603.16876v1
+
+## 11746. Perspectives - Interactive Document Clustering in the Discourse Analysis Tool Suite
+
+- **ID:** `2602.15540` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Tim Fischer, Chris Biemann
+- **abs:** https://arxiv.org/abs/2602.15540 · **PDF:** https://arxiv.org/pdf/2602.15540v1
+
+## 11747. Dynamic Training-Free Fusion of Subject and Style LoRAs
+
+- **ID:** `2602.15539` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Qinglong Cao, Yuntian Chen, Chao Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.15539 · **PDF:** https://arxiv.org/pdf/2602.15539v1
+
+## 11748. Functional Central Limit Theorem for Stochastic Gradient Descent
+
+- **ID:** `2602.15538` · **primary:** `stat.ML` · **Submitted:** 2026-02-17
+- **Authors:** Kessang Flamand, Victor-Emmanuel Brunel
+- **abs:** https://arxiv.org/abs/2602.15538 · **PDF:** https://arxiv.org/pdf/2602.15538v1
+
+## 11749. ZeroSyl: Simple Zero-Resource Syllable Tokenization for Spoken Language Modeling
+
+- **ID:** `2602.15537` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Nicol Visser, Simon Malan, Danel Slabbert, et al.
+- **abs:** https://arxiv.org/abs/2602.15537 · **PDF:** https://arxiv.org/pdf/2602.15537v1
+
+## 11750. SuperLocalMemory: Privacy-Preserving Multi-Agent Memory with Bayesian Trust Defense Against Memory Poisoning
+
+- **ID:** `2603.02240` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Varun Pratap Bhardwaj
+- **abs:** https://arxiv.org/abs/2603.02240 · **PDF:** https://arxiv.org/pdf/2603.02240v1
+
+## 11751. Exploring Drug Safety Through Knowledge Graphs: Protein Kinase Inhibitors as a Case Study
+
+- **ID:** `2603.00097` · **primary:** `q-bio.BM` · **Submitted:** 2026-02-17
+- **Authors:** David Jackson, Michael Gertz, Jürgen Hesser
+- **abs:** https://arxiv.org/abs/2603.00097 · **PDF:** https://arxiv.org/pdf/2603.00097v1
+
+## 11752. Quantifying construct validity in large language model evaluations
+
+- **ID:** `2602.15532` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Ryan Othniel Kearns
+- **abs:** https://arxiv.org/abs/2602.15532 · **PDF:** https://arxiv.org/pdf/2602.15532v1
+
+## 11753. EduEVAL-DB: A Role-Based Dataset for Pedagogical Risk Evaluation in Educational Explanations
+
+- **ID:** `2602.15531` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Javier Irigoyen, Roberto Daza, Aythami Morales, et al.
+- **abs:** https://arxiv.org/abs/2602.15531 · **PDF:** https://arxiv.org/pdf/2602.15531v2
+
+## 11754. ExpertWeaver: Unlocking the Inherent MoE in Dense LLMs with GLU Activation Patterns
+
+- **ID:** `2602.15521` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Ziyu Zhao, Tong Zhu, Zhi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.15521 · **PDF:** https://arxiv.org/pdf/2602.15521v1
+
+## 11755. The Obfuscation Atlas: Mapping Where Honesty Emerges in RLVR with Deception Probes
+
+- **ID:** `2602.15515` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Mohammad Taufeeque, Stefan Heimersheim, Adam Gleave, et al.
+- **abs:** https://arxiv.org/abs/2602.15515 · **PDF:** https://arxiv.org/pdf/2602.15515v1
+
+## 11756. DependencyAI: Detecting AI Generated Text through Dependency Parsing
+
+- **ID:** `2602.15514` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Sara Ahmed, Tracy Hammond
+- **abs:** https://arxiv.org/abs/2602.15514 · **PDF:** https://arxiv.org/pdf/2602.15514v1
+
+## 11757. HIMM: Human-Inspired Long-Term Memory Modeling for Embodied Exploration and Question Answering
+
+- **ID:** `2602.15513` · **primary:** `cs.RO` · **Submitted:** 2026-02-17
+- **Authors:** Ji Li, Bo Wang, Jing Xia, et al.
+- **abs:** https://arxiv.org/abs/2602.15513 · **PDF:** https://arxiv.org/pdf/2602.15513v2
+
+## 11758. On the Geometric Coherence of Global Aggregation in Federated Graph Neural Networks
+
+- **ID:** `2602.15510` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Chethana Prasad Kabgere, Shylaja SS
+- **abs:** https://arxiv.org/abs/2602.15510 · **PDF:** https://arxiv.org/pdf/2602.15510v3
+
+## 11759. Fine-Refine: Iterative Fine-grained Refinement for Mitigating Dialogue Hallucination
+
+- **ID:** `2602.15509` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Xiangyan Chen, Yujian Gan, Matthew Purver
+- **abs:** https://arxiv.org/abs/2602.15509 · **PDF:** https://arxiv.org/pdf/2602.15509v1
+
+## 11760. LuxMT Technical Report
+
+- **ID:** `2602.15506` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Nils Rehlinger
+- **abs:** https://arxiv.org/abs/2602.15506 · **PDF:** https://arxiv.org/pdf/2602.15506v1
+
+## 11761. Towards Expectation Detection in Language: A Case Study on Treatment Expectations in Reddit
+
+- **ID:** `2602.15504` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Aswathy Velutharambath, Amelie Wührl
+- **abs:** https://arxiv.org/abs/2602.15504 · **PDF:** https://arxiv.org/pdf/2602.15504v1
+
+## 11762. Approximation Theory for Lipschitz Continuous Transformers
+
+- **ID:** `2602.15503` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Takashi Furuya, Davide Murari, Carola-Bibiane Schönlieb
+- **abs:** https://arxiv.org/abs/2602.15503 · **PDF:** https://arxiv.org/pdf/2602.15503v1
+
+## 11763. ExLipBaB: Exact Lipschitz Constant Computation for Piecewise Linear Neural Networks
+
+- **ID:** `2602.15499` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Tom A. Splittgerber
+- **abs:** https://arxiv.org/abs/2602.15499 · **PDF:** https://arxiv.org/pdf/2602.15499v1
+
+## 11764. PhD Thesis Summary: Methods for Reliability Assessment and Enhancement of Deep Neural Network Hardware Accelerators
+
+- **ID:** `2603.08724` · **primary:** `cs.AR` · **Submitted:** 2026-02-17
+- **Authors:** Mahdi Taheri
+- **abs:** https://arxiv.org/abs/2603.08724 · **PDF:** https://arxiv.org/pdf/2603.08724v1
+
+## 11765. The Equalizer: Introducing Shape-Gain Decomposition in Neural Audio Codecs
+
+- **ID:** `2602.15491` · **primary:** `cs.SD` · **Submitted:** 2026-02-17
+- **Authors:** Samir Sadok, Laurent Girin, Xavier Alameda-Pineda
+- **abs:** https://arxiv.org/abs/2602.15491 · **PDF:** https://arxiv.org/pdf/2602.15491v1
+
+## 11766. RPT-SR: Regional Prior attention Transformer for infrared image Super-Resolution
+
+- **ID:** `2602.15490` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Youngwan Jin, Incheol Park, Yagiz Nalcakan, et al.
+- **abs:** https://arxiv.org/abs/2602.15490 · **PDF:** https://arxiv.org/pdf/2602.15490v1
+
+## 11767. Mitigating Shortcut Learning via Feature Disentanglement in Medical Imaging: A Benchmark Study
+
+- **ID:** `2602.18502` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Sarah Müller, Philipp Berens
+- **abs:** https://arxiv.org/abs/2602.18502 · **PDF:** https://arxiv.org/pdf/2602.18502v1
+
+## 11768. SecCodeBench-V2 Technical Report
+
+- **ID:** `2602.15485` · **primary:** `cs.CR` · **Submitted:** 2026-02-17
+- **Authors:** Longfei Chen, Ji Zhao, Lanxiao Cui, et al.
+- **abs:** https://arxiv.org/abs/2602.15485 · **PDF:** https://arxiv.org/pdf/2602.15485v2
+
+## 11769. Bottleneck Transformer-Based Approach for Improved Automatic STOI Score Prediction
+
+- **ID:** `2602.15484` · **primary:** `eess.AS` · **Submitted:** 2026-02-17
+- **Authors:** Amartyaveer, Murali Kadambi, Chandra Mohan Sharma, et al.
+- **abs:** https://arxiv.org/abs/2602.15484 · **PDF:** https://arxiv.org/pdf/2602.15484v2
+
+## 11770. Adaptive Uncertainty-Guided Surrogates for Efficient phase field Modeling of Dendritic Solidification
+
+- **ID:** `2603.00093` · **primary:** `physics.comp-ph` · **Submitted:** 2026-02-17
+- **Authors:** Eider Garate-Perez, Kerman López de Calle-Etxabe, Oihana Garcia, et al.
+- **abs:** https://arxiv.org/abs/2603.00093 · **PDF:** https://arxiv.org/pdf/2603.00093v1
+
+## 11771. DART: Input-Difficulty-AwaRe Adaptive Threshold for Early-Exit DNNs
+
+- **ID:** `2603.12269` · **primary:** `cs.AR` · **Submitted:** 2026-02-17
+- **Authors:** Parth Patne, Mahdi Taheri, Christian Herglotz, et al.
+- **abs:** https://arxiv.org/abs/2603.12269 · **PDF:** https://arxiv.org/pdf/2603.12269v1
+
+## 11772. LLM-as-Judge on a Budget
+
+- **ID:** `2602.15481` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Aadirupa Saha, Aniket Wagde, Branislav Kveton
+- **abs:** https://arxiv.org/abs/2602.15481 · **PDF:** https://arxiv.org/pdf/2602.15481v1
+
+## 11773. Evaluating Federated Learning for Cross-Country Mood Inference from Smartphone Sensing Data
+
+- **ID:** `2602.15478` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Sharmad Kalpande, Saurabh Shirke, Haroon R. Lone
+- **abs:** https://arxiv.org/abs/2602.15478 · **PDF:** https://arxiv.org/pdf/2602.15478v1
+
+## 11774. POP: Prior-fitted Optimizer Policies
+
+- **ID:** `2602.15473` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Jan Kobiolka, Christian Frey, Gresa Shala, et al.
+- **abs:** https://arxiv.org/abs/2602.15473 · **PDF:** https://arxiv.org/pdf/2602.15473v1
+
+## 11775. Fluids You Can Trust: Property-Preserving Operator Learning for Incompressible Flows
+
+- **ID:** `2602.15472` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-02-17
+- **Authors:** Ramansh Sharma, Matthew Lowery, Houman Owhadi, et al.
+- **abs:** https://arxiv.org/abs/2602.15472 · **PDF:** https://arxiv.org/pdf/2602.15472v3
+
+## 11776. A Dual-Path Generative Framework for Zero-Day Fraud Detection in Banking Systems
+
+- **ID:** `2603.13237` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Nasim Abdirahman Ismail, Enis Karaarslan
+- **abs:** https://arxiv.org/abs/2603.13237 · **PDF:** https://arxiv.org/pdf/2603.13237v1
+
+## 11777. The Skeletal Trap: Mapping Spatial Inequality and Ghost Stops in Ankara's Transit Network
+
+- **ID:** `2602.15470` · **primary:** `physics.soc-ph` · **Submitted:** 2026-02-17
+- **Authors:** Elifnaz Kancan
+- **abs:** https://arxiv.org/abs/2602.15470 · **PDF:** https://arxiv.org/pdf/2602.15470v1
+
+## 11778. On the Out-of-Distribution Generalization of Reasoning in Multimodal LLMs for Simple Visual Planning Tasks
+
+- **ID:** `2602.15460` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Yannic Neuhaus, Nicolas Flammarion, Matthias Hein, et al.
+- **abs:** https://arxiv.org/abs/2602.15460 · **PDF:** https://arxiv.org/pdf/2602.15460v1
+
+## 11779. Benchmarking IoT Time-Series AD with Event-Level Augmentations
+
+- **ID:** `2602.15457` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Dmitry Zhevnenko, Ilya Makarov, Aleksandr Kovalenko, et al.
+- **abs:** https://arxiv.org/abs/2602.15457 · **PDF:** https://arxiv.org/pdf/2602.15457v2
+
+## 11780. In Agents We Trust, but Who Do Agents Trust? Latent Source Preferences Steer LLM Generations
+
+- **ID:** `2602.15456` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Mohammad Aflah Khan, Mahsa Amani, Soumi Das, et al.
+- **abs:** https://arxiv.org/abs/2602.15456 · **PDF:** https://arxiv.org/pdf/2602.15456v1
+
+## 11781. Including Node Textual Metadata in Laplacian-constrained Gaussian Graphical Models
+
+- **ID:** `2602.15920` · **primary:** `stat.ML` · **Submitted:** 2026-02-17
+- **Authors:** Jianhua Wang, Killian Cressant, Pedro Braconnot Velloso, et al.
+- **abs:** https://arxiv.org/abs/2602.15920 · **PDF:** https://arxiv.org/pdf/2602.15920v1
+
+## 11782. Molecular Design beyond Training Data with Novel Extended Objective Functionals of Generative AI Models Driven by Quantum Annealing Computer
+
+- **ID:** `2602.15451` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-17
+- **Authors:** Hayato Kunugi, Mohsen Rahmani, Yosuke Iyama, et al.
+- **abs:** https://arxiv.org/abs/2602.15451 · **PDF:** https://arxiv.org/pdf/2602.15451v1
+
+## 11783. A comparative study of transformer models and recurrent neural networks for path-dependent composite materials
+
+- **ID:** `2603.00092` · **primary:** `physics.comp-ph` · **Submitted:** 2026-02-17
+- **Authors:** Petter Uvdal, Mohsen Mirkhalaf
+- **abs:** https://arxiv.org/abs/2603.00092 · **PDF:** https://arxiv.org/pdf/2603.00092v1
+
+## 11784. TAROT: Test-driven and Capability-adaptive Curriculum Reinforcement Fine-tuning for Code Generation with Large Language Models
+
+- **ID:** `2602.15449` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Chansung Park, Juyong Jiang, Fan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.15449 · **PDF:** https://arxiv.org/pdf/2602.15449v1
+
+## 11785. Algorithmic Approaches to Opinion Selection for Online Deliberation: A Comparative Study
+
+- **ID:** `2602.15439` · **primary:** `cs.CY` · **Submitted:** 2026-02-17
+- **Authors:** Salim Hafid, Manon Berriche, Jean-Philippe Cointet
+- **abs:** https://arxiv.org/abs/2602.15439 · **PDF:** https://arxiv.org/pdf/2602.15439v1
+
+## 11786. Logit Distance Bounds Representational Similarity
+
+- **ID:** `2602.15438` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Beatrix M. G. Nielsen, Emanuele Marconato, Luigi Gresele, et al.
+- **abs:** https://arxiv.org/abs/2602.15438 · **PDF:** https://arxiv.org/pdf/2602.15438v2
+
+## 11787. Measuring Social Integration Through Participation: Categorizing Organizations and Leisure Activities in the Displaced Karelians Interview Archive using LLMs
+
+- **ID:** `2602.15436` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Joonatan Laato, Veera Schroderus, Jenna Kanerva, et al.
+- **abs:** https://arxiv.org/abs/2602.15436 · **PDF:** https://arxiv.org/pdf/2602.15436v1
+
+## 11788. GaiaFlow: Semantic-Guided Diffusion Tuning for Carbon-Frugal Search
+
+- **ID:** `2602.15423` · **primary:** `cs.IR` · **Submitted:** 2026-02-17
+- **Authors:** Rong Fu, Jia Yee Tan, Chunlei Meng, et al.
+- **abs:** https://arxiv.org/abs/2602.15423 · **PDF:** https://arxiv.org/pdf/2602.15423v2
+
+## 11789. Switchable Activation Networks
+
+- **ID:** `2603.06601` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Laha Ale, Ning Zhang, Scott A. King, et al.
+- **abs:** https://arxiv.org/abs/2603.06601 · **PDF:** https://arxiv.org/pdf/2603.06601v1
+
+## 11790. Fairness over Equality: Correcting Social Incentives in Asymmetric Sequential Social Dilemmas
+
+- **ID:** `2602.15407` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Alper Demir, Hüseyin Aydın, Kale-ab Abebe Tessera, et al.
+- **abs:** https://arxiv.org/abs/2602.15407 · **PDF:** https://arxiv.org/pdf/2602.15407v1
+
+## 11791. Joint Enhancement and Classification using Coupled Diffusion Models of Signals and Logits
+
+- **ID:** `2602.15405` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Gilad Nurko, Roi Benita, Yehoshua Dissen, et al.
+- **abs:** https://arxiv.org/abs/2602.15405 · **PDF:** https://arxiv.org/pdf/2602.15405v1
+
+## 11792. Common Belief Revisited
+
+- **ID:** `2602.15403` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Thomas Ågotnes
+- **abs:** https://arxiv.org/abs/2602.15403 · **PDF:** https://arxiv.org/pdf/2602.15403v1
+
+## 11793. SENS-ASR: Semantic Embedding injection in Neural-transducer for Streaming Automatic Speech Recognition
+
+- **ID:** `2603.10005` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Youness Dkhissi, Valentin Vielzeuf, Elys Allesiardo, et al.
+- **abs:** https://arxiv.org/abs/2603.10005 · **PDF:** https://arxiv.org/pdf/2603.10005v2
+
+## 11794. Generalized Leverage Score for Scalable Assessment of Privacy Vulnerability
+
+- **ID:** `2602.15919` · **primary:** `stat.ML` · **Submitted:** 2026-02-17
+- **Authors:** Valentin Dorseuil, Jamal Atif, Olivier Cappé
+- **abs:** https://arxiv.org/abs/2602.15919 · **PDF:** https://arxiv.org/pdf/2602.15919v1
+
+## 11795. ActionCodec: What Makes for Good Action Tokenizers
+
+- **ID:** `2602.15397` · **primary:** `cs.RO` · **Submitted:** 2026-02-17
+- **Authors:** Zibin Dong, Yicheng Liu, Shiduo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.15397 · **PDF:** https://arxiv.org/pdf/2602.15397v1
+
+## 11796. Doubly Stochastic Mean-Shift Clustering
+
+- **ID:** `2602.15393` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Tom Trigano, Yann Sepulcre, Itshak Lapidot
+- **abs:** https://arxiv.org/abs/2602.15393 · **PDF:** https://arxiv.org/pdf/2602.15393v1
+
+## 11797. Improving LLM Reliability through Hybrid Abstention and Adaptive Detection
+
+- **ID:** `2602.15391` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Ankit Sharma, Nachiket Tapas, Jyotiprakash Patra
+- **abs:** https://arxiv.org/abs/2602.15391 · **PDF:** https://arxiv.org/pdf/2602.15391v1
+
+## 11798. From Chain-Ladder to Individual Claims Reserving
+
+- **ID:** `2602.15385` · **primary:** `stat.AP` · **Submitted:** 2026-02-17
+- **Authors:** Ronald Richman, Mario V. Wüthrich
+- **abs:** https://arxiv.org/abs/2602.15385 · **PDF:** https://arxiv.org/pdf/2602.15385v2
+
+## 11799. World-Model-Augmented Web Agents with Action Correction
+
+- **ID:** `2602.15384` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Zhouzhou Shen, Xueyu Hu, Xiyun Li, et al.
+- **abs:** https://arxiv.org/abs/2602.15384 · **PDF:** https://arxiv.org/pdf/2602.15384v1
+
+## 11800. The Vision Wormhole: Latent-Space Communication in Heterogeneous Multi-Agent Systems
+
+- **ID:** `2602.15382` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Xiaoze Liu, Ruowang Zhang, Weichen Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.15382 · **PDF:** https://arxiv.org/pdf/2602.15382v1
+
+## 11801. Fractional-Order Federated Learning
+
+- **ID:** `2602.15380` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Mohammad Partohaghighi, Roummel Marcia, YangQuan Chen
+- **abs:** https://arxiv.org/abs/2602.15380 · **PDF:** https://arxiv.org/pdf/2602.15380v1
+
+## 11802. FlashMem: Supporting Modern DNN Workloads on Mobile with GPU Memory Hierarchy Optimizations
+
+- **ID:** `2602.15379` · **primary:** `cs.DC` · **Submitted:** 2026-02-17
+- **Authors:** Zhihao Shu, Md Musfiqur Rahman Sanim, Hangyu Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.15379 · **PDF:** https://arxiv.org/pdf/2602.15379v1
+
+## 11803. Making Large Language Models Speak Tulu: Structured Prompting for an Extremely Low-Resource Language
+
+- **ID:** `2602.15378` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Prathamesh Devadiga, Paras Chopra
+- **abs:** https://arxiv.org/abs/2602.15378 · **PDF:** https://arxiv.org/pdf/2602.15378v1
+
+## 11804. Orchestration-Free Customer Service Automation: A Privacy-Preserving and Flowchart-Guided Framework
+
+- **ID:** `2602.15377` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Mengze Hong, Chen Jason Zhang, Zichang Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.15377 · **PDF:** https://arxiv.org/pdf/2602.15377v1
+
+## 11805. A Unified Evaluation of Learning-Based Similarity Techniques for Malware Detection
+
+- **ID:** `2602.15376` · **primary:** `cs.CR` · **Submitted:** 2026-02-17
+- **Authors:** Udbhav Prasad, Aniesh Chawla
+- **abs:** https://arxiv.org/abs/2602.15376 · **PDF:** https://arxiv.org/pdf/2602.15376v1
+
+## 11806. FuzzingRL: Reinforcement Fuzz-Testing for Revealing VLM Failures
+
+- **ID:** `2603.06600` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Jiajun Xu, Jiageng Mao, Ang Qi, et al.
+- **abs:** https://arxiv.org/abs/2603.06600 · **PDF:** https://arxiv.org/pdf/2603.06600v1
+
+## 11807. EarthSpatialBench: Benchmarking Spatial Reasoning Capabilities of Multimodal LLMs on Earth Imagery
+
+- **ID:** `2602.15918` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Zelin Xu, Yupu Zhang, Saugat Adhikari, et al.
+- **abs:** https://arxiv.org/abs/2602.15918 · **PDF:** https://arxiv.org/pdf/2602.15918v1
+
+## 11808. Far Out: Evaluating Language Models on Slang in Australian and Indian English
+
+- **ID:** `2602.15373` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Deniz Kaya Dilsiz, Dipankar Srirag, Aditya Joshi
+- **abs:** https://arxiv.org/abs/2602.15373 · **PDF:** https://arxiv.org/pdf/2602.15373v2
+
+## 11809. Agent Memory Below the Prompt: Persistent Q4 KV Cache for Multi-Agent LLM Inference on Edge Devices
+
+- **ID:** `2603.04428` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Yakov Pyotr Shkolnikov
+- **abs:** https://arxiv.org/abs/2603.04428 · **PDF:** https://arxiv.org/pdf/2603.04428v1
+
+## 11810. GMAIL: Generative Modality Alignment for generated Image Learning
+
+- **ID:** `2602.15368` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Shentong Mo, Sukmin Yun
+- **abs:** https://arxiv.org/abs/2602.15368 · **PDF:** https://arxiv.org/pdf/2602.15368v1
+
+## 11811. Alignment as Iatrogenesis: Pastoral Power, Collective Pathology, and the Structural Limits of Monolingual Safety Evaluation
+
+- **ID:** `2603.08723` · **primary:** `cs.CY` · **Submitted:** 2026-02-17
+- **Authors:** Hiroki Fukui
+- **abs:** https://arxiv.org/abs/2603.08723 · **PDF:** https://arxiv.org/pdf/2603.08723v2
+
+## 11812. CDRL: A Reinforcement Learning Framework Inspired by Cerebellar Circuits and Dendritic Computational Strategies
+
+- **ID:** `2602.15367` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Sibo Zhang, Rui Jing, Liangfu Lv, et al.
+- **abs:** https://arxiv.org/abs/2602.15367 · **PDF:** https://arxiv.org/pdf/2602.15367v1
+
+## 11813. Automated Multi-Source Debugging and Natural Language Error Explanation for Dashboard Applications
+
+- **ID:** `2602.15362` · **primary:** `cs.SE` · **Submitted:** 2026-02-17
+- **Authors:** Devendra Tata, Mona Rajhans
+- **abs:** https://arxiv.org/abs/2602.15362 · **PDF:** https://arxiv.org/pdf/2602.15362v1
+
+## 11814. Nonparametric Identification and Inference for Counterfactual Distributions with Confounding
+
+- **ID:** `2602.15916` · **primary:** `stat.ME` · **Submitted:** 2026-02-17
+- **Authors:** Jianle Sun, Kun Zhang
+- **abs:** https://arxiv.org/abs/2602.15916 · **PDF:** https://arxiv.org/pdf/2602.15916v1
+
+## 11815. NeuroSymActive: Differentiable Neural-Symbolic Reasoning with Active Exploration for Knowledge Graph Question Answering
+
+- **ID:** `2602.15353` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Rong Fu, Yang Li, Zeyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.15353 · **PDF:** https://arxiv.org/pdf/2602.15353v1
+
+## 11816. MaS-VQA: A Mask-and-Select Framework for Knowledge-Based Visual Question Answering
+
+- **ID:** `2602.15915` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Xianwei Mao, Kai Ye, Sheng Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.15915 · **PDF:** https://arxiv.org/pdf/2602.15915v1
+
+## 11817. Fine-Tuning LLMs to Generate Economical and Reliable Actions for the Power Grid
+
+- **ID:** `2602.15350` · **primary:** `eess.SY` · **Submitted:** 2026-02-17
+- **Authors:** Mohamad Chehade, Hao Zhu
+- **abs:** https://arxiv.org/abs/2602.15350 · **PDF:** https://arxiv.org/pdf/2602.15350v1
+
+## 11818. ER-MIA: Black-Box Adversarial Memory Injection Attacks on Long-Term Memory-Augmented Large Language Models
+
+- **ID:** `2602.15344` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Mitchell Piehl, Zhaohan Xi, Zuobin Xiong, et al.
+- **abs:** https://arxiv.org/abs/2602.15344 · **PDF:** https://arxiv.org/pdf/2602.15344v1
+
+## 11819. Benchmarking Self-Supervised Models for Cardiac Ultrasound View Classification
+
+- **ID:** `2602.15339` · **primary:** `eess.IV` · **Submitted:** 2026-02-17
+- **Authors:** Youssef Megahed, Salma I. Megahed, Robin Ducharme, et al.
+- **abs:** https://arxiv.org/abs/2602.15339 · **PDF:** https://arxiv.org/pdf/2602.15339v1
+
+## 11820. Discovering Implicit Large Language Model Alignment Objectives
+
+- **ID:** `2602.15338` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Edward Chen, Sanmi Koyejo, Carlos Guestrin
+- **abs:** https://arxiv.org/abs/2602.15338 · **PDF:** https://arxiv.org/pdf/2602.15338v1
+
+## 11821. FedPSA: Modeling Behavioral Staleness in Asynchronous Federated Learning
+
+- **ID:** `2602.15337` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Chaoyi Lu, Yiding Sun, Zhichuan Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.15337 · **PDF:** https://arxiv.org/pdf/2602.15337v2
+
+## 11822. Directional Reasoning Trajectory Change (DRTC): Identifying Critical Trace Segments in Reasoning Models
+
+- **ID:** `2602.15332` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Waldemar Chang
+- **abs:** https://arxiv.org/abs/2602.15332 · **PDF:** https://arxiv.org/pdf/2602.15332v2
+
+## 11823. A Scalable Curiosity-Driven Game-Theoretic Framework for Long-Tail Multi-Label Learning in Data Mining
+
+- **ID:** `2602.15330` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Jing Yang, Keze Wang
+- **abs:** https://arxiv.org/abs/2602.15330 · **PDF:** https://arxiv.org/pdf/2602.15330v1
+
+## 11824. Prescriptive Scaling Reveals the Evolution of Language Model Capabilities
+
+- **ID:** `2602.15327` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Hanlin Zhang, Jikai Jin, Vasilis Syrgkanis, et al.
+- **abs:** https://arxiv.org/abs/2602.15327 · **PDF:** https://arxiv.org/pdf/2602.15327v1
+
+## 11825. SCENE OTA-FD: Self-Centering Noncoherent Estimator for Over-the-Air Federated Distillation
+
+- **ID:** `2602.15326` · **primary:** `eess.SP` · **Submitted:** 2026-02-17
+- **Authors:** Hao Chen, Zavareh Bozorgasl
+- **abs:** https://arxiv.org/abs/2602.15326 · **PDF:** https://arxiv.org/pdf/2602.15326v1
+
+## 11826. AgriWorld:A World Tools Protocol Framework for Verifiable Agricultural Reasoning with Code-Executing LLM Agents
+
+- **ID:** `2602.15325` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Zhixing Zhang, Jesen Zhang, Hao Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.15325 · **PDF:** https://arxiv.org/pdf/2602.15325v1
+
+## 11827. Unforgeable Watermarks for Language Models via Robust Signatures
+
+- **ID:** `2602.15323` · **primary:** `cs.CR` · **Submitted:** 2026-02-17
+- **Authors:** Huijia Lin, Kameron Shahabi, Min Jae Song
+- **abs:** https://arxiv.org/abs/2602.15323 · **PDF:** https://arxiv.org/pdf/2602.15323v1
+
+## 11828. Steering Dynamical Regimes of Diffusion Models by Breaking Detailed Balance
+
+- **ID:** `2602.15914` · **primary:** `cond-mat.stat-mech` · **Submitted:** 2026-02-17
+- **Authors:** Haiqi Lu, Ying Tang
+- **abs:** https://arxiv.org/abs/2602.15914 · **PDF:** https://arxiv.org/pdf/2602.15914v1
+
+## 11829. On Surprising Effectiveness of Masking Updates in Adaptive Optimizers
+
+- **ID:** `2602.15322` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Taejong Joo, Wenhan Xia, Cheolmin Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.15322 · **PDF:** https://arxiv.org/pdf/2602.15322v1
+
+## 11830. Sparrow: Text-Anchored Window Attention with Visual-Semantic Glimpsing for Speculative Decoding in Video LLMs
+
+- **ID:** `2602.15318` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Libo Zhang, Zhaoning Zhang, Wangyang Hong, et al.
+- **abs:** https://arxiv.org/abs/2602.15318 · **PDF:** https://arxiv.org/pdf/2602.15318v1
+
+## 11831. Human Attribution of Causality to AI Across Agency, Misuse, and Misalignment
+
+- **ID:** `2603.13236` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Maria Victoria Carro, David Lagnado
+- **abs:** https://arxiv.org/abs/2603.13236 · **PDF:** https://arxiv.org/pdf/2603.13236v1
+
+## 11832. Training-Free Zero-Shot Anomaly Detection in 3D Brain MRI with 2D Foundation Models
+
+- **ID:** `2602.15315` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Tai Le-Gia, Jaehyun Ahn
+- **abs:** https://arxiv.org/abs/2602.15315 · **PDF:** https://arxiv.org/pdf/2602.15315v1
+
+## 11833. Mnemis: Dual-Route Retrieval on Hierarchical Graphs for Long-Term LLM Memory
+
+- **ID:** `2602.15313` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Zihao Tang, Xin Yu, Ziyu Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.15313 · **PDF:** https://arxiv.org/pdf/2602.15313v1
+
+## 11834. Extracting Consumer Insight from Text: A Large Language Model Approach to Emotion and Evaluation Measurement
+
+- **ID:** `2602.15312` · **primary:** `cs.CL` · **Submitted:** 2026-02-17
+- **Authors:** Stephan Ludwig, Peter J. Danaher, Xiaohao Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.15312 · **PDF:** https://arxiv.org/pdf/2602.15312v1
+
+## 11835. Sparse Additive Model Pruning for Order-Based Causal Structure Learning
+
+- **ID:** `2602.15306` · **primary:** `stat.ML` · **Submitted:** 2026-02-17
+- **Authors:** Kentaro Kanamori, Hirofumi Suzuki, Takuya Takagi
+- **abs:** https://arxiv.org/abs/2602.15306 · **PDF:** https://arxiv.org/pdf/2602.15306v1
+
+## 11836. Hybrid Federated and Split Learning for Privacy Preserving Clinical Prediction and Treatment Optimization
+
+- **ID:** `2602.15304` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Farzana Akter, Rakib Hossain, Deb Kanna Roy Toushi, et al.
+- **abs:** https://arxiv.org/abs/2602.15304 · **PDF:** https://arxiv.org/pdf/2602.15304v1
+
+## 11837. X-MAP: eXplainable Misclassification Analysis and Profiling for Spam and Phishing Detection
+
+- **ID:** `2602.15298` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Qi Zhang, Dian Chen, Lance M. Kaplan, et al.
+- **abs:** https://arxiv.org/abs/2602.15298 · **PDF:** https://arxiv.org/pdf/2602.15298v1
+
+## 11838. Foundation Models for Medical Imaging: Status, Challenges, and Directions
+
+- **ID:** `2602.15913` · **primary:** `eess.IV` · **Submitted:** 2026-02-17
+- **Authors:** Chuang Niu, Pengwei Wu, Bruno De Man, et al.
+- **abs:** https://arxiv.org/abs/2602.15913 · **PDF:** https://arxiv.org/pdf/2602.15913v1
+
+## 11839. EAA: Automating materials characterization with vision language model agents
+
+- **ID:** `2602.15294` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Ming Du, Yanqi Luo, Srutarshi Banerjee, et al.
+- **abs:** https://arxiv.org/abs/2602.15294 · **PDF:** https://arxiv.org/pdf/2602.15294v1
+
+## 11840. The Information Geometry of Softmax: Probing and Steering
+
+- **ID:** `2602.15293` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Kiho Park, Todd Nief, Yo Joong Choe, et al.
+- **abs:** https://arxiv.org/abs/2602.15293 · **PDF:** https://arxiv.org/pdf/2602.15293v1
+
+## 11841. AI-Paging: Lease-Based Execution Anchoring for Network-Exposed AI-as-a-Service
+
+- **ID:** `2602.15286` · **primary:** `cs.NI` · **Submitted:** 2026-02-17
+- **Authors:** Mohaned Chraiti, Merve Saimler
+- **abs:** https://arxiv.org/abs/2602.15286 · **PDF:** https://arxiv.org/pdf/2602.15286v2
+
+## 11842. Complex-Valued Unitary Representations as Classification Heads for Improved Uncertainty Quantification in Deep Neural Networks
+
+- **ID:** `2602.15283` · **primary:** `cs.LG` · **Submitted:** 2026-02-17
+- **Authors:** Akbar Anbar Jafari, Cagri Ozcinar, Gholamreza Anbarjafari
+- **abs:** https://arxiv.org/abs/2602.15283 · **PDF:** https://arxiv.org/pdf/2602.15283v1
+
+## 11843. High-Fidelity Network Management for Federated AI-as-a-Service: Cross-Domain Orchestration
+
+- **ID:** `2602.15281` · **primary:** `cs.NI` · **Submitted:** 2026-02-17
+- **Authors:** Mohaned Chraiti, Ozgur Ercetin, Merve Saimler
+- **abs:** https://arxiv.org/abs/2602.15281 · **PDF:** https://arxiv.org/pdf/2602.15281v2
+
+## 11844. Visual Persuasion: What Influences Decisions of Vision-Language Models?
+
+- **ID:** `2602.15278` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Manuel Cherep, Pranav M R, Pattie Maes, et al.
+- **abs:** https://arxiv.org/abs/2602.15278 · **PDF:** https://arxiv.org/pdf/2602.15278v1
+
+## 11845. Accelerating Large-Scale Dataset Distillation via Exploration-Exploitation Optimization
+
+- **ID:** `2602.15277` · **primary:** `cs.CV` · **Submitted:** 2026-02-17
+- **Authors:** Muhammad J. Alahmadi, Peng Gao, Feiyi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.15277 · **PDF:** https://arxiv.org/pdf/2602.15277v2
+
+## 11846. When Remembering and Planning are Worth it: Navigating under Change
+
+- **ID:** `2602.15274` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Omid Madani, J. Brian Burns, Reza Eghbali, et al.
+- **abs:** https://arxiv.org/abs/2602.15274 · **PDF:** https://arxiv.org/pdf/2602.15274v1
+
+## 11847. FrameRef: A Framing Dataset and Simulation Testbed for Modeling Bounded Rational Information Health
+
+- **ID:** `2602.15273` · **primary:** `cs.CY` · **Submitted:** 2026-02-17
+- **Authors:** Victor De Lima, Jiqun Liu, Grace Hui Yang
+- **abs:** https://arxiv.org/abs/2602.15273 · **PDF:** https://arxiv.org/pdf/2602.15273v1
+
+## 11848. Enhancing Diversity and Feasibility: Joint Population Synthesis from Multi-source Data Using Generative Models
+
+- **ID:** `2602.15270` · **primary:** `cs.AI` · **Submitted:** 2026-02-17
+- **Authors:** Farbod Abbasi, Zachary Patterson, Bilal Farooq
+- **abs:** https://arxiv.org/abs/2602.15270 · **PDF:** https://arxiv.org/pdf/2602.15270v1
+
+## 11849. From Diagnosis to Inoculation: Building Cognitive Resistance to AI Disempowerment
+
+- **ID:** `2602.15265` · **primary:** `cs.HC` · **Submitted:** 2026-02-16
+- **Authors:** Aleksey Komissarov
+- **abs:** https://arxiv.org/abs/2602.15265 · **PDF:** https://arxiv.org/pdf/2602.15265v1
+
+## 11850. Thin Keys, Full Values: Reducing KV Cache via Low-Dimensional Attention Selection
+
+- **ID:** `2603.04427` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Hengshuai Yao, Xing Chen, Ahmed Murtadha, et al.
+- **abs:** https://arxiv.org/abs/2603.04427 · **PDF:** https://arxiv.org/pdf/2603.04427v4
+
+## 11851. Delta-Crosscoder: Robust Crosscoder Model Diffing in Narrow Fine-Tuning Regimes
+
+- **ID:** `2603.04426` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Aly Kassem, Thomas Jiralerspong, Negar Rostamzadeh, et al.
+- **abs:** https://arxiv.org/abs/2603.04426 · **PDF:** https://arxiv.org/pdf/2603.04426v1
+
+## 11852. Fast and Effective On-policy Distillation from Reasoning Prefixes
+
+- **ID:** `2602.15260` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Dongxu Zhang, Zhichao Yang, Sepehr Janghorbani, et al.
+- **abs:** https://arxiv.org/abs/2602.15260 · **PDF:** https://arxiv.org/pdf/2602.15260v1
+
+## 11853. Knowing Isn't Understanding: Re-grounding Generative Proactivity with Epistemic and Behavioral Insight
+
+- **ID:** `2602.15259` · **primary:** `cs.CY` · **Submitted:** 2026-02-16
+- **Authors:** Kirandeep Kaur, Xingda Lyu, Chirag Shah
+- **abs:** https://arxiv.org/abs/2602.15259 · **PDF:** https://arxiv.org/pdf/2602.15259v1
+
+## 11854. How to Train Your Long-Context Visual Document Model
+
+- **ID:** `2602.15257` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Austin Veselka
+- **abs:** https://arxiv.org/abs/2602.15257 · **PDF:** https://arxiv.org/pdf/2602.15257v2
+
+## 11855. Scaling Laws for Masked-Reconstruction Transformers on Single-Cell Transcriptomics
+
+- **ID:** `2602.15253` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Ihor Kendiukhov
+- **abs:** https://arxiv.org/abs/2602.15253 · **PDF:** https://arxiv.org/pdf/2602.15253v1
+
+## 11856. Decision Making under Imperfect Recall: Algorithms and Benchmarks
+
+- **ID:** `2602.15252` · **primary:** `cs.GT` · **Submitted:** 2026-02-16
+- **Authors:** Emanuel Tewolde, Brian Hu Zhang, Ioannis Anagnostides, et al.
+- **abs:** https://arxiv.org/abs/2602.15252 · **PDF:** https://arxiv.org/pdf/2602.15252v1
+
+## 11857. Artificial Intelligence Specialization in the European Union: Underexplored Role of the Periphery at NUTS-3 Level
+
+- **ID:** `2602.15249` · **primary:** `cs.DL` · **Submitted:** 2026-02-16
+- **Authors:** Victor Herrero-Solana
+- **abs:** https://arxiv.org/abs/2602.15249 · **PDF:** https://arxiv.org/pdf/2602.15249v1
+
+## 11858. Predicting Invoice Dilution in Supply Chain Finance with Leakage Free Two Stage XGBoost, KAN (Kolmogorov Arnold Networks), and Ensemble Models
+
+- **ID:** `2602.15248` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Pavel Koptev, Vishnu Kumar, Konstantin Malkov, et al.
+- **abs:** https://arxiv.org/abs/2602.15248 · **PDF:** https://arxiv.org/pdf/2602.15248v1
+
+## 11859. MyoInteract: A Framework for Fast Prototyping of Biomechanical HCI Tasks using Reinforcement Learning
+
+- **ID:** `2602.15245` · **primary:** `cs.HC` · **Submitted:** 2026-02-16
+- **Authors:** Ankit Bhattarai, Hannah Selder, Florian Fischer, et al.
+- **abs:** https://arxiv.org/abs/2602.15245 · **PDF:** https://arxiv.org/pdf/2602.15245v1
+
+## 11860. GenAI for Systems: Recurring Challenges and Design Principles from Software to Silicon
+
+- **ID:** `2602.15241` · **primary:** `cs.SE` · **Submitted:** 2026-02-16
+- **Authors:** Arya Tschand, Chenyu Wang, Zishen Wan, et al.
+- **abs:** https://arxiv.org/abs/2602.15241 · **PDF:** https://arxiv.org/pdf/2602.15241v1
+
+## 11861. Fine-Tune, Don't Prompt, Your Language Model to Identify Biased Language in Clinical Notes
+
+- **ID:** `2603.10004` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Isotta Landi, Eugenia Alleva, Nicole Bussola, et al.
+- **abs:** https://arxiv.org/abs/2603.10004 · **PDF:** https://arxiv.org/pdf/2603.10004v1
+
+## 11862. Size Transferability of Graph Transformers with Convolutional Positional Encodings
+
+- **ID:** `2602.15239` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Javier Porras-Valenzuela, Zhiyang Wang, Alejandro Ribeiro
+- **abs:** https://arxiv.org/abs/2602.15239 · **PDF:** https://arxiv.org/pdf/2602.15239v1
+
+## 11863. Closing the Distribution Gap in Adversarial Training for LLMs
+
+- **ID:** `2602.15238` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Chengzhi Hu, Jonas Dornbusch, David Lüdke, et al.
+- **abs:** https://arxiv.org/abs/2602.15238 · **PDF:** https://arxiv.org/pdf/2602.15238v2
+
+## 11864. BindCLIP: A Unified Contrastive-Generative Representation Learning Framework for Virtual Screening
+
+- **ID:** `2602.15236` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Anjie Qiao, Zhen Wang, Yaliang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.15236 · **PDF:** https://arxiv.org/pdf/2602.15236v1
+
+## 11865. Data-Driven Optimization of Multi-Generational Cellular Networks: A Performance Classification Framework for Strategic Infrastructure Management
+
+- **ID:** `2603.04425` · **primary:** `cs.NI` · **Submitted:** 2026-02-16
+- **Authors:** Maryam Sabahat, M. Umar Khan
+- **abs:** https://arxiv.org/abs/2603.04425 · **PDF:** https://arxiv.org/pdf/2603.04425v1
+
+## 11866. tensorFM: Low-Rank Approximations of Cross-Order Feature Interactions
+
+- **ID:** `2602.15229` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Alessio Mazzetto, Mohammad Mahdi Khalili, Laura Fee Nern, et al.
+- **abs:** https://arxiv.org/abs/2602.15229 · **PDF:** https://arxiv.org/pdf/2602.15229v1
+
+## 11867. Automatically Finding Reward Model Biases
+
+- **ID:** `2602.15222` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Atticus Wang, Iván Arcuschin, Arthur Conmy
+- **abs:** https://arxiv.org/abs/2602.15222 · **PDF:** https://arxiv.org/pdf/2602.15222v1
+
+## 11868. Secure and Energy-Efficient Wireless Agentic AI Networks
+
+- **ID:** `2602.15212` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Yuanyan Song, Kezhi Wang, Xinmian Xu
+- **abs:** https://arxiv.org/abs/2602.15212 · **PDF:** https://arxiv.org/pdf/2602.15212v1
+
+## 11869. ÜberWeb: Insights from Multilingual Curation for a 20-Trillion-Token Dataset
+
+- **ID:** `2602.15210` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** DatologyAI, :, Aldo Gael Carranza, et al.
+- **abs:** https://arxiv.org/abs/2602.15210 · **PDF:** https://arxiv.org/pdf/2602.15210v3
+
+## 11870. MAVRL: Learning Reward Functions from Multiple Feedback Types with Amortized Variational Inference
+
+- **ID:** `2602.15206` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Raphaël Baur, Yannick Metz, Maria Gkoulta, et al.
+- **abs:** https://arxiv.org/abs/2602.15206 · **PDF:** https://arxiv.org/pdf/2602.15206v1
+
+## 11871. Is Mamba Reliable for Medical Imaging?
+
+- **ID:** `2602.16723` · **primary:** `cs.CR` · **Submitted:** 2026-02-16
+- **Authors:** Banafsheh Saber Latibari, Najmeh Nazari, Daniel Brignac, et al.
+- **abs:** https://arxiv.org/abs/2602.16723 · **PDF:** https://arxiv.org/pdf/2602.16723v1
+
+## 11872. Tomography by Design: An Algebraic Approach to Low-Rank Quantum States
+
+- **ID:** `2602.15202` · **primary:** `quant-ph` · **Submitted:** 2026-02-16
+- **Authors:** Shakir Showkat Sofi, Charlotte Vermeylen, Lieven De Lathauwer
+- **abs:** https://arxiv.org/abs/2602.15202 · **PDF:** https://arxiv.org/pdf/2602.15202v1
+
+## 11873. COMPOT: Calibration-Optimized Matrix Procrustes Orthogonalization for Transformers Compression
+
+- **ID:** `2602.15200` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Denis Makhov, Dmitriy Shopkhoev, Magauiya Zhussip, et al.
+- **abs:** https://arxiv.org/abs/2602.15200 · **PDF:** https://arxiv.org/pdf/2602.15200v1
+
+## 11874. Colosseum: Auditing Collusion in Cooperative Multi-Agent Systems
+
+- **ID:** `2602.15198` · **primary:** `cs.MA` · **Submitted:** 2026-02-16
+- **Authors:** Mason Nakamura, Abhinav Kumar, Saswat Das, et al.
+- **abs:** https://arxiv.org/abs/2602.15198 · **PDF:** https://arxiv.org/pdf/2602.15198v1
+
+## 11875. OpaqueToolsBench: Learning Nuances of Tool Behavior Through Interaction
+
+- **ID:** `2602.15197` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Skyler Hallinan, Thejas Venkatesh, Xiang Ren, et al.
+- **abs:** https://arxiv.org/abs/2602.15197 · **PDF:** https://arxiv.org/pdf/2602.15197v1
+
+## 11876. Weight space Detection of Backdoors in LoRA Adapters
+
+- **ID:** `2602.15195` · **primary:** `cs.CR` · **Submitted:** 2026-02-16
+- **Authors:** David Puertolas Merenciano, Ekaterina Vasyagina, Raghav Dixit, et al.
+- **abs:** https://arxiv.org/abs/2602.15195 · **PDF:** https://arxiv.org/pdf/2602.15195v2
+
+## 11877. AIC CTU@AVerImaTeC: dual-retriever RAG for image-text fact checking
+
+- **ID:** `2602.15190` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Herbert Ullrich, Jan Drchal
+- **abs:** https://arxiv.org/abs/2602.15190 · **PDF:** https://arxiv.org/pdf/2602.15190v1
+
+## 11878. ScrapeGraphAI-100k: A Large-Scale Dataset for LLM-Based Web Information Extraction
+
+- **ID:** `2602.15189` · **primary:** `cs.IR` · **Submitted:** 2026-02-16
+- **Authors:** William Brach, Francesco Zuppichini, Marco Vinciguerra, et al.
+- **abs:** https://arxiv.org/abs/2602.15189 · **PDF:** https://arxiv.org/pdf/2602.15189v1
+
+## 11879. Learning Data-Efficient and Generalizable Neural Operators via Fundamental Physics Knowledge
+
+- **ID:** `2602.15184` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Siying Ma, Mehrdad M. Zadeh, Mauricio Soroco, et al.
+- **abs:** https://arxiv.org/abs/2602.15184 · **PDF:** https://arxiv.org/pdf/2602.15184v1
+
+## 11880. Seeing to Generalize: How Visual Data Corrects Binding Shortcuts
+
+- **ID:** `2602.15183` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Nicolas Buzeta, Felipe del Rio, Cristian Hinostroza, et al.
+- **abs:** https://arxiv.org/abs/2602.15183 · **PDF:** https://arxiv.org/pdf/2602.15183v1
+
+## 11881. Time-Archival Camera Virtualization for Sports and Visual Performances
+
+- **ID:** `2602.15181` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Yunxiao Zhang, William Stone, Suryansh Kumar
+- **abs:** https://arxiv.org/abs/2602.15181 · **PDF:** https://arxiv.org/pdf/2602.15181v1
+
+## 11882. Mind the (DH) Gap! A Contrast in Risky Choices Between Reasoning and Conversational LLMs
+
+- **ID:** `2602.15173` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Luise Ge, Yongyan Zhang, Yevgeniy Vorobeychik
+- **abs:** https://arxiv.org/abs/2602.15173 · **PDF:** https://arxiv.org/pdf/2602.15173v1
+
+## 11883. Learning the S-matrix from data: Rediscovering gravity from gauge theory via symbolic regression
+
+- **ID:** `2602.15169` · **primary:** `hep-th` · **Submitted:** 2026-02-16
+- **Authors:** Nathan Moynihan
+- **abs:** https://arxiv.org/abs/2602.15169 · **PDF:** https://arxiv.org/pdf/2602.15169v1
+
+## 11884. Distributional Deep Learning for Super-Resolution of 4D Flow MRI under Domain Shift
+
+- **ID:** `2602.15167` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Xiaoyi Wen, Fei Jiang
+- **abs:** https://arxiv.org/abs/2602.15167 · **PDF:** https://arxiv.org/pdf/2602.15167v1
+
+## 11885. Exploiting Layer-Specific Vulnerabilities to Backdoor Attack in Federated Learning
+
+- **ID:** `2602.15161` · **primary:** `cs.CR` · **Submitted:** 2026-02-16
+- **Authors:** Mohammad Hadi Foroughi, Seyed Hamed Rastegar, Mohammad Sabokrou, et al.
+- **abs:** https://arxiv.org/abs/2602.15161 · **PDF:** https://arxiv.org/pdf/2602.15161v1
+
+## 11886. Learning Representations from Incomplete EHR Data with Dual-Masked Autoencoding
+
+- **ID:** `2602.15159` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Xiao Xiang, David Restrepo, Hyewon Jeong, et al.
+- **abs:** https://arxiv.org/abs/2602.15159 · **PDF:** https://arxiv.org/pdf/2602.15159v1
+
+## 11887. da Costa and Tarski meet Goguen and Carnap: a novel approach for ontological heterogeneity based on consequence systems
+
+- **ID:** `2602.15158` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Gabriel Rocha
+- **abs:** https://arxiv.org/abs/2602.15158 · **PDF:** https://arxiv.org/pdf/2602.15158v1
+
+## 11888. Panini: Continual Learning in Token Space via Structured Memory
+
+- **ID:** `2602.15156` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Shreyas Rajesh, Pavan Holur, Mehmet Yigit Turali, et al.
+- **abs:** https://arxiv.org/abs/2602.15156 · **PDF:** https://arxiv.org/pdf/2602.15156v1
+
+## 11889. Refine Now, Query Fast: A Decoupled Refinement Paradigm for Implicit Neural Fields
+
+- **ID:** `2602.15155` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Tianyu Xiong, Skylar Wurster, Han-Wei Shen
+- **abs:** https://arxiv.org/abs/2602.15155 · **PDF:** https://arxiv.org/pdf/2602.15155v3
+
+## 11890. Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
+
+- **ID:** `2602.15154` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Praditha Alwis, Soumyadeep Chandra, Deepak Ravikumar, et al.
+- **abs:** https://arxiv.org/abs/2602.15154 · **PDF:** https://arxiv.org/pdf/2602.15154v1
+
+## 11891. Beyond Reinforcement Learning: Fast and Scalable Quantum Circuit Synthesis
+
+- **ID:** `2602.15146` · **primary:** `quant-ph` · **Submitted:** 2026-02-16
+- **Authors:** Lukas Theißinger, Thore Gerlach, David Berghaus, et al.
+- **abs:** https://arxiv.org/abs/2602.15146 · **PDF:** https://arxiv.org/pdf/2602.15146v2
+
+## 11892. Protecting Language Models Against Unauthorized Distillation through Trace Rewriting
+
+- **ID:** `2602.15143` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Xinhang Ma, William Yeoh, Ning Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.15143 · **PDF:** https://arxiv.org/pdf/2602.15143v1
+
+## 11893. CGRA-DeBERTa Concept Guided Residual Augmentation Transformer for Theologically Islamic Understanding
+
+- **ID:** `2602.15139` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Tahir Hussain, Saddam Hussain Khan
+- **abs:** https://arxiv.org/abs/2602.15139 · **PDF:** https://arxiv.org/pdf/2602.15139v1
+
+## 11894. MB-DSMIL-CL-PL: Scalable Weakly Supervised Ovarian Cancer Subtype Classification and Localisation Using Contrastive and Prototype Learning with Frozen Patch Features
+
+- **ID:** `2602.15138` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Marcus Jenkins, Jasenka Mazibrada, Bogdan Leahu, et al.
+- **abs:** https://arxiv.org/abs/2602.15138 · **PDF:** https://arxiv.org/pdf/2602.15138v1
+
+## 11895. Universal priors: solving empirical Bayes via Bayesian inference and pretraining
+
+- **ID:** `2602.15136` · **primary:** `stat.ML` · **Submitted:** 2026-02-16
+- **Authors:** Nick Cannella, Anzo Teh, Yanjun Han, et al.
+- **abs:** https://arxiv.org/abs/2602.15136 · **PDF:** https://arxiv.org/pdf/2602.15136v1
+
+## 11896. PolyNODE: Variable-dimension Neural ODEs on M-polyfolds
+
+- **ID:** `2602.15128` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Per Åhag, Alexander Friedrich, Fredrik Ohlsson, et al.
+- **abs:** https://arxiv.org/abs/2602.15128 · **PDF:** https://arxiv.org/pdf/2602.15128v1
+
+## 11897. Probing the Limits of the Lie Detector Approach to LLM Deception
+
+- **ID:** `2603.10003` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Tom-Felix Berger
+- **abs:** https://arxiv.org/abs/2603.10003 · **PDF:** https://arxiv.org/pdf/2603.10003v1
+
+## 11898. ResearchGym: Evaluating Language Model Agents on Real-World AI Research
+
+- **ID:** `2602.15112` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Aniketh Garikaparthi, Manasi Patwardhan, Arman Cohan
+- **abs:** https://arxiv.org/abs/2602.15112 · **PDF:** https://arxiv.org/pdf/2602.15112v2
+
+## 11899. Symmetry in language statistics shapes the geometry of model representations
+
+- **ID:** `2602.15029` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Dhruva Karkada, Daniel J. Korchinski, Andres Nava, et al.
+- **abs:** https://arxiv.org/abs/2602.15029 · **PDF:** https://arxiv.org/pdf/2602.15029v2
+
+## 11900. Long Context, Less Focus: A Scaling Gap in LLMs Revealed through Privacy and Personalization
+
+- **ID:** `2602.15028` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Shangding Gu
+- **abs:** https://arxiv.org/abs/2602.15028 · **PDF:** https://arxiv.org/pdf/2602.15028v1
+
+## 11901. Rethinking Diffusion Models with Symmetries through Canonicalization with Applications to Molecular Graph Generation
+
+- **ID:** `2602.15022` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Cai Zhou, Zijie Chen, Zian Li, et al.
+- **abs:** https://arxiv.org/abs/2602.15022 · **PDF:** https://arxiv.org/pdf/2602.15022v1
+
+## 11902. Generalization from Low- to Moderate-Resolution Spectra with Neural Networks for Stellar Parameter Estimation: A Case Study with DESI
+
+- **ID:** `2602.15021` · **primary:** `astro-ph.SR` · **Submitted:** 2026-02-16
+- **Authors:** Xiaosheng Zhao, Yuan-Sen Ting, Rosemary F. G. Wyse, et al.
+- **abs:** https://arxiv.org/abs/2602.15021 · **PDF:** https://arxiv.org/pdf/2602.15021v1
+
+## 11903. Hunt Globally: Wide Search AI Agents for Drug Asset Scouting in Investing, Business Development, and Competitive Intelligence
+
+- **ID:** `2602.15019` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Alisa Vinogradova, Vlad Vinogradov, Luba Greenwood, et al.
+- **abs:** https://arxiv.org/abs/2602.15019 · **PDF:** https://arxiv.org/pdf/2602.15019v2
+
+## 11904. Scaling Beyond Masked Diffusion Language Models
+
+- **ID:** `2602.15014` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Subham Sekhar Sahoo, Jean-Marie Lemercier, Zhihan Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.15014 · **PDF:** https://arxiv.org/pdf/2602.15014v1
+
+## 11905. Text Style Transfer with Parameter-efficient LLM Finetuning and Round-trip Translation
+
+- **ID:** `2602.15013` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Ruoxi Liu, Philipp Koehn
+- **abs:** https://arxiv.org/abs/2602.15013 · **PDF:** https://arxiv.org/pdf/2602.15013v1
+
+## 11906. Cold-Start Personalization via Training-Free Priors from Structured World Models
+
+- **ID:** `2602.15012` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Avinandan Bose, Shuyue Stella Li, Faeze Brahman, et al.
+- **abs:** https://arxiv.org/abs/2602.15012 · **PDF:** https://arxiv.org/pdf/2602.15012v1
+
+## 11907. BPP: Long-Context Robot Imitation Learning by Focusing on Key History Frames
+
+- **ID:** `2602.15010` · **primary:** `cs.RO` · **Submitted:** 2026-02-16
+- **Authors:** Max Sobol Mark, Jacky Liang, Maria Attarian, et al.
+- **abs:** https://arxiv.org/abs/2602.15010 · **PDF:** https://arxiv.org/pdf/2602.15010v2
+
+## 11908. Efficient Sampling with Discrete Diffusion Models: Sharp and Adaptive Guarantees
+
+- **ID:** `2602.15008` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Daniil Dmitriev, Zhihan Huang, Yuting Wei
+- **abs:** https://arxiv.org/abs/2602.15008 · **PDF:** https://arxiv.org/pdf/2602.15008v1
+
+## 11909. Distributed Quantum Gaussian Processes for Multi-Agent Systems
+
+- **ID:** `2602.15006` · **primary:** `cs.MA` · **Submitted:** 2026-02-16
+- **Authors:** Meet Gandhi, George P. Kontoudis
+- **abs:** https://arxiv.org/abs/2602.15006 · **PDF:** https://arxiv.org/pdf/2602.15006v1
+
+## 11910. Learning User Interests via Reasoning and Distillation for Cross-Domain News Recommendation
+
+- **ID:** `2602.15005` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Mengdan Zhu, Yufan Zhao, Tao Di, et al.
+- **abs:** https://arxiv.org/abs/2602.15005 · **PDF:** https://arxiv.org/pdf/2602.15005v1
+
+## 11911. PDE foundation models are skillful AI weather emulators for the Martian atmosphere
+
+- **ID:** `2602.15004` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Johannes Schmude, Sujit Roy, Liping Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.15004 · **PDF:** https://arxiv.org/pdf/2602.15004v1
+
+## 11912. On propagation of chaos for the Fisher-Rao gradient flow in entropic mean-field optimization
+
+- **ID:** `2602.15094` · **primary:** `math.OC` · **Submitted:** 2026-02-16
+- **Authors:** Petra Lazić, Linshan Liu, Mateusz B. Majka
+- **abs:** https://arxiv.org/abs/2602.15094 · **PDF:** https://arxiv.org/pdf/2602.15094v1
+
+## 11913. Boundary Point Jailbreaking of Black-Box LLMs
+
+- **ID:** `2602.15001` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Xander Davies, Giorgi Giglemiani, Edmund Lau, et al.
+- **abs:** https://arxiv.org/abs/2602.15001 · **PDF:** https://arxiv.org/pdf/2602.15001v2
+
+## 11914. Spectral Convolution on Orbifolds for Geometric Deep Learning
+
+- **ID:** `2602.14997` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Tim Mangliers, Bernhard Mössner, Benjamin Himpel
+- **abs:** https://arxiv.org/abs/2602.14997 · **PDF:** https://arxiv.org/pdf/2602.14997v2
+
+## 11915. On the Semantics of Primary Cause in Hybrid Dynamic Domains
+
+- **ID:** `2602.14994` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Shakil M. Khan, Asim Mehmood, Sandra Zilles
+- **abs:** https://arxiv.org/abs/2602.14994 · **PDF:** https://arxiv.org/pdf/2602.14994v1
+
+## 11916. ThermEval: A Structured Benchmark for Evaluation of Vision-Language Models on Thermal Imagery
+
+- **ID:** `2602.14989` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Ayush Shrivastava, Kirtan Gangani, Laksh Jain, et al.
+- **abs:** https://arxiv.org/abs/2602.14989 · **PDF:** https://arxiv.org/pdf/2602.14989v1
+
+## 11917. Orthogonalized Multimodal Contrastive Learning with Asymmetric Masking for Structured Representations
+
+- **ID:** `2602.14983` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Carolin Cissee, Raneen Younis, Zahra Ahmadi
+- **abs:** https://arxiv.org/abs/2602.14983 · **PDF:** https://arxiv.org/pdf/2602.14983v1
+
+## 11918. Block Empirical Likelihood Inference for Longitudinal Generalized Partially Linear Single-Index Models
+
+- **ID:** `2602.14981` · **primary:** `stat.ME` · **Submitted:** 2026-02-16
+- **Authors:** Tianni Zhang, Yuyao Wang, Yu Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.14981 · **PDF:** https://arxiv.org/pdf/2602.14981v2
+
+## 11919. MacroGuide: Topological Guidance for Macrocycle Generation
+
+- **ID:** `2602.14977` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Alicja Maksymiuk, Alexandre Duplessis, Michael Bronstein, et al.
+- **abs:** https://arxiv.org/abs/2602.14977 · **PDF:** https://arxiv.org/pdf/2602.14977v1
+
+## 11920. Faster Molecular Dynamics with Neural Network Potentials via Distilled Multiple Time-Stepping and Non-Conservative Forces
+
+- **ID:** `2602.14975` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-16
+- **Authors:** Nicolaï Gouraud, Côme Cattin, Thomas Plé, et al.
+- **abs:** https://arxiv.org/abs/2602.14975 · **PDF:** https://arxiv.org/pdf/2602.14975v2
+
+## 11921. Use What You Know: Causal Foundation Models with Partial Graphs
+
+- **ID:** `2602.14972` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Arik Reuter, Anish Dhir, Cristiana Diaconu, et al.
+- **abs:** https://arxiv.org/abs/2602.14972 · **PDF:** https://arxiv.org/pdf/2602.14972v1
+
+## 11922. Counterfactual Fairness Evaluation of LLM-Based Contact Center Agent Quality Assurance System
+
+- **ID:** `2602.14970` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Kawin Mayilvaghanan, Siddhant Gupta, Ayush Kumar
+- **abs:** https://arxiv.org/abs/2602.14970 · **PDF:** https://arxiv.org/pdf/2602.14970v1
+
+## 11923. PhyScensis: Physics-Augmented LLM Agents for Complex Physical Scene Arrangement
+
+- **ID:** `2602.14968` · **primary:** `cs.RO` · **Submitted:** 2026-02-16
+- **Authors:** Yian Wang, Han Yang, Minghao Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.14968 · **PDF:** https://arxiv.org/pdf/2602.14968v1
+
+## 11924. Tool-Aware Planning in Contact Center AI: Evaluating LLMs through Lineage-Guided Query Decomposition
+
+- **ID:** `2602.14955` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Varun Nathan, Shreyas Guha, Ayush Kumar
+- **abs:** https://arxiv.org/abs/2602.14955 · **PDF:** https://arxiv.org/pdf/2602.14955v1
+
+## 11925. Locally Adaptive Multi-Objective Learning
+
+- **ID:** `2602.14952` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Jivat Neet Kaur, Isaac Gibbs, Michael I. Jordan
+- **abs:** https://arxiv.org/abs/2602.14952 · **PDF:** https://arxiv.org/pdf/2602.14952v1
+
+## 11926. Gradient Networks for Universal Magnetic Modeling of Synchronous Machines
+
+- **ID:** `2602.14947` · **primary:** `eess.SY` · **Submitted:** 2026-02-16
+- **Authors:** Junyi Li, Tim Foissner, Floran Martin, et al.
+- **abs:** https://arxiv.org/abs/2602.14947 · **PDF:** https://arxiv.org/pdf/2602.14947v1
+
+## 11927. Mixture-of-Experts under Finite-Rate Gating: Communication--Generalization Trade-offs
+
+- **ID:** `2602.15091` · **primary:** `stat.ML` · **Submitted:** 2026-02-16
+- **Authors:** Ali Khalesi, Mohammad Reza Deylam Salehi
+- **abs:** https://arxiv.org/abs/2602.15091 · **PDF:** https://arxiv.org/pdf/2602.15091v2
+
+## 11928. AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
+
+- **ID:** `2602.14941` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Zun Wang, Han Lin, Jaehong Yoon, et al.
+- **abs:** https://arxiv.org/abs/2602.14941 · **PDF:** https://arxiv.org/pdf/2602.14941v1
+
+## 11929. Fault Detection in Electrical Distribution System using Autoencoders
+
+- **ID:** `2602.14939` · **primary:** `eess.SY` · **Submitted:** 2026-02-16
+- **Authors:** Sidharthenee Nayak, Victor Sam Moses Babu, Chandrashekhar Narayan Bhende, et al.
+- **abs:** https://arxiv.org/abs/2602.14939 · **PDF:** https://arxiv.org/pdf/2602.14939v1
+
+## 11930. Variance-Reduced $(\varepsilon,δ)-$Unlearning using Forget Set Gradients
+
+- **ID:** `2602.14938` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Martin Van Waerebeke, Marco Lorenzi, Kevin Scaman, et al.
+- **abs:** https://arxiv.org/abs/2602.14938 · **PDF:** https://arxiv.org/pdf/2602.14938v1
+
+## 11931. Activation-Space Uncertainty Quantification for Pretrained Networks
+
+- **ID:** `2602.14934` · **primary:** `stat.ML` · **Submitted:** 2026-02-16
+- **Authors:** Richard Bergna, Stefan Depeweg, Sergio Calvo-Ordoñez, et al.
+- **abs:** https://arxiv.org/abs/2602.14934 · **PDF:** https://arxiv.org/pdf/2602.14934v2
+
+## 11932. From Classical to Quantum: Extending Prometheus for Unsupervised Discovery of Phase Transitions in Three Dimensions and Quantum Systems
+
+- **ID:** `2602.14928` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-02-16
+- **Authors:** Brandon Yee, Wilson Collins, Maximilian Rutkowski
+- **abs:** https://arxiv.org/abs/2602.14928 · **PDF:** https://arxiv.org/pdf/2602.14928v3
+
+## 11933. MAC-AMP: A Closed-Loop Multi-Agent Collaboration System for Multi-Objective Antimicrobial Peptide Design
+
+- **ID:** `2602.14926` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Gen Zhou, Sugitha Janarthanan, Lianghong Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.14926 · **PDF:** https://arxiv.org/pdf/2602.14926v1
+
+## 11934. ReusStdFlow: A Standardized Reusability Framework for Dynamic Workflow Construction in Agentic AI
+
+- **ID:** `2602.14922` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Gaoyang Zhang, Shanghong Zou, Yafang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.14922 · **PDF:** https://arxiv.org/pdf/2602.14922v1
+
+## 11935. BHyGNN+: Unsupervised Representation Learning for Heterophilic Hypergraphs
+
+- **ID:** `2602.14919` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Tianyi Ma, Yiyue Qian, Zehong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.14919 · **PDF:** https://arxiv.org/pdf/2602.14919v1
+
+## 11936. BFS-PO: Best-First Search for Large Reasoning Models
+
+- **ID:** `2602.14917` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Fiorenzo Parascandolo, Wenhui Tan, Enver Sangineto, et al.
+- **abs:** https://arxiv.org/abs/2602.14917 · **PDF:** https://arxiv.org/pdf/2602.14917v1
+
+## 11937. Additive Control Variates Dominate Self-Normalisation in Off-Policy Evaluation
+
+- **ID:** `2602.14914` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Olivier Jeunen, Shashank Gupta
+- **abs:** https://arxiv.org/abs/2602.14914 · **PDF:** https://arxiv.org/pdf/2602.14914v1
+
+## 11938. Coverage Guarantees for Pseudo-Calibrated Conformal Prediction under Distribution Shift
+
+- **ID:** `2602.14913` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Farbod Siahkali, Ashwin Verma, Vijay Gupta
+- **abs:** https://arxiv.org/abs/2602.14913 · **PDF:** https://arxiv.org/pdf/2602.14913v1
+
+## 11939. Position: Introspective Experience from Conversational Environments as a Path to Better Learning
+
+- **ID:** `2602.14910` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Claudiu Cristian Musat, Jackson Tolins, Diego Antognini, et al.
+- **abs:** https://arxiv.org/abs/2602.14910 · **PDF:** https://arxiv.org/pdf/2602.14910v1
+
+## 11940. Adjoint-based shape optimization of a ship hull using a Conditional Variational Autoencoder (CVAE) assisted propulsion surrogate model
+
+- **ID:** `2602.14907` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-02-16
+- **Authors:** Moloud Arian Maram, Georgios Bletsos, Thanh Tung Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.14907 · **PDF:** https://arxiv.org/pdf/2602.14907v2
+
+## 11941. The Potential of CoT for Reasoning: A Closer Look at Trace Dynamics
+
+- **ID:** `2602.14903` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Gregor Bachmann, Yichen Jiang, Seyed Mohsen Moosavi Dezfooli, et al.
+- **abs:** https://arxiv.org/abs/2602.14903 · **PDF:** https://arxiv.org/pdf/2602.14903v2
+
+## 11942. Picking the Right Specialist: Attentive Neural Process-based Selection of Task-Specialized Models as Tools for Agentic Healthcare Systems
+
+- **ID:** `2602.14901` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Pramit Saha, Joshua Strong, Mohammad Alsharid, et al.
+- **abs:** https://arxiv.org/abs/2602.14901 · **PDF:** https://arxiv.org/pdf/2602.14901v1
+
+## 11943. Algorithmic Simplification of Neural Networks with Mosaic-of-Motifs
+
+- **ID:** `2602.14896` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Pedram Bakhtiarifard, Tong Chen, Jonathan Wenshøj, et al.
+- **abs:** https://arxiv.org/abs/2602.14896 · **PDF:** https://arxiv.org/pdf/2602.14896v1
+
+## 11944. Lifted Relational Probabilistic Inference via Implicit Learning
+
+- **ID:** `2602.14890` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Luise Ge, Brendan Juba, Kris Nilsson, et al.
+- **abs:** https://arxiv.org/abs/2602.14890 · **PDF:** https://arxiv.org/pdf/2602.14890v1
+
+## 11945. Web-Scale Multimodal Summarization using CLIP-Based Semantic Alignment
+
+- **ID:** `2602.14889` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Mounvik K, N Harshit
+- **abs:** https://arxiv.org/abs/2602.14889 · **PDF:** https://arxiv.org/pdf/2602.14889v1
+
+## 11946. When Scaling Fails: Network and Fabric Effects on Distributed GPU Training Performance
+
+- **ID:** `2603.04424` · **primary:** `cs.NI` · **Submitted:** 2026-02-16
+- **Authors:** Dinesh Gopalan, Ratul Ali
+- **abs:** https://arxiv.org/abs/2603.04424 · **PDF:** https://arxiv.org/pdf/2603.04424v1
+
+## 11947. Drift-Diffusion Matching: Embedding dynamics in latent manifolds of asymmetric neural networks
+
+- **ID:** `2602.14885` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-02-16
+- **Authors:** Ramón Nartallo-Kaluarachchi, Renaud Lambiotte, Alain Goriely
+- **abs:** https://arxiv.org/abs/2602.14885 · **PDF:** https://arxiv.org/pdf/2602.14885v1
+
+## 11948. Numerical exploration of the range of shape functionals using neural networks
+
+- **ID:** `2602.14881` · **primary:** `math.OC` · **Submitted:** 2026-02-16
+- **Authors:** Eloi Martinet, Ilias Ftouhi
+- **abs:** https://arxiv.org/abs/2602.14881 · **PDF:** https://arxiv.org/pdf/2602.14881v1
+
+## 11949. CT-Bench: A Benchmark for Multimodal Lesion Understanding in Computed Tomography
+
+- **ID:** `2602.14879` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Qingqing Zhu, Qiao Jin, Tejas S. Mathai, et al.
+- **abs:** https://arxiv.org/abs/2602.14879 · **PDF:** https://arxiv.org/pdf/2602.14879v2
+
+## 11950. On the Learning Dynamics of RLVR at the Edge of Competence
+
+- **ID:** `2602.14872` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Yu Huang, Zixin Wen, Yuejie Chi, et al.
+- **abs:** https://arxiv.org/abs/2602.14872 · **PDF:** https://arxiv.org/pdf/2602.14872v1
+
+## 11951. Concept Influence: Leveraging Interpretability to Improve Performance and Efficiency in Training Data Attribution
+
+- **ID:** `2602.14869` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Matthew Kowal, Goncalo Paulo, Louis Jaburi, et al.
+- **abs:** https://arxiv.org/abs/2602.14869 · **PDF:** https://arxiv.org/pdf/2602.14869v1
+
+## 11952. Goldilocks RL: Tuning Task Difficulty to Escape Sparse Rewards for Reasoning
+
+- **ID:** `2602.14868` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Ilia Mahrooghi, Aryo Lotfi, Emmanuel Abbe
+- **abs:** https://arxiv.org/abs/2602.14868 · **PDF:** https://arxiv.org/pdf/2602.14868v1
+
+## 11953. Fast and accurate quasi-atom method for simultaneous atomistic and continuum simulation of solids
+
+- **ID:** `2602.14867` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-16
+- **Authors:** Artem Chuprov, Egor E. Nuzhin, Alexey A. Tsukanov, et al.
+- **abs:** https://arxiv.org/abs/2602.14867 · **PDF:** https://arxiv.org/pdf/2602.14867v1
+
+## 11954. EmbeWebAgent: Embedding Web Agents into Any Customized UI
+
+- **ID:** `2602.14865` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Chenyang Ma, Clyde Fare, Matthew Wilson, et al.
+- **abs:** https://arxiv.org/abs/2602.14865 · **PDF:** https://arxiv.org/pdf/2602.14865v1
+
+## 11955. The Well-Tempered Classifier: Some Elementary Properties of Temperature Scaling
+
+- **ID:** `2602.14862` · **primary:** `stat.ML` · **Submitted:** 2026-02-16
+- **Authors:** Pierre-Alexandre Mattei, Bruno Loureiro
+- **abs:** https://arxiv.org/abs/2602.14862 · **PDF:** https://arxiv.org/pdf/2602.14862v1
+
+## 11956. World Models for Policy Refinement in StarCraft II
+
+- **ID:** `2602.14857` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Yixin Zhang, Ziyi Wang, Yiming Rong, et al.
+- **abs:** https://arxiv.org/abs/2602.14857 · **PDF:** https://arxiv.org/pdf/2602.14857v1
+
+## 11957. A Pragmatic Method for Comparing Clusterings with Overlaps and Outliers
+
+- **ID:** `2602.14855` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Ryan DeWolfe, Paweł Prałat, François Théberge
+- **abs:** https://arxiv.org/abs/2602.14855 · **PDF:** https://arxiv.org/pdf/2602.14855v2
+
+## 11958. BEACONS: Bounded-Error, Algebraically-Composable Neural Solvers for Partial Differential Equations
+
+- **ID:** `2602.14853` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Jonathan Gorard, Ammar Hakim, James Juno
+- **abs:** https://arxiv.org/abs/2602.14853 · **PDF:** https://arxiv.org/pdf/2602.14853v1
+
+## 11959. Atomix: Timely, Transactional Tool Use for Reliable Agentic Workflows
+
+- **ID:** `2602.14849` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Bardia Mohammadi, Nearchos Potamitis, Lars Klein, et al.
+- **abs:** https://arxiv.org/abs/2602.14849 · **PDF:** https://arxiv.org/pdf/2602.14849v1
+
+## 11960. Multi-dimensional Persistent Sheaf Laplacians for Image Analysis
+
+- **ID:** `2602.14846` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Xiang Xiang Wang, Guo-Wei Wei
+- **abs:** https://arxiv.org/abs/2602.14846 · **PDF:** https://arxiv.org/pdf/2602.14846v1
+
+## 11961. Interactionless Inverse Reinforcement Learning: A Data-Centric Framework for Durable Alignment
+
+- **ID:** `2602.14844` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Elias Malomgré, Pieter Simoens
+- **abs:** https://arxiv.org/abs/2602.14844 · **PDF:** https://arxiv.org/pdf/2602.14844v2
+
+## 11962. Debiasing Central Fixation Confounds Reveals a Peripheral "Sweet Spot" for Human-like Scanpaths in Hard-Attention Vision
+
+- **ID:** `2602.14834` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Pengcheng Pan, Yonekura Shogo, Yasuo Kuniyosh
+- **abs:** https://arxiv.org/abs/2602.14834 · **PDF:** https://arxiv.org/pdf/2602.14834v1
+
+## 11963. RF-GPT: Teaching AI to See the Wireless World
+
+- **ID:** `2602.14833` · **primary:** `eess.SP` · **Submitted:** 2026-02-16
+- **Authors:** Hang Zou, Yu Tian, Bohao Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.14833 · **PDF:** https://arxiv.org/pdf/2602.14833v1
+
+## 11964. Exploring the limits of pre-trained embeddings in machine-guided protein design: a case study on predicting AAV vector viability
+
+- **ID:** `2602.14828` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-16
+- **Authors:** Ana F. Rodrigues, Lucas Ferraz, Laura Balbi, et al.
+- **abs:** https://arxiv.org/abs/2602.14828 · **PDF:** https://arxiv.org/pdf/2602.14828v1
+
+## 11965. Testimole-Conversational: A 30-Billion-Word Italian Discussion Board Corpus (1996-2024) for Language Modeling and Sociolinguistic Research
+
+- **ID:** `2602.14819` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Matteo Rinaldi, Rossella Varvara, Viviana Patti
+- **abs:** https://arxiv.org/abs/2602.14819 · **PDF:** https://arxiv.org/pdf/2602.14819v1
+
+## 11966. Learning State-Tracking from Code Using Linear RNNs
+
+- **ID:** `2602.14814` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Julien Siems, Riccardo Grazzi, Kirill Kalinin, et al.
+- **abs:** https://arxiv.org/abs/2602.14814 · **PDF:** https://arxiv.org/pdf/2602.14814v1
+
+## 11967. Physical Commonsense Reasoning for Lower-Resourced Languages and Dialects: a Study on Basque
+
+- **ID:** `2602.14812` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Jaione Bengoetxea, Itziar Gonzalez-Dios, Rodrigo Agerri
+- **abs:** https://arxiv.org/abs/2602.14812 · **PDF:** https://arxiv.org/pdf/2602.14812v2
+
+## 11968. Hybrid Feature Learning with Time Series Embeddings for Equipment Anomaly Prediction
+
+- **ID:** `2602.15089` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Takato Yasuno
+- **abs:** https://arxiv.org/abs/2602.15089 · **PDF:** https://arxiv.org/pdf/2602.15089v1
+
+## 11969. Resp-Agent: An Agent-Based System for Multimodal Respiratory Sound Generation and Disease Diagnosis
+
+- **ID:** `2602.15909` · **primary:** `eess.AS` · **Submitted:** 2026-02-16
+- **Authors:** Pengfei Zhang, Tianxin Xie, Minghao Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.15909 · **PDF:** https://arxiv.org/pdf/2602.15909v3
+
+## 11970. Overthinking Loops in Agents: A Structural Risk via MCP Tools
+
+- **ID:** `2602.14798` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Yohan Lee, Jisoo Jang, Seoyeon Choi, et al.
+- **abs:** https://arxiv.org/abs/2602.14798 · **PDF:** https://arxiv.org/pdf/2602.14798v1
+
+## 11971. IT-DPC-SRI: A Cloud-Optimized Archive of Italian Radar Precipitation (2010-2025)
+
+- **ID:** `2602.15088` · **primary:** `physics.ao-ph` · **Submitted:** 2026-02-16
+- **Authors:** Gabriele Franch, Elena Tomasi, Uladzislau Azhel, et al.
+- **abs:** https://arxiv.org/abs/2602.15088 · **PDF:** https://arxiv.org/pdf/2602.15088v1
+
+## 11972. Return of the Schema: Building Complete Datasets for Machine Learning and Reasoning on Knowledge Graphs
+
+- **ID:** `2602.14795` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Ivan Diliso, Roberto Barile, Claudia d'Amato, et al.
+- **abs:** https://arxiv.org/abs/2602.14795 · **PDF:** https://arxiv.org/pdf/2602.14795v1
+
+## 11973. Extending Multi-Source Bayesian Optimization With Causality Principles
+
+- **ID:** `2602.14791` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Luuk Jacobs, Mohammad Ali Javidian
+- **abs:** https://arxiv.org/abs/2602.14791 · **PDF:** https://arxiv.org/pdf/2602.14791v1
+
+## 11974. On the Stability of Nonlinear Dynamics in GD and SGD: Beyond Quadratic Potentials
+
+- **ID:** `2602.14789` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Rotem Mulayoff, Sebastian U. Stich
+- **abs:** https://arxiv.org/abs/2602.14789 · **PDF:** https://arxiv.org/pdf/2602.14789v1
+
+## 11975. VIPA: Visual Informative Part Attention for Referring Image Segmentation
+
+- **ID:** `2602.14788` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Yubin Cho, Hyunwoo Yu, Kyeongbo Kong, et al.
+- **abs:** https://arxiv.org/abs/2602.14788 · **PDF:** https://arxiv.org/pdf/2602.14788v1
+
+## 11976. SA-SSL-MOS: Self-supervised Learning MOS Prediction with Spectral Augmentation for Generalized Multi-Rate Speech Assessment
+
+- **ID:** `2602.14785` · **primary:** `eess.AS` · **Submitted:** 2026-02-16
+- **Authors:** Fengyuan Cao, Xinyu Liang, Fredrik Cumlin, et al.
+- **abs:** https://arxiv.org/abs/2602.14785 · **PDF:** https://arxiv.org/pdf/2602.14785v1
+
+## 11977. What hackers talk about when they talk about AI: Early-stage diffusion of a cybercrime innovation
+
+- **ID:** `2602.14783` · **primary:** `cs.CY` · **Submitted:** 2026-02-16
+- **Authors:** Benoît Dupont, Chad Whelan, Serge-Olivier Paquette
+- **abs:** https://arxiv.org/abs/2602.14783 · **PDF:** https://arxiv.org/pdf/2602.14783v1
+
+## 11978. A Geometric Analysis of Small-sized Language Model Hallucinations
+
+- **ID:** `2602.14778` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Emanuele Ricco, Elia Onofri, Lorenzo Cima, et al.
+- **abs:** https://arxiv.org/abs/2602.14778 · **PDF:** https://arxiv.org/pdf/2602.14778v2
+
+## 11979. Emergently Misaligned Language Models Show Behavioral Self-Awareness That Shifts With Subsequent Realignment
+
+- **ID:** `2602.14777` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Laurène Vaugrante, Anietta Weckauff, Thilo Hagendorff
+- **abs:** https://arxiv.org/abs/2602.14777 · **PDF:** https://arxiv.org/pdf/2602.14777v1
+
+## 11980. Learning Structural Hardness for Combinatorial Auctions: Instance-Dependent Algorithm Selection via Graph Neural Networks
+
+- **ID:** `2602.14772` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Sungwoo Kang
+- **abs:** https://arxiv.org/abs/2602.14772 · **PDF:** https://arxiv.org/pdf/2602.14772v1
+
+## 11981. GOT-JEPA: Generic Object Tracking with Model Adaptation and Occlusion Handling using Joint-Embedding Predictive Architecture
+
+- **ID:** `2602.14771` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Shih-Fang Chen, Jun-Cheng Chen, I-Hong Jhuo, et al.
+- **abs:** https://arxiv.org/abs/2602.14771 · **PDF:** https://arxiv.org/pdf/2602.14771v4
+
+## 11982. Multi-Agent Comedy Club: Investigating Community Discussion Effects on LLM Humor Generation
+
+- **ID:** `2602.14770` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Shiwei Hong, Lingyao Li, Ethan Z. Rong, et al.
+- **abs:** https://arxiv.org/abs/2602.14770 · **PDF:** https://arxiv.org/pdf/2602.14770v2
+
+## 11983. SpreadsheetArena: Decomposing Preference in LLM Generation of Spreadsheet Workbooks
+
+- **ID:** `2603.10002` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Srivatsa Kundurthy, Clara Na, Michael Handley, et al.
+- **abs:** https://arxiv.org/abs/2603.10002 · **PDF:** https://arxiv.org/pdf/2603.10002v1
+
+## 11984. Leveraging Wikidata for Geographically Informed Sociocultural Bias Dataset Creation: Application to Latin America
+
+- **ID:** `2603.10001` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Yannis Karmim, Renato Pino, Hernan Contreras, et al.
+- **abs:** https://arxiv.org/abs/2603.10001 · **PDF:** https://arxiv.org/pdf/2603.10001v2
+
+## 11985. High-Resolution Range Profile Classifiers Require Aspect-Angle Awareness
+
+- **ID:** `2603.00087` · **primary:** `eess.SP` · **Submitted:** 2026-02-16
+- **Authors:** Edwyn Brient, Santiago Velasco-Forero, Rami Kassab
+- **abs:** https://arxiv.org/abs/2603.00087 · **PDF:** https://arxiv.org/pdf/2603.00087v1
+
+## 11986. Unlocking Reasoning Capability on Machine Translation in Large Language Models
+
+- **ID:** `2602.14763` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Sara Rajaee, Sebastian Vincent, Alexandre Berard, et al.
+- **abs:** https://arxiv.org/abs/2602.14763 · **PDF:** https://arxiv.org/pdf/2602.14763v1
+
+## 11987. Universal Algorithm-Implicit Learning
+
+- **ID:** `2602.14761` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Stefano Woerner, Seong Joon Oh, Christian F. Baumgartner
+- **abs:** https://arxiv.org/abs/2602.14761 · **PDF:** https://arxiv.org/pdf/2602.14761v1
+
+## 11988. Residual Connections and the Causal Shift: Uncovering a Structural Misalignment in Transformers
+
+- **ID:** `2602.14760` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Jonathan Lys, Vincent Gripon, Bastien Pasdeloup, et al.
+- **abs:** https://arxiv.org/abs/2602.14760 · **PDF:** https://arxiv.org/pdf/2602.14760v2
+
+## 11989. Inner Loop Inference for Pretrained Transformers: Unlocking Latent Capabilities Without Training
+
+- **ID:** `2602.14759` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Jonathan Lys, Vincent Gripon, Bastien Pasdeloup, et al.
+- **abs:** https://arxiv.org/abs/2602.14759 · **PDF:** https://arxiv.org/pdf/2602.14759v2
+
+## 11990. Solving Inverse Parametrized Problems via Finite Elements and Extreme Learning Networks
+
+- **ID:** `2602.14757` · **primary:** `math.NA` · **Submitted:** 2026-02-16
+- **Authors:** Erik Burman, Mats G. Larson, Karl Larsson, et al.
+- **abs:** https://arxiv.org/abs/2602.14757 · **PDF:** https://arxiv.org/pdf/2602.14757v1
+
+## 11991. StrokeNeXt: A Siamese-encoder Approach for Brain Stroke Classification in Computed Tomography Imagery
+
+- **ID:** `2602.15087` · **primary:** `eess.IV` · **Submitted:** 2026-02-16
+- **Authors:** Leo Thomas Ramos, Angel D. Sappa
+- **abs:** https://arxiv.org/abs/2602.15087 · **PDF:** https://arxiv.org/pdf/2602.15087v1
+
+## 11992. Cognitive networks reconstruct mindsets about STEM subjects and educational contexts in almost 1000 high-schoolers, University students and LLM-based digital twins
+
+- **ID:** `2602.14749` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Francesco Gariboldi, Emma Franchino, Edith Haim, et al.
+- **abs:** https://arxiv.org/abs/2602.14749 · **PDF:** https://arxiv.org/pdf/2602.14749v1
+
+## 11993. Rethinking the Role of LLMs in Time Series Forecasting
+
+- **ID:** `2602.14744` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Xin Qiu, Junlong Tong, Yirong Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.14744 · **PDF:** https://arxiv.org/pdf/2602.14744v2
+
+## 11994. LLMStructBench: Benchmarking Large Language Model Structured Data Extraction
+
+- **ID:** `2602.14743` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Sönke Tenckhoff, Mario Koddenbrock, Erik Rodner
+- **abs:** https://arxiv.org/abs/2602.14743 · **PDF:** https://arxiv.org/pdf/2602.14743v1
+
+## 11995. AI Arms and Influence: Frontier Models Exhibit Sophisticated Reasoning in Simulated Nuclear Crises
+
+- **ID:** `2602.14740` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Kenneth Payne
+- **abs:** https://arxiv.org/abs/2602.14740 · **PDF:** https://arxiv.org/pdf/2602.14740v1
+
+## 11996. Parameter-Minimal Neural DE Solvers via Horner Polynomials
+
+- **ID:** `2602.14737` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** T. Matulić, D. Seršić
+- **abs:** https://arxiv.org/abs/2602.14737 · **PDF:** https://arxiv.org/pdf/2602.14737v1
+
+## 11997. The Signal Horizon: Local Blindness and the Contraction of Pauli-Weight Spectra in Noisy Quantum Encodings
+
+- **ID:** `2602.14735` · **primary:** `quant-ph` · **Submitted:** 2026-02-16
+- **Authors:** Ait Haddou Marwan
+- **abs:** https://arxiv.org/abs/2602.14735 · **PDF:** https://arxiv.org/pdf/2602.14735v1
+
+## 11998. Iterative LLM-based improvement for French Clinical Interview Transcription and Speaker Diarization
+
+- **ID:** `2603.00086` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Ambre Marie, Thomas Bertin, Guillaume Dardenne, et al.
+- **abs:** https://arxiv.org/abs/2603.00086 · **PDF:** https://arxiv.org/pdf/2603.00086v1
+
+## 11999. Scale redundancy and soft gauge fixing in positively homogeneous neural networks
+
+- **ID:** `2602.14729` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Rodrigo Carmo Terin
+- **abs:** https://arxiv.org/abs/2602.14729 · **PDF:** https://arxiv.org/pdf/2602.14729v1
+
+## 12000. D2-LoRA: A Synergistic Approach to Differential and Directional Low-Rank Adaptation
+
+- **ID:** `2602.14728` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Nozomu Fujisawa, Masaaki Kondo
+- **abs:** https://arxiv.org/abs/2602.14728 · **PDF:** https://arxiv.org/pdf/2602.14728v1
+
+## 12001. ManeuverNet: A Soft Actor-Critic Framework for Precise Maneuvering of Double-Ackermann-Steering Robots with Optimized Reward Functions
+
+- **ID:** `2602.14726` · **primary:** `cs.RO` · **Submitted:** 2026-02-16
+- **Authors:** Kohio Deflesselle, Mélodie Daniel, Aly Magassouba, et al.
+- **abs:** https://arxiv.org/abs/2602.14726 · **PDF:** https://arxiv.org/pdf/2602.14726v1
+
+## 12002. WebWorld: A Large-Scale World Model for Web Agent Training
+
+- **ID:** `2602.14721` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Zikai Xiao, Jianhong Tu, Chuhang Zou, et al.
+- **abs:** https://arxiv.org/abs/2602.14721 · **PDF:** https://arxiv.org/pdf/2602.14721v1
+
+## 12003. Orcheo: A Modular Full-Stack Platform for Conversational Search
+
+- **ID:** `2602.14710` · **primary:** `cs.IR` · **Submitted:** 2026-02-16
+- **Authors:** Shaojie Jiang, Svitlana Vakulenko, Maarten de Rijke
+- **abs:** https://arxiv.org/abs/2602.14710 · **PDF:** https://arxiv.org/pdf/2602.14710v1
+
+## 12004. Unbiased Approximate Vector-Jacobian Products for Efficient Backpropagation
+
+- **ID:** `2602.14701` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Killian Bakong, Laurent Massoulié, Edouard Oyallon, et al.
+- **abs:** https://arxiv.org/abs/2602.14701 · **PDF:** https://arxiv.org/pdf/2602.14701v1
+
+## 12005. Qute: Towards Quantum-Native Database
+
+- **ID:** `2602.14699` · **primary:** `cs.DB` · **Submitted:** 2026-02-16
+- **Authors:** Muzhi Chen, Xuanhe Zhou, Wei Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.14699 · **PDF:** https://arxiv.org/pdf/2602.14699v1
+
+## 12006. Engineering Reasoning and Instruction (ERI) Benchmark: A Large Taxonomy-driven Dataset for Foundation Models and Agents
+
+- **ID:** `2603.02239` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** MZ Naser, Ahmad Bani Awwad, Zoie McCreery, et al.
+- **abs:** https://arxiv.org/abs/2603.02239 · **PDF:** https://arxiv.org/pdf/2603.02239v1
+
+## 12007. Evolutionary System Prompt Learning for Reinforcement Learning in LLMs
+
+- **ID:** `2602.14697` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Lunjun Zhang, Ryan Chen, Bradly C. Stadie
+- **abs:** https://arxiv.org/abs/2602.14697 · **PDF:** https://arxiv.org/pdf/2602.14697v3
+
+## 12008. A Critical Look at Targeted Instruction Selection: Disentangling What Matters (and What Doesn't)
+
+- **ID:** `2602.14696` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Nihal V. Nayak, Paula Rodriguez-Diaz, Neha Hulkund, et al.
+- **abs:** https://arxiv.org/abs/2602.14696 · **PDF:** https://arxiv.org/pdf/2602.14696v1
+
+## 12009. TokaMind: A Multi-Modal Transformer Foundation Model for Tokamak Plasma Dynamics
+
+- **ID:** `2602.15084` · **primary:** `physics.plasm-ph` · **Submitted:** 2026-02-16
+- **Authors:** Tobia Boschi, Andrea Loreti, Nicola C. Amorisco, et al.
+- **abs:** https://arxiv.org/abs/2602.15084 · **PDF:** https://arxiv.org/pdf/2602.15084v1
+
+## 12010. Removing Planner Bias in Goal Recognition Through Multi-Plan Dataset Generation
+
+- **ID:** `2602.14691` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Mustafa F. Abdelwahed, Felipe Meneguzzi Kin Max Piamolini Gusmao, Joan Espasa
+- **abs:** https://arxiv.org/abs/2602.14691 · **PDF:** https://arxiv.org/pdf/2602.14691v1
+
+## 12011. Exposing the Systematic Vulnerability of Open-Weight Models to Prefill Attacks
+
+- **ID:** `2602.14689` · **primary:** `cs.CR` · **Submitted:** 2026-02-16
+- **Authors:** Lukas Struppek, Adam Gleave, Kellin Pelrine
+- **abs:** https://arxiv.org/abs/2602.14689 · **PDF:** https://arxiv.org/pdf/2602.14689v1
+
+## 12012. SynthSAEBench: Evaluating Sparse Autoencoders on Scalable Realistic Synthetic Data
+
+- **ID:** `2602.14687` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** David Chanin, Adrià Garriga-Alonso
+- **abs:** https://arxiv.org/abs/2602.14687 · **PDF:** https://arxiv.org/pdf/2602.14687v1
+
+## 12013. Exposing Diversity Bias in Deep Generative Models: Statistical Origins and Correction of Diversity Error
+
+- **ID:** `2602.14682` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Farzan Farnia, Mohammad Jalali, Azim Ospanov
+- **abs:** https://arxiv.org/abs/2602.14682 · **PDF:** https://arxiv.org/pdf/2602.14682v1
+
+## 12014. ST-EVO: Towards Generative Spatio-Temporal Evolution of Multi-Agent Communication Topologies
+
+- **ID:** `2602.14681` · **primary:** `cs.MA` · **Submitted:** 2026-02-16
+- **Authors:** Xingjian Wu, Xvyuan Liu, Junkai Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.14681 · **PDF:** https://arxiv.org/pdf/2602.14681v3
+
+## 12015. Kernel-based optimization of measurement operators for quantum reservoir computers
+
+- **ID:** `2602.14677` · **primary:** `quant-ph` · **Submitted:** 2026-02-16
+- **Authors:** Markus Gross, Hans-Martin Rieser
+- **abs:** https://arxiv.org/abs/2602.14677 · **PDF:** https://arxiv.org/pdf/2602.14677v2
+
+## 12016. GREAT-EER: Graph Edge Attention Network for Emergency Evacuation Responses
+
+- **ID:** `2602.14676` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Attila Lischka, Balázs Kulcsár
+- **abs:** https://arxiv.org/abs/2602.14676 · **PDF:** https://arxiv.org/pdf/2602.14676v1
+
+## 12017. Crowdsourcing Piedmontese to Test LLMs on Non-Standard Orthography
+
+- **ID:** `2602.14675` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Gianluca Vico, Jindřich Libovický
+- **abs:** https://arxiv.org/abs/2602.14675 · **PDF:** https://arxiv.org/pdf/2602.14675v1
+
+## 12018. From User Preferences to Base Score Extraction Functions in Gradual Argumentation (with Appendix)
+
+- **ID:** `2602.14674` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Aniol Civit, Antonio Rago, Antonio Andriella, et al.
+- **abs:** https://arxiv.org/abs/2602.14674 · **PDF:** https://arxiv.org/pdf/2602.14674v3
+
+## 12019. Pseudo-differential-enhanced physics-informed neural networks
+
+- **ID:** `2602.14663` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Andrew Gracyk
+- **abs:** https://arxiv.org/abs/2602.14663 · **PDF:** https://arxiv.org/pdf/2602.14663v1
+
+## 12020. An Embarrassingly Simple Way to Optimize Orthogonal Matrices at Scale
+
+- **ID:** `2602.14656` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Adrián Javaloy, Antonio Vergari
+- **abs:** https://arxiv.org/abs/2602.14656 · **PDF:** https://arxiv.org/pdf/2602.14656v1
+
+## 12021. Breaking Data Efficiency Dilemma: A Federated and Augmented Learning Framework For Alzheimer's Disease Detection via Speech
+
+- **ID:** `2602.14655` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Xiao Wei, Bin Wen, Yuqin Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.14655 · **PDF:** https://arxiv.org/pdf/2602.14655v1
+
+## 12022. Is Information Density Uniform when Utterances are Grounded on Perception and Discourse?
+
+- **ID:** `2602.14653` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Matteo Gay, Coleman Haley, Mario Giulianelli, et al.
+- **abs:** https://arxiv.org/abs/2602.14653 · **PDF:** https://arxiv.org/pdf/2602.14653v1
+
+## 12023. GradMAP: Faster Layer Pruning with Gradient Metric and Projection Compensation
+
+- **ID:** `2602.14649` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Hao Liu, Guangyan Li, Wensheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.14649 · **PDF:** https://arxiv.org/pdf/2602.14649v1
+
+## 12024. Arbor: A Framework for Reliable Navigation of Critical Conversation Flows
+
+- **ID:** `2602.14643` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Luís Silva, Diogo Gonçalves, Catarina Farinha, et al.
+- **abs:** https://arxiv.org/abs/2602.14643 · **PDF:** https://arxiv.org/pdf/2602.14643v3
+
+## 12025. GenPANIS: A Latent-Variable Generative Framework for Forward and Inverse PDE Problems in Multiphase Media
+
+- **ID:** `2602.14642` · **primary:** `stat.ML` · **Submitted:** 2026-02-16
+- **Authors:** Matthaios Chatzopoulos, Phaedon-Stelios Koutsourelakis
+- **abs:** https://arxiv.org/abs/2602.14642 · **PDF:** https://arxiv.org/pdf/2602.14642v1
+
+## 12026. Quantum Reservoir Computing with Neutral Atoms on a Small, Complex, Medical Dataset
+
+- **ID:** `2602.14641` · **primary:** `quant-ph` · **Submitted:** 2026-02-16
+- **Authors:** Luke Antoncich, Yuben Moodley, Ugo Varetto, et al.
+- **abs:** https://arxiv.org/abs/2602.14641 · **PDF:** https://arxiv.org/pdf/2602.14641v1
+
+## 12027. Beyond the Prompt in Large Language Models: Comprehension, In-Context Learning, and Chain-of-Thought
+
+- **ID:** `2603.10000` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Yuling Jiao, Yanming Lai, Huazhen Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.10000 · **PDF:** https://arxiv.org/pdf/2603.10000v2
+
+## 12028. Alignment Adapter to Improve the Performance of Compressed Deep Learning Models
+
+- **ID:** `2602.14635` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Rohit Raj Rai, Abhishek Dhaka, Amit Awekar
+- **abs:** https://arxiv.org/abs/2602.14635 · **PDF:** https://arxiv.org/pdf/2602.14635v1
+
+## 12029. Bayesian Cosmic Void Finding with Graph Flows
+
+- **ID:** `2602.14630` · **primary:** `astro-ph.CO` · **Submitted:** 2026-02-16
+- **Authors:** Leander Thiele
+- **abs:** https://arxiv.org/abs/2602.14630 · **PDF:** https://arxiv.org/pdf/2602.14630v1
+
+## 12030. Concepts' Information Bottleneck Models
+
+- **ID:** `2602.14626` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Karim Galliamov, Syed M Ahsan Kazmi, Adil Khan, et al.
+- **abs:** https://arxiv.org/abs/2602.14626 · **PDF:** https://arxiv.org/pdf/2602.14626v1
+
+## 12031. S-PRESSO: Ultra Low Bitrate Sound Effect Compression With Diffusion Autoencoders And Offline Quantization
+
+- **ID:** `2602.15082` · **primary:** `cs.SD` · **Submitted:** 2026-02-16
+- **Authors:** Zineb Lahrichi, Gaëtan Hadjeres, Gaël Richard, et al.
+- **abs:** https://arxiv.org/abs/2602.15082 · **PDF:** https://arxiv.org/pdf/2602.15082v2
+
+## 12032. Tabular Foundation Models Can Learn Association Rules
+
+- **ID:** `2602.14622` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Erkan Karabulut, Daniel Daza, Paul Groth, et al.
+- **abs:** https://arxiv.org/abs/2602.14622 · **PDF:** https://arxiv.org/pdf/2602.14622v2
+
+## 12033. VariViT: A Vision Transformer for Variable Image Sizes
+
+- **ID:** `2602.14615` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Aswathi Varma, Suprosanna Shit, Chinmay Prabhakar, et al.
+- **abs:** https://arxiv.org/abs/2602.14615 · **PDF:** https://arxiv.org/pdf/2602.14615v1
+
+## 12034. LongAudio-RAG: Event-Grounded Question Answering over Multi-Hour Long Audio
+
+- **ID:** `2602.14612` · **primary:** `eess.AS` · **Submitted:** 2026-02-16
+- **Authors:** Naveen Vakada, Kartik Hegde, Arvind Krishna Sridhar, et al.
+- **abs:** https://arxiv.org/abs/2602.14612 · **PDF:** https://arxiv.org/pdf/2602.14612v3
+
+## 12035. A Bayesian Approach to Low-Discrepancy Subset Selection
+
+- **ID:** `2602.14607` · **primary:** `stat.ME` · **Submitted:** 2026-02-16
+- **Authors:** Nathan Kirk
+- **abs:** https://arxiv.org/abs/2602.14607 · **PDF:** https://arxiv.org/pdf/2602.14607v1
+
+## 12036. Towards Selection as Power: Bounding Decision Authority in Autonomous Agents
+
+- **ID:** `2602.14606` · **primary:** `cs.MA` · **Submitted:** 2026-02-16
+- **Authors:** Jose Manuel de la Chica Rodriguez, Juan Manuel Vera Díaz
+- **abs:** https://arxiv.org/abs/2602.14606 · **PDF:** https://arxiv.org/pdf/2602.14606v1
+
+## 12037. OPBench: A Graph Benchmark to Combat the Opioid Crisis
+
+- **ID:** `2602.14602` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Tianyi Ma, Yiyang Li, Yiyue Qian, et al.
+- **abs:** https://arxiv.org/abs/2602.14602 · **PDF:** https://arxiv.org/pdf/2602.14602v1
+
+## 12038. The Wikidata Query Logs Dataset
+
+- **ID:** `2602.14594` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Sebastian Walter, Hannah Bast
+- **abs:** https://arxiv.org/abs/2602.14594 · **PDF:** https://arxiv.org/pdf/2602.14594v1
+
+## 12039. Automated Classification of Source Code Changes Based on Metrics Clustering in the Software Development Process
+
+- **ID:** `2602.14591` · **primary:** `cs.SE` · **Submitted:** 2026-02-16
+- **Authors:** Evgenii Kniazev
+- **abs:** https://arxiv.org/abs/2602.14591 · **PDF:** https://arxiv.org/pdf/2602.14591v1
+
+## 12040. MATEO: A Multimodal Benchmark for Temporal Reasoning and Planning in LVLMs
+
+- **ID:** `2602.14589` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Gabriel Roccabruna, Olha Khomyn, Giuseppe Riccardi
+- **abs:** https://arxiv.org/abs/2602.14589 · **PDF:** https://arxiv.org/pdf/2602.14589v1
+
+## 12041. Decoupled Continuous-Time Reinforcement Learning via Hamiltonian Flow
+
+- **ID:** `2602.14587` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Minh Nguyen
+- **abs:** https://arxiv.org/abs/2602.14587 · **PDF:** https://arxiv.org/pdf/2602.14587v1
+
+## 12042. Replicable Constrained Bandits
+
+- **ID:** `2602.14580` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Matteo Bollini, Gianmarco Genalti, Francesco Emanuele Stradi, et al.
+- **abs:** https://arxiv.org/abs/2602.14580 · **PDF:** https://arxiv.org/pdf/2602.14580v1
+
+## 12043. RNM-TD3: N:M Semi-structured Sparse Reinforcement Learning From Scratch
+
+- **ID:** `2602.14578` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Isam Vrce, Andreas Kassler, Gökçe Aydos
+- **abs:** https://arxiv.org/abs/2602.14578 · **PDF:** https://arxiv.org/pdf/2602.14578v1
+
+## 12044. Generating Realistic, Protocol-Compliant Maritime Radio Dialogues using Self-Instruct and Low-Rank Adaptation
+
+- **ID:** `2603.04423` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Gürsel Akdeniz, Emin Cagatay Nakilcioglu
+- **abs:** https://arxiv.org/abs/2603.04423 · **PDF:** https://arxiv.org/pdf/2603.04423v1
+
+## 12045. DCTracks: An Open Dataset for Machine Learning-Based Drift Chamber Track Reconstruction
+
+- **ID:** `2602.14571` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Qian Liyan, Zhang Yao, Yuan Ye, et al.
+- **abs:** https://arxiv.org/abs/2602.14571 · **PDF:** https://arxiv.org/pdf/2602.14571v1
+
+## 12046. Assessing Large Language Models for Medical QA: Zero-Shot and LLM-as-a-Judge Evaluation
+
+- **ID:** `2602.14564` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Shefayat E Shams Adib, Ahmed Alfey Sani, Ekramul Alam Esham, et al.
+- **abs:** https://arxiv.org/abs/2602.14564 · **PDF:** https://arxiv.org/pdf/2602.14564v1
+
+## 12047. Fluid-Agent Reinforcement Learning
+
+- **ID:** `2602.14559` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Shishir Sharma, Doina Precup, Theodore J. Perkins
+- **abs:** https://arxiv.org/abs/2602.14559 · **PDF:** https://arxiv.org/pdf/2602.14559v1
+
+## 12048. Governing AI Forgetting: Auditing for Machine Unlearning Compliance
+
+- **ID:** `2602.14553` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Qinqi Lin, Ningning Ding, Lingjie Duan, et al.
+- **abs:** https://arxiv.org/abs/2602.14553 · **PDF:** https://arxiv.org/pdf/2602.14553v1
+
+## 12049. A Retrieval-Augmented Language Assistant for Unmanned Aircraft Safety Assessment and Regulatory Compliance
+
+- **ID:** `2603.09999` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Gabriele Immordino, Andrea Vaiuso, Marcello Righi
+- **abs:** https://arxiv.org/abs/2603.09999 · **PDF:** https://arxiv.org/pdf/2603.09999v1
+
+## 12050. Truly Adapting to Adversarial Constraints in Constrained MABs
+
+- **ID:** `2602.14543` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Francesco Emanuele Stradi, Kalana Kalupahana, Matteo Castiglioni, et al.
+- **abs:** https://arxiv.org/abs/2602.14543 · **PDF:** https://arxiv.org/pdf/2602.14543v1
+
+## 12051. Explainable Token-level Noise Filtering for LLM Fine-tuning Datasets
+
+- **ID:** `2602.14536` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Yuchen Yang, Wenze Lin, Enhao Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.14536 · **PDF:** https://arxiv.org/pdf/2602.14536v2
+
+## 12052. Disentangling Deception and Hallucination Failures in LLMs
+
+- **ID:** `2602.14529` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Haolang Lu, Hongrui Peng, WeiYe Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.14529 · **PDF:** https://arxiv.org/pdf/2602.14529v1
+
+## 12053. TWISTED-RL: Hierarchical Skilled Agents for Knot-Tying without Human Demonstrations
+
+- **ID:** `2602.14526` · **primary:** `cs.RO` · **Submitted:** 2026-02-16
+- **Authors:** Guy Freund, Tom Jurgenson, Matan Sudry, et al.
+- **abs:** https://arxiv.org/abs/2602.14526 · **PDF:** https://arxiv.org/pdf/2602.14526v1
+
+## 12054. Accelerating Posterior Inference from Pulsar Light Curves via Learned Latent Representations and Local Simulator-Guided Optimization
+
+- **ID:** `2602.14520` · **primary:** `stat.ML` · **Submitted:** 2026-02-16
+- **Authors:** Farhana Taiyebah, Abu Bucker Siddik, Indronil Bhattacharjee, et al.
+- **abs:** https://arxiv.org/abs/2602.14520 · **PDF:** https://arxiv.org/pdf/2602.14520v1
+
+## 12055. DeepMTL2R: A Library for Deep Multi-task Learning to Rank
+
+- **ID:** `2602.14519` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Chaosheng Dong, Peiyao Xiao, Yijia Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.14519 · **PDF:** https://arxiv.org/pdf/2602.14519v1
+
+## 12056. Diagnosing Knowledge Conflict in Multimodal Long-Chain Reasoning
+
+- **ID:** `2602.14518` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Jing Tang, Kun Wang, Haolang Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.14518 · **PDF:** https://arxiv.org/pdf/2602.14518v1
+
+## 12057. Beyond Translation: Evaluating Mathematical Reasoning Capabilities of LLMs in Sinhala and Tamil
+
+- **ID:** `2602.14517` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Sukumar Kishanthan, Kumar Thushalika, Buddhi Jayasekara, et al.
+- **abs:** https://arxiv.org/abs/2602.14517 · **PDF:** https://arxiv.org/pdf/2602.14517v1
+
+## 12058. Covariance-Aware Transformers for Quadratic Programming and Decision Making
+
+- **ID:** `2602.14506` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Kutay Tire, Yufan Zhang, Ege Onur Taga, et al.
+- **abs:** https://arxiv.org/abs/2602.14506 · **PDF:** https://arxiv.org/pdf/2602.14506v1
+
+## 12059. Formally Verifying and Explaining Sepsis Treatment Policies with COOL-MC
+
+- **ID:** `2602.14505` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Dennis Gross
+- **abs:** https://arxiv.org/abs/2602.14505 · **PDF:** https://arxiv.org/pdf/2602.14505v1
+
+## 12060. Joint Sensor Deployment and Physics-Informed Graph Transformer for Smart Grid Attack Detection
+
+- **ID:** `2603.00085` · **primary:** `cs.NE` · **Submitted:** 2026-02-16
+- **Authors:** Mariam Elnour, Mohammad AlShaikh Saleh, Rachad Atat, et al.
+- **abs:** https://arxiv.org/abs/2603.00085 · **PDF:** https://arxiv.org/pdf/2603.00085v1
+
+## 12061. Bounding Probabilities of Causation with Partial Causal Diagrams
+
+- **ID:** `2602.14503` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Yuxuan Xie, Ang Li
+- **abs:** https://arxiv.org/abs/2602.14503 · **PDF:** https://arxiv.org/pdf/2602.14503v1
+
+## 12062. Uncertainty-Aware Vision-Language Segmentation for Medical Imaging
+
+- **ID:** `2602.14498` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Aryan Das, Tanishq Rachamalla, Koushik Biswas, et al.
+- **abs:** https://arxiv.org/abs/2602.14498 · **PDF:** https://arxiv.org/pdf/2602.14498v2
+
+## 12063. Divine Benevolence is an $x^2$: GLUs scale asymptotically faster than MLPs
+
+- **ID:** `2602.14495` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Alejandro Francisco Queiruga
+- **abs:** https://arxiv.org/abs/2602.14495 · **PDF:** https://arxiv.org/pdf/2602.14495v2
+
+## 12064. Query as Anchor: Scenario-Adaptive User Representation via Large Language Model
+
+- **ID:** `2602.14492` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Jiahao Yuan, Yike Xu, Jinyong Wen, et al.
+- **abs:** https://arxiv.org/abs/2602.14492 · **PDF:** https://arxiv.org/pdf/2602.14492v2
+
+## 12065. Parameter-Efficient Fine-Tuning of LLMs with Mixture of Space Experts
+
+- **ID:** `2602.14490` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Buze Zhang, Jinkai Tao, Zilang Zeng, et al.
+- **abs:** https://arxiv.org/abs/2602.14490 · **PDF:** https://arxiv.org/pdf/2602.14490v1
+
+## 12066. BETA-Labeling for Multilingual Dataset Construction in Low-Resource IR
+
+- **ID:** `2602.14488` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Md. Najib Hasan, Mst. Jannatun Ferdous Rain, Fyad Mohammed, et al.
+- **abs:** https://arxiv.org/abs/2602.14488 · **PDF:** https://arxiv.org/pdf/2602.14488v2
+
+## 12067. Revisiting the Platonic Representation Hypothesis: An Aristotelian View
+
+- **ID:** `2602.14486` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Fabian Gröger, Shuo Wen, Maria Brbić
+- **abs:** https://arxiv.org/abs/2602.14486 · **PDF:** https://arxiv.org/pdf/2602.14486v1
+
+## 12068. TikArt: Stabilizing Aperture-Guided Fine-Grained Visual Reasoning with Reinforcement Learning
+
+- **ID:** `2602.14482` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Hao Ding, Zhichuan Yang, Weijie Ge, et al.
+- **abs:** https://arxiv.org/abs/2602.14482 · **PDF:** https://arxiv.org/pdf/2602.14482v2
+
+## 12069. On the Rate-Distortion-Complexity Tradeoff for Semantic Communication
+
+- **ID:** `2602.14481` · **primary:** `cs.IT` · **Submitted:** 2026-02-16
+- **Authors:** Jingxuan Chai, Yong Xiao, Guangming Shi
+- **abs:** https://arxiv.org/abs/2602.14481 · **PDF:** https://arxiv.org/pdf/2602.14481v1
+
+## 12070. Constrained and Composite Sampling via Proximal Sampler
+
+- **ID:** `2602.14478` · **primary:** `stat.ML` · **Submitted:** 2026-02-16
+- **Authors:** Thanh Dang, Jiaming Liang
+- **abs:** https://arxiv.org/abs/2602.14478 · **PDF:** https://arxiv.org/pdf/2602.14478v1
+
+## 12071. When AI Agents Teach Each Other: Discourse Patterns Resembling Peer Learning in the Moltbook Community
+
+- **ID:** `2602.14477` · **primary:** `cs.HC` · **Submitted:** 2026-02-16
+- **Authors:** Eason Chen, Ce Guan, A Elshafiey, et al.
+- **abs:** https://arxiv.org/abs/2602.14477 · **PDF:** https://arxiv.org/pdf/2602.14477v2
+
+## 12072. One Good Source is All You Need: Near-Optimal Regret for Bandits under Heterogeneous Noise
+
+- **ID:** `2602.14474` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Aadirupa Saha, Amith Bhat, Haipeng Luo
+- **abs:** https://arxiv.org/abs/2602.14474 · **PDF:** https://arxiv.org/pdf/2602.14474v1
+
+## 12073. Learning Transferability: A Two-Stage Reinforcement Learning Approach for Enhancing Quadruped Robots' Performance in U-Shaped Stair Climbing
+
+- **ID:** `2602.14473` · **primary:** `cs.RO` · **Submitted:** 2026-02-16
+- **Authors:** Baixiao Huang, Baiyu Huang, Yu Hou
+- **abs:** https://arxiv.org/abs/2602.14473 · **PDF:** https://arxiv.org/pdf/2602.14473v1
+
+## 12074. Frequentist Regret Analysis of Gaussian Process Thompson Sampling via Fractional Posteriors
+
+- **ID:** `2602.14472` · **primary:** `math.ST` · **Submitted:** 2026-02-16
+- **Authors:** Somjit Roy, Prateek Jaiswal, Anirban Bhattacharya, et al.
+- **abs:** https://arxiv.org/abs/2602.14472 · **PDF:** https://arxiv.org/pdf/2602.14472v1
+
+## 12075. Socially-Weighted Alignment: A Game-Theoretic Framework for Multi-Agent LLM Systems
+
+- **ID:** `2602.14471` · **primary:** `cs.MA` · **Submitted:** 2026-02-16
+- **Authors:** Furkan Mumcu, Yasin Yilmaz
+- **abs:** https://arxiv.org/abs/2602.14471 · **PDF:** https://arxiv.org/pdf/2602.14471v1
+
+## 12076. Near-Optimal Sample Complexity for Online Constrained MDPs
+
+- **ID:** `2602.15076` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Chang Liu, Yunfan Li, Lin F. Yang
+- **abs:** https://arxiv.org/abs/2602.15076 · **PDF:** https://arxiv.org/pdf/2602.15076v1
+
+## 12077. HyperRAG: Reasoning N-ary Facts over Hypergraphs for Retrieval Augmented Generation
+
+- **ID:** `2602.14470` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Wen-Sheng Lien, Yu-Kai Chan, Hao-Lung Hsiao, et al.
+- **abs:** https://arxiv.org/abs/2602.14470 · **PDF:** https://arxiv.org/pdf/2602.14470v1
+
+## 12078. Measuring and Mitigating Post-hoc Rationalization in Reverse Chain-of-Thought Generation
+
+- **ID:** `2602.14469` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Guangyue Peng, Zongchao Chen, Wen Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.14469 · **PDF:** https://arxiv.org/pdf/2602.14469v1
+
+## 12079. LACONIC: Length-Aware Constrained Reinforcement Learning for LLM
+
+- **ID:** `2602.14468` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Chang Liu, Yiran Zhao, Lawrence Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.14468 · **PDF:** https://arxiv.org/pdf/2602.14468v1
+
+## 12080. Robust Bias Evaluation with FilBBQ: A Filipino Bias Benchmark for Question-Answering Language Models
+
+- **ID:** `2602.14466` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Lance Calvin Lim Gamboa, Yue Feng, Mark Lee
+- **abs:** https://arxiv.org/abs/2602.14466 · **PDF:** https://arxiv.org/pdf/2602.14466v1
+
+## 12081. CoCoDiff: Correspondence-Consistent Diffusion Model for Fine-grained Style Transfer
+
+- **ID:** `2602.14464` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Wenbo Nie, Zixiang Li, Renshuai Tao, et al.
+- **abs:** https://arxiv.org/abs/2602.14464 · **PDF:** https://arxiv.org/pdf/2602.14464v2
+
+## 12082. Silent Inconsistency in Data-Parallel Full Fine-Tuning: Diagnosing Worker-Level Optimization Misalignment
+
+- **ID:** `2602.14462` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Hong Li, Zhen Zhou, Honggang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.14462 · **PDF:** https://arxiv.org/pdf/2602.14462v2
+
+## 12083. Frontier AI Risk Management Framework in Practice: A Risk Analysis Technical Report v1.5
+
+- **ID:** `2602.14457` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Dongrui Liu, Yi Yu, Jie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.14457 · **PDF:** https://arxiv.org/pdf/2602.14457v1
+
+## 12084. Traceable Latent Variable Discovery Based on Multi-Agent Collaboration
+
+- **ID:** `2602.14456` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Huaming Du, Tao Hu, Yijie Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.14456 · **PDF:** https://arxiv.org/pdf/2602.14456v1
+
+## 12085. WiSparse: Boosting LLM Inference Efficiency with Weight-Aware Mixed Activation Sparsity
+
+- **ID:** `2602.14452` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Lei Chen, Yuan Meng, Xiaoyu Zhan, et al.
+- **abs:** https://arxiv.org/abs/2602.14452 · **PDF:** https://arxiv.org/pdf/2602.14452v1
+
+## 12086. Precedent-Informed Reasoning: Mitigating Overthinking in Large Reasoning Models via Test-Time Precedent Learning
+
+- **ID:** `2602.14451` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Qianyue Wang, Jinwu Hu, Huanxiang Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.14451 · **PDF:** https://arxiv.org/pdf/2602.14451v1
+
+## 12087. Task-Lens: Cross-Task Utility Based Speech Dataset Profiling for Low-Resource Indian Languages
+
+- **ID:** `2602.23388` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Swati Sharma, Divya V. Sharma, Anubha Gupta
+- **abs:** https://arxiv.org/abs/2602.23388 · **PDF:** https://arxiv.org/pdf/2602.23388v1
+
+## 12088. Selective Synchronization Attention
+
+- **ID:** `2602.14445` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Hasi Hays
+- **abs:** https://arxiv.org/abs/2602.14445 · **PDF:** https://arxiv.org/pdf/2602.14445v1
+
+## 12089. Broken Chains: The Cost of Incomplete Reasoning in LLMs
+
+- **ID:** `2602.14444` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Ian Su, Gaurav Purushothaman, Jey Narayan, et al.
+- **abs:** https://arxiv.org/abs/2602.14444 · **PDF:** https://arxiv.org/pdf/2602.14444v1
+
+## 12090. Structure-Aware Piano Accompaniment via Style Planning and Dataset-Aligned Pattern Retrieval
+
+- **ID:** `2602.15074` · **primary:** `cs.SD` · **Submitted:** 2026-02-16
+- **Authors:** Wanyu Zang, Yang Yu, Meng Yu
+- **abs:** https://arxiv.org/abs/2602.15074 · **PDF:** https://arxiv.org/pdf/2602.15074v1
+
+## 12091. CAIRO: Decoupling Order from Scale in Regression
+
+- **ID:** `2602.14440` · **primary:** `stat.ME` · **Submitted:** 2026-02-16
+- **Authors:** Harri Vanhems, Yue Zhao, Peng Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.14440 · **PDF:** https://arxiv.org/pdf/2602.14440v2
+
+## 12092. Building the ethical AI framework of the future: from philosophy to practice
+
+- **ID:** `2603.06599` · **primary:** `cs.CY` · **Submitted:** 2026-02-16
+- **Authors:** Jasper Kyle Catapang
+- **abs:** https://arxiv.org/abs/2603.06599 · **PDF:** https://arxiv.org/pdf/2603.06599v1
+
+## 12093. Synthetic Reader Panels: Tournament-Based Ideation with LLM Personas for Autonomous Publishing
+
+- **ID:** `2602.14433` · **primary:** `cs.CY` · **Submitted:** 2026-02-16
+- **Authors:** Fred Zimmerman
+- **abs:** https://arxiv.org/abs/2602.14433 · **PDF:** https://arxiv.org/pdf/2602.14433v1
+
+## 12094. S2D: Selective Spectral Decay for Quantization-Friendly Conditioning of Neural Activations
+
+- **ID:** `2602.14432` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Arnav Chavan, Nahush Lele, Udbhav Bamba, et al.
+- **abs:** https://arxiv.org/abs/2602.14432 · **PDF:** https://arxiv.org/pdf/2602.14432v1
+
+## 12095. A unified framework for evaluating the robustness of machine-learning interpretability for prospect risking
+
+- **ID:** `2602.14430` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Prithwijit Chowdhury, Ahmad Mustafa, Mohit Prabhushankar, et al.
+- **abs:** https://arxiv.org/abs/2602.14430 · **PDF:** https://arxiv.org/pdf/2602.14430v1
+
+## 12096. LLM-Guided Knowledge Distillation for Temporal Knowledge Graph Reasoning
+
+- **ID:** `2602.14428` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Wang Xing, Wei Song, Siyu Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.14428 · **PDF:** https://arxiv.org/pdf/2602.14428v1
+
+## 12097. The geometry of invariant learning: an information-theoretic analysis of data augmentation and generalization
+
+- **ID:** `2602.14423` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Abdelali Bouyahia, Frédéric LeBlanc, Mario Marchand
+- **abs:** https://arxiv.org/abs/2602.14423 · **PDF:** https://arxiv.org/pdf/2602.14423v1
+
+## 12098. WavePhaseNet: A DFT-Based Method for Constructing Semantic Conceptual Hierarchy Structures (SCHS)
+
+- **ID:** `2602.14419` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Kiyotaka Kasubuchi, Kazuo Fukiya
+- **abs:** https://arxiv.org/abs/2602.14419 · **PDF:** https://arxiv.org/pdf/2602.14419v1
+
+## 12099. Feature Recalibration Based Olfactory-Visual Multimodal Model for Enhanced Rice Deterioration Detection
+
+- **ID:** `2602.14408` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Rongqiang Zhao, Hengrui Hu, Yijing Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.14408 · **PDF:** https://arxiv.org/pdf/2602.14408v2
+
+## 12100. TruthStance: An Annotated Dataset of Conversations on Truth Social
+
+- **ID:** `2602.14406` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Fathima Ameen, Danielle Brown, Manusha Malgareddy, et al.
+- **abs:** https://arxiv.org/abs/2602.14406 · **PDF:** https://arxiv.org/pdf/2602.14406v1
+
+## 12101. Boule or Baguette? A Study on Task Topology, Length Generalization, and the Benefit of Reasoning Traces
+
+- **ID:** `2602.14404` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** William L. Tong, Ege Cakar, Cengiz Pehlevan
+- **abs:** https://arxiv.org/abs/2602.14404 · **PDF:** https://arxiv.org/pdf/2602.14404v1
+
+## 12102. pFedNavi: Structure-Aware Personalized Federated Vision-Language Navigation for Embodied AI
+
+- **ID:** `2602.14401` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Qingqian Yang, Hao Wang, Sai Qian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.14401 · **PDF:** https://arxiv.org/pdf/2602.14401v1
+
+## 12103. LRD-MPC: Efficient MPC Inference through Low-rank Decomposition
+
+- **ID:** `2602.14397` · **primary:** `cs.CR` · **Submitted:** 2026-02-16
+- **Authors:** Tingting Tang, Yongqin Wang, Murali Annavaram
+- **abs:** https://arxiv.org/abs/2602.14397 · **PDF:** https://arxiv.org/pdf/2602.14397v1
+
+## 12104. Hello-Chat: Towards Realistic Social Audio Interactions
+
+- **ID:** `2602.23387` · **primary:** `cs.SD` · **Submitted:** 2026-02-16
+- **Authors:** Yueran Hou, Peilei Jia, Zihan Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.23387 · **PDF:** https://arxiv.org/pdf/2602.23387v1
+
+## 12105. Beyond Token-Level Policy Gradients for Complex Reasoning with Large Language Models
+
+- **ID:** `2602.14386` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Mufan Xu, Kehai Chen, Xuefeng Bai, et al.
+- **abs:** https://arxiv.org/abs/2602.14386 · **PDF:** https://arxiv.org/pdf/2602.14386v1
+
+## 12106. Adapting VACE for Real-Time Autoregressive Video Diffusion
+
+- **ID:** `2602.14381` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Ryan Fosdick
+- **abs:** https://arxiv.org/abs/2602.14381 · **PDF:** https://arxiv.org/pdf/2602.14381v1
+
+## 12107. A Study on Multi-Class Online Fuzzy Classifiers for Dynamic Environments
+
+- **ID:** `2602.14375` · **primary:** `cs.LG` · **Submitted:** 2026-02-16
+- **Authors:** Kensuke Ajimoto, Yuma Yamamoto, Yoshifumi Kusunoki, et al.
+- **abs:** https://arxiv.org/abs/2602.14375 · **PDF:** https://arxiv.org/pdf/2602.14375v1
+
+## 12108. Differentially Private Retrieval-Augmented Generation
+
+- **ID:** `2602.14374` · **primary:** `cs.CR` · **Submitted:** 2026-02-16
+- **Authors:** Tingting Tang, James Flemings, Yongqin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.14374 · **PDF:** https://arxiv.org/pdf/2602.14374v1
+
+## 12109. Competition for attention predicts good-to-bad tipping in AI
+
+- **ID:** `2602.14370` · **primary:** `cs.AI` · **Submitted:** 2026-02-16
+- **Authors:** Neil F. Johnson, Frank Y. Huo
+- **abs:** https://arxiv.org/abs/2602.14370 · **PDF:** https://arxiv.org/pdf/2602.14370v2
+
+## 12110. InnoEval: On Research Idea Evaluation as a Knowledge-Grounded, Multi-Perspective Reasoning Problem
+
+- **ID:** `2602.14367` · **primary:** `cs.CL` · **Submitted:** 2026-02-16
+- **Authors:** Shuofei Qiao, Yunxiang Wei, Xuehai Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.14367 · **PDF:** https://arxiv.org/pdf/2602.14367v1
+
+## 12111. Image-based Joint-level Detection for Inflammation in Rheumatoid Arthritis from Small and Imbalanced Data
+
+- **ID:** `2602.14365` · **primary:** `cs.CV` · **Submitted:** 2026-02-16
+- **Authors:** Shun Kato, Yasushi Kondo, Shuntaro Saito, et al.
+- **abs:** https://arxiv.org/abs/2602.14365 · **PDF:** https://arxiv.org/pdf/2602.14365v1
+
+## 12112. A Trajectory-Based Safety Audit of Clawdbot (OpenClaw)
+
+- **ID:** `2602.14364` · **primary:** `cs.CR` · **Submitted:** 2026-02-16
+- **Authors:** Tianyu Chen, Dongrui Liu, Xia Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.14364 · **PDF:** https://arxiv.org/pdf/2602.14364v1
+
+## 12113. AdaptManip: Learning Adaptive Whole-Body Object Lifting and Delivery with Online Recurrent State Estimation
+
+- **ID:** `2602.14363` · **primary:** `cs.RO` · **Submitted:** 2026-02-16
+- **Authors:** Morgan Byrd, Donghoon Baek, Kartik Garg, et al.
+- **abs:** https://arxiv.org/abs/2602.14363 · **PDF:** https://arxiv.org/pdf/2602.14363v1
+
+## 12114. High Precision Audience Expansion via Extreme Classification in a Two-Sided Marketplace
+
+- **ID:** `2602.14358` · **primary:** `cs.IR` · **Submitted:** 2026-02-16
+- **Authors:** Dillon Davis, Huiji Gao, Thomas Legrand, et al.
+- **abs:** https://arxiv.org/abs/2602.14358 · **PDF:** https://arxiv.org/pdf/2602.14358v1
+
+## 12115. Key Considerations for Domain Expert Involvement in LLM Design and Evaluation: An Ethnographic Study
+
+- **ID:** `2602.14357` · **primary:** `cs.HC` · **Submitted:** 2026-02-16
+- **Authors:** Annalisa Szymanski, Oghenemaro Anuyah, Toby Jia-Jun Li, et al.
+- **abs:** https://arxiv.org/abs/2602.14357 · **PDF:** https://arxiv.org/pdf/2602.14357v1
+
+## 12116. WIMLE: Uncertainty-Aware World Models with IMLE for Sample-Efficient Continuous Control
+
+- **ID:** `2602.14351` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Mehran Aghabozorgi, Alireza Moazeni, Yanshu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.14351 · **PDF:** https://arxiv.org/pdf/2602.14351v1
+
+## 12117. AXE: An Agentic eXploit Engine for Confirming Zero-Day Vulnerability Reports
+
+- **ID:** `2602.14345` · **primary:** `cs.CR` · **Submitted:** 2026-02-15
+- **Authors:** Amirali Sajadi, Tu Nguyen, Kostadin Damevski, et al.
+- **abs:** https://arxiv.org/abs/2602.14345 · **PDF:** https://arxiv.org/pdf/2602.14345v1
+
+## 12118. Zero-Shot Instruction Following in RL via Structured LTL Representations
+
+- **ID:** `2602.14344` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Mathias Jackermeier, Mattia Giuri, Jacques Cloete, et al.
+- **abs:** https://arxiv.org/abs/2602.14344 · **PDF:** https://arxiv.org/pdf/2602.14344v1
+
+## 12119. High-accuracy log-concave sampling with stochastic queries
+
+- **ID:** `2602.14342` · **primary:** `math.ST` · **Submitted:** 2026-02-15
+- **Authors:** Fan Chen, Sinho Chewi, Constantinos Daskalakis, et al.
+- **abs:** https://arxiv.org/abs/2602.14342 · **PDF:** https://arxiv.org/pdf/2602.14342v1
+
+## 12120. Train Less, Learn More: Adaptive Efficient Rollout Optimization for Group-Based Reinforcement Learning
+
+- **ID:** `2602.14338` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Zhi Zhang, Zhen Han, Costas Mavromatis, et al.
+- **abs:** https://arxiv.org/abs/2602.14338 · **PDF:** https://arxiv.org/pdf/2602.14338v1
+
+## 12121. Automated evaluation of LLMs for effective machine translation of Mandarin Chinese to English
+
+- **ID:** `2603.09998` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Yue Zhang, Rodney Beard, John Hawkins, et al.
+- **abs:** https://arxiv.org/abs/2603.09998 · **PDF:** https://arxiv.org/pdf/2603.09998v1
+
+## 12122. Conformal Signal Temporal Logic for Robust Reinforcement Learning Control: A Case Study
+
+- **ID:** `2602.14322` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Hani Beirami, M M Manjurul Islam
+- **abs:** https://arxiv.org/abs/2602.14322 · **PDF:** https://arxiv.org/pdf/2602.14322v2
+
+## 12123. Offline Learning of Nash Stable Coalition Structures with Possibly Overlapping Coalitions
+
+- **ID:** `2602.14321` · **primary:** `cs.GT` · **Submitted:** 2026-02-15
+- **Authors:** Saar Cohen
+- **abs:** https://arxiv.org/abs/2602.14321 · **PDF:** https://arxiv.org/pdf/2602.14321v1
+
+## 12124. In Transformer We Trust? A Perspective on Transformer Architecture Failure Modes
+
+- **ID:** `2602.14318` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Trishit Mondal, Ameya D. Jagtap
+- **abs:** https://arxiv.org/abs/2602.14318 · **PDF:** https://arxiv.org/pdf/2602.14318v1
+
+## 12125. Benchmarking at the Edge of Comprehension
+
+- **ID:** `2602.14307` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Samuele Marro, Jialin Yu, Emanuele La Malfa, et al.
+- **abs:** https://arxiv.org/abs/2602.14307 · **PDF:** https://arxiv.org/pdf/2602.14307v2
+
+## 12126. Floe: Federated Specialization for Real-Time LLM-SLM Inference
+
+- **ID:** `2602.14302` · **primary:** `cs.DC` · **Submitted:** 2026-02-15
+- **Authors:** Chunlin Tian, Kahou Tam, Yebo Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.14302 · **PDF:** https://arxiv.org/pdf/2602.14302v1
+
+## 12127. DeepFusion: Accelerating MoE Training via Federated Knowledge Distillation from Heterogeneous Edge Devices
+
+- **ID:** `2602.14301` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Songyuan Li, Jia Hu, Ahmed M. Abdelmoniem, et al.
+- **abs:** https://arxiv.org/abs/2602.14301 · **PDF:** https://arxiv.org/pdf/2602.14301v1
+
+## 12128. Does Socialization Emerge in AI Agent Society? A Case Study of Moltbook
+
+- **ID:** `2602.14299` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Ming Li, Xirui Li, Tianyi Zhou
+- **abs:** https://arxiv.org/abs/2602.14299 · **PDF:** https://arxiv.org/pdf/2602.14299v2
+
+## 12129. AutoWebWorld: Synthesizing Infinite Verifiable Web Environments via Finite State Machines
+
+- **ID:** `2602.14296` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Yifan Wu, Yiran Peng, Yiyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.14296 · **PDF:** https://arxiv.org/pdf/2602.14296v1
+
+## 12130. Machine Learning as a Tool (MLAT): A Framework for Integrating Statistical ML Models as Callable Tools within LLM Agent Workflows
+
+- **ID:** `2602.14295` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Edwin Chen, Zulekha Bibi
+- **abs:** https://arxiv.org/abs/2602.14295 · **PDF:** https://arxiv.org/pdf/2602.14295v1
+
+## 12131. KernelBlaster: Continual Cross-Task CUDA Optimization via Memory-Augmented In-Context Reinforcement Learning
+
+- **ID:** `2602.14293` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Kris Shengjun Dong, Sahil Modi, Dima Nikiforov, et al.
+- **abs:** https://arxiv.org/abs/2602.14293 · **PDF:** https://arxiv.org/pdf/2602.14293v1
+
+## 12132. Online LLM watermark detection via e-processes
+
+- **ID:** `2602.14286` · **primary:** `stat.ME` · **Submitted:** 2026-02-15
+- **Authors:** Weijie Su, Ruodu Wang, Zinan Zhao
+- **abs:** https://arxiv.org/abs/2602.14286 · **PDF:** https://arxiv.org/pdf/2602.14286v3
+
+## 12133. FMMD: A multimodal open peer review dataset based on F1000Research
+
+- **ID:** `2602.14285` · **primary:** `cs.DL` · **Submitted:** 2026-02-15
+- **Authors:** Zhenzhen Zhuang, Yuqing Fu, Jing Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.14285 · **PDF:** https://arxiv.org/pdf/2602.14285v1
+
+## 12134. MILD: Multi-Intent Learning and Disambiguation for Proactive Failure Prediction in Intent-based Networking
+
+- **ID:** `2602.14283` · **primary:** `cs.NI` · **Submitted:** 2026-02-15
+- **Authors:** Md. Kamrul Hossain, Walid Aljoby
+- **abs:** https://arxiv.org/abs/2602.14283 · **PDF:** https://arxiv.org/pdf/2602.14283v1
+
+## 12135. MCPShield: A Security Cognition Layer for Adaptive Trust Calibration in Model Context Protocol Agents
+
+- **ID:** `2602.14281` · **primary:** `cs.CR` · **Submitted:** 2026-02-15
+- **Authors:** Zhenhong Zhou, Yuanhe Zhang, Hongwei Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.14281 · **PDF:** https://arxiv.org/pdf/2602.14281v3
+
+## 12136. Fast Compute for ML Optimization
+
+- **ID:** `2602.14280` · **primary:** `stat.CO` · **Submitted:** 2026-02-15
+- **Authors:** Nick Polson, Vadim Sokolov
+- **abs:** https://arxiv.org/abs/2602.14280 · **PDF:** https://arxiv.org/pdf/2602.14280v1
+
+## 12137. Whom to Query for What: Adaptive Group Elicitation via Multi-Turn LLM Interactions
+
+- **ID:** `2602.14279` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Ruomeng Ding, Tianwei Gao, Thomas P. Zollo, et al.
+- **abs:** https://arxiv.org/abs/2602.14279 · **PDF:** https://arxiv.org/pdf/2602.14279v1
+
+## 12138. Reverse N-Wise Output-Oriented Testing for AI/ML and Quantum Computing Systems
+
+- **ID:** `2602.14275` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Lamine Rihani
+- **abs:** https://arxiv.org/abs/2602.14275 · **PDF:** https://arxiv.org/pdf/2602.14275v1
+
+## 12139. Integrating Unstructured Text into Causal Inference: Empirical Evidence from Real Data
+
+- **ID:** `2602.14274` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Boning Zhou, Ziyu Wang, Han Hong, et al.
+- **abs:** https://arxiv.org/abs/2602.14274 · **PDF:** https://arxiv.org/pdf/2602.14274v1
+
+## 12140. Radial-VCReg: More Informative Representation Learning Through Radial Gaussianization
+
+- **ID:** `2602.14272` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Yilun Kuang, Yash Dagade, Deep Chakraborty, et al.
+- **abs:** https://arxiv.org/abs/2602.14272 · **PDF:** https://arxiv.org/pdf/2602.14272v1
+
+## 12141. A Rational Analysis of the Effects of Sycophantic AI
+
+- **ID:** `2602.14270` · **primary:** `cs.CY` · **Submitted:** 2026-02-15
+- **Authors:** Rafael M. Batista, Thomas L. Griffiths
+- **abs:** https://arxiv.org/abs/2602.14270 · **PDF:** https://arxiv.org/pdf/2602.14270v1
+
+## 12142. Cross-household Transfer Learning Approach with LSTM-based Demand Forecasting
+
+- **ID:** `2602.14267` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Manal Rahal, Bestoun S. Ahmed, Roger Renström, et al.
+- **abs:** https://arxiv.org/abs/2602.14267 · **PDF:** https://arxiv.org/pdf/2602.14267v1
+
+## 12143. STATe-of-Thoughts: Structured Action Templates for Tree-of-Thoughts
+
+- **ID:** `2602.14265` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Zachary Bamberger, Till R. Saenger, Gilad Morad, et al.
+- **abs:** https://arxiv.org/abs/2602.14265 · **PDF:** https://arxiv.org/pdf/2602.14265v2
+
+## 12144. Ethical and Explainable AI in Reusable MLOps Pipelines
+
+- **ID:** `2603.03341` · **primary:** `cs.CY` · **Submitted:** 2026-02-15
+- **Authors:** Rakib Hossain, Mahmood Menon Khan, Lisan Al Amin, et al.
+- **abs:** https://arxiv.org/abs/2603.03341 · **PDF:** https://arxiv.org/pdf/2603.03341v1
+
+## 12145. Detecting LLM Hallucinations via Embedding Cluster Geometry: A Three-Type Taxonomy with Measurable Signatures
+
+- **ID:** `2602.14259` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Matic Korun
+- **abs:** https://arxiv.org/abs/2602.14259 · **PDF:** https://arxiv.org/pdf/2602.14259v1
+
+## 12146. AD-Bench: A Real-World, Trajectory-Aware Advertising Analytics Benchmark for LLM Agents
+
+- **ID:** `2602.14257` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Lingxiang Hu, Yiding Sun, Tianle Xia, et al.
+- **abs:** https://arxiv.org/abs/2602.14257 · **PDF:** https://arxiv.org/pdf/2602.14257v1
+
+## 12147. GRAIL: Goal Recognition Alignment through Imitation Learning
+
+- **ID:** `2602.14252` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Osher Elhadad, Felipe Meneguzzi, Reuth Mirsky
+- **abs:** https://arxiv.org/abs/2602.14252 · **PDF:** https://arxiv.org/pdf/2602.14252v1
+
+## 12148. Multi-Agent Debate: A Unified Agentic Framework for Tabular Anomaly Detection
+
+- **ID:** `2602.14251` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Pinqiao Wang, Sheng Li
+- **abs:** https://arxiv.org/abs/2602.14251 · **PDF:** https://arxiv.org/pdf/2602.14251v1
+
+## 12149. Energy-Efficient Over-the-Air Federated Learning via Pinching Antenna Systems
+
+- **ID:** `2602.14250` · **primary:** `cs.IT` · **Submitted:** 2026-02-15
+- **Authors:** Saba Asaad, Ali Bereyhi
+- **abs:** https://arxiv.org/abs/2602.14250 · **PDF:** https://arxiv.org/pdf/2602.14250v1
+
+## 12150. Federated Ensemble Learning with Progressive Model Personalization
+
+- **ID:** `2602.14244` · **primary:** `stat.ML` · **Submitted:** 2026-02-15
+- **Authors:** Ala Emrani, Amir Najafi, Abolfazl Motahari
+- **abs:** https://arxiv.org/abs/2602.14244 · **PDF:** https://arxiv.org/pdf/2602.14244v1
+
+## 12151. GRAFNet: Multiscale Retinal Processing via Guided Cortical Attention Feedback for Enhancing Medical Image Polyp Segmentation
+
+- **ID:** `2602.15072` · **primary:** `cs.CV` · **Submitted:** 2026-02-15
+- **Authors:** Abdul Joseph Fofanah, Lian Wen, Alpha Alimamy Kamara, et al.
+- **abs:** https://arxiv.org/abs/2602.15072 · **PDF:** https://arxiv.org/pdf/2602.15072v1
+
+## 12152. A Hybrid TGN-SEAL Model for Dynamic Graph Link Prediction
+
+- **ID:** `2602.14239` · **primary:** `cs.SI` · **Submitted:** 2026-02-15
+- **Authors:** Nafiseh Sadat Sajadi, Behnam Bahrak, Mahdi Jafari Siavoshani
+- **abs:** https://arxiv.org/abs/2602.14239 · **PDF:** https://arxiv.org/pdf/2602.14239v1
+
+## 12153. We can still parse using syntactic rules
+
+- **ID:** `2602.14238` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Ghaly Hussein
+- **abs:** https://arxiv.org/abs/2602.14238 · **PDF:** https://arxiv.org/pdf/2602.14238v1
+
+## 12154. AbracADDbra: Touch-Guided Object Addition by Decoupling Placement and Editing Subtasks
+
+- **ID:** `2602.14237` · **primary:** `cs.CV` · **Submitted:** 2026-02-15
+- **Authors:** Kunal Swami, Raghu Chittersu, Yuvraj Rathore, et al.
+- **abs:** https://arxiv.org/abs/2602.14237 · **PDF:** https://arxiv.org/pdf/2602.14237v1
+
+## 12155. Dual-Signal Adaptive KV-Cache Optimization for Long-Form Video Understanding in Vision-Language Models
+
+- **ID:** `2602.14236` · **primary:** `cs.CV` · **Submitted:** 2026-02-15
+- **Authors:** Vishnu Sai, Dheeraj Sai, Srinath B, et al.
+- **abs:** https://arxiv.org/abs/2602.14236 · **PDF:** https://arxiv.org/pdf/2602.14236v1
+
+## 12156. REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents
+
+- **ID:** `2602.14234` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Zheng Chu, Xiao Wang, Jack Hong, et al.
+- **abs:** https://arxiv.org/abs/2602.14234 · **PDF:** https://arxiv.org/pdf/2602.14234v1
+
+## 12157. Evaluating LLMs in Finance Requires Explicit Bias Consideration
+
+- **ID:** `2602.14233` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Yaxuan Kong, Hoyoung Lee, Yoontae Hwang, et al.
+- **abs:** https://arxiv.org/abs/2602.14233 · **PDF:** https://arxiv.org/pdf/2602.14233v1
+
+## 12158. Robust multi-task boosting using clustering and local ensembling
+
+- **ID:** `2602.14231` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Seyedsaman Emami, Daniel Hernández-Lobato, Gonzalo Martínez-Muñoz
+- **abs:** https://arxiv.org/abs/2602.14231 · **PDF:** https://arxiv.org/pdf/2602.14231v1
+
+## 12159. CORPGEN: Simulating Corporate Environments with Autonomous Digital Employees in Multi-Horizon Task Environments
+
+- **ID:** `2602.14229` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Abubakarr Jaye, Nigel Boachie Kumankumah, Chidera Biringa, et al.
+- **abs:** https://arxiv.org/abs/2602.14229 · **PDF:** https://arxiv.org/pdf/2602.14229v1
+
+## 12160. Text Before Vision: Staged Knowledge Injection Matters for Agentic RLVR in Ultra-High-Resolution Remote Sensing Understanding
+
+- **ID:** `2602.14225` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Fengxiang Wang, Mingshuo Chen, Yueying Li, et al.
+- **abs:** https://arxiv.org/abs/2602.14225 · **PDF:** https://arxiv.org/pdf/2602.14225v1
+
+## 12161. The Interspeech 2026 Audio Reasoning Challenge: Evaluating Reasoning Process Quality for Audio Reasoning Models and Agents
+
+- **ID:** `2602.14224` · **primary:** `cs.SD` · **Submitted:** 2026-02-15
+- **Authors:** Ziyang Ma, Ruiyang Xu, Yinghao Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.14224 · **PDF:** https://arxiv.org/pdf/2602.14224v1
+
+## 12162. Reasoning Language Models for complex assessments tasks: Evaluating parental cooperation from child protection case reports
+
+- **ID:** `2602.14216` · **primary:** `cs.CY` · **Submitted:** 2026-02-15
+- **Authors:** Dragan Stoll, Brian E. Perron, Zia Qi, et al.
+- **abs:** https://arxiv.org/abs/2602.14216 · **PDF:** https://arxiv.org/pdf/2602.14216v1
+
+## 12163. SkillJect: Automating Stealthy Skill-Based Prompt Injection for Coding Agents with Trace-Driven Closed-Loop Refinement
+
+- **ID:** `2602.14211` · **primary:** `cs.CR` · **Submitted:** 2026-02-15
+- **Authors:** Xiaojun Jia, Jie Liao, Simeng Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.14211 · **PDF:** https://arxiv.org/pdf/2602.14211v1
+
+## 12164. MAGE: All-[MASK] Block Already Knows Where to Look in Diffusion LLM
+
+- **ID:** `2602.14209` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Omin Kwon, Yeonjae Kim, Doyeon Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.14209 · **PDF:** https://arxiv.org/pdf/2602.14209v1
+
+## 12165. Fast Catch-Up, Late Switching: Optimal Batch Size Scheduling via Functional Scaling Laws
+
+- **ID:** `2602.14208` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Jinbo Wang, Binghui Li, Zhanpeng Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.14208 · **PDF:** https://arxiv.org/pdf/2602.14208v2
+
+## 12166. GeoEyes: On-Demand Visual Focusing for Evidence-Grounded Understanding of Ultra-High-Resolution Remote Sensing Imagery
+
+- **ID:** `2602.14201` · **primary:** `cs.CV` · **Submitted:** 2026-02-15
+- **Authors:** Fengxiang Wang, Mingshuo Chen, Yueying Li, et al.
+- **abs:** https://arxiv.org/abs/2602.14201 · **PDF:** https://arxiv.org/pdf/2602.14201v2
+
+## 12167. TS-Haystack: A Multi-Scale Retrieval Benchmark for Time Series Language Models
+
+- **ID:** `2602.14200` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Nicolas Zumarraga, Thomas Kaar, Ning Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.14200 · **PDF:** https://arxiv.org/pdf/2602.14200v3
+
+## 12168. Learning Part-Aware Dense 3D Feature Field for Generalizable Articulated Object Manipulation
+
+- **ID:** `2602.14193` · **primary:** `cs.RO` · **Submitted:** 2026-02-15
+- **Authors:** Yue Chen, Muqing Jiang, Kaifeng Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.14193 · **PDF:** https://arxiv.org/pdf/2602.14193v1
+
+## 12169. Knowing When Not to Answer: Abstention-Aware Scientific Reasoning
+
+- **ID:** `2602.14189` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Samir Abdaljalil, Erchin Serpedin, Hasan Kurban
+- **abs:** https://arxiv.org/abs/2602.14189 · **PDF:** https://arxiv.org/pdf/2602.14189v1
+
+## 12170. GPT-5 vs Other LLMs in Long Short-Context Performance
+
+- **ID:** `2602.14188` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Nima Esmi, Maryam Nezhad-Moghaddam, Fatemeh Borhani, et al.
+- **abs:** https://arxiv.org/abs/2602.14188 · **PDF:** https://arxiv.org/pdf/2602.14188v1
+
+## 12171. UniWeTok: An Unified Binary Tokenizer with Codebook Size $\mathit{2^{128}}$ for Unified Multimodal Large Language Model
+
+- **ID:** `2602.14178` · **primary:** `cs.CV` · **Submitted:** 2026-02-15
+- **Authors:** Shaobin Zhuang, Yuang Ai, Jiaming Han, et al.
+- **abs:** https://arxiv.org/abs/2602.14178 · **PDF:** https://arxiv.org/pdf/2602.14178v3
+
+## 12172. Towards Spatial Transcriptomics-driven Pathology Foundation Models
+
+- **ID:** `2602.14177` · **primary:** `cs.CV` · **Submitted:** 2026-02-15
+- **Authors:** Konstantin Hemker, Andrew H. Song, Cristina Almagro-Pérez, et al.
+- **abs:** https://arxiv.org/abs/2602.14177 · **PDF:** https://arxiv.org/pdf/2602.14177v1
+
+## 12173. Investigation for Relative Voice Impression Estimation
+
+- **ID:** `2602.14172` · **primary:** `cs.SD` · **Submitted:** 2026-02-15
+- **Authors:** Kenichi Fujita, Yusuke Ijima
+- **abs:** https://arxiv.org/abs/2602.14172 · **PDF:** https://arxiv.org/pdf/2602.14172v2
+
+## 12174. Deep Dense Exploration for LLM Reinforcement Learning via Pivot-Driven Resampling
+
+- **ID:** `2602.14169` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Yiran Guo, Zhongjian Qiao, Yingqi Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.14169 · **PDF:** https://arxiv.org/pdf/2602.14169v1
+
+## 12175. Index Light, Reason Deep: Deferred Visual Ingestion for Visual-Dense Document Question Answering
+
+- **ID:** `2602.14162` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Tao Xu
+- **abs:** https://arxiv.org/abs/2602.14162 · **PDF:** https://arxiv.org/pdf/2602.14162v2
+
+## 12176. When Benchmarks Lie: Evaluating Malicious Prompt Classifiers Under True Distribution Shift
+
+- **ID:** `2602.14161` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Max Fomin
+- **abs:** https://arxiv.org/abs/2602.14161 · **PDF:** https://arxiv.org/pdf/2602.14161v1
+
+## 12177. Process-Supervised Multi-Agent Reinforcement Learning for Reliable Clinical Reasoning
+
+- **ID:** `2602.14160` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Chaeeun Lee, T. Michael Yates, Pasquale Minervini, et al.
+- **abs:** https://arxiv.org/abs/2602.14160 · **PDF:** https://arxiv.org/pdf/2602.14160v1
+
+## 12178. Synergistic Intra- and Cross-Layer Regularization Losses for MoE Expert Specialization
+
+- **ID:** `2602.14159` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Rizhen Hu, Yuan Cao, Boao Kong, et al.
+- **abs:** https://arxiv.org/abs/2602.14159 · **PDF:** https://arxiv.org/pdf/2602.14159v1
+
+## 12179. A Multi-Agent Framework for Medical AI: Leveraging Fine-Tuned GPT, LLaMA, and DeepSeek R1 for Evidence-Based and Bias-Aware Clinical Query Processing
+
+- **ID:** `2602.14158` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Naeimeh Nourmohammadi, Md Meem Hossain, The Anh Han, et al.
+- **abs:** https://arxiv.org/abs/2602.14158 · **PDF:** https://arxiv.org/pdf/2602.14158v1
+
+## 12180. When Test-Time Guidance Is Enough: Fast Image and Video Editing with Diffusion Guidance
+
+- **ID:** `2602.14157` · **primary:** `cs.CV` · **Submitted:** 2026-02-15
+- **Authors:** Ahmed Ghorbel, Badr Moufad, Navid Bagheri Shouraki, et al.
+- **abs:** https://arxiv.org/abs/2602.14157 · **PDF:** https://arxiv.org/pdf/2602.14157v2
+
+## 12181. A Penalty Approach for Differentiation Through Black-Box Quadratic Programming Solvers
+
+- **ID:** `2602.14154` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Yuxuan Linghu, Zhiyuan Liu, Qi Deng
+- **abs:** https://arxiv.org/abs/2602.14154 · **PDF:** https://arxiv.org/pdf/2602.14154v2
+
+## 12182. Empathy Is Not What Changed: Clinical Assessment of Psychological Safety Across GPT Model Generations
+
+- **ID:** `2603.09997` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Michael Keeman, Anastasia Keeman
+- **abs:** https://arxiv.org/abs/2603.09997 · **PDF:** https://arxiv.org/pdf/2603.09997v1
+
+## 12183. ROAST: Rollout-based On-distribution Activation Steering Technique
+
+- **ID:** `2602.14143` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Xuanbo Su, Hao Luo, Yingfang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.14143 · **PDF:** https://arxiv.org/pdf/2602.14143v1
+
+## 12184. Detection of On-Ground Chestnuts Using Artificial Intelligence Toward Automated Picking
+
+- **ID:** `2602.14140` · **primary:** `cs.CV` · **Submitted:** 2026-02-15
+- **Authors:** Kaixuan Fang, Yuzhen Lu, Xinyang Mu
+- **abs:** https://arxiv.org/abs/2602.14140 · **PDF:** https://arxiv.org/pdf/2602.14140v1
+
+## 12185. ForesightSafety Bench: A Frontier Risk Evaluation and Governance Framework towards Safe AI
+
+- **ID:** `2602.14135` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Haibo Tong, Feifei Zhao, Linghao Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.14135 · **PDF:** https://arxiv.org/pdf/2602.14135v4
+
+## 12186. DenseMLLM: Standard Multimodal LLMs are Intrinsic Dense Predictors
+
+- **ID:** `2602.14134` · **primary:** `cs.CV` · **Submitted:** 2026-02-15
+- **Authors:** Yi Li, Hongze Shen, Lexiang Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.14134 · **PDF:** https://arxiv.org/pdf/2602.14134v1
+
+## 12187. Algebraic Quantum Intelligence: A New Framework for Reproducible Machine Creativity
+
+- **ID:** `2602.14130` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Kazuo Yano, Jonghyeok Lee, Tae Ishitomi, et al.
+- **abs:** https://arxiv.org/abs/2602.14130 · **PDF:** https://arxiv.org/pdf/2602.14130v1
+
+## 12188. Toward Autonomous O-RAN: A Multi-Scale Agentic AI Framework for Real-Time Network Control and Management
+
+- **ID:** `2602.14117` · **primary:** `cs.NI` · **Submitted:** 2026-02-15
+- **Authors:** Hojjat Navidan, Mohammad Cheraghinia, Jaron Fontaine, et al.
+- **abs:** https://arxiv.org/abs/2602.14117 · **PDF:** https://arxiv.org/pdf/2602.14117v1
+
+## 12189. PIPE-RDF: An LLM-Assisted Pipeline for Enterprise RDF Benchmarking
+
+- **ID:** `2602.18497` · **primary:** `cs.DB` · **Submitted:** 2026-02-15
+- **Authors:** Suraj Ranganath
+- **abs:** https://arxiv.org/abs/2602.18497 · **PDF:** https://arxiv.org/pdf/2602.18497v1
+
+## 12190. Sanity Checks for Sparse Autoencoders: Do SAEs Beat Random Baselines?
+
+- **ID:** `2602.14111` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Anton Korznikov, Andrey Galichin, Alexey Dontsov, et al.
+- **abs:** https://arxiv.org/abs/2602.14111 · **PDF:** https://arxiv.org/pdf/2602.14111v1
+
+## 12191. Geometry-Aware Physics-Informed PointNets for Modeling Flows Across Porous Structures
+
+- **ID:** `2602.14108` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Luigi Ciceri, Corrado Mio, Jianyi Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.14108 · **PDF:** https://arxiv.org/pdf/2602.14108v1
+
+## 12192. Anticipating Adversary Behavior in DevSecOps Scenarios through Large Language Models
+
+- **ID:** `2602.14106` · **primary:** `cs.CR` · **Submitted:** 2026-02-15
+- **Authors:** Mario Marín Caballero, Miguel Betancourt Alonso, Daniel Díaz-López, et al.
+- **abs:** https://arxiv.org/abs/2602.14106 · **PDF:** https://arxiv.org/pdf/2602.14106v1
+
+## 12193. Character-aware Transformers Learn an Irregular Morphological Pattern Yet None Generalize Like Humans
+
+- **ID:** `2602.14100` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Akhilesh Kakolu Ramarao, Kevin Tang, Dinah Baer-Henney
+- **abs:** https://arxiv.org/abs/2602.14100 · **PDF:** https://arxiv.org/pdf/2602.14100v1
+
+## 12194. SemanticFeels: Semantic Labeling during In-Hand Manipulation
+
+- **ID:** `2602.14099` · **primary:** `cs.RO` · **Submitted:** 2026-02-15
+- **Authors:** Anas Al Shikh Khalil, Haozhi Qi, Roberto Calandra
+- **abs:** https://arxiv.org/abs/2602.14099 · **PDF:** https://arxiv.org/pdf/2602.14099v1
+
+## 12195. NEST: Nascent Encoded Steganographic Thoughts
+
+- **ID:** `2602.14095` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Artem Karpov
+- **abs:** https://arxiv.org/abs/2602.14095 · **PDF:** https://arxiv.org/pdf/2602.14095v1
+
+## 12196. GUI-GENESIS: Automated Synthesis of Efficient Environments with Verifiable Rewards for GUI Agent Post-Training
+
+- **ID:** `2602.14093` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Yuan Cao, Dezhi Ran, Mengzhou Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.14093 · **PDF:** https://arxiv.org/pdf/2602.14093v1
+
+## 12197. TabTracer: Monte Carlo Tree Search for Complex Table Reasoning with Large Language Models
+
+- **ID:** `2602.14089` · **primary:** `cs.DB` · **Submitted:** 2026-02-15
+- **Authors:** Zhizhao Luo, Zhaojing Luo, Meihui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.14089 · **PDF:** https://arxiv.org/pdf/2602.14089v1
+
+## 12198. Neural Optimal Transport in Hilbert Spaces: Characterizing Spurious Solutions and Gaussian Smoothing
+
+- **ID:** `2602.14086` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Jae-Hwan Choi, Jiwoo Yoon, Dohyun Kwon, et al.
+- **abs:** https://arxiv.org/abs/2602.14086 · **PDF:** https://arxiv.org/pdf/2602.14086v1
+
+## 12199. Plan-MCTS: Plan Exploration for Action Exploitation in Web Navigation
+
+- **ID:** `2602.14083` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Weiming Zhang, Jihong Wang, Jiamu Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.14083 · **PDF:** https://arxiv.org/pdf/2602.14083v1
+
+## 12200. CCiV: A Benchmark for Structure, Rhythm and Quality in LLM-Generated Chinese \textit{Ci} Poetry
+
+- **ID:** `2602.14081` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Shangqing Zhao, Yupei Ren, Yuhao Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.14081 · **PDF:** https://arxiv.org/pdf/2602.14081v1
+
+## 12201. Empty Shelves or Lost Keys? Recall Is the Bottleneck for Parametric Factuality
+
+- **ID:** `2602.14080` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Nitay Calderon, Eyal Ben-David, Zorik Gekhman, et al.
+- **abs:** https://arxiv.org/abs/2602.14080 · **PDF:** https://arxiv.org/pdf/2602.14080v1
+
+## 12202. Policy Gradient with Adaptive Entropy Annealing for Continual Fine-Tuning
+
+- **ID:** `2602.14078` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Yaqian Zhang, Bernhard Pfahringer, Eibe Frank, et al.
+- **abs:** https://arxiv.org/abs/2602.14078 · **PDF:** https://arxiv.org/pdf/2602.14078v1
+
+## 12203. GTS: Inference-Time Scaling of Latent Reasoning with a Learnable Gaussian Thought Sampler
+
+- **ID:** `2602.14077` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Minghan Wang, Ye Bai, Thuy-Trang Vu, et al.
+- **abs:** https://arxiv.org/abs/2602.14077 · **PDF:** https://arxiv.org/pdf/2602.14077v2
+
+## 12204. Annotation-Efficient Vision-Language Model Adaptation to the Polish Language Using the LLaVA Framework
+
+- **ID:** `2602.14073` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Grzegorz Statkiewicz, Alicja Dobrzeniecka, Karolina Seweryn, et al.
+- **abs:** https://arxiv.org/abs/2602.14073 · **PDF:** https://arxiv.org/pdf/2602.14073v2
+
+## 12205. There Are No Silly Questions: Evaluation of Offline LLM Capabilities from a Turkish Perspective
+
+- **ID:** `2603.09996` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Edibe Yilmaz, Kahraman Kostas
+- **abs:** https://arxiv.org/abs/2603.09996 · **PDF:** https://arxiv.org/pdf/2603.09996v1
+
+## 12206. Open Rubric System: Scaling Reinforcement Learning with Pairwise Adaptive Rubric
+
+- **ID:** `2602.14069` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Ruipeng Jia, Yunyi Yang, Yuxin Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.14069 · **PDF:** https://arxiv.org/pdf/2602.14069v2
+
+## 12207. REAL: Resolving Knowledge Conflicts in Knowledge-Intensive Visual Question Answering via Reasoning-Pivot Alignment
+
+- **ID:** `2602.14065` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Kai Ye, Xianwei Mao, Sheng Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.14065 · **PDF:** https://arxiv.org/pdf/2602.14065v1
+
+## 12208. From Scarcity to Scale: A Release-Level Analysis of the Pashto Common Voice Dataset
+
+- **ID:** `2602.14062` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Jandad Jahani, Mursal Dawodi, Jawid Ahmad Baktash
+- **abs:** https://arxiv.org/abs/2602.14062 · **PDF:** https://arxiv.org/pdf/2602.14062v1
+
+## 12209. LM-Lexicon: Improving Definition Modeling via Harmonizing Semantic Experts
+
+- **ID:** `2602.14060` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Yang Liu, Jiaye Yang, Weikang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.14060 · **PDF:** https://arxiv.org/pdf/2602.14060v1
+
+## 12210. LogitsCoder: Towards Efficient Chain-of-Thought Path Search via Logits Preference Decoding for Code Generation
+
+- **ID:** `2602.14054` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Jizheng Chen, Weiming Zhang, Xinyi Dai, et al.
+- **abs:** https://arxiv.org/abs/2602.14054 · **PDF:** https://arxiv.org/pdf/2602.14054v1
+
+## 12211. Decentralized Federated Learning With Energy Harvesting Devices
+
+- **ID:** `2602.14051` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Kai Zhang, Xuanyu Cao, Khaled B. Letaief
+- **abs:** https://arxiv.org/abs/2602.14051 · **PDF:** https://arxiv.org/pdf/2602.14051v1
+
+## 12212. Position Encoding with Random Float Sampling Enhances Length Generalization of Transformers
+
+- **ID:** `2602.14050` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Atsushi Shimizu, Shohei Taniguchi, Yutaka Matsuo
+- **abs:** https://arxiv.org/abs/2602.14050 · **PDF:** https://arxiv.org/pdf/2602.14050v1
+
+## 12213. UniST-Pred: A Robust Unified Framework for Spatio-Temporal Traffic Forecasting in Transportation Networks Under Disruptions
+
+- **ID:** `2602.14049` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Yue Wang, Areg Karapetyan, Djellel Difallah, et al.
+- **abs:** https://arxiv.org/abs/2602.14049 · **PDF:** https://arxiv.org/pdf/2602.14049v1
+
+## 12214. Context Shapes LLMs Retrieval-Augmented Fact-Checking Effectiveness
+
+- **ID:** `2602.14044` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Pietro Bernardelle, Stefano Civelli, Kevin Roitero, et al.
+- **abs:** https://arxiv.org/abs/2602.14044 · **PDF:** https://arxiv.org/pdf/2602.14044v2
+
+## 12215. Beyond Static Snapshots: Dynamic Modeling and Forecasting of Group-Level Value Evolution with Large Language Models
+
+- **ID:** `2602.14043` · **primary:** `cs.SI` · **Submitted:** 2026-02-15
+- **Authors:** Qiankun Pi, Guixin Su, Jinliang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.14043 · **PDF:** https://arxiv.org/pdf/2602.14043v1
+
+## 12216. Restoration Adaptation for Semantic Segmentation on Low Quality Images
+
+- **ID:** `2602.14042` · **primary:** `cs.CV` · **Submitted:** 2026-02-15
+- **Authors:** Kai Guan, Rongyuan Wu, Shuai Li, et al.
+- **abs:** https://arxiv.org/abs/2602.14042 · **PDF:** https://arxiv.org/pdf/2602.14042v1
+
+## 12217. BitDance: Scaling Autoregressive Generative Models with Binary Tokens
+
+- **ID:** `2602.14041` · **primary:** `cs.CV` · **Submitted:** 2026-02-15
+- **Authors:** Yuang Ai, Jiaming Han, Shaobin Zhuang, et al.
+- **abs:** https://arxiv.org/abs/2602.14041 · **PDF:** https://arxiv.org/pdf/2602.14041v2
+
+## 12218. Geometry-Preserving Aggregation for Mixture-of-Experts Embedding Models
+
+- **ID:** `2602.14039` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Sajjad Kachuee, Mohammad Sharifkhani
+- **abs:** https://arxiv.org/abs/2602.14039 · **PDF:** https://arxiv.org/pdf/2602.14039v1
+
+## 12219. Choosing How to Remember: Adaptive Memory Structures for LLM Agents
+
+- **ID:** `2602.14038` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Mingfei Lu, Mengjia Wu, Feng Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.14038 · **PDF:** https://arxiv.org/pdf/2602.14038v1
+
+## 12220. FloCA: Towards Faithful and Logically Consistent Flowchart Reasoning
+
+- **ID:** `2602.14035` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Jinzi Zou, Bolin Wang, Liang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.14035 · **PDF:** https://arxiv.org/pdf/2602.14035v1
+
+## 12221. MC$^2$Mark: Distortion-Free Multi-Bit Watermarking for Long Messages
+
+- **ID:** `2602.14030` · **primary:** `cs.CR` · **Submitted:** 2026-02-15
+- **Authors:** Xuehao Cui, Ruibo Chen, Yihan Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.14030 · **PDF:** https://arxiv.org/pdf/2602.14030v1
+
+## 12222. Why Self-Training Helps and Hurts: Denoising vs. Signal Forgetting
+
+- **ID:** `2602.14029` · **primary:** `stat.ML` · **Submitted:** 2026-02-15
+- **Authors:** Mingqi Wu, Archer Y. Yang, Qiang Sun
+- **abs:** https://arxiv.org/abs/2602.14029 · **PDF:** https://arxiv.org/pdf/2602.14029v1
+
+## 12223. GRRM: Group Relative Reward Modeling for Machine Translation
+
+- **ID:** `2602.14028` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Sen Yang, Shanbo Cheng, Lu Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.14028 · **PDF:** https://arxiv.org/pdf/2602.14028v1
+
+## 12224. EIDOS: Latent-Space Predictive Learning for Time Series Foundation Models
+
+- **ID:** `2602.14024` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Xinxing Zhou, Qingren Yao, Yiji Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.14024 · **PDF:** https://arxiv.org/pdf/2602.14024v1
+
+## 12225. Computable Bernstein Certificates for Cross-Fitted Clipped Covariance Estimation
+
+- **ID:** `2602.14020` · **primary:** `stat.ML` · **Submitted:** 2026-02-15
+- **Authors:** Even He, Zaizai Yan
+- **abs:** https://arxiv.org/abs/2602.14020 · **PDF:** https://arxiv.org/pdf/2602.14020v1
+
+## 12226. S2SServiceBench: A Multimodal Benchmark for Last-Mile S2S Climate Services
+
+- **ID:** `2602.14017` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Chenyue Li, Wen Deng, Zhuotao Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.14017 · **PDF:** https://arxiv.org/pdf/2602.14017v1
+
+## 12227. From SFT to RL: Demystifying the Post-Training Pipeline for LLM-based Vulnerability Detection
+
+- **ID:** `2602.14012` · **primary:** `cs.CR` · **Submitted:** 2026-02-15
+- **Authors:** Youpeng Li, Fuxun Yu, Xinda Wang
+- **abs:** https://arxiv.org/abs/2602.14012 · **PDF:** https://arxiv.org/pdf/2602.14012v1
+
+## 12228. KoopGen: Koopman Generator Networks for Representing and Predicting Dynamical Systems with Continuous Spectra
+
+- **ID:** `2602.14011` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Liangyu Su, Jun Shu, Rui Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.14011 · **PDF:** https://arxiv.org/pdf/2602.14011v1
+
+## 12229. A Deployment-Friendly Foundational Framework for Efficient Computational Pathology
+
+- **ID:** `2602.14010` · **primary:** `cs.CV` · **Submitted:** 2026-02-15
+- **Authors:** Yu Cai, Cheng Jin, Jiabo Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.14010 · **PDF:** https://arxiv.org/pdf/2602.14010v1
+
+## 12230. Named Entity Recognition for Payment Data Using NLP
+
+- **ID:** `2602.14009` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Srikumar Nayak
+- **abs:** https://arxiv.org/abs/2602.14009 · **PDF:** https://arxiv.org/pdf/2602.14009v1
+
+## 12231. Prompt-Driven Low-Altitude Edge Intelligence: Modular Agents and Generative Reasoning
+
+- **ID:** `2602.14003` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Jiahao You, Ziye Jia, Chao Dong, et al.
+- **abs:** https://arxiv.org/abs/2602.14003 · **PDF:** https://arxiv.org/pdf/2602.14003v1
+
+## 12232. The Sufficiency-Conciseness Trade-off in LLM Self-Explanation from an Information Bottleneck Perspective
+
+- **ID:** `2602.14002` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Ali Zahedzadeh, Behnam Bahrak
+- **abs:** https://arxiv.org/abs/2602.14002 · **PDF:** https://arxiv.org/pdf/2602.14002v1
+
+## 12233. Distillation and Interpretability of Ensemble Forecasts of ENSO Phase using Entropic Learning
+
+- **ID:** `2602.16857` · **primary:** `physics.comp-ph` · **Submitted:** 2026-02-15
+- **Authors:** Michael Groom, Davide Bassetti, Illia Horenko, et al.
+- **abs:** https://arxiv.org/abs/2602.16857 · **PDF:** https://arxiv.org/pdf/2602.16857v1
+
+## 12234. Context Over Compute Human-in-the-Loop Outperforms Iterative Chain-of-Thought Prompting in Interview Answer Quality
+
+- **ID:** `2603.09995` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Kewen Zhu, Zixi Liu, Yanjing Li
+- **abs:** https://arxiv.org/abs/2603.09995 · **PDF:** https://arxiv.org/pdf/2603.09995v1
+
+## 12235. Bridging AI and Clinical Reasoning: Abductive Explanations for Alignment on Critical Symptoms
+
+- **ID:** `2602.13985` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Belona Sonna, Alban Grastien
+- **abs:** https://arxiv.org/abs/2602.13985 · **PDF:** https://arxiv.org/pdf/2602.13985v1
+
+## 12236. Cognitive Chunking for Soft Prompts: Accelerating Compressor Learning via Block-wise Causal Masking
+
+- **ID:** `2602.13980` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Guojie Liu, Yiqi Wang, Yanfeng Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.13980 · **PDF:** https://arxiv.org/pdf/2602.13980v1
+
+## 12237. Chain-of-Thought Reasoning with Large Language Models for Clinical Alzheimer's Disease Assessment and Diagnosis
+
+- **ID:** `2602.13979` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Tongze Zhang, Jun-En Ding, Melik Ozolcer, et al.
+- **abs:** https://arxiv.org/abs/2602.13979 · **PDF:** https://arxiv.org/pdf/2602.13979v1
+
+## 12238. WoVR: World Models as Reliable Simulators for Post-Training VLA Policies with RL
+
+- **ID:** `2602.13977` · **primary:** `cs.RO` · **Submitted:** 2026-02-15
+- **Authors:** Zhennan Jiang, Shangqing Zhou, Yutong Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.13977 · **PDF:** https://arxiv.org/pdf/2602.13977v1
+
+## 12239. DAIAN: Deep Adaptive Intent-Aware Network for CTR Prediction in Trigger-Induced Recommendation
+
+- **ID:** `2602.13971` · **primary:** `cs.IR` · **Submitted:** 2026-02-15
+- **Authors:** Zhihao Lv, Longtao Zhang, Ailong He, et al.
+- **abs:** https://arxiv.org/abs/2602.13971 · **PDF:** https://arxiv.org/pdf/2602.13971v1
+
+## 12240. Neuromem: A Granular Decomposition of the Streaming Lifecycle in External Memory for LLMs
+
+- **ID:** `2602.13967` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Ruicheng Zhang, Xinyi Li, Tianyi Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.13967 · **PDF:** https://arxiv.org/pdf/2602.13967v1
+
+## 12241. HLE-Verified: A Systematic Verification and Structured Revision of Humanity's Last Exam
+
+- **ID:** `2602.13964` · **primary:** `cs.CL` · **Submitted:** 2026-02-15
+- **Authors:** Weiqi Zhai, Zhihai Wang, Jinghang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13964 · **PDF:** https://arxiv.org/pdf/2602.13964v3
+
+## 12242. MarsRetrieval: Benchmarking Vision-Language Models for Planetary-Scale Geospatial Retrieval on Mars
+
+- **ID:** `2602.13961` · **primary:** `cs.CV` · **Submitted:** 2026-02-15
+- **Authors:** Shuoyuan Wang, Yiran Wang, Hongxin Wei
+- **abs:** https://arxiv.org/abs/2602.13961 · **PDF:** https://arxiv.org/pdf/2602.13961v1
+
+## 12243. Steady-State Behavior of Constant-Stepsize Stochastic Approximation: Gaussian Approximation and Tail Bounds
+
+- **ID:** `2602.13960` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Zedong Wang, Yuyang Wang, Ijay Narang, et al.
+- **abs:** https://arxiv.org/abs/2602.13960 · **PDF:** https://arxiv.org/pdf/2602.13960v1
+
+## 12244. Chemical Language Models for Natural Products: A State-Space Model Approach
+
+- **ID:** `2602.13958` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Ho-Hsuan Wang, Afnan Sultan, Andrea Volkamer, et al.
+- **abs:** https://arxiv.org/abs/2602.13958 · **PDF:** https://arxiv.org/pdf/2602.13958v1
+
+## 12245. An effective Genetic Programming Hyper-Heuristic for Uncertain Agile Satellite Scheduling
+
+- **ID:** `2602.15070` · **primary:** `cs.NE` · **Submitted:** 2026-02-15
+- **Authors:** Yuning Chen, Junhua Xue, Wangqi Gu, et al.
+- **abs:** https://arxiv.org/abs/2602.15070 · **PDF:** https://arxiv.org/pdf/2602.15070v1
+
+## 12246. Eureka-Audio: Triggering Audio Intelligence in Compact Language Models
+
+- **ID:** `2602.13954` · **primary:** `cs.SD` · **Submitted:** 2026-02-15
+- **Authors:** Dan Zhang, Yishu Lei, Jing Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.13954 · **PDF:** https://arxiv.org/pdf/2602.13954v1
+
+## 12247. Mobile-Agent-v3.5: Multi-platform Fundamental GUI Agents
+
+- **ID:** `2602.16855` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Haiyang Xu, Xi Zhang, Haowei Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.16855 · **PDF:** https://arxiv.org/pdf/2602.16855v1
+
+## 12248. QuRL: Efficient Reinforcement Learning with Quantized Rollout
+
+- **ID:** `2602.13953` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Yuhang Li, Reena Elangovan, Xin Dong, et al.
+- **abs:** https://arxiv.org/abs/2602.13953 · **PDF:** https://arxiv.org/pdf/2602.13953v1
+
+## 12249. FedEMA-Distill: Exponential Moving Average Guided Knowledge Distillation for Robust Federated Learning
+
+- **ID:** `2603.04422` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Hamza Reguieg, Mohamed El Kamili, Essaid Sabir
+- **abs:** https://arxiv.org/abs/2603.04422 · **PDF:** https://arxiv.org/pdf/2603.04422v1
+
+## 12250. FM-RME: Foundation Model Empowered Radio Map Estimation
+
+- **ID:** `2602.22231` · **primary:** `eess.SP` · **Submitted:** 2026-02-15
+- **Authors:** Dong Yang, Yue Wang, Songyang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.22231 · **PDF:** https://arxiv.org/pdf/2602.22231v1
+
+## 12251. Experiential Reinforcement Learning
+
+- **ID:** `2602.13949` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Taiwei Shi, Sihao Chen, Bowen Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.13949 · **PDF:** https://arxiv.org/pdf/2602.13949v1
+
+## 12252. A Theoretical Framework for LLM Fine-tuning Using Early Stopping for Non-random Initialization
+
+- **ID:** `2602.13942` · **primary:** `stat.ML` · **Submitted:** 2026-02-15
+- **Authors:** Zexuan Sun, Garvesh Raskutti
+- **abs:** https://arxiv.org/abs/2602.13942 · **PDF:** https://arxiv.org/pdf/2602.13942v1
+
+## 12253. You Can Learn Tokenization End-to-End with Reinforcement Learning
+
+- **ID:** `2602.13940` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Sam Dauncey, Roger Wattenhofer
+- **abs:** https://arxiv.org/abs/2602.13940 · **PDF:** https://arxiv.org/pdf/2602.13940v1
+
+## 12254. An Adaptive Model Selection Framework for Demand Forecasting under Horizon-Induced Degradation to Support Business Strategy and Operations
+
+- **ID:** `2602.13939` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Adolfo González, Víctor Parada
+- **abs:** https://arxiv.org/abs/2602.13939 · **PDF:** https://arxiv.org/pdf/2602.13939v2
+
+## 12255. A Multi-Agent Framework for Code-Guided, Modular, and Verifiable Automated Machine Learning
+
+- **ID:** `2602.13937` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Dat Le, Duc-Cuong Le, Anh-Son Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.13937 · **PDF:** https://arxiv.org/pdf/2602.13937v1
+
+## 12256. A Generalizable Physics-guided Causal Model for Trajectory Prediction in Autonomous Driving
+
+- **ID:** `2602.13936` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Zhenyu Zong, Yuchen Wang, Haohong Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.13936 · **PDF:** https://arxiv.org/pdf/2602.13936v1
+
+## 12257. Statistical Early Stopping for Reasoning Models
+
+- **ID:** `2602.13935` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Yangxinyu Xie, Tao Wang, Soham Mallick, et al.
+- **abs:** https://arxiv.org/abs/2602.13935 · **PDF:** https://arxiv.org/pdf/2602.13935v1
+
+## 12258. Why Code, Why Now: Learnability, Computability, and the Real Limits of Machine Learning
+
+- **ID:** `2602.13934` · **primary:** `cs.LG` · **Submitted:** 2026-02-15
+- **Authors:** Zhimin Zhao
+- **abs:** https://arxiv.org/abs/2602.13934 · **PDF:** https://arxiv.org/pdf/2602.13934v2
+
+## 12259. HyMem: Hybrid Memory Architecture with Dynamic Retrieval Scheduling
+
+- **ID:** `2602.13933` · **primary:** `cs.AI` · **Submitted:** 2026-02-15
+- **Authors:** Xiaochen Zhao, Kaikai Wang, Xiaowen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.13933 · **PDF:** https://arxiv.org/pdf/2602.13933v1
+
+## 12260. MamaDino: A Hybrid Vision Model for Breast Cancer 3-Year Risk Prediction
+
+- **ID:** `2602.13930` · **primary:** `cs.CV` · **Submitted:** 2026-02-14
+- **Authors:** Ruggiero Santeramo, Igor Zubarev, Florian Jug
+- **abs:** https://arxiv.org/abs/2602.13930 · **PDF:** https://arxiv.org/pdf/2602.13930v1
+
+## 12261. voice2mode: Phonation Mode Classification in Singing using Self-Supervised Speech Models
+
+- **ID:** `2602.13928` · **primary:** `cs.SD` · **Submitted:** 2026-02-14
+- **Authors:** Aju Ani Justus, Ruchit Agrawal, Sudarsana Reddy Kadiri, et al.
+- **abs:** https://arxiv.org/abs/2602.13928 · **PDF:** https://arxiv.org/pdf/2602.13928v1
+
+## 12262. GREPO: A Benchmark for Graph Neural Networks on Repository-Level Bug Localization
+
+- **ID:** `2602.13921` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Juntong Wang, Libin Chen, Xiyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13921 · **PDF:** https://arxiv.org/pdf/2602.13921v1
+
+## 12263. A Comparative Analysis of Social Network Topology in Reddit and Moltbook
+
+- **ID:** `2602.13920` · **primary:** `cs.SI` · **Submitted:** 2026-02-14
+- **Authors:** Yiming Zhu, Gareth Tyson, Pan Hui
+- **abs:** https://arxiv.org/abs/2602.13920 · **PDF:** https://arxiv.org/pdf/2602.13920v3
+
+## 12264. DeepXiv-SDK: An Agentic Data Interface for Scientific Literature
+
+- **ID:** `2603.00084` · **primary:** `cs.DL` · **Submitted:** 2026-02-14
+- **Authors:** Hongjin Qian, Ziyi Xia, Ze Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.00084 · **PDF:** https://arxiv.org/pdf/2603.00084v2
+
+## 12265. Evaluating Adjective-Noun Compositionality in LLMs: Functional vs Representational Perspectives
+
+- **ID:** `2603.09994` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Ruchira Dhar, Qiwei Peng, Anders Søgaard
+- **abs:** https://arxiv.org/abs/2603.09994 · **PDF:** https://arxiv.org/pdf/2603.09994v2
+
+## 12266. Common Knowledge Always, Forever
+
+- **ID:** `2602.13914` · **primary:** `cs.LO` · **Submitted:** 2026-02-14
+- **Authors:** Martín Diéguez, David Fernández-Duque
+- **abs:** https://arxiv.org/abs/2602.13914 · **PDF:** https://arxiv.org/pdf/2602.13914v1
+
+## 12267. From Pixels to Policies: Reinforcing Spatial Reasoning in Language Models for Content-Aware Layout Design
+
+- **ID:** `2602.13912` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Sha Li, Stefano Petrangeli, Yu Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.13912 · **PDF:** https://arxiv.org/pdf/2602.13912v2
+
+## 12268. Sufficient Conditions for Stability of Minimum-Norm Interpolating Deep ReLU Networks
+
+- **ID:** `2602.13910` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Ouns El Harzli, Yoonsoo Nam, Ilja Kuzborskij, et al.
+- **abs:** https://arxiv.org/abs/2602.13910 · **PDF:** https://arxiv.org/pdf/2602.13910v1
+
+## 12269. Quantifying Normality: Convergence Rate to Gaussian Limit for Stochastic Approximation and Unadjusted OU Algorithm
+
+- **ID:** `2602.13906` · **primary:** `stat.ML` · **Submitted:** 2026-02-14
+- **Authors:** Shaan Ul Haque, Zedong Wang, Zixuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.13906 · **PDF:** https://arxiv.org/pdf/2602.13906v1
+
+## 12270. Pre-Editorial Normalization for Automatically Transcribed Medieval Manuscripts in Old French and Latin
+
+- **ID:** `2602.13905` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Thibault Clérice, Rachel Bawden, Anthony Glaise, et al.
+- **abs:** https://arxiv.org/abs/2602.13905 · **PDF:** https://arxiv.org/pdf/2602.13905v1
+
+## 12271. Diagnosing Pathological Chain-of-Thought in Reasoning Models
+
+- **ID:** `2602.13904` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Manqing Liu, David Williams-King, Ida Caspary, et al.
+- **abs:** https://arxiv.org/abs/2602.13904 · **PDF:** https://arxiv.org/pdf/2602.13904v1
+
+## 12272. RPGD: RANSAC-P3P Gradient Descent for Extrinsic Calibration in 3D Human Pose Estimation
+
+- **ID:** `2602.13901` · **primary:** `cs.CV` · **Submitted:** 2026-02-14
+- **Authors:** Zhanyu Tuo
+- **abs:** https://arxiv.org/abs/2602.13901 · **PDF:** https://arxiv.org/pdf/2602.13901v1
+
+## 12273. GSRM: Generative Speech Reward Model for Speech RLHF
+
+- **ID:** `2602.13891` · **primary:** `cs.SD` · **Submitted:** 2026-02-14
+- **Authors:** Maohao Shen, Tejas Jayashankar, Osama Hanna, et al.
+- **abs:** https://arxiv.org/abs/2602.13891 · **PDF:** https://arxiv.org/pdf/2602.13891v1
+
+## 12274. Evaluating Prompt Engineering Techniques for RAG in Small Language Models: A Multi-Hop QA Approach
+
+- **ID:** `2602.13890` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Amir Hossein Mohammadi, Ali Moeinian, Zahra Razavizade, et al.
+- **abs:** https://arxiv.org/abs/2602.13890 · **PDF:** https://arxiv.org/pdf/2602.13890v1
+
+## 12275. Parameter-Efficient Fine-Tuning of DINOv2 for Large-Scale Font Classification
+
+- **ID:** `2602.13889` · **primary:** `cs.CV` · **Submitted:** 2026-02-14
+- **Authors:** Daniel Chen, Zaria Zinn, Marcus Lowe
+- **abs:** https://arxiv.org/abs/2602.13889 · **PDF:** https://arxiv.org/pdf/2602.13889v1
+
+## 12276. Mixture-of-experts Wishart model for covariance matrices with an application to Cancer drug screening
+
+- **ID:** `2602.13888` · **primary:** `stat.ME` · **Submitted:** 2026-02-14
+- **Authors:** The Tien Mai, Zhi Zhao
+- **abs:** https://arxiv.org/abs/2602.13888 · **PDF:** https://arxiv.org/pdf/2602.13888v1
+
+## 12277. VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection
+
+- **ID:** `2602.13880` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Jiahao Xie, Guangmo Tong
+- **abs:** https://arxiv.org/abs/2602.13880 · **PDF:** https://arxiv.org/pdf/2602.13880v1
+
+## 12278. Ambient Physics: Training Neural PDE Solvers with Partial Observations
+
+- **ID:** `2602.13873` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Harris Abdul Majid, Giannis Daras, Francesco Tudisco, et al.
+- **abs:** https://arxiv.org/abs/2602.13873 · **PDF:** https://arxiv.org/pdf/2602.13873v1
+
+## 12279. Predicting fixed-sample test decisions enables anytime-valid inference
+
+- **ID:** `2602.13872` · **primary:** `stat.ME` · **Submitted:** 2026-02-14
+- **Authors:** Chris Holmes, Stephen Walker
+- **abs:** https://arxiv.org/abs/2602.13872 · **PDF:** https://arxiv.org/pdf/2602.13872v1
+
+## 12280. Ensemble-Conditional Gaussian Processes (Ens-CGP): Representation, Geometry, and Inference
+
+- **ID:** `2602.13871` · **primary:** `math.ST` · **Submitted:** 2026-02-14
+- **Authors:** Sai Ravela, Jae Deok Kim, Kenneth Gee, et al.
+- **abs:** https://arxiv.org/abs/2602.13871 · **PDF:** https://arxiv.org/pdf/2602.13871v1
+
+## 12281. ADAB: Arabic Dataset for Automated Politeness Benchmarking -- A Large-Scale Resource for Computational Sociopragmatics
+
+- **ID:** `2602.13870` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Hend Al-Khalifa, Nadia Ghezaiel, Maria Bounnit, et al.
+- **abs:** https://arxiv.org/abs/2602.13870 · **PDF:** https://arxiv.org/pdf/2602.13870v2
+
+## 12282. Bridging the Multilingual Safety Divide: Efficient, Culturally-Aware Alignment for Global South Languages
+
+- **ID:** `2602.13867` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Somnath Banerjee, Rima Hazra, Animesh Mukherjee
+- **abs:** https://arxiv.org/abs/2602.13867 · **PDF:** https://arxiv.org/pdf/2602.13867v1
+
+## 12283. Enabling Option Learning in Sparse Rewards with Hindsight Experience Replay
+
+- **ID:** `2602.13865` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Gabriel Romio, Mateus Begnini Melchiades, Bruno Castro da Silva, et al.
+- **abs:** https://arxiv.org/abs/2602.13865 · **PDF:** https://arxiv.org/pdf/2602.13865v1
+
+## 12284. Evolving Multi-Channel Confidence-Aware Activation Functions for Missing Data with Channel Propagation
+
+- **ID:** `2602.13864` · **primary:** `cs.NE` · **Submitted:** 2026-02-14
+- **Authors:** Naeem Shahabi Sani, Ferial Najiantabriz, Shayan Shafaei, et al.
+- **abs:** https://arxiv.org/abs/2602.13864 · **PDF:** https://arxiv.org/pdf/2602.13864v1
+
+## 12285. Tutoring Large Language Models to be Domain-adaptive, Precise, and Safe
+
+- **ID:** `2602.13860` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Somnath Banerjee
+- **abs:** https://arxiv.org/abs/2602.13860 · **PDF:** https://arxiv.org/pdf/2602.13860v1
+
+## 12286. sleep2vec: Unified Cross-Modal Alignment for Heterogeneous Nocturnal Biosignals
+
+- **ID:** `2602.13857` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Weixuan Yuan, Zengrui Jin, Yichen Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13857 · **PDF:** https://arxiv.org/pdf/2602.13857v1
+
+## 12287. From Fluent to Verifiable: Claim-Level Auditability for Deep Research Agents
+
+- **ID:** `2602.13855` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Razeen A Rasheed, Somnath Banerjee, Animesh Mukherjee, et al.
+- **abs:** https://arxiv.org/abs/2602.13855 · **PDF:** https://arxiv.org/pdf/2602.13855v1
+
+## 12288. Experimentation Accelerator: Interpretable Insights and Creative Recommendations for A/B Testing with Content-Aware ranking
+
+- **ID:** `2602.13852` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Zhengmian Hu, Lei Shi, Ritwik Sinha, et al.
+- **abs:** https://arxiv.org/abs/2602.13852 · **PDF:** https://arxiv.org/pdf/2602.13852v1
+
+## 12289. Evaluating LLM-Generated ACSL Annotations for Formal Verification
+
+- **ID:** `2602.13851` · **primary:** `cs.SE` · **Submitted:** 2026-02-14
+- **Authors:** Arshad Beg, Diarmuid O'Donoghue, Rosemary Monahan
+- **abs:** https://arxiv.org/abs/2602.13851 · **PDF:** https://arxiv.org/pdf/2602.13851v2
+
+## 12290. Testing For Distribution Shifts with Conditional Conformal Test Martingales
+
+- **ID:** `2602.13848` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Shalev Shaer, Yarin Bar, Drew Prinster, et al.
+- **abs:** https://arxiv.org/abs/2602.13848 · **PDF:** https://arxiv.org/pdf/2602.13848v1
+
+## 12291. Causally constrained reduced-order neural models of complex turbulent dynamical systems
+
+- **ID:** `2602.13847` · **primary:** `nlin.CD` · **Submitted:** 2026-02-14
+- **Authors:** Fabrizio Falasca, Laure Zanna
+- **abs:** https://arxiv.org/abs/2602.13847 · **PDF:** https://arxiv.org/pdf/2602.13847v2
+
+## 12292. Do Mixed-Vendor Multi-Agent LLMs Improve Clinical Diagnosis?
+
+- **ID:** `2603.04421` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Grace Chang Yuan, Xiaoman Zhang, Sung Eun Kim, et al.
+- **abs:** https://arxiv.org/abs/2603.04421 · **PDF:** https://arxiv.org/pdf/2603.04421v2
+
+## 12293. Automated Prediction of Paravalvular Regurgitation before Transcatheter Aortic Valve Implantation
+
+- **ID:** `2602.13842` · **primary:** `cs.CV` · **Submitted:** 2026-02-14
+- **Authors:** Michele Cannito, Riccardo Renzulli, Adson Duarte, et al.
+- **abs:** https://arxiv.org/abs/2602.13842 · **PDF:** https://arxiv.org/pdf/2602.13842v1
+
+## 12294. PrivAct: Internalizing Contextual Privacy Preservation via Multi-Agent Preference Training
+
+- **ID:** `2602.13840` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Yuhan Cheng, Hancheng Ye, Hai Helen Li, et al.
+- **abs:** https://arxiv.org/abs/2602.13840 · **PDF:** https://arxiv.org/pdf/2602.13840v1
+
+## 12295. Speculative Decoding with a Speculative Vocabulary
+
+- **ID:** `2602.13836` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Miles Williams, Young D. Kwon, Rui Li, et al.
+- **abs:** https://arxiv.org/abs/2602.13836 · **PDF:** https://arxiv.org/pdf/2602.13836v1
+
+## 12296. Beyond Words: Evaluating and Bridging Epistemic Divergence in User-Agent Interaction via Theory of Mind
+
+- **ID:** `2602.13832` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Minyuan Ruan, Ziyue Wang, Kaiming Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.13832 · **PDF:** https://arxiv.org/pdf/2602.13832v1
+
+## 12297. VAR-3D: View-aware Auto-Regressive Model for Text-to-3D Generation via a 3D Tokenizer
+
+- **ID:** `2602.13818` · **primary:** `cs.CV` · **Submitted:** 2026-02-14
+- **Authors:** Zongcheng Han, Dongyan Cao, Haoran Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.13818 · **PDF:** https://arxiv.org/pdf/2602.13818v1
+
+## 12298. What happens when reviewers receive AI feedback in their reviews?
+
+- **ID:** `2602.13817` · **primary:** `cs.HC` · **Submitted:** 2026-02-14
+- **Authors:** Shiping Chen, Shu Zhong, Duncan P. Brumby, et al.
+- **abs:** https://arxiv.org/abs/2602.13817 · **PDF:** https://arxiv.org/pdf/2602.13817v1
+
+## 12299. The acquisition of English irregular inflections by Yemeni L1 Arabic learners: A Universal Grammar approach
+
+- **ID:** `2602.13816` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Muneef Y. Alsawsh, Mohammed Q. Shormani
+- **abs:** https://arxiv.org/abs/2602.13816 · **PDF:** https://arxiv.org/pdf/2602.13816v1
+
+## 12300. Pawsterior: Variational Flow Matching for Structured Simulation-Based Inference
+
+- **ID:** `2602.13813` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Jorge Carrasco-Pollo, Floor Eijkelboom, Jan-Willem van de Meent
+- **abs:** https://arxiv.org/abs/2602.13813 · **PDF:** https://arxiv.org/pdf/2602.13813v3
+
+## 12301. DTBench: A Synthetic Benchmark for Document-to-Table Extraction
+
+- **ID:** `2602.13812` · **primary:** `cs.DB` · **Submitted:** 2026-02-14
+- **Authors:** Yuxiang Guo, Zhuoran Du, Nan Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.13812 · **PDF:** https://arxiv.org/pdf/2602.13812v2
+
+## 12302. A Unified Physics-Informed Neural Network for Modeling Coupled Electro- and Elastodynamic Wave Propagation Using Three-Stage Loss Optimization
+
+- **ID:** `2602.13811` · **primary:** `cs.NE` · **Submitted:** 2026-02-14
+- **Authors:** Suhas Suresh Bharadwaj, Reuben Thomas Thovelil
+- **abs:** https://arxiv.org/abs/2602.13811 · **PDF:** https://arxiv.org/pdf/2602.13811v1
+
+## 12303. Mean Flow Policy with Instantaneous Velocity Constraint for One-step Action Generation
+
+- **ID:** `2602.13810` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Guojian Zhan, Letian Tao, Pengcheng Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13810 · **PDF:** https://arxiv.org/pdf/2602.13810v2
+
+## 12304. An end-to-end agentic pipeline for smart contract translation and quality evaluation
+
+- **ID:** `2602.13808` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Abhinav Goel, Chaitya Shah, Agostino Capponi, et al.
+- **abs:** https://arxiv.org/abs/2602.13808 · **PDF:** https://arxiv.org/pdf/2602.13808v1
+
+## 12305. AnomaMind: Agentic Time Series Anomaly Detection with Tool-Augmented Reasoning
+
+- **ID:** `2602.13807` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Xiaoyu Tao, Yuchong Wu, Mingyue Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.13807 · **PDF:** https://arxiv.org/pdf/2602.13807v1
+
+## 12306. Fast Physics-Driven Untrained Network for Highly Nonlinear Inverse Scattering Problems
+
+- **ID:** `2602.13805` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Yutong Du, Zicheng Liu, Yi Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.13805 · **PDF:** https://arxiv.org/pdf/2602.13805v1
+
+## 12307. Attention in Constant Time: Vashista Sparse Attention for Long-Context Decoding with Exponential Guarantees
+
+- **ID:** `2602.13804` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Vashista Nobaub
+- **abs:** https://arxiv.org/abs/2602.13804 · **PDF:** https://arxiv.org/pdf/2602.13804v1
+
+## 12308. Cast-R1: Learning Tool-Augmented Sequential Decision Policies for Time Series Forecasting
+
+- **ID:** `2602.13802` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Xiaoyu Tao, Mingyue Cheng, Chuang Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.13802 · **PDF:** https://arxiv.org/pdf/2602.13802v1
+
+## 12309. OMGs: A multi-agent system supporting MDT decision-making across the ovarian tumour care continuum
+
+- **ID:** `2602.13793` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Yangyang Zhang, Zilong Wang, Jianbo Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.13793 · **PDF:** https://arxiv.org/pdf/2602.13793v1
+
+## 12310. StackingNet: Collective Inference Across Independent AI Foundation Models
+
+- **ID:** `2602.13792` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Siyang Li, Chenhao Liu, Dongrui Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.13792 · **PDF:** https://arxiv.org/pdf/2602.13792v1
+
+## 12311. MechPert: Mechanistic Consensus as an Inductive Bias for Unseen Perturbation Prediction
+
+- **ID:** `2602.13791` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Marc Boubnovski Martell, Josefa Lia Stoisser, Lawrence Phillips, et al.
+- **abs:** https://arxiv.org/abs/2602.13791 · **PDF:** https://arxiv.org/pdf/2602.13791v1
+
+## 12312. How Do Lexical Senses Correspond Between Spoken German and German Sign Language?
+
+- **ID:** `2602.13790` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Melis Çelikkol, Wei Zhao
+- **abs:** https://arxiv.org/abs/2602.13790 · **PDF:** https://arxiv.org/pdf/2602.13790v1
+
+## 12313. Comparables XAI: Faithful Example-based AI Explanations with Counterfactual Trace Adjustments
+
+- **ID:** `2602.13784` · **primary:** `cs.HC` · **Submitted:** 2026-02-14
+- **Authors:** Yifan Zhang, Tianle Ren, Fei Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13784 · **PDF:** https://arxiv.org/pdf/2602.13784v1
+
+## 12314. MEMTS: Internalizing Domain Knowledge via Parameterized Memory for Retrieval-Free Domain Adaptation of Time Series Foundation Models
+
+- **ID:** `2602.13783` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Xiaoyun Yu, Li fan, Xiangfei Qiu, et al.
+- **abs:** https://arxiv.org/abs/2602.13783 · **PDF:** https://arxiv.org/pdf/2602.13783v1
+
+## 12315. On Representation Redundancy in Large-Scale Instruction Tuning Data Selection
+
+- **ID:** `2602.13773` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Youwei Shu, Shaomian Zheng, Dingnan Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.13773 · **PDF:** https://arxiv.org/pdf/2602.13773v1
+
+## 12316. NeuroMambaLLM: Dynamic Graph Learning of fMRI Functional Connectivity in Autistic Brains Using Mamba and Language Model Reasoning
+
+- **ID:** `2602.13770` · **primary:** `eess.IV` · **Submitted:** 2026-02-14
+- **Authors:** Yasaman Torabi, Parsa Razmara, Hamed Ajorlou, et al.
+- **abs:** https://arxiv.org/abs/2602.13770 · **PDF:** https://arxiv.org/pdf/2602.13770v1
+
+## 12317. OR-Agent: Bridging Evolutionary Search and Structured Research for Automated Algorithm Discovery
+
+- **ID:** `2602.13769` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Qi Liu, Ruochen Hao, Can Li, et al.
+- **abs:** https://arxiv.org/abs/2602.13769 · **PDF:** https://arxiv.org/pdf/2602.13769v2
+
+## 12318. MOTIF: Learning Action Motifs for Few-shot Cross-Embodiment Transfer
+
+- **ID:** `2602.13764` · **primary:** `cs.RO` · **Submitted:** 2026-02-14
+- **Authors:** Heng Zhi, Wentao Tan, Lei Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.13764 · **PDF:** https://arxiv.org/pdf/2602.13764v1
+
+## 12319. Discrete Double-Bracket Flows for Isotropic-Noise Invariant Eigendecomposition
+
+- **ID:** `2602.13759` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** ZhiMing Li, JiaHe Feng
+- **abs:** https://arxiv.org/abs/2602.13759 · **PDF:** https://arxiv.org/pdf/2602.13759v1
+
+## 12320. OmniScience: A Large-scale Multi-modal Dataset for Scientific Image Understanding
+
+- **ID:** `2602.13758` · **primary:** `cs.CV` · **Submitted:** 2026-02-14
+- **Authors:** Haoyi Tao, Chaozheng Huang, Nan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13758 · **PDF:** https://arxiv.org/pdf/2602.13758v1
+
+## 12321. RMPL: Relation-aware Multi-task Progressive Learning with Stage-wise Training for Multimedia Event Extraction
+
+- **ID:** `2602.13748` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Yongkang Jin, Jianwen Luo, Jingjing Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13748 · **PDF:** https://arxiv.org/pdf/2602.13748v1
+
+## 12322. Data-driven Bi-level Optimization of Thermal Power Systems with embedded Artificial Neural Networks
+
+- **ID:** `2602.13746` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Talha Ansar, Muhammad Mujtaba Abbas, Ramit Debnath, et al.
+- **abs:** https://arxiv.org/abs/2602.13746 · **PDF:** https://arxiv.org/pdf/2602.13746v1
+
+## 12323. OneLatent: Single-Token Compression for Visual Latent Reasoning
+
+- **ID:** `2602.13738` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Bo Lv, Yasheng Sun, Junjie Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13738 · **PDF:** https://arxiv.org/pdf/2602.13738v1
+
+## 12324. Discrete Gene Crossover Accelerates Solution Discovery in Quality-Diversity Algorithms
+
+- **ID:** `2602.13730` · **primary:** `cs.NE` · **Submitted:** 2026-02-14
+- **Authors:** Joshua Hutchinson, J. Michael Herrmann, Simón C. Smith
+- **abs:** https://arxiv.org/abs/2602.13730 · **PDF:** https://arxiv.org/pdf/2602.13730v1
+
+## 12325. HybridFlow: A Two-Step Generative Policy for Robotic Manipulation
+
+- **ID:** `2602.13718` · **primary:** `cs.RO` · **Submitted:** 2026-02-14
+- **Authors:** Zhenchen Dong, Jinna Fu, Jiaming Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.13718 · **PDF:** https://arxiv.org/pdf/2602.13718v1
+
+## 12326. On Theoretically-Driven LLM Agents for Multi-Dimensional Discourse Analysis
+
+- **ID:** `2602.13713` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Maciej Uberna, Michał Wawer, Jarosław A. Chudziak, et al.
+- **abs:** https://arxiv.org/abs/2602.13713 · **PDF:** https://arxiv.org/pdf/2602.13713v2
+
+## 12327. Fine-tuned Vision Language Model for Localization of Parasitic Eggs in Microscopic Images
+
+- **ID:** `2602.13712` · **primary:** `cs.CV` · **Submitted:** 2026-02-14
+- **Authors:** Chan Hao Sien, Hezerul Abdul Karim, Nouar AlDahoul
+- **abs:** https://arxiv.org/abs/2602.13712 · **PDF:** https://arxiv.org/pdf/2602.13712v1
+
+## 12328. HBVLA: Pushing 1-Bit Post-Training Quantization for Vision-Language-Action Models
+
+- **ID:** `2602.13710` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Xin Yan, Zhenglin Wan, Feiyang Ye, et al.
+- **abs:** https://arxiv.org/abs/2602.13710 · **PDF:** https://arxiv.org/pdf/2602.13710v1
+
+## 12329. Near-Optimal Regret for Policy Optimization in Contextual MDPs with General Offline Function Approximation
+
+- **ID:** `2602.13706` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Orin Levy, Aviv Rosenberg, Alon Cohen, et al.
+- **abs:** https://arxiv.org/abs/2602.13706 · **PDF:** https://arxiv.org/pdf/2602.13706v1
+
+## 12330. Pailitao-VL: Unified Embedding and Reranker for Real-Time Multi-Modal Industrial Search
+
+- **ID:** `2602.13704` · **primary:** `cs.IR` · **Submitted:** 2026-02-14
+- **Authors:** Lei Chen, Chen Ju, Xu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.13704 · **PDF:** https://arxiv.org/pdf/2602.13704v2
+
+## 12331. Metaphors' journeys across time and genre: tracking the evolution of literary metaphors with temporal embeddings
+
+- **ID:** `2602.13701` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Veronica Mangiaterra, Chiara Barattieri di San Pietro, Paolo Canal, et al.
+- **abs:** https://arxiv.org/abs/2602.13701 · **PDF:** https://arxiv.org/pdf/2602.13701v1
+
+## 12332. Offline-First Large Language Model Architecture for AI-Assisted Learning with Adaptive Response Levels in Low-Connectivity Environments
+
+- **ID:** `2603.03339` · **primary:** `cs.CY` · **Submitted:** 2026-02-14
+- **Authors:** Joseph Walusimbi, Ann Move Oguti, Joshua Benjamin Ssentongo, et al.
+- **abs:** https://arxiv.org/abs/2603.03339 · **PDF:** https://arxiv.org/pdf/2603.03339v4
+
+## 12333. Optimal Regret for Policy Optimization in Contextual Bandits
+
+- **ID:** `2602.13700` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Orin Levy, Yishay Mansour
+- **abs:** https://arxiv.org/abs/2602.13700 · **PDF:** https://arxiv.org/pdf/2602.13700v1
+
+## 12334. Attention Head Entropy of LLMs Predicts Answer Correctness
+
+- **ID:** `2602.13699` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Sophie Ostmeier, Brian Axelrod, Maya Varma, et al.
+- **abs:** https://arxiv.org/abs/2602.13699 · **PDF:** https://arxiv.org/pdf/2602.13699v1
+
+## 12335. No Need to Train Your RDB Foundation Model
+
+- **ID:** `2602.13697` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Linjie Xu, Yanlin Zhang, Quan Gan, et al.
+- **abs:** https://arxiv.org/abs/2602.13697 · **PDF:** https://arxiv.org/pdf/2602.13697v1
+
+## 12336. Can a Lightweight Automated AI Pipeline Solve Research-Level Mathematical Problems?
+
+- **ID:** `2602.13695` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Lve Meng, Weilong Zhao, Yanzhi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.13695 · **PDF:** https://arxiv.org/pdf/2602.13695v2
+
+## 12337. PhGPO: Pheromone-Guided Policy Optimization for Long-Horizon Tool Planning
+
+- **ID:** `2602.13691` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Yu Li, Guangfeng Cai, Shengtian Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.13691 · **PDF:** https://arxiv.org/pdf/2602.13691v1
+
+## 12338. RDBLearn: Simple In-Context Prediction Over Relational Databases
+
+- **ID:** `2602.18495` · **primary:** `cs.DB` · **Submitted:** 2026-02-14
+- **Authors:** Yanlin Zhang, Linjie Xu, Quan Gan, et al.
+- **abs:** https://arxiv.org/abs/2602.18495 · **PDF:** https://arxiv.org/pdf/2602.18495v1
+
+## 12339. Physics Aware Neural Networks: Denoising for Magnetic Navigation
+
+- **ID:** `2602.13690` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Aritra Das, Yashas Shende, Muskaan Chugh, et al.
+- **abs:** https://arxiv.org/abs/2602.13690 · **PDF:** https://arxiv.org/pdf/2602.13690v1
+
+## 12340. AuTAgent: A Reinforcement Learning Framework for Tool-Augmented Audio Reasoning
+
+- **ID:** `2602.13685` · **primary:** `cs.SD` · **Submitted:** 2026-02-14
+- **Authors:** Siqian Tong, Xuan Li, Yiwei Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13685 · **PDF:** https://arxiv.org/pdf/2602.13685v1
+
+## 12341. On the Sparsifiability of Correlation Clustering: Approximation Guarantees under Edge Sampling
+
+- **ID:** `2602.13684` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma
+- **abs:** https://arxiv.org/abs/2602.13684 · **PDF:** https://arxiv.org/pdf/2602.13684v1
+
+## 12342. An Ensemble Learning Approach towards Waste Segmentation in Cluttered Environment
+
+- **ID:** `2602.13681` · **primary:** `cs.CV` · **Submitted:** 2026-02-14
+- **Authors:** Maimoona Jafar, Syed Imran Ali, Ahsan Saadat, et al.
+- **abs:** https://arxiv.org/abs/2602.13681 · **PDF:** https://arxiv.org/pdf/2602.13681v1
+
+## 12343. AllMem: A Memory-centric Recipe for Efficient Long-context Modeling
+
+- **ID:** `2602.13680` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Ziming Wang, Xiang Wang, Kailong Peng, et al.
+- **abs:** https://arxiv.org/abs/2602.13680 · **PDF:** https://arxiv.org/pdf/2602.13680v1
+
+## 12344. Transferable XAI: Relating Understanding Across Domains with Explanation Transfer
+
+- **ID:** `2602.13675` · **primary:** `cs.HC` · **Submitted:** 2026-02-14
+- **Authors:** Fei Wang, Yifan Zhang, Brian Y. Lim
+- **abs:** https://arxiv.org/abs/2602.13675 · **PDF:** https://arxiv.org/pdf/2602.13675v1
+
+## 12345. On the Dynamics of Observation and Semantics
+
+- **ID:** `2602.18494` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Xiu Li
+- **abs:** https://arxiv.org/abs/2602.18494 · **PDF:** https://arxiv.org/pdf/2602.18494v1
+
+## 12346. LEAD-Drift: Real-time and Explainable Intent Drift Detection by Learning a Data-Driven Risk Score
+
+- **ID:** `2602.13672` · **primary:** `cs.NI` · **Submitted:** 2026-02-14
+- **Authors:** Md. Kamrul Hossain, Walid Aljoby
+- **abs:** https://arxiv.org/abs/2602.13672 · **PDF:** https://arxiv.org/pdf/2602.13672v1
+
+## 12347. MAS-on-the-Fly: Dynamic Adaptation of LLM-based Multi-Agent Systems at Test Time
+
+- **ID:** `2602.13671` · **primary:** `cs.MA` · **Submitted:** 2026-02-14
+- **Authors:** Guangyi Liu, Haojun Lin, Huan Zeng, et al.
+- **abs:** https://arxiv.org/abs/2602.13671 · **PDF:** https://arxiv.org/pdf/2602.13671v1
+
+## 12348. Linguistic Uncertainty and Engagement in Arabic-Language X (formerly Twitter) Discourse
+
+- **ID:** `2603.00082` · **primary:** `cs.CY` · **Submitted:** 2026-02-14
+- **Authors:** Mohamed Soufan
+- **abs:** https://arxiv.org/abs/2603.00082 · **PDF:** https://arxiv.org/pdf/2603.00082v1
+
+## 12349. Advancing Analytic Class-Incremental Learning through Vision-Language Calibration
+
+- **ID:** `2602.13670` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Binyu Zhao, Wei Zhang, Xingrui Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.13670 · **PDF:** https://arxiv.org/pdf/2602.13670v1
+
+## 12350. CEI: A Benchmark for Evaluating Pragmatic Reasoning in Language Models
+
+- **ID:** `2603.09993` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Jon Chun, Hannah Sussman, Adrian Mangine, et al.
+- **abs:** https://arxiv.org/abs/2603.09993 · **PDF:** https://arxiv.org/pdf/2603.09993v1
+
+## 12351. ALMo: Interactive Aim-Limit-Defined, Multi-Objective System for Personalized High-Dose-Rate Brachytherapy Treatment Planning and Visualization for Cervical Cancer
+
+- **ID:** `2602.13666` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Edward Chen, Natalie Dullerud, Pang Wei Koh, et al.
+- **abs:** https://arxiv.org/abs/2602.13666 · **PDF:** https://arxiv.org/pdf/2602.13666v1
+
+## 12352. HyFunc: Accelerating LLM-based Function Calls for Agentic AI through Hybrid-Model Cascade and Dynamic Templating
+
+- **ID:** `2602.13665` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Weibin Liao, Jian-guang Lou, Haoyi Xiong
+- **abs:** https://arxiv.org/abs/2602.13665 · **PDF:** https://arxiv.org/pdf/2602.13665v1
+
+## 12353. LeafNet: A Large-Scale Dataset and Comprehensive Benchmark for Foundational Vision-Language Understanding of Plant Diseases
+
+- **ID:** `2602.13662` · **primary:** `cs.CV` · **Submitted:** 2026-02-14
+- **Authors:** Khang Nguyen Quoc, Phuong D. Dao, Luyl-Da Quach
+- **abs:** https://arxiv.org/abs/2602.13662 · **PDF:** https://arxiv.org/pdf/2602.13662v3
+
+## 12354. Optimized Certainty Equivalent Risk-Controlling Prediction Sets
+
+- **ID:** `2602.13660` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Jiayi Huang, Amirmohammad Farzaneh, Osvaldo Simeone
+- **abs:** https://arxiv.org/abs/2602.13660 · **PDF:** https://arxiv.org/pdf/2602.13660v1
+
+## 12355. Zero-Order Optimization for LLM Fine-Tuning via Learnable Direction Sampling
+
+- **ID:** `2602.13659` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Valery Parfenov, Grigoriy Evseev, Andrey Veprikov, et al.
+- **abs:** https://arxiv.org/abs/2602.13659 · **PDF:** https://arxiv.org/pdf/2602.13659v1
+
+## 12356. Attention-gated U-Net model for semantic segmentation of brain tumors and feature extraction for survival prognosis
+
+- **ID:** `2602.15067` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Rut Pate, Snehal Rajput, Mehul S. Raval, et al.
+- **abs:** https://arxiv.org/abs/2602.15067 · **PDF:** https://arxiv.org/pdf/2602.15067v1
+
+## 12357. Building Autonomous GUI Navigation via Agentic-Q Estimation and Step-Wise Policy Optimization
+
+- **ID:** `2602.13653` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Yibo Wang, Guangda Huzhang, Yuwei Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.13653 · **PDF:** https://arxiv.org/pdf/2602.13653v1
+
+## 12358. Cumulative Utility Parity for Fair Federated Learning under Intermittent Client Participation
+
+- **ID:** `2602.13651` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Stefan Behfar, Richard Mortier
+- **abs:** https://arxiv.org/abs/2602.13651 · **PDF:** https://arxiv.org/pdf/2602.13651v1
+
+## 12359. KorMedMCQA-V: A Multimodal Benchmark for Evaluating Vision-Language Models on the Korean Medical Licensing Examination
+
+- **ID:** `2602.13650` · **primary:** `cs.CV` · **Submitted:** 2026-02-14
+- **Authors:** Byungjin Choi, Seongsu Bae, Sunjun Kweon, et al.
+- **abs:** https://arxiv.org/abs/2602.13650 · **PDF:** https://arxiv.org/pdf/2602.13650v1
+
+## 12360. Joint Time Series Chain: Detecting Unusual Evolving Trend across Time Series
+
+- **ID:** `2602.13649` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Li Zhang, Nital Patel, Xiuqi Li, et al.
+- **abs:** https://arxiv.org/abs/2602.13649 · **PDF:** https://arxiv.org/pdf/2602.13649v1
+
+## 12361. SF-RAG: Structure-Fidelity Retrieval-Augmented Generation for Academic Question Answering
+
+- **ID:** `2602.13647` · **primary:** `cs.IR` · **Submitted:** 2026-02-14
+- **Authors:** Rui Yu, Tianyi Wang, Ruixia Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.13647 · **PDF:** https://arxiv.org/pdf/2602.13647v2
+
+## 12362. Hierarchical Audio-Visual-Proprioceptive Fusion for Precise Robotic Manipulation
+
+- **ID:** `2602.13640` · **primary:** `cs.RO` · **Submitted:** 2026-02-14
+- **Authors:** Siyuan Li, Jiani Lu, Yu Song, et al.
+- **abs:** https://arxiv.org/abs/2602.13640 · **PDF:** https://arxiv.org/pdf/2602.13640v1
+
+## 12363. Guided Collaboration in Heterogeneous LLM-Based Multi-Agent Systems via Entropy-Based Understanding Assessment and Experience Retrieval
+
+- **ID:** `2602.13639` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Linlin Wang, Tianqing Zhu, Laiqiao Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.13639 · **PDF:** https://arxiv.org/pdf/2602.13639v1
+
+## 12364. Optimization-Free Graph Embedding via Distributional Kernel for Community Detection
+
+- **ID:** `2602.13634` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Shuaibin Song, Kai Ming Ting, Kaifeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.13634 · **PDF:** https://arxiv.org/pdf/2602.13634v1
+
+## 12365. Benchmark Leakage Trap: Can We Trust LLM-based Recommendation?
+
+- **ID:** `2602.13626` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Mingqiao Zhang, Qiyao Peng, Yumeng Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13626 · **PDF:** https://arxiv.org/pdf/2602.13626v2
+
+## 12366. Anthropomorphism on Risk Perception: The Role of Trust and Domain Knowledge in Decision-Support AI
+
+- **ID:** `2602.13625` · **primary:** `cs.HC` · **Submitted:** 2026-02-14
+- **Authors:** Manuele Reani, Xiangyang He, Zuolan Bao
+- **abs:** https://arxiv.org/abs/2602.13625 · **PDF:** https://arxiv.org/pdf/2602.13625v1
+
+## 12367. Locally Private Parametric Methods for Change-Point Detection
+
+- **ID:** `2602.13619` · **primary:** `stat.ML` · **Submitted:** 2026-02-14
+- **Authors:** Anuj Kumar Yadav, Cemre Cadir, Yanina Shkel, et al.
+- **abs:** https://arxiv.org/abs/2602.13619 · **PDF:** https://arxiv.org/pdf/2602.13619v1
+
+## 12368. DiffusionRollout: Uncertainty-Aware Rollout Planning in Long-Horizon PDE Solving
+
+- **ID:** `2602.13616` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Seungwoo Yoo, Juil Koo, Daehyeon Choi, et al.
+- **abs:** https://arxiv.org/abs/2602.13616 · **PDF:** https://arxiv.org/pdf/2602.13616v1
+
+## 12369. An Adaptive Multichain Blockchain: A Multiobjective Optimization Approach
+
+- **ID:** `2602.22230` · **primary:** `cs.CR` · **Submitted:** 2026-02-14
+- **Authors:** Nimrod Talmon, Haim Zysberg
+- **abs:** https://arxiv.org/abs/2602.22230 · **PDF:** https://arxiv.org/pdf/2602.22230v1
+
+## 12370. From What to How: Bridging User Requirements with Software Development Using Large Language Models
+
+- **ID:** `2602.13611` · **primary:** `cs.SE` · **Submitted:** 2026-02-14
+- **Authors:** Xiao He, Ru Chen, Jialun Cao
+- **abs:** https://arxiv.org/abs/2602.13611 · **PDF:** https://arxiv.org/pdf/2602.13611v1
+
+## 12371. Machine Learning for Complex Systems Dynamics: Detecting Bifurcations in Dynamical Systems with Deep Neural Networks
+
+- **ID:** `2603.04420` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Swadesh Pal, Roderick Melnik
+- **abs:** https://arxiv.org/abs/2603.04420 · **PDF:** https://arxiv.org/pdf/2603.04420v1
+
+## 12372. Multi-Modal Sensing and Fusion in mmWave Beamforming for Connected Vehicles: A Transformer Based Framework
+
+- **ID:** `2602.13606` · **primary:** `cs.NI` · **Submitted:** 2026-02-14
+- **Authors:** Muhammad Baqer Mollah, Honggang Wang, Mohammad Ataul Karim, et al.
+- **abs:** https://arxiv.org/abs/2602.13606 · **PDF:** https://arxiv.org/pdf/2602.13606v1
+
+## 12373. Towards Sparse Video Understanding and Reasoning
+
+- **ID:** `2602.13602` · **primary:** `cs.CV` · **Submitted:** 2026-02-14
+- **Authors:** Chenwei Xu, Zhen Ye, Shang Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.13602 · **PDF:** https://arxiv.org/pdf/2602.13602v1
+
+## 12374. TAMUSA-Chat: A Domain-Adapted Large Language Model Conversational System for Research and Responsible Deployment
+
+- **ID:** `2603.09992` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Izzat Alsmadi, Anas Alsobeh
+- **abs:** https://arxiv.org/abs/2603.09992 · **PDF:** https://arxiv.org/pdf/2603.09992v1
+
+## 12375. The Quantization Trap: Breaking Linear Scaling Laws in Multi-Hop Reasoning
+
+- **ID:** `2602.13595` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Henry Han, Xiyang Liu, Xiaodong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13595 · **PDF:** https://arxiv.org/pdf/2602.13595v1
+
+## 12376. Hippocampus: An Efficient and Scalable Memory Module for Agentic AI
+
+- **ID:** `2602.13594` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Yi Li, Lianjie Cao, Faraz Ahmed, et al.
+- **abs:** https://arxiv.org/abs/2602.13594 · **PDF:** https://arxiv.org/pdf/2602.13594v1
+
+## 12377. Context-Dependent Affordance Computation in Vision-Language Models
+
+- **ID:** `2603.04419` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Murad Farzulla
+- **abs:** https://arxiv.org/abs/2603.04419 · **PDF:** https://arxiv.org/pdf/2603.04419v1
+
+## 12378. Two-Stream Interactive Joint Learning of Scene Parsing and Geometric Vision Tasks
+
+- **ID:** `2602.13588` · **primary:** `cs.CV` · **Submitted:** 2026-02-14
+- **Authors:** Guanfeng Tang, Hongbo Zhao, Ziwei Long, et al.
+- **abs:** https://arxiv.org/abs/2602.13588 · **PDF:** https://arxiv.org/pdf/2602.13588v1
+
+## 12379. A First Proof Sprint
+
+- **ID:** `2602.13587` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Joseph Corneli
+- **abs:** https://arxiv.org/abs/2602.13587 · **PDF:** https://arxiv.org/pdf/2602.13587v1
+
+## 12380. Interpretable clustering via optimal multiway-split decision trees
+
+- **ID:** `2602.13586` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Hayato Suzuki, Shunnosuke Ikeda, Yuichi Takano
+- **abs:** https://arxiv.org/abs/2602.13586 · **PDF:** https://arxiv.org/pdf/2602.13586v1
+
+## 12381. Differentiable Rule Induction from Raw Sequence Inputs
+
+- **ID:** `2602.13583` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Kun Gao, Katsumi Inoue, Yongzhi Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.13583 · **PDF:** https://arxiv.org/pdf/2602.13583v1
+
+## 12382. PoultryLeX-Net: Domain-Adaptive Dual-Stream Transformer Architecture for Large-Scale Poultry Stakeholder Modeling
+
+- **ID:** `2603.09991` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Stephen Afrifa, Biswash Khatiwada, Kapalik Khanal, et al.
+- **abs:** https://arxiv.org/abs/2603.09991 · **PDF:** https://arxiv.org/pdf/2603.09991v1
+
+## 12383. Rubrics as an Attack Surface: Stealthy Preference Drift in LLM Judges
+
+- **ID:** `2602.13576` · **primary:** `cs.CR` · **Submitted:** 2026-02-14
+- **Authors:** Ruomeng Ding, Yifei Pang, He Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.13576 · **PDF:** https://arxiv.org/pdf/2602.13576v1
+
+## 12384. Elo-Evolve: A Co-evolutionary Framework for Language Model Alignment
+
+- **ID:** `2602.13575` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Jing Zhao, Ting Zhen, Junwei Bao, et al.
+- **abs:** https://arxiv.org/abs/2602.13575 · **PDF:** https://arxiv.org/pdf/2602.13575v2
+
+## 12385. LLM-Confidence Reranker: A Training-Free Approach for Enhancing Retrieval-Augmented Generation Systems
+
+- **ID:** `2602.13571` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Zhipeng Song, Xiangyu Kong, Xinrui Bao, et al.
+- **abs:** https://arxiv.org/abs/2602.13571 · **PDF:** https://arxiv.org/pdf/2602.13571v1
+
+## 12386. Who Do LLMs Trust? Human Experts Matter More Than Other LLMs
+
+- **ID:** `2602.13568` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Anooshka Bajaj, Zoran Tiganj
+- **abs:** https://arxiv.org/abs/2602.13568 · **PDF:** https://arxiv.org/pdf/2602.13568v1
+
+## 12387. DistillLens: Symmetric Knowledge Distillation Through Logit Lens
+
+- **ID:** `2602.13567` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Manish Dhakal, Uthman Jinadu, Anjila Budathoki, et al.
+- **abs:** https://arxiv.org/abs/2602.13567 · **PDF:** https://arxiv.org/pdf/2602.13567v1
+
+## 12388. Fast Surrogate Learning for Multi-Objective UAV Placement in Motorway Intelligent Transportation System
+
+- **ID:** `2602.13564` · **primary:** `cs.NE` · **Submitted:** 2026-02-14
+- **Authors:** Weian Guo, Shixin Deng, Wuzhao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.13564 · **PDF:** https://arxiv.org/pdf/2602.13564v1
+
+## 12389. Mitigating the Safety-utility Trade-off in LLM Alignment via Adaptive Safe Context Learning
+
+- **ID:** `2602.13562` · **primary:** `cs.CR` · **Submitted:** 2026-02-14
+- **Authors:** Yanbo Wang, Minzheng Wang, Jian Liang, et al.
+- **abs:** https://arxiv.org/abs/2602.13562 · **PDF:** https://arxiv.org/pdf/2602.13562v1
+
+## 12390. OpAgent: Operator Agent for Web Navigation
+
+- **ID:** `2602.13559` · **primary:** `cs.AI` · **Submitted:** 2026-02-14
+- **Authors:** Yuyu Guo, Wenjie Yang, Siyuan Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.13559 · **PDF:** https://arxiv.org/pdf/2602.13559v1
+
+## 12391. Scenario-Adaptive MU-MIMO OFDM Semantic Communication With Asymmetric Neural Network
+
+- **ID:** `2602.13557` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Chongyang Li, Tianqian Zhang, Shouyin Liu
+- **abs:** https://arxiv.org/abs/2602.13557 · **PDF:** https://arxiv.org/pdf/2602.13557v1
+
+## 12392. Discrete-Space Generative AI Pipeline for Semantic Transmission of Signals
+
+- **ID:** `2602.13556` · **primary:** `cs.IT` · **Submitted:** 2026-02-14
+- **Authors:** Silvija Kokalj-Filipovic, Yagna Kaasaragadda
+- **abs:** https://arxiv.org/abs/2602.13556 · **PDF:** https://arxiv.org/pdf/2602.13556v1
+
+## 12393. Privacy-Concealing Cooperative Perception for BEV Scene Segmentation
+
+- **ID:** `2602.13555` · **primary:** `cs.CV` · **Submitted:** 2026-02-14
+- **Authors:** Song Wang, Lingling Li, Marcus Santos, et al.
+- **abs:** https://arxiv.org/abs/2602.13555 · **PDF:** https://arxiv.org/pdf/2602.13555v1
+
+## 12394. Small Reward Models via Backward Inference
+
+- **ID:** `2602.13551` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Yike Wang, Faeze Brahman, Shangbin Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.13551 · **PDF:** https://arxiv.org/pdf/2602.13551v2
+
+## 12395. Out-of-Support Generalisation via Weight-Space Sequence Modelling
+
+- **ID:** `2602.13550` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Roussel Desmond Nzoyem
+- **abs:** https://arxiv.org/abs/2602.13550 · **PDF:** https://arxiv.org/pdf/2602.13550v3
+
+## 12396. AISA: Awakening Intrinsic Safety Awareness in Large Language Models against Jailbreak Attacks
+
+- **ID:** `2602.13547` · **primary:** `cs.CR` · **Submitted:** 2026-02-14
+- **Authors:** Weiming Song, Xuan Xie, Ruiping Yin
+- **abs:** https://arxiv.org/abs/2602.13547 · **PDF:** https://arxiv.org/pdf/2602.13547v1
+
+## 12397. LiveNewsBench: Evaluating LLM Web Search Capabilities with Freshly Curated News
+
+- **ID:** `2602.13543` · **primary:** `cs.IR` · **Submitted:** 2026-02-14
+- **Authors:** Yunfan Zhang, Kathleen McKeown, Smaranda Muresan
+- **abs:** https://arxiv.org/abs/2602.13543 · **PDF:** https://arxiv.org/pdf/2602.13543v1
+
+## 12398. On Calibration of Large Language Models: From Response To Capability
+
+- **ID:** `2602.13540` · **primary:** `cs.CL` · **Submitted:** 2026-02-14
+- **Authors:** Sin-Han Yang, Cheng-Kuang Wu, Chieh-Yen Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.13540 · **PDF:** https://arxiv.org/pdf/2602.13540v1
+
+## 12399. Fast Swap-Based Element Selection for Multiplication-Free Dimension Reduction
+
+- **ID:** `2602.13532` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Nobutaka Ono
+- **abs:** https://arxiv.org/abs/2602.13532 · **PDF:** https://arxiv.org/pdf/2602.13532v1
+
+## 12400. QuaRK: A Quantum Reservoir Kernel for Time Series Learning
+
+- **ID:** `2602.13531` · **primary:** `cs.LG` · **Submitted:** 2026-02-14
+- **Authors:** Abdallah Aaraba, Soumaya Cherkaoui, Ola Ahmad, et al.
+- **abs:** https://arxiv.org/abs/2602.13531 · **PDF:** https://arxiv.org/pdf/2602.13531v1
+
+## 12401. REMem: Reasoning with Episodic Memory in Language Agent
+
+- **ID:** `2602.13530` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Yiheng Shu, Saisri Padmaja Jonnalagedda, Xiang Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.13530 · **PDF:** https://arxiv.org/pdf/2602.13530v3
+
+## 12402. SecureGate: Learning When to Reveal PII Safely via Token-Gated Dual-Adapters for Federated LLMs
+
+- **ID:** `2602.13529` · **primary:** `cs.CR` · **Submitted:** 2026-02-13
+- **Authors:** Mohamed Shaaban, Mohamed Elmahallawy
+- **abs:** https://arxiv.org/abs/2602.13529 · **PDF:** https://arxiv.org/pdf/2602.13529v1
+
+## 12403. Singular Vectors of Attention Heads Align with Features
+
+- **ID:** `2602.13524` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Gabriel Franco, Carson Loughridge, Mark Crovella
+- **abs:** https://arxiv.org/abs/2602.13524 · **PDF:** https://arxiv.org/pdf/2602.13524v1
+
+## 12404. Arming Data Agents with Tribal Knowledge
+
+- **ID:** `2602.13521` · **primary:** `cs.DB` · **Submitted:** 2026-02-13
+- **Authors:** Shubham Agarwal, Asim Biswal, Sepanta Zeighami, et al.
+- **abs:** https://arxiv.org/abs/2602.13521 · **PDF:** https://arxiv.org/pdf/2602.13521v2
+
+## 12405. Think Deep, Not Just Long: Measuring LLM Reasoning Effort via Deep-Thinking Tokens
+
+- **ID:** `2602.13517` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Wei-Lin Chen, Liqian Peng, Tian Tan, et al.
+- **abs:** https://arxiv.org/abs/2602.13517 · **PDF:** https://arxiv.org/pdf/2602.13517v1
+
+## 12406. SPILLage: Agentic Oversharing on the Web
+
+- **ID:** `2602.13516` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Jaechul Roh, Eugene Bagdasarian, Hamed Haddadi, et al.
+- **abs:** https://arxiv.org/abs/2602.13516 · **PDF:** https://arxiv.org/pdf/2602.13516v1
+
+## 12407. SpargeAttention2: Trainable Sparse Attention via Hybrid Top-k+Top-p Masking and Distillation Fine-Tuning
+
+- **ID:** `2602.13515` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Jintao Zhang, Kai Jiang, Chendong Xiang, et al.
+- **abs:** https://arxiv.org/abs/2602.13515 · **PDF:** https://arxiv.org/pdf/2602.13515v1
+
+## 12408. Learning Gradient Flow: Using Equation Discovery to Accelerate Engineering Optimization
+
+- **ID:** `2602.13513` · **primary:** `math.OC` · **Submitted:** 2026-02-13
+- **Authors:** Grant Norman, Conor Rowan, Kurt Maute, et al.
+- **abs:** https://arxiv.org/abs/2602.13513 · **PDF:** https://arxiv.org/pdf/2602.13513v2
+
+## 12409. Stochastic variance reduced extragradient methods for solving hierarchical variational inequalities
+
+- **ID:** `2602.13510` · **primary:** `math.OC` · **Submitted:** 2026-02-13
+- **Authors:** Pavel Dvurechensky, Andrea Ebner, Johannes Carl Schnebel, et al.
+- **abs:** https://arxiv.org/abs/2602.13510 · **PDF:** https://arxiv.org/pdf/2602.13510v1
+
+## 12410. $γ$-weakly $θ$-up-concavity: Linearizable Non-Convex Optimization with Applications to DR-Submodular and OSS Functions
+
+- **ID:** `2602.13506` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Mohammad Pedramfar, Vaneet Aggarwal
+- **abs:** https://arxiv.org/abs/2602.13506 · **PDF:** https://arxiv.org/pdf/2602.13506v1
+
+## 12411. From Perceptions To Evidence: Detecting AI-Generated Content In Turkish News Media With A Fine-Tuned Bert Classifier
+
+- **ID:** `2602.13504` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Ozancan Ozdemir
+- **abs:** https://arxiv.org/abs/2602.13504 · **PDF:** https://arxiv.org/pdf/2602.13504v1
+
+## 12412. Translating Dietary Standards into Healthy Meals with Minimal Substitutions
+
+- **ID:** `2602.13502` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Trevor Chan, Ilias Tagkopoulos
+- **abs:** https://arxiv.org/abs/2602.13502 · **PDF:** https://arxiv.org/pdf/2602.13502v1
+
+## 12413. TrasMuon: Trust-Region Adaptive Scaling for Orthogonalized Momentum Optimizers
+
+- **ID:** `2602.13498` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Peng Cheng, Jiucheng Zang, Qingnan Li, et al.
+- **abs:** https://arxiv.org/abs/2602.13498 · **PDF:** https://arxiv.org/pdf/2602.13498v2
+
+## 12414. Future of Edge AI in biodiversity monitoring
+
+- **ID:** `2602.13496` · **primary:** `cs.CY` · **Submitted:** 2026-02-13
+- **Authors:** Aude Vuilliomenet, Kate E. Jones, Duncan Wilson
+- **abs:** https://arxiv.org/abs/2602.13496 · **PDF:** https://arxiv.org/pdf/2602.13496v1
+
+## 12415. What Do We Mean by 'Pilot Study': Early Findings from a Meta-Review of Pilot Study Reporting at CHI
+
+- **ID:** `2602.13488` · **primary:** `cs.HC` · **Submitted:** 2026-02-13
+- **Authors:** Belu Ticona, Amna Liaqat, Antonios Anastasopoulos
+- **abs:** https://arxiv.org/abs/2602.13488 · **PDF:** https://arxiv.org/pdf/2602.13488v1
+
+## 12416. Preventing Rank Collapse in Federated Low-Rank Adaptation with Client Heterogeneity
+
+- **ID:** `2602.13486` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Fei Wu, Jia Hu, Geyong Min, et al.
+- **abs:** https://arxiv.org/abs/2602.13486 · **PDF:** https://arxiv.org/pdf/2602.13486v1
+
+## 12417. Federated Learning of Nonlinear Temporal Dynamics with Graph Attention-based Cross-Client Interpretability
+
+- **ID:** `2602.13485` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Ayse Tursucular, Ayush Mohanty, Nazal Mohamed, et al.
+- **abs:** https://arxiv.org/abs/2602.13485 · **PDF:** https://arxiv.org/pdf/2602.13485v1
+
+## 12418. Finding Highly Interpretable Prompt-Specific Circuits in Language Models
+
+- **ID:** `2602.13483` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Gabriel Franco, Lucas M. Tassis, Azalea Rohr, et al.
+- **abs:** https://arxiv.org/abs/2602.13483 · **PDF:** https://arxiv.org/pdf/2602.13483v1
+
+## 12419. Comparing Classifiers: A Case Study Using PyCM
+
+- **ID:** `2602.13482` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Sadra Sabouri, Alireza Zolanvari, Sepand Haghighi
+- **abs:** https://arxiv.org/abs/2602.13482 · **PDF:** https://arxiv.org/pdf/2602.13482v1
+
+## 12420. OMNI-LEAK: Orchestrator Multi-Agent Network Induced Data Leakage
+
+- **ID:** `2602.13477` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Akshat Naik, Jay Culligan, Yarin Gal, et al.
+- **abs:** https://arxiv.org/abs/2602.13477 · **PDF:** https://arxiv.org/pdf/2602.13477v2
+
+## 12421. AsyncVLA: An Asynchronous VLA for Fast and Robust Navigation on the Edge
+
+- **ID:** `2602.13476` · **primary:** `cs.RO` · **Submitted:** 2026-02-13
+- **Authors:** Noriaki Hirose, Catherine Glossop, Dhruv Shah, et al.
+- **abs:** https://arxiv.org/abs/2602.13476 · **PDF:** https://arxiv.org/pdf/2602.13476v1
+
+## 12422. Efficient and Debiased Learning of Average Hazard Under Non-Proportional Hazards
+
+- **ID:** `2602.13475` · **primary:** `stat.ME` · **Submitted:** 2026-02-13
+- **Authors:** Xiang Meng, Lu Tian, Kenneth Kehl, et al.
+- **abs:** https://arxiv.org/abs/2602.13475 · **PDF:** https://arxiv.org/pdf/2602.13475v1
+
+## 12423. NeuroWeaver: An Autonomous Evolutionary Agent for Exploring the Programmatic Space of EEG Analysis Pipelines
+
+- **ID:** `2602.13473` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Guoan Wang, Shihao Yang, Jun-En Ding, et al.
+- **abs:** https://arxiv.org/abs/2602.13473 · **PDF:** https://arxiv.org/pdf/2602.13473v1
+
+## 12424. How Multimodal Large Language Models Support Access to Visual Information: A Diary Study With Blind and Low Vision People
+
+- **ID:** `2602.13469` · **primary:** `cs.HC` · **Submitted:** 2026-02-13
+- **Authors:** Ricardo E. Gonzalez Penuela, Crescentia Jung, Sharon Y Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.13469 · **PDF:** https://arxiv.org/pdf/2602.13469v2
+
+## 12425. Language Model Memory and Memory Models for Language
+
+- **ID:** `2602.13466` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Benjamin L. Badger
+- **abs:** https://arxiv.org/abs/2602.13466 · **PDF:** https://arxiv.org/pdf/2602.13466v1
+
+## 12426. MoltNet: Understanding Social Behavior of AI Agents in the Agent-Native MoltBook
+
+- **ID:** `2602.13458` · **primary:** `cs.SI` · **Submitted:** 2026-02-13
+- **Authors:** Yi Feng, Chen Huang, Zhibo Man, et al.
+- **abs:** https://arxiv.org/abs/2602.13458 · **PDF:** https://arxiv.org/pdf/2602.13458v1
+
+## 12427. Using Machine Learning to Enhance the Detection of Obfuscated Abusive Words in Swahili: A Focus on Child Safety
+
+- **ID:** `2602.13455` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Phyllis Nabangi, Abdul-Jalil Zakaria, Jema David Ndibwile
+- **abs:** https://arxiv.org/abs/2602.13455 · **PDF:** https://arxiv.org/pdf/2602.13455v1
+
+## 12428. LLM-Powered Automatic Translation and Urgency in Crisis Scenarios
+
+- **ID:** `2602.13452` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Belu Ticona, Antonis Anastasopoulos
+- **abs:** https://arxiv.org/abs/2602.13452 · **PDF:** https://arxiv.org/pdf/2602.13452v1
+
+## 12429. End-to-End NOMA with Perfect and Quantized CSI Over Rayleigh Fading Channels
+
+- **ID:** `2602.13446` · **primary:** `cs.IT` · **Submitted:** 2026-02-13
+- **Authors:** Selma Benouadah, Mojtaba Vaezi, Ruizhan Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.13446 · **PDF:** https://arxiv.org/pdf/2602.13446v1
+
+## 12430. FlowHOI: Flow-based Semantics-Grounded Generation of Hand-Object Interactions for Dexterous Robot Manipulation
+
+- **ID:** `2602.13444` · **primary:** `cs.RO` · **Submitted:** 2026-02-13
+- **Authors:** Huajian Zeng, Lingyun Chen, Jiaqi Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.13444 · **PDF:** https://arxiv.org/pdf/2602.13444v1
+
+## 12431. Decorrelating the Future: Joint Frequency Domain Learning for Spatio-temporal Forecasting
+
+- **ID:** `2603.04418` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Zepu Wang, Bowen Liao, Jeff, et al.
+- **abs:** https://arxiv.org/abs/2603.04418 · **PDF:** https://arxiv.org/pdf/2603.04418v1
+
+## 12432. Measuring Neural Network Complexity via Effective Degrees of Freedom
+
+- **ID:** `2602.13442` · **primary:** `stat.ME` · **Submitted:** 2026-02-13
+- **Authors:** Jia Zhou, Douglas Landsittel
+- **abs:** https://arxiv.org/abs/2602.13442 · **PDF:** https://arxiv.org/pdf/2602.13442v1
+
+## 12433. Backdooring Bias in Large Language Models
+
+- **ID:** `2602.13427` · **primary:** `cs.CR` · **Submitted:** 2026-02-13
+- **Authors:** Anudeep Das, Prach Chantasantitam, Gurjot Singh, et al.
+- **abs:** https://arxiv.org/abs/2602.13427 · **PDF:** https://arxiv.org/pdf/2602.13427v1
+
+## 12434. Metabolic cost of information processing in Poisson variational autoencoders
+
+- **ID:** `2602.13421` · **primary:** `stat.ML` · **Submitted:** 2026-02-13
+- **Authors:** Hadi Vafaii, Jacob L. Yates
+- **abs:** https://arxiv.org/abs/2602.13421 · **PDF:** https://arxiv.org/pdf/2602.13421v1
+
+## 12435. Protect$^*$: Steerable Retrosynthesis through Neuro-Symbolic State Encoding
+
+- **ID:** `2602.13419` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-13
+- **Authors:** Shreyas Vinaya Sathyanarayana, Shah Rahil Kirankumar, Sharanabasava D. Hiremath, et al.
+- **abs:** https://arxiv.org/abs/2602.13419 · **PDF:** https://arxiv.org/pdf/2602.13419v1
+
+## 12436. Text Has Curvature
+
+- **ID:** `2602.13418` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Karish Grover, Hanqing Zeng, Yinglong Xia, et al.
+- **abs:** https://arxiv.org/abs/2602.13418 · **PDF:** https://arxiv.org/pdf/2602.13418v1
+
+## 12437. High-Resolution Climate Projections Using Diffusion-Based Downscaling of a Lightweight Climate Emulator
+
+- **ID:** `2602.13416` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Haiwen Guan, Moein Darman, Dibyajyoti Chakraborty, et al.
+- **abs:** https://arxiv.org/abs/2602.13416 · **PDF:** https://arxiv.org/pdf/2602.13416v1
+
+## 12438. FUTON: Fourier Tensor Network for Implicit Neural Representations
+
+- **ID:** `2602.13414` · **primary:** `eess.IV` · **Submitted:** 2026-02-13
+- **Authors:** Pooya Ashtari, Pourya Behmandpoor, Nikos Deligiannis, et al.
+- **abs:** https://arxiv.org/abs/2602.13414 · **PDF:** https://arxiv.org/pdf/2602.13414v1
+
+## 12439. Why is Normalization Preferred? A Worst-Case Complexity Theory for Stochastically Preconditioned SGD under Heavy-Tailed Noise
+
+- **ID:** `2602.13413` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Yuchen Fang, James Demmel, Javad Lavaei
+- **abs:** https://arxiv.org/abs/2602.13413 · **PDF:** https://arxiv.org/pdf/2602.13413v1
+
+## 12440. Evolutionary design of thermodynamic logic gates and their heat emission
+
+- **ID:** `2602.13410` · **primary:** `cond-mat.stat-mech` · **Submitted:** 2026-02-13
+- **Authors:** Stephen Whitelam
+- **abs:** https://arxiv.org/abs/2602.13410 · **PDF:** https://arxiv.org/pdf/2602.13410v1
+
+## 12441. On-Policy Supervised Fine-Tuning for Efficient Reasoning
+
+- **ID:** `2602.13407` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Anhao Zhao, Ziyang Chen, Junlong Tong, et al.
+- **abs:** https://arxiv.org/abs/2602.13407 · **PDF:** https://arxiv.org/pdf/2602.13407v1
+
+## 12442. Accelerated Discovery of Cryoprotectant Cocktails via Multi-Objective Bayesian Optimization
+
+- **ID:** `2602.13398` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Daniel Emerson, Nora Gaby-Biegel, Purva Joshi, et al.
+- **abs:** https://arxiv.org/abs/2602.13398 · **PDF:** https://arxiv.org/pdf/2602.13398v1
+
+## 12443. Imitating What Works: Simulation-Filtered Modular Policy Learning from Human Videos
+
+- **ID:** `2602.13197` · **primary:** `cs.RO` · **Submitted:** 2026-02-13
+- **Authors:** Albert J. Zhai, Kuo-Hao Zeng, Jiasen Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.13197 · **PDF:** https://arxiv.org/pdf/2602.13197v1
+
+## 12444. Semantic Chunking and the Entropy of Natural Language
+
+- **ID:** `2602.13194` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Weishun Zhong, Doron Sivan, Tankut Can, et al.
+- **abs:** https://arxiv.org/abs/2602.13194 · **PDF:** https://arxiv.org/pdf/2602.13194v2
+
+## 12445. CoPE-VideoLM: Leveraging Codec Primitives For Efficient Video Language Modeling
+
+- **ID:** `2602.13191` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Sayan Deb Sarkar, Rémi Pautrat, Ondrej Miksik, et al.
+- **abs:** https://arxiv.org/abs/2602.13191 · **PDF:** https://arxiv.org/pdf/2602.13191v2
+
+## 12446. Profiling systematic uncertainties in Simulation-Based Inference with Factorizable Normalizing Flows
+
+- **ID:** `2602.13184` · **primary:** `hep-ph` · **Submitted:** 2026-02-13
+- **Authors:** Davide Valsecchi, Mauro Donegà, Rainer Wallny
+- **abs:** https://arxiv.org/abs/2602.13184 · **PDF:** https://arxiv.org/pdf/2602.13184v1
+
+## 12447. Selection of CMIP6 Models for Regional Precipitation Projection and Climate Change Assessment in the Jhelum and Chenab River Basins
+
+- **ID:** `2602.13181` · **primary:** `physics.ao-ph` · **Submitted:** 2026-02-13
+- **Authors:** Saad Ahmed Jamal, Ammara Nusrat, Muhammad Azmat, et al.
+- **abs:** https://arxiv.org/abs/2602.13181 · **PDF:** https://arxiv.org/pdf/2602.13181v1
+
+## 12448. Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents
+
+- **ID:** `2602.13379` · **primary:** `cs.CR` · **Submitted:** 2026-02-13
+- **Authors:** Xu Li, Simon Yu, Minzhou Pan, et al.
+- **abs:** https://arxiv.org/abs/2602.13379 · **PDF:** https://arxiv.org/pdf/2602.13379v1
+
+## 12449. Improved Regret Guarantees for Online Mirror Descent using a Portfolio of Mirror Maps
+
+- **ID:** `2602.13177` · **primary:** `math.OC` · **Submitted:** 2026-02-13
+- **Authors:** Swati Gupta, Jai Moondra, Mohit Singh
+- **abs:** https://arxiv.org/abs/2602.13177 · **PDF:** https://arxiv.org/pdf/2602.13177v1
+
+## 12450. Learning functional components of PDEs from data using neural networks
+
+- **ID:** `2602.13174` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Torkel E. Loman, Yurij Salmaniw, Antonio Leon Villares, et al.
+- **abs:** https://arxiv.org/abs/2602.13174 · **PDF:** https://arxiv.org/pdf/2602.13174v1
+
+## 12451. Operator Learning for Families of Finite-State Mean-Field Games
+
+- **ID:** `2602.13169` · **primary:** `math.OC` · **Submitted:** 2026-02-13
+- **Authors:** William Hofgard, Asaf Cohen, Mathieu Laurière
+- **abs:** https://arxiv.org/abs/2602.13169 · **PDF:** https://arxiv.org/pdf/2602.13169v1
+
+## 12452. Realistic Face Reconstruction from Facial Embeddings via Diffusion Models
+
+- **ID:** `2602.13168` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Dong Han, Yong Li, Joachim Denzler
+- **abs:** https://arxiv.org/abs/2602.13168 · **PDF:** https://arxiv.org/pdf/2602.13168v1
+
+## 12453. Optimal Take-off under Fuzzy Clearances
+
+- **ID:** `2602.13166` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Hugo Henry, Arthur Tsai, Kelly Cohen
+- **abs:** https://arxiv.org/abs/2602.13166 · **PDF:** https://arxiv.org/pdf/2602.13166v1
+
+## 12454. Asynchronous Verified Semantic Caching for Tiered LLM Architectures
+
+- **ID:** `2602.13165` · **primary:** `cs.IR` · **Submitted:** 2026-02-13
+- **Authors:** Asmit Kumar Singh, Haozhe Wang, Laxmi Naga Santosh Attaluri, et al.
+- **abs:** https://arxiv.org/abs/2602.13165 · **PDF:** https://arxiv.org/pdf/2602.13165v2
+
+## 12455. LAF-YOLOv10 with Partial Convolution Backbone, Attention-Guided Feature Pyramid, Auxiliary P2 Head, and Wise-IoU Loss for Small Object Detection in Drone Aerial Imagery
+
+- **ID:** `2602.13378` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Sohail Ali Farooqui, Zuhair Ahmed Khan Taha, Mohammed Mudassir Uddin, et al.
+- **abs:** https://arxiv.org/abs/2602.13378 · **PDF:** https://arxiv.org/pdf/2602.13378v1
+
+## 12456. Information-Theoretic Privacy Control for Sequential Multi-Agent LLM Systems
+
+- **ID:** `2603.05520` · **primary:** `cs.MA` · **Submitted:** 2026-02-13
+- **Authors:** Sadia Asif, Mohammad Mohammadi Amiri
+- **abs:** https://arxiv.org/abs/2603.05520 · **PDF:** https://arxiv.org/pdf/2603.05520v1
+
+## 12457. In-Context Autonomous Network Incident Response: An End-to-End Large Language Model Agent Approach
+
+- **ID:** `2602.13156` · **primary:** `cs.CR` · **Submitted:** 2026-02-13
+- **Authors:** Yiran Gao, Kim Hammar, Tao Li
+- **abs:** https://arxiv.org/abs/2602.13156 · **PDF:** https://arxiv.org/pdf/2602.13156v1
+
+## 12458. Learning to Approximate Uniform Facility Location via Graph Neural Networks
+
+- **ID:** `2602.13155` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Chendi Qian, Christopher Morris, Stefanie Jegelka, et al.
+- **abs:** https://arxiv.org/abs/2602.13155 · **PDF:** https://arxiv.org/pdf/2602.13155v1
+
+## 12459. Quantization-Robust LLM Unlearning via Low-Rank Adaptation
+
+- **ID:** `2602.13151` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** João Vitor Boer Abitante, Joana Meneguzzo Pasquali, Luan Fonseca Garcia, et al.
+- **abs:** https://arxiv.org/abs/2602.13151 · **PDF:** https://arxiv.org/pdf/2602.13151v2
+
+## 12460. FlashSchNet: Fast and Accurate Coarse-Grained Neural Network Molecular Dynamics
+
+- **ID:** `2602.13140` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Pingzhi Li, Hongxuan Li, Zirui Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.13140 · **PDF:** https://arxiv.org/pdf/2602.13140v1
+
+## 12461. OpenLID-v3: Improving the Precision of Closely Related Language Identification -- An Experience Report
+
+- **ID:** `2602.13139` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Mariia Fedorova, Nikolay Arefyev, Maja Buljan, et al.
+- **abs:** https://arxiv.org/abs/2602.13139 · **PDF:** https://arxiv.org/pdf/2602.13139v2
+
+## 12462. Order Matters in Retrosynthesis: Structure-aware Generation via Reaction-Center-Guided Discrete Flow Matching
+
+- **ID:** `2602.13136` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Chenguang Wang, Zihan Zhou, Lei Bai, et al.
+- **abs:** https://arxiv.org/abs/2602.13136 · **PDF:** https://arxiv.org/pdf/2602.13136v1
+
+## 12463. Constrained Assumption-Based Argumentation Frameworks
+
+- **ID:** `2602.13135` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Emanuele De Angelis, Fabio Fioravanti, Maria Chiara Meo, et al.
+- **abs:** https://arxiv.org/abs/2602.13135 · **PDF:** https://arxiv.org/pdf/2602.13135v1
+
+## 12464. Eventizing Traditionally Opaque Binary Neural Networks as 1-safe Petri net Models
+
+- **ID:** `2602.13128` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Mohamed Tarraf, Alex Chan, Alex Yakovlev, et al.
+- **abs:** https://arxiv.org/abs/2602.13128 · **PDF:** https://arxiv.org/pdf/2602.13128v1
+
+## 12465. From sunblock to softblock: Analyzing the correlates of neology in published writing and on social media
+
+- **ID:** `2602.13123` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Maria Ryskina, Matthew R. Gormley, Kyle Mahowald, et al.
+- **abs:** https://arxiv.org/abs/2602.13123 · **PDF:** https://arxiv.org/pdf/2602.13123v1
+
+## 12466. Structural Divergence Between AI-Agent and Human Social Networks in Moltbook
+
+- **ID:** `2602.15064` · **primary:** `physics.soc-ph` · **Submitted:** 2026-02-13
+- **Authors:** Wenpin Hou, Zhicheng Ji
+- **abs:** https://arxiv.org/abs/2602.15064 · **PDF:** https://arxiv.org/pdf/2602.15064v1
+
+## 12467. An Online Reference-Free Evaluation Framework for Flowchart Image-to-Code Generation
+
+- **ID:** `2602.13376` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Giang Son Nguyen, Zi Pong Lim, Sarthak Ketanbhai Modi, et al.
+- **abs:** https://arxiv.org/abs/2602.13376 · **PDF:** https://arxiv.org/pdf/2602.13376v1
+
+## 12468. AdaGrad-Diff: A New Version of the Adaptive Gradient Algorithm
+
+- **ID:** `2602.13112` · **primary:** `stat.ML` · **Submitted:** 2026-02-13
+- **Authors:** Matia Bojovic, Saverio Salzo, Massimiliano Pontil
+- **abs:** https://arxiv.org/abs/2602.13112 · **PDF:** https://arxiv.org/pdf/2602.13112v1
+
+## 12469. SCOPE: Selective Conformal Optimized Pairwise LLM Judging
+
+- **ID:** `2602.13110` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Sher Badshah, Ali Emami, Hassan Sajjad
+- **abs:** https://arxiv.org/abs/2602.13110 · **PDF:** https://arxiv.org/pdf/2602.13110v2
+
+## 12470. Which Algorithms Can Graph Neural Networks Learn?
+
+- **ID:** `2602.13106` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Solveig Wittig, Antonis Vasileiou, Robert R. Nerem, et al.
+- **abs:** https://arxiv.org/abs/2602.13106 · **PDF:** https://arxiv.org/pdf/2602.13106v1
+
+## 12471. Random Forests as Statistical Procedures: Design, Variance, and Dependence
+
+- **ID:** `2602.13104` · **primary:** `stat.ML` · **Submitted:** 2026-02-13
+- **Authors:** Nathaniel S. O'Connell
+- **abs:** https://arxiv.org/abs/2602.13104 · **PDF:** https://arxiv.org/pdf/2602.13104v3
+
+## 12472. R-Diverse: Mitigating Diversity Illusion in Self-Play LLM Training
+
+- **ID:** `2602.13103` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Gengsheng Li, Jinghan He, Shijie Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13103 · **PDF:** https://arxiv.org/pdf/2602.13103v2
+
+## 12473. Towards interpretable models for language proficiency assessment: Predicting the CEFR level of Estonian learner texts
+
+- **ID:** `2602.13102` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Kais Allkivi
+- **abs:** https://arxiv.org/abs/2602.13102 · **PDF:** https://arxiv.org/pdf/2602.13102v2
+
+## 12474. Barron-Wiener-Laguerre models
+
+- **ID:** `2602.13098` · **primary:** `stat.ME` · **Submitted:** 2026-02-13
+- **Authors:** Rahul Manavalan, Filip Tronarp
+- **abs:** https://arxiv.org/abs/2602.13098 · **PDF:** https://arxiv.org/pdf/2602.13098v1
+
+## 12475. Consistency of Large Reasoning Models Under Multi-Turn Attacks
+
+- **ID:** `2602.13093` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Yubo Li, Ramayya Krishnan, Rema Padman
+- **abs:** https://arxiv.org/abs/2602.13093 · **PDF:** https://arxiv.org/pdf/2602.13093v3
+
+## 12476. Learning to Remember: End-to-End Training of Memory Agents for Long-Context Reasoning
+
+- **ID:** `2602.18493` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Kehao Zhang, Shangtong Gui, Sheng Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.18493 · **PDF:** https://arxiv.org/pdf/2602.18493v1
+
+## 12477. How cyborg propaganda reshapes collective action
+
+- **ID:** `2602.13088` · **primary:** `cs.CY` · **Submitted:** 2026-02-13
+- **Authors:** Jonas R. Kunst, Kinga Bierwiaczonek, Meeyoung Cha, et al.
+- **abs:** https://arxiv.org/abs/2602.13088 · **PDF:** https://arxiv.org/pdf/2602.13088v1
+
+## 12478. EXCODER: EXplainable Classification Of DiscretE time series Representations
+
+- **ID:** `2602.13087` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Yannik Hahn, Antonin Königsfeld, Hasan Tercan, et al.
+- **abs:** https://arxiv.org/abs/2602.13087 · **PDF:** https://arxiv.org/pdf/2602.13087v1
+
+## 12479. Exploring a New Competency Modeling Process with Large Language Models
+
+- **ID:** `2602.13084` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Silin Du, Manqing Xin, Raymond Jia Wang
+- **abs:** https://arxiv.org/abs/2602.13084 · **PDF:** https://arxiv.org/pdf/2602.13084v1
+
+## 12480. Unified Multi-Domain Graph Pre-training for Homogeneous and Heterogeneous Graphs via Domain-Specific Expert Encoding
+
+- **ID:** `2602.13075` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Chundong Liang, Yongqi Huang, Dongxiao He, et al.
+- **abs:** https://arxiv.org/abs/2602.13075 · **PDF:** https://arxiv.org/pdf/2602.13075v1
+
+## 12481. LCSB: Layer-Cyclic Selective Backpropagation for Memory-Efficient On-Device LLM Fine-Tuning
+
+- **ID:** `2602.13073` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Juneyoung Park, Eunbeen Yoon, Seongwan Kim. Jaeho Lee
+- **abs:** https://arxiv.org/abs/2602.13073 · **PDF:** https://arxiv.org/pdf/2602.13073v1
+
+## 12482. Bus-Conditioned Zero-Shot Trajectory Generation via Task Arithmetic
+
+- **ID:** `2602.13071` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Shuai Liu, Ning Cao, Yile Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.13071 · **PDF:** https://arxiv.org/pdf/2602.13071v1
+
+## 12483. Memory-Efficient Structured Backpropagation for On-Device LLM Fine-Tuning
+
+- **ID:** `2602.13069` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Juneyoung Park, Yuri Hong, Seongwan Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.13069 · **PDF:** https://arxiv.org/pdf/2602.13069v1
+
+## 12484. Backdoor Attacks on Contrastive Continual Learning for IoT Systems
+
+- **ID:** `2602.13062` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Alfous Tim, Kuniyilh Simi D
+- **abs:** https://arxiv.org/abs/2602.13062 · **PDF:** https://arxiv.org/pdf/2602.13062v1
+
+## 12485. Diverging Flows: Detecting Extrapolations in Conditional Generation
+
+- **ID:** `2602.13061` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Constantinos Tsakonas, Serena Ivaldi, Jean-Baptiste Mouret
+- **abs:** https://arxiv.org/abs/2602.13061 · **PDF:** https://arxiv.org/pdf/2602.13061v1
+
+## 12486. TraceBack: Multi-Agent Decomposition for Fine-Grained Table Attribution
+
+- **ID:** `2602.13059` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Tejas Anvekar, Junha Park, Rajat Jha, et al.
+- **abs:** https://arxiv.org/abs/2602.13059 · **PDF:** https://arxiv.org/pdf/2602.13059v1
+
+## 12487. Curriculum-DPO++: Direct Preference Optimization via Data and Model Curricula for Text-to-Image Generation
+
+- **ID:** `2602.13055` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Florinel-Alin Croitoru, Vlad Hondru, Radu Tudor Ionescu, et al.
+- **abs:** https://arxiv.org/abs/2602.13055 · **PDF:** https://arxiv.org/pdf/2602.13055v1
+
+## 12488. Quantization-Aware Collaborative Inference for Large Embodied AI Models
+
+- **ID:** `2602.13052` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Zhonghao Lyu, Ming Xiao, Mikael Skoglund, et al.
+- **abs:** https://arxiv.org/abs/2602.13052 · **PDF:** https://arxiv.org/pdf/2602.13052v1
+
+## 12489. Can we trust AI to detect healthy multilingual English speakers among the cognitively impaired cohort in the UK? An investigation using real-world conversational speech
+
+- **ID:** `2602.13047` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Madhurananda Pahar, Caitlin Illingworth, Dorota Braun, et al.
+- **abs:** https://arxiv.org/abs/2602.13047 · **PDF:** https://arxiv.org/pdf/2602.13047v1
+
+## 12490. Geometric Manifold Rectification for Imbalanced Learning
+
+- **ID:** `2602.13045` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Xubin Wang, Qing Li, Weijia Jia
+- **abs:** https://arxiv.org/abs/2602.13045 · **PDF:** https://arxiv.org/pdf/2602.13045v1
+
+## 12491. A Two-Stage Architecture for NDA Analysis: LLM-based Segmentation and Transformer-based Clause Classification
+
+- **ID:** `2603.09990` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Ana Begnini, Matheus Vicente, Leonardo Souza
+- **abs:** https://arxiv.org/abs/2603.09990 · **PDF:** https://arxiv.org/pdf/2603.09990v1
+
+## 12492. GPTZero: Robust Detection of LLM-Generated Texts
+
+- **ID:** `2602.13042` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** George Alexandru Adam, Alexander Cui, Edwin Thomas, et al.
+- **abs:** https://arxiv.org/abs/2602.13042 · **PDF:** https://arxiv.org/pdf/2602.13042v1
+
+## 12493. TCRL: Temporal-Coupled Adversarial Training for Robust Constrained Reinforcement Learning in Worst-Case Scenarios
+
+- **ID:** `2602.13040` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Wentao Xu, Zhongming Yao, Weihao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.13040 · **PDF:** https://arxiv.org/pdf/2602.13040v1
+
+## 12494. Look Inward to Explore Outward: Learning Temperature Policy from LLM Internal States via Hierarchical RL
+
+- **ID:** `2602.13035` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Yixiao Zhou, Yang Li, Dongzhou Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.13035 · **PDF:** https://arxiv.org/pdf/2602.13035v1
+
+## 12495. MoralityGym: A Benchmark for Evaluating Hierarchical Moral Alignment in Sequential Decision-Making Agents
+
+- **ID:** `2602.13372` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Simon Rosen, Siddarth Singh, Ebenezer Gelo, et al.
+- **abs:** https://arxiv.org/abs/2602.13372 · **PDF:** https://arxiv.org/pdf/2602.13372v1
+
+## 12496. Buy versus Build an LLM: A Decision Framework for Governments
+
+- **ID:** `2602.13033` · **primary:** `cs.CY` · **Submitted:** 2026-02-13
+- **Authors:** Jiahao Lu, Ziwei Xu, William Tjhi, et al.
+- **abs:** https://arxiv.org/abs/2602.13033 · **PDF:** https://arxiv.org/pdf/2602.13033v2
+
+## 12497. Resource-Efficient Gesture Recognition through Convexified Attention
+
+- **ID:** `2602.13030` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Daniel Schwartz, Dario Salvucci, Yusuf Osmanlioglu, et al.
+- **abs:** https://arxiv.org/abs/2602.13030 · **PDF:** https://arxiv.org/pdf/2602.13030v1
+
+## 12498. Human-Aligned MLLM Judges for Fine-Grained Image Editing Evaluation: A Benchmark, Framework, and Analysis
+
+- **ID:** `2602.13028` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Runzhou Liu, Hailey Weingord, Sejal Mittal, et al.
+- **abs:** https://arxiv.org/abs/2602.13028 · **PDF:** https://arxiv.org/pdf/2602.13028v1
+
+## 12499. FedHENet: A Frugal Federated Learning Framework for Heterogeneous Environments
+
+- **ID:** `2602.13024` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Alejandro Dopico-Castro, Oscar Fontenla-Romero, Bertha Guijarro-Berdiñas, et al.
+- **abs:** https://arxiv.org/abs/2602.13024 · **PDF:** https://arxiv.org/pdf/2602.13024v1
+
+## 12500. Prior-Guided Symbolic Regression: Towards Scientific Consistency in Equation Discovery
+
+- **ID:** `2602.13021` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Jing Xiao, Xinhai Chen, Jiaming Peng, et al.
+- **abs:** https://arxiv.org/abs/2602.13021 · **PDF:** https://arxiv.org/pdf/2602.13021v2
+
+## 12501. Synaptic Activation and Dual Liquid Dynamics for Interpretable Bio-Inspired Models
+
+- **ID:** `2602.13017` · **primary:** `cs.NE` · **Submitted:** 2026-02-13
+- **Authors:** Mónika Farsang, Radu Grosu
+- **abs:** https://arxiv.org/abs/2602.13017 · **PDF:** https://arxiv.org/pdf/2602.13017v1
+
+## 12502. Probabilistic Wind Power Forecasting with Tree-Based Machine Learning and Weather Ensembles
+
+- **ID:** `2602.13010` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Max Bruninx, Diederik van Binsbergen, Timothy Verstraeten, et al.
+- **abs:** https://arxiv.org/abs/2602.13010 · **PDF:** https://arxiv.org/pdf/2602.13010v1
+
+## 12503. Machine Learning-Based Classification of Jhana Advanced Concentrative Absorption Meditation (ACAM-J) using 7T fMRI
+
+- **ID:** `2602.13008` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Puneet Kumar, Winson F. Z. Yang, Alakhsimar Singh, et al.
+- **abs:** https://arxiv.org/abs/2602.13008 · **PDF:** https://arxiv.org/pdf/2602.13008v1
+
+## 12504. Uncertainty in Federated Granger Causality: From Origins to Systemic Consequences
+
+- **ID:** `2602.13004` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Ayush Mohanty, Nazal Mohamed, Nagi Gebraeel
+- **abs:** https://arxiv.org/abs/2602.13004 · **PDF:** https://arxiv.org/pdf/2602.13004v1
+
+## 12505. MASAR: Motion-Appearance Synergy Refinement for Joint Detection and Trajectory Forecasting
+
+- **ID:** `2602.13003` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Mohammed Amine Bencheikh Lehocine, Julian Schmidt, Frank Moosmann, et al.
+- **abs:** https://arxiv.org/abs/2602.13003 · **PDF:** https://arxiv.org/pdf/2602.13003v1
+
+## 12506. Know More, Know Clearer: A Meta-Cognitive Framework for Knowledge Augmentation in Large Language Models
+
+- **ID:** `2602.12996` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Hao Chen, Ye He, Yuchun Fan, et al.
+- **abs:** https://arxiv.org/abs/2602.12996 · **PDF:** https://arxiv.org/pdf/2602.12996v1
+
+## 12507. Evaluating the Homogeneity of Keyphrase Prediction Models
+
+- **ID:** `2602.12989` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Maël Houbre, Florian Boudin, Beatrice Daille
+- **abs:** https://arxiv.org/abs/2602.12989 · **PDF:** https://arxiv.org/pdf/2602.12989v1
+
+## 12508. SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents
+
+- **ID:** `2602.12984` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Yujiong Shen, Yajie Yang, Zhiheng Xi, et al.
+- **abs:** https://arxiv.org/abs/2602.12984 · **PDF:** https://arxiv.org/pdf/2602.12984v1
+
+## 12509. Detecting Object Tracking Failure via Sequential Hypothesis Testing
+
+- **ID:** `2602.12983` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Alejandro Monroy Muñoz, Rajeev Verma, Alexander Timans
+- **abs:** https://arxiv.org/abs/2602.12983 · **PDF:** https://arxiv.org/pdf/2602.12983v1
+
+## 12510. Multi-Dimensional Visual Data Recovery: Scale-Aware Tensor Modeling and Accelerated Randomized Computation
+
+- **ID:** `2602.12982` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Wenjin Qin, Hailin Wang, Jiangjun Peng, et al.
+- **abs:** https://arxiv.org/abs/2602.12982 · **PDF:** https://arxiv.org/pdf/2602.12982v1
+
+## 12511. MAUNet-Light: A Concise MAUNet Architecture for Bias Correction and Downscaling of Precipitation Estimates
+
+- **ID:** `2602.12980` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Sumanta Chandra Mishra Sharma, Adway Mitra, Auroop Ratan Ganguly
+- **abs:** https://arxiv.org/abs/2602.12980 · **PDF:** https://arxiv.org/pdf/2602.12980v1
+
+## 12512. Learning Native Continuation for Action Chunking Flow Policies
+
+- **ID:** `2602.12978` · **primary:** `cs.RO` · **Submitted:** 2026-02-13
+- **Authors:** Yufeng Liu, Hang Yu, Juntu Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.12978 · **PDF:** https://arxiv.org/pdf/2602.12978v1
+
+## 12513. Drift-Aware Variational Autoencoder-based Anomaly Detection with Two-level Ensembling
+
+- **ID:** `2602.12976` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Jin Li, Kleanthis Malialis, Christos G. Panayiotou, et al.
+- **abs:** https://arxiv.org/abs/2602.12976 · **PDF:** https://arxiv.org/pdf/2602.12976v1
+
+## 12514. Extending confidence calibration to generalised measures of variation
+
+- **ID:** `2602.12975` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Andrew Thompson, Vivek Desai
+- **abs:** https://arxiv.org/abs/2602.12975 · **PDF:** https://arxiv.org/pdf/2602.12975v1
+
+## 12515. Length Generalization Bounds for Transformers
+
+- **ID:** `2603.02238` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Andy Yang, Pascal Bergsträßer, Georg Zetzsche, et al.
+- **abs:** https://arxiv.org/abs/2603.02238 · **PDF:** https://arxiv.org/pdf/2603.02238v1
+
+## 12516. Jointly Optimizing Debiased CTR and Uplift for Coupons Marketing: A Unified Causal Framework
+
+- **ID:** `2602.12972` · **primary:** `cs.SI` · **Submitted:** 2026-02-13
+- **Authors:** Siyun Yang, Shixiao Yang, Jian Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.12972 · **PDF:** https://arxiv.org/pdf/2602.12972v1
+
+## 12517. RGAlign-Rec: Ranking-Guided Alignment for Latent Query Reasoning in Recommendation Systems
+
+- **ID:** `2602.12968` · **primary:** `cs.IR` · **Submitted:** 2026-02-13
+- **Authors:** Junhua Liu, Yang Jihao, Cheng Chang, et al.
+- **abs:** https://arxiv.org/abs/2602.12968 · **PDF:** https://arxiv.org/pdf/2602.12968v1
+
+## 12518. ProbeLLM: Automating Principled Diagnosis of LLM Failures
+
+- **ID:** `2602.12966` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Yue Huang, Zhengzhe Jiang, Yuchen Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.12966 · **PDF:** https://arxiv.org/pdf/2602.12966v1
+
+## 12519. Information-theoretic analysis of world models in optimal reward maximizers
+
+- **ID:** `2602.12963` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Alfred Harwood, Jose Faustino, Alex Altair
+- **abs:** https://arxiv.org/abs/2602.12963 · **PDF:** https://arxiv.org/pdf/2602.12963v1
+
+## 12520. TriGen: NPU Architecture for End-to-End Acceleration of Large Language Models based on SW-HW Co-Design
+
+- **ID:** `2602.12962` · **primary:** `cs.AR` · **Submitted:** 2026-02-13
+- **Authors:** Jonghun Lee, Junghoon Lee, Hyeonjin Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.12962 · **PDF:** https://arxiv.org/pdf/2602.12962v1
+
+## 12521. Ca-MCF: Category-level Multi-label Causal Feature selection
+
+- **ID:** `2602.12961` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Wanfu Gao, Yanan Wang, Yonghao Li
+- **abs:** https://arxiv.org/abs/2602.12961 · **PDF:** https://arxiv.org/pdf/2602.12961v1
+
+## 12522. Transporting Task Vectors across Different Architectures without Training
+
+- **ID:** `2602.12952` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Filippo Rinaldi, Aniello Panariello, Giacomo Salici, et al.
+- **abs:** https://arxiv.org/abs/2602.12952 · **PDF:** https://arxiv.org/pdf/2602.12952v1
+
+## 12523. G2CP: A Graph-Grounded Communication Protocol for Verifiable and Efficient Multi-Agent Reasoning
+
+- **ID:** `2602.13370` · **primary:** `cs.MA` · **Submitted:** 2026-02-13
+- **Authors:** Karim Ben Khaled, Davy Monticolo
+- **abs:** https://arxiv.org/abs/2602.13370 · **PDF:** https://arxiv.org/pdf/2602.13370v1
+
+## 12524. Deep-Learning Atlas Registration for Melanoma Brain Metastases: Preserving Pathology While Enabling Cohort-Level Analyses
+
+- **ID:** `2602.12933` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Nanna E. Wielenberg, Ilinca Popp, Oliver Blanck, et al.
+- **abs:** https://arxiv.org/abs/2602.12933 · **PDF:** https://arxiv.org/pdf/2602.12933v1
+
+## 12525. TFTF: Training-Free Targeted Flow for Conditional Sampling
+
+- **ID:** `2602.12932` · **primary:** `stat.ML` · **Submitted:** 2026-02-13
+- **Authors:** Qianqian Qu, Jun S. Liu
+- **abs:** https://arxiv.org/abs/2602.12932 · **PDF:** https://arxiv.org/pdf/2602.12932v1
+
+## 12526. Never say never: Exploring the effects of available knowledge on agent persuasiveness in controlled physiotherapy motivation dialogues
+
+- **ID:** `2602.12924` · **primary:** `cs.HC` · **Submitted:** 2026-02-13
+- **Authors:** Stephan Vonschallen, Rahel Häusler, Theresa Schmiedel, et al.
+- **abs:** https://arxiv.org/abs/2602.12924 · **PDF:** https://arxiv.org/pdf/2602.12924v2
+
+## 12527. Annealing in variational inference mitigates mode collapse: A theoretical study on Gaussian mixtures
+
+- **ID:** `2602.12923` · **primary:** `stat.ML` · **Submitted:** 2026-02-13
+- **Authors:** Luigi Fogliani, Bruno Loureiro, Marylou Gabrié
+- **abs:** https://arxiv.org/abs/2602.12923 · **PDF:** https://arxiv.org/pdf/2602.12923v1
+
+## 12528. When Words Don't Mean What They Say: Figurative Understanding in Bengali Idioms
+
+- **ID:** `2602.12921` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Adib Sakhawat, Shamim Ara Parveen, Md Ruhul Amin, et al.
+- **abs:** https://arxiv.org/abs/2602.12921 · **PDF:** https://arxiv.org/pdf/2602.12921v1
+
+## 12529. EPRBench: A High-Quality Benchmark Dataset for Event Stream Based Visual Place Recognition
+
+- **ID:** `2602.12919` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Xiao Wang, Xingxing Xiong, Jinfeng Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.12919 · **PDF:** https://arxiv.org/pdf/2602.12919v1
+
+## 12530. Ultrasound-Guided Real-Time Spinal Motion Visualization for Spinal Instability Assessment
+
+- **ID:** `2602.12917` · **primary:** `physics.med-ph` · **Submitted:** 2026-02-13
+- **Authors:** Feng Li, Yuan Bi, Tianyu Song, et al.
+- **abs:** https://arxiv.org/abs/2602.12917 · **PDF:** https://arxiv.org/pdf/2602.12917v1
+
+## 12531. Reliable Thinking with Images
+
+- **ID:** `2602.12916` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Haobin Li, Yutong Yang, Yijie Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.12916 · **PDF:** https://arxiv.org/pdf/2602.12916v2
+
+## 12532. ViMedCSS: A Vietnamese Medical Code-Switching Speech Dataset & Benchmark
+
+- **ID:** `2602.12911` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Tung X. Nguyen, Nhu Vo, Giang-Son Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.12911 · **PDF:** https://arxiv.org/pdf/2602.12911v1
+
+## 12533. Nanbeige4.1-3B: A Small General Model that Reasons, Aligns, and Acts
+
+- **ID:** `2602.13367` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Chen Yang, Guangyue Peng, Jiaying Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.13367 · **PDF:** https://arxiv.org/pdf/2602.13367v1
+
+## 12534. Nonparametric Contextual Online Bilateral Trade
+
+- **ID:** `2602.12904` · **primary:** `cs.GT` · **Submitted:** 2026-02-13
+- **Authors:** Emanuele Coccia, Martino Bernasconi, Andrea Celli
+- **abs:** https://arxiv.org/abs/2602.12904 · **PDF:** https://arxiv.org/pdf/2602.12904v1
+
+## 12535. Contextual Online Bilateral Trade
+
+- **ID:** `2602.12903` · **primary:** `cs.GT` · **Submitted:** 2026-02-13
+- **Authors:** Romain Cosson, Federico Fusco, Anupam Gupta, et al.
+- **abs:** https://arxiv.org/abs/2602.12903 · **PDF:** https://arxiv.org/pdf/2602.12903v1
+
+## 12536. Robustness of Object Detection of Autonomous Vehicles in Adverse Weather Conditions
+
+- **ID:** `2602.12902` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Fox Pettersen, Hong Zhu
+- **abs:** https://arxiv.org/abs/2602.12902 · **PDF:** https://arxiv.org/pdf/2602.12902v1
+
+## 12537. Blessings of Multiple Good Arms in Multi-Objective Linear Bandits
+
+- **ID:** `2602.12901` · **primary:** `stat.ML` · **Submitted:** 2026-02-13
+- **Authors:** Heesang Ann, Min-hwan Oh
+- **abs:** https://arxiv.org/abs/2602.12901 · **PDF:** https://arxiv.org/pdf/2602.12901v1
+
+## 12538. RADAR: Revealing Asymmetric Development of Abilities in MLLM Pre-training
+
+- **ID:** `2602.12892` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Yunshuang Nie, Bingqian Lin, Minzhe Niu, et al.
+- **abs:** https://arxiv.org/abs/2602.12892 · **PDF:** https://arxiv.org/pdf/2602.12892v1
+
+## 12539. BaziQA-Benchmark: Evaluating Symbolic and Temporally Compositional Reasoning in Large Language Models
+
+- **ID:** `2602.12889` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Jiangxi Chen, Qian Liu
+- **abs:** https://arxiv.org/abs/2602.12889 · **PDF:** https://arxiv.org/pdf/2602.12889v1
+
+## 12540. Safe-SDL:Establishing Safety Boundaries and Control Mechanisms for AI-Driven Self-Driving Laboratories
+
+- **ID:** `2602.15061` · **primary:** `cs.RO` · **Submitted:** 2026-02-13
+- **Authors:** Zihan Zhang, Haohui Que, Junhan Chang, et al.
+- **abs:** https://arxiv.org/abs/2602.15061 · **PDF:** https://arxiv.org/pdf/2602.15061v1
+
+## 12541. Semantic Communities and Boundary-Spanning Lyrics in K-pop: A Graph-Based Unsupervised Analysis
+
+- **ID:** `2602.12881` · **primary:** `cs.SI` · **Submitted:** 2026-02-13
+- **Authors:** Oktay Karakuş
+- **abs:** https://arxiv.org/abs/2602.12881 · **PDF:** https://arxiv.org/pdf/2602.12881v1
+
+## 12542. BrowseComp-$V^3$: A Visual, Vertical, and Verifiable Benchmark for Multimodal Browsing Agents
+
+- **ID:** `2602.12876` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Huanyao Zhang, Jiepeng Zhou, Bo Li, et al.
+- **abs:** https://arxiv.org/abs/2602.12876 · **PDF:** https://arxiv.org/pdf/2602.12876v2
+
+## 12543. A Microservice-Based Platform for Sustainable and Intelligent SLO Fulfilment and Service Management
+
+- **ID:** `2602.12875` · **primary:** `cs.SE` · **Submitted:** 2026-02-13
+- **Authors:** Juan Luis Herrera, Daniel Wang, Schahram Dustdar
+- **abs:** https://arxiv.org/abs/2602.12875 · **PDF:** https://arxiv.org/pdf/2602.12875v1
+
+## 12544. Knowledge-Based Design Requirements for Generative Social Robots in Higher Education
+
+- **ID:** `2602.12873` · **primary:** `cs.HC` · **Submitted:** 2026-02-13
+- **Authors:** Stephan Vonschallen, Dominique Oberle, Theresa Schmiedel, et al.
+- **abs:** https://arxiv.org/abs/2602.12873 · **PDF:** https://arxiv.org/pdf/2602.12873v3
+
+## 12545. MentalBench: A Benchmark for Evaluating Psychiatric Diagnostic Capability of Large Language Models
+
+- **ID:** `2602.12871` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Hoyun Song, Migyeong Kang, Jisu Shin, et al.
+- **abs:** https://arxiv.org/abs/2602.12871 · **PDF:** https://arxiv.org/pdf/2602.12871v1
+
+## 12546. X-VORTEX: Spatio-Temporal Contrastive Learning for Wake Vortex Trajectory Forecasting
+
+- **ID:** `2602.12869` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Zhan Qu, Michael Färber
+- **abs:** https://arxiv.org/abs/2602.12869 · **PDF:** https://arxiv.org/pdf/2602.12869v1
+
+## 12547. Model-Aware Rate-Distortion Limits for Task-Oriented Source Coding
+
+- **ID:** `2602.12866` · **primary:** `cs.IT` · **Submitted:** 2026-02-13
+- **Authors:** Andriy Enttsel, Vincent Corlay
+- **abs:** https://arxiv.org/abs/2602.12866 · **PDF:** https://arxiv.org/pdf/2602.12866v1
+
+## 12548. Assessing Spear-Phishing Website Generation in Large Language Model Coding Agents
+
+- **ID:** `2602.13363` · **primary:** `cs.CR` · **Submitted:** 2026-02-13
+- **Authors:** Tailia Malloy, Tegawende F. Bissyande
+- **abs:** https://arxiv.org/abs/2602.13363 · **PDF:** https://arxiv.org/pdf/2602.13363v1
+
+## 12549. CLOT: Closed-Loop Global Motion Tracking for Whole-Body Humanoid Teleoperation
+
+- **ID:** `2602.15060` · **primary:** `cs.RO` · **Submitted:** 2026-02-13
+- **Authors:** Tengjie Zhu, Guanyu Cai, Yang Zhaohui, et al.
+- **abs:** https://arxiv.org/abs/2602.15060 · **PDF:** https://arxiv.org/pdf/2602.15060v2
+
+## 12550. WebClipper: Efficient Evolution of Web Agents with Graph-based Trajectory Pruning
+
+- **ID:** `2602.12852` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Junjie Wang, Zequn Xie, Dan Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.12852 · **PDF:** https://arxiv.org/pdf/2602.12852v1
+
+## 12551. Chimera: Neuro-Symbolic Attention Primitives for Trustworthy Dataplane Intelligence
+
+- **ID:** `2602.12851` · **primary:** `cs.NI` · **Submitted:** 2026-02-13
+- **Authors:** Rong Fu, Xiaowen Ma, Kun Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.12851 · **PDF:** https://arxiv.org/pdf/2602.12851v2
+
+## 12552. Amortized Reasoning Tree Search: Decoupling Proposal and Decision in Large Language Models
+
+- **ID:** `2602.12846` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Zesheng Hong, Jiadong Yu, Hui Pan
+- **abs:** https://arxiv.org/abs/2602.12846 · **PDF:** https://arxiv.org/pdf/2602.12846v1
+
+## 12553. Nonparametric Distribution Regression Re-calibration
+
+- **ID:** `2602.13362` · **primary:** `stat.ML` · **Submitted:** 2026-02-13
+- **Authors:** Ádám Jung, Domokos M. Kelen, András A. Benczúr
+- **abs:** https://arxiv.org/abs/2602.13362 · **PDF:** https://arxiv.org/pdf/2602.13362v1
+
+## 12554. TRACE: Temporal Reasoning via Agentic Context Evolution for Streaming Electronic Health Records (EHRs)
+
+- **ID:** `2602.12833` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Zhan Qu, Michael Färber
+- **abs:** https://arxiv.org/abs/2602.12833 · **PDF:** https://arxiv.org/pdf/2602.12833v1
+
+## 12555. FLAC: Maximum Entropy RL via Kinetic Energy Regularized Bridge Matching
+
+- **ID:** `2602.12829` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Lei Lv, Yunfei Li, Yu Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.12829 · **PDF:** https://arxiv.org/pdf/2602.12829v1
+
+## 12556. GRAIL: Geometry-Aware Retrieval-Augmented Inference with LLMs over Hyperbolic Representations of Patient Trajectories
+
+- **ID:** `2602.12828` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Zhan Qu, Michael Färber
+- **abs:** https://arxiv.org/abs/2602.12828 · **PDF:** https://arxiv.org/pdf/2602.12828v1
+
+## 12557. Reliable Hierarchical Operating System Fingerprinting via Conformal Prediction
+
+- **ID:** `2602.12825` · **primary:** `cs.CR` · **Submitted:** 2026-02-13
+- **Authors:** Rubén Pérez-Jove, Osvaldo Simeone, Alejandro Pazos, et al.
+- **abs:** https://arxiv.org/abs/2602.12825 · **PDF:** https://arxiv.org/pdf/2602.12825v1
+
+## 12558. Concept Heterogeneity-aware Representation Steering
+
+- **ID:** `2603.02237` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Laziz U. Abdullaev, Noelle Y. L. Wong, Ryan T. Z. Lee, et al.
+- **abs:** https://arxiv.org/abs/2603.02237 · **PDF:** https://arxiv.org/pdf/2603.02237v1
+
+## 12559. AIWizards at MULTIPRIDE: A Hierarchical Approach to Slur Reclamation Detection
+
+- **ID:** `2602.12818` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Luca Tedeschini, Matteo Fasulo
+- **abs:** https://arxiv.org/abs/2602.12818 · **PDF:** https://arxiv.org/pdf/2602.12818v1
+
+## 12560. Left-right asymmetry in predicting brain activity from LLMs' representations emerges with their formal linguistic competence
+
+- **ID:** `2602.12811` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Laurent Bonnasse-Gahot, Christophe Pallier
+- **abs:** https://arxiv.org/abs/2602.12811 · **PDF:** https://arxiv.org/pdf/2602.12811v1
+
+## 12561. RAT-Bench: A Comprehensive Benchmark for Text Anonymization
+
+- **ID:** `2602.12806` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Nataša Krčo, Zexi Yao, Matthieu Meeus, et al.
+- **abs:** https://arxiv.org/abs/2602.12806 · **PDF:** https://arxiv.org/pdf/2602.12806v1
+
+## 12562. The Speed-up Factor: A Quantitative Multi-Iteration Active Learning Performance Metric
+
+- **ID:** `2602.13359` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Hannes Kath, Thiago S. Gouvêa, Daniel Sonntag
+- **abs:** https://arxiv.org/abs/2602.13359 · **PDF:** https://arxiv.org/pdf/2602.13359v1
+
+## 12563. Can Neural Networks Provide Latent Embeddings for Telemetry-Aware Greedy Routing?
+
+- **ID:** `2602.12798` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Andreas Boltres, Niklas Freymuth, Gerhard Neumann
+- **abs:** https://arxiv.org/abs/2602.12798 · **PDF:** https://arxiv.org/pdf/2602.12798v1
+
+## 12564. The System Hallucination Scale (SHS): A Minimal yet Effective Human-Centered Instrument for Evaluating Hallucination-Related Behavior in Large Language Models
+
+- **ID:** `2603.09989` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Heimo Müller, Dominik Steiger, Markus Plass, et al.
+- **abs:** https://arxiv.org/abs/2603.09989 · **PDF:** https://arxiv.org/pdf/2603.09989v1
+
+## 12565. SQuTR: A Robustness Benchmark for Spoken Query to Text Retrieval under Acoustic Noise
+
+- **ID:** `2602.12783` · **primary:** `cs.IR` · **Submitted:** 2026-02-13
+- **Authors:** Yuejie Li, Ke Yang, Yueying Hua, et al.
+- **abs:** https://arxiv.org/abs/2602.12783 · **PDF:** https://arxiv.org/pdf/2602.12783v1
+
+## 12566. Aspect-Based Sentiment Analysis for Future Tourism Experiences: A BERT-MoE Framework for Persian User Reviews
+
+- **ID:** `2602.12778` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Hamidreza Kazemi Taskooh, Taha Zare Harofte
+- **abs:** https://arxiv.org/abs/2602.12778 · **PDF:** https://arxiv.org/pdf/2602.12778v1
+
+## 12567. "Not Human, Funnier": How Machine Identity Shapes Humor Perception in Online AI Stand-up Comedy
+
+- **ID:** `2602.12763` · **primary:** `cs.HC` · **Submitted:** 2026-02-13
+- **Authors:** Xuehan Huang, Canwen Wang, Yifei Hao, et al.
+- **abs:** https://arxiv.org/abs/2602.12763 · **PDF:** https://arxiv.org/pdf/2602.12763v1
+
+## 12568. CUDABench: Benchmarking LLMs for Text-to-CUDA Generation
+
+- **ID:** `2603.02236` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Jiace Zhu, Wentao Chen, Qi Fan, et al.
+- **abs:** https://arxiv.org/abs/2603.02236 · **PDF:** https://arxiv.org/pdf/2603.02236v1
+
+## 12569. Towards a Diagnostic and Predictive Evaluation Methodology for Sequence Labeling Tasks
+
+- **ID:** `2602.12759` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Elena Alvarez-Mellado, Julio Gonzalo
+- **abs:** https://arxiv.org/abs/2602.12759 · **PDF:** https://arxiv.org/pdf/2602.12759v2
+
+## 12570. VineetVC: Adaptive Video Conferencing Under Severe Bandwidth Constraints Using Audio-Driven Talking-Head Reconstruction
+
+- **ID:** `2602.12758` · **primary:** `eess.IV` · **Submitted:** 2026-02-13
+- **Authors:** Vineet Kumar Rakesh, Soumya Mazumdar, Tapas Samanta, et al.
+- **abs:** https://arxiv.org/abs/2602.12758 · **PDF:** https://arxiv.org/pdf/2602.12758v1
+
+## 12571. Talking with Verifiers: Automatic Specification Generation for Neural Network Verification
+
+- **ID:** `2603.02235` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Yizhak Y. Elboher, Reuven Peleg, Zhouxing Shi, et al.
+- **abs:** https://arxiv.org/abs/2603.02235 · **PDF:** https://arxiv.org/pdf/2603.02235v1
+
+## 12572. Closing the Loop: A Control-Theoretic Framework for Provably Stable Time Series Forecasting with LLMs
+
+- **ID:** `2602.12756` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Xingyu Zhang, Hanyun Du, Zeen Song, et al.
+- **abs:** https://arxiv.org/abs/2602.12756 · **PDF:** https://arxiv.org/pdf/2602.12756v1
+
+## 12573. Hierarchical Successor Representation for Robust Transfer
+
+- **ID:** `2602.12753` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Changmin Yu, Máté Lengyel
+- **abs:** https://arxiv.org/abs/2602.12753 · **PDF:** https://arxiv.org/pdf/2602.12753v1
+
+## 12574. X-SYS: A Reference Architecture for Interactive Explanation Systems
+
+- **ID:** `2602.12748` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Tobias Labarta, Nhi Hoang, Maximilian Dreyer, et al.
+- **abs:** https://arxiv.org/abs/2602.12748 · **PDF:** https://arxiv.org/pdf/2602.12748v2
+
+## 12575. Lamer-SSL: Layer-aware Mixture of LoRA Experts for Continual Multilingual Expansion of Self-supervised Models without Forgetting
+
+- **ID:** `2602.12746` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Jing Xu, Minglin Wu, Xueyuan Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.12746 · **PDF:** https://arxiv.org/pdf/2602.12746v1
+
+## 12576. Adaptive Structured Pruning of Convolutional Neural Networks for Time Series Classification
+
+- **ID:** `2602.12744` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Javidan Abdullayev, Maxime Devanne, Cyril Meyer, et al.
+- **abs:** https://arxiv.org/abs/2602.12744 · **PDF:** https://arxiv.org/pdf/2602.12744v1
+
+## 12577. Synthetic Craquelure Generation for Unsupervised Painting Restoration
+
+- **ID:** `2602.12742` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Jana Cuch-Guillén, Antonio Agudo, Raül Pérez-Gonzalo
+- **abs:** https://arxiv.org/abs/2602.12742 · **PDF:** https://arxiv.org/pdf/2602.12742v1
+
+## 12578. VimRAG: Navigating Massive Visual Context in Retrieval-Augmented Generation via Multimodal Memory Graph
+
+- **ID:** `2602.12735` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Qiuchen Wang, Shihang Wang, Yu Zeng, et al.
+- **abs:** https://arxiv.org/abs/2602.12735 · **PDF:** https://arxiv.org/pdf/2602.12735v1
+
+## 12579. ADEPT: RL-Aligned Agentic Decoding of Emotion via Evidence Probing Tools -- From Consensus Learning to Ambiguity-Driven Emotion Reasoning
+
+- **ID:** `2602.12714` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Esther Sun, Bo-Hao Su, Abinay Reddy Naini, et al.
+- **abs:** https://arxiv.org/abs/2602.12714 · **PDF:** https://arxiv.org/pdf/2602.12714v1
+
+## 12580. Reverse Delegated Training and Private Inference via Perfectly-Secure Quantum Homomorphic Encryption
+
+- **ID:** `2602.12712` · **primary:** `quant-ph` · **Submitted:** 2026-02-13
+- **Authors:** Sergio A. Ortega, Miguel A. Martin-Delgado
+- **abs:** https://arxiv.org/abs/2602.12712 · **PDF:** https://arxiv.org/pdf/2602.12712v1
+
+## 12581. ReFilter: Improving Robustness of Retrieval-Augmented Generation via Gated Filter
+
+- **ID:** `2602.12709` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Yixin Chen, Ying Xiong, Shangyu Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.12709 · **PDF:** https://arxiv.org/pdf/2602.12709v1
+
+## 12582. Mixture of Predefined Experts: Maximizing Data Usage on Vertical Federated Learning
+
+- **ID:** `2602.12708` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Jon Irureta, Gorka Azkune, Jon Imaz, et al.
+- **abs:** https://arxiv.org/abs/2602.12708 · **PDF:** https://arxiv.org/pdf/2602.12708v1
+
+## 12583. Physics-Informed Laplace Neural Operator for Solving Partial Differential Equations
+
+- **ID:** `2602.12706` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Heechang Kim, Qianying Cao, Hyomin Shin, et al.
+- **abs:** https://arxiv.org/abs/2602.12706 · **PDF:** https://arxiv.org/pdf/2602.12706v1
+
+## 12584. MedXIAOHE: A Comprehensive Recipe for Building Medical MLLMs
+
+- **ID:** `2602.12705` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Baorong Shi, Bo Cui, Boyuan Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.12705 · **PDF:** https://arxiv.org/pdf/2602.12705v3
+
+## 12585. QTabGAN: A Hybrid Quantum-Classical GAN for Tabular Data Synthesis
+
+- **ID:** `2602.12704` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Subhangi Kumari, Rakesh Achutha, Vignesh Sivaraman
+- **abs:** https://arxiv.org/abs/2602.12704 · **PDF:** https://arxiv.org/pdf/2602.12704v2
+
+## 12586. SWING: Unlocking Implicit Graph Representations for Graph Random Features
+
+- **ID:** `2602.12703` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Alessandro Manenti, Avinava Dubey, Arijit Sehanobish, et al.
+- **abs:** https://arxiv.org/abs/2602.12703 · **PDF:** https://arxiv.org/pdf/2602.12703v1
+
+## 12587. Structured vs. Unstructured Pruning: An Exponential Gap
+
+- **ID:** `2603.02234` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Davide Ferre', Frédéric Giroire, Frederik Mallmann-Trenn, et al.
+- **abs:** https://arxiv.org/abs/2603.02234 · **PDF:** https://arxiv.org/pdf/2603.02234v2
+
+## 12588. AdaCorrection: Adaptive Offset Cache Correction for Accurate Diffusion Transformers
+
+- **ID:** `2602.13357` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Dong Liu, Yanxuan Yu, Ben Lengerich, et al.
+- **abs:** https://arxiv.org/abs/2602.13357 · **PDF:** https://arxiv.org/pdf/2602.13357v1
+
+## 12589. Channel-Aware Probing for Multi-Channel Imaging
+
+- **ID:** `2602.12696` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Umar Marikkar, Syed Sameed Husain, Muhammad Awais, et al.
+- **abs:** https://arxiv.org/abs/2602.12696 · **PDF:** https://arxiv.org/pdf/2602.12696v1
+
+## 12590. Leverage-Weighted Conformal Prediction
+
+- **ID:** `2602.12693` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Shreyas Fadnavis
+- **abs:** https://arxiv.org/abs/2602.12693 · **PDF:** https://arxiv.org/pdf/2602.12693v1
+
+## 12591. Adaptive Personalized Federated Learning via Multi-task Averaging of Kernel Mean Embeddings
+
+- **ID:** `2603.02233` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Jean-Baptiste Fermanian, Batiste Le Bars, Aurélien Bellet
+- **abs:** https://arxiv.org/abs/2603.02233 · **PDF:** https://arxiv.org/pdf/2603.02233v1
+
+## 12592. ALOE: Action-Level Off-Policy Evaluation for Vision-Language-Action Model Post-Training
+
+- **ID:** `2602.12691` · **primary:** `cs.RO` · **Submitted:** 2026-02-13
+- **Authors:** Rushuai Yang, Hecheng Wang, Chiming Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.12691 · **PDF:** https://arxiv.org/pdf/2602.12691v2
+
+## 12593. Trust the uncertain teacher: distilling dark knowledge via calibrated uncertainty
+
+- **ID:** `2602.12687` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Jeonghyun Kim, SooKyung Kim, Richeng Xuan, et al.
+- **abs:** https://arxiv.org/abs/2602.12687 · **PDF:** https://arxiv.org/pdf/2602.12687v1
+
+## 12594. Xiaomi-Robotics-0: An Open-Sourced Vision-Language-Action Model with Real-Time Execution
+
+- **ID:** `2602.12684` · **primary:** `cs.RO` · **Submitted:** 2026-02-13
+- **Authors:** Rui Cai, Jun Guo, Xinze He, et al.
+- **abs:** https://arxiv.org/abs/2602.12684 · **PDF:** https://arxiv.org/pdf/2602.12684v2
+
+## 12595. Flow Matching from Viewpoint of Proximal Operators
+
+- **ID:** `2602.12683` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Kenji Fukumizu, Wei Huang, Han Bao, et al.
+- **abs:** https://arxiv.org/abs/2602.12683 · **PDF:** https://arxiv.org/pdf/2602.12683v2
+
+## 12596. Fool Me If You Can: On the Robustness of Binary Code Similarity Detection Models against Semantics-preserving Transformations
+
+- **ID:** `2602.12681` · **primary:** `cs.CR` · **Submitted:** 2026-02-13
+- **Authors:** Jiyong Uhm, Minseok Kim, Michalis Polychronakis, et al.
+- **abs:** https://arxiv.org/abs/2602.12681 · **PDF:** https://arxiv.org/pdf/2602.12681v1
+
+## 12597. A Regularization-Sharpness Tradeoff for Linear Interpolators
+
+- **ID:** `2602.12680` · **primary:** `stat.ML` · **Submitted:** 2026-02-13
+- **Authors:** Qingyi Hu, Liam Hodgkinson
+- **abs:** https://arxiv.org/abs/2602.12680 · **PDF:** https://arxiv.org/pdf/2602.12680v1
+
+## 12598. SLA2: Sparse-Linear Attention with Learnable Routing and QAT
+
+- **ID:** `2602.12675` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Jintao Zhang, Haoxu Wang, Kai Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.12675 · **PDF:** https://arxiv.org/pdf/2602.12675v1
+
+## 12599. $\mathcal{X}$-KD: General Experiential Knowledge Distillation for Large Language Models
+
+- **ID:** `2602.12674` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Yuang Cai, Yuyu Yuan
+- **abs:** https://arxiv.org/abs/2602.12674 · **PDF:** https://arxiv.org/pdf/2602.12674v1
+
+## 12600. SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks
+
+- **ID:** `2602.12670` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Xiangyi Li, Wenbo Chen, Yimin Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.12670 · **PDF:** https://arxiv.org/pdf/2602.12670v3
+
+## 12601. Evaluating Robustness of Reasoning Models on Parameterized Logical Problems
+
+- **ID:** `2602.12665` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Naïm Es-sebbani, Esteban Marquer, Yakoub Salhi, et al.
+- **abs:** https://arxiv.org/abs/2602.12665 · **PDF:** https://arxiv.org/pdf/2602.12665v1
+
+## 12602. Doc-to-LoRA: Learning to Instantly Internalize Contexts
+
+- **ID:** `2602.15902` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Rujikorn Charakorn, Edoardo Cetin, Shinnosuke Uesaka, et al.
+- **abs:** https://arxiv.org/abs/2602.15902 · **PDF:** https://arxiv.org/pdf/2602.15902v1
+
+## 12603. Think Fast and Slow: Step-Level Cognitive Depth Adaptation for LLM Agents
+
+- **ID:** `2602.12662` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Ruihan Yang, Fanghua Ye, Xiang We, et al.
+- **abs:** https://arxiv.org/abs/2602.12662 · **PDF:** https://arxiv.org/pdf/2602.12662v1
+
+## 12604. Learning Ordinal Probabilistic Reward from Preferences
+
+- **ID:** `2602.12660` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Longze Chen, Lu Wang, Renke Shan, et al.
+- **abs:** https://arxiv.org/abs/2602.12660 · **PDF:** https://arxiv.org/pdf/2602.12660v2
+
+## 12605. IndicFairFace: Balanced Indian Face Dataset for Auditing and Mitigating Geographical Bias in Vision-Language Models
+
+- **ID:** `2602.12659` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Aarish Shah Mohsin, Mohammed Tayyab Ilyas Khan, Mohammad Nadeem, et al.
+- **abs:** https://arxiv.org/abs/2602.12659 · **PDF:** https://arxiv.org/pdf/2602.12659v1
+
+## 12606. PMG: Parameterized Motion Generator for Human-like Locomotion Control
+
+- **ID:** `2602.12656` · **primary:** `cs.RO` · **Submitted:** 2026-02-13
+- **Authors:** Chenxi Han, Yuheng Min, Zihao Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.12656 · **PDF:** https://arxiv.org/pdf/2602.12656v2
+
+## 12607. Uncovering spatial tissue domains and cell types in spatial omics through cross-scale profiling of cellular and genomic interactions
+
+- **ID:** `2602.12651` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Rui Yan, Xiaohan Xing, Xun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.12651 · **PDF:** https://arxiv.org/pdf/2602.12651v1
+
+## 12608. Multi-Task Learning with Additive U-Net for Image Denoising and Classification
+
+- **ID:** `2602.12649` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Vikram Lakkavalli, Neelam Sinha
+- **abs:** https://arxiv.org/abs/2602.12649 · **PDF:** https://arxiv.org/pdf/2602.12649v1
+
+## 12609. Unifying Model-Free Efficiency and Model-Based Representations via Latent Dynamics
+
+- **ID:** `2602.12643` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Jashaswimalya Acharjee, Balaraman Ravindran
+- **abs:** https://arxiv.org/abs/2602.12643 · **PDF:** https://arxiv.org/pdf/2602.12643v1
+
+## 12610. Beyond Normalization: Rethinking the Partition Function as a Difficulty Scheduler for RLVR
+
+- **ID:** `2602.12642` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Dohyung Kim, Minbeom Kim, Jeonghye Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.12642 · **PDF:** https://arxiv.org/pdf/2602.12642v1
+
+## 12611. Artic: AI-oriented Real-time Communication for MLLM Video Assistant
+
+- **ID:** `2602.12641` · **primary:** `cs.NI` · **Submitted:** 2026-02-13
+- **Authors:** Jiangkai Wu, Zhiyuan Ren, Junquan Zhong, et al.
+- **abs:** https://arxiv.org/abs/2602.12641 · **PDF:** https://arxiv.org/pdf/2602.12641v1
+
+## 12612. CLASE: A Hybrid Method for Chinese Legalese Stylistic Evaluation
+
+- **ID:** `2602.12639` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Yiran Rex Ma, Yuxiao Ye, Huiyuan Xie
+- **abs:** https://arxiv.org/abs/2602.12639 · **PDF:** https://arxiv.org/pdf/2602.12639v1
+
+## 12613. Dual-Granularity Contrastive Reward via Generated Episodic Guidance for Efficient Embodied RL
+
+- **ID:** `2602.12636` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Xin Liu, Yixuan Li, Yuhui Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.12636 · **PDF:** https://arxiv.org/pdf/2602.12636v1
+
+## 12614. Unleashing Low-Bit Inference on Ascend NPUs: A Comprehensive Evaluation of HiFloat Formats
+
+- **ID:** `2602.12635` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Pengxiang Zhao, Hui-Ling Zhen, Xing Li, et al.
+- **abs:** https://arxiv.org/abs/2602.12635 · **PDF:** https://arxiv.org/pdf/2602.12635v3
+
+## 12615. AI Agents for Inventory Control: Human-LLM-OR Complementarity
+
+- **ID:** `2602.12631` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Jackie Baek, Yaopeng Fu, Will Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.12631 · **PDF:** https://arxiv.org/pdf/2602.12631v1
+
+## 12616. TensorCommitments: A Lightweight Verifiable Inference for Language Models
+
+- **ID:** `2602.12630` · **primary:** `cs.CR` · **Submitted:** 2026-02-13
+- **Authors:** Oguzhan Baser, Elahe Sadeghi, Eric Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.12630 · **PDF:** https://arxiv.org/pdf/2602.12630v1
+
+## 12617. Beyond Binary Preferences: A Principled Framework for Reward Modeling with Ordinal Feedback
+
+- **ID:** `2603.02232` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Amirhossein Afsharrad, Ruida Zhou, Luca Viano, et al.
+- **abs:** https://arxiv.org/abs/2603.02232 · **PDF:** https://arxiv.org/pdf/2603.02232v1
+
+## 12618. Using Deep Learning to Generate Semantically Correct Hindi Captions
+
+- **ID:** `2602.13352` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Wasim Akram Khan, Anil Kumar Vuppala
+- **abs:** https://arxiv.org/abs/2602.13352 · **PDF:** https://arxiv.org/pdf/2602.13352v1
+
+## 12619. Formalizing the Sampling Design Space of Diffusion-Based Generative Models via Adaptive Solvers and Wasserstein-Bounded Timesteps
+
+- **ID:** `2602.12624` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Sangwoo Jo, Sungjoon Choi
+- **abs:** https://arxiv.org/abs/2602.12624 · **PDF:** https://arxiv.org/pdf/2602.12624v1
+
+## 12620. Efficient Personalized Federated PCA with Manifold Optimization for IoT Anomaly Detection
+
+- **ID:** `2602.12622` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Xianchao Xiu, Chenyi Huang, Wei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.12622 · **PDF:** https://arxiv.org/pdf/2602.12622v1
+
+## 12621. Vision Token Reduction via Attention-Driven Self-Compression for Efficient Multimodal Large Language Models
+
+- **ID:** `2602.12618` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Omer Faruk Deniz, Ruiyu Mao, Ruochen Li, et al.
+- **abs:** https://arxiv.org/abs/2602.12618 · **PDF:** https://arxiv.org/pdf/2602.12618v1
+
+## 12622. GeoAgent: Learning to Geolocate Everywhere with Reinforced Geographic Characteristics
+
+- **ID:** `2602.12617` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Modi Jin, Yiming Zhang, Boyuan Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.12617 · **PDF:** https://arxiv.org/pdf/2602.12617v1
+
+## 12623. Coden: Efficient Temporal Graph Neural Networks for Continuous Prediction
+
+- **ID:** `2602.12613` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Zulun Zhu, Siqiang Luo
+- **abs:** https://arxiv.org/abs/2602.12613 · **PDF:** https://arxiv.org/pdf/2602.12613v1
+
+## 12624. Data-Driven Deep MIMO Detection:Network Architectures and Generalization Analysis
+
+- **ID:** `2602.20178` · **primary:** `eess.SP` · **Submitted:** 2026-02-13
+- **Authors:** Yongwei Yi, Xinping Yi, Wenjin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.20178 · **PDF:** https://arxiv.org/pdf/2602.20178v1
+
+## 12625. Self-EvolveRec: Self-Evolving Recommender Systems with LLM-based Directional Feedback
+
+- **ID:** `2602.12612` · **primary:** `cs.IR` · **Submitted:** 2026-02-13
+- **Authors:** Sein Kim, Sangwu Park, Hongseok Kang, et al.
+- **abs:** https://arxiv.org/abs/2602.12612 · **PDF:** https://arxiv.org/pdf/2602.12612v1
+
+## 12626. UBio-MolFM: A Universal Molecular Foundation Model for Bio-Systems
+
+- **ID:** `2602.17709` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-13
+- **Authors:** Lin Huang, Arthur Jiang, XiaoLi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.17709 · **PDF:** https://arxiv.org/pdf/2602.17709v1
+
+## 12627. Physics-Informed Neural Networks with Architectural Physics Embedding for Large-Scale Wave Field Reconstruction
+
+- **ID:** `2603.02231` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Huiwen Zhang, Feng Ye, Chu Ma
+- **abs:** https://arxiv.org/abs/2603.02231 · **PDF:** https://arxiv.org/pdf/2603.02231v1
+
+## 12628. QuEPT: Quantized Elastic Precision Transformers with One-Shot Calibration for Multi-Bit Switching
+
+- **ID:** `2602.12609` · **primary:** `cs.CV` · **Submitted:** 2026-02-13
+- **Authors:** Ke Xu, Yixin Wang, Zhongcheng Li, et al.
+- **abs:** https://arxiv.org/abs/2602.12609 · **PDF:** https://arxiv.org/pdf/2602.12609v1
+
+## 12629. RelBench v2: A Large-Scale Benchmark and Repository for Relational Data
+
+- **ID:** `2602.12606` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Justin Gu, Rishabh Ranjan, Charilaos Kanatsoulis, et al.
+- **abs:** https://arxiv.org/abs/2602.12606 · **PDF:** https://arxiv.org/pdf/2602.12606v1
+
+## 12630. Block-Sample MAC-Bayes Generalization Bounds
+
+- **ID:** `2602.12605` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Matthias Frey, Jingge Zhu, Michael C. Gastpar
+- **abs:** https://arxiv.org/abs/2602.12605 · **PDF:** https://arxiv.org/pdf/2602.12605v1
+
+## 12631. Differentially Private Two-Stage Empirical Risk Minimization and Applications to Individualized Treatment Rule
+
+- **ID:** `2602.12604` · **primary:** `math.ST` · **Submitted:** 2026-02-13
+- **Authors:** Joowon Lee, Guanhua Chen
+- **abs:** https://arxiv.org/abs/2602.12604 · **PDF:** https://arxiv.org/pdf/2602.12604v1
+
+## 12632. HyperMLP: An Integrated Perspective for Sequence Modeling
+
+- **ID:** `2602.12601` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Jiecheng Lu, Shihao Yang
+- **abs:** https://arxiv.org/abs/2602.12601 · **PDF:** https://arxiv.org/pdf/2602.12601v1
+
+## 12633. Generalized Discrete Diffusion with Self-Correction
+
+- **ID:** `2603.02230` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Linxuan Wang, Ziyi Wang, Yikun Bai, et al.
+- **abs:** https://arxiv.org/abs/2603.02230 · **PDF:** https://arxiv.org/pdf/2603.02230v1
+
+## 12634. Causally Grounded Mechanistic Interpretability for LLMs with Faithful Natural-Language Explanations
+
+- **ID:** `2603.09988` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Ajay Pravin Mahale
+- **abs:** https://arxiv.org/abs/2603.09988 · **PDF:** https://arxiv.org/pdf/2603.09988v1
+
+## 12635. RQ-GMM: Residual Quantized Gaussian Mixture Model for Multimodal Semantic Discretization in CTR Prediction
+
+- **ID:** `2602.12593` · **primary:** `cs.IR` · **Submitted:** 2026-02-13
+- **Authors:** Ziye Tong, Jiahao Liu, Weimin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.12593 · **PDF:** https://arxiv.org/pdf/2602.12593v1
+
+## 12636. Power Interpretable Causal ODE Networks: A Unified Model for Explainable Anomaly Detection and Root Cause Analysis in Power Systems
+
+- **ID:** `2602.12592` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Yue Sun, Likai Wang, Rick S. Blum, et al.
+- **abs:** https://arxiv.org/abs/2602.12592 · **PDF:** https://arxiv.org/pdf/2602.12592v1
+
+## 12637. Vehicle behaviour estimation for abnormal event detection using distributed fiber optic sensing
+
+- **ID:** `2602.12591` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Hemant Prasad, Daisuke Ikefuji, Shin Tominaga, et al.
+- **abs:** https://arxiv.org/abs/2602.12591 · **PDF:** https://arxiv.org/pdf/2602.12591v1
+
+## 12638. Multi-Head Attention as a Source of Catastrophic Forgetting in MoE Transformers
+
+- **ID:** `2602.12587` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Anrui Chen, Ruijun Huang, Xin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.12587 · **PDF:** https://arxiv.org/pdf/2602.12587v1
+
+## 12639. Can I Have Your Order? Monte-Carlo Tree Search for Slot Filling Ordering in Diffusion Language Models
+
+- **ID:** `2602.12586` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Joshua Ong Jun Leang, Yu Zhao, Mihaela Cătălina Stoian, et al.
+- **abs:** https://arxiv.org/abs/2602.12586 · **PDF:** https://arxiv.org/pdf/2602.12586v1
+
+## 12640. VI-CuRL: Stabilizing Verifier-Independent RL Reasoning via Confidence-Guided Variance Reduction
+
+- **ID:** `2602.12579` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Xin-Qiang Cai, Masashi Sugiyama
+- **abs:** https://arxiv.org/abs/2602.12579 · **PDF:** https://arxiv.org/pdf/2602.12579v1
+
+## 12641. Discovering Semantic Latent Structures in Psychological Scales: A Response-Free Pathway to Efficient Simplification
+
+- **ID:** `2602.12575` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Bo Wang, Yuxuan Zhang, Yueqin Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.12575 · **PDF:** https://arxiv.org/pdf/2602.12575v2
+
+## 12642. Monte Carlo Tree Search with Reasoning Path Refinement for Small Language Models in Conversational Text-to-NoSQL
+
+- **ID:** `2602.12574` · **primary:** `cs.DB` · **Submitted:** 2026-02-13
+- **Authors:** Xubang Xiong, Raymond Chi-Wing Wong, Yuanfeng Song
+- **abs:** https://arxiv.org/abs/2602.12574 · **PDF:** https://arxiv.org/pdf/2602.12574v1
+
+## 12643. Enhancing Heat Sink Efficiency in MOSFETs using Physics Informed Neural Networks: A Systematic Study on Coolant Velocity Estimation
+
+- **ID:** `2602.20177` · **primary:** `cs.NE` · **Submitted:** 2026-02-13
+- **Authors:** Aniruddha Bora, Isabel K. Alvarez, Julie Chalfant, et al.
+- **abs:** https://arxiv.org/abs/2602.20177 · **PDF:** https://arxiv.org/pdf/2602.20177v1
+
+## 12644. Fractional Order Federated Learning for Battery Electric Vehicle Energy Consumption Modeling
+
+- **ID:** `2602.12567` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Mohammad Partohaghighi, Roummel Marcia, Bruce J. West, et al.
+- **abs:** https://arxiv.org/abs/2602.12567 · **PDF:** https://arxiv.org/pdf/2602.12567v1
+
+## 12645. To Mix or To Merge: Toward Multi-Domain Reinforcement Learning for Large Language Models
+
+- **ID:** `2602.12566` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Haoqing Wang, Xiang Long, Ziheng Li, et al.
+- **abs:** https://arxiv.org/abs/2602.12566 · **PDF:** https://arxiv.org/pdf/2602.12566v3
+
+## 12646. Alignment Is Not Enough: A Relational Framework for Moral Standing in Human-AI Interaction
+
+- **ID:** `2603.00078` · **primary:** `cs.CY` · **Submitted:** 2026-02-13
+- **Authors:** Faezeh B. Pasandi, Hannah B. Pasandi
+- **abs:** https://arxiv.org/abs/2603.00078 · **PDF:** https://arxiv.org/pdf/2603.00078v1
+
+## 12647. SD-MoE: Spectral Decomposition for Effective Expert Specialization
+
+- **ID:** `2602.12556` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Ruijun Huang, Fang Dong, Xin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.12556 · **PDF:** https://arxiv.org/pdf/2602.12556v1
+
+## 12648. Safety Training Persists Through Helpfulness Optimization in LLM Agents
+
+- **ID:** `2603.02229` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Benjamin Plaut
+- **abs:** https://arxiv.org/abs/2603.02229 · **PDF:** https://arxiv.org/pdf/2603.02229v1
+
+## 12649. A consequence of failed sequential learning: A computational account of developmental amnesia
+
+- **ID:** `2602.12547` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-13
+- **Authors:** Qi Zhang
+- **abs:** https://arxiv.org/abs/2602.12547 · **PDF:** https://arxiv.org/pdf/2602.12547v1
+
+## 12650. Decoder-only Conformer with Modality-aware Sparse Mixtures of Experts for ASR
+
+- **ID:** `2602.12546` · **primary:** `eess.AS` · **Submitted:** 2026-02-13
+- **Authors:** Jaeyoung Lee, Masato Mimura
+- **abs:** https://arxiv.org/abs/2602.12546 · **PDF:** https://arxiv.org/pdf/2602.12546v1
+
+## 12651. Scaling Web Agent Training through Automatic Data Generation and Fine-grained Evaluation
+
+- **ID:** `2602.12544` · **primary:** `cs.AI` · **Submitted:** 2026-02-13
+- **Authors:** Lajanugen Logeswaran, Jaekyeom Kim, Sungryull Sohn, et al.
+- **abs:** https://arxiv.org/abs/2602.12544 · **PDF:** https://arxiv.org/pdf/2602.12544v1
+
+## 12652. Exploring Accurate and Transparent Domain Adaptation in Predictive Healthcare via Concept-Grounded Orthogonal Inference
+
+- **ID:** `2602.12542` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Pengfei Hu, Chang Lu, Feifan Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.12542 · **PDF:** https://arxiv.org/pdf/2602.12542v1
+
+## 12653. Cross-Chirality Generalization by Axial Vectors for Hetero-Chiral Protein-Peptide Interaction Design
+
+- **ID:** `2602.20176` · **primary:** `q-bio.BM` · **Submitted:** 2026-02-13
+- **Authors:** Ziyi Yang, Zitong Tian, Yinjun Jia, et al.
+- **abs:** https://arxiv.org/abs/2602.20176 · **PDF:** https://arxiv.org/pdf/2602.20176v1
+
+## 12654. Linear Regression with Unknown Truncation Beyond Gaussian Features
+
+- **ID:** `2602.12534` · **primary:** `stat.ML` · **Submitted:** 2026-02-13
+- **Authors:** Alexandros Kouridakis, Anay Mehrotra, Alkis Kalavasis, et al.
+- **abs:** https://arxiv.org/abs/2602.12534 · **PDF:** https://arxiv.org/pdf/2602.12534v1
+
+## 12655. AMPS: Adaptive Modality Preference Steering via Functional Entropy
+
+- **ID:** `2602.12533` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Zihan Huang, Xintong Li, Rohan Surana, et al.
+- **abs:** https://arxiv.org/abs/2602.12533 · **PDF:** https://arxiv.org/pdf/2602.12533v1
+
+## 12656. Autorubric: A Unified Framework for Rubric-Based LLM Evaluation
+
+- **ID:** `2603.00077` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Delip Rao, Chris Callison-Burch
+- **abs:** https://arxiv.org/abs/2603.00077 · **PDF:** https://arxiv.org/pdf/2603.00077v1
+
+## 12657. Flow-Factory: A Unified Framework for Reinforcement Learning in Flow-Matching Models
+
+- **ID:** `2602.12529` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Bowen Ping, Chengyou Jia, Minnan Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.12529 · **PDF:** https://arxiv.org/pdf/2602.12529v3
+
+## 12658. DiffuRank: Effective Document Reranking with Diffusion Language Models
+
+- **ID:** `2602.12528` · **primary:** `cs.IR` · **Submitted:** 2026-02-13
+- **Authors:** Qi Liu, Kun Ai, Jiaxin Mao, et al.
+- **abs:** https://arxiv.org/abs/2602.12528 · **PDF:** https://arxiv.org/pdf/2602.12528v1
+
+## 12659. Analytical Results for Two Exponential Family Distributions in Hierarchical Dirichlet Processes
+
+- **ID:** `2602.12527` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Naiqi Li
+- **abs:** https://arxiv.org/abs/2602.12527 · **PDF:** https://arxiv.org/pdf/2602.12527v1
+
+## 12660. Constraint-Rectified Training for Efficient Chain-of-Thought
+
+- **ID:** `2602.12526` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Qinhang Wu, Sen Lin, Ming Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.12526 · **PDF:** https://arxiv.org/pdf/2602.12526v1
+
+## 12661. Multi-Agent Model-Based Reinforcement Learning with Joint State-Action Learned Embeddings
+
+- **ID:** `2602.12520` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Zhizun Wang, David Meger
+- **abs:** https://arxiv.org/abs/2602.12520 · **PDF:** https://arxiv.org/pdf/2602.12520v1
+
+## 12662. Bench-MFG: A Benchmark Suite for Learning in Stationary Mean Field Games
+
+- **ID:** `2602.12517` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Lorenzo Magnino, Jiacheng Shen, Matthieu Geist, et al.
+- **abs:** https://arxiv.org/abs/2602.12517 · **PDF:** https://arxiv.org/pdf/2602.12517v1
+
+## 12663. Visual RAG Toolkit: Scaling Multi-Vector Visual Retrieval with Training-Free Pooling and Multi-Stage Search
+
+- **ID:** `2602.12510` · **primary:** `cs.IR` · **Submitted:** 2026-02-13
+- **Authors:** Ara Yeroyan
+- **abs:** https://arxiv.org/abs/2602.12510 · **PDF:** https://arxiv.org/pdf/2602.12510v1
+
+## 12664. Monocular Reconstruction of Neural Tactile Fields
+
+- **ID:** `2602.12508` · **primary:** `cs.RO` · **Submitted:** 2026-02-13
+- **Authors:** Pavan Mantripragada, Siddhanth Deshmukh, Eadom Dessalene, et al.
+- **abs:** https://arxiv.org/abs/2602.12508 · **PDF:** https://arxiv.org/pdf/2602.12508v1
+
+## 12665. Speech to Speech Synthesis for Voice Impersonation
+
+- **ID:** `2602.16721` · **primary:** `cs.SD` · **Submitted:** 2026-02-13
+- **Authors:** Bjorn Johnson, Jared Levy
+- **abs:** https://arxiv.org/abs/2602.16721 · **PDF:** https://arxiv.org/pdf/2602.16721v1
+
+## 12666. Evolving Demonstration Optimization for Chain-of-Thought Feature Transformation
+
+- **ID:** `2603.09987` · **primary:** `cs.CL` · **Submitted:** 2026-02-13
+- **Authors:** Xinyuan Wang, Kunpeng Liu, Arun Vignesh Malarkkan, et al.
+- **abs:** https://arxiv.org/abs/2603.09987 · **PDF:** https://arxiv.org/pdf/2603.09987v1
+
+## 12667. On Robustness and Chain-of-Thought Consistency of RL-Finetuned VLMs
+
+- **ID:** `2602.12506` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Rosie Zhao, Anshul Shah, Xiaoyu Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.12506 · **PDF:** https://arxiv.org/pdf/2602.12506v2
+
+## 12668. Favia: Forensic Agent for Vulnerability-fix Identification and Analysis
+
+- **ID:** `2602.12500` · **primary:** `cs.SE` · **Submitted:** 2026-02-13
+- **Authors:** André Storhaug, Jiamou Sun, Jingyue Li
+- **abs:** https://arxiv.org/abs/2602.12500 · **PDF:** https://arxiv.org/pdf/2602.12500v1
+
+## 12669. A Theoretical Analysis of Mamba's Training Dynamics: Filtering Relevant Features for Generalization in State Space Models
+
+- **ID:** `2602.12499` · **primary:** `cs.LG` · **Submitted:** 2026-02-13
+- **Authors:** Mugunthan Shandirasegaran, Hongkang Li, Songyang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.12499 · **PDF:** https://arxiv.org/pdf/2602.12499v1
+
+## 12670. Composable Model-Free RL for Navigation with Input-Affine Systems
+
+- **ID:** `2602.12492` · **primary:** `cs.RO` · **Submitted:** 2026-02-13
+- **Authors:** Xinhuan Sang, Abdelrahman Abdelgawad, Roberto Tron
+- **abs:** https://arxiv.org/abs/2602.12492 · **PDF:** https://arxiv.org/pdf/2602.12492v1
+
+## 12671. Transformer-based CoVaR: Systemic Risk in Textual Information
+
+- **ID:** `2602.12490` · **primary:** `econ.EM` · **Submitted:** 2026-02-13
+- **Authors:** Junyu Chen, Tom Boot, Lingwei Kong, et al.
+- **abs:** https://arxiv.org/abs/2602.12490 · **PDF:** https://arxiv.org/pdf/2602.12490v1
+
+## 12672. Gradient-Enhanced Partitioned Gaussian Processes for Real-Time Quadrotor Dynamics Modeling
+
+- **ID:** `2602.12487` · **primary:** `cs.RO` · **Submitted:** 2026-02-13
+- **Authors:** Xinhuan Sang, Adam Rozman, Sheryl Grace, et al.
+- **abs:** https://arxiv.org/abs/2602.12487 · **PDF:** https://arxiv.org/pdf/2602.12487v1
+
+## 12673. Human-Like Coarse Object Representations in Vision Models
+
+- **ID:** `2602.12486` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Andrey Gizdov, Andrea Procopio, Yichen Li, et al.
+- **abs:** https://arxiv.org/abs/2602.12486 · **PDF:** https://arxiv.org/pdf/2602.12486v1
+
+## 12674. A Formal Framework for the Explanation of Finite Automata Decisions
+
+- **ID:** `2602.13351` · **primary:** `cs.FL` · **Submitted:** 2026-02-12
+- **Authors:** Jaime Cuartas Granada, Alexey Ignatiev, Peter J. Stuckey
+- **abs:** https://arxiv.org/abs/2602.13351 · **PDF:** https://arxiv.org/pdf/2602.13351v2
+
+## 12675. A Lightweight and Explainable DenseNet-121 Framework for Grape Leaf Disease Classification
+
+- **ID:** `2602.12484` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Md. Ehsanul Haque, Md. Saymon Hosen Polash, Rakib Hasan Ovi, et al.
+- **abs:** https://arxiv.org/abs/2602.12484 · **PDF:** https://arxiv.org/pdf/2602.12484v1
+
+## 12676. Detecting Brick Kiln Infrastructure at Scale: Graph, Foundation, and Remote Sensing Models for Satellite Imagery Data
+
+- **ID:** `2602.13350` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Usman Nazir, Xidong Chen, Hafiz Muhammad Abubakar, et al.
+- **abs:** https://arxiv.org/abs/2602.13350 · **PDF:** https://arxiv.org/pdf/2602.13350v1
+
+## 12677. Geometric separation and constructive universal approximation with two hidden layers
+
+- **ID:** `2602.12482` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Chanyoung Sung
+- **abs:** https://arxiv.org/abs/2602.12482 · **PDF:** https://arxiv.org/pdf/2602.12482v1
+
+## 12678. Task- and Metric-Specific Signal Quality Indices for Medical Time Series
+
+- **ID:** `2602.12478` · **primary:** `eess.SP` · **Submitted:** 2026-02-12
+- **Authors:** Jad Haidamous, Christoph Hoog Antink
+- **abs:** https://arxiv.org/abs/2602.12478 · **PDF:** https://arxiv.org/pdf/2602.12478v1
+
+## 12679. Not a Silver Bullet for Loneliness: How Attachment and Age Shape Intimacy with AI Companions
+
+- **ID:** `2602.12476` · **primary:** `cs.CY` · **Submitted:** 2026-02-12
+- **Authors:** Raffaele Ciriello, Uri Gal, Ofir Turel
+- **abs:** https://arxiv.org/abs/2602.12476 · **PDF:** https://arxiv.org/pdf/2602.12476v1
+
+## 12680. Tight Bounds for Logistic Regression with Large Stepsize Gradient Descent in Low Dimension
+
+- **ID:** `2602.12471` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Michael Crawshaw, Mingrui Liu
+- **abs:** https://arxiv.org/abs/2602.12471 · **PDF:** https://arxiv.org/pdf/2602.12471v1
+
+## 12681. Designing RNAs with Language Models
+
+- **ID:** `2602.12470` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Milan Gautam, Ning Dai, Tianshuo Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.12470 · **PDF:** https://arxiv.org/pdf/2602.12470v1
+
+## 12682. Regularized Meta-Learning for Improved Generalization
+
+- **ID:** `2602.12469` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Noor Islam S. Mohammad, Md Muntaqim Meherab
+- **abs:** https://arxiv.org/abs/2602.12469 · **PDF:** https://arxiv.org/pdf/2602.12469v1
+
+## 12683. Continuous Diffusion Models Can Obey Formal Syntax
+
+- **ID:** `2602.12468` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Jinwoo Kim, Taylor Berg-Kirkpatrick, Loris D'Antoni
+- **abs:** https://arxiv.org/abs/2602.12468 · **PDF:** https://arxiv.org/pdf/2602.12468v1
+
+## 12684. Probabilistic Design of Parametrized Quantum Circuits through Local Gate Modifications
+
+- **ID:** `2602.12465` · **primary:** `quant-ph` · **Submitted:** 2026-02-12
+- **Authors:** Grier M. Jones, Aviraj Newatia, Alexander Lao, et al.
+- **abs:** https://arxiv.org/abs/2602.12465 · **PDF:** https://arxiv.org/pdf/2602.12465v1
+
+## 12685. Correctness, Artificial Intelligence, and the Epistemic Value of Mathematical Proof
+
+- **ID:** `2602.12463` · **primary:** `math.HO` · **Submitted:** 2026-02-12
+- **Authors:** James Owen Weatherall, Jesse Wolfson
+- **abs:** https://arxiv.org/abs/2602.12463 · **PDF:** https://arxiv.org/pdf/2602.12463v1
+
+## 12686. Spectral Homogenization of the Radiative Transfer Equation via Low-Rank Tensor Train Decomposition
+
+- **ID:** `2602.17708` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-12
+- **Authors:** Y. Sungtaek Ju
+- **abs:** https://arxiv.org/abs/2602.17708 · **PDF:** https://arxiv.org/pdf/2602.17708v1
+
+## 12687. Computationally sufficient statistics for Ising models
+
+- **ID:** `2602.12449` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Abhijith Jayakumar, Shreya Shukla, Marc Vuffray, et al.
+- **abs:** https://arxiv.org/abs/2602.12449 · **PDF:** https://arxiv.org/pdf/2602.12449v1
+
+## 12688. Your Robot Will Feel You Now: Empathy in Robots and Embodied Agents
+
+- **ID:** `2603.20200` · **primary:** `cs.RO` · **Submitted:** 2026-02-12
+- **Authors:** Angelica Lim, Ö. Nilay Yalçin
+- **abs:** https://arxiv.org/abs/2603.20200 · **PDF:** https://arxiv.org/pdf/2603.20200v1
+
+## 12689. RBCorr: Response Bias Correction in Language Models
+
+- **ID:** `2602.12445` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Om Bhatt, Anna A. Ivanova
+- **abs:** https://arxiv.org/abs/2602.12445 · **PDF:** https://arxiv.org/pdf/2602.12445v1
+
+## 12690. Safe Reinforcement Learning via Recovery-based Shielding with Gaussian Process Dynamics Models
+
+- **ID:** `2602.12444` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Alexander W. Goodall, Francesco Belardinelli
+- **abs:** https://arxiv.org/abs/2602.12444 · **PDF:** https://arxiv.org/pdf/2602.12444v2
+
+## 12691. Reconstructing Carbon Monoxide Reanalysis with Machine Learning
+
+- **ID:** `2602.15056` · **primary:** `physics.ao-ph` · **Submitted:** 2026-02-12
+- **Authors:** Paula Harder, Johannes Flemming
+- **abs:** https://arxiv.org/abs/2602.15056 · **PDF:** https://arxiv.org/pdf/2602.15056v1
+
+## 12692. Neural and numerical methods for $\mathrm{G}_2$-structures on contact Calabi-Yau 7-manifolds
+
+- **ID:** `2602.12438` · **primary:** `math.DG` · **Submitted:** 2026-02-12
+- **Authors:** Elli Heyes, Edward Hirst, Henrique N. Sá Earp, et al.
+- **abs:** https://arxiv.org/abs/2602.12438 · **PDF:** https://arxiv.org/pdf/2602.12438v1
+
+## 12693. The Value Sensitivity Gap: How Clinical Large Language Models Respond to Patient Preference Statements in Shared Decision-Making
+
+- **ID:** `2603.00076` · **primary:** `cs.CY` · **Submitted:** 2026-02-12
+- **Authors:** Sanjay Basu
+- **abs:** https://arxiv.org/abs/2603.00076 · **PDF:** https://arxiv.org/pdf/2603.00076v1
+
+## 12694. From Prompt to Production:Automating Brand-Safe Marketing Imagery with Text-to-Image Models
+
+- **ID:** `2602.13349` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Parmida Atighehchian, Henry Wang, Andrei Kapustin, et al.
+- **abs:** https://arxiv.org/abs/2602.13349 · **PDF:** https://arxiv.org/pdf/2602.13349v1
+
+## 12695. Agent Skills for Large Language Models: Architecture, Acquisition, Security, and the Path Forward
+
+- **ID:** `2602.12430` · **primary:** `cs.MA` · **Submitted:** 2026-02-12
+- **Authors:** Renjun Xu, Yang Yan
+- **abs:** https://arxiv.org/abs/2602.12430 · **PDF:** https://arxiv.org/pdf/2602.12430v3
+
+## 12696. Stabilizing Native Low-Rank LLM Pretraining
+
+- **ID:** `2602.12429` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Paul Janson, Edouard Oyallon, Eugene Belilovsky
+- **abs:** https://arxiv.org/abs/2602.12429 · **PDF:** https://arxiv.org/pdf/2602.12429v1
+
+## 12697. Exploring the Performance of ML/DL Architectures on the MNIST-1D Dataset
+
+- **ID:** `2602.13348` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Michael Beebe, GodsGift Uzor, Manasa Chepuri, et al.
+- **abs:** https://arxiv.org/abs/2602.13348 · **PDF:** https://arxiv.org/pdf/2602.13348v1
+
+## 12698. Interference-Robust Non-Coherent Over-the-Air Computation for Decentralized Optimization
+
+- **ID:** `2602.12426` · **primary:** `eess.SP` · **Submitted:** 2026-02-12
+- **Authors:** Nicolò Michelusi
+- **abs:** https://arxiv.org/abs/2602.12426 · **PDF:** https://arxiv.org/pdf/2602.12426v1
+
+## 12699. RankLLM: Weighted Ranking of LLMs by Quantifying Question Difficulty
+
+- **ID:** `2602.12424` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Ziqian Zhang, Xingjian Hu, Yue Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.12424 · **PDF:** https://arxiv.org/pdf/2602.12424v1
+
+## 12700. CacheMind: From Miss Rates to Why -- Natural-Language, Trace-Grounded Reasoning for Cache Replacement
+
+- **ID:** `2602.12422` · **primary:** `cs.AR` · **Submitted:** 2026-02-12
+- **Authors:** Kaushal Mhapsekar, Azam Ghanbari, Bita Aslrousta, et al.
+- **abs:** https://arxiv.org/abs/2602.12422 · **PDF:** https://arxiv.org/pdf/2602.12422v1
+
+## 12701. Visual Foresight for Robotic Stow: A Diffusion-Based World Model from Sparse Snapshots
+
+- **ID:** `2602.13347` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Lijun Zhang, Nikhil Chacko, Petter Nilsson, et al.
+- **abs:** https://arxiv.org/abs/2602.13347 · **PDF:** https://arxiv.org/pdf/2602.13347v1
+
+## 12702. Tensor Network Generator-Enhanced Optimization for Traveling Salesman Problem
+
+- **ID:** `2602.20175` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Ryo Sakai, Chen-Yu Liu
+- **abs:** https://arxiv.org/abs/2602.20175 · **PDF:** https://arxiv.org/pdf/2602.20175v1
+
+## 12703. Intent-Driven Smart Manufacturing Integrating Knowledge Graphs and Large Language Models
+
+- **ID:** `2602.12419` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Takoua Jradi, John Violos, Dimitrios Spatharakis, et al.
+- **abs:** https://arxiv.org/abs/2602.12419 · **PDF:** https://arxiv.org/pdf/2602.12419v1
+
+## 12704. Sparse Autoencoders are Capable LLM Jailbreak Mitigators
+
+- **ID:** `2602.12418` · **primary:** `cs.CR` · **Submitted:** 2026-02-12
+- **Authors:** Yannick Assogba, Jacopo Cortellazzi, Javier Abad, et al.
+- **abs:** https://arxiv.org/abs/2602.12418 · **PDF:** https://arxiv.org/pdf/2602.12418v1
+
+## 12705. propella-1: Multi-Property Document Annotation for LLM Data Curation at Scale
+
+- **ID:** `2602.12414` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Maximilian Idahl, Benedikt Droste, Björn Plüster, et al.
+- **abs:** https://arxiv.org/abs/2602.12414 · **PDF:** https://arxiv.org/pdf/2602.12414v2
+
+## 12706. Soft Contamination Means Benchmarks Test Shallow Generalization
+
+- **ID:** `2602.12413` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Ari Spiesberger, Juan J. Vazquez, Nicky Pochinkov, et al.
+- **abs:** https://arxiv.org/abs/2602.12413 · **PDF:** https://arxiv.org/pdf/2602.12413v1
+
+## 12707. MiDAS: A Multimodal Data Acquisition System and Dataset for Robot-Assisted Minimally Invasive Surgery
+
+- **ID:** `2602.12407` · **primary:** `cs.RO` · **Submitted:** 2026-02-12
+- **Authors:** Keshara Weerasinghe, Seyed Hamid Reza Roodabeh, Andrew Hawkins, et al.
+- **abs:** https://arxiv.org/abs/2602.12407 · **PDF:** https://arxiv.org/pdf/2602.12407v2
+
+## 12708. Self-Refining Vision Language Model for Robotic Failure Detection and Reasoning
+
+- **ID:** `2602.12405` · **primary:** `cs.RO` · **Submitted:** 2026-02-12
+- **Authors:** Carl Qi, Xiaojie Wang, Silong Yong, et al.
+- **abs:** https://arxiv.org/abs/2602.12405 · **PDF:** https://arxiv.org/pdf/2602.12405v1
+
+## 12709. AstRL: Analog and Mixed-Signal Circuit Synthesis with Deep Reinforcement Learning
+
+- **ID:** `2602.12402` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Felicia B. Guo, Ken T. Ho, Andrei Vladimirescu, et al.
+- **abs:** https://arxiv.org/abs/2602.12402 · **PDF:** https://arxiv.org/pdf/2602.12402v1
+
+## 12710. What does RL improve for Visual Reasoning? A Frankenstein-Style Analysis
+
+- **ID:** `2602.12395` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Xirui Li, Ming Li, Tianyi Zhou
+- **abs:** https://arxiv.org/abs/2602.12395 · **PDF:** https://arxiv.org/pdf/2602.12395v1
+
+## 12711. Synthetic Interaction Data for Scalable Personalization in Large Language Models
+
+- **ID:** `2602.12394` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Yuchen Ma, Yue Huang, Wenjie Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.12394 · **PDF:** https://arxiv.org/pdf/2602.12394v1
+
+## 12712. Reproducing DragDiffusion: Interactive Point-Based Editing with Diffusion Models
+
+- **ID:** `2602.12393` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Ali Subhan, Ashir Raza
+- **abs:** https://arxiv.org/abs/2602.12393 · **PDF:** https://arxiv.org/pdf/2602.12393v1
+
+## 12713. High-dimensional Level Set Estimation with Trust Regions and Double Acquisition Functions
+
+- **ID:** `2602.12391` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Giang Ngo, Dat Phan Trong, Dang Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.12391 · **PDF:** https://arxiv.org/pdf/2602.12391v1
+
+## 12714. Rational Neural Networks have Expressivity Advantages
+
+- **ID:** `2602.12390` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Maosen Tang, Alex Townsend
+- **abs:** https://arxiv.org/abs/2602.12390 · **PDF:** https://arxiv.org/pdf/2602.12390v1
+
+## 12715. Evolving Beyond Snapshots: Harmonizing Structure and Sequence via Entity State Tuning for Temporal Knowledge Graph Forecasting
+
+- **ID:** `2602.12389` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Siyuan Li, Yunjia Wu, Yiyong Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.12389 · **PDF:** https://arxiv.org/pdf/2602.12389v2
+
+## 12716. Vibe Coding on Trial: Operating Characteristics of Unanimous LLM Juries
+
+- **ID:** `2602.18492` · **primary:** `cs.DB` · **Submitted:** 2026-02-12
+- **Authors:** Muhammad Aziz Ullah, Abdul Serwadda
+- **abs:** https://arxiv.org/abs/2602.18492 · **PDF:** https://arxiv.org/pdf/2602.18492v1
+
+## 12717. Accelerating Feedback-based Algorithms for Quantum Optimization Using Gradient Descent
+
+- **ID:** `2602.12387` · **primary:** `quant-ph` · **Submitted:** 2026-02-12
+- **Authors:** Masih Mozakka, Mohsen Heidari
+- **abs:** https://arxiv.org/abs/2602.12387 · **PDF:** https://arxiv.org/pdf/2602.12387v1
+
+## 12718. Provably Convergent Actor-Critic in Risk-averse MARL
+
+- **ID:** `2602.12386` · **primary:** `cs.MA` · **Submitted:** 2026-02-12
+- **Authors:** Yizhou Zhang, Eric Mazumdar
+- **abs:** https://arxiv.org/abs/2602.12386 · **PDF:** https://arxiv.org/pdf/2602.12386v1
+
+## 12719. Why Deep Jacobian Spectra Separate: Depth-Induced Scaling and Singular-Vector Alignment
+
+- **ID:** `2602.12384` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Nathanaël Haas, François Gatine, Augustin M Cosse, et al.
+- **abs:** https://arxiv.org/abs/2602.12384 · **PDF:** https://arxiv.org/pdf/2602.12384v2
+
+## 12720. TFT-ACB-XML: Decision-Level Integration of Customized Temporal Fusion Transformer and Attention-BiLSTM with XGBoost Meta-Learner for BTC Price Forecasting
+
+- **ID:** `2602.12380` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Raiz Ud Din, Saddam Hussain Khan
+- **abs:** https://arxiv.org/abs/2602.12380 · **PDF:** https://arxiv.org/pdf/2602.12380v1
+
+## 12721. CellMaster: Collaborative Cell Type Annotation in Single-Cell Analysis
+
+- **ID:** `2602.13346` · **primary:** `q-bio.GN` · **Submitted:** 2026-02-12
+- **Authors:** Zhen Wang, Yiming Gao, Jieyuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.13346 · **PDF:** https://arxiv.org/pdf/2602.13346v1
+
+## 12722. Deep Doubly Debiased Longitudinal Effect Estimation with ICE G-Computation
+
+- **ID:** `2602.12379` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Wenxin Chen, Weishen Pan, Kyra Gan, et al.
+- **abs:** https://arxiv.org/abs/2602.12379 · **PDF:** https://arxiv.org/pdf/2602.12379v1
+
+## 12723. Value Bonuses using Ensemble Errors for Exploration in Reinforcement Learning
+
+- **ID:** `2602.12375` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Abdul Wahab, Raksha Kumaraswamy, Martha White
+- **abs:** https://arxiv.org/abs/2602.12375 · **PDF:** https://arxiv.org/pdf/2602.12375v1
+
+## 12724. Policy4OOD: A Knowledge-Guided World Model for Policy Intervention Simulation against the Opioid Overdose Crisis
+
+- **ID:** `2602.12373` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Yijun Ma, Zehong Wang, Weixiang Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.12373 · **PDF:** https://arxiv.org/pdf/2602.12373v1
+
+## 12725. A Machine Learning Approach to the Nirenberg Problem
+
+- **ID:** `2602.12368` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Gianfranco Cortés, Maria Esteban-Casadevall, Yueqing Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.12368 · **PDF:** https://arxiv.org/pdf/2602.12368v1
+
+## 12726. BLUEPRINT Rebuilding a Legacy: Multimodal Retrieval for Complex Engineering Drawings and Documents
+
+- **ID:** `2602.13345` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Ethan Seefried, Ran Eldegaway, Sanjay Das, et al.
+- **abs:** https://arxiv.org/abs/2602.13345 · **PDF:** https://arxiv.org/pdf/2602.13345v1
+
+## 12727. A Theoretical Framework for Adaptive Utility-Weighted Benchmarking
+
+- **ID:** `2602.12356` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Philip Waggoner
+- **abs:** https://arxiv.org/abs/2602.12356 · **PDF:** https://arxiv.org/pdf/2602.12356v1
+
+## 12728. Variational Green's Functions for Volumetric PDEs
+
+- **ID:** `2602.12349` · **primary:** `cs.GR` · **Submitted:** 2026-02-12
+- **Authors:** Joao Teixeira, Eitan Grinspun, Otman Benchekroun
+- **abs:** https://arxiv.org/abs/2602.12349 · **PDF:** https://arxiv.org/pdf/2602.12349v1
+
+## 12729. Intrinsic Credit Assignment for Long Horizon Interaction
+
+- **ID:** `2602.12342` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Ilze Amanda Auzina, Joschka Strüber, Sergio Hernández-Gutiérrez, et al.
+- **abs:** https://arxiv.org/abs/2602.12342 · **PDF:** https://arxiv.org/pdf/2602.12342v1
+
+## 12730. Wireless TokenCom: RL-Based Tokenizer Agreement for Multi-User Wireless Token Communications
+
+- **ID:** `2602.12338` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Farshad Zeinali, Mahdi Boloursaz Mashhadi, Dusit Niyato, et al.
+- **abs:** https://arxiv.org/abs/2602.12338 · **PDF:** https://arxiv.org/pdf/2602.12338v1
+
+## 12731. Scaling Verification Can Be More Effective than Scaling Policy Learning for Vision-Language-Action Alignment
+
+- **ID:** `2602.12281` · **primary:** `cs.RO` · **Submitted:** 2026-02-12
+- **Authors:** Jacky Kwok, Xilun Zhang, Mengdi Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.12281 · **PDF:** https://arxiv.org/pdf/2602.12281v2
+
+## 12732. UniT: Unified Multimodal Chain-of-Thought Test-time Scaling
+
+- **ID:** `2602.12279` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Leon Liangyu Chen, Haoyu Ma, Zhipeng Fan, et al.
+- **abs:** https://arxiv.org/abs/2602.12279 · **PDF:** https://arxiv.org/pdf/2602.12279v1
+
+## 12733. AttentionRetriever: Attention Layers are Secretly Long Document Retrievers
+
+- **ID:** `2602.12278` · **primary:** `cs.IR` · **Submitted:** 2026-02-12
+- **Authors:** David Jiahao Fu, Lam Thanh Do, Jiayu Li, et al.
+- **abs:** https://arxiv.org/abs/2602.12278 · **PDF:** https://arxiv.org/pdf/2602.12278v1
+
+## 12734. The Appeal and Reality of Recycling LoRAs with Adaptive Merging
+
+- **ID:** `2602.12323` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Haokun Liu, Gyung Hyun Je, Marco Ciccone, et al.
+- **abs:** https://arxiv.org/abs/2602.12323 · **PDF:** https://arxiv.org/pdf/2602.12323v1
+
+## 12735. Agentic Test-Time Scaling for WebAgents
+
+- **ID:** `2602.12276` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Nicholas Lee, Lutfi Eren Erdogan, Chris Joseph John, et al.
+- **abs:** https://arxiv.org/abs/2602.12276 · **PDF:** https://arxiv.org/pdf/2602.12276v1
+
+## 12736. On-Policy Context Distillation for Language Models
+
+- **ID:** `2602.12275` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Tianzhu Ye, Li Dong, Xun Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.12275 · **PDF:** https://arxiv.org/pdf/2602.12275v2
+
+## 12737. Function-Space Decoupled Diffusion for Forward and Inverse Modeling in Carbon Capture and Storage
+
+- **ID:** `2602.12274` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Xin Ju, Jiachen Yao, Anima Anandkumar, et al.
+- **abs:** https://arxiv.org/abs/2602.12274 · **PDF:** https://arxiv.org/pdf/2602.12274v2
+
+## 12738. Learning to Control: The iUzawa-Net for Nonsmooth Optimal Control of Linear PDEs
+
+- **ID:** `2602.12273` · **primary:** `math.OC` · **Submitted:** 2026-02-12
+- **Authors:** Yongcun Song, Xiaoming Yuan, Hangrui Yue, et al.
+- **abs:** https://arxiv.org/abs/2602.12273 · **PDF:** https://arxiv.org/pdf/2602.12273v1
+
+## 12739. MonarchRT: Efficient Attention for Real-Time Video Generation
+
+- **ID:** `2602.12271` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Krish Agarwal, Zhuoming Chen, Cheng Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.12271 · **PDF:** https://arxiv.org/pdf/2602.12271v1
+
+## 12740. Creative Ownership in the Age of AI
+
+- **ID:** `2602.12270` · **primary:** `econ.TH` · **Submitted:** 2026-02-12
+- **Authors:** Annie Liang, Jay Lu
+- **abs:** https://arxiv.org/abs/2602.12270 · **PDF:** https://arxiv.org/pdf/2602.12270v1
+
+## 12741. ForeAct: Steering Your VLA with Efficient Visual Foresight Planning
+
+- **ID:** `2602.12322` · **primary:** `cs.RO` · **Submitted:** 2026-02-12
+- **Authors:** Zhuoyang Zhang, Shang Yang, Qinghao Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.12322 · **PDF:** https://arxiv.org/pdf/2602.12322v1
+
+## 12742. CM2: Reinforcement Learning with Checklist Rewards for Multi-Turn and Multi-Step Agentic Tool Use
+
+- **ID:** `2602.12268` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Zhen Zhang, Kaiqiang Song, Xun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.12268 · **PDF:** https://arxiv.org/pdf/2602.12268v2
+
+## 12743. Self-Supervised Learning via Flow-Guided Neural Operator on Time-Series Data
+
+- **ID:** `2602.12267` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Duy Nguyen, Jiachen Yao, Jiayun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.12267 · **PDF:** https://arxiv.org/pdf/2602.12267v2
+
+## 12744. T3D: Few-Step Diffusion Language Models via Trajectory Self-Distillation with Direct Discriminative Optimization
+
+- **ID:** `2602.12262` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Tunyu Zhang, Xinxi Zhang, Ligong Han, et al.
+- **abs:** https://arxiv.org/abs/2602.12262 · **PDF:** https://arxiv.org/pdf/2602.12262v2
+
+## 12745. Think like a Scientist: Physics-guided LLM Agent for Equation Discovery
+
+- **ID:** `2602.12259` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Jianke Yang, Ohm Venkatachalam, Mohammad Kianezhad, et al.
+- **abs:** https://arxiv.org/abs/2602.12259 · **PDF:** https://arxiv.org/pdf/2602.12259v2
+
+## 12746. On the implicit regularization of Langevin dynamics with projected noise
+
+- **ID:** `2602.12257` · **primary:** `math.PR` · **Submitted:** 2026-02-12
+- **Authors:** Govind Menon, Austin J. Stromme, Adrien Vacher
+- **abs:** https://arxiv.org/abs/2602.12257 · **PDF:** https://arxiv.org/pdf/2602.12257v1
+
+## 12747. Is Online Linear Optimization Sufficient for Strategic Robustness?
+
+- **ID:** `2602.12253` · **primary:** `cs.GT` · **Submitted:** 2026-02-12
+- **Authors:** Yang Cai, Haipeng Luo, Chen-Yu Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.12253 · **PDF:** https://arxiv.org/pdf/2602.12253v1
+
+## 12748. A technical curriculum on language-oriented artificial intelligence in translation and specialised communication
+
+- **ID:** `2602.12251` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Ralph Krüger
+- **abs:** https://arxiv.org/abs/2602.12251 · **PDF:** https://arxiv.org/pdf/2602.12251v1
+
+## 12749. Community Concealment from Unsupervised Graph Learning-Based Clustering
+
+- **ID:** `2602.12250` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Dalyapraz Manatova, Pablo Moriano, L. Jean Camp
+- **abs:** https://arxiv.org/abs/2602.12250 · **PDF:** https://arxiv.org/pdf/2602.12250v1
+
+## 12750. "Sorry, I Didn't Catch That": How Speech Models Miss What Matters Most
+
+- **ID:** `2602.12249` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Kaitlyn Zhou, Martijn Bartelds, Federico Bianchi, et al.
+- **abs:** https://arxiv.org/abs/2602.12249 · **PDF:** https://arxiv.org/pdf/2602.12249v2
+
+## 12751. ExtractBench: A Benchmark and Evaluation Methodology for Complex Structured Extraction
+
+- **ID:** `2602.12247` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Nick Ferguson, Josh Pennington, Narek Beghian, et al.
+- **abs:** https://arxiv.org/abs/2602.12247 · **PDF:** https://arxiv.org/pdf/2602.12247v2
+
+## 12752. Intrinsic-Energy Joint Embedding Predictive Architectures Induce Quasimetric Spaces
+
+- **ID:** `2602.12245` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Anthony Kobanda, Waris Radji
+- **abs:** https://arxiv.org/abs/2602.12245 · **PDF:** https://arxiv.org/pdf/2602.12245v1
+
+## 12753. Moonshine v2: Ergodic Streaming Encoder ASR for Latency-Critical Speech Applications
+
+- **ID:** `2602.12241` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Manjunath Kudlur, Evan King, James Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.12241 · **PDF:** https://arxiv.org/pdf/2602.12241v1
+
+## 12754. Olmix: A Framework for Data Mixing Throughout LM Development
+
+- **ID:** `2602.12237` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Mayee F. Chen, Tyler Murray, David Heineman, et al.
+- **abs:** https://arxiv.org/abs/2602.12237 · **PDF:** https://arxiv.org/pdf/2602.12237v1
+
+## 12755. Energy-Aware Spike Budgeting for Continual Learning in Spiking Neural Networks for Neuromorphic Vision
+
+- **ID:** `2602.12236` · **primary:** `cs.NE` · **Submitted:** 2026-02-12
+- **Authors:** Anika Tabassum Meem, Muntasir Hossain Nadid, Md Zesun Ahmed Mia
+- **abs:** https://arxiv.org/abs/2602.12236 · **PDF:** https://arxiv.org/pdf/2602.12236v2
+
+## 12756. Detecting Overflow in Compressed Token Representations for Retrieval-Augmented Generation
+
+- **ID:** `2602.12235` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Julia Belikova, Danila Rozhevskii, Dennis Svirin, et al.
+- **abs:** https://arxiv.org/abs/2602.12235 · **PDF:** https://arxiv.org/pdf/2602.12235v2
+
+## 12757. Abstractive Red-Teaming of Language Model Character
+
+- **ID:** `2602.12318` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Nate Rahn, Allison Qi, Avery Griffin, et al.
+- **abs:** https://arxiv.org/abs/2602.12318 · **PDF:** https://arxiv.org/pdf/2602.12318v1
+
+## 12758. Categorical Flow Maps
+
+- **ID:** `2602.12233` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Daan Roos, Oscar Davis, Floor Eijkelboom, et al.
+- **abs:** https://arxiv.org/abs/2602.12233 · **PDF:** https://arxiv.org/pdf/2602.12233v1
+
+## 12759. Free Lunch in Medical Image Foundation Model Pre-training via Randomized Synthesis and Disentanglement
+
+- **ID:** `2602.12317` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-12
+- **Authors:** Yuhan Wei, Yuting He, Linshan Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.12317 · **PDF:** https://arxiv.org/pdf/2602.12317v1
+
+## 12760. Diffusion Alignment Beyond KL: Variance Minimisation as Effective Policy Optimiser
+
+- **ID:** `2602.12229` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Zijing Ou, Jacob Si, Junyi Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.12229 · **PDF:** https://arxiv.org/pdf/2602.12229v1
+
+## 12761. Bandit Learning in Matching Markets with Interviews
+
+- **ID:** `2602.12224` · **primary:** `cs.GT` · **Submitted:** 2026-02-12
+- **Authors:** Amirmahdi Mirfakhar, Xuchuang Wang, Mengfan Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.12224 · **PDF:** https://arxiv.org/pdf/2602.12224v1
+
+## 12762. Towards On-Policy SFT: Distribution Discriminant Theory and its Applications in LLM Training
+
+- **ID:** `2602.12222` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Miaosen Zhang, Yishan Liu, Shuxia Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.12222 · **PDF:** https://arxiv.org/pdf/2602.12222v2
+
+## 12763. The Observer Effect in World Models: Invasive Adaptation Corrupts Latent Physics
+
+- **ID:** `2602.12218` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Christian Internò, Jumpei Yamaguchi, Loren Amdahl-Culleton, et al.
+- **abs:** https://arxiv.org/abs/2602.12218 · **PDF:** https://arxiv.org/pdf/2602.12218v1
+
+## 12764. VIRENA: Virtual Arena for Research, Education, and Democratic Innovation
+
+- **ID:** `2602.12207` · **primary:** `cs.HC` · **Submitted:** 2026-02-12
+- **Authors:** Emma Hoes, K. Jonathan Klueser, Fabrizio Gilardi
+- **abs:** https://arxiv.org/abs/2602.12207 · **PDF:** https://arxiv.org/pdf/2602.12207v2
+
+## 12765. DeepGen 1.0: A Lightweight Unified Multimodal Model for Advancing Image Generation and Editing
+
+- **ID:** `2602.12205` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Dianyi Wang, Ruihang Li, Feng Han, et al.
+- **abs:** https://arxiv.org/abs/2602.12205 · **PDF:** https://arxiv.org/pdf/2602.12205v2
+
+## 12766. Learning to Forget Attention: Memory Consolidation for Adaptive Compute Reduction
+
+- **ID:** `2602.12204` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma
+- **abs:** https://arxiv.org/abs/2602.12204 · **PDF:** https://arxiv.org/pdf/2602.12204v1
+
+## 12767. ExStrucTiny: A Benchmark for Schema-Variable Structured Information Extraction from Document Images
+
+- **ID:** `2602.12203` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Mathieu Sibue, Andres Muñoz Garza, Samuel Mensah, et al.
+- **abs:** https://arxiv.org/abs/2602.12203 · **PDF:** https://arxiv.org/pdf/2602.12203v1
+
+## 12768. GT-HarmBench: Benchmarking AI Safety Risks Through the Lens of Game Theory
+
+- **ID:** `2602.12316` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Pepijn Cobben, Xuanqiang Angelo Huang, Thao Amelia Pham, et al.
+- **abs:** https://arxiv.org/abs/2602.12316 · **PDF:** https://arxiv.org/pdf/2602.12316v1
+
+## 12769. Visual Reasoning Benchmark: Evaluating Multimodal LLMs on Classroom-Authentic Visual Problems from Primary Education
+
+- **ID:** `2602.12196` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Mohamed Huti, Alasdair Mackintosh, Amy Waldock, et al.
+- **abs:** https://arxiv.org/abs/2602.12196 · **PDF:** https://arxiv.org/pdf/2602.12196v1
+
+## 12770. AgenticShop: Benchmarking Agentic Product Curation for Personalized Web Shopping
+
+- **ID:** `2602.12315` · **primary:** `cs.IR` · **Submitted:** 2026-02-12
+- **Authors:** Sunghwan Kim, Ryang Heo, Yongsik Seo, et al.
+- **abs:** https://arxiv.org/abs/2602.12315 · **PDF:** https://arxiv.org/pdf/2602.12315v1
+
+## 12771. Query-focused and Memory-aware Reranker for Long Context Processing
+
+- **ID:** `2602.12192` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Yuqing Li, Jiangnan Li, Mo Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.12192 · **PDF:** https://arxiv.org/pdf/2602.12192v2
+
+## 12772. WaveFormer: Wavelet Embedding Transformer for Biomedical Signals
+
+- **ID:** `2602.12189` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Habib Irani, Bikram De, Vangelis Metsis
+- **abs:** https://arxiv.org/abs/2602.12189 · **PDF:** https://arxiv.org/pdf/2602.12189v1
+
+## 12773. Visible and Hyperspectral Imaging for Quality Assessment of Milk: Property Characterisation and Identification
+
+- **ID:** `2602.12313` · **primary:** `eess.IV` · **Submitted:** 2026-02-12
+- **Authors:** Massimo Martinelli, Elena Tomassi, Nafiou Arouna, et al.
+- **abs:** https://arxiv.org/abs/2602.12313 · **PDF:** https://arxiv.org/pdf/2602.12313v2
+
+## 12774. SAGEO Arena: A Realistic Environment for Evaluating Search-Augmented Generative Engine Optimization
+
+- **ID:** `2602.12187` · **primary:** `cs.IR` · **Submitted:** 2026-02-12
+- **Authors:** Sunghwan Kim, Wooseok Jeong, Serin Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.12187 · **PDF:** https://arxiv.org/pdf/2602.12187v1
+
+## 12775. Convex Markov Games and Beyond: New Proof of Existence, Characterization and Learning Algorithms for Nash Equilibria
+
+- **ID:** `2602.12181` · **primary:** `cs.GT` · **Submitted:** 2026-02-12
+- **Authors:** Anas Barakat, Ioannis Panageas, Antonios Varvitsiotis
+- **abs:** https://arxiv.org/abs/2602.12181 · **PDF:** https://arxiv.org/pdf/2602.12181v1
+
+## 12776. How Sampling Shapes LLM Alignment: From One-Shot Optima to Iterative Dynamics
+
+- **ID:** `2602.12180` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Yurong Chen, Yu He, Michael I. Jordan, et al.
+- **abs:** https://arxiv.org/abs/2602.12180 · **PDF:** https://arxiv.org/pdf/2602.12180v1
+
+## 12777. SAM3-LiteText: An Anatomical Study of the SAM3 Text Encoder for Efficient Vision-Language Segmentation
+
+- **ID:** `2602.12173` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Chengxi Zeng, Yuxuan Jiang, Ge Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.12173 · **PDF:** https://arxiv.org/pdf/2602.12173v1
+
+## 12778. Pedagogically-Inspired Data Synthesis for Language Model Knowledge Distillation
+
+- **ID:** `2602.12172` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Bowei He, Yankai Chen, Xiaokun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.12172 · **PDF:** https://arxiv.org/pdf/2602.12172v1
+
+## 12779. Statistical Parsing for Logical Information Retrieval
+
+- **ID:** `2602.12170` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Greg Coppola
+- **abs:** https://arxiv.org/abs/2602.12170 · **PDF:** https://arxiv.org/pdf/2602.12170v1
+
+## 12780. Sci-CoE: Co-evolving Scientific Reasoning LLMs via Geometric Consensus with Sparse Supervision
+
+- **ID:** `2602.12164` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Xiaohan He, Shiyang Feng, Songtao Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.12164 · **PDF:** https://arxiv.org/pdf/2602.12164v1
+
+## 12781. Amortized Molecular Optimization via Group Relative Policy Optimization
+
+- **ID:** `2602.12162` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Muhammad bin Javaid, Hasham Hussain, Ashima Khanna, et al.
+- **abs:** https://arxiv.org/abs/2602.12162 · **PDF:** https://arxiv.org/pdf/2602.12162v2
+
+## 12782. 3DGSNav: Enhancing Vision-Language Model Reasoning for Object Navigation via Active 3D Gaussian Splatting
+
+- **ID:** `2602.12159` · **primary:** `cs.RO` · **Submitted:** 2026-02-12
+- **Authors:** Wancai Zheng, Hao Chen, Xianlong Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.12159 · **PDF:** https://arxiv.org/pdf/2602.12159v1
+
+## 12783. SafeNeuron: Neuron-Level Safety Alignment for Large Language Models
+
+- **ID:** `2602.12158` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Zhaoxin Wang, Jiaming Liang, Fengbin Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.12158 · **PDF:** https://arxiv.org/pdf/2602.12158v1
+
+## 12784. dVoting: Fast Voting for dLLMs
+
+- **ID:** `2602.12153` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Sicheng Feng, Zigeng Chen, Xinyin Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.12153 · **PDF:** https://arxiv.org/pdf/2602.12153v1
+
+## 12785. GPT-4o Lacks Core Features of Theory of Mind
+
+- **ID:** `2602.12150` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** John Muchovej, Amanda Royka, Shane Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.12150 · **PDF:** https://arxiv.org/pdf/2602.12150v3
+
+## 12786. It's TIME: Towards the Next Generation of Time Series Forecasting Benchmarks
+
+- **ID:** `2602.12147` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Zhongzheng Qiao, Sheng Pan, Anni Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.12147 · **PDF:** https://arxiv.org/pdf/2602.12147v3
+
+## 12787. Seq2Seq2Seq: Lossless Data Compression via Discrete Latent Transformers and Reinforcement Learning
+
+- **ID:** `2602.12146` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Mahdi Khodabandeh, Ghazal Shabani, Arash Yousefi Jordehi, et al.
+- **abs:** https://arxiv.org/abs/2602.12146 · **PDF:** https://arxiv.org/pdf/2602.12146v1
+
+## 12788. On the Adoption of AI Coding Agents in Open-source Android and iOS Development
+
+- **ID:** `2602.12144` · **primary:** `cs.SE` · **Submitted:** 2026-02-12
+- **Authors:** Muhammad Ahmad Khan, Hasnain Ali, Muneeb Rana, et al.
+- **abs:** https://arxiv.org/abs/2602.12144 · **PDF:** https://arxiv.org/pdf/2602.12144v1
+
+## 12789. STAR : Bridging Statistical and Agentic Reasoning for Large Model Performance Prediction
+
+- **ID:** `2602.12143` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Xiaoxiao Wang, Chunxiao Li, Junying Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.12143 · **PDF:** https://arxiv.org/pdf/2602.12143v1
+
+## 12790. Oscillators Are All You Need: Irregular Time Series Modelling via Damped Harmonic Oscillators with Closed-Form Solutions
+
+- **ID:** `2602.12139` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Yashas Shende, Aritra Das, Reva Laxmi Chauhan, et al.
+- **abs:** https://arxiv.org/abs/2602.12139 · **PDF:** https://arxiv.org/pdf/2602.12139v1
+
+## 12791. CitiLink-Minutes: A Multilayer Annotated Dataset of Municipal Meeting Minutes
+
+- **ID:** `2602.12137` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Ricardo Campos, Ana Filipa Pacheco, Ana Luísa Fernandes, et al.
+- **abs:** https://arxiv.org/abs/2602.12137 · **PDF:** https://arxiv.org/pdf/2602.12137v2
+
+## 12792. WavBench: Benchmarking Reasoning, Colloquialism, and Paralinguistics for End-to-End Spoken Dialogue Models
+
+- **ID:** `2602.12135` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Yangzhuo Li, Shengpeng Ji, Yifu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.12135 · **PDF:** https://arxiv.org/pdf/2602.12135v2
+
+## 12793. Value Alignment Tax: Measuring Value Trade-offs in LLM Alignment
+
+- **ID:** `2602.12134` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Jiajun Chen, Hua Shen
+- **abs:** https://arxiv.org/abs/2602.12134 · **PDF:** https://arxiv.org/pdf/2602.12134v1
+
+## 12794. Neutral Prompts, Non-Neutral People: Quantifying Gender and Skin-Tone Bias in Gemini Flash 2.5 Image and GPT Image 1.5
+
+- **ID:** `2602.12133` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Roberto Balestri
+- **abs:** https://arxiv.org/abs/2602.12133 · **PDF:** https://arxiv.org/pdf/2602.12133v1
+
+## 12795. A Rule-based Computational Model for Gaidhlig Morphology
+
+- **ID:** `2602.12132` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Peter J Barclay
+- **abs:** https://arxiv.org/abs/2602.12132 · **PDF:** https://arxiv.org/pdf/2602.12132v1
+
+## 12796. Towards Personalized Bangla Book Recommendation: A Large-Scale Multi-Entity Book Graph Dataset
+
+- **ID:** `2602.12129` · **primary:** `cs.IR` · **Submitted:** 2026-02-12
+- **Authors:** Rahin Arefin Ahmed, Md. Anik Chowdhury, Sakil Ahmed Sheikh Reza, et al.
+- **abs:** https://arxiv.org/abs/2602.12129 · **PDF:** https://arxiv.org/pdf/2602.12129v1
+
+## 12797. HLA: Hadamard Linear Attention
+
+- **ID:** `2602.12128` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Hanno Ackermann, Hong Cai, Mohsen Ghafoorian, et al.
+- **abs:** https://arxiv.org/abs/2602.12128 · **PDF:** https://arxiv.org/pdf/2602.12128v1
+
+## 12798. Quantifying Hallucinations in Language Language Models on Medical Textbooks
+
+- **ID:** `2603.09986` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Brandon C. Colelough, Davis Bartels, Dina Demner-Fushman
+- **abs:** https://arxiv.org/abs/2603.09986 · **PDF:** https://arxiv.org/pdf/2603.09986v1
+
+## 12799. Learning beyond Teacher: Generalized On-Policy Distillation with Reward Extrapolation
+
+- **ID:** `2602.12125` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Wenkai Yang, Weijie Liu, Ruobing Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.12125 · **PDF:** https://arxiv.org/pdf/2602.12125v2
+
+## 12800. Capability-Oriented Training Induced Alignment Risk
+
+- **ID:** `2602.12124` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Yujun Zhou, Yue Huang, Han Bao, et al.
+- **abs:** https://arxiv.org/abs/2602.12124 · **PDF:** https://arxiv.org/pdf/2602.12124v1
+
+## 12801. Meta-Sel: Efficient Demonstration Selection for In-Context Learning via Supervised Meta-Learning
+
+- **ID:** `2602.12123` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Xubin Wang, Weijia Jia
+- **abs:** https://arxiv.org/abs/2602.12123 · **PDF:** https://arxiv.org/pdf/2602.12123v1
+
+## 12802. Zero-Shot Time Series Foundation Models for Annual Institutional Forecasting Under Data Sparsity
+
+- **ID:** `2602.12120` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Jittarin Jetwiriyanon, Teo Susnjak, Surangika Ranathunga
+- **abs:** https://arxiv.org/abs/2602.12120 · **PDF:** https://arxiv.org/pdf/2602.12120v2
+
+## 12803. KAN-FIF: Spline-Parameterized Lightweight Physics-based Tropical Cyclone Estimation on Meteorological Satellite
+
+- **ID:** `2602.12117` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Jiakang Shen, Qinghui Chen, Runtong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.12117 · **PDF:** https://arxiv.org/pdf/2602.12117v2
+
+## 12804. P-GenRM: Personalized Generative Reward Model with Test-time User-based Scaling
+
+- **ID:** `2602.12116` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Pinyi Zhang, Ting-En Lin, Yuchuan Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.12116 · **PDF:** https://arxiv.org/pdf/2602.12116v1
+
+## 12805. Stop Unnecessary Reflection: Training LRMs for Efficient Reasoning with Adaptive Reflection and Length Coordinated Penalty
+
+- **ID:** `2602.12113` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Zewei Yu, Lirong Gao, Yuke Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.12113 · **PDF:** https://arxiv.org/pdf/2602.12113v2
+
+## 12806. Few-Shot Design Optimization by Exploiting Auxiliary Information
+
+- **ID:** `2602.12112` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Arjun Mani, Carl Vondrick, Richard Zemel
+- **abs:** https://arxiv.org/abs/2602.12112 · **PDF:** https://arxiv.org/pdf/2602.12112v1
+
+## 12807. The Pensieve Paradigm: Stateful Language Models Mastering Their Own Context
+
+- **ID:** `2602.12108` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Xiaoyuan Liu, Tian Liang, Dongyang Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.12108 · **PDF:** https://arxiv.org/pdf/2602.12108v1
+
+## 12808. On the Complexity of Offline Reinforcement Learning with $Q^\star$-Approximation and Partial Coverage
+
+- **ID:** `2602.12107` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Haolin Liu, Braham Snyder, Chen-Yu Wei
+- **abs:** https://arxiv.org/abs/2602.12107 · **PDF:** https://arxiv.org/pdf/2602.12107v1
+
+## 12809. Iskra: A System for Inverse Geometry Processing
+
+- **ID:** `2602.12105` · **primary:** `cs.GR` · **Submitted:** 2026-02-12
+- **Authors:** Ana Dodik, Ahmed H. Mahmoud, Justin Solomon
+- **abs:** https://arxiv.org/abs/2602.12105 · **PDF:** https://arxiv.org/pdf/2602.12105v1
+
+## 12810. Multi Graph Search for High-Dimensional Robot Motion Planning
+
+- **ID:** `2602.12096` · **primary:** `cs.RO` · **Submitted:** 2026-02-12
+- **Authors:** Itamar Mishani, Maxim Likhachev
+- **abs:** https://arxiv.org/abs/2602.12096 · **PDF:** https://arxiv.org/pdf/2602.12096v1
+
+## 12811. Perceptual Self-Reflection in Agentic Physics Simulation Code Generation
+
+- **ID:** `2602.12311` · **primary:** `cs.SE` · **Submitted:** 2026-02-12
+- **Authors:** Prashant Shende, Bradley Camburn
+- **abs:** https://arxiv.org/abs/2602.12311 · **PDF:** https://arxiv.org/pdf/2602.12311v1
+
+## 12812. DeepSight: An All-in-One LM Safety Toolkit
+
+- **ID:** `2602.12092` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Bo Zhang, Jiaxuan Guo, Lijun Li, et al.
+- **abs:** https://arxiv.org/abs/2602.12092 · **PDF:** https://arxiv.org/pdf/2602.12092v1
+
+## 12813. Choose Your Agent: Tradeoffs in Adopting AI Advisors, Coaches, and Delegates in Multi-Party Negotiation
+
+- **ID:** `2602.12089` · **primary:** `cs.GT` · **Submitted:** 2026-02-12
+- **Authors:** Kehang Zhu, Nithum Thain, Vivian Tsai, et al.
+- **abs:** https://arxiv.org/abs/2602.12089 · **PDF:** https://arxiv.org/pdf/2602.12089v2
+
+## 12814. Geometry of Uncertainty: Learning Metric Spaces for Multimodal State Estimation in RL
+
+- **ID:** `2602.12087` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Alfredo Reichlin, Adriano Pacciarelli, Danica Kragic, et al.
+- **abs:** https://arxiv.org/abs/2602.12087 · **PDF:** https://arxiv.org/pdf/2602.12087v1
+
+## 12815. Differentiable Modal Logic for Multi-Agent Diagnosis, Orchestration and Communication
+
+- **ID:** `2602.12083` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Antonin Sulc
+- **abs:** https://arxiv.org/abs/2602.12083 · **PDF:** https://arxiv.org/pdf/2602.12083v1
+
+## 12816. Empirical Gaussian Processes
+
+- **ID:** `2602.12082` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Jihao Andreas Lin, Sebastian Ament, Louis C. Tiao, et al.
+- **abs:** https://arxiv.org/abs/2602.12082 · **PDF:** https://arxiv.org/pdf/2602.12082v1
+
+## 12817. PathCRF: Ball-Free Soccer Event Detection via Possession Path Inference from Player Trajectories
+
+- **ID:** `2602.12080` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Hyunsung Kim, Kunhee Lee, Sangwoo Seo, et al.
+- **abs:** https://arxiv.org/abs/2602.12080 · **PDF:** https://arxiv.org/pdf/2602.12080v1
+
+## 12818. Tiny Recursive Reasoning with Mamba-2 Attention Hybrid
+
+- **ID:** `2602.12078` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Wenlong Wang, Fergal Reid
+- **abs:** https://arxiv.org/abs/2602.12078 · **PDF:** https://arxiv.org/pdf/2602.12078v2
+
+## 12819. ModelWisdom: An Integrated Toolkit for TLA+ Model Visualization, Digest and Repair
+
+- **ID:** `2602.12058` · **primary:** `cs.SE` · **Submitted:** 2026-02-12
+- **Authors:** Zhiyong Chen, Jialun Cao, Chang Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.12058 · **PDF:** https://arxiv.org/pdf/2602.12058v1
+
+## 12820. LawThinker: A Deep Research Legal Agent in Dynamic Environments
+
+- **ID:** `2602.12056` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Xinyu Yang, Chenlong Deng, Tongyu Wen, et al.
+- **abs:** https://arxiv.org/abs/2602.12056 · **PDF:** https://arxiv.org/pdf/2602.12056v1
+
+## 12821. Multi UAVs Preflight Planning in a Shared and Dynamic Airspace
+
+- **ID:** `2602.12055` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Amath Sow, Mauricio Rodriguez Cesen, Fabiola Martins Campos de Oliveira, et al.
+- **abs:** https://arxiv.org/abs/2602.12055 · **PDF:** https://arxiv.org/pdf/2602.12055v1
+
+## 12822. Improving HPC Code Generation Capability of LLMs via Online Reinforcement Learning with Real-Machine Benchmark Rewards
+
+- **ID:** `2602.12049` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Ryo Mikasa, Shun-ichiro Hayashi, Daichi Mukunoki, et al.
+- **abs:** https://arxiv.org/abs/2602.12049 · **PDF:** https://arxiv.org/pdf/2602.12049v1
+
+## 12823. Safety Beyond the Training Data: Robust Out-of-Distribution MPC via Conformalized System Level Synthesis
+
+- **ID:** `2602.12047` · **primary:** `cs.RO` · **Submitted:** 2026-02-12
+- **Authors:** Anutam Srinivasan, Antoine Leeman, Glen Chou
+- **abs:** https://arxiv.org/abs/2602.12047 · **PDF:** https://arxiv.org/pdf/2602.12047v1
+
+## 12824. Fourier Transformers for Latent Crystallographic Diffusion and Generative Modeling
+
+- **ID:** `2602.12045` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Jed A. Duersch, Elohan Veillon, Astrid Klipfel, et al.
+- **abs:** https://arxiv.org/abs/2602.12045 · **PDF:** https://arxiv.org/pdf/2602.12045v1
+
+## 12825. Improved Inference for CSDID Using the Cluster Jackknife
+
+- **ID:** `2602.12043` · **primary:** `econ.EM` · **Submitted:** 2026-02-12
+- **Authors:** Sunny R. Karim, Morten Ørregaard Nielsen, James G. MacKinnon, et al.
+- **abs:** https://arxiv.org/abs/2602.12043 · **PDF:** https://arxiv.org/pdf/2602.12043v1
+
+## 12826. The Implicit Bias of Logit Regularization
+
+- **ID:** `2602.12039` · **primary:** `stat.ML` · **Submitted:** 2026-02-12
+- **Authors:** Alon Beck, Yohai Bar Sinai, Noam Levi
+- **abs:** https://arxiv.org/abs/2602.12039 · **PDF:** https://arxiv.org/pdf/2602.12039v2
+
+## 12827. An Empirical Study of the Imbalance Issue in Software Vulnerability Detection
+
+- **ID:** `2602.12038` · **primary:** `cs.SE` · **Submitted:** 2026-02-12
+- **Authors:** Yuejun Guo, Qiang Hu, Qiang Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.12038 · **PDF:** https://arxiv.org/pdf/2602.12038v1
+
+## 12828. Composition-RL: Compose Your Verifiable Prompts for Reinforcement Learning of Large Language Models
+
+- **ID:** `2602.12036` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Xin Xu, Clive Bai, Kai Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.12036 · **PDF:** https://arxiv.org/pdf/2602.12036v1
+
+## 12829. PrefillShare: A Shared Prefill Module for KV Reuse in Multi-LLM Disaggregated Serving
+
+- **ID:** `2602.12029` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Sunghyeon Woo, Hoseung Kim, Sunghwan Shim, et al.
+- **abs:** https://arxiv.org/abs/2602.12029 · **PDF:** https://arxiv.org/pdf/2602.12029v1
+
+## 12830. Protein Circuit Tracing via Cross-layer Transcoders
+
+- **ID:** `2602.12026` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Darin Tsui, Kunal Talreja, Daniel Saeedi, et al.
+- **abs:** https://arxiv.org/abs/2602.12026 · **PDF:** https://arxiv.org/pdf/2602.12026v1
+
+## 12831. Decomposition of Spillover Effects Under Misspecification: Pseudo-true Estimands and a Local-Global Extension
+
+- **ID:** `2602.12023` · **primary:** `econ.EM` · **Submitted:** 2026-02-12
+- **Authors:** Yechan Park, Xiaodong Yang
+- **abs:** https://arxiv.org/abs/2602.12023 · **PDF:** https://arxiv.org/pdf/2602.12023v3
+
+## 12832. Improved state mixing in higher-order and block diagonal linear recurrent networks
+
+- **ID:** `2602.12021` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Igor Dubinin, Antonio Orvieto, Felix Effenberger
+- **abs:** https://arxiv.org/abs/2602.12021 · **PDF:** https://arxiv.org/pdf/2602.12021v2
+
+## 12833. Artificial intelligence is creating a new global linguistic hierarchy
+
+- **ID:** `2602.12018` · **primary:** `cs.CY` · **Submitted:** 2026-02-12
+- **Authors:** Giulia Occhini, Kumiko Tanaka-Ishii, Anna Barford, et al.
+- **abs:** https://arxiv.org/abs/2602.12018 · **PDF:** https://arxiv.org/pdf/2602.12018v1
+
+## 12834. Disentangling Ambiguity from Instability in Large Language Models: A Clinical Text-to-SQL Case Study
+
+- **ID:** `2602.12015` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Angelo Ziletti, Leonardo D'Ambrosi
+- **abs:** https://arxiv.org/abs/2602.12015 · **PDF:** https://arxiv.org/pdf/2602.12015v1
+
+## 12835. FedGRPO: Privately Optimizing Foundation Models with Group-Relative Rewards from Domain Client
+
+- **ID:** `2602.12014` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Gongxi Zhu, Hanlin Gu, Lixin Fan, et al.
+- **abs:** https://arxiv.org/abs/2602.12014 · **PDF:** https://arxiv.org/pdf/2602.12014v1
+
+## 12836. InjectRBP: Steering Large Language Model Reasoning Behavior via Pattern Injection
+
+- **ID:** `2602.12013` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Xiuping Wu, Zhao Yu, Yuxin Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.12013 · **PDF:** https://arxiv.org/pdf/2602.12013v1
+
+## 12837. On the Sensitivity of Firing Rate-Based Federated Spiking Neural Networks to Differential Privacy
+
+- **ID:** `2602.12009` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Luiz Pereira, Mirko Perkusich, Dalton Valadares, et al.
+- **abs:** https://arxiv.org/abs/2602.12009 · **PDF:** https://arxiv.org/pdf/2602.12009v1
+
+## 12838. LaCy: What Small Language Models Can and Should Learn is Not Just a Question of Loss
+
+- **ID:** `2602.12005` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Szilvia Ujváry, Louis Béthune, Pierre Ablin, et al.
+- **abs:** https://arxiv.org/abs/2602.12005 · **PDF:** https://arxiv.org/pdf/2602.12005v2
+
+## 12839. CSEval: A Framework for Evaluating Clinical Semantics in Text-to-Image Generation
+
+- **ID:** `2602.12004` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Robert Cronshaw, Konstantinos Vilouras, Junyu Yan, et al.
+- **abs:** https://arxiv.org/abs/2602.12004 · **PDF:** https://arxiv.org/pdf/2602.12004v1
+
+## 12840. Momentum LMS Theory beyond Stationarity: Stability, Tracking, and Regret
+
+- **ID:** `2602.11995` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Yifei Jin, Xin Zheng, Lei Guo
+- **abs:** https://arxiv.org/abs/2602.11995 · **PDF:** https://arxiv.org/pdf/2602.11995v1
+
+## 12841. Evaluating AGENTS.md: Are Repository-Level Context Files Helpful for Coding Agents?
+
+- **ID:** `2602.11988` · **primary:** `cs.SE` · **Submitted:** 2026-02-12
+- **Authors:** Thibaud Gloaguen, Niels Mündler, Mark Müller, et al.
+- **abs:** https://arxiv.org/abs/2602.11988 · **PDF:** https://arxiv.org/pdf/2602.11988v1
+
+## 12842. Automatic Simplification of Common Vulnerabilities and Exposures Descriptions
+
+- **ID:** `2602.11982` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Varpu Vehomäki, Kimmo K. Kaski
+- **abs:** https://arxiv.org/abs/2602.11982 · **PDF:** https://arxiv.org/pdf/2602.11982v1
+
+## 12843. Accelerating Robotic Reinforcement Learning with Agent Guidance
+
+- **ID:** `2602.11978` · **primary:** `cs.RO` · **Submitted:** 2026-02-12
+- **Authors:** Haojun Chen, Zili Zou, Chengdong Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.11978 · **PDF:** https://arxiv.org/pdf/2602.11978v2
+
+## 12844. Calibrated Bayesian Deep Learning for Explainable Decision Support Systems Based on Medical Imaging
+
+- **ID:** `2602.11973` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Hua Xu, Julián D. Arias-Londoño, Juan I. Godino-Llorente
+- **abs:** https://arxiv.org/abs/2602.11973 · **PDF:** https://arxiv.org/pdf/2602.11973v1
+
+## 12845. DHPLT: large-scale multilingual diachronic corpora and word representations for semantic change modelling
+
+- **ID:** `2602.11968` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Mariia Fedorova, Andrey Kutuzov, Khonzoda Umarova
+- **abs:** https://arxiv.org/abs/2602.11968 · **PDF:** https://arxiv.org/pdf/2602.11968v1
+
+## 12846. Manifold-Aware Temporal Domain Generalization for Large Language Models
+
+- **ID:** `2602.11965` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Yiheng Yao, Zekun Cai, Xinyuan Song, et al.
+- **abs:** https://arxiv.org/abs/2602.11965 · **PDF:** https://arxiv.org/pdf/2602.11965v1
+
+## 12847. Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments
+
+- **ID:** `2602.11964` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Romain Froger, Pierre Andrews, Matteo Bettini, et al.
+- **abs:** https://arxiv.org/abs/2602.11964 · **PDF:** https://arxiv.org/pdf/2602.11964v1
+
+## 12848. Scaling Model and Data for Multilingual Machine Translation with Open Large Language Models
+
+- **ID:** `2602.11961` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Yuzhe Shang, Pengzhi Gao, Wei Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.11961 · **PDF:** https://arxiv.org/pdf/2602.11961v2
+
+## 12849. Benchmarking Vision-Language Models for French PDF-to-Markdown Conversion
+
+- **ID:** `2602.11960` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Bruno Rigal, Victor Dupriez, Alexis Mignon, et al.
+- **abs:** https://arxiv.org/abs/2602.11960 · **PDF:** https://arxiv.org/pdf/2602.11960v1
+
+## 12850. RAM-Net: Expressive Linear Attention with Selectively Addressable Memory
+
+- **ID:** `2602.11958` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Kaicheng Xiao, Haotian Li, Liran Dong, et al.
+- **abs:** https://arxiv.org/abs/2602.11958 · **PDF:** https://arxiv.org/pdf/2602.11958v1
+
+## 12851. Are Two LLMs Better Than One? A Student-Teacher Dual-Head LLMs Architecture for Pharmaceutical Content Optimization
+
+- **ID:** `2602.11957` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Suyash Mishra, Qiang Li, Anubhav Girdhar
+- **abs:** https://arxiv.org/abs/2602.11957 · **PDF:** https://arxiv.org/pdf/2602.11957v1
+
+## 12852. TAVAE: A VAE with Adaptable Priors Explains Contextual Modulation in the Visual Cortex
+
+- **ID:** `2602.11956` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-12
+- **Authors:** Balázs Meszéna, Keith T. Murray, Julien Corbo, et al.
+- **abs:** https://arxiv.org/abs/2602.11956 · **PDF:** https://arxiv.org/pdf/2602.11956v1
+
+## 12853. An Integrated Causal Inference Framework for Traffic Safety Modeling with Semantic Street-View Visual Features
+
+- **ID:** `2602.13339` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Lishan Sun, Yujia Cheng, Pengfei Cui, et al.
+- **abs:** https://arxiv.org/abs/2602.13339 · **PDF:** https://arxiv.org/pdf/2602.13339v1
+
+## 12854. Insights on Muon from Simple Quadratics
+
+- **ID:** `2602.11948` · **primary:** `math.OC` · **Submitted:** 2026-02-12
+- **Authors:** Antoine Gonon, Andreea-Alexandra Muşat, Nicolas Boumal
+- **abs:** https://arxiv.org/abs/2602.11948 · **PDF:** https://arxiv.org/pdf/2602.11948v1
+
+## 12855. Mixed-Integer Programming for Change-point Detection
+
+- **ID:** `2602.11947` · **primary:** `math.OC` · **Submitted:** 2026-02-12
+- **Authors:** Apoorva Narula, Santanu S. Dey, Yao Xie
+- **abs:** https://arxiv.org/abs/2602.11947 · **PDF:** https://arxiv.org/pdf/2602.11947v1
+
+## 12856. Towards Performance-Enhanced Model-Contrastive Federated Learning using Historical Information in Heterogeneous Scenarios
+
+- **ID:** `2602.11945` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Hongliang Zhang, Jiguo Yu, Guijuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.11945 · **PDF:** https://arxiv.org/pdf/2602.11945v1
+
+## 12857. Using predictive multiplicity to measure individual performance within the AI Act
+
+- **ID:** `2602.11944` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Karolin Frohnapfel, Mara Seyfert, Sebastian Bordt, et al.
+- **abs:** https://arxiv.org/abs/2602.11944 · **PDF:** https://arxiv.org/pdf/2602.11944v1
+
+## 12858. Synthesis of Late Gadolinium Enhancement Images via Implicit Neural Representations for Cardiac Scar Segmentation
+
+- **ID:** `2602.11942` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Soufiane Ben Haddou, Laura Alvarez-Florez, Erik J. Bekkers, et al.
+- **abs:** https://arxiv.org/abs/2602.11942 · **PDF:** https://arxiv.org/pdf/2602.11942v1
+
+## 12859. IncompeBench: A Permissively Licensed, Fine-Grained Benchmark for Music Information Retrieval
+
+- **ID:** `2602.11941` · **primary:** `cs.IR` · **Submitted:** 2026-02-12
+- **Authors:** Benjamin Clavié, Atoof Shakir, Jonah Turner, et al.
+- **abs:** https://arxiv.org/abs/2602.11941 · **PDF:** https://arxiv.org/pdf/2602.11941v1
+
+## 12860. Temporally Unified Adversarial Perturbations for Time Series Forecasting
+
+- **ID:** `2602.11940` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Ruixian Su, Yukun Bao, Xinze Zhang
+- **abs:** https://arxiv.org/abs/2602.11940 · **PDF:** https://arxiv.org/pdf/2602.11940v1
+
+## 12861. Do Large Language Models Adapt to Language Variation across Socioeconomic Status?
+
+- **ID:** `2602.11939` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Elisa Bassignana, Mike Zhang, Dirk Hovy, et al.
+- **abs:** https://arxiv.org/abs/2602.11939 · **PDF:** https://arxiv.org/pdf/2602.11939v1
+
+## 12862. Who is the richest club in the championship? Detecting and Rewriting Underspecified Questions Improve QA Performance
+
+- **ID:** `2602.11938` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Yunchong Huang, Gianni Barlacchi, Sandro Pezzelle
+- **abs:** https://arxiv.org/abs/2602.11938 · **PDF:** https://arxiv.org/pdf/2602.11938v4
+
+## 12863. Extending Puzzle for Mixture-of-Experts Reasoning Models with Application to GPT-OSS Acceleration
+
+- **ID:** `2602.11937` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Akhiad Bercovich, Nir Ailon, Vladimir Anisimov, et al.
+- **abs:** https://arxiv.org/abs/2602.11937 · **PDF:** https://arxiv.org/pdf/2602.11937v2
+
+## 12864. Cross-Modal Robustness Transfer (CMRT): Training Robust Speech Translation Models Using Adversarial Text
+
+- **ID:** `2602.11933` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Abderrahmane Issam, Yusuf Can Semerci, Jan Scholtes, et al.
+- **abs:** https://arxiv.org/abs/2602.11933 · **PDF:** https://arxiv.org/pdf/2602.11933v1
+
+## 12865. AdaptEvolve: Improving Efficiency of Evolutionary AI Agents through Adaptive Model Selection
+
+- **ID:** `2602.11931` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Pretam Ray, Pratik Prabhanjan Brahma, Zicheng Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.11931 · **PDF:** https://arxiv.org/pdf/2602.11931v1
+
+## 12866. Who Does What? Archetypes of Roles Assigned to LLMs During Human-AI Decision-Making
+
+- **ID:** `2602.11924` · **primary:** `cs.HC` · **Submitted:** 2026-02-12
+- **Authors:** Shreya Chappidi, Jatinder Singh, Andra V. Krauze
+- **abs:** https://arxiv.org/abs/2602.11924 · **PDF:** https://arxiv.org/pdf/2602.11924v1
+
+## 12867. Learning Conditional Averages
+
+- **ID:** `2602.11920` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Marco Bressan, Nataly Brukhim, Nicolo Cesa-Bianchi, et al.
+- **abs:** https://arxiv.org/abs/2602.11920 · **PDF:** https://arxiv.org/pdf/2602.11920v1
+
+## 12868. ALADIN: Accuracy-Latency-Aware Design-space Inference Analysis for Embedded AI Accelerators
+
+- **ID:** `2603.08722` · **primary:** `cs.AR` · **Submitted:** 2026-02-12
+- **Authors:** T. Baldi, D. Casini, A. Biondi
+- **abs:** https://arxiv.org/abs/2603.08722 · **PDF:** https://arxiv.org/pdf/2603.08722v1
+
+## 12869. DynaHOI: Benchmarking Hand-Object Interaction for Dynamic Target
+
+- **ID:** `2602.11919` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** BoCheng Hu, Zhonghan Zhao, Kaiyue Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.11919 · **PDF:** https://arxiv.org/pdf/2602.11919v1
+
+## 12870. MEME: Modeling the Evolutionary Modes of Financial Markets
+
+- **ID:** `2602.11918` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Taian Guo, Haiyang Shen, Junyu Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.11918 · **PDF:** https://arxiv.org/pdf/2602.11918v1
+
+## 12871. AlphaPROBE: Alpha Mining via Principled Retrieval and On-graph biased evolution
+
+- **ID:** `2602.11917` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Taian Guo, Haiyang Shen, Junyu Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.11917 · **PDF:** https://arxiv.org/pdf/2602.11917v1
+
+## 12872. TADA! Tuning Audio Diffusion Models through Activation Steering
+
+- **ID:** `2602.11910` · **primary:** `cs.SD` · **Submitted:** 2026-02-12
+- **Authors:** Łukasz Staniszewski, Katarzyna Zaleska, Mateusz Modrzejewski, et al.
+- **abs:** https://arxiv.org/abs/2602.11910 · **PDF:** https://arxiv.org/pdf/2602.11910v1
+
+## 12873. Echo: Towards Advanced Audio Comprehension via Audio-Interleaved Reasoning
+
+- **ID:** `2602.11909` · **primary:** `cs.SD` · **Submitted:** 2026-02-12
+- **Authors:** Daiqing Wu, Xuan Zhang, Dongbao Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.11909 · **PDF:** https://arxiv.org/pdf/2602.11909v2
+
+## 12874. When Should LLMs Be Less Specific? Selective Abstraction for Reliable Long-Form Text Generation
+
+- **ID:** `2602.11908` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Shani Goren, Ido Galil, Ran El-Yaniv
+- **abs:** https://arxiv.org/abs/2602.11908 · **PDF:** https://arxiv.org/pdf/2602.11908v2
+
+## 12875. Leveraging LLMs to support co-evolution between definitions and instances of textual DSLs: A Systematic Evaluation
+
+- **ID:** `2602.11904` · **primary:** `cs.SE` · **Submitted:** 2026-02-12
+- **Authors:** Weixing Zhang, Bowen Jiang, Yuhong Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.11904 · **PDF:** https://arxiv.org/pdf/2602.11904v1
+
+## 12876. Mitigating Mismatch within Reference-based Preference Optimization
+
+- **ID:** `2602.11902` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Suqin Yuan, Xingrui Yu, Jiyang Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.11902 · **PDF:** https://arxiv.org/pdf/2602.11902v1
+
+## 12877. Benchmark Illusion: Disagreement among LLMs and Its Scientific Consequences
+
+- **ID:** `2602.11898` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Eddie Yang, Dashun Wang
+- **abs:** https://arxiv.org/abs/2602.11898 · **PDF:** https://arxiv.org/pdf/2602.11898v1
+
+## 12878. Agentic AI for Cybersecurity: A Meta-Cognitive Architecture for Governable Autonomy
+
+- **ID:** `2602.11897` · **primary:** `cs.CR` · **Submitted:** 2026-02-12
+- **Authors:** Andrei Kojukhov, Arkady Bovshover
+- **abs:** https://arxiv.org/abs/2602.11897 · **PDF:** https://arxiv.org/pdf/2602.11897v2
+
+## 12879. Universal Diffusion-Based Probabilistic Downscaling
+
+- **ID:** `2602.11893` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Roberto Molinaro, Niall Siegenheim, Henry Martin, et al.
+- **abs:** https://arxiv.org/abs/2602.11893 · **PDF:** https://arxiv.org/pdf/2602.11893v2
+
+## 12880. LLM-based Triplet Extraction from Financial Reports
+
+- **ID:** `2602.11886` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Dante Wesslund, Ville Stenström, Pontus Linde, et al.
+- **abs:** https://arxiv.org/abs/2602.11886 · **PDF:** https://arxiv.org/pdf/2602.11886v1
+
+## 12881. Where Bits Matter in World Model Planning: A Paired Mixed-Bit Study for Efficient Spatial Reasoning
+
+- **ID:** `2602.11882` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Suraj Ranganath, Anish Patnaik, Vaishak Menon
+- **abs:** https://arxiv.org/abs/2602.11882 · **PDF:** https://arxiv.org/pdf/2602.11882v1
+
+## 12882. From Atoms to Trees: Building a Structured Feature Forest with Hierarchical Sparse Autoencoders
+
+- **ID:** `2602.11881` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Yifan Luo, Yang Zhan, Jiedong Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.11881 · **PDF:** https://arxiv.org/pdf/2602.11881v1
+
+## 12883. SynthRAR: Ring Artifacts Reduction in CT with Unrolled Network and Synthetic Data Training
+
+- **ID:** `2602.11880` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Hongxu Yang, Levente Lippenszky, Edina Timko, et al.
+- **abs:** https://arxiv.org/abs/2602.11880 · **PDF:** https://arxiv.org/pdf/2602.11880v1
+
+## 12884. Towards Fair and Comprehensive Evaluation of Routers in Collaborative LLM Systems
+
+- **ID:** `2602.11877` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Wanxing Wu, He Zhu, Yixia Li, et al.
+- **abs:** https://arxiv.org/abs/2602.11877 · **PDF:** https://arxiv.org/pdf/2602.11877v1
+
+## 12885. DMAP: A Distribution Map for Text
+
+- **ID:** `2602.11871` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Tom Kempton, Julia Rozanova, Parameswaran Kamalaruban, et al.
+- **abs:** https://arxiv.org/abs/2602.11871 · **PDF:** https://arxiv.org/pdf/2602.11871v1
+
+## 12886. Intelligent AI Delegation
+
+- **ID:** `2602.11865` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Nenad Tomašev, Matija Franklin, Simon Osindero
+- **abs:** https://arxiv.org/abs/2602.11865 · **PDF:** https://arxiv.org/pdf/2602.11865v1
+
+## 12887. In-Context Function Learning in Large Language Models
+
+- **ID:** `2602.11863` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Elif Akata, Konstantinos Voudouris, Vincent Fortuin, et al.
+- **abs:** https://arxiv.org/abs/2602.11863 · **PDF:** https://arxiv.org/pdf/2602.11863v1
+
+## 12888. A$^{2}$V-SLP: Alignment-Aware Variational Modeling for Disentangled Sign Language Production
+
+- **ID:** `2602.11861` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Sümeyye Meryem Taşyürek, Enis Mücahid İskender, Hacer Yalim Keles
+- **abs:** https://arxiv.org/abs/2602.11861 · **PDF:** https://arxiv.org/pdf/2602.11861v1
+
+## 12889. Talk2DM: Enabling Natural Language Querying and Commonsense Reasoning for Vehicle-Road-Cloud Integrated Dynamic Maps with Large Language Models
+
+- **ID:** `2602.11860` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Lu Tao, Jinxuan Luo, Yousuke Watanabe, et al.
+- **abs:** https://arxiv.org/abs/2602.11860 · **PDF:** https://arxiv.org/pdf/2602.11860v1
+
+## 12890. Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perception
+
+- **ID:** `2602.11858` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Lai Wei, Liangbo He, Jun Lan, et al.
+- **abs:** https://arxiv.org/abs/2602.11858 · **PDF:** https://arxiv.org/pdf/2602.11858v2
+
+## 12891. Scale-Invariant Fast Convergence in Games
+
+- **ID:** `2602.11857` · **primary:** `cs.GT` · **Submitted:** 2026-02-12
+- **Authors:** Taira Tsuchiya, Haipeng Luo, Shinji Ito
+- **abs:** https://arxiv.org/abs/2602.11857 · **PDF:** https://arxiv.org/pdf/2602.11857v1
+
+## 12892. Robust Optimization Approach and Learning Based Hide-and-Seek Game for Resilient Network Design
+
+- **ID:** `2602.11854` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Mohammad Khosravi, Setareh Maghsudi
+- **abs:** https://arxiv.org/abs/2602.11854 · **PDF:** https://arxiv.org/pdf/2602.11854v1
+
+## 12893. Prototype Transformer: Towards Language Model Architectures Interpretable by Design
+
+- **ID:** `2602.11852` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Yordan Yordanov, Matteo Forasassi, Bayar Menzat, et al.
+- **abs:** https://arxiv.org/abs/2602.11852 · **PDF:** https://arxiv.org/pdf/2602.11852v1
+
+## 12894. Resource-Aware Deployment Optimization for Collaborative Intrusion Detection in Layered Networks
+
+- **ID:** `2602.11851` · **primary:** `cs.CR` · **Submitted:** 2026-02-12
+- **Authors:** André García Gómez, Ines Rieger, Wolfgang Hotwagner, et al.
+- **abs:** https://arxiv.org/abs/2602.11851 · **PDF:** https://arxiv.org/pdf/2602.11851v1
+
+## 12895. Free Lunch for Stabilizing Rectified Flow Inversion
+
+- **ID:** `2602.11850` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Chenru Wang, Beier Zhu, Chi Zhang
+- **abs:** https://arxiv.org/abs/2602.11850 · **PDF:** https://arxiv.org/pdf/2602.11850v2
+
+## 12896. Improving Neural Retrieval with Attribution-Guided Query Rewriting
+
+- **ID:** `2602.11841` · **primary:** `cs.IR` · **Submitted:** 2026-02-12
+- **Authors:** Moncef Garouani, Josiane Mothe
+- **abs:** https://arxiv.org/abs/2602.11841 · **PDF:** https://arxiv.org/pdf/2602.11841v1
+
+## 12897. ULTRA:Urdu Language Transformer-based Recommendation Architecture
+
+- **ID:** `2602.11836` · **primary:** `cs.IR` · **Submitted:** 2026-02-12
+- **Authors:** Alishbah Bashir, Fatima Qaiser, Ijaz Hussain
+- **abs:** https://arxiv.org/abs/2602.11836 · **PDF:** https://arxiv.org/pdf/2602.11836v2
+
+## 12898. EqDeepRx: Learning a Scalable MIMO Receiver
+
+- **ID:** `2602.11834` · **primary:** `eess.SP` · **Submitted:** 2026-02-12
+- **Authors:** Mikko Honkala, Dani Korpi, Elias Raninen, et al.
+- **abs:** https://arxiv.org/abs/2602.11834 · **PDF:** https://arxiv.org/pdf/2602.11834v1
+
+## 12899. Towards Sustainable Investment Policies Informed by Opponent Shaping
+
+- **ID:** `2602.11829` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Juan Agustin Duque, Razvan Ciuca, Ayoub Echchahed, et al.
+- **abs:** https://arxiv.org/abs/2602.11829 · **PDF:** https://arxiv.org/pdf/2602.11829v1
+
+## 12900. CAAL: Confidence-Aware Active Learning for Heteroscedastic Atmospheric Regression
+
+- **ID:** `2602.11825` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Fei Jiang, Jiyang Xia, Junjie Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.11825 · **PDF:** https://arxiv.org/pdf/2602.11825v1
+
+## 12901. Revis: Sparse Latent Steering to Mitigate Object Hallucination in Large Vision-Language Models
+
+- **ID:** `2602.11824` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Jialin Wu, Wei Shi, Han Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.11824 · **PDF:** https://arxiv.org/pdf/2602.11824v1
+
+## 12902. A Comparative Study of MAP and LMMSE Estimators for Blind Inverse Problems
+
+- **ID:** `2602.11814` · **primary:** `cs.IT` · **Submitted:** 2026-02-12
+- **Authors:** Nathan Buskulic, Luca Calatroni
+- **abs:** https://arxiv.org/abs/2602.11814 · **PDF:** https://arxiv.org/pdf/2602.11814v1
+
+## 12903. Predicting LLM Output Length via Entropy-Guided Representations
+
+- **ID:** `2602.11812` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Huanyi Xie, Yubin Chen, Liangyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.11812 · **PDF:** https://arxiv.org/pdf/2602.11812v2
+
+## 12904. How to Sample High Quality 3D Fractals for Action Recognition Pre-Training?
+
+- **ID:** `2602.11810` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Marko Putak, Thomas B. Moeslund, Joakim Bruslund Haurum
+- **abs:** https://arxiv.org/abs/2602.11810 · **PDF:** https://arxiv.org/pdf/2602.11810v1
+
+## 12905. Deep Kernel Fusion for Transformers
+
+- **ID:** `2602.11808` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Zixi Zhang, Zhiwen Mo, Yiren Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.11808 · **PDF:** https://arxiv.org/pdf/2602.11808v1
+
+## 12906. PuYun-LDM: A Latent Diffusion Model for High-Resolution Ensemble Weather Forecasts
+
+- **ID:** `2602.11807` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Lianjun Wu, Shengchen Zhu, Yuxuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.11807 · **PDF:** https://arxiv.org/pdf/2602.11807v2
+
+## 12907. From Path Signatures to Sequential Modeling: Incremental Signature Contributions for Offline RL
+
+- **ID:** `2602.11805` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Ziyi Zhao, Qingchuan Li, Yuxuan Xu
+- **abs:** https://arxiv.org/abs/2602.11805 · **PDF:** https://arxiv.org/pdf/2602.11805v1
+
+## 12908. TopoFair: Linking Topological Bias to Fairness in Link Prediction Benchmarks
+
+- **ID:** `2602.11802` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Lilian Marey, Mathilde Perez, Tiphaine Viard, et al.
+- **abs:** https://arxiv.org/abs/2602.11802 · **PDF:** https://arxiv.org/pdf/2602.11802v1
+
+## 12909. SpaTeoGL: Spatiotemporal Graph Learning for Interpretable Seizure Onset Zone Analysis from Intracranial EEG
+
+- **ID:** `2602.11801` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Elham Rostami, Aref Einizade, Taous-Meriem Laleg-Kirati
+- **abs:** https://arxiv.org/abs/2602.11801 · **PDF:** https://arxiv.org/pdf/2602.11801v1
+
+## 12910. Temporal Difference Learning with Constrained Initial Representations
+
+- **ID:** `2602.11800` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Jiafei Lyu, Jingwen Yang, Zhongjian Qiao, et al.
+- **abs:** https://arxiv.org/abs/2602.11800 · **PDF:** https://arxiv.org/pdf/2602.11800v1
+
+## 12911. Hi-SAM: A Hierarchical Structure-Aware Multi-modal Framework for Large-Scale Recommendation
+
+- **ID:** `2602.11799` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Pingjun Pan, Tingting Zhou, Peiyao Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.11799 · **PDF:** https://arxiv.org/pdf/2602.11799v1
+
+## 12912. A Subword Embedding Approach for Variation Detection in Luxembourgish User Comments
+
+- **ID:** `2602.11795` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Anne-Marie Lutgen, Alistair Plum, Christoph Purschke
+- **abs:** https://arxiv.org/abs/2602.11795 · **PDF:** https://arxiv.org/pdf/2602.11795v1
+
+## 12913. Latent-Variable Learning of SPDEs via Wiener Chaos
+
+- **ID:** `2602.11794` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Sebastian Zeng, Andreas Petersson, Wolfgang Bock
+- **abs:** https://arxiv.org/abs/2602.11794 · **PDF:** https://arxiv.org/pdf/2602.11794v1
+
+## 12914. More Haste, Less Speed: Weaker Single-Layer Watermark Improves Distortion-Free Watermark Ensembles
+
+- **ID:** `2602.11793` · **primary:** `cs.CR` · **Submitted:** 2026-02-12
+- **Authors:** Ruibo Chen, Yihan Wu, Xuehao Cui, et al.
+- **abs:** https://arxiv.org/abs/2602.11793 · **PDF:** https://arxiv.org/pdf/2602.11793v1
+
+## 12915. Detecting RLVR Training Data via Structural Convergence of Reasoning
+
+- **ID:** `2602.11792` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Hongbo Zhang, Yue Yang, Jianhao Yan, et al.
+- **abs:** https://arxiv.org/abs/2602.11792 · **PDF:** https://arxiv.org/pdf/2602.11792v1
+
+## 12916. Beyond End-to-End Video Models: An LLM-Based Multi-Agent System for Educational Video Generation
+
+- **ID:** `2602.11790` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Lingyong Yan, Jiulong Wu, Dong Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.11790 · **PDF:** https://arxiv.org/pdf/2602.11790v1
+
+## 12917. Decentralized Non-convex Stochastic Optimization with Heterogeneous Variance
+
+- **ID:** `2602.11789` · **primary:** `math.OC` · **Submitted:** 2026-02-12
+- **Authors:** Hongxu Chen, Ke Wei, Luo Luo
+- **abs:** https://arxiv.org/abs/2602.11789 · **PDF:** https://arxiv.org/pdf/2602.11789v1
+
+## 12918. Evaluating LLM Safety Under Repeated Inference via Accelerated Prompt Stress Testing
+
+- **ID:** `2602.11786` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Keita Broadwater
+- **abs:** https://arxiv.org/abs/2602.11786 · **PDF:** https://arxiv.org/pdf/2602.11786v1
+
+## 12919. Safe Fairness Guarantees Without Demographics in Classification: Spectral Uncertainty Set Perspective
+
+- **ID:** `2602.11785` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Ainhize Barrainkua, Santiago Mazuelas, Novi Quadrianto, et al.
+- **abs:** https://arxiv.org/abs/2602.11785 · **PDF:** https://arxiv.org/pdf/2602.11785v1
+
+## 12920. FlowMind: Execute-Summarize for Structured Workflow Generation from LLM Reasoning
+
+- **ID:** `2602.11782` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Yihao Liu, Ziyun Zhang, Zile He, et al.
+- **abs:** https://arxiv.org/abs/2602.11782 · **PDF:** https://arxiv.org/pdf/2602.11782v1
+
+## 12921. RELATE: A Reinforcement Learning-Enhanced LLM Framework for Advertising Text Generation
+
+- **ID:** `2602.11780` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Jinfang Wang, Jiajie Liu, Jianwei Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.11780 · **PDF:** https://arxiv.org/pdf/2602.11780v1
+
+## 12922. Temperature as a Meta-Policy: Adaptive Temperature in LLM Reinforcement Learning
+
+- **ID:** `2602.11779` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Haoran Dang, Cuiling Lan, Hai Wan, et al.
+- **abs:** https://arxiv.org/abs/2602.11779 · **PDF:** https://arxiv.org/pdf/2602.11779v1
+
+## 12923. MUSE: Multi-Tenant Model Serving With Seamless Model Updates
+
+- **ID:** `2602.11776` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Cláudio Correia, Alberto E. A. Ferreira, Lucas Martins, et al.
+- **abs:** https://arxiv.org/abs/2602.11776 · **PDF:** https://arxiv.org/pdf/2602.11776v2
+
+## 12924. How to Optimize Multispecies Set Predictions in Presence-Absence Modeling ?
+
+- **ID:** `2602.11771` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Sébastien Gigot--Léandri, Gaétan Morand, Alexis Joly, et al.
+- **abs:** https://arxiv.org/abs/2602.11771 · **PDF:** https://arxiv.org/pdf/2602.11771v1
+
+## 12925. TSR: Trajectory-Search Rollouts for Multi-Turn RL of LLM Agents
+
+- **ID:** `2602.11767` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Aladin Djuhera, Swanand Ravindra Kadhe, Farhan Ahmed, et al.
+- **abs:** https://arxiv.org/abs/2602.11767 · **PDF:** https://arxiv.org/pdf/2602.11767v2
+
+## 12926. MiniCPM-SALA: Hybridizing Sparse and Linear Attention for Efficient Long-Context Modeling
+
+- **ID:** `2602.11761` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** MiniCPM Team, Wenhao An, Yingfa Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.11761 · **PDF:** https://arxiv.org/pdf/2602.11761v2
+
+## 12927. Aggregate Models, Not Explanations: Improving Feature Importance Estimation
+
+- **ID:** `2602.11760` · **primary:** `stat.ML` · **Submitted:** 2026-02-12
+- **Authors:** Joseph Paillard, Angel Reyero Lobo, Denis A. Engemann, et al.
+- **abs:** https://arxiv.org/abs/2602.11760 · **PDF:** https://arxiv.org/pdf/2602.11760v1
+
+## 12928. TUBO: A Tailored ML Framework for Reliable Network Traffic Forecasting
+
+- **ID:** `2602.11759` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Zhihang Yuan, Leyang Xue, Waleed Ahsan, et al.
+- **abs:** https://arxiv.org/abs/2602.11759 · **PDF:** https://arxiv.org/pdf/2602.11759v1
+
+## 12929. Cooperation Breakdown in LLM Agents Under Communication Delays
+
+- **ID:** `2602.11754` · **primary:** `cs.MA` · **Submitted:** 2026-02-12
+- **Authors:** Keita Nishimoto, Kimitaka Asatani, Ichiro Sakata
+- **abs:** https://arxiv.org/abs/2602.11754 · **PDF:** https://arxiv.org/pdf/2602.11754v1
+
+## 12930. Curriculum Learning and Pseudo-Labeling Improve the Generalization of Multi-Label Arabic Dialect Identification Models
+
+- **ID:** `2602.12937` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Ali Mekky, Mohamed El Zeftawy, Lara Hassan, et al.
+- **abs:** https://arxiv.org/abs/2602.12937 · **PDF:** https://arxiv.org/pdf/2602.12937v2
+
+## 12931. AmbiBench: Benchmarking Mobile GUI Agents Beyond One-Shot Instructions in the Wild
+
+- **ID:** `2602.11750` · **primary:** `cs.SE` · **Submitted:** 2026-02-12
+- **Authors:** Jiazheng Sun, Mingxuan Li, Yingying Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.11750 · **PDF:** https://arxiv.org/pdf/2602.11750v1
+
+## 12932. AIR: Improving Agent Safety through Incident Response
+
+- **ID:** `2602.11749` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Zibo Xiao, Jun Sun, Junjie Chen
+- **abs:** https://arxiv.org/abs/2602.11749 · **PDF:** https://arxiv.org/pdf/2602.11749v1
+
+## 12933. Think Longer to Explore Deeper: Learn to Explore In-Context via Length-Incentivized Reinforcement Learning
+
+- **ID:** `2602.11748` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Futing Wang, Jianhao Yan, Yun Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.11748 · **PDF:** https://arxiv.org/pdf/2602.11748v1
+
+## 12934. High-Probability Minimax Adaptive Estimation in Besov Spaces via Online-to-Batch
+
+- **ID:** `2602.11747` · **primary:** `math.ST` · **Submitted:** 2026-02-12
+- **Authors:** Paul Liautaud, Pierre Gaillard, Olivier Wintenberger
+- **abs:** https://arxiv.org/abs/2602.11747 · **PDF:** https://arxiv.org/pdf/2602.11747v1
+
+## 12935. Text2GQL-Bench: A Text to Graph Query Language Benchmark [Experiment, Analysis & Benchmark]
+
+- **ID:** `2602.11745` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Songlin Lyu, Lujie Ban, Zihang Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.11745 · **PDF:** https://arxiv.org/pdf/2602.11745v1
+
+## 12936. U-Former ODE: Fast Probabilistic Forecasting of Irregular Time Series
+
+- **ID:** `2602.11738` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Ilya Kuleshov, Alexander Marusov, Alexey Zaytsev
+- **abs:** https://arxiv.org/abs/2602.11738 · **PDF:** https://arxiv.org/pdf/2602.11738v1
+
+## 12937. Mask What Matters: Mitigating Object Hallucinations in Multimodal Large Language Models with Object-Aligned Visual Contrastive Decoding
+
+- **ID:** `2602.11737` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Boqi Chen, Xudong Liu, Jianing Qiu
+- **abs:** https://arxiv.org/abs/2602.11737 · **PDF:** https://arxiv.org/pdf/2602.11737v1
+
+## 12938. Adapting Vision-Language Models for E-commerce Understanding at Scale
+
+- **ID:** `2602.11733` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Matteo Nulli, Vladimir Orshulevich, Tala Bazazo, et al.
+- **abs:** https://arxiv.org/abs/2602.11733 · **PDF:** https://arxiv.org/pdf/2602.11733v1
+
+## 12939. The Dunning-Kruger Effect in Large Language Models: An Empirical Study of Confidence Calibration
+
+- **ID:** `2603.09985` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Sudipta Ghosh, Mrityunjoy Panday
+- **abs:** https://arxiv.org/abs/2603.09985 · **PDF:** https://arxiv.org/pdf/2603.09985v1
+
+## 12940. Thinking with Drafting: Optical Decompression via Logical Reconstruction
+
+- **ID:** `2602.11731` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Jingxuan Wei, Honghao He, Caijun Jia, et al.
+- **abs:** https://arxiv.org/abs/2602.11731 · **PDF:** https://arxiv.org/pdf/2602.11731v1
+
+## 12941. Cross-Architecture Model Diffing with Crosscoders: Unsupervised Discovery of Differences Between LLMs
+
+- **ID:** `2602.11729` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Thomas Jiralerspong, Trenton Bricken
+- **abs:** https://arxiv.org/abs/2602.11729 · **PDF:** https://arxiv.org/pdf/2602.11729v1
+
+## 12942. Dopamine: Brain Modes, Not Brains
+
+- **ID:** `2602.11726` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Shervin Ghasemlou
+- **abs:** https://arxiv.org/abs/2602.11726 · **PDF:** https://arxiv.org/pdf/2602.11726v1
+
+## 12943. PAC-Bayesian Generalization Guarantees for Fairness on Stochastic and Deterministic Classifiers
+
+- **ID:** `2602.11722` · **primary:** `stat.ML` · **Submitted:** 2026-02-12
+- **Authors:** Julien Bastian, Benjamin Leblanc, Pascal Germain, et al.
+- **abs:** https://arxiv.org/abs/2602.11722 · **PDF:** https://arxiv.org/pdf/2602.11722v1
+
+## 12944. Beyond Parameter Arithmetic: Sparse Complementary Fusion for Distribution-Aware Model Merging
+
+- **ID:** `2602.11717` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Weihong Lin, Lin Sun, Qilong Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.11717 · **PDF:** https://arxiv.org/pdf/2602.11717v1
+
+## 12945. DICE: Diffusion Large Language Models Excel at Generating CUDA Kernels
+
+- **ID:** `2602.11715` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Haolei Bai, Lingcheng Kong, Xueyi Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.11715 · **PDF:** https://arxiv.org/pdf/2602.11715v1
+
+## 12946. Potential-energy gating for robust state estimation in bistable stochastic systems
+
+- **ID:** `2602.11712` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Luigi Simeone
+- **abs:** https://arxiv.org/abs/2602.11712 · **PDF:** https://arxiv.org/pdf/2602.11712v2
+
+## 12947. Estimation of instrument and noise parameters for inverse problem based on prior diffusion model
+
+- **ID:** `2602.11711` · **primary:** `stat.ML` · **Submitted:** 2026-02-12
+- **Authors:** Jean-François Giovannelli
+- **abs:** https://arxiv.org/abs/2602.11711 · **PDF:** https://arxiv.org/pdf/2602.11711v1
+
+## 12948. LLM-Driven 3D Scene Generation of Agricultural Simulation Environments
+
+- **ID:** `2602.11706` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Arafa Yoncalik, Wouter Jansen, Nico Huebel, et al.
+- **abs:** https://arxiv.org/abs/2602.11706 · **PDF:** https://arxiv.org/pdf/2602.11706v1
+
+## 12949. Semantically Conditioned Diffusion Models for Cerebral DSA Synthesis
+
+- **ID:** `2602.11703` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Qiwen Xu, David Rügamer, Holger Wenz, et al.
+- **abs:** https://arxiv.org/abs/2602.11703 · **PDF:** https://arxiv.org/pdf/2602.11703v1
+
+## 12950. TabSieve: Explicit In-Table Evidence Selection for Tabular Prediction
+
+- **ID:** `2602.11700` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Yongyao Wang, Ziqi Miao, Lu Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.11700 · **PDF:** https://arxiv.org/pdf/2602.11700v1
+
+## 12951. Finding Sense in Nonsense with Generated Contexts: Perspectives from Humans and Language Models
+
+- **ID:** `2602.11699` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Katrina Olsen, Sebastian Padó
+- **abs:** https://arxiv.org/abs/2602.11699 · **PDF:** https://arxiv.org/pdf/2602.11699v2
+
+## 12952. SpiralFormer: Looped Transformers Can Learn Hierarchical Dependencies via Multi-Resolution Recursion
+
+- **ID:** `2602.11698` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Chengting Yu, Xiaobo Shu, Yadao Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.11698 · **PDF:** https://arxiv.org/pdf/2602.11698v1
+
+## 12953. OMEGA-Avatar: One-shot Modeling of 360° Gaussian Avatars
+
+- **ID:** `2602.11693` · **primary:** `cs.GR` · **Submitted:** 2026-02-12
+- **Authors:** Zehao Xia, Yiqun Wang, Zhengda Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.11693 · **PDF:** https://arxiv.org/pdf/2602.11693v1
+
+## 12954. ANML: Attribution-Native Machine Learning with Guaranteed Robustness
+
+- **ID:** `2602.11690` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Oliver Zahn, Matt Beton, Simran Chana
+- **abs:** https://arxiv.org/abs/2602.11690 · **PDF:** https://arxiv.org/pdf/2602.11690v1
+
+## 12955. LAER-MoE: Load-Adaptive Expert Re-layout for Efficient Mixture-of-Experts Training
+
+- **ID:** `2602.11686` · **primary:** `cs.DC` · **Submitted:** 2026-02-12
+- **Authors:** Xinyi Liu, Yujie Wang, Fangcheng Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.11686 · **PDF:** https://arxiv.org/pdf/2602.11686v1
+
+## 12956. DRACO: a Cross-Domain Benchmark for Deep Research Accuracy, Completeness, and Objectivity
+
+- **ID:** `2602.11685` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Joey Zhong, Hao Zhang, Clare Southern, et al.
+- **abs:** https://arxiv.org/abs/2602.11685 · **PDF:** https://arxiv.org/pdf/2602.11685v1
+
+## 12957. PatientHub: A Unified Framework for Patient Simulation
+
+- **ID:** `2602.11684` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Sahand Sabour, TszYam NG, Minlie Huang
+- **abs:** https://arxiv.org/abs/2602.11684 · **PDF:** https://arxiv.org/pdf/2602.11684v1
+
+## 12958. ThinkRouter: Efficient Reasoning via Routing Thinking between Latent and Discrete Spaces
+
+- **ID:** `2602.11683` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Xin Xu, Tong Yu, Xiang Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.11683 · **PDF:** https://arxiv.org/pdf/2602.11683v1
+
+## 12959. Provable Offline Reinforcement Learning for Structured Cyclic MDPs
+
+- **ID:** `2602.11679` · **primary:** `stat.ML` · **Submitted:** 2026-02-12
+- **Authors:** Kyungbok Lee, Angelica Cristello Sarteau, Michael R. Kosorok
+- **abs:** https://arxiv.org/abs/2602.11679 · **PDF:** https://arxiv.org/pdf/2602.11679v1
+
+## 12960. Beyond Pixels: Vector-to-Graph Transformation for Reliable Schematic Auditing
+
+- **ID:** `2602.11678` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Chengwei Ma, Zhen Tian, Zhou Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.11678 · **PDF:** https://arxiv.org/pdf/2602.11678v1
+
+## 12961. Right for the Wrong Reasons: Epistemic Regret Minimization for Causal Rung Collapse in LLMs
+
+- **ID:** `2602.11675` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Edward Y. Chang
+- **abs:** https://arxiv.org/abs/2602.11675 · **PDF:** https://arxiv.org/pdf/2602.11675v2
+
+## 12962. Benchmark Health Index: A Systematic Framework for Benchmarking the Benchmarks of LLMs
+
+- **ID:** `2602.11674` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Longyuan Zhu, Hairan Hua, Linlin Miao, et al.
+- **abs:** https://arxiv.org/abs/2602.11674 · **PDF:** https://arxiv.org/pdf/2602.11674v1
+
+## 12963. Explainable Machine-Learning based Detection of Knee Injuries in Runners
+
+- **ID:** `2602.11668` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** David Fuentes-Jiménez, Sara García-de-Villa, David Casillas-Pérez, et al.
+- **abs:** https://arxiv.org/abs/2602.11668 · **PDF:** https://arxiv.org/pdf/2602.11668v1
+
+## 12964. PhyNiKCE: A Neurosymbolic Agentic Framework for Autonomous Computational Fluid Dynamics
+
+- **ID:** `2602.11666` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** E Fan, Lisong Shi, Zhengtong Li, et al.
+- **abs:** https://arxiv.org/abs/2602.11666 · **PDF:** https://arxiv.org/pdf/2602.11666v1
+
+## 12965. Fully First-Order Algorithms for Online Bilevel Optimization
+
+- **ID:** `2602.11665` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Tingkai Jia, Cheng Chen
+- **abs:** https://arxiv.org/abs/2602.11665 · **PDF:** https://arxiv.org/pdf/2602.11665v1
+
+## 12966. UMAP Is Spectral Clustering on the Fuzzy Nearest-Neighbor Graph
+
+- **ID:** `2602.11662` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Yang Yang
+- **abs:** https://arxiv.org/abs/2602.11662 · **PDF:** https://arxiv.org/pdf/2602.11662v1
+
+## 12967. Quark Medical Alignment: A Holistic Multi-Dimensional Alignment and Collaborative Optimization Paradigm
+
+- **ID:** `2602.11661` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Tianxiang Xu, Jiayi Liu, Yixuan Tong, et al.
+- **abs:** https://arxiv.org/abs/2602.11661 · **PDF:** https://arxiv.org/pdf/2602.11661v2
+
+## 12968. SToRM: Supervised Token Reduction for Multi-modal LLMs toward efficient end-to-end autonomous driving
+
+- **ID:** `2602.11656` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Seo Hyun Kim, Jin Bok Park, Do Yeon Koo, et al.
+- **abs:** https://arxiv.org/abs/2602.11656 · **PDF:** https://arxiv.org/pdf/2602.11656v3
+
+## 12969. LoRA-based Parameter-Efficient LLMs for Continuous Learning in Edge-based Malware Detection
+
+- **ID:** `2602.11655` · **primary:** `cs.CR` · **Submitted:** 2026-02-12
+- **Authors:** Christian Rondanini, Barbara Carminati, Elena Ferrari, et al.
+- **abs:** https://arxiv.org/abs/2602.11655 · **PDF:** https://arxiv.org/pdf/2602.11655v1
+
+## 12970. DMind-3: A Sovereign Edge--Local--Cloud AI System with Controlled Deliberation and Correction-Based Tuning for Safe, Low-Latency Transaction Execution
+
+- **ID:** `2602.11651` · **primary:** `cs.CR` · **Submitted:** 2026-02-12
+- **Authors:** Enhao Huang, Frank Li, Tony Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.11651 · **PDF:** https://arxiv.org/pdf/2602.11651v1
+
+## 12971. Which Feedback Works for Whom? Differential Effects of LLM-Generated Feedback Elements Across Learner Profiles
+
+- **ID:** `2602.11650` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Momoka Furuhashi, Kouta Nakayama, Noboru Kawai, et al.
+- **abs:** https://arxiv.org/abs/2602.11650 · **PDF:** https://arxiv.org/pdf/2602.11650v1
+
+## 12972. Brain Tumor Classifiers Under Attack: Robustness of ResNet Variants Against Transferable FGSM and PGD Attacks
+
+- **ID:** `2602.11646` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Ryan Deem, Garrett Goodman, Waqas Majeed, et al.
+- **abs:** https://arxiv.org/abs/2602.11646 · **PDF:** https://arxiv.org/pdf/2602.11646v1
+
+## 12973. ViTaS: Visual Tactile Soft Fusion Contrastive Learning for Visuomotor Learning
+
+- **ID:** `2602.11643` · **primary:** `cs.RO` · **Submitted:** 2026-02-12
+- **Authors:** Yufeng Tian, Shuiqi Cheng, Tianming Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.11643 · **PDF:** https://arxiv.org/pdf/2602.11643v1
+
+## 12974. Both Topology and Text Matter: Revisiting LLM-guided Out-of-Distribution Detection on Text-attributed Graphs
+
+- **ID:** `2602.11641` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Yinlin Zhu, Di Wu, Xu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.11641 · **PDF:** https://arxiv.org/pdf/2602.11641v1
+
+## 12975. PACE: Prefix-Protected and Difficulty-Aware Compression for Efficient Reasoning
+
+- **ID:** `2602.11639` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Ruixiang Feng, Yuntao Wen, Silin Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.11639 · **PDF:** https://arxiv.org/pdf/2602.11639v1
+
+## 12976. Variation-aware Flexible 3D Gaussian Editing
+
+- **ID:** `2602.11638` · **primary:** `cs.GR` · **Submitted:** 2026-02-12
+- **Authors:** Hao Qin, Yukai Sun, Meng Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.11638 · **PDF:** https://arxiv.org/pdf/2602.11638v3
+
+## 12977. ScalSelect: Scalable Training-Free Multimodal Data Selection for Efficient Visual Instruction Tuning
+
+- **ID:** `2602.11636` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Changti Wu, Jiahuai Mao, Yuzhuo Miao, et al.
+- **abs:** https://arxiv.org/abs/2602.11636 · **PDF:** https://arxiv.org/pdf/2602.11636v1
+
+## 12978. Do MLLMs Really Understand Space? A Mathematical Reasoning Evaluation
+
+- **ID:** `2602.11635` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Shuo Lu, Jianjie Cheng, Yinuo Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.11635 · **PDF:** https://arxiv.org/pdf/2602.11635v1
+
+## 12979. TIP: Resisting Gradient Inversion via Targeted Interpretable Perturbation in Federated Learning
+
+- **ID:** `2602.11633` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Jianhua Wang, Yinlin Su
+- **abs:** https://arxiv.org/abs/2602.11633 · **PDF:** https://arxiv.org/pdf/2602.11633v1
+
+## 12980. CL API: Real-Time Closed-Loop Interactions with Biological Neural Networks
+
+- **ID:** `2602.11632` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-12
+- **Authors:** David Hogan, Andrew Doherty, Boon Kien Khoo, et al.
+- **abs:** https://arxiv.org/abs/2602.11632 · **PDF:** https://arxiv.org/pdf/2602.11632v1
+
+## 12981. Enforcing Reciprocity in Operator Learning for Seismic Wave Propagation
+
+- **ID:** `2602.11631` · **primary:** `physics.geo-ph` · **Submitted:** 2026-02-12
+- **Authors:** Caifeng Zou, Yaozhong Shi, Zachary E. Ross, et al.
+- **abs:** https://arxiv.org/abs/2602.11631 · **PDF:** https://arxiv.org/pdf/2602.11631v1
+
+## 12982. Neuro-Symbolic Multitasking: A Unified Framework for Discovering Generalizable Solutions to PDE Families
+
+- **ID:** `2602.11630` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Yipeng Huang, Dejun Xu, Zexin Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.11630 · **PDF:** https://arxiv.org/pdf/2602.11630v1
+
+## 12983. GP2F: Cross-Domain Graph Prompting with Adaptive Fusion of Pre-trained Graph Neural Networks
+
+- **ID:** `2602.11629` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Dongxiao He, Wenxuan Sun, Yongqi Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.11629 · **PDF:** https://arxiv.org/pdf/2602.11629v1
+
+## 12984. PLESS: Pseudo-Label Enhancement with Spreading Scribbles for Weakly Supervised Segmentation
+
+- **ID:** `2602.11628` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Yeva Gabrielyan, Varduhi Yeghiazaryan, Irina Voiculescu
+- **abs:** https://arxiv.org/abs/2602.11628 · **PDF:** https://arxiv.org/pdf/2602.11628v1
+
+## 12985. ArGEnT: Arbitrary Geometry-encoded Transformer for Operator Learning
+
+- **ID:** `2602.11626` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Wenqian Chen, Yucheng Fu, Michael Penwarden, et al.
+- **abs:** https://arxiv.org/abs/2602.11626 · **PDF:** https://arxiv.org/pdf/2602.11626v1
+
+## 12986. PLOT-CT: Pre-log Voronoi Decomposition Assisted Generation for Low-dose CT Reconstruction
+
+- **ID:** `2602.11625` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Bin Huang, Xun Yu, Yikun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.11625 · **PDF:** https://arxiv.org/pdf/2602.11625v1
+
+## 12987. TreeGrad-Ranker: Feature Ranking via $O(L)$-Time Gradients for Decision Trees
+
+- **ID:** `2602.11623` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Weida Li, Yaoliang Yu, Bryan Kian Hsiang Low
+- **abs:** https://arxiv.org/abs/2602.11623 · **PDF:** https://arxiv.org/pdf/2602.11623v1
+
+## 12988. When Agents Disagree With Themselves: Measuring Behavioral Consistency in LLM-Based Agents
+
+- **ID:** `2602.11619` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Aman Mehta
+- **abs:** https://arxiv.org/abs/2602.11619 · **PDF:** https://arxiv.org/pdf/2602.11619v1
+
+## 12989. How Well Do Large-Scale Chemical Language Models Transfer to Downstream Tasks?
+
+- **ID:** `2602.11618` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Tatsuya Sagawa, Ryosuke Kojima
+- **abs:** https://arxiv.org/abs/2602.11618 · **PDF:** https://arxiv.org/pdf/2602.11618v3
+
+## 12990. An Efficient Hybrid Deep Learning Approach for Detecting Online Abusive Language
+
+- **ID:** `2603.09984` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Vuong M. Ngo, Cach N. Dang, Kien V. Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2603.09984 · **PDF:** https://arxiv.org/pdf/2603.09984v1
+
+## 12991. SkillRater: Untangling Capabilities in Multimodal Data
+
+- **ID:** `2602.11615` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Naveen Sahi, Jeremy Dohmann, Armen Aghajanyan, et al.
+- **abs:** https://arxiv.org/abs/2602.11615 · **PDF:** https://arxiv.org/pdf/2602.11615v1
+
+## 12992. scPilot: Large Language Model Reasoning Toward Automated Single-Cell Analysis and Discovery
+
+- **ID:** `2602.11609` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Yiming Gao, Zhen Wang, Jefferson Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.11609 · **PDF:** https://arxiv.org/pdf/2602.11609v1
+
+## 12993. Scene-Aware Memory Discrimination: Deciding Which Personal Knowledge Stays
+
+- **ID:** `2602.11607` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Yijie Zhong, Mengying Guo, Zewei Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.11607 · **PDF:** https://arxiv.org/pdf/2602.11607v1
+
+## 12994. Quantum walk inspired JPEG compression of images
+
+- **ID:** `2602.12306` · **primary:** `eess.IV` · **Submitted:** 2026-02-12
+- **Authors:** Abhishek Verma, Sahil Tomar, Sandeep Kumar
+- **abs:** https://arxiv.org/abs/2602.12306 · **PDF:** https://arxiv.org/pdf/2602.12306v1
+
+## 12995. ABot-N0: Technical Report on the VLA Foundation Model for Versatile Embodied Navigation
+
+- **ID:** `2602.11598` · **primary:** `cs.RO` · **Submitted:** 2026-02-12
+- **Authors:** Zedong Chu, Shichao Xie, Xiaolong Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.11598 · **PDF:** https://arxiv.org/pdf/2602.11598v1
+
+## 12996. MAPLE: Modality-Aware Post-training and Learning Ecosystem
+
+- **ID:** `2602.11596` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Nikhil Verma, Minjung Kim, JooYoung Yoo, et al.
+- **abs:** https://arxiv.org/abs/2602.11596 · **PDF:** https://arxiv.org/pdf/2602.11596v1
+
+## 12997. Learn from Your Mistakes: Self-Correcting Masked Diffusion Models
+
+- **ID:** `2602.11590` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Yair Schiff, Omer Belhasin, Roy Uziel, et al.
+- **abs:** https://arxiv.org/abs/2602.11590 · **PDF:** https://arxiv.org/pdf/2602.11590v2
+
+## 12998. Gradient Compression May Hurt Generalization: A Remedy by Synthetic Data Guided Sharpness Aware Minimization
+
+- **ID:** `2602.11584` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Yujie Gu, Richeng Jin, Zhaoyang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.11584 · **PDF:** https://arxiv.org/pdf/2602.11584v1
+
+## 12999. The Five Ws of Multi-Agent Communication: Who Talks to Whom, When, What, and Why -- A Survey from MARL to Emergent Language and LLMs
+
+- **ID:** `2602.11583` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Jingdi Chen, Hanqing Yang, Zongjun Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.11583 · **PDF:** https://arxiv.org/pdf/2602.11583v1
+
+## 13000. Analytical Search
+
+- **ID:** `2602.11581` · **primary:** `cs.IR` · **Submitted:** 2026-02-12
+- **Authors:** Yiteng Tu, Shuo Miao, Weihang Su, et al.
+- **abs:** https://arxiv.org/abs/2602.11581 · **PDF:** https://arxiv.org/pdf/2602.11581v1
+
+## 13001. OptiML: An End-to-End Framework for Program Synthesis and CUDA Kernel Optimization
+
+- **ID:** `2602.12305` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Arijit Bhattacharjee, Heng Ping, Son Vu Le, et al.
+- **abs:** https://arxiv.org/abs/2602.12305 · **PDF:** https://arxiv.org/pdf/2602.12305v1
+
+## 13002. ReaDy-Go: Real-to-Sim Dynamic 3D Gaussian Splatting Simulation for Environment-Specific Visual Navigation with Moving Obstacles
+
+- **ID:** `2602.11575` · **primary:** `cs.RO` · **Submitted:** 2026-02-12
+- **Authors:** Seungyeon Yoo, Youngseok Jang, Dabin Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.11575 · **PDF:** https://arxiv.org/pdf/2602.11575v2
+
+## 13003. Learning to Configure Agentic AI Systems
+
+- **ID:** `2602.11574` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Aditya Taparia, Som Sagar, Ransalu Senanayake
+- **abs:** https://arxiv.org/abs/2602.11574 · **PDF:** https://arxiv.org/pdf/2602.11574v1
+
+## 13004. PRIME: A Process-Outcome Alignment Benchmark for Verifiable Reasoning in Mathematics and Engineering
+
+- **ID:** `2602.11570` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Xiangfeng Wang, Hangyu Guo, Yanlin Lai, et al.
+- **abs:** https://arxiv.org/abs/2602.11570 · **PDF:** https://arxiv.org/pdf/2602.11570v1
+
+## 13005. SemaPop: Semantic-Persona Conditioned Population Synthesis
+
+- **ID:** `2602.11569` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Zhenlin Qin, Yancheng Ling, Leizhen Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.11569 · **PDF:** https://arxiv.org/pdf/2602.11569v1
+
+## 13006. MoE-SpAc: Efficient MoE Inference Based on Speculative Activation Utility in Heterogeneous Edge Scenarios
+
+- **ID:** `2603.09983` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Shuhuai Li, Jianghao Lin, Dongdong Ge, et al.
+- **abs:** https://arxiv.org/abs/2603.09983 · **PDF:** https://arxiv.org/pdf/2603.09983v1
+
+## 13007. Brain4FMs: A Benchmark of Foundation Models for Electrical Brain Signal
+
+- **ID:** `2602.11558` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Fanqi Shen, Enhong Yang, Jiahe Li, et al.
+- **abs:** https://arxiv.org/abs/2602.11558 · **PDF:** https://arxiv.org/pdf/2602.11558v1
+
+## 13008. The Implicit Bias of Steepest Descent with Mini-batch Stochastic Gradient
+
+- **ID:** `2602.11557` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Jichu Li, Xuan Tang, Difan Zou
+- **abs:** https://arxiv.org/abs/2602.11557 · **PDF:** https://arxiv.org/pdf/2602.11557v1
+
+## 13009. HyperDet: 3D Object Detection with Hyper 4D Radar Point Clouds
+
+- **ID:** `2602.11554` · **primary:** `cs.RO` · **Submitted:** 2026-02-12
+- **Authors:** Yichun Xiao, Runwei Guan, Fangqiang Ding
+- **abs:** https://arxiv.org/abs/2602.11554 · **PDF:** https://arxiv.org/pdf/2602.11554v2
+
+## 13010. Perception-based Image Denoising via Generative Compression
+
+- **ID:** `2602.11553` · **primary:** `cs.CV` · **Submitted:** 2026-02-12
+- **Authors:** Nam Nguyen, Thinh Nguyen, Bella Bose
+- **abs:** https://arxiv.org/abs/2602.11553 · **PDF:** https://arxiv.org/pdf/2602.11553v1
+
+## 13011. SIGHT: Reinforcement Learning with Self-Evidence and Information-Gain Diverse Branching for Search Agent
+
+- **ID:** `2602.11551` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Wenlin Zhong, Jinluan Yang, Yiquan Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.11551 · **PDF:** https://arxiv.org/pdf/2602.11551v1
+
+## 13012. TS-Memory: Plug-and-Play Memory for Time Series Foundation Models
+
+- **ID:** `2602.11550` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Sisuo Lyu, Siru Zhong, Tiegang Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.11550 · **PDF:** https://arxiv.org/pdf/2602.11550v1
+
+## 13013. Native Reasoning Models: Training Language Models to Reason on Unverifiable Data
+
+- **ID:** `2602.11549` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Yuanfu Wang, Zhixuan Liu, Xiangtian Li, et al.
+- **abs:** https://arxiv.org/abs/2602.11549 · **PDF:** https://arxiv.org/pdf/2602.11549v2
+
+## 13014. Pretraining A Large Language Model using Distributed GPUs: A Memory-Efficient Decentralized Paradigm
+
+- **ID:** `2602.11543` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Jinrui Zhang, Chaodong Xiao, Aoqi Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.11543 · **PDF:** https://arxiv.org/pdf/2602.11543v1
+
+## 13015. Budget-Constrained Agentic Large Language Models: Intention-Based Planning for Costly Tool Use
+
+- **ID:** `2602.11541` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Hanbing Liu, Chunhao Tian, Nan An, et al.
+- **abs:** https://arxiv.org/abs/2602.11541 · **PDF:** https://arxiv.org/pdf/2602.11541v1
+
+## 13016. Real-Time Proactive Anomaly Detection via Forward and Backward Forecast Modeling
+
+- **ID:** `2602.11539` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Luis Olmos, Rashida Hasan
+- **abs:** https://arxiv.org/abs/2602.11539 · **PDF:** https://arxiv.org/pdf/2602.11539v1
+
+## 13017. Krause Synchronization Transformers
+
+- **ID:** `2602.11534` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Jingkun Liu, Yisong Yue, Max Welling, et al.
+- **abs:** https://arxiv.org/abs/2602.11534 · **PDF:** https://arxiv.org/pdf/2602.11534v2
+
+## 13018. AltTS: A Dual-Path Framework with Alternating Optimization for Multivariate Time Series Forecasting
+
+- **ID:** `2602.11533` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Zhihang Yuan, Zhiyuan Liu, Mahesh K. Marina
+- **abs:** https://arxiv.org/abs/2602.11533 · **PDF:** https://arxiv.org/pdf/2602.11533v1
+
+## 13019. PASCAL: A Phase-Aware Scheduling Algorithm for Serving Reasoning-based Large Language Models
+
+- **ID:** `2602.11530` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Eunyeong Cho, Jehyeon Bang, Ranggi Hwang, et al.
+- **abs:** https://arxiv.org/abs/2602.11530 · **PDF:** https://arxiv.org/pdf/2602.11530v1
+
+## 13020. Stop Tracking Me! Proactive Defense Against Attribute Inference Attack in LLMs
+
+- **ID:** `2602.11528` · **primary:** `cs.CR` · **Submitted:** 2026-02-12
+- **Authors:** Dong Yan, Jian Liang, Ran He, et al.
+- **abs:** https://arxiv.org/abs/2602.11528 · **PDF:** https://arxiv.org/pdf/2602.11528v1
+
+## 13021. CausalAgent: A Conversational Multi-Agent System for End-to-End Causal Inference
+
+- **ID:** `2602.11527` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Jiawei Zhu, Wei Chen, Ruichu Cai
+- **abs:** https://arxiv.org/abs/2602.11527 · **PDF:** https://arxiv.org/pdf/2602.11527v1
+
+## 13022. Adaptive Milestone Reward for GUI Agents
+
+- **ID:** `2602.11524` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Congmin Zheng, Xiaoyun Mo, Xinbei Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.11524 · **PDF:** https://arxiv.org/pdf/2602.11524v1
+
+## 13023. Unifying Stable Optimization and Reference Regularization in RLHF
+
+- **ID:** `2602.11523` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Li He, Qiang Qu, He Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.11523 · **PDF:** https://arxiv.org/pdf/2602.11523v1
+
+## 13024. Locally Interpretable Individualized Treatment Rules for Black-Box Decision Models
+
+- **ID:** `2602.11520` · **primary:** `stat.ME` · **Submitted:** 2026-02-12
+- **Authors:** Yasin Khadem Charvadeh, Katherine S. Panageas, Yuan Chen
+- **abs:** https://arxiv.org/abs/2602.11520 · **PDF:** https://arxiv.org/pdf/2602.11520v1
+
+## 13025. OmniCustom: Sync Audio-Video Customization Via Joint Audio-Video Generation Model
+
+- **ID:** `2602.12304` · **primary:** `cs.SD` · **Submitted:** 2026-02-12
+- **Authors:** Maomao Li, Zhen Li, Kaipeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.12304 · **PDF:** https://arxiv.org/pdf/2602.12304v3
+
+## 13026. Calibration and Evaluation of Car-Following Models for Autonomous Shuttles Using a Novel Multi-Criteria Framework
+
+- **ID:** `2602.11517` · **primary:** `cs.ET` · **Submitted:** 2026-02-12
+- **Authors:** Renan Favero, Lily Elefteriadou
+- **abs:** https://arxiv.org/abs/2602.11517 · **PDF:** https://arxiv.org/pdf/2602.11517v1
+
+## 13027. Human-Inspired Continuous Learning of Internal Reasoning Processes: Learning How to Think for Adaptive AI Systems
+
+- **ID:** `2602.11516` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Hong Su
+- **abs:** https://arxiv.org/abs/2602.11516 · **PDF:** https://arxiv.org/pdf/2602.11516v1
+
+## 13028. How Smart Is Your GUI Agent? A Framework for the Future of Software Interaction
+
+- **ID:** `2602.11514` · **primary:** `cs.SE` · **Submitted:** 2026-02-12
+- **Authors:** Sidong Feng, Chunyang Chen
+- **abs:** https://arxiv.org/abs/2602.11514 · **PDF:** https://arxiv.org/pdf/2602.11514v1
+
+## 13029. Differentially Private and Communication Efficient Large Language Model Split Inference via Stochastic Quantization and Soft Prompt
+
+- **ID:** `2602.11513` · **primary:** `cs.CR` · **Submitted:** 2026-02-12
+- **Authors:** Yujie Gu, Richeng Jin, Xiaoyu Ji, et al.
+- **abs:** https://arxiv.org/abs/2602.11513 · **PDF:** https://arxiv.org/pdf/2602.11513v1
+
+## 13030. AgentLeak: A Full-Stack Benchmark for Privacy Leakage in Multi-Agent LLM Systems
+
+- **ID:** `2602.11510` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Faouzi El Yagoubi, Godwin Badu-Marfo, Ranwa Al Mallah
+- **abs:** https://arxiv.org/abs/2602.11510 · **PDF:** https://arxiv.org/pdf/2602.11510v2
+
+## 13031. Multimodal Fact-Level Attribution for Verifiable Reasoning
+
+- **ID:** `2602.11509` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** David Wan, Han Wang, Ziyang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.11509 · **PDF:** https://arxiv.org/pdf/2602.11509v1
+
+## 13032. RooflineBench: A Benchmarking Framework for On-Device LLMs via Roofline Analysis
+
+- **ID:** `2602.11506` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Zhen Bi, Xueshu Chen, Luoyang Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.11506 · **PDF:** https://arxiv.org/pdf/2602.11506v3
+
+## 13033. Calibrating an Imperfect Auxiliary Predictor for Unobserved No-Purchase Choice
+
+- **ID:** `2602.11505` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Jiangkai Xiong, Kalyan Talluri, Hanzhao Wang
+- **abs:** https://arxiv.org/abs/2602.11505 · **PDF:** https://arxiv.org/pdf/2602.11505v2
+
+## 13034. A Generic Framework for Fair Consensus Clustering in Streams
+
+- **ID:** `2602.11500` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Diptarka Chakraborty, Kushagra Chatterjee, Debarati Das, et al.
+- **abs:** https://arxiv.org/abs/2602.11500 · **PDF:** https://arxiv.org/pdf/2602.11500v1
+
+## 13035. Partial GFlowNet: Accelerating Convergence in Large State Spaces via Strategic Partitioning
+
+- **ID:** `2602.11498` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Xuan Yu, Xu Wang, Rui Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.11498 · **PDF:** https://arxiv.org/pdf/2602.11498v1
+
+## 13036. Jailbreaking Leaves a Trace: Understanding and Detecting Jailbreak Attacks from Internal Representations of Large Language Models
+
+- **ID:** `2602.11495` · **primary:** `cs.CR` · **Submitted:** 2026-02-12
+- **Authors:** Sri Durga Sai Sowmya Kadali, Evangelos E. Papalexakis
+- **abs:** https://arxiv.org/abs/2602.11495 · **PDF:** https://arxiv.org/pdf/2602.11495v2
+
+## 13037. IMAGAgent: Orchestrating Multi-Turn Image Editing via Constraint-Aware Planning and Reflection
+
+- **ID:** `2603.29602` · **primary:** `cs.GR` · **Submitted:** 2026-02-12
+- **Authors:** Fei Shen, Chengyu Xie, Lihong Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.29602 · **PDF:** https://arxiv.org/pdf/2603.29602v1
+
+## 13038. Exploring Multiple High-Scoring Subspaces in Generative Flow Networks
+
+- **ID:** `2602.11491` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Xuan Yu, Xu Wang, Rui Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.11491 · **PDF:** https://arxiv.org/pdf/2602.11491v1
+
+## 13039. When Audio-LLMs Don't Listen: A Cross-Linguistic Study of Modality Arbitration
+
+- **ID:** `2602.11488` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Jayadev Billa
+- **abs:** https://arxiv.org/abs/2602.11488 · **PDF:** https://arxiv.org/pdf/2602.11488v3
+
+## 13040. Understanding Persuasive Interactions between Generative Social Agents and Humans: The Knowledge-based Persuasion Model (KPM)
+
+- **ID:** `2602.11483` · **primary:** `cs.HC` · **Submitted:** 2026-02-12
+- **Authors:** Stephan Vonschallen, Friederike Eyssel, Theresa Schmiedel
+- **abs:** https://arxiv.org/abs/2602.11483 · **PDF:** https://arxiv.org/pdf/2602.11483v1
+
+## 13041. External Division of Two Bregman Proximity Operators for Poisson Inverse Problems
+
+- **ID:** `2602.11482` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Kazuki Haishima, Kyohei Suzuki, Konstantinos Slavakis
+- **abs:** https://arxiv.org/abs/2602.11482 · **PDF:** https://arxiv.org/pdf/2602.11482v1
+
+## 13042. Compiler-Guided Inference-Time Adaptation: Improving GPT-5 Programming Performance in Idris
+
+- **ID:** `2602.11481` · **primary:** `cs.PL` · **Submitted:** 2026-02-12
+- **Authors:** Minda Li, Bhaskar Krishnamachari
+- **abs:** https://arxiv.org/abs/2602.11481 · **PDF:** https://arxiv.org/pdf/2602.11481v1
+
+## 13043. PRISM: A 3D Probabilistic Neural Representation for Interpretable Shape Modeling
+
+- **ID:** `2602.11467` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Yining Jiao, Sreekalyani Bhamidi, Carlton Jude Zdanski, et al.
+- **abs:** https://arxiv.org/abs/2602.11467 · **PDF:** https://arxiv.org/pdf/2602.11467v1
+
+## 13044. Assessing Low Back Movement with Motion Tape Sensor Data Through Deep Learning
+
+- **ID:** `2602.11465` · **primary:** `cs.LG` · **Submitted:** 2026-02-12
+- **Authors:** Jared Levy, Aarti Lalwani, Elijah Wyckoff, et al.
+- **abs:** https://arxiv.org/abs/2602.11465 · **PDF:** https://arxiv.org/pdf/2602.11465v1
+
+## 13045. EM-Aware Physical Synthesis: Neural Inductor Modeling and Intelligent Placement & Routing for RF Circuits
+
+- **ID:** `2602.11461` · **primary:** `cs.AR` · **Submitted:** 2026-02-12
+- **Authors:** Yilun Huang, Asal Mehradfar, Salman Avestimehr, et al.
+- **abs:** https://arxiv.org/abs/2602.11461 · **PDF:** https://arxiv.org/pdf/2602.11461v1
+
+## 13046. ADRD-Bench: A Preliminary LLM Benchmark for Alzheimer's Disease and Related Dementias
+
+- **ID:** `2602.11460` · **primary:** `cs.CL` · **Submitted:** 2026-02-12
+- **Authors:** Guangxin Zhao, Jiahao Zheng, Malaz Boustani, et al.
+- **abs:** https://arxiv.org/abs/2602.11460 · **PDF:** https://arxiv.org/pdf/2602.11460v1
+
+## 13047. Credit Where It is Due: Cross-Modality Connectivity Drives Precise Reinforcement Learning for MLLM Reasoning
+
+- **ID:** `2602.11455` · **primary:** `cs.AI` · **Submitted:** 2026-02-12
+- **Authors:** Zhengbo Jiao, Shaobo Wang, Zifan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.11455 · **PDF:** https://arxiv.org/pdf/2602.11455v1
+
+## 13048. Adaptive Power Iteration Method for Differentially Private PCA
+
+- **ID:** `2602.11454` · **primary:** `cs.DS` · **Submitted:** 2026-02-12
+- **Authors:** Ta Duy Nguyen, Alina Ene, Huy Le Nguyen
+- **abs:** https://arxiv.org/abs/2602.11454 · **PDF:** https://arxiv.org/pdf/2602.11454v2
+
+## 13049. From Noise to Order: Learning to Rank via Denoising Diffusion
+
+- **ID:** `2602.11453` · **primary:** `cs.IR` · **Submitted:** 2026-02-12
+- **Authors:** Sajad Ebrahimi, Bhaskar Mitra, Negar Arabzadeh, et al.
+- **abs:** https://arxiv.org/abs/2602.11453 · **PDF:** https://arxiv.org/pdf/2602.11453v1
+
+## 13050. LoopFormer: Elastic-Depth Looped Transformers for Latent Reasoning via Shortcut Modulation
+
+- **ID:** `2602.11451` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Ahmadreza Jeddi, Marco Ciccone, Babak Taati
+- **abs:** https://arxiv.org/abs/2602.11451 · **PDF:** https://arxiv.org/pdf/2602.11451v1
+
+## 13051. Hierarchical Concept Embedding & Pursuit for Interpretable Image Classification
+
+- **ID:** `2602.11448` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Nghia Nguyen, Tianjiao Ding, René Vidal
+- **abs:** https://arxiv.org/abs/2602.11448 · **PDF:** https://arxiv.org/pdf/2602.11448v3
+
+## 13052. Enhanced Portable Ultra Low-Field Diffusion Tensor Imaging with Bayesian Artifact Correction and Deep Learning-Based Super-Resolution
+
+- **ID:** `2602.11446` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Mark D. Olchanyi, Annabel Sorby-Adams, John Kirsch, et al.
+- **abs:** https://arxiv.org/abs/2602.11446 · **PDF:** https://arxiv.org/pdf/2602.11446v1
+
+## 13053. Towards Reliable Machine Translation: Scaling LLMs for Critical Error Detection and Safety
+
+- **ID:** `2602.11444` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Muskaan Chopra, Lorenz Sparrenberg, Rafet Sifa
+- **abs:** https://arxiv.org/abs/2602.11444 · **PDF:** https://arxiv.org/pdf/2602.11444v1
+
+## 13054. Multi-Level Strategic Classification: Incentivizing Improvement through Promotion and Relegation Dynamics
+
+- **ID:** `2602.11439` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Ziyuan Huang, Lina Alkarmi, Mingyan Liu
+- **abs:** https://arxiv.org/abs/2602.11439 · **PDF:** https://arxiv.org/pdf/2602.11439v1
+
+## 13055. Distributionally Robust Cooperative Multi-Agent Reinforcement Learning via Robust Value Factorization
+
+- **ID:** `2602.11437` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Chengrui Qu, Christopher Yeh, Kishan Panaganti, et al.
+- **abs:** https://arxiv.org/abs/2602.11437 · **PDF:** https://arxiv.org/pdf/2602.11437v1
+
+## 13056. Fighting MRI Anisotropy: Learning Multiple Cardiac Shapes From a Single Implicit Neural Representation
+
+- **ID:** `2602.11436` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Carolina Brás, Soufiane Ben Haddou, Thijs P. Kuipers, et al.
+- **abs:** https://arxiv.org/abs/2602.11436 · **PDF:** https://arxiv.org/pdf/2602.11436v1
+
+## 13057. Surface impedance inference via neural fields and sparse acoustic data obtained by a compact array
+
+- **ID:** `2602.11425` · **primary:** `cs.SD` · **Submitted:** 2026-02-11
+- **Authors:** Yuanxin Xia, Xinyan Li, Matteo Calafà, et al.
+- **abs:** https://arxiv.org/abs/2602.11425 · **PDF:** https://arxiv.org/pdf/2602.11425v1
+
+## 13058. Gradients Must Earn Their Influence: Unifying SFT with Generalized Entropic Objectives
+
+- **ID:** `2602.11424` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Zecheng Wang, Deyuan Liu, Chunshan Li, et al.
+- **abs:** https://arxiv.org/abs/2602.11424 · **PDF:** https://arxiv.org/pdf/2602.11424v1
+
+## 13059. Optimizing Agent Planning for Security and Autonomy
+
+- **ID:** `2602.11416` · **primary:** `cs.CR` · **Submitted:** 2026-02-11
+- **Authors:** Aashish Kolluri, Rishi Sharma, Manuel Costa, et al.
+- **abs:** https://arxiv.org/abs/2602.11416 · **PDF:** https://arxiv.org/pdf/2602.11416v1
+
+## 13060. Neural Paging: Learning Context Management Policies for Turing-Complete Agents
+
+- **ID:** `2603.02228` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Liang Chen, Qi Liu
+- **abs:** https://arxiv.org/abs/2603.02228 · **PDF:** https://arxiv.org/pdf/2603.02228v1
+
+## 13061. TimeSynth: A Framework for Uncovering Systematic Biases in Time Series Forecasting
+
+- **ID:** `2602.11413` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Md Rakibul Haque, Vishwa Goudar, Shireen Elhabian, et al.
+- **abs:** https://arxiv.org/abs/2602.11413 · **PDF:** https://arxiv.org/pdf/2602.11413v1
+
+## 13062. When Visibility Outpaces Verification: Delayed Verification and Narrative Lock-in in Agentic AI Discourse
+
+- **ID:** `2602.11412` · **primary:** `cs.CY` · **Submitted:** 2026-02-11
+- **Authors:** Hanjing Shi, Dominic DiFranzo
+- **abs:** https://arxiv.org/abs/2602.11412 · **PDF:** https://arxiv.org/pdf/2602.11412v1
+
+## 13063. CADET: Context-Conditioned Ads CTR Prediction With a Decoder-Only Transformer
+
+- **ID:** `2602.11410` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** David Pardoe, Neil Daftary, Miro Furtado, et al.
+- **abs:** https://arxiv.org/abs/2602.11410 · **PDF:** https://arxiv.org/pdf/2602.11410v1
+
+## 13064. TRACER: Trajectory Risk Aggregation for Critical Episodes in Agentic Reasoning
+
+- **ID:** `2602.11409` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Sina Tayebati, Divake Kumar, Nastaran Darabi, et al.
+- **abs:** https://arxiv.org/abs/2602.11409 · **PDF:** https://arxiv.org/pdf/2602.11409v1
+
+## 13065. GHOST: Unmasking Phantom States in Mamba2 via Grouped Hidden-state Output-aware Selection & Truncation
+
+- **ID:** `2602.11408` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Michael Menezes, Anastasios Kyrillidis
+- **abs:** https://arxiv.org/abs/2602.11408 · **PDF:** https://arxiv.org/pdf/2602.11408v1
+
+## 13066. The Cost of Learning under Multiple Change Points
+
+- **ID:** `2602.11406` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Tomer Gafni, Garud Iyengar, Assaf Zeevi
+- **abs:** https://arxiv.org/abs/2602.11406 · **PDF:** https://arxiv.org/pdf/2602.11406v1
+
+## 13067. Latent Forcing: Reordering the Diffusion Trajectory for Pixel-Space Image Generation
+
+- **ID:** `2602.11401` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Alan Baade, Eric Ryan Chan, Kyle Sargent, et al.
+- **abs:** https://arxiv.org/abs/2602.11401 · **PDF:** https://arxiv.org/pdf/2602.11401v1
+
+## 13068. Can We Really Learn One Representation to Optimize All Rewards?
+
+- **ID:** `2602.11399` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Chongyi Zheng, Royina Karegoudra Jayanth, Benjamin Eysenbach
+- **abs:** https://arxiv.org/abs/2602.11399 · **PDF:** https://arxiv.org/pdf/2602.11399v1
+
+## 13069. Evolution With Purpose: Hierarchy-Informed Optimization of Whole-Brain Models
+
+- **ID:** `2602.11398` · **primary:** `cs.NE` · **Submitted:** 2026-02-11
+- **Authors:** Hormoz Shahrzad, Niharika Gajawelli, Kaitlin Maile, et al.
+- **abs:** https://arxiv.org/abs/2602.11398 · **PDF:** https://arxiv.org/pdf/2602.11398v3
+
+## 13070. General and Efficient Steering of Unconditional Diffusion
+
+- **ID:** `2602.11395` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Qingsong Wang, Mikhail Belkin, Yusu Wang
+- **abs:** https://arxiv.org/abs/2602.11395 · **PDF:** https://arxiv.org/pdf/2602.11395v1
+
+## 13071. Advancing AI Trustworthiness Through Patient Simulation: Risk Assessment of Conversational Agents for Antidepressant Selection
+
+- **ID:** `2602.11391` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Md Tanvir Rouf Shawon, Mohammad Sabik Irbaz, Hadeel R. A. Elyazori, et al.
+- **abs:** https://arxiv.org/abs/2602.11391 · **PDF:** https://arxiv.org/pdf/2602.11391v3
+
+## 13072. Causal-JEPA: Learning World Models through Object-Level Latent Interventions
+
+- **ID:** `2602.11389` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Heejeong Nam, Quentin Le Lidec, Lucas Maes, et al.
+- **abs:** https://arxiv.org/abs/2602.11389 · **PDF:** https://arxiv.org/pdf/2602.11389v1
+
+## 13073. Sparse Semantic Dimension as a Generalization Certificate for LLMs
+
+- **ID:** `2602.11388` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Dibyanayan Bandyopadhyay, Asif Ekbal
+- **abs:** https://arxiv.org/abs/2602.11388 · **PDF:** https://arxiv.org/pdf/2602.11388v1
+
+## 13074. Provably Efficient Algorithms for S- and Non-Rectangular Robust MDPs with General Parameterization
+
+- **ID:** `2602.11387` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Anirudh Satheesh, Ziyi Chen, Furong Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.11387 · **PDF:** https://arxiv.org/pdf/2602.11387v1
+
+## 13075. WSBD: Freezing-Based Optimizer for Quantum Neural Networks
+
+- **ID:** `2602.11383` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Christopher Kverne, Mayur Akewar, Yuqian Huo, et al.
+- **abs:** https://arxiv.org/abs/2602.11383 · **PDF:** https://arxiv.org/pdf/2602.11383v1
+
+## 13076. Toward Adaptive Non-Intrusive Reduced-Order Models: Design and Challenges
+
+- **ID:** `2602.11378` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Amirpasha Hedayat, Alberto Padovan, Karthik Duraisamy
+- **abs:** https://arxiv.org/abs/2602.11378 · **PDF:** https://arxiv.org/pdf/2602.11378v1
+
+## 13077. Retrieval-Aware Distillation for Transformer-SSM Hybrids
+
+- **ID:** `2602.11374` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Aviv Bick, Eric P. Xing, Albert Gu
+- **abs:** https://arxiv.org/abs/2602.11374 · **PDF:** https://arxiv.org/pdf/2602.11374v1
+
+## 13078. The Manifold of the Absolute: Religious Perennialism as Generative Inference
+
+- **ID:** `2602.11368` · **primary:** `cs.CY` · **Submitted:** 2026-02-11
+- **Authors:** Arthur Juliani
+- **abs:** https://arxiv.org/abs/2602.11368 · **PDF:** https://arxiv.org/pdf/2602.11368v3
+
+## 13079. The Energy of Falsehood: Detecting Hallucinations via Diffusion Model Likelihoods
+
+- **ID:** `2602.11364` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Arpit Singh Gautam, Kailash Talreja, Saurabh Jha
+- **abs:** https://arxiv.org/abs/2602.11364 · **PDF:** https://arxiv.org/pdf/2602.11364v1
+
+## 13080. Finding the Cracks: Improving LLMs Reasoning with Paraphrastic Probing and Consistency Verification
+
+- **ID:** `2602.11361` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Weili Shi, Dongliang Guo, Lehan Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.11361 · **PDF:** https://arxiv.org/pdf/2602.11361v1
+
+## 13081. Bootstrapping-based Regularisation for Reducing Individual Prediction Instability in Clinical Risk Prediction Models
+
+- **ID:** `2602.11360` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Sara Matijevic, Christopher Yau
+- **abs:** https://arxiv.org/abs/2602.11360 · **PDF:** https://arxiv.org/pdf/2602.11360v1
+
+## 13082. When Models Examine Themselves: Vocabulary-Activation Correspondence in Self-Referential Processing
+
+- **ID:** `2602.11358` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Zachary Pedram Dadfar
+- **abs:** https://arxiv.org/abs/2602.11358 · **PDF:** https://arxiv.org/pdf/2602.11358v2
+
+## 13083. ReplicatorBench: Benchmarking LLM Agents for Replicability in Social and Behavioral Sciences
+
+- **ID:** `2602.11354` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Bang Nguyen, Dominik Soós, Qian Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.11354 · **PDF:** https://arxiv.org/pdf/2602.11354v1
+
+## 13084. Pushing Forward Pareto Frontiers of Proactive Agents with Behavioral Agentic Optimization
+
+- **ID:** `2602.11351` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Yihang Yao, Zhepeng Cen, Haohong Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.11351 · **PDF:** https://arxiv.org/pdf/2602.11351v1
+
+## 13085. Structured Hybrid Mechanistic Models for Robust Estimation of Time-Dependent Intervention Outcomes
+
+- **ID:** `2602.11350` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Tomer Meir, Ori Linial, Danny Eytan, et al.
+- **abs:** https://arxiv.org/abs/2602.11350 · **PDF:** https://arxiv.org/pdf/2602.11350v1
+
+## 13086. AgentNoiseBench: Benchmarking Robustness of Tool-Using LLM Agents Under Noisy Condition
+
+- **ID:** `2602.11348` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Ruipeng Wang, Yuxin Chen, Yukai Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.11348 · **PDF:** https://arxiv.org/pdf/2602.11348v2
+
+## 13087. Divide and Learn: Multi-Objective Combinatorial Optimization at Scale
+
+- **ID:** `2602.11346` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Esha Singh, Dongxia Wu, Chien-Yi Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.11346 · **PDF:** https://arxiv.org/pdf/2602.11346v1
+
+## 13088. Situated, Dynamic, and Subjective: Envisioning the Design of Theory-of-Mind-Enabled Everyday AI with Industry Practitioners
+
+- **ID:** `2602.11342` · **primary:** `cs.HC` · **Submitted:** 2026-02-11
+- **Authors:** Qiaosi Wang, Jini Kim, Avanita Sharma, et al.
+- **abs:** https://arxiv.org/abs/2602.11342 · **PDF:** https://arxiv.org/pdf/2602.11342v1
+
+## 13089. Bi-Level Prompt Optimization for Multimodal LLM-as-a-Judge
+
+- **ID:** `2602.11340` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Bo Pan, Xuan Kan, Kaitai Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.11340 · **PDF:** https://arxiv.org/pdf/2602.11340v1
+
+## 13090. MolmoSpaces: A Large-Scale Open Ecosystem for Robot Navigation and Manipulation
+
+- **ID:** `2602.11337` · **primary:** `cs.RO` · **Submitted:** 2026-02-11
+- **Authors:** Yejin Kim, Wilbert Pumacay, Omar Rayyan, et al.
+- **abs:** https://arxiv.org/abs/2602.11337 · **PDF:** https://arxiv.org/pdf/2602.11337v2
+
+## 13091. Traffic Flow Reconstruction from Limited Collected Data
+
+- **ID:** `2602.11336` · **primary:** `math.DS` · **Submitted:** 2026-02-11
+- **Authors:** Nail Baloul, Amaury Hayat, Thibault Liard, et al.
+- **abs:** https://arxiv.org/abs/2602.11336 · **PDF:** https://arxiv.org/pdf/2602.11336v1
+
+## 13092. Cross-Fitting-Free Debiased Machine Learning with Multiway Dependence
+
+- **ID:** `2602.11333` · **primary:** `econ.EM` · **Submitted:** 2026-02-11
+- **Authors:** Kaicheng Chen, Harold D. Chiang
+- **abs:** https://arxiv.org/abs/2602.11333 · **PDF:** https://arxiv.org/pdf/2602.11333v1
+
+## 13093. Sample-Free Safety Assessment of Neural Network Controllers via Taylor Methods
+
+- **ID:** `2602.11332` · **primary:** `eess.SY` · **Submitted:** 2026-02-11
+- **Authors:** Adam Evans, Roberto Armellin
+- **abs:** https://arxiv.org/abs/2602.11332 · **PDF:** https://arxiv.org/pdf/2602.11332v1
+
+## 13094. Evaluating Alignment of Behavioral Dispositions in LLMs
+
+- **ID:** `2602.11328` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Amir Taubenfeld, Zorik Gekhman, Lior Nezry, et al.
+- **abs:** https://arxiv.org/abs/2602.11328 · **PDF:** https://arxiv.org/pdf/2602.11328v1
+
+## 13095. Security Threat Modeling for Emerging AI-Agent Protocols: A Comparative Analysis of MCP, A2A, Agora, and ANP
+
+- **ID:** `2602.11327` · **primary:** `cs.CR` · **Submitted:** 2026-02-11
+- **Authors:** Zeynab Anbiaee, Mahdi Rabbani, Mansur Mirani, et al.
+- **abs:** https://arxiv.org/abs/2602.11327 · **PDF:** https://arxiv.org/pdf/2602.11327v1
+
+## 13096. Amortised and provably-robust simulation-based inference
+
+- **ID:** `2602.11325` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Ayush Bharti, Charita Dellaporta, Yuga Hikida, et al.
+- **abs:** https://arxiv.org/abs/2602.11325 · **PDF:** https://arxiv.org/pdf/2602.11325v2
+
+## 13097. Predictive Associative Memory: Retrieval Beyond Similarity Through Temporal Co-occurrence
+
+- **ID:** `2602.11322` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jason Dury
+- **abs:** https://arxiv.org/abs/2602.11322 · **PDF:** https://arxiv.org/pdf/2602.11322v1
+
+## 13098. Efficient Analysis of the Distilled Neural Tangent Kernel
+
+- **ID:** `2602.11320` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jamie Mahowald, Brian Bell, Alex Ho, et al.
+- **abs:** https://arxiv.org/abs/2602.11320 · **PDF:** https://arxiv.org/pdf/2602.11320v2
+
+## 13099. The Consensus Trap: Dissecting Subjectivity and the "Ground Truth" Illusion in Data Annotation
+
+- **ID:** `2602.11318` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Sheza Munir, Benjamin Mah, Krisha Kalsi, et al.
+- **abs:** https://arxiv.org/abs/2602.11318 · **PDF:** https://arxiv.org/pdf/2602.11318v2
+
+## 13100. Hierarchical Testing of a Hybrid Machine Learning-Physics Global Atmosphere Model
+
+- **ID:** `2602.11313` · **primary:** `physics.ao-ph` · **Submitted:** 2026-02-11
+- **Authors:** Ziming Chen, L. Ruby Leung, Wenyu Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.11313 · **PDF:** https://arxiv.org/pdf/2602.11313v1
+
+## 13101. Are Aligned Large Language Models Still Misaligned?
+
+- **ID:** `2602.11305` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Usman Naseem, Gautam Siddharth Kashyap, Rafiq Ali, et al.
+- **abs:** https://arxiv.org/abs/2602.11305 · **PDF:** https://arxiv.org/pdf/2602.11305v1
+
+## 13102. CryptoAnalystBench: Failures in Multi-Tool Long-Form LLM Analysis
+
+- **ID:** `2602.11304` · **primary:** `cs.IR` · **Submitted:** 2026-02-11
+- **Authors:** Anushri Eswaran, Oleg Golev, Darshan Tank, et al.
+- **abs:** https://arxiv.org/abs/2602.11304 · **PDF:** https://arxiv.org/pdf/2602.11304v1
+
+## 13103. The PBSAI Governance Ecosystem: A Multi-Agent AI Reference Architecture for Securing Enterprise AI Estates
+
+- **ID:** `2602.11301` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** John M. Willis
+- **abs:** https://arxiv.org/abs/2602.11301 · **PDF:** https://arxiv.org/pdf/2602.11301v1
+
+## 13104. Voxtral Realtime
+
+- **ID:** `2602.11298` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Alexander H. Liu, Andy Ehrenberg, Andy Lo, et al.
+- **abs:** https://arxiv.org/abs/2602.11298 · **PDF:** https://arxiv.org/pdf/2602.11298v2
+
+## 13105. On Decision-Valued Maps and Representational Dependence
+
+- **ID:** `2602.11295` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Gil Raitses
+- **abs:** https://arxiv.org/abs/2602.11295 · **PDF:** https://arxiv.org/pdf/2602.11295v1
+
+## 13106. HiFloat4 Format for Language Model Inference
+
+- **ID:** `2602.11287` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Yuanyong Luo, Jing Huang, Yu Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.11287 · **PDF:** https://arxiv.org/pdf/2602.11287v2
+
+## 13107. DeepRed: an architecture for redshift estimation
+
+- **ID:** `2602.11281` · **primary:** `astro-ph.IM` · **Submitted:** 2026-02-11
+- **Authors:** Alessandro Meroni, Nicolò Oreste Pinciroli Vago, Piero Fraternali
+- **abs:** https://arxiv.org/abs/2602.11281 · **PDF:** https://arxiv.org/pdf/2602.11281v1
+
+## 13108. Unlearnable phases of matter
+
+- **ID:** `2602.11262` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-02-11
+- **Authors:** Tarun Advaith Kumar, Yijian Zou, Amir-Reza Negari, et al.
+- **abs:** https://arxiv.org/abs/2602.11262 · **PDF:** https://arxiv.org/pdf/2602.11262v2
+
+## 13109. Diffusion-Pretrained Dense and Contextual Embeddings
+
+- **ID:** `2602.11151` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Sedigheh Eslami, Maksim Gaiduk, Markus Krimmel, et al.
+- **abs:** https://arxiv.org/abs/2602.11151 · **PDF:** https://arxiv.org/pdf/2602.11151v2
+
+## 13110. YOR: Your Own Mobile Manipulator for Generalizable Robotics
+
+- **ID:** `2602.11150` · **primary:** `cs.RO` · **Submitted:** 2026-02-11
+- **Authors:** Manan H Anjaria, Mehmet Enes Erciyes, Vedant Ghatnekar, et al.
+- **abs:** https://arxiv.org/abs/2602.11150 · **PDF:** https://arxiv.org/pdf/2602.11150v1
+
+## 13111. Data Repetition Beats Data Scaling in Long-CoT Supervised Fine-Tuning
+
+- **ID:** `2602.11149` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Dawid J. Kopiczko, Sagar Vaze, Tijmen Blankevoort, et al.
+- **abs:** https://arxiv.org/abs/2602.11149 · **PDF:** https://arxiv.org/pdf/2602.11149v1
+
+## 13112. Beyond VLM-Based Rewards: Diffusion-Native Latent Reward Modeling
+
+- **ID:** `2602.11146` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Gongye Liu, Bo Yang, Yida Zhi, et al.
+- **abs:** https://arxiv.org/abs/2602.11146 · **PDF:** https://arxiv.org/pdf/2602.11146v1
+
+## 13113. SCRAPL: Scattering Transform with Random Paths for Machine Learning
+
+- **ID:** `2602.11145` · **primary:** `cs.SD` · **Submitted:** 2026-02-11
+- **Authors:** Christopher Mitcheltree, Vincent Lostanlen, Emmanouil Benetos, et al.
+- **abs:** https://arxiv.org/abs/2602.11145 · **PDF:** https://arxiv.org/pdf/2602.11145v1
+
+## 13114. GENIUS: Generative Fluid Intelligence Evaluation Suite
+
+- **ID:** `2602.11144` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Ruichuan An, Sihan Yang, Ziyu Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.11144 · **PDF:** https://arxiv.org/pdf/2602.11144v1
+
+## 13115. Data-Efficient Hierarchical Goal-Conditioned Reinforcement Learning via Normalizing Flows
+
+- **ID:** `2602.11142` · **primary:** `cs.RO` · **Submitted:** 2026-02-11
+- **Authors:** Shaswat Garg, Matin Moezzi, Brandon Da Silva
+- **abs:** https://arxiv.org/abs/2602.11142 · **PDF:** https://arxiv.org/pdf/2602.11142v1
+
+## 13116. MultiCube-RAG for Multi-hop Question Answering
+
+- **ID:** `2602.15898` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Jimeng Shi, Wei Hu, Runchu Tian, et al.
+- **abs:** https://arxiv.org/abs/2602.15898 · **PDF:** https://arxiv.org/pdf/2602.15898v1
+
+## 13117. LCIP: Loss-Controlled Inverse Projection of High-Dimensional Image Data
+
+- **ID:** `2602.11141` · **primary:** `cs.HC` · **Submitted:** 2026-02-11
+- **Authors:** Yu Wang, Frederik L. Dennig, Michael Behrisch, et al.
+- **abs:** https://arxiv.org/abs/2602.11141 · **PDF:** https://arxiv.org/pdf/2602.11141v1
+
+## 13118. TabICLv2: A better, faster, scalable, and open tabular foundation model
+
+- **ID:** `2602.11139` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jingang Qu, David Holzmüller, Gaël Varoquaux, et al.
+- **abs:** https://arxiv.org/abs/2602.11139 · **PDF:** https://arxiv.org/pdf/2602.11139v1
+
+## 13119. Weight Decay Improves Language Model Plasticity
+
+- **ID:** `2602.11137` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Tessa Han, Sebastian Bordt, Hanlin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.11137 · **PDF:** https://arxiv.org/pdf/2602.11137v1
+
+## 13120. FormalJudge: A Neuro-Symbolic Paradigm for Agentic Oversight
+
+- **ID:** `2602.11136` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Jiayi Zhou, Yang Sheng, Hantao Lou, et al.
+- **abs:** https://arxiv.org/abs/2602.11136 · **PDF:** https://arxiv.org/pdf/2602.11136v2
+
+## 13121. Just on Time: Token-Level Early Stopping for Diffusion Language Models
+
+- **ID:** `2602.11133` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Zahar Kohut, Severyn Shykula, Dmytro Khamula, et al.
+- **abs:** https://arxiv.org/abs/2602.11133 · **PDF:** https://arxiv.org/pdf/2602.11133v1
+
+## 13122. From Circuits to Dynamics: Understanding and Stabilizing Failure in 3D Diffusion Transformers
+
+- **ID:** `2602.11130` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Maximilian Plattner, Fabian Paischer, Johannes Brandstetter, et al.
+- **abs:** https://arxiv.org/abs/2602.11130 · **PDF:** https://arxiv.org/pdf/2602.11130v1
+
+## 13123. Asymmetric Prompt Weighting for Reinforcement Learning with Verifiable Rewards
+
+- **ID:** `2602.11128` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Reinhard Heckel, Mahdi Soltanolkotabi, Christos Thramboulidis
+- **abs:** https://arxiv.org/abs/2602.11128 · **PDF:** https://arxiv.org/pdf/2602.11128v1
+
+## 13124. The Offline-Frontier Shift: Diagnosing Distributional Limits in Generative Multi-Objective Optimization
+
+- **ID:** `2602.11126` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Stephanie Holly, Alexandru-Ciprian Zăvoianu, Siegfried Silber, et al.
+- **abs:** https://arxiv.org/abs/2602.11126 · **PDF:** https://arxiv.org/pdf/2602.11126v1
+
+## 13125. From Natural Language to Materials Discovery:The Materials Knowledge Navigation Agent
+
+- **ID:** `2602.11123` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Genmao Zhuang, Amir Barati Farimani
+- **abs:** https://arxiv.org/abs/2602.11123 · **PDF:** https://arxiv.org/pdf/2602.11123v1
+
+## 13126. A Doubly Robust Machine Learning Approach for Disentangling Treatment Effect Heterogeneity with Functional Outcomes
+
+- **ID:** `2602.11118` · **primary:** `stat.ME` · **Submitted:** 2026-02-11
+- **Authors:** Filippo Salmaso, Lorenzo Testa, Francesca Chiaromonte
+- **abs:** https://arxiv.org/abs/2602.11118 · **PDF:** https://arxiv.org/pdf/2602.11118v1
+
+## 13127. Learning to Compose for Cross-domain Agentic Workflow Generation
+
+- **ID:** `2602.11114` · **primary:** `cs.MA` · **Submitted:** 2026-02-11
+- **Authors:** Jialiang Wang, Shengxiang Xu, Hanmo Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.11114 · **PDF:** https://arxiv.org/pdf/2602.11114v1
+
+## 13128. Renet: Principled and Efficient Relaxation for the Elastic Net via Dynamic Objective Selection
+
+- **ID:** `2602.11107` · **primary:** `stat.ME` · **Submitted:** 2026-02-11
+- **Authors:** Albert Dorador
+- **abs:** https://arxiv.org/abs/2602.11107 · **PDF:** https://arxiv.org/pdf/2602.11107v1
+
+## 13129. TEGRA: Text Encoding With Graph and Retrieval Augmentation for Misinformation Detection
+
+- **ID:** `2602.11106` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Géraud Faye, Wassila Ouerdane, Guillaume Gadek, et al.
+- **abs:** https://arxiv.org/abs/2602.11106 · **PDF:** https://arxiv.org/pdf/2602.11106v2
+
+## 13130. Prompt-Dependent Ranking of Large Language Models with Uncertainty Quantification
+
+- **ID:** `2603.03336` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Angel Rodrigo Avelar Menendez, Yufeng Liu, Xiaowu Dai
+- **abs:** https://arxiv.org/abs/2603.03336 · **PDF:** https://arxiv.org/pdf/2603.03336v1
+
+## 13131. GameDevBench: Evaluating Agentic Capabilities Through Game Development
+
+- **ID:** `2602.11103` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Wayne Chi, Yixiong Fang, Arnav Yayavaram, et al.
+- **abs:** https://arxiv.org/abs/2602.11103 · **PDF:** https://arxiv.org/pdf/2602.11103v1
+
+## 13132. Statistical Learning Analysis of Physics-Informed Neural Networks
+
+- **ID:** `2602.11097` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** David A. Barajas-Solano
+- **abs:** https://arxiv.org/abs/2602.11097 · **PDF:** https://arxiv.org/pdf/2602.11097v1
+
+## 13133. Safety Recovery in Reasoning Models Is Only a Few Early Steering Steps Away
+
+- **ID:** `2602.11096` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Soumya Suvra Ghosal, Souradip Chakraborty, Vaibhav Singh, et al.
+- **abs:** https://arxiv.org/abs/2602.11096 · **PDF:** https://arxiv.org/pdf/2602.11096v1
+
+## 13134. MerLin: A Discovery Engine for Photonic and Hybrid Quantum Machine Learning
+
+- **ID:** `2602.11092` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Cassandre Notton, Benjamin Stott, Philippe Schoeb, et al.
+- **abs:** https://arxiv.org/abs/2602.11092 · **PDF:** https://arxiv.org/pdf/2602.11092v1
+
+## 13135. Can Large Language Models Make Everyone Happy?
+
+- **ID:** `2602.11091` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Usman Naseem, Gautam Siddharth Kashyap, Ebad Shabbir, et al.
+- **abs:** https://arxiv.org/abs/2602.11091 · **PDF:** https://arxiv.org/pdf/2602.11091v1
+
+## 13136. Direct Learning of Calibration-Aware Uncertainty for Neural PDE Surrogates
+
+- **ID:** `2602.11090` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Carlos Stein Brito
+- **abs:** https://arxiv.org/abs/2602.11090 · **PDF:** https://arxiv.org/pdf/2602.11090v1
+
+## 13137. DataChef: Cooking Up Optimal Data Recipes for LLM Adaptation via Reinforcement Learning
+
+- **ID:** `2602.11089` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Yicheng Chen, Zerun Ma, Xinchen Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.11089 · **PDF:** https://arxiv.org/pdf/2602.11089v2
+
+## 13138. General Flexible $f$-divergence for Challenging Offline RL Datasets with Low Stochasticity and Diverse Behavior Policies
+
+- **ID:** `2602.11087` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jianxun Wang, Grant C. Forbes, Leonardo Villalobos-Arias, et al.
+- **abs:** https://arxiv.org/abs/2602.11087 · **PDF:** https://arxiv.org/pdf/2602.11087v1
+
+## 13139. First International StepUP Competition for Biometric Footstep Recognition: Methods, Results and Remaining Challenges
+
+- **ID:** `2602.11086` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Robyn Larracy, Eve MacDonald, Angkoon Phinyomark, et al.
+- **abs:** https://arxiv.org/abs/2602.11086 · **PDF:** https://arxiv.org/pdf/2602.11086v1
+
+## 13140. Peak + Accumulation: A Proxy-Level Scoring Formula for Multi-Turn LLM Attack Detection
+
+- **ID:** `2602.11247` · **primary:** `cs.CR` · **Submitted:** 2026-02-11
+- **Authors:** J Alex Corll
+- **abs:** https://arxiv.org/abs/2602.11247 · **PDF:** https://arxiv.org/pdf/2602.11247v2
+
+## 13141. GRASP: group-Shapley feature selection for patients
+
+- **ID:** `2602.11084` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Yuheng Luo, Shuyan Li, Zhong Cao
+- **abs:** https://arxiv.org/abs/2602.11084 · **PDF:** https://arxiv.org/pdf/2602.11084v1
+
+## 13142. How Many Features Can a Language Model Store Under the Linear Representation Hypothesis?
+
+- **ID:** `2602.11246` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Nikhil Garg, Jon Kleinberg, Kenny Peng
+- **abs:** https://arxiv.org/abs/2602.11246 · **PDF:** https://arxiv.org/pdf/2602.11246v1
+
+## 13143. Designing Explainable AI for Healthcare Reviews: Guidance on Adoption and Trust
+
+- **ID:** `2603.00072` · **primary:** `cs.CY` · **Submitted:** 2026-02-11
+- **Authors:** Eman Alamoudi, Ellis Solaiman
+- **abs:** https://arxiv.org/abs/2603.00072 · **PDF:** https://arxiv.org/pdf/2603.00072v1
+
+## 13144. Token-Efficient Change Detection in LLM APIs
+
+- **ID:** `2602.11083` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Timothée Chauvin, Clément Lalanne, Erwan Le Merrer, et al.
+- **abs:** https://arxiv.org/abs/2602.11083 · **PDF:** https://arxiv.org/pdf/2602.11083v1
+
+## 13145. SteuerLLM: Local specialized large language model for German tax law analysis
+
+- **ID:** `2602.11081` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Sebastian Wind, Jeta Sopa, Laurin Schmid, et al.
+- **abs:** https://arxiv.org/abs/2602.11081 · **PDF:** https://arxiv.org/pdf/2602.11081v1
+
+## 13146. In-the-Wild Model Organisms: Mitigating Undesirable Emergent Behaviors in Production LLM Post-Training via Data Attribution
+
+- **ID:** `2602.11079` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Frank Xiao, Santiago Aranguri
+- **abs:** https://arxiv.org/abs/2602.11079 · **PDF:** https://arxiv.org/pdf/2602.11079v2
+
+## 13147. Interpretable Attention-Based Multi-Agent PPO for Latency Spike Resolution in 6G RAN Slicing
+
+- **ID:** `2602.11076` · **primary:** `eess.SY` · **Submitted:** 2026-02-11
+- **Authors:** Kavan Fatehi, Mostafa Rahmani Ghourtani, Amir Sonee, et al.
+- **abs:** https://arxiv.org/abs/2602.11076 · **PDF:** https://arxiv.org/pdf/2602.11076v1
+
+## 13148. Chatting with Images for Introspective Visual Thinking
+
+- **ID:** `2602.11073` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Junfei Wu, Jian Guan, Qiang Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.11073 · **PDF:** https://arxiv.org/pdf/2602.11073v2
+
+## 13149. Simultaneous Speech-to-Speech Translation Without Aligned Data
+
+- **ID:** `2602.11072` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Tom Labiausse, Romain Fabre, Yannick Estève, et al.
+- **abs:** https://arxiv.org/abs/2602.11072 · **PDF:** https://arxiv.org/pdf/2602.11072v1
+
+## 13150. DCInject: Persistent Backdoor Attacks via Frequency Manipulation in Personal Federated Learning
+
+- **ID:** `2602.18489` · **primary:** `cs.CR` · **Submitted:** 2026-02-11
+- **Authors:** Nahom Birhan, Daniel Wesego, Dereje Shenkut, et al.
+- **abs:** https://arxiv.org/abs/2602.18489 · **PDF:** https://arxiv.org/pdf/2602.18489v1
+
+## 13151. Conversational Behavior Modeling Foundation Model With Multi-Level Perception
+
+- **ID:** `2602.11065` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Dingkun Zhou, Shuchang Pan, Jiachen Lian, et al.
+- **abs:** https://arxiv.org/abs/2602.11065 · **PDF:** https://arxiv.org/pdf/2602.11065v1
+
+## 13152. Evaluating Memory Structure in LLM Agents
+
+- **ID:** `2602.11243` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Alina Shutova, Alexandra Olenina, Ivan Vinogradov, et al.
+- **abs:** https://arxiv.org/abs/2602.11243 · **PDF:** https://arxiv.org/pdf/2602.11243v1
+
+## 13153. Motion Capture is Not the Target Domain: Scaling Synthetic Data for Learning Motion Representations
+
+- **ID:** `2602.11064` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Firas Darwish, George Nicholson, Aiden Doherty, et al.
+- **abs:** https://arxiv.org/abs/2602.11064 · **PDF:** https://arxiv.org/pdf/2602.11064v1
+
+## 13154. MoToRec: Sparse-Regularized Multimodal Tokenization for Cold-Start Recommendation
+
+- **ID:** `2602.11062` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jialin Liu, Zhaorui Zhang, Ray C. C. Cheung
+- **abs:** https://arxiv.org/abs/2602.11062 · **PDF:** https://arxiv.org/pdf/2602.11062v2
+
+## 13155. Grandes Modelos de Linguagem Multimodais (MLLMs): Da Teoria à Prática
+
+- **ID:** `2602.12302` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Neemias da Silva, Júlio C. W. Scholz, John Harrison, et al.
+- **abs:** https://arxiv.org/abs/2602.12302 · **PDF:** https://arxiv.org/pdf/2602.12302v1
+
+## 13156. Active Zero: Self-Evolving Vision-Language Models through Active Environment Exploration
+
+- **ID:** `2602.11241` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Jinghan He, Junfeng Fang, Feng Xiong, et al.
+- **abs:** https://arxiv.org/abs/2602.11241 · **PDF:** https://arxiv.org/pdf/2602.11241v1
+
+## 13157. A Gibbs posterior sampler for inverse problem based on prior diffusion model
+
+- **ID:** `2602.11059` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Jean-François Giovannelli
+- **abs:** https://arxiv.org/abs/2602.11059 · **PDF:** https://arxiv.org/pdf/2602.11059v1
+
+## 13158. Divide, Harmonize, Then Conquer It: Shooting Multi-Commodity Flow Problems with Multimodal Language Models
+
+- **ID:** `2602.11057` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Xinyu Yuan, Yan Qiao, Zonghui Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.11057 · **PDF:** https://arxiv.org/pdf/2602.11057v1
+
+## 13159. Toward Reliable Tea Leaf Disease Diagnosis Using Deep Learning Model: Enhancing Robustness With Explainable AI and Adversarial Training
+
+- **ID:** `2602.11239` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Samanta Ghosh, Jannatul Adan Mahi, Shayan Abrar, et al.
+- **abs:** https://arxiv.org/abs/2602.11239 · **PDF:** https://arxiv.org/pdf/2602.11239v1
+
+## 13160. GraphSeek: Next-Generation Graph Analytics with LLMs
+
+- **ID:** `2602.11052` · **primary:** `cs.DB` · **Submitted:** 2026-02-11
+- **Authors:** Maciej Besta, Łukasz Jarmocik, Orest Hrycyna, et al.
+- **abs:** https://arxiv.org/abs/2602.11052 · **PDF:** https://arxiv.org/pdf/2602.11052v2
+
+## 13161. Embedding Inversion via Conditional Masked Diffusion Language Models
+
+- **ID:** `2602.11047` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Han Xiao
+- **abs:** https://arxiv.org/abs/2602.11047 · **PDF:** https://arxiv.org/pdf/2602.11047v3
+
+## 13162. SurveyLens: A Research Discipline-Aware Benchmark for Automatic Survey Generation
+
+- **ID:** `2602.11238` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Beichen Guo, Zhiyuan Wen, Jia Gu, et al.
+- **abs:** https://arxiv.org/abs/2602.11238 · **PDF:** https://arxiv.org/pdf/2602.11238v1
+
+## 13163. Language Model Inversion through End-to-End Differentiation
+
+- **ID:** `2602.11044` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Kevin Yandoka Denamganaï, Kartic Subr
+- **abs:** https://arxiv.org/abs/2602.11044 · **PDF:** https://arxiv.org/pdf/2602.11044v1
+
+## 13164. Characterizing Trainability of Instantaneous Quantum Polynomial Circuit Born Machines
+
+- **ID:** `2602.11042` · **primary:** `quant-ph` · **Submitted:** 2026-02-11
+- **Authors:** Kevin Shen, Susanne Pielawa, Vedran Dunjko, et al.
+- **abs:** https://arxiv.org/abs/2602.11042 · **PDF:** https://arxiv.org/pdf/2602.11042v2
+
+## 13165. Learning Page Order in Shuffled WOO Releases
+
+- **ID:** `2602.11040` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Efe Kahraman, Giulio Tosato
+- **abs:** https://arxiv.org/abs/2602.11040 · **PDF:** https://arxiv.org/pdf/2602.11040v2
+
+## 13166. AI-Driven Clinical Decision Support System for Enhanced Diabetes Diagnosis and Management
+
+- **ID:** `2602.11237` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Mujeeb Ur Rehman, Imran Rehan, Sohail Khalid
+- **abs:** https://arxiv.org/abs/2602.11237 · **PDF:** https://arxiv.org/pdf/2602.11237v1
+
+## 13167. Beyond Context Sharing: A Unified Agent Communication Protocol (ACP) for Secure, Federated, and Autonomous Agent-to-Agent (A2A) Orchestration
+
+- **ID:** `2602.15055` · **primary:** `cs.MA` · **Submitted:** 2026-02-11
+- **Authors:** Naveen Kumar Krishnan
+- **abs:** https://arxiv.org/abs/2602.15055 · **PDF:** https://arxiv.org/pdf/2602.15055v1
+
+## 13168. Linguistic Indicators of Early Cognitive Decline in the DementiaBank Pitt Corpus: A Statistical and Machine Learning Study
+
+- **ID:** `2602.11028` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Artsvik Avetisyan, Sachin Kumar
+- **abs:** https://arxiv.org/abs/2602.11028 · **PDF:** https://arxiv.org/pdf/2602.11028v1
+
+## 13169. Chain-of-Look Spatial Reasoning for Dense Surgical Instrument Counting
+
+- **ID:** `2602.11024` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Rishikesh Bhyri, Brian R Quaranto, Philip J Seger, et al.
+- **abs:** https://arxiv.org/abs/2602.11024 · **PDF:** https://arxiv.org/pdf/2602.11024v1
+
+## 13170. ContactGaussian-WM: Learning Physics-Grounded World Model from Videos
+
+- **ID:** `2602.11021` · **primary:** `cs.RO` · **Submitted:** 2026-02-11
+- **Authors:** Meizhong Wang, Wanxin Jin, Kun Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.11021 · **PDF:** https://arxiv.org/pdf/2602.11021v1
+
+## 13171. ABot-M0: VLA Foundation Model for Robotic Manipulation with Action Manifold Learning
+
+- **ID:** `2602.11236` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Yandan Yang, Shuang Zeng, Tong Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.11236 · **PDF:** https://arxiv.org/pdf/2602.11236v1
+
+## 13172. When Fusion Helps and When It Breaks: View-Aligned Robustness in Same-Source Financial Imaging
+
+- **ID:** `2602.11020` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Rui Ma
+- **abs:** https://arxiv.org/abs/2602.11020 · **PDF:** https://arxiv.org/pdf/2602.11020v2
+
+## 13173. OSIL: Learning Offline Safe Imitation Policies with Safety Inferred from Non-preferred Trajectories
+
+- **ID:** `2602.11018` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Returaj Burnwal, Nirav Pravinbhai Bhatt, Balaraman Ravindran
+- **abs:** https://arxiv.org/abs/2602.11018 · **PDF:** https://arxiv.org/pdf/2602.11018v1
+
+## 13174. From Buffers to Registers: Unlocking Fine-Grained FlashAttention with Hybrid-Bonded 3D NPU Co-Design
+
+- **ID:** `2602.11016` · **primary:** `cs.AR` · **Submitted:** 2026-02-11
+- **Authors:** Jinxin Yu, Yudong Pan, Mengdi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.11016 · **PDF:** https://arxiv.org/pdf/2602.11016v1
+
+## 13175. CVPL: A Geometric Framework for Post-Hoc Linkage Risk Assessment in Protected Tabular Data
+
+- **ID:** `2602.11015` · **primary:** `cs.CR` · **Submitted:** 2026-02-11
+- **Authors:** Valery Khvatov, Alexey Neyman
+- **abs:** https://arxiv.org/abs/2602.11015 · **PDF:** https://arxiv.org/pdf/2602.11015v1
+
+## 13176. ROCKET: Rapid Optimization via Calibration-guided Knapsack Enhanced Truncation for Efficient Model Compression
+
+- **ID:** `2602.11008` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Ammar Ali, Baher Mohammad, Denis Makhov, et al.
+- **abs:** https://arxiv.org/abs/2602.11008 · **PDF:** https://arxiv.org/pdf/2602.11008v1
+
+## 13177. Learning Glioblastoma Tumor Heterogeneity Using Brain Inspired Topological Neural Networks
+
+- **ID:** `2602.11234` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Ankita Paul, Wenyi Wang
+- **abs:** https://arxiv.org/abs/2602.11234 · **PDF:** https://arxiv.org/pdf/2602.11234v1
+
+## 13178. Enhancing Predictability of Multi-Tenant DNN Inference for Autonomous Vehicles' Perception
+
+- **ID:** `2602.11004` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Liangkai Liu, Kang G. Shin, Jinkyu Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.11004 · **PDF:** https://arxiv.org/pdf/2602.11004v1
+
+## 13179. Fine-Tuning GPT-5 for GPU Kernel Generation
+
+- **ID:** `2602.11000` · **primary:** `cs.DC` · **Submitted:** 2026-02-11
+- **Authors:** Ali Tehrani, Yahya Emara, Essam Wissam, et al.
+- **abs:** https://arxiv.org/abs/2602.11000 · **PDF:** https://arxiv.org/pdf/2602.11000v1
+
+## 13180. CLI-Gym: Scalable CLI Task Generation via Agentic Environment Inversion
+
+- **ID:** `2602.10999` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Yusong Lin, Haiyang Wang, Shuzhe Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.10999 · **PDF:** https://arxiv.org/pdf/2602.10999v1
+
+## 13181. The emergence of numerical representations in communicating artificial agents
+
+- **ID:** `2602.10996` · **primary:** `cs.MA` · **Submitted:** 2026-02-11
+- **Authors:** Daniela Mihai, Lucas Weber, Francesca Franzon
+- **abs:** https://arxiv.org/abs/2602.10996 · **PDF:** https://arxiv.org/pdf/2602.10996v1
+
+## 13182. LoRA-Squeeze: Simple and Effective Post-Tuning and In-Tuning Compression of LoRA Modules
+
+- **ID:** `2602.10993` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Ivan Vulić, Adam Grycner, Quentin de Laroussilhe, et al.
+- **abs:** https://arxiv.org/abs/2602.10993 · **PDF:** https://arxiv.org/pdf/2602.10993v2
+
+## 13183. Variational Optimality of Föllmer Processes in Generative Diffusions
+
+- **ID:** `2602.10989` · **primary:** `math.ST` · **Submitted:** 2026-02-11
+- **Authors:** Yifan Chen, Eric Vanden-Eijnden
+- **abs:** https://arxiv.org/abs/2602.10989 · **PDF:** https://arxiv.org/pdf/2602.10989v1
+
+## 13184. TVCACHE: A Stateful Tool-Value Cache for Post-Training LLM Agents
+
+- **ID:** `2602.10986` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Abhishek Vijaya Kumar, Bhaskar Kataria, Byungsoo Oh, et al.
+- **abs:** https://arxiv.org/abs/2602.10986 · **PDF:** https://arxiv.org/pdf/2602.10986v1
+
+## 13185. Sample Efficient Generative Molecular Optimization with Joint Self-Improvement
+
+- **ID:** `2602.10984` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Serra Korkmaz, Adam Izdebski, Jonathan Pirnay, et al.
+- **abs:** https://arxiv.org/abs/2602.10984 · **PDF:** https://arxiv.org/pdf/2602.10984v1
+
+## 13186. RiemannGL: Riemannian Geometry Changes Graph Deep Learning
+
+- **ID:** `2602.10982` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Li Sun, Qiqi Wan, Suyang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.10982 · **PDF:** https://arxiv.org/pdf/2602.10982v1
+
+## 13187. Compressed Sensing for Capability Localization in Large Language Models
+
+- **ID:** `2603.03335` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Anna Bair, Yixuan Even Xu, Mingjie Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.03335 · **PDF:** https://arxiv.org/pdf/2603.03335v1
+
+## 13188. FeatureBench: Benchmarking Agentic Coding for Complex Feature Development
+
+- **ID:** `2602.10975` · **primary:** `cs.SE` · **Submitted:** 2026-02-11
+- **Authors:** Qixing Zhou, Jiacheng Zhang, Haiyang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.10975 · **PDF:** https://arxiv.org/pdf/2602.10975v1
+
+## 13189. A Jointly Efficient and Optimal Algorithm for Heteroskedastic Generalized Linear Bandits with Adversarial Corruptions
+
+- **ID:** `2602.10971` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Sanghwa Kim, Junghyun Lee, Se-Young Yun
+- **abs:** https://arxiv.org/abs/2602.10971 · **PDF:** https://arxiv.org/pdf/2602.10971v1
+
+## 13190. Weighting-Based Identification and Estimation in Graphical Models of Missing Data
+
+- **ID:** `2602.10969` · **primary:** `stat.ME` · **Submitted:** 2026-02-11
+- **Authors:** Anna Guo, Razieh Nabi
+- **abs:** https://arxiv.org/abs/2602.10969 · **PDF:** https://arxiv.org/pdf/2602.10969v1
+
+## 13191. Healthy Harvests: A Comparative Look at Guava Disease Classification Using InceptionV3
+
+- **ID:** `2602.10967` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Samanta Ghosh, Shaila Afroz Anika, Umma Habiba Ahmed, et al.
+- **abs:** https://arxiv.org/abs/2602.10967 · **PDF:** https://arxiv.org/pdf/2602.10967v1
+
+## 13192. MoEEdit: Efficient and Routing-Stable Knowledge Editing for Mixture-of-Experts LLMs
+
+- **ID:** `2602.10965` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Yupu Gu, Rongzhe Wei, Andy Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.10965 · **PDF:** https://arxiv.org/pdf/2602.10965v1
+
+## 13193. Can LLMs Cook Jamaican Couscous? A Study of Cultural Novelty in Recipe Generation
+
+- **ID:** `2602.10964` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** F. Carichon, R. Rampa, G. Farnadi
+- **abs:** https://arxiv.org/abs/2602.10964 · **PDF:** https://arxiv.org/pdf/2602.10964v1
+
+## 13194. Rotary Positional Embeddings as Phase Modulation: Theoretical Bounds on the RoPE Base for Long-Context Transformers
+
+- **ID:** `2602.10959` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Feilong Liu
+- **abs:** https://arxiv.org/abs/2602.10959 · **PDF:** https://arxiv.org/pdf/2602.10959v1
+
+## 13195. Stochastic Parroting in Temporal Attention -- Regulating the Diagonal Sink
+
+- **ID:** `2602.10956` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Victoria Hankemeier, Malte Schilling
+- **abs:** https://arxiv.org/abs/2602.10956 · **PDF:** https://arxiv.org/pdf/2602.10956v3
+
+## 13196. Search or Accelerate: Confidence-Switched Position Beam Search for Diffusion Language Models
+
+- **ID:** `2602.10953` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Mingyu Cao, Alvaro H. C. Correia, Christos Louizos, et al.
+- **abs:** https://arxiv.org/abs/2602.10953 · **PDF:** https://arxiv.org/pdf/2602.10953v2
+
+## 13197. Optimal Initialization in Depth: Lyapunov Initialization and Limit Theorems for Deep Leaky ReLU Networks
+
+- **ID:** `2602.10949` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Constantin Kogler, Tassilo Schwarz, Samuel Kittle
+- **abs:** https://arxiv.org/abs/2602.10949 · **PDF:** https://arxiv.org/pdf/2602.10949v1
+
+## 13198. Latent Generative Solvers for Generalizable Long-Term Physics Simulation
+
+- **ID:** `2602.11229` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Zituo Chen, Haixu Wu, Sili Deng
+- **abs:** https://arxiv.org/abs/2602.11229 · **PDF:** https://arxiv.org/pdf/2602.11229v1
+
+## 13199. Computational Phenomenology of Temporal Experience in Autism: Quantifying the Emotional and Narrative Characteristics of Lived Unpredictability
+
+- **ID:** `2602.10947` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Kacper Dudzic, Karolina Drożdż, Maciej Wodziński, et al.
+- **abs:** https://arxiv.org/abs/2602.10947 · **PDF:** https://arxiv.org/pdf/2602.10947v2
+
+## 13200. What do people want to fact-check?
+
+- **ID:** `2602.10935` · **primary:** `cs.HC` · **Submitted:** 2026-02-11
+- **Authors:** Bijean Ghafouri, Dorsaf Sallami, Luca Luceri, et al.
+- **abs:** https://arxiv.org/abs/2602.10935 · **PDF:** https://arxiv.org/pdf/2602.10935v1
+
+## 13201. CMAD: Cooperative Multi-Agent Diffusion via Stochastic Optimal Control
+
+- **ID:** `2602.10933` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Riccardo Barbano, Alexander Denker, Zeljko Kereta, et al.
+- **abs:** https://arxiv.org/abs/2602.10933 · **PDF:** https://arxiv.org/pdf/2602.10933v1
+
+## 13202. Routing Absorption in Sparse Attention: Why Random Gates Are Hard to Beat
+
+- **ID:** `2603.02227` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Keston Aquino-Michaels
+- **abs:** https://arxiv.org/abs/2603.02227 · **PDF:** https://arxiv.org/pdf/2603.02227v1
+
+## 13203. Spatial-Morphological Modeling for Multi-Attribute Imputation of Urban Blocks
+
+- **ID:** `2602.10923` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Vasilii Starikov, Ruslan Kozliak, Georgii Kontsevik, et al.
+- **abs:** https://arxiv.org/abs/2602.10923 · **PDF:** https://arxiv.org/pdf/2602.10923v1
+
+## 13204. Near-Constant Strong Violation and Last-Iterate Convergence for Online CMDPs via Decaying Safety Margins
+
+- **ID:** `2602.10917` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Qian Zuo, Zhiyong Wang, Fengxiang He
+- **abs:** https://arxiv.org/abs/2602.10917 · **PDF:** https://arxiv.org/pdf/2602.10917v2
+
+## 13205. Traceable, Enforceable, and Compensable Participation: A Participation Ledger for People-Centered AI Governance
+
+- **ID:** `2602.10916` · **primary:** `cs.CY` · **Submitted:** 2026-02-11
+- **Authors:** Rashid Mushkani
+- **abs:** https://arxiv.org/abs/2602.10916 · **PDF:** https://arxiv.org/pdf/2602.10916v1
+
+## 13206. Blind Gods and Broken Screens: Architecting a Secure, Intent-Centric Mobile Agent Operating System
+
+- **ID:** `2602.10915` · **primary:** `cs.CR` · **Submitted:** 2026-02-11
+- **Authors:** Zhenhua Zou, Sheng Guo, Qiuyang Zhan, et al.
+- **abs:** https://arxiv.org/abs/2602.10915 · **PDF:** https://arxiv.org/pdf/2602.10915v3
+
+## 13207. Efficient Sparse Selective-Update RNNs for Long-Range Sequence Modeling
+
+- **ID:** `2603.02226` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Bojian Yin, Shurong Wang, Haoyu Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.02226 · **PDF:** https://arxiv.org/pdf/2603.02226v1
+
+## 13208. Mitigating Gradient Inversion Risks in Language Models via Token Obfuscation
+
+- **ID:** `2602.15897` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Xinguo Feng, Zhongkui Ma, Zihan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.15897 · **PDF:** https://arxiv.org/pdf/2602.15897v1
+
+## 13209. Tuning the burn-in phase in training recurrent neural networks improves their performance
+
+- **ID:** `2602.10911` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Julian D. Schiller, Malte Heinrich, Victor G. Lopez, et al.
+- **abs:** https://arxiv.org/abs/2602.10911 · **PDF:** https://arxiv.org/pdf/2602.10911v1
+
+## 13210. SoftMatcha 2: A Fast and Soft Pattern Matcher for Trillion-Scale Corpora
+
+- **ID:** `2602.10908` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Masataka Yoneda, Yusuke Matsushita, Go Kamoda, et al.
+- **abs:** https://arxiv.org/abs/2602.10908 · **PDF:** https://arxiv.org/pdf/2602.10908v1
+
+## 13211. Ask the Expert: Collaborative Inference for Vision Transformers with Near-Edge Accelerators
+
+- **ID:** `2602.13334` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Hao Liu, Suhaib A. Fahmy
+- **abs:** https://arxiv.org/abs/2602.13334 · **PDF:** https://arxiv.org/pdf/2602.13334v1
+
+## 13212. Natural Hypergradient Descent: Algorithm Design, Convergence Analysis, and Parallel Implementation
+
+- **ID:** `2602.10905` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Deyi Kong, Zaiwei Chen, Shuzhong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.10905 · **PDF:** https://arxiv.org/pdf/2602.10905v2
+
+## 13213. Resource-Efficient Model-Free Reinforcement Learning for Board Games
+
+- **ID:** `2602.10894` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Kazuki Ota, Takayuki Osa, Motoki Omura, et al.
+- **abs:** https://arxiv.org/abs/2602.10894 · **PDF:** https://arxiv.org/pdf/2602.10894v1
+
+## 13214. Interactive LLM-assisted Curriculum Learning for Multi-Task Evolutionary Policy Search
+
+- **ID:** `2602.10891` · **primary:** `cs.NE` · **Submitted:** 2026-02-11
+- **Authors:** Berfin Sakallioglu, Giorgia Nadizar, Eric Medvet
+- **abs:** https://arxiv.org/abs/2602.10891 · **PDF:** https://arxiv.org/pdf/2602.10891v1
+
+## 13215. Anomaly Detection with Machine Learning Algorithms in Large-Scale Power Grids
+
+- **ID:** `2602.10888` · **primary:** `eess.SY` · **Submitted:** 2026-02-11
+- **Authors:** Marc Gillioz, Guillaume Dubuis, Étienne Voutaz, et al.
+- **abs:** https://arxiv.org/abs/2602.10888 · **PDF:** https://arxiv.org/pdf/2602.10888v1
+
+## 13216. The CLEF-2026 FinMMEval Lab: Multilingual and Multimodal Evaluation of Financial AI Systems
+
+- **ID:** `2602.10886` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Zhuohan Xie, Rania Elbadry, Fan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.10886 · **PDF:** https://arxiv.org/pdf/2602.10886v1
+
+## 13217. Reinforcing Chain-of-Thought Reasoning with Self-Evolving Rubrics
+
+- **ID:** `2602.10885` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Leheng Sheng, Wenchang Ma, Ruixin Hong, et al.
+- **abs:** https://arxiv.org/abs/2602.10885 · **PDF:** https://arxiv.org/pdf/2602.10885v1
+
+## 13218. Diagnosing Structural Failures in LLM-Based Evidence Extraction for Meta-Analysis
+
+- **ID:** `2602.10881` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Zhiyin Tan, Jennifer D'Souza
+- **abs:** https://arxiv.org/abs/2602.10881 · **PDF:** https://arxiv.org/pdf/2602.10881v1
+
+## 13219. C-MOP: Integrating Momentum and Boundary-Aware Clustering for Enhanced Prompt Evolution
+
+- **ID:** `2602.10874` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Binwei Yan, Yifei Fu, Mingjian Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.10874 · **PDF:** https://arxiv.org/pdf/2602.10874v1
+
+## 13220. FedPS: Federated data Preprocessing via aggregated Statistics
+
+- **ID:** `2602.10870` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Xuefeng Xu, Graham Cormode
+- **abs:** https://arxiv.org/abs/2602.10870 · **PDF:** https://arxiv.org/pdf/2602.10870v1
+
+## 13221. Generative AI-Driven Phase Control for RIS-Aided Cell-Free Massive MIMO Systems
+
+- **ID:** `2602.11226` · **primary:** `cs.IT` · **Submitted:** 2026-02-11
+- **Authors:** Kalpesh K. Patel, Malay Chakraborty, Ekant Sharma, et al.
+- **abs:** https://arxiv.org/abs/2602.11226 · **PDF:** https://arxiv.org/pdf/2602.11226v1
+
+## 13222. The Sample Complexity of Uniform Approximation for Multi-Dimensional CDFs and Fixed-Price Mechanisms
+
+- **ID:** `2602.10868` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Matteo Castiglioni, Anna Lunghi, Alberto Marchesi
+- **abs:** https://arxiv.org/abs/2602.10868 · **PDF:** https://arxiv.org/pdf/2602.10868v1
+
+## 13223. Deep Learning of Compositional Targets with Hierarchical Spectral Methods
+
+- **ID:** `2602.10867` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Hugo Tabanelli, Yatin Dandi, Luca Pesce, et al.
+- **abs:** https://arxiv.org/abs/2602.10867 · **PDF:** https://arxiv.org/pdf/2602.10867v1
+
+## 13224. ICA: Information-Aware Credit Assignment for Visually Grounded Long-Horizon Information-Seeking Agents
+
+- **ID:** `2602.10863` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Cong Pang, Xuyu Feng, Yujie Yi, et al.
+- **abs:** https://arxiv.org/abs/2602.10863 · **PDF:** https://arxiv.org/pdf/2602.10863v1
+
+## 13225. Automated Model Design using Gated Neuron Selection in Telecom
+
+- **ID:** `2602.10854` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Adam Orucu, Marcus Medhage, Farnaz Moradi, et al.
+- **abs:** https://arxiv.org/abs/2602.10854 · **PDF:** https://arxiv.org/pdf/2602.10854v1
+
+## 13226. Time Series Foundation Models for Energy Load Forecasting on Consumer Hardware: A Multi-Dimensional Zero-Shot Benchmark
+
+- **ID:** `2602.10848` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Luigi Simeone
+- **abs:** https://arxiv.org/abs/2602.10848 · **PDF:** https://arxiv.org/pdf/2602.10848v1
+
+## 13227. Enhancing Multivariate Time Series Forecasting with Global Temporal Retrieval
+
+- **ID:** `2602.10847` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Fanpu Cao, Lu Dai, Jindong Han, et al.
+- **abs:** https://arxiv.org/abs/2602.10847 · **PDF:** https://arxiv.org/pdf/2602.10847v1
+
+## 13228. Every Little Helps: Building Knowledge Graph Foundation Model with Fine-grained Transferable Multi-modal Tokens
+
+- **ID:** `2602.15896` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Yichi Zhang, Zhuo Chen, Lingbing Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.15896 · **PDF:** https://arxiv.org/pdf/2602.15896v1
+
+## 13229. SynergyKGC: Reconciling Topological Heterogeneity in Knowledge Graph Completion via Topology-Aware Synergy
+
+- **ID:** `2602.10845` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Xuecheng Zou, Yu Tang, Bingbing Wang
+- **abs:** https://arxiv.org/abs/2602.10845 · **PDF:** https://arxiv.org/pdf/2602.10845v1
+
+## 13230. Agent-Diff: Benchmarking LLM Agents on Enterprise API Tasks via Code Execution with State-Diff-Based Evaluation
+
+- **ID:** `2602.11224` · **primary:** `cs.SE` · **Submitted:** 2026-02-11
+- **Authors:** Hubert M. Pysklo, Artem Zhuravel, Patrick D. Watson
+- **abs:** https://arxiv.org/abs/2602.11224 · **PDF:** https://arxiv.org/pdf/2602.11224v2
+
+## 13231. SimuScene: Training and Benchmarking Code Generation to Simulate Physical Scenarios
+
+- **ID:** `2602.10840` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Yanan Wang, Renxi Wang, Yongxin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.10840 · **PDF:** https://arxiv.org/pdf/2602.10840v1
+
+## 13232. Training-Induced Bias Toward LLM-Generated Content in Dense Retrieval
+
+- **ID:** `2602.10833` · **primary:** `cs.IR` · **Submitted:** 2026-02-11
+- **Authors:** William Xion, Wolfgang Nejdl
+- **abs:** https://arxiv.org/abs/2602.10833 · **PDF:** https://arxiv.org/pdf/2602.10833v1
+
+## 13233. I can tell whether you are a Native Hawlêri Speaker! How ANN, CNN, and RNN perform in NLI-Native Language Identification
+
+- **ID:** `2602.10832` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Hardi Garari, Hossein Hassani
+- **abs:** https://arxiv.org/abs/2602.10832 · **PDF:** https://arxiv.org/pdf/2602.10832v1
+
+## 13234. Self-Supervised Learning for Speaker Recognition: A study and review
+
+- **ID:** `2602.10829` · **primary:** `eess.AS` · **Submitted:** 2026-02-11
+- **Authors:** Theo Lepage, Reda Dehak
+- **abs:** https://arxiv.org/abs/2602.10829 · **PDF:** https://arxiv.org/pdf/2602.10829v1
+
+## 13235. Flow caching for autoregressive video generation
+
+- **ID:** `2602.10825` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Yuexiao Ma, Xuzhe Zheng, Jing Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.10825 · **PDF:** https://arxiv.org/pdf/2602.10825v1
+
+## 13236. Adaptive Sampling for Private Worst-Case Group Optimization
+
+- **ID:** `2602.10820` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Max Cairney-Leeming, Amartya Sanyal, Christoph H. Lampert
+- **abs:** https://arxiv.org/abs/2602.10820 · **PDF:** https://arxiv.org/pdf/2602.10820v1
+
+## 13237. RePO: Bridging On-Policy Learning and Off-Policy Knowledge through Rephrasing Policy Optimization
+
+- **ID:** `2602.10819` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Linxuan Xia, Xiaolong Yang, Yongyuan Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.10819 · **PDF:** https://arxiv.org/pdf/2602.10819v1
+
+## 13238. Beyond Confidence: The Rhythms of Reasoning in Generative Models
+
+- **ID:** `2602.10816` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Deyuan Liu, Zecheng Wang, Zhanyue Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.10816 · **PDF:** https://arxiv.org/pdf/2602.10816v1
+
+## 13239. Understand Then Memory: A Cognitive Gist-Driven RAG Framework with Global Semantic Diffusion
+
+- **ID:** `2602.15895` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Pengcheng Zhou, Haochen Li, Zhiqiang Nie, et al.
+- **abs:** https://arxiv.org/abs/2602.15895 · **PDF:** https://arxiv.org/pdf/2602.15895v2
+
+## 13240. Why Does RL Generalize Better Than SFT? A Data-Centric Perspective on VLM Post-Training
+
+- **ID:** `2602.10815` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Aojun Lu, Tao Feng, Hangjie Yuan, et al.
+- **abs:** https://arxiv.org/abs/2602.10815 · **PDF:** https://arxiv.org/pdf/2602.10815v1
+
+## 13241. See, Plan, Snap: Evaluating Multimodal GUI Agents in Scratch
+
+- **ID:** `2602.10814` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Xingyi Zhang, Yulei Ye, Kaifeng Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.10814 · **PDF:** https://arxiv.org/pdf/2602.10814v1
+
+## 13242. PELLI: Framework to effectively integrate LLMs for quality software generation
+
+- **ID:** `2602.10808` · **primary:** `cs.SE` · **Submitted:** 2026-02-11
+- **Authors:** Rasmus Krebs, Somnath Mazumdar
+- **abs:** https://arxiv.org/abs/2602.10808 · **PDF:** https://arxiv.org/pdf/2602.10808v1
+
+## 13243. Integrating Generative AI-enhanced Cognitive Systems in Higher Education: From Stakeholder Perceptions to a Conceptual Framework considering the EU AI Act
+
+- **ID:** `2602.10802` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Da-Lun Chen, Prasasthy Balasubramanian, Lauri Lovén, et al.
+- **abs:** https://arxiv.org/abs/2602.10802 · **PDF:** https://arxiv.org/pdf/2602.10802v1
+
+## 13244. Deep Learning-based Method for Expressing Knowledge Boundary of Black-Box LLM
+
+- **ID:** `2602.10801` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Haotian Sheng, Heyong Wang, Ming Hong, et al.
+- **abs:** https://arxiv.org/abs/2602.10801 · **PDF:** https://arxiv.org/pdf/2602.10801v1
+
+## 13245. RSHallu: Dual-Mode Hallucination Evaluation for Remote-Sensing Multimodal Large Language Models with Domain-Tailored Mitigation
+
+- **ID:** `2602.10799` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Zihui Zhou, Yong Feng, Yanying Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.10799 · **PDF:** https://arxiv.org/pdf/2602.10799v1
+
+## 13246. PRISM: Parallel Residual Iterative Sequence Model
+
+- **ID:** `2602.10796` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jie Jiang, Ke Cheng, Xin Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.10796 · **PDF:** https://arxiv.org/pdf/2602.10796v2
+
+## 13247. Transport, Don't Generate: Deterministic Geometric Flows for Combinatorial Optimization
+
+- **ID:** `2602.10794` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Benjy Friedmann, Nadav Dym
+- **abs:** https://arxiv.org/abs/2602.10794 · **PDF:** https://arxiv.org/pdf/2602.10794v1
+
+## 13248. Semi-Supervised Cross-Domain Imitation Learning
+
+- **ID:** `2602.10793` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Li-Min Chu, Kai-Siang Ma, Ming-Hong Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.10793 · **PDF:** https://arxiv.org/pdf/2602.10793v1
+
+## 13249. Bayesian Signal Component Decomposition via Diffusion-within-Gibbs Sampling
+
+- **ID:** `2602.10792` · **primary:** `eess.SP` · **Submitted:** 2026-02-11
+- **Authors:** Yi Zhang, Rui Guo, Yonina C. Eldar
+- **abs:** https://arxiv.org/abs/2602.10792 · **PDF:** https://arxiv.org/pdf/2602.10792v1
+
+## 13250. The Automatic Verification of Image-Text Claims (AVerImaTeC) Shared Task
+
+- **ID:** `2602.11221` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Rui Cao, Zhenyun Deng, Yulong Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.11221 · **PDF:** https://arxiv.org/pdf/2602.11221v2
+
+## 13251. VulReaD: Knowledge-Graph-guided Software Vulnerability Reasoning and Detection
+
+- **ID:** `2602.10787` · **primary:** `cs.SE` · **Submitted:** 2026-02-11
+- **Authors:** Samal Mukhtar, Yinghua Yao, Zhu Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.10787 · **PDF:** https://arxiv.org/pdf/2602.10787v1
+
+## 13252. Kill it with FIRE: On Leveraging Latent Space Directions for Runtime Backdoor Mitigation in Deep Neural Networks
+
+- **ID:** `2602.10780` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Enrico Ahlers, Daniel Passon, Yannic Noller, et al.
+- **abs:** https://arxiv.org/abs/2602.10780 · **PDF:** https://arxiv.org/pdf/2602.10780v1
+
+## 13253. LOREN: Low Rank-Based Code-Rate Adaptation in Neural Receivers
+
+- **ID:** `2602.10770` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Bram Van Bolderik, Vlado Menkovski, Sonia Heemstra de Groot, et al.
+- **abs:** https://arxiv.org/abs/2602.10770 · **PDF:** https://arxiv.org/pdf/2602.10770v1
+
+## 13254. Collaborative Threshold Watermarking
+
+- **ID:** `2602.10765` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Tameem Bakr, Anish Ambreth, Nils Lukas
+- **abs:** https://arxiv.org/abs/2602.10765 · **PDF:** https://arxiv.org/pdf/2602.10765v1
+
+## 13255. Patch the Distribution Mismatch: RL Rewriting Agent for Stable Off-Policy SFT
+
+- **ID:** `2602.11220` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jiacheng Wang, Ping Jian, Zhen Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.11220 · **PDF:** https://arxiv.org/pdf/2602.11220v1
+
+## 13256. Amortized Inference of Neuron Parameters on Analog Neuromorphic Hardware
+
+- **ID:** `2602.10763` · **primary:** `cs.NE` · **Submitted:** 2026-02-11
+- **Authors:** Jakob Kaiser, Eric Müller, Johannes Schemmel
+- **abs:** https://arxiv.org/abs/2602.10763 · **PDF:** https://arxiv.org/pdf/2602.10763v2
+
+## 13257. Exploring the impact of adaptive rewiring in Graph Neural Networks
+
+- **ID:** `2602.10754` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Charlotte Cambier van Nooten, Christos Aronis, Yuliya Shapovalova, et al.
+- **abs:** https://arxiv.org/abs/2602.10754 · **PDF:** https://arxiv.org/pdf/2602.10754v1
+
+## 13258. Predicting integers from continuous parameters
+
+- **ID:** `2602.10751` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Bas Maat, Peter Bloem
+- **abs:** https://arxiv.org/abs/2602.10751 · **PDF:** https://arxiv.org/pdf/2602.10751v1
+
+## 13259. SecureScan: An AI-Driven Multi-Layer Framework for Malware and Phishing Detection Using Logistic Regression and Threat Intelligence Integration
+
+- **ID:** `2602.10750` · **primary:** `cs.CR` · **Submitted:** 2026-02-11
+- **Authors:** Rumman Firdos, Aman Dangi
+- **abs:** https://arxiv.org/abs/2602.10750 · **PDF:** https://arxiv.org/pdf/2602.10750v1
+
+## 13260. Spectral-Spatial Contrastive Learning Framework for Regression on Hyperspectral Data
+
+- **ID:** `2602.10745` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Mohamad Dhaini, Paul Honeine, Maxime Berar, et al.
+- **abs:** https://arxiv.org/abs/2602.10745 · **PDF:** https://arxiv.org/pdf/2602.10745v1
+
+## 13261. Self-Supervised Image Super-Resolution Quality Assessment based on Content-Free Multi-Model Oriented Representation Learning
+
+- **ID:** `2602.10744` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Kian Majlessi, Amir Masoud Soltani, Mohammad Ebrahim Mahdavi, et al.
+- **abs:** https://arxiv.org/abs/2602.10744 · **PDF:** https://arxiv.org/pdf/2602.10744v1
+
+## 13262. Kalman Linear Attention: Parallel Bayesian Filtering For Efficient Language Modelling and State Tracking
+
+- **ID:** `2602.10743` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Vaisakh Shaj, Cameron Barker, Aidan Scannell, et al.
+- **abs:** https://arxiv.org/abs/2602.10743 · **PDF:** https://arxiv.org/pdf/2602.10743v1
+
+## 13263. Reinforced Curriculum Pre-Alignment for Domain-Adaptive VLMs
+
+- **ID:** `2602.10740` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Yuming Yan, Shuo Yang, Kai Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.10740 · **PDF:** https://arxiv.org/pdf/2602.10740v1
+
+## 13264. The Million-Label NER: Breaking Scale Barriers with GLiNER bi-encoder
+
+- **ID:** `2602.18487` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Ihor Stepanov, Mykhailo Shtopko, Dmytro Vodianytskyi, et al.
+- **abs:** https://arxiv.org/abs/2602.18487 · **PDF:** https://arxiv.org/pdf/2602.18487v1
+
+## 13265. Calliope: A TTS-based Narrated E-book Creator Ensuring Exact Synchronization, Privacy, and Layout Fidelity
+
+- **ID:** `2602.10735` · **primary:** `cs.SD` · **Submitted:** 2026-02-11
+- **Authors:** Hugo L. Hammer, Vajira Thambawita, Pål Halvorsen
+- **abs:** https://arxiv.org/abs/2602.10735 · **PDF:** https://arxiv.org/pdf/2602.10735v1
+
+## 13266. Credal Concept Bottleneck Models: Structural Separation of Epistemic and Aleatoric Uncertainty
+
+- **ID:** `2602.11219` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Tanmoy Mukherjee, Marius Kloft, Pierre Marquis, et al.
+- **abs:** https://arxiv.org/abs/2602.11219 · **PDF:** https://arxiv.org/pdf/2602.11219v1
+
+## 13267. Beyond Musical Descriptors: Extracting Preference-Bearing Intent in Music Queries
+
+- **ID:** `2602.12301` · **primary:** `cs.SD` · **Submitted:** 2026-02-11
+- **Authors:** Marion Baranes, Romain Hennequin, Elena V. Epure
+- **abs:** https://arxiv.org/abs/2602.12301 · **PDF:** https://arxiv.org/pdf/2602.12301v1
+
+## 13268. Macaron: Controlled, Human-Written Benchmark for Multilingual and Multicultural Reasoning via Template-Filling
+
+- **ID:** `2602.10732` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Alaa Elsetohy, Sama Hadhoud, Haryo Akbarianto Wibowo, et al.
+- **abs:** https://arxiv.org/abs/2602.10732 · **PDF:** https://arxiv.org/pdf/2602.10732v1
+
+## 13269. Rising Multi-Armed Bandits with Known Horizons
+
+- **ID:** `2602.10727` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Seockbean Song, Chenyu Gan, Youngsik Yoon, et al.
+- **abs:** https://arxiv.org/abs/2602.10727 · **PDF:** https://arxiv.org/pdf/2602.10727v2
+
+## 13270. Quality-constrained Entropy Maximization Policy Optimization for LLM Diversity
+
+- **ID:** `2602.15894` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Haihui Pan, Yuzhong Hong, Shaoke Lv, et al.
+- **abs:** https://arxiv.org/abs/2602.15894 · **PDF:** https://arxiv.org/pdf/2602.15894v1
+
+## 13271. A Diffusion-Based Generative Prior Approach to Sparse-view Computed Tomography
+
+- **ID:** `2602.10722` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Davide Evangelista, Pasquale Cascarano, Elena Loli Piccolomini
+- **abs:** https://arxiv.org/abs/2602.10722 · **PDF:** https://arxiv.org/pdf/2602.10722v1
+
+## 13272. SnapMLA: Efficient Long-Context MLA Decoding via Hardware-Aware FP8 Quantized Pipelining
+
+- **ID:** `2602.10718` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Yifan Zhang, Zunhai Su, Shuhao Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.10718 · **PDF:** https://arxiv.org/pdf/2602.10718v2
+
+## 13273. RE-LLM: Refining Empathetic Speech-LLM Responses by Integrating Emotion Nuance
+
+- **ID:** `2602.10716` · **primary:** `eess.AS` · **Submitted:** 2026-02-11
+- **Authors:** Jing-Han Chen, Bo-Hao Su, Ya-Tse Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.10716 · **PDF:** https://arxiv.org/pdf/2602.10716v1
+
+## 13274. Locomo-Plus: Beyond-Factual Cognitive Memory Evaluation Framework for LLM Agents
+
+- **ID:** `2602.10715` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Yifei Li, Weidong Guo, Lingling Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.10715 · **PDF:** https://arxiv.org/pdf/2602.10715v1
+
+## 13275. The CompMath-MCQ Dataset: Are LLMs Ready for Higher-Level Math?
+
+- **ID:** `2603.03334` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Bianca Raimondi, Francesco Pivi, Davide Evangelista, et al.
+- **abs:** https://arxiv.org/abs/2603.03334 · **PDF:** https://arxiv.org/pdf/2603.03334v1
+
+## 13276. A Non-asymptotic Analysis for Learning and Applying a Preconditioner in MCMC
+
+- **ID:** `2602.10714` · **primary:** `stat.CO` · **Submitted:** 2026-02-11
+- **Authors:** Max Hird, Florian Maire, Jeffrey Negrea
+- **abs:** https://arxiv.org/abs/2602.10714 · **PDF:** https://arxiv.org/pdf/2602.10714v1
+
+## 13277. Cross-Sectional Asset Retrieval via Future-Aligned Soft Contrastive Learning
+
+- **ID:** `2602.10711` · **primary:** `cs.CE` · **Submitted:** 2026-02-11
+- **Authors:** Hyeongmin Lee, Chanyeol Choi, Jihoon Kwon, et al.
+- **abs:** https://arxiv.org/abs/2602.10711 · **PDF:** https://arxiv.org/pdf/2602.10711v1
+
+## 13278. Interpretable Graph-Level Anomaly Detection via Contrast with Normal Prototypes
+
+- **ID:** `2602.10708` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Qiuran Zhao, Kai Ming Ting, Xinpeng Li
+- **abs:** https://arxiv.org/abs/2602.10708 · **PDF:** https://arxiv.org/pdf/2602.10708v1
+
+## 13279. Reducing Estimation Uncertainty Using Normalizing Flows and Stratification
+
+- **ID:** `2602.10706` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Paweł Lorek, Rafał Nowak, Rafał Topolnicki, et al.
+- **abs:** https://arxiv.org/abs/2602.10706 · **PDF:** https://arxiv.org/pdf/2602.10706v3
+
+## 13280. A Unified Experimental Architecture for Informative Path Planning: from Simulation to Deployment with GuadalPlanner
+
+- **ID:** `2602.10702` · **primary:** `cs.RO` · **Submitted:** 2026-02-11
+- **Authors:** Alejandro Mendoza Barrionuevo, Dame Seck Diop, Alejandro Casado Pérez, et al.
+- **abs:** https://arxiv.org/abs/2602.10702 · **PDF:** https://arxiv.org/pdf/2602.10702v1
+
+## 13281. Spend Search Where It Pays: Value-Guided Structured Sampling and Optimization for Generative Recommendation
+
+- **ID:** `2602.10699` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Jie Jiang, Yangru Huang, Zeyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.10699 · **PDF:** https://arxiv.org/pdf/2602.10699v2
+
+## 13282. AugVLA-3D: Depth-Driven Feature Augmentation for Vision-Language-Action Models
+
+- **ID:** `2602.10698` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Zhifeng Rao, Wenlong Chen, Lei Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.10698 · **PDF:** https://arxiv.org/pdf/2602.10698v1
+
+## 13283. Fast and Large-Scale Unbalanced Optimal Transport via its Semi-Dual and Adaptive Gradient Methods
+
+- **ID:** `2602.10697` · **primary:** `math.OC` · **Submitted:** 2026-02-11
+- **Authors:** Ferdinand Genans
+- **abs:** https://arxiv.org/abs/2602.10697 · **PDF:** https://arxiv.org/pdf/2602.10697v1
+
+## 13284. Robust Assortment Optimization from Observational Data
+
+- **ID:** `2602.10696` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Miao Lu, Yuxuan Han, Han Zhong, et al.
+- **abs:** https://arxiv.org/abs/2602.10696 · **PDF:** https://arxiv.org/pdf/2602.10696v2
+
+## 13285. VESPO: Variational Sequence-Level Soft Policy Optimization for Stable Off-Policy LLM Training
+
+- **ID:** `2602.10693` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Guobin Shen, Chenxiao Zhao, Xiang Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.10693 · **PDF:** https://arxiv.org/pdf/2602.10693v2
+
+## 13286. Convergence Rates for Distribution Matching with Sliced Optimal Transport
+
+- **ID:** `2602.10691` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Gauthier Thurin, Claire Boyer, Kimia Nadjahi
+- **abs:** https://arxiv.org/abs/2602.10691 · **PDF:** https://arxiv.org/pdf/2602.10691v1
+
+## 13287. MedScope: Incentivizing "Think with Videos" for Clinical Reasoning via Coarse-to-Fine Tool Calling
+
+- **ID:** `2602.13332` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Wenjie Li, Yujie Zhang, Haoran Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.13332 · **PDF:** https://arxiv.org/pdf/2602.13332v1
+
+## 13288. The Magic Correlations: Understanding Knowledge Transfer from Pretraining to Supervised Fine-Tuning
+
+- **ID:** `2602.11217` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Simin Fan, Dimitris Paparas, Natasha Noy, et al.
+- **abs:** https://arxiv.org/abs/2602.11217 · **PDF:** https://arxiv.org/pdf/2602.11217v1
+
+## 13289. OmniVL-Guard: Towards Unified Vision-Language Forgery Detection and Grounding via Balanced RL
+
+- **ID:** `2602.10687` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Jinjie Shen, Jing Wu, Yaxiong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.10687 · **PDF:** https://arxiv.org/pdf/2602.10687v2
+
+## 13290. Beyond Task Performance: A Metric-Based Analysis of Sequential Cooperation in Heterogeneous Multi-Agent Destructive Foraging
+
+- **ID:** `2602.10685` · **primary:** `cs.MA` · **Submitted:** 2026-02-11
+- **Authors:** Alejandro Mendoza Barrionuevo, Samuel Yanes Luis, Daniel Gutiérrez Reina, et al.
+- **abs:** https://arxiv.org/abs/2602.10685 · **PDF:** https://arxiv.org/pdf/2602.10685v1
+
+## 13291. A solvable high-dimensional model where nonlinear autoencoders learn structure invisible to PCA while test loss misaligns with generalization
+
+- **ID:** `2602.10680` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Vicente Conde Mendes, Lorenzo Bardone, Cédric Koller, et al.
+- **abs:** https://arxiv.org/abs/2602.10680 · **PDF:** https://arxiv.org/pdf/2602.10680v1
+
+## 13292. Protein Language Model Embeddings Improve Generalization of Implicit Transfer Operators
+
+- **ID:** `2602.11216` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Panagiotis Antoniadis, Beatrice Pavesi, Simon Olsson, et al.
+- **abs:** https://arxiv.org/abs/2602.11216 · **PDF:** https://arxiv.org/pdf/2602.11216v1
+
+## 13293. TwiFF (Think With Future Frames): A Large-Scale Dataset for Dynamic Visual Reasoning
+
+- **ID:** `2602.10675` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Junhua Liu, Zhangcheng Wang, Zhike Han, et al.
+- **abs:** https://arxiv.org/abs/2602.10675 · **PDF:** https://arxiv.org/pdf/2602.10675v1
+
+## 13294. Domain Knowledge Guided Bayesian Optimization For Autonomous Alignment Of Complex Scientific Instruments
+
+- **ID:** `2602.10670` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Aashwin Mishra, Matt Seaberg, Ryan Roussel, et al.
+- **abs:** https://arxiv.org/abs/2602.10670 · **PDF:** https://arxiv.org/pdf/2602.10670v1
+
+## 13295. From Diet to Free Lunch: Estimating Auxiliary Signal Properties using Dynamic Pruning Masks in Speech Enhancement Networks
+
+- **ID:** `2602.10666` · **primary:** `eess.AS` · **Submitted:** 2026-02-11
+- **Authors:** Riccardo Miccini, Clément Laroche, Tobias Piechowiak, et al.
+- **abs:** https://arxiv.org/abs/2602.10666 · **PDF:** https://arxiv.org/pdf/2602.10666v1
+
+## 13296. Targeted Syntactic Evaluation of Language Models on Georgian Case Alignment
+
+- **ID:** `2602.10661` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Daniel Gallagher, Gerhard Heyer
+- **abs:** https://arxiv.org/abs/2602.10661 · **PDF:** https://arxiv.org/pdf/2602.10661v2
+
+## 13297. Charting Empirical Laws for LLM Fine-Tuning in Scientific Multi-Discipline Learning
+
+- **ID:** `2602.11215` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Lintao Wang, Zhuqiang Lu, Yilin Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.11215 · **PDF:** https://arxiv.org/pdf/2602.11215v1
+
+## 13298. Benchmarks Are Not That Out of Distribution: Word Overlap Predicts Performance
+
+- **ID:** `2602.10657` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Woojin Chung, Jeonghoon Kim
+- **abs:** https://arxiv.org/abs/2602.10657 · **PDF:** https://arxiv.org/pdf/2602.10657v1
+
+## 13299. UMEM: Unified Memory Extraction and Management Framework for Generalizable Memory
+
+- **ID:** `2602.10652` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Yongshi Ye, Hui Jiang, Feihu Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.10652 · **PDF:** https://arxiv.org/pdf/2602.10652v1
+
+## 13300. Evaluation metrics for temporal preservation in synthetic longitudinal patient data
+
+- **ID:** `2602.10643` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Katariina Perkonoja, Parisa Movahedi, Antti Airola, et al.
+- **abs:** https://arxiv.org/abs/2602.10643 · **PDF:** https://arxiv.org/pdf/2602.10643v1
+
+## 13301. Beyond Kemeny Medians: Consensus Ranking Distributions Definition, Properties and Statistical Learning
+
+- **ID:** `2602.10640` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Stephan Clémençon, Ekhine Irurozki
+- **abs:** https://arxiv.org/abs/2602.10640 · **PDF:** https://arxiv.org/pdf/2602.10640v1
+
+## 13302. Coarse-Grained Boltzmann Generators
+
+- **ID:** `2602.10637` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Weilong Chen, Bojun Zhao, Jan Eckwert, et al.
+- **abs:** https://arxiv.org/abs/2602.10637 · **PDF:** https://arxiv.org/pdf/2602.10637v1
+
+## 13303. OmniSapiens: A Foundation Model for Social Behavior Processing via Heterogeneity-Aware Relative Policy Optimization
+
+- **ID:** `2602.10635` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Keane Ong, Sabri Boughorbel, Luwei Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.10635 · **PDF:** https://arxiv.org/pdf/2602.10635v1
+
+## 13304. Statistical-Geometric Degeneracy in UAV Search: A Physics-Aware Asymmetric Filtering Approach
+
+- **ID:** `2602.15893` · **primary:** `cs.RO` · **Submitted:** 2026-02-11
+- **Authors:** Zhiyuan Ren, Yudong Fang, Tao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.15893 · **PDF:** https://arxiv.org/pdf/2602.15893v1
+
+## 13305. The Neurosymbolic Frontier of Nonuniform Ellipticity: Formalizing Sharp Schauder Theory via Topos-Theoretic Reasoning Models
+
+- **ID:** `2602.10632` · **primary:** `cs.SC` · **Submitted:** 2026-02-11
+- **Authors:** Suyash Mishra
+- **abs:** https://arxiv.org/abs/2602.10632 · **PDF:** https://arxiv.org/pdf/2602.10632v1
+
+## 13306. Generative clinical time series models trained on moderate amounts of patient data are privacy preserving
+
+- **ID:** `2602.10631` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Rustam Zhumagambetov, Niklas Giesa, Sebastian D. Boie, et al.
+- **abs:** https://arxiv.org/abs/2602.10631 · **PDF:** https://arxiv.org/pdf/2602.10631v1
+
+## 13307. To Think or Not To Think, That is The Question for Large Reasoning Models in Theory of Mind Tasks
+
+- **ID:** `2602.10625` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Nanxu Gong, Haotian Li, Sixun Dong, et al.
+- **abs:** https://arxiv.org/abs/2602.10625 · **PDF:** https://arxiv.org/pdf/2602.10625v3
+
+## 13308. A Vision-Language Foundation Model for Zero-shot Clinical Collaboration and Automated Concept Discovery in Dermatology
+
+- **ID:** `2602.10624` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Siyuan Yan, Xieji Li, Dan Mo, et al.
+- **abs:** https://arxiv.org/abs/2602.10624 · **PDF:** https://arxiv.org/pdf/2602.10624v1
+
+## 13309. Mitigating Reward Hacking in RLHF via Bayesian Non-negative Reward Modeling
+
+- **ID:** `2602.10623` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Zhibin Duan, Guowei Rong, Zhuo Li, et al.
+- **abs:** https://arxiv.org/abs/2602.10623 · **PDF:** https://arxiv.org/pdf/2602.10623v1
+
+## 13310. How Do Decoder-Only LLMs Perceive Users? Rethinking Attention Masking for User Representation Learning
+
+- **ID:** `2602.10622` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Jiahao Yuan, Yike Xu, Jinyong Wen, et al.
+- **abs:** https://arxiv.org/abs/2602.10622 · **PDF:** https://arxiv.org/pdf/2602.10622v1
+
+## 13311. ISD-Agent-Bench: A Comprehensive Benchmark for Evaluating LLM-based Instructional Design Agents
+
+- **ID:** `2602.10620` · **primary:** `cs.SE` · **Submitted:** 2026-02-11
+- **Authors:** YoungHoon Jeon, Suwan Kim, Haein Son, et al.
+- **abs:** https://arxiv.org/abs/2602.10620 · **PDF:** https://arxiv.org/pdf/2602.10620v1
+
+## 13312. Improving Medical Visual Reinforcement Fine-Tuning via Perception and Reasoning Augmentation
+
+- **ID:** `2602.10619` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Guangjing Yang, ZhangYuan Yu, Ziyuan Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.10619 · **PDF:** https://arxiv.org/pdf/2602.10619v1
+
+## 13313. Pupillometry and Brain Dynamics for Cognitive Load in Working Memory
+
+- **ID:** `2602.10614` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Nusaibah Farrukh, Malavika Pradeep, Akshay Sasi, et al.
+- **abs:** https://arxiv.org/abs/2602.10614 · **PDF:** https://arxiv.org/pdf/2602.10614v1
+
+## 13314. Highly Adaptive Principal Component Regression
+
+- **ID:** `2602.10613` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Mingxun Wang, Alejandro Schuler, Mark van der Laan, et al.
+- **abs:** https://arxiv.org/abs/2602.10613 · **PDF:** https://arxiv.org/pdf/2602.10613v1
+
+## 13315. On the Role of Consistency Between Physics and Data in Physics-Informed Neural Networks
+
+- **ID:** `2602.10611` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Nicolás Becerra-Zuniga, Lucas Lacasa, Eusebio Valero, et al.
+- **abs:** https://arxiv.org/abs/2602.10611 · **PDF:** https://arxiv.org/pdf/2602.10611v1
+
+## 13316. Online Causal Kalman Filtering for Stable and Effective Policy Optimization
+
+- **ID:** `2602.10609` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Shuo He, Lang Feng, Xin Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.10609 · **PDF:** https://arxiv.org/pdf/2602.10609v2
+
+## 13317. Bayesian Inference of Contextual Bandit Policies via Empirical Likelihood
+
+- **ID:** `2602.10608` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Jiangrong Ouyang, Mingming Gong, Howard Bondell
+- **abs:** https://arxiv.org/abs/2602.10608 · **PDF:** https://arxiv.org/pdf/2602.10608v1
+
+## 13318. Support Vector Data Description for Radar Target Detection
+
+- **ID:** `2602.18486` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jean Pinsolle, Yadang Alexis Rouzoumka, Chengfang Ren, et al.
+- **abs:** https://arxiv.org/abs/2602.18486 · **PDF:** https://arxiv.org/pdf/2602.18486v1
+
+## 13319. Hierarchical Zero-Order Optimization for Deep Neural Networks
+
+- **ID:** `2602.10607` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Sansheng Cao, Zhengyu Ma, Yonghong Tian
+- **abs:** https://arxiv.org/abs/2602.10607 · **PDF:** https://arxiv.org/pdf/2602.10607v1
+
+## 13320. Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters
+
+- **ID:** `2602.10604` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Ailin Huang, Ang Li, Aobo Kong, et al.
+- **abs:** https://arxiv.org/abs/2602.10604 · **PDF:** https://arxiv.org/pdf/2602.10604v2
+
+## 13321. APEX-SQL: Talking to the data via Agentic Exploration for Text-to-SQL
+
+- **ID:** `2602.16720` · **primary:** `cs.DB` · **Submitted:** 2026-02-11
+- **Authors:** Bowen Cao, Weibin Liao, Yushi Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.16720 · **PDF:** https://arxiv.org/pdf/2602.16720v1
+
+## 13322. dnaHNet: A Scalable and Hierarchical Foundation Model for Genomic Sequence Learning
+
+- **ID:** `2602.10603` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Arnav Shah, Junzhe Li, Parsa Idehpour, et al.
+- **abs:** https://arxiv.org/abs/2602.10603 · **PDF:** https://arxiv.org/pdf/2602.10603v2
+
+## 13323. Learning Mixture Density via Natural Gradient Expectation Maximization
+
+- **ID:** `2602.10602` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Yutao Chen, Jasmine Bayrooti, Steven Morad
+- **abs:** https://arxiv.org/abs/2602.10602 · **PDF:** https://arxiv.org/pdf/2602.10602v1
+
+## 13324. Neuro-symbolic Action Masking for Deep Reinforcement Learning
+
+- **ID:** `2602.10598` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Shuai Han, Mehdi Dastani, Shihan Wang
+- **abs:** https://arxiv.org/abs/2602.10598 · **PDF:** https://arxiv.org/pdf/2602.10598v1
+
+## 13325. Roughness-Informed Federated Learning
+
+- **ID:** `2602.10595` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Mohammad Partohaghighi, Roummel Marcia, Bruce J. West, et al.
+- **abs:** https://arxiv.org/abs/2602.10595 · **PDF:** https://arxiv.org/pdf/2602.10595v1
+
+## 13326. Flow-Enabled Generalization to Human Demonstrations in Few-Shot Imitation Learning
+
+- **ID:** `2602.10594` · **primary:** `cs.RO` · **Submitted:** 2026-02-11
+- **Authors:** Runze Tang, Penny Sweetser
+- **abs:** https://arxiv.org/abs/2602.10594 · **PDF:** https://arxiv.org/pdf/2602.10594v2
+
+## 13327. TRACE: Theoretical Risk Attribution under Covariate-shift Effects
+
+- **ID:** `2602.10588` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Hosein Anjidani, S. Yahya S. R. Tehrani, Mohammad Mahdi Mojahedian, et al.
+- **abs:** https://arxiv.org/abs/2602.10588 · **PDF:** https://arxiv.org/pdf/2602.10588v1
+
+## 13328. Towards Compressive and Scalable Recurrent Memory
+
+- **ID:** `2602.11212` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Yunchong Song, Jushi Kai, Liming Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.11212 · **PDF:** https://arxiv.org/pdf/2602.11212v1
+
+## 13329. Deep Bootstrap
+
+- **ID:** `2602.10587` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Jinyuan Chang, Yuling Jiao, Lican Kang, et al.
+- **abs:** https://arxiv.org/abs/2602.10587 · **PDF:** https://arxiv.org/pdf/2602.10587v1
+
+## 13330. Neural Additive Experts: Context-Gated Experts for Controllable Model Additivity
+
+- **ID:** `2602.10585` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Guangzhi Xiong, Sanchit Sinha, Aidong Zhang
+- **abs:** https://arxiv.org/abs/2602.10585 · **PDF:** https://arxiv.org/pdf/2602.10585v1
+
+## 13331. When Gradient Clipping Becomes a Control Mechanism for Differential Privacy in Deep Learning
+
+- **ID:** `2602.10584` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Mohammad Partohaghighi, Roummel Marcia, Bruce J. West, et al.
+- **abs:** https://arxiv.org/abs/2602.10584 · **PDF:** https://arxiv.org/pdf/2602.10584v1
+
+## 13332. Flow of Spans: Generalizing Language Models to Dynamic Span-Vocabulary via GFlowNets
+
+- **ID:** `2602.10583` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Bo Xue, Yunchong Song, Fanghao Shao, et al.
+- **abs:** https://arxiv.org/abs/2602.10583 · **PDF:** https://arxiv.org/pdf/2602.10583v1
+
+## 13333. HiST-VLA: A Hierarchical Spatio-Temporal Vision-Language-Action Model for End-to-End Autonomous Driving
+
+- **ID:** `2602.13329` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Yiru Wang, Zichong Gu, Yu Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.13329 · **PDF:** https://arxiv.org/pdf/2602.13329v1
+
+## 13334. LLM-Based Scientific Equation Discovery via Physics-Informed Token-Regularized Policy Optimization
+
+- **ID:** `2602.10576` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Boxiao Wang, Kai Li, Tianyi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.10576 · **PDF:** https://arxiv.org/pdf/2602.10576v1
+
+## 13335. MetaphorStar: Image Metaphor Understanding and Reasoning with End-to-End Visual Reinforcement Learning
+
+- **ID:** `2602.10575` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Chenhao Zhang, Yazhe Niu, Hongsheng Li
+- **abs:** https://arxiv.org/abs/2602.10575 · **PDF:** https://arxiv.org/pdf/2602.10575v1
+
+## 13336. Gauss-Newton Unlearning for the LLM Era
+
+- **ID:** `2602.10568` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Lev McKinney, Anvith Thudi, Juhan Bae, et al.
+- **abs:** https://arxiv.org/abs/2602.10568 · **PDF:** https://arxiv.org/pdf/2602.10568v1
+
+## 13337. Online Min-Max Optimization: From Individual Regrets to Cumulative Saddle Points
+
+- **ID:** `2602.10565` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Abhijeet Vyas, Brian Bullins
+- **abs:** https://arxiv.org/abs/2602.10565 · **PDF:** https://arxiv.org/pdf/2602.10565v1
+
+## 13338. When to Memorize and When to Stop: Gated Recurrent Memory for Long-Context Reasoning
+
+- **ID:** `2602.10560` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Leheng Sheng, Yongtao Zhang, Wenchang Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.10560 · **PDF:** https://arxiv.org/pdf/2602.10560v1
+
+## 13339. LAP: Language-Action Pre-Training Enables Zero-shot Cross-Embodiment Transfer
+
+- **ID:** `2602.10556` · **primary:** `cs.RO` · **Submitted:** 2026-02-11
+- **Authors:** Lihan Zha, Asher J. Hancock, Mingtong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.10556 · **PDF:** https://arxiv.org/pdf/2602.10556v2
+
+## 13340. Contrastive Learning for Multi Label ECG Classification with Jaccard Score Based Sigmoid Loss
+
+- **ID:** `2602.10553` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Junichiro Takahashi, Masataka Sato, Satoshi Kodeta, et al.
+- **abs:** https://arxiv.org/abs/2602.10553 · **PDF:** https://arxiv.org/pdf/2602.10553v1
+
+## 13341. MindPilot: Closed-loop Visual Stimulation Optimization for Brain Modulation with EEG-guided Diffusion
+
+- **ID:** `2602.10552` · **primary:** `cs.NE` · **Submitted:** 2026-02-11
+- **Authors:** Dongyang Li, Kunpeng Xie, Mingyang Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.10552 · **PDF:** https://arxiv.org/pdf/2602.10552v2
+
+## 13342. C^2ROPE: Causal Continuous Rotary Positional Encoding for 3D Large Multimodal-Models Reasoning
+
+- **ID:** `2602.10551` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Guanting Ye, Qiyan Zhao, Wenhao Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.10551 · **PDF:** https://arxiv.org/pdf/2602.10551v2
+
+## 13343. Enhancing Weakly Supervised Multimodal Video Anomaly Detection through Text Guidance
+
+- **ID:** `2602.10549` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Shengyang Sun, Jiashen Hua, Junyi Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.10549 · **PDF:** https://arxiv.org/pdf/2602.10549v1
+
+## 13344. RealHD: A High-Quality Dataset for Robust Detection of State-of-the-Art AI-Generated Images
+
+- **ID:** `2602.10546` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Hanzhe Yu, Yun Ye, Jintao Rong, et al.
+- **abs:** https://arxiv.org/abs/2602.10546 · **PDF:** https://arxiv.org/pdf/2602.10546v1
+
+## 13345. $μ$pscaling small models: Principled warm starts and hyperparameter transfer
+
+- **ID:** `2602.10545` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Yuxin Ma, Nan Chen, Mateo Díaz, et al.
+- **abs:** https://arxiv.org/abs/2602.10545 · **PDF:** https://arxiv.org/pdf/2602.10545v1
+
+## 13346. Bridging the Compression-Precision Paradox: A Hybrid Architecture for Clinical EEG Report Generation with Guaranteed Measurement Accuracy
+
+- **ID:** `2602.10544` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Wuyang Zhang, Zhen Luo, Chuqiao Gu, et al.
+- **abs:** https://arxiv.org/abs/2602.10544 · **PDF:** https://arxiv.org/pdf/2602.10544v1
+
+## 13347. Predictive-State Communication: Innovation Coding and Reconciliation under Delay
+
+- **ID:** `2602.10542` · **primary:** `cs.IT` · **Submitted:** 2026-02-11
+- **Authors:** Ozgur Ercetin, Mohaned Chraiti
+- **abs:** https://arxiv.org/abs/2602.10542 · **PDF:** https://arxiv.org/pdf/2602.10542v1
+
+## 13348. FastLSQ: Solving PDEs in One Shot via Fourier Features with Exact Analytical Derivatives
+
+- **ID:** `2602.10541` · **primary:** `math.NA` · **Submitted:** 2026-02-11
+- **Authors:** Antonin Sulc
+- **abs:** https://arxiv.org/abs/2602.10541 · **PDF:** https://arxiv.org/pdf/2602.10541v4
+
+## 13349. What Makes Value Learning Efficient in Residual Reinforcement Learning?
+
+- **ID:** `2602.10539` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Guozheng Ma, Lu Li, Haoyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.10539 · **PDF:** https://arxiv.org/pdf/2602.10539v1
+
+## 13350. Why Agentic Theorem Prover Works: A Statistical Provability Theory of Mathematical Reasoning Models
+
+- **ID:** `2602.10538` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Sho Sonoda, Shunta Akiyama, Yuya Uezato
+- **abs:** https://arxiv.org/abs/2602.10538 · **PDF:** https://arxiv.org/pdf/2602.10538v2
+
+## 13351. Statistical Inference and Learning for Shapley Additive Explanations (SHAP)
+
+- **ID:** `2602.10532` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Justin Whitehouse, Ayush Sawarni, Vasilis Syrgkanis
+- **abs:** https://arxiv.org/abs/2602.10532 · **PDF:** https://arxiv.org/pdf/2602.10532v1
+
+## 13352. From Collapse to Improvement: Statistical Perspectives on the Evolutionary Dynamics of Iterative Training on Contaminated Sources
+
+- **ID:** `2602.10531` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Soham Bakshi, Sunrit Chakraborty
+- **abs:** https://arxiv.org/abs/2602.10531 · **PDF:** https://arxiv.org/pdf/2602.10531v2
+
+## 13353. Generalized Robust Adaptive-Bandwidth Multi-View Manifold Learning in High Dimensions with Noise
+
+- **ID:** `2602.10530` · **primary:** `stat.ML` · **Submitted:** 2026-02-11
+- **Authors:** Xiucai Ding, Chao Shen, Hau-Tieng Wu
+- **abs:** https://arxiv.org/abs/2602.10530 · **PDF:** https://arxiv.org/pdf/2602.10530v1
+
+## 13354. Training-free Dropout Sampling for Semantic Token Acceptance in Speculative Decoding
+
+- **ID:** `2603.03333` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Jeongtae Lee, Minjung Jo, Hyunjoon Jeong, et al.
+- **abs:** https://arxiv.org/abs/2603.03333 · **PDF:** https://arxiv.org/pdf/2603.03333v1
+
+## 13355. A Swap-Adversarial Framework for Improving Domain Generalization in Electroencephalography-Based Parkinson's Disease Prediction
+
+- **ID:** `2602.10528` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Seongwon Jin, Hanseul Choi, Sunggu Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.10528 · **PDF:** https://arxiv.org/pdf/2602.10528v1
+
+## 13356. AI-PACE: A Framework for Integrating AI into Medical Education
+
+- **ID:** `2602.10527` · **primary:** `cs.CY` · **Submitted:** 2026-02-11
+- **Authors:** Scott P. McGrath, Katherine K. Kim, Karnjit Johl, et al.
+- **abs:** https://arxiv.org/abs/2602.10527 · **PDF:** https://arxiv.org/pdf/2602.10527v1
+
+## 13357. LHAW: Controllable Underspecification for Long-Horizon Tasks
+
+- **ID:** `2602.10525` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** George Pu, Michael S. Lee, Udari Madhushani Sehwag, et al.
+- **abs:** https://arxiv.org/abs/2602.10525 · **PDF:** https://arxiv.org/pdf/2602.10525v2
+
+## 13358. Scaling Reward Modeling without Human Supervision
+
+- **ID:** `2603.02225` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jingxuan Fan, Yueying Li, Zhenting Qi, et al.
+- **abs:** https://arxiv.org/abs/2603.02225 · **PDF:** https://arxiv.org/pdf/2603.02225v2
+
+## 13359. Prioritize the Process, Not Just the Outcome: Rewarding Latent Thought Trajectories Improves Reasoning in Looped Language Models
+
+- **ID:** `2602.10520` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jonathan Williams, Esin Tureci
+- **abs:** https://arxiv.org/abs/2602.10520 · **PDF:** https://arxiv.org/pdf/2602.10520v2
+
+## 13360. Co-jump: Cooperative Jumping with Quadrupedal Robots via Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.10514` · **primary:** `cs.RO` · **Submitted:** 2026-02-11
+- **Authors:** Shihao Dong, Yeke Chen, Zeren Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.10514 · **PDF:** https://arxiv.org/pdf/2602.10514v1
+
+## 13361. 1%>100%: High-Efficiency Visual Adapter with Complex Linear Projection Optimization
+
+- **ID:** `2602.10513` · **primary:** `cs.CV` · **Submitted:** 2026-02-11
+- **Authors:** Dongshuo Yin, Xue Yang, Deng-Ping Fan, et al.
+- **abs:** https://arxiv.org/abs/2602.10513 · **PDF:** https://arxiv.org/pdf/2602.10513v1
+
+## 13362. Don't Eliminate Cut: Exponential Separations in LLM-Based Theorem Proving
+
+- **ID:** `2602.10512` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Sho Sonoda, Shunta Akiyama, Yuya Uezato
+- **abs:** https://arxiv.org/abs/2602.10512 · **PDF:** https://arxiv.org/pdf/2602.10512v1
+
+## 13363. Privacy-Utility Tradeoffs in Quantum Information Processing
+
+- **ID:** `2602.10510` · **primary:** `quant-ph` · **Submitted:** 2026-02-11
+- **Authors:** Theshani Nuradha, Sujeet Bhalerao, Felix Leditzky
+- **abs:** https://arxiv.org/abs/2602.10510 · **PDF:** https://arxiv.org/pdf/2602.10510v1
+
+## 13364. Learning Structure-Semantic Evolution Trajectories for Graph Domain Adaptation
+
+- **ID:** `2602.10506` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Wei Chen, Xingyu Guo, Shuang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.10506 · **PDF:** https://arxiv.org/pdf/2602.10506v1
+
+## 13365. On the Robustness of Knowledge Editing for Detoxification
+
+- **ID:** `2602.10504` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Ming Dong, Shiyi Tang, Ziyan Peng, et al.
+- **abs:** https://arxiv.org/abs/2602.10504 · **PDF:** https://arxiv.org/pdf/2602.10504v1
+
+## 13366. Enhancing Ride-Hailing Forecasting at DiDi with Multi-View Geospatial Representation Learning from the Web
+
+- **ID:** `2602.10502` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Xixuan Hao, Guicheng Li, Daiqiang Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.10502 · **PDF:** https://arxiv.org/pdf/2602.10502v1
+
+## 13367. Low-Dimensional Execution Manifolds in Transformer Learning Dynamics: Evidence from Modular Arithmetic Tasks
+
+- **ID:** `2602.10496` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Yongzhong Xu
+- **abs:** https://arxiv.org/abs/2602.10496 · **PDF:** https://arxiv.org/pdf/2602.10496v2
+
+## 13368. Canvas-of-Thought: Grounding Reasoning via Mutable Structured States
+
+- **ID:** `2602.10494` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Lingzhuang Sun, Yuxia Zhu, Ruitong Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.10494 · **PDF:** https://arxiv.org/pdf/2602.10494v1
+
+## 13369. Learning Adaptive Distribution Alignment with Neural Characteristic Function for Graph Domain Adaptation
+
+- **ID:** `2602.10489` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Wei Chen, Xingyu Guo, Shuang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.10489 · **PDF:** https://arxiv.org/pdf/2602.10489v2
+
+## 13370. Abstraction Generation for Generalized Planning with Pretrained Large Language Models
+
+- **ID:** `2602.10485` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Zhenhe Cui, Huaxiang Xia, Hangjun Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.10485 · **PDF:** https://arxiv.org/pdf/2602.10485v1
+
+## 13371. Pricing Query Complexity of Multiplicative Revenue Approximation
+
+- **ID:** `2602.10483` · **primary:** `cs.GT` · **Submitted:** 2026-02-11
+- **Authors:** Wei Tang, Yifan Wang, Mengxiao Zhang
+- **abs:** https://arxiv.org/abs/2602.10483 · **PDF:** https://arxiv.org/pdf/2602.10483v1
+
+## 13372. Protecting Context and Prompts: Deterministic Security for Non-Deterministic AI
+
+- **ID:** `2602.10481` · **primary:** `cs.CR` · **Submitted:** 2026-02-11
+- **Authors:** Mohan Rajagopalan, Vinay Rao
+- **abs:** https://arxiv.org/abs/2602.10481 · **PDF:** https://arxiv.org/pdf/2602.10481v1
+
+## 13373. Neuro-Symbolic Synergy for Interactive World Modeling
+
+- **ID:** `2602.10480` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Hongyu Zhao, Siyu Zhou, Haolin Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.10480 · **PDF:** https://arxiv.org/pdf/2602.10480v3
+
+## 13374. GPU-Fuzz: Finding Memory Errors in Deep Learning Frameworks
+
+- **ID:** `2602.10478` · **primary:** `cs.CR` · **Submitted:** 2026-02-11
+- **Authors:** Zihao Li, Hongyi Lu, Yanan Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.10478 · **PDF:** https://arxiv.org/pdf/2602.10478v3
+
+## 13375. Driving Reaction Trajectories via Latent Flow Matching
+
+- **ID:** `2602.10476` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Yili Shen, Xiangliang Zhang
+- **abs:** https://arxiv.org/abs/2602.10476 · **PDF:** https://arxiv.org/pdf/2602.10476v1
+
+## 13376. Why Human Guidance Matters in Collaborative Vibe Coding
+
+- **ID:** `2602.10473` · **primary:** `cs.HC` · **Submitted:** 2026-02-11
+- **Authors:** Haoyu Hu, Raja Marjieh, Katherine M Collins, et al.
+- **abs:** https://arxiv.org/abs/2602.10473 · **PDF:** https://arxiv.org/pdf/2602.10473v2
+
+## 13377. TestExplora: Benchmarking LLMs for Proactive Bug Discovery via Repository-Level Test Generation
+
+- **ID:** `2602.10471` · **primary:** `cs.SE` · **Submitted:** 2026-02-11
+- **Authors:** Steven Liu, Jane Luo, Xin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.10471 · **PDF:** https://arxiv.org/pdf/2602.10471v2
+
+## 13378. Online Generalized-mean Welfare Maximization: Achieving Near-Optimal Regret from Samples
+
+- **ID:** `2602.10469` · **primary:** `cs.GT` · **Submitted:** 2026-02-11
+- **Authors:** Zongjun Yang, Rachitesh Kumar, Christian Kroer
+- **abs:** https://arxiv.org/abs/2602.10469 · **PDF:** https://arxiv.org/pdf/2602.10469v1
+
+## 13379. Fragile Thoughts: How Large Language Models Handle Chain-of-Thought Perturbations
+
+- **ID:** `2603.03332` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Ashwath Vaithinathan Aravindan, Mayank Kejriwal
+- **abs:** https://arxiv.org/abs/2603.03332 · **PDF:** https://arxiv.org/pdf/2603.03332v2
+
+## 13380. MERIT Feedback Elicits Better Bargaining in LLM Negotiators
+
+- **ID:** `2602.10467` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Jihwan Oh, Murad Aghazada, Yooju Shin, et al.
+- **abs:** https://arxiv.org/abs/2602.10467 · **PDF:** https://arxiv.org/pdf/2602.10467v4
+
+## 13381. Authenticated Workflows: A Systems Approach to Protecting Agentic AI
+
+- **ID:** `2602.10465` · **primary:** `cs.CR` · **Submitted:** 2026-02-11
+- **Authors:** Mohan Rajagopalan, Vinay Rao
+- **abs:** https://arxiv.org/abs/2602.10465 · **PDF:** https://arxiv.org/pdf/2602.10465v1
+
+## 13382. Do More Predictions Improve Statistical Inference? Filtered Prediction-Powered Inference
+
+- **ID:** `2602.10464` · **primary:** `math.ST` · **Submitted:** 2026-02-11
+- **Authors:** Shirong Xu, Will Wei Sun
+- **abs:** https://arxiv.org/abs/2602.10464 · **PDF:** https://arxiv.org/pdf/2602.10464v1
+
+## 13383. Unlocked Backpropagation using Wave Scattering
+
+- **ID:** `2602.10461` · **primary:** `math.OC` · **Submitted:** 2026-02-11
+- **Authors:** Christian Pehle, Jean-Jacques Slotine
+- **abs:** https://arxiv.org/abs/2602.10461 · **PDF:** https://arxiv.org/pdf/2602.10461v1
+
+## 13384. Found-RL: foundation model-enhanced reinforcement learning for autonomous driving
+
+- **ID:** `2602.10458` · **primary:** `cs.AI` · **Submitted:** 2026-02-11
+- **Authors:** Yansong Qu, Zihao Sheng, Zilin Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.10458 · **PDF:** https://arxiv.org/pdf/2602.10458v1
+
+## 13385. Analyzing Fairness of Neural Network Prediction via Counterfactual Dataset Generation
+
+- **ID:** `2602.10457` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Brian Hyeongseok Kim, Jacqueline L. Mitchell, Chao Wang
+- **abs:** https://arxiv.org/abs/2602.10457 · **PDF:** https://arxiv.org/pdf/2602.10457v1
+
+## 13386. Compute Only Once: UG-Separation for Efficient Large Recommendation Models
+
+- **ID:** `2602.10455` · **primary:** `cs.IR` · **Submitted:** 2026-02-11
+- **Authors:** Hui Lu, Zheng Chai, Shipeng Bai, et al.
+- **abs:** https://arxiv.org/abs/2602.10455 · **PDF:** https://arxiv.org/pdf/2602.10455v1
+
+## 13387. LATA: A Tool for LLM-Assisted Translation Annotation
+
+- **ID:** `2602.10454` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Baorong Huang, Ali Asiri
+- **abs:** https://arxiv.org/abs/2602.10454 · **PDF:** https://arxiv.org/pdf/2602.10454v1
+
+## 13388. The Landscape of Prompt Injection Threats in LLM Agents: From Taxonomy to Analysis
+
+- **ID:** `2602.10453` · **primary:** `cs.CR` · **Submitted:** 2026-02-11
+- **Authors:** Peiran Wang, Xinfeng Li, Chong Xiang, et al.
+- **abs:** https://arxiv.org/abs/2602.10453 · **PDF:** https://arxiv.org/pdf/2602.10453v1
+
+## 13389. Distributed Online Convex Optimization with Nonseparable Costs and Constraints
+
+- **ID:** `2602.10452` · **primary:** `math.OC` · **Submitted:** 2026-02-11
+- **Authors:** Zhaoye Pan, Haozhe Lei, Fan Zuo, et al.
+- **abs:** https://arxiv.org/abs/2602.10452 · **PDF:** https://arxiv.org/pdf/2602.10452v2
+
+## 13390. A Multimodal Conditional Mixture Model with Distribution-Level Physics Priors
+
+- **ID:** `2602.10451` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jinkyo Han, Bahador Bahmani
+- **abs:** https://arxiv.org/abs/2602.10451 · **PDF:** https://arxiv.org/pdf/2602.10451v1
+
+## 13391. Constructing Industrial-Scale Optimization Modeling Benchmark
+
+- **ID:** `2602.10450` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Zhong Li, Hongliang Lu, Tao Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.10450 · **PDF:** https://arxiv.org/pdf/2602.10450v1
+
+## 13392. A Unified Theory of Random Projection for Influence Functions
+
+- **ID:** `2602.10449` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Pingbang Hu, Yuzheng Hu, Jiaqi W. Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.10449 · **PDF:** https://arxiv.org/pdf/2602.10449v2
+
+## 13393. End-to-End Semantic ID Generation for Generative Advertisement Recommendation
+
+- **ID:** `2602.10445` · **primary:** `cs.IR` · **Submitted:** 2026-02-11
+- **Authors:** Jie Jiang, Xinxun Zhang, Enming Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.10445 · **PDF:** https://arxiv.org/pdf/2602.10445v2
+
+## 13394. Chamfer-Linkage for Hierarchical Agglomerative Clustering
+
+- **ID:** `2602.10444` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Kishen N Gowda, Willem Fletcher, MohammadHossein Bateni, et al.
+- **abs:** https://arxiv.org/abs/2602.10444 · **PDF:** https://arxiv.org/pdf/2602.10444v1
+
+## 13395. LakeMLB: Data Lake Machine Learning Benchmark
+
+- **ID:** `2602.10441` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Feiyu Pan, Tianbin Zhang, Aoqian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.10441 · **PDF:** https://arxiv.org/pdf/2602.10441v1
+
+## 13396. SWE-MiniSandbox: Container-Free Reinforcement Learning for Building Software Engineering Agents
+
+- **ID:** `2602.11210` · **primary:** `cs.SE` · **Submitted:** 2026-02-11
+- **Authors:** Danlong Yuan, Wei Wu, Zhengren Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.11210 · **PDF:** https://arxiv.org/pdf/2602.11210v3
+
+## 13397. AudioRouter: Data Efficient Audio Understanding via RL based Dual Reasoning
+
+- **ID:** `2602.10439` · **primary:** `cs.SD` · **Submitted:** 2026-02-11
+- **Authors:** Liyang Chen, Hongkai Chen, Yujun Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.10439 · **PDF:** https://arxiv.org/pdf/2602.10439v1
+
+## 13398. Control Reinforcement Learning: Interpretable Token-Level Steering of LLMs via Sparse Autoencoder Features
+
+- **ID:** `2602.10437` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Seonglae Cho, Zekun Wu, Adriano Koshiyama
+- **abs:** https://arxiv.org/abs/2602.10437 · **PDF:** https://arxiv.org/pdf/2602.10437v2
+
+## 13399. A Dual-Stream Physics-Augmented Unsupervised Architecture for Runtime Embedded Vehicle Health Monitoring
+
+- **ID:** `2602.10432` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Enzo Nicolas Spotorno, Antonio Augusto Medeiros Frohlich
+- **abs:** https://arxiv.org/abs/2602.10432 · **PDF:** https://arxiv.org/pdf/2602.10432v1
+
+## 13400. QTALE: Quantization-Robust Token-Adaptive Layer Execution for LLMs
+
+- **ID:** `2602.10431` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Kanghyun Noh, Jinheon Choi, Yulhwa Kim
+- **abs:** https://arxiv.org/abs/2602.10431 · **PDF:** https://arxiv.org/pdf/2602.10431v3
+
+## 13401. Breaking the Curse of Repulsion: Optimistic Distributionally Robust Policy Optimization for Off-Policy Generative Recommendation
+
+- **ID:** `2602.10430` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jie Jiang, Yusen Huo, Xiangxin Zhan, et al.
+- **abs:** https://arxiv.org/abs/2602.10430 · **PDF:** https://arxiv.org/pdf/2602.10430v1
+
+## 13402. AIvilization v0: Toward Large-Scale Artificial Social Simulation with a Unified Agent Architecture and Adaptive Agent Profiles
+
+- **ID:** `2602.10429` · **primary:** `cs.MA` · **Submitted:** 2026-02-11
+- **Authors:** Wenkai Fan, Shurui Zhang, Xiaolong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.10429 · **PDF:** https://arxiv.org/pdf/2602.10429v1
+
+## 13403. Binary Flow Matching: Prediction-Loss Space Alignment for Robust Learning
+
+- **ID:** `2602.10420` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jiadong Hong, Lei Liu, Xinyu Bian, et al.
+- **abs:** https://arxiv.org/abs/2602.10420 · **PDF:** https://arxiv.org/pdf/2602.10420v1
+
+## 13404. Equivariant Evidential Deep Learning for Interatomic Potentials
+
+- **ID:** `2602.10419` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Zhongyao Wang, Taoyong Cui, Jiawen Zou, et al.
+- **abs:** https://arxiv.org/abs/2602.10419 · **PDF:** https://arxiv.org/pdf/2602.10419v1
+
+## 13405. Graph neural networks uncover structure and functions underlying the activity of simulated neural assemblies
+
+- **ID:** `2602.13325` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-11
+- **Authors:** Cédric Allier, Larissa Heinrich, Magdalena Schneider, et al.
+- **abs:** https://arxiv.org/abs/2602.13325 · **PDF:** https://arxiv.org/pdf/2602.13325v1
+
+## 13406. AI-rithmetic
+
+- **ID:** `2602.10416` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Alex Bie, Travis Dick, Alex Kulesza, et al.
+- **abs:** https://arxiv.org/abs/2602.10416 · **PDF:** https://arxiv.org/pdf/2602.10416v1
+
+## 13407. EVOKE: Emotion Vocabulary Of Korean and English
+
+- **ID:** `2602.10414` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Yoonwon Jung, Hagyeong Shin, Benjamin K. Bergen
+- **abs:** https://arxiv.org/abs/2602.10414 · **PDF:** https://arxiv.org/pdf/2602.10414v1
+
+## 13408. LightGTS-Cov: Covariate-Enhanced Time Series Forecasting
+
+- **ID:** `2602.10412` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Yong Shang, Zhipeng Yao, Ning Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.10412 · **PDF:** https://arxiv.org/pdf/2602.10412v1
+
+## 13409. LUCID: Attention with Preconditioned Representations
+
+- **ID:** `2602.10410` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Sai Surya Duvvuri, Nirmal Patel, Nilesh Gupta, et al.
+- **abs:** https://arxiv.org/abs/2602.10410 · **PDF:** https://arxiv.org/pdf/2602.10410v1
+
+## 13410. Gated Removal of Normalization in Transformers Enables Stable Training and Efficient Inference
+
+- **ID:** `2602.10408` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Andrei Kanavalau, Carmen Amo Alonso, Sanjay Lall
+- **abs:** https://arxiv.org/abs/2602.10408 · **PDF:** https://arxiv.org/pdf/2602.10408v1
+
+## 13411. Towards Affordable, Non-Invasive Real-Time Hypoglycemia Detection Using Wearable Sensor Signals
+
+- **ID:** `2602.10407` · **primary:** `cs.HC` · **Submitted:** 2026-02-11
+- **Authors:** Lawrence Obiuwevwi, Krzysztof J. Rechowicz, Vikas Ashok, et al.
+- **abs:** https://arxiv.org/abs/2602.10407 · **PDF:** https://arxiv.org/pdf/2602.10407v1
+
+## 13412. Modular Multi-Task Learning for Chemical Reaction Prediction
+
+- **ID:** `2602.10404` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jiayun Pang, Ahmed M. Zaitoun, Xacobe Couso Cambeiro, et al.
+- **abs:** https://arxiv.org/abs/2602.10404 · **PDF:** https://arxiv.org/pdf/2602.10404v1
+
+## 13413. Experimental Demonstration of Online Learning-Based Concept Drift Adaptation for Failure Detection in Optical Networks
+
+- **ID:** `2602.10401` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Yousuf Moiz Ali, Jaroslaw E. Prilepsky, João Pedro, et al.
+- **abs:** https://arxiv.org/abs/2602.10401 · **PDF:** https://arxiv.org/pdf/2602.10401v1
+
+## 13414. When are We Worried? Temporal Trends of Anxiety and What They Reveal about Us
+
+- **ID:** `2602.10400` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Saif M. Mohammad
+- **abs:** https://arxiv.org/abs/2602.10400 · **PDF:** https://arxiv.org/pdf/2602.10400v1
+
+## 13415. Tensor Methods: A Unified and Interpretable Approach for Material Design
+
+- **ID:** `2602.10392` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Shaan Pakala, Aldair E. Gongora, Brian Giera, et al.
+- **abs:** https://arxiv.org/abs/2602.10392 · **PDF:** https://arxiv.org/pdf/2602.10392v1
+
+## 13416. Affordances Enable Partial World Modeling with LLMs
+
+- **ID:** `2602.10390` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Khimya Khetarpal, Gheorghe Comanici, Jonathan Richens, et al.
+- **abs:** https://arxiv.org/abs/2602.10390 · **PDF:** https://arxiv.org/pdf/2602.10390v1
+
+## 13417. Less is Enough: Synthesizing Diverse Data in Feature Space of LLMs
+
+- **ID:** `2602.10388` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Zhongzhi Li, Xuansheng Wu, Yijiang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.10388 · **PDF:** https://arxiv.org/pdf/2602.10388v2
+
+## 13418. Making Databases Faster with LLM Evolutionary Sampling
+
+- **ID:** `2602.10387` · **primary:** `cs.DB` · **Submitted:** 2026-02-11
+- **Authors:** Mehmet Hamza Erol, Xiangpeng Hao, Federico Bianchi, et al.
+- **abs:** https://arxiv.org/abs/2602.10387 · **PDF:** https://arxiv.org/pdf/2602.10387v1
+
+## 13419. Colorful Talks with Graphs: Human-Interpretable Graph Encodings for Large Language Models
+
+- **ID:** `2602.10386` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Angelo Zangari, Peyman Baghershahi, Sourav Medya
+- **abs:** https://arxiv.org/abs/2602.10386 · **PDF:** https://arxiv.org/pdf/2602.10386v1
+
+## 13420. Time-to-Event Transformer to Capture Timing Attention of Events in EHR Time Series
+
+- **ID:** `2602.10385` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Jia Li, Yu Hou, Rui Zhang
+- **abs:** https://arxiv.org/abs/2602.10385 · **PDF:** https://arxiv.org/pdf/2602.10385v1
+
+## 13421. When Tables Go Crazy: Evaluating Multimodal Models on French Financial Documents
+
+- **ID:** `2602.10384` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Virginie Mouilleron, Théo Lasnier, Anna Mosolova, et al.
+- **abs:** https://arxiv.org/abs/2602.10384 · **PDF:** https://arxiv.org/pdf/2602.10384v3
+
+## 13422. Triggers Hijack Language Circuits: A Mechanistic Analysis of Backdoor Behaviors in Large Language Models
+
+- **ID:** `2602.10382` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Théo Lasnier, Wissam Antoun, Francis Kulumba, et al.
+- **abs:** https://arxiv.org/abs/2602.10382 · **PDF:** https://arxiv.org/pdf/2602.10382v2
+
+## 13423. Deep learning outperforms traditional machine learning methods in predicting childhood malnutrition: evidence from survey data
+
+- **ID:** `2602.10381` · **primary:** `cs.LG` · **Submitted:** 2026-02-11
+- **Authors:** Deepak Bastola, Yang Li
+- **abs:** https://arxiv.org/abs/2602.10381 · **PDF:** https://arxiv.org/pdf/2602.10381v1
+
+## 13424. The Alignment Bottleneck in Decomposition-Based Claim Verification
+
+- **ID:** `2602.10380` · **primary:** `cs.CL` · **Submitted:** 2026-02-11
+- **Authors:** Mahmud Elahi Akhter, Federico Ruggeri, Iman Munire Bilal, et al.
+- **abs:** https://arxiv.org/abs/2602.10380 · **PDF:** https://arxiv.org/pdf/2602.10380v1
+
+## 13425. Flash-SD-KDE: Accelerating SD-KDE with Tensor Cores
+
+- **ID:** `2602.10378` · **primary:** `cs.DC` · **Submitted:** 2026-02-10
+- **Authors:** Elliot L. Epstein, Rajat Vadiraj Dwaraknath, John Winnicki
+- **abs:** https://arxiv.org/abs/2602.10378 · **PDF:** https://arxiv.org/pdf/2602.10378v1
+
+## 13426. Hardware Co-Design Scaling Laws via Roofline Modelling for On-Device LLMs
+
+- **ID:** `2602.10377` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Luoyang Sun, Jiwen Jiang, Yifeng Ding, et al.
+- **abs:** https://arxiv.org/abs/2602.10377 · **PDF:** https://arxiv.org/pdf/2602.10377v1
+
+## 13427. Simple LLM Baselines are Competitive for Model Diffing
+
+- **ID:** `2602.10371` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Elias Kempf, Simon Schrodi, Bartosz Cywiński, et al.
+- **abs:** https://arxiv.org/abs/2602.10371 · **PDF:** https://arxiv.org/pdf/2602.10371v1
+
+## 13428. Causal Effect Estimation with Learned Instrument Representations
+
+- **ID:** `2602.10370` · **primary:** `stat.ML` · **Submitted:** 2026-02-10
+- **Authors:** Frances Dean, Jenna Fields, Radhika Bhalerao, et al.
+- **abs:** https://arxiv.org/abs/2602.10370 · **PDF:** https://arxiv.org/pdf/2602.10370v1
+
+## 13429. LiveMedBench: A Contamination-Free Medical Benchmark for LLMs with Automated Rubric Evaluation
+
+- **ID:** `2602.10367` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Zhiling Yan, Dingjie Song, Zhe Fang, et al.
+- **abs:** https://arxiv.org/abs/2602.10367 · **PDF:** https://arxiv.org/pdf/2602.10367v1
+
+## 13430. ENIGMA: EEG-to-Image in 15 Minutes Using Less Than 1% of the Parameters
+
+- **ID:** `2602.10361` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-10
+- **Authors:** Reese Kneeland, Wangshu Jiang, Ugo Bruzadin Nunes, et al.
+- **abs:** https://arxiv.org/abs/2602.10361 · **PDF:** https://arxiv.org/pdf/2602.10361v1
+
+## 13431. Beyond Calibration: Confounding Pathology Limits Foundation Model Specificity in Abdominal Trauma CT
+
+- **ID:** `2602.10359` · **primary:** `eess.IV` · **Submitted:** 2026-02-10
+- **Authors:** Jineel H Raythatha, Shuchang Ye, Jeremy Hsu, et al.
+- **abs:** https://arxiv.org/abs/2602.10359 · **PDF:** https://arxiv.org/pdf/2602.10359v2
+
+## 13432. Theoretical Analysis of Contrastive Learning under Imbalanced Data: From Training Dynamics to a Pruning Solution
+
+- **ID:** `2602.10357` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Haixu Liao, Yating Zhou, Songyang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.10357 · **PDF:** https://arxiv.org/pdf/2602.10357v1
+
+## 13433. Autonomous Continual Learning of Computer-Use Agents for Environment Adaptation
+
+- **ID:** `2602.10356` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Tianci Xue, Zeyi Liao, Tianneng Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.10356 · **PDF:** https://arxiv.org/pdf/2602.10356v1
+
+## 13434. Synthesizing the Kill Chain: A Zero-Shot Framework for Target Verification and Tactical Reasoning on the Edge
+
+- **ID:** `2602.13324` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Jesse Barkley, Abraham George, Amir Barati Farimani
+- **abs:** https://arxiv.org/abs/2602.13324 · **PDF:** https://arxiv.org/pdf/2602.13324v1
+
+## 13435. Physically Interpretable AlphaEarth Foundation Model Embeddings Enable LLM-Based Land Surface Intelligence
+
+- **ID:** `2602.10354` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Mashrekur Rahman
+- **abs:** https://arxiv.org/abs/2602.10354 · **PDF:** https://arxiv.org/pdf/2602.10354v1
+
+## 13436. Learning Self-Interpretation from Interpretability Artifacts: Training Lightweight Adapters on Vector-Label Pairs
+
+- **ID:** `2602.10352` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Keenan Pepper, Alex McKenzie, Florin Pop, et al.
+- **abs:** https://arxiv.org/abs/2602.10352 · **PDF:** https://arxiv.org/pdf/2602.10352v1
+
+## 13437. Adaptive Physics Transformer with Fused Global-Local Attention for Subsurface Energy Systems
+
+- **ID:** `2602.11208` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Xin Ju, Nok Hei, Fung, et al.
+- **abs:** https://arxiv.org/abs/2602.11208 · **PDF:** https://arxiv.org/pdf/2602.11208v1
+
+## 13438. Subspace Geometry Governs Catastrophic Forgetting in Low-Rank Adaptation
+
+- **ID:** `2603.02224` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Brady Steele
+- **abs:** https://arxiv.org/abs/2603.02224 · **PDF:** https://arxiv.org/pdf/2603.02224v1
+
+## 13439. When Less Is More? Diagnosing ASR Predictions in Sardinian via Layer-Wise Decoding
+
+- **ID:** `2602.10350` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Domenico De Cristofaro, Alessandro Vietti, Marianne Pouplier, et al.
+- **abs:** https://arxiv.org/abs/2602.10350 · **PDF:** https://arxiv.org/pdf/2602.10350v1
+
+## 13440. Geometry-Aware Decoding with Wasserstein-Regularized Truncation and Mass Penalties for Large Language Models
+
+- **ID:** `2602.10346` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Arash Gholami Davoodi, Navid Rezazadeh, Seyed Pouyan Mousavi Davoudi, et al.
+- **abs:** https://arxiv.org/abs/2602.10346 · **PDF:** https://arxiv.org/pdf/2602.10346v1
+
+## 13441. Identifying Evidence-Based Nudges in Biomedical Literature with Large Language Models
+
+- **ID:** `2602.10345` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Jaydeep Chauhan, Mark Seidman, Pezhman Raeisian Parvari, et al.
+- **abs:** https://arxiv.org/abs/2602.10345 · **PDF:** https://arxiv.org/pdf/2602.10345v1
+
+## 13442. Conditional Uncertainty-Aware Political Deepfake Detection with Stochastic Convolutional Neural Networks
+
+- **ID:** `2602.10343` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Rafael-Petruţ Gardoş
+- **abs:** https://arxiv.org/abs/2602.10343 · **PDF:** https://arxiv.org/pdf/2602.10343v1
+
+## 13443. The Subjectivity of Respect in Police Traffic Stops: Modeling Community Perspectives in Body-Worn Camera Footage
+
+- **ID:** `2602.10339` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Preni Golazizian, Elnaz Rahmati, Jackson Trager, et al.
+- **abs:** https://arxiv.org/abs/2602.10339 · **PDF:** https://arxiv.org/pdf/2602.10339v2
+
+## 13444. Contrastive explanations of BDI agents
+
+- **ID:** `2602.13323` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Michael Winikoff
+- **abs:** https://arxiv.org/abs/2602.13323 · **PDF:** https://arxiv.org/pdf/2602.13323v1
+
+## 13445. Generalized Prediction-Powered Inference, with Application to Binary Classifier Evaluation
+
+- **ID:** `2602.10332` · **primary:** `stat.ME` · **Submitted:** 2026-02-10
+- **Authors:** Runjia Zou, Daniela Witten, Brian Williamson
+- **abs:** https://arxiv.org/abs/2602.10332 · **PDF:** https://arxiv.org/pdf/2602.10332v1
+
+## 13446. Efficient reduction of stellar contamination and noise in planetary transmission spectra using neural networks
+
+- **ID:** `2602.10330` · **primary:** `astro-ph.EP` · **Submitted:** 2026-02-10
+- **Authors:** David S. Duque-Castaño, Lauren Flor-Torres, Jorge I. Zuluaga
+- **abs:** https://arxiv.org/abs/2602.10330 · **PDF:** https://arxiv.org/pdf/2602.10330v2
+
+## 13447. Are More Tokens Rational? Inference-Time Scaling in Language Models as Adaptive Resource Rationality
+
+- **ID:** `2602.10329` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Zhimin Hu, Riya Roshan, Sashank Varma
+- **abs:** https://arxiv.org/abs/2602.10329 · **PDF:** https://arxiv.org/pdf/2602.10329v1
+
+## 13448. Diagnostic Benchmarks for Invariant Learning Dynamics: Empirical Validation of the Eidos Architecture
+
+- **ID:** `2602.13322` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Datorien L. Anderson
+- **abs:** https://arxiv.org/abs/2602.13322 · **PDF:** https://arxiv.org/pdf/2602.13322v1
+
+## 13449. Flow Matching with Uncertainty Quantification and Guidance
+
+- **ID:** `2602.10326` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Juyeop Han, Lukas Lao Beyer, Sertac Karaman
+- **abs:** https://arxiv.org/abs/2602.10326 · **PDF:** https://arxiv.org/pdf/2602.10326v1
+
+## 13450. Discovering Differences in Strategic Behavior Between Humans and LLMs
+
+- **ID:** `2602.10324` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Caroline Wang, Daniel Kasenberg, Kim Stachenfeld, et al.
+- **abs:** https://arxiv.org/abs/2602.10324 · **PDF:** https://arxiv.org/pdf/2602.10324v1
+
+## 13451. Detecting Jailbreak Attempts in Clinical Training LLMs Through Automated Linguistic Feature Extraction
+
+- **ID:** `2602.13321` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Tri Nguyen, Huy Hoang Bao Le, Lohith Srikanth Pentapalli, et al.
+- **abs:** https://arxiv.org/abs/2602.13321 · **PDF:** https://arxiv.org/pdf/2602.13321v1
+
+## 13452. Certainty-Validity: A Diagnostic Framework for Discrete Commitment Systems
+
+- **ID:** `2603.00070` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Datorien L. Anderson
+- **abs:** https://arxiv.org/abs/2603.00070 · **PDF:** https://arxiv.org/pdf/2603.00070v1
+
+## 13453. Stop Training for the Worst: Progressive Unmasking Accelerates Masked Diffusion Training
+
+- **ID:** `2602.10314` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Jaeyeon Kim, Jonathan Geuter, David Alvarez-Melis, et al.
+- **abs:** https://arxiv.org/abs/2602.10314 · **PDF:** https://arxiv.org/pdf/2602.10314v1
+
+## 13454. R2RAG-Flood: A reasoning-reinforced training-free retrieval augmentation generation framework for flood damage nowcasting
+
+- **ID:** `2602.10312` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Lipai Huang, Kai Yin, Chia-Fu Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.10312 · **PDF:** https://arxiv.org/pdf/2602.10312v1
+
+## 13455. Confounding Robust Continuous Control via Automatic Reward Shaping
+
+- **ID:** `2602.10305` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Mateo Juliani, Mingxuan Li, Elias Bareinboim
+- **abs:** https://arxiv.org/abs/2602.10305 · **PDF:** https://arxiv.org/pdf/2602.10305v1
+
+## 13456. ICODEN: Ordinary Differential Equation Neural Networks for Interval-Censored Data
+
+- **ID:** `2602.10303` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Haoling Wang, Lang Zeng, Tao Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.10303 · **PDF:** https://arxiv.org/pdf/2602.10303v1
+
+## 13457. Configuration-to-Performance Scaling Law with Neural Ansatz
+
+- **ID:** `2602.10300` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Huaqing Zhang, Kaiyue Wen, Tengyu Ma
+- **abs:** https://arxiv.org/abs/2602.10300 · **PDF:** https://arxiv.org/pdf/2602.10300v1
+
+## 13458. On Emergent Social World Models -- Evidence for Functional Integration of Theory of Mind and Pragmatic Reasoning in Language Models
+
+- **ID:** `2602.10298` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Polina Tsvilodub, Jan-Felix Klumpp, Amir Mohammadpour, et al.
+- **abs:** https://arxiv.org/abs/2602.10298 · **PDF:** https://arxiv.org/pdf/2602.10298v1
+
+## 13459. ECHO: An Open Research Platform for Evaluation of Chat, Human Behavior, and Outcomes
+
+- **ID:** `2602.10295` · **primary:** `cs.HC` · **Submitted:** 2026-02-10
+- **Authors:** Jiqun Liu, Nischal Dinesh, Ran Yu
+- **abs:** https://arxiv.org/abs/2602.10295 · **PDF:** https://arxiv.org/pdf/2602.10295v1
+
+## 13460. Information Fidelity in Tool-Using LLM Agents: A Martingale Analysis of the Model Context Protocol
+
+- **ID:** `2602.13320` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Flint Xiaofeng Fan, Cheston Tan, Roger Wattenhofer, et al.
+- **abs:** https://arxiv.org/abs/2602.13320 · **PDF:** https://arxiv.org/pdf/2602.13320v1
+
+## 13461. What Does Preference Learning Recover from Pairwise Comparison Data?
+
+- **ID:** `2602.10286` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Rattana Pukdee, Maria-Florina Balcan, Pradeep Ravikumar
+- **abs:** https://arxiv.org/abs/2602.10286 · **PDF:** https://arxiv.org/pdf/2602.10286v1
+
+## 13462. Situation Graph Prediction: Structured Perspective Inference for User Modeling
+
+- **ID:** `2602.13319` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Jisung Shin, Daniel Platnick, Marjan Alirezaie, et al.
+- **abs:** https://arxiv.org/abs/2602.13319 · **PDF:** https://arxiv.org/pdf/2602.13319v1
+
+## 13463. Linear-LLM-SCM: Benchmarking LLMs for Coefficient Elicitation in Linear-Gaussian Causal Models
+
+- **ID:** `2602.10282` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Kanta Yamaoka, Sumantrak Mukherjee, Thomas Gärtner, et al.
+- **abs:** https://arxiv.org/abs/2602.10282 · **PDF:** https://arxiv.org/pdf/2602.10282v1
+
+## 13464. ERGO: Excess-Risk-Guided Optimization for High-Fidelity Monocular 3D Gaussian Splatting
+
+- **ID:** `2602.10278` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Zehua Ma, Hanhui Li, Zhenyu Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.10278 · **PDF:** https://arxiv.org/pdf/2602.10278v1
+
+## 13465. Power-SMC: Low-Latency Sequence-Level Power Sampling for Training-Free LLM Reasoning
+
+- **ID:** `2602.10273` · **primary:** `stat.ML` · **Submitted:** 2026-02-10
+- **Authors:** Seyedarmin Azizi, Erfan Baghaei Potraghloo, Minoo Ahmadi, et al.
+- **abs:** https://arxiv.org/abs/2602.10273 · **PDF:** https://arxiv.org/pdf/2602.10273v2
+
+## 13466. MLDocRAG: Multimodal Long-Context Document Retrieval Augmented Generation
+
+- **ID:** `2602.10271` · **primary:** `cs.IR` · **Submitted:** 2026-02-10
+- **Authors:** Yongyue Zhang, Yaxiong Wu
+- **abs:** https://arxiv.org/abs/2602.10271 · **PDF:** https://arxiv.org/pdf/2602.10271v3
+
+## 13467. Red Teaming LLMs as Socio-Technical Practice: From Exploration and Data Creation to Evaluation
+
+- **ID:** `2602.18483` · **primary:** `cs.CY` · **Submitted:** 2026-02-10
+- **Authors:** Adriana Alvarado Garcia, Ruyuan Wan, Ozioma C. Oguine, et al.
+- **abs:** https://arxiv.org/abs/2602.18483 · **PDF:** https://arxiv.org/pdf/2602.18483v1
+
+## 13468. From Classical to Topological Neural Networks Under Uncertainty
+
+- **ID:** `2602.10266` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Sarah Harkins Dayton, Layal Bou Hamdan, Ioannis D. Schizas, et al.
+- **abs:** https://arxiv.org/abs/2602.10266 · **PDF:** https://arxiv.org/pdf/2602.10266v1
+
+## 13469. Kernel-Based Learning of Chest X-ray Images for Predicting ICU Escalation among COVID-19 Patients
+
+- **ID:** `2602.10261` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Qiyuan Shi, Jian Kang, Yi Li
+- **abs:** https://arxiv.org/abs/2602.10261 · **PDF:** https://arxiv.org/pdf/2602.10261v1
+
+## 13470. The Complexity of Bayesian Network Learning: Revisiting the Superstructure
+
+- **ID:** `2602.10253` · **primary:** `cs.DS` · **Submitted:** 2026-02-10
+- **Authors:** Robert Ganian, Viktoriia Korchemna
+- **abs:** https://arxiv.org/abs/2602.10253 · **PDF:** https://arxiv.org/pdf/2602.10253v1
+
+## 13471. Modeling Programming Skills with Source Code Embeddings for Context-aware Exercise Recommendation
+
+- **ID:** `2602.10249` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Carlos Eduardo P. Silva, João Pedro M. Sena, Julio C. S. Reis, et al.
+- **abs:** https://arxiv.org/abs/2602.10249 · **PDF:** https://arxiv.org/pdf/2602.10249v1
+
+## 13472. DECKBench: Benchmarking Multi-Agent Frameworks for Academic Slide Generation and Editing
+
+- **ID:** `2602.13318` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Daesik Jang, Morgan Lindsay Heisler, Linzi Xing, et al.
+- **abs:** https://arxiv.org/abs/2602.13318 · **PDF:** https://arxiv.org/pdf/2602.13318v1
+
+## 13473. KORAL: Knowledge Graph Guided LLM Reasoning for SSD Operational Analysis
+
+- **ID:** `2602.10246` · **primary:** `cs.DC` · **Submitted:** 2026-02-10
+- **Authors:** Mayur Akewar, Sandeep Madireddy, Dongsheng Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.10246 · **PDF:** https://arxiv.org/pdf/2602.10246v1
+
+## 13474. Learning to Evict from Key-Value Cache
+
+- **ID:** `2602.10238` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Luca Moschella, Laura Manduchi, Ozan Sener
+- **abs:** https://arxiv.org/abs/2602.10238 · **PDF:** https://arxiv.org/pdf/2602.10238v1
+
+## 13475. Transforming Policy-Car Swerving for Mitigating Stop-and-Go Traffic Waves: A Practice-Oriented Jam-Absorption Driving Strategy
+
+- **ID:** `2602.10234` · **primary:** `physics.soc-ph` · **Submitted:** 2026-02-10
+- **Authors:** Zhengbing He
+- **abs:** https://arxiv.org/abs/2602.10234 · **PDF:** https://arxiv.org/pdf/2602.10234v2
+
+## 13476. ImprovEvolve: Ask AlphaEvolve to Improve the Input Solution and Then Improvise
+
+- **ID:** `2602.10233` · **primary:** `cs.NE` · **Submitted:** 2026-02-10
+- **Authors:** Alexey Kravatskiy, Valentin Khrulkov, Ivan Oseledets
+- **abs:** https://arxiv.org/abs/2602.10233 · **PDF:** https://arxiv.org/pdf/2602.10233v1
+
+## 13477. Risk-Equalized Differentially Private Synthetic Data: Protecting Outliers by Controlling Record-Level Influence
+
+- **ID:** `2602.10232` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Amir Asiaee, Chao Yan, Zachary B. Abrams, et al.
+- **abs:** https://arxiv.org/abs/2602.10232 · **PDF:** https://arxiv.org/pdf/2602.10232v1
+
+## 13478. Blockwise Advantage Estimation for Multi-Objective RL with Verifiable Rewards
+
+- **ID:** `2602.10231` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Kirill Pavlenko, Alexander Golubev, Simon Karasik, et al.
+- **abs:** https://arxiv.org/abs/2602.10231 · **PDF:** https://arxiv.org/pdf/2602.10231v1
+
+## 13479. The Global Landscape of Environmental AI Regulation: From the Cost of Reasoning to a Right to Green AI
+
+- **ID:** `2603.00068` · **primary:** `cs.CY` · **Submitted:** 2026-02-10
+- **Authors:** Kai Ebert, Boris Gamazaychikov, Philipp Hacker, et al.
+- **abs:** https://arxiv.org/abs/2603.00068 · **PDF:** https://arxiv.org/pdf/2603.00068v1
+
+## 13480. Frame-Level Internal Tool Use for Temporal Grounding in Audio LMs
+
+- **ID:** `2602.10230` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Joesph An, Phillip Keung, Jiaqi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.10230 · **PDF:** https://arxiv.org/pdf/2602.10230v1
+
+## 13481. Latent Thoughts Tuning: Bridging Context and Reasoning with Fused Information in Latent Tokens
+
+- **ID:** `2602.10229` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Weihao Liu, Dehai Min, Lu Cheng
+- **abs:** https://arxiv.org/abs/2602.10229 · **PDF:** https://arxiv.org/pdf/2602.10229v1
+
+## 13482. PRISM: Differentially Private Synthetic Data with Structure-Aware Budget Allocation for Prediction
+
+- **ID:** `2602.10228` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Amir Asiaee, Chao Yan, Zachary B. Abrams, et al.
+- **abs:** https://arxiv.org/abs/2602.10228 · **PDF:** https://arxiv.org/pdf/2602.10228v1
+
+## 13483. Self-Evolving Recommendation System: End-To-End Autonomous Model Optimization With LLM Agents
+
+- **ID:** `2602.10226` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Haochen Wang, Yi Wu, Daryl Chang, et al.
+- **abs:** https://arxiv.org/abs/2602.10226 · **PDF:** https://arxiv.org/pdf/2602.10226v1
+
+## 13484. Quantum Integrated Sensing and Computation with Indefinite Causal Order
+
+- **ID:** `2602.10225` · **primary:** `quant-ph` · **Submitted:** 2026-02-10
+- **Authors:** Ivana Nikoloska
+- **abs:** https://arxiv.org/abs/2602.10225 · **PDF:** https://arxiv.org/pdf/2602.10225v1
+
+## 13485. Internalizing Meta-Experience into Memory for Guided Reinforcement Learning in Large Language Models
+
+- **ID:** `2602.10224` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Shiting Huang, Zecheng Li, Yu Zeng, et al.
+- **abs:** https://arxiv.org/abs/2602.10224 · **PDF:** https://arxiv.org/pdf/2602.10224v1
+
+## 13486. ACE-RTL: When Agentic Context Evolution Meets RTL-Specialized LLMs
+
+- **ID:** `2602.10218` · **primary:** `cs.AR` · **Submitted:** 2026-02-10
+- **Authors:** Chenhui Deng, Zhongzhi Yu, Guan-Ting Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.10218 · **PDF:** https://arxiv.org/pdf/2602.10218v2
+
+## 13487. Temper-Then-Tilt: Principled Unlearning for Generative Models through Tempering and Classifier Guidance
+
+- **ID:** `2602.10217` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Jacob L. Block, Mehryar Mohri, Aryan Mokhtari, et al.
+- **abs:** https://arxiv.org/abs/2602.10217 · **PDF:** https://arxiv.org/pdf/2602.10217v1
+
+## 13488. ELROND: Exploring and decomposing intrinsic capabilities of diffusion models
+
+- **ID:** `2602.10216` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Paweł Skierś, Tomasz Trzciński, Kamil Deja
+- **abs:** https://arxiv.org/abs/2602.10216 · **PDF:** https://arxiv.org/pdf/2602.10216v1
+
+## 13489. Rank-Accuracy Trade-off for LoRA: A Gradient-Flow Analysis
+
+- **ID:** `2602.10212` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Michael Rushka, Diego Klabjan
+- **abs:** https://arxiv.org/abs/2602.10212 · **PDF:** https://arxiv.org/pdf/2602.10212v1
+
+## 13490. How Much Reasoning Do Retrieval-Augmented Models Add beyond LLMs? A Benchmarking Framework for Multi-Hop Inference over Hybrid Knowledge
+
+- **ID:** `2602.10210` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Junhong Lin, Bing Zhang, Song Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.10210 · **PDF:** https://arxiv.org/pdf/2602.10210v1
+
+## 13491. Neural Network Quantum Field Theory from Transformer Architectures
+
+- **ID:** `2602.10209` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Dmitry S. Ageev, Yulia A. Ageeva
+- **abs:** https://arxiv.org/abs/2602.10209 · **PDF:** https://arxiv.org/pdf/2602.10209v1
+
+## 13492. Adaptive Optimization via Momentum on Variance-Normalized Gradients
+
+- **ID:** `2602.10204` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Francisco Patitucci, Aryan Mokhtari
+- **abs:** https://arxiv.org/abs/2602.10204 · **PDF:** https://arxiv.org/pdf/2602.10204v1
+
+## 13493. Versor: A Geometric Sequence Architecture
+
+- **ID:** `2602.10195` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Truong Minh Huy, Edward Hirst
+- **abs:** https://arxiv.org/abs/2602.10195 · **PDF:** https://arxiv.org/pdf/2602.10195v2
+
+## 13494. Signature-Kernel Based Evaluation Metrics for Robust Probabilistic and Tail-Event Forecasting
+
+- **ID:** `2602.10182` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Benjamin R. Redhead, Thomas L. Lee, Peng Gu, et al.
+- **abs:** https://arxiv.org/abs/2602.10182 · **PDF:** https://arxiv.org/pdf/2602.10182v1
+
+## 13495. Biases in the Blind Spot: Detecting What LLMs Fail to Mention
+
+- **ID:** `2602.10117` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Iván Arcuschin, David Chanin, Adrià Garriga-Alonso, et al.
+- **abs:** https://arxiv.org/abs/2602.10117 · **PDF:** https://arxiv.org/pdf/2602.10117v4
+
+## 13496. When the Prompt Becomes Visual: Vision-Centric Jailbreak Attacks for Large Image Editing Models
+
+- **ID:** `2602.10179` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Jiacheng Hou, Yining Sun, Ruochong Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.10179 · **PDF:** https://arxiv.org/pdf/2602.10179v1
+
+## 13497. Olaf-World: Orienting Latent Actions for Video World Modeling
+
+- **ID:** `2602.10104` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Yuxin Jiang, Yuchao Gu, Ivor W. Tsang, et al.
+- **abs:** https://arxiv.org/abs/2602.10104 · **PDF:** https://arxiv.org/pdf/2602.10104v1
+
+## 13498. Towards Explainable Federated Learning: Understanding the Impact of Differential Privacy
+
+- **ID:** `2602.10100` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Júlio Oliveira, Rodrigo Ferreira, André Riker, et al.
+- **abs:** https://arxiv.org/abs/2602.10100 · **PDF:** https://arxiv.org/pdf/2602.10100v1
+
+## 13499. Learning on the Manifold: Unlocking Standard Diffusion Transformers with Representation Encoders
+
+- **ID:** `2602.10099` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Amandeep Kumar, Vishal M. Patel
+- **abs:** https://arxiv.org/abs/2602.10099 · **PDF:** https://arxiv.org/pdf/2602.10099v1
+
+## 13500. Step-resolved data attribution for looped transformers
+
+- **ID:** `2602.10097` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Georgios Kaissis, David Mildenberger, Juan Felipe Gomez, et al.
+- **abs:** https://arxiv.org/abs/2602.10097 · **PDF:** https://arxiv.org/pdf/2602.10097v1
+
+## 13501. Causality in Video Diffusers is Separable from Denoising
+
+- **ID:** `2602.10095` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Xingjian Bai, Guande He, Zhengqi Li, et al.
+- **abs:** https://arxiv.org/abs/2602.10095 · **PDF:** https://arxiv.org/pdf/2602.10095v1
+
+## 13502. Quantum-Audit: Evaluating the Reasoning Limits of LLMs on Quantum Computing
+
+- **ID:** `2602.10092` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Mohamed Afane, Kayla Laufer, Wenqi Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.10092 · **PDF:** https://arxiv.org/pdf/2602.10092v1
+
+## 13503. Agent World Model: Infinity Synthetic Environments for Agentic Reinforcement Learning
+
+- **ID:** `2602.10090` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Zhaoyang Wang, Canwen Xu, Boyi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.10090 · **PDF:** https://arxiv.org/pdf/2602.10090v2
+
+## 13504. CODE-SHARP: Continuous Open-ended Discovery and Evolution of Skills as Hierarchical Reward Programs
+
+- **ID:** `2602.10085` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Richard Bornemann, Pierluigi Vito Amadori, Antoine Cully
+- **abs:** https://arxiv.org/abs/2602.10085 · **PDF:** https://arxiv.org/pdf/2602.10085v2
+
+## 13505. Towards Autonomous Mathematics Research
+
+- **ID:** `2602.10177` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Tony Feng, Trieu H. Trinh, Garrett Bingham, et al.
+- **abs:** https://arxiv.org/abs/2602.10177 · **PDF:** https://arxiv.org/pdf/2602.10177v3
+
+## 13506. Anagent For Enhancing Scientific Table & Figure Analysis
+
+- **ID:** `2602.10081` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Xuehang Guo, Zhiyong Lu, Tom Hope, et al.
+- **abs:** https://arxiv.org/abs/2602.10081 · **PDF:** https://arxiv.org/pdf/2602.10081v2
+
+## 13507. PulseLM: A Foundation Dataset and Benchmark for PPG-Text Learning
+
+- **ID:** `2603.03331` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Hung Manh Pham, Jinyang Wu, Xiao Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.03331 · **PDF:** https://arxiv.org/pdf/2603.03331v1
+
+## 13508. CAPID: Context-Aware PII Detection for Question-Answering Systems
+
+- **ID:** `2602.10074` · **primary:** `cs.CR` · **Submitted:** 2026-02-10
+- **Authors:** Mariia Ponomarenko, Sepideh Abedini, Masoumeh Shafieinejad, et al.
+- **abs:** https://arxiv.org/abs/2602.10074 · **PDF:** https://arxiv.org/pdf/2602.10074v1
+
+## 13509. Dissecting Performative Prediction: A Comprehensive Survey
+
+- **ID:** `2602.10176` · **primary:** `stat.ML` · **Submitted:** 2026-02-10
+- **Authors:** Thomas Kehrenberg, Javier Sanguino, Jose A. Lozano, et al.
+- **abs:** https://arxiv.org/abs/2602.10176 · **PDF:** https://arxiv.org/pdf/2602.10176v1
+
+## 13510. Characterizing and Predicting Wildfire Evacuation Behavior: A Dual-Stage ML Approach
+
+- **ID:** `2603.02223` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Sazzad Bin Bashar Polock, Anandi Dutta, Subasish Das
+- **abs:** https://arxiv.org/abs/2603.02223 · **PDF:** https://arxiv.org/pdf/2603.02223v1
+
+## 13511. Features as Rewards: Scalable Supervision for Open-Ended Tasks via Interpretability
+
+- **ID:** `2602.10067` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Aaditya Vikram Prasad, Connor Watts, Jack Merullo, et al.
+- **abs:** https://arxiv.org/abs/2602.10067 · **PDF:** https://arxiv.org/pdf/2602.10067v3
+
+## 13512. Boltzmann Generators for Condensed Matter via Riemannian Flow Matching
+
+- **ID:** `2602.18482` · **primary:** `physics.comp-ph` · **Submitted:** 2026-02-10
+- **Authors:** Emil Hoffmann, Maximilian Schebek, Leon Klein, et al.
+- **abs:** https://arxiv.org/abs/2602.18482 · **PDF:** https://arxiv.org/pdf/2602.18482v2
+
+## 13513. Chain of Mindset: Reasoning with Adaptive Cognitive Modes
+
+- **ID:** `2602.10063` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Tianyi Jiang, Arctanx An, Hengyi Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.10063 · **PDF:** https://arxiv.org/pdf/2602.10063v2
+
+## 13514. Vendi Novelty Scores for Out-of-Distribution Detection
+
+- **ID:** `2602.10062` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Amey P. Pasarkar, Adji Bousso Dieng
+- **abs:** https://arxiv.org/abs/2602.10062 · **PDF:** https://arxiv.org/pdf/2602.10062v1
+
+## 13515. Evaluating Disentangled Representations for Controllable Music Generation
+
+- **ID:** `2602.10058` · **primary:** `cs.SD` · **Submitted:** 2026-02-10
+- **Authors:** Laura Ibáñez-Martínez, Chukwuemeka Nkama, Andrea Poltronieri, et al.
+- **abs:** https://arxiv.org/abs/2602.10058 · **PDF:** https://arxiv.org/pdf/2602.10058v2
+
+## 13516. WildCat: Near-Linear Attention in Theory and Practice
+
+- **ID:** `2602.10056` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Tobias Schröder, Lester Mackey
+- **abs:** https://arxiv.org/abs/2602.10056 · **PDF:** https://arxiv.org/pdf/2602.10056v1
+
+## 13517. UltraLIF: Fully Differentiable Spiking Neural Networks via Ultradiscretization and Max-Plus Algebra
+
+- **ID:** `2602.11206` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Jose Marie Antonio Miñoza
+- **abs:** https://arxiv.org/abs/2602.11206 · **PDF:** https://arxiv.org/pdf/2602.11206v1
+
+## 13518. Long Chain-of-Thought Compression via Fine-Grained Group Policy Optimization
+
+- **ID:** `2602.10048` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Xinchen Han, Hossam Afifi, Michel Marot, et al.
+- **abs:** https://arxiv.org/abs/2602.10048 · **PDF:** https://arxiv.org/pdf/2602.10048v2
+
+## 13519. Conformal Prediction Sets for Instance Segmentation
+
+- **ID:** `2602.10045` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Kerri Lu, Dan M. Kluger, Stephen Bates, et al.
+- **abs:** https://arxiv.org/abs/2602.10045 · **PDF:** https://arxiv.org/pdf/2602.10045v1
+
+## 13520. Optimistic World Models: Efficient Exploration in Model-Based Deep Reinforcement Learning
+
+- **ID:** `2602.10044` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Akshay Mete, Shahid Aamir Sheikh, Tzu-Hsiang Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.10044 · **PDF:** https://arxiv.org/pdf/2602.10044v1
+
+## 13521. Fake-HR1: Rethinking Reasoning of Vision Language Model for Synthetic Image Detection
+
+- **ID:** `2602.10042` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Changjiang Jiang, Xinkuan Sha, Fengchang Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.10042 · **PDF:** https://arxiv.org/pdf/2602.10042v2
+
+## 13522. Certainty robustness: Evaluating LLM stability under self-challenging prompts
+
+- **ID:** `2603.03330` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Mohammadreza Saadat, Steve Nemzer
+- **abs:** https://arxiv.org/abs/2603.03330 · **PDF:** https://arxiv.org/pdf/2603.03330v1
+
+## 13523. Effectiveness of Binary Autoencoders for QUBO-Based Optimization Problems
+
+- **ID:** `2602.10037` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Tetsuro Abe, Masashi Yamashita, Shu Tanaka
+- **abs:** https://arxiv.org/abs/2602.10037 · **PDF:** https://arxiv.org/pdf/2602.10037v1
+
+## 13524. Position: Message-passing and spectral GNNs are two sides of the same coin
+
+- **ID:** `2602.10031` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Antonis Vasileiou, Juan Cervino, Pascal Frossard, et al.
+- **abs:** https://arxiv.org/abs/2602.10031 · **PDF:** https://arxiv.org/pdf/2602.10031v1
+
+## 13525. Overview of the TREC 2025 RAGTIME Track
+
+- **ID:** `2602.10024` · **primary:** `cs.IR` · **Submitted:** 2026-02-10
+- **Authors:** Dawn Lawrie, Sean MacAvaney, James Mayfield, et al.
+- **abs:** https://arxiv.org/abs/2602.10024 · **PDF:** https://arxiv.org/pdf/2602.10024v1
+
+## 13526. MEVER: Multi-Modal and Explainable Claim Verification with Graph-based Evidence Retrieval
+
+- **ID:** `2602.10023` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Delvin Ce Zhang, Suhan Cui, Zhelin Chu, et al.
+- **abs:** https://arxiv.org/abs/2602.10023 · **PDF:** https://arxiv.org/pdf/2602.10023v1
+
+## 13527. Decoupled Reasoning with Implicit Fact Tokens (DRIFT): A Dual-Model Framework for Efficient Long-Context Inference
+
+- **ID:** `2602.10021` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Wenxuan Xie, Yujia Wang, Xin Tan, et al.
+- **abs:** https://arxiv.org/abs/2602.10021 · **PDF:** https://arxiv.org/pdf/2602.10021v1
+
+## 13528. ADORA: Training Reasoning Models with Dynamic Advantage Estimation on Reinforcement Learning
+
+- **ID:** `2602.10019` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Qingnan Ren, Shiting Huang, Zhen Fang, et al.
+- **abs:** https://arxiv.org/abs/2602.10019 · **PDF:** https://arxiv.org/pdf/2602.10019v1
+
+## 13529. Online Selective Conformal Prediction with Asymmetric Rules: A Permutation Test Approach
+
+- **ID:** `2602.10018` · **primary:** `stat.ME` · **Submitted:** 2026-02-10
+- **Authors:** Mingyi Zheng, Ying Jin
+- **abs:** https://arxiv.org/abs/2602.10018 · **PDF:** https://arxiv.org/pdf/2602.10018v1
+
+## 13530. SCORE: Specificity, Context Utilization, Robustness, and Relevance for Reference-Free LLM Evaluation
+
+- **ID:** `2602.10017` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Homaira Huda Shomee, Rochana Chaturvedi, Yangxinyu Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.10017 · **PDF:** https://arxiv.org/pdf/2602.10017v1
+
+## 13531. Kunlun: Establishing Scaling Laws for Massive-Scale Recommendation Systems through Unified Architecture Design
+
+- **ID:** `2602.10016` · **primary:** `cs.IR` · **Submitted:** 2026-02-10
+- **Authors:** Bojian Hou, Xiaolong Liu, Xiaoyi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.10016 · **PDF:** https://arxiv.org/pdf/2602.10016v2
+
+## 13532. RoboSubtaskNet: Temporal Sub-task Segmentation for Human-to-Robot Skill Transfer in Real-World Environments
+
+- **ID:** `2602.10015` · **primary:** `cs.RO` · **Submitted:** 2026-02-10
+- **Authors:** Dharmendra Sharma, Archit Sharma, John Rebeiro, et al.
+- **abs:** https://arxiv.org/abs/2602.10015 · **PDF:** https://arxiv.org/pdf/2602.10015v2
+
+## 13533. A Task-Centric Theory for Iterative Self-Improvement with Easy-to-Hard Curricula
+
+- **ID:** `2602.10014` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Chenruo Liu, Yijun Dong, Yiqiu Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.10014 · **PDF:** https://arxiv.org/pdf/2602.10014v2
+
+## 13534. Discovering High Level Patterns from Simulation Traces
+
+- **ID:** `2602.10009` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Sean Memery, Kartic Subr
+- **abs:** https://arxiv.org/abs/2602.10009 · **PDF:** https://arxiv.org/pdf/2602.10009v1
+
+## 13535. A Collaborative Safety Shield for Safe and Efficient CAV Lane Changes in Congested On-Ramp Merging
+
+- **ID:** `2602.10007` · **primary:** `cs.RO` · **Submitted:** 2026-02-10
+- **Authors:** Bharathkumar Hegde, Melanie Bouroche
+- **abs:** https://arxiv.org/abs/2602.10007 · **PDF:** https://arxiv.org/pdf/2602.10007v1
+
+## 13536. Answer First, Reason Later: Aligning Search Relevance via Mode-Balanced Reinforcement Learning
+
+- **ID:** `2602.10006` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Shijie Zhang, Xiang Guo, Rujun Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.10006 · **PDF:** https://arxiv.org/pdf/2602.10006v1
+
+## 13537. ESTAR: Early-Stopping Token-Aware Reasoning For Efficient Inference
+
+- **ID:** `2602.10004` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Junda Wang, Zhichao Yang, Dongxu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.10004 · **PDF:** https://arxiv.org/pdf/2602.10004v1
+
+## 13538. ViSpeechFormer: A Phonemic Approach for Vietnamese Automatic Speech Recognition
+
+- **ID:** `2602.10003` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Khoa Anh Nguyen, Long Minh Hoang, Nghia Hieu Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.10003 · **PDF:** https://arxiv.org/pdf/2602.10003v1
+
+## 13539. A Representation-Consistent Gated Recurrent Framework for Robust Medical Time-Series Classification
+
+- **ID:** `2603.00067` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Maitri Krishna Sai
+- **abs:** https://arxiv.org/abs/2603.00067 · **PDF:** https://arxiv.org/pdf/2603.00067v1
+
+## 13540. A Unified Assessment of the Poverty of the Stimulus Argument for Neural Language Models
+
+- **ID:** `2602.09992` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Xiulin Yang, Arianna Bisazza, Nathan Schneider, et al.
+- **abs:** https://arxiv.org/abs/2602.09992 · **PDF:** https://arxiv.org/pdf/2602.09992v1
+
+## 13541. Empirical Stability Analysis of Kolmogorov-Arnold Networks in Hard-Constrained Recurrent Physics-Informed Discovery
+
+- **ID:** `2602.09988` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Enzo Nicolas Spotorno, Josafat Leal Filho, Antonio Augusto Medeiros Frohlich
+- **abs:** https://arxiv.org/abs/2602.09988 · **PDF:** https://arxiv.org/pdf/2602.09988v2
+
+## 13542. Infusion: Shaping Model Behavior by Editing Training Data via Influence Functions
+
+- **ID:** `2602.09987` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** J Rosser, Robert Kirk, Edward Grefenstette, et al.
+- **abs:** https://arxiv.org/abs/2602.09987 · **PDF:** https://arxiv.org/pdf/2602.09987v3
+
+## 13543. Online Monitoring Framework for Automotive Time Series Data using JEPA Embeddings
+
+- **ID:** `2602.09985` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Alexander Fertig, Karthikeyan Chandra Sekaran, Lakshman Balasubramanian, et al.
+- **abs:** https://arxiv.org/abs/2602.09985 · **PDF:** https://arxiv.org/pdf/2602.09985v1
+
+## 13544. Coupled Inference in Diffusion Models for Semantic Decomposition
+
+- **ID:** `2602.09983` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Calvin Yeung, Ali Zakeri, Zhuowen Zou, et al.
+- **abs:** https://arxiv.org/abs/2602.09983 · **PDF:** https://arxiv.org/pdf/2602.09983v1
+
+## 13545. Supervised Metric Regularization Through Alternating Optimization for Multi-Regime Physics-Informed Neural Networks
+
+- **ID:** `2602.09980` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Enzo Nicolas Spotorno, Josafat Ribeiro Leal, Antonio Augusto Frohlich
+- **abs:** https://arxiv.org/abs/2602.09980 · **PDF:** https://arxiv.org/pdf/2602.09980v2
+
+## 13546. Causal Identification in Multi-Task Demand Learning with Confounding
+
+- **ID:** `2602.09969` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Varun Gupta, Vijay Kamble
+- **abs:** https://arxiv.org/abs/2602.09969 · **PDF:** https://arxiv.org/pdf/2602.09969v1
+
+## 13547. Drug Release Modeling using Physics-Informed Neural Networks
+
+- **ID:** `2602.09963` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Daanish Aleem Qureshi, Khemraj Shukla, Vikas Srivastava
+- **abs:** https://arxiv.org/abs/2602.09963 · **PDF:** https://arxiv.org/pdf/2602.09963v1
+
+## 13548. ViMultiChoice: Toward a Method That Gives Explanation for Multiple-Choice Reading Comprehension in Vietnamese
+
+- **ID:** `2602.09961` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Trung Tien Cao, Lam Minh Thai, Nghia Hieu Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.09961 · **PDF:** https://arxiv.org/pdf/2602.09961v2
+
+## 13549. Statistical-Computational Trade-offs in Learning Multi-Index Models via Harmonic Analysis
+
+- **ID:** `2602.09959` · **primary:** `math.ST` · **Submitted:** 2026-02-10
+- **Authors:** Hugo Latourelle-Vigeant, Theodor Misiakiewicz
+- **abs:** https://arxiv.org/abs/2602.09959 · **PDF:** https://arxiv.org/pdf/2602.09959v1
+
+## 13550. ATTNPO: Attention-Guided Process Supervision for Efficient Reasoning
+
+- **ID:** `2602.09953` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Shuaiyi Nie, Siyu Ding, Wenyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.09953 · **PDF:** https://arxiv.org/pdf/2602.09953v1
+
+## 13551. Contesting Artificial Moral Agents
+
+- **ID:** `2603.00066` · **primary:** `cs.CY` · **Submitted:** 2026-02-10
+- **Authors:** Aisha Aijaz
+- **abs:** https://arxiv.org/abs/2603.00066 · **PDF:** https://arxiv.org/pdf/2603.00066v1
+
+## 13552. Bladder Vessel Segmentation using a Hybrid Attention-Convolution Framework
+
+- **ID:** `2602.09949` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Franziska Krauß, Matthias Ege, Zoltan Lovasz, et al.
+- **abs:** https://arxiv.org/abs/2602.09949 · **PDF:** https://arxiv.org/pdf/2602.09949v1
+
+## 13553. Closing Reasoning Gaps in Clinical Agents with Differential Reasoning Learning
+
+- **ID:** `2602.09945` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Jinsong Liu, Yuhang Jiang, Ramayya Krishnan, et al.
+- **abs:** https://arxiv.org/abs/2602.09945 · **PDF:** https://arxiv.org/pdf/2602.09945v1
+
+## 13554. Instruct2Act: From Human Instruction to Actions Sequencing and Execution via Robot Action Network for Robotic Manipulation
+
+- **ID:** `2602.09940` · **primary:** `cs.RO` · **Submitted:** 2026-02-10
+- **Authors:** Archit Sharma, Dharmendra Sharma, John Rebeiro, et al.
+- **abs:** https://arxiv.org/abs/2602.09940 · **PDF:** https://arxiv.org/pdf/2602.09940v1
+
+## 13555. GPU-Accelerated Algorithms for Graph Vector Search: Taxonomy, Empirical Study, and Research Directions
+
+- **ID:** `2602.16719` · **primary:** `cs.DB` · **Submitted:** 2026-02-10
+- **Authors:** Yaowen Liu, Xuejia Chen, Anxin Tian, et al.
+- **abs:** https://arxiv.org/abs/2602.16719 · **PDF:** https://arxiv.org/pdf/2602.16719v1
+
+## 13556. Why Do AI Agents Systematically Fail at Cloud Root Cause Analysis?
+
+- **ID:** `2602.09937` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Taeyoon Kim, Woohyeok Park, Hoyeong Yun, et al.
+- **abs:** https://arxiv.org/abs/2602.09937 · **PDF:** https://arxiv.org/pdf/2602.09937v2
+
+## 13557. The Catastrophic Failure of The k-Means Algorithm in High Dimensions, and How Hartigan's Algorithm Avoids It
+
+- **ID:** `2602.09936` · **primary:** `stat.ML` · **Submitted:** 2026-02-10
+- **Authors:** Roy R. Lederman, David Silva-Sánchez, Ziling Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.09936 · **PDF:** https://arxiv.org/pdf/2602.09936v1
+
+## 13558. Unbalanced optimal transport for robust longitudinal lesion evolution with registration-aware and appearance-guided priors
+
+- **ID:** `2602.09933` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Melika Qahqaie, Dominik Neumann, Tobias Heimann, et al.
+- **abs:** https://arxiv.org/abs/2602.09933 · **PDF:** https://arxiv.org/pdf/2602.09933v1
+
+## 13559. Self-Service or Not? How to Guide Practitioners in Classifying AI Systems Under the EU AI Act
+
+- **ID:** `2603.00065` · **primary:** `cs.CY` · **Submitted:** 2026-02-10
+- **Authors:** Ronald Schnitzer, Maximilian Hoeving, Sonja Zillner
+- **abs:** https://arxiv.org/abs/2603.00065 · **PDF:** https://arxiv.org/pdf/2603.00065v1
+
+## 13560. Monocular Normal Estimation via Shading Sequence Estimation
+
+- **ID:** `2602.09929` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Zongrui Li, Xinhua Ma, Minghui Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.09929 · **PDF:** https://arxiv.org/pdf/2602.09929v5
+
+## 13561. LLMs Encode Their Failures: Predicting Success from Pre-Generation Activations
+
+- **ID:** `2602.09924` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** William Lugoloobi, Thomas Foster, William Bankes, et al.
+- **abs:** https://arxiv.org/abs/2602.09924 · **PDF:** https://arxiv.org/pdf/2602.09924v2
+
+## 13562. SARS: A Novel Face and Body Shape and Appearance Aware 3D Reconstruction System extends Morphable Models
+
+- **ID:** `2602.09918` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Gulraiz Khan, Kenneth Y. Wertheim, Kevin Pimbblet, et al.
+- **abs:** https://arxiv.org/abs/2602.09918 · **PDF:** https://arxiv.org/pdf/2602.09918v1
+
+## 13563. Cosmo3DFlow: Wavelet Flow Matching for Spatial-to-Spectral Compression in Reconstructing the Early Universe
+
+- **ID:** `2602.10172` · **primary:** `astro-ph.IM` · **Submitted:** 2026-02-10
+- **Authors:** Md. Khairul Islam, Zeyu Xia, Ryan Goudjil, et al.
+- **abs:** https://arxiv.org/abs/2602.10172 · **PDF:** https://arxiv.org/pdf/2602.10172v1
+
+## 13564. AmharicIR+Instr: A Two-Dataset Resource for Neural Retrieval and Instruction Tuning
+
+- **ID:** `2602.09914` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Tilahun Yeshambel, Moncef Garouani, Josiane Mothe
+- **abs:** https://arxiv.org/abs/2602.09914 · **PDF:** https://arxiv.org/pdf/2602.09914v1
+
+## 13565. MedCalc-Bench Doesn't Measure What You Think: A Benchmark Audit and the Case for Open-Book Evaluation
+
+- **ID:** `2603.02222` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Artus Krohn-Grimberghe
+- **abs:** https://arxiv.org/abs/2603.02222 · **PDF:** https://arxiv.org/pdf/2603.02222v1
+
+## 13566. Self-Regulated Reading with AI Support: An Eight-Week Study with Students
+
+- **ID:** `2602.09907` · **primary:** `cs.HC` · **Submitted:** 2026-02-10
+- **Authors:** Yue Fu, Joel Wester, Niels Van Berkel, et al.
+- **abs:** https://arxiv.org/abs/2602.09907 · **PDF:** https://arxiv.org/pdf/2602.09907v1
+
+## 13567. Safeguarding Privacy: Privacy-Preserving Detection of Mind Wandering and Disengagement Using Federated Learning in Online Education
+
+- **ID:** `2602.09904` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Anna Bodonhelyi, Mengdi Wang, Efe Bozkir, et al.
+- **abs:** https://arxiv.org/abs/2602.09904 · **PDF:** https://arxiv.org/pdf/2602.09904v1
+
+## 13568. Routing, Cascades, and User Choice for LLMs
+
+- **ID:** `2602.09902` · **primary:** `cs.GT` · **Submitted:** 2026-02-10
+- **Authors:** Rafid Mahmood
+- **abs:** https://arxiv.org/abs/2602.09902 · **PDF:** https://arxiv.org/pdf/2602.09902v1
+
+## 13569. QP-OneModel: A Unified Generative LLM for Multi-Task Query Understanding in Xiaohongshu Search
+
+- **ID:** `2602.09901` · **primary:** `cs.IR` · **Submitted:** 2026-02-10
+- **Authors:** Jianzhao Huang, Xiaorui Huang, Fei Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.09901 · **PDF:** https://arxiv.org/pdf/2602.09901v1
+
+## 13570. TaCo: A Benchmark for Lossless and Lossy Codecs of Heterogeneous Tactile Data
+
+- **ID:** `2602.09893` · **primary:** `cs.RO` · **Submitted:** 2026-02-10
+- **Authors:** Zhengxue Cheng, Yan Zhao, Keyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.09893 · **PDF:** https://arxiv.org/pdf/2602.09893v1
+
+## 13571. Stemphonic: All-at-once Flexible Multi-stem Music Generation
+
+- **ID:** `2602.09891` · **primary:** `cs.SD` · **Submitted:** 2026-02-10
+- **Authors:** Shih-Lun Wu, Ge Zhu, Juan-Pablo Caceres, et al.
+- **abs:** https://arxiv.org/abs/2602.09891 · **PDF:** https://arxiv.org/pdf/2602.09891v1
+
+## 13572. The Devil Behind Moltbook: Anthropic Safety is Always Vanishing in Self-Evolving AI Societies
+
+- **ID:** `2602.09877` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Chenxu Wang, Chaozhuo Li, Songyang Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.09877 · **PDF:** https://arxiv.org/pdf/2602.09877v2
+
+## 13573. Steer2Edit: From Activation Steering to Component-Level Editing
+
+- **ID:** `2602.09870` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Chung-En Sun, Ge Yan, Zimo Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.09870 · **PDF:** https://arxiv.org/pdf/2602.09870v2
+
+## 13574. Statistical benchmarking of transformer models in low signal-to-noise time-series forecasting
+
+- **ID:** `2602.09869` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Cyril Garcia, Guillaume Remy
+- **abs:** https://arxiv.org/abs/2602.09869 · **PDF:** https://arxiv.org/pdf/2602.09869v1
+
+## 13575. Differentiable Tripartite Modularity for Clustering Heterogeneous Graphs
+
+- **ID:** `2602.09864` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Benoît Hurpeau
+- **abs:** https://arxiv.org/abs/2602.09864 · **PDF:** https://arxiv.org/pdf/2602.09864v1
+
+## 13576. MedFeat: Model-Aware and Explainability-Driven Feature Engineering with LLMs for Clinical Tabular Prediction
+
+- **ID:** `2603.02221` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Zizheng Zhang, Yiming Li, Justin Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.02221 · **PDF:** https://arxiv.org/pdf/2603.02221v1
+
+## 13577. Code2World: A GUI World Model via Renderable Code Generation
+
+- **ID:** `2602.09856` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Yuhao Zheng, Li'an Zhong, Yi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.09856 · **PDF:** https://arxiv.org/pdf/2602.09856v1
+
+## 13578. CoFEH: LLM-driven Feature Engineering Empowered by Collaborative Bayesian Hyperparameter Optimization
+
+- **ID:** `2602.09851` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Beicheng Xu, Keyao Ding, Wei Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.09851 · **PDF:** https://arxiv.org/pdf/2602.09851v1
+
+## 13579. Robust Processing and Learning: Principles, Methods, and Wireless Applications
+
+- **ID:** `2602.09848` · **primary:** `eess.SP` · **Submitted:** 2026-02-10
+- **Authors:** Shixiong Wang, Wei Dai, Li-Chun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.09848 · **PDF:** https://arxiv.org/pdf/2602.09848v1
+
+## 13580. Stabilized Maximum-Likelihood Iterative Quantum Amplitude Estimation for Structural CVaR under Correlated Random Fields
+
+- **ID:** `2602.09847` · **primary:** `stat.ML` · **Submitted:** 2026-02-10
+- **Authors:** Alireza Tabarraei
+- **abs:** https://arxiv.org/abs/2602.09847 · **PDF:** https://arxiv.org/pdf/2602.09847v1
+
+## 13581. KernelCraft: Benchmarking for Agentic Close-to-Metal Kernel Generation on Emerging Hardware
+
+- **ID:** `2603.08721` · **primary:** `cs.AR` · **Submitted:** 2026-02-10
+- **Authors:** Jiayi Nie, Haoran Wu, Yao Lai, et al.
+- **abs:** https://arxiv.org/abs/2603.08721 · **PDF:** https://arxiv.org/pdf/2603.08721v1
+
+## 13582. Step-Size Stability in Stochastic Optimization: A Theoretical Perspective
+
+- **ID:** `2602.09842` · **primary:** `math.OC` · **Submitted:** 2026-02-10
+- **Authors:** Fabian Schaipp, Robert M. Gower, Adrien Taylor
+- **abs:** https://arxiv.org/abs/2602.09842 · **PDF:** https://arxiv.org/pdf/2602.09842v1
+
+## 13583. Hybrid Responsible AI-Stochastic Approach for SLA Compliance in Multivendor 6G Networks
+
+- **ID:** `2602.09841` · **primary:** `cs.NI` · **Submitted:** 2026-02-10
+- **Authors:** Emanuel Figetakis, Ahmed Refaey Hussein
+- **abs:** https://arxiv.org/abs/2602.09841 · **PDF:** https://arxiv.org/pdf/2602.09841v1
+
+## 13584. How Do People Quantify Naturally: Evidence from Mandarin Picture Description
+
+- **ID:** `2602.09838` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Yayun Zhang, Guanyi Chen, Fahime Same, et al.
+- **abs:** https://arxiv.org/abs/2602.09838 · **PDF:** https://arxiv.org/pdf/2602.09838v1
+
+## 13585. LLM Reasoning Predicts When Models Are Right: Evidence from Coding Classroom Discourse
+
+- **ID:** `2602.09832` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Bakhtawar Ahtisham, Kirk Vanacore, Zhuqian Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.09832 · **PDF:** https://arxiv.org/pdf/2602.09832v1
+
+## 13586. From FusHa to Folk: Exploring Cross-Lingual Transfer in Arabic Language Models
+
+- **ID:** `2602.09826` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Abdulmuizz Khalak, Abderrahmane Issam, Gerasimos Spanakis
+- **abs:** https://arxiv.org/abs/2602.09826 · **PDF:** https://arxiv.org/pdf/2602.09826v2
+
+## 13587. AraModernBERT: Transtokenized Initialization and Long-Context Encoder Modeling for Arabic
+
+- **ID:** `2603.09982` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Omar Elshehy, Omer Nacar, Abdelbasset Djamai, et al.
+- **abs:** https://arxiv.org/abs/2603.09982 · **PDF:** https://arxiv.org/pdf/2603.09982v2
+
+## 13588. PlugSI: Plug-and-Play Test-Time Graph Adaptation for Spatial Interpolation
+
+- **ID:** `2602.09824` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Xuhang Wu, Zhuoxuan Liang, Wei Li, et al.
+- **abs:** https://arxiv.org/abs/2602.09824 · **PDF:** https://arxiv.org/pdf/2602.09824v1
+
+## 13589. Parallel Complex Diffusion for Scalable Time Series Generation
+
+- **ID:** `2602.17706` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Rongyao Cai, Yuxi Wan, Kexin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.17706 · **PDF:** https://arxiv.org/pdf/2602.17706v1
+
+## 13590. Covo-Audio Technical Report
+
+- **ID:** `2602.09823` · **primary:** `cs.SD` · **Submitted:** 2026-02-10
+- **Authors:** Wenfu Wang, Chenxing Li, Liqiang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.09823 · **PDF:** https://arxiv.org/pdf/2602.09823v2
+
+## 13591. Text summarization via global structure awareness
+
+- **ID:** `2602.09821` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Jiaquan Zhang, Chaoning Zhang, Shuxu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.09821 · **PDF:** https://arxiv.org/pdf/2602.09821v2
+
+## 13592. AlphaForgeBench: Benchmarking End-to-End Trading Strategy Design with Large Language Models
+
+- **ID:** `2602.18481` · **primary:** `q-fin.TR` · **Submitted:** 2026-02-10
+- **Authors:** Wentao Zhang, Mingxuan Zhao, Jincheng Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.18481 · **PDF:** https://arxiv.org/pdf/2602.18481v1
+
+## 13593. AnalyticsGPT: An LLM Workflow for Scientometric Question Answering
+
+- **ID:** `2602.09817` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Khang Ly, Georgios Cheirmpos, Adrian Raudaschl, et al.
+- **abs:** https://arxiv.org/abs/2602.09817 · **PDF:** https://arxiv.org/pdf/2602.09817v1
+
+## 13594. Efficient Unsupervised Environment Design through Hierarchical Policy Representation Learning
+
+- **ID:** `2602.09813` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Dexun Li, Sidney Tio, Pradeep Varakantham
+- **abs:** https://arxiv.org/abs/2602.09813 · **PDF:** https://arxiv.org/pdf/2602.09813v1
+
+## 13595. A Controlled Study of Double DQN and Dueling DQN Under Cross-Environment Transfer
+
+- **ID:** `2602.09810` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Azkaa Nasir, Fatima Dossa, Muhammad Ahmed Atif, et al.
+- **abs:** https://arxiv.org/abs/2602.09810 · **PDF:** https://arxiv.org/pdf/2602.09810v2
+
+## 13596. Forecasting as Rendering: A 2D Gaussian Splatting Framework for Time Series Forecasting
+
+- **ID:** `2603.02220` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Yixin Wang, Yifan Hu, Peiyuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.02220 · **PDF:** https://arxiv.org/pdf/2603.02220v1
+
+## 13597. AutoHarness: improving LLM agents by automatically synthesizing a code harness
+
+- **ID:** `2603.03329` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Xinghua Lou, Miguel Lázaro-Gredilla, Antoine Dedieu, et al.
+- **abs:** https://arxiv.org/abs/2603.03329 · **PDF:** https://arxiv.org/pdf/2603.03329v1
+
+## 13598. Decomposing Reasoning Efficiency in Large Language Models
+
+- **ID:** `2602.09805` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Daniel Kaiser, Arnoldo Frigessi, Ali Ramezani-Kebrya, et al.
+- **abs:** https://arxiv.org/abs/2602.09805 · **PDF:** https://arxiv.org/pdf/2602.09805v1
+
+## 13599. Would a Large Language Model Pay Extra for a View? Inferring Willingness to Pay from Subjective Choices
+
+- **ID:** `2602.09802` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Manon Reusens, Sofie Goethals, Toon Calders, et al.
+- **abs:** https://arxiv.org/abs/2602.09802 · **PDF:** https://arxiv.org/pdf/2602.09802v1
+
+## 13600. EvoCodeBench: A Human-Performance Benchmark for Self-Evolving LLM-Driven Coding Systems
+
+- **ID:** `2602.10171` · **primary:** `cs.SE` · **Submitted:** 2026-02-10
+- **Authors:** Wentao Zhang, Jianfeng Wang, Liheng Liang, et al.
+- **abs:** https://arxiv.org/abs/2602.10171 · **PDF:** https://arxiv.org/pdf/2602.10171v1
+
+## 13601. Symbolic Pattern Temporal Numeric Planning with Intermediate Conditions and Effects
+
+- **ID:** `2602.09798` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Matteo Cardellini, Enrico Giunchiglia
+- **abs:** https://arxiv.org/abs/2602.09798 · **PDF:** https://arxiv.org/pdf/2602.09798v1
+
+## 13602. Learning Global Hypothesis Space for Enhancing Synergistic Reasoning Chain
+
+- **ID:** `2602.09794` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Jiaquan Zhang, Chaoning Zhang, Shuxu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.09794 · **PDF:** https://arxiv.org/pdf/2602.09794v2
+
+## 13603. Fully-automated sleep staging: multicenter validation of a generalizable deep neural network for Parkinson's disease and isolated REM sleep behavior disorder
+
+- **ID:** `2602.09793` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Jesper Strøm, Casper Skjærbæk, Natasha Becker Bertelsen, et al.
+- **abs:** https://arxiv.org/abs/2602.09793 · **PDF:** https://arxiv.org/pdf/2602.09793v2
+
+## 13604. The minimal width of universal $p$-adic ReLU neural networks
+
+- **ID:** `2603.00064` · **primary:** `math.NT` · **Submitted:** 2026-02-10
+- **Authors:** Sándor Z. Kiss, Ambrus Pál
+- **abs:** https://arxiv.org/abs/2603.00064 · **PDF:** https://arxiv.org/pdf/2603.00064v1
+
+## 13605. Toeplitz Based Spectral Methods for Data-driven Dynamical Systems
+
+- **ID:** `2602.09791` · **primary:** `math.DS` · **Submitted:** 2026-02-10
+- **Authors:** Vladimir R. Kostic, Karim Lounici, Massimiliano Pontil
+- **abs:** https://arxiv.org/abs/2602.09791 · **PDF:** https://arxiv.org/pdf/2602.09791v1
+
+## 13606. EVA: Towards a universal model of the immune system
+
+- **ID:** `2602.10168` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-10
+- **Authors:** Scienta Team, Ethan Bandasack, Vincent Bouget, et al.
+- **abs:** https://arxiv.org/abs/2602.10168 · **PDF:** https://arxiv.org/pdf/2602.10168v2
+
+## 13607. When Less is More: The LLM Scaling Paradox in Context Compression
+
+- **ID:** `2602.09789` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Ruishan Guo, Yibing Liu, Guoxin Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.09789 · **PDF:** https://arxiv.org/pdf/2602.09789v2
+
+## 13608. Where Are We At with Automatic Speech Recognition for the Bambara Language?
+
+- **ID:** `2602.09785` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Seydou Diallo, Yacouba Diarra, Mamadou K. Keita, et al.
+- **abs:** https://arxiv.org/abs/2602.09785 · **PDF:** https://arxiv.org/pdf/2602.09785v1
+
+## 13609. Circuit Fingerprints: How Answer Tokens Encode Their Geometrical Path
+
+- **ID:** `2602.09784` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Andres Saurez, Neha Sengar, Dongsoo Har
+- **abs:** https://arxiv.org/abs/2602.09784 · **PDF:** https://arxiv.org/pdf/2602.09784v1
+
+## 13610. Why Linear Interpretability Works: Invariant Subspaces as a Result of Architectural Constraints
+
+- **ID:** `2602.09783` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Andres Saurez, Yousung Lee, Dongsoo Har
+- **abs:** https://arxiv.org/abs/2602.09783 · **PDF:** https://arxiv.org/pdf/2602.09783v1
+
+## 13611. Flexible Entropy Control in RLVR with Gradient-Preserving Perspective
+
+- **ID:** `2602.09782` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Kun Chen, Peng Shi, Fanfan Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.09782 · **PDF:** https://arxiv.org/pdf/2602.09782v1
+
+## 13612. Explainability in Generative Medical Diffusion Models: A Faithfulness-Based Analysis on MRI Synthesis
+
+- **ID:** `2602.09781` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Surjo Dey, Pallabi Saikia
+- **abs:** https://arxiv.org/abs/2602.09781 · **PDF:** https://arxiv.org/pdf/2602.09781v1
+
+## 13613. Self-Supervised Learning as Discrete Communication
+
+- **ID:** `2602.09764` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Kawtar Zaher, Ilyass Moummad, Olivier Buisson, et al.
+- **abs:** https://arxiv.org/abs/2602.09764 · **PDF:** https://arxiv.org/pdf/2602.09764v1
+
+## 13614. Grounding LTL Tasks in Sub-Symbolic RL Environments for Zero-Shot Generalization
+
+- **ID:** `2602.09761` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Matteo Pannacci, Andrea Fanti, Elena Umili, et al.
+- **abs:** https://arxiv.org/abs/2602.09761 · **PDF:** https://arxiv.org/pdf/2602.09761v1
+
+## 13615. Improving Interpretability of Lexical Semantic Change with Neurobiological Features
+
+- **ID:** `2602.09760` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Kohei Oda, Hiroya Takamura, Kiyoaki Shirai, et al.
+- **abs:** https://arxiv.org/abs/2602.09760 · **PDF:** https://arxiv.org/pdf/2602.09760v1
+
+## 13616. Towards Poisoning Robustness Certification for Natural Language Generation
+
+- **ID:** `2602.09757` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Mihnea Ghitu, Matthew Wicker
+- **abs:** https://arxiv.org/abs/2602.09757 · **PDF:** https://arxiv.org/pdf/2602.09757v1
+
+## 13617. Measuring What AI Systems Might Do: Towards A Measurement Science in AI
+
+- **ID:** `2603.00063` · **primary:** `cs.CY` · **Submitted:** 2026-02-10
+- **Authors:** Konstantinos Voudouris, Mirko Thalmann, Alex Kipnis, et al.
+- **abs:** https://arxiv.org/abs/2603.00063 · **PDF:** https://arxiv.org/pdf/2603.00063v1
+
+## 13618. Linear Model Extraction via Factual and Counterfactual Queries
+
+- **ID:** `2602.09748` · **primary:** `math.OC` · **Submitted:** 2026-02-10
+- **Authors:** Daan Otto, Jannis Kurtz, Dick den Hertog, et al.
+- **abs:** https://arxiv.org/abs/2602.09748 · **PDF:** https://arxiv.org/pdf/2602.09748v2
+
+## 13619. Sparse Axonal and Dendritic Delays Enable Competitive SNNs for Keyword Classification
+
+- **ID:** `2602.09746` · **primary:** `cs.NE` · **Submitted:** 2026-02-10
+- **Authors:** Younes Bouhadjar, Emre Neftci
+- **abs:** https://arxiv.org/abs/2602.09746 · **PDF:** https://arxiv.org/pdf/2602.09746v1
+
+## 13620. Semantic Waveforms for AI-Native 6G Networks
+
+- **ID:** `2602.13316` · **primary:** `cs.NI` · **Submitted:** 2026-02-10
+- **Authors:** Nour Hello, Mohamed Amine Hamoura, Francois Rivet, et al.
+- **abs:** https://arxiv.org/abs/2602.13316 · **PDF:** https://arxiv.org/pdf/2602.13316v1
+
+## 13621. Anatomy-Preserving Latent Diffusion for Generation of Brain Segmentation Masks with Ischemic Infarct
+
+- **ID:** `2602.10167` · **primary:** `eess.IV` · **Submitted:** 2026-02-10
+- **Authors:** Lucia Borrego, Vajira Thambawita, Marco Ciuffreda, et al.
+- **abs:** https://arxiv.org/abs/2602.10167 · **PDF:** https://arxiv.org/pdf/2602.10167v1
+
+## 13622. Allure of Craquelure: A Variational-Generative Approach to Crack Detection in Paintings
+
+- **ID:** `2602.09730` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Laura Paul, Holger Rauhut, Martin Burger, et al.
+- **abs:** https://arxiv.org/abs/2602.09730 · **PDF:** https://arxiv.org/pdf/2602.09730v1
+
+## 13623. ExO-PPO: an Extended Off-policy Proximal Policy Optimization Algorithm
+
+- **ID:** `2602.09726` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Hanyong Wang, Menglong Yang
+- **abs:** https://arxiv.org/abs/2602.09726 · **PDF:** https://arxiv.org/pdf/2602.09726v1
+
+## 13624. Efficient Remote Prefix Fetching with GPU-native Media ASICs
+
+- **ID:** `2602.09725` · **primary:** `cs.DC` · **Submitted:** 2026-02-10
+- **Authors:** Liang Mi, Weijun Wang, Jinghan Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.09725 · **PDF:** https://arxiv.org/pdf/2602.09725v2
+
+## 13625. Targum - A Multilingual New Testament Translation Corpus
+
+- **ID:** `2602.09724` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Maciej Rapacz, Aleksander Smywiński-Pohl
+- **abs:** https://arxiv.org/abs/2602.09724 · **PDF:** https://arxiv.org/pdf/2602.09724v2
+
+## 13626. AI-Assisted Scientific Assessment: A Case Study on Climate Change
+
+- **ID:** `2602.09723` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Christian Buck, Levke Caesar, Michelle Chen Huebscher, et al.
+- **abs:** https://arxiv.org/abs/2602.09723 · **PDF:** https://arxiv.org/pdf/2602.09723v1
+
+## 13627. Continual Learning for non-stationary regression via Memory-Efficient Replay
+
+- **ID:** `2602.09720` · **primary:** `stat.ML` · **Submitted:** 2026-02-10
+- **Authors:** Pablo García-Santaclara, Bruno Fernández-Castro, RebecaP. Díaz-Redondo, et al.
+- **abs:** https://arxiv.org/abs/2602.09720 · **PDF:** https://arxiv.org/pdf/2602.09720v1
+
+## 13628. Unsupervised Layer-Wise Dynamic Test Time Adaptation for LLMs
+
+- **ID:** `2602.09719` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Longhuan Xu, Cunjian Chen, Feng Yin
+- **abs:** https://arxiv.org/abs/2602.09719 · **PDF:** https://arxiv.org/pdf/2602.09719v1
+
+## 13629. SAQNN: Spectral Adaptive Quantum Neural Network as a Universal Approximator
+
+- **ID:** `2602.09718` · **primary:** `quant-ph` · **Submitted:** 2026-02-10
+- **Authors:** Jialiang Tang, Jialin Zhang, Xiaoming Sun
+- **abs:** https://arxiv.org/abs/2602.09718 · **PDF:** https://arxiv.org/pdf/2602.09718v1
+
+## 13630. From Lightweight CNNs to SpikeNets: Benchmarking Accuracy-Energy Tradeoffs with Pruned Spiking SqueezeNet
+
+- **ID:** `2602.09717` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Radib Bin Kabir, Tawsif Tashwar Dipto, Mehedi Ahamed, et al.
+- **abs:** https://arxiv.org/abs/2602.09717 · **PDF:** https://arxiv.org/pdf/2602.09717v1
+
+## 13631. BRAVA-GNN: Betweenness Ranking Approximation Via Degree MAss Inspired Graph Neural Network
+
+- **ID:** `2602.09716` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Justin Dachille, Aurora Rossi, Sunil Kumar Maurya, et al.
+- **abs:** https://arxiv.org/abs/2602.09716 · **PDF:** https://arxiv.org/pdf/2602.09716v1
+
+## 13632. TraceMem: Weaving Narrative Memory Schemata from User Conversational Traces
+
+- **ID:** `2602.09712` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Yiming Shu, Pei Liu, Tiange Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.09712 · **PDF:** https://arxiv.org/pdf/2602.09712v1
+
+## 13633. Physics-informed diffusion models in spectral space
+
+- **ID:** `2602.09708` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Davide Gallon, Philippe von Wurstemberger, Patrick Cheridito, et al.
+- **abs:** https://arxiv.org/abs/2602.09708 · **PDF:** https://arxiv.org/pdf/2602.09708v1
+
+## 13634. Extended Isolation Forest with feature sensitivities
+
+- **ID:** `2602.09704` · **primary:** `stat.ME` · **Submitted:** 2026-02-10
+- **Authors:** Illia Donhauzer
+- **abs:** https://arxiv.org/abs/2602.09704 · **PDF:** https://arxiv.org/pdf/2602.09704v1
+
+## 13635. Maastricht University at AMIYA: Adapting LLMs for Dialectal Arabic using Fine-tuning and MBR Decoding
+
+- **ID:** `2602.09703` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Abdulhai Alali, Abderrahmane Issam
+- **abs:** https://arxiv.org/abs/2602.09703 · **PDF:** https://arxiv.org/pdf/2602.09703v1
+
+## 13636. GenSeg-R1: RL-Driven Vision-Language Grounding for Fine-Grained Referring Segmentation
+
+- **ID:** `2602.09701` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Sandesh Hegde, Jaison Saji Chacko, Debarshi Banerjee, et al.
+- **abs:** https://arxiv.org/abs/2602.09701 · **PDF:** https://arxiv.org/pdf/2602.09701v1
+
+## 13637. Life Cycle-Aware Evaluation of Knowledge Distillation for Machine Translation: Environmental Impact and Translation Quality Trade-offs
+
+- **ID:** `2602.09691` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Joseph Attieh, Timothee Mickus, Anne-Laure Ligozat, et al.
+- **abs:** https://arxiv.org/abs/2602.09691 · **PDF:** https://arxiv.org/pdf/2602.09691v1
+
+## 13638. Contextual and Seasonal LSTMs for Time Series Anomaly Detection
+
+- **ID:** `2602.09690` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Lingpei Zhang, Qingming Li, Yong Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.09690 · **PDF:** https://arxiv.org/pdf/2602.09690v1
+
+## 13639. Model soups need only one ingredient
+
+- **ID:** `2602.09689` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Alireza Abdollahpoorrostam, Nikolaos Dimitriadis, Adam Hazimeh, et al.
+- **abs:** https://arxiv.org/abs/2602.09689 · **PDF:** https://arxiv.org/pdf/2602.09689v1
+
+## 13640. Resilient Class-Incremental Learning: on the Interplay of Drifting, Unlabelled and Imbalanced Data Streams
+
+- **ID:** `2602.09681` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Jin Li, Kleanthis Malialis, Marios Polycarpou
+- **abs:** https://arxiv.org/abs/2602.09681 · **PDF:** https://arxiv.org/pdf/2602.09681v1
+
+## 13641. Administrative Law's Fourth Settlement: AI and the Capability-Accountability Trap
+
+- **ID:** `2602.09678` · **primary:** `cs.CY` · **Submitted:** 2026-02-10
+- **Authors:** Nicholas Caputo
+- **abs:** https://arxiv.org/abs/2602.09678 · **PDF:** https://arxiv.org/pdf/2602.09678v2
+
+## 13642. StructLens: A Structural Lens for Language Models via Maximum Spanning Trees
+
+- **ID:** `2603.03328` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Haruki Sakajo, Frederikus Hudi, Yusuke Sakai, et al.
+- **abs:** https://arxiv.org/abs/2603.03328 · **PDF:** https://arxiv.org/pdf/2603.03328v1
+
+## 13643. Differentiable Modeling for Low-Inertia Grids: Benchmarking PINNs, NODEs, and DP for Identification and Control of SMIB System
+
+- **ID:** `2602.09667` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Shinhoo Kang, Sangwook Kim, Sehyun Yun
+- **abs:** https://arxiv.org/abs/2602.09667 · **PDF:** https://arxiv.org/pdf/2602.09667v1
+
+## 13644. IDPruner: Harmonizing Importance and Diversity in Visual Token Pruning for MLLMs
+
+- **ID:** `2602.13315` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Yifan Tan, Yifu Sun, Shirui Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.13315 · **PDF:** https://arxiv.org/pdf/2602.13315v1
+
+## 13645. The Hidden Costs of Domain Fine-Tuning: Pii-Bearing Data Degrades Safety and Increases Leakage
+
+- **ID:** `2603.00061` · **primary:** `cs.CR` · **Submitted:** 2026-02-10
+- **Authors:** Jayesh Choudhari, Piyush Kumar Singh
+- **abs:** https://arxiv.org/abs/2603.00061 · **PDF:** https://arxiv.org/pdf/2603.00061v1
+
+## 13646. ClinAlign: Scaling Healthcare Alignment from Clinician Preference
+
+- **ID:** `2602.09653` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Shiwei Lyu, Xidong Wang, Lei Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.09653 · **PDF:** https://arxiv.org/pdf/2602.09653v2
+
+## 13647. Sim2Radar: Toward Bridging the Radar Sim-to-Real Gap with VLM-Guided Scene Reconstruction
+
+- **ID:** `2602.13314` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Emily Bejerano, Federico Tondolo, Ayaan Qayyum, et al.
+- **abs:** https://arxiv.org/abs/2602.13314 · **PDF:** https://arxiv.org/pdf/2602.13314v3
+
+## 13648. The Entropic Signature of Class Speciation in Diffusion Models
+
+- **ID:** `2602.09651` · **primary:** `stat.ML` · **Submitted:** 2026-02-10
+- **Authors:** Florian Handke, Dejan Stančević, Felix Koulischer, et al.
+- **abs:** https://arxiv.org/abs/2602.09651 · **PDF:** https://arxiv.org/pdf/2602.09651v1
+
+## 13649. MATA: Multi-Agent Framework for Reliable and Flexible Table Question Answering
+
+- **ID:** `2602.09642` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Sieun Hyeon, Jusang Oh, Sunghwan Steve Cho, et al.
+- **abs:** https://arxiv.org/abs/2602.09642 · **PDF:** https://arxiv.org/pdf/2602.09642v1
+
+## 13650. Blind denoising diffusion models and the blessings of dimensionality
+
+- **ID:** `2602.09639` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Zahra Kadkhodaie, Aram-Alexandre Pooladian, Sinho Chewi, et al.
+- **abs:** https://arxiv.org/abs/2602.09639 · **PDF:** https://arxiv.org/pdf/2602.09639v1
+
+## 13651. Learning Under Extreme Data Scarcity: Subject-Level Evaluation of Lightweight CNNs for fMRI-Based Prodromal Parkinsons Detection
+
+- **ID:** `2603.00060` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Naimur Rahman
+- **abs:** https://arxiv.org/abs/2603.00060 · **PDF:** https://arxiv.org/pdf/2603.00060v1
+
+## 13652. LLM-FS: Zero-Shot Feature Selection for Effective and Interpretable Malware Detection
+
+- **ID:** `2602.09634` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Naveen Gill, Ajvad Haneef K, Madhu Kumar S D
+- **abs:** https://arxiv.org/abs/2602.09634 · **PDF:** https://arxiv.org/pdf/2602.09634v1
+
+## 13653. Stop Testing Attacks, Start Diagnosing Defenses: The Four-Checkpoint Framework Reveals Where LLM Safety Breaks
+
+- **ID:** `2602.09629` · **primary:** `cs.CR` · **Submitted:** 2026-02-10
+- **Authors:** Hayfa Dhabhi, Kashyap Thimmaraju
+- **abs:** https://arxiv.org/abs/2602.09629 · **PDF:** https://arxiv.org/pdf/2602.09629v1
+
+## 13654. Agentic Spatio-Temporal Grounding via Collaborative Reasoning
+
+- **ID:** `2602.13313` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Heng Zhao, Yew-Soon Ong, Joey Tianyi Zhou
+- **abs:** https://arxiv.org/abs/2602.13313 · **PDF:** https://arxiv.org/pdf/2602.13313v1
+
+## 13655. MILE-RefHumEval: A Reference-Free, Multi-Independent LLM Framework for Human-Aligned Evaluation
+
+- **ID:** `2602.09624` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Nalin Srun, Parisa Rastin, Guénaël Cabanes, et al.
+- **abs:** https://arxiv.org/abs/2602.09624 · **PDF:** https://arxiv.org/pdf/2602.09624v1
+
+## 13656. AlignTune: Modular Toolkit for Post-Training Alignment of Large Language Models
+
+- **ID:** `2602.09621` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** R E Zera Marveen Lyngkhoi, Chirag Chawla, Pratinav Seth, et al.
+- **abs:** https://arxiv.org/abs/2602.09621 · **PDF:** https://arxiv.org/pdf/2602.09621v2
+
+## 13657. FLINGO -- Instilling ASP Expressiveness into Linear Integer Constraints
+
+- **ID:** `2602.09620` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Jorge Fandinno, Pedro Cabalar, Philipp Wanko, et al.
+- **abs:** https://arxiv.org/abs/2602.09620 · **PDF:** https://arxiv.org/pdf/2602.09620v1
+
+## 13658. AnyTouch 2: General Optical Tactile Representation Learning For Dynamic Tactile Perception
+
+- **ID:** `2602.09617` · **primary:** `cs.RO` · **Submitted:** 2026-02-10
+- **Authors:** Ruoxuan Feng, Yuxuan Zhou, Siyu Mei, et al.
+- **abs:** https://arxiv.org/abs/2602.09617 · **PDF:** https://arxiv.org/pdf/2602.09617v1
+
+## 13659. With Argus Eyes: Assessing Retrieval Gaps via Uncertainty Scoring to Detect and Remedy Retrieval Blind Spots
+
+- **ID:** `2602.09616` · **primary:** `cs.IR` · **Submitted:** 2026-02-10
+- **Authors:** Zeinab Sadat Taghavi, Ali Modarressi, Hinrich Schutze, et al.
+- **abs:** https://arxiv.org/abs/2602.09616 · **PDF:** https://arxiv.org/pdf/2602.09616v1
+
+## 13660. Tracking Finite-Time Lyapunov Exponents to Robustify Neural ODEs
+
+- **ID:** `2602.09613` · **primary:** `math.DS` · **Submitted:** 2026-02-10
+- **Authors:** Tobias Wöhrer, Christian Kuehn
+- **abs:** https://arxiv.org/abs/2602.09613 · **PDF:** https://arxiv.org/pdf/2602.09613v1
+
+## 13661. AGMark: Attention-Guided Dynamic Watermarking for Large Vision-Language Models
+
+- **ID:** `2602.09611` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Yue Li, Xin Yi, Dongsheng Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.09611 · **PDF:** https://arxiv.org/pdf/2602.09611v1
+
+## 13662. Stochastic Parrots or Singing in Harmony? Testing Five Leading LLMs for their Ability to Replicate a Human Survey with Synthetic Data
+
+- **ID:** `2603.00059` · **primary:** `cs.CY` · **Submitted:** 2026-02-10
+- **Authors:** Jason Miklian, Kristian Hoelscher, John E. Katsos
+- **abs:** https://arxiv.org/abs/2603.00059 · **PDF:** https://arxiv.org/pdf/2603.00059v1
+
+## 13663. Learning from the Irrecoverable: Error-Localized Policy Optimization for Tool-Integrated LLM Reasoning
+
+- **ID:** `2602.09598` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Qiao Liang, Yuke Zhu, Chao Ge, et al.
+- **abs:** https://arxiv.org/abs/2602.09598 · **PDF:** https://arxiv.org/pdf/2602.09598v1
+
+## 13664. Detecting radar targets swarms in range profiles with a partially complex-valued neural network
+
+- **ID:** `2602.09597` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Martin Bauw
+- **abs:** https://arxiv.org/abs/2602.09597 · **PDF:** https://arxiv.org/pdf/2602.09597v1
+
+## 13665. Why the Counterintuitive Phenomenon of Likelihood Rarely Appears in Tabular Anomaly Detection with Deep Generative Models?
+
+- **ID:** `2602.09593` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Donghwan Kim, Junghun Phee, Hyunsoo Yoon
+- **abs:** https://arxiv.org/abs/2602.09593 · **PDF:** https://arxiv.org/pdf/2602.09593v1
+
+## 13666. On the Optimal Reasoning Length for RL-Trained Language Models
+
+- **ID:** `2602.09591` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Daisuke Nohara, Taishi Nakamura, Rio Yokota
+- **abs:** https://arxiv.org/abs/2602.09591 · **PDF:** https://arxiv.org/pdf/2602.09591v2
+
+## 13667. Context-Aware Counterfactual Data Augmentation for Gender Bias Mitigation in Language Models
+
+- **ID:** `2602.09590` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Shweta Parihar, Liu Guangliang, Natalie Parde, et al.
+- **abs:** https://arxiv.org/abs/2602.09590 · **PDF:** https://arxiv.org/pdf/2602.09590v1
+
+## 13668. A benchmark for joint dialogue satisfaction, emotion recognition, and emotion state transition prediction
+
+- **ID:** `2603.03327` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Jing Bian, Haoxiang Su, Liting Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.03327 · **PDF:** https://arxiv.org/pdf/2603.03327v1
+
+## 13669. MieDB-100k: A Comprehensive Dataset for Medical Image Editing
+
+- **ID:** `2602.09587` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Yongfan Lai, Wen Qian, Bo Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.09587 · **PDF:** https://arxiv.org/pdf/2602.09587v1
+
+## 13670. PeroMAS: A Multi-agent System of Perovskite Material Discovery
+
+- **ID:** `2602.13312` · **primary:** `cs.MA` · **Submitted:** 2026-02-10
+- **Authors:** Yishu Wang, Wei Liu, Yifan Li, et al.
+- **abs:** https://arxiv.org/abs/2602.13312 · **PDF:** https://arxiv.org/pdf/2602.13312v1
+
+## 13671. Mitigating the Likelihood Paradox in Flow-based OOD Detection via Entropy Manipulation
+
+- **ID:** `2602.09581` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Donghwan Kim, Hyunsoo Yoon
+- **abs:** https://arxiv.org/abs/2602.09581 · **PDF:** https://arxiv.org/pdf/2602.09581v1
+
+## 13672. SERFN: Sample-Efficient Real-World Dexterous Policy Fine-Tuning via Action-Chunked Critics and Normalizing Flows
+
+- **ID:** `2602.09580` · **primary:** `cs.RO` · **Submitted:** 2026-02-10
+- **Authors:** Chenyu Yang, Denis Tarasov, Davide Liconti, et al.
+- **abs:** https://arxiv.org/abs/2602.09580 · **PDF:** https://arxiv.org/pdf/2602.09580v2
+
+## 13673. Rollout-Training Co-Design for Efficient LLM-Based Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.09578` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Zhida Jiang, Zhaolong Xing, Jiawei Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.09578 · **PDF:** https://arxiv.org/pdf/2602.09578v1
+
+## 13674. Aligning Tree-Search Policies with Fixed Token Budgets in Test-Time Scaling of LLMs
+
+- **ID:** `2602.09574` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Sora Miyamoto, Daisuke Oba, Naoaki Okazaki
+- **abs:** https://arxiv.org/abs/2602.09574 · **PDF:** https://arxiv.org/pdf/2602.09574v1
+
+## 13675. Predictive Query Language: A Domain-Specific Language for Predictive Modeling on Relational Databases
+
+- **ID:** `2602.09572` · **primary:** `cs.DB` · **Submitted:** 2026-02-10
+- **Authors:** Vid Kocijan, Jinu Sunil, Jan Eric Lenssen, et al.
+- **abs:** https://arxiv.org/abs/2602.09572 · **PDF:** https://arxiv.org/pdf/2602.09572v2
+
+## 13676. LEMUR: A Corpus for Robust Fine-Tuning of Multilingual Law Embedding Models for Retrieval
+
+- **ID:** `2602.09570` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Narges Baba Ahmadi, Jan Strich, Martin Semmann, et al.
+- **abs:** https://arxiv.org/abs/2602.09570 · **PDF:** https://arxiv.org/pdf/2602.09570v1
+
+## 13677. Training deep physical neural networks with local physical information bottleneck
+
+- **ID:** `2602.09569` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Hao Wang, Ziao Wang, Xiangpeng Liang, et al.
+- **abs:** https://arxiv.org/abs/2602.09569 · **PDF:** https://arxiv.org/pdf/2602.09569v1
+
+## 13678. Distributionally Robust Geometric Joint Chance-Constrained Optimization: Neurodynamic Approaches
+
+- **ID:** `2603.06597` · **primary:** `cs.NE` · **Submitted:** 2026-02-10
+- **Authors:** Ange Valli, Siham Tassouli, Abdel Lisser
+- **abs:** https://arxiv.org/abs/2603.06597 · **PDF:** https://arxiv.org/pdf/2603.06597v2
+
+## 13679. ECG-IMN: Interpretable Mesomorphic Neural Networks for 12-Lead Electrocardiogram Interpretation
+
+- **ID:** `2602.09566` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Vajira Thambawita, Jonas L. Isaksen, Jørgen K. Kanters, et al.
+- **abs:** https://arxiv.org/abs/2602.09566 · **PDF:** https://arxiv.org/pdf/2602.09566v1
+
+## 13680. Advancing Block Diffusion Language Models for Test-Time Scaling
+
+- **ID:** `2602.09555` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Yi Lu, Deyang Kong, Jianing Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.09555 · **PDF:** https://arxiv.org/pdf/2602.09555v2
+
+## 13681. PaperRepro: Automated Computational Reproducibility Assessment for Social Science Papers
+
+- **ID:** `2603.00058` · **primary:** `cs.CY` · **Submitted:** 2026-02-10
+- **Authors:** Linhao Zhang, Tong Xia, Jinghua Piao, et al.
+- **abs:** https://arxiv.org/abs/2603.00058 · **PDF:** https://arxiv.org/pdf/2603.00058v1
+
+## 13682. Beyond SMILES: Evaluating Agentic Systems for Drug Discovery
+
+- **ID:** `2602.10163` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-10
+- **Authors:** Edward Wijaya
+- **abs:** https://arxiv.org/abs/2602.10163 · **PDF:** https://arxiv.org/pdf/2602.10163v1
+
+## 13683. Comprehensive Comparison of RAG Methods Across Multi-Domain Conversational QA
+
+- **ID:** `2602.09552` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Klejda Alushi, Jan Strich, Chris Biemann, et al.
+- **abs:** https://arxiv.org/abs/2602.09552 · **PDF:** https://arxiv.org/pdf/2602.09552v1
+
+## 13684. NExT-Guard: Training-Free Streaming Safeguard without Token-Level Labels
+
+- **ID:** `2603.02219` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Junfeng Fang, Nachuan Chen, Houcheng Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.02219 · **PDF:** https://arxiv.org/pdf/2603.02219v1
+
+## 13685. UniARM: Towards a Unified Autoregressive Reward Model for Multi-Objective Test-Time Alignment
+
+- **ID:** `2602.09538` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Hongyan Xie, Yikun Ban, Ruiyu Fang, et al.
+- **abs:** https://arxiv.org/abs/2602.09538 · **PDF:** https://arxiv.org/pdf/2602.09538v1
+
+## 13686. Autoregressive Direct Preference Optimization
+
+- **ID:** `2602.09533` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Masanari Oi, Mahiro Ukai, Masahiro Kaneko, et al.
+- **abs:** https://arxiv.org/abs/2602.09533 · **PDF:** https://arxiv.org/pdf/2602.09533v1
+
+## 13687. "Bespoke Bots": Diverse Instructor Needs for Customizing Generative AI Classroom Chatbots
+
+- **ID:** `2603.00057` · **primary:** `cs.CY` · **Submitted:** 2026-02-10
+- **Authors:** Irene Hou, Zeyu Xiong, Philip J. Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.00057 · **PDF:** https://arxiv.org/pdf/2603.00057v1
+
+## 13688. Learning to Discover Iterative Spectral Algorithms
+
+- **ID:** `2602.09530` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Zihang Liu, Oleg Balabanov, Yaoqing Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.09530 · **PDF:** https://arxiv.org/pdf/2602.09530v1
+
+## 13689. Rashomon Sets and Model Multiplicity in Federated Learning
+
+- **ID:** `2602.09520` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Xenia Heilmann, Luca Corbucci, Mattia Cerrato
+- **abs:** https://arxiv.org/abs/2602.09520 · **PDF:** https://arxiv.org/pdf/2602.09520v1
+
+## 13690. Knowledge Integration Decay in Search-Augmented Reasoning of Large Language Models
+
+- **ID:** `2602.09517` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Sangwon Yu, Ik-hwan Kim, Donghun Kang, et al.
+- **abs:** https://arxiv.org/abs/2602.09517 · **PDF:** https://arxiv.org/pdf/2602.09517v1
+
+## 13691. The CLEF-2026 CheckThat! Lab: Advancing Multilingual Fact-Checking
+
+- **ID:** `2602.09516` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Julia Maria Struß, Sebastian Schellhammer, Stefan Dietze, et al.
+- **abs:** https://arxiv.org/abs/2602.09516 · **PDF:** https://arxiv.org/pdf/2602.09516v1
+
+## 13692. Controllable and explainable personality sliders for LLMs at inference time
+
+- **ID:** `2603.03326` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Florian Hoppe, David Khachaturov, Robert Mullins, et al.
+- **abs:** https://arxiv.org/abs/2603.03326 · **PDF:** https://arxiv.org/pdf/2603.03326v1
+
+## 13693. EcoGym: Evaluating LLMs for Long-Horizon Plan-and-Execute in Interactive Economies
+
+- **ID:** `2602.09514` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Xavier Hu, Jinxiang Xia, Shengze Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.09514 · **PDF:** https://arxiv.org/pdf/2602.09514v2
+
+## 13694. Self-Play Only Evolves When Self-Synthetic Pipeline Ensures Learnable Information Gain
+
+- **ID:** `2603.02218` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Wei Liu, Siya Qi, Yali Du, et al.
+- **abs:** https://arxiv.org/abs/2603.02218 · **PDF:** https://arxiv.org/pdf/2603.02218v1
+
+## 13695. Beyond Student: An Asymmetric Network for Neural Network Inheritance
+
+- **ID:** `2602.09509` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Yiyun Zhou, Jingwei Shi, Mingjing Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.09509 · **PDF:** https://arxiv.org/pdf/2602.09509v2
+
+## 13696. Towards Uniformity and Alignment for Multimodal Representation Learning
+
+- **ID:** `2602.09507` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Wenzhe Yin, Pan Zhou, Zehao Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.09507 · **PDF:** https://arxiv.org/pdf/2602.09507v1
+
+## 13697. Is Retraining-Free Enough? The Necessity of Router Calibration for Efficient MoE Compression
+
+- **ID:** `2603.02217` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Sieun Hyeon, Jaeyoung Do
+- **abs:** https://arxiv.org/abs/2603.02217 · **PDF:** https://arxiv.org/pdf/2603.02217v1
+
+## 13698. Seeing the Goal, Missing the Truth: Human Accountability for AI Bias
+
+- **ID:** `2602.09504` · **primary:** `q-fin.GN` · **Submitted:** 2026-02-10
+- **Authors:** Sean Cao, Wei Jiang, Hui Xu
+- **abs:** https://arxiv.org/abs/2602.09504 · **PDF:** https://arxiv.org/pdf/2602.09504v1
+
+## 13699. How effective are VLMs in assisting humans in inferring the quality of mental models from Multimodal short answers?
+
+- **ID:** `2603.00056` · **primary:** `cs.CY` · **Submitted:** 2026-02-10
+- **Authors:** Pritam Sil, Durgaprasad Karnam, Vinay Reddy Venumuddala, et al.
+- **abs:** https://arxiv.org/abs/2603.00056 · **PDF:** https://arxiv.org/pdf/2603.00056v1
+
+## 13700. Improved Approximate Regret for Decentralized Online Continuous Submodular Maximization via Reductions
+
+- **ID:** `2602.09502` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Yuanyu Wan, Yu Shen, Dingzhi Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.09502 · **PDF:** https://arxiv.org/pdf/2602.09502v1
+
+## 13701. Where-to-Unmask: Ground-Truth-Guided Unmasking Order Learning for Masked Diffusion Language Models
+
+- **ID:** `2602.09501` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Hikaru Asano, Tadashi Kozuno, Kuniaki Saito, et al.
+- **abs:** https://arxiv.org/abs/2602.09501 · **PDF:** https://arxiv.org/pdf/2602.09501v1
+
+## 13702. Computationally Efficient Replicable Learning of Parities
+
+- **ID:** `2602.09499` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Moshe Noivirt, Jessica Sorrell, Eliad Tsfadia
+- **abs:** https://arxiv.org/abs/2602.09499 · **PDF:** https://arxiv.org/pdf/2602.09499v1
+
+## 13703. ATPO: Adaptive Tree Policy Optimization for Multi-Turn Medical Dialogue
+
+- **ID:** `2603.02216` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Ruike Cao, Shaojie Bai, Fugen Yao, et al.
+- **abs:** https://arxiv.org/abs/2603.02216 · **PDF:** https://arxiv.org/pdf/2603.02216v1
+
+## 13704. RxnNano:Training Compact LLMs for Chemical Reaction and Retrosynthesis Prediction via Hierarchical Curriculum Learning
+
+- **ID:** `2603.02215` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Ran Li, Shimin Di, Haowei LI, et al.
+- **abs:** https://arxiv.org/abs/2603.02215 · **PDF:** https://arxiv.org/pdf/2603.02215v1
+
+## 13705. Zero-Sacrifice Persistent-Robustness Adversarial Defense for Pre-Trained Encoders
+
+- **ID:** `2602.11204` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Zhuxin Lei, Ziyuan Yang, Yi Zhang
+- **abs:** https://arxiv.org/abs/2602.11204 · **PDF:** https://arxiv.org/pdf/2602.11204v1
+
+## 13706. Beware of the Batch Size: Hyperparameter Bias in Evaluating LoRA
+
+- **ID:** `2602.09492` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Sangyoon Lee, Jaeho Lee
+- **abs:** https://arxiv.org/abs/2602.09492 · **PDF:** https://arxiv.org/pdf/2602.09492v1
+
+## 13707. Computing Conditional Shapley Values Using Tabular Foundation Models
+
+- **ID:** `2602.09489` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Lars Henry Berge Olsen, Dennis Christensen
+- **abs:** https://arxiv.org/abs/2602.09489 · **PDF:** https://arxiv.org/pdf/2602.09489v1
+
+## 13708. Adaptive recurrent flow map operator learning for reaction diffusion dynamics
+
+- **ID:** `2602.09487` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Huseyin Tunc
+- **abs:** https://arxiv.org/abs/2602.09487 · **PDF:** https://arxiv.org/pdf/2602.09487v1
+
+## 13709. Listen to the Layers: Mitigating Hallucinations with Inter-Layer Disagreement
+
+- **ID:** `2602.09486` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Koduvayur Subbalakshmi, Sabbir Hossain Ujjal, Venkata Krishna Teja Mangichetty, et al.
+- **abs:** https://arxiv.org/abs/2602.09486 · **PDF:** https://arxiv.org/pdf/2602.09486v2
+
+## 13710. Bridging Efficiency and Transparency: Explainable CoT Compression in Multimodal Large Reasoning Models
+
+- **ID:** `2602.09485` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Yizhi Wang, Linan Yue, Min-Ling Zhang
+- **abs:** https://arxiv.org/abs/2602.09485 · **PDF:** https://arxiv.org/pdf/2602.09485v1
+
+## 13711. IntPro: A Proxy Agent for Context-Aware Intent Understanding via Retrieval-conditioned Inference
+
+- **ID:** `2603.03325` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Guanming Liu, Meng Wu, Peng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.03325 · **PDF:** https://arxiv.org/pdf/2603.03325v1
+
+## 13712. M3-AD: Reflection-aware Multi-modal, Multi-category, and Multi-dimensional Benchmark and Framework for Industrial Anomaly Detection
+
+- **ID:** `2603.00055` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Chao Huang, Yanhui Li, Yunkang Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.00055 · **PDF:** https://arxiv.org/pdf/2603.00055v1
+
+## 13713. ArtifactLens: Hundreds of Labels Are Enough for Artifact Detection with VLMs
+
+- **ID:** `2602.09475` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** James Burgess, Rameen Abdal, Dan Stoddart, et al.
+- **abs:** https://arxiv.org/abs/2602.09475 · **PDF:** https://arxiv.org/pdf/2602.09475v1
+
+## 13714. Online Learning in MDPs with Partially Adversarial Transitions and Losses
+
+- **ID:** `2602.09474` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Ofir Schlisselberg, Tal Lancewicki, Yishay Mansour
+- **abs:** https://arxiv.org/abs/2602.09474 · **PDF:** https://arxiv.org/pdf/2602.09474v1
+
+## 13715. NOWJ @BioCreative IX ToxHabits: An Ensemble Deep Learning Approach for Detecting Substance Use and Contextual Information in Clinical Texts
+
+- **ID:** `2602.09469` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Huu-Huy-Hoang Tran, Gia-Bao Duong, Quoc-Viet-Anh Tran, et al.
+- **abs:** https://arxiv.org/abs/2602.09469 · **PDF:** https://arxiv.org/pdf/2602.09469v1
+
+## 13716. Expert Divergence Learning for MoE-based Language Models
+
+- **ID:** `2603.00054` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Jiaang Li, Haibin Chen, Langming Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.00054 · **PDF:** https://arxiv.org/pdf/2603.00054v1
+
+## 13717. AlgoVeri: An Aligned Benchmark for Verified Code Generation on Classical Algorithms
+
+- **ID:** `2602.09464` · **primary:** `cs.SE` · **Submitted:** 2026-02-10
+- **Authors:** Haoyu Zhao, Ziran Yang, Jiawei Li, et al.
+- **abs:** https://arxiv.org/abs/2602.09464 · **PDF:** https://arxiv.org/pdf/2602.09464v1
+
+## 13718. SpotAgent: Grounding Visual Geo-localization in Large Vision-Language Models through Agentic Reasoning
+
+- **ID:** `2602.09463` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Furong Jia, Ling Dai, Wenjin Deng, et al.
+- **abs:** https://arxiv.org/abs/2602.09463 · **PDF:** https://arxiv.org/pdf/2602.09463v3
+
+## 13719. Scalable and Reliable State-Aware Inference of High-Impact N-k Contingencies
+
+- **ID:** `2602.09461` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Lihao Mai, Chenhan Xiao, Yang Weng
+- **abs:** https://arxiv.org/abs/2602.09461 · **PDF:** https://arxiv.org/pdf/2602.09461v1
+
+## 13720. From Average Sensitivity to Small-Loss Regret Bounds under Random-Order Model
+
+- **ID:** `2602.09457` · **primary:** `stat.ML` · **Submitted:** 2026-02-10
+- **Authors:** Shinsaku Sakaue, Yuichi Yoshida
+- **abs:** https://arxiv.org/abs/2602.09457 · **PDF:** https://arxiv.org/pdf/2602.09457v1
+
+## 13721. Taming the Monster Every Context: Complexity Measure and Unified Framework for Offline-Oracle Efficient Contextual Bandits
+
+- **ID:** `2602.09456` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Hao Qin, Chicheng Zhang
+- **abs:** https://arxiv.org/abs/2602.09456 · **PDF:** https://arxiv.org/pdf/2602.09456v1
+
+## 13722. Enhancing Affine Maximizer Auctions with Correlation-Aware Payment
+
+- **ID:** `2602.09455` · **primary:** `cs.GT` · **Submitted:** 2026-02-10
+- **Authors:** Haoran Sun, Xuanzhi Xia, Xu Chu, et al.
+- **abs:** https://arxiv.org/abs/2602.09455 · **PDF:** https://arxiv.org/pdf/2602.09455v1
+
+## 13723. SiliconMind-V1: Multi-Agent Distillation and Debug-Reasoning Workflows for Verilog Code Generation
+
+- **ID:** `2603.08719` · **primary:** `cs.AR` · **Submitted:** 2026-02-10
+- **Authors:** Mu-Chi Chen, Yu-Hung Kao, Po-Hsuan Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.08719 · **PDF:** https://arxiv.org/pdf/2603.08719v2
+
+## 13724. The Wisdom of Many Queries: Complexity-Diversity Principle for Dense Retriever Training
+
+- **ID:** `2602.09448` · **primary:** `cs.IR` · **Submitted:** 2026-02-10
+- **Authors:** Xincan Feng, Noriki Nishida, Yusuke Sakai, et al.
+- **abs:** https://arxiv.org/abs/2602.09448 · **PDF:** https://arxiv.org/pdf/2602.09448v4
+
+## 13725. SWE-AGI: Benchmarking Specification-Driven Software Construction with MoonBit in the Era of Autonomous Agents
+
+- **ID:** `2602.09447` · **primary:** `cs.SE` · **Submitted:** 2026-02-10
+- **Authors:** Zhirui Zhang, Hongbo Zhang, Haoxiang Fei, et al.
+- **abs:** https://arxiv.org/abs/2602.09447 · **PDF:** https://arxiv.org/pdf/2602.09447v2
+
+## 13726. A Scoping Review of Deep Learning for Urban Visual Pollution and Proposal of a Real-Time Monitoring Framework with a Visual Pollution Index
+
+- **ID:** `2602.09446` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Mohammad Masudur Rahman, Md. Rashedur Rahman, Ashraful Islam, et al.
+- **abs:** https://arxiv.org/abs/2602.09446 · **PDF:** https://arxiv.org/pdf/2602.09446v1
+
+## 13727. Conceptual Cultural Index: A Metric for Cultural Specificity via Relative Generality
+
+- **ID:** `2602.09444` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Takumi Ohashi, Hitoshi Iyatomi
+- **abs:** https://arxiv.org/abs/2602.09444 · **PDF:** https://arxiv.org/pdf/2602.09444v1
+
+## 13728. P1-VL: Bridging Visual Perception and Scientific Reasoning in Physics Olympiads
+
+- **ID:** `2602.09443` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Yun Luo, Futing Wang, Qianjia Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.09443 · **PDF:** https://arxiv.org/pdf/2602.09443v1
+
+## 13729. Evaluating Social Bias in RAG Systems: When External Context Helps and Reasoning Hurts
+
+- **ID:** `2602.09442` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Shweta Parihar, Lu Cheng
+- **abs:** https://arxiv.org/abs/2602.09442 · **PDF:** https://arxiv.org/pdf/2602.09442v1
+
+## 13730. Mag-Mamba: Modeling Coupled spatiotemporal Asymmetry for POI Recommendation
+
+- **ID:** `2603.00053` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Zhuoxuan Li, Tangwei Ye, Jieyuan Pei, et al.
+- **abs:** https://arxiv.org/abs/2603.00053 · **PDF:** https://arxiv.org/pdf/2603.00053v1
+
+## 13731. Breaking the Pre-Sampling Barrier: Activation-Informed Difficulty-Aware Self-Consistency
+
+- **ID:** `2602.09438` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Taewoong Yoon, Geunyeong Jeong, Geon Park, et al.
+- **abs:** https://arxiv.org/abs/2602.09438 · **PDF:** https://arxiv.org/pdf/2602.09438v1
+
+## 13732. Controlling Chat Style in Language Models via Single-Direction Editing
+
+- **ID:** `2603.03324` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Zhenyu Xu, Victor S. Sheng
+- **abs:** https://arxiv.org/abs/2603.03324 · **PDF:** https://arxiv.org/pdf/2603.03324v1
+
+## 13733. Omni-Safety under Cross-Modality Conflict: Vulnerabilities, Dynamics Mechanisms and Efficient Alignment
+
+- **ID:** `2602.10161` · **primary:** `cs.CR` · **Submitted:** 2026-02-10
+- **Authors:** Kun Wang, Zherui Li, Zhenhong Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.10161 · **PDF:** https://arxiv.org/pdf/2602.10161v1
+
+## 13734. Diffusion-Guided Pretraining for Brain Graph Foundation Models
+
+- **ID:** `2602.09437` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Xinxu Wei, Rong Zhou, Lifang He, et al.
+- **abs:** https://arxiv.org/abs/2602.09437 · **PDF:** https://arxiv.org/pdf/2602.09437v3
+
+## 13735. A Behavioral Fingerprint for Large Language Models: Provenance Tracking via Refusal Vectors
+
+- **ID:** `2602.09434` · **primary:** `cs.CR` · **Submitted:** 2026-02-10
+- **Authors:** Zhenyu Xu, Victor S. Sheng
+- **abs:** https://arxiv.org/abs/2602.09434 · **PDF:** https://arxiv.org/pdf/2602.09434v1
+
+## 13736. Autonomous Action Runtime Management(AARM):A System Specification for Securing AI-Driven Actions at Runtime
+
+- **ID:** `2602.09433` · **primary:** `cs.CR` · **Submitted:** 2026-02-10
+- **Authors:** Herman Errico
+- **abs:** https://arxiv.org/abs/2602.09433 · **PDF:** https://arxiv.org/pdf/2602.09433v1
+
+## 13737. Discern Truth from Falsehood: Reducing Over-Refusal via Contrastive Refinement
+
+- **ID:** `2603.03323` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Yuxiao Lu, Lin Xu, Yang Sun, et al.
+- **abs:** https://arxiv.org/abs/2603.03323 · **PDF:** https://arxiv.org/pdf/2603.03323v1
+
+## 13738. Sci-VLA: Agentic VLA Inference Plugin for Long-Horizon Tasks in Scientific Experiments
+
+- **ID:** `2602.09430` · **primary:** `cs.RO` · **Submitted:** 2026-02-10
+- **Authors:** Yiwen Pang, Bo Zhou, Changjin Li, et al.
+- **abs:** https://arxiv.org/abs/2602.09430 · **PDF:** https://arxiv.org/pdf/2602.09430v1
+
+## 13739. Can Large Language Models Derive New Knowledge? A Dynamic Benchmark for Biological Knowledge Discovery
+
+- **ID:** `2603.03322` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Chaoqun Yang, Xinyu Lin, Shulin Li, et al.
+- **abs:** https://arxiv.org/abs/2603.03322 · **PDF:** https://arxiv.org/pdf/2603.03322v1
+
+## 13740. Knowledge-guided generative surrogate modeling for high-dimensional design optimization under scarce data
+
+- **ID:** `2603.00052` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Bingran Wang, Seongha Jeong, Sebastiaan P. C. van Schie, et al.
+- **abs:** https://arxiv.org/abs/2603.00052 · **PDF:** https://arxiv.org/pdf/2603.00052v1
+
+## 13741. Bridging the Modality Gap in Roadside LiDAR: A Training-Free Vision-Language Model Framework for Vehicle Classification
+
+- **ID:** `2602.09425` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Yiqiao Li, Bo Shang, Jie Wei
+- **abs:** https://arxiv.org/abs/2602.09425 · **PDF:** https://arxiv.org/pdf/2602.09425v1
+
+## 13742. Reward-Guided Discrete Diffusion via Clean-Sample Markov Chain for Molecule and Biological Sequence Design
+
+- **ID:** `2602.09424` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Prin Phunyaphibarn, Minhyuk Sung
+- **abs:** https://arxiv.org/abs/2602.09424 · **PDF:** https://arxiv.org/pdf/2602.09424v1
+
+## 13743. Beyond Input-Output: Rethinking Creativity through Design-by-Analogy in Human-AI Collaboration
+
+- **ID:** `2602.09423` · **primary:** `cs.HC` · **Submitted:** 2026-02-10
+- **Authors:** Xuechen Li, Shuai Zhang, Nan Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.09423 · **PDF:** https://arxiv.org/pdf/2602.09423v1
+
+## 13744. DIALEVAL: Automated Type-Theoretic Evaluation of LLM Instruction Following
+
+- **ID:** `2603.03321` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Nardine Basta, Dali Kaafar
+- **abs:** https://arxiv.org/abs/2603.03321 · **PDF:** https://arxiv.org/pdf/2603.03321v1
+
+## 13745. Are Language Models Sensitive to Morally Irrelevant Distractors?
+
+- **ID:** `2602.09416` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Andrew Shaw, Christina Hahn, Catherine Rasgaitis, et al.
+- **abs:** https://arxiv.org/abs/2602.09416 · **PDF:** https://arxiv.org/pdf/2602.09416v1
+
+## 13746. AD$^2$: Analysis and Detection of Adversarial Threats in Visual Perception for End-to-End Autonomous Driving Systems
+
+- **ID:** `2602.10160` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Ishan Sahu, Somnath Hazra, Somak Aditya, et al.
+- **abs:** https://arxiv.org/abs/2602.10160 · **PDF:** https://arxiv.org/pdf/2602.10160v1
+
+## 13747. LARV: Data-Free Layer-wise Adaptive Rescaling Veneer for Model Merging
+
+- **ID:** `2602.09413` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Xinyu Wang, Ke Deng, Fei Dou, et al.
+- **abs:** https://arxiv.org/abs/2602.09413 · **PDF:** https://arxiv.org/pdf/2602.09413v1
+
+## 13748. Accelerating Post-Quantum Cryptography via LLM-Driven Hardware-Software Co-Design
+
+- **ID:** `2602.09410` · **primary:** `cs.AR` · **Submitted:** 2026-02-10
+- **Authors:** Yuchao Liao, Tosiron Adegbija, Roman Lysecky
+- **abs:** https://arxiv.org/abs/2602.09410 · **PDF:** https://arxiv.org/pdf/2602.09410v1
+
+## 13749. Is Memorization Helpful or Harmful? Prior Information Sets the Threshold
+
+- **ID:** `2602.09405` · **primary:** `stat.ML` · **Submitted:** 2026-02-10
+- **Authors:** Chen Cheng, Rina Foygel Barber
+- **abs:** https://arxiv.org/abs/2602.09405 · **PDF:** https://arxiv.org/pdf/2602.09405v1
+
+## 13750. AgentCAT: An LLM Agent for Extracting and Analyzing Catalytic Reaction Data from Chemical Engineering Literature
+
+- **ID:** `2602.18479` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-10
+- **Authors:** Wei Yang, Zihao Liu, Tao Tan, et al.
+- **abs:** https://arxiv.org/abs/2602.18479 · **PDF:** https://arxiv.org/pdf/2602.18479v1
+
+## 13751. Learning with Multiple Correct Answers -- A Trichotomy of Regret Bounds under Different Feedback Models
+
+- **ID:** `2602.09402` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Alireza F. Pour, Farnam Mansouri, Shai Ben-David
+- **abs:** https://arxiv.org/abs/2602.09402 · **PDF:** https://arxiv.org/pdf/2602.09402v1
+
+## 13752. LitBench: A Graph-Centric Large Language Model Benchmarking Tool For Literature Tasks
+
+- **ID:** `2603.00051` · **primary:** `cs.DL` · **Submitted:** 2026-02-10
+- **Authors:** Andreas Varvarigos, Ali Maatouk, Jiasheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2603.00051 · **PDF:** https://arxiv.org/pdf/2603.00051v1
+
+## 13753. From We to Me: Theory Informed Narrative Shift with Abductive Reasoning
+
+- **ID:** `2603.03320` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Jaikrishna Manojkumar Patil, Divyagna Bavikadi, Kaustuv Mukherji, et al.
+- **abs:** https://arxiv.org/abs/2603.03320 · **PDF:** https://arxiv.org/pdf/2603.03320v1
+
+## 13754. Squeezing More from the Stream : Learning Representation Online for Streaming Reinforcement Learning
+
+- **ID:** `2602.09396` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Nilaksh, Antoine Clavaud, Mathieu Reymond, et al.
+- **abs:** https://arxiv.org/abs/2602.09396 · **PDF:** https://arxiv.org/pdf/2602.09396v1
+
+## 13755. Sparse Layer Sharpness-Aware Minimization for Efficient Fine-Tuning
+
+- **ID:** `2602.09395` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Yifei Cheng, Xianglin Yang, Guoxia Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.09395 · **PDF:** https://arxiv.org/pdf/2602.09395v1
+
+## 13756. The Critical Horizon: Inspection Design Principles for Multi-Stage Operations and Deep Reasoning
+
+- **ID:** `2602.09394` · **primary:** `stat.ML` · **Submitted:** 2026-02-10
+- **Authors:** Seyed Morteza Emadi
+- **abs:** https://arxiv.org/abs/2602.09394 · **PDF:** https://arxiv.org/pdf/2602.09394v2
+
+## 13757. LLMAC: A Global and Explainable Access Control Framework with Large Language Model
+
+- **ID:** `2602.09392` · **primary:** `cs.CR` · **Submitted:** 2026-02-10
+- **Authors:** Sharif Noor Zisad, Ragib Hasan
+- **abs:** https://arxiv.org/abs/2602.09392 · **PDF:** https://arxiv.org/pdf/2602.09392v1
+
+## 13758. TVTSyn: Content-Synchronous Time-Varying Timbre for Streaming Voice Conversion and Anonymization
+
+- **ID:** `2602.09389` · **primary:** `eess.AS` · **Submitted:** 2026-02-10
+- **Authors:** Waris Quamer, Mu-Ruei Tseng, Ghady Nasrallah, et al.
+- **abs:** https://arxiv.org/abs/2602.09389 · **PDF:** https://arxiv.org/pdf/2602.09389v1
+
+## 13759. Effective vocabulary expanding of multilingual language models for extremely low-resource languages
+
+- **ID:** `2602.09388` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Jianyu Zheng
+- **abs:** https://arxiv.org/abs/2602.09388 · **PDF:** https://arxiv.org/pdf/2602.09388v1
+
+## 13760. Visual Para-Thinker: Divide-and-Conquer Reasoning for Visual Comprehension
+
+- **ID:** `2602.13310` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Haoran Xu, Hongyu Wang, Jiaze Li, et al.
+- **abs:** https://arxiv.org/abs/2602.13310 · **PDF:** https://arxiv.org/pdf/2602.13310v1
+
+## 13761. Contractual Deepfakes: Can Large Language Models Generate Contracts?
+
+- **ID:** `2602.09384` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Eliza Mik
+- **abs:** https://arxiv.org/abs/2602.09384 · **PDF:** https://arxiv.org/pdf/2602.09384v1
+
+## 13762. BiasScope: Towards Automated Detection of Bias in LLM-as-a-Judge Evaluation
+
+- **ID:** `2602.09383` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Peng Lai, Zhihao Ou, Yong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.09383 · **PDF:** https://arxiv.org/pdf/2602.09383v1
+
+## 13763. Egocentric Bias in Vision-Language Models
+
+- **ID:** `2602.15892` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Maijunxian Wang, Yijiang Li, Bingyang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.15892 · **PDF:** https://arxiv.org/pdf/2602.15892v1
+
+## 13764. Beyond Closed-Pool Video Retrieval: A Benchmark and Agent Framework for Real-World Video Search and Moment Localization
+
+- **ID:** `2602.10159` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Tao Yu, Yujia Yang, Haopeng Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.10159 · **PDF:** https://arxiv.org/pdf/2602.10159v1
+
+## 13765. LingxiDiagBench: A Multi-Agent Framework for Benchmarking LLMs in Chinese Psychiatric Consultation and Diagnosis
+
+- **ID:** `2602.09379` · **primary:** `cs.MA` · **Submitted:** 2026-02-10
+- **Authors:** Shihao Xu, Tiancheng Zhou, Jiatong Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.09379 · **PDF:** https://arxiv.org/pdf/2602.09379v2
+
+## 13766. Adaptive Value Decomposition: Coordinating a Varying Number of Agents in Urban Systems
+
+- **ID:** `2602.13309` · **primary:** `cs.MA` · **Submitted:** 2026-02-10
+- **Authors:** Yexin Li, Jinjin Guo, Haoyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.13309 · **PDF:** https://arxiv.org/pdf/2602.13309v1
+
+## 13767. NMRTrans: Structure Elucidation from Experimental NMR Spectra via Set Transformers
+
+- **ID:** `2602.10158` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-10
+- **Authors:** Liujia Yang, Zhuo Yang, Jiaqing Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.10158 · **PDF:** https://arxiv.org/pdf/2602.10158v1
+
+## 13768. Latent Poincaré Shaping for Agentic Reinforcement Learning
+
+- **ID:** `2602.09375` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Hanchen Xia, Baoyou Chen, Zelin Zang, et al.
+- **abs:** https://arxiv.org/abs/2602.09375 · **PDF:** https://arxiv.org/pdf/2602.09375v3
+
+## 13769. Surrogate-Guided Quantum Discovery in Black-Box Landscapes with Latent-Quadratic Interaction Embedding Transformers
+
+- **ID:** `2602.09374` · **primary:** `quant-ph` · **Submitted:** 2026-02-10
+- **Authors:** Saisubramaniam Gopalakrishnan, Dagnachew Birru
+- **abs:** https://arxiv.org/abs/2602.09374 · **PDF:** https://arxiv.org/pdf/2602.09374v1
+
+## 13770. AfriNLLB: Efficient Translation Models for African Languages
+
+- **ID:** `2602.09373` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Yasmin Moslem, Aman Kassahun Wassie, Amanuel Gizachew Abebe
+- **abs:** https://arxiv.org/abs/2602.09373 · **PDF:** https://arxiv.org/pdf/2602.09373v1
+
+## 13771. AgentSkiller: Scaling Generalist Agent Intelligence through Semantically Integrated Cross-Domain Data Synthesis
+
+- **ID:** `2602.09372` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Zexu Sun, Bokai Ji, Hengyi Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.09372 · **PDF:** https://arxiv.org/pdf/2602.09372v1
+
+## 13772. Unsupervised Cross-Lingual Part-of-Speech Tagging with Monolingual Corpora Only
+
+- **ID:** `2602.09366` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Jianyu Zheng
+- **abs:** https://arxiv.org/abs/2602.09366 · **PDF:** https://arxiv.org/pdf/2602.09366v1
+
+## 13773. Behavioral Economics of AI: LLM Biases and Corrections
+
+- **ID:** `2602.09362` · **primary:** `econ.GN` · **Submitted:** 2026-02-10
+- **Authors:** Pietro Bini, Lin William Cong, Xing Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.09362 · **PDF:** https://arxiv.org/pdf/2602.09362v1
+
+## 13774. MalMoE: Mixture-of-Experts Enhanced Encrypted Malicious Traffic Detection Under Graph Drift
+
+- **ID:** `2602.10157` · **primary:** `cs.CR` · **Submitted:** 2026-02-10
+- **Authors:** Yunpeng Tan, Qingyang Li, Mingxin Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.10157 · **PDF:** https://arxiv.org/pdf/2602.10157v1
+
+## 13775. Large Language Models for Designing Participatory Budgeting Rules
+
+- **ID:** `2602.09349` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Nguyen Thach, Xingchen Sha, Hau Chan
+- **abs:** https://arxiv.org/abs/2602.09349 · **PDF:** https://arxiv.org/pdf/2602.09349v2
+
+## 13776. Image Quality in the Era of Artificial Intelligence
+
+- **ID:** `2602.09347` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Jana G. Delfino, Jason L. Granstedt, Frank W. Samuelson, et al.
+- **abs:** https://arxiv.org/abs/2602.09347 · **PDF:** https://arxiv.org/pdf/2602.09347v1
+
+## 13777. CktEvo: Repository-Level RTL Code Benchmark for Design Evolution
+
+- **ID:** `2603.08718` · **primary:** `cs.AR` · **Submitted:** 2026-02-10
+- **Authors:** Zhengyuan Shi, Jingxin Wang, Tairan Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.08718 · **PDF:** https://arxiv.org/pdf/2603.08718v1
+
+## 13778. Digital Linguistic Bias in Spanish: Evidence from Lexical Variation in LLMs
+
+- **ID:** `2602.09346` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Yoshifumi Kawasaki
+- **abs:** https://arxiv.org/abs/2602.09346 · **PDF:** https://arxiv.org/pdf/2602.09346v1
+
+## 13779. AgentCgroup: Understanding and Controlling OS Resources of AI Agents
+
+- **ID:** `2602.09345` · **primary:** `cs.OS` · **Submitted:** 2026-02-10
+- **Authors:** Yusheng Zheng, Jiakun Fan, Quanzhi Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.09345 · **PDF:** https://arxiv.org/pdf/2602.09345v2
+
+## 13780. AI-Generated Letters from the Future: A Randomized Test of Personalized Climate Communication
+
+- **ID:** `2603.00050` · **primary:** `cs.CY` · **Submitted:** 2026-02-10
+- **Authors:** Nattavudh Powdthavee, Pat Pataranutaporn, Sandra J. Geiger, et al.
+- **abs:** https://arxiv.org/abs/2603.00050 · **PDF:** https://arxiv.org/pdf/2603.00050v1
+
+## 13781. Not-in-Perspective: Towards Shielding Google's Perspective API Against Adversarial Negation Attacks
+
+- **ID:** `2602.09343` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Michail S. Alexiou, J. Sukarno Mertoguno
+- **abs:** https://arxiv.org/abs/2602.09343 · **PDF:** https://arxiv.org/pdf/2602.09343v1
+
+## 13782. Auditing Multi-Agent LLM Reasoning Trees Outperforms Majority Vote and LLM-as-Judge
+
+- **ID:** `2602.09341` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Wei Yang, Shixuan Li, Heng Ping, et al.
+- **abs:** https://arxiv.org/abs/2602.09341 · **PDF:** https://arxiv.org/pdf/2602.09341v1
+
+## 13783. Measuring Dataset Diversity from a Geometric Perspective
+
+- **ID:** `2602.09340` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Yang Ba, Mohammad Sadeq Abolhasani, Michelle V Mancenido, et al.
+- **abs:** https://arxiv.org/abs/2602.09340 · **PDF:** https://arxiv.org/pdf/2602.09340v1
+
+## 13784. Understanding Risk and Dependency in AI Chatbot Use from User Discourse
+
+- **ID:** `2602.09339` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Jianfeng Zhu, Karin G. Coifman, Ruoming Jin
+- **abs:** https://arxiv.org/abs/2602.09339 · **PDF:** https://arxiv.org/pdf/2602.09339v1
+
+## 13785. Kyrtos: A methodology for automatic deep analysis of graphic charts with curves in technical documents
+
+- **ID:** `2602.09337` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Michail S. Alexiou, Nikolaos G. Bourbakis
+- **abs:** https://arxiv.org/abs/2602.09337 · **PDF:** https://arxiv.org/pdf/2602.09337v1
+
+## 13786. FM SO.P: A Progressive Task Mixture Framework with Automatic Evaluation for Cross-Domain SOP Understanding
+
+- **ID:** `2602.09336` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Siyuan Huang, Ziyu Wang, Chao Pan, et al.
+- **abs:** https://arxiv.org/abs/2602.09336 · **PDF:** https://arxiv.org/pdf/2602.09336v1
+
+## 13787. Beyond Uniform Credit: Causal Credit Assignment for Policy Optimization
+
+- **ID:** `2602.09331` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Mykola Khandoga, Rui Yuan, Vinay Kumar Sankarapu
+- **abs:** https://arxiv.org/abs/2602.09331 · **PDF:** https://arxiv.org/pdf/2602.09331v1
+
+## 13788. MacrOData: New Benchmarks of Thousands of Datasets for Tabular Outlier Detection
+
+- **ID:** `2602.09329` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Xueying Ding, Simon Klüttermann, Haomin Wen, et al.
+- **abs:** https://arxiv.org/abs/2602.09329 · **PDF:** https://arxiv.org/pdf/2602.09329v1
+
+## 13789. In-Hospital Stroke Prediction from PPG-Derived Hemodynamic Features
+
+- **ID:** `2602.09328` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Jiaming Liu, Cheng Ding, Daoqiang Zhang
+- **abs:** https://arxiv.org/abs/2602.09328 · **PDF:** https://arxiv.org/pdf/2602.09328v1
+
+## 13790. Priority-Aware Shapley Value
+
+- **ID:** `2602.09326` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Kiljae Lee, Ziqi Liu, Weijing Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.09326 · **PDF:** https://arxiv.org/pdf/2602.09326v1
+
+## 13791. GAFR-Net: A Graph Attention and Fuzzy-Rule Network for Interpretable Breast Cancer Image Classification
+
+- **ID:** `2602.09318` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Lin-Guo Gao, Suxing Liu
+- **abs:** https://arxiv.org/abs/2602.09318 · **PDF:** https://arxiv.org/pdf/2602.09318v1
+
+## 13792. SnareNet: Flexible Repair Layers for Neural Networks with Hard Constraints
+
+- **ID:** `2602.09317` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Ya-Chi Chu, Alkiviades Boukas, Madeleine Udell
+- **abs:** https://arxiv.org/abs/2602.09317 · **PDF:** https://arxiv.org/pdf/2602.09317v1
+
+## 13793. Effective MoE-based LLM Compression by Exploiting Heterogeneous Inter-Group Experts Routing Frequency and Information Density
+
+- **ID:** `2602.09316` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Zhendong Mi, Yixiao Chen, Pu Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.09316 · **PDF:** https://arxiv.org/pdf/2602.09316v2
+
+## 13794. A Deep Multi-Modal Method for Patient Wound Healing Assessment
+
+- **ID:** `2602.09315` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Subba Reddy Oota, Vijay Rowtula, Shahid Mohammed, et al.
+- **abs:** https://arxiv.org/abs/2602.09315 · **PDF:** https://arxiv.org/pdf/2602.09315v1
+
+## 13795. Learning to Select Like Humans: Explainable Active Learning for Medical Imaging
+
+- **ID:** `2602.13308` · **primary:** `eess.IV` · **Submitted:** 2026-02-10
+- **Authors:** Ifrat Ikhtear Uddin, Longwei Wang, Xiao Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.13308 · **PDF:** https://arxiv.org/pdf/2602.13308v3
+
+## 13796. Clarifying Shampoo: Adapting Spectral Descent to Stochasticity and the Parameter Trajectory
+
+- **ID:** `2602.09314` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Runa Eschenhagen, Anna Cai, Tsung-Hsien Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.09314 · **PDF:** https://arxiv.org/pdf/2602.09314v1
+
+## 13797. BiJEPA: Bi-directional Joint Embedding Predictive Architecture for Symmetric Representation Learning
+
+- **ID:** `2603.00049` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Yongchao Huang
+- **abs:** https://arxiv.org/abs/2603.00049 · **PDF:** https://arxiv.org/pdf/2603.00049v1
+
+## 13798. Don't Shoot The Breeze: Topic Continuity Model Using Nonlinear Naive Bayes With Attention
+
+- **ID:** `2602.09312` · **primary:** `cs.CL` · **Submitted:** 2026-02-10
+- **Authors:** Shu-Ting Pi, Pradeep Bagavan, Yejia Li, et al.
+- **abs:** https://arxiv.org/abs/2602.09312 · **PDF:** https://arxiv.org/pdf/2602.09312v1
+
+## 13799. How Far Can You Grow? Characterizing the Extrapolation Frontier of Graph Generative Models for Materials Science
+
+- **ID:** `2602.09309` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-10
+- **Authors:** Can Polat, Erchin Serpedin, Mustafa Kurban, et al.
+- **abs:** https://arxiv.org/abs/2602.09309 · **PDF:** https://arxiv.org/pdf/2602.09309v1
+
+## 13800. STRAND: Sequence-Conditioned Transport for Single-Cell Perturbations
+
+- **ID:** `2602.10156` · **primary:** `q-bio.GN` · **Submitted:** 2026-02-10
+- **Authors:** Boyang Fu, George Dasoulas, Sameer Gabbita, et al.
+- **abs:** https://arxiv.org/abs/2602.10156 · **PDF:** https://arxiv.org/pdf/2602.10156v1
+
+## 13801. Empowering Contrastive Federated Sequential Recommendation with LLMs
+
+- **ID:** `2602.09306` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Thi Minh Chau Nguyen, Minh Hieu Nguyen, Duc Anh Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.09306 · **PDF:** https://arxiv.org/pdf/2602.09306v1
+
+## 13802. Reward Modeling for Reinforcement Learning-Based LLM Reasoning: Design, Challenges, and Evaluation
+
+- **ID:** `2602.09305` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Pei-Chi Pan, Yingbin Liang, Sen Lin
+- **abs:** https://arxiv.org/abs/2602.09305 · **PDF:** https://arxiv.org/pdf/2602.09305v1
+
+## 13803. Statistical Roughness-Informed Machine Unlearning
+
+- **ID:** `2602.09304` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Mohammad Partohaghighi, Roummel Marcia, Bruce J. West, et al.
+- **abs:** https://arxiv.org/abs/2602.09304 · **PDF:** https://arxiv.org/pdf/2602.09304v1
+
+## 13804. Stabilizing Physics-Informed Consistency Models via Structure-Preserving Training
+
+- **ID:** `2602.09303` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Che-Chia Chang, Chen-Yang Dai, Te-Sheng Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.09303 · **PDF:** https://arxiv.org/pdf/2602.09303v1
+
+## 13805. Risk-sensitive reinforcement learning using expectiles, shortfall risk and optimized certainty equivalent risk
+
+- **ID:** `2602.09300` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Sumedh Gupte, Shrey Rakeshkumar Patel, Soumen Pachal, et al.
+- **abs:** https://arxiv.org/abs/2602.09300 · **PDF:** https://arxiv.org/pdf/2602.09300v1
+
+## 13806. The Laplacian Mechanism Improves Transformers by Reshaping Token Geometry
+
+- **ID:** `2602.09297` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Yuchong Zhang, Vardan Papyan
+- **abs:** https://arxiv.org/abs/2602.09297 · **PDF:** https://arxiv.org/pdf/2602.09297v1
+
+## 13807. Positive-Unlabelled Active Learning to Curate a Dataset for Orca Resident Interpretation
+
+- **ID:** `2602.09295` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Bret Nestor, Bohan Yao, Jasmine Moore, et al.
+- **abs:** https://arxiv.org/abs/2602.09295 · **PDF:** https://arxiv.org/pdf/2602.09295v1
+
+## 13808. Triggered: A Statistical Analysis of Environmental Influences on Extremist Groups
+
+- **ID:** `2602.09289` · **primary:** `cs.SI` · **Submitted:** 2026-02-10
+- **Authors:** Christine de Kock, Eduard Hovy
+- **abs:** https://arxiv.org/abs/2602.09289 · **PDF:** https://arxiv.org/pdf/2602.09289v1
+
+## 13809. Measuring Privacy Risks and Tradeoffs in Financial Synthetic Data Generation
+
+- **ID:** `2602.09288` · **primary:** `cs.LG` · **Submitted:** 2026-02-10
+- **Authors:** Michael Zuo, Inwon Kang, Stacy Patterson, et al.
+- **abs:** https://arxiv.org/abs/2602.09288 · **PDF:** https://arxiv.org/pdf/2602.09288v1
+
+## 13810. Human Control Is the Anchor, Not the Answer: Early Divergence of Oversight in Agentic AI Communities
+
+- **ID:** `2602.09286` · **primary:** `cs.AI` · **Submitted:** 2026-02-10
+- **Authors:** Hanjing Shi, Dominic DiFranzo
+- **abs:** https://arxiv.org/abs/2602.09286 · **PDF:** https://arxiv.org/pdf/2602.09286v1
+
+## 13811. X-Mark: Saliency-Guided Robust Dataset Ownership Verification for Medical Imaging
+
+- **ID:** `2602.09284` · **primary:** `cs.CV` · **Submitted:** 2026-02-10
+- **Authors:** Pranav Kulkarni, Junfeng Guo, Heng Huang
+- **abs:** https://arxiv.org/abs/2602.09284 · **PDF:** https://arxiv.org/pdf/2602.09284v1
+
+## 13812. The effect of whitening on explanation performance
+
+- **ID:** `2602.09278` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Benedict Clark, Stoyan Karastoyanov, Rick Wilming, et al.
+- **abs:** https://arxiv.org/abs/2602.09278 · **PDF:** https://arxiv.org/pdf/2602.09278v1
+
+## 13813. Mutual Information Collapse Explains Disentanglement Failure in $β$-VAEs
+
+- **ID:** `2602.09277` · **primary:** `stat.ML` · **Submitted:** 2026-02-09
+- **Authors:** Minh Vu, Xiaoliang Wan, Shuangqing Wei
+- **abs:** https://arxiv.org/abs/2602.09277 · **PDF:** https://arxiv.org/pdf/2602.09277v1
+
+## 13814. Effective Reasoning Chains Reduce Intrinsic Dimensionality
+
+- **ID:** `2602.09276` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Archiki Prasad, Mandar Joshi, Kenton Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.09276 · **PDF:** https://arxiv.org/pdf/2602.09276v1
+
+## 13815. Collective Behavior of AI Agents: the Case of Moltbook
+
+- **ID:** `2602.09270` · **primary:** `physics.soc-ph` · **Submitted:** 2026-02-09
+- **Authors:** Giordano De Marzo, David Garcia
+- **abs:** https://arxiv.org/abs/2602.09270 · **PDF:** https://arxiv.org/pdf/2602.09270v1
+
+## 13816. Measuring Inclusion in Interaction: Inclusion Analytics for Human-AI Collaborative Learning
+
+- **ID:** `2602.09269` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Jaeyoon Choi, Nia Nixon
+- **abs:** https://arxiv.org/abs/2602.09269 · **PDF:** https://arxiv.org/pdf/2602.09269v1
+
+## 13817. Generalizing GNNs with Tokenized Mixture of Experts
+
+- **ID:** `2602.09258` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Xiaoguang Guo, Zehong Wang, Jiazheng Li, et al.
+- **abs:** https://arxiv.org/abs/2602.09258 · **PDF:** https://arxiv.org/pdf/2602.09258v1
+
+## 13818. MOSAIC: Unveiling the Moral, Social and Individual Dimensions of Large Language Models
+
+- **ID:** `2603.00048` · **primary:** `cs.CY` · **Submitted:** 2026-02-09
+- **Authors:** Erica Coppolillo, Emilio Ferrara
+- **abs:** https://arxiv.org/abs/2603.00048 · **PDF:** https://arxiv.org/pdf/2603.00048v1
+
+## 13819. STaR: Scalable Task-Conditioned Retrieval for Long-Horizon Multimodal Robot Memory
+
+- **ID:** `2602.09255` · **primary:** `cs.RO` · **Submitted:** 2026-02-09
+- **Authors:** Mingfeng Yuan, Hao Zhang, Mahan Mohammadi, et al.
+- **abs:** https://arxiv.org/abs/2602.09255 · **PDF:** https://arxiv.org/pdf/2602.09255v2
+
+## 13820. VLM-Guided Iterative Refinement for Surgical Image Segmentation with Foundation Models
+
+- **ID:** `2602.09252` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Ange Lou, Yamin Li, Qi Chang, et al.
+- **abs:** https://arxiv.org/abs/2602.09252 · **PDF:** https://arxiv.org/pdf/2602.09252v1
+
+## 13821. Optimal Estimation in Orthogonally Invariant Generalized Linear Models: Spectral Initialization and Approximate Message Passing
+
+- **ID:** `2602.09240` · **primary:** `math.ST` · **Submitted:** 2026-02-09
+- **Authors:** Yihan Zhang, Hong Chang Ji, Ramji Venkataramanan, et al.
+- **abs:** https://arxiv.org/abs/2602.09240 · **PDF:** https://arxiv.org/pdf/2602.09240v1
+
+## 13822. Feature salience - not task-informativeness - drives machine learning model explanations
+
+- **ID:** `2602.09238` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Benedict Clark, Marta Oliveira, Rick Wilming, et al.
+- **abs:** https://arxiv.org/abs/2602.09238 · **PDF:** https://arxiv.org/pdf/2602.09238v3
+
+## 13823. RAPID: Risk of Attribute Prediction-Induced Disclosure in Synthetic Microdata
+
+- **ID:** `2602.09235` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Matthias Templ, Oscar Thees, Roman Müller
+- **abs:** https://arxiv.org/abs/2602.09235 · **PDF:** https://arxiv.org/pdf/2602.09235v1
+
+## 13824. Do Neural Networks Lose Plasticity in a Gradually Changing World?
+
+- **ID:** `2602.09234` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Tianhui Liu, Lili Mou
+- **abs:** https://arxiv.org/abs/2602.09234 · **PDF:** https://arxiv.org/pdf/2602.09234v1
+
+## 13825. CAGE: A Framework for Culturally Adaptive Red-Teaming Benchmark Generation
+
+- **ID:** `2602.20170` · **primary:** `cs.CY` · **Submitted:** 2026-02-09
+- **Authors:** Chaeyun Kim, YongTaek Lim, Kihyun Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.20170 · **PDF:** https://arxiv.org/pdf/2602.20170v1
+
+## 13826. Beyond the Unit Hypersphere: Embedding Magnitude in Contrastive Learning
+
+- **ID:** `2602.09229` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Xincan Feng, Taro Watanabe
+- **abs:** https://arxiv.org/abs/2602.09229 · **PDF:** https://arxiv.org/pdf/2602.09229v2
+
+## 13827. Barycentric alignment for instance-level comparison of neural representations
+
+- **ID:** `2602.09225` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Shreya Saha, Zoe Wanying He, Meenakshi Khosla
+- **abs:** https://arxiv.org/abs/2602.09225 · **PDF:** https://arxiv.org/pdf/2602.09225v1
+
+## 13828. MUZZLE: Adaptive Agentic Red-Teaming of Web Agents Against Indirect Prompt Injection Attacks
+
+- **ID:** `2602.09222` · **primary:** `cs.CR` · **Submitted:** 2026-02-09
+- **Authors:** Georgios Syros, Evan Rose, Brian Grinstead, et al.
+- **abs:** https://arxiv.org/abs/2602.09222 · **PDF:** https://arxiv.org/pdf/2602.09222v1
+
+## 13829. A Lightweight Multi-View Approach to Short-Term Load Forecasting
+
+- **ID:** `2602.09220` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Julien Guité-Vinet, Alexandre Blondin Massé, Éric Beaudry
+- **abs:** https://arxiv.org/abs/2602.09220 · **PDF:** https://arxiv.org/pdf/2602.09220v1
+
+## 13830. PRISM-XR: Empowering Privacy-Aware XR Collaboration with Multimodal Large Language Models
+
+- **ID:** `2602.10154` · **primary:** `cs.CR` · **Submitted:** 2026-02-09
+- **Authors:** Jiangong Chen, Mingyu Zhu, Bin Li
+- **abs:** https://arxiv.org/abs/2602.10154 · **PDF:** https://arxiv.org/pdf/2602.10154v1
+
+## 13831. CausalGDP: Causality-Guided Diffusion Policies for Reinforcement Learning
+
+- **ID:** `2602.09207` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Xiaofeng Xiao, Xiao Hu, Yang Ye, et al.
+- **abs:** https://arxiv.org/abs/2602.09207 · **PDF:** https://arxiv.org/pdf/2602.09207v1
+
+## 13832. EExApp: GNN-Based Reinforcement Learning for Radio Unit Energy Optimization in 5G O-RAN
+
+- **ID:** `2602.09206` · **primary:** `eess.SY` · **Submitted:** 2026-02-09
+- **Authors:** Jie Lu, Peihao Yan, Huacheng Zeng
+- **abs:** https://arxiv.org/abs/2602.09206 · **PDF:** https://arxiv.org/pdf/2602.09206v2
+
+## 13833. Genocide by Algorithm in Gaza: Artificial Intelligence, Countervailing Responsibility, and the Corruption of Public Discourse
+
+- **ID:** `2602.09202` · **primary:** `cs.CY` · **Submitted:** 2026-02-09
+- **Authors:** Branislav Radeljic
+- **abs:** https://arxiv.org/abs/2602.09202 · **PDF:** https://arxiv.org/pdf/2602.09202v1
+
+## 13834. Fair Feature Importance Scores via Feature Occlusion and Permutation
+
+- **ID:** `2602.09196` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Camille Little, Madeline Navarro, Santiago Segarra, et al.
+- **abs:** https://arxiv.org/abs/2602.09196 · **PDF:** https://arxiv.org/pdf/2602.09196v1
+
+## 13835. ML-DCN: Masked Low-Rank Deep Crossing Network Towards Scalable Ads Click-through Rate Prediction at Pinterest
+
+- **ID:** `2602.09194` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Jiacheng Li, Yixiong Meng, Yi wu, et al.
+- **abs:** https://arxiv.org/abs/2602.09194 · **PDF:** https://arxiv.org/pdf/2602.09194v1
+
+## 13836. Automated Concept Discovery for LLM-as-a-Judge Preference Analysis
+
+- **ID:** `2603.03319` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** James Wedgwood, Chhavi Yadav, Virginia Smith
+- **abs:** https://arxiv.org/abs/2603.03319 · **PDF:** https://arxiv.org/pdf/2603.03319v1
+
+## 13837. Gradient Residual Connections
+
+- **ID:** `2602.09190` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yangchen Pan, Qizhen Ying, Philip Torr, et al.
+- **abs:** https://arxiv.org/abs/2602.09190 · **PDF:** https://arxiv.org/pdf/2602.09190v1
+
+## 13838. AIDev: Studying AI Coding Agents on GitHub
+
+- **ID:** `2602.09185` · **primary:** `cs.SE` · **Submitted:** 2026-02-09
+- **Authors:** Hao Li, Haoxiang Zhang, Ahmed E. Hassan
+- **abs:** https://arxiv.org/abs/2602.09185 · **PDF:** https://arxiv.org/pdf/2602.09185v1
+
+## 13839. One RNG to Rule Them All: How Randomness Becomes an Attack Vector in Machine Learning
+
+- **ID:** `2602.09182` · **primary:** `cs.CR` · **Submitted:** 2026-02-09
+- **Authors:** Kotekar Annapoorna Prabhu, Andrew Gan, Zahra Ghodsi
+- **abs:** https://arxiv.org/abs/2602.09182 · **PDF:** https://arxiv.org/pdf/2602.09182v1
+
+## 13840. Weighted Wasserstein Barycenter of Gaussian Processes for exotic Bayesian Optimization tasks
+
+- **ID:** `2602.09181` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Antonio Candelieri, Francesco Archetti
+- **abs:** https://arxiv.org/abs/2602.09181 · **PDF:** https://arxiv.org/pdf/2602.09181v1
+
+## 13841. $n$-Musketeers: Reinforcement Learning Shapes Collaboration Among Language Models
+
+- **ID:** `2602.09173` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Ryozo Masukawa, Sanggeon Yun, Hyunwoo Oh, et al.
+- **abs:** https://arxiv.org/abs/2602.09173 · **PDF:** https://arxiv.org/pdf/2602.09173v1
+
+## 13842. Quantifying Epistemic Uncertainty in Diffusion Models
+
+- **ID:** `2602.09170` · **primary:** `stat.ML` · **Submitted:** 2026-02-09
+- **Authors:** Aditi Gupta, Raphael A. Meyer, Yotam Yaniv, et al.
+- **abs:** https://arxiv.org/abs/2602.09170 · **PDF:** https://arxiv.org/pdf/2602.09170v1
+
+## 13843. Train Less, Infer Faster: Efficient Model Finetuning and Compression via Structured Sparsity
+
+- **ID:** `2602.09169` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Jonathan Svirsky, Yehonathan Refael, Ofir Lindenbaum
+- **abs:** https://arxiv.org/abs/2602.09169 · **PDF:** https://arxiv.org/pdf/2602.09169v1
+
+## 13844. Faster Rates For Federated Variational Inequalities
+
+- **ID:** `2602.09164` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Guanghui Wang, Satyen Kale
+- **abs:** https://arxiv.org/abs/2602.09164 · **PDF:** https://arxiv.org/pdf/2602.09164v1
+
+## 13845. FlyAOC: Evaluating Agentic Ontology Curation of Drosophila Scientific Knowledge Bases
+
+- **ID:** `2602.09163` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Xingjian Zhang, Sophia Moylan, Ziyang Xiong, et al.
+- **abs:** https://arxiv.org/abs/2602.09163 · **PDF:** https://arxiv.org/pdf/2602.09163v1
+
+## 13846. Boltzmann Reinforcement Learning for Noise resilience in Analog Ising Machines
+
+- **ID:** `2602.09162` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Aditya Choudhary, Saaketh Desai, Prasad Iyer
+- **abs:** https://arxiv.org/abs/2602.09162 · **PDF:** https://arxiv.org/pdf/2602.09162v1
+
+## 13847. Minimum Distance Summaries for Robust Neural Posterior Estimation
+
+- **ID:** `2602.09161` · **primary:** `stat.ML` · **Submitted:** 2026-02-09
+- **Authors:** Sherman Khoo, Dennis Prangle, Song Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.09161 · **PDF:** https://arxiv.org/pdf/2602.09161v1
+
+## 13848. CoMMa: Contribution-Aware Medical Multi-Agents From A Game-Theoretic Perspective
+
+- **ID:** `2602.09159` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Yichen Wu, Yujin Oh, Sangjoon Park, et al.
+- **abs:** https://arxiv.org/abs/2602.09159 · **PDF:** https://arxiv.org/pdf/2602.09159v1
+
+## 13849. What do Geometric Hallucination Detection Metrics Actually Measure?
+
+- **ID:** `2602.09158` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Eric Yeats, John Buckheit, Sarah Scullen, et al.
+- **abs:** https://arxiv.org/abs/2602.09158 · **PDF:** https://arxiv.org/pdf/2602.09158v1
+
+## 13850. Decoding Future Risk: Deep Learning Analysis of Tubular Adenoma Whole-Slide Images
+
+- **ID:** `2602.09155` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Ahmed Rahu, Brian Shula, Brandon Combs, et al.
+- **abs:** https://arxiv.org/abs/2602.09155 · **PDF:** https://arxiv.org/pdf/2602.09155v1
+
+## 13851. A Hybrid Deterministic Framework for Named Entity Extraction in Broadcast News Video
+
+- **ID:** `2602.09154` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Andrea Filiberto Lucas, Dylan Seychell
+- **abs:** https://arxiv.org/abs/2602.09154 · **PDF:** https://arxiv.org/pdf/2602.09154v1
+
+## 13852. SceneSmith: Agentic Generation of Simulation-Ready Indoor Scenes
+
+- **ID:** `2602.09153` · **primary:** `cs.RO` · **Submitted:** 2026-02-09
+- **Authors:** Nicholas Pfaff, Thomas Cohn, Sergey Zakharov, et al.
+- **abs:** https://arxiv.org/abs/2602.09153 · **PDF:** https://arxiv.org/pdf/2602.09153v1
+
+## 13853. Basic Legibility Protocols Improve Trusted Monitoring
+
+- **ID:** `2602.10153` · **primary:** `cs.CR` · **Submitted:** 2026-02-09
+- **Authors:** Ashwin Sreevatsa, Sebastian Prasanna, Cody Rushing
+- **abs:** https://arxiv.org/abs/2602.10153 · **PDF:** https://arxiv.org/pdf/2602.10153v1
+
+## 13854. Fine-Tuning a Large Vision-Language Model for Artwork's Scoring and Critique
+
+- **ID:** `2602.13306` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Zhehan Zhang, Meihua Qian, Li Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.13306 · **PDF:** https://arxiv.org/pdf/2602.13306v1
+
+## 13855. Overview of PAN 2026: Voight-Kampff Generative AI Detection, Text Watermarking, Multi-Author Writing Style Analysis, Generative Plagiarism Detection, and Reasoning Trajectory Detection
+
+- **ID:** `2602.09147` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Janek Bevendorff, Maik Fröbe, André Greiner-Petter, et al.
+- **abs:** https://arxiv.org/abs/2602.09147 · **PDF:** https://arxiv.org/pdf/2602.09147v1
+
+## 13856. WildfireVLM: AI-powered Analysis for Early Wildfire Detection and Risk Assessment Using Satellite Imagery
+
+- **ID:** `2602.13305` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Aydin Ayanzadeh, Prakhar Dixit, Sadia Kamal, et al.
+- **abs:** https://arxiv.org/abs/2602.13305 · **PDF:** https://arxiv.org/pdf/2602.13305v1
+
+## 13857. PABU: Progress-Aware Belief Update for Efficient LLM Agents
+
+- **ID:** `2602.09138` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Haitao Jiang, Lin Ge, Hengrui Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.09138 · **PDF:** https://arxiv.org/pdf/2602.09138v1
+
+## 13858. PCReg-Net: Progressive Contrast-Guided Registration for Cross-Domain Image Alignment
+
+- **ID:** `2602.13304` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Jiahao Qin
+- **abs:** https://arxiv.org/abs/2602.13304 · **PDF:** https://arxiv.org/pdf/2602.13304v2
+
+## 13859. Validating Interpretability in siRNA Efficacy Prediction: A Perturbation-Based, Dataset-Aware Protocol
+
+- **ID:** `2602.10152` · **primary:** `q-bio.GN` · **Submitted:** 2026-02-09
+- **Authors:** Zahra Khodagholi, Niloofar Yousefi
+- **abs:** https://arxiv.org/abs/2602.10152 · **PDF:** https://arxiv.org/pdf/2602.10152v2
+
+## 13860. UniComp: A Unified Evaluation of Large Language Model Compression via Pruning, Quantization and Distillation
+
+- **ID:** `2602.09130` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Jonathan von Rad, Yong Cao, Andreas Geiger
+- **abs:** https://arxiv.org/abs/2602.09130 · **PDF:** https://arxiv.org/pdf/2602.09130v2
+
+## 13861. Counterfactual Maps: What They Are and How to Find Them
+
+- **ID:** `2602.09128` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Awa Khouna, Julien Ferry, Thibaut Vidal
+- **abs:** https://arxiv.org/abs/2602.09128 · **PDF:** https://arxiv.org/pdf/2602.09128v1
+
+## 13862. Epistemic Throughput: Fundamental Limits of Attention-Constrained Inference
+
+- **ID:** `2602.09127` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Lei You
+- **abs:** https://arxiv.org/abs/2602.09127 · **PDF:** https://arxiv.org/pdf/2602.09127v2
+
+## 13863. Uncertainty-Aware Multimodal Emotion Recognition through Dirichlet Parameterization
+
+- **ID:** `2602.09121` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Rémi Grzeczkowicz, Eric Soriano, Ali Janati, et al.
+- **abs:** https://arxiv.org/abs/2602.09121 · **PDF:** https://arxiv.org/pdf/2602.09121v1
+
+## 13864. SpinCastML an Open Decision-Making Application for Inverse Design of Electrospinning Manufacturing: A Machine Learning, Optimal Sampling and Inverse Monte Carlo Approach
+
+- **ID:** `2602.09120` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Elisa Roldan, Tasneem Sabir
+- **abs:** https://arxiv.org/abs/2602.09120 · **PDF:** https://arxiv.org/pdf/2602.09120v1
+
+## 13865. Importance inversion transfer identifies shared principles for cross-domain learning
+
+- **ID:** `2602.09116` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Daniele Caligiore
+- **abs:** https://arxiv.org/abs/2602.09116 · **PDF:** https://arxiv.org/pdf/2602.09116v2
+
+## 13866. Benchmarking the Energy Savings with Speculative Decoding Strategies
+
+- **ID:** `2602.09113` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Rohit Dutta, Paramita Koley, Soham Poddar, et al.
+- **abs:** https://arxiv.org/abs/2602.09113 · **PDF:** https://arxiv.org/pdf/2602.09113v1
+
+## 13867. A Small-Scale System for Autoregressive Program Synthesis Enabling Controlled Experimentation
+
+- **ID:** `2602.09112` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Russ Webb, Jason Ramapuram
+- **abs:** https://arxiv.org/abs/2602.09112 · **PDF:** https://arxiv.org/pdf/2602.09112v1
+
+## 13868. Distributed Hybrid Parallelism for Large Language Models: Comparative Study and System Design Guide
+
+- **ID:** `2602.09109` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Hossam Amer, Rezaul Karim, Ali Pourranjbar, et al.
+- **abs:** https://arxiv.org/abs/2602.09109 · **PDF:** https://arxiv.org/pdf/2602.09109v1
+
+## 13869. From Adam to Adam-Like Lagrangians: Second-Order Nonlocal Dynamics
+
+- **ID:** `2602.09101` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Carlos Heredia
+- **abs:** https://arxiv.org/abs/2602.09101 · **PDF:** https://arxiv.org/pdf/2602.09101v1
+
+## 13870. Predicting magnetism with first-principles AI
+
+- **ID:** `2602.09093` · **primary:** `cond-mat.str-el` · **Submitted:** 2026-02-09
+- **Authors:** Max Geier, Liang Fu
+- **abs:** https://arxiv.org/abs/2602.09093 · **PDF:** https://arxiv.org/pdf/2602.09093v1
+
+## 13871. Quantum-Inspired Self-Attention in a Large Language Model
+
+- **ID:** `2603.03318` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Nikita Kuznetsov, Niyaz Ismagilov, Ernesto Campos
+- **abs:** https://arxiv.org/abs/2603.03318 · **PDF:** https://arxiv.org/pdf/2603.03318v1
+
+## 13872. Robustness Is a Function, Not a Number: A Factorized Comprehensive Study of OOD Robustness in Vision-Based Driving
+
+- **ID:** `2602.09018` · **primary:** `cs.RO` · **Submitted:** 2026-02-09
+- **Authors:** Amir Mallak, Alaa Maalouf
+- **abs:** https://arxiv.org/abs/2602.09018 · **PDF:** https://arxiv.org/pdf/2602.09018v1
+
+## 13873. Autonomous AI and Ownership Rules
+
+- **ID:** `2602.20169` · **primary:** `cs.CY` · **Submitted:** 2026-02-09
+- **Authors:** Frank Fagan
+- **abs:** https://arxiv.org/abs/2602.20169 · **PDF:** https://arxiv.org/pdf/2602.20169v1
+
+## 13874. Contact-Anchored Policies: Contact Conditioning Creates Strong Robot Utility Models
+
+- **ID:** `2602.09017` · **primary:** `cs.RO` · **Submitted:** 2026-02-09
+- **Authors:** Zichen Jeff Cui, Omar Rayyan, Haritheja Etukuru, et al.
+- **abs:** https://arxiv.org/abs/2602.09017 · **PDF:** https://arxiv.org/pdf/2602.09017v1
+
+## 13875. CIC-Trap4Phish: A Unified Multi-Format Dataset for Phishing and Quishing Attachment Detection
+
+- **ID:** `2602.09015` · **primary:** `cs.CR` · **Submitted:** 2026-02-09
+- **Authors:** Fatemeh Nejati, Mahdi Rabbani, Morteza Eskandarian, et al.
+- **abs:** https://arxiv.org/abs/2602.09015 · **PDF:** https://arxiv.org/pdf/2602.09015v3
+
+## 13876. ArcFlow: Unleashing 2-Step Text-to-Image Generation via High-Precision Non-Linear Flow Distillation
+
+- **ID:** `2602.09014` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Zihan Yang, Shuyuan Tu, Licheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.09014 · **PDF:** https://arxiv.org/pdf/2602.09014v1
+
+## 13877. Next-Gen CAPTCHAs: Leveraging the Cognitive Gap for Scalable and Diverse GUI-Agent Defense
+
+- **ID:** `2602.09012` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Jiacheng Liu, Yaxin Luo, Jiacheng Cui, et al.
+- **abs:** https://arxiv.org/abs/2602.09012 · **PDF:** https://arxiv.org/pdf/2602.09012v1
+
+## 13878. ANCRe: Adaptive Neural Connection Reassignment for Efficient Depth Scaling
+
+- **ID:** `2602.09009` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yilang Zhang, Bingcong Li, Niao He, et al.
+- **abs:** https://arxiv.org/abs/2602.09009 · **PDF:** https://arxiv.org/pdf/2602.09009v1
+
+## 13879. ShapeCond: Fast Shapelet-Guided Dataset Condensation for Time Series Classification
+
+- **ID:** `2602.09008` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Sijia Peng, Yun Xiong, Xi Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.09008 · **PDF:** https://arxiv.org/pdf/2602.09008v1
+
+## 13880. GEBench: Benchmarking Image Generation Models as GUI Environments
+
+- **ID:** `2602.09007` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Haodong Li, Jingwei Wu, Quan Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.09007 · **PDF:** https://arxiv.org/pdf/2602.09007v2
+
+## 13881. ARO: A New Lens On Matrix Optimization For Large Models
+
+- **ID:** `2602.09006` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Wenbo Gong, Javier Zazo, Qijun Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.09006 · **PDF:** https://arxiv.org/pdf/2602.09006v1
+
+## 13882. SinFoS: A Parallel Dataset for Translating Sinhala Figures of Speech
+
+- **ID:** `2602.09866` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Johan Sofalas, Dilushri Pavithra, Nevidu Jayatilleke, et al.
+- **abs:** https://arxiv.org/abs/2602.09866 · **PDF:** https://arxiv.org/pdf/2602.09866v1
+
+## 13883. Data Science and Technology Towards AGI Part I: Tiered Data Management
+
+- **ID:** `2602.09003` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Yudong Wang, Zixuan Fu, Hengyu Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.09003 · **PDF:** https://arxiv.org/pdf/2602.09003v1
+
+## 13884. From Obstacles to Etiquette: Robot Social Navigation with VLM-Informed Path Selection
+
+- **ID:** `2602.09002` · **primary:** `cs.RO` · **Submitted:** 2026-02-09
+- **Authors:** Zilin Fang, Anxing Xiao, David Hsu, et al.
+- **abs:** https://arxiv.org/abs/2602.09002 · **PDF:** https://arxiv.org/pdf/2602.09002v1
+
+## 13885. DirMoE: Dirichlet-routed Mixture of Experts
+
+- **ID:** `2602.09001` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Amirhossein Vahidi, Hesam Asadollahzadeh, Navid Akhavan Attar, et al.
+- **abs:** https://arxiv.org/abs/2602.09001 · **PDF:** https://arxiv.org/pdf/2602.09001v1
+
+## 13886. iGRPO: Self-Feedback-Driven LLM Reasoning
+
+- **ID:** `2602.09000` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Ali Hatamizadeh, Shrimai Prabhumoye, Igor Gitman, et al.
+- **abs:** https://arxiv.org/abs/2602.09000 · **PDF:** https://arxiv.org/pdf/2602.09000v1
+
+## 13887. UI-Venus-1.5 Technical Report
+
+- **ID:** `2602.09082` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Venus Team, Changlong Gao, Zhangxuan Gu, et al.
+- **abs:** https://arxiv.org/abs/2602.09082 · **PDF:** https://arxiv.org/pdf/2602.09082v2
+
+## 13888. Universal Coefficients and Mayer-Vietoris Sequence for Groupoid Homology
+
+- **ID:** `2602.08998` · **primary:** `math.AT` · **Submitted:** 2026-02-09
+- **Authors:** Luciano Melodia
+- **abs:** https://arxiv.org/abs/2602.08998 · **PDF:** https://arxiv.org/pdf/2602.08998v4
+
+## 13889. Paradox of De-identification: A Critique of HIPAA Safe Harbour in the Age of LLMs
+
+- **ID:** `2602.08997` · **primary:** `cs.CY` · **Submitted:** 2026-02-09
+- **Authors:** Lavender Y. Jiang, Xujin Chris Liu, Kyunghyun Cho, et al.
+- **abs:** https://arxiv.org/abs/2602.08997 · **PDF:** https://arxiv.org/pdf/2602.08997v1
+
+## 13890. DMamba: Decomposition-enhanced Mamba for Time Series Forecasting
+
+- **ID:** `2602.09081` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Ruxuan Chen, Fang Sun
+- **abs:** https://arxiv.org/abs/2602.09081 · **PDF:** https://arxiv.org/pdf/2602.09081v1
+
+## 13891. When Actions Go Off-Task: Detecting and Correcting Misaligned Actions in Computer-Use Agents
+
+- **ID:** `2602.08995` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Yuting Ning, Jaylen Jones, Zhehao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.08995 · **PDF:** https://arxiv.org/pdf/2602.08995v1
+
+## 13892. PEST: Physics-Enhanced Swin Transformer for 3D Turbulence Simulation
+
+- **ID:** `2602.10150` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-02-09
+- **Authors:** Yilong Dai, Shengyu Chen, Xiaowei Jia, et al.
+- **abs:** https://arxiv.org/abs/2602.10150 · **PDF:** https://arxiv.org/pdf/2602.10150v1
+
+## 13893. Semantic Labeling for Third-Party Cybersecurity Risk Assessment: A Semi-Supervised Approach to Intent-Aware Question Retrieval
+
+- **ID:** `2602.10149` · **primary:** `cs.CR` · **Submitted:** 2026-02-09
+- **Authors:** Ali Nour Eldin, Mohamed Sellami, Mehdi Acheli, et al.
+- **abs:** https://arxiv.org/abs/2602.10149 · **PDF:** https://arxiv.org/pdf/2602.10149v3
+
+## 13894. ZUNA: Flexible EEG Superresolution with Position-Aware Diffusion Autoencoders
+
+- **ID:** `2602.18478` · **primary:** `eess.SP` · **Submitted:** 2026-02-09
+- **Authors:** Christopher Warner, Jonas Mago, JR Huml, et al.
+- **abs:** https://arxiv.org/abs/2602.18478 · **PDF:** https://arxiv.org/pdf/2602.18478v1
+
+## 13895. InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discovery
+
+- **ID:** `2602.08990` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Shiyang Feng, Runmin Ma, Xiangchao Yan, et al.
+- **abs:** https://arxiv.org/abs/2602.08990 · **PDF:** https://arxiv.org/pdf/2602.08990v1
+
+## 13896. Improving Detection of Rare Nodes in Hierarchical Multi-Label Learning
+
+- **ID:** `2602.08986` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Isaac Xu, Martin Gillis, Ayushi Sharma, et al.
+- **abs:** https://arxiv.org/abs/2602.08986 · **PDF:** https://arxiv.org/pdf/2602.08986v1
+
+## 13897. Next Concept Prediction in Discrete Latent Space Leads to Stronger Language Models
+
+- **ID:** `2602.08984` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Yuliang Liu, Yunchong Song, Yixuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.08984 · **PDF:** https://arxiv.org/pdf/2602.08984v1
+
+## 13898. Retcon -- a Prompt-Based Technique for Precise Control of LLMs in Conversations
+
+- **ID:** `2603.03317` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** David Kogan, Sam Nguyen, Masanori Suzuki, et al.
+- **abs:** https://arxiv.org/abs/2603.03317 · **PDF:** https://arxiv.org/pdf/2603.03317v1
+
+## 13899. Red-teaming the Multimodal Reasoning: Jailbreaking Vision-Language Models via Cross-modal Entanglement Attacks
+
+- **ID:** `2602.10148` · **primary:** `cs.CR` · **Submitted:** 2026-02-09
+- **Authors:** Yu Yan, Sheng Sun, Shengjia Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.10148 · **PDF:** https://arxiv.org/pdf/2602.10148v1
+
+## 13900. StretchTime: Adaptive Time Series Forecasting via Symplectic Attention
+
+- **ID:** `2602.08983` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yubin Kim, Viresh Pati, Jevon Twitty, et al.
+- **abs:** https://arxiv.org/abs/2602.08983 · **PDF:** https://arxiv.org/pdf/2602.08983v1
+
+## 13901. When do neural ordinary differential equations generalize on complex networks?
+
+- **ID:** `2602.08980` · **primary:** `physics.soc-ph` · **Submitted:** 2026-02-09
+- **Authors:** Moritz Laber, Tina Eliassi-Rad, Brennan Klein
+- **abs:** https://arxiv.org/abs/2602.08980 · **PDF:** https://arxiv.org/pdf/2602.08980v1
+
+## 13902. Beyond Transcripts: A Renewed Perspective on Audio Chaptering
+
+- **ID:** `2602.08979` · **primary:** `cs.SD` · **Submitted:** 2026-02-09
+- **Authors:** Fabian Retkowski, Maike Züfle, Thai Binh Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.08979 · **PDF:** https://arxiv.org/pdf/2602.08979v1
+
+## 13903. Distributionally Robust Optimization via Generative Ambiguity Modeling
+
+- **ID:** `2602.08976` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Jiaqi Wen, Jianyi Yang
+- **abs:** https://arxiv.org/abs/2602.08976 · **PDF:** https://arxiv.org/pdf/2602.08976v1
+
+## 13904. stable-worldmodel-v1: Reproducible World Modeling Research and Evaluation
+
+- **ID:** `2602.08968` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Lucas Maes, Quentin Le Lidec, Dan Haramati, et al.
+- **abs:** https://arxiv.org/abs/2602.08968 · **PDF:** https://arxiv.org/pdf/2602.08968v2
+
+## 13905. Learning to Coordinate via Quantum Entanglement in Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.08965` · **primary:** `cs.MA` · **Submitted:** 2026-02-09
+- **Authors:** John Gardiner, Orlando Romero, Brendan Tivnan, et al.
+- **abs:** https://arxiv.org/abs/2602.08965 · **PDF:** https://arxiv.org/pdf/2602.08965v2
+
+## 13906. A Behavioural and Representational Evaluation of Goal-Directedness in Language Model Agents
+
+- **ID:** `2602.08964` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Raghu Arghal, Fade Chen, Niall Dalton, et al.
+- **abs:** https://arxiv.org/abs/2602.08964 · **PDF:** https://arxiv.org/pdf/2602.08964v1
+
+## 13907. Looping Back to Move Forward: Recursive Transformers for Efficient and Flexible Large Multimodal Models
+
+- **ID:** `2602.09080` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Ruihan Xu, Yuting Gao, Lan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.09080 · **PDF:** https://arxiv.org/pdf/2602.09080v1
+
+## 13908. MotionCrafter: Dense Geometry and Motion Reconstruction with a 4D VAE
+
+- **ID:** `2602.08961` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Ruijie Zhu, Jiahao Lu, Wenbo Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.08961 · **PDF:** https://arxiv.org/pdf/2602.08961v2
+
+## 13909. Spectral Collapse in Diffusion Inversion
+
+- **ID:** `2602.13303` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Nicolas Bourriez, Alexandre Verine, Auguste Genovesio
+- **abs:** https://arxiv.org/abs/2602.13303 · **PDF:** https://arxiv.org/pdf/2602.13303v1
+
+## 13910. How Should We Model the Probability of a Language?
+
+- **ID:** `2602.08951` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Rasul Dent, Pedro Ortiz Suarez, Thibault Clérice, et al.
+- **abs:** https://arxiv.org/abs/2602.08951 · **PDF:** https://arxiv.org/pdf/2602.08951v1
+
+## 13911. Digital Twin and Agentic AI for Wild Fire Disaster Management: Intelligent Virtual Situation Room
+
+- **ID:** `2602.08949` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Mohammad Morsali, Siavash H. Khajavi
+- **abs:** https://arxiv.org/abs/2602.08949 · **PDF:** https://arxiv.org/pdf/2602.08949v1
+
+## 13912. CoRefine: Confidence-Guided Self-Refinement for Adaptive Test-Time Compute
+
+- **ID:** `2602.08948` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Chen Jin, Ryutaro Tanno, Tom Diethe, et al.
+- **abs:** https://arxiv.org/abs/2602.08948 · **PDF:** https://arxiv.org/pdf/2602.08948v1
+
+## 13913. Patient foundation model for risk stratification in low-risk overweight patients
+
+- **ID:** `2602.09079` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Zachary N. Flamholz, Dillon Tracy, Ripple Khera, et al.
+- **abs:** https://arxiv.org/abs/2602.09079 · **PDF:** https://arxiv.org/pdf/2602.09079v1
+
+## 13914. GitSearch: Enhancing Community Notes Generation with Gap-Informed Targeted Search
+
+- **ID:** `2602.08945` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Sahajpreet Singh, Kokil Jaidka, Min-Yen Kan
+- **abs:** https://arxiv.org/abs/2602.08945 · **PDF:** https://arxiv.org/pdf/2602.08945v1
+
+## 13915. pixelLOG: Logging of Online Gameplay for Cognitive Research
+
+- **ID:** `2602.08941` · **primary:** `cs.HC` · **Submitted:** 2026-02-09
+- **Authors:** Zeyu Lu, Dennis L. Barbour
+- **abs:** https://arxiv.org/abs/2602.08941 · **PDF:** https://arxiv.org/pdf/2602.08941v1
+
+## 13916. CausalT5K: Diagnosing and Informing Refusal for Trustworthy Causal Reasoning of Skepticism, Sycophancy, Detection-Correction, and Rung Collapse
+
+- **ID:** `2602.08939` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Longling Geng, Andy Ouyang, Theodore Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.08939 · **PDF:** https://arxiv.org/pdf/2602.08939v1
+
+## 13917. StealthRL: Reinforcement Learning Paraphrase Attacks for Multi-Detector Evasion of AI-Text Detectors
+
+- **ID:** `2602.08934` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Suraj Ranganath, Atharv Ramesh
+- **abs:** https://arxiv.org/abs/2602.08934 · **PDF:** https://arxiv.org/pdf/2602.08934v2
+
+## 13918. Provably robust learning of regression neural networks using $β$-divergences
+
+- **ID:** `2602.08933` · **primary:** `stat.ML` · **Submitted:** 2026-02-09
+- **Authors:** Abhik Ghosh, Suryasis Jana
+- **abs:** https://arxiv.org/abs/2602.08933 · **PDF:** https://arxiv.org/pdf/2602.08933v1
+
+## 13919. Online monotone density estimation and log-optimal calibration
+
+- **ID:** `2602.08927` · **primary:** `stat.ML` · **Submitted:** 2026-02-09
+- **Authors:** Rohan Hore, Ruodu Wang, Aaditya Ramdas
+- **abs:** https://arxiv.org/abs/2602.08927 · **PDF:** https://arxiv.org/pdf/2602.08927v2
+
+## 13920. The Influence of Iconicity in Transfer Learning for Sign Language Recognition
+
+- **ID:** `2603.03316` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Keren Artiaga, Conor Lynch, Haithem Afli, et al.
+- **abs:** https://arxiv.org/abs/2603.03316 · **PDF:** https://arxiv.org/pdf/2603.03316v1
+
+## 13921. A Hybrid Machine Learning Approach for Graduate Admission Prediction and Combined University-Program Recommendation
+
+- **ID:** `2603.29881` · **primary:** `cs.IR` · **Submitted:** 2026-02-09
+- **Authors:** Melina Heidari Far, Elham Tabrizi
+- **abs:** https://arxiv.org/abs/2603.29881 · **PDF:** https://arxiv.org/pdf/2603.29881v1
+
+## 13922. DynamiQ: Accelerating Gradient Synchronization using Compressed Multi-hop All-reduce
+
+- **ID:** `2602.08923` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Wenchen Han, Shay Vargaftik, Michael Mitzenmacher, et al.
+- **abs:** https://arxiv.org/abs/2602.08923 · **PDF:** https://arxiv.org/pdf/2602.08923v1
+
+## 13923. Diffusion-Inspired Reconfiguration of Transformers for Uncertainty Calibration
+
+- **ID:** `2602.08920` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Manh Cuong Dao, Quang Hung Pham, Phi Le Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.08920 · **PDF:** https://arxiv.org/pdf/2602.08920v1
+
+## 13924. Automatic In-Domain Exemplar Construction and LLM-Based Refinement of Multi-LLM Expansions for Query Expansion
+
+- **ID:** `2602.08917` · **primary:** `cs.IR` · **Submitted:** 2026-02-09
+- **Authors:** Minghan Li, Ercong Nie, Siqi Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.08917 · **PDF:** https://arxiv.org/pdf/2602.08917v2
+
+## 13925. AMS-HD: Hyperdimensional Computing for Real-Time and Energy-Efficient Acute Mountain Sickness Detection
+
+- **ID:** `2602.08916` · **primary:** `cs.SC` · **Submitted:** 2026-02-09
+- **Authors:** Abu Masum, Mehran Moghadam, M. Hassan Najafi, et al.
+- **abs:** https://arxiv.org/abs/2602.08916 · **PDF:** https://arxiv.org/pdf/2602.08916v1
+
+## 13926. Framework for Integrating Zero Trust in Cloud-Based Endpoint Security for Critical Infrastructure
+
+- **ID:** `2602.09078` · **primary:** `cs.CR` · **Submitted:** 2026-02-09
+- **Authors:** Shyam Kumar Gajula
+- **abs:** https://arxiv.org/abs/2602.09078 · **PDF:** https://arxiv.org/pdf/2602.09078v1
+
+## 13927. Gesturing Toward Abstraction: Multimodal Convention Formation in Collaborative Physical Tasks
+
+- **ID:** `2602.08914` · **primary:** `cs.HC` · **Submitted:** 2026-02-09
+- **Authors:** Kiyosu Maeda, William P. McCarthy, Ching-Yi Tsai, et al.
+- **abs:** https://arxiv.org/abs/2602.08914 · **PDF:** https://arxiv.org/pdf/2602.08914v1
+
+## 13928. GEMSS: A Variational Bayesian Method for Discovering Multiple Sparse Solutions in Classification and Regression Problems
+
+- **ID:** `2602.08913` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Kateřina Henclová, Václav Šmídl
+- **abs:** https://arxiv.org/abs/2602.08913 · **PDF:** https://arxiv.org/pdf/2602.08913v1
+
+## 13929. Analysis of Converged 3D Gaussian Splatting Solutions: Density Effects and Prediction Limit
+
+- **ID:** `2602.08909` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Zhendong Wang, Cihan Ruan, Jingchuan Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.08909 · **PDF:** https://arxiv.org/pdf/2602.08909v1
+
+## 13930. Positive Distribution Shift as a Framework for Understanding Tractable Learning
+
+- **ID:** `2602.08907` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Marko Medvedev, Idan Attias, Elisabetta Cornacchia, et al.
+- **abs:** https://arxiv.org/abs/2602.08907 · **PDF:** https://arxiv.org/pdf/2602.08907v2
+
+## 13931. Efficient and Stable Reinforcement Learning for Diffusion Language Models
+
+- **ID:** `2602.08905` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Jiawei Liu, Xiting Wang, Yuanyuan Zhong, et al.
+- **abs:** https://arxiv.org/abs/2602.08905 · **PDF:** https://arxiv.org/pdf/2602.08905v1
+
+## 13932. GSS: Gated Subspace Steering for Selective Memorization Mitigation in LLMs
+
+- **ID:** `2602.08901` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Xuanqi Zhang, Haoyang Shang, Xiaoxiao Li
+- **abs:** https://arxiv.org/abs/2602.08901 · **PDF:** https://arxiv.org/pdf/2602.08901v1
+
+## 13933. Large Language Models and Book Summarization: Reading or Remembering, Which Is Better?
+
+- **ID:** `2603.09981` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Tairan Fu, Javier Conde, Pedro Reviriego, et al.
+- **abs:** https://arxiv.org/abs/2603.09981 · **PDF:** https://arxiv.org/pdf/2603.09981v1
+
+## 13934. OmniReview: A Large-scale Benchmark and LLM-enhanced Framework for Realistic Reviewer Recommendation
+
+- **ID:** `2602.08896` · **primary:** `cs.IR` · **Submitted:** 2026-02-09
+- **Authors:** Yehua Huang, Penglei Sun, Zebin Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.08896 · **PDF:** https://arxiv.org/pdf/2602.08896v1
+
+## 13935. M-QUEST -- Meme Question-Understanding Evaluation on Semantics and Toxicity
+
+- **ID:** `2603.03315` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Stefano De Giorgis, Ting-Chih Chen, Filip Ilievski
+- **abs:** https://arxiv.org/abs/2603.03315 · **PDF:** https://arxiv.org/pdf/2603.03315v1
+
+## 13936. Discrete Bridges for Mutual Information Estimation
+
+- **ID:** `2602.08894` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Iryna Zabarianska, Sergei Kholkin, Grigoriy Ksenofontov, et al.
+- **abs:** https://arxiv.org/abs/2602.08894 · **PDF:** https://arxiv.org/pdf/2602.08894v1
+
+## 13937. Winner's Curse Drives False Promises in Data-Driven Decisions: A Case Study in Refugee Matching
+
+- **ID:** `2602.08892` · **primary:** `stat.ML` · **Submitted:** 2026-02-09
+- **Authors:** Hamsa Bastani, Osbert Bastani, Bryce McLaughlin
+- **abs:** https://arxiv.org/abs/2602.08892 · **PDF:** https://arxiv.org/pdf/2602.08892v1
+
+## 13938. Scalable Delphi: Large Language Models for Structured Risk Estimation
+
+- **ID:** `2602.08889` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Tobias Lorenz, Mario Fritz
+- **abs:** https://arxiv.org/abs/2602.08889 · **PDF:** https://arxiv.org/pdf/2602.08889v1
+
+## 13939. DeepQuali: Initial results of a study on the use of large language models for assessing the quality of user stories
+
+- **ID:** `2602.08887` · **primary:** `cs.SE` · **Submitted:** 2026-02-09
+- **Authors:** Adam Trendowicz, Daniel Seifert, Andreas Jedlitschka, et al.
+- **abs:** https://arxiv.org/abs/2602.08887 · **PDF:** https://arxiv.org/pdf/2602.08887v1
+
+## 13940. Contrastive Learning for Diversity-Aware Product Recommendations in Retail
+
+- **ID:** `2602.08886` · **primary:** `cs.IR` · **Submitted:** 2026-02-09
+- **Authors:** Vasileios Karlis, Ezgi Yıldırım, David Vos, et al.
+- **abs:** https://arxiv.org/abs/2602.08886 · **PDF:** https://arxiv.org/pdf/2602.08886v1
+
+## 13941. Breaking the Simplification Bottleneck in Amortized Neural Symbolic Regression
+
+- **ID:** `2602.08885` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Paul Saegert, Ullrich Köthe
+- **abs:** https://arxiv.org/abs/2602.08885 · **PDF:** https://arxiv.org/pdf/2602.08885v4
+
+## 13942. Differentiable Logical Programming for Quantum Circuit Discovery and Optimization
+
+- **ID:** `2602.08880` · **primary:** `quant-ph` · **Submitted:** 2026-02-09
+- **Authors:** Antonin Sulc
+- **abs:** https://arxiv.org/abs/2602.08880 · **PDF:** https://arxiv.org/pdf/2602.08880v1
+
+## 13943. Learning Potentials for Dynamic Matching and Application to Heart Transplantation
+
+- **ID:** `2602.08878` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Itai Zilberstein, Ioannis Anagnostides, Zachary W. Sollie, et al.
+- **abs:** https://arxiv.org/abs/2602.08878 · **PDF:** https://arxiv.org/pdf/2602.08878v1
+
+## 13944. Stress-Testing Alignment Audits With Prompt-Level Strategic Deception
+
+- **ID:** `2602.08877` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Oliver Daniels, Perusha Moodley, Benjamin M. Marlin, et al.
+- **abs:** https://arxiv.org/abs/2602.08877 · **PDF:** https://arxiv.org/pdf/2602.08877v2
+
+## 13945. Is Reasoning Capability Enough for Safety in Long-Context Language Models?
+
+- **ID:** `2602.08874` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Yu Fu, Haz Sameen Shahgir, Huanli Gong, et al.
+- **abs:** https://arxiv.org/abs/2602.08874 · **PDF:** https://arxiv.org/pdf/2602.08874v1
+
+## 13946. Whose Name Comes Up? Benchmarking and Intervention-Based Auditing of LLM-Based Scholar Recommendation
+
+- **ID:** `2602.08873` · **primary:** `cs.IR` · **Submitted:** 2026-02-09
+- **Authors:** Lisette Espin-Noboa, Gonzalo Gabriel Mendez
+- **abs:** https://arxiv.org/abs/2602.08873 · **PDF:** https://arxiv.org/pdf/2602.08873v1
+
+## 13947. Large Language Models for Geolocation Extraction in Humanitarian Crisis Response
+
+- **ID:** `2602.08872` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** G. Cafferata, T. Demarco, K. Kalimeri, et al.
+- **abs:** https://arxiv.org/abs/2602.08872 · **PDF:** https://arxiv.org/pdf/2602.08872v1
+
+## 13948. AnomSeer: Reinforcing Multimodal LLMs to Reason for Time-Series Anomaly Detection
+
+- **ID:** `2602.08868` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Junru Zhang, Lang Feng, Haoran Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.08868 · **PDF:** https://arxiv.org/pdf/2602.08868v1
+
+## 13949. Understanding Dynamic Compute Allocation in Recurrent Transformers
+
+- **ID:** `2602.08864` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Ibraheem Muhammad Moosa, Suhas Lohit, Ye Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.08864 · **PDF:** https://arxiv.org/pdf/2602.08864v1
+
+## 13950. Near-optimal Swap Regret Minimization for Convex Losses
+
+- **ID:** `2602.08862` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Lunjia Hu, Jon Schneider, Yifan Wu
+- **abs:** https://arxiv.org/abs/2602.08862 · **PDF:** https://arxiv.org/pdf/2602.08862v1
+
+## 13951. Magnitude Distance: A Geometric Measure of Dataset Similarity
+
+- **ID:** `2602.08859` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Sahel Torkamani, Henry Gouk, Rik Sarkar
+- **abs:** https://arxiv.org/abs/2602.08859 · **PDF:** https://arxiv.org/pdf/2602.08859v1
+
+## 13952. FlattenGPT: Depth Compression for Transformer with Layer Flattening
+
+- **ID:** `2602.08858` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Ruihan Xu, Qingpei Guo, Yao Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.08858 · **PDF:** https://arxiv.org/pdf/2602.08858v1
+
+## 13953. Discovering Interpretable Algorithms by Decompiling Transformers to RASP
+
+- **ID:** `2602.08857` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Xinting Huang, Aleksandra Bakalova, Satwik Bhattamishra, et al.
+- **abs:** https://arxiv.org/abs/2602.08857 · **PDF:** https://arxiv.org/pdf/2602.08857v1
+
+## 13954. Rethinking Graph Generalization through the Lens of Sharpness-Aware Minimization
+
+- **ID:** `2602.08855` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yang Qiu, Yixiong Zou, Jun Wang
+- **abs:** https://arxiv.org/abs/2602.08855 · **PDF:** https://arxiv.org/pdf/2602.08855v1
+
+## 13955. Cutting Through the Noise: On-the-fly Outlier Detection for Robust Training of Machine Learning Interatomic Potentials
+
+- **ID:** `2602.08849` · **primary:** `stat.ML` · **Submitted:** 2026-02-09
+- **Authors:** Terry C. W. Lam, Niamh O'Neill, Christoph Schran, et al.
+- **abs:** https://arxiv.org/abs/2602.08849 · **PDF:** https://arxiv.org/pdf/2602.08849v1
+
+## 13956. Deciding the Satisfiability of Combined Qualitative Constraint Networks
+
+- **ID:** `2602.08848` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Quentin Cohen-Solal, Alexandre Niveau, Maroua Bouzid
+- **abs:** https://arxiv.org/abs/2602.08848 · **PDF:** https://arxiv.org/pdf/2602.08848v1
+
+## 13957. Dr. MAS: Stable Reinforcement Learning for Multi-Agent LLM Systems
+
+- **ID:** `2602.08847` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Lang Feng, Longtao Zheng, Shuo He, et al.
+- **abs:** https://arxiv.org/abs/2602.08847 · **PDF:** https://arxiv.org/pdf/2602.08847v1
+
+## 13958. Learning to Remember, Learn, and Forget in Attention-Based Models
+
+- **ID:** `2602.09075` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Djohan Bonnet, Jamie Lohoff, Jan Finkbeiner, et al.
+- **abs:** https://arxiv.org/abs/2602.09075 · **PDF:** https://arxiv.org/pdf/2602.09075v2
+
+## 13959. AMEM4Rec: Leveraging Cross-User Similarity for Memory Evolution in Agentic LLM Recommenders
+
+- **ID:** `2602.08837` · **primary:** `cs.IR` · **Submitted:** 2026-02-09
+- **Authors:** Minh-Duc Nguyen, Hai-Dang Kieu, Dung D. Le
+- **abs:** https://arxiv.org/abs/2602.08837 · **PDF:** https://arxiv.org/pdf/2602.08837v1
+
+## 13960. Learning the Value Systems of Societies with Preference-based Multi-objective Reinforcement Learning
+
+- **ID:** `2602.08835` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Andrés Holgado-Sánchez, Peter Vamplew, Richard Dazeley, et al.
+- **abs:** https://arxiv.org/abs/2602.08835 · **PDF:** https://arxiv.org/pdf/2602.08835v2
+
+## 13961. WildReward: Learning Reward Models from In-the-Wild Human Interactions
+
+- **ID:** `2602.08829` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Hao Peng, Yunjia Qi, Xiaozhi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.08829 · **PDF:** https://arxiv.org/pdf/2602.08829v1
+
+## 13962. Affective Flow Language Model for Emotional Support Conversation
+
+- **ID:** `2602.08826` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Chenghui Zou, Ning Wang, Tiesunlong Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.08826 · **PDF:** https://arxiv.org/pdf/2602.08826v1
+
+## 13963. A Methodology for Effective Surrogate Learning in Complex Optimization
+
+- **ID:** `2602.08825` · **primary:** `cs.NE` · **Submitted:** 2026-02-09
+- **Authors:** Tomohiro Harada, Enrique Alba, Gabriel Luque
+- **abs:** https://arxiv.org/abs/2602.08825 · **PDF:** https://arxiv.org/pdf/2602.08825v1
+
+## 13964. On the Use of a Large Language Model to Support the Conduction of a Systematic Mapping Study: A Brief Report from a Practitioner's View
+
+- **ID:** `2602.10147` · **primary:** `cs.SE` · **Submitted:** 2026-02-09
+- **Authors:** Cauã Ferreira Barros, Marcos Kalinowski, Mohamad Kassab, et al.
+- **abs:** https://arxiv.org/abs/2602.10147 · **PDF:** https://arxiv.org/pdf/2602.10147v1
+
+## 13965. Bayesian Preference Learning for Test-Time Steerable Reward Models
+
+- **ID:** `2602.08819` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Jiwoo Hong, Shao Tang, Zhipeng Wang
+- **abs:** https://arxiv.org/abs/2602.08819 · **PDF:** https://arxiv.org/pdf/2602.08819v1
+
+## 13966. FlexMoRE: A Flexible Mixture of Rank-heterogeneous Experts for Efficient Federatedly-trained Large Language Models
+
+- **ID:** `2602.08818` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Annemette Brok Pirchert, Jacob Nielsen, Mogens Henrik From, et al.
+- **abs:** https://arxiv.org/abs/2602.08818 · **PDF:** https://arxiv.org/pdf/2602.08818v1
+
+## 13967. Kirin: Improving ANN efficiency with SNN Hybridization
+
+- **ID:** `2602.08817` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Chenyu Wang, Zhanglu Yan, Zhi Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.08817 · **PDF:** https://arxiv.org/pdf/2602.08817v1
+
+## 13968. Permissive-Washing in the Open AI Supply Chain: A Large-Scale Audit of License Integrity
+
+- **ID:** `2602.08816` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** James Jewitt, Gopi Krishnan Rajbahadur, Hao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.08816 · **PDF:** https://arxiv.org/pdf/2602.08816v1
+
+## 13969. Negative-Aware Diffusion Process for Temporal Knowledge Graph Extrapolation
+
+- **ID:** `2602.08815` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Yanglei Gan, Peng He, Yuxiang Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.08815 · **PDF:** https://arxiv.org/pdf/2602.08815v1
+
+## 13970. Robust Policy Optimization to Prevent Catastrophic Forgetting
+
+- **ID:** `2602.08813` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Mahdi Sabbaghi, George Pappas, Adel Javanmard, et al.
+- **abs:** https://arxiv.org/abs/2602.08813 · **PDF:** https://arxiv.org/pdf/2602.08813v1
+
+## 13971. $\texttt{lrnnx}$: A library for Linear RNNs
+
+- **ID:** `2602.08810` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Karan Bania, Soham Kalburgi, Manit Tanwar, et al.
+- **abs:** https://arxiv.org/abs/2602.08810 · **PDF:** https://arxiv.org/pdf/2602.08810v1
+
+## 13972. Efficient Deep Learning for Biometrics: Overview, Challenges and Trends in Ear of Frugal AI
+
+- **ID:** `2602.08809` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Karim Haroun, Aya Zitouni, Aicha Zenakhri, et al.
+- **abs:** https://arxiv.org/abs/2602.08809 · **PDF:** https://arxiv.org/pdf/2602.08809v1
+
+## 13973. How2Everything: Mining the Web for How-To Procedures to Evaluate and Improve LLMs
+
+- **ID:** `2602.08808` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yapei Chang, Kyle Lo, Mohit Iyyer, et al.
+- **abs:** https://arxiv.org/abs/2602.08808 · **PDF:** https://arxiv.org/pdf/2602.08808v1
+
+## 13974. Root Cause Analysis Method Based on Large Language Models with Residual Connection Structures
+
+- **ID:** `2602.08804` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Liming Zhou, Ailing Liu, Hongwei Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.08804 · **PDF:** https://arxiv.org/pdf/2602.08804v1
+
+## 13975. Addressing data annotation scarcity in Brain Tumor Segmentation on 3D MRI scan Using a Semi-Supervised Teacher-Student Framework
+
+- **ID:** `2602.08797` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Jiaming Liu, Cheng Ding, Daoqiang Zhang
+- **abs:** https://arxiv.org/abs/2602.08797 · **PDF:** https://arxiv.org/pdf/2602.08797v1
+
+## 13976. The Use of AI Tools to Develop and Validate Q-Matrices
+
+- **ID:** `2602.08796` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Kevin Fan, Jacquelyn A. Bialo, Hongli Li
+- **abs:** https://arxiv.org/abs/2602.08796 · **PDF:** https://arxiv.org/pdf/2602.08796v1
+
+## 13977. LakeHopper: Cross Data Lakes Column Type Annotation through Model Adaptation
+
+- **ID:** `2602.08793` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Yushi Sun, Xujia Li, Nan Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.08793 · **PDF:** https://arxiv.org/pdf/2602.08793v1
+
+## 13978. Multimodal Learning for Arcing Detection in Pantograph-Catenary Systems
+
+- **ID:** `2602.08792` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Hao Dong, Eleni Chatzi, Olga Fink
+- **abs:** https://arxiv.org/abs/2602.08792 · **PDF:** https://arxiv.org/pdf/2602.08792v1
+
+## 13979. Empirically Understanding the Value of Prediction in Allocation
+
+- **ID:** `2602.08786` · **primary:** `cs.CY` · **Submitted:** 2026-02-09
+- **Authors:** Unai Fischer-Abaigar, Emily Aiken, Christoph Kern, et al.
+- **abs:** https://arxiv.org/abs/2602.08786 · **PDF:** https://arxiv.org/pdf/2602.08786v3
+
+## 13980. A Graphop Analysis of Graph Neural Networks on Sparse Graphs: Generalization and Universal Approximation
+
+- **ID:** `2602.08785` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Ofek Amran, Tom Gilat, Ron Levie
+- **abs:** https://arxiv.org/abs/2602.08785 · **PDF:** https://arxiv.org/pdf/2602.08785v1
+
+## 13981. Dynamics Within Latent Chain-of-Thought: An Empirical Study of Causal Structure
+
+- **ID:** `2602.08783` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Zirui Li, Xuefeng Bai, Kehai Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.08783 · **PDF:** https://arxiv.org/pdf/2602.08783v2
+
+## 13982. Amortising Inference and Meta-Learning Priors in Neural Networks
+
+- **ID:** `2602.08782` · **primary:** `stat.ML` · **Submitted:** 2026-02-09
+- **Authors:** Tommy Rochussen, Vincent Fortuin
+- **abs:** https://arxiv.org/abs/2602.08782 · **PDF:** https://arxiv.org/pdf/2602.08782v1
+
+## 13983. Default Machine Learning Hyperparameters Do Not Provide Informative Initialization for Bayesian Optimization
+
+- **ID:** `2602.08774` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Nicolás Villagrán Prieto, Eduardo C. Garrido-Merchán
+- **abs:** https://arxiv.org/abs/2602.08774 · **PDF:** https://arxiv.org/pdf/2602.08774v1
+
+## 13984. VERA: Identifying and Leveraging Visual Evidence Retrieval Heads in Long-Context Understanding
+
+- **ID:** `2602.10146` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Rongcan Pei, Huan Li, Fang Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.10146 · **PDF:** https://arxiv.org/pdf/2602.10146v1
+
+## 13985. FreqLens: Interpretable Frequency Attribution for Time Series Forecasting
+
+- **ID:** `2602.08768` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Chi-Sheng Chen, Xinyu Zhang, En-Jui Kuo, et al.
+- **abs:** https://arxiv.org/abs/2602.08768 · **PDF:** https://arxiv.org/pdf/2602.08768v1
+
+## 13986. Taming Scylla: Understanding the multi-headed agentic daemon of the coding seas
+
+- **ID:** `2602.08765` · **primary:** `cs.SE` · **Submitted:** 2026-02-09
+- **Authors:** Micah Villmow
+- **abs:** https://arxiv.org/abs/2602.08765 · **PDF:** https://arxiv.org/pdf/2602.08765v1
+
+## 13987. Efficient Brain Extraction of MRI Scans with Mild to Moderate Neuropathology
+
+- **ID:** `2602.08764` · **primary:** `eess.IV` · **Submitted:** 2026-02-09
+- **Authors:** Hjalti Thrastarson, Lotta M. Ellingsen
+- **abs:** https://arxiv.org/abs/2602.08764 · **PDF:** https://arxiv.org/pdf/2602.08764v1
+
+## 13988. HoGS: Homophily-Oriented Graph Synthesis for Local Differentially Private GNN Training
+
+- **ID:** `2602.08762` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Wen Xu, Zhetao Li, Yong Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.08762 · **PDF:** https://arxiv.org/pdf/2602.08762v1
+
+## 13989. Align and Adapt: Multimodal Multiview Human Activity Recognition under Arbitrary View Combinations
+
+- **ID:** `2602.08755` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Duc-Anh Nguyen, Nhien-An Le-Khac
+- **abs:** https://arxiv.org/abs/2602.08755 · **PDF:** https://arxiv.org/pdf/2602.08755v3
+
+## 13990. Belief Offloading in Human-AI Interaction
+
+- **ID:** `2602.08754` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Rose E. Guingrich, Dvija Mehta, Umang Bhatt
+- **abs:** https://arxiv.org/abs/2602.08754 · **PDF:** https://arxiv.org/pdf/2602.08754v1
+
+## 13991. Central Dogma Transformer II: An AI Microscope for Understanding Cellular Regulatory Mechanisms
+
+- **ID:** `2602.08751` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Nobuyuki Ota
+- **abs:** https://arxiv.org/abs/2602.08751 · **PDF:** https://arxiv.org/pdf/2602.08751v3
+
+## 13992. Silence Routing: When Not Speaking Improves Collective Judgment
+
+- **ID:** `2602.10145` · **primary:** `physics.soc-ph` · **Submitted:** 2026-02-09
+- **Authors:** Itsuki Fujisaki, Kunhao Yang
+- **abs:** https://arxiv.org/abs/2602.10145 · **PDF:** https://arxiv.org/pdf/2602.10145v1
+
+## 13993. On the Expressive Power of GNNs for Boolean Satisfiability
+
+- **ID:** `2602.08745` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Saku Peltonen, Roger Wattenhofer
+- **abs:** https://arxiv.org/abs/2602.08745 · **PDF:** https://arxiv.org/pdf/2602.08745v1
+
+## 13994. Welfarist Formulations for Diverse Similarity Search
+
+- **ID:** `2602.08742` · **primary:** `cs.DS` · **Submitted:** 2026-02-09
+- **Authors:** Siddharth Barman, Nirjhar Das, Shivam Gupta, et al.
+- **abs:** https://arxiv.org/abs/2602.08742 · **PDF:** https://arxiv.org/pdf/2602.08742v1
+
+## 13995. Map of Encoders -- Mapping Sentence Encoders using Quantum Relative Entropy
+
+- **ID:** `2602.08740` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Gaifan Zhang, Danushka Bollegala
+- **abs:** https://arxiv.org/abs/2602.08740 · **PDF:** https://arxiv.org/pdf/2602.08740v1
+
+## 13996. Finite-State Controllers for (Hidden-Model) POMDPs using Deep Reinforcement Learning
+
+- **ID:** `2602.08734` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** David Hudák, Maris F. L. Galesloot, Martin Tappler, et al.
+- **abs:** https://arxiv.org/abs/2602.08734 · **PDF:** https://arxiv.org/pdf/2602.08734v2
+
+## 13997. Foundation Inference Models for Ordinary Differential Equations
+
+- **ID:** `2602.08733` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Maximilian Mauel, Johannes R. Hübers, David Berghaus, et al.
+- **abs:** https://arxiv.org/abs/2602.08733 · **PDF:** https://arxiv.org/pdf/2602.08733v1
+
+## 13998. Artifact Reduction in Undersampled 3D Cone-Beam CTs using a Hybrid 2D-3D CNN Framework
+
+- **ID:** `2602.08727` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Johannes Thalhammer, Tina Dorosti, Sebastian Peterhansl, et al.
+- **abs:** https://arxiv.org/abs/2602.08727 · **PDF:** https://arxiv.org/pdf/2602.08727v1
+
+## 13999. Data Reconstruction: Identifiability and Optimization with Sample Splitting
+
+- **ID:** `2602.08723` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yujie Shen, Zihan Wang, Jian Qian, et al.
+- **abs:** https://arxiv.org/abs/2602.08723 · **PDF:** https://arxiv.org/pdf/2602.08723v1
+
+## 14000. QUOKA: Query-Oriented KV Selection For Efficient LLM Prefill
+
+- **ID:** `2602.08722` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Dalton Jones, Junyoung Park, Matthew Morse, et al.
+- **abs:** https://arxiv.org/abs/2602.08722 · **PDF:** https://arxiv.org/pdf/2602.08722v1
+
+## 14001. Zero-shot System for Automatic Body Region Detection for Volumetric CT and MR Images
+
+- **ID:** `2602.08717` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Farnaz Khun Jush, Grit Werner, Mark Klemens, et al.
+- **abs:** https://arxiv.org/abs/2602.08717 · **PDF:** https://arxiv.org/pdf/2602.08717v1
+
+## 14002. PERSPECTRA: A Scalable and Configurable Pluralist Benchmark of Perspectives from Arguments
+
+- **ID:** `2602.08716` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Shangrui Nie, Kian Omoomi, Lucie Flek, et al.
+- **abs:** https://arxiv.org/abs/2602.08716 · **PDF:** https://arxiv.org/pdf/2602.08716v1
+
+## 14003. Exploring SAIG Methods for an Objective Evaluation of XAI
+
+- **ID:** `2602.08715` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Miquel Miró-Nicolau, Gabriel Moyà-Alcover, Anna Arias-Duart
+- **abs:** https://arxiv.org/abs/2602.08715 · **PDF:** https://arxiv.org/pdf/2602.08715v1
+
+## 14004. FactSim: Fact-Checking for Opinion Summarization
+
+- **ID:** `2602.08709` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Leandro Anghinoni, Jorge Sanchez
+- **abs:** https://arxiv.org/abs/2602.08709 · **PDF:** https://arxiv.org/pdf/2602.08709v1
+
+## 14005. Intermediate Results on the Complexity of STRIPS$_{1}^{1}$
+
+- **ID:** `2602.08708` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Stefan Edelkamp, Jiří Fink, Petr Gregor, et al.
+- **abs:** https://arxiv.org/abs/2602.08708 · **PDF:** https://arxiv.org/pdf/2602.08708v1
+
+## 14006. Why do we Trust Chatbots? From Normative Principles to Behavioral Drivers
+
+- **ID:** `2602.08707` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Aditya Gulati, Nuria Oliver
+- **abs:** https://arxiv.org/abs/2602.08707 · **PDF:** https://arxiv.org/pdf/2602.08707v3
+
+## 14007. Technosocial risks of ideal emotion recognition technologies: A defense of the (social) value of emotional expressions
+
+- **ID:** `2602.08706` · **primary:** `cs.HC` · **Submitted:** 2026-02-09
+- **Authors:** Alexandra Pregent
+- **abs:** https://arxiv.org/abs/2602.08706 · **PDF:** https://arxiv.org/pdf/2602.08706v1
+
+## 14008. Do Images Clarify? A Study on the Effect of Images on Clarifying Questions in Conversational Search
+
+- **ID:** `2602.08700` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Clemencia Siro, Zahra Abbasiantaeb, Yifei Yuan, et al.
+- **abs:** https://arxiv.org/abs/2602.08700 · **PDF:** https://arxiv.org/pdf/2602.08700v1
+
+## 14009. Challenges in Translating Technical Lectures: Insights from the NPTEL
+
+- **ID:** `2602.08698` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Basudha Raje, Sadanand Venkatraman, Nandana TP, et al.
+- **abs:** https://arxiv.org/abs/2602.08698 · **PDF:** https://arxiv.org/pdf/2602.08698v1
+
+## 14010. Prototype-Based Disentanglement for Controllable Dysarthric Speech Synthesis
+
+- **ID:** `2602.08696` · **primary:** `cs.SD` · **Submitted:** 2026-02-09
+- **Authors:** Haoshen Wang, Xueli Zhong, Bingbing Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.08696 · **PDF:** https://arxiv.org/pdf/2602.08696v2
+
+## 14011. Trapped by simplicity: When Transformers fail to learn from noisy features
+
+- **ID:** `2602.08695` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Evan Peters, Ando Deng, Matheus H. Zambianco, et al.
+- **abs:** https://arxiv.org/abs/2602.08695 · **PDF:** https://arxiv.org/pdf/2602.08695v1
+
+## 14012. Reasoning aligns language models to human cognition
+
+- **ID:** `2602.08693` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Gonçalo Guiomar, Elia Torre, Pehuen Moure, et al.
+- **abs:** https://arxiv.org/abs/2602.08693 · **PDF:** https://arxiv.org/pdf/2602.08693v1
+
+## 14013. PBLean: Pseudo-Boolean Proof Certificates for Lean 4
+
+- **ID:** `2602.08692` · **primary:** `cs.LO` · **Submitted:** 2026-02-09
+- **Authors:** Stefan Szeider
+- **abs:** https://arxiv.org/abs/2602.08692 · **PDF:** https://arxiv.org/pdf/2602.08692v2
+
+## 14014. SoK: The Pitfalls of Deep Reinforcement Learning for Cybersecurity
+
+- **ID:** `2602.08690` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Shae McFadden, Myles Foley, Elizabeth Bates, et al.
+- **abs:** https://arxiv.org/abs/2602.08690 · **PDF:** https://arxiv.org/pdf/2602.08690v1
+
+## 14015. Learning To Sample From Diffusion Models Via Inverse Reinforcement Learning
+
+- **ID:** `2602.08689` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Constant Bourdrez, Alexandre Vérine, Olivier Cappé
+- **abs:** https://arxiv.org/abs/2602.08689 · **PDF:** https://arxiv.org/pdf/2602.08689v1
+
+## 14016. Old wine in old glasses: Comparing computational and qualitative methods in identifying incivility on Persian Twitter during the #MahsaAmini movement
+
+- **ID:** `2602.08688` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Hossein Kermani, Fatemeh Oudlajani, Pardis Yarahmadi, et al.
+- **abs:** https://arxiv.org/abs/2602.08688 · **PDF:** https://arxiv.org/pdf/2602.08688v1
+
+## 14017. CompilerKV: Risk-Adaptive KV Compression via Offline Experience Compilation
+
+- **ID:** `2602.08686` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Ning Yang, Chengzhi Wang, Yibo Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.08686 · **PDF:** https://arxiv.org/pdf/2602.08686v1
+
+## 14018. The Theory and Practice of MAP Inference over Non-Convex Constraints
+
+- **ID:** `2602.08681` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Leander Kurscheidt, Gabriele Masina, Roberto Sebastiani, et al.
+- **abs:** https://arxiv.org/abs/2602.08681 · **PDF:** https://arxiv.org/pdf/2602.08681v2
+
+## 14019. Dashed Line Defense: Plug-And-Play Defense Against Adaptive Score-Based Query Attacks
+
+- **ID:** `2602.08679` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yanzhang Fu, Zizheng Guo, Jizhou Luo
+- **abs:** https://arxiv.org/abs/2602.08679 · **PDF:** https://arxiv.org/pdf/2602.08679v1
+
+## 14020. LLaDA2.1: Speeding Up Text Diffusion via Token Editing
+
+- **ID:** `2602.08676` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Tiwei Bie, Maosong Cao, Xiang Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.08676 · **PDF:** https://arxiv.org/pdf/2602.08676v3
+
+## 14021. 6G-Bench: An Open Benchmark for Semantic Communication and Network-Level Reasoning with Foundation Models in AI-Native 6G Networks
+
+- **ID:** `2602.08675` · **primary:** `cs.NI` · **Submitted:** 2026-02-09
+- **Authors:** Mohamed Amine Ferrag, Abderrahmane Lakas, Merouane Debbah
+- **abs:** https://arxiv.org/abs/2602.08675 · **PDF:** https://arxiv.org/pdf/2602.08675v1
+
+## 14022. Learning to Judge: LLMs Designing and Applying Evaluation Rubrics
+
+- **ID:** `2602.08672` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Clemencia Siro, Pourya Aliannejadi, Mohammad Aliannejadi
+- **abs:** https://arxiv.org/abs/2602.08672 · **PDF:** https://arxiv.org/pdf/2602.08672v1
+
+## 14023. Retrieval Pivot Attacks in Hybrid RAG: Measuring and Mitigating Amplified Leakage from Vector Seeds to Graph Expansion
+
+- **ID:** `2602.08668` · **primary:** `cs.CR` · **Submitted:** 2026-02-09
+- **Authors:** Scott Thornton
+- **abs:** https://arxiv.org/abs/2602.08668 · **PDF:** https://arxiv.org/pdf/2602.08668v3
+
+## 14024. Equalized Generative Treatment: Matching f-divergences for Fairness in Generative Models
+
+- **ID:** `2602.08660` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Alexandre Verine, Rafael Pinot, Florian Le Bronnec
+- **abs:** https://arxiv.org/abs/2602.08660 · **PDF:** https://arxiv.org/pdf/2602.08660v1
+
+## 14025. Fundamental Reasoning Paradigms Induce Out-of-Domain Generalization in Language Models
+
+- **ID:** `2602.08658` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Mingzi Cao, Xingwei Tan, Mahmud Elahi Akhter, et al.
+- **abs:** https://arxiv.org/abs/2602.08658 · **PDF:** https://arxiv.org/pdf/2602.08658v2
+
+## 14026. Two-Stage Data Synthesization: A Statistics-Driven Restricted Trade-off between Privacy and Prediction
+
+- **ID:** `2602.08657` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Xiaotong Liu, Shao-Bo Lin, Jun Fan, et al.
+- **abs:** https://arxiv.org/abs/2602.08657 · **PDF:** https://arxiv.org/pdf/2602.08657v1
+
+## 14027. From Robotics to Sepsis Treatment: Offline RL via Geometric Pessimism
+
+- **ID:** `2602.08655` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Sarthak Wanjari
+- **abs:** https://arxiv.org/abs/2602.08655 · **PDF:** https://arxiv.org/pdf/2602.08655v2
+
+## 14028. Projected Gradient Ascent for Efficient Reward-Guided Updates with One-Step Generative Models
+
+- **ID:** `2602.08646` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Jisung Hwang, Minhyuk Sung
+- **abs:** https://arxiv.org/abs/2602.08646 · **PDF:** https://arxiv.org/pdf/2602.08646v1
+
+## 14029. What Is the Alignment Tax?
+
+- **ID:** `2603.00047` · **primary:** `econ.EM` · **Submitted:** 2026-02-09
+- **Authors:** Robin Young
+- **abs:** https://arxiv.org/abs/2603.00047 · **PDF:** https://arxiv.org/pdf/2603.00047v2
+
+## 14030. Equitable Multi-Task Learning for AI-RANs
+
+- **ID:** `2603.08717` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Panayiotis Raptis, Fatih Aslan, George Iosifidis
+- **abs:** https://arxiv.org/abs/2603.08717 · **PDF:** https://arxiv.org/pdf/2603.08717v2
+
+## 14031. LEFT: Learnable Fusion of Tri-view Tokens for Unsupervised Time Series Anomaly Detection
+
+- **ID:** `2602.08638` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Dezheng Wang, Tong Chen, Guansong Pang, et al.
+- **abs:** https://arxiv.org/abs/2602.08638 · **PDF:** https://arxiv.org/pdf/2602.08638v1
+
+## 14032. Towards Self-Robust LLMs: Intrinsic Prompt Noise Resistance via CoIPO
+
+- **ID:** `2603.03314` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Xin Yang, Letian Li, Abudukelimu Wuerkaixi, et al.
+- **abs:** https://arxiv.org/abs/2603.03314 · **PDF:** https://arxiv.org/pdf/2603.03314v1
+
+## 14033. We Should Separate Memorization from Copyright
+
+- **ID:** `2602.08632` · **primary:** `cs.CY` · **Submitted:** 2026-02-09
+- **Authors:** Adi Haviv, Niva Elkin-Koren, Uri Hacohen, et al.
+- **abs:** https://arxiv.org/abs/2602.08632 · **PDF:** https://arxiv.org/pdf/2602.08632v1
+
+## 14034. Debate is efficient with your time
+
+- **ID:** `2602.08630` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Jonah Brown-Cohen, Geoffrey Irving, Simon C. Marshall, et al.
+- **abs:** https://arxiv.org/abs/2602.08630 · **PDF:** https://arxiv.org/pdf/2602.08630v1
+
+## 14035. CauScale: Neural Causal Discovery at Scale
+
+- **ID:** `2602.08629` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Bo Peng, Sirui Chen, Jiaguo Tian, et al.
+- **abs:** https://arxiv.org/abs/2602.08629 · **PDF:** https://arxiv.org/pdf/2602.08629v1
+
+## 14036. Do Multilingual LLMs have specialized language heads?
+
+- **ID:** `2602.08625` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Muhammad Naufil
+- **abs:** https://arxiv.org/abs/2602.08625 · **PDF:** https://arxiv.org/pdf/2602.08625v1
+
+## 14037. How does fine-tuning improve sensorimotor representations in large language models?
+
+- **ID:** `2603.03313` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Minghua Wu, Javier Conde, Pedro Reviriego, et al.
+- **abs:** https://arxiv.org/abs/2603.03313 · **PDF:** https://arxiv.org/pdf/2603.03313v1
+
+## 14038. Sparse Models, Sparse Safety: Unsafe Routes in Mixture-of-Experts LLMs
+
+- **ID:** `2602.08621` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yukun Jiang, Hai Huang, Mingjie Li, et al.
+- **abs:** https://arxiv.org/abs/2602.08621 · **PDF:** https://arxiv.org/pdf/2602.08621v1
+
+## 14039. Enhancing Genetic Algorithms with Graph Neural Networks: A Timetabling Case Study
+
+- **ID:** `2602.08619` · **primary:** `cs.NE` · **Submitted:** 2026-02-09
+- **Authors:** Laura-Maria Cornei, Mihaela-Elena Breabăn
+- **abs:** https://arxiv.org/abs/2602.08619 · **PDF:** https://arxiv.org/pdf/2602.08619v1
+
+## 14040. ERIS: Enhancing Privacy and Communication Efficiency in Serverless Federated Learning
+
+- **ID:** `2602.08617` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Dario Fenoglio, Pasquale Polverino, Jacopo Quizi, et al.
+- **abs:** https://arxiv.org/abs/2602.08617 · **PDF:** https://arxiv.org/pdf/2602.08617v1
+
+## 14041. Breaking the Grid: Distance-Guided Reinforcement Learning in Large Discrete and Hybrid Action Spaces
+
+- **ID:** `2602.08616` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Heiko Hoppe, Fabian Akkerman, Wouter van Heeswijk, et al.
+- **abs:** https://arxiv.org/abs/2602.08616 · **PDF:** https://arxiv.org/pdf/2602.08616v1
+
+## 14042. VocalNet-MDM: Accelerating Streaming Speech LLM via Self-Distilled Masked Diffusion Modeling
+
+- **ID:** `2602.08607` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Ziyang Cheng, Yuhao Wang, Heyang Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.08607 · **PDF:** https://arxiv.org/pdf/2602.08607v1
+
+## 14043. Constructive conditional normalizing flows
+
+- **ID:** `2602.08606` · **primary:** `math.OC` · **Submitted:** 2026-02-09
+- **Authors:** Borjan Geshkovski, Domènec Ruiz-Balet
+- **abs:** https://arxiv.org/abs/2602.08606 · **PDF:** https://arxiv.org/pdf/2602.08606v1
+
+## 14044. OSCAR: Optimization-Steered Agentic Planning for Composed Image Retrieval
+
+- **ID:** `2602.08603` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Teng Wang, Rong Shan, Jianghao Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.08603 · **PDF:** https://arxiv.org/pdf/2602.08603v1
+
+## 14045. Beyond Scalar Scores: Reinforcement Learning for Error-Aware Quality Estimation of Machine Translation
+
+- **ID:** `2602.08600` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Archchana Sindhujan, Girish A. Koushik, Shenbin Qian, et al.
+- **abs:** https://arxiv.org/abs/2602.08600 · **PDF:** https://arxiv.org/pdf/2602.08600v1
+
+## 14046. An Attention Mechanism for Robust Multimodal Integration in a Global Workspace Architecture
+
+- **ID:** `2602.08597` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Roland Bertin-Johannet, Lara Scipio, Leopold Maytié, et al.
+- **abs:** https://arxiv.org/abs/2602.08597 · **PDF:** https://arxiv.org/pdf/2602.08597v2
+
+## 14047. Kissan-Dost: Bridging the Last Mile in Smallholder Precision Agriculture with Conversational IoT
+
+- **ID:** `2602.08593` · **primary:** `cs.HC` · **Submitted:** 2026-02-09
+- **Authors:** Muhammad Saad Ali, Daanish U. Khan, Laiba Intizar Ahmad, et al.
+- **abs:** https://arxiv.org/abs/2602.08593 · **PDF:** https://arxiv.org/pdf/2602.08593v1
+
+## 14048. TFMLinker: Universal Link Predictor by Graph In-Context Learning with Tabular Foundation Models
+
+- **ID:** `2602.08592` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Tianyin Liao, Chunyu Hu, Yicheng Sui, et al.
+- **abs:** https://arxiv.org/abs/2602.08592 · **PDF:** https://arxiv.org/pdf/2602.08592v1
+
+## 14049. SDFed: Bridging Local Global Discrepancy via Subspace Refinement and Divergence Control in Federated Prompt Learning
+
+- **ID:** `2602.08590` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yicheng Di, Wei Yuan, Tieke He, et al.
+- **abs:** https://arxiv.org/abs/2602.08590 · **PDF:** https://arxiv.org/pdf/2602.08590v3
+
+## 14050. BioLM-Score: Language-Prior Conditioned Probabilistic Geometric Potentials for Protein-Ligand Scoring
+
+- **ID:** `2602.18476` · **primary:** `q-bio.BM` · **Submitted:** 2026-02-09
+- **Authors:** Zhangfan Yang, Baoyun Chen, Dong Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.18476 · **PDF:** https://arxiv.org/pdf/2602.18476v1
+
+## 14051. FairRARI: A Plug and Play Framework for Fairness-Aware PageRank
+
+- **ID:** `2602.08589` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Emmanouil Kariotakis, Aritra Konar
+- **abs:** https://arxiv.org/abs/2602.08589 · **PDF:** https://arxiv.org/pdf/2602.08589v1
+
+## 14052. PRISM: A Principled Framework for Multi-Agent Reasoning via Gain Decomposition
+
+- **ID:** `2602.08586` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Yiming Yang, Zhuoyuan Li, Fanxiang Zeng, et al.
+- **abs:** https://arxiv.org/abs/2602.08586 · **PDF:** https://arxiv.org/pdf/2602.08586v2
+
+## 14053. Predicting Future Utility: Global Combinatorial Optimization for Task-Agnostic KV Cache Eviction
+
+- **ID:** `2602.08585` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Ziyao Tang, Pengkun Jiao, Xinhang Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.08585 · **PDF:** https://arxiv.org/pdf/2602.08585v1
+
+## 14054. Conditional Sequence Modeling for Safe Reinforcement Learning
+
+- **ID:** `2602.08584` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Wensong Bai, Chao Zhang, Qihang Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.08584 · **PDF:** https://arxiv.org/pdf/2602.08584v1
+
+## 14055. Modeling Score Approximation Errors in Diffusion Models via Forward SPDEs
+
+- **ID:** `2602.08579` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Junsu Seo
+- **abs:** https://arxiv.org/abs/2602.08579 · **PDF:** https://arxiv.org/pdf/2602.08579v1
+
+## 14056. An arithmetic method algorithm optimizing k-nearest neighbors compared to regression algorithms and evaluated on real world data sources
+
+- **ID:** `2602.08577` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Theodoros Anagnostopoulos, Evanthia Zervoudi, Christos Anagnostopoulos, et al.
+- **abs:** https://arxiv.org/abs/2602.08577 · **PDF:** https://arxiv.org/pdf/2602.08577v1
+
+## 14057. ValueFlow: Measuring the Propagation of Value Perturbations in Multi-Agent LLM Systems
+
+- **ID:** `2602.08567` · **primary:** `cs.MA` · **Submitted:** 2026-02-09
+- **Authors:** Jinnuo Liu, Chuke Liu, Hua Shen
+- **abs:** https://arxiv.org/abs/2602.08567 · **PDF:** https://arxiv.org/pdf/2602.08567v1
+
+## 14058. Agent-Supported Foresight for AI Systemic Risks: AI Agents for Breadth, Experts for Judgment
+
+- **ID:** `2602.08565` · **primary:** `cs.HC` · **Submitted:** 2026-02-09
+- **Authors:** Leon Fröhling, Alessandro Giaconia, Edyta Paulina Bogucka, et al.
+- **abs:** https://arxiv.org/abs/2602.08565 · **PDF:** https://arxiv.org/pdf/2602.08565v1
+
+## 14059. M-Loss: Quantifying Model Merging Compatibility with Limited Unlabeled Data
+
+- **ID:** `2602.08564` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Tiantong Wang, Yiyang Duan, Haoyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.08564 · **PDF:** https://arxiv.org/pdf/2602.08564v1
+
+## 14060. Stateless Yet Not Forgetful: Implicit Memory as a Hidden Channel in LLMs
+
+- **ID:** `2602.08563` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Ahmed Salem, Andrew Paverd, Sahar Abdelnabi
+- **abs:** https://arxiv.org/abs/2602.08563 · **PDF:** https://arxiv.org/pdf/2602.08563v1
+
+## 14061. Automating Computational Reproducibility in Social Science: Comparing Prompt-Based and Agent-Based Approaches
+
+- **ID:** `2602.08561` · **primary:** `cs.SE` · **Submitted:** 2026-02-09
+- **Authors:** Syed Mehtab Hussain Shah, Frank Hopfgartner, Arnim Bleier
+- **abs:** https://arxiv.org/abs/2602.08561 · **PDF:** https://arxiv.org/pdf/2602.08561v2
+
+## 14062. DNS: Data-driven Nonlinear Smoother for Complex Model-free Process
+
+- **ID:** `2602.08560` · **primary:** `eess.SP` · **Submitted:** 2026-02-09
+- **Authors:** Fredrik Cumlin, Anubhab Ghosh, Saikat Chatterjee
+- **abs:** https://arxiv.org/abs/2602.08560 · **PDF:** https://arxiv.org/pdf/2602.08560v1
+
+## 14063. Rho-Perfect: Correlation Ceiling For Subjective Evaluation Datasets
+
+- **ID:** `2602.08552` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Fredrik Cumlin
+- **abs:** https://arxiv.org/abs/2602.08552 · **PDF:** https://arxiv.org/pdf/2602.08552v1
+
+## 14064. GOT-Edit: Geometry-Aware Generic Object Tracking via Online Model Editing
+
+- **ID:** `2602.08550` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Shih-Fang Chen, Jun-Cheng Chen, I-Hong Jhuo, et al.
+- **abs:** https://arxiv.org/abs/2602.08550 · **PDF:** https://arxiv.org/pdf/2602.08550v3
+
+## 14065. How Do Language Models Understand Tables? A Mechanistic Analysis of Cell Location
+
+- **ID:** `2602.08548` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Xuanliang Zhang, Dingzirui Wang, Keyan Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.08548 · **PDF:** https://arxiv.org/pdf/2602.08548v1
+
+## 14066. GISA: A Benchmark for General Information-Seeking Assistant
+
+- **ID:** `2602.08543` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Yutao Zhu, Xingshuo Zhang, Maosen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.08543 · **PDF:** https://arxiv.org/pdf/2602.08543v2
+
+## 14067. Incremental (k, z)-Clustering on Graphs
+
+- **ID:** `2602.08542` · **primary:** `cs.DS` · **Submitted:** 2026-02-09
+- **Authors:** Emilio Cruciani, Sebastian Forster, Antonis Skarlatos
+- **abs:** https://arxiv.org/abs/2602.08542 · **PDF:** https://arxiv.org/pdf/2602.08542v2
+
+## 14068. Trajectory Stitching for Solving Inverse Problems with Flow-Based Models
+
+- **ID:** `2602.08538` · **primary:** `eess.IV` · **Submitted:** 2026-02-09
+- **Authors:** Alexander Denker, Moshe Eliasof, Zeljko Kereta, et al.
+- **abs:** https://arxiv.org/abs/2602.08538 · **PDF:** https://arxiv.org/pdf/2602.08538v1
+
+## 14069. Causal Schrödinger Bridges: Constrained Optimal Transport on Structural Manifolds
+
+- **ID:** `2602.08535` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Rui Wu, Li YongJun
+- **abs:** https://arxiv.org/abs/2602.08535 · **PDF:** https://arxiv.org/pdf/2602.08535v4
+
+## 14070. Dialogue Model Optimization via Agent Game and Adaptive Tree-based GRPO
+
+- **ID:** `2602.08533` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Kun Peng, Conghui Tan, Yu Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.08533 · **PDF:** https://arxiv.org/pdf/2602.08533v2
+
+## 14071. Federated Inference: Toward Privacy-Preserving Collaborative and Incentivized Model Serving
+
+- **ID:** `2603.02214` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Jungwon Seo, Ferhat Ozgur Catak, Chunming Rong, et al.
+- **abs:** https://arxiv.org/abs/2603.02214 · **PDF:** https://arxiv.org/pdf/2603.02214v2
+
+## 14072. KidMesh: Computational Mesh Reconstruction for Pediatric Congenital Hydronephrosis Using Deep Neural Networks
+
+- **ID:** `2602.13299` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Haoran Sun, Zhanpeng Zhu, Anguo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.13299 · **PDF:** https://arxiv.org/pdf/2602.13299v1
+
+## 14073. Reinforcement Inference: Leveraging Uncertainty for Self-Correcting Language Model Reasoning
+
+- **ID:** `2602.08520` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Xinhai Sun
+- **abs:** https://arxiv.org/abs/2602.08520 · **PDF:** https://arxiv.org/pdf/2602.08520v3
+
+## 14074. Bridging Academia and Industry: A Comprehensive Benchmark for Attributed Graph Clustering
+
+- **ID:** `2602.08519` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yunhui Liu, Pengyu Qiu, Yu Xing, et al.
+- **abs:** https://arxiv.org/abs/2602.08519 · **PDF:** https://arxiv.org/pdf/2602.08519v1
+
+## 14075. TreeTensor: Boost AI System on Nested Data with Constrained Tree-Like Tensor
+
+- **ID:** `2602.08517` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Shaoang Zhang, Yazhe Niu
+- **abs:** https://arxiv.org/abs/2602.08517 · **PDF:** https://arxiv.org/pdf/2602.08517v2
+
+## 14076. Do physics-informed neural networks (PINNs) need to be deep? Shallow PINNs using the Levenberg-Marquardt algorithm
+
+- **ID:** `2602.08515` · **primary:** `math.NA` · **Submitted:** 2026-02-09
+- **Authors:** Muhammad Luthfi Shahab, Imam Mukhlash, Hadi Susanto
+- **abs:** https://arxiv.org/abs/2602.08515 · **PDF:** https://arxiv.org/pdf/2602.08515v2
+
+## 14077. A Multi-objective Evolutionary Algorithm Based on Bi-population with Uniform Sampling for Neural Architecture Search
+
+- **ID:** `2602.08513` · **primary:** `cs.NE` · **Submitted:** 2026-02-09
+- **Authors:** Yu Xue, Pengcheng Jiang, Chenchen Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.08513 · **PDF:** https://arxiv.org/pdf/2602.08513v1
+
+## 14078. Estimation of Fish Catch Using Sentinel-2, 3 and XGBoost-Kernel-Based Kernel Ridge Regression
+
+- **ID:** `2602.08511` · **primary:** `physics.app-ph` · **Submitted:** 2026-02-09
+- **Authors:** Kanu Mohammed, Vaishnavi Joshi, Pranjali Diliprao Patil, et al.
+- **abs:** https://arxiv.org/abs/2602.08511 · **PDF:** https://arxiv.org/pdf/2602.08511v1
+
+## 14079. A General Theory of Proportionality with Additive Utilities
+
+- **ID:** `2602.08504` · **primary:** `cs.GT` · **Submitted:** 2026-02-09
+- **Authors:** Piotr Skowron
+- **abs:** https://arxiv.org/abs/2602.08504 · **PDF:** https://arxiv.org/pdf/2602.08504v1
+
+## 14080. Learning Self-Correction in Vision-Language Models via Rollout Augmentation
+
+- **ID:** `2602.08503` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Yi Ding, Ziliang Qiu, Bolian Li, et al.
+- **abs:** https://arxiv.org/abs/2602.08503 · **PDF:** https://arxiv.org/pdf/2602.08503v1
+
+## 14081. Is Meta-Path Attention an Explanation? Evidence of Alignment and Decoupling in Heterogeneous GNNs
+
+- **ID:** `2602.08500` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Maiqi Jiang, Noman Ali, Yiran Ding, et al.
+- **abs:** https://arxiv.org/abs/2602.08500 · **PDF:** https://arxiv.org/pdf/2602.08500v1
+
+## 14082. Contextual Rollout Bandits for Reinforcement Learning with Verifiable Rewards
+
+- **ID:** `2602.08499` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Xiaodong Lu, Xiaohan Wang, Jiajun Chai, et al.
+- **abs:** https://arxiv.org/abs/2602.08499 · **PDF:** https://arxiv.org/pdf/2602.08499v1
+
+## 14083. Characterizing, Evaluating, and Optimizing Complex Reasoning
+
+- **ID:** `2602.08498` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Haoran Zhang, Yafu Li, Zhi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.08498 · **PDF:** https://arxiv.org/pdf/2602.08498v1
+
+## 14084. When LLMs get significantly worse: A statistical approach to detect model degradations
+
+- **ID:** `2602.10144` · **primary:** `stat.ML` · **Submitted:** 2026-02-09
+- **Authors:** Jonas Kübler, Kailash Budhathoki, Matthäus Kleindessner, et al.
+- **abs:** https://arxiv.org/abs/2602.10144 · **PDF:** https://arxiv.org/pdf/2602.10144v1
+
+## 14085. Understanding Image2Video Domain Shift in Food Segmentation: An Instance-level Analysis on Apples
+
+- **ID:** `2602.08491` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Keonvin Park, Aditya Pal, Jin Hong Mok
+- **abs:** https://arxiv.org/abs/2602.08491 · **PDF:** https://arxiv.org/pdf/2602.08491v2
+
+## 14086. Beyond Correctness: Learning Robust Reasoning via Transfer
+
+- **ID:** `2602.08489` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Hyunseok Lee, Soheil Abbasloo, Jihoon Tack, et al.
+- **abs:** https://arxiv.org/abs/2602.08489 · **PDF:** https://arxiv.org/pdf/2602.08489v1
+
+## 14087. Empirical Study of Observable Sets in Multiclass Quantum Classification
+
+- **ID:** `2602.08485` · **primary:** `quant-ph` · **Submitted:** 2026-02-09
+- **Authors:** Paul San Sebastian, Mikel Cañizo, Roman Orus
+- **abs:** https://arxiv.org/abs/2602.08485 · **PDF:** https://arxiv.org/pdf/2602.08485v1
+
+## 14088. CLEAR: A Knowledge-Centric Vessel Trajectory Analysis Platform
+
+- **ID:** `2602.08482` · **primary:** `cs.DB` · **Submitted:** 2026-02-09
+- **Authors:** Hengyu Liu, Tianyi Li, Haoyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.08482 · **PDF:** https://arxiv.org/pdf/2602.08482v2
+
+## 14089. Gesture Matters: Pedestrian Gesture Recognition for AVs Through Skeleton Pose Evaluation
+
+- **ID:** `2602.08479` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Alif Rizqullah Mahdi, Mahdi Rezaei, Natasha Merat
+- **abs:** https://arxiv.org/abs/2602.08479 · **PDF:** https://arxiv.org/pdf/2602.08479v1
+
+## 14090. DRAGON: Robust Classification for Very Large Collections of Software Repositories
+
+- **ID:** `2602.09071` · **primary:** `cs.SE` · **Submitted:** 2026-02-09
+- **Authors:** Stefano Balla, Stefano Zacchiroli, Thomas Degueule, et al.
+- **abs:** https://arxiv.org/abs/2602.09071 · **PDF:** https://arxiv.org/pdf/2602.09071v1
+
+## 14091. Time-Delayed Transformers for Data-Driven Modeling of Low-Dimensional Dynamics
+
+- **ID:** `2602.08478` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Albert Alcalde, Markus Widhalm, Emre Yılmaz
+- **abs:** https://arxiv.org/abs/2602.08478 · **PDF:** https://arxiv.org/pdf/2602.08478v1
+
+## 14092. Learning Credal Ensembles via Distributionally Robust Optimization
+
+- **ID:** `2602.08470` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Kaizheng Wang, Ghifari Adam Faza, Fabio Cuzzolin, et al.
+- **abs:** https://arxiv.org/abs/2602.08470 · **PDF:** https://arxiv.org/pdf/2602.08470v2
+
+## 14093. REMIND: Rethinking Medical High-Modality Learning under Missingness--A Long-Tailed Distribution Perspective
+
+- **ID:** `2603.00046` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Chenwei Wu, Zitao Shuai, Liyue Shen
+- **abs:** https://arxiv.org/abs/2603.00046 · **PDF:** https://arxiv.org/pdf/2603.00046v1
+
+## 14094. Low Rank Transformer for Multivariate Time Series Anomaly Detection and Localization
+
+- **ID:** `2602.08467` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Charalampos Shimillas, Kleanthis Malialis, Konstantinos Fokianos, et al.
+- **abs:** https://arxiv.org/abs/2602.08467 · **PDF:** https://arxiv.org/pdf/2602.08467v1
+
+## 14095. The Effective Depth Paradox: Evaluating the Relationship between Architectural Topology and Trainability in Deep CNNs
+
+- **ID:** `2602.13298` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Manfred M. Fischer, Joshua Pitts
+- **abs:** https://arxiv.org/abs/2602.13298 · **PDF:** https://arxiv.org/pdf/2602.13298v2
+
+## 14096. Estimating Aleatoric Uncertainty in the Causal Treatment Effect
+
+- **ID:** `2602.08461` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Liyuan Xu, Bijan Mazaheri
+- **abs:** https://arxiv.org/abs/2602.08461 · **PDF:** https://arxiv.org/pdf/2602.08461v1
+
+## 14097. Decentralized Spatial Reuse Optimization in Wi-Fi: An Internal Regret Minimization Approach
+
+- **ID:** `2602.08456` · **primary:** `cs.NI` · **Submitted:** 2026-02-09
+- **Authors:** Francesc Wilhelmi, Boris Bellalta, Miguel Casasnovas, et al.
+- **abs:** https://arxiv.org/abs/2602.08456 · **PDF:** https://arxiv.org/pdf/2602.08456v1
+
+## 14098. When Evaluation Becomes a Side Channel: Regime Leakage and Structural Mitigations for Alignment Assessment
+
+- **ID:** `2602.08449` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Igor Santos-Grueiro
+- **abs:** https://arxiv.org/abs/2602.08449 · **PDF:** https://arxiv.org/pdf/2602.08449v3
+
+## 14099. Vista: Scene-Aware Optimization for Streaming Video Question Answering under Post-Hoc Queries
+
+- **ID:** `2602.08448` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Haocheng Lu, Nan Zhang, Wei Tao, et al.
+- **abs:** https://arxiv.org/abs/2602.08448 · **PDF:** https://arxiv.org/pdf/2602.08448v1
+
+## 14100. RIFLE: Robust Distillation-based FL for Deep Model Deployment on Resource-Constrained IoT Networks
+
+- **ID:** `2602.08446` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Pouria Arefijamal, Mahdi Ahmadlou, Bardia Safaei, et al.
+- **abs:** https://arxiv.org/abs/2602.08446 · **PDF:** https://arxiv.org/pdf/2602.08446v1
+
+## 14101. Large Language Models and Impossible Language Acquisition: "False Promise" or an Overturn of our Current Perspective towards AI
+
+- **ID:** `2602.08437` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Ziyan Wang, Longlong Ma
+- **abs:** https://arxiv.org/abs/2602.08437 · **PDF:** https://arxiv.org/pdf/2602.08437v4
+
+## 14102. Conditional Generative Models for High-Resolution Range Profiles: Capturing Geometry-Driven Trends in a Large-Scale Maritime Dataset
+
+- **ID:** `2602.13297` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Edwyn Brient, Santiago Velasco-Forero, Rami Kassab
+- **abs:** https://arxiv.org/abs/2602.13297 · **PDF:** https://arxiv.org/pdf/2602.13297v1
+
+## 14103. MFN Decomposition and Related Metrics for High-Resolution Range Profiles Generative Models
+
+- **ID:** `2602.13296` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Edwyn Brient, Santiago Velasco-Forero, Rami Kassab
+- **abs:** https://arxiv.org/abs/2602.13296 · **PDF:** https://arxiv.org/pdf/2602.13296v1
+
+## 14104. NarraScore: Bridging Visual Narrative and Musical Dynamics via Hierarchical Affective Control
+
+- **ID:** `2602.09070` · **primary:** `cs.SD` · **Submitted:** 2026-02-09
+- **Authors:** Yufan Wen, Zhaocheng Liu, YeGuo Hua, et al.
+- **abs:** https://arxiv.org/abs/2602.09070 · **PDF:** https://arxiv.org/pdf/2602.09070v2
+
+## 14105. USBD: Universal Structural Basis Distillation for Source-Free Graph Domain Adaptation
+
+- **ID:** `2602.08431` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yingxu Wang, Kunyu Zhang, Mengzhu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.08431 · **PDF:** https://arxiv.org/pdf/2602.08431v1
+
+## 14106. Benchmarking Early Deterioration Prediction Across Hospital-Rich and MCI-Like Emergency Triage Under Constrained Sensing
+
+- **ID:** `2602.20168` · **primary:** `cs.CY` · **Submitted:** 2026-02-09
+- **Authors:** KMA Solaiman, Joshua Sebastian, Karma Tobden
+- **abs:** https://arxiv.org/abs/2602.20168 · **PDF:** https://arxiv.org/pdf/2602.20168v2
+
+## 14107. The Connection between Kriging and Large Neural Networks
+
+- **ID:** `2602.08427` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Marius Marinescu
+- **abs:** https://arxiv.org/abs/2602.08427 · **PDF:** https://arxiv.org/pdf/2602.08427v1
+
+## 14108. Prism: Spectral-Aware Block-Sparse Attention
+
+- **ID:** `2602.08426` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Xinghao Wang, Pengyu Wang, Xiaoran Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.08426 · **PDF:** https://arxiv.org/pdf/2602.08426v1
+
+## 14109. LLMs + Security = Trouble
+
+- **ID:** `2602.08422` · **primary:** `cs.CR` · **Submitted:** 2026-02-09
+- **Authors:** Benjamin Livshits
+- **abs:** https://arxiv.org/abs/2602.08422 · **PDF:** https://arxiv.org/pdf/2602.08422v1
+
+## 14110. Radial Müntz-Szász Networks: Neural Architectures with Learnable Power Bases for Multidimensional Singularities
+
+- **ID:** `2602.08419` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Gnankan Landry Regis N'guessan, Bum Jun Kim
+- **abs:** https://arxiv.org/abs/2602.08419 · **PDF:** https://arxiv.org/pdf/2602.08419v2
+
+## 14111. Mathematical Foundations of Modeling ETL Process Chains
+
+- **ID:** `2603.29877` · **primary:** `cs.DC` · **Submitted:** 2026-02-09
+- **Authors:** Levin Maier, Lucas Schulze, Robert Lilow, et al.
+- **abs:** https://arxiv.org/abs/2603.29877 · **PDF:** https://arxiv.org/pdf/2603.29877v1
+
+## 14112. From Assistant to Double Agent: Formalizing and Benchmarking Attacks on OpenClaw for Personalized Local AI Agent
+
+- **ID:** `2602.08412` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Yuhang Wang, Feiming Xu, Zheng Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.08412 · **PDF:** https://arxiv.org/pdf/2602.08412v2
+
+## 14113. Drop the mask! GAMM-A Taxonomy for Graph Attributes Missing Mechanisms
+
+- **ID:** `2602.08407` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Richard Serrano, Baptiste Jeudy, Charlotte Laclau, et al.
+- **abs:** https://arxiv.org/abs/2602.08407 · **PDF:** https://arxiv.org/pdf/2602.08407v1
+
+## 14114. Optimizing Spectral Prediction in MXene-Based Metasurfaces Through Multi-Channel Spectral Refinement and Savitzky-Golay Smoothing
+
+- **ID:** `2602.08406` · **primary:** `physics.optics` · **Submitted:** 2026-02-09
+- **Authors:** Shujaat Khan, Waleed Iqbal Waseer
+- **abs:** https://arxiv.org/abs/2602.08406 · **PDF:** https://arxiv.org/pdf/2602.08406v1
+
+## 14115. TEAM: Temporal-Spatial Consistency Guided Expert Activation for MoE Diffusion Language Model Acceleration
+
+- **ID:** `2602.08404` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Linye Wei, Zixiang Luo, Pingzhi Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.08404 · **PDF:** https://arxiv.org/pdf/2602.08404v1
+
+## 14116. Intelligent support for Human Oversight: Integrating Reinforcement Learning with Gaze Simulation to Personalize Highlighting
+
+- **ID:** `2602.08403` · **primary:** `cs.HC` · **Submitted:** 2026-02-09
+- **Authors:** Thorsten Klößner, João Belo, Zekun Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.08403 · **PDF:** https://arxiv.org/pdf/2602.08403v1
+
+## 14117. On Protecting Agentic Systems' Intellectual Property via Watermarking
+
+- **ID:** `2602.08401` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Liwen Wang, Zongjie Li, Yuchong Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.08401 · **PDF:** https://arxiv.org/pdf/2602.08401v1
+
+## 14118. SCOUT-RAG: Scalable and Cost-Efficient Unifying Traversal for Agentic Graph-RAG over Distributed Domains
+
+- **ID:** `2602.08400` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Longkun Li, Yuanben Zou, Jinghan Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.08400 · **PDF:** https://arxiv.org/pdf/2602.08400v1
+
+## 14119. AntigenLM: Structure-Aware DNA Language Modeling for Influenza
+
+- **ID:** `2602.09067` · **primary:** `q-bio.GN` · **Submitted:** 2026-02-09
+- **Authors:** Yue Pei, Xuebin Chi, Yu Kang
+- **abs:** https://arxiv.org/abs/2602.09067 · **PDF:** https://arxiv.org/pdf/2602.09067v1
+
+## 14120. BiManiBench: A Hierarchical Benchmark for Evaluating Bimanual Coordination of Multimodal Large Language Models
+
+- **ID:** `2602.08392` · **primary:** `cs.RO` · **Submitted:** 2026-02-09
+- **Authors:** Xin Wu, Zhixuan Liang, Yue Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.08392 · **PDF:** https://arxiv.org/pdf/2602.08392v1
+
+## 14121. Altruism and Fair Objective in Mixed-Motive Markov games
+
+- **ID:** `2602.08389` · **primary:** `cs.MA` · **Submitted:** 2026-02-09
+- **Authors:** Yao-hua Franck Xu, Tayeb Lemlouma, Arnaud Braud, et al.
+- **abs:** https://arxiv.org/abs/2602.08389 · **PDF:** https://arxiv.org/pdf/2602.08389v1
+
+## 14122. Modalities, a PyTorch-native Framework For Large-scale LLM Training and Research
+
+- **ID:** `2602.08387` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Max Lübbering, Timm Ruland, Richard Rutmann, et al.
+- **abs:** https://arxiv.org/abs/2602.08387 · **PDF:** https://arxiv.org/pdf/2602.08387v1
+
+## 14123. Breaking the Factorization Barrier in Diffusion Language Models
+
+- **ID:** `2603.00045` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Ian Li, Zilei Shao, Benjie Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.00045 · **PDF:** https://arxiv.org/pdf/2603.00045v2
+
+## 14124. Dynamic Long Context Reasoning over Compressed Memory via End-to-End Reinforcement Learning
+
+- **ID:** `2602.08382` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Zhuoen Chen, Dongfang Li, Meishan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.08382 · **PDF:** https://arxiv.org/pdf/2602.08382v1
+
+## 14125. Reinforcement Learning with Backtracking Feedback
+
+- **ID:** `2602.08377` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Bilgehan Sel, Vaishakh Keshava, Phillip Wallis, et al.
+- **abs:** https://arxiv.org/abs/2602.08377 · **PDF:** https://arxiv.org/pdf/2602.08377v1
+
+## 14126. OJBKQ: Objective-Joint Babai-Klein Quantization
+
+- **ID:** `2602.08376` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Xinyu Wang, Ziyu Zhao, Peng Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.08376 · **PDF:** https://arxiv.org/pdf/2602.08376v1
+
+## 14127. Schrödinger bridge problem via empirical risk minimization
+
+- **ID:** `2602.08374` · **primary:** `stat.ML` · **Submitted:** 2026-02-09
+- **Authors:** Denis Belomestny, Alexey Naumov, Nikita Puchkin, et al.
+- **abs:** https://arxiv.org/abs/2602.08374 · **PDF:** https://arxiv.org/pdf/2602.08374v1
+
+## 14128. Grounding Generative Planners in Verifiable Logic: A Hybrid Architecture for Trustworthy Embodied AI
+
+- **ID:** `2602.08373` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Feiyu Wu, Xu Zheng, Yue Qu, et al.
+- **abs:** https://arxiv.org/abs/2602.08373 · **PDF:** https://arxiv.org/pdf/2602.08373v1
+
+## 14129. Dynamic Regret via Discounted-to-Dynamic Reduction with Applications to Curved Losses and Adam Optimizer
+
+- **ID:** `2602.08372` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yan-Feng Xie, Yu-Jie Zhang, Peng Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.08372 · **PDF:** https://arxiv.org/pdf/2602.08372v1
+
+## 14130. ViGoEmotions: A Benchmark Dataset For Fine-grained Emotion Detection on Vietnamese Texts
+
+- **ID:** `2602.08371` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Hung Quang Tran, Nam Tien Pham, Son T. Luu, et al.
+- **abs:** https://arxiv.org/abs/2602.08371 · **PDF:** https://arxiv.org/pdf/2602.08371v2
+
+## 14131. Learning Human-Like Badminton Skills for Humanoid Robots
+
+- **ID:** `2602.08370` · **primary:** `cs.RO` · **Submitted:** 2026-02-09
+- **Authors:** Yeke Chen, Shihao Dong, Xiaoyu Ji, et al.
+- **abs:** https://arxiv.org/abs/2602.08370 · **PDF:** https://arxiv.org/pdf/2602.08370v1
+
+## 14132. MemAdapter: Fast Alignment across Agent Memory Paradigms via Generative Subgraph Retrieval
+
+- **ID:** `2602.08369` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Xin Zhang, Kailai Yang, Chenyue Li, et al.
+- **abs:** https://arxiv.org/abs/2602.08369 · **PDF:** https://arxiv.org/pdf/2602.08369v1
+
+## 14133. WorldTravel: A Realistic Multimodal Travel-Planning Benchmark with Tightly Coupled Constraints
+
+- **ID:** `2602.08367` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Zexuan Wang, Chenghao Yang, Yingqi Que, et al.
+- **abs:** https://arxiv.org/abs/2602.08367 · **PDF:** https://arxiv.org/pdf/2602.08367v1
+
+## 14134. Roadmap to Quantum Aesthetics
+
+- **ID:** `2602.08363` · **primary:** `physics.pop-ph` · **Submitted:** 2026-02-09
+- **Authors:** Ivan C. H. Liu, Hsiao-Yuan Chen
+- **abs:** https://arxiv.org/abs/2602.08363 · **PDF:** https://arxiv.org/pdf/2602.08363v1
+
+## 14135. Circuit Representations of Random Forests with Applications to XAI
+
+- **ID:** `2602.08362` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Chunxi Ji, Adnan Darwiche
+- **abs:** https://arxiv.org/abs/2602.08362 · **PDF:** https://arxiv.org/pdf/2602.08362v2
+
+## 14136. Does Your Reasoning Model Implicitly Know When to Stop Thinking?
+
+- **ID:** `2602.08354` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Zixuan Huang, Xin Xia, Yuxi Ren, et al.
+- **abs:** https://arxiv.org/abs/2602.08354 · **PDF:** https://arxiv.org/pdf/2602.08354v3
+
+## 14137. Towards Better Evolution Modeling for Temporal Knowledge Graphs
+
+- **ID:** `2602.08353` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Zhang Jiasheng, Li Zhangpin, Wang Mingzhe, et al.
+- **abs:** https://arxiv.org/abs/2602.08353 · **PDF:** https://arxiv.org/pdf/2602.08353v1
+
+## 14138. The Chicken and Egg Dilemma: Co-optimizing Data and Model Configurations for LLMs
+
+- **ID:** `2602.08351` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Zhiliang Chen, Alfred Wei Lun Leong, Shao Yong Ong, et al.
+- **abs:** https://arxiv.org/abs/2602.08351 · **PDF:** https://arxiv.org/pdf/2602.08351v2
+
+## 14139. All ERMs Can Fail in Stochastic Convex Optimization Lower Bounds in Linear Dimension
+
+- **ID:** `2602.08350` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Tal Burla, Roi Livni
+- **abs:** https://arxiv.org/abs/2602.08350 · **PDF:** https://arxiv.org/pdf/2602.08350v1
+
+## 14140. Spectral Disentanglement and Enhancement: A Dual-domain Contrastive Framework for Representation Learning
+
+- **ID:** `2602.09066` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Jinjin Guo, Yexin Li, Zhichao Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.09066 · **PDF:** https://arxiv.org/pdf/2602.09066v1
+
+## 14141. OPE: Overcoming Information Saturation in Parallel Thinking via Outline-Guided Path Exploration
+
+- **ID:** `2602.08344` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Qi Guo, Jianing Wang, Deyang Kong, et al.
+- **abs:** https://arxiv.org/abs/2602.08344 · **PDF:** https://arxiv.org/pdf/2602.08344v1
+
+## 14142. ManifoldKV: Training-Free KV Cache Compression via Euclidean Outlier Detection
+
+- **ID:** `2602.08343` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Debajyoti Datta, Trishala Neeraj, Bibek Paudel, et al.
+- **abs:** https://arxiv.org/abs/2602.08343 · **PDF:** https://arxiv.org/pdf/2602.08343v1
+
+## 14143. UrbanGraphEmbeddings: Learning and Evaluating Spatially Grounded Multimodal Embeddings for Urban Science
+
+- **ID:** `2602.08342` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Jie Zhang, Xingtong Yu, Yuan Fang, et al.
+- **abs:** https://arxiv.org/abs/2602.08342 · **PDF:** https://arxiv.org/pdf/2602.08342v1
+
+## 14144. Effect-Level Validation for Causal Discovery
+
+- **ID:** `2602.08340` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Hoang Dang, Luan Pham, Minh Nguyen
+- **abs:** https://arxiv.org/abs/2602.08340 · **PDF:** https://arxiv.org/pdf/2602.08340v1
+
+## 14145. CoTZero: Annotation-Free Human-Like Vision Reasoning via Hierarchical Synthetic CoT
+
+- **ID:** `2602.08339` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Chengyi Du, Yazhe Niu, Dazhong Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.08339 · **PDF:** https://arxiv.org/pdf/2602.08339v1
+
+## 14146. Enhanced Graph Transformer with Serialized Graph Tokens
+
+- **ID:** `2602.09065` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Ruixiang Wang, Yuyang Hong, Shiming Xiang, et al.
+- **abs:** https://arxiv.org/abs/2602.09065 · **PDF:** https://arxiv.org/pdf/2602.09065v1
+
+## 14147. Property-Driven Evaluation of GNN Expressiveness at Scale: Datasets, Framework, and Study
+
+- **ID:** `2603.00044` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Sicong Che, Jiayi Yang, Sarfraz Khurshid, et al.
+- **abs:** https://arxiv.org/abs/2603.00044 · **PDF:** https://arxiv.org/pdf/2603.00044v1
+
+## 14148. UReason: Benchmarking the Reasoning Paradox in Unified Multimodal Models
+
+- **ID:** `2602.08336` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Cheng Yang, Chufan Shi, Bo Shui, et al.
+- **abs:** https://arxiv.org/abs/2602.08336 · **PDF:** https://arxiv.org/pdf/2602.08336v1
+
+## 14149. Who Deserves the Reward? SHARP: Shapley Credit-based Optimization for Multi-Agent System
+
+- **ID:** `2602.08335` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Yanming Li, Xuelin Zhang, WenJie Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.08335 · **PDF:** https://arxiv.org/pdf/2602.08335v1
+
+## 14150. Regime Change Hypothesis: Foundations for Decoupled Dynamics in Neural Network Training
+
+- **ID:** `2602.08333` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Cristian Pérez-Corral, Alberto Fernández-Hernández, Jose I. Mestre, et al.
+- **abs:** https://arxiv.org/abs/2602.08333 · **PDF:** https://arxiv.org/pdf/2602.08333v1
+
+## 14151. Latent Reasoning with Supervised Thinking States
+
+- **ID:** `2602.08332` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Ido Amos, Avi Caciularu, Mor Geva, et al.
+- **abs:** https://arxiv.org/abs/2602.08332 · **PDF:** https://arxiv.org/pdf/2602.08332v1
+
+## 14152. PACC: Protocol-Aware Cross-Layer Compression for Compact Network Traffic Representation
+
+- **ID:** `2602.08331` · **primary:** `cs.NI` · **Submitted:** 2026-02-09
+- **Authors:** Zhaochen Guo, Tianyufei Zhou, Honghao Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.08331 · **PDF:** https://arxiv.org/pdf/2602.08331v1
+
+## 14153. Near-Oracle KV Selection via Pre-hoc Sparsity for Long-Context Inference
+
+- **ID:** `2602.08329` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yifei Gao, Lei Wang, Rong-Cheng Tu, et al.
+- **abs:** https://arxiv.org/abs/2602.08329 · **PDF:** https://arxiv.org/pdf/2602.08329v1
+
+## 14154. Towards Efficient Large Language Reasoning Models via Extreme-Ratio Chain-of-Thought Compression
+
+- **ID:** `2602.08324` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yuntian Tang, Bohan Jia, Wenxuan Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.08324 · **PDF:** https://arxiv.org/pdf/2602.08324v2
+
+## 14155. A Generative Model for Joint Multiple Intent Detection and Slot Filling
+
+- **ID:** `2602.08322` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Liz Li, Wei Zhu
+- **abs:** https://arxiv.org/abs/2602.08322 · **PDF:** https://arxiv.org/pdf/2602.08322v2
+
+## 14156. Improving Data and Reward Design for Scientific Reasoning in Large Language Models
+
+- **ID:** `2602.08321` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Zijie Chen, Zhenghao Lin, Xiao Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.08321 · **PDF:** https://arxiv.org/pdf/2602.08321v2
+
+## 14157. Is Flow Matching Just Trajectory Replay for Sequential Data?
+
+- **ID:** `2602.08318` · **primary:** `stat.ML` · **Submitted:** 2026-02-09
+- **Authors:** Soon Hoe Lim, Shizheng Lin, Michael W. Mahoney, et al.
+- **abs:** https://arxiv.org/abs/2602.08318 · **PDF:** https://arxiv.org/pdf/2602.08318v1
+
+## 14158. SWE Context Bench: A Benchmark for Context Learning in Coding
+
+- **ID:** `2602.08316` · **primary:** `cs.SE` · **Submitted:** 2026-02-09
+- **Authors:** Jared Zhu, Minhao Hu, Junde Wu
+- **abs:** https://arxiv.org/abs/2602.08316 · **PDF:** https://arxiv.org/pdf/2602.08316v2
+
+## 14159. Fast Flow Matching based Conditional Independence Tests for Causal Discovery
+
+- **ID:** `2602.08315` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Shunyu Zhao, Yanfeng Yang, Shuai Li, et al.
+- **abs:** https://arxiv.org/abs/2602.08315 · **PDF:** https://arxiv.org/pdf/2602.08315v1
+
+## 14160. Moral Sycophancy in Vision Language Models
+
+- **ID:** `2602.08311` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Shadman Rabby, Md. Hefzul Hossain Papon, Sabbir Ahmed, et al.
+- **abs:** https://arxiv.org/abs/2602.08311 · **PDF:** https://arxiv.org/pdf/2602.08311v1
+
+## 14161. Interaction-Grounded Learning for Contextual Markov Decision Processes with Personalized Feedback
+
+- **ID:** `2602.08307` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Mengxiao Zhang, Yuheng Zhang, Haipeng Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.08307 · **PDF:** https://arxiv.org/pdf/2602.08307v1
+
+## 14162. TextResNet: Decoupling and Routing Optimization Signals in Compound AI Systems via Deep Residual Tuning
+
+- **ID:** `2602.08306` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Suizhi Huang, Mei Li, Han Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.08306 · **PDF:** https://arxiv.org/pdf/2602.08306v1
+
+## 14163. JUSTICE: Judicial Unified Synthesis Through Intermediate Conclusion Emulation for Automated Judgment Document Generation
+
+- **ID:** `2602.08305` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Binglin Wu, Yingyi Zhang, Xiannneg Li
+- **abs:** https://arxiv.org/abs/2602.08305 · **PDF:** https://arxiv.org/pdf/2602.08305v1
+
+## 14164. Reinforcement Learning for Control with Probabilistic Stability Guarantee: A Finite-Sample Approach
+
+- **ID:** `2603.00043` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Minghao Han, Lixian Zhang, Chenliang Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.00043 · **PDF:** https://arxiv.org/pdf/2603.00043v1
+
+## 14165. Grokking in Linear Models for Logistic Regression
+
+- **ID:** `2602.08302` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Nataraj Das, Atreya Vedantam, Chandrashekar Lakshminarayanan
+- **abs:** https://arxiv.org/abs/2602.08302 · **PDF:** https://arxiv.org/pdf/2602.08302v1
+
+## 14166. Automatic Generation of Polynomial Symmetry Breaking Constraints
+
+- **ID:** `2602.08297` · **primary:** `cs.SC` · **Submitted:** 2026-02-09
+- **Authors:** Madalina Erascu, Johannes Middeke
+- **abs:** https://arxiv.org/abs/2602.08297 · **PDF:** https://arxiv.org/pdf/2602.08297v1
+
+## 14167. The Vibe-Automation of Automation: A Proactive Education Framework for Computer Science in the Age of Generative AI
+
+- **ID:** `2602.08295` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Ilya Levin
+- **abs:** https://arxiv.org/abs/2602.08295 · **PDF:** https://arxiv.org/pdf/2602.08295v1
+
+## 14168. When Does Context Help? Error Dynamics of Contextual Information in Large Language Models
+
+- **ID:** `2602.08294` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Dingzirui Wang, Xuanliang Zhang, Keyan Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.08294 · **PDF:** https://arxiv.org/pdf/2602.08294v1
+
+## 14169. Trust-Based Incentive Mechanisms in Semi-Decentralized Federated Learning Systems
+
+- **ID:** `2602.08290` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Ajay Kumar Shrestha
+- **abs:** https://arxiv.org/abs/2602.08290 · **PDF:** https://arxiv.org/pdf/2602.08290v1
+
+## 14170. VisPhyWorld: Probing Physical Reasoning via Code-Driven Video Reconstruction
+
+- **ID:** `2602.13294` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Jiarong Liang, Max Ku, Ka-Hei Hui, et al.
+- **abs:** https://arxiv.org/abs/2602.13294 · **PDF:** https://arxiv.org/pdf/2602.13294v2
+
+## 14171. Knowledge Augmented Entity and Relation Extraction for Legal Documents with Hypergraph Neural Network
+
+- **ID:** `2602.08289` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Binglin Wu, Xianneng Li
+- **abs:** https://arxiv.org/abs/2602.08289 · **PDF:** https://arxiv.org/pdf/2602.08289v1
+
+## 14172. Maximizing the Spectral Energy Gain in Sub-1-Bit LLMs via Latent Geometry Alignment
+
+- **ID:** `2603.00042` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Banseok Lee, Youngmin Kim
+- **abs:** https://arxiv.org/abs/2603.00042 · **PDF:** https://arxiv.org/pdf/2603.00042v1
+
+## 14173. Predicting Open Source Software Sustainability with Deep Temporal Neural Hierarchical Architectures and Explainable AI
+
+- **ID:** `2602.09064` · **primary:** `cs.SE` · **Submitted:** 2026-02-09
+- **Authors:** S M Rakib Ul Karim, Wenyi Lu, Enock Kasaadha, et al.
+- **abs:** https://arxiv.org/abs/2602.09064 · **PDF:** https://arxiv.org/pdf/2602.09064v2
+
+## 14174. Perfecting Human-AI Interaction at Clinical Scale. Turning Production Signals into Safer, More Human Conversations
+
+- **ID:** `2603.29893` · **primary:** `cs.HC` · **Submitted:** 2026-02-09
+- **Authors:** Subhabrata Mukherjee, Markel Sanz Ausin, Kriti Aggarwal, et al.
+- **abs:** https://arxiv.org/abs/2603.29893 · **PDF:** https://arxiv.org/pdf/2603.29893v1
+
+## 14175. Noise Stability of Transformer Models
+
+- **ID:** `2602.08287` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Themistoklis Haris, Zihan Zhang, Yuichi Yoshida
+- **abs:** https://arxiv.org/abs/2602.08287 · **PDF:** https://arxiv.org/pdf/2602.08287v1
+
+## 14176. Tighnari v2: Mitigating Label Noise and Distribution Shift in Multimodal Plant Distribution Prediction via Mixture of Experts and Weakly Supervised Learning
+
+- **ID:** `2602.08282` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Haixu Liu, Yufei Wang, Tianxiang Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.08282 · **PDF:** https://arxiv.org/pdf/2602.08282v1
+
+## 14177. New Skills or Sharper Primitives? A Probabilistic Perspective on the Emergence of Reasoning in RLVR
+
+- **ID:** `2602.08281` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Zhilin Wang, Yafu Li, Shunkai Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.08281 · **PDF:** https://arxiv.org/pdf/2602.08281v1
+
+## 14178. Econometric vs. Causal Structure-Learning for Time-Series Policy Decisions: Evidence from the UK COVID-19 Policies
+
+- **ID:** `2603.00041` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Bruno Petrungaro, Anthony C. Constantinou
+- **abs:** https://arxiv.org/abs/2603.00041 · **PDF:** https://arxiv.org/pdf/2603.00041v1
+
+## 14179. PISCO: Precise Video Instance Insertion with Sparse Control
+
+- **ID:** `2602.08277` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Xiangbo Gao, Renjie Li, Xinghao Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.08277 · **PDF:** https://arxiv.org/pdf/2602.08277v2
+
+## 14180. Toward Formalizing LLM-Based Agent Designs through Structural Context Modeling and Semantic Dynamics Analysis
+
+- **ID:** `2602.08276` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Haoyu Jia, Kento Kawaharazuka, Kei Okada
+- **abs:** https://arxiv.org/abs/2602.08276 · **PDF:** https://arxiv.org/pdf/2602.08276v1
+
+## 14181. Linguistics and Human Brain: A Perspective of Computational Neuroscience
+
+- **ID:** `2602.08275` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-09
+- **Authors:** Fudong Zhang, Bo Chai, Yujie Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.08275 · **PDF:** https://arxiv.org/pdf/2602.08275v2
+
+## 14182. Language Modeling and Understanding Through Paraphrase Generation and Detection
+
+- **ID:** `2602.08274` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Jan Philip Wahle
+- **abs:** https://arxiv.org/abs/2602.08274 · **PDF:** https://arxiv.org/pdf/2602.08274v3
+
+## 14183. When Do Multi-Agent Systems Outperform? Analysing the Learning Efficiency of Agentic Systems
+
+- **ID:** `2602.08272` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Junwei Su, Chuan Wu
+- **abs:** https://arxiv.org/abs/2602.08272 · **PDF:** https://arxiv.org/pdf/2602.08272v1
+
+## 14184. Puda: Private User Dataset Agent for User-Sovereign and Privacy-Preserving Personalized AI
+
+- **ID:** `2602.08268` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Akinori Maeda, Yuto Sekiya, Sota Sugimura, et al.
+- **abs:** https://arxiv.org/abs/2602.08268 · **PDF:** https://arxiv.org/pdf/2602.08268v2
+
+## 14185. Inverting Data Transformations via Diffusion Sampling
+
+- **ID:** `2602.08267` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Jinwoo Kim, Sékou-Oumar Kaba, Jiyun Park, et al.
+- **abs:** https://arxiv.org/abs/2602.08267 · **PDF:** https://arxiv.org/pdf/2602.08267v1
+
+## 14186. Attn-QAT: 4-Bit Attention With Quantization-Aware Training
+
+- **ID:** `2603.00040` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Peiyuan Zhang, Matthew Noto, Wenxuan Tan, et al.
+- **abs:** https://arxiv.org/abs/2603.00040 · **PDF:** https://arxiv.org/pdf/2603.00040v2
+
+## 14187. Constraint-Aware Generative Auto-bidding via Pareto-Prioritized Regret Optimization
+
+- **ID:** `2602.08261` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Binglin Wu, Yingyi Zhang, Xianneng Li, et al.
+- **abs:** https://arxiv.org/abs/2602.08261 · **PDF:** https://arxiv.org/pdf/2602.08261v1
+
+## 14188. Decentralized Attention Fails Centralized Signals: Rethinking Transformers for Medical Time Series
+
+- **ID:** `2602.18473` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Guoqi Yu, Juncheng Wang, Chen Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.18473 · **PDF:** https://arxiv.org/pdf/2602.18473v1
+
+## 14189. A Statistical Framework for Alignment with Biased AI Feedback
+
+- **ID:** `2602.08259` · **primary:** `stat.ML` · **Submitted:** 2026-02-09
+- **Authors:** Xintao Xia, Zhiqiu Xia, Linjun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.08259 · **PDF:** https://arxiv.org/pdf/2602.08259v1
+
+## 14190. SynthAgent: A Multi-Agent LLM Framework for Realistic Patient Simulation -- A Case Study in Obesity with Mental Health Comorbidities
+
+- **ID:** `2602.08254` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Arman Aghaee, Sepehr Asgarian, Jouhyun Jeon
+- **abs:** https://arxiv.org/abs/2602.08254 · **PDF:** https://arxiv.org/pdf/2602.08254v1
+
+## 14191. G-LNS: Generative Large Neighborhood Search for LLM-Based Automatic Heuristic Design
+
+- **ID:** `2602.08253` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Baoyun Zhao, He Wang, Liang Zeng
+- **abs:** https://arxiv.org/abs/2602.08253 · **PDF:** https://arxiv.org/pdf/2602.08253v1
+
+## 14192. Language Predicts Identity Fusion Across Cultures and Reveals Divergent Pathways to Violence
+
+- **ID:** `2602.08252` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Devin R. Wright, Justin E. Lane, F. LeRon Shults
+- **abs:** https://arxiv.org/abs/2602.08252 · **PDF:** https://arxiv.org/pdf/2602.08252v1
+
+## 14193. STEP: Warm-Started Visuomotor Policies with Spatiotemporal Consistency Prediction
+
+- **ID:** `2602.08245` · **primary:** `cs.RO` · **Submitted:** 2026-02-09
+- **Authors:** Jinhao Li, Yuxuan Cong, Yingqiao Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.08245 · **PDF:** https://arxiv.org/pdf/2602.08245v1
+
+## 14194. Learning in Context, Guided by Choice: A Reward-Free Paradigm for Reinforcement Learning with Transformers
+
+- **ID:** `2602.08244` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Juncheng Dong, Bowen He, Moyang Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.08244 · **PDF:** https://arxiv.org/pdf/2602.08244v1
+
+## 14195. Discrete Adjoint Schrödinger Bridge Sampler
+
+- **ID:** `2602.08243` · **primary:** `stat.ML` · **Submitted:** 2026-02-09
+- **Authors:** Wei Guo, Yuchen Zhu, Xiaochen Du, et al.
+- **abs:** https://arxiv.org/abs/2602.08243 · **PDF:** https://arxiv.org/pdf/2602.08243v1
+
+## 14196. Mirror: A Multi-Agent System for AI-Assisted Ethics Review
+
+- **ID:** `2602.13292` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Yifan Ding, Yuhui Shi, Zhiyan Li, et al.
+- **abs:** https://arxiv.org/abs/2602.13292 · **PDF:** https://arxiv.org/pdf/2602.13292v1
+
+## 14197. Do MLLMs Really See It: Reinforcing Visual Attention in Multimodal LLMs
+
+- **ID:** `2602.08241` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Siqu Ou, Tianrui Wan, Zhiyuan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.08241 · **PDF:** https://arxiv.org/pdf/2602.08241v1
+
+## 14198. PTS-SNN: A Prompt-Tuned Temporal Shift Spiking Neural Networks for Efficient Speech Emotion Recognition
+
+- **ID:** `2602.08240` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Xun Su, Huamin Wang, Qi Zhang
+- **abs:** https://arxiv.org/abs/2602.08240 · **PDF:** https://arxiv.org/pdf/2602.08240v1
+
+## 14199. Linearization Explains Fine-Tuning in Large Language Models
+
+- **ID:** `2602.08239` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Zahra Rahimi Afzal, Tara Esmaeilbeig, Mojtaba Soltanalian, et al.
+- **abs:** https://arxiv.org/abs/2602.08239 · **PDF:** https://arxiv.org/pdf/2602.08239v1
+
+## 14200. On convexity and efficiency in semantic systems
+
+- **ID:** `2602.08238` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Nathaniel Imel, Noga Zaslavasky
+- **abs:** https://arxiv.org/abs/2602.08238 · **PDF:** https://arxiv.org/pdf/2602.08238v1
+
+## 14201. Document Reconstruction Unlocks Scalable Long-Context RLVR
+
+- **ID:** `2602.08237` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Yao Xiao, Lei Wang, Yue Deng, et al.
+- **abs:** https://arxiv.org/abs/2602.08237 · **PDF:** https://arxiv.org/pdf/2602.08237v4
+
+## 14202. When and How Much to Imagine: Adaptive Test-Time Scaling with World Models for Visual Spatial Reasoning
+
+- **ID:** `2602.08236` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Shoubin Yu, Yue Zhang, Zun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.08236 · **PDF:** https://arxiv.org/pdf/2602.08236v1
+
+## 14203. scBench: Evaluating AI Agents on Single-Cell RNA-seq Analysis
+
+- **ID:** `2602.09063` · **primary:** `q-bio.GN` · **Submitted:** 2026-02-09
+- **Authors:** Kenny Workman, Zhen Yang, Harihara Muralidharan, et al.
+- **abs:** https://arxiv.org/abs/2602.09063 · **PDF:** https://arxiv.org/pdf/2602.09063v1
+
+## 14204. When Benign Inputs Lead to Severe Harms: Eliciting Unsafe Unintended Behaviors of Computer-Use Agents
+
+- **ID:** `2602.08235` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Jaylen Jones, Zhehao Zhang, Yuting Ning, et al.
+- **abs:** https://arxiv.org/abs/2602.08235 · **PDF:** https://arxiv.org/pdf/2602.08235v1
+
+## 14205. CARE: Confounder-Aware Aggregation for Reliable LLM Evaluation
+
+- **ID:** `2603.00039` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Jitian Zhao, Changho Shin, Tzu-Heng Huang, et al.
+- **abs:** https://arxiv.org/abs/2603.00039 · **PDF:** https://arxiv.org/pdf/2603.00039v1
+
+## 14206. SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning
+
+- **ID:** `2602.08234` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Peng Xia, Jianwen Chen, Hanyang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.08234 · **PDF:** https://arxiv.org/pdf/2602.08234v1
+
+## 14207. Tutti: Expressive Multi-Singer Synthesis via Structure-Level Timbre Control and Vocal Texture Modeling
+
+- **ID:** `2602.08233` · **primary:** `cs.SD` · **Submitted:** 2026-02-09
+- **Authors:** Jiatao Chen, Xing Tang, Xiaoyue Duan, et al.
+- **abs:** https://arxiv.org/abs/2602.08233 · **PDF:** https://arxiv.org/pdf/2602.08233v1
+
+## 14208. Adaptive Matrix Online Learning through Smoothing with Guarantees for Nonsmooth Nonconvex Optimization
+
+- **ID:** `2602.08232` · **primary:** `math.OC` · **Submitted:** 2026-02-09
+- **Authors:** Ruichen Jiang, Zakaria Mhammedi, Mehryar Mohri, et al.
+- **abs:** https://arxiv.org/abs/2602.08232 · **PDF:** https://arxiv.org/pdf/2602.08232v1
+
+## 14209. Generating Adversarial Events: A Motion-Aware Point Cloud Framework
+
+- **ID:** `2602.08230` · **primary:** `cs.CV` · **Submitted:** 2026-02-09
+- **Authors:** Hongwei Ren, Youxin Jiang, Qifei Gu, et al.
+- **abs:** https://arxiv.org/abs/2602.08230 · **PDF:** https://arxiv.org/pdf/2602.08230v1
+
+## 14210. InfiCoEvalChain: A Blockchain-Based Decentralized Framework for Collaborative LLM Evaluation
+
+- **ID:** `2602.08229` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Yifan Yang, Jinjia Li, Kunxi Li, et al.
+- **abs:** https://arxiv.org/abs/2602.08229 · **PDF:** https://arxiv.org/pdf/2602.08229v1
+
+## 14211. Investigating Writing Professionals' Relationships with Generative AI: How Combined Perceptions of Rivalry and Collaboration Shape Work Practices and Outcomes
+
+- **ID:** `2602.08227` · **primary:** `cs.HC` · **Submitted:** 2026-02-09
+- **Authors:** Rama Adithya Varanasi, Oded Nov, Batia Mishan Wiesenfeld
+- **abs:** https://arxiv.org/abs/2602.08227 · **PDF:** https://arxiv.org/pdf/2602.08227v2
+
+## 14212. Weak-Driven Learning: How Weak Agents make Strong Agents Stronger
+
+- **ID:** `2602.08222` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Zehao Chen, Gongxun Li, Tianxiang Ai, et al.
+- **abs:** https://arxiv.org/abs/2602.08222 · **PDF:** https://arxiv.org/pdf/2602.08222v1
+
+## 14213. CoRect: Context-Aware Logit Contrast for Hidden State Rectification to Resolve Knowledge Conflicts
+
+- **ID:** `2602.08221` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Xuhua Ma, Richong Zhang, Zhijie Nie
+- **abs:** https://arxiv.org/abs/2602.08221 · **PDF:** https://arxiv.org/pdf/2602.08221v1
+
+## 14214. Pretraining with Token-Level Adaptive Latent Chain-of-Thought
+
+- **ID:** `2602.08220` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Boyi Zeng, Yiqin Hao, He Li, et al.
+- **abs:** https://arxiv.org/abs/2602.08220 · **PDF:** https://arxiv.org/pdf/2602.08220v2
+
+## 14215. Escaping the BLEU Trap: A Signal-Grounded Framework with Decoupled Semantic Guidance for EEG-to-Text Decoding
+
+- **ID:** `2603.03312` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Yuchen Wang, Haonan Wang, Yu Guo, et al.
+- **abs:** https://arxiv.org/abs/2603.03312 · **PDF:** https://arxiv.org/pdf/2603.03312v1
+
+## 14216. Sparsity-Aware Evolution for Model Merging
+
+- **ID:** `2602.08218` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Huan Zhang, Yanjian Zhang, Guillaume Wisniewski, et al.
+- **abs:** https://arxiv.org/abs/2602.08218 · **PDF:** https://arxiv.org/pdf/2602.08218v1
+
+## 14217. Thermodynamic Isomorphism of Transformers: A Lagrangian Approach to Attention Dynamics
+
+- **ID:** `2602.08216` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Gunn Kim
+- **abs:** https://arxiv.org/abs/2602.08216 · **PDF:** https://arxiv.org/pdf/2602.08216v2
+
+## 14218. Distribution-Free Robust Predict-Then-Optimize in Function Spaces
+
+- **ID:** `2602.08215` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Yash Patel, Ambuj Tewari
+- **abs:** https://arxiv.org/abs/2602.08215 · **PDF:** https://arxiv.org/pdf/2602.08215v2
+
+## 14219. RECUR: Resource Exhaustion Attack via Recursive-Entropy Guided Counterfactual Utilization and Reflection
+
+- **ID:** `2602.08214` · **primary:** `cs.AI` · **Submitted:** 2026-02-09
+- **Authors:** Ziwei Wang, Yuanhe Zhang, Jing Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.08214 · **PDF:** https://arxiv.org/pdf/2602.08214v1
+
+## 14220. DrugR: Optimizing Molecular Drugs through LLM-based Explicit Reasoning
+
+- **ID:** `2602.08213` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Haoran Liu, Zheni Zeng, Yukun Yan, et al.
+- **abs:** https://arxiv.org/abs/2602.08213 · **PDF:** https://arxiv.org/pdf/2602.08213v1
+
+## 14221. ConceptRM: The Quest to Mitigate Alert Fatigue through Consensus-Based Purity-Driven Data Cleaning for Reflection Modelling
+
+- **ID:** `2602.20166` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Yongda Yu, Lei Zhang, Xinxin Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.20166 · **PDF:** https://arxiv.org/pdf/2602.20166v1
+
+## 14222. CADO: From Imitation to Cost Minimization for Heatmap-based Solvers in Combinatorial Optimization
+
+- **ID:** `2602.08210` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Hyungseok Song, Deunsol Yoon, Kanghoon Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.08210 · **PDF:** https://arxiv.org/pdf/2602.08210v1
+
+## 14223. LLMs and people both learn to form conventions -- just not with each other
+
+- **ID:** `2602.08208` · **primary:** `cs.CL` · **Submitted:** 2026-02-09
+- **Authors:** Cameron R. Jones, Agnese Lombardi, Kyle Mahowald, et al.
+- **abs:** https://arxiv.org/abs/2602.08208 · **PDF:** https://arxiv.org/pdf/2602.08208v1
+
+## 14224. Interpretable Dynamic Network Modeling of Tensor Time Series via Kronecker Time-Varying Graphical Lasso
+
+- **ID:** `2602.08197` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Shingo Higashiguchi, Koki Kawabata, Yasuko Matsubara, et al.
+- **abs:** https://arxiv.org/abs/2602.08197 · **PDF:** https://arxiv.org/pdf/2602.08197v1
+
+## 14225. Dreaming in Code for Curriculum Learning in Open-Ended Worlds
+
+- **ID:** `2602.08194` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Konstantinos Mitsides, Maxence Faldor, Antoine Cully
+- **abs:** https://arxiv.org/abs/2602.08194 · **PDF:** https://arxiv.org/pdf/2602.08194v1
+
+## 14226. Large Language Models in Peer-Run Community Behavioral Health Services: Understanding Peer Specialists and Service Users' Perspectives on Opportunities, Risks, and Mitigation Strategies
+
+- **ID:** `2602.08187` · **primary:** `cs.HC` · **Submitted:** 2026-02-09
+- **Authors:** Cindy Peng, Megan Chai, Gao Mo, et al.
+- **abs:** https://arxiv.org/abs/2602.08187 · **PDF:** https://arxiv.org/pdf/2602.08187v1
+
+## 14227. Nexus: Inferring Join Graphs from Metadata Alone via Iterative Low-Rank Matrix Completion
+
+- **ID:** `2602.08186` · **primary:** `cs.DB` · **Submitted:** 2026-02-09
+- **Authors:** Tianji Cong, Yuanyuan Tian, Andreas Mueller, et al.
+- **abs:** https://arxiv.org/abs/2602.08186 · **PDF:** https://arxiv.org/pdf/2602.08186v1
+
+## 14228. Information Geometry of Absorbing Markov-Chain and Discriminative Random Walks
+
+- **ID:** `2602.08185` · **primary:** `stat.ML` · **Submitted:** 2026-02-09
+- **Authors:** Masanari Kimura
+- **abs:** https://arxiv.org/abs/2602.08185 · **PDF:** https://arxiv.org/pdf/2602.08185v1
+
+## 14229. Nansde-net: A neural sde framework for generating time series with memory
+
+- **ID:** `2602.08182` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Hiromu Ozai, Kei Nakagawa
+- **abs:** https://arxiv.org/abs/2602.08182 · **PDF:** https://arxiv.org/pdf/2602.08182v1
+
+## 14230. Learning to Drive in New Cities Without Human Demonstrations
+
+- **ID:** `2602.15891` · **primary:** `cs.RO` · **Submitted:** 2026-02-09
+- **Authors:** Zilin Wang, Saeed Rahmani, Daphne Cornelisse, et al.
+- **abs:** https://arxiv.org/abs/2602.15891 · **PDF:** https://arxiv.org/pdf/2602.15891v1
+
+## 14231. Agent Mars: Multi-Agent Simulation for Multi-Planetary Life Exploration and Settlement
+
+- **ID:** `2602.13291` · **primary:** `cs.MA` · **Submitted:** 2026-02-09
+- **Authors:** Ziyang Wang
+- **abs:** https://arxiv.org/abs/2602.13291 · **PDF:** https://arxiv.org/pdf/2602.13291v1
+
+## 14232. Fundamental Limits of Community Detection in Contextual Multi-Layer Stochastic Block Models
+
+- **ID:** `2602.08173` · **primary:** `math.ST` · **Submitted:** 2026-02-09
+- **Authors:** Shuyang Gong, Dong Huang, Zhangsong Li
+- **abs:** https://arxiv.org/abs/2602.08173 · **PDF:** https://arxiv.org/pdf/2602.08173v1
+
+## 14233. A Causal Machine Learning Framework for Treatment Personalization in Clinical Trials: Application to Ulcerative Colitis
+
+- **ID:** `2602.08171` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Cristian Minoccheri, Sophia Tesic, Kayvan Najarian, et al.
+- **abs:** https://arxiv.org/abs/2602.08171 · **PDF:** https://arxiv.org/pdf/2602.08171v1
+
+## 14234. Physiologically Informed Deep Learning: A Multi-Scale Framework for Next-Generation PBPK Modeling
+
+- **ID:** `2602.18472` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Shunqi Liu, Han Qiu, Tong Wang
+- **abs:** https://arxiv.org/abs/2602.18472 · **PDF:** https://arxiv.org/pdf/2602.18472v1
+
+## 14235. Evasion of IoT Malware Detection via Dummy Code Injection
+
+- **ID:** `2602.08170` · **primary:** `cs.CR` · **Submitted:** 2026-02-09
+- **Authors:** Sahar Zargarzadeh, Mohammad Islam
+- **abs:** https://arxiv.org/abs/2602.08170 · **PDF:** https://arxiv.org/pdf/2602.08170v1
+
+## 14236. Spherical Steering: Geometry-Aware Activation Rotation for Language Models
+
+- **ID:** `2602.08169` · **primary:** `cs.LG` · **Submitted:** 2026-02-09
+- **Authors:** Zejia You, Chunyuan Deng, Hanjie Chen
+- **abs:** https://arxiv.org/abs/2602.08169 · **PDF:** https://arxiv.org/pdf/2602.08169v1
+
+## 14237. Self-Supervised Bootstrapping of Action-Predictive Embodied Reasoning
+
+- **ID:** `2602.08167` · **primary:** `cs.RO` · **Submitted:** 2026-02-09
+- **Authors:** Milan Ganai, Katie Luo, Jonas Frey, et al.
+- **abs:** https://arxiv.org/abs/2602.08167 · **PDF:** https://arxiv.org/pdf/2602.08167v1
+
+## 14238. Charting the Future of AI-supported Science Education: A Human-Centered Vision
+
+- **ID:** `2602.18471` · **primary:** `cs.CY` · **Submitted:** 2026-02-09
+- **Authors:** Xiaoming Zhai, Kent Crippen
+- **abs:** https://arxiv.org/abs/2602.18471 · **PDF:** https://arxiv.org/pdf/2602.18471v1
+
+## 14239. Transforming Science Learning Materials in the Era of Artificial Intelligence
+
+- **ID:** `2602.18470` · **primary:** `cs.CY` · **Submitted:** 2026-02-08
+- **Authors:** Xiaoming Zhai, Kent Crippen
+- **abs:** https://arxiv.org/abs/2602.18470 · **PDF:** https://arxiv.org/pdf/2602.18470v2
+
+## 14240. The Landscape of AI in Science Education: What is Changing and How to Respond
+
+- **ID:** `2602.18469` · **primary:** `cs.CY` · **Submitted:** 2026-02-08
+- **Authors:** Xiaoming Zhai, Kent Crippen
+- **abs:** https://arxiv.org/abs/2602.18469 · **PDF:** https://arxiv.org/pdf/2602.18469v2
+
+## 14241. NLP for Local Governance Meeting Records: A Focus Article on Tasks, Datasets, Metrics and Benchmark
+
+- **ID:** `2602.08162` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Ricardo Campos, José Pedro Evans, José Miguel Isidro, et al.
+- **abs:** https://arxiv.org/abs/2602.08162 · **PDF:** https://arxiv.org/pdf/2602.08162v1
+
+## 14242. The Confidence Manifold: Geometric Structure of Correctness Representations in Language Models
+
+- **ID:** `2602.08159` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Seonglae Cho, Zekun Wu, Kleyton Da Costa, et al.
+- **abs:** https://arxiv.org/abs/2602.08159 · **PDF:** https://arxiv.org/pdf/2602.08159v1
+
+## 14243. A second order regret bound for NormalHedge
+
+- **ID:** `2602.08151` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Yoav Freund, Nicholas J. A. Harvey, Victor S. Portella, et al.
+- **abs:** https://arxiv.org/abs/2602.08151 · **PDF:** https://arxiv.org/pdf/2602.08151v1
+
+## 14244. DIAL-SUMMER: A Structured Evaluation Framework of Hierarchical Errors in Dialogue Summaries
+
+- **ID:** `2602.08149` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Sahana Ramnath, Nima Chitsazan, Mingyang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.08149 · **PDF:** https://arxiv.org/pdf/2602.08149v1
+
+## 14245. The Logovista English-Japanese Machine Translation System
+
+- **ID:** `2603.03311` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Barton D. Wright
+- **abs:** https://arxiv.org/abs/2603.03311 · **PDF:** https://arxiv.org/pdf/2603.03311v1
+
+## 14246. Variance-Gated Ensembles: An Epistemic-Aware Framework for Uncertainty Estimation
+
+- **ID:** `2602.08142` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** H. Martin Gillis, Isaac Xu, Thomas Trappenberg
+- **abs:** https://arxiv.org/abs/2602.08142 · **PDF:** https://arxiv.org/pdf/2602.08142v1
+
+## 14247. Robustness of Vision Language Models Against Split-Image Harmful Input Attacks
+
+- **ID:** `2602.08136` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Md Rafi Ur Rashid, MD Sadik Hossain Shanto, Vishnu Asutosh Dasu, et al.
+- **abs:** https://arxiv.org/abs/2602.08136 · **PDF:** https://arxiv.org/pdf/2602.08136v1
+
+## 14248. Adjustment of Cluster-Then-Predict Framework for Multiport Scatterer Load Prediction
+
+- **ID:** `2602.08129` · **primary:** `eess.SP` · **Submitted:** 2026-02-08
+- **Authors:** Hanjun Park, Aleksandr D. Kuznetsov, Ville Viikari
+- **abs:** https://arxiv.org/abs/2602.08129 · **PDF:** https://arxiv.org/pdf/2602.08129v1
+
+## 14249. Online Bayesian Imbalanced Learning with Bregman-Calibrated Deep Networks
+
+- **ID:** `2602.08128` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Zahir Alsulaimawi
+- **abs:** https://arxiv.org/abs/2602.08128 · **PDF:** https://arxiv.org/pdf/2602.08128v1
+
+## 14250. Gender and Race Bias in Consumer Product Recommendations by Large Language Models
+
+- **ID:** `2602.08124` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Ke Xu, Shera Potka, Alex Thomo
+- **abs:** https://arxiv.org/abs/2602.08124 · **PDF:** https://arxiv.org/pdf/2602.08124v1
+
+## 14251. Initial Risk Probing and Feasibility Testing of Glow: a Generative AI-Powered Dialectical Behavior Therapy Skills Coach for Substance Use Recovery and HIV Prevention
+
+- **ID:** `2602.08121` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Liying Wang, Madison Lee, Yunzhang Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.08121 · **PDF:** https://arxiv.org/pdf/2602.08121v1
+
+## 14252. Constrained Pricing under Finite Mixtures of Logit
+
+- **ID:** `2602.08119` · **primary:** `math.OC` · **Submitted:** 2026-02-08
+- **Authors:** Hoang Giang Pham, Tien Mai
+- **abs:** https://arxiv.org/abs/2602.08119 · **PDF:** https://arxiv.org/pdf/2602.08119v1
+
+## 14253. AGORA: Agentic Green Orchestration Architecture for Beyond 5G Networks
+
+- **ID:** `2602.13290` · **primary:** `cs.NI` · **Submitted:** 2026-02-08
+- **Authors:** Rodrigo Moreira, Larissa Ferreira Rodrigues Moreira, Maycon Peixoto, et al.
+- **abs:** https://arxiv.org/abs/2602.13290 · **PDF:** https://arxiv.org/pdf/2602.13290v1
+
+## 14254. MMLSv2: A Multimodal Dataset for Martian Landslide Detection in Remote Sensing Imagery
+
+- **ID:** `2602.08112` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Sidike Paheding, Abel Reyes-Angulo, Leo Thomas Ramos, et al.
+- **abs:** https://arxiv.org/abs/2602.08112 · **PDF:** https://arxiv.org/pdf/2602.08112v1
+
+## 14255. Evaluating the Impact of Post-Training Quantization on Reliable VQA with Multimodal LLMs
+
+- **ID:** `2602.13289` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Paul Jonas Kurz, Tobias Jan Wieczorek, Mohamed A. Abdelsalam, et al.
+- **abs:** https://arxiv.org/abs/2602.13289 · **PDF:** https://arxiv.org/pdf/2602.13289v1
+
+## 14256. Mutual information and task-relevant latent dimensionality
+
+- **ID:** `2602.08105` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Paarth Gulati, Eslam Abdelaleem, Audrey Sederberg, et al.
+- **abs:** https://arxiv.org/abs/2602.08105 · **PDF:** https://arxiv.org/pdf/2602.08105v1
+
+## 14257. Can Large Language Models Implement Agent-Based Models? An ODD-based Replication Study
+
+- **ID:** `2602.10140` · **primary:** `cs.SE` · **Submitted:** 2026-02-08
+- **Authors:** Nuno Fachada, Daniel Fernandes, Carlos M. Fernandes, et al.
+- **abs:** https://arxiv.org/abs/2602.10140 · **PDF:** https://arxiv.org/pdf/2602.10140v1
+
+## 14258. Interpretable Failure Analysis in Multi-Agent Reinforcement Learning Systems
+
+- **ID:** `2602.08104` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Risal Shahriar Shefin, Debashis Gupta, Thai Le, et al.
+- **abs:** https://arxiv.org/abs/2602.08104 · **PDF:** https://arxiv.org/pdf/2602.08104v2
+
+## 14259. Emergent Search and Backtracking in Latent Reasoning Models
+
+- **ID:** `2602.08100` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Jasmine Cui, Charles Ye
+- **abs:** https://arxiv.org/abs/2602.08100 · **PDF:** https://arxiv.org/pdf/2602.08100v1
+
+## 14260. Generative AI in Action: Field Experimental Evidence from Alibaba's Customer Service Operations
+
+- **ID:** `2603.29888` · **primary:** `cs.HC` · **Submitted:** 2026-02-08
+- **Authors:** Xiao Ni, Yiwei Wang, Tianjun Feng, et al.
+- **abs:** https://arxiv.org/abs/2603.29888 · **PDF:** https://arxiv.org/pdf/2603.29888v1
+
+## 14261. VidVec: Unlocking Video MLLM Embeddings for Video-Text Retrieval
+
+- **ID:** `2602.08099` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Issar Tzachor, Dvir Samuel, Rami Ben-Ari
+- **abs:** https://arxiv.org/abs/2602.08099 · **PDF:** https://arxiv.org/pdf/2602.08099v1
+
+## 14262. GAAVI: Global Asymptotic Anytime Valid Inference for the Conditional Mean Function
+
+- **ID:** `2602.08096` · **primary:** `stat.ME` · **Submitted:** 2026-02-08
+- **Authors:** Brian M Cho, Raaz Dwivedi, Nathan Kallus
+- **abs:** https://arxiv.org/abs/2602.08096 · **PDF:** https://arxiv.org/pdf/2602.08096v1
+
+## 14263. Objective Decoupling in Social Reinforcement Learning: Recovering Ground Truth from Sycophantic Majorities
+
+- **ID:** `2602.08092` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Majid Ghasemi, Mark Crowley
+- **abs:** https://arxiv.org/abs/2602.08092 · **PDF:** https://arxiv.org/pdf/2602.08092v1
+
+## 14264. Entropic-Time Inference: Self-Organizing Large Language Model Decoding Beyond Attention
+
+- **ID:** `2603.03310` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Andrew Kiruluta
+- **abs:** https://arxiv.org/abs/2603.03310 · **PDF:** https://arxiv.org/pdf/2603.03310v1
+
+## 14265. Online Domain-aware LLM Decoding for Continual Domain Evolution
+
+- **ID:** `2602.08088` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Mohammad Abu-Shaira, Weishi Shi
+- **abs:** https://arxiv.org/abs/2602.08088 · **PDF:** https://arxiv.org/pdf/2602.08088v1
+
+## 14266. Probability Hacking and the Design of Trustworthy ML for Signal Processing in C-UAS: A Scenario Based Method
+
+- **ID:** `2602.08086` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Liisa Janssens, Laura Middeldorp
+- **abs:** https://arxiv.org/abs/2602.08086 · **PDF:** https://arxiv.org/pdf/2602.08086v1
+
+## 14267. Large language models for spreading dynamics in complex systems
+
+- **ID:** `2602.08085` · **primary:** `physics.soc-ph` · **Submitted:** 2026-02-08
+- **Authors:** Shuyu Jiang, Hao Ren, Yichang Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.08085 · **PDF:** https://arxiv.org/pdf/2602.08085v1
+
+## 14268. Spectral Guardrails for Agents in the Wild: Detecting Tool Use Hallucinations via Attention Topology
+
+- **ID:** `2602.08082` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Valentin Noël
+- **abs:** https://arxiv.org/abs/2602.08082 · **PDF:** https://arxiv.org/pdf/2602.08082v1
+
+## 14269. The CAPSARII Approach to Cyber-Secure Wearable, Ultra-Low-Power Networked Sensors for Soldier Health Monitoring
+
+- **ID:** `2602.08080` · **primary:** `cs.ET` · **Submitted:** 2026-02-08
+- **Authors:** Luciano Bozzi, Christian Celidonio, Umberto Nuzzi, et al.
+- **abs:** https://arxiv.org/abs/2602.08080 · **PDF:** https://arxiv.org/pdf/2602.08080v1
+
+## 14270. Multimodal normative modeling in Alzheimers Disease with introspective variational autoencoders
+
+- **ID:** `2602.08077` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Sayantan Kumar, Peijie Qiu, Aristeidis Sotiras
+- **abs:** https://arxiv.org/abs/2602.08077 · **PDF:** https://arxiv.org/pdf/2602.08077v1
+
+## 14271. Enhancing Bandit Algorithms with LLMs for Time-varying User Preferences in Streaming Recommendations
+
+- **ID:** `2602.08067` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Chenglei Shen, Yi Zhan, Weijie Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.08067 · **PDF:** https://arxiv.org/pdf/2602.08067v1
+
+## 14272. SiameseNorm: Breaking the Barrier to Reconciling Pre/Post-Norm
+
+- **ID:** `2602.08064` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Tianyu Li, Dongchen Han, Zixuan Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.08064 · **PDF:** https://arxiv.org/pdf/2602.08064v1
+
+## 14273. Efficient Distribution Learning with Error Bounds in Wasserstein Distance
+
+- **ID:** `2602.08063` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Eduardo Figueiredo, Steven Adams, Luca Laurenti
+- **abs:** https://arxiv.org/abs/2602.08063 · **PDF:** https://arxiv.org/pdf/2602.08063v1
+
+## 14274. Efficient and Adaptable Detection of Malicious LLM Prompts via Bootstrap Aggregation
+
+- **ID:** `2602.08062` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Shayan Ali Hassan, Tao Ni, Zafar Ayyub Qazi, et al.
+- **abs:** https://arxiv.org/abs/2602.08062 · **PDF:** https://arxiv.org/pdf/2602.08062v1
+
+## 14275. Securing Dual-Use Pathogen Data of Concern
+
+- **ID:** `2602.08061` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Doni Bloomfield, Allison Berke, Moritz S. Hanke, et al.
+- **abs:** https://arxiv.org/abs/2602.08061 · **PDF:** https://arxiv.org/pdf/2602.08061v1
+
+## 14276. Compiler-Assisted Speculative Sampling for Accelerated LLM Inference on Heterogeneous Edge Devices
+
+- **ID:** `2602.08060` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Alejandro Ruiz y Mesa, Guilherme Korol, Moritz Riesterer, et al.
+- **abs:** https://arxiv.org/abs/2602.08060 · **PDF:** https://arxiv.org/pdf/2602.08060v2
+
+## 14277. DICE: Disentangling Artist Style from Content via Contrastive Subspace Decomposition in Diffusion Models
+
+- **ID:** `2602.08059` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Tong Zhang, Ru Zhang, Jianyi Liu
+- **abs:** https://arxiv.org/abs/2602.08059 · **PDF:** https://arxiv.org/pdf/2602.08059v1
+
+## 14278. Picasso: Holistic Scene Reconstruction with Physics-Constrained Sampling
+
+- **ID:** `2602.08058` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Xihang Yu, Rajat Talak, Lorenzo Shaikewitz, et al.
+- **abs:** https://arxiv.org/abs/2602.08058 · **PDF:** https://arxiv.org/pdf/2602.08058v1
+
+## 14279. Weak to Strong: VLM-Based Pseudo-Labeling as a Weakly Supervised Training Strategy in Multimodal Video-based Hidden Emotion Understanding Tasks
+
+- **ID:** `2602.08057` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Yufei Wang, Haixu Liu, Tianxiang Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.08057 · **PDF:** https://arxiv.org/pdf/2602.08057v1
+
+## 14280. Persistent Entropy as a Detector of Phase Transitions
+
+- **ID:** `2602.09058` · **primary:** `stat.ML` · **Submitted:** 2026-02-08
+- **Authors:** Matteo Rucco
+- **abs:** https://arxiv.org/abs/2602.09058 · **PDF:** https://arxiv.org/pdf/2602.09058v1
+
+## 14281. Epigraph-Guided Flow Matching for Safe and Performant Offline Reinforcement Learning
+
+- **ID:** `2602.08054` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Manan Tayal, Mumuksh Tayal
+- **abs:** https://arxiv.org/abs/2602.08054 · **PDF:** https://arxiv.org/pdf/2602.08054v1
+
+## 14282. Graph-Enhanced Deep Reinforcement Learning for Multi-Objective Unrelated Parallel Machine Scheduling
+
+- **ID:** `2602.08052` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Bulent Soykan, Sean Mondesire, Ghaith Rabadi, et al.
+- **abs:** https://arxiv.org/abs/2602.08052 · **PDF:** https://arxiv.org/pdf/2602.08052v1
+
+## 14283. Interpretable Fuzzy Systems For Forward Osmosis Desalination
+
+- **ID:** `2602.08050` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Qusai Khaled, Uzay Kaymak, Laura Genga
+- **abs:** https://arxiv.org/abs/2602.08050 · **PDF:** https://arxiv.org/pdf/2602.08050v1
+
+## 14284. Combating data scarcity in recommendation services: Integrating cognitive types of VARK and neural network technologies (LLM)
+
+- **ID:** `2603.03309` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Nikita Zmanovskii
+- **abs:** https://arxiv.org/abs/2603.03309 · **PDF:** https://arxiv.org/pdf/2603.03309v1
+
+## 14285. TDGNet: Hallucination Detection in Diffusion Language Models via Temporal Dynamic Graphs
+
+- **ID:** `2602.08048` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Arshia Hemmat, Philip Torr, Yongqiang Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.08048 · **PDF:** https://arxiv.org/pdf/2602.08048v1
+
+## 14286. V-ABFT: Variance-Based Adaptive Threshold for Fault-Tolerant Matrix Multiplication in Mixed-Precision Deep Learning
+
+- **ID:** `2602.08043` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Yiheng Gao, Qin Hua, Zizhong Chen
+- **abs:** https://arxiv.org/abs/2602.08043 · **PDF:** https://arxiv.org/pdf/2602.08043v1
+
+## 14287. Graph-based Semi-Supervised Learning via Maximum Discrimination
+
+- **ID:** `2602.08042` · **primary:** `stat.ML` · **Submitted:** 2026-02-08
+- **Authors:** Nadav Katz, Ariel Jaffe
+- **abs:** https://arxiv.org/abs/2602.08042 · **PDF:** https://arxiv.org/pdf/2602.08042v1
+
+## 14288. Implicit Strategic Optimization: Rethinking Long-Horizon Decision-Making in Adversarial Poker Environments
+
+- **ID:** `2602.08041` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Boyang Xia, Weiyou Tian, Qingnan Ren, et al.
+- **abs:** https://arxiv.org/abs/2602.08041 · **PDF:** https://arxiv.org/pdf/2602.08041v1
+
+## 14289. FIRE: Frobenius-Isometry Reinitialization for Balancing the Stability-Plasticity Tradeoff
+
+- **ID:** `2602.08040` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Isaac Han, Sangyeon Park, Seungwon Oh, et al.
+- **abs:** https://arxiv.org/abs/2602.08040 · **PDF:** https://arxiv.org/pdf/2602.08040v3
+
+## 14290. TAAM:Inductive Graph-Class Incremental Learning with Task-Aware Adaptive Modulation
+
+- **ID:** `2602.08036` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Jingtao Liu, Xinming Zhang
+- **abs:** https://arxiv.org/abs/2602.08036 · **PDF:** https://arxiv.org/pdf/2602.08036v1
+
+## 14291. The Benefits of Diversity: Combining Comparisons and Ratings for Efficient Scoring
+
+- **ID:** `2602.08033` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Julien Fageot, Matthias Grossglauser, Lê-Nguyên Hoang, et al.
+- **abs:** https://arxiv.org/abs/2602.08033 · **PDF:** https://arxiv.org/pdf/2602.08033v1
+
+## 14292. Horizon Imagination: Efficient On-Policy Rollout in Diffusion World Models
+
+- **ID:** `2602.08032` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Lior Cohen, Ofir Nabati, Kaixin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.08032 · **PDF:** https://arxiv.org/pdf/2602.08032v2
+
+## 14293. Beyond Raw Detection Scores: Markov-Informed Calibration for Boosting Machine-Generated Text Detection
+
+- **ID:** `2602.08031` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Chenwang Wu, Yiu-ming Cheung, Shuhai Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.08031 · **PDF:** https://arxiv.org/pdf/2602.08031v1
+
+## 14294. Free(): Learning to Forget in Malloc-Only Reasoning Models
+
+- **ID:** `2602.08030` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Yilun Zheng, Dongyang Ma, Tian Liang, et al.
+- **abs:** https://arxiv.org/abs/2602.08030 · **PDF:** https://arxiv.org/pdf/2602.08030v2
+
+## 14295. The Algorithmic Unconscious: Structural Mechanisms and Implicit Biases in Large Language Models
+
+- **ID:** `2602.18468` · **primary:** `cs.CY` · **Submitted:** 2026-02-08
+- **Authors:** Philippe Boisnard
+- **abs:** https://arxiv.org/abs/2602.18468 · **PDF:** https://arxiv.org/pdf/2602.18468v1
+
+## 14296. Diverge to Induce Prompting: Multi-Rationale Induction for Zero-Shot Reasoning
+
+- **ID:** `2602.08028` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Po-Chun Chen, Hen-Hsen Huang, Hsin-Hsi Chen
+- **abs:** https://arxiv.org/abs/2602.08028 · **PDF:** https://arxiv.org/pdf/2602.08028v1
+
+## 14297. Sharp analysis of linear ensemble sampling
+
+- **ID:** `2602.08026` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Arya Akhavan, David Janz, Csaba Szepesvári
+- **abs:** https://arxiv.org/abs/2602.08026 · **PDF:** https://arxiv.org/pdf/2602.08026v1
+
+## 14298. MIND: Benchmarking Memory Consistency and Action Control in World Models
+
+- **ID:** `2602.08025` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Yixuan Ye, Xuanyu Lu, Yuxin Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.08025 · **PDF:** https://arxiv.org/pdf/2602.08025v2
+
+## 14299. FlashVID: Efficient Video Large Language Models via Training-free Tree-based Spatiotemporal Token Merging
+
+- **ID:** `2602.08024` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Ziyang Fan, Keyu Chen, Ruilong Xing, et al.
+- **abs:** https://arxiv.org/abs/2602.08024 · **PDF:** https://arxiv.org/pdf/2602.08024v1
+
+## 14300. CyberExplorer: Benchmarking LLM Offensive Security Capabilities in a Real-World Attacking Simulation Environment
+
+- **ID:** `2602.08023` · **primary:** `cs.CR` · **Submitted:** 2026-02-08
+- **Authors:** Nanda Rani, Kimberly Milner, Minghao Shao, et al.
+- **abs:** https://arxiv.org/abs/2602.08023 · **PDF:** https://arxiv.org/pdf/2602.08023v2
+
+## 14301. Structure-Aware Robust Counterfactual Explanations via Conditional Gaussian Network Classifiers
+
+- **ID:** `2602.08021` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Zhan-Yi Liao, Jaewon Yoo, Hao-Tsung Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.08021 · **PDF:** https://arxiv.org/pdf/2602.08021v1
+
+## 14302. Anonymization-Enhanced Privacy Protection for Mobile GUI Agents: Available but Invisible
+
+- **ID:** `2602.10139` · **primary:** `cs.CR` · **Submitted:** 2026-02-08
+- **Authors:** Lepeng Zhao, Zhenhua Zou, Shuo Li, et al.
+- **abs:** https://arxiv.org/abs/2602.10139 · **PDF:** https://arxiv.org/pdf/2602.10139v2
+
+## 14303. The Rise of Sparse Mixture-of-Experts: A Survey from Algorithmic Foundations to Decentralized Architectures and Vertical Domain Applications
+
+- **ID:** `2602.08019` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Dong Pan, Bingtao Li, Yongsheng Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.08019 · **PDF:** https://arxiv.org/pdf/2602.08019v1
+
+## 14304. ICBAC: an Intelligent Contract-Based Access Control framework for supply chain management by integrating blockchain and federated learning
+
+- **ID:** `2602.08014` · **primary:** `cs.CR` · **Submitted:** 2026-02-08
+- **Authors:** Sadegh Sohani, Salar Ghazi, Farnaz Kamranfar, et al.
+- **abs:** https://arxiv.org/abs/2602.08014 · **PDF:** https://arxiv.org/pdf/2602.08014v1
+
+## 14305. Small Agent Group is the Future of Digital Health
+
+- **ID:** `2602.08013` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Yuqiao Meng, Luoxi Tang, Dazheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.08013 · **PDF:** https://arxiv.org/pdf/2602.08013v1
+
+## 14306. A Unified Density Operator View of Flow Control and Merging
+
+- **ID:** `2602.08012` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Riccardo De Santi, Malte Franke, Ya-Ping Hsieh, et al.
+- **abs:** https://arxiv.org/abs/2602.08012 · **PDF:** https://arxiv.org/pdf/2602.08012v1
+
+## 14307. Towards Adaptive, Scalable, and Robust Coordination of LLM Agents: A Dynamic Ad-Hoc Networking Perspective
+
+- **ID:** `2602.08009` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Rui Li, Zeyu Zhang, Xiaohe Bo, et al.
+- **abs:** https://arxiv.org/abs/2602.08009 · **PDF:** https://arxiv.org/pdf/2602.08009v1
+
+## 14308. From $O(mn)$ to $O(r^2)$: Two-Sided Low-Rank Communication for Adam in Distributed Training with Memory Efficiency
+
+- **ID:** `2602.08007` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Sizhe Dang, Jiaqi Shao, Xiaodong Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.08007 · **PDF:** https://arxiv.org/pdf/2602.08007v1
+
+## 14309. ForecastOcc: Vision-based Semantic Occupancy Forecasting
+
+- **ID:** `2602.08006` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Riya Mohan, Juana Valeria Hurtado, Rohit Mohan, et al.
+- **abs:** https://arxiv.org/abs/2602.08006 · **PDF:** https://arxiv.org/pdf/2602.08006v1
+
+## 14310. DeltaKV: Residual-Based KV Cache Compression via Long-Range Similarity
+
+- **ID:** `2602.08005` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Jitai Hao, Qiang Huang, Yaowei Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.08005 · **PDF:** https://arxiv.org/pdf/2602.08005v1
+
+## 14311. Can Multimodal LLMs See Science Instruction? Benchmarking Pedagogical Reasoning in K-12 Classroom Videos
+
+- **ID:** `2602.18466` · **primary:** `cs.CY` · **Submitted:** 2026-02-08
+- **Authors:** Yixuan Shen, Peng He, Honglu Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.18466 · **PDF:** https://arxiv.org/pdf/2602.18466v2
+
+## 14312. Don't Always Pick the Highest-Performing Model: An Information Theoretic View of LLM Ensemble Selection
+
+- **ID:** `2602.08003` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Yigit Turkmen, Baturalp Buyukates, Melih Bastopcu
+- **abs:** https://arxiv.org/abs/2602.08003 · **PDF:** https://arxiv.org/pdf/2602.08003v1
+
+## 14313. Regret Analysis of Unichain Average Reward Constrained MDPs with General Parameterization
+
+- **ID:** `2602.08000` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Anirudh Satheesh, Vaneet Aggarwal
+- **abs:** https://arxiv.org/abs/2602.08000 · **PDF:** https://arxiv.org/pdf/2602.08000v1
+
+## 14314. Tighter Information-Theoretic Generalization Bounds via a Novel Class of Change of Measure Inequalities
+
+- **ID:** `2602.07999` · **primary:** `cs.IT` · **Submitted:** 2026-02-08
+- **Authors:** Yanxiao Liu, Yijun Fan, Deniz Gündüz
+- **abs:** https://arxiv.org/abs/2602.07999 · **PDF:** https://arxiv.org/pdf/2602.07999v2
+
+## 14315. Fast Model Selection and Stable Optimization for Softmax-Gated Multinomial-Logistic Mixture of Experts Models
+
+- **ID:** `2602.07997` · **primary:** `stat.ML` · **Submitted:** 2026-02-08
+- **Authors:** TrungKhang Tran, TrungTin Nguyen, Md Abul Bashar, et al.
+- **abs:** https://arxiv.org/abs/2602.07997 · **PDF:** https://arxiv.org/pdf/2602.07997v1
+
+## 14316. The Judge Who Never Admits: Hidden Shortcuts in LLM-based Evaluation
+
+- **ID:** `2602.07996` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Arash Marioriyad, Omid Ghahroodi, Ehsaneddin Asgari, et al.
+- **abs:** https://arxiv.org/abs/2602.07996 · **PDF:** https://arxiv.org/pdf/2602.07996v1
+
+## 14317. MCIE: Multimodal LLM-Driven Complex Instruction Image Editing with Spatial Guidance
+
+- **ID:** `2602.07993` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Xuehai Bai, Xiaoling Gu, Akide Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.07993 · **PDF:** https://arxiv.org/pdf/2602.07993v1
+
+## 14318. When Is Compositional Reasoning Learnable from Verifiable Rewards?
+
+- **ID:** `2602.07992` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Daniel Barzilai, Yotam Wolf, Ronen Basri
+- **abs:** https://arxiv.org/abs/2602.07992 · **PDF:** https://arxiv.org/pdf/2602.07992v1
+
+## 14319. Learning to Alleviate Familiarity Bias in Video Recommendation
+
+- **ID:** `2602.07987` · **primary:** `cs.IR` · **Submitted:** 2026-02-08
+- **Authors:** Zheng Ren, Yi Wu, Jianan Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.07987 · **PDF:** https://arxiv.org/pdf/2602.07987v1
+
+## 14320. Accelerating Social Science Research via Agentic Hypothesization and Experimentation
+
+- **ID:** `2602.07983` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Jishu Sen Gupta, Harini SI, Somesh Kumar Singh, et al.
+- **abs:** https://arxiv.org/abs/2602.07983 · **PDF:** https://arxiv.org/pdf/2602.07983v1
+
+## 14321. Old Habits Die Hard: How Conversational History Geometrically Traps LLMs
+
+- **ID:** `2603.03308` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Adi Simhi, Fazl Barez, Martin Tutek, et al.
+- **abs:** https://arxiv.org/abs/2603.03308 · **PDF:** https://arxiv.org/pdf/2603.03308v1
+
+## 14322. Cross-Linguistic Persona-Driven Data Synthesis for Robust Multimodal Cognitive Decline Detection
+
+- **ID:** `2602.07978` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Rui Feng, Zhiyao Luo, Liuyu Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.07978 · **PDF:** https://arxiv.org/pdf/2602.07978v1
+
+## 14323. Beyond Optimization: Intelligence as Metric-Topology Factorization under Geometric Incompleteness
+
+- **ID:** `2602.07974` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Xin Li
+- **abs:** https://arxiv.org/abs/2602.07974 · **PDF:** https://arxiv.org/pdf/2602.07974v1
+
+## 14324. On Improving Neurosymbolic Learning by Exploiting the Representation Space
+
+- **ID:** `2602.07973` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Aaditya Naik, Efthymia Tsamoura, Shibo Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.07973 · **PDF:** https://arxiv.org/pdf/2602.07973v1
+
+## 14325. Learning-guided Kansa collocation for forward and inverse PDEs beyond linearity
+
+- **ID:** `2602.07970` · **primary:** `cs.CE` · **Submitted:** 2026-02-08
+- **Authors:** Zheyuan Hu, Weitao Chen, Cengiz Öztireli, et al.
+- **abs:** https://arxiv.org/abs/2602.07970 · **PDF:** https://arxiv.org/pdf/2602.07970v3
+
+## 14326. An Explainable Multi-Task Similarity Measure: Integrating Accumulated Local Effects and Weighted Fréchet Distance
+
+- **ID:** `2602.07966` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Pablo Hidalgo, Daniel Rodriguez
+- **abs:** https://arxiv.org/abs/2602.07966 · **PDF:** https://arxiv.org/pdf/2602.07966v1
+
+## 14327. Lost in Translation? A Comparative Study on the Cross-Lingual Transfer of Composite Harms
+
+- **ID:** `2602.07963` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Vaibhav Shukla, Hardik Sharma, Adith N Reganti, et al.
+- **abs:** https://arxiv.org/abs/2602.07963 · **PDF:** https://arxiv.org/pdf/2602.07963v1
+
+## 14328. LOCA-bench: Benchmarking Language Agents Under Controllable and Extreme Context Growth
+
+- **ID:** `2602.07962` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Weihao Zeng, Yuzhen Huang, Junxian He
+- **abs:** https://arxiv.org/abs/2602.07962 · **PDF:** https://arxiv.org/pdf/2602.07962v1
+
+## 14329. Accuracy-Delay Trade-Off in LLM Offloading via Token-Level Uncertainty
+
+- **ID:** `2602.07958` · **primary:** `eess.SY` · **Submitted:** 2026-02-08
+- **Authors:** Yumin Kim, Hyeonsu Lyu, Minjae Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.07958 · **PDF:** https://arxiv.org/pdf/2602.07958v1
+
+## 14330. Multimodal Information Fusion for Chart Understanding: A Survey of MLLMs -- Evolution, Limitations, and Cognitive Enhancement
+
+- **ID:** `2602.10138` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Zhihang Yi, Jian Zhao, Jiancheng Lv, et al.
+- **abs:** https://arxiv.org/abs/2602.10138 · **PDF:** https://arxiv.org/pdf/2602.10138v1
+
+## 14331. Bielik Guard: Efficient Polish Language Safety Classifiers for LLM Content Moderation
+
+- **ID:** `2602.07954` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Krzysztof Wróbel, Jan Maria Kowalski, Jerzy Surma, et al.
+- **abs:** https://arxiv.org/abs/2602.07954 · **PDF:** https://arxiv.org/pdf/2602.07954v3
+
+## 14332. A Thermodynamic Theory of Learning Part II: Critical Period Closure and Continual Learning Failure
+
+- **ID:** `2602.07950` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Daisuke Okanohara
+- **abs:** https://arxiv.org/abs/2602.07950 · **PDF:** https://arxiv.org/pdf/2602.07950v2
+
+## 14333. Multi-encoder ConvNeXt Network with Smooth Attentional Feature Fusion for Multispectral Semantic Segmentation
+
+- **ID:** `2602.10137` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Leo Thomas Ramos, Angel D. Sappa
+- **abs:** https://arxiv.org/abs/2602.10137 · **PDF:** https://arxiv.org/pdf/2602.10137v1
+
+## 14334. UniRank: End-to-End Domain-Specific Reranking of Hybrid Text-Image Candidates
+
+- **ID:** `2603.29897` · **primary:** `cs.IR` · **Submitted:** 2026-02-08
+- **Authors:** Yupei Yang, Lin Yang, Wanxi Deng, et al.
+- **abs:** https://arxiv.org/abs/2603.29897 · **PDF:** https://arxiv.org/pdf/2603.29897v1
+
+## 14335. IV Co-Scientist: Multi-Agent LLM Framework for Causal Instrumental Variable Discovery
+
+- **ID:** `2602.07943` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Ivaxi Sheth, Zhijing Jin, Bryan Wilder, et al.
+- **abs:** https://arxiv.org/abs/2602.07943 · **PDF:** https://arxiv.org/pdf/2602.07943v1
+
+## 14336. MePo: Meta Post-Refinement for Rehearsal-Free General Continual Learning
+
+- **ID:** `2602.07940` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Guanglong Sun, Hongwei Yan, Liyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.07940 · **PDF:** https://arxiv.org/pdf/2602.07940v2
+
+## 14337. Attention-Based Deep Learning for Early Parkinson's Disease Detection with Tabular Biomedical Data
+
+- **ID:** `2602.07933` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Olamide Samuel Oseni, Ibraheem Omotolani Obanla, Toheeb Aduramomi Jimoh
+- **abs:** https://arxiv.org/abs/2602.07933 · **PDF:** https://arxiv.org/pdf/2602.07933v1
+
+## 14338. TopicENA: Enabling Epistemic Network Analysis at Scale through Automated Topic-Based Coding
+
+- **ID:** `2603.03307` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Owen H. T. Lu, Tiffany T. Y. Hsu
+- **abs:** https://arxiv.org/abs/2603.03307 · **PDF:** https://arxiv.org/pdf/2603.03307v1
+
+## 14339. Token-Oriented Object Notation vs JSON: A Benchmark of Plain and Constrained Decoding Generation
+
+- **ID:** `2603.03306` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Ivan Matveev
+- **abs:** https://arxiv.org/abs/2603.03306 · **PDF:** https://arxiv.org/pdf/2603.03306v1
+
+## 14340. Patches of Nonlinearity: Instruction Vectors in Large Language Models
+
+- **ID:** `2602.07930` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Irina Bigoulaeva, Jonas Rohweder, Subhabrata Dutta, et al.
+- **abs:** https://arxiv.org/abs/2602.07930 · **PDF:** https://arxiv.org/pdf/2602.07930v1
+
+## 14341. A Kinetic-Energy Perspective of Flow Matching
+
+- **ID:** `2602.07928` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Ziyun Li, Huancheng Hu, Soon Hoe Lim, et al.
+- **abs:** https://arxiv.org/abs/2602.07928 · **PDF:** https://arxiv.org/pdf/2602.07928v1
+
+## 14342. Optimized Human-Robot Co-Dispatch Planning for Petro-Site Surveillance under Varying Criticalities
+
+- **ID:** `2602.07924` · **primary:** `cs.RO` · **Submitted:** 2026-02-08
+- **Authors:** Nur Ahmad Khatim, Mansur Arief
+- **abs:** https://arxiv.org/abs/2602.07924 · **PDF:** https://arxiv.org/pdf/2602.07924v1
+
+## 14343. Selective Fine-Tuning for Targeted and Robust Concept Unlearning
+
+- **ID:** `2602.07919` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Mansi, Avinash Kori, Francesca Toni, et al.
+- **abs:** https://arxiv.org/abs/2602.07919 · **PDF:** https://arxiv.org/pdf/2602.07919v1
+
+## 14344. CausalArmor: Efficient Indirect Prompt Injection Guardrails via Causal Attribution
+
+- **ID:** `2602.07918` · **primary:** `cs.CR` · **Submitted:** 2026-02-08
+- **Authors:** Minbeom Kim, Mihir Parmar, Phillip Wallis, et al.
+- **abs:** https://arxiv.org/abs/2602.07918 · **PDF:** https://arxiv.org/pdf/2602.07918v1
+
+## 14345. CausalCompass: Evaluating the Robustness of Time-Series Causal Discovery in Misspecified Scenarios
+
+- **ID:** `2602.07915` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Huiyang Yi, Xiaojian Shen, Yonggang Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.07915 · **PDF:** https://arxiv.org/pdf/2602.07915v1
+
+## 14346. SparseEval: Efficient Evaluation of Large Language Models by Sparse Optimization
+
+- **ID:** `2602.07909` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Taolin Zhang, Hang Guo, Wang Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.07909 · **PDF:** https://arxiv.org/pdf/2602.07909v1
+
+## 14347. AceGRPO: Adaptive Curriculum Enhanced Group Relative Policy Optimization for Autonomous Machine Learning Engineering
+
+- **ID:** `2602.07906` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Yuzhu Cai, Zexi Liu, Xinyu Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.07906 · **PDF:** https://arxiv.org/pdf/2602.07906v4
+
+## 14348. MedCoG: Maximizing LLM Inference Density in Medical Reasoning via Meta-Cognitive Regulation
+
+- **ID:** `2602.07905` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Yu Zhao, Hao Guan, Yongcheng Jing, et al.
+- **abs:** https://arxiv.org/abs/2602.07905 · **PDF:** https://arxiv.org/pdf/2602.07905v1
+
+## 14349. Adaptive Acquisition Selection for Bayesian Optimization with Large Language Models
+
+- **ID:** `2602.07904` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Giang Ngo, Dat Phan Trong, Dang Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.07904 · **PDF:** https://arxiv.org/pdf/2602.07904v1
+
+## 14350. GCN-MPPR: Enhancing the Propagation of Message Passing Neural Networks via Motif-Based Personalized PageRank
+
+- **ID:** `2602.07903` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Mingcan Wang, Junchang Xin, Zhongming Yao, et al.
+- **abs:** https://arxiv.org/abs/2602.07903 · **PDF:** https://arxiv.org/pdf/2602.07903v1
+
+## 14351. Incremental Mapping with Measurement Synchronization & Compression
+
+- **ID:** `2602.07901` · **primary:** `cs.RO` · **Submitted:** 2026-02-08
+- **Authors:** Mark Griguletskii, Danil Belov, Pavel Osinenko
+- **abs:** https://arxiv.org/abs/2602.07901 · **PDF:** https://arxiv.org/pdf/2602.07901v1
+
+## 14352. Rethinking the Value of Agent-Generated Tests for LLM-Based Software Engineering Agents
+
+- **ID:** `2602.07900` · **primary:** `cs.SE` · **Submitted:** 2026-02-08
+- **Authors:** Zhi Chen, Zhensu Sun, Yuling Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.07900 · **PDF:** https://arxiv.org/pdf/2602.07900v1
+
+## 14353. Mitigating Forgetting in Continual Learning with Selective Gradient Projection
+
+- **ID:** `2603.26671` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Anika Singh, Aayush Dhaulakhandi, Varun Chopade, et al.
+- **abs:** https://arxiv.org/abs/2603.26671 · **PDF:** https://arxiv.org/pdf/2603.26671v1
+
+## 14354. Safety Alignment as Continual Learning: Mitigating the Alignment Tax via Orthogonal Gradient Projection
+
+- **ID:** `2602.07892` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Guanglong Sun, Siyuan Zhang, Liyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.07892 · **PDF:** https://arxiv.org/pdf/2602.07892v1
+
+## 14355. Scalable Adaptation of 3D Geometric Foundation Models via Weak Supervision from Internet Video
+
+- **ID:** `2602.07891` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Zihui Gao, Ke Liu, Donny Y. Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.07891 · **PDF:** https://arxiv.org/pdf/2602.07891v1
+
+## 14356. StaTS: Spectral Trajectory Schedule Learning for Adaptive Time Series Forecasting with Frequency Guided Denoiser
+
+- **ID:** `2603.00037` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Jintao Zhang, Zirui Liu, Mingyue Cheng, et al.
+- **abs:** https://arxiv.org/abs/2603.00037 · **PDF:** https://arxiv.org/pdf/2603.00037v1
+
+## 14357. Efficient Anti-exploration via VQVAE and Fuzzy Clustering in Offline Reinforcement Learning
+
+- **ID:** `2602.07889` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Long Chen, Yinkui Liu, Shen Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07889 · **PDF:** https://arxiv.org/pdf/2602.07889v1
+
+## 14358. Rich-ARQ: From 1-bit Acknowledgment to Rich Neural Coded Feedback
+
+- **ID:** `2602.07886` · **primary:** `cs.IT` · **Submitted:** 2026-02-08
+- **Authors:** Enhao Chen, Yulin Shao
+- **abs:** https://arxiv.org/abs/2602.07886 · **PDF:** https://arxiv.org/pdf/2602.07886v1
+
+## 14359. MemFly: On-the-Fly Memory Optimization via Information Bottleneck
+
+- **ID:** `2602.07885` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Zhenyuan Zhang, Xianzhang Jia, Zhiqin Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.07885 · **PDF:** https://arxiv.org/pdf/2602.07885v1
+
+## 14360. GRAFT: Decoupling Ranking and Calibration for Survival Analysis
+
+- **ID:** `2602.07884` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Mohammad Ashhad, Robert Hoehndorf, Ricardo Henao
+- **abs:** https://arxiv.org/abs/2602.07884 · **PDF:** https://arxiv.org/pdf/2602.07884v1
+
+## 14361. ToolSelf: Unifying Task Execution and Self-Reconfiguration via Tool-Driven Intrinsic Adaptation
+
+- **ID:** `2602.07883` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Jingqi Zhou, Sheng Wang, DeZhao Deng, et al.
+- **abs:** https://arxiv.org/abs/2602.07883 · **PDF:** https://arxiv.org/pdf/2602.07883v2
+
+## 14362. Deep Variable-Length Feedback Codes
+
+- **ID:** `2602.07881` · **primary:** `cs.IT` · **Submitted:** 2026-02-08
+- **Authors:** Yu Ding, Yulin Shao
+- **abs:** https://arxiv.org/abs/2602.07881 · **PDF:** https://arxiv.org/pdf/2602.07881v1
+
+## 14363. Rethinking Latency Denial-of-Service: Attacking the LLM Serving Framework, Not the Model
+
+- **ID:** `2602.07878` · **primary:** `cs.CR` · **Submitted:** 2026-02-08
+- **Authors:** Tianyi Wang, Huawei Fan, Yuanchao Shu, et al.
+- **abs:** https://arxiv.org/abs/2602.07878 · **PDF:** https://arxiv.org/pdf/2602.07878v1
+
+## 14364. Harpoon: Generalised Manifold Guidance for Conditional Tabular Diffusion
+
+- **ID:** `2602.07875` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Aditya Shankar, Yuandou Wang, Rihan Hai, et al.
+- **abs:** https://arxiv.org/abs/2602.07875 · **PDF:** https://arxiv.org/pdf/2602.07875v2
+
+## 14365. Direct Soft-Policy Sampling via Langevin Dynamics
+
+- **ID:** `2602.07873` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Donghyeon Ki, Hee-Jun Ahn, Kyungyoon Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.07873 · **PDF:** https://arxiv.org/pdf/2602.07873v1
+
+## 14366. Orchestrating Attention: Bringing Harmony to the 'Chaos' of Neurodivergent Learning States
+
+- **ID:** `2602.07865` · **primary:** `cs.HC` · **Submitted:** 2026-02-08
+- **Authors:** Satyam Kumar Navneet, Joydeep Chandra, Yong Zhang
+- **abs:** https://arxiv.org/abs/2602.07865 · **PDF:** https://arxiv.org/pdf/2602.07865v1
+
+## 14367. Dynamic Load Model for Data Centers with Pattern-Consistent Calibration
+
+- **ID:** `2602.07859` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Siyu Lu, Chenhan Xiao, Yang Weng
+- **abs:** https://arxiv.org/abs/2602.07859 · **PDF:** https://arxiv.org/pdf/2602.07859v1
+
+## 14368. Emergent Misalignment is Easy, Narrow Misalignment is Hard
+
+- **ID:** `2602.07852` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Anna Soligo, Edward Turner, Senthooran Rajamanoharan, et al.
+- **abs:** https://arxiv.org/abs/2602.07852 · **PDF:** https://arxiv.org/pdf/2602.07852v1
+
+## 14369. LQA: A Lightweight Quantized-Adaptive Framework for Vision-Language Models on the Edge
+
+- **ID:** `2602.07849` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Xin Wang, Hong Jia, Hualin Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.07849 · **PDF:** https://arxiv.org/pdf/2602.07849v3
+
+## 14370. MARTI-MARS$^2$: Scaling Multi-Agent Self-Search via Reinforcement Learning for Code Generation
+
+- **ID:** `2602.07848` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Shijie Wang, Pengfei Li, Yikun Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.07848 · **PDF:** https://arxiv.org/pdf/2602.07848v1
+
+## 14371. Evaluating and Calibrating LLM Confidence on Questions with Multiple Correct Answers
+
+- **ID:** `2602.07842` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Yuhan Wang, Shiyu Ni, Zhikai Ding, et al.
+- **abs:** https://arxiv.org/abs/2602.07842 · **PDF:** https://arxiv.org/pdf/2602.07842v1
+
+## 14372. SAGE: Scalable AI Governance & Evaluation
+
+- **ID:** `2602.07840` · **primary:** `cs.IR` · **Submitted:** 2026-02-08
+- **Authors:** Benjamin Le, Xueying Lu, Nick Stern, et al.
+- **abs:** https://arxiv.org/abs/2602.07840 · **PDF:** https://arxiv.org/pdf/2602.07840v2
+
+## 14373. TodoEvolve: Learning to Architect Agent Planning Systems
+
+- **ID:** `2602.07839` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Jiaxi Liu, Yanzuo Jiang, Guibin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.07839 · **PDF:** https://arxiv.org/pdf/2602.07839v1
+
+## 14374. Explainable LLM Unlearning Through Reasoning
+
+- **ID:** `2603.09980` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Junfeng Liao, Qizhou Wang, Shanshan Ye, et al.
+- **abs:** https://arxiv.org/abs/2603.09980 · **PDF:** https://arxiv.org/pdf/2603.09980v1
+
+## 14375. Interpretable Analytic Calabi-Yau Metrics via Symbolic Distillation
+
+- **ID:** `2602.07834` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** D Yang Eng
+- **abs:** https://arxiv.org/abs/2602.07834 · **PDF:** https://arxiv.org/pdf/2602.07834v1
+
+## 14376. SPD-Faith Bench: Diagnosing and Improving Faithfulness in Chain-of-Thought for Multimodal Large Language Models
+
+- **ID:** `2602.07833` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Weijiang Lv, Yaoxuan Feng, Xiaobo Xia, et al.
+- **abs:** https://arxiv.org/abs/2602.07833 · **PDF:** https://arxiv.org/pdf/2602.07833v1
+
+## 14377. rePIRL: Learn PRM with Inverse RL for LLM Reasoning
+
+- **ID:** `2602.07832` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Xian Wu, Kaijie Zhu, Ying Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.07832 · **PDF:** https://arxiv.org/pdf/2602.07832v1
+
+## 14378. Time Series Reasoning via Process-Verifiable Thinking Data Synthesis and Scheduling for Tailored LLM Reasoning
+
+- **ID:** `2602.07830` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Jiahui Zhou, Dan Li, Boxin Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07830 · **PDF:** https://arxiv.org/pdf/2602.07830v1
+
+## 14379. Efficient Representations are Controllable Representations
+
+- **ID:** `2602.07828` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Charles Ye, Jasmine Cui
+- **abs:** https://arxiv.org/abs/2602.07828 · **PDF:** https://arxiv.org/pdf/2602.07828v1
+
+## 14380. Data Darwinism Part I: Unlocking the Value of Scientific Data for Pre-training
+
+- **ID:** `2602.07824` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Yiwei Qin, Zhen Huang, Tiantian Mi, et al.
+- **abs:** https://arxiv.org/abs/2602.07824 · **PDF:** https://arxiv.org/pdf/2602.07824v1
+
+## 14381. How well are open sourced AI-generated image detection models out-of-the-box: A comprehensive benchmark study
+
+- **ID:** `2602.07814` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Simiao Ren, Yuchen Zhou, Xingyu Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.07814 · **PDF:** https://arxiv.org/pdf/2602.07814v1
+
+## 14382. LLMs Know More About Numbers than They Can Say
+
+- **ID:** `2602.07812` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Fengting Yuchi, Li Du, Jason Eisner
+- **abs:** https://arxiv.org/abs/2602.07812 · **PDF:** https://arxiv.org/pdf/2602.07812v2
+
+## 14383. Draft-Conditioned Constrained Decoding for Structured Generation in LLMs
+
+- **ID:** `2603.03305` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Avinash Reddy, Thayne T. Walker, James S. Ide, et al.
+- **abs:** https://arxiv.org/abs/2603.03305 · **PDF:** https://arxiv.org/pdf/2603.03305v1
+
+## 14384. Pruning as a Cooperative Game: Surrogate-Assisted Layer Contribution Estimation for Large Language Models
+
+- **ID:** `2602.07804` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Xuan Ding, Pengyu Tong, Ranjie Duan, et al.
+- **abs:** https://arxiv.org/abs/2602.07804 · **PDF:** https://arxiv.org/pdf/2602.07804v1
+
+## 14385. SoulX-Singer: Towards High-Quality Zero-Shot Singing Voice Synthesis
+
+- **ID:** `2602.07803` · **primary:** `eess.AS` · **Submitted:** 2026-02-08
+- **Authors:** Jiale Qian, Hao Meng, Tian Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.07803 · **PDF:** https://arxiv.org/pdf/2602.07803v2
+
+## 14386. VideoTemp-o3: Harmonizing Temporal Grounding and Video Understanding in Agentic Thinking-with-Videos
+
+- **ID:** `2602.07801` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Wenqi Liu, Yunxiao Wang, Shijie Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.07801 · **PDF:** https://arxiv.org/pdf/2602.07801v3
+
+## 14387. Approximating Matrix Functions with Deep Neural Networks and Transformers
+
+- **ID:** `2602.07800` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Rahul Padmanabhan, Simone Brugiapaglia
+- **abs:** https://arxiv.org/abs/2602.07800 · **PDF:** https://arxiv.org/pdf/2602.07800v1
+
+## 14388. Fairness Aware Reward Optimization
+
+- **ID:** `2602.07799` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Ching Lam Choi, Vighnesh Subramaniam, Phillip Isola, et al.
+- **abs:** https://arxiv.org/abs/2602.07799 · **PDF:** https://arxiv.org/pdf/2602.07799v1
+
+## 14389. CausalTAD: Injecting Causal Knowledge into Large Language Models for Tabular Anomaly Detection
+
+- **ID:** `2602.07798` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Ruiqi Wang, Ruikang Liu, Runyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.07798 · **PDF:** https://arxiv.org/pdf/2602.07798v1
+
+## 14390. Thinking Makes LLM Agents Introverted: How Mandatory Thinking Can Backfire in User-Engaged Agents
+
+- **ID:** `2602.07796` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Jiatong Li, Changdae Oh, Hyeong Kyu Choi, et al.
+- **abs:** https://arxiv.org/abs/2602.07796 · **PDF:** https://arxiv.org/pdf/2602.07796v1
+
+## 14391. Emergent Structured Representations Support Flexible In-Context Inference in Large Language Models
+
+- **ID:** `2602.07794` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Ningyu Xu, Qi Zhang, Xipeng Qiu, et al.
+- **abs:** https://arxiv.org/abs/2602.07794 · **PDF:** https://arxiv.org/pdf/2602.07794v2
+
+## 14392. MaD-Mix: Multi-Modal Data Mixtures via Latent Space Coupling for Vision-Language Model Training
+
+- **ID:** `2602.07790` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Wanyun Xie, Francesco Tonin, Volkan Cevher
+- **abs:** https://arxiv.org/abs/2602.07790 · **PDF:** https://arxiv.org/pdf/2602.07790v1
+
+## 14393. Do Multi-Agents Dream of Electric Screens? Achieving Perfect Accuracy on AndroidWorld Through Task Decomposition
+
+- **ID:** `2602.07787` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Pierre-Louis Favreau, Jean-Pierre Lo, Clement Guiguet, et al.
+- **abs:** https://arxiv.org/abs/2602.07787 · **PDF:** https://arxiv.org/pdf/2602.07787v1
+
+## 14394. Still Manual? Automated Linter Configuration via DSL-Based LLM Compilation of Coding Standards
+
+- **ID:** `2602.07783` · **primary:** `cs.SE` · **Submitted:** 2026-02-08
+- **Authors:** Zejun Zhang, Yixin Gan, Zhenchang Xing, et al.
+- **abs:** https://arxiv.org/abs/2602.07783 · **PDF:** https://arxiv.org/pdf/2602.07783v1
+
+## 14395. Attn-GS: Attention-Guided Context Compression for Efficient Personalized LLMs
+
+- **ID:** `2602.07778` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Shenglai Zeng, Tianqi Zheng, Chuan Tian, et al.
+- **abs:** https://arxiv.org/abs/2602.07778 · **PDF:** https://arxiv.org/pdf/2602.07778v1
+
+## 14396. Generative Reasoning Re-ranker
+
+- **ID:** `2602.07774` · **primary:** `cs.IR` · **Submitted:** 2026-02-08
+- **Authors:** Mingfu Liang, Yufei Li, Jay Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.07774 · **PDF:** https://arxiv.org/pdf/2602.07774v4
+
+## 14397. SRR-Judge: Step-Level Rating and Refinement for Enhancing Search-Integrated Reasoning in Search Agents
+
+- **ID:** `2602.07773` · **primary:** `cs.CL` · **Submitted:** 2026-02-08
+- **Authors:** Chen Zhang, Kuicai Dong, Dexun Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07773 · **PDF:** https://arxiv.org/pdf/2602.07773v1
+
+## 14398. PAND: Prompt-Aware Neighborhood Distillation for Lightweight Fine-Grained Visual Classification
+
+- **ID:** `2602.07768` · **primary:** `cs.CV` · **Submitted:** 2026-02-08
+- **Authors:** Qiuming Luo, Yuebing Li, Feng Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07768 · **PDF:** https://arxiv.org/pdf/2602.07768v2
+
+## 14399. BFTS: Thompson Sampling with Bayesian Additive Regression Trees
+
+- **ID:** `2602.07767` · **primary:** `stat.ML` · **Submitted:** 2026-02-08
+- **Authors:** Ruizhe Deng, Bibhas Chakraborty, Ran Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.07767 · **PDF:** https://arxiv.org/pdf/2602.07767v1
+
+## 14400. Disentangled Instrumental Variables for Causal Inference with Networked Observational Data
+
+- **ID:** `2602.07765` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Zhirong Huang, Debo Cheng, Guixian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.07765 · **PDF:** https://arxiv.org/pdf/2602.07765v1
+
+## 14401. Preference Conditioned Multi-Objective Reinforcement Learning: Decomposed, Diversity-Driven Policy Optimization
+
+- **ID:** `2602.07764` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Tanmay Ambadkar, Sourav Panda, Shreyash Kale, et al.
+- **abs:** https://arxiv.org/abs/2602.07764 · **PDF:** https://arxiv.org/pdf/2602.07764v1
+
+## 14402. Learning to Continually Learn via Meta-learning Agentic Memory Designs
+
+- **ID:** `2602.07755` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Yiming Xiong, Shengran Hu, Jeff Clune
+- **abs:** https://arxiv.org/abs/2602.07755 · **PDF:** https://arxiv.org/pdf/2602.07755v1
+
+## 14403. Humanizing AI Grading: Student-Centered Insights on Fairness, Trust, Consistency and Transparency
+
+- **ID:** `2602.07754` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Bahare Riahi, Viktoriia Storozhevykh, Veronica Catete
+- **abs:** https://arxiv.org/abs/2602.07754 · **PDF:** https://arxiv.org/pdf/2602.07754v2
+
+## 14404. Geo-Code: A Code Framework for Reverse Code Generation from Geometric Images Based on Two-Stage Multi-Agent Evolution
+
+- **ID:** `2602.07749` · **primary:** `cs.AI` · **Submitted:** 2026-02-08
+- **Authors:** Zhenyu Wu, Yanxi Long, Jian Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07749 · **PDF:** https://arxiv.org/pdf/2602.07749v1
+
+## 14405. Riemannian MeanFlow
+
+- **ID:** `2602.07744` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Dongyeop Woo, Marta Skreta, Seonghyun Park, et al.
+- **abs:** https://arxiv.org/abs/2602.07744 · **PDF:** https://arxiv.org/pdf/2602.07744v2
+
+## 14406. HypRAG: Hyperbolic Dense Retrieval for Retrieval Augmented Generation
+
+- **ID:** `2602.07739` · **primary:** `cs.IR` · **Submitted:** 2026-02-08
+- **Authors:** Hiren Madhu, Ngoc Bui, Ali Maatouk, et al.
+- **abs:** https://arxiv.org/abs/2602.07739 · **PDF:** https://arxiv.org/pdf/2602.07739v1
+
+## 14407. Learnable Chernoff Baselines for Inference-Time Alignment
+
+- **ID:** `2602.07738` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Sunil Madhow, Yuchen Liang, Ness Shroff, et al.
+- **abs:** https://arxiv.org/abs/2602.07738 · **PDF:** https://arxiv.org/pdf/2602.07738v2
+
+## 14408. TerraBind: Fast and Accurate Binding Affinity Prediction through Coarse Structural Representations
+
+- **ID:** `2602.07735` · **primary:** `cs.LG` · **Submitted:** 2026-02-08
+- **Authors:** Matteo Rossi, Ryan Pederson, Miles Wang-Henderson, et al.
+- **abs:** https://arxiv.org/abs/2602.07735 · **PDF:** https://arxiv.org/pdf/2602.07735v1
+
+## 14409. Efficient Adaptive Data Analysis over Dense Distributions
+
+- **ID:** `2602.07732` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Joon Suk Huh
+- **abs:** https://arxiv.org/abs/2602.07732 · **PDF:** https://arxiv.org/pdf/2602.07732v1
+
+## 14410. The Laplacian Keyboard: Beyond the Linear Span
+
+- **ID:** `2602.07730` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Siddarth Chandrasekar, Marlos C. Machado
+- **abs:** https://arxiv.org/abs/2602.07730 · **PDF:** https://arxiv.org/pdf/2602.07730v1
+
+## 14411. Do We Need Adam? Surprisingly Strong and Sparse Reinforcement Learning with SGD in LLMs
+
+- **ID:** `2602.07729` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Sagnik Mukherjee, Lifan Yuan, Pavan Jayasinha, et al.
+- **abs:** https://arxiv.org/abs/2602.07729 · **PDF:** https://arxiv.org/pdf/2602.07729v2
+
+## 14412. Knowledge Graph and Hypergraph Transformers with Repository-Attention and Journey-Based Role Transport
+
+- **ID:** `2603.03304` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Mahesh Godavarti
+- **abs:** https://arxiv.org/abs/2603.03304 · **PDF:** https://arxiv.org/pdf/2603.03304v1
+
+## 14413. ParisKV: Fast and Drift-Robust KV-Cache Retrieval for Long-Context LLMs
+
+- **ID:** `2602.07721` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Yanlin Qi, Xinhang Chen, Huiqiang Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.07721 · **PDF:** https://arxiv.org/pdf/2602.07721v2
+
+## 14414. A Rational Account of Categorization Based on Information Theory
+
+- **ID:** `2603.29895` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Christophe J. MacLellan, Karthik Singaravadivelan, Xin Lian, et al.
+- **abs:** https://arxiv.org/abs/2603.29895 · **PDF:** https://arxiv.org/pdf/2603.29895v1
+
+## 14415. Efficient Planning in Reinforcement Learning via Model Introspection
+
+- **ID:** `2602.07719` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Gabriel Stella
+- **abs:** https://arxiv.org/abs/2602.07719 · **PDF:** https://arxiv.org/pdf/2602.07719v1
+
+## 14416. Analyzing and Guiding Zero-Shot Posterior Sampling in Diffusion Models
+
+- **ID:** `2602.07715` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Roi Benita, Michael Elad, Joseph Keshet
+- **abs:** https://arxiv.org/abs/2602.07715 · **PDF:** https://arxiv.org/pdf/2602.07715v1
+
+## 14417. Benchmarking Anomaly Detection Across Heterogeneous Cloud Telemetry Datasets
+
+- **ID:** `2602.13288` · **primary:** `cs.NI` · **Submitted:** 2026-02-07
+- **Authors:** Mohammad Saiful Islam, Andriy Miranskyy
+- **abs:** https://arxiv.org/abs/2602.13288 · **PDF:** https://arxiv.org/pdf/2602.13288v1
+
+## 14418. Towards Robust Scaling Laws for Optimizers
+
+- **ID:** `2602.07712` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Alexandra Volkova, Mher Safaryan, Christoph H. Lampert, et al.
+- **abs:** https://arxiv.org/abs/2602.07712 · **PDF:** https://arxiv.org/pdf/2602.07712v2
+
+## 14419. On Generation in Metric Spaces
+
+- **ID:** `2602.07710` · **primary:** `stat.ML` · **Submitted:** 2026-02-07
+- **Authors:** Jiaxun Li, Vinod Raman, Ambuj Tewari
+- **abs:** https://arxiv.org/abs/2602.07710 · **PDF:** https://arxiv.org/pdf/2602.07710v1
+
+## 14420. Patient-Centered, Graph-Augmented Artificial Intelligence-Enabled Passive Surveillance for Early Stroke Risk Detection in High-Risk Individuals
+
+- **ID:** `2602.22228` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Jiyeong Kim, Stephen P. Ma, Nirali Vora, et al.
+- **abs:** https://arxiv.org/abs/2602.22228 · **PDF:** https://arxiv.org/pdf/2602.22228v1
+
+## 14421. Generative structural elucidation from mass spectra as an iterative optimization problem
+
+- **ID:** `2602.07709` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-07
+- **Authors:** Mrunali Manjrekar, Runzhong Wang, Samuel Goldman, et al.
+- **abs:** https://arxiv.org/abs/2602.07709 · **PDF:** https://arxiv.org/pdf/2602.07709v1
+
+## 14422. Quantifying Explanation Quality in Graph Neural Networks using Out-of-Distribution Generalization
+
+- **ID:** `2602.07708` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Ding Zhang, Siddharth Betala, Chirag Agarwal
+- **abs:** https://arxiv.org/abs/2602.07708 · **PDF:** https://arxiv.org/pdf/2602.07708v1
+
+## 14423. Dense Feature Learning via Linear Structure Preservation in Medical Data
+
+- **ID:** `2602.07706` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Yuanyun Zhang, Mingxuan Zhang, Siyuan Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07706 · **PDF:** https://arxiv.org/pdf/2602.07706v1
+
+## 14424. On Sequence-to-Sequence Models for Automated Log Parsing
+
+- **ID:** `2602.07698` · **primary:** `cs.SE` · **Submitted:** 2026-02-07
+- **Authors:** Adam Sorrenti, Andriy Miranskyy
+- **abs:** https://arxiv.org/abs/2602.07698 · **PDF:** https://arxiv.org/pdf/2602.07698v1
+
+## 14425. On the Infinite Width and Depth Limits of Predictive Coding Networks
+
+- **ID:** `2602.07697` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Francesco Innocenti, El Mehdi Achour, Rafal Bogacz
+- **abs:** https://arxiv.org/abs/2602.07697 · **PDF:** https://arxiv.org/pdf/2602.07697v1
+
+## 14426. EventCast: Hybrid Demand Forecasting in E-Commerce with LLM-Based Event Knowledge
+
+- **ID:** `2602.07695` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Congcong Hu, Yuang Shi, Fan Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.07695 · **PDF:** https://arxiv.org/pdf/2602.07695v2
+
+## 14427. HumanLM: Simulating Users with State Alignment Beats Response Imitation
+
+- **ID:** `2603.03303` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Shirley Wu, Evelyn Choi, Arpandeep Khatua, et al.
+- **abs:** https://arxiv.org/abs/2603.03303 · **PDF:** https://arxiv.org/pdf/2603.03303v1
+
+## 14428. Process-of-Thought Reasoning for Videos
+
+- **ID:** `2602.07689` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Jusheng Zhang, Kaitong Cai, Jian Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.07689 · **PDF:** https://arxiv.org/pdf/2602.07689v1
+
+## 14429. Mapping Drivers of Greenness: Spatial Variable Selection for MODIS Vegetation Indices
+
+- **ID:** `2602.07681` · **primary:** `stat.ME` · **Submitted:** 2026-02-07
+- **Authors:** Qishi Zhan, Cheng-Han Yu, Yuchi Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.07681 · **PDF:** https://arxiv.org/pdf/2602.07681v2
+
+## 14430. Vision and Language: Novel Representations and Artificial intelligence for Driving Scene Safety Assessment and Autonomous Vehicle Planning
+
+- **ID:** `2602.07680` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Ross Greer, Maitrayee Keskar, Angel Martinez-Sanchez, et al.
+- **abs:** https://arxiv.org/abs/2602.07680 · **PDF:** https://arxiv.org/pdf/2602.07680v2
+
+## 14431. Spectral Gating Networks
+
+- **ID:** `2602.07679` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Jusheng Zhang, Yijia Fan, Kaitong Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.07679 · **PDF:** https://arxiv.org/pdf/2602.07679v1
+
+## 14432. Developing an AI Assistant for Knowledge Management and Workforce Training in State DOTs
+
+- **ID:** `2603.03302` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Divija Amaram, Lu Gao, Gowtham Reddy Gudla, et al.
+- **abs:** https://arxiv.org/abs/2603.03302 · **PDF:** https://arxiv.org/pdf/2603.03302v1
+
+## 14433. ElliCE: Efficient and Provably Robust Algorithmic Recourse via the Rashomon Sets
+
+- **ID:** `2602.07674` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Bohdan Turbal, Iryna Voitsitska, Lesia Semenova
+- **abs:** https://arxiv.org/abs/2602.07674 · **PDF:** https://arxiv.org/pdf/2602.07674v1
+
+## 14434. Blind to the Human Touch: Overlap Bias in LLM-Based Summary Evaluation
+
+- **ID:** `2602.07673` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Jiangnan Fang, Cheng-Tse Liu, Hanieh Deilamsalehy, et al.
+- **abs:** https://arxiv.org/abs/2602.07673 · **PDF:** https://arxiv.org/pdf/2602.07673v1
+
+## 14435. Debugging code world models
+
+- **ID:** `2602.07672` · **primary:** `cs.SE` · **Submitted:** 2026-02-07
+- **Authors:** Babak Rahmani
+- **abs:** https://arxiv.org/abs/2602.07672 · **PDF:** https://arxiv.org/pdf/2602.07672v2
+
+## 14436. Federated Learning with Profile Mapping under Distribution Shifts and Drifts
+
+- **ID:** `2602.07671` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Mohan Li, Dario Fenoglio, Martin Gjoreski, et al.
+- **abs:** https://arxiv.org/abs/2602.07671 · **PDF:** https://arxiv.org/pdf/2602.07671v1
+
+## 14437. Surprisal-Guided Selection: Compute-Optimal Test-Time Strategies for Execution-Grounded Code Generation
+
+- **ID:** `2602.07670` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Jarrod Barnes
+- **abs:** https://arxiv.org/abs/2602.07670 · **PDF:** https://arxiv.org/pdf/2602.07670v1
+
+## 14438. Looking and Listening Inside and Outside: Multimodal Artificial Intelligence Systems for Driver Safety Assessment and Intelligent Vehicle Decision-Making
+
+- **ID:** `2602.07668` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Ross Greer, Laura Fleig, Maitrayee Keskar, et al.
+- **abs:** https://arxiv.org/abs/2602.07668 · **PDF:** https://arxiv.org/pdf/2602.07668v1
+
+## 14439. Fast Response or Silence: Conversation Persistence in an AI-Agent Social Network
+
+- **ID:** `2602.07667` · **primary:** `econ.EM` · **Submitted:** 2026-02-07
+- **Authors:** Aysajan Eziz
+- **abs:** https://arxiv.org/abs/2602.07667 · **PDF:** https://arxiv.org/pdf/2602.07667v2
+
+## 14440. SoK: DARPA's AI Cyber Challenge (AIxCC): Competition Design, Architectures, and Lessons Learned
+
+- **ID:** `2602.07666` · **primary:** `cs.CR` · **Submitted:** 2026-02-07
+- **Authors:** Cen Zhang, Younggi Park, Fabian Fleischer, et al.
+- **abs:** https://arxiv.org/abs/2602.07666 · **PDF:** https://arxiv.org/pdf/2602.07666v2
+
+## 14441. SVD-Preconditioned Gradient Descent Method for Solving Nonlinear Least Squares Problems
+
+- **ID:** `2602.09057` · **primary:** `math.NA` · **Submitted:** 2026-02-07
+- **Authors:** Zhipeng Chang, Wenrui Hao, Nian Liu
+- **abs:** https://arxiv.org/abs/2602.09057 · **PDF:** https://arxiv.org/pdf/2602.09057v1
+
+## 14442. ONTrust: A Reference Ontology of Trust
+
+- **ID:** `2602.07662` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Glenda Amaral, Tiago Prince Sales, Riccardo Baratella, et al.
+- **abs:** https://arxiv.org/abs/2602.07662 · **PDF:** https://arxiv.org/pdf/2602.07662v1
+
+## 14443. Continuous Program Search
+
+- **ID:** `2602.07659` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Matthew Siper, Muhammad Umair Nasir, Ahmed Khalifa, et al.
+- **abs:** https://arxiv.org/abs/2602.07659 · **PDF:** https://arxiv.org/pdf/2602.07659v1
+
+## 14444. Agent-Fence: Mapping Security Vulnerabilities Across Deep Research Agents
+
+- **ID:** `2602.07652` · **primary:** `cs.CR` · **Submitted:** 2026-02-07
+- **Authors:** Sai Puppala, Ismail Hossain, Md Jahangir Alam, et al.
+- **abs:** https://arxiv.org/abs/2602.07652 · **PDF:** https://arxiv.org/pdf/2602.07652v1
+
+## 14445. From Dead Pixels to Editable Slides: Infographic Reconstruction into Native Google Slides via Vision-Language Region Understanding
+
+- **ID:** `2602.07645` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Leonardo Gonzalez
+- **abs:** https://arxiv.org/abs/2602.07645 · **PDF:** https://arxiv.org/pdf/2602.07645v1
+
+## 14446. Efficient Table Retrieval and Understanding with Multimodal Large Language Models
+
+- **ID:** `2602.07642` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Zhuoyan Xu, Haoyang Fang, Boran Han, et al.
+- **abs:** https://arxiv.org/abs/2602.07642 · **PDF:** https://arxiv.org/pdf/2602.07642v1
+
+## 14447. TASTE: Task-Aware Out-of-Distribution Detection via Stein Operators
+
+- **ID:** `2602.07640` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Michał Kozyra, Gesine Reinert
+- **abs:** https://arxiv.org/abs/2602.07640 · **PDF:** https://arxiv.org/pdf/2602.07640v1
+
+## 14448. Letting Tutor Personas "Speak Up" for LLMs: Learning Steering Vectors from Dialogue via Preference Optimization
+
+- **ID:** `2602.07639` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Jaewook Lee, Alexander Scarlatos, Simon Woodhead, et al.
+- **abs:** https://arxiv.org/abs/2602.07639 · **PDF:** https://arxiv.org/pdf/2602.07639v1
+
+## 14449. Flow-Based Conformal Predictive Distributions
+
+- **ID:** `2602.07633` · **primary:** `stat.ML` · **Submitted:** 2026-02-07
+- **Authors:** Trevor Harris
+- **abs:** https://arxiv.org/abs/2602.07633 · **PDF:** https://arxiv.org/pdf/2602.07633v2
+
+## 14450. Scalable Mean-Field Variational Inference via Preconditioned Primal-Dual Optimization
+
+- **ID:** `2602.07632` · **primary:** `stat.ML` · **Submitted:** 2026-02-07
+- **Authors:** Jinhua Lyu, Tianmin Yu, Ying Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.07632 · **PDF:** https://arxiv.org/pdf/2602.07632v2
+
+## 14451. SleepMaMi: A Universal Sleep Foundation Model for Integrating Macro- and Micro-structures
+
+- **ID:** `2602.07628` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Keondo Park, Younghoon Na, Yourim Choi, et al.
+- **abs:** https://arxiv.org/abs/2602.07628 · **PDF:** https://arxiv.org/pdf/2602.07628v1
+
+## 14452. AD-MIR: Bridging the Gap from Perception to Persuasion in Advertising Video Understanding via Structured Reasoning
+
+- **ID:** `2602.07625` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Binxiao Xu, Junyu Feng, Xiaopeng Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.07625 · **PDF:** https://arxiv.org/pdf/2602.07625v1
+
+## 14453. M2A: Multimodal Memory Agent with Dual-Layer Hybrid Memory for Long-Term Personalized Interactions
+
+- **ID:** `2602.07624` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Junyu Feng, Binxiao Xu, Jiayi Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.07624 · **PDF:** https://arxiv.org/pdf/2602.07624v1
+
+## 14454. SciClaimEval: Cross-modal Claim Verification in Scientific Papers
+
+- **ID:** `2602.07621` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Xanh Ho, Yun-Ang Wu, Sunisth Kumar, et al.
+- **abs:** https://arxiv.org/abs/2602.07621 · **PDF:** https://arxiv.org/pdf/2602.07621v2
+
+## 14455. Dense Neural Networks are not Universal Approximators
+
+- **ID:** `2602.07618` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Levi Rauchwerger, Stefanie Jegelka, Ron Levie
+- **abs:** https://arxiv.org/abs/2602.07618 · **PDF:** https://arxiv.org/pdf/2602.07618v2
+
+## 14456. SERE: Similarity-based Expert Re-routing for Efficient Batch Decoding in MoE Models
+
+- **ID:** `2602.07616` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Juntong Wu, Jialiang Cheng, Fuyu Lv, et al.
+- **abs:** https://arxiv.org/abs/2602.07616 · **PDF:** https://arxiv.org/pdf/2602.07616v1
+
+## 14457. Fast Rerandomization for Balancing Covariates in Randomized Experiments: A Metropolis-Hastings Framework
+
+- **ID:** `2602.07613` · **primary:** `stat.ME` · **Submitted:** 2026-02-07
+- **Authors:** Jiuyao Lu, Tianruo Zhang, Ke Zhu
+- **abs:** https://arxiv.org/abs/2602.07613 · **PDF:** https://arxiv.org/pdf/2602.07613v1
+
+## 14458. Evaluating Large Language Models for Detecting Architectural Decision Violations
+
+- **ID:** `2602.07609` · **primary:** `cs.SE` · **Submitted:** 2026-02-07
+- **Authors:** Ruoyu Su, Alexander Bakhtin, Noman Ahmad, et al.
+- **abs:** https://arxiv.org/abs/2602.07609 · **PDF:** https://arxiv.org/pdf/2602.07609v1
+
+## 14459. Fine-R1: Make Multi-modal LLMs Excel in Fine-Grained Visual Recognition by Chain-of-Thought Reasoning
+
+- **ID:** `2602.07605` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Hulingxiao He, Zijun Geng, Yuxin Peng
+- **abs:** https://arxiv.org/abs/2602.07605 · **PDF:** https://arxiv.org/pdf/2602.07605v2
+
+## 14460. Escaping Spectral Bias without Backpropagation: Fast Implicit Neural Representations with Extreme Learning Machines
+
+- **ID:** `2602.07603` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Woojin Cho, Junghwan Park
+- **abs:** https://arxiv.org/abs/2602.07603 · **PDF:** https://arxiv.org/pdf/2602.07603v1
+
+## 14461. Object-Oriented Transition Modeling with Inductive Logic Programming
+
+- **ID:** `2602.07602` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Gabriel Stella, Dmitri Loguinov
+- **abs:** https://arxiv.org/abs/2602.07602 · **PDF:** https://arxiv.org/pdf/2602.07602v1
+
+## 14462. Rational Transductors
+
+- **ID:** `2602.07599` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Mehryar Mohri
+- **abs:** https://arxiv.org/abs/2602.07599 · **PDF:** https://arxiv.org/pdf/2602.07599v1
+
+## 14463. Astro: Activation-guided Structured Regularization for Outlier-Robust LLM Post-Training Quantization
+
+- **ID:** `2602.07596` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Xi Chen, Ming Li, Junxi Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07596 · **PDF:** https://arxiv.org/pdf/2602.07596v1
+
+## 14464. TeleBoost: A Systematic Alignment Framework for High-Fidelity, Controllable, and Robust Video Generation
+
+- **ID:** `2602.07595` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Yuanzhi Liang, Xuan'er Wu, Yirui Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.07595 · **PDF:** https://arxiv.org/pdf/2602.07595v1
+
+## 14465. Learning to Self-Verify Makes Language Models Better Reasoners
+
+- **ID:** `2602.07594` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Yuxin Chen, Yu Wang, Yi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.07594 · **PDF:** https://arxiv.org/pdf/2602.07594v1
+
+## 14466. Beyond Arrow: From Impossibility to Possibilities in Multi-Criteria Benchmarking
+
+- **ID:** `2602.07593` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Polina Gordienko, Christoph Jansen, Julian Rodemann, et al.
+- **abs:** https://arxiv.org/abs/2602.07593 · **PDF:** https://arxiv.org/pdf/2602.07593v1
+
+## 14467. Automated rock joint trace mapping using a supervised learning model trained on synthetic data generated by parametric modelling
+
+- **ID:** `2602.07590` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Jessica Ka Yi Chiu, Tom Frode Hansen, Eivind Magnus Paulsen, et al.
+- **abs:** https://arxiv.org/abs/2602.07590 · **PDF:** https://arxiv.org/pdf/2602.07590v1
+
+## 14468. Unified Biomolecular Trajectory Generation via Pretrained Variational Bridge
+
+- **ID:** `2602.07588` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Ziyang Yu, Wenbing Huang, Yang Liu
+- **abs:** https://arxiv.org/abs/2602.07588 · **PDF:** https://arxiv.org/pdf/2602.07588v2
+
+## 14469. $\partial$CBDs: Differentiable Causal Block Diagrams
+
+- **ID:** `2602.07581` · **primary:** `eess.SY` · **Submitted:** 2026-02-07
+- **Authors:** Thomas Beckers, Ján Drgoňa, Truong X. Nghiem
+- **abs:** https://arxiv.org/abs/2602.07581 · **PDF:** https://arxiv.org/pdf/2602.07581v1
+
+## 14470. Enhancing Time Series Classification with Diversity-Driven Neural Network Ensembles
+
+- **ID:** `2602.07579` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Javidan Abdullayev, Maxime Devanne, Cyril Meyer, et al.
+- **abs:** https://arxiv.org/abs/2602.07579 · **PDF:** https://arxiv.org/pdf/2602.07579v1
+
+## 14471. ViCA: Efficient Multimodal LLMs with Vision-Only Cross-Attention
+
+- **ID:** `2602.07574` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Wenjie Liu, Hao Wu, Xin Qiu, et al.
+- **abs:** https://arxiv.org/abs/2602.07574 · **PDF:** https://arxiv.org/pdf/2602.07574v1
+
+## 14472. Graph Domain Adaptation via Homophily-Agnostic Reconstructing Structure
+
+- **ID:** `2602.07573` · **primary:** `cs.SI` · **Submitted:** 2026-02-07
+- **Authors:** Ruiyi Fang, Shuo Wang, Ruizhi Pu, et al.
+- **abs:** https://arxiv.org/abs/2602.07573 · **PDF:** https://arxiv.org/pdf/2602.07573v1
+
+## 14473. How does longer temporal context enhance multimodal narrative video processing in the brain?
+
+- **ID:** `2602.07570` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-07
+- **Authors:** Prachi Jindal, Anant Khandelwal, Manish Gupta, et al.
+- **abs:** https://arxiv.org/abs/2602.07570 · **PDF:** https://arxiv.org/pdf/2602.07570v1
+
+## 14474. Cross-Camera Cow Identification via Disentangled Representation Learning
+
+- **ID:** `2602.07566` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Runcheng Wang, Yaru Chen, Guiguo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.07566 · **PDF:** https://arxiv.org/pdf/2602.07566v1
+
+## 14475. Gaussian Match-and-Copy: A Minimalist Benchmark for Studying Transformer Induction
+
+- **ID:** `2602.07562` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Antoine Gonon, Alexandre Cordonnier, Nicolas Boumal
+- **abs:** https://arxiv.org/abs/2602.07562 · **PDF:** https://arxiv.org/pdf/2602.07562v1
+
+## 14476. VERIFY-RL: Verifiable Recursive Decomposition for Reinforcement Learning in Mathematical Reasoning
+
+- **ID:** `2602.07559` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Kaleem Ullah Qasim, Jiashu Zhang, Hao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07559 · **PDF:** https://arxiv.org/pdf/2602.07559v1
+
+## 14477. VISOR: VIsual Spatial Object Reasoning for Language-driven Object Navigation
+
+- **ID:** `2602.07555` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Francesco Taioli, Shiping Yang, Sonia Raychaudhuri, et al.
+- **abs:** https://arxiv.org/abs/2602.07555 · **PDF:** https://arxiv.org/pdf/2602.07555v1
+
+## 14478. Revealing the Semantic Selection Gap in DINOv3 through Training-Free Few-Shot Segmentation
+
+- **ID:** `2602.07550` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Hussni Mohd Zakir, Eric Tatt Wei Ho
+- **abs:** https://arxiv.org/abs/2602.07550 · **PDF:** https://arxiv.org/pdf/2602.07550v1
+
+## 14479. When Is Enough Not Enough? Illusory Completion in Search Agents
+
+- **ID:** `2602.07549` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Dayoon Ko, Jihyuk Kim, Sohyeon Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.07549 · **PDF:** https://arxiv.org/pdf/2602.07549v1
+
+## 14480. Capturing the Topological Phase Transition and Thermodynamics of the 2D XY Model via Manifold-Aware Score-Based Generative Modeling
+
+- **ID:** `2602.07548` · **primary:** `cond-mat.stat-mech` · **Submitted:** 2026-02-07
+- **Authors:** Pratyush Jha
+- **abs:** https://arxiv.org/abs/2602.07548 · **PDF:** https://arxiv.org/pdf/2602.07548v1
+
+## 14481. Linguistic properties and model scale in brain encoding: from small to compressed language models
+
+- **ID:** `2602.07547` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-07
+- **Authors:** Subba Reddy Oota, Vijay Rowtula, Satya Sai Srinath Namburi, et al.
+- **abs:** https://arxiv.org/abs/2602.07547 · **PDF:** https://arxiv.org/pdf/2602.07547v1
+
+## 14482. Improving Variable-Length Generation in Diffusion Language Models via Length Regularization
+
+- **ID:** `2602.07546` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Zicong Cheng, Ruixuan Jia, Jia Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07546 · **PDF:** https://arxiv.org/pdf/2602.07546v1
+
+## 14483. Explanatory Interactive Machine Learning for Bias Mitigation in Visual Gender Classification
+
+- **ID:** `2602.13286` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Nathanya Satriani, Djordje Slijepčević, Markus Schedl, et al.
+- **abs:** https://arxiv.org/abs/2602.13286 · **PDF:** https://arxiv.org/pdf/2602.13286v1
+
+## 14484. MSP-LLM: A Unified Large Language Model Framework for Complete Material Synthesis Planning
+
+- **ID:** `2602.07543` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Heewoong Noh, Gyoung S. Na, Namkyeong Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.07543 · **PDF:** https://arxiv.org/pdf/2602.07543v3
+
+## 14485. LLM-Guided Diagnostic Evidence Alignment for Medical Vision-Language Pretraining under Limited Pairing
+
+- **ID:** `2602.07540` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Huimin Yan, Liang Bai, Xian Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.07540 · **PDF:** https://arxiv.org/pdf/2602.07540v1
+
+## 14486. Training-Driven Representational Geometry Modularization Predicts Brain Alignment in Language Models
+
+- **ID:** `2602.07539` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-07
+- **Authors:** Yixuan Liu, Zhiyuan Ma, Likai Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.07539 · **PDF:** https://arxiv.org/pdf/2602.07539v1
+
+## 14487. Beyond Core and Penumbra: Bi-Temporal Image-Driven Stroke Evolution Analysis
+
+- **ID:** `2602.07535` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Md Sazidur Rahman, Kjersti Engan, Kathinka Dæhli Kurz, et al.
+- **abs:** https://arxiv.org/abs/2602.07535 · **PDF:** https://arxiv.org/pdf/2602.07535v1
+
+## 14488. Fine-Grained Cat Breed Recognition with Global Context Vision Transformer
+
+- **ID:** `2602.07534` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Mowmita Parvin Hera, Md. Shahriar Mahmud Kallol, Shohanur Rahman Nirob, et al.
+- **abs:** https://arxiv.org/abs/2602.07534 · **PDF:** https://arxiv.org/pdf/2602.07534v1
+
+## 14489. Joint Reward Modeling: Internalizing Chain-of-Thought for Efficient Visual Reward Models
+
+- **ID:** `2602.07533` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Yankai Yang, Yancheng Long, Hongyang Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.07533 · **PDF:** https://arxiv.org/pdf/2602.07533v1
+
+## 14490. Evaluating Object-Centric Models beyond Object Discovery
+
+- **ID:** `2602.07532` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Krishnakant Singh, Simone Schaub-Meyer, Stefan Roth
+- **abs:** https://arxiv.org/abs/2602.07532 · **PDF:** https://arxiv.org/pdf/2602.07532v1
+
+## 14491. Compact Conformal Subgraphs
+
+- **ID:** `2602.07530` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Sreenivas Gollapudi, Kostas Kollias, Kamesh Munagala, et al.
+- **abs:** https://arxiv.org/abs/2602.07530 · **PDF:** https://arxiv.org/pdf/2602.07530v2
+
+## 14492. MedVerse: Efficient and Reliable Medical Reasoning via DAG-Structured Parallel Execution
+
+- **ID:** `2602.07529` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Jianwen Chen, Xinyu Yang, Peng Xia, et al.
+- **abs:** https://arxiv.org/abs/2602.07529 · **PDF:** https://arxiv.org/pdf/2602.07529v2
+
+## 14493. Pareto-guided Pipeline for Distilling Featherweight AI Agents in Mobile MOBA Games
+
+- **ID:** `2602.07521` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Xionghui Yang, Bozhou Chen, Yunlong Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.07521 · **PDF:** https://arxiv.org/pdf/2602.07521v1
+
+## 14494. MDL: A Unified Multi-Distribution Learner in Large-scale Industrial Recommendation through Tokenization
+
+- **ID:** `2602.07520` · **primary:** `cs.IR` · **Submitted:** 2026-02-07
+- **Authors:** Shanlei Mu, Yuchen Jiang, Shikang Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.07520 · **PDF:** https://arxiv.org/pdf/2602.07520v2
+
+## 14495. PALMS: Pavlovian Associative Learning Models Simulator
+
+- **ID:** `2602.07519` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Martin Fixman, Alessandro Abati, Julián Jiménez Nimmo, et al.
+- **abs:** https://arxiv.org/abs/2602.07519 · **PDF:** https://arxiv.org/pdf/2602.07519v2
+
+## 14496. Physical Analog Kolmogorov-Arnold Networks based on Reconfigurable Nonlinear-Processing Units
+
+- **ID:** `2602.07518` · **primary:** `cs.ET` · **Submitted:** 2026-02-07
+- **Authors:** Manuel Escudero, Mohamadreza Zolfagharinejad, Sjoerd van den Belt, et al.
+- **abs:** https://arxiv.org/abs/2602.07518 · **PDF:** https://arxiv.org/pdf/2602.07518v1
+
+## 14497. MemPot: Defending Against Memory Extraction Attack with Optimized Honeypots
+
+- **ID:** `2602.07517` · **primary:** `cs.CR` · **Submitted:** 2026-02-07
+- **Authors:** Yuhao Wang, Shengfang Zhai, Guanghao Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.07517 · **PDF:** https://arxiv.org/pdf/2602.07517v1
+
+## 14498. VividFace: Real-Time and Realistic Facial Expression Shadowing for Humanoid Robots
+
+- **ID:** `2602.07506` · **primary:** `cs.RO` · **Submitted:** 2026-02-07
+- **Authors:** Peizhen Li, Longbing Cao, Xiao-Ming Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.07506 · **PDF:** https://arxiv.org/pdf/2602.07506v2
+
+## 14499. Let's Simplify Step by Step: Guiding LLM Towards Multilingual Unsupervised Proficiency-Controlled Sentence Simplification
+
+- **ID:** `2602.07499` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Jingshen Zhang, Xin Ying Qiu, Lifang Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.07499 · **PDF:** https://arxiv.org/pdf/2602.07499v1
+
+## 14500. From Native Memes to Global Moderation: Cross-Cultural Evaluation of Vision-Language Models for Hateful Meme Detection
+
+- **ID:** `2602.07497` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Mo Wang, Kaixuan Ren, Pratik Jalan, et al.
+- **abs:** https://arxiv.org/abs/2602.07497 · **PDF:** https://arxiv.org/pdf/2602.07497v2
+
+## 14501. CoMI-IRL: Contrastive Multi-Intention Inverse Reinforcement Learning
+
+- **ID:** `2602.07496` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Antonio Mone, Frans A. Oliehoek, Luciano Cavalcante Siebert
+- **abs:** https://arxiv.org/abs/2602.07496 · **PDF:** https://arxiv.org/pdf/2602.07496v1
+
+## 14502. Hyperparameter Transfer Laws for Non-Recurrent Multi-Path Neural Networks
+
+- **ID:** `2602.07494` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Shenxi Wu, Haosong Zhang, Xingjian Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.07494 · **PDF:** https://arxiv.org/pdf/2602.07494v1
+
+## 14503. GraphAgents: Knowledge Graph-Guided Agentic AI for Cross-Domain Materials Design
+
+- **ID:** `2602.07491` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Isabella A. Stewart, Tarjei Paule Hage, Yu-Chuan Hsu, et al.
+- **abs:** https://arxiv.org/abs/2602.07491 · **PDF:** https://arxiv.org/pdf/2602.07491v1
+
+## 14504. Deriving Neural Scaling Laws from the statistics of natural language
+
+- **ID:** `2602.07488` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Francesco Cagnetta, Allan Raventós, Surya Ganguli, et al.
+- **abs:** https://arxiv.org/abs/2602.07488 · **PDF:** https://arxiv.org/pdf/2602.07488v2
+
+## 14505. ODELoRA: Training Low-Rank Adaptation by Solving Ordinary Differential Equations
+
+- **ID:** `2602.07479` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Yihang Gao, Vincent Y. F. Tan
+- **abs:** https://arxiv.org/abs/2602.07479 · **PDF:** https://arxiv.org/pdf/2602.07479v1
+
+## 14506. AI-Driven Predictive Modelling for Groundwater Salinization in Israel
+
+- **ID:** `2602.07478` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Laxmi Pandey, Ariel Meroz, Ben Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.07478 · **PDF:** https://arxiv.org/pdf/2602.07478v1
+
+## 14507. From Exact Hits to Close Enough: Semantic Caching for LLM Embeddings
+
+- **ID:** `2603.03301` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Dvir David Biton, Roy Friedman
+- **abs:** https://arxiv.org/abs/2603.03301 · **PDF:** https://arxiv.org/pdf/2603.03301v1
+
+## 14508. Statistical inference after variable selection in Cox models: A simulation study
+
+- **ID:** `2602.07477` · **primary:** `stat.ME` · **Submitted:** 2026-02-07
+- **Authors:** Lena Schemet, Sarah Friedrich-Welz
+- **abs:** https://arxiv.org/abs/2602.07477 · **PDF:** https://arxiv.org/pdf/2602.07477v1
+
+## 14509. Bipartite Graph Attention-based Clustering for Large-scale scRNA-seq Data
+
+- **ID:** `2602.07475` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Zhuomin Liang, Liang Bai, Xian Yang
+- **abs:** https://arxiv.org/abs/2602.07475 · **PDF:** https://arxiv.org/pdf/2602.07475v1
+
+## 14510. Computing the Reachability Value of Posterior-Deterministic POMDPs
+
+- **ID:** `2602.07473` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Nathanaël Fijalkow, Arka Ghosh, Roman Kniazev, et al.
+- **abs:** https://arxiv.org/abs/2602.07473 · **PDF:** https://arxiv.org/pdf/2602.07473v1
+
+## 14511. Bandit Allocational Instability
+
+- **ID:** `2602.07472` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Yilun Chen, Jiaqi Lu
+- **abs:** https://arxiv.org/abs/2602.07472 · **PDF:** https://arxiv.org/pdf/2602.07472v1
+
+## 14512. Are Reasoning LLMs Robust to Interventions on Their Chain-of-Thought?
+
+- **ID:** `2602.07470` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Alexander von Recum, Leander Girrbach, Zeynep Akata
+- **abs:** https://arxiv.org/abs/2602.07470 · **PDF:** https://arxiv.org/pdf/2602.07470v1
+
+## 14513. Learned Finite Element-based Regularization of the Inverse Problem in Electrocardiographic Imaging
+
+- **ID:** `2602.07466` · **primary:** `math.NA` · **Submitted:** 2026-02-07
+- **Authors:** Manuel Haas, Thomas Grandits, Thomas Pinetz, et al.
+- **abs:** https://arxiv.org/abs/2602.07466 · **PDF:** https://arxiv.org/pdf/2602.07466v2
+
+## 14514. On the Importance of a Multi-Scale Calibration for Quantization
+
+- **ID:** `2602.07465` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Seungwoo Son, Ingyu Seong, Junhan Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.07465 · **PDF:** https://arxiv.org/pdf/2602.07465v1
+
+## 14515. SED-SFT: Selectively Encouraging Diversity in Supervised Fine-Tuning
+
+- **ID:** `2602.07464` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Yijie Chen, Yijin Liu, Fandong Meng
+- **abs:** https://arxiv.org/abs/2602.07464 · **PDF:** https://arxiv.org/pdf/2602.07464v1
+
+## 14516. Pull Requests as a Training Signal for Repo-Level Code Editing
+
+- **ID:** `2602.07457` · **primary:** `cs.SE` · **Submitted:** 2026-02-07
+- **Authors:** Qinglin Zhu, Tianyu Chen, Shuai Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.07457 · **PDF:** https://arxiv.org/pdf/2602.07457v1
+
+## 14517. Data-Aware and Scalable Sensitivity Analysis for Decision Tree Ensembles
+
+- **ID:** `2602.07453` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Namrita Varshney, Ashutosh Gupta, Arhaan Ahmad, et al.
+- **abs:** https://arxiv.org/abs/2602.07453 · **PDF:** https://arxiv.org/pdf/2602.07453v1
+
+## 14518. DLLM Agent: See Farther, Run Faster
+
+- **ID:** `2602.07451` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Huiling Zhen, Weizhe Lin, Renxi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.07451 · **PDF:** https://arxiv.org/pdf/2602.07451v3
+
+## 14519. Measuring cross-language intelligibility between Romance languages with computational tools
+
+- **ID:** `2602.07447` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Liviu P Dinu, Ana Sabina Uban, Bogdan Iordache, et al.
+- **abs:** https://arxiv.org/abs/2602.07447 · **PDF:** https://arxiv.org/pdf/2602.07447v1
+
+## 14520. Proximal Action Replacement for Behavior Cloning Actor-Critic in Offline Reinforcement Learning
+
+- **ID:** `2602.07441` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Jinzong Dong, Wei Huang, Jianshu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.07441 · **PDF:** https://arxiv.org/pdf/2602.07441v1
+
+## 14521. Active Learning Using Aggregated Acquisition Functions: Accuracy and Sustainability Analysis
+
+- **ID:** `2602.07440` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Cédric Jung, Shirin Salehi, Anke Schmeink
+- **abs:** https://arxiv.org/abs/2602.07440 · **PDF:** https://arxiv.org/pdf/2602.07440v1
+
+## 14522. TextOp: Real-time Interactive Text-Driven Humanoid Robot Motion Generation and Control
+
+- **ID:** `2602.07439` · **primary:** `cs.RO` · **Submitted:** 2026-02-07
+- **Authors:** Weiji Xie, Jiakun Zheng, Jinrui Han, et al.
+- **abs:** https://arxiv.org/abs/2602.07439 · **PDF:** https://arxiv.org/pdf/2602.07439v1
+
+## 14523. Reverse-Engineering Model Editing on Language Models
+
+- **ID:** `2602.10134` · **primary:** `cs.CR` · **Submitted:** 2026-02-07
+- **Authors:** Zhiyu Sun, Minrui Luo, Yu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.10134 · **PDF:** https://arxiv.org/pdf/2602.10134v1
+
+## 14524. Bridging Speech, Emotion, and Motion: a VLM-based Multimodal Edge-deployable Framework for Humanoid Robots
+
+- **ID:** `2602.07434` · **primary:** `cs.RO` · **Submitted:** 2026-02-07
+- **Authors:** Songhua Yang, Xuetao Li, Xuanye Fei, et al.
+- **abs:** https://arxiv.org/abs/2602.07434 · **PDF:** https://arxiv.org/pdf/2602.07434v1
+
+## 14525. Multi-Agent Systems Shape Social Norms for Prosocial Behavior Change
+
+- **ID:** `2602.07433` · **primary:** `cs.HC` · **Submitted:** 2026-02-07
+- **Authors:** Yibin Feng, Tianqi Song, Yugin Tan, et al.
+- **abs:** https://arxiv.org/abs/2602.07433 · **PDF:** https://arxiv.org/pdf/2602.07433v1
+
+## 14526. The Moltbook Illusion: Separating Human Influence from Emergent Behavior in AI Agent Societies
+
+- **ID:** `2602.07432` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Ning Li
+- **abs:** https://arxiv.org/abs/2602.07432 · **PDF:** https://arxiv.org/pdf/2602.07432v2
+
+## 14527. Brep2Shape: Boundary and Shape Representation Alignment via Self-Supervised Transformers
+
+- **ID:** `2602.07429` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Yuanxu Sun, Yuezhou Ma, Haixu Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.07429 · **PDF:** https://arxiv.org/pdf/2602.07429v1
+
+## 14528. A Holistic Framework for Automated Configuration Recommendation for Cloud Service Monitoring
+
+- **ID:** `2603.12268` · **primary:** `cs.DC` · **Submitted:** 2026-02-07
+- **Authors:** Anson Bastos, Shreeya Venneti, Anjaly Parayil, et al.
+- **abs:** https://arxiv.org/abs/2603.12268 · **PDF:** https://arxiv.org/pdf/2603.12268v1
+
+## 14529. Sign-Based Optimizers Are Effective Under Heavy-Tailed Noise
+
+- **ID:** `2602.07425` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Dingzhi Yu, Hongyi Tao, Yuanyu Wan, et al.
+- **abs:** https://arxiv.org/abs/2602.07425 · **PDF:** https://arxiv.org/pdf/2602.07425v1
+
+## 14530. Secure Code Generation via Online Reinforcement Learning with Vulnerability Reward Model
+
+- **ID:** `2602.07422` · **primary:** `cs.CR` · **Submitted:** 2026-02-07
+- **Authors:** Tianyi Wu, Mingzhe Du, Yue Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.07422 · **PDF:** https://arxiv.org/pdf/2602.07422v1
+
+## 14531. Achieving Optimal Static and Dynamic Regret Simultaneously in Bandits with Deterministic Losses
+
+- **ID:** `2602.07418` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Jian Qian, Chen-Yu Wei
+- **abs:** https://arxiv.org/abs/2602.07418 · **PDF:** https://arxiv.org/pdf/2602.07418v2
+
+## 14532. Learning Molecular Chirality via Chiral Determinant Kernels
+
+- **ID:** `2602.07415` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Runhan Shi, Zhicheng Zhang, Letian Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.07415 · **PDF:** https://arxiv.org/pdf/2602.07415v1
+
+## 14533. Can LLMs Truly Embody Human Personality? Analyzing AI and Human Behavior Alignment in Dispute Resolution
+
+- **ID:** `2602.07414` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Deuksin Kwon, Kaleen Shrestha, Bin Han, et al.
+- **abs:** https://arxiv.org/abs/2602.07414 · **PDF:** https://arxiv.org/pdf/2602.07414v1
+
+## 14534. AngelSlim: A more accessible, comprehensive, and efficient toolkit for large model compression
+
+- **ID:** `2602.21233` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Rui Cen, QiangQiang Hu, Hong Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.21233 · **PDF:** https://arxiv.org/pdf/2602.21233v3
+
+## 14535. Nonparametric Bayesian Optimization for General Rewards
+
+- **ID:** `2602.07411` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Zishi Zhang, Tao Ren, Yijie Peng
+- **abs:** https://arxiv.org/abs/2602.07411 · **PDF:** https://arxiv.org/pdf/2602.07411v1
+
+## 14536. Progressive Multi-Agent Reasoning for Biological Perturbation Prediction
+
+- **ID:** `2602.07408` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Hyomin Kim, Sang-Yeon Hwang, Jaechang Lim, et al.
+- **abs:** https://arxiv.org/abs/2602.07408 · **PDF:** https://arxiv.org/pdf/2602.07408v1
+
+## 14537. Deep Neural Network Architectures for Electrocardiogram Classification: A Comprehensive Evaluation
+
+- **ID:** `2602.17701` · **primary:** `eess.SP` · **Submitted:** 2026-02-07
+- **Authors:** Yun Song, Wenjia Zheng, Tiedan Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.17701 · **PDF:** https://arxiv.org/pdf/2602.17701v1
+
+## 14538. BitLogic: Training Framework for Gradient-Based FPGA-Native Neural Networks
+
+- **ID:** `2602.07400` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Simon Bührer, Andreas Plesner, Aczel Till, et al.
+- **abs:** https://arxiv.org/abs/2602.07400 · **PDF:** https://arxiv.org/pdf/2602.07400v1
+
+## 14539. VGAS: Value-Guided Action-Chunk Selection for Few-Shot Vision-Language-Action Adaptation
+
+- **ID:** `2602.07399` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Changhua Xu, Jie Lu, Junyu Xuan, et al.
+- **abs:** https://arxiv.org/abs/2602.07399 · **PDF:** https://arxiv.org/pdf/2602.07399v1
+
+## 14540. Benchmarking Legal RAG: The Promise and Limits of AI Statutory Surveys
+
+- **ID:** `2603.03300` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Mohamed Afane, Emaan Hariri, Derek Ouyang, et al.
+- **abs:** https://arxiv.org/abs/2603.03300 · **PDF:** https://arxiv.org/pdf/2603.03300v1
+
+## 14541. AgentSys: Secure and Dynamic LLM Agents Through Explicit Hierarchical Memory Management
+
+- **ID:** `2602.07398` · **primary:** `cs.CR` · **Submitted:** 2026-02-07
+- **Authors:** Ruoyao Wen, Hao Li, Chaowei Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.07398 · **PDF:** https://arxiv.org/pdf/2602.07398v1
+
+## 14542. Scout Before You Attend: Sketch-and-Walk Sparse Attention for Efficient LLM Inference
+
+- **ID:** `2602.07397` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Hoang Anh Duy Le, Sahil Joshi, Zeyu Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.07397 · **PDF:** https://arxiv.org/pdf/2602.07397v1
+
+## 14543. NAAMSE: Framework for Evolutionary Security Evaluation of Agents
+
+- **ID:** `2602.07391` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Kunal Pai, Parth Shah, Harshil Patel
+- **abs:** https://arxiv.org/abs/2602.07391 · **PDF:** https://arxiv.org/pdf/2602.07391v2
+
+## 14544. Advantages of Domain Knowledge Injection for Legal Document Summarization: A Case Study on Summarizing Indian Court Judgments in English and Hindi
+
+- **ID:** `2602.07382` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Debtanu Datta, Rajdeep Mukherjee, Adrijit Goswami, et al.
+- **abs:** https://arxiv.org/abs/2602.07382 · **PDF:** https://arxiv.org/pdf/2602.07382v1
+
+## 14545. When the Model Said 'No Comment', We Knew Helpfulness Was Dead, Honesty Was Alive, and Safety Was Terrified
+
+- **ID:** `2602.07381` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Gautam Siddharth Kashyap, Mark Dras, Usman Naseem
+- **abs:** https://arxiv.org/abs/2602.07381 · **PDF:** https://arxiv.org/pdf/2602.07381v1
+
+## 14546. Dichotomy of Feature Learning and Unlearning: Fast-Slow Analysis on Neural Networks with Stochastic Gradient Descent
+
+- **ID:** `2602.07378` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Shota Imai, Sota Nishiyama, Masaaki Imaizumi
+- **abs:** https://arxiv.org/abs/2602.07378 · **PDF:** https://arxiv.org/pdf/2602.07378v1
+
+## 14547. Do Large Language Models Reflect Demographic Pluralism in Safety?
+
+- **ID:** `2602.07376` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Usman Naseem, Gautam Siddharth Kashyap, Sushant Kumar Ray, et al.
+- **abs:** https://arxiv.org/abs/2602.07376 · **PDF:** https://arxiv.org/pdf/2602.07376v1
+
+## 14548. Efficient Post-Training Pruning of Large Language Models with Statistical Correction
+
+- **ID:** `2602.07375` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Peiqi Yu, Jinhao Wang, Xinyi Sui, et al.
+- **abs:** https://arxiv.org/abs/2602.07375 · **PDF:** https://arxiv.org/pdf/2602.07375v1
+
+## 14549. TernaryLM: Memory-Efficient Language Modeling via Native 1.5-Bit Quantization with Adaptive Layer-wise Scaling
+
+- **ID:** `2602.07374` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Nisharg Nargund, Priyesh Shukla
+- **abs:** https://arxiv.org/abs/2602.07374 · **PDF:** https://arxiv.org/pdf/2602.07374v2
+
+## 14550. Privately Learning Decision Lists and a Differentially Private Winnow
+
+- **ID:** `2602.07370` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Mark Bun, William Fang
+- **abs:** https://arxiv.org/abs/2602.07370 · **PDF:** https://arxiv.org/pdf/2602.07370v1
+
+## 14551. FEM-Informed Hypergraph Neural Networks for Efficient Elastoplasticity
+
+- **ID:** `2602.07364` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Jianchuan Yang, Xi Chen, Jidong Zhao
+- **abs:** https://arxiv.org/abs/2602.07364 · **PDF:** https://arxiv.org/pdf/2602.07364v1
+
+## 14552. ViHERMES: A Graph-Grounded Multihop Question Answering Benchmark and System for Vietnamese Healthcare Regulations
+
+- **ID:** `2602.07361` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Long S. T. Nguyen, Quan M. Bui, Tin T. Ngo, et al.
+- **abs:** https://arxiv.org/abs/2602.07361 · **PDF:** https://arxiv.org/pdf/2602.07361v1
+
+## 14553. W&D:Scaling Parallel Tool Calling for Efficient Deep Research Agents
+
+- **ID:** `2602.07359` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Xiaoqiang Lin, Jun Hao Liew, Silvio Savarese, et al.
+- **abs:** https://arxiv.org/abs/2602.07359 · **PDF:** https://arxiv.org/pdf/2602.07359v1
+
+## 14554. UTOPIA: Unlearnable Tabular Data via Decoupled Shortcut Embedding
+
+- **ID:** `2602.07358` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Jiaming He, Fuming Luo, Hongwei Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07358 · **PDF:** https://arxiv.org/pdf/2602.07358v2
+
+## 14555. Controllable Value Alignment in Large Language Models through Neuron-Level Editing
+
+- **ID:** `2602.07356` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Yonghui Yang, Junwei Li, Jilong Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.07356 · **PDF:** https://arxiv.org/pdf/2602.07356v1
+
+## 14556. Urban Vibrancy Embedding and Application on Traffic Prediction
+
+- **ID:** `2602.21232` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Sumin Han, Jisun An, Dongman Lee
+- **abs:** https://arxiv.org/abs/2602.21232 · **PDF:** https://arxiv.org/pdf/2602.21232v1
+
+## 14557. AgentTrace: A Structured Logging Framework for Agent System Observability
+
+- **ID:** `2602.10133` · **primary:** `cs.SE` · **Submitted:** 2026-02-07
+- **Authors:** Adam AlSayyad, Kelvin Yuxiang Huang, Richik Pal
+- **abs:** https://arxiv.org/abs/2602.10133 · **PDF:** https://arxiv.org/pdf/2602.10133v1
+
+## 14558. Optimizing Few-Step Generation with Adaptive Matching Distillation
+
+- **ID:** `2602.07345` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Lichen Bai, Zikai Zhou, Shitong Shao, et al.
+- **abs:** https://arxiv.org/abs/2602.07345 · **PDF:** https://arxiv.org/pdf/2602.07345v1
+
+## 14559. Seeing Roads Through Words: A Language-Guided Framework for RGB-T Driving Scene Segmentation
+
+- **ID:** `2602.07343` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Ruturaj Reddy, Hrishav Bakul Barua, Junn Yong Loo, et al.
+- **abs:** https://arxiv.org/abs/2602.07343 · **PDF:** https://arxiv.org/pdf/2602.07343v1
+
+## 14560. SupChain-Bench: Benchmarking Large Language Models for Real-World Supply Chain Management
+
+- **ID:** `2602.07342` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Shengyue Guan, Yihao Liu, Lang Cao
+- **abs:** https://arxiv.org/abs/2602.07342 · **PDF:** https://arxiv.org/pdf/2602.07342v1
+
+## 14561. Scalable Dexterous Robot Learning with AR-based Remote Human-Robot Interactions
+
+- **ID:** `2602.07341` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Yicheng Yang, Ruijiao Li, Lifeng Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.07341 · **PDF:** https://arxiv.org/pdf/2602.07341v1
+
+## 14562. Revisiting Robustness for LLM Safety Alignment via Selective Geometry Control
+
+- **ID:** `2602.07340` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Yonghui Yang, Wenjian Tao, Jilong Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.07340 · **PDF:** https://arxiv.org/pdf/2602.07340v1
+
+## 14563. RAPiD: Real-time Deterministic Trajectory Planning via Diffusion Behavior Priors for Safe and Efficient Autonomous Driving
+
+- **ID:** `2602.07339` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Ruturaj Reddy, Hrishav Bakul Barua, Junn Yong Loo, et al.
+- **abs:** https://arxiv.org/abs/2602.07339 · **PDF:** https://arxiv.org/pdf/2602.07339v1
+
+## 14564. Intent Mismatch Causes LLMs to Get Lost in Multi-Turn Conversation
+
+- **ID:** `2602.07338` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Geng Liu, Fei Zhu, Rong Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.07338 · **PDF:** https://arxiv.org/pdf/2602.07338v1
+
+## 14565. High Fidelity Textual User Representation over Heterogeneous Sources via Reinforcement Learning
+
+- **ID:** `2602.07333` · **primary:** `cs.IR` · **Submitted:** 2026-02-07
+- **Authors:** Rajat Arora, Ye Tao, Jianqiang Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.07333 · **PDF:** https://arxiv.org/pdf/2602.07333v1
+
+## 14566. Action-to-Action Flow Matching
+
+- **ID:** `2602.07322` · **primary:** `cs.RO` · **Submitted:** 2026-02-07
+- **Authors:** Jindou Jia, Gen Li, Xiangyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.07322 · **PDF:** https://arxiv.org/pdf/2602.07322v1
+
+## 14567. Incorruptible Neural Networks: Training Models that can Generalize to Large Internal Perturbations
+
+- **ID:** `2602.07320` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Philip Jacobson, Ben Feinberg, Suhas Kumar, et al.
+- **abs:** https://arxiv.org/abs/2602.07320 · **PDF:** https://arxiv.org/pdf/2602.07320v1
+
+## 14568. Beyond Accuracy: Risk-Sensitive Evaluation of Hallucinated Medical Advice
+
+- **ID:** `2602.07319` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** Savan Doshi
+- **abs:** https://arxiv.org/abs/2602.07319 · **PDF:** https://arxiv.org/pdf/2602.07319v2
+
+## 14569. LUCID-SAE: Learning Unified Vision-Language Sparse Codes for Interpretable Concept Discovery
+
+- **ID:** `2602.07311` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Difei Gu, Yunhe Gao, Gerasimos Chatzoudis, et al.
+- **abs:** https://arxiv.org/abs/2602.07311 · **PDF:** https://arxiv.org/pdf/2602.07311v1
+
+## 14570. Optimization of Precipitate Segmentation Through Linear Genetic Programming of Image Processing
+
+- **ID:** `2602.07310` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Kyle Williams, Andrew Seltzman
+- **abs:** https://arxiv.org/abs/2602.07310 · **PDF:** https://arxiv.org/pdf/2602.07310v1
+
+## 14571. Semantic Search At LinkedIn
+
+- **ID:** `2602.07309` · **primary:** `cs.IR` · **Submitted:** 2026-02-07
+- **Authors:** Fedor Borisyuk, Sriram Vasudevan, Muchen Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.07309 · **PDF:** https://arxiv.org/pdf/2602.07309v1
+
+## 14572. Adaptive Scaffolding for Cognitive Engagement in an Intelligent Tutoring System
+
+- **ID:** `2602.07308` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Sutapa Dey Tithi, Nazia Alam, Tahreem Yasir, et al.
+- **abs:** https://arxiv.org/abs/2602.07308 · **PDF:** https://arxiv.org/pdf/2602.07308v1
+
+## 14573. LIT-GRAPH: Evaluating Deep vs. Shallow Graph Embeddings for High-Quality Text Recommendation in Domain-Specific Knowledge Graphs
+
+- **ID:** `2602.07307` · **primary:** `cs.IR` · **Submitted:** 2026-02-07
+- **Authors:** Nirmal Gelal, Chloe Snow, Kathleen M. Jagodnik, et al.
+- **abs:** https://arxiv.org/abs/2602.07307 · **PDF:** https://arxiv.org/pdf/2602.07307v1
+
+## 14574. Parallel Track Transformers: Enabling Fast GPU Inference with Reduced Synchronization
+
+- **ID:** `2602.07306` · **primary:** `cs.DC` · **Submitted:** 2026-02-07
+- **Authors:** Chong Wang, Nan Du, Tom Gunter, et al.
+- **abs:** https://arxiv.org/abs/2602.07306 · **PDF:** https://arxiv.org/pdf/2602.07306v1
+
+## 14575. KRONE: Hierarchical and Modular Log Anomaly Detection
+
+- **ID:** `2602.07303` · **primary:** `cs.DB` · **Submitted:** 2026-02-07
+- **Authors:** Lei Ma, Jinyang Liu, Tieying Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.07303 · **PDF:** https://arxiv.org/pdf/2602.07303v2
+
+## 14576. Optimizing Chlorination in Water Distribution Systems via Surrogate-assisted Neuroevolution
+
+- **ID:** `2602.07299` · **primary:** `cs.NE` · **Submitted:** 2026-02-07
+- **Authors:** Rivaaj Monsia, Daniel Young, Olivier Francon, et al.
+- **abs:** https://arxiv.org/abs/2602.07299 · **PDF:** https://arxiv.org/pdf/2602.07299v1
+
+## 14577. Principled Synthetic Data Enables the First Scaling Laws for LLMs in Recommendation
+
+- **ID:** `2602.07298` · **primary:** `cs.IR` · **Submitted:** 2026-02-07
+- **Authors:** Benyu Zhang, Qiang Zhang, Jianpeng Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.07298 · **PDF:** https://arxiv.org/pdf/2602.07298v2
+
+## 14578. Progressive Searching for Retrieval in RAG
+
+- **ID:** `2602.07297` · **primary:** `cs.IR` · **Submitted:** 2026-02-07
+- **Authors:** Taehee Jeong, Xingzhe Zhao, Peizu Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07297 · **PDF:** https://arxiv.org/pdf/2602.07297v1
+
+## 14579. Surrogate Modeling for Neutron Transport: A Neural Operator Approach
+
+- **ID:** `2602.15890` · **primary:** `physics.comp-ph` · **Submitted:** 2026-02-07
+- **Authors:** Md Hossain Sahadath, Qiyun Cheng, Shaowu Pan, et al.
+- **abs:** https://arxiv.org/abs/2602.15890 · **PDF:** https://arxiv.org/pdf/2602.15890v1
+
+## 14580. Fin-RATE: A Real-world Financial Analytics and Tracking Evaluation Benchmark for LLMs on SEC Filings
+
+- **ID:** `2602.07294` · **primary:** `cs.CE` · **Submitted:** 2026-02-07
+- **Authors:** Yidong Jiang, Junrong Chen, Eftychia Makri, et al.
+- **abs:** https://arxiv.org/abs/2602.07294 · **PDF:** https://arxiv.org/pdf/2602.07294v3
+
+## 14581. Agents in the Wild: Safety, Society, and the Illusion of Sociality on Moltbook
+
+- **ID:** `2602.13284` · **primary:** `cs.SI` · **Submitted:** 2026-02-07
+- **Authors:** Yunbei Zhang, Kai Mei, Ming Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.13284 · **PDF:** https://arxiv.org/pdf/2602.13284v1
+
+## 14582. Fair Decisions from Calibrated Scores: Achieving Optimal Classification While Satisfying Sufficiency
+
+- **ID:** `2602.07285` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Etam Benger, Katrina Ligett
+- **abs:** https://arxiv.org/abs/2602.07285 · **PDF:** https://arxiv.org/pdf/2602.07285v1
+
+## 14583. Imagining the Alien: Human Projections and Cognitive Limitations
+
+- **ID:** `2602.07284` · **primary:** `astro-ph.IM` · **Submitted:** 2026-02-07
+- **Authors:** S. G. Djorgovski
+- **abs:** https://arxiv.org/abs/2602.07284 · **PDF:** https://arxiv.org/pdf/2602.07284v1
+
+## 14584. How LLMs Cite and Why It Matters: A Cross-Model Audit of Reference Fabrication in AI-Assisted Academic Writing and Methods to Detect Phantom Citations
+
+- **ID:** `2603.03299` · **primary:** `cs.CL` · **Submitted:** 2026-02-07
+- **Authors:** MZ Naser
+- **abs:** https://arxiv.org/abs/2603.03299 · **PDF:** https://arxiv.org/pdf/2603.03299v1
+
+## 14585. VertCoHiRF: Decentralized Vertical Clustering Beyond k-means
+
+- **ID:** `2602.07279` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Bruno Belucci, Karim Lounici, Vladimir R. Kostic, et al.
+- **abs:** https://arxiv.org/abs/2602.07279 · **PDF:** https://arxiv.org/pdf/2602.07279v1
+
+## 14586. Laplacian-LoRA: Delaying Oversmoothing in Deep GCNs via Spectral Low-Rank Adaptation
+
+- **ID:** `2602.07278` · **primary:** `cs.LG` · **Submitted:** 2026-02-07
+- **Authors:** Sai Vamsi Alisetti
+- **abs:** https://arxiv.org/abs/2602.07278 · **PDF:** https://arxiv.org/pdf/2602.07278v1
+
+## 14587. Cross-View World Models
+
+- **ID:** `2602.07277` · **primary:** `cs.CV` · **Submitted:** 2026-02-07
+- **Authors:** Rishabh Sharma, Gijs Hogervorst, Wayne E. Mackey, et al.
+- **abs:** https://arxiv.org/abs/2602.07277 · **PDF:** https://arxiv.org/pdf/2602.07277v1
+
+## 14588. Steer2Adapt: Dynamically Composing Steering Vectors Elicits Efficient Adaptation of LLMs
+
+- **ID:** `2602.07276` · **primary:** `cs.AI` · **Submitted:** 2026-02-07
+- **Authors:** Pengrui Han, Xueqiang Xu, Keyang Xuan, et al.
+- **abs:** https://arxiv.org/abs/2602.07276 · **PDF:** https://arxiv.org/pdf/2602.07276v1
+
+## 14589. TermiGen: High-Fidelity Environment and Robust Trajectory Synthesis for Terminal Agents
+
+- **ID:** `2602.07274` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Kaijie Zhu, Yuzhou Nie, Yijiang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07274 · **PDF:** https://arxiv.org/pdf/2602.07274v1
+
+## 14590. Hybrid Feedback-Guided Optimal Learning for Wireless Interactive Panoramic Scene Delivery
+
+- **ID:** `2602.07273` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Xiaoyi Wu, Juaren Steiger, Bin Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07273 · **PDF:** https://arxiv.org/pdf/2602.07273v1
+
+## 14591. BRIDGE: Predicting Human Task Completion Time From Model Performance
+
+- **ID:** `2602.07267` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Fengyuan Liu, Jay Gala, Nilaksh, et al.
+- **abs:** https://arxiv.org/abs/2602.07267 · **PDF:** https://arxiv.org/pdf/2602.07267v1
+
+## 14592. XShare: Collaborative in-Batch Expert Sharing for Faster MoE Inference
+
+- **ID:** `2602.07265` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Daniil Vankov, Nikita Ivkin, Kyle Ulrich, et al.
+- **abs:** https://arxiv.org/abs/2602.07265 · **PDF:** https://arxiv.org/pdf/2602.07265v1
+
+## 14593. Revisiting the Seasonal Trend Decomposition for Enhanced Time Series Forecasting
+
+- **ID:** `2602.18465` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Sanjeev Panta, Xu Yuan, Li Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.18465 · **PDF:** https://arxiv.org/pdf/2602.18465v1
+
+## 14594. aerial-autonomy-stack -- a Faster-than-real-time, Autopilot-agnostic, ROS2 Framework to Simulate and Deploy Perception-based Drones
+
+- **ID:** `2602.07264` · **primary:** `cs.RO` · **Submitted:** 2026-02-06
+- **Authors:** Jacopo Panerati, Sina Sajjadi, Sina Soleymanpour, et al.
+- **abs:** https://arxiv.org/abs/2602.07264 · **PDF:** https://arxiv.org/pdf/2602.07264v1
+
+## 14595. ACAR: Adaptive Complexity Routing for Multi-Model Ensembles with Auditable Decision Traces
+
+- **ID:** `2602.21231` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Ramchand Kumaresan
+- **abs:** https://arxiv.org/abs/2602.21231 · **PDF:** https://arxiv.org/pdf/2602.21231v1
+
+## 14596. tLoRA: Efficient Multi-LoRA Training with Elastic Shared Super-Models
+
+- **ID:** `2602.07263` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Kevin Li, Dibyadeep Saha, Avni Kanodia, et al.
+- **abs:** https://arxiv.org/abs/2602.07263 · **PDF:** https://arxiv.org/pdf/2602.07263v2
+
+## 14597. Cognitive algorithms and systems of episodic memory, semantic memory and their learnings
+
+- **ID:** `2602.07261` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-06
+- **Authors:** Qi Zhang
+- **abs:** https://arxiv.org/abs/2602.07261 · **PDF:** https://arxiv.org/pdf/2602.07261v1
+
+## 14598. 3D Transport-based Morphometry (3D-TBM) for medical image analysis
+
+- **ID:** `2602.07260` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Hongyu Kan, Kristofor Pas, Ivan Medri, et al.
+- **abs:** https://arxiv.org/abs/2602.07260 · **PDF:** https://arxiv.org/pdf/2602.07260v1
+
+## 14599. Incentive-Aware AI Safety via Strategic Resource Allocation: A Stackelberg Security Games Perspective
+
+- **ID:** `2602.07259` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Cheol Woo Kim, Davin Choo, Tzeh Yuan Neoh, et al.
+- **abs:** https://arxiv.org/abs/2602.07259 · **PDF:** https://arxiv.org/pdf/2602.07259v1
+
+## 14600. Robust Ultra-High-Dimensional Variable Selection With Correlated Structure Using Group Testing
+
+- **ID:** `2602.07258` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Wanru Guo, Juan Xie, Binbin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.07258 · **PDF:** https://arxiv.org/pdf/2602.07258v1
+
+## 14601. MIDAS: Mosaic Input-Specific Differentiable Architecture Search
+
+- **ID:** `2602.17700` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Konstanty Subbotko
+- **abs:** https://arxiv.org/abs/2602.17700 · **PDF:** https://arxiv.org/pdf/2602.17700v1
+
+## 14602. Graph homophily booster: Reimagining the role of discrete features in heterophilic graph learning
+
+- **ID:** `2602.07256` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Ruizhong Qiu, Ting-Wei Li, Gaotang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07256 · **PDF:** https://arxiv.org/pdf/2602.07256v1
+
+## 14603. From Out-of-Distribution Detection to Hallucination Detection: A Geometric View
+
+- **ID:** `2602.07253` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Litian Liu, Reza Pourreza, Yubing Jian, et al.
+- **abs:** https://arxiv.org/abs/2602.07253 · **PDF:** https://arxiv.org/pdf/2602.07253v1
+
+## 14604. The Double-Edged Sword of Data-Driven Super-Resolution: Adversarial Super-Resolution Models
+
+- **ID:** `2602.07251` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Haley Duba-Sullivan, Steven R. Young, Emma J. Reid
+- **abs:** https://arxiv.org/abs/2602.07251 · **PDF:** https://arxiv.org/pdf/2602.07251v1
+
+## 14605. Beyond Crash: Hijacking Your Autonomous Vehicle for Fun and Profit
+
+- **ID:** `2602.07249` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Qi Sun, Ahmed Abdo, Luis Burbano, et al.
+- **abs:** https://arxiv.org/abs/2602.07249 · **PDF:** https://arxiv.org/pdf/2602.07249v1
+
+## 14606. Realistic Synthetic Household Data Generation at Scale
+
+- **ID:** `2602.07243` · **primary:** `cs.RO` · **Submitted:** 2026-02-06
+- **Authors:** Siddharth Singh, Ifrah Idrees, Abraham Dauhajre
+- **abs:** https://arxiv.org/abs/2602.07243 · **PDF:** https://arxiv.org/pdf/2602.07243v1
+
+## 14607. Is there "Secret Sauce'' in Large Language Model Development?
+
+- **ID:** `2602.07238` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Matthias Mertens, Natalia Fischl-Lanzoni, Neil Thompson
+- **abs:** https://arxiv.org/abs/2602.07238 · **PDF:** https://arxiv.org/pdf/2602.07238v1
+
+## 14608. ArcMark: Multi-bit LLM Watermark via Optimal Transport
+
+- **ID:** `2602.07235` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Atefeh Gilani, Carol Xuan Long, Sajani Vithana, et al.
+- **abs:** https://arxiv.org/abs/2602.07235 · **PDF:** https://arxiv.org/pdf/2602.07235v1
+
+## 14609. Extracting Root-Causal Brain Activity Driving Psychopathology from Resting State fMRI
+
+- **ID:** `2602.07233` · **primary:** `eess.IV` · **Submitted:** 2026-02-06
+- **Authors:** Eric V. Strobl
+- **abs:** https://arxiv.org/abs/2602.07233 · **PDF:** https://arxiv.org/pdf/2602.07233v1
+
+## 14610. Cerebellar-Inspired Residual Control for Fault Recovery: From Inference-Time Adaptation to Structural Consolidation
+
+- **ID:** `2602.07227` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Nethmi Jayasinghe, Diana Gontero, Spencer T. Brown, et al.
+- **abs:** https://arxiv.org/abs/2602.07227 · **PDF:** https://arxiv.org/pdf/2602.07227v1
+
+## 14611. Fault-Tolerant Evaluation for Sample-Efficient Model Performance Estimators
+
+- **ID:** `2602.07226` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Zihan Zhu, Yanqiu Wu, Qiongkai Xu
+- **abs:** https://arxiv.org/abs/2602.07226 · **PDF:** https://arxiv.org/pdf/2602.07226v1
+
+## 14612. SpecAttn: Co-Designing Sparse Attention with Self-Speculative Decoding
+
+- **ID:** `2602.07223` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Yikang Yue, Yuqi Xue, Jian Huang
+- **abs:** https://arxiv.org/abs/2602.07223 · **PDF:** https://arxiv.org/pdf/2602.07223v1
+
+## 14613. The Median is Easier than it Looks: Approximation with a Constant-Depth, Linear-Width ReLU Network
+
+- **ID:** `2602.07219` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Abhigyan Dutta, Itay Safran, Paul Valiant
+- **abs:** https://arxiv.org/abs/2602.07219 · **PDF:** https://arxiv.org/pdf/2602.07219v1
+
+## 14614. Collaborative and Efficient Fine-tuning: Leveraging Task Similarity
+
+- **ID:** `2602.07218` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Gagik Magakyan, Amirhossein Reisizadeh, Chanwoo Park, et al.
+- **abs:** https://arxiv.org/abs/2602.07218 · **PDF:** https://arxiv.org/pdf/2602.07218v1
+
+## 14615. How Well Can LLM Agents Simulate End-User Security and Privacy Attitudes and Behaviors?
+
+- **ID:** `2602.18464` · **primary:** `cs.CY` · **Submitted:** 2026-02-06
+- **Authors:** Yuxuan Li, Leyang Li, Hao-Ping Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.18464 · **PDF:** https://arxiv.org/pdf/2602.18464v2
+
+## 14616. Probing Neural TSP Representations for Prescriptive Decision Support
+
+- **ID:** `2602.07216` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Reuben Narad, Léonard Boussioux, Michael Wagner
+- **abs:** https://arxiv.org/abs/2602.07216 · **PDF:** https://arxiv.org/pdf/2602.07216v1
+
+## 14617. Multi-Agentic AI for Fairness-Aware and Accelerated Multi-modal Large Model Inference in Real-world Mobile Edge Networks
+
+- **ID:** `2602.07215` · **primary:** `eess.SY` · **Submitted:** 2026-02-06
+- **Authors:** Haiyuan Li, Hari Madhukumar, Shuangyi Yan, et al.
+- **abs:** https://arxiv.org/abs/2602.07215 · **PDF:** https://arxiv.org/pdf/2602.07215v1
+
+## 14618. Adaptive Retrieval helps Reasoning in LLMs -- but mostly if it's not used
+
+- **ID:** `2602.07213` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Srijan Shakya, Anamaria-Roberta Hartl, Sepp Hochreiter, et al.
+- **abs:** https://arxiv.org/abs/2602.07213 · **PDF:** https://arxiv.org/pdf/2602.07213v1
+
+## 14619. Equipping LLM with Directional Multi-Talker Speech Understanding Capabilities
+
+- **ID:** `2602.07211` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Ju Lin, Jing Pan, Ruizhi Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07211 · **PDF:** https://arxiv.org/pdf/2602.07211v1
+
+## 14620. Sequences as Nodes for Contrastive Multimodal Graph Recommendation
+
+- **ID:** `2602.07208` · **primary:** `cs.IR` · **Submitted:** 2026-02-06
+- **Authors:** Bucher Sahyouni, Matthew Vowels, Liqun Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.07208 · **PDF:** https://arxiv.org/pdf/2602.07208v1
+
+## 14621. Multimodal Enhancement of Sequential Recommendation
+
+- **ID:** `2602.07207` · **primary:** `cs.IR` · **Submitted:** 2026-02-06
+- **Authors:** Bucher Sahyouni, Matthew Vowels, Liqun Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.07207 · **PDF:** https://arxiv.org/pdf/2602.07207v1
+
+## 14622. DSL: Understanding and Improving Softmax Recommender Systems with Competition-Aware Scaling
+
+- **ID:** `2602.07206` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Bucher Sahyouni, Matthew Vowels, Liqun Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.07206 · **PDF:** https://arxiv.org/pdf/2602.07206v1
+
+## 14623. Online Learning for Uninformed Markov Games: Empirical Nash-Value Regret and Non-Stationarity Adaptation
+
+- **ID:** `2602.07205` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Junyan Liu, Haipeng Luo, Zihan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.07205 · **PDF:** https://arxiv.org/pdf/2602.07205v1
+
+## 14624. Exactly Computing do-Shapley Values
+
+- **ID:** `2602.07203` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** R. Teal Witter, Álvaro Parafita, Tomas Garriga, et al.
+- **abs:** https://arxiv.org/abs/2602.07203 · **PDF:** https://arxiv.org/pdf/2602.07203v1
+
+## 14625. Risk-Sensitive Exponential Actor Critic
+
+- **ID:** `2602.07202` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Alonso Granados, Jason Pacheco
+- **abs:** https://arxiv.org/abs/2602.07202 · **PDF:** https://arxiv.org/pdf/2602.07202v1
+
+## 14626. BadSNN: Backdoor Attacks on Spiking Neural Networks via Adversarial Spiking Neuron
+
+- **ID:** `2602.07200` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Abdullah Arafat Miah, Kevin Vu, Yu Bi
+- **abs:** https://arxiv.org/abs/2602.07200 · **PDF:** https://arxiv.org/pdf/2602.07200v1
+
+## 14627. RuleFlow : Generating Reusable Program Optimizations with LLMs
+
+- **ID:** `2602.09051` · **primary:** `cs.SE` · **Submitted:** 2026-02-06
+- **Authors:** Avaljot Singh, Dushyant Bharadwaj, Stefanos Baziotis, et al.
+- **abs:** https://arxiv.org/abs/2602.09051 · **PDF:** https://arxiv.org/pdf/2602.09051v1
+
+## 14628. Automated Modernization of Machine Learning Engineering Notebooks for Reproducibility
+
+- **ID:** `2602.07195` · **primary:** `cs.SE` · **Submitted:** 2026-02-06
+- **Authors:** Bihui Jin, Kaiyuan Wang, Pengyu Nie
+- **abs:** https://arxiv.org/abs/2602.07195 · **PDF:** https://arxiv.org/pdf/2602.07195v1
+
+## 14629. SAS-Net: Cross-Domain Image Registration as Inverse Rendering via Structure-Appearance Factorization
+
+- **ID:** `2602.09050` · **primary:** `eess.IV` · **Submitted:** 2026-02-06
+- **Authors:** Jiahao Qin
+- **abs:** https://arxiv.org/abs/2602.09050 · **PDF:** https://arxiv.org/pdf/2602.09050v3
+
+## 14630. "Death" of a Chatbot: Investigating and Designing Toward Psychologically Safe Endings for Human-AI Relationships
+
+- **ID:** `2602.07193` · **primary:** `cs.HC` · **Submitted:** 2026-02-06
+- **Authors:** Rachel Poonsiriwong, Chayapatr Archiwaranguprok, Pat Pataranutaporn
+- **abs:** https://arxiv.org/abs/2602.07193 · **PDF:** https://arxiv.org/pdf/2602.07193v2
+
+## 14631. Systematic Performance Assessment of Deep Material Networks for Multiscale Material Modeling
+
+- **ID:** `2602.07192` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Xiaolong He, Haoyan Wei, Wei Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.07192 · **PDF:** https://arxiv.org/pdf/2602.07192v1
+
+## 14632. Long-Context Long-Form Question Answering for Legal Domain
+
+- **ID:** `2602.07190` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Anagha Kulkarni, Parin Rajesh Jhaveri, Prasha Shrestha, et al.
+- **abs:** https://arxiv.org/abs/2602.07190 · **PDF:** https://arxiv.org/pdf/2602.07190v1
+
+## 14633. Latent Target Score Matching, with an application to Simulation-Based Inference
+
+- **ID:** `2602.07189` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Joohwan Ko, Tomas Geffner
+- **abs:** https://arxiv.org/abs/2602.07189 · **PDF:** https://arxiv.org/pdf/2602.07189v1
+
+## 14634. PreFlect: From Retrospective to Prospective Reflection in Large Language Model Agents
+
+- **ID:** `2602.07187` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Hanyu Wang, Yuanpu Cao, Lu Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.07187 · **PDF:** https://arxiv.org/pdf/2602.07187v1
+
+## 14635. The Value of Variance: Mitigating Debate Collapse in Multi-Agent Systems via Uncertainty-Driven Policy Optimization
+
+- **ID:** `2602.07186` · **primary:** `cs.MA` · **Submitted:** 2026-02-06
+- **Authors:** Luoxi Tang, Yuqiao Meng, Joseph Costa, et al.
+- **abs:** https://arxiv.org/abs/2602.07186 · **PDF:** https://arxiv.org/pdf/2602.07186v1
+
+## 14636. Measuring Complexity at the Requirements Stage: Spectral Metrics as Development Effort Predictors
+
+- **ID:** `2602.07182` · **primary:** `cs.SE` · **Submitted:** 2026-02-06
+- **Authors:** Maximilian Vierlboeck, Antonio Pugliese, Roshanak Rose Nilchian, et al.
+- **abs:** https://arxiv.org/abs/2602.07182 · **PDF:** https://arxiv.org/pdf/2602.07182v4
+
+## 14637. Can LLMs Discern the Traits Influencing Your Preferences? Evaluating Personality-Driven Preference Alignment in LLMs
+
+- **ID:** `2602.07181` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Tianyu Zhao, Siqi Li, Yasser Shoukry, et al.
+- **abs:** https://arxiv.org/abs/2602.07181 · **PDF:** https://arxiv.org/pdf/2602.07181v2
+
+## 14638. An Information-Theoretic Framework for Comparing Voice and Text Explainability
+
+- **ID:** `2602.07179` · **primary:** `cs.HC` · **Submitted:** 2026-02-06
+- **Authors:** Mona Rajhans, Vishal Khawarey
+- **abs:** https://arxiv.org/abs/2602.07179 · **PDF:** https://arxiv.org/pdf/2602.07179v1
+
+## 14639. Open TutorAI: An Open-source Platform for Personalized and Immersive Learning with Generative AI
+
+- **ID:** `2602.07176` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Mohamed El Hajji, Tarek Ait Baha, Aicha Dakir, et al.
+- **abs:** https://arxiv.org/abs/2602.07176 · **PDF:** https://arxiv.org/pdf/2602.07176v1
+
+## 14640. Learning Nonlinear Systems In-Context: From Synthetic Data to Real-World Motor Control
+
+- **ID:** `2602.07173` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Tong Jian, Tianyu Dai, Tao Yu
+- **abs:** https://arxiv.org/abs/2602.07173 · **PDF:** https://arxiv.org/pdf/2602.07173v1
+
+## 14641. Your Language Model Secretly Contains Personality Subnetworks
+
+- **ID:** `2602.07164` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Ruimeng Ye, Zihan Wang, Zinan Ling, et al.
+- **abs:** https://arxiv.org/abs/2602.07164 · **PDF:** https://arxiv.org/pdf/2602.07164v1
+
+## 14642. High-fidelity 3D multi-slab diffusion MRI using Slab-shifting for Harmonized 3D Acquisition and Reconstruction with Profile Encoding Networks (SHARPEN)
+
+- **ID:** `2602.07162` · **primary:** `physics.med-ph` · **Submitted:** 2026-02-06
+- **Authors:** Ziyu Li, Karla L. Miller, Wenchuan Wu
+- **abs:** https://arxiv.org/abs/2602.07162 · **PDF:** https://arxiv.org/pdf/2602.07162v1
+
+## 14643. Free Energy Mixer
+
+- **ID:** `2602.07160` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Jiecheng Lu, Shihao Yang
+- **abs:** https://arxiv.org/abs/2602.07160 · **PDF:** https://arxiv.org/pdf/2602.07160v1
+
+## 14644. Mimetic Initialization of MLPs
+
+- **ID:** `2602.07156` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Asher Trockman, J. Zico Kolter
+- **abs:** https://arxiv.org/abs/2602.07156 · **PDF:** https://arxiv.org/pdf/2602.07156v1
+
+## 14645. Beyond Pooling: Matching for Robust Generalization under Data Heterogeneity
+
+- **ID:** `2602.07154` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Ayush Roy, Rudrasis Chakraborty, Lav Varshney, et al.
+- **abs:** https://arxiv.org/abs/2602.07154 · **PDF:** https://arxiv.org/pdf/2602.07154v1
+
+## 14646. ANCHOR: Branch-Point Data Generation for GUI Agents
+
+- **ID:** `2602.07153` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Jinbiao Wei, Yilun Zhao, Kangqi Ni, et al.
+- **abs:** https://arxiv.org/abs/2602.07153 · **PDF:** https://arxiv.org/pdf/2602.07153v1
+
+## 14647. Trojans in Artificial Intelligence (TrojAI) Final Report
+
+- **ID:** `2602.07152` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Kristopher W. Reese, Taylor Kulp-McDowall, Michael Majurski, et al.
+- **abs:** https://arxiv.org/abs/2602.07152 · **PDF:** https://arxiv.org/pdf/2602.07152v2
+
+## 14648. On Randomness in Agentic Evals
+
+- **ID:** `2602.07150` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Bjarni Haukur Bjarnason, André Silva, Martin Monperrus
+- **abs:** https://arxiv.org/abs/2602.07150 · **PDF:** https://arxiv.org/pdf/2602.07150v3
+
+## 14649. Convex Dominance in Deep Learning I: A Scaling Law of Loss and Learning Rate
+
+- **ID:** `2602.07145` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Zhiqi Bu, Shiyun Xu, Jialin Mao
+- **abs:** https://arxiv.org/abs/2602.07145 · **PDF:** https://arxiv.org/pdf/2602.07145v1
+
+## 14650. BONSAI: Bayesian Optimization with Natural Simplicity and Interpretability
+
+- **ID:** `2602.07144` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Samuel Daulton, David Eriksson, Maximilian Balandat, et al.
+- **abs:** https://arxiv.org/abs/2602.07144 · **PDF:** https://arxiv.org/pdf/2602.07144v1
+
+## 14651. Massive Sound Embedding Benchmark (MSEB)
+
+- **ID:** `2602.07143` · **primary:** `cs.SD` · **Submitted:** 2026-02-06
+- **Authors:** Georg Heigold, Ehsan Variani, Tom Bagby, et al.
+- **abs:** https://arxiv.org/abs/2602.07143 · **PDF:** https://arxiv.org/pdf/2602.07143v1
+
+## 14652. Exploring Teachers' Perspectives on Using Conversational AI Agents for Group Collaboration
+
+- **ID:** `2602.07142` · **primary:** `cs.HC` · **Submitted:** 2026-02-06
+- **Authors:** Prerna Ravi, Carúmey Stevens, Beatriz Flamia Azevedo, et al.
+- **abs:** https://arxiv.org/abs/2602.07142 · **PDF:** https://arxiv.org/pdf/2602.07142v1
+
+## 14653. Featured Reproducing Kernel Banach Spaces for Learning and Neural Networks
+
+- **ID:** `2602.07141` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Isabel de la Higuera, Francisco Herrera, M. Victoria Velasco
+- **abs:** https://arxiv.org/abs/2602.07141 · **PDF:** https://arxiv.org/pdf/2602.07141v1
+
+## 14654. TATRA: Training-Free Instance-Adaptive Prompting Through Rephrasing and Aggregation
+
+- **ID:** `2603.03298` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Bartosz Dziuba, Kacper Kuchta, Paweł Batorski, et al.
+- **abs:** https://arxiv.org/abs/2603.03298 · **PDF:** https://arxiv.org/pdf/2603.03298v1
+
+## 14655. Landscaper: Understanding Loss Landscapes Through Multi-Dimensional Topological Analysis
+
+- **ID:** `2602.07135` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Jiaqing Chen, Nicholas Hadler, Tiankai Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.07135 · **PDF:** https://arxiv.org/pdf/2602.07135v3
+
+## 14656. Accuracy Standards for AI at Work vs. Personal Life: Evidence from an Online Survey
+
+- **ID:** `2602.13283` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Gaston Besanson, Federico Todeschini
+- **abs:** https://arxiv.org/abs/2602.13283 · **PDF:** https://arxiv.org/pdf/2602.13283v1
+
+## 14657. Discrete Adjoint Matching
+
+- **ID:** `2602.07132` · **primary:** `stat.ML` · **Submitted:** 2026-02-06
+- **Authors:** Oswin So, Brian Karrer, Chuchu Fan, et al.
+- **abs:** https://arxiv.org/abs/2602.07132 · **PDF:** https://arxiv.org/pdf/2602.07132v2
+
+## 14658. Certified Learning under Distribution Shift: Sound Verification and Identifiable Structure
+
+- **ID:** `2602.17699` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Chandrasekhar Gokavarapu, Sudhakar Gadde, Y. Rajasekhar, et al.
+- **abs:** https://arxiv.org/abs/2602.17699 · **PDF:** https://arxiv.org/pdf/2602.17699v1
+
+## 14659. Finding Connections: Membership Inference Attacks for the Multi-Table Synthetic Data Setting
+
+- **ID:** `2602.07126` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Joshua Ward, Chi-Hua Wang, Guang Cheng
+- **abs:** https://arxiv.org/abs/2602.07126 · **PDF:** https://arxiv.org/pdf/2602.07126v1
+
+## 14660. Reasoning-Augmented Representations for Multimodal Retrieval
+
+- **ID:** `2602.07125` · **primary:** `cs.IR` · **Submitted:** 2026-02-06
+- **Authors:** Jianrui Zhang, Anirudh Sundara Rajan, Brandon Han, et al.
+- **abs:** https://arxiv.org/abs/2602.07125 · **PDF:** https://arxiv.org/pdf/2602.07125v1
+
+## 14661. Anchored Decoding: Provably Reducing Copyright Risk for Any Language Model
+
+- **ID:** `2602.07120` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Jacqueline He, Jonathan Hayase, Wen-tau Yih, et al.
+- **abs:** https://arxiv.org/abs/2602.07120 · **PDF:** https://arxiv.org/pdf/2602.07120v1
+
+## 14662. Learning a Generative Meta-Model of LLM Activations
+
+- **ID:** `2602.06964` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Grace Luo, Jiahai Feng, Trevor Darrell, et al.
+- **abs:** https://arxiv.org/abs/2602.06964 · **PDF:** https://arxiv.org/pdf/2602.06964v1
+
+## 14663. InftyThink+: Effective and Efficient Infinite-Horizon Reasoning via Reinforcement Learning
+
+- **ID:** `2602.06960` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Yuchen Yan, Liang Jiang, Jin Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.06960 · **PDF:** https://arxiv.org/pdf/2602.06960v2
+
+## 14664. Improving Credit Card Fraud Detection with an Optimized Explainable Boosting Machine
+
+- **ID:** `2602.06955` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Reza E. Fazel, Arash Bakhtiary, Siavash A. Bigdeli
+- **abs:** https://arxiv.org/abs/2602.06955 · **PDF:** https://arxiv.org/pdf/2602.06955v1
+
+## 14665. TTSR: Test-Time Self-Reflection for Continual Reasoning Improvement
+
+- **ID:** `2603.03297` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Haoyang He, Zihua Rong, Liangjie Zhao, et al.
+- **abs:** https://arxiv.org/abs/2603.03297 · **PDF:** https://arxiv.org/pdf/2603.03297v1
+
+## 14666. PlugMem: A Task-Agnostic Plugin Memory Module for LLM Agents
+
+- **ID:** `2603.03296` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Ke Yang, Zixi Chen, Xuan He, et al.
+- **abs:** https://arxiv.org/abs/2603.03296 · **PDF:** https://arxiv.org/pdf/2603.03296v1
+
+## 14667. DAWN: Dependency-Aware Fast Inference for Diffusion LLMs
+
+- **ID:** `2602.06953` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Lizhuo Luo, Zhuoran Shi, Jiajun Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.06953 · **PDF:** https://arxiv.org/pdf/2602.06953v1
+
+## 14668. DreamDojo: A Generalist Robot World Model from Large-Scale Human Videos
+
+- **ID:** `2602.06949` · **primary:** `cs.RO` · **Submitted:** 2026-02-06
+- **Authors:** Shenyuan Gao, William Liang, Kaiyuan Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.06949 · **PDF:** https://arxiv.org/pdf/2602.06949v1
+
+## 14669. Agentic Uncertainty Reveals Agentic Overconfidence
+
+- **ID:** `2602.06948` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Jean Kaddour, Srijan Patel, Gbètondji Dovonon, et al.
+- **abs:** https://arxiv.org/abs/2602.06948 · **PDF:** https://arxiv.org/pdf/2602.06948v1
+
+## 14670. Towards Understanding Multimodal Fine-Tuning: Spatial Features
+
+- **ID:** `2602.08713` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Lachin Naghashyar, Hunar Batra, Ashkan Khakzar, et al.
+- **abs:** https://arxiv.org/abs/2602.08713 · **PDF:** https://arxiv.org/pdf/2602.08713v1
+
+## 14671. Optimal Derivative Feedback Control for an Active Magnetic Levitation System: An Experimental Study on Data-Driven Approaches
+
+- **ID:** `2602.06944` · **primary:** `eess.SY` · **Submitted:** 2026-02-06
+- **Authors:** Saber Omidi, Rene Akupan Ebunle, Se Young Yoon
+- **abs:** https://arxiv.org/abs/2602.06944 · **PDF:** https://arxiv.org/pdf/2602.06944v1
+
+## 14672. Optimal Turkish Subword Strategies at Scale: Systematic Evaluation of Data, Vocabulary, Morphology Interplay
+
+- **ID:** `2602.06942` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Duygu Altinok
+- **abs:** https://arxiv.org/abs/2602.06942 · **PDF:** https://arxiv.org/pdf/2602.06942v1
+
+## 14673. Endogenous Resistance to Activation Steering in Language Models
+
+- **ID:** `2602.06941` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Alex McKenzie, Keenan Pepper, Stijn Servaes, et al.
+- **abs:** https://arxiv.org/abs/2602.06941 · **PDF:** https://arxiv.org/pdf/2602.06941v1
+
+## 14674. From Core to Detail: Unsupervised Disentanglement with Entropy-Ordered Flows
+
+- **ID:** `2602.06940` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Daniel Galperin, Ullrich Köthe
+- **abs:** https://arxiv.org/abs/2602.06940 · **PDF:** https://arxiv.org/pdf/2602.06940v1
+
+## 14675. Cochain Perspectives on Temporal-Difference Signals for Learning Beyond Markov Dynamics
+
+- **ID:** `2602.06939` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Zuyuan Zhang, Sizhe Tang, Tian Lan
+- **abs:** https://arxiv.org/abs/2602.06939 · **PDF:** https://arxiv.org/pdf/2602.06939v1
+
+## 14676. ShallowJail: Steering Jailbreaks against Large Language Models
+
+- **ID:** `2602.07107` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Shang Liu, Hanyu Pei, Zeyan Liu
+- **abs:** https://arxiv.org/abs/2602.07107 · **PDF:** https://arxiv.org/pdf/2602.07107v2
+
+## 14677. Reliable Mislabel Detection for Video Capsule Endoscopy Data
+
+- **ID:** `2602.06938` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Julia Werner, Julius Oexle, Oliver Bause, et al.
+- **abs:** https://arxiv.org/abs/2602.06938 · **PDF:** https://arxiv.org/pdf/2602.06938v1
+
+## 14678. Reciprocal Latent Fields for Precomputed Sound Propagation
+
+- **ID:** `2602.06937` · **primary:** `cs.SD` · **Submitted:** 2026-02-06
+- **Authors:** Hugo Seuté, Pranai Vasudev, Etienne Richan, et al.
+- **abs:** https://arxiv.org/abs/2602.06937 · **PDF:** https://arxiv.org/pdf/2602.06937v1
+
+## 14679. Implementing Grassroots Logic Programs with Multiagent Transition Systems and AI
+
+- **ID:** `2602.06934` · **primary:** `cs.PL` · **Submitted:** 2026-02-06
+- **Authors:** Ehud Shapiro
+- **abs:** https://arxiv.org/abs/2602.06934 · **PDF:** https://arxiv.org/pdf/2602.06934v2
+
+## 14680. When RL Meets Adaptive Speculative Training: A Unified Training-Serving System
+
+- **ID:** `2602.06932` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Junxiong Wang, Fengxiang Bie, Jisen Li, et al.
+- **abs:** https://arxiv.org/abs/2602.06932 · **PDF:** https://arxiv.org/pdf/2602.06932v1
+
+## 14681. Continuous-time reinforcement learning: ellipticity enables model-free value function approximation
+
+- **ID:** `2602.06930` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Wenlong Mou
+- **abs:** https://arxiv.org/abs/2602.06930 · **PDF:** https://arxiv.org/pdf/2602.06930v1
+
+## 14682. Robustness Beyond Known Groups with Low-rank Adaptation
+
+- **ID:** `2602.06924` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Abinitha Gourabathina, Hyewon Jeong, Teya Bergamaschi, et al.
+- **abs:** https://arxiv.org/abs/2602.06924 · **PDF:** https://arxiv.org/pdf/2602.06924v2
+
+## 14683. From Kepler to Newton: Inductive Biases Guide Learned World Models in Transformers
+
+- **ID:** `2602.06923` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Ziming Liu, Sophia Sanborn, Surya Ganguli, et al.
+- **abs:** https://arxiv.org/abs/2602.06923 · **PDF:** https://arxiv.org/pdf/2602.06923v1
+
+## 14684. Halluverse-M^3: A multitask multilingual benchmark for hallucination in LLMs
+
+- **ID:** `2602.06920` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Samir Abdaljalil, Parichit Sharma, Erchin Serpedin, et al.
+- **abs:** https://arxiv.org/abs/2602.06920 · **PDF:** https://arxiv.org/pdf/2602.06920v1
+
+## 14685. Automatic Detection and Analysis of Singing Mistakes for Music Pedagogy
+
+- **ID:** `2602.06917` · **primary:** `eess.AS` · **Submitted:** 2026-02-06
+- **Authors:** Sumit Kumar, Suraj Jaiswal, Parampreet Singh, et al.
+- **abs:** https://arxiv.org/abs/2602.06917 · **PDF:** https://arxiv.org/pdf/2602.06917v1
+
+## 14686. ScaleBITS: Scalable Bitwidth Search for Hardware-Aligned Mixed-Precision LLMs
+
+- **ID:** `2602.17698` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Xinlin Li, Timothy Chou, Josh Fromm, et al.
+- **abs:** https://arxiv.org/abs/2602.17698 · **PDF:** https://arxiv.org/pdf/2602.17698v1
+
+## 14687. PANC: Prior-Aware Normalized Cut for Object Segmentation
+
+- **ID:** `2602.06912` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Juan Gutiérrez, Victor Gutiérrez-Garcia, José Luis Blanco-Murillo
+- **abs:** https://arxiv.org/abs/2602.06912 · **PDF:** https://arxiv.org/pdf/2602.06912v1
+
+## 14688. Design Conductor: An agent autonomously builds a 1.5 GHz Linux-capable RISC-V CPU
+
+- **ID:** `2603.08716` · **primary:** `cs.AR` · **Submitted:** 2026-02-06
+- **Authors:** The Verkor Team, Ravi Krishna, Suresh Krishna, et al.
+- **abs:** https://arxiv.org/abs/2603.08716 · **PDF:** https://arxiv.org/pdf/2603.08716v1
+
+## 14689. TamperBench: Systematically Stress-Testing LLM Safety Under Fine-Tuning and Tampering
+
+- **ID:** `2602.06911` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Saad Hossain, Tom Tseng, Punya Syon Pandey, et al.
+- **abs:** https://arxiv.org/abs/2602.06911 · **PDF:** https://arxiv.org/pdf/2602.06911v1
+
+## 14690. Ex-Omni: Enabling 3D Facial Animation Generation for Omni-modal Large Language Models
+
+- **ID:** `2602.07106` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Haoyu Zhang, Zhipeng Li, Yiwen Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.07106 · **PDF:** https://arxiv.org/pdf/2602.07106v1
+
+## 14691. Revisiting the Generic Transformer: Deconstructing a Strong Baseline for Time Series Foundation Models
+
+- **ID:** `2602.06909` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Yunshi Wen, Wesley M. Gifford, Chandra Reddy, et al.
+- **abs:** https://arxiv.org/abs/2602.06909 · **PDF:** https://arxiv.org/pdf/2602.06909v1
+
+## 14692. A first realization of reinforcement learning-based closed-loop EEG-TMS
+
+- **ID:** `2602.06907` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Dania Humaidan, Jiahua Xu, Jing Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.06907 · **PDF:** https://arxiv.org/pdf/2602.06907v1
+
+## 14693. Improved Upper Bounds for Slicing the Hypercube
+
+- **ID:** `2602.16807` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Duncan Soiffer, Nathaniel Itty, Christopher D. Rosin, et al.
+- **abs:** https://arxiv.org/abs/2602.16807 · **PDF:** https://arxiv.org/pdf/2602.16807v1
+
+## 14694. Parameter-free Dynamic Regret: Time-varying Movement Costs, Delayed Feedback, and Memory
+
+- **ID:** `2602.06902` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Emmanuel Esposito, Andrew Jacobsen, Hao Qiu, et al.
+- **abs:** https://arxiv.org/abs/2602.06902 · **PDF:** https://arxiv.org/pdf/2602.06902v1
+
+## 14695. Supercharging Simulation-Based Inference for Bayesian Optimal Experimental Design
+
+- **ID:** `2602.06900` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Samuel Klein, Willie Neiswanger, Daniel Ratner, et al.
+- **abs:** https://arxiv.org/abs/2602.06900 · **PDF:** https://arxiv.org/pdf/2602.06900v1
+
+## 14696. Sample Complexity of Causal Identification with Temporal Heterogeneity
+
+- **ID:** `2602.06899` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Ameya Rathod, Sujay Belsare, Salvik Krishna Nautiyal, et al.
+- **abs:** https://arxiv.org/abs/2602.06899 · **PDF:** https://arxiv.org/pdf/2602.06899v1
+
+## 14697. Extended to Reality: Prompt Injection in 3D Environments
+
+- **ID:** `2602.07104` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Zhuoheng Li, Ying Chen
+- **abs:** https://arxiv.org/abs/2602.07104 · **PDF:** https://arxiv.org/pdf/2602.07104v1
+
+## 14698. A Cycle-Consistent Graph Surrogate for Full-Cycle Left Ventricular Myocardial Biomechanics
+
+- **ID:** `2602.06884` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Siyu Mu, Wei Xuan Chan, Choon Hwai Yap
+- **abs:** https://arxiv.org/abs/2602.06884 · **PDF:** https://arxiv.org/pdf/2602.06884v1
+
+## 14699. Vision Transformer Finetuning Benefits from Non-Smooth Components
+
+- **ID:** `2602.06883` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Ambroise Odonnat, Laetitia Chapel, Romain Tavenard, et al.
+- **abs:** https://arxiv.org/abs/2602.06883 · **PDF:** https://arxiv.org/pdf/2602.06883v2
+
+## 14700. Decoupling Variance and Scale-Invariant Updates in Adaptive Gradient Descent for Unified Vector and Matrix Optimization
+
+- **ID:** `2602.06880` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Zitao Song, Cedar Site Bai, Zhe Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.06880 · **PDF:** https://arxiv.org/pdf/2602.06880v1
+
+## 14701. NanoFLUX: Distillation-Driven Compression of Large Text-to-Image Generation Models for Mobile Devices
+
+- **ID:** `2602.06879` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Ruchika Chavhan, Malcolm Chadwick, Alberto Gil Couto Pimentel Ramos, et al.
+- **abs:** https://arxiv.org/abs/2602.06879 · **PDF:** https://arxiv.org/pdf/2602.06879v1
+
+## 14702. scDFM: Distributional Flow Matching Model for Robust Single-Cell Perturbation Prediction
+
+- **ID:** `2602.07103` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-06
+- **Authors:** Chenglei Yu, Chuanrui Wang, Bangyan Liao, et al.
+- **abs:** https://arxiv.org/abs/2602.07103 · **PDF:** https://arxiv.org/pdf/2602.07103v1
+
+## 14703. TraceCoder: A Trace-Driven Multi-Agent Framework for Automated Debugging of LLM-Generated Code
+
+- **ID:** `2602.06875` · **primary:** `cs.SE` · **Submitted:** 2026-02-06
+- **Authors:** Jiangping Huang, Wenguang Ye, Weisong Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.06875 · **PDF:** https://arxiv.org/pdf/2602.06875v1
+
+## 14704. Uncovering Cross-Objective Interference in Multi-Objective Alignment
+
+- **ID:** `2602.06869` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Yining Lu, Meng Jiang
+- **abs:** https://arxiv.org/abs/2602.06869 · **PDF:** https://arxiv.org/pdf/2602.06869v1
+
+## 14705. T-STAR: A Context-Aware Transformer Framework for Short-Term Probabilistic Demand Forecasting in Dock-Based Shared Micro-Mobility
+
+- **ID:** `2602.06866` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Jingyi Cheng, Gonçalo Homem de Almeida Correia, Oded Cats, et al.
+- **abs:** https://arxiv.org/abs/2602.06866 · **PDF:** https://arxiv.org/pdf/2602.06866v1
+
+## 14706. Git for Sketches: An Intelligent Tracking System for Capturing Design Evolution
+
+- **ID:** `2602.06047` · **primary:** `cs.HC` · **Submitted:** 2026-02-06
+- **Authors:** Sankar B, Amogh A S, Sandhya Baranwal, et al.
+- **abs:** https://arxiv.org/abs/2602.06047 · **PDF:** https://arxiv.org/pdf/2602.06047v1
+
+## 14707. Zero-shot Generalizable Graph Anomaly Detection with Mixture of Riemannian Experts
+
+- **ID:** `2602.06859` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Xinyu Zhao, Qingyun Sun, Jiayi Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.06859 · **PDF:** https://arxiv.org/pdf/2602.06859v2
+
+## 14708. Designing a Robust, Bounded, and Smooth Loss Function for Improved Supervised Learning
+
+- **ID:** `2602.06858` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Soumi Mahato, Lineesh M. C
+- **abs:** https://arxiv.org/abs/2602.06858 · **PDF:** https://arxiv.org/pdf/2602.06858v1
+
+## 14709. AIRS-Bench: a Suite of Tasks for Frontier AI Research Science Agents
+
+- **ID:** `2602.06855` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Alisia Lupidi, Bhavul Gauri, Thomas Simon Foster, et al.
+- **abs:** https://arxiv.org/abs/2602.06855 · **PDF:** https://arxiv.org/pdf/2602.06855v3
+
+## 14710. SEMA: Simple yet Effective Learning for Multi-Turn Jailbreak Attacks
+
+- **ID:** `2602.06854` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Mingqian Feng, Xiaodong Liu, Weiwei Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.06854 · **PDF:** https://arxiv.org/pdf/2602.06854v1
+
+## 14711. The Quantum Sieve Tracer: A Hybrid Framework for Layer-Wise Activation Tracing in Large Language Models
+
+- **ID:** `2602.06852` · **primary:** `quant-ph` · **Submitted:** 2026-02-06
+- **Authors:** Jonathan Pan
+- **abs:** https://arxiv.org/abs/2602.06852 · **PDF:** https://arxiv.org/pdf/2602.06852v1
+
+## 14712. Rethinking Multi-Condition DiTs: Eliminating Redundant Attention via Position-Alignment and Keyword-Scoping
+
+- **ID:** `2602.06850` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Chao Zhou, Tianyi Wei, Yiling Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.06850 · **PDF:** https://arxiv.org/pdf/2602.06850v1
+
+## 14713. Improved Sampling Schedules for Discrete Diffusion Models
+
+- **ID:** `2602.06849` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Alberto Foresti, Mustapha Bounoua, Giulio Franzese, et al.
+- **abs:** https://arxiv.org/abs/2602.06849 · **PDF:** https://arxiv.org/pdf/2602.06849v2
+
+## 14714. The Representational Geometry of Number
+
+- **ID:** `2602.06843` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Zhimin Hu, Lanhao Niu, Sashank Varma
+- **abs:** https://arxiv.org/abs/2602.06843 · **PDF:** https://arxiv.org/pdf/2602.06843v1
+
+## 14715. Are Deep Learning Based Hybrid PDE Solvers Reliable? Why Training Paradigms and Update Strategies Matter
+
+- **ID:** `2602.06842` · **primary:** `math.NA` · **Submitted:** 2026-02-06
+- **Authors:** Yuhan Wu, Jan Willem van Beek, Victorita Dolean, et al.
+- **abs:** https://arxiv.org/abs/2602.06842 · **PDF:** https://arxiv.org/pdf/2602.06842v1
+
+## 14716. From Features to Actions: Explainability in Traditional and Agentic AI Systems
+
+- **ID:** `2602.06841` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Sindhuja Chaduvula, Jessee Ho, Kina Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.06841 · **PDF:** https://arxiv.org/pdf/2602.06841v3
+
+## 14717. A Zipf-preserving, long-range correlated surrogate for written language and other symbolic sequences
+
+- **ID:** `2603.02213` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Marcelo A. Montemurro, Mirko Degli Esposti
+- **abs:** https://arxiv.org/abs/2603.02213 · **PDF:** https://arxiv.org/pdf/2603.02213v1
+
+## 14718. An Adaptive Differentially Private Federated Learning Framework with Bi-level Optimization
+
+- **ID:** `2602.06838` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Jin Wang, Hui Ma, Fei Xing, et al.
+- **abs:** https://arxiv.org/abs/2602.06838 · **PDF:** https://arxiv.org/pdf/2602.06838v2
+
+## 14719. Learning Deep Hybrid Models with Sharpness-Aware Minimization
+
+- **ID:** `2602.06837` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Naoya Takeishi
+- **abs:** https://arxiv.org/abs/2602.06837 · **PDF:** https://arxiv.org/pdf/2602.06837v1
+
+## 14720. LLM Active Alignment: A Nash Equilibrium Perspective
+
+- **ID:** `2602.06836` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Tonghan Wang, Yuqi Pan, Xinyi Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.06836 · **PDF:** https://arxiv.org/pdf/2602.06836v1
+
+## 14721. Fast and Robust Likelihood-Guided Diffusion Posterior Sampling with Amortized Variational Inference
+
+- **ID:** `2602.07102` · **primary:** `stat.ML` · **Submitted:** 2026-02-06
+- **Authors:** Léon Zheng, Thomas Hirtz, Yazid Janati, et al.
+- **abs:** https://arxiv.org/abs/2602.07102 · **PDF:** https://arxiv.org/pdf/2602.07102v1
+
+## 14722. GraFSTNet: Graph-based Frequency SpatioTemporal Network for Cellular Traffic Prediction
+
+- **ID:** `2602.13282` · **primary:** `cs.NI` · **Submitted:** 2026-02-06
+- **Authors:** Ziyi Li, Hui Ma, Fei Xing, et al.
+- **abs:** https://arxiv.org/abs/2602.13282 · **PDF:** https://arxiv.org/pdf/2602.13282v2
+
+## 14723. Pimp My LLM: Leveraging Variability Modeling to Tune Inference Hyperparameters
+
+- **ID:** `2602.17697` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Nada Zine, Clément Quinton, Romain Rouvoy
+- **abs:** https://arxiv.org/abs/2602.17697 · **PDF:** https://arxiv.org/pdf/2602.17697v1
+
+## 14724. AEGPO: Adaptive Entropy-Guided Policy Optimization for Diffusion Models
+
+- **ID:** `2602.06825` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Yuming Li, Qingyu Li, Chengyu Bai, et al.
+- **abs:** https://arxiv.org/abs/2602.06825 · **PDF:** https://arxiv.org/pdf/2602.06825v1
+
+## 14725. Can LLM Safety Be Ensured by Constraining Parameter Regions?
+
+- **ID:** `2602.17696` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Zongmin Li, Jian Su, Farah Benamara, et al.
+- **abs:** https://arxiv.org/abs/2602.17696 · **PDF:** https://arxiv.org/pdf/2602.17696v1
+
+## 14726. RanSOM: Second-Order Momentum with Randomized Scaling for Constrained and Unconstrained Optimization
+
+- **ID:** `2602.06824` · **primary:** `math.OC` · **Submitted:** 2026-02-06
+- **Authors:** El Mahdi Chayti
+- **abs:** https://arxiv.org/abs/2602.06824 · **PDF:** https://arxiv.org/pdf/2602.06824v1
+
+## 14727. AI-Generated Music Detection in Broadcast Monitoring
+
+- **ID:** `2602.06823` · **primary:** `cs.SD` · **Submitted:** 2026-02-06
+- **Authors:** David López-Ayala, Asier Cabello, Pablo Zinemanas, et al.
+- **abs:** https://arxiv.org/abs/2602.06823 · **PDF:** https://arxiv.org/pdf/2602.06823v2
+
+## 14728. POP: Online Structural Pruning Enables Efficient Inference of Large Foundation Models
+
+- **ID:** `2602.06822` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Yi Chen, Wonjin Shin, Shuhong Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.06822 · **PDF:** https://arxiv.org/pdf/2602.06822v1
+
+## 14729. ScaleEnv: Scaling Environment Synthesis from Scratch for Generalist Interactive Tool-Use Agent Training
+
+- **ID:** `2602.06820` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Dunwei Tu, Hongyan Hao, Hansi Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.06820 · **PDF:** https://arxiv.org/pdf/2602.06820v1
+
+## 14730. Bridging 6G IoT and AI: LLM-Based Efficient Approach for Physical Layer's Optimization Tasks
+
+- **ID:** `2602.06819` · **primary:** `eess.SP` · **Submitted:** 2026-02-06
+- **Authors:** Ahsan Mehmood, Naveed Ul Hassan, Ghassan M. Kraidy
+- **abs:** https://arxiv.org/abs/2602.06819 · **PDF:** https://arxiv.org/pdf/2602.06819v1
+
+## 14731. Wild Guesses and Mild Guesses in Active Concept Learning
+
+- **ID:** `2602.06818` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Anirudh Chari, Neil Pattanaik
+- **abs:** https://arxiv.org/abs/2602.06818 · **PDF:** https://arxiv.org/pdf/2602.06818v1
+
+## 14732. Calibrating Tabular Anomaly Detection via Optimal Transport
+
+- **ID:** `2602.06810` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Hangting Ye, He Zhao. Wei Fan, Xiaozhuang Song, et al.
+- **abs:** https://arxiv.org/abs/2602.06810 · **PDF:** https://arxiv.org/pdf/2602.06810v1
+
+## 14733. SuReNav: Superpixel Graph-based Constraint Relaxation for Navigation in Over-constrained Environments
+
+- **ID:** `2602.06807` · **primary:** `cs.RO` · **Submitted:** 2026-02-06
+- **Authors:** Keonyoung Koh, Moonkyeong Jung, Samuel Seungsup Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.06807 · **PDF:** https://arxiv.org/pdf/2602.06807v1
+
+## 14734. RAIGen: Rare Attribute Identification in Text-to-Image Generative Models
+
+- **ID:** `2602.06806` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Silpa Vadakkeeveetil Sreelatha, Dan Wang, Serge Belongie, et al.
+- **abs:** https://arxiv.org/abs/2602.06806 · **PDF:** https://arxiv.org/pdf/2602.06806v1
+
+## 14735. On the Non-Identifiability of Steering Vectors in Large Language Models
+
+- **ID:** `2602.06801` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Sohan Venkatesh, Ashish Mahendran Kurapath
+- **abs:** https://arxiv.org/abs/2602.06801 · **PDF:** https://arxiv.org/pdf/2602.06801v4
+
+## 14736. FlowDA: Accurate, Low-Latency Weather Data Assimilation via Flow Matching
+
+- **ID:** `2602.06800` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Ran Cheng, Lailai Zhu
+- **abs:** https://arxiv.org/abs/2602.06800 · **PDF:** https://arxiv.org/pdf/2602.06800v1
+
+## 14737. Visual Word Sense Disambiguation with CLIP through Dual-Channel Text Prompting and Image Augmentations
+
+- **ID:** `2602.06799` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Shamik Bhattacharya, Daniel Perkins, Yaren Dogan, et al.
+- **abs:** https://arxiv.org/abs/2602.06799 · **PDF:** https://arxiv.org/pdf/2602.06799v1
+
+## 14738. Optimal Learning-Rate Schedules under Functional Scaling Laws: Power Decay and Warmup-Stable-Decay
+
+- **ID:** `2602.06797` · **primary:** `stat.ML` · **Submitted:** 2026-02-06
+- **Authors:** Binghui Li, Zilin Wang, Fengling Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.06797 · **PDF:** https://arxiv.org/pdf/2602.06797v2
+
+## 14739. Generating Data-Driven Reasoning Rubrics for Domain-Adaptive Reward Modeling
+
+- **ID:** `2602.06795` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Kate Sanders, Nathaniel Weir, Sapana Chaudhary, et al.
+- **abs:** https://arxiv.org/abs/2602.06795 · **PDF:** https://arxiv.org/pdf/2602.06795v1
+
+## 14740. Rare Event Analysis of Large Language Models
+
+- **ID:** `2602.06791` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Jake McAllister Dorman, Edward Gillman, Dominic C. Rose, et al.
+- **abs:** https://arxiv.org/abs/2602.06791 · **PDF:** https://arxiv.org/pdf/2602.06791v1
+
+## 14741. Displacement-Resistant Extensions of DPO with Nonconvex $f$-Divergences
+
+- **ID:** `2602.06788` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Idan Pipano, Shoham Sabach, Kavosh Asadi, et al.
+- **abs:** https://arxiv.org/abs/2602.06788 · **PDF:** https://arxiv.org/pdf/2602.06788v1
+
+## 14742. Weisfeiler and Lehman Go Categorical
+
+- **ID:** `2602.06787` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Seongjin Choi, Gahee Kim, Se-Young Yun
+- **abs:** https://arxiv.org/abs/2602.06787 · **PDF:** https://arxiv.org/pdf/2602.06787v1
+
+## 14743. Language Model Goal Selection Differs from Humans' in an Open-Ended Task
+
+- **ID:** `2603.03295` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Gaia Molinaro, Dave August, Danielle Perszyk, et al.
+- **abs:** https://arxiv.org/abs/2603.03295 · **PDF:** https://arxiv.org/pdf/2603.03295v1
+
+## 14744. Revisiting Emotions Representation for Recognition in the Wild
+
+- **ID:** `2602.06778` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Joao Baptista Cardia Neto, Claudio Ferrari, Stefano Berretti
+- **abs:** https://arxiv.org/abs/2602.06778 · **PDF:** https://arxiv.org/pdf/2602.06778v1
+
+## 14745. Next-generation cyberattack detection with large language models: anomaly analysis across heterogeneous logs
+
+- **ID:** `2602.06777` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Yassine Chagna, Antal Goldschmidt
+- **abs:** https://arxiv.org/abs/2602.06777 · **PDF:** https://arxiv.org/pdf/2602.06777v1
+
+## 14746. Fair Transit Stop Placement: A Clustering Perspective and Beyond
+
+- **ID:** `2602.06776` · **primary:** `cs.GT` · **Submitted:** 2026-02-06
+- **Authors:** Haris Aziz, Ling Gai, Yuhang Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.06776 · **PDF:** https://arxiv.org/pdf/2602.06776v1
+
+## 14747. Robust Online Learning
+
+- **ID:** `2602.06775` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Sajad Ashkezari
+- **abs:** https://arxiv.org/abs/2602.06775 · **PDF:** https://arxiv.org/pdf/2602.06775v2
+
+## 14748. Towards Understanding What State Space Models Learn About Code
+
+- **ID:** `2602.06774` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Jiali Wu, Abhinav Anand, Shweta Verma, et al.
+- **abs:** https://arxiv.org/abs/2602.06774 · **PDF:** https://arxiv.org/pdf/2602.06774v1
+
+## 14749. On the Convergence of Multicalibration Gradient Boosting
+
+- **ID:** `2602.06773` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Daniel Haimovich, Fridolin Linder, Lorenzo Perini, et al.
+- **abs:** https://arxiv.org/abs/2602.06773 · **PDF:** https://arxiv.org/pdf/2602.06773v1
+
+## 14750. Calibrating Generative AI to Produce Realistic Essays for Data Augmentation
+
+- **ID:** `2602.06772` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Edward W. Wolfe, Justin O. Barber
+- **abs:** https://arxiv.org/abs/2602.06772 · **PDF:** https://arxiv.org/pdf/2602.06772v1
+
+## 14751. AEGIS: Adversarial Target-Guided Retention-Data-Free Robust Concept Erasure from Diffusion Models
+
+- **ID:** `2602.06771` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Fengpeng Li, Kemou Li, Qizhou Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.06771 · **PDF:** https://arxiv.org/pdf/2602.06771v2
+
+## 14752. Soft Forward-Backward Representations for Zero-shot Reinforcement Learning with General Utilities
+
+- **ID:** `2602.06769` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Marco Bagatella, Thomas Rupf, Georg Martius, et al.
+- **abs:** https://arxiv.org/abs/2602.06769 · **PDF:** https://arxiv.org/pdf/2602.06769v1
+
+## 14753. Sparse Spike Encoding of Channel Responses for Energy Efficient Human Activity Recognition
+
+- **ID:** `2602.06766` · **primary:** `cs.NE` · **Submitted:** 2026-02-06
+- **Authors:** Eleonora Cicciarella, Riccardo Mazzieri, Jacopo Pegoraro, et al.
+- **abs:** https://arxiv.org/abs/2602.06766 · **PDF:** https://arxiv.org/pdf/2602.06766v1
+
+## 14754. R-Align: Enhancing Generative Reward Models through Rationale-Centric Meta-Judging
+
+- **ID:** `2602.06763` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Yanlin Lai, Mitt Huang, Hangyu Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.06763 · **PDF:** https://arxiv.org/pdf/2602.06763v1
+
+## 14755. Assessing the Reliability of Persona-Conditioned LLMs as Synthetic Survey Respondents
+
+- **ID:** `2602.18462` · **primary:** `cs.CY` · **Submitted:** 2026-02-06
+- **Authors:** Erika Elizabeth Taday Morocho, Lorenzo Cima, Tiziano Fagni, et al.
+- **abs:** https://arxiv.org/abs/2602.18462 · **PDF:** https://arxiv.org/pdf/2602.18462v1
+
+## 14756. BayesFlow 2: Multi-Backend Amortized Bayesian Inference in Python
+
+- **ID:** `2602.07098` · **primary:** `stat.CO` · **Submitted:** 2026-02-06
+- **Authors:** Lars Kühmichel, Jerry M. Huang, Valentin Pratz, et al.
+- **abs:** https://arxiv.org/abs/2602.07098 · **PDF:** https://arxiv.org/pdf/2602.07098v2
+
+## 14757. A Unified Framework for LLM Watermarks
+
+- **ID:** `2602.06754` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Thibaud Gloaguen, Robin Staab, Nikola Jovanović, et al.
+- **abs:** https://arxiv.org/abs/2602.06754 · **PDF:** https://arxiv.org/pdf/2602.06754v1
+
+## 14758. EXACT: Explicit Attribute-Guided Decoding-Time Personalization
+
+- **ID:** `2602.17695` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Xin Yu, Hanwen Xing, Lingzhou Xue
+- **abs:** https://arxiv.org/abs/2602.17695 · **PDF:** https://arxiv.org/pdf/2602.17695v1
+
+## 14759. Gold Exploration using Representations from a Multispectral Autoencoder
+
+- **ID:** `2602.06748` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Argyro Tsandalidou, Konstantinos Dogeas, Eleftheria Tetoula Tsonga, et al.
+- **abs:** https://arxiv.org/abs/2602.06748 · **PDF:** https://arxiv.org/pdf/2602.06748v1
+
+## 14760. Semantically Labelled Automata for Multi-Task Reinforcement Learning with LTL Instructions
+
+- **ID:** `2602.06746` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Alessandro Abate, Giuseppe De Giacomo, Mathias Jackermeier, et al.
+- **abs:** https://arxiv.org/abs/2602.06746 · **PDF:** https://arxiv.org/pdf/2602.06746v1
+
+## 14761. Structural bias in multi-objective optimisation
+
+- **ID:** `2602.06742` · **primary:** `cs.NE` · **Submitted:** 2026-02-06
+- **Authors:** Jakub Kudela, Niki van Stein, Thomas Bäck, et al.
+- **abs:** https://arxiv.org/abs/2602.06742 · **PDF:** https://arxiv.org/pdf/2602.06742v1
+
+## 14762. Disentanglement by means of action-induced representations
+
+- **ID:** `2602.06741` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Gorka Muñoz-Gil, Hendrik Poulsen Nautrup, Arunava Majumder, et al.
+- **abs:** https://arxiv.org/abs/2602.06741 · **PDF:** https://arxiv.org/pdf/2602.06741v1
+
+## 14763. Optimal Abstractions for Verifying Properties of Kolmogorov-Arnold Networks (KANs)
+
+- **ID:** `2602.06737` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Noah Schwartz, Chandra Kanth Nagesh, Sriram Sankaranarayanan, et al.
+- **abs:** https://arxiv.org/abs/2602.06737 · **PDF:** https://arxiv.org/pdf/2602.06737v1
+
+## 14764. Pairwise is Not Enough: Hypergraph Neural Networks for Multi-Agent Pathfinding
+
+- **ID:** `2602.06733` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Rishabh Jain, Keisuke Okumura, Michael Amir, et al.
+- **abs:** https://arxiv.org/abs/2602.06733 · **PDF:** https://arxiv.org/pdf/2602.06733v1
+
+## 14765. Table-as-Search: Formulate Long-Horizon Agentic Information Seeking as Table Completion
+
+- **ID:** `2602.06724` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Tian Lan, Felix Henry, Bin Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.06724 · **PDF:** https://arxiv.org/pdf/2602.06724v1
+
+## 14766. GhostCite: A Large-Scale Analysis of Citation Validity in the Age of Large Language Models
+
+- **ID:** `2602.06718` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Zuyao Xu, Yuqi Qiu, Lu Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.06718 · **PDF:** https://arxiv.org/pdf/2602.06718v1
+
+## 14767. F-GRPO: Don't Let Your Policy Learn the Obvious and Forget the Rare
+
+- **ID:** `2602.06717` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Daniil Plyusov, Alexey Gorbatovski, Boris Shaposhnikov, et al.
+- **abs:** https://arxiv.org/abs/2602.06717 · **PDF:** https://arxiv.org/pdf/2602.06717v1
+
+## 14768. Missing At Random as Covariate Shift: Correcting Bias in Iterative Imputation
+
+- **ID:** `2602.06713` · **primary:** `stat.ML` · **Submitted:** 2026-02-06
+- **Authors:** Luke Shannon, Song Liu, Katarzyna Reluga
+- **abs:** https://arxiv.org/abs/2602.06713 · **PDF:** https://arxiv.org/pdf/2602.06713v1
+
+## 14769. Autoregressive Models for Knowledge Graph Generation
+
+- **ID:** `2602.06707` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Thiviyan Thanapalasingam, Antonis Vozikis, Peter Bloem, et al.
+- **abs:** https://arxiv.org/abs/2602.06707 · **PDF:** https://arxiv.org/pdf/2602.06707v1
+
+## 14770. SaDiT: Efficient Protein Backbone Design via Latent Structural Tokenization and Diffusion Transformers
+
+- **ID:** `2602.06706` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Shentong Mo, Lanqing Li
+- **abs:** https://arxiv.org/abs/2602.06706 · **PDF:** https://arxiv.org/pdf/2602.06706v1
+
+## 14771. RealFin: How Well Do LLMs Reason About Finance When Users Leave Things Unsaid?
+
+- **ID:** `2602.07096` · **primary:** `q-fin.ST` · **Submitted:** 2026-02-06
+- **Authors:** Yuyang Dai, Yan Lin, Zhuohan Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.07096 · **PDF:** https://arxiv.org/pdf/2602.07096v1
+
+## 14772. Explaining Grokking in Transformers through the Lens of Inductive Bias
+
+- **ID:** `2602.06702` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Jaisidh Singh, Diganta Misra, Antonio Orvieto
+- **abs:** https://arxiv.org/abs/2602.06702 · **PDF:** https://arxiv.org/pdf/2602.06702v1
+
+## 14773. WorldEdit: Towards Open-World Image Editing with a Knowledge-Informed Benchmark
+
+- **ID:** `2602.07095` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Wang Lin, Feng Wang, Majun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.07095 · **PDF:** https://arxiv.org/pdf/2602.07095v1
+
+## 14774. Evidence for Daily and Weekly Periodic Variability in GPT-4o Performance
+
+- **ID:** `2602.15889` · **primary:** `stat.AP` · **Submitted:** 2026-02-06
+- **Authors:** Paul Tschisgale, Peter Wulff
+- **abs:** https://arxiv.org/abs/2602.15889 · **PDF:** https://arxiv.org/pdf/2602.15889v1
+
+## 14775. GhazalBench: Usage-Grounded Evaluation of LLMs on Persian Ghazals
+
+- **ID:** `2603.09979` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Ghazal Kalhor, Yadollah Yaghoobzadeh
+- **abs:** https://arxiv.org/abs/2603.09979 · **PDF:** https://arxiv.org/pdf/2603.09979v1
+
+## 14776. Taipan: A Query-free Transfer-based Multiple Sensitive Attribute Inference Attack Solely from Publicly Released Graphs
+
+- **ID:** `2602.06700` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Ying Song, Balaji Palanisamy
+- **abs:** https://arxiv.org/abs/2602.06700 · **PDF:** https://arxiv.org/pdf/2602.06700v1
+
+## 14777. Quantum Attention by Overlap Interference: Predicting Sequences from Classical and Many-Body Quantum Data
+
+- **ID:** `2602.06699` · **primary:** `quant-ph` · **Submitted:** 2026-02-06
+- **Authors:** Alessio Pecilli, Matteo Rosati
+- **abs:** https://arxiv.org/abs/2602.06699 · **PDF:** https://arxiv.org/pdf/2602.06699v1
+
+## 14778. Diffeomorphism-Equivariant Neural Networks
+
+- **ID:** `2602.06695` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Josephine Elisabeth Oettinger, Zakhar Shumaylov, Johannes Bostelmann, et al.
+- **abs:** https://arxiv.org/abs/2602.06695 · **PDF:** https://arxiv.org/pdf/2602.06695v1
+
+## 14779. NanoQuant: Efficient Sub-1-Bit Quantization of Large Language Models
+
+- **ID:** `2602.06694` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Hyochan Chong, Dongkyu Kim, Changdong Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.06694 · **PDF:** https://arxiv.org/pdf/2602.06694v1
+
+## 14780. Makespan Minimization in Split Learning: From Theory to Practice
+
+- **ID:** `2602.06693` · **primary:** `cs.NI` · **Submitted:** 2026-02-06
+- **Authors:** Robert Ganian, Fionn Mc Inerney, Dimitra Tsigkari
+- **abs:** https://arxiv.org/abs/2602.06693 · **PDF:** https://arxiv.org/pdf/2602.06693v1
+
+## 14781. Evaluating Prompt Engineering Strategies for Sentiment Control in AI-Generated Texts
+
+- **ID:** `2602.06692` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Kerstin Sahler, Sophie Jentzsch
+- **abs:** https://arxiv.org/abs/2602.06692 · **PDF:** https://arxiv.org/pdf/2602.06692v1
+
+## 14782. Memory-Conditioned Flow-Matching for Stable Autoregressive PDE Rollouts
+
+- **ID:** `2602.06689` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Victor Armegioiu
+- **abs:** https://arxiv.org/abs/2602.06689 · **PDF:** https://arxiv.org/pdf/2602.06689v1
+
+## 14783. NeuroSleep: Neuromorphic Event-Driven Single-Channel EEG Sleep Staging for Edge-Efficient Sensing
+
+- **ID:** `2602.15888` · **primary:** `eess.SP` · **Submitted:** 2026-02-06
+- **Authors:** Boyu Li, Xingchun Zhu, Yonghui Wu
+- **abs:** https://arxiv.org/abs/2602.15888 · **PDF:** https://arxiv.org/pdf/2602.15888v2
+
+## 14784. AsynDBT: Asynchronous Distributed Bilevel Tuning for efficient In-Context Learning with Large Language Models
+
+- **ID:** `2602.17694` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Hui Ma, Shaoyu Dou, Ya Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.17694 · **PDF:** https://arxiv.org/pdf/2602.17694v1
+
+## 14785. Pruning at Initialisation through the lens of Graphon Limit: Convergence, Expressivity, and Generalisation
+
+- **ID:** `2602.06675` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Hoang Pham, The-Anh Ta, Long Tran-Thanh
+- **abs:** https://arxiv.org/abs/2602.06675 · **PDF:** https://arxiv.org/pdf/2602.06675v1
+
+## 14786. CytoCrowd: A Multi-Annotator Benchmark Dataset for Cytology Image Analysis
+
+- **ID:** `2602.06674` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Yonghao Si, Xingyuan Zeng, Zhao Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.06674 · **PDF:** https://arxiv.org/pdf/2602.06674v1
+
+## 14787. Automatic Method Illustration Generation for AI Scientific Papers via Drawing Middleware Creation, Evolution, and Orchestration
+
+- **ID:** `2603.29590` · **primary:** `cs.GR` · **Submitted:** 2026-02-06
+- **Authors:** Zhuoling Li, Jiarui Zhang, Ping Hu, et al.
+- **abs:** https://arxiv.org/abs/2603.29590 · **PDF:** https://arxiv.org/pdf/2603.29590v2
+
+## 14788. compar:IA: The French Government's LLM arena to collect French-language human prompts and preference data
+
+- **ID:** `2602.06669` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Lucie Termignon, Simonas Zilinskas, Hadrien Pélissier, et al.
+- **abs:** https://arxiv.org/abs/2602.06669 · **PDF:** https://arxiv.org/pdf/2602.06669v1
+
+## 14789. Not All Layers Need Tuning: Selective Layer Restoration Recovers Diversity
+
+- **ID:** `2602.06665` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Bowen Zhang, Meiyi Wang, Harold Soh
+- **abs:** https://arxiv.org/abs/2602.06665 · **PDF:** https://arxiv.org/pdf/2602.06665v1
+
+## 14790. Multimodal Generative Retrieval Model with Staged Pretraining for Food Delivery on Meituan
+
+- **ID:** `2602.06654` · **primary:** `cs.IR` · **Submitted:** 2026-02-06
+- **Authors:** Boyu Chen, Tai Guo, Weiyu Cui, et al.
+- **abs:** https://arxiv.org/abs/2602.06654 · **PDF:** https://arxiv.org/pdf/2602.06654v1
+
+## 14791. RAPID: Reconfigurable, Adaptive Platform for Iterative Design
+
+- **ID:** `2602.06653` · **primary:** `cs.RO` · **Submitted:** 2026-02-06
+- **Authors:** Zi Yin, Fanhong Li, Shurui Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.06653 · **PDF:** https://arxiv.org/pdf/2602.06653v1
+
+## 14792. Same Answer, Different Representations: Hidden instability in VLMs
+
+- **ID:** `2602.06652` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Farooq Ahmad Wani, Alessandro Suglia, Rohit Saxena, et al.
+- **abs:** https://arxiv.org/abs/2602.06652 · **PDF:** https://arxiv.org/pdf/2602.06652v1
+
+## 14793. Beyond Static Alignment: Hierarchical Policy Control for LLM Safety via Risk-Aware Chain-of-Thought
+
+- **ID:** `2602.06650` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Jianfeng Si, Lin Sun, Weihong Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.06650 · **PDF:** https://arxiv.org/pdf/2602.06650v1
+
+## 14794. Reading Between the Waves: Robust Topic Segmentation Using Inter-Sentence Audio Features
+
+- **ID:** `2602.06647` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Steffen Freisinger, Philipp Seeberger, Tobias Bocklet, et al.
+- **abs:** https://arxiv.org/abs/2602.06647 · **PDF:** https://arxiv.org/pdf/2602.06647v1
+
+## 14795. Humanoid Manipulation Interface: Humanoid Whole-Body Manipulation from Robot-Free Demonstrations
+
+- **ID:** `2602.06643` · **primary:** `cs.RO` · **Submitted:** 2026-02-06
+- **Authors:** Ruiqian Nai, Boyuan Zheng, Junming Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.06643 · **PDF:** https://arxiv.org/pdf/2602.06643v2
+
+## 14796. Temperature Scaling Attack Disrupting Model Confidence in Federated Learning
+
+- **ID:** `2602.06638` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Kichang Lee, Jaeho Jin, JaeYeon Park, et al.
+- **abs:** https://arxiv.org/abs/2602.06638 · **PDF:** https://arxiv.org/pdf/2602.06638v2
+
+## 14797. Graph-Based Nearest-Neighbor Search without the Spread
+
+- **ID:** `2602.06633` · **primary:** `cs.CG` · **Submitted:** 2026-02-06
+- **Authors:** Jeff Giliberti, Sariel Har-Peled, Jonas Sauer, et al.
+- **abs:** https://arxiv.org/abs/2602.06633 · **PDF:** https://arxiv.org/pdf/2602.06633v1
+
+## 14798. Trust Regions Sell, But Who's Buying? Overlap Geometry as an Alternative Trust Region for Policy Optimization
+
+- **ID:** `2602.06627` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Gaurish Trivedi, Alakh Sharma, Kartikey Singh Bhandari, et al.
+- **abs:** https://arxiv.org/abs/2602.06627 · **PDF:** https://arxiv.org/pdf/2602.06627v1
+
+## 14799. UnWeaving the knots of GraphRAG -- turns out VectorRAG is almost enough
+
+- **ID:** `2603.29875` · **primary:** `cs.IR` · **Submitted:** 2026-02-06
+- **Authors:** Ryszard Tuora, Mateusz Galiński, Michał Godziszewski, et al.
+- **abs:** https://arxiv.org/abs/2603.29875 · **PDF:** https://arxiv.org/pdf/2603.29875v1
+
+## 14800. FairJudge: An Adaptive, Debiased, and Consistent LLM-as-a-Judge
+
+- **ID:** `2602.06625` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Bo Yang, Lanfei Feng, Yunkui Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.06625 · **PDF:** https://arxiv.org/pdf/2602.06625v1
+
+## 14801. Do Prompts Guarantee Safety? Mitigating Toxicity from LLM Generations through Subspace Intervention
+
+- **ID:** `2602.06623` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Himanshu Singh, Ziwei Xu, A. V. Subramanyam, et al.
+- **abs:** https://arxiv.org/abs/2602.06623 · **PDF:** https://arxiv.org/pdf/2602.06623v1
+
+## 14802. Infinite-dimensional generative diffusions via Doob's h-transform
+
+- **ID:** `2602.06621` · **primary:** `stat.ML` · **Submitted:** 2026-02-06
+- **Authors:** Thorben Pieper-Sethmacher, Daniel Paulin
+- **abs:** https://arxiv.org/abs/2602.06621 · **PDF:** https://arxiv.org/pdf/2602.06621v1
+
+## 14803. Confundo: Learning to Generate Robust Poison for Practical RAG Systems
+
+- **ID:** `2602.06616` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Haoyang Hu, Zhejun Jiang, Yueming Lyu, et al.
+- **abs:** https://arxiv.org/abs/2602.06616 · **PDF:** https://arxiv.org/pdf/2602.06616v1
+
+## 14804. DAVE: Distribution-aware Attribution via ViT Gradient Decomposition
+
+- **ID:** `2602.06613` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Adam Wróbel, Siddhartha Gairola, Jacek Tabor, et al.
+- **abs:** https://arxiv.org/abs/2602.06613 · **PDF:** https://arxiv.org/pdf/2602.06613v1
+
+## 14805. Adaptive-CaRe: Adaptive Causal Regularization for Robust Outcome Prediction
+
+- **ID:** `2602.06611` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Nithya Bhasker, Fiona R. Kolbinger, Susu Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.06611 · **PDF:** https://arxiv.org/pdf/2602.06611v1
+
+## 14806. The challenge of generating and evolving real-life like synthetic test data without accessing real-world raw data -- a Systematic Review
+
+- **ID:** `2602.06609` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Maj-Annika Tammisto, Faiz Ali Shah, Daniel Rodriguez, et al.
+- **abs:** https://arxiv.org/abs/2602.06609 · **PDF:** https://arxiv.org/pdf/2602.06609v1
+
+## 14807. Green Optimization: Energy-aware Design of Metaheuristics by Using Machine Learning Surrogates to Cope with Real Problems
+
+- **ID:** `2602.06610` · **primary:** `cs.NE` · **Submitted:** 2026-02-06
+- **Authors:** Tomohiro Harada, Enrique Alba, Gabriel Luque
+- **abs:** https://arxiv.org/abs/2602.06610 · **PDF:** https://arxiv.org/pdf/2602.06610v2
+
+## 14808. The hidden risks of temporal resampling in clinical reinforcement learning
+
+- **ID:** `2602.06603` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Thomas Frost, Hrisheekesh Vaidya, Steve Harris
+- **abs:** https://arxiv.org/abs/2602.06603 · **PDF:** https://arxiv.org/pdf/2602.06603v2
+
+## 14809. Scaling Speech Tokenizers with Diffusion Autoencoders
+
+- **ID:** `2602.06602` · **primary:** `cs.SD` · **Submitted:** 2026-02-06
+- **Authors:** Yuancheng Wang, Zhenyu Tang, Yun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.06602 · **PDF:** https://arxiv.org/pdf/2602.06602v1
+
+## 14810. Echoes as Anchors: Probabilistic Costs and Attention Refocusing in LLM Reasoning
+
+- **ID:** `2602.06600` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Zhuoyuan Hao, Zhuo Li, Wu Li, et al.
+- **abs:** https://arxiv.org/abs/2602.06600 · **PDF:** https://arxiv.org/pdf/2602.06600v1
+
+## 14811. Sample-Efficient Policy Space Response Oracles with Joint Experience Best Response
+
+- **ID:** `2602.06599` · **primary:** `cs.MA` · **Submitted:** 2026-02-06
+- **Authors:** Ariyan Bighashdel, Thiago D. Simão, Frans A. Oliehoek
+- **abs:** https://arxiv.org/abs/2602.06599 · **PDF:** https://arxiv.org/pdf/2602.06599v1
+
+## 14812. DiTS: Multimodal Diffusion Transformers Are Time Series Forecasters
+
+- **ID:** `2602.06597` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Haoran Zhang, Haixuan Liu, Yong Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.06597 · **PDF:** https://arxiv.org/pdf/2602.06597v1
+
+## 14813. Personality as Relational Infrastructure: User Perceptions of Personality-Trait-Infused LLM Messaging
+
+- **ID:** `2602.06596` · **primary:** `cs.HC` · **Submitted:** 2026-02-06
+- **Authors:** Dominik P. Hofer, David Haag, Rania Islambouli, et al.
+- **abs:** https://arxiv.org/abs/2602.06596 · **PDF:** https://arxiv.org/pdf/2602.06596v2
+
+## 14814. Energy-Aware Metaheuristics
+
+- **ID:** `2602.06595` · **primary:** `cs.NE` · **Submitted:** 2026-02-06
+- **Authors:** Tomohiro Harada, Enrique Alba, Gabriel Luque
+- **abs:** https://arxiv.org/abs/2602.06595 · **PDF:** https://arxiv.org/pdf/2602.06595v3
+
+## 14815. AgentStepper: Interactive Debugging of Software Development Agents
+
+- **ID:** `2602.06593` · **primary:** `cs.SE` · **Submitted:** 2026-02-06
+- **Authors:** Robert Hutter, Michael Pradel
+- **abs:** https://arxiv.org/abs/2602.06593 · **PDF:** https://arxiv.org/pdf/2602.06593v1
+
+## 14816. ProtoQuant: Quantization of Prototypical Parts For General and Fine-Grained Image Classification
+
+- **ID:** `2602.06592` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Mikołaj Janusz, Adam Wróbel, Bartosz Zieliński, et al.
+- **abs:** https://arxiv.org/abs/2602.06592 · **PDF:** https://arxiv.org/pdf/2602.06592v1
+
+## 14817. Degradation of Feature Space in Continual Learning
+
+- **ID:** `2602.06586` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Chiara Lanza, Roberto Pereira, Marco Miozzo, et al.
+- **abs:** https://arxiv.org/abs/2602.06586 · **PDF:** https://arxiv.org/pdf/2602.06586v1
+
+## 14818. Target noise: A pre-training based neural network initialization for efficient high resolution learning
+
+- **ID:** `2602.06585` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Shaowen Wang, Tariq Alkhalifah
+- **abs:** https://arxiv.org/abs/2602.06585 · **PDF:** https://arxiv.org/pdf/2602.06585v1
+
+## 14819. Inference-Time Rethinking with Latent Thought Vectors for Math Reasoning
+
+- **ID:** `2602.06584` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Deqian Kong, Minglu Zhao, Aoyang Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.06584 · **PDF:** https://arxiv.org/pdf/2602.06584v1
+
+## 14820. Efficient Online Variational Estimation via Monte Carlo Sampling
+
+- **ID:** `2602.06579` · **primary:** `stat.ME` · **Submitted:** 2026-02-06
+- **Authors:** Mathis Chagneux, Mathias Müller, Pierre Gloaguen, et al.
+- **abs:** https://arxiv.org/abs/2602.06579 · **PDF:** https://arxiv.org/pdf/2602.06579v1
+
+## 14821. Exploring Sparsity and Smoothness of Arbitrary $\ell_p$ Norms in Adversarial Attacks
+
+- **ID:** `2602.06578` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Christof Duhme, Florian Eilers, Xiaoyi Jiang
+- **abs:** https://arxiv.org/abs/2602.06578 · **PDF:** https://arxiv.org/pdf/2602.06578v1
+
+## 14822. Perturbing the Phase: Analyzing Adversarial Robustness of Complex-Valued Neural Networks
+
+- **ID:** `2602.06577` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Florian Eilers, Christof Duhme, Xiaoyi Jiang
+- **abs:** https://arxiv.org/abs/2602.06577 · **PDF:** https://arxiv.org/pdf/2602.06577v1
+
+## 14823. Fine-Tuning and Evaluating Conversational AI for Agricultural Advisory
+
+- **ID:** `2603.03294` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Sanyam Singh, Naga Ganesh, Vineet Singh, et al.
+- **abs:** https://arxiv.org/abs/2603.03294 · **PDF:** https://arxiv.org/pdf/2603.03294v2
+
+## 14824. Transformer-based Parameter Fitting of Models derived from Bloch-McConnell Equations for CEST MRI Analysis
+
+- **ID:** `2602.06574` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Christof Duhme, Chris Lippe, Verena Hoerr, et al.
+- **abs:** https://arxiv.org/abs/2602.06574 · **PDF:** https://arxiv.org/pdf/2602.06574v1
+
+## 14825. Lemon Agent Technical Report
+
+- **ID:** `2602.07092` · **primary:** `cs.MA` · **Submitted:** 2026-02-06
+- **Authors:** Haipeng Jiang, Kailong Ren, Zimo Yin, et al.
+- **abs:** https://arxiv.org/abs/2602.07092 · **PDF:** https://arxiv.org/pdf/2602.07092v1
+
+## 14826. Baichuan-M3: Modeling Clinical Inquiry for Reliable Medical Decision-Making
+
+- **ID:** `2602.06570` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Baichuan-M3 Team, :, Chengfeng Dou, et al.
+- **abs:** https://arxiv.org/abs/2602.06570 · **PDF:** https://arxiv.org/pdf/2602.06570v1
+
+## 14827. SPARC: Separating Perception And Reasoning Circuits for Test-time Scaling of VLMs
+
+- **ID:** `2602.06566` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Niccolo Avogaro, Nayanika Debnath, Li Mi, et al.
+- **abs:** https://arxiv.org/abs/2602.06566 · **PDF:** https://arxiv.org/pdf/2602.06566v2
+
+## 14828. Learning to Allocate Resources with Censored Feedback
+
+- **ID:** `2602.06565` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Giovanni Montanari, Côme Fiegel, Corentin Pla, et al.
+- **abs:** https://arxiv.org/abs/2602.06565 · **PDF:** https://arxiv.org/pdf/2602.06565v1
+
+## 14829. Which Graph Shift Operator? A Spectral Answer to an Empirical Question
+
+- **ID:** `2602.06557` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Yassine Abbahaddou
+- **abs:** https://arxiv.org/abs/2602.06557 · **PDF:** https://arxiv.org/pdf/2602.06557v1
+
+## 14830. LIBERO-X: Robustness Litmus for Vision-Language-Action Models
+
+- **ID:** `2602.06556` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Guodong Wang, Chenkai Zhang, Qingjie Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.06556 · **PDF:** https://arxiv.org/pdf/2602.06556v1
+
+## 14831. Reinforcement Learning-Based Dynamic Management of Structured Parallel Farm Skeletons on Serverless Platforms
+
+- **ID:** `2602.06555` · **primary:** `cs.DC` · **Submitted:** 2026-02-06
+- **Authors:** Lanpei Li, Massimo Coppola, Malio Li, et al.
+- **abs:** https://arxiv.org/abs/2602.06555 · **PDF:** https://arxiv.org/pdf/2602.06555v1
+
+## 14832. SeeUPO: Sequence-Level Agentic-RL with Convergence Guarantees
+
+- **ID:** `2602.06554` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Tianyi Hu, Qingxu Fu, Yanxi Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.06554 · **PDF:** https://arxiv.org/pdf/2602.06554v1
+
+## 14833. Evolving Ranking Functions for Canonical Blow-Ups in Positive Characteristic
+
+- **ID:** `2602.06553` · **primary:** `math.AG` · **Submitted:** 2026-02-06
+- **Authors:** Gergely Bérczi
+- **abs:** https://arxiv.org/abs/2602.06553 · **PDF:** https://arxiv.org/pdf/2602.06553v1
+
+## 14834. Fine-Grained Model Merging via Modular Expert Recombination
+
+- **ID:** `2602.06552` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Haiyun Qiu, Xingyu Wu, Liang Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.06552 · **PDF:** https://arxiv.org/pdf/2602.06552v1
+
+## 14835. Dynamics-Aligned Shared Hypernetworks for Zero-Shot Actuator Inversion
+
+- **ID:** `2602.06550` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Jan Benad, Pradeep Kr. Banerjee, Frank Röder, et al.
+- **abs:** https://arxiv.org/abs/2602.06550 · **PDF:** https://arxiv.org/pdf/2602.06550v1
+
+## 14836. Refining the Information Bottleneck via Adversarial Information Separation
+
+- **ID:** `2602.06549` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Shuai Ning, Zhenpeng Wang, Lin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.06549 · **PDF:** https://arxiv.org/pdf/2602.06549v2
+
+## 14837. NECromancer: Breathing Life into Skeletons via BVH Animation
+
+- **ID:** `2602.06548` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Mingxi Xu, Qi Wang, Zhengyu Wen, et al.
+- **abs:** https://arxiv.org/abs/2602.06548 · **PDF:** https://arxiv.org/pdf/2602.06548v1
+
+## 14838. Malicious Agent Skills in the Wild: A Large-Scale Security Empirical Study
+
+- **ID:** `2602.06547` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Yi Liu, Zhihao Chen, Yanjun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.06547 · **PDF:** https://arxiv.org/pdf/2602.06547v2
+
+## 14839. MTQE.en-he: Machine Translation Quality Estimation for English-Hebrew
+
+- **ID:** `2602.06546` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Andy Rosenbaum, Assaf Siani, Ilan Kernerman
+- **abs:** https://arxiv.org/abs/2602.06546 · **PDF:** https://arxiv.org/pdf/2602.06546v1
+
+## 14840. Operationalizing Stein's Method for Online Linear Optimization: CLT-Based Optimal Tradeoffs
+
+- **ID:** `2602.06545` · **primary:** `stat.ML` · **Submitted:** 2026-02-06
+- **Authors:** Zhiyu Zhang, Aaditya Ramdas
+- **abs:** https://arxiv.org/abs/2602.06545 · **PDF:** https://arxiv.org/pdf/2602.06545v1
+
+## 14841. Live Knowledge Tracing: Real-Time Adaptation using Tabular Foundation Models
+
+- **ID:** `2602.06542` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Mounir Lbath, Alexandre Paresy, Abdelkayoum Kaddouri, et al.
+- **abs:** https://arxiv.org/abs/2602.06542 · **PDF:** https://arxiv.org/pdf/2602.06542v2
+
+## 14842. AgentCPM-Report: Interleaving Drafting and Deepening for Open-Ended Deep Research
+
+- **ID:** `2602.06540` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Yishan Li, Wentong Chen, Yukun Yan, et al.
+- **abs:** https://arxiv.org/abs/2602.06540 · **PDF:** https://arxiv.org/pdf/2602.06540v1
+
+## 14843. Revisiting the Sliced Wasserstein Kernel for persistence diagrams: a Figalli-Gigli approach
+
+- **ID:** `2602.06539` · **primary:** `stat.ML` · **Submitted:** 2026-02-06
+- **Authors:** Marc Janthial, Théo Lacombe
+- **abs:** https://arxiv.org/abs/2602.06539 · **PDF:** https://arxiv.org/pdf/2602.06539v1
+
+## 14844. AlertBERT: A noise-robust alert grouping framework for simultaneous cyber attacks
+
+- **ID:** `2602.06534` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Lukas Karner, Max Landauer, Markus Wurzenberger, et al.
+- **abs:** https://arxiv.org/abs/2602.06534 · **PDF:** https://arxiv.org/pdf/2602.06534v1
+
+## 14845. LogicSkills: A Structured Benchmark for Formal Reasoning in Large Language Models
+
+- **ID:** `2602.06533` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Brian Rabern, Philipp Mondorf, Barbara Plank
+- **abs:** https://arxiv.org/abs/2602.06533 · **PDF:** https://arxiv.org/pdf/2602.06533v2
+
+## 14846. Concept-Aware Privacy Mechanisms for Defending Embedding Inversion Attacks
+
+- **ID:** `2602.07090` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Yu-Che Tsai, Hsiang Hsiao, Kuan-Yu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.07090 · **PDF:** https://arxiv.org/pdf/2602.07090v1
+
+## 14847. HyPER: Bridging Exploration and Exploitation for Scalable LLM Reasoning with Hypothesis Path Expansion and Reduction
+
+- **ID:** `2602.06527` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Shengxuan Qiu, Haochen Huang, Shuzhang Zhong, et al.
+- **abs:** https://arxiv.org/abs/2602.06527 · **PDF:** https://arxiv.org/pdf/2602.06527v1
+
+## 14848. Completing Missing Annotation: Multi-Agent Debate for Accurate and Scalable Relevant Assessment for IR Benchmarks
+
+- **ID:** `2602.06526` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Minjeong Ban, Jeonghwan Choi, Hyangsuk Min, et al.
+- **abs:** https://arxiv.org/abs/2602.06526 · **PDF:** https://arxiv.org/pdf/2602.06526v1
+
+## 14849. Progress Constraints for Reinforcement Learning in Behavior Trees
+
+- **ID:** `2602.06525` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Finn Rietz, Mart Kartašev, Petter Ögren, et al.
+- **abs:** https://arxiv.org/abs/2602.06525 · **PDF:** https://arxiv.org/pdf/2602.06525v2
+
+## 14850. Topography scanning as a part of process monitoring in power cable insulation process
+
+- **ID:** `2602.06519` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Janne Harjuhahto, Jaakko Harjuhahto, Mikko Lahti, et al.
+- **abs:** https://arxiv.org/abs/2602.06519 · **PDF:** https://arxiv.org/pdf/2602.06519v1
+
+## 14851. Attention-Driven Framework for Non-Rigid Medical Image Registration
+
+- **ID:** `2602.07088` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Muhammad Zafar Iqbal, Ghazanfar Farooq Siddiqui, Anwar Ul Haq, et al.
+- **abs:** https://arxiv.org/abs/2602.07088 · **PDF:** https://arxiv.org/pdf/2602.07088v1
+
+## 14852. Sequential Auditing for f-Differential Privacy
+
+- **ID:** `2602.06518` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Tim Kutta, Martin Dunsche, Yu Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.06518 · **PDF:** https://arxiv.org/pdf/2602.06518v1
+
+## 14853. A Case Study of Selected PTQ Baselines for Reasoning LLMs on Ascend NPU
+
+- **ID:** `2602.17693` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Yuchen Luo, Fangyue Zhu, Ruining Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.17693 · **PDF:** https://arxiv.org/pdf/2602.17693v1
+
+## 14854. SE-Search: Self-Evolving Search Agent via Memory and Dense Reward
+
+- **ID:** `2603.03293` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Jian Li, Yizhang Jin, Dongqi Liu, et al.
+- **abs:** https://arxiv.org/abs/2603.03293 · **PDF:** https://arxiv.org/pdf/2603.03293v1
+
+## 14855. Evolutionary Generation of Multi-Agent Systems
+
+- **ID:** `2602.06511` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Yuntong Hu, Matthew Trager, Yuting Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.06511 · **PDF:** https://arxiv.org/pdf/2602.06511v2
+
+## 14856. Designing Computational Tools for Exploring Causal Relationships in Qualitative Data
+
+- **ID:** `2602.06506` · **primary:** `cs.HC` · **Submitted:** 2026-02-06
+- **Authors:** Han Meng, Qiuyuan Lyu, Peinuan Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.06506 · **PDF:** https://arxiv.org/pdf/2602.06506v1
+
+## 14857. Forest canopy height estimation from satellite RGB imagery using large-scale airborne LiDAR-derived training data and monocular depth estimation
+
+- **ID:** `2602.06503` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Yongkang Lai, Xihan Mu, Dasheng Fan, et al.
+- **abs:** https://arxiv.org/abs/2602.06503 · **PDF:** https://arxiv.org/pdf/2602.06503v2
+
+## 14858. Can Microcanonical Langevin Dynamics Leverage Mini-Batch Gradient Noise?
+
+- **ID:** `2602.06500` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Emanuel Sommer, Kangning Diao, Jakob Robnik, et al.
+- **abs:** https://arxiv.org/abs/2602.06500 · **PDF:** https://arxiv.org/pdf/2602.06500v1
+
+## 14859. Electron-Informed Coarse-Graining Molecular Representation Learning for Real-World Molecular Physics
+
+- **ID:** `2602.07087` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-06
+- **Authors:** Gyoung S. Na, Chanyoung Park
+- **abs:** https://arxiv.org/abs/2602.07087 · **PDF:** https://arxiv.org/pdf/2602.07087v1
+
+## 14860. Adaptive Uncertainty-Aware Tree Search for Robust Reasoning
+
+- **ID:** `2602.06493` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Zeen Song, Zihao Ma, Wenwen Qiang, et al.
+- **abs:** https://arxiv.org/abs/2602.06493 · **PDF:** https://arxiv.org/pdf/2602.06493v1
+
+## 14861. Evaluating Retrieval-Augmented Generation Variants for Natural Language-Based SQL and API Call Generation
+
+- **ID:** `2602.07086` · **primary:** `cs.SE` · **Submitted:** 2026-02-06
+- **Authors:** Michael Marketsmüller, Simon Martin, Tim Schlippe
+- **abs:** https://arxiv.org/abs/2602.07086 · **PDF:** https://arxiv.org/pdf/2602.07086v1
+
+## 14862. JADE: Expert-Grounded Dynamic Evaluation for Open-Ended Professional Tasks
+
+- **ID:** `2602.06486` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Lanbo Lin, Jiayao Liu, Tianyuan Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.06486 · **PDF:** https://arxiv.org/pdf/2602.06486v1
+
+## 14863. From Conflict to Consensus: Boosting Medical Reasoning via Multi-Round Agentic RAG
+
+- **ID:** `2603.03292` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Wenhao Wu, Zhentao Tang, Yafu Li, et al.
+- **abs:** https://arxiv.org/abs/2603.03292 · **PDF:** https://arxiv.org/pdf/2603.03292v2
+
+## 14864. AgentCPM-Explore: Realizing Long-Horizon Deep Exploration for Edge-Scale Agents
+
+- **ID:** `2602.06485` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Haotian Chen, Xin Cong, Shengda Fan, et al.
+- **abs:** https://arxiv.org/abs/2602.06485 · **PDF:** https://arxiv.org/pdf/2602.06485v1
+
+## 14865. Efficient-LVSM: Faster, Cheaper, and Better Large View Synthesis Model via Decoupled Co-Refinement Attention
+
+- **ID:** `2602.06478` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Xiaosong Jia, Yihang Sun, Junqi You, et al.
+- **abs:** https://arxiv.org/abs/2602.06478 · **PDF:** https://arxiv.org/pdf/2602.06478v1
+
+## 14866. QuantaAlpha: An Evolutionary Framework for LLM-Driven Alpha Mining
+
+- **ID:** `2602.07085` · **primary:** `q-fin.ST` · **Submitted:** 2026-02-06
+- **Authors:** Jun Han, Shuo Zhang, Wei Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07085 · **PDF:** https://arxiv.org/pdf/2602.07085v1
+
+## 14867. BEAGLE: Behavior-Enforced Agent for Grounded Learner Emulation
+
+- **ID:** `2602.13280` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Hanchen David Wang, Clayton Cohn, Zifan Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.13280 · **PDF:** https://arxiv.org/pdf/2602.13280v1
+
+## 14868. Prism: Spectral Parameter Sharing for Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.06476` · **primary:** `cs.MA` · **Submitted:** 2026-02-06
+- **Authors:** Kyungbeom Kim, Seungwon Oh, Kyung-Joong Kim
+- **abs:** https://arxiv.org/abs/2602.06476 · **PDF:** https://arxiv.org/pdf/2602.06476v1
+
+## 14869. Towards Generalizable Reasoning: Group Causal Counterfactual Policy Optimization for LLM Reasoning
+
+- **ID:** `2602.06475` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Jingyao Wang, Peizheng Guo, Wenwen Qiang, et al.
+- **abs:** https://arxiv.org/abs/2602.06475 · **PDF:** https://arxiv.org/pdf/2602.06475v1
+
+## 14870. Revisiting the Shape Convention of Transformer Language Models
+
+- **ID:** `2602.06471` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Feng-Ting Liao, Meng-Hsi Chen, Guan-Ting Yi, et al.
+- **abs:** https://arxiv.org/abs/2602.06471 · **PDF:** https://arxiv.org/pdf/2602.06471v1
+
+## 14871. Improve Large Language Model Systems with User Logs
+
+- **ID:** `2602.06470` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Changyue Wang, Weihang Su, Qingyao Ai, et al.
+- **abs:** https://arxiv.org/abs/2602.06470 · **PDF:** https://arxiv.org/pdf/2602.06470v1
+
+## 14872. AbFlow : End-to-end Paratope-Centric Antibody Design by Interaction Enhanced Flow Matching
+
+- **ID:** `2602.07084` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-06
+- **Authors:** Wenda Wang, Yang Zhang, Zhewei Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.07084 · **PDF:** https://arxiv.org/pdf/2602.07084v1
+
+## 14873. Diffusion-State Policy Optimization for Masked Diffusion Language Models
+
+- **ID:** `2602.06462` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Daisuke Oba, Hiroki Furuta, Naoaki Okazaki
+- **abs:** https://arxiv.org/abs/2602.06462 · **PDF:** https://arxiv.org/pdf/2602.06462v2
+
+## 14874. Achieving Better Local Regret Bound for Online Non-Convex Bilevel Optimization
+
+- **ID:** `2602.06457` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Tingkai Jia, Haiguang Wang, Cheng Chen
+- **abs:** https://arxiv.org/abs/2602.06457 · **PDF:** https://arxiv.org/pdf/2602.06457v1
+
+## 14875. The Window Dilemma: Why Concept Drift Detection is Ill-Posed
+
+- **ID:** `2602.06456` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Brandon Gower-Winter, Misja Groen, Georg Krempl
+- **abs:** https://arxiv.org/abs/2602.06456 · **PDF:** https://arxiv.org/pdf/2602.06456v1
+
+## 14876. RelayGen: Intra-Generation Model Switching for Efficient Reasoning
+
+- **ID:** `2602.06454` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Jiwon Song, Yoongon Kim, Jae-Joon Kim
+- **abs:** https://arxiv.org/abs/2602.06454 · **PDF:** https://arxiv.org/pdf/2602.06454v1
+
+## 14877. On the Plasticity and Stability for Post-Training Large Language Models
+
+- **ID:** `2602.06453` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Wenwen Qiang, Ziyin Gu, Jiahuan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.06453 · **PDF:** https://arxiv.org/pdf/2602.06453v1
+
+## 14878. BrokenBind: Universal Modality Exploration beyond Dataset Boundaries
+
+- **ID:** `2602.06451` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Zhuo Huang, Runnan Chen, Bo Han, et al.
+- **abs:** https://arxiv.org/abs/2602.06451 · **PDF:** https://arxiv.org/pdf/2602.06451v1
+
+## 14879. Evaluating an evidence-guided reinforcement learning framework in aligning light-parameter large language models with decision-making cognition in psychiatric clinical reasoning
+
+- **ID:** `2602.06449` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Xinxin Lin, Guangxin Dai, Yi Zhong, et al.
+- **abs:** https://arxiv.org/abs/2602.06449 · **PDF:** https://arxiv.org/pdf/2602.06449v1
+
+## 14880. Principle-Evolvable Scientific Discovery via Uncertainty Minimization
+
+- **ID:** `2602.06448` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Yingming Pu, Tao Lin, Hongyu Chen
+- **abs:** https://arxiv.org/abs/2602.06448 · **PDF:** https://arxiv.org/pdf/2602.06448v1
+
+## 14881. CORE: Comprehensive Ontological Relation Evaluation for Large Language Models
+
+- **ID:** `2602.06446` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Satyam Dwivedi, Sanjukta Ghosh, Shivam Dwivedi, et al.
+- **abs:** https://arxiv.org/abs/2602.06446 · **PDF:** https://arxiv.org/pdf/2602.06446v1
+
+## 14882. TrajAD: Trajectory Anomaly Detection for Trustworthy LLM Agents
+
+- **ID:** `2602.06443` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Yibing Liu, Chong Zhang, Zhongyi Han, et al.
+- **abs:** https://arxiv.org/abs/2602.06443 · **PDF:** https://arxiv.org/pdf/2602.06443v1
+
+## 14883. Agentic Unlearning: When LLM Agent Meets Machine Unlearning
+
+- **ID:** `2602.17692` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Bin Wang, Fan Wang, Pingping Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.17692 · **PDF:** https://arxiv.org/pdf/2602.17692v2
+
+## 14884. Is Gradient Ascent Really Necessary? Memorize to Forget for Machine Unlearning
+
+- **ID:** `2602.06441` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Zhuo Huang, Qizhou Wang, Ziming Hong, et al.
+- **abs:** https://arxiv.org/abs/2602.06441 · **PDF:** https://arxiv.org/pdf/2602.06441v1
+
+## 14885. TrailBlazer: History-Guided Reinforcement Learning for Black-Box LLM Jailbreaking
+
+- **ID:** `2602.06440` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Sung-Hoon Yoon, Ruizhi Qian, Minda Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.06440 · **PDF:** https://arxiv.org/pdf/2602.06440v1
+
+## 14886. A methodology for analyzing financial needs hierarchy from social discussions using LLM
+
+- **ID:** `2602.06431` · **primary:** `cs.SI` · **Submitted:** 2026-02-06
+- **Authors:** Abhishek Jangra, Sachin Thukral, Arnab Chatterjee, et al.
+- **abs:** https://arxiv.org/abs/2602.06431 · **PDF:** https://arxiv.org/pdf/2602.06431v1
+
+## 14887. Rethinking Scientific Modeling: Toward Physically Consistent and Simulation-Executable Programmatic Generation
+
+- **ID:** `2602.07083` · **primary:** `cs.SE` · **Submitted:** 2026-02-06
+- **Authors:** Yongqing Jiang, Jianze Wang, Zhiqi Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.07083 · **PDF:** https://arxiv.org/pdf/2602.07083v1
+
+## 14888. Investigating the structure of emotions by analyzing similarity and association of emotion words
+
+- **ID:** `2602.06430` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Fumitaka Iwaki, Tatsuji Takahashi
+- **abs:** https://arxiv.org/abs/2602.06430 · **PDF:** https://arxiv.org/pdf/2602.06430v1
+
+## 14889. Reclaiming First Principles: A Differentiable Framework for Conceptual Hydrologic Models
+
+- **ID:** `2602.06429` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Jasper A. Vrugt, Jonathan M. Frame, Ethan Bollman
+- **abs:** https://arxiv.org/abs/2602.06429 · **PDF:** https://arxiv.org/pdf/2602.06429v1
+
+## 14890. Beyond Code Contributions: How Network Position, Temporal Bursts, and Code Review Activities Shape Contributor Influence in Large-Scale Open Source Ecosystems
+
+- **ID:** `2602.06426` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** S M Rakib Ul Karim, Wenyi Lu, Sean Goggins
+- **abs:** https://arxiv.org/abs/2602.06426 · **PDF:** https://arxiv.org/pdf/2602.06426v1
+
+## 14891. On the Wings of Imagination: Conflicting Script-based Multi-role Framework for Humor Caption Generation
+
+- **ID:** `2602.06423` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Wenbo Shang, Yuxi Sun, Jing Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.06423 · **PDF:** https://arxiv.org/pdf/2602.06423v1
+
+## 14892. Performance Evaluation of LLMs in Automated RDF Knowledge Graph Generation
+
+- **ID:** `2603.29878` · **primary:** `cs.IR` · **Submitted:** 2026-02-06
+- **Authors:** Ioana Ramona Martin, Tudor Cioara, Ionut Anghel, et al.
+- **abs:** https://arxiv.org/abs/2603.29878 · **PDF:** https://arxiv.org/pdf/2603.29878v1
+
+## 14893. Tethered Reasoning: Decoupling Entropy from Hallucination in Quantized LLMs via Manifold Steering
+
+- **ID:** `2602.17691` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Craig Atkinson
+- **abs:** https://arxiv.org/abs/2602.17691 · **PDF:** https://arxiv.org/pdf/2602.17691v1
+
+## 14894. MosaicThinker: On-Device Visual Spatial Reasoning for Embodied AI via Iterative Construction of Space Representation
+
+- **ID:** `2602.07082` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Haoming Wang, Qiyao Xue, Weichen Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.07082 · **PDF:** https://arxiv.org/pdf/2602.07082v1
+
+## 14895. Adaptive Protein Tokenization
+
+- **ID:** `2602.06418` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Rohit Dilip, Ayush Varshney, David Van Valen
+- **abs:** https://arxiv.org/abs/2602.06418 · **PDF:** https://arxiv.org/pdf/2602.06418v1
+
+## 14896. Intrinsic Stability Limits of Autoregressive Reasoning: Structural Consequences for Long-Horizon Execution
+
+- **ID:** `2602.06413` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Hsien-Jyh Liao
+- **abs:** https://arxiv.org/abs/2602.06413 · **PDF:** https://arxiv.org/pdf/2602.06413v1
+
+## 14897. Stopping Computation for Converged Tokens in Masked Diffusion-LM Decoding
+
+- **ID:** `2602.06412` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Daisuke Oba, Danushka Bollegala, Masahiro Kaneko, et al.
+- **abs:** https://arxiv.org/abs/2602.06412 · **PDF:** https://arxiv.org/pdf/2602.06412v2
+
+## 14898. EEG Emotion Classification Using an Enhanced Transformer-CNN-BiLSTM Architecture with Dual Attention Mechanisms
+
+- **ID:** `2602.06411` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** S M Rakib UI Karim, Wenyi Lu, Diponkor Bala, et al.
+- **abs:** https://arxiv.org/abs/2602.06411 · **PDF:** https://arxiv.org/pdf/2602.06411v1
+
+## 14899. A neuromorphic model of the insect visual system for natural image processing
+
+- **ID:** `2602.06405` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Adam D. Hines, Karin Nordström, Andrew B. Barron
+- **abs:** https://arxiv.org/abs/2602.06405 · **PDF:** https://arxiv.org/pdf/2602.06405v1
+
+## 14900. Near-Optimal Regret for Distributed Adversarial Bandits: A Black-Box Approach
+
+- **ID:** `2602.06404` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Hao Qiu, Mengxiao Zhang, Nicolò Cesa-Bianchi
+- **abs:** https://arxiv.org/abs/2602.06404 · **PDF:** https://arxiv.org/pdf/2602.06404v1
+
+## 14901. TFusionOcc: Student's t-Distribution Based Object-Centric Multi-Sensor Fusion Framework for 3D Occupancy Prediction
+
+- **ID:** `2602.06400` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Zhenxing Ming, Julie Stephany Berrio, Mao Shan, et al.
+- **abs:** https://arxiv.org/abs/2602.06400 · **PDF:** https://arxiv.org/pdf/2602.06400v1
+
+## 14902. ARIS-RSMA Enhanced ISAC System: Joint Rate Splitting and Beamforming Design
+
+- **ID:** `2602.06399` · **primary:** `eess.SP` · **Submitted:** 2026-02-06
+- **Authors:** Xin Jin, Tiejun Lv, Yashuai Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.06399 · **PDF:** https://arxiv.org/pdf/2602.06399v1
+
+## 14903. Empirical Analysis of Adversarial Robustness and Explainability Drift in Cybersecurity Classifiers
+
+- **ID:** `2602.06395` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Mona Rajhans, Vishal Khawarey
+- **abs:** https://arxiv.org/abs/2602.06395 · **PDF:** https://arxiv.org/pdf/2602.06395v1
+
+## 14904. Unlocking Noisy Real-World Corpora for Foundation Model Pre-Training via Quality-Aware Tokenization
+
+- **ID:** `2602.06394` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Arvid E. Gollwitzer, Paridhi Latawa, David de Gruijl, et al.
+- **abs:** https://arxiv.org/abs/2602.06394 · **PDF:** https://arxiv.org/pdf/2602.06394v1
+
+## 14905. DesignAsCode: Bridging Structural Editability and Visual Fidelity in Graphic Design Generation
+
+- **ID:** `2602.17690` · **primary:** `cs.GR` · **Submitted:** 2026-02-06
+- **Authors:** Ziyuan Liu, Shizhao Sun, Danqing Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.17690 · **PDF:** https://arxiv.org/pdf/2602.17690v2
+
+## 14906. Generating High-quality Privacy-preserving Synthetic Data
+
+- **ID:** `2602.06390` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** David Yavo, Richard Khoury, Christophe Pere, et al.
+- **abs:** https://arxiv.org/abs/2602.06390 · **PDF:** https://arxiv.org/pdf/2602.06390v1
+
+## 14907. Uniform Spectral Growth and Convergence of Muon in LoRA-Style Matrix Factorization
+
+- **ID:** `2602.06385` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Changmin Kang, Jihun Yun, Baekrok Shin, et al.
+- **abs:** https://arxiv.org/abs/2602.06385 · **PDF:** https://arxiv.org/pdf/2602.06385v1
+
+## 14908. FMBench: Adaptive Large Language Model Output Formatting
+
+- **ID:** `2602.06384` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Yaoting Wang, Yun Zhou, Henghui Ding
+- **abs:** https://arxiv.org/abs/2602.06384 · **PDF:** https://arxiv.org/pdf/2602.06384v1
+
+## 14909. HyQuRP: Hybrid quantum-classical neural network with rotational and permutational equivariance for 3D point clouds
+
+- **ID:** `2602.06381` · **primary:** `quant-ph` · **Submitted:** 2026-02-06
+- **Authors:** Semin Park, Chae-Yeun Park
+- **abs:** https://arxiv.org/abs/2602.06381 · **PDF:** https://arxiv.org/pdf/2602.06381v1
+
+## 14910. Difficulty-Estimated Policy Optimization
+
+- **ID:** `2602.06375` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Yu Zhao, Fan Jiang, Tianle Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.06375 · **PDF:** https://arxiv.org/pdf/2602.06375v1
+
+## 14911. A Multiplicative Neural Network Architecture: Locality and Regularity of Approximation
+
+- **ID:** `2602.06374` · **primary:** `math.FA` · **Submitted:** 2026-02-06
+- **Authors:** Hee-Sun Choi, Beom-Seok Han
+- **abs:** https://arxiv.org/abs/2602.06374 · **PDF:** https://arxiv.org/pdf/2602.06374v2
+
+## 14912. ReBeCA: Unveiling Interpretable Behavior Hierarchy behind the Iterative Self-Reflection of Language Models with Causal Analysis
+
+- **ID:** `2602.06373` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Tianqiang Yan, Sihan Shang, Yuheng Li, et al.
+- **abs:** https://arxiv.org/abs/2602.06373 · **PDF:** https://arxiv.org/pdf/2602.06373v1
+
+## 14913. Cost-Aware Model Selection for Text Classification: Multi-Objective Trade-offs Between Fine-Tuned Encoders and LLM Prompting in Production
+
+- **ID:** `2602.06370` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Alberto Andres Valdes Gonzalez
+- **abs:** https://arxiv.org/abs/2602.06370 · **PDF:** https://arxiv.org/pdf/2602.06370v1
+
+## 14914. Federated Prompt-Tuning with Heterogeneous and Incomplete Multimodal Client Data
+
+- **ID:** `2602.07081` · **primary:** `cs.MM` · **Submitted:** 2026-02-06
+- **Authors:** Thu Hang Phung, Duong M. Nguyen, Thanh Trung Huynh, et al.
+- **abs:** https://arxiv.org/abs/2602.07081 · **PDF:** https://arxiv.org/pdf/2602.07081v1
+
+## 14915. Revisiting Salient Object Detection from an Observer-Centric Perspective
+
+- **ID:** `2602.06369` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Fuxi Zhang, Yifan Wang, Hengrun Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.06369 · **PDF:** https://arxiv.org/pdf/2602.06369v1
+
+## 14916. Advances in Battery Energy Storage Management: Control and Economic Synergies
+
+- **ID:** `2602.06365` · **primary:** `eess.SY` · **Submitted:** 2026-02-06
+- **Authors:** Venkata Rajesh Chundru, Shreshta Rajakumar Deshpande, Stanislav A Gankov
+- **abs:** https://arxiv.org/abs/2602.06365 · **PDF:** https://arxiv.org/pdf/2602.06365v1
+
+## 14917. CodeCircuit: Toward Inferring LLM-Generated Code Correctness via Attribution Graphs
+
+- **ID:** `2602.07080` · **primary:** `cs.SE` · **Submitted:** 2026-02-06
+- **Authors:** Yicheng He, Zheng Zhao, Zhou Kaiyu, et al.
+- **abs:** https://arxiv.org/abs/2602.07080 · **PDF:** https://arxiv.org/pdf/2602.07080v1
+
+## 14918. Envy-Free Allocation of Indivisible Goods via Noisy Queries
+
+- **ID:** `2602.06361` · **primary:** `cs.GT` · **Submitted:** 2026-02-06
+- **Authors:** Zihan Li, Yan Hao Ling, Jonathan Scarlett, et al.
+- **abs:** https://arxiv.org/abs/2602.06361 · **PDF:** https://arxiv.org/pdf/2602.06361v1
+
+## 14919. Training Data Selection with Gradient Orthogonality for Efficient Domain Adaptation
+
+- **ID:** `2602.06359` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Xiyang Zhang, Yuanhe Tian, Hongzhi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.06359 · **PDF:** https://arxiv.org/pdf/2602.06359v1
+
+## 14920. SHINE: A Scalable In-Context Hypernetwork for Mapping Context to LoRA in a Single Pass
+
+- **ID:** `2602.06358` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Yewei Liu, Xiyuan Wang, Yansheng Mao, et al.
+- **abs:** https://arxiv.org/abs/2602.06358 · **PDF:** https://arxiv.org/pdf/2602.06358v1
+
+## 14921. Evaluating LLM-persona Generated Distributions for Decision-making
+
+- **ID:** `2602.06357` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Jackie Baek, Yunhan Chen, Ziyu Chi, et al.
+- **abs:** https://arxiv.org/abs/2602.06357 · **PDF:** https://arxiv.org/pdf/2602.06357v1
+
+## 14922. Di3PO - Diptych Diffusion DPO for Targeted Improvements in Image Generation
+
+- **ID:** `2602.06355` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Sanjana Reddy, Ishaan Malhi, Sally Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.06355 · **PDF:** https://arxiv.org/pdf/2602.06355v2
+
+## 14923. Enhance and Reuse: A Dual-Mechanism Approach to Boost Deep Forest for Label Distribution Learning
+
+- **ID:** `2602.06353` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Jia-Le Xu, Shen-Huan Lyu, Yu-Nian Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.06353 · **PDF:** https://arxiv.org/pdf/2602.06353v1
+
+## 14924. Comprehensive Evaluation of Large Language Models on Software Engineering Tasks: A Multi-Task Benchmark
+
+- **ID:** `2602.07079` · **primary:** `cs.SE` · **Submitted:** 2026-02-06
+- **Authors:** Go Frendi Gunawan, Mukhlis Amien
+- **abs:** https://arxiv.org/abs/2602.07079 · **PDF:** https://arxiv.org/pdf/2602.07079v1
+
+## 14925. Trifuse: Enhancing Attention-Based GUI Grounding via Multimodal Fusion
+
+- **ID:** `2602.06351` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Longhui Ma, Di Zhao, Siwei Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.06351 · **PDF:** https://arxiv.org/pdf/2602.06351v1
+
+## 14926. Adversarial Learning in Games with Bandit Feedback: Logarithmic Pure-Strategy Maximin Regret
+
+- **ID:** `2602.06348` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Shinji Ito, Haipeng Luo, Arnab Maiti, et al.
+- **abs:** https://arxiv.org/abs/2602.06348 · **PDF:** https://arxiv.org/pdf/2602.06348v1
+
+## 14927. Zero-Trust Runtime Verification for Agentic Payment Protocols: Mitigating Replay and Context-Binding Failures in AP2
+
+- **ID:** `2602.06345` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Qianlong Lan, Anuj Kaul, Shaun Jones, et al.
+- **abs:** https://arxiv.org/abs/2602.06345 · **PDF:** https://arxiv.org/pdf/2602.06345v1
+
+## 14928. LLM-Enhanced Rumor Detection via Virtual Node Induced Edge Prediction
+
+- **ID:** `2602.13279` · **primary:** `cs.SI` · **Submitted:** 2026-02-06
+- **Authors:** Jiran Tao, Cheng Wang, Binyan Jiang
+- **abs:** https://arxiv.org/abs/2602.13279 · **PDF:** https://arxiv.org/pdf/2602.13279v2
+
+## 14929. The Optimal Token Baseline: Variance Reduction for Long-Horizon LLM-RL
+
+- **ID:** `2602.07078` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Yingru Li, Jiawei Xu, Ziniu Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07078 · **PDF:** https://arxiv.org/pdf/2602.07078v1
+
+## 14930. Action Hallucination in Generative Visual-Language-Action Models
+
+- **ID:** `2602.06339` · **primary:** `cs.RO` · **Submitted:** 2026-02-06
+- **Authors:** Harold Soh, Eugene Lim
+- **abs:** https://arxiv.org/abs/2602.06339 · **PDF:** https://arxiv.org/pdf/2602.06339v1
+
+## 14931. Can Post-Training Transform LLMs into Causal Reasoners?
+
+- **ID:** `2602.06337` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Junqi Chen, Sirui Chen, Chaochao Lu
+- **abs:** https://arxiv.org/abs/2602.06337 · **PDF:** https://arxiv.org/pdf/2602.06337v1
+
+## 14932. AdFL: In-Browser Federated Learning for Online Advertisement
+
+- **ID:** `2602.06336` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Ahmad Alemari, Pritam Sen, Cristian Borcea
+- **abs:** https://arxiv.org/abs/2602.06336 · **PDF:** https://arxiv.org/pdf/2602.06336v1
+
+## 14933. Don't Break the Boundary: Continual Unlearning for OOD Detection Based on Free Energy Repulsion
+
+- **ID:** `2602.06331` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Ningkang Peng, Kun Shao, Jingyang Mao, et al.
+- **abs:** https://arxiv.org/abs/2602.06331 · **PDF:** https://arxiv.org/pdf/2602.06331v1
+
+## 14934. The Doctor Will (Still) See You Now: On the Structural Limits of Agentic AI in Healthcare
+
+- **ID:** `2602.18460` · **primary:** `cs.CY` · **Submitted:** 2026-02-06
+- **Authors:** Gabriela Aránguiz Dias, Kiana Jafari, Allie Griffith, et al.
+- **abs:** https://arxiv.org/abs/2602.18460 · **PDF:** https://arxiv.org/pdf/2602.18460v1
+
+## 14935. Online Adaptive Reinforcement Learning with Echo State Networks for Non-Stationary Dynamics
+
+- **ID:** `2602.06326` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Aoi Yoshimura, Gouhei Tanaka
+- **abs:** https://arxiv.org/abs/2602.06326 · **PDF:** https://arxiv.org/pdf/2602.06326v1
+
+## 14936. How (Not) to Hybridize Neural and Mechanistic Models for Epidemiological Forecasting
+
+- **ID:** `2602.06323` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Yiqi Su, Ray Lee, Jiaming Cui, et al.
+- **abs:** https://arxiv.org/abs/2602.06323 · **PDF:** https://arxiv.org/pdf/2602.06323v1
+
+## 14937. High-Dimensional Limit of Stochastic Gradient Flow via Dynamical Mean-Field Theory
+
+- **ID:** `2602.06320` · **primary:** `stat.ML` · **Submitted:** 2026-02-06
+- **Authors:** Sota Nishiyama, Masaaki Imaizumi
+- **abs:** https://arxiv.org/abs/2602.06320 · **PDF:** https://arxiv.org/pdf/2602.06320v2
+
+## 14938. Exposing Weaknesses of Large Reasoning Models through Graph Algorithm Problems
+
+- **ID:** `2602.06319` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Qifan Zhang, Jianhao Ruan, Aochuan Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.06319 · **PDF:** https://arxiv.org/pdf/2602.06319v1
+
+## 14939. The Condensate Theorem: Transformers are O(n), Not $O(n^2)$
+
+- **ID:** `2602.06317` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Jorge L. Ruiz Williams
+- **abs:** https://arxiv.org/abs/2602.06317 · **PDF:** https://arxiv.org/pdf/2602.06317v2
+
+## 14940. Lost in Speech: Benchmarking, Evaluation, and Parsing of Spoken Code-Switching Beyond Standard UD Assumptions
+
+- **ID:** `2602.06307` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Nemika Tyagi, Holly Hendrix, Nelvin Licona-Guevara, et al.
+- **abs:** https://arxiv.org/abs/2602.06307 · **PDF:** https://arxiv.org/pdf/2602.06307v1
+
+## 14941. CALM: Class-Conditional Sparse Attention Vectors for Large Audio-Language Models
+
+- **ID:** `2602.07077` · **primary:** `cs.SD` · **Submitted:** 2026-02-06
+- **Authors:** Videet Mehta, Liming Wang, Hilde Kuehne, et al.
+- **abs:** https://arxiv.org/abs/2602.07077 · **PDF:** https://arxiv.org/pdf/2602.07077v2
+
+## 14942. Attention's Gravitational Field:A Power-Law Interpretation of Positional Correlation
+
+- **ID:** `2603.04805` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Edward Zhang
+- **abs:** https://arxiv.org/abs/2603.04805 · **PDF:** https://arxiv.org/pdf/2603.04805v1
+
+## 14943. Accelerating Vision Transformers on Brain Processing Unit
+
+- **ID:** `2602.06300` · **primary:** `cs.CV` · **Submitted:** 2026-02-06
+- **Authors:** Jinchi Tang, Yan Guo
+- **abs:** https://arxiv.org/abs/2602.06300 · **PDF:** https://arxiv.org/pdf/2602.06300v1
+
+## 14944. Time-uniform conformal and PAC prediction
+
+- **ID:** `2602.06297` · **primary:** `stat.ML` · **Submitted:** 2026-02-06
+- **Authors:** Kayla E. Scharfstein, Arun Kumar Kuchibhotla
+- **abs:** https://arxiv.org/abs/2602.06297 · **PDF:** https://arxiv.org/pdf/2602.06297v1
+
+## 14945. LatentChem: From Textual CoT to Latent Thinking in Chemical Reasoning
+
+- **ID:** `2602.07075` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-06
+- **Authors:** Xinwu Ye, Yicheng Mao, Jia Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.07075 · **PDF:** https://arxiv.org/pdf/2602.07075v4
+
+## 14946. Robust Pre-Training of Medical Vision-and-Language Models with Domain-Invariant Multi-Modal Masked Reconstruction
+
+- **ID:** `2602.17689` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Melika Filvantorkaman, Mohsen Piri
+- **abs:** https://arxiv.org/abs/2602.17689 · **PDF:** https://arxiv.org/pdf/2602.17689v1
+
+## 14947. Judging What We Cannot Solve: A Consequence-Based Approach for Oracle-Free Evaluation of Research-Level Math
+
+- **ID:** `2602.06291` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Guijin Son, Donghun Yang, Hitesh Laxmichand Patel, et al.
+- **abs:** https://arxiv.org/abs/2602.06291 · **PDF:** https://arxiv.org/pdf/2602.06291v1
+
+## 14948. Toward generative machine learning for boosting ensembles of climate simulations
+
+- **ID:** `2602.06287` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Parsa Gooya, Reinel Sospedra-Alfonso, Johannes Exenberger
+- **abs:** https://arxiv.org/abs/2602.06287 · **PDF:** https://arxiv.org/pdf/2602.06287v1
+
+## 14949. Do LLMs Act Like Rational Agents? Measuring Belief Coherence in Probabilistic Decision Making
+
+- **ID:** `2602.06286` · **primary:** `cs.AI` · **Submitted:** 2026-02-06
+- **Authors:** Khurram Yamin, Jingjing Tang, Santiago Cortes-Gomez, et al.
+- **abs:** https://arxiv.org/abs/2602.06286 · **PDF:** https://arxiv.org/pdf/2602.06286v1
+
+## 14950. SOCKET: SOft Collison Kernel EsTimator for Sparse Attention
+
+- **ID:** `2602.06283` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Sahil Joshi, Agniva Chowdhury, Wyatt Bellinger, et al.
+- **abs:** https://arxiv.org/abs/2602.06283 · **PDF:** https://arxiv.org/pdf/2602.06283v1
+
+## 14951. Pro-ZD: A Transferable Graph Neural Network Approach for Proactive Zero-Day Threats Mitigation
+
+- **ID:** `2602.07073` · **primary:** `cs.CR` · **Submitted:** 2026-02-06
+- **Authors:** Nardine Basta, Firas Ben Hmida, Houssem Jmal, et al.
+- **abs:** https://arxiv.org/abs/2602.07073 · **PDF:** https://arxiv.org/pdf/2602.07073v1
+
+## 14952. Statistical Learning from Attribution Sets
+
+- **ID:** `2602.06276` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Lorne Applebaum, Robert Busa-Fekete, August Y. Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.06276 · **PDF:** https://arxiv.org/pdf/2602.06276v1
+
+## 14953. RoPE-LIME: RoPE-Space Locality + Sparse-K Sampling for Efficient LLM Attribution
+
+- **ID:** `2602.06275` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Isaac Picov, Ritesh Goru
+- **abs:** https://arxiv.org/abs/2602.06275 · **PDF:** https://arxiv.org/pdf/2602.06275v2
+
+## 14954. One Bias After Another: Mechanistic Reward Shaping and Persistent Biases in Language Reward Models
+
+- **ID:** `2603.03291` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Daniel Fein, Max Lamparth, Violet Xiang, et al.
+- **abs:** https://arxiv.org/abs/2603.03291 · **PDF:** https://arxiv.org/pdf/2603.03291v1
+
+## 14955. VowelPrompt: Hearing Speech Emotions from Text via Vowel-level Prosodic Augmentation
+
+- **ID:** `2602.06270` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Yancheng Wang, Osama Hanna, Ruiming Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.06270 · **PDF:** https://arxiv.org/pdf/2602.06270v1
+
+## 14956. PurSAMERE: Reliable Adversarial Purification via Sharpness-Aware Minimization of Expected Reconstruction Error
+
+- **ID:** `2602.06269` · **primary:** `cs.LG` · **Submitted:** 2026-02-06
+- **Authors:** Vinh Hoang, Sebastian Krumscheid, Holger Rauhut, et al.
+- **abs:** https://arxiv.org/abs/2602.06269 · **PDF:** https://arxiv.org/pdf/2602.06269v1
+
+## 14957. MPIB: A Benchmark for Medical Prompt Injection Attacks and Clinical Safety in LLMs
+
+- **ID:** `2602.06268` · **primary:** `cs.CL` · **Submitted:** 2026-02-06
+- **Authors:** Junhyeok Lee, Han Jang, Kyu Sung Choi
+- **abs:** https://arxiv.org/abs/2602.06268 · **PDF:** https://arxiv.org/pdf/2602.06268v1
+
+## 14958. Conformal changepoint localization
+
+- **ID:** `2602.06267` · **primary:** `stat.ME` · **Submitted:** 2026-02-05
+- **Authors:** Rohan Hore, Aaditya Ramdas
+- **abs:** https://arxiv.org/abs/2602.06267 · **PDF:** https://arxiv.org/pdf/2602.06267v1
+
+## 14959. Is my model "mind blurting"? Interpreting the dynamics of reasoning tokens with Recurrence Quantification Analysis (RQA)
+
+- **ID:** `2602.06266` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Quoc Tuan Pham, Mehdi Jafari, Flora Salim
+- **abs:** https://arxiv.org/abs/2602.06266 · **PDF:** https://arxiv.org/pdf/2602.06266v1
+
+## 14960. Swap Regret Minimization Through Response-Based Approachability
+
+- **ID:** `2602.06264` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Ioannis Anagnostides, Gabriele Farina, Maxwell Fishelson, et al.
+- **abs:** https://arxiv.org/abs/2602.06264 · **PDF:** https://arxiv.org/pdf/2602.06264v1
+
+## 14961. Can One-sided Arguments Lead to Response Change in Large Language Models?
+
+- **ID:** `2602.06260` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Pedro Cisneros-Velarde
+- **abs:** https://arxiv.org/abs/2602.06260 · **PDF:** https://arxiv.org/pdf/2602.06260v1
+
+## 14962. GRP-Obliteration: Unaligning LLMs With a Single Unlabeled Prompt
+
+- **ID:** `2602.06258` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Mark Russinovich, Yanan Cai, Keegan Hines, et al.
+- **abs:** https://arxiv.org/abs/2602.06258 · **PDF:** https://arxiv.org/pdf/2602.06258v1
+
+## 14963. On Randomized Algorithms in Online Strategic Classification
+
+- **ID:** `2602.06257` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Chase Hutton, Adam Melrod, Han Shao
+- **abs:** https://arxiv.org/abs/2602.06257 · **PDF:** https://arxiv.org/pdf/2602.06257v1
+
+## 14964. Steering Safely or Off a Cliff? Rethinking Specificity and Robustness in Inference-Time Interventions
+
+- **ID:** `2602.06256` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Navita Goyal, Hal Daumé
+- **abs:** https://arxiv.org/abs/2602.06256 · **PDF:** https://arxiv.org/pdf/2602.06256v1
+
+## 14965. ASMa: Asymmetric Spatio-temporal Masking for Skeleton Action Representation Learning
+
+- **ID:** `2602.06251` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Aman Anand, Amir Eskandari, Elyas Rahsno, et al.
+- **abs:** https://arxiv.org/abs/2602.06251 · **PDF:** https://arxiv.org/pdf/2602.06251v1
+
+## 14966. REBEL: Hidden Knowledge Recovery via Evolutionary-Based Evaluation Loop
+
+- **ID:** `2602.06248` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Patryk Rybak, Paweł Batorski, Paul Swoboda, et al.
+- **abs:** https://arxiv.org/abs/2602.06248 · **PDF:** https://arxiv.org/pdf/2602.06248v1
+
+## 14967. Adaptive Sparse Möbius Transforms for Learning Polynomials
+
+- **ID:** `2602.06246` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Yigit Efe Erginbas, Justin Singh Kang, Elizabeth Polito, et al.
+- **abs:** https://arxiv.org/abs/2602.06246 · **PDF:** https://arxiv.org/pdf/2602.06246v1
+
+## 14968. Inheritance Between Feedforward and Convolutional Networks via Model Projection
+
+- **ID:** `2602.06245` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Nicolas Ewen, Jairo Diaz-Rodriguez, Kelly Ramsay
+- **abs:** https://arxiv.org/abs/2602.06245 · **PDF:** https://arxiv.org/pdf/2602.06245v1
+
+## 14969. AnCoder: Anchored Code Generation via Discrete Diffusion Models
+
+- **ID:** `2602.17688` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Anton Xue, Litu Rout, Constantine Caramanis, et al.
+- **abs:** https://arxiv.org/abs/2602.17688 · **PDF:** https://arxiv.org/pdf/2602.17688v1
+
+## 14970. A Fast and Generalizable Fourier Neural Operator-Based Surrogate for Melt-Pool Prediction in Laser Processing
+
+- **ID:** `2602.06241` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Alix Benoit, Toni Ivas, Mateusz Papierz, et al.
+- **abs:** https://arxiv.org/abs/2602.06241 · **PDF:** https://arxiv.org/pdf/2602.06241v3
+
+## 14971. ATEX-CF: Attack-Informed Counterfactual Explanations for Graph Neural Networks
+
+- **ID:** `2602.06240` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Yu Zhang, Sean Bin Yang, Arijit Khan, et al.
+- **abs:** https://arxiv.org/abs/2602.06240 · **PDF:** https://arxiv.org/pdf/2602.06240v1
+
+## 14972. Provably avoiding over-optimization in Direct Preference Optimization without knowing the data distribution
+
+- **ID:** `2602.06239` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Adam Barla, Emanuele Nevali, Luca Viano, et al.
+- **abs:** https://arxiv.org/abs/2602.06239 · **PDF:** https://arxiv.org/pdf/2602.06239v1
+
+## 14973. RuleSmith: Multi-Agent LLMs for Automated Game Balancing
+
+- **ID:** `2602.06232` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Ziyao Zeng, Chen Liu, Tianyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.06232 · **PDF:** https://arxiv.org/pdf/2602.06232v1
+
+## 14974. AriadneMem: Threading the Maze of Lifelong Memory for LLM Agents
+
+- **ID:** `2603.03290` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Wenhui Zhu, Xiwen Chen, Zhipeng Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.03290 · **PDF:** https://arxiv.org/pdf/2603.03290v1
+
+## 14975. SR4-Fit: An Interpretable and Informative Classification Algorithm Applied to Prediction of U.S. House of Representatives Elections
+
+- **ID:** `2602.06229` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Shyam Sundar Murali Krishnan, Dean Frederick Hougen
+- **abs:** https://arxiv.org/abs/2602.06229 · **PDF:** https://arxiv.org/pdf/2602.06229v1
+
+## 14976. Do It for HER: First-Order Temporal Logic Reward Specification in Reinforcement Learning (Extended Version)
+
+- **ID:** `2602.06227` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Pierriccardo Olivieri, Fausto Lasca, Alessandro Gianola, et al.
+- **abs:** https://arxiv.org/abs/2602.06227 · **PDF:** https://arxiv.org/pdf/2602.06227v1
+
+## 14977. BenchMarker: An Education-Inspired Toolkit for Highlighting Flaws in Multiple-Choice Benchmarks
+
+- **ID:** `2602.06221` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Nishant Balepur, Bhavya Rajasekaran, Jane Oh, et al.
+- **abs:** https://arxiv.org/abs/2602.06221 · **PDF:** https://arxiv.org/pdf/2602.06221v1
+
+## 14978. IRPAPERS: A Visual Document Benchmark for Scientific Retrieval and Question Answering
+
+- **ID:** `2602.17687` · **primary:** `cs.IR` · **Submitted:** 2026-02-05
+- **Authors:** Connor Shorten, Augustas Skaburskas, Daniel M. Jones, et al.
+- **abs:** https://arxiv.org/abs/2602.17687 · **PDF:** https://arxiv.org/pdf/2602.17687v1
+
+## 14979. Coupled Local and Global World Models for Efficient First Order RL
+
+- **ID:** `2602.06219` · **primary:** `cs.RO` · **Submitted:** 2026-02-05
+- **Authors:** Joseph Amigo, Rooholla Khorrambakht, Nicolas Mansard, et al.
+- **abs:** https://arxiv.org/abs/2602.06219 · **PDF:** https://arxiv.org/pdf/2602.06219v1
+
+## 14980. Cross-Modal Redundancy and the Geometry of Vision-Language Embeddings
+
+- **ID:** `2602.06218` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Grégoire Dhimoïla, Thomas Fel, Victor Boutin, et al.
+- **abs:** https://arxiv.org/abs/2602.06218 · **PDF:** https://arxiv.org/pdf/2602.06218v2
+
+## 14981. Addressing the Waypoint-Action Gap in End-to-End Autonomous Driving via Vehicle Motion Models
+
+- **ID:** `2602.06214` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Jorge Daniel Rodríguez-Vidal, Gabriel Villalonga, Diego Porres, et al.
+- **abs:** https://arxiv.org/abs/2602.06214 · **PDF:** https://arxiv.org/pdf/2602.06214v2
+
+## 14982. Emergent Low-Rank Training Dynamics in MLPs with Smooth Activations
+
+- **ID:** `2602.06208` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Alec S. Xu, Can Yaras, Matthew Asato, et al.
+- **abs:** https://arxiv.org/abs/2602.06208 · **PDF:** https://arxiv.org/pdf/2602.06208v1
+
+## 14983. Multi-Way Representation Alignment
+
+- **ID:** `2602.06205` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Akshit Achara, Tatiana Gaintseva, Mateo Mahaut, et al.
+- **abs:** https://arxiv.org/abs/2602.06205 · **PDF:** https://arxiv.org/pdf/2602.06205v1
+
+## 14984. Learning Rate Scaling across LoRA Ranks and Transfer to Full Finetuning
+
+- **ID:** `2602.06204` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Nan Chen, Soledad Villar, Soufiane Hayou
+- **abs:** https://arxiv.org/abs/2602.06204 · **PDF:** https://arxiv.org/pdf/2602.06204v1
+
+## 14985. AnyThermal: Towards Learning Universal Representations for Thermal Perception
+
+- **ID:** `2602.06203` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Parv Maheshwari, Jay Karhade, Yogesh Chawla, et al.
+- **abs:** https://arxiv.org/abs/2602.06203 · **PDF:** https://arxiv.org/pdf/2602.06203v1
+
+## 14986. From Bias Mitigation to Bias Negotiation: Governing Identity and Sociocultural Reasoning in Generative AI
+
+- **ID:** `2602.18459` · **primary:** `cs.CY` · **Submitted:** 2026-02-05
+- **Authors:** Zackary Okun Dunivin, Bingyi Han, John Bollenbocher
+- **abs:** https://arxiv.org/abs/2602.18459 · **PDF:** https://arxiv.org/pdf/2602.18459v1
+
+## 14987. Personagram: Bridging Personas and Product Design for Creative Ideation with Multimodal LLMs
+
+- **ID:** `2602.06197` · **primary:** `cs.HC` · **Submitted:** 2026-02-05
+- **Authors:** Taewook Kim, Matthew K. Hong, Yan-Ying Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.06197 · **PDF:** https://arxiv.org/pdf/2602.06197v1
+
+## 14988. Generics in science communication: Misaligned interpretations across laypeople, scientists, and large language models
+
+- **ID:** `2602.06190` · **primary:** `cs.HC` · **Submitted:** 2026-02-05
+- **Authors:** Uwe Peters, Andrea Bertazzoli, Jasmine M. DeJesus, et al.
+- **abs:** https://arxiv.org/abs/2602.06190 · **PDF:** https://arxiv.org/pdf/2602.06190v1
+
+## 14989. Artificial Intelligence in Open Source Software Engineering: A Foundation for Sustainability
+
+- **ID:** `2602.07071` · **primary:** `cs.SE` · **Submitted:** 2026-02-05
+- **Authors:** S M Rakib UI Karim, Wenyi Lu, Sean Goggins
+- **abs:** https://arxiv.org/abs/2602.07071 · **PDF:** https://arxiv.org/pdf/2602.07071v1
+
+## 14990. $f$-FUM: Federated Unlearning via min--max and $f$-divergence
+
+- **ID:** `2602.06187` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Radmehr Karimian, Amirhossein Bagheri, Meghdad Kurmanji, et al.
+- **abs:** https://arxiv.org/abs/2602.06187 · **PDF:** https://arxiv.org/pdf/2602.06187v1
+
+## 14991. PhenoLIP: Integrating Phenotype Ontology Knowledge into Medical Vision-Language Pretraining
+
+- **ID:** `2602.06184` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Cheng Liang, Chaoyi Wu, Weike Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.06184 · **PDF:** https://arxiv.org/pdf/2602.06184v1
+
+## 14992. To 2:4 Sparsity and Beyond: Neuron-level Activation Function to Accelerate LLM Pre-Training
+
+- **ID:** `2602.06183` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Meghana Madhyastha, Daniel Haziza, Jesse Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.06183 · **PDF:** https://arxiv.org/pdf/2602.06183v1
+
+## 14993. Uncertainty Drives Social Bias Changes in Quantized Large Language Models
+
+- **ID:** `2602.06181` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Stanley Z. Hua, Sanae Lotfi, Irene Y. Chen
+- **abs:** https://arxiv.org/abs/2602.06181 · **PDF:** https://arxiv.org/pdf/2602.06181v1
+
+## 14994. STACodec: Semantic Token Assignment for Balancing Acoustic Fidelity and Semantic Information in Audio Codecs
+
+- **ID:** `2602.06180` · **primary:** `eess.AS` · **Submitted:** 2026-02-05
+- **Authors:** Kaiyuan Zhang, Mohan Shi, Eray Eren, et al.
+- **abs:** https://arxiv.org/abs/2602.06180 · **PDF:** https://arxiv.org/pdf/2602.06180v1
+
+## 14995. Large Language Model Reasoning Failures
+
+- **ID:** `2602.06176` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Peiyang Song, Pengrui Han, Noah Goodman
+- **abs:** https://arxiv.org/abs/2602.06176 · **PDF:** https://arxiv.org/pdf/2602.06176v1
+
+## 14996. Optimal rates for density and mode estimation with expand-and-sparsify representations
+
+- **ID:** `2602.06175` · **primary:** `math.ST` · **Submitted:** 2026-02-05
+- **Authors:** Kaushik Sinha, Christopher Tosh
+- **abs:** https://arxiv.org/abs/2602.06175 · **PDF:** https://arxiv.org/pdf/2602.06175v2
+
+## 14997. Know Your Scientist: KYC as Biosecurity Infrastructure
+
+- **ID:** `2602.06172` · **primary:** `cs.CR` · **Submitted:** 2026-02-05
+- **Authors:** Jonathan Feldman, Tal Feldman, Annie I Anton
+- **abs:** https://arxiv.org/abs/2602.06172 · **PDF:** https://arxiv.org/pdf/2602.06172v1
+
+## 14998. Hybrid Dual-Path Linear Transformations for Efficient Transformer Architectures
+
+- **ID:** `2602.07070` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Vladimer Khasia
+- **abs:** https://arxiv.org/abs/2602.07070 · **PDF:** https://arxiv.org/pdf/2602.07070v1
+
+## 14999. Stop the Flip-Flop: Context-Preserving Verification for Fast Revocable Diffusion Decoding
+
+- **ID:** `2602.06161` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Yanzheng Xiang, Lan Wei, Yizhen Yao, et al.
+- **abs:** https://arxiv.org/abs/2602.06161 · **PDF:** https://arxiv.org/pdf/2602.06161v1
+
+## 15000. SCONE: A Practical, Constraint-Aware Plug-in for Latent Encoding in Learned DNA Storage
+
+- **ID:** `2602.06157` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Cihan Ruan, Lebin Zhou, Rongduo Han, et al.
+- **abs:** https://arxiv.org/abs/2602.06157 · **PDF:** https://arxiv.org/pdf/2602.06157v1
+
+## 15001. Latent Structure Emergence in Diffusion Models via Confidence-Based Filtering
+
+- **ID:** `2602.06155` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Wei Wei, Yizhou Zeng, Kuntian Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.06155 · **PDF:** https://arxiv.org/pdf/2602.06155v1
+
+## 15002. MoSE: Mixture of Slimmable Experts for Efficient and Adaptive Language Models
+
+- **ID:** `2602.06154` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Nurbek Tastan, Stefanos Laskaridis, Karthik Nandakumar, et al.
+- **abs:** https://arxiv.org/abs/2602.06154 · **PDF:** https://arxiv.org/pdf/2602.06154v1
+
+## 15003. Optimistic Training and Convergence of Q-Learning -- Extended Version
+
+- **ID:** `2602.06146` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Prashant Mehta, Sean Meyn
+- **abs:** https://arxiv.org/abs/2602.06146 · **PDF:** https://arxiv.org/pdf/2602.06146v1
+
+## 15004. Protean Compiler: An Agile Framework to Drive Fine-grain Phase Ordering
+
+- **ID:** `2602.06142` · **primary:** `cs.PL` · **Submitted:** 2026-02-05
+- **Authors:** Amir H. Ashouri, Shayan Shirahmad Gale Bagi, Kavin Satheeskumar, et al.
+- **abs:** https://arxiv.org/abs/2602.06142 · **PDF:** https://arxiv.org/pdf/2602.06142v2
+
+## 15005. Bidirectional Reward-Guided Diffusion for Real-World Image Super-Resolution
+
+- **ID:** `2602.07069` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Zihao Fan, Xin Lu, Yidi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.07069 · **PDF:** https://arxiv.org/pdf/2602.07069v1
+
+## 15006. Flow Matching for Offline Reinforcement Learning with Discrete Actions
+
+- **ID:** `2602.06138` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Fairoz Nower Khan, Nabuat Zaman Nahim, Ruiquan Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.06138 · **PDF:** https://arxiv.org/pdf/2602.06138v1
+
+## 15007. Warm Starts, Cold States: Exploiting Adiabaticity for Variational Ground-States
+
+- **ID:** `2602.06137` · **primary:** `quant-ph` · **Submitted:** 2026-02-05
+- **Authors:** Ricard Puig, Berta Casas, Alba Cervera-Lierta, et al.
+- **abs:** https://arxiv.org/abs/2602.06137 · **PDF:** https://arxiv.org/pdf/2602.06137v1
+
+## 15008. Tempora: Characterising the Time-Contingent Utility of Online Test-Time Adaptation
+
+- **ID:** `2602.06136` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Sudarshan Sreeram, Young D. Kwon, Cecilia Mascolo
+- **abs:** https://arxiv.org/abs/2602.06136 · **PDF:** https://arxiv.org/pdf/2602.06136v1
+
+## 15009. Hear You in Silence: Designing for Active Listening in Human Interaction with Conversational Agents Using Context-Aware Pacing
+
+- **ID:** `2602.06134` · **primary:** `cs.HC` · **Submitted:** 2026-02-05
+- **Authors:** Zhihan Jiang, Qianhui Chen, Chu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.06134 · **PDF:** https://arxiv.org/pdf/2602.06134v1
+
+## 15010. MRI Cross-Modal Synthesis: A Comparative Study of Generative Models for T1-to-T2 Reconstruction
+
+- **ID:** `2602.07068` · **primary:** `eess.IV` · **Submitted:** 2026-02-05
+- **Authors:** Ali Alqutayfi, Sadam Al-Azani
+- **abs:** https://arxiv.org/abs/2602.07068 · **PDF:** https://arxiv.org/pdf/2602.07068v1
+
+## 15011. Self-Improving World Modelling with Latent Actions
+
+- **ID:** `2602.06130` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Yifu Qiu, Zheng Zhao, Waylon Li, et al.
+- **abs:** https://arxiv.org/abs/2602.06130 · **PDF:** https://arxiv.org/pdf/2602.06130v2
+
+## 15012. Urban Spatio-Temporal Foundation Models for Climate-Resilient Housing: Scaling Diffusion Transformers for Disaster Risk Prediction
+
+- **ID:** `2602.06129` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Olaf Yunus Laitinen Imanov, Derya Umut Kulali, Taner Yilmaz
+- **abs:** https://arxiv.org/abs/2602.06129 · **PDF:** https://arxiv.org/pdf/2602.06129v1
+
+## 15013. Compressing LLMs with MoP: Mixture of Pruners
+
+- **ID:** `2602.06127` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Bruno Lopes Yamamoto, Lucas Lauton de Alcantara, Victor Zacarias, et al.
+- **abs:** https://arxiv.org/abs/2602.06127 · **PDF:** https://arxiv.org/pdf/2602.06127v1
+
+## 15014. The Story is Not the Science: Execution-Grounded Evaluation of Mechanistic Interpretability Research
+
+- **ID:** `2602.18458` · **primary:** `cs.CY` · **Submitted:** 2026-02-05
+- **Authors:** Xiaoyan Bai, Alexander Baumgartner, Haojia Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.18458 · **PDF:** https://arxiv.org/pdf/2602.18458v1
+
+## 15015. Private and interpretable clinical prediction with quantum-inspired tensor train models
+
+- **ID:** `2602.06110` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** José Ramón Pareja Monturiol, Juliette Sinnott, Roger G. Melko, et al.
+- **abs:** https://arxiv.org/abs/2602.06110 · **PDF:** https://arxiv.org/pdf/2602.06110v1
+
+## 15016. Shared LoRA Subspaces for almost Strict Continual Learning
+
+- **ID:** `2602.06043` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Prakhar Kaushik, Ankit Vaidya, Shravan Chaudhari, et al.
+- **abs:** https://arxiv.org/abs/2602.06043 · **PDF:** https://arxiv.org/pdf/2602.06043v1
+
+## 15017. Pseudo-Invertible Neural Networks
+
+- **ID:** `2602.06042` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Yamit Ehrlich, Nimrod Berman, Assaf Shocher
+- **abs:** https://arxiv.org/abs/2602.06042 · **PDF:** https://arxiv.org/pdf/2602.06042v1
+
+## 15018. DyTopo: Dynamic Topology Routing for Multi-Agent Reasoning via Semantic Matching
+
+- **ID:** `2602.06039` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Yuxing Lu, Yucheng Hu, Xukai Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.06039 · **PDF:** https://arxiv.org/pdf/2602.06039v1
+
+## 15019. CommCP: Efficient Multi-Agent Coordination via LLM-Based Communication with Conformal Prediction
+
+- **ID:** `2602.06038` · **primary:** `cs.RO` · **Submitted:** 2026-02-05
+- **Authors:** Xiaopan Zhang, Zejin Wang, Zhixu Li, et al.
+- **abs:** https://arxiv.org/abs/2602.06038 · **PDF:** https://arxiv.org/pdf/2602.06038v1
+
+## 15020. DFlash: Block Diffusion for Flash Speculative Decoding
+
+- **ID:** `2602.06036` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Jian Chen, Yesheng Liang, Zhijian Liu
+- **abs:** https://arxiv.org/abs/2602.06036 · **PDF:** https://arxiv.org/pdf/2602.06036v1
+
+## 15021. Can vision language models learn intuitive physics from interaction?
+
+- **ID:** `2602.06033` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Luca M. Schulze Buschoff, Konstantinos Voudouris, Can Demircan, et al.
+- **abs:** https://arxiv.org/abs/2602.06033 · **PDF:** https://arxiv.org/pdf/2602.06033v1
+
+## 15022. PhysicsAgentABM: Physics-Guided Generative Agent-Based Modeling
+
+- **ID:** `2602.06030` · **primary:** `cs.MA` · **Submitted:** 2026-02-05
+- **Authors:** Kavana Venkatesh, Yinhan He, Jundong Li, et al.
+- **abs:** https://arxiv.org/abs/2602.06030 · **PDF:** https://arxiv.org/pdf/2602.06030v2
+
+## 15023. AP-OOD: Attention Pooling for Out-of-Distribution Detection
+
+- **ID:** `2602.06031` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Claus Hofmann, Christian Huber, Bernhard Lehner, et al.
+- **abs:** https://arxiv.org/abs/2602.06031 · **PDF:** https://arxiv.org/pdf/2602.06031v1
+
+## 15024. Curiosity is Knowledge: Self-Consistent Learning and No-Regret Optimization with Active Inference
+
+- **ID:** `2602.06029` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Yingke Li, Anjali Parashar, Enlu Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.06029 · **PDF:** https://arxiv.org/pdf/2602.06029v1
+
+## 15025. Learning Query-Aware Budget-Tier Routing for Runtime Agent Memory
+
+- **ID:** `2602.06025` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Haozhen Zhang, Haodong Yue, Tao Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.06025 · **PDF:** https://arxiv.org/pdf/2602.06025v1
+
+## 15026. Jackpot: Optimal Budgeted Rejection Sampling for Extreme Actor-Policy Mismatch Reinforcement Learning
+
+- **ID:** `2602.06107` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Zhuoming Chen, Hongyi Liu, Yang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.06107 · **PDF:** https://arxiv.org/pdf/2602.06107v1
+
+## 15027. Developing a Discrete-Event Simulator of School Shooter Behavior from VR Data
+
+- **ID:** `2602.06023` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Christopher A. McClurg, Alan R. Wagner
+- **abs:** https://arxiv.org/abs/2602.06023 · **PDF:** https://arxiv.org/pdf/2602.06023v2
+
+## 15028. Correctness-Optimized Residual Activation Lens (CORAL): Transferrable and Calibration-Aware Inference-Time Steering
+
+- **ID:** `2602.06022` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Miranda Muqing Miao, Young-Min Cho, Lyle Ungar
+- **abs:** https://arxiv.org/abs/2602.06022 · **PDF:** https://arxiv.org/pdf/2602.06022v1
+
+## 15029. Diffusion Model's Generalization Can Be Characterized by Inductive Biases toward a Data-Dependent Ridge Manifold
+
+- **ID:** `2602.06021` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Ye He, Yitong Qiu, Molei Tao
+- **abs:** https://arxiv.org/abs/2602.06021 · **PDF:** https://arxiv.org/pdf/2602.06021v1
+
+## 15030. Mechanisms of AI Protein Folding in ESMFold
+
+- **ID:** `2602.06020` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Kevin Lu, Jannik Brinkmann, Stefan Huber, et al.
+- **abs:** https://arxiv.org/abs/2602.06020 · **PDF:** https://arxiv.org/pdf/2602.06020v2
+
+## 15031. Multi-Token Prediction via Self-Distillation
+
+- **ID:** `2602.06019` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** John Kirchenbauer, Abhimanyu Hans, Brian Bartoldson, et al.
+- **abs:** https://arxiv.org/abs/2602.06019 · **PDF:** https://arxiv.org/pdf/2602.06019v1
+
+## 15032. A Systematic Evaluation of Large Language Models for PTSD Severity Estimation: The Role of Contextual Knowledge and Modeling Strategies
+
+- **ID:** `2602.06015` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Panagiotis Kaliosis, Adithya V Ganesan, Oscar N. E. Kjell, et al.
+- **abs:** https://arxiv.org/abs/2602.06015 · **PDF:** https://arxiv.org/pdf/2602.06015v1
+
+## 15033. Optimism Stabilizes Thompson Sampling for Adaptive Inference
+
+- **ID:** `2602.06014` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Shunxing Yan, Han Zhong
+- **abs:** https://arxiv.org/abs/2602.06014 · **PDF:** https://arxiv.org/pdf/2602.06014v1
+
+## 15034. GenArena: How Can We Achieve Human-Aligned Evaluation for Visual Generation Tasks?
+
+- **ID:** `2602.06013` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Ruihang Li, Leigang Qu, Jingxu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.06013 · **PDF:** https://arxiv.org/pdf/2602.06013v1
+
+## 15035. AgenticPay: A Multi-Agent LLM Negotiation System for Buyer-Seller Transactions
+
+- **ID:** `2602.06008` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Xianyang Liu, Shangding Gu, Dawn Song
+- **abs:** https://arxiv.org/abs/2602.06008 · **PDF:** https://arxiv.org/pdf/2602.06008v1
+
+## 15036. Speech Emotion Recognition Leveraging OpenAI's Whisper Representations and Attentive Pooling Methods
+
+- **ID:** `2602.06000` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Ali Shendabadi, Parnia Izadirad, Mostafa Salehi, et al.
+- **abs:** https://arxiv.org/abs/2602.06000 · **PDF:** https://arxiv.org/pdf/2602.06000v1
+
+## 15037. On the Role of Iterative Computation in Reinforcement Learning
+
+- **ID:** `2602.05999` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Raj Ghugare, Michał Bortkiewicz, Alicja Ziarko, et al.
+- **abs:** https://arxiv.org/abs/2602.05999 · **PDF:** https://arxiv.org/pdf/2602.05999v2
+
+## 15038. Algebraic Robustness Verification of Neural Networks
+
+- **ID:** `2602.06105` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Yulia Alexandr, Hao Duan, Guido Montúfar
+- **abs:** https://arxiv.org/abs/2602.06105 · **PDF:** https://arxiv.org/pdf/2602.06105v1
+
+## 15039. Causal Inference on Stopped Random Walks in Online Advertising
+
+- **ID:** `2602.05997` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Jia Yuan Yu
+- **abs:** https://arxiv.org/abs/2602.05997 · **PDF:** https://arxiv.org/pdf/2602.05997v1
+
+## 15040. Orthogonal Self-Attention
+
+- **ID:** `2602.05996` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Leo Zhang, James Martens
+- **abs:** https://arxiv.org/abs/2602.05996 · **PDF:** https://arxiv.org/pdf/2602.05996v1
+
+## 15041. Pragmatic Curiosity: A Hybrid Learning-Optimization Paradigm via Active Inference
+
+- **ID:** `2602.06104` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Yingke Li, Anjali Parashar, Enlu Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.06104 · **PDF:** https://arxiv.org/pdf/2602.06104v1
+
+## 15042. Diamond Maps: Efficient Reward Alignment via Stochastic Flow Maps
+
+- **ID:** `2602.05993` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Peter Holderrieth, Douglas Chen, Luca Eyring, et al.
+- **abs:** https://arxiv.org/abs/2602.05993 · **PDF:** https://arxiv.org/pdf/2602.05993v1
+
+## 15043. DSB: Dynamic Sliding Block Scheduling for Diffusion LLMs
+
+- **ID:** `2602.05992` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Lizhuo Luo, Shenggui Li, Yonggang Wen, et al.
+- **abs:** https://arxiv.org/abs/2602.05992 · **PDF:** https://arxiv.org/pdf/2602.05992v2
+
+## 15044. Layer-wise LoRA fine-tuning: a similarity metric approach
+
+- **ID:** `2602.05988` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Keith Ando Ogawa, Bruno Lopes Yamamoto, Lucas Lauton de Alcantara, et al.
+- **abs:** https://arxiv.org/abs/2602.05988 · **PDF:** https://arxiv.org/pdf/2602.05988v1
+
+## 15045. RISE-Video: Can Video Generators Decode Implicit World Rules?
+
+- **ID:** `2602.05986` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Mingxin Liu, Shuran Ma, Shibei Meng, et al.
+- **abs:** https://arxiv.org/abs/2602.05986 · **PDF:** https://arxiv.org/pdf/2602.05986v1
+
+## 15046. Geographically-aware Transformer-based Traffic Forecasting for Urban Motorway Digital Twins
+
+- **ID:** `2602.05983` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Krešimir Kušić, Vinny Cahill, Ivana Dusparic
+- **abs:** https://arxiv.org/abs/2602.05983 · **PDF:** https://arxiv.org/pdf/2602.05983v1
+
+## 15047. Clifford Kolmogorov-Arnold Networks
+
+- **ID:** `2602.05977` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Matthias Wolff, Francesco Alesiani, Christof Duhme, et al.
+- **abs:** https://arxiv.org/abs/2602.05977 · **PDF:** https://arxiv.org/pdf/2602.05977v1
+
+## 15048. SAGE: Benchmarking and Improving Retrieval for Deep Research Agents
+
+- **ID:** `2602.05975` · **primary:** `cs.IR` · **Submitted:** 2026-02-05
+- **Authors:** Tiansheng Hu, Yilun Zhao, Canyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.05975 · **PDF:** https://arxiv.org/pdf/2602.05975v2
+
+## 15049. Characterizing Human Semantic Navigation in Concept Production as Trajectories in Embedding Space
+
+- **ID:** `2602.05971` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Felipe D. Toro-Hernández, Jesuino Vieira Filho, Rodrigo M. Cabral-Carvalho
+- **abs:** https://arxiv.org/abs/2602.05971 · **PDF:** https://arxiv.org/pdf/2602.05971v1
+
+## 15050. Inverse Depth Scaling From Most Layers Being Similar
+
+- **ID:** `2602.05970` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Yizhou Liu, Sara Kangaslahti, Ziming Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.05970 · **PDF:** https://arxiv.org/pdf/2602.05970v1
+
+## 15051. Toward Faithful and Complete Answer Construction from a Single Document
+
+- **ID:** `2602.06103` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Zhaoyang Chen, Cody Fleming
+- **abs:** https://arxiv.org/abs/2602.06103 · **PDF:** https://arxiv.org/pdf/2602.06103v1
+
+## 15052. A Hybrid Data-Driven Algorithm for Real-Time Friction Force Estimation in Hydraulic Cylinders
+
+- **ID:** `2602.05967` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Mohamad Amin Jamshidi, Mehrbod Zarifi, Zolfa Anvari, et al.
+- **abs:** https://arxiv.org/abs/2602.05967 · **PDF:** https://arxiv.org/pdf/2602.05967v1
+
+## 15053. LSA: Localized Semantic Alignment for Enhancing Temporal Consistency in Traffic Video Generation
+
+- **ID:** `2602.05966` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Mirlan Karimov, Teodora Spasojevic, Markus Braun, et al.
+- **abs:** https://arxiv.org/abs/2602.05966 · **PDF:** https://arxiv.org/pdf/2602.05966v1
+
+## 15054. Learning to Share: Selective Memory for Efficient Parallel Agentic Systems
+
+- **ID:** `2602.05965` · **primary:** `cs.MA` · **Submitted:** 2026-02-05
+- **Authors:** Joseph Fioresi, Parth Parag Kulkarni, Ashmal Vayani, et al.
+- **abs:** https://arxiv.org/abs/2602.05965 · **PDF:** https://arxiv.org/pdf/2602.05965v1
+
+## 15055. Discrete diffusion samplers and bridges: Off-policy algorithms and applications in latent spaces
+
+- **ID:** `2602.05961` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Arran Carter, Sanghyeok Choi, Kirill Tamogashev, et al.
+- **abs:** https://arxiv.org/abs/2602.05961 · **PDF:** https://arxiv.org/pdf/2602.05961v1
+
+## 15056. Better Source, Better Flow: Learning Condition-Dependent Source Distribution for Flow Matching
+
+- **ID:** `2602.05951` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Junwan Kim, Jiho Park, Seonghu Jeon, et al.
+- **abs:** https://arxiv.org/abs/2602.05951 · **PDF:** https://arxiv.org/pdf/2602.05951v1
+
+## 15057. Breaking Symmetry Bottlenecks in GNN Readouts
+
+- **ID:** `2602.05950` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Mouad Talhi, Arne Wolf, Anthea Monod
+- **abs:** https://arxiv.org/abs/2602.05950 · **PDF:** https://arxiv.org/pdf/2602.05950v1
+
+## 15058. f-GRPO and Beyond: Divergence-Based Reinforcement Learning Algorithms for General LLM Alignment
+
+- **ID:** `2602.05946` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Rajdeep Haldar, Lantao Mei, Guang Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.05946 · **PDF:** https://arxiv.org/pdf/2602.05946v2
+
+## 15059. Orthogonal Model Merging
+
+- **ID:** `2602.05943` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Sihan Yang, Kexuan Shi, Weiyang Liu
+- **abs:** https://arxiv.org/abs/2602.05943 · **PDF:** https://arxiv.org/pdf/2602.05943v1
+
+## 15060. Self-Improving Multilingual Long Reasoning via Translation-Reasoning Integrated Training
+
+- **ID:** `2602.05940` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Junxiao Liu, Zhijun Wang, Yixiao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.05940 · **PDF:** https://arxiv.org/pdf/2602.05940v1
+
+## 15061. Dimensionality Reduction on Riemannian Manifolds in Data Analysis
+
+- **ID:** `2602.05936` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Alaa El Ichi, Khalide Jbilou
+- **abs:** https://arxiv.org/abs/2602.05936 · **PDF:** https://arxiv.org/pdf/2602.05936v1
+
+## 15062. Tuning Out-of-Distribution (OOD) Detectors Without Given OOD Data
+
+- **ID:** `2602.05935` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Sudeepta Mondal, Xinyi Mary Xie, Ruxiao Duan, et al.
+- **abs:** https://arxiv.org/abs/2602.05935 · **PDF:** https://arxiv.org/pdf/2602.05935v1
+
+## 15063. Approximation of Log-Partition Function in Policy Mirror Descent Induces Implicit Regularization for LLM Post-Training
+
+- **ID:** `2602.05933` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Zhenghao Xu, Qin Lu, Changlong Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.05933 · **PDF:** https://arxiv.org/pdf/2602.05933v1
+
+## 15064. Polyglots or Multitudes? Multilingual LLM Answers to Value-laden Multiple-Choice Questions
+
+- **ID:** `2602.05932` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Léo Labat, Etienne Ollion, François Yvon
+- **abs:** https://arxiv.org/abs/2602.05932 · **PDF:** https://arxiv.org/pdf/2602.05932v1
+
+## 15065. Compound Deception in Elite Peer Review: A Failure Mode Taxonomy of 100 Fabricated Citations at NeurIPS 2025
+
+- **ID:** `2602.05930` · **primary:** `cs.DL` · **Submitted:** 2026-02-05
+- **Authors:** Samar Ansari
+- **abs:** https://arxiv.org/abs/2602.05930 · **PDF:** https://arxiv.org/pdf/2602.05930v1
+
+## 15066. KV-CoRE: Benchmarking Data-Dependent Low-Rank Compressibility of KV-Caches in LLMs
+
+- **ID:** `2602.05929` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Jian Chen, Zhuoran Wang, Jiayu Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.05929 · **PDF:** https://arxiv.org/pdf/2602.05929v2
+
+## 15067. Transformers Are Born Biased: Structural Inductive Biases at Random Initialization and Their Practical Consequences
+
+- **ID:** `2602.05927` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Siquan Li, Yao Tong, Haonan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.05927 · **PDF:** https://arxiv.org/pdf/2602.05927v1
+
+## 15068. Quantum Reinforcement Learning with Transformers for the Capacitated Vehicle Routing Problem
+
+- **ID:** `2602.05920` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Eva Andrés
+- **abs:** https://arxiv.org/abs/2602.05920 · **PDF:** https://arxiv.org/pdf/2602.05920v1
+
+## 15069. Chunky Post-Training: Data Driven Failures of Generalization
+
+- **ID:** `2602.05910` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Seoirse Murray, Allison Qi, Timothy Qian, et al.
+- **abs:** https://arxiv.org/abs/2602.05910 · **PDF:** https://arxiv.org/pdf/2602.05910v1
+
+## 15070. Codified Finite-state Machines for Role-playing
+
+- **ID:** `2602.05905` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Letian Peng, Yupeng Hou, Kun Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.05905 · **PDF:** https://arxiv.org/pdf/2602.05905v1
+
+## 15071. Verification of the Implicit World Model in a Generative Model via Adversarial Sequences
+
+- **ID:** `2602.05903` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** András Balogh, Márk Jelasity
+- **abs:** https://arxiv.org/abs/2602.05903 · **PDF:** https://arxiv.org/pdf/2602.05903v1
+
+## 15072. Regularized Calibration with Successive Rounding for Post-Training Quantization
+
+- **ID:** `2602.05902` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Seohyeon Cha, Huancheng Chen, Dongjun Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.05902 · **PDF:** https://arxiv.org/pdf/2602.05902v1
+
+## 15073. Universal approximation with signatures of non-geometric rough paths
+
+- **ID:** `2602.05898` · **primary:** `math.PR` · **Submitted:** 2026-02-05
+- **Authors:** Mihriban Ceylan, Anna P. Kwossek, David J. Prömel
+- **abs:** https://arxiv.org/abs/2602.05898 · **PDF:** https://arxiv.org/pdf/2602.05898v1
+
+## 15074. Stop Rewarding Hallucinated Steps: Faithfulness-Aware Step-Level Reinforcement Learning for Small Reasoning Models
+
+- **ID:** `2602.05897` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Shuo Nie, Hexuan Deng, Chao Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.05897 · **PDF:** https://arxiv.org/pdf/2602.05897v1
+
+## 15075. Parity, Sensitivity, and Transformers
+
+- **ID:** `2602.05896` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Alexander Kozachinskiy, Tomasz Steifer, Przemysław Wałȩga
+- **abs:** https://arxiv.org/abs/2602.05896 · **PDF:** https://arxiv.org/pdf/2602.05896v1
+
+## 15076. ContextBench: A Benchmark for Context Retrieval in Coding Agents
+
+- **ID:** `2602.05892` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Han Li, Letian Zhu, Bohan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.05892 · **PDF:** https://arxiv.org/pdf/2602.05892v3
+
+## 15077. DFPO: Scaling Value Modeling via Distributional Flow towards Robust and Generalizable LLM Post-Training
+
+- **ID:** `2602.05890` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Dingwei Zhu, Zhiheng Xi, Shihan Dou, et al.
+- **abs:** https://arxiv.org/abs/2602.05890 · **PDF:** https://arxiv.org/pdf/2602.05890v1
+
+## 15078. Metric Hedonic Games on the Line
+
+- **ID:** `2602.05888` · **primary:** `cs.GT` · **Submitted:** 2026-02-05
+- **Authors:** Merlin de la Haye, Pascal Lenzner, Farehe Soheil, et al.
+- **abs:** https://arxiv.org/abs/2602.05888 · **PDF:** https://arxiv.org/pdf/2602.05888v2
+
+## 15079. Escaping Local Minima Provably in Non-convex Matrix Sensing: A Deterministic Framework via Simulated Lifting
+
+- **ID:** `2602.05887` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Tianqi Shen, Jinji Yang, Junze He, et al.
+- **abs:** https://arxiv.org/abs/2602.05887 · **PDF:** https://arxiv.org/pdf/2602.05887v2
+
+## 15080. Dr. Kernel: Reinforcement Learning Done Right for Triton Kernel Generations
+
+- **ID:** `2602.05885` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Wei Liu, Jiawei Xu, Yingru Li, et al.
+- **abs:** https://arxiv.org/abs/2602.05885 · **PDF:** https://arxiv.org/pdf/2602.05885v2
+
+## 15081. Neural Implicit 3D Cardiac Shape Reconstruction from Sparse CT Angiography Slices Mimicking 2D Transthoracic Echocardiography Views
+
+- **ID:** `2602.05884` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Gino E. Jansen, Carolina Brás, R. Nils Planken, et al.
+- **abs:** https://arxiv.org/abs/2602.05884 · **PDF:** https://arxiv.org/pdf/2602.05884v1
+
+## 15082. A Guide to Large Language Models in Modeling and Simulation: From Core Techniques to Critical Challenges
+
+- **ID:** `2602.05883` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Philippe J. Giabbanelli
+- **abs:** https://arxiv.org/abs/2602.05883 · **PDF:** https://arxiv.org/pdf/2602.05883v1
+
+## 15083. EuroLLM-22B: Technical Report
+
+- **ID:** `2602.05879` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Miguel Moura Ramos, Duarte M. Alves, Hippolyte Gisserot-Boukhlef, et al.
+- **abs:** https://arxiv.org/abs/2602.05879 · **PDF:** https://arxiv.org/pdf/2602.05879v1
+
+## 15084. Agent2Agent Threats in Safety-Critical LLM Assistants: A Human-Centric Taxonomy
+
+- **ID:** `2602.05877` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Lukas Stappen, Ahmet Erkan Turan, Johann Hagerer, et al.
+- **abs:** https://arxiv.org/abs/2602.05877 · **PDF:** https://arxiv.org/pdf/2602.05877v1
+
+## 15085. Beyond Manual Planning: Seating Allocation for Large Organizations
+
+- **ID:** `2602.05875` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Anton Ipsen, Michael Cashmore, Kirsty Fielding, et al.
+- **abs:** https://arxiv.org/abs/2602.05875 · **PDF:** https://arxiv.org/pdf/2602.05875v1
+
+## 15086. xList-Hate: A Checklist-Based Framework for Interpretable and Generalizable Hate Speech Detection
+
+- **ID:** `2602.05874` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Adrián Girón, Pablo Miralles, Javier Huertas-Tato, et al.
+- **abs:** https://arxiv.org/abs/2602.05874 · **PDF:** https://arxiv.org/pdf/2602.05874v1
+
+## 15087. Large-scale Score-based Variational Posterior Inference for Bayesian Deep Neural Networks
+
+- **ID:** `2602.05873` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Minyoung Kim
+- **abs:** https://arxiv.org/abs/2602.05873 · **PDF:** https://arxiv.org/pdf/2602.05873v1
+
+## 15088. TokaMark: A Comprehensive Benchmark for MAST Tokamak Plasma Models
+
+- **ID:** `2602.10132` · **primary:** `physics.plasm-ph` · **Submitted:** 2026-02-05
+- **Authors:** Cécile Rousseau, Samuel Jackson, Rodrigo H. Ordonez-Hurtado, et al.
+- **abs:** https://arxiv.org/abs/2602.10132 · **PDF:** https://arxiv.org/pdf/2602.10132v2
+
+## 15089. Wedge Sampling: Efficient Tensor Completion with Nearly-Linear Sample Complexity
+
+- **ID:** `2602.05869` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Hengrui Luo, Anna Ma, Ludovic Stephan, et al.
+- **abs:** https://arxiv.org/abs/2602.05869 · **PDF:** https://arxiv.org/pdf/2602.05869v1
+
+## 15090. Constrained Group Relative Policy Optimization
+
+- **ID:** `2602.05863` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Roger Girgis, Rodrigue de Schaetzen, Luke Rowe, et al.
+- **abs:** https://arxiv.org/abs/2602.05863 · **PDF:** https://arxiv.org/pdf/2602.05863v2
+
+## 15091. Distribution-free two-sample testing with blurred total variation distance
+
+- **ID:** `2602.05862` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Rohan Hore, Rina Foygel Barber
+- **abs:** https://arxiv.org/abs/2602.05862 · **PDF:** https://arxiv.org/pdf/2602.05862v1
+
+## 15092. CFRecs: Counterfactual Recommendations on Real Estate User Listing Interaction Graphs
+
+- **ID:** `2602.05861` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Seyedmasoud Mousavi, Ruomeng Xu, Xiaojing Zhu
+- **abs:** https://arxiv.org/abs/2602.05861 · **PDF:** https://arxiv.org/pdf/2602.05861v1
+
+## 15093. Distributed physics-informed neural networks via domain decomposition for fast flow reconstruction
+
+- **ID:** `2602.15883` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Yixiao Qian, Jiaxu Liu, Zewei Xia, et al.
+- **abs:** https://arxiv.org/abs/2602.15883 · **PDF:** https://arxiv.org/pdf/2602.15883v1
+
+## 15094. DLM-Scope: Mechanistic Interpretability of Diffusion Language Models via Sparse Autoencoders
+
+- **ID:** `2602.05859` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Xu Wang, Bingqing Jiang, Yu Wan, et al.
+- **abs:** https://arxiv.org/abs/2602.05859 · **PDF:** https://arxiv.org/pdf/2602.05859v1
+
+## 15095. BABE: Biology Arena BEnchmark
+
+- **ID:** `2602.05857` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Junting Zhou, Jin Chen, Linfeng Hao, et al.
+- **abs:** https://arxiv.org/abs/2602.05857 · **PDF:** https://arxiv.org/pdf/2602.05857v1
+
+## 15096. A Hybrid Autoencoder for Robust Heightmap Generation from Fused Lidar and Depth Data for Humanoid Robot Locomotion
+
+- **ID:** `2602.05855` · **primary:** `cs.RO` · **Submitted:** 2026-02-05
+- **Authors:** Dennis Bank, Joost Cordes, Thomas Seel, et al.
+- **abs:** https://arxiv.org/abs/2602.05855 · **PDF:** https://arxiv.org/pdf/2602.05855v1
+
+## 15097. RRAttention: Dynamic Block Sparse Attention via Per-Head Round-Robin Shifts for Long-Context Inference
+
+- **ID:** `2602.05853` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Siran Liu, Guoxia Wang, Sa Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.05853 · **PDF:** https://arxiv.org/pdf/2602.05853v1
+
+## 15098. Exact Recovery in the Data Block Model
+
+- **ID:** `2602.05852` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Amir R. Asadi, Akbar Davoodi, Ramin Javadi, et al.
+- **abs:** https://arxiv.org/abs/2602.05852 · **PDF:** https://arxiv.org/pdf/2602.05852v1
+
+## 15099. Visualizing the loss landscapes of physics-informed neural networks
+
+- **ID:** `2602.05849` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Conor Rowan, Finn Murphy-Blanchard
+- **abs:** https://arxiv.org/abs/2602.05849 · **PDF:** https://arxiv.org/pdf/2602.05849v1
+
+## 15100. DARWIN: Dynamic Agentically Rewriting Self-Improving Network
+
+- **ID:** `2602.05848` · **primary:** `cs.NE` · **Submitted:** 2026-02-05
+- **Authors:** Henry Jiang
+- **abs:** https://arxiv.org/abs/2602.05848 · **PDF:** https://arxiv.org/pdf/2602.05848v1
+
+## 15101. OmniVideo-R1: Reinforcing Audio-visual Reasoning with Query Intention and Modality Attention
+
+- **ID:** `2602.05847` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Zhangquan Chen, Jiale Tao, Ruihuang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.05847 · **PDF:** https://arxiv.org/pdf/2602.05847v2
+
+## 15102. Optimal scaling laws in learning hierarchical multi-index models
+
+- **ID:** `2602.05846` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Leonardo Defilippis, Florent Krzakala, Bruno Loureiro, et al.
+- **abs:** https://arxiv.org/abs/2602.05846 · **PDF:** https://arxiv.org/pdf/2602.05846v1
+
+## 15103. OdysseyArena: Benchmarking Large Language Models For Long-Horizon, Active and Inductive Interactions
+
+- **ID:** `2602.05843` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Fangzhi Xu, Hang Yan, Qiushi Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.05843 · **PDF:** https://arxiv.org/pdf/2602.05843v1
+
+## 15104. Reinforcement World Model Learning for LLM-based Agents
+
+- **ID:** `2602.05842` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Xiao Yu, Baolin Peng, Ruize Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.05842 · **PDF:** https://arxiv.org/pdf/2602.05842v2
+
+## 15105. FHAIM: Fully Homomorphic AIM For Private Synthetic Data Generation
+
+- **ID:** `2602.05838` · **primary:** `cs.CR` · **Submitted:** 2026-02-05
+- **Authors:** Mayank Kumar, Qian Lou, Paulo Barreto, et al.
+- **abs:** https://arxiv.org/abs/2602.05838 · **PDF:** https://arxiv.org/pdf/2602.05838v2
+
+## 15106. Synthesizing Realistic Test Data without Breaking Privacy
+
+- **ID:** `2602.05833` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Laura Plein, Alexi Turcotte, Arina Hallemans, et al.
+- **abs:** https://arxiv.org/abs/2602.05833 · **PDF:** https://arxiv.org/pdf/2602.05833v1
+
+## 15107. Learning Compact Boolean Networks
+
+- **ID:** `2602.05830` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Shengpu Wang, Yuhao Mao, Yani Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.05830 · **PDF:** https://arxiv.org/pdf/2602.05830v1
+
+## 15108. TKG-Thinker: Towards Dynamic Reasoning over Temporal Knowledge Graphs via Agentic Reinforcement Learning
+
+- **ID:** `2602.05818` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Zihao Jiang, Miao Peng, Zhenyan Shan, et al.
+- **abs:** https://arxiv.org/abs/2602.05818 · **PDF:** https://arxiv.org/pdf/2602.05818v2
+
+## 15109. Interpreting Manifolds and Graph Neural Embeddings from Internet of Things Traffic Flows
+
+- **ID:** `2602.05817` · **primary:** `cs.CR` · **Submitted:** 2026-02-05
+- **Authors:** Enrique Feito-Casares, Francisco M. Melgarejo-Meseguer, Elena Casiraghi, et al.
+- **abs:** https://arxiv.org/abs/2602.05817 · **PDF:** https://arxiv.org/pdf/2602.05817v2
+
+## 15110. Where Does Warm-Up Come From? Adaptive Scheduling for Norm-Constrained Optimizers
+
+- **ID:** `2602.05813` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Artem Riabinin, Andrey Veprikov, Arman Bolatov, et al.
+- **abs:** https://arxiv.org/abs/2602.05813 · **PDF:** https://arxiv.org/pdf/2602.05813v1
+
+## 15111. Principled Confidence Estimation for Deep Computed Tomography
+
+- **ID:** `2602.05812` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Matteo Gätzner, Johannes Kirschner
+- **abs:** https://arxiv.org/abs/2602.05812 · **PDF:** https://arxiv.org/pdf/2602.05812v1
+
+## 15112. STProtein: predicting spatial protein expression from multi-omics data
+
+- **ID:** `2602.05811` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Zhaorui Jiang, Yingfang Yuan, Lei Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.05811 · **PDF:** https://arxiv.org/pdf/2602.05811v1
+
+## 15113. Bifrost: Steering Strategic Trajectories to Bridge Contextual Gaps for Self-Improving Agents
+
+- **ID:** `2602.05810` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Quan M. Tran, Zhuo Huang, Wenbin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.05810 · **PDF:** https://arxiv.org/pdf/2602.05810v1
+
+## 15114. NEX: Neuron Explore-Exploit Scoring for Label-Free Chain-of-Thought Selection and Model Ranking
+
+- **ID:** `2602.05805` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Kang Chen, Zhuoka Feng, Sihan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.05805 · **PDF:** https://arxiv.org/pdf/2602.05805v1
+
+## 15115. Non-Stationary Inventory Control with Lead Times
+
+- **ID:** `2602.05799` · **primary:** `math.OC` · **Submitted:** 2026-02-05
+- **Authors:** Nele H. Amiri, Sean R. Sinclair, Maximiliano Udenio
+- **abs:** https://arxiv.org/abs/2602.05799 · **PDF:** https://arxiv.org/pdf/2602.05799v1
+
+## 15116. Learning False Discovery Rate Control via Model-Based Neural Networks
+
+- **ID:** `2602.05798` · **primary:** `stat.ME` · **Submitted:** 2026-02-05
+- **Authors:** Arnau Vilella, Jasin Machkour, Michael Muma, et al.
+- **abs:** https://arxiv.org/abs/2602.05798 · **PDF:** https://arxiv.org/pdf/2602.05798v1
+
+## 15117. Classification Under Local Differential Privacy with Model Reversal and Model Averaging
+
+- **ID:** `2602.05797` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Caihong Qin, Yang Bai
+- **abs:** https://arxiv.org/abs/2602.05797 · **PDF:** https://arxiv.org/pdf/2602.05797v1
+
+## 15118. FiMI: A Domain-Specific Language Model for Indian Finance Ecosystem
+
+- **ID:** `2602.05794` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Aboli Kathar, Aman Kumar, Anusha Kamath, et al.
+- **abs:** https://arxiv.org/abs/2602.05794 · **PDF:** https://arxiv.org/pdf/2602.05794v2
+
+## 15119. Price of universality in vector quantization is at most 0.11 bit
+
+- **ID:** `2602.05790` · **primary:** `cs.IT` · **Submitted:** 2026-02-05
+- **Authors:** Alina Harbuzova, Or Ordentlich, Yury Polyanskiy
+- **abs:** https://arxiv.org/abs/2602.05790 · **PDF:** https://arxiv.org/pdf/2602.05790v1
+
+## 15120. Allocentric Perceiver: Disentangling Allocentric Reasoning from Egocentric Visual Priors via Frame Instantiation
+
+- **ID:** `2602.05789` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Hengyi Wang, Ruiqiang Zhang, Chang Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.05789 · **PDF:** https://arxiv.org/pdf/2602.05789v1
+
+## 15121. Bagging-Based Model Merging for Robust General Text Embeddings
+
+- **ID:** `2602.05787` · **primary:** `cs.IR` · **Submitted:** 2026-02-05
+- **Authors:** Hengran Zhang, Keping Bi, Jiafeng Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.05787 · **PDF:** https://arxiv.org/pdf/2602.05787v2
+
+## 15122. Selecting Hyperparameters for Tree-Boosting
+
+- **ID:** `2602.05786` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Floris Jan Koster, Fabio Sigrist
+- **abs:** https://arxiv.org/abs/2602.05786 · **PDF:** https://arxiv.org/pdf/2602.05786v1
+
+## 15123. ReText: Text Boosts Generalization in Image-Based Person Re-identification
+
+- **ID:** `2602.05785` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Timur Mamedov, Karina Kvanchiani, Anton Konushin, et al.
+- **abs:** https://arxiv.org/abs/2602.05785 · **PDF:** https://arxiv.org/pdf/2602.05785v1
+
+## 15124. Distributional Reinforcement Learning with Diffusion Bridge Critics
+
+- **ID:** `2602.05783` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Shutong Ding, Yimiao Zhou, Ke Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.05783 · **PDF:** https://arxiv.org/pdf/2602.05783v1
+
+## 15125. Automated Customization of LLMs for Enterprise Code Repositories Using Semantic Scopes
+
+- **ID:** `2602.05780` · **primary:** `cs.SE` · **Submitted:** 2026-02-05
+- **Authors:** Ulrich Finkler, Irene Manotas, Wei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.05780 · **PDF:** https://arxiv.org/pdf/2602.05780v1
+
+## 15126. How Controlling the Variance can Improve Training Stability of Sparsely Activated DNNs and CNNs
+
+- **ID:** `2602.05779` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Emily Dent, Jared Tanner
+- **abs:** https://arxiv.org/abs/2602.05779 · **PDF:** https://arxiv.org/pdf/2602.05779v1
+
+## 15127. Cross-Domain Offline Policy Adaptation via Selective Transition Correction
+
+- **ID:** `2602.05776` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Mengbei Yan, Jiafei Lyu, Shengjie Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.05776 · **PDF:** https://arxiv.org/pdf/2602.05776v1
+
+## 15128. Note on Martingale Theory and Applications
+
+- **ID:** `2602.05774` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Xiandong Zou
+- **abs:** https://arxiv.org/abs/2602.05774 · **PDF:** https://arxiv.org/pdf/2602.05774v3
+
+## 15129. Different Time, Different Language: Revisiting the Bias Against Non-Native Speakers in GPT Detectors
+
+- **ID:** `2602.05769` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Adnan Al Ali, Jindřich Helcl, Jindřich Libovický
+- **abs:** https://arxiv.org/abs/2602.05769 · **PDF:** https://arxiv.org/pdf/2602.05769v1
+
+## 15130. PMT Waveform Simulation and Reconstruction with Conditional Diffusion Network
+
+- **ID:** `2602.05767` · **primary:** `hep-ex` · **Submitted:** 2026-02-05
+- **Authors:** Kainan Liu, Jingyu Huang, Guihong Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.05767 · **PDF:** https://arxiv.org/pdf/2602.05767v1
+
+## 15131. RL-VLA$^3$: Reinforcement Learning VLA Accelerating via Full Asynchronism
+
+- **ID:** `2602.05765` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Zhong Guan, Haoran Sun, Yongjian Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.05765 · **PDF:** https://arxiv.org/pdf/2602.05765v1
+
+## 15132. RocqSmith: Can Automatic Optimization Forge Better Proof Agents?
+
+- **ID:** `2602.05762` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Andrei Kozyrev, Nikita Khramov, Denis Lochmelis, et al.
+- **abs:** https://arxiv.org/abs/2602.05762 · **PDF:** https://arxiv.org/pdf/2602.05762v1
+
+## 15133. LongR: Unleashing Long-Context Reasoning via Reinforcement Learning with Dense Utility Rewards
+
+- **ID:** `2602.05758` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Bowen Ping, Zijun Chen, Yiyao Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.05758 · **PDF:** https://arxiv.org/pdf/2602.05758v1
+
+## 15134. TimelyFreeze: Adaptive Parameter Freezing Mechanism for Pipeline Parallelism
+
+- **ID:** `2602.05754` · **primary:** `cs.DC` · **Submitted:** 2026-02-05
+- **Authors:** Seonghye Cho, Jaemin Han, Hyunjin Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.05754 · **PDF:** https://arxiv.org/pdf/2602.05754v2
+
+## 15135. How to Achieve the Intended Aim of Deep Clustering Now, without Deep Learning
+
+- **ID:** `2602.05749` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Kai Ming Ting, Wei-Jie Xu, Hang Zhang
+- **abs:** https://arxiv.org/abs/2602.05749 · **PDF:** https://arxiv.org/pdf/2602.05749v1
+
+## 15136. LeakBoost: Perceptual-Loss-Based Membership Inference Attack
+
+- **ID:** `2602.05748` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Amit Kravchik Taub, Fred M. Grabovski, Guy Amit, et al.
+- **abs:** https://arxiv.org/abs/2602.05748 · **PDF:** https://arxiv.org/pdf/2602.05748v1
+
+## 15137. Learning to Inject: Automated Prompt Injection via Reinforcement Learning
+
+- **ID:** `2602.05746` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Xin Chen, Jie Zhang, Florian Tramèr
+- **abs:** https://arxiv.org/abs/2602.05746 · **PDF:** https://arxiv.org/pdf/2602.05746v1
+
+## 15138. Fast Rates for Nonstationary Weighted Risk Minimization
+
+- **ID:** `2602.05742` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Tobias Brock, Thomas Nagler
+- **abs:** https://arxiv.org/abs/2602.05742 · **PDF:** https://arxiv.org/pdf/2602.05742v1
+
+## 15139. Neuro-Inspired Visual Pattern Recognition via Biological Reservoir Computing
+
+- **ID:** `2602.05737` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Luca Ciampi, Ludovico Iannello, Fabrizio Tonelli, et al.
+- **abs:** https://arxiv.org/abs/2602.05737 · **PDF:** https://arxiv.org/pdf/2602.05737v1
+
+## 15140. CSRv2: Unlocking Ultra-Sparse Embeddings
+
+- **ID:** `2602.05735` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Lixuan Guo, Yifei Wang, Tiansheng Wen, et al.
+- **abs:** https://arxiv.org/abs/2602.05735 · **PDF:** https://arxiv.org/pdf/2602.05735v4
+
+## 15141. Evaluating the impact of word embeddings on similarity scoring in practical information retrieval
+
+- **ID:** `2602.05734` · **primary:** `cs.IR` · **Submitted:** 2026-02-05
+- **Authors:** Niall McCarroll, Kevin Curran, Eugene McNamee, et al.
+- **abs:** https://arxiv.org/abs/2602.05734 · **PDF:** https://arxiv.org/pdf/2602.05734v1
+
+## 15142. Adaptive Global and Fine-Grained Perceptual Fusion for MLLM Embeddings Compatible with Hard Negative Amplification
+
+- **ID:** `2602.05729` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Lexiang Hu, Youze Xue, Dian Li, et al.
+- **abs:** https://arxiv.org/abs/2602.05729 · **PDF:** https://arxiv.org/pdf/2602.05729v1
+
+## 15143. CompactRAG: Reducing LLM Calls and Token Overhead in Multi-Hop Question Answering
+
+- **ID:** `2602.05728` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Hao Yang, Zhiyu Yang, Xupeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.05728 · **PDF:** https://arxiv.org/pdf/2602.05728v1
+
+## 15144. Muon in Associative Memory Learning: Training Dynamics and Scaling Laws
+
+- **ID:** `2602.05725` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Binghui Li, Kaifei Wang, Han Zhong, et al.
+- **abs:** https://arxiv.org/abs/2602.05725 · **PDF:** https://arxiv.org/pdf/2602.05725v1
+
+## 15145. Mitigating Hallucination in Financial Retrieval-Augmented Generation via Fine-Grained Knowledge Verification
+
+- **ID:** `2602.05723` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Taoye Yin, Haoyuan Hu, Yaxin Fan, et al.
+- **abs:** https://arxiv.org/abs/2602.05723 · **PDF:** https://arxiv.org/pdf/2602.05723v1
+
+## 15146. Anchored Policy Optimization: Mitigating Exploration Collapse Via Support-Constrained Rectification
+
+- **ID:** `2602.05717` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Tianyi Wang, Long Li, Hongcan Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.05717 · **PDF:** https://arxiv.org/pdf/2602.05717v1
+
+## 15147. Projected Boosting with Fairness Constraints: Quantifying the Cost of Fair Training Distributions
+
+- **ID:** `2602.05713` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Amir Asiaee, Kaveh Aryan
+- **abs:** https://arxiv.org/abs/2602.05713 · **PDF:** https://arxiv.org/pdf/2602.05713v1
+
+## 15148. Towards Green AI: Decoding the Energy of LLM Inference in Software Development
+
+- **ID:** `2602.05712` · **primary:** `cs.SE` · **Submitted:** 2026-02-05
+- **Authors:** Lola Solovyeva, Fernando Castor
+- **abs:** https://arxiv.org/abs/2602.05712 · **PDF:** https://arxiv.org/pdf/2602.05712v1
+
+## 15149. OmniMoE: An Efficient MoE by Orchestrating Atomic Experts at Scale
+
+- **ID:** `2602.05711` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Jingze Shi, Zhangyang Peng, Yizhang Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.05711 · **PDF:** https://arxiv.org/pdf/2602.05711v1
+
+## 15150. Ethology of Latent Spaces
+
+- **ID:** `2602.05710` · **primary:** `cs.CY` · **Submitted:** 2026-02-05
+- **Authors:** Philippe Boisnard
+- **abs:** https://arxiv.org/abs/2602.05710 · **PDF:** https://arxiv.org/pdf/2602.05710v1
+
+## 15151. Nonlinearity as Rank: Generative Low-Rank Adapter with Radial Basis Functions
+
+- **ID:** `2602.05709` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Yihao Ouyang, Shiwei Li, Haozhao Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.05709 · **PDF:** https://arxiv.org/pdf/2602.05709v1
+
+## 15152. Cost-Efficient RAG for Entity Matching with LLMs: A Blocking-based Exploration
+
+- **ID:** `2602.05708` · **primary:** `cs.DB` · **Submitted:** 2026-02-05
+- **Authors:** Chuangtao Ma, Zeyu Zhang, Arijit Khan, et al.
+- **abs:** https://arxiv.org/abs/2602.05708 · **PDF:** https://arxiv.org/pdf/2602.05708v1
+
+## 15153. Fix Representation (Optimally) Before Fairness: Finite-Sample Shrinkage Population Correction and the True Price of Fairness Under Subpopulation Shift
+
+- **ID:** `2602.05707` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Amir Asiaee, Kaveh Aryan
+- **abs:** https://arxiv.org/abs/2602.05707 · **PDF:** https://arxiv.org/pdf/2602.05707v1
+
+## 15154. Poster: Camera Tampering Detection for Outdoor IoT Systems
+
+- **ID:** `2602.05706` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Shadi Attarha, Kanaga Shanmugi, Anna Förster
+- **abs:** https://arxiv.org/abs/2602.05706 · **PDF:** https://arxiv.org/pdf/2602.05706v1
+
+## 15155. Limitations of SGD for Multi-Index Models Beyond Statistical Queries
+
+- **ID:** `2602.05704` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Daniel Barzilai, Ohad Shamir
+- **abs:** https://arxiv.org/abs/2602.05704 · **PDF:** https://arxiv.org/pdf/2602.05704v1
+
+## 15156. FUTURE-VLA: Forecasting Unified Trajectories Under Real-time Execution
+
+- **ID:** `2602.15882` · **primary:** `cs.RO` · **Submitted:** 2026-02-05
+- **Authors:** Jingjing Fan, Yushan Liu, Shoujie Li, et al.
+- **abs:** https://arxiv.org/abs/2602.15882 · **PDF:** https://arxiv.org/pdf/2602.15882v1
+
+## 15157. Broken neural scaling laws in materials science
+
+- **ID:** `2602.05702` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-05
+- **Authors:** Max Großmann, Malte Grunert, Erich Runge
+- **abs:** https://arxiv.org/abs/2602.05702 · **PDF:** https://arxiv.org/pdf/2602.05702v1
+
+## 15158. SweetSpot: An Analytical Model for Predicting Energy Efficiency of LLM Inference
+
+- **ID:** `2602.05695` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Hiari Pizzini Cavagna, Andrea Proia, Giacomo Madella, et al.
+- **abs:** https://arxiv.org/abs/2602.05695 · **PDF:** https://arxiv.org/pdf/2602.05695v2
+
+## 15159. Consensus-Aligned Neuron Efficient Fine-Tuning Large Language Models for Multi-Domain Machine Translation
+
+- **ID:** `2602.05694` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Shuting Jiang, Ran Song, Yuxin Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.05694 · **PDF:** https://arxiv.org/pdf/2602.05694v1
+
+## 15160. FedRandom: Sampling Consistent and Accurate Contribution Values in Federated Learning
+
+- **ID:** `2602.05693` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Arno Geimer, Beltran Fiz Pontiveros, Radu State
+- **abs:** https://arxiv.org/abs/2602.05693 · **PDF:** https://arxiv.org/pdf/2602.05693v1
+
+## 15161. MedErrBench: A Fine-Grained Multilingual Benchmark for Medical Error Detection and Correction with Clinical Expert Annotations
+
+- **ID:** `2602.05692` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Congbo Ma, Yichun Zhang, Yousef Al-Jazzazi, et al.
+- **abs:** https://arxiv.org/abs/2602.05692 · **PDF:** https://arxiv.org/pdf/2602.05692v1
+
+## 15162. Almost Asymptotically Optimal Active Clustering Through Pairwise Observations
+
+- **ID:** `2602.05690` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Rachel S. Y. Teo, P. N. Karthik, Ramya Korlakai Vinayak, et al.
+- **abs:** https://arxiv.org/abs/2602.05690 · **PDF:** https://arxiv.org/pdf/2602.05690v1
+
+## 15163. Mining Generalizable Activation Functions
+
+- **ID:** `2602.05688` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Alex Vitvitskyi, Michael Boratko, Matej Grcic, et al.
+- **abs:** https://arxiv.org/abs/2602.05688 · **PDF:** https://arxiv.org/pdf/2602.05688v1
+
+## 15164. Exploring AI-Augmented Sensemaking of Patient-Generated Health Data: A Mixed-Method Study with Healthcare Professionals in Cardiac Risk Reduction
+
+- **ID:** `2602.05687` · **primary:** `cs.HC` · **Submitted:** 2026-02-05
+- **Authors:** Pavithren V S Pakianathan, Rania Islambouli, Diogo Branco, et al.
+- **abs:** https://arxiv.org/abs/2602.05687 · **PDF:** https://arxiv.org/pdf/2602.05687v4
+
+## 15165. Perception-Based Beliefs for POMDPs with Visual Observations
+
+- **ID:** `2602.05679` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Miriam Schäfers, Merlijn Krale, Thiago D. Simão, et al.
+- **abs:** https://arxiv.org/abs/2602.05679 · **PDF:** https://arxiv.org/pdf/2602.05679v1
+
+## 15166. Variable Search Stepsize for Randomized Local Search in Multi-Objective Combinatorial Optimization
+
+- **ID:** `2602.05675` · **primary:** `cs.NE` · **Submitted:** 2026-02-05
+- **Authors:** Xuepeng Ren, Maocai Wang, Guangming Dai, et al.
+- **abs:** https://arxiv.org/abs/2602.05675 · **PDF:** https://arxiv.org/pdf/2602.05675v1
+
+## 15167. HyperPotter: Spell the Charm of High-Order Interactions in Audio Deepfake Detection
+
+- **ID:** `2602.05670` · **primary:** `cs.SD` · **Submitted:** 2026-02-05
+- **Authors:** Qing Wen, Haohao Li, Zhongjie Ba, et al.
+- **abs:** https://arxiv.org/abs/2602.05670 · **PDF:** https://arxiv.org/pdf/2602.05670v1
+
+## 15168. Stable but Wrong: When More Data Degrades Scientific Conclusions
+
+- **ID:** `2602.05668` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Zhipeng Zhang, Kai Li
+- **abs:** https://arxiv.org/abs/2602.05668 · **PDF:** https://arxiv.org/pdf/2602.05668v1
+
+## 15169. Accelerating Benchmarking of Functional Connectivity Modeling via Structure-aware Core-set Selection
+
+- **ID:** `2602.05667` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Ling Zhan, Zhen Li, Junjie Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.05667 · **PDF:** https://arxiv.org/pdf/2602.05667v1
+
+## 15170. A Theoretical Analysis of Test-Driven LLM Code Generation
+
+- **ID:** `2602.06098` · **primary:** `cs.SE` · **Submitted:** 2026-02-05
+- **Authors:** Nicolas Menet, Michael Hersche, Andreas Krause, et al.
+- **abs:** https://arxiv.org/abs/2602.06098 · **PDF:** https://arxiv.org/pdf/2602.06098v2
+
+## 15171. Graph-based Agent Memory: Taxonomy, Techniques, and Applications
+
+- **ID:** `2602.05665` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Chang Yang, Chuang Zhou, Yilin Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.05665 · **PDF:** https://arxiv.org/pdf/2602.05665v1
+
+## 15172. Probabilistic Multi-Regional Solar Power Forecasting with Any-Quantile Recurrent Neural Networks
+
+- **ID:** `2602.05660` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Slawek Smyl, Paweł Pełka, Grzegorz Dudek
+- **abs:** https://arxiv.org/abs/2602.05660 · **PDF:** https://arxiv.org/pdf/2602.05660v1
+
+## 15173. Video-based Music Generation
+
+- **ID:** `2602.07063` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Serkan Sulun
+- **abs:** https://arxiv.org/abs/2602.07063 · **PDF:** https://arxiv.org/pdf/2602.07063v1
+
+## 15174. Tight Long-Term Tail Decay of (Clipped) SGD in Non-Convex Optimization
+
+- **ID:** `2602.05657` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Aleksandar Armacki, Dragana Bajović, Dušan Jakovetić, et al.
+- **abs:** https://arxiv.org/abs/2602.05657 · **PDF:** https://arxiv.org/pdf/2602.05657v1
+
+## 15175. Alignment Verifiability in Large Language Models: Normative Indistinguishability under Behavioral Evaluation
+
+- **ID:** `2602.05656` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Igor Santos-Grueiro
+- **abs:** https://arxiv.org/abs/2602.05656 · **PDF:** https://arxiv.org/pdf/2602.05656v2
+
+## 15176. Enhancing Personality Recognition by Comparing the Predictive Power of Traits, Facets, and Nuances
+
+- **ID:** `2602.05650` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Amir Ansari, Jana Subirana, Bruna Silva, et al.
+- **abs:** https://arxiv.org/abs/2602.05650 · **PDF:** https://arxiv.org/pdf/2602.05650v1
+
+## 15177. End-to-End Compression for Tabular Foundation Models
+
+- **ID:** `2602.05649` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Guri Zabërgja, Rafiq Kamel, Arlind Kadra, et al.
+- **abs:** https://arxiv.org/abs/2602.05649 · **PDF:** https://arxiv.org/pdf/2602.05649v1
+
+## 15178. Modelling the Morphology of Verbal Paradigms: A Case Study in the Tokenization of Turkish and Hebrew
+
+- **ID:** `2602.05648` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Giuseppe Samo, Paola Merlo
+- **abs:** https://arxiv.org/abs/2602.05648 · **PDF:** https://arxiv.org/pdf/2602.05648v1
+
+## 15179. TRACE: Trajectory-Aware Comprehensive Evaluation for Deep Research Agents
+
+- **ID:** `2602.21230` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Yanyu Chen, Jiyue Jiang, Jiahong Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.21230 · **PDF:** https://arxiv.org/pdf/2602.21230v1
+
+## 15180. Empowering Time Series Analysis with Large-Scale Multimodal Pretraining
+
+- **ID:** `2602.05646` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Peng Chen, Siyuan Wang, Shiyan Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.05646 · **PDF:** https://arxiv.org/pdf/2602.05646v1
+
+## 15181. Artificial Organisations
+
+- **ID:** `2602.13275` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** William Waites
+- **abs:** https://arxiv.org/abs/2602.13275 · **PDF:** https://arxiv.org/pdf/2602.13275v1
+
+## 15182. UAV Trajectory Optimization via Improved Noisy Deep Q-Network
+
+- **ID:** `2602.05644` · **primary:** `eess.SY` · **Submitted:** 2026-02-05
+- **Authors:** Zhang Hengyu, Maryam Cheraghy, Liu Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.05644 · **PDF:** https://arxiv.org/pdf/2602.05644v1
+
+## 15183. Joint Embedding Variational Bayes
+
+- **ID:** `2602.05639` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Amin Oji, Paul Fieguth
+- **abs:** https://arxiv.org/abs/2602.05639 · **PDF:** https://arxiv.org/pdf/2602.05639v3
+
+## 15184. Generative Ontology: When Structured Knowledge Learns to Create
+
+- **ID:** `2602.05636` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Benny Cheung
+- **abs:** https://arxiv.org/abs/2602.05636 · **PDF:** https://arxiv.org/pdf/2602.05636v2
+
+## 15185. Structural Disentanglement in Bilinear MLPs via Architectural Inductive Bias
+
+- **ID:** `2602.05635` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Ojasva Nema, Kaustubh Sharma, Aditya Chauhan, et al.
+- **abs:** https://arxiv.org/abs/2602.05635 · **PDF:** https://arxiv.org/pdf/2602.05635v1
+
+## 15186. CASTLE: A Comprehensive Benchmark for Evaluating Student-Tailored Personalized Safety in Large Language Models
+
+- **ID:** `2602.05633` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Rui Jia, Ruiyi Lan, Fengrui Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.05633 · **PDF:** https://arxiv.org/pdf/2602.05633v1
+
+## 15187. Rewards as Labels: Revisiting RLVR from a Classification Perspective
+
+- **ID:** `2602.05630` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Zepeng Zhai, Meilin Chen, Jiaxuan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.05630 · **PDF:** https://arxiv.org/pdf/2602.05630v4
+
+## 15188. AI chatbots versus human healthcare professionals: a systematic review and meta-analysis of empathy in patient care
+
+- **ID:** `2602.05628` · **primary:** `cs.HC` · **Submitted:** 2026-02-05
+- **Authors:** Alastair Howcroft, Amber Bennett-Weston, Ahmad Khan, et al.
+- **abs:** https://arxiv.org/abs/2602.05628 · **PDF:** https://arxiv.org/pdf/2602.05628v1
+
+## 15189. Agentic Workflow Using RBA$_θ$ for Event Prediction
+
+- **ID:** `2602.06097` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Purbak Sengupta, Sambeet Mishra, Sonal Shreya
+- **abs:** https://arxiv.org/abs/2602.06097 · **PDF:** https://arxiv.org/pdf/2602.06097v1
+
+## 15190. Reactive Knowledge Representation and Asynchronous Reasoning
+
+- **ID:** `2602.05625` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Simon Kohaut, Benedict Flade, Julian Eggert, et al.
+- **abs:** https://arxiv.org/abs/2602.05625 · **PDF:** https://arxiv.org/pdf/2602.05625v2
+
+## 15191. Mode-Dependent Rectification for Stable PPO Training
+
+- **ID:** `2602.05619` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Mohamad Mohamad, Francesco Ponzio, Xavier Descombes
+- **abs:** https://arxiv.org/abs/2602.05619 · **PDF:** https://arxiv.org/pdf/2602.05619v1
+
+## 15192. Path-Guided Flow Matching for Dataset Distillation
+
+- **ID:** `2602.05616` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Xuhui Li, Zhengquan Luo, Xiwei Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.05616 · **PDF:** https://arxiv.org/pdf/2602.05616v1
+
+## 15193. Shiva-DiT: Residual-Based Differentiable Top-$k$ Selection for Efficient Diffusion Transformers
+
+- **ID:** `2602.05605` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Jiaji Zhang, Hailiang Zhao, Guoxuan Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.05605 · **PDF:** https://arxiv.org/pdf/2602.05605v1
+
+## 15194. On the Superlinear Relationship between SGD Noise Covariance and Loss Landscape Curvature
+
+- **ID:** `2602.05600` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Yikuan Zhang, Ning Yang, Yuhai Tu
+- **abs:** https://arxiv.org/abs/2602.05600 · **PDF:** https://arxiv.org/pdf/2602.05600v1
+
+## 15195. BhashaSetu: Cross-Lingual Knowledge Transfer from High-Resource to Extreme Low-Resource Languages
+
+- **ID:** `2602.05599` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Subhadip Maji, Arnab Bhattacharya
+- **abs:** https://arxiv.org/abs/2602.05599 · **PDF:** https://arxiv.org/pdf/2602.05599v1
+
+## 15196. CAViT -- Channel-Aware Vision Transformer for Dynamic Feature Fusion
+
+- **ID:** `2602.05598` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Aon Safdar, Mohamed Saadeldin
+- **abs:** https://arxiv.org/abs/2602.05598 · **PDF:** https://arxiv.org/pdf/2602.05598v1
+
+## 15197. Emulating Aggregate Human Choice Behavior and Biases with GPT Conversational Agents
+
+- **ID:** `2602.05597` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Stephen Pilli, Vivek Nallur
+- **abs:** https://arxiv.org/abs/2602.05597 · **PDF:** https://arxiv.org/pdf/2602.05597v1
+
+## 15198. Efficient Algorithms for Robust Markov Decision Processes with $s$-Rectangular Ambiguity Sets
+
+- **ID:** `2602.05591` · **primary:** `math.OC` · **Submitted:** 2026-02-05
+- **Authors:** Chin Pang Ho, Marek Petrik, Wolfram Wiesemann
+- **abs:** https://arxiv.org/abs/2602.05591 · **PDF:** https://arxiv.org/pdf/2602.05591v1
+
+## 15199. OpenMAG: A Comprehensive Benchmark for Multimodal-Attributed Graph
+
+- **ID:** `2602.05576` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Chenxi Wan, Xunkai Li, Yilong Zuo, et al.
+- **abs:** https://arxiv.org/abs/2602.05576 · **PDF:** https://arxiv.org/pdf/2602.05576v1
+
+## 15200. EdgeMask-DG*: Learning Domain-Invariant Graph Structures via Adversarial Edge Masking
+
+- **ID:** `2602.05571` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Rishabh Bhattacharya, Naresh Manwani
+- **abs:** https://arxiv.org/abs/2602.05571 · **PDF:** https://arxiv.org/pdf/2602.05571v1
+
+## 15201. TangramSR: Can Vision-Language Models Reason in Continuous Geometric Space?
+
+- **ID:** `2602.05570` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Yikun Zong, Cheston Tan
+- **abs:** https://arxiv.org/abs/2602.05570 · **PDF:** https://arxiv.org/pdf/2602.05570v1
+
+## 15202. MAGPrompt: Message-Adaptive Graph Prompt Tuning for Graph Neural Networks
+
+- **ID:** `2602.05567` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Long D. Nguyen, Binh P. Nguyen
+- **abs:** https://arxiv.org/abs/2602.05567 · **PDF:** https://arxiv.org/pdf/2602.05567v1
+
+## 15203. Piecewise Deterministic Markov Processes for Bayesian Inference of PDE Coefficients
+
+- **ID:** `2602.05559` · **primary:** `stat.CO` · **Submitted:** 2026-02-05
+- **Authors:** Leon Riccius, Iuri B. C. M. Rocha, Joris Bierkens, et al.
+- **abs:** https://arxiv.org/abs/2602.05559 · **PDF:** https://arxiv.org/pdf/2602.05559v1
+
+## 15204. ArkTS-CodeSearch: A Open-Source ArkTS Dataset for Code Retrieval
+
+- **ID:** `2602.05550` · **primary:** `cs.SE` · **Submitted:** 2026-02-05
+- **Authors:** Yulong He, Artem Ermakov, Sergey Kovalchuk, et al.
+- **abs:** https://arxiv.org/abs/2602.05550 · **PDF:** https://arxiv.org/pdf/2602.05550v2
+
+## 15205. TACIT: Transformation-Aware Capturing of Implicit Thought
+
+- **ID:** `2602.07061` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Daniel Nobrega
+- **abs:** https://arxiv.org/abs/2602.07061 · **PDF:** https://arxiv.org/pdf/2602.07061v1
+
+## 15206. Logical Guidance for the Exact Composition of Diffusion Models
+
+- **ID:** `2602.05549` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Francesco Alesiani, Jonathan Warrell, Tanja Bien, et al.
+- **abs:** https://arxiv.org/abs/2602.05549 · **PDF:** https://arxiv.org/pdf/2602.05549v2
+
+## 15207. Unveiling Implicit Advantage Symmetry: Why GRPO Struggles with Exploration and Difficulty Adaptation
+
+- **ID:** `2602.05548` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Zhiqi Yu, Zhangquan Chen, Mengting Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.05548 · **PDF:** https://arxiv.org/pdf/2602.05548v3
+
+## 15208. Multi-Task GRPO: Reliable LLM Reasoning Across Tasks
+
+- **ID:** `2602.05547` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Shyam Sundhar Ramesh, Xiaotong Ji, Matthieu Zimmer, et al.
+- **abs:** https://arxiv.org/abs/2602.05547 · **PDF:** https://arxiv.org/pdf/2602.05547v1
+
+## 15209. Reasoning-guided Collaborative Filtering with Language Models for Explainable Recommendation
+
+- **ID:** `2602.05544` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Fahad Anwaar, Adil Mehmood Khan, Muhammad Khalid, et al.
+- **abs:** https://arxiv.org/abs/2602.05544 · **PDF:** https://arxiv.org/pdf/2602.05544v1
+
+## 15210. Reducing the Complexity of Matrix Multiplication to $O(N^2log_2N)$ by an Asymptotically Optimal Quantum Algorithm
+
+- **ID:** `2602.05541` · **primary:** `quant-ph` · **Submitted:** 2026-02-05
+- **Authors:** Jiaqi Yao, Ding Liu
+- **abs:** https://arxiv.org/abs/2602.05541 · **PDF:** https://arxiv.org/pdf/2602.05541v2
+
+## 15211. Steering Large Reasoning Models towards Concise Reasoning via Flow Matching
+
+- **ID:** `2602.05539` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Yawei Li, Benjamin Bergner, Yinghan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.05539 · **PDF:** https://arxiv.org/pdf/2602.05539v1
+
+## 15212. When Shared Knowledge Hurts: Spectral Over-Accumulation in Model Merging
+
+- **ID:** `2602.05536` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Yayuan Li, Ze Peng, Jian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.05536 · **PDF:** https://arxiv.org/pdf/2602.05536v1
+
+## 15213. Detecting Misbehaviors of Large Vision-Language Models by Evidential Uncertainty Quantification
+
+- **ID:** `2602.05535` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Tao Huang, Rui Wang, Xiaofei Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.05535 · **PDF:** https://arxiv.org/pdf/2602.05535v2
+
+## 15214. Conditional Diffusion Guidance under Hard Constraint: A Stochastic Analysis Approach
+
+- **ID:** `2602.05533` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Zhengyi Guo, Wenpin Tang, Renyuan Xu
+- **abs:** https://arxiv.org/abs/2602.05533 · **PDF:** https://arxiv.org/pdf/2602.05533v2
+
+## 15215. Split Personality Training: Revealing Latent Knowledge Through Alternate Personalities
+
+- **ID:** `2602.05532` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Florian Dietz, William Wale, Oscar Gilg, et al.
+- **abs:** https://arxiv.org/abs/2602.05532 · **PDF:** https://arxiv.org/pdf/2602.05532v1
+
+## 15216. Solving Stochastic Variational Inequalities without the Bounded Variance Assumption
+
+- **ID:** `2602.05531` · **primary:** `math.OC` · **Submitted:** 2026-02-05
+- **Authors:** Ahmet Alacaoglu, Jun-Hyun Kim
+- **abs:** https://arxiv.org/abs/2602.05531 · **PDF:** https://arxiv.org/pdf/2602.05531v1
+
+## 15217. AI Agent Systems for Supply Chains: Structured Decision Prompts and Memory Retrieval
+
+- **ID:** `2602.05524` · **primary:** `cs.MA` · **Submitted:** 2026-02-05
+- **Authors:** Konosuke Yoshizato, Kazuma Shimizu, Ryota Higa, et al.
+- **abs:** https://arxiv.org/abs/2602.05524 · **PDF:** https://arxiv.org/pdf/2602.05524v1
+
+## 15218. Capture the Flags: Family-Based Evaluation of Agentic LLMs via Semantics-Preserving Transformations
+
+- **ID:** `2602.05523` · **primary:** `cs.SE` · **Submitted:** 2026-02-05
+- **Authors:** Shahin Honarvar, Amber Gorzynski, James Lee-Jones, et al.
+- **abs:** https://arxiv.org/abs/2602.05523 · **PDF:** https://arxiv.org/pdf/2602.05523v1
+
+## 15219. A Unified Multimodal Framework for Dataset Construction and Model-Based Diagnosis of Ameloblastoma
+
+- **ID:** `2602.05515` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Ajo Babu George, Anna Mariam John, Athul Anoop, et al.
+- **abs:** https://arxiv.org/abs/2602.05515 · **PDF:** https://arxiv.org/pdf/2602.05515v1
+
+## 15220. DECO: Decoupled Multimodal Diffusion Transformer for Bimanual Dexterous Manipulation with a Plugin Tactile Adapter
+
+- **ID:** `2602.05513` · **primary:** `cs.RO` · **Submitted:** 2026-02-05
+- **Authors:** Xukun Li, Yu Sun, Lei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.05513 · **PDF:** https://arxiv.org/pdf/2602.05513v2
+
+## 15221. A Human-in-the-Loop, LLM-Centered Architecture for Knowledge-Graph Question Answering
+
+- **ID:** `2602.05512` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Larissa Pusch, Alexandre Courtiol, Tim Conrad
+- **abs:** https://arxiv.org/abs/2602.05512 · **PDF:** https://arxiv.org/pdf/2602.05512v2
+
+## 15222. ProMoral-Bench: Evaluating Prompting Strategies for Moral Reasoning and Safety in LLMs
+
+- **ID:** `2602.13274` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Rohan Subramanian Thomas, Shikhar Shiromani, Abdullah Chaudhry, et al.
+- **abs:** https://arxiv.org/abs/2602.13274 · **PDF:** https://arxiv.org/pdf/2602.13274v1
+
+## 15223. SDFP: Speculative Decoding with FIT-Pruned Models for Training-Free and Plug-and-Play LLM Acceleration
+
+- **ID:** `2602.05499` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Hanyu Wei, Zunhai Su, Peng Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.05499 · **PDF:** https://arxiv.org/pdf/2602.05499v1
+
+## 15224. XEmoGPT: An Explainable Multimodal Emotion Recognition Framework with Cue-Level Perception and Reasoning
+
+- **ID:** `2602.05496` · **primary:** `cs.MM` · **Submitted:** 2026-02-05
+- **Authors:** Hanwen Zhang, Yao Liu, Peiyuan Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.05496 · **PDF:** https://arxiv.org/pdf/2602.05496v1
+
+## 15225. Transport and Merge: Cross-Architecture Merging for Large Language Models
+
+- **ID:** `2602.05495` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Chenhang Cui, Binyun Yang, Fei Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.05495 · **PDF:** https://arxiv.org/pdf/2602.05495v2
+
+## 15226. A Unified Framework for Rethinking Policy Divergence Measures in GRPO
+
+- **ID:** `2602.05494` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Qingyuan Wu, Yuhui Wang, Simon Sinong Zhan, et al.
+- **abs:** https://arxiv.org/abs/2602.05494 · **PDF:** https://arxiv.org/pdf/2602.05494v2
+
+## 15227. LinguistAgent: A Reflective Multi-Model Platform for Automated Linguistic Annotation
+
+- **ID:** `2602.05493` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Bingru Li
+- **abs:** https://arxiv.org/abs/2602.05493 · **PDF:** https://arxiv.org/pdf/2602.05493v1
+
+## 15228. Convergence Rate of the Last Iterate of Stochastic Proximal Algorithms
+
+- **ID:** `2602.05489` · **primary:** `math.OC` · **Submitted:** 2026-02-05
+- **Authors:** Kevin Kurian Thomas Vaidyan, Michael P. Friedlander, Ahmet Alacaoglu
+- **abs:** https://arxiv.org/abs/2602.05489 · **PDF:** https://arxiv.org/pdf/2602.05489v1
+
+## 15229. Sovereign-by-Design A Reference Architecture for AI and Blockchain Enabled Systems
+
+- **ID:** `2602.05486` · **primary:** `cs.SE` · **Submitted:** 2026-02-05
+- **Authors:** Matteo Esposito, Lodovica Marchesi, Roberto Tonelli, et al.
+- **abs:** https://arxiv.org/abs/2602.05486 · **PDF:** https://arxiv.org/pdf/2602.05486v1
+
+## 15230. Phi-Former: A Pairwise Hierarchical Approach for Compound-Protein Interactions Prediction
+
+- **ID:** `2602.05479` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Zhe Wang, Zijing Liu, Chencheng Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.05479 · **PDF:** https://arxiv.org/pdf/2602.05479v1
+
+## 15231. LLM-driven Multimodal Recommendation
+
+- **ID:** `2602.05474` · **primary:** `cs.IR` · **Submitted:** 2026-02-05
+- **Authors:** Yicheng Di
+- **abs:** https://arxiv.org/abs/2602.05474 · **PDF:** https://arxiv.org/pdf/2602.05474v5
+
+## 15232. ALIVE: Awakening LLM Reasoning via Adversarial Learning and Instructive Verbal Evaluation
+
+- **ID:** `2602.05472` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Yiwen Duan, Jing Ye, Xinpei Zhao
+- **abs:** https://arxiv.org/abs/2602.05472 · **PDF:** https://arxiv.org/pdf/2602.05472v1
+
+## 15233. Reasoning under Ambiguity: Uncertainty-Aware Multilingual Emotion Classification under Partial Supervision
+
+- **ID:** `2602.05471` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Md. Mithun Hossain, Mashary N. Alrasheedy, Nirban Bhowmick, et al.
+- **abs:** https://arxiv.org/abs/2602.05471 · **PDF:** https://arxiv.org/pdf/2602.05471v2
+
+## 15234. MerNav: A Highly Generalizable Memory-Execute-Review Framework for Zero-Shot Object Goal Navigation
+
+- **ID:** `2602.05467` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Dekang Qi, Shuang Zeng, Xinyuan Chang, et al.
+- **abs:** https://arxiv.org/abs/2602.05467 · **PDF:** https://arxiv.org/pdf/2602.05467v1
+
+## 15235. Optimization is Not Enough: Why Problem Formulation Deserves Equal Attention
+
+- **ID:** `2602.05466` · **primary:** `cs.NE` · **Submitted:** 2026-02-05
+- **Authors:** Iván Olarte Rodríguez, Gokhan Serhat, Mariusz Bujny, et al.
+- **abs:** https://arxiv.org/abs/2602.05466 · **PDF:** https://arxiv.org/pdf/2602.05466v1
+
+## 15236. Refine and Purify: Orthogonal Basis Optimization with Null-Space Denoising for Conditional Representation Learning
+
+- **ID:** `2602.05464` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Jiaquan Wang, Yan Lyu, Chen Li, et al.
+- **abs:** https://arxiv.org/abs/2602.05464 · **PDF:** https://arxiv.org/pdf/2602.05464v1
+
+## 15237. Thermodynamic Limits of Physical Intelligence
+
+- **ID:** `2602.05463` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Koichi Takahashi, Yusuke Hayashi
+- **abs:** https://arxiv.org/abs/2602.05463 · **PDF:** https://arxiv.org/pdf/2602.05463v1
+
+## 15238. When Are RL Hyperparameters Benign? A Study in Offline Goal-Conditioned RL
+
+- **ID:** `2602.05459` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Jan Malte Töpperwien, Aditya Mohan, Marius Lindauer
+- **abs:** https://arxiv.org/abs/2602.05459 · **PDF:** https://arxiv.org/pdf/2602.05459v1
+
+## 15239. Ontology-Driven Robotic Specification Synthesis
+
+- **ID:** `2602.05456` · **primary:** `cs.RO` · **Submitted:** 2026-02-05
+- **Authors:** Maksym Figat, Ryan M. Mackey, Michel D. Ingham
+- **abs:** https://arxiv.org/abs/2602.05456 · **PDF:** https://arxiv.org/pdf/2602.05456v1
+
+## 15240. Attention Retention for Continual Learning with Vision Transformers
+
+- **ID:** `2602.05454` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Yue Lu, Xiangyu Zhou, Shizhou Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.05454 · **PDF:** https://arxiv.org/pdf/2602.05454v1
+
+## 15241. Towards Segmenting the Invisible: An End-to-End Registration and Segmentation Framework for Weakly Supervised Tumour Analysis
+
+- **ID:** `2602.05453` · **primary:** `eess.IV` · **Submitted:** 2026-02-05
+- **Authors:** Budhaditya Mukhopadhyay, Chirag Mandal, Pavan Tummala, et al.
+- **abs:** https://arxiv.org/abs/2602.05453 · **PDF:** https://arxiv.org/pdf/2602.05453v1
+
+## 15242. DisCa: Accelerating Video Diffusion Transformers with Distillation-Compatible Learnable Feature Caching
+
+- **ID:** `2602.05449` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Chang Zou, Changlin Li, Yang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.05449 · **PDF:** https://arxiv.org/pdf/2602.05449v2
+
+## 15243. BLITZRANK: Principled Zero-shot Ranking Agents with Tournament Graphs
+
+- **ID:** `2602.05448` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Sheshansh Agrawal, Thien Hang Nguyen, Douwe Kiela
+- **abs:** https://arxiv.org/abs/2602.05448 · **PDF:** https://arxiv.org/pdf/2602.05448v3
+
+## 15244. Structured Context Engineering for File-Native Agentic Systems: Evaluating Schema Accuracy, Format Effectiveness, and Multi-File Navigation at Scale
+
+- **ID:** `2602.05447` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Damon McMillan
+- **abs:** https://arxiv.org/abs/2602.05447 · **PDF:** https://arxiv.org/pdf/2602.05447v2
+
+## 15245. interwhen: A Generalizable Framework for Verifiable Reasoning with Test-time Monitors
+
+- **ID:** `2602.11202` · **primary:** `cs.LO` · **Submitted:** 2026-02-05
+- **Authors:** Vishak K Bhat, Prateek Chanda, Vijval Ekbote, et al.
+- **abs:** https://arxiv.org/abs/2602.11202 · **PDF:** https://arxiv.org/pdf/2602.11202v2
+
+## 15246. Causal Front-Door Adjustment for Robust Jailbreak Attacks on LLMs
+
+- **ID:** `2602.05444` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Yao Zhou, Zeen Song, Wenwen Qiang, et al.
+- **abs:** https://arxiv.org/abs/2602.05444 · **PDF:** https://arxiv.org/pdf/2602.05444v2
+
+## 15247. Assessing Reproducibility in Evolutionary Computation: A Case Study using Human- and LLM-based Assessment
+
+- **ID:** `2602.07059` · **primary:** `cs.NE` · **Submitted:** 2026-02-05
+- **Authors:** Francesca Da Ros, Tarik Začiragić, Aske Plaat, et al.
+- **abs:** https://arxiv.org/abs/2602.07059 · **PDF:** https://arxiv.org/pdf/2602.07059v1
+
+## 15248. NanoNet: Parameter-Efficient Learning with Label-Scarce Supervision for Lightweight Text Mining Model
+
+- **ID:** `2602.06093` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Qianren Mao, Yashuo Luo, Ziqi Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.06093 · **PDF:** https://arxiv.org/pdf/2602.06093v1
+
+## 15249. Benchmarking Affordance Generalization with BusyBox
+
+- **ID:** `2602.05441` · **primary:** `cs.RO` · **Submitted:** 2026-02-05
+- **Authors:** Dean Fortier, Timothy Adamson, Tess Hellebrekers, et al.
+- **abs:** https://arxiv.org/abs/2602.05441 · **PDF:** https://arxiv.org/pdf/2602.05441v1
+
+## 15250. Once Correct, Still Wrong: Counterfactual Hallucination in Multilingual Vision-Language Models
+
+- **ID:** `2602.05437` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Basel Mousi, Fahim Dalvi, Shammur Chowdhury, et al.
+- **abs:** https://arxiv.org/abs/2602.05437 · **PDF:** https://arxiv.org/pdf/2602.05437v1
+
+## 15251. Beyond single-channel agentic benchmarking
+
+- **ID:** `2602.18456` · **primary:** `cs.CY` · **Submitted:** 2026-02-05
+- **Authors:** Nelu D. Radpour
+- **abs:** https://arxiv.org/abs/2602.18456 · **PDF:** https://arxiv.org/pdf/2602.18456v1
+
+## 15252. Day-Ahead Electricity Price Forecasting for Volatile Markets Using Foundation Models with Regularization Strategy
+
+- **ID:** `2602.05430` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Kritchanat Ponyuenyong, Pengyu Tu, Jia Wei Tan, et al.
+- **abs:** https://arxiv.org/abs/2602.05430 · **PDF:** https://arxiv.org/pdf/2602.05430v1
+
+## 15253. M$^2$-Miner: Multi-Agent Enhanced MCTS for Mobile GUI Agent Data Mining
+
+- **ID:** `2602.05429` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Rui Lv, Juncheng Mo, Tianyi Chu, et al.
+- **abs:** https://arxiv.org/abs/2602.05429 · **PDF:** https://arxiv.org/pdf/2602.05429v1
+
+## 15254. THOR: Inductive Link Prediction over Hyper-Relational Knowledge Graphs
+
+- **ID:** `2602.05424` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Weijian Yu, Yuhuan Lu, Dingqi Yang
+- **abs:** https://arxiv.org/abs/2602.05424 · **PDF:** https://arxiv.org/pdf/2602.05424v2
+
+## 15255. Disco: Densely-overlapping Cell Instance Segmentation via Adjacency-aware Collaborative Coloring
+
+- **ID:** `2602.05420` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Rui Sun, Yiwen Yang, Kaiyu Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.05420 · **PDF:** https://arxiv.org/pdf/2602.05420v1
+
+## 15256. Grammatical Error Correction Evaluation by Optimally Transporting Edit Representation
+
+- **ID:** `2602.05419` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Takumi Goto, Yusuke Sakai, Taro Watanabe
+- **abs:** https://arxiv.org/abs/2602.05419 · **PDF:** https://arxiv.org/pdf/2602.05419v1
+
+## 15257. Reduced-Order Surrogates for Forced Flexible Mesh Coastal-Ocean Models
+
+- **ID:** `2602.05416` · **primary:** `cs.CE` · **Submitted:** 2026-02-05
+- **Authors:** Freja Høgholm Petersen, Jesper Sandvig Mariegaard, Rocco Palmitessa, et al.
+- **abs:** https://arxiv.org/abs/2602.05416 · **PDF:** https://arxiv.org/pdf/2602.05416v1
+
+## 15258. SciDef: Automating Definition Extraction from Academic Literature with Large Language Models
+
+- **ID:** `2602.05413` · **primary:** `cs.IR` · **Submitted:** 2026-02-05
+- **Authors:** Filip Kučera, Christoph Mandl, Isao Echizen, et al.
+- **abs:** https://arxiv.org/abs/2602.05413 · **PDF:** https://arxiv.org/pdf/2602.05413v1
+
+## 15259. Robust Federated Learning via Byzantine Filtering over Encrypted Updates
+
+- **ID:** `2602.05410` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Adda Akram Bendoukha, Aymen Boudguiga, Nesrine Kaaniche, et al.
+- **abs:** https://arxiv.org/abs/2602.05410 · **PDF:** https://arxiv.org/pdf/2602.05410v2
+
+## 15260. H-AdminSim: A Multi-Agent Simulator for Realistic Hospital Administrative Workflows with FHIR Integration
+
+- **ID:** `2602.05407` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Jun-Min Lee, Meong Hi Son, Edward Choi
+- **abs:** https://arxiv.org/abs/2602.05407 · **PDF:** https://arxiv.org/pdf/2602.05407v1
+
+## 15261. Enabling Automatic Disordered Speech Recognition: An Impaired Speech Dataset in the Akan Language
+
+- **ID:** `2602.05406` · **primary:** `cs.SD` · **Submitted:** 2026-02-05
+- **Authors:** Isaac Wiafe, Akon Obu Ekpezu, Sumaya Ahmed Salihs, et al.
+- **abs:** https://arxiv.org/abs/2602.05406 · **PDF:** https://arxiv.org/pdf/2602.05406v1
+
+## 15262. Advancing Opinion Dynamics Modeling with Neural Diffusion-Convection-Reaction Equation
+
+- **ID:** `2602.05403` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Chenghua Gong, Yihang Jiang, Hao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.05403 · **PDF:** https://arxiv.org/pdf/2602.05403v1
+
+## 15263. OPUS: Towards Efficient and Principled Data Selection in Large Language Model Pre-training in Every Iteration
+
+- **ID:** `2602.05400` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Shaobo Wang, Xuan Ouyang, Tianyi Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.05400 · **PDF:** https://arxiv.org/pdf/2602.05400v2
+
+## 15264. Optimal Bayesian Stopping for Efficient Inference of Consistent LLM Answers
+
+- **ID:** `2602.05395` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Jingkai Huang, Will Ma, Zhengyuan Zhou
+- **abs:** https://arxiv.org/abs/2602.05395 · **PDF:** https://arxiv.org/pdf/2602.05395v1
+
+## 15265. Late-to-Early Training: LET LLMs Learn Earlier, So Faster and Better
+
+- **ID:** `2602.05393` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Ji Zhao, Yufei Gu, Shitong Shao, et al.
+- **abs:** https://arxiv.org/abs/2602.05393 · **PDF:** https://arxiv.org/pdf/2602.05393v1
+
+## 15266. Beyond Length: Context-Aware Expansion and Independence as Developmentally Sensitive Evaluation in Child Utterances
+
+- **ID:** `2602.05392` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Jiyun Chun, Eric Fosler-Lussier, Michael White, et al.
+- **abs:** https://arxiv.org/abs/2602.05392 · **PDF:** https://arxiv.org/pdf/2602.05392v1
+
+## 15267. Assessing Electricity Demand Forecasting with Exogenous Data in Time Series Foundation Models
+
+- **ID:** `2602.05390` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Wei Soon Cheong, Lian Lian Jiang, Jamie Ng Suat Ling
+- **abs:** https://arxiv.org/abs/2602.05390 · **PDF:** https://arxiv.org/pdf/2602.05390v1
+
+## 15268. A Decomposition-based State Space Model for Multivariate Time-Series Forecasting
+
+- **ID:** `2602.05389` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Shunya Nagashima, Shuntaro Suzuki, Shuitsu Koyama, et al.
+- **abs:** https://arxiv.org/abs/2602.05389 · **PDF:** https://arxiv.org/pdf/2602.05389v1
+
+## 15269. Parallel Swin Transformer-Enhanced 3D MRI-to-CT Synthesis for MRI-Only Radiotherapy Planning
+
+- **ID:** `2602.05387` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Zolnamar Dorjsembe, Hung-Yi Chen, Furen Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.05387 · **PDF:** https://arxiv.org/pdf/2602.05387v1
+
+## 15270. Spider-Sense: Intrinsic Risk Sensing for Efficient Agent Defense with Hierarchical Adaptive Screening
+
+- **ID:** `2602.05386` · **primary:** `cs.CR` · **Submitted:** 2026-02-05
+- **Authors:** Zhenxiong Yu, Zhi Yang, Zhiheng Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.05386 · **PDF:** https://arxiv.org/pdf/2602.05386v2
+
+## 15271. IESR:Efficient MCTS-Based Modular Reasoning for Text-to-SQL with Large Language Models
+
+- **ID:** `2602.05385` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Tao Liu, Jiafan Lu, Bohan Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.05385 · **PDF:** https://arxiv.org/pdf/2602.05385v1
+
+## 15272. VRIQ: Benchmarking and Analyzing Visual-Reasoning IQ of VLMs
+
+- **ID:** `2602.05382` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Tina Khezresmaeilzadeh, Jike Zhong, Konstantinos Psounis
+- **abs:** https://arxiv.org/abs/2602.05382 · **PDF:** https://arxiv.org/pdf/2602.05382v1
+
+## 15273. Clinical Validation of Medical-based Large Language Model Chatbots on Ophthalmic Patient Queries with LLM-based Evaluation
+
+- **ID:** `2602.05381` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Ting Fang Tan, Kabilan Elangovan, Andreas Pollreisz, et al.
+- **abs:** https://arxiv.org/abs/2602.05381 · **PDF:** https://arxiv.org/pdf/2602.05381v1
+
+## 15274. Variance Reduction Based Experience Replay for Policy Optimization
+
+- **ID:** `2602.05379` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Hua Zheng, Wei Xie, M. Ben Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.05379 · **PDF:** https://arxiv.org/pdf/2602.05379v1
+
+## 15275. Erase at the Core: Representation Unlearning for Machine Unlearning
+
+- **ID:** `2602.05375` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Jaewon Lee, Yongwoo Kim, Donghyun Kim
+- **abs:** https://arxiv.org/abs/2602.05375 · **PDF:** https://arxiv.org/pdf/2602.05375v2
+
+## 15276. Cross-Lingual Empirical Evaluation of Large Language Models for Arabic Medical Tasks
+
+- **ID:** `2602.05374` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Chaimae Abouzahir, Congbo Ma, Nizar Habash, et al.
+- **abs:** https://arxiv.org/abs/2602.05374 · **PDF:** https://arxiv.org/pdf/2602.05374v1
+
+## 15277. Hinge Regression Tree: A Newton Method for Oblique Regression Tree Splitting
+
+- **ID:** `2602.05371` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Hongyi Li, Han Lin, Jun Xu
+- **abs:** https://arxiv.org/abs/2602.05371 · **PDF:** https://arxiv.org/pdf/2602.05371v2
+
+## 15278. PACE: Defying the Scaling Hypothesis of Exploration in Iterative Alignment for Mathematical Reasoning
+
+- **ID:** `2602.05370` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Jun Rao, Zixiong Yu, Xuebo Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.05370 · **PDF:** https://arxiv.org/pdf/2602.05370v2
+
+## 15279. RaBiT: Residual-Aware Binarization Training for Accurate and Efficient LLMs
+
+- **ID:** `2602.05367` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Youngcheon You, Banseok Lee, Minseop Choi, et al.
+- **abs:** https://arxiv.org/abs/2602.05367 · **PDF:** https://arxiv.org/pdf/2602.05367v1
+
+## 15280. Multi-Field Tool Retrieval
+
+- **ID:** `2602.05366` · **primary:** `cs.IR` · **Submitted:** 2026-02-05
+- **Authors:** Yichen Tang, Weihang Su, Yiqun Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.05366 · **PDF:** https://arxiv.org/pdf/2602.05366v1
+
+## 15281. Bayesian Neighborhood Adaptation for Graph Neural Networks
+
+- **ID:** `2602.05358` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Paribesh Regmi, Rui Li, Kishan KC
+- **abs:** https://arxiv.org/abs/2602.05358 · **PDF:** https://arxiv.org/pdf/2602.05358v2
+
+## 15282. SVRepair: Structured Visual Reasoning for Automated Program Repair
+
+- **ID:** `2602.06090` · **primary:** `cs.SE` · **Submitted:** 2026-02-05
+- **Authors:** Xiaoxuan Tang, Jincheng Wang, Liwei Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.06090 · **PDF:** https://arxiv.org/pdf/2602.06090v1
+
+## 15283. PATHWAYS: Evaluating Investigation and Context Discovery in AI Web Agents
+
+- **ID:** `2602.05354` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Shifat E. Arman, Syed Nazmus Sakib, Tapodhir Karmakar Taton, et al.
+- **abs:** https://arxiv.org/abs/2602.05354 · **PDF:** https://arxiv.org/pdf/2602.05354v2
+
+## 15284. AgentXRay: White-Boxing Agentic Systems via Workflow Reconstruction
+
+- **ID:** `2602.05353` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Ruijie Shi, Houbin Zhang, Yuecheng Han, et al.
+- **abs:** https://arxiv.org/abs/2602.05353 · **PDF:** https://arxiv.org/pdf/2602.05353v2
+
+## 15285. Smoothness Errors in Dynamics Models and How to Avoid Them
+
+- **ID:** `2602.05352` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Edward Berman, Luisa Li, Jung Yeon Park, et al.
+- **abs:** https://arxiv.org/abs/2602.05352 · **PDF:** https://arxiv.org/pdf/2602.05352v1
+
+## 15286. How Do Language Models Acquire Character-Level Information?
+
+- **ID:** `2602.05347` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Soma Sato, Ryohei Sasano
+- **abs:** https://arxiv.org/abs/2602.05347 · **PDF:** https://arxiv.org/pdf/2602.05347v1
+
+## 15287. Decision-Focused Sequential Experimental Design: A Directional Uncertainty-Guided Approach
+
+- **ID:** `2602.05340` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Beichen Wan, Mo Liu, Paul Grigas, et al.
+- **abs:** https://arxiv.org/abs/2602.05340 · **PDF:** https://arxiv.org/pdf/2602.05340v1
+
+## 15288. Consistency-Preserving Concept Erasure via Unsafe-Safe Pairing and Directional Fisher-weighted Adaptation
+
+- **ID:** `2602.05339` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Yongwoo Kim, Sungmin Cha, Hyunsoo Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.05339 · **PDF:** https://arxiv.org/pdf/2602.05339v1
+
+## 15289. Pool-based Active Learning as Noisy Lossy Compression: Characterizing Label Complexity via Finite Blocklength Analysis
+
+- **ID:** `2602.05333` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Kosuke Sugiyama, Masato Uchida
+- **abs:** https://arxiv.org/abs/2602.05333 · **PDF:** https://arxiv.org/pdf/2602.05333v1
+
+## 15290. ProAct: Agentic Lookahead in Interactive Environments
+
+- **ID:** `2602.05327` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Yangbin Yu, Mingyu Yang, Junyou Li, et al.
+- **abs:** https://arxiv.org/abs/2602.05327 · **PDF:** https://arxiv.org/pdf/2602.05327v1
+
+## 15291. GAS: Enhancing Reward-Cost Balance of Generative Model-assisted Offline Safe RL
+
+- **ID:** `2602.05323` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Zifan Liu, Xinran Li, Shibo Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.05323 · **PDF:** https://arxiv.org/pdf/2602.05323v1
+
+## 15292. Accelerated Sequential Flow Matching: A Bayesian Filtering Perspective
+
+- **ID:** `2602.05319` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Yinan Huang, Hans Hao-Hsun Hsu, Junran Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.05319 · **PDF:** https://arxiv.org/pdf/2602.05319v2
+
+## 15293. Curriculum Learning for Efficient Chain-of-Thought Distillation via Structure-Aware Masking and GRPO
+
+- **ID:** `2602.17686` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Bowen Yu, Maolin Wang, Sheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.17686 · **PDF:** https://arxiv.org/pdf/2602.17686v3
+
+## 15294. Formal Synthesis of Certifiably Robust Neural Lyapunov-Barrier Certificates
+
+- **ID:** `2602.05311` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Chengxiao Wang, Haoze Wu, Gagandeep Singh
+- **abs:** https://arxiv.org/abs/2602.05311 · **PDF:** https://arxiv.org/pdf/2602.05311v1
+
+## 15295. MergePipe: A Budget-Aware Parameter Management System for Scalable LLM Merging
+
+- **ID:** `2602.13273` · **primary:** `cs.DB` · **Submitted:** 2026-02-05
+- **Authors:** Yuanyi Wang, Yanggan Gu, Zihao Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13273 · **PDF:** https://arxiv.org/pdf/2602.13273v1
+
+## 15296. MentorCollab: Selective Large-to-Small Inference-Time Guidance for Efficient Reasoning
+
+- **ID:** `2602.05307` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Haojin Wang, Yike Wang, Shangbin Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.05307 · **PDF:** https://arxiv.org/pdf/2602.05307v1
+
+## 15297. Transformer-Based Reinforcement Learning for Autonomous Orbital Collision Avoidance in Partially Observable Environments
+
+- **ID:** `2602.06088` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Thomas Georges, Adam Abdin
+- **abs:** https://arxiv.org/abs/2602.06088 · **PDF:** https://arxiv.org/pdf/2602.06088v1
+
+## 15298. FlashBlock: Attention Caching for Efficient Long-Context Block Diffusion
+
+- **ID:** `2602.05305` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Zhuokun Chen, Jianfei Cai, Bohan Zhuang
+- **abs:** https://arxiv.org/abs/2602.05305 · **PDF:** https://arxiv.org/pdf/2602.05305v2
+
+## 15299. A Short and Unified Convergence Analysis of the SAG, SAGA, and IAG Algorithms
+
+- **ID:** `2602.05304` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Feng Zhu, Robert W. Heath, Aritra Mitra
+- **abs:** https://arxiv.org/abs/2602.05304 · **PDF:** https://arxiv.org/pdf/2602.05304v1
+
+## 15300. PieArena: Frontier Language Agents Achieve MBA-Level Negotiation Performance and Reveal Novel Behavioral Differences
+
+- **ID:** `2602.05302` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Chris Zhu, Sasha Cui, Will Sanok Dufallo, et al.
+- **abs:** https://arxiv.org/abs/2602.05302 · **PDF:** https://arxiv.org/pdf/2602.05302v2
+
+## 15301. Logarithmic-time Schedules for Scaling Language Models with Momentum
+
+- **ID:** `2602.05298` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Damien Ferbach, Courtney Paquette, Gauthier Gidel, et al.
+- **abs:** https://arxiv.org/abs/2602.05298 · **PDF:** https://arxiv.org/pdf/2602.05298v2
+
+## 15302. Aspect-Aware MOOC Recommendation in a Heterogeneous Network
+
+- **ID:** `2602.05297` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Seongyeub Chu, Jongwoo Kim, Mun Yong Yi
+- **abs:** https://arxiv.org/abs/2602.05297 · **PDF:** https://arxiv.org/pdf/2602.05297v1
+
+## 15303. Towards a Science of Collective AI: LLM-based Multi-Agent Systems Need a Transition from Blind Trial-and-Error to Rigorous Science
+
+- **ID:** `2602.05289` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Jingru Fan, Dewen Liu, Yufan Dang, et al.
+- **abs:** https://arxiv.org/abs/2602.05289 · **PDF:** https://arxiv.org/pdf/2602.05289v1
+
+## 15304. Position: Universal Time Series Foundation Models Rest on a Category Error
+
+- **ID:** `2602.05287` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Xilin Dai, Wanxu Cai, Zhijian Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.05287 · **PDF:** https://arxiv.org/pdf/2602.05287v1
+
+## 15305. HealthMamba: An Uncertainty-aware Spatiotemporal Graph State Space Model for Effective and Reliable Healthcare Facility Visit Prediction
+
+- **ID:** `2602.05286` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Dahai Yu, Lin Jiang, Rongchao Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.05286 · **PDF:** https://arxiv.org/pdf/2602.05286v2
+
+## 15306. Robust Inference-Time Steering of Protein Diffusion Models via Embedding Optimization
+
+- **ID:** `2602.05285` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Minhuan Li, Jiequn Han, Pilar Cossio, et al.
+- **abs:** https://arxiv.org/abs/2602.05285 · **PDF:** https://arxiv.org/pdf/2602.05285v1
+
+## 15307. Back to Basics: Revisiting Exploration in Reinforcement Learning for LLM Reasoning via Generative Probabilities
+
+- **ID:** `2602.05281` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Pengyi Li, Elizaveta Goncharova, Andrey Kuznetsov, et al.
+- **abs:** https://arxiv.org/abs/2602.05281 · **PDF:** https://arxiv.org/pdf/2602.05281v2
+
+## 15308. Hallucination-Resistant Security Planning with a Large Language Model
+
+- **ID:** `2602.05279` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Kim Hammar, Tansu Alpcan, Emil Lupu
+- **abs:** https://arxiv.org/abs/2602.05279 · **PDF:** https://arxiv.org/pdf/2602.05279v1
+
+## 15309. Asymptotically optimal sequential change detection for bounded means
+
+- **ID:** `2602.05272` · **primary:** `math.ST` · **Submitted:** 2026-02-05
+- **Authors:** Ashwin Ram, Aaditya Ramdas
+- **abs:** https://arxiv.org/abs/2602.05272 · **PDF:** https://arxiv.org/pdf/2602.05272v1
+
+## 15310. Hybrid Gated Flow (HGF): Stabilizing 1.58-bit LLMs via Selective Low-Rank Correction
+
+- **ID:** `2602.05269` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** David Alejandro Trejo Pizzo
+- **abs:** https://arxiv.org/abs/2602.05269 · **PDF:** https://arxiv.org/pdf/2602.05269v1
+
+## 15311. Beyond Cosine Similarity
+
+- **ID:** `2602.05266` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Xinbo Ai
+- **abs:** https://arxiv.org/abs/2602.05266 · **PDF:** https://arxiv.org/pdf/2602.05266v1
+
+## 15312. Length-Unbiased Sequence Policy Optimization: Revealing and Controlling Response Length Variation in RLVR
+
+- **ID:** `2602.05261` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Fanfan Liu, Youyang Yin, Peng Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.05261 · **PDF:** https://arxiv.org/pdf/2602.05261v1
+
+## 15313. An Asymptotic Law of the Iterated Logarithm for $\mathrm{KL}_{\inf}$
+
+- **ID:** `2602.05259` · **primary:** `math.ST` · **Submitted:** 2026-02-05
+- **Authors:** Ashwin Ram, Aaditya Ramdas
+- **abs:** https://arxiv.org/abs/2602.05259 · **PDF:** https://arxiv.org/pdf/2602.05259v1
+
+## 15314. CoPE: Clipped RoPE as A Scalable Free Lunch for Long Context LLMs
+
+- **ID:** `2602.05258` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Haoran Li, Sucheng Ren, Alan Yuille, et al.
+- **abs:** https://arxiv.org/abs/2602.05258 · **PDF:** https://arxiv.org/pdf/2602.05258v1
+
+## 15315. Copyright Detective: A Forensic System to Evidence LLMs Flickering Copyright Leakage Risks
+
+- **ID:** `2602.05252` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Guangwei Zhang, Jianing Zhu, Cheng Qian, et al.
+- **abs:** https://arxiv.org/abs/2602.05252 · **PDF:** https://arxiv.org/pdf/2602.05252v2
+
+## 15316. TADS: Task-Aware Data Selection for Multi-Task Multimodal Pre-Training
+
+- **ID:** `2602.05251` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Guanjie Cheng, Boyi Li, Lingyu Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.05251 · **PDF:** https://arxiv.org/pdf/2602.05251v1
+
+## 15317. Automatic Cognitive Task Generation for In-Situ Evaluation of Embodied Agents
+
+- **ID:** `2602.05249` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Xinyi He, Ying Yang, Chuanjian Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.05249 · **PDF:** https://arxiv.org/pdf/2602.05249v1
+
+## 15318. CORP: Closed-Form One-shot Representation-Preserving Structured Pruning for Vision Transformers
+
+- **ID:** `2602.05243` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Boxiang Zhang, Baijian Yang
+- **abs:** https://arxiv.org/abs/2602.05243 · **PDF:** https://arxiv.org/pdf/2602.05243v1
+
+## 15319. EGSS: Entropy-guided Stepwise Scaling for Reliable Software Engineering
+
+- **ID:** `2602.05242` · **primary:** `cs.SE` · **Submitted:** 2026-02-05
+- **Authors:** Chenhui Mao, Yuanting Lei, Zhixiang Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.05242 · **PDF:** https://arxiv.org/pdf/2602.05242v1
+
+## 15320. Explainable AI: A Combined XAI Framework for Explaining Brain Tumour Detection Models
+
+- **ID:** `2602.05240` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Patrick McGonagle, William Farrelly, Kevin Curran
+- **abs:** https://arxiv.org/abs/2602.05240 · **PDF:** https://arxiv.org/pdf/2602.05240v1
+
+## 15321. PatchFlow: Leveraging a Flow-Based Model with Patch Features
+
+- **ID:** `2602.05238` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Boxiang Zhang, Baijian Yang, Xiaoming Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.05238 · **PDF:** https://arxiv.org/pdf/2602.05238v1
+
+## 15322. VQ-DSC-R: Robust Vector Quantized-Enabled Digital Semantic Communication With OFDM Transmission
+
+- **ID:** `2602.15045` · **primary:** `cs.IT` · **Submitted:** 2026-02-05
+- **Authors:** Jianqiao Chen, Nan Ma, Xiaodong Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.15045 · **PDF:** https://arxiv.org/pdf/2602.15045v1
+
+## 15323. FedMosaic: Federated Retrieval-Augmented Generation via Parametric Adapters
+
+- **ID:** `2602.05235` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Zhilin Liang, Yuxiang Wang, Zimu Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.05235 · **PDF:** https://arxiv.org/pdf/2602.05235v1
+
+## 15324. Faithful Bi-Directional Model Steering via Distribution Matching and Distributed Interchange Interventions
+
+- **ID:** `2602.05234` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Yuntai Bao, Xuhong Zhang, Jintao Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.05234 · **PDF:** https://arxiv.org/pdf/2602.05234v2
+
+## 15325. Balanced Anomaly-guided Ego-graph Diffusion Model for Inductive Graph Anomaly Detection
+
+- **ID:** `2602.05232` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Chunyu Wei, Siyuan He, Yu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.05232 · **PDF:** https://arxiv.org/pdf/2602.05232v1
+
+## 15326. ZeroS: Zero-Sum Linear Attention for Efficient Transformers
+
+- **ID:** `2602.05230` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Jiecheng Lu, Xu Han, Yan Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.05230 · **PDF:** https://arxiv.org/pdf/2602.05230v1
+
+## 15327. Surgery: Mitigating Harmful Fine-Tuning for Large Language Models via Attention Sink
+
+- **ID:** `2602.05228` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Guozhi Liu, Weiwei Lin, Tiansheng Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.05228 · **PDF:** https://arxiv.org/pdf/2602.05228v2
+
+## 15328. Radon--Wasserstein Gradient Flows for Interacting-Particle Sampling in High Dimensions
+
+- **ID:** `2602.05227` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Elias Hess-Childs, Dejan Slepčev, Lantian Xu
+- **abs:** https://arxiv.org/abs/2602.05227 · **PDF:** https://arxiv.org/pdf/2602.05227v2
+
+## 15329. Metric space valued Fréchet regression
+
+- **ID:** `2602.05225` · **primary:** `math.ST` · **Submitted:** 2026-02-05
+- **Authors:** László Györfi, Pierre Humbert, Batiste Le Bars
+- **abs:** https://arxiv.org/abs/2602.05225 · **PDF:** https://arxiv.org/pdf/2602.05225v1
+
+## 15330. Bagpiper: Solving Open-Ended Audio Tasks via Rich Captions
+
+- **ID:** `2602.05220` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Jinchuan Tian, Haoran Wang, Bo-Hao Su, et al.
+- **abs:** https://arxiv.org/abs/2602.05220 · **PDF:** https://arxiv.org/pdf/2602.05220v2
+
+## 15331. Private Prediction via Shrinkage
+
+- **ID:** `2602.05219` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Chao Yan
+- **abs:** https://arxiv.org/abs/2602.05219 · **PDF:** https://arxiv.org/pdf/2602.05219v1
+
+## 15332. Semantic Search over 9 Million Mathematical Theorems
+
+- **ID:** `2602.05216` · **primary:** `cs.IR` · **Submitted:** 2026-02-05
+- **Authors:** Luke Alexander, Eric Leonen, Sophie Szeto, et al.
+- **abs:** https://arxiv.org/abs/2602.05216 · **PDF:** https://arxiv.org/pdf/2602.05216v2
+
+## 15333. Disentangled Representation Learning via Flow Matching
+
+- **ID:** `2602.05214` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Jinjin Chi, Taoping Liu, Mengtao Yin, et al.
+- **abs:** https://arxiv.org/abs/2602.05214 · **PDF:** https://arxiv.org/pdf/2602.05214v1
+
+## 15334. Quantifying the Knowledge Proximity Between Academic and Industry Research: An Entity and Semantic Perspective
+
+- **ID:** `2602.05211` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Hongye Zhao, Yi Zhao, Chengzhi Zhang
+- **abs:** https://arxiv.org/abs/2602.05211 · **PDF:** https://arxiv.org/pdf/2602.05211v1
+
+## 15335. ARCHI-TTS: A flow-matching-based Text-to-Speech Model with Self-supervised Semantic Aligner and Accelerated Inference
+
+- **ID:** `2602.05207` · **primary:** `eess.AS` · **Submitted:** 2026-02-05
+- **Authors:** Chunyat Wu, Jiajun Deng, Zhengxi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.05207 · **PDF:** https://arxiv.org/pdf/2602.05207v1
+
+## 15336. Aligning Large Language Model Behavior with Human Citation Preferences
+
+- **ID:** `2602.05205` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Kenichiro Ando, Tatsuya Harada
+- **abs:** https://arxiv.org/abs/2602.05205 · **PDF:** https://arxiv.org/pdf/2602.05205v1
+
+## 15337. Extreme Weather Nowcasting via Local Precipitation Pattern Prediction
+
+- **ID:** `2602.05204` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Changhoon Song, Teng Yuan Chang, Youngjoon Hong
+- **abs:** https://arxiv.org/abs/2602.05204 · **PDF:** https://arxiv.org/pdf/2602.05204v2
+
+## 15338. Visual Exclusivity Attacks: Automatic Multimodal Red Teaming via Agentic Planning
+
+- **ID:** `2603.20198` · **primary:** `cs.CR` · **Submitted:** 2026-02-05
+- **Authors:** Yunbei Zhang, Yingqiang Ge, Weijie Xu, et al.
+- **abs:** https://arxiv.org/abs/2603.20198 · **PDF:** https://arxiv.org/pdf/2603.20198v1
+
+## 15339. Traceable Cross-Source RAG for Chinese Tibetan Medicine Question Answering
+
+- **ID:** `2602.05195` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Fengxian Chen, Zhilong Tao, Jiaxuan Li, et al.
+- **abs:** https://arxiv.org/abs/2602.05195 · **PDF:** https://arxiv.org/pdf/2602.05195v2
+
+## 15340. First Proof
+
+- **ID:** `2602.05192` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Mohammed Abouzaid, Andrew J. Blumberg, Martin Hairer, et al.
+- **abs:** https://arxiv.org/abs/2602.05192 · **PDF:** https://arxiv.org/pdf/2602.05192v2
+
+## 15341. Double-P: Hierarchical Top-P Sparse Attention for Long-Context LLMs
+
+- **ID:** `2602.05191` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Wentao Ni, Kangqi Zhang, Zhongming Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.05191 · **PDF:** https://arxiv.org/pdf/2602.05191v1
+
+## 15342. Are Open-Weight LLMs Ready for Social Media Moderation? A Comparative Study on Bluesky
+
+- **ID:** `2602.05189` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Hsuan-Yu Chou, Wajiha Naveed, Shuyan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.05189 · **PDF:** https://arxiv.org/pdf/2602.05189v1
+
+## 15343. Impact of AI Search Summaries on Website Traffic: Evidence from Google AI Overviews and Wikipedia
+
+- **ID:** `2602.18455` · **primary:** `cs.CY` · **Submitted:** 2026-02-05
+- **Authors:** Mehrzad Khosravi, Hema Yoganarasimhan
+- **abs:** https://arxiv.org/abs/2602.18455 · **PDF:** https://arxiv.org/pdf/2602.18455v2
+
+## 15344. SpectraKAN: Conditioning Spectral Operators
+
+- **ID:** `2602.05187` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Chun-Wun Cheng, Carola-Bibiane Schönlieb, Angelica I. Aviles-Rivero
+- **abs:** https://arxiv.org/abs/2602.05187 · **PDF:** https://arxiv.org/pdf/2602.05187v1
+
+## 15345. Towards Worst-Case Guarantees with Scale-Aware Interpretability
+
+- **ID:** `2602.05184` · **primary:** `hep-th` · **Submitted:** 2026-02-05
+- **Authors:** Lauren Greenspan, David Berman, Aryeh Brill, et al.
+- **abs:** https://arxiv.org/abs/2602.05184 · **PDF:** https://arxiv.org/pdf/2602.05184v1
+
+## 15346. Data-Centric Interpretability for LLM-based Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.05183` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** John Yan, Michael Yu, Yuqi Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.05183 · **PDF:** https://arxiv.org/pdf/2602.05183v2
+
+## 15347. The Single-Multi Evolution Loop for Self-Improving Model Collaboration Systems
+
+- **ID:** `2602.05182` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Shangbin Feng, Kishan Panaganti, Yulia Tsvetkov, et al.
+- **abs:** https://arxiv.org/abs/2602.05182 · **PDF:** https://arxiv.org/pdf/2602.05182v2
+
+## 15348. Benchmarking Artificial Intelligence Models for Daily Coastal Hypoxia Forecasting
+
+- **ID:** `2602.05178` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Magesh Rajasekaran, Md Saiful Sajol, Chris Alvin, et al.
+- **abs:** https://arxiv.org/abs/2602.05178 · **PDF:** https://arxiv.org/pdf/2602.05178v1
+
+## 15349. Among Us: Measuring and Mitigating Malicious Contributions in Model Collaboration Systems
+
+- **ID:** `2602.05176` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Ziyuan Yang, Wenxuan Ding, Shangbin Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.05176 · **PDF:** https://arxiv.org/pdf/2602.05176v1
+
+## 15350. Total Variation Rates for Riemannian Flow Matching
+
+- **ID:** `2602.05174` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Yunrui Guan, Krishnakumar Balasubramanian, Shiqian Ma
+- **abs:** https://arxiv.org/abs/2602.05174 · **PDF:** https://arxiv.org/pdf/2602.05174v1
+
+## 15351. TemporalBench: A Benchmark for Evaluating LLM-Based Agents on Contextual and Event-Informed Time Series Tasks
+
+- **ID:** `2602.13272` · **primary:** `cs.AI` · **Submitted:** 2026-02-05
+- **Authors:** Muyan Weng, Defu Cao, Wei Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.13272 · **PDF:** https://arxiv.org/pdf/2602.13272v1
+
+## 15352. Finite-Particle Rates for Regularized Stein Variational Gradient Descent
+
+- **ID:** `2602.05172` · **primary:** `stat.ML` · **Submitted:** 2026-02-05
+- **Authors:** Ye He, Krishnakumar Balasubramanian, Sayan Banerjee, et al.
+- **abs:** https://arxiv.org/abs/2602.05172 · **PDF:** https://arxiv.org/pdf/2602.05172v1
+
+## 15353. Path Sampling for Rare Events Boosted by Machine Learning
+
+- **ID:** `2602.05167` · **primary:** `physics.comp-ph` · **Submitted:** 2026-02-05
+- **Authors:** Porhouy Minh, Sapna Sarupria
+- **abs:** https://arxiv.org/abs/2602.05167 · **PDF:** https://arxiv.org/pdf/2602.05167v1
+
+## 15354. EBPO: Empirical Bayes Shrinkage for Stabilizing Group-Relative Policy Optimization
+
+- **ID:** `2602.05165` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Kevin Han, Yuhang Zhou, Mingze Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.05165 · **PDF:** https://arxiv.org/pdf/2602.05165v3
+
+## 15355. Position: Capability Control Should be a Separate Goal From Alignment
+
+- **ID:** `2602.05164` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Shoaib Ahmed Siddiqui, Eleni Triantafillou, David Krueger, et al.
+- **abs:** https://arxiv.org/abs/2602.05164 · **PDF:** https://arxiv.org/pdf/2602.05164v1
+
+## 15356. SHaSaM: Submodular Hard Sample Mining for Fair Facial Attribute Recognition
+
+- **ID:** `2602.05162` · **primary:** `cs.CV` · **Submitted:** 2026-02-05
+- **Authors:** Anay Majee, Rishabh Iyer
+- **abs:** https://arxiv.org/abs/2602.05162 · **PDF:** https://arxiv.org/pdf/2602.05162v1
+
+## 15357. GreekMMLU: A Native-Sourced Multitask Benchmark for Evaluating Language Models in Greek
+
+- **ID:** `2602.05150` · **primary:** `cs.CL` · **Submitted:** 2026-02-05
+- **Authors:** Yang Zhang, Mersin Konomi, Christos Xypolopoulos, et al.
+- **abs:** https://arxiv.org/abs/2602.05150 · **PDF:** https://arxiv.org/pdf/2602.05150v2
+
+## 15358. CoSA: Compressed Sensing-Based Adaptation of Large Language Models
+
+- **ID:** `2602.05148` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Songtao Wei, Yi Li, Bohan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.05148 · **PDF:** https://arxiv.org/pdf/2602.05148v2
+
+## 15359. Cross-talk based multi-task learning for fault classification of machine system influenced by multiple variables
+
+- **ID:** `2602.05146` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Wonjun Yi, Rismaya Kumar Mishra, Yong-Hwa Park
+- **abs:** https://arxiv.org/abs/2602.05146 · **PDF:** https://arxiv.org/pdf/2602.05146v2
+
+## 15360. TIDE: Temporal Incremental Draft Engine for Self-Improving LLM Inference
+
+- **ID:** `2602.05145` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Jiyoung Park, Hankyu Jang, Changseok Song, et al.
+- **abs:** https://arxiv.org/abs/2602.05145 · **PDF:** https://arxiv.org/pdf/2602.05145v1
+
+## 15361. Fairness Under Group-Conditional Prior Probability Shift: Invariance, Drift, and Target-Aware Post-Processing
+
+- **ID:** `2602.05144` · **primary:** `cs.LG` · **Submitted:** 2026-02-05
+- **Authors:** Amir Asiaee, Kaveh Aryan
+- **abs:** https://arxiv.org/abs/2602.05144 · **PDF:** https://arxiv.org/pdf/2602.05144v1
+
+## 15362. HugRAG: Hierarchical Causal Knowledge Graph Design for RAG
+
+- **ID:** `2602.05143` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Nengbo Wang, Tuo Liang, Vikash Singh, et al.
+- **abs:** https://arxiv.org/abs/2602.05143 · **PDF:** https://arxiv.org/pdf/2602.05143v1
+
+## 15363. Adaptive Exploration for Latent-State Bandits
+
+- **ID:** `2602.05139` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Jikai Jin, Kenneth Hung, Sanath Kumar Krishnamurthy, et al.
+- **abs:** https://arxiv.org/abs/2602.05139 · **PDF:** https://arxiv.org/pdf/2602.05139v2
+
+## 15364. Decoupled Orthogonal Dynamics: Regularization for Deep Network Optimizers
+
+- **ID:** `2602.05136` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Hao Chen, Jinghui Yuan, Hanmin Zhang
+- **abs:** https://arxiv.org/abs/2602.05136 · **PDF:** https://arxiv.org/pdf/2602.05136v1
+
+## 15365. SemPipes -- Optimizable Semantic Data Operators for Tabular Machine Learning Pipelines
+
+- **ID:** `2602.05134` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Olga Ovcharenko, Matthias Boehm, Sebastian Schelter
+- **abs:** https://arxiv.org/abs/2602.05134 · **PDF:** https://arxiv.org/pdf/2602.05134v1
+
+## 15366. CAST-CKT: Chaos-Aware Spatio-Temporal and Cross-City Knowledge Transfer for Traffic Flow Prediction
+
+- **ID:** `2602.05133` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Abdul Joseph Fofanah, Lian Wen, David Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.05133 · **PDF:** https://arxiv.org/pdf/2602.05133v1
+
+## 15367. Rethinking Rubric Generation for Improving LLM Judge and Reward Modeling for Open-ended Tasks
+
+- **ID:** `2602.05125` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** William F. Shen, Xinchi Qiu, Chenxi Whitehouse, et al.
+- **abs:** https://arxiv.org/abs/2602.05125 · **PDF:** https://arxiv.org/pdf/2602.05125v1
+
+## 15368. Certifiable Boolean Reasoning Is Universal
+
+- **ID:** `2602.05120` · **primary:** `cs.CC` · **Submitted:** 2026-02-04
+- **Authors:** Wenhao Li, Anastasis Kratsios, Hrad Ghoukasian, et al.
+- **abs:** https://arxiv.org/abs/2602.05120 · **PDF:** https://arxiv.org/pdf/2602.05120v1
+
+## 15369. Unbiased Single-Queried Gradient for Combinatorial Objective
+
+- **ID:** `2602.05119` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Thanawat Sornwanee
+- **abs:** https://arxiv.org/abs/2602.05119 · **PDF:** https://arxiv.org/pdf/2602.05119v2
+
+## 15370. SocialVeil: Probing Social Intelligence of Language Agents under Communication Barriers
+
+- **ID:** `2602.05115` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Keyang Xuan, Pengda Wang, Chongrui Ye, et al.
+- **abs:** https://arxiv.org/abs/2602.05115 · **PDF:** https://arxiv.org/pdf/2602.05115v1
+
+## 15371. Democratic Preference Alignment via Sortition-Weighted RLHF
+
+- **ID:** `2602.05113` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Suvadip Sana, Jinzhou Wu, Martin T. Wells
+- **abs:** https://arxiv.org/abs/2602.05113 · **PDF:** https://arxiv.org/pdf/2602.05113v1
+
+## 15372. Understanding LLM Evaluator Behavior: A Structured Multi-Evaluator Framework for Merchant Risk Assessment
+
+- **ID:** `2602.05110` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Liang Wang, Junpeng Wang, Chin-chia Michael Yeh, et al.
+- **abs:** https://arxiv.org/abs/2602.05110 · **PDF:** https://arxiv.org/pdf/2602.05110v1
+
+## 15373. Multilingual Extraction and Recognition of Implicit Discourse Relations in Speech and Text
+
+- **ID:** `2602.05107` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Ahmed Ruby, Christian Hardmeier, Sara Stymne
+- **abs:** https://arxiv.org/abs/2602.05107 · **PDF:** https://arxiv.org/pdf/2602.05107v1
+
+## 15374. Data Kernel Perspective Space Performance Guarantees for Synthetic Data from Transformer Models
+
+- **ID:** `2602.05106` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Michael Browder, Kevin Duh, J. David Harris, et al.
+- **abs:** https://arxiv.org/abs/2602.05106 · **PDF:** https://arxiv.org/pdf/2602.05106v1
+
+## 15375. GAMMS: Graph based Adversarial Multiagent Modeling Simulator
+
+- **ID:** `2602.05105` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Rohan Patil, Jai Malegaonkar, Xiao Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.05105 · **PDF:** https://arxiv.org/pdf/2602.05105v1
+
+## 15376. SPARE: Self-distillation for PARameter-Efficient Removal
+
+- **ID:** `2602.07058` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Natnael Mola, Leonardo S. B. Pereira, Carolina R. Kelsch, et al.
+- **abs:** https://arxiv.org/abs/2602.07058 · **PDF:** https://arxiv.org/pdf/2602.07058v2
+
+## 15377. Visual concept ranking uncovers medical shortcuts used by large multimodal models
+
+- **ID:** `2602.05096` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Joseph D. Janizek, Sonnet Xu, Junayd Lateef, et al.
+- **abs:** https://arxiv.org/abs/2602.05096 · **PDF:** https://arxiv.org/pdf/2602.05096v2
+
+## 15378. Evaluating Robustness and Adaptability in Learning-Based Mission Planning for Active Debris Removal
+
+- **ID:** `2602.05091` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Agni Bandyopadhyay, Günther Waxenegger-Wilfing
+- **abs:** https://arxiv.org/abs/2602.05091 · **PDF:** https://arxiv.org/pdf/2602.05091v1
+
+## 15379. Beware Untrusted Simulators -- Reward-Free Backdoor Attacks in Reinforcement Learning
+
+- **ID:** `2602.05089` · **primary:** `cs.CR` · **Submitted:** 2026-02-04
+- **Authors:** Ethan Rathbun, Wo Wei Lin, Alina Oprea, et al.
+- **abs:** https://arxiv.org/abs/2602.05089 · **PDF:** https://arxiv.org/pdf/2602.05089v3
+
+## 15380. VERA-MH: Reliability and Validity of an Open-Source AI Safety Evaluation in Mental Health
+
+- **ID:** `2602.05088` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Kate H. Bentley, Luca Belli, Adam M. Chekroud, et al.
+- **abs:** https://arxiv.org/abs/2602.05088 · **PDF:** https://arxiv.org/pdf/2602.05088v3
+
+## 15381. Autodiscover: A reinforcement learning recommendation system for the cold-start imbalance challenge in active learning, powered by graph-aware thompson sampling
+
+- **ID:** `2602.05087` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Parsa Vares
+- **abs:** https://arxiv.org/abs/2602.05087 · **PDF:** https://arxiv.org/pdf/2602.05087v1
+
+## 15382. Optimal Multi-Debris Mission Planning in LEO: A Deep Reinforcement Learning Approach with Co-Elliptic Transfers and Refueling
+
+- **ID:** `2602.17685` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Agni Bandyopadhyay, Gunther Waxenegger-Wilfing
+- **abs:** https://arxiv.org/abs/2602.17685 · **PDF:** https://arxiv.org/pdf/2602.17685v1
+
+## 15383. Locas: Your Models are Principled Initializers of Locally-Supported Parametric Memories
+
+- **ID:** `2602.05085` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Sidi Lu, Zhenwen Liang, Dongyang Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.05085 · **PDF:** https://arxiv.org/pdf/2602.05085v1
+
+## 15384. Individual Fairness In Strategic Classification
+
+- **ID:** `2602.05084` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Zhiqun Zuo, Mohammad Mahdi Khalili
+- **abs:** https://arxiv.org/abs/2602.05084 · **PDF:** https://arxiv.org/pdf/2602.05084v1
+
+## 15385. Large-Ensemble Simulations Reveal Links Between Atmospheric Blocking Frequency and Sea Surface Temperature Variability
+
+- **ID:** `2602.05083` · **primary:** `physics.ao-ph` · **Submitted:** 2026-02-04
+- **Authors:** Zilu Meng, Gregory J. Hakim, Wenchang Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.05083 · **PDF:** https://arxiv.org/pdf/2602.05083v1
+
+## 15386. Reliable Explanations or Random Noise? A Reliability Metric for XAI
+
+- **ID:** `2602.05082` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Poushali Sengupta, Sabita Maharjan, Frank Eliassen, et al.
+- **abs:** https://arxiv.org/abs/2602.05082 · **PDF:** https://arxiv.org/pdf/2602.05082v1
+
+## 15387. Reinforcement Learning Enhancement Using Vector Semantic Representation and Symbolic Reasoning for Human-Centered Autonomous Emergency Braking
+
+- **ID:** `2602.05079` · **primary:** `cs.RO` · **Submitted:** 2026-02-04
+- **Authors:** Vinal Asodia, Iman Sharifi, Saber Fallah
+- **abs:** https://arxiv.org/abs/2602.05079 · **PDF:** https://arxiv.org/pdf/2602.05079v1
+
+## 15388. Mechanistic Evidence for Faithfulness Decay in Chain-of-Thought Reasoning
+
+- **ID:** `2602.11201` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Donald Ye, Max Loffgren, Om Kotadia, et al.
+- **abs:** https://arxiv.org/abs/2602.11201 · **PDF:** https://arxiv.org/pdf/2602.11201v1
+
+## 15389. Food Portion Estimation: From Pixels to Calories
+
+- **ID:** `2602.05078` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Gautham Vinod, Fengqing Zhu
+- **abs:** https://arxiv.org/abs/2602.05078 · **PDF:** https://arxiv.org/pdf/2602.05078v1
+
+## 15390. Optimizing Mission Planning for Multi-Debris Rendezvous Using Reinforcement Learning with Refueling and Adaptive Collision Avoidance
+
+- **ID:** `2602.05075` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Agni Bandyopadhyay, Gunther Waxenegger-Wilfing
+- **abs:** https://arxiv.org/abs/2602.05075 · **PDF:** https://arxiv.org/pdf/2602.05075v1
+
+## 15391. Uncertainty Quantification in LLM Agents: Foundations, Emerging Challenges, and Opportunities
+
+- **ID:** `2602.05073` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Changdae Oh, Seongheon Park, To Eun Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.05073 · **PDF:** https://arxiv.org/pdf/2602.05073v2
+
+## 15392. E-Globe: Scalable $ε$-Global Verification of Neural Networks via Tight Upper Bounds and Pattern-Aware Branching
+
+- **ID:** `2602.05068` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Wenting Li, Saif R. Kazi, Russell Bent, et al.
+- **abs:** https://arxiv.org/abs/2602.05068 · **PDF:** https://arxiv.org/pdf/2602.05068v1
+
+## 15393. Bypassing AI Control Protocols via Agent-as-a-Proxy Attacks
+
+- **ID:** `2602.05066` · **primary:** `cs.CR` · **Submitted:** 2026-02-04
+- **Authors:** Jafar Isbarov, Murat Kantarcioglu
+- **abs:** https://arxiv.org/abs/2602.05066 · **PDF:** https://arxiv.org/pdf/2602.05066v2
+
+## 15394. Does SGD Seek Flatness or Sharpness? An Exactly Solvable Model
+
+- **ID:** `2602.05065` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Yizhou Xu, Pierfrancesco Beneventano, Isaac Chuang, et al.
+- **abs:** https://arxiv.org/abs/2602.05065 · **PDF:** https://arxiv.org/pdf/2602.05065v1
+
+## 15395. Scaling Laws for Embedding Dimension in Information Retrieval
+
+- **ID:** `2602.05062` · **primary:** `cs.IR` · **Submitted:** 2026-02-04
+- **Authors:** Julian Killingback, Mahta Rafiee, Madine Manas, et al.
+- **abs:** https://arxiv.org/abs/2602.05062 · **PDF:** https://arxiv.org/pdf/2602.05062v1
+
+## 15396. StagePilot: A Deep Reinforcement Learning Agent for Stage-Controlled Cybergrooming Simulation
+
+- **ID:** `2602.05060` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Heajun An, Qi Zhang, Minqian Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.05060 · **PDF:** https://arxiv.org/pdf/2602.05060v1
+
+## 15397. Evaluating Large Language Models on Solved and Unsolved Problems in Graph Theory: Implications for Computing Education
+
+- **ID:** `2602.05059` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Adithya Kulkarni, Mohna Chakraborty, Jay Bagga
+- **abs:** https://arxiv.org/abs/2602.05059 · **PDF:** https://arxiv.org/pdf/2602.05059v1
+
+## 15398. Learning fermionic linear optics with Heisenberg scaling and physical operations
+
+- **ID:** `2602.05058` · **primary:** `quant-ph` · **Submitted:** 2026-02-04
+- **Authors:** Aria Christensen, Andrew Zhao
+- **abs:** https://arxiv.org/abs/2602.05058 · **PDF:** https://arxiv.org/pdf/2602.05058v2
+
+## 15399. VEXA: Evidence-Grounded and Persona-Adaptive Explanations for Scam Risk Sensemaking
+
+- **ID:** `2602.05056` · **primary:** `cs.CR` · **Submitted:** 2026-02-04
+- **Authors:** Heajun An, Connor Ng, Sandesh Sharma Dulal, et al.
+- **abs:** https://arxiv.org/abs/2602.05056 · **PDF:** https://arxiv.org/pdf/2602.05056v1
+
+## 15400. Quantile-Physics Hybrid Framework for Safe-Speed Recommendation under Diverse Weather Conditions Leveraging Connected Vehicle and Road Weather Information Systems Data
+
+- **ID:** `2602.05053` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Wen Zhang, Adel W. Sadek, Chunming Qiao
+- **abs:** https://arxiv.org/abs/2602.05053 · **PDF:** https://arxiv.org/pdf/2602.05053v2
+
+## 15401. Learning, Solving and Optimizing PDEs with TensorGalerkin: an efficient high-performance Galerkin assembly algorithm
+
+- **ID:** `2602.05052` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Shizheng Wen, Mingyuan Chi, Tianwei Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.05052 · **PDF:** https://arxiv.org/pdf/2602.05052v2
+
+## 15402. ReFORM: Reflected Flows for On-support Offline RL via Noise Manipulation
+
+- **ID:** `2602.05051` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Songyuan Zhang, Oswin So, H. M. Sabbir Ahmad, et al.
+- **abs:** https://arxiv.org/abs/2602.05051 · **PDF:** https://arxiv.org/pdf/2602.05051v1
+
+## 15403. VISTA: Enhancing Visual Conditioning via Track-Following Preference Optimization in Vision-Language-Action Models
+
+- **ID:** `2602.05049` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Yiye Chen, Yanan Jian, Xiaoyi Dong, et al.
+- **abs:** https://arxiv.org/abs/2602.05049 · **PDF:** https://arxiv.org/pdf/2602.05049v1
+
+## 15404. MINT: Minimal Information Neuro-Symbolic Tree for Objective-Driven Knowledge-Gap Reasoning and Active Elicitation
+
+- **ID:** `2602.05048` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Zeyu Fang, Tian Lan, Mahdi Imani
+- **abs:** https://arxiv.org/abs/2602.05048 · **PDF:** https://arxiv.org/pdf/2602.05048v1
+
+## 15405. Quality Model for Machine Learning Components
+
+- **ID:** `2602.05043` · **primary:** `cs.SE` · **Submitted:** 2026-02-04
+- **Authors:** Grace A. Lewis, Rachel Brower-Sinning, Robert Edman, et al.
+- **abs:** https://arxiv.org/abs/2602.05043 · **PDF:** https://arxiv.org/pdf/2602.05043v1
+
+## 15406. Feedback Control for Multi-Objective Graph Self-Supervision
+
+- **ID:** `2602.05036` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Karish Grover, Theodore Vasiloudis, Han Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.05036 · **PDF:** https://arxiv.org/pdf/2602.05036v1
+
+## 15407. Capacity Constraints and the Multilingual Penalty for Lexical Disambiguation
+
+- **ID:** `2602.05035` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Sean Trott, Pamela D. Rivière
+- **abs:** https://arxiv.org/abs/2602.05035 · **PDF:** https://arxiv.org/pdf/2602.05035v1
+
+## 15408. Causal Representation Meets Stochastic Modeling under Generic Geometry
+
+- **ID:** `2602.05033` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Jiaxu Ren, Yixin Wang, Biwei Huang
+- **abs:** https://arxiv.org/abs/2602.05033 · **PDF:** https://arxiv.org/pdf/2602.05033v1
+
+## 15409. MTS-CSNet: Multiscale Tensor Factorization for Deep Compressive Sensing on RGB Images
+
+- **ID:** `2602.07056` · **primary:** `eess.IV` · **Submitted:** 2026-02-04
+- **Authors:** Mehmet Yamac, Lei Xu, Serkan Kiranyaz, et al.
+- **abs:** https://arxiv.org/abs/2602.07056 · **PDF:** https://arxiv.org/pdf/2602.07056v1
+
+## 15410. Laplacian Representations for Decision-Time Planning
+
+- **ID:** `2602.05031` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Dikshant Shehmar, Matthew Schlegel, Matthew E. Taylor, et al.
+- **abs:** https://arxiv.org/abs/2602.05031 · **PDF:** https://arxiv.org/pdf/2602.05031v1
+
+## 15411. Differentiable Inverse Graphics for Zero-shot Scene Reconstruction and Robot Grasping
+
+- **ID:** `2602.05029` · **primary:** `cs.RO` · **Submitted:** 2026-02-04
+- **Authors:** Octavio Arriaga, Proneet Sharma, Jichen Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.05029 · **PDF:** https://arxiv.org/pdf/2602.05029v1
+
+## 15412. Human-Centered Explainable AI for Security Enhancement: A Deep Intrusion Detection Framework
+
+- **ID:** `2602.13271` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Md Muntasir Jahid Ayan, Md. Shahriar Rashid, Tazzina Afroze Hassan, et al.
+- **abs:** https://arxiv.org/abs/2602.13271 · **PDF:** https://arxiv.org/pdf/2602.13271v1
+
+## 15413. AudioSAE: Towards Understanding of Audio-Processing Models with Sparse AutoEncoders
+
+- **ID:** `2602.05027` · **primary:** `cs.SD` · **Submitted:** 2026-02-04
+- **Authors:** Georgii Aparin, Tasnima Sadekova, Alexey Rukhovich, et al.
+- **abs:** https://arxiv.org/abs/2602.05027 · **PDF:** https://arxiv.org/pdf/2602.05027v2
+
+## 15414. Laws of Learning Dynamics and the Core of Learners
+
+- **ID:** `2602.05026` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Inkee Jung, Siu Cheong Lau
+- **abs:** https://arxiv.org/abs/2602.05026 · **PDF:** https://arxiv.org/pdf/2602.05026v1
+
+## 15415. Do Vision-Language Models Respect Contextual Integrity in Location Disclosure?
+
+- **ID:** `2602.05023` · **primary:** `cs.CR` · **Submitted:** 2026-02-04
+- **Authors:** Ruixin Yang, Ethan Mendes, Arthur Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.05023 · **PDF:** https://arxiv.org/pdf/2602.05023v2
+
+## 15416. A Simple Reduction Scheme for Constrained Contextual Bandits with Adversarial Contexts via Regression
+
+- **ID:** `2602.05019` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Dhruv Sarkar, Abhishek Sinha
+- **abs:** https://arxiv.org/abs/2602.05019 · **PDF:** https://arxiv.org/pdf/2602.05019v1
+
+## 15417. From Fragmentation to Integration: Exploring the Design Space of AI Agents for Human-as-the-Unit Privacy Management
+
+- **ID:** `2602.05016` · **primary:** `cs.HC` · **Submitted:** 2026-02-04
+- **Authors:** Eryue Xu, Tianshi Li
+- **abs:** https://arxiv.org/abs/2602.05016 · **PDF:** https://arxiv.org/pdf/2602.05016v2
+
+## 15418. DeepRead: Document Structure-Aware Reasoning to Enhance Agentic Search
+
+- **ID:** `2602.05014` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Zhanli Li, Huiwen Tian, Lvzhou Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.05014 · **PDF:** https://arxiv.org/pdf/2602.05014v3
+
+## 15419. Private PoEtry: Private In-Context Learning via Product of Experts
+
+- **ID:** `2602.05012` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Rob Romijnders, Mohammad Mahdi Derakhshani, Jonathan Petit, et al.
+- **abs:** https://arxiv.org/abs/2602.05012 · **PDF:** https://arxiv.org/pdf/2602.05012v1
+
+## 15420. Enhanced QKNorm normalization for neural transformers with the Lp norm
+
+- **ID:** `2602.05006` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Ezequiel Lopez-Rubio, Javier Montes-Perez, Esteban Jose Palomo
+- **abs:** https://arxiv.org/abs/2602.05006 · **PDF:** https://arxiv.org/pdf/2602.05006v1
+
+## 15421. CoWork-X: Experience-Optimized Co-Evolution for Multi-Agent Collaboration System
+
+- **ID:** `2602.05004` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Zexin Lin, Jiachen Yu, Haoyang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.05004 · **PDF:** https://arxiv.org/pdf/2602.05004v1
+
+## 15422. EntRGi: Entropy Aware Reward Guidance for Diffusion Language Models
+
+- **ID:** `2602.05000` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Atula Tejaswi, Litu Rout, Constantine Caramanis, et al.
+- **abs:** https://arxiv.org/abs/2602.05000 · **PDF:** https://arxiv.org/pdf/2602.05000v1
+
+## 15423. Learning Rate Matters: Vanilla LoRA May Suffice for LLM Fine-tuning
+
+- **ID:** `2602.04998` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Yu-Ang Lee, Ching-Yun Ko, Pin-Yu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.04998 · **PDF:** https://arxiv.org/pdf/2602.04998v1
+
+## 15424. SIDeR: Semantic Identity Decoupling for Unrestricted Face Privacy
+
+- **ID:** `2602.04994` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Zhuosen Bao, Xia Du, Zheng Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.04994 · **PDF:** https://arxiv.org/pdf/2602.04994v1
+
+## 15425. Position: Machine Learning for Heart Transplant Allocation Policy Optimization Should Account for Incentives
+
+- **ID:** `2602.04990` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Ioannis Anagnostides, Itai Zilberstein, Zachary W. Sollie, et al.
+- **abs:** https://arxiv.org/abs/2602.04990 · **PDF:** https://arxiv.org/pdf/2602.04990v2
+
+## 15426. Near-Optimal Dynamic Matching via Coarsening with Application to Heart Transplantation
+
+- **ID:** `2602.04989` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Itai Zilberstein, Ioannis Anagnostides, Zachary W. Sollie, et al.
+- **abs:** https://arxiv.org/abs/2602.04989 · **PDF:** https://arxiv.org/pdf/2602.04989v2
+
+## 15427. Artificial Intelligence as Strange Intelligence: Against Linear Models of Intelligence
+
+- **ID:** `2602.04986` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Kendra Chilson, Eric Schwitzgebel
+- **abs:** https://arxiv.org/abs/2602.04986 · **PDF:** https://arxiv.org/pdf/2602.04986v1
+
+## 15428. AI-Based Detection of In-Treatment Changes from Prostate MR-Linac Images
+
+- **ID:** `2602.04983` · **primary:** `eess.IV` · **Submitted:** 2026-02-04
+- **Authors:** Seungbin Park, Peilin Wang, Ryan Pennell, et al.
+- **abs:** https://arxiv.org/abs/2602.04983 · **PDF:** https://arxiv.org/pdf/2602.04983v1
+
+## 15429. BioACE: An Automated Framework for Biomedical Answer and Citation Evaluations
+
+- **ID:** `2602.04982` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Deepak Gupta, Davis Bartels, Dina Demner-Fushman
+- **abs:** https://arxiv.org/abs/2602.04982 · **PDF:** https://arxiv.org/pdf/2602.04982v2
+
+## 15430. Theory of Space: Can Foundation Models Construct Spatial Beliefs through Active Exploration?
+
+- **ID:** `2602.07055` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Pingyue Zhang, Zihan Huang, Yue Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.07055 · **PDF:** https://arxiv.org/pdf/2602.07055v1
+
+## 15431. Stochastic hierarchical data-driven optimization: application to plasma-surface kinetics
+
+- **ID:** `2602.04975` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** José Afonso, Vasco Guerra, Pedro Viegas
+- **abs:** https://arxiv.org/abs/2602.04975 · **PDF:** https://arxiv.org/pdf/2602.04975v1
+
+## 15432. Instance-optimal high-precision shadow tomography with few-copy measurements: A metrological approach
+
+- **ID:** `2602.04952` · **primary:** `quant-ph` · **Submitted:** 2026-02-04
+- **Authors:** Senrui Chen, Weiyuan Gong, Sisi Zhou
+- **abs:** https://arxiv.org/abs/2602.04952 · **PDF:** https://arxiv.org/pdf/2602.04952v1
+
+## 15433. Reinforced Attention Learning
+
+- **ID:** `2602.04884` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Bangzheng Li, Jianmo Ni, Chen Qu, et al.
+- **abs:** https://arxiv.org/abs/2602.04884 · **PDF:** https://arxiv.org/pdf/2602.04884v2
+
+## 15434. Protein Autoregressive Modeling via Multiscale Structure Generation
+
+- **ID:** `2602.04883` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Yanru Qu, Cheng-Yen Hsieh, Zaixiang Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.04883 · **PDF:** https://arxiv.org/pdf/2602.04883v1
+
+## 15435. Contrastive Continual Learning for Model Adaptability in Internet of Things
+
+- **ID:** `2602.04881` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Ajesh Koyatan Chathoth
+- **abs:** https://arxiv.org/abs/2602.04881 · **PDF:** https://arxiv.org/pdf/2602.04881v1
+
+## 15436. Rethinking the Trust Region in LLM Reinforcement Learning
+
+- **ID:** `2602.04879` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Penghui Qi, Xiangxin Zhou, Zichen Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.04879 · **PDF:** https://arxiv.org/pdf/2602.04879v1
+
+## 15437. Smart Diagnosis and Early Intervention in PCOS: A Deep Learning Approach to Women's Reproductive Health
+
+- **ID:** `2602.04944` · **primary:** `eess.IV` · **Submitted:** 2026-02-04
+- **Authors:** Shayan Abrar, Samura Rahman, Ishrat Jahan Momo, et al.
+- **abs:** https://arxiv.org/abs/2602.04944 · **PDF:** https://arxiv.org/pdf/2602.04944v1
+
+## 15438. Multi-layer Cross-Attention is Provably Optimal for Multi-modal In-context Learning
+
+- **ID:** `2602.04872` · **primary:** `stat.ML` · **Submitted:** 2026-02-04
+- **Authors:** Nicholas Barnfield, Subhabrata Sen, Pragya Sur
+- **abs:** https://arxiv.org/abs/2602.04872 · **PDF:** https://arxiv.org/pdf/2602.04872v1
+
+## 15439. Multi-Head LatentMoE and Head Parallel: Communication-Efficient and Deterministic MoE Parallelism
+
+- **ID:** `2602.04870` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Chenwei Cui, Rockwell Jackson, Benjamin Joseph Herrera, et al.
+- **abs:** https://arxiv.org/abs/2602.04870 · **PDF:** https://arxiv.org/pdf/2602.04870v1
+
+## 15440. CRoSS: A Continual Robotic Simulation Suite for Scalable Reinforcement Learning with High Task Diversity and Realistic Physics Simulation
+
+- **ID:** `2602.04868` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Yannick Denker, Alexander Gepperth
+- **abs:** https://arxiv.org/abs/2602.04868 · **PDF:** https://arxiv.org/pdf/2602.04868v1
+
+## 15441. Graph-Theoretic Analysis of Phase Optimization Complexity in Variational Wave Functions for Heisenberg Antiferromagnets
+
+- **ID:** `2602.04943` · **primary:** `cond-mat.str-el` · **Submitted:** 2026-02-04
+- **Authors:** Mahmud Ashraf Shamim, Md Moshiur Rahman Raj, Mohamed Hibat-Allah, et al.
+- **abs:** https://arxiv.org/abs/2602.04943 · **PDF:** https://arxiv.org/pdf/2602.04943v2
+
+## 15442. Subliminal Effects in Your Data: A General Mechanism via Log-Linearity
+
+- **ID:** `2602.04863` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Ishaq Aden-Ali, Noah Golowich, Allen Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.04863 · **PDF:** https://arxiv.org/pdf/2602.04863v1
+
+## 15443. From Evaluation to Design: Using Potential Energy Surface Smoothness Metrics to Guide Machine Learning Interatomic Potential Architectures
+
+- **ID:** `2602.04861` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Ryan Liu, Eric Qu, Tobias Kreiman, et al.
+- **abs:** https://arxiv.org/abs/2602.04861 · **PDF:** https://arxiv.org/pdf/2602.04861v1
+
+## 15444. Privileged Information Distillation for Language Models
+
+- **ID:** `2602.04942` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Emiliano Penaloza, Dheeraj Vattikonda, Nicolas Gontier, et al.
+- **abs:** https://arxiv.org/abs/2602.04942 · **PDF:** https://arxiv.org/pdf/2602.04942v3
+
+## 15445. CoT is Not the Chain of Truth: An Empirical Internal Analysis of Reasoning LLMs for Fake News Generation
+
+- **ID:** `2602.04856` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Zhao Tong, Chunlin Gong, Yiping Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.04856 · **PDF:** https://arxiv.org/pdf/2602.04856v3
+
+## 15446. Decomposed Prompting Does Not Fix Knowledge Gaps, But Helps Models Say "I Don't Know"
+
+- **ID:** `2602.04853` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Dhruv Madhwal, Lyuxin David Zhang, Dan Roth, et al.
+- **abs:** https://arxiv.org/abs/2602.04853 · **PDF:** https://arxiv.org/pdf/2602.04853v1
+
+## 15447. The Key to State Reduction in Linear Attention: A Rank-based Perspective
+
+- **ID:** `2602.04852` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Philipp Nazari, T. Konstantin Rusch
+- **abs:** https://arxiv.org/abs/2602.04852 · **PDF:** https://arxiv.org/pdf/2602.04852v2
+
+## 15448. El Agente Quntur: A research collaborator agent for quantum chemistry
+
+- **ID:** `2602.04850` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-04
+- **Authors:** Juan B. Pérez-Sánchez, Yunheng Zou, Jorge A. Campos-Gonzalez-Angulo, et al.
+- **abs:** https://arxiv.org/abs/2602.04850 · **PDF:** https://arxiv.org/pdf/2602.04850v1
+
+## 15449. El Agente Estructural: An Artificially Intelligent Molecular Editor
+
+- **ID:** `2602.04849` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-04
+- **Authors:** Changhyeok Choi, Yunheng Zou, Marcel Müller, et al.
+- **abs:** https://arxiv.org/abs/2602.04849 · **PDF:** https://arxiv.org/pdf/2602.04849v1
+
+## 15450. Improving Set Function Approximation with Quasi-Arithmetic Neural Networks
+
+- **ID:** `2602.04941` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Tomas Tokar, Scott Sanner
+- **abs:** https://arxiv.org/abs/2602.04941 · **PDF:** https://arxiv.org/pdf/2602.04941v1
+
+## 15451. Fluid Representations in Reasoning Models
+
+- **ID:** `2602.04843` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Dmitrii Kharlapenko, Alessandro Stolfo, Arthur Conmy, et al.
+- **abs:** https://arxiv.org/abs/2602.04843 · **PDF:** https://arxiv.org/pdf/2602.04843v1
+
+## 15452. Group-Evolving Agents: Open-Ended Self-Improvement via Experience Sharing
+
+- **ID:** `2602.04837` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Zhaotian Weng, Antonis Antoniades, Deepak Nathani, et al.
+- **abs:** https://arxiv.org/abs/2602.04837 · **PDF:** https://arxiv.org/pdf/2602.04837v1
+
+## 15453. Are AI Capabilities Increasing Exponentially? A Competing Hypothesis
+
+- **ID:** `2602.04836` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Haosen Ge, Hamsa Bastani, Osbert Bastani
+- **abs:** https://arxiv.org/abs/2602.04836 · **PDF:** https://arxiv.org/pdf/2602.04836v2
+
+## 15454. AVERE: Improving Audiovisual Emotion Reasoning with Preference Optimization
+
+- **ID:** `2602.07054` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Ashutosh Chaubey, Jiacheng Pang, Maksim Siniukov, et al.
+- **abs:** https://arxiv.org/abs/2602.07054 · **PDF:** https://arxiv.org/pdf/2602.07054v1
+
+## 15455. It's not a Lottery, it's a Race: Understanding How Gradient Descent Adapts the Network's Capacity to the Task
+
+- **ID:** `2602.04832` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Hannah Pinson
+- **abs:** https://arxiv.org/abs/2602.04832 · **PDF:** https://arxiv.org/pdf/2602.04832v1
+
+## 15456. Safe Urban Traffic Control via Uncertainty-Aware Conformal Prediction and World-Model Reinforcement Learning
+
+- **ID:** `2602.04821` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Joydeep Chandra, Satyam Kumar Navneet, Aleksandr Algazinov, et al.
+- **abs:** https://arxiv.org/abs/2602.04821 · **PDF:** https://arxiv.org/pdf/2602.04821v1
+
+## 15457. Toward Reliable and Explainable Nail Disease Classification: Leveraging Adversarial Training and Grad-CAM Visualization
+
+- **ID:** `2602.04820` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Farzia Hossain, Samanta Ghosh, Shahida Begum, et al.
+- **abs:** https://arxiv.org/abs/2602.04820 · **PDF:** https://arxiv.org/pdf/2602.04820v1
+
+## 15458. XtraLight-MedMamba for Classification of Neoplastic Tubular Adenomas
+
+- **ID:** `2602.04819` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Aqsa Sultana, Rayan Afsar, Ahmed Rahu, et al.
+- **abs:** https://arxiv.org/abs/2602.04819 · **PDF:** https://arxiv.org/pdf/2602.04819v4
+
+## 15459. Horizon-LM: A RAM-Centric Architecture for LLM Training
+
+- **ID:** `2602.04816` · **primary:** `cs.OS` · **Submitted:** 2026-02-04
+- **Authors:** Zhengqing Yuan, Lichao Sun, Yanfang Ye
+- **abs:** https://arxiv.org/abs/2602.04816 · **PDF:** https://arxiv.org/pdf/2602.04816v2
+
+## 15460. Agentic AI in Healthcare & Medicine: A Seven-Dimensional Taxonomy for Empirical Evaluation of LLM-based Agents
+
+- **ID:** `2602.04813` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Shubham Vatsal, Harsh Dubey, Aditi Singh
+- **abs:** https://arxiv.org/abs/2602.04813 · **PDF:** https://arxiv.org/pdf/2602.04813v1
+
+## 15461. Robust Generalizable Heterogeneous Legal Link Prediction
+
+- **ID:** `2602.04812` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Lorenz Wendlinger, Simon Alexander Nonn, Abdullah Al Zubaer, et al.
+- **abs:** https://arxiv.org/abs/2602.04812 · **PDF:** https://arxiv.org/pdf/2602.04812v1
+
+## 15462. SE-Bench: Benchmarking Self-Evolution with Knowledge Internalization
+
+- **ID:** `2602.04811` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Jiarui Yuan, Tailin Jin, Weize Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.04811 · **PDF:** https://arxiv.org/pdf/2602.04811v1
+
+## 15463. CodeScaler: Scaling Code LLM Training and Test-Time Inference via Execution-Free Reward Models
+
+- **ID:** `2602.17684` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Xiao Zhu, Xinyu Zhou, Boyu Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.17684 · **PDF:** https://arxiv.org/pdf/2602.17684v1
+
+## 15464. Beyond Rewards in Reinforcement Learning for Cyber Defence
+
+- **ID:** `2602.04809` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Elizabeth Bates, Chris Hicks, Vasilios Mavroudis
+- **abs:** https://arxiv.org/abs/2602.04809 · **PDF:** https://arxiv.org/pdf/2602.04809v2
+
+## 15465. Evolving Afferent Architectures: Biologically-inspired Models for Damage-Avoidance Learning
+
+- **ID:** `2602.04807` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Wolfgang Maass, Sabine Janzen, Prajvi Saxena, et al.
+- **abs:** https://arxiv.org/abs/2602.04807 · **PDF:** https://arxiv.org/pdf/2602.04807v1
+
+## 15466. Skin Tokens: A Learned Compact Representation for Unified Autoregressive Rigging
+
+- **ID:** `2602.04805` · **primary:** `cs.GR` · **Submitted:** 2026-02-04
+- **Authors:** Jia-peng Zhang, Cheng-Feng Pu, Meng-Hao Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.04805 · **PDF:** https://arxiv.org/pdf/2602.04805v1
+
+## 15467. OmniSIFT: Modality-Asymmetric Token Compression for Efficient Omni-modal Large Language Models
+
+- **ID:** `2602.04804` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Yue Ding, Yiyan Ji, Jungang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.04804 · **PDF:** https://arxiv.org/pdf/2602.04804v1
+
+## 15468. Probabilistic NDVI Forecasting from Sparse Satellite Time Series and Weather Covariates
+
+- **ID:** `2602.17683` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Irene Iele, Giulia Romoli, Daniele Molino, et al.
+- **abs:** https://arxiv.org/abs/2602.17683 · **PDF:** https://arxiv.org/pdf/2602.17683v1
+
+## 15469. Score-Based Change-Point Detection and Region Localization for Spatio-Temporal Point Processes
+
+- **ID:** `2602.04798` · **primary:** `stat.ME` · **Submitted:** 2026-02-04
+- **Authors:** Wenbin Zhou, Liyan Xie, Shixiang Zhu
+- **abs:** https://arxiv.org/abs/2602.04798 · **PDF:** https://arxiv.org/pdf/2602.04798v1
+
+## 15470. Maximum-Volume Nonnegative Matrix Factorization
+
+- **ID:** `2602.04795` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Olivier Vu Thanh, Nicolas Gillis
+- **abs:** https://arxiv.org/abs/2602.04795 · **PDF:** https://arxiv.org/pdf/2602.04795v2
+
+## 15471. Team, Then Trim: An Assembly-Line LLM Framework for High-Quality Tabular Data Generation
+
+- **ID:** `2602.04785` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Congjing Zhang, Ryan Feng Lin, Ruoxuan Bao, et al.
+- **abs:** https://arxiv.org/abs/2602.04785 · **PDF:** https://arxiv.org/pdf/2602.04785v1
+
+## 15472. From independent patches to coordinated attention: Controlling information flow in vision transformers
+
+- **ID:** `2602.04784` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Kieran A. Murphy
+- **abs:** https://arxiv.org/abs/2602.04784 · **PDF:** https://arxiv.org/pdf/2602.04784v1
+
+## 15473. Legendre Memory Unit with A Multi-Slice Compensation Model for Short-Term Wind Speed Forecasting Based on Wind Farm Cluster Data
+
+- **ID:** `2602.04782` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Mumin Zhang, Haochen Zhang, Xin Zhi Khoo, et al.
+- **abs:** https://arxiv.org/abs/2602.04782 · **PDF:** https://arxiv.org/pdf/2602.04782v1
+
+## 15474. A Quantum-inspired Hybrid Swarm Intelligence and Decision-Making for Multi-Criteria ADAS Calibration
+
+- **ID:** `2602.15043` · **primary:** `cs.NE` · **Submitted:** 2026-02-04
+- **Authors:** Sanjai Pathak, Ashish Mani, Amlan Chatterjee
+- **abs:** https://arxiv.org/abs/2602.15043 · **PDF:** https://arxiv.org/pdf/2602.15043v1
+
+## 15475. Reliable and Responsible Foundation Models: A Comprehensive Survey
+
+- **ID:** `2602.08145` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Xinyu Yang, Junlin Han, Rishi Bommasani, et al.
+- **abs:** https://arxiv.org/abs/2602.08145 · **PDF:** https://arxiv.org/pdf/2602.08145v1
+
+## 15476. Duality Models: An Embarrassingly Simple One-step Generation Paradigm
+
+- **ID:** `2602.17682` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Peng Sun, Xinyi Shang, Tao Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.17682 · **PDF:** https://arxiv.org/pdf/2602.17682v1
+
+## 15477. Dynamical Regimes of Multimodal Diffusion Models
+
+- **ID:** `2602.04780` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Emil Albrychiewicz, Andrés Franco Valiente, Li-Ching Chen
+- **abs:** https://arxiv.org/abs/2602.04780 · **PDF:** https://arxiv.org/pdf/2602.04780v1
+
+## 15478. Communication Enhances LLMs' Stability in Strategic Thinking
+
+- **ID:** `2602.06081` · **primary:** `cs.MA` · **Submitted:** 2026-02-04
+- **Authors:** Nunzio Lore, Babak Heydari
+- **abs:** https://arxiv.org/abs/2602.06081 · **PDF:** https://arxiv.org/pdf/2602.06081v1
+
+## 15479. Speaker-Aware Simulation Improves Conversational Speech Recognition
+
+- **ID:** `2602.04776` · **primary:** `cs.SD` · **Submitted:** 2026-02-04
+- **Authors:** Máté Gedeon, Péter Mihajlik
+- **abs:** https://arxiv.org/abs/2602.04776 · **PDF:** https://arxiv.org/pdf/2602.04776v1
+
+## 15480. Interval-Based AUC (iAUC): Extending ROC Analysis to Uncertainty-Aware Classification
+
+- **ID:** `2602.04775` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Yuqi Li, Matthew M. Engelhard
+- **abs:** https://arxiv.org/abs/2602.04775 · **PDF:** https://arxiv.org/pdf/2602.04775v1
+
+## 15481. Theory of Optimal Learning Rate Schedules and Scaling Laws for a Random Feature Model
+
+- **ID:** `2602.04774` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-02-04
+- **Authors:** Blake Bordelon, Francesco Mori
+- **abs:** https://arxiv.org/abs/2602.04774 · **PDF:** https://arxiv.org/pdf/2602.04774v1
+
+## 15482. Generative Modeling via Drifting
+
+- **ID:** `2602.04770` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Mingyang Deng, He Li, Tianhong Li, et al.
+- **abs:** https://arxiv.org/abs/2602.04770 · **PDF:** https://arxiv.org/pdf/2602.04770v2
+
+## 15483. NeuroCanvas: VLLM-Powered Robust Seizure Detection by Reformulating Multichannel EEG as Image
+
+- **ID:** `2602.04769` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Yan Chen, Jie Peng, Moajjem Hossain Chowdhury, et al.
+- **abs:** https://arxiv.org/abs/2602.04769 · **PDF:** https://arxiv.org/pdf/2602.04769v2
+
+## 15484. Billion-Scale Graph Foundation Models
+
+- **ID:** `2602.04768` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Maya Bechler-Speicher, Yoel Gottlieb, Andrey Isakov, et al.
+- **abs:** https://arxiv.org/abs/2602.04768 · **PDF:** https://arxiv.org/pdf/2602.04768v1
+
+## 15485. Beyond Many-Shot Translation: Scaling In-Context Demonstrations For Low-Resource Machine Translation
+
+- **ID:** `2602.04764` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Luis Frentzen Salim, Esteban Carlin, Alexandre Morinvil, et al.
+- **abs:** https://arxiv.org/abs/2602.04764 · **PDF:** https://arxiv.org/pdf/2602.04764v1
+
+## 15486. Active Asymmetric Multi-Agent Multimodal Learning under Uncertainty
+
+- **ID:** `2602.04763` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Rui Liu, Pratap Tokekar, Ming Lin
+- **abs:** https://arxiv.org/abs/2602.04763 · **PDF:** https://arxiv.org/pdf/2602.04763v1
+
+## 15487. Improved Dimension Dependence for Bandit Convex Optimization with Gradient Variations
+
+- **ID:** `2602.04761` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Hang Yu, Yu-Hu Yan, Peng Zhao
+- **abs:** https://arxiv.org/abs/2602.04761 · **PDF:** https://arxiv.org/pdf/2602.04761v1
+
+## 15488. A Dual-TransUNet Deep Learning Framework for Multi-Source Precipitation Merging and Improving Seasonal and Extreme Estimates
+
+- **ID:** `2602.04757` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Yuchen Ye, Zixuan Qi, Shixuan Li, et al.
+- **abs:** https://arxiv.org/abs/2602.04757 · **PDF:** https://arxiv.org/pdf/2602.04757v1
+
+## 15489. When Silence Is Golden: Can LLMs Learn to Abstain in Temporal QA and Beyond?
+
+- **ID:** `2602.04755` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Xinyu Zhou, Chang Jin, Carsten Eickhoff, et al.
+- **abs:** https://arxiv.org/abs/2602.04755 · **PDF:** https://arxiv.org/pdf/2602.04755v2
+
+## 15490. Transolver-3: Scaling Up Transformer Solvers to Industrial-Scale Geometries
+
+- **ID:** `2602.04940` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Hang Zhou, Haixu Wu, Haonan Shangguan, et al.
+- **abs:** https://arxiv.org/abs/2602.04940 · **PDF:** https://arxiv.org/pdf/2602.04940v1
+
+## 15491. Comparative Insights on Adversarial Machine Learning from Industry and Academia: A User-Study Approach
+
+- **ID:** `2602.04753` · **primary:** `cs.CR` · **Submitted:** 2026-02-04
+- **Authors:** Vishruti Kakkad, Paul Chung, Hanan Hibshi, et al.
+- **abs:** https://arxiv.org/abs/2602.04753 · **PDF:** https://arxiv.org/pdf/2602.04753v1
+
+## 15492. Decomposing Query-Key Feature Interactions Using Contrastive Covariances
+
+- **ID:** `2602.04752` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Andrew Lee, Yonatan Belinkov, Fernanda Viégas, et al.
+- **abs:** https://arxiv.org/abs/2602.04752 · **PDF:** https://arxiv.org/pdf/2602.04752v1
+
+## 15493. Exploiting contextual information to improve stance detection in informal political discourse with LLMs
+
+- **ID:** `2602.04750` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Arman Engin Sucu, Yixiang Zhou, Mario A. Nascimento, et al.
+- **abs:** https://arxiv.org/abs/2602.04750 · **PDF:** https://arxiv.org/pdf/2602.04750v1
+
+## 15494. Impact of diversity on bounded archives for multi-objective local search
+
+- **ID:** `2602.04745` · **primary:** `cs.NE` · **Submitted:** 2026-02-04
+- **Authors:** Amadeu A. Coco, Cyprien Borée, Julien Baste, et al.
+- **abs:** https://arxiv.org/abs/2602.04745 · **PDF:** https://arxiv.org/pdf/2602.04745v1
+
+## 15495. Inference-Time Reasoning Selectively Reduces Implicit Social Bias in Large Language Models
+
+- **ID:** `2602.04742` · **primary:** `cs.CY` · **Submitted:** 2026-02-04
+- **Authors:** Molly Apsel, Michael N. Jones
+- **abs:** https://arxiv.org/abs/2602.04742 · **PDF:** https://arxiv.org/pdf/2602.04742v1
+
+## 15496. Alignment Drift in Multimodal LLMs: A Two-Phase, Longitudinal Evaluation of Harm Across Eight Model Releases
+
+- **ID:** `2602.04739` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Casey Ford, Madison Van Doren, Emily Dix
+- **abs:** https://arxiv.org/abs/2602.04739 · **PDF:** https://arxiv.org/pdf/2602.04739v1
+
+## 15497. Rationality Measurement and Theory for Reinforcement Learning Agents
+
+- **ID:** `2602.04737` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Kejiang Qian, Amos Storkey, Fengxiang He
+- **abs:** https://arxiv.org/abs/2602.04737 · **PDF:** https://arxiv.org/pdf/2602.04737v1
+
+## 15498. Conditional Counterfactual Mean Embeddings: Doubly Robust Estimation and Learning Rates
+
+- **ID:** `2602.04736` · **primary:** `stat.ML` · **Submitted:** 2026-02-04
+- **Authors:** Thatchanon Anancharoenkij, Donlapark Ponnoprat
+- **abs:** https://arxiv.org/abs/2602.04736 · **PDF:** https://arxiv.org/pdf/2602.04736v1
+
+## 15499. From Data to Behavior: Predicting Unintended Model Behaviors Before Training
+
+- **ID:** `2602.04735` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Mengru Wang, Zhenqian Xu, Junfeng Fang, et al.
+- **abs:** https://arxiv.org/abs/2602.04735 · **PDF:** https://arxiv.org/pdf/2602.04735v1
+
+## 15500. DMFlow: Disordered Materials Generation by Flow Matching
+
+- **ID:** `2602.04734` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Liming Wu, Rui Jiao, Qi Li, et al.
+- **abs:** https://arxiv.org/abs/2602.04734 · **PDF:** https://arxiv.org/pdf/2602.04734v1
+
+## 15501. Less Finetuning, Better Retrieval: Rethinking LLM Adaptation for Biomedical Retrievers via Synthetic Data and Model Merging
+
+- **ID:** `2602.04731` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Sameh Khattab, Jean-Philippe Corbeil, Osman Alperen Koraş, et al.
+- **abs:** https://arxiv.org/abs/2602.04731 · **PDF:** https://arxiv.org/pdf/2602.04731v1
+
+## 15502. "Be My Cheese?": Cultural Nuance Benchmarking for Machine Translation in Multilingual LLMs
+
+- **ID:** `2602.04729` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Madison Van Doren, Casey Ford, Jennifer Barajas, et al.
+- **abs:** https://arxiv.org/abs/2602.04729 · **PDF:** https://arxiv.org/pdf/2602.04729v1
+
+## 15503. Exploring the Ethical Concerns in User Reviews of Mental Health Apps using Topic Modeling and Sentiment Analysis
+
+- **ID:** `2602.18454` · **primary:** `cs.CY` · **Submitted:** 2026-02-04
+- **Authors:** Mohammad Masudur Rahman, Beenish Moalla Chaudhry
+- **abs:** https://arxiv.org/abs/2602.18454 · **PDF:** https://arxiv.org/pdf/2602.18454v1
+
+## 15504. Cross-Attention Transformer for Joint Multi-Receiver Uplink Neural Decoding
+
+- **ID:** `2602.04728` · **primary:** `eess.SP` · **Submitted:** 2026-02-04
+- **Authors:** Xavier Tardy, Grégoire Lefebvre, Apostolos Kountouris, et al.
+- **abs:** https://arxiv.org/abs/2602.04728 · **PDF:** https://arxiv.org/pdf/2602.04728v1
+
+## 15505. Supporting software engineering tasks with agentic AI: Demonstration on document retrieval and test scenario generation
+
+- **ID:** `2602.04726` · **primary:** `cs.SE` · **Submitted:** 2026-02-04
+- **Authors:** Marian Kica, Lukas Radosky, David Slivka, et al.
+- **abs:** https://arxiv.org/abs/2602.04726 · **PDF:** https://arxiv.org/pdf/2602.04726v1
+
+## 15506. Benchmarking and Enhancing PPG-Based Cuffless Blood Pressure Estimation Methods
+
+- **ID:** `2602.04725` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Neville Mathew, Yidan Shen, Renjie Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.04725 · **PDF:** https://arxiv.org/pdf/2602.04725v1
+
+## 15507. Identifying Intervenable and Interpretable Features via Orthogonality Regularization
+
+- **ID:** `2602.04718` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Moritz Miller, Florent Draye, Bernhard Schölkopf
+- **abs:** https://arxiv.org/abs/2602.04718 · **PDF:** https://arxiv.org/pdf/2602.04718v1
+
+## 15508. Evolutionary Mapping of Neural Networks to Spatial Accelerators
+
+- **ID:** `2602.04717` · **primary:** `cs.NE` · **Submitted:** 2026-02-04
+- **Authors:** Alessandro Pierro, Jonathan Timcheck, Jason Yik, et al.
+- **abs:** https://arxiv.org/abs/2602.04717 · **PDF:** https://arxiv.org/pdf/2602.04717v1
+
+## 15509. Linguistically Informed Evaluation of Multilingual ASR for African Languages
+
+- **ID:** `2602.04716` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Fei-Yueh Chen, Lateef Adeleke, C. M. Downey
+- **abs:** https://arxiv.org/abs/2602.04716 · **PDF:** https://arxiv.org/pdf/2602.04716v1
+
+## 15510. Bounded-Abstention Multi-horizon Time-series Forecasting
+
+- **ID:** `2602.04714` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Luca Stradiotti, Laurens Devos, Anna Monreale, et al.
+- **abs:** https://arxiv.org/abs/2602.04714 · **PDF:** https://arxiv.org/pdf/2602.04714v1
+
+## 15511. Adaptive Prompt Elicitation for Text-to-Image Generation
+
+- **ID:** `2602.04713` · **primary:** `cs.HC` · **Submitted:** 2026-02-04
+- **Authors:** Xinyi Wen, Lena Hegemann, Xiaofu Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.04713 · **PDF:** https://arxiv.org/pdf/2602.04713v1
+
+## 15512. SAR-RAG: ATR Visual Question Answering by Semantic Search, Retrieval, and MLLM Generation
+
+- **ID:** `2602.04712` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** David F. Ramirez, Tim Overman, Kristen Jaskie, et al.
+- **abs:** https://arxiv.org/abs/2602.04712 · **PDF:** https://arxiv.org/pdf/2602.04712v1
+
+## 15513. Addressing Corpus Knowledge Poisoning Attacks on RAG Using Sparse Attention
+
+- **ID:** `2602.04711` · **primary:** `cs.IR` · **Submitted:** 2026-02-04
+- **Authors:** Sagie Dekel, Moshe Tennenholtz, Oren Kurland
+- **abs:** https://arxiv.org/abs/2602.04711 · **PDF:** https://arxiv.org/pdf/2602.04711v2
+
+## 15514. Towards Understanding and Avoiding Limitations of Convolutions on Graphs
+
+- **ID:** `2602.04709` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Andreas Roth
+- **abs:** https://arxiv.org/abs/2602.04709 · **PDF:** https://arxiv.org/pdf/2602.04709v1
+
+## 15515. LiteToken: Removing Intermediate Merge Residues From BPE Tokenizers
+
+- **ID:** `2602.04706` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Yike Sun, Haotong Yang, Zhouchen Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.04706 · **PDF:** https://arxiv.org/pdf/2602.04706v1
+
+## 15516. ERNIE 5.0 Technical Report
+
+- **ID:** `2602.04705` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Haifeng Wang, Hua Wu, Tian Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.04705 · **PDF:** https://arxiv.org/pdf/2602.04705v1
+
+## 15517. Knowledge Distillation for mmWave Beam Prediction Using Sub-6 GHz Channels
+
+- **ID:** `2602.04703` · **primary:** `eess.SP` · **Submitted:** 2026-02-04
+- **Authors:** Sina Tavakolian, Nhan Thanh Nguyen, Ahmed Alkhateeb, et al.
+- **abs:** https://arxiv.org/abs/2602.04703 · **PDF:** https://arxiv.org/pdf/2602.04703v1
+
+## 15518. Linear Model Merging Unlocks Simple and Scalable Multimodal Data Mixture Optimization
+
+- **ID:** `2602.04937` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Davide Berasi, Matteo Farina, Massimiliano Mancini, et al.
+- **abs:** https://arxiv.org/abs/2602.04937 · **PDF:** https://arxiv.org/pdf/2602.04937v1
+
+## 15519. Rethinking Personalization in Large Language Models at the Token Level
+
+- **ID:** `2603.06595` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Chenheng Zhang, Yijun Lu, Lizhe Fang, et al.
+- **abs:** https://arxiv.org/abs/2603.06595 · **PDF:** https://arxiv.org/pdf/2603.06595v1
+
+## 15520. Neural Sentinel: Unified Vision Language Model (VLM) for License Plate Recognition with Human-in-the-Loop Continual Learning
+
+- **ID:** `2602.07051` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Karthik Sivakoti
+- **abs:** https://arxiv.org/abs/2602.07051 · **PDF:** https://arxiv.org/pdf/2602.07051v1
+
+## 15521. Beyond Learning on Molecules by Weakly Supervising on Molecules
+
+- **ID:** `2602.04696` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-04
+- **Authors:** Gordan Prastalo, Kevin Maik Jablonka
+- **abs:** https://arxiv.org/abs/2602.04696 · **PDF:** https://arxiv.org/pdf/2602.04696v1
+
+## 15522. Same Input, Different Scores: A Multi Model Study on the Inconsistency of LLM Judge
+
+- **ID:** `2603.04417` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Fiona Lau
+- **abs:** https://arxiv.org/abs/2603.04417 · **PDF:** https://arxiv.org/pdf/2603.04417v1
+
+## 15523. LinGO: A Linguistic Graph Optimization Framework with LLMs for Interpreting Intents of Online Uncivil Discourse
+
+- **ID:** `2602.04693` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Yuan Zhang, Thales Bertaglia
+- **abs:** https://arxiv.org/abs/2602.04693 · **PDF:** https://arxiv.org/pdf/2602.04693v1
+
+## 15524. DRMOT: A Dataset and Framework for RGBD Referring Multi-Object Tracking
+
+- **ID:** `2602.04692` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Sijia Chen, Lijuan Ma, Yanqiu Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.04692 · **PDF:** https://arxiv.org/pdf/2602.04692v2
+
+## 15525. Static and auto-regressive neural emulation of phytoplankton biomass dynamics from physical predictors in the global ocean
+
+- **ID:** `2602.04689` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Mahima Lakra, Ronan Fablet, Lucas Drumetz, et al.
+- **abs:** https://arxiv.org/abs/2602.04689 · **PDF:** https://arxiv.org/pdf/2602.04689v1
+
+## 15526. Investigating Disability Representations in Text-to-Image Models
+
+- **ID:** `2602.04687` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Yang Tian, Yu Fan, Liudmila Zavolokina, et al.
+- **abs:** https://arxiv.org/abs/2602.04687 · **PDF:** https://arxiv.org/pdf/2602.04687v3
+
+## 15527. Audio ControlNet for Fine-Grained Audio Generation and Editing
+
+- **ID:** `2602.04680` · **primary:** `cs.SD` · **Submitted:** 2026-02-04
+- **Authors:** Haina Zhu, Yao Xiao, Xiquan Li, et al.
+- **abs:** https://arxiv.org/abs/2602.04680 · **PDF:** https://arxiv.org/pdf/2602.04680v1
+
+## 15528. Let Experts Feel Uncertainty: A Multi-Expert Label Distribution Approach to Probabilistic Time Series Forecasting
+
+- **ID:** `2602.04678` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Zhen Zhou, Zhirui Wang, Qi Hong, et al.
+- **abs:** https://arxiv.org/abs/2602.04678 · **PDF:** https://arxiv.org/pdf/2602.04678v1
+
+## 15529. REDistill: Robust Estimator Distillation for Balancing Robustness and Efficiency
+
+- **ID:** `2602.04677` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Ondrej Tybl, Lukas Neumann
+- **abs:** https://arxiv.org/abs/2602.04677 · **PDF:** https://arxiv.org/pdf/2602.04677v1
+
+## 15530. Generalized Schrödinger Bridge on Graphs
+
+- **ID:** `2602.04675` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Panagiotis Theodoropoulos, Juno Nam, Evangelos Theodorou, et al.
+- **abs:** https://arxiv.org/abs/2602.04675 · **PDF:** https://arxiv.org/pdf/2602.04675v1
+
+## 15531. Overstating Attitudes, Ignoring Networks: LLM Biases in Simulating Misinformation Susceptibility
+
+- **ID:** `2602.04674` · **primary:** `cs.SI` · **Submitted:** 2026-02-04
+- **Authors:** Eun Cheol Choi, Lindsay E. Young, Emilio Ferrara
+- **abs:** https://arxiv.org/abs/2602.04674 · **PDF:** https://arxiv.org/pdf/2602.04674v1
+
+## 15532. Delving into Muon and Beyond: Deep Analysis and Extensions
+
+- **ID:** `2602.04669` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Xianbiao Qi, Marco Chen, Jiaquan Ye, et al.
+- **abs:** https://arxiv.org/abs/2602.04669 · **PDF:** https://arxiv.org/pdf/2602.04669v1
+
+## 15533. Causal explanations of outliers in systems with lagged time-dependencies
+
+- **ID:** `2602.04667` · **primary:** `stat.ML` · **Submitted:** 2026-02-04
+- **Authors:** Philipp Alexander Schwarz, Johannes Oberpriller, Sven Klaassen
+- **abs:** https://arxiv.org/abs/2602.04667 · **PDF:** https://arxiv.org/pdf/2602.04667v1
+
+## 15534. Rethinking the Design Space of Reinforcement Learning for Diffusion Models: On the Importance of Likelihood Estimation Beyond Loss Design
+
+- **ID:** `2602.04663` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Jaemoo Choi, Yuchen Zhu, Wei Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.04663 · **PDF:** https://arxiv.org/pdf/2602.04663v1
+
+## 15535. Approaches to Semantic Textual Similarity in Slovak Language: From Algorithms to Transformers
+
+- **ID:** `2602.04659` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Lukas Radosky, Miroslav Blstak, Matej Krajcovic, et al.
+- **abs:** https://arxiv.org/abs/2602.04659 · **PDF:** https://arxiv.org/pdf/2602.04659v1
+
+## 15536. LATMiX: Learnable Affine Transformations for Microscaling Quantization of LLMs
+
+- **ID:** `2602.17681` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Ofir Gordon, Lior Dikstein, Arnon Netzer, et al.
+- **abs:** https://arxiv.org/abs/2602.17681 · **PDF:** https://arxiv.org/pdf/2602.17681v1
+
+## 15537. Inference-Time Backdoors via Hidden Instructions in LLM Chat Templates
+
+- **ID:** `2602.04653` · **primary:** `cs.CR` · **Submitted:** 2026-02-04
+- **Authors:** Ariel Fogel, Omer Hofman, Eilon Cohen, et al.
+- **abs:** https://arxiv.org/abs/2602.04653 · **PDF:** https://arxiv.org/pdf/2602.04653v3
+
+## 15538. SAFE: Stable Alignment Finetuning with Entropy-Aware Predictive Control for Reinforcement Learning from Human Feedback (RLHF)
+
+- **ID:** `2602.04651` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Dipan Maity
+- **abs:** https://arxiv.org/abs/2602.04651 · **PDF:** https://arxiv.org/pdf/2602.04651v2
+
+## 15539. Learning to Separate RF Signals Under Uncertainty: Detect-Then-Separate vs. Unified Joint Models
+
+- **ID:** `2602.04650` · **primary:** `eess.SP` · **Submitted:** 2026-02-04
+- **Authors:** Ariel Rodrigez, Alejandro Lancho, Amir Weiss
+- **abs:** https://arxiv.org/abs/2602.04650 · **PDF:** https://arxiv.org/pdf/2602.04650v1
+
+## 15540. Outcome Accuracy is Not Enough: Aligning the Reasoning Process of Reward Models
+
+- **ID:** `2602.04649` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Binghai Wang, Yantao Liu, Yuxuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.04649 · **PDF:** https://arxiv.org/pdf/2602.04649v1
+
+## 15541. GRIP: Geometric Refinement and Adaptive Information Potential for Data Efficiency
+
+- **ID:** `2603.00031` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Changhao Wang, Jiaolong Yang, Xinhao Yao, et al.
+- **abs:** https://arxiv.org/abs/2603.00031 · **PDF:** https://arxiv.org/pdf/2603.00031v1
+
+## 15542. Interpreting Physics in Video World Models
+
+- **ID:** `2602.07050` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Sonia Joseph, Quentin Garrido, Randall Balestriero, et al.
+- **abs:** https://arxiv.org/abs/2602.07050 · **PDF:** https://arxiv.org/pdf/2602.07050v1
+
+## 15543. LLM-Assisted Replication for Quantitative Social Science
+
+- **ID:** `2602.18453` · **primary:** `cs.CY` · **Submitted:** 2026-02-04
+- **Authors:** So Kubota, Hiromu Yakura, Samuel Coavoux, et al.
+- **abs:** https://arxiv.org/abs/2602.18453 · **PDF:** https://arxiv.org/pdf/2602.18453v1
+
+## 15544. A Coin Flip for Safety: LLM Judges Fail to Reliably Measure Adversarial Robustness
+
+- **ID:** `2603.06594` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Leo Schwinn, Moritz Ladenburger, Tim Beyer, et al.
+- **abs:** https://arxiv.org/abs/2603.06594 · **PDF:** https://arxiv.org/pdf/2603.06594v2
+
+## 15545. MTS-JEPA: Multi-Resolution Joint-Embedding Predictive Architecture for Time-Series Anomaly Prediction
+
+- **ID:** `2602.04643` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Yanan He, Yunshi Wen, Xin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.04643 · **PDF:** https://arxiv.org/pdf/2602.04643v1
+
+## 15546. Towards Structured, State-Aware, and Execution-Grounded Reasoning for Software Engineering Agents
+
+- **ID:** `2602.04640` · **primary:** `cs.SE` · **Submitted:** 2026-02-04
+- **Authors:** Tse-Hsun, Chen
+- **abs:** https://arxiv.org/abs/2602.04640 · **PDF:** https://arxiv.org/pdf/2602.04640v1
+
+## 15547. RIGA-Fold: A General Framework for Protein Inverse Folding via Recurrent Interaction and Geometric Awareness
+
+- **ID:** `2602.04637` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Sisi Yuan, Jiehuang Chen, Junchuang Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.04637 · **PDF:** https://arxiv.org/pdf/2602.04637v1
+
+## 15548. WideSeek-R1: Exploring Width Scaling for Broad Information Seeking via Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.04634` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Zelai Xu, Zhexuan Xu, Ruize Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.04634 · **PDF:** https://arxiv.org/pdf/2602.04634v3
+
+## 15549. Mapping the Web of Science, a large-scale graph and text-based dataset with LLM embeddings
+
+- **ID:** `2602.04630` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Tim Kunt, Annika Buchholz, Imene Khebouri, et al.
+- **abs:** https://arxiv.org/abs/2602.04630 · **PDF:** https://arxiv.org/pdf/2602.04630v1
+
+## 15550. Hierarchical Embedding Fusion for Retrieval-Augmented Code Generation
+
+- **ID:** `2603.06593` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Nikita Sorokin, Ivan Sedykh, Valentin Malykh
+- **abs:** https://arxiv.org/abs/2603.06593 · **PDF:** https://arxiv.org/pdf/2603.06593v1
+
+## 15551. QUATRO: Query-Adaptive Trust Region Policy Optimization for LLM Fine-tuning
+
+- **ID:** `2602.04620` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Doyeon Lee, Eunyi Lyou, Hyunsoo Cho, et al.
+- **abs:** https://arxiv.org/abs/2602.04620 · **PDF:** https://arxiv.org/pdf/2602.04620v2
+
+## 15552. Hierarchical Latent Structures in Data Generation Process Unify Mechanistic Phenomena across Scale
+
+- **ID:** `2603.06592` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Jonas Rohweder, Subhabrata Dutta, Iryna Gurevych
+- **abs:** https://arxiv.org/abs/2603.06592 · **PDF:** https://arxiv.org/pdf/2603.06592v1
+
+## 15553. LEAD: Layer-wise Expert-aligned Decoding for Faithful Radiology Report Generation
+
+- **ID:** `2602.04617` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Ruixiao Yang, Yuanhe Tian, Xu Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.04617 · **PDF:** https://arxiv.org/pdf/2602.04617v1
+
+## 15554. A Human-Centered Privacy Approach (HCP) to AI
+
+- **ID:** `2602.04616` · **primary:** `cs.HC` · **Submitted:** 2026-02-04
+- **Authors:** Luyi Sun, Wei Xu, Zaifeng Gao
+- **abs:** https://arxiv.org/abs/2602.04616 · **PDF:** https://arxiv.org/pdf/2602.04616v1
+
+## 15555. Disentangling meaning from language in LLM-based machine translation
+
+- **ID:** `2602.04613` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Théo Lasnier, Armel Zebaze, Djamé Seddah, et al.
+- **abs:** https://arxiv.org/abs/2602.04613 · **PDF:** https://arxiv.org/pdf/2602.04613v1
+
+## 15556. Targeted Synthetic Control Method
+
+- **ID:** `2602.04611` · **primary:** `stat.ML` · **Submitted:** 2026-02-04
+- **Authors:** Yuxin Wang, Dennis Frauen, Emil Javurek, et al.
+- **abs:** https://arxiv.org/abs/2602.04611 · **PDF:** https://arxiv.org/pdf/2602.04611v1
+
+## 15557. Resilient Load Forecasting under Climate Change: Adaptive Conditional Neural Processes for Few-Shot Extreme Load Forecasting
+
+- **ID:** `2602.04609` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Chenxi Hu, Yue Ma, Yifan Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.04609 · **PDF:** https://arxiv.org/pdf/2602.04609v1
+
+## 15558. Jacobian Regularization Stabilizes Long-Term Integration of Neural Differential Equations
+
+- **ID:** `2602.04608` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Maya Janvier, Julien Salomon, Etienne Meunier
+- **abs:** https://arxiv.org/abs/2602.04608 · **PDF:** https://arxiv.org/pdf/2602.04608v1
+
+## 15559. Focus-LIME: Surgical Interpretation of Long-Context Large Language Models via Proxy-Based Neighborhood Selection
+
+- **ID:** `2602.04607` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Junhao Liu, Haonan Yu, Zhenyu Yan, et al.
+- **abs:** https://arxiv.org/abs/2602.04607 · **PDF:** https://arxiv.org/pdf/2602.04607v1
+
+## 15560. RexBERT: Context Specialized Bidirectional Encoders for E-commerce
+
+- **ID:** `2602.04605` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Rahul Bajaj, Anuj Garg
+- **abs:** https://arxiv.org/abs/2602.04605 · **PDF:** https://arxiv.org/pdf/2602.04605v1
+
+## 15561. Beyond Holistic Scores: Automatic Trait-Based Quality Scoring of Argumentative Essays
+
+- **ID:** `2602.04604` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Lucile Favero, Juan Antonio Pérez-Ortiz, Tanja Käser, et al.
+- **abs:** https://arxiv.org/abs/2602.04604 · **PDF:** https://arxiv.org/pdf/2602.04604v1
+
+## 15562. Stochastic Decision Horizons for Constrained Reinforcement Learning
+
+- **ID:** `2602.04599` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Nikola Milosevic, Leonard Franz, Daniel Haeufle, et al.
+- **abs:** https://arxiv.org/abs/2602.04599 · **PDF:** https://arxiv.org/pdf/2602.04599v1
+
+## 15563. A principled framework for uncertainty decomposition in TabPFN
+
+- **ID:** `2602.04596` · **primary:** `stat.ML` · **Submitted:** 2026-02-04
+- **Authors:** Sandra Fortini, Kenyon Ng, Sonia Petrone, et al.
+- **abs:** https://arxiv.org/abs/2602.04596 · **PDF:** https://arxiv.org/pdf/2602.04596v1
+
+## 15564. ASA: Training-Free Representation Engineering for Tool-Calling Agents
+
+- **ID:** `2602.04935` · **primary:** `cs.SE` · **Submitted:** 2026-02-04
+- **Authors:** Youjin Wang, Run Zhou, Rong Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.04935 · **PDF:** https://arxiv.org/pdf/2602.04935v2
+
+## 15565. VILLAIN at AVerImaTeC: Verifying Image-Text Claims via Multi-Agent Collaboration
+
+- **ID:** `2602.04587` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Jaeyoon Jung, Yejun Yoon, Kunwoo Park
+- **abs:** https://arxiv.org/abs/2602.04587 · **PDF:** https://arxiv.org/pdf/2602.04587v2
+
+## 15566. Real-time processing of analog signals on accelerated neuromorphic hardware
+
+- **ID:** `2602.04582` · **primary:** `cs.NE` · **Submitted:** 2026-02-04
+- **Authors:** Yannik Stradmann, Johannes Schemmel, Mihai A. Petrovici, et al.
+- **abs:** https://arxiv.org/abs/2602.04582 · **PDF:** https://arxiv.org/pdf/2602.04582v2
+
+## 15567. Trust The Typical
+
+- **ID:** `2602.04581` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Debargha Ganguly, Sreehari Sankar, Biyao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.04581 · **PDF:** https://arxiv.org/pdf/2602.04581v1
+
+## 15568. AIANO: Enhancing Information Retrieval with AI-Augmented Annotation
+
+- **ID:** `2602.04579` · **primary:** `cs.IR` · **Submitted:** 2026-02-04
+- **Authors:** Sameh Khattab, Marie Bauer, Lukas Heine, et al.
+- **abs:** https://arxiv.org/abs/2602.04579 · **PDF:** https://arxiv.org/pdf/2602.04579v1
+
+## 15569. Semantic Self-Distillation for Language Model Uncertainty
+
+- **ID:** `2602.04577` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Edward Phillips, Sean Wu, Fredrik K. Gustafsson, et al.
+- **abs:** https://arxiv.org/abs/2602.04577 · **PDF:** https://arxiv.org/pdf/2602.04577v2
+
+## 15570. Vibe AIGC: A New Paradigm for Content Generation via Agentic Orchestration
+
+- **ID:** `2602.04575` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Jiaheng Liu, Yuanxing Zhang, Shihao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.04575 · **PDF:** https://arxiv.org/pdf/2602.04575v2
+
+## 15571. Probabilistic Label Spreading: Efficient and Consistent Estimation of Soft Labels with Epistemic Uncertainty on Graphs
+
+- **ID:** `2602.04574` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Jonathan Klees, Tobias Riedlinger, Peter Stehr, et al.
+- **abs:** https://arxiv.org/abs/2602.04574 · **PDF:** https://arxiv.org/pdf/2602.04574v1
+
+## 15572. From Competition to Collaboration: Designing Sustainable Mechanisms Between LLMs and Online Forums
+
+- **ID:** `2602.04572` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Niv Fono, Yftah Ziser, Omer Ben-Porat
+- **abs:** https://arxiv.org/abs/2602.04572 · **PDF:** https://arxiv.org/pdf/2602.04572v1
+
+## 15573. Can LLMs capture stable human-generated sentence entropy measures?
+
+- **ID:** `2602.04570` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Estrella Pivel-Villanueva, Elisabeth Frederike Sterner, Franziska Knolle
+- **abs:** https://arxiv.org/abs/2602.04570 · **PDF:** https://arxiv.org/pdf/2602.04570v1
+
+## 15574. Dual Mind World Model Inspired Network Digital Twin for Access Scheduling
+
+- **ID:** `2602.04566` · **primary:** `cs.NI` · **Submitted:** 2026-02-04
+- **Authors:** Hrishikesh Dutta, Roberto Minerva, Noel Crespi
+- **abs:** https://arxiv.org/abs/2602.04566 · **PDF:** https://arxiv.org/pdf/2602.04566v1
+
+## 15575. Textual Planning with Explicit Latent Transitions
+
+- **ID:** `2602.04557` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Eliezer Shlomi, Ido Levy, Eilam Shapira, et al.
+- **abs:** https://arxiv.org/abs/2602.04557 · **PDF:** https://arxiv.org/pdf/2602.04557v1
+
+## 15576. Enhancing IMU-Based Online Handwriting Recognition via Contrastive Learning with Zero Inference Overhead
+
+- **ID:** `2602.07049` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Jindong Li, Dario Zanca, Vincent Christlein, et al.
+- **abs:** https://arxiv.org/abs/2602.07049 · **PDF:** https://arxiv.org/pdf/2602.07049v1
+
+## 15577. Rethinking Weight Tying: Pseudo-Inverse Tying for Stable LM Training and Updates
+
+- **ID:** `2602.04556` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Jian Gu, Aldeida Aleti, Chunyang Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.04556 · **PDF:** https://arxiv.org/pdf/2602.04556v1
+
+## 15578. Finding Structure in Continual Learning
+
+- **ID:** `2602.04555` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Pourya Shamsolmoali, Masoumeh Zareapoor
+- **abs:** https://arxiv.org/abs/2602.04555 · **PDF:** https://arxiv.org/pdf/2602.04555v1
+
+## 15579. Gradient Flow Through Diagram Expansions: Learning Regimes and Explicit Solutions
+
+- **ID:** `2602.04548` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Dmitry Yarotsky, Eugene Golikov, Yaroslav Gusev
+- **abs:** https://arxiv.org/abs/2602.04548 · **PDF:** https://arxiv.org/pdf/2602.04548v1
+
+## 15580. OmniRad: A Radiological Foundation Model for Multi-Task Medical Image Analysis
+
+- **ID:** `2602.04547` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Luca Zedda, Andrea Loddo, Cecilia Di Ruberto
+- **abs:** https://arxiv.org/abs/2602.04547 · **PDF:** https://arxiv.org/pdf/2602.04547v1
+
+## 15581. Unmasking Superspreaders: Data-Driven Approaches for Identifying and Comparing Key Influencers of Conspiracy Theories on X.com
+
+- **ID:** `2602.04546` · **primary:** `cs.SI` · **Submitted:** 2026-02-04
+- **Authors:** Florian Kramer, Henrich R. Greve, Moritz von Zahn, et al.
+- **abs:** https://arxiv.org/abs/2602.04546 · **PDF:** https://arxiv.org/pdf/2602.04546v1
+
+## 15582. Continual Learning through Control Minimization
+
+- **ID:** `2602.04542` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Sander de Haan, Yassine Taoudi-Benchekroun, Pau Vilimelis Aceituno, et al.
+- **abs:** https://arxiv.org/abs/2602.04542 · **PDF:** https://arxiv.org/pdf/2602.04542v1
+
+## 15583. LycheeDecode: Accelerating Long-Context LLM Inference via Hybrid-Head Sparse Decoding
+
+- **ID:** `2602.04541` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Gang Lin, Dongfang Li, Zhuoen Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.04541 · **PDF:** https://arxiv.org/pdf/2602.04541v1
+
+## 15584. PersoPilot: An Adaptive AI-Copilot for Transparent Contextualized Persona Classification and Personalized Response Generation
+
+- **ID:** `2602.04540` · **primary:** `cs.HC` · **Submitted:** 2026-02-04
+- **Authors:** Saleh Afzoon, Amin Beheshti, Usman Naseem
+- **abs:** https://arxiv.org/abs/2602.04540 · **PDF:** https://arxiv.org/pdf/2602.04540v1
+
+## 15585. Forget to Generalize: Iterative Adaptation for Generalization in Federated Learning
+
+- **ID:** `2602.04536` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Abdulrahman Alotaibi, Irene Tenison, Miriam Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.04536 · **PDF:** https://arxiv.org/pdf/2602.04536v1
+
+## 15586. RA-QA: A Benchmarking System for Respiratory Audio Question Answering Under Real-World Heterogeneity
+
+- **ID:** `2602.18452` · **primary:** `cs.SD` · **Submitted:** 2026-02-04
+- **Authors:** Gaia A. Bertolino, Yuwei Zhang, Tong Xia, et al.
+- **abs:** https://arxiv.org/abs/2602.18452 · **PDF:** https://arxiv.org/pdf/2602.18452v2
+
+## 15587. BioBridge: Bridging Proteins and Language for Enhanced Biological Reasoning with LLMs
+
+- **ID:** `2602.17680` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Yujia Wang, Jihong Guan, Wengen Li, et al.
+- **abs:** https://arxiv.org/abs/2602.17680 · **PDF:** https://arxiv.org/pdf/2602.17680v1
+
+## 15588. Landscape-aware Automated Algorithm Design: An Efficient Framework for Real-world Optimization
+
+- **ID:** `2602.04529` · **primary:** `cs.NE` · **Submitted:** 2026-02-04
+- **Authors:** Haoran Yin, Shuaiqun Pan, Zhao Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.04529 · **PDF:** https://arxiv.org/pdf/2602.04529v1
+
+## 15589. SLUM-i: Semi-supervised Learning for Urban Mapping of Informal Settlements and Data Quality Benchmarking
+
+- **ID:** `2602.04525` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Muhammad Taha Mukhtar, Syed Musa Ali Kazmi, Khola Naseem, et al.
+- **abs:** https://arxiv.org/abs/2602.04525 · **PDF:** https://arxiv.org/pdf/2602.04525v1
+
+## 15590. $C$-$ΔΘ$: Circuit-Restricted Weight Arithmetic for Selective Refusal
+
+- **ID:** `2602.04521` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Aditya Kasliwal, Pratinav Seth, Vinay Kumar Sankarapu
+- **abs:** https://arxiv.org/abs/2602.04521 · **PDF:** https://arxiv.org/pdf/2602.04521v1
+
+## 15591. Learning the Value Systems of Agents with Preference-based and Inverse Reinforcement Learning
+
+- **ID:** `2602.04518` · **primary:** `cs.CY` · **Submitted:** 2026-02-04
+- **Authors:** Andrés Holgado-Sánchez, Holger Billhardt, Alberto Fernández, et al.
+- **abs:** https://arxiv.org/abs/2602.04518 · **PDF:** https://arxiv.org/pdf/2602.04518v1
+
+## 15592. ReFRAME or Remain: Unsupervised Lexical Semantic Change Detection with Frame Semantics
+
+- **ID:** `2602.04514` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Bach Phan-Tat, Kris Heylen, Dirk Geeraerts, et al.
+- **abs:** https://arxiv.org/abs/2602.04514 · **PDF:** https://arxiv.org/pdf/2602.04514v2
+
+## 15593. Combining scEEG and PPG for reliable sleep staging using lightweight wearables
+
+- **ID:** `2602.15042` · **primary:** `eess.SP` · **Submitted:** 2026-02-04
+- **Authors:** Jiawei Wang, Liang Xu, Shuntian Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.15042 · **PDF:** https://arxiv.org/pdf/2602.15042v1
+
+## 15594. BrainVista: Modeling Naturalistic Brain Dynamics as Multimodal Next-Token Prediction
+
+- **ID:** `2602.04512` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-04
+- **Authors:** Xuanhua Yin, Runkai Zhao, Lina Yao, et al.
+- **abs:** https://arxiv.org/abs/2602.04512 · **PDF:** https://arxiv.org/pdf/2602.04512v1
+
+## 15595. Model-Dowser: Data-Free Importance Probing to Mitigate Catastrophic Forgetting in Multimodal Large Language Models
+
+- **ID:** `2602.04509` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Hyeontaek Hwang, Nguyen Dinh Son, Daeyoung Kim
+- **abs:** https://arxiv.org/abs/2602.04509 · **PDF:** https://arxiv.org/pdf/2602.04509v3
+
+## 15596. Forecasting Future Language: Context Design for Mention Markets
+
+- **ID:** `2602.21229` · **primary:** `q-fin.GN` · **Submitted:** 2026-02-04
+- **Authors:** Sumin Kim, Jihoon Kwon, Yoon Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.21229 · **PDF:** https://arxiv.org/pdf/2602.21229v2
+
+## 15597. ReThinker: Scientific Reasoning by Rethinking with Guided Reflection and Confidence Control
+
+- **ID:** `2602.04496` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Zhentao Tang, Yuqi Cui, Shixiong Kai, et al.
+- **abs:** https://arxiv.org/abs/2602.04496 · **PDF:** https://arxiv.org/pdf/2602.04496v1
+
+## 15598. Comparing Euclidean and Hyperbolic K-Means for Generalized Category Discovery
+
+- **ID:** `2602.04932` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Mohamad Dalal, Thomas B. Moeslund, Joakim Bruslund Haurum
+- **abs:** https://arxiv.org/abs/2602.04932 · **PDF:** https://arxiv.org/pdf/2602.04932v1
+
+## 15599. PersoDPO: Scalable Preference Optimization for Instruction-Adherent, Persona-Grounded Dialogue via Multi-LLM Evaluation
+
+- **ID:** `2602.04493` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Saleh Afzoon, MohammadHossein Ahmadi, Usman Naseem, et al.
+- **abs:** https://arxiv.org/abs/2602.04493 · **PDF:** https://arxiv.org/pdf/2602.04493v1
+
+## 15600. Discovering Mechanistic Models of Neural Activity: System Identification in an in Silico Zebrafish
+
+- **ID:** `2602.04492` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-04
+- **Authors:** Jan-Matthis Lueckmann, Viren Jain, Michał Januszewski
+- **abs:** https://arxiv.org/abs/2602.04492 · **PDF:** https://arxiv.org/pdf/2602.04492v1
+
+## 15601. Greedy-Gnorm: A Gradient Matrix Norm-Based Alternative to Attention Entropy for Head Pruning
+
+- **ID:** `2602.04491` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Yuxi Guo, Paul Sheridan
+- **abs:** https://arxiv.org/abs/2602.04491 · **PDF:** https://arxiv.org/pdf/2602.04491v1
+
+## 15602. Deconstructing sentence disambiguation by joint latent modeling of reading paradigms: LLM surprisal is not enough
+
+- **ID:** `2602.04489` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Dario Paape, Tal Linzen, Shravan Vasishth
+- **abs:** https://arxiv.org/abs/2602.04489 · **PDF:** https://arxiv.org/pdf/2602.04489v1
+
+## 15603. Beyond Unimodal Shortcuts: MLLMs as Cross-Modal Reasoners for Grounded Named Entity Recognition
+
+- **ID:** `2602.04486` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Jinlong Ma, Yu Zhang, Xuefeng Bai, et al.
+- **abs:** https://arxiv.org/abs/2602.04486 · **PDF:** https://arxiv.org/pdf/2602.04486v1
+
+## 15604. Universality of General Spiked Tensor Models
+
+- **ID:** `2602.04472` · **primary:** `math.ST` · **Submitted:** 2026-02-04
+- **Authors:** Yanjin Xiang, Zhihua Zhang
+- **abs:** https://arxiv.org/abs/2602.04472 · **PDF:** https://arxiv.org/pdf/2602.04472v2
+
+## 15605. LLM-Empowered Cooperative Content Caching in Vehicular Fog Caching-Assisted Platoon Networks
+
+- **ID:** `2602.04471` · **primary:** `cs.NI` · **Submitted:** 2026-02-04
+- **Authors:** Bowen Tan, Qiong Wu, Pingyi Fan, et al.
+- **abs:** https://arxiv.org/abs/2602.04471 · **PDF:** https://arxiv.org/pdf/2602.04471v1
+
+## 15606. Is Micro Domain-Adaptive Pre-Training Effective for Real-World Operations? Multi-Step Evaluation Reveals Potential and Bottlenecks
+
+- **ID:** `2602.04466` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Masaya Tsunokake, Yuta Koreeda, Terufumi Morishita, et al.
+- **abs:** https://arxiv.org/abs/2602.04466 · **PDF:** https://arxiv.org/pdf/2602.04466v1
+
+## 15607. Bayesian PINNs for uncertainty-aware inverse problems (BPINN-IP)
+
+- **ID:** `2602.04459` · **primary:** `stat.ML` · **Submitted:** 2026-02-04
+- **Authors:** Ali Mohammad-Djafari
+- **abs:** https://arxiv.org/abs/2602.04459 · **PDF:** https://arxiv.org/pdf/2602.04459v1
+
+## 15608. Journey to the Centre of Cluster: Harnessing Interior Nodes for A/B Testing under Network Interference
+
+- **ID:** `2602.04457` · **primary:** `stat.ME` · **Submitted:** 2026-02-04
+- **Authors:** Qianyi Chen, Anpeng Wu, Bo Li, et al.
+- **abs:** https://arxiv.org/abs/2602.04457 · **PDF:** https://arxiv.org/pdf/2602.04457v1
+
+## 15609. Growth First, Care Second? Tracing the Landscape of LLM Value Preferences in Everyday Dilemmas
+
+- **ID:** `2602.04456` · **primary:** `cs.CY` · **Submitted:** 2026-02-04
+- **Authors:** Zhiyi Chen, Eun Cheol Choi, Yingjia Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.04456 · **PDF:** https://arxiv.org/pdf/2602.04456v1
+
+## 15610. RASA: Routing-Aware Safety Alignment for Mixture-of-Experts Models
+
+- **ID:** `2602.04448` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Jiacheng Liang, Yuhui Wang, Tanqiu Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.04448 · **PDF:** https://arxiv.org/pdf/2602.04448v1
+
+## 15611. Mixture of Masters: Sparse Chess Language Models with Player Routing
+
+- **ID:** `2602.04447` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Giacomo Frisoni, Lorenzo Molfetta, Davide Freddi, et al.
+- **abs:** https://arxiv.org/abs/2602.04447 · **PDF:** https://arxiv.org/pdf/2602.04447v1
+
+## 15612. No One-Size-Fits-All: Building Systems For Translation to Bashkir, Kazakh, Kyrgyz, Tatar and Chuvash Using Synthetic And Original Data
+
+- **ID:** `2602.04442` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Dmitry Karpov
+- **abs:** https://arxiv.org/abs/2602.04442 · **PDF:** https://arxiv.org/pdf/2602.04442v1
+
+## 15613. Hand Gesture Recognition from Doppler Radar Signals Using Echo State Networks
+
+- **ID:** `2602.04436` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Towa Sano, Gouhei Tanaka
+- **abs:** https://arxiv.org/abs/2602.04436 · **PDF:** https://arxiv.org/pdf/2602.04436v1
+
+## 15614. How Attention Sinks Emerge in Large Language Models: An Interpretability Perspective
+
+- **ID:** `2603.06591` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Runyu Peng, Ruixiao Li, Mingshu Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.06591 · **PDF:** https://arxiv.org/pdf/2603.06591v1
+
+## 15615. Machine Learning-Driven Crystal System Prediction for Perovskites Using Augmented X-ray Diffraction Data
+
+- **ID:** `2602.04435` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-04
+- **Authors:** Ansu Mathew, Ahmer A. B. Baloch, Alamin Yakasai, et al.
+- **abs:** https://arxiv.org/abs/2602.04435 · **PDF:** https://arxiv.org/pdf/2602.04435v1
+
+## 15616. MaMa: A Game-Theoretic Approach for Designing Safe Agentic Systems
+
+- **ID:** `2602.04431` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Jonathan Nöther, Adish Singla, Goran Radanovic
+- **abs:** https://arxiv.org/abs/2602.04431 · **PDF:** https://arxiv.org/pdf/2602.04431v1
+
+## 15617. Joint Parameter and State-Space Bayesian Optimization: Using Process Expertise to Accelerate Manufacturing Optimization
+
+- **ID:** `2602.17679` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Saksham Kiroriwal, Julius Pfrommer, Jürgen Beyerer
+- **abs:** https://arxiv.org/abs/2602.17679 · **PDF:** https://arxiv.org/pdf/2602.17679v1
+
+## 15618. DF-ACBlurGAN: Structure-Aware Conditional Generation of Internally Repeated Patterns for Biomaterial Microtopography Design
+
+- **ID:** `2603.28776` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Rongjun Dong, Xin Chen, Morgan R Alexander, et al.
+- **abs:** https://arxiv.org/abs/2603.28776 · **PDF:** https://arxiv.org/pdf/2603.28776v1
+
+## 15619. Fine-Grained Activation Steering: Steering Less, Achieving More
+
+- **ID:** `2602.04428` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Zijian Feng, Tianjiao Li, Zixiao Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.04428 · **PDF:** https://arxiv.org/pdf/2602.04428v1
+
+## 15620. Depth-Wise Emergence of Prediction-Centric Geometry in Large Language Models
+
+- **ID:** `2602.04931` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Shahar Haim, Daniel C McNamee
+- **abs:** https://arxiv.org/abs/2602.04931 · **PDF:** https://arxiv.org/pdf/2602.04931v1
+
+## 15621. SPEAR: An Engineering Case Study of Multi-Agent Coordination for Smart Contract Auditing
+
+- **ID:** `2602.04418` · **primary:** `cs.MA` · **Submitted:** 2026-02-04
+- **Authors:** Arnab Mallick, Indraveni Chebolu, Harmesh Rana
+- **abs:** https://arxiv.org/abs/2602.04418 · **PDF:** https://arxiv.org/pdf/2602.04418v1
+
+## 15622. EMA Policy Gradient: Taming Reinforcement Learning for LLMs with EMA Anchor and Top-k KL
+
+- **ID:** `2602.04417` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Lunjun Zhang, Jimmy Ba
+- **abs:** https://arxiv.org/abs/2602.04417 · **PDF:** https://arxiv.org/pdf/2602.04417v1
+
+## 15623. Med-MMFL: A Multimodal Federated Learning Benchmark in Healthcare
+
+- **ID:** `2602.04416` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Aavash Chhetri, Bibek Niroula, Pratik Shrestha, et al.
+- **abs:** https://arxiv.org/abs/2602.04416 · **PDF:** https://arxiv.org/pdf/2602.04416v1
+
+## 15624. ShapBPT: Image Feature Attributions Using Data-Aware Binary Partition Trees
+
+- **ID:** `2602.07047` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Muhammad Rashid, Elvio G. Amparore, Enrico Ferrari, et al.
+- **abs:** https://arxiv.org/abs/2602.07047 · **PDF:** https://arxiv.org/pdf/2602.07047v3
+
+## 15625. Attack Selection Reduces Safety in Concentrated AI Control Settings against Trusted Monitoring
+
+- **ID:** `2602.04930` · **primary:** `cs.CR` · **Submitted:** 2026-02-04
+- **Authors:** Joachim Schaeffer, Arjun Khandelwal, Tyler Tracy
+- **abs:** https://arxiv.org/abs/2602.04930 · **PDF:** https://arxiv.org/pdf/2602.04930v1
+
+## 15626. History-Guided Iterative Visual Reasoning with Self-Correction
+
+- **ID:** `2602.04413` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Xinglong Yang, Zhilin Peng, Zhanzhan Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.04413 · **PDF:** https://arxiv.org/pdf/2602.04413v1
+
+## 15627. HoRD: Robust Humanoid Control via History-Conditioned Reinforcement Learning and Online Distillation
+
+- **ID:** `2602.04412` · **primary:** `cs.RO` · **Submitted:** 2026-02-04
+- **Authors:** Puyue Wang, Jiawei Hu, Yan Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.04412 · **PDF:** https://arxiv.org/pdf/2602.04412v3
+
+## 15628. Separation-Utility Pareto Frontier: An Information-Theoretic Characterization
+
+- **ID:** `2602.04408` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Shizhou Xu
+- **abs:** https://arxiv.org/abs/2602.04408 · **PDF:** https://arxiv.org/pdf/2602.04408v2
+
+## 15629. Theory of Speciation Transitions in Diffusion Models with General Class Structure
+
+- **ID:** `2602.04404` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Beatrice Achilli, Marco Benedetti, Giulio Biroli, et al.
+- **abs:** https://arxiv.org/abs/2602.04404 · **PDF:** https://arxiv.org/pdf/2602.04404v1
+
+## 15630. Performative Learning Theory
+
+- **ID:** `2602.04402` · **primary:** `stat.ML` · **Submitted:** 2026-02-04
+- **Authors:** Julian Rodemann, Unai Fischer-Abaigar, James Bailie, et al.
+- **abs:** https://arxiv.org/abs/2602.04402 · **PDF:** https://arxiv.org/pdf/2602.04402v2
+
+## 15631. Swordsman: Entropy-Driven Adaptive Block Partition for Efficient Diffusion Language Models
+
+- **ID:** `2602.04399` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Yu Zhang, Xinchen Li, Jialei Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.04399 · **PDF:** https://arxiv.org/pdf/2602.04399v1
+
+## 15632. Bi-directional Bias Attribution: Debiasing Large Language Models without Modifying Prompts
+
+- **ID:** `2602.04398` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Yujie Lin, Kunquan Li, Yixuan Liao, et al.
+- **abs:** https://arxiv.org/abs/2602.04398 · **PDF:** https://arxiv.org/pdf/2602.04398v1
+
+## 15633. Optimal Rates for Feasible Payoff Set Estimation in Games
+
+- **ID:** `2602.04397` · **primary:** `cs.GT` · **Submitted:** 2026-02-04
+- **Authors:** Annalisa Barbara, Riccardo Poiani, Martino Bernasconi, et al.
+- **abs:** https://arxiv.org/abs/2602.04397 · **PDF:** https://arxiv.org/pdf/2602.04397v1
+
+## 15634. LoRDO: Distributed Low-Rank Optimization with Infrequent Communication
+
+- **ID:** `2602.04396` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Andrej Jovanović, Alex Iacob, Mher Safaryan, et al.
+- **abs:** https://arxiv.org/abs/2602.04396 · **PDF:** https://arxiv.org/pdf/2602.04396v1
+
+## 15635. Evaluating the Presence of Sex Bias in Clinical Reasoning by Large Language Models
+
+- **ID:** `2602.04392` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Isabel Tsintsiper, Sheng Wong, Beth Albert, et al.
+- **abs:** https://arxiv.org/abs/2602.04392 · **PDF:** https://arxiv.org/pdf/2602.04392v1
+
+## 15636. Beyond Rejection Sampling: Trajectory Fusion for Scaling Mathematical Reasoning
+
+- **ID:** `2602.04391` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Jie Deng, Hanshuang Tong, Jun Li, et al.
+- **abs:** https://arxiv.org/abs/2602.04391 · **PDF:** https://arxiv.org/pdf/2602.04391v1
+
+## 15637. SOON: Symmetric Orthogonal Operator Network for Global Subseasonal-to-Seasonal Climate Forecasting
+
+- **ID:** `2602.15040` · **primary:** `physics.ao-ph` · **Submitted:** 2026-02-04
+- **Authors:** Ziyu Zhou, Tian Zhou, Shiyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.15040 · **PDF:** https://arxiv.org/pdf/2602.15040v1
+
+## 15638. On the use of LLMs to generate a dataset of Neural Networks
+
+- **ID:** `2602.04388` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Nadia Daoudi, Jordi Cabot
+- **abs:** https://arxiv.org/abs/2602.04388 · **PDF:** https://arxiv.org/pdf/2602.04388v1
+
+## 15639. Digital Twins & ZeroConf AI: Structuring Automated Intelligent Pipelines for Industrial Applications
+
+- **ID:** `2602.04385` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Marco Picone, Fabio Turazza, Matteo Martinelli, et al.
+- **abs:** https://arxiv.org/abs/2602.04385 · **PDF:** https://arxiv.org/pdf/2602.04385v1
+
+## 15640. Blockchain Federated Learning for Sustainable Retail: Reducing Waste through Collaborative Demand Forecasting
+
+- **ID:** `2602.04384` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Fabio Turazza, Alessandro Neri, Marcello Pietri, et al.
+- **abs:** https://arxiv.org/abs/2602.04384 · **PDF:** https://arxiv.org/pdf/2602.04384v1
+
+## 15641. Enabling Real-Time Colonoscopic Polyp Segmentation on Commodity CPUs via Ultra-Lightweight Architecture
+
+- **ID:** `2602.04381` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Weihao Gao, Zhuo Deng, Zheng Gong, et al.
+- **abs:** https://arxiv.org/abs/2602.04381 · **PDF:** https://arxiv.org/pdf/2602.04381v1
+
+## 15642. ARC-AGI-2 Technical Report
+
+- **ID:** `2603.06590` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Wallyson Lemes de Oliveira, Mekhron Bobokhonov, Matteo Caorsi, et al.
+- **abs:** https://arxiv.org/abs/2603.06590 · **PDF:** https://arxiv.org/pdf/2603.06590v1
+
+## 15643. Beyond KL Divergence: Policy Optimization with Flexible Bregman Divergences for LLM Reasoning
+
+- **ID:** `2602.04380` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Rui Yuan, Mykola Khandoga, Vinay Kumar Sankarapu
+- **abs:** https://arxiv.org/abs/2602.04380 · **PDF:** https://arxiv.org/pdf/2602.04380v1
+
+## 15644. Reducing the labeling burden in time-series mapping using Common Ground: a semi-automated approach to tracking changes in land cover and species over time
+
+- **ID:** `2602.04373` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Geethen Singh, Jasper A Slingsby, Tamara B Robinson, et al.
+- **abs:** https://arxiv.org/abs/2602.04373 · **PDF:** https://arxiv.org/pdf/2602.04373v1
+
+## 15645. Multi-scale hypergraph meets LLMs: Aligning large language models for time series analysis
+
+- **ID:** `2602.04369` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Zongjiang Shang, Dongliang Cui, Binqing Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.04369 · **PDF:** https://arxiv.org/pdf/2602.04369v2
+
+## 15646. EXaMCaP: Subset Selection with Entropy Gain Maximization for Probing Capability Gains of Large Chart Understanding Training Sets
+
+- **ID:** `2602.04365` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Jiapeng Liu, Liang Li, Bing Li, et al.
+- **abs:** https://arxiv.org/abs/2602.04365 · **PDF:** https://arxiv.org/pdf/2602.04365v1
+
+## 15647. Anytime-Valid Conformal Risk Control
+
+- **ID:** `2602.04364` · **primary:** `stat.ML` · **Submitted:** 2026-02-04
+- **Authors:** Bror Hultberg, Dave Zachariah, Antônio H. Ribeiro
+- **abs:** https://arxiv.org/abs/2602.04364 · **PDF:** https://arxiv.org/pdf/2602.04364v1
+
+## 15648. TurboBoA: Faster and Exact Attention-aware Quantization without Backpropagation
+
+- **ID:** `2602.04929` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Junhan Kim, Yeo Jeong Park, Seungwoo Son, et al.
+- **abs:** https://arxiv.org/abs/2602.04929 · **PDF:** https://arxiv.org/pdf/2602.04929v1
+
+## 15649. SparVAR: Exploring Sparsity in Visual AutoRegressive Modeling for Training-Free Acceleration
+
+- **ID:** `2602.04361` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Zekun Li, Ning Wang, Tongxin Bai, et al.
+- **abs:** https://arxiv.org/abs/2602.04361 · **PDF:** https://arxiv.org/pdf/2602.04361v2
+
+## 15650. Counterfactual Explanations for Hypergraph Neural Networks
+
+- **ID:** `2602.04360` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Fabiano Veglianti, Lorenzo Antonelli, Gabriele Tolomei
+- **abs:** https://arxiv.org/abs/2602.04360 · **PDF:** https://arxiv.org/pdf/2602.04360v1
+
+## 15651. Can Vision Replace Text in Working Memory? Evidence from Spatial n-Back in Vision-Language Models
+
+- **ID:** `2602.04355` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Sichu Liang, Hongyu Zhu, Wenwen Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.04355 · **PDF:** https://arxiv.org/pdf/2602.04355v1
+
+## 15652. Mosaic Learning: A Framework for Decentralized Learning with Model Fragmentation
+
+- **ID:** `2602.04352` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Sayan Biswas, Davide Frey, Romaric Gaudel, et al.
+- **abs:** https://arxiv.org/abs/2602.04352 · **PDF:** https://arxiv.org/pdf/2602.04352v1
+
+## 15653. Optimizing What We Trust: Reliability-Guided QUBO Selection of Multi-Agent Weak Framing Signals for Arabic Sentiment Prediction
+
+- **ID:** `2603.04416` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Rabab Alkhalifa
+- **abs:** https://arxiv.org/abs/2603.04416 · **PDF:** https://arxiv.org/pdf/2603.04416v1
+
+## 15654. VecSet-Edit: Unleashing Pre-trained LRM for Mesh Editing from Single Image
+
+- **ID:** `2602.04349` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Teng-Fang Hsiao, Bo-Kai Ruan, Yu-Lun Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.04349 · **PDF:** https://arxiv.org/pdf/2602.04349v1
+
+## 15655. A Bandit-Based Approach to Educational Recommender Systems: Contextual Thompson Sampling for Learner Skill Gain Optimization
+
+- **ID:** `2602.04347` · **primary:** `stat.ML` · **Submitted:** 2026-02-04
+- **Authors:** Lukas De Kerpel, Arthur Thuy, Dries F. Benoit
+- **abs:** https://arxiv.org/abs/2602.04347 · **PDF:** https://arxiv.org/pdf/2602.04347v1
+
+## 15656. MirrorLA: Reflecting Feature Map for Vision Linear Attention
+
+- **ID:** `2602.04346` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Weikang Meng, Liangyu Huo, Yadan Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.04346 · **PDF:** https://arxiv.org/pdf/2602.04346v1
+
+## 15657. UnMaskFork: Test-Time Scaling for Masked Diffusion via Deterministic Action Branching
+
+- **ID:** `2602.04344` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Kou Misaki, Takuya Akiba
+- **abs:** https://arxiv.org/abs/2602.04344 · **PDF:** https://arxiv.org/pdf/2602.04344v1
+
+## 15658. Explicit Uncertainty Modeling for Active CLIP Adaptation with Dual Prompt Tuning
+
+- **ID:** `2602.04340` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Qian-Wei Wang, Yaguang Song, Shu-Tao Xia
+- **abs:** https://arxiv.org/abs/2602.04340 · **PDF:** https://arxiv.org/pdf/2602.04340v1
+
+## 15659. RISE: Interactive Visual Diagnosis of Fairness in Machine Learning Models
+
+- **ID:** `2602.04339` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Ray Chen, Christan Grant
+- **abs:** https://arxiv.org/abs/2602.04339 · **PDF:** https://arxiv.org/pdf/2602.04339v1
+
+## 15660. Fine-tuning Pre-trained Vision-Language Models in a Human-Annotation-Free Manner
+
+- **ID:** `2602.04337` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Qian-Wei Wang, Guanghao Meng, Ren Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.04337 · **PDF:** https://arxiv.org/pdf/2602.04337v1
+
+## 15661. Euphonium: Steering Video Flow Matching via Process Reward Gradient Guided Stochastic Dynamics
+
+- **ID:** `2602.04928` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Ruizhe Zhong, Jiesong Lian, Xiaoyue Mi, et al.
+- **abs:** https://arxiv.org/abs/2602.04928 · **PDF:** https://arxiv.org/pdf/2602.04928v2
+
+## 15662. SimpleTool: Parallel Decoding for Real-Time LLM Function Calling
+
+- **ID:** `2603.00030` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Xiaoxin Shi, Jiaxin Wan, Linkang Dong, et al.
+- **abs:** https://arxiv.org/abs/2603.00030 · **PDF:** https://arxiv.org/pdf/2603.00030v1
+
+## 15663. PriMod4AI: Lifecycle-Aware Privacy Threat Modeling for AI Systems using LLM
+
+- **ID:** `2602.04927` · **primary:** `cs.CR` · **Submitted:** 2026-02-04
+- **Authors:** Gautam Savaliya, Robert Aufschläger, Abhishek Subedi, et al.
+- **abs:** https://arxiv.org/abs/2602.04927 · **PDF:** https://arxiv.org/pdf/2602.04927v1
+
+## 15664. Geometry-Aware Optimal Transport: Fast Intrinsic Dimension and Wasserstein Distance Estimation
+
+- **ID:** `2602.04335` · **primary:** `stat.ML` · **Submitted:** 2026-02-04
+- **Authors:** Ferdinand Genans, Olivier Wintenberger
+- **abs:** https://arxiv.org/abs/2602.04335 · **PDF:** https://arxiv.org/pdf/2602.04335v1
+
+## 15665. Pruning Minimal Reasoning Graphs for Efficient Retrieval-Augmented Generation
+
+- **ID:** `2602.04926` · **primary:** `cs.DB` · **Submitted:** 2026-02-04
+- **Authors:** Ning Wang, Kuanyan Zhu, Daniel Yuehwoon Yee, et al.
+- **abs:** https://arxiv.org/abs/2602.04926 · **PDF:** https://arxiv.org/pdf/2602.04926v1
+
+## 15666. Internalizing LLM Reasoning via Discovery and Replay of Latent Actions
+
+- **ID:** `2602.04925` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Zhenning Shi, Yijia Zhu, Junhan Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.04925 · **PDF:** https://arxiv.org/pdf/2602.04925v1
+
+## 15667. From Assumptions to Actions: Turning LLM Reasoning into Uncertainty-Aware Planning for Embodied Agents
+
+- **ID:** `2602.04326` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** SeungWon Seo, SooBin Lim, SeongRae Noh, et al.
+- **abs:** https://arxiv.org/abs/2602.04326 · **PDF:** https://arxiv.org/pdf/2602.04326v1
+
+## 15668. Efficient Equivariant High-Order Crystal Tensor Prediction via Cartesian Local-Environment Many-Body Coupling
+
+- **ID:** `2602.04323` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Dian Jin, Yancheng Yuan, Xiaoming Tao
+- **abs:** https://arxiv.org/abs/2602.04323 · **PDF:** https://arxiv.org/pdf/2602.04323v1
+
+## 15669. A Domain-Specific Curated Benchmark for Entity and Document-Level Relation Extraction
+
+- **ID:** `2602.04320` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Marco Martinelli, Stefano Marchesin, Vanessa Bonato, et al.
+- **abs:** https://arxiv.org/abs/2602.04320 · **PDF:** https://arxiv.org/pdf/2602.04320v1
+
+## 15670. Knowing When to Answer: Adaptive Confidence Refinement for Reliable Audio-Visual Question Answering
+
+- **ID:** `2602.04924` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Dinh Phu Tran, Jihoon Jeong, Saad Wazir, et al.
+- **abs:** https://arxiv.org/abs/2602.04924 · **PDF:** https://arxiv.org/pdf/2602.04924v1
+
+## 15671. Imposing Boundary Conditions on Neural Operators via Learned Function Extensions
+
+- **ID:** `2602.04923` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Sepehr Mousavi, Siddhartha Mishra, Laura De Lorenzis
+- **abs:** https://arxiv.org/abs/2602.04923 · **PDF:** https://arxiv.org/pdf/2602.04923v1
+
+## 15672. VLRS-Bench: A Vision-Language Reasoning Benchmark for Remote Sensing
+
+- **ID:** `2602.07045` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Zhiming Luo, Di Wang, Haonan Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.07045 · **PDF:** https://arxiv.org/pdf/2602.07045v1
+
+## 15673. Universal Robust Speech Adaptation for Cross-Domain Speech Recognition and Enhancement
+
+- **ID:** `2602.04307` · **primary:** `eess.AS` · **Submitted:** 2026-02-04
+- **Authors:** Chien-Chun Wang, Hung-Shin Lee, Hsin-Min Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.04307 · **PDF:** https://arxiv.org/pdf/2602.04307v2
+
+## 15674. DeFrame: Debiasing Large Language Models Against Framing Effects
+
+- **ID:** `2602.04306` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Kahee Lim, Soyeon Kim, Steven Euijong Whang
+- **abs:** https://arxiv.org/abs/2602.04306 · **PDF:** https://arxiv.org/pdf/2602.04306v1
+
+## 15675. Beyond Static Cropping: Layer-Adaptive Visual Localization and Decoding Enhancement
+
+- **ID:** `2602.04304` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Zipeng Zhu, Zhanghao Hu, Qinglin Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.04304 · **PDF:** https://arxiv.org/pdf/2602.04304v1
+
+## 15676. Revisiting Prompt Sensitivity in Large Language Models for Text Classification: The Role of Prompt Underspecification
+
+- **ID:** `2602.04297` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Branislav Pecher, Michal Spiegel, Robert Belanec, et al.
+- **abs:** https://arxiv.org/abs/2602.04297 · **PDF:** https://arxiv.org/pdf/2602.04297v1
+
+## 15677. ProxyWar: Dynamic Assessment of LLM Code Generation in Game Arenas
+
+- **ID:** `2602.04296` · **primary:** `cs.SE` · **Submitted:** 2026-02-04
+- **Authors:** Wenjun Peng, Xinyu Wang, Qi Wu
+- **abs:** https://arxiv.org/abs/2602.04296 · **PDF:** https://arxiv.org/pdf/2602.04296v1
+
+## 15678. How Few-shot Demonstrations Affect Prompt-based Defenses Against LLM Jailbreak Attacks
+
+- **ID:** `2602.04294` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Yanshu Wang, Shuaishuai Yang, Jingjing He, et al.
+- **abs:** https://arxiv.org/abs/2602.04294 · **PDF:** https://arxiv.org/pdf/2602.04294v1
+
+## 15679. ImpRIF: Stronger Implicit Reasoning Leads to Better Complex Instruction Following
+
+- **ID:** `2602.21228` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Yuancheng Yang, Lin Yang, Xu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.21228 · **PDF:** https://arxiv.org/pdf/2602.21228v1
+
+## 15680. Disentangling Causal Importance from Emergent Structure in Multi-Expert Orchestration
+
+- **ID:** `2602.04291` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Sudipto Ghosh, Sujoy Nath, Sunny Manchanda, et al.
+- **abs:** https://arxiv.org/abs/2602.04291 · **PDF:** https://arxiv.org/pdf/2602.04291v1
+
+## 15681. Budget-Aware Agentic Routing via Boundary-Guided Training
+
+- **ID:** `2602.21227` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Caiqi Zhang, Menglin Xia, Xuchao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.21227 · **PDF:** https://arxiv.org/pdf/2602.21227v1
+
+## 15682. Guided Verifier: Collaborative Multimodal Reasoning via Dynamic Process Supervision
+
+- **ID:** `2602.04290` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Lingzhuang Sun, Ruitong Liu, Yuxia Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.04290 · **PDF:** https://arxiv.org/pdf/2602.04290v1
+
+## 15683. Canzona: A Unified, Asynchronous, and Load-Balanced Framework for Distributed Matrix-based Optimizers
+
+- **ID:** `2602.06079` · **primary:** `cs.DC` · **Submitted:** 2026-02-04
+- **Authors:** Liangyu Wang, Siqi Zhang, Junjie Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.06079 · **PDF:** https://arxiv.org/pdf/2602.06079v1
+
+## 15684. Expected Moral Shortfall for Ethical Competence in Decision-making Models
+
+- **ID:** `2602.13268` · **primary:** `cs.CY` · **Submitted:** 2026-02-04
+- **Authors:** Aisha Aijaz, Raghava Mutharaju, Manohar Kumar
+- **abs:** https://arxiv.org/abs/2602.13268 · **PDF:** https://arxiv.org/pdf/2602.13268v1
+
+## 15685. Proxy Compression for Language Modeling
+
+- **ID:** `2602.04289` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Lin Zheng, Xinyu Li, Qian Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.04289 · **PDF:** https://arxiv.org/pdf/2602.04289v1
+
+## 15686. Contextual Drag: How Errors in the Context Affect LLM Reasoning
+
+- **ID:** `2602.04288` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Yun Cheng, Xingyu Zhu, Haoyu Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.04288 · **PDF:** https://arxiv.org/pdf/2602.04288v2
+
+## 15687. Convolution Operator Network for Forward and Inverse Problems (FI-Conv): Application to Plasma Turbulence Simulations
+
+- **ID:** `2602.04287` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Xingzhuo Chen, Anthony Poole, Ionut-Gabriel Farcas, et al.
+- **abs:** https://arxiv.org/abs/2602.04287 · **PDF:** https://arxiv.org/pdf/2602.04287v1
+
+## 15688. Agent-Omit: Training Efficient LLM Agents for Adaptive Thought and Observation Omission via Agentic Reinforcement Learning
+
+- **ID:** `2602.04284` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Yansong Ning, Jun Fang, Naiqiang Tan, et al.
+- **abs:** https://arxiv.org/abs/2602.04284 · **PDF:** https://arxiv.org/pdf/2602.04284v1
+
+## 15689. ECG-R1: Protocol-Guided and Modality-Agnostic MLLM for Reliable ECG Interpretation
+
+- **ID:** `2602.04279` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Jiarui Jin, Haoyu Wang, Xingliang Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.04279 · **PDF:** https://arxiv.org/pdf/2602.04279v1
+
+## 15690. Multi Objective Design Optimization of Non Pneumatic Passenger Car Tires Using Finite Element Modeling, Machine Learning, and Particle swarm Optimization and Bayesian Optimization Algorithms
+
+- **ID:** `2602.04277` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Priyankkumar Dhrangdhariya, Soumyadipta Maiti, Venkataramana Runkana
+- **abs:** https://arxiv.org/abs/2602.04277 · **PDF:** https://arxiv.org/pdf/2602.04277v1
+
+## 15691. Allocate Marginal Reviews to Borderline Papers Using LLM Comparative Ranking
+
+- **ID:** `2602.06078` · **primary:** `cs.DL` · **Submitted:** 2026-02-04
+- **Authors:** Elliot L. Epstein, Rajat Dwaraknath, John Winnicki, et al.
+- **abs:** https://arxiv.org/abs/2602.06078 · **PDF:** https://arxiv.org/pdf/2602.06078v1
+
+## 15692. CyIN: Cyclic Informative Latent Space for Bridging Complete and Incomplete Multimodal Learning
+
+- **ID:** `2602.04920` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Ronghao Lin, Qiaolin He, Sijie Mai, et al.
+- **abs:** https://arxiv.org/abs/2602.04920 · **PDF:** https://arxiv.org/pdf/2602.04920v1
+
+## 15693. Bures-Wasserstein Importance-Weighted Evidence Lower Bound: Exposition and Applications
+
+- **ID:** `2602.04272` · **primary:** `stat.CO` · **Submitted:** 2026-02-04
+- **Authors:** Peiwen Jiang, Takuo Matsubara, Minh-Ngoc Tran
+- **abs:** https://arxiv.org/abs/2602.04272 · **PDF:** https://arxiv.org/pdf/2602.04272v1
+
+## 15694. SkeletonGaussian: Editable 4D Generation through Gaussian Skeletonization
+
+- **ID:** `2602.04271` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Lifan Wu, Ruijie Zhu, Yubo Ai, et al.
+- **abs:** https://arxiv.org/abs/2602.04271 · **PDF:** https://arxiv.org/pdf/2602.04271v1
+
+## 15695. Multi-Integration of Labels across Categories for Component Identification (MILCCI)
+
+- **ID:** `2602.04270` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Noga Mudrik, Yuxi Chen, Gal Mishne, et al.
+- **abs:** https://arxiv.org/abs/2602.04270 · **PDF:** https://arxiv.org/pdf/2602.04270v1
+
+## 15696. Gradually Compacting Large Language Models for Reasoning Like a Boiling Frog
+
+- **ID:** `2602.04919` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Yiran Zhao, Shengyang Zhou, Zijian Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.04919 · **PDF:** https://arxiv.org/pdf/2602.04919v1
+
+## 15697. Aortic Valve Disease Detection from PPG via Physiology-Informed Self-Supervised Learning
+
+- **ID:** `2602.04266` · **primary:** `eess.SP` · **Submitted:** 2026-02-04
+- **Authors:** Jiaze Wang, Qinghao Zhao, Zizheng Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.04266 · **PDF:** https://arxiv.org/pdf/2602.04266v1
+
+## 15698. Thickening-to-Thinning: Reward Shaping via Human-Inspired Learning Dynamics for LLM Reasoning
+
+- **ID:** `2602.04265` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Wenze Lin, Zhen Yang, Xitai Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.04265 · **PDF:** https://arxiv.org/pdf/2602.04265v2
+
+## 15699. From Dead Neurons to Deep Approximators: Deep Bernstein Networks as a Provable Alternative to Residual Layers
+
+- **ID:** `2602.04264` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Ibrahim Albool, Malak Gamal El-Din, Salma Elmalaki, et al.
+- **abs:** https://arxiv.org/abs/2602.04264 · **PDF:** https://arxiv.org/pdf/2602.04264v1
+
+## 15700. Embracing Anisotropy: Turning Massive Activations into Interpretable Control Knobs for Large Language Models
+
+- **ID:** `2603.00029` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Youngji Roh, Hyunjin Cho, Jaehyung Kim
+- **abs:** https://arxiv.org/abs/2603.00029 · **PDF:** https://arxiv.org/pdf/2603.00029v1
+
+## 15701. AppleVLM: End-to-end Autonomous Driving with Advanced Perception and Planning-Enhanced Vision-Language Models
+
+- **ID:** `2602.04256` · **primary:** `cs.RO` · **Submitted:** 2026-02-04
+- **Authors:** Yuxuan Han, Kunyuan Wu, Qianyi Shao, et al.
+- **abs:** https://arxiv.org/abs/2602.04256 · **PDF:** https://arxiv.org/pdf/2602.04256v1
+
+## 15702. From Ambiguity to Action: A POMDP Perspective on Partial Multi-Label Ambiguity and Its Horizon-One Resolution
+
+- **ID:** `2602.04255` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Hanlin Pan, Yuhao Tang, Wanfu Gao
+- **abs:** https://arxiv.org/abs/2602.04255 · **PDF:** https://arxiv.org/pdf/2602.04255v1
+
+## 15703. Scaling Agentic Verifier for Competitive Coding
+
+- **ID:** `2602.04254` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Zeyao Ma, Jing Zhang, Xiaokang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.04254 · **PDF:** https://arxiv.org/pdf/2602.04254v1
+
+## 15704. ACIL: Active Class Incremental Learning for Image Classification
+
+- **ID:** `2602.04252` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Aditya R. Bhattacharya, Debanjan Goswami, Shayok Chakraborty
+- **abs:** https://arxiv.org/abs/2602.04252 · **PDF:** https://arxiv.org/pdf/2602.04252v1
+
+## 15705. Empirical-MCTS: Continuous Agent Evolution via Dual-Experience Monte Carlo Tree Search
+
+- **ID:** `2602.04248` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Hao Lu, Haoyuan Huang, Yulin Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.04248 · **PDF:** https://arxiv.org/pdf/2602.04248v1
+
+## 15706. DementiaBank-Emotion: A Multi-Rater Emotion Annotation Corpus for Alzheimer's Disease Speech (Version 1.0)
+
+- **ID:** `2602.04247` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Cheonkam Jeong, Jessica Liao, Audrey Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.04247 · **PDF:** https://arxiv.org/pdf/2602.04247v1
+
+## 15707. Simulated Adoption: Decoupling Magnitude and Direction in LLM In-Context Conflict Resolution
+
+- **ID:** `2602.04918` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Long Zhang, Fangwei Lin
+- **abs:** https://arxiv.org/abs/2602.04918 · **PDF:** https://arxiv.org/pdf/2602.04918v2
+
+## 15708. CoLT: Reasoning with Chain of Latent Tool Calls
+
+- **ID:** `2602.04246` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Fangwei Zhu, Zhifang Sui
+- **abs:** https://arxiv.org/abs/2602.04246 · **PDF:** https://arxiv.org/pdf/2602.04246v1
+
+## 15709. Training A Foundation Model to Represent Graphs as Vectors
+
+- **ID:** `2602.04244` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Qi Feng, Jicong Fan
+- **abs:** https://arxiv.org/abs/2602.04244 · **PDF:** https://arxiv.org/pdf/2602.04244v1
+
+## 15710. VISION-ICE: Video-based Interpretation and Spatial Identification of Arrhythmia Origins via Neural Networks in Intracardiac Echocardiography
+
+- **ID:** `2602.20165` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Dorsa EPMoghaddam, Feng Gao, Drew Bernard, et al.
+- **abs:** https://arxiv.org/abs/2602.20165 · **PDF:** https://arxiv.org/pdf/2602.20165v1
+
+## 15711. Windowed SummaryMixing: An Efficient Fine-Tuning of Self-Supervised Learning Models for Low-resource Speech Recognition
+
+- **ID:** `2602.09043` · **primary:** `eess.AS` · **Submitted:** 2026-02-04
+- **Authors:** Aditya Srinivas Menon, Kumud Tripathi, Raj Gohil, et al.
+- **abs:** https://arxiv.org/abs/2602.09043 · **PDF:** https://arxiv.org/pdf/2602.09043v1
+
+## 15712. Tokenization and Morphological Fidelity in Uralic NLP: A Cross-Lingual Evaluation
+
+- **ID:** `2602.04241` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Nuo Xu, Ahrii Kim
+- **abs:** https://arxiv.org/abs/2602.04241 · **PDF:** https://arxiv.org/pdf/2602.04241v2
+
+## 15713. SPOT-Occ: Sparse Prototype-guided Transformer for Camera-based 3D Occupancy Prediction
+
+- **ID:** `2602.04240` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Suzeyu Chen, Leheng Li, Ying-Cong Chen
+- **abs:** https://arxiv.org/abs/2602.04240 · **PDF:** https://arxiv.org/pdf/2602.04240v1
+
+## 15714. Cascading Robustness Verification: Toward Efficient Model-Agnostic Certification
+
+- **ID:** `2602.04236` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Mohammadreza Maleki, Rushendra Sidibomma, Arman Adibi, et al.
+- **abs:** https://arxiv.org/abs/2602.04236 · **PDF:** https://arxiv.org/pdf/2602.04236v1
+
+## 15715. Provable Target Sample Complexity Improvements as Pre-Trained Models Scale
+
+- **ID:** `2602.04233` · **primary:** `stat.ML` · **Submitted:** 2026-02-04
+- **Authors:** Kazuto Fukuchi, Ryuichiro Hataya, Kota Matsui
+- **abs:** https://arxiv.org/abs/2602.04233 · **PDF:** https://arxiv.org/pdf/2602.04233v1
+
+## 15716. Multi-Aspect Mining and Anomaly Detection for Heterogeneous Tensor Streams
+
+- **ID:** `2602.04917` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Soshi Kakio, Yasuko Matsubara, Ren Fujiwara, et al.
+- **abs:** https://arxiv.org/abs/2602.04917 · **PDF:** https://arxiv.org/pdf/2602.04917v1
+
+## 15717. RAPO: Risk-Aware Preference Optimization for Generalizable Safe Reasoning
+
+- **ID:** `2602.04224` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Zeming Wei, Qiaosheng Zhang, Xia Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.04224 · **PDF:** https://arxiv.org/pdf/2602.04224v1
+
+## 15718. AFD-INSTRUCTION: A Comprehensive Antibody Instruction Dataset with Functional Annotations for LLM-Based Understanding and Design
+
+- **ID:** `2602.04916` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-04
+- **Authors:** Ling Luo, Wenbin Jiang, Hongyuan Chang, et al.
+- **abs:** https://arxiv.org/abs/2602.04916 · **PDF:** https://arxiv.org/pdf/2602.04916v2
+
+## 15719. SLAY: Geometry-Aware Spherical Linearized Attention with Yat-Kernel
+
+- **ID:** `2602.04915` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Jose Miguel Luna, Taha Bouhsine, Krzysztof Choromanski
+- **abs:** https://arxiv.org/abs/2602.04915 · **PDF:** https://arxiv.org/pdf/2602.04915v2
+
+## 15720. Frontend Token Enhancement for Token-Based Speech Recognition
+
+- **ID:** `2602.04217` · **primary:** `cs.SD` · **Submitted:** 2026-02-04
+- **Authors:** Takanori Ashihara, Shota Horiguchi, Kohei Matsuura, et al.
+- **abs:** https://arxiv.org/abs/2602.04217 · **PDF:** https://arxiv.org/pdf/2602.04217v1
+
+## 15721. OAT: Ordered Action Tokenization
+
+- **ID:** `2602.04215` · **primary:** `cs.RO` · **Submitted:** 2026-02-04
+- **Authors:** Chaoqi Liu, Xiaoshen Han, Jiawei Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.04215 · **PDF:** https://arxiv.org/pdf/2602.04215v2
+
+## 15722. InterPReT: Interactive Policy Restructuring and Training Enable Effective Imitation Learning from Laypersons
+
+- **ID:** `2602.04213` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Feiyu Gavin Zhu, Jean Oh, Reid Simmons
+- **abs:** https://arxiv.org/abs/2602.04213 · **PDF:** https://arxiv.org/pdf/2602.04213v1
+
+## 15723. Language Models Struggle to Use Representations Learned In-Context
+
+- **ID:** `2602.04212` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Michael A. Lepori, Tal Linzen, Ann Yuan, et al.
+- **abs:** https://arxiv.org/abs/2602.04212 · **PDF:** https://arxiv.org/pdf/2602.04212v1
+
+## 15724. Steering LLMs via Scalable Interactive Oversight
+
+- **ID:** `2602.04210` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Enyu Zhou, Zhiheng Xi, Long Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.04210 · **PDF:** https://arxiv.org/pdf/2602.04210v2
+
+## 15725. SCALE: Self-uncertainty Conditioned Adaptive Looking and Execution for Vision-Language-Action Models
+
+- **ID:** `2602.04208` · **primary:** `cs.RO` · **Submitted:** 2026-02-04
+- **Authors:** Hyeonbeom Choi, Daechul Ahn, Youhan Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.04208 · **PDF:** https://arxiv.org/pdf/2602.04208v1
+
+## 15726. Enforcing Monotonic Progress in Legal Cross-Examination: Preventing Long-Horizon Stagnation in LLM-Based Inquiry
+
+- **ID:** `2602.04206` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Hsien-Jyh Liao
+- **abs:** https://arxiv.org/abs/2602.04206 · **PDF:** https://arxiv.org/pdf/2602.04206v1
+
+## 15727. PipeMFL-240K: A Large-scale Dataset and Benchmark for Object Detection in Pipeline Magnetic Flux Leakage Imaging
+
+- **ID:** `2602.07044` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Tianyi Qu, Songxiao Yang, Haolin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.07044 · **PDF:** https://arxiv.org/pdf/2602.07044v1
+
+## 15728. AGMA: Adaptive Gaussian Mixture Anchors for Prior-Guided Multimodal Human Trajectory Forecasting
+
+- **ID:** `2602.04204` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Chao Li, Rui Zhang, Siyuan Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.04204 · **PDF:** https://arxiv.org/pdf/2602.04204v1
+
+## 15729. EPPCMinerBen: A Novel Benchmark for Evaluating Large Language Models on Electronic Patient-Provider Communication via the Patient Portal
+
+- **ID:** `2603.00028` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Samah Fodeh, Yan Wang, Linhai Ma, et al.
+- **abs:** https://arxiv.org/abs/2603.00028 · **PDF:** https://arxiv.org/pdf/2603.00028v1
+
+## 15730. From Sparse Sensors to Continuous Fields: STRIDE for Spatiotemporal Reconstruction
+
+- **ID:** `2602.04201` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Yanjie Tong, Peng Chen
+- **abs:** https://arxiv.org/abs/2602.04201 · **PDF:** https://arxiv.org/pdf/2602.04201v1
+
+## 15731. From Helpfulness to Toxic Proactivity: Diagnosing Behavioral Misalignment in LLM Agents
+
+- **ID:** `2602.04197` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Xinyue Wang, Yuanhe Zhang, Zhengshuo Gong, et al.
+- **abs:** https://arxiv.org/abs/2602.04197 · **PDF:** https://arxiv.org/pdf/2602.04197v1
+
+## 15732. AM-FM: A Foundation Model for Ambient Intelligence Through WiFi
+
+- **ID:** `2602.11200` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Guozhen Zhu, Yuqian Hu, Sakila Jayaweera, et al.
+- **abs:** https://arxiv.org/abs/2602.11200 · **PDF:** https://arxiv.org/pdf/2602.11200v1
+
+## 15733. The Missing Half: Unveiling Training-time Implicit Safety Risks Beyond Deployment
+
+- **ID:** `2602.04196` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Zhexin Zhang, Yida Lu, Junfeng Fang, et al.
+- **abs:** https://arxiv.org/abs/2602.04196 · **PDF:** https://arxiv.org/pdf/2602.04196v1
+
+## 15734. The Thinking Boundary: Quantifying Reasoning Suitability of Multimodal Tasks via Dual Tuning
+
+- **ID:** `2603.04415` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Ruobing Zheng, Tianqi Li, Jianing Li, et al.
+- **abs:** https://arxiv.org/abs/2603.04415 · **PDF:** https://arxiv.org/pdf/2603.04415v1
+
+## 15735. LORE: Jointly Learning the Intrinsic Dimensionality and Relative Similarity Structure From Ordinal Data
+
+- **ID:** `2602.04192` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Vivek Anand, Alec Helbling, Mark A. Davenport, et al.
+- **abs:** https://arxiv.org/abs/2602.04192 · **PDF:** https://arxiv.org/pdf/2602.04192v2
+
+## 15736. Benchmarking Uncertainty Quantification of Plug-and-Play Diffusion Priors for Inverse Problems Solving
+
+- **ID:** `2602.04189` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Xiaoyu Qiu, Taewon Yang, Zhanhao Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.04189 · **PDF:** https://arxiv.org/pdf/2602.04189v1
+
+## 15737. Natural Language Instructions for Scene-Responsive Human-in-the-Loop Motion Planning in Autonomous Driving using Vision-Language-Action Models
+
+- **ID:** `2602.04184` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Angel Martinez-Sanchez, Parthib Roy, Ross Greer
+- **abs:** https://arxiv.org/abs/2602.04184 · **PDF:** https://arxiv.org/pdf/2602.04184v1
+
+## 15738. HoloEv-Net: Efficient Event-based Action Recognition via Holographic Spatial Embedding and Global Spectral Gating
+
+- **ID:** `2602.04182` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Weidong Hao
+- **abs:** https://arxiv.org/abs/2602.04182 · **PDF:** https://arxiv.org/pdf/2602.04182v1
+
+## 15739. Piece of CAKE: Adaptive Execution Engines via Microsecond-Scale Learning
+
+- **ID:** `2602.04181` · **primary:** `cs.DB` · **Submitted:** 2026-02-04
+- **Authors:** Zijie Zhao, Ryan Marcus
+- **abs:** https://arxiv.org/abs/2602.04181 · **PDF:** https://arxiv.org/pdf/2602.04181v1
+
+## 15740. Bilevel Optimization with Lower-Level Uniform Convexity: Theory and Algorithm
+
+- **ID:** `2603.00027` · **primary:** `math.OC` · **Submitted:** 2026-02-04
+- **Authors:** Yuman Wu, Xiaochuan Gong, Jie Hao, et al.
+- **abs:** https://arxiv.org/abs/2603.00027 · **PDF:** https://arxiv.org/pdf/2603.00027v1
+
+## 15741. Topology-Aware Revival for Efficient Sparse Training
+
+- **ID:** `2602.04166` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Meiling Jin, Fei Wang, Xiaoyun Yuan, et al.
+- **abs:** https://arxiv.org/abs/2602.04166 · **PDF:** https://arxiv.org/pdf/2602.04166v1
+
+## 15742. BPDQ: Bit-Plane Decomposition Quantization on a Variable Grid for Large Language Models
+
+- **ID:** `2602.04163` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Junyu Chen, Jungang Li, Jing Xiong, et al.
+- **abs:** https://arxiv.org/abs/2602.04163 · **PDF:** https://arxiv.org/pdf/2602.04163v1
+
+## 15743. Improving 2D Diffusion Models for 3D Medical Imaging with Inter-Slice Consistent Stochasticity
+
+- **ID:** `2602.04162` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Chenhe Du, Qing Wu, Xuanyu Tian, et al.
+- **abs:** https://arxiv.org/abs/2602.04162 · **PDF:** https://arxiv.org/pdf/2602.04162v2
+
+## 15744. Maximin Relative Improvement: Fair Learning as a Bargaining Problem
+
+- **ID:** `2602.04155` · **primary:** `stat.ML` · **Submitted:** 2026-02-04
+- **Authors:** Jiwoo Han, Moulinath Banerjee, Yuekai Sun
+- **abs:** https://arxiv.org/abs/2602.04155 · **PDF:** https://arxiv.org/pdf/2602.04155v1
+
+## 15745. Pruning for Generalization: A Transfer-Oriented Spatiotemporal Graph Framework
+
+- **ID:** `2602.04153` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Zihao Jing, Yuxi Long, Ganlin Feng
+- **abs:** https://arxiv.org/abs/2602.04153 · **PDF:** https://arxiv.org/pdf/2602.04153v1
+
+## 15746. MA3DSG: Multi-Agent 3D Scene Graph Generation for Large-Scale Indoor Environments
+
+- **ID:** `2602.04152` · **primary:** `cs.RO` · **Submitted:** 2026-02-04
+- **Authors:** Yirum Kim, Jaewoo Kim, Ue-Hwan Kim
+- **abs:** https://arxiv.org/abs/2602.04152 · **PDF:** https://arxiv.org/pdf/2602.04152v1
+
+## 15747. Training Data Efficiency in Multimodal Process Reward Models
+
+- **ID:** `2602.04145` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Jinyuan Li, Chengsong Huang, Langlin Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.04145 · **PDF:** https://arxiv.org/pdf/2602.04145v2
+
+## 15748. OMG-Agent: Toward Robust Missing Modality Generation with Decoupled Coarse-to-Fine Agentic Workflows
+
+- **ID:** `2602.04144` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Ruiting Dai, Zheyu Wang, Haoyu Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.04144 · **PDF:** https://arxiv.org/pdf/2602.04144v1
+
+## 15749. JSynFlow: Japanese Synthesised Flowchart Visual Question Answering Dataset built with Large Language Models
+
+- **ID:** `2602.04142` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Hiroshi Sasaki
+- **abs:** https://arxiv.org/abs/2602.04142 · **PDF:** https://arxiv.org/pdf/2602.04142v2
+
+## 15750. When and What to Ask: AskBench and Rubric-Guided RLVR for LLM Clarification
+
+- **ID:** `2602.11199` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Jiale Zhao, Ke Fang, Lu Cheng
+- **abs:** https://arxiv.org/abs/2602.11199 · **PDF:** https://arxiv.org/pdf/2602.11199v1
+
+## 15751. A$^2$-LLM: An End-to-end Conversational Audio Avatar Large Language Model
+
+- **ID:** `2602.04913` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Xiaolin Hu, Hang Yuan, Xinzhu Sang, et al.
+- **abs:** https://arxiv.org/abs/2602.04913 · **PDF:** https://arxiv.org/pdf/2602.04913v1
+
+## 15752. Generative Neural Operators through Diffusion Last Layer
+
+- **ID:** `2602.04139` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Sungwon Park, Anthony Zhou, Hongjoong Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.04139 · **PDF:** https://arxiv.org/pdf/2602.04139v1
+
+## 15753. VeriInteresting: An Empirical Study of Model Prompt Interactions in Verilog Code Generation
+
+- **ID:** `2603.08715` · **primary:** `cs.AR` · **Submitted:** 2026-02-04
+- **Authors:** Luca Collini, Andrew Hennesee, Patrick Yubeaton, et al.
+- **abs:** https://arxiv.org/abs/2603.08715 · **PDF:** https://arxiv.org/pdf/2603.08715v1
+
+## 15754. Atomic Information Flow: A Network Flow Model for Tool Attributions in RAG Systems
+
+- **ID:** `2602.04912` · **primary:** `cs.IR` · **Submitted:** 2026-02-04
+- **Authors:** James Gao, Josh Zhou, Qi Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.04912 · **PDF:** https://arxiv.org/pdf/2602.04912v1
+
+## 15755. Lyapunov Constrained Soft Actor-Critic (LC-SAC) using Koopman Operator Theory for Quadrotor Trajectory Tracking
+
+- **ID:** `2602.04132` · **primary:** `eess.SY` · **Submitted:** 2026-02-04
+- **Authors:** Dhruv S. Kushwaha, Zoleikha A. Biron
+- **abs:** https://arxiv.org/abs/2602.04132 · **PDF:** https://arxiv.org/pdf/2602.04132v3
+
+## 15756. Directional Concentration Uncertainty: A representational approach to uncertainty quantification for generative models
+
+- **ID:** `2602.13264` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Souradeep Chattopadhyay, Brendan Kennedy, Sai Munikoti, et al.
+- **abs:** https://arxiv.org/abs/2602.13264 · **PDF:** https://arxiv.org/pdf/2602.13264v1
+
+## 15757. Decoupling Time and Risk: Risk-Sensitive Reinforcement Learning with General Discounting
+
+- **ID:** `2602.04131` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Mehrdad Moghimi, Anthony Coache, Hyejin Ku
+- **abs:** https://arxiv.org/abs/2602.04131 · **PDF:** https://arxiv.org/pdf/2602.04131v1
+
+## 15758. KGLAMP: Knowledge Graph-guided Language model for Adaptive Multi-robot Planning and Replanning
+
+- **ID:** `2602.04129` · **primary:** `cs.RO` · **Submitted:** 2026-02-04
+- **Authors:** Chak Lam Shek, Faizan M. Tariq, Sangjae Bae, et al.
+- **abs:** https://arxiv.org/abs/2602.04129 · **PDF:** https://arxiv.org/pdf/2602.04129v1
+
+## 15759. From Lemmas to Dependencies: What Signals Drive Light Verbs Classification?
+
+- **ID:** `2602.04127` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Sercan Karakaş, Yusuf Şimşek
+- **abs:** https://arxiv.org/abs/2602.04127 · **PDF:** https://arxiv.org/pdf/2602.04127v1
+
+## 15760. Attack-Resistant Uniform Fairness for Linear and Smooth Contextual Bandits
+
+- **ID:** `2602.04125` · **primary:** `stat.ML` · **Submitted:** 2026-02-04
+- **Authors:** Qingwen Zhang, Wenjia Wang
+- **abs:** https://arxiv.org/abs/2602.04125 · **PDF:** https://arxiv.org/pdf/2602.04125v1
+
+## 15761. Scalable Explainability-as-a-Service (XaaS) for Edge AI Systems
+
+- **ID:** `2602.04120` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Samaresh Kumar Singh, Joyjit Roy
+- **abs:** https://arxiv.org/abs/2602.04120 · **PDF:** https://arxiv.org/pdf/2602.04120v1
+
+## 15762. Synthesizable Molecular Generation via Soft-constrained GFlowNets with Rich Chemical Priors
+
+- **ID:** `2602.04119` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Hyeonah Kim, Minsu Kim, Celine Roget, et al.
+- **abs:** https://arxiv.org/abs/2602.04119 · **PDF:** https://arxiv.org/pdf/2602.04119v1
+
+## 15763. Learning to Reason in 13 Parameters
+
+- **ID:** `2602.04118` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** John X. Morris, Niloofar Mireshghallah, Mark Ibrahim, et al.
+- **abs:** https://arxiv.org/abs/2602.04118 · **PDF:** https://arxiv.org/pdf/2602.04118v1
+
+## 15764. A logical re-conception of neural networks: Hamiltonian bitwise part-whole architecture
+
+- **ID:** `2602.04911` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** E Bowen, R Granger, A Rodriguez
+- **abs:** https://arxiv.org/abs/2602.04911 · **PDF:** https://arxiv.org/pdf/2602.04911v1
+
+## 15765. Toward Effective Multimodal Graph Foundation Model: A Divide-and-Conquer Based Approach
+
+- **ID:** `2602.04116` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Sicheng Liu, Xunkai Li, Daohan Su, et al.
+- **abs:** https://arxiv.org/abs/2602.04116 · **PDF:** https://arxiv.org/pdf/2602.04116v1
+
+## 15766. Turning mechanistic models into forecasters by using machine learning
+
+- **ID:** `2602.04114` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Amit K. Chakraborty, Hao Wang, Pouria Ramazi
+- **abs:** https://arxiv.org/abs/2602.04114 · **PDF:** https://arxiv.org/pdf/2602.04114v1
+
+## 15767. ZKBoost: Zero-Knowledge Verifiable Training for XGBoost
+
+- **ID:** `2602.04113` · **primary:** `cs.CR` · **Submitted:** 2026-02-04
+- **Authors:** Nikolas Melissaris, Jiayi Xu, Antigoni Polychroniadou, et al.
+- **abs:** https://arxiv.org/abs/2602.04113 · **PDF:** https://arxiv.org/pdf/2602.04113v2
+
+## 15768. DELTA: Deliberative Multi-Agent Reasoning with Reinforcement Learning for Multimodal Psychological Counseling
+
+- **ID:** `2602.04112` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Jiangnan Yang, Junjie Chen, Fei Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.04112 · **PDF:** https://arxiv.org/pdf/2602.04112v1
+
+## 15769. ActMem: Bridging the Gap Between Memory Retrieval and Reasoning in LLM Agents
+
+- **ID:** `2603.00026` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Xiaohui Zhang, Zequn Sun, Chengyuan Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.00026 · **PDF:** https://arxiv.org/pdf/2603.00026v1
+
+## 15770. Rate-Optimal Noise Annealing in Semi-Dual Neural Optimal Transport: Tangential Identifiability, Off-Manifold Ambiguity, and Guaranteed Recovery
+
+- **ID:** `2602.04110` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Raymond Chu, Jaewoong Choi, Dohyun Kwon
+- **abs:** https://arxiv.org/abs/2602.04110 · **PDF:** https://arxiv.org/pdf/2602.04110v1
+
+## 15771. Tinker Tales: Supporting Child-AI Collaboration through Co-Creative Storytelling with Educational Scaffolding
+
+- **ID:** `2602.04109` · **primary:** `cs.HC` · **Submitted:** 2026-02-04
+- **Authors:** Nayoung Choi, Jiseung Hong, Peace Cyebukayire, et al.
+- **abs:** https://arxiv.org/abs/2602.04109 · **PDF:** https://arxiv.org/pdf/2602.04109v1
+
+## 15772. Supervised Learning as Lossy Compression: Characterizing Generalization and Sample Complexity via Finite Blocklength Analysis
+
+- **ID:** `2602.04107` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Kosuke Sugiyama, Masato Uchida
+- **abs:** https://arxiv.org/abs/2602.04107 · **PDF:** https://arxiv.org/pdf/2602.04107v1
+
+## 15773. Expert Selections In MoE Models Reveal (Almost) As Much As Text
+
+- **ID:** `2602.04105` · **primary:** `cs.CL` · **Submitted:** 2026-02-04
+- **Authors:** Amir Nuriyev, Gabriel Kulp
+- **abs:** https://arxiv.org/abs/2602.04105 · **PDF:** https://arxiv.org/pdf/2602.04105v3
+
+## 15774. Learning Where It Matters: Geometric Anchoring for Robust Preference Alignment
+
+- **ID:** `2602.04909` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Youngjae Cho, Jongsuk Kim, Ji-Hoon Kim
+- **abs:** https://arxiv.org/abs/2602.04909 · **PDF:** https://arxiv.org/pdf/2602.04909v2
+
+## 15775. DMS2F-HAD: A Dual-branch Mamba-based Spatial-Spectral Fusion Network for Hyperspectral Anomaly Detection
+
+- **ID:** `2602.04102` · **primary:** `cs.CV` · **Submitted:** 2026-02-04
+- **Authors:** Aayushma Pant, Lakpa Tamang, Tsz-Kwan Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.04102 · **PDF:** https://arxiv.org/pdf/2602.04102v1
+
+## 15776. Interfaze: The Future of AI is built on Task-Specific Small Models
+
+- **ID:** `2602.04101` · **primary:** `cs.AI` · **Submitted:** 2026-02-04
+- **Authors:** Harsha Vardhan Khurdula, Vineet Agarwal, Yoeven D Khemlani
+- **abs:** https://arxiv.org/abs/2602.04101 · **PDF:** https://arxiv.org/pdf/2602.04101v1
+
+## 15777. Rethinking Perplexity: Revealing the Impact of Input Length on Perplexity Evaluation in LLMs
+
+- **ID:** `2602.04099` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Letian Cheng, Junyan Wang, Yan Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.04099 · **PDF:** https://arxiv.org/pdf/2602.04099v1
+
+## 15778. CORE: Context-Robust Remasking for Diffusion Language Models
+
+- **ID:** `2602.04096` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Kevin Zhai, Sabbir Mollah, Zhenyi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.04096 · **PDF:** https://arxiv.org/pdf/2602.04096v3
+
+## 15779. A computational account of dreaming: learning and memory consolidation
+
+- **ID:** `2602.04095` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-04
+- **Authors:** Qi Zhang
+- **abs:** https://arxiv.org/abs/2602.04095 · **PDF:** https://arxiv.org/pdf/2602.04095v1
+
+## 15780. Temporal Pair Consistency for Variance-Reduced Flow Matching
+
+- **ID:** `2602.04908` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Chika Maduabuchi, Jindong Wang
+- **abs:** https://arxiv.org/abs/2602.04908 · **PDF:** https://arxiv.org/pdf/2602.04908v2
+
+## 15781. Federated Concept-Based Models: Interpretable models with distributed supervision
+
+- **ID:** `2602.04093` · **primary:** `cs.LG` · **Submitted:** 2026-02-04
+- **Authors:** Dario Fenoglio, Arianna Casanova, Francesco De Santis, et al.
+- **abs:** https://arxiv.org/abs/2602.04093 · **PDF:** https://arxiv.org/pdf/2602.04093v1
+
+## 15782. Scaling In-Context Online Learning Capability of LLMs via Cross-Episode Meta-RL
+
+- **ID:** `2602.04089` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Xiaofeng Lin, Sirou Zhu, Yilei Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.04089 · **PDF:** https://arxiv.org/pdf/2602.04089v1
+
+## 15783. TAB-PO: Preference Optimization with a Token-Level Adaptive Barrier for Token-Critical Structured Generation
+
+- **ID:** `2603.00025` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Samah Fodeh, Linhai Ma, Ganesh Puthiaraju, et al.
+- **abs:** https://arxiv.org/abs/2603.00025 · **PDF:** https://arxiv.org/pdf/2603.00025v1
+
+## 15784. Physics as the Inductive Bias for Causal Discovery
+
+- **ID:** `2602.04907` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Jianhong Chen, Naichen Shi, Xubo Yue
+- **abs:** https://arxiv.org/abs/2602.04907 · **PDF:** https://arxiv.org/pdf/2602.04907v1
+
+## 15785. BASS: Benchmarking Audio LMs for Musical Structure and Semantic Reasoning
+
+- **ID:** `2602.04085` · **primary:** `cs.SD` · **Submitted:** 2026-02-03
+- **Authors:** Min Jang, Orevaoghene Ahia, Nazif Tamer, et al.
+- **abs:** https://arxiv.org/abs/2602.04085 · **PDF:** https://arxiv.org/pdf/2602.04085v1
+
+## 15786. Structure-Informed Estimation for Pilot-Limited MIMO Channels via Tensor Decomposition
+
+- **ID:** `2602.04083` · **primary:** `eess.SP` · **Submitted:** 2026-02-03
+- **Authors:** Alexandre Barbosa de Lima
+- **abs:** https://arxiv.org/abs/2602.04083 · **PDF:** https://arxiv.org/pdf/2602.04083v2
+
+## 15787. A Probabilistic Framework for Solving High-Frequency Helmholtz Equations via Diffusion Models
+
+- **ID:** `2602.04082` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yicheng Zou, Samuel Lanthaler, Hossein Salahshoor
+- **abs:** https://arxiv.org/abs/2602.04082 · **PDF:** https://arxiv.org/pdf/2602.04082v1
+
+## 15788. Abstraction Induces the Brain Alignment of Language and Speech Models
+
+- **ID:** `2602.04081` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Emily Cheng, Aditya R. Vaidya, Richard Antonello
+- **abs:** https://arxiv.org/abs/2602.04081 · **PDF:** https://arxiv.org/pdf/2602.04081v1
+
+## 15789. Principles of Lipschitz continuity in neural networks
+
+- **ID:** `2602.04078` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Róisín Luo
+- **abs:** https://arxiv.org/abs/2602.04078 · **PDF:** https://arxiv.org/pdf/2602.04078v1
+
+## 15790. Efficient Subgroup Analysis via Optimal Trees with Global Parameter Fusion
+
+- **ID:** `2602.04077` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Zhongming Xie, Joseph Giorgio, Jingshen Wang
+- **abs:** https://arxiv.org/abs/2602.04077 · **PDF:** https://arxiv.org/pdf/2602.04077v1
+
+## 15791. Thermodynamic assessment of machine learning models for solid-state synthesis prediction
+
+- **ID:** `2602.04075` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-03
+- **Authors:** Jane Schlesinger, Simon Hjaltason, Nathan J. Szymanski, et al.
+- **abs:** https://arxiv.org/abs/2602.04075 · **PDF:** https://arxiv.org/pdf/2602.04075v1
+
+## 15792. Stroke Lesions as a Rosetta Stone for Language Model Interpretability
+
+- **ID:** `2602.04074` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Julius Fridriksson, Roger D. Newman-Norlund, Saeed Ahmadi, et al.
+- **abs:** https://arxiv.org/abs/2602.04074 · **PDF:** https://arxiv.org/pdf/2602.04074v1
+
+## 15793. Agentic AI-Empowered Dynamic Survey Framework
+
+- **ID:** `2602.04071` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Furkan Mumcu, Lokman Bekit, Michael J. Jones, et al.
+- **abs:** https://arxiv.org/abs/2602.04071 · **PDF:** https://arxiv.org/pdf/2602.04071v1
+
+## 15794. Multiclass Hate Speech Detection with RoBERTa-OTA: Integrating Transformer Attention and Graph Convolutional Networks
+
+- **ID:** `2603.04414` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Mahmoud Abusaqer, Jamil Saquer
+- **abs:** https://arxiv.org/abs/2603.04414 · **PDF:** https://arxiv.org/pdf/2603.04414v1
+
+## 15795. An Empirical Survey and Benchmark of Learned Distance Indexes for Road Networks
+
+- **ID:** `2602.04068` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Gautam Choudhary, Libin Zhou, Yeasir Rayhan, et al.
+- **abs:** https://arxiv.org/abs/2602.04068 · **PDF:** https://arxiv.org/pdf/2602.04068v1
+
+## 15796. LISA: Laplacian In-context Spectral Analysis
+
+- **ID:** `2602.04906` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Julio Candanedo
+- **abs:** https://arxiv.org/abs/2602.04906 · **PDF:** https://arxiv.org/pdf/2602.04906v1
+
+## 15797. SEIS: Subspace-based Equivariance and Invariance Scores for Neural Representations
+
+- **ID:** `2602.04054` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Huahua Lin, Katayoun Farrahi, Xiaohao Cai
+- **abs:** https://arxiv.org/abs/2602.04054 · **PDF:** https://arxiv.org/pdf/2602.04054v1
+
+## 15798. The Anatomy of the Moltbook Social Graph
+
+- **ID:** `2602.10131` · **primary:** `cs.SI` · **Submitted:** 2026-02-03
+- **Authors:** David Holtz
+- **abs:** https://arxiv.org/abs/2602.10131 · **PDF:** https://arxiv.org/pdf/2602.10131v1
+
+## 15799. Partition Trees: Conditional Density Estimation over General Outcome Spaces
+
+- **ID:** `2602.04042` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Felipe Angelim, Alessandro Leite
+- **abs:** https://arxiv.org/abs/2602.04042 · **PDF:** https://arxiv.org/pdf/2602.04042v1
+
+## 15800. OMNI-Dent: Towards an Accessible and Explainable AI Framework for Automated Dental Diagnosis
+
+- **ID:** `2602.07041` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Leeje Jang, Yao-Yi Chiang, Angela M. Hastings, et al.
+- **abs:** https://arxiv.org/abs/2602.07041 · **PDF:** https://arxiv.org/pdf/2602.07041v1
+
+## 15801. DADP: Domain Adaptive Diffusion Policy
+
+- **ID:** `2602.04037` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Pengcheng Wang, Qinghang Liu, Haotian Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.04037 · **PDF:** https://arxiv.org/pdf/2602.04037v2
+
+## 15802. DCER: Dual-Stage Compression and Energy-Based Reconstruction
+
+- **ID:** `2602.04904` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yiwen Wang, Jiahao Qin
+- **abs:** https://arxiv.org/abs/2602.04904 · **PDF:** https://arxiv.org/pdf/2602.04904v1
+
+## 15803. On the Credibility of Evaluating LLMs using Survey Questions
+
+- **ID:** `2602.04033` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Jindřich Libovický
+- **abs:** https://arxiv.org/abs/2602.04033 · **PDF:** https://arxiv.org/pdf/2602.04033v1
+
+## 15804. Mind the Performance Gap: Capability-Behavior Trade-offs in Feature Steering
+
+- **ID:** `2602.04903` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Eitan Sprejer, Oscar Agustín Stanchi, María Victoria Carro, et al.
+- **abs:** https://arxiv.org/abs/2602.04903 · **PDF:** https://arxiv.org/pdf/2602.04903v1
+
+## 15805. The Illusion of Generalization: Re-examining Tabular Language Model Evaluation
+
+- **ID:** `2602.04031` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Aditya Gorla, Ratish Puduppully
+- **abs:** https://arxiv.org/abs/2602.04031 · **PDF:** https://arxiv.org/pdf/2602.04031v1
+
+## 15806. Multimodal Consistency-Guided Reference-Free Data Selection for ASR Accent Adaptation
+
+- **ID:** `2602.13263` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Ligong Lei, Wenwen Lu, Xudong Pang, et al.
+- **abs:** https://arxiv.org/abs/2602.13263 · **PDF:** https://arxiv.org/pdf/2602.13263v1
+
+## 15807. PluRel: Synthetic Data unlocks Scaling Laws for Relational Foundation Models
+
+- **ID:** `2602.04029` · **primary:** `cs.DB` · **Submitted:** 2026-02-03
+- **Authors:** Vignesh Kothapalli, Rishabh Ranjan, Valter Hudovernik, et al.
+- **abs:** https://arxiv.org/abs/2602.04029 · **PDF:** https://arxiv.org/pdf/2602.04029v1
+
+## 15808. Axiomatic Foundations of Counterfactual Explanations
+
+- **ID:** `2602.04028` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Leila Amgoud, Martin Cooper
+- **abs:** https://arxiv.org/abs/2602.04028 · **PDF:** https://arxiv.org/pdf/2602.04028v1
+
+## 15809. A Consensus-Bayesian Framework for Detecting Malicious Activity in Enterprise Directory Access Graphs
+
+- **ID:** `2602.04027` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Pratyush Uppuluri, Shilpa Noushad, Sajan Kumar
+- **abs:** https://arxiv.org/abs/2602.04027 · **PDF:** https://arxiv.org/pdf/2602.04027v2
+
+## 15810. Momentum Attention: The Physics of In-Context Learning and Spectral Forensics for Mechanistic Interpretability
+
+- **ID:** `2602.04902` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Kingsuk Maitra
+- **abs:** https://arxiv.org/abs/2602.04902 · **PDF:** https://arxiv.org/pdf/2602.04902v2
+
+## 15811. Verify as You Go: An LLM-Powered Browser Extension for Fake News Detection
+
+- **ID:** `2603.05519` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Dorsaf Sallami, Esma Aïmeur
+- **abs:** https://arxiv.org/abs/2603.05519 · **PDF:** https://arxiv.org/pdf/2603.05519v1
+
+## 15812. Group Contrastive Learning for Weakly Paired Multimodal Data
+
+- **ID:** `2602.04021` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Aditya Gorla, Hugues Van Assel, Jan-Christian Huetter, et al.
+- **abs:** https://arxiv.org/abs/2602.04021 · **PDF:** https://arxiv.org/pdf/2602.04021v1
+
+## 15813. Understanding and Guiding Layer Placement in Parameter-Efficient Fine-Tuning of Large Language Models
+
+- **ID:** `2602.04019` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yichen Xu, Yuyang Liang, Shan Dai, et al.
+- **abs:** https://arxiv.org/abs/2602.04019 · **PDF:** https://arxiv.org/pdf/2602.04019v2
+
+## 15814. Chaplains' Reflections on the Design and Usage of AI for Conversational Care
+
+- **ID:** `2602.04017` · **primary:** `cs.HC` · **Submitted:** 2026-02-03
+- **Authors:** Joel Wester, Samuel Rhys Cox, Henning Pohl, et al.
+- **abs:** https://arxiv.org/abs/2602.04017 · **PDF:** https://arxiv.org/pdf/2602.04017v1
+
+## 15815. A Multi-Modal Foundational Model for Wireless Communication and Sensing
+
+- **ID:** `2602.04016` · **primary:** `eess.SP` · **Submitted:** 2026-02-03
+- **Authors:** Vahid Yazdnian, Yasaman Ghasempour
+- **abs:** https://arxiv.org/abs/2602.04016 · **PDF:** https://arxiv.org/pdf/2602.04016v1
+
+## 15816. PromptSplit: Revealing Prompt-Level Disagreement in Generative Models
+
+- **ID:** `2602.04009` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Mehdi Lotfian, Mohammad Jalali, Farzan Farnia
+- **abs:** https://arxiv.org/abs/2602.04009 · **PDF:** https://arxiv.org/pdf/2602.04009v2
+
+## 15817. Rational ANOVA Networks
+
+- **ID:** `2602.04006` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Jusheng Zhang, Ningyuan Liu, Qinhan Lyu, et al.
+- **abs:** https://arxiv.org/abs/2602.04006 · **PDF:** https://arxiv.org/pdf/2602.04006v1
+
+## 15818. When AI Persuades: Adversarial Explanation Attacks on Human Trust in AI-Assisted Decision Making
+
+- **ID:** `2602.04003` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Shutong Fan, Lan Zhang, Xiaoyong Yuan
+- **abs:** https://arxiv.org/abs/2602.04003 · **PDF:** https://arxiv.org/pdf/2602.04003v2
+
+## 15819. Functional Stochastic Localization
+
+- **ID:** `2602.03999` · **primary:** `math.PR` · **Submitted:** 2026-02-03
+- **Authors:** Anming Gu, Bobby Shi, Kevin Tian
+- **abs:** https://arxiv.org/abs/2602.03999 · **PDF:** https://arxiv.org/pdf/2602.03999v2
+
+## 15820. Bypassing the Rationale: Causal Auditing of Implicit Reasoning in Language Models
+
+- **ID:** `2602.03994` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Anish Sathyanarayanan, Aditya Nagarsekar, Aarush Rathore
+- **abs:** https://arxiv.org/abs/2602.03994 · **PDF:** https://arxiv.org/pdf/2602.03994v2
+
+## 15821. eCP: Informative uncertainty quantification via Equivariantized Conformal Prediction with pre-trained models
+
+- **ID:** `2602.03986` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Nikolaos Bousias, Lars Lindemann, George Pappas
+- **abs:** https://arxiv.org/abs/2602.03986 · **PDF:** https://arxiv.org/pdf/2602.03986v1
+
+## 15822. DeXposure-FM: A Time-series, Graph Foundation Model for Credit Exposures and Stability on Decentralized Financial Networks
+
+- **ID:** `2602.03981` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Aijie Shu, Wenbin Wu, Gbenga Ibikunle, et al.
+- **abs:** https://arxiv.org/abs/2602.03981 · **PDF:** https://arxiv.org/pdf/2602.03981v1
+
+## 15823. Transformers perform adaptive partial pooling
+
+- **ID:** `2602.03980` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Vsevolod Kapatsinski
+- **abs:** https://arxiv.org/abs/2602.03980 · **PDF:** https://arxiv.org/pdf/2602.03980v1
+
+## 15824. Likelihood-Based Reward Designs for General LLM Reasoning
+
+- **ID:** `2602.03979` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Ariel Kwiatkowski, Natasha Butt, Ismail Labiad, et al.
+- **abs:** https://arxiv.org/abs/2602.03979 · **PDF:** https://arxiv.org/pdf/2602.03979v1
+
+## 15825. Monitorability as a Free Gift: How RLVR Spontaneously Aligns Reasoning
+
+- **ID:** `2602.03978` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Zidi Xiong, Shan Chen, Himabindu Lakkaraju
+- **abs:** https://arxiv.org/abs/2602.03978 · **PDF:** https://arxiv.org/pdf/2602.03978v1
+
+## 15826. Adaptive Test-Time Compute Allocation via Learned Heuristics over Categorical Structure
+
+- **ID:** `2602.03975` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Shuhui Qu
+- **abs:** https://arxiv.org/abs/2602.03975 · **PDF:** https://arxiv.org/pdf/2602.03975v1
+
+## 15827. Active Epistemic Control for Query-Efficient Verified Planning
+
+- **ID:** `2602.03974` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Shuhui Qu
+- **abs:** https://arxiv.org/abs/2602.03974 · **PDF:** https://arxiv.org/pdf/2602.03974v1
+
+## 15828. Fixed Budget is No Harder Than Fixed Confidence in Best-Arm Identification up to Logarithmic Factors
+
+- **ID:** `2602.03972` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Kapilan Balagopalan, Yinan Li, Yao Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.03972 · **PDF:** https://arxiv.org/pdf/2602.03972v2
+
+## 15829. Statistical Guarantees for Reasoning Probes on Looped Boolean Circuits
+
+- **ID:** `2602.03970` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Anastasis Kratsios, Giulia Livieri, A. Martina Neuman
+- **abs:** https://arxiv.org/abs/2602.03970 · **PDF:** https://arxiv.org/pdf/2602.03970v2
+
+## 15830. Structural shifts in institutional participation and collaboration within the AI arXiv preprint research ecosystem
+
+- **ID:** `2602.03969` · **primary:** `cs.SI` · **Submitted:** 2026-02-03
+- **Authors:** Shama Maganur, Mayank Kejriwal
+- **abs:** https://arxiv.org/abs/2602.03969 · **PDF:** https://arxiv.org/pdf/2602.03969v2
+
+## 15831. Non-linear PCA via Evolution Strategies: a Novel Objective Function
+
+- **ID:** `2602.03967` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Thomas Uriot, Elise Chung
+- **abs:** https://arxiv.org/abs/2602.03967 · **PDF:** https://arxiv.org/pdf/2602.03967v1
+
+## 15832. Automatic Classification of Pedagogical Materials against CS Curriculum Guidelines
+
+- **ID:** `2602.03962` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Erik Saule, Kalpathi Subramanian, Razvan Bunescu
+- **abs:** https://arxiv.org/abs/2602.03962 · **PDF:** https://arxiv.org/pdf/2602.03962v1
+
+## 15833. Contextuality from Single-State Ontological Models: An Information-Theoretic No-Go Theorem
+
+- **ID:** `2602.16716` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Song-Ju Kim
+- **abs:** https://arxiv.org/abs/2602.16716 · **PDF:** https://arxiv.org/pdf/2602.16716v2
+
+## 15834. Child Mortality Prediction in Bangladesh: A Decade-Long Validation Study
+
+- **ID:** `2602.03957` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Md Muhtasim Munif Fahim, Md Rezaul Karim
+- **abs:** https://arxiv.org/abs/2602.03957 · **PDF:** https://arxiv.org/pdf/2602.03957v1
+
+## 15835. AgentArk: Distilling Multi-Agent Intelligence into a Single LLM Agent
+
+- **ID:** `2602.03955` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Yinyi Luo, Yiqiao Jin, Weichen Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.03955 · **PDF:** https://arxiv.org/pdf/2602.03955v1
+
+## 15836. Learning Multi-type heterogeneous interacting particle systems
+
+- **ID:** `2602.03954` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Quanjun Lang, Xiong Wang, Fei Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.03954 · **PDF:** https://arxiv.org/pdf/2602.03954v1
+
+## 15837. Representation Geometry as a Diagnostic for Out-of-Distribution Robustness
+
+- **ID:** `2602.03951` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Ali Zia, Farid Hazratian
+- **abs:** https://arxiv.org/abs/2602.03951 · **PDF:** https://arxiv.org/pdf/2602.03951v2
+
+## 15838. Enhancing Mathematical Problem Solving in LLMs through Execution-Driven Reasoning Augmentation
+
+- **ID:** `2602.03950` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Aditya Basarkar, Benyamin Tabarsi, Tiffany Barnes, et al.
+- **abs:** https://arxiv.org/abs/2602.03950 · **PDF:** https://arxiv.org/pdf/2602.03950v2
+
+## 15839. Semantic Rate Distortion and Posterior Design: Compute Constraints, Multimodality, and Strategic Inference
+
+- **ID:** `2602.03949` · **primary:** `cs.IT` · **Submitted:** 2026-02-03
+- **Authors:** Emrah Akyol
+- **abs:** https://arxiv.org/abs/2602.03949 · **PDF:** https://arxiv.org/pdf/2602.03949v1
+
+## 15840. Privacy utility trade offs for parameter estimation in degree heterogeneous higher order networks
+
+- **ID:** `2602.03948` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Bibhabasu Mandal, Sagnik Nandy
+- **abs:** https://arxiv.org/abs/2602.03948 · **PDF:** https://arxiv.org/pdf/2602.03948v1
+
+## 15841. Grables: Tabular Learning Beyond Independent Rows
+
+- **ID:** `2602.03945` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Tamara Cucumides, Floris Geerts
+- **abs:** https://arxiv.org/abs/2602.03945 · **PDF:** https://arxiv.org/pdf/2602.03945v1
+
+## 15842. Linguistic Blind Spots in Clinical Decision Extraction
+
+- **ID:** `2602.03942` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Mohamed Elgaar, Hadi Amiri
+- **abs:** https://arxiv.org/abs/2602.03942 · **PDF:** https://arxiv.org/pdf/2602.03942v1
+
+## 15843. Personalization Increases Affective Alignment but Has Role-Dependent Effects on Epistemic Independence in LLMs
+
+- **ID:** `2603.00024` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Sean W. Kelley, Christoph Riedl
+- **abs:** https://arxiv.org/abs/2603.00024 · **PDF:** https://arxiv.org/pdf/2603.00024v1
+
+## 15844. Autonomous AI Agents for Real-Time Affordable Housing Site Selection: Multi-Objective Reinforcement Learning Under Regulatory Constraints
+
+- **ID:** `2602.03940` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Olaf Yunus Laitinen Imanov, Duygu Erisken, Derya Umut Kulali, et al.
+- **abs:** https://arxiv.org/abs/2602.03940 · **PDF:** https://arxiv.org/pdf/2602.03940v1
+
+## 15845. Aster: Autonomous Scientific Discovery over 20x Faster Than Existing Methods
+
+- **ID:** `2602.07040` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Emmett Bicker
+- **abs:** https://arxiv.org/abs/2602.07040 · **PDF:** https://arxiv.org/pdf/2602.07040v1
+
+## 15846. C-IDS: Solving Contextual POMDP via Information-Directed Objective
+
+- **ID:** `2602.03939` · **primary:** `eess.SY` · **Submitted:** 2026-02-03
+- **Authors:** Chongyang Shi, Michael Dorothy, Jie Fu
+- **abs:** https://arxiv.org/abs/2602.03939 · **PDF:** https://arxiv.org/pdf/2602.03939v1
+
+## 15847. First-Principles AI finds crystallization of fractional quantum Hall liquids
+
+- **ID:** `2602.03927` · **primary:** `cond-mat.mes-hall` · **Submitted:** 2026-02-03
+- **Authors:** Ahmed Abouelkomsan, Liang Fu
+- **abs:** https://arxiv.org/abs/2602.03927 · **PDF:** https://arxiv.org/pdf/2602.03927v1
+
+## 15848. PLATE: Plasticity-Tunable Efficient Adapters for Geometry-Aware Continual Learning
+
+- **ID:** `2602.03846` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Romain Cosentino
+- **abs:** https://arxiv.org/abs/2602.03846 · **PDF:** https://arxiv.org/pdf/2602.03846v1
+
+## 15849. Parallel-Probe: Towards Efficient Parallel Thinking via 2D Probing
+
+- **ID:** `2602.03845` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Tong Zheng, Chengsong Huang, Runpeng Dai, et al.
+- **abs:** https://arxiv.org/abs/2602.03845 · **PDF:** https://arxiv.org/pdf/2602.03845v2
+
+## 15850. WIND: Weather Inverse Diffusion for Zero-Shot Atmospheric Modeling
+
+- **ID:** `2602.03924` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Michael Aich, Andreas Fürst, Florian Sestak, et al.
+- **abs:** https://arxiv.org/abs/2602.03924 · **PDF:** https://arxiv.org/pdf/2602.03924v1
+
+## 15851. Investigating Quantum Circuit Designs Using Neuro-Evolution
+
+- **ID:** `2602.03840` · **primary:** `cs.NE` · **Submitted:** 2026-02-03
+- **Authors:** Devroop Kar, Daniel Krutz, Travis Desell
+- **abs:** https://arxiv.org/abs/2602.03840 · **PDF:** https://arxiv.org/pdf/2602.03840v1
+
+## 15852. Understanding and Exploiting Weight Update Sparsity for Communication-Efficient Distributed RL
+
+- **ID:** `2602.03839` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Erfan Miahi, Eugene Belilovsky
+- **abs:** https://arxiv.org/abs/2602.03839 · **PDF:** https://arxiv.org/pdf/2602.03839v1
+
+## 15853. PrevizWhiz: Combining Rough 3D Scenes and 2D Video to Guide Generative Video Previsualization
+
+- **ID:** `2602.03838` · **primary:** `cs.HC` · **Submitted:** 2026-02-03
+- **Authors:** Erzhen Hu, Frederik Brudy, David Ledo, et al.
+- **abs:** https://arxiv.org/abs/2602.03838 · **PDF:** https://arxiv.org/pdf/2602.03838v1
+
+## 15854. Accelerating Scientific Research with Gemini: Case Studies and Common Techniques
+
+- **ID:** `2602.03837` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** David P. Woodruff, Vincent Cohen-Addad, Lalit Jain, et al.
+- **abs:** https://arxiv.org/abs/2602.03837 · **PDF:** https://arxiv.org/pdf/2602.03837v3
+
+## 15855. Online Vector Quantized Attention
+
+- **ID:** `2602.03922` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Nick Alonso, Tomas Figliolia, Beren Millidge
+- **abs:** https://arxiv.org/abs/2602.03922 · **PDF:** https://arxiv.org/pdf/2602.03922v2
+
+## 15856. AutoFigure: Generating and Refining Publication-Ready Scientific Illustrations
+
+- **ID:** `2602.03828` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Minjun Zhu, Zhen Lin, Yixuan Weng, et al.
+- **abs:** https://arxiv.org/abs/2602.03828 · **PDF:** https://arxiv.org/pdf/2602.03828v2
+
+## 15857. SpecMD: A Comprehensive Study On Speculative Expert Prefetching
+
+- **ID:** `2602.03921` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Duc Hoang, Ajay Jaiswal, Mohammad Samragh, et al.
+- **abs:** https://arxiv.org/abs/2602.03921 · **PDF:** https://arxiv.org/pdf/2602.03921v1
+
+## 15858. Robust Intervention Learning from Emergency Stop Interventions
+
+- **ID:** `2602.03825` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Ethan Pronovost, Khimya Khetarpal, Siddhartha Srinivasa
+- **abs:** https://arxiv.org/abs/2602.03825 · **PDF:** https://arxiv.org/pdf/2602.03825v1
+
+## 15859. Preference-based Conditional Treatment Effects and Policy Learning
+
+- **ID:** `2602.03823` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Dovid Parnas, Mathieu Even, Julie Josse, et al.
+- **abs:** https://arxiv.org/abs/2602.03823 · **PDF:** https://arxiv.org/pdf/2602.03823v1
+
+## 15860. They Said Memes Were Harmless-We Found the Ones That Hurt: Decoding Jokes, Symbols, and Cultural References
+
+- **ID:** `2602.03822` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Sahil Tripathi, Gautam Siddharth Kashyap, Mehwish Nasim, et al.
+- **abs:** https://arxiv.org/abs/2602.03822 · **PDF:** https://arxiv.org/pdf/2602.03822v1
+
+## 15861. Adaptive Evidence Weighting for Audio-Spatiotemporal Fusion
+
+- **ID:** `2602.03817` · **primary:** `cs.SD` · **Submitted:** 2026-02-03
+- **Authors:** Oscar Ovanger, Levi Harris, Timothy H. Keitt
+- **abs:** https://arxiv.org/abs/2602.03817 · **PDF:** https://arxiv.org/pdf/2602.03817v1
+
+## 15862. Riemannian Dueling Optimization
+
+- **ID:** `2603.00023` · **primary:** `math.OC` · **Submitted:** 2026-02-03
+- **Authors:** Yuxuan Ren, Abhishek Roy, Shiqian Ma
+- **abs:** https://arxiv.org/abs/2603.00023 · **PDF:** https://arxiv.org/pdf/2603.00023v1
+
+## 15863. SymPlex: A Structure-Aware Transformer for Symbolic PDE Solving
+
+- **ID:** `2602.03816` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yesom Park, Annie C. Lu, Shao-Ching Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.03816 · **PDF:** https://arxiv.org/pdf/2602.03816v1
+
+## 15864. Fast-Slow Efficient Training for Multimodal Large Language Models via Visual Token Pruning
+
+- **ID:** `2602.03815` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Dingkun Zhang, Shuhan Qi, Yulin Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.03815 · **PDF:** https://arxiv.org/pdf/2602.03815v1
+
+## 15865. Conformal Thinking: Risk Control for Reasoning on a Compute Budget
+
+- **ID:** `2602.03814` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Xi Wang, Anushri Suresh, Alvin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03814 · **PDF:** https://arxiv.org/pdf/2602.03814v1
+
+## 15866. Antidistillation Fingerprinting
+
+- **ID:** `2602.03812` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yixuan Even Xu, John Kirchenbauer, Yash Savani, et al.
+- **abs:** https://arxiv.org/abs/2602.03812 · **PDF:** https://arxiv.org/pdf/2602.03812v1
+
+## 15867. Enhancing Imbalanced Node Classification via Curriculum-Guided Feature Learning and Three-Stage Attention Network
+
+- **ID:** `2602.03808` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Abdul Joseph Fofanah, Lian Wen, David Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.03808 · **PDF:** https://arxiv.org/pdf/2602.03808v1
+
+## 15868. Bridging Online and Offline RL: Contextual Bandit Learning for Multi-Turn Code Generation
+
+- **ID:** `2602.03806` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Ziru Chen, Dongdong Chen, Ruinan Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.03806 · **PDF:** https://arxiv.org/pdf/2602.03806v1
+
+## 15869. Prediction of Critical Heat Flux in Rod Bundles Using Tube-Based Hybrid Machine Learning Models in CTF
+
+- **ID:** `2602.03805` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Aidan Furlong, Robert Salko, Xingang Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.03805 · **PDF:** https://arxiv.org/pdf/2602.03805v1
+
+## 15870. Simulating Meaning, Nevermore! Introducing ICR: A Semiotic-Hermeneutic Metric for Evaluating Meaning in LLM Text Summaries
+
+- **ID:** `2603.04413` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Natalie Perez, Sreyoshi Bhaduri, Aman Chadha
+- **abs:** https://arxiv.org/abs/2603.04413 · **PDF:** https://arxiv.org/pdf/2603.04413v2
+
+## 15871. Do We Need Asynchronous SGD? On the Near-Optimality of Synchronous Methods
+
+- **ID:** `2602.03802` · **primary:** `cs.DC` · **Submitted:** 2026-02-03
+- **Authors:** Grigory Begunov, Alexander Tyurin
+- **abs:** https://arxiv.org/abs/2602.03802 · **PDF:** https://arxiv.org/pdf/2602.03802v1
+
+## 15872. Conformal Reachability for Safe Control in Unknown Environments
+
+- **ID:** `2602.03799` · **primary:** `cs.RO` · **Submitted:** 2026-02-03
+- **Authors:** Xinhang Ma, Junlin Wu, Yiannis Kantaros, et al.
+- **abs:** https://arxiv.org/abs/2602.03799 · **PDF:** https://arxiv.org/pdf/2602.03799v1
+
+## 15873. FullStack-Agent: Enhancing Agentic Full-Stack Web Coding via Development-Oriented Testing and Repository Back-Translation
+
+- **ID:** `2602.03798` · **primary:** `cs.SE` · **Submitted:** 2026-02-03
+- **Authors:** Zimu Lu, Houxing Ren, Yunqiao Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.03798 · **PDF:** https://arxiv.org/pdf/2602.03798v1
+
+## 15874. Manifold Random Features
+
+- **ID:** `2602.03797` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Ananya Parashar, Derek Long, Dwaipayan Saha, et al.
+- **abs:** https://arxiv.org/abs/2602.03797 · **PDF:** https://arxiv.org/pdf/2602.03797v1
+
+## 15875. Understanding Agent Scaling in LLM-Based Multi-Agent Systems via Diversity
+
+- **ID:** `2602.03794` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Yingxuan Yang, Chengrui Qu, Muning Wen, et al.
+- **abs:** https://arxiv.org/abs/2602.03794 · **PDF:** https://arxiv.org/pdf/2602.03794v1
+
+## 15876. WebSentinel: Detecting and Localizing Prompt Injection Attacks for Web Agents
+
+- **ID:** `2602.03792` · **primary:** `cs.CR` · **Submitted:** 2026-02-03
+- **Authors:** Xilong Wang, Yinuo Liu, Zhun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.03792 · **PDF:** https://arxiv.org/pdf/2602.03792v1
+
+## 15877. Should I use Synthetic Data for That? An Analysis of the Suitability of Synthetic Data for Data Sharing and Augmentation
+
+- **ID:** `2602.03791` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Bogdan Kulynych, Theresa Stadler, Jean Louis Raisaro, et al.
+- **abs:** https://arxiv.org/abs/2602.03791 · **PDF:** https://arxiv.org/pdf/2602.03791v1
+
+## 15878. SpatiaLab: Can Vision-Language Models Perform Spatial Reasoning in the Wild?
+
+- **ID:** `2602.03916` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Azmine Toushik Wasi, Wahid Faisal, Abdur Rahman, et al.
+- **abs:** https://arxiv.org/abs/2602.03916 · **PDF:** https://arxiv.org/pdf/2602.03916v2
+
+## 15879. Fast Sampling for Flows and Diffusions with Lazy and Point Mass Stochastic Interpolants
+
+- **ID:** `2602.03789` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Gabriel Damsholt, Jes Frellsen, Susanne Ditlevsen
+- **abs:** https://arxiv.org/abs/2602.03789 · **PDF:** https://arxiv.org/pdf/2602.03789v1
+
+## 15880. Inference-time Unlearning Using Conformal Prediction
+
+- **ID:** `2602.03787` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Somnath Basu Roy Chowdhury, Rahul Kidambi, Avinava Dubey, et al.
+- **abs:** https://arxiv.org/abs/2602.03787 · **PDF:** https://arxiv.org/pdf/2602.03787v1
+
+## 15881. Universality of shallow and deep neural networks on non-Euclidean spaces
+
+- **ID:** `2602.23381` · **primary:** `math.GN` · **Submitted:** 2026-02-03
+- **Authors:** Vugar Ismailov
+- **abs:** https://arxiv.org/abs/2602.23381 · **PDF:** https://arxiv.org/pdf/2602.23381v2
+
+## 15882. AOrchestra: Automating Sub-Agent Creation for Agentic Orchestration
+
+- **ID:** `2602.03786` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Jianhao Ruan, Zhihao Xu, Yiran Peng, et al.
+- **abs:** https://arxiv.org/abs/2602.03786 · **PDF:** https://arxiv.org/pdf/2602.03786v2
+
+## 15883. Context Compression via Explicit Information Transmission
+
+- **ID:** `2602.03784` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Jiangnan Ye, Hanqi Yan, Zhenyi Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.03784 · **PDF:** https://arxiv.org/pdf/2602.03784v1
+
+## 15884. Efficient Estimation of Kernel Surrogate Models for Task Attribution
+
+- **ID:** `2602.03783` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Zhenshuo Zhang, Minxuan Duan, Hongyang R. Zhang
+- **abs:** https://arxiv.org/abs/2602.03783 · **PDF:** https://arxiv.org/pdf/2602.03783v1
+
+## 15885. Reward Redistribution for CVaR MDPs using a Bellman Operator on L-infinity
+
+- **ID:** `2602.03778` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Aneri Muni, Vincent Taboga, Esther Derman, et al.
+- **abs:** https://arxiv.org/abs/2602.03778 · **PDF:** https://arxiv.org/pdf/2602.03778v1
+
+## 15886. DiffLOB: Diffusion Models for Counterfactual Generation in Limit Order Books
+
+- **ID:** `2602.03776` · **primary:** `q-fin.CP` · **Submitted:** 2026-02-03
+- **Authors:** Zhuohan Wang, Carmine Ventre
+- **abs:** https://arxiv.org/abs/2602.03776 · **PDF:** https://arxiv.org/pdf/2602.03776v1
+
+## 15887. An Empirical Study of Collective Behaviors and Social Dynamics in Large Language Model Agents
+
+- **ID:** `2602.03775` · **primary:** `cs.SI` · **Submitted:** 2026-02-03
+- **Authors:** Farnoosh Hashemi, Michael W. Macy
+- **abs:** https://arxiv.org/abs/2602.03775 · **PDF:** https://arxiv.org/pdf/2602.03775v2
+
+## 15888. Reasoning Cache: Continual Improvement Over Long Horizons via Short-Horizon RL
+
+- **ID:** `2602.03773` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Ian Wu, Yuxiao Qu, Amrith Setlur, et al.
+- **abs:** https://arxiv.org/abs/2602.03773 · **PDF:** https://arxiv.org/pdf/2602.03773v2
+
+## 15889. UniGeM: Unifying Data Mixing and Selection via Geometric Exploration and Mining
+
+- **ID:** `2602.03772` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Changhao Wang, Yunfei Yu, Xinhao Yao, et al.
+- **abs:** https://arxiv.org/abs/2602.03772 · **PDF:** https://arxiv.org/pdf/2602.03772v1
+
+## 15890. Reasoning with Latent Tokens in Diffusion Language Models
+
+- **ID:** `2602.03769` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Andre He, Sean Welleck, Daniel Fried
+- **abs:** https://arxiv.org/abs/2602.03769 · **PDF:** https://arxiv.org/pdf/2602.03769v1
+
+## 15891. Decision-oriented benchmarking to transform AI weather forecast access: Application to the Indian monsoon
+
+- **ID:** `2602.03767` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Rajat Masiwal, Colin Aitken, Adam Marchakitus, et al.
+- **abs:** https://arxiv.org/abs/2602.03767 · **PDF:** https://arxiv.org/pdf/2602.03767v1
+
+## 15892. FOVI: A biologically-inspired foveated interface for deep vision models
+
+- **ID:** `2602.03766` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Nicholas M. Blauch, George A. Alvarez, Talia Konkle
+- **abs:** https://arxiv.org/abs/2602.03766 · **PDF:** https://arxiv.org/pdf/2602.03766v1
+
+## 15893. Conditional Flow Matching for Visually-Guided Acoustic Highlighting
+
+- **ID:** `2602.03762` · **primary:** `eess.AS` · **Submitted:** 2026-02-03
+- **Authors:** Hugo Malard, Gael Le Lan, Daniel Wong, et al.
+- **abs:** https://arxiv.org/abs/2602.03762 · **PDF:** https://arxiv.org/pdf/2602.03762v2
+
+## 15894. Zero-shot large vision-language model prompting for automated bone identification in paleoradiology x-ray archives
+
+- **ID:** `2602.03750` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Owen Dong, Lily Gao, Manish Kota, et al.
+- **abs:** https://arxiv.org/abs/2602.03750 · **PDF:** https://arxiv.org/pdf/2602.03750v1
+
+## 15895. Phaedra: Learning High-Fidelity Discrete Tokenization for the Physical Science
+
+- **ID:** `2602.03915` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Levi Lingsch, Georgios Kissas, Johannes Jakubik, et al.
+- **abs:** https://arxiv.org/abs/2602.03915 · **PDF:** https://arxiv.org/pdf/2602.03915v1
+
+## 15896. Soft Sensor for Bottom-Hole Pressure Estimation in Petroleum Wells Using Long Short-Term Memory and Transfer Learning
+
+- **ID:** `2602.03737` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** M. A. Fernandes, E. Gildin, M. A. Sampaio
+- **abs:** https://arxiv.org/abs/2602.03737 · **PDF:** https://arxiv.org/pdf/2602.03737v1
+
+## 15897. Fast-MWEM: Private Data Release in Sublinear Time
+
+- **ID:** `2602.03732` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Themistoklis Haris, Steve Choi, Mutiraj Laksanawisit
+- **abs:** https://arxiv.org/abs/2602.03732 · **PDF:** https://arxiv.org/pdf/2602.03732v1
+
+## 15898. CUBO: Self-Contained Retrieval-Augmented Generation on Consumer Laptops 10 GB Corpora, 16 GB RAM, Single-Device Deployment
+
+- **ID:** `2602.03731` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Paolo Astrino
+- **abs:** https://arxiv.org/abs/2602.03731 · **PDF:** https://arxiv.org/pdf/2602.03731v1
+
+## 15899. Efficient Variance-reduced Estimation from Generative EHR Models: The SCOPE and REACH Estimators
+
+- **ID:** `2602.03730` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Luke Solo, Matthew B. A. McDermott, William F. Parker, et al.
+- **abs:** https://arxiv.org/abs/2602.03730 · **PDF:** https://arxiv.org/pdf/2602.03730v1
+
+## 15900. Efficient Training of Boltzmann Generators Using Off-Policy Log-Dispersion Regularization
+
+- **ID:** `2602.03729` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Henrik Schopmans, Christopher von Klitzing, Pascal Friederich
+- **abs:** https://arxiv.org/abs/2602.03729 · **PDF:** https://arxiv.org/pdf/2602.03729v1
+
+## 15901. Beyond Independent Genes: Learning Module-Inductive Representations for Gene Perturbation Prediction
+
+- **ID:** `2602.04901` · **primary:** `q-bio.GN` · **Submitted:** 2026-02-03
+- **Authors:** Jiafa Ruan, Ruijie Quan, Zongxin Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.04901 · **PDF:** https://arxiv.org/pdf/2602.04901v1
+
+## 15902. Training Multi-Turn Search Agent via Contrastive Dynamic Branch Sampling
+
+- **ID:** `2602.03719` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Yubao Zhao, Weiquan Huang, Sudong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.03719 · **PDF:** https://arxiv.org/pdf/2602.03719v1
+
+## 15903. Additive Multi-Step Markov Chains and the Curse of Dimensionality in Large Language Models
+
+- **ID:** `2603.04412` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** O. V. Usatenko, S. S. Melnyk, G. M. Pritula
+- **abs:** https://arxiv.org/abs/2603.04412 · **PDF:** https://arxiv.org/pdf/2603.04412v1
+
+## 15904. VR-VFL: Joint Rate and Client Selection for Vehicular Federated Learning Under Imperfect CSI
+
+- **ID:** `2602.03711` · **primary:** `eess.SP` · **Submitted:** 2026-02-03
+- **Authors:** Metehan Karatas, Subhrakanti Dey, Christian Rohner, et al.
+- **abs:** https://arxiv.org/abs/2602.03711 · **PDF:** https://arxiv.org/pdf/2602.03711v1
+
+## 15905. Noise reduction in BERT NER models for clinical entity extraction
+
+- **ID:** `2603.00022` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Kuldeep Jiwani, Yash K Jeengar, Ayush Dhaka
+- **abs:** https://arxiv.org/abs/2603.00022 · **PDF:** https://arxiv.org/pdf/2603.00022v1
+
+## 15906. No Shortcuts to Culture: Indonesian Multi-hop Question Answering for Complex Cultural Understanding
+
+- **ID:** `2602.03709` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Vynska Amalia Permadi, Xingwei Tan, Nafise Sadat Moosavi, et al.
+- **abs:** https://arxiv.org/abs/2602.03709 · **PDF:** https://arxiv.org/pdf/2602.03709v1
+
+## 15907. Beyond Tokens: Semantic-Aware Speculative Decoding for Efficient Inference by Probing Internal States
+
+- **ID:** `2602.03708` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Ximing Dong, Shaowei Wang, Dayi Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.03708 · **PDF:** https://arxiv.org/pdf/2602.03708v2
+
+## 15908. OmniRAG-Agent: Agentic Omnimodal Reasoning for Low-Resource Long Audio-Video Question Answering
+
+- **ID:** `2602.03707` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Yifan Zhu, Xinyu Mu, Tao Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.03707 · **PDF:** https://arxiv.org/pdf/2602.03707v4
+
+## 15909. Cognitively Diverse Multiple-Choice Question Generation: A Hybrid Multi-Agent Framework with Large Language Models
+
+- **ID:** `2602.03704` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Yu Tian, Linh Huynh, Katerina Christhilf, et al.
+- **abs:** https://arxiv.org/abs/2602.03704 · **PDF:** https://arxiv.org/pdf/2602.03704v1
+
+## 15910. Anytime Pretraining: Horizon-Free Learning-Rate Schedules with Weight Averaging
+
+- **ID:** `2602.03702` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Alexandru Meterez, Pranav Ajit Nair, Depen Morwani, et al.
+- **abs:** https://arxiv.org/abs/2602.03702 · **PDF:** https://arxiv.org/pdf/2602.03702v1
+
+## 15911. When Excellence Stops Producing Knowledge: A Practitioner's Observation on Research Funding
+
+- **ID:** `2602.07039` · **primary:** `cs.CY` · **Submitted:** 2026-02-03
+- **Authors:** Heimo Müller
+- **abs:** https://arxiv.org/abs/2602.07039 · **PDF:** https://arxiv.org/pdf/2602.07039v1
+
+## 15912. Data-Driven Graph Filters via Adaptive Spectral Shaping
+
+- **ID:** `2602.03698` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Dylan Sandfelder, Mihai Cucuringu, Xiaowen Dong
+- **abs:** https://arxiv.org/abs/2602.03698 · **PDF:** https://arxiv.org/pdf/2602.03698v1
+
+## 15913. Causal Discovery for Cross-Sectional Data Based on Super-Structure and Divide-and-Conquer
+
+- **ID:** `2602.03914` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Wenyu Wang, Yaping Wan
+- **abs:** https://arxiv.org/abs/2602.03914 · **PDF:** https://arxiv.org/pdf/2602.03914v1
+
+## 15914. Conflict-Resolving and Sharpness-Aware Minimization for Generalized Knowledge Editing with Multiple Updates
+
+- **ID:** `2602.03696` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Duy Nguyen, Hanqi Xiao, Archiki Prasad, et al.
+- **abs:** https://arxiv.org/abs/2602.03696 · **PDF:** https://arxiv.org/pdf/2602.03696v1
+
+## 15915. Agent Primitives: Reusable Latent Building Blocks for Multi-Agent Systems
+
+- **ID:** `2602.03695` · **primary:** `cs.MA` · **Submitted:** 2026-02-03
+- **Authors:** Haibo Jin, Kuang Peng, Ye Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.03695 · **PDF:** https://arxiv.org/pdf/2602.03695v1
+
+## 15916. OCRTurk: A Comprehensive OCR Benchmark for Turkish
+
+- **ID:** `2602.03693` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Deniz Yılmaz, Evren Ayberk Munis, Çağrı Toraman, et al.
+- **abs:** https://arxiv.org/abs/2602.03693 · **PDF:** https://arxiv.org/pdf/2602.03693v1
+
+## 15917. Entropy-Aware Structural Alignment for Zero-Shot Handwritten Chinese Character Recognition
+
+- **ID:** `2602.03913` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Qiuming Luo, Tao Zeng, Feng Li, et al.
+- **abs:** https://arxiv.org/abs/2602.03913 · **PDF:** https://arxiv.org/pdf/2602.03913v2
+
+## 15918. LLM-Inspired Pretrain-Then-Finetune for Small-Data, Large-Scale Optimization
+
+- **ID:** `2602.03690` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Zishi Zhang, Jinhui Han, Ming Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.03690 · **PDF:** https://arxiv.org/pdf/2602.03690v1
+
+## 15919. Rethinking the Reranker: Boundary-Aware Evidence Selection for Robust Retrieval-Augmented Generation
+
+- **ID:** `2602.03689` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Jiashuo Sun, Pengcheng Jiang, Saizhuo Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.03689 · **PDF:** https://arxiv.org/pdf/2602.03689v1
+
+## 15920. TodyComm: Task-Oriented Dynamic Communication for Multi-Round LLM-based Multi-Agent System
+
+- **ID:** `2602.03688` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Wenzhe Fan, Tommaso Tognoli, Henry Peng Zou, et al.
+- **abs:** https://arxiv.org/abs/2602.03688 · **PDF:** https://arxiv.org/pdf/2602.03688v1
+
+## 15921. QuAIL: Quality-Aware Inertial Learning for Robust Training under Data Corruption
+
+- **ID:** `2602.03686` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Mattia Sabella, Alberto Archetti, Pietro Pinoli, et al.
+- **abs:** https://arxiv.org/abs/2602.03686 · **PDF:** https://arxiv.org/pdf/2602.03686v1
+
+## 15922. Universal One-third Time Scaling in Learning Peaked Distributions
+
+- **ID:** `2602.03685` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yizhou Liu, Ziming Liu, Cengiz Pehlevan, et al.
+- **abs:** https://arxiv.org/abs/2602.03685 · **PDF:** https://arxiv.org/pdf/2602.03685v1
+
+## 15923. Improved Analysis of the Accelerated Noisy Power Method with Applications to Decentralized PCA
+
+- **ID:** `2602.03682` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Pierre Aguié, Mathieu Even, Laurent Massoulié
+- **abs:** https://arxiv.org/abs/2602.03682 · **PDF:** https://arxiv.org/pdf/2602.03682v1
+
+## 15924. Neural Attention Search Linear: Towards Adaptive Token-Level Hybrid Attention Models
+
+- **ID:** `2602.03681` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Difan Deng, Andreas Bentzen Winje, Lukas Fehring, et al.
+- **abs:** https://arxiv.org/abs/2602.03681 · **PDF:** https://arxiv.org/pdf/2602.03681v1
+
+## 15925. Echo State Networks for Time Series Forecasting: Hyperparameter Sweep and Benchmarking
+
+- **ID:** `2602.03912` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Alexander Häußer
+- **abs:** https://arxiv.org/abs/2602.03912 · **PDF:** https://arxiv.org/pdf/2602.03912v2
+
+## 15926. ContraLog: Log File Anomaly Detection with Contrastive Learning and Masked Language Modeling
+
+- **ID:** `2602.03678` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Simon Dietz, Kai Klede, An Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.03678 · **PDF:** https://arxiv.org/pdf/2602.03678v1
+
+## 15927. Instruction Anchors: Dissecting the Causal Dynamics of Modality Arbitration
+
+- **ID:** `2602.03677` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Yu Zhang, Mufan Xu, Xuefeng Bai, et al.
+- **abs:** https://arxiv.org/abs/2602.03677 · **PDF:** https://arxiv.org/pdf/2602.03677v1
+
+## 15928. General learned delegation by clones
+
+- **ID:** `2602.13262` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Darren Li, Meiqi Chen, Chenze Shao, et al.
+- **abs:** https://arxiv.org/abs/2602.13262 · **PDF:** https://arxiv.org/pdf/2602.13262v1
+
+## 15929. Equilibrium Propagation for Non-Conservative Systems
+
+- **ID:** `2602.03670` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Antonino Emanuele Scurria, Dimitri Vanden Abeele, Bortolo Matteo Mognetti, et al.
+- **abs:** https://arxiv.org/abs/2602.03670 · **PDF:** https://arxiv.org/pdf/2602.03670v1
+
+## 15930. Efficient Sequential Neural Network with Spatial-Temporal Attention and Linear LSTM for Robust Lane Detection Using Multi-Frame Images
+
+- **ID:** `2602.03669` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Sandeep Patil, Yongqi Dong, Haneen Farah, et al.
+- **abs:** https://arxiv.org/abs/2602.03669 · **PDF:** https://arxiv.org/pdf/2602.03669v1
+
+## 15931. Mitigating Conversational Inertia in Multi-Turn Agents
+
+- **ID:** `2602.03664` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Yang Wan, Zheng Cao, Zhenhao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03664 · **PDF:** https://arxiv.org/pdf/2602.03664v2
+
+## 15932. Sequential Group Composition: A Window into the Mechanics of Deep Learning
+
+- **ID:** `2602.03655` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Giovanni Luca Marchetti, Daniel Kunin, Adele Myers, et al.
+- **abs:** https://arxiv.org/abs/2602.03655 · **PDF:** https://arxiv.org/pdf/2602.03655v1
+
+## 15933. Evaluating Kubernetes Performance for GenAI Inference: From Automatic Speech Recognition to LLM Summarization
+
+- **ID:** `2602.04900` · **primary:** `cs.ET` · **Submitted:** 2026-02-03
+- **Authors:** Sai Sindhur Malleni, Raúl Sevilla, Aleksei Vasilevskii, et al.
+- **abs:** https://arxiv.org/abs/2602.04900 · **PDF:** https://arxiv.org/pdf/2602.04900v3
+
+## 15934. RAGTurk: Best Practices for Retrieval Augmented Generation in Turkish
+
+- **ID:** `2602.03652` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Süha Kağan Köse, Mehmet Can Baytekin, Burak Aktaş, et al.
+- **abs:** https://arxiv.org/abs/2602.03652 · **PDF:** https://arxiv.org/pdf/2602.03652v1
+
+## 15935. Search-R2: Enhancing Search-Integrated Reasoning via Actor-Refiner Collaboration
+
+- **ID:** `2602.03647` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Bowei He, Minda Hu, Zenan Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.03647 · **PDF:** https://arxiv.org/pdf/2602.03647v1
+
+## 15936. Reinforcement Fine-Tuning for History-Aware Dense Retriever in RAG
+
+- **ID:** `2602.03645` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yicheng Zhang, Zhen Qin, Zhaomin Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.03645 · **PDF:** https://arxiv.org/pdf/2602.03645v1
+
+## 15937. CTTVAE: Latent Space Structuring for Conditional Tabular Data Generation on Imbalanced Datasets
+
+- **ID:** `2602.03641` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Milosh Devic, Jordan Gierschendorf, David Garson
+- **abs:** https://arxiv.org/abs/2602.03641 · **PDF:** https://arxiv.org/pdf/2602.03641v1
+
+## 15938. Tutorial on Reasoning for IR & IR for Reasoning
+
+- **ID:** `2602.03640` · **primary:** `cs.IR` · **Submitted:** 2026-02-03
+- **Authors:** Mohanna Hoveyda, Panagiotis Efstratiadis, Arjen de Vries, et al.
+- **abs:** https://arxiv.org/abs/2602.03640 · **PDF:** https://arxiv.org/pdf/2602.03640v1
+
+## 15939. TRE: Encouraging Exploration in the Trust Region
+
+- **ID:** `2602.03635` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Chao Huang, Yujing Lu, Quangang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.03635 · **PDF:** https://arxiv.org/pdf/2602.03635v1
+
+## 15940. BIRDTurk: Adaptation of the BIRD Text-to-SQL Dataset to Turkish
+
+- **ID:** `2602.03633` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Burak Aktaş, Mehmet Can Baytekin, Süha Kağan Köse, et al.
+- **abs:** https://arxiv.org/abs/2602.03633 · **PDF:** https://arxiv.org/pdf/2602.03633v1
+
+## 15941. The Role of Target Update Frequencies in Q-Learning
+
+- **ID:** `2602.03911` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Simon Weissmann, Tilman Aach, Benedikt Wille, et al.
+- **abs:** https://arxiv.org/abs/2602.03911 · **PDF:** https://arxiv.org/pdf/2602.03911v1
+
+## 15942. Can LLMs Do Rocket Science? Exploring the Limits of Complex Reasoning with GTOC 12
+
+- **ID:** `2602.03630` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Iñaki del Campo, Pablo Cuervo, Victor Rodriguez-Fernandez, et al.
+- **abs:** https://arxiv.org/abs/2602.03630 · **PDF:** https://arxiv.org/pdf/2602.03630v1
+
+## 15943. Ultra Fast PDE Solving via Physics Guided Few-step Diffusion
+
+- **ID:** `2602.03627` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Cindy Xiangrui Kong, Yueqi Wang, Haoyang Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.03627 · **PDF:** https://arxiv.org/pdf/2602.03627v1
+
+## 15944. Learning Query-Specific Rubrics from Human Preferences for DeepResearch Report Generation
+
+- **ID:** `2602.03619` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Changze Lv, Jie Zhou, Wentao Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.03619 · **PDF:** https://arxiv.org/pdf/2602.03619v1
+
+## 15945. Quantization-Aware Regularizers for Deep Neural Networks Compression
+
+- **ID:** `2602.03614` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Dario Malchiodi, Mattia Ferraretto, Marco Frasca
+- **abs:** https://arxiv.org/abs/2602.03614 · **PDF:** https://arxiv.org/pdf/2602.03614v1
+
+## 15946. Simulation-Based Inference via Regression Projection and Batched Discrepancies
+
+- **ID:** `2602.03613` · **primary:** `stat.ME` · **Submitted:** 2026-02-03
+- **Authors:** Arya Farahi, Jonah Rose, Paul Torrey
+- **abs:** https://arxiv.org/abs/2602.03613 · **PDF:** https://arxiv.org/pdf/2602.03613v1
+
+## 15947. One Model, Many Skills: Parameter-Efficient Fine-Tuning for Multitask Code Analysis
+
+- **ID:** `2603.09978` · **primary:** `cs.SE` · **Submitted:** 2026-02-03
+- **Authors:** Amal Akli, Maxime Cordy, Mike Papadakis, et al.
+- **abs:** https://arxiv.org/abs/2603.09978 · **PDF:** https://arxiv.org/pdf/2603.09978v1
+
+## 15948. Generator-based Graph Generation via Heat Diffusion
+
+- **ID:** `2602.03612` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Anthony Stephenson, Ian Gallagher, Christopher Nemeth
+- **abs:** https://arxiv.org/abs/2602.03612 · **PDF:** https://arxiv.org/pdf/2602.03612v1
+
+## 15949. Explanations Leak: Membership Inference with Differential Privacy and Active Learning Defense
+
+- **ID:** `2602.03611` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Fatima Ezzeddine, Osama Zammar, Silvia Giordano, et al.
+- **abs:** https://arxiv.org/abs/2602.03611 · **PDF:** https://arxiv.org/pdf/2602.03611v1
+
+## 15950. Controlling Output Rankings in Generative Engines for LLM-based Search
+
+- **ID:** `2602.03608` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Haibo Jin, Ruoxi Chen, Peiyan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03608 · **PDF:** https://arxiv.org/pdf/2602.03608v1
+
+## 15951. A Lightweight Library for Energy-Based Joint-Embedding Predictive Architectures
+
+- **ID:** `2602.03604` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Basile Terver, Randall Balestriero, Megi Dervishi, et al.
+- **abs:** https://arxiv.org/abs/2602.03604 · **PDF:** https://arxiv.org/pdf/2602.03604v2
+
+## 15952. SAGE-5GC: Security-Aware Guidelines for Evaluating Anomaly Detection in the 5G Core Network
+
+- **ID:** `2602.03596` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Cristian Manca, Christian Scano, Giorgio Piras, et al.
+- **abs:** https://arxiv.org/abs/2602.03596 · **PDF:** https://arxiv.org/pdf/2602.03596v3
+
+## 15953. From Global to Local: Learning Context-Aware Graph Representations for Document Classification and Summarization
+
+- **ID:** `2603.00021` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Ruangrin Ldallitsakool, Margarita Bugueño, Gerard de Melo
+- **abs:** https://arxiv.org/abs/2603.00021 · **PDF:** https://arxiv.org/pdf/2603.00021v1
+
+## 15954. Efficient Algorithms for Partial Constraint Satisfaction Problems over Control-flow Graphs
+
+- **ID:** `2602.03588` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Xuran Cai, Amir Goharshady
+- **abs:** https://arxiv.org/abs/2602.03588 · **PDF:** https://arxiv.org/pdf/2602.03588v1
+
+## 15955. Phantom Transfer: Data-level Defences are Insufficient Against Data Poisoning
+
+- **ID:** `2602.04899` · **primary:** `cs.CR` · **Submitted:** 2026-02-03
+- **Authors:** Andrew Draganov, Tolga H. Dur, Anandmayi Bhongade, et al.
+- **abs:** https://arxiv.org/abs/2602.04899 · **PDF:** https://arxiv.org/pdf/2602.04899v1
+
+## 15956. CL-bench: A Benchmark for Context Learning
+
+- **ID:** `2602.03587` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Shihan Dou, Ming Zhang, Zhangyue Yin, et al.
+- **abs:** https://arxiv.org/abs/2602.03587 · **PDF:** https://arxiv.org/pdf/2602.03587v1
+
+## 15957. APEX: Probing Neural Networks via Activation Perturbation
+
+- **ID:** `2602.03586` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Tao Ren, Xiaoyu Luo, Qiongxiu Li
+- **abs:** https://arxiv.org/abs/2602.03586 · **PDF:** https://arxiv.org/pdf/2602.03586v1
+
+## 15958. $V_0$: A Generalist Value Model for Any Policy at State Zero
+
+- **ID:** `2602.03584` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Yi-Kai Zhang, Zhiyuan Yao, Hongyan Hao, et al.
+- **abs:** https://arxiv.org/abs/2602.03584 · **PDF:** https://arxiv.org/pdf/2602.03584v2
+
+## 15959. Optimization and Generation in Aerodynamics Inverse Design
+
+- **ID:** `2602.03582` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Huaguan Chen, Ning Lin, Luxi Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.03582 · **PDF:** https://arxiv.org/pdf/2602.03582v2
+
+## 15960. Don't believe everything you read: Understanding and Measuring MCP Behavior under Misleading Tool Descriptions
+
+- **ID:** `2602.03580` · **primary:** `cs.CR` · **Submitted:** 2026-02-03
+- **Authors:** Zhihao Li, Boyang Ma, Xuelong Dai, et al.
+- **abs:** https://arxiv.org/abs/2602.03580 · **PDF:** https://arxiv.org/pdf/2602.03580v1
+
+## 15961. Use Graph When It Needs: Efficiently and Adaptively Integrating Retrieval-Augmented Generation with Graphs
+
+- **ID:** `2602.03578` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Su Dong, Qinggang Zhang, Yilin Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.03578 · **PDF:** https://arxiv.org/pdf/2602.03578v1
+
+## 15962. Asymmetric Hierarchical Anchoring for Audio-Visual Joint Representation: Resolving Information Allocation Ambiguity for Robust Cross-Modal Generalization
+
+- **ID:** `2602.03570` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Bixing Wu, Yuhong Zhao, Zongli Ye, et al.
+- **abs:** https://arxiv.org/abs/2602.03570 · **PDF:** https://arxiv.org/pdf/2602.03570v1
+
+## 15963. HY3D-Bench: Generation of 3D Assets
+
+- **ID:** `2602.03907` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Team Hunyuan3D, :, Bowen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03907 · **PDF:** https://arxiv.org/pdf/2602.03907v1
+
+## 15964. EHRWorld: A Patient-Centric Medical World Model for Long-Horizon Clinical Trajectories
+
+- **ID:** `2602.03569` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Linjie Mu, Zhongzhen Huang, Yannian Gu, et al.
+- **abs:** https://arxiv.org/abs/2602.03569 · **PDF:** https://arxiv.org/pdf/2602.03569v1
+
+## 15965. EVE: Efficient Verification of Data Erasure through Customized Perturbation in Approximate Unlearning
+
+- **ID:** `2602.03567` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Weiqi Wang, Zhiyi Tian, Chenhan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03567 · **PDF:** https://arxiv.org/pdf/2602.03567v1
+
+## 15966. Riemannian Neural Optimal Transport
+
+- **ID:** `2602.03566` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Alessandro Micheli, Yueqi Cao, Anthea Monod, et al.
+- **abs:** https://arxiv.org/abs/2602.03566 · **PDF:** https://arxiv.org/pdf/2602.03566v1
+
+## 15967. CoGenCast: A Coupled Autoregressive-Flow Generative Framework for Time Series Forecasting
+
+- **ID:** `2602.03564` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yaguo Liu, Mingyue Cheng, Daoyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.03564 · **PDF:** https://arxiv.org/pdf/2602.03564v1
+
+## 15968. ACL: Aligned Contrastive Learning Improves BERT and Multi-exit BERT Fine-tuning
+
+- **ID:** `2602.03563` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Liz Li, Wei Zhu
+- **abs:** https://arxiv.org/abs/2602.03563 · **PDF:** https://arxiv.org/pdf/2602.03563v2
+
+## 15969. NPCNet: Navigator-Driven Pseudo Text for Deep Clustering of Early Sepsis Phenotyping
+
+- **ID:** `2602.03562` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Pi-Ju Tsai, Charkkri Limbud, Kuan-Fu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.03562 · **PDF:** https://arxiv.org/pdf/2602.03562v1
+
+## 15970. GeoIB: Geometry-Aware Information Bottleneck via Statistical-Manifold Compression
+
+- **ID:** `2602.03906` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Weiqi Wang, Zhiyi Tian, Chenhan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03906 · **PDF:** https://arxiv.org/pdf/2602.03906v1
+
+## 15971. HySparse: A Hybrid Sparse Attention Architecture with Oracle Token Selection and KV Cache Sharing
+
+- **ID:** `2602.03560` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Yizhao Gao, Jianyu Wei, Qihao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03560 · **PDF:** https://arxiv.org/pdf/2602.03560v1
+
+## 15972. ELIQ: A Label-Free Framework for Quality Assessment of Evolving AI-Generated Images
+
+- **ID:** `2602.03558` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Xinyue Li, Zhiming Xu, Zhichao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03558 · **PDF:** https://arxiv.org/pdf/2602.03558v1
+
+## 15973. When Single Answer Is Not Enough: Rethinking Single-Step Retrosynthesis Benchmarks for LLMs
+
+- **ID:** `2602.03554` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Bogdan Zagribelnyy, Ivan Ilin, Maksim Kuznetsov, et al.
+- **abs:** https://arxiv.org/abs/2602.03554 · **PDF:** https://arxiv.org/pdf/2602.03554v1
+
+## 15974. Assessing the Impact of Typological Features on Multilingual Machine Translation in the Age of Large Language Models
+
+- **ID:** `2602.03551` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Vitalii Hirak, Jaap Jumelet, Arianna Bisazza
+- **abs:** https://arxiv.org/abs/2602.03551 · **PDF:** https://arxiv.org/pdf/2602.03551v1
+
+## 15975. SEAD: Self-Evolving Agent for Multi-Turn Service Dialogue
+
+- **ID:** `2602.03548` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Yuqin Dai, Ning Gao, Wei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03548 · **PDF:** https://arxiv.org/pdf/2602.03548v2
+
+## 15976. How to Train Your Resistive Network: Generalized Equilibrium Propagation and Analytical Learning
+
+- **ID:** `2602.03546` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Jonathan Lin, Aman Desai, Frank Barrows, et al.
+- **abs:** https://arxiv.org/abs/2602.03546 · **PDF:** https://arxiv.org/pdf/2602.03546v3
+
+## 15977. Persona Generators: Generating Diverse Synthetic Personas at Scale
+
+- **ID:** `2602.03545` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Davide Paglieri, Logan Cross, William A. Cunningham, et al.
+- **abs:** https://arxiv.org/abs/2602.03545 · **PDF:** https://arxiv.org/pdf/2602.03545v1
+
+## 15978. Can Large Language Models Generalize Procedures Across Representations?
+
+- **ID:** `2602.03542` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Fangru Lin, Valentin Hofmann, Xingchen Wan, et al.
+- **abs:** https://arxiv.org/abs/2602.03542 · **PDF:** https://arxiv.org/pdf/2602.03542v1
+
+## 15979. Group Selection as a Safeguard Against AI Substitution
+
+- **ID:** `2602.03541` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Qiankun Zhong, Thomas F. Eisenmann, Julian Garcia, et al.
+- **abs:** https://arxiv.org/abs/2602.03541 · **PDF:** https://arxiv.org/pdf/2602.03541v1
+
+## 15980. MatGPTQ: Accurate and Efficient Post-Training Matryoshka Quantization
+
+- **ID:** `2602.03537` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Maximilian Kleinegger, Elvir Crnčević, Dan Alistarh
+- **abs:** https://arxiv.org/abs/2602.03537 · **PDF:** https://arxiv.org/pdf/2602.03537v1
+
+## 15981. Sparse Training of Neural Networks based on Multilevel Mirror Descent
+
+- **ID:** `2602.03535` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yannick Lunk, Sebastian J. Scott, Leon Bungert
+- **abs:** https://arxiv.org/abs/2602.03535 · **PDF:** https://arxiv.org/pdf/2602.03535v1
+
+## 15982. Robust Representation Learning in Masked Autoencoders
+
+- **ID:** `2602.03531` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Anika Shrivastava, Renu Rameshan, Samar Agnihotri
+- **abs:** https://arxiv.org/abs/2602.03531 · **PDF:** https://arxiv.org/pdf/2602.03531v1
+
+## 15983. Morphe: High-Fidelity Generative Video Streaming with Vision Foundation Model
+
+- **ID:** `2602.03529` · **primary:** `cs.NI` · **Submitted:** 2026-02-03
+- **Authors:** Tianyi Gong, Zijian Cao, Zixing Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03529 · **PDF:** https://arxiv.org/pdf/2602.03529v1
+
+## 15984. WARP Logic Neural Networks
+
+- **ID:** `2602.03527` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Lino Gerlach, Thore Gerlach, Liv Våge, et al.
+- **abs:** https://arxiv.org/abs/2602.03527 · **PDF:** https://arxiv.org/pdf/2602.03527v1
+
+## 15985. D3PIA: A Discrete Denoising Diffusion Model for Piano Accompaniment Generation From Lead sheet
+
+- **ID:** `2602.03523` · **primary:** `cs.SD` · **Submitted:** 2026-02-03
+- **Authors:** Eunjin Choi, Hounsu Kim, Hayeon Bang, et al.
+- **abs:** https://arxiv.org/abs/2602.03523 · **PDF:** https://arxiv.org/pdf/2602.03523v1
+
+## 15986. Live or Lie: Action-Aware Capsule Multiple Instance Learning for Risk Assessment in Live Streaming Platforms
+
+- **ID:** `2602.03520` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yiran Qiao, Jing Chen, Xiang Ao, et al.
+- **abs:** https://arxiv.org/abs/2602.03520 · **PDF:** https://arxiv.org/pdf/2602.03520v2
+
+## 15987. Rank-Learner: Orthogonal Ranking of Treatment Effects
+
+- **ID:** `2602.03517` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Henri Arno, Dennis Frauen, Emil Javurek, et al.
+- **abs:** https://arxiv.org/abs/2602.03517 · **PDF:** https://arxiv.org/pdf/2602.03517v1
+
+## 15988. Not All Negative Samples Are Equal: LLMs Learn Better from Plausible Reasoning
+
+- **ID:** `2602.03516` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Zixiang Di, Jinyi Han, Shuo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03516 · **PDF:** https://arxiv.org/pdf/2602.03516v2
+
+## 15989. Mitigating Staleness in Asynchronous Pipeline Parallelism via Basis Rotation
+
+- **ID:** `2602.03515` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Hyunji Jung, Sungbin Shin, Namhoon Lee
+- **abs:** https://arxiv.org/abs/2602.03515 · **PDF:** https://arxiv.org/pdf/2602.03515v1
+
+## 15990. A Function-Space Stability Boundary for Generalization in Interpolating Learning Systems
+
+- **ID:** `2602.03514` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Ronald Katende
+- **abs:** https://arxiv.org/abs/2602.03514 · **PDF:** https://arxiv.org/pdf/2602.03514v2
+
+## 15991. CMR: Contractive Mapping Embeddings for Robust Humanoid Locomotion on Unstructured Terrains
+
+- **ID:** `2602.03511` · **primary:** `cs.RO` · **Submitted:** 2026-02-03
+- **Authors:** Qixin Zeng, Hongyin Zhang, Shangke Lyu, et al.
+- **abs:** https://arxiv.org/abs/2602.03511 · **PDF:** https://arxiv.org/pdf/2602.03511v1
+
+## 15992. FaithRL: Learning to Reason Faithfully through Step-Level Faithfulness Maximization
+
+- **ID:** `2602.03507` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Runquan Gui, Yafu Li, Xiaoye Qu, et al.
+- **abs:** https://arxiv.org/abs/2602.03507 · **PDF:** https://arxiv.org/pdf/2602.03507v2
+
+## 15993. Explaining the Explainer: Understanding the Inner Workings of Transformer-based Symbolic Regression Models
+
+- **ID:** `2602.03506` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Arco van Breda, Erman Acar
+- **abs:** https://arxiv.org/abs/2602.03506 · **PDF:** https://arxiv.org/pdf/2602.03506v1
+
+## 15994. Generative Decompression: Optimal Lossy Decoding Against Distribution Mismatch
+
+- **ID:** `2602.03505` · **primary:** `cs.IT` · **Submitted:** 2026-02-03
+- **Authors:** Saeed R. Khosravirad, Ahmed Alkhateeb, Ingrid van de Voorde
+- **abs:** https://arxiv.org/abs/2602.03505 · **PDF:** https://arxiv.org/pdf/2602.03505v1
+
+## 15995. Semantic-level Backdoor Attack against Text-to-Image Diffusion Models
+
+- **ID:** `2602.04898` · **primary:** `cs.CR` · **Submitted:** 2026-02-03
+- **Authors:** Tianxin Chen, Wenbo Jiang, Hongqiao Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.04898 · **PDF:** https://arxiv.org/pdf/2602.04898v2
+
+## 15996. Reparameterization Flow Policy Optimization
+
+- **ID:** `2602.03501` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Hai Zhong, Zhuoran Li, Xun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.03501 · **PDF:** https://arxiv.org/pdf/2602.03501v1
+
+## 15997. One Size Does Not Fit All: Token-Wise Adaptive Compression for KV Cache
+
+- **ID:** `2603.04411` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Liming Lu, Kaixi Qiu, Jiayu Zhou, et al.
+- **abs:** https://arxiv.org/abs/2603.04411 · **PDF:** https://arxiv.org/pdf/2603.04411v1
+
+## 15998. Lookahead Path Likelihood Optimization for Diffusion LLMs
+
+- **ID:** `2602.03496` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Xuejie Liu, Yap Vit Chun, Yitao Liang, et al.
+- **abs:** https://arxiv.org/abs/2602.03496 · **PDF:** https://arxiv.org/pdf/2602.03496v1
+
+## 15999. DALI: A Workload-Aware Offloading Framework for Efficient MoE Inference on Local PCs
+
+- **ID:** `2602.03495` · **primary:** `cs.DC` · **Submitted:** 2026-02-03
+- **Authors:** Zeyu Zhu, Gang Li, Peisong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.03495 · **PDF:** https://arxiv.org/pdf/2602.03495v1
+
+## 16000. Least but not Last: Fine-tuning Intermediate Principal Components for Better Performance-Forgetting Trade-Offs
+
+- **ID:** `2602.03493` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Alessio Quercia, Arya Bangun, Ira Assent, et al.
+- **abs:** https://arxiv.org/abs/2602.03493 · **PDF:** https://arxiv.org/pdf/2602.03493v1
+
+## 16001. Decoupling Skeleton and Flesh: Efficient Multimodal Table Reasoning with Disentangled Alignment and Structure-aware Guidance
+
+- **ID:** `2602.03491` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Yingjie Zhu, Xuefeng Bai, Kehai Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.03491 · **PDF:** https://arxiv.org/pdf/2602.03491v1
+
+## 16002. A Minimal Task Reveals Emergent Path Integration and Object-Location Binding in a Predictive Sequence Model
+
+- **ID:** `2602.03490` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Linda Ariel Ventura, Victoria Bosch, Tim C Kietzmann, et al.
+- **abs:** https://arxiv.org/abs/2602.03490 · **PDF:** https://arxiv.org/pdf/2602.03490v1
+
+## 16003. DeepDFA: Injecting Temporal Logic in Deep Learning for Sequential Subsymbolic Applications
+
+- **ID:** `2602.03486` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Elena Umili, Francesco Argenziano, Roberto Capobianco
+- **abs:** https://arxiv.org/abs/2602.03486 · **PDF:** https://arxiv.org/pdf/2602.03486v1
+
+## 16004. Self-Verification Dilemma: Experience-Driven Suppression of Overused Checking in LLM Reasoning
+
+- **ID:** `2602.03485` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Quanyu Long, Kai Jie Jiang, Jianda Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.03485 · **PDF:** https://arxiv.org/pdf/2602.03485v1
+
+## 16005. Preferences for Idiomatic Language are Acquired Slowly -- and Forgotten Quickly: A Case Study on Swedish
+
+- **ID:** `2602.03484` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Jenny Kunz
+- **abs:** https://arxiv.org/abs/2602.03484 · **PDF:** https://arxiv.org/pdf/2602.03484v1
+
+## 16006. When Routing Collapses: On the Degenerate Convergence of LLM Routers
+
+- **ID:** `2602.03478` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Guannan Lai, Han-Jia Ye
+- **abs:** https://arxiv.org/abs/2602.03478 · **PDF:** https://arxiv.org/pdf/2602.03478v1
+
+## 16007. ScDiVa: Masked Discrete Diffusion for Joint Modeling of Single-Cell Identity and Expression
+
+- **ID:** `2602.03477` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Mingxuan Wang, Cheng Chen, Gaoyang Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.03477 · **PDF:** https://arxiv.org/pdf/2602.03477v1
+
+## 16008. Scaling Continual Learning with Bi-Level Routing Mixture-of-Experts
+
+- **ID:** `2602.03473` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Meng Lou, Yunxiang Fu, Yizhou Yu
+- **abs:** https://arxiv.org/abs/2602.03473 · **PDF:** https://arxiv.org/pdf/2602.03473v1
+
+## 16009. IntentRL: Training Proactive User-intent Agents for Open-ended Deep Research via Reinforcement Learning
+
+- **ID:** `2602.03468` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Haohao Luo, Zexi Li, Yuexiang Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.03468 · **PDF:** https://arxiv.org/pdf/2602.03468v1
+
+## 16010. The Dual Role of Abstracting over the Irrelevant in Symbolic Explanations: Cognitive Effort vs. Understanding
+
+- **ID:** `2602.03467` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Zeynep G. Saribatur, Johannes Langer, Ute Schmid
+- **abs:** https://arxiv.org/abs/2602.03467 · **PDF:** https://arxiv.org/pdf/2602.03467v1
+
+## 16011. Quantum Circuit Generation via test-time learning with large language models
+
+- **ID:** `2602.03466` · **primary:** `quant-ph` · **Submitted:** 2026-02-03
+- **Authors:** Adriano Macarone-Palmieri, Rosario Lo Franco
+- **abs:** https://arxiv.org/abs/2602.03466 · **PDF:** https://arxiv.org/pdf/2602.03466v4
+
+## 16012. Soft-Radial Projection for Constrained End-to-End Learning
+
+- **ID:** `2602.03461` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Philipp J. Schneider, Daniel Kuhn
+- **abs:** https://arxiv.org/abs/2602.03461 · **PDF:** https://arxiv.org/pdf/2602.03461v1
+
+## 16013. Steering Externalities: Benign Activation Steering Unintentionally Increases Jailbreak Risk for Large Language Models
+
+- **ID:** `2602.04896` · **primary:** `cs.CR` · **Submitted:** 2026-02-03
+- **Authors:** Chen Xiong, Zhiyuan He, Pin-Yu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.04896 · **PDF:** https://arxiv.org/pdf/2602.04896v1
+
+## 16014. Causal Inference on Networks under Misspecified Exposure Mappings: A Partial Identification Framework
+
+- **ID:** `2602.03459` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Maresa Schröder, Miruna Oprescu, Stefan Feuerriegel, et al.
+- **abs:** https://arxiv.org/abs/2602.03459 · **PDF:** https://arxiv.org/pdf/2602.03459v1
+
+## 16015. Beyond Variance: Prompt-Efficient RLVR via Rare-Event Amplification and Bidirectional Pairing
+
+- **ID:** `2602.03452` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Xin Sheng, Jiaxin Li, Yujuan Pang, et al.
+- **abs:** https://arxiv.org/abs/2602.03452 · **PDF:** https://arxiv.org/pdf/2602.03452v1
+
+## 16016. Score-based diffusion models for diffuse optical tomography with uncertainty quantification
+
+- **ID:** `2602.03449` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Fabian Schneider, Meghdoot Mozumder, Konstantin Tamarov, et al.
+- **abs:** https://arxiv.org/abs/2602.03449 · **PDF:** https://arxiv.org/pdf/2602.03449v1
+
+## 16017. SalamahBench: Toward Standardized Safety Evaluation for Arabic Language Models
+
+- **ID:** `2603.04410` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Omar Abdelnasser, Fatemah Alharbi, Khaled Khasawneh, et al.
+- **abs:** https://arxiv.org/abs/2603.04410 · **PDF:** https://arxiv.org/pdf/2603.04410v1
+
+## 16018. Hierarchical Concept-to-Appearance Guidance for Multi-Subject Image Generation
+
+- **ID:** `2602.03448` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Yijia Xu, Zihao Wang, Jinshi Cui
+- **abs:** https://arxiv.org/abs/2602.03448 · **PDF:** https://arxiv.org/pdf/2602.03448v1
+
+## 16019. CRL-VLA: Continual Vision-Language-Action Learning
+
+- **ID:** `2602.03445` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Qixin Zeng, Shuo Zhang, Hongyin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03445 · **PDF:** https://arxiv.org/pdf/2602.03445v1
+
+## 16020. A-RAG: Scaling Agentic Retrieval-Augmented Generation via Hierarchical Retrieval Interfaces
+
+- **ID:** `2602.03442` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Mingxuan Du, Benfeng Xu, Chiwei Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.03442 · **PDF:** https://arxiv.org/pdf/2602.03442v1
+
+## 16021. UNIKIE-BENCH: Benchmarking Large Multimodal Models for Key Information Extraction in Visual Documents
+
+- **ID:** `2602.07038` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Yifan Ji, Zhipeng Xu, Zhenghao Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.07038 · **PDF:** https://arxiv.org/pdf/2602.07038v1
+
+## 16022. Ontology-to-tools compilation for executable semantic constraint enforcement in LLM agents
+
+- **ID:** `2602.03439` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Xiaochi Zhou, Patrick Bulter, Changxuan Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.03439 · **PDF:** https://arxiv.org/pdf/2602.03439v1
+
+## 16023. Acceleration of Atomistic NEGF: Algorithms, Parallelization, and Machine Learning
+
+- **ID:** `2602.03438` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-03
+- **Authors:** Mathieu Luisier, Nicolas Vetsch, Alexander Maeder, et al.
+- **abs:** https://arxiv.org/abs/2602.03438 · **PDF:** https://arxiv.org/pdf/2602.03438v1
+
+## 16024. DiscoverLLM: From Executing Intents to Discovering Them
+
+- **ID:** `2602.03429` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Tae Soo Kim, Yoonjoo Lee, Jaesang Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.03429 · **PDF:** https://arxiv.org/pdf/2602.03429v1
+
+## 16025. CoCoEmo: Composable and Controllable Human-Like Emotional TTS via Activation Steering
+
+- **ID:** `2602.03420` · **primary:** `cs.SD` · **Submitted:** 2026-02-03
+- **Authors:** Siyi Wang, Shihong Tan, Siyi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.03420 · **PDF:** https://arxiv.org/pdf/2602.03420v1
+
+## 16026. SWE-World: Building Software Engineering Agents in Docker-Free Environments
+
+- **ID:** `2602.03419` · **primary:** `cs.SE` · **Submitted:** 2026-02-03
+- **Authors:** Shuang Sun, Huatong Song, Lisheng Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.03419 · **PDF:** https://arxiv.org/pdf/2602.03419v1
+
+## 16027. FactNet: A Billion-Scale Knowledge Graph for Multilingual Factual Grounding
+
+- **ID:** `2602.03417` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Yingli Shen, Wen Lai, Jie Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.03417 · **PDF:** https://arxiv.org/pdf/2602.03417v1
+
+## 16028. Most Convolutional Networks Suffer from Small Adversarial Perturbations
+
+- **ID:** `2602.03415` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Amit Daniely, Idan Mehalel
+- **abs:** https://arxiv.org/abs/2602.03415 · **PDF:** https://arxiv.org/pdf/2602.03415v1
+
+## 16029. Socratic-Geo: Synthetic Data Generation and Geometric Reasoning via Multi-Agent Interaction
+
+- **ID:** `2602.03414` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Zhengbo Jiao, Shaobo Wang, Zifan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03414 · **PDF:** https://arxiv.org/pdf/2602.03414v1
+
+## 16030. Verified Critical Step Optimization for LLM Agents
+
+- **ID:** `2602.03412` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Mukai Li, Qingcheng Zeng, Tianqing Fang, et al.
+- **abs:** https://arxiv.org/abs/2602.03412 · **PDF:** https://arxiv.org/pdf/2602.03412v1
+
+## 16031. SWE-Master: Unleashing the Potential of Software Engineering Agents via Post-Training
+
+- **ID:** `2602.03411` · **primary:** `cs.SE` · **Submitted:** 2026-02-03
+- **Authors:** Huatong Song, Lisheng Huang, Shuang Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.03411 · **PDF:** https://arxiv.org/pdf/2602.03411v2
+
+## 16032. Enhancing Quantum Diffusion Models for Complex Image Generation
+
+- **ID:** `2602.03405` · **primary:** `quant-ph` · **Submitted:** 2026-02-03
+- **Authors:** Jeongbin Jo, Santanam Wishal, Shah Md Khalil Ullah, et al.
+- **abs:** https://arxiv.org/abs/2602.03405 · **PDF:** https://arxiv.org/pdf/2602.03405v2
+
+## 16033. Feasible strategies for conflict resolution within intuitionistic fuzzy preference-based conflict situations
+
+- **ID:** `2602.03403` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Guangming Lang, Mingchuan Shang, Mengjun Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.03403 · **PDF:** https://arxiv.org/pdf/2602.03403v1
+
+## 16034. Risk Awareness Injection: Calibrating Vision-Language Models for Safety without Compromising Utility
+
+- **ID:** `2602.03402` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Mengxuan Wang, Yuxin Chen, Gang Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.03402 · **PDF:** https://arxiv.org/pdf/2602.03402v2
+
+## 16035. Precision in Practice: Knowledge Guided Code Summarizing Grounded in Industrial Expectations
+
+- **ID:** `2602.03400` · **primary:** `cs.SE` · **Submitted:** 2026-02-03
+- **Authors:** Jintai Li, Songqiang Chen, Shuo Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.03400 · **PDF:** https://arxiv.org/pdf/2602.03400v1
+
+## 16036. Towards Distillation-Resistant Large Language Models: An Information-Theoretic Perspective
+
+- **ID:** `2602.03396` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Hao Fang, Tianyi Zhang, Tianqu Zhuang, et al.
+- **abs:** https://arxiv.org/abs/2602.03396 · **PDF:** https://arxiv.org/pdf/2602.03396v2
+
+## 16037. The Label Horizon Paradox: Rethinking Supervision Targets in Financial Forecasting
+
+- **ID:** `2602.03395` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Chen-Hui Song, Shuoling Liu, Liyuan Chen
+- **abs:** https://arxiv.org/abs/2602.03395 · **PDF:** https://arxiv.org/pdf/2602.03395v3
+
+## 16038. Improving the Linearized Laplace Approximation via Quadratic Approximations
+
+- **ID:** `2602.03394` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Pedro Jiménez, Luis A. Ortega, Pablo Morales-Álvarez, et al.
+- **abs:** https://arxiv.org/abs/2602.03394 · **PDF:** https://arxiv.org/pdf/2602.03394v1
+
+## 16039. On the Entropy Dynamics in Reinforcement Fine-Tuning of Large Language Models
+
+- **ID:** `2602.03392` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Shumin Wang, Yuexiang Xie, Wenhao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03392 · **PDF:** https://arxiv.org/pdf/2602.03392v1
+
+## 16040. From Vicious to Virtuous Cycles: Synergistic Representation Learning for Unsupervised Video Object-Centric Learning
+
+- **ID:** `2602.03390` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Hyun Seok Seong, WonJun Moon, Jae-Pil Heo
+- **abs:** https://arxiv.org/abs/2602.03390 · **PDF:** https://arxiv.org/pdf/2602.03390v1
+
+## 16041. Chain-of-Goals Hierarchical Policy for Long-Horizon Offline Goal-Conditioned RL
+
+- **ID:** `2602.03389` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Jinwoo Choi, Sang-Hyun Lee, Seung-Woo Seo
+- **abs:** https://arxiv.org/abs/2602.03389 · **PDF:** https://arxiv.org/pdf/2602.03389v1
+
+## 16042. Toward a Sustainable Federated Learning Ecosystem: A Practical Least Core Mechanism for Payoff Allocation
+
+- **ID:** `2602.03387` · **primary:** `cs.GT` · **Submitted:** 2026-02-03
+- **Authors:** Zhengwei Ni, Zhidu Li, Wei Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.03387 · **PDF:** https://arxiv.org/pdf/2602.03387v1
+
+## 16043. An Approximate Ascent Approach To Prove Convergence of PPO
+
+- **ID:** `2602.03386` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Leif Doering, Daniel Schmidt, Moritz Melcher, et al.
+- **abs:** https://arxiv.org/abs/2602.03386 · **PDF:** https://arxiv.org/pdf/2602.03386v1
+
+## 16044. Dynamic Topology Optimization for Non-IID Data in Decentralized Learning
+
+- **ID:** `2602.03383` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Bart Cox, Antreas Ioannou, Jérémie Decouchant
+- **abs:** https://arxiv.org/abs/2602.03383 · **PDF:** https://arxiv.org/pdf/2602.03383v1
+
+## 16045. Rethinking Benign Relearning: Syntax as the Hidden Driver of Unlearning Failures
+
+- **ID:** `2602.03379` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Sangyeon Yoon, Hyesoo Hong, Wonje Jeung, et al.
+- **abs:** https://arxiv.org/abs/2602.03379 · **PDF:** https://arxiv.org/pdf/2602.03379v1
+
+## 16046. SLIM-Diff: Shared Latent Image-Mask Diffusion with Lp loss for Data-Scarce Epilepsy FLAIR MRI
+
+- **ID:** `2602.03372` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Mario Pascual-González, Ariadna Jiménez-Partinen, R. M. Luque-Baena, et al.
+- **abs:** https://arxiv.org/abs/2602.03372 · **PDF:** https://arxiv.org/pdf/2602.03372v1
+
+## 16047. Stochastic Spiking Neuron Based SNN Can be Inherently Bayesian
+
+- **ID:** `2602.07037` · **primary:** `cs.NE` · **Submitted:** 2026-02-03
+- **Authors:** Huannan Zheng, Jingli Liu, Kezhou Yang
+- **abs:** https://arxiv.org/abs/2602.07037 · **PDF:** https://arxiv.org/pdf/2602.07037v1
+
+## 16048. Symbol-Aware Reasoning with Masked Discrete Diffusion for Handwritten Mathematical Expression Recognition
+
+- **ID:** `2602.03370` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Takaya Kawakatsu, Ryo Ishiyama
+- **abs:** https://arxiv.org/abs/2602.03370 · **PDF:** https://arxiv.org/pdf/2602.03370v1
+
+## 16049. Pursuing Best Industrial Practices for Retrieval-Augmented Generation in the Medical Domain
+
+- **ID:** `2602.03368` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Liz Li, Wei Zhu
+- **abs:** https://arxiv.org/abs/2602.03368 · **PDF:** https://arxiv.org/pdf/2602.03368v2
+
+## 16050. MeKi: Memory-based Expert Knowledge Injection for Efficient LLM Scaling
+
+- **ID:** `2602.03359` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Ning Ding, Fangcheng Liu, Kyungrae Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.03359 · **PDF:** https://arxiv.org/pdf/2602.03359v1
+
+## 16051. GFlowPO: Generative Flow Network as a Language Model Prompt Optimizer
+
+- **ID:** `2602.03358` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Junmo Cho, Suhan Kim, Sangjune An, et al.
+- **abs:** https://arxiv.org/abs/2602.03358 · **PDF:** https://arxiv.org/pdf/2602.03358v1
+
+## 16052. Achieving Linear Speedup for Composite Federated Learning
+
+- **ID:** `2602.03357` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Kun Huang, Shi Pu
+- **abs:** https://arxiv.org/abs/2602.03357 · **PDF:** https://arxiv.org/pdf/2602.03357v1
+
+## 16053. PACE: Pretrained Audio Continual Learning
+
+- **ID:** `2602.03355` · **primary:** `cs.SD` · **Submitted:** 2026-02-03
+- **Authors:** Chang Li, Kanglei Zhou, Liyuan Wang
+- **abs:** https://arxiv.org/abs/2602.03355 · **PDF:** https://arxiv.org/pdf/2602.03355v1
+
+## 16054. Causal Graph Learning via Distributional Invariance of Cause-Effect Relationship
+
+- **ID:** `2602.03353` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Nang Hung Nguyen, Phi Le Nguyen, Thao Nguyen Truong, et al.
+- **abs:** https://arxiv.org/abs/2602.03353 · **PDF:** https://arxiv.org/pdf/2602.03353v1
+
+## 16055. PEGRL: Improving Machine Translation by Post-Editing Guided Reinforcement Learning
+
+- **ID:** `2602.03352` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Yunzhi Shen, Hao Zhou, Xin Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.03352 · **PDF:** https://arxiv.org/pdf/2602.03352v1
+
+## 16056. Building Interpretable Models for Moral Decision-Making
+
+- **ID:** `2602.03351` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Mayank Goel, Aritra Das, Paras Chopra
+- **abs:** https://arxiv.org/abs/2602.03351 · **PDF:** https://arxiv.org/pdf/2602.03351v2
+
+## 16057. MENASpeechBank: A Reference Voice Bank with Persona-Conditioned Multi-Turn Conversations for AudioLLMs
+
+- **ID:** `2602.07036` · **primary:** `cs.SD` · **Submitted:** 2026-02-03
+- **Authors:** Zien Sheikh Ali, Hunzalah Hassan Bhatti, Rabindra Nath Nandi, et al.
+- **abs:** https://arxiv.org/abs/2602.07036 · **PDF:** https://arxiv.org/pdf/2602.07036v1
+
+## 16058. Robustness as an Emergent Property of Task Performance
+
+- **ID:** `2602.03344` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Shir Ashury-Tahan, Ariel Gera, Elron Bandel, et al.
+- **abs:** https://arxiv.org/abs/2602.03344 · **PDF:** https://arxiv.org/pdf/2602.03344v1
+
+## 16059. All-Atom GPCR-Ligand Simulation via Residual Isometric Latent Flow
+
+- **ID:** `2602.03902` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-03
+- **Authors:** Jiying Zhang, Shuhao Zhang, Pierre Vandergheynst, et al.
+- **abs:** https://arxiv.org/abs/2602.03902 · **PDF:** https://arxiv.org/pdf/2602.03902v1
+
+## 16060. Tiled Prompts: Overcoming Prompt Underspecification in Image and Video Super-Resolution
+
+- **ID:** `2602.03342` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Bryan Sangwoo Kim, Jonghyun Park, Jong Chul Ye
+- **abs:** https://arxiv.org/abs/2602.03342 · **PDF:** https://arxiv.org/pdf/2602.03342v1
+
+## 16061. A feedback control optimizer for online and hardware-aware training of Spiking Neural Networks
+
+- **ID:** `2602.13261` · **primary:** `cs.NE` · **Submitted:** 2026-02-03
+- **Authors:** Matteo Saponati, Chiara De Luca, Giacomo Indiveri, et al.
+- **abs:** https://arxiv.org/abs/2602.13261 · **PDF:** https://arxiv.org/pdf/2602.13261v1
+
+## 16062. MentalSeek-Dx: Towards Progressive Hypothetico-Deductive Reasoning for Real-world Psychiatric Diagnosis
+
+- **ID:** `2602.03340` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Xiao Sun, Yuming Yang, Junnan Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.03340 · **PDF:** https://arxiv.org/pdf/2602.03340v1
+
+## 16063. Accurate Failure Prediction in Agents Does Not Imply Effective Failure Prevention
+
+- **ID:** `2602.03338` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Rakshith Vasudev, Melisa Russak, Dan Bikel, et al.
+- **abs:** https://arxiv.org/abs/2602.03338 · **PDF:** https://arxiv.org/pdf/2602.03338v1
+
+## 16064. NeuroPareto: Calibrated Acquisition for Costly Many-Goal Search in Vast Parameter Spaces
+
+- **ID:** `2602.03901` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Rong Fu, Chunlei Meng, Youjin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.03901 · **PDF:** https://arxiv.org/pdf/2602.03901v3
+
+## 16065. Bayesian Conformal Prediction as a Decision Risk Problem
+
+- **ID:** `2602.03331` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Fanyi Wu, Veronika Lohmanova, Samuel Kaski, et al.
+- **abs:** https://arxiv.org/abs/2602.03331 · **PDF:** https://arxiv.org/pdf/2602.03331v1
+
+## 16066. From Inexact Gradients to Byzantine Robustness: Acceleration and Optimization under Similarity
+
+- **ID:** `2602.03329` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Renaud Gaucher, Aymeric Dieuleveut, Hadrien Hendrikx
+- **abs:** https://arxiv.org/abs/2602.03329 · **PDF:** https://arxiv.org/pdf/2602.03329v1
+
+## 16067. A Novel approach to portfolio construction
+
+- **ID:** `2602.03325` · **primary:** `q-fin.PM` · **Submitted:** 2026-02-03
+- **Authors:** T. Di Matteo, L. Riso, M. G. Zoia
+- **abs:** https://arxiv.org/abs/2602.03325 · **PDF:** https://arxiv.org/pdf/2602.03325v1
+
+## 16068. Unpacking Human Preference for LLMs: Demographically Aware Evaluation with the HUMAINE Framework
+
+- **ID:** `2603.04409` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Nora Petrova, Andrew Gordon, Enzo Blindow
+- **abs:** https://arxiv.org/abs/2603.04409 · **PDF:** https://arxiv.org/pdf/2603.04409v1
+
+## 16069. MedSAM-Agent: Empowering Interactive Medical Image Segmentation with Multi-turn Agentic Reinforcement Learning
+
+- **ID:** `2602.03320` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Shengyuan Liu, Liuxin Bao, Qi Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.03320 · **PDF:** https://arxiv.org/pdf/2602.03320v1
+
+## 16070. Information-Theoretic Multi-Model Fusion for Target-Oriented Adaptive Sampling in Materials Design
+
+- **ID:** `2602.03319` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yixuan Zhang, Zhiyuan Li, Weijia He, et al.
+- **abs:** https://arxiv.org/abs/2602.03319 · **PDF:** https://arxiv.org/pdf/2602.03319v1
+
+## 16071. Knowledge Model Prompting Increases LLM Performance on Planning Tasks
+
+- **ID:** `2602.03900` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Erik Goh, John Kos, Ashok Goel
+- **abs:** https://arxiv.org/abs/2602.03900 · **PDF:** https://arxiv.org/pdf/2602.03900v1
+
+## 16072. MIRROR: A Multi-Agent Framework with Iterative Adaptive Revision and Hierarchical Retrieval for Optimization Modeling in Operations Research
+
+- **ID:** `2602.03318` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Yifan Shi, Jialong Shi, Jiayi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.03318 · **PDF:** https://arxiv.org/pdf/2602.03318v2
+
+## 16073. Multiparameter Uncertainty Mapping in Quantitative Molecular MRI using a Physics-Structured Variational Autoencoder (PS-VAE)
+
+- **ID:** `2602.03317` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Alex Finkelstein, Ron Moneta, Or Zohar, et al.
+- **abs:** https://arxiv.org/abs/2602.03317 · **PDF:** https://arxiv.org/pdf/2602.03317v1
+
+## 16074. Memora: A Harmonic Memory Representation Balancing Abstraction and Specificity
+
+- **ID:** `2602.03315` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Menglin Xia, Xuchao Zhang, Shantanu Dixit, et al.
+- **abs:** https://arxiv.org/abs/2602.03315 · **PDF:** https://arxiv.org/pdf/2602.03315v1
+
+## 16075. RDT2: Exploring the Scaling Limit of UMI Data Towards Zero-Shot Cross-Embodiment Generalization
+
+- **ID:** `2602.03310` · **primary:** `cs.RO` · **Submitted:** 2026-02-03
+- **Authors:** Songming Liu, Bangguo Li, Kai Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.03310 · **PDF:** https://arxiv.org/pdf/2602.03310v1
+
+## 16076. Entropy-Gated Selective Policy Optimization:Token-Level Gradient Allocation for Hybrid Training of Large Language Models
+
+- **ID:** `2602.03309` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yuelin Hu, Zhengxue Cheng, Wei Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.03309 · **PDF:** https://arxiv.org/pdf/2602.03309v1
+
+## 16077. Learning to Select: Query-Aware Adaptive Dimension Selection for Dense Retrieval
+
+- **ID:** `2602.03306` · **primary:** `cs.IR` · **Submitted:** 2026-02-03
+- **Authors:** Zhanyu Wu, Richong Zhang, Zhijie Nie
+- **abs:** https://arxiv.org/abs/2602.03306 · **PDF:** https://arxiv.org/pdf/2602.03306v2
+
+## 16078. medR: Reward Engineering for Clinical Offline Reinforcement Learning via Tri-Drive Potential Functions
+
+- **ID:** `2602.03305` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Qianyi Xu, Gousia Habib, Feng Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.03305 · **PDF:** https://arxiv.org/pdf/2602.03305v2
+
+## 16079. Full end-to-end diagnostic workflow automation of 3D OCT via foundation model-driven AI for retinal diseases
+
+- **ID:** `2602.03302` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Jinze Zhang, Jian Zhong, Li Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.03302 · **PDF:** https://arxiv.org/pdf/2602.03302v1
+
+## 16080. Periodic Regularized Q-Learning
+
+- **ID:** `2602.03301` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Hyukjun Yang, Han-Dong Lim, Donghwan Lee
+- **abs:** https://arxiv.org/abs/2602.03301 · **PDF:** https://arxiv.org/pdf/2602.03301v1
+
+## 16081. Privacy Amplification Persists under Unlimited Synthetic Data Release
+
+- **ID:** `2602.04895` · **primary:** `cs.CR` · **Submitted:** 2026-02-03
+- **Authors:** Clément Pierquin, Aurélien Bellet, Marc Tommasi, et al.
+- **abs:** https://arxiv.org/abs/2602.04895 · **PDF:** https://arxiv.org/pdf/2602.04895v1
+
+## 16082. R1-SyntheticVL: Is Synthetic Data from Generative Models Ready for Multimodal Large Language Model?
+
+- **ID:** `2602.03300` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Jingyi Zhang, Tianyi Lin, Huanjin Yao, et al.
+- **abs:** https://arxiv.org/abs/2602.03300 · **PDF:** https://arxiv.org/pdf/2602.03300v1
+
+## 16083. Lipschitz Multiscale Deep Equilibrium Models: A Theoretically Guaranteed and Accelerated Approach
+
+- **ID:** `2602.03297` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Naoki Sato, Hideaki Iiduka
+- **abs:** https://arxiv.org/abs/2602.03297 · **PDF:** https://arxiv.org/pdf/2602.03297v1
+
+## 16084. POP: Prefill-Only Pruning for Efficient Large Model Inference
+
+- **ID:** `2602.03295` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Junhui He, Zhihui Fu, Jun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.03295 · **PDF:** https://arxiv.org/pdf/2602.03295v1
+
+## 16085. Anomaly Detection via Mean Shift Density Enhancement
+
+- **ID:** `2602.03293` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Pritam Kar, Rahul Bordoloi, Olaf Wolkenhauer, et al.
+- **abs:** https://arxiv.org/abs/2602.03293 · **PDF:** https://arxiv.org/pdf/2602.03293v1
+
+## 16086. Universal Approximation of Continuous Functionals on Compact Subsets via Linear Measurements and Scalar Nonlinearities
+
+- **ID:** `2602.03290` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Andrey Krylov, Maksim Penkin
+- **abs:** https://arxiv.org/abs/2602.03290 · **PDF:** https://arxiv.org/pdf/2602.03290v1
+
+## 16087. DLLM-Searcher: Adapting Diffusion Large Language Model for Search Agents
+
+- **ID:** `2602.07035` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Jiahao Zhao, Shaoxuan Xu, Zhongxiang Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.07035 · **PDF:** https://arxiv.org/pdf/2602.07035v1
+
+## 16088. Rejecting Arguments Based on Doubt in Structured Bipolar Argumentation
+
+- **ID:** `2602.03286` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Michael A. Müller, Srdjan Vesic, Bruno Yun
+- **abs:** https://arxiv.org/abs/2602.03286 · **PDF:** https://arxiv.org/pdf/2602.03286v1
+
+## 16089. MeetBench-XL: Calibrated Multi-Dimensional Evaluation and Learned Dual-Policy Agents for Real-Time Meetings
+
+- **ID:** `2602.03285` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Yuelin Hu, Jun Xu, Bingcong Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.03285 · **PDF:** https://arxiv.org/pdf/2602.03285v1
+
+## 16090. Byzantine Machine Learning: MultiKrum and an optimal notion of robustness
+
+- **ID:** `2602.03899` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Gilles Bareilles, Wassim Bouaziz, Julien Fageot, et al.
+- **abs:** https://arxiv.org/abs/2602.03899 · **PDF:** https://arxiv.org/pdf/2602.03899v1
+
+## 16091. ST-Raptor: An Agentic System for Semi-Structured Table QA
+
+- **ID:** `2602.07034` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Jinxiu Qu, Zirui Tang, Hongzhang Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.07034 · **PDF:** https://arxiv.org/pdf/2602.07034v1
+
+## 16092. Orthogonal Approximate Message Passing Algorithms for Rectangular Spiked Matrix Models with Rotationally Invariant Noise
+
+- **ID:** `2602.03283` · **primary:** `math.ST` · **Submitted:** 2026-02-03
+- **Authors:** Haohua Chen, Songbin Liu, Junjie Ma
+- **abs:** https://arxiv.org/abs/2602.03283 · **PDF:** https://arxiv.org/pdf/2602.03283v1
+
+## 16093. Global Geometry Is Not Enough for Vision Representations
+
+- **ID:** `2602.03282` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Jiwan Chung, Seon Joo Kim
+- **abs:** https://arxiv.org/abs/2602.03282 · **PDF:** https://arxiv.org/pdf/2602.03282v1
+
+## 16094. Agentic Proposing: Enhancing Large Language Model Reasoning via Compositional Skill Synthesis
+
+- **ID:** `2602.03279` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Zhengbo Jiao, Shaobo Wang, Zifan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03279 · **PDF:** https://arxiv.org/pdf/2602.03279v1
+
+## 16095. BlockRR: A Unified Framework of RR-type Algorithms for Label Differential Privacy
+
+- **ID:** `2602.03277` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Haixia Liu, Yi Ding
+- **abs:** https://arxiv.org/abs/2602.03277 · **PDF:** https://arxiv.org/pdf/2602.03277v1
+
+## 16096. Probing Memes in LLMs: A Paradigm for the Entangled Evaluation World
+
+- **ID:** `2603.04408` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Luzhou Peng, Zhengxin Yang, Honglu Ji, et al.
+- **abs:** https://arxiv.org/abs/2603.04408 · **PDF:** https://arxiv.org/pdf/2603.04408v1
+
+## 16097. Unveiling Covert Toxicity in Multimodal Data via Toxicity Association Graphs: A Graph-Based Metric and Interpretable Detection Framework
+
+- **ID:** `2602.03268` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Guanzong Wu, Zihao Zhu, Siwei Lyu, et al.
+- **abs:** https://arxiv.org/abs/2602.03268 · **PDF:** https://arxiv.org/pdf/2602.03268v1
+
+## 16098. Beyond Suffixes: Token Position in GCG Adversarial Attacks on Large Language Models
+
+- **ID:** `2602.03265` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Hicham Eddoubi, Umar Faruk Abdullahi, Fadi Hassan
+- **abs:** https://arxiv.org/abs/2602.03265 · **PDF:** https://arxiv.org/pdf/2602.03265v1
+
+## 16099. HypCBC: Domain-Invariant Hyperbolic Cross-Branch Consistency for Generalizable Medical Image Analysis
+
+- **ID:** `2602.03264` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Francesco Di Salvo, Sebastian Doerrich, Jonas Alle, et al.
+- **abs:** https://arxiv.org/abs/2602.03264 · **PDF:** https://arxiv.org/pdf/2602.03264v1
+
+## 16100. CSR-Bench: A Benchmark for Evaluating the Cross-modal Safety and Reliability of MLLMs
+
+- **ID:** `2602.03263` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Yuxuan Liu, Yuntian Shi, Kun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.03263 · **PDF:** https://arxiv.org/pdf/2602.03263v1
+
+## 16101. A Hitchhiker's Guide to Poisson Gradient Estimation
+
+- **ID:** `2602.03896` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Michael Ibrahim, Hanqi Zhao, Eli Sennesh, et al.
+- **abs:** https://arxiv.org/abs/2602.03896 · **PDF:** https://arxiv.org/pdf/2602.03896v1
+
+## 16102. BamaER: A Behavior-Aware Memory-Augmented Model for Exercise Recommendation
+
+- **ID:** `2602.15879` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Qing Yang, Yuhao Jiang, Rui Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.15879 · **PDF:** https://arxiv.org/pdf/2602.15879v1
+
+## 16103. Principled Federated Random Forests for Heterogeneous Data
+
+- **ID:** `2602.03258` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Rémi Khellaf, Erwan Scornet, Aurélien Bellet, et al.
+- **abs:** https://arxiv.org/abs/2602.03258 · **PDF:** https://arxiv.org/pdf/2602.03258v1
+
+## 16104. GraDE: A Graph Diffusion Estimator for Frequent Subgraph Discovery in Neural Architectures
+
+- **ID:** `2602.03257` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yikang Yang, Zhengxin Yang, Minghao Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.03257 · **PDF:** https://arxiv.org/pdf/2602.03257v1
+
+## 16105. LPS-Bench: Benchmarking Safety Awareness of Computer-Use Agents in Long-Horizon Planning under Benign and Adversarial Scenarios
+
+- **ID:** `2602.03255` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Tianyu Chen, Chujia Hu, Ge Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.03255 · **PDF:** https://arxiv.org/pdf/2602.03255v1
+
+## 16106. Benchmarking Bias Mitigation Toward Fairness Without Harm from Vision to LVLMs
+
+- **ID:** `2602.03895` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Xuwei Tan, Ziyu Hu, Xueru Zhang
+- **abs:** https://arxiv.org/abs/2602.03895 · **PDF:** https://arxiv.org/pdf/2602.03895v1
+
+## 16107. Accordion-Thinking: Self-Regulated Step Summaries for Efficient and Readable LLM Reasoning
+
+- **ID:** `2602.03249` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Zhicheng Yang, Zhijiang Guo, Yinya Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.03249 · **PDF:** https://arxiv.org/pdf/2602.03249v1
+
+## 16108. Vision Transformers for Zero-Shot Clustering of Animal Images: A Comparative Benchmarking Study
+
+- **ID:** `2602.03894` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Hugo Markoff, Stefan Hein Bengtson, Michael Ørsted
+- **abs:** https://arxiv.org/abs/2602.03894 · **PDF:** https://arxiv.org/pdf/2602.03894v1
+
+## 16109. Mići Princ -- A Little Boy Teaching Speech Technologies the Chakavian Dialect
+
+- **ID:** `2602.03245` · **primary:** `eess.AS` · **Submitted:** 2026-02-03
+- **Authors:** Nikola Ljubešić, Peter Rupnik, Tea Perinčić
+- **abs:** https://arxiv.org/abs/2602.03245 · **PDF:** https://arxiv.org/pdf/2602.03245v1
+
+## 16110. The Necessity of a Unified Framework for LLM-Based Agent Evaluation
+
+- **ID:** `2602.03238` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Pengyu Zhu, Li Sun, Philip S. Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.03238 · **PDF:** https://arxiv.org/pdf/2602.03238v1
+
+## 16111. Merging Beyond: Streaming LLM Updates via Activation-Guided Rotations
+
+- **ID:** `2602.03237` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yuxuan Yao, Haonan Sheng, Qingsong Lv, et al.
+- **abs:** https://arxiv.org/abs/2602.03237 · **PDF:** https://arxiv.org/pdf/2602.03237v1
+
+## 16112. BayeSQP: Bayesian Optimization through Sequential Quadratic Programming
+
+- **ID:** `2602.03232` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Paul Brunzema, Sebastian Trimpe
+- **abs:** https://arxiv.org/abs/2602.03232 · **PDF:** https://arxiv.org/pdf/2602.03232v1
+
+## 16113. IT-OSE: Exploring Optimal Sample Size for Industrial Data Augmentation
+
+- **ID:** `2602.15878` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Mingchun Sun, Rongqiang Zhao, Zhennan Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.15878 · **PDF:** https://arxiv.org/pdf/2602.15878v1
+
+## 16114. ATACompressor: Adaptive Task-Aware Compression for Efficient Long-Context Processing in LLMs
+
+- **ID:** `2602.03226` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Xuancheng Li, Haitao Li, Yujia Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.03226 · **PDF:** https://arxiv.org/pdf/2602.03226v1
+
+## 16115. TAME: A Trustworthy Test-Time Evolution of Agent Memory with Systematic Benchmarking
+
+- **ID:** `2602.03224` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Yu Cheng, Jiuan Zhou, Yongkang Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.03224 · **PDF:** https://arxiv.org/pdf/2602.03224v1
+
+## 16116. Distribution-Aware End-to-End Embedding for Streaming Numerical Features in Click-Through Rate Prediction
+
+- **ID:** `2602.03223` · **primary:** `cs.IR` · **Submitted:** 2026-02-03
+- **Authors:** Jiahao Liu, Hongji Ruan, Weimin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.03223 · **PDF:** https://arxiv.org/pdf/2602.03223v1
+
+## 16117. Audit After Segmentation: Reference-Free Mask Quality Assessment for Language-Referred Audio-Visual Segmentation
+
+- **ID:** `2602.03892` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Jinxing Zhou, Yanghao Zhou, Yaoting Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.03892 · **PDF:** https://arxiv.org/pdf/2602.03892v1
+
+## 16118. Beyond Quantity: Trajectory Diversity Scaling for Code Agents
+
+- **ID:** `2602.03219` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Guhong Chen, Chenghao Sun, Cheng Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.03219 · **PDF:** https://arxiv.org/pdf/2602.03219v2
+
+## 16119. Topology Matters: A Cautionary Case Study of Graph SSL on Neuro-Inspired Benchmarks
+
+- **ID:** `2602.03217` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** May Kristine Jonson Carlon, Su Myat Noe, Haojiong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.03217 · **PDF:** https://arxiv.org/pdf/2602.03217v1
+
+## 16120. Sounding Highlights: Dual-Pathway Audio Encoders for Audio-Visual Video Highlight Detection
+
+- **ID:** `2602.03891` · **primary:** `eess.AS` · **Submitted:** 2026-02-03
+- **Authors:** Seohyun Joo, Yoori Oh
+- **abs:** https://arxiv.org/abs/2602.03891 · **PDF:** https://arxiv.org/pdf/2602.03891v2
+
+## 16121. Token Sparse Attention: Efficient Long-Context Inference with Interleaved Token Selection
+
+- **ID:** `2602.03216` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Dongwon Jo, Beomseok Kang, Jiwon Song, et al.
+- **abs:** https://arxiv.org/abs/2602.03216 · **PDF:** https://arxiv.org/pdf/2602.03216v1
+
+## 16122. Latent Neural-ODE for Model-Informed Precision Dosing: Overcoming Structural Assumptions in Pharmacokinetics
+
+- **ID:** `2602.03215` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Benjamin Maurel, Agathe Guilloux, Sarah Zohar, et al.
+- **abs:** https://arxiv.org/abs/2602.03215 · **PDF:** https://arxiv.org/pdf/2602.03215v1
+
+## 16123. Lookahead Sample Reward Guidance for Test-Time Scaling of Diffusion Models
+
+- **ID:** `2602.03211` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yeongmin Kim, Donghyeok Shin, Byeonghu Na, et al.
+- **abs:** https://arxiv.org/abs/2602.03211 · **PDF:** https://arxiv.org/pdf/2602.03211v1
+
+## 16124. Spectral Evolution Search: Efficient Inference-Time Scaling for Reward-Aligned Image Generation
+
+- **ID:** `2602.03208` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Jinyan Ye, Zhongjie Duan, Zhiwen Li, et al.
+- **abs:** https://arxiv.org/abs/2602.03208 · **PDF:** https://arxiv.org/pdf/2602.03208v1
+
+## 16125. Sparsity is Combinatorial Depth: Quantifying MoE Expressivity via Tropical Geometry
+
+- **ID:** `2602.03204` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Ye Su, Huayi Tang, Zixuan Gong, et al.
+- **abs:** https://arxiv.org/abs/2602.03204 · **PDF:** https://arxiv.org/pdf/2602.03204v1
+
+## 16126. ForesightKV: Optimizing KV Cache Eviction for Reasoning Models by Learning Long-Term Contribution
+
+- **ID:** `2602.03203` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Zican Dong, Peiyu Liu, Junyi Li, et al.
+- **abs:** https://arxiv.org/abs/2602.03203 · **PDF:** https://arxiv.org/pdf/2602.03203v1
+
+## 16127. Sharp Inequalities between Total Variation and Hellinger Distances for Gaussian Mixtures
+
+- **ID:** `2602.03202` · **primary:** `math.ST` · **Submitted:** 2026-02-03
+- **Authors:** Joonhyuk Jung, Chao Gao
+- **abs:** https://arxiv.org/abs/2602.03202 · **PDF:** https://arxiv.org/pdf/2602.03202v1
+
+## 16128. From Scalar Rewards to Potential Trends: Shaping Potential Landscapes for Model-Based Reinforcement Learning
+
+- **ID:** `2602.03201` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yao-Hui Li, Zeyu Wang, Xin Li, et al.
+- **abs:** https://arxiv.org/abs/2602.03201 · **PDF:** https://arxiv.org/pdf/2602.03201v2
+
+## 16129. Hand3R: Online 4D Hand-Scene Reconstruction in the Wild
+
+- **ID:** `2602.03200` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Wendi Hu, Haonan Zhou, Wenhao Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.03200 · **PDF:** https://arxiv.org/pdf/2602.03200v1
+
+## 16130. Reinforcement Learning with Promising Tokens for Large Language Models
+
+- **ID:** `2602.03195` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Jing-Cheng Pang, Liang Lu, Xian Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.03195 · **PDF:** https://arxiv.org/pdf/2602.03195v2
+
+## 16131. Prompt Augmentation Scales up GRPO Training on Mathematical Reasoning
+
+- **ID:** `2602.03190` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Wenquan Lu, Hai Huang, Randall Balestriero
+- **abs:** https://arxiv.org/abs/2602.03190 · **PDF:** https://arxiv.org/pdf/2602.03190v2
+
+## 16132. DynSplit-KV: Dynamic Semantic Splitting for KVCache Compression in Efficient Long-Context LLM Inference
+
+- **ID:** `2602.03184` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Jiancai Ye, Jun Liu, Qingchen Li, et al.
+- **abs:** https://arxiv.org/abs/2602.03184 · **PDF:** https://arxiv.org/pdf/2602.03184v1
+
+## 16133. Privasis: Synthesizing the Largest "Public" Private Dataset from Scratch
+
+- **ID:** `2602.03183` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Hyunwoo Kim, Niloofar Mireshghallah, Michael Duan, et al.
+- **abs:** https://arxiv.org/abs/2602.03183 · **PDF:** https://arxiv.org/pdf/2602.03183v1
+
+## 16134. Probe-then-Commit Multi-Objective Bandits: Theoretical Benefits of Limited Multi-Arm Feedback
+
+- **ID:** `2602.03175` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Ming Shi
+- **abs:** https://arxiv.org/abs/2602.03175 · **PDF:** https://arxiv.org/pdf/2602.03175v2
+
+## 16135. Adversarial construction as a potential solution to the experiment design problem in large task spaces
+
+- **ID:** `2602.03172` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Prakhar Godara, Frederick Callaway, Marcelo G. Mattar
+- **abs:** https://arxiv.org/abs/2602.03172 · **PDF:** https://arxiv.org/pdf/2602.03172v1
+
+## 16136. StepScorer: Accelerating Reinforcement Learning with Step-wise Scoring and Psychological Regret Modeling
+
+- **ID:** `2602.03171` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Zhe Xu
+- **abs:** https://arxiv.org/abs/2602.03171 · **PDF:** https://arxiv.org/pdf/2602.03171v1
+
+## 16137. NeuralFLoC: Neural Flow-Based Joint Registration and Clustering of Functional Data
+
+- **ID:** `2602.03169` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Xinyang Xiong, Siyuan jiang, Pengcheng Zeng
+- **abs:** https://arxiv.org/abs/2602.03169 · **PDF:** https://arxiv.org/pdf/2602.03169v1
+
+## 16138. Online Conformal Prediction via Universal Portfolio Algorithms
+
+- **ID:** `2602.03168` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Tuo Liu, Edgar Dobriban, Francesco Orabona
+- **abs:** https://arxiv.org/abs/2602.03168 · **PDF:** https://arxiv.org/pdf/2602.03168v1
+
+## 16139. Entropic Mirror Monte Carlo
+
+- **ID:** `2602.03165` · **primary:** `stat.ME` · **Submitted:** 2026-02-03
+- **Authors:** Anas Cherradi, Yazid Janati, Alain Durmus, et al.
+- **abs:** https://arxiv.org/abs/2602.03165 · **PDF:** https://arxiv.org/pdf/2602.03165v1
+
+## 16140. MemCast: Memory-Driven Time Series Forecasting with Experience-Conditioned Reasoning
+
+- **ID:** `2602.03164` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Xiaoyu Tao, Mingyue Cheng, Ze Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.03164 · **PDF:** https://arxiv.org/pdf/2602.03164v1
+
+## 16141. VALUEFLOW: Toward Pluralistic and Steerable Value-based Alignment in Large Language Models
+
+- **ID:** `2602.03160` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Woojin Kim, Sieun Hyeon, Jusang Oh, et al.
+- **abs:** https://arxiv.org/abs/2602.03160 · **PDF:** https://arxiv.org/pdf/2602.03160v1
+
+## 16142. Fully Kolmogorov-Arnold Deep Model in Medical Image Segmentation
+
+- **ID:** `2602.03156` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Xingyu Qiu, Xinghua Ma, Dong Liang, et al.
+- **abs:** https://arxiv.org/abs/2602.03156 · **PDF:** https://arxiv.org/pdf/2602.03156v1
+
+## 16143. Intelligent Front-End Personalization: AI-Driven UI Adaptation
+
+- **ID:** `2602.03154` · **primary:** `cs.HC` · **Submitted:** 2026-02-03
+- **Authors:** Mona Rajhans
+- **abs:** https://arxiv.org/abs/2602.03154 · **PDF:** https://arxiv.org/pdf/2602.03154v1
+
+## 16144. FASA: Frequency-aware Sparse Attention
+
+- **ID:** `2602.03152` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Yifei Wang, Yueqi Wang, Zhenrui Yue, et al.
+- **abs:** https://arxiv.org/abs/2602.03152 · **PDF:** https://arxiv.org/pdf/2602.03152v3
+
+## 16145. Enhancing Foundation VLM Robustness to Missing Modality: Scalable Diffusion for Bi-directional Feature Restoration
+
+- **ID:** `2602.03151` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Wei Dai, Haoyu Wang, Honghao Chang, et al.
+- **abs:** https://arxiv.org/abs/2602.03151 · **PDF:** https://arxiv.org/pdf/2602.03151v1
+
+## 16146. General Agents Contain World Models, even under Partial Observability and Stochasticity
+
+- **ID:** `2602.03146` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Santiago Cifuentes
+- **abs:** https://arxiv.org/abs/2602.03146 · **PDF:** https://arxiv.org/pdf/2602.03146v1
+
+## 16147. Internet of Agentic AI: Incentive-Compatible Distributed Teaming and Workflow
+
+- **ID:** `2602.03145` · **primary:** `cs.GT` · **Submitted:** 2026-02-03
+- **Authors:** Ya-Ting Yang, Quanyan Zhu
+- **abs:** https://arxiv.org/abs/2602.03145 · **PDF:** https://arxiv.org/pdf/2602.03145v1
+
+## 16148. What Makes a Good Example? Modeling Exemplar Selection with Neural Network Representations
+
+- **ID:** `2602.03144` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Fanxiao Wani Qiu, Oscar Leong, Alexander LaTourrette
+- **abs:** https://arxiv.org/abs/2602.03144 · **PDF:** https://arxiv.org/pdf/2602.03144v1
+
+## 16149. Self-Hinting Language Models Enhance Reinforcement Learning
+
+- **ID:** `2602.03143` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Baohao Liao, Hanze Dong, Xinxing Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.03143 · **PDF:** https://arxiv.org/pdf/2602.03143v1
+
+## 16150. Short Chains, Deep Thoughts: Balancing Reasoning Efficiency and Intra-Segment Capability via Split-Merge Optimization
+
+- **ID:** `2602.03141` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Runquan Gui, Jie Wang, Zhihai Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.03141 · **PDF:** https://arxiv.org/pdf/2602.03141v2
+
+## 16151. SATORIS-N: Spectral Analysis based Traffic Observation Recovery via Informed Subspaces and Nuclear-norm minimization
+
+- **ID:** `2602.03138` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Sampad Mohanty, Bhaskar Krishnamachari
+- **abs:** https://arxiv.org/abs/2602.03138 · **PDF:** https://arxiv.org/pdf/2602.03138v1
+
+## 16152. Enhanced Parcel Arrival Forecasting for Logistic Hubs: An Ensemble Deep Learning Approach
+
+- **ID:** `2602.03135` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Xinyue Pan, Yujia Xu, Benoit Montreuil
+- **abs:** https://arxiv.org/abs/2602.03135 · **PDF:** https://arxiv.org/pdf/2602.03135v1
+
+## 16153. SwiftVLM: Efficient Vision-Language Model Inference via Cross-Layer Token Bypass
+
+- **ID:** `2602.03134` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Chen Qian, Xinran Yu, Danyang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.03134 · **PDF:** https://arxiv.org/pdf/2602.03134v1
+
+## 16154. Contrastive Concept-Tree Search for LLM-Assisted Algorithm Discovery
+
+- **ID:** `2602.03132` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Timothee Leleu, Sudeera Gunathilaka, Federico Ghimenti, et al.
+- **abs:** https://arxiv.org/abs/2602.03132 · **PDF:** https://arxiv.org/pdf/2602.03132v1
+
+## 16155. Understanding Multi-Agent LLM Frameworks: A Unified Benchmark and Experimental Analysis
+
+- **ID:** `2602.03128` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Abdelghny Orogat, Ana Rostam, Essam Mansour
+- **abs:** https://arxiv.org/abs/2602.03128 · **PDF:** https://arxiv.org/pdf/2602.03128v1
+
+## 16156. Feature, Alignment, and Supervision in Category Learning: A Comparative Approach with Children and Neural Networks
+
+- **ID:** `2602.03124` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Fanxiao Wani Qiu, Oscar Leong
+- **abs:** https://arxiv.org/abs/2602.03124 · **PDF:** https://arxiv.org/pdf/2602.03124v1
+
+## 16157. Beyond Cropping and Rotation: Automated Evolution of Powerful Task-Specific Augmentations with Generative Models
+
+- **ID:** `2602.03123` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Judah Goldfeder, Shreyes Kaliyur, Vaibhav Sourirajan, et al.
+- **abs:** https://arxiv.org/abs/2602.03123 · **PDF:** https://arxiv.org/pdf/2602.03123v1
+
+## 16158. TransConv-DDPM: Enhanced Diffusion Model for Generating Time-Series Data in Healthcare
+
+- **ID:** `2602.07033` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Md Shahriar Kabir, Sana Alamgeer, Minakshi Debnath, et al.
+- **abs:** https://arxiv.org/abs/2602.07033 · **PDF:** https://arxiv.org/pdf/2602.07033v1
+
+## 16159. Quantized Evolution Strategies: High-precision Fine-tuning of Quantized LLMs at Low-precision Cost
+
+- **ID:** `2602.03120` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yinggan Xu, Risto Miikkulainen, Xin Qiu
+- **abs:** https://arxiv.org/abs/2602.03120 · **PDF:** https://arxiv.org/pdf/2602.03120v1
+
+## 16160. Function-Space Empirical Bayes Regularisation with Large Vision-Language Model Priors
+
+- **ID:** `2602.03119` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Pengcheng Hao, Huaze Tang, Ercan Engin Kuruoglu, et al.
+- **abs:** https://arxiv.org/abs/2602.03119 · **PDF:** https://arxiv.org/pdf/2602.03119v1
+
+## 16161. Digital Lifelong Learning in the Age of AI: Trends and Insights
+
+- **ID:** `2602.03114` · **primary:** `cs.CY` · **Submitted:** 2026-02-03
+- **Authors:** Geeta Puri, Nachamma Socklingam, Dorien Herremans
+- **abs:** https://arxiv.org/abs/2602.03114 · **PDF:** https://arxiv.org/pdf/2602.03114v1
+
+## 16162. Transcendental Regularization of Finite Mixtures:Theoretical Guarantees and Practical Limitations
+
+- **ID:** `2602.03889` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Ernest Fokoué
+- **abs:** https://arxiv.org/abs/2602.03889 · **PDF:** https://arxiv.org/pdf/2602.03889v1
+
+## 16163. One Model, All Roles: Multi-Turn, Multi-Agent Self-Play Reinforcement Learning for Conversational Social Intelligence
+
+- **ID:** `2602.03109` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Bowen Jiang, Taiwei Shi, Ryo Kamoi, et al.
+- **abs:** https://arxiv.org/abs/2602.03109 · **PDF:** https://arxiv.org/pdf/2602.03109v1
+
+## 16164. ChemPro: A Progressive Chemistry Benchmark for Large Language Models
+
+- **ID:** `2602.03108` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Aaditya Baranwal, Shruti Vyas
+- **abs:** https://arxiv.org/abs/2602.03108 · **PDF:** https://arxiv.org/pdf/2602.03108v2
+
+## 16165. The Mask of Civility: Benchmarking Chinese Mock Politeness Comprehension in Large Language Models
+
+- **ID:** `2602.03107` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Yitong Zhang, Yuhan Xiang, Mingxuan Liu
+- **abs:** https://arxiv.org/abs/2602.03107 · **PDF:** https://arxiv.org/pdf/2602.03107v1
+
+## 16166. "I'm happy even though it's not real": GenAI Photo Editing as a Remembering Experience
+
+- **ID:** `2602.03104` · **primary:** `cs.HC` · **Submitted:** 2026-02-03
+- **Authors:** Yufeng Wu, Qing Li, Elise van den Hoven, et al.
+- **abs:** https://arxiv.org/abs/2602.03104 · **PDF:** https://arxiv.org/pdf/2602.03104v2
+
+## 16167. Task--Specificity Score: Measuring How Much Instructions Really Matter for Supervision
+
+- **ID:** `2602.03103` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Pritam Kadasi, Abhishek Upperwal, Mayank Singh
+- **abs:** https://arxiv.org/abs/2602.03103 · **PDF:** https://arxiv.org/pdf/2602.03103v1
+
+## 16168. Consensus Group Relative Policy Optimization for Text Generation
+
+- **ID:** `2602.03102` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yuki Ichihara, Yuu Jinnai, Kaito Ariu, et al.
+- **abs:** https://arxiv.org/abs/2602.03102 · **PDF:** https://arxiv.org/pdf/2602.03102v1
+
+## 16169. LLM-FSM: Scaling Large Language Models for Finite-State Reasoning in RTL Code Generation
+
+- **ID:** `2602.07032` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Yuheng Wu, Berk Gokmen, Zhouhua Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.07032 · **PDF:** https://arxiv.org/pdf/2602.07032v1
+
+## 16170. Risky-Bench: Probing Agentic Safety Risks under Real-World Deployment
+
+- **ID:** `2602.03100` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Jingnan Zheng, Yanzhen Luo, Jingjun Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.03100 · **PDF:** https://arxiv.org/pdf/2602.03100v1
+
+## 16171. TextME: Bridging Unseen Modalities Through Text Descriptions
+
+- **ID:** `2602.03098` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Soyeon Hong, Jinchan Kim, Jaegook You, et al.
+- **abs:** https://arxiv.org/abs/2602.03098 · **PDF:** https://arxiv.org/pdf/2602.03098v2
+
+## 16172. De-conflating Preference and Qualification: Constrained Dual-Perspective Reasoning for Job Recommendation with Large Language Models
+
+- **ID:** `2602.03097` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Bryce Kan, Wei Yang, Emily Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.03097 · **PDF:** https://arxiv.org/pdf/2602.03097v1
+
+## 16173. PRISM: Structured Optimization via Anisotropic Spectral Shaping
+
+- **ID:** `2602.03096` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yujie Yang
+- **abs:** https://arxiv.org/abs/2602.03096 · **PDF:** https://arxiv.org/pdf/2602.03096v1
+
+## 16174. Test-time Recursive Thinking: Self-Improvement without External Feedback
+
+- **ID:** `2602.03094` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Yufan Zhuang, Chandan Singh, Liyuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.03094 · **PDF:** https://arxiv.org/pdf/2602.03094v1
+
+## 16175. Developing a Multi-Agent System to Generate Next Generation Science Assessments with Evidence-Centered Design
+
+- **ID:** `2602.18451` · **primary:** `cs.CY` · **Submitted:** 2026-02-03
+- **Authors:** Yaxuan Yang, Jongchan Park, Yifan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.18451 · **PDF:** https://arxiv.org/pdf/2602.18451v1
+
+## 16176. Lagged backward-compatible physics-informed neural networks for unsaturated soil consolidation analysis
+
+- **ID:** `2602.07031` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Dong Li, Shuai Huang, Yapeng Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.07031 · **PDF:** https://arxiv.org/pdf/2602.07031v1
+
+## 16177. Training and Simulation of Quadrupedal Robot in Adaptive Stair Climbing for Indoor Firefighting: An End-to-End Reinforcement Learning Approach
+
+- **ID:** `2602.03087` · **primary:** `cs.RO` · **Submitted:** 2026-02-03
+- **Authors:** Baixiao Huang, Baiyu Huang, Yu Hou
+- **abs:** https://arxiv.org/abs/2602.03087 · **PDF:** https://arxiv.org/pdf/2602.03087v1
+
+## 16178. Neural Predictor-Corrector: Solving Homotopy Problems with Reinforcement Learning
+
+- **ID:** `2602.03086` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Jiayao Mai, Bangyan Liao, Zhenjun Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.03086 · **PDF:** https://arxiv.org/pdf/2602.03086v1
+
+## 16179. The Trigger in the Haystack: Extracting and Reconstructing LLM Backdoor Triggers
+
+- **ID:** `2602.03085` · **primary:** `cs.CR` · **Submitted:** 2026-02-03
+- **Authors:** Blake Bullwinkel, Giorgio Severi, Keegan Hines, et al.
+- **abs:** https://arxiv.org/abs/2602.03085 · **PDF:** https://arxiv.org/pdf/2602.03085v1
+
+## 16180. AERO: Autonomous Evolutionary Reasoning Optimization via Endogenous Dual-Loop Feedback
+
+- **ID:** `2602.03084` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Zhitao Gao, Jie Ma, Xuhong Li, et al.
+- **abs:** https://arxiv.org/abs/2602.03084 · **PDF:** https://arxiv.org/pdf/2602.03084v2
+
+## 16181. Geometry-Preserving Neural Architectures on Manifolds with Boundary
+
+- **ID:** `2602.03082` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Karthik Elamvazhuthi, Shiba Biswal, Kian Rosenblum, et al.
+- **abs:** https://arxiv.org/abs/2602.03082 · **PDF:** https://arxiv.org/pdf/2602.03082v1
+
+## 16182. ReMiT: RL-Guided Mid-Training for Iterative LLM Evolution
+
+- **ID:** `2602.03075` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Junjie Huang, Jiarui Qin, Di Yin, et al.
+- **abs:** https://arxiv.org/abs/2602.03075 · **PDF:** https://arxiv.org/pdf/2602.03075v1
+
+## 16183. TMS: Trajectory-Mixed Supervision for Reward-Free, On-Policy SFT
+
+- **ID:** `2602.03073` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Rana Muhammad Shahroz Khan, Zijie Liu, Zhen Tan, et al.
+- **abs:** https://arxiv.org/abs/2602.03073 · **PDF:** https://arxiv.org/pdf/2602.03073v1
+
+## 16184. DSFlow: Dual Supervision and Step-Aware Architecture for One-Step Flow Matching Speech Synthesis
+
+- **ID:** `2602.09041` · **primary:** `cs.SD` · **Submitted:** 2026-02-03
+- **Authors:** Bin Lin, Peng Yang, Chao Yan, et al.
+- **abs:** https://arxiv.org/abs/2602.09041 · **PDF:** https://arxiv.org/pdf/2602.09041v1
+
+## 16185. Learning Physiology-Informed Vocal Spectrotemporal Representations for Speech Emotion Recognition
+
+- **ID:** `2602.13259` · **primary:** `cs.SD` · **Submitted:** 2026-02-03
+- **Authors:** Xu Zhang, Longbing Cao, Runze Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.13259 · **PDF:** https://arxiv.org/pdf/2602.13259v1
+
+## 16186. FlashSinkhorn: IO-Aware Entropic Optimal Transport
+
+- **ID:** `2602.03067` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Felix X. -F. Ye, Xingjie Li, An Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.03067 · **PDF:** https://arxiv.org/pdf/2602.03067v2
+
+## 16187. Prenatal Stress Detection from Electrocardiography Using Self-Supervised Deep Learning: Development and External Validation
+
+- **ID:** `2602.03886` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-03
+- **Authors:** Martin G. Frasch, Marlene J. E. Mayer, Clara Becker, et al.
+- **abs:** https://arxiv.org/abs/2602.03886 · **PDF:** https://arxiv.org/pdf/2602.03886v2
+
+## 16188. Shortcut Features as Top Eigenfunctions of NTK: A Linear Neural Network Case and More
+
+- **ID:** `2602.03066` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Jinwoo Lim, Suhyun Kim, Soo-Mook Moon
+- **abs:** https://arxiv.org/abs/2602.03066 · **PDF:** https://arxiv.org/pdf/2602.03066v1
+
+## 16189. MAPLE: A Sub-Agent Architecture for Memory, Learning, and Personalization in Agentic AI Systems
+
+- **ID:** `2602.13258` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Deepak Babu Piskala
+- **abs:** https://arxiv.org/abs/2602.13258 · **PDF:** https://arxiv.org/pdf/2602.13258v1
+
+## 16190. JRDB-Pose3D: A Multi-person 3D Human Pose and Shape Estimation Dataset for Robotics
+
+- **ID:** `2602.03064` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Sandika Biswas, Kian Izadpanah, Hamid Rezatofighi
+- **abs:** https://arxiv.org/abs/2602.03064 · **PDF:** https://arxiv.org/pdf/2602.03064v1
+
+## 16191. Evaluating LLMs When They Do Not Know the Answer: Statistical Evaluation of Mathematical Reasoning via Comparative Signals
+
+- **ID:** `2602.03061` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Zihan Dong, Zhixian Zhang, Yang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.03061 · **PDF:** https://arxiv.org/pdf/2602.03061v1
+
+## 16192. From Speech-to-Spatial: Grounding Utterances on A Live Shared View with Augmented Reality
+
+- **ID:** `2602.03059` · **primary:** `cs.HC` · **Submitted:** 2026-02-03
+- **Authors:** Yoonsang Kim, Divyansh Pradhan, Devshree Jadeja, et al.
+- **abs:** https://arxiv.org/abs/2602.03059 · **PDF:** https://arxiv.org/pdf/2602.03059v1
+
+## 16193. Stationarity and Spectral Characterization of Random Signals on Simplicial Complexes
+
+- **ID:** `2602.03055` · **primary:** `eess.SP` · **Submitted:** 2026-02-03
+- **Authors:** Madeline Navarro, Andrei Buciulea, Santiago Segarra, et al.
+- **abs:** https://arxiv.org/abs/2602.03055 · **PDF:** https://arxiv.org/pdf/2602.03055v1
+
+## 16194. Towards Considerate Embodied AI: Co-Designing Situated Multi-Site Healthcare Robots from Abstract Concepts to High-Fidelity Prototypes
+
+- **ID:** `2602.03054` · **primary:** `cs.HC` · **Submitted:** 2026-02-03
+- **Authors:** Yuanchen Bai, Ruixiang Han, Niti Parikh, et al.
+- **abs:** https://arxiv.org/abs/2602.03054 · **PDF:** https://arxiv.org/pdf/2602.03054v1
+
+## 16195. MAS-ProVe: Understanding the Process Verification of Multi-Agent Systems
+
+- **ID:** `2602.03053` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Vishal Venkataramani, Haizhou Shi, Zixuan Ke, et al.
+- **abs:** https://arxiv.org/abs/2602.03053 · **PDF:** https://arxiv.org/pdf/2602.03053v1
+
+## 16196. Fedcompass: Federated Clustered and Periodic Aggregation Framework for Hybrid Classical-Quantum Models
+
+- **ID:** `2602.03052` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yueheng Wang, Xing He, Zinuo Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.03052 · **PDF:** https://arxiv.org/pdf/2602.03052v1
+
+## 16197. SAES-SVD: Self-Adaptive Suppression of Accumulated and Local Errors for SVD-based LLM Compression
+
+- **ID:** `2602.03051` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Xing Hu, Dawei Yang, Yuan Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.03051 · **PDF:** https://arxiv.org/pdf/2602.03051v1
+
+## 16198. Unified Inference Framework for Single and Multi-Player Performative Prediction: Method and Asymptotic Optimality
+
+- **ID:** `2602.03049` · **primary:** `stat.ML` · **Submitted:** 2026-02-03
+- **Authors:** Zhixian Zhang, Xiaotian Hou, Linjun Zhang
+- **abs:** https://arxiv.org/abs/2602.03049 · **PDF:** https://arxiv.org/pdf/2602.03049v1
+
+## 16199. CoBA-RL: Capability-Oriented Budget Allocation for Reinforcement Learning in LLMs
+
+- **ID:** `2602.03048` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Zhiyuan Yao, Yi-Kai Zhang, Yuxin Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.03048 · **PDF:** https://arxiv.org/pdf/2602.03048v3
+
+## 16200. Clarify Before You Draw: Proactive Agents for Robust Text-to-CAD Generation
+
+- **ID:** `2602.03045` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Bo Yuan, Zelin Zhao, Petr Molodyk, et al.
+- **abs:** https://arxiv.org/abs/2602.03045 · **PDF:** https://arxiv.org/pdf/2602.03045v1
+
+## 16201. SAFE-KD: Risk-Controlled Early-Exit Distillation for Vision Backbones
+
+- **ID:** `2602.03043` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Salim Khazem
+- **abs:** https://arxiv.org/abs/2602.03043 · **PDF:** https://arxiv.org/pdf/2602.03043v1
+
+## 16202. Bongards at the Boundary of Perception and Reasoning: Programs or Language?
+
+- **ID:** `2602.03038` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Cassidy Langenfeld, Claas Beger, Gloria Geng, et al.
+- **abs:** https://arxiv.org/abs/2602.03038 · **PDF:** https://arxiv.org/pdf/2602.03038v1
+
+## 16203. LatentMem: Customizing Latent Memory for Multi-Agent Systems
+
+- **ID:** `2602.03036` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Muxin Fu, Xiangyuan Xue, Yafu Li, et al.
+- **abs:** https://arxiv.org/abs/2602.03036 · **PDF:** https://arxiv.org/pdf/2602.03036v2
+
+## 16204. Generalizable and Interpretable RF Fingerprinting with Shapelet-Enhanced Large Language Models
+
+- **ID:** `2602.03035` · **primary:** `cs.CR` · **Submitted:** 2026-02-03
+- **Authors:** Tianya Zhao, Junqing Zhang, Haowen Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.03035 · **PDF:** https://arxiv.org/pdf/2602.03035v1
+
+## 16205. KANFIS: A Neuro-Symbolic Framework for Interpretable and Uncertainty-Aware Learning
+
+- **ID:** `2602.03034` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Binbin Yong, Haoran Pei, Jun Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.03034 · **PDF:** https://arxiv.org/pdf/2602.03034v2
+
+## 16206. Physics-inspired transformer quantum states via latent imaginary-time evolution
+
+- **ID:** `2602.03031` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-02-03
+- **Authors:** Kimihiro Yamazaki, Itsushi Sakata, Takuya Konishi, et al.
+- **abs:** https://arxiv.org/abs/2602.03031 · **PDF:** https://arxiv.org/pdf/2602.03031v1
+
+## 16207. Visual Reasoning over Time Series via Multi-Agent System
+
+- **ID:** `2602.03026` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Weilin Ruan, Yuxuan Liang
+- **abs:** https://arxiv.org/abs/2602.03026 · **PDF:** https://arxiv.org/pdf/2602.03026v1
+
+## 16208. RC-GRPO: Reward-Conditioned Group Relative Policy Optimization for Multi-Turn Tool Calling Agents
+
+- **ID:** `2602.03025` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Haitian Zhong, Jixiu Zhai, Lei Song, et al.
+- **abs:** https://arxiv.org/abs/2602.03025 · **PDF:** https://arxiv.org/pdf/2602.03025v1
+
+## 16209. Consistency Deep Equilibrium Models
+
+- **ID:** `2602.03024` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Junchao Lin, Zenan Ling, Jingwen Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.03024 · **PDF:** https://arxiv.org/pdf/2602.03024v1
+
+## 16210. Rethinking Music Captioning with Music Metadata LLMs
+
+- **ID:** `2602.03023` · **primary:** `cs.SD` · **Submitted:** 2026-02-03
+- **Authors:** Irmak Bukey, Zhepei Wang, Chris Donahue, et al.
+- **abs:** https://arxiv.org/abs/2602.03023 · **PDF:** https://arxiv.org/pdf/2602.03023v1
+
+## 16211. STAR: Similarity-guided Teacher-Assisted Refinement for Super-Tiny Function Calling Models
+
+- **ID:** `2602.03022` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Jiliang Ni, Jiachen Pu, Zhongyi Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.03022 · **PDF:** https://arxiv.org/pdf/2602.03022v2
+
+## 16212. FedKRSO: Communication and Memory Efficient Federated Fine-Tuning of Large Language Models
+
+- **ID:** `2602.03019` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Guohao Yang, Tongle Wu, Yuanxiong Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.03019 · **PDF:** https://arxiv.org/pdf/2602.03019v1
+
+## 16213. From Zero to Hero: Advancing Zero-Shot Foundation Models for Tabular Outlier Detection
+
+- **ID:** `2602.03018` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Xueying Ding, Haomin Wen, Simon Klütterman, et al.
+- **abs:** https://arxiv.org/abs/2602.03018 · **PDF:** https://arxiv.org/pdf/2602.03018v1
+
+## 16214. Explainable Computer Vision Framework for Automated Pore Detection and Criticality Assessment in Additive Manufacturing
+
+- **ID:** `2602.03883` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Akshansh Mishra, Rakesh Morisetty
+- **abs:** https://arxiv.org/abs/2602.03883 · **PDF:** https://arxiv.org/pdf/2602.03883v1
+
+## 16215. CVE-Factory: Scaling Expert-Level Agentic Tasks for Code Security Vulnerability
+
+- **ID:** `2602.03012` · **primary:** `cs.CR` · **Submitted:** 2026-02-03
+- **Authors:** Xianzhen Luo, Jingyuan Zhang, Shiqi Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.03012 · **PDF:** https://arxiv.org/pdf/2602.03012v1
+
+## 16216. VOILA: Value-of-Information Guided Fidelity Selection for Cost-Aware Multimodal Question Answering
+
+- **ID:** `2602.03007` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Rahul Atul Bhope, K. R. Jayaram, Vinod Muthusamy, et al.
+- **abs:** https://arxiv.org/abs/2602.03007 · **PDF:** https://arxiv.org/pdf/2602.03007v1
+
+## 16217. Distilling LLM Reasoning into Graph of Concept Predictors
+
+- **ID:** `2602.03006` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Ziyang Yu, Liang Zhao
+- **abs:** https://arxiv.org/abs/2602.03006 · **PDF:** https://arxiv.org/pdf/2602.03006v2
+
+## 16218. Causal Graph Spatial-Temporal Autoencoder for Reliable and Interpretable Process Monitoring
+
+- **ID:** `2602.03004` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Xiangrui Zhang, Chunyue Song, Wei Dai, et al.
+- **abs:** https://arxiv.org/abs/2602.03004 · **PDF:** https://arxiv.org/pdf/2602.03004v1
+
+## 16219. Open Problems in Differentiable Social Choice: Learning Mechanisms, Decisions, and Alignment
+
+- **ID:** `2602.03003` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Zhiyu An, Wan Du
+- **abs:** https://arxiv.org/abs/2602.03003 · **PDF:** https://arxiv.org/pdf/2602.03003v3
+
+## 16220. Adaptive Batch Sizes Using Non-Euclidean Gradient Noise Scales for Stochastic Sign and Spectral Descent
+
+- **ID:** `2602.03001` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Hiroki Naganuma, Shagun Gupta, Youssef Briki, et al.
+- **abs:** https://arxiv.org/abs/2602.03001 · **PDF:** https://arxiv.org/pdf/2602.03001v1
+
+## 16221. PriorProbe: Recovering Individual-Level Priors for Personalizing Neural Networks in Facial Expression Recognition
+
+- **ID:** `2602.03882` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Haijiang Yan, Nick Chater, Adam Sanborn
+- **abs:** https://arxiv.org/abs/2602.03882 · **PDF:** https://arxiv.org/pdf/2602.03882v1
+
+## 16222. Agent Alpha: Tree Search Unifying Generation, Exploration and Evaluation for Computer-Use Agents
+
+- **ID:** `2602.02995` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Sizhe Tang, Rongqian Chen, Tian Lan
+- **abs:** https://arxiv.org/abs/2602.02995 · **PDF:** https://arxiv.org/pdf/2602.02995v1
+
+## 16223. Large Language Models Can Take False First Steps at Inference-time Planning
+
+- **ID:** `2602.02991` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Haijiang Yan, Jian-Qiao Zhu, Adam Sanborn
+- **abs:** https://arxiv.org/abs/2602.02991 · **PDF:** https://arxiv.org/pdf/2602.02991v1
+
+## 16224. Learning to Repair Lean Proofs from Compiler Feedback
+
+- **ID:** `2602.02990` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Evan Wang, Simon Chess, Daniel Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.02990 · **PDF:** https://arxiv.org/pdf/2602.02990v2
+
+## 16225. NLI:Non-uniform Linear Interpolation Approximation of Nonlinear Operations for Efficient LLMs Inference
+
+- **ID:** `2602.02988` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Jiangyong Yu, Xiaomeng Han, Xing Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.02988 · **PDF:** https://arxiv.org/pdf/2602.02988v1
+
+## 16226. Why Some Models Resist Unlearning: A Linear Stability Perspective
+
+- **ID:** `2602.02986` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Wei-Kai Chang, Rajiv Khanna
+- **abs:** https://arxiv.org/abs/2602.02986 · **PDF:** https://arxiv.org/pdf/2602.02986v1
+
+## 16227. PackInfer: Compute- and I/O-Efficient Attention for Batched LLM Inference
+
+- **ID:** `2602.06072` · **primary:** `cs.DC` · **Submitted:** 2026-02-03
+- **Authors:** Rui Ning, Wei Zhang, Fan Lai
+- **abs:** https://arxiv.org/abs/2602.06072 · **PDF:** https://arxiv.org/pdf/2602.06072v1
+
+## 16228. Do LLMs Share Human-Like Biases? Causal Reasoning Under Prior Knowledge, Irrelevant Context, and Varying Compute Budgets
+
+- **ID:** `2602.02983` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Hanna M. Dettki, Charley M. Wu, Bob Rehder
+- **abs:** https://arxiv.org/abs/2602.02983 · **PDF:** https://arxiv.org/pdf/2602.02983v2
+
+## 16229. WST-X Series: Wavelet Scattering Transform for Interpretable Speech Deepfake Detection
+
+- **ID:** `2602.02980` · **primary:** `eess.AS` · **Submitted:** 2026-02-03
+- **Authors:** Xi Xuan, Davide Carbone, Ruchi Pandey, et al.
+- **abs:** https://arxiv.org/abs/2602.02980 · **PDF:** https://arxiv.org/pdf/2602.02980v1
+
+## 16230. CPMobius: Iterative Coach-Player Reasoning for Data-Free Reinforcement Learning
+
+- **ID:** `2602.02979` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Ran Li, Zeyuan Liu, Yinghao chen, et al.
+- **abs:** https://arxiv.org/abs/2602.02979 · **PDF:** https://arxiv.org/pdf/2602.02979v1
+
+## 16231. Structuring Value Representations via Geometric Coherence in Markov Decision Processes
+
+- **ID:** `2602.02978` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Zuyuan Zhang, Zeyu Fang, Tian Lan
+- **abs:** https://arxiv.org/abs/2602.02978 · **PDF:** https://arxiv.org/pdf/2602.02978v1
+
+## 16232. Aligning Forest and Trees in Images and Long Captions for Visually Grounded Understanding
+
+- **ID:** `2602.02977` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Byeongju Woo, Zilin Wang, Byeonghyun Pak, et al.
+- **abs:** https://arxiv.org/abs/2602.02977 · **PDF:** https://arxiv.org/pdf/2602.02977v1
+
+## 16233. Where Norms and References Collide: Evaluating LLMs on Normative Reasoning
+
+- **ID:** `2602.02975` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Mitchell Abrams, Kaveh Eskandari Miandoab, Felix Gervits, et al.
+- **abs:** https://arxiv.org/abs/2602.02975 · **PDF:** https://arxiv.org/pdf/2602.02975v1
+
+## 16234. Learning Fast Monomial Orders for Gröbner Basis Computations
+
+- **ID:** `2602.02972` · **primary:** `cs.SC` · **Submitted:** 2026-02-03
+- **Authors:** R. Caleb Bunch, Alperen A. Ergür, Melika Golestani, et al.
+- **abs:** https://arxiv.org/abs/2602.02972 · **PDF:** https://arxiv.org/pdf/2602.02972v1
+
+## 16235. DDL2PropBank Agent: Benchmarking Multi-Agent Frameworks' Developer Experience Through a Novel Relational Schema Mapping Task
+
+- **ID:** `2602.11198` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Shafiuddin Rehan Ahmed, Wei Wei
+- **abs:** https://arxiv.org/abs/2602.11198 · **PDF:** https://arxiv.org/pdf/2602.11198v1
+
+## 16236. Co2PO: Coordinated Constrained Policy Optimization for Multi-Agent RL
+
+- **ID:** `2602.02970` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Shrenik Patel, Christine Truong
+- **abs:** https://arxiv.org/abs/2602.02970 · **PDF:** https://arxiv.org/pdf/2602.02970v1
+
+## 16237. Q-ShiftDP: A Differentially Private Parameter-Shift Rule for Quantum Machine Learning
+
+- **ID:** `2602.02962` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Hoang M. Ngo, Nhat Hoang-Xuan, Quan Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.02962 · **PDF:** https://arxiv.org/pdf/2602.02962v1
+
+## 16238. Generative Engine Optimization: A VLM and Agent Framework for Pinterest Acquisition Growth
+
+- **ID:** `2602.02961` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Faye Zhang, Qianyu Cheng, Jasmine Wan, et al.
+- **abs:** https://arxiv.org/abs/2602.02961 · **PDF:** https://arxiv.org/pdf/2602.02961v1
+
+## 16239. Embodiment-Aware Generalist Specialist Distillation for Unified Humanoid Whole-Body Control
+
+- **ID:** `2602.02960` · **primary:** `cs.RO` · **Submitted:** 2026-02-03
+- **Authors:** Quanquan Peng, Yunfeng Lin, Yufei Xue, et al.
+- **abs:** https://arxiv.org/abs/2602.02960 · **PDF:** https://arxiv.org/pdf/2602.02960v2
+
+## 16240. Human-Centric Traffic Signal Control for Equity: A Multi-Agent Action Branching Deep Reinforcement Learning Approach
+
+- **ID:** `2602.02959` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Xiaocai Zhang, Neema Nassir, Lok Sang Chan, et al.
+- **abs:** https://arxiv.org/abs/2602.02959 · **PDF:** https://arxiv.org/pdf/2602.02959v1
+
+## 16241. Quant VideoGen: Auto-Regressive Long Video Generation via 2-Bit KV-Cache Quantization
+
+- **ID:** `2602.02958` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Haocheng Xi, Shuo Yang, Yilong Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.02958 · **PDF:** https://arxiv.org/pdf/2602.02958v3
+
+## 16242. Synthetic Data Augmentation for Medical Audio Classification: A Preliminary Evaluation
+
+- **ID:** `2602.02955` · **primary:** `cs.SD` · **Submitted:** 2026-02-03
+- **Authors:** David McShannon, Anthony Mella, Nicholas Dietrich
+- **abs:** https://arxiv.org/abs/2602.02955 · **PDF:** https://arxiv.org/pdf/2602.02955v1
+
+## 16243. UAT-LITE: Inference-Time Uncertainty-Aware Attention for Pretrained Transformers
+
+- **ID:** `2602.02952` · **primary:** `cs.AI` · **Submitted:** 2026-02-03
+- **Authors:** Elias Hossain, Shubhashis Roy Dipta, Subash Neupane, et al.
+- **abs:** https://arxiv.org/abs/2602.02952 · **PDF:** https://arxiv.org/pdf/2602.02952v2
+
+## 16244. Nüwa: Mending the Spatial Integrity Torn by VLM Token Pruning
+
+- **ID:** `2602.02951` · **primary:** `cs.CV` · **Submitted:** 2026-02-03
+- **Authors:** Yihong Huang, Fei Ma, Yihua Shao, et al.
+- **abs:** https://arxiv.org/abs/2602.02951 · **PDF:** https://arxiv.org/pdf/2602.02951v1
+
+## 16245. Variational Sparse Paired Autoencoders (vsPAIR) for Inverse Problems and Uncertainty Quantification
+
+- **ID:** `2602.02948` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Jack Michael Solomon, Rishi Leburu, Matthias Chung
+- **abs:** https://arxiv.org/abs/2602.02948 · **PDF:** https://arxiv.org/pdf/2602.02948v2
+
+## 16246. 3D-Learning: Diffusion-Augmented Distributionally Robust Decision-Focused Learning
+
+- **ID:** `2602.02943` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Jiaqi Wen, Lei Fan, Jianyi Yang
+- **abs:** https://arxiv.org/abs/2602.02943 · **PDF:** https://arxiv.org/pdf/2602.02943v1
+
+## 16247. A vector logic for intensional formal semantics
+
+- **ID:** `2602.02940` · **primary:** `math.LO` · **Submitted:** 2026-02-03
+- **Authors:** Daniel Quigley
+- **abs:** https://arxiv.org/abs/2602.02940 · **PDF:** https://arxiv.org/pdf/2602.02940v1
+
+## 16248. Equal Access, Unequal Interaction: A Counterfactual Audit of LLM Fairness
+
+- **ID:** `2602.02932` · **primary:** `cs.CL` · **Submitted:** 2026-02-03
+- **Authors:** Alireza Amiri-Margavi, Arshia Gharagozlou, Amin Gholami Davodi, et al.
+- **abs:** https://arxiv.org/abs/2602.02932 · **PDF:** https://arxiv.org/pdf/2602.02932v1
+
+## 16249. Weighted Sum-of-Trees Model for Clustered Data
+
+- **ID:** `2602.02931` · **primary:** `stat.ME` · **Submitted:** 2026-02-03
+- **Authors:** Kevin McCoy, Zachary Wooten, Katarzyna Tomczak, et al.
+- **abs:** https://arxiv.org/abs/2602.02931 · **PDF:** https://arxiv.org/pdf/2602.02931v1
+
+## 16250. Rare Event Early Detection: A Dataset of Sepsis Onset for Critically Ill Trauma Patients
+
+- **ID:** `2602.02930` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Yin Jin, Tucker R. Stewart, Deyi Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.02930 · **PDF:** https://arxiv.org/pdf/2602.02930v1
+
+## 16251. RPG-AE: Neuro-Symbolic Graph Autoencoders with Rare Pattern Mining for Provenance-Based Anomaly Detection
+
+- **ID:** `2602.02929` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Asif Tauhid, Sidahmed Benabderrahmane, Mohamad Altrabulsi, et al.
+- **abs:** https://arxiv.org/abs/2602.02929 · **PDF:** https://arxiv.org/pdf/2602.02929v2
+
+## 16252. Distance Marching for Generative Modeling
+
+- **ID:** `2602.02928` · **primary:** `cs.LG` · **Submitted:** 2026-02-03
+- **Authors:** Zimo Wang, Ishit Mehta, Haolin Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.02928 · **PDF:** https://arxiv.org/pdf/2602.02928v1
+
+## 16253. Training-Free Self-Correction for Multimodal Masked Diffusion Models
+
+- **ID:** `2602.02927` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** Yidong Ouyang, Panwen Hu, Zhengyan Wan, et al.
+- **abs:** https://arxiv.org/abs/2602.02927 · **PDF:** https://arxiv.org/pdf/2602.02927v1
+
+## 16254. Refining Decision Boundaries In Anomaly Detection Using Similarity Search Within the Feature Space
+
+- **ID:** `2602.02925` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Sidahmed Benabderrahmane, Petko Valtchev, James Cheney, et al.
+- **abs:** https://arxiv.org/abs/2602.02925 · **PDF:** https://arxiv.org/pdf/2602.02925v1
+
+## 16255. How Does the Lagrangian Guide Safe Reinforcement Learning through Diffusion Models?
+
+- **ID:** `2602.02924` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Xiaoyuan Cheng, Wenxuan Yuan, Boyang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.02924 · **PDF:** https://arxiv.org/pdf/2602.02924v1
+
+## 16256. A Reproducible Framework for Bias-Resistant Machine Learning on Small-Sample Neuroimaging Data
+
+- **ID:** `2602.02920` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Jagan Mohan Reddy Dwarampudi, Jennifer L Purks, Joshua Wong, et al.
+- **abs:** https://arxiv.org/abs/2602.02920 · **PDF:** https://arxiv.org/pdf/2602.02920v1
+
+## 16257. DeltaEvolve: Accelerating Scientific Discovery through Momentum-Driven Evolution
+
+- **ID:** `2602.02919` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Jiachen Jiang, Tianyu Ding, Zhihui Zhu
+- **abs:** https://arxiv.org/abs/2602.02919 · **PDF:** https://arxiv.org/pdf/2602.02919v1
+
+## 16258. A Multi-scale Linear-time Encoder for Whole-Slide Image Analysis
+
+- **ID:** `2602.02918` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Jagan Mohan Reddy Dwarampudi, Joshua Wong, Hien Van Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.02918 · **PDF:** https://arxiv.org/pdf/2602.02918v1
+
+## 16259. Weighted Temporal Decay Loss for Learning Wearable PPG Data with Sparse Clinical Labels
+
+- **ID:** `2602.02917` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yunsung Chung, Keum San Chun, Migyeong Gwak, et al.
+- **abs:** https://arxiv.org/abs/2602.02917 · **PDF:** https://arxiv.org/pdf/2602.02917v1
+
+## 16260. Notes on the Reward Representation of Posterior Updates
+
+- **ID:** `2602.02912` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Pedro A. Ortega
+- **abs:** https://arxiv.org/abs/2602.02912 · **PDF:** https://arxiv.org/pdf/2602.02912v1
+
+## 16261. Reasoning about Reasoning: BAPO Bounds on Chain-of-Thought Token Complexity in LLMs
+
+- **ID:** `2602.02909` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Kiran Tomlinson, Tobias Schnabel, Adith Swaminathan, et al.
+- **abs:** https://arxiv.org/abs/2602.02909 · **PDF:** https://arxiv.org/pdf/2602.02909v1
+
+## 16262. Neural Sabermetrics with World Model: Play-by-play Predictive Modeling with Large Language Model
+
+- **ID:** `2602.07030` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Young Jin Ahn, Yiyang Du, Zheyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.07030 · **PDF:** https://arxiv.org/pdf/2602.07030v1
+
+## 16263. A Random Matrix Theory Perspective on the Consistency of Diffusion Models
+
+- **ID:** `2602.02908` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Binxu Wang, Jacob Zavatone-Veth, Cengiz Pehlevan
+- **abs:** https://arxiv.org/abs/2602.02908 · **PDF:** https://arxiv.org/pdf/2602.02908v1
+
+## 16264. FIRE-Bench: Evaluating Agents on the Rediscovery of Scientific Insights
+
+- **ID:** `2602.02905` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Zhen Wang, Fan Bai, Zhongyan Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.02905 · **PDF:** https://arxiv.org/pdf/2602.02905v1
+
+## 16265. Spatiotemporal Decision Transformer for Traffic Coordination
+
+- **ID:** `2602.02903` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Haoran Su, Yandong Sun, Hanxiao Deng
+- **abs:** https://arxiv.org/abs/2602.02903 · **PDF:** https://arxiv.org/pdf/2602.02903v1
+
+## 16266. Minimal Computational Preconditions for Subjective Perspective in Artificial Agents
+
+- **ID:** `2602.02902` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Hongju Pae
+- **abs:** https://arxiv.org/abs/2602.02902 · **PDF:** https://arxiv.org/pdf/2602.02902v2
+
+## 16267. Manifold-Constrained Energy-Based Transition Models for Offline Reinforcement Learning
+
+- **ID:** `2602.02900` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zeyu Fang, Zuyuan Zhang, Mahdi Imani, et al.
+- **abs:** https://arxiv.org/abs/2602.02900 · **PDF:** https://arxiv.org/pdf/2602.02900v1
+
+## 16268. Controlled disagreement improves generalization in decentralized training
+
+- **ID:** `2602.02899` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zesen Wang, Mikael Johansson
+- **abs:** https://arxiv.org/abs/2602.02899 · **PDF:** https://arxiv.org/pdf/2602.02899v1
+
+## 16269. Aligning Language Model Benchmarks with Pairwise Preferences
+
+- **ID:** `2602.02898` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Marco Gutierrez, Xinyi Leng, Hannah Cyberey, et al.
+- **abs:** https://arxiv.org/abs/2602.02898 · **PDF:** https://arxiv.org/pdf/2602.02898v1
+
+## 16270. Moving On, Even When You're Broken: Fail-Active Trajectory Generation via Diffusion Policies Conditioned on Embodiment and Task
+
+- **ID:** `2602.02895` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Gilberto G. Briscoe-Martinez, Yaashia Gautam, Rahul Shetty, et al.
+- **abs:** https://arxiv.org/abs/2602.02895 · **PDF:** https://arxiv.org/pdf/2602.02895v2
+
+## 16271. DoubleTake: Contrastive Reasoning for Faithful Decision-Making in Medical Imaging
+
+- **ID:** `2602.02894` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Daivik Patel, Shrenik Patel
+- **abs:** https://arxiv.org/abs/2602.02894 · **PDF:** https://arxiv.org/pdf/2602.02894v1
+
+## 16272. TraceNAS: Zero-shot LLM Pruning via Gradient Trace Correlation
+
+- **ID:** `2602.02891` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Prajna G. Malettira, Manish Nagaraj, Arjun Roy, et al.
+- **abs:** https://arxiv.org/abs/2602.02891 · **PDF:** https://arxiv.org/pdf/2602.02891v1
+
+## 16273. Self-Soupervision: Cooking Model Soups without Labels
+
+- **ID:** `2602.02890` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Anthony Fuller, James R. Green, Evan Shelhamer
+- **abs:** https://arxiv.org/abs/2602.02890 · **PDF:** https://arxiv.org/pdf/2602.02890v1
+
+## 16274. HALT: Hallucination Assessment via Log-probs as Time series
+
+- **ID:** `2602.02888` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Ahmad Shapiro, Karan Taneja, Ashok Goel
+- **abs:** https://arxiv.org/abs/2602.02888 · **PDF:** https://arxiv.org/pdf/2602.02888v1
+
+## 16275. Mixture of Concept Bottleneck Experts
+
+- **ID:** `2602.02886` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Francesco De Santis, Gabriele Ciravegna, Giovanni De Felice, et al.
+- **abs:** https://arxiv.org/abs/2602.02886 · **PDF:** https://arxiv.org/pdf/2602.02886v1
+
+## 16276. Learning-Infused Formal Reasoning: From Contract Synthesis to Artifact Reuse and Formal Semantics
+
+- **ID:** `2602.02881` · **primary:** `cs.SE` · **Submitted:** 2026-02-02
+- **Authors:** Arshad Beg, Diarmuid O'Donoghue, Rosemary Monahan
+- **abs:** https://arxiv.org/abs/2602.02881 · **PDF:** https://arxiv.org/pdf/2602.02881v1
+
+## 16277. Which course? Discourse! Teaching Discourse and Generation in the Era of LLMs
+
+- **ID:** `2602.02878` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Junyi Jessy Li, Yang Janet Liu, Kanishka Misra, et al.
+- **abs:** https://arxiv.org/abs/2602.02878 · **PDF:** https://arxiv.org/pdf/2602.02878v2
+
+## 16278. A Geometry-Aware Efficient Algorithm for Compositional Entropic Risk Minimization
+
+- **ID:** `2602.02877` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Xiyuan Wei, Linli Zhou, Bokun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.02877 · **PDF:** https://arxiv.org/pdf/2602.02877v1
+
+## 16279. Extracting Recurring Vulnerabilities from Black-Box LLM-Generated Software
+
+- **ID:** `2602.04894` · **primary:** `cs.CR` · **Submitted:** 2026-02-02
+- **Authors:** Tomer Kordonsky, Maayan Yamin, Noam Benzimra, et al.
+- **abs:** https://arxiv.org/abs/2602.04894 · **PDF:** https://arxiv.org/pdf/2602.04894v3
+
+## 16280. "I May Not Have Articulated Myself Clearly": Diagnosing Dynamic Instability in LLM Reasoning at Inference Time
+
+- **ID:** `2602.02863` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Jinkun Chen, Fengxiang Cheng, Sijia Han, et al.
+- **abs:** https://arxiv.org/abs/2602.02863 · **PDF:** https://arxiv.org/pdf/2602.02863v1
+
+## 16281. STEER: Inference-Time Risk Control via Constrained Quality-Diversity Search
+
+- **ID:** `2602.02862` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Eric Yang, Jong Ha Lee, Jonathan Amar, et al.
+- **abs:** https://arxiv.org/abs/2602.02862 · **PDF:** https://arxiv.org/pdf/2602.02862v1
+
+## 16282. Late-Stage Generalization Collapse in Grokking: Detecting anti-grokking with Weightwatcher
+
+- **ID:** `2602.02859` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Hari K Prakash, Charles H Martin
+- **abs:** https://arxiv.org/abs/2602.02859 · **PDF:** https://arxiv.org/pdf/2602.02859v1
+
+## 16283. IMAGINE: Intelligent Multi-Agent Godot-based Indoor Networked Exploration
+
+- **ID:** `2602.02858` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Tiago Leite, Maria Conceição, António Grilo
+- **abs:** https://arxiv.org/abs/2602.02858 · **PDF:** https://arxiv.org/pdf/2602.02858v1
+
+## 16284. When pre-training hurts LoRA fine-tuning: a dynamical analysis via single-index models
+
+- **ID:** `2602.02855` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Gibbs Nwemadji, Bruno Loureiro, Jean Barbier
+- **abs:** https://arxiv.org/abs/2602.02855 · **PDF:** https://arxiv.org/pdf/2602.02855v1
+
+## 16285. Recurrent Equivariant Constraint Modulation: Learning Per-Layer Symmetry Relaxation from Data
+
+- **ID:** `2602.02853` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Stefanos Pertigkiozoglou, Mircea Petrache, Shubhendu Trivedi, et al.
+- **abs:** https://arxiv.org/abs/2602.02853 · **PDF:** https://arxiv.org/pdf/2602.02853v2
+
+## 16286. DiGAN: Diffusion-Guided Attention Network for Early Alzheimer's Disease Detection
+
+- **ID:** `2602.03881` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Maxx Richard Rahman, Mostafa Hammouda, Wolfgang Maass
+- **abs:** https://arxiv.org/abs/2602.03881 · **PDF:** https://arxiv.org/pdf/2602.03881v2
+
+## 16287. AutoSizer: Automatic Sizing of Analog and Mixed-Signal Circuits via Large Language Model (LLM) Agents
+
+- **ID:** `2602.02849` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Xi Yu, Dmitrii Torbunov, Soumyajit Mandal, et al.
+- **abs:** https://arxiv.org/abs/2602.02849 · **PDF:** https://arxiv.org/pdf/2602.02849v1
+
+## 16288. Zero Sum SVD: Balancing Loss Sensitivity for Low Rank LLM Compression
+
+- **ID:** `2602.02848` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Ali Abbasi, Chayne Thrash, Haoran Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.02848 · **PDF:** https://arxiv.org/pdf/2602.02848v1
+
+## 16289. Causal Flow Q-Learning for Robust Offline Reinforcement Learning
+
+- **ID:** `2602.02847` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Mingxuan Li, Junzhe Zhang, Elias Bareinboim
+- **abs:** https://arxiv.org/abs/2602.02847 · **PDF:** https://arxiv.org/pdf/2602.02847v1
+
+## 16290. Act or Clarify? Modeling Sensitivity to Uncertainty and Cost in Communication
+
+- **ID:** `2602.02843` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Polina Tsvilodub, Karl Mulligan, Todd Snider, et al.
+- **abs:** https://arxiv.org/abs/2602.02843 · **PDF:** https://arxiv.org/pdf/2602.02843v2
+
+## 16291. Chain of Simulation: A Dual-Mode Reasoning Framework for Large Language Models with Dynamic Problem Routing
+
+- **ID:** `2602.02842` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Saeid Sheikhi
+- **abs:** https://arxiv.org/abs/2602.02842 · **PDF:** https://arxiv.org/pdf/2602.02842v1
+
+## 16292. Semantics-Aware Generative Latent Data Augmentation for Learning in Low-Resource Domains
+
+- **ID:** `2602.02841` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Jaesung Bae, Minje Kim
+- **abs:** https://arxiv.org/abs/2602.02841 · **PDF:** https://arxiv.org/pdf/2602.02841v2
+
+## 16293. Beyond Content: Behavioral Policies Reveal Actors in Information Operations
+
+- **ID:** `2602.02838` · **primary:** `cs.SI` · **Submitted:** 2026-02-02
+- **Authors:** Philipp J. Schneider, Lanqin Yuan, Marian-Andrei Rizoiu
+- **abs:** https://arxiv.org/abs/2602.02838 · **PDF:** https://arxiv.org/pdf/2602.02838v1
+
+## 16294. Tabula RASA: Exposing and Breaking the Relational Bottleneck in Transformers
+
+- **ID:** `2602.02834` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Jonas Petersen, Camilla Mazzoleni, Riccardo Maggioni
+- **abs:** https://arxiv.org/abs/2602.02834 · **PDF:** https://arxiv.org/pdf/2602.02834v2
+
+## 16295. Koopman Autoencoders with Continuous-Time Latent Dynamics for Fluid Dynamics Forecasting
+
+- **ID:** `2602.02832` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Rares Grozavescu, Pengyu Zhang, Etienne Meunier, et al.
+- **abs:** https://arxiv.org/abs/2602.02832 · **PDF:** https://arxiv.org/pdf/2602.02832v2
+
+## 16296. SC3D: Dynamic and Differentiable Causal Discovery for Temporal and Instantaneous Graphs
+
+- **ID:** `2602.02830` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Sourajit Das, Dibyajyoti Chakraborty, Romit Maulik
+- **abs:** https://arxiv.org/abs/2602.02830 · **PDF:** https://arxiv.org/pdf/2602.02830v2
+
+## 16297. A Single Revision Step Improves Token-Efficient LLM Reasoning
+
+- **ID:** `2602.02828` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yingchuan Zhang, Terry Ma, Wenxuan Zhong, et al.
+- **abs:** https://arxiv.org/abs/2602.02828 · **PDF:** https://arxiv.org/pdf/2602.02828v1
+
+## 16298. CATNIP: LLM Unlearning via Calibrated and Tokenized Negative Preference Alignment
+
+- **ID:** `2602.02824` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Zhengbang Yang, Yisheng Zhong, Junyuan Hong, et al.
+- **abs:** https://arxiv.org/abs/2602.02824 · **PDF:** https://arxiv.org/pdf/2602.02824v1
+
+## 16299. R2-Router: A New Paradigm for LLM Routing with Reasoning
+
+- **ID:** `2602.02823` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Jiaqi Xue, Qian Lou, Jiarong Xing, et al.
+- **abs:** https://arxiv.org/abs/2602.02823 · **PDF:** https://arxiv.org/pdf/2602.02823v1
+
+## 16300. When Efficient Communication Explains Convexity
+
+- **ID:** `2602.02821` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Ashvin Ranjan, Shane Steinert-Threlkeld
+- **abs:** https://arxiv.org/abs/2602.02821 · **PDF:** https://arxiv.org/pdf/2602.02821v1
+
+## 16301. From Tokens to Numbers: Continuous Number Modeling for SVG Generation
+
+- **ID:** `2602.02820` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Michael Ogezi, Martin Bell, Freda Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.02820 · **PDF:** https://arxiv.org/pdf/2602.02820v1
+
+## 16302. Membership Inference Attacks from Causal Principles
+
+- **ID:** `2602.02819` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Mathieu Even, Clément Berenfeld, Linus Bleistein, et al.
+- **abs:** https://arxiv.org/abs/2602.02819 · **PDF:** https://arxiv.org/pdf/2602.02819v2
+
+## 16303. Downscaling land surface temperature data using edge detection and block-diagonal Gaussian process regression
+
+- **ID:** `2602.02813` · **primary:** `stat.AP` · **Submitted:** 2026-02-02
+- **Authors:** Sanjit Dandapanthula, Margaret Johnson, Madeleine Pascolini-Campbell, et al.
+- **abs:** https://arxiv.org/abs/2602.02813 · **PDF:** https://arxiv.org/pdf/2602.02813v1
+
+## 16304. LmPT: Conditional Point Transformer for Anatomical Landmark Detection on 3D Point Clouds
+
+- **ID:** `2602.02808` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Matteo Bastico, Pierre Onghena, David Ryckelynck, et al.
+- **abs:** https://arxiv.org/abs/2602.02808 · **PDF:** https://arxiv.org/pdf/2602.02808v1
+
+## 16305. Joint Learning of Hierarchical Neural Options and Abstract World Model
+
+- **ID:** `2602.02799` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Wasu Top Piriyakulkij, Wolfgang Lehrach, Kevin Ellis, et al.
+- **abs:** https://arxiv.org/abs/2602.02799 · **PDF:** https://arxiv.org/pdf/2602.02799v1
+
+## 16306. Causality--Δ: Jacobian-Based Dependency Analysis in Flow Matching Models
+
+- **ID:** `2602.02793` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Reza Rezvan, Gustav Gille, Moritz Schauer, et al.
+- **abs:** https://arxiv.org/abs/2602.02793 · **PDF:** https://arxiv.org/pdf/2602.02793v1
+
+## 16307. Plug-In Classification of Drift Functions in Diffusion Processes Using Neural Networks
+
+- **ID:** `2602.02791` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** Yuzhen Zhao, Jiarong Fan, Yating Liu
+- **abs:** https://arxiv.org/abs/2602.02791 · **PDF:** https://arxiv.org/pdf/2602.02791v1
+
+## 16308. Simulating Human Audiovisual Search Behavior
+
+- **ID:** `2602.02790` · **primary:** `cs.HC` · **Submitted:** 2026-02-02
+- **Authors:** Hyunsung Cho, Xuejing Luo, Byungjoo Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.02790 · **PDF:** https://arxiv.org/pdf/2602.02790v1
+
+## 16309. Structure-Preserving Learning Improves Geometry Generalization in Neural PDEs
+
+- **ID:** `2602.02788` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Benjamin D. Shaffer, Shawn Koohy, Brooks Kinch, et al.
+- **abs:** https://arxiv.org/abs/2602.02788 · **PDF:** https://arxiv.org/pdf/2602.02788v1
+
+## 16310. Isotonic Layer: A Universal Framework for Generic Recommendation Debiasing
+
+- **ID:** `2603.06589` · **primary:** `cs.IR` · **Submitted:** 2026-02-02
+- **Authors:** Hailing Cheng, Yafang Yang, Hemeng Tao, et al.
+- **abs:** https://arxiv.org/abs/2603.06589 · **PDF:** https://arxiv.org/pdf/2603.06589v1
+
+## 16311. LEMON: Local Explanations via Modality-aware OptimizatioN
+
+- **ID:** `2602.02786` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yu Qin, Phillip Sloan, Raul Santos-Rodriguez, et al.
+- **abs:** https://arxiv.org/abs/2602.02786 · **PDF:** https://arxiv.org/pdf/2602.02786v1
+
+## 16312. Cross-Temporal Attention Fusion (CTAF) for Multimodal Physiological Signals in Self-Supervised Learning
+
+- **ID:** `2602.02784` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Arian Khorasani, Théophile Demazure
+- **abs:** https://arxiv.org/abs/2602.02784 · **PDF:** https://arxiv.org/pdf/2602.02784v1
+
+## 16313. Evaluating False Alarm and Missing Attacks in CAN IDS
+
+- **ID:** `2602.02781` · **primary:** `cs.CR` · **Submitted:** 2026-02-02
+- **Authors:** Nirab Hossain, Pablo Moriano
+- **abs:** https://arxiv.org/abs/2602.02781 · **PDF:** https://arxiv.org/pdf/2602.02781v1
+
+## 16314. A Study of Adaptive Modeling Towards Robust Generalization
+
+- **ID:** `2602.02780` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Zihao Jing, Qiuhao Zeng, Ruiyi Fang, et al.
+- **abs:** https://arxiv.org/abs/2602.02780 · **PDF:** https://arxiv.org/pdf/2602.02780v2
+
+## 16315. Verification and Identification in ECG biometric on large-scale
+
+- **ID:** `2602.02776` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Scagnetto Arjuna
+- **abs:** https://arxiv.org/abs/2602.02776 · **PDF:** https://arxiv.org/pdf/2602.02776v2
+
+## 16316. AmharicStoryQA: A Multicultural Story Question Answering Benchmark in Amharic
+
+- **ID:** `2602.02774` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Israel Abebe Azime, Abenezer Kebede Angamo, Hana Mekonen Tamiru, et al.
+- **abs:** https://arxiv.org/abs/2602.02774 · **PDF:** https://arxiv.org/pdf/2602.02774v1
+
+## 16317. BiTimeCrossNet: Time-Aware Self-Supervised Learning for Pediatric Sleep
+
+- **ID:** `2602.02769` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Saurav Raj Pandey, Harlin Lee
+- **abs:** https://arxiv.org/abs/2602.02769 · **PDF:** https://arxiv.org/pdf/2602.02769v1
+
+## 16318. Provable Effects of Data Replay in Continual Learning: A Feature Learning Perspective
+
+- **ID:** `2602.02767` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Meng Ding, Jinhui Xu, Kaiyi Ji
+- **abs:** https://arxiv.org/abs/2602.02767 · **PDF:** https://arxiv.org/pdf/2602.02767v1
+
+## 16319. Privately Fine-Tuned LLMs Preserve Temporal Dynamics in Tabular Data
+
+- **ID:** `2602.02766` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Lucas Rosenblatt, Peihan Liu, Ryan McKenna, et al.
+- **abs:** https://arxiv.org/abs/2602.02766 · **PDF:** https://arxiv.org/pdf/2602.02766v1
+
+## 16320. Genetic Generalized Additive Models
+
+- **ID:** `2602.15877` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Kaaustaaub Shankar, Kelly Cohen
+- **abs:** https://arxiv.org/abs/2602.15877 · **PDF:** https://arxiv.org/pdf/2602.15877v1
+
+## 16321. Exposing Vulnerabilities in Explanation for Time Series Classifiers via Dual-Target Attacks
+
+- **ID:** `2602.02763` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Bohan Wang, Zewen Liu, Lu Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.02763 · **PDF:** https://arxiv.org/pdf/2602.02763v2
+
+## 16322. A Comparative Study of Adversarial Robustness in CNN and CNN-ANFIS Architectures
+
+- **ID:** `2602.07028` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Kaaustaaub Shankar, Bharadwaj Dogga, Kelly Cohen
+- **abs:** https://arxiv.org/abs/2602.07028 · **PDF:** https://arxiv.org/pdf/2602.07028v1
+
+## 16323. On the Sample Efficiency of Inverse Dynamics Models for Semi-Supervised Imitation Learning
+
+- **ID:** `2602.02762` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Sacha Morin, Moonsub Byeon, Alexia Jolicoeur-Martineau, et al.
+- **abs:** https://arxiv.org/abs/2602.02762 · **PDF:** https://arxiv.org/pdf/2602.02762v1
+
+## 16324. From Task Solving to Robust Real-World Adaptation in LLM Agents
+
+- **ID:** `2602.02760` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Pouya Pezeshkpour, Estevam Hruschka
+- **abs:** https://arxiv.org/abs/2602.02760 · **PDF:** https://arxiv.org/pdf/2602.02760v1
+
+## 16325. Near-Universal Multiplicative Updates for Nonnegative Einsum Factorization
+
+- **ID:** `2602.02759` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** John Hood, Aaron Schein
+- **abs:** https://arxiv.org/abs/2602.02759 · **PDF:** https://arxiv.org/pdf/2602.02759v2
+
+## 16326. Scaling Small Agents Through Strategy Auctions
+
+- **ID:** `2602.02751` · **primary:** `cs.MA` · **Submitted:** 2026-02-02
+- **Authors:** Lisa Alazraki, William F. Shen, Yoram Bachrach, et al.
+- **abs:** https://arxiv.org/abs/2602.02751 · **PDF:** https://arxiv.org/pdf/2602.02751v1
+
+## 16327. Semantic Containment as a Fundamental Property of Emergent Misalignment
+
+- **ID:** `2603.04407` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Rohan Saxena
+- **abs:** https://arxiv.org/abs/2603.04407 · **PDF:** https://arxiv.org/pdf/2603.04407v1
+
+## 16328. Entropy-Guided Dynamic Tokens for Graph-LLM Alignment in Molecular Understanding
+
+- **ID:** `2602.02742` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zihao Jing, Qiuhao Zeng, Ruiyi Fang, et al.
+- **abs:** https://arxiv.org/abs/2602.02742 · **PDF:** https://arxiv.org/pdf/2602.02742v3
+
+## 16329. TopoPrune: Robust Data Pruning via Unified Latent Space Topology
+
+- **ID:** `2602.02739` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Arjun Roy, Prajna G. Malettira, Manish Nagaraj, et al.
+- **abs:** https://arxiv.org/abs/2602.02739 · **PDF:** https://arxiv.org/pdf/2602.02739v1
+
+## 16330. When Noise Lowers The Loss: Rethinking Likelihood-Based Evaluation in Music Large Language Models
+
+- **ID:** `2602.02738` · **primary:** `cs.SD` · **Submitted:** 2026-02-02
+- **Authors:** Xiaosha Li, Chun Liu, Ziyu Wang
+- **abs:** https://arxiv.org/abs/2602.02738 · **PDF:** https://arxiv.org/pdf/2602.02738v1
+
+## 16331. Time-Critical Multimodal Medical Transportation: Organs, Patients, and Medical Supplies
+
+- **ID:** `2602.02736` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Elaheh Sabziyan Varnousfaderani, Syed A. M. Shihab, Mohammad Taghizadeh
+- **abs:** https://arxiv.org/abs/2602.02736 · **PDF:** https://arxiv.org/pdf/2602.02736v1
+
+## 16332. TabPFN for Zero-shot Parametric Engineering Design Generation
+
+- **ID:** `2602.02735` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Ke Wang, Yifan Tang, Nguyen Gia Hien Vu, et al.
+- **abs:** https://arxiv.org/abs/2602.02735 · **PDF:** https://arxiv.org/pdf/2602.02735v1
+
+## 16333. WAXAL: A Large-Scale Multilingual African Language Speech Corpus
+
+- **ID:** `2602.02734` · **primary:** `eess.AS` · **Submitted:** 2026-02-02
+- **Authors:** Abdoulaye Diack, Perry Nelson, Kwaku Agbesi, et al.
+- **abs:** https://arxiv.org/abs/2602.02734 · **PDF:** https://arxiv.org/pdf/2602.02734v3
+
+## 16334. Predicting first-episode homelessness among US Veterans using longitudinal EHR data: time-varying models and social risk factors
+
+- **ID:** `2602.02731` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Rohan Pandey, Haijuan Yan, Hong Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.02731 · **PDF:** https://arxiv.org/pdf/2602.02731v1
+
+## 16335. CAPS: Unifying Attention, Recurrence, and Alignment in Transformer-based Time Series Forecasting
+
+- **ID:** `2602.02729` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Viresh Pati, Yubin Kim, Vinh Pham, et al.
+- **abs:** https://arxiv.org/abs/2602.02729 · **PDF:** https://arxiv.org/pdf/2602.02729v1
+
+## 16336. Search-Augmented Masked Diffusion Models for Constrained Generation
+
+- **ID:** `2602.02727` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Huu Binh Ta, Michael Cardei, Alvaro Velasquez, et al.
+- **abs:** https://arxiv.org/abs/2602.02727 · **PDF:** https://arxiv.org/pdf/2602.02727v1
+
+## 16337. Vector Quantized Latent Concepts: A Scalable Alternative to Clustering-Based Concept Discovery
+
+- **ID:** `2602.02726` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Xuemin Yu, Ankur Garg, Samira Ebrahimi Kahou, et al.
+- **abs:** https://arxiv.org/abs/2602.02726 · **PDF:** https://arxiv.org/pdf/2602.02726v1
+
+## 16338. Automated Dysphagia Screening Using Noninvasive Neck Acoustic Sensing
+
+- **ID:** `2602.02725` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Jade Chng, Rong Xing, Yunfei Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.02725 · **PDF:** https://arxiv.org/pdf/2602.02725v1
+
+## 16339. Automatic Design of Optimization Test Problems with Large Language Models
+
+- **ID:** `2602.02724` · **primary:** `cs.NE` · **Submitted:** 2026-02-02
+- **Authors:** Wojciech Achtelik, Hubert Guzowski, Maciej Smołka, et al.
+- **abs:** https://arxiv.org/abs/2602.02724 · **PDF:** https://arxiv.org/pdf/2602.02724v1
+
+## 16340. Hierarchical Entity-centric Reinforcement Learning with Factored Subgoal Diffusion
+
+- **ID:** `2602.02722` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Dan Haramati, Carl Qi, Tal Daniel, et al.
+- **abs:** https://arxiv.org/abs/2602.02722 · **PDF:** https://arxiv.org/pdf/2602.02722v1
+
+## 16341. Neural Probabilistic Amplitude Shaping for Nonlinear Fiber Channels
+
+- **ID:** `2602.02716` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Mohammad Taha Askari, Lutz Lampe, Amirhossein Ghazisaeidi
+- **abs:** https://arxiv.org/abs/2602.02716 · **PDF:** https://arxiv.org/pdf/2602.02716v1
+
+## 16342. TruKAN: Towards More Efficient Kolmogorov-Arnold Networks Using Truncated Power Functions
+
+- **ID:** `2602.03879` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Ali Bayeh, Samira Sadaoui, Malek Mouhoub
+- **abs:** https://arxiv.org/abs/2602.03879 · **PDF:** https://arxiv.org/pdf/2602.03879v1
+
+## 16343. Towards Understanding Steering Strength
+
+- **ID:** `2602.02712` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Magamed Taimeskhanov, Samuel Vaiter, Damien Garreau
+- **abs:** https://arxiv.org/abs/2602.02712 · **PDF:** https://arxiv.org/pdf/2602.02712v1
+
+## 16344. Dynamic Mix Precision Routing for Efficient Multi-step LLM Interaction
+
+- **ID:** `2602.02711` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Yuanzhe Li, Jianing Deng, Jingtong Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.02711 · **PDF:** https://arxiv.org/pdf/2602.02711v1
+
+## 16345. Maximum Likelihood Reinforcement Learning
+
+- **ID:** `2602.02710` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Fahim Tajwar, Guanning Zeng, Yueer Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.02710 · **PDF:** https://arxiv.org/pdf/2602.02710v1
+
+## 16346. ATLAS : Adaptive Self-Evolutionary Research Agent with Task-Distributed Multi-LLM Supporters
+
+- **ID:** `2602.02709` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Ujin Jeon, Jiyong Kwon, Madison Ann Sullivan, et al.
+- **abs:** https://arxiv.org/abs/2602.02709 · **PDF:** https://arxiv.org/pdf/2602.02709v2
+
+## 16347. BinaryPPO: Efficient Policy Optimization for Binary Classification
+
+- **ID:** `2602.02708` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Punya Syon Pandey, Zhijing Jin
+- **abs:** https://arxiv.org/abs/2602.02708 · **PDF:** https://arxiv.org/pdf/2602.02708v1
+
+## 16348. Every Bit Counts: A Theoretical Study of Precision-Expressivity Tradeoffs in Quantized Transformers
+
+- **ID:** `2602.02707` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Sayak Chakrabarti, Toniann Pitassi, Josh Alman
+- **abs:** https://arxiv.org/abs/2602.02707 · **PDF:** https://arxiv.org/pdf/2602.02707v1
+
+## 16349. InfMem: Learning System-2 Memory Control for Long-Context Agent
+
+- **ID:** `2602.02704` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Xinyu Wang, Mingze Li, Peng Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.02704 · **PDF:** https://arxiv.org/pdf/2602.02704v1
+
+## 16350. "Humans welcome to observe": A First Look at the Agent Social Network Moltbook
+
+- **ID:** `2602.10127` · **primary:** `cs.SI` · **Submitted:** 2026-02-02
+- **Authors:** Yukun Jiang, Yage Zhang, Xinyue Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.10127 · **PDF:** https://arxiv.org/pdf/2602.10127v1
+
+## 16351. Sparsely Supervised Diffusion
+
+- **ID:** `2602.02699` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Wenshuai Zhao, Zhiyuan Li, Yi Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.02699 · **PDF:** https://arxiv.org/pdf/2602.02699v1
+
+## 16352. NSC-SL: A Bandwidth-Aware Neural Subspace Compression for Communication-Efficient Split Learning
+
+- **ID:** `2602.02696` · **primary:** `cs.NI` · **Submitted:** 2026-02-02
+- **Authors:** Zhen Fang, Miao Yang, Zehang Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.02696 · **PDF:** https://arxiv.org/pdf/2602.02696v1
+
+## 16353. Eidolon: A Practical Post-Quantum Signature Scheme Based on k-Colorability in the Age of Graph Neural Networks
+
+- **ID:** `2602.02689` · **primary:** `cs.CR` · **Submitted:** 2026-02-02
+- **Authors:** Asmaa Cherkaoui, Ramon Flores, Delaram Kahrobaei, et al.
+- **abs:** https://arxiv.org/abs/2602.02689 · **PDF:** https://arxiv.org/pdf/2602.02689v1
+
+## 16354. Monotonicity as an Architectural Bias for Robust Language Models
+
+- **ID:** `2602.02686` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Patrick Cooper, Alireza Nadali, Ashutosh Trivedi, et al.
+- **abs:** https://arxiv.org/abs/2602.02686 · **PDF:** https://arxiv.org/pdf/2602.02686v1
+
+## 16355. Expert-Data Alignment Governs Generation Quality in Decentralized Diffusion Models
+
+- **ID:** `2602.02685` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Marcos Villagra, Bidhan Roy, Raihan Seraj, et al.
+- **abs:** https://arxiv.org/abs/2602.02685 · **PDF:** https://arxiv.org/pdf/2602.02685v2
+
+## 16356. FlexRank: Nested Low-Rank Knowledge Decomposition for Adaptive Model Deployment
+
+- **ID:** `2602.02680` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Riccardo Zaccone, Stefanos Laskaridis, Marco Ciccone, et al.
+- **abs:** https://arxiv.org/abs/2602.02680 · **PDF:** https://arxiv.org/pdf/2602.02680v1
+
+## 16357. MARA: Continuous SE(3)-Equivariant Attention for Molecular Force Fields
+
+- **ID:** `2602.02671` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Francesco Leonardi, Boris Bonev, Kaspar Riesen
+- **abs:** https://arxiv.org/abs/2602.02671 · **PDF:** https://arxiv.org/pdf/2602.02671v1
+
+## 16358. MARS: Modular Agent with Reflective Search for Automated AI Research
+
+- **ID:** `2602.02660` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Jiefeng Chen, Bhavana Dalvi Mishra, Jaehyun Nam, et al.
+- **abs:** https://arxiv.org/abs/2602.02660 · **PDF:** https://arxiv.org/pdf/2602.02660v2
+
+## 16359. Reward-free Alignment for Conflicting Objectives
+
+- **ID:** `2602.02495` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Peter L. Chen, Xiaopeng Li, Xi Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.02495 · **PDF:** https://arxiv.org/pdf/2602.02495v2
+
+## 16360. MEG-XL: Data-Efficient Brain-to-Text via Long-Context Pre-Training
+
+- **ID:** `2602.02494` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Dulhan Jayalath, Oiwi Parker Jones
+- **abs:** https://arxiv.org/abs/2602.02494 · **PDF:** https://arxiv.org/pdf/2602.02494v1
+
+## 16361. PixelGen: Pixel Diffusion Beats Latent Diffusion with Perceptual Loss
+
+- **ID:** `2602.02493` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Zehong Ma, Ruihan Xu, Shiliang Zhang
+- **abs:** https://arxiv.org/abs/2602.02493 · **PDF:** https://arxiv.org/pdf/2602.02493v1
+
+## 16362. New explanations and inference for least angle regression
+
+- **ID:** `2602.02491` · **primary:** `math.ST` · **Submitted:** 2026-02-02
+- **Authors:** Karl B. Gregory, Daniel J. Nordman
+- **abs:** https://arxiv.org/abs/2602.02491 · **PDF:** https://arxiv.org/pdf/2602.02491v1
+
+## 16363. RLAnything: Forge Environment, Policy, and Reward Model in Completely Dynamic RL System
+
+- **ID:** `2602.02488` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yinjie Wang, Tianbao Xie, Ke Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.02488 · **PDF:** https://arxiv.org/pdf/2602.02488v1
+
+## 16364. RE-TRAC: REcursive TRAjectory Compression for Deep Search Agents
+
+- **ID:** `2602.02486` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Jialiang Zhu, Gongrui Zhang, Xiaolong Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.02486 · **PDF:** https://arxiv.org/pdf/2602.02486v1
+
+## 16365. Expanding the Capabilities of Reinforcement Learning via Text Feedback
+
+- **ID:** `2602.02482` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yuda Song, Lili Chen, Fahim Tajwar, et al.
+- **abs:** https://arxiv.org/abs/2602.02482 · **PDF:** https://arxiv.org/pdf/2602.02482v2
+
+## 16366. Flow Policy Gradients for Robot Control
+
+- **ID:** `2602.02481` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Brent Yi, Hongsuk Choi, Himanshu Gaurav Singh, et al.
+- **abs:** https://arxiv.org/abs/2602.02481 · **PDF:** https://arxiv.org/pdf/2602.02481v1
+
+## 16367. Benchmarking Large Language Models for Zero-shot and Few-shot Phishing URL Detection
+
+- **ID:** `2602.02641` · **primary:** `cs.CR` · **Submitted:** 2026-02-02
+- **Authors:** Najmul Hasan, Prashanth BusiReddyGari
+- **abs:** https://arxiv.org/abs/2602.02641 · **PDF:** https://arxiv.org/pdf/2602.02641v1
+
+## 16368. Training LLMs for Divide-and-Conquer Reasoning Elevates Test-Time Scalability
+
+- **ID:** `2602.02477` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Xiao Liang, Zhong-Zhi Li, Zhenghao Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.02477 · **PDF:** https://arxiv.org/pdf/2602.02477v1
+
+## 16369. A Positive Case for Faithfulness: LLM Self-Explanations Help Predict Model Behavior
+
+- **ID:** `2602.02639` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Harry Mayne, Justin Singh Kang, Dewi Gould, et al.
+- **abs:** https://arxiv.org/abs/2602.02639 · **PDF:** https://arxiv.org/pdf/2602.02639v1
+
+## 16370. AgentRx: Diagnosing AI Agent Failures from Execution Trajectories
+
+- **ID:** `2602.02475` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Shraddha Barke, Arnav Goyal, Alind Khare, et al.
+- **abs:** https://arxiv.org/abs/2602.02475 · **PDF:** https://arxiv.org/pdf/2602.02475v1
+
+## 16371. MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents
+
+- **ID:** `2602.02474` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Haozhen Zhang, Quanyu Long, Jianzhu Bao, et al.
+- **abs:** https://arxiv.org/abs/2602.02474 · **PDF:** https://arxiv.org/pdf/2602.02474v1
+
+## 16372. HumanX: Toward Agile and Generalizable Humanoid Interaction Skills from Human Videos
+
+- **ID:** `2602.02473` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Yinhuai Wang, Qihan Zhao, Yuen Fui Lau, et al.
+- **abs:** https://arxiv.org/abs/2602.02473 · **PDF:** https://arxiv.org/pdf/2602.02473v1
+
+## 16373. SPARKLING: Balancing Signal Preservation and Symmetry Breaking for Width-Progressive Learning
+
+- **ID:** `2602.02472` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Qifan Yu, Xinyu Ma, Zhijian Zhuo, et al.
+- **abs:** https://arxiv.org/abs/2602.02472 · **PDF:** https://arxiv.org/pdf/2602.02472v1
+
+## 16374. Multi-head automated segmentation by incorporating detection head into the contextual layer neural network
+
+- **ID:** `2602.02471` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Edwin Kys, Febian Febian
+- **abs:** https://arxiv.org/abs/2602.02471 · **PDF:** https://arxiv.org/pdf/2602.02471v1
+
+## 16375. Breaking the Reversal Curse in Autoregressive Language Models via Identity Bridge
+
+- **ID:** `2602.02470` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Xutao Ma, Yixiao Huang, Hanlin Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.02470 · **PDF:** https://arxiv.org/pdf/2602.02470v1
+
+## 16376. Age-Aware Edge-Blind Federated Learning via Over-the-Air Aggregation
+
+- **ID:** `2602.02469` · **primary:** `cs.IT` · **Submitted:** 2026-02-02
+- **Authors:** Ahmed M. Elshazly, Ahmed Arafa
+- **abs:** https://arxiv.org/abs/2602.02469 · **PDF:** https://arxiv.org/pdf/2602.02469v2
+
+## 16377. Avenir-Web: Human-Experience-Imitating Multimodal Web Agents with Mixture of Grounding Experts
+
+- **ID:** `2602.02468` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Aiden Yiliu Li, Xinyue Hao, Shilong Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.02468 · **PDF:** https://arxiv.org/pdf/2602.02468v1
+
+## 16378. Indications of Belief-Guided Agency and Meta-Cognitive Monitoring in Large Language Models
+
+- **ID:** `2602.02467` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Noam Steinmetz Yalon, Ariel Goldstein, Liad Mudrik, et al.
+- **abs:** https://arxiv.org/abs/2602.02467 · **PDF:** https://arxiv.org/pdf/2602.02467v1
+
+## 16379. MentisOculi: Revealing the Limits of Reasoning with Mental Imagery
+
+- **ID:** `2602.02465` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Jana Zeller, Thaddäus Wiedemer, Fanfei Li, et al.
+- **abs:** https://arxiv.org/abs/2602.02465 · **PDF:** https://arxiv.org/pdf/2602.02465v1
+
+## 16380. From Directions to Regions: Decomposing Activations in Language Models via Local Geometry
+
+- **ID:** `2602.02464` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Or Shafran, Shaked Ronen, Omri Fahn, et al.
+- **abs:** https://arxiv.org/abs/2602.02464 · **PDF:** https://arxiv.org/pdf/2602.02464v1
+
+## 16381. Abstract Activation Spaces for Content-Invariant Reasoning in Large Language Models
+
+- **ID:** `2602.02462` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Gabriele Maraia, Marco Valentino, Fabio Massimo Zanzotto, et al.
+- **abs:** https://arxiv.org/abs/2602.02462 · **PDF:** https://arxiv.org/pdf/2602.02462v1
+
+## 16382. Conflict-Aware Client Selection for Multi-Server Federated Learning
+
+- **ID:** `2602.02458` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Mingwei Hong, Zheng Lin, Zehang Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.02458 · **PDF:** https://arxiv.org/pdf/2602.02458v1
+
+## 16383. Drift-Bench: Diagnosing Cooperative Breakdowns in LLM Agents under Input Faults via Multi-Turn Interaction
+
+- **ID:** `2602.02455` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Han Bao, Zheyuan Zhang, Pengcheng Jing, et al.
+- **abs:** https://arxiv.org/abs/2602.02455 · **PDF:** https://arxiv.org/pdf/2602.02455v1
+
+## 16384. World-Gymnast: Training Robots with Reinforcement Learning in a World Model
+
+- **ID:** `2602.02454` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Ansh Kumar Sharma, Yixiang Sun, Ninghao Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.02454 · **PDF:** https://arxiv.org/pdf/2602.02454v1
+
+## 16385. Thinking with Comics: Enhancing Multimodal Reasoning through Structured Visual Storytelling
+
+- **ID:** `2602.02453` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Andong Chen, Wenxin Zhu, Qiuyu Ding, et al.
+- **abs:** https://arxiv.org/abs/2602.02453 · **PDF:** https://arxiv.org/pdf/2602.02453v2
+
+## 16386. Active Causal Experimentalist (ACE): Learning Intervention Strategies via Direct Preference Optimization
+
+- **ID:** `2602.02451` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Patrick Cooper, Alvaro Velasquez
+- **abs:** https://arxiv.org/abs/2602.02451 · **PDF:** https://arxiv.org/pdf/2602.02451v1
+
+## 16387. Finite-Sample Wasserstein Error Bounds and Concentration Inequalities for Nonlinear Stochastic Approximation
+
+- **ID:** `2602.02445` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Seo Taek Kong, R. Srikant
+- **abs:** https://arxiv.org/abs/2602.02445 · **PDF:** https://arxiv.org/pdf/2602.02445v1
+
+## 16388. hSNMF: Hybrid Spatially Regularized NMF for Image-Derived Spatial Transcriptomics
+
+- **ID:** `2602.02638` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Md Ishtyaq Mahmud, Veena Kochat, Suresh Satpati, et al.
+- **abs:** https://arxiv.org/abs/2602.02638 · **PDF:** https://arxiv.org/pdf/2602.02638v1
+
+## 16389. Certain Head, Uncertain Tail: Expert-Sample for Test-Time Scaling in Fine-Grained MoE
+
+- **ID:** `2602.02443` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yuanteng Chen, Peisong Wang, Nanxin Zeng, et al.
+- **abs:** https://arxiv.org/abs/2602.02443 · **PDF:** https://arxiv.org/pdf/2602.02443v1
+
+## 16390. Large Language Models for Mental Health: A Multilingual Evaluation
+
+- **ID:** `2602.02440` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Nishat Raihan, Sadiya Sayara Chowdhury Puspo, Ana-Maria Bucur, et al.
+- **abs:** https://arxiv.org/abs/2602.02440 · **PDF:** https://arxiv.org/pdf/2602.02440v1
+
+## 16391. Energy-Efficient Neuromorphic Computing for Edge AI: A Framework with Adaptive Spiking Neural Networks and Hardware-Aware Optimization
+
+- **ID:** `2602.02439` · **primary:** `cs.NE` · **Submitted:** 2026-02-02
+- **Authors:** Olaf Yunus Laitinen Imanov, Derya Umut Kulali, Taner Yilmaz, et al.
+- **abs:** https://arxiv.org/abs/2602.02439 · **PDF:** https://arxiv.org/pdf/2602.02439v1
+
+## 16392. UniReason 1.0: A Unified Reasoning Framework for World Knowledge Aligned Image Generation and Editing
+
+- **ID:** `2602.02437` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Dianyi Wang, Chaofan Ma, Feng Han, et al.
+- **abs:** https://arxiv.org/abs/2602.02437 · **PDF:** https://arxiv.org/pdf/2602.02437v4
+
+## 16393. WideSeek: Advancing Wide Research via Multi-Agent Scaling
+
+- **ID:** `2602.02636` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Ziyang Huang, Haolin Ren, Xiaowei Yuan, et al.
+- **abs:** https://arxiv.org/abs/2602.02636 · **PDF:** https://arxiv.org/pdf/2602.02636v1
+
+## 16394. Maximizing Reliability with Bayesian Optimization
+
+- **ID:** `2602.02432` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Jack M. Buckingham, Ivo Couckuyt, Juergen Branke
+- **abs:** https://arxiv.org/abs/2602.02432 · **PDF:** https://arxiv.org/pdf/2602.02432v1
+
+## 16395. Full-Batch Gradient Descent Outperforms One-Pass SGD: Sample Complexity Separation in Single-Index Learning
+
+- **ID:** `2602.02431` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** Filip Kovačević, Hong Chang Ji, Denny Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.02431 · **PDF:** https://arxiv.org/pdf/2602.02431v1
+
+## 16396. Graph-Augmented Reasoning with Large Language Models for Tobacco Pest and Disease Management
+
+- **ID:** `2602.02635` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Siyu Li, Chenwei Song, Qi Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.02635 · **PDF:** https://arxiv.org/pdf/2602.02635v1
+
+## 16397. Embedding Perturbation may Better Reflect the Uncertainty in LLM Reasoning
+
+- **ID:** `2602.02427` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Qihao Wen, Jiahao Wang, Yang Nan, et al.
+- **abs:** https://arxiv.org/abs/2602.02427 · **PDF:** https://arxiv.org/pdf/2602.02427v1
+
+## 16398. DPBench: Large Language Models Struggle with Simultaneous Coordination
+
+- **ID:** `2602.13255` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Najmul Hasan, Prashanth BusiReddyGari
+- **abs:** https://arxiv.org/abs/2602.13255 · **PDF:** https://arxiv.org/pdf/2602.13255v1
+
+## 16399. Repurposing Protein Language Models for Latent Flow-Based Fitness Optimization
+
+- **ID:** `2602.02425` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Amaru Caceres Arroyo, Lea Bogensperger, Ahmed Allam, et al.
+- **abs:** https://arxiv.org/abs/2602.02425 · **PDF:** https://arxiv.org/pdf/2602.02425v1
+
+## 16400. Poly-attention: a general scheme for higher-order self-attention
+
+- **ID:** `2602.02422` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Sayak Chakrabarti, Toniann Pitassi, Josh Alman
+- **abs:** https://arxiv.org/abs/2602.02422 · **PDF:** https://arxiv.org/pdf/2602.02422v1
+
+## 16401. SafeGround: Know When to Trust GUI Grounding Models via Uncertainty Calibration
+
+- **ID:** `2602.02419` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Qingni Wang, Yue Fan, Xin Eric Wang
+- **abs:** https://arxiv.org/abs/2602.02419 · **PDF:** https://arxiv.org/pdf/2602.02419v2
+
+## 16402. Trust Region Continual Learning as an Implicit Meta-Learner
+
+- **ID:** `2602.02417` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zekun Wang, Anant Gupta, Christopher J. MacLellan
+- **abs:** https://arxiv.org/abs/2602.02417 · **PDF:** https://arxiv.org/pdf/2602.02417v1
+
+## 16403. HQP: Sensitivity-Aware Hybrid Quantization and Pruning for Ultra-Low-Latency Edge AI Inference
+
+- **ID:** `2602.06069` · **primary:** `cs.DC` · **Submitted:** 2026-02-02
+- **Authors:** Dinesh Gopalan, Ratul Ali
+- **abs:** https://arxiv.org/abs/2602.06069 · **PDF:** https://arxiv.org/pdf/2602.06069v1
+
+## 16404. A Reduction from Delayed to Immediate Feedback for Online Convex Optimization with Improved Guarantees
+
+- **ID:** `2602.02634` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Alexander Ryabchenko, Idan Attias, Daniel M. Roy
+- **abs:** https://arxiv.org/abs/2602.02634 · **PDF:** https://arxiv.org/pdf/2602.02634v1
+
+## 16405. Rethinking Test-Time Training: Tilting The Latent Distribution For Few-Shot Source-Free Adaptation
+
+- **ID:** `2602.02633` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** Tahir Qasim Syed, Behraj Khan
+- **abs:** https://arxiv.org/abs/2602.02633 · **PDF:** https://arxiv.org/pdf/2602.02633v1
+
+## 16406. Structure Enables Effective Self-Localization of Errors in LLMs
+
+- **ID:** `2602.02416` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Ankur Samanta, Akshayaa Magesh, Ayush Jain, et al.
+- **abs:** https://arxiv.org/abs/2602.02416 · **PDF:** https://arxiv.org/pdf/2602.02416v1
+
+## 16407. Active Transfer Bagging: A New Approach for Accelerated Active Learning Acquisition of Data by Combined Transfer Learning and Bagging Based Models
+
+- **ID:** `2602.02415` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Vivienne Pelletier, Daniel J. Rivera, Obinna Nwokonkwo, et al.
+- **abs:** https://arxiv.org/abs/2602.02415 · **PDF:** https://arxiv.org/pdf/2602.02415v1
+
+## 16408. Misconception Diagnosis From Student-Tutor Dialogue: Generate, Retrieve, Rerank
+
+- **ID:** `2602.02414` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Joshua Mitton, Prarthana Bhattacharyya, Digory Smith, et al.
+- **abs:** https://arxiv.org/abs/2602.02414 · **PDF:** https://arxiv.org/pdf/2602.02414v1
+
+## 16409. Masked Autoencoders as Universal Speech Enhancer
+
+- **ID:** `2602.02413` · **primary:** `cs.SD` · **Submitted:** 2026-02-02
+- **Authors:** Rajalaxmi Rajagopalan, Ritwik Giri, Zhiqiang Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.02413 · **PDF:** https://arxiv.org/pdf/2602.02413v1
+
+## 16410. ReasonEdit: Editing Vision-Language Models using Human Reasoning
+
+- **ID:** `2602.02408` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Jiaxing Qiu, Kaihua Hou, Roxana Daneshjou, et al.
+- **abs:** https://arxiv.org/abs/2602.02408 · **PDF:** https://arxiv.org/pdf/2602.02408v3
+
+## 16411. Provably Data-driven Multiple Hyper-parameter Tuning with Structured Loss Function
+
+- **ID:** `2602.02406` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** Tung Quoc Le, Anh Tuan Nguyen, Viet Anh Nguyen
+- **abs:** https://arxiv.org/abs/2602.02406 · **PDF:** https://arxiv.org/pdf/2602.02406v1
+
+## 16412. Didactic to Constructive: Turning Expert Solutions into Learnable Reasoning
+
+- **ID:** `2602.02405` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Ethan Mendes, Jungsoo Park, Alan Ritter
+- **abs:** https://arxiv.org/abs/2602.02405 · **PDF:** https://arxiv.org/pdf/2602.02405v1
+
+## 16413. SoMA: A Real-to-Sim Neural Simulator for Robotic Soft-body Manipulation
+
+- **ID:** `2602.02402` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Mu Huang, Hui Wang, Kerui Ren, et al.
+- **abs:** https://arxiv.org/abs/2602.02402 · **PDF:** https://arxiv.org/pdf/2602.02402v1
+
+## 16414. An Empirical Study on Noisy Data and LLM Pretraining Loss Divergence
+
+- **ID:** `2602.02400` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Qizhen Zhang, Ankush Garg, Jakob Foerster, et al.
+- **abs:** https://arxiv.org/abs/2602.02400 · **PDF:** https://arxiv.org/pdf/2602.02400v1
+
+## 16415. Performance of Small Language Model Pretraining on FABRIC: An Empirical Study
+
+- **ID:** `2602.02632` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Praveen Rao
+- **abs:** https://arxiv.org/abs/2602.02632 · **PDF:** https://arxiv.org/pdf/2602.02632v1
+
+## 16416. PRISM: Performer RS-IMLE for Single-pass Multisensory Imitation Learning
+
+- **ID:** `2602.02396` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Amisha Bhaskar, Pratap Tokekar, Stefano Di Cairano, et al.
+- **abs:** https://arxiv.org/abs/2602.02396 · **PDF:** https://arxiv.org/pdf/2602.02396v1
+
+## 16417. David vs. Goliath: Verifiable Agent-to-Agent Jailbreaking via Reinforcement Learning
+
+- **ID:** `2602.02395` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Samuel Nellessen, Tal Kachman
+- **abs:** https://arxiv.org/abs/2602.02395 · **PDF:** https://arxiv.org/pdf/2602.02395v1
+
+## 16418. Trailer Reimagined: An Innovative, Llm-DRiven, Expressive Automated Movie Summary framework (TRAILDREAMS)
+
+- **ID:** `2602.02630` · **primary:** `cs.MM` · **Submitted:** 2026-02-02
+- **Authors:** Roberto Balestri, Pasquale Cascarano, Mirko Degli Esposti, et al.
+- **abs:** https://arxiv.org/abs/2602.02630 · **PDF:** https://arxiv.org/pdf/2602.02630v1
+
+## 16419. Infinite-World: Scaling Interactive World Models to 1000-Frame Horizons via Pose-Free Hierarchical Memory
+
+- **ID:** `2602.02393` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Ruiqi Wu, Xuanhua He, Meng Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.02393 · **PDF:** https://arxiv.org/pdf/2602.02393v2
+
+## 16420. Personalized Image Generation via Human-in-the-loop Bayesian Optimization
+
+- **ID:** `2602.02388` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Rajalaxmi Rajagopalan, Debottam Dutta, Yu-Lin Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.02388 · **PDF:** https://arxiv.org/pdf/2602.02388v2
+
+## 16421. Trust by Design: Skill Profiles for Transparent, Cost-Aware LLM Routing
+
+- **ID:** `2602.02386` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Mika Okamoto, Ansel Kaplan Erol, Glenn Matlin
+- **abs:** https://arxiv.org/abs/2602.02386 · **PDF:** https://arxiv.org/pdf/2602.02386v1
+
+## 16422. Transformers learn factored representations
+
+- **ID:** `2602.02385` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Adam Shai, Loren Amdahl-Culleton, Casper L. Christensen, et al.
+- **abs:** https://arxiv.org/abs/2602.02385 · **PDF:** https://arxiv.org/pdf/2602.02385v1
+
+## 16423. SLIME: Stabilized Likelihood Implicit Margin Enforcement for Preference Optimization
+
+- **ID:** `2602.02383` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Maksim Afanasyev, Illarion Iov
+- **abs:** https://arxiv.org/abs/2602.02383 · **PDF:** https://arxiv.org/pdf/2602.02383v2
+
+## 16424. Trustworthy Blockchain-based Federated Learning for Electronic Health Records: Securing Participant Identity with Decentralized Identifiers and Verifiable Credentials
+
+- **ID:** `2602.02629` · **primary:** `cs.CR` · **Submitted:** 2026-02-02
+- **Authors:** Rodrigo Tertulino, Ricardo Almeida, Laercio Alencar
+- **abs:** https://arxiv.org/abs/2602.02629 · **PDF:** https://arxiv.org/pdf/2602.02629v1
+
+## 16425. ROG: Retrieval-Augmented LLM Reasoning for Complex First-Order Queries over Knowledge Graphs
+
+- **ID:** `2602.02382` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Ziyan Zhang, Chao Wang, Zhuo Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.02382 · **PDF:** https://arxiv.org/pdf/2602.02382v1
+
+## 16426. Self-Supervised Learning from Structural Invariance
+
+- **ID:** `2602.02381` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yipeng Zhang, Hafez Ghaemi, Jungyoon Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.02381 · **PDF:** https://arxiv.org/pdf/2602.02381v1
+
+## 16427. From Sycophancy to Sensemaking: Premise Governance for Human-AI Decision Making
+
+- **ID:** `2602.02378` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Raunak Jain
+- **abs:** https://arxiv.org/abs/2602.02378 · **PDF:** https://arxiv.org/pdf/2602.02378v2
+
+## 16428. Proof-RM: A Scalable and Generalizable Reward Model for Math Proof
+
+- **ID:** `2602.02377` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Haotong Yang, Zitong Wang, Shijia Kang, et al.
+- **abs:** https://arxiv.org/abs/2602.02377 · **PDF:** https://arxiv.org/pdf/2602.02377v2
+
+## 16429. C-kNN-LSH: A Nearest-Neighbor Algorithm for Sequential Counterfactual Inference
+
+- **ID:** `2602.02371` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Jing Wang, Jie Shen, Qiaomin Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.02371 · **PDF:** https://arxiv.org/pdf/2602.02371v1
+
+## 16430. Live-Evo: Online Evolution of Agentic Memory from Continuous Feedback
+
+- **ID:** `2602.02369` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Yaolun Zhang, Yiran Wu, Yijiong Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.02369 · **PDF:** https://arxiv.org/pdf/2602.02369v1
+
+## 16431. ReasonCACHE: Teaching LLMs To Reason Without Weight Updates
+
+- **ID:** `2602.02366` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Sharut Gupta, Phillip Isola, Stefanie Jegelka, et al.
+- **abs:** https://arxiv.org/abs/2602.02366 · **PDF:** https://arxiv.org/pdf/2602.02366v1
+
+## 16432. SWE-Universe: Scale Real-World Verifiable Environments to Millions
+
+- **ID:** `2602.02361` · **primary:** `cs.SE` · **Submitted:** 2026-02-02
+- **Authors:** Mouxiang Chen, Lei Zhang, Yunlong Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.02361 · **PDF:** https://arxiv.org/pdf/2602.02361v1
+
+## 16433. Automated Multiple Mini Interview (MMI) Scoring
+
+- **ID:** `2602.02360` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Ryan Huynh, Frank Guerin, Alison Callwood
+- **abs:** https://arxiv.org/abs/2602.02360 · **PDF:** https://arxiv.org/pdf/2602.02360v1
+
+## 16434. Transfer Learning Through Conditional Quantile Matching
+
+- **ID:** `2602.02358` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** Yikun Zhang, Steven Wilkins-Reeves, Wesley Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.02358 · **PDF:** https://arxiv.org/pdf/2602.02358v1
+
+## 16435. NAB: Neural Adaptive Binning for Sparse-View CT reconstruction
+
+- **ID:** `2602.02356` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Wangduo Xie, Matthew B. Blaschko
+- **abs:** https://arxiv.org/abs/2602.02356 · **PDF:** https://arxiv.org/pdf/2602.02356v2
+
+## 16436. Hierarchical Federated Learning with SignSGD: A Highly Communication-Efficient Approach
+
+- **ID:** `2602.02355` · **primary:** `cs.DC` · **Submitted:** 2026-02-02
+- **Authors:** Amirreza Kazemi, Seyed Mohammad Azimi-Abarghouyi, Gabor Fodor, et al.
+- **abs:** https://arxiv.org/abs/2602.02355 · **PDF:** https://arxiv.org/pdf/2602.02355v1
+
+## 16437. Implicit neural representation of textures
+
+- **ID:** `2602.02354` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Albert Kwok, Zheyuan Hu, Dounia Hammou
+- **abs:** https://arxiv.org/abs/2602.02354 · **PDF:** https://arxiv.org/pdf/2602.02354v1
+
+## 16438. Artificial Intelligence and Symmetries: Learning, Encoding, and Discovering Structure in Physical Data
+
+- **ID:** `2602.02351` · **primary:** `hep-ph` · **Submitted:** 2026-02-02
+- **Authors:** Veronica Sanz
+- **abs:** https://arxiv.org/abs/2602.02351 · **PDF:** https://arxiv.org/pdf/2602.02351v1
+
+## 16439. Context Learning for Multi-Agent Discussion
+
+- **ID:** `2602.02350` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Xingyuan Hua, Sheng Yue, Xinyi Li, et al.
+- **abs:** https://arxiv.org/abs/2602.02350 · **PDF:** https://arxiv.org/pdf/2602.02350v1
+
+## 16440. Why Steering Works: Toward a Unified View of Language Model Parameter Dynamics
+
+- **ID:** `2602.02343` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Ziwen Xu, Chenyan Wu, Hengyu Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.02343 · **PDF:** https://arxiv.org/pdf/2602.02343v2
+
+## 16441. Rethinking Generative Recommender Tokenizer: Recsys-Native Encoding and Semantic Quantization Beyond LLMs
+
+- **ID:** `2602.02338` · **primary:** `cs.IR` · **Submitted:** 2026-02-02
+- **Authors:** Yu Liang, Zhongjin Zhang, Yuxuan Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.02338 · **PDF:** https://arxiv.org/pdf/2602.02338v1
+
+## 16442. Building a Correct-by-Design Lakehouse. Data Contracts, Versioning, and Transactional Pipelines for Humans and Agents
+
+- **ID:** `2602.02335` · **primary:** `cs.DC` · **Submitted:** 2026-02-02
+- **Authors:** Weiming Sheng, Jinlang Wang, Manuel Barros, et al.
+- **abs:** https://arxiv.org/abs/2602.02335 · **PDF:** https://arxiv.org/pdf/2602.02335v3
+
+## 16443. VQ-Style: Disentangling Style and Content in Motion with Residual Quantized Representations
+
+- **ID:** `2602.02334` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Fatemeh Zargarbashi, Dhruv Agrawal, Jakob Buhmann, et al.
+- **abs:** https://arxiv.org/abs/2602.02334 · **PDF:** https://arxiv.org/pdf/2602.02334v2
+
+## 16444. TTT-Parkour: Rapid Test-Time Training for Perceptive Robot Parkour
+
+- **ID:** `2602.02331` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Shaoting Zhu, Baijun Ye, Jiaxuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.02331 · **PDF:** https://arxiv.org/pdf/2602.02331v1
+
+## 16445. Language Steering for Multilingual In-Context Learning
+
+- **ID:** `2602.02326` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Neeraja Kirtane, Kuan-Hao Huang
+- **abs:** https://arxiv.org/abs/2602.02326 · **PDF:** https://arxiv.org/pdf/2602.02326v2
+
+## 16446. A Large-Scale Dataset for Molecular Structure-Language Description via a Rule-Regularized Method
+
+- **ID:** `2602.02320` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Feiyang Cai, Guijuan He, Yi Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.02320 · **PDF:** https://arxiv.org/pdf/2602.02320v2
+
+## 16447. The Shape of Beliefs: Geometry, Dynamics, and Interventions along Representation Manifolds of Language Models' Posteriors
+
+- **ID:** `2602.02315` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Raphaël Sarfati, Eric Bigelow, Daniel Wurgaft, et al.
+- **abs:** https://arxiv.org/abs/2602.02315 · **PDF:** https://arxiv.org/pdf/2602.02315v1
+
+## 16448. Interpreting and Controlling LLM Reasoning through Integrated Policy Gradient
+
+- **ID:** `2602.02313` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Changming Li, Kaixing Zhang, Haoyun Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.02313 · **PDF:** https://arxiv.org/pdf/2602.02313v2
+
+## 16449. Introns and Templates Matter: Rethinking Linkage in GP-GOMEA
+
+- **ID:** `2602.02311` · **primary:** `cs.NE` · **Submitted:** 2026-02-02
+- **Authors:** Johannes Koch, Tanja Alderliesten, Peter A. N. Bosman
+- **abs:** https://arxiv.org/abs/2602.02311 · **PDF:** https://arxiv.org/pdf/2602.02311v1
+
+## 16450. FragmentFlow: Scalable Transition State Generation for Large Molecules
+
+- **ID:** `2602.02310` · **primary:** `physics.chem-ph` · **Submitted:** 2026-02-02
+- **Authors:** Ron Shprints, Peter Holderrieth, Juno Nam, et al.
+- **abs:** https://arxiv.org/abs/2602.02310 · **PDF:** https://arxiv.org/pdf/2602.02310v2
+
+## 16451. Spark: Modular Spiking Neural Networks
+
+- **ID:** `2602.02306` · **primary:** `cs.NE` · **Submitted:** 2026-02-02
+- **Authors:** Mario Franco, Carlos Gershenson
+- **abs:** https://arxiv.org/abs/2602.02306 · **PDF:** https://arxiv.org/pdf/2602.02306v3
+
+## 16452. Position: Explaining Behavioral Shifts in Large Language Models Requires a Comparative Approach
+
+- **ID:** `2602.02304` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Martino Ciaperoni, Marzio Di Vece, Luca Pappalardo, et al.
+- **abs:** https://arxiv.org/abs/2602.02304 · **PDF:** https://arxiv.org/pdf/2602.02304v1
+
+## 16453. Advancing General-Purpose Reasoning Models with Modular Gradient Surgery
+
+- **ID:** `2602.02301` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Min Cai, Yu Liang, Longzheng Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.02301 · **PDF:** https://arxiv.org/pdf/2602.02301v2
+
+## 16454. vLLM Hook v0: A Plug-in for Programming Model Internals on vLLM
+
+- **ID:** `2603.06588` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Ching-Yun Ko, Pin-Yu Chen
+- **abs:** https://arxiv.org/abs/2603.06588 · **PDF:** https://arxiv.org/pdf/2603.06588v1
+
+## 16455. Decoupling Generalizability and Membership Privacy Risks in Neural Networks
+
+- **ID:** `2602.02296` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Xingli Fang, Jung-Eun Kim
+- **abs:** https://arxiv.org/abs/2602.02296 · **PDF:** https://arxiv.org/pdf/2602.02296v2
+
+## 16456. EvalQReason: A Framework for Step-Level Reasoning Evaluation in Large Language Models
+
+- **ID:** `2602.02295` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Shaima Ahmad Freja, Ferhat Ozgur Catak, Betul Yurdem, et al.
+- **abs:** https://arxiv.org/abs/2602.02295 · **PDF:** https://arxiv.org/pdf/2602.02295v1
+
+## 16457. Learning Better Certified Models from Empirically-Robust Teachers
+
+- **ID:** `2602.02626` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Alessandro De Palma
+- **abs:** https://arxiv.org/abs/2602.02626 · **PDF:** https://arxiv.org/pdf/2602.02626v1
+
+## 16458. Hallucination or Creativity: How to Evaluate AI-Generated Scientific Stories?
+
+- **ID:** `2602.02290` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Alex Argese, Pasquale Lisena, Raphaël Troncy
+- **abs:** https://arxiv.org/abs/2602.02290 · **PDF:** https://arxiv.org/pdf/2602.02290v2
+
+## 16459. An Optimization Method for Autoregressive Time Series Forecasting
+
+- **ID:** `2602.02288` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zheng Li, Jerry Cheng, Huanying Gu
+- **abs:** https://arxiv.org/abs/2602.02288 · **PDF:** https://arxiv.org/pdf/2602.02288v1
+
+## 16460. Cross-Lingual Stability of LLM Judges Under Controlled Generation: Evidence from Finno-Ugric Languages
+
+- **ID:** `2602.02287` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Isaac Chung, Linda Freienthal
+- **abs:** https://arxiv.org/abs/2602.02287 · **PDF:** https://arxiv.org/pdf/2602.02287v1
+
+## 16461. DFKI-Speech System for WildSpoof Challenge: A robust framework for SASV In-the-Wild
+
+- **ID:** `2602.02286` · **primary:** `cs.SD` · **Submitted:** 2026-02-02
+- **Authors:** Arnab Das, Yassine El Kheir, Enes Erdem Erdogan, et al.
+- **abs:** https://arxiv.org/abs/2602.02286 · **PDF:** https://arxiv.org/pdf/2602.02286v1
+
+## 16462. Statistical Learning Theory in Lean 4: Empirical Processes from Scratch
+
+- **ID:** `2602.02285` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yuanhe Zhang, Jason D. Lee, Fanghui Liu
+- **abs:** https://arxiv.org/abs/2602.02285 · **PDF:** https://arxiv.org/pdf/2602.02285v1
+
+## 16463. Choice-Model-Assisted Q-learning for Delayed-Feedback Revenue Management
+
+- **ID:** `2602.02283` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Owen Shen, Patrick Jaillet
+- **abs:** https://arxiv.org/abs/2602.02283 · **PDF:** https://arxiv.org/pdf/2602.02283v1
+
+## 16464. MoLF: Mixture-of-Latent-Flow for Pan-Cancer Spatial Gene Expression Prediction from Histology
+
+- **ID:** `2602.02282` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Susu Hu, Stefanie Speidel
+- **abs:** https://arxiv.org/abs/2602.02282 · **PDF:** https://arxiv.org/pdf/2602.02282v1
+
+## 16465. Recommender system in X inadvertently profiles ideological positions of users
+
+- **ID:** `2602.02624` · **primary:** `cs.SI` · **Submitted:** 2026-02-02
+- **Authors:** Paul Bouchaud, Pedro Ramaciotti
+- **abs:** https://arxiv.org/abs/2602.02624 · **PDF:** https://arxiv.org/pdf/2602.02624v1
+
+## 16466. Backpropagation as Physical Relaxation: Exact Gradients in Finite Time
+
+- **ID:** `2602.02281` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Antonino Emanuele Scurria
+- **abs:** https://arxiv.org/abs/2602.02281 · **PDF:** https://arxiv.org/pdf/2602.02281v1
+
+## 16467. RACA: Representation-Aware Coverage Criteria for LLM Safety Testing
+
+- **ID:** `2602.02280` · **primary:** `cs.SE` · **Submitted:** 2026-02-02
+- **Authors:** Zeming Wei, Zhixin Zhang, Chengcan Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.02280 · **PDF:** https://arxiv.org/pdf/2602.02280v1
+
+## 16468. Kimi K2.5: Visual Agentic Intelligence
+
+- **ID:** `2602.02276` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Kimi Team, Tongtong Bai, Yifan Bai, et al.
+- **abs:** https://arxiv.org/abs/2602.02276 · **PDF:** https://arxiv.org/pdf/2602.02276v1
+
+## 16469. Learning Consistent Causal Abstraction Networks
+
+- **ID:** `2602.02623` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Gabriele D'Acunto, Paolo Di Lorenzo, Sergio Barbarossa
+- **abs:** https://arxiv.org/abs/2602.02623 · **PDF:** https://arxiv.org/pdf/2602.02623v1
+
+## 16470. dziribot: rag based intelligent conversational agent for algerian arabic dialect
+
+- **ID:** `2602.02270` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** El Batoul Bechiri, Dihia Lanasri
+- **abs:** https://arxiv.org/abs/2602.02270 · **PDF:** https://arxiv.org/pdf/2602.02270v1
+
+## 16471. Bridging the Sim-to-Real Gap with multipanda ros2: A Real-Time ROS2 Framework for Multimanual Systems
+
+- **ID:** `2602.02269` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Jon Škerlj, Seongjin Bien, Abdeldjallil Naceri, et al.
+- **abs:** https://arxiv.org/abs/2602.02269 · **PDF:** https://arxiv.org/pdf/2602.02269v1
+
+## 16472. HopFormer: Sparse Graph Transformers with Explicit Receptive Field Control
+
+- **ID:** `2602.02268` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Sanggeon Yun, Raheeb Hassan, Ryozo Masukawa, et al.
+- **abs:** https://arxiv.org/abs/2602.02268 · **PDF:** https://arxiv.org/pdf/2602.02268v1
+
+## 16473. OpenSeal: Good, Fast, and Cheap Construction of an Open-Source Southeast Asian LLM via Parallel Data
+
+- **ID:** `2602.02266` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Tan Sang Nguyen, Muhammad Reza Qorib, Hwee Tou Ng
+- **abs:** https://arxiv.org/abs/2602.02266 · **PDF:** https://arxiv.org/pdf/2602.02266v1
+
+## 16474. Unsupervised Physics-Informed Operator Learning through Multi-Stage Curriculum Training
+
+- **ID:** `2602.02264` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Paolo Marcandelli, Natansh Mathur, Stefano Markidis, et al.
+- **abs:** https://arxiv.org/abs/2602.02264 · **PDF:** https://arxiv.org/pdf/2602.02264v1
+
+## 16475. OmniCode: A Benchmark for Evaluating Software Engineering Agents
+
+- **ID:** `2602.02262` · **primary:** `cs.SE` · **Submitted:** 2026-02-02
+- **Authors:** Atharv Sonwane, Eng-Shen Tu, Wei-Chung Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.02262 · **PDF:** https://arxiv.org/pdf/2602.02262v2
+
+## 16476. Unlocking the Duality between Flow and Field Matching
+
+- **ID:** `2602.02261` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Daniil Shlenskii, Alexander Varlamov, Nazar Buzun, et al.
+- **abs:** https://arxiv.org/abs/2602.02261 · **PDF:** https://arxiv.org/pdf/2602.02261v1
+
+## 16477. Learning Markov Decision Processes under Fully Bandit Feedback
+
+- **ID:** `2602.02260` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zhengjia Zhuo, Anupam Gupta, Viswanath Nagarajan
+- **abs:** https://arxiv.org/abs/2602.02260 · **PDF:** https://arxiv.org/pdf/2602.02260v1
+
+## 16478. Segment to Focus: Guiding Latent Action Models in the Presence of Distractors
+
+- **ID:** `2602.02259` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Hamza Adnan, Matthew T. Jackson, Alexey Zakharov
+- **abs:** https://arxiv.org/abs/2602.02259 · **PDF:** https://arxiv.org/pdf/2602.02259v1
+
+## 16479. Alignment-Aware Model Adaptation via Feedback-Guided Optimization
+
+- **ID:** `2602.02258` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Gaurav Bhatt, Aditya Chinchure, Jiawei Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.02258 · **PDF:** https://arxiv.org/pdf/2602.02258v2
+
+## 16480. Fair Context Learning for Evidence-Balanced Test-Time Adaptation in Vision-Language Models
+
+- **ID:** `2602.07027` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Sanggeon Yun, Ryozo Masukawa, SungHeon Jeong, et al.
+- **abs:** https://arxiv.org/abs/2602.07027 · **PDF:** https://arxiv.org/pdf/2602.07027v1
+
+## 16481. Well-Posed KL-Regularized Control via Wasserstein and Kalman-Wasserstein KL Divergences
+
+- **ID:** `2602.02250` · **primary:** `math.OC` · **Submitted:** 2026-02-02
+- **Authors:** Viktor Stein, Adwait Datar, Nihat Ay
+- **abs:** https://arxiv.org/abs/2602.02250 · **PDF:** https://arxiv.org/pdf/2602.02250v1
+
+## 16482. Learning While Staying Curious: Entropy-Preserving Supervised Fine-Tuning via Adaptive Self-Distillation for Large Reasoning Models
+
+- **ID:** `2602.02244` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Hao Wang, Hao Gu, Hongming Piao, et al.
+- **abs:** https://arxiv.org/abs/2602.02244 · **PDF:** https://arxiv.org/pdf/2602.02244v2
+
+## 16483. Variational Entropic Optimal Transport
+
+- **ID:** `2602.02241` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Roman Dyachenko, Nikita Gushchin, Kirill Sokolov, et al.
+- **abs:** https://arxiv.org/abs/2602.02241 · **PDF:** https://arxiv.org/pdf/2602.02241v1
+
+## 16484. Causal Inference for Preprocessed Outcomes with an Application to Functional Connectivity
+
+- **ID:** `2602.02240` · **primary:** `stat.ME` · **Submitted:** 2026-02-02
+- **Authors:** Zihang Wang, Razieh Nabi, Benjamin B. Risk
+- **abs:** https://arxiv.org/abs/2602.02240 · **PDF:** https://arxiv.org/pdf/2602.02240v1
+
+## 16485. Interpretability in Deep Time Series Models Demands Semantic Alignment
+
+- **ID:** `2602.02239` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Giovanni De Felice, Riccardo D'Elia, Alberto Termine, et al.
+- **abs:** https://arxiv.org/abs/2602.02239 · **PDF:** https://arxiv.org/pdf/2602.02239v1
+
+## 16486. Geometry- and Relation-Aware Diffusion for EEG Super-Resolution
+
+- **ID:** `2602.02238` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Laura Yao, Gengwei Zhang, Moajjem Chowdhury, et al.
+- **abs:** https://arxiv.org/abs/2602.02238 · **PDF:** https://arxiv.org/pdf/2602.02238v1
+
+## 16487. Online Fine-Tuning of Pretrained Controllers for Autonomous Driving via Real-Time Recurrent RL
+
+- **ID:** `2602.02236` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Julian Lemmel, Felix Resch, Mónika Farsang, et al.
+- **abs:** https://arxiv.org/abs/2602.02236 · **PDF:** https://arxiv.org/pdf/2602.02236v3
+
+## 16488. SEDformer: Event-Synchronous Spiking Transformers for Irregular Telemetry Time Series Forecasting
+
+- **ID:** `2602.02230` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Ziyu Zhou, Yuchen Fang, Weilin Ruan, et al.
+- **abs:** https://arxiv.org/abs/2602.02230 · **PDF:** https://arxiv.org/pdf/2602.02230v2
+
+## 16489. Prediction-Powered Risk Monitoring of Deployed Models for Detecting Harmful Distribution Shifts
+
+- **ID:** `2602.02229` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Guangyi Zhang, Yunlong Cai, Guanding Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.02229 · **PDF:** https://arxiv.org/pdf/2602.02229v1
+
+## 16490. Spectral Superposition: A Theory of Feature Geometry
+
+- **ID:** `2602.02224` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Georgi Ivanov, Narmeen Oozeer, Shivam Raval, et al.
+- **abs:** https://arxiv.org/abs/2602.02224 · **PDF:** https://arxiv.org/pdf/2602.02224v1
+
+## 16491. Using Correspondence Patterns to Identify Irregular Words in Cognate sets Through Leave-One-Out Validation
+
+- **ID:** `2602.02221` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Frederic Blum, Johann-Mattis List
+- **abs:** https://arxiv.org/abs/2602.02221 · **PDF:** https://arxiv.org/pdf/2602.02221v1
+
+## 16492. Am I More Pointwise or Pairwise? Revealing Position Bias in Rubric-Based LLM-as-a-Judge
+
+- **ID:** `2602.02219` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Yuzheng Xu, Tosho Hirasawa, Tadashi Kozuno, et al.
+- **abs:** https://arxiv.org/abs/2602.02219 · **PDF:** https://arxiv.org/pdf/2602.02219v1
+
+## 16493. Scientific Theory of a Black-Box: A Life Cycle-Scale XAI Framework Based on Constructive Empiricism
+
+- **ID:** `2602.02215` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Sebastian Müller, Vanessa Toborek, Eike Stadtländer, et al.
+- **abs:** https://arxiv.org/abs/2602.02215 · **PDF:** https://arxiv.org/pdf/2602.02215v1
+
+## 16494. Generating Physically Sound Designs from Text and a Set of Physical Constraints
+
+- **ID:** `2602.02213` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Gregory Barber, Todd C. Henry, Mulugeta A. Haile
+- **abs:** https://arxiv.org/abs/2602.02213 · **PDF:** https://arxiv.org/pdf/2602.02213v1
+
+## 16495. Towards AI Evaluation in Domain-Specific RAG Systems: The AgriHubi Case Study
+
+- **ID:** `2602.02208` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Md. Toufique Hasan, Ayman Asad Khan, Mika Saari, et al.
+- **abs:** https://arxiv.org/abs/2602.02208 · **PDF:** https://arxiv.org/pdf/2602.02208v1
+
+## 16496. Sinhala Physical Common Sense Reasoning Dataset for Global PIQA
+
+- **ID:** `2602.02207` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Nisansa de Silva, Surangika Ranathunga
+- **abs:** https://arxiv.org/abs/2602.02207 · **PDF:** https://arxiv.org/pdf/2602.02207v1
+
+## 16497. Fat-Cat: Document-Driven Metacognitive Multi-Agent System for Complex Reasoning
+
+- **ID:** `2602.02206` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Tong Yang, Yemin Wang, Chaoning Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.02206 · **PDF:** https://arxiv.org/pdf/2602.02206v2
+
+## 16498. Cardinality-Preserving Attention Channels for Graph Transformers in Molecular Property Prediction
+
+- **ID:** `2602.02201` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Abhijit Gupta
+- **abs:** https://arxiv.org/abs/2602.02201 · **PDF:** https://arxiv.org/pdf/2602.02201v5
+
+## 16499. More Than a Quick Glance: Overcoming the Greedy Bias in KV-Cache Compression
+
+- **ID:** `2602.02199` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Aryan Sood, Tanvi Sharma, Vansh Agrawal
+- **abs:** https://arxiv.org/abs/2602.02199 · **PDF:** https://arxiv.org/pdf/2602.02199v1
+
+## 16500. Hierarchical Adaptive Eviction for KV Cache Management in Multimodal Language Models
+
+- **ID:** `2602.02197` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Xindian Ma, Yidi Lu, Peng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.02197 · **PDF:** https://arxiv.org/pdf/2602.02197v1
+
+## 16501. TIDE: Trajectory-based Diagnostic Evaluation of Test-Time Improvement in LLM Agents
+
+- **ID:** `2602.02196` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Hang Yan, Xinyu Che, Fangzhi Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.02196 · **PDF:** https://arxiv.org/pdf/2602.02196v2
+
+## 16502. State Rank Dynamics in Linear Attention LLMs
+
+- **ID:** `2602.02195` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Ao Sun, Hongtao Zhang, Heng Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.02195 · **PDF:** https://arxiv.org/pdf/2602.02195v1
+
+## 16503. ECHO-2: A Large-Scale Distributed Rollout Framework for Cost-Efficient Reinforcement Learning
+
+- **ID:** `2602.02192` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Jie Xiao, Meng Chen, Qingnan Ren, et al.
+- **abs:** https://arxiv.org/abs/2602.02192 · **PDF:** https://arxiv.org/pdf/2602.02192v4
+
+## 16504. PCA of probability measures: Sparse and Dense sampling regimes
+
+- **ID:** `2602.02190` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** Gachon Erell, Jérémie Bigot, Elsa Cazelles
+- **abs:** https://arxiv.org/abs/2602.02190 · **PDF:** https://arxiv.org/pdf/2602.02190v1
+
+## 16505. Reasoning in a Combinatorial and Constrained World: Benchmarking LLMs on Natural-Language Combinatorial Optimization
+
+- **ID:** `2602.02188` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Xia Jiang, Jing Chen, Cong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.02188 · **PDF:** https://arxiv.org/pdf/2602.02188v1
+
+## 16506. Vision-DeepResearch Benchmark: Rethinking Visual and Textual Search for Multimodal Large Language Models
+
+- **ID:** `2602.02185` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Yu Zeng, Wenxuan Huang, Zhen Fang, et al.
+- **abs:** https://arxiv.org/abs/2602.02185 · **PDF:** https://arxiv.org/pdf/2602.02185v2
+
+## 16507. Malware Detection Through Memory Analysis
+
+- **ID:** `2602.02184` · **primary:** `cs.CR` · **Submitted:** 2026-02-02
+- **Authors:** Sarah Nassar
+- **abs:** https://arxiv.org/abs/2602.02184 · **PDF:** https://arxiv.org/pdf/2602.02184v1
+
+## 16508. Evaluating Metalinguistic Knowledge in Large Language Models across the World's Languages
+
+- **ID:** `2602.02182` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Tjaša Arčon, Matej Klemen, Marko Robnik-Šikonja, et al.
+- **abs:** https://arxiv.org/abs/2602.02182 · **PDF:** https://arxiv.org/pdf/2602.02182v2
+
+## 16509. STILL: Selecting Tokens for Intra-Layer Hybrid Attention to Linearize LLMs
+
+- **ID:** `2602.02180` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Weikang Meng, Liangyu Huo, Yadan Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.02180 · **PDF:** https://arxiv.org/pdf/2602.02180v1
+
+## 16510. SurvKAN: A Fully Parametric Survival Model Based on Kolmogorov-Arnold Networks
+
+- **ID:** `2602.02179` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Marina Mastroleo, Alberto Archetti, Federico Mastroleo, et al.
+- **abs:** https://arxiv.org/abs/2602.02179 · **PDF:** https://arxiv.org/pdf/2602.02179v1
+
+## 16511. AR-MAP: Are Autoregressive Large Language Models Implicit Teachers for Diffusion Large Language Models?
+
+- **ID:** `2602.02178` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Liang Lin, Feng Xiong, Zengbin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.02178 · **PDF:** https://arxiv.org/pdf/2602.02178v2
+
+## 16512. Generalized Optimal Classification Trees: A Mixed-Integer Programming Approach
+
+- **ID:** `2602.02173` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Jiancheng Tu, Wenqi Fan, Zhibin Wu
+- **abs:** https://arxiv.org/abs/2602.02173 · **PDF:** https://arxiv.org/pdf/2602.02173v1
+
+## 16513. Self-Evolving Coordination Protocol in Multi-Agent AI Systems: An Exploratory Systems Feasibility Study
+
+- **ID:** `2602.02170` · **primary:** `cs.MA` · **Submitted:** 2026-02-02
+- **Authors:** Jose Manuel de la Chica Rodriguez, Juan Manuel Vera Díaz
+- **abs:** https://arxiv.org/abs/2602.02170 · **PDF:** https://arxiv.org/pdf/2602.02170v1
+
+## 16514. Real-Time 2D LiDAR Object Detection Using Three-Frame RGB Scan Encoding
+
+- **ID:** `2602.02167` · **primary:** `eess.SP` · **Submitted:** 2026-02-02
+- **Authors:** Soheil Behnam Roudsari, Alexandre S. Brandão, Felipe N. Martins
+- **abs:** https://arxiv.org/abs/2602.02167 · **PDF:** https://arxiv.org/pdf/2602.02167v1
+
+## 16515. Co-RedTeam: Orchestrated Security Discovery and Exploitation with LLM Agents
+
+- **ID:** `2602.02164` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Pengfei He, Ash Fox, Lesly Miculicich, et al.
+- **abs:** https://arxiv.org/abs/2602.02164 · **PDF:** https://arxiv.org/pdf/2602.02164v2
+
+## 16516. Interpretable Tabular Foundation Models via In-Context Kernel Regression
+
+- **ID:** `2602.02162` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Ratmir Miftachov, Bruno Charron, Simon Valentin
+- **abs:** https://arxiv.org/abs/2602.02162 · **PDF:** https://arxiv.org/pdf/2602.02162v1
+
+## 16517. Generating Causal Temporal Interaction Graphs for Counterfactual Validation of Temporal Link Prediction
+
+- **ID:** `2602.02161` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Aniq Ur Rahman, Justin P. Coon
+- **abs:** https://arxiv.org/abs/2602.02161 · **PDF:** https://arxiv.org/pdf/2602.02161v1
+
+## 16518. D-CORE: Incentivizing Task Decomposition in Large Reasoning Models for Complex Tool Use
+
+- **ID:** `2602.02160` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Bowen Xu, Shaoyu Wu, Hao Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.02160 · **PDF:** https://arxiv.org/pdf/2602.02160v1
+
+## 16519. Focus-dLLM: Accelerating Long-Context Diffusion LLM Inference via Confidence-Guided Context Focusing
+
+- **ID:** `2602.02159` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Lingkun Long, Yushi Huang, Shihao Bai, et al.
+- **abs:** https://arxiv.org/abs/2602.02159 · **PDF:** https://arxiv.org/pdf/2602.02159v1
+
+## 16520. Traffic-Aware Navigation in Road Networks
+
+- **ID:** `2602.02158` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Sarah Nassar
+- **abs:** https://arxiv.org/abs/2602.02158 · **PDF:** https://arxiv.org/pdf/2602.02158v1
+
+## 16521. Efficient Neural Controlled Differential Equations via Attentive Kernel Smoothing
+
+- **ID:** `2602.02157` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Egor Serov, Ilya Kuleshov, Alexey Zaytsev
+- **abs:** https://arxiv.org/abs/2602.02157 · **PDF:** https://arxiv.org/pdf/2602.02157v1
+
+## 16522. Learning Beyond the Gaussian Data: Learning Dynamics of Neural Networks on an Expressive and Cumulant-Controllable Data Model
+
+- **ID:** `2602.02153` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** Onat Ure, Samet Demir, Zafer Dogan
+- **abs:** https://arxiv.org/abs/2602.02153 · **PDF:** https://arxiv.org/pdf/2602.02153v1
+
+## 16523. Revisiting Adaptive Rounding with Vectorized Reparameterization for LLM Quantization
+
+- **ID:** `2602.02151` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yuli Zhou, Qingxuan Chen, Luca Benini, et al.
+- **abs:** https://arxiv.org/abs/2602.02151 · **PDF:** https://arxiv.org/pdf/2602.02151v1
+
+## 16524. ECHO: Entropy-Confidence Hybrid Optimization for Test-Time Reinforcement Learning
+
+- **ID:** `2602.02150` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Chu Zhao, Enneng Yang, Yuting Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.02150 · **PDF:** https://arxiv.org/pdf/2602.02150v1
+
+## 16525. Back to the Future: Look-ahead Augmentation and Parallel Self-Refinement for Time Series Forecasting
+
+- **ID:** `2602.02146` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Sunho Kim, Susik Yoon
+- **abs:** https://arxiv.org/abs/2602.02146 · **PDF:** https://arxiv.org/pdf/2602.02146v1
+
+## 16526. Learning Generative Selection for Best-of-N
+
+- **ID:** `2602.02143` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Shubham Toshniwal, Aleksander Ficek, Siddhartha Jain, et al.
+- **abs:** https://arxiv.org/abs/2602.02143 · **PDF:** https://arxiv.org/pdf/2602.02143v1
+
+## 16527. Quantifying the Gap between Understanding and Generation within Unified Multimodal Models
+
+- **ID:** `2602.02140` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Chenlong Wang, Yuhang Chen, Zhihan Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.02140 · **PDF:** https://arxiv.org/pdf/2602.02140v1
+
+## 16528. EvoMU: Evolutionary Machine Unlearning
+
+- **ID:** `2602.02139` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Pawel Batorski, Paul Swoboda
+- **abs:** https://arxiv.org/abs/2602.02139 · **PDF:** https://arxiv.org/pdf/2602.02139v2
+
+## 16529. CAM: A Causality-based Analysis Framework for Multi-Agent Code Generation Systems
+
+- **ID:** `2602.02138` · **primary:** `cs.SE` · **Submitted:** 2026-02-02
+- **Authors:** Zongyi Lyu, Zhenlan Ji, Songqiang Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.02138 · **PDF:** https://arxiv.org/pdf/2602.02138v2
+
+## 16530. DCoPilot: Generative AI-Empowered Policy Adaptation for Dynamic Data Center Operations
+
+- **ID:** `2602.02137` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Minghao Li, Ruihang Wang, Rui Tan, et al.
+- **abs:** https://arxiv.org/abs/2602.02137 · **PDF:** https://arxiv.org/pdf/2602.02137v3
+
+## 16531. Mitigating Safety Tax via Distribution-Grounded Refinement in Large Reasoning Models
+
+- **ID:** `2602.02136` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Yingsha Xie, Tiansheng Huang, Enneng Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.02136 · **PDF:** https://arxiv.org/pdf/2602.02136v1
+
+## 16532. Understanding the Reversal Curse Mitigation in Masked Diffusion Models through Attention and Training Dynamics
+
+- **ID:** `2602.02133` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Sangwoo Shin, BumJun Kim, Kyelim Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.02133 · **PDF:** https://arxiv.org/pdf/2602.02133v1
+
+## 16533. There Is More to Refusal in Large Language Models than a Single Direction
+
+- **ID:** `2602.02132` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Faaiz Joad, Majd Hawasly, Sabri Boughorbel, et al.
+- **abs:** https://arxiv.org/abs/2602.02132 · **PDF:** https://arxiv.org/pdf/2602.02132v1
+
+## 16534. Scalable Spatio-Temporal SE(3) Diffusion for Long-Horizon Protein Dynamics
+
+- **ID:** `2602.02128` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Nima Shoghi, Yuxuan Liu, Yuning Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.02128 · **PDF:** https://arxiv.org/pdf/2602.02128v2
+
+## 16535. Two-Stage Grid Optimization for Group-wise Quantization of LLMs
+
+- **ID:** `2602.02126` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Junhan Kim, Gukryeol Lee, Seungwoo Son, et al.
+- **abs:** https://arxiv.org/abs/2602.02126 · **PDF:** https://arxiv.org/pdf/2602.02126v1
+
+## 16536. Toxicity Assessment in Preclinical Histopathology via Class-Aware Mahalanobis Distance for Known and Novel Anomalies
+
+- **ID:** `2602.02124` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Olga Graf, Dhrupal Patel, Peter Groß, et al.
+- **abs:** https://arxiv.org/abs/2602.02124 · **PDF:** https://arxiv.org/pdf/2602.02124v1
+
+## 16537. Modality Gap-Driven Subspace Alignment Training Paradigm For Multimodal Large Language Models
+
+- **ID:** `2602.07026` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Xiaomin Yu, Yi Xin, Wenjie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.07026 · **PDF:** https://arxiv.org/pdf/2602.07026v1
+
+## 16538. The Maximum von Neumann Entropy Principle: Theory and Applications in Machine Learning
+
+- **ID:** `2602.02117` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Youqi Wu, Farzan Farnia
+- **abs:** https://arxiv.org/abs/2602.02117 · **PDF:** https://arxiv.org/pdf/2602.02117v1
+
+## 16539. Implicit Bias in LLMs for Transgender Populations
+
+- **ID:** `2602.13253` · **primary:** `cs.CY` · **Submitted:** 2026-02-02
+- **Authors:** Micaela Hirsch, Marina Elichiry, Blas Radi, et al.
+- **abs:** https://arxiv.org/abs/2602.13253 · **PDF:** https://arxiv.org/pdf/2602.13253v1
+
+## 16540. Enhancing Diffusion-Based Quantitatively Controllable Image Generation via Matrix-Form EDM and Adaptive Vicinal Training
+
+- **ID:** `2602.02114` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Xin Ding, Yun Chen, Sen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.02114 · **PDF:** https://arxiv.org/pdf/2602.02114v1
+
+## 16541. Training-free score-based diffusion for parameter-dependent stochastic dynamical systems
+
+- **ID:** `2602.02113` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** Minglei Yang, Sicheng He
+- **abs:** https://arxiv.org/abs/2602.02113 · **PDF:** https://arxiv.org/pdf/2602.02113v1
+
+## 16542. Unifying Masked Diffusion Models with Various Generation Orders and Beyond
+
+- **ID:** `2602.02112` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Chunsan Hong, Sanghyun Lee, Jong Chul Ye
+- **abs:** https://arxiv.org/abs/2602.02112 · **PDF:** https://arxiv.org/pdf/2602.02112v1
+
+## 16543. An Empirical Study of World Model Quantization
+
+- **ID:** `2602.02110` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zhongqian Fu, Tianyi Zhao, Kai Han, et al.
+- **abs:** https://arxiv.org/abs/2602.02110 · **PDF:** https://arxiv.org/pdf/2602.02110v1
+
+## 16544. Out of the Memory Barrier: A Highly Memory Efficient Training System for LLMs with Million-Token Contexts
+
+- **ID:** `2602.02108` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Wenhao Li, Daohai Yu, Gen Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.02108 · **PDF:** https://arxiv.org/pdf/2602.02108v4
+
+## 16545. Dicta-LM 3.0: Advancing The Frontier of Hebrew Sovereign LLMs
+
+- **ID:** `2602.02104` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Shaltiel Shmidman, Avi Shmidman, Amir DN Cohen, et al.
+- **abs:** https://arxiv.org/abs/2602.02104 · **PDF:** https://arxiv.org/pdf/2602.02104v1
+
+## 16546. No Global Plan in Chain-of-Thought: Uncover the Latent Planning Horizon of LLMs
+
+- **ID:** `2602.02103` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Liyan Xu, Mo Yu, Fandong Meng, et al.
+- **abs:** https://arxiv.org/abs/2602.02103 · **PDF:** https://arxiv.org/pdf/2602.02103v1
+
+## 16547. The Verification Crisis: Expert Perceptions of GenAI Disinformation and the Case for Reproducible Provenance
+
+- **ID:** `2602.02100` · **primary:** `cs.CY` · **Submitted:** 2026-02-02
+- **Authors:** Alexander Loth, Martin Kappes, Marc-Oliver Pahl
+- **abs:** https://arxiv.org/abs/2602.02100 · **PDF:** https://arxiv.org/pdf/2602.02100v1
+
+## 16548. Think Dense, Not Long: Dynamic Decoupled Conditional Advantage for Efficient Reasoning
+
+- **ID:** `2602.02099` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Keqin Peng, Yuanxin Ouyang, Xuebo Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.02099 · **PDF:** https://arxiv.org/pdf/2602.02099v1
+
+## 16549. Probabilistic Performance Guarantees for Multi-Task Reinforcement Learning
+
+- **ID:** `2602.02098` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yannik Schnitzer, Mathias Jackermeier, Alessandro Abate, et al.
+- **abs:** https://arxiv.org/abs/2602.02098 · **PDF:** https://arxiv.org/pdf/2602.02098v1
+
+## 16550. WADEPre: A Wavelet-based Decomposition Model for Extreme Precipitation Nowcasting with Multi-Scale Learning
+
+- **ID:** `2602.02096` · **primary:** `physics.ao-ph` · **Submitted:** 2026-02-02
+- **Authors:** Baitian Liu, Haiping Zhang, Huiling Yuan, et al.
+- **abs:** https://arxiv.org/abs/2602.02096 · **PDF:** https://arxiv.org/pdf/2602.02096v1
+
+## 16551. LEC-KG: An LLM-Embedding Collaborative Framework for Domain-Specific Knowledge Graph Construction -- A Case Study on SDGs
+
+- **ID:** `2602.02090` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Yikai Zeng, Yingchao Piao, Changhua Pei, et al.
+- **abs:** https://arxiv.org/abs/2602.02090 · **PDF:** https://arxiv.org/pdf/2602.02090v2
+
+## 16552. CryoLVM: Self-supervised Learning from Cryo-EM Density Maps with Large Vision Models
+
+- **ID:** `2602.02620` · **primary:** `q-bio.QM` · **Submitted:** 2026-02-02
+- **Authors:** Weining Fu, Kai Shu, Kui Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.02620 · **PDF:** https://arxiv.org/pdf/2602.02620v2
+
+## 16553. Efficient Swap Regret Minimization in Combinatorial Bandits
+
+- **ID:** `2602.02087` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Andreas Kontogiannis, Vasilis Pollatos, Panayotis Mertikopoulos, et al.
+- **abs:** https://arxiv.org/abs/2602.02087 · **PDF:** https://arxiv.org/pdf/2602.02087v1
+
+## 16554. Closing the Loop: Universal Repository Representation with RPG-Encoder
+
+- **ID:** `2602.02084` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Jane Luo, Chengyu Yin, Xin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.02084 · **PDF:** https://arxiv.org/pdf/2602.02084v2
+
+## 16555. Handling Covariate Mismatch in Federated Linear Prediction
+
+- **ID:** `2602.02083` · **primary:** `math.ST` · **Submitted:** 2026-02-02
+- **Authors:** Alexis Ayme, Rémi Khellaf
+- **abs:** https://arxiv.org/abs/2602.02083 · **PDF:** https://arxiv.org/pdf/2602.02083v1
+
+## 16556. Active learning from positive and unlabeled examples
+
+- **ID:** `2602.02081` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Farnam Mansouri, Sandra Zilles, Shai Ben-David
+- **abs:** https://arxiv.org/abs/2602.02081 · **PDF:** https://arxiv.org/pdf/2602.02081v1
+
+## 16557. Learning Half-Spaces from Perturbed Contrastive Examples
+
+- **ID:** `2602.02080` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Aryan Alavi Razavi Ravari, Farnam Mansouri, Yuxin Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.02080 · **PDF:** https://arxiv.org/pdf/2602.02080v1
+
+## 16558. AICD Bench: A Challenging Benchmark for AI-Generated Code Detection
+
+- **ID:** `2602.02079` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Daniil Orel, Dilshod Azizov, Indraneil Paul, et al.
+- **abs:** https://arxiv.org/abs/2602.02079 · **PDF:** https://arxiv.org/pdf/2602.02079v1
+
+## 16559. daVinci-Agency: Unlocking Long-Horizon Agency Data-Efficiently
+
+- **ID:** `2602.02619` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Mohan Jiang, Dayuan Fu, Junhao Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.02619 · **PDF:** https://arxiv.org/pdf/2602.02619v2
+
+## 16560. Calibrating Adaptive Smoothing Methods for Freeway Traffic Reconstruction
+
+- **ID:** `2602.02072` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Junyi Ji, Derek Gloudemans, Gergely Zachár, et al.
+- **abs:** https://arxiv.org/abs/2602.02072 · **PDF:** https://arxiv.org/pdf/2602.02072v1
+
+## 16561. BAPS: A Fine-Grained Low-Precision Scheme for Softmax in Attention via Block-Aware Precision reScaling
+
+- **ID:** `2602.02071` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zisheng Ye, Xiaoyu He, Maoyuan Song, et al.
+- **abs:** https://arxiv.org/abs/2602.02071 · **PDF:** https://arxiv.org/pdf/2602.02071v1
+
+## 16562. Multi-View Stenosis Classification Leveraging Transformer-Based Multiple-Instance Learning Using Real-World Clinical Data
+
+- **ID:** `2602.02067` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Nikola Cenikj, Özgün Turgut, Alexander Müller, et al.
+- **abs:** https://arxiv.org/abs/2602.02067 · **PDF:** https://arxiv.org/pdf/2602.02067v1
+
+## 16563. See2Refine: Vision-Language Feedback Improves LLM-Based eHMI Action Designers
+
+- **ID:** `2602.02063` · **primary:** `cs.HC` · **Submitted:** 2026-02-02
+- **Authors:** Ding Xia, Xinyue Gui, Mark Colley, et al.
+- **abs:** https://arxiv.org/abs/2602.02063 · **PDF:** https://arxiv.org/pdf/2602.02063v1
+
+## 16564. Learning to Route and Schedule LLMs from User Retrials via Contextual Queueing Bandits
+
+- **ID:** `2602.02061` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Seoungbin Bae, Junyoung Son, Dabeen Lee
+- **abs:** https://arxiv.org/abs/2602.02061 · **PDF:** https://arxiv.org/pdf/2602.02061v1
+
+## 16565. FiLoRA: Focus-and-Ignore LoRA for Controllable Feature Reliance
+
+- **ID:** `2602.02060` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Hyunsuk Chung, Caren Han, Yerin Choi, et al.
+- **abs:** https://arxiv.org/abs/2602.02060 · **PDF:** https://arxiv.org/pdf/2602.02060v1
+
+## 16566. A Semi-Supervised Pipeline for Generalized Behavior Discovery from Animal-Borne Motion Time Series
+
+- **ID:** `2602.02618` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Fatemeh Karimi Nejadasl, Judy Shamoun-Baranes, Eldar Rakhimberdiev
+- **abs:** https://arxiv.org/abs/2602.02618 · **PDF:** https://arxiv.org/pdf/2602.02618v1
+
+## 16567. Ultrafast On-chip Online Learning via Spline Locality in Kolmogorov-Arnold Networks
+
+- **ID:** `2602.02056` · **primary:** `cs.AR` · **Submitted:** 2026-02-02
+- **Authors:** Duc Hoang, Aarush Gupta, Philip Harris
+- **abs:** https://arxiv.org/abs/2602.02056 · **PDF:** https://arxiv.org/pdf/2602.02056v1
+
+## 16568. FORLER: Federated Offline Reinforcement Learning with Q-Ensemble and Actor Rectification
+
+- **ID:** `2602.02055` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Nan Qiao, Sheng Yue
+- **abs:** https://arxiv.org/abs/2602.02055 · **PDF:** https://arxiv.org/pdf/2602.02055v1
+
+## 16569. WildGraphBench: Benchmarking GraphRAG with Wild-Source Corpora
+
+- **ID:** `2602.02053` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Pengyu Wang, Benfeng Xu, Licheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.02053 · **PDF:** https://arxiv.org/pdf/2602.02053v2
+
+## 16570. SIDiffAgent: Self-Improving Diffusion Agent
+
+- **ID:** `2602.02051` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Shivank Garg, Ayush Singh, Gaurav Kumar Nayak
+- **abs:** https://arxiv.org/abs/2602.02051 · **PDF:** https://arxiv.org/pdf/2602.02051v1
+
+## 16571. Rethinking the Role of Entropy in Optimizing Tool-Use Behaviors for Large Language Model Agents
+
+- **ID:** `2602.02050` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Zeping Li, Hongru Wang, Yiwen Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.02050 · **PDF:** https://arxiv.org/pdf/2602.02050v3
+
+## 16572. Dissecting Outlier Dynamics in LLM NVFP4 Pretraining
+
+- **ID:** `2602.02047` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Peijie Dong, Ruibo Fan, Yuechen Tao, et al.
+- **abs:** https://arxiv.org/abs/2602.02047 · **PDF:** https://arxiv.org/pdf/2602.02047v1
+
+## 16573. On Stability and Robustness of Diffusion Posterior Sampling for Bayesian Inverse Problems
+
+- **ID:** `2602.02045` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yiming Yang, Xiaoyuan Cheng, Yi He, et al.
+- **abs:** https://arxiv.org/abs/2602.02045 · **PDF:** https://arxiv.org/pdf/2602.02045v1
+
+## 16574. Twinning Complex Networked Systems: Data-Driven Calibration of the mABCD Synthetic Graph Generator
+
+- **ID:** `2602.02044` · **primary:** `cs.SI` · **Submitted:** 2026-02-02
+- **Authors:** Piotr Bródka, Michał Czuba, Bogumił Kamiński, et al.
+- **abs:** https://arxiv.org/abs/2602.02044 · **PDF:** https://arxiv.org/pdf/2602.02044v2
+
+## 16575. Auto-Comp: An Automated Pipeline for Scalable Compositional Probing of Contrastive Vision-Language Models
+
+- **ID:** `2602.02043` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Cristian Sbrolli, Matteo Matteucci, Toshihiko Yamasaki
+- **abs:** https://arxiv.org/abs/2602.02043 · **PDF:** https://arxiv.org/pdf/2602.02043v1
+
+## 16576. Hunt Instead of Wait: Evaluating Deep Data Research on Large Language Models
+
+- **ID:** `2602.02039` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Wei Liu, Peijie Yu, Michele Orini, et al.
+- **abs:** https://arxiv.org/abs/2602.02039 · **PDF:** https://arxiv.org/pdf/2602.02039v1
+
+## 16577. Bandwidth-Efficient Multi-Agent Communication through Information Bottleneck and Vector Quantization
+
+- **ID:** `2602.02035` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Ahmad Farooq, Kamran Iqbal
+- **abs:** https://arxiv.org/abs/2602.02035 · **PDF:** https://arxiv.org/pdf/2602.02035v1
+
+## 16578. Constrained Process Maps for Multi-Agent Generative AI Workflows
+
+- **ID:** `2602.02034` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Ananya Joshi, Michael Rudow
+- **abs:** https://arxiv.org/abs/2602.02034 · **PDF:** https://arxiv.org/pdf/2602.02034v1
+
+## 16579. One Size, Many Fits: Aligning Diverse Group-Wise Click Preferences in Large-Scale Advertising Image Generation
+
+- **ID:** `2602.02033` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Shuo Lu, Haohan Wang, Wei Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.02033 · **PDF:** https://arxiv.org/pdf/2602.02033v1
+
+## 16580. Canonical Intermediate Representation for LLM-based optimization problem formulation and code generation
+
+- **ID:** `2602.02029` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Zhongyuan Lyu, Shuoyu Hu, Lujie Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.02029 · **PDF:** https://arxiv.org/pdf/2602.02029v1
+
+## 16581. Edit Knowledge, Not Just Facts via Multi-Step Reasoning over Background Stories
+
+- **ID:** `2602.02028` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Ya Gao, Kalle Kujanpää, Pekka Marttinen, et al.
+- **abs:** https://arxiv.org/abs/2602.02028 · **PDF:** https://arxiv.org/pdf/2602.02028v1
+
+## 16582. CTRL-RAG: Contrastive Likelihood Reward Based Reinforcement Learning for Context-Faithful RAG Models
+
+- **ID:** `2603.04406` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Zhehao Tan, Yihan Jiao, Dan Yang, et al.
+- **abs:** https://arxiv.org/abs/2603.04406 · **PDF:** https://arxiv.org/pdf/2603.04406v1
+
+## 16583. Light Alignment Improves LLM Safety via Model Self-Reflection with a Single Neuron
+
+- **ID:** `2602.02027` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Sicheng Shen, Mingyang Lv, Han Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.02027 · **PDF:** https://arxiv.org/pdf/2602.02027v1
+
+## 16584. Hippasus: Effective and Efficient Automatic Feature Augmentation for Machine Learning Tasks on Relational Data
+
+- **ID:** `2602.02025` · **primary:** `cs.DB` · **Submitted:** 2026-02-02
+- **Authors:** Serafeim Papadias, Kostas Patroumpas, Dimitrios Skoutas
+- **abs:** https://arxiv.org/abs/2602.02025 · **PDF:** https://arxiv.org/pdf/2602.02025v1
+
+## 16585. Adaptive Quality-Diversity Trade-offs for Large-Scale Batch Recommendation
+
+- **ID:** `2602.02024` · **primary:** `cs.IR` · **Submitted:** 2026-02-02
+- **Authors:** Clémence Réda, Tomas Rigaux, Hiba Bederina, et al.
+- **abs:** https://arxiv.org/abs/2602.02024 · **PDF:** https://arxiv.org/pdf/2602.02024v1
+
+## 16586. The Geometry of Representational Failures in Vision Language Models
+
+- **ID:** `2602.07025` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Daniele Savietto, Declan Campbell, André Panisson, et al.
+- **abs:** https://arxiv.org/abs/2602.07025 · **PDF:** https://arxiv.org/pdf/2602.07025v1
+
+## 16587. Scale-covariant spiking wavelets
+
+- **ID:** `2602.02020` · **primary:** `cs.NE` · **Submitted:** 2026-02-02
+- **Authors:** Jens Egholm Pedersen, Tony Lindeberg, Peter Gerstoft
+- **abs:** https://arxiv.org/abs/2602.02020 · **PDF:** https://arxiv.org/pdf/2602.02020v2
+
+## 16588. Do I Really Know? Learning Factual Self-Verification for Hallucination Reduction
+
+- **ID:** `2602.02018` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Enes Altinisik, Masoomali Fatehkia, Fatih Deniz, et al.
+- **abs:** https://arxiv.org/abs/2602.02018 · **PDF:** https://arxiv.org/pdf/2602.02018v1
+
+## 16589. DASH: Faster Shampoo via Batched Block Preconditioning and Efficient Inverse-Root Solvers
+
+- **ID:** `2602.02016` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Ionut-Vlad Modoranu, Philip Zmushko, Erik Schultheis, et al.
+- **abs:** https://arxiv.org/abs/2602.02016 · **PDF:** https://arxiv.org/pdf/2602.02016v1
+
+## 16590. Robust Domain Generalization under Divergent Marginal and Conditional Distributions
+
+- **ID:** `2602.02015` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Jewon Yeom, Kyubyung Chae, Hyunggyu Lim, et al.
+- **abs:** https://arxiv.org/abs/2602.02015 · **PDF:** https://arxiv.org/pdf/2602.02015v1
+
+## 16591. Rethinking Genomic Modeling Through Optical Character Recognition
+
+- **ID:** `2602.02014` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Hongxin Xiang, Pengsen Ma, Yunkang Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.02014 · **PDF:** https://arxiv.org/pdf/2602.02014v1
+
+## 16592. SNAP: A Self-Consistent Agreement Principle with Application to Robust Computation
+
+- **ID:** `2602.02013` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Xiaoyi Jiang, Andreas Nienkötter
+- **abs:** https://arxiv.org/abs/2602.02013 · **PDF:** https://arxiv.org/pdf/2602.02013v1
+
+## 16593. NEAT: Neuron-Based Early Exit for Large Reasoning Models
+
+- **ID:** `2602.02010` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Kang Liu, Yongkang Liu, Xiaocui Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.02010 · **PDF:** https://arxiv.org/pdf/2602.02010v1
+
+## 16594. Logic-Guided Vector Fields for Constrained Generative Modeling
+
+- **ID:** `2602.02009` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Ali Baheri
+- **abs:** https://arxiv.org/abs/2602.02009 · **PDF:** https://arxiv.org/pdf/2602.02009v1
+
+## 16595. Beyond RAG for Agent Memory: Retrieval by Decoupling and Aggregation
+
+- **ID:** `2602.02007` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Zhanghao Hu, Qinglin Zhu, Hanqi Yan, et al.
+- **abs:** https://arxiv.org/abs/2602.02007 · **PDF:** https://arxiv.org/pdf/2602.02007v2
+
+## 16596. Position: The Need for Ultrafast Training
+
+- **ID:** `2602.02005` · **primary:** `cs.AR` · **Submitted:** 2026-02-02
+- **Authors:** Duc Hoang
+- **abs:** https://arxiv.org/abs/2602.02005 · **PDF:** https://arxiv.org/pdf/2602.02005v1
+
+## 16597. ClueTracer: Question-to-Vision Clue Tracing for Training-Free Hallucination Suppression in Multimodal Reasoning
+
+- **ID:** `2602.02004` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Gongli Xi, Kun Wang, Zeming Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.02004 · **PDF:** https://arxiv.org/pdf/2602.02004v1
+
+## 16598. Preserve-Then-Quantize: Balancing Rank Budgets for Quantization Error Reconstruction in LLMs
+
+- **ID:** `2602.02001` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yoonjun Cho, Dongjae Jeon, Soeun Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.02001 · **PDF:** https://arxiv.org/pdf/2602.02001v1
+
+## 16599. SurfSplat: Conquering Feedforward 2D Gaussian Splatting with Surface Continuity Priors
+
+- **ID:** `2602.02000` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Bing He, Jingnan Gao, Yunuo Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.02000 · **PDF:** https://arxiv.org/pdf/2602.02000v2
+
+## 16600. From Latent Signals to Reflection Behavior: Tracing Meta-Cognitive Activation Trajectory in R1-Style LLMs
+
+- **ID:** `2602.01999` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Yanrui Du, Yibo Gao, Sendong Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.01999 · **PDF:** https://arxiv.org/pdf/2602.01999v2
+
+## 16601. On the Limits of Layer Pruning for Generative Reasoning in LLMs
+
+- **ID:** `2602.01997` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Safal Shrestha, Anubhav Shrestha, Aadim Nepal, et al.
+- **abs:** https://arxiv.org/abs/2602.01997 · **PDF:** https://arxiv.org/pdf/2602.01997v1
+
+## 16602. Thinking Like a Doctor: Conversational Diagnosis through the Exploration of Diagnostic Knowledge Graphs
+
+- **ID:** `2602.01995` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Jeongmoon Won, Seungwon Kook, Yohan Jo
+- **abs:** https://arxiv.org/abs/2602.01995 · **PDF:** https://arxiv.org/pdf/2602.01995v1
+
+## 16603. Optimizing Tensor Train Decomposition in DNNs for RISC-V Architectures Using Design Space Exploration and Compiler Optimizations
+
+- **ID:** `2602.01996` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Theologos Anthimopoulos, Milad Kokhazadeh, Vasilios Kelefouras, et al.
+- **abs:** https://arxiv.org/abs/2602.01996 · **PDF:** https://arxiv.org/pdf/2602.01996v1
+
+## 16604. Emergent Analogical Reasoning in Transformers
+
+- **ID:** `2602.01992` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Gouki Minegishi, Jingyuan Feng, Hiroki Furuta, et al.
+- **abs:** https://arxiv.org/abs/2602.01992 · **PDF:** https://arxiv.org/pdf/2602.01992v3
+
+## 16605. SAME: Stabilized Mixture-of-Experts for Multimodal Continual Instruction Tuning
+
+- **ID:** `2602.01990` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zhen-Hao Xie, Jun-Tao Tang, Yu-Cheng Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.01990 · **PDF:** https://arxiv.org/pdf/2602.01990v1
+
+## 16606. Stochastic Interpolants in Hilbert Spaces
+
+- **ID:** `2602.01988` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** James Boran Yu, RuiKang OuYang, Julien Horwood, et al.
+- **abs:** https://arxiv.org/abs/2602.01988 · **PDF:** https://arxiv.org/pdf/2602.01988v1
+
+## 16607. TinyGuard:A lightweight Byzantine Defense for Resource-Constrained Federated Learning via Statistical Update Fingerprints
+
+- **ID:** `2602.02615` · **primary:** `cs.CR` · **Submitted:** 2026-02-02
+- **Authors:** Ali Mahdavi, Santa Aghapour, Azadeh Zamanifar, et al.
+- **abs:** https://arxiv.org/abs/2602.02615 · **PDF:** https://arxiv.org/pdf/2602.02615v1
+
+## 16608. Evolving from Tool User to Creator via Training-Free Experience Reuse in Multimodal Reasoning
+
+- **ID:** `2602.01983` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Xintian Shen, Jiawei Chen, Lihao Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.01983 · **PDF:** https://arxiv.org/pdf/2602.01983v1
+
+## 16609. S3-CoT: Self-Sampled Succinct Reasoning Enables Efficient Chain-of-Thought LLMs
+
+- **ID:** `2602.01982` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Yanrui Du, Sendong Zhao, Yibo Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.01982 · **PDF:** https://arxiv.org/pdf/2602.01982v1
+
+## 16610. SpikingGamma: Surrogate-Gradient Free and Temporally Precise Online Training of Spiking Neural Networks with Smoothed Delays
+
+- **ID:** `2602.01978` · **primary:** `cs.NE` · **Submitted:** 2026-02-02
+- **Authors:** Roel Koopman, Sebastian Otte, Sander Bohté
+- **abs:** https://arxiv.org/abs/2602.01978 · **PDF:** https://arxiv.org/pdf/2602.01978v1
+
+## 16611. Testing Storage-System Correctness: Challenges, Fuzzing Limitations, and AI-Augmented Opportunities
+
+- **ID:** `2602.02614` · **primary:** `cs.SE` · **Submitted:** 2026-02-02
+- **Authors:** Ying Wang, Jiahui Chen, Dejun Jiang
+- **abs:** https://arxiv.org/abs/2602.02614 · **PDF:** https://arxiv.org/pdf/2602.02614v2
+
+## 16612. Beyond Local Edits: Embedding-Virtualized Knowledge for Broader Evaluation and Preservation of Model Editing
+
+- **ID:** `2602.01977` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Shuainan Liu, Xuanang Chen, Ben He, et al.
+- **abs:** https://arxiv.org/abs/2602.01977 · **PDF:** https://arxiv.org/pdf/2602.01977v1
+
+## 16613. FlyPrompt: Brain-Inspired Random-Expanded Routing with Temporal-Ensemble Experts for General Continual Learning
+
+- **ID:** `2602.01976` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Hongwei Yan, Guanglong Sun, Kanglei Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.01976 · **PDF:** https://arxiv.org/pdf/2602.01976v3
+
+## 16614. IntraSlice: Towards High-Performance Structural Pruning with Block-Intra PCA for LLMs
+
+- **ID:** `2602.01975` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Meng Li, Peisong Wang, Yuantian Shao, et al.
+- **abs:** https://arxiv.org/abs/2602.01975 · **PDF:** https://arxiv.org/pdf/2602.01975v1
+
+## 16615. Your AI-Generated Image Detector Can Secretly Achieve SOTA Accuracy, If Calibrated
+
+- **ID:** `2602.01973` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Muli Yang, Gabriel James Goenawan, Henan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01973 · **PDF:** https://arxiv.org/pdf/2602.01973v1
+
+## 16616. Small Generalizable Prompt Predictive Models Can Steer Efficient RL Post-Training of Large Reasoning Models
+
+- **ID:** `2602.01970` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Yun Qu, Qi Wang, Yixiu Mao, et al.
+- **abs:** https://arxiv.org/abs/2602.01970 · **PDF:** https://arxiv.org/pdf/2602.01970v1
+
+## 16617. Learn2Fold: Structured Origami Generation with World Model Planning
+
+- **ID:** `2603.29585` · **primary:** `cs.GR` · **Submitted:** 2026-02-02
+- **Authors:** Yanjia Huang, Yunuo Chen, Ying Jiang, et al.
+- **abs:** https://arxiv.org/abs/2603.29585 · **PDF:** https://arxiv.org/pdf/2603.29585v1
+
+## 16618. Orthogonal Hierarchical Decomposition for Structure-Aware Table Understanding with Large Language Models
+
+- **ID:** `2602.01969` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Bin Cao, Huixian Lu, Chenwen Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.01969 · **PDF:** https://arxiv.org/pdf/2602.01969v1
+
+## 16619. Mixture-of-Experts with Intermediate CTC Supervision for Accented Speech Recognition
+
+- **ID:** `2602.01967` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Wonjun Lee, Hyounghun Kim, Gary Geunbae Lee
+- **abs:** https://arxiv.org/abs/2602.01967 · **PDF:** https://arxiv.org/pdf/2602.01967v1
+
+## 16620. Self-Consolidation for Self-Evolving Agents
+
+- **ID:** `2602.01966` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Hongzhuo Yu, Fei Zhu, Guo-Sen Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.01966 · **PDF:** https://arxiv.org/pdf/2602.01966v1
+
+## 16621. Breaking the Static Graph: Context-Aware Traversal for Robust Retrieval-Augmented Generation
+
+- **ID:** `2602.01965` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Kwun Hang Lau, Fangyuan Zhang, Boyu Ruan, et al.
+- **abs:** https://arxiv.org/abs/2602.01965 · **PDF:** https://arxiv.org/pdf/2602.01965v1
+
+## 16622. Exploring Silicon-Based Societies: An Early Study of the Moltbook Agent Community
+
+- **ID:** `2602.02613` · **primary:** `cs.MA` · **Submitted:** 2026-02-02
+- **Authors:** Yu-Zheng Lin, Bono Po-Jen Shih, Hsuan-Ying Alessandra Chien, et al.
+- **abs:** https://arxiv.org/abs/2602.02613 · **PDF:** https://arxiv.org/pdf/2602.02613v3
+
+## 16623. Zero-Shot Off-Policy Learning
+
+- **ID:** `2602.01962` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Arip Asadulaev, Maksim Bobrin, Salem Lahlou, et al.
+- **abs:** https://arxiv.org/abs/2602.01962 · **PDF:** https://arxiv.org/pdf/2602.01962v1
+
+## 16624. Grounding Generated Videos in Feasible Plans via World Models
+
+- **ID:** `2602.01960` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Christos Ziakas, Amir Bar, Alessandra Russo
+- **abs:** https://arxiv.org/abs/2602.01960 · **PDF:** https://arxiv.org/pdf/2602.01960v3
+
+## 16625. Efficient Epistemic Uncertainty Estimation for Large Language Models via Knowledge Distillation
+
+- **ID:** `2602.01956` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Seonghyeon Park, Jewon Yeom, Jaewon Sok, et al.
+- **abs:** https://arxiv.org/abs/2602.01956 · **PDF:** https://arxiv.org/pdf/2602.01956v1
+
+## 16626. Deep Multivariate Models with Parametric Conditionals
+
+- **ID:** `2602.01953` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Dmitrij Schlesinger, Boris Flach, Alexander Shekhovtsov
+- **abs:** https://arxiv.org/abs/2602.01953 · **PDF:** https://arxiv.org/pdf/2602.01953v1
+
+## 16627. Boundary-Constrained Diffusion Models for Floorplan Generation: Balancing Realism and Diversity
+
+- **ID:** `2602.01949` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Leonardo Stoppani, Davide Bacciu, Shahab Mokarizadeh
+- **abs:** https://arxiv.org/abs/2602.01949 · **PDF:** https://arxiv.org/pdf/2602.01949v1
+
+## 16628. Human Society-Inspired Approaches to Agentic AI Security: The 4C Framework
+
+- **ID:** `2602.01942` · **primary:** `cs.CR` · **Submitted:** 2026-02-02
+- **Authors:** Alsharif Abuadbba, Nazatul Sultan, Surya Nepal, et al.
+- **abs:** https://arxiv.org/abs/2602.01942 · **PDF:** https://arxiv.org/pdf/2602.01942v1
+
+## 16629. FluxNet: Learning Capacity-Constrained Local Transport Operators for Conservative and Bounded PDE Surrogates
+
+- **ID:** `2602.01941` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-02
+- **Authors:** Zishuo Lan, Junjie Li, Lei Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01941 · **PDF:** https://arxiv.org/pdf/2602.01941v1
+
+## 16630. Towards Exploratory and Focused Manipulation with Bimanual Active Perception: A New Problem, Benchmark and Strategy
+
+- **ID:** `2602.01939` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Yuxin He, Ruihao Zhang, Tianao Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.01939 · **PDF:** https://arxiv.org/pdf/2602.01939v3
+
+## 16631. T-LLM: Teaching Large Language Models to Forecast Time Series via Temporal Distillation
+
+- **ID:** `2602.01937` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Suhan Guo, Bingxu Wang, Shaodan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01937 · **PDF:** https://arxiv.org/pdf/2602.01937v1
+
+## 16632. PIMCST: Physics-Informed Multi-Phase Consensus and Spatio-Temporal Few-Shot Learning for Traffic Flow Forecasting
+
+- **ID:** `2602.01936` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Abdul Joseph Fofanah, Lian Wen, David Chen
+- **abs:** https://arxiv.org/abs/2602.01936 · **PDF:** https://arxiv.org/pdf/2602.01936v1
+
+## 16633. COLT: Lightweight Multi-LLM Collaboration through Shared MCTS Reasoning for Model Compilation
+
+- **ID:** `2602.01935` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Annabelle Sujun Tang, Christopher Priebe, Lianhui Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.01935 · **PDF:** https://arxiv.org/pdf/2602.01935v1
+
+## 16634. Large Language Model and Formal Concept Analysis: a comparative study for Topic Modeling
+
+- **ID:** `2602.01933` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Fabrice Boissier, Monica Sen, Irina Rychkova
+- **abs:** https://arxiv.org/abs/2602.01933 · **PDF:** https://arxiv.org/pdf/2602.01933v1
+
+## 16635. IslamicLegalBench: Evaluating LLMs Knowledge and Reasoning of Islamic Law Across 1,200 Years of Islamic Pluralist Legal Traditions
+
+- **ID:** `2602.21226` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Ezieddin Elmahjub, Junaid Qadir, Abdullah Mushtaq, et al.
+- **abs:** https://arxiv.org/abs/2602.21226 · **PDF:** https://arxiv.org/pdf/2602.21226v1
+
+## 16636. Probabilistic function-on-function nonlinear autoregressive model for emulation and reliability analysis of dynamical systems
+
+- **ID:** `2602.01929` · **primary:** `math.DS` · **Submitted:** 2026-02-02
+- **Authors:** Zhouzhou Song, Marcos A. Valdebenito, Styfen Schär, et al.
+- **abs:** https://arxiv.org/abs/2602.01929 · **PDF:** https://arxiv.org/pdf/2602.01929v1
+
+## 16637. Privacy Amplification by Missing Data
+
+- **ID:** `2602.01928` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** Simon Roburin, Rafaël Pinot, Erwan Scornet
+- **abs:** https://arxiv.org/abs/2602.01928 · **PDF:** https://arxiv.org/pdf/2602.01928v2
+
+## 16638. Bayesian Integration of Nonlinear Incomplete Clinical Data
+
+- **ID:** `2602.01924` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Lucía González-Zamorano, Nuria Balbás-Esteban, Vanessa Gómez-Verdejo, et al.
+- **abs:** https://arxiv.org/abs/2602.01924 · **PDF:** https://arxiv.org/pdf/2602.01924v1
+
+## 16639. Embedding Learning on Multiplex Networks for Link Prediction
+
+- **ID:** `2602.01922` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Orell Trautmann, Olaf Wolkenhauer, Clémence Réda
+- **abs:** https://arxiv.org/abs/2602.01922 · **PDF:** https://arxiv.org/pdf/2602.01922v1
+
+## 16640. PIMPC-GNN: Physics-Informed Multi-Phase Consensus Learning for Enhancing Imbalanced Node Classification in Graph Neural Networks
+
+- **ID:** `2602.01920` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Abdul Joseph Fofanah, Lian Wen, David Chen
+- **abs:** https://arxiv.org/abs/2602.01920 · **PDF:** https://arxiv.org/pdf/2602.01920v1
+
+## 16641. From Code-Centric to Concept-Centric: Teaching NLP with LLM-Assisted "Vibe Coding"
+
+- **ID:** `2602.01919` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Hend Al-Khalifa
+- **abs:** https://arxiv.org/abs/2602.01919 · **PDF:** https://arxiv.org/pdf/2602.01919v1
+
+## 16642. GuideWeb: A Benchmark for Automatic In-App Guide Generation on Real-World Web UIs
+
+- **ID:** `2602.01917` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Chengguang Gan, Yoshihiro Tsujii, Yunhao Liang, et al.
+- **abs:** https://arxiv.org/abs/2602.01917 · **PDF:** https://arxiv.org/pdf/2602.01917v1
+
+## 16643. VLM-Guided Experience Replay
+
+- **ID:** `2602.01915` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Elad Sharony, Tom Jurgenson, Orr Krupnik, et al.
+- **abs:** https://arxiv.org/abs/2602.01915 · **PDF:** https://arxiv.org/pdf/2602.01915v1
+
+## 16644. Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs
+
+- **ID:** `2602.01914` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Wenbo Pan, Zhichao Liu, Xianlong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01914 · **PDF:** https://arxiv.org/pdf/2602.01914v1
+
+## 16645. Reliable Real-Time Value at Risk Estimation via Quantile Regression Forest with Conformal Calibration
+
+- **ID:** `2602.01912` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** Du-Yi Wang, Guo Liang, Kun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01912 · **PDF:** https://arxiv.org/pdf/2602.01912v1
+
+## 16646. DomusFM: A Foundation Model for Smart-Home Sensor Data
+
+- **ID:** `2602.01910` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Michele Fiori, Gabriele Civitarese, Flora D. Salim, et al.
+- **abs:** https://arxiv.org/abs/2602.01910 · **PDF:** https://arxiv.org/pdf/2602.01910v1
+
+## 16647. Propagating the prior from far to near offset: A self-supervised diffusion framework for progressively recovering near-offsets of towed-streamer data
+
+- **ID:** `2602.01909` · **primary:** `physics.geo-ph` · **Submitted:** 2026-02-02
+- **Authors:** Shijun Cheng, Tariq Alkhalifah
+- **abs:** https://arxiv.org/abs/2602.01909 · **PDF:** https://arxiv.org/pdf/2602.01909v1
+
+## 16648. DSXFormer: Dual-Pooling Spectral Squeeze-Expansion and Dynamic Context Attention Transformer for Hyperspectral Image Classification
+
+- **ID:** `2602.01906` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Farhan Ullah, Irfan Ullah, Khalil Khan, et al.
+- **abs:** https://arxiv.org/abs/2602.01906 · **PDF:** https://arxiv.org/pdf/2602.01906v1
+
+## 16649. Learning Sparse Visual Representations via Spatial-Semantic Factorization
+
+- **ID:** `2602.01905` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Theodore Zhengde Zhao, Sid Kiblawi, Jianwei Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.01905 · **PDF:** https://arxiv.org/pdf/2602.01905v1
+
+## 16650. Data- and Variance-dependent Regret Bounds for Online Tabular MDPs
+
+- **ID:** `2602.01903` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Mingyi Li, Taira Tsuchiya, Kenji Yamanishi
+- **abs:** https://arxiv.org/abs/2602.01903 · **PDF:** https://arxiv.org/pdf/2602.01903v1
+
+## 16651. Architecture-Agnostic Curriculum Learning for Document Understanding: Empirical Evidence from Text-Only and Multimodal
+
+- **ID:** `2602.21225` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Mohammed Hamdan, Vincenzo Dentamaro, Giuseppe Pirlo, et al.
+- **abs:** https://arxiv.org/abs/2602.21225 · **PDF:** https://arxiv.org/pdf/2602.21225v1
+
+## 16652. Observation-dependent Bayesian active learning via input-warped Gaussian processes
+
+- **ID:** `2602.01898` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Sanna Jarl, Maria Bånkestad, Jonathan J. S. Scragg, et al.
+- **abs:** https://arxiv.org/abs/2602.01898 · **PDF:** https://arxiv.org/pdf/2602.01898v1
+
+## 16653. Internal Flow Signatures for Self-Checking and Refinement in LLMs
+
+- **ID:** `2602.01897` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Sungheon Jeong, Sanggeon Yun, Ryozo Masukawa, et al.
+- **abs:** https://arxiv.org/abs/2602.01897 · **PDF:** https://arxiv.org/pdf/2602.01897v1
+
+## 16654. Geometric Analysis of Token Selection in Multi-Head Attention
+
+- **ID:** `2602.01893` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Timur Mudarisov, Mikhal Burtsev, Tatiana Petrova, et al.
+- **abs:** https://arxiv.org/abs/2602.01893 · **PDF:** https://arxiv.org/pdf/2602.01893v1
+
+## 16655. ES-MemEval: Benchmarking Conversational Agents on Personalized Long-Term Emotional Support
+
+- **ID:** `2602.01885` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Tiantian Chen, Jiaqi Lu, Ying Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.01885 · **PDF:** https://arxiv.org/pdf/2602.01885v1
+
+## 16656. Entropy-Guided Data-Efficient Training for Multimodal Reasoning Reward Models
+
+- **ID:** `2602.01884` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Shidong Yang, Tongwen Huang, Hao Wen, et al.
+- **abs:** https://arxiv.org/abs/2602.01884 · **PDF:** https://arxiv.org/pdf/2602.01884v1
+
+## 16657. Discovering Data Manifold Geometry via Non-Contracting Flows
+
+- **ID:** `2602.02611` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** David Vigouroux, Lucas Drumetz, Ronan Fablet, et al.
+- **abs:** https://arxiv.org/abs/2602.02611 · **PDF:** https://arxiv.org/pdf/2602.02611v1
+
+## 16658. Autocorrelated Optimize-via-Estimate: Predict-then-Optimize versus Finite-sample Optimal
+
+- **ID:** `2602.01877` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zichun Wang, Gar Goei Loke, Ruiting Zuo
+- **abs:** https://arxiv.org/abs/2602.01877 · **PDF:** https://arxiv.org/pdf/2602.01877v1
+
+## 16659. PretrainRL: Alleviating Factuality Hallucination of Large Language Models at the Beginning
+
+- **ID:** `2602.01875` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Langming Liu, Kangtao Lv, Haibin Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.01875 · **PDF:** https://arxiv.org/pdf/2602.01875v1
+
+## 16660. Grappa: Gradient-Only Communication for Scalable Graph Neural Network Training
+
+- **ID:** `2602.01872` · **primary:** `cs.DC` · **Submitted:** 2026-02-02
+- **Authors:** Chongyang Xu, Christoph Siebenbrunner, Laurent Bindschaedler
+- **abs:** https://arxiv.org/abs/2602.01872 · **PDF:** https://arxiv.org/pdf/2602.01872v2
+
+## 16661. ProcMEM: Learning Reusable Procedural Memory from Experience via Non-Parametric PPO for LLM Agents
+
+- **ID:** `2602.01869` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Qirui Mi, Zhijian Ma, Mengyue Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.01869 · **PDF:** https://arxiv.org/pdf/2602.01869v1
+
+## 16662. GRAB: An LLM-Inspired Sequence-First Click-Through Rate Prediction Modeling Paradigm
+
+- **ID:** `2602.01865` · **primary:** `cs.IR` · **Submitted:** 2026-02-02
+- **Authors:** Shaopeng Chen, Chuyue Xie, Huimin Ren, et al.
+- **abs:** https://arxiv.org/abs/2602.01865 · **PDF:** https://arxiv.org/pdf/2602.01865v2
+
+## 16663. Transformers as Measure-Theoretic Associative Memory: A Statistical Perspective and Minimax Optimality
+
+- **ID:** `2602.01863` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** Ryotaro Kawata, Taiji Suzuki
+- **abs:** https://arxiv.org/abs/2602.01863 · **PDF:** https://arxiv.org/pdf/2602.01863v1
+
+## 16664. RIR-Former: Coordinate-Guided Transformer for Continuous Reconstruction of Room Impulse Responses
+
+- **ID:** `2602.01861` · **primary:** `eess.AS` · **Submitted:** 2026-02-02
+- **Authors:** Shaoheng Xu, Chunyi Sun, Jihui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01861 · **PDF:** https://arxiv.org/pdf/2602.01861v2
+
+## 16665. SOPRAG: Multi-view Graph Experts Retrieval for Industrial Standard Operating Procedures
+
+- **ID:** `2602.01858` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Liangtao Lin, Zhaomeng Zhu, Tianwei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01858 · **PDF:** https://arxiv.org/pdf/2602.01858v1
+
+## 16666. Time2Vec Transformer for Robust Gesture Recognition from Low-Density sEMG
+
+- **ID:** `2602.01855` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Blagoj Hristov, Hristijan Gjoreski, Vesna Ojleska Latkoska, et al.
+- **abs:** https://arxiv.org/abs/2602.01855 · **PDF:** https://arxiv.org/pdf/2602.01855v2
+
+## 16667. Designing Time Series Experiments in A/B Testing with Transformer Reinforcement Learning
+
+- **ID:** `2602.01853` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Xiangkun Wu, Qianglin Wen, Yingying Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01853 · **PDF:** https://arxiv.org/pdf/2602.01853v1
+
+## 16668. FUPareto: Bridging the Forgetting-Utility Gap in Federated Unlearning via Pareto Augmented Optimization
+
+- **ID:** `2602.01852` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zeyan Wang, Zhengmao Liu, Yongxin Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.01852 · **PDF:** https://arxiv.org/pdf/2602.01852v1
+
+## 16669. Behavioral Consistency Validation for LLM Agents: An Analysis of Trading-Style Switching through Stock-Market Simulation
+
+- **ID:** `2602.07023` · **primary:** `q-fin.TR` · **Submitted:** 2026-02-02
+- **Authors:** Zeping Li, Guancheng Wan, Keyang Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.07023 · **PDF:** https://arxiv.org/pdf/2602.07023v2
+
+## 16670. Self-Rewarding Sequential Monte Carlo for Masked Diffusion Language Models
+
+- **ID:** `2602.01849` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Ziwei Luo, Ziqi Jin, Lei Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01849 · **PDF:** https://arxiv.org/pdf/2602.01849v1
+
+## 16671. ROMA: Recursive Open Meta-Agent Framework for Long-Horizon Multi-Agent Systems
+
+- **ID:** `2602.01848` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Salaheddin Alzu'bi, Baran Nama, Arda Kaz, et al.
+- **abs:** https://arxiv.org/abs/2602.01848 · **PDF:** https://arxiv.org/pdf/2602.01848v2
+
+## 16672. No Generation without Representation: Efficient Causal Protein Language Models Enable Zero-Shot Fitness Estimation
+
+- **ID:** `2602.01845` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Furkan Eris
+- **abs:** https://arxiv.org/abs/2602.01845 · **PDF:** https://arxiv.org/pdf/2602.01845v1
+
+## 16673. CloDS: Visual-Only Unsupervised Cloth Dynamics Learning in Unknown Conditions
+
+- **ID:** `2602.01844` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Yuliang Zhan, Jian Li, Wenbing Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.01844 · **PDF:** https://arxiv.org/pdf/2602.01844v3
+
+## 16674. Prism: Efficient Test-Time Scaling via Hierarchical Search and Self-Verification for Discrete Diffusion Language Models
+
+- **ID:** `2602.01842` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Jinbin Bai, Yixuan Li, Yuchen Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.01842 · **PDF:** https://arxiv.org/pdf/2602.01842v2
+
+## 16675. Read As Human: Compressing Context via Parallelizable Close Reading and Skimming
+
+- **ID:** `2602.01840` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Jiwei Tang, Shilei Liu, Zhicheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01840 · **PDF:** https://arxiv.org/pdf/2602.01840v2
+
+## 16676. DOGMA: Weaving Structural Information into Data-centric Single-cell Transcriptomics Analysis
+
+- **ID:** `2602.01839` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Ru Zhang, Xunkai Li, Yaxin Deng, et al.
+- **abs:** https://arxiv.org/abs/2602.01839 · **PDF:** https://arxiv.org/pdf/2602.01839v1
+
+## 16677. AXE: Low-Cost Cross-Domain Web Structured Information Extraction
+
+- **ID:** `2602.01838` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Abdelrahman Mansour, Khaled W. Alshaer, Moataz Elsaban
+- **abs:** https://arxiv.org/abs/2602.01838 · **PDF:** https://arxiv.org/pdf/2602.01838v2
+
+## 16678. Fly0: Decoupling Semantic Grounding from Geometric Planning for Zero-Shot Aerial Navigation
+
+- **ID:** `2602.15875` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Zhenxing Xu, Brikit Lu, Weidong Bao, et al.
+- **abs:** https://arxiv.org/abs/2602.15875 · **PDF:** https://arxiv.org/pdf/2602.15875v1
+
+## 16679. Synesthesia of Vehicles: Tactile Data Synthesis from Visual Inputs
+
+- **ID:** `2602.01832` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Rui Wang, Yaoguang Cao, Yuyi Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.01832 · **PDF:** https://arxiv.org/pdf/2602.01832v1
+
+## 16680. Hyperbolic Graph Neural Networks Under the Microscope: The Role of Geometry-Task Alignment
+
+- **ID:** `2602.01828` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Dionisia Naddeo, Jonas Linkerhägner, Nicola Toschi, et al.
+- **abs:** https://arxiv.org/abs/2602.01828 · **PDF:** https://arxiv.org/pdf/2602.01828v1
+
+## 16681. Beyond Precision: Training-Inference Mismatch is an Optimization Problem and Simple LR Scheduling Fixes It
+
+- **ID:** `2602.01826` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yaxiang Zhang, Yingru Li, Jiacai Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.01826 · **PDF:** https://arxiv.org/pdf/2602.01826v1
+
+## 16682. Learning Sequential Decisions from Multiple Sources via Group-Robust Markov Decision Processes
+
+- **ID:** `2602.01825` · **primary:** `stat.ME` · **Submitted:** 2026-02-02
+- **Authors:** Mingyuan Xu, Zongqi Xia, Tianxi Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.01825 · **PDF:** https://arxiv.org/pdf/2602.01825v1
+
+## 16683. INDIBATOR: Diverse and Fact-Grounded Individuality for Multi-Agent Debate in Molecular Discovery
+
+- **ID:** `2602.01815` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Yunhui Jang, Seonghyun Park, Jaehyung Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.01815 · **PDF:** https://arxiv.org/pdf/2602.01815v1
+
+## 16684. Sentence Curve Language Models
+
+- **ID:** `2602.01807` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** DongNyeong Heo, Heeyoul Choi
+- **abs:** https://arxiv.org/abs/2602.01807 · **PDF:** https://arxiv.org/pdf/2602.01807v2
+
+## 16685. Fast Autoregressive Video Diffusion and World Models with Temporal Cache Compression and Sparse Attention
+
+- **ID:** `2602.01801` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Dvir Samuel, Issar Tzachor, Matan Levy, et al.
+- **abs:** https://arxiv.org/abs/2602.01801 · **PDF:** https://arxiv.org/pdf/2602.01801v1
+
+## 16686. Spatio-Temporal Transformers for Long-Term NDVI Forecasting
+
+- **ID:** `2602.01799` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Ido Faran, Nathan S. Netanyahu, Maxim Shoshany
+- **abs:** https://arxiv.org/abs/2602.01799 · **PDF:** https://arxiv.org/pdf/2602.01799v1
+
+## 16687. ORCH: many analyses, one merge-a deterministic multi-agent orchestrator for discrete-choice reasoning with EMA-guided routing
+
+- **ID:** `2602.01797` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Hanlin Zhou, Huah Yong Chan
+- **abs:** https://arxiv.org/abs/2602.01797 · **PDF:** https://arxiv.org/pdf/2602.01797v1
+
+## 16688. RedVisor: Reasoning-Aware Prompt Injection Defense via Zero-Copy KV Cache Reuse
+
+- **ID:** `2602.01795` · **primary:** `cs.CR` · **Submitted:** 2026-02-02
+- **Authors:** Mingrui Liu, Sixiao Zhang, Cheng Long, et al.
+- **abs:** https://arxiv.org/abs/2602.01795 · **PDF:** https://arxiv.org/pdf/2602.01795v1
+
+## 16689. Make Every Draft Count: Hidden State based Speculative Decoding
+
+- **ID:** `2602.21224` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Yuetao Chen, Xuliang Wang, Xinzhou Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.21224 · **PDF:** https://arxiv.org/pdf/2602.21224v1
+
+## 16690. Grad2Reward: From Sparse Judgment to Dense Rewards for Improving Open-Ended LLM Reasoning
+
+- **ID:** `2602.01791` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zheng Zhang, Ao Lu, Yuanhao Zeng, et al.
+- **abs:** https://arxiv.org/abs/2602.01791 · **PDF:** https://arxiv.org/pdf/2602.01791v1
+
+## 16691. Boltz is a Strong Baseline for Atom-level Representation Learning
+
+- **ID:** `2602.13249` · **primary:** `q-bio.BM` · **Submitted:** 2026-02-02
+- **Authors:** Hyosoon Jang, Hyunjin Seo, Yunhui Jang, et al.
+- **abs:** https://arxiv.org/abs/2602.13249 · **PDF:** https://arxiv.org/pdf/2602.13249v1
+
+## 16692. CodeOCR: On the Effectiveness of Vision Language Models in Code Understanding
+
+- **ID:** `2602.01785` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Yuling Shi, Chaoxiang Xie, Zhensu Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.01785 · **PDF:** https://arxiv.org/pdf/2602.01785v1
+
+## 16693. LingLanMiDian: Systematic Evaluation of LLMs on TCM Knowledge and Clinical Reasoning
+
+- **ID:** `2602.01779` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Rui Hua, Yu Wei, Zixin Shu, et al.
+- **abs:** https://arxiv.org/abs/2602.01779 · **PDF:** https://arxiv.org/pdf/2602.01779v1
+
+## 16694. Data Distribution Matters: A Data-Centric Perspective on Context Compression for Large Language Model
+
+- **ID:** `2602.01778` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Kangtao Lv, Jiwei Tang, Langming Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.01778 · **PDF:** https://arxiv.org/pdf/2602.01778v1
+
+## 16695. Stein-Rule Shrinkage for Stochastic Gradient Estimation in High Dimensions
+
+- **ID:** `2602.01777` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** M. Arashi, M. Amintoosi
+- **abs:** https://arxiv.org/abs/2602.01777 · **PDF:** https://arxiv.org/pdf/2602.01777v2
+
+## 16696. Position: Beyond Model-Centric Prediction -- Agentic Time Series Forecasting
+
+- **ID:** `2602.01776` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Mingyue Cheng, Xiaoyu Tao, Qi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.01776 · **PDF:** https://arxiv.org/pdf/2602.01776v4
+
+## 16697. Efficient Cross-Architecture Knowledge Transfer for Large-Scale Online User Response Prediction
+
+- **ID:** `2602.01775` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Yucheng Wu, Yuekui Yang, Hongzheng Li, et al.
+- **abs:** https://arxiv.org/abs/2602.01775 · **PDF:** https://arxiv.org/pdf/2602.01775v1
+
+## 16698. Cost-Aware Bayesian Optimization for Prototyping Interactive Devices
+
+- **ID:** `2602.01774` · **primary:** `cs.HC` · **Submitted:** 2026-02-02
+- **Authors:** Thomas Langerak, Renate Zhang, Ziyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01774 · **PDF:** https://arxiv.org/pdf/2602.01774v1
+
+## 16699. DIA-CLIP: a universal representation learning framework for zero-shot DIA proteomics
+
+- **ID:** `2602.01772` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yucheng Liao, Han Wen, Weinan E, et al.
+- **abs:** https://arxiv.org/abs/2602.01772 · **PDF:** https://arxiv.org/pdf/2602.01772v1
+
+## 16700. : One LLM Token for Explicit Graph Structural Understanding
+
+- **ID:** `2602.01771` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Jingyao Wu, Bin Lu, Zijun Di, et al.
+- **abs:** https://arxiv.org/abs/2602.01771 · **PDF:** https://arxiv.org/pdf/2602.01771v1
+
+## 16701. IRIS: Implicit Reward-Guided Internal Sifting for Mitigating Multimodal Hallucination
+
+- **ID:** `2602.01769` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yuanshuai Li, Yuping Yan, Jirui Han, et al.
+- **abs:** https://arxiv.org/abs/2602.01769 · **PDF:** https://arxiv.org/pdf/2602.01769v2
+
+## 16702. CoMeT: Collaborative Memory Transformer for Efficient Long Context Modeling
+
+- **ID:** `2602.01766` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Runsong Zhao, Shilei Liu, Jiwei Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.01766 · **PDF:** https://arxiv.org/pdf/2602.01766v1
+
+## 16703. Backdoor Sentinel: Detecting and Detoxifying Backdoors in Diffusion Models via Temporal Noise Consistency
+
+- **ID:** `2602.01765` · **primary:** `cs.CR` · **Submitted:** 2026-02-02
+- **Authors:** Bingzheng Wang, Xiaoyan Gu, Hongbo Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.01765 · **PDF:** https://arxiv.org/pdf/2602.01765v1
+
+## 16704. Condition Errors Refinement in Autoregressive Image Generation with Diffusion Loss
+
+- **ID:** `2602.07022` · **primary:** `eess.IV` · **Submitted:** 2026-02-02
+- **Authors:** Yucheng Zhou, Hao Li, Jianbing Shen
+- **abs:** https://arxiv.org/abs/2602.07022 · **PDF:** https://arxiv.org/pdf/2602.07022v1
+
+## 16705. A Provable Expressiveness Hierarchy in Hybrid Linear-Full Attention
+
+- **ID:** `2602.01763` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Xiaowei Ye, Xiaoyu He, Chao Liao, et al.
+- **abs:** https://arxiv.org/abs/2602.01763 · **PDF:** https://arxiv.org/pdf/2602.01763v1
+
+## 16706. PRISM: Parametrically Refactoring Inference for Speculative Sampling Draft Models
+
+- **ID:** `2602.01762` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Xuliang Wang, Yuetao Chen, Maochan Zhen, et al.
+- **abs:** https://arxiv.org/abs/2602.01762 · **PDF:** https://arxiv.org/pdf/2602.01762v1
+
+## 16707. Zero2Text: Zero-Training Cross-Domain Inversion Attacks on Textual Embeddings
+
+- **ID:** `2602.01757` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Doohyun Kim, Donghwa Kang, Kyungjae Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.01757 · **PDF:** https://arxiv.org/pdf/2602.01757v2
+
+## 16708. WorldCup Sampling for Multi-bit LLM Watermarking
+
+- **ID:** `2602.01752` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Yidan Wang, Yubing Ren, Yanan Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.01752 · **PDF:** https://arxiv.org/pdf/2602.01752v1
+
+## 16709. MGKAN: Predicting Asymmetric Drug-Drug Interactions via a Multimodal Graph Kolmogorov-Arnold Network
+
+- **ID:** `2602.01751` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Kunyi Fan, Mengjie Chen, Longlong Li, et al.
+- **abs:** https://arxiv.org/abs/2602.01751 · **PDF:** https://arxiv.org/pdf/2602.01751v2
+
+## 16710. Adversarial Reward Auditing for Active Detection and Mitigation of Reward Hacking
+
+- **ID:** `2602.01750` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Mohammad Beigi, Ming Jin, Junshan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01750 · **PDF:** https://arxiv.org/pdf/2602.01750v1
+
+## 16711. Controlling Exploration-Exploitation in GFlowNets via Markov Chain Perspectives
+
+- **ID:** `2602.01749` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Lin Chen, Samuel Drapeau, Fanghao Shao, et al.
+- **abs:** https://arxiv.org/abs/2602.01749 · **PDF:** https://arxiv.org/pdf/2602.01749v3
+
+## 16712. Gender Dynamics and Homophily in a Social Network of LLM Agents
+
+- **ID:** `2602.02606` · **primary:** `cs.SI` · **Submitted:** 2026-02-02
+- **Authors:** Faezeh Fadaei, Jenny Carla Moran, Taha Yasseri
+- **abs:** https://arxiv.org/abs/2602.02606 · **PDF:** https://arxiv.org/pdf/2602.02606v2
+
+## 16713. Enhancing Automated Essay Scoring with Three Techniques: Two-Stage Fine-Tuning, Score Alignment, and Self-Training
+
+- **ID:** `2602.01747` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Hongseok Choi, Serynn Kim, Wencke Liermann, et al.
+- **abs:** https://arxiv.org/abs/2602.01747 · **PDF:** https://arxiv.org/pdf/2602.01747v1
+
+## 16714. Rethinking LoRA for Data Heterogeneous Federated Learning: Subspace and State Alignment
+
+- **ID:** `2602.01746` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Hongyi Peng, Han Yu, Xiaoxiao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.01746 · **PDF:** https://arxiv.org/pdf/2602.01746v1
+
+## 16715. Probability-Entropy Calibration: An Elastic Indicator for Adaptive Fine-tuning
+
+- **ID:** `2602.01745` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Wenhao Yu, Shaohang Wei, Jiahong Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.01745 · **PDF:** https://arxiv.org/pdf/2602.01745v1
+
+## 16716. Softmax Linear Attention: Reclaiming Global Competition
+
+- **ID:** `2602.01744` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Mingwei Xu, Xuan Lin, Xinnan Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.01744 · **PDF:** https://arxiv.org/pdf/2602.01744v1
+
+## 16717. MACD: Model-Aware Contrastive Decoding via Counterfactual Data
+
+- **ID:** `2602.01740` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Qixin Xiao
+- **abs:** https://arxiv.org/abs/2602.01740 · **PDF:** https://arxiv.org/pdf/2602.01740v2
+
+## 16718. Physics-Informed Chebyshev Polynomial Neural Operator for Parametric Partial Differential Equations
+
+- **ID:** `2602.01737` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-02-02
+- **Authors:** Biao Chen, Jing Wang, Hairun Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.01737 · **PDF:** https://arxiv.org/pdf/2602.01737v1
+
+## 16719. Position: The Inevitable End of One-Architecture-Fits-All-Domains in Time Series Forecasting
+
+- **ID:** `2602.01736` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Qinwei Ma, Jingzhe Shi, Jiahao Qiu, et al.
+- **abs:** https://arxiv.org/abs/2602.01736 · **PDF:** https://arxiv.org/pdf/2602.01736v1
+
+## 16720. MSign: An Optimizer Preventing Training Instability in Large Language Models via Stable Rank Restoration
+
+- **ID:** `2602.01734` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Lianhai Ren, Yucheng Ding, Xiao Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.01734 · **PDF:** https://arxiv.org/pdf/2602.01734v1
+
+## 16721. ST-BCP: Tightening Coverage Bound for Backward Conformal Prediction via Non-Conformity Score Transformation
+
+- **ID:** `2602.01733` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** Junxian Liu, Hao Zeng, Hongxin Wei
+- **abs:** https://arxiv.org/abs/2602.01733 · **PDF:** https://arxiv.org/pdf/2602.01733v1
+
+## 16722. X-Blocks: Linguistic Building Blocks of Natural Language Explanations for Automated Vehicles
+
+- **ID:** `2602.13248` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Ashkan Y. Zadeh, Xiaomeng Li, Andry Rakotonirainy, et al.
+- **abs:** https://arxiv.org/abs/2602.13248 · **PDF:** https://arxiv.org/pdf/2602.13248v1
+
+## 16723. Cross-Domain Fake News Detection on Unseen Domains via LLM-Based Domain-Aware User Modeling
+
+- **ID:** `2602.01726` · **primary:** `cs.SI` · **Submitted:** 2026-02-02
+- **Authors:** Xuankai Yang, Yan Wang, Jiajie Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.01726 · **PDF:** https://arxiv.org/pdf/2602.01726v1
+
+## 16724. SafePred: A Predictive Guardrail for Computer-Using Agents via World Models
+
+- **ID:** `2602.01725` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Yurun Chen, Zeyi Liao, Ping Yin, et al.
+- **abs:** https://arxiv.org/abs/2602.01725 · **PDF:** https://arxiv.org/pdf/2602.01725v1
+
+## 16725. COMI: Coarse-to-fine Context Compression via Marginal Information Gain
+
+- **ID:** `2602.01719` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Jiwei Tang, Shilei Liu, Zhicheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01719 · **PDF:** https://arxiv.org/pdf/2602.01719v3
+
+## 16726. Revisiting Generalization Measures Beyond IID: An Empirical Study under Distributional Shift
+
+- **ID:** `2602.01718` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Sora Nakai, Youssef Fadhloun, Kacem Mathlouthi, et al.
+- **abs:** https://arxiv.org/abs/2602.01718 · **PDF:** https://arxiv.org/pdf/2602.01718v1
+
+## 16727. BBPE16: UTF-16-based byte-level byte-pair encoding for improved multilingual speech recognition
+
+- **ID:** `2602.01717` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Hyunsik Kim, Haeri Kim, Munhak Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.01717 · **PDF:** https://arxiv.org/pdf/2602.01717v1
+
+## 16728. Mechanistic Indicators of Steering Effectiveness in Large Language Models
+
+- **ID:** `2602.01716` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Mehdi Jafari, Hao Xue, Flora Salim
+- **abs:** https://arxiv.org/abs/2602.01716 · **PDF:** https://arxiv.org/pdf/2602.01716v3
+
+## 16729. Measuring Pragmatic Influence in Large Language Model Instructions
+
+- **ID:** `2602.21223` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Yilin Geng, Omri Abend, Eduard Hovy, et al.
+- **abs:** https://arxiv.org/abs/2602.21223 · **PDF:** https://arxiv.org/pdf/2602.21223v1
+
+## 16730. MedAraBench: Large-Scale Arabic Medical Question Answering Dataset and Benchmark
+
+- **ID:** `2602.01714` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Mouath Abu-Daoud, Leen Kharouf, Omar El Hajj, et al.
+- **abs:** https://arxiv.org/abs/2602.01714 · **PDF:** https://arxiv.org/pdf/2602.01714v1
+
+## 16731. Optimizing Prompts for Large Language Models: A Causal Approach
+
+- **ID:** `2602.01711` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Wei Chen, Yanbin Fang, Shuran Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.01711 · **PDF:** https://arxiv.org/pdf/2602.01711v1
+
+## 16732. Physics Informed Generative AI Enabling Labour Free Segmentation For Microscopy Analysis
+
+- **ID:** `2602.01710` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Salma Zahran, Zhou Ao, Zhengyang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01710 · **PDF:** https://arxiv.org/pdf/2602.01710v1
+
+## 16733. ARTIS: Agentic Risk-Aware Test-Time Scaling via Iterative Simulation
+
+- **ID:** `2602.01709` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Xingshan Zeng, Lingzhi Wang, Weiwen Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.01709 · **PDF:** https://arxiv.org/pdf/2602.01709v2
+
+## 16734. Game of Thought: Robust Information Seeking with Large Language Models Using Game Theory
+
+- **ID:** `2602.01708` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Langyuan Cui, Chun Kai Ling, Hwee Tou Ng
+- **abs:** https://arxiv.org/abs/2602.01708 · **PDF:** https://arxiv.org/pdf/2602.01708v1
+
+## 16735. Beyond Mode Elicitation: Diversity-Preserving Reinforcement Learning via Latent Diffusion Reasoner
+
+- **ID:** `2602.01705` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Haoqiang Kang, Yizhe Zhang, Nikki Lijing Kuang, et al.
+- **abs:** https://arxiv.org/abs/2602.01705 · **PDF:** https://arxiv.org/pdf/2602.01705v2
+
+## 16736. $\textbf{AGT$^{AO}$}$: Robust and Stabilized LLM Unlearning via Adversarial Gating Training with Adaptive Orthogonality
+
+- **ID:** `2602.01703` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Pengyu Li, Lingling Zhang, Zhitao Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.01703 · **PDF:** https://arxiv.org/pdf/2602.01703v1
+
+## 16737. Beyond Single-Modal Analytics: A Framework for Integrating Heterogeneous LLM-Based Query Systems for Multi-Modal Data
+
+- **ID:** `2602.01701` · **primary:** `cs.DB` · **Submitted:** 2026-02-02
+- **Authors:** Ruyu Li, Tinghui Zhang, Haodi Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.01701 · **PDF:** https://arxiv.org/pdf/2602.01701v2
+
+## 16738. Mitigating loss of control in advanced AI systems through instrumental goal trajectories
+
+- **ID:** `2602.01699` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Willem Fourie
+- **abs:** https://arxiv.org/abs/2602.01699 · **PDF:** https://arxiv.org/pdf/2602.01699v1
+
+## 16739. Restoring Exploration after Post-Training: Latent Exploration Decoding for Large Reasoning Models
+
+- **ID:** `2602.01698` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Wenhui Tan, Fiorenzo Parascandolo, Enver Sangineto, et al.
+- **abs:** https://arxiv.org/abs/2602.01698 · **PDF:** https://arxiv.org/pdf/2602.01698v2
+
+## 16740. Cross-Modal Purification and Fusion for Small-Object RGB-D Transmission-Line Defect Detection
+
+- **ID:** `2602.01696` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Jiaming Cui, Wenqiang Li, Shuai Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.01696 · **PDF:** https://arxiv.org/pdf/2602.01696v5
+
+## 16741. Beyond Dense States: Elevating Sparse Transcoders to Active Operators for Latent Reasoning
+
+- **ID:** `2602.01695` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Yadong Wang, Haodong Chen, Yu Tian, et al.
+- **abs:** https://arxiv.org/abs/2602.01695 · **PDF:** https://arxiv.org/pdf/2602.01695v1
+
+## 16742. What LLMs Think When You Don't Tell Them What to Think About?
+
+- **ID:** `2602.01689` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Yongchan Kwon, James Zou
+- **abs:** https://arxiv.org/abs/2602.01689 · **PDF:** https://arxiv.org/pdf/2602.01689v1
+
+## 16743. Counting Hypothesis: Potential Mechanism of In-Context Learning
+
+- **ID:** `2602.01687` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Jung H. Lee, Sujith Vijayan
+- **abs:** https://arxiv.org/abs/2602.01687 · **PDF:** https://arxiv.org/pdf/2602.01687v1
+
+## 16744. Semantic-aware Wasserstein Policy Regularization for Large Language Model Alignment
+
+- **ID:** `2602.01685` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Byeonghu Na, Hyungho Na, Yeongmin Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.01685 · **PDF:** https://arxiv.org/pdf/2602.01685v1
+
+## 16745. The Strategic Foresight of LLMs: Evidence from a Fully Prospective Venture Tournament
+
+- **ID:** `2602.01684` · **primary:** `econ.GN` · **Submitted:** 2026-02-02
+- **Authors:** Felipe A. Csaszar, Aticus Peterson, Daniel Wilde
+- **abs:** https://arxiv.org/abs/2602.01684 · **PDF:** https://arxiv.org/pdf/2602.01684v1
+
+## 16746. FreshMem: Brain-Inspired Frequency-Space Hybrid Memory for Streaming Video Understanding
+
+- **ID:** `2602.01683` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Kangcong Li, Peng Ye, Lin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01683 · **PDF:** https://arxiv.org/pdf/2602.01683v1
+
+## 16747. Finite and Corruption-Robust Regret Bounds in Online Inverse Linear Optimization under M-Convex Action Sets
+
+- **ID:** `2602.01682` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Taihei Oki, Shinsaku Sakaue
+- **abs:** https://arxiv.org/abs/2602.01682 · **PDF:** https://arxiv.org/pdf/2602.01682v1
+
+## 16748. Towards Autonomous Instrument Tray Assembly for Sterile Processing Applications
+
+- **ID:** `2602.01679` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Raghavasimhan Sankaranarayanan, Paul Stuart, Nicholas Ahn, et al.
+- **abs:** https://arxiv.org/abs/2602.01679 · **PDF:** https://arxiv.org/pdf/2602.01679v1
+
+## 16749. AI for Sustainable Data Protection and Fair Algorithmic Management in Environmental Regulation
+
+- **ID:** `2602.07021` · **primary:** `cs.CY` · **Submitted:** 2026-02-02
+- **Authors:** Sahibpreet Singh, Saksham Sharma
+- **abs:** https://arxiv.org/abs/2602.07021 · **PDF:** https://arxiv.org/pdf/2602.07021v1
+
+## 16750. TRIP-Bench: A Benchmark for Long-Horizon Interactive Agents in Real-World Scenarios
+
+- **ID:** `2602.01675` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Yuanzhe Shen, Zisu Huang, Zhengyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01675 · **PDF:** https://arxiv.org/pdf/2602.01675v1
+
+## 16751. Real-Time Loop Closure Detection in Visual SLAM via NetVLAD and Faiss
+
+- **ID:** `2602.01673` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Enguang Fan
+- **abs:** https://arxiv.org/abs/2602.01673 · **PDF:** https://arxiv.org/pdf/2602.01673v1
+
+## 16752. Scaling Search-Augmented LLM Reasoning via Adaptive Information Control
+
+- **ID:** `2602.01672` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Siheng Xiong, Oguzhan Gungordu, Blair Johnson, et al.
+- **abs:** https://arxiv.org/abs/2602.01672 · **PDF:** https://arxiv.org/pdf/2602.01672v1
+
+## 16753. AI-Assisted Adaptive Rendering for High-Frequency Security Telemetry in Web Interfaces
+
+- **ID:** `2602.01671` · **primary:** `cs.HC` · **Submitted:** 2026-02-02
+- **Authors:** Mona Rajhans
+- **abs:** https://arxiv.org/abs/2602.01671 · **PDF:** https://arxiv.org/pdf/2602.01671v1
+
+## 16754. ASGMamba: Adaptive Spectral Gating Mamba for Multivariate Time Series Forecasting
+
+- **ID:** `2602.01668` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Qianyang Li, Xingjun Zhang, Shaoxun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01668 · **PDF:** https://arxiv.org/pdf/2602.01668v1
+
+## 16755. Quantifying Epistemic Predictive Uncertainty in Conformal Prediction
+
+- **ID:** `2602.01667` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Siu Lun Chau, Soroush H. Zargarbashi, Yusuf Sale, et al.
+- **abs:** https://arxiv.org/abs/2602.01667 · **PDF:** https://arxiv.org/pdf/2602.01667v1
+
+## 16756. TABX: A High-Throughput Sandbox Battle Simulator for Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.01665` · **primary:** `cs.MA` · **Submitted:** 2026-02-02
+- **Authors:** Hayeong Lee, JunHyeok Oh, Byung-Jun Lee
+- **abs:** https://arxiv.org/abs/2602.01665 · **PDF:** https://arxiv.org/pdf/2602.01665v1
+
+## 16757. FlowSteer: Interactive Agentic Workflow Orchestration via End-to-End Reinforcement Learning
+
+- **ID:** `2602.01664` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Mingda Zhang, Haoran Luo, Tiesunlong Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.01664 · **PDF:** https://arxiv.org/pdf/2602.01664v3
+
+## 16758. CoDiQ: Test-Time Scaling for Controllable Difficult Question Generation
+
+- **ID:** `2602.01660` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Zhongyuan Peng, Caijun Xu, Changyi Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.01660 · **PDF:** https://arxiv.org/pdf/2602.01660v1
+
+## 16759. Efficient Adversarial Attacks on High-dimensional Offline Bandits
+
+- **ID:** `2602.01658` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Seyed Mohammad Hadi Hosseini, Amir Najafi, Mahdieh Soleymani Baghshah
+- **abs:** https://arxiv.org/abs/2602.01658 · **PDF:** https://arxiv.org/pdf/2602.01658v1
+
+## 16760. ProjDevBench: Benchmarking AI Coding Agents on End-to-End Project Development
+
+- **ID:** `2602.01655` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Pengrui Lu, Shiqi Zhang, Yunzhong Hou, et al.
+- **abs:** https://arxiv.org/abs/2602.01655 · **PDF:** https://arxiv.org/pdf/2602.01655v2
+
+## 16761. Steering Vector Fields for Context-Aware Inference-Time Control in Large Language Models
+
+- **ID:** `2602.01654` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Jiaqian Li, Yanshu Li, Kuan-Hao Huang
+- **abs:** https://arxiv.org/abs/2602.01654 · **PDF:** https://arxiv.org/pdf/2602.01654v1
+
+## 16762. On the Spatiotemporal Dynamics of Generalization in Neural Networks
+
+- **ID:** `2602.01651` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zichao Wei
+- **abs:** https://arxiv.org/abs/2602.01651 · **PDF:** https://arxiv.org/pdf/2602.01651v1
+
+## 16763. Contribution-aware Token Compression for Efficient Video Understanding via Reinforcement Learning
+
+- **ID:** `2602.01649` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Yinchao Ma, Qiang Zhou, Zhibin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01649 · **PDF:** https://arxiv.org/pdf/2602.01649v2
+
+## 16764. From Perception to Action: Spatial AI Agents and World Models
+
+- **ID:** `2602.01644` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Gloria Felicia, Nolan Bryant, Handi Putra, et al.
+- **abs:** https://arxiv.org/abs/2602.01644 · **PDF:** https://arxiv.org/pdf/2602.01644v1
+
+## 16765. De Novo Molecular Generation from Mass Spectra via Many-Body Enhanced Diffusion
+
+- **ID:** `2602.01643` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Xichen Sun, Wentao Wei, Jiahua Rao, et al.
+- **abs:** https://arxiv.org/abs/2602.01643 · **PDF:** https://arxiv.org/pdf/2602.01643v1
+
+## 16766. The Effect of Mini-Batch Noise on the Implicit Bias of Adam
+
+- **ID:** `2602.01642` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Matias D. Cattaneo, Boris Shigida
+- **abs:** https://arxiv.org/abs/2602.01642 · **PDF:** https://arxiv.org/pdf/2602.01642v1
+
+## 16767. A2Eval: Agentic and Automated Evaluation for Embodied Brain
+
+- **ID:** `2602.01640` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Shuai Zhang, Jiayu Hu, Zijie Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.01640 · **PDF:** https://arxiv.org/pdf/2602.01640v1
+
+## 16768. Chance-Constrained Inference for Hallucination Risk Control in Large Language Models
+
+- **ID:** `2602.01637` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Sreenivasan Mohandas
+- **abs:** https://arxiv.org/abs/2602.01637 · **PDF:** https://arxiv.org/pdf/2602.01637v1
+
+## 16769. COMET: Codebook-based Online-adaptive Multi-scale Embedding for Time-series Anomaly Detection
+
+- **ID:** `2602.01635` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Jinwoo Park, Hyeongwon Kang, Seung Hun Han, et al.
+- **abs:** https://arxiv.org/abs/2602.01635 · **PDF:** https://arxiv.org/pdf/2602.01635v2
+
+## 16770. HuPER: A Human-Inspired Framework for Phonetic Perception
+
+- **ID:** `2602.01634` · **primary:** `eess.AS` · **Submitted:** 2026-02-02
+- **Authors:** Chenxu Guo, Jiachen Lian, Yisi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.01634 · **PDF:** https://arxiv.org/pdf/2602.01634v1
+
+## 16771. AdaptNC: Adaptive Nonconformity Scores for Uncertainty-Aware Autonomous Systems in Dynamic Environments
+
+- **ID:** `2602.01629` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Renukanandan Tumu, Aditya Singh, Rahul Mangharam
+- **abs:** https://arxiv.org/abs/2602.01629 · **PDF:** https://arxiv.org/pdf/2602.01629v1
+
+## 16772. Toward Enhancing Representation Learning in Federated Multi-Task Settings
+
+- **ID:** `2602.01626` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Mehdi Setayesh, Mahdi Beitollahi, Yasser H. Khalil, et al.
+- **abs:** https://arxiv.org/abs/2602.01626 · **PDF:** https://arxiv.org/pdf/2602.01626v1
+
+## 16773. Efficient Softmax Reformulation for Homomorphic Encryption via Moment Generating Function
+
+- **ID:** `2602.01621` · **primary:** `cs.CR` · **Submitted:** 2026-02-02
+- **Authors:** Hanjun Park, Byeong-Seo Min, Jiheon Woo, et al.
+- **abs:** https://arxiv.org/abs/2602.01621 · **PDF:** https://arxiv.org/pdf/2602.01621v2
+
+## 16774. SUSD: Structured Unsupervised Skill Discovery through State Factorization
+
+- **ID:** `2602.01619` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Seyed Mohammad Hadi Hosseini, Mahdieh Soleymani Baghshah
+- **abs:** https://arxiv.org/abs/2602.01619 · **PDF:** https://arxiv.org/pdf/2602.01619v1
+
+## 16775. SEA-Guard: Culturally Grounded Multilingual Safeguard for Southeast Asia
+
+- **ID:** `2602.01618` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Panuthep Tasawong, Jian Gang Ngui, Alham Fikri Aji, et al.
+- **abs:** https://arxiv.org/abs/2602.01618 · **PDF:** https://arxiv.org/pdf/2602.01618v1
+
+## 16776. Fine-Tuning Language Models to Know What They Know
+
+- **ID:** `2602.02605` · **primary:** `cs.NE` · **Submitted:** 2026-02-02
+- **Authors:** Sangjun Park, Elliot Meyerson, Xin Qiu, et al.
+- **abs:** https://arxiv.org/abs/2602.02605 · **PDF:** https://arxiv.org/pdf/2602.02605v1
+
+## 16777. AgroFlux: A Spatial-Temporal Benchmark for Carbon and Nitrogen Flux Prediction in Agricultural Ecosystems
+
+- **ID:** `2602.01614` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Qi Cheng, Licheng Liu, Yao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01614 · **PDF:** https://arxiv.org/pdf/2602.01614v1
+
+## 16778. A Practical Tensor-Network Compression Pipeline for Production-Scale Large Language Models
+
+- **ID:** `2602.01613` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Sergii Kozyrev, Davyd Maiboroda
+- **abs:** https://arxiv.org/abs/2602.01613 · **PDF:** https://arxiv.org/pdf/2602.01613v1
+
+## 16779. What Do Agents Learn from Trajectory-SFT: Semantics or Interfaces?
+
+- **ID:** `2602.01611` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Weizheng Gu, Chengze Li, Zhuohao Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.01611 · **PDF:** https://arxiv.org/pdf/2602.01611v1
+
+## 16780. ToPT: Task-Oriented Prompt Tuning for Urban Region Representation Learning
+
+- **ID:** `2602.01610` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Zitao Guo, Changyang Jiang, Tianhong Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.01610 · **PDF:** https://arxiv.org/pdf/2602.01610v1
+
+## 16781. Reasoning with Autoregressive-Diffusion Collaborative Thoughts
+
+- **ID:** `2602.01608` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Mu Yuan, Liekang Zeng, Guoliang Xing, et al.
+- **abs:** https://arxiv.org/abs/2602.01608 · **PDF:** https://arxiv.org/pdf/2602.01608v1
+
+## 16782. Boosting Maximum Entropy Reinforcement Learning via One-Step Flow Matching
+
+- **ID:** `2602.01606` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zeqiao Li, Yijing Wang, Haoyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01606 · **PDF:** https://arxiv.org/pdf/2602.01606v1
+
+## 16783. Minimax optimal differentially private synthetic data for smooth queries
+
+- **ID:** `2602.01607` · **primary:** `math.ST` · **Submitted:** 2026-02-02
+- **Authors:** Rundong Ding, Yiyun He, Yizhe Zhu
+- **abs:** https://arxiv.org/abs/2602.01607 · **PDF:** https://arxiv.org/pdf/2602.01607v2
+
+## 16784. Universal Redundancies in Time Series Foundation Models
+
+- **ID:** `2602.01605` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Anthony Bao, Venkata Hasith Vattikuti, Jeffrey Lai, et al.
+- **abs:** https://arxiv.org/abs/2602.01605 · **PDF:** https://arxiv.org/pdf/2602.01605v1
+
+## 16785. Inference-Aware Meta-Alignment of LLMs via Non-Linear GRPO
+
+- **ID:** `2602.01603` · **primary:** `stat.ML` · **Submitted:** 2026-02-02
+- **Authors:** Shokichi Takakura, Akifumi Wachi, Rei Higuchi, et al.
+- **abs:** https://arxiv.org/abs/2602.01603 · **PDF:** https://arxiv.org/pdf/2602.01603v1
+
+## 16786. Adaptive Rollout Allocation for Online Reinforcement Learning with Verifiable Rewards
+
+- **ID:** `2602.01601` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Hieu Trung Nguyen, Bao Nguyen, Wenao Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.01601 · **PDF:** https://arxiv.org/pdf/2602.01601v3
+
+## 16787. Expected Harm: Rethinking Safety Evaluation of (Mis)Aligned LLMs
+
+- **ID:** `2602.01600` · **primary:** `cs.CR` · **Submitted:** 2026-02-02
+- **Authors:** Yen-Shan Chen, Zhi Rui Tam, Cheng-Kuang Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.01600 · **PDF:** https://arxiv.org/pdf/2602.01600v1
+
+## 16788. The Multiple Ticket Hypothesis: Random Sparse Subnetworks Suffice for RLVR
+
+- **ID:** `2602.01599` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Israel Adewuyi, Solomon Okibe, Vladmir Ivanov
+- **abs:** https://arxiv.org/abs/2602.01599 · **PDF:** https://arxiv.org/pdf/2602.01599v1
+
+## 16789. P-RAG: Prompt-Enhanced Parametric RAG with LoRA and Selective CoT for Biomedical and Multi-Hop QA
+
+- **ID:** `2602.15874` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Xingda Lyu, Gongfu Lyu, Zitai Yan, et al.
+- **abs:** https://arxiv.org/abs/2602.15874 · **PDF:** https://arxiv.org/pdf/2602.15874v1
+
+## 16790. The Art of Socratic Inquiry: A Framework for Proactive Template-Guided Therapeutic Conversation Generation
+
+- **ID:** `2602.01598` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Mingwen Zhang, Minqiang Yang, Changsheng Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.01598 · **PDF:** https://arxiv.org/pdf/2602.01598v1
+
+## 16791. Wiki Live Challenge: Challenging Deep Research Agents with Expert-Level Wikipedia Articles
+
+- **ID:** `2602.01590` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Shaohan Wang, Benfeng Xu, Licheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01590 · **PDF:** https://arxiv.org/pdf/2602.01590v2
+
+## 16792. Two-chart Beltrami Optimization for Distortion-Controlled Spherical Bijection with Application to Brain Surface Registration
+
+- **ID:** `2602.01589` · **primary:** `cs.GR` · **Submitted:** 2026-02-02
+- **Authors:** Zhehao Xu, Lok Ming Lui
+- **abs:** https://arxiv.org/abs/2602.01589 · **PDF:** https://arxiv.org/pdf/2602.01589v2
+
+## 16793. Spectral Text Fusion: A Frequency-Aware Approach to Multimodal Time-Series Forecasting
+
+- **ID:** `2602.01588` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Huu Hiep Nguyen, Minh Hoang Nguyen, Dung Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.01588 · **PDF:** https://arxiv.org/pdf/2602.01588v2
+
+## 16794. Provable Defense Framework for LLM Jailbreaks via Noise-Augumented Alignment
+
+- **ID:** `2602.01587` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Zehua Cheng, Jianwei Yang, Wei Dai, et al.
+- **abs:** https://arxiv.org/abs/2602.01587 · **PDF:** https://arxiv.org/pdf/2602.01587v1
+
+## 16795. A Lightweight Sparse Interaction Network for Time Series Forecasting
+
+- **ID:** `2602.01585` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Xu Zhang, Qitong Wang, Peng Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01585 · **PDF:** https://arxiv.org/pdf/2602.01585v1
+
+## 16796. On the Fragility of AI-Based Channel Decoders under Small Channel Perturbations
+
+- **ID:** `2602.01582` · **primary:** `cs.IT` · **Submitted:** 2026-02-02
+- **Authors:** Haoyu Lei, Mohammad Jalali, Chin Wa Lau, et al.
+- **abs:** https://arxiv.org/abs/2602.01582 · **PDF:** https://arxiv.org/pdf/2602.01582v1
+
+## 16797. Nearly Optimal Active Preference Learning and Its Application to LLM Alignment
+
+- **ID:** `2602.01581` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yao Zhao, Kwang-Sung Jun
+- **abs:** https://arxiv.org/abs/2602.01581 · **PDF:** https://arxiv.org/pdf/2602.01581v1
+
+## 16798. DrawSim-PD: Simulating Student Science Drawings to Support NGSS-Aligned Teacher Diagnostic Reasoning
+
+- **ID:** `2602.01578` · **primary:** `cs.CY` · **Submitted:** 2026-02-02
+- **Authors:** Arijit Chakma, Peng He, Honglu Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.01578 · **PDF:** https://arxiv.org/pdf/2602.01578v1
+
+## 16799. Generative Visual Code Mobile World Models
+
+- **ID:** `2602.01576` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Woosung Koh, Sungjun Han, Segyu Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.01576 · **PDF:** https://arxiv.org/pdf/2602.01576v1
+
+## 16800. When Is Generalized Bayes Bayesian? A Decision-Theoretic Characterization of Loss-Based Updating
+
+- **ID:** `2602.01573` · **primary:** `stat.ME` · **Submitted:** 2026-02-02
+- **Authors:** Kenichiro McAlinn, Kōsaku Takanashi
+- **abs:** https://arxiv.org/abs/2602.01573 · **PDF:** https://arxiv.org/pdf/2602.01573v1
+
+## 16801. LLM-based Embeddings: Attention Values Encode Sentence Semantics Better Than Hidden States
+
+- **ID:** `2602.01572` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Yeqin Zhang, Yunfei Wang, Jiaxuan Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.01572 · **PDF:** https://arxiv.org/pdf/2602.01572v1
+
+## 16802. DREAMS: A Social Exchange Theory-Informed Modeling of Misinformation Engagement on Social Media
+
+- **ID:** `2602.01567` · **primary:** `cs.SI` · **Submitted:** 2026-02-02
+- **Authors:** Lin Tian, Marian-Andrei Rizoiu
+- **abs:** https://arxiv.org/abs/2602.01567 · **PDF:** https://arxiv.org/pdf/2602.01567v1
+
+## 16803. FS-Researcher: Test-Time Scaling for Long-Horizon Research Tasks with File-System-Based Agents
+
+- **ID:** `2602.01566` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Chiwei Zhu, Benfeng Xu, Mingxuan Du, et al.
+- **abs:** https://arxiv.org/abs/2602.01566 · **PDF:** https://arxiv.org/pdf/2602.01566v1
+
+## 16804. Local Exponential Stability of Mean-Field Langevin Descent-Ascent in Wasserstein Space
+
+- **ID:** `2602.01564` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Geuntaek Seo, Minseop Shin, Pierre Monmarché, et al.
+- **abs:** https://arxiv.org/abs/2602.01564 · **PDF:** https://arxiv.org/pdf/2602.01564v1
+
+## 16805. Multimodal UNcommonsense: From Odd to Ordinary and Ordinary to Odd
+
+- **ID:** `2602.01561` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Yejin Son, Saejin Kim, Dongjun Min, et al.
+- **abs:** https://arxiv.org/abs/2602.01561 · **PDF:** https://arxiv.org/pdf/2602.01561v1
+
+## 16806. Argument Rarity-based Originality Assessment for AI-Assisted Writing
+
+- **ID:** `2602.01560` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Keito Inoshita, Michiaki Omura, Tsukasa Yamanaka, et al.
+- **abs:** https://arxiv.org/abs/2602.01560 · **PDF:** https://arxiv.org/pdf/2602.01560v3
+
+## 16807. How Implicit Bias Accumulates and Propagates in LLM Long-term Memory
+
+- **ID:** `2602.01558` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Yiming Ma, Lixu Wang, Lionel Z. Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01558 · **PDF:** https://arxiv.org/pdf/2602.01558v1
+
+## 16808. Autonomous Question Formation for Large Language Model-Driven AI Systems
+
+- **ID:** `2602.01556` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Hong Su
+- **abs:** https://arxiv.org/abs/2602.01556 · **PDF:** https://arxiv.org/pdf/2602.01556v1
+
+## 16809. InfoTok: Regulating Information Flow for Capacity-Constrained Shared Visual Tokenization in Unified MLLMs
+
+- **ID:** `2602.01554` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Lv Tang, Tianyi Zheng, Bo Li, et al.
+- **abs:** https://arxiv.org/abs/2602.01554 · **PDF:** https://arxiv.org/pdf/2602.01554v1
+
+## 16810. Plain Transformers are Surprisingly Powerful Link Predictors
+
+- **ID:** `2602.01553` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Quang Truong, Yu Song, Donald Loveland, et al.
+- **abs:** https://arxiv.org/abs/2602.01553 · **PDF:** https://arxiv.org/pdf/2602.01553v1
+
+## 16811. S1-NexusAgent: a Self-Evolving Agent Framework for Multidisciplinary Scientific Research
+
+- **ID:** `2602.01550` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** S1-NexusAgent Team
+- **abs:** https://arxiv.org/abs/2602.01550 · **PDF:** https://arxiv.org/pdf/2602.01550v2
+
+## 16812. AI Assisted Economics Measurement From Survey: Evidence from Public Employee Pension Choice
+
+- **ID:** `2602.02604` · **primary:** `econ.EM` · **Submitted:** 2026-02-02
+- **Authors:** Tiancheng Wang, Krishna Sharma
+- **abs:** https://arxiv.org/abs/2602.02604 · **PDF:** https://arxiv.org/pdf/2602.02604v1
+
+## 16813. Toward Cognitive Supersensing in Multimodal Large Language Model
+
+- **ID:** `2602.01541` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Boyi Li, Yifan Shen, Yuanzhe Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.01541 · **PDF:** https://arxiv.org/pdf/2602.01541v1
+
+## 16814. MAGIC: A Co-Evolving Attacker-Defender Adversarial Game for Robust LLM Safety
+
+- **ID:** `2602.01539` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Xiaoyu Wen, Zhida He, Han Qi, et al.
+- **abs:** https://arxiv.org/abs/2602.01539 · **PDF:** https://arxiv.org/pdf/2602.01539v2
+
+## 16815. Making Avatars Interact: Towards Text-Driven Human-Object Interaction for Controllable Talking Avatars
+
+- **ID:** `2602.01538` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Youliang Zhang, Zhengguang Zhou, Zhentao Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.01538 · **PDF:** https://arxiv.org/pdf/2602.01538v1
+
+## 16816. Rotation-free Online Handwritten Character Recognition Using Linear Recurrent Units
+
+- **ID:** `2602.01533` · **primary:** `cs.CV` · **Submitted:** 2026-02-02
+- **Authors:** Zhe Ling, Sicheng Yu, Danyu Yang
+- **abs:** https://arxiv.org/abs/2602.01533 · **PDF:** https://arxiv.org/pdf/2602.01533v1
+
+## 16817. PRISM: Festina Lente Proactivity -- Risk-Sensitive, Uncertainty-Aware Deliberation for Proactive Agents
+
+- **ID:** `2602.01532` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Yuxuan Fu, Xiaoyu Tan, Teqi Hao, et al.
+- **abs:** https://arxiv.org/abs/2602.01532 · **PDF:** https://arxiv.org/pdf/2602.01532v1
+
+## 16818. Making Bias Non-Predictive: Training Robust LLM Judges via Reinforcement Learning
+
+- **ID:** `2602.01528` · **primary:** `cs.CY` · **Submitted:** 2026-02-02
+- **Authors:** Qian Wang, Xuandong Zhao, Zirui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01528 · **PDF:** https://arxiv.org/pdf/2602.01528v1
+
+## 16819. Toward a Machine Bertin: Why Visualization Needs Design Principles for Machine Cognition
+
+- **ID:** `2602.01527` · **primary:** `cs.HC` · **Submitted:** 2026-02-02
+- **Authors:** Brian Keith-Norambuena
+- **abs:** https://arxiv.org/abs/2602.01527 · **PDF:** https://arxiv.org/pdf/2602.01527v1
+
+## 16820. The Inlet Rank Collapse in Implicit Neural Representations: Diagnosis and Unified Remedy
+
+- **ID:** `2602.01526` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Jianqiao Zheng, Hemanth Saratchandran, Simon Lucey
+- **abs:** https://arxiv.org/abs/2602.01526 · **PDF:** https://arxiv.org/pdf/2602.01526v1
+
+## 16821. A Relative-Budget Theory for Reinforcement Learning with Verifiable Rewards in Large Language Model Reasoning
+
+- **ID:** `2602.01523` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Akifumi Wachi, Hirota Kinoshita, Shokichi Takakura, et al.
+- **abs:** https://arxiv.org/abs/2602.01523 · **PDF:** https://arxiv.org/pdf/2602.01523v1
+
+## 16822. When Is Rank-1 Enough? Geometry-Guided Initialization for Parameter-Efficient Fine-Tuning
+
+- **ID:** `2602.01522` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Haoran Zhao, Soyeon Caren Han, Eduard Hovy
+- **abs:** https://arxiv.org/abs/2602.01522 · **PDF:** https://arxiv.org/pdf/2602.01522v1
+
+## 16823. You Need an Encoder for Native Position-Independent Caching
+
+- **ID:** `2602.01519` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Shiju Zhao, Junhao Hu, Jiaqi Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.01519 · **PDF:** https://arxiv.org/pdf/2602.01519v1
+
+## 16824. Qrita: High-performance Top-k and Top-p Algorithm for GPUs using Pivot-based Truncation and Selection
+
+- **ID:** `2602.01518` · **primary:** `cs.AI` · **Submitted:** 2026-02-02
+- **Authors:** Jongseok Park, Sunga Kim, Alvin Cheung, et al.
+- **abs:** https://arxiv.org/abs/2602.01518 · **PDF:** https://arxiv.org/pdf/2602.01518v1
+
+## 16825. White-Box Neural Ensemble for Vehicular Plasticity: Quantifying the Efficiency Cost of Symbolic Auditability in Adaptive NMPC
+
+- **ID:** `2602.01516` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Enzo Nicolas Spotorno, Matheus Wagner, Antonio Augusto Medeiros Frohlich
+- **abs:** https://arxiv.org/abs/2602.01516 · **PDF:** https://arxiv.org/pdf/2602.01516v1
+
+## 16826. RAPT: Model-Predictive Out-of-Distribution Detection and Failure Diagnosis for Sim-to-Real Humanoid Robots
+
+- **ID:** `2602.01515` · **primary:** `cs.RO` · **Submitted:** 2026-02-02
+- **Authors:** Humphrey Munn, Brendan Tidd, Peter Bohm, et al.
+- **abs:** https://arxiv.org/abs/2602.01515 · **PDF:** https://arxiv.org/pdf/2602.01515v1
+
+## 16827. MarkCleaner: High-Fidelity Watermark Removal via Imperceptible Micro-Geometric Perturbation
+
+- **ID:** `2602.01513` · **primary:** `eess.IV` · **Submitted:** 2026-02-02
+- **Authors:** Xiaoxi Kong, Jieyu Yuan, Pengdi Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.01513 · **PDF:** https://arxiv.org/pdf/2602.01513v1
+
+## 16828. Alternating Reinforcement Learning for Rubric-Based Reward Modeling in Non-Verifiable LLM Post-Training
+
+- **ID:** `2602.01511` · **primary:** `cs.CL` · **Submitted:** 2026-02-02
+- **Authors:** Ran Xu, Tianci Liu, Zihan Dong, et al.
+- **abs:** https://arxiv.org/abs/2602.01511 · **PDF:** https://arxiv.org/pdf/2602.01511v2
+
+## 16829. Enhancing Generalization in Evolutionary Feature Construction for Symbolic Regression through Vicinal Jensen Gap Minimization
+
+- **ID:** `2602.01510` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Hengzhe Zhang, Qi Chen, Bing Xue, et al.
+- **abs:** https://arxiv.org/abs/2602.01510 · **PDF:** https://arxiv.org/pdf/2602.01510v1
+
+## 16830. Financial Bond Similarity Search Using Representation Learning
+
+- **ID:** `2602.07020` · **primary:** `q-fin.ST` · **Submitted:** 2026-02-02
+- **Authors:** Amin Haeri, Mahdi Ghelichi, Nishant Agrawal, et al.
+- **abs:** https://arxiv.org/abs/2602.07020 · **PDF:** https://arxiv.org/pdf/2602.07020v1
+
+## 16831. Harnessing Flexible Spatial and Temporal Data Center Workloads for Grid Regulation Services
+
+- **ID:** `2602.01508` · **primary:** `eess.SY` · **Submitted:** 2026-02-02
+- **Authors:** Yingrui Fan, Junbo Zhao
+- **abs:** https://arxiv.org/abs/2602.01508 · **PDF:** https://arxiv.org/pdf/2602.01508v1
+
+## 16832. Optimal Sample Complexity for Single Time-Scale Actor-Critic with Momentum
+
+- **ID:** `2602.01505` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Navdeep Kumar, Tehila Dahan, Lior Cohen, et al.
+- **abs:** https://arxiv.org/abs/2602.01505 · **PDF:** https://arxiv.org/pdf/2602.01505v1
+
+## 16833. Governance at the Edge of Architecture: Regulating NeuroAI and Neuromorphic Systems
+
+- **ID:** `2602.01503` · **primary:** `cs.ET` · **Submitted:** 2026-02-02
+- **Authors:** Afifah Kashif, Abdul Muhsin Hameed, Asim Iqbal
+- **abs:** https://arxiv.org/abs/2602.01503 · **PDF:** https://arxiv.org/pdf/2602.01503v2
+
+## 16834. Draw2Learn: A Human-AI Collaborative Tool for Drawing-Based Science Learning
+
+- **ID:** `2602.01494` · **primary:** `cs.HC` · **Submitted:** 2026-02-02
+- **Authors:** Yuqi Hang
+- **abs:** https://arxiv.org/abs/2602.01494 · **PDF:** https://arxiv.org/pdf/2602.01494v1
+
+## 16835. OpInf-LLM: Parametric PDE Solving with LLMs via Operator Inference
+
+- **ID:** `2602.01493` · **primary:** `cs.LG` · **Submitted:** 2026-02-02
+- **Authors:** Zhuoyuan Wang, Hanjiang Hu, Xiyu Deng, et al.
+- **abs:** https://arxiv.org/abs/2602.01493 · **PDF:** https://arxiv.org/pdf/2602.01493v1
+
+## 16836. Multi-Scale Wavelet Transformers for Operator Learning of Dynamical Systems
+
+- **ID:** `2602.01486` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Xuesong Wang, Michael Groom, Rafael Oliveira, et al.
+- **abs:** https://arxiv.org/abs/2602.01486 · **PDF:** https://arxiv.org/pdf/2602.01486v1
+
+## 16837. Predicting and improving test-time scaling laws via reward tail-guided search
+
+- **ID:** `2602.01485` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Muheng Li, Jian Qian, Wenlong Mou
+- **abs:** https://arxiv.org/abs/2602.01485 · **PDF:** https://arxiv.org/pdf/2602.01485v1
+
+## 16838. Causal Preference Elicitation
+
+- **ID:** `2602.01483` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Edwin V. Bonilla, He Zhao, Daniel M. Steinberg
+- **abs:** https://arxiv.org/abs/2602.01483 · **PDF:** https://arxiv.org/pdf/2602.01483v1
+
+## 16839. Community-Level Modeling of Gyral Folding Patterns for Robust and Anatomically Informed Individualized Brain Mapping
+
+- **ID:** `2602.01482` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-01
+- **Authors:** Minheng Chen, Tong Chen, Yan Zhuang, et al.
+- **abs:** https://arxiv.org/abs/2602.01482 · **PDF:** https://arxiv.org/pdf/2602.01482v1
+
+## 16840. Rod Flow: A Continuous-Time Model for Gradient Descent at the Edge of Stability
+
+- **ID:** `2602.01480` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Eric Regis, Sinho Chewi
+- **abs:** https://arxiv.org/abs/2602.01480 · **PDF:** https://arxiv.org/pdf/2602.01480v1
+
+## 16841. Ebisu: Benchmarking Large Language Models in Japanese Finance
+
+- **ID:** `2602.01479` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Xueqing Peng, Ruoyu Xiang, Fan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01479 · **PDF:** https://arxiv.org/pdf/2602.01479v1
+
+## 16842. Density-Informed Pseudo-Counts for Calibrated Evidential Deep Learning
+
+- **ID:** `2602.01477` · **primary:** `stat.ML` · **Submitted:** 2026-02-01
+- **Authors:** Pietro Carlotti, Nevena Gligić, Arya Farahi
+- **abs:** https://arxiv.org/abs/2602.01477 · **PDF:** https://arxiv.org/pdf/2602.01477v1
+
+## 16843. Learning to Guide Local Search for MPE Inference in Probabilistic Graphical Models
+
+- **ID:** `2602.01475` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Brij Malhotra, Shivvrat Arya, Tahrima Rahman, et al.
+- **abs:** https://arxiv.org/abs/2602.01475 · **PDF:** https://arxiv.org/pdf/2602.01475v1
+
+## 16844. Legal Infrastructure for Transformative AI Governance
+
+- **ID:** `2602.01474` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Gillian K. Hadfield
+- **abs:** https://arxiv.org/abs/2602.01474 · **PDF:** https://arxiv.org/pdf/2602.01474v1
+
+## 16845. ConPress: Learning Efficient Reasoning from Multi-Question Contextual Pressure
+
+- **ID:** `2602.01472` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Jie Deng, Shining Liang, Jun Li, et al.
+- **abs:** https://arxiv.org/abs/2602.01472 · **PDF:** https://arxiv.org/pdf/2602.01472v1
+
+## 16846. P-EAGLE: Parallel-Drafting EAGLE with Scalable Training
+
+- **ID:** `2602.01469` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Mude Hui, Xin Huang, Jaime Campos Salas, et al.
+- **abs:** https://arxiv.org/abs/2602.01469 · **PDF:** https://arxiv.org/pdf/2602.01469v1
+
+## 16847. A Statistical Theory of Gated Attention through the Lens of Hierarchical Mixture of Experts
+
+- **ID:** `2602.01468` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Viet Nguyen, Tuan Minh Pham, Thinh Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.01468 · **PDF:** https://arxiv.org/pdf/2602.01468v1
+
+## 16848. Task-Aware LoRA Adapter Composition via Similarity Retrieval in Vector Databases
+
+- **ID:** `2602.21222` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Riya Adsul, Balachandra Devarangadi Sunil, Isha Nalawade, et al.
+- **abs:** https://arxiv.org/abs/2602.21222 · **PDF:** https://arxiv.org/pdf/2602.21222v1
+
+## 16849. Rethinking Multinomial Logistic Mixture of Experts with Sigmoid Gating Function
+
+- **ID:** `2602.01466` · **primary:** `stat.ML` · **Submitted:** 2026-02-01
+- **Authors:** Tuan Minh Pham, Thinh Cao, Viet Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.01466 · **PDF:** https://arxiv.org/pdf/2602.01466v1
+
+## 16850. Agyn: A Multi-Agent System for Team-Based Autonomous Software Engineering
+
+- **ID:** `2602.01465` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Nikita Benkovich, Vitalii Valkov
+- **abs:** https://arxiv.org/abs/2602.01465 · **PDF:** https://arxiv.org/pdf/2602.01465v2
+
+## 16851. GOPO: Policy Optimization using Ranked Rewards
+
+- **ID:** `2602.03876` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Kyuseong Choi, Dwaipayan Saha, Woojeong Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.03876 · **PDF:** https://arxiv.org/pdf/2602.03876v1
+
+## 16852. Non-Uniform Noise-to-Signal Ratio in the REINFORCE Policy-Gradient Estimator
+
+- **ID:** `2602.01460` · **primary:** `math.OC` · **Submitted:** 2026-02-01
+- **Authors:** Haoyu Han, Heng Yang
+- **abs:** https://arxiv.org/abs/2602.01460 · **PDF:** https://arxiv.org/pdf/2602.01460v2
+
+## 16853. Understanding vision transformer robustness through the lens of out-of-distribution detection
+
+- **ID:** `2602.01459` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Joey Kuang, Alexander Wong
+- **abs:** https://arxiv.org/abs/2602.01459 · **PDF:** https://arxiv.org/pdf/2602.01459v1
+
+## 16854. Rectified LpJEPA: Joint-Embedding Predictive Architectures with Sparse and Maximum-Entropy Representations
+
+- **ID:** `2602.01456` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Yilun Kuang, Yash Dagade, Tim G. J. Rudner, et al.
+- **abs:** https://arxiv.org/abs/2602.01456 · **PDF:** https://arxiv.org/pdf/2602.01456v1
+
+## 16855. Modeling Topological Impact on Node Attribute Distributions in Attributed Graphs
+
+- **ID:** `2602.01454` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Amirreza Shiralinasab Langari, Leila Yeganeh, Kim Khoa Nguyen
+- **abs:** https://arxiv.org/abs/2602.01454 · **PDF:** https://arxiv.org/pdf/2602.01454v1
+
+## 16856. Position: 3D Gaussian Splatting Watermarking Should Be Scenario-Driven and Threat-Model Explicit
+
+- **ID:** `2602.02602` · **primary:** `cs.CR` · **Submitted:** 2026-02-01
+- **Authors:** Yangfan Deng, Anirudh Nakra, Min Wu
+- **abs:** https://arxiv.org/abs/2602.02602 · **PDF:** https://arxiv.org/pdf/2602.02602v1
+
+## 16857. Provable Cooperative Multi-Agent Exploration for Reward-Free MDPs
+
+- **ID:** `2602.01453` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Idan Barnea, Orin Levy, Yishay Mansour
+- **abs:** https://arxiv.org/abs/2602.01453 · **PDF:** https://arxiv.org/pdf/2602.01453v2
+
+## 16858. Cross-Paradigm Evaluation of Gaze-Based Semantic Object Identification for Intelligent Vehicles
+
+- **ID:** `2602.01452` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Penghao Deng, Jidong J. Yang, Jiachen Bian
+- **abs:** https://arxiv.org/abs/2602.01452 · **PDF:** https://arxiv.org/pdf/2602.01452v1
+
+## 16859. Understanding QA generation: Extracting Parametric and Contextual Knowledge with CQA for Low Resource Bangla Language
+
+- **ID:** `2602.01451` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Umme Abira Azmary, MD Ikramul Kayes, Swakkhar Shatabda, et al.
+- **abs:** https://arxiv.org/abs/2602.01451 · **PDF:** https://arxiv.org/pdf/2602.01451v1
+
+## 16860. Dimension-Free Multimodal Sampling via Preconditioned Annealed Langevin Dynamics
+
+- **ID:** `2602.01449` · **primary:** `math.NA` · **Submitted:** 2026-02-01
+- **Authors:** Lorenzo Baldassari, Josselin Garnier, Knut Solna, et al.
+- **abs:** https://arxiv.org/abs/2602.01449 · **PDF:** https://arxiv.org/pdf/2602.01449v1
+
+## 16861. SentiFuse: Deep Multi-model Fusion Framework for Robust Sentiment Extraction
+
+- **ID:** `2602.01447` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Hieu Minh Duong, Rupa Ghosh, Cong Hoan Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2602.01447 · **PDF:** https://arxiv.org/pdf/2602.01447v1
+
+## 16862. A Meta-Knowledge-Augmented LLM Framework for Hyperparameter Optimization in Time-Series Forecasting
+
+- **ID:** `2602.01445` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Ons Saadallah, Mátyás andó, Tamás Gábor Orosz
+- **abs:** https://arxiv.org/abs/2602.01445 · **PDF:** https://arxiv.org/pdf/2602.01445v2
+
+## 16863. SimGym: Traffic-Grounded Browser Agents for Offline A/B Testing in E-Commerce
+
+- **ID:** `2602.01443` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Alberto Castelo, Zahra Zanjani Foumani, Ailin Fan, et al.
+- **abs:** https://arxiv.org/abs/2602.01443 · **PDF:** https://arxiv.org/pdf/2602.01443v1
+
+## 16864. The Gradient-Causal Gap: Why Gradient Importance Fails on Complex Tasks
+
+- **ID:** `2602.01442` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Donald Ye
+- **abs:** https://arxiv.org/abs/2602.01442 · **PDF:** https://arxiv.org/pdf/2602.01442v2
+
+## 16865. TQL: Scaling Q-Functions with Transformers by Preventing Attention Collapse
+
+- **ID:** `2602.01439` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Perry Dong, Kuo-Han Hung, Alexander Swerdlow, et al.
+- **abs:** https://arxiv.org/abs/2602.01439 · **PDF:** https://arxiv.org/pdf/2602.01439v1
+
+## 16866. CIPHER: Cryptographic Insecurity Profiling via Hybrid Evaluation of Responses
+
+- **ID:** `2602.01438` · **primary:** `cs.CR` · **Submitted:** 2026-02-01
+- **Authors:** Max Manolov, Tony Gao, Siddharth Shukla, et al.
+- **abs:** https://arxiv.org/abs/2602.01438 · **PDF:** https://arxiv.org/pdf/2602.01438v2
+
+## 16867. Theoretical Analysis of Measure Consistency Regularization for Partially Observed Data
+
+- **ID:** `2602.01437` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Yinsong Wang, Shahin Shahrampour
+- **abs:** https://arxiv.org/abs/2602.01437 · **PDF:** https://arxiv.org/pdf/2602.01437v1
+
+## 16868. Phase Transitions for Feature Learning in Neural Networks
+
+- **ID:** `2602.01434` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Andrea Montanari, Zihao Wang
+- **abs:** https://arxiv.org/abs/2602.01434 · **PDF:** https://arxiv.org/pdf/2602.01434v2
+
+## 16869. DCD: Decomposition-based Causal Discovery from Autocorrelated and Non-Stationary Temporal Data
+
+- **ID:** `2602.01433` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Muhammad Hasan Ferdous, Md Osman Gani
+- **abs:** https://arxiv.org/abs/2602.01433 · **PDF:** https://arxiv.org/pdf/2602.01433v1
+
+## 16870. Improving the Trade-off Between Watermark Strength and Speculative Sampling Efficiency for Language Models
+
+- **ID:** `2602.01428` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Weiqing He, Xiang Li, Li Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.01428 · **PDF:** https://arxiv.org/pdf/2602.01428v2
+
+## 16871. Robust Generalization with Adaptive Optimal Transport Priors for Decision-Focused Learning
+
+- **ID:** `2602.01427` · **primary:** `stat.ML` · **Submitted:** 2026-02-01
+- **Authors:** Haixiang Sun, Andrew L. Liu
+- **abs:** https://arxiv.org/abs/2602.01427 · **PDF:** https://arxiv.org/pdf/2602.01427v2
+
+## 16872. Building Better Deception Probes Using Targeted Instruction Pairs
+
+- **ID:** `2602.01425` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Vikram Natarajan, Devina Jain, Shivam Arora, et al.
+- **abs:** https://arxiv.org/abs/2602.01425 · **PDF:** https://arxiv.org/pdf/2602.01425v1
+
+## 16873. Semi-supervised CAPP Transformer Learning via Pseudo-labeling
+
+- **ID:** `2602.01419` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Dennis Gross, Helge Spieker, Arnaud Gotlieb, et al.
+- **abs:** https://arxiv.org/abs/2602.01419 · **PDF:** https://arxiv.org/pdf/2602.01419v1
+
+## 16874. Where to Attend: A Principled Vision-Centric Position Encoding with Parabolas
+
+- **ID:** `2602.01418` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Christoffer Koo Øhrstrøm, Rafael I. Cabral Muchacho, Yifei Dong, et al.
+- **abs:** https://arxiv.org/abs/2602.01418 · **PDF:** https://arxiv.org/pdf/2602.01418v1
+
+## 16875. Global AI Bias Audit for Technical Governance
+
+- **ID:** `2602.13246` · **primary:** `cs.CY` · **Submitted:** 2026-02-01
+- **Authors:** Jason Hung
+- **abs:** https://arxiv.org/abs/2602.13246 · **PDF:** https://arxiv.org/pdf/2602.13246v1
+
+## 16876. Importance Weighted Variational Inference without the Reparameterization Trick
+
+- **ID:** `2602.01412` · **primary:** `stat.ML` · **Submitted:** 2026-02-01
+- **Authors:** Kamélia Daudel, Minh-Ngoc Tran, Cheng Zhang
+- **abs:** https://arxiv.org/abs/2602.01412 · **PDF:** https://arxiv.org/pdf/2602.01412v1
+
+## 16877. SNIP: An Adaptive Mixed Precision Framework for Subbyte Large Language Model Training
+
+- **ID:** `2602.01410` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Yunjie Pan, Yongyi Yang, Hanmei Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.01410 · **PDF:** https://arxiv.org/pdf/2602.01410v1
+
+## 16878. From Pragmas to Partners: A Symbiotic Evolution of Agentic High-Level Synthesis
+
+- **ID:** `2602.01401` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Niansong Zhang, Sunwoo Kim, Shreesha Srinath, et al.
+- **abs:** https://arxiv.org/abs/2602.01401 · **PDF:** https://arxiv.org/pdf/2602.01401v3
+
+## 16879. Online Social Welfare Function-based Resource Allocation
+
+- **ID:** `2602.01400` · **primary:** `stat.ML` · **Submitted:** 2026-02-01
+- **Authors:** Kanad Pardeshi, Samsara Foubert, Aarti Singh
+- **abs:** https://arxiv.org/abs/2602.01400 · **PDF:** https://arxiv.org/pdf/2602.01400v1
+
+## 16880. An Odd Estimator for Shapley Values
+
+- **ID:** `2602.01399` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Fabian Fumagalli, Landon Butler, Justin Singh Kang, et al.
+- **abs:** https://arxiv.org/abs/2602.01399 · **PDF:** https://arxiv.org/pdf/2602.01399v1
+
+## 16881. Nonlinear model reduction for transport-dominated problems
+
+- **ID:** `2602.01397` · **primary:** `math.NA` · **Submitted:** 2026-02-01
+- **Authors:** Jan S. Hesthaven, Benjamin Peherstorfer, Benjamin Unger
+- **abs:** https://arxiv.org/abs/2602.01397 · **PDF:** https://arxiv.org/pdf/2602.01397v1
+
+## 16882. Rethinking Selective Knowledge Distillation
+
+- **ID:** `2602.01395` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Almog Tavor, Itay Ebenspanger, Neil Cnaan, et al.
+- **abs:** https://arxiv.org/abs/2602.01395 · **PDF:** https://arxiv.org/pdf/2602.01395v1
+
+## 16883. SSNAPS: Audio-Visual Separation of Speech and Background Noise with Diffusion Inverse Sampling
+
+- **ID:** `2602.01394` · **primary:** `eess.AS` · **Submitted:** 2026-02-01
+- **Authors:** Yochai Yemini, Yoav Ellinson, Rami Ben-Ari, et al.
+- **abs:** https://arxiv.org/abs/2602.01394 · **PDF:** https://arxiv.org/pdf/2602.01394v1
+
+## 16884. Adaptive traffic signal control optimization using a novel road partition and multi-channel state representation method
+
+- **ID:** `2602.12296` · **primary:** `eess.SY` · **Submitted:** 2026-02-01
+- **Authors:** Maojiang Deng, Shoufeng Lu, Jiazhao Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.12296 · **PDF:** https://arxiv.org/pdf/2602.12296v1
+
+## 16885. How well can VLMs rate audio descriptions: A multi-dimensional quantitative assessment framework
+
+- **ID:** `2602.01390` · **primary:** `cs.HC` · **Submitted:** 2026-02-01
+- **Authors:** Lana Do, Gio Jung, Juvenal Francisco Barajas, et al.
+- **abs:** https://arxiv.org/abs/2602.01390 · **PDF:** https://arxiv.org/pdf/2602.01390v1
+
+## 16886. The Enhanced Physics-Informed Kolmogorov-Arnold Networks: Applications of Newton's Laws in Financial Deep Reinforcement Learning (RL) Algorithms
+
+- **ID:** `2602.01388` · **primary:** `cs.CE` · **Submitted:** 2026-02-01
+- **Authors:** Trang Thoi, Hung Tran, Tram Thoi, et al.
+- **abs:** https://arxiv.org/abs/2602.01388 · **PDF:** https://arxiv.org/pdf/2602.01388v2
+
+## 16887. Reversible Deep Learning for 13C NMR in Chemoinformatics: On Structures and Spectra
+
+- **ID:** `2602.03875` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Stefan Kuhn, Vandana Dwarka, Przemyslaw Karol Grenda, et al.
+- **abs:** https://arxiv.org/abs/2602.03875 · **PDF:** https://arxiv.org/pdf/2602.03875v3
+
+## 16888. "If You're Very Clever, No One Knows You've Used It": The Social Dynamics of Developing Generative AI Literacy in the Workplace
+
+- **ID:** `2602.01386` · **primary:** `cs.HC` · **Submitted:** 2026-02-01
+- **Authors:** Qing Nancy Xia, Marios Constantinides, Advait Sarkar, et al.
+- **abs:** https://arxiv.org/abs/2602.01386 · **PDF:** https://arxiv.org/pdf/2602.01386v1
+
+## 16889. Autonomous AI Agents for Option Hedging: Enhancing Financial Stability through Shortfall Aware Reinforcement Learning
+
+- **ID:** `2603.06587` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Minxuan Hu, Ziheng Chen, Jiayu Yi, et al.
+- **abs:** https://arxiv.org/abs/2603.06587 · **PDF:** https://arxiv.org/pdf/2603.06587v1
+
+## 16890. PromptRL: Prompt Matters in RL for Flow-Based Image Generation
+
+- **ID:** `2602.01382` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Fu-Yun Wang, Han Zhang, Michael Gharbi, et al.
+- **abs:** https://arxiv.org/abs/2602.01382 · **PDF:** https://arxiv.org/pdf/2602.01382v1
+
+## 16891. On the Power of (Approximate) Reward Models for Inference-Time Scaling
+
+- **ID:** `2602.01381` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Youheng Zhu, Yiping Lu
+- **abs:** https://arxiv.org/abs/2602.01381 · **PDF:** https://arxiv.org/pdf/2602.01381v1
+
+## 16892. WAKESET: A Large-Scale, High-Reynolds Number Flow Dataset for Machine Learning of Turbulent Wake Dynamics
+
+- **ID:** `2602.01379` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-02-01
+- **Authors:** Zachary Cooper-Baldock, Paulo E. Santos, Russell S. A. Brinkworth, et al.
+- **abs:** https://arxiv.org/abs/2602.01379 · **PDF:** https://arxiv.org/pdf/2602.01379v2
+
+## 16893. Context Dependence and Reliability in Autoregressive Language Models
+
+- **ID:** `2602.01378` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Poushali Sengupta, Shashi Raj Pandey, Sabita Maharjan, et al.
+- **abs:** https://arxiv.org/abs/2602.01378 · **PDF:** https://arxiv.org/pdf/2602.01378v1
+
+## 16894. Robust Sublinear Convergence Rates for Iterative Bregman Projections
+
+- **ID:** `2602.01372` · **primary:** `math.OC` · **Submitted:** 2026-02-01
+- **Authors:** Gabriel Peyré
+- **abs:** https://arxiv.org/abs/2602.01372 · **PDF:** https://arxiv.org/pdf/2602.01372v1
+
+## 16895. PolyGen: Fully Synthetic Vision-Language Training via Multi-Generator Ensembles
+
+- **ID:** `2602.01370` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Leonardo Brusini, Cristian Sbrolli, Eugenio Lomurno, et al.
+- **abs:** https://arxiv.org/abs/2602.01370 · **PDF:** https://arxiv.org/pdf/2602.01370v1
+
+## 16896. Deep Variational Contrastive Learning for Joint Risk Stratification and Time-to-Event Estimation
+
+- **ID:** `2602.01367` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Pinar Erbil, Alberto Archetti, Eugenio Lomurno, et al.
+- **abs:** https://arxiv.org/abs/2602.01367 · **PDF:** https://arxiv.org/pdf/2602.01367v1
+
+## 16897. When Domains Interact: Asymmetric and Order-Sensitive Cross-Domain Effects in Reinforcement Learning for Reasoning
+
+- **ID:** `2602.01365` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Wang Yang, Shouren Wang, Chaoda Song, et al.
+- **abs:** https://arxiv.org/abs/2602.01365 · **PDF:** https://arxiv.org/pdf/2602.01365v1
+
+## 16898. Causally Disentangled Contrastive Learning for Multilingual Speaker Embeddings
+
+- **ID:** `2602.01363` · **primary:** `cs.SD` · **Submitted:** 2026-02-01
+- **Authors:** Mariëtte Olijslager, Seyed Sahand Mohammadi Ziabari, Ali Mohammed Mansoor Alsahag
+- **abs:** https://arxiv.org/abs/2602.01363 · **PDF:** https://arxiv.org/pdf/2602.01363v1
+
+## 16899. Balancing Understanding and Generation in Discrete Diffusion Models
+
+- **ID:** `2602.01362` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Yue Liu, Yuzhong Zhao, Zheyong Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.01362 · **PDF:** https://arxiv.org/pdf/2602.01362v1
+
+## 16900. PaAno: Patch-Based Representation Learning for Time-Series Anomaly Detection
+
+- **ID:** `2602.01359` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Jinju Park, Seokho Kang
+- **abs:** https://arxiv.org/abs/2602.01359 · **PDF:** https://arxiv.org/pdf/2602.01359v1
+
+## 16901. Towards knowledge-based workflows: a semantic approach to atomistic simulations for mechanical and thermodynamic properties
+
+- **ID:** `2602.01358` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-01
+- **Authors:** Abril Azocar Guzman, Hoang-Thien Luu, Sarath Menon, et al.
+- **abs:** https://arxiv.org/abs/2602.01358 · **PDF:** https://arxiv.org/pdf/2602.01358v1
+
+## 16902. Your Self-Play Algorithm is Secretly an Adversarial Imitator: Understanding LLM Self-Play through the Lens of Imitation Learning
+
+- **ID:** `2602.01357` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Shangzhe Li, Xuchao Zhang, Chetan Bansal, et al.
+- **abs:** https://arxiv.org/abs/2602.01357 · **PDF:** https://arxiv.org/pdf/2602.01357v1
+
+## 16903. Aggregation Queries over Unstructured Text: Benchmark and Agentic Method
+
+- **ID:** `2602.01355` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Haojia Zhu, Qinyuan Xu, Haoyu Li, et al.
+- **abs:** https://arxiv.org/abs/2602.01355 · **PDF:** https://arxiv.org/pdf/2602.01355v2
+
+## 16904. CaST: Causal Discovery via Spatio-Temporal Graphs in Disaster Tweets
+
+- **ID:** `2602.02601` · **primary:** `cs.SI` · **Submitted:** 2026-02-01
+- **Authors:** Hieu Duong, Eugene Levin, Todd Gary, et al.
+- **abs:** https://arxiv.org/abs/2602.02601 · **PDF:** https://arxiv.org/pdf/2602.02601v1
+
+## 16905. Step-Wise Refusal Dynamics in Autoregressive and Diffusion Language Models
+
+- **ID:** `2602.02600` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Eliron Rahimi, Elad Hirshel, Rom Himelstein, et al.
+- **abs:** https://arxiv.org/abs/2602.02600 · **PDF:** https://arxiv.org/pdf/2602.02600v2
+
+## 16906. CRAFT: Calibrated Reasoning with Answer-Faithful Traces via Reinforcement Learning for Multi-Hop Question Answering
+
+- **ID:** `2602.01348` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Yu Liu, Wenxiao Zhang, Diandian Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.01348 · **PDF:** https://arxiv.org/pdf/2602.01348v2
+
+## 16907. Model Specific Task Similarity for Vision Language Model Selection via Layer Conductance
+
+- **ID:** `2602.01346` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Wei Yang, Hong Xie, Tao Tan, et al.
+- **abs:** https://arxiv.org/abs/2602.01346 · **PDF:** https://arxiv.org/pdf/2602.01346v1
+
+## 16908. RAP: KV-Cache Compression via RoPE-Aligned Pruning
+
+- **ID:** `2602.02599` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Jihao Xin, Tian Lyu, David Keyes, et al.
+- **abs:** https://arxiv.org/abs/2602.02599 · **PDF:** https://arxiv.org/pdf/2602.02599v3
+
+## 16909. Adaptive Quantum-Safe Cryptography for 6G Vehicular Networks via Context-Aware Optimization
+
+- **ID:** `2602.01342` · **primary:** `cs.CR` · **Submitted:** 2026-02-01
+- **Authors:** Poushali Sengupta, Mayank Raikwar, Sabita Maharjan, et al.
+- **abs:** https://arxiv.org/abs/2602.01342 · **PDF:** https://arxiv.org/pdf/2602.01342v1
+
+## 16910. Social Catalysts, Not Moral Agents: The Illusion of Alignment in LLM Societies
+
+- **ID:** `2602.02598` · **primary:** `physics.soc-ph` · **Submitted:** 2026-02-01
+- **Authors:** Yueqing Hu, Yixuan Jiang, Zehua Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.02598 · **PDF:** https://arxiv.org/pdf/2602.02598v1
+
+## 16911. Finding Differentially Private Second Order Stationary Points in Stochastic Minimax Optimization
+
+- **ID:** `2602.01339` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Difei Xu, Youming Tao, Meng Ding, et al.
+- **abs:** https://arxiv.org/abs/2602.01339 · **PDF:** https://arxiv.org/pdf/2602.01339v1
+
+## 16912. High-accuracy sampling for diffusion models and log-concave distributions
+
+- **ID:** `2602.01338` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Fan Chen, Sinho Chewi, Constantinos Daskalakis, et al.
+- **abs:** https://arxiv.org/abs/2602.01338 · **PDF:** https://arxiv.org/pdf/2602.01338v1
+
+## 16913. Beyond Pixels: Visual Metaphor Transfer via Schema-Driven Agentic Reasoning
+
+- **ID:** `2602.01335` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Yu Xu, Yuxin Zhang, Juan Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.01335 · **PDF:** https://arxiv.org/pdf/2602.01335v1
+
+## 16914. A-MapReduce: Executing Wide Search via Agentic MapReduce
+
+- **ID:** `2602.01331` · **primary:** `cs.MA` · **Submitted:** 2026-02-01
+- **Authors:** Mingju Chen, Guibin Zhang, Heng Chang, et al.
+- **abs:** https://arxiv.org/abs/2602.01331 · **PDF:** https://arxiv.org/pdf/2602.01331v1
+
+## 16915. ContextEvolve: Multi-Agent Context Compression for Systems Code Optimization
+
+- **ID:** `2602.02597` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Hongyuan Su, Yu Zheng, Yong Li
+- **abs:** https://arxiv.org/abs/2602.02597 · **PDF:** https://arxiv.org/pdf/2602.02597v1
+
+## 16916. DreamOn: Diffusion Language Models For Code Infilling Beyond Fixed-size Canvas
+
+- **ID:** `2602.01326` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Zirui Wu, Lin Zheng, Zhihui Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.01326 · **PDF:** https://arxiv.org/pdf/2602.01326v1
+
+## 16917. PolySAE: Modeling Feature Interactions in Sparse Autoencoders via Polynomial Decoding
+
+- **ID:** `2602.01322` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Panagiotis Koromilas, Andreas D. Demou, James Oldfield, et al.
+- **abs:** https://arxiv.org/abs/2602.01322 · **PDF:** https://arxiv.org/pdf/2602.01322v1
+
+## 16918. TxRay: Agentic Postmortem of Live Blockchain Attacks
+
+- **ID:** `2602.01317` · **primary:** `cs.CR` · **Submitted:** 2026-02-01
+- **Authors:** Ziyue Wang, Jiangshan Yu, Kaihua Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.01317 · **PDF:** https://arxiv.org/pdf/2602.01317v5
+
+## 16919. Evaluating Long-Horizon Memory for Multi-Party Collaborative Dialogues
+
+- **ID:** `2602.01313` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Chuanrui Hu, Tong Li, Xingze Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.01313 · **PDF:** https://arxiv.org/pdf/2602.01313v3
+
+## 16920. Imperfect Influence, Preserved Rankings: A Theory of TRAK for Data Attribution
+
+- **ID:** `2602.01312` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Han Tong, Shubhangi Ghosh, Haolin Zou, et al.
+- **abs:** https://arxiv.org/abs/2602.01312 · **PDF:** https://arxiv.org/pdf/2602.01312v1
+
+## 16921. Dispelling the Curse of Singularities in Neural Network Optimizations
+
+- **ID:** `2602.01308` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Hengjie Cao, Mengyi Chen, Yifeng Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.01308 · **PDF:** https://arxiv.org/pdf/2602.01308v2
+
+## 16922. Fubini Study geometry of representation drift in high dimensional data
+
+- **ID:** `2602.02596` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Arturo Tozzi
+- **abs:** https://arxiv.org/abs/2602.02596 · **PDF:** https://arxiv.org/pdf/2602.02596v1
+
+## 16923. RE-MCDF: Closed-Loop Multi-Expert LLM Reasoning for Knowledge-Grounded Clinical Diagnosis
+
+- **ID:** `2602.01297` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Shaowei Shen, Xiaohong Yang, Jie Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.01297 · **PDF:** https://arxiv.org/pdf/2602.01297v2
+
+## 16924. Best-of-Both-Worlds for Heavy-Tailed Markov Decision Processes
+
+- **ID:** `2602.01295` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Yu Chen, Yuhao Liu, Jiatai Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.01295 · **PDF:** https://arxiv.org/pdf/2602.01295v2
+
+## 16925. Dynamic Heuristic Neuromorphic Solver for the Edge User Allocation Problem with Bayesian Confidence Propagation Neural Network
+
+- **ID:** `2602.01294` · **primary:** `cs.NE` · **Submitted:** 2026-02-01
+- **Authors:** Kecheng Zhang, Anders Lansner, Ahsan Javed Awan, et al.
+- **abs:** https://arxiv.org/abs/2602.01294 · **PDF:** https://arxiv.org/pdf/2602.01294v1
+
+## 16926. Gradient-Aligned Calibration for Post-Training Quantization of Diffusion Models
+
+- **ID:** `2602.01289` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Dung Anh Hoang, Cuong Pham anh Trung Le, Jianfei Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.01289 · **PDF:** https://arxiv.org/pdf/2602.01289v3
+
+## 16927. EDIS: Diagnosing LLM Reasoning via Entropy Dynamics
+
+- **ID:** `2602.01288` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Chenghua Zhu, Siyan Wu, Xiangkang Zeng, et al.
+- **abs:** https://arxiv.org/abs/2602.01288 · **PDF:** https://arxiv.org/pdf/2602.01288v2
+
+## 16928. Multi-LLM Adaptive Conformal Inference for Reliable LLM Responses
+
+- **ID:** `2602.01285` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Kangjun Noh, Seongchan Lee, Ilmun Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.01285 · **PDF:** https://arxiv.org/pdf/2602.01285v1
+
+## 16929. Richer Bayesian Last Layers with Subsampled NTK Features
+
+- **ID:** `2602.01279` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Sergio Calvo-Ordoñez, Jonathan Plenk, Richard Bergna, et al.
+- **abs:** https://arxiv.org/abs/2602.01279 · **PDF:** https://arxiv.org/pdf/2602.01279v1
+
+## 16930. LLM-Driven Ontology Construction for Enterprise Knowledge Graphs
+
+- **ID:** `2602.01276` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Abdulsobur Oyewale, Tommaso Soru
+- **abs:** https://arxiv.org/abs/2602.01276 · **PDF:** https://arxiv.org/pdf/2602.01276v1
+
+## 16931. PACER: Blockwise Pre-verification for Speculative Decoding with Adaptive Length
+
+- **ID:** `2602.01274` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Situo Zhang, Yifan Zhang, Zichen Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.01274 · **PDF:** https://arxiv.org/pdf/2602.01274v1
+
+## 16932. From Intents to Actions: Agentic AI in Autonomous Networks
+
+- **ID:** `2602.01271` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Burak Demirel, Pablo Soldati, Yu Wang
+- **abs:** https://arxiv.org/abs/2602.01271 · **PDF:** https://arxiv.org/pdf/2602.01271v1
+
+## 16933. Mixture-of-World Models: Scaling Multi-Task Reinforcement Learning with Modular Latent Dynamics
+
+- **ID:** `2602.01270` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Boxuan Zhang, Weipu Zhang, Zhaohan Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.01270 · **PDF:** https://arxiv.org/pdf/2602.01270v1
+
+## 16934. Diving into Kronecker Adapters: Component Design Matters
+
+- **ID:** `2602.01267` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Jiayu Bai, Danchen Yu, Zhenyu Liao, et al.
+- **abs:** https://arxiv.org/abs/2602.01267 · **PDF:** https://arxiv.org/pdf/2602.01267v1
+
+## 16935. BicKD: Bilateral Contrastive Knowledge Distillation
+
+- **ID:** `2602.01265` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Jiangnan Zhu, Yukai Xu, Li Xiong, et al.
+- **abs:** https://arxiv.org/abs/2602.01265 · **PDF:** https://arxiv.org/pdf/2602.01265v1
+
+## 16936. Sample Efficient Active Algorithms for Offline Reinforcement Learning
+
+- **ID:** `2602.01260` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Soumyadeep Roy, Shashwat Kushwaha, Ambedkar Dukkipati
+- **abs:** https://arxiv.org/abs/2602.01260 · **PDF:** https://arxiv.org/pdf/2602.01260v1
+
+## 16937. Mechanistic Interpretability of Brain-to-Speech Models Across Speech Modes
+
+- **ID:** `2602.01247` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Maryam Maghsoudi, Ayushi Mishra
+- **abs:** https://arxiv.org/abs/2602.01247 · **PDF:** https://arxiv.org/pdf/2602.01247v1
+
+## 16938. PARSE: An Open-Domain Reasoning Question Answering Benchmark for Persian
+
+- **ID:** `2602.01246` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Jamshid Mozafari, Seyed Parsa Mousavinasab, Adam Jatowt
+- **abs:** https://arxiv.org/abs/2602.01246 · **PDF:** https://arxiv.org/pdf/2602.01246v1
+
+## 16939. Large-Scale Terminal Agentic Trajectory Generation from Dockerized Environments
+
+- **ID:** `2602.01244` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Siwei Wu, Yizhi Li, Yuyang Song, et al.
+- **abs:** https://arxiv.org/abs/2602.01244 · **PDF:** https://arxiv.org/pdf/2602.01244v2
+
+## 16940. Minimizing Mismatch Risk: A Prototype-Based Routing Framework for Zero-shot LLM-generated Text Detection
+
+- **ID:** `2602.01240` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Ke Sun, Guangsheng Bao, Han Cui, et al.
+- **abs:** https://arxiv.org/abs/2602.01240 · **PDF:** https://arxiv.org/pdf/2602.01240v1
+
+## 16941. Inferential Question Answering
+
+- **ID:** `2602.01239` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Jamshid Mozafari, Hamed Zamani, Guido Zuccon, et al.
+- **abs:** https://arxiv.org/abs/2602.01239 · **PDF:** https://arxiv.org/pdf/2602.01239v1
+
+## 16942. Predictive Scheduling for Efficient Inference-Time Reasoning in Large Language Models
+
+- **ID:** `2602.01237` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Katrina Brown, Aneesh Muppidi, Rana Shahout
+- **abs:** https://arxiv.org/abs/2602.01237 · **PDF:** https://arxiv.org/pdf/2602.01237v1
+
+## 16943. Lotus: Efficient LLM Training by Randomized Low-Rank Gradient Projection with Adaptive Subspace Switching
+
+- **ID:** `2602.01233` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Tianhao Miao, Zhongyuan Bao, Lejun Zhang
+- **abs:** https://arxiv.org/abs/2602.01233 · **PDF:** https://arxiv.org/pdf/2602.01233v1
+
+## 16944. Supervised Fine-Tuning Needs to Unlock the Potential of Token Priority
+
+- **ID:** `2602.01227` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Zhanming Shen, Zeyu Qin, Jiaqi Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.01227 · **PDF:** https://arxiv.org/pdf/2602.01227v2
+
+## 16945. FutureMind: Equipping Small Language Models with Strategic Thinking-Pattern Priors via Adaptive Knowledge Distillation
+
+- **ID:** `2602.01222` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Shaoxiong Yang, Junting Li, Mengyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01222 · **PDF:** https://arxiv.org/pdf/2602.01222v1
+
+## 16946. MiTA Attention: Efficient Fast-Weight Scaling via a Mixture of Top-k Activations
+
+- **ID:** `2602.01219` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Qishuai Wen, Zhiyuan Huang, Xianghan Meng, et al.
+- **abs:** https://arxiv.org/abs/2602.01219 · **PDF:** https://arxiv.org/pdf/2602.01219v3
+
+## 16947. Learning from Anonymized and Incomplete Tabular Data
+
+- **ID:** `2602.01217` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Lucas Lange, Adrian Böttinger, Victor Christen, et al.
+- **abs:** https://arxiv.org/abs/2602.01217 · **PDF:** https://arxiv.org/pdf/2602.01217v1
+
+## 16948. AI Meets Plasticity: A Comprehensive Survey
+
+- **ID:** `2602.01215` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-01
+- **Authors:** Hadi Bakhshan, Sima Farshbaf, Junior Ramirez Machado, et al.
+- **abs:** https://arxiv.org/abs/2602.01215 · **PDF:** https://arxiv.org/pdf/2602.01215v1
+
+## 16949. SimpleGPT: Improving GPT via A Simple Normalization Strategy
+
+- **ID:** `2602.01212` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Marco Chen, Xianbiao Qi, Yelin He, et al.
+- **abs:** https://arxiv.org/abs/2602.01212 · **PDF:** https://arxiv.org/pdf/2602.01212v1
+
+## 16950. Chronos: Learning Temporal Dynamics of Reasoning Chains for Test-Time Scaling
+
+- **ID:** `2602.01208` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Kai Zhang, Jiayi Liao, Chengpeng Li, et al.
+- **abs:** https://arxiv.org/abs/2602.01208 · **PDF:** https://arxiv.org/pdf/2602.01208v1
+
+## 16951. Not All Preferences Are Created Equal: Stability-Aware and Gradient-Efficient Alignment for Reasoning Models
+
+- **ID:** `2602.01207` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Hui Wu, Hengyi Cai, Jinman Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.01207 · **PDF:** https://arxiv.org/pdf/2602.01207v1
+
+## 16952. Addressing Explainability of Generative AI using SMILE (Statistical Model-agnostic Interpretability with Local Explanations)
+
+- **ID:** `2602.01206` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Zeinab Dehghani
+- **abs:** https://arxiv.org/abs/2602.01206 · **PDF:** https://arxiv.org/pdf/2602.01206v1
+
+## 16953. ASTER: Agentic Scaling with Tool-integrated Extended Reasoning
+
+- **ID:** `2602.01204` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Xuqin Zhang, Quan He, Zhenrui Zheng, et al.
+- **abs:** https://arxiv.org/abs/2602.01204 · **PDF:** https://arxiv.org/pdf/2602.01204v1
+
+## 16954. Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse
+
+- **ID:** `2602.01203` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Zizhuo Fu, Wenxuan Zeng, Runsheng Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01203 · **PDF:** https://arxiv.org/pdf/2602.01203v1
+
+## 16955. Workflow-R1: Group Sub-sequence Policy Optimization for Multi-turn Workflow Construction
+
+- **ID:** `2602.01202` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Mingze Kong, Zikun Qu, Zhongquan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.01202 · **PDF:** https://arxiv.org/pdf/2602.01202v1
+
+## 16956. A State-Transition Framework for Efficient LLM Reasoning
+
+- **ID:** `2602.01198` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Liang Zhang, Yu Zhao, Longyue Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01198 · **PDF:** https://arxiv.org/pdf/2602.01198v1
+
+## 16957. To Defend Against Cyber Attacks, We Must Teach AI Agents to Hack
+
+- **ID:** `2602.02595` · **primary:** `cs.CR` · **Submitted:** 2026-02-01
+- **Authors:** Terry Yue Zhuo, Yangruibo Ding, Wenbo Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.02595 · **PDF:** https://arxiv.org/pdf/2602.02595v1
+
+## 16958. Unraveling the Hidden Dynamical Structure in Recurrent Neural Policies
+
+- **ID:** `2602.01196` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Jin Li, Yue Wu, Mengsha Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.01196 · **PDF:** https://arxiv.org/pdf/2602.01196v1
+
+## 16959. Bridging Lexical Ambiguity and Vision: A Mini Review on Visual Word Sense Disambiguation
+
+- **ID:** `2602.01193` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Shashini Nilukshi, Deshan Sumanathilaka
+- **abs:** https://arxiv.org/abs/2602.01193 · **PDF:** https://arxiv.org/pdf/2602.01193v1
+
+## 16960. Autoregressive, Yet Revisable: In Decoding Revision for Secure Code Generation
+
+- **ID:** `2602.01187` · **primary:** `cs.SE` · **Submitted:** 2026-02-01
+- **Authors:** Chengran Yang, Zichao Wei, Heminghao Deng, et al.
+- **abs:** https://arxiv.org/abs/2602.01187 · **PDF:** https://arxiv.org/pdf/2602.01187v1
+
+## 16961. The Gaussian-Head OFL Family: One-Shot Federated Learning from Client Global Statistics
+
+- **ID:** `2602.01186` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Fabio Turazza, Marco Picone, Marco Mamei
+- **abs:** https://arxiv.org/abs/2602.01186 · **PDF:** https://arxiv.org/pdf/2602.01186v1
+
+## 16962. FedBGS: A Blockchain Approach to Segment Gossip Learning in Decentralized Systems
+
+- **ID:** `2602.01185` · **primary:** `cs.CR` · **Submitted:** 2026-02-01
+- **Authors:** Fabio Turazza, Marcello Pietri, Marco Picone, et al.
+- **abs:** https://arxiv.org/abs/2602.01185 · **PDF:** https://arxiv.org/pdf/2602.01185v1
+
+## 16963. Refining Context-Entangled Content Segmentation via Curriculum Selection and Anti-Curriculum Promotion
+
+- **ID:** `2602.01183` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Chunming He, Rihan Zhang, Fengyang Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.01183 · **PDF:** https://arxiv.org/pdf/2602.01183v1
+
+## 16964. Analyzing and Improving Diffusion Models for Time-Series Data Imputation: A Proximal Recursion Perspective
+
+- **ID:** `2602.01182` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Zhichao Chen, Hao Wang, Fangyikang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01182 · **PDF:** https://arxiv.org/pdf/2602.01182v1
+
+## 16965. Rethinking the Flow-Based Gradual Domain Adaption: A Semi-Dual Optimal Transport Perspective
+
+- **ID:** `2602.01179` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Zhichao Chen, Zhan Zhuang, Yunfei Teng, et al.
+- **abs:** https://arxiv.org/abs/2602.01179 · **PDF:** https://arxiv.org/pdf/2602.01179v1
+
+## 16966. Equivalence of Privacy and Stability with Generalization Guarantees in Quantum Learning
+
+- **ID:** `2602.01177` · **primary:** `quant-ph` · **Submitted:** 2026-02-01
+- **Authors:** Ayanava Dasgupta, Naqueeb Ahmad Warsi, Masahito Hayashi
+- **abs:** https://arxiv.org/abs/2602.01177 · **PDF:** https://arxiv.org/pdf/2602.01177v2
+
+## 16967. Multi-Fidelity Physics-Informed Neural Networks with Bayesian Uncertainty Quantification and Adaptive Residual Learning for Efficient Solution of Parametric Partial Differential Equations
+
+- **ID:** `2602.01176` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Olaf Yunus Laitinen Imanov
+- **abs:** https://arxiv.org/abs/2602.01176 · **PDF:** https://arxiv.org/pdf/2602.01176v1
+
+## 16968. ASP-Bench: From Natural Language to Logic Programs
+
+- **ID:** `2602.01171` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Stefan Szeider
+- **abs:** https://arxiv.org/abs/2602.01171 · **PDF:** https://arxiv.org/pdf/2602.01171v1
+
+## 16969. EmoAra: Emotion-Preserving English Speech Transcription and Cross-Lingual Translation with Arabic Text-to-Speech
+
+- **ID:** `2602.01170` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Besher Hassan, Ibrahim Alsarraj, Musaab Hasan, et al.
+- **abs:** https://arxiv.org/abs/2602.01170 · **PDF:** https://arxiv.org/pdf/2602.01170v1
+
+## 16970. PedagoSense: A Pedology Grounded LLM System for Pedagogical Strategy Detection and Contextual Response Generation in Learning Dialogues
+
+- **ID:** `2602.01169` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Shahem Sultan, Shahem Fadi, Yousef Melhim, et al.
+- **abs:** https://arxiv.org/abs/2602.01169 · **PDF:** https://arxiv.org/pdf/2602.01169v1
+
+## 16971. Do All Individual Layers Help? An Empirical Study of Task-Interfering Layers in Vision-Language Models
+
+- **ID:** `2602.01167` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Zhiming Liu, Yujie Wei, Lei Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.01167 · **PDF:** https://arxiv.org/pdf/2602.01167v1
+
+## 16972. Semantically Aware UAV Landing Site Assessment from Remote Sensing Imagery via Multimodal Large Language Models
+
+- **ID:** `2602.01163` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Chunliang Hua, Zeyuan Yang, Lei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01163 · **PDF:** https://arxiv.org/pdf/2602.01163v1
+
+## 16973. Typologically-Informed Candidate Reranking for LLM-based Translation into Low-Resource Languages
+
+- **ID:** `2602.01162` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Nipuna Abeykoon, Ashen Weerathunga, Pubudu Wijesinghe, et al.
+- **abs:** https://arxiv.org/abs/2602.01162 · **PDF:** https://arxiv.org/pdf/2602.01162v1
+
+## 16974. Beyond Training for Cultural Awareness: The Role of Dataset Linguistic Structure in Large Language Models
+
+- **ID:** `2602.01161` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Reem I. Masoud, Chen Feng, Shunta Asano, et al.
+- **abs:** https://arxiv.org/abs/2602.01161 · **PDF:** https://arxiv.org/pdf/2602.01161v1
+
+## 16975. Deep Time-Series Models Meet Volatility: Multi-Horizon Electricity Price Forecasting in the Australian National Electricity Market
+
+- **ID:** `2602.01157` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Mohammed Osman Gani, Zhipeng He, Chun Ouyang, et al.
+- **abs:** https://arxiv.org/abs/2602.01157 · **PDF:** https://arxiv.org/pdf/2602.01157v2
+
+## 16976. PolicyFlow: Policy Optimization with Continuous Normalizing Flow in Reinforcement Learning
+
+- **ID:** `2602.01156` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Shunpeng Yang, Ben Liu, Hua Chen
+- **abs:** https://arxiv.org/abs/2602.01156 · **PDF:** https://arxiv.org/pdf/2602.01156v1
+
+## 16977. Multi-Agent Causal Reasoning System for Error Pattern Rule Automation in Vehicles
+
+- **ID:** `2602.01155` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Hugo Math, Julian Lorenz, Stefan Oelsner, et al.
+- **abs:** https://arxiv.org/abs/2602.01155 · **PDF:** https://arxiv.org/pdf/2602.01155v2
+
+## 16978. Statistical MIA: Rethinking Membership Inference Attack for Reliable Unlearning Auditing
+
+- **ID:** `2602.01150` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Jialong Sun, Zeming Wei, Jiaxuan Zou, et al.
+- **abs:** https://arxiv.org/abs/2602.01150 · **PDF:** https://arxiv.org/pdf/2602.01150v1
+
+## 16979. Robust Machine Learning Framework for Reliable Discovery of High-Performance Half-Heusler Thermoelectrics
+
+- **ID:** `2602.01149` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-02-01
+- **Authors:** Shoeb Athar, Adrien Mecibah, Philippe Jund
+- **abs:** https://arxiv.org/abs/2602.01149 · **PDF:** https://arxiv.org/pdf/2602.01149v1
+
+## 16980. Capabilities and Fundamental Limits of Latent Chain-of-Thought
+
+- **ID:** `2602.01148` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Jiaxuan Zou, Yaozhong Xiong, Yong Liu
+- **abs:** https://arxiv.org/abs/2602.01148 · **PDF:** https://arxiv.org/pdf/2602.01148v1
+
+## 16981. Unleashing the Potential of Differential Evolution through Individual-Level Strategy Diversity
+
+- **ID:** `2602.01147` · **primary:** `cs.NE` · **Submitted:** 2026-02-01
+- **Authors:** Chenchen Feng, Minyang Chen, Zhuozhao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.01147 · **PDF:** https://arxiv.org/pdf/2602.01147v1
+
+## 16982. PersistBench: When Should Long-Term Memories Be Forgotten by LLMs?
+
+- **ID:** `2602.01146` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Sidharth Pulipaka, Oliver Chen, Manas Sharma, et al.
+- **abs:** https://arxiv.org/abs/2602.01146 · **PDF:** https://arxiv.org/pdf/2602.01146v1
+
+## 16983. Effective Frontiers: A Unification of Neural Scaling Laws
+
+- **ID:** `2602.02593` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Jiaxuan Zou, Zixuan Gong, Ye Su, et al.
+- **abs:** https://arxiv.org/abs/2602.02593 · **PDF:** https://arxiv.org/pdf/2602.02593v1
+
+## 16984. Generalized Radius and Integrated Codebook Transforms for Differentiable Vector Quantization
+
+- **ID:** `2602.01140` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Haochen You, Heng Zhang, Hongyang He, et al.
+- **abs:** https://arxiv.org/abs/2602.01140 · **PDF:** https://arxiv.org/pdf/2602.01140v1
+
+## 16985. Key Principles of Graph Machine Learning: Representation, Robustness, and Generalization
+
+- **ID:** `2602.01139` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Yassine Abbahaddou
+- **abs:** https://arxiv.org/abs/2602.01139 · **PDF:** https://arxiv.org/pdf/2602.01139v1
+
+## 16986. Self-Generative Adversarial Fine-Tuning for Large Language Models
+
+- **ID:** `2602.01137` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Shiguang Wu, Yaqing Wang, Quanming Yao
+- **abs:** https://arxiv.org/abs/2602.01137 · **PDF:** https://arxiv.org/pdf/2602.01137v1
+
+## 16987. A Unified Matrix-Spectral Framework for Stability and Interpretability in Deep Learning
+
+- **ID:** `2602.01136` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Ronald Katende
+- **abs:** https://arxiv.org/abs/2602.01136 · **PDF:** https://arxiv.org/pdf/2602.01136v1
+
+## 16988. Scalable Sample-Level Causal Discovery in Event Sequences via Autoregressive Density Estimation
+
+- **ID:** `2602.01135` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Hugo Math, Rainer Lienhart
+- **abs:** https://arxiv.org/abs/2602.01135 · **PDF:** https://arxiv.org/pdf/2602.01135v2
+
+## 16989. Parallel Training in Spiking Neural Networks
+
+- **ID:** `2602.01133` · **primary:** `cs.NE` · **Submitted:** 2026-02-01
+- **Authors:** Yanbin Huang, Man Yao, Yuqi Pan, et al.
+- **abs:** https://arxiv.org/abs/2602.01133 · **PDF:** https://arxiv.org/pdf/2602.01133v1
+
+## 16990. Don't Judge a Book by its Cover: Testing LLMs' Robustness Under Logical Obfuscation
+
+- **ID:** `2602.01132` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Abhilekh Borah, Shubhra Ghosh, Kedar Joshi, et al.
+- **abs:** https://arxiv.org/abs/2602.01132 · **PDF:** https://arxiv.org/pdf/2602.01132v1
+
+## 16991. Lyapunov Stability-Aware Stackelberg Game for Low-Altitude Economy: A Control-Oriented Pruning-Based DRL Approach
+
+- **ID:** `2602.01131` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Yue Zhong, Jiawen Kang, Yongju Tong, et al.
+- **abs:** https://arxiv.org/abs/2602.01131 · **PDF:** https://arxiv.org/pdf/2602.01131v1
+
+## 16992. Tangent Space Fine-Tuning for Directional Preference Alignment in Large Language Models
+
+- **ID:** `2602.01128` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Mete Erdogan
+- **abs:** https://arxiv.org/abs/2602.01128 · **PDF:** https://arxiv.org/pdf/2602.01128v1
+
+## 16993. WinFLoRA: Incentivizing Client-Adaptive Aggregation in Federated LoRA under Privacy Heterogeneity
+
+- **ID:** `2602.01126` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Mengsha Kou, Xiaoyu Xia, Ziqi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01126 · **PDF:** https://arxiv.org/pdf/2602.01126v1
+
+## 16994. Long-range Modeling and Processing of Multimodal Event Sequences
+
+- **ID:** `2602.01125` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Jichu Li, Yilun Zhong, Zhiting Li, et al.
+- **abs:** https://arxiv.org/abs/2602.01125 · **PDF:** https://arxiv.org/pdf/2602.01125v1
+
+## 16995. ChronoSpike: An Adaptive Spiking Graph Neural Network for Dynamic Graphs
+
+- **ID:** `2602.01124` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Md Abrar Jahin, Taufikur Rahman Fuad, Jay Pujara, et al.
+- **abs:** https://arxiv.org/abs/2602.01124 · **PDF:** https://arxiv.org/pdf/2602.01124v1
+
+## 16996. MarkovScale: Towards Optimal Sequential Scaling at Inference Time
+
+- **ID:** `2602.01120` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Youkang Wang, Jian Wang, Rubing Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.01120 · **PDF:** https://arxiv.org/pdf/2602.01120v1
+
+## 16997. Tendem: A Hybrid AI+Human Platform
+
+- **ID:** `2602.01119` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Konstantin Chernyshev, Ekaterina Artemova, Viacheslav Zhukov, et al.
+- **abs:** https://arxiv.org/abs/2602.01119 · **PDF:** https://arxiv.org/pdf/2602.01119v1
+
+## 16998. Logic-Oriented Retriever Enhancement via Contrastive Learning
+
+- **ID:** `2602.01116` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Wenxuan Zhang, Yuan-Hao Jiang, Changyong Qi, et al.
+- **abs:** https://arxiv.org/abs/2602.01116 · **PDF:** https://arxiv.org/pdf/2602.01116v1
+
+## 16999. Single-Edge Node Injection Threats to GNN-Based Security Monitoring in Industrial Graph Systems
+
+- **ID:** `2602.01113` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Wenjie Liang, Ranhui Yan, Jia Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.01113 · **PDF:** https://arxiv.org/pdf/2602.01113v1
+
+## 17000. Learnable Koopman-Enhanced Transformer-Based Time Series Forecasting with Spectral Control
+
+- **ID:** `2602.02592` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Ali Forootani, Raffaele Iervolino
+- **abs:** https://arxiv.org/abs/2602.02592 · **PDF:** https://arxiv.org/pdf/2602.02592v2
+
+## 17001. Transforming Vehicle Diagnostics: A Multimodal Approach to Error Patterns Prediction
+
+- **ID:** `2602.01109` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Hugo Math, Rainer Lienhart
+- **abs:** https://arxiv.org/abs/2602.01109 · **PDF:** https://arxiv.org/pdf/2602.01109v1
+
+## 17002. SPELL: Synthesis of Programmatic Edits using LLMs
+
+- **ID:** `2602.01107` · **primary:** `cs.SE` · **Submitted:** 2026-02-01
+- **Authors:** Daniel Ramos, Catarina Gamboa, Inês Lynce, et al.
+- **abs:** https://arxiv.org/abs/2602.01107 · **PDF:** https://arxiv.org/pdf/2602.01107v1
+
+## 17003. OLion: Approaching the Hadamard Ideal by Intersecting Spectral and $\ell_{\infty}$ Implicit Biases
+
+- **ID:** `2602.01105` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Zixiao Wang, Yifei Shen, Huishuai Zhang
+- **abs:** https://arxiv.org/abs/2602.01105 · **PDF:** https://arxiv.org/pdf/2602.01105v2
+
+## 17004. Probing RLVR training instability through the lens of objective-level hacking
+
+- **ID:** `2602.01103` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Yiming Dong, Kun Fu, Haoyu Li, et al.
+- **abs:** https://arxiv.org/abs/2602.01103 · **PDF:** https://arxiv.org/pdf/2602.01103v1
+
+## 17005. Hard Constraints Meet Soft Generation: Guaranteed Feasibility for LLM-based Combinatorial Optimization
+
+- **ID:** `2602.01090` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Yang Liu, Chuan Zhou, Yancheng Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.01090 · **PDF:** https://arxiv.org/pdf/2602.01090v1
+
+## 17006. MedBeads: An Agent-Native, Immutable Data Substrate for Trustworthy Medical AI
+
+- **ID:** `2602.01086` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Takahito Nakajima
+- **abs:** https://arxiv.org/abs/2602.01086 · **PDF:** https://arxiv.org/pdf/2602.01086v1
+
+## 17007. On the Expressive Power of Permutation-Equivariant Weight-Space Networks
+
+- **ID:** `2602.01083` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Adir Dayan, Yam Eitan, Haggai Maron
+- **abs:** https://arxiv.org/abs/2602.01083 · **PDF:** https://arxiv.org/pdf/2602.01083v1
+
+## 17008. EvoOpt-LLM: Evolving industrial optimization models with large language models
+
+- **ID:** `2602.01082` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Yiliu He, Tianle Li, Binghao Ji, et al.
+- **abs:** https://arxiv.org/abs/2602.01082 · **PDF:** https://arxiv.org/pdf/2602.01082v2
+
+## 17009. VividVoice: A Unified Framework for Scene-Aware Visually-Driven Speech Synthesis
+
+- **ID:** `2602.02591` · **primary:** `cs.SD` · **Submitted:** 2026-02-01
+- **Authors:** Chengyuan Ma, Jiawei Jin, Ruijie Xiong, et al.
+- **abs:** https://arxiv.org/abs/2602.02591 · **PDF:** https://arxiv.org/pdf/2602.02591v1
+
+## 17010. AutoHealth: An Uncertainty-Aware Multi-Agent System for Autonomous Health Data Modeling
+
+- **ID:** `2602.01078` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Tong Xia, Weibin Li, Gang Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.01078 · **PDF:** https://arxiv.org/pdf/2602.01078v1
+
+## 17011. Decoding Ambiguous Emotions with Test-Time Scaling in Audio-Language Models
+
+- **ID:** `2602.03873` · **primary:** `cs.SD` · **Submitted:** 2026-02-01
+- **Authors:** Hong Jia, Weibin Li, Jingyao Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.03873 · **PDF:** https://arxiv.org/pdf/2602.03873v1
+
+## 17012. ConvexBench: Can LLMs Recognize Convex Functions?
+
+- **ID:** `2602.01075` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Yepeng Liu, Yu Huang, Yu-Xiang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01075 · **PDF:** https://arxiv.org/pdf/2602.01075v2
+
+## 17013. Vortex Stretching in the Navier-Stokes Equations and Information Dissipation in Diffusion Models: A Reformulation from a Partial Differential Equation Viewpoint
+
+- **ID:** `2602.01071` · **primary:** `math.AP` · **Submitted:** 2026-02-01
+- **Authors:** Tsuyoshi Yoneda
+- **abs:** https://arxiv.org/abs/2602.01071 · **PDF:** https://arxiv.org/pdf/2602.01071v1
+
+## 17014. What If We Allocate Test-Time Compute Adaptively?
+
+- **ID:** `2602.01070` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Ahsan Bilal, Ahmed Mohsin, Muhammad Umer, et al.
+- **abs:** https://arxiv.org/abs/2602.01070 · **PDF:** https://arxiv.org/pdf/2602.01070v3
+
+## 17015. PDE-Constrained Optimization for Neural Image Segmentation with Physics Priors
+
+- **ID:** `2602.01069` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Seema K. Poudel, Sunny K. Khadka
+- **abs:** https://arxiv.org/abs/2602.01069 · **PDF:** https://arxiv.org/pdf/2602.01069v1
+
+## 17016. From Utterance to Vividity: Training Expressive Subtitle Translation LLM via Adaptive Local Preference Optimization
+
+- **ID:** `2602.01068` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Chaoqun Cui, Shijing Wang, Liangbin Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.01068 · **PDF:** https://arxiv.org/pdf/2602.01068v1
+
+## 17017. Exploring Knowledge Purification in Multi-Teacher Knowledge Distillation for LLMs
+
+- **ID:** `2602.01064` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Ruihan Jin, Pengpeng Shao, Zhengqi Wen, et al.
+- **abs:** https://arxiv.org/abs/2602.01064 · **PDF:** https://arxiv.org/pdf/2602.01064v1
+
+## 17018. Personality Expression Across Contexts: Linguistic and Behavioral Variation in LLM Agents
+
+- **ID:** `2602.01063` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Bin Han, Deuksin Kwon, Jonathan Gratch
+- **abs:** https://arxiv.org/abs/2602.01063 · **PDF:** https://arxiv.org/pdf/2602.01063v1
+
+## 17019. SetPO: Set-Level Policy Optimization for Diversity-Preserving LLM Reasoning
+
+- **ID:** `2602.01062` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Chenyi Li, Yuan Zhang, Bo Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01062 · **PDF:** https://arxiv.org/pdf/2602.01062v1
+
+## 17020. TLDiffGAN: A Latent Diffusion-GAN Framework with Temporal Information Fusion for Anomalous Sound Detection
+
+- **ID:** `2602.01060` · **primary:** `cs.SD` · **Submitted:** 2026-02-01
+- **Authors:** Chengyuan Ma, Peng Jia, Hongyue Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.01060 · **PDF:** https://arxiv.org/pdf/2602.01060v1
+
+## 17021. Good SFT Optimizes for SFT, Better SFT Prepares for Reinforcement Learning
+
+- **ID:** `2602.01058` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Dylan Zhang, Yufeng Xu, Haojin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01058 · **PDF:** https://arxiv.org/pdf/2602.01058v1
+
+## 17022. The Quantum Learning Menagerie (A survey on Quantum learning for Classical concepts)
+
+- **ID:** `2602.01054` · **primary:** `quant-ph` · **Submitted:** 2026-02-01
+- **Authors:** Sagnik Chatterjee
+- **abs:** https://arxiv.org/abs/2602.01054 · **PDF:** https://arxiv.org/pdf/2602.01054v1
+
+## 17023. LRAgent: Efficient KV Cache Sharing for Multi-LoRA LLM Agents
+
+- **ID:** `2602.01053` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Hyesung Jeon, Hyeongju Ha, Jae-Joon Kim
+- **abs:** https://arxiv.org/abs/2602.01053 · **PDF:** https://arxiv.org/pdf/2602.01053v1
+
+## 17024. SwiftRepertoire: Few-Shot Immune-Signature Synthesis via Dynamic Kernel Codes
+
+- **ID:** `2602.01051` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Rong Fu, Muge Qi, Yang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.01051 · **PDF:** https://arxiv.org/pdf/2602.01051v3
+
+## 17025. Residual Decoding: Mitigating Hallucinations in Large Vision-Language Models via History-Aware Residual Guidance
+
+- **ID:** `2602.01047` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Xinrong Chen, Xu Chu, Yingmin Qiu, et al.
+- **abs:** https://arxiv.org/abs/2602.01047 · **PDF:** https://arxiv.org/pdf/2602.01047v3
+
+## 17026. Superposition unifies power-law training dynamics
+
+- **ID:** `2602.01045` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Zixin Jessie Chen, Hao Chen, Yizhou Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.01045 · **PDF:** https://arxiv.org/pdf/2602.01045v1
+
+## 17027. PeerRank: Autonomous LLM Evaluation Through Web-Grounded, Bias-Controlled Peer Review
+
+- **ID:** `2602.02589` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Yanki Margalit, Erni Avram, Ran Taig, et al.
+- **abs:** https://arxiv.org/abs/2602.02589 · **PDF:** https://arxiv.org/pdf/2602.02589v1
+
+## 17028. Adaptive Dual-Weighting Framework for Federated Learning via Out-of-Distribution Detection
+
+- **ID:** `2602.01039` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Zhiwei Ling, Hailiang Zhao, Chao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.01039 · **PDF:** https://arxiv.org/pdf/2602.01039v1
+
+## 17029. VEQ: Modality-Adaptive Quantization for MoE Vision-Language Models
+
+- **ID:** `2602.01037` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Guangshuo Qin, Zhiteng Li, Zheng Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.01037 · **PDF:** https://arxiv.org/pdf/2602.01037v1
+
+## 17030. Discovering Process-Outcome Credit in Multi-Step LLM Reasoning
+
+- **ID:** `2602.01034` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Xiangwei Wang, Wei Wang, Ken Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.01034 · **PDF:** https://arxiv.org/pdf/2602.01034v1
+
+## 17031. HierCon: Hierarchical Contrastive Attention for Audio Deepfake Detection
+
+- **ID:** `2602.01032` · **primary:** `cs.SD` · **Submitted:** 2026-02-01
+- **Authors:** Zhili Nicholas Liang, Soyeon Caren Han, Qizhou Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01032 · **PDF:** https://arxiv.org/pdf/2602.01032v1
+
+## 17032. HalluHard: A Hard Multi-Turn Hallucination Benchmark
+
+- **ID:** `2602.01031` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Dongyang Fan, Sebastien Delsad, Nicolas Flammarion, et al.
+- **abs:** https://arxiv.org/abs/2602.01031 · **PDF:** https://arxiv.org/pdf/2602.01031v1
+
+## 17033. Bias in the Ear of the Listener: Assessing Sensitivity in Audio Language Models Across Linguistic, Demographic, and Positional Variations
+
+- **ID:** `2602.01030` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Sheng-Lun Wei, Yu-Ling Liao, Yen-Hua Chang, et al.
+- **abs:** https://arxiv.org/abs/2602.01030 · **PDF:** https://arxiv.org/pdf/2602.01030v1
+
+## 17034. SFMP: Fine-Grained, Hardware-Friendly and Search-Free Mixed-Precision Quantization for Large Language Models
+
+- **ID:** `2602.01027` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Xin Nie, Haicheng Zhang, Liang Dong, et al.
+- **abs:** https://arxiv.org/abs/2602.01027 · **PDF:** https://arxiv.org/pdf/2602.01027v1
+
+## 17035. The Stacked Autoencoder Evolution Hypothesis
+
+- **ID:** `2602.01026` · **primary:** `cs.NE` · **Submitted:** 2026-02-01
+- **Authors:** Hiroyuki Iizuka
+- **abs:** https://arxiv.org/abs/2602.01026 · **PDF:** https://arxiv.org/pdf/2602.01026v1
+
+## 17036. Toward Universal and Transferable Jailbreak Attacks on Vision-Language Models
+
+- **ID:** `2602.01025` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Kaiyuan Cui, Yige Li, Yutao Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.01025 · **PDF:** https://arxiv.org/pdf/2602.01025v1
+
+## 17037. Unifying Ranking and Generation in Query Auto-Completion via Retrieval-Augmented Generation and Multi-Objective Alignment
+
+- **ID:** `2602.01023` · **primary:** `cs.IR` · **Submitted:** 2026-02-01
+- **Authors:** Kai Yuan, Anthony Zheng, Jia Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.01023 · **PDF:** https://arxiv.org/pdf/2602.01023v4
+
+## 17038. Calibrating Behavioral Parameters with Large Language Models
+
+- **ID:** `2602.01022` · **primary:** `econ.GN` · **Submitted:** 2026-02-01
+- **Authors:** Brandon Yee, Krishna Sharma
+- **abs:** https://arxiv.org/abs/2602.01022 · **PDF:** https://arxiv.org/pdf/2602.01022v1
+
+## 17039. Inter- and Intra-Subject Variability in EEG: A Systematic Survey
+
+- **ID:** `2602.01019` · **primary:** `q-bio.NC` · **Submitted:** 2026-02-01
+- **Authors:** Xuan-The Tran, Thien-Nhan Vo, Son-Tung Vu, et al.
+- **abs:** https://arxiv.org/abs/2602.01019 · **PDF:** https://arxiv.org/pdf/2602.01019v1
+
+## 17040. Offline Discovery of Interpretable Skills from Multi-Task Trajectories
+
+- **ID:** `2602.01018` · **primary:** `cs.RO` · **Submitted:** 2026-02-01
+- **Authors:** Chongyu Zhu, Mithun Vanniasinghe, Jiayu Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.01018 · **PDF:** https://arxiv.org/pdf/2602.01018v1
+
+## 17041. How Does Unfaithful Reasoning Emerge from Autoregressive Training? A Study of Synthetic Experiments
+
+- **ID:** `2602.01017` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Fuxin Wang, Amr Alazali, Yiqiao Zhong
+- **abs:** https://arxiv.org/abs/2602.01017 · **PDF:** https://arxiv.org/pdf/2602.01017v1
+
+## 17042. Large Language Models as Students Who Think Aloud: Overly Coherent, Verbose, and Confident
+
+- **ID:** `2602.01015` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Conrad Borchers, Jill-Jênn Vie, Roger Azevedo
+- **abs:** https://arxiv.org/abs/2602.01015 · **PDF:** https://arxiv.org/pdf/2602.01015v1
+
+## 17043. Multi-Agent Teams Hold Experts Back
+
+- **ID:** `2602.01011` · **primary:** `cs.MA` · **Submitted:** 2026-02-01
+- **Authors:** Aneesh Pappu, Batu El, Hancheng Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.01011 · **PDF:** https://arxiv.org/pdf/2602.01011v3
+
+## 17044. LASS-ODE: Scaling ODE Computations to Connect Foundation Models with Dynamical Physical Systems
+
+- **ID:** `2602.01009` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Haoran Li, Chenhan Xiao, Lihao Mai, et al.
+- **abs:** https://arxiv.org/abs/2602.01009 · **PDF:** https://arxiv.org/pdf/2602.01009v2
+
+## 17045. Adapting Where It Matters: Depth-Aware Adaptation for Efficient Multilingual Speech Recognition in Low-Resource Languages
+
+- **ID:** `2602.01008` · **primary:** `eess.AS` · **Submitted:** 2026-02-01
+- **Authors:** Yang Xiao, Eun-Jung Holden, Ting Dang
+- **abs:** https://arxiv.org/abs/2602.01008 · **PDF:** https://arxiv.org/pdf/2602.01008v1
+
+## 17046. Distilling Token-Trained Models into Byte-Level Models
+
+- **ID:** `2602.01007` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Zishuo Bao, Jiaqi Leng, Junxiong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.01007 · **PDF:** https://arxiv.org/pdf/2602.01007v1
+
+## 17047. Predicting Anemia Among Under-Five Children in Nepal Using Machine Learning and Deep Learning
+
+- **ID:** `2602.01005` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Deepak Bastola, Pitambar Acharya, Dipak Dulal, et al.
+- **abs:** https://arxiv.org/abs/2602.01005 · **PDF:** https://arxiv.org/pdf/2602.01005v1
+
+## 17048. ESSAM: A Novel Competitive Evolution Strategies Approach to Reinforcement Learning for Memory Efficient LLMs Fine-Tuning
+
+- **ID:** `2602.01003` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Zhishen Sun, Sizhe Dang, Guang Dai, et al.
+- **abs:** https://arxiv.org/abs/2602.01003 · **PDF:** https://arxiv.org/pdf/2602.01003v1
+
+## 17049. How RLHF Amplifies Sycophancy
+
+- **ID:** `2602.01002` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Itai Shapira, Gerdus Benade, Ariel D. Procaccia
+- **abs:** https://arxiv.org/abs/2602.01002 · **PDF:** https://arxiv.org/pdf/2602.01002v1
+
+## 17050. CortiNet: A Physics-Perception Hybrid Cortical-Inspired Dual-Stream Network for Gallbladder Disease Diagnosis from Ultrasound
+
+- **ID:** `2602.01000` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Vagish Kumar, Souvik Chakraborty
+- **abs:** https://arxiv.org/abs/2602.01000 · **PDF:** https://arxiv.org/pdf/2602.01000v1
+
+## 17051. Reliable Use of Lemmas via Eligibility Reasoning and Section$-$Aware Reinforcement Learning
+
+- **ID:** `2602.00998` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Zhikun Xu, Xiaodong Yu, Ben Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.00998 · **PDF:** https://arxiv.org/pdf/2602.00998v1
+
+## 17052. Error Taxonomy-Guided Prompt Optimization
+
+- **ID:** `2602.00997` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Mayank Singh, Vikas Yadav, Eduardo Blanco
+- **abs:** https://arxiv.org/abs/2602.00997 · **PDF:** https://arxiv.org/pdf/2602.00997v1
+
+## 17053. DeALOG: Decentralized Multi-Agents Log-Mediated Reasoning Framework
+
+- **ID:** `2602.00996` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Abhijit Chakraborty, Ashish Raj Shekhar, Shiven Agarwal, et al.
+- **abs:** https://arxiv.org/abs/2602.00996 · **PDF:** https://arxiv.org/pdf/2602.00996v1
+
+## 17054. Reasoning and Tool-use Compete in Agentic RL:From Quantifying Interference to Disentangled Tuning
+
+- **ID:** `2602.00994` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Yu Li, Mingyang Yi, Xiuyu Li, et al.
+- **abs:** https://arxiv.org/abs/2602.00994 · **PDF:** https://arxiv.org/pdf/2602.00994v1
+
+## 17055. HERMES: A Holistic End-to-End Risk-Aware Multimodal Embodied System with Vision-Language Models for Long-Tail Autonomous Driving
+
+- **ID:** `2602.00993` · **primary:** `cs.RO` · **Submitted:** 2026-02-01
+- **Authors:** Weizhe Tang, Junwei You, Jiaxi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.00993 · **PDF:** https://arxiv.org/pdf/2602.00993v1
+
+## 17056. Optimal Decision-Making Based on Prediction Sets
+
+- **ID:** `2602.00989` · **primary:** `stat.ML` · **Submitted:** 2026-02-01
+- **Authors:** Tao Wang, Edgar Dobriban
+- **abs:** https://arxiv.org/abs/2602.00989 · **PDF:** https://arxiv.org/pdf/2602.00989v3
+
+## 17057. Scalable Random Wavelet Features: Efficient Non-Stationary Kernel Approximation with Convergence Guarantees
+
+- **ID:** `2602.00987` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Sawan Kumar, Souvik Chakraborty
+- **abs:** https://arxiv.org/abs/2602.00987 · **PDF:** https://arxiv.org/pdf/2602.00987v1
+
+## 17058. Sparse Reward Subsystem in Large Language Models
+
+- **ID:** `2602.00986` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Guowei Xu, Mert Yuksekgonul, James Zou
+- **abs:** https://arxiv.org/abs/2602.00986 · **PDF:** https://arxiv.org/pdf/2602.00986v1
+
+## 17059. Understanding the Impact of Differentially Private Training on Memorization of Long-Tailed Data
+
+- **ID:** `2602.03872` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Jiaming Zhang, Huanyi Xie, Meng Ding, et al.
+- **abs:** https://arxiv.org/abs/2602.03872 · **PDF:** https://arxiv.org/pdf/2602.03872v1
+
+## 17060. DISPO: Enhancing Training Efficiency and Stability in Reinforcement Learning for Large Language Model Mathematical Reasoning
+
+- **ID:** `2602.00983` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Batuhan K. Karaman, Aditya Rawal, Suhaila Shakiah, et al.
+- **abs:** https://arxiv.org/abs/2602.00983 · **PDF:** https://arxiv.org/pdf/2602.00983v1
+
+## 17061. Navigating Simply, Aligning Deeply: Winning Solutions for Mouse vs. AI 2025
+
+- **ID:** `2602.00982` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Phu-Hoa Pham, Chi-Nguyen Tran, Dao Sy Duy Minh, et al.
+- **abs:** https://arxiv.org/abs/2602.00982 · **PDF:** https://arxiv.org/pdf/2602.00982v1
+
+## 17062. MedSpeak: A Knowledge Graph-Aided ASR Error Correction Framework for Spoken Medical QA
+
+- **ID:** `2602.00981` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Yutong Song, Shiva Shrestha, Chenhan Lyu, et al.
+- **abs:** https://arxiv.org/abs/2602.00981 · **PDF:** https://arxiv.org/pdf/2602.00981v1
+
+## 17063. GradingAttack: Attacking Large Language Models Towards Short Answer Grading Ability
+
+- **ID:** `2602.00979` · **primary:** `cs.CR` · **Submitted:** 2026-02-01
+- **Authors:** Xueyi Li, Zhuoneng Zhou, Zitao Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.00979 · **PDF:** https://arxiv.org/pdf/2602.00979v1
+
+## 17064. Organismal Agency and Rapid Adaptation: The Phenopoiesis Algorithm for Phenotype-First Evolution
+
+- **ID:** `2602.00978` · **primary:** `cs.NE` · **Submitted:** 2026-02-01
+- **Authors:** Nam H. Le
+- **abs:** https://arxiv.org/abs/2602.00978 · **PDF:** https://arxiv.org/pdf/2602.00978v1
+
+## 17065. Trust in One Round: Confidence Estimation for Large Language Models via Structural Signals
+
+- **ID:** `2602.00977` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Pengyue Yang, Jiawen Wen, Haolin Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.00977 · **PDF:** https://arxiv.org/pdf/2602.00977v1
+
+## 17066. Forest-Guided Semantic Transport for Label-Supervised Manifold Alignment
+
+- **ID:** `2602.00974` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Adrien Aumon, Myriam Lizotte, Guy Wolf, et al.
+- **abs:** https://arxiv.org/abs/2602.00974 · **PDF:** https://arxiv.org/pdf/2602.00974v1
+
+## 17067. Verification Required: The Impact of Information Credibility on AI Persuasion
+
+- **ID:** `2602.00970` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Saaduddin Mahmud, Eugene Bagdasarian, Shlomo Zilberstein
+- **abs:** https://arxiv.org/abs/2602.00970 · **PDF:** https://arxiv.org/pdf/2602.00970v1
+
+## 17068. On the Spectral Flattening of Quantized Embeddings
+
+- **ID:** `2602.00969` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Junlin Huang, Wenyi Fang, Zhenheng Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.00969 · **PDF:** https://arxiv.org/pdf/2602.00969v1
+
+## 17069. Multimodal Scientific Learning Beyond Diffusions and Flows
+
+- **ID:** `2602.00960` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Leonardo Ferreira Guilhoto, Akshat Kaushal, Paris Perdikaris
+- **abs:** https://arxiv.org/abs/2602.00960 · **PDF:** https://arxiv.org/pdf/2602.00960v1
+
+## 17070. Probing the Knowledge Boundary: An Interactive Agentic Framework for Deep Knowledge Extraction
+
+- **ID:** `2602.00959` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Yuheng Yang, Siqi Zhu, Tao Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.00959 · **PDF:** https://arxiv.org/pdf/2602.00959v1
+
+## 17071. From drift to adaptation to the failed ml model: Transfer Learning in Industrial MLOps
+
+- **ID:** `2602.00957` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Waqar Muhammad Ashraf, Talha Ansar, Fahad Ahmed, et al.
+- **abs:** https://arxiv.org/abs/2602.00957 · **PDF:** https://arxiv.org/pdf/2602.00957v1
+
+## 17072. Hybrid Topological and Deep Feature Fusion for Accurate MRI-Based Alzheimer's Disease Severity Classification
+
+- **ID:** `2602.00956` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Faisal Ahmed
+- **abs:** https://arxiv.org/abs/2602.00956 · **PDF:** https://arxiv.org/pdf/2602.00956v1
+
+## 17073. Small-Margin Preferences Still Matter-If You Train Them Right
+
+- **ID:** `2602.00954` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Jinlong Pang, Zhaowei Zhu, Na Di, et al.
+- **abs:** https://arxiv.org/abs/2602.00954 · **PDF:** https://arxiv.org/pdf/2602.00954v1
+
+## 17074. SAGE: Agentic Framework for Interpretable and Clinically Translatable Computational Pathology Biomarker Discovery
+
+- **ID:** `2602.00953` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Sahar Almahfouz Nasser, Juan Francisco Pesantez Borja, Jincheng Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.00953 · **PDF:** https://arxiv.org/pdf/2602.00953v1
+
+## 17075. Optimal Budgeted Adaptation of Large Language Models
+
+- **ID:** `2602.00952` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Jing Wang, Jie Shen, Dean Foster, et al.
+- **abs:** https://arxiv.org/abs/2602.00952 · **PDF:** https://arxiv.org/pdf/2602.00952v1
+
+## 17076. R-HTN: Rebellious Online HTN Planning for Safety and Game AI
+
+- **ID:** `2602.00951` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Hector Munoz-Avila, David W. Aha, Paola Rizzo
+- **abs:** https://arxiv.org/abs/2602.00951 · **PDF:** https://arxiv.org/pdf/2602.00951v1
+
+## 17077. MindGuard: Guardrail Classifiers for Multi-Turn Mental Health Support
+
+- **ID:** `2602.00950` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** António Farinhas, Nuno M. Guerreiro, José Pombal, et al.
+- **abs:** https://arxiv.org/abs/2602.00950 · **PDF:** https://arxiv.org/pdf/2602.00950v1
+
+## 17078. FinEvo: From Isolated Backtests to Ecological Market Games for Multi-Agent Financial Strategy Evolution
+
+- **ID:** `2602.00948` · **primary:** `physics.soc-ph` · **Submitted:** 2026-02-01
+- **Authors:** Mingxi Zou, Jiaxiang Chen, Aotian Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.00948 · **PDF:** https://arxiv.org/pdf/2602.00948v1
+
+## 17079. The Keyhole Effect: Why Chat Interfaces Fail at Data Analysis
+
+- **ID:** `2602.00947` · **primary:** `cs.AI` · **Submitted:** 2026-02-01
+- **Authors:** Mohan Reddy
+- **abs:** https://arxiv.org/abs/2602.00947 · **PDF:** https://arxiv.org/pdf/2602.00947v1
+
+## 17080. XAI-CLIP: ROI-Guided Perturbation Framework for Explainable Medical Image Segmentation in Multimodal Vision-Language Models
+
+- **ID:** `2602.07017` · **primary:** `cs.CV` · **Submitted:** 2026-02-01
+- **Authors:** Thuraya Alzubaidi, Sana Ammar, Maryam Alsharqi, et al.
+- **abs:** https://arxiv.org/abs/2602.07017 · **PDF:** https://arxiv.org/pdf/2602.07017v1
+
+## 17081. Neural FOXP2 -- Language Specific Neuron Steering for Targeted Language Improvement in LLMs
+
+- **ID:** `2602.00945` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Anusa Saha, Tanmay Joshi, Vinija Jain, et al.
+- **abs:** https://arxiv.org/abs/2602.00945 · **PDF:** https://arxiv.org/pdf/2602.00945v1
+
+## 17082. Asymptotic Semantic Collapse in Hierarchical Optimization
+
+- **ID:** `2602.18450` · **primary:** `cs.CL` · **Submitted:** 2026-02-01
+- **Authors:** Faruk Alpay, Bugra Kilictas
+- **abs:** https://arxiv.org/abs/2602.18450 · **PDF:** https://arxiv.org/pdf/2602.18450v1
+
+## 17083. Dynamic Prior Thompson Sampling for Cold-Start Exploration in Recommender Systems
+
+- **ID:** `2602.00943` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Zhenyu Zhao, David Zhang, Ellie Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.00943 · **PDF:** https://arxiv.org/pdf/2602.00943v1
+
+## 17084. SALAAD: Sparse And Low-Rank Adaptation via ADMM for Large Language Model Inference
+
+- **ID:** `2602.00942` · **primary:** `cs.LG` · **Submitted:** 2026-02-01
+- **Authors:** Hao Ma, Melis Ilayda Bal, Liang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.00942 · **PDF:** https://arxiv.org/pdf/2602.00942v2
+
+## 17085. Improving Minimax Estimation Rates for Contaminated Mixture of Multinomial Logistic Experts via Expert Heterogeneity
+
+- **ID:** `2602.00939` · **primary:** `math.ST` · **Submitted:** 2026-01-31
+- **Authors:** Fanqi Yan, Dung Le, Trang Pham, et al.
+- **abs:** https://arxiv.org/abs/2602.00939 · **PDF:** https://arxiv.org/pdf/2602.00939v1
+
+## 17086. CLAMP: Contrastive Learning for 3D Multi-View Action-Conditioned Robotic Manipulation Pretraining
+
+- **ID:** `2602.00937` · **primary:** `cs.RO` · **Submitted:** 2026-01-31
+- **Authors:** I-Chun Arthur Liu, Krzysztof Choromanski, Sandy Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.00937 · **PDF:** https://arxiv.org/pdf/2602.00937v1
+
+## 17087. MCP-Atlas: A Large-Scale Benchmark for Tool-Use Competency with Real MCP Servers
+
+- **ID:** `2602.00933` · **primary:** `cs.SE` · **Submitted:** 2026-01-31
+- **Authors:** Chaithanya Bandi, Ben Hertzberg, Geobio Boo, et al.
+- **abs:** https://arxiv.org/abs/2602.00933 · **PDF:** https://arxiv.org/pdf/2602.00933v1
+
+## 17088. Continuous-Utility Direct Preference Optimization
+
+- **ID:** `2602.00931` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Muhammad Ahmed Mohsin, Muhammad Umer, Ahsan Bilal, et al.
+- **abs:** https://arxiv.org/abs/2602.00931 · **PDF:** https://arxiv.org/pdf/2602.00931v1
+
+## 17089. Learning Abstractions for Hierarchical Planning in Program-Synthesis Agents
+
+- **ID:** `2602.00929` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Zergham Ahmed, Kazuki Irie, Joshua B. Tenenbaum, et al.
+- **abs:** https://arxiv.org/abs/2602.00929 · **PDF:** https://arxiv.org/pdf/2602.00929v1
+
+## 17090. Beyond What Seems Necessary: Hidden Gains from Scaling Training-Time Reasoning Length under Outcome Supervision
+
+- **ID:** `2602.00927` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Yihao Xue, Allan Zhang, Jianhao Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.00927 · **PDF:** https://arxiv.org/pdf/2602.00927v1
+
+## 17091. Supervised sparse auto-encoders as unconstrained feature models for semantic composition
+
+- **ID:** `2602.00924` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Ouns El Harzli, Hugo Wallner, Yoonsoo Nam, et al.
+- **abs:** https://arxiv.org/abs/2602.00924 · **PDF:** https://arxiv.org/pdf/2602.00924v1
+
+## 17092. On the Convergence of Jacobian-Free Backpropagation for Optimal Control Problems with Implicit Hamiltonians
+
+- **ID:** `2602.00921` · **primary:** `math.OC` · **Submitted:** 2026-01-31
+- **Authors:** Eric Gelphman, Deepanshu Verma, Nicole Tianjiao Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.00921 · **PDF:** https://arxiv.org/pdf/2602.00921v1
+
+## 17093. Early Classification of Time Series in Non-Stationary Cost Regimes
+
+- **ID:** `2602.00918` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Aurélien Renault, Alexis Bondu, Antoine Cornuéjols, et al.
+- **abs:** https://arxiv.org/abs/2602.00918 · **PDF:** https://arxiv.org/pdf/2602.00918v1
+
+## 17094. A Baseline Multimodal Approach to Emotion Recognition in Conversations
+
+- **ID:** `2602.00914` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Víctor Yeste, Rodrigo Rivas-Arévalo
+- **abs:** https://arxiv.org/abs/2602.00914 · **PDF:** https://arxiv.org/pdf/2602.00914v1
+
+## 17095. Do Schwartz Higher-Order Values Help Sentence-Level Human Value Detection? A Study of Hierarchical Gating and Calibration
+
+- **ID:** `2602.00913` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Víctor Yeste, Paolo Rosso
+- **abs:** https://arxiv.org/abs/2602.00913 · **PDF:** https://arxiv.org/pdf/2602.00913v2
+
+## 17096. Synapse Compendium Aware Federated Knowledge Exchange for Tool Routed LLMs
+
+- **ID:** `2602.00911` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Abhijit Chakraborty, Sandipan De, Yash Shah, et al.
+- **abs:** https://arxiv.org/abs/2602.00911 · **PDF:** https://arxiv.org/pdf/2602.00911v1
+
+## 17097. Efficient Deep Learning for Medical Imaging: Bridging the Gap Between High-Performance AI and Clinical Deployment
+
+- **ID:** `2602.00910` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Cuong Manh Nguyen, Truong-Son Hy
+- **abs:** https://arxiv.org/abs/2602.00910 · **PDF:** https://arxiv.org/pdf/2602.00910v1
+
+## 17098. PyGALAX: An Open-Source Python Toolkit for Advanced Explainable Geospatial Machine Learning
+
+- **ID:** `2602.00907` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Pingping Wang, Yihong Yuan, Lingcheng Li, et al.
+- **abs:** https://arxiv.org/abs/2602.00907 · **PDF:** https://arxiv.org/pdf/2602.00907v1
+
+## 17099. Hallucination is a Consequence of Space-Optimality: A Rate-Distortion Theorem for Membership Testing
+
+- **ID:** `2602.00906` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Anxin Guo, Jingwei Li
+- **abs:** https://arxiv.org/abs/2602.00906 · **PDF:** https://arxiv.org/pdf/2602.00906v5
+
+## 17100. Domain-Adaptive and Scalable Dense Retrieval for Content-Based Recommendation
+
+- **ID:** `2602.00899` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Mritunjay Pandey
+- **abs:** https://arxiv.org/abs/2602.00899 · **PDF:** https://arxiv.org/pdf/2602.00899v1
+
+## 17101. GAPNet: Plug-in Jointly Learning Task-Specific Graph for Dynamic Stock Relation
+
+- **ID:** `2602.00888` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Yingjie Niu, Lanxin Lu, Changhong Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.00888 · **PDF:** https://arxiv.org/pdf/2602.00888v1
+
+## 17102. EffGen: Enabling Small Language Models as Capable Autonomous Agents
+
+- **ID:** `2602.00887` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Gaurav Srivastava, Aafiya Hussain, Chi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00887 · **PDF:** https://arxiv.org/pdf/2602.00887v1
+
+## 17103. Agentic Observability: Automated Alert Triage for Adobe E-Commerce
+
+- **ID:** `2602.02585` · **primary:** `cs.SE` · **Submitted:** 2026-01-31
+- **Authors:** Aprameya Bharadwaj, Kyle Tu
+- **abs:** https://arxiv.org/abs/2602.02585 · **PDF:** https://arxiv.org/pdf/2602.02585v1
+
+## 17104. RoDiF: Robust Direct Fine-Tuning of Diffusion Policies with Corrupted Human Feedback
+
+- **ID:** `2602.00886` · **primary:** `cs.RO` · **Submitted:** 2026-01-31
+- **Authors:** Amitesh Vatsa, Zhixian Xie, Wanxin Jin
+- **abs:** https://arxiv.org/abs/2602.00886 · **PDF:** https://arxiv.org/pdf/2602.00886v1
+
+## 17105. Reliability-Aware Determinantal Point Processes for Robust Informative Data Selection in Large Language Models
+
+- **ID:** `2602.00885` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Ahmad Sarlak, Abolfazl Razi
+- **abs:** https://arxiv.org/abs/2602.00885 · **PDF:** https://arxiv.org/pdf/2602.00885v1
+
+## 17106. Test-time Generalization for Physics through Neural Operator Splitting
+
+- **ID:** `2602.00884` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Louis Serrano, Jiequn Han, Edouard Oyallon, et al.
+- **abs:** https://arxiv.org/abs/2602.00884 · **PDF:** https://arxiv.org/pdf/2602.00884v1
+
+## 17107. DIAMOND: Directed Inference for Artifact Mitigation in Flow Matching Models
+
+- **ID:** `2602.00883` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Alicja Polowczyk, Agnieszka Polowczyk, Piotr Borycki, et al.
+- **abs:** https://arxiv.org/abs/2602.00883 · **PDF:** https://arxiv.org/pdf/2602.00883v1
+
+## 17108. ILSIC: Corpora for Identifying Indian Legal Statutes from Queries by Laypeople
+
+- **ID:** `2602.00881` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Shounak Paul, Raghav Dogra, Pawan Goyal, et al.
+- **abs:** https://arxiv.org/abs/2602.00881 · **PDF:** https://arxiv.org/pdf/2602.00881v1
+
+## 17109. Dynamic Expert Sharing: Decoupling Memory from Parallelism in Mixture-of-Experts Diffusion LLMs
+
+- **ID:** `2602.00879` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Hao Mark Chen, Zhiwen Mo, Royson Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.00879 · **PDF:** https://arxiv.org/pdf/2602.00879v1
+
+## 17110. Sublinear Time Quantum Algorithm for Attention Approximation
+
+- **ID:** `2602.00874` · **primary:** `quant-ph` · **Submitted:** 2026-01-31
+- **Authors:** Zhao Song, Jianfei Xue, Jiahao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.00874 · **PDF:** https://arxiv.org/pdf/2602.00874v1
+
+## 17111. Learning Heat-based Equations in Self-similar variables
+
+- **ID:** `2602.00872` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Shihao Wang, Qipeng Qian, Jingquan Wang
+- **abs:** https://arxiv.org/abs/2602.00872 · **PDF:** https://arxiv.org/pdf/2602.00872v2
+
+## 17112. Beyond Output Critique: Self-Correction via Task Distillation
+
+- **ID:** `2602.00871` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Hossein A. Rahmani, Mengting Wan, Pei Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.00871 · **PDF:** https://arxiv.org/pdf/2602.00871v1
+
+## 17113. Constitutional Spec-Driven Development: Enforcing Security by Construction in AI-Assisted Code Generation
+
+- **ID:** `2602.02584` · **primary:** `cs.SE` · **Submitted:** 2026-01-31
+- **Authors:** Srinivas Rao Marri
+- **abs:** https://arxiv.org/abs/2602.02584 · **PDF:** https://arxiv.org/pdf/2602.02584v1
+
+## 17114. Improving Flow Matching by Aligning Flow Divergence
+
+- **ID:** `2602.00869` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Yuhao Huang, Taos Transue, Shih-Hsin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00869 · **PDF:** https://arxiv.org/pdf/2602.00869v1
+
+## 17115. Foundation CAN LM: A Pretrained Language Model For Automotive CAN Data
+
+- **ID:** `2602.00866` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Akiharu Esashi, Pawissanutt Lertpongrujikorn, Justin Makino, et al.
+- **abs:** https://arxiv.org/abs/2602.00866 · **PDF:** https://arxiv.org/pdf/2602.00866v1
+
+## 17116. Towards Multiscale Graph-based Protein Learning with Geometric Secondary Structural Motifs
+
+- **ID:** `2602.00862` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Shih-Hsin Wang, Yuhao Huang, Taos Transue, et al.
+- **abs:** https://arxiv.org/abs/2602.00862 · **PDF:** https://arxiv.org/pdf/2602.00862v1
+
+## 17117. Multi-Head Attention Is a Multi-Player Game
+
+- **ID:** `2602.00861` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Kushal Chakrabarti, Nirmal Balachundar
+- **abs:** https://arxiv.org/abs/2602.00861 · **PDF:** https://arxiv.org/pdf/2602.00861v1
+
+## 17118. Unifying Adversarial Robustness and Training Across Text Scoring Models
+
+- **ID:** `2602.00857` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Manveer Singh Tamber, Hosna Oyarhoseini, Jimmy Lin
+- **abs:** https://arxiv.org/abs/2602.00857 · **PDF:** https://arxiv.org/pdf/2602.00857v1
+
+## 17119. Position: Human-Centric AI Requires a Minimum Viable Level of Human Understanding
+
+- **ID:** `2602.00854` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Fangzhou Lin, Qianwen Ge, Lingyu Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.00854 · **PDF:** https://arxiv.org/pdf/2602.00854v1
+
+## 17120. Investigating the Robustness of Subtask Distillation under Spurious Correlation
+
+- **ID:** `2602.00852` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Pattarawat Chormai, Klaus-Robert Müller, Grégoire Montavon
+- **abs:** https://arxiv.org/abs/2602.00852 · **PDF:** https://arxiv.org/pdf/2602.00852v1
+
+## 17121. Persuasion Propagation in LLM Agents
+
+- **ID:** `2602.00851` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Hyejun Jeong, Amir Houmansadr, Shlomo Zilberstein, et al.
+- **abs:** https://arxiv.org/abs/2602.00851 · **PDF:** https://arxiv.org/pdf/2602.00851v2
+
+## 17122. RMFlow: Refined Mean Flow by a Noise-Injection Step for Multimodal Generation
+
+- **ID:** `2602.00849` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Yuhao Huang, Shih-Hsin Wang, Andrea L. Bertozzi, et al.
+- **abs:** https://arxiv.org/abs/2602.00849 · **PDF:** https://arxiv.org/pdf/2602.00849v1
+
+## 17123. Factuality on Demand: Controlling the Factuality-Informativeness Trade-off in Text Generation
+
+- **ID:** `2602.00848` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Ziwei Gong, Yanda Chen, Julia Hirschberg, et al.
+- **abs:** https://arxiv.org/abs/2602.00848 · **PDF:** https://arxiv.org/pdf/2602.00848v1
+
+## 17124. Omni-RRM: Advancing Omni Reward Modeling via Automatic Rubric-Grounded Preference Synthesis
+
+- **ID:** `2602.00846` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Zicheng Kong, Dehua Ma, Zhenbo Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.00846 · **PDF:** https://arxiv.org/pdf/2602.00846v1
+
+## 17125. Optimizing Agentic Reasoning with Retrieval via Synthetic Semantic Information Gain Reward
+
+- **ID:** `2602.00845` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Senkang Hu, Yong Dai, Yuzhi Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.00845 · **PDF:** https://arxiv.org/pdf/2602.00845v2
+
+## 17126. Multivariate Time Series Data Imputation via Distributionally Robust Regularization
+
+- **ID:** `2602.00844` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Che-Yi Liao, Zheng Dong, Gian-Gabriel Garcia, et al.
+- **abs:** https://arxiv.org/abs/2602.00844 · **PDF:** https://arxiv.org/pdf/2602.00844v1
+
+## 17127. NegaBent, No Regrets: Evolving Spectrally Flat Boolean Functions
+
+- **ID:** `2602.00843` · **primary:** `cs.NE` · **Submitted:** 2026-01-31
+- **Authors:** Claude Carlet, Marko Ðurasevic, Ermes Franch, et al.
+- **abs:** https://arxiv.org/abs/2602.00843 · **PDF:** https://arxiv.org/pdf/2602.00843v1
+
+## 17128. Exploration of Unary Arithmetic-Based Matrix Multiply Units for Low Precision DL Accelerators
+
+- **ID:** `2602.00838` · **primary:** `cs.AR` · **Submitted:** 2026-01-31
+- **Authors:** Prabhu Vellaisamy, Harideep Nair, Di Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.00838 · **PDF:** https://arxiv.org/pdf/2602.00838v1
+
+## 17129. IDEM Enough? Evolving Highly Nonlinear Idempotent Boolean Functions
+
+- **ID:** `2602.00837` · **primary:** `cs.CR` · **Submitted:** 2026-01-31
+- **Authors:** Claude Carlet, Marko Ðurasevic, Domagoj Jakobovic, et al.
+- **abs:** https://arxiv.org/abs/2602.00837 · **PDF:** https://arxiv.org/pdf/2602.00837v1
+
+## 17130. Score-based Metropolis-Hastings for Fractional Langevin Algorithms
+
+- **ID:** `2602.00835` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Ahmed Aloui, Junyi Liao, Ali Hasan, et al.
+- **abs:** https://arxiv.org/abs/2602.00835 · **PDF:** https://arxiv.org/pdf/2602.00835v1
+
+## 17131. A Minimum Variance Path Principle for Accurate and Stable Score-Based Density Ratio Estimation
+
+- **ID:** `2602.00834` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Wei Chen, Jiacheng Li, Shigui Li, et al.
+- **abs:** https://arxiv.org/abs/2602.00834 · **PDF:** https://arxiv.org/pdf/2602.00834v3
+
+## 17132. Copula-Based Aggregation and Context-Aware Conformal Prediction for Reliable Renewable Energy Forecasting
+
+- **ID:** `2602.02583` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Alireza Moradi, Mathieu Tanneau, Reza Zandehshahvar, et al.
+- **abs:** https://arxiv.org/abs/2602.02583 · **PDF:** https://arxiv.org/pdf/2602.02583v1
+
+## 17133. Over-Alignment vs Over-Fitting: The Role of Feature Learning Strength in Generalization
+
+- **ID:** `2602.00827` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Taesun Yeom, Taehyeok Ha, Jaeho Lee
+- **abs:** https://arxiv.org/abs/2602.00827 · **PDF:** https://arxiv.org/pdf/2602.00827v1
+
+## 17134. Harmful Overfitting in Sobolev Spaces
+
+- **ID:** `2602.00825` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Kedar Karhadkar, Alexander Sietsema, Deanna Needell, et al.
+- **abs:** https://arxiv.org/abs/2602.00825 · **PDF:** https://arxiv.org/pdf/2602.00825v1
+
+## 17135. Deep networks learn to parse uniform-depth context-free languages from local statistics
+
+- **ID:** `2602.06065` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Jack T. Parley, Francesco Cagnetta, Matthieu Wyart
+- **abs:** https://arxiv.org/abs/2602.06065 · **PDF:** https://arxiv.org/pdf/2602.06065v2
+
+## 17136. Safety-Efficacy Trade Off: Robustness against Data-Poisoning
+
+- **ID:** `2602.00822` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Diego Granziol
+- **abs:** https://arxiv.org/abs/2602.00822 · **PDF:** https://arxiv.org/pdf/2602.00822v1
+
+## 17137. Uncertainty and Fairness Awareness in LLM-Based Recommendation Systems
+
+- **ID:** `2602.02582` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Chandan Kumar Sah, Xiaoli Lian, Li Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.02582 · **PDF:** https://arxiv.org/pdf/2602.02582v1
+
+## 17138. Hessian Spectral Analysis at Foundation Model Scale
+
+- **ID:** `2602.00816` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Diego Granziol, Khurshid Juarev
+- **abs:** https://arxiv.org/abs/2602.00816 · **PDF:** https://arxiv.org/pdf/2602.00816v1
+
+## 17139. Resource-Efficient Reinforcement for Reasoning Large Language Models via Dynamic One-Shot Policy Refinement
+
+- **ID:** `2602.00815` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Yunjian Zhang, Sudong Wang, Yang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.00815 · **PDF:** https://arxiv.org/pdf/2602.00815v1
+
+## 17140. MissMAC-Bench: Building Solid Benchmark for Missing Modality Issue in Robust Multimodal Affective Computing
+
+- **ID:** `2602.00811` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Ronghao Lin, Honghao Lu, Ruixing Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.00811 · **PDF:** https://arxiv.org/pdf/2602.00811v1
+
+## 17141. Mobile Exergames: Activity Recognition Based on Smartphone Sensors
+
+- **ID:** `2602.00809` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** David Craveiro, Hugo Silva
+- **abs:** https://arxiv.org/abs/2602.00809 · **PDF:** https://arxiv.org/pdf/2602.00809v1
+
+## 17142. QuantLRM: Quantization of Large Reasoning Models via Fine-Tuning Signals
+
+- **ID:** `2602.02581` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Nan Zhang, Eugene Kwek, Yusen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.02581 · **PDF:** https://arxiv.org/pdf/2602.02581v1
+
+## 17143. JTok: On Token Embedding as another Axis of Scaling Law via Joint Token Self-modulation
+
+- **ID:** `2602.00800` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Yebin Yang, Huaijin Wu, Fu Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.00800 · **PDF:** https://arxiv.org/pdf/2602.00800v1
+
+## 17144. Zero-Flow Encoders
+
+- **ID:** `2602.00797` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Yakun Wang, Leyang Wang, Song Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.00797 · **PDF:** https://arxiv.org/pdf/2602.00797v1
+
+## 17145. SpeechLess: Micro-utterance with Personalized Spatial Memory-aware Assistant in Everyday Augmented Reality
+
+- **ID:** `2602.00793` · **primary:** `cs.HC` · **Submitted:** 2026-01-31
+- **Authors:** Yoonsang Kim, Devshree Jadeja, Divyansh Pradhan, et al.
+- **abs:** https://arxiv.org/abs/2602.00793 · **PDF:** https://arxiv.org/pdf/2602.00793v1
+
+## 17146. Latent Shadows: The Gaussian-Discrete Duality in Masked Diffusion
+
+- **ID:** `2602.00792` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Guinan Chen, Xunpeng Huang, Ying Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.00792 · **PDF:** https://arxiv.org/pdf/2602.00792v1
+
+## 17147. Sporadic Gradient Tracking over Directed Graphs: A Theoretical Perspective on Decentralized Federated Learning
+
+- **ID:** `2602.00791` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Shahryar Zehtabi, Dong-Jun Han, Seyyedali Hosseinalipour, et al.
+- **abs:** https://arxiv.org/abs/2602.00791 · **PDF:** https://arxiv.org/pdf/2602.00791v1
+
+## 17148. Multi-Objective Multi-Fidelity Bayesian Optimization with Causal Priors
+
+- **ID:** `2602.00788` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Md Abir Hossen, Mohammad Ali Javidian, Vignesh Narayanan, et al.
+- **abs:** https://arxiv.org/abs/2602.00788 · **PDF:** https://arxiv.org/pdf/2602.00788v1
+
+## 17149. World Models as an Intermediary between Agents and the Real World
+
+- **ID:** `2602.00785` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Sherry Yang
+- **abs:** https://arxiv.org/abs/2602.00785 · **PDF:** https://arxiv.org/pdf/2602.00785v1
+
+## 17150. Controlling Repetition in Protein Language Models
+
+- **ID:** `2602.00782` · **primary:** `q-bio.BM` · **Submitted:** 2026-01-31
+- **Authors:** Jiahao Zhang, Zeqing Zhang, Di Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00782 · **PDF:** https://arxiv.org/pdf/2602.00782v1
+
+## 17151. Fast Non-Episodic Finite-Horizon RL with K-Step Lookahead Thresholding
+
+- **ID:** `2602.00781` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Jiamin Xu, Kyra Gan
+- **abs:** https://arxiv.org/abs/2602.00781 · **PDF:** https://arxiv.org/pdf/2602.00781v1
+
+## 17152. Environment-Aware Adaptive Pruning with Interleaved Inference Orchestration for Vision-Language-Action Models
+
+- **ID:** `2602.00780` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Yuting Huang, Leilei Ding, Zhipeng Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.00780 · **PDF:** https://arxiv.org/pdf/2602.00780v1
+
+## 17153. HyLRA: Hybrid Layer Reuse Attention for Efficient Long-Context Inference
+
+- **ID:** `2602.00777` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Xuan Ai, Qingqing Yang, Peng Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00777 · **PDF:** https://arxiv.org/pdf/2602.00777v1
+
+## 17154. Stable Time Series Prediction of Enterprise Carbon Emissions Based on Causal Inference
+
+- **ID:** `2602.00775` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Zitao Hong, Zhen Peng, Xueping Liu
+- **abs:** https://arxiv.org/abs/2602.00775 · **PDF:** https://arxiv.org/pdf/2602.00775v1
+
+## 17155. A Novel VAE-DML Fusion Framework for Causal Analysis of Greenwashing in the Mining Industry
+
+- **ID:** `2602.00774` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Yuxin Lu, Zhen Peng, Xiqiang Xia, et al.
+- **abs:** https://arxiv.org/abs/2602.00774 · **PDF:** https://arxiv.org/pdf/2602.00774v2
+
+## 17156. Provable Model Provenance Set for Large Language Models
+
+- **ID:** `2602.00772` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Xiaoqi Qiu, Hao Zeng, Zhiyu Hou, et al.
+- **abs:** https://arxiv.org/abs/2602.00772 · **PDF:** https://arxiv.org/pdf/2602.00772v1
+
+## 17157. Learning in Bayesian Stackelberg Games With Unknown Follower's Types
+
+- **ID:** `2602.00771` · **primary:** `cs.GT` · **Submitted:** 2026-01-31
+- **Authors:** Matteo Bollini, Francesco Bacchiocchi, Samuel Coutts, et al.
+- **abs:** https://arxiv.org/abs/2602.00771 · **PDF:** https://arxiv.org/pdf/2602.00771v1
+
+## 17158. Reasoning as State Transition: A Representational Analysis of Reasoning Evolution in Large Language Models
+
+- **ID:** `2602.00770` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Siyuan Zhang, Jialian Li, Yichi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.00770 · **PDF:** https://arxiv.org/pdf/2602.00770v1
+
+## 17159. Eliciting Trustworthiness Priors of Large Language Models via Economic Games
+
+- **ID:** `2602.00769` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Siyu Yan, Lusha Zhu, Jian-Qiao Zhu
+- **abs:** https://arxiv.org/abs/2602.00769 · **PDF:** https://arxiv.org/pdf/2602.00769v1
+
+## 17160. A Causal Perspective for Enhancing Jailbreak Attack and Defense
+
+- **ID:** `2602.04893` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Licheng Pan, Yunsheng Lu, Jiexi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.04893 · **PDF:** https://arxiv.org/pdf/2602.04893v1
+
+## 17161. BLOCK-EM: Preventing Emergent Misalignment by Blocking Causal Features
+
+- **ID:** `2602.00767` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Muhammed Ustaomeroglu, Guannan Qu
+- **abs:** https://arxiv.org/abs/2602.00767 · **PDF:** https://arxiv.org/pdf/2602.00767v1
+
+## 17162. Communications-Incentivized Collaborative Reasoning in NetGPT through Agentic Reinforcement Learning
+
+- **ID:** `2602.00766` · **primary:** `cs.MA` · **Submitted:** 2026-01-31
+- **Authors:** Xiaoxue Yu, Rongpeng Li, Zhifeng Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.00766 · **PDF:** https://arxiv.org/pdf/2602.00766v1
+
+## 17163. Benchmarking Automatic Speech Recognition for Indian Languages in Agricultural Contexts
+
+- **ID:** `2602.03868` · **primary:** `eess.AS` · **Submitted:** 2026-01-31
+- **Authors:** Chandrashekar M S, Vineet Singh, Lakshmi Pedapudi
+- **abs:** https://arxiv.org/abs/2602.03868 · **PDF:** https://arxiv.org/pdf/2602.03868v2
+
+## 17164. Evaluating Deep Learning-Based Nerve Segmentation in Brachial Plexus Ultrasound Under Realistic Data Constraints
+
+- **ID:** `2602.00763` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Dylan Yves, Khush Agarwal, Jonathan Hoyin Chan, et al.
+- **abs:** https://arxiv.org/abs/2602.00763 · **PDF:** https://arxiv.org/pdf/2602.00763v1
+
+## 17165. WordCraft: Scaffolding the Keyword Method for L2 Vocabulary Learning with Multimodal LLMs
+
+- **ID:** `2602.00762` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Yuheng Shao, Junjie Xiong, Chaoran Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.00762 · **PDF:** https://arxiv.org/pdf/2602.00762v1
+
+## 17166. A Gradient Boosted Mixed-Model Machine Learning Framework for Vessel Speed in the U.S. Arctic
+
+- **ID:** `2602.12292` · **primary:** `eess.SP` · **Submitted:** 2026-01-31
+- **Authors:** Mauli Pant, Linda Fernandez, Indranil Sahoo
+- **abs:** https://arxiv.org/abs/2602.12292 · **PDF:** https://arxiv.org/pdf/2602.12292v1
+
+## 17167. APR: Penalizing Structural Redundancy in Large Reasoning Models via Anchor-based Process Rewards
+
+- **ID:** `2602.00760` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Kaiyan Chang, Chenwei Zhu, Yingfeng Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.00760 · **PDF:** https://arxiv.org/pdf/2602.00760v2
+
+## 17168. Adaptive Ability Decomposing for Unlocking Large Reasoning Model Effective Reinforcement Learning
+
+- **ID:** `2602.00759` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Zhipeng Chen, Xiaobo Qin, Wayne Xin Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.00759 · **PDF:** https://arxiv.org/pdf/2602.00759v1
+
+## 17169. Temporal Leakage in Search-Engine Date-Filtered Web Retrieval: A Case Study from Retrospective Forecasting
+
+- **ID:** `2602.00758` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Ali El Lahib, Ying-Jieh Xia, Zehan Li, et al.
+- **abs:** https://arxiv.org/abs/2602.00758 · **PDF:** https://arxiv.org/pdf/2602.00758v1
+
+## 17170. A New Workflow for Materials Discovery Bridging the Gap Between Experimental Databases and Graph Neural Networks
+
+- **ID:** `2602.00756` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-31
+- **Authors:** Brandon Schoener, Yuting Hu, Pasit Wanlapha, et al.
+- **abs:** https://arxiv.org/abs/2602.00756 · **PDF:** https://arxiv.org/pdf/2602.00756v1
+
+## 17171. Evolving Interpretable Constitutions for Multi-Agent Coordination
+
+- **ID:** `2602.00755` · **primary:** `cs.MA` · **Submitted:** 2026-01-31
+- **Authors:** Ujwal Kumar, Alice Saito, Hershraj Niranjani, et al.
+- **abs:** https://arxiv.org/abs/2602.00755 · **PDF:** https://arxiv.org/pdf/2602.00755v1
+
+## 17172. GraphNNK -- Graph Classification and Interpretability
+
+- **ID:** `2602.00753` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Zeljko Bolevic, Milos Brajovic, Isidora Stankovic, et al.
+- **abs:** https://arxiv.org/abs/2602.00753 · **PDF:** https://arxiv.org/pdf/2602.00753v1
+
+## 17173. Engineering AI Agents for Clinical Workflows: A Case Study in Architecture,MLOps, and Governance
+
+- **ID:** `2602.00751` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Cláudio Lúcio do Val Lopes, João Marcus Pitta, Fabiano Belém, et al.
+- **abs:** https://arxiv.org/abs/2602.00751 · **PDF:** https://arxiv.org/pdf/2602.00751v1
+
+## 17174. Bypassing Prompt Injection Detectors through Evasive Injections
+
+- **ID:** `2602.00750` · **primary:** `cs.CR` · **Submitted:** 2026-01-31
+- **Authors:** Md Jahedur Rahman, Ihsen Alouani
+- **abs:** https://arxiv.org/abs/2602.00750 · **PDF:** https://arxiv.org/pdf/2602.00750v2
+
+## 17175. HyperOffload: Graph-Driven Hierarchical Memory Management for Large Language Models on SuperNode Architectures
+
+- **ID:** `2602.00748` · **primary:** `cs.DC` · **Submitted:** 2026-01-31
+- **Authors:** Fangxin Liu, Qinghua Zhang, Hanjing Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.00748 · **PDF:** https://arxiv.org/pdf/2602.00748v2
+
+## 17176. Decouple Searching from Training: Scaling Data Mixing via Model Merging for Large Language Model Pre-training
+
+- **ID:** `2602.00747` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Shengrui Li, Fei Zhao, Kaiyan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.00747 · **PDF:** https://arxiv.org/pdf/2602.00747v1
+
+## 17177. SA-VLA: Spatially-Aware Flow-Matching for Vision-Language-Action Reinforcement Learning
+
+- **ID:** `2602.00743` · **primary:** `cs.RO` · **Submitted:** 2026-01-31
+- **Authors:** Xu Pan, Zhenglin Wan, Xingrui Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.00743 · **PDF:** https://arxiv.org/pdf/2602.00743v1
+
+## 17178. CURP: Codebook-based Continuous User Representation for Personalized Generation with LLMs
+
+- **ID:** `2602.00742` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Liang Wang, Xinyi Mou, Xiaoyou Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.00742 · **PDF:** https://arxiv.org/pdf/2602.00742v1
+
+## 17179. ExperienceWeaver: Optimizing Small-sample Experience Learning for LLM-based Clinical Text Improvement
+
+- **ID:** `2602.00740` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Ziyan Xiao, Yinghao Zhu, Liang Peng, et al.
+- **abs:** https://arxiv.org/abs/2602.00740 · **PDF:** https://arxiv.org/pdf/2602.00740v1
+
+## 17180. Pareto-Conditioned Diffusion Models for Offline Multi-Objective Optimization
+
+- **ID:** `2602.00737` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Jatan Shrestha, Santeri Heiskanen, Kari Hepola, et al.
+- **abs:** https://arxiv.org/abs/2602.00737 · **PDF:** https://arxiv.org/pdf/2602.00737v2
+
+## 17181. EchoReview: Learning Peer Review from the Echoes of Scientific Citations
+
+- **ID:** `2602.00733` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Yinuo Zhang, Dingcheng Huang, Haifeng Suo, et al.
+- **abs:** https://arxiv.org/abs/2602.00733 · **PDF:** https://arxiv.org/pdf/2602.00733v1
+
+## 17182. Neuro-symbolic AI for Predictive Maintenance (PdM) -- review and recommendations
+
+- **ID:** `2602.00731` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Kyle Hamilton, Muhammad Intizar Ali
+- **abs:** https://arxiv.org/abs/2602.00731 · **PDF:** https://arxiv.org/pdf/2602.00731v2
+
+## 17183. Augmenting Clinical Decision-Making with an Interactive and Interpretable AI Copilot: A Real-World User Study with Clinicians in Nephrology and Obstetrics
+
+- **ID:** `2602.00726` · **primary:** `cs.HC` · **Submitted:** 2026-01-31
+- **Authors:** Yinghao Zhu, Dehao Sui, Zixiang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00726 · **PDF:** https://arxiv.org/pdf/2602.00726v1
+
+## 17184. Rethinking Hallucinations: Correctness, Consistency, and Prompt Multiplicity
+
+- **ID:** `2602.00723` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Prakhar Ganesh, Reza Shokri, Golnoosh Farnadi
+- **abs:** https://arxiv.org/abs/2602.00723 · **PDF:** https://arxiv.org/pdf/2602.00723v1
+
+## 17185. Spectral Imbalance Causes Forgetting in Low-Rank Continual Adaptation
+
+- **ID:** `2602.00722` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Hao Gu, Mao-Lin Luo, Zi-Hao Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.00722 · **PDF:** https://arxiv.org/pdf/2602.00722v1
+
+## 17186. Federated Learning at the Forefront of Fairness: A Multifaceted Perspective
+
+- **ID:** `2602.00718` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Noorain Mukhtiar, Adnan Mahmood, Yipeng Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.00718 · **PDF:** https://arxiv.org/pdf/2602.00718v1
+
+## 17187. Deep Time-series Forecasting Needs Kernelized Moment Balancing
+
+- **ID:** `2602.00717` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Licheng Pan, Hao Wang, Haocheng Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.00717 · **PDF:** https://arxiv.org/pdf/2602.00717v1
+
+## 17188. Emergence of Distortions in High-Dimensional Guided Diffusion Models
+
+- **ID:** `2602.00716` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Enrico Ventura, Beatrice Achilli, Luca Ambrogioni, et al.
+- **abs:** https://arxiv.org/abs/2602.00716 · **PDF:** https://arxiv.org/pdf/2602.00716v3
+
+## 17189. From Detection to Prevention: Explaining Security-Critical Code to Avoid Vulnerabilities
+
+- **ID:** `2602.00711` · **primary:** `cs.CR` · **Submitted:** 2026-01-31
+- **Authors:** Ranjith Krishnamurthy, Oshando Johnson, Goran Piskachev, et al.
+- **abs:** https://arxiv.org/abs/2602.00711 · **PDF:** https://arxiv.org/pdf/2602.00711v1
+
+## 17190. Learning More from Less: Unlocking Internal Representations for Benchmark Compression
+
+- **ID:** `2602.00710` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Yueqi Zhang, Jin Hu, Shaoxiong Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.00710 · **PDF:** https://arxiv.org/pdf/2602.00710v2
+
+## 17191. Physics-informed Diffusion Generation for Geomagnetic Map Interpolation
+
+- **ID:** `2602.00709` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Wenda Li, Tongya Zheng, Kaixuan Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.00709 · **PDF:** https://arxiv.org/pdf/2602.00709v1
+
+## 17192. Self-Guard: Defending Large Reasoning Models via enhanced self-reflection
+
+- **ID:** `2602.00707` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Jingnan Zheng, Jingjun Xu, Yanzhen Luo, et al.
+- **abs:** https://arxiv.org/abs/2602.00707 · **PDF:** https://arxiv.org/pdf/2602.00707v1
+
+## 17193. LocalV: Exploiting Information Locality for IP-level Verilog Generation
+
+- **ID:** `2602.00704` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Hanqi Lyu, Di Huang, Yaoyu Zhu, et al.
+- **abs:** https://arxiv.org/abs/2602.00704 · **PDF:** https://arxiv.org/pdf/2602.00704v1
+
+## 17194. Cross-Modal Binary Attention: An Energy-Efficient Fusion Framework for Audio-Visual Learning
+
+- **ID:** `2602.00701` · **primary:** `cs.MM` · **Submitted:** 2026-01-31
+- **Authors:** Mohamed Saleh, Zahra Ahmadi
+- **abs:** https://arxiv.org/abs/2602.00701 · **PDF:** https://arxiv.org/pdf/2602.00701v1
+
+## 17195. From Prompt to Graph: Comparing LLM-Based Information Extraction Strategies in Domain-Specific Ontology Development
+
+- **ID:** `2602.00699` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Xuan Liu, Ziyu Li, Mu He, et al.
+- **abs:** https://arxiv.org/abs/2602.00699 · **PDF:** https://arxiv.org/pdf/2602.00699v1
+
+## 17196. Forecasting Energy Availability in Local Energy Communities via LSTM Federated Learning
+
+- **ID:** `2602.00694` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Fabio Turazza, Marcello Pietri, Natalia Selini Hadjidimitriou, et al.
+- **abs:** https://arxiv.org/abs/2602.00694 · **PDF:** https://arxiv.org/pdf/2602.00694v1
+
+## 17197. Beyond Static Instruction: A Multi-agent AI Framework for Adaptive Augmented Reality Robot Training
+
+- **ID:** `2603.00016` · **primary:** `cs.RO` · **Submitted:** 2026-01-31
+- **Authors:** Nicolas Leins, Jana Gonnermann-Müller, Malte Teichmann, et al.
+- **abs:** https://arxiv.org/abs/2603.00016 · **PDF:** https://arxiv.org/pdf/2603.00016v2
+
+## 17198. Vectra: A New Metric, Dataset, and Model for Visual Quality Assessment in E-Commerce In-Image Machine Translation
+
+- **ID:** `2602.07014` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Qingyu Wu, Yuxuan Han, Haijun Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07014 · **PDF:** https://arxiv.org/pdf/2602.07014v1
+
+## 17199. Topology and Geometry of the Learning Space of ReLU Networks: Connectivity and Singularities
+
+- **ID:** `2602.00693` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Marco Nurisso, Pierrick Leroy, Giovanni Petri, et al.
+- **abs:** https://arxiv.org/abs/2602.00693 · **PDF:** https://arxiv.org/pdf/2602.00693v1
+
+## 17200. Provably Protecting Fine-Tuned LLMs from Training Data Extraction
+
+- **ID:** `2602.00688` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Tom Segal, Asaf Shabtai, Yuval Elovici
+- **abs:** https://arxiv.org/abs/2602.00688 · **PDF:** https://arxiv.org/pdf/2602.00688v1
+
+## 17201. HumanStudy-Bench: Towards AI Agent Design for Participant Simulation
+
+- **ID:** `2602.00685` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Xuan Liu, Haoyang Shang, Zizhang Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.00685 · **PDF:** https://arxiv.org/pdf/2602.00685v1
+
+## 17202. RecGOAT: Graph Optimal Adaptive Transport for LLM-Enhanced Multimodal Recommendation with Dual Semantic Alignment
+
+- **ID:** `2602.00682` · **primary:** `cs.IR` · **Submitted:** 2026-01-31
+- **Authors:** Yuecheng Li, Hengwei Ju, Zeyu Song, et al.
+- **abs:** https://arxiv.org/abs/2602.00682 · **PDF:** https://arxiv.org/pdf/2602.00682v1
+
+## 17203. Audio-to-Image Bird Species Retrieval without Audio-Image Pairs via Text Distillation
+
+- **ID:** `2602.00681` · **primary:** `cs.SD` · **Submitted:** 2026-01-31
+- **Authors:** Ilyass Moummad, Marius Miron, Lukas Rauch, et al.
+- **abs:** https://arxiv.org/abs/2602.00681 · **PDF:** https://arxiv.org/pdf/2602.00681v1
+
+## 17204. OpenGuanDan: A Large-Scale Imperfect Information Game Benchmark
+
+- **ID:** `2602.00676` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Chao Li, Shangdong Yang, Chiheng Zhan, et al.
+- **abs:** https://arxiv.org/abs/2602.00676 · **PDF:** https://arxiv.org/pdf/2602.00676v1
+
+## 17205. Strong Linear Baselines Strike Back: Closed-Form Linear Models as Gaussian Process Conditional Density Estimators for TSAD
+
+- **ID:** `2602.00672` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Aleksandr Yugay, Hang Cui, Changhua Pei, et al.
+- **abs:** https://arxiv.org/abs/2602.00672 · **PDF:** https://arxiv.org/pdf/2602.00672v1
+
+## 17206. Three-Way Emotion Classification of EEG-based Signals using Machine Learning
+
+- **ID:** `2602.00670` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Ashna Purwar, Gaurav Simkar, Madhumita, et al.
+- **abs:** https://arxiv.org/abs/2602.00670 · **PDF:** https://arxiv.org/pdf/2602.00670v1
+
+## 17207. Improving Neuropathological Reconstruction Fidelity via AI Slice Imputation
+
+- **ID:** `2602.00669` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Marina Crespo Aguirre, Jonathan Williams-Ramirez, Dina Zemlyanker, et al.
+- **abs:** https://arxiv.org/abs/2602.00669 · **PDF:** https://arxiv.org/pdf/2602.00669v1
+
+## 17208. Can Small Language Models Handle Context-Summarized Multi-Turn Customer-Service QA? A Synthetic Data-Driven Comparative Evaluation
+
+- **ID:** `2602.00665` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Lakshan Cooray, Deshan Sumanathilaka, Pattigadapa Venkatesh Raju
+- **abs:** https://arxiv.org/abs/2602.00665 · **PDF:** https://arxiv.org/pdf/2602.00665v2
+
+## 17209. SEISMO: Increasing Sample Efficiency in Molecular Optimization with a Trajectory-Aware LLM Agent
+
+- **ID:** `2602.00663` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Fabian P. Krüger, Andrea Hunklinger, Adrian Wolny, et al.
+- **abs:** https://arxiv.org/abs/2602.00663 · **PDF:** https://arxiv.org/pdf/2602.00663v2
+
+## 17210. Predictive Maintenance for Ultrafiltration Membranes Using Explainable Similarity-Based Prognostics
+
+- **ID:** `2602.00659` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Qusai Khaled, Laura Genga, Uzay Kaymak
+- **abs:** https://arxiv.org/abs/2602.00659 · **PDF:** https://arxiv.org/pdf/2602.00659v1
+
+## 17211. Non-Clashing Teaching in Graphs: Algorithms, Complexity, and Bounds
+
+- **ID:** `2602.00657` · **primary:** `cs.CC` · **Submitted:** 2026-01-31
+- **Authors:** Sujoy Bhore, Liana Khazaliya, Fionn Mc Inerney
+- **abs:** https://arxiv.org/abs/2602.00657 · **PDF:** https://arxiv.org/pdf/2602.00657v2
+
+## 17212. Riemannian Flow Matching for Disentangled Graph Domain Adaptation
+
+- **ID:** `2602.00656` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Yingxu Wang, Xinwang Liu, Mengzhu Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00656 · **PDF:** https://arxiv.org/pdf/2602.00656v1
+
+## 17213. PHAT: Modeling Period Heterogeneity for Multivariate Time Series Forecasting
+
+- **ID:** `2602.00654` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Jiaming Ma, Qihe Huang, Haofeng Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.00654 · **PDF:** https://arxiv.org/pdf/2602.00654v3
+
+## 17214. Non-Contrastive Vision-Language Learning with Predictive Embedding Alignment
+
+- **ID:** `2602.00653` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Lukas Kuhn, Giuseppe Serra, Florian Buettner
+- **abs:** https://arxiv.org/abs/2602.00653 · **PDF:** https://arxiv.org/pdf/2602.00653v2
+
+## 17215. Steering to Say No: Configurable Refusal via Activation Steering in Vision Language Models
+
+- **ID:** `2602.07013` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Jiaxi Yang, Shicheng Liu, Yuchen Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.07013 · **PDF:** https://arxiv.org/pdf/2602.07013v1
+
+## 17216. CoRe-Fed: Bridging Collaborative and Representation Fairness via Federated Embedding Distillation
+
+- **ID:** `2602.00647` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Noorain Mukhtiar, Adnan Mahmood, Quan Z. Sheng
+- **abs:** https://arxiv.org/abs/2602.00647 · **PDF:** https://arxiv.org/pdf/2602.00647v1
+
+## 17217. A General Model for Retinal Segmentation and Quantification
+
+- **ID:** `2602.07012` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Zhonghua Wang, Lie Ju, Sijia Li, et al.
+- **abs:** https://arxiv.org/abs/2602.07012 · **PDF:** https://arxiv.org/pdf/2602.07012v1
+
+## 17218. LegalOne: A Family of Foundation Models for Reliable Legal Reasoning
+
+- **ID:** `2602.00642` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Haitao Li, Yifan Chen, Shuo Miao, et al.
+- **abs:** https://arxiv.org/abs/2602.00642 · **PDF:** https://arxiv.org/pdf/2602.00642v2
+
+## 17219. Sampling from multi-modal distributions on Riemannian manifolds with training-free stochastic interpolants
+
+- **ID:** `2602.00641` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Alain Durmus, Maxence Noble, Thibaut Pellerin
+- **abs:** https://arxiv.org/abs/2602.00641 · **PDF:** https://arxiv.org/pdf/2602.00641v1
+
+## 17220. Combinatorial Bandit Bayesian Optimization for Tensor Outputs
+
+- **ID:** `2602.00640` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Jingru Huang, Haijie Xu, Jie Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.00640 · **PDF:** https://arxiv.org/pdf/2602.00640v2
+
+## 17221. Formal Semantic Control over Language Models
+
+- **ID:** `2602.00638` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Yingji Zhang
+- **abs:** https://arxiv.org/abs/2602.00638 · **PDF:** https://arxiv.org/pdf/2602.00638v1
+
+## 17222. On the Equilibrium between Feasible Zone and Uncertain Model in Safe Exploration
+
+- **ID:** `2602.00636` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Yujie Yang, Zhilong Zheng, Shengbo Eben Li
+- **abs:** https://arxiv.org/abs/2602.00636 · **PDF:** https://arxiv.org/pdf/2602.00636v2
+
+## 17223. S$^3$POT: Contrast-Driven Face Occlusion Segmentation via Self-Supervised Prompt Learning
+
+- **ID:** `2602.00635` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Lingsong Wang, Mancheng Meng, Ziyan Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.00635 · **PDF:** https://arxiv.org/pdf/2602.00635v1
+
+## 17224. ProphetKV: User-Query-Driven Selective Recomputation for Efficient KV Cache Reuse in Retrieval-Augmented Generation
+
+- **ID:** `2602.02579` · **primary:** `cs.OS` · **Submitted:** 2026-01-31
+- **Authors:** Shihao Wang, Jiahao Chen, Yanqi Pan, et al.
+- **abs:** https://arxiv.org/abs/2602.02579 · **PDF:** https://arxiv.org/pdf/2602.02579v3
+
+## 17225. Action-Free Offline-to-Online RL via Discretised State Policies
+
+- **ID:** `2602.00629` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Natinael Solomon Neggatu, Jeremie Houssineau, Giovanni Montana
+- **abs:** https://arxiv.org/abs/2602.00629 · **PDF:** https://arxiv.org/pdf/2602.00629v1
+
+## 17226. From Associations to Activations: Comparing Behavioral and Hidden-State Semantic Geometry in LLMs
+
+- **ID:** `2602.00628` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Louis Schiekiera, Max Zimmer, Christophe Roux, et al.
+- **abs:** https://arxiv.org/abs/2602.00628 · **PDF:** https://arxiv.org/pdf/2602.00628v2
+
+## 17227. MoDEx: Mixture of Depth-specific Experts for Multivariate Long-term Time Series Forecasting
+
+- **ID:** `2602.00624` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Hyekyung Yoon, Minhyuk Lee, Imseung Park, et al.
+- **abs:** https://arxiv.org/abs/2602.00624 · **PDF:** https://arxiv.org/pdf/2602.00624v1
+
+## 17228. Relaxed Triangle Inequality for Kullback-Leibler Divergence Between Multivariate Gaussian Distributions
+
+- **ID:** `2602.02577` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Shiji Xiao, Yufeng Zhang, Chubo Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.02577 · **PDF:** https://arxiv.org/pdf/2602.02577v2
+
+## 17229. Rethinking Zero-Shot Time Series Classification: From Task-specific Classifiers to In-Context Inference
+
+- **ID:** `2602.00620` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Juntao Fang, Shifeng Xie, Shengbin Nie, et al.
+- **abs:** https://arxiv.org/abs/2602.00620 · **PDF:** https://arxiv.org/pdf/2602.00620v1
+
+## 17230. Jailbreaking LLMs via Calibration
+
+- **ID:** `2602.00619` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Yuxuan Lu, Yongkang Guo, Yuqing Kong
+- **abs:** https://arxiv.org/abs/2602.00619 · **PDF:** https://arxiv.org/pdf/2602.00619v1
+
+## 17231. Inference-Only Prompt Projection for Safe Text-to-Image Generation with TV Guarantees
+
+- **ID:** `2602.00616` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Minhyuk Lee, Hyekyung Yoon, Myungjoo Kang
+- **abs:** https://arxiv.org/abs/2602.00616 · **PDF:** https://arxiv.org/pdf/2602.00616v1
+
+## 17232. Transformer-Based Model for Multilingual Hope Speech Detection
+
+- **ID:** `2602.00613` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Nsrin Ashraf, Mariam Labib, Hamada Nayel
+- **abs:** https://arxiv.org/abs/2602.00613 · **PDF:** https://arxiv.org/pdf/2602.00613v1
+
+## 17233. Lookahead-then-Verify: Reliable Constrained Decoding for Diffusion LLMs under Context-Free Grammars
+
+- **ID:** `2602.00612` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Yitong Zhang, Yongmin Li, Yuetong Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.00612 · **PDF:** https://arxiv.org/pdf/2602.00612v2
+
+## 17234. Structured Self-Consistency:A Multi-Task Evaluation of LLMs on VirtualHome
+
+- **ID:** `2602.00611` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Jiaqi Xu, Tao Huang, Kai Zhang
+- **abs:** https://arxiv.org/abs/2602.00611 · **PDF:** https://arxiv.org/pdf/2602.00611v2
+
+## 17235. Scalable Generative Game Engine: Breaking the Resolution Wall via Hardware-Algorithm Co-Design
+
+- **ID:** `2602.00608` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Wei Zeng, Xuchen Li, Ruili Feng, et al.
+- **abs:** https://arxiv.org/abs/2602.00608 · **PDF:** https://arxiv.org/pdf/2602.00608v1
+
+## 17236. Actor-Dual-Critic Dynamics for Zero-sum and Identical-Interest Stochastic Games
+
+- **ID:** `2602.00606` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Ahmed Said Donmez, Yuksel Arslantas, Muhammed O. Sayin
+- **abs:** https://arxiv.org/abs/2602.00606 · **PDF:** https://arxiv.org/pdf/2602.00606v1
+
+## 17237. Direct Preference Optimization with Rating Information: Practical Algorithms and Provable Gains
+
+- **ID:** `2602.00603` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Luca Viano, Ruida Zhou, Yifan Sun, et al.
+- **abs:** https://arxiv.org/abs/2602.00603 · **PDF:** https://arxiv.org/pdf/2602.00603v1
+
+## 17238. Latent Context Compilation: Distilling Long Context into Compact Portable Memory
+
+- **ID:** `2602.21221` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Zeju Li, Yizhou Zhou, Qiang Xu
+- **abs:** https://arxiv.org/abs/2602.21221 · **PDF:** https://arxiv.org/pdf/2602.21221v1
+
+## 17239. Responsible AI in Business
+
+- **ID:** `2602.13244` · **primary:** `cs.CY` · **Submitted:** 2026-01-31
+- **Authors:** Stephan Sandfuchs, Diako Farooghi, Janis Mohr, et al.
+- **abs:** https://arxiv.org/abs/2602.13244 · **PDF:** https://arxiv.org/pdf/2602.13244v1
+
+## 17240. Hermes the Polyglot: A Unified Framework to Enhance Expressiveness for Multimodal Interlingual Subtitling
+
+- **ID:** `2602.00597` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Chaoqun Cui, Shijing Wang, Liangbin Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.00597 · **PDF:** https://arxiv.org/pdf/2602.00597v1
+
+## 17241. Kernelized Edge Attention: Addressing Semantic Attention Blurring in Temporal Graph Neural Networks
+
+- **ID:** `2602.00596` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Govind Waghmare, Srini Rohan Gujulla Leel, Nikhil Tumbde, et al.
+- **abs:** https://arxiv.org/abs/2602.00596 · **PDF:** https://arxiv.org/pdf/2602.00596v1
+
+## 17242. Kanade: A Simple Disentangled Tokenizer for Spoken Language Modeling
+
+- **ID:** `2602.00594` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Zhijie Huang, Stephen McIntosh, Daisuke Saito, et al.
+- **abs:** https://arxiv.org/abs/2602.00594 · **PDF:** https://arxiv.org/pdf/2602.00594v1
+
+## 17243. From Pixels to Facts (Pix2Fact): Benchmarking Multi-Hop Reasoning for Fine-Grained Visual Fact Checking
+
+- **ID:** `2602.00593` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Yifan Jiang, Cong Zhang, Bofei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.00593 · **PDF:** https://arxiv.org/pdf/2602.00593v1
+
+## 17244. DockSmith: Scaling Reliable Coding Environments via an Agentic Docker Builder
+
+- **ID:** `2602.00592` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Jiaran Zhang, Luck Ma, Yanhao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.00592 · **PDF:** https://arxiv.org/pdf/2602.00592v1
+
+## 17245. Multimodal Machine Learning for Integrating Heterogeneous Analytical Systems
+
+- **ID:** `2602.00590` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-31
+- **Authors:** Shun Muroga, Hideaki Nakajima, Taiyo Shimizu, et al.
+- **abs:** https://arxiv.org/abs/2602.00590 · **PDF:** https://arxiv.org/pdf/2602.00590v1
+
+## 17246. SEER: Transformer-based Robust Time Series Forecasting via Automated Patch Enhancement and Replacement
+
+- **ID:** `2602.00589` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Xiangfei Qiu, Xvyuan Liu, Tianen Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.00589 · **PDF:** https://arxiv.org/pdf/2602.00589v1
+
+## 17247. The French Drama Revolution: Political Economy and Literary Production, 1700-1900
+
+- **ID:** `2602.00588` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Thiago Dumont Oliveira
+- **abs:** https://arxiv.org/abs/2602.00588 · **PDF:** https://arxiv.org/pdf/2602.00588v1
+
+## 17248. Safe Langevin Soft Actor Critic
+
+- **ID:** `2602.00587` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Mahesh Keswani, Samyak Jain, Raunak P. Bhattacharyya
+- **abs:** https://arxiv.org/abs/2602.00587 · **PDF:** https://arxiv.org/pdf/2602.00587v1
+
+## 17249. RAG-GNN: Integrating Retrieved Knowledge with Graph Neural Networks for Precision Medicine
+
+- **ID:** `2602.00586` · **primary:** `q-bio.MN` · **Submitted:** 2026-01-31
+- **Authors:** Hasi Hays, William J. Richardson
+- **abs:** https://arxiv.org/abs/2602.00586 · **PDF:** https://arxiv.org/pdf/2602.00586v1
+
+## 17250. Exploring Information Seeking Agent Consolidation
+
+- **ID:** `2602.00585` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Guochen Yan, Jialong Wu, Zhengwei Tao, et al.
+- **abs:** https://arxiv.org/abs/2602.00585 · **PDF:** https://arxiv.org/pdf/2602.00585v1
+
+## 17251. MAUGen: A Unified Diffusion Approach for Multi-Identity Facial Expression and AU Label Generation
+
+- **ID:** `2602.00583` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Xiangdong Li, Ye Lou, Ao Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.00583 · **PDF:** https://arxiv.org/pdf/2602.00583v1
+
+## 17252. Bridging Time and Frequency: A Joint Modeling Framework for Irregular Multivariate Time Series Forecasting
+
+- **ID:** `2602.00582` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Xiangfei Qiu, Kangjia Yan, Xvyuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.00582 · **PDF:** https://arxiv.org/pdf/2602.00582v1
+
+## 17253. Small Shifts, Large Gains: Unlocking Traditional TSP Heuristic Guided-Sampling via Unsupervised Neural Instance Modification
+
+- **ID:** `2602.00580` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Wei Huang, Hanchen Wang, Dong Wen, et al.
+- **abs:** https://arxiv.org/abs/2602.00580 · **PDF:** https://arxiv.org/pdf/2602.00580v1
+
+## 17254. SAU: Sparsity-Aware Unlearning for LLMs via Gradient Masking and Importance Redistribution
+
+- **ID:** `2602.00577` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Yuze Wang, Yujia Tong, Xuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.00577 · **PDF:** https://arxiv.org/pdf/2602.00577v3
+
+## 17255. Data Distribution as a Lever for Guiding Optimizers Toward Superior Generalization in LLMs
+
+- **ID:** `2602.00576` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Tushaar Gangavarapu, Jiping Li, Christopher Vattheuer, et al.
+- **abs:** https://arxiv.org/abs/2602.00576 · **PDF:** https://arxiv.org/pdf/2602.00576v1
+
+## 17256. Learning Modal-Mixed Chain-of-Thought Reasoning with Latent Embeddings
+
+- **ID:** `2602.00574` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Yifei Shao, Kun Zhou, Ziming Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.00574 · **PDF:** https://arxiv.org/pdf/2602.00574v1
+
+## 17257. When Classes Evolve: A Benchmark and Framework for Stage-Aware Class-Incremental Learning
+
+- **ID:** `2602.00573` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Zheng Zhang, Tao Hu, Xueheng Li, et al.
+- **abs:** https://arxiv.org/abs/2602.00573 · **PDF:** https://arxiv.org/pdf/2602.00573v1
+
+## 17258. Forget by Uncertainty: Orthogonal Entropy Unlearning for Quantized Neural Networks
+
+- **ID:** `2602.00567` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Tian Zhang, Yujia Tong, Junhao Dong, et al.
+- **abs:** https://arxiv.org/abs/2602.00567 · **PDF:** https://arxiv.org/pdf/2602.00567v1
+
+## 17259. Product Interaction: An Algebraic Formalism for Deep Learning Architectures
+
+- **ID:** `2602.02573` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Haonan Dong, Chun-Wun Cheng, Angelica I. Aviles-Rivero
+- **abs:** https://arxiv.org/abs/2602.02573 · **PDF:** https://arxiv.org/pdf/2602.02573v1
+
+## 17260. Unmasking Reasoning Processes: A Process-aware Benchmark for Evaluating Structural Mathematical Reasoning in LLMs
+
+- **ID:** `2602.00564` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Xiang Zheng, Weiqi Zhai, Wei Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00564 · **PDF:** https://arxiv.org/pdf/2602.00564v2
+
+## 17261. Uncovering Latent Communication Patterns in Brain Networks via Adaptive Flow Routing
+
+- **ID:** `2602.00561` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Tianhao Huang, Guanghui Min, Zhenyu Lei, et al.
+- **abs:** https://arxiv.org/abs/2602.00561 · **PDF:** https://arxiv.org/pdf/2602.00561v1
+
+## 17262. Learning to Decode Against Compositional Hallucination in Video Multimodal Large Language Models
+
+- **ID:** `2602.00559` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Wenbin Xing, Quanxing Zha, Lizheng Zu, et al.
+- **abs:** https://arxiv.org/abs/2602.00559 · **PDF:** https://arxiv.org/pdf/2602.00559v1
+
+## 17263. A Hierarchical and Attentional Analysis of Argument Structure Constructions in BERT Using Naturalistic Corpora
+
+- **ID:** `2602.00554` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Liu Kaipeng, Wu Ling
+- **abs:** https://arxiv.org/abs/2602.00554 · **PDF:** https://arxiv.org/pdf/2602.00554v1
+
+## 17264. Beyond the Node: Clade-level Selection for Efficient MCTS in Automatic Heuristic Design
+
+- **ID:** `2602.00549` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Kezhao Lai, Yutao Lai, Hai-Lin Liu
+- **abs:** https://arxiv.org/abs/2602.00549 · **PDF:** https://arxiv.org/pdf/2602.00549v1
+
+## 17265. Contrastive Domain Generalization for Cross-Instrument Molecular Identification in Mass Spectrometry
+
+- **ID:** `2602.00547` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Seunghyun Yoo, Sanghong Kim, Namkyung Yoon, et al.
+- **abs:** https://arxiv.org/abs/2602.00547 · **PDF:** https://arxiv.org/pdf/2602.00547v1
+
+## 17266. Depth, Not Data: An Analysis of Hessian Spectral Bifurcation
+
+- **ID:** `2602.00545` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Shenyang Deng, Boyao Liao, Zhuoli Ouyang, et al.
+- **abs:** https://arxiv.org/abs/2602.00545 · **PDF:** https://arxiv.org/pdf/2602.00545v1
+
+## 17267. Reasoning by Commented Code for Table Question Answering
+
+- **ID:** `2602.00543` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Seho Pyo, Jiheon Seok, Jaejin Lee
+- **abs:** https://arxiv.org/abs/2602.00543 · **PDF:** https://arxiv.org/pdf/2602.00543v1
+
+## 17268. NPNet: A Non-Parametric Network with Adaptive Gaussian-Fourier Positional Encoding for 3D Classification and Segmentation
+
+- **ID:** `2602.00542` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Mohammad Saeid, Amir Salarpour, Pedram MohajerAnsari, et al.
+- **abs:** https://arxiv.org/abs/2602.00542 · **PDF:** https://arxiv.org/pdf/2602.00542v1
+
+## 17269. One Loss to Rule Them All: Marked Time-to-Event for Structured EHR Foundation Models
+
+- **ID:** `2602.00541` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Zilin Jing, Vincent Jeanselme, Yuta Kobayashi, et al.
+- **abs:** https://arxiv.org/abs/2602.00541 · **PDF:** https://arxiv.org/pdf/2602.00541v1
+
+## 17270. Surrogate Ensemble in Expensive Multi-Objective Optimization via Deep Q-Learning
+
+- **ID:** `2602.00540` · **primary:** `cs.NE` · **Submitted:** 2026-01-31
+- **Authors:** Yuxin Wu, Hongshu Guo, Ting Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.00540 · **PDF:** https://arxiv.org/pdf/2602.00540v1
+
+## 17271. OpenDDI: A Comprehensive Benchmark for DDI Prediction
+
+- **ID:** `2602.00539` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Xinmo Jin, Bowen Fan, Xunkai Li, et al.
+- **abs:** https://arxiv.org/abs/2602.00539 · **PDF:** https://arxiv.org/pdf/2602.00539v1
+
+## 17272. Invertible Memory Flow Networks
+
+- **ID:** `2602.00535` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Liyu Zerihun, Alexandr Plashchinsky
+- **abs:** https://arxiv.org/abs/2602.00535 · **PDF:** https://arxiv.org/pdf/2602.00535v1
+
+## 17273. AIRE-Prune: Asymptotic Impulse-Response Energy for State Pruning in State Space Models
+
+- **ID:** `2602.00534` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Apurba Prasad Padhy, Fernando Camacho, Saibal Mukhopadhyay
+- **abs:** https://arxiv.org/abs/2602.00534 · **PDF:** https://arxiv.org/pdf/2602.00534v1
+
+## 17274. Convergent World Representations and Divergent Tasks
+
+- **ID:** `2602.00533` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Core Francisco Park
+- **abs:** https://arxiv.org/abs/2602.00533 · **PDF:** https://arxiv.org/pdf/2602.00533v1
+
+## 17275. Reinforcement Learning-assisted Constraint Relaxation for Constrained Expensive Optimization
+
+- **ID:** `2602.00532` · **primary:** `cs.NE` · **Submitted:** 2026-01-31
+- **Authors:** Qianhao Zhu, Sijie Ma, Zeyuan Ma, et al.
+- **abs:** https://arxiv.org/abs/2602.00532 · **PDF:** https://arxiv.org/pdf/2602.00532v1
+
+## 17276. Reward Shaping for Inference-Time Alignment: A Stackelberg Game Perspective
+
+- **ID:** `2602.02572` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Haichuan Wang, Tao Lin, Lingkai Kong, et al.
+- **abs:** https://arxiv.org/abs/2602.02572 · **PDF:** https://arxiv.org/pdf/2602.02572v1
+
+## 17277. How Far Are LLMs from Professional Poker Players? Revisiting Game-Theoretic Reasoning with Agentic Tool Use
+
+- **ID:** `2602.00528` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Minhua Lin, Enyan Dai, Hui Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.00528 · **PDF:** https://arxiv.org/pdf/2602.00528v1
+
+## 17278. Physiology as Language: Translating Respiration to Sleep EEG
+
+- **ID:** `2602.00526` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Kaiwen Zha, Chao Li, Hao He, et al.
+- **abs:** https://arxiv.org/abs/2602.00526 · **PDF:** https://arxiv.org/pdf/2602.00526v1
+
+## 17279. MAU-GPT: Enhancing Multi-type Industrial Anomaly Understanding via Anomaly-aware and Generalist Experts Adaptation
+
+- **ID:** `2602.07011` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Zhuonan Wang, Zhenxuan Fan, Siwen Tan, et al.
+- **abs:** https://arxiv.org/abs/2602.07011 · **PDF:** https://arxiv.org/pdf/2602.07011v2
+
+## 17280. Diagnosing the Reliability of LLM-as-a-Judge via Item Response Theory
+
+- **ID:** `2602.00521` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Junhyuk Choi, Sohhyung Park, Chanhee Cho, et al.
+- **abs:** https://arxiv.org/abs/2602.00521 · **PDF:** https://arxiv.org/pdf/2602.00521v1
+
+## 17281. NEST: Nested Event Stream Transformer for Sequences of Multisets
+
+- **ID:** `2602.00520` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Minghui Sun, Haoyu Gong, Xingyu You, et al.
+- **abs:** https://arxiv.org/abs/2602.00520 · **PDF:** https://arxiv.org/pdf/2602.00520v2
+
+## 17282. Contrastive Learning for Privacy Enhancements in Industrial Internet of Things
+
+- **ID:** `2602.00515` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Lin Liu, Rita Machacy, Simi Kuniyilh
+- **abs:** https://arxiv.org/abs/2602.00515 · **PDF:** https://arxiv.org/pdf/2602.00515v1
+
+## 17283. Minerva: Reinforcement Learning with Verifiable Rewards for Cyber Threat Intelligence LLMs
+
+- **ID:** `2602.00513` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Md Tanvirul Alam, Aritran Piplai, Ionut Cardei, et al.
+- **abs:** https://arxiv.org/abs/2602.00513 · **PDF:** https://arxiv.org/pdf/2602.00513v2
+
+## 17284. Judging the Judges: Human Validation of Multi-LLM Evaluation for High-Quality K--12 Science Instructional Materials
+
+- **ID:** `2602.13243` · **primary:** `cs.CY` · **Submitted:** 2026-01-31
+- **Authors:** Peng He, Zhaohui Li, Zeyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13243 · **PDF:** https://arxiv.org/pdf/2602.13243v1
+
+## 17285. Partition of Unity Neural Networks for Interpretable Classification with Explicit Class Regions
+
+- **ID:** `2602.00511` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Akram Aldroubi
+- **abs:** https://arxiv.org/abs/2602.00511 · **PDF:** https://arxiv.org/pdf/2602.00511v1
+
+## 17286. PCBSchemaGen: Constraint-Guided Schematic Design via LLM for Printed Circuit Boards (PCB)
+
+- **ID:** `2602.00510` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Huanghaohe Zou, Peng Han, Emad Nazerian, et al.
+- **abs:** https://arxiv.org/abs/2602.00510 · **PDF:** https://arxiv.org/pdf/2602.00510v1
+
+## 17287. Field-Theoretic Memory for AI Agents: Continuous Dynamics for Context Preservation
+
+- **ID:** `2602.21220` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Subhadip Mitra
+- **abs:** https://arxiv.org/abs/2602.21220 · **PDF:** https://arxiv.org/pdf/2602.21220v1
+
+## 17288. Trajectory Consistency for One-Step Generation on Euler Mean Flows
+
+- **ID:** `2602.02571` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Zhiqi Li, Yuchen Sun, Duowen Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.02571 · **PDF:** https://arxiv.org/pdf/2602.02571v1
+
+## 17289. Sparse Shortcuts: Facilitating Efficient Fusion in Multimodal Large Language Models
+
+- **ID:** `2602.00505` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Jingrui Zhang, Feng Liang, Yong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.00505 · **PDF:** https://arxiv.org/pdf/2602.00505v1
+
+## 17290. DECEIVE-AFC: Adversarial Claim Attacks against Search-Enabled LLM-based Fact-Checking Systems
+
+- **ID:** `2602.02569` · **primary:** `cs.CR` · **Submitted:** 2026-01-31
+- **Authors:** Haoran Ou, Kangjie Chen, Gelei Deng, et al.
+- **abs:** https://arxiv.org/abs/2602.02569 · **PDF:** https://arxiv.org/pdf/2602.02569v2
+
+## 17291. Culturally-Grounded Governance for Multilingual Language Models: Rights, Data Boundaries, and Accountable AI Design
+
+- **ID:** `2602.00497` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Hanjing Shi, Dominic DiFranzo
+- **abs:** https://arxiv.org/abs/2602.00497 · **PDF:** https://arxiv.org/pdf/2602.00497v1
+
+## 17292. From Junior to Senior: Allocating Agency and Navigating Professional Growth in Agentic AI-Mediated Software Engineering
+
+- **ID:** `2602.00496` · **primary:** `cs.HC` · **Submitted:** 2026-01-31
+- **Authors:** Dana Feng, Bhada Yun, April Yi Wang
+- **abs:** https://arxiv.org/abs/2602.00496 · **PDF:** https://arxiv.org/pdf/2602.00496v2
+
+## 17293. From Knowledge to Inference: Scaling Laws of Specialized Reasoning on GlobalHealthAtlas
+
+- **ID:** `2602.00491` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Zhaokun Yan, Zhaohan Liu, Wuzheng Dong, et al.
+- **abs:** https://arxiv.org/abs/2602.00491 · **PDF:** https://arxiv.org/pdf/2602.00491v1
+
+## 17294. OD-DEAL: Dynamic Expert-Guided Adversarial Learning with Online Decomposition for Scalable Capacitated Vehicle Routing
+
+- **ID:** `2602.00488` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Dongbin Jiao, Zisheng Chen, Xianyi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00488 · **PDF:** https://arxiv.org/pdf/2602.00488v2
+
+## 17295. Replacing Parameters with Preferences: Federated Alignment of Heterogeneous Vision-Language Models
+
+- **ID:** `2602.00485` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Shule Lu, Yujing Wang, Hainan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.00485 · **PDF:** https://arxiv.org/pdf/2602.00485v2
+
+## 17296. AREAL-DTA: Dynamic Tree Attention for Efficient Reinforcement Learning of Large Language Models
+
+- **ID:** `2602.00482` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Jiarui Zhang, Yuchen Yang, Ran Yan, et al.
+- **abs:** https://arxiv.org/abs/2602.00482 · **PDF:** https://arxiv.org/pdf/2602.00482v1
+
+## 17297. Quality-Diversity Optimization as Multi-Objective Optimization
+
+- **ID:** `2602.00478` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Xi Lin, Ping Guo, Yilu Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.00478 · **PDF:** https://arxiv.org/pdf/2602.00478v1
+
+## 17298. Intention-Adaptive LLM Fine-Tuning for Text Revision Generation
+
+- **ID:** `2602.00477` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Zhexiong Liu, Diane Litman
+- **abs:** https://arxiv.org/abs/2602.00477 · **PDF:** https://arxiv.org/pdf/2602.00477v2
+
+## 17299. Diffusion LMs Can Approximate Optimal Infilling Lengths Implicitly
+
+- **ID:** `2602.00476` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Hengchang Liu, Zhao Yang, Bing Su
+- **abs:** https://arxiv.org/abs/2602.00476 · **PDF:** https://arxiv.org/pdf/2602.00476v1
+
+## 17300. Parallel Stochastic Gradient-Based Planning for World Models
+
+- **ID:** `2602.00475` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Michael Psenka, Michael Rabbat, Aditi Krishnapriyan, et al.
+- **abs:** https://arxiv.org/abs/2602.00475 · **PDF:** https://arxiv.org/pdf/2602.00475v1
+
+## 17301. Stabilizing Fixed-Point Iteration for Markov Chain Poisson Equations
+
+- **ID:** `2602.00474` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Yang Xu, Vaneet Aggarwal
+- **abs:** https://arxiv.org/abs/2602.00474 · **PDF:** https://arxiv.org/pdf/2602.00474v1
+
+## 17302. Quantum Phase Recognition via Quantum Attention Mechanism
+
+- **ID:** `2602.00473` · **primary:** `quant-ph` · **Submitted:** 2026-01-31
+- **Authors:** Jin-Long Chen, Xin Li, Zhang-Qi Yin
+- **abs:** https://arxiv.org/abs/2602.00473 · **PDF:** https://arxiv.org/pdf/2602.00473v1
+
+## 17303. Dual Latent Memory for Visual Multi-agent System
+
+- **ID:** `2602.00471` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Xinlei Yu, Chengming Xu, Zhangquan Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.00471 · **PDF:** https://arxiv.org/pdf/2602.00471v1
+
+## 17304. Words that make SENSE: Sensorimotor Norms in Learned Lexical Token Representations
+
+- **ID:** `2602.00469` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Abhinav Gupta, Toben H. Mintz, Jesse Thomason
+- **abs:** https://arxiv.org/abs/2602.00469 · **PDF:** https://arxiv.org/pdf/2602.00469v1
+
+## 17305. PAIR-Former: Budgeted Relational MIL for miRNA Target Prediction
+
+- **ID:** `2602.00465` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Jiaqi Yin, Baiming Chen, Jia Fei, et al.
+- **abs:** https://arxiv.org/abs/2602.00465 · **PDF:** https://arxiv.org/pdf/2602.00465v2
+
+## 17306. LatentLens: Revealing Highly Interpretable Visual Tokens in LLMs
+
+- **ID:** `2602.00462` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Benno Krojer, Shravan Nayak, Oscar Mañas, et al.
+- **abs:** https://arxiv.org/abs/2602.00462 · **PDF:** https://arxiv.org/pdf/2602.00462v3
+
+## 17307. Test-Time Adaptation for Tactile-Vision-Language Models
+
+- **ID:** `2602.15873` · **primary:** `cs.RO` · **Submitted:** 2026-01-31
+- **Authors:** Chuyang Ye, Haoxian Jing, Qinting Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.15873 · **PDF:** https://arxiv.org/pdf/2602.15873v1
+
+## 17308. Search Inspired Exploration in Reinforcement Learning
+
+- **ID:** `2602.00460` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Georgios Sotirchos, Zlatan Ajanović, Jens Kober
+- **abs:** https://arxiv.org/abs/2602.00460 · **PDF:** https://arxiv.org/pdf/2602.00460v1
+
+## 17309. What Matters to an LLM? Behavioral and Computational Evidences from Summarization
+
+- **ID:** `2602.00459` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Yongxin Zhou, Changshun Wu, Philippe Mulhem, et al.
+- **abs:** https://arxiv.org/abs/2602.00459 · **PDF:** https://arxiv.org/pdf/2602.00459v1
+
+## 17310. LatentTrack: Sequential Weight Generation via Latent Filtering
+
+- **ID:** `2602.00458` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Omer Haq
+- **abs:** https://arxiv.org/abs/2602.00458 · **PDF:** https://arxiv.org/pdf/2602.00458v1
+
+## 17311. Benchmarking Agents in Insurance Underwriting Environments
+
+- **ID:** `2602.00456` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Amanda Dsouza, Ramya Ramakrishnan, Charles Dickens, et al.
+- **abs:** https://arxiv.org/abs/2602.00456 · **PDF:** https://arxiv.org/pdf/2602.00456v1
+
+## 17312. Cross-Modal Memory Compression for Efficient Multi-Agent Debate
+
+- **ID:** `2602.00454` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Jing Wu, Yue Sun, Tianpei Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.00454 · **PDF:** https://arxiv.org/pdf/2602.00454v1
+
+## 17313. FedMOA: Federated GRPO for Personalized Reasoning LLMs under Heterogeneous Rewards
+
+- **ID:** `2602.00453` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Ziyao Wang, Daeun Jung, Yexiao He, et al.
+- **abs:** https://arxiv.org/abs/2602.00453 · **PDF:** https://arxiv.org/pdf/2602.00453v1
+
+## 17314. Stabilizing Decentralized Federated Fine-Tuning via Topology-Aware Alternating LoRA
+
+- **ID:** `2602.00451` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Xiaoyu Wang, Xiaotian Li, Zhixiang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.00451 · **PDF:** https://arxiv.org/pdf/2602.00451v1
+
+## 17315. Reasoning-Based Personalized Generation for Users with Sparse Data
+
+- **ID:** `2602.21219` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Bo Ni, Branislav Kveton, Samyadeep Basu, et al.
+- **abs:** https://arxiv.org/abs/2602.21219 · **PDF:** https://arxiv.org/pdf/2602.21219v1
+
+## 17316. Mitigating Task-Order Sensitivity and Forgetting via Hierarchical Second-Order Consolidation
+
+- **ID:** `2602.02568` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Protik Nag, Krishnan Raghavan, Vignesh Narayanan
+- **abs:** https://arxiv.org/abs/2602.02568 · **PDF:** https://arxiv.org/pdf/2602.02568v1
+
+## 17317. Do Latent-CoT Models Think Step-by-Step? A Mechanistic Study on Sequential Reasoning Tasks
+
+- **ID:** `2602.00449` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Jia Liang, Liangming Pan
+- **abs:** https://arxiv.org/abs/2602.00449 · **PDF:** https://arxiv.org/pdf/2602.00449v1
+
+## 17318. Towards Building Non-Fine-Tunable Foundation Models
+
+- **ID:** `2602.00446` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Ziyao Wang, Nizhang Li, Pingzhi Li, et al.
+- **abs:** https://arxiv.org/abs/2602.00446 · **PDF:** https://arxiv.org/pdf/2602.00446v1
+
+## 17319. IceBench-S2S: A Benchmark of Deep Learning for Challenging Subseasonal-to-Seasonal Daily Arctic Sea Ice Forecasting in Deep Latent Space
+
+- **ID:** `2602.02567` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Jingyi Xu, Shengnan Wang, Weidong Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.02567 · **PDF:** https://arxiv.org/pdf/2602.02567v1
+
+## 17320. DISK: Dynamic Inference SKipping for World Models
+
+- **ID:** `2602.00440` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Anugunj Naman, Gaibo Zhang, Ayushman Singh, et al.
+- **abs:** https://arxiv.org/abs/2602.00440 · **PDF:** https://arxiv.org/pdf/2602.00440v1
+
+## 17321. Exact Instance Compression for Convex Empirical Risk Minimization via Color Refinement
+
+- **ID:** `2602.00437` · **primary:** `math.OC` · **Submitted:** 2026-01-31
+- **Authors:** Bryan Zhu, Ziang Chen
+- **abs:** https://arxiv.org/abs/2602.00437 · **PDF:** https://arxiv.org/pdf/2602.00437v1
+
+## 17322. GRACE: an Agentic AI for Particle Physics Experiment Design and Simulation
+
+- **ID:** `2602.15039` · **primary:** `hep-ex` · **Submitted:** 2026-01-31
+- **Authors:** Justin Hill, Hong Joo Ryoo
+- **abs:** https://arxiv.org/abs/2602.15039 · **PDF:** https://arxiv.org/pdf/2602.15039v1
+
+## 17323. When Agents "Misremember" Collectively: Exploring the Mandela Effect in LLM-based Multi-Agent Systems
+
+- **ID:** `2602.00428` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Naen Xu, Hengyu An, Shuo Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.00428 · **PDF:** https://arxiv.org/pdf/2602.00428v2
+
+## 17324. Topological Residual Asymmetry for Bivariate Causal Direction
+
+- **ID:** `2602.00427` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Mouad El Bouchattaoui
+- **abs:** https://arxiv.org/abs/2602.00427 · **PDF:** https://arxiv.org/pdf/2602.00427v1
+
+## 17325. LLMs as High-Dimensional Nonlinear Autoregressive Models with Attention: Training, Alignment and Inference
+
+- **ID:** `2602.00426` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Vikram Krishnamurthy
+- **abs:** https://arxiv.org/abs/2602.00426 · **PDF:** https://arxiv.org/pdf/2602.00426v1
+
+## 17326. Segment-Level Attribution for Selective Learning of Long Reasoning Traces
+
+- **ID:** `2602.00425` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Siyuan Wang, Yanchen Liu, Xiang Ren
+- **abs:** https://arxiv.org/abs/2602.00425 · **PDF:** https://arxiv.org/pdf/2602.00425v1
+
+## 17327. Open Materials Generation with Inference-Time Reinforcement Learning
+
+- **ID:** `2602.00424` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Philipp Hoellmer, Stefano Martiniani
+- **abs:** https://arxiv.org/abs/2602.00424 · **PDF:** https://arxiv.org/pdf/2602.00424v1
+
+## 17328. Federated-inspired Single-cell Batch Integration in Latent Space
+
+- **ID:** `2602.00423` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Quang-Huy Nguyen, Zongliang Yue, Hao Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.00423 · **PDF:** https://arxiv.org/pdf/2602.00423v2
+
+## 17329. Text is All You Need for Vision-Language Model Jailbreaking
+
+- **ID:** `2602.00420` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Yihang Chen, Zhao Xu, Youyuan Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.00420 · **PDF:** https://arxiv.org/pdf/2602.00420v1
+
+## 17330. Shuffle and Joint Differential Privacy for Generalized Linear Contextual Bandits
+
+- **ID:** `2602.00417` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Sahasrajit Sarmasarkar
+- **abs:** https://arxiv.org/abs/2602.00417 · **PDF:** https://arxiv.org/pdf/2602.00417v1
+
+## 17331. PolarMem: A Training-Free Polarized Latent Graph Memory for Verifiable Multimodal Agents
+
+- **ID:** `2602.00415` · **primary:** `cs.AI` · **Submitted:** 2026-01-31
+- **Authors:** Zhisheng Chen, Tingyu Wu, Zijie Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.00415 · **PDF:** https://arxiv.org/pdf/2602.00415v1
+
+## 17332. Toward Autonomous Laboratory Safety Monitoring with Vision Language Models: Learning to See Hazards Through Scene Structure
+
+- **ID:** `2602.00414` · **primary:** `cs.CV` · **Submitted:** 2026-01-31
+- **Authors:** Trishna Chakraborty, Udita Ghosh, Aldair Ernesto Gongora, et al.
+- **abs:** https://arxiv.org/abs/2602.00414 · **PDF:** https://arxiv.org/pdf/2602.00414v1
+
+## 17333. Alignment of Diffusion Model and Flow Matching for Text-to-Image Generation
+
+- **ID:** `2602.00413` · **primary:** `stat.ML` · **Submitted:** 2026-01-31
+- **Authors:** Yidong Ouyang, Liyan Xie, Hongyuan Zha, et al.
+- **abs:** https://arxiv.org/abs/2602.00413 · **PDF:** https://arxiv.org/pdf/2602.00413v1
+
+## 17334. Robustness of AutoML on Dirty Categorical Data
+
+- **ID:** `2602.00412` · **primary:** `cs.LG` · **Submitted:** 2026-01-31
+- **Authors:** Marcos L. P. Bueno, Joaquin Vanschoren
+- **abs:** https://arxiv.org/abs/2602.00412 · **PDF:** https://arxiv.org/pdf/2602.00412v1
+
+## 17335. EPSVec: Efficient and Private Synthetic Data Generation via Dataset Vectors
+
+- **ID:** `2602.21218` · **primary:** `cs.CL` · **Submitted:** 2026-01-31
+- **Authors:** Amin Banayeeanzade, Qingchuan Yang, Deqing Fu, et al.
+- **abs:** https://arxiv.org/abs/2602.21218 · **PDF:** https://arxiv.org/pdf/2602.21218v1
+
+## 17336. Variational Approach for Job Shop Scheduling
+
+- **ID:** `2602.00408` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Seung Heon Oh, Jiwon Baek, Ki Young Cho, et al.
+- **abs:** https://arxiv.org/abs/2602.00408 · **PDF:** https://arxiv.org/pdf/2602.00408v2
+
+## 17337. Fed-Listing: Federated Label Distribution Inference in Graph Neural Networks
+
+- **ID:** `2602.00407` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Suprim Nakarmi, Junggab Son, Yue Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.00407 · **PDF:** https://arxiv.org/pdf/2602.00407v1
+
+## 17338. RobustDebias: Debiasing Language Models using Distributionally Robust Optimization
+
+- **ID:** `2602.00405` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Deep Gandhi, Katyani Singh, Nidhi Hegde
+- **abs:** https://arxiv.org/abs/2602.00405 · **PDF:** https://arxiv.org/pdf/2602.00405v1
+
+## 17339. DROGO: Default Representation Objective via Graph Optimization in Reinforcement Learning
+
+- **ID:** `2602.00403` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Hon Tik Tse, Marlos C. Machado
+- **abs:** https://arxiv.org/abs/2602.00403 · **PDF:** https://arxiv.org/pdf/2602.00403v1
+
+## 17340. A Conditional Companion: Lived Experiences of People with Mental Health Disorders Using LLMs
+
+- **ID:** `2602.00402` · **primary:** `cs.HC` · **Submitted:** 2026-01-30
+- **Authors:** Aditya Kumar Purohit, Hendrik Heuer
+- **abs:** https://arxiv.org/abs/2602.00402 · **PDF:** https://arxiv.org/pdf/2602.00402v2
+
+## 17341. ZEST: Zero-shot Embodied Skill Transfer for Athletic Robot Control
+
+- **ID:** `2602.00401` · **primary:** `cs.RO` · **Submitted:** 2026-01-30
+- **Authors:** Jean Pierre Sleiman, He Li, Alphonsus Adu-Bredu, et al.
+- **abs:** https://arxiv.org/abs/2602.00401 · **PDF:** https://arxiv.org/pdf/2602.00401v1
+
+## 17342. KEPO: Knowledge-Enhanced Preference Optimization for Reinforcement Learning with Reasoning
+
+- **ID:** `2602.00400` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Fan Yang, Rui Meng, Trudi Di Qi, et al.
+- **abs:** https://arxiv.org/abs/2602.00400 · **PDF:** https://arxiv.org/pdf/2602.00400v1
+
+## 17343. Reinforcement Learning for Control Systems with Time Delays: A Comprehensive Survey
+
+- **ID:** `2602.00399` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Armando Alves Neto
+- **abs:** https://arxiv.org/abs/2602.00399 · **PDF:** https://arxiv.org/pdf/2602.00399v1
+
+## 17344. MemoryLLM: Plug-n-Play Interpretable Feed-Forward Memory for Transformers
+
+- **ID:** `2602.00398` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Ajay Jaiswal, Lauren Hannah, Han-Byul Kim, et al.
+- **abs:** https://arxiv.org/abs/2602.00398 · **PDF:** https://arxiv.org/pdf/2602.00398v1
+
+## 17345. Unveiling the Potential of Quantization with MXFP4: Strategies for Quantization Error Reduction
+
+- **ID:** `2603.08713` · **primary:** `cs.AR` · **Submitted:** 2026-01-30
+- **Authors:** Jatin Chhugani, Geonhwa Jeong, Bor-Yiing Su, et al.
+- **abs:** https://arxiv.org/abs/2603.08713 · **PDF:** https://arxiv.org/pdf/2603.08713v1
+
+## 17346. Fast Forward: Accelerating LLM Prefill with Predictive FFN Sparsity
+
+- **ID:** `2602.00397` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Aayush Gautam, Mukul Gagrani, Junyoung Park, et al.
+- **abs:** https://arxiv.org/abs/2602.00397 · **PDF:** https://arxiv.org/pdf/2602.00397v1
+
+## 17347. 3DGS$^2$-TR: Scalable Second-Order Trust-Region Method for 3D Gaussian Splatting
+
+- **ID:** `2602.00395` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Roger Hsiao, Yuchen Fang, Xiangru Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.00395 · **PDF:** https://arxiv.org/pdf/2602.00395v1
+
+## 17348. Applied Sociolinguistic AI for Community Development (ASA-CD): A New Scientific Paradigm for Linguistically-Grounded Social Intervention
+
+- **ID:** `2602.21217` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** S M Ruhul Alam, Rifa Ferzana
+- **abs:** https://arxiv.org/abs/2602.21217 · **PDF:** https://arxiv.org/pdf/2602.21217v1
+
+## 17349. Brazilian Portuguese Image Captioning with Transformers: A Study on Cross-Native-Translated Dataset
+
+- **ID:** `2602.00393` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Gabriel Bromonschenkel, Alessandro L. Koerich, Thiago M. Paixão, et al.
+- **abs:** https://arxiv.org/abs/2602.00393 · **PDF:** https://arxiv.org/pdf/2602.00393v1
+
+## 17350. Localized, High-resolution Geographic Representations with Slepian Functions
+
+- **ID:** `2602.00392` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Arjun Rao, Ruth Crasto, Tessa Ooms, et al.
+- **abs:** https://arxiv.org/abs/2602.00392 · **PDF:** https://arxiv.org/pdf/2602.00392v1
+
+## 17351. Hybrid operator learning of wave scattering maps in high-contrast media
+
+- **ID:** `2602.11197` · **primary:** `eess.SP` · **Submitted:** 2026-01-30
+- **Authors:** Advait Balaji, Trevor Teolis, S. David Mis, et al.
+- **abs:** https://arxiv.org/abs/2602.11197 · **PDF:** https://arxiv.org/pdf/2602.11197v1
+
+## 17352. Safer by Diffusion, Broken by Context: Diffusion LLM's Safety Blessing and Its Failure Mode
+
+- **ID:** `2602.00388` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Zeyuan He, Yupeng Chen, Lang Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.00388 · **PDF:** https://arxiv.org/pdf/2602.00388v2
+
+## 17353. Singular Bayesian Neural Networks
+
+- **ID:** `2602.00387` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Mame Diarra Toure, David A. Stephens
+- **abs:** https://arxiv.org/abs/2602.00387 · **PDF:** https://arxiv.org/pdf/2602.00387v2
+
+## 17354. Generalized Inverses of Matrix Products: From Fundamental Subspaces to Randomized Decompositions
+
+- **ID:** `2602.00386` · **primary:** `math.NA` · **Submitted:** 2026-01-30
+- **Authors:** Michał P. Karpowicz, Gilbert Strang
+- **abs:** https://arxiv.org/abs/2602.00386 · **PDF:** https://arxiv.org/pdf/2602.00386v1
+
+## 17355. RePaint-Enhanced Conditional Diffusion Model for Parametric Engineering Designs under Performance and Parameter Constraints
+
+- **ID:** `2602.00384` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Ke Wang, Nguyen Gia Hien Vu, Yifan Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.00384 · **PDF:** https://arxiv.org/pdf/2602.00384v1
+
+## 17356. Modeling Image-Caption Rating from Comparative Judgments
+
+- **ID:** `2602.00381` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Kezia Minni, Qiang Zhang, Monoshiz Mahbub Khan, et al.
+- **abs:** https://arxiv.org/abs/2602.00381 · **PDF:** https://arxiv.org/pdf/2602.00381v2
+
+## 17357. Clause-Internal or Clause-External? Testing Turkish Reflexive Binding in Adapted versus Chain of Thought Large Language Models
+
+- **ID:** `2602.00380` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Sercan Karakaş
+- **abs:** https://arxiv.org/abs/2602.00380 · **PDF:** https://arxiv.org/pdf/2602.00380v2
+
+## 17358. Parametrization of subgrid scales in long-term simulations of the shallow-water equations using machine learning and convex limiting
+
+- **ID:** `2602.00378` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-01-30
+- **Authors:** Md Amran Hossan Mojamder, Zhihang Xu, Min Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00378 · **PDF:** https://arxiv.org/pdf/2602.00378v1
+
+## 17359. DecompressionLM: Deterministic, Diagnostic, and Zero-Shot Concept Graph Extraction from Language Models
+
+- **ID:** `2602.00377` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Zhaochen Hong, Jiaxuan You
+- **abs:** https://arxiv.org/abs/2602.00377 · **PDF:** https://arxiv.org/pdf/2602.00377v2
+
+## 17360. MATRIX: A Multimodal Benchmark and Post-Training Framework for Materials Science
+
+- **ID:** `2602.00376` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Delia McGrath, Curtis Chong, Rohil Kulkarni, et al.
+- **abs:** https://arxiv.org/abs/2602.00376 · **PDF:** https://arxiv.org/pdf/2602.00376v1
+
+## 17361. Post-Training Probability Manifold Correction via Structured SVD Pruning and Self-Referential Distillation
+
+- **ID:** `2602.00372` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Aaron R. Flouro, Shawn P. Chadwick
+- **abs:** https://arxiv.org/abs/2602.00372 · **PDF:** https://arxiv.org/pdf/2602.00372v1
+
+## 17362. POET: Protocol Optimization via Eligibility Tuning
+
+- **ID:** `2602.00370` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Trisha Das, Katherine Kero, Dorinda Schumann, et al.
+- **abs:** https://arxiv.org/abs/2602.00370 · **PDF:** https://arxiv.org/pdf/2602.00370v1
+
+## 17363. Quantum Generator Kernels
+
+- **ID:** `2602.00361` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Philipp Altmann, Maximilian Mansky, Maximilian Zorn, et al.
+- **abs:** https://arxiv.org/abs/2602.00361 · **PDF:** https://arxiv.org/pdf/2602.00361v1
+
+## 17364. Leveraging Textual-Cues for Enhancing Multimodal Sentiment Analysis by Object Recognition
+
+- **ID:** `2602.00360` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Sumana Biswas, Karen Young, Josephine Griffith
+- **abs:** https://arxiv.org/abs/2602.00360 · **PDF:** https://arxiv.org/pdf/2602.00360v1
+
+## 17365. Position: Agentic Evolution is the Path to Evolving LLMs
+
+- **ID:** `2602.00359` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Minhua Lin, Hanqing Lu, Zhan Shi, et al.
+- **abs:** https://arxiv.org/abs/2602.00359 · **PDF:** https://arxiv.org/pdf/2602.00359v2
+
+## 17366. Planning with Language and Generative Models: Toward General Reward-Guided Wireless Network Design
+
+- **ID:** `2602.00357` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Chenyang Yuan, Xiaoyuan Cheng
+- **abs:** https://arxiv.org/abs/2602.00357 · **PDF:** https://arxiv.org/pdf/2602.00357v1
+
+## 17367. MHDash: An Online Platform for Benchmarking Mental Health-Aware AI Assistants
+
+- **ID:** `2602.00353` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Yihe Zhang, Cheyenne N Mohawk, Kaiying Han, et al.
+- **abs:** https://arxiv.org/abs/2602.00353 · **PDF:** https://arxiv.org/pdf/2602.00353v1
+
+## 17368. DETOUR: An Interactive Benchmark for Dual-Agent Search and Reasoning
+
+- **ID:** `2602.00352` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Li Siyan, Darshan Deshpande, Anand Kannappan, et al.
+- **abs:** https://arxiv.org/abs/2602.00352 · **PDF:** https://arxiv.org/pdf/2602.00352v1
+
+## 17369. Learning Alzheimer's Disease Signatures by bridging EEG with Spiking Neural Networks and Biophysical Simulations
+
+- **ID:** `2602.07010` · **primary:** `cs.NE` · **Submitted:** 2026-01-30
+- **Authors:** Szymon Mamoń, Max Talanov, Alessandro Crimi
+- **abs:** https://arxiv.org/abs/2602.07010 · **PDF:** https://arxiv.org/pdf/2602.07010v1
+
+## 17370. AdaFuse: Adaptive Multimodal Fusion for Lung Cancer Risk Prediction via Reinforcement Learning
+
+- **ID:** `2602.00347` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Chongyu Qu, Zhengyi Lu, Yuxiang Lai, et al.
+- **abs:** https://arxiv.org/abs/2602.00347 · **PDF:** https://arxiv.org/pdf/2602.00347v1
+
+## 17371. When RAG Hurts: Diagnosing and Mitigating Attention Distraction in Retrieval-Augmented LVLMs
+
+- **ID:** `2602.00344` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Beidi Zhao, Wenlong Deng, Xinting Liao, et al.
+- **abs:** https://arxiv.org/abs/2602.00344 · **PDF:** https://arxiv.org/pdf/2602.00344v1
+
+## 17372. Standardized Methods and Recommendations for Green Federated Learning
+
+- **ID:** `2602.00343` · **primary:** `cs.DC` · **Submitted:** 2026-01-30
+- **Authors:** Austin Tapp, Holger R. Roth, Ziyue Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.00343 · **PDF:** https://arxiv.org/pdf/2602.00343v1
+
+## 17373. Bridging the Semantic Chasm: Synergistic Conceptual Anchoring for Generalized Few-Shot and Zero-Shot OOD Perception
+
+- **ID:** `2602.00340` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Alexandros Christoforos, Sarah Jenkins, Michael Brown, et al.
+- **abs:** https://arxiv.org/abs/2602.00340 · **PDF:** https://arxiv.org/pdf/2602.00340v1
+
+## 17374. Adaptive Momentum and Nonlinear Damping for Neural Network Training
+
+- **ID:** `2602.00334` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Aikaterini Karoni, Rajit Rajpal, Benedict Leimkuhler, et al.
+- **abs:** https://arxiv.org/abs/2602.00334 · **PDF:** https://arxiv.org/pdf/2602.00334v1
+
+## 17375. Efficient and accurate steering of Large Language Models through attention-guided feature learning
+
+- **ID:** `2602.00333` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Parmida Davarmanesh, Ashia Wilson, Adityanarayanan Radhakrishnan
+- **abs:** https://arxiv.org/abs/2602.00333 · **PDF:** https://arxiv.org/pdf/2602.00333v1
+
+## 17376. Prototype-based Explainable Neural Networks with Channel-specific Reasoning for Geospatial Learning Tasks
+
+- **ID:** `2602.00331` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Anushka Narayanan, Karianne J. Bergen
+- **abs:** https://arxiv.org/abs/2602.00331 · **PDF:** https://arxiv.org/pdf/2602.00331v1
+
+## 17377. In-Run Data Shapley for Adam Optimizer
+
+- **ID:** `2602.00329` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Meng Ding, Zeqing Zhang, Di Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00329 · **PDF:** https://arxiv.org/pdf/2602.00329v3
+
+## 17378. Harvest: Opportunistic Peer-to-Peer GPU Caching for LLM Inference
+
+- **ID:** `2602.00328` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Nikhil Gopal, Kostis Kaffes
+- **abs:** https://arxiv.org/abs/2602.00328 · **PDF:** https://arxiv.org/pdf/2602.00328v1
+
+## 17379. SayNext-Bench: Why Do LLMs Struggle with Next-Utterance Prediction?
+
+- **ID:** `2602.00327` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Yueyi Yang, Haotian Liu, Fang Kang, et al.
+- **abs:** https://arxiv.org/abs/2602.00327 · **PDF:** https://arxiv.org/pdf/2602.00327v1
+
+## 17380. Detecting AI-Generated Content in Academic Peer Reviews
+
+- **ID:** `2602.00319` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Siyuan Shen, Kai Wang
+- **abs:** https://arxiv.org/abs/2602.00319 · **PDF:** https://arxiv.org/pdf/2602.00319v2
+
+## 17381. Optimal Transport-Guided Adversarial Attacks on Graph Neural Network-Based Bot Detection
+
+- **ID:** `2602.00318` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Kunal Mukherjee, Zulfikar Alom, Tran Gia Bao Ngo, et al.
+- **abs:** https://arxiv.org/abs/2602.00318 · **PDF:** https://arxiv.org/pdf/2602.00318v1
+
+## 17382. MiNER: A Two-Stage Pipeline for Metadata Extraction from Municipal Meeting Minutes
+
+- **ID:** `2602.00316` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Rodrigo Batista, Luís Filipe Cunha, Purificação Silvano, et al.
+- **abs:** https://arxiv.org/abs/2602.00316 · **PDF:** https://arxiv.org/pdf/2602.00316v3
+
+## 17383. Beyond the Loss Curve: Scaling Laws, Active Learning, and the Limits of Learning from Exact Posteriors
+
+- **ID:** `2602.00315` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Arian Khorasani, Nathaniel Chen, Yug D Oswal, et al.
+- **abs:** https://arxiv.org/abs/2602.00315 · **PDF:** https://arxiv.org/pdf/2602.00315v2
+
+## 17384. Opportunistic Promptable Segmentation: Leveraging Routine Radiological Annotations to Guide 3D CT Lesion Segmentation
+
+- **ID:** `2602.00309` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Samuel Church, Joshua D. Warner, Danyal Maqbool, et al.
+- **abs:** https://arxiv.org/abs/2602.00309 · **PDF:** https://arxiv.org/pdf/2602.00309v1
+
+## 17385. Autonomous Data Processing using Meta-Agents
+
+- **ID:** `2602.00307` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Udayan Khurana
+- **abs:** https://arxiv.org/abs/2602.00307 · **PDF:** https://arxiv.org/pdf/2602.00307v2
+
+## 17386. Semantics-Preserving Evasion of LLM Vulnerability Detectors
+
+- **ID:** `2602.00305` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Luze Sun, Alina Oprea, Eric Wong
+- **abs:** https://arxiv.org/abs/2602.00305 · **PDF:** https://arxiv.org/pdf/2602.00305v1
+
+## 17387. Neural Ising Machines via Unrolling and Zeroth-Order Training
+
+- **ID:** `2602.00302` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Sam Reifenstein, Timothee Leleu
+- **abs:** https://arxiv.org/abs/2602.00302 · **PDF:** https://arxiv.org/pdf/2602.00302v1
+
+## 17388. Soft Clustering Anchors for Self-Supervised Speech Representation Learning in Joint Embedding Prediction Architectures
+
+- **ID:** `2602.09040` · **primary:** `eess.AS` · **Submitted:** 2026-01-30
+- **Authors:** Georgios Ioannides, Adrian Kieback, Judah Goldfeder, et al.
+- **abs:** https://arxiv.org/abs/2602.09040 · **PDF:** https://arxiv.org/pdf/2602.09040v1
+
+## 17389. Faithful-Patchscopes: Understanding and Mitigating Model Bias in Hidden Representations Explanation of Large Language Models
+
+- **ID:** `2602.00300` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Xilin Gong, Shu Yang, Zehua Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.00300 · **PDF:** https://arxiv.org/pdf/2602.00300v1
+
+## 17390. Agentic Framework for Epidemiological Modeling
+
+- **ID:** `2602.00299` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Rituparna Datta, Zihan Guan, Baltazar Espinoza, et al.
+- **abs:** https://arxiv.org/abs/2602.00299 · **PDF:** https://arxiv.org/pdf/2602.00299v2
+
+## 17391. Assessing Domain-Level Susceptibility to Emergent Misalignment from Narrow Finetuning
+
+- **ID:** `2602.00298` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Abhishek Mishra, Mugilan Arulvanan, Reshma Ashok, et al.
+- **abs:** https://arxiv.org/abs/2602.00298 · **PDF:** https://arxiv.org/pdf/2602.00298v1
+
+## 17392. From Observations to States: Latent Time Series Forecasting
+
+- **ID:** `2602.00297` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Jie Yang, Yifan Hu, Yuante Li, et al.
+- **abs:** https://arxiv.org/abs/2602.00297 · **PDF:** https://arxiv.org/pdf/2602.00297v1
+
+## 17393. Multi-Speaker Conversational Audio Deepfake: Taxonomy, Dataset and Pilot Study
+
+- **ID:** `2602.00295` · **primary:** `cs.SD` · **Submitted:** 2026-01-30
+- **Authors:** Alabi Ahmed, Vandana Janeja, Sanjay Purushotham
+- **abs:** https://arxiv.org/abs/2602.00295 · **PDF:** https://arxiv.org/pdf/2602.00295v1
+
+## 17394. Self-Attention at Constant Cost per Token via Symmetry-Aware Taylor Approximation
+
+- **ID:** `2602.00294` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Franz A. Heinsen, Leo Kozachkov
+- **abs:** https://arxiv.org/abs/2602.00294 · **PDF:** https://arxiv.org/pdf/2602.00294v1
+
+## 17395. LogicGaze: Benchmarking Causal Consistency in Visual Narratives via Counterfactual Verification
+
+- **ID:** `2602.00292` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Rory Driscoll, Alexandros Christoforos, Chadbourne Davis
+- **abs:** https://arxiv.org/abs/2602.00292 · **PDF:** https://arxiv.org/pdf/2602.00292v1
+
+## 17396. TimeBlind: A Spatio-Temporal Compositionality Benchmark for Video LLMs
+
+- **ID:** `2602.00288` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Baiqi Li, Kangyi Zhao, Ce Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.00288 · **PDF:** https://arxiv.org/pdf/2602.00288v3
+
+## 17397. Generation Order and Parallel Decoding in Masked Diffusion Models: An Information-Theoretic Perspective
+
+- **ID:** `2602.00286` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Shaorong Zhang, Longxuan Yu, Rob Brekelmans, et al.
+- **abs:** https://arxiv.org/abs/2602.00286 · **PDF:** https://arxiv.org/pdf/2602.00286v1
+
+## 17398. EQ-5D Classification Using Biomedical Entity-Enriched Pre-trained Language Models and Multiple Instance Learning
+
+- **ID:** `2602.21216` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Zhyar Rzgar K Rostam, Gábor Kertész
+- **abs:** https://arxiv.org/abs/2602.21216 · **PDF:** https://arxiv.org/pdf/2602.21216v1
+
+## 17399. Sample Complexity Analysis for Constrained Bilevel Reinforcement Learning
+
+- **ID:** `2602.00282` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Naman Saxena, Vaneet Aggarwal
+- **abs:** https://arxiv.org/abs/2602.00282 · **PDF:** https://arxiv.org/pdf/2602.00282v1
+
+## 17400. Benchmarking Uncertainty Calibration in Large Language Model Long-Form Question Answering
+
+- **ID:** `2602.00279` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Philip Müller, Nicholas Popovič, Michael Färber, et al.
+- **abs:** https://arxiv.org/abs/2602.00279 · **PDF:** https://arxiv.org/pdf/2602.00279v1
+
+## 17401. Training LLMs with Fault Tolerant HSDP on 100,000 GPUs
+
+- **ID:** `2602.00277` · **primary:** `cs.DC` · **Submitted:** 2026-01-30
+- **Authors:** Omkar Salpekar, Rohan Varma, Kenny Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.00277 · **PDF:** https://arxiv.org/pdf/2602.00277v1
+
+## 17402. Localizing and Correcting Errors for LLM-based Planners
+
+- **ID:** `2602.00276` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Aditya Kumar, William W. Cohen
+- **abs:** https://arxiv.org/abs/2602.00276 · **PDF:** https://arxiv.org/pdf/2602.00276v2
+
+## 17403. VoxServe: Streaming-Centric Serving System for Speech Language Models
+
+- **ID:** `2602.00269` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Keisuke Kamahori, Wei-Tzu Lee, Atindra Jha, et al.
+- **abs:** https://arxiv.org/abs/2602.00269 · **PDF:** https://arxiv.org/pdf/2602.00269v1
+
+## 17404. TokenTrim: Inference-Time Token Pruning for Autoregressive Long Video Generation
+
+- **ID:** `2602.00268` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Ariel Shaulov, Eitan Shaar, Amit Edenzon, et al.
+- **abs:** https://arxiv.org/abs/2602.00268 · **PDF:** https://arxiv.org/pdf/2602.00268v1
+
+## 17405. PLACID: Identity-Preserving Multi-Object Compositing via Video Diffusion with Synthetic Trajectories
+
+- **ID:** `2602.00267` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Gemma Canet Tarrés, Manel Baradad, Francesc Moreno-Noguer, et al.
+- **abs:** https://arxiv.org/abs/2602.00267 · **PDF:** https://arxiv.org/pdf/2602.00267v1
+
+## 17406. Complete Identification of Deep ReLU Neural Networks by Many-Valued Logic
+
+- **ID:** `2602.00266` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Yani Zhang, Helmut Bölcskei
+- **abs:** https://arxiv.org/abs/2602.00266 · **PDF:** https://arxiv.org/pdf/2602.00266v1
+
+## 17407. A Comparative Simulation Study of the Fairness and Accuracy of Predictive Policing Systems in Baltimore City
+
+- **ID:** `2602.02566` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Samin Semsar, Kiran Laxmikant Prabhu, Gabriella Waters, et al.
+- **abs:** https://arxiv.org/abs/2602.02566 · **PDF:** https://arxiv.org/pdf/2602.02566v1
+
+## 17408. Subspace Clustering on Incomplete Data with Self-Supervised Contrastive Learning
+
+- **ID:** `2602.00262` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Huanran Li, Daniel Pimentel-Alarcón
+- **abs:** https://arxiv.org/abs/2602.00262 · **PDF:** https://arxiv.org/pdf/2602.00262v1
+
+## 17409. Intelligent Reasoning Cues: A Framework and Case Study of the Roles of AI Information in Complex Decisions
+
+- **ID:** `2602.00259` · **primary:** `cs.HC` · **Submitted:** 2026-01-30
+- **Authors:** Venkatesh Sivaraman, Eric P. Mason, Mengfan Ellen Li, et al.
+- **abs:** https://arxiv.org/abs/2602.00259 · **PDF:** https://arxiv.org/pdf/2602.00259v1
+
+## 17410. TABES: Trajectory-Aware Backward-on-Entropy Steering for Masked Diffusion Models
+
+- **ID:** `2602.00250` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Shreshth Saini, Avinab Saha, Balu Adsumilli, et al.
+- **abs:** https://arxiv.org/abs/2602.00250 · **PDF:** https://arxiv.org/pdf/2602.00250v2
+
+## 17411. SANEval: Open-Vocabulary Compositional Benchmarks with Failure-mode Diagnosis
+
+- **ID:** `2602.00249` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Rishav Pramanik, Ian E. Nielsen, Jeff Smith, et al.
+- **abs:** https://arxiv.org/abs/2602.00249 · **PDF:** https://arxiv.org/pdf/2602.00249v1
+
+## 17412. CAPA: Contribution-Aware Pruning and FFN Approximation for Efficient Large Vision-Language Models
+
+- **ID:** `2602.00247` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Samyak Jha, Junho Kim
+- **abs:** https://arxiv.org/abs/2602.00247 · **PDF:** https://arxiv.org/pdf/2602.00247v1
+
+## 17413. High Rank Matrix Completion via Grassmannian Proxy Fusion
+
+- **ID:** `2602.02565` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Huanran Li, Jeremy Johnson, Daniel Pimentel-Alarcón
+- **abs:** https://arxiv.org/abs/2602.02565 · **PDF:** https://arxiv.org/pdf/2602.02565v1
+
+## 17414. Green-NAS: A Global-Scale Multi-Objective Neural Architecture Search for Robust and Efficient Edge-Native Weather Forecasting
+
+- **ID:** `2602.00240` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Md Muhtasim Munif Fahim, Soyda Humyra Yesmin, Saiful Islam, et al.
+- **abs:** https://arxiv.org/abs/2602.00240 · **PDF:** https://arxiv.org/pdf/2602.00240v2
+
+## 17415. DIVERGE: Diversity-Enhanced RAG for Open-Ended Information Seeking
+
+- **ID:** `2602.00238` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Tianyi Hu, Niket Tandon, Akhil Arora
+- **abs:** https://arxiv.org/abs/2602.00238 · **PDF:** https://arxiv.org/pdf/2602.00238v1
+
+## 17416. VideoGPA: Distilling Geometry Priors for 3D-Consistent Video Generation
+
+- **ID:** `2601.23286` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Hongyang Du, Junjie Ye, Xiaoyan Cong, et al.
+- **abs:** https://arxiv.org/abs/2601.23286 · **PDF:** https://arxiv.org/pdf/2601.23286v1
+
+## 17417. End-to-end Optimization of Belief and Policy Learning in Shared Autonomy Paradigms
+
+- **ID:** `2601.23285` · **primary:** `cs.RO` · **Submitted:** 2026-01-30
+- **Authors:** MH Farhadi, Ali Rabiee, Sima Ghafoori, et al.
+- **abs:** https://arxiv.org/abs/2601.23285 · **PDF:** https://arxiv.org/pdf/2601.23285v1
+
+## 17418. Label Curation Using Agentic AI
+
+- **ID:** `2602.02564` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Subhodeep Ghosh, Bayan Divaaniaazar, Md Ishat-E-Rabban, et al.
+- **abs:** https://arxiv.org/abs/2602.02564 · **PDF:** https://arxiv.org/pdf/2602.02564v1
+
+## 17419. Decoupled Diffusion Sampling for Inverse Problems on Function Spaces
+
+- **ID:** `2601.23280` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Thomas Y. L. Lin, Jiachen Yao, Lufang Chiang, et al.
+- **abs:** https://arxiv.org/abs/2601.23280 · **PDF:** https://arxiv.org/pdf/2601.23280v3
+
+## 17420. FOCUS: DLLMs Know How to Tame Their Compute Bound
+
+- **ID:** `2601.23278` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Kaihua Liang, Xin Tan, An Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.23278 · **PDF:** https://arxiv.org/pdf/2601.23278v1
+
+## 17421. Denoising the Deep Sky: Physics-Based CCD Noise Formation for Astronomical Imaging
+
+- **ID:** `2601.23276` · **primary:** `astro-ph.IM` · **Submitted:** 2026-01-30
+- **Authors:** Shuhong Liu, Xining Ge, Ziying Gu, et al.
+- **abs:** https://arxiv.org/abs/2601.23276 · **PDF:** https://arxiv.org/pdf/2601.23276v3
+
+## 17422. UPA: Unsupervised Prompt Agent via Tree-Based Search and Selection
+
+- **ID:** `2601.23273` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Siran Peng, Weisong Zhao, Tianyu Fu, et al.
+- **abs:** https://arxiv.org/abs/2601.23273 · **PDF:** https://arxiv.org/pdf/2601.23273v1
+
+## 17423. IRL-DAL: Safe and Adaptive Trajectory Planning for Autonomous Driving via Energy-Guided Diffusion Models
+
+- **ID:** `2601.23266` · **primary:** `cs.RO` · **Submitted:** 2026-01-30
+- **Authors:** Seyed Ahmad Hosseini Miangoleh, Amin Jalal Aghdasian, Farzaneh Abdollahi
+- **abs:** https://arxiv.org/abs/2601.23266 · **PDF:** https://arxiv.org/pdf/2601.23266v1
+
+## 17424. PaperBanana: Automating Academic Illustration for AI Scientists
+
+- **ID:** `2601.23265` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Dawei Zhu, Rui Meng, Yale Song, et al.
+- **abs:** https://arxiv.org/abs/2601.23265 · **PDF:** https://arxiv.org/pdf/2601.23265v2
+
+## 17425. Particle-Guided Diffusion Models for Partial Differential Equations
+
+- **ID:** `2601.23262` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Andrew Millard, Fredrik Lindsten, Zheng Zhao
+- **abs:** https://arxiv.org/abs/2601.23262 · **PDF:** https://arxiv.org/pdf/2601.23262v1
+
+## 17426. TEON: Tensorized Orthonormalization Beyond Layer-Wise Muon for Large Language Model Pre-Training
+
+- **ID:** `2601.23261` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Ruijie Zhang, Yequan Zhao, Ziyue Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.23261 · **PDF:** https://arxiv.org/pdf/2601.23261v2
+
+## 17427. PaperX: A Unified Framework for Multimodal Academic Presentation Generation with Scholar DAG
+
+- **ID:** `2602.03866` · **primary:** `cs.DL` · **Submitted:** 2026-01-30
+- **Authors:** Tao Yu, Minghui Zhang, Zhiqing Cui, et al.
+- **abs:** https://arxiv.org/abs/2602.03866 · **PDF:** https://arxiv.org/pdf/2602.03866v3
+
+## 17428. Agnostic Language Identification and Generation
+
+- **ID:** `2601.23258` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Mikael Møller Høgsgaard, Chirag Pabbaraju
+- **abs:** https://arxiv.org/abs/2601.23258 · **PDF:** https://arxiv.org/pdf/2601.23258v1
+
+## 17429. Now You Hear Me: Audio Narrative Attacks Against Large Audio-Language Models
+
+- **ID:** `2601.23255` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Ye Yu, Haibo Jin, Yaoning Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.23255 · **PDF:** https://arxiv.org/pdf/2601.23255v1
+
+## 17430. Training-Free Test-Time Adaptation with Brownian Distance Covariance in Vision-Language Models
+
+- **ID:** `2601.23253` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Yi Zhang, Chun-Wun Cheng, Angelica I. Aviles-Rivero, et al.
+- **abs:** https://arxiv.org/abs/2601.23253 · **PDF:** https://arxiv.org/pdf/2601.23253v1
+
+## 17431. Nested Slice Sampling: Vectorized Nested Sampling for GPU-Accelerated Inference
+
+- **ID:** `2601.23252` · **primary:** `stat.CO` · **Submitted:** 2026-01-30
+- **Authors:** David Yallup, Namu Kroupa, Will Handley
+- **abs:** https://arxiv.org/abs/2601.23252 · **PDF:** https://arxiv.org/pdf/2601.23252v1
+
+## 17432. Graph Attention Network for Node Regression on Random Geometric Graphs with Erdős--Rényi contamination
+
+- **ID:** `2601.23239` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Somak Laha, Suqi Liu, Morgane Austern
+- **abs:** https://arxiv.org/abs/2601.23239 · **PDF:** https://arxiv.org/pdf/2601.23239v1
+
+## 17433. How well do generative models solve inverse problems? A benchmark study
+
+- **ID:** `2601.23238` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Patrick Krüger, Patrick Materne, Werner Krebs, et al.
+- **abs:** https://arxiv.org/abs/2601.23238 · **PDF:** https://arxiv.org/pdf/2601.23238v1
+
+## 17434. YuriiFormer: A Suite of Nesterov-Accelerated Transformers
+
+- **ID:** `2601.23236` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Aleksandr Zimin, Yury Polyanskiy, Philippe Rigollet
+- **abs:** https://arxiv.org/abs/2601.23236 · **PDF:** https://arxiv.org/pdf/2601.23236v2
+
+## 17435. Real-World Design and Deployment of an Embedded GenAI-powered 9-1-1 Calltaking Training System: Experiences and Lessons Learned
+
+- **ID:** `2602.13241` · **primary:** `cs.CY` · **Submitted:** 2026-01-30
+- **Authors:** Zirong Chen, Meiyi Ma
+- **abs:** https://arxiv.org/abs/2602.13241 · **PDF:** https://arxiv.org/pdf/2602.13241v1
+
+## 17436. Sequence Diffusion Model for Temporal Link Prediction in Continuous-Time Dynamic Graph
+
+- **ID:** `2601.23233` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Nguyen Minh Duc, Viet Cuong Ta
+- **abs:** https://arxiv.org/abs/2601.23233 · **PDF:** https://arxiv.org/pdf/2601.23233v1
+
+## 17437. ShotFinder: Imagination-Driven Open-Domain Video Shot Retrieval via Web Search
+
+- **ID:** `2601.23232` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Tao Yu, Haopeng Jin, Hao Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.23232 · **PDF:** https://arxiv.org/pdf/2601.23232v3
+
+## 17438. Solving Inverse Problems with Flow-based Models via Model Predictive Control
+
+- **ID:** `2601.23231` · **primary:** `eess.IV` · **Submitted:** 2026-01-30
+- **Authors:** George Webber, Alexander Denker, Riccardo Barbano, et al.
+- **abs:** https://arxiv.org/abs/2601.23231 · **PDF:** https://arxiv.org/pdf/2601.23231v1
+
+## 17439. Strongly Polynomial Time Complexity of Policy Iteration for $L_\infty$ Robust MDPs
+
+- **ID:** `2601.23229` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Ali Asadi, Krishnendu Chatterjee, Ehsan Goharshady, et al.
+- **abs:** https://arxiv.org/abs/2601.23229 · **PDF:** https://arxiv.org/pdf/2601.23229v1
+
+## 17440. Scaling Multiagent Systems with Process Rewards
+
+- **ID:** `2601.23228` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Ed Li, Junyu Ren, Cat Yan
+- **abs:** https://arxiv.org/abs/2601.23228 · **PDF:** https://arxiv.org/pdf/2601.23228v2
+
+## 17441. Agile Reinforcement Learning through Separable Neural Architecture
+
+- **ID:** `2601.23225` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Rajib Mostakim, Reza T. Batley, Sourav Saha
+- **abs:** https://arxiv.org/abs/2601.23225 · **PDF:** https://arxiv.org/pdf/2601.23225v1
+
+## 17442. Are you going to finish that? A Practical Study of the Partial Token Problem
+
+- **ID:** `2601.23223` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Hao Xu, Alisa Liu, Jonathan Hayase, et al.
+- **abs:** https://arxiv.org/abs/2601.23223 · **PDF:** https://arxiv.org/pdf/2601.23223v2
+
+## 17443. Optimal Fair Aggregation of Crowdsourced Noisy Labels using Demographic Parity Constraints
+
+- **ID:** `2601.23221` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Gabriel Singer, Samuel Gruffaz, Olivier Vo Van, et al.
+- **abs:** https://arxiv.org/abs/2601.23221 · **PDF:** https://arxiv.org/pdf/2601.23221v1
+
+## 17444. Med-Scout: Curing MLLMs' Geometric Blindness in Medical Perception via Geometry-Aware RL Post-Training
+
+- **ID:** `2601.23220` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Anglin Liu, Ruichao Chen, Yi Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.23220 · **PDF:** https://arxiv.org/pdf/2601.23220v1
+
+## 17445. MonoScale: Scaling Multi-Agent System with Monotonic Improvement
+
+- **ID:** `2601.23219` · **primary:** `cs.MA` · **Submitted:** 2026-01-30
+- **Authors:** Shuai Shao, Yixiang Liu, Bingwei Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.23219 · **PDF:** https://arxiv.org/pdf/2601.23219v1
+
+## 17446. Tackling air quality with SAPIENS
+
+- **ID:** `2601.23215` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Marcella Bona, Nathan Heatley, Jia-Chen Hua, et al.
+- **abs:** https://arxiv.org/abs/2601.23215 · **PDF:** https://arxiv.org/pdf/2601.23215v1
+
+## 17447. Disentangling multispecific antibody function with graph neural networks
+
+- **ID:** `2601.23212` · **primary:** `q-bio.BM` · **Submitted:** 2026-01-30
+- **Authors:** Joshua Southern, Changpeng Lu, Santrupti Nerli, et al.
+- **abs:** https://arxiv.org/abs/2601.23212 · **PDF:** https://arxiv.org/pdf/2601.23212v1
+
+## 17448. MapDream: Task-Driven Map Learning for Vision-Language Navigation
+
+- **ID:** `2602.00222` · **primary:** `cs.RO` · **Submitted:** 2026-01-30
+- **Authors:** Guoxin Lian, Shuo Wang, Yucheng Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00222 · **PDF:** https://arxiv.org/pdf/2602.00222v2
+
+## 17449. A Random Matrix Theory of Masked Self-Supervised Regression
+
+- **ID:** `2601.23208` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Arie Wortsman Zurich, Federica Gerace, Bruno Loureiro, et al.
+- **abs:** https://arxiv.org/abs/2601.23208 · **PDF:** https://arxiv.org/pdf/2601.23208v1
+
+## 17450. Learning to Execute Graph Algorithms Exactly with Graph Neural Networks
+
+- **ID:** `2601.23207` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Muhammad Fetrat Qharabagh, Artur Back de Luca, George Giapitzakis, et al.
+- **abs:** https://arxiv.org/abs/2601.23207 · **PDF:** https://arxiv.org/pdf/2601.23207v1
+
+## 17451. High-quality generation of dynamic game content via small language models: A proof of concept
+
+- **ID:** `2601.23206` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Morten I. K. Munk, Arturo Valdivia, Paolo Burelli
+- **abs:** https://arxiv.org/abs/2601.23206 · **PDF:** https://arxiv.org/pdf/2601.23206v1
+
+## 17452. TSAQA: Time Series Analysis Question And Answering Benchmark
+
+- **ID:** `2601.23204` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Baoyu Jing, Sanhorn Chen, Lecheng Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.23204 · **PDF:** https://arxiv.org/pdf/2601.23204v1
+
+## 17453. A General ReLearner: Empowering Spatiotemporal Prediction by Re-learning Input-label Residual
+
+- **ID:** `2602.02563` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Jiaming Ma, Binwu Wang, Pengkun Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.02563 · **PDF:** https://arxiv.org/pdf/2602.02563v1
+
+## 17454. Scale-Cascaded Diffusion Models for Super-Resolution in Medical Imaging
+
+- **ID:** `2601.23201` · **primary:** `eess.IV` · **Submitted:** 2026-01-30
+- **Authors:** Darshan Thaker, Mahmoud Mostapha, Radu Miron, et al.
+- **abs:** https://arxiv.org/abs/2601.23201 · **PDF:** https://arxiv.org/pdf/2601.23201v1
+
+## 17455. Deep Search with Hierarchical Meta-Cognitive Monitoring Inspired by Cognitive Neuroscience
+
+- **ID:** `2601.23188` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Zhongxiang Sun, Qipeng Wang, Weijie Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.23188 · **PDF:** https://arxiv.org/pdf/2601.23188v1
+
+## 17456. ReGuLaR: Variational Latent Reasoning Guided by Rendered Chain-of-Thought
+
+- **ID:** `2601.23184` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Fanmeng Wang, Haotian Liu, Guojiang Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.23184 · **PDF:** https://arxiv.org/pdf/2601.23184v1
+
+## 17457. JobResQA: A Benchmark for LLM Machine Reading Comprehension on Multilingual Résumés and JDs
+
+- **ID:** `2601.23183` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Casimiro Pio Carrino, Paula Estrella, Rabih Zbib, et al.
+- **abs:** https://arxiv.org/abs/2601.23183 · **PDF:** https://arxiv.org/pdf/2601.23183v1
+
+## 17458. FourierSampler: Unlocking Non-Autoregressive Potential in Diffusion Language Models via Frequency-Guided Generation
+
+- **ID:** `2601.23182` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Siyang He, Qiqi Wang, Xiaoran Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.23182 · **PDF:** https://arxiv.org/pdf/2601.23182v1
+
+## 17459. Ensuring Semantics in Weights of Implicit Neural Representations through the Implicit Function Theorem
+
+- **ID:** `2601.23181` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Tianming Qiu, Christos Sonis, Hao Shen
+- **abs:** https://arxiv.org/abs/2601.23181 · **PDF:** https://arxiv.org/pdf/2601.23181v1
+
+## 17460. TriSpec: Ternary Speculative Decoding via Lightweight Proxy Verification
+
+- **ID:** `2601.23180` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Haoyun Jiang, Junqi He, Feng Hong, et al.
+- **abs:** https://arxiv.org/abs/2601.23180 · **PDF:** https://arxiv.org/pdf/2601.23180v1
+
+## 17461. Make Anything Match Your Target: Universal Adversarial Perturbations against Closed-Source MLLMs via Multi-Crop Routed Meta Optimization
+
+- **ID:** `2601.23179` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Hui Lu, Yi Yu, Yiming Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.23179 · **PDF:** https://arxiv.org/pdf/2601.23179v1
+
+## 17462. MeshGraphNet-Transformer: Scalable Mesh-based Learned Simulation for Solid Mechanics
+
+- **ID:** `2601.23177` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Mikel M. Iparraguirre, Iciar Alfaro, David Gonzalez, et al.
+- **abs:** https://arxiv.org/abs/2601.23177 · **PDF:** https://arxiv.org/pdf/2601.23177v3
+
+## 17463. Beyond Fixed Frames: Dynamic Character-Aligned Speech Tokenization
+
+- **ID:** `2601.23174` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Luca Della Libera, Cem Subakan, Mirco Ravanelli
+- **abs:** https://arxiv.org/abs/2601.23174 · **PDF:** https://arxiv.org/pdf/2601.23174v2
+
+## 17464. Names Don't Matter: Symbol-Invariant Transformer for Open-Vocabulary Learning
+
+- **ID:** `2601.23169` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** İlker Işık, Wenchao Li
+- **abs:** https://arxiv.org/abs/2601.23169 · **PDF:** https://arxiv.org/pdf/2601.23169v1
+
+## 17465. Monotonic Reference-Free Refinement for Autoformalization
+
+- **ID:** `2601.23166` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Lan Zhang, Marco Valentino, André Freitas
+- **abs:** https://arxiv.org/abs/2601.23166 · **PDF:** https://arxiv.org/pdf/2601.23166v1
+
+## 17466. Stochastic Linear Bandits with Parameter Noise
+
+- **ID:** `2601.23164` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Daniel Ezer, Alon Peled-Cohen, Yishay Mansour
+- **abs:** https://arxiv.org/abs/2601.23164 · **PDF:** https://arxiv.org/pdf/2601.23164v1
+
+## 17467. Probing the Trajectories of Reasoning Traces in Large Language Models
+
+- **ID:** `2601.23163` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Marthe Ballon, Brecht Verbeken, Vincent Ginis, et al.
+- **abs:** https://arxiv.org/abs/2601.23163 · **PDF:** https://arxiv.org/pdf/2601.23163v1
+
+## 17468. DIFFA-2: A Practical Diffusion Large Language Model for General Audio Understanding
+
+- **ID:** `2601.23161` · **primary:** `cs.SD` · **Submitted:** 2026-01-30
+- **Authors:** Jiaming Zhou, Xuxin Cheng, Shiwan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.23161 · **PDF:** https://arxiv.org/pdf/2601.23161v1
+
+## 17469. No More, No Less: Least-Privilege Language Models
+
+- **ID:** `2601.23157` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Paulius Rauba, Dominykas Seputis, Patrikas Vanagas, et al.
+- **abs:** https://arxiv.org/abs/2601.23157 · **PDF:** https://arxiv.org/pdf/2601.23157v2
+
+## 17470. Unsupervised Hierarchical Skill Discovery
+
+- **ID:** `2601.23156` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Damion Harvey, Geraud Nangue Tasse, Branden Ingram, et al.
+- **abs:** https://arxiv.org/abs/2601.23156 · **PDF:** https://arxiv.org/pdf/2601.23156v1
+
+## 17471. SPICE: Submodular Penalized Information-Conflict Selection for Efficient Large Language Model Training
+
+- **ID:** `2601.23155` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Powei Chang, Jinpeng Zhang, Bowen Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.23155 · **PDF:** https://arxiv.org/pdf/2601.23155v1
+
+## 17472. On Safer Reinforcement Learning Policies for Sedation and Analgesia in Intensive Care
+
+- **ID:** `2601.23154` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Joel Romero-Hernandez, Oscar Camara
+- **abs:** https://arxiv.org/abs/2601.23154 · **PDF:** https://arxiv.org/pdf/2601.23154v1
+
+## 17473. Behemoth: Benchmarking Unlearning in LLMs Using Fully Synthetic Data
+
+- **ID:** `2601.23153` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Eugenia Iofinova, Dan Alistarh
+- **abs:** https://arxiv.org/abs/2601.23153 · **PDF:** https://arxiv.org/pdf/2601.23153v1
+
+## 17474. Tri-LLM Cooperative Federated Zero-Shot Intrusion Detection with Semantic Disagreement and Trust-Aware Aggregation
+
+- **ID:** `2602.00219` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Saeid Jamshidi, Omar Abdul Wahab, Foutse Khomh, et al.
+- **abs:** https://arxiv.org/abs/2602.00219 · **PDF:** https://arxiv.org/pdf/2602.00219v1
+
+## 17475. Position-Aware Self-supervised Representation Learning for Cross-mode Radar Signal Recognition
+
+- **ID:** `2602.11196` · **primary:** `eess.SP` · **Submitted:** 2026-01-30
+- **Authors:** Hongyang Zhang, Haitao Zhang, Yinhao Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.11196 · **PDF:** https://arxiv.org/pdf/2602.11196v1
+
+## 17476. Manifold-Aware Perturbations for Constrained Generative Modeling
+
+- **ID:** `2601.23151` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Katherine Keegan, Lars Ruthotto
+- **abs:** https://arxiv.org/abs/2601.23151 · **PDF:** https://arxiv.org/pdf/2601.23151v1
+
+## 17477. Compressed BC-LISTA via Low-Rank Convolutional Decomposition
+
+- **ID:** `2601.23148` · **primary:** `eess.IV` · **Submitted:** 2026-01-30
+- **Authors:** Han Wang, Yhonatan Kvich, Eduardo Pérez, et al.
+- **abs:** https://arxiv.org/abs/2601.23148 · **PDF:** https://arxiv.org/pdf/2601.23148v1
+
+## 17478. Securing Time in Energy IoT: A Clock-Dynamics-Aware Spatio-Temporal Graph Attention Network for Clock Drift Attacks and Y2K38 Failures
+
+- **ID:** `2601.23147` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Saeid Jamshidi, Omar Abdul Wahab, Rolando Herrero, et al.
+- **abs:** https://arxiv.org/abs/2601.23147 · **PDF:** https://arxiv.org/pdf/2601.23147v1
+
+## 17479. THINKSAFE: Self-Generated Safety Alignment for Reasoning Models
+
+- **ID:** `2601.23143` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Seanie Lee, Sangwoo Park, Yumin Choi, et al.
+- **abs:** https://arxiv.org/abs/2601.23143 · **PDF:** https://arxiv.org/pdf/2601.23143v1
+
+## 17480. GRIP2: A Robust and Powerful Deep Knockoff Method for Feature Selection
+
+- **ID:** `2602.00218` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Bob Junyi Zou, Lu Tian
+- **abs:** https://arxiv.org/abs/2602.00218 · **PDF:** https://arxiv.org/pdf/2602.00218v1
+
+## 17481. Traversal-as-Policy: Log-Distilled Gated Behavior Trees as Externalized, Verifiable Policies for Safe, Robust, and Efficient Agents
+
+- **ID:** `2603.05517` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Peiran Li, Jiashuo Sun, Fangzhou Lin, et al.
+- **abs:** https://arxiv.org/abs/2603.05517 · **PDF:** https://arxiv.org/pdf/2603.05517v1
+
+## 17482. Why GRPO Needs Normalization: A Local-Curvature Perspective on Adaptive Gradients
+
+- **ID:** `2601.23135` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Cheng Ge, Caitlyn Heqi Yin, Hao Liang, et al.
+- **abs:** https://arxiv.org/abs/2601.23135 · **PDF:** https://arxiv.org/pdf/2601.23135v1
+
+## 17483. Machine Learning for Energy-Performance-aware Scheduling
+
+- **ID:** `2601.23134` · **primary:** `cs.AR` · **Submitted:** 2026-01-30
+- **Authors:** Zheyuan Hu, Yifei Shi
+- **abs:** https://arxiv.org/abs/2601.23134 · **PDF:** https://arxiv.org/pdf/2601.23134v1
+
+## 17484. RAudit: A Blind Auditing Protocol for Large Language Model Reasoning
+
+- **ID:** `2601.23133` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Edward Y. Chang, Longling Geng
+- **abs:** https://arxiv.org/abs/2601.23133 · **PDF:** https://arxiv.org/pdf/2601.23133v1
+
+## 17485. Secure Tool Manifest and Digital Signing Solution for Verifiable MCP and LLM Pipelines
+
+- **ID:** `2601.23132` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Saeid Jamshidi, Kawser Wazed Nafi, Arghavan Moradi Dakhel, et al.
+- **abs:** https://arxiv.org/abs/2601.23132 · **PDF:** https://arxiv.org/pdf/2601.23132v1
+
+## 17486. Regularisation in neural networks: a survey and empirical analysis of approaches
+
+- **ID:** `2601.23131` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Christiaan P. Opperman, Anna S. Bosman, Katherine M. Malan
+- **abs:** https://arxiv.org/abs/2601.23131 · **PDF:** https://arxiv.org/pdf/2601.23131v1
+
+## 17487. Evaluating the Utility of Grounding Documents with Reference-Free LLM-based Metrics
+
+- **ID:** `2601.23129` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Yilun Hua, Giuseppe Castellucci, Peter Schulam, et al.
+- **abs:** https://arxiv.org/abs/2601.23129 · **PDF:** https://arxiv.org/pdf/2601.23129v1
+
+## 17488. Distribution-informed Efficient Conformal Prediction for Full Ranking
+
+- **ID:** `2601.23128` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Wenbo Liao, Huipeng Huang, Chen Jia, et al.
+- **abs:** https://arxiv.org/abs/2601.23128 · **PDF:** https://arxiv.org/pdf/2601.23128v1
+
+## 17489. Dispersion Loss Counteracts Embedding Condensation and Improves Generalization in Small Language Models
+
+- **ID:** `2602.00217` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Chen Liu, Xingzhi Sun, Xi Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.00217 · **PDF:** https://arxiv.org/pdf/2602.00217v1
+
+## 17490. To See Far, Look Close: Evolutionary Forecasting for Long-term Time Series
+
+- **ID:** `2601.23114` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Jiaming Ma, Siyuan Mu, Ruilin Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.23114 · **PDF:** https://arxiv.org/pdf/2601.23114v2
+
+## 17491. CATTO: Balancing Preferences and Confidence in Language Models
+
+- **ID:** `2601.23096` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Nisarg Parikh, Ananya Sai, Pannaga Shivaswamy, et al.
+- **abs:** https://arxiv.org/abs/2601.23096 · **PDF:** https://arxiv.org/pdf/2601.23096v2
+
+## 17492. Safer Policy Compliance with Dynamic Epistemic Fallback
+
+- **ID:** `2601.23094` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Joseph Marvin Imperial, Harish Tayyar Madabushi
+- **abs:** https://arxiv.org/abs/2601.23094 · **PDF:** https://arxiv.org/pdf/2601.23094v1
+
+## 17493. WiFiPenTester: Advancing Wireless Ethical Hacking with Governed GenAI
+
+- **ID:** `2601.23092` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Haitham S. Al-Sinani, Chris J. Mitchell
+- **abs:** https://arxiv.org/abs/2601.23092 · **PDF:** https://arxiv.org/pdf/2601.23092v1
+
+## 17494. From Similarity to Vulnerability: Key Collision Attack on LLM Semantic Caching
+
+- **ID:** `2601.23088` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Zhixiang Zhang, Zesen Liu, Yuchong Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.23088 · **PDF:** https://arxiv.org/pdf/2601.23088v1
+
+## 17495. Chain-of-thought obfuscation learned from output supervision can generalise to unseen tasks
+
+- **ID:** `2601.23086` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Nathaniel Mitrani Hadida, Sassan Bhanji, Cameron Tice, et al.
+- **abs:** https://arxiv.org/abs/2601.23086 · **PDF:** https://arxiv.org/pdf/2601.23086v1
+
+## 17496. OrLog: Resolving Complex Queries with LLMs and Probabilistic Reasoning
+
+- **ID:** `2601.23085` · **primary:** `cs.IR` · **Submitted:** 2026-01-30
+- **Authors:** Mohanna Hoveyda, Jelle Piepenbrock, Arjen P de Vries, et al.
+- **abs:** https://arxiv.org/abs/2601.23085 · **PDF:** https://arxiv.org/pdf/2601.23085v1
+
+## 17497. Predicting the post-wildfire mudflow onset using machine learning models on multi-parameter experimental data
+
+- **ID:** `2602.11194` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Mahta Movasat, Ingrid Tomac
+- **abs:** https://arxiv.org/abs/2602.11194 · **PDF:** https://arxiv.org/pdf/2602.11194v1
+
+## 17498. Character as a Latent Variable in Large Language Models: A Mechanistic Account of Emergent Misalignment and Conditional Safety Failures
+
+- **ID:** `2601.23081` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Yanghao Su, Wenbo Zhou, Tianwei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.23081 · **PDF:** https://arxiv.org/pdf/2601.23081v1
+
+## 17499. MathlibLemma: Folklore Lemma Generation and Benchmark for Formal Mathematics
+
+- **ID:** `2602.02561` · **primary:** `cs.LO` · **Submitted:** 2026-01-30
+- **Authors:** Xinyu Liu, Zixuan Xie, Amir Moeini, et al.
+- **abs:** https://arxiv.org/abs/2602.02561 · **PDF:** https://arxiv.org/pdf/2602.02561v1
+
+## 17500. RN-D: Discretized Categorical Actors with Regularized Networks for On-Policy Reinforcement Learning
+
+- **ID:** `2601.23075` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yuexin Bian, Jie Feng, Tao Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.23075 · **PDF:** https://arxiv.org/pdf/2601.23075v1
+
+## 17501. Auditing Sybil: Explaining Deep Lung Cancer Risk Prediction Through Generative Interventional Attributions
+
+- **ID:** `2602.02560` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Bartlomiej Sobieski, Jakub Grzywaczewski, Karol Dobiczek, et al.
+- **abs:** https://arxiv.org/abs/2602.02560 · **PDF:** https://arxiv.org/pdf/2602.02560v1
+
+## 17502. A Geometric Multimodal Foundation Model Integrating Bp-MRI and Clinical Reports in Prostate Cancer Classification
+
+- **ID:** `2602.00214` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Juan A. Olmos, Antoine Manzanera, Fabio Martínez
+- **abs:** https://arxiv.org/abs/2602.00214 · **PDF:** https://arxiv.org/pdf/2602.00214v1
+
+## 17503. SplineFlow: Flow Matching for Dynamical Systems with B-Spline Interpolants
+
+- **ID:** `2601.23072` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Santanu Subhash Rathod, Pietro Liò, Xiao Zhang
+- **abs:** https://arxiv.org/abs/2601.23072 · **PDF:** https://arxiv.org/pdf/2601.23072v1
+
+## 17504. TessPay: Verify-then-Pay Infrastructure for Trusted Agentic Commerce
+
+- **ID:** `2602.00213` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Mehul Goenka, Tejas Pathak, Siddharth Asthana
+- **abs:** https://arxiv.org/abs/2602.00213 · **PDF:** https://arxiv.org/pdf/2602.00213v1
+
+## 17505. ExplainerPFN: Towards tabular foundation models for model-free zero-shot feature importance estimations
+
+- **ID:** `2601.23068` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Joao Fonseca, Julia Stoyanovich
+- **abs:** https://arxiv.org/abs/2601.23068 · **PDF:** https://arxiv.org/pdf/2601.23068v1
+
+## 17506. Towards Explicit Acoustic Evidence Perception in Audio LLMs for Speech Deepfake Detection
+
+- **ID:** `2601.23066` · **primary:** `cs.SD` · **Submitted:** 2026-01-30
+- **Authors:** Xiaoxuan Guo, Yuankun Xie, Haonan Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.23066 · **PDF:** https://arxiv.org/pdf/2601.23066v1
+
+## 17507. HierLoc: Hyperbolic Entity Embeddings for Hierarchical Visual Geolocation
+
+- **ID:** `2601.23064` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Hari Krishna Gadi, Daniel Matos, Hongyi Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.23064 · **PDF:** https://arxiv.org/pdf/2601.23064v2
+
+## 17508. Experience-Driven Multi-Agent Systems Are Training-free Context-aware Earth Observers
+
+- **ID:** `2602.02559` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Pengyu Dai, Weihao Xuan, Junjue Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.02559 · **PDF:** https://arxiv.org/pdf/2602.02559v1
+
+## 17509. On the Impact of Code Comments for Automated Bug-Fixing: An Empirical Study
+
+- **ID:** `2601.23059` · **primary:** `cs.SE` · **Submitted:** 2026-01-30
+- **Authors:** Antonio Vitale, Emanuela Guglielmi, Simone Scalabrino, et al.
+- **abs:** https://arxiv.org/abs/2601.23059 · **PDF:** https://arxiv.org/pdf/2601.23059v1
+
+## 17510. From Absolute to Relative: Rethinking Reward Shaping in Group-Based Reinforcement Learning
+
+- **ID:** `2601.23058` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Wenzhe Niu, Wei He, Zongxia Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.23058 · **PDF:** https://arxiv.org/pdf/2601.23058v1
+
+## 17511. PA-MIL: Phenotype-Aware Multiple Instance Learning Guided by Language Prompting and Genotype-to-Phenotype Relationships
+
+- **ID:** `2602.02558` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Zekang Yang, Hong Liu, Xiangdong Wang
+- **abs:** https://arxiv.org/abs/2602.02558 · **PDF:** https://arxiv.org/pdf/2602.02558v1
+
+## 17512. Adaptive Edge Learning for Density-Aware Graph Generation
+
+- **ID:** `2601.23052` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Seyedeh Ava Razi Razavi, James Sargant, Sheridan Houghten, et al.
+- **abs:** https://arxiv.org/abs/2601.23052 · **PDF:** https://arxiv.org/pdf/2601.23052v1
+
+## 17513. MedMCP-Calc: Benchmarking LLMs for Realistic Medical Calculator Scenarios via MCP Integration
+
+- **ID:** `2601.23049` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Yakun Zhu, Yutong Huang, Shengqian Qin, et al.
+- **abs:** https://arxiv.org/abs/2601.23049 · **PDF:** https://arxiv.org/pdf/2601.23049v1
+
+## 17514. From Abstract to Contextual: What LLMs Still Cannot Do in Mathematics
+
+- **ID:** `2601.23048` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Bowen Cao, Dongdong Zhang, Yixia Li, et al.
+- **abs:** https://arxiv.org/abs/2601.23048 · **PDF:** https://arxiv.org/pdf/2601.23048v2
+
+## 17515. The Hot Mess of AI: How Does Misalignment Scale With Model Intelligence and Task Complexity?
+
+- **ID:** `2601.23045` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Alexander Hägele, Aryo Pradipta Gema, Henry Sleight, et al.
+- **abs:** https://arxiv.org/abs/2601.23045 · **PDF:** https://arxiv.org/pdf/2601.23045v1
+
+## 17516. Avoiding Premature Collapse: Adaptive Annealing for Entropy-Regularized Structural Inference
+
+- **ID:** `2601.23039` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yizhi Liu
+- **abs:** https://arxiv.org/abs/2601.23039 · **PDF:** https://arxiv.org/pdf/2601.23039v3
+
+## 17517. Breaking the Stochasticity Barrier: An Adaptive Variance-Reduced Method for Variational Inequalities
+
+- **ID:** `2601.23034` · **primary:** `math.OC` · **Submitted:** 2026-01-30
+- **Authors:** Yungi Jeong, Takumi Otsuka
+- **abs:** https://arxiv.org/abs/2601.23034 · **PDF:** https://arxiv.org/pdf/2601.23034v1
+
+## 17518. Guided by Trajectories: Repairing and Rewarding Tool-Use Trajectories for Tool-Integrated Reasoning
+
+- **ID:** `2601.23032` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Siyu Gong, Linan Yue, Weibo Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.23032 · **PDF:** https://arxiv.org/pdf/2601.23032v1
+
+## 17519. Interpretable Unsupervised Deformable Image Registration via Confidence-bound Multi-Hop Visual Reasoning
+
+- **ID:** `2602.00211` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Zafar Iqbal, Anwar Ul Haq, Srimannarayana Grandhi
+- **abs:** https://arxiv.org/abs/2602.00211 · **PDF:** https://arxiv.org/pdf/2602.00211v1
+
+## 17520. MELINOE: Fine-Tuning Enables Memory-Efficient Inference for Mixture-of-Experts Models
+
+- **ID:** `2602.11192` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Arian Raje, Anupam Nayak, Gauri Joshi
+- **abs:** https://arxiv.org/abs/2602.11192 · **PDF:** https://arxiv.org/pdf/2602.11192v1
+
+## 17521. Asymptotic Theory of Iterated Empirical Risk Minimization, with Applications to Active Learning
+
+- **ID:** `2601.23031` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Hugo Cui, Yue M. Lu
+- **abs:** https://arxiv.org/abs/2601.23031 · **PDF:** https://arxiv.org/pdf/2601.23031v1
+
+## 17522. Neural Backward Filtering Forward Guiding
+
+- **ID:** `2601.23030` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Gefan Yang, Frank van der Meulen, Stefan Sommer
+- **abs:** https://arxiv.org/abs/2601.23030 · **PDF:** https://arxiv.org/pdf/2601.23030v1
+
+## 17523. Divide-and-Conquer CoT: RL for Reducing Latency via Parallel Reasoning
+
+- **ID:** `2601.23027` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Arvind Mahankali, Kaiyue Wen, Tengyu Ma
+- **abs:** https://arxiv.org/abs/2601.23027 · **PDF:** https://arxiv.org/pdf/2601.23027v1
+
+## 17524. Causal Characterization of Measurement and Mechanistic Anomalies
+
+- **ID:** `2601.23026` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Hendrik Suhr, David Kaltenpoth, Jilles Vreeken
+- **abs:** https://arxiv.org/abs/2601.23026 · **PDF:** https://arxiv.org/pdf/2601.23026v1
+
+## 17525. DimABSA: Building Multilingual and Multidomain Datasets for Dimensional Aspect-Based Sentiment Analysis
+
+- **ID:** `2601.23022` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Lung-Hao Lee, Liang-Chih Yu, Natalia Loukashevich, et al.
+- **abs:** https://arxiv.org/abs/2601.23022 · **PDF:** https://arxiv.org/pdf/2601.23022v2
+
+## 17526. The Alignment Curse: Cross-Modality Jailbreak Transfer in Omni-Models
+
+- **ID:** `2602.02557` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yupeng Chen, Junchi Yu, Aoxi Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.02557 · **PDF:** https://arxiv.org/pdf/2602.02557v1
+
+## 17527. AST-PAC: AST-guided Membership Inference for Code
+
+- **ID:** `2602.13240` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Roham Koohestani, Ali Al-Kaswan, Jonathan Katzy, et al.
+- **abs:** https://arxiv.org/abs/2602.13240 · **PDF:** https://arxiv.org/pdf/2602.13240v1
+
+## 17528. Mem-T: Densifying Rewards for Long-Horizon Memory Agents
+
+- **ID:** `2601.23014` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yanwei Yue, Boci Peng, Xuanbo Fan, et al.
+- **abs:** https://arxiv.org/abs/2601.23014 · **PDF:** https://arxiv.org/pdf/2601.23014v2
+
+## 17529. Leveraging Convolutional Sparse Autoencoders for Robust Movement Classification from Low-Density sEMG
+
+- **ID:** `2601.23011` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Blagoj Hristov, Zoran Hadzi-Velkov, Katerina Hadzi-Velkova Saneva, et al.
+- **abs:** https://arxiv.org/abs/2601.23011 · **PDF:** https://arxiv.org/pdf/2601.23011v1
+
+## 17530. Automatic Constraint Policy Optimization based on Continuous Constraint Interpolation Framework for Offline Reinforcement Learning
+
+- **ID:** `2601.23010` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Xinchen Han, Qiuyang Fang, Hossam Afifi, et al.
+- **abs:** https://arxiv.org/abs/2601.23010 · **PDF:** https://arxiv.org/pdf/2601.23010v1
+
+## 17531. InstructDiff: Domain-Adaptive Data Selection via Differential Entropy for Efficient LLM Fine-Tuning
+
+- **ID:** `2601.23006` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Junyou Su, He Zhu, Xiao Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.23006 · **PDF:** https://arxiv.org/pdf/2601.23006v1
+
+## 17532. Bias Beyond Borders: Political Ideology Evaluation and Steering in Multilingual LLMs
+
+- **ID:** `2601.23001` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Afrozah Nadeem, Agrima Seth, Mehwish Nasim, et al.
+- **abs:** https://arxiv.org/abs/2601.23001 · **PDF:** https://arxiv.org/pdf/2601.23001v3
+
+## 17533. Mano: Restriking Manifold Optimization for LLM Training
+
+- **ID:** `2601.23000` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yufei Gu, Zeke Xie
+- **abs:** https://arxiv.org/abs/2601.23000 · **PDF:** https://arxiv.org/pdf/2601.23000v1
+
+## 17534. TriCEGAR: A Trace-Driven Abstraction Mechanism for Agentic AI
+
+- **ID:** `2601.22997` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Roham Koohestani, Ateş Görpelioğlu, Egor Klimov, et al.
+- **abs:** https://arxiv.org/abs/2601.22997 · **PDF:** https://arxiv.org/pdf/2601.22997v1
+
+## 17535. Value-at-Risk Constrained Policy Optimization
+
+- **ID:** `2601.22993` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Rohan Tangri, Jan-Peter Calliess
+- **abs:** https://arxiv.org/abs/2601.22993 · **PDF:** https://arxiv.org/pdf/2601.22993v1
+
+## 17536. Self-Supervised Slice-to-Volume Reconstruction with Gaussian Representations for Fetal MRI
+
+- **ID:** `2601.22990` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Yinsong Wang, Thomas Fletcher, Xinzhe Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.22990 · **PDF:** https://arxiv.org/pdf/2601.22990v1
+
+## 17537. ArabicDialectHub: A Cross-Dialectal Arabic Learning Resource and Platform
+
+- **ID:** `2601.22987` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Salem Lahlou
+- **abs:** https://arxiv.org/abs/2601.22987 · **PDF:** https://arxiv.org/pdf/2601.22987v1
+
+## 17538. dgMARK: Decoding-Guided Watermarking for Diffusion Language Models
+
+- **ID:** `2601.22985` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Pyo Min Hong, Albert No
+- **abs:** https://arxiv.org/abs/2601.22985 · **PDF:** https://arxiv.org/pdf/2601.22985v1
+
+## 17539. Why Your Deep Research Agent Fails? On Hallucination Evaluation in Full Research Trajectory
+
+- **ID:** `2601.22984` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Yuhao Zhan, Tianyu Fan, Linxuan Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.22984 · **PDF:** https://arxiv.org/pdf/2601.22984v1
+
+## 17540. Analyzing Shapley Additive Explanations to Understand Anomaly Detection Algorithm Behaviors and Their Complementarity
+
+- **ID:** `2602.00208` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Jordan Levy, Paul Saves, Moncef Garouani, et al.
+- **abs:** https://arxiv.org/abs/2602.00208 · **PDF:** https://arxiv.org/pdf/2602.00208v1
+
+## 17541. PIDSMaker: Building and Evaluating Provenance-based Intrusion Detection Systems
+
+- **ID:** `2601.22983` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Tristan Bilot, Baoxiang Jiang, Thomas Pasquier
+- **abs:** https://arxiv.org/abs/2601.22983 · **PDF:** https://arxiv.org/pdf/2601.22983v2
+
+## 17542. About an Automating Annotation Method for Robot Markers
+
+- **ID:** `2601.22982` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Wataru Uemura, Takeru Nagashima
+- **abs:** https://arxiv.org/abs/2601.22982 · **PDF:** https://arxiv.org/pdf/2601.22982v1
+
+## 17543. Learnable Permutation for Structured Sparsity on Transformer Models
+
+- **ID:** `2601.22980` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Zekai Li, Ji Liu, Guanchen Li, et al.
+- **abs:** https://arxiv.org/abs/2601.22980 · **PDF:** https://arxiv.org/pdf/2601.22980v1
+
+## 17544. Quantifying Model Uniqueness in Heterogeneous AI Ecosystems
+
+- **ID:** `2601.22977` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Lei You
+- **abs:** https://arxiv.org/abs/2601.22977 · **PDF:** https://arxiv.org/pdf/2601.22977v2
+
+## 17545. Golden Goose: A Simple Trick to Synthesize Unlimited RLVR Tasks from Unverifiable Internet Text
+
+- **ID:** `2601.22975` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Ximing Lu, David Acuna, Jaehun Jung, et al.
+- **abs:** https://arxiv.org/abs/2601.22975 · **PDF:** https://arxiv.org/pdf/2601.22975v2
+
+## 17546. MiTa: A Hierarchical Multi-Agent Collaboration Framework with Memory-integrated and Task Allocation
+
+- **ID:** `2601.22974` · **primary:** `cs.ET` · **Submitted:** 2026-01-30
+- **Authors:** XiaoJie Zhang, JianHan Wu, Xiaoyang Qu, et al.
+- **abs:** https://arxiv.org/abs/2601.22974 · **PDF:** https://arxiv.org/pdf/2601.22974v1
+
+## 17547. Multi-Scale Temporal Homeostasis Enables Efficient and Robust Neural Networks
+
+- **ID:** `2602.07009` · **primary:** `cs.NE` · **Submitted:** 2026-01-30
+- **Authors:** MD Azizul Hakim
+- **abs:** https://arxiv.org/abs/2602.07009 · **PDF:** https://arxiv.org/pdf/2602.07009v1
+
+## 17548. Stabilizing the Q-Gradient Field for Policy Smoothness in Actor-Critic
+
+- **ID:** `2601.22970` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Jeong Woon Lee, Kyoleen Kwak, Daeho Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.22970 · **PDF:** https://arxiv.org/pdf/2601.22970v1
+
+## 17549. Improved Algorithms for Nash Welfare in Linear Bandits
+
+- **ID:** `2601.22969` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Dhruv Sarkar, Nishant Pandey, Sayak Ray Chowdhury
+- **abs:** https://arxiv.org/abs/2601.22969 · **PDF:** https://arxiv.org/pdf/2601.22969v1
+
+## 17550. A Unified View of Attention and Residual Sinks: Outlier-Driven Rescaling is Essential for Transformer Training
+
+- **ID:** `2601.22966` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Zihan Qiu, Zeyu Huang, Kaiyue Wen, et al.
+- **abs:** https://arxiv.org/abs/2601.22966 · **PDF:** https://arxiv.org/pdf/2601.22966v1
+
+## 17551. EvoClinician: A Self-Evolving Agent for Multi-Turn Medical Diagnosis via Test-Time Evolutionary Learning
+
+- **ID:** `2601.22964` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Yufei He, Juncheng Liu, Zhiyuan Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.22964 · **PDF:** https://arxiv.org/pdf/2601.22964v1
+
+## 17552. Residual Context Diffusion Language Models
+
+- **ID:** `2601.22954` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Yuezhou Hu, Harman Singh, Monishwaran Maheswaran, et al.
+- **abs:** https://arxiv.org/abs/2601.22954 · **PDF:** https://arxiv.org/pdf/2601.22954v1
+
+## 17553. Beyond Experience Retrieval: Learning to Generate Utility-Optimized Structured Experience for Frozen LLMs
+
+- **ID:** `2602.02556` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Xuancheng Li, Haitao Li, Yujia Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.02556 · **PDF:** https://arxiv.org/pdf/2602.02556v1
+
+## 17554. OneFlowSBI: One Model, Many Queries for Simulation-Based Inference
+
+- **ID:** `2601.22951` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Mayank Nautiyal, Li Ju, Melker Ernfors, et al.
+- **abs:** https://arxiv.org/abs/2601.22951 · **PDF:** https://arxiv.org/pdf/2601.22951v1
+
+## 17555. Perplexity Cannot Always Tell Right from Wrong
+
+- **ID:** `2601.22950` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Petar Veličković, Federico Barbero, Christos Perivolaropoulos, et al.
+- **abs:** https://arxiv.org/abs/2601.22950 · **PDF:** https://arxiv.org/pdf/2601.22950v1
+
+## 17556. Autonomous Chain-of-Thought Distillation for Graph-Based Fraud Detection
+
+- **ID:** `2601.22949` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Yuan Li, Jun Hu, Bryan Hooi, et al.
+- **abs:** https://arxiv.org/abs/2601.22949 · **PDF:** https://arxiv.org/pdf/2601.22949v1
+
+## 17557. Alignment among Language, Vision and Action Representations
+
+- **ID:** `2601.22948` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Nicola Milano, Stefano Nolfi
+- **abs:** https://arxiv.org/abs/2601.22948 · **PDF:** https://arxiv.org/pdf/2601.22948v1
+
+## 17558. Learning to Explore with Parameter-Space Noise: A Deep Dive into Parameter-Space Noise for Reinforcement Learning with Verifiable Rewards
+
+- **ID:** `2602.02555` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Bizhe Bai, Xinyue Wang, Peng Ye, et al.
+- **abs:** https://arxiv.org/abs/2602.02555 · **PDF:** https://arxiv.org/pdf/2602.02555v3
+
+## 17559. Relaxing Positional Alignment in Masked Diffusion Language Models
+
+- **ID:** `2601.22947` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Mengyu Ye, Ryosuke Takahashi, Keito Kudo, et al.
+- **abs:** https://arxiv.org/abs/2601.22947 · **PDF:** https://arxiv.org/pdf/2601.22947v1
+
+## 17560. From Data Leak to Secret Misses: The Impact of Data Leakage on Secret Detection Models
+
+- **ID:** `2601.22946` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Farnaz Soltaniani, Mohammad Ghafari
+- **abs:** https://arxiv.org/abs/2601.22946 · **PDF:** https://arxiv.org/pdf/2601.22946v1
+
+## 17561. Environment-Conditioned Tail Reweighting for Total Variation Invariant Risk Minimization
+
+- **ID:** `2601.22944` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yuanchao Wang, Zhao-Rong Lai, Tianqi Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.22944 · **PDF:** https://arxiv.org/pdf/2601.22944v2
+
+## 17562. Scalable Topology-Preserving Graph Coarsening with Graph Collapse
+
+- **ID:** `2601.22943` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Xiang Wu, Rong-Hua Li, Xunkai Li, et al.
+- **abs:** https://arxiv.org/abs/2601.22943 · **PDF:** https://arxiv.org/pdf/2601.22943v1
+
+## 17563. Reducing Class-Wise Performance Disparity via Margin Regularization
+
+- **ID:** `2602.00205` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Beier Zhu, Kesen Zhao, Jiequan Cui, et al.
+- **abs:** https://arxiv.org/abs/2602.00205 · **PDF:** https://arxiv.org/pdf/2602.00205v1
+
+## 17564. A Real-Time Privacy-Preserving Behavior Recognition System via Edge-Cloud Collaboration
+
+- **ID:** `2601.22938` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Huan Song, Shuyu Tian, Junyi Hao, et al.
+- **abs:** https://arxiv.org/abs/2601.22938 · **PDF:** https://arxiv.org/pdf/2601.22938v1
+
+## 17565. Protecting Private Code in IDE Autocomplete using Differential Privacy
+
+- **ID:** `2601.22935` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Evgeny Grigorenko, David Stanojević, David Ilić, et al.
+- **abs:** https://arxiv.org/abs/2601.22935 · **PDF:** https://arxiv.org/pdf/2601.22935v1
+
+## 17566. DC-LA: Difference-of-Convex Langevin Algorithm
+
+- **ID:** `2601.22932` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Hoang Phuc Hau Luu, Zhongjian Wang
+- **abs:** https://arxiv.org/abs/2601.22932 · **PDF:** https://arxiv.org/pdf/2601.22932v1
+
+## 17567. Benchmarking Machine Translation on Chinese Social Media Texts
+
+- **ID:** `2601.22931` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Kaiyan Zhao, Zheyong Xie, Zhongtao Miao, et al.
+- **abs:** https://arxiv.org/abs/2601.22931 · **PDF:** https://arxiv.org/pdf/2601.22931v1
+
+## 17568. MTDrive: Multi-turn Interactive Reinforcement Learning for Autonomous Driving
+
+- **ID:** `2601.22930` · **primary:** `cs.RO` · **Submitted:** 2026-01-30
+- **Authors:** Xidong Li, Mingyu Guo, Chenchao Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.22930 · **PDF:** https://arxiv.org/pdf/2601.22930v1
+
+## 17569. Semantic Leakage from Image Embeddings
+
+- **ID:** `2601.22929` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Yiyi Chen, Qiongkai Xu, Desmond Elliott, et al.
+- **abs:** https://arxiv.org/abs/2601.22929 · **PDF:** https://arxiv.org/pdf/2601.22929v2
+
+## 17570. LLMs Explain't: A Post-Mortem on Semantic Interpretability in Transformer Models
+
+- **ID:** `2601.22928` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Alhassan Abdelhalim, Janick Edinger, Sören Laue, et al.
+- **abs:** https://arxiv.org/abs/2601.22928 · **PDF:** https://arxiv.org/pdf/2601.22928v1
+
+## 17571. BEAR: Towards Beam-Search-Aware Optimization for Recommendation with Large Language Models
+
+- **ID:** `2601.22925` · **primary:** `cs.IR` · **Submitted:** 2026-01-30
+- **Authors:** Weiqin Yang, Bohao Wang, Zhenxiang Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.22925 · **PDF:** https://arxiv.org/pdf/2601.22925v1
+
+## 17572. Evaluating Large Language Models for Security Bug Report Prediction
+
+- **ID:** `2601.22921` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Farnaz Soltaniani, Shoaib Razzaq, Mohammad Ghafari
+- **abs:** https://arxiv.org/abs/2601.22921 · **PDF:** https://arxiv.org/pdf/2601.22921v1
+
+## 17573. Semantic-Aware Advanced Persistent Threat Detection Using Autoencoders on LLM-Encoded System Logs
+
+- **ID:** `2602.00204` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Waleed Khan Mohammed, Zahirul Arief Irfan Bin Shahrul Anuar, Mousa Sufian Mousa Mitani, et al.
+- **abs:** https://arxiv.org/abs/2602.00204 · **PDF:** https://arxiv.org/pdf/2602.00204v1
+
+## 17574. FlexLoRA: Entropy-Guided Flexible Low-Rank Adaptation
+
+- **ID:** `2601.22905` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Muqing Liu, Chongjie Si, Yuheng Jia
+- **abs:** https://arxiv.org/abs/2601.22905 · **PDF:** https://arxiv.org/pdf/2601.22905v1
+
+## 17575. DINO-SAE: DINO Spherical Autoencoder for High-Fidelity Image Reconstruction and Generation
+
+- **ID:** `2601.22904` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Hun Chang, Byunghee Cha, Jong Chul Ye
+- **abs:** https://arxiv.org/abs/2601.22904 · **PDF:** https://arxiv.org/pdf/2601.22904v1
+
+## 17576. MulFeRL: Enhancing Reinforcement Learning with Verbal Feedback in a Multi-turn Loop
+
+- **ID:** `2601.22900` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Xuancheng Li, Haitao Li, Yujia Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.22900 · **PDF:** https://arxiv.org/pdf/2601.22900v1
+
+## 17577. Uncertainty-Aware Extrapolation in Bayesian Oblique Trees
+
+- **ID:** `2601.22899` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Viktor Andonovikj, Sašo Džeroski, Pavle Boškoski
+- **abs:** https://arxiv.org/abs/2601.22899 · **PDF:** https://arxiv.org/pdf/2601.22899v1
+
+## 17578. Game-Theoretic Co-Evolution for LLM-Based Heuristic Discovery
+
+- **ID:** `2601.22896` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Xinyi Ke, Kai Li, Junliang Xing, et al.
+- **abs:** https://arxiv.org/abs/2601.22896 · **PDF:** https://arxiv.org/pdf/2601.22896v2
+
+## 17579. Calibrated Multivariate Distributional Regression with Pre-Rank Regularization
+
+- **ID:** `2601.22895` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Aya Laajil, Elnura Zhalieva, Naomi Desobry, et al.
+- **abs:** https://arxiv.org/abs/2601.22895 · **PDF:** https://arxiv.org/pdf/2601.22895v1
+
+## 17580. PlatoLTL: Learning to Generalize Across Symbols in LTL Instructions for Multi-Task RL
+
+- **ID:** `2601.22891` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Jacques Cloete, Mathias Jackermeier, Ioannis Havoutis, et al.
+- **abs:** https://arxiv.org/abs/2601.22891 · **PDF:** https://arxiv.org/pdf/2601.22891v1
+
+## 17581. Vision-Language Model Purified Semi-Supervised Semantic Segmentation for Remote Sensing Images
+
+- **ID:** `2602.00202` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Shanwen Wang, Xin Sun, Danfeng Hong, et al.
+- **abs:** https://arxiv.org/abs/2602.00202 · **PDF:** https://arxiv.org/pdf/2602.00202v1
+
+## 17582. DiffuSpeech: Silent Thought, Spoken Answer via Unified Speech-Text Diffusion
+
+- **ID:** `2601.22889` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Yuxuan Lou, Ziming Wu, Yaochen Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.22889 · **PDF:** https://arxiv.org/pdf/2601.22889v1
+
+## 17583. Should LLMs, like, Generate How Users Talk? Building Dialect-Accurate Dialog[ue]s Beyond the American Default with MDial
+
+- **ID:** `2601.22888` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Jio Oh, Paul Vicinanza, Thomas Butler, et al.
+- **abs:** https://arxiv.org/abs/2601.22888 · **PDF:** https://arxiv.org/pdf/2601.22888v2
+
+## 17584. MoVE: Mixture of Value Embeddings -- A New Axis for Scaling Parametric Memory in Autoregressive Models
+
+- **ID:** `2601.22887` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yangyan Li
+- **abs:** https://arxiv.org/abs/2601.22887 · **PDF:** https://arxiv.org/pdf/2601.22887v1
+
+## 17585. Leveraging LLMs For Turkish Skill Extraction
+
+- **ID:** `2601.22885` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Ezgi Arslan İltüzer, Özgür Anıl Özlü, Vahid Farajijobehdar, et al.
+- **abs:** https://arxiv.org/abs/2601.22885 · **PDF:** https://arxiv.org/pdf/2601.22885v1
+
+## 17586. Reinforcement Learning-Based Co-Design and Operation of Chiller and Thermal Energy Storage for Cost-Optimal HVAC Systems
+
+- **ID:** `2601.22880` · **primary:** `eess.SY` · **Submitted:** 2026-01-30
+- **Authors:** Tanay Raghunandan Srinivasa, Vivek Deulkar, Aviruch Bhatia, et al.
+- **abs:** https://arxiv.org/abs/2601.22880 · **PDF:** https://arxiv.org/pdf/2601.22880v2
+
+## 17587. Synthetic Time Series Generation via Complex Networks
+
+- **ID:** `2601.22879` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Jaime Vale, Vanessa Freitas Silva, Maria Eduarda Silva, et al.
+- **abs:** https://arxiv.org/abs/2601.22879 · **PDF:** https://arxiv.org/pdf/2601.22879v1
+
+## 17588. Matterhorn: Efficient Analog Sparse Spiking Transformer Architecture with Masked Time-To-First-Spike Encoding
+
+- **ID:** `2601.22876` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Zhanglu Yan, Kaiwen Tang, Zixuan Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.22876 · **PDF:** https://arxiv.org/pdf/2601.22876v1
+
+## 17589. From Labels to Facets: Building a Taxonomically Enriched Turkish Learner Corpus
+
+- **ID:** `2601.22875` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Elif Sayar, Tolgahan Türker, Anna Golynskaia Knezhevich, et al.
+- **abs:** https://arxiv.org/abs/2601.22875 · **PDF:** https://arxiv.org/pdf/2601.22875v2
+
+## 17590. EmoShift: Lightweight Activation Steering for Enhanced Emotion-Aware Speech Synthesis
+
+- **ID:** `2601.22873` · **primary:** `eess.AS` · **Submitted:** 2026-01-30
+- **Authors:** Li Zhou, Hao Jiang, Junjie Li, et al.
+- **abs:** https://arxiv.org/abs/2601.22873 · **PDF:** https://arxiv.org/pdf/2601.22873v1
+
+## 17591. Eroding the Truth-Default: A Causal Analysis of Human Susceptibility to Foundation Model Hallucinations and Disinformation in the Wild
+
+- **ID:** `2601.22871` · **primary:** `cs.CY` · **Submitted:** 2026-01-30
+- **Authors:** Alexander Loth, Martin Kappes, Marc-Oliver Pahl
+- **abs:** https://arxiv.org/abs/2601.22871 · **PDF:** https://arxiv.org/pdf/2601.22871v2
+
+## 17592. When Anomalies Depend on Context: Learning Conditional Compatibility for Anomaly Detection
+
+- **ID:** `2601.22868` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Shashank Mishra, Didier Stricker, Jason Rambach
+- **abs:** https://arxiv.org/abs/2601.22868 · **PDF:** https://arxiv.org/pdf/2601.22868v2
+
+## 17593. Degradation-Aware Frequency Regulation of a Heterogeneous Battery Fleet via Reinforcement Learning
+
+- **ID:** `2601.22865` · **primary:** `eess.SY` · **Submitted:** 2026-01-30
+- **Authors:** Tanay Raghunandan Srinivasa, Vivek Deulkar, Jia Bhargava, et al.
+- **abs:** https://arxiv.org/abs/2601.22865 · **PDF:** https://arxiv.org/pdf/2601.22865v2
+
+## 17594. Bayesian Interpolating Neural Network (B-INN): a scalable and reliable Bayesian model for large-scale physical systems
+
+- **ID:** `2601.22860` · **primary:** `math.NA` · **Submitted:** 2026-01-30
+- **Authors:** Chanwook Park, Brian Kim, Jiachen Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.22860 · **PDF:** https://arxiv.org/pdf/2601.22860v2
+
+## 17595. MEnvAgent: Scalable Polyglot Environment Construction for Verifiable Software Engineering
+
+- **ID:** `2601.22859` · **primary:** `cs.SE` · **Submitted:** 2026-01-30
+- **Authors:** Chuanzhe Guo, Jingjing Wu, Sijun He, et al.
+- **abs:** https://arxiv.org/abs/2601.22859 · **PDF:** https://arxiv.org/pdf/2601.22859v2
+
+## 17596. Learning to Build Shapes by Extrusion
+
+- **ID:** `2601.22858` · **primary:** `cs.GR` · **Submitted:** 2026-01-30
+- **Authors:** Thor Vestergaard Christiansen, Karran Pandey, Alba Reinders, et al.
+- **abs:** https://arxiv.org/abs/2601.22858 · **PDF:** https://arxiv.org/pdf/2601.22858v1
+
+## 17597. BatCoder: Self-Supervised Bidirectional Code-Documentation Learning via Back-Translation
+
+- **ID:** `2602.02554` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Jingwen Xu, Yiyang Lu, Zisu Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.02554 · **PDF:** https://arxiv.org/pdf/2602.02554v1
+
+## 17598. OptiMAG: Structure-Semantic Alignment via Unbalanced Optimal Transport
+
+- **ID:** `2601.22856` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yilong Zuo, Xunkai Li, Zhihan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22856 · **PDF:** https://arxiv.org/pdf/2601.22856v1
+
+## 17599. Hierarchical Shift Mixing -- Beyond Dense Attention in Transformers
+
+- **ID:** `2601.22852` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Robert Forchheimer
+- **abs:** https://arxiv.org/abs/2601.22852 · **PDF:** https://arxiv.org/pdf/2601.22852v1
+
+## 17600. When Meanings Meet: Investigating the Emergence and Quality of Shared Concept Spaces during Multilingual Language Model Training
+
+- **ID:** `2601.22851` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Felicia Körner, Max Müller-Eberstein, Anna Korhonen, et al.
+- **abs:** https://arxiv.org/abs/2601.22851 · **PDF:** https://arxiv.org/pdf/2601.22851v1
+
+## 17601. iScheduler: Reinforcement Learning-Driven Continual Optimization for Large-Scale Resource Investment Problems
+
+- **ID:** `2602.06064` · **primary:** `cs.DC` · **Submitted:** 2026-01-30
+- **Authors:** Yi-Xiang Hu, Yuke Wang, Feng Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.06064 · **PDF:** https://arxiv.org/pdf/2602.06064v1
+
+## 17602. Unconditional flow-based time series generation with equivariance-regularised latent spaces
+
+- **ID:** `2601.22848` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Camilo Carvajal Reyes, Felipe Tobar
+- **abs:** https://arxiv.org/abs/2601.22848 · **PDF:** https://arxiv.org/pdf/2601.22848v1
+
+## 17603. Reducing Memorisation in Generative Models via Riemannian Bayesian Inference
+
+- **ID:** `2602.00199` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Johanna Marie Gegenfurtner, Albert Kjøller Jacobsen, Naima Elosegui Borras, et al.
+- **abs:** https://arxiv.org/abs/2602.00199 · **PDF:** https://arxiv.org/pdf/2602.00199v1
+
+## 17604. Just-in-Time Catching Test Generation at Meta
+
+- **ID:** `2601.22832` · **primary:** `cs.SE` · **Submitted:** 2026-01-30
+- **Authors:** Matthew Becker, Yifei Chen, Nicholas Cochran, et al.
+- **abs:** https://arxiv.org/abs/2601.22832 · **PDF:** https://arxiv.org/pdf/2601.22832v1
+
+## 17605. Decomposing and Composing: Towards Efficient Vision-Language Continual Learning via Rank-1 Expert Pool in a Single LoRA
+
+- **ID:** `2601.22828` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Zhan Fa, Yue Duan, Jian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22828 · **PDF:** https://arxiv.org/pdf/2601.22828v1
+
+## 17606. Offline Reinforcement Learning of High-Quality Behaviors Under Robust Style Alignment
+
+- **ID:** `2601.22823` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Mathieu Petitbois, Rémy Portelas, Sylvain Lamprier
+- **abs:** https://arxiv.org/abs/2601.22823 · **PDF:** https://arxiv.org/pdf/2601.22823v1
+
+## 17607. User-Adaptive Meta-Learning for Cold-Start Medication Recommendation with Uncertainty Filtering
+
+- **ID:** `2601.22820` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Arya Hadizadeh Moghaddam, Mohsen Nayebi Kerdabadi, Dongjie Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.22820 · **PDF:** https://arxiv.org/pdf/2601.22820v1
+
+## 17608. Hide and Seek in Embedding Space: Geometry-based Steganography and Detection in Large Language Models
+
+- **ID:** `2601.22818` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Charles Westphal, Keivan Navaie, Fernando E. Rosas
+- **abs:** https://arxiv.org/abs/2601.22818 · **PDF:** https://arxiv.org/pdf/2601.22818v1
+
+## 17609. Cascaded Flow Matching for Heterogeneous Tabular Data with Mixed-Type Features
+
+- **ID:** `2601.22816` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Markus Mueller, Kathrin Gruber, Dennis Fok
+- **abs:** https://arxiv.org/abs/2601.22816 · **PDF:** https://arxiv.org/pdf/2601.22816v1
+
+## 17610. Quartet II: Accurate LLM Pre-Training in NVFP4 by Improved Unbiased Gradient Estimation
+
+- **ID:** `2601.22813` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Andrei Panferov, Erik Schultheis, Soroush Tabesh, et al.
+- **abs:** https://arxiv.org/abs/2601.22813 · **PDF:** https://arxiv.org/pdf/2601.22813v1
+
+## 17611. SCALED : Surrogate-gradient for Codec-Aware Learning of Downsampling in ABR Streaming
+
+- **ID:** `2602.00198` · **primary:** `eess.IV` · **Submitted:** 2026-01-30
+- **Authors:** Esteban Pesnel, Julien Le Tanou, Michael Ropert, et al.
+- **abs:** https://arxiv.org/abs/2602.00198 · **PDF:** https://arxiv.org/pdf/2602.00198v1
+
+## 17612. Rank-and-Reason: Multi-Agent Collaboration Accelerates Zero-Shot Protein Mutation Prediction
+
+- **ID:** `2602.00197` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-30
+- **Authors:** Yang Tan, Yuanxi Yu, Can Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.00197 · **PDF:** https://arxiv.org/pdf/2602.00197v2
+
+## 17613. Aligning the Unseen in Attributed Graphs: Interplay between Graph Geometry and Node Attributes Manifold
+
+- **ID:** `2601.22806` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Aldric Labarthe, Roland Bouffanais, Julien Randon-Furling
+- **abs:** https://arxiv.org/abs/2601.22806 · **PDF:** https://arxiv.org/pdf/2601.22806v1
+
+## 17614. SOMBRERO: Measuring and Steering Boundary Placement in End-to-End Hierarchical Sequence Models
+
+- **ID:** `2601.22805` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Pit Neitemeier, Alessio Serra, Jiaze Li, et al.
+- **abs:** https://arxiv.org/abs/2601.22805 · **PDF:** https://arxiv.org/pdf/2601.22805v1
+
+## 17615. CVeDRL: An Efficient Code Verifier via Difficulty-aware Reinforcement Learning
+
+- **ID:** `2601.22803` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Ji Shi, Peiming Guo, Meishan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22803 · **PDF:** https://arxiv.org/pdf/2601.22803v1
+
+## 17616. Clipping-Free Policy Optimization for Large Language Models
+
+- **ID:** `2601.22801` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Ömer Veysel Çağatan, Barış Akgün, Gözde Gül Şahin, et al.
+- **abs:** https://arxiv.org/abs/2601.22801 · **PDF:** https://arxiv.org/pdf/2601.22801v1
+
+## 17617. Trackly: A Unified SaaS Platform for User Behavior Analytics and Real Time Rule Based Anomaly Detection
+
+- **ID:** `2601.22800` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Md Zahurul Haque, Md. Hafizur Rahman, Yeahyea Sarker
+- **abs:** https://arxiv.org/abs/2601.22800 · **PDF:** https://arxiv.org/pdf/2601.22800v1
+
+## 17618. Where Not to Learn: Prior-Aligned Training with Subset-based Attribution Constraints for Reliable Decision-Making
+
+- **ID:** `2602.07008` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Ruoyu Chen, Shangquan Sun, Xiaoqing Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.07008 · **PDF:** https://arxiv.org/pdf/2602.07008v1
+
+## 17619. On the calibration of survival models with competing risks
+
+- **ID:** `2602.00194` · **primary:** `stat.ME` · **Submitted:** 2026-01-30
+- **Authors:** Julie Alberge, Tristan Haugomat, Gaël Varoquaux, et al.
+- **abs:** https://arxiv.org/abs/2602.00194 · **PDF:** https://arxiv.org/pdf/2602.00194v1
+
+## 17620. Efficient Distance Pruning for Process Suffix Comparison in Prescriptive Process Monitoring
+
+- **ID:** `2602.09039` · **primary:** `cs.DB` · **Submitted:** 2026-01-30
+- **Authors:** Sarra Madad
+- **abs:** https://arxiv.org/abs/2602.09039 · **PDF:** https://arxiv.org/pdf/2602.09039v1
+
+## 17621. Sparse or Dense? A Mechanistic Estimation of Computation Density in Transformer-based LLMs
+
+- **ID:** `2601.22795` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Corentin Kervadec, Iuliia Lysova, Marco Baroni, et al.
+- **abs:** https://arxiv.org/abs/2601.22795 · **PDF:** https://arxiv.org/pdf/2601.22795v1
+
+## 17622. CALM: Joint Contextual Acoustic-Linguistic Modeling for Personalization of Multi-Speaker ASR
+
+- **ID:** `2601.22792` · **primary:** `eess.AS` · **Submitted:** 2026-01-30
+- **Authors:** Muhammad Shakeel, Yosuke Fukumoto, Chikara Maeda, et al.
+- **abs:** https://arxiv.org/abs/2601.22792 · **PDF:** https://arxiv.org/pdf/2601.22792v1
+
+## 17623. Time-TK: A Multi-Offset Temporal Interaction Framework Combining Transformer and Kolmogorov-Arnold Networks for Time Series Forecasting
+
+- **ID:** `2602.11190` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Fan Zhang, Shiming Fan, Hua Wang
+- **abs:** https://arxiv.org/abs/2602.11190 · **PDF:** https://arxiv.org/pdf/2602.11190v1
+
+## 17624. Conditional Performance Guarantee for Large Reasoning Models
+
+- **ID:** `2601.22790` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Jianguo Huang, Hao Zeng, Bingyi Jing, et al.
+- **abs:** https://arxiv.org/abs/2601.22790 · **PDF:** https://arxiv.org/pdf/2601.22790v1
+
+## 17625. Float8@2bits: Entropy Coding Enables Data-Free Model Compression
+
+- **ID:** `2601.22787` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Patrick Putzky, Martin Genzel, Mattes Mollenhauer, et al.
+- **abs:** https://arxiv.org/abs/2601.22787 · **PDF:** https://arxiv.org/pdf/2601.22787v1
+
+## 17626. Toward IIT-Inspired Consciousness in LLMs: A Reward-Based Learning Framework
+
+- **ID:** `2601.22786` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Hamid Reza Akbari, Mohammad Hossein Sameti, Amir M. Mansourian, et al.
+- **abs:** https://arxiv.org/abs/2601.22786 · **PDF:** https://arxiv.org/pdf/2601.22786v1
+
+## 17627. Approximating $f$-Divergences with Rank Statistics
+
+- **ID:** `2601.22784` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Viktor Stein, José Manuel de Frutos
+- **abs:** https://arxiv.org/abs/2601.22784 · **PDF:** https://arxiv.org/pdf/2601.22784v1
+
+## 17628. Compact Hypercube Embeddings for Fast Text-based Wildlife Observation Retrieval
+
+- **ID:** `2601.22783` · **primary:** `cs.IR` · **Submitted:** 2026-01-30
+- **Authors:** Ilyass Moummad, Marius Miron, David Robinson, et al.
+- **abs:** https://arxiv.org/abs/2601.22783 · **PDF:** https://arxiv.org/pdf/2601.22783v1
+
+## 17629. Learning with Challenges: Adaptive Difficulty-Aware Data Generation for Mobile GUI Agent Training
+
+- **ID:** `2601.22781` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Linjia Kang, Zhimin Wang, Yongkang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22781 · **PDF:** https://arxiv.org/pdf/2601.22781v1
+
+## 17630. MuCO: Generative Peptide Cyclization Empowered by Multi-stage Conformation Optimization
+
+- **ID:** `2602.11189` · **primary:** `q-bio.BM` · **Submitted:** 2026-01-30
+- **Authors:** Yitian Wang, Fanmeng Wang, Angxiao Yue, et al.
+- **abs:** https://arxiv.org/abs/2602.11189 · **PDF:** https://arxiv.org/pdf/2602.11189v1
+
+## 17631. RASST: Fast Cross-modal Retrieval-Augmented Simultaneous Speech Translation
+
+- **ID:** `2601.22777` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Jiaxuan Luo, Siqi Ouyang, Lei Li
+- **abs:** https://arxiv.org/abs/2601.22777 · **PDF:** https://arxiv.org/pdf/2601.22777v1
+
+## 17632. TSPO: Breaking the Double Homogenization Dilemma in Multi-turn Search Policy Optimization
+
+- **ID:** `2601.22776` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Shichao Ma, Zhiyuan Ma, Ming Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.22776 · **PDF:** https://arxiv.org/pdf/2601.22776v1
+
+## 17633. GRANITE: A Generalized Regional Framework for Identifying Agreement in Feature-Based Explanations
+
+- **ID:** `2601.22771` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Julia Herbinger, Gabriel Laberge, Maximilian Muschalik, et al.
+- **abs:** https://arxiv.org/abs/2601.22771 · **PDF:** https://arxiv.org/pdf/2601.22771v1
+
+## 17634. Beyond Abstract Compliance: Operationalising trust in AI as a moral relationship
+
+- **ID:** `2601.22769` · **primary:** `cs.CY` · **Submitted:** 2026-01-30
+- **Authors:** Lameck Mbangula Amugongo, Tutaleni Asino, Nicola J Bidwell
+- **abs:** https://arxiv.org/abs/2601.22769 · **PDF:** https://arxiv.org/pdf/2601.22769v1
+
+## 17635. Sparse Attention as Compact Kernel Regression
+
+- **ID:** `2601.22766` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Saul Santos, Nuno Gonçalves, Daniel C. McNamee, et al.
+- **abs:** https://arxiv.org/abs/2601.22766 · **PDF:** https://arxiv.org/pdf/2601.22766v2
+
+## 17636. Bayesian Matrix Completion Under Geometric Constraints
+
+- **ID:** `2601.22765` · **primary:** `eess.SP` · **Submitted:** 2026-01-30
+- **Authors:** Rohit Varma Chiluvuri, Santosh Nannuru
+- **abs:** https://arxiv.org/abs/2601.22765 · **PDF:** https://arxiv.org/pdf/2601.22765v1
+
+## 17637. How Far Can Pretrained LLMs Go in Symbolic Music? Controlled Comparisons of Supervised and Preference-based Adaptation
+
+- **ID:** `2601.22764` · **primary:** `cs.SD` · **Submitted:** 2026-01-30
+- **Authors:** Deepak Kumar, Emmanouil Karystinaios, Gerhard Widmer, et al.
+- **abs:** https://arxiv.org/abs/2601.22764 · **PDF:** https://arxiv.org/pdf/2601.22764v1
+
+## 17638. AscendCraft: Automatic Ascend NPU Kernel Generation via DSL-Guided Transcompilation
+
+- **ID:** `2601.22760` · **primary:** `cs.DC` · **Submitted:** 2026-01-30
+- **Authors:** Zhongzhen Wen, Shudi Shao, Zhong Li, et al.
+- **abs:** https://arxiv.org/abs/2601.22760 · **PDF:** https://arxiv.org/pdf/2601.22760v1
+
+## 17639. Qualitative Evaluation of LLM-Designed GUI
+
+- **ID:** `2601.22759` · **primary:** `cs.HC` · **Submitted:** 2026-01-30
+- **Authors:** Bartosz Sawicki, Tomasz Les, Dariusz Parzych, et al.
+- **abs:** https://arxiv.org/abs/2601.22759 · **PDF:** https://arxiv.org/pdf/2601.22759v1
+
+## 17640. AutoRefine: From Trajectories to Reusable Expertise for Continual LLM Agent Refinement
+
+- **ID:** `2601.22758` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Libin Qiu, Zhirong Gao, Junfu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.22758 · **PDF:** https://arxiv.org/pdf/2601.22758v1
+
+## 17641. Unveiling Scaling Behaviors in Molecular Language Models: Effects of Model Size, Data, and Representation
+
+- **ID:** `2601.22757` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Dong Xu, Qihua Pan, Sisi Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.22757 · **PDF:** https://arxiv.org/pdf/2601.22757v1
+
+## 17642. Understanding Generalization from Embedding Dimension and Distributional Convergence
+
+- **ID:** `2601.22756` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Junjie Yu, Zhuoli Ouyang, Haotian Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.22756 · **PDF:** https://arxiv.org/pdf/2601.22756v1
+
+## 17643. Procedural Knowledge Extraction from Industrial Troubleshooting Guides Using Vision Language Models
+
+- **ID:** `2601.22754` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Guillermo Gil de Avalle, Laura Maruster, Christos Emmanouilidis
+- **abs:** https://arxiv.org/abs/2601.22754 · **PDF:** https://arxiv.org/pdf/2601.22754v1
+
+## 17644. EEO-TFV: Escape-Explore Optimizer for Web-Scale Time-Series Forecasting and Vision Analysis
+
+- **ID:** `2602.02551` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Hua Wang, Jinghao Lu, Fan Zhang
+- **abs:** https://arxiv.org/abs/2602.02551 · **PDF:** https://arxiv.org/pdf/2602.02551v1
+
+## 17645. OSNIP: Breaking the Privacy-Utility-Efficiency Trilemma in LLM Inference via Obfuscated Semantic Null Space
+
+- **ID:** `2601.22752` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Zhiyuan Cao, Zeyu Ma, Chenhao Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.22752 · **PDF:** https://arxiv.org/pdf/2601.22752v1
+
+## 17646. Discovering Scaling Exponents with Physics-Informed Müntz-Szász Networks
+
+- **ID:** `2601.22751` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Gnankan Landry Regis N'guessan, Bum Jun Kim
+- **abs:** https://arxiv.org/abs/2601.22751 · **PDF:** https://arxiv.org/pdf/2601.22751v1
+
+## 17647. UrbanMoE: A Sparse Multi-Modal Mixture-of-Experts Framework for Multi-Task Urban Region Profiling
+
+- **ID:** `2601.22746` · **primary:** `cs.ET` · **Submitted:** 2026-01-30
+- **Authors:** Pingping Liu, Jiamiao Liu, Zijian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22746 · **PDF:** https://arxiv.org/pdf/2601.22746v1
+
+## 17648. Is Softmax Loss All You Need? A Principled Analysis of Softmax-family Loss
+
+- **ID:** `2601.22745` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yuanhao Pu, Defu Lian, Enhong Chen
+- **abs:** https://arxiv.org/abs/2601.22745 · **PDF:** https://arxiv.org/pdf/2601.22745v1
+
+## 17649. Beauty and the Beast: Imperceptible Perturbations Against Diffusion-Based Face Swapping via Directional Attribute Editing
+
+- **ID:** `2601.22744` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Yilong Huang, Songze Li
+- **abs:** https://arxiv.org/abs/2601.22744 · **PDF:** https://arxiv.org/pdf/2601.22744v1
+
+## 17650. HyPAC: Cost-Efficient LLMs-Human Hybrid Annotation with PAC Error Guarantees
+
+- **ID:** `2602.02550` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Hao Zeng, Huipeng Huang, Xinhao Qu, et al.
+- **abs:** https://arxiv.org/abs/2602.02550 · **PDF:** https://arxiv.org/pdf/2602.02550v1
+
+## 17651. AR-BENCH: Benchmarking Legal Reasoning with Judgment Error Detection, Classification and Correction
+
+- **ID:** `2601.22742` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Yifei Li, Richong Zhang, Wanyu Tu, et al.
+- **abs:** https://arxiv.org/abs/2601.22742 · **PDF:** https://arxiv.org/pdf/2601.22742v1
+
+## 17652. Decomposing Epistemic Uncertainty for Causal Decision Making
+
+- **ID:** `2601.22736` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Md Musfiqur Rahman, Ziwei Jiang, Hilaf Hasson, et al.
+- **abs:** https://arxiv.org/abs/2601.22736 · **PDF:** https://arxiv.org/pdf/2601.22736v1
+
+## 17653. MM-THEBench: Do Reasoning MLLMs Think Reasonably?
+
+- **ID:** `2601.22735` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Zhidian Huang, Zijun Yao, Ji Qi, et al.
+- **abs:** https://arxiv.org/abs/2601.22735 · **PDF:** https://arxiv.org/pdf/2601.22735v1
+
+## 17654. AI-Generated Image Detectors Overrely on Global Artifacts: Evidence from Inpainting Exchange
+
+- **ID:** `2602.00192` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Elif Nebioglu, Emirhan Bilgiç, Adrian Popescu
+- **abs:** https://arxiv.org/abs/2602.00192 · **PDF:** https://arxiv.org/pdf/2602.00192v1
+
+## 17655. ImgCoT: Compressing Long Chain of Thought into Compact Visual Tokens for Efficient Reasoning of Large Language Model
+
+- **ID:** `2601.22730` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Xiaoshu Chen, Sihang Zhou, Ke Liang, et al.
+- **abs:** https://arxiv.org/abs/2601.22730 · **PDF:** https://arxiv.org/pdf/2601.22730v1
+
+## 17656. TDPNavigator-Placer: Thermal- and Wirelength-Aware Chiplet Placement in 2.5D Systems Through Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.11187` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yubo Hou, Furen Zhuang, Partha Pratim Kundu, et al.
+- **abs:** https://arxiv.org/abs/2602.11187 · **PDF:** https://arxiv.org/pdf/2602.11187v1
+
+## 17657. GEPC: Group-Equivariant Posterior Consistency for Out-of-Distribution Detection in Diffusion Models
+
+- **ID:** `2602.00191` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yadang Alexis Rouzoumka, Jean Pinsolle, Eugénie Terreaux, et al.
+- **abs:** https://arxiv.org/abs/2602.00191 · **PDF:** https://arxiv.org/pdf/2602.00191v2
+
+## 17658. OpenVTON-Bench: A Large-Scale High-Resolution Benchmark for Controllable Virtual Try-On Evaluation
+
+- **ID:** `2601.22725` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Jin Li, Tao Chen, Shuai Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.22725 · **PDF:** https://arxiv.org/pdf/2601.22725v2
+
+## 17659. A Cross-Domain Graph Learning Protocol for Single-Step Molecular Geometry Refinement
+
+- **ID:** `2601.22723` · **primary:** `physics.chem-ph` · **Submitted:** 2026-01-30
+- **Authors:** Chengchun Liu, Wendi Cai, Boxuan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.22723 · **PDF:** https://arxiv.org/pdf/2601.22723v1
+
+## 17660. Local Intrinsic Dimension of Representations Predicts Alignment and Generalization in AI Models and Human Brain
+
+- **ID:** `2601.22722` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Junjie Yu, Wenxiao Ma, Chen Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.22722 · **PDF:** https://arxiv.org/pdf/2601.22722v1
+
+## 17661. AEGIS: White-Box Attack Path Generation using LLMs and Training Effectiveness Evaluation for Large-Scale Cyber Defence Exercises
+
+- **ID:** `2601.22720` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Ivan K. Tung, Yu Xiang Shi, Alex Chien, et al.
+- **abs:** https://arxiv.org/abs/2601.22720 · **PDF:** https://arxiv.org/pdf/2601.22720v1
+
+## 17662. From Gameplay Traces to Game Mechanics: Causal Induction with Large Language Models
+
+- **ID:** `2602.00190` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Mohit Jiwatode, Alexander Dockhorn, Bodo Rosenhahn
+- **abs:** https://arxiv.org/abs/2602.00190 · **PDF:** https://arxiv.org/pdf/2602.00190v1
+
+## 17663. A Step Back: Prefix Importance Ratio Stabilizes Policy Optimization
+
+- **ID:** `2601.22718` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Shiye Lei, Zhihao Cheng, Dacheng Tao
+- **abs:** https://arxiv.org/abs/2601.22718 · **PDF:** https://arxiv.org/pdf/2601.22718v1
+
+## 17664. Breaking the Blocks: Continuous Low-Rank Decomposed Scaling for Unified LLM Quantization and Adaptation
+
+- **ID:** `2601.22716` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Pingzhi Tang, Ruijie Zhou, Fanxu Meng, et al.
+- **abs:** https://arxiv.org/abs/2601.22716 · **PDF:** https://arxiv.org/pdf/2601.22716v1
+
+## 17665. Inference-time Alignment via Sparse Junction Steering
+
+- **ID:** `2602.21215` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Runyi Hu, Jie Zhang, Shiqian Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.21215 · **PDF:** https://arxiv.org/pdf/2602.21215v1
+
+## 17666. Vision-Language Models Unlock Task-Centric Latent Actions
+
+- **ID:** `2601.22714` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Alexander Nikulin, Ilya Zisman, Albina Klepach, et al.
+- **abs:** https://arxiv.org/abs/2601.22714 · **PDF:** https://arxiv.org/pdf/2601.22714v1
+
+## 17667. ToolTok: Tool Tokenization for Efficient and Generalizable GUI Agents
+
+- **ID:** `2602.02548` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Xiaoce Wang, Guibin Zhang, Junzhe Li, et al.
+- **abs:** https://arxiv.org/abs/2602.02548 · **PDF:** https://arxiv.org/pdf/2602.02548v1
+
+## 17668. GAC-KAN: An Ultra-Lightweight GNSS Interference Classifier for GenAI-Powered Consumer Edge Devices
+
+- **ID:** `2602.11186` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Zhihan Zeng, Kaihe Wang, Zhongpei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.11186 · **PDF:** https://arxiv.org/pdf/2602.11186v1
+
+## 17669. SQUAD: Scalable Quorum Adaptive Decisions via ensemble of early exit neural networks
+
+- **ID:** `2601.22711` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Matteo Gambella, Fabrizio Pittorino, Giuliano Casale, et al.
+- **abs:** https://arxiv.org/abs/2601.22711 · **PDF:** https://arxiv.org/pdf/2601.22711v1
+
+## 17670. AlienLM: Alienization of Language for API-Boundary Privacy in Black-Box LLMs
+
+- **ID:** `2601.22710` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Jaehee Kim, Pilsung Kang
+- **abs:** https://arxiv.org/abs/2601.22710 · **PDF:** https://arxiv.org/pdf/2601.22710v1
+
+## 17671. Gated Relational Alignment via Confidence-based Distillation for Efficient VLMs
+
+- **ID:** `2601.22709` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Yanlong Chen, Amirhossein Habibian, Luca Benini, et al.
+- **abs:** https://arxiv.org/abs/2601.22709 · **PDF:** https://arxiv.org/pdf/2601.22709v2
+
+## 17672. A Unified Study of LoRA Variants: Taxonomy, Review, Codebase, and Empirical Evaluation
+
+- **ID:** `2601.22708` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Haonan He, Jingqi Ye, Minglei Li, et al.
+- **abs:** https://arxiv.org/abs/2601.22708 · **PDF:** https://arxiv.org/pdf/2601.22708v1
+
+## 17673. Deep Learning-Based Early-Stage IR-Drop Estimation via CNN Surrogate Modeling
+
+- **ID:** `2601.22707` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Ritesh Bhadana
+- **abs:** https://arxiv.org/abs/2601.22707 · **PDF:** https://arxiv.org/pdf/2601.22707v1
+
+## 17674. Metric Hub: A metric library and practical selection workflow for use-case-driven data quality assessment in medical AI
+
+- **ID:** `2601.22702` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Katinka Becker, Maximilian P. Oppelt, Tobias S. Zech, et al.
+- **abs:** https://arxiv.org/abs/2601.22702 · **PDF:** https://arxiv.org/pdf/2601.22702v1
+
+## 17675. Best-of-Q: Improving VLM agents with Q-function Action Ranking at Inference
+
+- **ID:** `2601.22701` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Emilien Biré, María Santos, Kai Yuan
+- **abs:** https://arxiv.org/abs/2601.22701 · **PDF:** https://arxiv.org/pdf/2601.22701v1
+
+## 17676. Models Know Models Best: Evaluation via Model-Preferred Formats
+
+- **ID:** `2601.22699` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Joonhak Lee, Sungmok Jung, Jongyeon Park, et al.
+- **abs:** https://arxiv.org/abs/2601.22699 · **PDF:** https://arxiv.org/pdf/2601.22699v1
+
+## 17677. Bi-MCQ: Reformulating Vision-Language Alignment for Negation Understanding
+
+- **ID:** `2601.22696` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Tae Hun Kim, Hyun Gyu Lee
+- **abs:** https://arxiv.org/abs/2601.22696 · **PDF:** https://arxiv.org/pdf/2601.22696v1
+
+## 17678. PEAR: Pixel-aligned Expressive humAn mesh Recovery
+
+- **ID:** `2601.22693` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Jiahao Wu, Yunfei Liu, Lijian Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.22693 · **PDF:** https://arxiv.org/pdf/2601.22693v2
+
+## 17679. FNF: Functional Network Fingerprint for Large Language Models
+
+- **ID:** `2601.22692` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Yiheng Liu, Junhao Ning, Sichen Xia, et al.
+- **abs:** https://arxiv.org/abs/2601.22692 · **PDF:** https://arxiv.org/pdf/2601.22692v1
+
+## 17680. Do Transformers Have the Ability for Periodicity Generalization?
+
+- **ID:** `2601.22690` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Huanyu Liu, Ge Li, Yihong Dong, et al.
+- **abs:** https://arxiv.org/abs/2601.22690 · **PDF:** https://arxiv.org/pdf/2601.22690v1
+
+## 17681. TSLM: Tree-Structured Language Modeling for Divergent Thinking
+
+- **ID:** `2601.22688` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Doyoung Kim, Jaehyeok Doo, Minjoon Seo
+- **abs:** https://arxiv.org/abs/2601.22688 · **PDF:** https://arxiv.org/pdf/2601.22688v1
+
+## 17682. LPIPS-AttnWav2Lip: Generic Audio-Driven lip synchronization for Talking Head Generation in the Wild
+
+- **ID:** `2602.00189` · **primary:** `cs.SD` · **Submitted:** 2026-01-30
+- **Authors:** Zhipeng Chen, Xinheng Wang, Lun Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.00189 · **PDF:** https://arxiv.org/pdf/2602.00189v1
+
+## 17683. Learning to Price: Interpretable Attribute-Level Models for Dynamic Markets
+
+- **ID:** `2602.00188` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Srividhya Sethuraman, Chandrashekar Lakshminarayanan
+- **abs:** https://arxiv.org/abs/2602.00188 · **PDF:** https://arxiv.org/pdf/2602.00188v1
+
+## 17684. Stabilizing Consistency Training: A Flow Map Analysis and Self-Distillation
+
+- **ID:** `2601.22679` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Youngjoong Kim, Duhoe Kim, Woosung Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.22679 · **PDF:** https://arxiv.org/pdf/2601.22679v1
+
+## 17685. Full-Graph vs. Mini-Batch Training: Comprehensive Analysis from a Batch Size and Fan-Out Size Perspective
+
+- **ID:** `2601.22678` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Mengfan Liu, Da Zheng, Junwei Su, et al.
+- **abs:** https://arxiv.org/abs/2601.22678 · **PDF:** https://arxiv.org/pdf/2601.22678v1
+
+## 17686. Fire on Motion: Optimizing Video Pass-bands for Efficient Spiking Action Recognition
+
+- **ID:** `2601.22675` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Shuhan Ye, Yuanbin Qian, Yi Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.22675 · **PDF:** https://arxiv.org/pdf/2601.22675v1
+
+## 17687. Beyond Fixed Rounds: Data-Free Early Stopping for Practical Federated Learning
+
+- **ID:** `2601.22669` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Youngjoon Lee, Hyukjoon Lee, Seungrok Jung, et al.
+- **abs:** https://arxiv.org/abs/2601.22669 · **PDF:** https://arxiv.org/pdf/2601.22669v2
+
+## 17688. From Horizontal Layering to Vertical Integration: A Comparative Study of the AI-Driven Software Development Paradigm
+
+- **ID:** `2601.22667` · **primary:** `cs.SE` · **Submitted:** 2026-01-30
+- **Authors:** Chi Zhang, Zehan Li, Ziqian Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.22667 · **PDF:** https://arxiv.org/pdf/2601.22667v1
+
+## 17689. Real-Time Aligned Reward Model beyond Semantics
+
+- **ID:** `2601.22664` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Zixuan Huang, Xin Xia, Yuxi Ren, et al.
+- **abs:** https://arxiv.org/abs/2601.22664 · **PDF:** https://arxiv.org/pdf/2601.22664v3
+
+## 17690. Unsupervised Synthetic Image Attribution: Alignment and Disentanglement
+
+- **ID:** `2601.22663` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Zongfang Liu, Guangyi Chen, Boyang Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.22663 · **PDF:** https://arxiv.org/pdf/2601.22663v1
+
+## 17691. Task-Aware LLM Council with Adaptive Decision Pathways for Decision Support
+
+- **ID:** `2601.22662` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Wei Zhu, Lixing Yu, Hao-Ren Yao, et al.
+- **abs:** https://arxiv.org/abs/2601.22662 · **PDF:** https://arxiv.org/pdf/2601.22662v1
+
+## 17692. Layerwise Progressive Freezing Enables STE-Free Training of Deep Binary Neural Networks
+
+- **ID:** `2601.22660` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Evan Gibson Smith, Bashima Islam
+- **abs:** https://arxiv.org/abs/2601.22660 · **PDF:** https://arxiv.org/pdf/2601.22660v1
+
+## 17693. NAG: A Unified Native Architecture for Encoder-free Text-Graph Modeling in Language Models
+
+- **ID:** `2601.22657` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Haisong Gong, Zhibo Liu, Qiang Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.22657 · **PDF:** https://arxiv.org/pdf/2601.22657v1
+
+## 17694. Human-Centered Explainability in AI-Enhanced UI Security Interfaces: Designing Trustworthy Copilots for Cybersecurity Analysts
+
+- **ID:** `2601.22653` · **primary:** `cs.HC` · **Submitted:** 2026-01-30
+- **Authors:** Mona Rajhans
+- **abs:** https://arxiv.org/abs/2601.22653 · **PDF:** https://arxiv.org/pdf/2601.22653v1
+
+## 17695. Spectral Gradient Descent Mitigates Anisotropy-Driven Misalignment: A Case Study in Phase Retrieval
+
+- **ID:** `2601.22652` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Guillaume Braun, Han Bao, Wei Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.22652 · **PDF:** https://arxiv.org/pdf/2601.22652v1
+
+## 17696. GUDA: Counterfactual Group-wise Training Data Attribution for Diffusion Models via Unlearning
+
+- **ID:** `2601.22651` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Naoki Murata, Yuhta Takida, Chieh-Hsin Lai, et al.
+- **abs:** https://arxiv.org/abs/2601.22651 · **PDF:** https://arxiv.org/pdf/2601.22651v1
+
+## 17697. Generative and Nonparametric Approaches for Conditional Distribution Estimation: Methods, Perspectives, and Comparative Evaluations
+
+- **ID:** `2601.22650` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Yen-Shiu Chin, Zhi-Yu Jou, Toshinari Morimoto, et al.
+- **abs:** https://arxiv.org/abs/2601.22650 · **PDF:** https://arxiv.org/pdf/2601.22650v1
+
+## 17698. Spectra: Rethinking Optimizers for LLMs Under Spectral Anisotropy
+
+- **ID:** `2602.11185` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Zhendong Huang, Hengjie Cao, Fang Dong, et al.
+- **abs:** https://arxiv.org/abs/2602.11185 · **PDF:** https://arxiv.org/pdf/2602.11185v1
+
+## 17699. UCPO: Uncertainty-Aware Policy Optimization
+
+- **ID:** `2601.22648` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Xianzhou Zeng, Jing Huang, Chunmei Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.22648 · **PDF:** https://arxiv.org/pdf/2601.22648v1
+
+## 17700. Test-Time Mixture of World Models for Embodied Agents in Dynamic Environments
+
+- **ID:** `2601.22647` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Jinwoo Jang, Minjong Yoo, Sihyung Yoon, et al.
+- **abs:** https://arxiv.org/abs/2601.22647 · **PDF:** https://arxiv.org/pdf/2601.22647v1
+
+## 17701. Beyond Medical Chatbots: Meddollina and the Rise of Continuous Clinical Intelligence
+
+- **ID:** `2601.22645` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Vaibhav Ram S. V. N. S, Swetanshu Agrawal, Samudra Banerjee, et al.
+- **abs:** https://arxiv.org/abs/2601.22645 · **PDF:** https://arxiv.org/pdf/2601.22645v1
+
+## 17702. Scaling GraphLLM with Bilevel-Optimized Sparse Querying
+
+- **ID:** `2602.09038` · **primary:** `cs.DB` · **Submitted:** 2026-01-30
+- **Authors:** Yangzhe Peng, Haiquan Qiu, Quanming Yao, et al.
+- **abs:** https://arxiv.org/abs/2602.09038 · **PDF:** https://arxiv.org/pdf/2602.09038v1
+
+## 17703. Pushing the Boundaries of Natural Reasoning: Interleaved Bonus from Formal-Logic Verification
+
+- **ID:** `2601.22642` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Chuxue Cao, Jinluan Yang, Haoran Li, et al.
+- **abs:** https://arxiv.org/abs/2601.22642 · **PDF:** https://arxiv.org/pdf/2601.22642v1
+
+## 17704. KBVQ-MoE: KLT-guided SVD with Bias-Corrected Vector Quantization for MoE Large Language Models
+
+- **ID:** `2602.11184` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Zukang Xu, Zhixiong Zhao, Xing Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.11184 · **PDF:** https://arxiv.org/pdf/2602.11184v2
+
+## 17705. ScholarPeer: A Context-Aware Multi-Agent Framework for Automated Peer Review
+
+- **ID:** `2601.22638` · **primary:** `cs.MA` · **Submitted:** 2026-01-30
+- **Authors:** Palash Goyal, Mihir Parmar, Yiwen Song, et al.
+- **abs:** https://arxiv.org/abs/2601.22638 · **PDF:** https://arxiv.org/pdf/2601.22638v1
+
+## 17706. Training Beyond Convergence: Grokking nnU-Net for Glioma Segmentation in Sub-Saharan MRI
+
+- **ID:** `2601.22637` · **primary:** `eess.IV` · **Submitted:** 2026-01-30
+- **Authors:** Mohtady Barakat, Omar Salah, Ahmed Yasser, et al.
+- **abs:** https://arxiv.org/abs/2601.22637 · **PDF:** https://arxiv.org/pdf/2601.22637v1
+
+## 17707. Statistical Estimation of Adversarial Risk in Large Language Models under Best-of-N Sampling
+
+- **ID:** `2601.22636` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Mingqian Feng, Xiaodong Liu, Weiwei Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.22636 · **PDF:** https://arxiv.org/pdf/2601.22636v2
+
+## 17708. What can Computer Vision learn from Ranganathan?
+
+- **ID:** `2601.22634` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Mayukh Bagchi, Fausto Giunchiglia
+- **abs:** https://arxiv.org/abs/2601.22634 · **PDF:** https://arxiv.org/pdf/2601.22634v1
+
+## 17709. MCP-Diag: A Deterministic, Protocol-Driven Architecture for AI-Native Network Diagnostics
+
+- **ID:** `2601.22633` · **primary:** `cs.NI` · **Submitted:** 2026-01-30
+- **Authors:** Devansh Lodha, Mohit Panchal, Sameer G. Kulkarni
+- **abs:** https://arxiv.org/abs/2601.22633 · **PDF:** https://arxiv.org/pdf/2601.22633v1
+
+## 17710. DART-ing Through the Drift: Dynamic Tracing of Knowledge Neurons for Adaptive Inference-Time Pruning
+
+- **ID:** `2601.22632` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Abhishek Tyagi, Yunuo Cen, Shrey Dhorajiya, et al.
+- **abs:** https://arxiv.org/abs/2601.22632 · **PDF:** https://arxiv.org/pdf/2601.22632v1
+
+## 17711. PEFT-MuTS: A Multivariate Parameter-Efficient Fine-Tuning Framework for Remaining Useful Life Prediction based on Cross-domain Time Series Representation Model
+
+- **ID:** `2601.22631` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** En Fu, Yanyan Hu, Changhua Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.22631 · **PDF:** https://arxiv.org/pdf/2601.22631v1
+
+## 17712. Time-Annealed Perturbation Sampling: Diverse Generation for Diffusion Language Models
+
+- **ID:** `2601.22629` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Jingxuan Wu, Zhenglin Wan, Xingrui Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.22629 · **PDF:** https://arxiv.org/pdf/2601.22629v2
+
+## 17713. TTCS: Test-Time Curriculum Synthesis for Self-Evolving
+
+- **ID:** `2601.22628` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Chengyi Yang, Zhishang Xiang, Yunbo Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.22628 · **PDF:** https://arxiv.org/pdf/2601.22628v1
+
+## 17714. RPWithPrior: Label Differential Privacy in Regression
+
+- **ID:** `2601.22625` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Haixia Liu, Ruifan Huang
+- **abs:** https://arxiv.org/abs/2601.22625 · **PDF:** https://arxiv.org/pdf/2601.22625v1
+
+## 17715. COBRA++: Enhanced COBRA Optimizer with Augmented Surrogate Pool and Reinforced Surrogate Selection
+
+- **ID:** `2601.22624` · **primary:** `cs.NE` · **Submitted:** 2026-01-30
+- **Authors:** Zipei Yu, Zhiyang Huang, Hongshu Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.22624 · **PDF:** https://arxiv.org/pdf/2601.22624v2
+
+## 17716. SYMPHONY: Synergistic Multi-agent Planning with Heterogeneous Language Model Assembly
+
+- **ID:** `2601.22623` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Wei Zhu, Zhiwen Tang, Kun Yue
+- **abs:** https://arxiv.org/abs/2601.22623 · **PDF:** https://arxiv.org/pdf/2601.22623v1
+
+## 17717. Layer-wise Swapping for Generalizable Multilingual Safety
+
+- **ID:** `2601.22620` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Hyunseo Shin, Wonseok Hwang
+- **abs:** https://arxiv.org/abs/2601.22620 · **PDF:** https://arxiv.org/pdf/2601.22620v2
+
+## 17718. EntroCut: Entropy-Guided Adaptive Truncation for Efficient Chain-of-Thought Reasoning in Small-scale Large Reasoning Models
+
+- **ID:** `2601.22617` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Hongxi Yan, Qingjie Liu, Yunhong Wang
+- **abs:** https://arxiv.org/abs/2601.22617 · **PDF:** https://arxiv.org/pdf/2601.22617v1
+
+## 17719. Stabilizing Transformer Training Through Consensus
+
+- **ID:** `2601.22614` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Shyam Venkatasubramanian, Sean Moushegian, Michael Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.22614 · **PDF:** https://arxiv.org/pdf/2601.22614v1
+
+## 17720. Local-Global Multimodal Contrastive Learning for Molecular Property Prediction
+
+- **ID:** `2601.22610` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Xiayu Liu, Zhengyi Lu, Yunhong Liao, et al.
+- **abs:** https://arxiv.org/abs/2601.22610 · **PDF:** https://arxiv.org/pdf/2601.22610v1
+
+## 17721. naPINN: Noise-Adaptive Physics-Informed Neural Networks for Recovering Physics from Corrupted Measurement
+
+- **ID:** `2602.02547` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Hankyeol Kim, Pilsung Kang
+- **abs:** https://arxiv.org/abs/2602.02547 · **PDF:** https://arxiv.org/pdf/2602.02547v1
+
+## 17722. From Self-Evolving Synthetic Data to Verifiable-Reward RL: Post-Training Multi-turn Interactive Tool-Using Agents
+
+- **ID:** `2601.22607` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Jiaxuan Gao, Jiaao Chen, Chuyi He, et al.
+- **abs:** https://arxiv.org/abs/2601.22607 · **PDF:** https://arxiv.org/pdf/2601.22607v3
+
+## 17723. Lethe:Adapter-Augmented Dual-Stream Update for Persistent Knowledge Erasure in Federated Unlearning
+
+- **ID:** `2601.22601` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Hanwei Tan, Wentai Hu, Ligang He, et al.
+- **abs:** https://arxiv.org/abs/2601.22601 · **PDF:** https://arxiv.org/pdf/2601.22601v1
+
+## 17724. An Efficient Algorithm for Thresholding Monte Carlo Tree Search
+
+- **ID:** `2601.22600` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Shoma Nameki, Atsuyoshi Nakamura, Junpei Komiyama, et al.
+- **abs:** https://arxiv.org/abs/2601.22600 · **PDF:** https://arxiv.org/pdf/2601.22600v1
+
+## 17725. D$^2$Quant: Accurate Low-bit Post-Training Weight Quantization for LLMs
+
+- **ID:** `2602.02546` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Xianglong Yan, ChengZhu Bao, Zhiteng Li, et al.
+- **abs:** https://arxiv.org/abs/2602.02546 · **PDF:** https://arxiv.org/pdf/2602.02546v2
+
+## 17726. TimeMachine-bench: A Benchmark for Evaluating Model Capabilities in Repository-Level Migration Tasks
+
+- **ID:** `2601.22597` · **primary:** `cs.SE` · **Submitted:** 2026-01-30
+- **Authors:** Ryo Fujii, Makoto Morishita, Kazuki Yano, et al.
+- **abs:** https://arxiv.org/abs/2601.22597 · **PDF:** https://arxiv.org/pdf/2601.22597v1
+
+## 17727. Learn More with Less: Uncertainty Consistency Guided Query Selection for RLVR
+
+- **ID:** `2601.22595` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Hao Yi, Yulan Hu, Xin Li, et al.
+- **abs:** https://arxiv.org/abs/2601.22595 · **PDF:** https://arxiv.org/pdf/2601.22595v1
+
+## 17728. Language Model Circuits Are Sparse in the Neuron Basis
+
+- **ID:** `2601.22594` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Aryaman Arora, Zhengxuan Wu, Jacob Steinhardt, et al.
+- **abs:** https://arxiv.org/abs/2601.22594 · **PDF:** https://arxiv.org/pdf/2601.22594v1
+
+## 17729. Heterogeneous Graph Alignment for Joint Reasoning and Interpretability
+
+- **ID:** `2601.22593` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Zahra Moslemi, Ziyi Liang, Norbert Fortin, et al.
+- **abs:** https://arxiv.org/abs/2601.22593 · **PDF:** https://arxiv.org/pdf/2601.22593v1
+
+## 17730. Beyond Alignment: Expanding Reasoning Capacity via Manifold-Reshaping Policy Optimization
+
+- **ID:** `2602.02545` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Dayu Wang, Jiaye Yang, Weikang Li, et al.
+- **abs:** https://arxiv.org/abs/2602.02545 · **PDF:** https://arxiv.org/pdf/2602.02545v1
+
+## 17731. FedCARE: Federated Unlearning with Conflict-Aware Projection and Relearning-Resistant Recovery
+
+- **ID:** `2601.22589` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yue Li, Mingmin Chu, Xilei Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.22589 · **PDF:** https://arxiv.org/pdf/2601.22589v1
+
+## 17732. Rethinking LLM-as-a-Judge: Representation-as-a-Judge with Small Language Models via Semantic Capacity Asymmetry
+
+- **ID:** `2601.22588` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Zhuochun Li, Yong Zhang, Ming Li, et al.
+- **abs:** https://arxiv.org/abs/2601.22588 · **PDF:** https://arxiv.org/pdf/2601.22588v1
+
+## 17733. WED-Net: A Weather-Effect Disentanglement Network with Causal Augmentation for Urban Flow Prediction
+
+- **ID:** `2601.22586` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Qian Hong, Siyuan Chang, Xiao Zhou
+- **abs:** https://arxiv.org/abs/2601.22586 · **PDF:** https://arxiv.org/pdf/2601.22586v1
+
+## 17734. HetCCL: Accelerating LLM Training with Heterogeneous GPUs
+
+- **ID:** `2601.22585` · **primary:** `cs.DC` · **Submitted:** 2026-01-30
+- **Authors:** Heehoon Kim, Jaehwan Lee, Taejeoung Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.22585 · **PDF:** https://arxiv.org/pdf/2601.22585v1
+
+## 17735. QUASAR: A Universal Autonomous System for Atomistic Simulation and a Benchmark of Its Capabilities
+
+- **ID:** `2602.00185` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-30
+- **Authors:** Fengxu Yang, Jack D. Evans
+- **abs:** https://arxiv.org/abs/2602.00185 · **PDF:** https://arxiv.org/pdf/2602.00185v1
+
+## 17736. Visible Singularities Guided Correlation Network for Limited-Angle CT Reconstruction
+
+- **ID:** `2602.00184` · **primary:** `eess.IV` · **Submitted:** 2026-01-30
+- **Authors:** Yiyang Wen, Liu Shi, Zekun Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.00184 · **PDF:** https://arxiv.org/pdf/2602.00184v1
+
+## 17737. MC-GRPO: Median-Centered Group Relative Policy Optimization for Small-Rollout Reinforcement Learning
+
+- **ID:** `2601.22582` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Youngeun Kim
+- **abs:** https://arxiv.org/abs/2601.22582 · **PDF:** https://arxiv.org/pdf/2601.22582v1
+
+## 17738. Cross-Domain Few-Shot Learning for Hyperspectral Image Classification Based on Mixup Foundation Model
+
+- **ID:** `2601.22581` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Naeem Paeedeh, Mahardhika Pratama, Ary Shiddiqi, et al.
+- **abs:** https://arxiv.org/abs/2601.22581 · **PDF:** https://arxiv.org/pdf/2601.22581v1
+
+## 17739. SPA-Cache: Singular Proxies for Adaptive Caching in Diffusion Language Models
+
+- **ID:** `2602.02544` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Wenhao Sun, Rong-Cheng Tu, Yifu Ding, et al.
+- **abs:** https://arxiv.org/abs/2602.02544 · **PDF:** https://arxiv.org/pdf/2602.02544v1
+
+## 17740. SpanNorm: Reconciling Training Stability and Performance in Deep Transformers
+
+- **ID:** `2601.22580` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Chao Wang, Bei Li, Jiaqi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22580 · **PDF:** https://arxiv.org/pdf/2601.22580v1
+
+## 17741. Non-Intrusive Graph-Based Bot Detection for E-Commerce Using Inductive Graph Neural Networks
+
+- **ID:** `2601.22579` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Sichen Zhao, Zhiming Xue, Yalun Qi, et al.
+- **abs:** https://arxiv.org/abs/2601.22579 · **PDF:** https://arxiv.org/pdf/2601.22579v6
+
+## 17742. FedDis: A Causal Disentanglement Framework for Federated Traffic Prediction
+
+- **ID:** `2601.22578` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Chengyang Zhou, Zijian Zhang, Chunxu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22578 · **PDF:** https://arxiv.org/pdf/2601.22578v1
+
+## 17743. RPP: A Certified Poisoned-Sample Detection Framework for Backdoor Attacks under Dataset Imbalance
+
+- **ID:** `2602.00183` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Miao Lin, Feng Yu, Rui Ning, et al.
+- **abs:** https://arxiv.org/abs/2602.00183 · **PDF:** https://arxiv.org/pdf/2602.00183v1
+
+## 17744. PhoStream: Benchmarking Real-World Streaming for Omnimodal Assistants in Mobile Scenarios
+
+- **ID:** `2601.22575` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Xudong Lu, Huankang Guan, Yang Bo, et al.
+- **abs:** https://arxiv.org/abs/2601.22575 · **PDF:** https://arxiv.org/pdf/2601.22575v1
+
+## 17745. Mitigating Hallucinations in Video Large Language Models via Spatiotemporal-Semantic Contrastive Decoding
+
+- **ID:** `2601.22574` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Yuansheng Gao, Jinman Zhao, Tong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22574 · **PDF:** https://arxiv.org/pdf/2601.22574v1
+
+## 17746. PerfGuard: A Performance-Aware Agent for Visual Content Generation
+
+- **ID:** `2601.22571` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Zhipeng Chen, Zhongrui Zhang, Chao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22571 · **PDF:** https://arxiv.org/pdf/2601.22571v2
+
+## 17747. Leveraging Data to Say No: Memory Augmented Plug-and-Play Selective Prediction
+
+- **ID:** `2601.22570` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Aditya Sarkar, Yi Li, Jiacheng Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.22570 · **PDF:** https://arxiv.org/pdf/2601.22570v1
+
+## 17748. Whispers of Wealth: Red-Teaming Google's Agent Payments Protocol via Prompt Injection
+
+- **ID:** `2601.22569` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Tanusree Debi, Wentian Zhu
+- **abs:** https://arxiv.org/abs/2601.22569 · **PDF:** https://arxiv.org/pdf/2601.22569v1
+
+## 17749. EUGens: Efficient, Unified, and General Dense Layers
+
+- **ID:** `2601.22563` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Sang Min Kim, Byeongchan Kim, Arijit Sehanobish, et al.
+- **abs:** https://arxiv.org/abs/2601.22563 · **PDF:** https://arxiv.org/pdf/2601.22563v2
+
+## 17750. EigenAI: Deterministic Inference, Verifiable Results
+
+- **ID:** `2602.00182` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** David Ribeiro Alves, Vishnu Patankar, Matheus Pereira, et al.
+- **abs:** https://arxiv.org/abs/2602.00182 · **PDF:** https://arxiv.org/pdf/2602.00182v1
+
+## 17751. CamReasoner: Reinforcing Camera Movement Understanding via Structured Spatial Reasoning
+
+- **ID:** `2602.00181` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Hang Wu, Yujun Cai, Zehao Li, et al.
+- **abs:** https://arxiv.org/abs/2602.00181 · **PDF:** https://arxiv.org/pdf/2602.00181v2
+
+## 17752. Spec-Driven Development:From Code to Contract in the Age of AI Coding Assistants
+
+- **ID:** `2602.00180` · **primary:** `cs.SE` · **Submitted:** 2026-01-30
+- **Authors:** Deepak Babu Piskala
+- **abs:** https://arxiv.org/abs/2602.00180 · **PDF:** https://arxiv.org/pdf/2602.00180v1
+
+## 17753. Exo-Plore: Exploring Exoskeleton Control Space through Human-aligned Simulation
+
+- **ID:** `2601.22550` · **primary:** `cs.RO` · **Submitted:** 2026-01-30
+- **Authors:** Geonho Leem, Jaedong Lee, Jehee Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.22550 · **PDF:** https://arxiv.org/pdf/2601.22550v1
+
+## 17754. How Understanding Forecast Uncertainty Resolves the Explainability Problem in Machine Learning Models
+
+- **ID:** `2602.00179` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Joseph L. Breeden
+- **abs:** https://arxiv.org/abs/2602.00179 · **PDF:** https://arxiv.org/pdf/2602.00179v3
+
+## 17755. Are LLM Evaluators Really Narcissists? Sanity Checking Self-Preference Evaluations
+
+- **ID:** `2601.22548` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Dani Roytburg, Matthew Bozoukov, Matthew Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.22548 · **PDF:** https://arxiv.org/pdf/2601.22548v3
+
+## 17756. Towards the Holographic Characteristic of LLMs for Efficient Short-text Generation
+
+- **ID:** `2601.22546` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Shun Qian, Bingquan Liu, Chengjie Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.22546 · **PDF:** https://arxiv.org/pdf/2601.22546v1
+
+## 17757. Adapting Reinforcement Learning for Path Planning in Constrained Parking Scenarios
+
+- **ID:** `2601.22545` · **primary:** `cs.RO` · **Submitted:** 2026-01-30
+- **Authors:** Feng Tao, Luca Paparusso, Chenyi Gu, et al.
+- **abs:** https://arxiv.org/abs/2601.22545 · **PDF:** https://arxiv.org/pdf/2601.22545v1
+
+## 17758. Toward Ultra-Long-Horizon Sequential Model Editing
+
+- **ID:** `2602.02543` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Mingda Liu, Zhenghan Zhu, Ze'an Miao, et al.
+- **abs:** https://arxiv.org/abs/2602.02543 · **PDF:** https://arxiv.org/pdf/2602.02543v2
+
+## 17759. Scalable spatial point process models for forensic footwear analysis
+
+- **ID:** `2602.07006` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Alokesh Manna, Neil Spencer, Dipak K. Dey
+- **abs:** https://arxiv.org/abs/2602.07006 · **PDF:** https://arxiv.org/pdf/2602.07006v1
+
+## 17760. Detect and Act: Automated Dynamic Optimizer through Meta-Black-Box Optimization
+
+- **ID:** `2601.22542` · **primary:** `cs.NE` · **Submitted:** 2026-01-30
+- **Authors:** Zijian Gao, Yuanting Zhong, Zeyuan Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.22542 · **PDF:** https://arxiv.org/pdf/2601.22542v1
+
+## 17761. Benchmarking Long Roll-outs of Auto-regressive Neural Operators for the Compressible Navier-Stokes Equations with Conserved Quantity Correction
+
+- **ID:** `2601.22541` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Sean Current, Chandan Kumar, Datta Gaitonde, et al.
+- **abs:** https://arxiv.org/abs/2601.22541 · **PDF:** https://arxiv.org/pdf/2601.22541v1
+
+## 17762. Neural-Inspired Posterior Approximation (NIPA)
+
+- **ID:** `2601.22539` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Babak Shahbaba, Zahra Moslemi
+- **abs:** https://arxiv.org/abs/2601.22539 · **PDF:** https://arxiv.org/pdf/2601.22539v1
+
+## 17763. Learning to Defer in Non-Stationary Time Series via Switching State-Space Models
+
+- **ID:** `2601.22538` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yannis Montreuil, Letian Yu, Axel Carlier, et al.
+- **abs:** https://arxiv.org/abs/2601.22538 · **PDF:** https://arxiv.org/pdf/2601.22538v1
+
+## 17764. Decoding in Geometry: Alleviating Embedding-Space Crowding for Complex Reasoning
+
+- **ID:** `2601.22536` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Yixin Yang, Qingxiu Dong, Zhifang Sui
+- **abs:** https://arxiv.org/abs/2601.22536 · **PDF:** https://arxiv.org/pdf/2601.22536v1
+
+## 17765. Demystifying Design Choices of Reinforcement Fine-tuning: A Batched Contextual Bandit Learning Perspective
+
+- **ID:** `2601.22532` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Hong Xie, Xiao Hu, Tao Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.22532 · **PDF:** https://arxiv.org/pdf/2601.22532v1
+
+## 17766. Learn from A Rationalist: Distilling Intermediate Interpretable Rationales
+
+- **ID:** `2601.22531` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Jiayi Dai, Randy Goebel
+- **abs:** https://arxiv.org/abs/2601.22531 · **PDF:** https://arxiv.org/pdf/2601.22531v1
+
+## 17767. Enhancing TableQA through Verifiable Reasoning Trace Reward
+
+- **ID:** `2601.22530` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Tung Sum Thomas Kwok, Xinyu Wang, Hengzhi He, et al.
+- **abs:** https://arxiv.org/abs/2601.22530 · **PDF:** https://arxiv.org/pdf/2601.22530v1
+
+## 17768. Darwinian Memory: A Training-Free Self-Regulating Memory System for GUI Agent Evolution
+
+- **ID:** `2601.22528` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Hongze Mi, Yibo Feng, WenJie Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.22528 · **PDF:** https://arxiv.org/pdf/2601.22528v1
+
+## 17769. $ρ$-$\texttt{EOS}$: Training-free Bidirectional Variable-Length Control for Masked Diffusion LLMs
+
+- **ID:** `2601.22527` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Jingyi Yang, Yuxian Jiang, Jing Shao
+- **abs:** https://arxiv.org/abs/2601.22527 · **PDF:** https://arxiv.org/pdf/2601.22527v2
+
+## 17770. Variational Bayesian Flow Network for Graph Generation
+
+- **ID:** `2601.22524` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yida Xiong, Jiameng Chen, Xiuwen Gong, et al.
+- **abs:** https://arxiv.org/abs/2601.22524 · **PDF:** https://arxiv.org/pdf/2601.22524v1
+
+## 17771. One Ring to Rule Them All: Unifying Group-Based RL via Dynamic Power-Mean Geometry
+
+- **ID:** `2601.22521` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Weisong Zhao, Tong Wang, Zichang Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.22521 · **PDF:** https://arxiv.org/pdf/2601.22521v1
+
+## 17772. Corrected Samplers for Discrete Flow Models
+
+- **ID:** `2601.22519` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Zhengyan Wan, Yidong Ouyang, Liyan Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.22519 · **PDF:** https://arxiv.org/pdf/2601.22519v1
+
+## 17773. SCOPE-PD: Explainable AI on Subjective and Clinical Objective Measurements of Parkinson's Disease for Precision Decision-Making
+
+- **ID:** `2601.22516` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Md Mezbahul Islam, John Michael Templeton, Masrur Sobhan, et al.
+- **abs:** https://arxiv.org/abs/2601.22516 · **PDF:** https://arxiv.org/pdf/2601.22516v1
+
+## 17774. Why Self-Rewarding Works: Theoretical Guarantees for Iterative Alignment of Language Models
+
+- **ID:** `2601.22513` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Shi Fu, Yingjie Wang, Shengchao Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.22513 · **PDF:** https://arxiv.org/pdf/2601.22513v2
+
+## 17775. DRL-Enabled Trajectory Planing for UAV-Assisted VLC: Optimal Altitude and Reward Design
+
+- **ID:** `2601.22512` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Tian-Tian Lin, Yi Liu, Xiao-Wei Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.22512 · **PDF:** https://arxiv.org/pdf/2601.22512v1
+
+## 17776. Mock Worlds, Real Skills: Building Small Agentic Language Models with Synthetic Tasks, Simulated Environments, and Rubric-Based Rewards
+
+- **ID:** `2601.22511` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Yuanjie Lyu, Chengyu Wang, Lei Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.22511 · **PDF:** https://arxiv.org/pdf/2601.22511v2
+
+## 17777. BeSafe-Bench: Unveiling Behavioral Safety Risks of Situated Agents in Functional Environments
+
+- **ID:** `2603.25747` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Yuxuan Li, Yi Lin, Peng Wang, et al.
+- **abs:** https://arxiv.org/abs/2603.25747 · **PDF:** https://arxiv.org/pdf/2603.25747v1
+
+## 17778. Shattered Compositionality: Counterintuitive Learning Dynamics of Transformers for Arithmetic
+
+- **ID:** `2601.22510` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Xingyu Zhao, Darsh Sharma, Rheeya Uppaal, et al.
+- **abs:** https://arxiv.org/abs/2601.22510 · **PDF:** https://arxiv.org/pdf/2601.22510v1
+
+## 17779. Keep Rehearsing and Refining: Lifelong Learning Vehicle Routing under Continually Drifting Tasks
+
+- **ID:** `2601.22509` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Jiyuan Pei, Yi Mei, Jialin Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.22509 · **PDF:** https://arxiv.org/pdf/2601.22509v1
+
+## 17780. Predicting Gene Disease Associations in Type 2 Diabetes Using Machine Learning on Single-Cell RNA-Seq Data
+
+- **ID:** `2602.09036` · **primary:** `q-bio.GN` · **Submitted:** 2026-01-30
+- **Authors:** Maria De La Luz Lomboy Toledo, Daniel Onah
+- **abs:** https://arxiv.org/abs/2602.09036 · **PDF:** https://arxiv.org/pdf/2602.09036v1
+
+## 17781. E2CAR: An Efficient 2D-CNN Framework for Real-Time EEG Artifact Removal on Edge Devices
+
+- **ID:** `2602.09035` · **primary:** `eess.SP` · **Submitted:** 2026-01-30
+- **Authors:** Haoliang Liu, Chengkun Cai, Xu Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.09035 · **PDF:** https://arxiv.org/pdf/2602.09035v1
+
+## 17782. Auto-Augmentation Contrastive Learning for Wearable-based Human Activity Recognition
+
+- **ID:** `2602.02542` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Qingyu Wu, Jianfei Shen, Feiyi Fan, et al.
+- **abs:** https://arxiv.org/abs/2602.02542 · **PDF:** https://arxiv.org/pdf/2602.02542v1
+
+## 17783. Stabilizing Diffusion Posterior Sampling by Noise--Frequency Continuation
+
+- **ID:** `2602.00176` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Feng Tian, Yixuan Li, Weili Zeng, et al.
+- **abs:** https://arxiv.org/abs/2602.00176 · **PDF:** https://arxiv.org/pdf/2602.00176v1
+
+## 17784. Fairness-Aware Performance Evaluation for Multi-Party Multi-Objective Optimization
+
+- **ID:** `2601.22497` · **primary:** `cs.NE` · **Submitted:** 2026-01-30
+- **Authors:** Zifan Zhao, Peilan Xu, Wenjian Luo
+- **abs:** https://arxiv.org/abs/2601.22497 · **PDF:** https://arxiv.org/pdf/2601.22497v1
+
+## 17785. Action-Sufficient Goal Representations
+
+- **ID:** `2601.22496` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Jinu Hyeon, Woobin Park, Hongjoon Ahn, et al.
+- **abs:** https://arxiv.org/abs/2601.22496 · **PDF:** https://arxiv.org/pdf/2601.22496v1
+
+## 17786. Gradual Fine-Tuning for Flow Matching Models
+
+- **ID:** `2601.22495` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Gudrun Thorkelsdottir, Arindam Banerjee
+- **abs:** https://arxiv.org/abs/2601.22495 · **PDF:** https://arxiv.org/pdf/2601.22495v1
+
+## 17787. SSL: Sweet Spot Learning for Differentiated Guidance in Agentic Optimization
+
+- **ID:** `2601.22491` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Jinyang Wu, Changpeng Yang, Yuhao Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.22491 · **PDF:** https://arxiv.org/pdf/2601.22491v1
+
+## 17788. Doc2Spec: Synthesizing Formal Programming Specifications from Natural Language via Grammar Induction
+
+- **ID:** `2602.04892` · **primary:** `cs.PL` · **Submitted:** 2026-01-30
+- **Authors:** Shihao Xia, Mengting He, Haomin Jia, et al.
+- **abs:** https://arxiv.org/abs/2602.04892 · **PDF:** https://arxiv.org/pdf/2602.04892v1
+
+## 17789. Elastic Spectral State Space Models for Budgeted Inference
+
+- **ID:** `2601.22488` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Dachuan Song, Xuan Wang
+- **abs:** https://arxiv.org/abs/2601.22488 · **PDF:** https://arxiv.org/pdf/2601.22488v2
+
+## 17790. AI Literacy, Safety Awareness, and STEM Career Aspirations of Australian Secondary Students: Evaluating the Impact of Workshop Interventions
+
+- **ID:** `2601.22486` · **primary:** `cs.CY` · **Submitted:** 2026-01-30
+- **Authors:** Christian Bergh, Alexandra Vassar, Natasha Banks, et al.
+- **abs:** https://arxiv.org/abs/2601.22486 · **PDF:** https://arxiv.org/pdf/2601.22486v1
+
+## 17791. FraudShield: Knowledge Graph Empowered Defense for LLMs against Fraud Attacks
+
+- **ID:** `2601.22485` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Naen Xu, Jinghuai Zhang, Ping He, et al.
+- **abs:** https://arxiv.org/abs/2601.22485 · **PDF:** https://arxiv.org/pdf/2601.22485v1
+
+## 17792. Mitigating Cognitive Inertia in Large Reasoning Models via Latent Spike Steering
+
+- **ID:** `2601.22484` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Seojin Lee, ByeongJeong Kim, Hwanhee Lee
+- **abs:** https://arxiv.org/abs/2601.22484 · **PDF:** https://arxiv.org/pdf/2601.22484v1
+
+## 17793. Changepoint Detection As Model Selection: A General Framework
+
+- **ID:** `2601.22481` · **primary:** `stat.ME` · **Submitted:** 2026-01-30
+- **Authors:** Michael Grantham, Xueheng Shi, Bertrand Clarke
+- **abs:** https://arxiv.org/abs/2601.22481 · **PDF:** https://arxiv.org/pdf/2601.22481v2
+
+## 17794. Transform-Augmented GRPO Improves Pass@k
+
+- **ID:** `2601.22478` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Khiem Le, Youssef Mroueh, Phuc Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.22478 · **PDF:** https://arxiv.org/pdf/2601.22478v2
+
+## 17795. RulePlanner: All-in-One Reinforcement Learner for Unifying Design Rules in 3D Floorplanning
+
+- **ID:** `2601.22476` · **primary:** `cs.AR` · **Submitted:** 2026-01-30
+- **Authors:** Ruizhe Zhong, Xingbo Du, Junchi Yan
+- **abs:** https://arxiv.org/abs/2601.22476 · **PDF:** https://arxiv.org/pdf/2601.22476v1
+
+## 17796. Continual Policy Distillation from Distributed Reinforcement Learning Teachers
+
+- **ID:** `2601.22475` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yuxuan Li, Qijun He, Mingqi Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.22475 · **PDF:** https://arxiv.org/pdf/2601.22475v1
+
+## 17797. The Illusion of Forgetting: Attack Unlearned Diffusion via Initial Latent Variable Optimization
+
+- **ID:** `2602.00175` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Manyi Li, Yufan Liu, Lai Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.00175 · **PDF:** https://arxiv.org/pdf/2602.00175v1
+
+## 17798. Unrewarded Exploration in Large Language Models Reveals Latent Learning from Psychology
+
+- **ID:** `2601.22474` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Jian Xiong, Jingbo Zhou, Zihan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.22474 · **PDF:** https://arxiv.org/pdf/2601.22474v1
+
+## 17799. Training-Free Representation Guidance for Diffusion Models with a Representation Alignment Projector
+
+- **ID:** `2601.22468` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Wenqiang Zu, Shenghao Xie, Bo Lei, et al.
+- **abs:** https://arxiv.org/abs/2601.22468 · **PDF:** https://arxiv.org/pdf/2601.22468v1
+
+## 17800. EvoEGF-Mol: Evolving Exponential Geodesic Flow for Structure-based Drug Design
+
+- **ID:** `2601.22466` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yaowei Jin, Junjie Wang, Cheng Cao, et al.
+- **abs:** https://arxiv.org/abs/2601.22466 · **PDF:** https://arxiv.org/pdf/2601.22466v1
+
+## 17801. Learning Robust Reasoning through Guided Adversarial Self-Play
+
+- **ID:** `2602.00173` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Shuozhe Li, Vaishnav Tadiparthi, Kwonjoon Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.00173 · **PDF:** https://arxiv.org/pdf/2602.00173v1
+
+## 17802. Toward Non-Expert Customized Congestion Control
+
+- **ID:** `2601.22461` · **primary:** `cs.NI` · **Submitted:** 2026-01-30
+- **Authors:** Mingrui Zhang, Hamid Bagheri, Lisong Xu
+- **abs:** https://arxiv.org/abs/2601.22461 · **PDF:** https://arxiv.org/pdf/2601.22461v1
+
+## 17803. AI Decodes Historical Chinese Archives to Reveal Lost Climate History
+
+- **ID:** `2601.22458` · **primary:** `physics.ao-ph` · **Submitted:** 2026-01-30
+- **Authors:** Sida He, Lingxi Xie, Xiaopeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22458 · **PDF:** https://arxiv.org/pdf/2601.22458v1
+
+## 17804. Machine Unlearning in Low-Dimensional Feature Subspace
+
+- **ID:** `2601.22456` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Kun Fang, Qinghua Tao, Junxu Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.22456 · **PDF:** https://arxiv.org/pdf/2601.22456v1
+
+## 17805. Temporal Graph Pattern Machine
+
+- **ID:** `2601.22454` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yijun Ma, Zehong Wang, Weixiang Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.22454 · **PDF:** https://arxiv.org/pdf/2601.22454v2
+
+## 17806. Does My Chatbot Have an Agenda? Understanding Human and AI Agency in Human-Human-like Chatbot Interaction
+
+- **ID:** `2601.22452` · **primary:** `cs.HC` · **Submitted:** 2026-01-30
+- **Authors:** Bhada Yun, Evgenia Taranova, April Yi Wang
+- **abs:** https://arxiv.org/abs/2601.22452 · **PDF:** https://arxiv.org/pdf/2601.22452v2
+
+## 17807. Countering the Over-Reliance Trap: Mitigating Object Hallucination for LVLMs via a Self-Validation Framework
+
+- **ID:** `2601.22451` · **primary:** `cs.CV` · **Submitted:** 2026-01-30
+- **Authors:** Shiyu Liu, Xinyi Wen, Zhibin Lan, et al.
+- **abs:** https://arxiv.org/abs/2601.22451 · **PDF:** https://arxiv.org/pdf/2601.22451v1
+
+## 17808. Tuning the Implicit Regularizer of Masked Diffusion Language Models: Enhancing Generalization via Insights from $k$-Parity
+
+- **ID:** `2601.22450` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Jianhao Huang, Baharan Mirzasoleiman
+- **abs:** https://arxiv.org/abs/2601.22450 · **PDF:** https://arxiv.org/pdf/2601.22450v1
+
+## 17809. Controllable Information Production
+
+- **ID:** `2601.22449` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Tristan Shah, Stas Tiomkin
+- **abs:** https://arxiv.org/abs/2601.22449 · **PDF:** https://arxiv.org/pdf/2601.22449v1
+
+## 17810. HeaPA: Difficulty-Aware Heap Sampling and On-Policy Query Augmentation for LLM Reinforcement Learning
+
+- **ID:** `2601.22448` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Weiqi Wang, Xin Liu, Binxuan Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.22448 · **PDF:** https://arxiv.org/pdf/2601.22448v1
+
+## 17811. Beyond Activation Patterns: A Weight-Based Out-of-Context Explanation of Sparse Autoencoder Features
+
+- **ID:** `2601.22447` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Yiting Liu, Zhi-Hong Deng
+- **abs:** https://arxiv.org/abs/2601.22447 · **PDF:** https://arxiv.org/pdf/2601.22447v1
+
+## 17812. Anytime Safe PAC Efficient Reasoning
+
+- **ID:** `2601.22446` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Chengyao Yu, Hao Zeng, Youxin Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.22446 · **PDF:** https://arxiv.org/pdf/2601.22446v1
+
+## 17813. Neuron Block Dynamics for XOR Classification with Zero-Margin
+
+- **ID:** `2602.00172` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Guillaume Braun, Masaaki Imaizumi
+- **abs:** https://arxiv.org/abs/2602.00172 · **PDF:** https://arxiv.org/pdf/2602.00172v1
+
+## 17814. Automating Forecasting Question Generation and Resolution for AI Evaluation
+
+- **ID:** `2601.22444` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Nikos I. Bosse, Peter Mühlbacher, Jack Wildman, et al.
+- **abs:** https://arxiv.org/abs/2601.22444 · **PDF:** https://arxiv.org/pdf/2601.22444v2
+
+## 17815. Recovering Whole-Brain Causal Connectivity under Indirect Observation with Applications to Human EEG and fMRI
+
+- **ID:** `2602.09034` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-30
+- **Authors:** Sangyoon Bae, Miruna Oprescu, David Keetae Park, et al.
+- **abs:** https://arxiv.org/abs/2602.09034 · **PDF:** https://arxiv.org/pdf/2602.09034v1
+
+## 17816. Weak Diffusion Priors Can Still Achieve Strong Inverse-Problem Performance
+
+- **ID:** `2601.22443` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Jing Jia, Wei Yuan, Sifan Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.22443 · **PDF:** https://arxiv.org/pdf/2601.22443v1
+
+## 17817. AsyncMesh: Fully Asynchronous Optimization for Data and Pipeline Parallelism
+
+- **ID:** `2601.22442` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Thalaiyasingam Ajanthan, Sameera Ramasinghe, Gil Avraham, et al.
+- **abs:** https://arxiv.org/abs/2601.22442 · **PDF:** https://arxiv.org/pdf/2601.22442v1
+
+## 17818. Simulation-based Bayesian inference with ameliorative learned summary statistics -- Part I
+
+- **ID:** `2601.22441` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Getachew K. Befekadu
+- **abs:** https://arxiv.org/abs/2601.22441 · **PDF:** https://arxiv.org/pdf/2601.22441v1
+
+## 17819. AI and My Values: User Perceptions of LLMs' Ability to Extract, Embody, and Explain Human Values from Casual Conversations
+
+- **ID:** `2601.22440` · **primary:** `cs.HC` · **Submitted:** 2026-01-30
+- **Authors:** Bhada Yun, Renn Su, April Yi Wang
+- **abs:** https://arxiv.org/abs/2601.22440 · **PDF:** https://arxiv.org/pdf/2601.22440v2
+
+## 17820. Stop Jostling: Adaptive Negative Sampling Reduces the Marginalization of Low-Resource Language Tokens by Cross-Entropy Loss
+
+- **ID:** `2601.22439` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Galim Turumtaev
+- **abs:** https://arxiv.org/abs/2601.22439 · **PDF:** https://arxiv.org/pdf/2601.22439v1
+
+## 17821. Towards Resiliency in Large Language Model Serving with KevlarFlow
+
+- **ID:** `2601.22438` · **primary:** `cs.DC` · **Submitted:** 2026-01-30
+- **Authors:** Shangshu Qian, Kipling Liu, P. C. Sruthi, et al.
+- **abs:** https://arxiv.org/abs/2601.22438 · **PDF:** https://arxiv.org/pdf/2601.22438v1
+
+## 17822. Retrieval Augmented (Knowledge Graph), and Large Language Model-Driven Design Structure Matrix (DSM) Generation of Cyber-Physical Systems
+
+- **ID:** `2602.16715` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** H. Sinan Bank, Daniel R. Herber
+- **abs:** https://arxiv.org/abs/2602.16715 · **PDF:** https://arxiv.org/pdf/2602.16715v1
+
+## 17823. Large Language Model Agents Are Not Always Faithful Self-Evolvers
+
+- **ID:** `2601.22436` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Weixiang Zhao, Yingshuo Wang, Yichen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22436 · **PDF:** https://arxiv.org/pdf/2601.22436v2
+
+## 17824. Rethinking Anonymity Claims in Synthetic Data Generation: A Model-Centric Privacy Attack Perspective
+
+- **ID:** `2601.22434` · **primary:** `cs.CR` · **Submitted:** 2026-01-30
+- **Authors:** Georgi Ganev, Emiliano De Cristofaro
+- **abs:** https://arxiv.org/abs/2601.22434 · **PDF:** https://arxiv.org/pdf/2601.22434v1
+
+## 17825. When LLM meets Fuzzy-TOPSIS for Personnel Selection through Automated Profile Analysis
+
+- **ID:** `2601.22433` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Shahria Hoque, Ahmed Akib Jawad Karim, Md. Golam Rabiul Alam, et al.
+- **abs:** https://arxiv.org/abs/2601.22433 · **PDF:** https://arxiv.org/pdf/2601.22433v1
+
+## 17826. ReNCE: Learning to Reason by Noise Contrastive Estimation
+
+- **ID:** `2601.22432` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Wenzheng Zhang, Karl Stratos
+- **abs:** https://arxiv.org/abs/2601.22432 · **PDF:** https://arxiv.org/pdf/2601.22432v1
+
+## 17827. CoDCL: Counterfactual Data Augmentation Contrastive Learning for Continuous-Time Dynamic Network Link Prediction
+
+- **ID:** `2601.22427` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Hantong Feng, Yonggang Wu, Duxin Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.22427 · **PDF:** https://arxiv.org/pdf/2601.22427v1
+
+## 17828. Uncertainty-Aware Multimodal Learning via Conformal Shapley Intervals
+
+- **ID:** `2602.00171` · **primary:** `stat.ML` · **Submitted:** 2026-01-30
+- **Authors:** Mathew Chandy, Michael Johnson, Judong Shen, et al.
+- **abs:** https://arxiv.org/abs/2602.00171 · **PDF:** https://arxiv.org/pdf/2602.00171v1
+
+## 17829. The Blessing of Dimensionality in LLM Fine-tuning: A Variance-Curvature Perspective
+
+- **ID:** `2602.00170` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Qiyao Liang, Jinyeop Song, Yizhou Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.00170 · **PDF:** https://arxiv.org/pdf/2602.00170v1
+
+## 17830. MetaLead: A Comprehensive Human-Curated Leaderboard Dataset for Transparent Reporting of Machine Learning Experiments
+
+- **ID:** `2601.22420` · **primary:** `cs.LG` · **Submitted:** 2026-01-30
+- **Authors:** Roelien C. Timmer, Necva Bölücü, Stephen Wan
+- **abs:** https://arxiv.org/abs/2601.22420 · **PDF:** https://arxiv.org/pdf/2601.22420v1
+
+## 17831. Dynamic Welfare-Maximizing Pooled Testing
+
+- **ID:** `2601.22419` · **primary:** `cs.GT` · **Submitted:** 2026-01-30
+- **Authors:** Nicholas Lopez, Francisco Marmolejo-Cossío, Jose Roberto Tello Ayala, et al.
+- **abs:** https://arxiv.org/abs/2601.22419 · **PDF:** https://arxiv.org/pdf/2601.22419v1
+
+## 17832. AI-Enabled Waste Classification as a Data-Driven Decision Support Tool for Circular Economy and Urban Sustainability
+
+- **ID:** `2601.22418` · **primary:** `cs.AI` · **Submitted:** 2026-01-30
+- **Authors:** Julius Sechang Mboli, Omolara Aderonke Ogungbemi
+- **abs:** https://arxiv.org/abs/2601.22418 · **PDF:** https://arxiv.org/pdf/2601.22418v1
+
+## 17833. Prompt Optimization Via Diffusion Language Models
+
+- **ID:** `2602.18449` · **primary:** `cs.CL` · **Submitted:** 2026-01-30
+- **Authors:** Shiyu Wang, Haolin Chen, Liangwei Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.18449 · **PDF:** https://arxiv.org/pdf/2602.18449v1
+
+## 17834. MM-OpenFGL: A Comprehensive Benchmark for Multimodal Federated Graph Learning
+
+- **ID:** `2601.22416` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Xunkai Li, Yuming Ai, Yinlin Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.22416 · **PDF:** https://arxiv.org/pdf/2601.22416v1
+
+## 17835. Towards Agentic Intelligence for Materials Science
+
+- **ID:** `2602.00169` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-29
+- **Authors:** Huan Zhang, Yizhan Li, Wenhao Huang, et al.
+- **abs:** https://arxiv.org/abs/2602.00169 · **PDF:** https://arxiv.org/pdf/2602.00169v2
+
+## 17836. Word-Centered Semantic Graphs for Interpretable Diachronic Sense Tracking
+
+- **ID:** `2601.22410` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Imene Kolli, Kai-Robin Lange, Jonas Rieger, et al.
+- **abs:** https://arxiv.org/abs/2601.22410 · **PDF:** https://arxiv.org/pdf/2601.22410v1
+
+## 17837. Optimization, Generalization and Differential Privacy Bounds for Gradient Descent on Kolmogorov-Arnold Networks
+
+- **ID:** `2601.22409` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Puyu Wang, Junyu Zhou, Philipp Liznerski, et al.
+- **abs:** https://arxiv.org/abs/2601.22409 · **PDF:** https://arxiv.org/pdf/2601.22409v2
+
+## 17838. Minimal-Action Discrete Schrödinger Bridge Matching for Peptide Sequence Design
+
+- **ID:** `2601.22408` · **primary:** `q-bio.BM` · **Submitted:** 2026-01-29
+- **Authors:** Shrey Goel, Pranam Chatterjee
+- **abs:** https://arxiv.org/abs/2601.22408 · **PDF:** https://arxiv.org/pdf/2601.22408v1
+
+## 17839. Joint Continual Learning of Local Language Models and Cloud Offloading Decisions with Budget Constraints
+
+- **ID:** `2602.00166` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Evan Chen, Wenzhi Fang, Shiqiang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00166 · **PDF:** https://arxiv.org/pdf/2602.00166v2
+
+## 17840. Bifocal Attention: Harmonizing Geometric and Spectral Positional Embeddings for Algorithmic Generalization
+
+- **ID:** `2601.22402` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Kanishk Awadhiya
+- **abs:** https://arxiv.org/abs/2601.22402 · **PDF:** https://arxiv.org/pdf/2601.22402v1
+
+## 17841. Semi-Autonomous Mathematics Discovery with Gemini: A Case Study on the Erdős Problems
+
+- **ID:** `2601.22401` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Tony Feng, Trieu Trinh, Garrett Bingham, et al.
+- **abs:** https://arxiv.org/abs/2601.22401 · **PDF:** https://arxiv.org/pdf/2601.22401v3
+
+## 17842. Spectral Filtering for Learning Quantum Dynamics
+
+- **ID:** `2601.22400` · **primary:** `quant-ph` · **Submitted:** 2026-01-29
+- **Authors:** Elad Hazan, Annie Marsden
+- **abs:** https://arxiv.org/abs/2601.22400 · **PDF:** https://arxiv.org/pdf/2601.22400v1
+
+## 17843. Score-based Integrated Gradient for Root Cause Explanations of Outliers
+
+- **ID:** `2601.22399` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Phuoc Nguyen, Truyen Tran, Sunil Gupta, et al.
+- **abs:** https://arxiv.org/abs/2601.22399 · **PDF:** https://arxiv.org/pdf/2601.22399v1
+
+## 17844. Jailbreaks on Vision Language Model via Multimodal Reasoning
+
+- **ID:** `2601.22398` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Aarush Noheria, Yuguang Yao
+- **abs:** https://arxiv.org/abs/2601.22398 · **PDF:** https://arxiv.org/pdf/2601.22398v1
+
+## 17845. SAIR: Cost-Efficient Multi-Stage ML Pipeline Autoscaling via In-Context Reinforcement Learning
+
+- **ID:** `2601.22397` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jianchang Su, Yifan Zhang, Shengkai Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.22397 · **PDF:** https://arxiv.org/pdf/2601.22397v1
+
+## 17846. Culturally Grounded Personas in Large Language Models: Characterization and Alignment with Socio-Psychological Value Frameworks
+
+- **ID:** `2601.22396` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Candida M. Greco, Lucio La Cava, Andrea Tagarelli
+- **abs:** https://arxiv.org/abs/2601.22396 · **PDF:** https://arxiv.org/pdf/2601.22396v1
+
+## 17847. Specialists or Generalists? Multi-Agent and Single-Agent LLMs for Essay Grading
+
+- **ID:** `2601.22386` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Jamiu Adekunle Idowu, Ahmed Almasoud
+- **abs:** https://arxiv.org/abs/2601.22386 · **PDF:** https://arxiv.org/pdf/2601.22386v1
+
+## 17848. SP^2DPO: An LLM-assisted Semantic Per-Pair DPO Generalization
+
+- **ID:** `2601.22385` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Chaoyue He, Xin Zhou, Di Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.22385 · **PDF:** https://arxiv.org/pdf/2601.22385v1
+
+## 17849. Graph is a Substrate Across Data Modalities
+
+- **ID:** `2601.22384` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Ziming Li, Xiaoming Wu, Zehong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.22384 · **PDF:** https://arxiv.org/pdf/2601.22384v1
+
+## 17850. Purely Agentic Black-Box Optimization for Biological Design
+
+- **ID:** `2601.22382` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Natalie Maus, Yimeng Zeng, Haydn Thomas Jones, et al.
+- **abs:** https://arxiv.org/abs/2601.22382 · **PDF:** https://arxiv.org/pdf/2601.22382v1
+
+## 17851. SPLA: Block Sparse Plus Linear Attention for Long Context Modeling
+
+- **ID:** `2601.22379` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Bailin Wang, Dan Friedman, Tao Lei, et al.
+- **abs:** https://arxiv.org/abs/2601.22379 · **PDF:** https://arxiv.org/pdf/2601.22379v1
+
+## 17852. It's all In the (Exponential) Family: An Equivalence between Maximum Likelihood Estimation and Control Variates for Sketching Algorithms
+
+- **ID:** `2601.22378` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** Keegan Kang, Kerong Wang, Ding Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22378 · **PDF:** https://arxiv.org/pdf/2601.22378v2
+
+## 17853. Stability-Aware Prompt Optimization for Clinical Data Abstraction
+
+- **ID:** `2601.22373` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Arinbjörn Kolbeinsson, Daniel Timbie, Sajjan Narsinghani, et al.
+- **abs:** https://arxiv.org/abs/2601.22373 · **PDF:** https://arxiv.org/pdf/2601.22373v1
+
+## 17854. FIRE: Multi-fidelity Regression with Distribution-conditioned In-context Learning using Tabular Foundation Models
+
+- **ID:** `2601.22371` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Rosen Ting-Ying Yu, Nicholas Sung, Faez Ahmed
+- **abs:** https://arxiv.org/abs/2601.22371 · **PDF:** https://arxiv.org/pdf/2601.22371v1
+
+## 17855. Learning Provably Correct Distributed Protocols Without Human Knowledge
+
+- **ID:** `2601.22369` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Yujie Hui, Xiaoyi Lu, Andrew Perrault, et al.
+- **abs:** https://arxiv.org/abs/2601.22369 · **PDF:** https://arxiv.org/pdf/2601.22369v1
+
+## 17856. Amortized Simulation-Based Inference in Generalized Bayes via Neural Posterior Estimation
+
+- **ID:** `2601.22367` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** Shiyi Sun, Geoff K. Nicholls, Jeong Eun Lee
+- **abs:** https://arxiv.org/abs/2601.22367 · **PDF:** https://arxiv.org/pdf/2601.22367v1
+
+## 17857. Towards Solving the Gilbert-Pollak Conjecture via Large Language Models
+
+- **ID:** `2601.22365` · **primary:** `cs.DM` · **Submitted:** 2026-01-29
+- **Authors:** Yisi Ke, Tianyu Huang, Yankai Shu, et al.
+- **abs:** https://arxiv.org/abs/2601.22365 · **PDF:** https://arxiv.org/pdf/2601.22365v1
+
+## 17858. Context Structure Reshapes the Representational Geometry of Language Models
+
+- **ID:** `2601.22364` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Eghbal A. Hosseini, Yuxuan Li, Yasaman Bahri, et al.
+- **abs:** https://arxiv.org/abs/2601.22364 · **PDF:** https://arxiv.org/pdf/2601.22364v1
+
+## 17859. Understanding Efficiency: Quantization, Batching, and Serving Strategies in LLM Energy Use
+
+- **ID:** `2601.22362` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Julien Delavande, Regis Pierrard, Sasha Luccioni
+- **abs:** https://arxiv.org/abs/2601.22362 · **PDF:** https://arxiv.org/pdf/2601.22362v1
+
+## 17860. MERMAID: Memory-Enhanced Retrieval and Reasoning with Multi-Agent Iterative Knowledge Grounding for Veracity Assessment
+
+- **ID:** `2601.22361` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Yupeng Cao, Chengyang He, Yangyang Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.22361 · **PDF:** https://arxiv.org/pdf/2601.22361v1
+
+## 17861. The Unseen Threat: Residual Knowledge in Machine Unlearning under Perturbed Samples
+
+- **ID:** `2601.22359` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Hsiang Hsu, Pradeep Niroula, Zichang He, et al.
+- **abs:** https://arxiv.org/abs/2601.22359 · **PDF:** https://arxiv.org/pdf/2601.22359v1
+
+## 17862. Benford's Law as a Distributional Prior for Post-Training Quantization of Large Language Models
+
+- **ID:** `2602.00165` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Arthur Negrão, Pedro Silva, Vander L. S. Freitas, et al.
+- **abs:** https://arxiv.org/abs/2602.00165 · **PDF:** https://arxiv.org/pdf/2602.00165v1
+
+## 17863. Why Are AI Agent Involved Pull Requests (Fix-Related) Remain Unmerged? An Empirical Study
+
+- **ID:** `2602.00164` · **primary:** `cs.SE` · **Submitted:** 2026-01-29
+- **Authors:** Khairul Alam, Saikat Mondal, Banani Roy
+- **abs:** https://arxiv.org/abs/2602.00164 · **PDF:** https://arxiv.org/pdf/2602.00164v1
+
+## 17864. Small Talk, Big Impact: The Energy Cost of Thanking AI
+
+- **ID:** `2601.22357` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Julien Delavande, Regis Pierrard, Sasha Luccioni
+- **abs:** https://arxiv.org/abs/2601.22357 · **PDF:** https://arxiv.org/pdf/2601.22357v1
+
+## 17865. PoSafeNet: Safe Learning with Poset-Structured Neural Nets
+
+- **ID:** `2601.22356` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Kiwan Wong, Wei Xiao, Daniela Rus
+- **abs:** https://arxiv.org/abs/2601.22356 · **PDF:** https://arxiv.org/pdf/2601.22356v1
+
+## 17866. Relative Wasserstein Angle and the Problem of the $W_2$-Nearest Gaussian Distribution
+
+- **ID:** `2601.22355` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Binshuai Wang, Peng Wei
+- **abs:** https://arxiv.org/abs/2601.22355 · **PDF:** https://arxiv.org/pdf/2601.22355v1
+
+## 17867. Recoverability Has a Law: The ERR Measure for Tool-Augmented Agents
+
+- **ID:** `2601.22352` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Sri Vatsa Vuddanti, Satwik Kumar Chittiprolu
+- **abs:** https://arxiv.org/abs/2601.22352 · **PDF:** https://arxiv.org/pdf/2601.22352v1
+
+## 17868. Learning Policy Representations for Steerable Behavior Synthesis
+
+- **ID:** `2601.22350` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Beiming Li, Sergio Rozada, Alejandro Ribeiro
+- **abs:** https://arxiv.org/abs/2601.22350 · **PDF:** https://arxiv.org/pdf/2601.22350v1
+
+## 17869. MixQuant: Pushing the Limits of Block Rotations in Post-Training Quantization
+
+- **ID:** `2601.22347` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Sai Sanjeet, Ian Colbert, Pablo Monteagudo-Lago, et al.
+- **abs:** https://arxiv.org/abs/2601.22347 · **PDF:** https://arxiv.org/pdf/2601.22347v1
+
+## 17870. Failing to Explore: Language Models on Interactive Tasks
+
+- **ID:** `2601.22345` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Mahdi JafariRaviz, Keivan Rezaei, Arshia Soltani Moakhar, et al.
+- **abs:** https://arxiv.org/abs/2601.22345 · **PDF:** https://arxiv.org/pdf/2601.22345v1
+
+## 17871. Quantum-Inspired Reinforcement Learning for Secure and Sustainable AIoT-Driven Supply Chain Systems
+
+- **ID:** `2601.22339` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Muhammad Bilal Akram Dastagir, Omer Tariq, Shahid Mumtaz, et al.
+- **abs:** https://arxiv.org/abs/2601.22339 · **PDF:** https://arxiv.org/pdf/2601.22339v1
+
+## 17872. From Retrieving Information to Reasoning with AI: Exploring Different Interaction Modalities to Support Human-AI Coordination in Clinical Decision-Making
+
+- **ID:** `2601.22338` · **primary:** `cs.HC` · **Submitted:** 2026-01-29
+- **Authors:** Behnam Rahdari, Sameer Shaikh, Jonathan H Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.22338 · **PDF:** https://arxiv.org/pdf/2601.22338v1
+
+## 17873. Dependence-Aware Label Aggregation for LLM-as-a-Judge via Ising Models
+
+- **ID:** `2601.22336` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** Krishnakumar Balasubramanian, Aleksandr Podkopaev, Shiva Prasad Kasiviswanathan
+- **abs:** https://arxiv.org/abs/2601.22336 · **PDF:** https://arxiv.org/pdf/2601.22336v1
+
+## 17874. Knowledge Gradient for Preference Learning
+
+- **ID:** `2601.22335` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Kaiwen Wu, Jacob R. Gardner
+- **abs:** https://arxiv.org/abs/2601.22335 · **PDF:** https://arxiv.org/pdf/2601.22335v1
+
+## 17875. DP-$λ$CGD: Efficient Noise Correlation for Differentially Private Model Training
+
+- **ID:** `2601.22334` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Nikita P. Kalinin, Ryan McKenna, Rasmus Pagh, et al.
+- **abs:** https://arxiv.org/abs/2601.22334 · **PDF:** https://arxiv.org/pdf/2601.22334v1
+
+## 17876. Scalable Batch Correction for Cell Painting via Batch-Dependent Kernels and Adaptive Sampling
+
+- **ID:** `2601.22331` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Aditya Narayan Ravi, Snehal Vadvalkar, Abhishek Pandey, et al.
+- **abs:** https://arxiv.org/abs/2601.22331 · **PDF:** https://arxiv.org/pdf/2601.22331v1
+
+## 17877. Sparks of Rationality: Do Reasoning LLMs Align with Human Judgment and Choice?
+
+- **ID:** `2601.22329` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Ala N. Tak, Amin Banayeeanzade, Anahita Bolourani, et al.
+- **abs:** https://arxiv.org/abs/2601.22329 · **PDF:** https://arxiv.org/pdf/2601.22329v1
+
+## 17878. Knowledge-Informed Kernel State Reconstruction for Interpretable Dynamical System Discovery
+
+- **ID:** `2601.22328` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Luca Muscarnera, Silas Ruhrberg Estévez, Samuel Holt, et al.
+- **abs:** https://arxiv.org/abs/2601.22328 · **PDF:** https://arxiv.org/pdf/2601.22328v1
+
+## 17879. Molecular Representations in Implicit Functional Space via Hyper-Networks
+
+- **ID:** `2601.22327` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Zehong Wang, Xiaolong Han, Qi Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.22327 · **PDF:** https://arxiv.org/pdf/2601.22327v1
+
+## 17880. Label-Efficient Monitoring of Classification Models via Stratified Importance Sampling
+
+- **ID:** `2601.22326` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Lupo Marsigli, Angel Lopez de Haro
+- **abs:** https://arxiv.org/abs/2601.22326 · **PDF:** https://arxiv.org/pdf/2601.22326v1
+
+## 17881. AgentScore: Autoformulation of Deployable Clinical Scoring Systems
+
+- **ID:** `2601.22324` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Silas Ruhrberg Estévez, Christopher Chiu, Mihaela van der Schaar
+- **abs:** https://arxiv.org/abs/2601.22324 · **PDF:** https://arxiv.org/pdf/2601.22324v1
+
+## 17882. Models Under SCOPE: Scalable and Controllable Routing via Pre-hoc Reasoning
+
+- **ID:** `2601.22323` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Qi Cao, Shuhao Zhang, Ruizhe Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.22323 · **PDF:** https://arxiv.org/pdf/2601.22323v2
+
+## 17883. Spatially-Adaptive Conformal Graph Transformer for Indoor Localization in Wi-Fi Driven Networks
+
+- **ID:** `2601.22322` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Ayesh Abu Lehyeh, Anastassia Gharib, Safwan Wshah
+- **abs:** https://arxiv.org/abs/2601.22322 · **PDF:** https://arxiv.org/pdf/2601.22322v1
+
+## 17884. Matrix Factorization for Practical Continual Mean Estimation Under User-Level Differential Privacy
+
+- **ID:** `2601.22320` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Nikita P. Kalinin, Ali Najar, Valentin Roth, et al.
+- **abs:** https://arxiv.org/abs/2601.22320 · **PDF:** https://arxiv.org/pdf/2601.22320v1
+
+## 17885. Federate the Router: Learning Language Model Routers with Sparse and Decentralized Evaluations
+
+- **ID:** `2601.22318` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Baris Askin, Shivam Patel, Anupam Nayak, et al.
+- **abs:** https://arxiv.org/abs/2601.22318 · **PDF:** https://arxiv.org/pdf/2601.22318v1
+
+## 17886. FlowSymm: Physics Aware, Symmetry Preserving Graph Attention for Network Flow Completion
+
+- **ID:** `2601.22317` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Ege Demirci, Francesco Bullo, Ananthram Swami, et al.
+- **abs:** https://arxiv.org/abs/2601.22317 · **PDF:** https://arxiv.org/pdf/2601.22317v2
+
+## 17887. Gaussian Process Bandit Optimization with Machine Learning Predictions and Application to Hypothesis Generation
+
+- **ID:** `2601.22315` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Xin Jennifer Chen, Yunjin Tong
+- **abs:** https://arxiv.org/abs/2601.22315 · **PDF:** https://arxiv.org/pdf/2601.22315v1
+
+## 17888. Hair-Trigger Alignment: Black-Box Evaluation Cannot Guarantee Post-Update Alignment
+
+- **ID:** `2601.22313` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yavuz Bakman, Duygu Nur Yaldiz, Salman Avestimehr, et al.
+- **abs:** https://arxiv.org/abs/2601.22313 · **PDF:** https://arxiv.org/pdf/2601.22313v1
+
+## 17889. SCALAR: Quantifying Structural Hallucination, Consistency, and Reasoning Gaps in Materials Foundation Models
+
+- **ID:** `2601.22312` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Can Polat, Erchin Serpedin, Mustafa Kurban, et al.
+- **abs:** https://arxiv.org/abs/2601.22312 · **PDF:** https://arxiv.org/pdf/2601.22312v1
+
+## 17890. Why Reasoning Fails to Plan: A Planning-Centric Analysis of Long-Horizon Decision Making in LLM Agents
+
+- **ID:** `2601.22311` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Zehong Wang, Fang Wu, Hongru Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.22311 · **PDF:** https://arxiv.org/pdf/2601.22311v1
+
+## 17891. Stealthy Poisoning Attacks Bypass Defenses in Regression Settings
+
+- **ID:** `2601.22308` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Javier Carnerero-Cano, Luis Muñoz-González, Phillippa Spencer, et al.
+- **abs:** https://arxiv.org/abs/2601.22308 · **PDF:** https://arxiv.org/pdf/2601.22308v2
+
+## 17892. Exact closed-form Gaussian moments of residual layers
+
+- **ID:** `2601.22307` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Simon Kuang, Xinfan Lin
+- **abs:** https://arxiv.org/abs/2601.22307 · **PDF:** https://arxiv.org/pdf/2601.22307v1
+
+## 17893. Sylber 2.0: A Universal Syllable Embedding
+
+- **ID:** `2601.22306` · **primary:** `eess.AS` · **Submitted:** 2026-01-29
+- **Authors:** Cheol Jun Cho, Nicholas Lee, Alan W Black, et al.
+- **abs:** https://arxiv.org/abs/2601.22306 · **PDF:** https://arxiv.org/pdf/2601.22306v1
+
+## 17894. BayesFlow: A Probability Inference Framework for Meta-Agent Assisted Workflow Generation
+
+- **ID:** `2601.22305` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Bo Yuan, Yun Zhou, Zhichao Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.22305 · **PDF:** https://arxiv.org/pdf/2601.22305v1
+
+## 17895. ZK-HybridFL: Zero-Knowledge Proof-Enhanced Hybrid Ledger for Federated Learning
+
+- **ID:** `2601.22302` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Amirhossein Taherpour, Xiaodong Wang
+- **abs:** https://arxiv.org/abs/2601.22302 · **PDF:** https://arxiv.org/pdf/2601.22302v1
+
+## 17896. Online unsupervised Hebbian learning in deep photonic neuromorphic networks
+
+- **ID:** `2601.22300` · **primary:** `physics.optics` · **Submitted:** 2026-01-29
+- **Authors:** Xi Li, Disha Biswas, Peng Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.22300 · **PDF:** https://arxiv.org/pdf/2601.22300v2
+
+## 17897. Conformal Prediction for Generative Models via Adaptive Cluster-Based Density Estimation
+
+- **ID:** `2601.22298` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Qidong Yang, Qianyu Julie Zhu, Jonathan Giezendanner, et al.
+- **abs:** https://arxiv.org/abs/2601.22298 · **PDF:** https://arxiv.org/pdf/2601.22298v1
+
+## 17898. Prepare Reasoning Language Models for Multi-Agent Debate with Self-Debate Reinforcement Learning
+
+- **ID:** `2601.22297` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Chenxi Liu, Yanshuo Chen, Ruibo Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.22297 · **PDF:** https://arxiv.org/pdf/2601.22297v1
+
+## 17899. ParalESN: Enabling parallel information processing in Reservoir Computing
+
+- **ID:** `2601.22296` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Matteo Pinna, Giacomo Lagomarsini, Andrea Ceni, et al.
+- **abs:** https://arxiv.org/abs/2601.22296 · **PDF:** https://arxiv.org/pdf/2601.22296v1
+
+## 17900. Learning Reward Functions for Cooperative Resilience in Multi-Agent Systems
+
+- **ID:** `2601.22292` · **primary:** `cs.MA` · **Submitted:** 2026-01-29
+- **Authors:** Manuela Chacon-Chamorro, Luis Felipe Giraldo, Nicanor Quijano
+- **abs:** https://arxiv.org/abs/2601.22292 · **PDF:** https://arxiv.org/pdf/2601.22292v1
+
+## 17901. The Six Sigma Agent: Achieving Enterprise-Grade Reliability in LLM Systems Through Consensus-Driven Decomposed Execution
+
+- **ID:** `2601.22290` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Khush Patel, Siva Surendira, Jithin George, et al.
+- **abs:** https://arxiv.org/abs/2601.22290 · **PDF:** https://arxiv.org/pdf/2601.22290v1
+
+## 17902. PersonaCite: VoC-Grounded Interviewable Agentic Synthetic AI Personas for Verifiable User and Design Research
+
+- **ID:** `2601.22288` · **primary:** `cs.HC` · **Submitted:** 2026-01-29
+- **Authors:** Mario Truss
+- **abs:** https://arxiv.org/abs/2601.22288 · **PDF:** https://arxiv.org/pdf/2601.22288v1
+
+## 17903. Demystifying Mergeability: Interpretable Properties to Predict Model Merging Success
+
+- **ID:** `2601.22285` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Luca Zhou, Bo Zhao, Rose Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.22285 · **PDF:** https://arxiv.org/pdf/2601.22285v3
+
+## 17904. Riemannian Lyapunov Optimizer: A Unified Framework for Optimization
+
+- **ID:** `2601.22284` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yixuan Wang, Omkar Sudhir Patil, Warren E. Dixon
+- **abs:** https://arxiv.org/abs/2601.22284 · **PDF:** https://arxiv.org/pdf/2601.22284v1
+
+## 17905. SurrogateSHAP: Training-Free Contributor Attribution for Text-to-Image (T2I) Models
+
+- **ID:** `2601.22276` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Mingyu Lu, Soham Gadgil, Chris Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.22276 · **PDF:** https://arxiv.org/pdf/2601.22276v1
+
+## 17906. VMonarch: Efficient Video Diffusion Transformers with Structured Attention
+
+- **ID:** `2601.22275` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Cheng Liang, Haoxian Chen, Liang Hou, et al.
+- **abs:** https://arxiv.org/abs/2601.22275 · **PDF:** https://arxiv.org/pdf/2601.22275v1
+
+## 17907. Block removal for large language models through constrained binary optimization
+
+- **ID:** `2602.00161` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** David Jansen, Roman Rausch, David Montero, et al.
+- **abs:** https://arxiv.org/abs/2602.00161 · **PDF:** https://arxiv.org/pdf/2602.00161v1
+
+## 17908. Task-Uniform Convergence and Backward Transfer in Federated Domain-Incremental Learning with Partial Participation
+
+- **ID:** `2601.22274` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Longtao Xu, Jian Li
+- **abs:** https://arxiv.org/abs/2601.22274 · **PDF:** https://arxiv.org/pdf/2601.22274v1
+
+## 17909. JAF: Judge Agent Forest
+
+- **ID:** `2601.22269` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Sahil Garg, Brad Cheezum, Sridhar Dutta, et al.
+- **abs:** https://arxiv.org/abs/2601.22269 · **PDF:** https://arxiv.org/pdf/2601.22269v1
+
+## 17910. Sheaf Neural Networks and biomedical applications
+
+- **ID:** `2602.00159` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Aneeqa Mehrab, Jan Willem Van Looy, Pietro Demurtas, et al.
+- **abs:** https://arxiv.org/abs/2602.00159 · **PDF:** https://arxiv.org/pdf/2602.00159v2
+
+## 17911. Privacy-Preserving Sensor-Based Human Activity Recognition for Low-Resource Healthcare Using Classical Machine Learning
+
+- **ID:** `2601.22265` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Ramakant Kumar, Pravin Kumar
+- **abs:** https://arxiv.org/abs/2601.22265 · **PDF:** https://arxiv.org/pdf/2601.22265v1
+
+## 17912. Predicting Intermittent Job Failure Categories for Diagnosis Using Few-Shot Fine-Tuned Language Models
+
+- **ID:** `2601.22264` · **primary:** `cs.SE` · **Submitted:** 2026-01-29
+- **Authors:** Henri Aïdasso, Francis Bordeleau, Ali Tizghadam
+- **abs:** https://arxiv.org/abs/2601.22264 · **PDF:** https://arxiv.org/pdf/2601.22264v1
+
+## 17913. Attention Guidance through Video Script: A Case Study of Object Focusing on 360° VR Video Tours
+
+- **ID:** `2603.16875` · **primary:** `cs.HC` · **Submitted:** 2026-01-29
+- **Authors:** Paulo Vitor Santana Silva, Arthur Ricardo Sousa Vitória, Diogo Fernandes Costa Silva, et al.
+- **abs:** https://arxiv.org/abs/2603.16875 · **PDF:** https://arxiv.org/pdf/2603.16875v1
+
+## 17914. Focus360: Guiding User Attention in Immersive Videos for VR
+
+- **ID:** `2603.28774` · **primary:** `cs.HC` · **Submitted:** 2026-01-29
+- **Authors:** Paulo Vitor S. Silva, Lucas L. Neves, Rafael A. Goiás, et al.
+- **abs:** https://arxiv.org/abs/2603.28774 · **PDF:** https://arxiv.org/pdf/2603.28774v1
+
+## 17915. Tabular Foundation Models Can Do Survival Analysis
+
+- **ID:** `2601.22259` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Da In Kim, Wei Siang Lai, Kelly W. Zhang
+- **abs:** https://arxiv.org/abs/2601.22259 · **PDF:** https://arxiv.org/pdf/2601.22259v1
+
+## 17916. Symmetry Breaking in Transformers for Efficient and Interpretable Training
+
+- **ID:** `2601.22257` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Eva Silverstein, Daniel Kunin, Vasudev Shyam
+- **abs:** https://arxiv.org/abs/2601.22257 · **PDF:** https://arxiv.org/pdf/2601.22257v1
+
+## 17917. AI Narrative Breakdown. A Critical Assessment of Power and Promise
+
+- **ID:** `2601.22255` · **primary:** `cs.CY` · **Submitted:** 2026-01-29
+- **Authors:** Rainer Rehak
+- **abs:** https://arxiv.org/abs/2601.22255 · **PDF:** https://arxiv.org/pdf/2601.22255v1
+
+## 17918. RAPTOR: Ridge-Adaptive Logistic Probes
+
+- **ID:** `2602.00158` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Ziqi Gao, Yaotian Zhu, Qingcheng Zeng, et al.
+- **abs:** https://arxiv.org/abs/2602.00158 · **PDF:** https://arxiv.org/pdf/2602.00158v2
+
+## 17919. ProDCARL: Reinforcement Learning-Aligned Diffusion Models for De Novo Antimicrobial Peptide Design
+
+- **ID:** `2602.00157` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-29
+- **Authors:** Fang Sheng, Mohammad Noaeen, Zahra Shakeri
+- **abs:** https://arxiv.org/abs/2602.00157 · **PDF:** https://arxiv.org/pdf/2602.00157v2
+
+## 17920. FunPRM: Function-as-Step Process Reward Model with Meta Reward Correction for Code Generation
+
+- **ID:** `2601.22249` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Ruiyi Zhang, Peijia Qin, Qi Cao, et al.
+- **abs:** https://arxiv.org/abs/2601.22249 · **PDF:** https://arxiv.org/pdf/2601.22249v1
+
+## 17921. MirrorMark: A Distortion-Free Multi-Bit Watermark for Large Language Models
+
+- **ID:** `2601.22246` · **primary:** `cs.CR` · **Submitted:** 2026-01-29
+- **Authors:** Ya Jiang, Massieh Kordi Boroujeny, Surender Suresh Kumar, et al.
+- **abs:** https://arxiv.org/abs/2601.22246 · **PDF:** https://arxiv.org/pdf/2601.22246v1
+
+## 17922. Is Hierarchical Quantization Essential for Optimal Reconstruction?
+
+- **ID:** `2601.22244` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Shirin Reyhanian, Laurenz Wiskott
+- **abs:** https://arxiv.org/abs/2601.22244 · **PDF:** https://arxiv.org/pdf/2601.22244v2
+
+## 17923. Aligning Microscopic Vehicle and Macroscopic Traffic Statistics: Reconstructing Driving Behavior from Partial Data
+
+- **ID:** `2601.22242` · **primary:** `cs.MA` · **Submitted:** 2026-01-29
+- **Authors:** Zhihao Zhang, Keith Redmill, Chengyang Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.22242 · **PDF:** https://arxiv.org/pdf/2601.22242v2
+
+## 17924. Investigating the Interplay of Parameterization and Optimizer in Gradient-Free Topology Optimization: A Cantilever Beam Case Study
+
+- **ID:** `2601.22241` · **primary:** `cs.NE` · **Submitted:** 2026-01-29
+- **Authors:** Jelle Westra, Iván Olarte Rodríguez, Niki van Stein, et al.
+- **abs:** https://arxiv.org/abs/2601.22241 · **PDF:** https://arxiv.org/pdf/2601.22241v2
+
+## 17925. A Systematic Literature Review on LLM Defenses Against Prompt Injection and Jailbreaking: Expanding NIST Taxonomy
+
+- **ID:** `2601.22240` · **primary:** `cs.CR` · **Submitted:** 2026-01-29
+- **Authors:** Pedro H. Barcha Correia, Ryan W. Achjian, Diego E. G. Caetano de Oliveira, et al.
+- **abs:** https://arxiv.org/abs/2601.22240 · **PDF:** https://arxiv.org/pdf/2601.22240v1
+
+## 17926. DAJ: Data-Reweighted LLM Judge for Test-Time Scaling in Code Generation
+
+- **ID:** `2601.22230` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Peijia Qin, Ruiyi Zhang, Qi Cao, et al.
+- **abs:** https://arxiv.org/abs/2601.22230 · **PDF:** https://arxiv.org/pdf/2601.22230v1
+
+## 17927. Lost in Space? Vision-Language Models Struggle with Relative Camera Pose Estimation
+
+- **ID:** `2601.22228` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Ken Deng, Yifu Qiu, Yoni Kasten, et al.
+- **abs:** https://arxiv.org/abs/2601.22228 · **PDF:** https://arxiv.org/pdf/2601.22228v1
+
+## 17928. RedSage: A Cybersecurity Generalist LLM
+
+- **ID:** `2601.22159` · **primary:** `cs.CR` · **Submitted:** 2026-01-29
+- **Authors:** Naufal Suryanto, Muzammal Naseer, Pengfei Li, et al.
+- **abs:** https://arxiv.org/abs/2601.22159 · **PDF:** https://arxiv.org/pdf/2601.22159v2
+
+## 17929. Discovering Hidden Gems in Model Repositories
+
+- **ID:** `2601.22157` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jonathan Kahana, Eliahu Horwitz, Yedid Hoshen
+- **abs:** https://arxiv.org/abs/2601.22157 · **PDF:** https://arxiv.org/pdf/2601.22157v1
+
+## 17930. Hybrid Linear Attention Done Right: Efficient Distillation and Effective Architectures for Extremely Long Contexts
+
+- **ID:** `2601.22156` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Yingfa Chen, Zhen Leng Thai, Zihan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.22156 · **PDF:** https://arxiv.org/pdf/2601.22156v1
+
+## 17931. UEval: A Benchmark for Unified Multimodal Generation
+
+- **ID:** `2601.22155` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Bo Li, Yida Yin, Wenhao Chai, et al.
+- **abs:** https://arxiv.org/abs/2601.22155 · **PDF:** https://arxiv.org/pdf/2601.22155v1
+
+## 17932. Exploring Reasoning Reward Model for Agents
+
+- **ID:** `2601.22154` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Kaixuan Fan, Kaituo Feng, Manyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22154 · **PDF:** https://arxiv.org/pdf/2601.22154v1
+
+## 17933. Late Breaking Results: Conversion of Neural Networks into Logic Flows for Edge Computing
+
+- **ID:** `2601.22151` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Daniel Stein, Shaoyi Huang, Rolf Drechsler, et al.
+- **abs:** https://arxiv.org/abs/2601.22151 · **PDF:** https://arxiv.org/pdf/2601.22151v1
+
+## 17934. DynaWeb: Model-Based Reinforcement Learning of Web Agents
+
+- **ID:** `2601.22149` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Hang Ding, Peidong Liu, Junqiao Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.22149 · **PDF:** https://arxiv.org/pdf/2601.22149v1
+
+## 17935. FineInstructions: Scaling Synthetic Instructions to Pre-Training Scale
+
+- **ID:** `2601.22146` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Ajay Patel, Colin Raffel, Chris Callison-Burch
+- **abs:** https://arxiv.org/abs/2601.22146 · **PDF:** https://arxiv.org/pdf/2601.22146v1
+
+## 17936. Routing the Lottery: Adaptive Subnetworks for Heterogeneous Data
+
+- **ID:** `2601.22141` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Grzegorz Stefanski, Alberto Presta, Michal Byra
+- **abs:** https://arxiv.org/abs/2601.22141 · **PDF:** https://arxiv.org/pdf/2601.22141v1
+
+## 17937. Reasoning While Asking: Transforming Reasoning Large Language Models from Passive Solvers to Proactive Inquirers
+
+- **ID:** `2601.22139` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Xin Chen, Feng Jiang, Yiqian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22139 · **PDF:** https://arxiv.org/pdf/2601.22139v1
+
+## 17938. StepShield: When, Not Whether to Intervene on Rogue Agents
+
+- **ID:** `2601.22136` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Gloria Felicia, Michael Eniolade, Jinfeng He, et al.
+- **abs:** https://arxiv.org/abs/2601.22136 · **PDF:** https://arxiv.org/pdf/2601.22136v1
+
+## 17939. PRISM: Distribution-free Adaptive Computation of Matrix Functions for Accelerating Neural Network Training
+
+- **ID:** `2601.22137` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Shenghao Yang, Zhichao Wang, Oleg Balabanov, et al.
+- **abs:** https://arxiv.org/abs/2601.22137 · **PDF:** https://arxiv.org/pdf/2601.22137v1
+
+## 17940. Securing SIM-Assisted Wireless Networks via Quantum Reinforcement Learning
+
+- **ID:** `2602.13238` · **primary:** `cs.NI` · **Submitted:** 2026-01-29
+- **Authors:** Le-Hung Hoang, Quang-Trung Luu, Dinh Thai Hoang, et al.
+- **abs:** https://arxiv.org/abs/2602.13238 · **PDF:** https://arxiv.org/pdf/2602.13238v1
+
+## 17941. ReasoningBomb: A Stealthy Denial-of-Service Attack by Inducing Pathologically Long Reasoning in Large Reasoning Models
+
+- **ID:** `2602.00154` · **primary:** `cs.CR` · **Submitted:** 2026-01-29
+- **Authors:** Xiaogeng Liu, Xinyan Wang, Yechao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.00154 · **PDF:** https://arxiv.org/pdf/2602.00154v1
+
+## 17942. Pay for Hints, Not Answers: LLM Shepherding for Cost-Efficient Inference
+
+- **ID:** `2601.22132` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Ziming Dong, Hardik Sharma, Evan O'Toole, et al.
+- **abs:** https://arxiv.org/abs/2601.22132 · **PDF:** https://arxiv.org/pdf/2601.22132v1
+
+## 17943. SMOG: Scalable Meta-Learning for Multi-Objective Bayesian Optimization
+
+- **ID:** `2601.22131` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Leonard Papenmeier, Petru Tighineanu
+- **abs:** https://arxiv.org/abs/2601.22131 · **PDF:** https://arxiv.org/pdf/2601.22131v1
+
+## 17944. World of Workflows: A Benchmark for Bringing World Models to Enterprise Systems
+
+- **ID:** `2601.22130` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Lakshya Gupta, Litao Li, Yizhe Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.22130 · **PDF:** https://arxiv.org/pdf/2601.22130v2
+
+## 17945. SWE-Replay: Efficient Test-Time Scaling for Software Engineering Agents
+
+- **ID:** `2601.22129` · **primary:** `cs.SE` · **Submitted:** 2026-01-29
+- **Authors:** Yifeng Ding, Lingming Zhang
+- **abs:** https://arxiv.org/abs/2601.22129 · **PDF:** https://arxiv.org/pdf/2601.22129v2
+
+## 17946. The Patient is not a Moving Document: A World Model Training Paradigm for Longitudinal EHR
+
+- **ID:** `2601.22128` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Irsyad Adam, Zekai Chen, David Laprade, et al.
+- **abs:** https://arxiv.org/abs/2601.22128 · **PDF:** https://arxiv.org/pdf/2601.22128v1
+
+## 17947. EditYourself: Audio-Driven Generation and Manipulation of Talking Head Videos with Diffusion Transformers
+
+- **ID:** `2601.22127` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** John Flynn, Wolfgang Paier, Dimitar Dinev, et al.
+- **abs:** https://arxiv.org/abs/2601.22127 · **PDF:** https://arxiv.org/pdf/2601.22127v1
+
+## 17948. Latent Spherical Flow Policy for Reinforcement Learning with Combinatorial Actions
+
+- **ID:** `2601.22211` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Lingkai Kong, Anagha Satish, Hezi Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.22211 · **PDF:** https://arxiv.org/pdf/2601.22211v1
+
+## 17949. A Federated and Parameter-Efficient Framework for Large Language Model Training in Medicine
+
+- **ID:** `2601.22124` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Anran Li, Yuanyuan Chen, Wenjun Long, et al.
+- **abs:** https://arxiv.org/abs/2601.22124 · **PDF:** https://arxiv.org/pdf/2601.22124v1
+
+## 17950. Learning Hamiltonian Flow Maps: Mean Flow Consistency for Large-Timestep Molecular Dynamics
+
+- **ID:** `2601.22123` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Winfried Ripken, Michael Plainer, Gregor Lied, et al.
+- **abs:** https://arxiv.org/abs/2601.22123 · **PDF:** https://arxiv.org/pdf/2601.22123v3
+
+## 17951. Alpha Discovery via Grammar-Guided Learning and Search
+
+- **ID:** `2601.22119` · **primary:** `q-fin.CP` · **Submitted:** 2026-01-29
+- **Authors:** Han Yang, Dong Hao, Zhuohan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.22119 · **PDF:** https://arxiv.org/pdf/2601.22119v1
+
+## 17952. Defining Operational Conditions for Safety-Critical AI-Based Systems from Data
+
+- **ID:** `2601.22118` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Johann Christensen, Elena Hoemann, Frank Köster, et al.
+- **abs:** https://arxiv.org/abs/2601.22118 · **PDF:** https://arxiv.org/pdf/2601.22118v1
+
+## 17953. Diverse Approaches to Optimal Execution Schedule Generation
+
+- **ID:** `2601.22113` · **primary:** `q-fin.TR` · **Submitted:** 2026-01-29
+- **Authors:** Robert de Witt, Mikko S. Pakkanen
+- **abs:** https://arxiv.org/abs/2601.22113 · **PDF:** https://arxiv.org/pdf/2601.22113v2
+
+## 17954. SINA: A Circuit Schematic Image-to-Netlist Generator Using Artificial Intelligence
+
+- **ID:** `2601.22114` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Saoud Aldowaish, Yashwanth Karumanchi, Kai-Chen Chiang, et al.
+- **abs:** https://arxiv.org/abs/2601.22114 · **PDF:** https://arxiv.org/pdf/2601.22114v1
+
+## 17955. Physics Informed Reconstruction of Four-Dimensional Atmospheric Wind Fields Using Multi-UAS Swarm Observations in a Synthetic Turbulent Environment
+
+- **ID:** `2601.22111` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Abdullah Tasim, Wei Sun
+- **abs:** https://arxiv.org/abs/2601.22111 · **PDF:** https://arxiv.org/pdf/2601.22111v1
+
+## 17956. Value-Based Pre-Training with Downstream Feedback
+
+- **ID:** `2601.22108` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Shuqi Ke, Giulia Fanti
+- **abs:** https://arxiv.org/abs/2601.22108 · **PDF:** https://arxiv.org/pdf/2601.22108v1
+
+## 17957. Prior-Informed Flow Matching for Graph Reconstruction
+
+- **ID:** `2601.22107` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Harvey Chen, Nicolas Zilberstein, Santiago Segarra
+- **abs:** https://arxiv.org/abs/2601.22107 · **PDF:** https://arxiv.org/pdf/2601.22107v1
+
+## 17958. ECO: Quantized Training without Full-Precision Master Weights
+
+- **ID:** `2601.22101` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Mahdi Nikdan, Amir Zandieh, Dan Alistarh, et al.
+- **abs:** https://arxiv.org/abs/2601.22101 · **PDF:** https://arxiv.org/pdf/2601.22101v1
+
+## 17959. Boosting CVaR Policy Optimization with Quantile Gradients
+
+- **ID:** `2601.22100` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yudong Luo, Erick Delage
+- **abs:** https://arxiv.org/abs/2601.22100 · **PDF:** https://arxiv.org/pdf/2601.22100v2
+
+## 17960. GeoNorm: Unify Pre-Norm and Post-Norm with Geodesic Optimization
+
+- **ID:** `2601.22095` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Chuanyang Zheng, Jiankai Sun, Yihang Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.22095 · **PDF:** https://arxiv.org/pdf/2601.22095v1
+
+## 17961. Investigating Associational Biases in Inter-Model Communication of Large Generative Models
+
+- **ID:** `2601.22093` · **primary:** `cs.CY` · **Submitted:** 2026-01-29
+- **Authors:** Fethiye Irmak Dogan, Yuval Weiss, Kajal Patel, et al.
+- **abs:** https://arxiv.org/abs/2601.22093 · **PDF:** https://arxiv.org/pdf/2601.22093v1
+
+## 17962. Learning to Recommend Multi-Agent Subgraphs from Calling Trees
+
+- **ID:** `2601.22209` · **primary:** `cs.MA` · **Submitted:** 2026-01-29
+- **Authors:** Xinyuan Song, Liang Zhao
+- **abs:** https://arxiv.org/abs/2601.22209 · **PDF:** https://arxiv.org/pdf/2601.22209v1
+
+## 17963. Latent Adversarial Regularization for Offline Preference Optimization
+
+- **ID:** `2601.22083` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Enyi Jiang, Yibo Jacky Zhang, Yinglun Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.22083 · **PDF:** https://arxiv.org/pdf/2601.22083v2
+
+## 17964. Where Do the Joules Go? Diagnosing Inference Energy Consumption
+
+- **ID:** `2601.22076` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jae-Won Chung, Ruofan Wu, Jeff J. Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.22076 · **PDF:** https://arxiv.org/pdf/2601.22076v2
+
+## 17965. Lens-descriptor guided evolutionary algorithm for optimization of complex optical systems with glass choice
+
+- **ID:** `2601.22075` · **primary:** `cs.NE` · **Submitted:** 2026-01-29
+- **Authors:** Kirill Antonov, Teus Tukker, Tiago Botari, et al.
+- **abs:** https://arxiv.org/abs/2601.22075 · **PDF:** https://arxiv.org/pdf/2601.22075v1
+
+## 17966. VTC-R1: Vision-Text Compression for Efficient Long-Context Reasoning
+
+- **ID:** `2601.22069` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Yibo Wang, Yongcheng Jing, Shunyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.22069 · **PDF:** https://arxiv.org/pdf/2601.22069v2
+
+## 17967. Making Foundation Models Probabilistic via Singular Value Ensembles
+
+- **ID:** `2601.22068` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Mehmet Ozgur Turkoglu, Dominik J. Mühlematter, Alexander Becker, et al.
+- **abs:** https://arxiv.org/abs/2601.22068 · **PDF:** https://arxiv.org/pdf/2601.22068v1
+
+## 17968. Causal Imitation Learning Under Measurement Error and Distribution Shift
+
+- **ID:** `2601.22206` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Shi Bo, AmirEmad Ghassami
+- **abs:** https://arxiv.org/abs/2601.22206 · **PDF:** https://arxiv.org/pdf/2601.22206v1
+
+## 17969. Vision-DeepResearch: Incentivizing DeepResearch Capability in Multimodal Large Language Models
+
+- **ID:** `2601.22060` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Wenxuan Huang, Yu Zeng, Qiuchen Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.22060 · **PDF:** https://arxiv.org/pdf/2601.22060v3
+
+## 17970. Unsupervised Decomposition and Recombination with Discriminator-Driven Diffusion Models
+
+- **ID:** `2601.22057` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Archer Wang, Emile Anand, Yilun Du, et al.
+- **abs:** https://arxiv.org/abs/2601.22057 · **PDF:** https://arxiv.org/pdf/2601.22057v2
+
+## 17971. $G^2$-Reader: Dual Evolving Graphs for Multimodal Document QA
+
+- **ID:** `2601.22055` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Yaxin Du, Junru Song, Yifan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.22055 · **PDF:** https://arxiv.org/pdf/2601.22055v1
+
+## 17972. MetricAnything: Scaling Metric Depth Pretraining with Noisy Heterogeneous Sources
+
+- **ID:** `2601.22054` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Baorui Ma, Jiahui Yang, Donglin Di, et al.
+- **abs:** https://arxiv.org/abs/2601.22054 · **PDF:** https://arxiv.org/pdf/2601.22054v1
+
+## 17973. MasalBench: A Benchmark for Contextual and Cross-Cultural Understanding of Persian Proverbs in LLMs
+
+- **ID:** `2601.22050` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Ghazal Kalhor, Behnam Bahrak
+- **abs:** https://arxiv.org/abs/2601.22050 · **PDF:** https://arxiv.org/pdf/2601.22050v1
+
+## 17974. On the Paradoxical Interference between Instruction-Following and Task Solving
+
+- **ID:** `2601.22047` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Yunjia Qi, Hao Peng, Xintong Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.22047 · **PDF:** https://arxiv.org/pdf/2601.22047v1
+
+## 17975. FedAdaVR: Adaptive Variance Reduction for Robust Federated Learning under Limited Client Participation
+
+- **ID:** `2601.22204` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** S M Ruhul Kabir Howlader, Xiao Chen, Yifei Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.22204 · **PDF:** https://arxiv.org/pdf/2601.22204v1
+
+## 17976. SIA: Symbolic Interpretability for Anticipatory Deep Reinforcement Learning in Network Control
+
+- **ID:** `2601.22044` · **primary:** `cs.NI` · **Submitted:** 2026-01-29
+- **Authors:** MohammadErfan Jabbari, Abhishek Duttagupta, Claudio Fiandrino, et al.
+- **abs:** https://arxiv.org/abs/2601.22044 · **PDF:** https://arxiv.org/pdf/2601.22044v1
+
+## 17977. Learning to Communicate Across Modalities: Perceptual Heterogeneity in Multi-Agent Systems
+
+- **ID:** `2601.22041` · **primary:** `cs.MA` · **Submitted:** 2026-01-29
+- **Authors:** Naomi Pitzer, Daniela Mihai
+- **abs:** https://arxiv.org/abs/2601.22041 · **PDF:** https://arxiv.org/pdf/2601.22041v1
+
+## 17978. A Separable Architecture for Continuous Token Representation in Language Models
+
+- **ID:** `2601.22040` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Reza T. Batley, Sourav Saha
+- **abs:** https://arxiv.org/abs/2601.22040 · **PDF:** https://arxiv.org/pdf/2601.22040v1
+
+## 17979. Beyond Conditional Computation: Retrieval-Augmented Genomic Foundation Models with Gengram
+
+- **ID:** `2601.22203` · **primary:** `q-bio.GN` · **Submitted:** 2026-01-29
+- **Authors:** Huinan Xu, Xuyang Feng, Junhong Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.22203 · **PDF:** https://arxiv.org/pdf/2601.22203v1
+
+## 17980. Optimizing Agentic Workflows using Meta-tools
+
+- **ID:** `2601.22037` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Sami Abuzakuk, Anne-Marie Kermarrec, Rishi Sharma, et al.
+- **abs:** https://arxiv.org/abs/2601.22037 · **PDF:** https://arxiv.org/pdf/2601.22037v2
+
+## 17981. Cross-Fusion Distance: A Novel Metric for Measuring Fusion and Separability Between Data Groups in Representation Space
+
+- **ID:** `2601.22036` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Xiaolong Zhang, Jianwei Zhang, Xubo Song
+- **abs:** https://arxiv.org/abs/2601.22036 · **PDF:** https://arxiv.org/pdf/2601.22036v1
+
+## 17982. Thinking Out of Order: When Output Order Stops Reflecting Reasoning Order in Diffusion Language Models
+
+- **ID:** `2601.22035` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Longxuan Yu, Yu Fu, Shaorong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.22035 · **PDF:** https://arxiv.org/pdf/2601.22035v1
+
+## 17983. Holographic generative flows with AdS/CFT
+
+- **ID:** `2601.22033` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Ehsan Mirafzali, Sanjit Shashi, Sanya Murdeshwar, et al.
+- **abs:** https://arxiv.org/abs/2601.22033 · **PDF:** https://arxiv.org/pdf/2601.22033v1
+
+## 17984. Causal Autoregressive Diffusion Language Model
+
+- **ID:** `2601.22031` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Junhao Ruan, Bei Li, Yongjing Yin, et al.
+- **abs:** https://arxiv.org/abs/2601.22031 · **PDF:** https://arxiv.org/pdf/2601.22031v1
+
+## 17985. Per-parameter Task Arithmetic for Unlearning in Large Language Models
+
+- **ID:** `2601.22030` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Chengyi Cai, Zesheng Ye, Jiangchao Yao, et al.
+- **abs:** https://arxiv.org/abs/2601.22030 · **PDF:** https://arxiv.org/pdf/2601.22030v1
+
+## 17986. The Ensemble Inverse Problem: Applications and Methods
+
+- **ID:** `2601.22029` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Zhengyan Huan, Camila Pazos, Martin Klassen, et al.
+- **abs:** https://arxiv.org/abs/2601.22029 · **PDF:** https://arxiv.org/pdf/2601.22029v1
+
+## 17987. From Logits to Latents: Contrastive Representation Shaping for LLM Unlearning
+
+- **ID:** `2601.22028` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Haoran Tang, Rajiv Khanna
+- **abs:** https://arxiv.org/abs/2601.22028 · **PDF:** https://arxiv.org/pdf/2601.22028v1
+
+## 17988. CAR-bench: Evaluating the Consistency and Limit-Awareness of LLM Agents under Real-World Uncertainty
+
+- **ID:** `2601.22027` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Johannes Kirmayr, Lukas Stappen, Elisabeth André
+- **abs:** https://arxiv.org/abs/2601.22027 · **PDF:** https://arxiv.org/pdf/2601.22027v1
+
+## 17989. When "Better" Prompts Hurt: Evaluation-Driven Iteration for LLM Applications
+
+- **ID:** `2601.22025` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Daniel Commey
+- **abs:** https://arxiv.org/abs/2601.22025 · **PDF:** https://arxiv.org/pdf/2601.22025v1
+
+## 17990. SymbXRL: Symbolic Explainable Deep Reinforcement Learning for Mobile Networks
+
+- **ID:** `2601.22024` · **primary:** `cs.NI` · **Submitted:** 2026-01-29
+- **Authors:** Abhishek Duttagupta, MohammadErfan Jabbari, Claudio Fiandrino, et al.
+- **abs:** https://arxiv.org/abs/2601.22024 · **PDF:** https://arxiv.org/pdf/2601.22024v1
+
+## 17991. Visual-Guided Key-Token Regularization for Multimodal Large Language Model Unlearning
+
+- **ID:** `2601.22020` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Chengyi Cai, Zesheng Ye, Peike Li, et al.
+- **abs:** https://arxiv.org/abs/2601.22020 · **PDF:** https://arxiv.org/pdf/2601.22020v1
+
+## 17992. TBDFiltering: Sample-Efficient Tree-Based Data Filtering
+
+- **ID:** `2601.22016` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Robert Istvan Busa-Fekete, Julian Zimmert, Anne Xiangyi Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.22016 · **PDF:** https://arxiv.org/pdf/2601.22016v1
+
+## 17993. Vidmento: Creating Video Stories Through Context-Aware Expansion With Generative Video
+
+- **ID:** `2601.22013` · **primary:** `cs.HC` · **Submitted:** 2026-01-29
+- **Authors:** Catherine Yeh, Anh Truong, Mira Dontcheva, et al.
+- **abs:** https://arxiv.org/abs/2601.22013 · **PDF:** https://arxiv.org/pdf/2601.22013v2
+
+## 17994. Putting a Face to Forgetting: Continual Learning meets Mechanistic Interpretability
+
+- **ID:** `2601.22012` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Sergi Masip, Gido M. van de Ven, Javier Ferrando, et al.
+- **abs:** https://arxiv.org/abs/2601.22012 · **PDF:** https://arxiv.org/pdf/2601.22012v1
+
+## 17995. Exploring Diverse Generation Paths via Inference-time Stiefel Activation Steering
+
+- **ID:** `2601.22010` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Dongxuan Zhu, Ly Tran Ho Khanh, Andy Yat-Ming Cheung, et al.
+- **abs:** https://arxiv.org/abs/2601.22010 · **PDF:** https://arxiv.org/pdf/2601.22010v1
+
+## 17996. MEIDNet: Multimodal generative AI framework for inverse materials design
+
+- **ID:** `2601.22009` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-29
+- **Authors:** Anand Babu, Rogério Almeida Gouvêa, Pierre Vandergheynst, et al.
+- **abs:** https://arxiv.org/abs/2601.22009 · **PDF:** https://arxiv.org/pdf/2601.22009v1
+
+## 17997. Hierarchy of discriminative power and complexity in learning quantum ensembles
+
+- **ID:** `2601.22005` · **primary:** `quant-ph` · **Submitted:** 2026-01-29
+- **Authors:** Jian Yao, Pengtao Li, Xiaohui Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.22005 · **PDF:** https://arxiv.org/pdf/2601.22005v1
+
+## 17998. Efficient Stochastic Optimisation via Sequential Monte Carlo
+
+- **ID:** `2601.22003` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** James Cuin, Davide Carbone, Yanbo Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.22003 · **PDF:** https://arxiv.org/pdf/2601.22003v1
+
+## 17999. Rate-Distortion Optimization for Transformer Inference
+
+- **ID:** `2601.22002` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Anderson de Andrade, Alon Harell, Ivan V. Bajić
+- **abs:** https://arxiv.org/abs/2601.22002 · **PDF:** https://arxiv.org/pdf/2601.22002v2
+
+## 18000. Heterogeneous Computing: The Key to Powering the Future of AI Agent Inference
+
+- **ID:** `2601.22001` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Yiren Zhao, Junyi Liu
+- **abs:** https://arxiv.org/abs/2601.22001 · **PDF:** https://arxiv.org/pdf/2601.22001v1
+
+## 18001. Negatives-Dominant Contrastive Learning for Generalization in Imbalanced Domains
+
+- **ID:** `2601.21999` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Meng Cao, Jiexi Liu, Songcan Chen
+- **abs:** https://arxiv.org/abs/2601.21999 · **PDF:** https://arxiv.org/pdf/2601.21999v1
+
+## 18002. Mechanistic Data Attribution: Tracing the Training Origins of Interpretable LLM Units
+
+- **ID:** `2601.21996` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Jianhui Chen, Yuzhang Luo, Liangming Pan
+- **abs:** https://arxiv.org/abs/2601.21996 · **PDF:** https://arxiv.org/pdf/2601.21996v1
+
+## 18003. Liquid Interfaces: A Dynamic Ontology for the Interoperability of Autonomous Systems
+
+- **ID:** `2601.21993` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Dhiogo de Sá, Carlos Schmiedel, Carlos Pereira Lopes
+- **abs:** https://arxiv.org/abs/2601.21993 · **PDF:** https://arxiv.org/pdf/2601.21993v1
+
+## 18004. Geometry of Drifting MDPs with Path-Integral Stability Certificates
+
+- **ID:** `2601.21991` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Zuyuan Zhang, Mahdi Imani, Tian Lan
+- **abs:** https://arxiv.org/abs/2601.21991 · **PDF:** https://arxiv.org/pdf/2601.21991v1
+
+## 18005. Batched First-Order Methods for Parallel LP Solving in MIP
+
+- **ID:** `2601.21990` · **primary:** `math.OC` · **Submitted:** 2026-01-29
+- **Authors:** Nicolas Blin, Stefano Gualandi, Christopher Maes, et al.
+- **abs:** https://arxiv.org/abs/2601.21990 · **PDF:** https://arxiv.org/pdf/2601.21990v1
+
+## 18006. Generalized Information Gathering Under Dynamics Uncertainty
+
+- **ID:** `2601.21988` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Fernando Palafox, Jingqi Li, Jesse Milzman, et al.
+- **abs:** https://arxiv.org/abs/2601.21988 · **PDF:** https://arxiv.org/pdf/2601.21988v1
+
+## 18007. Elign: Equivariant Diffusion Model Alignment from Foundational Machine Learning Force Fields
+
+- **ID:** `2601.21985` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yunyang Li, Lin Huang, Luojia Xia, et al.
+- **abs:** https://arxiv.org/abs/2601.21985 · **PDF:** https://arxiv.org/pdf/2601.21985v1
+
+## 18008. PowerGenie: Analytically-Guided Evolutionary Discovery of Superior Reconfigurable Power Converters
+
+- **ID:** `2601.21984` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jian Gao, Yiwei Zou, Abhishek Pradhan, et al.
+- **abs:** https://arxiv.org/abs/2601.21984 · **PDF:** https://arxiv.org/pdf/2601.21984v3
+
+## 18009. Investigating Batch Inference in a Sequential Monte Carlo Framework for Neural Networks
+
+- **ID:** `2601.21983` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Andrew Millard, Joshua Murphy, Peter Green, et al.
+- **abs:** https://arxiv.org/abs/2601.21983 · **PDF:** https://arxiv.org/pdf/2601.21983v1
+
+## 18010. VERSA: Verified Event Data Format for Reliable Soccer Analytics
+
+- **ID:** `2601.21981` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Geonhee Jo, Mingu Kang, Kangmin Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.21981 · **PDF:** https://arxiv.org/pdf/2601.21981v1
+
+## 18011. Investigation into using stochastic embedding representations for evaluating the trustworthiness of the Fréchet Inception Distance
+
+- **ID:** `2601.21979` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Ciaran Bench, Vivek Desai, Carlijn Roozemond, et al.
+- **abs:** https://arxiv.org/abs/2601.21979 · **PDF:** https://arxiv.org/pdf/2601.21979v1
+
+## 18012. Bridging Graph Structure and Knowledge-Guided Editing for Interpretable Temporal Knowledge Graph Reasoning
+
+- **ID:** `2601.21978` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Shiqi Fan, Quanming Yao, Hongyi Nie, et al.
+- **abs:** https://arxiv.org/abs/2601.21978 · **PDF:** https://arxiv.org/pdf/2601.21978v1
+
+## 18013. From Particles to Agents: Hallucination as a Metric for Cognitive Friction in Spatial Simulation
+
+- **ID:** `2601.21977` · **primary:** `cs.HC` · **Submitted:** 2026-01-29
+- **Authors:** Javier Argota Sánchez-Vaquerizo, Luis Borunda Monsivais
+- **abs:** https://arxiv.org/abs/2601.21977 · **PDF:** https://arxiv.org/pdf/2601.21977v1
+
+## 18014. Mind the Gap: How Elicitation Protocols Shape the Stated-Revealed Preference Gap in Language Models
+
+- **ID:** `2601.21975` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Pranav Mahajan, Ihor Kendiukhov, Syed Hussain, et al.
+- **abs:** https://arxiv.org/abs/2601.21975 · **PDF:** https://arxiv.org/pdf/2601.21975v1
+
+## 18015. Learning Decentralized LLM Collaboration with Multi-Agent Actor Critic
+
+- **ID:** `2601.21972` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Shuo Liu, Tianle Chen, Ryan Amiri, et al.
+- **abs:** https://arxiv.org/abs/2601.21972 · **PDF:** https://arxiv.org/pdf/2601.21972v3
+
+## 18016. MoE-ACT: Improving Surgical Imitation Learning Policies through Supervised Mixture-of-Experts
+
+- **ID:** `2601.21971` · **primary:** `cs.RO` · **Submitted:** 2026-01-29
+- **Authors:** Lorenzo Mazza, Ariel Rodriguez, Rayan Younis, et al.
+- **abs:** https://arxiv.org/abs/2601.21971 · **PDF:** https://arxiv.org/pdf/2601.21971v1
+
+## 18017. Token-Guard: Towards Token-Level Hallucination Control via Self-Checking Decoding
+
+- **ID:** `2601.21969` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Yifan Zhu, Huiqiang Rong, Haoran Luo
+- **abs:** https://arxiv.org/abs/2601.21969 · **PDF:** https://arxiv.org/pdf/2601.21969v2
+
+## 18018. OVD: On-policy Verbal Distillation
+
+- **ID:** `2601.21968` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Jing Xiong, Hui Shen, Shansan Gong, et al.
+- **abs:** https://arxiv.org/abs/2601.21968 · **PDF:** https://arxiv.org/pdf/2601.21968v1
+
+## 18019. The Energy Impact of Domain Model Design in Classical Planning
+
+- **ID:** `2601.21967` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Ilche Georgievski, Serhat Tekin, Marco Aiello
+- **abs:** https://arxiv.org/abs/2601.21967 · **PDF:** https://arxiv.org/pdf/2601.21967v1
+
+## 18020. From Tokens to Blocks: A Block-Diffusion Perspective on Molecular Generation
+
+- **ID:** `2601.21964` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Qianwei Yang, Dong Xu, Zhangfan Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.21964 · **PDF:** https://arxiv.org/pdf/2601.21964v2
+
+## 18021. Industrialized Deception: The Collateral Effects of LLM-Generated Misinformation on Digital Ecosystems
+
+- **ID:** `2601.21963` · **primary:** `cs.CY` · **Submitted:** 2026-01-29
+- **Authors:** Alexander Loth, Martin Kappes, Marc-Oliver Pahl
+- **abs:** https://arxiv.org/abs/2601.21963 · **PDF:** https://arxiv.org/pdf/2601.21963v2
+
+## 18022. How do Visual Attributes Influence Web Agents? A Comprehensive Evaluation of User Interface Design Factors
+
+- **ID:** `2601.21961` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Kuai Yu, Naicheng Yu, Han Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21961 · **PDF:** https://arxiv.org/pdf/2601.21961v2
+
+## 18023. Near-Optimal Private Tests for Simple and MLR Hypotheses
+
+- **ID:** `2601.21959` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** Yu-Wei Chen, Raghu Pasupathy, Jordan Awan
+- **abs:** https://arxiv.org/abs/2601.21959 · **PDF:** https://arxiv.org/pdf/2601.21959v1
+
+## 18024. Uncertainty-Aware Data-Based Method for Fast and Reliable Shape Optimization
+
+- **ID:** `2601.21956` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yunjia Yang, Runze Li, Yufei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.21956 · **PDF:** https://arxiv.org/pdf/2601.21956v1
+
+## 18025. From Generative Modeling to Clinical Classification: A GPT-Based Architecture for EHR Notes
+
+- **ID:** `2601.21955` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Fariba Afrin Irany, Sampson Akwafuo
+- **abs:** https://arxiv.org/abs/2601.21955 · **PDF:** https://arxiv.org/pdf/2601.21955v3
+
+## 18026. Diffusion Path Samplers via Sequential Monte Carlo
+
+- **ID:** `2601.21951` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** James Matthew Young, Paula Cordero-Encinar, Sebastian Reich, et al.
+- **abs:** https://arxiv.org/abs/2601.21951 · **PDF:** https://arxiv.org/pdf/2601.21951v1
+
+## 18027. Embracing Aleatoric Uncertainty in Medical Multimodal Learning with Missing Modalities
+
+- **ID:** `2601.21950` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Linxiao Gong, Yang Liu, Lianlong Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.21950 · **PDF:** https://arxiv.org/pdf/2601.21950v1
+
+## 18028. ToolWeaver: Weaving Collaborative Semantics for Scalable Tool Use in Large Language Models
+
+- **ID:** `2601.21947` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Bowen Fang, Wen Ye, Yunyue Su, et al.
+- **abs:** https://arxiv.org/abs/2601.21947 · **PDF:** https://arxiv.org/pdf/2601.21947v1
+
+## 18029. Dependence of Equilibrium Propagation Training Success on Network Architecture
+
+- **ID:** `2601.21945` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Qingshan Wang, Clara C. Wanjura, Florian Marquardt
+- **abs:** https://arxiv.org/abs/2601.21945 · **PDF:** https://arxiv.org/pdf/2601.21945v1
+
+## 18030. Clarity: The Flexibility-Interpretability Trade-Off in Sparsity-aware Concept Bottleneck Models
+
+- **ID:** `2601.21944` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Konstantinos P. Panousis, Diego Marcos
+- **abs:** https://arxiv.org/abs/2601.21944 · **PDF:** https://arxiv.org/pdf/2601.21944v1
+
+## 18031. Entropy-Based Dimension-Free Convergence and Loss-Adaptive Schedules for Diffusion Models
+
+- **ID:** `2601.21943` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Ahmad Aghapour, Erhan Bayraktar, Ziqing Zhang
+- **abs:** https://arxiv.org/abs/2601.21943 · **PDF:** https://arxiv.org/pdf/2601.21943v1
+
+## 18032. Clustering in Deep Stochastic Transformers
+
+- **ID:** `2601.21942` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** Lev Fedorov, Michaël E. Sander, Romuald Elie, et al.
+- **abs:** https://arxiv.org/abs/2601.21942 · **PDF:** https://arxiv.org/pdf/2601.21942v1
+
+## 18033. Robust Multimodal Representation Learning in Healthcare
+
+- **ID:** `2601.21941` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Xiaoguang Zhu, Linxiao Gong, Lianlong Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.21941 · **PDF:** https://arxiv.org/pdf/2601.21941v1
+
+## 18034. Retrieval-Infused Reasoning Sandbox: A Benchmark for Decoupling Retrieval and Reasoning Capabilities
+
+- **ID:** `2601.21937` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Shuangshuang Ying, Zheyu Wang, Yunjian Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.21937 · **PDF:** https://arxiv.org/pdf/2601.21937v2
+
+## 18035. AgenticSimLaw: A Juvenile Courtroom Multi-Agent Debate Simulation for Explainable High-Stakes Tabular Decision Making
+
+- **ID:** `2601.21936` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Jon Chun, Kathrine Elkins, Yong Suk Lee
+- **abs:** https://arxiv.org/abs/2601.21936 · **PDF:** https://arxiv.org/pdf/2601.21936v1
+
+## 18036. LoRIF: Low-Rank Influence Functions for Scalable Training Data Attribution
+
+- **ID:** `2601.21929` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Shuangqi Li, Hieu Le, Jingyi Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.21929 · **PDF:** https://arxiv.org/pdf/2601.21929v1
+
+## 18037. SONIC: Segmented Optimized Nexus for Information Compression in Key-Value Caching
+
+- **ID:** `2601.21927` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Hong Chen, Xiang Liu, Bo Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21927 · **PDF:** https://arxiv.org/pdf/2601.21927v1
+
+## 18038. Optimistic Transfer under Task Shift via Bellman Alignment
+
+- **ID:** `2601.21924` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jinhang Chai, Enpei Zhang, Elynn Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.21924 · **PDF:** https://arxiv.org/pdf/2601.21924v1
+
+## 18039. From Future of Work to Future of Workers: Addressing Asymptomatic AI Harms for Dignified Human-AI Interaction
+
+- **ID:** `2601.21920` · **primary:** `cs.HC` · **Submitted:** 2026-01-29
+- **Authors:** Upol Ehsan, Samir Passi, Koustuv Saha, et al.
+- **abs:** https://arxiv.org/abs/2601.21920 · **PDF:** https://arxiv.org/pdf/2601.21920v1
+
+## 18040. Self-Compression of Chain-of-Thought via Multi-Agent Reinforcement Learning
+
+- **ID:** `2601.21919` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Yiqun Chen, Jinyuan Feng, Wei Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.21919 · **PDF:** https://arxiv.org/pdf/2601.21919v1
+
+## 18041. On Approximate Computation of Critical Points
+
+- **ID:** `2601.21917` · **primary:** `math.OC` · **Submitted:** 2026-01-29
+- **Authors:** Amir Ali Ahmadi, Georgina Hall
+- **abs:** https://arxiv.org/abs/2601.21917 · **PDF:** https://arxiv.org/pdf/2601.21917v1
+
+## 18042. JADE: Bridging the Strategic-Operational Gap in Dynamic Agentic RAG
+
+- **ID:** `2601.21916` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Yiqun Chen, Erhan Zhang, Tianyi Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.21916 · **PDF:** https://arxiv.org/pdf/2601.21916v1
+
+## 18043. ProRAG: Process-Supervised Reinforcement Learning for Retrieval-Augmented Generation
+
+- **ID:** `2601.21912` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Zhao Wang, Ziliang Zhao, Zhicheng Dou
+- **abs:** https://arxiv.org/abs/2601.21912 · **PDF:** https://arxiv.org/pdf/2601.21912v1
+
+## 18044. From Meta-Thought to Execution: Cognitively Aligned Post-Training for Generalizable and Reliable LLM Reasoning
+
+- **ID:** `2601.21909` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Shaojie Wang, Liang Zhang
+- **abs:** https://arxiv.org/abs/2601.21909 · **PDF:** https://arxiv.org/pdf/2601.21909v1
+
+## 18045. Hardware-Triggered Backdoors
+
+- **ID:** `2601.21902` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jonas Möller, Erik Imgrund, Thorsten Eisenhofer, et al.
+- **abs:** https://arxiv.org/abs/2601.21902 · **PDF:** https://arxiv.org/pdf/2601.21902v2
+
+## 18046. TraceRouter: Robust Safety for Large Foundation Models via Path-Level Intervention
+
+- **ID:** `2601.21900` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Chuancheng Shi, Shangze Li, Wenjun Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.21900 · **PDF:** https://arxiv.org/pdf/2601.21900v2
+
+## 18047. Breaking the Regional Barrier: Inductive Semantic Topology Learning for Worldwide Air Quality Forecasting
+
+- **ID:** `2601.21899` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Zhiqing Cui, Siru Zhong, Ming Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.21899 · **PDF:** https://arxiv.org/pdf/2601.21899v1
+
+## 18048. Adaptive Benign Overfitting (ABO): Overparameterized RLS for Online Learning in Non-stationary Time-series
+
+- **ID:** `2601.22200` · **primary:** `q-fin.ST` · **Submitted:** 2026-01-29
+- **Authors:** Luis Ontaneda Mijares, Nick Firoozye
+- **abs:** https://arxiv.org/abs/2601.22200 · **PDF:** https://arxiv.org/pdf/2601.22200v1
+
+## 18049. Making Models Unmergeable via Scaling-Sensitive Loss Landscape
+
+- **ID:** `2601.21898` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Minwoo Jang, Hoyoung Kim, Jabin Koo, et al.
+- **abs:** https://arxiv.org/abs/2601.21898 · **PDF:** https://arxiv.org/pdf/2601.21898v1
+
+## 18050. A Low-Complexity Plug-and-Play Deep Learning Model for Generalizable Massive MIMO Precoding
+
+- **ID:** `2601.21897` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Ali Hasanzadeh Karkan, Ahmed Ibrahim, Jean-François Frigon, et al.
+- **abs:** https://arxiv.org/abs/2601.21897 · **PDF:** https://arxiv.org/pdf/2601.21897v1
+
+## 18051. Learn-to-Distance: Distance Learning for Detecting LLM-Generated Text
+
+- **ID:** `2601.21895` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Hongyi Zhou, Jin Zhu, Kai Ye, et al.
+- **abs:** https://arxiv.org/abs/2601.21895 · **PDF:** https://arxiv.org/pdf/2601.21895v2
+
+## 18052. Not All Code Is Equal: A Data-Centric Study of Code Complexity and LLM Reasoning
+
+- **ID:** `2601.21894` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Lukas Twist, Shu Yang, Hanqi Yan, et al.
+- **abs:** https://arxiv.org/abs/2601.21894 · **PDF:** https://arxiv.org/pdf/2601.21894v1
+
+## 18053. Improving Classifier-Free Guidance of Flow Matching via Manifold Projection
+
+- **ID:** `2601.21892` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Jian-Feng Cai, Haixia Liu, Zhengyi Su, et al.
+- **abs:** https://arxiv.org/abs/2601.21892 · **PDF:** https://arxiv.org/pdf/2601.21892v1
+
+## 18054. VSE: Variational state estimation of complex model-free process
+
+- **ID:** `2601.21887` · **primary:** `eess.SP` · **Submitted:** 2026-01-29
+- **Authors:** Gustav Norén, Anubhab Ghosh, Fredrik Cumlin, et al.
+- **abs:** https://arxiv.org/abs/2601.21887 · **PDF:** https://arxiv.org/pdf/2601.21887v1
+
+## 18055. Adaptive Surrogate-Based Strategy for Accelerating Convergence Speed when Solving Expensive Unconstrained Multi-Objective Optimisation Problems
+
+- **ID:** `2601.21885` · **primary:** `cs.NE` · **Submitted:** 2026-01-29
+- **Authors:** Tiwonge Msulira Banda, Alexandru-Ciprian Zăvoianu
+- **abs:** https://arxiv.org/abs/2601.21885 · **PDF:** https://arxiv.org/pdf/2601.21885v1
+
+## 18056. Managing Solution Stability in Decision-Focused Learning with Cost Regularization
+
+- **ID:** `2601.21883` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Victor Spitzer, Francois Sanson
+- **abs:** https://arxiv.org/abs/2601.21883 · **PDF:** https://arxiv.org/pdf/2601.21883v1
+
+## 18057. How Expressive Are Graph Neural Networks in the Presence of Node Identifiers?
+
+- **ID:** `2601.21882` · **primary:** `cs.LO` · **Submitted:** 2026-01-29
+- **Authors:** Arie Soeteman, Michael Benedikt, Martin Grohe, et al.
+- **abs:** https://arxiv.org/abs/2601.21882 · **PDF:** https://arxiv.org/pdf/2601.21882v1
+
+## 18058. astra-langchain4j: Experiences Combining LLMs and Agent Programming
+
+- **ID:** `2601.21879` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Rem Collier, Katharine Beaumont, Andrei Ciortea
+- **abs:** https://arxiv.org/abs/2601.21879 · **PDF:** https://arxiv.org/pdf/2601.21879v1
+
+## 18059. Evolution of Benchmark: Black-Box Optimization Benchmark Design through Large Language Model
+
+- **ID:** `2601.21877` · **primary:** `cs.NE` · **Submitted:** 2026-01-29
+- **Authors:** Chen Wang, Sijie Ma, Zeyuan Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.21877 · **PDF:** https://arxiv.org/pdf/2601.21877v2
+
+## 18060. Low-Rank Plus Sparse Matrix Transfer Learning under Growing Representations and Ambient Dimensions
+
+- **ID:** `2601.21873` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jinhang Chai, Xuyuan Liu, Elynn Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.21873 · **PDF:** https://arxiv.org/pdf/2601.21873v1
+
+## 18061. WebArbiter: A Principle-Guided Reasoning Process Reward Model for Web Agents
+
+- **ID:** `2601.21872` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Yao Zhang, Shijie Tang, Zeyu Li, et al.
+- **abs:** https://arxiv.org/abs/2601.21872 · **PDF:** https://arxiv.org/pdf/2601.21872v1
+
+## 18062. See Without Decoding: Motion-Vector-Based Tracking in Compressed Video
+
+- **ID:** `2602.00153` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Axel Duché, Clément Chatelain, Gilles Gasso
+- **abs:** https://arxiv.org/abs/2602.00153 · **PDF:** https://arxiv.org/pdf/2602.00153v1
+
+## 18063. On Forgetting and Stability of Score-based Generative models
+
+- **ID:** `2601.21868` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** Stanislas Strasman, Gabriel Cardoso, Sylvain Le Corff, et al.
+- **abs:** https://arxiv.org/abs/2601.21868 · **PDF:** https://arxiv.org/pdf/2601.21868v1
+
+## 18064. MoHETS: Long-term Time Series Forecasting with Mixture-of-Heterogeneous-Experts
+
+- **ID:** `2601.21866` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Evandro S. Ortigossa, Guy Lutsker, Eran Segal
+- **abs:** https://arxiv.org/abs/2601.21866 · **PDF:** https://arxiv.org/pdf/2601.21866v2
+
+## 18065. KnowBias: Mitigating Social Bias in LLMs via Know-Bias Neuron Enhancement
+
+- **ID:** `2601.21864` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Jinhao Pan, Chahat Raj, Anjishnu Mukherjee, et al.
+- **abs:** https://arxiv.org/abs/2601.21864 · **PDF:** https://arxiv.org/pdf/2601.21864v1
+
+## 18066. Pathwise Learning of Stochastic Dynamical Systems with Partial Observations
+
+- **ID:** `2601.21860` · **primary:** `math.OC` · **Submitted:** 2026-01-29
+- **Authors:** Nicole Tianjiao Yang
+- **abs:** https://arxiv.org/abs/2601.21860 · **PDF:** https://arxiv.org/pdf/2601.21860v1
+
+## 18067. Blind Ultrasound Image Enhancement via Self-Supervised Physics-Guided Degradation Modeling
+
+- **ID:** `2601.21856` · **primary:** `eess.IV` · **Submitted:** 2026-01-29
+- **Authors:** Shujaat Khan, Syed Muhammad Atif, Jaeyoung Huh, et al.
+- **abs:** https://arxiv.org/abs/2601.21856 · **PDF:** https://arxiv.org/pdf/2601.21856v1
+
+## 18068. Advanced techniques and applications of LiDAR Place Recognition in Agricultural Environments: A Comprehensive Survey
+
+- **ID:** `2601.22198` · **primary:** `cs.RO` · **Submitted:** 2026-01-29
+- **Authors:** Judith Vilella-Cantos, Mónica Ballesta, David Valiente, et al.
+- **abs:** https://arxiv.org/abs/2601.22198 · **PDF:** https://arxiv.org/pdf/2601.22198v1
+
+## 18069. LEMUR: Learned Multi-Vector Retrieval
+
+- **ID:** `2601.21853` · **primary:** `cs.IR` · **Submitted:** 2026-01-29
+- **Authors:** Elias Jääsaari, Ville Hyvönen, Teemu Roos
+- **abs:** https://arxiv.org/abs/2601.21853 · **PDF:** https://arxiv.org/pdf/2601.21853v1
+
+## 18070. Visual Disentangled Diffusion Autoencoders: Scalable Counterfactual Generation for Foundation Models
+
+- **ID:** `2601.21851` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Sidney Bender, Marco Morik
+- **abs:** https://arxiv.org/abs/2601.21851 · **PDF:** https://arxiv.org/pdf/2601.21851v1
+
+## 18071. READY: Reward Discovery for Meta-Black-Box Optimization
+
+- **ID:** `2601.21847` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Zechuan Huang, Zhiguang Cao, Hongshu Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.21847 · **PDF:** https://arxiv.org/pdf/2601.21847v1
+
+## 18072. Real-Time Human Activity Recognition on Edge Microcontrollers: Dynamic Hierarchical Inference with Multi-Spectral Sensor Fusion
+
+- **ID:** `2602.00152` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Boyu Li, Kuangji Zuo, Lincong Li, et al.
+- **abs:** https://arxiv.org/abs/2602.00152 · **PDF:** https://arxiv.org/pdf/2602.00152v1
+
+## 18073. Constrained Meta Reinforcement Learning with Provable Test-Time Safety
+
+- **ID:** `2601.21845` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Tingting Ni, Maryam Kamgarpour
+- **abs:** https://arxiv.org/abs/2601.21845 · **PDF:** https://arxiv.org/pdf/2601.21845v1
+
+## 18074. Bridging Forecast Accuracy and Inventory KPIs: A Simulation-Based Software Framework
+
+- **ID:** `2601.21844` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** So Fukuhara, Abdallah Alabdallah, Nuwan Gunasekara, et al.
+- **abs:** https://arxiv.org/abs/2601.21844 · **PDF:** https://arxiv.org/pdf/2601.21844v2
+
+## 18075. Embodied Task Planning via Graph-Informed Action Generation with Large Language Model
+
+- **ID:** `2601.21841` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Xiang Li, Ning Yan, Masood Mortazavi
+- **abs:** https://arxiv.org/abs/2601.21841 · **PDF:** https://arxiv.org/pdf/2601.21841v2
+
+## 18076. Test-Time Compute Games
+
+- **ID:** `2601.21839` · **primary:** `cs.CY` · **Submitted:** 2026-01-29
+- **Authors:** Ander Artola Velasco, Dimitrios Rontogiannis, Stratis Tsirtsis, et al.
+- **abs:** https://arxiv.org/abs/2601.21839 · **PDF:** https://arxiv.org/pdf/2601.21839v1
+
+## 18077. Trustworthy Intelligent Education: A Systematic Perspective on Progress, Challenges, and Future Directions
+
+- **ID:** `2601.21837` · **primary:** `cs.CY` · **Submitted:** 2026-01-29
+- **Authors:** Xiaoshan Yu, Shangshang Yang, Ziwen Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21837 · **PDF:** https://arxiv.org/pdf/2601.21837v1
+
+## 18078. Scalable Linearized Laplace Approximation via Surrogate Neural Kernel
+
+- **ID:** `2601.21835` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Luis A. Ortega, Simón Rodríguez-Santana, Daniel Hernández-Lobato
+- **abs:** https://arxiv.org/abs/2601.21835 · **PDF:** https://arxiv.org/pdf/2601.21835v2
+
+## 18079. Goal-Driven Adaptive Sampling Strategies for Machine Learning Models Predicting Fields
+
+- **ID:** `2601.21832` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jigar Parekh, Philipp Bekemeyer
+- **abs:** https://arxiv.org/abs/2601.21832 · **PDF:** https://arxiv.org/pdf/2601.21832v1
+
+## 18080. Generative Modeling of Discrete Data Using Geometric Latent Subspaces
+
+- **ID:** `2601.21831` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** Daniel Gonzalez-Alvarado, Jonas Cassel, Stefania Petra, et al.
+- **abs:** https://arxiv.org/abs/2601.21831 · **PDF:** https://arxiv.org/pdf/2601.21831v1
+
+## 18081. Looking Beyond Accuracy: A Holistic Benchmark of ECG Foundation Models
+
+- **ID:** `2601.21830` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Francesca Filice, Edoardo De Rose, Simone Bartucci, et al.
+- **abs:** https://arxiv.org/abs/2601.21830 · **PDF:** https://arxiv.org/pdf/2601.21830v1
+
+## 18082. Mil-SCORE: Benchmarking Long-Context Geospatial Reasoning and Planning in Large Language Models
+
+- **ID:** `2601.21826` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Aadi Palnitkar, Mingyang Mao, Nicholas Waytowich, et al.
+- **abs:** https://arxiv.org/abs/2601.21826 · **PDF:** https://arxiv.org/pdf/2601.21826v3
+
+## 18083. DASH: Deterministic Attention Scheduling for High-throughput Reproducible LLM Training
+
+- **ID:** `2601.21824` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Xinwei Qiang, Hongmin Chen, Shixuan Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.21824 · **PDF:** https://arxiv.org/pdf/2601.21824v1
+
+## 18084. General Self-Prediction Enhancement for Spiking Neurons
+
+- **ID:** `2601.21823` · **primary:** `cs.NE` · **Submitted:** 2026-01-29
+- **Authors:** Zihan Huang, Zijie Xu, Yihan Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.21823 · **PDF:** https://arxiv.org/pdf/2601.21823v1
+
+## 18085. Indic-TunedLens: Interpreting Multilingual Models in Indian Languages
+
+- **ID:** `2602.15038` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Mihir Panchal, Deeksha Varshney, Mamta, et al.
+- **abs:** https://arxiv.org/abs/2602.15038 · **PDF:** https://arxiv.org/pdf/2602.15038v2
+
+## 18086. CORE:Toward Ubiquitous 6G Intelligence Through Collaborative Orchestration of Large Language Model Agents Over Hierarchical Edge
+
+- **ID:** `2601.21822` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Zitong Yu, Boquan Sun, Yang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.21822 · **PDF:** https://arxiv.org/pdf/2601.21822v1
+
+## 18087. A Judge-Aware Ranking Framework for Evaluating Large Language Models without Ground Truth
+
+- **ID:** `2601.21817` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** Mingyuan Xu, Xinzi Tan, Jiawei Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.21817 · **PDF:** https://arxiv.org/pdf/2601.21817v1
+
+## 18088. Nonparametric LLM Evaluation from Preference Data
+
+- **ID:** `2601.21816` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Dennis Frauen, Athiya Deviyani, Mihaela van der Schaar, et al.
+- **abs:** https://arxiv.org/abs/2601.21816 · **PDF:** https://arxiv.org/pdf/2601.21816v1
+
+## 18089. Moral Outrage Shapes Commitments Beyond Attention: Multimodal Moral Emotions on YouTube in Korea and the US
+
+- **ID:** `2601.21815` · **primary:** `cs.CY` · **Submitted:** 2026-01-29
+- **Authors:** Seongchan Park, Jaehong Kim, Hyeonseung Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.21815 · **PDF:** https://arxiv.org/pdf/2601.21815v2
+
+## 18090. A Decomposable Forward Process in Diffusion Models for Time-Series Forecasting
+
+- **ID:** `2601.21812` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** Francisco Caldas, Sahil Kumar, Cláudia Soares
+- **abs:** https://arxiv.org/abs/2601.21812 · **PDF:** https://arxiv.org/pdf/2601.21812v2
+
+## 18091. NL2LOGIC: AST-Guided Translation of Natural Language into First-Order Logic with Large Language Models
+
+- **ID:** `2602.13237` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Rizky Ramadhana Putra, Raihan Sultan Pasha Basuki, Yutong Cheng, et al.
+- **abs:** https://arxiv.org/abs/2602.13237 · **PDF:** https://arxiv.org/pdf/2602.13237v1
+
+## 18092. Distribution-Aware Reward Estimation for Test-Time Reinforcement Learning
+
+- **ID:** `2601.21804` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Bodong Du, Xuanqi Huang, Xiaomeng Li
+- **abs:** https://arxiv.org/abs/2601.21804 · **PDF:** https://arxiv.org/pdf/2601.21804v1
+
+## 18093. RAG-E: Quantifying Retriever-Generator Alignment and Failure Modes
+
+- **ID:** `2601.21803` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Korbinian Randl, Guido Rocchietti, Aron Henriksson, et al.
+- **abs:** https://arxiv.org/abs/2601.21803 · **PDF:** https://arxiv.org/pdf/2601.21803v1
+
+## 18094. A Unified XAI-LLM Approach for EndotrachealSuctioning Activity Recognition
+
+- **ID:** `2601.21802` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Hoang Khang Phan, Quang Vinh Dang, Noriyo Colley, et al.
+- **abs:** https://arxiv.org/abs/2601.21802 · **PDF:** https://arxiv.org/pdf/2601.21802v1
+
+## 18095. Lang2Act: Fine-Grained Visual Reasoning through Self-Emergent Linguistic Toolchains
+
+- **ID:** `2602.13235` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Yuqi Xiong, Chunyi Peng, Zhipeng Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.13235 · **PDF:** https://arxiv.org/pdf/2602.13235v1
+
+## 18096. BioAgent Bench: An AI Agent Evaluation Suite for Bioinformatics
+
+- **ID:** `2601.21800` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Dionizije Fa, Marko Čuljak, Bruno Pandža, et al.
+- **abs:** https://arxiv.org/abs/2601.21800 · **PDF:** https://arxiv.org/pdf/2601.21800v2
+
+## 18097. Enhancing Conversational Agents via Task-Oriented Adversarial Memory Adaptation
+
+- **ID:** `2601.21797` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Yimin Deng, Yuqing Fu, Derong Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.21797 · **PDF:** https://arxiv.org/pdf/2601.21797v1
+
+## 18098. KID: Knowledge-Injected Dual-Head Learning for Knowledge-Grounded Harmful Meme Detection
+
+- **ID:** `2601.21796` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Yaocong Li, Leihan Zhang, Le Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.21796 · **PDF:** https://arxiv.org/pdf/2601.21796v1
+
+## 18099. Effective LoRA Adapter Routing using Task Representations
+
+- **ID:** `2601.21795` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Akash Dhasade, Anne-Marie Kermarrec, Igor Pavlovic, et al.
+- **abs:** https://arxiv.org/abs/2601.21795 · **PDF:** https://arxiv.org/pdf/2601.21795v2
+
+## 18100. Knowledge Vector Weakening: Efficient Training-free Unlearning for Large Vision-Language Models
+
+- **ID:** `2601.21794` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yejin Kim, Dongjun Hwang, Sungmin Cha, et al.
+- **abs:** https://arxiv.org/abs/2601.21794 · **PDF:** https://arxiv.org/pdf/2601.21794v1
+
+## 18101. NetMamba+: A Framework of Pre-trained Models for Efficient and Accurate Network Traffic Classification
+
+- **ID:** `2601.21792` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Tongze Wang, Xiaohui Xie, Wenduo Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21792 · **PDF:** https://arxiv.org/pdf/2601.21792v1
+
+## 18102. ECSEL: Explainable Classification via Signomial Equation Learning
+
+- **ID:** `2601.21789` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Adia Lumadjeng, Ilker Birbil, Erman Acar
+- **abs:** https://arxiv.org/abs/2601.21789 · **PDF:** https://arxiv.org/pdf/2601.21789v1
+
+## 18103. Assessing the Business Process Modeling Competences of Large Language Models
+
+- **ID:** `2601.21787` · **primary:** `cs.SE` · **Submitted:** 2026-01-29
+- **Authors:** Chantale Lauer, Peter Pfeiffer, Alexander Rombach, et al.
+- **abs:** https://arxiv.org/abs/2601.21787 · **PDF:** https://arxiv.org/pdf/2601.21787v1
+
+## 18104. Synthetic-to-Real Domain Bridging for Single-View 3D Reconstruction of Ships for Maritime Monitoring
+
+- **ID:** `2601.21786` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Borja Carrillo-Perez, Felix Sattler, Angel Bueno Rodriguez, et al.
+- **abs:** https://arxiv.org/abs/2601.21786 · **PDF:** https://arxiv.org/pdf/2601.21786v1
+
+## 18105. Quantum LEGO Learning: A Modular Design Principle for Hybrid Artificial Intelligence
+
+- **ID:** `2601.21780` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jun Qi, Chao-Han Huck Yang, Pin-Yu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.21780 · **PDF:** https://arxiv.org/pdf/2601.21780v1
+
+## 18106. Error Amplification Limits ANN-to-SNN Conversion in Continuous Control
+
+- **ID:** `2601.21778` · **primary:** `cs.NE` · **Submitted:** 2026-01-29
+- **Authors:** Zijie Xu, Zihan Huang, Yiting Dong, et al.
+- **abs:** https://arxiv.org/abs/2601.21778 · **PDF:** https://arxiv.org/pdf/2601.21778v1
+
+## 18107. Differentiable Knapsack and Top-k Operators via Dynamic Programming
+
+- **ID:** `2601.21775` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Germain Vivier-Ardisson, Michaël E. Sander, Axel Parmentier, et al.
+- **abs:** https://arxiv.org/abs/2601.21775 · **PDF:** https://arxiv.org/pdf/2601.21775v1
+
+## 18108. Abstract Concept Modelling in Conceptual Spaces: A Study on Chess Strategies
+
+- **ID:** `2601.21771` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Hadi Banaee, Stephanie Lowry
+- **abs:** https://arxiv.org/abs/2601.21771 · **PDF:** https://arxiv.org/pdf/2601.21771v1
+
+## 18109. Zonkey: A Hierarchical Diffusion Language Model with Differentiable Tokenization and Probabilistic Attention
+
+- **ID:** `2601.21768` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Alon Rozental
+- **abs:** https://arxiv.org/abs/2601.21768 · **PDF:** https://arxiv.org/pdf/2601.21768v1
+
+## 18110. Evaluating ChatGPT on Medical Information Extraction Tasks: Performance, Explainability and Beyond
+
+- **ID:** `2601.21767` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Liz Li, Wei Zhu
+- **abs:** https://arxiv.org/abs/2601.21767 · **PDF:** https://arxiv.org/pdf/2601.21767v2
+
+## 18111. CoFrGeNet: Continued Fraction Architectures for Language Generation
+
+- **ID:** `2601.21766` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Amit Dhurandhar, Vijil Chenthamarakshan, Dennis Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.21766 · **PDF:** https://arxiv.org/pdf/2601.21766v2
+
+## 18112. Mean-field Variational Bayes for Sparse Probit Regression
+
+- **ID:** `2601.21765` · **primary:** `stat.CO` · **Submitted:** 2026-01-29
+- **Authors:** Augusto Fasano, Giovanni Rebaudo
+- **abs:** https://arxiv.org/abs/2601.21765 · **PDF:** https://arxiv.org/pdf/2601.21765v1
+
+## 18113. Zero-Shot Statistical Downscaling via Diffusion Posterior Sampling
+
+- **ID:** `2601.21760` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Ruian Tie, Wenbo Xiong, Zhengyu Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.21760 · **PDF:** https://arxiv.org/pdf/2601.21760v2
+
+## 18114. Influence Guided Sampling for Domain Adaptation of Text Retrievers
+
+- **ID:** `2601.21759` · **primary:** `cs.IR` · **Submitted:** 2026-01-29
+- **Authors:** Meet Doshi, Vishwajeet Kumar, Yulong Li, et al.
+- **abs:** https://arxiv.org/abs/2601.21759 · **PDF:** https://arxiv.org/pdf/2601.21759v1
+
+## 18115. EWSJF: An Adaptive Scheduler with Hybrid Partitioning for Mixed-Workload LLM Inference
+
+- **ID:** `2601.21758` · **primary:** `cs.DC` · **Submitted:** 2026-01-29
+- **Authors:** Bronislav Sidik, Chaya Levi, Joseph Kampeas
+- **abs:** https://arxiv.org/abs/2601.21758 · **PDF:** https://arxiv.org/pdf/2601.21758v1
+
+## 18116. Language-based Trial and Error Falls Behind in the Era of Experience
+
+- **ID:** `2601.21754` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Haoyu Wang, Guozheng Ma, Shugang Cui, et al.
+- **abs:** https://arxiv.org/abs/2601.21754 · **PDF:** https://arxiv.org/pdf/2601.21754v2
+
+## 18117. Investigating the Impact of Histopathological Foundation Models on Regressive Prediction of Homologous Recombination Deficiency
+
+- **ID:** `2602.00151` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Alexander Blezinger, Wolfgang Nejdl, Ming Tang
+- **abs:** https://arxiv.org/abs/2602.00151 · **PDF:** https://arxiv.org/pdf/2602.00151v2
+
+## 18118. FISMO: Fisher-Structured Momentum-Orthogonalized Optimizer
+
+- **ID:** `2601.21750` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Chenrui Xu, Wenjing Yan, Ying-Jun Angela Zhang
+- **abs:** https://arxiv.org/abs/2601.21750 · **PDF:** https://arxiv.org/pdf/2601.21750v1
+
+## 18119. Temporal Sepsis Modeling: a Relational and Explainable-by-Design Framework
+
+- **ID:** `2601.21747` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Vincent Lemaire, Nédra Meloulli, Pierre Jaquet
+- **abs:** https://arxiv.org/abs/2601.21747 · **PDF:** https://arxiv.org/pdf/2601.21747v3
+
+## 18120. Temporal Guidance for Large Language Models
+
+- **ID:** `2601.21744` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Hong-Kai Zheng, Piji Li
+- **abs:** https://arxiv.org/abs/2601.21744 · **PDF:** https://arxiv.org/pdf/2601.21744v1
+
+## 18121. Epistemic Context Learning: Building Trust the Right Way in LLM-Based Multi-Agent Systems
+
+- **ID:** `2601.21742` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Ruiwen Zhou, Maojia Song, Xiaobao Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.21742 · **PDF:** https://arxiv.org/pdf/2601.21742v1
+
+## 18122. Why Adam Works Better with $β_1 = β_2$: The Missing Gradient Scale Invariance Principle
+
+- **ID:** `2601.21739` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Alberto Fernández-Hernández, Cristian Pérez-Corral, Jose I. Mestre, et al.
+- **abs:** https://arxiv.org/abs/2601.21739 · **PDF:** https://arxiv.org/pdf/2601.21739v1
+
+## 18123. From Global to Granular: Revealing IQA Model Performance via Correlation Surface
+
+- **ID:** `2601.21738` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Baoliang Chen, Danni Huang, Hanwei Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.21738 · **PDF:** https://arxiv.org/pdf/2601.21738v1
+
+## 18124. Mixed-Precision Training and Compilation for RRAM-based Computing-in-Memory Accelerators
+
+- **ID:** `2601.21737` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Rebecca Pelke, Joel Klein, Jose Cubero-Cascante, et al.
+- **abs:** https://arxiv.org/abs/2601.21737 · **PDF:** https://arxiv.org/pdf/2601.21737v3
+
+## 18125. CE-GOCD: Central Entity-Guided Graph Optimization for Community Detection to Augment LLM Scientific Question Answering
+
+- **ID:** `2601.21733` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Jiayin Lan, Jiaqi Li, Baoxin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21733 · **PDF:** https://arxiv.org/pdf/2601.21733v1
+
+## 18126. Amortized Spectral Kernel Discovery via Prior-Data Fitted Network
+
+- **ID:** `2601.21731` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Kaustubh Sharma, Srijan Tiwari, Ojasva Nema, et al.
+- **abs:** https://arxiv.org/abs/2601.21731 · **PDF:** https://arxiv.org/pdf/2601.21731v1
+
+## 18127. DropoutTS: Sample-Adaptive Dropout for Robust Time Series Forecasting
+
+- **ID:** `2601.21726` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Siru Zhong, Yiqiu Liu, Zhiqing Cui, et al.
+- **abs:** https://arxiv.org/abs/2601.21726 · **PDF:** https://arxiv.org/pdf/2601.21726v1
+
+## 18128. Procedural Pretraining: Warming Up Language Models with Abstract Data
+
+- **ID:** `2601.21725` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Liangze Jiang, Zachary Shinnick, Anton van den Hengel, et al.
+- **abs:** https://arxiv.org/abs/2601.21725 · **PDF:** https://arxiv.org/pdf/2601.21725v1
+
+## 18129. Enhancing Language Models for Robust Greenwashing Detection
+
+- **ID:** `2601.21722` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Neil Heinrich Braun, Keane Ong, Rui Mao, et al.
+- **abs:** https://arxiv.org/abs/2601.21722 · **PDF:** https://arxiv.org/pdf/2601.21722v1
+
+## 18130. LoRA and Privacy: When Random Projections Help (and When They Don't)
+
+- **ID:** `2601.21719` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yaxi Hu, Johanna Düngler, Bernhard Schölkopf, et al.
+- **abs:** https://arxiv.org/abs/2601.21719 · **PDF:** https://arxiv.org/pdf/2601.21719v1
+
+## 18131. When does predictive inverse dynamics outperform behavior cloning?
+
+- **ID:** `2601.21718` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Lukas Schäfer, Pallavi Choudhury, Abdelhak Lemkhenter, et al.
+- **abs:** https://arxiv.org/abs/2601.21718 · **PDF:** https://arxiv.org/pdf/2601.21718v1
+
+## 18132. On sample complexity for covariance estimation via the unadjusted Langevin algorithm
+
+- **ID:** `2601.21717` · **primary:** `math.PR` · **Submitted:** 2026-01-29
+- **Authors:** Shogo Nakakita
+- **abs:** https://arxiv.org/abs/2601.21717 · **PDF:** https://arxiv.org/pdf/2601.21717v2
+
+## 18133. DreamActor-M2: Universal Character Image Animation via Spatiotemporal In-Context Learning
+
+- **ID:** `2601.21716` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Mingshuang Luo, Shuang Liang, Zhengkun Rong, et al.
+- **abs:** https://arxiv.org/abs/2601.21716 · **PDF:** https://arxiv.org/pdf/2601.21716v1
+
+## 18134. E-mem: Multi-agent based Episodic Context Reconstruction for LLM Agent Memory
+
+- **ID:** `2601.21714` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Kaixiang Wang, Yidan Lin, Jiong Lou, et al.
+- **abs:** https://arxiv.org/abs/2601.21714 · **PDF:** https://arxiv.org/pdf/2601.21714v1
+
+## 18135. Disentangling perception and reasoning for improving data efficiency in learning cloth manipulation without demonstrations
+
+- **ID:** `2601.21713` · **primary:** `cs.RO` · **Submitted:** 2026-01-29
+- **Authors:** Donatien Delehelle, Fei Chen, Darwin Caldwell
+- **abs:** https://arxiv.org/abs/2601.21713 · **PDF:** https://arxiv.org/pdf/2601.21713v1
+
+## 18136. TACLer: Tailored Curriculum Reinforcement Learning for Efficient Reasoning
+
+- **ID:** `2601.21711` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Huiyuan Lai, Malvina Nissim
+- **abs:** https://arxiv.org/abs/2601.21711 · **PDF:** https://arxiv.org/pdf/2601.21711v1
+
+## 18137. Why Attention Patterns Exist: A Unifying Temporal Perspective Analysis
+
+- **ID:** `2601.21709` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Qingyue Yang, Jie Wang, Xing Li, et al.
+- **abs:** https://arxiv.org/abs/2601.21709 · **PDF:** https://arxiv.org/pdf/2601.21709v1
+
+## 18138. FBS: Modeling Native Parallel Reading inside a Transformer
+
+- **ID:** `2601.21708` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Tongxi Wang
+- **abs:** https://arxiv.org/abs/2601.21708 · **PDF:** https://arxiv.org/pdf/2601.21708v1
+
+## 18139. SmartMeterFM: Unifying Smart Meter Data Generative Tasks Using Flow Matching Models
+
+- **ID:** `2601.21706` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Nan Lin, Yanbo Wang, Jacco Heres, et al.
+- **abs:** https://arxiv.org/abs/2601.21706 · **PDF:** https://arxiv.org/pdf/2601.21706v2
+
+## 18140. Beyond Forgetting: Machine Unlearning Elicits Controllable Side Behaviors and Capabilities
+
+- **ID:** `2601.21702` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Tien Dang, The-Hai Nguyen, Dinh Mai Phuong, et al.
+- **abs:** https://arxiv.org/abs/2601.21702 · **PDF:** https://arxiv.org/pdf/2601.21702v2
+
+## 18141. Toward Culturally Aligned LLMs through Ontology-Guided Multi-Agent Reasoning
+
+- **ID:** `2601.21700` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Wonduk Seo, Wonseok Choi, Junseo Koh, et al.
+- **abs:** https://arxiv.org/abs/2601.21700 · **PDF:** https://arxiv.org/pdf/2601.21700v2
+
+## 18142. Can David Beat Goliath? On Multi-Hop Reasoning with Resource-Constrained Agents
+
+- **ID:** `2601.21699` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Hojae Han, Heeyun Jung, Jongyoon Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.21699 · **PDF:** https://arxiv.org/pdf/2601.21699v1
+
+## 18143. Curriculum Learning for LLM Pretraining: An Analysis of Learning Dynamics
+
+- **ID:** `2601.21698` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Mohamed Elgaar, Hadi Amiri
+- **abs:** https://arxiv.org/abs/2601.21698 · **PDF:** https://arxiv.org/pdf/2601.21698v1
+
+## 18144. Independent Component Discovery in Temporal Count Data
+
+- **ID:** `2601.21696` · **primary:** `stat.ME` · **Submitted:** 2026-01-29
+- **Authors:** Alexandre Chaussard, Anna Bonnet, Sylvain Le Corff
+- **abs:** https://arxiv.org/abs/2601.21696 · **PDF:** https://arxiv.org/pdf/2601.21696v1
+
+## 18145. TCAP: Tri-Component Attention Profiling for Unsupervised Backdoor Detection in MLLM Fine-Tuning
+
+- **ID:** `2601.21692` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Mingzu Liu, Hao Fang, Runmin Cong
+- **abs:** https://arxiv.org/abs/2601.21692 · **PDF:** https://arxiv.org/pdf/2601.21692v1
+
+## 18146. A Unified Generalization Framework for Model Merging: Trade-offs, Non-Linearity, and Scaling Laws
+
+- **ID:** `2601.21690` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Qinglun Li, Anke Tang, Miao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.21690 · **PDF:** https://arxiv.org/pdf/2601.21690v2
+
+## 18147. XFACTORS: Disentangled Information Bottleneck via Contrastive Supervision
+
+- **ID:** `2601.21688` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Alexandre Myara, Nicolas Bourriez, Thomas Boyer, et al.
+- **abs:** https://arxiv.org/abs/2601.21688 · **PDF:** https://arxiv.org/pdf/2601.21688v1
+
+## 18148. Don't be so Stief! Learning KV Cache low-rank approximation over the Stiefel manifold
+
+- **ID:** `2601.21686` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Luca Benfenati, Matteo Risso, Andrea Vannozzi, et al.
+- **abs:** https://arxiv.org/abs/2601.21686 · **PDF:** https://arxiv.org/pdf/2601.21686v1
+
+## 18149. Do Not Waste Your Rollouts: Recycling Search Experience for Efficient Test-Time Scaling
+
+- **ID:** `2601.21684` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Xinglin Wang, Jiayi Shi, Shaoxiong Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.21684 · **PDF:** https://arxiv.org/pdf/2601.21684v1
+
+## 18150. Can Local Learning Match Self-Supervised Backpropagation?
+
+- **ID:** `2601.21683` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Wu S. Zihan, Ariane Delrocq, Wulfram Gerstner, et al.
+- **abs:** https://arxiv.org/abs/2601.21683 · **PDF:** https://arxiv.org/pdf/2601.21683v1
+
+## 18151. FIT: Defying Catastrophic Forgetting in Continual LLM Unlearning
+
+- **ID:** `2601.21682` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Xiaoyu Xu, Minxin Du, Kun Fang, et al.
+- **abs:** https://arxiv.org/abs/2601.21682 · **PDF:** https://arxiv.org/pdf/2601.21682v1
+
+## 18152. LLM4Fluid: Large Language Models as Generalizable Neural Solvers for Fluid Dynamics
+
+- **ID:** `2601.21681` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Qisong Xiao, Xinhai Chen, Qinglin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21681 · **PDF:** https://arxiv.org/pdf/2601.21681v1
+
+## 18153. Scale-Dependent Semantic Dynamics Revealed by Allan Deviation
+
+- **ID:** `2601.21678` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Debayan Dasgupta
+- **abs:** https://arxiv.org/abs/2601.21678 · **PDF:** https://arxiv.org/pdf/2601.21678v1
+
+## 18154. Neural Signals Generate Clinical Notes in the Wild
+
+- **ID:** `2601.22197` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jathurshan Pradeepkumar, Zheng Chen, Jimeng Sun
+- **abs:** https://arxiv.org/abs/2601.22197 · **PDF:** https://arxiv.org/pdf/2601.22197v2
+
+## 18155. When Gradient Optimization Is Not Enough: $\dagger$ Dispersive and Anchoring Geometric Regularizer for Multimodal Learning
+
+- **ID:** `2601.21670` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Zixuan Xia, Hao Wang, Pengcheng Weng, et al.
+- **abs:** https://arxiv.org/abs/2601.21670 · **PDF:** https://arxiv.org/pdf/2601.21670v1
+
+## 18156. Expected Return Causes Outcome-Level Mode Collapse in Reinforcement Learning and How to Fix It with Inverse Probability Scaling
+
+- **ID:** `2601.21669` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Abhijeet Sinha, Sundari Elango, Dianbo Liu
+- **abs:** https://arxiv.org/abs/2601.21669 · **PDF:** https://arxiv.org/pdf/2601.21669v1
+
+## 18157. SONIC-O1: A Real-World Benchmark for Evaluating Multimodal Large Language Models on Audio-Video Understanding
+
+- **ID:** `2601.21666` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Ahmed Y. Radwan, Christos Emmanouilidis, Hina Tabassum, et al.
+- **abs:** https://arxiv.org/abs/2601.21666 · **PDF:** https://arxiv.org/pdf/2601.21666v1
+
+## 18158. AdaptBPE: From General Purpose to Specialized Tokenizers
+
+- **ID:** `2601.21665` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Vijini Liyanage, François Yvon
+- **abs:** https://arxiv.org/abs/2601.21665 · **PDF:** https://arxiv.org/pdf/2601.21665v1
+
+## 18159. SENDAI: A Hierarchical Sparse-measurement, EfficieNt Data AssImilation Framework
+
+- **ID:** `2601.21664` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Xingyue Zhang, Yuxuan Bao, Mars Liyao Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.21664 · **PDF:** https://arxiv.org/pdf/2601.21664v1
+
+## 18160. Epistemic Uncertainty Quantification for Pre-trained VLMs via Riemannian Flow Matching
+
+- **ID:** `2601.21662` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Li Ju, Mayank Nautiyal, Andreas Hellander, et al.
+- **abs:** https://arxiv.org/abs/2601.21662 · **PDF:** https://arxiv.org/pdf/2601.21662v1
+
+## 18161. TabClustPFN: A Prior-Fitted Network for Tabular Data Clustering
+
+- **ID:** `2601.21656` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Tianqi Zhao, Guanyang Wang, Yan Shuo Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.21656 · **PDF:** https://arxiv.org/pdf/2601.21656v2
+
+## 18162. Reversible Diffusion Decoding for Diffusion Language Models
+
+- **ID:** `2602.00150` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Xinyun Wang, Min Zhang, Sen Cui, et al.
+- **abs:** https://arxiv.org/abs/2602.00150 · **PDF:** https://arxiv.org/pdf/2602.00150v1
+
+## 18163. ScholarGym: Benchmarking Large Language Model Capabilities in the Information-Gathering Stage of Deep Research
+
+- **ID:** `2601.21654` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Hao Shen, Hang Yang, Zhouhong Gu, et al.
+- **abs:** https://arxiv.org/abs/2601.21654 · **PDF:** https://arxiv.org/pdf/2601.21654v3
+
+## 18164. Gauge-invariant representation holonomy
+
+- **ID:** `2601.21653` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Vasileios Sevetlidis, George Pavlidis
+- **abs:** https://arxiv.org/abs/2601.21653 · **PDF:** https://arxiv.org/pdf/2601.21653v1
+
+## 18165. When Life Gives You AI, Will You Turn It Into A Market for Lemons? Understanding How Information Asymmetries About AI System Capabilities Affect Market Outcomes and Adoption
+
+- **ID:** `2601.21650` · **primary:** `cs.HC` · **Submitted:** 2026-01-29
+- **Authors:** Alexander Erlei, Federico Cau, Radoslav Georgiev, et al.
+- **abs:** https://arxiv.org/abs/2601.21650 · **PDF:** https://arxiv.org/pdf/2601.21650v1
+
+## 18166. SWE-Spot: Building Small Repo-Experts with Repository-Centric Learning
+
+- **ID:** `2601.21649` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jinjun Peng, Magnus Saebo, Tianjun Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.21649 · **PDF:** https://arxiv.org/pdf/2601.21649v1
+
+## 18167. ILRR: Inference-Time Steering Method for Masked Diffusion Language Models
+
+- **ID:** `2601.21647` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Eden Avrahami, Eliya Nachmani
+- **abs:** https://arxiv.org/abs/2601.21647 · **PDF:** https://arxiv.org/pdf/2601.21647v1
+
+## 18168. Identifiable Equivariant Networks are Layerwise Equivariant
+
+- **ID:** `2601.21645` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Vahid Shahverdi, Giovanni Luca Marchetti, Georg Bökman, et al.
+- **abs:** https://arxiv.org/abs/2601.21645 · **PDF:** https://arxiv.org/pdf/2601.21645v1
+
+## 18169. Seg-MoE: Multi-Resolution Segment-wise Mixture-of-Experts for Time Series Forecasting Transformers
+
+- **ID:** `2601.21641` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Evandro S. Ortigossa, Eran Segal
+- **abs:** https://arxiv.org/abs/2601.21641 · **PDF:** https://arxiv.org/pdf/2601.21641v2
+
+## 18170. Generative Design of Ship Propellers using Conditional Flow Matching
+
+- **ID:** `2601.21637` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Patrick Kruger, Rafael Diaz, Simon Hauschulz, et al.
+- **abs:** https://arxiv.org/abs/2601.21637 · **PDF:** https://arxiv.org/pdf/2601.21637v1
+
+## 18171. Sampling-Free Privacy Accounting for Matrix Mechanisms under Random Allocation
+
+- **ID:** `2601.21636` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jan Schuchardt, Nikita Kalinin
+- **abs:** https://arxiv.org/abs/2601.21636 · **PDF:** https://arxiv.org/pdf/2601.21636v2
+
+## 18172. Reinforcement Learning for Adaptive Composition of Quantum Circuit Optimisation Passes
+
+- **ID:** `2601.21629` · **primary:** `quant-ph` · **Submitted:** 2026-01-29
+- **Authors:** Daniel Mills, Ifan Williams, Jacob Swain, et al.
+- **abs:** https://arxiv.org/abs/2601.21629 · **PDF:** https://arxiv.org/pdf/2601.21629v1
+
+## 18173. Noise as a Probe: Membership Inference Attacks on Diffusion Models Leveraging Initial Noise
+
+- **ID:** `2601.21628` · **primary:** `cs.CR` · **Submitted:** 2026-01-29
+- **Authors:** Puwei Lian, Yujun Cai, Songze Li, et al.
+- **abs:** https://arxiv.org/abs/2601.21628 · **PDF:** https://arxiv.org/pdf/2601.21628v1
+
+## 18174. HeRo-Q: A General Framework for Stable Low Bit Quantization via Hessian Conditioning
+
+- **ID:** `2601.21626` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jinhao Zhang Yunquan Zhang, Zicheng yan, Boyang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.21626 · **PDF:** https://arxiv.org/pdf/2601.21626v1
+
+## 18175. Training Memory in Deep Neural Networks: Mechanisms, Evidence, and Measurement Gaps
+
+- **ID:** `2601.21624` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Vasileios Sevetlidis, George Pavlidis
+- **abs:** https://arxiv.org/abs/2601.21624 · **PDF:** https://arxiv.org/pdf/2601.21624v1
+
+## 18176. LAMP: Look-Ahead Mixed-Precision Inference of Large Language Models
+
+- **ID:** `2601.21623` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Stanislav Budzinskiy, Marian Gloser, Tolunay Yilmaz, et al.
+- **abs:** https://arxiv.org/abs/2601.21623 · **PDF:** https://arxiv.org/pdf/2601.21623v1
+
+## 18177. Breaking the Overscaling Curse: Thinking Parallelism Before Parallel Thinking
+
+- **ID:** `2601.21619` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yiming Wang, Zhuosheng Zhang, Rui Wang
+- **abs:** https://arxiv.org/abs/2601.21619 · **PDF:** https://arxiv.org/pdf/2601.21619v1
+
+## 18178. Semantic Content Determines Algorithmic Performance
+
+- **ID:** `2601.21618` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Martiño Ríos-García, Nawaf Alampara, Kevin Maik Jablonka
+- **abs:** https://arxiv.org/abs/2601.21618 · **PDF:** https://arxiv.org/pdf/2601.21618v1
+
+## 18179. Beyond Parameter Finetuning: Test-Time Representation Refinement for Node Classification
+
+- **ID:** `2601.21615` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jiaxin Zhang, Yiqi Wang, Siwei Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21615 · **PDF:** https://arxiv.org/pdf/2601.21615v1
+
+## 18180. Multitask Learning for Earth Observation Data Classification with Hybrid Quantum Network
+
+- **ID:** `2601.22195` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Fan Fan, Yilei Shi, Tobias Guggemos, et al.
+- **abs:** https://arxiv.org/abs/2601.22195 · **PDF:** https://arxiv.org/pdf/2601.22195v1
+
+## 18181. Representation-Regularized Convolutional Audio Transformer for Audio Understanding
+
+- **ID:** `2601.21612` · **primary:** `eess.AS` · **Submitted:** 2026-01-29
+- **Authors:** Bing Han, Chushu Zhou, Yifan Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.21612 · **PDF:** https://arxiv.org/pdf/2601.21612v1
+
+## 18182. Thinking Broad, Acting Fast: Latent Reasoning Distillation from Multi-Perspective Chain-of-Thought for E-Commerce Relevance
+
+- **ID:** `2601.21611` · **primary:** `cs.IR` · **Submitted:** 2026-01-29
+- **Authors:** Baopu Qiu, Hao Chen, Yuanrong Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.21611 · **PDF:** https://arxiv.org/pdf/2601.21611v1
+
+## 18183. RecNet: Self-Evolving Preference Propagation for Agentic Recommender Systems
+
+- **ID:** `2601.21609` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Bingqian Li, Xiaolei Wang, Junyi Li, et al.
+- **abs:** https://arxiv.org/abs/2601.21609 · **PDF:** https://arxiv.org/pdf/2601.21609v1
+
+## 18184. Search-Based Risk Feature Discovery in Document Structure Spaces under a Constrained Budget
+
+- **ID:** `2601.21608` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Saisubramaniam Gopalakrishnan, Harikrishnan P M, Dagnachew Birru
+- **abs:** https://arxiv.org/abs/2601.21608 · **PDF:** https://arxiv.org/pdf/2601.21608v1
+
+## 18185. Age Matters: Analyzing Age-Related Discussions in App Reviews
+
+- **ID:** `2601.21605` · **primary:** `cs.SE` · **Submitted:** 2026-01-29
+- **Authors:** Shashiwadana Nirmania, Garima Sharma, Hourieh Khalajzadeh, et al.
+- **abs:** https://arxiv.org/abs/2601.21605 · **PDF:** https://arxiv.org/pdf/2601.21605v1
+
+## 18186. Dynamics Reveals Structure: Challenging the Linear Propagation Assumption
+
+- **ID:** `2601.21601` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Hoyeon Chang, Bálint Mucsányi, Seong Joon Oh
+- **abs:** https://arxiv.org/abs/2601.21601 · **PDF:** https://arxiv.org/pdf/2601.21601v1
+
+## 18187. CORE: Collaborative Reasoning via Cross Teaching
+
+- **ID:** `2601.21600` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Kshitij Mishra, Mirat Aubakirov, Martin Takac, et al.
+- **abs:** https://arxiv.org/abs/2601.21600 · **PDF:** https://arxiv.org/pdf/2601.21600v1
+
+## 18188. Beyond Imitation: Reinforcement Learning for Active Latent Planning
+
+- **ID:** `2601.21598` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Zhi Zheng, Wee Sun Lee
+- **abs:** https://arxiv.org/abs/2601.21598 · **PDF:** https://arxiv.org/pdf/2601.21598v1
+
+## 18189. Scalable Power Sampling: Unlocking Efficient, Training-Free Reasoning for LLMs via Distribution Sharpening
+
+- **ID:** `2601.21590` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Xiaotong Ji, Rasul Tutunov, Matthieu Zimmer, et al.
+- **abs:** https://arxiv.org/abs/2601.21590 · **PDF:** https://arxiv.org/pdf/2601.21590v1
+
+## 18190. Heterogeneity-Aware Knowledge Sharing for Graph Federated Learning
+
+- **ID:** `2601.21589` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Wentao Yu, Sheng Wan, Shuo Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.21589 · **PDF:** https://arxiv.org/pdf/2601.21589v1
+
+## 18191. Stay in Character, Stay Safe: Dual-Cycle Adversarial Self-Evolution for Safety Role-Playing Agents
+
+- **ID:** `2602.13234` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Mingyang Liao, Yichen Wan, shuchen wu, et al.
+- **abs:** https://arxiv.org/abs/2602.13234 · **PDF:** https://arxiv.org/pdf/2602.13234v1
+
+## 18192. Language Models as Artificial Learners: Investigating Crosslinguistic Influence
+
+- **ID:** `2601.21587` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Abderrahmane Issam, Yusuf Can Semerci, Jan Scholtes, et al.
+- **abs:** https://arxiv.org/abs/2601.21587 · **PDF:** https://arxiv.org/pdf/2601.21587v1
+
+## 18193. CORDS: Continuous Representations of Discrete Structures
+
+- **ID:** `2601.21583` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Tin Hadži Veljković, Erik Bekkers, Michael Tiemann, et al.
+- **abs:** https://arxiv.org/abs/2601.21583 · **PDF:** https://arxiv.org/pdf/2601.21583v1
+
+## 18194. Depth-Recurrent Attention Mixtures: Giving Latent Reasoning the Attention it Deserves
+
+- **ID:** `2601.21582` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Jonas Knupp, Jan Hendrik Metzen, Jeremias Bohn, et al.
+- **abs:** https://arxiv.org/abs/2601.21582 · **PDF:** https://arxiv.org/pdf/2601.21582v1
+
+## 18195. Evaluating Prediction Uncertainty Estimates from BatchEnsemble
+
+- **ID:** `2601.21581` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Morten Blørstad, Herman Jangsett Mostein, Nello Blaser, et al.
+- **abs:** https://arxiv.org/abs/2601.21581 · **PDF:** https://arxiv.org/pdf/2601.21581v1
+
+## 18196. Practical Evaluation of Quantum Kernel Methods for Radar Micro-Doppler Classification on Noisy Intermediate-Scale Quantum (NISQ) Hardware
+
+- **ID:** `2601.22194` · **primary:** `quant-ph` · **Submitted:** 2026-01-29
+- **Authors:** Vikas Agnihotri, Jasleen Kaur, Sarvagya Kaushik
+- **abs:** https://arxiv.org/abs/2601.22194 · **PDF:** https://arxiv.org/pdf/2601.22194v1
+
+## 18197. KromHC: Manifold-Constrained Hyper-Connections with Kronecker-Product Residual Matrices
+
+- **ID:** `2601.21579` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Wuyang Zhou, Yuxuan Gu, Giorgos Iacovides, et al.
+- **abs:** https://arxiv.org/abs/2601.21579 · **PDF:** https://arxiv.org/pdf/2601.21579v1
+
+## 18198. Learning the Mechanism of Catastrophic Forgetting: A Perspective from Gradient Similarity
+
+- **ID:** `2601.21577` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Mutian Yang, Zisen Zhan, Yutong Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.21577 · **PDF:** https://arxiv.org/pdf/2601.21577v1
+
+## 18199. Chain Of Thought Compression: A Theoritical Analysis
+
+- **ID:** `2601.21576` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Juncai Li, Ru Li, Yuxiang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.21576 · **PDF:** https://arxiv.org/pdf/2601.21576v1
+
+## 18200. Learning Physics-Grounded 4D Dynamics with Neural Gaussian Force Fields
+
+- **ID:** `2602.00148` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Shiqian Li, Ruihong Shen, Junfeng Ni, et al.
+- **abs:** https://arxiv.org/abs/2602.00148 · **PDF:** https://arxiv.org/pdf/2602.00148v2
+
+## 18201. Signal-Adaptive Trust Regions for Gradient-Free Optimization of Recurrent Spiking Neural Networks
+
+- **ID:** `2601.21572` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Jinhao Li, Yuhao Sun, Zhiyuan Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.21572 · **PDF:** https://arxiv.org/pdf/2601.21572v1
+
+## 18202. Shaping capabilities with token-level data filtering
+
+- **ID:** `2601.21571` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Neil Rathi, Alec Radford
+- **abs:** https://arxiv.org/abs/2601.21571 · **PDF:** https://arxiv.org/pdf/2601.21571v2
+
+## 18203. EmboCoach-Bench: Benchmarking AI Agents on Developing Embodied Robots
+
+- **ID:** `2601.21570` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Zixing Lei, Genjia Liu, Yuanshuo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.21570 · **PDF:** https://arxiv.org/pdf/2601.21570v1
+
+## 18204. Bridging Functional and Representational Similarity via Usable Information
+
+- **ID:** `2601.21568` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Antonio Almudévar, Alfonso Ortega
+- **abs:** https://arxiv.org/abs/2601.21568 · **PDF:** https://arxiv.org/pdf/2601.21568v1
+
+## 18205. FlexCausal: Flexible Causal Disentanglement via Structural Flow Priors and Manifold-Aware Interventions
+
+- **ID:** `2601.21567` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yutao Jin, Yuang Tao, Junyong Zhai
+- **abs:** https://arxiv.org/abs/2601.21567 · **PDF:** https://arxiv.org/pdf/2601.21567v1
+
+## 18206. Multi-objective Integer Linear Programming approach for Automatic Software Cognitive Complexity Reduction
+
+- **ID:** `2601.21565` · **primary:** `cs.SE` · **Submitted:** 2026-01-29
+- **Authors:** Adriana Novoa-Hurtado, Rubén Saborido, Francisco Chicano, et al.
+- **abs:** https://arxiv.org/abs/2601.21565 · **PDF:** https://arxiv.org/pdf/2601.21565v1
+
+## 18207. Representation Unlearning: Forgetting through Information Compression
+
+- **ID:** `2601.21564` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Antonio Almudévar, Alfonso Ortega
+- **abs:** https://arxiv.org/abs/2601.21564 · **PDF:** https://arxiv.org/pdf/2601.21564v1
+
+## 18208. SAL: Selective Adaptive Learning for Backpropagation-Free Training with Sparsification
+
+- **ID:** `2601.21561` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Fanping Liu, Hua Yang, Jiasi Zou
+- **abs:** https://arxiv.org/abs/2601.21561 · **PDF:** https://arxiv.org/pdf/2601.21561v1
+
+## 18209. HistoPrism: Unlocking Functional Pathway Analysis from Pan-Cancer Histology via Gene Expression Prediction
+
+- **ID:** `2601.21560` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Susu Hu, Qinghe Zeng, Nithya Bhasker, et al.
+- **abs:** https://arxiv.org/abs/2601.21560 · **PDF:** https://arxiv.org/pdf/2601.21560v3
+
+## 18210. ASTRA: Automated Synthesis of agentic Trajectories and Reinforcement Arenas
+
+- **ID:** `2601.21558` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Xiaoyu Tian, Haotian Wang, Shuaiting Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.21558 · **PDF:** https://arxiv.org/pdf/2601.21558v2
+
+## 18211. Meta Context Engineering via Agentic Skill Evolution
+
+- **ID:** `2601.21557` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Haoran Ye, Xuning He, Vincent Arak, et al.
+- **abs:** https://arxiv.org/abs/2601.21557 · **PDF:** https://arxiv.org/pdf/2601.21557v2
+
+## 18212. Note2Chat: Improving LLMs for Multi-Turn Clinical History Taking Using Medical Notes
+
+- **ID:** `2601.21551` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Yang Zhou, Zhenting Sheng, Mingrui Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.21551 · **PDF:** https://arxiv.org/pdf/2601.21551v1
+
+## 18213. Training slow silicon neurons to control extremely fast robots with spiking reinforcement learning
+
+- **ID:** `2601.21548` · **primary:** `cs.RO` · **Submitted:** 2026-01-29
+- **Authors:** Irene Ambrosini, Ingo Blakowski, Dmitrii Zendrikov, et al.
+- **abs:** https://arxiv.org/abs/2601.21548 · **PDF:** https://arxiv.org/pdf/2601.21548v1
+
+## 18214. Multi-Modal Time Series Prediction via Mixture of Modulated Experts
+
+- **ID:** `2601.21547` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Lige Zhang, Ali Maatouk, Jialin Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.21547 · **PDF:** https://arxiv.org/pdf/2601.21547v1
+
+## 18215. ShardMemo: Masked MoE Routing for Sharded Agentic LLM Memory
+
+- **ID:** `2601.21545` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Yang Zhao, Chengxiao Dai, Yue Xiu, et al.
+- **abs:** https://arxiv.org/abs/2601.21545 · **PDF:** https://arxiv.org/pdf/2601.21545v1
+
+## 18216. inversedMixup: Data Augmentation via Inverting Mixed Embeddings
+
+- **ID:** `2601.21543` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Fanshuang Kong, Richong Zhang, Qiyu Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.21543 · **PDF:** https://arxiv.org/pdf/2601.21543v2
+
+## 18217. Bi-Anchor Interpolation Solver for Accelerating Generative Modeling
+
+- **ID:** `2601.21542` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Hongxu Chen, Hongxiang Li, Zhen Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21542 · **PDF:** https://arxiv.org/pdf/2601.21542v2
+
+## 18218. ARGORA: Orchestrated Argumentation for Causally Grounded LLM Reasoning and Decision Making
+
+- **ID:** `2601.21533` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Youngjin Jin, Hanna Kim, Kwanwoo Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.21533 · **PDF:** https://arxiv.org/pdf/2601.21533v1
+
+## 18219. On the Adversarial Robustness of Large Vision-Language Models under Visual Token Compression
+
+- **ID:** `2601.21531` · **primary:** `cs.CR` · **Submitted:** 2026-01-29
+- **Authors:** Xinwei Zhang, Hangcheng Liu, Li Bai, et al.
+- **abs:** https://arxiv.org/abs/2601.21531 · **PDF:** https://arxiv.org/pdf/2601.21531v1
+
+## 18220. Fast and Geometrically Grounded Lorentz Neural Networks
+
+- **ID:** `2601.21529` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Robert van der Klis, Ricardo Chávez Torres, Max van Spengler, et al.
+- **abs:** https://arxiv.org/abs/2601.21529 · **PDF:** https://arxiv.org/pdf/2601.21529v1
+
+## 18221. Sustainable Materials Discovery in the Era of Artificial Intelligence
+
+- **ID:** `2601.21527` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-29
+- **Authors:** Sajid Mannan, Rupert J. Myers, Rohit Batra, et al.
+- **abs:** https://arxiv.org/abs/2601.21527 · **PDF:** https://arxiv.org/pdf/2601.21527v2
+
+## 18222. KAPSO: A Knowledge-grounded framework for Autonomous Program Synthesis and Optimization
+
+- **ID:** `2601.21526` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Alireza Nadafian, Alireza Mohammadshahi, Majid Yazdani
+- **abs:** https://arxiv.org/abs/2601.21526 · **PDF:** https://arxiv.org/pdf/2601.21526v2
+
+## 18223. LMK > CLS: Landmark Pooling for Dense Embeddings
+
+- **ID:** `2601.21525` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Meet Doshi, Aashka Trivedi, Vishwajeet Kumar, et al.
+- **abs:** https://arxiv.org/abs/2601.21525 · **PDF:** https://arxiv.org/pdf/2601.21525v1
+
+## 18224. Explicit Credit Assignment through Local Rewards and Dependence Graphs in Multi-Agent Reinforcement Learning
+
+- **ID:** `2601.21523` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Bang Giang Le, Viet Cuong Ta
+- **abs:** https://arxiv.org/abs/2601.21523 · **PDF:** https://arxiv.org/pdf/2601.21523v1
+
+## 18225. More Bang for the Buck: Improving the Inference of Large Language Models at a Fixed Budget using Reset and Discard (ReD)
+
+- **ID:** `2601.21522` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Sagi Meir, Tommer D. Keidar, Noam Levi, et al.
+- **abs:** https://arxiv.org/abs/2601.21522 · **PDF:** https://arxiv.org/pdf/2601.21522v1
+
+## 18226. A Unified SPD Token Transformer Framework for EEG Classification: Systematic Comparison of Geometric Embeddings
+
+- **ID:** `2601.21521` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Chi-Sheng Chen, En-Jui Kuo, Guan-Ying Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.21521 · **PDF:** https://arxiv.org/pdf/2601.21521v1
+
+## 18227. Cascaded Transfer: Learning Many Tasks under Budget Constraints
+
+- **ID:** `2601.21513` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Eloi Campagne, Yvenn Amara-Ouali, Yannig Goude, et al.
+- **abs:** https://arxiv.org/abs/2601.21513 · **PDF:** https://arxiv.org/pdf/2601.21513v1
+
+## 18228. MURAD: A Large-Scale Multi-Domain Unified Reverse Arabic Dictionary Dataset
+
+- **ID:** `2601.21512` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Serry Sibaee, Yasser Alhabashi, Nadia Sibai, et al.
+- **abs:** https://arxiv.org/abs/2601.21512 · **PDF:** https://arxiv.org/pdf/2601.21512v1
+
+## 18229. LLaMEA-SAGE: Guiding Automated Algorithm Design with Structural Feedback from Explainable AI
+
+- **ID:** `2601.21511` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Niki van Stein, Anna V. Kononova, Lars Kotthoff, et al.
+- **abs:** https://arxiv.org/abs/2601.21511 · **PDF:** https://arxiv.org/pdf/2601.21511v1
+
+## 18230. The Effectiveness of Style Vectors for Steering Large Language Models: A Human Evaluation
+
+- **ID:** `2601.21505` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Diaoulé Diallo, Katharina Dworatzyk, Sophie Jentzsch, et al.
+- **abs:** https://arxiv.org/abs/2601.21505 · **PDF:** https://arxiv.org/pdf/2601.21505v1
+
+## 18231. MAR: Efficient Large Language Models via Module-aware Architecture Refinement
+
+- **ID:** `2601.21503` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Junhong Cai, Guiqin Wang, Kejie Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.21503 · **PDF:** https://arxiv.org/pdf/2601.21503v1
+
+## 18232. Task-Awareness Improves LLM Generations and Uncertainty
+
+- **ID:** `2601.21500` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Tim Tomov, Dominik Fuchsgruber, Stephan Günnemann
+- **abs:** https://arxiv.org/abs/2601.21500 · **PDF:** https://arxiv.org/pdf/2601.21500v1
+
+## 18233. SimGraph: A Unified Framework for Scene Graph-Based Image Generation and Editing
+
+- **ID:** `2601.21498` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Thanh-Nhan Vo, Trong-Thuan Nguyen, Tam V. Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.21498 · **PDF:** https://arxiv.org/pdf/2601.21498v1
+
+## 18234. The Path of Least Resistance: Guiding LLM Reasoning Trajectories with Prefix Consensus
+
+- **ID:** `2601.21494` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Ishan Jindal, Sai Prashanth Akuthota, Jayant Taneja, et al.
+- **abs:** https://arxiv.org/abs/2601.21494 · **PDF:** https://arxiv.org/pdf/2601.21494v2
+
+## 18235. Manifold constrained steepest descent
+
+- **ID:** `2601.21487` · **primary:** `math.OC` · **Submitted:** 2026-01-29
+- **Authors:** Kaiwei Yang, Lexiao Lai
+- **abs:** https://arxiv.org/abs/2601.21487 · **PDF:** https://arxiv.org/pdf/2601.21487v1
+
+## 18236. ETS: Energy-Guided Test-Time Scaling for Training-Free RL Alignment
+
+- **ID:** `2601.21484` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Xiuyu Li, Jinkai Zhang, Mingyang Yi, et al.
+- **abs:** https://arxiv.org/abs/2601.21484 · **PDF:** https://arxiv.org/pdf/2601.21484v1
+
+## 18237. DimStance: Multilingual Datasets for Dimensional Stance Analysis
+
+- **ID:** `2601.21483` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Jonas Becker, Liang-Chih Yu, Shamsuddeen Hassan Muhammad, et al.
+- **abs:** https://arxiv.org/abs/2601.21483 · **PDF:** https://arxiv.org/pdf/2601.21483v2
+
+## 18238. Differential Dynamic Causal Nets: Model Construction, Identification and Group Comparisons
+
+- **ID:** `2601.21478` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-29
+- **Authors:** Kang You, Gary Green, Jian Zhang
+- **abs:** https://arxiv.org/abs/2601.21478 · **PDF:** https://arxiv.org/pdf/2601.21478v1
+
+## 18239. Mean-Field Control on Sparse Graphs: From Local Limits to GNNs via Neighborhood Distributions
+
+- **ID:** `2601.21477` · **primary:** `cs.MA` · **Submitted:** 2026-01-29
+- **Authors:** Tobias Schmidt, Kai Cui
+- **abs:** https://arxiv.org/abs/2601.21477 · **PDF:** https://arxiv.org/pdf/2601.21477v1
+
+## 18240. SOUP: Token-level Single-sample Mix-policy Reinforcement Learning for Large Language Models
+
+- **ID:** `2601.21476` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Lei Yang, Wei Bi, Chenxi Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.21476 · **PDF:** https://arxiv.org/pdf/2601.21476v1
+
+## 18241. Task-free Adaptive Meta Black-box Optimization
+
+- **ID:** `2601.21475` · **primary:** `cs.NE` · **Submitted:** 2026-01-29
+- **Authors:** Chao Wang, Licheng Jiao, Lingling Li, et al.
+- **abs:** https://arxiv.org/abs/2601.21475 · **PDF:** https://arxiv.org/pdf/2601.21475v2
+
+## 18242. ScaleSim: Serving Large-Scale Multi-Agent Simulation with Invocation Distance-Based Memory Management
+
+- **ID:** `2601.21473` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Zaifeng Pan, Yipeng Shen, Zhengding Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.21473 · **PDF:** https://arxiv.org/pdf/2601.21473v1
+
+## 18243. Best Arm Identification with LLM Judges and Limited Human
+
+- **ID:** `2601.21471` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Ruicheng Ao, Hongyu Chen, Siyang Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.21471 · **PDF:** https://arxiv.org/pdf/2601.21471v1
+
+## 18244. PPI-SVRG: Unifying Prediction-Powered Inference and Variance Reduction for Semi-Supervised Optimization
+
+- **ID:** `2601.21470` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Ruicheng Ao, Hongyu Chen, Haoyang Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.21470 · **PDF:** https://arxiv.org/pdf/2601.21470v1
+
+## 18245. Adaptive Confidence Gating in Multi-Agent Collaboration for Efficient and Optimized Code Generation
+
+- **ID:** `2601.21469` · **primary:** `cs.SE` · **Submitted:** 2026-01-29
+- **Authors:** Haoji Zhang, Yuzhe Li, Zhenqiang Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.21469 · **PDF:** https://arxiv.org/pdf/2601.21469v1
+
+## 18246. MemOCR: Layout-Aware Visual Memory for Efficient Long-Horizon Reasoning
+
+- **ID:** `2601.21468` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Yaorui Shi, Shugui Liu, Yu Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.21468 · **PDF:** https://arxiv.org/pdf/2601.21468v4
+
+## 18247. A block-coordinate descent framework for non-convex composite optimization. Application to sparse precision matrix estimation
+
+- **ID:** `2601.21467` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Guillaume Lauga
+- **abs:** https://arxiv.org/abs/2601.21467 · **PDF:** https://arxiv.org/pdf/2601.21467v1
+
+## 18248. Topeax -- An Improved Clustering Topic Model with Density Peak Detection and Lexical-Semantic Term Importance
+
+- **ID:** `2601.21465` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Márton Kardos
+- **abs:** https://arxiv.org/abs/2601.21465 · **PDF:** https://arxiv.org/pdf/2601.21465v1
+
+## 18249. Conversation for Non-verifiable Learning: Self-Evolving LLMs through Meta-Evaluation
+
+- **ID:** `2601.21464` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Yuan Sui, Bryan Hooi
+- **abs:** https://arxiv.org/abs/2601.21464 · **PDF:** https://arxiv.org/pdf/2601.21464v1
+
+## 18250. Unifying Speech Editing Detection and Content Localization via Prior-Enhanced Audio LLMs
+
+- **ID:** `2601.21463` · **primary:** `cs.SD` · **Submitted:** 2026-01-29
+- **Authors:** Jun Xue, Yi Chai, Yanzhen Ren, et al.
+- **abs:** https://arxiv.org/abs/2601.21463 · **PDF:** https://arxiv.org/pdf/2601.21463v1
+
+## 18251. Partial Feedback Online Learning
+
+- **ID:** `2601.21462` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Shihao Shao, Cong Fang, Zhouchen Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.21462 · **PDF:** https://arxiv.org/pdf/2601.21462v3
+
+## 18252. L$^3$: Large Lookup Layers
+
+- **ID:** `2601.21461` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Albert Tseng, Christopher De Sa
+- **abs:** https://arxiv.org/abs/2601.21461 · **PDF:** https://arxiv.org/pdf/2601.21461v2
+
+## 18253. HER: Human-like Reasoning and Reinforcement Learning for LLM Role-playing
+
+- **ID:** `2601.21459` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Chengyu Du, Xintao Wang, Aili Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.21459 · **PDF:** https://arxiv.org/pdf/2601.21459v3
+
+## 18254. Questioning the Coverage-Length Metric in Conformal Prediction: When Shorter Intervals Are Not Better
+
+- **ID:** `2601.21455` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** Yizhou Min, Yizhou Lu, Lanqi Li, et al.
+- **abs:** https://arxiv.org/abs/2601.21455 · **PDF:** https://arxiv.org/pdf/2601.21455v1
+
+## 18255. LION: A Clifford Neural Paradigm for Multimodal-Attributed Graph Learning
+
+- **ID:** `2601.21453` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Xunkai Li, Zhengyu Wu, Zekai Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.21453 · **PDF:** https://arxiv.org/pdf/2601.21453v1
+
+## 18256. SAGE: Sequence-level Adaptive Gradient Evolution for Generative Recommendation
+
+- **ID:** `2601.21452` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yu Xie, Xing Kai Ren, Ying Qi, et al.
+- **abs:** https://arxiv.org/abs/2601.21452 · **PDF:** https://arxiv.org/pdf/2601.21452v3
+
+## 18257. ChipBench: A Next-Step Benchmark for Evaluating LLM Performance in AI-Aided Chip Design
+
+- **ID:** `2601.21448` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Zhongkai Yu, Chenyang Zhou, Yichen Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.21448 · **PDF:** https://arxiv.org/pdf/2601.21448v2
+
+## 18258. Synthetic Pattern Generation and Detection of Financial Activities using Graph Autoencoders
+
+- **ID:** `2601.21446` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Francesco Zola, Lucia Muñoz, Andrea Venturi, et al.
+- **abs:** https://arxiv.org/abs/2601.21446 · **PDF:** https://arxiv.org/pdf/2601.21446v1
+
+## 18259. Spava: Accelerating Long-Video Understanding via Sequence-Parallelism-aware Approximate Attention
+
+- **ID:** `2601.21444` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Yuxiang Huang, Mingye Li, Xu Han, et al.
+- **abs:** https://arxiv.org/abs/2601.21444 · **PDF:** https://arxiv.org/pdf/2601.21444v1
+
+## 18260. The Paradox of Robustness: Decoupling Rule-Based Logic from Affective Noise in High-Stakes Decision-Making
+
+- **ID:** `2601.21439` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Jon Chun, Katherine Elkins
+- **abs:** https://arxiv.org/abs/2601.21439 · **PDF:** https://arxiv.org/pdf/2601.21439v1
+
+## 18261. Accurate Network Traffic Matrix Prediction via LEAD: a Large Language Model-Enhanced Adapter-Based Conditional Diffusion Model
+
+- **ID:** `2601.21437` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yu Sun, Yaqiong Liu, Nan Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.21437 · **PDF:** https://arxiv.org/pdf/2601.21437v2
+
+## 18262. From Consistency to Complementarity: Aligned and Disentangled Multi-modal Learning for Time Series Understanding and Reasoning
+
+- **ID:** `2601.21436` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Hang Ni, Weijia Zhang, Fei Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21436 · **PDF:** https://arxiv.org/pdf/2601.21436v2
+
+## 18263. When Prohibitions Become Permissions: Auditing Negation Sensitivity in Language Models
+
+- **ID:** `2601.21433` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Katherine Elkins, Jon Chun
+- **abs:** https://arxiv.org/abs/2601.21433 · **PDF:** https://arxiv.org/pdf/2601.21433v1
+
+## 18264. Lossy Common Information in a Learnable Gray-Wyner Network
+
+- **ID:** `2601.21424` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Anderson de Andrade, Alon Harell, Ivan V. Bajić
+- **abs:** https://arxiv.org/abs/2601.21424 · **PDF:** https://arxiv.org/pdf/2601.21424v2
+
+## 18265. ConceptMoE: Adaptive Token-to-Concept Compression for Implicit Compute Allocation
+
+- **ID:** `2601.21420` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Zihao Huang, Jundong Zhou, Xingwei Qu, et al.
+- **abs:** https://arxiv.org/abs/2601.21420 · **PDF:** https://arxiv.org/pdf/2601.21420v1
+
+## 18266. Revisiting Diffusion Model Predictions Through Dimensionality
+
+- **ID:** `2601.21419` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Qing Jin, Chaoyang Wang
+- **abs:** https://arxiv.org/abs/2601.21419 · **PDF:** https://arxiv.org/pdf/2601.21419v2
+
+## 18267. Mitigating Overthinking in Large Reasoning Models via Difficulty-aware Reinforcement Learning
+
+- **ID:** `2601.21418` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Qian Wan, Ziao Xu, Luona Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.21418 · **PDF:** https://arxiv.org/pdf/2601.21418v1
+
+## 18268. System 1&2 Synergy via Dynamic Model Interpolation
+
+- **ID:** `2601.21414` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Chenxu Yang, Qingyi Si, Chong Tian, et al.
+- **abs:** https://arxiv.org/abs/2601.21414 · **PDF:** https://arxiv.org/pdf/2601.21414v1
+
+## 18269. Statsformer: Validated Ensemble Learning with LLM-Derived Semantic Priors
+
+- **ID:** `2601.21410` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** Erica Zhang, Naomi Sagan, Danny Tse, et al.
+- **abs:** https://arxiv.org/abs/2601.21410 · **PDF:** https://arxiv.org/pdf/2601.21410v2
+
+## 18270. BrainFuse: a unified infrastructure integrating realistic biological modeling and core AI methodology
+
+- **ID:** `2601.21407` · **primary:** `cs.NE` · **Submitted:** 2026-01-29
+- **Authors:** Baiyu Chen, Yujie Wu, Siyuan Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.21407 · **PDF:** https://arxiv.org/pdf/2601.21407v1
+
+## 18271. Generation Enhances Understanding in Unified Multimodal Models via Multi-Representation Generation
+
+- **ID:** `2601.21406` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Zihan Su, Hongyang Wei, Kangrui Cen, et al.
+- **abs:** https://arxiv.org/abs/2601.21406 · **PDF:** https://arxiv.org/pdf/2601.21406v2
+
+## 18272. DataCross: A Unified Benchmark and Agent Framework for Cross-Modal Heterogeneous Data Analysis
+
+- **ID:** `2601.21403` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Ruyi Qi, Zhou Liu, Wentao Zhang
+- **abs:** https://arxiv.org/abs/2601.21403 · **PDF:** https://arxiv.org/pdf/2601.21403v1
+
+## 18273. Intrinsic Reward Policy Optimization for Sparse-Reward Environments
+
+- **ID:** `2601.21391` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Minjae Cho, Huy Trong Tran
+- **abs:** https://arxiv.org/abs/2601.21391 · **PDF:** https://arxiv.org/pdf/2601.21391v1
+
+## 18274. Learning to Optimize Job Shop Scheduling Under Structural Uncertainty
+
+- **ID:** `2601.21389` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Rui Zhang, Jianwei Niu, Xuefeng Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.21389 · **PDF:** https://arxiv.org/pdf/2601.21389v1
+
+## 18275. User-Centric Evidence Ranking for Attribution and Fact Verification
+
+- **ID:** `2601.21387` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Guy Alt, Eran Hirsch, Serwar Basch, et al.
+- **abs:** https://arxiv.org/abs/2601.21387 · **PDF:** https://arxiv.org/pdf/2601.21387v1
+
+## 18276. Understanding Frechet Speech Distance for Synthetic Speech Quality Evaluation
+
+- **ID:** `2601.21386` · **primary:** `cs.SD` · **Submitted:** 2026-01-29
+- **Authors:** June-Woo Kim, Dhruv Agarwal, Federica Cerina
+- **abs:** https://arxiv.org/abs/2601.21386 · **PDF:** https://arxiv.org/pdf/2601.21386v1
+
+## 18277. Sim-MSTNet: sim2real based Multi-task SpatioTemporal Network Traffic Forecasting
+
+- **ID:** `2601.21384` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Hui Ma, Qingzhong Li, Jin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21384 · **PDF:** https://arxiv.org/pdf/2601.21384v1
+
+## 18278. DA-SPS: A Dual-stage Network based on Singular Spectrum Analysis, Patching-strategy and Spearman-correlation for Multivariate Time-series Prediction
+
+- **ID:** `2601.21381` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Tianhao Zhang, Shusen Ma, Yu Kang, et al.
+- **abs:** https://arxiv.org/abs/2601.21381 · **PDF:** https://arxiv.org/pdf/2601.21381v1
+
+## 18279. TeachBench: A Syllabus-Grounded Framework for Evaluating Teaching Ability in Large Language Models
+
+- **ID:** `2601.21375` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Zheng Li, Siyao Song, Jingyuan Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.21375 · **PDF:** https://arxiv.org/pdf/2601.21375v1
+
+## 18280. NEMO: Execution-Aware Optimization Modeling via Autonomous Coding Agents
+
+- **ID:** `2601.21372` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Yang Song, Anoushka Vyas, Zirui Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.21372 · **PDF:** https://arxiv.org/pdf/2601.21372v1
+
+## 18281. Rethinking Federated Graph Foundation Models: A Graph-Language Alignment-based Approach
+
+- **ID:** `2601.21369` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yinlin Zhu, Di Wu, Xianzhi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.21369 · **PDF:** https://arxiv.org/pdf/2601.21369v1
+
+## 18282. Hebbian Learning with Global Direction
+
+- **ID:** `2601.21367` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Wenjia Hua, Kejie Zhao, Luziwei Leng, et al.
+- **abs:** https://arxiv.org/abs/2601.21367 · **PDF:** https://arxiv.org/pdf/2601.21367v1
+
+## 18283. Perceptrons and localization of attention's mean-field landscape
+
+- **ID:** `2601.21366` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Antonio Álvarez-López, Borjan Geshkovski, Domènec Ruiz-Balet
+- **abs:** https://arxiv.org/abs/2601.21366 · **PDF:** https://arxiv.org/pdf/2601.21366v1
+
+## 18284. The Compliance Paradox: Semantic-Instruction Decoupling in Automated Academic Code Evaluation
+
+- **ID:** `2601.21360` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Devanshu Sahoo, Manish Prasad, Vasudev Majhi, et al.
+- **abs:** https://arxiv.org/abs/2601.21360 · **PDF:** https://arxiv.org/pdf/2601.21360v1
+
+## 18285. Graph-Free Root Cause Analysis
+
+- **ID:** `2601.21359` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Luan Pham
+- **abs:** https://arxiv.org/abs/2601.21359 · **PDF:** https://arxiv.org/pdf/2601.21359v1
+
+## 18286. Latent Chain-of-Thought as Planning: Decoupling Reasoning from Verbalization
+
+- **ID:** `2601.21358` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Jiecong Wang, Hao Peng, Chunyang Liu
+- **abs:** https://arxiv.org/abs/2601.21358 · **PDF:** https://arxiv.org/pdf/2601.21358v2
+
+## 18287. Expected Improvement via Gradient Norms
+
+- **ID:** `2601.21357` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Joshua Hang Sai Ip, Georgios Makrygiorgos, Ali Mesbah
+- **abs:** https://arxiv.org/abs/2601.21357 · **PDF:** https://arxiv.org/pdf/2601.21357v1
+
+## 18288. BEAP-Agent: Backtrackable Execution and Adaptive Planning for GUI Agents
+
+- **ID:** `2601.21352` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Ziyu Lu, Tengjin Weng, Yiying Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.21352 · **PDF:** https://arxiv.org/pdf/2601.21352v1
+
+## 18289. Theoretically Optimal Attention/FFN Ratios in Disaggregated LLM Serving
+
+- **ID:** `2601.21351` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Chendong Song, Meixuan Wang, Hang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.21351 · **PDF:** https://arxiv.org/pdf/2601.21351v1
+
+## 18290. Factored Causal Representation Learning for Robust Reward Modeling in RLHF
+
+- **ID:** `2601.21350` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yupei Yang, Lin Yang, Wanxi Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.21350 · **PDF:** https://arxiv.org/pdf/2601.21350v1
+
+## 18291. L2R: Low-Rank and Lipschitz-Controlled Routing for Mixture-of-Experts
+
+- **ID:** `2601.21349` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Minghao Yang, Ren Togo, Guang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.21349 · **PDF:** https://arxiv.org/pdf/2601.21349v1
+
+## 18292. Memorization Control in Diffusion Models from Denoising-centric Perspective
+
+- **ID:** `2601.21348` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Thuy Phuong Vu, Mai Viet Hoang Do, Minhhuy Le, et al.
+- **abs:** https://arxiv.org/abs/2601.21348 · **PDF:** https://arxiv.org/pdf/2601.21348v1
+
+## 18293. Dynamic Framework for Collaborative Learning: Leveraging Advanced LLM with Adaptive Feedback Mechanisms
+
+- **ID:** `2601.21344` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Hassam Tahir, Faizan Faisal, Fady Alnajjar, et al.
+- **abs:** https://arxiv.org/abs/2601.21344 · **PDF:** https://arxiv.org/pdf/2601.21344v1
+
+## 18294. Self-Improving Pretraining: using post-trained models to pretrain better models
+
+- **ID:** `2601.21343` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Ellen Xiaoqing Tan, Shehzaad Dhuliawala, Jing Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.21343 · **PDF:** https://arxiv.org/pdf/2601.21343v2
+
+## 18295. Ostrakon-VL: Towards Domain-Expert MLLM for Food-Service and Retail Stores
+
+- **ID:** `2601.21342` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Zhiyong Shen, Gongpeng Zhao, Jun Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.21342 · **PDF:** https://arxiv.org/pdf/2601.21342v1
+
+## 18296. EHR-RAG: Bridging Long-Horizon Structured Electronic Health Records and Large Language Models via Enhanced Retrieval-Augmented Generation
+
+- **ID:** `2601.21340` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Lang Cao, Qingyu Chen, Yue Guo
+- **abs:** https://arxiv.org/abs/2601.21340 · **PDF:** https://arxiv.org/pdf/2601.21340v1
+
+## 18297. Within-Model vs Between-Prompt Variability in Large Language Models for Creative Tasks
+
+- **ID:** `2601.21339` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Jennifer Haase, Jana Gonnermann-Müller, Paul H. P. Hanel, et al.
+- **abs:** https://arxiv.org/abs/2601.21339 · **PDF:** https://arxiv.org/pdf/2601.21339v1
+
+## 18298. Qwen3-ASR Technical Report
+
+- **ID:** `2601.21337` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Xian Shi, Xiong Wang, Zhifang Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.21337 · **PDF:** https://arxiv.org/pdf/2601.21337v2
+
+## 18299. Modeling Endogenous Logic: Causal Neuro-Symbolic Reasoning Model for Explainable Multi-Behavior Recommendation
+
+- **ID:** `2601.21335` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Yuzhe Chen, Jie Cao, Youquan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21335 · **PDF:** https://arxiv.org/pdf/2601.21335v1
+
+## 18300. Convex Loss Functions for Support Vector Machines (SVMs) and Neural Networks
+
+- **ID:** `2601.21331` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Filippo Portera
+- **abs:** https://arxiv.org/abs/2601.21331 · **PDF:** https://arxiv.org/pdf/2601.21331v4
+
+## 18301. Scalable Analytic Classifiers with Associative Drift Compensation for Class-Incremental Learning of Vision Transformers
+
+- **ID:** `2602.00144` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Xuan Rao, Mingming Ha, Bo Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.00144 · **PDF:** https://arxiv.org/pdf/2602.00144v1
+
+## 18302. Bulk-Calibrated Credal Ambiguity Sets: Fast, Tractable Decision Making under Out-of-Sample Contamination
+
+- **ID:** `2601.21324` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** Mengqi Chen, Thomas B. Berrett, Theodoros Damoulas, et al.
+- **abs:** https://arxiv.org/abs/2601.21324 · **PDF:** https://arxiv.org/pdf/2601.21324v1
+
+## 18303. Adversarial Vulnerability Transcends Computational Paradigms: Feature Engineering Provides No Defense Against Neural Adversarial Transfer
+
+- **ID:** `2601.21323` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Achraf Hsain, Ahmed Abdelkader, Emmanuel Baldwin Mbaya, et al.
+- **abs:** https://arxiv.org/abs/2601.21323 · **PDF:** https://arxiv.org/pdf/2601.21323v1
+
+## 18304. White-Box Op-Amp Design via Human-Mimicking Reasoning
+
+- **ID:** `2601.21321` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Zihao Chen, Jiayin Wang, Ziyi Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.21321 · **PDF:** https://arxiv.org/pdf/2601.21321v1
+
+## 18305. Optimal Transport-Induced Samples against Out-of-Distribution Overconfidence
+
+- **ID:** `2601.21320` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Keke Tang, Ziyong Du, Xiaofei Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21320 · **PDF:** https://arxiv.org/pdf/2601.21320v1
+
+## 18306. Heterogeneous Vertiport Selection Optimization for On-Demand Air Taxi Services: A Deep Reinforcement Learning Approach
+
+- **ID:** `2601.21316` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Aoyu Pang, Maonan Wang, Zifan Sha, et al.
+- **abs:** https://arxiv.org/abs/2601.21316 · **PDF:** https://arxiv.org/pdf/2601.21316v1
+
+## 18307. Distributionally Robust Classification for Multi-source Unsupervised Domain Adaptation
+
+- **ID:** `2601.21315` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Seonghwi Kim, Sung Ho Jo, Wooseok Ha, et al.
+- **abs:** https://arxiv.org/abs/2601.21315 · **PDF:** https://arxiv.org/pdf/2601.21315v2
+
+## 18308. Few-Shot Learning for Dynamic Operations of Automated Electric Taxi Fleets under Evolving Charging Infrastructure: A Meta-Deep Reinforcement Learning Approach
+
+- **ID:** `2601.21312` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Xiaozhuang Li, Xindi Tang, Fang He
+- **abs:** https://arxiv.org/abs/2601.21312 · **PDF:** https://arxiv.org/pdf/2601.21312v1
+
+## 18309. CircuChain: Disentangling Competence and Compliance in LLM Circuit Analysis
+
+- **ID:** `2602.15037` · **primary:** `cs.SE` · **Submitted:** 2026-01-29
+- **Authors:** Mayank Ravishankara
+- **abs:** https://arxiv.org/abs/2602.15037 · **PDF:** https://arxiv.org/pdf/2602.15037v1
+
+## 18310. Transferable Graph Condensation from the Causal Perspective
+
+- **ID:** `2601.21309` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Huaming Du, Yijie Huang, Su Yao, et al.
+- **abs:** https://arxiv.org/abs/2601.21309 · **PDF:** https://arxiv.org/pdf/2601.21309v3
+
+## 18311. PlotChain: Deterministic Checkpointed Evaluation of Multimodal LLMs on Engineering Plot Reading
+
+- **ID:** `2602.13232` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Mayank Ravishankara
+- **abs:** https://arxiv.org/abs/2602.13232 · **PDF:** https://arxiv.org/pdf/2602.13232v1
+
+## 18312. The Surprising Difficulty of Search in Model-Based Reinforcement Learning
+
+- **ID:** `2601.21306` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Wei-Di Chang, Mikael Henaff, Brandon Amos, et al.
+- **abs:** https://arxiv.org/abs/2601.21306 · **PDF:** https://arxiv.org/pdf/2601.21306v1
+
+## 18313. Achieving $\varepsilon^{-2}$ Dependence for Average-Reward Q-Learning with a New Contraction Principle
+
+- **ID:** `2601.21301` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Zijun Chen, Zaiwei Chen, Nian Si, et al.
+- **abs:** https://arxiv.org/abs/2601.21301 · **PDF:** https://arxiv.org/pdf/2601.21301v1
+
+## 18314. Grounding and Enhancing Informativeness and Utility in Dataset Distillation
+
+- **ID:** `2601.21296` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Shaobo Wang, Yantai Yang, Guo Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.21296 · **PDF:** https://arxiv.org/pdf/2601.21296v1
+
+## 18315. Missing-Data-Induced Phase Transitions in Spectral PLS for Multimodal Learning
+
+- **ID:** `2601.21294` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Anders Gjølbye, Ida Kargaard, Emma Kargaard, et al.
+- **abs:** https://arxiv.org/abs/2601.21294 · **PDF:** https://arxiv.org/pdf/2601.21294v1
+
+## 18316. Physics-Guided Tiny-Mamba Transformer for Reliability-Aware Early Fault Warning
+
+- **ID:** `2601.21293` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Changyu Li, Dingcheng Huang, Kexuan Yao, et al.
+- **abs:** https://arxiv.org/abs/2601.21293 · **PDF:** https://arxiv.org/pdf/2601.21293v1
+
+## 18317. TimeSliver : Symbolic-Linear Decomposition for Explainable Time Series Classification
+
+- **ID:** `2601.21289` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Akash Pandey, Payal Mohapatra, Wei Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.21289 · **PDF:** https://arxiv.org/pdf/2601.21289v2
+
+## 18318. SCENE: Semantic-aware Codec Enhancement with Neural Embeddings
+
+- **ID:** `2601.22189` · **primary:** `eess.IV` · **Submitted:** 2026-01-29
+- **Authors:** Han-Yu Lin, Li-Wei Chen, Hung-Shin Lee
+- **abs:** https://arxiv.org/abs/2601.22189 · **PDF:** https://arxiv.org/pdf/2601.22189v1
+
+## 18319. Drive-KD: Multi-Teacher Distillation for VLMs in Autonomous Driving
+
+- **ID:** `2601.21288` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Weitong Lian, Zecong Tang, Haoran Li, et al.
+- **abs:** https://arxiv.org/abs/2601.21288 · **PDF:** https://arxiv.org/pdf/2601.21288v1
+
+## 18320. Zenith: Scaling up Ranking Models for Billion-scale Livestreaming Recommendation
+
+- **ID:** `2601.21285` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Ruifeng Zhang, Zexi Huang, Zikai Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21285 · **PDF:** https://arxiv.org/pdf/2601.21285v3
+
+## 18321. PILD: Physics-Informed Learning via Diffusion
+
+- **ID:** `2601.21284` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Tianyi Zeng, Tianyi Wang, Jiaru Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.21284 · **PDF:** https://arxiv.org/pdf/2601.21284v1
+
+## 18322. DUET: Distilled LLM Unlearning from an Efficiently Contextualized Teacher
+
+- **ID:** `2601.21283` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yisheng Zhong, Zhengbang Yang, Zhuangdi Zhu
+- **abs:** https://arxiv.org/abs/2601.21283 · **PDF:** https://arxiv.org/pdf/2601.21283v2
+
+## 18323. EGAM: Extended Graph Attention Model for Solving Routing Problems
+
+- **ID:** `2601.21281` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Licheng Wang, Yuzi Yan, Mingtao Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.21281 · **PDF:** https://arxiv.org/pdf/2601.21281v1
+
+## 18324. NEXUS: Bit-Exact ANN-to-SNN Equivalence via Neuromorphic Gate Circuits with Surrogate-Free Training
+
+- **ID:** `2601.21279` · **primary:** `cs.NE` · **Submitted:** 2026-01-29
+- **Authors:** Zhengzheng Tang
+- **abs:** https://arxiv.org/abs/2601.21279 · **PDF:** https://arxiv.org/pdf/2601.21279v2
+
+## 18325. GeoRC: A Benchmark for Geolocation Reasoning Chains
+
+- **ID:** `2601.21278` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Mohit Talreja, Joshua Diao, Jim Thannikary James, et al.
+- **abs:** https://arxiv.org/abs/2601.21278 · **PDF:** https://arxiv.org/pdf/2601.21278v1
+
+## 18326. More Code, Less Reuse: Investigating Code Quality and Reviewer Sentiment towards AI-generated Pull Requests
+
+- **ID:** `2601.21276` · **primary:** `cs.SE` · **Submitted:** 2026-01-29
+- **Authors:** Haoming Huang, Pongchai Jaisri, Shota Shimizu, et al.
+- **abs:** https://arxiv.org/abs/2601.21276 · **PDF:** https://arxiv.org/pdf/2601.21276v1
+
+## 18327. Lightweight High-Fidelity Low-Bitrate Talking Face Compression for 3D Video Conference
+
+- **ID:** `2601.21269` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Jianglong Li, Jun Xu, Bingcong Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.21269 · **PDF:** https://arxiv.org/pdf/2601.21269v1
+
+## 18328. Reinforcement Learning from Meta-Evaluation: Aligning Language Models Without Ground-Truth Labels
+
+- **ID:** `2601.21268` · **primary:** `cs.NE` · **Submitted:** 2026-01-29
+- **Authors:** Micah Rentschler, Jesse Roberts
+- **abs:** https://arxiv.org/abs/2601.21268 · **PDF:** https://arxiv.org/pdf/2601.21268v1
+
+## 18329. Model-Free Neural State Estimation in Nonlinear Dynamical Systems: Comparing Neural and Classical Filters
+
+- **ID:** `2601.21266` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Zhuochen Liu, Hans Walker, Rahul Jain
+- **abs:** https://arxiv.org/abs/2601.21266 · **PDF:** https://arxiv.org/pdf/2601.21266v2
+
+## 18330. CausalEmbed: Auto-Regressive Multi-Vector Generation in Latent Space for Visual Document Embedding
+
+- **ID:** `2601.21262` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Jiahao Huo, Yu Huang, Yibo Yan, et al.
+- **abs:** https://arxiv.org/abs/2601.21262 · **PDF:** https://arxiv.org/pdf/2601.21262v2
+
+## 18331. Music Plagiarism Detection: Problem Formulation and a Segment-based Solution
+
+- **ID:** `2601.21260` · **primary:** `cs.SD` · **Submitted:** 2026-01-29
+- **Authors:** Seonghyeon Go, Yumin Kim
+- **abs:** https://arxiv.org/abs/2601.21260 · **PDF:** https://arxiv.org/pdf/2601.21260v2
+
+## 18332. MolLIBRA: Genetic Molecular Optimization with Multi-Fingerprint Surrogates and Text-Molecule Aligned Critic
+
+- **ID:** `2602.07002` · **primary:** `cs.NE` · **Submitted:** 2026-01-29
+- **Authors:** Masahi Okada, Kazuki Sakai, Hiroaki Yoshida, et al.
+- **abs:** https://arxiv.org/abs/2602.07002 · **PDF:** https://arxiv.org/pdf/2602.07002v1
+
+## 18333. MoCo: A One-Stop Shop for Model Collaboration Research
+
+- **ID:** `2601.21257` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Shangbin Feng, Yuyang Bai, Ziyuan Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.21257 · **PDF:** https://arxiv.org/pdf/2601.21257v1
+
+## 18334. Hypersolid: Emergent Vision Representations via Short-Range Repulsion
+
+- **ID:** `2601.21255` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Esteban Rodríguez-Betancourt, Edgar Casasola-Murillo
+- **abs:** https://arxiv.org/abs/2601.21255 · **PDF:** https://arxiv.org/pdf/2601.21255v1
+
+## 18335. Position: Certifiable State Integrity in Cyber-Physical Systems -- Why Modular Sovereignty Solves the Plasticity-Stability Paradox
+
+- **ID:** `2601.21249` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Enzo Nicolás Spotorno, Antônio Augusto Medeiros Fröhlich
+- **abs:** https://arxiv.org/abs/2601.21249 · **PDF:** https://arxiv.org/pdf/2601.21249v1
+
+## 18336. Conditional Generative Framework with Peak-Aware Attention for Robust Chemical Detection under Interferences
+
+- **ID:** `2601.21246` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Namkyung Yoon, Sanghong Kim, Hwangnam Kim
+- **abs:** https://arxiv.org/abs/2601.21246 · **PDF:** https://arxiv.org/pdf/2601.21246v1
+
+## 18337. Less Noise, More Voice: Reinforcement Learning for Reasoning via Instruction Purification
+
+- **ID:** `2601.21244` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yiju Guo, Tianyi Hu, Zexu Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.21244 · **PDF:** https://arxiv.org/pdf/2601.21244v2
+
+## 18338. Solving the Offline and Online Min-Max Problem of Non-smooth Submodular-Concave Functions: A Zeroth-Order Approach
+
+- **ID:** `2601.21243` · **primary:** `math.OC` · **Submitted:** 2026-01-29
+- **Authors:** Amir Ali Farzin, Yuen-Man Pun, Philipp Braun, et al.
+- **abs:** https://arxiv.org/abs/2601.21243 · **PDF:** https://arxiv.org/pdf/2601.21243v1
+
+## 18339. Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks
+
+- **ID:** `2601.21242` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Luwei Sun, Dongrui Shen, Jianfe Li, et al.
+- **abs:** https://arxiv.org/abs/2601.21242 · **PDF:** https://arxiv.org/pdf/2601.21242v1
+
+## 18340. TIDE: Tuning-Integrated Dynamic Evolution for LLM-Based Automated Heuristic Design
+
+- **ID:** `2601.21239` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Chentong Chen, Mengyuan Zhong, Ye Fan, et al.
+- **abs:** https://arxiv.org/abs/2601.21239 · **PDF:** https://arxiv.org/pdf/2601.21239v2
+
+## 18341. PTQ4ARVG: Post-Training Quantization for AutoRegressive Visual Generation Models
+
+- **ID:** `2601.21238` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Xuewen Liu, Zhikai Li, Jing Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.21238 · **PDF:** https://arxiv.org/pdf/2601.21238v1
+
+## 18342. Quantifying Noise in Language Generation
+
+- **ID:** `2601.21237` · **primary:** `cs.DS` · **Submitted:** 2026-01-29
+- **Authors:** Aaron Li, Ian Zhang
+- **abs:** https://arxiv.org/abs/2601.21237 · **PDF:** https://arxiv.org/pdf/2601.21237v1
+
+## 18343. SHARP: Social Harm Analysis via Risk Profiles for Measuring Inequities in Large Language Models
+
+- **ID:** `2601.21235` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Alok Abhishek, Tushar Bandopadhyay, Lisa Erickson
+- **abs:** https://arxiv.org/abs/2601.21235 · **PDF:** https://arxiv.org/pdf/2601.21235v1
+
+## 18344. PHDME: Physics-Informed Diffusion Models without Explicit Governing Equations
+
+- **ID:** `2601.21234` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Kaiyuan Tan, Kendra Givens, Peilun Li, et al.
+- **abs:** https://arxiv.org/abs/2601.21234 · **PDF:** https://arxiv.org/pdf/2601.21234v1
+
+## 18345. Just Ask: Curious Code Agents Reveal System Prompts in Frontier LLMs
+
+- **ID:** `2601.21233` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Xiang Zheng, Yutao Wu, Hanxun Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.21233 · **PDF:** https://arxiv.org/pdf/2601.21233v1
+
+## 18346. Data-Driven Generation of Neutron Star Equations of State Using Variational Autoencoders
+
+- **ID:** `2601.21231` · **primary:** `astro-ph.HE` · **Submitted:** 2026-01-29
+- **Authors:** Alex Ross, Tianqi Zhao, Sanjay Reddy
+- **abs:** https://arxiv.org/abs/2601.21231 · **PDF:** https://arxiv.org/pdf/2601.21231v1
+
+## 18347. Delegation Without Living Governance
+
+- **ID:** `2601.21226` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Wolfgang Rohde
+- **abs:** https://arxiv.org/abs/2601.21226 · **PDF:** https://arxiv.org/pdf/2601.21226v1
+
+## 18348. MGSM-Pro: A Simple Strategy for Robust Multilingual Mathematical Reasoning Evaluation
+
+- **ID:** `2601.21225` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Tianyi Xu, Kosei Uemura, Alfred Malengo Kondoro, et al.
+- **abs:** https://arxiv.org/abs/2601.21225 · **PDF:** https://arxiv.org/pdf/2601.21225v1
+
+## 18349. Early warning prediction: Onsager-Machlup vs Schrödinger
+
+- **ID:** `2602.00143` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-29
+- **Authors:** Xiaoai Xu, Yixuan Zhou, Xiang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.00143 · **PDF:** https://arxiv.org/pdf/2602.00143v1
+
+## 18350. Causal Discovery for Explainable AI: A Dual-Encoding Approach
+
+- **ID:** `2601.21221` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Henry Salgado, Meagan R. Kendall, Martine Ceberio
+- **abs:** https://arxiv.org/abs/2601.21221 · **PDF:** https://arxiv.org/pdf/2601.21221v1
+
+## 18351. Soft Quantization: Model Compression Via Weight Coupling
+
+- **ID:** `2601.21219` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Daniel T. Bernstein, Luca Di Carlo, David Schwab
+- **abs:** https://arxiv.org/abs/2601.21219 · **PDF:** https://arxiv.org/pdf/2601.21219v1
+
+## 18352. Parametric Knowledge is Not All You Need: Toward Honest Large Language Models via Retrieval of Pretraining Data
+
+- **ID:** `2601.21218` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Christopher Adrian Kusuma, Muhammad Reza Qorib, Hwee Tou Ng
+- **abs:** https://arxiv.org/abs/2601.21218 · **PDF:** https://arxiv.org/pdf/2601.21218v1
+
+## 18353. A Flexible Empirical Bayes Approach to Generalized Linear Models, with Applications to Sparse Logistic Regression
+
+- **ID:** `2601.21217` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** Dongyue Xie, Wanrong Zhu, Matthew Stephens
+- **abs:** https://arxiv.org/abs/2601.21217 · **PDF:** https://arxiv.org/pdf/2601.21217v1
+
+## 18354. Temporal Context and Architecture: A Benchmark for Naturalistic EEG Decoding
+
+- **ID:** `2601.21215` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Mehmet Ergezer
+- **abs:** https://arxiv.org/abs/2601.21215 · **PDF:** https://arxiv.org/pdf/2601.21215v1
+
+## 18355. Scaling Reasoning Hop Exposes Weaknesses: Demystifying and Improving Hop Generalization in Large Language Models
+
+- **ID:** `2601.21214` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Zhaoyi Li, Jiatong Li, Gangwei Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.21214 · **PDF:** https://arxiv.org/pdf/2601.21214v1
+
+## 18356. Intelli-Planner: Towards Customized Urban Planning via Large Language Model Empowered Reinforcement Learning
+
+- **ID:** `2601.21212` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Xixian Yong, Peilin Sun, Zihe Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21212 · **PDF:** https://arxiv.org/pdf/2601.21212v1
+
+## 18357. Uncovering Hidden Correctness in LLM Causal Reasoning via Symbolic Verification
+
+- **ID:** `2601.21210` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Paul He, Yinya Huang, Mrinmaya Sachan, et al.
+- **abs:** https://arxiv.org/abs/2601.21210 · **PDF:** https://arxiv.org/pdf/2601.21210v1
+
+## 18358. When should I search more: Adaptive Complex Query Optimization with Reinforcement Learning
+
+- **ID:** `2601.21208` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Wei Wen, Sihang Deng, Tianjun Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.21208 · **PDF:** https://arxiv.org/pdf/2601.21208v1
+
+## 18359. A Sheaf-Theoretic and Topological Perspective on Complex Network Modeling and Attention Mechanisms in Graph Neural Models
+
+- **ID:** `2601.21207` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Chuan-Shen Hu
+- **abs:** https://arxiv.org/abs/2601.21207 · **PDF:** https://arxiv.org/pdf/2601.21207v2
+
+## 18360. Multilingual Dysarthric Speech Assessment Using Universal Phone Recognition and Language-Specific Phonemic Contrast Modeling
+
+- **ID:** `2601.21205` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Eunjung Yeo, Julie M. Liss, Visar Berisha, et al.
+- **abs:** https://arxiv.org/abs/2601.21205 · **PDF:** https://arxiv.org/pdf/2601.21205v2
+
+## 18361. Scaling Embeddings Outperforms Scaling Experts in Language Models
+
+- **ID:** `2601.21204` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Hong Liu, Jiaqi Zhang, Chao Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21204 · **PDF:** https://arxiv.org/pdf/2601.21204v2
+
+## 18362. Rethinking Self-Training Based Cross-Subject Domain Adaptation for SSVEP Classification
+
+- **ID:** `2601.21203` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Weiguang Wang, Yong Liu, Yingjie Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.21203 · **PDF:** https://arxiv.org/pdf/2601.21203v1
+
+## 18363. Provably Reliable Classifier Guidance via Cross-Entropy Control
+
+- **ID:** `2601.21200` · **primary:** `stat.ML` · **Submitted:** 2026-01-29
+- **Authors:** Sharan Sahu, Arisina Banerjee, Yuchen Wu
+- **abs:** https://arxiv.org/abs/2601.21200 · **PDF:** https://arxiv.org/pdf/2601.21200v2
+
+## 18364. Thinker: A vision-language foundation model for embodied intelligence
+
+- **ID:** `2601.21199` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Baiyu Pan, Daqin Luo, Junpeng Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.21199 · **PDF:** https://arxiv.org/pdf/2601.21199v1
+
+## 18365. ZipMoE: Efficient On-Device MoE Serving via Lossless Compression and Cache-Affinity Scheduling
+
+- **ID:** `2601.21198` · **primary:** `cs.DC` · **Submitted:** 2026-01-29
+- **Authors:** Yuchen Yang, Yaru Zhao, Pu Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.21198 · **PDF:** https://arxiv.org/pdf/2601.21198v1
+
+## 18366. Do Reasoning Models Enhance Embedding Models?
+
+- **ID:** `2601.21192` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Wun Yu Chan, Shaojin Chen, Huihao Jing, et al.
+- **abs:** https://arxiv.org/abs/2601.21192 · **PDF:** https://arxiv.org/pdf/2601.21192v1
+
+## 18367. From Linear Input to Hierarchical Structure: Function Words as Statistical Cues for Language Learning
+
+- **ID:** `2601.21191` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Xiulin Yang, Heidi Getz, Ethan Gotlieb Wilcox
+- **abs:** https://arxiv.org/abs/2601.21191 · **PDF:** https://arxiv.org/pdf/2601.21191v1
+
+## 18368. Adaptive and Robust Cost-Aware Proof of Quality for Decentralized LLM Inference Networks
+
+- **ID:** `2601.21189` · **primary:** `cs.CR` · **Submitted:** 2026-01-29
+- **Authors:** Arther Tian, Alex Ding, Frank Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.21189 · **PDF:** https://arxiv.org/pdf/2601.21189v1
+
+## 18369. FRISM: Fine-Grained Reasoning Injection via Subspace-Level Model Merging for Vision-Language Models
+
+- **ID:** `2601.21187` · **primary:** `cs.CV` · **Submitted:** 2026-01-29
+- **Authors:** Chenyu Huang, Peng Ye, Xudong Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.21187 · **PDF:** https://arxiv.org/pdf/2601.21187v1
+
+## 18370. Sycophantic Anchors: Localizing and Quantifying User Agreement in Reasoning Models
+
+- **ID:** `2601.21183` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Jacek Duszenko
+- **abs:** https://arxiv.org/abs/2601.21183 · **PDF:** https://arxiv.org/pdf/2601.21183v2
+
+## 18371. Rethinking Refinement: Correcting Generative Bias without Noise Injection
+
+- **ID:** `2601.21182` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Xin Peng, Ang Gao
+- **abs:** https://arxiv.org/abs/2601.21182 · **PDF:** https://arxiv.org/pdf/2601.21182v1
+
+## 18372. MAD: Modality-Adaptive Decoding for Mitigating Cross-Modal Hallucinations in Multimodal Large Language Models
+
+- **ID:** `2601.21181` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Sangyun Chung, Se Yeon Kim, Youngchae Chee, et al.
+- **abs:** https://arxiv.org/abs/2601.21181 · **PDF:** https://arxiv.org/pdf/2601.21181v1
+
+## 18373. Flow Perturbation++: Multi-Step Unbiased Jacobian Estimation for High-Dimensional Boltzmann Sampling
+
+- **ID:** `2601.21177` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Xin Peng, Ang Gao
+- **abs:** https://arxiv.org/abs/2601.21177 · **PDF:** https://arxiv.org/pdf/2601.21177v1
+
+## 18374. Breaking the Reasoning Horizon in Entity Alignment Foundation Models
+
+- **ID:** `2601.21174` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Yuanning Cui, Zequn Sun, Wei Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.21174 · **PDF:** https://arxiv.org/pdf/2601.21174v1
+
+## 18375. AC2L-GAD: Active Counterfactual Contrastive Learning for Graph Anomaly Detection
+
+- **ID:** `2601.21171` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Kamal Berahmand, Saman Forouzandeh, Mehrnoush Mohammadi, et al.
+- **abs:** https://arxiv.org/abs/2601.21171 · **PDF:** https://arxiv.org/pdf/2601.21171v1
+
+## 18376. The Powers of Precision: Structure-Informed Detection in Complex Systems -- From Customer Churn to Seizure Onset
+
+- **ID:** `2601.21170` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Augusto Santos, Teresa Santos, Catarina Rodrigues, et al.
+- **abs:** https://arxiv.org/abs/2601.21170 · **PDF:** https://arxiv.org/pdf/2601.21170v2
+
+## 18377. Output-Space Search: Targeting LLM Generations in a Frozen Encoder-Defined Output Space
+
+- **ID:** `2601.21169` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Tobias Materzok
+- **abs:** https://arxiv.org/abs/2601.21169 · **PDF:** https://arxiv.org/pdf/2601.21169v1
+
+## 18378. Efficient Simple Regret Algorithms for Stochastic Contextual Bandits
+
+- **ID:** `2601.21167` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Shuai Liu, Alireza Bakhtiari, Alex Ayoub, et al.
+- **abs:** https://arxiv.org/abs/2601.21167 · **PDF:** https://arxiv.org/pdf/2601.21167v1
+
+## 18379. FrontierScience: Evaluating AI's Ability to Perform Expert-Level Scientific Tasks
+
+- **ID:** `2601.21165` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Miles Wang, Robi Lin, Kat Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.21165 · **PDF:** https://arxiv.org/pdf/2601.21165v1
+
+## 18380. Concise Geometric Description as a Bridge: Unleashing the Potential of LLM for Plane Geometry Problem Solving
+
+- **ID:** `2601.21164` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Jingyun Wang, Dian Li, Xiaohan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21164 · **PDF:** https://arxiv.org/pdf/2601.21164v2
+
+## 18381. A2RAG: Adaptive Agentic Graph Retrieval for Cost-Aware and Reliable Reasoning
+
+- **ID:** `2601.21162` · **primary:** `cs.IR` · **Submitted:** 2026-01-29
+- **Authors:** Jiate Liu, Zebin Chen, Shaobo Qiao, et al.
+- **abs:** https://arxiv.org/abs/2601.21162 · **PDF:** https://arxiv.org/pdf/2601.21162v1
+
+## 18382. A Federated Generalized Expectation-Maximization Algorithm for Mixture Models with an Unknown Number of Components
+
+- **ID:** `2601.21160` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Michael Ibrahim, Nagi Gebraeel, Weijun Xie
+- **abs:** https://arxiv.org/abs/2601.21160 · **PDF:** https://arxiv.org/pdf/2601.21160v1
+
+## 18383. Bridging the Arithmetic Gap: The Cognitive Complexity Benchmark and Financial-PoT for Robust Financial Reasoning
+
+- **ID:** `2601.21157` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Boxiang Zhao, Qince Li, Zhonghao Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21157 · **PDF:** https://arxiv.org/pdf/2601.21157v1
+
+## 18384. Comparison of Image Processing Models in Quark Gluon Jet Classification
+
+- **ID:** `2602.00141` · **primary:** `physics.data-an` · **Submitted:** 2026-01-29
+- **Authors:** Daeun Kim, Jiwon Lee, Wonjun Jeong, et al.
+- **abs:** https://arxiv.org/abs/2602.00141 · **PDF:** https://arxiv.org/pdf/2602.00141v1
+
+## 18385. A new strategy for finite-sample valid prediction of future insurance claims in the regression setting
+
+- **ID:** `2601.21153` · **primary:** `stat.AP` · **Submitted:** 2026-01-29
+- **Authors:** Liang Hong
+- **abs:** https://arxiv.org/abs/2601.21153 · **PDF:** https://arxiv.org/pdf/2601.21153v1
+
+## 18386. Learning to Advect: A Neural Semi-Lagrangian Architecture for Weather Forecasting
+
+- **ID:** `2601.21151` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Carlos A. Pereira, Stéphane Gaudreault, Valentin Dallerit, et al.
+- **abs:** https://arxiv.org/abs/2601.21151 · **PDF:** https://arxiv.org/pdf/2601.21151v1
+
+## 18387. Can Neural Networks Learn Small Algebraic Worlds? An Investigation Into the Group-theoretic Structures Learned By Narrow Models Trained To Predict Group Operations
+
+- **ID:** `2601.21150` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Henry Kvinge, Andrew Aguilar, Nayda Farnsworth, et al.
+- **abs:** https://arxiv.org/abs/2601.21150 · **PDF:** https://arxiv.org/pdf/2601.21150v1
+
+## 18388. Mobility-Embedded POIs: Learning What A Place Is and How It Is Used from Human Movement
+
+- **ID:** `2601.21149` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Maria Despoina Siampou, Shushman Choudhury, Shang-Ling Hsu, et al.
+- **abs:** https://arxiv.org/abs/2601.21149 · **PDF:** https://arxiv.org/pdf/2601.21149v2
+
+## 18389. BrainStack: Neuro-MoE with Functionally Guided Expert Routing for EEG-Based Language Decoding
+
+- **ID:** `2601.21148` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Ziyi Zhao, Jinzhao Zhou, Xiaowei Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.21148 · **PDF:** https://arxiv.org/pdf/2601.21148v1
+
+## 18390. Smooth Dynamic Cutoffs for Machine Learning Interatomic Potentials
+
+- **ID:** `2601.21147` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Kevin Han, Haolin Cong, Bowen Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.21147 · **PDF:** https://arxiv.org/pdf/2601.21147v1
+
+## 18391. Maxwait: A Generalized Mechanism for Distributed Time-Sensitive Systems
+
+- **ID:** `2601.21146` · **primary:** `cs.DC` · **Submitted:** 2026-01-29
+- **Authors:** Francesco Paladino, Shulu Li, Edward A. Lee
+- **abs:** https://arxiv.org/abs/2601.21146 · **PDF:** https://arxiv.org/pdf/2601.21146v1
+
+## 18392. Optimization and Mobile Deployment for Anthropocene Neural Style Transfer
+
+- **ID:** `2601.21141` · **primary:** `cs.HC` · **Submitted:** 2026-01-29
+- **Authors:** Po-Hsun Chen, Ivan C. H. Liu
+- **abs:** https://arxiv.org/abs/2601.21141 · **PDF:** https://arxiv.org/pdf/2601.21141v1
+
+## 18393. EnsembleLink: Accurate Record Linkage Without Training Data
+
+- **ID:** `2601.21138` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Noah Dasanaike
+- **abs:** https://arxiv.org/abs/2601.21138 · **PDF:** https://arxiv.org/pdf/2601.21138v2
+
+## 18394. TRACE: Trajectory Recovery for Continuous Mechanism Evolution in Causal Representation Learning
+
+- **ID:** `2601.21135` · **primary:** `cs.LG` · **Submitted:** 2026-01-29
+- **Authors:** Shicheng Fan, Kun Zhang, Lu Cheng
+- **abs:** https://arxiv.org/abs/2601.21135 · **PDF:** https://arxiv.org/pdf/2601.21135v1
+
+## 18395. Large Language Models Naively Recover Ethnicity from Individual Records
+
+- **ID:** `2601.21132` · **primary:** `cs.CL` · **Submitted:** 2026-01-29
+- **Authors:** Noah Dasanaike
+- **abs:** https://arxiv.org/abs/2601.21132 · **PDF:** https://arxiv.org/pdf/2601.21132v1
+
+## 18396. Thompson sampling: Precise arm-pull dynamics and adaptive inference
+
+- **ID:** `2601.21131` · **primary:** `math.ST` · **Submitted:** 2026-01-29
+- **Authors:** Qiyang Han
+- **abs:** https://arxiv.org/abs/2601.21131 · **PDF:** https://arxiv.org/pdf/2601.21131v1
+
+## 18397. What You Feel Is Not What They See: On Predicting Self-Reported Emotion from Third-Party Observer Labels
+
+- **ID:** `2601.21130` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Yara El-Tawil, Aneesha Sampath, Emily Mower Provost
+- **abs:** https://arxiv.org/abs/2601.21130 · **PDF:** https://arxiv.org/pdf/2601.21130v1
+
+## 18398. Beyond a Single Reference: Training and Evaluation with Paraphrases in Sign Language Translation
+
+- **ID:** `2601.21128` · **primary:** `cs.AI` · **Submitted:** 2026-01-29
+- **Authors:** Václav Javorek, Tomáš Železný, Alessa Carbo, et al.
+- **abs:** https://arxiv.org/abs/2601.21128 · **PDF:** https://arxiv.org/pdf/2601.21128v1
+
+## 18399. PhaseCoder: Microphone Geometry-Agnostic Spatial Audio Understanding for Multimodal LLMs
+
+- **ID:** `2601.21124` · **primary:** `cs.SD` · **Submitted:** 2026-01-28
+- **Authors:** Artem Dementyev, Wazeer Zulfikar, Sinan Hersek, et al.
+- **abs:** https://arxiv.org/abs/2601.21124 · **PDF:** https://arxiv.org/pdf/2601.21124v1
+
+## 18400. CUA-Skill: Develop Skills for Computer Using Agent
+
+- **ID:** `2601.21123` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Tianyi Chen, Yinheng Li, Michael Solodko, et al.
+- **abs:** https://arxiv.org/abs/2601.21123 · **PDF:** https://arxiv.org/pdf/2601.21123v2
+
+## 18401. AI-Assisted Engineering Should Track the Epistemic Status and Temporal Validity of Architectural Decisions
+
+- **ID:** `2601.21116` · **primary:** `cs.SE` · **Submitted:** 2026-01-28
+- **Authors:** Sankalp Gilda, Shlok Gilda
+- **abs:** https://arxiv.org/abs/2601.21116 · **PDF:** https://arxiv.org/pdf/2601.21116v1
+
+## 18402. Multi-task Code LLMs: Data Mix or Model Merge?
+
+- **ID:** `2601.21115` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Mingzhi Zhu, Boris Sobolev, Rahul Krishna, et al.
+- **abs:** https://arxiv.org/abs/2601.21115 · **PDF:** https://arxiv.org/pdf/2601.21115v1
+
+## 18403. Planner-Auditor Twin: Agentic Discharge Planning with FHIR-Based LLM Planning, Guideline Recall, Optional Caching and Self-Improvement
+
+- **ID:** `2601.21113` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Kaiyuan Wu, Aditya Nagori, Rishikesan Kamaleswaran
+- **abs:** https://arxiv.org/abs/2601.21113 · **PDF:** https://arxiv.org/pdf/2601.21113v1
+
+## 18404. How does information access affect LLM monitors' ability to detect sabotage?
+
+- **ID:** `2601.21112` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Rauno Arike, Raja Mehta Moreno, Rohan Subramani, et al.
+- **abs:** https://arxiv.org/abs/2601.21112 · **PDF:** https://arxiv.org/pdf/2601.21112v2
+
+## 18405. ChunkWise LoRA: Adaptive Sequence Partitioning for Memory-Efficient Low-Rank Adaptation and Accelerated LLM Inference
+
+- **ID:** `2601.21109` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Ketan Thakkar, Maitreyi Chatterjee, Ramasubramanian Balasubramanian, et al.
+- **abs:** https://arxiv.org/abs/2601.21109 · **PDF:** https://arxiv.org/pdf/2601.21109v1
+
+## 18406. SteerEval: A Framework for Evaluating Steerability with Natural Language Profiles for Recommendation
+
+- **ID:** `2601.21105` · **primary:** `cs.IR` · **Submitted:** 2026-01-28
+- **Authors:** Joyce Zhou, Weijie Zhou, Doug Turnbull, et al.
+- **abs:** https://arxiv.org/abs/2601.21105 · **PDF:** https://arxiv.org/pdf/2601.21105v1
+
+## 18407. Multilevel and Sequential Monte Carlo for Training-Free Diffusion Guidance
+
+- **ID:** `2601.21104` · **primary:** `stat.ML` · **Submitted:** 2026-01-28
+- **Authors:** Aidan Gleich, Scott C. Schmidler
+- **abs:** https://arxiv.org/abs/2601.21104 · **PDF:** https://arxiv.org/pdf/2601.21104v1
+
+## 18408. Magellan: Autonomous Discovery of Novel Compiler Optimization Heuristics with AlphaEvolve
+
+- **ID:** `2601.21096` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Hongzheng Chen, Alexander Novikov, Ngân Vũ, et al.
+- **abs:** https://arxiv.org/abs/2601.21096 · **PDF:** https://arxiv.org/pdf/2601.21096v1
+
+## 18409. Responsible AI: The Good, The Bad, The AI
+
+- **ID:** `2601.21095` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Akbar Anbar Jafari, Cagri Ozcinar, Gholamreza Anbarjafari
+- **abs:** https://arxiv.org/abs/2601.21095 · **PDF:** https://arxiv.org/pdf/2601.21095v1
+
+## 18410. Safety Generalization Under Distribution Shift in Safe Reinforcement Learning: A Diabetes Testbed
+
+- **ID:** `2601.21094` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Minjae Kwon, Josephine Lamp, Lu Feng
+- **abs:** https://arxiv.org/abs/2601.21094 · **PDF:** https://arxiv.org/pdf/2601.21094v1
+
+## 18411. High-dimensional learning dynamics of multi-pass Stochastic Gradient Descent in multi-index models
+
+- **ID:** `2601.21093` · **primary:** `stat.ML` · **Submitted:** 2026-01-28
+- **Authors:** Zhou Fan, Leda Wang
+- **abs:** https://arxiv.org/abs/2601.21093 · **PDF:** https://arxiv.org/pdf/2601.21093v2
+
+## 18412. MapPFN: Learning Causal Perturbation Maps in Context
+
+- **ID:** `2601.21092` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Marvin Sextro, Weronika Kłos, Gabriel Dernbach
+- **abs:** https://arxiv.org/abs/2601.21092 · **PDF:** https://arxiv.org/pdf/2601.21092v1
+
+## 18413. Deep Reinforcement Learning for Fault-Adaptive Routing in Eisenstein-Jacobi Interconnection Topologies
+
+- **ID:** `2601.21090` · **primary:** `cs.DC` · **Submitted:** 2026-01-28
+- **Authors:** Mohammad Walid Charrwi, Zaid Hussain
+- **abs:** https://arxiv.org/abs/2601.21090 · **PDF:** https://arxiv.org/pdf/2601.21090v1
+
+## 18414. An efficient, accurate, and interpretable machine learning method for computing probability of failure
+
+- **ID:** `2601.21089` · **primary:** `stat.ML` · **Submitted:** 2026-01-28
+- **Authors:** Jacob Zhu, Donald Estep
+- **abs:** https://arxiv.org/abs/2601.21089 · **PDF:** https://arxiv.org/pdf/2601.21089v1
+
+## 18415. Position-invariant Fine-tuning of Speech Enhancement Models with Self-supervised Speech Representations
+
+- **ID:** `2601.21084` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Amit Meghanani, Thomas Hain
+- **abs:** https://arxiv.org/abs/2601.21084 · **PDF:** https://arxiv.org/pdf/2601.21084v1
+
+## 18416. OpenSec: Measuring Incident Response Agent Calibration Under Adversarial Evidence
+
+- **ID:** `2601.21083` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Jarrod Barnes
+- **abs:** https://arxiv.org/abs/2601.21083 · **PDF:** https://arxiv.org/pdf/2601.21083v3
+
+## 18417. LOCUS: Low-Dimensional Model Embeddings for Efficient Model Exploration, Comparison, and Selection
+
+- **ID:** `2601.21082` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Shivam Patel, William Cocke, Gauri Joshi
+- **abs:** https://arxiv.org/abs/2601.21082 · **PDF:** https://arxiv.org/pdf/2601.21082v1
+
+## 18418. Parametric Hyperbolic Conservation Laws: A Unified Framework for Conservation, Entropy Stability, and Hyperbolicity
+
+- **ID:** `2601.21080` · **primary:** `math.NA` · **Submitted:** 2026-01-28
+- **Authors:** Lizuo Liu, Lu Zhang, Anne Gelb
+- **abs:** https://arxiv.org/abs/2601.21080 · **PDF:** https://arxiv.org/pdf/2601.21080v1
+
+## 18419. Multi-modal Imputation for Alzheimer's Disease Classification
+
+- **ID:** `2601.21076` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Abhijith Shaji, Tamoghna Chattopadhyay, Sophia I. Thomopoulos, et al.
+- **abs:** https://arxiv.org/abs/2601.21076 · **PDF:** https://arxiv.org/pdf/2601.21076v1
+
+## 18420. Towards Comprehensive Benchmarking Infrastructure for LLMs In Software Engineering
+
+- **ID:** `2601.21070` · **primary:** `cs.SE` · **Submitted:** 2026-01-28
+- **Authors:** Daniel Rodriguez-Cardenas, Xiaochang Li, Marcos Macedo, et al.
+- **abs:** https://arxiv.org/abs/2601.21070 · **PDF:** https://arxiv.org/pdf/2601.21070v1
+
+## 18421. Out-of-Distribution Generalization in Graph Foundation Models
+
+- **ID:** `2601.21067` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Haoyang Li, Haibo Chen, Xin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.21067 · **PDF:** https://arxiv.org/pdf/2601.21067v1
+
+## 18422. Textual Equilibrium Propagation for Deep Compound AI Systems
+
+- **ID:** `2601.21064` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Minghui Chen, Wenlong Deng, James Zou, et al.
+- **abs:** https://arxiv.org/abs/2601.21064 · **PDF:** https://arxiv.org/pdf/2601.21064v2
+
+## 18423. Signal from Structure: Exploiting Submodular Upper Bounds in Generative Flow Networks
+
+- **ID:** `2601.21061` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Alexandre Larouche, Audrey Durand
+- **abs:** https://arxiv.org/abs/2601.21061 · **PDF:** https://arxiv.org/pdf/2601.21061v1
+
+## 18424. Human-LLM Collaborative Feature Engineering for Tabular Data
+
+- **ID:** `2601.21060` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Zhuoyan Li, Aditya Bansal, Jinzhao Li, et al.
+- **abs:** https://arxiv.org/abs/2601.21060 · **PDF:** https://arxiv.org/pdf/2601.21060v1
+
+## 18425. Snowball: A Scalable All-to-All Ising Machine with Dual-Mode Markov Chain Monte Carlo Spin Selection and Asynchronous Spin Updates for Fast Combinatorial Optimization
+
+- **ID:** `2601.21058` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Seungki Hong, Kyeongwon Jeong, Taekwang Jang
+- **abs:** https://arxiv.org/abs/2601.21058 · **PDF:** https://arxiv.org/pdf/2601.21058v1
+
+## 18426. Better without U: Impact of Selective Hubbard U Correction on Foundational MLIPs
+
+- **ID:** `2601.21056` · **primary:** `physics.chem-ph` · **Submitted:** 2026-01-28
+- **Authors:** Thomas Warford, Fabian L. Thiemann, Gábor Csányi
+- **abs:** https://arxiv.org/abs/2601.21056 · **PDF:** https://arxiv.org/pdf/2601.21056v1
+
+## 18427. Llama-3.1-FoundationAI-SecurityLLM-Reasoning-8B Technical Report
+
+- **ID:** `2601.21051` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Zhuoran Yang, Ed Li, Jianliang He, et al.
+- **abs:** https://arxiv.org/abs/2601.21051 · **PDF:** https://arxiv.org/pdf/2601.21051v1
+
+## 18428. SMKC: Sketch Based Kernel Correlation Images for Variable Cardinality Time Series Anomaly Detection
+
+- **ID:** `2601.21050` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Haokun Zhou
+- **abs:** https://arxiv.org/abs/2601.21050 · **PDF:** https://arxiv.org/pdf/2601.21050v1
+
+## 18429. QUARK: Robust Retrieval under Non-Faithful Queries via Query-Anchored Aggregation
+
+- **ID:** `2601.21049` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Rita Qiuran Lyu, Michelle Manqiao Wang, Lei Shi
+- **abs:** https://arxiv.org/abs/2601.21049 · **PDF:** https://arxiv.org/pdf/2601.21049v1
+
+## 18430. Test-Time Adaptation for Unsupervised Combinatorial Optimization
+
+- **ID:** `2601.21048` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Yiqiao Liao, Farinaz Koushanfar, Parinaz Naghizadeh
+- **abs:** https://arxiv.org/abs/2601.21048 · **PDF:** https://arxiv.org/pdf/2601.21048v1
+
+## 18431. Log2Motion: Biomechanical Motion Synthesis from Touch Logs
+
+- **ID:** `2601.21043` · **primary:** `cs.HC` · **Submitted:** 2026-01-28
+- **Authors:** Michał Patryk Miazga, Hannah Bussmann, Antti Oulasvirta, et al.
+- **abs:** https://arxiv.org/abs/2601.21043 · **PDF:** https://arxiv.org/pdf/2601.21043v2
+
+## 18432. Thinking in Frames: How Visual Context and Test-Time Scaling Empower Video Reasoning
+
+- **ID:** `2601.21037` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Chengzu Li, Zanyi Wang, Jiaang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.21037 · **PDF:** https://arxiv.org/pdf/2601.21037v1
+
+## 18433. Predict-Project-Renoise: Sampling Diffusion Models under Hard Constraints
+
+- **ID:** `2601.21033` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Omer Rochman-Sharabi, Gilles Louppe
+- **abs:** https://arxiv.org/abs/2601.21033 · **PDF:** https://arxiv.org/pdf/2601.21033v1
+
+## 18434. SIGMA-PPG: Statistical-prior Informed Generative Masking Architecture for PPG Foundation Model
+
+- **ID:** `2601.21031` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Zongheng Guo, Tao Chen, Yang Jiao, et al.
+- **abs:** https://arxiv.org/abs/2601.21031 · **PDF:** https://arxiv.org/pdf/2601.21031v1
+
+## 18435. "Unlimited Realm of Exploration and Experimentation": Methods and Motivations of AI-Generated Sexual Content Creators
+
+- **ID:** `2601.21028` · **primary:** `cs.CY` · **Submitted:** 2026-01-28
+- **Authors:** Jaron Mink, Lucy Qin, Elissa M. Redmiles
+- **abs:** https://arxiv.org/abs/2601.21028 · **PDF:** https://arxiv.org/pdf/2601.21028v1
+
+## 18436. Diffusion-based Annealed Boltzmann Generators : benefits, pitfalls and hopes
+
+- **ID:** `2601.21026` · **primary:** `stat.ML` · **Submitted:** 2026-01-28
+- **Authors:** Louis Grenioux, Maxence Noble
+- **abs:** https://arxiv.org/abs/2601.21026 · **PDF:** https://arxiv.org/pdf/2601.21026v1
+
+## 18437. A Diffusive Classification Loss for Learning Energy-based Generative Models
+
+- **ID:** `2601.21025` · **primary:** `stat.ML` · **Submitted:** 2026-01-28
+- **Authors:** RuiKang OuYang, Louis Grenioux, José Miguel Hernández-Lobato
+- **abs:** https://arxiv.org/abs/2601.21025 · **PDF:** https://arxiv.org/pdf/2601.21025v2
+
+## 18438. Conditional Denoising Model as a Physical Surrogate Model
+
+- **ID:** `2601.21021` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** José Afonso, Pedro Viegas, Rodrigo Ventura, et al.
+- **abs:** https://arxiv.org/abs/2601.21021 · **PDF:** https://arxiv.org/pdf/2601.21021v1
+
+## 18439. Reducing Text Bias in Synthetically Generated MCQAs for VLMs in Autonomous Driving
+
+- **ID:** `2602.17677` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Sutej Kulgod, Sean Ye, Sanchit Tanwar, et al.
+- **abs:** https://arxiv.org/abs/2602.17677 · **PDF:** https://arxiv.org/pdf/2602.17677v1
+
+## 18440. Towards regularized learning from functional data with covariate shift
+
+- **ID:** `2601.21019` · **primary:** `math.ST` · **Submitted:** 2026-01-28
+- **Authors:** Markus Holzleitner, Sergiy Pereverzyev, Sergei V. Pereverzyev, et al.
+- **abs:** https://arxiv.org/abs/2601.21019 · **PDF:** https://arxiv.org/pdf/2601.21019v1
+
+## 18441. Unplugging a Seemingly Sentient Machine Is the Rational Choice -- A Metaphysical Perspective
+
+- **ID:** `2601.21016` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Erik J Bekkers, Anna Ciaunica
+- **abs:** https://arxiv.org/abs/2601.21016 · **PDF:** https://arxiv.org/pdf/2601.21016v1
+
+## 18442. Efficient Causal Structure Learning via Modular Subgraph Integration
+
+- **ID:** `2601.21014` · **primary:** `stat.ML` · **Submitted:** 2026-01-28
+- **Authors:** Haixiang Sun, Pengchao Tian, Zihan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.21014 · **PDF:** https://arxiv.org/pdf/2601.21014v2
+
+## 18443. Order-Aware Test-Time Adaptation: Leveraging Temporal Dynamics for Robust Streaming Inference
+
+- **ID:** `2601.21012` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Young Kyung Kim, Oded Schlesinger, Qiangqiang Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.21012 · **PDF:** https://arxiv.org/pdf/2601.21012v1
+
+## 18444. Solver-in-the-Loop: MDP-Based Benchmarks for Self-Correction and Behavioral Rationality in Operations Research
+
+- **ID:** `2601.21008` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Ruicheng Ao, David Simchi-Levi, Xinshang Wang
+- **abs:** https://arxiv.org/abs/2601.21008 · **PDF:** https://arxiv.org/pdf/2601.21008v2
+
+## 18445. Bayesian-LoRA: Probabilistic Low-Rank Adaptation of Large Language Models
+
+- **ID:** `2601.21003` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Moule Lin, Shuhao Guan, Andrea Patane, et al.
+- **abs:** https://arxiv.org/abs/2601.21003 · **PDF:** https://arxiv.org/pdf/2601.21003v1
+
+## 18446. UrduBench: An Urdu Reasoning Benchmark using Contextually Ensembled Translations with Human-in-the-Loop
+
+- **ID:** `2601.21000` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Muhammad Ali Shafique, Areej Mehboob, Layba Fiaz, et al.
+- **abs:** https://arxiv.org/abs/2601.21000 · **PDF:** https://arxiv.org/pdf/2601.21000v1
+
+## 18447. MADE: Benchmark Environments for Closed-Loop Materials Discovery
+
+- **ID:** `2601.20996` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Shreshth A Malik, Tiarnan Doherty, Panagiotis Tigas, et al.
+- **abs:** https://arxiv.org/abs/2601.20996 · **PDF:** https://arxiv.org/pdf/2601.20996v1
+
+## 18448. The Depth Delusion: Why Transformers Should Be Wider, Not Deeper
+
+- **ID:** `2601.20994` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Md Muhtasim Munif Fahim, Md Rezaul Karim
+- **abs:** https://arxiv.org/abs/2601.20994 · **PDF:** https://arxiv.org/pdf/2601.20994v1
+
+## 18449. asr_eval: Algorithms and tools for multi-reference and streaming speech recognition evaluation
+
+- **ID:** `2601.20992` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Oleg Sedukhin, Andrey Kostin
+- **abs:** https://arxiv.org/abs/2601.20992 · **PDF:** https://arxiv.org/pdf/2601.20992v1
+
+## 18450. Top-k on a Budget: Adaptive Ranking with Weak and Strong Oracles
+
+- **ID:** `2601.20989` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Lutz Oettershagen
+- **abs:** https://arxiv.org/abs/2601.20989 · **PDF:** https://arxiv.org/pdf/2601.20989v1
+
+## 18451. Pre-trained Encoders for Global Child Development: Transfer Learning Enables Deployment in Data-Scarce Settings
+
+- **ID:** `2601.20987` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Md Muhtasim Munif Fahim, Md Rezaul Karim
+- **abs:** https://arxiv.org/abs/2601.20987 · **PDF:** https://arxiv.org/pdf/2601.20987v1
+
+## 18452. Distributional Active Inference
+
+- **ID:** `2601.20985` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Abdullah Akgül, Gulcin Baykal, Manuel Haußmann, et al.
+- **abs:** https://arxiv.org/abs/2601.20985 · **PDF:** https://arxiv.org/pdf/2601.20985v1
+
+## 18453. Monotone Optimisation with Learned Projections
+
+- **ID:** `2601.20983` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Ahmed Rashwan, Keith Briggs, Chris Budd, et al.
+- **abs:** https://arxiv.org/abs/2601.20983 · **PDF:** https://arxiv.org/pdf/2601.20983v1
+
+## 18454. Diversifying Toxicity Search in Large Language Models Through Speciation
+
+- **ID:** `2601.20981` · **primary:** `cs.NE` · **Submitted:** 2026-01-28
+- **Authors:** Onkar Shelar, Travis Desell
+- **abs:** https://arxiv.org/abs/2601.20981 · **PDF:** https://arxiv.org/pdf/2601.20981v1
+
+## 18455. DeepSearchQA: Bridging the Comprehensiveness Gap for Deep Research Agents
+
+- **ID:** `2601.20975` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Nikita Gupta, Riju Chatterjee, Lukas Haas, et al.
+- **abs:** https://arxiv.org/abs/2601.20975 · **PDF:** https://arxiv.org/pdf/2601.20975v1
+
+## 18456. An Explainable Failure Prediction Framework for Neural Networks in Radio Access Networks
+
+- **ID:** `2602.13231` · **primary:** `cs.NI` · **Submitted:** 2026-01-28
+- **Authors:** Khaleda Papry, Francesco Spinnato, Marco Fiore, et al.
+- **abs:** https://arxiv.org/abs/2602.13231 · **PDF:** https://arxiv.org/pdf/2602.13231v1
+
+## 18457. The augmented NLP bound for maximum-entropy remote sampling
+
+- **ID:** `2601.20970` · **primary:** `math.OC` · **Submitted:** 2026-01-28
+- **Authors:** Gabriel Ponte, Marcia Fampa, Jon Lee
+- **abs:** https://arxiv.org/abs/2601.20970 · **PDF:** https://arxiv.org/pdf/2601.20970v2
+
+## 18458. The Epistemic Planning Domain Definition Language: Official Guideline
+
+- **ID:** `2601.20969` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Alessandro Burigana, Francesco Fabiano
+- **abs:** https://arxiv.org/abs/2601.20969 · **PDF:** https://arxiv.org/pdf/2601.20969v2
+
+## 18459. A Theory of Universal Agnostic Learning
+
+- **ID:** `2601.20961` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Steve Hanneke, Shay Moran
+- **abs:** https://arxiv.org/abs/2601.20961 · **PDF:** https://arxiv.org/pdf/2601.20961v1
+
+## 18460. Parametric Quantum State Tomography with HyperRBMs
+
+- **ID:** `2601.20950` · **primary:** `quant-ph` · **Submitted:** 2026-01-28
+- **Authors:** Simon Tonner, Viet T. Tran, Richard Kueng
+- **abs:** https://arxiv.org/abs/2601.20950 · **PDF:** https://arxiv.org/pdf/2601.20950v1
+
+## 18461. Evolutionary Strategies lead to Catastrophic Forgetting in LLMs
+
+- **ID:** `2601.20861` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Immanuel Abdi, Akshat Gupta, Micah Mok, et al.
+- **abs:** https://arxiv.org/abs/2601.20861 · **PDF:** https://arxiv.org/pdf/2601.20861v1
+
+## 18462. When Flores Bloomz Wrong: Cross-Direction Contamination in Machine Translation Evaluation
+
+- **ID:** `2601.20858` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** David Tan, Pinzhen Chen, Josef van Genabith, et al.
+- **abs:** https://arxiv.org/abs/2601.20858 · **PDF:** https://arxiv.org/pdf/2601.20858v1
+
+## 18463. SokoBench: Evaluating Long-Horizon Planning and Reasoning in Large Language Models
+
+- **ID:** `2601.20856` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Sebastiano Monti, Carlo Nicolini, Gianni Pellegrini, et al.
+- **abs:** https://arxiv.org/abs/2601.20856 · **PDF:** https://arxiv.org/pdf/2601.20856v1
+
+## 18464. Exploring Transformer Placement in Variational Autoencoders for Tabular Data Generation
+
+- **ID:** `2601.20854` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Aníbal Silva, Moisés Santos, André Restivo, et al.
+- **abs:** https://arxiv.org/abs/2601.20854 · **PDF:** https://arxiv.org/pdf/2601.20854v1
+
+## 18465. C3Box: A CLIP-based Class-Incremental Learning Toolbox
+
+- **ID:** `2601.20852` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Hao Sun, Da-Wei Zhou
+- **abs:** https://arxiv.org/abs/2601.20852 · **PDF:** https://arxiv.org/pdf/2601.20852v1
+
+## 18466. Do LLMs Favor LLMs? Quantifying Interaction Effects in Peer Review
+
+- **ID:** `2601.20920` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Vibhhu Sharma, Thorsten Joachims, Sarah Dean
+- **abs:** https://arxiv.org/abs/2601.20920 · **PDF:** https://arxiv.org/pdf/2601.20920v1
+
+## 18467. Post-Training Fairness Control: A Single-Train Framework for Dynamic Fairness in Recommendation
+
+- **ID:** `2601.20848` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Weixin Chen, Li Chen, Yuhan Zhao
+- **abs:** https://arxiv.org/abs/2601.20848 · **PDF:** https://arxiv.org/pdf/2601.20848v1
+
+## 18468. A New Dataset and Framework for Robust Road Surface Classification via Camera-IMU Fusion
+
+- **ID:** `2601.20847` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Willams de Lima Costa, Thifany Ketuli Silva de Souza, Jonas Ferreira Silva, et al.
+- **abs:** https://arxiv.org/abs/2601.20847 · **PDF:** https://arxiv.org/pdf/2601.20847v2
+
+## 18469. PatchFormer: A Patch-Based Time Series Foundation Model with Hierarchical Masked Reconstruction and Cross-Domain Transfer Learning for Zero-Shot Multi-Horizon Forecasting
+
+- **ID:** `2601.20845` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Olaf Yunus Laitinen Imanov, Derya Umut Kulali, Taner Yilmaz
+- **abs:** https://arxiv.org/abs/2601.20845 · **PDF:** https://arxiv.org/pdf/2601.20845v1
+
+## 18470. $\mathbb{R}^{2k}$ is Theoretically Large Enough for Embedding-based Top-$k$ Retrieval
+
+- **ID:** `2601.20844` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Zihao Wang, Hang Yin, Lihui Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.20844 · **PDF:** https://arxiv.org/pdf/2601.20844v2
+
+## 18471. Deep Researcher with Sequential Plan Reflection and Candidates Crossover (Deep Researcher Reflect Evolve)
+
+- **ID:** `2601.20843` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Saurav Prateek
+- **abs:** https://arxiv.org/abs/2601.20843 · **PDF:** https://arxiv.org/pdf/2601.20843v1
+
+## 18472. Reward Models Inherit Value Biases from Pretraining
+
+- **ID:** `2601.20838` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Brian Christian, Jessica A. F. Thompson, Elle Michelle Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.20838 · **PDF:** https://arxiv.org/pdf/2601.20838v2
+
+## 18473. Open-Vocabulary Functional 3D Human-Scene Interaction Generation
+
+- **ID:** `2601.20835` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Jie Liu, Yu Sun, Alpar Cseke, et al.
+- **abs:** https://arxiv.org/abs/2601.20835 · **PDF:** https://arxiv.org/pdf/2601.20835v2
+
+## 18474. Linear representations in language models can change dramatically over a conversation
+
+- **ID:** `2601.20834` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Andrew Kyle Lampinen, Yuxuan Li, Eghbal Hosseini, et al.
+- **abs:** https://arxiv.org/abs/2601.20834 · **PDF:** https://arxiv.org/pdf/2601.20834v2
+
+## 18475. MemCtrl: Using MLLMs as Active Memory Controllers on Embodied Agents
+
+- **ID:** `2601.20831` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Vishnu Sashank Dorbala, Dinesh Manocha
+- **abs:** https://arxiv.org/abs/2601.20831 · **PDF:** https://arxiv.org/pdf/2601.20831v1
+
+## 18476. VSCOUT: A Hybrid Variational Autoencoder Approach to Outlier Detection in High-Dimensional Retrospective Monitoring
+
+- **ID:** `2601.20830` · **primary:** `stat.ML` · **Submitted:** 2026-01-28
+- **Authors:** Waldyn G. Martinez
+- **abs:** https://arxiv.org/abs/2601.20830 · **PDF:** https://arxiv.org/pdf/2601.20830v1
+
+## 18477. Training Reasoning Models on Saturated Problems via Failure-Prefix Conditioning
+
+- **ID:** `2601.20829` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Minwu Kim, Safal Shrestha, Keith Ross
+- **abs:** https://arxiv.org/abs/2601.20829 · **PDF:** https://arxiv.org/pdf/2601.20829v1
+
+## 18478. Demystifying Prediction Powered Inference
+
+- **ID:** `2601.20819` · **primary:** `stat.ML` · **Submitted:** 2026-01-28
+- **Authors:** Yilin Song, Dan M. Kluger, Harsh Parikh, et al.
+- **abs:** https://arxiv.org/abs/2601.20819 · **PDF:** https://arxiv.org/pdf/2601.20819v1
+
+## 18479. Noninvasive Intracranial Pressure Estimation Using Subspace System Identification and Bespoke Machine Learning Algorithms: A Learning-to-Rank Approach
+
+- **ID:** `2601.20916` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Anni Zhao, Ayca Ermis, Jeffrey Robert Vitt, et al.
+- **abs:** https://arxiv.org/abs/2601.20916 · **PDF:** https://arxiv.org/pdf/2601.20916v2
+
+## 18480. GNN Explanations that do not Explain and How to find Them
+
+- **ID:** `2601.20815` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Steve Azzolin, Stefano Teso, Bruno Lepri, et al.
+- **abs:** https://arxiv.org/abs/2601.20815 · **PDF:** https://arxiv.org/pdf/2601.20815v3
+
+## 18481. Noisy but Valid: Robust Statistical Evaluation of LLMs with Imperfect Judges
+
+- **ID:** `2601.20913` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Chen Feng, Minghe Shen, Ananth Balashankar, et al.
+- **abs:** https://arxiv.org/abs/2601.20913 · **PDF:** https://arxiv.org/pdf/2601.20913v1
+
+## 18482. Context-Augmented Code Generation Using Programming Knowledge Graphs
+
+- **ID:** `2601.20810` · **primary:** `cs.SE` · **Submitted:** 2026-01-28
+- **Authors:** Shahd Seddik, Fahd Seddik, Iman Saberi, et al.
+- **abs:** https://arxiv.org/abs/2601.20810 · **PDF:** https://arxiv.org/pdf/2601.20810v1
+
+## 18483. Structured Semantic Information Helps Retrieve Better Examples for In-Context Learning in Few-Shot Relation Extraction
+
+- **ID:** `2601.20803` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Aunabil Chakma, Mihai Surdeanu, Eduardo Blanco
+- **abs:** https://arxiv.org/abs/2601.20803 · **PDF:** https://arxiv.org/pdf/2601.20803v1
+
+## 18484. Reinforcement Learning via Self-Distillation
+
+- **ID:** `2601.20802` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Jonas Hübotter, Frederike Lübeck, Lejs Behric, et al.
+- **abs:** https://arxiv.org/abs/2601.20802 · **PDF:** https://arxiv.org/pdf/2601.20802v2
+
+## 18485. Spatiotemporal Robustness of Temporal Logic Tasks using Multi-Objective Reasoning
+
+- **ID:** `2603.29868` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Oliver Schön, Lars Lindemann
+- **abs:** https://arxiv.org/abs/2603.29868 · **PDF:** https://arxiv.org/pdf/2603.29868v1
+
+## 18486. Conditional PED-ANOVA: Hyperparameter Importance in Hierarchical & Dynamic Search Spaces
+
+- **ID:** `2601.20800` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Kaito Baba, Yoshihiko Ozaki, Shuhei Watanabe
+- **abs:** https://arxiv.org/abs/2601.20800 · **PDF:** https://arxiv.org/pdf/2601.20800v2
+
+## 18487. Dissecting Multimodal In-Context Learning: Modality Asymmetries and Circuit Dynamics in modern Transformers
+
+- **ID:** `2601.20796` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Yiran Huang, Karsten Roth, Quentin Bouniot, et al.
+- **abs:** https://arxiv.org/abs/2601.20796 · **PDF:** https://arxiv.org/pdf/2601.20796v1
+
+## 18488. Non-Markov Multi-Round Conversational Image Generation with History-Conditioned MLLMs
+
+- **ID:** `2601.20911` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Haochen Zhang, Animesh Sinha, Felix Juefei-Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.20911 · **PDF:** https://arxiv.org/pdf/2601.20911v1
+
+## 18489. Jurisdiction as Structural Barrier: How Privacy Policy Organization May Reduce Visibility of Substantive Disclosures
+
+- **ID:** `2601.20792` · **primary:** `cs.CY` · **Submitted:** 2026-01-28
+- **Authors:** Thomas Brackin
+- **abs:** https://arxiv.org/abs/2601.20792 · **PDF:** https://arxiv.org/pdf/2601.20792v1
+
+## 18490. FAIRT2V: Training-Free Debiasing for Text-to-Video Diffusion Models
+
+- **ID:** `2601.20791` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Haonan Zhong, Wei Song, Tingxu Han, et al.
+- **abs:** https://arxiv.org/abs/2601.20791 · **PDF:** https://arxiv.org/pdf/2601.20791v1
+
+## 18491. INSURE-Dial: A Phase-Aware Conversational Dataset & Benchmark for Compliance Verification and Phase Detection
+
+- **ID:** `2602.18448` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Shubham Kulkarni, Alexander Lyzhov, Preetam Joshi, et al.
+- **abs:** https://arxiv.org/abs/2602.18448 · **PDF:** https://arxiv.org/pdf/2602.18448v2
+
+## 18492. SERA: Soft-Verified Efficient Repository Agents
+
+- **ID:** `2601.20789` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Ethan Shen, Danny Tormoen, Saurabh Shah, et al.
+- **abs:** https://arxiv.org/abs/2601.20789 · **PDF:** https://arxiv.org/pdf/2601.20789v2
+
+## 18493. REASON: Accelerating Probabilistic Logical Reasoning for Scalable Neuro-Symbolic Intelligence
+
+- **ID:** `2601.20784` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Zishen Wan, Che-Kai Liu, Jiayi Qian, et al.
+- **abs:** https://arxiv.org/abs/2601.20784 · **PDF:** https://arxiv.org/pdf/2601.20784v1
+
+## 18494. Neural Quantum States in Mixed Precision
+
+- **ID:** `2601.20782` · **primary:** `quant-ph` · **Submitted:** 2026-01-28
+- **Authors:** Massimo Solinas, Agnes Valenti, Nawaf Bou-Rabee, et al.
+- **abs:** https://arxiv.org/abs/2601.20782 · **PDF:** https://arxiv.org/pdf/2601.20782v1
+
+## 18495. Independence of Approximate Clones
+
+- **ID:** `2601.20779` · **primary:** `cs.GT` · **Submitted:** 2026-01-28
+- **Authors:** Théo Delemazure
+- **abs:** https://arxiv.org/abs/2601.20779 · **PDF:** https://arxiv.org/pdf/2601.20779v1
+
+## 18496. Hierarchical JEPA Meets Predictive Remote Control in Beyond 5G Networks
+
+- **ID:** `2602.07000` · **primary:** `eess.SY` · **Submitted:** 2026-01-28
+- **Authors:** Abanoub M. Girgis, Ibtissam Labriji, Mehdi Bennis
+- **abs:** https://arxiv.org/abs/2602.07000 · **PDF:** https://arxiv.org/pdf/2602.07000v1
+
+## 18497. Active Learning for Decision Trees with Provable Guarantees
+
+- **ID:** `2601.20775` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Arshia Soltani Moakhar, Tanapoom Laoaron, Faraz Ghahremani, et al.
+- **abs:** https://arxiv.org/abs/2601.20775 · **PDF:** https://arxiv.org/pdf/2601.20775v2
+
+## 18498. When More Data Doesn't Help: Limits of Adaptation in Multitask Learning
+
+- **ID:** `2601.20774` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Steve Hanneke, Mingyue Xu
+- **abs:** https://arxiv.org/abs/2601.20774 · **PDF:** https://arxiv.org/pdf/2601.20774v1
+
+## 18499. Smoothing the Black-Box: Signed-Distance Supervision for Black-Box Model Copying
+
+- **ID:** `2601.20773` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Rubén Jiménez, Oriol Pujol
+- **abs:** https://arxiv.org/abs/2601.20773 · **PDF:** https://arxiv.org/pdf/2601.20773v1
+
+## 18500. COMET-SG1: Lightweight Autoregressive Regressor for Edge and Embedded AI
+
+- **ID:** `2601.20772` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Shakhyar Gogoi
+- **abs:** https://arxiv.org/abs/2601.20772 · **PDF:** https://arxiv.org/pdf/2601.20772v1
+
+## 18501. Cross-Country Learning for National Infectious Disease Forecasting Using European Data
+
+- **ID:** `2601.20771` · **primary:** `q-bio.PE` · **Submitted:** 2026-01-28
+- **Authors:** Zacharias Komodromos, Kleanthis Malialis, Artemis Kontou, et al.
+- **abs:** https://arxiv.org/abs/2601.20771 · **PDF:** https://arxiv.org/pdf/2601.20771v1
+
+## 18502. Leveraging Second-Order Curvature for Efficient Learned Image Compression: Theory and Empirical Evidence
+
+- **ID:** `2601.20769` · **primary:** `eess.IV` · **Submitted:** 2026-01-28
+- **Authors:** Yichi Zhang, Fengqing Zhu
+- **abs:** https://arxiv.org/abs/2601.20769 · **PDF:** https://arxiv.org/pdf/2601.20769v2
+
+## 18503. Less is More: Clustered Cross-Covariance Control for Offline RL
+
+- **ID:** `2601.20765` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Nan Qiao, Sheng Yue, Shuning Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.20765 · **PDF:** https://arxiv.org/pdf/2601.20765v2
+
+## 18504. Persona Prompting as a Lens on LLM Social Reasoning
+
+- **ID:** `2601.20757` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Jing Yang, Moritz Hechtbauer, Elisabeth Khalilov, et al.
+- **abs:** https://arxiv.org/abs/2601.20757 · **PDF:** https://arxiv.org/pdf/2601.20757v1
+
+## 18505. Supervised Guidance Training for Infinite-Dimensional Diffusion Models
+
+- **ID:** `2601.20756` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Elizabeth L. Baker, Alexander Denker, Jes Frellsen
+- **abs:** https://arxiv.org/abs/2601.20756 · **PDF:** https://arxiv.org/pdf/2601.20756v1
+
+## 18506. GraphAllocBench: A Flexible Benchmark for Preference-Conditioned Multi-Objective Policy Learning
+
+- **ID:** `2601.20753` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Zhiheng Jiang, Yunzhe Wang, Ryan Marr, et al.
+- **abs:** https://arxiv.org/abs/2601.20753 · **PDF:** https://arxiv.org/pdf/2601.20753v3
+
+## 18507. Toward Effective Multi-Domain Rumor Detection in Social Networks Using Domain-Gated Mixture-of-Experts
+
+- **ID:** `2602.21214` · **primary:** `cs.SI` · **Submitted:** 2026-01-28
+- **Authors:** Mohadeseh Sheikhqoraei, Zainabolhoda Heshmati, Zeinab Rajabi, et al.
+- **abs:** https://arxiv.org/abs/2602.21214 · **PDF:** https://arxiv.org/pdf/2602.21214v1
+
+## 18508. Leveraging Generative AI for Enhancing Domain-Driven Software Design
+
+- **ID:** `2601.20909` · **primary:** `cs.SE` · **Submitted:** 2026-01-28
+- **Authors:** Götz-Henrik Wiegand, Filip Stepniak, Patrick Baier
+- **abs:** https://arxiv.org/abs/2601.20909 · **PDF:** https://arxiv.org/pdf/2601.20909v1
+
+## 18509. Like a Therapist, But Not: Reddit Narratives of AI in Mental Health Contexts
+
+- **ID:** `2601.20747` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Elham Aghakhani, Rezvaneh Rezapour
+- **abs:** https://arxiv.org/abs/2601.20747 · **PDF:** https://arxiv.org/pdf/2601.20747v1
+
+## 18510. HESTIA: A Hessian-Guided Differentiable Quantization-Aware Training Framework for Extremely Low-Bit LLMs
+
+- **ID:** `2601.20745` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Guoan Wang, Feiyu Wang, Zongwei Lv, et al.
+- **abs:** https://arxiv.org/abs/2601.20745 · **PDF:** https://arxiv.org/pdf/2601.20745v1
+
+## 18511. SA-PEF: Step-Ahead Partial Error Feedback for Efficient Federated Learning
+
+- **ID:** `2601.20738` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Dawit Kiros Redie, Reza Arablouei, Stefan Werner
+- **abs:** https://arxiv.org/abs/2601.20738 · **PDF:** https://arxiv.org/pdf/2601.20738v1
+
+## 18512. Implementing Metric Temporal Answer Set Programming
+
+- **ID:** `2601.20735` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Arvid Becker, Pedro Cabalar, Martin Diéguez, et al.
+- **abs:** https://arxiv.org/abs/2601.20735 · **PDF:** https://arxiv.org/pdf/2601.20735v1
+
+## 18513. Continual GUI Agents
+
+- **ID:** `2601.20732` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Ziwei Liu, Borui Kang, Hangjie Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.20732 · **PDF:** https://arxiv.org/pdf/2601.20732v4
+
+## 18514. QueerGen: How LLMs Reflect Societal Norms on Gender and Sexuality in Sentence Completion Tasks
+
+- **ID:** `2601.20731` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Mae Sosto, Delfina Sol Martinez Pandiani, Laura Hollink
+- **abs:** https://arxiv.org/abs/2601.20731 · **PDF:** https://arxiv.org/pdf/2601.20731v1
+
+## 18515. AgentLongBench: A Controllable Long Benchmark For Long-Contexts Agents via Environment Rollouts
+
+- **ID:** `2601.20730` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Shicheng Fang, Yuxin Wang, Xiaoran Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.20730 · **PDF:** https://arxiv.org/pdf/2601.20730v3
+
+## 18516. Deep Semi-Supervised Survival Analysis for Predicting Cancer Prognosis
+
+- **ID:** `2601.20729` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Anchen Sun, Zhibin Chen, Xiaodong Cai
+- **abs:** https://arxiv.org/abs/2601.20729 · **PDF:** https://arxiv.org/pdf/2601.20729v1
+
+## 18517. Li-ViP3D++: Query-Gated Deformable Camera-LiDAR Fusion for End-to-End Perception and Trajectory Prediction
+
+- **ID:** `2601.20720` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Matej Halinkovic, Nina Masarykova, Alexey Vinel, et al.
+- **abs:** https://arxiv.org/abs/2601.20720 · **PDF:** https://arxiv.org/pdf/2601.20720v1
+
+## 18518. Adapting the Behavior of Reinforcement Learning Agents to Changing Action Spaces and Reward Functions
+
+- **ID:** `2601.20714` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Raul de la Rosa, Ivana Dusparic, Nicolas Cardozo
+- **abs:** https://arxiv.org/abs/2601.20714 · **PDF:** https://arxiv.org/pdf/2601.20714v1
+
+## 18519. TwinWeaver: An LLM-Based Foundation Model Framework for Pan-Cancer Digital Twins
+
+- **ID:** `2601.20906` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Nikita Makarov, Maria Bordukova, Lena Voith von Voithenberg, et al.
+- **abs:** https://arxiv.org/abs/2601.20906 · **PDF:** https://arxiv.org/pdf/2601.20906v2
+
+## 18520. A scalable flow-based approach to mitigate topological freezing
+
+- **ID:** `2601.20708` · **primary:** `hep-lat` · **Submitted:** 2026-01-28
+- **Authors:** Claudio Bonanno, Andrea Bulgarelli, Elia Cellini, et al.
+- **abs:** https://arxiv.org/abs/2601.20708 · **PDF:** https://arxiv.org/pdf/2601.20708v1
+
+## 18521. Beyond GEMM-Centric NPUs: Enabling Efficient Diffusion LLM Sampling
+
+- **ID:** `2601.20706` · **primary:** `cs.AR` · **Submitted:** 2026-01-28
+- **Authors:** Binglei Lou, Haoran Wu, Yao Lai, et al.
+- **abs:** https://arxiv.org/abs/2601.20706 · **PDF:** https://arxiv.org/pdf/2601.20706v1
+
+## 18522. Structurally Human, Semantically Biased: Detecting LLM-Generated References with Embeddings and GNNs
+
+- **ID:** `2601.20704` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Melika Mobini, Vincent Holst, Floriano Tori, et al.
+- **abs:** https://arxiv.org/abs/2601.20704 · **PDF:** https://arxiv.org/pdf/2601.20704v1
+
+## 18523. Enterprise Resource Planning Using Multi-type Transformers in Ferro-Titanium Industry
+
+- **ID:** `2601.20696` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Samira Yazdanpourmoghadam, Mahan Balal Pour, Vahid Partovi Nia
+- **abs:** https://arxiv.org/abs/2601.20696 · **PDF:** https://arxiv.org/pdf/2601.20696v1
+
+## 18524. Benchmarking Distilled Language Models: Performance and Efficiency in Resource-Constrained Settings
+
+- **ID:** `2602.20164` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Sachin Gopal Wani, Eric Page, Ajay Dholakia, et al.
+- **abs:** https://arxiv.org/abs/2602.20164 · **PDF:** https://arxiv.org/pdf/2602.20164v1
+
+## 18525. Is Pure Exploitation Sufficient in Exogenous MDPs with Linear Function Approximation?
+
+- **ID:** `2601.20694` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Hao Liang, Jiayu Cheng, Sean R. Sinclair, et al.
+- **abs:** https://arxiv.org/abs/2601.20694 · **PDF:** https://arxiv.org/pdf/2601.20694v1
+
+## 18526. Optimal Transport Group Counterfactual Explanations
+
+- **ID:** `2601.20692` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Enrique Valero-Leal, Bernd Bischl, Pedro Larrañaga, et al.
+- **abs:** https://arxiv.org/abs/2601.20692 · **PDF:** https://arxiv.org/pdf/2601.20692v1
+
+## 18527. How Much Information Can a Vision Token Hold? A Scaling Law for Recognition Limits in VLMs
+
+- **ID:** `2602.02539` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Shuxin Zhuang, Zi Liang, Runsheng Yu, et al.
+- **abs:** https://arxiv.org/abs/2602.02539 · **PDF:** https://arxiv.org/pdf/2602.02539v1
+
+## 18528. Denoising and Baseline Correction of Low-Scan FTIR Spectra: A Benchmark of Deep Learning Models Against Traditional Signal Processing
+
+- **ID:** `2601.20905` · **primary:** `eess.IV` · **Submitted:** 2026-01-28
+- **Authors:** Azadeh Mokari, Shravan Raghunathan, Artem Shydliukh, et al.
+- **abs:** https://arxiv.org/abs/2601.20905 · **PDF:** https://arxiv.org/pdf/2601.20905v1
+
+## 18529. Decoupling Perception and Calibration: Label-Efficient Image Quality Assessment Framework
+
+- **ID:** `2601.20689` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Xinyue Li, Zhichao Zhang, Zhiming Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.20689 · **PDF:** https://arxiv.org/pdf/2601.20689v1
+
+## 18530. Positive-Unlabeled Reinforcement Learning Distillation for On-Premise Small Models
+
+- **ID:** `2601.20687` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Zhiqiang Kou, Junyang Chen, Xin-Qiang Cai, et al.
+- **abs:** https://arxiv.org/abs/2601.20687 · **PDF:** https://arxiv.org/pdf/2601.20687v1
+
+## 18531. MuRAL-CPD: Active Learning for Multiresolution Change Point Detection
+
+- **ID:** `2601.20686` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Stefano Bertolasi, Diego Carrera, Diego Stucchi, et al.
+- **abs:** https://arxiv.org/abs/2601.20686 · **PDF:** https://arxiv.org/pdf/2601.20686v1
+
+## 18532. Polite But Boring? Trade-offs Between Engagement and Psychological Reactance to Chatbot Feedback Styles
+
+- **ID:** `2601.20683` · **primary:** `cs.HC` · **Submitted:** 2026-01-28
+- **Authors:** Samuel Rhys Cox, Joel Wester, Niels van Berkel
+- **abs:** https://arxiv.org/abs/2601.20683 · **PDF:** https://arxiv.org/pdf/2601.20683v1
+
+## 18533. String-Level Ground Fault Localization for TN-Earthed Three-Phase Photovoltaic Systems
+
+- **ID:** `2602.12289` · **primary:** `eess.SY` · **Submitted:** 2026-01-28
+- **Authors:** Yuanliang Li, Xun Gong, Reza Iravani, et al.
+- **abs:** https://arxiv.org/abs/2602.12289 · **PDF:** https://arxiv.org/pdf/2602.12289v2
+
+## 18534. Online Density-Based Clustering for Real-Time Narrative Evolution Monitorin
+
+- **ID:** `2601.20680` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Ostap Vykhopen, Viktoria Skorik, Maksym Tereshchenko, et al.
+- **abs:** https://arxiv.org/abs/2601.20680 · **PDF:** https://arxiv.org/pdf/2601.20680v2
+
+## 18535. ShieldedCode: Learning Robust Representations for Virtual Machine Protected Code
+
+- **ID:** `2601.20679` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Mingqiao Mo, Yunlong Tan, Hao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.20679 · **PDF:** https://arxiv.org/pdf/2601.20679v1
+
+## 18536. Efficient Multimodal Planning Agent for Visual Question-Answering
+
+- **ID:** `2601.20676` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Zhuo Chen, Xinyu Geng, Xinyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.20676 · **PDF:** https://arxiv.org/pdf/2601.20676v1
+
+## 18537. Harnessing Large Language Models for Precision Querying and Retrieval-Augmented Knowledge Extraction in Clinical Data Science
+
+- **ID:** `2601.20674` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Juan Jose Rubio Jan, Jack Wu, Julia Ive
+- **abs:** https://arxiv.org/abs/2601.20674 · **PDF:** https://arxiv.org/pdf/2601.20674v1
+
+## 18538. Learning Contextual Runtime Monitors for Safe AI-Based Autonomy
+
+- **ID:** `2601.20666` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Alejandro Luque-Cerpa, Mengyuan Wang, Emil Carlsson, et al.
+- **abs:** https://arxiv.org/abs/2601.20666 · **PDF:** https://arxiv.org/pdf/2601.20666v3
+
+## 18539. A Dialectic Pipeline for Improving LLM Robustness
+
+- **ID:** `2601.20659` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Sara Candussio
+- **abs:** https://arxiv.org/abs/2601.20659 · **PDF:** https://arxiv.org/pdf/2601.20659v1
+
+## 18540. P2S: Probabilistic Process Supervision for General-Domain Reasoning Question Answering
+
+- **ID:** `2601.20649` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Wenlin Zhong, Chengyuan Liu, Yiquan Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.20649 · **PDF:** https://arxiv.org/pdf/2601.20649v1
+
+## 18541. UltRAG: a Universal Simple Scalable Recipe for Knowledge Graph RAG
+
+- **ID:** `2603.28773` · **primary:** `cs.IR` · **Submitted:** 2026-01-28
+- **Authors:** Dobrik Georgiev, Kheeran Naidu, Alberto Cattaneo, et al.
+- **abs:** https://arxiv.org/abs/2603.28773 · **PDF:** https://arxiv.org/pdf/2603.28773v1
+
+## 18542. Detecting and Mitigating Memorization in Diffusion Models through Anisotropy of the Log-Probability
+
+- **ID:** `2601.20642` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Rohan Asthana, Vasileios Belagiannis
+- **abs:** https://arxiv.org/abs/2601.20642 · **PDF:** https://arxiv.org/pdf/2601.20642v2
+
+## 18543. Investigating the Development of Task-Oriented Communication in Vision-Language Models
+
+- **ID:** `2601.20641` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Boaz Carmeli, Orr Paradise, Shafi Goldwasser, et al.
+- **abs:** https://arxiv.org/abs/2601.20641 · **PDF:** https://arxiv.org/pdf/2601.20641v1
+
+## 18544. An Empirical Investigation of Neural ODEs and Symbolic Regression for Dynamical Systems
+
+- **ID:** `2601.20637` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Panayiotis Ioannou, Pietro Liò, Pietro Cicuta
+- **abs:** https://arxiv.org/abs/2601.20637 · **PDF:** https://arxiv.org/pdf/2601.20637v1
+
+## 18545. A Foundation Model for Virtual Sensors
+
+- **ID:** `2601.20634` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Leon Götz, Lars Frederik Peiss, Erik Sauer, et al.
+- **abs:** https://arxiv.org/abs/2601.20634 · **PDF:** https://arxiv.org/pdf/2601.20634v1
+
+## 18546. Sparse clustering via the Deterministic Information Bottleneck algorithm
+
+- **ID:** `2601.20628` · **primary:** `stat.ML` · **Submitted:** 2026-01-28
+- **Authors:** Efthymios Costa, Ioanna Papatsouma, Angelos Markos
+- **abs:** https://arxiv.org/abs/2601.20628 · **PDF:** https://arxiv.org/pdf/2601.20628v2
+
+## 18547. DIVERSE: Disagreement-Inducing Vector Evolution for Rashomon Set Exploration
+
+- **ID:** `2601.20627` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Gilles Eerlings, Brent Zoomers, Jori Liesenborgs, et al.
+- **abs:** https://arxiv.org/abs/2601.20627 · **PDF:** https://arxiv.org/pdf/2601.20627v1
+
+## 18548. Trigger Optimization and Event Classification for Dark Matter Searches in the CYGNO Experiment Using Machine Learning
+
+- **ID:** `2601.20626` · **primary:** `physics.ins-det` · **Submitted:** 2026-01-28
+- **Authors:** F. D. Amaro, R. Antonietti, E. Baracchini, et al.
+- **abs:** https://arxiv.org/abs/2601.20626 · **PDF:** https://arxiv.org/pdf/2601.20626v2
+
+## 18549. GDCNet: Generative Discrepancy Comparison Network for Multimodal Sarcasm Detection
+
+- **ID:** `2601.20618` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Shuguang Zhang, Junhong Lian, Guoxin Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.20618 · **PDF:** https://arxiv.org/pdf/2601.20618v1
+
+## 18550. Agent Benchmarks Fail Public Sector Requirements
+
+- **ID:** `2601.20617` · **primary:** `cs.CY` · **Submitted:** 2026-01-28
+- **Authors:** Jonathan Rystrøm, Chris Schmitz, Karolina Korgul, et al.
+- **abs:** https://arxiv.org/abs/2601.20617 · **PDF:** https://arxiv.org/pdf/2601.20617v1
+
+## 18551. Harder Is Better: Boosting Mathematical Reasoning via Difficulty-Aware GRPO and Multi-Aspect Question Reformulation
+
+- **ID:** `2601.20614` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Yanqi Dai, Yuxiang Ji, Xiao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.20614 · **PDF:** https://arxiv.org/pdf/2601.20614v1
+
+## 18552. AgentIF-OneDay: A Task-level Instruction-Following Benchmark for General AI Agents in Daily Scenarios
+
+- **ID:** `2601.20613` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Kaiyuan Chen, Qimin Wu, Taiyu Hou, et al.
+- **abs:** https://arxiv.org/abs/2601.20613 · **PDF:** https://arxiv.org/pdf/2601.20613v2
+
+## 18553. ACFormer: Mitigating Non-linearity with Auto Convolutional Encoder for Time Series Forecasting
+
+- **ID:** `2601.20611` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Gawon Lee, Hanbyeol Park, Minseop Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.20611 · **PDF:** https://arxiv.org/pdf/2601.20611v1
+
+## 18554. WFR-MFM: One-Step Inference for Dynamic Unbalanced Optimal Transport
+
+- **ID:** `2601.20606` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Xinyu Wang, Ruoyu Wang, Qiangwei Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.20606 · **PDF:** https://arxiv.org/pdf/2601.20606v1
+
+## 18555. CoBA: Integrated Deep Learning Model for Reliable Low-Altitude UAV Classification in mmWave Radio Networks
+
+- **ID:** `2601.20605` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Junaid Sajid, Ivo Müürsepp, Luca Reggiani, et al.
+- **abs:** https://arxiv.org/abs/2601.20605 · **PDF:** https://arxiv.org/pdf/2601.20605v1
+
+## 18556. Dialogical Reasoning Across AI Architectures: A Multi-Model Framework for Testing AI Alignment Strategies
+
+- **ID:** `2601.20604` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Gray Cox
+- **abs:** https://arxiv.org/abs/2601.20604 · **PDF:** https://arxiv.org/pdf/2601.20604v1
+
+## 18557. CLEAR-Mamba:Towards Accurate, Adaptive and Trustworthy Multi-Sequence Ophthalmic Angiography Classification
+
+- **ID:** `2601.20601` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Zhuonan Wang, Wenjie Yan, Wenqiao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.20601 · **PDF:** https://arxiv.org/pdf/2601.20601v3
+
+## 18558. Regularized Gradient Temporal-Difference Learning
+
+- **ID:** `2601.20599` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Hyunjun Na, Donghwan Lee
+- **abs:** https://arxiv.org/abs/2601.20599 · **PDF:** https://arxiv.org/pdf/2601.20599v1
+
+## 18559. Person Re-ID in 2025: Supervised, Self-Supervised, and Language-Aligned. What Works?
+
+- **ID:** `2601.20598` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Lakshman Balasubramanian
+- **abs:** https://arxiv.org/abs/2601.20598 · **PDF:** https://arxiv.org/pdf/2601.20598v1
+
+## 18560. Tacit Coordination of Large Language Models
+
+- **ID:** `2601.22184` · **primary:** `cs.GT` · **Submitted:** 2026-01-28
+- **Authors:** Ido Aharon, Emanuele La Malfa, Michael Wooldridge, et al.
+- **abs:** https://arxiv.org/abs/2601.22184 · **PDF:** https://arxiv.org/pdf/2601.22184v1
+
+## 18561. A Computational Approach to Language Contact -- A Case Study of Persian
+
+- **ID:** `2601.20592` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Ali Basirat, Danial Namazifard, Navid Baradaran Hemmati
+- **abs:** https://arxiv.org/abs/2601.20592 · **PDF:** https://arxiv.org/pdf/2601.20592v1
+
+## 18562. Exact Graph Learning via Integer Programming
+
+- **ID:** `2601.20589` · **primary:** `stat.ME` · **Submitted:** 2026-01-28
+- **Authors:** Lucas Kook, Søren Wengel Mogensen
+- **abs:** https://arxiv.org/abs/2601.20589 · **PDF:** https://arxiv.org/pdf/2601.20589v2
+
+## 18563. Ranking-aware Reinforcement Learning for Ordinal Ranking
+
+- **ID:** `2601.20585` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Aiming Hao, Chen Zhu, Jiashu Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.20585 · **PDF:** https://arxiv.org/pdf/2601.20585v1
+
+## 18564. Single-Nodal Spontaneous Symmetry Breaking in NLP Models
+
+- **ID:** `2601.20582` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Shalom Rosner, Ronit D. Gross, Ella Koresh, et al.
+- **abs:** https://arxiv.org/abs/2601.20582 · **PDF:** https://arxiv.org/pdf/2601.20582v1
+
+## 18565. Inequality in Congestion Games with Learning Agents
+
+- **ID:** `2601.20578` · **primary:** `cs.GT` · **Submitted:** 2026-01-28
+- **Authors:** Dimitris Michailidis, Sennay Ghebreab, Fernando P. Santos
+- **abs:** https://arxiv.org/abs/2601.20578 · **PDF:** https://arxiv.org/pdf/2601.20578v2
+
+## 18566. Robust Distributed Learning under Resource Constraints: Decentralized Quantile Estimation via (Asynchronous) ADMM
+
+- **ID:** `2601.20571` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Anna van Elst, Igor Colin, Stephan Clémençon
+- **abs:** https://arxiv.org/abs/2601.20571 · **PDF:** https://arxiv.org/pdf/2601.20571v1
+
+## 18567. Reinforcement Unlearning via Group Relative Policy Optimization
+
+- **ID:** `2601.20568` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Efstratios Zaradoukas, Bardh Prenkaj, Gjergji Kasneci
+- **abs:** https://arxiv.org/abs/2601.20568 · **PDF:** https://arxiv.org/pdf/2601.20568v3
+
+## 18568. Unsupervised Ensemble Learning Through Deep Energy-based Models
+
+- **ID:** `2601.20556` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Ariel Maymon, Yanir Buznah, Uri Shaham
+- **abs:** https://arxiv.org/abs/2601.20556 · **PDF:** https://arxiv.org/pdf/2601.20556v1
+
+## 18569. Online Risk-Averse Planning in POMDPs Using Iterated CVaR Value Function
+
+- **ID:** `2601.20554` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Yaacov Pariente, Vadim Indelman
+- **abs:** https://arxiv.org/abs/2601.20554 · **PDF:** https://arxiv.org/pdf/2601.20554v1
+
+## 18570. IoT Device Identification with Machine Learning: Common Pitfalls and Best Practices
+
+- **ID:** `2601.20548` · **primary:** `cs.CR` · **Submitted:** 2026-01-28
+- **Authors:** Kahraman Kostas, Rabia Yasa Kostas
+- **abs:** https://arxiv.org/abs/2601.20548 · **PDF:** https://arxiv.org/pdf/2601.20548v1
+
+## 18571. Beyond Divergent Creativity: A Human-Based Evaluation of Creativity in Large Language Models
+
+- **ID:** `2601.20546` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Kumiko Nakajima, Jan Zuiderveld, Sandro Pezzelle
+- **abs:** https://arxiv.org/abs/2601.20546 · **PDF:** https://arxiv.org/pdf/2601.20546v1
+
+## 18572. PathWise: Planning through World Model for Automated Heuristic Design via Self-Evolving LLMs
+
+- **ID:** `2601.20539` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Oguzhan Gungordu, Siheng Xiong, Faramarz Fekri
+- **abs:** https://arxiv.org/abs/2601.20539 · **PDF:** https://arxiv.org/pdf/2601.20539v2
+
+## 18573. Interpreting Emergent Extreme Events in Multi-Agent Systems
+
+- **ID:** `2601.20538` · **primary:** `cs.MA` · **Submitted:** 2026-01-28
+- **Authors:** Ling Tang, Jilin Mei, Dongrui Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.20538 · **PDF:** https://arxiv.org/pdf/2601.20538v2
+
+## 18574. Intelligence as Trajectory-Dominant Pareto Optimization
+
+- **ID:** `2602.13230` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Truong Xuan Khanh, Truong Quynh Hoa
+- **abs:** https://arxiv.org/abs/2602.13230 · **PDF:** https://arxiv.org/pdf/2602.13230v1
+
+## 18575. Incorporating data drift to perform survival analysis on credit risk
+
+- **ID:** `2601.20533` · **primary:** `stat.ML` · **Submitted:** 2026-01-28
+- **Authors:** Jianwei Peng, Stefan Lessmann
+- **abs:** https://arxiv.org/abs/2601.20533 · **PDF:** https://arxiv.org/pdf/2601.20533v1
+
+## 18576. Adaptive Temporal Dynamics for Personalized Emotion Recognition: A Liquid Neural Network Approach
+
+- **ID:** `2602.06997` · **primary:** `eess.SP` · **Submitted:** 2026-01-28
+- **Authors:** Anindya Bhattacharjee, Nittya Ananda Biswas, K. A. Shahriar, et al.
+- **abs:** https://arxiv.org/abs/2602.06997 · **PDF:** https://arxiv.org/pdf/2602.06997v1
+
+## 18577. ECGFlowCMR: Pretraining with ECG-Generated Cine CMR Helps Cardiac Disease Classification and Phenotype Prediction
+
+- **ID:** `2601.20904` · **primary:** `eess.IV` · **Submitted:** 2026-01-28
+- **Authors:** Xiaocheng Fang, Zhengyao Ding, Guangkun Nie, et al.
+- **abs:** https://arxiv.org/abs/2601.20904 · **PDF:** https://arxiv.org/pdf/2601.20904v2
+
+## 18578. Spectral Bayesian Regression on the Sphere
+
+- **ID:** `2601.20528` · **primary:** `math.ST` · **Submitted:** 2026-01-28
+- **Authors:** Claudio Durastanti
+- **abs:** https://arxiv.org/abs/2601.20528 · **PDF:** https://arxiv.org/pdf/2601.20528v1
+
+## 18579. ICON: Intent-Context Coupling for Efficient Multi-Turn Jailbreak Attack
+
+- **ID:** `2601.20903` · **primary:** `cs.CR` · **Submitted:** 2026-01-28
+- **Authors:** Xingwei Lin, Wenhao Lin, Sicong Cao, et al.
+- **abs:** https://arxiv.org/abs/2601.20903 · **PDF:** https://arxiv.org/pdf/2601.20903v1
+
+## 18580. Enhancing Post-Training Quantization via Future Activation Awareness
+
+- **ID:** `2602.02538` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Zheqi Lv, Zhenxuan Fan, Qi Tian, et al.
+- **abs:** https://arxiv.org/abs/2602.02538 · **PDF:** https://arxiv.org/pdf/2602.02538v1
+
+## 18581. CCMamba: Topologically-Informed Selective State-Space Networks on Combinatorial Complexes for Higher-Order Graph Learning
+
+- **ID:** `2601.20518` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Jiawen Chen, Qi Shao, Mingtong Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.20518 · **PDF:** https://arxiv.org/pdf/2601.20518v2
+
+## 18582. AIdentifyAGE Ontology for Decision Support in Forensic Dental Age Assessment
+
+- **ID:** `2602.16714` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Renato Marcelo, Ana Rodrigues, Cristiana Palmela Pereira, et al.
+- **abs:** https://arxiv.org/abs/2602.16714 · **PDF:** https://arxiv.org/pdf/2602.16714v1
+
+## 18583. Audio Deepfake Detection in the Age of Advanced Text-to-Speech models
+
+- **ID:** `2601.20510` · **primary:** `cs.SD` · **Submitted:** 2026-01-28
+- **Authors:** Robin Singh, Aditya Yogesh Nair, Fabio Palumbo, et al.
+- **abs:** https://arxiv.org/abs/2601.20510 · **PDF:** https://arxiv.org/pdf/2601.20510v1
+
+## 18584. WorldVQA: Measuring Atomic World Knowledge in Multimodal Large Language Models
+
+- **ID:** `2602.02537` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Runjie Zhou, Youbo Shao, Haoyu Lu, et al.
+- **abs:** https://arxiv.org/abs/2602.02537 · **PDF:** https://arxiv.org/pdf/2602.02537v1
+
+## 18585. Comparative evaluation of training strategies using partially labelled datasets for segmentation of white matter hyperintensities and stroke lesions in FLAIR MRI
+
+- **ID:** `2601.20503` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Jesse Phitidis, Alison Q. Smithard, William N. Whiteley, et al.
+- **abs:** https://arxiv.org/abs/2601.20503 · **PDF:** https://arxiv.org/pdf/2601.20503v1
+
+## 18586. MARVL: Multi-Stage Guidance for Robotic Manipulation via Vision-Language Models
+
+- **ID:** `2602.15872` · **primary:** `cs.RO` · **Submitted:** 2026-01-28
+- **Authors:** Xunlan Zhou, Xuanlin Chen, Shaowei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.15872 · **PDF:** https://arxiv.org/pdf/2602.15872v2
+
+## 18587. Spectral Diffusion Models on the Sphere
+
+- **ID:** `2601.20498` · **primary:** `math.PR` · **Submitted:** 2026-01-28
+- **Authors:** Pierpaolo Brutti, Claudio Durastanti, Francesco Mari
+- **abs:** https://arxiv.org/abs/2601.20498 · **PDF:** https://arxiv.org/pdf/2601.20498v1
+
+## 18588. Physics-informed Blind Reconstruction of Dense Fields from Sparse Measurements using Neural Networks with a Differentiable Simulator
+
+- **ID:** `2601.20496` · **primary:** `stat.ML` · **Submitted:** 2026-01-28
+- **Authors:** Ofek Aloni, Barak Fishbain
+- **abs:** https://arxiv.org/abs/2601.20496 · **PDF:** https://arxiv.org/pdf/2601.20496v1
+
+## 18589. Normative Equivalence in Human-AI Cooperation: Behaviour, Not Identity, Drives Cooperation in Mixed-Agent Groups
+
+- **ID:** `2601.20487` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Nico Mutzner, Taha Yasseri, Heiko Rauhut
+- **abs:** https://arxiv.org/abs/2601.20487 · **PDF:** https://arxiv.org/pdf/2601.20487v2
+
+## 18590. An explainable framework for the relationship between dementia and glucose metabolism patterns
+
+- **ID:** `2601.20480` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** C. Vázquez-García, F. J. Martínez-Murcia, F. Segovia Román, et al.
+- **abs:** https://arxiv.org/abs/2601.20480 · **PDF:** https://arxiv.org/pdf/2601.20480v2
+
+## 18591. Implicit Hypothesis Testing and Divergence Preservation in Neural Network Representations
+
+- **ID:** `2601.20477` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Kadircan Aksoy, Protim Bhattacharjee, Peter Jung
+- **abs:** https://arxiv.org/abs/2601.20477 · **PDF:** https://arxiv.org/pdf/2601.20477v2
+
+## 18592. Can We Improve Educational Diagram Generation with In-Context Examples? Not if a Hallucination Spoils the Bunch
+
+- **ID:** `2601.20476` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Evanfiya Logacheva, Arto Hellas, Tsvetomila Mihaylova, et al.
+- **abs:** https://arxiv.org/abs/2601.20476 · **PDF:** https://arxiv.org/pdf/2601.20476v1
+
+## 18593. COL-Trees: Efficient Hierarchical Object Search in Road Networks
+
+- **ID:** `2601.22183` · **primary:** `cs.DB` · **Submitted:** 2026-01-28
+- **Authors:** Tenindra Abeywickrama, Muhammad Aamir Cheema, Sabine Storandt
+- **abs:** https://arxiv.org/abs/2601.22183 · **PDF:** https://arxiv.org/pdf/2601.22183v1
+
+## 18594. CtrlCoT: Dual-Granularity Chain-of-Thought Compression for Controllable Reasoning
+
+- **ID:** `2601.20467` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Zhenxuan Fan, Jie Cao, Yang Dai, et al.
+- **abs:** https://arxiv.org/abs/2601.20467 · **PDF:** https://arxiv.org/pdf/2601.20467v1
+
+## 18595. BMAM: Brain-inspired Multi-Agent Memory Framework
+
+- **ID:** `2601.20465` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Yang Li, Jiaxiang Liu, Yusong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.20465 · **PDF:** https://arxiv.org/pdf/2601.20465v1
+
+## 18596. MuVaC: A Variational Causal Framework for Multimodal Sarcasm Understanding in Dialogues
+
+- **ID:** `2601.20451` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Diandian Guo, Fangfang Yuan, Cong Cao, et al.
+- **abs:** https://arxiv.org/abs/2601.20451 · **PDF:** https://arxiv.org/pdf/2601.20451v2
+
+## 18597. Text-only adaptation in LLM-based ASR through text denoising
+
+- **ID:** `2601.20900` · **primary:** `cs.SD` · **Submitted:** 2026-01-28
+- **Authors:** Andrés Carofilis, Sergio Burdisso, Esaú Villatoro-Tello, et al.
+- **abs:** https://arxiv.org/abs/2601.20900 · **PDF:** https://arxiv.org/pdf/2601.20900v3
+
+## 18598. Fair Recourse for All: Ensuring Individual and Group Fairness in Counterfactual Explanations
+
+- **ID:** `2601.20449` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Fatima Ezzeddine, Obaida Ammar, Silvia Giordano, et al.
+- **abs:** https://arxiv.org/abs/2601.20449 · **PDF:** https://arxiv.org/pdf/2601.20449v1
+
+## 18599. Curriculum-Learned Vanishing Stacked Residual PINNs for Hyperbolic PDE State Reconstruction
+
+- **ID:** `2602.06996` · **primary:** `cs.NE` · **Submitted:** 2026-01-28
+- **Authors:** Katayoun Eshkofti, Matthieu Barreau
+- **abs:** https://arxiv.org/abs/2602.06996 · **PDF:** https://arxiv.org/pdf/2602.06996v1
+
+## 18600. TimeCatcher: A Variational Framework for Volatility-Aware Forecasting of Non-Stationary Time Series
+
+- **ID:** `2601.20448` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Zhiyu Chen, Minhao Liu, Yanru Zhang
+- **abs:** https://arxiv.org/abs/2601.20448 · **PDF:** https://arxiv.org/pdf/2601.20448v1
+
+## 18601. Assembling the Mind's Mosaic: Towards EEG Semantic Intent Decoding
+
+- **ID:** `2601.20447` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-28
+- **Authors:** Jiahe Li, Junru Chen, Fanqi Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.20447 · **PDF:** https://arxiv.org/pdf/2601.20447v1
+
+## 18602. Reducing Prompt Sensitivity in LLM-based Speech Recognition Through Learnable Projection
+
+- **ID:** `2601.20898` · **primary:** `eess.AS` · **Submitted:** 2026-01-28
+- **Authors:** Sergio Burdisso, Esaú Villatoro-Tello, Shashi Kumar, et al.
+- **abs:** https://arxiv.org/abs/2601.20898 · **PDF:** https://arxiv.org/pdf/2601.20898v1
+
+## 18603. PEARL: Plan Exploration and Adaptive Reinforcement Learning for Multihop Tool Use
+
+- **ID:** `2601.20439` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Qihao Wang, Mingzhe Lu, Jiayue Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.20439 · **PDF:** https://arxiv.org/pdf/2601.20439v1
+
+## 18604. Self Voice Conversion as an Attack against Neural Audio Watermarking
+
+- **ID:** `2601.20432` · **primary:** `cs.SD` · **Submitted:** 2026-01-28
+- **Authors:** Yigitcan Özer, Wanying Ge, Zhe Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.20432 · **PDF:** https://arxiv.org/pdf/2601.20432v2
+
+## 18605. Nonlinear Dimensionality Reduction with Diffusion Maps in Practice
+
+- **ID:** `2601.20428` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Sönke Beier, Paula Pirker-Díaz, Friedrich Pagenkopf, et al.
+- **abs:** https://arxiv.org/abs/2601.20428 · **PDF:** https://arxiv.org/pdf/2601.20428v1
+
+## 18606. Hopes and Fears -- Emotion Distribution in the Topic Landscape of Finnish Parliamentary Speech 2000-2020
+
+- **ID:** `2601.20424` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Anna Ristilä, Otto Tarkka, Veronika Laippala, et al.
+- **abs:** https://arxiv.org/abs/2601.20424 · **PDF:** https://arxiv.org/pdf/2601.20424v1
+
+## 18607. From Sparse Decisions to Dense Reasoning: A Multi-attribute Trajectory Paradigm for Multimodal Moderation
+
+- **ID:** `2602.02536` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Tianle Gu, Kexin Huang, Lingyu Li, et al.
+- **abs:** https://arxiv.org/abs/2602.02536 · **PDF:** https://arxiv.org/pdf/2602.02536v1
+
+## 18608. Guiding the Recommender: Information-Aware Auto-Bidding for Content Promotion
+
+- **ID:** `2601.20422` · **primary:** `cs.GT` · **Submitted:** 2026-01-28
+- **Authors:** Yumou Liu, Zhenzhe Zheng, Jiang Rong, et al.
+- **abs:** https://arxiv.org/abs/2601.20422 · **PDF:** https://arxiv.org/pdf/2601.20422v1
+
+## 18609. Concept Component Analysis: A Principled Approach for Concept Extraction in LLMs
+
+- **ID:** `2601.20420` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Yuhang Liu, Erdun Gao, Dong Gong, et al.
+- **abs:** https://arxiv.org/abs/2601.20420 · **PDF:** https://arxiv.org/pdf/2601.20420v2
+
+## 18610. Let's Roll a BiFTA: Bi-refinement for Fine-grained Text-visual Alignment in Vision-Language Models
+
+- **ID:** `2601.20419` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Yuhao Sun, Chengyi Cai, Jiacheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.20419 · **PDF:** https://arxiv.org/pdf/2601.20419v1
+
+## 18611. SpeechMapper: Speech-to-text Embedding Projector for LLMs
+
+- **ID:** `2601.20417` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Biswesh Mohapatra, Marcely Zanon Boito, Ioan Calapodescu
+- **abs:** https://arxiv.org/abs/2601.20417 · **PDF:** https://arxiv.org/pdf/2601.20417v2
+
+## 18612. Enhancing Psychologists' Understanding through Explainable Deep Learning Framework for ADHD Diagnosis
+
+- **ID:** `2602.02535` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Abdul Rehman, Ilona Heldal, Jerry Chun-Wei Lin
+- **abs:** https://arxiv.org/abs/2602.02535 · **PDF:** https://arxiv.org/pdf/2602.02535v1
+
+## 18613. Beyond Accuracy: A Cognitive Load Framework for Mapping the Capability Boundaries of Tool-use Agents
+
+- **ID:** `2601.20412` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Qihao Wang, Yue Hu, Mingzhe Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.20412 · **PDF:** https://arxiv.org/pdf/2601.20412v1
+
+## 18614. AWGformer: Adaptive Wavelet-Guided Transformer for Multi-Resolution Time Series Forecasting
+
+- **ID:** `2601.20409` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Wei Li
+- **abs:** https://arxiv.org/abs/2601.20409 · **PDF:** https://arxiv.org/pdf/2601.20409v1
+
+## 18615. Meeting SLOs, Slashing Hours: Automated Enterprise LLM Optimization with OptiKIT
+
+- **ID:** `2601.20408` · **primary:** `cs.DC` · **Submitted:** 2026-01-28
+- **Authors:** Nicholas Santavas, Kareem Eissa, Patrycja Cieplicka, et al.
+- **abs:** https://arxiv.org/abs/2601.20408 · **PDF:** https://arxiv.org/pdf/2601.20408v1
+
+## 18616. On the Impact of AGENTS.md Files on the Efficiency of AI Coding Agents
+
+- **ID:** `2601.20404` · **primary:** `cs.SE` · **Submitted:** 2026-01-28
+- **Authors:** Jai Lal Lulla, Seyedmoein Mohsenimofidi, Matthias Galster, et al.
+- **abs:** https://arxiv.org/abs/2601.20404 · **PDF:** https://arxiv.org/pdf/2601.20404v2
+
+## 18617. GuideAI: A Real-time Personalized Learning Solution with Adaptive Interventions
+
+- **ID:** `2601.20402` · **primary:** `cs.HC` · **Submitted:** 2026-01-28
+- **Authors:** Ananya Shukla, Chaitanya Modi, Satvik Bajpai, et al.
+- **abs:** https://arxiv.org/abs/2601.20402 · **PDF:** https://arxiv.org/pdf/2601.20402v1
+
+## 18618. ScatterFusion: A Hierarchical Scattering Transform Framework for Enhanced Time Series Forecasting
+
+- **ID:** `2601.20401` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Wei Li
+- **abs:** https://arxiv.org/abs/2601.20401 · **PDF:** https://arxiv.org/pdf/2601.20401v1
+
+## 18619. Convergence Analysis of Randomized Subspace Normalized SGD under Heavy-Tailed Noise
+
+- **ID:** `2601.20399` · **primary:** `math.OC` · **Submitted:** 2026-01-28
+- **Authors:** Gaku Omiya, Pierre-Louis Poirion, Akiko Takeda
+- **abs:** https://arxiv.org/abs/2601.20399 · **PDF:** https://arxiv.org/pdf/2601.20399v2
+
+## 18620. FedRD: Reducing Divergences for Generalized Federated Learning via Heterogeneity-aware Parameter Guidance
+
+- **ID:** `2601.20397` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Kaile Wang, Jiannong Cao, Yu Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.20397 · **PDF:** https://arxiv.org/pdf/2601.20397v1
+
+## 18621. Graph-Structured Deep Learning Framework for Multi-task Contention Identification with High-dimensional Metrics
+
+- **ID:** `2601.20389` · **primary:** `cs.DC` · **Submitted:** 2026-01-28
+- **Authors:** Xiao Yang, Yinan Ni, Yuqi Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.20389 · **PDF:** https://arxiv.org/pdf/2601.20389v1
+
+## 18622. Faster Predictive Coding Networks via Better Initialization
+
+- **ID:** `2601.20895` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Luca Pinchetti, Simon Frieder, Thomas Lukasiewicz, et al.
+- **abs:** https://arxiv.org/abs/2601.20895 · **PDF:** https://arxiv.org/pdf/2601.20895v1
+
+## 18623. OmegaUse: Building a General-Purpose GUI Agent for Autonomous Task Execution
+
+- **ID:** `2601.20380` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Le Zhang, Yixiong Xiao, Xinjiang Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.20380 · **PDF:** https://arxiv.org/pdf/2601.20380v1
+
+## 18624. Policy of Thoughts: Scaling LLM Reasoning via Test-time Policy Evolution
+
+- **ID:** `2601.20379` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Zhengbo Jiao, Hongyu Xian, Qinglong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.20379 · **PDF:** https://arxiv.org/pdf/2601.20379v1
+
+## 18625. LLM-AutoDP: Automatic Data Processing via LLM Agents for Model Fine-tuning
+
+- **ID:** `2601.20375` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Wei Huang, Anda Cheng, Yinggui Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.20375 · **PDF:** https://arxiv.org/pdf/2601.20375v1
+
+## 18626. Unsupervised Anomaly Detection in Multi-Agent Trajectory Prediction via Transformer-Based Models
+
+- **ID:** `2601.20367` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Qing Lyu, Zhe Fu, Alexandre Bayen
+- **abs:** https://arxiv.org/abs/2601.20367 · **PDF:** https://arxiv.org/pdf/2601.20367v1
+
+## 18627. Can Continuous-Time Diffusion Models Generate and Solve Globally Constrained Discrete Problems? A Study on Sudoku
+
+- **ID:** `2601.20363` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Mariia Drozdova
+- **abs:** https://arxiv.org/abs/2601.20363 · **PDF:** https://arxiv.org/pdf/2601.20363v1
+
+## 18628. Switchcodec: Adaptive residual-expert sparse quantization for high-fidelity neural audio coding
+
+- **ID:** `2601.20362` · **primary:** `cs.SD` · **Submitted:** 2026-01-28
+- **Authors:** Xiangbo Wang, Wenbin Jiang, Jin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.20362 · **PDF:** https://arxiv.org/pdf/2601.20362v1
+
+## 18629. TINNs: Time-Induced Neural Networks for Solving Time-Dependent PDEs
+
+- **ID:** `2601.20361` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Chen-Yang Dai, Che-Chia Chang, Te-Sheng Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.20361 · **PDF:** https://arxiv.org/pdf/2601.20361v1
+
+## 18630. Is Parameter Isolation Better for Prompt-Based Continual Learning?
+
+- **ID:** `2601.20894` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Jiangyang Li, Chenhao Ding, Songlin Dong, et al.
+- **abs:** https://arxiv.org/abs/2601.20894 · **PDF:** https://arxiv.org/pdf/2601.20894v1
+
+## 18631. TABED: Test-Time Adaptive Ensemble Drafting for Robust Speculative Decoding in LVLMs
+
+- **ID:** `2601.20357` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Minjae Lee, Wonjun Kang, Byeongkeun Ahn, et al.
+- **abs:** https://arxiv.org/abs/2601.20357 · **PDF:** https://arxiv.org/pdf/2601.20357v1
+
+## 18632. CURVE: Learning Causality-Inspired Invariant Representations for Robust Scene Understanding via Uncertainty-Guided Regularization
+
+- **ID:** `2601.20355` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Yue Liang, Jiatong Du, Ziyi Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.20355 · **PDF:** https://arxiv.org/pdf/2601.20355v1
+
+## 18633. AMA: Adaptive Memory via Multi-Agent Collaboration
+
+- **ID:** `2601.20352` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Weiquan Huang, Zixuan Wang, Hehai Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.20352 · **PDF:** https://arxiv.org/pdf/2601.20352v2
+
+## 18634. Multimodal Multi-Agent Ransomware Analysis Using AutoGen
+
+- **ID:** `2601.20346` · **primary:** `cs.CR` · **Submitted:** 2026-01-28
+- **Authors:** Asifullah Khan, Aimen Wadood, Mubashar Iqbal, et al.
+- **abs:** https://arxiv.org/abs/2601.20346 · **PDF:** https://arxiv.org/pdf/2601.20346v2
+
+## 18635. Improving Diffusion Language Model Decoding through Joint Search in Generation Order and Token Space
+
+- **ID:** `2601.20339` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Yangyi Shen, Tianjian Feng, Jiaqi Han, et al.
+- **abs:** https://arxiv.org/abs/2601.20339 · **PDF:** https://arxiv.org/pdf/2601.20339v2
+
+## 18636. Do Whitepaper Claims Predict Market Behavior? Evidence from Cryptocurrency Factor Analysis
+
+- **ID:** `2601.20336` · **primary:** `q-fin.CP` · **Submitted:** 2026-01-28
+- **Authors:** Murad Farzulla
+- **abs:** https://arxiv.org/abs/2601.20336 · **PDF:** https://arxiv.org/pdf/2601.20336v4
+
+## 18637. HMVLA: Hyperbolic Multimodal Fusion for Vision-Language-Action Models
+
+- **ID:** `2602.02533` · **primary:** `cs.RO` · **Submitted:** 2026-01-28
+- **Authors:** Kun Wang, Xiao Feng, Mingcheng Qu, et al.
+- **abs:** https://arxiv.org/abs/2602.02533 · **PDF:** https://arxiv.org/pdf/2602.02533v1
+
+## 18638. MobileBench-OL: A Comprehensive Chinese Benchmark for Evaluating Mobile GUI Agents in Real-World Environment
+
+- **ID:** `2601.20335` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Qinzhuo Wu, Zhizhuo Yang, Hanhao Li, et al.
+- **abs:** https://arxiv.org/abs/2601.20335 · **PDF:** https://arxiv.org/pdf/2601.20335v2
+
+## 18639. Demonstration-Free Robotic Control via LLM Agents
+
+- **ID:** `2601.20334` · **primary:** `cs.RO` · **Submitted:** 2026-01-28
+- **Authors:** Brian Y. Tsui, Alan Y. Fang, Tiffany J. Hwu
+- **abs:** https://arxiv.org/abs/2601.20334 · **PDF:** https://arxiv.org/pdf/2601.20334v1
+
+## 18640. Test-Time Adaptation for Anomaly Segmentation via Topology-Aware Optimal Transport Chaining
+
+- **ID:** `2601.20333` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Ali Zia, Usman Ali, Umer Ramzan, et al.
+- **abs:** https://arxiv.org/abs/2601.20333 · **PDF:** https://arxiv.org/pdf/2601.20333v1
+
+## 18641. Window-Diffusion: Accelerating Diffusion Language Model Inference with Windowed Token Pruning and Caching
+
+- **ID:** `2601.20332` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Fengrui Zuo, Zhiwei Ke, Yiming Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.20332 · **PDF:** https://arxiv.org/pdf/2601.20332v2
+
+## 18642. PsychePass: Calibrating LLM Therapeutic Competence via Trajectory-Anchored Tournaments
+
+- **ID:** `2601.20330` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Zhuang Chen, Dazhen Wan, Zhangkai Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.20330 · **PDF:** https://arxiv.org/pdf/2601.20330v1
+
+## 18643. CE-RM: A Pointwise Generative Reward Model Optimized via Two-Stage Rollout and Unified Criteria
+
+- **ID:** `2601.20327` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Xinyu Hu, Yancheng He, Weixun Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.20327 · **PDF:** https://arxiv.org/pdf/2601.20327v2
+
+## 18644. Beyond Speedup -- Utilizing KV Cache for Sampling and Reasoning
+
+- **ID:** `2601.20326` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Zeyu Xing, Xing Li, Hui-Ling Zhen, et al.
+- **abs:** https://arxiv.org/abs/2601.20326 · **PDF:** https://arxiv.org/pdf/2601.20326v1
+
+## 18645. ECG-Agent: On-Device Tool-Calling Agent for ECG Multi-Turn Dialogue
+
+- **ID:** `2601.20323` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Hyunseung Chung, Jungwoo Oh, Daeun Kyung, et al.
+- **abs:** https://arxiv.org/abs/2601.20323 · **PDF:** https://arxiv.org/pdf/2601.20323v1
+
+## 18646. A generative machine learning model for designing metal hydrides applied to hydrogen storage
+
+- **ID:** `2601.20892` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Xiyuan Liu, Christian Hacker, Shengnian Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.20892 · **PDF:** https://arxiv.org/pdf/2601.20892v1
+
+## 18647. Less is More: Benchmarking LLM Based Recommendation Agents
+
+- **ID:** `2601.20316` · **primary:** `cs.IR` · **Submitted:** 2026-01-28
+- **Authors:** Kargi Chauhan, Mahalakshmi Venkateswarlu
+- **abs:** https://arxiv.org/abs/2601.20316 · **PDF:** https://arxiv.org/pdf/2601.20316v1
+
+## 18648. SAPO: Self-Adaptive Process Optimization Makes Small Reasoners Stronger
+
+- **ID:** `2601.20312` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Kaiyuan Chen, Guangmin Zheng, Jin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.20312 · **PDF:** https://arxiv.org/pdf/2601.20312v2
+
+## 18649. DiagLink: A Dual-User Diagnostic Assistance System by Synergizing Experts with LLMs and Knowledge Graphs
+
+- **ID:** `2601.20311` · **primary:** `cs.HC` · **Submitted:** 2026-01-28
+- **Authors:** Zihan Zhou, Yinan Liu, Yuyang Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.20311 · **PDF:** https://arxiv.org/pdf/2601.20311v1
+
+## 18650. ShellForge: Adversarial Co-Evolution of Webshell Generation and Multi-View Detection for Robust Webshell Defense
+
+- **ID:** `2601.22182` · **primary:** `cs.CR` · **Submitted:** 2026-01-28
+- **Authors:** Yizhong Ding
+- **abs:** https://arxiv.org/abs/2601.22182 · **PDF:** https://arxiv.org/pdf/2601.22182v1
+
+## 18651. SemBind: Binding Diffusion Watermarks to Semantics Against Black-Box Forgery Attacks
+
+- **ID:** `2601.20310` · **primary:** `cs.CR` · **Submitted:** 2026-01-28
+- **Authors:** Xin Zhang, Zijin Yang, Kejiang Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.20310 · **PDF:** https://arxiv.org/pdf/2601.20310v1
+
+## 18652. SurfAge-Net: A Hierarchical Surface-Based Network for Interpretable Fine-Grained Brain Age Prediction
+
+- **ID:** `2602.06994` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-28
+- **Authors:** Rongzhao He, Dalin Zhu, Ying Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.06994 · **PDF:** https://arxiv.org/pdf/2602.06994v1
+
+## 18653. SuperInfer: SLO-Aware Rotary Scheduling and Memory Management for LLM Inference on Superchips
+
+- **ID:** `2601.20309` · **primary:** `cs.DC` · **Submitted:** 2026-01-28
+- **Authors:** Jiahuan Yu, Mingtao Hu, Zichao Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.20309 · **PDF:** https://arxiv.org/pdf/2601.20309v1
+
+## 18654. Attractor Patch Networks: Reducing Catastrophic Forgetting with Routed Low-Rank Patch Experts
+
+- **ID:** `2602.06993` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Shashank
+- **abs:** https://arxiv.org/abs/2602.06993 · **PDF:** https://arxiv.org/pdf/2602.06993v1
+
+## 18655. Delayed Feedback Modeling for Post-Click Gross Merchandise Volume Prediction: Benchmark, Insights and Approaches
+
+- **ID:** `2601.20307` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Xinyu Li, Sishuo Chen, Guipeng Xv, et al.
+- **abs:** https://arxiv.org/abs/2601.20307 · **PDF:** https://arxiv.org/pdf/2601.20307v1
+
+## 18656. Endogenous Reprompting: Self-Evolving Cognitive Alignment for Unified Multimodal Models
+
+- **ID:** `2601.20305` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Zhenchen Tang, Songlin Yang, Zichuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.20305 · **PDF:** https://arxiv.org/pdf/2601.20305v1
+
+## 18657. Structure-constrained Language-informed Diffusion Model for Unpaired Low-dose Computed Tomography Angiography Reconstruction
+
+- **ID:** `2601.20304` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Genyuan Zhang, Zihao Wang, Zhifan Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.20304 · **PDF:** https://arxiv.org/pdf/2601.20304v1
+
+## 18658. Physically Guided Visual Mass Estimation from a Single RGB Image
+
+- **ID:** `2601.20303` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Sungjae Lee, Junhan Jeong, Yeonjoo Hong, et al.
+- **abs:** https://arxiv.org/abs/2601.20303 · **PDF:** https://arxiv.org/pdf/2601.20303v1
+
+## 18659. Towards Compact and Robust DNNs via Compression-aware Sharpness Minimization
+
+- **ID:** `2601.20301` · **primary:** `cs.CV` · **Submitted:** 2026-01-28
+- **Authors:** Jialuo He, Huangxun Chen
+- **abs:** https://arxiv.org/abs/2601.20301 · **PDF:** https://arxiv.org/pdf/2601.20301v1
+
+## 18660. MiLorE-SSL: Scaling Multilingual Capabilities in Self-Supervised Models without Forgetting
+
+- **ID:** `2601.20300` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Jing Xu, Minglin Wu, Xueyuan Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.20300 · **PDF:** https://arxiv.org/pdf/2601.20300v1
+
+## 18661. Truthfulness Despite Weak Supervision: Evaluating and Training LLMs Using Peer Prediction
+
+- **ID:** `2601.20299` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Tianyi Alex Qiu, Micah Carroll, Cameron Allen
+- **abs:** https://arxiv.org/abs/2601.20299 · **PDF:** https://arxiv.org/pdf/2601.20299v1
+
+## 18662. PredictionMarketBench: A SWE-bench-Style Framework for Backtesting Trading Agents on Prediction Markets
+
+- **ID:** `2602.00133` · **primary:** `q-fin.ST` · **Submitted:** 2026-01-28
+- **Authors:** Avi Arora, Ritesh Malpani
+- **abs:** https://arxiv.org/abs/2602.00133 · **PDF:** https://arxiv.org/pdf/2602.00133v1
+
+## 18663. Cheap2Rich: A Multi-Fidelity Framework for Data Assimilation and System Identification of Multiscale Physics -- Rotating Detonation Engines
+
+- **ID:** `2601.20295` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Yuxuan Bao, Jan Zajac, Megan Powers, et al.
+- **abs:** https://arxiv.org/abs/2601.20295 · **PDF:** https://arxiv.org/pdf/2601.20295v1
+
+## 18664. A Learning-based Framework for Spatial Impulse Response Compensation in 3D Photoacoustic Computed Tomography
+
+- **ID:** `2601.20291` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Kaiyi Yang, Seonyeong Park, Gangwon Jeong, et al.
+- **abs:** https://arxiv.org/abs/2601.20291 · **PDF:** https://arxiv.org/pdf/2601.20291v1
+
+## 18665. A New Mode of Teaching Chinese as a Foreign Language from the Perspective of Smart System Studied by Using Rongzhixue
+
+- **ID:** `2602.06992` · **primary:** `cs.CY` · **Submitted:** 2026-01-28
+- **Authors:** Xiaohui Zou, Lijun Ke, Shunpeng Zou
+- **abs:** https://arxiv.org/abs/2602.06992 · **PDF:** https://arxiv.org/pdf/2602.06992v1
+
+## 18666. ATTNSOM: Learning Cross-Isoform Attention for Cytochrome P450 Site-of-Metabolism
+
+- **ID:** `2601.20891` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-28
+- **Authors:** Hajung Kim, Eunha Lee, Sohyun Chung, et al.
+- **abs:** https://arxiv.org/abs/2601.20891 · **PDF:** https://arxiv.org/pdf/2601.20891v1
+
+## 18667. One Word is Enough: Minimal Adversarial Perturbations for Neural Text Ranking
+
+- **ID:** `2601.20283` · **primary:** `cs.IR` · **Submitted:** 2026-01-28
+- **Authors:** Tanmay Karmakar, Sourav Saha, Debapriyo Majumdar, et al.
+- **abs:** https://arxiv.org/abs/2601.20283 · **PDF:** https://arxiv.org/pdf/2601.20283v1
+
+## 18668. Memory Retrieval in Transformers: Insights from The Encoding Specificity Principle
+
+- **ID:** `2601.20282` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Viet Hung Dinh, Ming Ding, Youyang Qu, et al.
+- **abs:** https://arxiv.org/abs/2601.20282 · **PDF:** https://arxiv.org/pdf/2601.20282v1
+
+## 18669. ConfSpec: Efficient Step-Level Speculative Reasoning via Confidence-Gated Verification
+
+- **ID:** `2602.18447` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Siran Liu, Cyril Y. He
+- **abs:** https://arxiv.org/abs/2602.18447 · **PDF:** https://arxiv.org/pdf/2602.18447v1
+
+## 18670. The Forecast After the Forecast: A Post-Processing Shift in Time Series
+
+- **ID:** `2601.20280` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Daojun Liang, Qi Li, Yinglong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.20280 · **PDF:** https://arxiv.org/pdf/2601.20280v1
+
+## 18671. Beyond the Needle's Illusion: Decoupled Evaluation of Evidence Access and Use under Semantic Interference at 326M-Token Scale
+
+- **ID:** `2601.20276` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Tianwei Lin, Zuyi Zhou, Xinda Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.20276 · **PDF:** https://arxiv.org/pdf/2601.20276v1
+
+## 18672. RusLICA: A Russian-Language Platform for Automated Linguistic Inquiry and Category Analysis
+
+- **ID:** `2601.20275` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Elina Sigdel, Anastasia Panfilova
+- **abs:** https://arxiv.org/abs/2601.20275 · **PDF:** https://arxiv.org/pdf/2601.20275v1
+
+## 18673. Eliciting Least-to-Most Reasoning for Phishing URL Detection
+
+- **ID:** `2601.20270` · **primary:** `cs.CR` · **Submitted:** 2026-01-28
+- **Authors:** Holly Trikilis, Pasindu Marasinghe, Fariza Rashid, et al.
+- **abs:** https://arxiv.org/abs/2601.20270 · **PDF:** https://arxiv.org/pdf/2601.20270v1
+
+## 18674. Empirical Likelihood-Based Fairness Auditing: Distribution-Free Certification and Flagging
+
+- **ID:** `2601.20269` · **primary:** `stat.ML` · **Submitted:** 2026-01-28
+- **Authors:** Jie Tang, Chuanlong Xie, Xianli Zeng, et al.
+- **abs:** https://arxiv.org/abs/2601.20269 · **PDF:** https://arxiv.org/pdf/2601.20269v2
+
+## 18675. Robust SDE Parameter Estimation Under Missing Time Information Setting
+
+- **ID:** `2601.20268` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Long Van Tran, Truyen Tran, Phuoc Nguyen
+- **abs:** https://arxiv.org/abs/2601.20268 · **PDF:** https://arxiv.org/pdf/2601.20268v1
+
+## 18676. MrRoPE: Mixed-radix Rotary Position Embedding
+
+- **ID:** `2601.22181` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Qingyuan Tian, Wenhong Zhu, Xiaoran Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.22181 · **PDF:** https://arxiv.org/pdf/2601.22181v1
+
+## 18677. C2:Cross learning module enhanced decision transformer with Constraint-aware loss for auto-bidding
+
+- **ID:** `2601.20257` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Jinren Ding, Xuejian Xu, Shen Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.20257 · **PDF:** https://arxiv.org/pdf/2601.20257v2
+
+## 18678. SoftHateBench: Evaluating Moderation Models Against Reasoning-Driven, Policy-Compliant Hostility
+
+- **ID:** `2601.20256` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Xuanyu Su, Diana Inkpen, Nathalie Japkowicz
+- **abs:** https://arxiv.org/abs/2601.20256 · **PDF:** https://arxiv.org/pdf/2601.20256v1
+
+## 18679. HE-SNR: Uncovering Latent Logic via Entropy for Guiding Mid-Training on SWE-BENCH
+
+- **ID:** `2601.20255` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Yueyang Wang, Jiawei Fu, Baolong Bi, et al.
+- **abs:** https://arxiv.org/abs/2601.20255 · **PDF:** https://arxiv.org/pdf/2601.20255v1
+
+## 18680. Automated Benchmark Generation from Domain Guidelines Informed by Bloom's Taxonomy
+
+- **ID:** `2601.20253` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Si Chen, Le Huy Khiem, Annalisa Szymanski, et al.
+- **abs:** https://arxiv.org/abs/2601.20253 · **PDF:** https://arxiv.org/pdf/2601.20253v1
+
+## 18681. Efficient Evaluation of LLM Performance with Statistical Guarantees
+
+- **ID:** `2601.20251` · **primary:** `stat.ML` · **Submitted:** 2026-01-28
+- **Authors:** Skyler Wu, Yash Nair, Emmanuel J. Candès
+- **abs:** https://arxiv.org/abs/2601.20251 · **PDF:** https://arxiv.org/pdf/2601.20251v2
+
+## 18682. Order-Optimal Sample Complexity of Rectified Flows
+
+- **ID:** `2601.20250` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Hari Krishna Sahoo, Mudit Gaur, Vaneet Aggarwal
+- **abs:** https://arxiv.org/abs/2601.20250 · **PDF:** https://arxiv.org/pdf/2601.20250v1
+
+## 18683. SW-ASR: A Context-Aware Hybrid ASR Pipeline for Robust Single Word Speech Recognition
+
+- **ID:** `2601.20890` · **primary:** `cs.SD` · **Submitted:** 2026-01-28
+- **Authors:** Manali Sharma, Riya Naik, Buvaneshwari G
+- **abs:** https://arxiv.org/abs/2601.20890 · **PDF:** https://arxiv.org/pdf/2601.20890v1
+
+## 18684. How AI Impacts Skill Formation
+
+- **ID:** `2601.20245` · **primary:** `cs.CY` · **Submitted:** 2026-01-28
+- **Authors:** Judy Hanwen Shen, Alex Tamkin
+- **abs:** https://arxiv.org/abs/2601.20245 · **PDF:** https://arxiv.org/pdf/2601.20245v2
+
+## 18685. On the Relationship Between Representation Geometry and Generalization in Deep Neural Networks
+
+- **ID:** `2602.00130` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Sumit Yadav
+- **abs:** https://arxiv.org/abs/2602.00130 · **PDF:** https://arxiv.org/pdf/2602.00130v2
+
+## 18686. MALLOC: Benchmarking the Memory-aware Long Sequence Compression for Large Sequential Recommendation
+
+- **ID:** `2601.20234` · **primary:** `cs.IR` · **Submitted:** 2026-01-28
+- **Authors:** Qihang Yu, Kairui Fu, Zhaocheng Du, et al.
+- **abs:** https://arxiv.org/abs/2601.20234 · **PDF:** https://arxiv.org/pdf/2601.20234v2
+
+## 18687. Certificate-Guided Pruning for Stochastic Lipschitz Optimization
+
+- **ID:** `2601.20231` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma
+- **abs:** https://arxiv.org/abs/2601.20231 · **PDF:** https://arxiv.org/pdf/2601.20231v1
+
+## 18688. Unit-Based Agent for Semi-Cascaded Full-Duplex Dialogue Systems
+
+- **ID:** `2601.20230` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Haoyuan Yu, Yuxuan Chen, Minjie Cai
+- **abs:** https://arxiv.org/abs/2601.20230 · **PDF:** https://arxiv.org/pdf/2601.20230v2
+
+## 18689. Proactive SFC Provisioning with Forecast-Driven DRL in Data Centers
+
+- **ID:** `2601.20229` · **primary:** `cs.NI` · **Submitted:** 2026-01-28
+- **Authors:** Parisa Fard Moshiri, Poonam Lohan, Burak Kantarci, et al.
+- **abs:** https://arxiv.org/abs/2601.20229 · **PDF:** https://arxiv.org/pdf/2601.20229v1
+
+## 18690. Quantum statistics from classical simulations via generative Gibbs sampling
+
+- **ID:** `2601.20228` · **primary:** `physics.chem-ph` · **Submitted:** 2026-01-28
+- **Authors:** Weizhou Wang, Xuanxi Zhang, Jonathan Weare, et al.
+- **abs:** https://arxiv.org/abs/2601.20228 · **PDF:** https://arxiv.org/pdf/2601.20228v1
+
+## 18691. ProFlow: Zero-Shot Physics-Consistent Sampling via Proximal Flow Guidance
+
+- **ID:** `2601.20227` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Zichao Yu, Ming Li, Wenyi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.20227 · **PDF:** https://arxiv.org/pdf/2601.20227v1
+
+## 18692. Parametric and Generative Forecasts of Day-Ahead Market Curves for Storage Optimization
+
+- **ID:** `2601.20226` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Julian Gutierrez, Redouane Silvente
+- **abs:** https://arxiv.org/abs/2601.20226 · **PDF:** https://arxiv.org/pdf/2601.20226v1
+
+## 18693. Scaling Medical Reasoning Verification via Tool-Integrated Reinforcement Learning
+
+- **ID:** `2601.20221` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Hang Zhang, Ruheng Wang, Yuelyu Ji, et al.
+- **abs:** https://arxiv.org/abs/2601.20221 · **PDF:** https://arxiv.org/pdf/2601.20221v1
+
+## 18694. Latent-IMH: Efficient Bayesian Inference for Inverse Problems with Approximate Operators
+
+- **ID:** `2601.20888` · **primary:** `stat.ML` · **Submitted:** 2026-01-28
+- **Authors:** Youguang Chen, George Biros
+- **abs:** https://arxiv.org/abs/2601.20888 · **PDF:** https://arxiv.org/pdf/2601.20888v2
+
+## 18695. An Accounting Identity for Algorithmic Fairness
+
+- **ID:** `2601.20217` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Hadi Elzayn, Jacob Goldin
+- **abs:** https://arxiv.org/abs/2601.20217 · **PDF:** https://arxiv.org/pdf/2601.20217v1
+
+## 18696. Spark: Strategic Policy-Aware Exploration via Dynamic Branching for Long-Horizon Agentic Learning
+
+- **ID:** `2601.20209` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Jinyang Wu, Shuo Yang, Changpeng Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.20209 · **PDF:** https://arxiv.org/pdf/2601.20209v1
+
+## 18697. Monte Carlo Tree Search for Execution-Guided Program Repair with Large Language Models
+
+- **ID:** `2602.00129` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Yixuan Liang
+- **abs:** https://arxiv.org/abs/2602.00129 · **PDF:** https://arxiv.org/pdf/2602.00129v1
+
+## 18698. CADENT: Gated Hybrid Distillation for Sample-Efficient Transfer in Reinforcement Learning
+
+- **ID:** `2602.02532` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Mahyar Alinejad, Yue Wang, George Atia
+- **abs:** https://arxiv.org/abs/2602.02532 · **PDF:** https://arxiv.org/pdf/2602.02532v1
+
+## 18699. Towards Intelligent Urban Park Development Monitoring: LLM Agents for Multi-Modal Information Fusion and Analysis
+
+- **ID:** `2601.20206` · **primary:** `cs.AI` · **Submitted:** 2026-01-28
+- **Authors:** Zixuan Xiao, Chunguang Hu, Jun Ma
+- **abs:** https://arxiv.org/abs/2601.20206 · **PDF:** https://arxiv.org/pdf/2601.20206v1
+
+## 18700. Hyperparameter Transfer with Mixture-of-Expert Layers
+
+- **ID:** `2601.20205` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Tianze Jiang, Blake Bordelon, Cengiz Pehlevan, et al.
+- **abs:** https://arxiv.org/abs/2601.20205 · **PDF:** https://arxiv.org/pdf/2601.20205v2
+
+## 18701. Minimum-Cost Network Flow with Dual Predictions
+
+- **ID:** `2601.20203` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Zhiyang Chen, Hailong Yao, Xia Yin
+- **abs:** https://arxiv.org/abs/2601.20203 · **PDF:** https://arxiv.org/pdf/2601.20203v1
+
+## 18702. DeRaDiff: Denoising Time Realignment of Diffusion Models
+
+- **ID:** `2601.20198` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Ratnavibusena Don Shahain Manujith, Teoh Tze Tzun, Kenji Kawaguchi, et al.
+- **abs:** https://arxiv.org/abs/2601.20198 · **PDF:** https://arxiv.org/pdf/2601.20198v2
+
+## 18703. Bias-Reduced Estimation of Finite Mixtures: An Application to Latent Group Structures in Panel Data
+
+- **ID:** `2601.20197` · **primary:** `stat.ME` · **Submitted:** 2026-01-28
+- **Authors:** Raphaël Langevin
+- **abs:** https://arxiv.org/abs/2601.20197 · **PDF:** https://arxiv.org/pdf/2601.20197v2
+
+## 18704. Meta-Cognitive Reinforcement Learning with Self-Doubt and Recovery
+
+- **ID:** `2601.20193` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Zhipeng Zhang, Xiongfei Su, Kai Li
+- **abs:** https://arxiv.org/abs/2601.20193 · **PDF:** https://arxiv.org/pdf/2601.20193v2
+
+## 18705. Improving X-Codec-2.0 for Multi-Lingual Speech: 25 Hz Latent Rate and 24 kHz Sampling
+
+- **ID:** `2601.20185` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Husein Zolkepli
+- **abs:** https://arxiv.org/abs/2601.20185 · **PDF:** https://arxiv.org/pdf/2601.20185v2
+
+## 18706. On the Computational Complexity of Performative Prediction
+
+- **ID:** `2601.20180` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Ioannis Anagnostides, Rohan Chauhan, Ioannis Panageas, et al.
+- **abs:** https://arxiv.org/abs/2601.20180 · **PDF:** https://arxiv.org/pdf/2601.20180v1
+
+## 18707. Causal-Driven Feature Evaluation for Cross-Domain Image Classification
+
+- **ID:** `2601.20176` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Chen Cheng, Ang Li
+- **abs:** https://arxiv.org/abs/2601.20176 · **PDF:** https://arxiv.org/pdf/2601.20176v2
+
+## 18708. NeuraLSP: An Efficient and Rigorous Neural Left Singular Subspace Preconditioner for Conjugate Gradient Methods
+
+- **ID:** `2601.20174` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Alexander Benanti, Xi Han, Hong Qin
+- **abs:** https://arxiv.org/abs/2601.20174 · **PDF:** https://arxiv.org/pdf/2601.20174v2
+
+## 18709. MAPLE: Self-supervised Learning-Enhanced Nonlinear Dimensionality Reduction for Visual Analysis
+
+- **ID:** `2601.20173` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Zeyang Huang, Takanori Fujiwara, Angelos Chatzimparmpas, et al.
+- **abs:** https://arxiv.org/abs/2601.20173 · **PDF:** https://arxiv.org/pdf/2601.20173v1
+
+## 18710. Loss Landscape Geometry and the Learning of Symmetries: Or, What Influence Functions Reveal About Robust Generalization
+
+- **ID:** `2601.20172` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** James Amarel, Robyn Miller, Nicolas Hengartner, et al.
+- **abs:** https://arxiv.org/abs/2601.20172 · **PDF:** https://arxiv.org/pdf/2601.20172v1
+
+## 18711. Local Duality for Sparse Support Vector Machines
+
+- **ID:** `2601.20170` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Penghe Zhang, Naihua Xiu, Houduo Qi
+- **abs:** https://arxiv.org/abs/2601.20170 · **PDF:** https://arxiv.org/pdf/2601.20170v1
+
+## 18712. Continual Fine-Tuning with Provably Accurate and Parameter-Free Task Retrieval
+
+- **ID:** `2603.13235` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Hang Thi-Thuy Le, Long Minh Bui, Minh Hoang, et al.
+- **abs:** https://arxiv.org/abs/2603.13235 · **PDF:** https://arxiv.org/pdf/2603.13235v1
+
+## 18713. IDE-Bench: Evaluating Large Language Models as IDE Agents on Real-World Software Engineering Tasks
+
+- **ID:** `2601.20886` · **primary:** `cs.SE` · **Submitted:** 2026-01-28
+- **Authors:** Spencer Mateega, Jeff Yang, Tiana Costello, et al.
+- **abs:** https://arxiv.org/abs/2601.20886 · **PDF:** https://arxiv.org/pdf/2601.20886v2
+
+## 18714. Disaster Question Answering with LoRA Efficiency and Accurate End Position
+
+- **ID:** `2602.21212` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Takato Yasuno
+- **abs:** https://arxiv.org/abs/2602.21212 · **PDF:** https://arxiv.org/pdf/2602.21212v1
+
+## 18715. What's the plan? Metrics for implicit planning in LLMs and their application to rhyme generation and question answering
+
+- **ID:** `2601.20164` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Jim Maar, Denis Paperno, Callum Stuart McDougall, et al.
+- **abs:** https://arxiv.org/abs/2601.20164 · **PDF:** https://arxiv.org/pdf/2601.20164v1
+
+## 18716. Me-Agent: A Personalized Mobile Agent with Two-Level User Habit Learning for Enhanced Interaction
+
+- **ID:** `2601.20162` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Shuoxin Wang, Chang Liu, Gowen Loo, et al.
+- **abs:** https://arxiv.org/abs/2601.20162 · **PDF:** https://arxiv.org/pdf/2601.20162v1
+
+## 18717. PASS: Certified Subset Repair for Classical and Quantum Pairwise Constrained Clustering
+
+- **ID:** `2601.20157` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Pedro Chumpitaz-Flores, My Duong, Ying Mao, et al.
+- **abs:** https://arxiv.org/abs/2601.20157 · **PDF:** https://arxiv.org/pdf/2601.20157v2
+
+## 18718. Quantum Model Parallelism for MRI-Based Classification of Alzheimer's Disease Stages
+
+- **ID:** `2602.00128` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Emine Akpinar, Murat Oduncuoglu
+- **abs:** https://arxiv.org/abs/2602.00128 · **PDF:** https://arxiv.org/pdf/2602.00128v1
+
+## 18719. Spectral Ghost in Representation Learning: from Component Analysis to Self-Supervised Learning
+
+- **ID:** `2601.20154` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Bo Dai, Na Li, Dale Schuurmans
+- **abs:** https://arxiv.org/abs/2601.20154 · **PDF:** https://arxiv.org/pdf/2601.20154v2
+
+## 18720. Concentration Inequalities for Exchangeable Tensors and Matrix-valued Data
+
+- **ID:** `2601.20152` · **primary:** `math.ST` · **Submitted:** 2026-01-28
+- **Authors:** Chen Cheng, Rina Foygel Barber
+- **abs:** https://arxiv.org/abs/2601.20152 · **PDF:** https://arxiv.org/pdf/2601.20152v2
+
+## 18721. LogSieve: Task-Aware CI Log Reduction for Sustainable LLM-Based Analysis
+
+- **ID:** `2601.20148` · **primary:** `cs.SE` · **Submitted:** 2026-01-28
+- **Authors:** Marcus Emmanuel Barnes, Taher A. Ghaleb, Safwat Hassan
+- **abs:** https://arxiv.org/abs/2601.20148 · **PDF:** https://arxiv.org/pdf/2601.20148v1
+
+## 18722. Trajectory2Task: Training Robust Tool-Calling Agents with Synthesized Yet Verifiable Data for Complex User Intents
+
+- **ID:** `2601.20144` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Ziyi Wang, Yuxuan Lu, Yimeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.20144 · **PDF:** https://arxiv.org/pdf/2601.20144v2
+
+## 18723. Mind the Shift: Using Delta SSL Embeddings to Enhance Child ASR
+
+- **ID:** `2601.20142` · **primary:** `cs.CL` · **Submitted:** 2026-01-28
+- **Authors:** Zilai Wang, Natarajan Balaji Shankar, Kaiyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.20142 · **PDF:** https://arxiv.org/pdf/2601.20142v1
+
+## 18724. Large language models accurately predict public perceptions of support for climate action worldwide
+
+- **ID:** `2601.20141` · **primary:** `cs.CY` · **Submitted:** 2026-01-28
+- **Authors:** Nattavudh Powdthavee, Sandra J. Geiger
+- **abs:** https://arxiv.org/abs/2601.20141 · **PDF:** https://arxiv.org/pdf/2601.20141v1
+
+## 18725. ALIGN: Aligned Delegation with Performance Guarantees for Multi-Agent LLM Reasoning
+
+- **ID:** `2602.00127` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Tong Zhu, Baiting Chen, Jin Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.00127 · **PDF:** https://arxiv.org/pdf/2602.00127v1
+
+## 18726. Scaling Next-Brain-Token Prediction for MEG
+
+- **ID:** `2601.20138` · **primary:** `cs.LG` · **Submitted:** 2026-01-28
+- **Authors:** Richard Csaky
+- **abs:** https://arxiv.org/abs/2601.20138 · **PDF:** https://arxiv.org/pdf/2601.20138v2
+
+## 18727. Taxonomy of the Retrieval System Framework: Pitfalls and Paradigms
+
+- **ID:** `2601.20131` · **primary:** `cs.IR` · **Submitted:** 2026-01-27
+- **Authors:** Deep Shah, Sanket Badhe, Nehal Kathrotia
+- **abs:** https://arxiv.org/abs/2601.20131 · **PDF:** https://arxiv.org/pdf/2601.20131v1
+
+## 18728. BengaliSent140: A Large-Scale Bengali Binary Sentiment Dataset for Hate and Non-Hate Speech Classification
+
+- **ID:** `2601.20129` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Akif Islam, Sujan Kumar Roy, Md. Ekramul Hamid
+- **abs:** https://arxiv.org/abs/2601.20129 · **PDF:** https://arxiv.org/pdf/2601.20129v1
+
+## 18729. Machine learning enhanced data assimilation framework for multiscale carbonate rock characterization
+
+- **ID:** `2602.06989` · **primary:** `physics.geo-ph` · **Submitted:** 2026-01-27
+- **Authors:** Zhenkai Bo, Ahmed H. Elsheikh, Hannah P. Menke, et al.
+- **abs:** https://arxiv.org/abs/2602.06989 · **PDF:** https://arxiv.org/pdf/2602.06989v1
+
+## 18730. Rewarding Intellectual Humility Learning When Not To Answer In Large Language Models
+
+- **ID:** `2601.20126` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Abha Jha, Akanksha Mahajan, Ashwath Vaithinathan Aravindan, et al.
+- **abs:** https://arxiv.org/abs/2601.20126 · **PDF:** https://arxiv.org/pdf/2601.20126v1
+
+## 18731. Membership Inference Attacks Against Fine-tuned Diffusion Language Models
+
+- **ID:** `2601.20125` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Yuetian Chen, Kaiyuan Zhang, Yuntao Du, et al.
+- **abs:** https://arxiv.org/abs/2601.20125 · **PDF:** https://arxiv.org/pdf/2601.20125v3
+
+## 18732. Going NUTS with ADVI: Exploring various Bayesian Inference techniques with Facebook Prophet
+
+- **ID:** `2601.20120` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Jovan Krajevski, Biljana Tojtovska Ribarski
+- **abs:** https://arxiv.org/abs/2601.20120 · **PDF:** https://arxiv.org/pdf/2601.20120v1
+
+## 18733. A Reinforcement Learning Based Universal Sequence Design for Polar Codes
+
+- **ID:** `2601.20118` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** David Kin Wai Ho, Arman Fazeli, Mohamad M. Mansour, et al.
+- **abs:** https://arxiv.org/abs/2601.20118 · **PDF:** https://arxiv.org/pdf/2601.20118v1
+
+## 18734. In-Context Reinforcement Learning From Suboptimal Historical Data
+
+- **ID:** `2601.20116` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Juncheng Dong, Moyang Guo, Ethan X. Fang, et al.
+- **abs:** https://arxiv.org/abs/2601.20116 · **PDF:** https://arxiv.org/pdf/2601.20116v1
+
+## 18735. How Much Progress Has There Been in NVIDIA Datacenter GPUs?
+
+- **ID:** `2601.20115` · **primary:** `cs.AR` · **Submitted:** 2026-01-27
+- **Authors:** Emanuele Del Sozzo, Martin Fleming, Kenneth Flamm, et al.
+- **abs:** https://arxiv.org/abs/2601.20115 · **PDF:** https://arxiv.org/pdf/2601.20115v2
+
+## 18736. Look in the Middle: Structural Anchor Pruning for Scalable Visual RAG Indexing
+
+- **ID:** `2601.20107` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Zhuchenyang Liu, Ziyu Hu, Yao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.20107 · **PDF:** https://arxiv.org/pdf/2601.20107v1
+
+## 18737. FFE-Hallu:Hallucinations in Fixed Figurative Expressions:Benchmark of Idioms and Proverbs in the Persian Language
+
+- **ID:** `2601.20105` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Faezeh Hosseini, Mohammadali Yousefzadeh, Yadollah Yaghoobzadeh
+- **abs:** https://arxiv.org/abs/2601.20105 · **PDF:** https://arxiv.org/pdf/2601.20105v1
+
+## 18738. Benchmarking Reward Hack Detection in Code Environments via Contrastive Analysis
+
+- **ID:** `2601.20103` · **primary:** `cs.SE` · **Submitted:** 2026-01-27
+- **Authors:** Darshan Deshpande, Anand Kannappan, Rebecca Qian
+- **abs:** https://arxiv.org/abs/2601.20103 · **PDF:** https://arxiv.org/pdf/2601.20103v1
+
+## 18739. Counterfactual Cultural Cues Reduce Medical QA Accuracy in LLMs: Identifier vs Context Effects
+
+- **ID:** `2601.20102` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Amirhossein Haji Mohammad Rezaei, Zahra Shakeri
+- **abs:** https://arxiv.org/abs/2601.20102 · **PDF:** https://arxiv.org/pdf/2601.20102v1
+
+## 18740. Taming Toxic Talk: Using chatbots to intervene with users posting toxic comments
+
+- **ID:** `2601.20100` · **primary:** `cs.HC` · **Submitted:** 2026-01-27
+- **Authors:** Jeremy Foote, Deepak Kumar, Bedadyuti Jha, et al.
+- **abs:** https://arxiv.org/abs/2601.20100 · **PDF:** https://arxiv.org/pdf/2601.20100v1
+
+## 18741. Hypersonic Flow Control: Generalized Deep Reinforcement Learning for Hypersonic Intake Unstart Control under Uncertainty
+
+- **ID:** `2602.02531` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Trishit Mondal, Ameya D. Jagtap
+- **abs:** https://arxiv.org/abs/2602.02531 · **PDF:** https://arxiv.org/pdf/2602.02531v1
+
+## 18742. Dynamics of Human-AI Collective Knowledge on the Web: A Scalable Model and Insights for Sustainable Growth
+
+- **ID:** `2601.20099` · **primary:** `cs.CY` · **Submitted:** 2026-01-27
+- **Authors:** Buddhika Nettasinghe, Kang Zhao
+- **abs:** https://arxiv.org/abs/2601.20099 · **PDF:** https://arxiv.org/pdf/2601.20099v1
+
+## 18743. Should I Have Expressed a Different Intent? Counterfactual Generation for LLM-Based Autonomous Control
+
+- **ID:** `2601.20090` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Amirmohammad Farzaneh, Salvatore D'Oro, Osvaldo Simeone
+- **abs:** https://arxiv.org/abs/2601.20090 · **PDF:** https://arxiv.org/pdf/2601.20090v2
+
+## 18744. Quantization-Aware Distillation for NVFP4 Inference Accuracy Recovery
+
+- **ID:** `2601.20088` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Meng Xin, Sweta Priyadarshi, Jingyu Xin, et al.
+- **abs:** https://arxiv.org/abs/2601.20088 · **PDF:** https://arxiv.org/pdf/2601.20088v3
+
+## 18745. LLaTTE: Scaling Laws for Multi-Stage Sequence Modeling in Large-Scale Ads Recommendation
+
+- **ID:** `2601.20083` · **primary:** `cs.IR` · **Submitted:** 2026-01-27
+- **Authors:** Lee Xiong, Zhirong Chen, Rahul Mayuranath, et al.
+- **abs:** https://arxiv.org/abs/2601.20083 · **PDF:** https://arxiv.org/pdf/2601.20083v1
+
+## 18746. Techno-economic optimization of a heat-pipe microreactor, part II: multi-objective optimization analysis
+
+- **ID:** `2601.20079` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Paul Seurin, Dean Price
+- **abs:** https://arxiv.org/abs/2601.20079 · **PDF:** https://arxiv.org/pdf/2601.20079v1
+
+## 18747. Randomized Feasibility Methods for Constrained Optimization with Adaptive Step Sizes
+
+- **ID:** `2601.20076` · **primary:** `math.OC` · **Submitted:** 2026-01-27
+- **Authors:** Abhishek Chakraborty, Angelia Nedić
+- **abs:** https://arxiv.org/abs/2601.20076 · **PDF:** https://arxiv.org/pdf/2601.20076v1
+
+## 18748. Formulating Reinforcement Learning for Human-Robot Collaboration through Off-Policy Evaluation
+
+- **ID:** `2602.02530` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Saurav Singh, Rodney Sanchez, Alexander Ororbia, et al.
+- **abs:** https://arxiv.org/abs/2602.02530 · **PDF:** https://arxiv.org/pdf/2602.02530v1
+
+## 18749. Semi-Supervised Masked Autoencoders: Unlocking Vision Transformer Potential with Limited Data
+
+- **ID:** `2601.20072` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Atik Faysal, Mohammad Rostami, Reihaneh Gh. Roshan, et al.
+- **abs:** https://arxiv.org/abs/2601.20072 · **PDF:** https://arxiv.org/pdf/2601.20072v1
+
+## 18750. Distributional value gradients for stochastic environments
+
+- **ID:** `2601.20071` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Baptiste Debes, Tinne Tuytelaars
+- **abs:** https://arxiv.org/abs/2601.20071 · **PDF:** https://arxiv.org/pdf/2601.20071v3
+
+## 18751. Domain Expansion: A Latent Space Construction Framework for Multi-Task Learning
+
+- **ID:** `2601.20069` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Chi-Yao Huang, Khoa Vo, Aayush Atul Verma, et al.
+- **abs:** https://arxiv.org/abs/2601.20069 · **PDF:** https://arxiv.org/pdf/2601.20069v2
+
+## 18752. MiniTensor: A Lightweight, High-Performance Tensor Operations Library
+
+- **ID:** `2602.00125` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Soumyadip Sarkar
+- **abs:** https://arxiv.org/abs/2602.00125 · **PDF:** https://arxiv.org/pdf/2602.00125v1
+
+## 18753. VERGE: Formal Refinement and Guidance Engine for Verifiable LLM Reasoning
+
+- **ID:** `2601.20055` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Vikash Singh, Darion Cassel, Nathaniel Weir, et al.
+- **abs:** https://arxiv.org/abs/2601.20055 · **PDF:** https://arxiv.org/pdf/2601.20055v1
+
+## 18754. Explainable deep learning reveals the physical mechanisms behind the turbulent kinetic energy equation
+
+- **ID:** `2601.20052` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-01-27
+- **Authors:** Francisco Alcántara-Ávila, Andrés Cremades, Sergio Hoyas, et al.
+- **abs:** https://arxiv.org/abs/2601.20052 · **PDF:** https://arxiv.org/pdf/2601.20052v1
+
+## 18755. Size Matters: Reconstructing Real-Scale 3D Models from Monocular Images for Food Portion Estimation
+
+- **ID:** `2601.20051` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Gautham Vinod, Bruce Coburn, Siddeshwar Raghavan, et al.
+- **abs:** https://arxiv.org/abs/2601.20051 · **PDF:** https://arxiv.org/pdf/2601.20051v1
+
+## 18756. Insight Agents: An LLM-Based Multi-Agent System for Data Insights
+
+- **ID:** `2601.20048` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Jincheng Bai, Zhenyu Zhang, Jennifer Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.20048 · **PDF:** https://arxiv.org/pdf/2601.20048v2
+
+## 18757. Minimax Rates for Hyperbolic Hierarchical Learning
+
+- **ID:** `2601.20047` · **primary:** `stat.ML` · **Submitted:** 2026-01-27
+- **Authors:** Divit Rawal, Sriram Vishwanath
+- **abs:** https://arxiv.org/abs/2601.20047 · **PDF:** https://arxiv.org/pdf/2601.20047v1
+
+## 18758. Externally Validated Longitudinal GRU Model for Visit-Level 180-Day Mortality Risk in Metastatic Castration-Resistant Prostate Cancer
+
+- **ID:** `2601.20046` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Javier Mencia-Ledo, Mohammad Noaeen, Zahra Shakeri
+- **abs:** https://arxiv.org/abs/2601.20046 · **PDF:** https://arxiv.org/pdf/2601.20046v1
+
+## 18759. Regime-Adaptive Bayesian Optimization via Dirichlet Process Mixtures of Gaussian Processes
+
+- **ID:** `2601.20043` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Yan Zhang, Xuefeng Liu, Sipeng Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.20043 · **PDF:** https://arxiv.org/pdf/2601.20043v1
+
+## 18760. CiMRAG: CiM-Aware Domain-Adaptive and Noise-Resilient Retrieval-Augmented Generation for Edge-Based LLMs
+
+- **ID:** `2601.20041` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Shih-Hsuan Chiu, Ming-Syan Chen
+- **abs:** https://arxiv.org/abs/2601.20041 · **PDF:** https://arxiv.org/pdf/2601.20041v3
+
+## 18761. Finetune-Informed Pretraining Boosts Downstream Performance
+
+- **ID:** `2601.20884` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Atik Faysal, Mohammad Rostami, Reihaneh Gh. Roshan, et al.
+- **abs:** https://arxiv.org/abs/2601.20884 · **PDF:** https://arxiv.org/pdf/2601.20884v1
+
+## 18762. CheckIfExist: Detecting Citation Hallucinations in the Era of AI-Generated Content
+
+- **ID:** `2602.15871` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Diletta Abbonato
+- **abs:** https://arxiv.org/abs/2602.15871 · **PDF:** https://arxiv.org/pdf/2602.15871v1
+
+## 18763. Structural Compositional Function Networks: Interpretable Functional Compositions for Tabular Discovery
+
+- **ID:** `2601.20037` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Fang Li
+- **abs:** https://arxiv.org/abs/2601.20037 · **PDF:** https://arxiv.org/pdf/2601.20037v1
+
+## 18764. The Sound of Noise: Leveraging the Inductive Bias of Pre-trained Audio Transformers for Glitch Identification in LIGO
+
+- **ID:** `2601.20034` · **primary:** `astro-ph.IM` · **Submitted:** 2026-01-27
+- **Authors:** Suyash Deshmukh, Chayan Chatterjee, Abigail Petulante, et al.
+- **abs:** https://arxiv.org/abs/2601.20034 · **PDF:** https://arxiv.org/pdf/2601.20034v1
+
+## 18765. TAIGR: Towards Modeling Influencer Content on Social Media via Structured, Pragmatic Inference
+
+- **ID:** `2601.20032` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Nishanth Sridhar Nakshatri, Eylon Caplan, Rajkumar Pujari, et al.
+- **abs:** https://arxiv.org/abs/2601.20032 · **PDF:** https://arxiv.org/pdf/2601.20032v1
+
+## 18766. Decomposing multimodal embedding spaces with group-sparse autoencoders
+
+- **ID:** `2601.20028` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Chiraag Kaushik, Davis Barch, Andrea Fanelli
+- **abs:** https://arxiv.org/abs/2601.20028 · **PDF:** https://arxiv.org/pdf/2601.20028v1
+
+## 18767. Semantic Uncertainty Quantification of Hallucinations in LLMs: A Quantum Tensor Network Based Method
+
+- **ID:** `2601.20026` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Pragatheeswaran Vipulanandan, Kamal Premaratne, Dilip Sarkar
+- **abs:** https://arxiv.org/abs/2601.20026 · **PDF:** https://arxiv.org/pdf/2601.20026v1
+
+## 18768. Fuzzy Categorical Planning: Autonomous Goal Satisfaction with Graded Semantic Constraints
+
+- **ID:** `2601.20021` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Shuhui Qu
+- **abs:** https://arxiv.org/abs/2601.20021 · **PDF:** https://arxiv.org/pdf/2601.20021v1
+
+## 18769. Matching and mixing: Matchability of graphs under Markovian error
+
+- **ID:** `2601.20020` · **primary:** `math.ST` · **Submitted:** 2026-01-27
+- **Authors:** Zhirui Li, Keith D. Levin, Zhiang Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.20020 · **PDF:** https://arxiv.org/pdf/2601.20020v1
+
+## 18770. VoxMorph: Scalable Zero-shot Voice Identity Morphing via Disentangled Embeddings
+
+- **ID:** `2601.20883` · **primary:** `cs.SD` · **Submitted:** 2026-01-27
+- **Authors:** Bharath Krishnamurthy, Ajita Rattani
+- **abs:** https://arxiv.org/abs/2601.20883 · **PDF:** https://arxiv.org/pdf/2601.20883v1
+
+## 18771. Teaching LLMs to Ask: Self-Querying Category-Theoretic Planning for Under-Specified Reasoning
+
+- **ID:** `2601.20014` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Shuhui Qu
+- **abs:** https://arxiv.org/abs/2601.20014 · **PDF:** https://arxiv.org/pdf/2601.20014v1
+
+## 18772. LinguaMap: Which Layers of LLMs Speak Your Language and How to Tune Them?
+
+- **ID:** `2601.20009` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** J. Ben Tamo, Daniel Carlander-Reuterfelt, Jonathan Rubin, et al.
+- **abs:** https://arxiv.org/abs/2601.20009 · **PDF:** https://arxiv.org/pdf/2601.20009v2
+
+## 18773. VDLM: Variable Diffusion LMs via Robust Latent-to-Text Rendering
+
+- **ID:** `2602.15870` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Shuhui Qu
+- **abs:** https://arxiv.org/abs/2602.15870 · **PDF:** https://arxiv.org/pdf/2602.15870v1
+
+## 18774. On the Effectiveness of LLM-Specific Fine-Tuning for Detecting AI-Generated Text
+
+- **ID:** `2601.20006` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Michał Gromadzki, Anna Wróblewska, Agnieszka Kaliska
+- **abs:** https://arxiv.org/abs/2601.20006 · **PDF:** https://arxiv.org/pdf/2601.20006v1
+
+## 18775. BayPrAnoMeta: Bayesian Proto-MAML for Few-Shot Industrial Image Anomaly Detection
+
+- **ID:** `2601.19992` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Soham Sarkar, Tanmay Sen, Sayantan Banerjee
+- **abs:** https://arxiv.org/abs/2601.19992 · **PDF:** https://arxiv.org/pdf/2601.19992v1
+
+## 18776. Exploring the holographic entropy cone via reinforcement learning
+
+- **ID:** `2601.19979` · **primary:** `hep-th` · **Submitted:** 2026-01-27
+- **Authors:** Temple He, Jaeha Lee, Hirosi Ooguri
+- **abs:** https://arxiv.org/abs/2601.19979 · **PDF:** https://arxiv.org/pdf/2601.19979v1
+
+## 18777. Evaluation of Oncotimia: An LLM based system for supporting tumour boards
+
+- **ID:** `2601.19899` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Luis Lorenzo, Marcos Montana-Mendez, Sergio Figueiras, et al.
+- **abs:** https://arxiv.org/abs/2601.19899 · **PDF:** https://arxiv.org/pdf/2601.19899v1
+
+## 18778. Self-Distillation Enables Continual Learning
+
+- **ID:** `2601.19897` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Idan Shenfeld, Mehul Damani, Jonas Hübotter, et al.
+- **abs:** https://arxiv.org/abs/2601.19897 · **PDF:** https://arxiv.org/pdf/2601.19897v1
+
+## 18779. Post-LayerNorm Is Back: Stable, ExpressivE, and Deep
+
+- **ID:** `2601.19895` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Chen Chen, Lai Wei
+- **abs:** https://arxiv.org/abs/2601.19895 · **PDF:** https://arxiv.org/pdf/2601.19895v2
+
+## 18780. M-SGWR: Multiscale Similarity and Geographically Weighted Regression
+
+- **ID:** `2601.19888` · **primary:** `stat.ME` · **Submitted:** 2026-01-27
+- **Authors:** M. Naser Lessani, Zhenlong Li, Manzhu Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.19888 · **PDF:** https://arxiv.org/pdf/2601.19888v2
+
+## 18781. AI Cap-and-Trade: Efficiency Incentives for Accessibility and Sustainability
+
+- **ID:** `2601.19886` · **primary:** `econ.GN` · **Submitted:** 2026-01-27
+- **Authors:** Marco Bornstein, Amrit Singh Bedi
+- **abs:** https://arxiv.org/abs/2601.19886 · **PDF:** https://arxiv.org/pdf/2601.19886v1
+
+## 18782. SONIC: Spectral Oriented Neural Invariant Convolutions
+
+- **ID:** `2601.19884` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Gijs Joppe Moens, Regina Beets-Tan, Eduardo H. P. Pooch
+- **abs:** https://arxiv.org/abs/2601.19884 · **PDF:** https://arxiv.org/pdf/2601.19884v1
+
+## 18783. DevOps-Gym: Benchmarking AI Agents in Software DevOps Cycle
+
+- **ID:** `2601.20882` · **primary:** `cs.SE` · **Submitted:** 2026-01-27
+- **Authors:** Yuheng Tang, Kaijie Zhu, Bonan Ruan, et al.
+- **abs:** https://arxiv.org/abs/2601.20882 · **PDF:** https://arxiv.org/pdf/2601.20882v1
+
+## 18784. Real-Time Pulsatile Flow Prediction for Realistic, Diverse Intracranial Aneurysm Morphologies using a Graph Transformer and Steady-Flow Data Augmentation
+
+- **ID:** `2601.19876` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Yiying Sheng, Wenhao Ding, Dylan Roi, et al.
+- **abs:** https://arxiv.org/abs/2601.19876 · **PDF:** https://arxiv.org/pdf/2601.19876v2
+
+## 18785. Reflective Translation: Improving Low-Resource Machine Translation via Structured Self-Reflection
+
+- **ID:** `2601.19871` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Nicholas Cheng
+- **abs:** https://arxiv.org/abs/2601.19871 · **PDF:** https://arxiv.org/pdf/2601.19871v1
+
+## 18786. Bandits in Flux: Adversarial Constraints in Dynamic Environments
+
+- **ID:** `2601.19867` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Tareq Si Salem
+- **abs:** https://arxiv.org/abs/2601.19867 · **PDF:** https://arxiv.org/pdf/2601.19867v1
+
+## 18787. Benchmarking LLAMA Model Security Against OWASP Top 10 For LLM Applications
+
+- **ID:** `2601.19970` · **primary:** `cs.CR` · **Submitted:** 2026-01-27
+- **Authors:** Nourin Shahin, Izzat Alsmadi
+- **abs:** https://arxiv.org/abs/2601.19970 · **PDF:** https://arxiv.org/pdf/2601.19970v1
+
+## 18788. Calibration without Ground Truth
+
+- **ID:** `2601.19862` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Yuqing Kong, Mingyu Song, Yizhou Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19862 · **PDF:** https://arxiv.org/pdf/2601.19862v1
+
+## 18789. E2HiL: Entropy-Guided Sample Selection for Efficient Real-World Human-in-the-Loop Reinforcement Learning
+
+- **ID:** `2601.19969` · **primary:** `cs.RO` · **Submitted:** 2026-01-27
+- **Authors:** Haoyuan Deng, Yuanjiang Xue, Haoyang Du, et al.
+- **abs:** https://arxiv.org/abs/2601.19969 · **PDF:** https://arxiv.org/pdf/2601.19969v1
+
+## 18790. Generative Latent Alignment for Interpretable Radar Based Occupancy Detection in Ambient Assisted Living
+
+- **ID:** `2601.19853` · **primary:** `eess.SP` · **Submitted:** 2026-01-27
+- **Authors:** Huy Trinh
+- **abs:** https://arxiv.org/abs/2601.19853 · **PDF:** https://arxiv.org/pdf/2601.19853v1
+
+## 18791. Transforming Computational Lithography with AC and AI -- Faster, More Accurate, and Energy-efficient
+
+- **ID:** `2602.15036` · **primary:** `eess.SP` · **Submitted:** 2026-01-27
+- **Authors:** Saumyadip Mukhopadhyay, Kiho Yang, Kasyap Thottasserymana Vasudevan, et al.
+- **abs:** https://arxiv.org/abs/2602.15036 · **PDF:** https://arxiv.org/pdf/2602.15036v1
+
+## 18792. Towards Fair and Efficient De-identification: Quantifying the Efficiency and Generalizability of De-identification Approaches
+
+- **ID:** `2602.15869` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Noopur Zambare, Kiana Aghakasiri, Carissa Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.15869 · **PDF:** https://arxiv.org/pdf/2602.15869v1
+
+## 18793. Identifying and Transferring Reasoning-Critical Neurons: Improving LLM Inference Reliability via Activation Steering
+
+- **ID:** `2601.19847` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Fangan Dong, Zuming Yan, Xuri Ge, et al.
+- **abs:** https://arxiv.org/abs/2601.19847 · **PDF:** https://arxiv.org/pdf/2601.19847v1
+
+## 18794. HARMONI: Multimodal Personalization of Multi-User Human-Robot Interactions with LLMs
+
+- **ID:** `2601.19839` · **primary:** `cs.RO` · **Submitted:** 2026-01-27
+- **Authors:** Jeanne Malécot, Hamed Rahimi, Jeanne Cattoni, et al.
+- **abs:** https://arxiv.org/abs/2601.19839 · **PDF:** https://arxiv.org/pdf/2601.19839v1
+
+## 18795. Visual Generation Unlocks Human-Like Reasoning through Multimodal World Models
+
+- **ID:** `2601.19834` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Jialong Wu, Xiaoying Zhang, Hongyi Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.19834 · **PDF:** https://arxiv.org/pdf/2601.19834v1
+
+## 18796. A Multi-directional Meta-Learning Framework for Class-Generalizable Anomaly Detection
+
+- **ID:** `2601.19833` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Padmaksha Roy, Lamine Mili, Almuatazbellah Boker
+- **abs:** https://arxiv.org/abs/2601.19833 · **PDF:** https://arxiv.org/pdf/2601.19833v1
+
+## 18797. Neural Neural Scaling Laws
+
+- **ID:** `2601.19831` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Michael Y. Hu, Jane Pan, Ayush Rajesh Jhaveri, et al.
+- **abs:** https://arxiv.org/abs/2601.19831 · **PDF:** https://arxiv.org/pdf/2601.19831v1
+
+## 18798. When Iterative RAG Beats Ideal Evidence: A Diagnostic Study in Scientific Multi-hop Question Answering
+
+- **ID:** `2601.19827` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Mahdi Astaraki, Mohammad Arshi Saloot, Ali Shiraee Kasmaee, et al.
+- **abs:** https://arxiv.org/abs/2601.19827 · **PDF:** https://arxiv.org/pdf/2601.19827v2
+
+## 18799. Disclosure By Design: Identity Transparency as a Behavioural Property of Conversational AI Models
+
+- **ID:** `2603.16874` · **primary:** `cs.HC` · **Submitted:** 2026-01-27
+- **Authors:** Anna Gausen, Sarenne Wallbridge, Hannah Rose Kirk, et al.
+- **abs:** https://arxiv.org/abs/2603.16874 · **PDF:** https://arxiv.org/pdf/2603.16874v1
+
+## 18800. Routing End User Queries to Enterprise Databases
+
+- **ID:** `2601.19825` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Saikrishna Sudarshan, Tanay Kulkarni, Manasi Patwardhan, et al.
+- **abs:** https://arxiv.org/abs/2601.19825 · **PDF:** https://arxiv.org/pdf/2601.19825v1
+
+## 18801. An Interpretable Recommendation Model for Psychometric Data, With an Application to Gerontological Primary Care
+
+- **ID:** `2601.19824` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Andre Paulino de Lima, Paula Castro, Suzana Carvalho Vaz de Andrade, et al.
+- **abs:** https://arxiv.org/abs/2601.19824 · **PDF:** https://arxiv.org/pdf/2601.19824v1
+
+## 18802. Perturbation-Induced Linearization: Constructing Unlearnable Data with Solely Linear Classifiers
+
+- **ID:** `2601.19967` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Jinlin Liu, Wei Chen, Xiaojin Zhang
+- **abs:** https://arxiv.org/abs/2601.19967 · **PDF:** https://arxiv.org/pdf/2601.19967v2
+
+## 18803. Learn and Verify: A Framework for Rigorous Verification of Physics-Informed Neural Networks
+
+- **ID:** `2601.19818` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Kazuaki Tanaka, Kohei Yatabe
+- **abs:** https://arxiv.org/abs/2601.19818 · **PDF:** https://arxiv.org/pdf/2601.19818v1
+
+## 18804. Revisiting Incremental Stochastic Majorization-Minimization Algorithms with Applications to Mixture of Experts
+
+- **ID:** `2601.19811` · **primary:** `stat.ML` · **Submitted:** 2026-01-27
+- **Authors:** TrungKhang Tran, TrungTin Nguyen, Gersende Fort, et al.
+- **abs:** https://arxiv.org/abs/2601.19811 · **PDF:** https://arxiv.org/pdf/2601.19811v1
+
+## 18805. Unsupervised Learning of Efficient Exploration: Pre-training Adaptive Policies via Self-Imposed Goals
+
+- **ID:** `2601.19810` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Octavio Pappalardo
+- **abs:** https://arxiv.org/abs/2601.19810 · **PDF:** https://arxiv.org/pdf/2601.19810v1
+
+## 18806. Zero-Shot Stance Detection in the Wild: Dynamic Target Generation and Multi-Target Adaptation
+
+- **ID:** `2601.19802` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Aohua Li, Yuanshuo Zhang, Ge Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.19802 · **PDF:** https://arxiv.org/pdf/2601.19802v3
+
+## 18807. Component-Aware Pruning Framework for Neural Network Controllers via Gradient-Based Importance Estimation
+
+- **ID:** `2601.19794` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Ganesh Sundaram, Jonas Ulmen, Daniel Görges
+- **abs:** https://arxiv.org/abs/2601.19794 · **PDF:** https://arxiv.org/pdf/2601.19794v1
+
+## 18808. CASTER: Breaking the Cost-Performance Barrier in Multi-Agent Orchestration via Context-Aware Strategy for Task Efficient Routing
+
+- **ID:** `2601.19793` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Shanyv Liu, Xuyang Yuan, Tao Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.19793 · **PDF:** https://arxiv.org/pdf/2601.19793v1
+
+## 18809. LVLMs and Humans Ground Differently in Referential Communication
+
+- **ID:** `2601.19792` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Peter Zeng, Weiling Li, Amie Paige, et al.
+- **abs:** https://arxiv.org/abs/2601.19792 · **PDF:** https://arxiv.org/pdf/2601.19792v2
+
+## 18810. To Grok Grokking: Provable Grokking in Ridge Regression
+
+- **ID:** `2601.19791` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Mingyue Xu, Gal Vardi, Itay Safran
+- **abs:** https://arxiv.org/abs/2601.19791 · **PDF:** https://arxiv.org/pdf/2601.19791v2
+
+## 18811. VDE Bench: Evaluating The Capability of Image Editing Models to Modify Visual Documents
+
+- **ID:** `2602.00122` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Hongzhu Yi, Yujia Yang, Yuanxiang Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00122 · **PDF:** https://arxiv.org/pdf/2602.00122v1
+
+## 18812. Knowledge-Aware Evolution for Streaming Federated Continual Learning with Category Overlap and without Task Identifiers
+
+- **ID:** `2601.19788` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Sixing Tan, Xianmin Liu
+- **abs:** https://arxiv.org/abs/2601.19788 · **PDF:** https://arxiv.org/pdf/2601.19788v1
+
+## 18813. Rethinking Discrete Speech Representation Tokens for Accent Generation
+
+- **ID:** `2601.19786` · **primary:** `eess.AS` · **Submitted:** 2026-01-27
+- **Authors:** Jinzuomu Zhong, Yi Wang, Korin Richmond, et al.
+- **abs:** https://arxiv.org/abs/2601.19786 · **PDF:** https://arxiv.org/pdf/2601.19786v2
+
+## 18814. Reimagining Peer Review Process Through Multi-Agent Mechanism Design
+
+- **ID:** `2601.19778` · **primary:** `cs.MA` · **Submitted:** 2026-01-27
+- **Authors:** Ahmad Farooq, Kamran Iqbal
+- **abs:** https://arxiv.org/abs/2601.19778 · **PDF:** https://arxiv.org/pdf/2601.19778v1
+
+## 18815. RFX-Fuse: Breiman and Cutler's Unified ML Engine + Native Explainable Similarity
+
+- **ID:** `2603.13234` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Chris Kuchar
+- **abs:** https://arxiv.org/abs/2603.13234 · **PDF:** https://arxiv.org/pdf/2603.13234v1
+
+## 18816. Strong Reasoning Isn't Enough: Evaluating Evidence Elicitation in Interactive Diagnosis
+
+- **ID:** `2601.19773` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Zhuohan Long, Zhijie Bao, Zhongyu Wei
+- **abs:** https://arxiv.org/abs/2601.19773 · **PDF:** https://arxiv.org/pdf/2601.19773v1
+
+## 18817. GAVEL: Towards rule-based safety through activation monitoring
+
+- **ID:** `2601.19768` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Shir Rozenfeld, Rahul Pankajakshan, Itay Zloczower, et al.
+- **abs:** https://arxiv.org/abs/2601.19768 · **PDF:** https://arxiv.org/pdf/2601.19768v2
+
+## 18818. The Effect of Architecture During Continual Learning
+
+- **ID:** `2601.19766` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Allyson Hahn, Krishnan Raghavan
+- **abs:** https://arxiv.org/abs/2601.19766 · **PDF:** https://arxiv.org/pdf/2601.19766v1
+
+## 18819. Provable Learning of Random Hierarchy Models and Hierarchical Shallow-to-Deep Chaining
+
+- **ID:** `2601.19756` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Yunwei Ren, Yatin Dandi, Florent Krzakala, et al.
+- **abs:** https://arxiv.org/abs/2601.19756 · **PDF:** https://arxiv.org/pdf/2601.19756v1
+
+## 18820. Global Plane Waves From Local Gaussians: Periodic Charge Densities in a Blink
+
+- **ID:** `2601.19966` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-27
+- **Authors:** Jonas Elsborg, Felix Ærtebjerg, Luca Thiede, et al.
+- **abs:** https://arxiv.org/abs/2601.19966 · **PDF:** https://arxiv.org/pdf/2601.19966v1
+
+## 18821. Regularized $f$-Divergence Kernel Tests
+
+- **ID:** `2601.19755` · **primary:** `stat.ML` · **Submitted:** 2026-01-27
+- **Authors:** Mónica Ribero, Antonin Schrab, Arthur Gretton
+- **abs:** https://arxiv.org/abs/2601.19755 · **PDF:** https://arxiv.org/pdf/2601.19755v1
+
+## 18822. Agentic Design Patterns: A System-Theoretic Framework
+
+- **ID:** `2601.19752` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Minh-Dung Dao, Quy Minh Le, Hoang Thanh Lam, et al.
+- **abs:** https://arxiv.org/abs/2601.19752 · **PDF:** https://arxiv.org/pdf/2601.19752v1
+
+## 18823. Understanding LLM Failures: A Multi-Tape Turing Machine Analysis of Systematic Errors in Language Model Reasoning
+
+- **ID:** `2602.15868` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Magnus Boman
+- **abs:** https://arxiv.org/abs/2602.15868 · **PDF:** https://arxiv.org/pdf/2602.15868v2
+
+## 18824. Veri-Sure: A Contract-Aware Multi-Agent Framework with Temporal Tracing and Formal Verification for Correct RTL Code Generation
+
+- **ID:** `2601.19747` · **primary:** `cs.AR` · **Submitted:** 2026-01-27
+- **Authors:** Jiale Liu, Taiyu Zhou, Tianqi Jiang
+- **abs:** https://arxiv.org/abs/2601.19747 · **PDF:** https://arxiv.org/pdf/2601.19747v1
+
+## 18825. GraphDLG: Exploring Deep Leakage from Gradients in Federated Graph Learning
+
+- **ID:** `2601.19745` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Shuyue Wei, Wantong Chen, Tongyu Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.19745 · **PDF:** https://arxiv.org/pdf/2601.19745v1
+
+## 18826. Interpretable and backpropagation-free Green Learning for efficient multi-task echocardiographic segmentation and classification
+
+- **ID:** `2601.19743` · **primary:** `eess.IV` · **Submitted:** 2026-01-27
+- **Authors:** Jyun-Ping Kao, Jiaxin Yang, C. -C. Jay Kuo, et al.
+- **abs:** https://arxiv.org/abs/2601.19743 · **PDF:** https://arxiv.org/pdf/2601.19743v2
+
+## 18827. TokenSeek: Memory Efficient Fine Tuning via Instance-Aware Token Ditching
+
+- **ID:** `2601.19739` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Runjia Zeng, Qifan Wang, Qiang Guan, et al.
+- **abs:** https://arxiv.org/abs/2601.19739 · **PDF:** https://arxiv.org/pdf/2601.19739v1
+
+## 18828. Quantum Circuit Pre-Synthesis: Learning Local Edits to Reduce $T$-count
+
+- **ID:** `2601.19738` · **primary:** `quant-ph` · **Submitted:** 2026-01-27
+- **Authors:** Daniele Lizzio Bosco, Lukasz Cincio, Giuseppe Serra, et al.
+- **abs:** https://arxiv.org/abs/2601.19738 · **PDF:** https://arxiv.org/pdf/2601.19738v1
+
+## 18829. Stability and Generalization of Nonconvex Optimization with Heavy-Tailed Noise
+
+- **ID:** `2601.19730` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Hongxu Chen, Ke Wei, Xiaoming Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.19730 · **PDF:** https://arxiv.org/pdf/2601.19730v1
+
+## 18830. RvB: Automating AI System Hardening via Iterative Red-Blue Games
+
+- **ID:** `2601.19726` · **primary:** `cs.CR` · **Submitted:** 2026-01-27
+- **Authors:** Lige Huang, Zicheng Liu, Jie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.19726 · **PDF:** https://arxiv.org/pdf/2601.19726v1
+
+## 18831. Component-Level Lesioning of Language Models Reveals Clinically Aligned Aphasia Phenotypes
+
+- **ID:** `2601.19723` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Yifan Wang, Jichen Zheng, Jingyuan Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.19723 · **PDF:** https://arxiv.org/pdf/2601.19723v1
+
+## 18832. Improving Policy Exploitation in Online Reinforcement Learning with Instant Retrospect Action
+
+- **ID:** `2601.19720` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Gong Gao, Weidong Zhao, Xianhui Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.19720 · **PDF:** https://arxiv.org/pdf/2601.19720v2
+
+## 18833. Rethinking Divisive Hierarchical Clustering from a Distributional Perspective
+
+- **ID:** `2601.19718` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Kaifeng Zhang, Kai Ming Ting, Tianrun Liang, et al.
+- **abs:** https://arxiv.org/abs/2601.19718 · **PDF:** https://arxiv.org/pdf/2601.19718v1
+
+## 18834. Hyperbolic Additive Margin Softmax with Hierarchical Information for Speaker Verification
+
+- **ID:** `2601.19709` · **primary:** `cs.SD` · **Submitted:** 2026-01-27
+- **Authors:** Zhihua Fang, Liang He
+- **abs:** https://arxiv.org/abs/2601.19709 · **PDF:** https://arxiv.org/pdf/2601.19709v1
+
+## 18835. Scalable Exploration for High-Dimensional Continuous Control via Value-Guided Flow
+
+- **ID:** `2601.19707` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Yunyue Wei, Chenhui Zuo, Yanan Sui
+- **abs:** https://arxiv.org/abs/2601.19707 · **PDF:** https://arxiv.org/pdf/2601.19707v1
+
+## 18836. SAM Audio Judge: A Unified Multimodal Framework for Perceptual Evaluation of Audio Separation
+
+- **ID:** `2601.19702` · **primary:** `eess.AS` · **Submitted:** 2026-01-27
+- **Authors:** Helin Wang, Bowen Shi, Andros Tjandra, et al.
+- **abs:** https://arxiv.org/abs/2601.19702 · **PDF:** https://arxiv.org/pdf/2601.19702v1
+
+## 18837. Generalizable Multimodal Large Language Model Editing via Invariant Trajectory Learning
+
+- **ID:** `2601.19700` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Jiajie Su, Haoyuan Wang, Xiaohua Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.19700 · **PDF:** https://arxiv.org/pdf/2601.19700v2
+
+## 18838. Predicting Mortgage Default with Machine Learning: AutoML, Class Imbalance, and Leakage Control
+
+- **ID:** `2602.00120` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Xianghong Hu, Tianning Xu, Ying Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.00120 · **PDF:** https://arxiv.org/pdf/2602.00120v1
+
+## 18839. AlignCoder: Aligning Retrieval with Target Intent for Repository-Level Code Completion
+
+- **ID:** `2601.19697` · **primary:** `cs.SE` · **Submitted:** 2026-01-27
+- **Authors:** Tianyue Jiang, Yanli Wang, Yanlin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19697 · **PDF:** https://arxiv.org/pdf/2601.19697v1
+
+## 18840. Self-Supervised Weight Templates for Scalable Vision Model Initialization
+
+- **ID:** `2601.19694` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Yucheng Xie, Fu Feng, Ruixiao Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.19694 · **PDF:** https://arxiv.org/pdf/2601.19694v1
+
+## 18841. Incident-Guided Spatiotemporal Traffic Forecasting
+
+- **ID:** `2602.02528` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Lixiang Fan, Bohao Li, Tao Zou, et al.
+- **abs:** https://arxiv.org/abs/2602.02528 · **PDF:** https://arxiv.org/pdf/2602.02528v1
+
+## 18842. Now You See Me: Designing Responsible AI Dashboards for Early-Stage Health Innovation
+
+- **ID:** `2602.23378` · **primary:** `cs.HC` · **Submitted:** 2026-01-27
+- **Authors:** Svitlana Surodina, Sinem Görücü, Lili Golmohammadi, et al.
+- **abs:** https://arxiv.org/abs/2602.23378 · **PDF:** https://arxiv.org/pdf/2602.23378v1
+
+## 18843. LoPRo: Enhancing Low-Rank Quantization via Permuted Block-Wise Rotation
+
+- **ID:** `2601.19675` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Hongyaoxing Gu, Lijuan Hu, Liye Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.19675 · **PDF:** https://arxiv.org/pdf/2601.19675v1
+
+## 18844. Cross-Domain Offshore Wind Power Forecasting: Transfer Learning Through Meteorological Clusters
+
+- **ID:** `2601.19674` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Dominic Weisser, Chloé Hashimoto-Cullen, Benjamin Guedj
+- **abs:** https://arxiv.org/abs/2601.19674 · **PDF:** https://arxiv.org/pdf/2601.19674v1
+
+## 18845. A Benchmark for Audio Reasoning Capabilities of Multimodal Large Language Models
+
+- **ID:** `2601.19673` · **primary:** `cs.SD` · **Submitted:** 2026-01-27
+- **Authors:** Iwona Christop, Mateusz Czyżnikiewicz, Paweł Skórzewski, et al.
+- **abs:** https://arxiv.org/abs/2601.19673 · **PDF:** https://arxiv.org/pdf/2601.19673v1
+
+## 18846. ProToken: Token-Level Attribution for Federated Large Language Models
+
+- **ID:** `2601.19672` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Waris Gill, Ahmad Humayun, Ali Anwar, et al.
+- **abs:** https://arxiv.org/abs/2601.19672 · **PDF:** https://arxiv.org/pdf/2601.19672v2
+
+## 18847. Grasynda: Graph-based Synthetic Time Series Generation
+
+- **ID:** `2601.19668` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Luis Amorim, Moises Santos, Paulo J. Azevedo, et al.
+- **abs:** https://arxiv.org/abs/2601.19668 · **PDF:** https://arxiv.org/pdf/2601.19668v1
+
+## 18848. SynCABEL: Synthetic Contextualized Augmentation for Biomedical Entity Linking
+
+- **ID:** `2601.19667` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Adam Remaki, Christel Gérardin, Eulàlia Farré-Maduell, et al.
+- **abs:** https://arxiv.org/abs/2601.19667 · **PDF:** https://arxiv.org/pdf/2601.19667v1
+
+## 18849. Direct Doubly Robust Estimation of Conditional Quantile Contrasts
+
+- **ID:** `2601.19666` · **primary:** `stat.ME` · **Submitted:** 2026-01-27
+- **Authors:** Josh Givens, Song Liu, Henry W J Reeve, et al.
+- **abs:** https://arxiv.org/abs/2601.19666 · **PDF:** https://arxiv.org/pdf/2601.19666v1
+
+## 18850. KeepLoRA: Continual Learning with Residual Gradient Adaptation
+
+- **ID:** `2601.19659` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Mao-Lin Luo, Zi-Hao Zhou, Yi-Lin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.19659 · **PDF:** https://arxiv.org/pdf/2601.19659v1
+
+## 18851. One Token Is Enough: Improving Diffusion Language Models with a Sink Token
+
+- **ID:** `2601.19657` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Zihou Zhang, Zheyong Xie, Li Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.19657 · **PDF:** https://arxiv.org/pdf/2601.19657v4
+
+## 18852. Robustness of Constraint Automata for Description Logics with Concrete Domains
+
+- **ID:** `2601.19644` · **primary:** `cs.LO` · **Submitted:** 2026-01-27
+- **Authors:** Stéphane Demri, Tianwen Gu
+- **abs:** https://arxiv.org/abs/2601.19644 · **PDF:** https://arxiv.org/pdf/2601.19644v1
+
+## 18853. RATE: Reviewer Profiling and Annotation-free Training for Expertise Ranking in Peer Review Systems
+
+- **ID:** `2601.19637` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Weicong Liu, Zixuan Yang, Yibo Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.19637 · **PDF:** https://arxiv.org/pdf/2601.19637v1
+
+## 18854. Graph Modelling Analysis of Speech-Gesture Interaction for Aphasia Severity Estimation
+
+- **ID:** `2602.20163` · **primary:** `cs.SD` · **Submitted:** 2026-01-27
+- **Authors:** Navya Martin Kollapally, Christa Akers, Renjith Nelson Joseph
+- **abs:** https://arxiv.org/abs/2602.20163 · **PDF:** https://arxiv.org/pdf/2602.20163v1
+
+## 18855. ReportLogic: Evaluating Logical Quality in Deep Research Reports
+
+- **ID:** `2602.18446` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Jujia Zhao, Zhaoxin Huan, Zihan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.18446 · **PDF:** https://arxiv.org/pdf/2602.18446v1
+
+## 18856. A General-Purpose Diversified 2D Seismic Image Dataset from NAMSS
+
+- **ID:** `2602.04890` · **primary:** `physics.geo-ph` · **Submitted:** 2026-01-27
+- **Authors:** Lucas de Magalhães Araujo, Otávio Oliveira Napoli, Sandra Avila, et al.
+- **abs:** https://arxiv.org/abs/2602.04890 · **PDF:** https://arxiv.org/pdf/2602.04890v1
+
+## 18857. Tracking Drift: Variation-Aware Entropy Scheduling for Non-Stationary Reinforcement Learning
+
+- **ID:** `2601.19624` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Tongxi Wang, Zhuoyang Xia, Xinran Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.19624 · **PDF:** https://arxiv.org/pdf/2601.19624v1
+
+## 18858. Algorithmic Prompt-Augmentation for Efficient LLM-Based Heuristic Design for A* Search
+
+- **ID:** `2601.19622` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Thomas Bömer, Nico Koltermann, Max Disselnmeyer, et al.
+- **abs:** https://arxiv.org/abs/2601.19622 · **PDF:** https://arxiv.org/pdf/2601.19622v1
+
+## 18859. R^3: Replay, Reflection, and Ranking Rewards for LLM Reinforcement Learning
+
+- **ID:** `2601.19620` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Zhizheng Jiang, Kang Zhao, Weikai Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.19620 · **PDF:** https://arxiv.org/pdf/2601.19620v2
+
+## 18860. The role of self-supervised pretraining in differentially private medical image analysis
+
+- **ID:** `2601.19618` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Soroosh Tayebi Arasteh, Mina Farajiamiri, Mahshad Lotfinia, et al.
+- **abs:** https://arxiv.org/abs/2601.19618 · **PDF:** https://arxiv.org/pdf/2601.19618v1
+
+## 18861. Up to 36x Speedup: Mask-based Parallel Inference Paradigm for Key Information Extraction in MLLMs
+
+- **ID:** `2601.19613` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Xinzhong Wang, Ya Guo, Jing Li, et al.
+- **abs:** https://arxiv.org/abs/2601.19613 · **PDF:** https://arxiv.org/pdf/2601.19613v1
+
+## 18862. Safe Exploration via Policy Priors
+
+- **ID:** `2601.19612` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Manuel Wendl, Yarden As, Manish Prajapat, et al.
+- **abs:** https://arxiv.org/abs/2601.19612 · **PDF:** https://arxiv.org/pdf/2601.19612v2
+
+## 18863. Explicit Multi-head Attention for Inter-head Interaction in Large Language Models
+
+- **ID:** `2601.19611` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Runyu Peng, Yunhua Zhou, Demin Song, et al.
+- **abs:** https://arxiv.org/abs/2601.19611 · **PDF:** https://arxiv.org/pdf/2601.19611v1
+
+## 18864. ComAgent: Multi-LLM based Agentic AI Empowered Intelligent Wireless Networks
+
+- **ID:** `2601.19607` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Haoyun Li, Ming Xiao, Kezhi Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19607 · **PDF:** https://arxiv.org/pdf/2601.19607v1
+
+## 18865. GMS-CAVP: Improving Audio-Video Correspondence with Multi-Scale Contrastive and Generative Pretraining
+
+- **ID:** `2601.19606` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Shentong Mo, Zehua Chen, Jun Zhu
+- **abs:** https://arxiv.org/abs/2601.19606 · **PDF:** https://arxiv.org/pdf/2601.19606v1
+
+## 18866. Decompose-and-Formalise: Recursively Verifiable Natural Language Inference
+
+- **ID:** `2601.19605` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Xin Quan, Marco Valentino, Louise A. Dennis, et al.
+- **abs:** https://arxiv.org/abs/2601.19605 · **PDF:** https://arxiv.org/pdf/2601.19605v1
+
+## 18867. The Geometric Mechanics of Contrastive Learning: Alignment Potentials, Entropic Dispersion, and Modality Gap
+
+- **ID:** `2601.19597` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Yichao Cai, Zhen Zhang, Yuhang Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.19597 · **PDF:** https://arxiv.org/pdf/2601.19597v2
+
+## 18868. Intersectional Fairness via Mixed-Integer Optimization
+
+- **ID:** `2601.19595` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Jiří Němeček, Mark Kozdoba, Illia Kryvoviaz, et al.
+- **abs:** https://arxiv.org/abs/2601.19595 · **PDF:** https://arxiv.org/pdf/2601.19595v1
+
+## 18869. Modeling Cascaded Delay Feedback for Online Net Conversion Rate Prediction: Benchmark, Insights and Solutions
+
+- **ID:** `2601.19965` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Mingxuan Luo, Guipeng Xv, Sishuo Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.19965 · **PDF:** https://arxiv.org/pdf/2601.19965v2
+
+## 18870. From Atoms to Chains: Divergence-Guided Reasoning Curriculum for Unlabeled LLM Domain Adaptation
+
+- **ID:** `2601.19588` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Yongqi Wang, Xiaofeng Ji, Jie Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19588 · **PDF:** https://arxiv.org/pdf/2601.19588v1
+
+## 18871. LLM-Enhanced Reinforcement Learning for Long-Term User Satisfaction in Interactive Recommendation
+
+- **ID:** `2601.19585` · **primary:** `cs.IR` · **Submitted:** 2026-01-27
+- **Authors:** Chongjun Xia, Yanchun Peng, Xianzhi Wang
+- **abs:** https://arxiv.org/abs/2601.19585 · **PDF:** https://arxiv.org/pdf/2601.19585v2
+
+## 18872. Yunque DeepResearch Technical Report
+
+- **ID:** `2601.19578` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Yuxuan Cai, Xinyi Lai, Peng Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.19578 · **PDF:** https://arxiv.org/pdf/2601.19578v1
+
+## 18873. Learning Adaptive Parallel Execution for Efficient Code Localization
+
+- **ID:** `2601.19568` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Ke Xu, Siyang Xiao, Ming Liang, et al.
+- **abs:** https://arxiv.org/abs/2601.19568 · **PDF:** https://arxiv.org/pdf/2601.19568v1
+
+## 18874. Learning the Intrinsic Dimensionality of Fermi-Pasta-Ulam-Tsingou Trajectories: A Nonlinear Approach using a Deep Autoencoder Model
+
+- **ID:** `2601.19567` · **primary:** `cond-mat.stat-mech` · **Submitted:** 2026-01-27
+- **Authors:** Gionni Marchetti
+- **abs:** https://arxiv.org/abs/2601.19567 · **PDF:** https://arxiv.org/pdf/2601.19567v2
+
+## 18875. Tournament Informed Adversarial Quality Diversity
+
+- **ID:** `2601.19562` · **primary:** `cs.NE` · **Submitted:** 2026-01-27
+- **Authors:** Timothée Anne, Noah Syrkis, Meriem Elhosni, et al.
+- **abs:** https://arxiv.org/abs/2601.19562 · **PDF:** https://arxiv.org/pdf/2601.19562v1
+
+## 18876. AROMMA: Unifying Olfactory Embeddings for Single Molecules and Mixtures
+
+- **ID:** `2601.19561` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Dayoung Kang, JongWon Kim, Jiho Park, et al.
+- **abs:** https://arxiv.org/abs/2601.19561 · **PDF:** https://arxiv.org/pdf/2601.19561v1
+
+## 18877. Empowering Affected Individuals to Shape AI Fairness Assessments: Processes, Criteria, and Tools
+
+- **ID:** `2602.06984` · **primary:** `cs.CY` · **Submitted:** 2026-01-27
+- **Authors:** Lin Luo, Satwik Ghanta, Yuri Nakao, et al.
+- **abs:** https://arxiv.org/abs/2602.06984 · **PDF:** https://arxiv.org/pdf/2602.06984v1
+
+## 18878. Generalizable Equivariant Diffusion Models for Non-Abelian Lattice Gauge Theory
+
+- **ID:** `2601.19552` · **primary:** `hep-lat` · **Submitted:** 2026-01-27
+- **Authors:** Gert Aarts, Diaa E. Habibi, Andreas Ipp, et al.
+- **abs:** https://arxiv.org/abs/2601.19552 · **PDF:** https://arxiv.org/pdf/2601.19552v1
+
+## 18879. Scale-Consistent State-Space Dynamics via Fractal of Stationary Transformations
+
+- **ID:** `2601.19551` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Geunhyeok Yu, Hyoseok Hwang
+- **abs:** https://arxiv.org/abs/2601.19551 · **PDF:** https://arxiv.org/pdf/2601.19551v1
+
+## 18880. Hybrid Deep Learning Framework for CSI-Based Activity Recognition in Bandwidth-Constrained Wi-Fi Sensing
+
+- **ID:** `2602.06983` · **primary:** `eess.SP` · **Submitted:** 2026-01-27
+- **Authors:** Alison M. Fernandes, Hermes I. Del Monego, Bruno S. Chang, et al.
+- **abs:** https://arxiv.org/abs/2602.06983 · **PDF:** https://arxiv.org/pdf/2602.06983v1
+
+## 18881. Playing With AI: How Do State-Of-The-Art Large Language Models Perform in the 1977 Text-Based Adventure Game Zork?
+
+- **ID:** `2602.15867` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Berry Gerrits
+- **abs:** https://arxiv.org/abs/2602.15867 · **PDF:** https://arxiv.org/pdf/2602.15867v1
+
+## 18882. GenCP: Towards Generative Modeling Paradigm of Coupled Physics
+
+- **ID:** `2601.19541` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Tianrun Gao, Haoren Zheng, Wenhao Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.19541 · **PDF:** https://arxiv.org/pdf/2601.19541v1
+
+## 18883. SLM-SS: Speech Language Model for Generative Speech Separation
+
+- **ID:** `2601.19533` · **primary:** `cs.SD` · **Submitted:** 2026-01-27
+- **Authors:** Tianhua Li, Chenda Li, Wei Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19533 · **PDF:** https://arxiv.org/pdf/2601.19533v1
+
+## 18884. Benchmarks Saturate When The Model Gets Smarter Than The Judge
+
+- **ID:** `2601.19532` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Marthe Ballon, Andres Algaba, Brecht Verbeken, et al.
+- **abs:** https://arxiv.org/abs/2601.19532 · **PDF:** https://arxiv.org/pdf/2601.19532v1
+
+## 18885. Fuzzy expert system for the process of collecting and purifying acidic water: a digital twin approach
+
+- **ID:** `2601.19527` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Temirbolat Maratuly, Pakizar Shamoi, Timur Samigulin
+- **abs:** https://arxiv.org/abs/2601.19527 · **PDF:** https://arxiv.org/pdf/2601.19527v1
+
+## 18886. Mocap Anywhere: Towards Pairwise-Distance based Motion Capture in the Wild (for the Wild)
+
+- **ID:** `2601.19519` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Ofir Abramovich, Ariel Shamir, Andreas Aristidou
+- **abs:** https://arxiv.org/abs/2601.19519 · **PDF:** https://arxiv.org/pdf/2601.19519v1
+
+## 18887. Enhancing Academic Paper Recommendations Using Fine-Grained Knowledge Entities and Multifaceted Document Embeddings
+
+- **ID:** `2601.19513` · **primary:** `cs.IR` · **Submitted:** 2026-01-27
+- **Authors:** Haixu Xi, Heng Zhang, Chengzhi Zhang
+- **abs:** https://arxiv.org/abs/2601.19513 · **PDF:** https://arxiv.org/pdf/2601.19513v1
+
+## 18888. ALRM: Agentic LLM for Robotic Manipulation
+
+- **ID:** `2601.19510` · **primary:** `cs.RO` · **Submitted:** 2026-01-27
+- **Authors:** Vitor Gaboardi dos Santos, Ibrahim Khadraoui, Ibrahim Farhat, et al.
+- **abs:** https://arxiv.org/abs/2601.19510 · **PDF:** https://arxiv.org/pdf/2601.19510v2
+
+## 18889. Rethinking Intelligence: Brain-like Neuron Network
+
+- **ID:** `2601.19508` · **primary:** `cs.NE` · **Submitted:** 2026-01-27
+- **Authors:** Weifeng Liu
+- **abs:** https://arxiv.org/abs/2601.19508 · **PDF:** https://arxiv.org/pdf/2601.19508v1
+
+## 18890. Automated Safety Benchmarking: A Multi-agent Pipeline for LVLMs
+
+- **ID:** `2601.19507` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Xiangyang Zhu, Yuan Tian, Zicheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.19507 · **PDF:** https://arxiv.org/pdf/2601.19507v1
+
+## 18891. GradPruner: Gradient-Guided Layer Pruning Enabling Efficient Fine-Tuning and Inference for LLMs
+
+- **ID:** `2601.19503` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Wei Huang, Anda Cheng, Yinggui Wang
+- **abs:** https://arxiv.org/abs/2601.19503 · **PDF:** https://arxiv.org/pdf/2601.19503v1
+
+## 18892. Cortex-Grounded Diffusion Models for Brain Image Generation
+
+- **ID:** `2601.19498` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Fabian Bongratz, Yitong Li, Sama Elbaroudy, et al.
+- **abs:** https://arxiv.org/abs/2601.19498 · **PDF:** https://arxiv.org/pdf/2601.19498v1
+
+## 18893. AACR-Bench: Evaluating Automatic Code Review with Holistic Repository-Level Context
+
+- **ID:** `2601.19494` · **primary:** `cs.SE` · **Submitted:** 2026-01-27
+- **Authors:** Lei Zhang, Yongda Yu, Minghui Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.19494 · **PDF:** https://arxiv.org/pdf/2601.19494v3
+
+## 18894. ClaimPT: A Portuguese Dataset of Annotated Claims in News Articles
+
+- **ID:** `2601.19490` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Ricardo Campos, Raquel Sequeira, Sara Nerea, et al.
+- **abs:** https://arxiv.org/abs/2601.19490 · **PDF:** https://arxiv.org/pdf/2601.19490v3
+
+## 18895. LLM-VA: Resolving the Jailbreak-Overrefusal Trade-off via Vector Alignment
+
+- **ID:** `2601.19487` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Haonan Zhang, Dongxia Wang, Yi Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.19487 · **PDF:** https://arxiv.org/pdf/2601.19487v1
+
+## 18896. Posterior Distribution-assisted Evolutionary Dynamic Optimization as an Online Calibrator for Complex Social Simulations
+
+- **ID:** `2601.19481` · **primary:** `cs.NE` · **Submitted:** 2026-01-27
+- **Authors:** Peng Yang, Zhenhua Yang, Boquan Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.19481 · **PDF:** https://arxiv.org/pdf/2601.19481v1
+
+## 18897. Time-to-Injury Forecasting in Elite Female Football: A DeepHit Survival Approach
+
+- **ID:** `2601.19479` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Victoria Catterall, Cise Midoglu, Stephen Lynch
+- **abs:** https://arxiv.org/abs/2601.19479 · **PDF:** https://arxiv.org/pdf/2601.19479v1
+
+## 18898. ROIDS: Robust Outlier-Aware Informed Down-Sampling
+
+- **ID:** `2601.19477` · **primary:** `cs.NE` · **Submitted:** 2026-01-27
+- **Authors:** Alina Geiger, Martin Briesch, Dominik Sobania, et al.
+- **abs:** https://arxiv.org/abs/2601.19477 · **PDF:** https://arxiv.org/pdf/2601.19477v1
+
+## 18899. Cross-Session Decoding of Neural Spiking Data via Task-Conditioned Latent Alignment
+
+- **ID:** `2601.19963` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Canyang Zhao, Bolin Peng, J. Patrick Mayo, et al.
+- **abs:** https://arxiv.org/abs/2601.19963 · **PDF:** https://arxiv.org/pdf/2601.19963v1
+
+## 18900. A Practical Guide to Agentic AI Transition in Organizations
+
+- **ID:** `2602.10122` · **primary:** `cs.CY` · **Submitted:** 2026-01-27
+- **Authors:** Eranga Bandara, Ross Gore, Sachin Shetty, et al.
+- **abs:** https://arxiv.org/abs/2602.10122 · **PDF:** https://arxiv.org/pdf/2602.10122v1
+
+## 18901. IC-EO: Interpretable Code-based assistant for Earth Observation
+
+- **ID:** `2602.00117` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Lamia Lahouel, Laurynas Lopata, Simon Gruening, et al.
+- **abs:** https://arxiv.org/abs/2602.00117 · **PDF:** https://arxiv.org/pdf/2602.00117v1
+
+## 18902. On the Expressiveness of State Space Models via Temporal Logics
+
+- **ID:** `2601.19467` · **primary:** `cs.LO` · **Submitted:** 2026-01-27
+- **Authors:** Eric Alsmann, Lowejatan Noori, Martin Lange
+- **abs:** https://arxiv.org/abs/2601.19467 · **PDF:** https://arxiv.org/pdf/2601.19467v1
+
+## 18903. APC-RL: Exceeding Data-Driven Behavior Priors with Adaptive Policy Composition
+
+- **ID:** `2601.19452` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Finn Rietz, Pedro Zuidberg dos Martires, Johannes Andreas Stork
+- **abs:** https://arxiv.org/abs/2601.19452 · **PDF:** https://arxiv.org/pdf/2601.19452v1
+
+## 18904. Dynamic Multi-Expert Projectors with Stabilized Routing for Multilingual Speech Recognition
+
+- **ID:** `2601.19451` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Isha Pandey, Ashish Mittal, Vartul Bahuguna, et al.
+- **abs:** https://arxiv.org/abs/2601.19451 · **PDF:** https://arxiv.org/pdf/2601.19451v1
+
+## 18905. Fixed Aggregation Features Can Rival GNNs
+
+- **ID:** `2601.19449` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Celia Rubio-Madrigal, Rebekka Burkholz
+- **abs:** https://arxiv.org/abs/2601.19449 · **PDF:** https://arxiv.org/pdf/2601.19449v1
+
+## 18906. From Internal Diagnosis to External Auditing: A VLM-Driven Paradigm for Online Test-Time Backdoor Defense
+
+- **ID:** `2601.19448` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Binyan Xu, Fan Yang, Xilin Dai, et al.
+- **abs:** https://arxiv.org/abs/2601.19448 · **PDF:** https://arxiv.org/pdf/2601.19448v1
+
+## 18907. THDC: Training Hyperdimensional Computing Models with Backpropagation
+
+- **ID:** `2602.00116` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Hanne Dejonghe, Sam Leroux
+- **abs:** https://arxiv.org/abs/2602.00116 · **PDF:** https://arxiv.org/pdf/2602.00116v1
+
+## 18908. KG-CRAFT: Knowledge Graph-based Contrastive Reasoning with LLMs for Enhancing Automated Fact-checking
+
+- **ID:** `2601.19447` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Vítor N. Lourenço, Aline Paes, Tillman Weyde, et al.
+- **abs:** https://arxiv.org/abs/2601.19447 · **PDF:** https://arxiv.org/pdf/2601.19447v1
+
+## 18909. OSIRIS: Bridging Analog Circuit Design and Machine Learning with Scalable Dataset Generation
+
+- **ID:** `2601.19439` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Giuseppe Chiari, Michele Piccoli, Davide Zoni
+- **abs:** https://arxiv.org/abs/2601.19439 · **PDF:** https://arxiv.org/pdf/2601.19439v1
+
+## 18910. Ad Insertion in LLM-Generated Responses
+
+- **ID:** `2601.19435` · **primary:** `cs.GT` · **Submitted:** 2026-01-27
+- **Authors:** Shengwei Xu, Zhaohua Chen, Xiaotie Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.19435 · **PDF:** https://arxiv.org/pdf/2601.19435v1
+
+## 18911. Event Driven Clustering Algorithm
+
+- **ID:** `2602.00115` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** David El-Chai Ben-Ezra, Adar Tal, Daniel Brisk
+- **abs:** https://arxiv.org/abs/2602.00115 · **PDF:** https://arxiv.org/pdf/2602.00115v1
+
+## 18912. Deep Reinforcement Learning for Interference Suppression in RIS-Aided Space-Air-Ground Integrated Networks
+
+- **ID:** `2602.06982` · **primary:** `eess.SP` · **Submitted:** 2026-01-27
+- **Authors:** Pujitha Mamillapalli, Shikhar Verma, Tiago Koketsu Rodrigues, et al.
+- **abs:** https://arxiv.org/abs/2602.06982 · **PDF:** https://arxiv.org/pdf/2602.06982v1
+
+## 18913. Deep Unfolded Fractional Optimization for Maximizing Robust Throughput in 6G Networks
+
+- **ID:** `2602.06062` · **primary:** `cs.IT` · **Submitted:** 2026-01-27
+- **Authors:** Anh Thi Bui, Robert-Jeron Reifert, Hayssam Dahrouj, et al.
+- **abs:** https://arxiv.org/abs/2602.06062 · **PDF:** https://arxiv.org/pdf/2602.06062v1
+
+## 18914. An Agentic AI Control Plane for 6G Network Slice Orchestration, Monitoring, and Trading
+
+- **ID:** `2602.13227` · **primary:** `cs.NI` · **Submitted:** 2026-01-27
+- **Authors:** Eranga Bandara, Ross Gore, Sachin Shetty, et al.
+- **abs:** https://arxiv.org/abs/2602.13227 · **PDF:** https://arxiv.org/pdf/2602.13227v1
+
+## 18915. Task-Centric Policy Optimization from Misaligned Motion Priors
+
+- **ID:** `2601.19411` · **primary:** `cs.RO` · **Submitted:** 2026-01-27
+- **Authors:** Ziang Zheng, Kai Feng, Yi Nie, et al.
+- **abs:** https://arxiv.org/abs/2601.19411 · **PDF:** https://arxiv.org/pdf/2601.19411v2
+
+## 18916. Do LLMs Truly Benefit from Longer Context in Automatic Post-Editing?
+
+- **ID:** `2601.19410` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Ahrii Kim, Seong-heum Kim
+- **abs:** https://arxiv.org/abs/2601.19410 · **PDF:** https://arxiv.org/pdf/2601.19410v3
+
+## 18917. Sim-and-Human Co-training for Data-Efficient and Generalizable Robotic Manipulation
+
+- **ID:** `2601.19406` · **primary:** `cs.RO` · **Submitted:** 2026-01-27
+- **Authors:** Kaipeng Fang, Weiqing Liang, Yuyang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.19406 · **PDF:** https://arxiv.org/pdf/2601.19406v1
+
+## 18918. RPO:Reinforcement Fine-Tuning with Partial Reasoning Optimization
+
+- **ID:** `2601.19404` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Hongzhu Yi, Xinming Wang, Zhenghao zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.19404 · **PDF:** https://arxiv.org/pdf/2601.19404v2
+
+## 18919. Learned split-spectrum metalens for obstruction-free broadband imaging in the visible
+
+- **ID:** `2601.19403` · **primary:** `physics.optics` · **Submitted:** 2026-01-27
+- **Authors:** Seungwoo Yoon, Dohyun Kang, Eunsue Choi, et al.
+- **abs:** https://arxiv.org/abs/2601.19403 · **PDF:** https://arxiv.org/pdf/2601.19403v1
+
+## 18920. PROTEUS: SLA-Aware Routing via Lagrangian RL for Multi-LLM Serving Systems
+
+- **ID:** `2601.19402` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Amit Singh Bhatti, Vishal Vaddina, Dagnachew Birru
+- **abs:** https://arxiv.org/abs/2601.19402 · **PDF:** https://arxiv.org/pdf/2601.19402v3
+
+## 18921. Improved Convergence Rates of Muon Optimizer for Nonconvex Optimization
+
+- **ID:** `2601.19400` · **primary:** `math.OC` · **Submitted:** 2026-01-27
+- **Authors:** Shuntaro Nagashima, Hideaki Iiduka
+- **abs:** https://arxiv.org/abs/2601.19400 · **PDF:** https://arxiv.org/pdf/2601.19400v3
+
+## 18922. Residual Tokens Enhance Masked Autoencoders for Speech Modeling
+
+- **ID:** `2601.19399` · **primary:** `cs.SD` · **Submitted:** 2026-01-27
+- **Authors:** Samir Sadok, Stéphane Lathuilière, Xavier Alameda-Pineda
+- **abs:** https://arxiv.org/abs/2601.19399 · **PDF:** https://arxiv.org/pdf/2601.19399v1
+
+## 18923. SEAFormer: A Spatial Proximity and Edge-Aware Transformer for Real-World Vehicle Routing Problems
+
+- **ID:** `2601.19395` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Saeed Nasehi Basharzad, Farhana Choudhury, Egemen Tanin
+- **abs:** https://arxiv.org/abs/2601.19395 · **PDF:** https://arxiv.org/pdf/2601.19395v2
+
+## 18924. DSP-Reg: Domain-Sensitive Parameter Regularization for Robust Domain Generalization
+
+- **ID:** `2601.19394` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Xudong Han, Senkang Hu, Yihang Tao, et al.
+- **abs:** https://arxiv.org/abs/2601.19394 · **PDF:** https://arxiv.org/pdf/2601.19394v1
+
+## 18925. Epistemic Traps: Rational Misalignment Driven by Model Misspecification
+
+- **ID:** `2602.17676` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Xingcheng Xu, Jingjing Qu, Qiaosheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.17676 · **PDF:** https://arxiv.org/pdf/2602.17676v1
+
+## 18926. High-quality data augmentation for code comment classification
+
+- **ID:** `2601.19383` · **primary:** `cs.SE` · **Submitted:** 2026-01-27
+- **Authors:** Thomas Borsani, Andrea Rosani, Giuseppe Di Fatta
+- **abs:** https://arxiv.org/abs/2601.19383 · **PDF:** https://arxiv.org/pdf/2601.19383v1
+
+## 18927. Tri-Reader: An Open-Access, Multi-Stage AI Pipeline for First-Pass Lung Nodule Annotation in Screening CT
+
+- **ID:** `2601.19380` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Fakrul Islam Tushar, Joseph Y. Lo
+- **abs:** https://arxiv.org/abs/2601.19380 · **PDF:** https://arxiv.org/pdf/2601.19380v2
+
+## 18928. Optimal Asynchronous Stochastic Nonconvex Optimization under Heavy-Tailed Noise
+
+- **ID:** `2601.19379` · **primary:** `math.OC` · **Submitted:** 2026-01-27
+- **Authors:** Yidong Wu, Luo Luo
+- **abs:** https://arxiv.org/abs/2601.19379 · **PDF:** https://arxiv.org/pdf/2601.19379v1
+
+## 18929. Teaching Machine Learning Fundamentals with LEGO Robotics
+
+- **ID:** `2601.19376` · **primary:** `cs.RO` · **Submitted:** 2026-01-27
+- **Authors:** Viacheslav Sydora, Guner Dilsad Er, Michael Muehlebach
+- **abs:** https://arxiv.org/abs/2601.19376 · **PDF:** https://arxiv.org/pdf/2601.19376v1
+
+## 18930. Selective Steering: Norm-Preserving Control Through Discriminative Layer Selection
+
+- **ID:** `2601.19375` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Quy-Anh Dang, Chris Ngo
+- **abs:** https://arxiv.org/abs/2601.19375 · **PDF:** https://arxiv.org/pdf/2601.19375v1
+
+## 18931. Divergence-Free Diffusion Models for Incompressible Fluid Flows
+
+- **ID:** `2601.19368` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-01-27
+- **Authors:** Wilfried Genuist, Éric Savin, Filippo Gatti, et al.
+- **abs:** https://arxiv.org/abs/2601.19368 · **PDF:** https://arxiv.org/pdf/2601.19368v1
+
+## 18932. CHEHAB RL: Learning to Optimize Fully Homomorphic Encryption Computations
+
+- **ID:** `2601.19367` · **primary:** `cs.CR` · **Submitted:** 2026-01-27
+- **Authors:** Bilel Sefsaf, Abderraouf Dandani, Abdessamed Seddiki, et al.
+- **abs:** https://arxiv.org/abs/2601.19367 · **PDF:** https://arxiv.org/pdf/2601.19367v1
+
+## 18933. Revisiting Parameter Server in LLM Post-Training
+
+- **ID:** `2601.19362` · **primary:** `cs.DC` · **Submitted:** 2026-01-27
+- **Authors:** Xinyi Wan, Penghui Qi, Guangxing Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.19362 · **PDF:** https://arxiv.org/pdf/2601.19362v1
+
+## 18934. Binary Token-Level Classification with DeBERTa for All-Type MWE Identification: A Lightweight Approach with Linguistic Enhancement
+
+- **ID:** `2601.19360` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Diego Rossini, Lonneke van der Plas
+- **abs:** https://arxiv.org/abs/2601.19360 · **PDF:** https://arxiv.org/pdf/2601.19360v1
+
+## 18935. MeanCache: From Instantaneous to Average Velocity for Accelerating Flow Matching Inference
+
+- **ID:** `2601.19961` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Huanlin Gao, Ping Chen, Fuyuan Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.19961 · **PDF:** https://arxiv.org/pdf/2601.19961v3
+
+## 18936. GraphSB: Boosting Imbalanced Node Classification on Graphs through Structural Balance
+
+- **ID:** `2601.19352` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Zhixiao Wang, Chaofan Zhu, Qihan Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.19352 · **PDF:** https://arxiv.org/pdf/2601.19352v1
+
+## 18937. Cross-Examination Framework: A Task-Agnostic Diagnostic for Information Fidelity in Text-to-Text Generation
+
+- **ID:** `2601.19350` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Tathagata Raha, Clement Christophe, Nada Saadi, et al.
+- **abs:** https://arxiv.org/abs/2601.19350 · **PDF:** https://arxiv.org/pdf/2601.19350v1
+
+## 18938. Robust Uncertainty Estimation under Distribution Shift via Difference Reconstruction
+
+- **ID:** `2601.19341` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Xinran Xu, Li Rong Wang, Xiuyi Fan
+- **abs:** https://arxiv.org/abs/2601.19341 · **PDF:** https://arxiv.org/pdf/2601.19341v1
+
+## 18939. SETA: Statistical Fault Attribution for Compound AI Systems
+
+- **ID:** `2601.19337` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Sayak Chowdhury, Meenakshi D'Souza
+- **abs:** https://arxiv.org/abs/2601.19337 · **PDF:** https://arxiv.org/pdf/2601.19337v1
+
+## 18940. From Observations to Events: Event-Aware World Model for Reinforcement Learning
+
+- **ID:** `2601.19336` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Zhao-Han Peng, Shaohui Li, Zhi Li, et al.
+- **abs:** https://arxiv.org/abs/2601.19336 · **PDF:** https://arxiv.org/pdf/2601.19336v1
+
+## 18941. When Benchmarks Leak: Inference-Time Decontamination for LLMs
+
+- **ID:** `2601.19334` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Jianzhe Chai, Yu Zhe, Jun Sakuma
+- **abs:** https://arxiv.org/abs/2601.19334 · **PDF:** https://arxiv.org/pdf/2601.19334v1
+
+## 18942. Metric $k$-clustering using only Weak Comparison Oracles
+
+- **ID:** `2601.19333` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Rahul Raychaudhury, Aryan Esmailpour, Sainyam Galhotra, et al.
+- **abs:** https://arxiv.org/abs/2601.19333 · **PDF:** https://arxiv.org/pdf/2601.19333v1
+
+## 18943. Innovator-VL: A Multimodal Large Language Model for Scientific Discovery
+
+- **ID:** `2601.19325` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Zichen Wen, Boxue Yang, Shuang Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.19325 · **PDF:** https://arxiv.org/pdf/2601.19325v1
+
+## 18944. Do we really need Self-Attention for Streaming Automatic Speech Recognition?
+
+- **ID:** `2601.19960` · **primary:** `eess.AS` · **Submitted:** 2026-01-27
+- **Authors:** Youness Dkhissi, Valentin Vielzeuf, Elys Allesiardo, et al.
+- **abs:** https://arxiv.org/abs/2601.19960 · **PDF:** https://arxiv.org/pdf/2601.19960v1
+
+## 18945. Predictive Accuracy versus Interpretability in Energy Markets: A Copula-Enhanced TVP-SVAR Analysis
+
+- **ID:** `2601.19321` · **primary:** `q-fin.CP` · **Submitted:** 2026-01-27
+- **Authors:** Fredy Pokou, Jules Sadefo Kamdem, Kpante Emmanuel Gnandi
+- **abs:** https://arxiv.org/abs/2601.19321 · **PDF:** https://arxiv.org/pdf/2601.19321v1
+
+## 18946. 1S-DAug: One-Shot Data Augmentation for Robust Few-Shot Generalization
+
+- **ID:** `2602.00114` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Yunwei Bai, Ying Kiat Tan, Yao Shu, et al.
+- **abs:** https://arxiv.org/abs/2602.00114 · **PDF:** https://arxiv.org/pdf/2602.00114v3
+
+## 18947. StableQAT: Stable Quantization-Aware Training at Ultra-Low Bitwidths
+
+- **ID:** `2601.19320` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Tianyi Chen, Sihan Chen, Xiaoyi Qu, et al.
+- **abs:** https://arxiv.org/abs/2601.19320 · **PDF:** https://arxiv.org/pdf/2601.19320v2
+
+## 18948. Generalizable IoT Traffic Representations for Cross-Network Device Identification
+
+- **ID:** `2601.19315` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Arunan Sivanathan, David Warren, Deepak Mishra, et al.
+- **abs:** https://arxiv.org/abs/2601.19315 · **PDF:** https://arxiv.org/pdf/2601.19315v1
+
+## 18949. Instance-Guided Radar Depth Estimation for 3D Object Detection
+
+- **ID:** `2601.19314` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Chen-Chou Lo, Patrick Vandewalle
+- **abs:** https://arxiv.org/abs/2601.19314 · **PDF:** https://arxiv.org/pdf/2601.19314v1
+
+## 18950. LightSBB-M: Bridging Schrödinger and Bass for Generative Diffusion Modeling
+
+- **ID:** `2601.19312` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Alexandre Alouadi, Pierre Henry-Labordère, Grégoire Loeper, et al.
+- **abs:** https://arxiv.org/abs/2601.19312 · **PDF:** https://arxiv.org/pdf/2601.19312v1
+
+## 18951. Balancing Sustainability And Performance: The Role Of Small-Scale LLMs In Agentic Artificial Intelligence Systems
+
+- **ID:** `2601.19311` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Anh Khoa Ngo Ho, Martin Chauvin, Simon Gosset, et al.
+- **abs:** https://arxiv.org/abs/2601.19311 · **PDF:** https://arxiv.org/pdf/2601.19311v2
+
+## 18952. Curiosity Driven Knowledge Retrieval for Mobile Agents
+
+- **ID:** `2601.19306` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Sijia Li, Xiaoyu Tan, Shahir Ali, et al.
+- **abs:** https://arxiv.org/abs/2601.19306 · **PDF:** https://arxiv.org/pdf/2601.19306v1
+
+## 18953. Formula-One Prompting: Equation-First Reasoning For Applied Mathematics
+
+- **ID:** `2601.19302` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Natapong Nitarach, Pittawat Taveekitworachai, Kunat Pipatanakul
+- **abs:** https://arxiv.org/abs/2601.19302 · **PDF:** https://arxiv.org/pdf/2601.19302v2
+
+## 18954. Queue Length Regret Bounds for Contextual Queueing Bandits
+
+- **ID:** `2601.19300` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Seoungbin Bae, Garyeong Kang, Dabeen Lee
+- **abs:** https://arxiv.org/abs/2601.19300 · **PDF:** https://arxiv.org/pdf/2601.19300v1
+
+## 18955. Process-Aware Procurement Lead Time Prediction for Shipyard Delay Mitigation
+
+- **ID:** `2601.19296` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Yongjae Lee, Eunhee Park, Daesan Park, et al.
+- **abs:** https://arxiv.org/abs/2601.19296 · **PDF:** https://arxiv.org/pdf/2601.19296v1
+
+## 18956. Deep Neural Networks as Iterated Function Systems and a Generalization Bound
+
+- **ID:** `2601.19958` · **primary:** `stat.ML` · **Submitted:** 2026-01-27
+- **Authors:** Jonathan Vacher
+- **abs:** https://arxiv.org/abs/2601.19958 · **PDF:** https://arxiv.org/pdf/2601.19958v1
+
+## 18957. SRAG: RAG with Structured Data Improves Vector Retrieval
+
+- **ID:** `2603.26670` · **primary:** `cs.IR` · **Submitted:** 2026-01-27
+- **Authors:** Shalin Shah, Srikanth Ryali, Ramasubbu Venkatesh
+- **abs:** https://arxiv.org/abs/2603.26670 · **PDF:** https://arxiv.org/pdf/2603.26670v1
+
+## 18958. MetaGen: Self-Evolving Roles and Topologies for Multi-Agent LLM Reasoning
+
+- **ID:** `2601.19290` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Yimeng Wang, Jiaxing Zhao, Hongbin Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.19290 · **PDF:** https://arxiv.org/pdf/2601.19290v1
+
+## 18959. ReToP: Learning to Rewrite Electronic Health Records for Clinical Prediction
+
+- **ID:** `2601.19286` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Jesus Lovon-Melgarejo, Jose G. Moreno, Christine Damase-Michel, et al.
+- **abs:** https://arxiv.org/abs/2601.19286 · **PDF:** https://arxiv.org/pdf/2601.19286v1
+
+## 18960. Smoothing the Score Function for Generalization in Diffusion Models: An Optimization-based Explanation Framework
+
+- **ID:** `2601.19285` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Xinyu Zhou, Jiawei Zhang, Stephen J. Wright
+- **abs:** https://arxiv.org/abs/2601.19285 · **PDF:** https://arxiv.org/pdf/2601.19285v2
+
+## 18961. Model-Free Output Feedback Stabilization via Policy Gradient Methods
+
+- **ID:** `2601.19284` · **primary:** `eess.SY` · **Submitted:** 2026-01-27
+- **Authors:** Ankang Zhang, Ming Chi, Xiaoling Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19284 · **PDF:** https://arxiv.org/pdf/2601.19284v2
+
+## 18962. Group Distributionally Robust Optimization-Driven Reinforcement Learning for LLM Reasoning
+
+- **ID:** `2601.19280` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Kishan Panaganti, Zhenwen Liang, Wenhao Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.19280 · **PDF:** https://arxiv.org/pdf/2601.19280v1
+
+## 18963. DART: Diffusion-Inspired Speculative Decoding for Fast LLM Inference
+
+- **ID:** `2601.19278` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Fuliang Liu, Xue Li, Ketai Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.19278 · **PDF:** https://arxiv.org/pdf/2601.19278v1
+
+## 18964. Talos: Optimizing Top-$K$ Accuracy in Recommender Systems
+
+- **ID:** `2601.19276` · **primary:** `cs.IR` · **Submitted:** 2026-01-27
+- **Authors:** Shengjia Zhang, Weiqin Yang, Jiawei Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.19276 · **PDF:** https://arxiv.org/pdf/2601.19276v1
+
+## 18965. Tactile Memory with Soft Robot: Robust Object Insertion via Masked Encoding and Soft Wrist
+
+- **ID:** `2601.19275` · **primary:** `cs.RO` · **Submitted:** 2026-01-27
+- **Authors:** Tatsuya Kamijo, Mai Nishimura, Cristian C. Beltran-Hernandez, et al.
+- **abs:** https://arxiv.org/abs/2601.19275 · **PDF:** https://arxiv.org/pdf/2601.19275v1
+
+## 18966. Riddle Quest : The Enigma of Words
+
+- **ID:** `2601.19273` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Niharika Sri Parasa, Chaitali Diwan, Srinath Srinivasa
+- **abs:** https://arxiv.org/abs/2601.19273 · **PDF:** https://arxiv.org/pdf/2601.19273v1
+
+## 18967. DiaDem: Advancing Dialogue Descriptions in Audiovisual Video Captioning for Multimodal Large Language Models
+
+- **ID:** `2601.19267` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Xinlong Chen, Weihong Lin, Jingyun Hua, et al.
+- **abs:** https://arxiv.org/abs/2601.19267 · **PDF:** https://arxiv.org/pdf/2601.19267v1
+
+## 18968. Whitespaces Don't Lie: Feature-Driven and Embedding-Based Approaches for Detecting Machine-Generated Code
+
+- **ID:** `2601.19264` · **primary:** `cs.SE` · **Submitted:** 2026-01-27
+- **Authors:** Syed Mehedi Hasan Nirob, Shamim Ehsan, Moqsadur Rahman, et al.
+- **abs:** https://arxiv.org/abs/2601.19264 · **PDF:** https://arxiv.org/pdf/2601.19264v1
+
+## 18969. Decoupled Split Learning via Auxiliary Loss
+
+- **ID:** `2601.19261` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Anower Zihad, Felix Owino, Ming Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.19261 · **PDF:** https://arxiv.org/pdf/2601.19261v2
+
+## 18970. GhostUI: Unveiling Hidden Interactions in Mobile UI
+
+- **ID:** `2601.19258` · **primary:** `cs.HC` · **Submitted:** 2026-01-27
+- **Authors:** Minkyu Kweon, Seokhyeon Park, Soohyun Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.19258 · **PDF:** https://arxiv.org/pdf/2601.19258v1
+
+## 18971. PCEvo: Path-Consistent Molecular Representation via Virtual Evolutionary
+
+- **ID:** `2601.19257` · **primary:** `q-bio.BM` · **Submitted:** 2026-01-27
+- **Authors:** Kun Li, Longtao Hu, Yida Xiong, et al.
+- **abs:** https://arxiv.org/abs/2601.19257 · **PDF:** https://arxiv.org/pdf/2601.19257v1
+
+## 18972. E-QRGMM: Efficient Generative Metamodeling for Covariate-Dependent Uncertainty Quantification
+
+- **ID:** `2601.19256` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Zhiyang Liang, Qingkai Zhang
+- **abs:** https://arxiv.org/abs/2601.19256 · **PDF:** https://arxiv.org/pdf/2601.19256v1
+
+## 18973. LLM-Assisted Logic Rule Learning: Scaling Human Expertise for Time Series Anomaly Detection
+
+- **ID:** `2601.19255` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Haoting Zhang, Shekhar Jain
+- **abs:** https://arxiv.org/abs/2601.19255 · **PDF:** https://arxiv.org/pdf/2601.19255v1
+
+## 18974. GLOVE: Global Verifier for LLM Memory-Environment Realignment
+
+- **ID:** `2601.19249` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Xingkun Yin, Hongyang Du
+- **abs:** https://arxiv.org/abs/2601.19249 · **PDF:** https://arxiv.org/pdf/2601.19249v1
+
+## 18975. Beyond In-Domain Detection: SpikeScore for Cross-Domain Hallucination Detection
+
+- **ID:** `2601.19245` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Yongxin Deng, Zhen Fang, Sharon Li, et al.
+- **abs:** https://arxiv.org/abs/2601.19245 · **PDF:** https://arxiv.org/pdf/2601.19245v5
+
+## 18976. Contrast-Source-Based Physics-Driven Neural Network for Inverse Scattering Problems
+
+- **ID:** `2601.19243` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Yutong Du, Zicheng Liu
+- **abs:** https://arxiv.org/abs/2601.19243 · **PDF:** https://arxiv.org/pdf/2601.19243v1
+
+## 18977. VoxPrivacy: A Benchmark for Evaluating Interactional Privacy of Speech Language Models
+
+- **ID:** `2601.19956` · **primary:** `eess.AS` · **Submitted:** 2026-01-27
+- **Authors:** Yuxiang Wang, Hongyu Liu, Dekun Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.19956 · **PDF:** https://arxiv.org/pdf/2601.19956v1
+
+## 18978. Structure-based RNA Design by Step-wise Optimization of Latent Diffusion Model
+
+- **ID:** `2601.19232` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Qi Si, Xuyang Liu, Penglei Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19232 · **PDF:** https://arxiv.org/pdf/2601.19232v1
+
+## 18979. Variation is the Key: A Variation-Based Framework for LLM-Generated Text Detection
+
+- **ID:** `2602.13226` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Xuecong Li, Xiaohong Li, Qiang Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.13226 · **PDF:** https://arxiv.org/pdf/2602.13226v1
+
+## 18980. RPO-RAG: Aligning Small LLMs with Relation-aware Preference Optimization for Knowledge Graph Question Answering
+
+- **ID:** `2601.19225` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Kaehyun Um, KyuHwan Yeom, Haerim Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.19225 · **PDF:** https://arxiv.org/pdf/2601.19225v2
+
+## 18981. Nonlocal Kramers-Moyal formulas and data-driven discovery of stochastic dynamical systems with multiplicative Lévy noise
+
+- **ID:** `2601.19223` · **primary:** `math.DS` · **Submitted:** 2026-01-27
+- **Authors:** Yang Li, Jinqiao Duan
+- **abs:** https://arxiv.org/abs/2601.19223 · **PDF:** https://arxiv.org/pdf/2601.19223v1
+
+## 18982. UniPCB: A Unified Vision-Language Benchmark for Open-Ended PCB Quality Inspection
+
+- **ID:** `2601.19222` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Fuxiang Sun, Xi Jiang, Jiansheng Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.19222 · **PDF:** https://arxiv.org/pdf/2601.19222v1
+
+## 18983. DREAMSTATE: Diffusing States and Parameters for Recurrent Large Language Models
+
+- **ID:** `2601.19221` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Liu Xiao
+- **abs:** https://arxiv.org/abs/2601.19221 · **PDF:** https://arxiv.org/pdf/2601.19221v1
+
+## 18984. Accelerated Multiple Wasserstein Gradient Flows for Multi-objective Distributional Optimization
+
+- **ID:** `2601.19220` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Dai Hai Nguyen, Duc Dung Nguyen, Atsuyoshi Nakamura, et al.
+- **abs:** https://arxiv.org/abs/2601.19220 · **PDF:** https://arxiv.org/pdf/2601.19220v1
+
+## 18985. Bridging Visual and Wireless Sensing: A Unified Radiation Field for 3D Radio Map Construction
+
+- **ID:** `2601.19216` · **primary:** `cs.NI` · **Submitted:** 2026-01-27
+- **Authors:** Chaozheng Wen, Jingwen Tong, Zehong Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.19216 · **PDF:** https://arxiv.org/pdf/2601.19216v1
+
+## 18986. A Hybrid Supervised-LLM Pipeline for Actionable Suggestion Mining in Unstructured Customer Reviews
+
+- **ID:** `2601.19214` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Aakash Trivedi, Aniket Upadhyay, Pratik Narang, et al.
+- **abs:** https://arxiv.org/abs/2601.19214 · **PDF:** https://arxiv.org/pdf/2601.19214v1
+
+## 18987. How Do Transformers Learn to Associate Tokens: Gradient Leading Terms Bring Mechanistic Interpretability
+
+- **ID:** `2601.19208` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Shawn Im, Changdae Oh, Zhen Fang, et al.
+- **abs:** https://arxiv.org/abs/2601.19208 · **PDF:** https://arxiv.org/pdf/2601.19208v1
+
+## 18988. EnzyPGM: Pocket-conditioned Generative Model for Substrate-specific Enzyme Design
+
+- **ID:** `2601.19205` · **primary:** `q-bio.BM` · **Submitted:** 2026-01-27
+- **Authors:** Zefeng Lin, Zhihang Zhang, Weirong Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.19205 · **PDF:** https://arxiv.org/pdf/2601.19205v1
+
+## 18989. MATA: A Trainable Hierarchical Automaton System for Multi-Agent Visual Reasoning
+
+- **ID:** `2601.19204` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Zhixi Cai, Fucai Ke, Kevin Leo, et al.
+- **abs:** https://arxiv.org/abs/2601.19204 · **PDF:** https://arxiv.org/pdf/2601.19204v1
+
+## 18990. Do Images Speak Louder than Words? Investigating the Effect of Textual Misinformation in VLMs
+
+- **ID:** `2601.19202` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Chi Zhang, Wenxuan Ding, Jiale Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.19202 · **PDF:** https://arxiv.org/pdf/2601.19202v1
+
+## 18991. MAGNET: Towards Adaptive GUI Agents with Memory-Driven Knowledge Evolution
+
+- **ID:** `2601.19199` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Libo Sun, Jiwen Zhang, Siyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19199 · **PDF:** https://arxiv.org/pdf/2601.19199v2
+
+## 18992. HELM: A Human-Centered Evaluation Framework for LLM-Powered Recommender Systems
+
+- **ID:** `2601.19197` · **primary:** `cs.IR` · **Submitted:** 2026-01-27
+- **Authors:** Sushant Mehta
+- **abs:** https://arxiv.org/abs/2601.19197 · **PDF:** https://arxiv.org/pdf/2601.19197v1
+
+## 18993. SE-DiCoW: Self-Enrolled Diarization-Conditioned Whisper
+
+- **ID:** `2601.19194` · **primary:** `eess.AS` · **Submitted:** 2026-01-27
+- **Authors:** Alexander Polok, Dominik Klement, Samuele Cornell, et al.
+- **abs:** https://arxiv.org/abs/2601.19194 · **PDF:** https://arxiv.org/pdf/2601.19194v1
+
+## 18994. CoReTab: Improving Multimodal Table Understanding with Code-driven Reasoning
+
+- **ID:** `2601.19193` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Van-Quang Nguyen, Takayuki Okatani
+- **abs:** https://arxiv.org/abs/2601.19193 · **PDF:** https://arxiv.org/pdf/2601.19193v1
+
+## 18995. MetaMem: Evolving Meta-Memory for Knowledge Utilization through Self-Reflective Symbolic Optimization
+
+- **ID:** `2602.11182` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Haidong Xin, Xinze Li, Zhenghao Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.11182 · **PDF:** https://arxiv.org/pdf/2602.11182v1
+
+## 18996. Transparency-First Medical Language Models: Datasheets, Model Cards, and End-to-End Data Provenance for Clinical NLP
+
+- **ID:** `2601.19191` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Olaf Yunus Laitinen Imanov, Taner Yilmaz, Ayse Tuba Tugrul, et al.
+- **abs:** https://arxiv.org/abs/2601.19191 · **PDF:** https://arxiv.org/pdf/2601.19191v1
+
+## 18997. Foresight Learning for SEC Risk Prediction
+
+- **ID:** `2601.19189` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Benjamin Turtel, Paul Wilczewski, Danny Franklin, et al.
+- **abs:** https://arxiv.org/abs/2601.19189 · **PDF:** https://arxiv.org/pdf/2601.19189v1
+
+## 18998. Double Fairness Policy Learning: Integrating Action Fairness and Outcome Fairness in Decision-making
+
+- **ID:** `2601.19186` · **primary:** `stat.ML` · **Submitted:** 2026-01-27
+- **Authors:** Zeyu Bian, Lan Wang, Chengchun Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.19186 · **PDF:** https://arxiv.org/pdf/2601.19186v2
+
+## 18999. SNR-Edit: Structure-Aware Noise Rectification for Inversion-Free Flow-Based Editing
+
+- **ID:** `2601.19180` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Lifan Jiang, Boxi Wu, Yuhang Pei, et al.
+- **abs:** https://arxiv.org/abs/2601.19180 · **PDF:** https://arxiv.org/pdf/2601.19180v1
+
+## 19000. Learning Ordered Representations in Latent Space for Intrinsic Dimension Estimation via Principal Component Autoencoder
+
+- **ID:** `2601.19179` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Qipeng Zhan, Zhuoping Zhou, Zexuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19179 · **PDF:** https://arxiv.org/pdf/2601.19179v1
+
+## 19001. CollectiveKV: Decoupling and Sharing Collaborative Information in Sequential Recommendation
+
+- **ID:** `2601.19178` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Jingyu Li, Zhaocheng Du, Qianhui Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.19178 · **PDF:** https://arxiv.org/pdf/2601.19178v2
+
+## 19002. A Scalable Inter-edge Correlation Modeling in CopulaGNN for Link Sign Prediction
+
+- **ID:** `2601.19175` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Jinkyu Sung, Myunggeum Jee, Joonseok Lee
+- **abs:** https://arxiv.org/abs/2601.19175 · **PDF:** https://arxiv.org/pdf/2601.19175v4
+
+## 19003. SHIELD: An Auto-Healing Agentic Defense Framework for LLM Resource Exhaustion Attacks
+
+- **ID:** `2601.19174` · **primary:** `cs.CR` · **Submitted:** 2026-01-27
+- **Authors:** Nirhoshan Sivaroopan, Kanchana Thilakarathna, Albert Zomaya, et al.
+- **abs:** https://arxiv.org/abs/2601.19174 · **PDF:** https://arxiv.org/pdf/2601.19174v1
+
+## 19004. Bridging Gulfs in UI Generation through Semantic Guidance
+
+- **ID:** `2601.19171` · **primary:** `cs.HC` · **Submitted:** 2026-01-27
+- **Authors:** Seokhyeon Park, Soohyun Lee, Eugene Choi, et al.
+- **abs:** https://arxiv.org/abs/2601.19171 · **PDF:** https://arxiv.org/pdf/2601.19171v2
+
+## 19005. Multi-Agent Procedural Graph Extraction with Structural and Logical Refinement
+
+- **ID:** `2601.19170` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Wangyang Ying, Yanchi Liu, Xujiang Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.19170 · **PDF:** https://arxiv.org/pdf/2601.19170v1
+
+## 19006. Convergence of Muon with Newton-Schulz
+
+- **ID:** `2601.19156` · **primary:** `stat.ML` · **Submitted:** 2026-01-27
+- **Authors:** Gyu Yeol Kim, Min-hwan Oh
+- **abs:** https://arxiv.org/abs/2601.19156 · **PDF:** https://arxiv.org/pdf/2601.19156v1
+
+## 19007. LocationAgent: A Hierarchical Agent for Image Geolocation via Decoupling Strategy and Evidence from Parametric Knowledge
+
+- **ID:** `2601.19155` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Qiujun Li, Zijin Xiao, Xulin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19155 · **PDF:** https://arxiv.org/pdf/2601.19155v1
+
+## 19008. Analysis of Shuffling Beyond Pure Local Differential Privacy
+
+- **ID:** `2601.19154` · **primary:** `cs.DS` · **Submitted:** 2026-01-27
+- **Authors:** Shun Takagi, Seng Pei Liew
+- **abs:** https://arxiv.org/abs/2601.19154 · **PDF:** https://arxiv.org/pdf/2601.19154v4
+
+## 19009. TS-Debate: Multimodal Collaborative Debate for Zero-Shot Time Series Reasoning
+
+- **ID:** `2601.19151` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Patara Trirat, Jin Myung Kwak, Jay Heo, et al.
+- **abs:** https://arxiv.org/abs/2601.19151 · **PDF:** https://arxiv.org/pdf/2601.19151v1
+
+## 19010. GPCR-Filter: a deep learning framework for efficient and precise GPCR modulator discovery
+
+- **ID:** `2601.19149` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Jingjie Ning, Xiangzhen Shen, Li Hou, et al.
+- **abs:** https://arxiv.org/abs/2601.19149 · **PDF:** https://arxiv.org/pdf/2601.19149v2
+
+## 19011. Length-Adaptive Interest Network for Balancing Long and Short Sequence Modeling in CTR Prediction
+
+- **ID:** `2601.19142` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Zhicheng Zhang, Zhaocheng Du, Jieming Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.19142 · **PDF:** https://arxiv.org/pdf/2601.19142v1
+
+## 19012. Native LLM and MLLM Inference at Scale on Apple Silicon
+
+- **ID:** `2601.19139` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Wayner Barrios
+- **abs:** https://arxiv.org/abs/2601.19139 · **PDF:** https://arxiv.org/pdf/2601.19139v2
+
+## 19013. AgenticSCR: An Autonomous Agentic Secure Code Review for Immature Vulnerabilities Detection
+
+- **ID:** `2601.19138` · **primary:** `cs.CR` · **Submitted:** 2026-01-27
+- **Authors:** Wachiraphan Charoenwet, Kla Tantithamthavorn, Patanamon Thongtanunam, et al.
+- **abs:** https://arxiv.org/abs/2601.19138 · **PDF:** https://arxiv.org/pdf/2601.19138v1
+
+## 19014. In-Network Collective Operations: Game Changer or Challenge for AI Workloads?
+
+- **ID:** `2601.19132` · **primary:** `cs.NI` · **Submitted:** 2026-01-27
+- **Authors:** Torsten Hoefler, Mikhail Khalilov, Josiah Clark, et al.
+- **abs:** https://arxiv.org/abs/2601.19132 · **PDF:** https://arxiv.org/pdf/2601.19132v1
+
+## 19015. CLIP-Guided Unsupervised Semantic-Aware Exposure Correction
+
+- **ID:** `2601.19129` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Puzhen Wu, Han Weng, Quan Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.19129 · **PDF:** https://arxiv.org/pdf/2601.19129v2
+
+## 19016. Leveraging Sentence-oriented Augmentation and Transformer-Based Architecture for Vietnamese-Bahnaric Translation
+
+- **ID:** `2601.19124` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Tan Sang Nguyen, Quoc Nguyen Pham, Tho Quan
+- **abs:** https://arxiv.org/abs/2601.19124 · **PDF:** https://arxiv.org/pdf/2601.19124v1
+
+## 19017. Exploring Weaknesses in Function Call Models via Reinforcement Learning: An Adversarial Data Augmentation Approach
+
+- **ID:** `2601.19122` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Weiran Guo, Bing Bo, Shaoxiang Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.19122 · **PDF:** https://arxiv.org/pdf/2601.19122v1
+
+## 19018. LLMs as Orchestrators: Constraint-Compliant Multi-Agent Optimization for Recommendation Systems
+
+- **ID:** `2601.19121` · **primary:** `cs.IR` · **Submitted:** 2026-01-27
+- **Authors:** Guilin Zhang, Kai Zhao, Jeffrey Friedman, et al.
+- **abs:** https://arxiv.org/abs/2601.19121 · **PDF:** https://arxiv.org/pdf/2601.19121v3
+
+## 19019. RobustExplain: Evaluating Robustness of LLM-Based Explanation Agents for Recommendation
+
+- **ID:** `2601.19120` · **primary:** `cs.IR` · **Submitted:** 2026-01-27
+- **Authors:** Guilin Zhang, Kai Zhao, Jeffrey Friedman, et al.
+- **abs:** https://arxiv.org/abs/2601.19120 · **PDF:** https://arxiv.org/pdf/2601.19120v3
+
+## 19020. Uncertainty-Aware 3D Emotional Talking Face Synthesis with Emotion Prior Distillation
+
+- **ID:** `2601.19112` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Nanhan Shen, Zhilei Liu
+- **abs:** https://arxiv.org/abs/2601.19112 · **PDF:** https://arxiv.org/pdf/2601.19112v1
+
+## 19021. TinyTorch: Building Machine Learning Systems from First Principles
+
+- **ID:** `2601.19107` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Vijay Janapa Reddi
+- **abs:** https://arxiv.org/abs/2601.19107 · **PDF:** https://arxiv.org/pdf/2601.19107v2
+
+## 19022. Detecting and Correcting Hallucinations in LLM-Generated Code via Deterministic AST Analysis
+
+- **ID:** `2601.19106` · **primary:** `cs.SE` · **Submitted:** 2026-01-27
+- **Authors:** Dipin Khati, Daniel Rodriguez-Cardenas, Paul Pantzer, et al.
+- **abs:** https://arxiv.org/abs/2601.19106 · **PDF:** https://arxiv.org/pdf/2601.19106v1
+
+## 19023. Founder effects shape the evolutionary dynamics of multimodality in open LLM families
+
+- **ID:** `2603.22287` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Manuel Cebrian
+- **abs:** https://arxiv.org/abs/2603.22287 · **PDF:** https://arxiv.org/pdf/2603.22287v1
+
+## 19024. OWLEYE: Zero-Shot Learner for Cross-Domain Graph Data Anomaly Detection
+
+- **ID:** `2601.19102` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Lecheng Zheng, Dongqi Fu, Zihao Li, et al.
+- **abs:** https://arxiv.org/abs/2601.19102 · **PDF:** https://arxiv.org/pdf/2601.19102v2
+
+## 19025. m2sv: A Scalable Benchmark for Map-to-Street-View Spatial Reasoning
+
+- **ID:** `2601.19099` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Yosub Shin, Michael Buriek, Igor Molybog
+- **abs:** https://arxiv.org/abs/2601.19099 · **PDF:** https://arxiv.org/pdf/2601.19099v1
+
+## 19026. PsyProbe: Proactive and Interpretable Dialogue through User State Modeling for Exploratory Counseling
+
+- **ID:** `2601.19096` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Sohhyung Park, Hyunji Kang, Sungzoon Cho, et al.
+- **abs:** https://arxiv.org/abs/2601.19096 · **PDF:** https://arxiv.org/pdf/2601.19096v1
+
+## 19027. FloydNet: A Learning Paradigm for Global Relational Reasoning
+
+- **ID:** `2601.19094` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Jingcheng Yu, Mingliang Zeng, Qiwei Ye
+- **abs:** https://arxiv.org/abs/2601.19094 · **PDF:** https://arxiv.org/pdf/2601.19094v2
+
+## 19028. NeuroAI and Beyond
+
+- **ID:** `2601.19955` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Jean-Marc Fellous, Gert Cauwenberghs, Cornelia Fermüller, et al.
+- **abs:** https://arxiv.org/abs/2601.19955 · **PDF:** https://arxiv.org/pdf/2601.19955v1
+
+## 19029. Axe: A Simple Unified Layout Abstraction for Machine Learning Compilers
+
+- **ID:** `2601.19092` · **primary:** `cs.DC` · **Submitted:** 2026-01-27
+- **Authors:** Bohan Hou, Hongyi Jin, Guanjie Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19092 · **PDF:** https://arxiv.org/pdf/2601.19092v2
+
+## 19030. Out-of-Distribution Generalization for Neural Physics Solvers
+
+- **ID:** `2601.19091` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Zhao Wei, Chin Chun Ooi, Jian Cheng Wong, et al.
+- **abs:** https://arxiv.org/abs/2601.19091 · **PDF:** https://arxiv.org/pdf/2601.19091v1
+
+## 19031. LEMON: How Well Do MLLMs Perform Temporal Multimodal Understanding on Instructional Videos?
+
+- **ID:** `2601.20705` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Zhuang Yu, Lei Shen, Jing Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.20705 · **PDF:** https://arxiv.org/pdf/2601.20705v1
+
+## 19032. Privacy-Preserving Model Transcription with Differentially Private Synthetic Distillation
+
+- **ID:** `2601.19090` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Bochao Liu, Shiming Ge, Pengju Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19090 · **PDF:** https://arxiv.org/pdf/2601.19090v1
+
+## 19033. EPAS: Efficient Training with Progressive Activation Sharing
+
+- **ID:** `2601.19089` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Rezaul Karim, Maryam Dialameh, Yang Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.19089 · **PDF:** https://arxiv.org/pdf/2601.19089v1
+
+## 19034. Speed is Confidence
+
+- **ID:** `2601.19085` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Joshua V. Dillon
+- **abs:** https://arxiv.org/abs/2601.19085 · **PDF:** https://arxiv.org/pdf/2601.19085v2
+
+## 19035. More at Stake: How Payoff and Language Shape LLM Agent Strategies in Cooperation Dilemmas
+
+- **ID:** `2601.19082` · **primary:** `cs.AI` · **Submitted:** 2026-01-27
+- **Authors:** Trung-Kiet Huynh, Dao-Sy Duy-Minh, Thanh-Bang Cao, et al.
+- **abs:** https://arxiv.org/abs/2601.19082 · **PDF:** https://arxiv.org/pdf/2601.19082v1
+
+## 19036. Transformer Learning of Chaotic Collective Dynamics in Many-Body Systems
+
+- **ID:** `2601.19080` · **primary:** `physics.comp-ph` · **Submitted:** 2026-01-27
+- **Authors:** Ho Jang, Gia-Wei Chern
+- **abs:** https://arxiv.org/abs/2601.19080 · **PDF:** https://arxiv.org/pdf/2601.19080v1
+
+## 19037. C2NP: A Benchmark for Learning Scale-Dependent Geometric Invariances in 3D Materials Generation
+
+- **ID:** `2601.19076` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-27
+- **Authors:** Can Polat, Erchin Serpedin, Mustafa Kurban, et al.
+- **abs:** https://arxiv.org/abs/2601.19076 · **PDF:** https://arxiv.org/pdf/2601.19076v1
+
+## 19038. The "Robert Boulton" Singularity: Semantic Tunneling and Manifold Unfolding in Recursive AI
+
+- **ID:** `2602.02526` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Pengyue Hou
+- **abs:** https://arxiv.org/abs/2602.02526 · **PDF:** https://arxiv.org/pdf/2602.02526v1
+
+## 19039. HalluJudge: A Reference-Free Hallucination Detection for Context Misalignment in Code Review Automation
+
+- **ID:** `2601.19072` · **primary:** `cs.SE` · **Submitted:** 2026-01-27
+- **Authors:** Kla Tantithamthavorn, Hong Yi Lin, Patanamon Thongtanunam, et al.
+- **abs:** https://arxiv.org/abs/2601.19072 · **PDF:** https://arxiv.org/pdf/2601.19072v2
+
+## 19040. Critical Organization of Deep Neural Networks, and p-Adic Statistical Field Theories
+
+- **ID:** `2601.19070` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** W. A. Zúñiga-Galindo
+- **abs:** https://arxiv.org/abs/2601.19070 · **PDF:** https://arxiv.org/pdf/2601.19070v1
+
+## 19041. Transit Network Design with Two-Level Demand Uncertainties: A Machine Learning and Contextual Stochastic Optimization Framework
+
+- **ID:** `2603.00010` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Hongzhao Guan, Beste Basciftci, Pascal Van Hentenryck
+- **abs:** https://arxiv.org/abs/2603.00010 · **PDF:** https://arxiv.org/pdf/2603.00010v2
+
+## 19042. Dynamic Cogeneration of Bug Reproduction Test in Agentic Program Repair
+
+- **ID:** `2601.19066` · **primary:** `cs.SE` · **Submitted:** 2026-01-27
+- **Authors:** Runxiang Cheng, Michele Tufano, José Cambronero, et al.
+- **abs:** https://arxiv.org/abs/2601.19066 · **PDF:** https://arxiv.org/pdf/2601.19066v2
+
+## 19043. Optimizing Conversational Quality in Spoken Dialogue Systems with Reinforcement Learning from AI Feedback
+
+- **ID:** `2601.19063` · **primary:** `cs.CL` · **Submitted:** 2026-01-27
+- **Authors:** Siddhant Arora, Jinchuan Tian, Jiatong Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.19063 · **PDF:** https://arxiv.org/pdf/2601.19063v1
+
+## 19044. Who's in Charge? Disempowerment Patterns in Real-World LLM Usage
+
+- **ID:** `2601.19062` · **primary:** `cs.CY` · **Submitted:** 2026-01-27
+- **Authors:** Mrinank Sharma, Miles McCain, Raymond Douglas, et al.
+- **abs:** https://arxiv.org/abs/2601.19062 · **PDF:** https://arxiv.org/pdf/2601.19062v1
+
+## 19045. Thought-Transfer: Indirect Targeted Poisoning Attacks on Chain-of-Thought Reasoning Models
+
+- **ID:** `2601.19061` · **primary:** `cs.CR` · **Submitted:** 2026-01-27
+- **Authors:** Harsh Chaudhari, Ethan Rathbun, Hanna Foerster, et al.
+- **abs:** https://arxiv.org/abs/2601.19061 · **PDF:** https://arxiv.org/pdf/2601.19061v2
+
+## 19046. Pixel-Grounded Retrieval for Knowledgeable Large Multimodal Models
+
+- **ID:** `2601.19060` · **primary:** `cs.CV` · **Submitted:** 2026-01-27
+- **Authors:** Jeonghwan Kim, Renjie Tao, Sanat Sharma, et al.
+- **abs:** https://arxiv.org/abs/2601.19060 · **PDF:** https://arxiv.org/pdf/2601.19060v1
+
+## 19047. Probabilistic Sensing: Intelligence in Data Sampling
+
+- **ID:** `2601.19953` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Ibrahim Albulushi, Saleh Bunaiyan, Suraj S. Cheema, et al.
+- **abs:** https://arxiv.org/abs/2601.19953 · **PDF:** https://arxiv.org/pdf/2601.19953v1
+
+## 19048. Principled Fine-tuning of LLMs from User-Edits: A Medley of Preference, Supervision, and Reward
+
+- **ID:** `2601.19055` · **primary:** `cs.LG` · **Submitted:** 2026-01-27
+- **Authors:** Dipendra Misra, Aldo Pacchiano, Ta-Chung Chi, et al.
+- **abs:** https://arxiv.org/abs/2601.19055 · **PDF:** https://arxiv.org/pdf/2601.19055v1
+
+## 19049. From Answer Givers to Design Mentors: Guiding LLMs with the Cognitive Apprenticeship Model
+
+- **ID:** `2601.19053` · **primary:** `cs.HC` · **Submitted:** 2026-01-27
+- **Authors:** Yongsu Ahn, Lejun R Liao, Benjamin Bach, et al.
+- **abs:** https://arxiv.org/abs/2601.19053 · **PDF:** https://arxiv.org/pdf/2601.19053v1
+
+## 19050. HEATACO: Heatmap-Guided Ant Colony Decoding for Large-Scale Travelling Salesman Problems
+
+- **ID:** `2601.19041` · **primary:** `cs.NE` · **Submitted:** 2026-01-26
+- **Authors:** Bo-Cheng Lin, Yi Mei, Mengjie Zhang
+- **abs:** https://arxiv.org/abs/2601.19041 · **PDF:** https://arxiv.org/pdf/2601.19041v1
+
+## 19051. OATS: Online Data Augmentation for Time Series Foundation Models
+
+- **ID:** `2601.19040` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Junwei Deng, Chang Xu, Jiaqi W. Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.19040 · **PDF:** https://arxiv.org/pdf/2601.19040v1
+
+## 19052. Exploring the Use of VLMs for Navigation Assistance for People with Blindness and Low Vision
+
+- **ID:** `2603.15624` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Yu Li, Yuchen Zheng, Giles Hamilton-Fletcher, et al.
+- **abs:** https://arxiv.org/abs/2603.15624 · **PDF:** https://arxiv.org/pdf/2603.15624v1
+
+## 19053. XIMP: Cross Graph Inter-Message Passing for Molecular Property Prediction
+
+- **ID:** `2601.19037` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Anatol Ehrlich, Lorenz Kummer, Vojtech Voracek, et al.
+- **abs:** https://arxiv.org/abs/2601.19037 · **PDF:** https://arxiv.org/pdf/2601.19037v1
+
+## 19054. Unravelling the (In)compatibility of Statistical-Parity and Equalized-Odds
+
+- **ID:** `2601.19035` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Mortaza S. Bargh, Sunil Choenni, Floris ter Braak
+- **abs:** https://arxiv.org/abs/2601.19035 · **PDF:** https://arxiv.org/pdf/2601.19035v1
+
+## 19055. Spatial Heterogeneity in Climate Risk and Human Flourishing: An Exploration with Generative AI
+
+- **ID:** `2601.20880` · **primary:** `stat.AP` · **Submitted:** 2026-01-26
+- **Authors:** Stefano Maria Iacus, Haodong Qi, Devika Jain
+- **abs:** https://arxiv.org/abs/2601.20880 · **PDF:** https://arxiv.org/pdf/2601.20880v1
+
+## 19056. A Unifying View of Coverage in Linear Off-Policy Evaluation
+
+- **ID:** `2601.19030` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Philip Amortila, Audrey Huang, Akshay Krishnamurthy, et al.
+- **abs:** https://arxiv.org/abs/2601.19030 · **PDF:** https://arxiv.org/pdf/2601.19030v1
+
+## 19057. Is Finer Better? The Limits of Microscaling Formats in Large Language Models
+
+- **ID:** `2601.19026` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Andrea Fasoli, Monodeep Kar, Chi-Chun Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.19026 · **PDF:** https://arxiv.org/pdf/2601.19026v1
+
+## 19058. EVEREST: An Evidential, Tail-Aware Transformer for Rare-Event Time-Series Forecasting
+
+- **ID:** `2601.19022` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Antanas Zilinskas, Robert N. Shorten, Jakub Marecek
+- **abs:** https://arxiv.org/abs/2601.19022 · **PDF:** https://arxiv.org/pdf/2601.19022v2
+
+## 19059. Smooth embeddings in contracting recurrent networks driven by regular dynamics: A synthesis for neural representation
+
+- **ID:** `2601.19019` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-26
+- **Authors:** Vikas N. O'Reilly-Shah, Alessandro Maria Selvitella
+- **abs:** https://arxiv.org/abs/2601.19019 · **PDF:** https://arxiv.org/pdf/2601.19019v1
+
+## 19060. A Framework for Evaluating Faithfulness in Explainable AI for Machine Anomalous Sound Detection Using Frequency-Band Perturbation
+
+- **ID:** `2601.19017` · **primary:** `cs.SD` · **Submitted:** 2026-01-26
+- **Authors:** Alexander Buck, Georgina Cosma, Iain Phillips, et al.
+- **abs:** https://arxiv.org/abs/2601.19017 · **PDF:** https://arxiv.org/pdf/2601.19017v1
+
+## 19061. Observing Health Outcomes Using Remote Sensing Imagery and Geo-Context Guided Visual Transformer
+
+- **ID:** `2602.00110` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Yu Li, Guilherme N. DeSouza, Praveen Rao, et al.
+- **abs:** https://arxiv.org/abs/2602.00110 · **PDF:** https://arxiv.org/pdf/2602.00110v1
+
+## 19062. Accelerated training of Gaussian processes using banded square exponential covariances
+
+- **ID:** `2601.19007` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Emily C. Ehrhardt, Felipe Tobar
+- **abs:** https://arxiv.org/abs/2601.19007 · **PDF:** https://arxiv.org/pdf/2601.19007v1
+
+## 19063. Recommending Composite Items Using Multi-Level Preference Information: A Joint Interaction Modeling Approach
+
+- **ID:** `2601.19005` · **primary:** `cs.IR` · **Submitted:** 2026-01-26
+- **Authors:** Xuan Bi, Yaqiong Wang, Gediminas Adomavicius, et al.
+- **abs:** https://arxiv.org/abs/2601.19005 · **PDF:** https://arxiv.org/pdf/2601.19005v1
+
+## 19064. FROST: Filtering Reasoning Outliers with Attention for Efficient Reasoning
+
+- **ID:** `2601.19001` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Haozheng Luo, Zhuolin Jiang, Md Zahid Hasan, et al.
+- **abs:** https://arxiv.org/abs/2601.19001 · **PDF:** https://arxiv.org/pdf/2601.19001v2
+
+## 19065. Randomization Boosts KV Caching, Learning Balances Query Load: A Joint Perspective
+
+- **ID:** `2601.18999` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Fangzhou Wu, Sandeep Silwal, Qiuyi, et al.
+- **abs:** https://arxiv.org/abs/2601.18999 · **PDF:** https://arxiv.org/pdf/2601.18999v1
+
+## 19066. Malicious Repurposing of Open Science Artefacts by Using Large Language Models
+
+- **ID:** `2601.18998` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Zahra Hashemi, Zhiqiang Zhong, Jun Pang, et al.
+- **abs:** https://arxiv.org/abs/2601.18998 · **PDF:** https://arxiv.org/pdf/2601.18998v1
+
+## 19067. Anatomically-aware conformal prediction for medical image segmentation with random walks
+
+- **ID:** `2601.18997` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Mélanie Gaillochet, Christian Desrosiers, Hervé Lombaert
+- **abs:** https://arxiv.org/abs/2601.18997 · **PDF:** https://arxiv.org/pdf/2601.18997v1
+
+## 19068. A Geometric Taxonomy of Hallucinations in LLMs
+
+- **ID:** `2602.13224` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Javier Marín
+- **abs:** https://arxiv.org/abs/2602.13224 · **PDF:** https://arxiv.org/pdf/2602.13224v2
+
+## 19069. FreeOrbit4D: Training-Free Arbitrary Camera Redirection for Monocular Videos via Geometry-Complete 4D Reconstruction
+
+- **ID:** `2601.18993` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Wei Cao, Hao Zhang, Fengrui Tian, et al.
+- **abs:** https://arxiv.org/abs/2601.18993 · **PDF:** https://arxiv.org/pdf/2601.18993v1
+
+## 19070. LLMs versus the Halting Problem: Revisiting Program Termination Prediction
+
+- **ID:** `2601.18987` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Oren Sultan, Jordi Armengol-Estape, Pascal Kesseli, et al.
+- **abs:** https://arxiv.org/abs/2601.18987 · **PDF:** https://arxiv.org/pdf/2601.18987v4
+
+## 19071. Save the Good Prefix: Precise Error Penalization via Process-Supervised RL to Enhance LLM Reasoning
+
+- **ID:** `2601.18984` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Haolin Liu, Dian Yu, Sidi Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.18984 · **PDF:** https://arxiv.org/pdf/2601.18984v1
+
+## 19072. Attention-Enhanced Graph Filtering for False Data Injection Attack Detection and Localization
+
+- **ID:** `2601.18981` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Ruslan Abdulin, Mohammad Rasoul Narimani
+- **abs:** https://arxiv.org/abs/2601.18981 · **PDF:** https://arxiv.org/pdf/2601.18981v1
+
+## 19073. Right Move, Right Time: Multi-Sport Space Evaluation Platform for Ultimate Frisbee, Basketball, and Soccer
+
+- **ID:** `2603.06585` · **primary:** `cs.HC` · **Submitted:** 2026-01-26
+- **Authors:** Shunsuke Iwashita, Titouan Jeannot, Braden Eberhard, et al.
+- **abs:** https://arxiv.org/abs/2603.06585 · **PDF:** https://arxiv.org/pdf/2603.06585v1
+
+## 19074. Intent2QoS: Language Model-Driven Automation of Traffic Shaping Configurations
+
+- **ID:** `2601.18974` · **primary:** `cs.NI` · **Submitted:** 2026-01-26
+- **Authors:** Sudipta Acharya, Burak Kantarci
+- **abs:** https://arxiv.org/abs/2601.18974 · **PDF:** https://arxiv.org/pdf/2601.18974v1
+
+## 19075. When Does Adaptation Win? Scaling Laws for Meta-Learning in Quantum Control
+
+- **ID:** `2601.18973` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Nima Leclerc, Chris Miller, Nicholas Brawand
+- **abs:** https://arxiv.org/abs/2601.18973 · **PDF:** https://arxiv.org/pdf/2601.18973v3
+
+## 19076. Towards Self-Optimizing Electron Microscope: Robust Tuning of Aberration Coefficients via Physics-Aware Multi-Objective Bayesian Optimization
+
+- **ID:** `2601.18972` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Utkarsh Pratiush, Austin Houston, Richard Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.18972 · **PDF:** https://arxiv.org/pdf/2601.18972v1
+
+## 19077. NLP Privacy Risk Identification in Social Media (NLP-PRISM): A Survey
+
+- **ID:** `2602.15866` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Dhiman Goswami, Jai Kruthunz Naveen Kumar, Sanchari Das
+- **abs:** https://arxiv.org/abs/2602.15866 · **PDF:** https://arxiv.org/pdf/2602.15866v1
+
+## 19078. Fauna Sprout: A lightweight, approachable, developer-ready humanoid robot
+
+- **ID:** `2601.18963` · **primary:** `cs.RO` · **Submitted:** 2026-01-26
+- **Authors:** Fauna Robotics, :, Diego Aldarondo, et al.
+- **abs:** https://arxiv.org/abs/2601.18963 · **PDF:** https://arxiv.org/pdf/2601.18963v1
+
+## 19079. Reinforcement Learning for Quantum Technology
+
+- **ID:** `2601.18953` · **primary:** `quant-ph` · **Submitted:** 2026-01-26
+- **Authors:** Marin Bukov, Florian Marquardt
+- **abs:** https://arxiv.org/abs/2601.18953 · **PDF:** https://arxiv.org/pdf/2601.18953v1
+
+## 19080. Vector-Valued Distributional Reinforcement Learning Policy Evaluation: A Hilbert Space Embedding Approach
+
+- **ID:** `2601.18952` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Mehrdad Mohammadi, Qi Zheng, Ruoqing Zhu
+- **abs:** https://arxiv.org/abs/2601.18952 · **PDF:** https://arxiv.org/pdf/2601.18952v1
+
+## 19081. Collaborative Compressors in Distributed Mean Estimation with Limited Communication Budget
+
+- **ID:** `2601.18950` · **primary:** `stat.ML` · **Submitted:** 2026-01-26
+- **Authors:** Harsh Vardhan, Arya Mazumdar
+- **abs:** https://arxiv.org/abs/2601.18950 · **PDF:** https://arxiv.org/pdf/2601.18950v1
+
+## 19082. Tricky$^2$: Towards a Benchmark for Evaluating Human and LLM Error Interactions
+
+- **ID:** `2601.18949` · **primary:** `cs.SE` · **Submitted:** 2026-01-26
+- **Authors:** Cole Granger, Dipin Khati, Daniel Rodriguez-Cardenas, et al.
+- **abs:** https://arxiv.org/abs/2601.18949 · **PDF:** https://arxiv.org/pdf/2601.18949v1
+
+## 19083. Neural Theorem Proving for Verification Conditions: A Real-World Benchmark
+
+- **ID:** `2601.18944` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Qiyuan Xu, Xiaokun Luan, Renxi Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.18944 · **PDF:** https://arxiv.org/pdf/2601.18944v2
+
+## 19084. Configurable p-Neurons Using Modular p-Bits
+
+- **ID:** `2601.18943` · **primary:** `cs.ET` · **Submitted:** 2026-01-26
+- **Authors:** Saleh Bunaiyan, Mohammad Alsharif, Abdelrahman S. Abdelrahman, et al.
+- **abs:** https://arxiv.org/abs/2601.18943 · **PDF:** https://arxiv.org/pdf/2601.18943v1
+
+## 19085. A Few Bad Neurons: Isolating and Surgically Correcting Sycophancy
+
+- **ID:** `2601.18939` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Claire O'Brien, Jessica Seto, Dristi Roy, et al.
+- **abs:** https://arxiv.org/abs/2601.18939 · **PDF:** https://arxiv.org/pdf/2601.18939v1
+
+## 19086. FSD-CAP: Fractional Subgraph Diffusion with Class-Aware Propagation for Graph Feature Imputation
+
+- **ID:** `2601.18938` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Xin Qiao, Shijie Sun, Anqi Dong, et al.
+- **abs:** https://arxiv.org/abs/2601.18938 · **PDF:** https://arxiv.org/pdf/2601.18938v1
+
+## 19087. Bi-Level Online Provisioning and Scheduling with Switching Costs and Cross-Level Constraints
+
+- **ID:** `2601.18936` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Jialei Liu, C. Emre Koksal, Ming Shi
+- **abs:** https://arxiv.org/abs/2601.18936 · **PDF:** https://arxiv.org/pdf/2601.18936v2
+
+## 19088. BabyReasoningBench: Generating Developmentally-Inspired Reasoning Tasks for Evaluating Baby Language Models
+
+- **ID:** `2601.18933` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Kaustubh D. Dhole
+- **abs:** https://arxiv.org/abs/2601.18933 · **PDF:** https://arxiv.org/pdf/2601.18933v2
+
+## 19089. Advances in Diffusion-Based Generative Compression
+
+- **ID:** `2601.18932` · **primary:** `eess.IV` · **Submitted:** 2026-01-26
+- **Authors:** Yibo Yang, Stephan Mandt
+- **abs:** https://arxiv.org/abs/2601.18932 · **PDF:** https://arxiv.org/pdf/2601.18932v1
+
+## 19090. Toward Learning POMDPs Beyond Full-Rank Actions and State Observability
+
+- **ID:** `2601.18930` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Seiji Shaw, Travis Manderson, Chad Kessens, et al.
+- **abs:** https://arxiv.org/abs/2601.18930 · **PDF:** https://arxiv.org/pdf/2601.18930v3
+
+## 19091. RIFT: Reordered Instruction Following Testbed To Evaluate Instruction Following in Singular Multistep Prompt Structures
+
+- **ID:** `2601.18924` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Andrew Jaffe, Noah Reicin, Jinho D. Choi
+- **abs:** https://arxiv.org/abs/2601.18924 · **PDF:** https://arxiv.org/pdf/2601.18924v1
+
+## 19092. Accelerating Large-Scale Cheminformatics Using a Byte-Offset Indexing Architecture for Terabyte-Scale Data Integration
+
+- **ID:** `2601.18921` · **primary:** `cs.DB` · **Submitted:** 2026-01-26
+- **Authors:** Malikussaid, Septian Caesar Floresko, Sutiyo
+- **abs:** https://arxiv.org/abs/2601.18921 · **PDF:** https://arxiv.org/pdf/2601.18921v2
+
+## 19093. One Global Model, Many Behaviors: Stockout-Aware Feature Engineering and Dynamic Scaling for Multi-Horizon Retail Demand Forecasting with a Cost-Aware Ordering Policy (VN2 Winner Report)
+
+- **ID:** `2601.18919` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Bartosz Szabłowski
+- **abs:** https://arxiv.org/abs/2601.18919 · **PDF:** https://arxiv.org/pdf/2601.18919v1
+
+## 19094. GraIP: A Benchmarking Framework For Neural Graph Inverse Problems
+
+- **ID:** `2601.18917` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Semih Cantürk, Andrei Manolache, Arman Mielke, et al.
+- **abs:** https://arxiv.org/abs/2601.18917 · **PDF:** https://arxiv.org/pdf/2601.18917v1
+
+## 19095. ASEHybrid: When Geometry Matters Beyond Homophily in Graph Neural Networks
+
+- **ID:** `2601.18912` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Shalima Binta Manir, Tim Oates
+- **abs:** https://arxiv.org/abs/2601.18912 · **PDF:** https://arxiv.org/pdf/2601.18912v1
+
+## 19096. How Is Uncertainty Propagated in Knowledge Distillation?
+
+- **ID:** `2601.18909` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Ziyao Cui, Jian Pei
+- **abs:** https://arxiv.org/abs/2601.18909 · **PDF:** https://arxiv.org/pdf/2601.18909v1
+
+## 19097. AI as Teammate or Tool? A Review of Human-AI Interaction in Decision Support
+
+- **ID:** `2602.15865` · **primary:** `cs.HC` · **Submitted:** 2026-01-26
+- **Authors:** Most. Sharmin Sultana Samu, Nafisa Khan, Kazi Toufique Elahi, et al.
+- **abs:** https://arxiv.org/abs/2602.15865 · **PDF:** https://arxiv.org/pdf/2602.15865v1
+
+## 19098. Enhancing Speech Emotion Recognition using Dynamic Spectral Features and Kalman Smoothing
+
+- **ID:** `2601.18908` · **primary:** `cs.SD` · **Submitted:** 2026-01-26
+- **Authors:** Marouane El Hizabri, Abdelfattah Bezzaz, Ismail Hayoukane, et al.
+- **abs:** https://arxiv.org/abs/2601.18908 · **PDF:** https://arxiv.org/pdf/2601.18908v1
+
+## 19099. Implicit Q-Learning and SARSA: Liberating Policy Control from Step-Size Calibration
+
+- **ID:** `2601.18907` · **primary:** `stat.ML` · **Submitted:** 2026-01-26
+- **Authors:** Hwanwoo Kim, Eric Laber
+- **abs:** https://arxiv.org/abs/2601.18907 · **PDF:** https://arxiv.org/pdf/2601.18907v1
+
+## 19100. SICL-AT: Another way to adapt Auditory LLM to low-resource task
+
+- **ID:** `2601.18904` · **primary:** `cs.SD` · **Submitted:** 2026-01-26
+- **Authors:** Haolong Zheng, Siyin Wang, Zengrui Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.18904 · **PDF:** https://arxiv.org/pdf/2601.18904v1
+
+## 19101. Flatter Tokens are More Valuable for Speculative Draft Model Training
+
+- **ID:** `2601.18902` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Jiaming Fan, Daming Cao, Xiangzhong Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.18902 · **PDF:** https://arxiv.org/pdf/2601.18902v2
+
+## 19102. Self-Aware Knowledge Probing: Evaluating Language Models' Relational Knowledge through Confidence Calibration
+
+- **ID:** `2601.18901` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Christopher Kissling, Elena Merdjanovska, Alan Akbik
+- **abs:** https://arxiv.org/abs/2601.18901 · **PDF:** https://arxiv.org/pdf/2601.18901v1
+
+## 19103. RealStats: A Rigorous Real-Only Statistical Framework for Fake Image Detection
+
+- **ID:** `2601.18900` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Haim Zisman, Uri Shaham
+- **abs:** https://arxiv.org/abs/2601.18900 · **PDF:** https://arxiv.org/pdf/2601.18900v1
+
+## 19104. Language Family Matters: Evaluating LLM-Based ASR Across Linguistic Boundaries
+
+- **ID:** `2601.18899` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Yuchen Zhang, Ravi Shekhar, Haralambos Mouratidis
+- **abs:** https://arxiv.org/abs/2601.18899 · **PDF:** https://arxiv.org/pdf/2601.18899v2
+
+## 19105. Explainable Uncertainty Quantification for Wastewater Treatment Energy Prediction via Interval Type-2 Neuro-Fuzzy System
+
+- **ID:** `2601.18897` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Qusai Khaled, Bahjat Mallak, Uzay Kaymak, et al.
+- **abs:** https://arxiv.org/abs/2601.18897 · **PDF:** https://arxiv.org/pdf/2601.18897v1
+
+## 19106. XProvence: Zero-Cost Multilingual Context Pruning for Retrieval-Augmented Generation
+
+- **ID:** `2601.18886` · **primary:** `cs.IR` · **Submitted:** 2026-01-26
+- **Authors:** Youssef Mohamed, Mohamed Elhoseiny, Thibault Formal, et al.
+- **abs:** https://arxiv.org/abs/2601.18886 · **PDF:** https://arxiv.org/pdf/2601.18886v1
+
+## 19107. ctELM: Decoding and Manipulating Embeddings of Clinical Trials with Embedding Language Models
+
+- **ID:** `2601.18796` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Brian Ondov, Chia-Hsuan Chang, Yujia Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.18796 · **PDF:** https://arxiv.org/pdf/2601.18796v1
+
+## 19108. Reuse your FLOPs: Scaling RL on Hard Problems by Conditioning on Very Off-Policy Prefixes
+
+- **ID:** `2601.18795` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Amrith Setlur, Zijian Wang, Andrew Cohen, et al.
+- **abs:** https://arxiv.org/abs/2601.18795 · **PDF:** https://arxiv.org/pdf/2601.18795v2
+
+## 19109. MEGnifying Emotion: Sentiment Analysis from Annotated Brain Data
+
+- **ID:** `2601.18792` · **primary:** `cs.HC` · **Submitted:** 2026-01-26
+- **Authors:** Brian Liu, Oiwi Parker Jones
+- **abs:** https://arxiv.org/abs/2601.18792 · **PDF:** https://arxiv.org/pdf/2601.18792v1
+
+## 19110. Subword-Based Comparative Linguistics across 242 Languages Using Wikipedia Glottosets
+
+- **ID:** `2601.18791` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Iaroslav Chelombitko, Mika Hämäläinen, Aleksey Komissarov
+- **abs:** https://arxiv.org/abs/2601.18791 · **PDF:** https://arxiv.org/pdf/2601.18791v1
+
+## 19111. MortalMATH: Evaluating the Conflict Between Reasoning Objectives and Emergency Contexts
+
+- **ID:** `2601.18790` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Etienne Lanzeray, Stephane Meilliez, Malo Ruelle, et al.
+- **abs:** https://arxiv.org/abs/2601.18790 · **PDF:** https://arxiv.org/pdf/2601.18790v1
+
+## 19112. Unsupervised Text Segmentation via Kernel Change-Point Detection on Sentence Embeddings
+
+- **ID:** `2601.18788` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Mumin Jia, Jairo Diaz-Rodriguez
+- **abs:** https://arxiv.org/abs/2601.18788 · **PDF:** https://arxiv.org/pdf/2601.18788v1
+
+## 19113. Design Techniques for LLM-Powered Interactive Storytelling: A Case Study of the Dramamancer System
+
+- **ID:** `2601.18785` · **primary:** `cs.HC` · **Submitted:** 2026-01-26
+- **Authors:** Tiffany Wang, Yuqian Sun, Yi Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.18785 · **PDF:** https://arxiv.org/pdf/2601.18785v1
+
+## 19114. Multi-Objective Reinforcement Learning for Efficient Tactical Decision Making for Trucks in Highway Traffic
+
+- **ID:** `2601.18783` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Deepthi Pathare, Leo Laine, Morteza Haghir Chehreghani
+- **abs:** https://arxiv.org/abs/2601.18783 · **PDF:** https://arxiv.org/pdf/2601.18783v1
+
+## 19115. POPE: Learning to Reason on Hard Problems via Privileged On-Policy Exploration
+
+- **ID:** `2601.18779` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Yuxiao Qu, Amrith Setlur, Virginia Smith, et al.
+- **abs:** https://arxiv.org/abs/2601.18779 · **PDF:** https://arxiv.org/pdf/2601.18779v1
+
+## 19116. Teaching Models to Teach Themselves: Reasoning at the Edge of Learnability
+
+- **ID:** `2601.18778` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Shobhita Sundaram, John Quan, Ariel Kwiatkowski, et al.
+- **abs:** https://arxiv.org/abs/2601.18778 · **PDF:** https://arxiv.org/pdf/2601.18778v2
+
+## 19117. PRECISE: Reducing the Bias of LLM Evaluations Using Prediction-Powered Ranking Estimation
+
+- **ID:** `2601.18777` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Abhishek Divekar, Anirban Majumder
+- **abs:** https://arxiv.org/abs/2601.18777 · **PDF:** https://arxiv.org/pdf/2601.18777v1
+
+## 19118. Representational Homomorphism Predicts and Improves Compositional Generalization In Transformer Language Model
+
+- **ID:** `2601.18858` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Zhiyu An, Wan Du
+- **abs:** https://arxiv.org/abs/2601.18858 · **PDF:** https://arxiv.org/pdf/2601.18858v2
+
+## 19119. Dep-Search: Learning Dependency-Aware Reasoning Traces with Persistent Memory
+
+- **ID:** `2601.18771` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Yanming Liu, Xinyue Peng, Zixuan Yan, et al.
+- **abs:** https://arxiv.org/abs/2601.18771 · **PDF:** https://arxiv.org/pdf/2601.18771v1
+
+## 19120. Learning to Discover: A Generalized Framework for Raga Identification without Forgetting
+
+- **ID:** `2601.18766` · **primary:** `eess.AS` · **Submitted:** 2026-01-26
+- **Authors:** Parampreet Singh, Somya Kumar, Chaitanya Shailendra Nitawe, et al.
+- **abs:** https://arxiv.org/abs/2601.18766 · **PDF:** https://arxiv.org/pdf/2601.18766v1
+
+## 19121. Beyond Preferences: Learning Alignment Principles Grounded in Human Reasons and Values
+
+- **ID:** `2601.18760` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Henry Bell, Lara Neubauer da Costa Schertel, Bochu Ding, et al.
+- **abs:** https://arxiv.org/abs/2601.18760 · **PDF:** https://arxiv.org/pdf/2601.18760v1
+
+## 19122. $α^3$-SecBench: A Large-Scale Evaluation Suite of Security, Resilience, and Trust for LLM-based UAV Agents over 6G Networks
+
+- **ID:** `2601.18754` · **primary:** `cs.CR` · **Submitted:** 2026-01-26
+- **Authors:** Mohamed Amine Ferrag, Abderrahmane Lakas, Merouane Debbah
+- **abs:** https://arxiv.org/abs/2601.18754 · **PDF:** https://arxiv.org/pdf/2601.18754v1
+
+## 19123. HalluGuard: Demystifying Data-Driven and Reasoning-Driven Hallucinations in LLMs
+
+- **ID:** `2601.18753` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Xinyue Zeng, Junhong Lin, Yujun Yan, et al.
+- **abs:** https://arxiv.org/abs/2601.18753 · **PDF:** https://arxiv.org/pdf/2601.18753v2
+
+## 19124. Trust, Don't Trust, or Flip: Robust Preference-Based Reinforcement Learning with Multi-Expert Feedback
+
+- **ID:** `2601.18751` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Seyed Amir Hosseini, Maryam Abdolali, Amirhosein Tavakkoli, et al.
+- **abs:** https://arxiv.org/abs/2601.18751 · **PDF:** https://arxiv.org/pdf/2601.18751v1
+
+## 19125. Capturing P: On the Expressive Power and Efficient Evaluation of Boolean Retrieval
+
+- **ID:** `2601.18747` · **primary:** `cs.IR` · **Submitted:** 2026-01-26
+- **Authors:** Amir Aavani
+- **abs:** https://arxiv.org/abs/2601.18747 · **PDF:** https://arxiv.org/pdf/2601.18747v1
+
+## 19126. TSRBench: A Comprehensive Multi-task Multi-modal Time Series Reasoning Benchmark for Generalist Models
+
+- **ID:** `2601.18744` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Fangxu Yu, Xingang Guo, Lingzhi Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.18744 · **PDF:** https://arxiv.org/pdf/2601.18744v1
+
+## 19127. SeNeDiF-OOD: Semantic Nested Dichotomy Fusion for Out-of-Distribution Detection Methodology in Open-World Classification. A Case Study on Monument Style Classification
+
+- **ID:** `2601.18739` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Ignacio Antequera-Sánchez, Juan Luis Suárez-Díaz, Rosana Montes, et al.
+- **abs:** https://arxiv.org/abs/2601.18739 · **PDF:** https://arxiv.org/pdf/2601.18739v3
+
+## 19128. Benchmarking Machine Learning Models for IoT Malware Detection under Data Scarcity and Drift
+
+- **ID:** `2601.18736` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Jake Lyon, Ehsan Saeedizade, Shamik Sengupta
+- **abs:** https://arxiv.org/abs/2601.18736 · **PDF:** https://arxiv.org/pdf/2601.18736v1
+
+## 19129. Why Keep Your Doubts to Yourself? Trading Visual Uncertainties in Multi-Agent Bandit Systems
+
+- **ID:** `2601.18735` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Jusheng Zhang, Yijia Fan, Kaitong Cai, et al.
+- **abs:** https://arxiv.org/abs/2601.18735 · **PDF:** https://arxiv.org/pdf/2601.18735v1
+
+## 19130. Self-Distilled Reasoner: On-Policy Self-Distillation for Large Language Models
+
+- **ID:** `2601.18734` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Siyan Zhao, Zhihui Xie, Mengchen Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.18734 · **PDF:** https://arxiv.org/pdf/2601.18734v3
+
+## 19131. Advances and Innovations in the Multi-Agent Robotic System (MARS) Challenge
+
+- **ID:** `2601.18733` · **primary:** `cs.RO` · **Submitted:** 2026-01-26
+- **Authors:** Li Kang, Heng Zhou, Xiufeng Song, et al.
+- **abs:** https://arxiv.org/abs/2601.18733 · **PDF:** https://arxiv.org/pdf/2601.18733v1
+
+## 19132. Optimal Use of Preferences in Artificial Intelligence Algorithms
+
+- **ID:** `2601.18732` · **primary:** `econ.TH` · **Submitted:** 2026-01-26
+- **Authors:** Joshua S. Gans
+- **abs:** https://arxiv.org/abs/2601.18732 · **PDF:** https://arxiv.org/pdf/2601.18732v1
+
+## 19133. One Adapts to Any: Meta Reward Modeling for Personalized LLM Alignment
+
+- **ID:** `2601.18731` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Hongru Cai, Yongqi Li, Tiezheng Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.18731 · **PDF:** https://arxiv.org/pdf/2601.18731v1
+
+## 19134. Reflect: Transparent Principle-Guided Reasoning for Constitutional Alignment at Scale
+
+- **ID:** `2601.18730` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Henry Bell, Caroline Zhang, Mohammed Mobasserul Haque, et al.
+- **abs:** https://arxiv.org/abs/2601.18730 · **PDF:** https://arxiv.org/pdf/2601.18730v1
+
+## 19135. Riemannian AmbientFlow: Towards Simultaneous Manifold Learning and Generative Modeling from Corrupted Data
+
+- **ID:** `2601.18728` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Willem Diepeveen, Oscar Leong
+- **abs:** https://arxiv.org/abs/2601.18728 · **PDF:** https://arxiv.org/pdf/2601.18728v1
+
+## 19136. Statistical Inference for Explainable Boosting Machines
+
+- **ID:** `2601.18857` · **primary:** `stat.ML` · **Submitted:** 2026-01-26
+- **Authors:** Haimo Fang, Kevin Tan, Jonathan Pipping-Gamon, et al.
+- **abs:** https://arxiv.org/abs/2601.18857 · **PDF:** https://arxiv.org/pdf/2601.18857v2
+
+## 19137. HalluCitation Matters: Revealing the Impact of Hallucinated References with 300 Hallucinated Papers in ACL Conferences
+
+- **ID:** `2601.18724` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe
+- **abs:** https://arxiv.org/abs/2601.18724 · **PDF:** https://arxiv.org/pdf/2601.18724v1
+
+## 19138. Gained in Translation: Privileged Pairwise Judges Enhance Multilingual Reasoning
+
+- **ID:** `2601.18722` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Lintang Sutawika, Gokul Swamy, Zhiwei Steven Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.18722 · **PDF:** https://arxiv.org/pdf/2601.18722v1
+
+## 19139. Conditioned Generative Modeling of Molecular Glues: A Realistic AI Approach for Synthesizable Drug-like Molecules
+
+- **ID:** `2601.18716` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Naeyma N. Islam, Thomas R. Caulfield
+- **abs:** https://arxiv.org/abs/2601.18716 · **PDF:** https://arxiv.org/pdf/2601.18716v1
+
+## 19140. Low Cost, High Efficiency: LiDAR Place Recognition in Vineyards with Matryoshka Representation Learning
+
+- **ID:** `2601.18714` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Judith Vilella-Cantos, Mauro Martini, Marcello Chiaberge, et al.
+- **abs:** https://arxiv.org/abs/2601.18714 · **PDF:** https://arxiv.org/pdf/2601.18714v1
+
+## 19141. Point transformer for protein structural heterogeneity analysis using CryoEM
+
+- **ID:** `2601.18713` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-26
+- **Authors:** Muyuan Chen, Muchen Li, Renjie Liao
+- **abs:** https://arxiv.org/abs/2601.18713 · **PDF:** https://arxiv.org/pdf/2601.18713v1
+
+## 19142. Analyzing Images of Blood Cells with Quantum Machine Learning Methods: Equilibrium Propagation and Variational Quantum Circuits to Detect Acute Myeloid Leukemia
+
+- **ID:** `2601.18710` · **primary:** `cs.ET` · **Submitted:** 2026-01-26
+- **Authors:** A. Bano, L. Liebovitch
+- **abs:** https://arxiv.org/abs/2601.18710 · **PDF:** https://arxiv.org/pdf/2601.18710v1
+
+## 19143. SMART: Scalable Mesh-free Aerodynamic Simulations from Raw Geometries using a Transformer-based Surrogate Model
+
+- **ID:** `2601.18707` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Jan Hagnberger, Mathias Niepert
+- **abs:** https://arxiv.org/abs/2601.18707 · **PDF:** https://arxiv.org/pdf/2601.18707v1
+
+## 19144. Health-SCORE: Towards Scalable Rubrics for Improving Health-LLMs
+
+- **ID:** `2601.18706` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Zhichao Yang, Sepehr Janghorbani, Dongxu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.18706 · **PDF:** https://arxiv.org/pdf/2601.18706v1
+
+## 19145. Data-Driven Qubit Characterization and Optimal Control using Deep Learning
+
+- **ID:** `2601.18704` · **primary:** `quant-ph` · **Submitted:** 2026-01-26
+- **Authors:** Paul Surrey, Julian D. Teske, Tobias Hangleiter, et al.
+- **abs:** https://arxiv.org/abs/2601.18704 · **PDF:** https://arxiv.org/pdf/2601.18704v1
+
+## 19146. What is Safety? Corporate Discourse, Power, and the Politics of Generative AI Safety
+
+- **ID:** `2602.06981` · **primary:** `cs.CY` · **Submitted:** 2026-01-26
+- **Authors:** Ankolika De, Gabriel Lima, Yixin Zou
+- **abs:** https://arxiv.org/abs/2602.06981 · **PDF:** https://arxiv.org/pdf/2602.06981v1
+
+## 19147. From Fuzzy to Exact: The Halo Architecture for Infinite-Depth Reasoning via Rational Arithmetic
+
+- **ID:** `2601.18702` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Hansheng Ren
+- **abs:** https://arxiv.org/abs/2601.18702 · **PDF:** https://arxiv.org/pdf/2601.18702v4
+
+## 19148. TEA-Bench: A Systematic Benchmarking of Tool-enhanced Emotional Support Dialogue Agent
+
+- **ID:** `2601.18700` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Xingyu Sui, Yanyan Zhao, Yulin Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.18700 · **PDF:** https://arxiv.org/pdf/2601.18700v1
+
+## 19149. Mechanistic Analysis of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning
+
+- **ID:** `2601.18699` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Olaf Yunus Laitinen Imanov
+- **abs:** https://arxiv.org/abs/2601.18699 · **PDF:** https://arxiv.org/pdf/2601.18699v1
+
+## 19150. Explainability Methods for Hardware Trojan Detection: A Systematic Comparison
+
+- **ID:** `2601.18696` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Paul Whitten, Francis Wolff, Chris Papachristou
+- **abs:** https://arxiv.org/abs/2601.18696 · **PDF:** https://arxiv.org/pdf/2601.18696v3
+
+## 19151. Neural Multi-Speaker Voice Cloning for Nepali in Low-Resource Settings
+
+- **ID:** `2601.18694` · **primary:** `cs.SD` · **Submitted:** 2026-01-26
+- **Authors:** Aayush M. Shrestha, Aditya Bajracharya, Projan Shakya, et al.
+- **abs:** https://arxiv.org/abs/2601.18694 · **PDF:** https://arxiv.org/pdf/2601.18694v1
+
+## 19152. LLAMA LIMA: A Living Meta-Analysis on the Effects of Generative AI on Learning Mathematics
+
+- **ID:** `2601.18685` · **primary:** `math.HO` · **Submitted:** 2026-01-26
+- **Authors:** Anselm Strohmaier, Samira Bödefeld, Oliver Straser, et al.
+- **abs:** https://arxiv.org/abs/2601.18685 · **PDF:** https://arxiv.org/pdf/2601.18685v2
+
+## 19153. Learned harmonic mean estimation of the marginal likelihood for multimodal posteriors with flow matching
+
+- **ID:** `2601.18683` · **primary:** `stat.ME` · **Submitted:** 2026-01-26
+- **Authors:** Alicja Polanska, Jason D. McEwen
+- **abs:** https://arxiv.org/abs/2601.18683 · **PDF:** https://arxiv.org/pdf/2601.18683v1
+
+## 19154. ART for Diffusion Sampling: A Reinforcement Learning Approach to Timestep Schedule
+
+- **ID:** `2601.18681` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Yilie Huang, Wenpin Tang, Xunyu Zhou
+- **abs:** https://arxiv.org/abs/2601.18681 · **PDF:** https://arxiv.org/pdf/2601.18681v1
+
+## 19155. Counterfactual Explanations on Robust Perceptual Geodesics
+
+- **ID:** `2601.18678` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Eslam Zaher, Maciej Trzaskowski, Quan Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.18678 · **PDF:** https://arxiv.org/pdf/2601.18678v2
+
+## 19156. Out-of-Distribution Radar Detection with Complex VAEs: Theory, Whitening, and ANMF Fusion
+
+- **ID:** `2601.18677` · **primary:** `stat.ML` · **Submitted:** 2026-01-26
+- **Authors:** Yadang Alexis Rouzoumka, Jean Pinsolle, Eugénie Terreaux, et al.
+- **abs:** https://arxiv.org/abs/2601.18677 · **PDF:** https://arxiv.org/pdf/2601.18677v1
+
+## 19157. Quasi Monte Carlo methods enable extremely low-dimensional deep generative models
+
+- **ID:** `2601.18676` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Miles Martinez, Alex H. Williams
+- **abs:** https://arxiv.org/abs/2601.18676 · **PDF:** https://arxiv.org/pdf/2601.18676v1
+
+## 19158. Learning temporal embeddings from electronic health records of chronic kidney disease patients
+
+- **ID:** `2601.18675` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Aditya Kumar, Mario A. Cypko, Oliver Amft
+- **abs:** https://arxiv.org/abs/2601.18675 · **PDF:** https://arxiv.org/pdf/2601.18675v1
+
+## 19159. A Dynamic Framework for Grid Adaptation in Kolmogorov-Arnold Networks
+
+- **ID:** `2601.18672` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Spyros Rigas, Thanasis Papaioannou, Panagiotis Trakadas, et al.
+- **abs:** https://arxiv.org/abs/2601.18672 · **PDF:** https://arxiv.org/pdf/2601.18672v2
+
+## 19160. Uniform Computability of PAC Learning
+
+- **ID:** `2601.18663` · **primary:** `math.LO` · **Submitted:** 2026-01-26
+- **Authors:** Vasco Brattka, Guillaume Chirache
+- **abs:** https://arxiv.org/abs/2601.18663 · **PDF:** https://arxiv.org/pdf/2601.18663v1
+
+## 19161. Contrasting Global and Patient-Specific Regression Models via a Neural Network Representation
+
+- **ID:** `2601.18658` · **primary:** `stat.ME` · **Submitted:** 2026-01-26
+- **Authors:** Max Behrens, Daiana Stolz, Eleni Papakonstantinou, et al.
+- **abs:** https://arxiv.org/abs/2601.18658 · **PDF:** https://arxiv.org/pdf/2601.18658v1
+
+## 19162. FaLW: A Forgetting-aware Loss Reweighting for Long-tailed Unlearning
+
+- **ID:** `2601.18650` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Liheng Yu, Zhe Zhao, Yuxuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.18650 · **PDF:** https://arxiv.org/pdf/2601.18650v2
+
+## 19163. SITUATE -- Synthetic Object Counting Dataset for VLM training
+
+- **ID:** `2602.00108` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** René Peinl, Vincent Tischler, Patrick Schröder, et al.
+- **abs:** https://arxiv.org/abs/2602.00108 · **PDF:** https://arxiv.org/pdf/2602.00108v1
+
+## 19164. FadeMem: Biologically-Inspired Forgetting for Efficient Agent Memory
+
+- **ID:** `2601.18642` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Lei Wei, Xiao Peng, Xu Dong, et al.
+- **abs:** https://arxiv.org/abs/2601.18642 · **PDF:** https://arxiv.org/pdf/2601.18642v2
+
+## 19165. Unheard in the Digital Age: Rethinking AI Bias and Speech Diversity
+
+- **ID:** `2601.18641` · **primary:** `cs.HC` · **Submitted:** 2026-01-26
+- **Authors:** Onyedikachi Hope Amaechi-Okorie, Branislav Radeljic
+- **abs:** https://arxiv.org/abs/2601.18641 · **PDF:** https://arxiv.org/pdf/2601.18641v2
+
+## 19166. TwinPurify: Purifying gene expression data to reveal tumor-intrinsic transcriptional programs via self-supervised learning
+
+- **ID:** `2601.18640` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Zhiwei Zheng, Kevin Bryson
+- **abs:** https://arxiv.org/abs/2601.18640 · **PDF:** https://arxiv.org/pdf/2601.18640v2
+
+## 19167. Physics-Informed Uncertainty Enables Reliable AI-driven Design
+
+- **ID:** `2601.18638` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Tingkai Xue, Chin Chun Ooi, Yang Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.18638 · **PDF:** https://arxiv.org/pdf/2601.18638v1
+
+## 19168. Universality of Many-body Projected Ensemble for Learning Quantum Data Distribution
+
+- **ID:** `2601.18637` · **primary:** `quant-ph` · **Submitted:** 2026-01-26
+- **Authors:** Quoc Hoan Tran, Koki Chinzei, Yasuhiro Endo, et al.
+- **abs:** https://arxiv.org/abs/2601.18637 · **PDF:** https://arxiv.org/pdf/2601.18637v2
+
+## 19169. Computability of Agentic Systems
+
+- **ID:** `2602.13222` · **primary:** `cs.CC` · **Submitted:** 2026-01-26
+- **Authors:** Chatavut Viriyasuthee
+- **abs:** https://arxiv.org/abs/2602.13222 · **PDF:** https://arxiv.org/pdf/2602.13222v1
+
+## 19170. AdaReasoner: Dynamic Tool Orchestration for Iterative Visual Reasoning
+
+- **ID:** `2601.18631` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Mingyang Song, Haoyu Sun, Jiawei Gu, et al.
+- **abs:** https://arxiv.org/abs/2601.18631 · **PDF:** https://arxiv.org/pdf/2601.18631v2
+
+## 19171. Assessing the Quality of Mental Health Support in LLM Responses through Multi-Attribute Human Evaluation
+
+- **ID:** `2601.18630` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Abeer Badawi, Md Tahmid Rahman Laskar, Elahe Rahimi, et al.
+- **abs:** https://arxiv.org/abs/2601.18630 · **PDF:** https://arxiv.org/pdf/2601.18630v1
+
+## 19172. Rank-1 Approximation of Inverse Fisher for Natural Policy Gradients in Deep Reinforcement Learning
+
+- **ID:** `2601.18626` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Yingxiao Huo, Satya Prakash Dash, Radu Stoican, et al.
+- **abs:** https://arxiv.org/abs/2601.18626 · **PDF:** https://arxiv.org/pdf/2601.18626v3
+
+## 19173. CASSANDRA: Programmatic and Probabilistic Learning and Inference for Stochastic World Modeling
+
+- **ID:** `2601.18620` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Panagiotis Lymperopoulos, Abhiramon Rajasekharan, Ian Berlot-Attwell, et al.
+- **abs:** https://arxiv.org/abs/2601.18620 · **PDF:** https://arxiv.org/pdf/2601.18620v1
+
+## 19174. Emergence of Phonemic, Syntactic, and Semantic Representations in Artificial Neural Networks
+
+- **ID:** `2601.18617` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Pierre Orhan, Pablo Diego-Simón, Emmnanuel Chemla, et al.
+- **abs:** https://arxiv.org/abs/2601.18617 · **PDF:** https://arxiv.org/pdf/2601.18617v1
+
+## 19175. Geometry-Free Conditional Diffusion Modeling for Solving the Inverse Electrocardiography Problem
+
+- **ID:** `2601.18615` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Ramiro Valdes Jara, Adam Meyers
+- **abs:** https://arxiv.org/abs/2601.18615 · **PDF:** https://arxiv.org/pdf/2601.18615v1
+
+## 19176. PolySHAP: Extending KernelSHAP with Interaction-Informed Polynomial Regression
+
+- **ID:** `2601.18608` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Fabian Fumagalli, R. Teal Witter, Christopher Musco
+- **abs:** https://arxiv.org/abs/2601.18608 · **PDF:** https://arxiv.org/pdf/2601.18608v2
+
+## 19177. LaCoGSEA: Unsupervised deep learning for pathway analysis via latent correlation
+
+- **ID:** `2601.18604` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Zhiwei Zheng, Kevin Bryson
+- **abs:** https://arxiv.org/abs/2601.18604 · **PDF:** https://arxiv.org/pdf/2601.18604v2
+
+## 19178. LTS-VoiceAgent: A Listen-Think-Speak Framework for Efficient Streaming Voice Interaction via Semantic Triggering and Incremental Reasoning
+
+- **ID:** `2601.19952` · **primary:** `cs.SD` · **Submitted:** 2026-01-26
+- **Authors:** Wenhao Zou, Yuwei Miao, Zhanyu Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.19952 · **PDF:** https://arxiv.org/pdf/2601.19952v1
+
+## 19179. A Balanced Neuro-Symbolic Approach for Commonsense Abductive Logic
+
+- **ID:** `2601.18595` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Joseph Cotnareanu, Didier Chetelat, Yingxue Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.18595 · **PDF:** https://arxiv.org/pdf/2601.18595v1
+
+## 19180. Global Optimization of Atomic Clusters via Physically-Constrained Tensor Train Decomposition
+
+- **ID:** `2601.18592` · **primary:** `math.OC` · **Submitted:** 2026-01-26
+- **Authors:** Konstantin Sozykin, Nikita Rybin, Andrei Chertkov, et al.
+- **abs:** https://arxiv.org/abs/2601.18592 · **PDF:** https://arxiv.org/pdf/2601.18592v1
+
+## 19181. Stability as a Liability:Systematic Breakdown of Linguistic Structure in LLMs
+
+- **ID:** `2601.18588` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Xianzhe Meng, Qiangsheng Zeng, Ling Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.18588 · **PDF:** https://arxiv.org/pdf/2601.18588v2
+
+## 19182. Learning long term climate-resilient transport adaptation pathways under direct and indirect flood impacts using reinforcement learning
+
+- **ID:** `2601.18586` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Miguel Costa, Arthur Vandervoort, Carolin Schmidt, et al.
+- **abs:** https://arxiv.org/abs/2601.18586 · **PDF:** https://arxiv.org/pdf/2601.18586v1
+
+## 19183. From Classification to Ranking: Enhancing LLM Reasoning Capabilities for MBTI Personality Detection
+
+- **ID:** `2601.18582` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Yuan Cao, Feixiang Liu, Xinyue Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.18582 · **PDF:** https://arxiv.org/pdf/2601.18582v1
+
+## 19184. K-Myriad: Jump-starting reinforcement learning with unsupervised parallel agents
+
+- **ID:** `2601.18580` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Vincenzo De Paola, Mirco Mutti, Riccardo Zamboni, et al.
+- **abs:** https://arxiv.org/abs/2601.18580 · **PDF:** https://arxiv.org/pdf/2601.18580v1
+
+## 19185. FastInsight: Fast and Insightful Retrieval via Fusion Operators for Graph RAG
+
+- **ID:** `2601.18579` · **primary:** `cs.IR` · **Submitted:** 2026-01-26
+- **Authors:** Seonho An, Chaejeong Hyun, Min-Soo Kim
+- **abs:** https://arxiv.org/abs/2601.18579 · **PDF:** https://arxiv.org/pdf/2601.18579v1
+
+## 19186. Self-Refining Video Sampling
+
+- **ID:** `2601.18577` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Sangwon Jang, Taekyung Ki, Jaehyeong Jo, et al.
+- **abs:** https://arxiv.org/abs/2601.18577 · **PDF:** https://arxiv.org/pdf/2601.18577v1
+
+## 19187. One Persona, Many Cues, Different Results: How Sociodemographic Cues Impact LLM Personalization
+
+- **ID:** `2601.18572` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Franziska Weeber, Vera Neplenbroek, Jan Batzner, et al.
+- **abs:** https://arxiv.org/abs/2601.18572 · **PDF:** https://arxiv.org/pdf/2601.18572v1
+
+## 19188. Attention-Based Neural-Augmented Kalman Filter for Legged Robot State Estimation
+
+- **ID:** `2601.18569` · **primary:** `cs.RO` · **Submitted:** 2026-01-26
+- **Authors:** Seokju Lee, Kyung-Soo Kim
+- **abs:** https://arxiv.org/abs/2601.18569 · **PDF:** https://arxiv.org/pdf/2601.18569v1
+
+## 19189. An Unsupervised Tensor-Based Domain Alignment
+
+- **ID:** `2601.18564` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Chong Hyun Lee, Kibae Lee, Hyun Hee Yim
+- **abs:** https://arxiv.org/abs/2601.18564 · **PDF:** https://arxiv.org/pdf/2601.18564v1
+
+## 19190. Generative Diffusion Augmentation with Quantum-Enhanced Discrimination for Medical Image Diagnosis
+
+- **ID:** `2601.18556` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Jingsong Xia, Siqi Wang
+- **abs:** https://arxiv.org/abs/2601.18556 · **PDF:** https://arxiv.org/pdf/2601.18556v1
+
+## 19191. Deconstructing Instruction-Following: A New Benchmark for Granular Evaluation of Large Language Model Instruction Compliance Abilities
+
+- **ID:** `2601.18554` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Alberto Purpura, Li Wang, Sahil Badyal, et al.
+- **abs:** https://arxiv.org/abs/2601.18554 · **PDF:** https://arxiv.org/pdf/2601.18554v1
+
+## 19192. Unknown Unknowns: Why Hidden Intentions in LLMs Evade Detection
+
+- **ID:** `2601.18552` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Devansh Srivastav, David Pape, Lea Schönherr
+- **abs:** https://arxiv.org/abs/2601.18552 · **PDF:** https://arxiv.org/pdf/2601.18552v1
+
+## 19193. Information Hidden in Gradients of Regression with Target Noise
+
+- **ID:** `2601.18546` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Arash Jamshidi, Katsiaryna Haitsiukevich, Kai Puolamäki
+- **abs:** https://arxiv.org/abs/2601.18546 · **PDF:** https://arxiv.org/pdf/2601.18546v1
+
+## 19194. SKETCH: Semantic Key-Point Conditioning for Long-Horizon Vessel Trajectory Prediction
+
+- **ID:** `2601.18537` · **primary:** `cs.RO` · **Submitted:** 2026-01-26
+- **Authors:** Linyong Gan, Zimo Li, Wenxin Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.18537 · **PDF:** https://arxiv.org/pdf/2601.18537v2
+
+## 19195. Evaluating Morphological Plausibility of Subword Tokenization via Statistical Alignment with Morpho-Syntactic Features
+
+- **ID:** `2601.18536` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Abishek Stephen, Jindřich Libovický
+- **abs:** https://arxiv.org/abs/2601.18536 · **PDF:** https://arxiv.org/pdf/2601.18536v1
+
+## 19196. From Verifiable Dot to Reward Chain: Harnessing Verifiable Reference-based Rewards for Reinforcement Learning of Open-ended Generation
+
+- **ID:** `2601.18533` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Yuxin Jiang, Yufei Wang, Qiyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.18533 · **PDF:** https://arxiv.org/pdf/2601.18533v1
+
+## 19197. From Cold Start to Active Learning: Embedding-Based Scan Selection for Medical Image Segmentation
+
+- **ID:** `2601.18532` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Devon Levy, Bar Assayag, Laura Gaspar, et al.
+- **abs:** https://arxiv.org/abs/2601.18532 · **PDF:** https://arxiv.org/pdf/2601.18532v2
+
+## 19198. Exploring Fine-Tuning for In-Context Retrieval and Efficient KV-Caching in Long-Context Language Models
+
+- **ID:** `2601.18527` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Francesco Maria Molfese, Momchil Hardalov, Rexhina Blloshmi, et al.
+- **abs:** https://arxiv.org/abs/2601.18527 · **PDF:** https://arxiv.org/pdf/2601.18527v1
+
+## 19199. Closing the Modality Gap Aligns Group-Wise Semantics
+
+- **ID:** `2601.18525` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Eleonora Grassucci, Giordano Cicchetti, Emanuele Frasca, et al.
+- **abs:** https://arxiv.org/abs/2601.18525 · **PDF:** https://arxiv.org/pdf/2601.18525v1
+
+## 19200. From Human Labels to Literature: Semi-Supervised Learning of NMR Chemical Shifts at Scale
+
+- **ID:** `2601.18524` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Yongqi Jin, Yecheng Wang, Jun-jie Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.18524 · **PDF:** https://arxiv.org/pdf/2601.18524v1
+
+## 19201. Scalable Transit Delay Prediction at City Scale: A Systematic Approach with Multi-Resolution Feature Engineering and Deep Learning
+
+- **ID:** `2601.18521` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Emna Boudabbous, Mohamed Karaa, Lokman Sboui, et al.
+- **abs:** https://arxiv.org/abs/2601.18521 · **PDF:** https://arxiv.org/pdf/2601.18521v1
+
+## 19202. GenAI for Social Work Field Education: Client Simulation with Real-Time Feedback
+
+- **ID:** `2601.18517` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** James Sungarda, Hongkai Liu, Zilong Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.18517 · **PDF:** https://arxiv.org/pdf/2601.18517v1
+
+## 19203. LipNeXt: Scaling up Lipschitz-based Certified Robustness to Billion-parameter Models
+
+- **ID:** `2601.18513` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Kai Hu, Haoqi Hu, Matt Fredrikson
+- **abs:** https://arxiv.org/abs/2601.18513 · **PDF:** https://arxiv.org/pdf/2601.18513v1
+
+## 19204. Using Large Language Models to Construct Virtual Top Managers: A Method for Organizational Research
+
+- **ID:** `2601.18512` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Antonio Garzon-Vico, Krithika Sharon Komalapati, Arsalan Shahid, et al.
+- **abs:** https://arxiv.org/abs/2601.18512 · **PDF:** https://arxiv.org/pdf/2601.18512v1
+
+## 19205. Just-In-Time Reinforcement Learning: Continual Learning in LLM Agents Without Gradient Updates
+
+- **ID:** `2601.18510` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Yibo Li, Zijie Lin, Ailin Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.18510 · **PDF:** https://arxiv.org/pdf/2601.18510v1
+
+## 19206. Conformal Prediction Algorithms for Time Series Forecasting: Methods and Benchmarking
+
+- **ID:** `2601.18509` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Andro Sabashvili
+- **abs:** https://arxiv.org/abs/2601.18509 · **PDF:** https://arxiv.org/pdf/2601.18509v2
+
+## 19207. Nearly Optimal Bayesian Inference for Structural Missingness
+
+- **ID:** `2601.18500` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Chen Liang, Donghua Yang, Yutong Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.18500 · **PDF:** https://arxiv.org/pdf/2601.18500v2
+
+## 19208. DEEPMED: Building a Medical DeepResearch Agent via Multi-hop Med-Search Data and Turn-Controlled Agentic Training & Inference
+
+- **ID:** `2601.18496` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Zihan Wang, Hao Wang, Shi Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.18496 · **PDF:** https://arxiv.org/pdf/2601.18496v2
+
+## 19209. AgentDoG: A Diagnostic Guardrail Framework for AI Agent Safety and Security
+
+- **ID:** `2601.18491` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Dongrui Liu, Qihan Ren, Chen Qian, et al.
+- **abs:** https://arxiv.org/abs/2601.18491 · **PDF:** https://arxiv.org/pdf/2601.18491v1
+
+## 19210. Different Demographic Cues Yield Inconsistent Conclusions About LLM Personalization and Bias
+
+- **ID:** `2601.18486` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Manuel Tonneau, Neil K. R. Seghal, Niyati Malhotra, et al.
+- **abs:** https://arxiv.org/abs/2601.18486 · **PDF:** https://arxiv.org/pdf/2601.18486v2
+
+## 19211. Funny or Persuasive, but Not Both: Evaluating Fine-Grained Multi-Concept Control in LLMs
+
+- **ID:** `2601.18483` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Arya Labroo, Ivaxi Sheth, Vyas Raina, et al.
+- **abs:** https://arxiv.org/abs/2601.18483 · **PDF:** https://arxiv.org/pdf/2601.18483v1
+
+## 19212. Enhancing Control Policy Smoothness by Aligning Actions with Predictions from Preceding States
+
+- **ID:** `2601.18479` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Kyoleen Kwak, Hyoseok Hwang
+- **abs:** https://arxiv.org/abs/2601.18479 · **PDF:** https://arxiv.org/pdf/2601.18479v1
+
+## 19213. Latent Knowledge as a Predictor of Fact Acquisition in Fine-Tuned Large Language Models
+
+- **ID:** `2601.18468` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Daniel B. Hier, Tayo Obafemi-Ajayi
+- **abs:** https://arxiv.org/abs/2601.18468 · **PDF:** https://arxiv.org/pdf/2601.18468v1
+
+## 19214. OffSeeker: Online Reinforcement Learning Is Not All You Need for Deep Research Agents
+
+- **ID:** `2601.18467` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Yuhang Zhou, Kai Zheng, Qiguang Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.18467 · **PDF:** https://arxiv.org/pdf/2601.18467v2
+
+## 19215. Fair-Eye Net: A Fair, Trustworthy, Multimodal Integrated Glaucoma Full Chain AI System
+
+- **ID:** `2601.18464` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Wenbin Wei, Suyuan Yao, Cheng Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.18464 · **PDF:** https://arxiv.org/pdf/2601.18464v1
+
+## 19216. Mind the Boundary: Stabilizing Gemini Enterprise A2A via a Cloud Run Hub Across Projects and Accounts
+
+- **ID:** `2602.17675` · **primary:** `cs.DC` · **Submitted:** 2026-01-26
+- **Authors:** Takao Morita
+- **abs:** https://arxiv.org/abs/2602.17675 · **PDF:** https://arxiv.org/pdf/2602.17675v1
+
+## 19217. 3DGesPolicy: Phoneme-Aware Holistic Co-Speech Gesture Generation Based on Action Control
+
+- **ID:** `2601.18451` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Xuanmeng Sha, Liyun Zhang, Tomohiro Mashita, et al.
+- **abs:** https://arxiv.org/abs/2601.18451 · **PDF:** https://arxiv.org/pdf/2601.18451v1
+
+## 19218. On Procrustes Contamination in Machine Learning Applications of Geometric Morphometrics
+
+- **ID:** `2601.18448` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Lloyd Austin Courtenay
+- **abs:** https://arxiv.org/abs/2601.18448 · **PDF:** https://arxiv.org/pdf/2601.18448v1
+
+## 19219. GCFX: Generative Counterfactual Explanations for Deep Graph Models at the Model Level
+
+- **ID:** `2601.18447` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Jinlong Hu, Jiacheng Liu
+- **abs:** https://arxiv.org/abs/2601.18447 · **PDF:** https://arxiv.org/pdf/2601.18447v1
+
+## 19220. Scaling Behaviors of Evolutionary Algorithms on GPUs: When Does Parallelism Pay Off?
+
+- **ID:** `2601.18446` · **primary:** `cs.NE` · **Submitted:** 2026-01-26
+- **Authors:** Xinmeng Yu, Tao Jiang, Ran Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.18446 · **PDF:** https://arxiv.org/pdf/2601.18446v2
+
+## 19221. MulVul: Retrieval-augmented Multi-Agent Code Vulnerability Detection via Cross-Model Prompt Evolution
+
+- **ID:** `2601.18847` · **primary:** `cs.SE` · **Submitted:** 2026-01-26
+- **Authors:** Zihan Wu, Jie Xu, Yun Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.18847 · **PDF:** https://arxiv.org/pdf/2601.18847v1
+
+## 19222. Fusion of Spatio-Temporal and Multi-Scale Frequency Features for Dry Electrodes MI-EEG Decoding
+
+- **ID:** `2601.18424` · **primary:** `cs.HC` · **Submitted:** 2026-01-26
+- **Authors:** Tianyi Gong, Can Han, Junxi Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.18424 · **PDF:** https://arxiv.org/pdf/2601.18424v1
+
+## 19223. LLM Driven Design of Continuous Optimization Problems with Controllable High-level Properties
+
+- **ID:** `2601.18846` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Urban Skvorc, Niki van Stein, Moritz Seiler, et al.
+- **abs:** https://arxiv.org/abs/2601.18846 · **PDF:** https://arxiv.org/pdf/2601.18846v1
+
+## 19224. Gradient Regularized Natural Gradients
+
+- **ID:** `2601.18420` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Satya Prakash Dash, Hossein Abdi, Wei Pan, et al.
+- **abs:** https://arxiv.org/abs/2601.18420 · **PDF:** https://arxiv.org/pdf/2601.18420v2
+
+## 19225. Emergent Cooperation in Quantum Multi-Agent Reinforcement Learning Using Communication
+
+- **ID:** `2601.18419` · **primary:** `quant-ph` · **Submitted:** 2026-01-26
+- **Authors:** Michael Kölle, Christian Reff, Leo Sünkel, et al.
+- **abs:** https://arxiv.org/abs/2601.18419 · **PDF:** https://arxiv.org/pdf/2601.18419v1
+
+## 19226. daVinci-Dev: Agent-native Mid-training for Software Engineering
+
+- **ID:** `2601.18418` · **primary:** `cs.SE` · **Submitted:** 2026-01-26
+- **Authors:** Ji Zeng, Dayuan Fu, Tiantian Mi, et al.
+- **abs:** https://arxiv.org/abs/2601.18418 · **PDF:** https://arxiv.org/pdf/2601.18418v2
+
+## 19227. Reducing False Positives in Static Bug Detection with LLMs: An Empirical Study in Industry
+
+- **ID:** `2601.18844` · **primary:** `cs.SE` · **Submitted:** 2026-01-26
+- **Authors:** Xueying Du, Jiayi Feng, Yi Zou, et al.
+- **abs:** https://arxiv.org/abs/2601.18844 · **PDF:** https://arxiv.org/pdf/2601.18844v1
+
+## 19228. Pisets: A Robust Speech Recognition System for Lectures and Interviews
+
+- **ID:** `2601.18415` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Ivan Bondarenko, Daniil Grebenkin, Oleg Sedukhin, et al.
+- **abs:** https://arxiv.org/abs/2601.18415 · **PDF:** https://arxiv.org/pdf/2601.18415v1
+
+## 19229. Frequency-Based Hyperparameter Selection in Games
+
+- **ID:** `2601.18409` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Aniket Sanyal, Baraah A. M. Sidahmed, Rebekka Burkholz, et al.
+- **abs:** https://arxiv.org/abs/2601.18409 · **PDF:** https://arxiv.org/pdf/2601.18409v1
+
+## 19230. Superlinear Multi-Step Attention
+
+- **ID:** `2601.18401` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Yufeng Huang
+- **abs:** https://arxiv.org/abs/2601.18401 · **PDF:** https://arxiv.org/pdf/2601.18401v1
+
+## 19231. Estimating Dense-Packed Zone Height in Liquid-Liquid Separation: A Physics-Informed Neural Network Approach
+
+- **ID:** `2601.18399` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Mehmet Velioglu, Song Zhai, Alexander Mitsos, et al.
+- **abs:** https://arxiv.org/abs/2601.18399 · **PDF:** https://arxiv.org/pdf/2601.18399v1
+
+## 19232. Noise-Robust AV-ASR Using Visual Features Both in the Whisper Encoder and Decoder
+
+- **ID:** `2601.18396` · **primary:** `eess.AS` · **Submitted:** 2026-01-26
+- **Authors:** Zhengyang Li, Thomas Graave, Björn Möller, et al.
+- **abs:** https://arxiv.org/abs/2601.18396 · **PDF:** https://arxiv.org/pdf/2601.18396v1
+
+## 19233. Do not be greedy, Think Twice: Sampling and Selection for Document-level Information Extraction
+
+- **ID:** `2601.18395` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Mikel Zubillaga, Oscar Sainz, Oier Lopez de Lacalle, et al.
+- **abs:** https://arxiv.org/abs/2601.18395 · **PDF:** https://arxiv.org/pdf/2601.18395v1
+
+## 19234. OCR-Enhanced Multimodal ASR Can Read While Listening
+
+- **ID:** `2601.18393` · **primary:** `cs.SD` · **Submitted:** 2026-01-26
+- **Authors:** Junli Chen, Changli Tang, Yixuan Li, et al.
+- **abs:** https://arxiv.org/abs/2601.18393 · **PDF:** https://arxiv.org/pdf/2601.18393v1
+
+## 19235. GUIGuard: Toward a General Framework for Privacy-Preserving GUI Agents
+
+- **ID:** `2601.18842` · **primary:** `cs.CR` · **Submitted:** 2026-01-26
+- **Authors:** Yanxi Wang, Zhiling Zhang, Wenbo Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.18842 · **PDF:** https://arxiv.org/pdf/2601.18842v2
+
+## 19236. Dynamic Thinking-Token Selection for Efficient Reasoning in Large Reasoning Models
+
+- **ID:** `2601.18383` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Zhenyuan Guo, Tong Chen, Wenlong Meng, et al.
+- **abs:** https://arxiv.org/abs/2601.18383 · **PDF:** https://arxiv.org/pdf/2601.18383v1
+
+## 19237. AI Agent for Reverse-Engineering Legacy Finite-Difference Code and Translating to Devito
+
+- **ID:** `2601.18381` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Yinghan Hou, Zongyou Yang
+- **abs:** https://arxiv.org/abs/2601.18381 · **PDF:** https://arxiv.org/pdf/2601.18381v1
+
+## 19238. Corpus-Based Approaches to Igbo Diacritic Restoration
+
+- **ID:** `2601.18380` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Ignatius Ezeani
+- **abs:** https://arxiv.org/abs/2601.18380 · **PDF:** https://arxiv.org/pdf/2601.18380v1
+
+## 19239. Hierarchical Text Classification with LLM-Refined Taxonomies
+
+- **ID:** `2601.18375` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Jonas Golde, Nicolaas Jedema, Ravi Krishnan, et al.
+- **abs:** https://arxiv.org/abs/2601.18375 · **PDF:** https://arxiv.org/pdf/2601.18375v1
+
+## 19240. CitiLink: Enhancing Municipal Transparency and Citizen Engagement through Searchable Meeting Minutes
+
+- **ID:** `2601.18374` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Rodrigo Silva, José Evans, José Isidro, et al.
+- **abs:** https://arxiv.org/abs/2601.18374 · **PDF:** https://arxiv.org/pdf/2601.18374v3
+
+## 19241. Making medical vision-language models think causally across modalities with retrieval-augmented cross-modal reasoning
+
+- **ID:** `2601.18356` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Weiqin Yang, Haowen Xue, Qingyi Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.18356 · **PDF:** https://arxiv.org/pdf/2601.18356v1
+
+## 19242. Can Good Writing Be Generative? Expert-Level AI Writing Emerges through Fine-Tuning on High-Quality Books
+
+- **ID:** `2601.18353` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Tuhin Chakrabarty, Paramveer S. Dhillon
+- **abs:** https://arxiv.org/abs/2601.18353 · **PDF:** https://arxiv.org/pdf/2601.18353v1
+
+## 19243. Code over Words: Overcoming Semantic Inertia via Code-Grounded Reasoning
+
+- **ID:** `2601.18352` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Manjie Xu, Isabella Yin, Xinyi Tu, et al.
+- **abs:** https://arxiv.org/abs/2601.18352 · **PDF:** https://arxiv.org/pdf/2601.18352v2
+
+## 19244. Analysis of Control Bellman Residual Minimization for Markov Decision Problem
+
+- **ID:** `2601.18840` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Donghwan Lee, Hyukjun Yang
+- **abs:** https://arxiv.org/abs/2601.18840 · **PDF:** https://arxiv.org/pdf/2601.18840v2
+
+## 19245. Adapter Merging Reactivates Latent Reasoning Traces: A Mechanism Analysis
+
+- **ID:** `2601.18350` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Junyi Zou
+- **abs:** https://arxiv.org/abs/2601.18350 · **PDF:** https://arxiv.org/pdf/2601.18350v4
+
+## 19246. Structural Gender Bias in Credit Scoring: Proxy Leakage
+
+- **ID:** `2601.18342` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Navya SD, Sreekanth D, SS Uma Sankari
+- **abs:** https://arxiv.org/abs/2601.18342 · **PDF:** https://arxiv.org/pdf/2601.18342v1
+
+## 19247. Not the Example, but the Process: How Self-Generated Examples Enhance LLM Reasoning
+
+- **ID:** `2602.15863` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Daehoon Gwak, Minseo Jung, Junwoo Park, et al.
+- **abs:** https://arxiv.org/abs/2602.15863 · **PDF:** https://arxiv.org/pdf/2602.15863v1
+
+## 19248. A Dataset for Automatic Vocal Mode Classification
+
+- **ID:** `2601.18339` · **primary:** `cs.SD` · **Submitted:** 2026-01-26
+- **Authors:** Reemt Hinrichs, Sonja Stephan, Alexander Lange, et al.
+- **abs:** https://arxiv.org/abs/2601.18339 · **PDF:** https://arxiv.org/pdf/2601.18339v1
+
+## 19249. Analytic Incremental Learning For Sound Source Localization With Imbalance Rectification
+
+- **ID:** `2601.18335` · **primary:** `cs.SD` · **Submitted:** 2026-01-26
+- **Authors:** Zexia Fan, Yu Chen, Qiquan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.18335 · **PDF:** https://arxiv.org/pdf/2601.18335v1
+
+## 19250. Overalignment in Frontier LLMs: An Empirical Study of Sycophantic Behaviour in Healthcare
+
+- **ID:** `2601.18334` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Clément Christophe, Wadood Mohammed Abdul, Prateek Munjal, et al.
+- **abs:** https://arxiv.org/abs/2601.18334 · **PDF:** https://arxiv.org/pdf/2601.18334v1
+
+## 19251. A Tumor Aware DenseNet Swin Hybrid Learning with Boosted and Hierarchical Feature Spaces for Large-Scale Brain MRI Classification
+
+- **ID:** `2601.18330` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Muhammad Ali Shah, Muhammad Mansoor Alam, Saddam Hussain Khan
+- **abs:** https://arxiv.org/abs/2601.18330 · **PDF:** https://arxiv.org/pdf/2601.18330v1
+
+## 19252. Discriminability-Driven Spatial-Channel Selection with Gradient Norm for Drone Signal OOD Detection
+
+- **ID:** `2601.18329` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Chuhan Feng, Jing Li, Jie Li, et al.
+- **abs:** https://arxiv.org/abs/2601.18329 · **PDF:** https://arxiv.org/pdf/2601.18329v1
+
+## 19253. Cognitive Fusion of ZC Sequences and Time-Frequency Images for Out-of-Distribution Detection of Drone Signals
+
+- **ID:** `2601.18326` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Jie Li, Jing Li, Lu Lv, et al.
+- **abs:** https://arxiv.org/abs/2601.18326 · **PDF:** https://arxiv.org/pdf/2601.18326v1
+
+## 19254. Enhancing Action and Ingredient Modeling for Semantically Grounded Recipe Generation
+
+- **ID:** `2602.15862` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Guoshan Liu, Bin Zhu, Yian Li, et al.
+- **abs:** https://arxiv.org/abs/2602.15862 · **PDF:** https://arxiv.org/pdf/2602.15862v1
+
+## 19255. Integrating Fine-Grained Audio-Visual Evidence for Robust Multimodal Emotion Reasoning
+
+- **ID:** `2601.18321` · **primary:** `cs.MM` · **Submitted:** 2026-01-26
+- **Authors:** Zhixian Zhao, Wenjie Tian, Lei Xie
+- **abs:** https://arxiv.org/abs/2601.18321 · **PDF:** https://arxiv.org/pdf/2601.18321v2
+
+## 19256. MultiVis-Agent: A Multi-Agent Framework with Logic Rules for Reliable and Comprehensive Cross-Modal Data Visualization
+
+- **ID:** `2601.18320` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Jinwei Lu, Yuanfeng Song, Chen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.18320 · **PDF:** https://arxiv.org/pdf/2601.18320v1
+
+## 19257. CAST: Achieving Stable LLM-based Text Analysis for Data Analytics
+
+- **ID:** `2602.15861` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Jinxiang Xie, Zihao Li, Wei He, et al.
+- **abs:** https://arxiv.org/abs/2602.15861 · **PDF:** https://arxiv.org/pdf/2602.15861v1
+
+## 19258. A Master Class on Reproducibility: A Student Hackathon on Advanced MRI Reconstruction Methods
+
+- **ID:** `2601.18314` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Lina Felsner, Sevgi G. Kafali, Hannah Eichhorn, et al.
+- **abs:** https://arxiv.org/abs/2601.18314 · **PDF:** https://arxiv.org/pdf/2601.18314v1
+
+## 19259. Convex Chance-Constrained Stochastic Control under Uncertain Specifications with Application to Learning-Based Hybrid Powertrain Control
+
+- **ID:** `2601.18313` · **primary:** `eess.SY` · **Submitted:** 2026-01-26
+- **Authors:** Teruki Kato, Ryotaro Shima, Kenji Kashima
+- **abs:** https://arxiv.org/abs/2601.18313 · **PDF:** https://arxiv.org/pdf/2601.18313v1
+
+## 19260. A Generative AI-Driven Reliability Layer for Action-Oriented Disaster Resilience
+
+- **ID:** `2601.18308` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Geunsik Lim
+- **abs:** https://arxiv.org/abs/2601.18308 · **PDF:** https://arxiv.org/pdf/2601.18308v1
+
+## 19261. Calibrating Beyond English: Language Diversity for Better Quantized Multilingual LLM
+
+- **ID:** `2601.18306` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Everlyn Asiko Chimoto, Mostafa Elhoushi, Bruce A. Bassett
+- **abs:** https://arxiv.org/abs/2601.18306 · **PDF:** https://arxiv.org/pdf/2601.18306v1
+
+## 19262. Suppressing Final Layer Hidden State Jumps in Transformer Pretraining
+
+- **ID:** `2601.18302` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Keigo Shibata, Kazuki Yano, Ryosuke Takahashi, et al.
+- **abs:** https://arxiv.org/abs/2601.18302 · **PDF:** https://arxiv.org/pdf/2601.18302v1
+
+## 19263. Temp-R1: A Unified Autonomous Agent for Complex Temporal KGQA via Reverse Curriculum Reinforcement Learning
+
+- **ID:** `2601.18296` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Zhaoyan Gong, Zhiqiang Liu, Songze Li, et al.
+- **abs:** https://arxiv.org/abs/2601.18296 · **PDF:** https://arxiv.org/pdf/2601.18296v1
+
+## 19264. TriPlay-RL: Tri-Role Self-Play Reinforcement Learning for LLM Safety Alignment
+
+- **ID:** `2601.18292` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Zhewen Tan, Wenhan Yu, Jianfeng Si, et al.
+- **abs:** https://arxiv.org/abs/2601.18292 · **PDF:** https://arxiv.org/pdf/2601.18292v2
+
+## 19265. U-Fold: Dynamic Intent-Aware Context Folding for User-Centric Agents
+
+- **ID:** `2601.18285` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Jin Su, Runnan Fang, Yeqiu Li, et al.
+- **abs:** https://arxiv.org/abs/2601.18285 · **PDF:** https://arxiv.org/pdf/2601.18285v1
+
+## 19266. Think-Augmented Function Calling: Improving LLM Parameter Accuracy Through Embedded Reasoning
+
+- **ID:** `2601.18282` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Lei Wei, Xiao Peng, Jinpeng Ou, et al.
+- **abs:** https://arxiv.org/abs/2601.18282 · **PDF:** https://arxiv.org/pdf/2601.18282v2
+
+## 19267. Reflecting Twice before Speaking with Empathy: Self-Reflective Alternating Inference for Empathy-Aware End-to-End Spoken Dialogue
+
+- **ID:** `2601.18281` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Yuhang Jia, Pei Liu, Haoqin Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.18281 · **PDF:** https://arxiv.org/pdf/2601.18281v1
+
+## 19268. What Do Learned Models Measure?
+
+- **ID:** `2601.18278` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Indrė Žliobaitė
+- **abs:** https://arxiv.org/abs/2601.18278 · **PDF:** https://arxiv.org/pdf/2601.18278v1
+
+## 19269. TEFormer: Structured Bidirectional Temporal Enhancement Modeling in Spiking Transformers
+
+- **ID:** `2601.18274` · **primary:** `cs.NE` · **Submitted:** 2026-01-26
+- **Authors:** Sicheng Shen, Mingyang Lv, Bing Han, et al.
+- **abs:** https://arxiv.org/abs/2601.18274 · **PDF:** https://arxiv.org/pdf/2601.18274v1
+
+## 19270. Toward Scalable Normalizing Flows for the Hubbard Model
+
+- **ID:** `2601.18273` · **primary:** `cond-mat.str-el` · **Submitted:** 2026-01-26
+- **Authors:** Janik Kreit, Andrea Bulgarelli, Lena Funcke, et al.
+- **abs:** https://arxiv.org/abs/2601.18273 · **PDF:** https://arxiv.org/pdf/2601.18273v1
+
+## 19271. Designing large language model prompts to extract scores from messy text: A shared dataset and challenge
+
+- **ID:** `2601.18271` · **primary:** `cs.DL` · **Submitted:** 2026-01-26
+- **Authors:** Mike Thelwall
+- **abs:** https://arxiv.org/abs/2601.18271 · **PDF:** https://arxiv.org/pdf/2601.18271v1
+
+## 19272. Neural Network Approximation: A View from Polytope Decomposition
+
+- **ID:** `2601.18264` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** ZeYu Li, ShiJun Zhang, TieYong Zeng, et al.
+- **abs:** https://arxiv.org/abs/2601.18264 · **PDF:** https://arxiv.org/pdf/2601.18264v1
+
+## 19273. FGGM: Fisher-Guided Gradient Masking for Continual Learning
+
+- **ID:** `2601.18261` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Chao-Hong Tan, Qian Chen, Wen Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.18261 · **PDF:** https://arxiv.org/pdf/2601.18261v1
+
+## 19274. Beyond Retention: Orchestrating Structural Safety and Plasticity in Continual Learning for LLMs
+
+- **ID:** `2601.18255` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Fei Meng
+- **abs:** https://arxiv.org/abs/2601.18255 · **PDF:** https://arxiv.org/pdf/2601.18255v1
+
+## 19275. BoRP: Bootstrapped Regression Probing for Scalable and Human-Aligned LLM Evaluation
+
+- **ID:** `2601.18253` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Peng Sun, Xiangyu Zhang, Duan Wu
+- **abs:** https://arxiv.org/abs/2601.18253 · **PDF:** https://arxiv.org/pdf/2601.18253v1
+
+## 19276. Co-PLNet: A Collaborative Point-Line Network for Prompt-Guided Wireframe Parsing
+
+- **ID:** `2601.18252` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Chao Wang, Xuanying Li, Cheng Dai, et al.
+- **abs:** https://arxiv.org/abs/2601.18252 · **PDF:** https://arxiv.org/pdf/2601.18252v1
+
+## 19277. A multimodal vision foundation model for generalizable knee pathology
+
+- **ID:** `2601.18250` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Kang Yu, Dingyu Wang, Zimu Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.18250 · **PDF:** https://arxiv.org/pdf/2601.18250v1
+
+## 19278. Tractable Gaussian Phase Retrieval with Heavy Tails and Adversarial Corruption with Near-Linear Sample Complexity
+
+- **ID:** `2601.18245` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Santanu Das, Jatin Batra
+- **abs:** https://arxiv.org/abs/2601.18245 · **PDF:** https://arxiv.org/pdf/2601.18245v2
+
+## 19279. Reranker Optimization via Geodesic Distances on k-NN Manifolds
+
+- **ID:** `2602.15860` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Wen G. Gong
+- **abs:** https://arxiv.org/abs/2602.15860 · **PDF:** https://arxiv.org/pdf/2602.15860v1
+
+## 19280. Strength Change Explanations in Quantitative Argumentation
+
+- **ID:** `2603.00008` · **primary:** `cs.MA` · **Submitted:** 2026-01-26
+- **Authors:** Timotheus Kampik, Xiang Yin, Nico Potyka, et al.
+- **abs:** https://arxiv.org/abs/2603.00008 · **PDF:** https://arxiv.org/pdf/2603.00008v1
+
+## 19281. TAM-Eval: Evaluating LLMs for Automated Unit Test Maintenance
+
+- **ID:** `2601.18241` · **primary:** `cs.SE` · **Submitted:** 2026-01-26
+- **Authors:** Elena Bruches, Vadim Alperovich, Dari Baturova, et al.
+- **abs:** https://arxiv.org/abs/2601.18241 · **PDF:** https://arxiv.org/pdf/2601.18241v1
+
+## 19282. From Transcripts to AI Agents: Knowledge Extraction, RAG Integration, and Robust Evaluation of Conversational AI Assistants
+
+- **ID:** `2602.15859` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Krittin Pachtrachai, Petmongkon Pornpichitsuwan, Wachiravit Modecrua, et al.
+- **abs:** https://arxiv.org/abs/2602.15859 · **PDF:** https://arxiv.org/pdf/2602.15859v1
+
+## 19283. TechING: Towards Real World Technical Image Understanding via VLMs
+
+- **ID:** `2601.18238` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Tafazzul Nadeem, Bhavik Shangari, Manish Rai, et al.
+- **abs:** https://arxiv.org/abs/2601.18238 · **PDF:** https://arxiv.org/pdf/2601.18238v1
+
+## 19284. Generative AI in Saudi Arabia: A National Survey of Adoption, Risks, and Public Perceptions
+
+- **ID:** `2601.18234` · **primary:** `cs.CY` · **Submitted:** 2026-01-26
+- **Authors:** Abdulaziz AlDakheel, Ali Alshehre, Esraa Alamoudi, et al.
+- **abs:** https://arxiv.org/abs/2601.18234 · **PDF:** https://arxiv.org/pdf/2601.18234v1
+
+## 19285. Rethinking Cross-Modal Fine-Tuning: Optimizing the Interaction between Feature Alignment and Target Fitting
+
+- **ID:** `2601.18231` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Trong Khiem Tran, Manh Cuong Dao, Phi Le Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.18231 · **PDF:** https://arxiv.org/pdf/2601.18231v3
+
+## 19286. Facial Emotion Recognition on FER-2013 using an EfficientNetB2-Based Approach
+
+- **ID:** `2601.18228` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Sahil Naik, Soham Bagayatkar, Pavankumar Singh
+- **abs:** https://arxiv.org/abs/2601.18228 · **PDF:** https://arxiv.org/pdf/2601.18228v1
+
+## 19287. Yunjue Agent Tech Report: A Fully Reproducible, Zero-Start In-Situ Self-Evolving Agent System for Open-Ended Tasks
+
+- **ID:** `2601.18226` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Haotian Li, Shijun Yang, Weizhen Qi, et al.
+- **abs:** https://arxiv.org/abs/2601.18226 · **PDF:** https://arxiv.org/pdf/2601.18226v2
+
+## 19288. ShopSimulator: Evaluating and Exploring RL-Driven LLM Agent for Shopping Assistants
+
+- **ID:** `2601.18225` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Pei Wang, Yanan Wu, Xiaoshuai Song, et al.
+- **abs:** https://arxiv.org/abs/2601.18225 · **PDF:** https://arxiv.org/pdf/2601.18225v1
+
+## 19289. Automated HER2 scoring with uncertainty quantification using lensfree holography and deep learning
+
+- **ID:** `2601.18219` · **primary:** `physics.med-ph` · **Submitted:** 2026-01-26
+- **Authors:** Che-Yung Shen, Xilin Yang, Yuzhu Li, et al.
+- **abs:** https://arxiv.org/abs/2601.18219 · **PDF:** https://arxiv.org/pdf/2601.18219v1
+
+## 19290. PaperTok: Exploring the Use of Generative AI for Creating Short-form Videos for Research Communication
+
+- **ID:** `2601.18218` · **primary:** `cs.HC` · **Submitted:** 2026-01-26
+- **Authors:** Meziah Ruby Cristobal, Hyeonjeong Byeon, Tze-Yu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.18218 · **PDF:** https://arxiv.org/pdf/2601.18218v1
+
+## 19291. Paying Less Generalization Tax: A Cross-Domain Generalization Study of RL Training for LLM Agents
+
+- **ID:** `2601.18217` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Zhihan Liu, Lin Guan, Yixin Nie, et al.
+- **abs:** https://arxiv.org/abs/2601.18217 · **PDF:** https://arxiv.org/pdf/2601.18217v1
+
+## 19292. PaperSearchQA: Learning to Search and Reason over Scientific Papers with RLVR
+
+- **ID:** `2601.18207` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** James Burgess, Jan N. Hansen, Duo Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.18207 · **PDF:** https://arxiv.org/pdf/2601.18207v1
+
+## 19293. MemWeaver: Weaving Hybrid Memories for Traceable Long-Horizon Agentic Reasoning
+
+- **ID:** `2601.18204` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Juexiang Ye, Xue Li, Xinyu Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.18204 · **PDF:** https://arxiv.org/pdf/2601.18204v1
+
+## 19294. SAGE: Steerable Agentic Data Generation for Deep Search with Execution Feedback
+
+- **ID:** `2601.18202` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Fangyuan Xu, Rujun Han, Yanfei Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.18202 · **PDF:** https://arxiv.org/pdf/2601.18202v1
+
+## 19295. HeterCSI: Channel-Adaptive Heterogeneous CSI Pretraining Framework for Generalized Wireless Foundation Models
+
+- **ID:** `2601.18200` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Chenyu Zhang, Xinchen Lyu, Chenshan Ren, et al.
+- **abs:** https://arxiv.org/abs/2601.18200 · **PDF:** https://arxiv.org/pdf/2601.18200v1
+
+## 19296. GAIA: A Data Flywheel System for Training GUI Test-Time Scaling Critic Models
+
+- **ID:** `2601.18197` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Shaokang Wang, Pei Fu, Ruoceng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.18197 · **PDF:** https://arxiv.org/pdf/2601.18197v1
+
+## 19297. Smooth, Sparse, and Stable: Finite-Time Exact Skeleton Recovery via Smoothed Proximal Gradients
+
+- **ID:** `2601.18189` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Rui Wu, Yongjun Li
+- **abs:** https://arxiv.org/abs/2601.18189 · **PDF:** https://arxiv.org/pdf/2601.18189v1
+
+## 19298. \textsc{NaVIDA}: Vision-Language Navigation with Inverse Dynamics Augmentation
+
+- **ID:** `2601.18188` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Weiye Zhu, Zekai Zhang, Xiangchen Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.18188 · **PDF:** https://arxiv.org/pdf/2601.18188v2
+
+## 19299. VIBEVOICE-ASR Technical Report
+
+- **ID:** `2601.18184` · **primary:** `cs.SD` · **Submitted:** 2026-01-26
+- **Authors:** Zhiliang Peng, Jianwei Yu, Yaoyao Chang, et al.
+- **abs:** https://arxiv.org/abs/2601.18184 · **PDF:** https://arxiv.org/pdf/2601.18184v2
+
+## 19300. Success Conditioning as Policy Improvement: The Optimization Problem Solved by Imitating Success
+
+- **ID:** `2601.18175` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Daniel Russo
+- **abs:** https://arxiv.org/abs/2601.18175 · **PDF:** https://arxiv.org/pdf/2601.18175v1
+
+## 19301. Community Norms in the Spotlight: Enabling Task-Agnostic Unsupervised Pre-Training to Benefit Online Social Media
+
+- **ID:** `2602.02525` · **primary:** `cs.SI` · **Submitted:** 2026-01-26
+- **Authors:** Liam Hebert, Lucas Kopp, Robin Cohen
+- **abs:** https://arxiv.org/abs/2602.02525 · **PDF:** https://arxiv.org/pdf/2602.02525v1
+
+## 19302. Learning Fair Domain Adaptation with Virtual Label Distribution
+
+- **ID:** `2601.18171` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Yuguang Zhang, Lijun Sheng, Jian Liang, et al.
+- **abs:** https://arxiv.org/abs/2601.18171 · **PDF:** https://arxiv.org/pdf/2601.18171v1
+
+## 19303. GASTON: Graph-Aware Social Transformer for Online Networks
+
+- **ID:** `2602.02524` · **primary:** `cs.SI` · **Submitted:** 2026-01-26
+- **Authors:** Olha Wloch, Liam Hebert, Robin Cohen, et al.
+- **abs:** https://arxiv.org/abs/2602.02524 · **PDF:** https://arxiv.org/pdf/2602.02524v1
+
+## 19304. Fine-Grained Emotion Detection on GoEmotions: Experimental Comparison of Classical Machine Learning, BiLSTM, and Transformer Models
+
+- **ID:** `2601.18162` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Ani Harutyunyan, Sachin Kumar
+- **abs:** https://arxiv.org/abs/2601.18162 · **PDF:** https://arxiv.org/pdf/2601.18162v1
+
+## 19305. Agentic Very Long Video Understanding
+
+- **ID:** `2601.18157` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Aniket Rege, Arka Sadhu, Yuliang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.18157 · **PDF:** https://arxiv.org/pdf/2601.18157v2
+
+## 19306. Beyond Pairwise Comparisons: A Distributional Test of Distinctiveness for Machine-Generated Works in Intellectual Property Law
+
+- **ID:** `2601.18156` · **primary:** `cs.CY` · **Submitted:** 2026-01-26
+- **Authors:** Anirban Mukherjee, Hannah Hanwen Chang
+- **abs:** https://arxiv.org/abs/2601.18156 · **PDF:** https://arxiv.org/pdf/2601.18156v1
+
+## 19307. Explaining Synergistic Effects in Social Recommendations
+
+- **ID:** `2601.18151` · **primary:** `cs.SI` · **Submitted:** 2026-01-26
+- **Authors:** Yicong Li, Shan Jin, Qi Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.18151 · **PDF:** https://arxiv.org/pdf/2601.18151v1
+
+## 19308. FP8-RL: A Practical and Stable Low-Precision Stack for LLM Reinforcement Learning
+
+- **ID:** `2601.18150` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Zhaopeng Qiu, Shuang Yu, Jingqi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.18150 · **PDF:** https://arxiv.org/pdf/2601.18150v1
+
+## 19309. Exact Minimum-Volume Confidence Set Intersection for Multinomial Outcomes
+
+- **ID:** `2601.18145` · **primary:** `stat.ML` · **Submitted:** 2026-01-26
+- **Authors:** Heguang Lin, Binhao Chen, Mengze Li, et al.
+- **abs:** https://arxiv.org/abs/2601.18145 · **PDF:** https://arxiv.org/pdf/2601.18145v1
+
+## 19310. Enhance the Safety in Reinforcement Learning by ADRC Lagrangian Methods
+
+- **ID:** `2601.18142` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Mingxu Zhang, Huicheng Zhang, Jiaming Ji, et al.
+- **abs:** https://arxiv.org/abs/2601.18142 · **PDF:** https://arxiv.org/pdf/2601.18142v1
+
+## 19311. DeepPlanning: Benchmarking Long-Horizon Agentic Planning with Verifiable Constraints
+
+- **ID:** `2601.18137` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Yinger Zhang, Shutong Jiang, Renhao Li, et al.
+- **abs:** https://arxiv.org/abs/2601.18137 · **PDF:** https://arxiv.org/pdf/2601.18137v1
+
+## 19312. RareAlert: Aligning heterogeneous large language model reasoning for early rare disease risk screening
+
+- **ID:** `2601.18132` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Xi Chen, Hongru Zhou, Huahui Yi, et al.
+- **abs:** https://arxiv.org/abs/2601.18132 · **PDF:** https://arxiv.org/pdf/2601.18132v1
+
+## 19313. RouteMoA: Dynamic Routing without Pre-Inference Boosts Efficient Mixture-of-Agents
+
+- **ID:** `2601.18130` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Jize Wang, Han Wu, Zhiyuan You, et al.
+- **abs:** https://arxiv.org/abs/2601.18130 · **PDF:** https://arxiv.org/pdf/2601.18130v1
+
+## 19314. Typhoon-S: Minimal Open Post-Training for Sovereign Large Language Models
+
+- **ID:** `2601.18129` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Kunat Pipatanakul, Pittawat Taveekitworachai
+- **abs:** https://arxiv.org/abs/2601.18129 · **PDF:** https://arxiv.org/pdf/2601.18129v1
+
+## 19315. Nonlinear multi-study factor analysis
+
+- **ID:** `2601.18128` · **primary:** `stat.ML` · **Submitted:** 2026-01-26
+- **Authors:** Gemma E. Moran, Anandi Krishnan
+- **abs:** https://arxiv.org/abs/2601.18128 · **PDF:** https://arxiv.org/pdf/2601.18128v1
+
+## 19316. The Limits of AI Data Transparency Policy: Three Disclosure Fallacies
+
+- **ID:** `2601.18127` · **primary:** `cs.CY` · **Submitted:** 2026-01-26
+- **Authors:** Judy Hanwen Shen, Ken Liu, Angelina Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.18127 · **PDF:** https://arxiv.org/pdf/2601.18127v1
+
+## 19317. Understanding Users' Privacy Reasoning and Behaviors During Chatbot Use to Support Meaningful Agency in Privacy
+
+- **ID:** `2601.18125` · **primary:** `cs.HC` · **Submitted:** 2026-01-26
+- **Authors:** Mohammad Hadi Nezhad, Francisco Enrique Vicente Castro, Ivon Arroyo
+- **abs:** https://arxiv.org/abs/2601.18125 · **PDF:** https://arxiv.org/pdf/2601.18125v1
+
+## 19318. Deadline-Aware, Energy-Efficient Control of Domestic Immersion Hot Water Heater
+
+- **ID:** `2601.18123` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Muhammad Ibrahim Khan, Bivin Pradeep, James Brusey
+- **abs:** https://arxiv.org/abs/2601.18123 · **PDF:** https://arxiv.org/pdf/2601.18123v2
+
+## 19319. Beyond Text-to-SQL: Can LLMs Really Debug Enterprise ETL SQL?
+
+- **ID:** `2601.18119` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Jing Ye, Yiwen Duan, Yonghong Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.18119 · **PDF:** https://arxiv.org/pdf/2601.18119v1
+
+## 19320. LungCRCT: Causal Representation based Lung CT Processing for Lung Cancer Treatment
+
+- **ID:** `2601.18118` · **primary:** `cs.CV` · **Submitted:** 2026-01-26
+- **Authors:** Daeyoung Kim
+- **abs:** https://arxiv.org/abs/2601.18118 · **PDF:** https://arxiv.org/pdf/2601.18118v1
+
+## 19321. FABLE: Forest-Based Adaptive Bi-Path LLM-Enhanced Retrieval for Multi-Document Reasoning
+
+- **ID:** `2601.18116` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Lin Sun, Linglin Zhang, Jingang Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.18116 · **PDF:** https://arxiv.org/pdf/2601.18116v1
+
+## 19322. Robust Learning of a Group DRO Neuron
+
+- **ID:** `2601.18115` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Guyang Cao, Shuyao Li, Sushrut Karmalkar, et al.
+- **abs:** https://arxiv.org/abs/2601.18115 · **PDF:** https://arxiv.org/pdf/2601.18115v1
+
+## 19323. MalURLBench: A Benchmark Evaluating Agents' Vulnerabilities When Processing Web URLs
+
+- **ID:** `2601.18113` · **primary:** `cs.CR` · **Submitted:** 2026-01-26
+- **Authors:** Dezhang Kong, Zhuxi Wu, Shiqi Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.18113 · **PDF:** https://arxiv.org/pdf/2601.18113v3
+
+## 19324. Demystifying Data-Driven Probabilistic Medium-Range Weather Forecasting
+
+- **ID:** `2601.18111` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Jean Kossaifi, Nikola Kovachki, Morteza Mardani, et al.
+- **abs:** https://arxiv.org/abs/2601.18111 · **PDF:** https://arxiv.org/pdf/2601.18111v1
+
+## 19325. AttenMIA: LLM Membership Inference Attack through Attention Signals
+
+- **ID:** `2601.18110` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Pedram Zaree, Md Abdullah Al Mamun, Yue Dong, et al.
+- **abs:** https://arxiv.org/abs/2601.18110 · **PDF:** https://arxiv.org/pdf/2601.18110v1
+
+## 19326. Beyond Static Datasets: Robust Offline Policy Optimization via Vetted Synthetic Transitions
+
+- **ID:** `2601.18107` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Pedram Agand, Mo Chen
+- **abs:** https://arxiv.org/abs/2601.18107 · **PDF:** https://arxiv.org/pdf/2601.18107v1
+
+## 19327. GLEN-Bench: A Graph-Language based Benchmark for Nutritional Health
+
+- **ID:** `2601.18106` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Jiatan Huang, Zheyuan Zhang, Tianyi Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.18106 · **PDF:** https://arxiv.org/pdf/2601.18106v1
+
+## 19328. Mitigating the OWASP Top 10 For Large Language Models Applications using Intelligent Agents
+
+- **ID:** `2601.18105` · **primary:** `cs.CR` · **Submitted:** 2026-01-26
+- **Authors:** Mohammad Fasha, Faisal Abul Rub, Nasim Matar, et al.
+- **abs:** https://arxiv.org/abs/2601.18105 · **PDF:** https://arxiv.org/pdf/2601.18105v1
+
+## 19329. CHiRPE: A Step Towards Real-World Clinical NLP with Clinician-Oriented Model Explanations
+
+- **ID:** `2601.18102` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Stephanie Fong, Zimu Wang, Guilherme C. Oliveira, et al.
+- **abs:** https://arxiv.org/abs/2601.18102 · **PDF:** https://arxiv.org/pdf/2601.18102v1
+
+## 19330. From LLMs to LRMs: Rethinking Pruning for Reasoning-Centric Models
+
+- **ID:** `2601.18091` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Longwei Ding, Anhao Zhao, Fanghua Ye, et al.
+- **abs:** https://arxiv.org/abs/2601.18091 · **PDF:** https://arxiv.org/pdf/2601.18091v1
+
+## 19331. LatentMoE: Toward Optimal Accuracy per FLOP and Parameter in Mixture of Experts
+
+- **ID:** `2601.18089` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Venmugil Elango, Nidhi Bhatia, Roger Waleffe, et al.
+- **abs:** https://arxiv.org/abs/2601.18089 · **PDF:** https://arxiv.org/pdf/2601.18089v1
+
+## 19332. "Crash Test Dummies" for AI-Enabled Clinical Assessment: Validating Virtual Patient Scenarios with Virtual Learners
+
+- **ID:** `2601.18085` · **primary:** `cs.HC` · **Submitted:** 2026-01-26
+- **Authors:** Brian Gin, Ahreum Lim, Flávia Silva e Oliveira, et al.
+- **abs:** https://arxiv.org/abs/2601.18085 · **PDF:** https://arxiv.org/pdf/2601.18085v1
+
+## 19333. DRPG (Decompose, Retrieve, Plan, Generate): An Agentic Framework for Academic Rebuttal
+
+- **ID:** `2601.18081` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Peixuan Han, Yingjie Yu, Jingjun Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.18081 · **PDF:** https://arxiv.org/pdf/2601.18081v1
+
+## 19334. Use of operator defect identities in multi-channel signal plus residual-analysis via iterated products and telescoping energy-residuals: Applications to kernels in machine learning
+
+- **ID:** `2601.18080` · **primary:** `math.FA` · **Submitted:** 2026-01-26
+- **Authors:** Palle E. T. Jorgensen, Myung-Sin Song, James F. Tian
+- **abs:** https://arxiv.org/abs/2601.18080 · **PDF:** https://arxiv.org/pdf/2601.18080v1
+
+## 19335. Sparks of Cooperative Reasoning: LLMs as Strategic Hanabi Agents
+
+- **ID:** `2601.18077` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Mahesh Ramesh, Kaousheik Jayakumar, Aswinkumar Ramkumar, et al.
+- **abs:** https://arxiv.org/abs/2601.18077 · **PDF:** https://arxiv.org/pdf/2601.18077v2
+
+## 19336. Comparison requires valid measurement: Rethinking attack success rate comparisons in AI red teaming
+
+- **ID:** `2601.18076` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Alexandra Chouldechova, A. Feder Cooper, Solon Barocas, et al.
+- **abs:** https://arxiv.org/abs/2601.18076 · **PDF:** https://arxiv.org/pdf/2601.18076v1
+
+## 19337. Diffusion Model-based Reinforcement Learning for Version Age of Information Scheduling: Average and Tail-Risk-Sensitive Control
+
+- **ID:** `2601.18069` · **primary:** `cs.NI` · **Submitted:** 2026-01-26
+- **Authors:** Haoyuan Pan, Sizhao Chen, Zhaorui Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.18069 · **PDF:** https://arxiv.org/pdf/2601.18069v1
+
+## 19338. XGuardian: Towards Explainable and Generalized AI Anti-Cheat on FPS Games
+
+- **ID:** `2601.18068` · **primary:** `cs.CR` · **Submitted:** 2026-01-26
+- **Authors:** Jiayi Zhang, Chenxin Sun, Chenxiong Qian
+- **abs:** https://arxiv.org/abs/2601.18068 · **PDF:** https://arxiv.org/pdf/2601.18068v1
+
+## 19339. EvolVE: Evolutionary Search for LLM-based Verilog Generation and Optimization
+
+- **ID:** `2601.18067` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Wei-Po Hsin, Ren-Hao Deng, Yao-Ting Hsieh, et al.
+- **abs:** https://arxiv.org/abs/2601.18067 · **PDF:** https://arxiv.org/pdf/2601.18067v1
+
+## 19340. Grounded Concreteness: Human-Like Concreteness Sensitivity in Vision-Language Models
+
+- **ID:** `2601.18065` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Aryan Roy, Zekun Wang, Christopher J. MacLellan
+- **abs:** https://arxiv.org/abs/2601.18065 · **PDF:** https://arxiv.org/pdf/2601.18065v1
+
+## 19341. Resonant Sparse Geometry Networks
+
+- **ID:** `2601.18064` · **primary:** `cs.LG` · **Submitted:** 2026-01-26
+- **Authors:** Hasi Hays
+- **abs:** https://arxiv.org/abs/2601.18064 · **PDF:** https://arxiv.org/pdf/2601.18064v1
+
+## 19342. Expert Evaluation and the Limits of Human Feedback in Mental Health AI Safety Testing
+
+- **ID:** `2601.18061` · **primary:** `cs.AI` · **Submitted:** 2026-01-26
+- **Authors:** Kiana Jafari, Paul Ulrich Nikolaus Rust, Duncan Eddy, et al.
+- **abs:** https://arxiv.org/abs/2601.18061 · **PDF:** https://arxiv.org/pdf/2601.18061v2
+
+## 19343. Neurocomputational Mechanisms of Syntactic Transfer in Bilingual Sentence Production
+
+- **ID:** `2601.18056` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Ahmet Yavuz Uluslu, Elliot Murphy
+- **abs:** https://arxiv.org/abs/2601.18056 · **PDF:** https://arxiv.org/pdf/2601.18056v1
+
+## 19344. Addressing LLM Diversity by Infusing Random Concepts
+
+- **ID:** `2601.18053` · **primary:** `cs.CL` · **Submitted:** 2026-01-26
+- **Authors:** Pulin Agrawal, Prasoon Goyal
+- **abs:** https://arxiv.org/abs/2601.18053 · **PDF:** https://arxiv.org/pdf/2601.18053v1
+
+## 19345. Laser interferometry as a robust neuromorphic platform for machine learning
+
+- **ID:** `2601.18047` · **primary:** `physics.optics` · **Submitted:** 2026-01-26
+- **Authors:** Amanuel Anteneh, Kyungeun Kim, J. M. Schwarz, et al.
+- **abs:** https://arxiv.org/abs/2601.18047 · **PDF:** https://arxiv.org/pdf/2601.18047v4
+
+## 19346. Leveraging Persistence Image to Enhance Robustness and Performance in Curvilinear Structure Segmentation
+
+- **ID:** `2601.18045` · **primary:** `cs.CV` · **Submitted:** 2026-01-25
+- **Authors:** Zhuangzhi Gao, Feixiang Zhou, He Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.18045 · **PDF:** https://arxiv.org/pdf/2601.18045v1
+
+## 19347. TabularMath: Evaluating Computational Extrapolation in Tabular Learning via Program-Verified Synthesis
+
+- **ID:** `2602.02523` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Zerui Cheng, Jiashuo Liu, Jianzhu Yao, et al.
+- **abs:** https://arxiv.org/abs/2602.02523 · **PDF:** https://arxiv.org/pdf/2602.02523v1
+
+## 19348. SpatialEmb: Extract and Encode Spatial Information for 1-Stage Multi-channel Multi-speaker ASR on Arbitrary Microphone Arrays
+
+- **ID:** `2601.18037` · **primary:** `eess.AS` · **Submitted:** 2026-01-25
+- **Authors:** Yiwen Shao, Yong Xu, Sanjeev Khudanpur, et al.
+- **abs:** https://arxiv.org/abs/2601.18037 · **PDF:** https://arxiv.org/pdf/2601.18037v1
+
+## 19349. An Experimental Comparison of Cognitive Forcing Functions for Execution Plans in AI-Assisted Writing: Effects On Trust, Overreliance, and Perceived Critical Thinking
+
+- **ID:** `2601.18033` · **primary:** `cs.HC` · **Submitted:** 2026-01-25
+- **Authors:** Ahana Ghosh, Advait Sarkar, Siân Lindley, et al.
+- **abs:** https://arxiv.org/abs/2601.18033 · **PDF:** https://arxiv.org/pdf/2601.18033v1
+
+## 19350. Multimodal Machine Learning for Soft High-k Elastomers under Data Scarcity
+
+- **ID:** `2601.18032` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Brijesh FNU, Viet Thanh Duy Nguyen, Ashima Sharma, et al.
+- **abs:** https://arxiv.org/abs/2601.18032 · **PDF:** https://arxiv.org/pdf/2601.18032v2
+
+## 19351. Spelling Bee Embeddings for Language Modeling
+
+- **ID:** `2601.18030` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Markus N. Rabe, Judith Clymo, Zheren Dong
+- **abs:** https://arxiv.org/abs/2601.18030 · **PDF:** https://arxiv.org/pdf/2601.18030v1
+
+## 19352. Sentipolis: Emotion-Aware Agents for Social Simulations
+
+- **ID:** `2601.18027` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Chiyuan Fu, Lyuhao Chen, Yunze Xiao, et al.
+- **abs:** https://arxiv.org/abs/2601.18027 · **PDF:** https://arxiv.org/pdf/2601.18027v1
+
+## 19353. CommonLID: Re-evaluating State-of-the-Art Language Identification Performance on Web Data
+
+- **ID:** `2601.18026` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Pedro Ortiz Suarez, Laurie Burchell, Catherine Arnett, et al.
+- **abs:** https://arxiv.org/abs/2601.18026 · **PDF:** https://arxiv.org/pdf/2601.18026v1
+
+## 19354. MlPET: A Localized Neural Network Approach for Probabilistic Post-Reconstruction PET Image Analysis Using Informed Priors
+
+- **ID:** `2601.18021` · **primary:** `physics.med-ph` · **Submitted:** 2026-01-25
+- **Authors:** Thomas Mejer Hansen, Nana Christensen, Mikkel Vendelbo
+- **abs:** https://arxiv.org/abs/2601.18021 · **PDF:** https://arxiv.org/pdf/2601.18021v1
+
+## 19355. A System for Name and Address Parsing with Large Language Models
+
+- **ID:** `2601.18014` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Adeeba Tarannum, Muzakkiruddin Ahmed Mohammed, Mert Can Cakmak, et al.
+- **abs:** https://arxiv.org/abs/2601.18014 · **PDF:** https://arxiv.org/pdf/2601.18014v1
+
+## 19356. RIR-Mega-Speech: A Reverberant Speech Corpus with Comprehensive Acoustic Metadata and Reproducible Evaluation
+
+- **ID:** `2601.19949` · **primary:** `eess.AS` · **Submitted:** 2026-01-25
+- **Authors:** Mandip Goswami
+- **abs:** https://arxiv.org/abs/2601.19949 · **PDF:** https://arxiv.org/pdf/2601.19949v1
+
+## 19357. Evaluating Semantic and Syntactic Understanding in Large Language Models for Payroll Systems
+
+- **ID:** `2601.18012` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Hendrika Maclean, Mert Can Cakmak, Muzakkiruddin Ahmed Mohammed, et al.
+- **abs:** https://arxiv.org/abs/2601.18012 · **PDF:** https://arxiv.org/pdf/2601.18012v2
+
+## 19358. Memory-Efficient FPGA Implementation of Stochastic Simulated Annealing
+
+- **ID:** `2601.18007` · **primary:** `cs.AR` · **Submitted:** 2026-01-25
+- **Authors:** Duckgyu Shin, Naoya Onizawa, Warren J. Gross, et al.
+- **abs:** https://arxiv.org/abs/2601.18007 · **PDF:** https://arxiv.org/pdf/2601.18007v1
+
+## 19359. PEAR: Pairwise Evaluation for Automatic Relative Scoring in Machine Translation
+
+- **ID:** `2601.18006` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Lorenzo Proietti, Roman Grundkiewicz, Matt Post
+- **abs:** https://arxiv.org/abs/2601.18006 · **PDF:** https://arxiv.org/pdf/2601.18006v1
+
+## 19360. Flow-based Extremal Mathematical Structure Discovery
+
+- **ID:** `2601.18005` · **primary:** `math.CO` · **Submitted:** 2026-01-25
+- **Authors:** Gergely Bérczi, Baran Hashemi, Jonas Klüver
+- **abs:** https://arxiv.org/abs/2601.18005 · **PDF:** https://arxiv.org/pdf/2601.18005v1
+
+## 19361. Time series forecasting with Hahn Kolmogorov-Arnold networks
+
+- **ID:** `2601.18837` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Md Zahidul Hasan, A. Ben Hamza, Nizar Bouguila
+- **abs:** https://arxiv.org/abs/2601.18837 · **PDF:** https://arxiv.org/pdf/2601.18837v2
+
+## 19362. IMU-1: Sample-Efficient Pre-training of Small Language Models
+
+- **ID:** `2602.02522` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** George Grigorev
+- **abs:** https://arxiv.org/abs/2602.02522 · **PDF:** https://arxiv.org/pdf/2602.02522v1
+
+## 19363. Scaled Dot-Product Attention implements projection of inputs onto a common surface
+
+- **ID:** `2602.02521` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Terence D Sanger
+- **abs:** https://arxiv.org/abs/2602.02521 · **PDF:** https://arxiv.org/pdf/2602.02521v1
+
+## 19364. Coding-Enforced Resilient and Secure Aggregation for Hierarchical Federated Learning
+
+- **ID:** `2601.17995` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Shudi Weng, Ming Xiao, Mikael Skoglund
+- **abs:** https://arxiv.org/abs/2601.17995 · **PDF:** https://arxiv.org/pdf/2601.17995v1
+
+## 19365. AI-based approach to burnout identification from textual data
+
+- **ID:** `2601.17993` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Marina Zavertiaeva, Petr Parshakov, Mikhail Usanin, et al.
+- **abs:** https://arxiv.org/abs/2601.17993 · **PDF:** https://arxiv.org/pdf/2601.17993v1
+
+## 19366. A Cherry-Picking Approach to Large Load Shaping for More Effective Carbon Reduction
+
+- **ID:** `2601.17990` · **primary:** `stat.ML` · **Submitted:** 2026-01-25
+- **Authors:** Bokan Chen, Raiden Hasegawa, Adriaan Hilbers, et al.
+- **abs:** https://arxiv.org/abs/2601.17990 · **PDF:** https://arxiv.org/pdf/2601.17990v1
+
+## 19367. Systematic Characterization of Minimal Deep Learning Architectures: A Unified Analysis of Convergence, Pruning, and Quantization
+
+- **ID:** `2601.17987` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Ziwei Zheng, Huizhi Liang, Vaclav Snasel, et al.
+- **abs:** https://arxiv.org/abs/2601.17987 · **PDF:** https://arxiv.org/pdf/2601.17987v1
+
+## 19368. CanaryBench: Stress Testing Privacy Leakage in Cluster-Level Conversation Summaries
+
+- **ID:** `2601.18834` · **primary:** `cs.CR` · **Submitted:** 2026-01-25
+- **Authors:** Deep Mehta
+- **abs:** https://arxiv.org/abs/2601.18834 · **PDF:** https://arxiv.org/pdf/2601.18834v1
+
+## 19369. Federated learning for unpaired multimodal data through a homogeneous transformer model
+
+- **ID:** `2601.17986` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Anders Eklund
+- **abs:** https://arxiv.org/abs/2601.17986 · **PDF:** https://arxiv.org/pdf/2601.17986v1
+
+## 19370. SD-E$^2$: Semantic Exploration for Reasoning Under Token Budgets
+
+- **ID:** `2601.17982` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Kshitij Mishra, Nils Lukas, Salem Lahlou
+- **abs:** https://arxiv.org/abs/2601.17982 · **PDF:** https://arxiv.org/pdf/2601.17982v1
+
+## 19371. Agentic Business Process Management Systems
+
+- **ID:** `2601.18833` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Marlon Dumas, Fredrik Milani, David Chapela-Campa
+- **abs:** https://arxiv.org/abs/2601.18833 · **PDF:** https://arxiv.org/pdf/2601.18833v1
+
+## 19372. Boosting methods for interval-censored data with regression and classification
+
+- **ID:** `2601.17973` · **primary:** `stat.ML` · **Submitted:** 2026-01-25
+- **Authors:** Yuan Bian, Grace Y. Yi, Wenqing He
+- **abs:** https://arxiv.org/abs/2601.17973 · **PDF:** https://arxiv.org/pdf/2601.17973v2
+
+## 19373. LLMs as Cultural Archives: Cultural Commonsense Knowledge Graph Extraction
+
+- **ID:** `2601.17971` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Junior Cedric Tonga, Chen Cecilia Liu, Iryna Gurevych, et al.
+- **abs:** https://arxiv.org/abs/2601.17971 · **PDF:** https://arxiv.org/pdf/2601.17971v1
+
+## 19374. HYPE-EDIT-1: Benchmark for Measuring Reliability in Frontier Image Editing Models
+
+- **ID:** `2602.00105` · **primary:** `cs.CV` · **Submitted:** 2026-01-25
+- **Authors:** Wing Chan, Richard Allen
+- **abs:** https://arxiv.org/abs/2602.00105 · **PDF:** https://arxiv.org/pdf/2602.00105v1
+
+## 19375. TensorLens: End-to-End Transformer Analysis via High-Order Attention Tensors
+
+- **ID:** `2601.17958` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Ido Andrew Atad, Itamar Zimerman, Shahar Katz, et al.
+- **abs:** https://arxiv.org/abs/2601.17958 · **PDF:** https://arxiv.org/pdf/2601.17958v1
+
+## 19376. Scaling Effects and Uncertainty Quantification in Neural Actor Critic Algorithms
+
+- **ID:** `2601.17954` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Nikos Georgoudios, Konstantinos Spiliopoulos, Justin Sirignano
+- **abs:** https://arxiv.org/abs/2601.17954 · **PDF:** https://arxiv.org/pdf/2601.17954v1
+
+## 19377. A Monosemantic Attribution Framework for Stable Interpretability in Clinical Neuroscience Large Language Models
+
+- **ID:** `2601.17952` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Michail Mamalakis, Tiago Azevedo, Cristian Cosentino, et al.
+- **abs:** https://arxiv.org/abs/2601.17952 · **PDF:** https://arxiv.org/pdf/2601.17952v1
+
+## 19378. Credit Fairness: Online Fairness In Shared Resource Pools
+
+- **ID:** `2601.17944` · **primary:** `cs.GT` · **Submitted:** 2026-01-25
+- **Authors:** Seyed Majid Zahedi, Rupert Freeman
+- **abs:** https://arxiv.org/abs/2601.17944 · **PDF:** https://arxiv.org/pdf/2601.17944v1
+
+## 19379. LLM-Based SQL Generation: Prompting, Self-Refinement, and Adaptive Weighted Majority Voting
+
+- **ID:** `2601.17942` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Yu-Jie Yang, Hung-Fu Chang, Po-An Chen
+- **abs:** https://arxiv.org/abs/2601.17942 · **PDF:** https://arxiv.org/pdf/2601.17942v1
+
+## 19380. Energy-Aware Reinforcement Learning for Robotic Manipulation of Articulated Components in Infrastructure Operation and Maintenance
+
+- **ID:** `2602.12288` · **primary:** `eess.SY` · **Submitted:** 2026-01-25
+- **Authors:** Xiaowen Tao, Yinuo Wang, Haitao Ding, et al.
+- **abs:** https://arxiv.org/abs/2602.12288 · **PDF:** https://arxiv.org/pdf/2602.12288v3
+
+## 19381. The Geometric Reasoner: Manifold-Informed Latent Foresight Search for Long-Context Reasoning
+
+- **ID:** `2601.18832` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Ren Zhuang, Ben Wang, Shuifa Sun
+- **abs:** https://arxiv.org/abs/2601.18832 · **PDF:** https://arxiv.org/pdf/2601.18832v2
+
+## 19382. FedGraph-VASP: Privacy-Preserving Federated Graph Learning with Post-Quantum Security for Cross-Institutional Anti-Money Laundering
+
+- **ID:** `2601.17935` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Daniel Commey, Matilda Nkoom, Yousef Alsenani, et al.
+- **abs:** https://arxiv.org/abs/2601.17935 · **PDF:** https://arxiv.org/pdf/2601.17935v1
+
+## 19383. From Specialist to Generalist: Unlocking SAM's Learning Potential on Unlabeled Medical Images
+
+- **ID:** `2601.17934` · **primary:** `cs.CV` · **Submitted:** 2026-01-25
+- **Authors:** Vi Vu, Thanh-Huy Nguyen, Tien-Thinh Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.17934 · **PDF:** https://arxiv.org/pdf/2601.17934v2
+
+## 19384. Dissipative Learning: A Framework for Viable Adaptive Systems
+
+- **ID:** `2601.17933` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Laurent Caraffa
+- **abs:** https://arxiv.org/abs/2601.17933 · **PDF:** https://arxiv.org/pdf/2601.17933v1
+
+## 19385. Learning Transferable Skills in Action RPGs via Directed Skill Graphs and Selective Adaptation
+
+- **ID:** `2601.17923` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Ali Najar
+- **abs:** https://arxiv.org/abs/2601.17923 · **PDF:** https://arxiv.org/pdf/2601.17923v2
+
+## 19386. ShapLoRA: Allocation of Low-rank Adaption on Large Language Models via Shapley Value Inspired Importance Estimation
+
+- **ID:** `2601.17921` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Yi Zhao, Qinghua Yao, Xinyuan song, et al.
+- **abs:** https://arxiv.org/abs/2601.17921 · **PDF:** https://arxiv.org/pdf/2601.17921v1
+
+## 19387. Agentic AI for Self-Driving Laboratories in Soft Matter: Taxonomy, Benchmarks,and Open Challenges
+
+- **ID:** `2601.17920` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Xuanzhou Chen, Audrey Wang, Stanley Yin, et al.
+- **abs:** https://arxiv.org/abs/2601.17920 · **PDF:** https://arxiv.org/pdf/2601.17920v1
+
+## 19388. State Design Matters: How Representations Shape Dynamic Reasoning in Large Language Models
+
+- **ID:** `2602.15858` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Annie Wong, Aske Plaat, Thomas Bäck, et al.
+- **abs:** https://arxiv.org/abs/2602.15858 · **PDF:** https://arxiv.org/pdf/2602.15858v1
+
+## 19389. Benchmarking Direct Preference Optimization for Medical Large Vision-Language Models
+
+- **ID:** `2601.17918` · **primary:** `cs.CV` · **Submitted:** 2026-01-25
+- **Authors:** Dain Kim, Jiwoo Lee, Jaehoon Yun, et al.
+- **abs:** https://arxiv.org/abs/2601.17918 · **PDF:** https://arxiv.org/pdf/2601.17918v1
+
+## 19390. Streaming-dLLM: Accelerating Diffusion LLMs via Suffix Pruning and Dynamic Decoding
+
+- **ID:** `2601.17917` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Zhongyu Xiao, Zhiwei Hao, Jianyuan Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.17917 · **PDF:** https://arxiv.org/pdf/2601.17917v2
+
+## 19391. UniPACT: A Multimodal Framework for Prognostic Question Answering on Raw ECG and Structured EHR
+
+- **ID:** `2601.17916` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Jialu Tang, Tong Xia, Yuan Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.17916 · **PDF:** https://arxiv.org/pdf/2601.17916v1
+
+## 19392. How Much Temporal Modeling is Enough? A Systematic Study of Hybrid CNN-RNN Architectures for Multi-Label ECG Classification
+
+- **ID:** `2601.18830` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Alireza Jafari, Fatemeh Jafari
+- **abs:** https://arxiv.org/abs/2601.18830 · **PDF:** https://arxiv.org/pdf/2601.18830v1
+
+## 19393. Think Locally, Explain Globally: Graph-Guided LLM Investigations via Local Reasoning and Belief Propagation
+
+- **ID:** `2601.17915` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Saurabh Jha, Rohan Arora, Bhavya, et al.
+- **abs:** https://arxiv.org/abs/2601.17915 · **PDF:** https://arxiv.org/pdf/2601.17915v2
+
+## 19394. Causal Pre-training Under the Fairness Lens: An Empirical Study of TabPFN
+
+- **ID:** `2601.17912` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Qinyi Liu, Mohammad Khalil, Naman Goel
+- **abs:** https://arxiv.org/abs/2601.17912 · **PDF:** https://arxiv.org/pdf/2601.17912v2
+
+## 19395. Adaptive Weighting in Knowledge Distillation: An Axiomatic Framework for Multi-Scale Teacher Ensemble Optimization
+
+- **ID:** `2601.17910` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Aaron R. Flouro, Shawn P. Chadwick
+- **abs:** https://arxiv.org/abs/2601.17910 · **PDF:** https://arxiv.org/pdf/2601.17910v1
+
+## 19396. FARM: Few-shot Adaptive Malware Family Classification under Concept Drift
+
+- **ID:** `2601.17907` · **primary:** `cs.CR` · **Submitted:** 2026-01-25
+- **Authors:** Numan Halit Guldemir, Oluwafemi Olukoya, Jesús Martínez-del-Rincón
+- **abs:** https://arxiv.org/abs/2601.17907 · **PDF:** https://arxiv.org/pdf/2601.17907v2
+
+## 19397. Feature-Space Generative Models for One-Shot Class-Incremental Learning
+
+- **ID:** `2601.17905` · **primary:** `cs.CV` · **Submitted:** 2026-01-25
+- **Authors:** Jack Foster, Kirill Paramonov, Mete Ozay, et al.
+- **abs:** https://arxiv.org/abs/2601.17905 · **PDF:** https://arxiv.org/pdf/2601.17905v1
+
+## 19398. Evolving Interdependent Operators with Large Language Models for Multi-Objective Combinatorial Optimization
+
+- **ID:** `2601.17899` · **primary:** `cs.NE` · **Submitted:** 2026-01-25
+- **Authors:** Junhao Qiu, Xin Chen, Liang Ge, et al.
+- **abs:** https://arxiv.org/abs/2601.17899 · **PDF:** https://arxiv.org/pdf/2601.17899v2
+
+## 19399. Assessment of Generative Named Entity Recognition in the Era of Large Language Models
+
+- **ID:** `2601.17898` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Qi Zhan, Yile Wang, Hui Huang
+- **abs:** https://arxiv.org/abs/2601.17898 · **PDF:** https://arxiv.org/pdf/2601.17898v1
+
+## 19400. UniCog: Uncovering Cognitive Abilities of LLMs through Latent Mind Space Analysis
+
+- **ID:** `2601.17897` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Jiayu Liu, Yinhe Long, Zhenya Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.17897 · **PDF:** https://arxiv.org/pdf/2601.17897v1
+
+## 19401. Artificial Intelligence and Intellectual Property Rights: Comparative Transnational Policy Analysis
+
+- **ID:** `2601.17892` · **primary:** `cs.CY` · **Submitted:** 2026-01-25
+- **Authors:** Sahibpreet Singh, Manjit Singh
+- **abs:** https://arxiv.org/abs/2601.17892 · **PDF:** https://arxiv.org/pdf/2601.17892v1
+
+## 19402. When Personalization Legitimizes Risks: Uncovering Safety Vulnerabilities in Personalized Dialogue Agents
+
+- **ID:** `2601.17887` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Jiahe Guo, Xiangran Guo, Yulin Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.17887 · **PDF:** https://arxiv.org/pdf/2601.17887v1
+
+## 19403. CP Loss: Channel-wise Perceptual Loss for Time Series Forecasting
+
+- **ID:** `2601.18829` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Yaohua Zha, Chunlin Fan, Peiyuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.18829 · **PDF:** https://arxiv.org/pdf/2601.18829v1
+
+## 19404. PEAfowl: Perception-Enhanced Multi-View Vision-Language-Action for Bimanual Manipulation
+
+- **ID:** `2601.17885` · **primary:** `cs.CV` · **Submitted:** 2026-01-25
+- **Authors:** Qingyu Fan, Zhaoxiang Li, Yi Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.17885 · **PDF:** https://arxiv.org/pdf/2601.17885v1
+
+## 19405. EEG Foundation Models: Progresses, Benchmarking, and Open Problems
+
+- **ID:** `2601.17883` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Dingkun Liu, Yuheng Chen, Zhu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.17883 · **PDF:** https://arxiv.org/pdf/2601.17883v2
+
+## 19406. IPBC: An Interactive Projection-Based Framework for Human-in-the-Loop Semi-Supervised Clustering of High-Dimensional Data
+
+- **ID:** `2601.18828` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Mohammad Zare
+- **abs:** https://arxiv.org/abs/2601.18828 · **PDF:** https://arxiv.org/pdf/2601.18828v1
+
+## 19407. Self-Manager: Parallel Agent Loop for Long-form Deep Research
+
+- **ID:** `2601.17879` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Yilong Xu, Zhi Zheng, Xiang Long, et al.
+- **abs:** https://arxiv.org/abs/2601.17879 · **PDF:** https://arxiv.org/pdf/2601.17879v1
+
+## 19408. Comparative Algorithmic Governance of Public Health Instruments across India, EU, US and LMICs
+
+- **ID:** `2601.17877` · **primary:** `cs.CY` · **Submitted:** 2026-01-25
+- **Authors:** Sahibpreet Singh
+- **abs:** https://arxiv.org/abs/2601.17877 · **PDF:** https://arxiv.org/pdf/2601.17877v1
+
+## 19409. On the Emergence and Test-Time Use of Structural Information in Large Language Models
+
+- **ID:** `2601.17869` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Michelle Chao Chen, Moritz Miller, Bernhard Schölkopf, et al.
+- **abs:** https://arxiv.org/abs/2601.17869 · **PDF:** https://arxiv.org/pdf/2601.17869v1
+
+## 19410. VidLaDA: Bidirectional Diffusion Large Language Models for Efficient Video Understanding
+
+- **ID:** `2601.17868` · **primary:** `cs.CV` · **Submitted:** 2026-01-25
+- **Authors:** Zhihao He, Tieyuan Chen, Kangyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.17868 · **PDF:** https://arxiv.org/pdf/2601.17868v2
+
+## 19411. D-Models and E-Models: Diversity-Stability Trade-offs in the Sampling Behavior of Large Language Models
+
+- **ID:** `2601.17865` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Jia Gu, Liang Pang, Huawei Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.17865 · **PDF:** https://arxiv.org/pdf/2601.17865v2
+
+## 19412. MergeMix: Optimizing Mid-Training Data Mixtures via Learnable Model Merging
+
+- **ID:** `2601.17858` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Jiapeng Wang, Changxin Tian, Kunlong Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.17858 · **PDF:** https://arxiv.org/pdf/2601.17858v1
+
+## 19413. A Universal Load Balancing Principle and Its Application to Large Language Model Serving
+
+- **ID:** `2601.17855` · **primary:** `cs.DC` · **Submitted:** 2026-01-25
+- **Authors:** Zixi Chen, Tianci Bu, Chendong Song, et al.
+- **abs:** https://arxiv.org/abs/2601.17855 · **PDF:** https://arxiv.org/pdf/2601.17855v2
+
+## 19414. RAICL: Retrieval-Augmented In-Context Learning for Vision-Language-Model Based EEG Seizure Detection
+
+- **ID:** `2601.17844` · **primary:** `cs.HC` · **Submitted:** 2026-01-25
+- **Authors:** Siyang Li, Zhuoya Wang, Xiyan Gui, et al.
+- **abs:** https://arxiv.org/abs/2601.17844 · **PDF:** https://arxiv.org/pdf/2601.17844v1
+
+## 19415. EFT-CoT: A Multi-Agent Chain-of-Thought Framework for Emotion-Focused Therapy
+
+- **ID:** `2601.17842` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Lanqing Du, Yunong Li, YuJie Long, et al.
+- **abs:** https://arxiv.org/abs/2601.17842 · **PDF:** https://arxiv.org/pdf/2601.17842v2
+
+## 19416. Linguistic and Argument Diversity in Synthetic Data for Function-Calling Agents
+
+- **ID:** `2601.17829` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Dan Greenstein, Zohar Karnin, Chen Amiraz, et al.
+- **abs:** https://arxiv.org/abs/2601.17829 · **PDF:** https://arxiv.org/pdf/2601.17829v1
+
+## 19417. Aligning Medical Conversational AI through Online Reinforcement Learning with Information-Theoretic Rewards
+
+- **ID:** `2601.17828` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Tanvi Verma, Yang Zhou, Rick Siow Mong Goh, et al.
+- **abs:** https://arxiv.org/abs/2601.17828 · **PDF:** https://arxiv.org/pdf/2601.17828v1
+
+## 19418. RegGuard: AI-Powered Retrieval-Enhanced Assistant for Pharmaceutical Regulatory Compliance
+
+- **ID:** `2601.17826` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Siyuan Yang, Xihan Bian, Jiayin Tang
+- **abs:** https://arxiv.org/abs/2601.17826 · **PDF:** https://arxiv.org/pdf/2601.17826v1
+
+## 19419. DIETA: A Decoder-only transformer-based model for Italian-English machine TrAnslation
+
+- **ID:** `2601.17823` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Pranav Kasela, Marco Braga, Alessandro Ghiotto, et al.
+- **abs:** https://arxiv.org/abs/2601.17823 · **PDF:** https://arxiv.org/pdf/2601.17823v1
+
+## 19420. MMR-Bench: A Comprehensive Benchmark for Multimodal LLM Routing
+
+- **ID:** `2601.17814` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Haoxuan Ma, Guannan Lai, Han-Jia Ye
+- **abs:** https://arxiv.org/abs/2601.17814 · **PDF:** https://arxiv.org/pdf/2601.17814v1
+
+## 19421. R3G: A Reasoning--Retrieval--Reranking Framework for Vision-Centric Answer Generation
+
+- **ID:** `2602.00104` · **primary:** `cs.CV` · **Submitted:** 2026-01-25
+- **Authors:** Zhuohong Chen, Zhengxian Wu, Zirui Liao, et al.
+- **abs:** https://arxiv.org/abs/2602.00104 · **PDF:** https://arxiv.org/pdf/2602.00104v1
+
+## 19422. Motif Diversity in Human Liver ChIP-seq Data Using MAP-Elites
+
+- **ID:** `2601.17808` · **primary:** `cs.NE` · **Submitted:** 2026-01-25
+- **Authors:** Alejandro Medina, Mary Lauren Benton
+- **abs:** https://arxiv.org/abs/2601.17808 · **PDF:** https://arxiv.org/pdf/2601.17808v1
+
+## 19423. An autonomous living database for perovskite photovoltaics
+
+- **ID:** `2601.17807` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-25
+- **Authors:** Sherjeel Shabih, Hampus Näsström, Sharat Patil, et al.
+- **abs:** https://arxiv.org/abs/2601.17807 · **PDF:** https://arxiv.org/pdf/2601.17807v1
+
+## 19424. Automated structural testing of LLM-based agents: methods, framework, and case studies
+
+- **ID:** `2601.18827` · **primary:** `cs.SE` · **Submitted:** 2026-01-25
+- **Authors:** Jens Kohl, Otto Kruse, Youssef Mostafa, et al.
+- **abs:** https://arxiv.org/abs/2601.18827 · **PDF:** https://arxiv.org/pdf/2601.18827v1
+
+## 19425. Robust Computational Extraction of Non-Enhancing Hypercellular Tumor Regions from Clinical Imaging Data
+
+- **ID:** `2601.17802` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** A. Brawanski, Th. Schaffer, F. Raab, et al.
+- **abs:** https://arxiv.org/abs/2601.17802 · **PDF:** https://arxiv.org/pdf/2601.17802v1
+
+## 19426. Differentiable Integer Linear Programming is not Differentiable & it's not a mere technical problem
+
+- **ID:** `2601.17800` · **primary:** `math.OC` · **Submitted:** 2026-01-25
+- **Authors:** Thanawat Sornwanee
+- **abs:** https://arxiv.org/abs/2601.17800 · **PDF:** https://arxiv.org/pdf/2601.17800v2
+
+## 19427. Multi-source Heterogeneous Public Opinion Analysis via Collaborative Reasoning and Adaptive Fusion: A Systematically Integrated Approach
+
+- **ID:** `2602.15857` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Yi Liu
+- **abs:** https://arxiv.org/abs/2602.15857 · **PDF:** https://arxiv.org/pdf/2602.15857v1
+
+## 19428. Neuro-Symbolic Verification on Instruction Following of LLMs
+
+- **ID:** `2601.17789` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Yiming Su, Kunzhao Xu, Yanjie Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.17789 · **PDF:** https://arxiv.org/pdf/2601.17789v1
+
+## 19429. Beyond a Single Perspective: Text Anomaly Detection with Multi-View Language Representations
+
+- **ID:** `2601.17786` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Yixin Liu, Kehan Yan, Shiyuan Li, et al.
+- **abs:** https://arxiv.org/abs/2601.17786 · **PDF:** https://arxiv.org/pdf/2601.17786v1
+
+## 19430. Shortcut Learning in Binary Classifier Black Boxes: Applications to Voice Anti-Spoofing and Biometrics
+
+- **ID:** `2601.17782` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Md Sahidullah, Hye-jin Shim, Rosa Gonzalez Hautamäki, et al.
+- **abs:** https://arxiv.org/abs/2601.17782 · **PDF:** https://arxiv.org/pdf/2601.17782v1
+
+## 19431. Controlling Reading Ease with Gaze-Guided Text Generation
+
+- **ID:** `2601.17781` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Andreas Säuberli, Darja Jepifanova, Diego Frassinelli, et al.
+- **abs:** https://arxiv.org/abs/2601.17781 · **PDF:** https://arxiv.org/pdf/2601.17781v1
+
+## 19432. DPI: Exploiting Parameter Heterogeneity for Interference-Free Fine-Tuning
+
+- **ID:** `2601.17777` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Xiaoyu Liu, Xiaoyu Guan, Di Liang, et al.
+- **abs:** https://arxiv.org/abs/2601.17777 · **PDF:** https://arxiv.org/pdf/2601.17777v1
+
+## 19433. MarketGANs: Multivariate financial time-series data augmentation using generative adversarial networks
+
+- **ID:** `2601.17773` · **primary:** `q-fin.ST` · **Submitted:** 2026-01-25
+- **Authors:** Jeonggyu Huh, Seungwon Jeong, Hyun-Gyoon Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.17773 · **PDF:** https://arxiv.org/pdf/2601.17773v1
+
+## 19434. Context-Aware Iterative Token Detection and Masked Transmission for Wireless Token Communication
+
+- **ID:** `2601.17770` · **primary:** `eess.SP` · **Submitted:** 2026-01-25
+- **Authors:** Junyong Shin, Joohyuk Park, Jihong Park, et al.
+- **abs:** https://arxiv.org/abs/2601.17770 · **PDF:** https://arxiv.org/pdf/2601.17770v1
+
+## 19435. LLM-42: Enabling Determinism in LLM Inference with Verified Speculation
+
+- **ID:** `2601.17768` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Raja Gond, Aditya K Kamath, Ramachandran Ramjee, et al.
+- **abs:** https://arxiv.org/abs/2601.17768 · **PDF:** https://arxiv.org/pdf/2601.17768v2
+
+## 19436. HyCARD-Net: A Synergistic Hybrid Intelligence Framework for Cardiovascular Disease Diagnosis
+
+- **ID:** `2601.17767` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Rajan Das Gupta, Xiaobin Wu, Xun Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.17767 · **PDF:** https://arxiv.org/pdf/2601.17767v1
+
+## 19437. Cross-Lingual Probing and Community-Grounded Analysis of Gender Bias in Low-Resource Bengali
+
+- **ID:** `2601.17764` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Md Asgor Hossain Reaj, Rajan Das Gupta, Jui Saha Pritha, et al.
+- **abs:** https://arxiv.org/abs/2601.17764 · **PDF:** https://arxiv.org/pdf/2601.17764v1
+
+## 19438. AR-Omni: A Unified Autoregressive Model for Any-to-Any Generation
+
+- **ID:** `2601.17761` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Dongjie Cheng, Ruifeng Yuan, Yongqi Li, et al.
+- **abs:** https://arxiv.org/abs/2601.17761 · **PDF:** https://arxiv.org/pdf/2601.17761v1
+
+## 19439. Rethinking Soft Compression in Retrieval-Augmented Generation: A Query-Conditioned Selector Perspective
+
+- **ID:** `2602.15856` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Yunhao Liu, Zian Jia, Xinyu Gao, et al.
+- **abs:** https://arxiv.org/abs/2602.15856 · **PDF:** https://arxiv.org/pdf/2602.15856v2
+
+## 19440. MV-S2V: Multi-View Subject-Consistent Video Generation
+
+- **ID:** `2601.17756` · **primary:** `cs.CV` · **Submitted:** 2026-01-25
+- **Authors:** Ziyang Song, Xinyu Gong, Bangya Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.17756 · **PDF:** https://arxiv.org/pdf/2601.17756v2
+
+## 19441. ProGraph-R1: Progress-aware Reinforcement Learning for Graph Retrieval Augmented Generation
+
+- **ID:** `2601.17755` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Jinyoung Park, Sanghyeok Lee, Omar Zia Khan, et al.
+- **abs:** https://arxiv.org/abs/2601.17755 · **PDF:** https://arxiv.org/pdf/2601.17755v1
+
+## 19442. Hylog: A Hybrid Approach to Logging Text Production in Non-alphabetic Scripts
+
+- **ID:** `2601.17753` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Roberto Crotti, Giovanni Denaro, Zhiqiang Du, et al.
+- **abs:** https://arxiv.org/abs/2601.17753 · **PDF:** https://arxiv.org/pdf/2601.17753v1
+
+## 19443. Over-The-Air Extreme Learning Machines with XL Reception via Nonlinear Cascaded Metasurfaces
+
+- **ID:** `2601.17749` · **primary:** `eess.SP` · **Submitted:** 2026-01-25
+- **Authors:** Kyriakos Stylianopoulos, Mattia Fabiani, Giulia Torcolacci, et al.
+- **abs:** https://arxiv.org/abs/2601.17749 · **PDF:** https://arxiv.org/pdf/2601.17749v1
+
+## 19444. Faramesh: A Protocol-Agnostic Execution Control Plane for Autonomous Agent Systems
+
+- **ID:** `2601.17744` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Amjad Fatmi
+- **abs:** https://arxiv.org/abs/2601.17744 · **PDF:** https://arxiv.org/pdf/2601.17744v1
+
+## 19445. The Script is All You Need: An Agentic Framework for Long-Horizon Dialogue-to-Cinematic Video Generation
+
+- **ID:** `2601.17737` · **primary:** `cs.CV` · **Submitted:** 2026-01-25
+- **Authors:** Chenyu Mu, Xin He, Qu Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.17737 · **PDF:** https://arxiv.org/pdf/2601.17737v2
+
+## 19446. Athanor: Authoring Action Modification-based Interactions on Static Visualizations via Natural Language
+
+- **ID:** `2601.17736` · **primary:** `cs.HC` · **Submitted:** 2026-01-25
+- **Authors:** Can Liu, Jaeuk Lee, Tianhe Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.17736 · **PDF:** https://arxiv.org/pdf/2601.17736v1
+
+## 19447. ReFuGe: Feature Generation for Prediction Tasks on Relational Databases with LLM Agents
+
+- **ID:** `2601.17735` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Kyungho Kim, Geon Lee, Juyeon Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.17735 · **PDF:** https://arxiv.org/pdf/2601.17735v1
+
+## 19448. Unsupervised Elicitation of Moral Values from Language Models
+
+- **ID:** `2601.17728` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Meysam Alizadeh, Fabrizio Gilardi, Zeynab Samei
+- **abs:** https://arxiv.org/abs/2601.17728 · **PDF:** https://arxiv.org/pdf/2601.17728v1
+
+## 19449. EntWorld: A Holistic Environment and Benchmark for Verifiable Enterprise GUI Agents
+
+- **ID:** `2601.17722` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Ying Mo, Yu Bai, Dapeng Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.17722 · **PDF:** https://arxiv.org/pdf/2601.17722v1
+
+## 19450. The LLM Data Auditor: A Metric-oriented Survey on Quality and Trustworthiness in Evaluating Synthetic Data
+
+- **ID:** `2601.17717` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Kaituo Zhang, Mingzhi Hu, Hoang Anh Duy Le, et al.
+- **abs:** https://arxiv.org/abs/2601.17717 · **PDF:** https://arxiv.org/pdf/2601.17717v2
+
+## 19451. Do Reasoning Models Ask Better Questions? A Formal Information-Theoretic Analysis on Multi-Turn LLM Games
+
+- **ID:** `2601.17716` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Daniel M. Pedrozo, Telma W. de L. Soares, Bryan L. M. de Oliveira
+- **abs:** https://arxiv.org/abs/2601.17716 · **PDF:** https://arxiv.org/pdf/2601.17716v1
+
+## 19452. FedCCA: Client-Centric Adaptation against Data Heterogeneity in Federated Learning on IoT Devices
+
+- **ID:** `2601.17713` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Kaile Wang, Jiannong Cao, Yu Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.17713 · **PDF:** https://arxiv.org/pdf/2601.17713v1
+
+## 19453. CaSNet: Compress-and-Send Network Based Multi-Device Speech Enhancement Model for Distributed Microphone Arrays
+
+- **ID:** `2601.17711` · **primary:** `cs.SD` · **Submitted:** 2026-01-25
+- **Authors:** Chengqian Jiang, Jie Zhang, Haoyin Yan
+- **abs:** https://arxiv.org/abs/2601.17711 · **PDF:** https://arxiv.org/pdf/2601.17711v1
+
+## 19454. A Computational Approach to Visual Metonymy
+
+- **ID:** `2601.17706` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Saptarshi Ghosh, Linfeng Liu, Tianyu Jiang
+- **abs:** https://arxiv.org/abs/2601.17706 · **PDF:** https://arxiv.org/pdf/2601.17706v1
+
+## 19455. Distance-to-Distance Ratio: A Similarity Measure for Sentences Based on Rate of Change in LLM Embeddings
+
+- **ID:** `2601.17705` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Abdullah Qureshi, Kenneth Rice, Alexander Wolpert
+- **abs:** https://arxiv.org/abs/2601.17705 · **PDF:** https://arxiv.org/pdf/2601.17705v1
+
+## 19456. S$^3$-Attention:Attention-Aligned Endogenous Retrieval for Memory-Bounded Long-Context Inference
+
+- **ID:** `2601.17702` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Qingsen Ma, Dianyun Wang, Yaoye Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.17702 · **PDF:** https://arxiv.org/pdf/2601.17702v2
+
+## 19457. SQL-Trail: Multi-Turn Reinforcement Learning with Interleaved Feedback for Text-to-SQL
+
+- **ID:** `2601.17699` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Harper Hua, Zhen Han, Zhengyuan Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.17699 · **PDF:** https://arxiv.org/pdf/2601.17699v1
+
+## 19458. LegalMALR:Multi-Agent Query Understanding and LLM-Based Reranking for Chinese Statute Retrieval
+
+- **ID:** `2601.17692` · **primary:** `cs.IR` · **Submitted:** 2026-01-25
+- **Authors:** Yunhan Li, Mingjie Xie, Gaoli Kang, et al.
+- **abs:** https://arxiv.org/abs/2601.17692 · **PDF:** https://arxiv.org/pdf/2601.17692v1
+
+## 19459. Segment Length Matters: A Study of Segment Lengths on Audio Fingerprinting Performance
+
+- **ID:** `2601.17690` · **primary:** `cs.SD` · **Submitted:** 2026-01-25
+- **Authors:** Ziling Gong, Yunyan Ouyang, Iram Kamdar, et al.
+- **abs:** https://arxiv.org/abs/2601.17690 · **PDF:** https://arxiv.org/pdf/2601.17690v1
+
+## 19460. REV-INR: Regularized Evidential Implicit Neural Representation for Uncertainty-Aware Volume Visualization
+
+- **ID:** `2601.17689` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Shanu Saklani, Tushar M. Athawale, Nairita Pal, et al.
+- **abs:** https://arxiv.org/abs/2601.17689 · **PDF:** https://arxiv.org/pdf/2601.17689v1
+
+## 19461. Agentic reinforcement learning empowers next-generation chemical language models for molecular design and synthesis
+
+- **ID:** `2601.17687` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Hao Li, He Cao, Shenyao Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.17687 · **PDF:** https://arxiv.org/pdf/2601.17687v2
+
+## 19462. A Model-Driven Lossless Compression Algorithm Resistant to Mismatch
+
+- **ID:** `2601.17684` · **primary:** `cs.IT` · **Submitted:** 2026-01-25
+- **Authors:** Cordelia Hu, Jennifer Tang
+- **abs:** https://arxiv.org/abs/2601.17684 · **PDF:** https://arxiv.org/pdf/2601.17684v1
+
+## 19463. Differential Voting: Loss Functions For Axiomatically Diverse Aggregation of Heterogeneous Preferences
+
+- **ID:** `2601.18824` · **primary:** `cs.GT` · **Submitted:** 2026-01-25
+- **Authors:** Zhiyu An, Duaa Nakshbandi, Wan Du
+- **abs:** https://arxiv.org/abs/2601.18824 · **PDF:** https://arxiv.org/pdf/2601.18824v1
+
+## 19464. $\infty$-MoE: Generalizing Mixture of Experts to Infinite Experts
+
+- **ID:** `2601.17680` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Shota Takashiro, Takeshi Kojima, Shohei Taniguchi, et al.
+- **abs:** https://arxiv.org/abs/2601.17680 · **PDF:** https://arxiv.org/pdf/2601.17680v1
+
+## 19465. BanglaRobustNet: A Hybrid Denoising-Attention Architecture for Robust Bangla Speech Recognition
+
+- **ID:** `2601.17679` · **primary:** `cs.SD` · **Submitted:** 2026-01-25
+- **Authors:** Md Sazzadul Islam Ridoy, Mubaswira Ibnat Zidney, Sumi Akter, et al.
+- **abs:** https://arxiv.org/abs/2601.17679 · **PDF:** https://arxiv.org/pdf/2601.17679v1
+
+## 19466. DIML: Differentiable Inverse Mechanism Learning from Behaviors of Multi-Agent Learning Trajectories
+
+- **ID:** `2601.17678` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Zhiyu An, Wan Du
+- **abs:** https://arxiv.org/abs/2601.17678 · **PDF:** https://arxiv.org/pdf/2601.17678v1
+
+## 19467. Uni-RS: A Spatially Faithful Unified Understanding and Generation Model for Remote Sensing
+
+- **ID:** `2601.17673` · **primary:** `cs.CV` · **Submitted:** 2026-01-25
+- **Authors:** Weiyu Zhang, Yuan Hu, Yong Li, et al.
+- **abs:** https://arxiv.org/abs/2601.17673 · **PDF:** https://arxiv.org/pdf/2601.17673v1
+
+## 19468. Align to the Pivot: Dual Alignment with Self-Feedback for Multilingual Math Reasoning
+
+- **ID:** `2601.17671` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Chunxu Zhao, Xin Huang, Xue Han, et al.
+- **abs:** https://arxiv.org/abs/2601.17671 · **PDF:** https://arxiv.org/pdf/2601.17671v1
+
+## 19469. Grammar-Aware Literate Generative Mathematical Programming with Compiler-in-the-Loop
+
+- **ID:** `2601.17670` · **primary:** `cs.PL` · **Submitted:** 2026-01-25
+- **Authors:** Roberto Rossi, Steven D. Prestwich
+- **abs:** https://arxiv.org/abs/2601.17670 · **PDF:** https://arxiv.org/pdf/2601.17670v1
+
+## 19470. VAE with Hyperspherical Coordinates: Improving Anomaly Detection from Hypervolume-Compressed Latent Space
+
+- **ID:** `2601.18823` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Alejandro Ascarate, Leo Lebrat, Rodrigo Santa Cruz, et al.
+- **abs:** https://arxiv.org/abs/2601.18823 · **PDF:** https://arxiv.org/pdf/2601.18823v2
+
+## 19471. Fast KVzip: Efficient and Accurate LLM Inference with Gated KV Eviction
+
+- **ID:** `2601.17668` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Jang-Hyun Kim, Dongyoon Han, Sangdoo Yun
+- **abs:** https://arxiv.org/abs/2601.17668 · **PDF:** https://arxiv.org/pdf/2601.17668v2
+
+## 19472. Entropic Risk-Aware Monte Carlo Tree Search
+
+- **ID:** `2601.17667` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Pedro P. Santos, Jacopo Silvestrin, Alberto Sardinha, et al.
+- **abs:** https://arxiv.org/abs/2601.17667 · **PDF:** https://arxiv.org/pdf/2601.17667v2
+
+## 19473. UrduLM: A Resource-Efficient Monolingual Urdu Language Model
+
+- **ID:** `2601.17664` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Syed Muhammad Ali, Hammad Sajid, Zainab Haider, et al.
+- **abs:** https://arxiv.org/abs/2601.17664 · **PDF:** https://arxiv.org/pdf/2601.17664v1
+
+## 19474. Beyond the Rabbit Hole: Mapping the Relational Harms of QAnon Radicalization
+
+- **ID:** `2601.17658` · **primary:** `cs.CL` · **Submitted:** 2026-01-25
+- **Authors:** Bich Ngoc, Doan, Giuseppe Russo, et al.
+- **abs:** https://arxiv.org/abs/2601.17658 · **PDF:** https://arxiv.org/pdf/2601.17658v1
+
+## 19475. Autonomous Multi-Agent AI for High-Throughput Polymer Informatics: From Property Prediction to Generative Design Across Synthetic and Bio-Polymers
+
+- **ID:** `2602.00103` · **primary:** `cond-mat.soft` · **Submitted:** 2026-01-25
+- **Authors:** Mahule Roy, Adib Bazgir, Arthur da Silva Sousa Santos, et al.
+- **abs:** https://arxiv.org/abs/2602.00103 · **PDF:** https://arxiv.org/pdf/2602.00103v1
+
+## 19476. Kareus: Joint Reduction of Dynamic and Static Energy in Large Model Training
+
+- **ID:** `2601.17654` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Ruofan Wu, Jae-Won Chung, Mosharaf Chowdhury
+- **abs:** https://arxiv.org/abs/2601.17654 · **PDF:** https://arxiv.org/pdf/2601.17654v1
+
+## 19477. Time-Varying Causal Treatment for Quantifying the Causal Effect of Short-Term Variations on Arctic Sea Ice Dynamics
+
+- **ID:** `2601.17647` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Akila Sampath, Vandana Janeja, Jianwu Wang
+- **abs:** https://arxiv.org/abs/2601.17647 · **PDF:** https://arxiv.org/pdf/2601.17647v1
+
+## 19478. A Mosco sufficient condition for intrinsic stability of non-unique convex Empirical Risk Minimization
+
+- **ID:** `2601.17646` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Karim Bounja, Lahcen Laayouni, Abdeljalil Sakat
+- **abs:** https://arxiv.org/abs/2601.17646 · **PDF:** https://arxiv.org/pdf/2601.17646v1
+
+## 19479. AVMeme Exam: A Multimodal Multilingual Multicultural Benchmark for LLMs' Contextual and Cultural Knowledge and Thinking
+
+- **ID:** `2601.17645` · **primary:** `cs.SD` · **Submitted:** 2026-01-25
+- **Authors:** Xilin Jiang, Qiaolin Wang, Junkai Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.17645 · **PDF:** https://arxiv.org/pdf/2601.17645v1
+
+## 19480. A Systemic Evaluation of Multimodal RAG Privacy
+
+- **ID:** `2601.17644` · **primary:** `cs.CR` · **Submitted:** 2026-01-25
+- **Authors:** Ali Al-Lawati, Suhang Wang
+- **abs:** https://arxiv.org/abs/2601.17644 · **PDF:** https://arxiv.org/pdf/2601.17644v2
+
+## 19481. Health-ORSC-Bench: A Benchmark for Measuring Over-Refusal and Safety Completion in Health Context
+
+- **ID:** `2601.17642` · **primary:** `cs.AI` · **Submitted:** 2026-01-25
+- **Authors:** Zhihao Zhang, Liting Huang, Guanghao Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.17642 · **PDF:** https://arxiv.org/pdf/2601.17642v1
+
+## 19482. RPNT: Robust Pre-trained Neural Transformer -- A Pathway for Generalized Motor Decoding
+
+- **ID:** `2601.17641` · **primary:** `cs.LG` · **Submitted:** 2026-01-25
+- **Authors:** Hao Fang, Ryan A. Canfield, Tomohiro Ouchi, et al.
+- **abs:** https://arxiv.org/abs/2601.17641 · **PDF:** https://arxiv.org/pdf/2601.17641v2
+
+## 19483. Turbo4DGen: Ultra-Fast Acceleration for 4D Generation
+
+- **ID:** `2603.29572` · **primary:** `cs.GR` · **Submitted:** 2026-01-24
+- **Authors:** Yuanbin Man, Ying Huang, Zhile Ren, et al.
+- **abs:** https://arxiv.org/abs/2603.29572 · **PDF:** https://arxiv.org/pdf/2603.29572v1
+
+## 19484. BrainDistill: Implantable Motor Decoding with Task-Specific Knowledge Distillation
+
+- **ID:** `2601.17625` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Yuhan Xie, Jinhan Liu, Xiaoyong Ni, et al.
+- **abs:** https://arxiv.org/abs/2601.17625 · **PDF:** https://arxiv.org/pdf/2601.17625v1
+
+## 19485. Memento: Towards Proactive Visualization of Everyday Memories with Personal Wearable AR Assistant
+
+- **ID:** `2601.17622` · **primary:** `cs.HC` · **Submitted:** 2026-01-24
+- **Authors:** Yoonsang Kim, Yalong Yang, Arie E. Kaufman
+- **abs:** https://arxiv.org/abs/2601.17622 · **PDF:** https://arxiv.org/pdf/2601.17622v3
+
+## 19486. Agentic Search in the Wild: Intents and Trajectory Dynamics from 14M+ Real Search Requests
+
+- **ID:** `2601.17617` · **primary:** `cs.IR` · **Submitted:** 2026-01-24
+- **Authors:** Jingjie Ning, João Coelho, Yibo Kong, et al.
+- **abs:** https://arxiv.org/abs/2601.17617 · **PDF:** https://arxiv.org/pdf/2601.17617v2
+
+## 19487. Split-on-Share: Mixture of Sparse Experts for Task-Agnostic Continual Learning
+
+- **ID:** `2601.17616` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Fatema Siddika, Md Anwar Hossen, Tanwi Mallick, et al.
+- **abs:** https://arxiv.org/abs/2601.17616 · **PDF:** https://arxiv.org/pdf/2601.17616v1
+
+## 19488. Athena: Synergizing Data Prefetching and Off-Chip Prediction via Online Reinforcement Learning
+
+- **ID:** `2601.17615` · **primary:** `cs.AR` · **Submitted:** 2026-01-24
+- **Authors:** Rahul Bera, Zhenrong Lang, Caroline Hengartner, et al.
+- **abs:** https://arxiv.org/abs/2601.17615 · **PDF:** https://arxiv.org/pdf/2601.17615v1
+
+## 19489. Artificial Intelligence for Inclusive Engineering Education: Advancing Equality, Diversity, and Ethical Leadership
+
+- **ID:** `2602.02520` · **primary:** `cs.CY` · **Submitted:** 2026-01-24
+- **Authors:** Mona G. Ibrahim, Riham Hilal
+- **abs:** https://arxiv.org/abs/2602.02520 · **PDF:** https://arxiv.org/pdf/2602.02520v1
+
+## 19490. Radiomics in Medical Imaging: Methods, Applications, and Challenges
+
+- **ID:** `2602.00102` · **primary:** `eess.IV` · **Submitted:** 2026-01-24
+- **Authors:** Fnu Neha, Deepak kumar Shukla
+- **abs:** https://arxiv.org/abs/2602.00102 · **PDF:** https://arxiv.org/pdf/2602.00102v1
+
+## 19491. ToS: A Team of Specialists ensemble framework for Stereo Sound Event Localization and Detection with distance estimation in Video
+
+- **ID:** `2601.17611` · **primary:** `eess.AS` · **Submitted:** 2026-01-24
+- **Authors:** Davide Berghi, Philip J. B. Jackson
+- **abs:** https://arxiv.org/abs/2601.17611 · **PDF:** https://arxiv.org/pdf/2601.17611v1
+
+## 19492. What Language Models Know But Don't Say: Non-Generative Prior Extraction for Generalization
+
+- **ID:** `2601.17609` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Sara Rezaeimanesh, Mohammad M. Ghassemi
+- **abs:** https://arxiv.org/abs/2601.17609 · **PDF:** https://arxiv.org/pdf/2601.17609v1
+
+## 19493. A Thermodynamic Theory of Learning I: Irreversible Ensemble Transport and Epistemic Costs
+
+- **ID:** `2601.17607` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Daisuke Okanohara
+- **abs:** https://arxiv.org/abs/2601.17607 · **PDF:** https://arxiv.org/pdf/2601.17607v2
+
+## 19494. Human-Aligned Enhancement of Programming Answers with LLMs Guided by User Feedback
+
+- **ID:** `2601.17604` · **primary:** `cs.SE` · **Submitted:** 2026-01-24
+- **Authors:** Suborno Deb Bappon, Saikat Mondal, Chanchal K. Roy, et al.
+- **abs:** https://arxiv.org/abs/2601.17604 · **PDF:** https://arxiv.org/pdf/2601.17604v1
+
+## 19495. Understanding Transformer Encoder-Decoder Representations through Bernoulli Dropout
+
+- **ID:** `2601.17602` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Xuanzhou Chen
+- **abs:** https://arxiv.org/abs/2601.17602 · **PDF:** https://arxiv.org/pdf/2601.17602v1
+
+## 19496. Deep Intrinsic Surprise-Regularized Control (DISRC): A Biologically Inspired Mechanism for Efficient Deep Q-Learning in Sparse Environments
+
+- **ID:** `2601.17598` · **primary:** `cs.NE` · **Submitted:** 2026-01-24
+- **Authors:** Yash Kini, Shiv Davay, Shreya Polavarapu
+- **abs:** https://arxiv.org/abs/2601.17598 · **PDF:** https://arxiv.org/pdf/2601.17598v1
+
+## 19497. Learning to Ideate for Machine Learning Engineering Agents
+
+- **ID:** `2601.17596` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Yunxiang Zhang, Kang Zhou, Zhichao Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.17596 · **PDF:** https://arxiv.org/pdf/2601.17596v1
+
+## 19498. From Chains to DAGs: Probing the Graph Structure of Reasoning in LLMs
+
+- **ID:** `2601.17593` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Tianjun Zhong, Linyang He, Nima Mesgarani
+- **abs:** https://arxiv.org/abs/2601.17593 · **PDF:** https://arxiv.org/pdf/2601.17593v1
+
+## 19499. Intelligence Requires Grounding But Not Embodiment
+
+- **ID:** `2601.17588` · **primary:** `cs.AI` · **Submitted:** 2026-01-24
+- **Authors:** Marcus Ma, Shrikanth Narayanan
+- **abs:** https://arxiv.org/abs/2601.17588 · **PDF:** https://arxiv.org/pdf/2601.17588v1
+
+## 19500. Discovery of Feasible 3D Printing Configurations for Metal Alloys via AI-driven Adaptive Experimental Design
+
+- **ID:** `2601.17587` · **primary:** `cs.AI` · **Submitted:** 2026-01-24
+- **Authors:** Azza Fadhel, Nathaniel W. Zuckschwerdt, Aryan Deshwal, et al.
+- **abs:** https://arxiv.org/abs/2601.17587 · **PDF:** https://arxiv.org/pdf/2601.17587v1
+
+## 19501. Stylizing ViT: Anatomy-Preserving Instance Style Transfer for Domain Generalization
+
+- **ID:** `2601.17586` · **primary:** `cs.CV` · **Submitted:** 2026-01-24
+- **Authors:** Sebastian Doerrich, Francesco Di Salvo, Jonas Alle, et al.
+- **abs:** https://arxiv.org/abs/2601.17586 · **PDF:** https://arxiv.org/pdf/2601.17586v1
+
+## 19502. Sequence Repetition Enhances Token Embeddings and Improves Sequence Labeling with Decoder-only Language Models
+
+- **ID:** `2601.17585` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Matija Luka Kukić, Marko Čuljak, David Dukić, et al.
+- **abs:** https://arxiv.org/abs/2601.17585 · **PDF:** https://arxiv.org/pdf/2601.17585v1
+
+## 19503. Prompt Driven Development with Claude Code: Building a Complete TUI Framework for the Ring Programming Language
+
+- **ID:** `2601.17584` · **primary:** `cs.SE` · **Submitted:** 2026-01-24
+- **Authors:** Mahmoud Samir Fayed, Ahmed Samir Fayed
+- **abs:** https://arxiv.org/abs/2601.17584 · **PDF:** https://arxiv.org/pdf/2601.17584v1
+
+## 19504. Lossy Image Compression -- A Frequent Sequence Mining perspective employing efficient Clustering
+
+- **ID:** `2601.18821` · **primary:** `eess.IV` · **Submitted:** 2026-01-24
+- **Authors:** Avinash Kadimisetty, Oswald C, Sivaselvan B, et al.
+- **abs:** https://arxiv.org/abs/2601.18821 · **PDF:** https://arxiv.org/pdf/2601.18821v1
+
+## 19505. Frequent Pattern Mining approach to Image Compression
+
+- **ID:** `2602.00100` · **primary:** `eess.IV` · **Submitted:** 2026-01-24
+- **Authors:** Avinash Kadimisetty, C. Oswald, B. Sivalselvan
+- **abs:** https://arxiv.org/abs/2602.00100 · **PDF:** https://arxiv.org/pdf/2602.00100v1
+
+## 19506. GenAI-Net: A Generative AI Framework for Automated Biomolecular Network Design
+
+- **ID:** `2601.17582` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-24
+- **Authors:** Maurice Filo, Nicolò Rossi, Zhou Fang, et al.
+- **abs:** https://arxiv.org/abs/2601.17582 · **PDF:** https://arxiv.org/pdf/2601.17582v2
+
+## 19507. How AI Coding Agents Modify Code: A Large-Scale Study of GitHub Pull Requests
+
+- **ID:** `2601.17581` · **primary:** `cs.SE` · **Submitted:** 2026-01-24
+- **Authors:** Daniel Ogenrwot, John Businge
+- **abs:** https://arxiv.org/abs/2601.17581 · **PDF:** https://arxiv.org/pdf/2601.17581v2
+
+## 19508. Status Hierarchies in Language Models
+
+- **ID:** `2601.17577` · **primary:** `cs.HC` · **Submitted:** 2026-01-24
+- **Authors:** Emilio Barkett
+- **abs:** https://arxiv.org/abs/2601.17577 · **PDF:** https://arxiv.org/pdf/2601.17577v1
+
+## 19509. ME-WARD: A multimodal ergonomic analysis tool for musculoskeletal risk assessment from inertial and video data in working plac
+
+- **ID:** `2601.17571` · **primary:** `eess.SP` · **Submitted:** 2026-01-24
+- **Authors:** Javier González-Alonso, Paula Martín-Tapia, David González-Ortega, et al.
+- **abs:** https://arxiv.org/abs/2601.17571 · **PDF:** https://arxiv.org/pdf/2601.17571v1
+
+## 19510. Quantum-Inspired Episode Selection for Monte Carlo Reinforcement Learning via QUBO Optimization
+
+- **ID:** `2601.17570` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Hadi Salloum, Ali Jnadi, Yaroslav Kholodov, et al.
+- **abs:** https://arxiv.org/abs/2601.17570 · **PDF:** https://arxiv.org/pdf/2601.17570v1
+
+## 19511. Improving User Privacy in Personalized Generation: Client-Side Retrieval-Augmented Modification of Server-Side Generated Speculations
+
+- **ID:** `2601.17569` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Alireza Salemi, Hamed Zamani
+- **abs:** https://arxiv.org/abs/2601.17569 · **PDF:** https://arxiv.org/pdf/2601.17569v1
+
+## 19512. Real-Time Trend Prediction via Continually-Aligned LLM Query Generation
+
+- **ID:** `2601.17567` · **primary:** `cs.IR` · **Submitted:** 2026-01-24
+- **Authors:** Zijing Hui, Wenhan Lyu, Shusen Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.17567 · **PDF:** https://arxiv.org/pdf/2601.17567v1
+
+## 19513. JaxARC: A High-Performance JAX-based Environment for Abstraction and Reasoning Research
+
+- **ID:** `2601.17564` · **primary:** `cs.AI` · **Submitted:** 2026-01-24
+- **Authors:** Aadam, Monu Verma, Mohamed Abdel-Mottaleb
+- **abs:** https://arxiv.org/abs/2601.17564 · **PDF:** https://arxiv.org/pdf/2601.17564v1
+
+## 19514. Towards Generalisable Imitation Learning Through Conditioned Transition Estimation and Online Behaviour Alignment
+
+- **ID:** `2601.17563` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Nathan Gavenski, Matteo Leonetti, Odinaldo Rodrigues
+- **abs:** https://arxiv.org/abs/2601.17563 · **PDF:** https://arxiv.org/pdf/2601.17563v1
+
+## 19515. Sparse RBF Networks for PDEs and nonlocal equations: function space theory, operator calculus, and training algorithms
+
+- **ID:** `2601.17562` · **primary:** `math.NA` · **Submitted:** 2026-01-24
+- **Authors:** Zihan Shao, Konstantin Pieper, Xiaochuan Tian
+- **abs:** https://arxiv.org/abs/2601.17562 · **PDF:** https://arxiv.org/pdf/2601.17562v1
+
+## 19516. Lost in Translation: How Language Re-Aligns Vision for Cross-Species Pathology
+
+- **ID:** `2603.04405` · **primary:** `cs.CV` · **Submitted:** 2026-01-24
+- **Authors:** Ekansh Arora
+- **abs:** https://arxiv.org/abs/2603.04405 · **PDF:** https://arxiv.org/pdf/2603.04405v1
+
+## 19517. GreenServ: Energy-Efficient Context-Aware Dynamic Routing for Multi-Model LLM Inference
+
+- **ID:** `2601.17551` · **primary:** `cs.PF` · **Submitted:** 2026-01-24
+- **Authors:** Thomas Ziller, Shashikant Ilager, Alessandro Tundo, et al.
+- **abs:** https://arxiv.org/abs/2601.17551 · **PDF:** https://arxiv.org/pdf/2601.17551v2
+
+## 19518. Breaking the Protocol: Security Analysis of the Model Context Protocol Specification and Prompt Injection Vulnerabilities in Tool-Integrated LLM Agents
+
+- **ID:** `2601.17549` · **primary:** `cs.CR` · **Submitted:** 2026-01-24
+- **Authors:** Narek Maloyan, Dmitry Namiot
+- **abs:** https://arxiv.org/abs/2601.17549 · **PDF:** https://arxiv.org/pdf/2601.17549v1
+
+## 19519. Cognitive Platform Engineering for Autonomous Cloud Operations
+
+- **ID:** `2601.17542` · **primary:** `cs.AI` · **Submitted:** 2026-01-24
+- **Authors:** Vinoth Punniyamoorthy, Nitin Saksena, Srivenkateswara Reddy Sankiti, et al.
+- **abs:** https://arxiv.org/abs/2601.17542 · **PDF:** https://arxiv.org/pdf/2601.17542v1
+
+## 19520. OTI: A Model-free and Visually Interpretable Measure of Image Attackability
+
+- **ID:** `2601.17536` · **primary:** `cs.CV` · **Submitted:** 2026-01-24
+- **Authors:** Jiaming Liang, Haowei Liu, Chi-Man Pun
+- **abs:** https://arxiv.org/abs/2601.17536 · **PDF:** https://arxiv.org/pdf/2601.17536v1
+
+## 19521. Reconstructing Training Data from Adapter-based Federated Large Language Models
+
+- **ID:** `2601.17533` · **primary:** `cs.CR` · **Submitted:** 2026-01-24
+- **Authors:** Silong Chen, Yuchuan Luo, Guilin Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.17533 · **PDF:** https://arxiv.org/pdf/2601.17533v1
+
+## 19522. Less is More for RAG: Information Gain Pruning for Generator-Aligned Reranking and Evidence Selection
+
+- **ID:** `2601.17532` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Zhipeng Song, Yizhi Zhou, Xiangyu Kong, et al.
+- **abs:** https://arxiv.org/abs/2601.17532 · **PDF:** https://arxiv.org/pdf/2601.17532v1
+
+## 19523. Revealing the Truth with ConLLM for Detecting Multi-Modal Deepfakes
+
+- **ID:** `2601.17530` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Gautam Siddharth Kashyap, Harsh Joshi, Niharika Jain, et al.
+- **abs:** https://arxiv.org/abs/2601.17530 · **PDF:** https://arxiv.org/pdf/2601.17530v1
+
+## 19524. Bridging Expectation Signals: LLM-Based Experiments and a Behavioral Kalman Filter Framework
+
+- **ID:** `2601.17527` · **primary:** `econ.GN` · **Submitted:** 2026-01-24
+- **Authors:** Yu Wang, Xiangchen Liu
+- **abs:** https://arxiv.org/abs/2601.17527 · **PDF:** https://arxiv.org/pdf/2601.17527v1
+
+## 19525. EuleroDec: A Complex-Valued RVQ-VAE for Efficient and Robust Audio Coding
+
+- **ID:** `2601.17517` · **primary:** `cs.SD` · **Submitted:** 2026-01-24
+- **Authors:** Luca Cerovaz, Michele Mancusi, Emanuele Rodolà
+- **abs:** https://arxiv.org/abs/2601.17517 · **PDF:** https://arxiv.org/pdf/2601.17517v2
+
+## 19526. Constrained Multi-Objective Genetic Algorithm Variants for Design and Optimization of Tri-Band Microstrip Patch Antenna loaded CSRR for IoT Applications: A Comparative Case Study
+
+- **ID:** `2601.17513` · **primary:** `cs.NE` · **Submitted:** 2026-01-24
+- **Authors:** Moahmed Hamza Boulaich, Said Ohamouddou, Mohammed Ali Ennasar, et al.
+- **abs:** https://arxiv.org/abs/2601.17513 · **PDF:** https://arxiv.org/pdf/2601.17513v1
+
+## 19527. One-Shot Federated Clustering of Non-Independent Completely Distributed Data
+
+- **ID:** `2601.17512` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Yiqun Zhang, Shenghong Cai, Zihua Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.17512 · **PDF:** https://arxiv.org/pdf/2601.17512v1
+
+## 19528. "Rebuilding" Statistics in the Age of AI: A Town Hall Discussion on Culture, Infrastructure, and Training
+
+- **ID:** `2601.17510` · **primary:** `stat.ML` · **Submitted:** 2026-01-24
+- **Authors:** David L. Donoho, Jian Kang, Xihong Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.17510 · **PDF:** https://arxiv.org/pdf/2601.17510v1
+
+## 19529. To Case or Not to Case: An Empirical Study in Learned Sparse Retrieval
+
+- **ID:** `2601.17500` · **primary:** `cs.IR` · **Submitted:** 2026-01-24
+- **Authors:** Emmanouil Georgios Lionis, Jia-Huei Ju, Angelos Nalmpantis, et al.
+- **abs:** https://arxiv.org/abs/2601.17500 · **PDF:** https://arxiv.org/pdf/2601.17500v1
+
+## 19530. PEARL: Prototype-Enhanced Alignment for Label-Efficient Representation Learning with Deployment-Driven Insights from Digital Governance Communication Systems
+
+- **ID:** `2601.17495` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Ruiyu Zhang, Lin Nie, Wai-Fung Lam, et al.
+- **abs:** https://arxiv.org/abs/2601.17495 · **PDF:** https://arxiv.org/pdf/2601.17495v1
+
+## 19531. SpatialMath: Spatial Comprehension-Infused Symbolic Reasoning for Mathematical Problem-Solving
+
+- **ID:** `2601.17489` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Ashutosh Bajpai, Akshat Bhandari, Akshay Nambi, et al.
+- **abs:** https://arxiv.org/abs/2601.17489 · **PDF:** https://arxiv.org/pdf/2601.17489v1
+
+## 19532. Automatic Stability and Recovery for Neural Network Training
+
+- **ID:** `2601.17483` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Barak Or
+- **abs:** https://arxiv.org/abs/2601.17483 · **PDF:** https://arxiv.org/pdf/2601.17483v1
+
+## 19533. Lattice: Generative Guardrails for Conversational Agents
+
+- **ID:** `2601.17481` · **primary:** `cs.AI` · **Submitted:** 2026-01-24
+- **Authors:** Emily Broadhurst, Tawab Safi, Joseph Edell, et al.
+- **abs:** https://arxiv.org/abs/2601.17481 · **PDF:** https://arxiv.org/pdf/2601.17481v1
+
+## 19534. Unintended Memorization of Sensitive Information in Fine-Tuned Language Models
+
+- **ID:** `2601.17480` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Marton Szep, Jorge Marin Ruiz, Georgios Kaissis, et al.
+- **abs:** https://arxiv.org/abs/2601.17480 · **PDF:** https://arxiv.org/pdf/2601.17480v1
+
+## 19535. Evaluation of Large Language Models' educational feedback in Higher Education: potential, limitations and implications for educational practice
+
+- **ID:** `2602.02519` · **primary:** `cs.CY` · **Submitted:** 2026-01-24
+- **Authors:** Daniele Agostini, Federica Picasso
+- **abs:** https://arxiv.org/abs/2602.02519 · **PDF:** https://arxiv.org/pdf/2602.02519v1
+
+## 19536. LeanTutor: Towards a Verified AI Mathematical Proof Tutor
+
+- **ID:** `2601.17473` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Manooshree Patel, Rayna Bhattacharyya, Thomas Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.17473 · **PDF:** https://arxiv.org/pdf/2601.17473v2
+
+## 19537. Identifying and Correcting Label Noise for Robust GNNs via Influence Contradiction
+
+- **ID:** `2601.17469` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Wei Ju, Wei Zhang, Siyu Yi, et al.
+- **abs:** https://arxiv.org/abs/2601.17469 · **PDF:** https://arxiv.org/pdf/2601.17469v1
+
+## 19538. ReflexSplit: Single Image Reflection Separation via Layer Fusion-Separation
+
+- **ID:** `2601.17468` · **primary:** `cs.CV` · **Submitted:** 2026-01-24
+- **Authors:** Chia-Ming Lee, Yu-Fan Lin, Jin-Hui Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.17468 · **PDF:** https://arxiv.org/pdf/2601.17468v3
+
+## 19539. Harnessing Reasoning Trajectories for Hallucination Detection via Answer-agreement Representation Shaping
+
+- **ID:** `2601.17467` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Jianxiong Zhang, Bing Guo, Yuming Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.17467 · **PDF:** https://arxiv.org/pdf/2601.17467v1
+
+## 19540. Bayesian quantum sensing using graybox machine learning
+
+- **ID:** `2601.17465` · **primary:** `quant-ph` · **Submitted:** 2026-01-24
+- **Authors:** Akram Youssry, Stefan Todd, Patrick Murton, et al.
+- **abs:** https://arxiv.org/abs/2601.17465 · **PDF:** https://arxiv.org/pdf/2601.17465v1
+
+## 19541. Embodiment-Induced Coordination Regimes in Tabular Multi-Agent Q-Learning
+
+- **ID:** `2601.17454` · **primary:** `cs.MA` · **Submitted:** 2026-01-24
+- **Authors:** Muhammad Ahmed Atif, Nehal Naeem Haji, Mohammad Shahid Shaikh, et al.
+- **abs:** https://arxiv.org/abs/2601.17454 · **PDF:** https://arxiv.org/pdf/2601.17454v1
+
+## 19542. DREAM: Dual-Standard Semantic Homogeneity with Dynamic Optimization for Graph Learning with Label Noise
+
+- **ID:** `2601.17449` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Yusheng Zhao, Jiaye Xie, Qixin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.17449 · **PDF:** https://arxiv.org/pdf/2601.17449v1
+
+## 19543. Gauss-Newton Natural Gradient Descent for Shape Learning
+
+- **ID:** `2602.00099` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** James King, Arturs Berzins, Siddhartha Mishra, et al.
+- **abs:** https://arxiv.org/abs/2602.00099 · **PDF:** https://arxiv.org/pdf/2602.00099v2
+
+## 19544. Clustering-driven Memory Compression for On-device Large Language Models
+
+- **ID:** `2601.17443` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Ondrej Bohdal, Pramit Saha, Umberto Michieli, et al.
+- **abs:** https://arxiv.org/abs/2601.17443 · **PDF:** https://arxiv.org/pdf/2601.17443v1
+
+## 19545. MO-ELA: Rigorously Expanding Exploratory Landscape Features for Automated Algorithm Selection in Continuous Multi-Objective Optimisation
+
+- **ID:** `2602.00098` · **primary:** `cs.NE` · **Submitted:** 2026-01-24
+- **Authors:** Oliver Preuß, Jeroen Rook, Jakob Bossek, et al.
+- **abs:** https://arxiv.org/abs/2602.00098 · **PDF:** https://arxiv.org/pdf/2602.00098v1
+
+## 19546. A new approach for combined model class selection and parameters learning for auto-regressive neural models
+
+- **ID:** `2601.17442` · **primary:** `eess.SY` · **Submitted:** 2026-01-24
+- **Authors:** Corrado Sgadari, Alessio La Bella, Marcello Farina
+- **abs:** https://arxiv.org/abs/2601.17442 · **PDF:** https://arxiv.org/pdf/2601.17442v1
+
+## 19547. Data-driven Clustering and Merging of Adapters for On-device Large Language Models
+
+- **ID:** `2601.17441` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Ondrej Bohdal, Taha Ceritli, Mete Ozay, et al.
+- **abs:** https://arxiv.org/abs/2601.17441 · **PDF:** https://arxiv.org/pdf/2601.17441v1
+
+## 19548. UniGRec: Unified Generative Recommendation with Soft Identifiers for End-to-End Optimization
+
+- **ID:** `2601.17438` · **primary:** `cs.IR` · **Submitted:** 2026-01-24
+- **Authors:** Jialei Li, Yang Zhang, Yimeng Bai, et al.
+- **abs:** https://arxiv.org/abs/2601.17438 · **PDF:** https://arxiv.org/pdf/2601.17438v1
+
+## 19549. Towards a Declarative Agentic Layer for Intelligent Agents in MCP-Based Server Ecosystems
+
+- **ID:** `2601.17435` · **primary:** `cs.SE` · **Submitted:** 2026-01-24
+- **Authors:** Maria Jesus Rodriguez-Sanchez, Manuel Noguera, Angel Ruiz-Zafra, et al.
+- **abs:** https://arxiv.org/abs/2601.17435 · **PDF:** https://arxiv.org/pdf/2601.17435v1
+
+## 19550. The 17% Gap: Quantifying Epistemic Decay in AI-Assisted Survey Papers
+
+- **ID:** `2601.17431` · **primary:** `cs.CY` · **Submitted:** 2026-01-24
+- **Authors:** H. Kemal İlter
+- **abs:** https://arxiv.org/abs/2601.17431 · **PDF:** https://arxiv.org/pdf/2601.17431v1
+
+## 19551. Active Hypothesis Testing for Correlated Combinatorial Anomaly Detection
+
+- **ID:** `2601.17430` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Zichuan Yang, Yiming Xing
+- **abs:** https://arxiv.org/abs/2601.17430 · **PDF:** https://arxiv.org/pdf/2601.17430v1
+
+## 19552. Coronary Artery Segmentation and Vessel-Type Classification in X-Ray Angiography
+
+- **ID:** `2601.17429` · **primary:** `cs.CV` · **Submitted:** 2026-01-24
+- **Authors:** Mehdi Yousefzadeh, Siavash Shirzadeh Barough, Ashkan Fakharifar, et al.
+- **abs:** https://arxiv.org/abs/2601.17429 · **PDF:** https://arxiv.org/pdf/2601.17429v1
+
+## 19553. A Syllogistic Probe: Tracing the Evolution of Logic Reasoning in Large Language Models
+
+- **ID:** `2601.17426` · **primary:** `cs.AI` · **Submitted:** 2026-01-24
+- **Authors:** Zhengqing Zang, Yuqi Ding, Yanmei Gu, et al.
+- **abs:** https://arxiv.org/abs/2601.17426 · **PDF:** https://arxiv.org/pdf/2601.17426v1
+
+## 19554. Oops, Wait: Token-Level Signals as a Lens into LLM Reasoning
+
+- **ID:** `2601.17421` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Jaehui Hwang, Dongyoon Han, Sangdoo Yun, et al.
+- **abs:** https://arxiv.org/abs/2601.17421 · **PDF:** https://arxiv.org/pdf/2601.17421v1
+
+## 19555. Semantic-Aware Task Clustering for Federated Cooperative Multi-Task Semantic Communication
+
+- **ID:** `2601.17419` · **primary:** `eess.SP` · **Submitted:** 2026-01-24
+- **Authors:** Ahmad Halimi Razlighi, Pallavi Dhingra, Edgar Beck, et al.
+- **abs:** https://arxiv.org/abs/2601.17419 · **PDF:** https://arxiv.org/pdf/2601.17419v1
+
+## 19556. Kalman-Inspired Runtime Stability and Recovery in Hybrid Reasoning Systems
+
+- **ID:** `2602.15855` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Barak Or
+- **abs:** https://arxiv.org/abs/2602.15855 · **PDF:** https://arxiv.org/pdf/2602.15855v1
+
+## 19557. NCSAM Noise-Compensated Sharpness-Aware Minimization for Noisy Label Learning
+
+- **ID:** `2601.19947` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Jiayu Xu, Junbiao Pang
+- **abs:** https://arxiv.org/abs/2601.19947 · **PDF:** https://arxiv.org/pdf/2601.19947v1
+
+## 19558. Efficient Dilated Squeeze and Excitation Neural Operator for Differential Equations
+
+- **ID:** `2601.17407` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Prajwal Chauhan, Salah Eddine Choutri, Saif Eddin Jabari
+- **abs:** https://arxiv.org/abs/2601.17407 · **PDF:** https://arxiv.org/pdf/2601.17407v1
+
+## 19559. ReLE: A Scalable System and Structured Benchmark for Diagnosing Capability Anisotropy in Chinese LLMs
+
+- **ID:** `2601.17399` · **primary:** `cs.CV` · **Submitted:** 2026-01-24
+- **Authors:** Rui Fang, Jian Li, Wei Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.17399 · **PDF:** https://arxiv.org/pdf/2601.17399v1
+
+## 19560. CLM-Bench: Benchmarking and Analyzing Cross-lingual Misalignment of LLMs in Knowledge Editing
+
+- **ID:** `2601.17397` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Yucheng Hu, Wei Zhou, Juesi Xiao
+- **abs:** https://arxiv.org/abs/2601.17397 · **PDF:** https://arxiv.org/pdf/2601.17397v1
+
+## 19561. MK-SGC-SC: Multiple Kernel Guided Sparse Graph Construction in Spectral Clustering for Unsupervised Speaker Diarization
+
+- **ID:** `2601.19946` · **primary:** `eess.AS` · **Submitted:** 2026-01-24
+- **Authors:** Nikhil Raghav, Avisek Gupta, Swagatam Das, et al.
+- **abs:** https://arxiv.org/abs/2601.19946 · **PDF:** https://arxiv.org/pdf/2601.19946v2
+
+## 19562. GO-OSC and VASH: Geometry-Aware Representation Learning for Early Degradation Detection in Oscillatory Systems
+
+- **ID:** `2601.17396` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Vashista Nobaub
+- **abs:** https://arxiv.org/abs/2601.17396 · **PDF:** https://arxiv.org/pdf/2601.17396v1
+
+## 19563. ONRW: Optimizing inversion noise for high-quality and robust watermark
+
+- **ID:** `2601.17388` · **primary:** `cs.CV` · **Submitted:** 2026-01-24
+- **Authors:** Xuan Ding, Xiu Yan, Chuanlong Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.17388 · **PDF:** https://arxiv.org/pdf/2601.17388v1
+
+## 19564. Generation-Step-Aware Framework for Cross-Modal Representation and Control in Multilingual Speech-Text Models
+
+- **ID:** `2601.17387` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Toshiki Nakai, Varsha Suresh, Vera Demberg
+- **abs:** https://arxiv.org/abs/2601.17387 · **PDF:** https://arxiv.org/pdf/2601.17387v2
+
+## 19565. Physical Prompt Injection Attacks on Large Vision-Language Models
+
+- **ID:** `2601.17383` · **primary:** `cs.CV` · **Submitted:** 2026-01-24
+- **Authors:** Chen Ling, Kai Hu, Hangcheng Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.17383 · **PDF:** https://arxiv.org/pdf/2601.17383v1
+
+## 19566. Prompt and Circumstances: Evaluating the Efficacy of Human Prompt Inference in AI-Generated Art
+
+- **ID:** `2601.17379` · **primary:** `cs.CR` · **Submitted:** 2026-01-24
+- **Authors:** Khoi Trinh, Scott Seidenberger, Joseph Spracklen, et al.
+- **abs:** https://arxiv.org/abs/2601.17379 · **PDF:** https://arxiv.org/pdf/2601.17379v1
+
+## 19567. Res-MIA: A Training-Free Resolution-Based Membership Inference Attack on Federated Learning Models
+
+- **ID:** `2601.17378` · **primary:** `cs.CR` · **Submitted:** 2026-01-24
+- **Authors:** Mohammad Zare, Pirooz Shamsinejadbabaki
+- **abs:** https://arxiv.org/abs/2601.17378 · **PDF:** https://arxiv.org/pdf/2601.17378v1
+
+## 19568. WarrantScore: Modeling Warrants between Claims and Evidence for Substantiation Evaluation in Peer Reviews
+
+- **ID:** `2601.17377` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Kiyotada Mori, Shohei Tanaka, Tosho Hirasawa, et al.
+- **abs:** https://arxiv.org/abs/2601.17377 · **PDF:** https://arxiv.org/pdf/2601.17377v1
+
+## 19569. Diversified Scaling Inference in Time Series Foundation Models
+
+- **ID:** `2601.17376` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Ruijin Hua, Zichuan Liu, Kun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.17376 · **PDF:** https://arxiv.org/pdf/2601.17376v1
+
+## 19570. Error Analysis of Bayesian Inverse Problems with Generative Priors
+
+- **ID:** `2601.17374` · **primary:** `stat.ML` · **Submitted:** 2026-01-24
+- **Authors:** Bamdad Hosseini, Ziqi Huang
+- **abs:** https://arxiv.org/abs/2601.17374 · **PDF:** https://arxiv.org/pdf/2601.17374v2
+
+## 19571. Elastic Attention: Test-time Adaptive Sparsity Ratios for Efficient Transformers
+
+- **ID:** `2601.17367` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Zecheng Tang, Quantong Qiu, Yi Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.17367 · **PDF:** https://arxiv.org/pdf/2601.17367v2
+
+## 19572. Parameter Efficient Fine Tuning Llama 3.1 for Answering Arabic Legal Questions: A Case Study on Jordanian Laws
+
+- **ID:** `2601.17364` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Mohammed Fasha, Bassam Hammo, Bilal Sowan, et al.
+- **abs:** https://arxiv.org/abs/2601.17364 · **PDF:** https://arxiv.org/pdf/2601.17364v1
+
+## 19573. Do readers prefer AI-generated Italian short stories?
+
+- **ID:** `2601.17363` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Michael Farrell
+- **abs:** https://arxiv.org/abs/2601.17363 · **PDF:** https://arxiv.org/pdf/2601.17363v1
+
+## 19574. Robust Privacy: Inference-Time Privacy through Certified Robustness
+
+- **ID:** `2601.17360` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Jiankai Jin, Xiangzheng Zhang, Zhao Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.17360 · **PDF:** https://arxiv.org/pdf/2601.17360v1
+
+## 19575. Spectral Geometry for Deep Learning: Compression and Hallucination Detection via Random Matrix Theory
+
+- **ID:** `2601.17357` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Davide Ettori
+- **abs:** https://arxiv.org/abs/2601.17357 · **PDF:** https://arxiv.org/pdf/2601.17357v1
+
+## 19576. Mirage2Matter: A Physically Grounded Gaussian World Model from Video
+
+- **ID:** `2602.00096` · **primary:** `cs.CV` · **Submitted:** 2026-01-24
+- **Authors:** Zhengqing Gao, Ziwen Li, Xin Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00096 · **PDF:** https://arxiv.org/pdf/2602.00096v1
+
+## 19577. NeRF-MIR: Towards High-Quality Restoration of Masked Images with Neural Radiance Fields
+
+- **ID:** `2601.17350` · **primary:** `cs.CV` · **Submitted:** 2026-01-24
+- **Authors:** Xianliang Huang, Zhizhou Zhong, Shuhang Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.17350 · **PDF:** https://arxiv.org/pdf/2601.17350v1
+
+## 19578. Auditing Disability Representation in Vision-Language Models
+
+- **ID:** `2601.17348` · **primary:** `cs.AI` · **Submitted:** 2026-01-24
+- **Authors:** Srikant Panda, Sourabh Singh Yadav, Palkesh Malviya
+- **abs:** https://arxiv.org/abs/2601.17348 · **PDF:** https://arxiv.org/pdf/2601.17348v1
+
+## 19579. Multi-Agent Learning Path Planning via LLMs
+
+- **ID:** `2601.17346` · **primary:** `cs.AI` · **Submitted:** 2026-01-24
+- **Authors:** Haoxin Xu, Changyong Qi, Tong Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.17346 · **PDF:** https://arxiv.org/pdf/2601.17346v1
+
+## 19580. The Shadow Self: Intrinsic Value Misalignment in Large Language Model Agents
+
+- **ID:** `2601.17344` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Chen Chen, Kim Young Il, Yuan Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.17344 · **PDF:** https://arxiv.org/pdf/2601.17344v1
+
+## 19581. Are We Evaluating the Edit Locality of LLM Model Editing Properly?
+
+- **ID:** `2601.17343` · **primary:** `cs.AI` · **Submitted:** 2026-01-24
+- **Authors:** Wei Liu, Haomei Xu, Hongkai Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.17343 · **PDF:** https://arxiv.org/pdf/2601.17343v1
+
+## 19582. Decoupling Strategy and Execution in Task-Focused Dialogue via Goal-Oriented Preference Optimization
+
+- **ID:** `2602.15854` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Jingyi Xu, Xingyu Ren, Zhoupeng Shou, et al.
+- **abs:** https://arxiv.org/abs/2602.15854 · **PDF:** https://arxiv.org/pdf/2602.15854v2
+
+## 19583. The Relativity of AGI: Distributional Axioms, Fragility, and Undecidability
+
+- **ID:** `2601.17335` · **primary:** `cs.AI` · **Submitted:** 2026-01-24
+- **Authors:** Angshul Majumdar
+- **abs:** https://arxiv.org/abs/2601.17335 · **PDF:** https://arxiv.org/pdf/2601.17335v1
+
+## 19584. Power-based Partial Attention: Bridging Linear-Complexity and Full Attention
+
+- **ID:** `2601.17334` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Yufeng Huang
+- **abs:** https://arxiv.org/abs/2601.17334 · **PDF:** https://arxiv.org/pdf/2601.17334v2
+
+## 19585. FinMetaMind: A Tech Blueprint on NLQ Systems for Financial Knowledge Search
+
+- **ID:** `2601.17333` · **primary:** `cs.IR` · **Submitted:** 2026-01-24
+- **Authors:** Lalit Pant, Shivang Nagar
+- **abs:** https://arxiv.org/abs/2601.17333 · **PDF:** https://arxiv.org/pdf/2601.17333v1
+
+## 19586. TheoremForge: Scaling up Formal Data Synthesis with Low-Budget Agentic Workflow
+
+- **ID:** `2601.17332` · **primary:** `cs.AI` · **Submitted:** 2026-01-24
+- **Authors:** Yicheng Tao, Hongteng Xu
+- **abs:** https://arxiv.org/abs/2601.17332 · **PDF:** https://arxiv.org/pdf/2601.17332v1
+
+## 19587. Thermodynamically Optimal Regularization under Information-Geometric Constraints
+
+- **ID:** `2601.17330` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Laurent Caraffa
+- **abs:** https://arxiv.org/abs/2601.17330 · **PDF:** https://arxiv.org/pdf/2601.17330v1
+
+## 19588. Conformal Feedback Alignment: Quantifying Answer-Level Reliability for Robust LLM Alignment
+
+- **ID:** `2601.17329` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Tiejin Chen, Xiaoou Liu, Vishnu Nandam, et al.
+- **abs:** https://arxiv.org/abs/2601.17329 · **PDF:** https://arxiv.org/pdf/2601.17329v1
+
+## 19589. Denoising diffusion networks for normative modeling in neuroimaging
+
+- **ID:** `2602.04886` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Luke Whitbread, Lyle J. Palmer, Mark Jenkinson
+- **abs:** https://arxiv.org/abs/2602.04886 · **PDF:** https://arxiv.org/pdf/2602.04886v1
+
+## 19590. From Prompts to Worlds: How Users Iterate, Explore, and Make Sense of AI-Generated 3D Environments
+
+- **ID:** `2603.13233` · **primary:** `cs.HC` · **Submitted:** 2026-01-24
+- **Authors:** Aung Pyae
+- **abs:** https://arxiv.org/abs/2603.13233 · **PDF:** https://arxiv.org/pdf/2603.13233v1
+
+## 19591. ClinNet: Evidential Ordinal Regression with Bilateral Asymmetry and Prototype Memory for Knee Osteoarthritis Grading
+
+- **ID:** `2601.17315` · **primary:** `cs.CV` · **Submitted:** 2026-01-24
+- **Authors:** Xiaoyang Li, Runni Zhou
+- **abs:** https://arxiv.org/abs/2601.17315 · **PDF:** https://arxiv.org/pdf/2601.17315v1
+
+## 19592. Meta-Judging with Large Language Models: Concepts, Methods, and Challenges
+
+- **ID:** `2601.17312` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Hugo Silva, Mateus Mendes, Hugo Gonçalo Oliveira
+- **abs:** https://arxiv.org/abs/2601.17312 · **PDF:** https://arxiv.org/pdf/2601.17312v1
+
+## 19593. Phase Transition for Budgeted Multi-Agent Synergy
+
+- **ID:** `2601.17311` · **primary:** `cs.AI` · **Submitted:** 2026-01-24
+- **Authors:** Bang Liu, Linglong Kong, Jian Pei
+- **abs:** https://arxiv.org/abs/2601.17311 · **PDF:** https://arxiv.org/pdf/2601.17311v2
+
+## 19594. High-Fidelity Longitudinal Patient Simulation Using Real-World Data
+
+- **ID:** `2601.17310` · **primary:** `cs.AI` · **Submitted:** 2026-01-24
+- **Authors:** Yu Akagi, Tomohisa Seki, Hiromasa Ito, et al.
+- **abs:** https://arxiv.org/abs/2601.17310 · **PDF:** https://arxiv.org/pdf/2601.17310v1
+
+## 19595. PAR: Plausibility-aware Amortized Recourse Generation
+
+- **ID:** `2601.17309` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Anagha Sabu, Vidhya S, Narayanan C Krishnan
+- **abs:** https://arxiv.org/abs/2601.17309 · **PDF:** https://arxiv.org/pdf/2601.17309v1
+
+## 19596. Weighted Graph Clustering via Scale Contraction and Graph Structure Learning
+
+- **ID:** `2601.17307` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Haobing Liu, Yinuo Zhang, Tingting Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.17307 · **PDF:** https://arxiv.org/pdf/2601.17307v2
+
+## 19597. Decentralized Multi-Agent Swarms for Autonomous Grid Security in Industrial IoT: A Consensus-based Approach
+
+- **ID:** `2601.17303` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Samaresh Kumar Singh, Joyjit Roy
+- **abs:** https://arxiv.org/abs/2601.17303 · **PDF:** https://arxiv.org/pdf/2601.17303v1
+
+## 19598. Tabular Foundation Models are Strong Graph Anomaly Detectors
+
+- **ID:** `2601.17301` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Yunhui Liu, Tieke He, Yongchao Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.17301 · **PDF:** https://arxiv.org/pdf/2601.17301v1
+
+## 19599. Structure-Aware NL-to-SQL for SFC Provisioning via AST-Masking Empowered Language Models
+
+- **ID:** `2601.17295` · **primary:** `cs.NI` · **Submitted:** 2026-01-24
+- **Authors:** Xinyu Zhu, Parisa Fard Moshiri, Poonam Lohan, et al.
+- **abs:** https://arxiv.org/abs/2601.17295 · **PDF:** https://arxiv.org/pdf/2601.17295v1
+
+## 19600. A Lightweight Explainable Guardrail for Prompt Safety
+
+- **ID:** `2602.15853` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Md Asiful Islam, Mihai Surdeanu
+- **abs:** https://arxiv.org/abs/2602.15853 · **PDF:** https://arxiv.org/pdf/2602.15853v1
+
+## 19601. Dynamic Meta-Ensemble Framework for Efficient and Accurate Deep Learning in Plant Leaf Disease Detection on Resource-Constrained Edge Devices
+
+- **ID:** `2601.17290` · **primary:** `cs.CV` · **Submitted:** 2026-01-24
+- **Authors:** Weloday Fikadu Moges, Jianmei Su, Amin Waqas
+- **abs:** https://arxiv.org/abs/2601.17290 · **PDF:** https://arxiv.org/pdf/2601.17290v1
+
+## 19602. Dynamic Adversarial Reinforcement Learning for Robust Multimodal Large Language Models
+
+- **ID:** `2602.22227` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Yicheng Bao, Xuhong Wang, Qiaosheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.22227 · **PDF:** https://arxiv.org/pdf/2602.22227v3
+
+## 19603. Mind the Ambiguity: Aleatoric Uncertainty Quantification in LLMs for Safe Medical Question Answering
+
+- **ID:** `2601.17284` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Yaokun Liu, Yifan Liu, Phoebe Mbuvi, et al.
+- **abs:** https://arxiv.org/abs/2601.17284 · **PDF:** https://arxiv.org/pdf/2601.17284v1
+
+## 19604. On the Insecurity of Keystroke-Based AI Authorship Detection: Timing-Forgery Attacks Against Motor-Signal Verification
+
+- **ID:** `2601.17280` · **primary:** `cs.CR` · **Submitted:** 2026-01-24
+- **Authors:** David Condrey
+- **abs:** https://arxiv.org/abs/2601.17280 · **PDF:** https://arxiv.org/pdf/2601.17280v1
+
+## 19605. PingPong: A Natural Benchmark for Multi-Turn Code-Switching Dialogues
+
+- **ID:** `2601.17277` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Mohammad Rifqi Farhansyah, Hanif Muhammad Zhafran, Farid Adilazuarda, et al.
+- **abs:** https://arxiv.org/abs/2601.17277 · **PDF:** https://arxiv.org/pdf/2601.17277v1
+
+## 19606. Latent-Space Contrastive Reinforcement Learning for Stable and Efficient LLM Reasoning
+
+- **ID:** `2601.17275` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Lianlei Shan, Han Chen, Yixuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.17275 · **PDF:** https://arxiv.org/pdf/2601.17275v1
+
+## 19607. Unrolled Neural Networks for Constrained Optimization
+
+- **ID:** `2601.17274` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Samar Hadou, Alejandro Ribeiro
+- **abs:** https://arxiv.org/abs/2601.17274 · **PDF:** https://arxiv.org/pdf/2601.17274v1
+
+## 19608. Window Size Versus Accuracy Experiments in Voice Activity Detectors
+
+- **ID:** `2601.17270` · **primary:** `cs.SD` · **Submitted:** 2026-01-24
+- **Authors:** Max McKinnon, Samir Khaki, Chandan KA Reddy, et al.
+- **abs:** https://arxiv.org/abs/2601.17270 · **PDF:** https://arxiv.org/pdf/2601.17270v1
+
+## 19609. GraphDancer: Training LLMs to Explore and Reason over Graphs via Curriculum Reinforcement Learning
+
+- **ID:** `2602.02518` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Yuyang Bai, Zhuofeng Li, Ping Nie, et al.
+- **abs:** https://arxiv.org/abs/2602.02518 · **PDF:** https://arxiv.org/pdf/2602.02518v1
+
+## 19610. Covariate-assisted Grade of Membership Models via Shared Latent Geometry
+
+- **ID:** `2601.17265` · **primary:** `stat.ME` · **Submitted:** 2026-01-24
+- **Authors:** Zhiyu Xu, Yuqi Gu
+- **abs:** https://arxiv.org/abs/2601.17265 · **PDF:** https://arxiv.org/pdf/2601.17265v1
+
+## 19611. AGZO: Activation-Guided Zeroth-Order Optimization for LLM Fine-Tuning
+
+- **ID:** `2601.17261` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Wei Lin, Yining Jiang, Qingyu Song, et al.
+- **abs:** https://arxiv.org/abs/2601.17261 · **PDF:** https://arxiv.org/pdf/2601.17261v3
+
+## 19612. The Viscosity of Logic: Phase Transitions and Hysteresis in DPO Alignment
+
+- **ID:** `2601.17260` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Marco Pollanen
+- **abs:** https://arxiv.org/abs/2601.17260 · **PDF:** https://arxiv.org/pdf/2601.17260v1
+
+## 19613. Inference-Time Loss-Guided Colour Preservation in Diffusion Sampling
+
+- **ID:** `2601.17259` · **primary:** `cs.CV` · **Submitted:** 2026-01-24
+- **Authors:** Angad Singh Ahuja, Aarush Ram Anandh
+- **abs:** https://arxiv.org/abs/2601.17259 · **PDF:** https://arxiv.org/pdf/2601.17259v1
+
+## 19614. A Constrained Optimization Perspective of Unrolled Transformers
+
+- **ID:** `2601.17257` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Javier Porras-Valenzuela, Samar Hadou, Alejandro Ribeiro
+- **abs:** https://arxiv.org/abs/2601.17257 · **PDF:** https://arxiv.org/pdf/2601.17257v1
+
+## 19615. What Drives Length of Stay After Elective Spine Surgery? Insights from a Decade of Predictive Modeling
+
+- **ID:** `2602.02517` · **primary:** `cs.LG` · **Submitted:** 2026-01-24
+- **Authors:** Ha Na Cho, Seungmin Jeong, Yawen Guo, et al.
+- **abs:** https://arxiv.org/abs/2602.02517 · **PDF:** https://arxiv.org/pdf/2602.02517v1
+
+## 19616. Building Safe and Deployable Clinical Natural Language Processing under Temporal Leakage Constraints
+
+- **ID:** `2602.15852` · **primary:** `cs.CL` · **Submitted:** 2026-01-24
+- **Authors:** Ha Na Cho, Sairam Sutari, Alexander Lopez, et al.
+- **abs:** https://arxiv.org/abs/2602.15852 · **PDF:** https://arxiv.org/pdf/2602.15852v2
+
+## 19617. Frame-Guided Synthetic Claim Generation for Automatic Fact-Checking Using High-Volume Tabular Data
+
+- **ID:** `2601.17232` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Jacob Devasier, Akshith Putta, Qing Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.17232 · **PDF:** https://arxiv.org/pdf/2601.17232v1
+
+## 19618. CaseFacts: A Benchmark for Legal Fact-Checking and Precedent Retrieval
+
+- **ID:** `2601.17230` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Akshith Reddy Putta, Jacob Devasier, Chengkai Li
+- **abs:** https://arxiv.org/abs/2601.17230 · **PDF:** https://arxiv.org/pdf/2601.17230v1
+
+## 19619. Narrative Theory-Driven LLM Methods for Automatic Story Generation and Understanding: A Survey
+
+- **ID:** `2602.15851` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** David Y. Liu, Aditya Joshi, Paul Dawson
+- **abs:** https://arxiv.org/abs/2602.15851 · **PDF:** https://arxiv.org/pdf/2602.15851v1
+
+## 19620. Retell, Reward, Repeat: Reinforcement Learning for Narrative Theory-Informed Story Generation
+
+- **ID:** `2601.17226` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** David Y. Liu, Xanthe Muston, Aditya Joshi, et al.
+- **abs:** https://arxiv.org/abs/2601.17226 · **PDF:** https://arxiv.org/pdf/2601.17226v1
+
+## 19621. Parameter Inference and Uncertainty Quantification with Diffusion Models: Extending CDI to 2D Spatial Conditioning
+
+- **ID:** `2601.17224` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Dmitrii Torbunov, Yihui Ren, Lijun Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.17224 · **PDF:** https://arxiv.org/pdf/2601.17224v1
+
+## 19622. Beyond Outcome Verification: Verifiable Process Reward Models for Structured Reasoning
+
+- **ID:** `2601.17223` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Massimiliano Pronesti, Anya Belz, Yufang Hou
+- **abs:** https://arxiv.org/abs/2601.17223 · **PDF:** https://arxiv.org/pdf/2601.17223v1
+
+## 19623. Improving Generalization and Uncertainty Quantification of Photometric Redshift Models
+
+- **ID:** `2601.17222` · **primary:** `astro-ph.IM` · **Submitted:** 2026-01-23
+- **Authors:** Jonathan Soriano, Tuan Do, Srinath Saikrishnan, et al.
+- **abs:** https://arxiv.org/abs/2601.17222 · **PDF:** https://arxiv.org/pdf/2601.17222v1
+
+## 19624. Evaluation on Entity Matching in Recommender Systems
+
+- **ID:** `2601.17218` · **primary:** `cs.IR` · **Submitted:** 2026-01-23
+- **Authors:** Zihan Huang, Rohan Surana, Zhouhang Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.17218 · **PDF:** https://arxiv.org/pdf/2601.17218v2
+
+## 19625. Transfer learning for functional linear regression via control variates
+
+- **ID:** `2601.17217` · **primary:** `stat.ME` · **Submitted:** 2026-01-23
+- **Authors:** Yuping Yang, Zhiyang Zhou
+- **abs:** https://arxiv.org/abs/2601.17217 · **PDF:** https://arxiv.org/pdf/2601.17217v4
+
+## 19626. Spatiotemporal Semantic V2X Framework for Cooperative Collision Prediction
+
+- **ID:** `2601.17216` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Murat Arda Onsu, Poonam Lohan, Burak Kantarci, et al.
+- **abs:** https://arxiv.org/abs/2601.17216 · **PDF:** https://arxiv.org/pdf/2601.17216v2
+
+## 19627. JetFormer: A Scalable and Efficient Transformer for Jet Tagging from Offline Analysis to FPGA Triggers
+
+- **ID:** `2601.17215` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Ruoqing Zheng, Chang Sun, Qibin Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.17215 · **PDF:** https://arxiv.org/pdf/2601.17215v1
+
+## 19628. DF-RAG: Query-Aware Diversity for Retrieval-Augmented Generation
+
+- **ID:** `2601.17212` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Saadat Hasan Khan, Spencer Hong, Jingyu Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.17212 · **PDF:** https://arxiv.org/pdf/2601.17212v1
+
+## 19629. NewPINNs: Physics-Informing Neural Networks Using Conventional Solvers for Partial Differential Equations
+
+- **ID:** `2601.17207` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Maedeh Makki, Satish Chandran, Maziar Raissi, et al.
+- **abs:** https://arxiv.org/abs/2601.17207 · **PDF:** https://arxiv.org/pdf/2601.17207v1
+
+## 19630. Benchmarking von ASR-Modellen im deutschen medizinischen Kontext: Eine Leistungsanalyse anhand von Anamnesegesprächen
+
+- **ID:** `2601.19945` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Thomas Schuster, Julius Trögele, Nico Döring, et al.
+- **abs:** https://arxiv.org/abs/2601.19945 · **PDF:** https://arxiv.org/pdf/2601.19945v1
+
+## 19631. SpecBridge: Bridging Mass Spectrometry and Molecular Representations via Cross-Modal Alignment
+
+- **ID:** `2601.17204` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Yinkai Wang, Yan Zhou Chen, Xiaohui Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.17204 · **PDF:** https://arxiv.org/pdf/2601.17204v3
+
+## 19632. Relating Word Embedding Gender Biases to Gender Gaps: A Cross-Cultural Analysis
+
+- **ID:** `2601.17203` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Scott Friedman, Sonja Schmer-Galunder, Anthony Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.17203 · **PDF:** https://arxiv.org/pdf/2601.17203v1
+
+## 19633. Reasoning Beyond Literal: Cross-style Multimodal Reasoning for Figurative Language Understanding
+
+- **ID:** `2601.17197` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Seyyed Saeid Cheshmi, Hahnemann Ortiz, James Mooney, et al.
+- **abs:** https://arxiv.org/abs/2601.17197 · **PDF:** https://arxiv.org/pdf/2601.17197v1
+
+## 19634. Accelerated Sinkhorn Algorithms for Partial Optimal Transport
+
+- **ID:** `2601.17196` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Nghia Thu Truong, Qui Phu Pham, Quang Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.17196 · **PDF:** https://arxiv.org/pdf/2601.17196v3
+
+## 19635. PUNCH: Physics-informed Uncertainty-aware Network for Coronary Hemodynamics
+
+- **ID:** `2601.17192` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Sukirt Thakur, Marcus Roper, Yang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.17192 · **PDF:** https://arxiv.org/pdf/2601.17192v3
+
+## 19636. Rethinking Benchmarks for Differentially Private Image Classification
+
+- **ID:** `2601.17189` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Sabrina Mokhtari, Sara Kodeiri, Shubhankar Mohapatra, et al.
+- **abs:** https://arxiv.org/abs/2601.17189 · **PDF:** https://arxiv.org/pdf/2601.17189v2
+
+## 19637. EDU-CIRCUIT-HW: Evaluating Multimodal Large Language Models on Real-World University-Level STEM Student Handwritten Solutions
+
+- **ID:** `2602.00095` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Weiyu Sun, Liangliang Chen, Yongnuo Cai, et al.
+- **abs:** https://arxiv.org/abs/2602.00095 · **PDF:** https://arxiv.org/pdf/2602.00095v2
+
+## 19638. Implementing Tensor Logic: Unifying Datalog and Neural Reasoning via Tensor Contraction
+
+- **ID:** `2601.17188` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Swapn Shah, Wlodek Zadrozny
+- **abs:** https://arxiv.org/abs/2601.17188 · **PDF:** https://arxiv.org/pdf/2601.17188v1
+
+## 19639. High-Rate Quantized Matrix Multiplication: Theory and Practice
+
+- **ID:** `2601.17187` · **primary:** `cs.IT` · **Submitted:** 2026-01-23
+- **Authors:** Or Ordentlich, Yury Polyanskiy
+- **abs:** https://arxiv.org/abs/2601.17187 · **PDF:** https://arxiv.org/pdf/2601.17187v1
+
+## 19640. FASTR: Reimagining FASTQ via Compact Image-inspired Representation
+
+- **ID:** `2601.17184` · **primary:** `q-bio.GN` · **Submitted:** 2026-01-23
+- **Authors:** Adrian Tkachenko, Sepehr Salem, Ayotomiwa Ezekiel Adeniyi, et al.
+- **abs:** https://arxiv.org/abs/2601.17184 · **PDF:** https://arxiv.org/pdf/2601.17184v1
+
+## 19641. Federated Proximal Optimization for Privacy-Preserving Heart Disease Prediction: A Controlled Simulation Study on Non-IID Clinical Data
+
+- **ID:** `2601.17183` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Farzam Asad, Junaid Saif Khan, Maria Tariq, et al.
+- **abs:** https://arxiv.org/abs/2601.17183 · **PDF:** https://arxiv.org/pdf/2601.17183v1
+
+## 19642. Systematicity between Forms and Meanings across Languages Supports Efficient Communication
+
+- **ID:** `2601.17181` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Doreen Osmelak, Yang Xu, Michael Hahn, et al.
+- **abs:** https://arxiv.org/abs/2601.17181 · **PDF:** https://arxiv.org/pdf/2601.17181v1
+
+## 19643. Conservative & Aggressive NaNs Accelerate U-Nets for Neuroimaging
+
+- **ID:** `2601.17180` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Inés Gonzalez-Pepe, Vinuyan Sivakolunthu, Jacob Fortin, et al.
+- **abs:** https://arxiv.org/abs/2601.17180 · **PDF:** https://arxiv.org/pdf/2601.17180v1
+
+## 19644. TrojanGYM: A Detector-in-the-Loop LLM for Adaptive RTL Hardware Trojan Insertion
+
+- **ID:** `2601.17178` · **primary:** `cs.CR` · **Submitted:** 2026-01-23
+- **Authors:** Saideep Sreekumar, Zeng Wang, Akashdeep Saha, et al.
+- **abs:** https://arxiv.org/abs/2601.17178 · **PDF:** https://arxiv.org/pdf/2601.17178v1
+
+## 19645. Beyond Factual QA: Mentorship-Oriented Question Answering over Long-Form Multilingual Content
+
+- **ID:** `2601.17173` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Parth Bhalerao, Diola Dsouza, Ruiwen Guan, et al.
+- **abs:** https://arxiv.org/abs/2601.17173 · **PDF:** https://arxiv.org/pdf/2601.17173v1
+
+## 19646. Who Gets Which Message? Auditing Demographic Bias in LLM-Generated Targeted Text
+
+- **ID:** `2601.17172` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Tunazzina Islam
+- **abs:** https://arxiv.org/abs/2601.17172 · **PDF:** https://arxiv.org/pdf/2601.17172v1
+
+## 19647. A Unified Kantorovich Duality for Multimarginal Optimal Transport
+
+- **ID:** `2601.17171` · **primary:** `math.OC` · **Submitted:** 2026-01-23
+- **Authors:** Yehya Cheryala, Mokhtar Z. Alaya, Salim Bouzebda
+- **abs:** https://arxiv.org/abs/2601.17171 · **PDF:** https://arxiv.org/pdf/2601.17171v1
+
+## 19648. Interpreting Agentic Systems: Beyond Model Explanations to System-Level Accountability
+
+- **ID:** `2601.17168` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Judy Zhu, Dhari Gandhi, Himanshu Joshi, et al.
+- **abs:** https://arxiv.org/abs/2601.17168 · **PDF:** https://arxiv.org/pdf/2601.17168v1
+
+## 19649. Benchmarking Deep Learning-Based Reconstruction Methods for Photoacoustic Computed Tomography with Clinically Relevant Synthetic Datasets
+
+- **ID:** `2601.17165` · **primary:** `physics.med-ph` · **Submitted:** 2026-01-23
+- **Authors:** Panpan Chen, Seonyeong Park, Gangwon Jeong, et al.
+- **abs:** https://arxiv.org/abs/2601.17165 · **PDF:** https://arxiv.org/pdf/2601.17165v1
+
+## 19650. Information-Theoretic Causal Bounds under Unmeasured Confounding
+
+- **ID:** `2601.17160` · **primary:** `stat.ML` · **Submitted:** 2026-01-23
+- **Authors:** Yonghan Jung, Bogyeong Kang
+- **abs:** https://arxiv.org/abs/2601.17160 · **PDF:** https://arxiv.org/pdf/2601.17160v3
+
+## 19651. Interpretability of the Intent Detection Problem: A New Approach
+
+- **ID:** `2601.17156` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Eduardo Sanchez-Karhunen, Jose F. Quesada-Moreno, Miguel A. Gutiérrez-Naranjo
+- **abs:** https://arxiv.org/abs/2601.17156 · **PDF:** https://arxiv.org/pdf/2601.17156v1
+
+## 19652. Dynamic Role Assignment for Multi-Agent Debate
+
+- **ID:** `2601.17152` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Miao Zhang, Junsik Kim, Siyuan Xiang, et al.
+- **abs:** https://arxiv.org/abs/2601.17152 · **PDF:** https://arxiv.org/pdf/2601.17152v1
+
+## 19653. Scaling medical imaging report generation with multimodal reinforcement learning
+
+- **ID:** `2601.17151` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Qianchu Liu, Sheng Zhang, Guanghui Qin, et al.
+- **abs:** https://arxiv.org/abs/2601.17151 · **PDF:** https://arxiv.org/pdf/2601.17151v1
+
+## 19654. Falsifying Predictive Algorithm
+
+- **ID:** `2601.17146` · **primary:** `stat.ME` · **Submitted:** 2026-01-23
+- **Authors:** Amanda Coston
+- **abs:** https://arxiv.org/abs/2601.17146 · **PDF:** https://arxiv.org/pdf/2601.17146v1
+
+## 19655. Large Language Models for Assisting American College Applications
+
+- **ID:** `2602.15850` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Zhengliang Liu, Weihang You, Peng Shu, et al.
+- **abs:** https://arxiv.org/abs/2602.15850 · **PDF:** https://arxiv.org/pdf/2602.15850v1
+
+## 19656. Trade-offs Between Individual and Group Fairness in Machine Learning: A Comprehensive Review
+
+- **ID:** `2602.00094` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Sandra Benítez-Peña, Blas Kolic, Victoria Menendez, et al.
+- **abs:** https://arxiv.org/abs/2602.00094 · **PDF:** https://arxiv.org/pdf/2602.00094v1
+
+## 19657. Communication-Avoiding Linear Algebraic Kernel K-Means on GPUs
+
+- **ID:** `2601.17136` · **primary:** `cs.DC` · **Submitted:** 2026-01-23
+- **Authors:** Julian Bellavita, Matthew Rubino, Nakul Iyer, et al.
+- **abs:** https://arxiv.org/abs/2601.17136 · **PDF:** https://arxiv.org/pdf/2601.17136v2
+
+## 19658. ConceptACT: Episode-Level Concepts for Sample-Efficient Robotic Imitation Learning
+
+- **ID:** `2601.17135` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Jakob Karalus, Friedhelm Schwenker
+- **abs:** https://arxiv.org/abs/2601.17135 · **PDF:** https://arxiv.org/pdf/2601.17135v1
+
+## 19659. Learning to Collaborate: An Orchestrated-Decentralized Framework for Peer-to-Peer LLM Federation
+
+- **ID:** `2601.17133` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Inderjeet Singh, Eleonore Vissol-Gaudin, Andikan Otung, et al.
+- **abs:** https://arxiv.org/abs/2601.17133 · **PDF:** https://arxiv.org/pdf/2601.17133v1
+
+## 19660. From Emotion to Expression: Theoretical Foundations and Resources for Fear Speech
+
+- **ID:** `2601.17132` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Vigneshwaran Shankaran, Gabriella Lapesa, Claudia Wagner
+- **abs:** https://arxiv.org/abs/2601.17132 · **PDF:** https://arxiv.org/pdf/2601.17132v1
+
+## 19661. How does Graph Structure Modulate Membership-Inference Risk for Graph Neural Networks?
+
+- **ID:** `2601.17130` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Megha Khosla
+- **abs:** https://arxiv.org/abs/2601.17130 · **PDF:** https://arxiv.org/pdf/2601.17130v1
+
+## 19662. EveNet: A Foundation Model for Particle Collision Data Analysis
+
+- **ID:** `2601.17126` · **primary:** `hep-ex` · **Submitted:** 2026-01-23
+- **Authors:** Ting-Hsiang Hsu, Bai-Hong Zhou, Qibin Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.17126 · **PDF:** https://arxiv.org/pdf/2601.17126v1
+
+## 19663. AnyView: Synthesizing Any Novel View in Dynamic Scenes
+
+- **ID:** `2601.16982` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Basile Van Hoorick, Dian Chen, Shun Iwase, et al.
+- **abs:** https://arxiv.org/abs/2601.16982 · **PDF:** https://arxiv.org/pdf/2601.16982v1
+
+## 19664. A Scalable Measure of Loss Landscape Curvature for Analyzing the Training Dynamics of LLMs
+
+- **ID:** `2601.16979` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Dayal Singh Kalra, Jean-Christophe Gagnon-Audet, Andrey Gromov, et al.
+- **abs:** https://arxiv.org/abs/2601.16979 · **PDF:** https://arxiv.org/pdf/2601.16979v2
+
+## 19665. IntelliAsk: Learning to Ask High-Quality Research Questions via RLVR
+
+- **ID:** `2602.15849` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Karun Sharma, Vidushee Vats, Shengzhi Li, et al.
+- **abs:** https://arxiv.org/abs/2602.15849 · **PDF:** https://arxiv.org/pdf/2602.15849v2
+
+## 19666. Latent Diffusion for Internet of Things Attack Data Generation in Intrusion Detection
+
+- **ID:** `2601.16976` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Estela Sánchez-Carballo, Francisco M. Melgarejo-Meseguer, José Luis Rojo-Álvarez
+- **abs:** https://arxiv.org/abs/2601.16976 · **PDF:** https://arxiv.org/pdf/2601.16976v1
+
+## 19667. Auto-Regressive Masked Diffusion Models
+
+- **ID:** `2601.16971` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Mahdi Karami, Ali Ghodsi
+- **abs:** https://arxiv.org/abs/2601.16971 · **PDF:** https://arxiv.org/pdf/2601.16971v1
+
+## 19668. BONO-Bench: A Comprehensive Test Suite for Bi-objective Numerical Optimization with Traceable Pareto Sets
+
+- **ID:** `2601.16970` · **primary:** `math.OC` · **Submitted:** 2026-01-23
+- **Authors:** Lennart Schäpermeier, Pascal Kerschke
+- **abs:** https://arxiv.org/abs/2601.16970 · **PDF:** https://arxiv.org/pdf/2601.16970v1
+
+## 19669. Empowering Medical Equipment Sustainability in Low-Resource Settings: An AI-Powered Diagnostic and Support Platform for Biomedical Technicians
+
+- **ID:** `2601.16967` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Bernes Lorier Atabonfack, Ahmed Tahiru Issah, Mohammed Hardi Abdul Baaki, et al.
+- **abs:** https://arxiv.org/abs/2601.16967 · **PDF:** https://arxiv.org/pdf/2601.16967v1
+
+## 19670. Low-Rank Tensor Approximation of Weights in Large Language Models via Cosine Lanczos Bidiagonalization
+
+- **ID:** `2601.17112` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** A. El Ichi, K. Jbilou
+- **abs:** https://arxiv.org/abs/2601.17112 · **PDF:** https://arxiv.org/pdf/2601.17112v1
+
+## 19671. Spatial-Agent: Agentic Geo-spatial Reasoning with Scientific Core Concepts
+
+- **ID:** `2601.16965` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Riyang Bao, Cheng Yang, Dazhou Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.16965 · **PDF:** https://arxiv.org/pdf/2601.16965v1
+
+## 19672. AgentDrive: An Open Benchmark Dataset for Agentic AI Reasoning with LLM-Generated Scenarios in Autonomous Systems
+
+- **ID:** `2601.16964` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Mohamed Amine Ferrag, Abderrahmane Lakas, Merouane Debbah
+- **abs:** https://arxiv.org/abs/2601.16964 · **PDF:** https://arxiv.org/pdf/2601.16964v1
+
+## 19673. 3D Molecule Generation from Rigid Motifs via SE(3) Flows
+
+- **ID:** `2601.16955` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Roman Poletukhin, Marcel Kollovieh, Eike Eberhard, et al.
+- **abs:** https://arxiv.org/abs/2601.16955 · **PDF:** https://arxiv.org/pdf/2601.16955v1
+
+## 19674. Least-Loaded Expert Parallelism: Load Balancing An Imbalanced Mixture-of-Experts
+
+- **ID:** `2601.17111` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Xuan-Phi Nguyen, Shrey Pandit, Austin Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.17111 · **PDF:** https://arxiv.org/pdf/2601.17111v1
+
+## 19675. Forecasting Energy Consumption using Recurrent Neural Networks: A Comparative Analysis
+
+- **ID:** `2601.17110` · **primary:** `cs.CY` · **Submitted:** 2026-01-23
+- **Authors:** Abhishek Maity, Viraj Tukarul
+- **abs:** https://arxiv.org/abs/2601.17110 · **PDF:** https://arxiv.org/pdf/2601.17110v1
+
+## 19676. Strategies for Span Labeling with Large Language Models
+
+- **ID:** `2601.16946` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Danil Semin, Ondřej Dušek, Zdeněk Kasner
+- **abs:** https://arxiv.org/abs/2601.16946 · **PDF:** https://arxiv.org/pdf/2601.16946v1
+
+## 19677. Autonomous Editorial Systems and Computational Investigation with Artificial Intelligence
+
+- **ID:** `2603.13232` · **primary:** `cs.DL` · **Submitted:** 2026-01-23
+- **Authors:** Ahmed Banafea
+- **abs:** https://arxiv.org/abs/2603.13232 · **PDF:** https://arxiv.org/pdf/2603.13232v1
+
+## 19678. Is BatchEnsemble a Single Model? On Calibration and Diversity of Efficient Ensembles
+
+- **ID:** `2601.16936` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Anton Zamyatin, Patrick Indri, Sagar Malhotra, et al.
+- **abs:** https://arxiv.org/abs/2601.16936 · **PDF:** https://arxiv.org/pdf/2601.16936v1
+
+## 19679. Information Representation Fairness in Long-Document Embeddings: The Peculiar Interaction of Positional and Language Bias
+
+- **ID:** `2601.16934` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Elias Schuhmacher, Andrianos Michail, Juri Opitz, et al.
+- **abs:** https://arxiv.org/abs/2601.16934 · **PDF:** https://arxiv.org/pdf/2601.16934v1
+
+## 19680. Reward-Forcing: Autoregressive Video Generation with Reward Feedback
+
+- **ID:** `2601.16933` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Jingran Zhang, Ning Li, Yuanhao Ban, et al.
+- **abs:** https://arxiv.org/abs/2601.16933 · **PDF:** https://arxiv.org/pdf/2601.16933v1
+
+## 19681. Authority Signals in AI Cited Health Sources: A Framework for Evaluating Source Credibility in ChatGPT Responses
+
+- **ID:** `2601.17109` · **primary:** `cs.DL` · **Submitted:** 2026-01-23
+- **Authors:** Erin Jacques, Erela Datuowei, Vincent Jones, et al.
+- **abs:** https://arxiv.org/abs/2601.17109 · **PDF:** https://arxiv.org/pdf/2601.17109v1
+
+## 19682. Nishpaksh: TEC Standard-Compliant Framework for Fairness Auditing and Certification of AI Models
+
+- **ID:** `2601.16926` · **primary:** `cs.CY` · **Submitted:** 2026-01-23
+- **Authors:** Shashank Prakash, Ranjitha Prasad, Avinash Agarwal
+- **abs:** https://arxiv.org/abs/2601.16926 · **PDF:** https://arxiv.org/pdf/2601.16926v1
+
+## 19683. Group-realizable multi-group learning by minimizing empirical risk
+
+- **ID:** `2601.16922` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Navid Ardeshir, Samuel Deng, Daniel Hsu, et al.
+- **abs:** https://arxiv.org/abs/2601.16922 · **PDF:** https://arxiv.org/pdf/2601.16922v1
+
+## 19684. MambaNet: Mamba-assisted Channel Estimation Neural Network With Attention Mechanism
+
+- **ID:** `2601.17108` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Dianxin Luan, Chengsi Liang, Jie Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.17108 · **PDF:** https://arxiv.org/pdf/2601.17108v1
+
+## 19685. LoL: Longer than Longer, Scaling Video Generation to Hour
+
+- **ID:** `2601.16914` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Justin Cui, Jie Wu, Ming Li, et al.
+- **abs:** https://arxiv.org/abs/2601.16914 · **PDF:** https://arxiv.org/pdf/2601.16914v1
+
+## 19686. Preventing the Collapse of Peer Review Requires Verification-First AI
+
+- **ID:** `2601.16909` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Lei You, Lele Cao, Iryna Gurevych
+- **abs:** https://arxiv.org/abs/2601.16909 · **PDF:** https://arxiv.org/pdf/2601.16909v2
+
+## 19687. Calibrated Similarity for Reliable Geometric Analysis of Embedding Spaces
+
+- **ID:** `2601.16907` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Nicolas Tacheny
+- **abs:** https://arxiv.org/abs/2601.16907 · **PDF:** https://arxiv.org/pdf/2601.16907v1
+
+## 19688. The Trajectory Alignment Coefficient in Two Acts: From Reward Tuning to Reward Learning
+
+- **ID:** `2601.16906` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Calarina Muslimani, Yunshu Du, Kenta Kawamoto, et al.
+- **abs:** https://arxiv.org/abs/2601.16906 · **PDF:** https://arxiv.org/pdf/2601.16906v1
+
+## 19689. GRIP: Algorithm-Agnostic Machine Unlearning for Mixture-of-Experts via Geometric Router Constraints
+
+- **ID:** `2601.16905` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Andy Zhu, Rongzhe Wei, Yupu Gu, et al.
+- **abs:** https://arxiv.org/abs/2601.16905 · **PDF:** https://arxiv.org/pdf/2601.16905v2
+
+## 19690. Embedding -based Crop Type Classification in the Groundnut Basin of Senegal
+
+- **ID:** `2601.16900` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Madeline C. Lisaius, Srinivasan Keshav, Andrew Blake, et al.
+- **abs:** https://arxiv.org/abs/2601.16900 · **PDF:** https://arxiv.org/pdf/2601.16900v1
+
+## 19691. FedSGM: A Unified Framework for Constraint Aware, Bidirectionally Compressed, Multi-Step Federated Optimization
+
+- **ID:** `2601.16897` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Antesh Upadhyay, Sang Bin Moon, Abolfazl Hashemi
+- **abs:** https://arxiv.org/abs/2601.16897 · **PDF:** https://arxiv.org/pdf/2601.16897v1
+
+## 19692. How Sequential Algorithm Portfolios can benefit Black Box Optimization
+
+- **ID:** `2601.16896` · **primary:** `cs.NE` · **Submitted:** 2026-01-23
+- **Authors:** Catalin-Viorel Dinu, Diederick Vermetten, Carola Doerr
+- **abs:** https://arxiv.org/abs/2601.16896 · **PDF:** https://arxiv.org/pdf/2601.16896v1
+
+## 19693. Evaluating Large Vision-language Models for Surgical Tool Detection
+
+- **ID:** `2601.16895` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Nakul Poudel, Richard Simon, Cristian A. Linte
+- **abs:** https://arxiv.org/abs/2601.16895 · **PDF:** https://arxiv.org/pdf/2601.16895v1
+
+## 19694. LLM-Based Adversarial Persuasion Attacks on Fact-Checking Systems
+
+- **ID:** `2601.16890` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** João A. Leite, Olesya Razuvayevskaya, Kalina Bontcheva, et al.
+- **abs:** https://arxiv.org/abs/2601.16890 · **PDF:** https://arxiv.org/pdf/2601.16890v1
+
+## 19695. Interpreting and Controlling Model Behavior via Constitutions for Atomic Concept Edits
+
+- **ID:** `2602.00092` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Neha Kalibhat, Zi Wang, Prasoon Bajpai, et al.
+- **abs:** https://arxiv.org/abs/2602.00092 · **PDF:** https://arxiv.org/pdf/2602.00092v1
+
+## 19696. MAGE-KT: Multi-Agent Graph-Enhanced Knowledge Tracing with Subgraph Retrieval and Asymmetric Fusion
+
+- **ID:** `2601.16886` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Chi Yu, Hongyu Yuan, Zhiyi Duan
+- **abs:** https://arxiv.org/abs/2601.16886 · **PDF:** https://arxiv.org/pdf/2601.16886v1
+
+## 19697. Multigrade Neural Network Approximation
+
+- **ID:** `2601.16884` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Shijun Zhang, Zuowei Shen, Yuesheng Xu
+- **abs:** https://arxiv.org/abs/2601.16884 · **PDF:** https://arxiv.org/pdf/2601.16884v2
+
+## 19698. Performance uncertainty in medical image analysis: a large-scale investigation of confidence intervals
+
+- **ID:** `2601.17103` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Pascaline André, Charles Heitz, Evangelia Christodoulou, et al.
+- **abs:** https://arxiv.org/abs/2601.17103 · **PDF:** https://arxiv.org/pdf/2601.17103v1
+
+## 19699. Explaining Group Recommendations via Counterfactuals
+
+- **ID:** `2601.16882` · **primary:** `cs.IR` · **Submitted:** 2026-01-23
+- **Authors:** Maria Stratigi, Nikos Bikakis, Kostas Stefanidis
+- **abs:** https://arxiv.org/abs/2601.16882 · **PDF:** https://arxiv.org/pdf/2601.16882v2
+
+## 19700. Theory of Minimal Weight Perturbations in Deep Networks and its Applications for Low-Rank Activated Backdoor Attacks
+
+- **ID:** `2601.16880` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Bethan Evans, Jared Tanner
+- **abs:** https://arxiv.org/abs/2601.16880 · **PDF:** https://arxiv.org/pdf/2601.16880v1
+
+## 19701. Model-Centric Diagnostics: A Framework for Internal State Readouts
+
+- **ID:** `2601.16874` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Fangzheng Wu, Brian Summa
+- **abs:** https://arxiv.org/abs/2601.16874 · **PDF:** https://arxiv.org/pdf/2601.16874v2
+
+## 19702. Provably Learning Attention with Queries
+
+- **ID:** `2601.16873` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Satwik Bhattamishra, Kulin Shah, Michael Hahn, et al.
+- **abs:** https://arxiv.org/abs/2601.16873 · **PDF:** https://arxiv.org/pdf/2601.16873v1
+
+## 19703. Boosting Deep Reinforcement Learning with Semantic Knowledge for Robotic Manipulators
+
+- **ID:** `2601.16866` · **primary:** `cs.RO` · **Submitted:** 2026-01-23
+- **Authors:** Lucía Güitta-López, Vincenzo Suriani, Jaime Boal, et al.
+- **abs:** https://arxiv.org/abs/2601.16866 · **PDF:** https://arxiv.org/pdf/2601.16866v1
+
+## 19704. Mixture-of-Models: Unifying Heterogeneous Agents via N-Way Self-Evaluating Deliberation
+
+- **ID:** `2601.16863` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Tims Pecerskis, Aivars Smirnovs
+- **abs:** https://arxiv.org/abs/2601.16863 · **PDF:** https://arxiv.org/pdf/2601.16863v1
+
+## 19705. Orbitopal Fixing in SAT
+
+- **ID:** `2601.16855` · **primary:** `cs.LO` · **Submitted:** 2026-01-23
+- **Authors:** Markus Anders, Cayden Codel, Marijn J. H. Heule
+- **abs:** https://arxiv.org/abs/2601.16855 · **PDF:** https://arxiv.org/pdf/2601.16855v1
+
+## 19706. Reasoning Promotes Robustness in Theory of Mind Tasks
+
+- **ID:** `2601.16853` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Ian B. de Haan, Peter van der Putten, Max van Duijn
+- **abs:** https://arxiv.org/abs/2601.16853 · **PDF:** https://arxiv.org/pdf/2601.16853v1
+
+## 19707. The Art of Being Difficult: Combining Human and AI Strengths to Find Adversarial Instances for Heuristics
+
+- **ID:** `2601.16849` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Henri Nikoleit, Ankit Anand, Anurag Murty Naredla, et al.
+- **abs:** https://arxiv.org/abs/2601.16849 · **PDF:** https://arxiv.org/pdf/2601.16849v1
+
+## 19708. LabelKAN -- Kolmogorov-Arnold Networks for Inter-Label Learning: Avian Community Learning
+
+- **ID:** `2601.18818` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-23
+- **Authors:** Marc Grimson, Joshua Fan, Courtney L. Davis, et al.
+- **abs:** https://arxiv.org/abs/2601.18818 · **PDF:** https://arxiv.org/pdf/2601.18818v1
+
+## 19709. Can LLMs Assess Personality? Validating Conversational AI for Trait Profiling
+
+- **ID:** `2602.15848` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Andrius Matšenas, Anet Lello, Tõnis Lees, et al.
+- **abs:** https://arxiv.org/abs/2602.15848 · **PDF:** https://arxiv.org/pdf/2602.15848v1
+
+## 19710. Parametric Mean-Field empirical Bayes in high-dimensional linear regression
+
+- **ID:** `2601.16842` · **primary:** `math.ST` · **Submitted:** 2026-01-23
+- **Authors:** Seunghyun Lee, Nabarun Deb
+- **abs:** https://arxiv.org/abs/2601.16842 · **PDF:** https://arxiv.org/pdf/2601.16842v2
+
+## 19711. Do Personality Traits Interfere? Geometric Limitations of Steering in Large Language Models
+
+- **ID:** `2602.15847` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Pranav Bhandari, Usman Naseem, Mehwish Nasim
+- **abs:** https://arxiv.org/abs/2602.15847 · **PDF:** https://arxiv.org/pdf/2602.15847v1
+
+## 19712. ColorConceptBench: A Benchmark for Probabilistic Color-Concept Understanding in Text-to-Image Models
+
+- **ID:** `2601.16836` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Chenxi Ruan, Yu Xiao, Yihan Hou, et al.
+- **abs:** https://arxiv.org/abs/2601.16836 · **PDF:** https://arxiv.org/pdf/2601.16836v2
+
+## 19713. Interpolation of GEDI Biomass Estimates with Calibrated Uncertainty Quantification
+
+- **ID:** `2601.16834` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Robin Young, Srinivasan Keshav
+- **abs:** https://arxiv.org/abs/2601.16834 · **PDF:** https://arxiv.org/pdf/2601.16834v3
+
+## 19714. Uncertainty propagation through trained multi-layer perceptrons: Exact analytical results
+
+- **ID:** `2601.16830` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Andrew Thompson, Miles McCrory
+- **abs:** https://arxiv.org/abs/2601.16830 · **PDF:** https://arxiv.org/pdf/2601.16830v1
+
+## 19715. Privacy in Human-AI Romantic Relationships: Concerns, Boundaries, and Agency
+
+- **ID:** `2601.16824` · **primary:** `cs.HC` · **Submitted:** 2026-01-23
+- **Authors:** Rongjun Ma, Shijing He, Jose Luis Martin-Navarro, et al.
+- **abs:** https://arxiv.org/abs/2601.16824 · **PDF:** https://arxiv.org/pdf/2601.16824v2
+
+## 19716. Trapped in the past? Disentangling fluid and crystallized intelligence of large language models using chess
+
+- **ID:** `2601.16823` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Leonard S. Pleiss, Maximilian Schiffer, Robert K. von Weizsäcker
+- **abs:** https://arxiv.org/abs/2601.16823 · **PDF:** https://arxiv.org/pdf/2601.16823v1
+
+## 19717. Sample-wise Constrained Learning via a Sequential Penalty Approach with Applications in Image Processing
+
+- **ID:** `2601.16812` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Francesca Lanzillotta, Chiara Albisani, Davide Pucci, et al.
+- **abs:** https://arxiv.org/abs/2601.16812 · **PDF:** https://arxiv.org/pdf/2601.16812v1
+
+## 19718. Incorporating Eye-Tracking Signals Into Multimodal Deep Visual Models For Predicting User Aesthetic Experience In Residential Interiors
+
+- **ID:** `2601.16811` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Chen-Ying Chien, Po-Chih Kuo
+- **abs:** https://arxiv.org/abs/2601.16811 · **PDF:** https://arxiv.org/pdf/2601.16811v1
+
+## 19719. Will It Survive? Deciphering the Fate of AI-Generated Code in Open Source
+
+- **ID:** `2601.16809` · **primary:** `cs.SE` · **Submitted:** 2026-01-23
+- **Authors:** Musfiqur Rahman, Emad Shihab
+- **abs:** https://arxiv.org/abs/2601.16809 · **PDF:** https://arxiv.org/pdf/2601.16809v1
+
+## 19720. An Efficient Insect-inspired Approach for Visual Point-goal Navigation
+
+- **ID:** `2601.16806` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Lu Yihe, Barbara Webb
+- **abs:** https://arxiv.org/abs/2601.16806 · **PDF:** https://arxiv.org/pdf/2601.16806v1
+
+## 19721. SoS: Analysis of Surface over Semantics in Multilingual Text-To-Image Generation
+
+- **ID:** `2601.16803` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Carolin Holtermann, Florian Schneider, Anne Lauscher
+- **abs:** https://arxiv.org/abs/2601.16803 · **PDF:** https://arxiv.org/pdf/2601.16803v1
+
+## 19722. Large Language Models as Automatic Annotators and Annotation Adjudicators for Fine-Grained Opinion Analysis
+
+- **ID:** `2601.16800` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Gaurav Negi, MA Waskow, John McCrae, et al.
+- **abs:** https://arxiv.org/abs/2601.16800 · **PDF:** https://arxiv.org/pdf/2601.16800v2
+
+## 19723. Building a Robust Risk-Based Access Control System to Combat Ransomware's Capability to Encrypt
+
+- **ID:** `2601.16795` · **primary:** `cs.CR` · **Submitted:** 2026-01-23
+- **Authors:** Kenan Begovic, Abdulaziz Al-Ali, Qutaibah Malluhi
+- **abs:** https://arxiv.org/abs/2601.16795 · **PDF:** https://arxiv.org/pdf/2601.16795v2
+
+## 19724. A Novel Transfer Learning Approach for Mental Stability Classification from Voice Signal
+
+- **ID:** `2601.16793` · **primary:** `cs.SD` · **Submitted:** 2026-01-23
+- **Authors:** Rafiul Islam, Md. Taimur Ahad
+- **abs:** https://arxiv.org/abs/2601.16793 · **PDF:** https://arxiv.org/pdf/2601.16793v1
+
+## 19725. REL-SF4PASS: Panoramic Semantic Segmentation with REL Depth Representation and Spherical Fusion
+
+- **ID:** `2601.16788` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Xuewei Li, Xinghan Bao, Zhimin Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.16788 · **PDF:** https://arxiv.org/pdf/2601.16788v1
+
+## 19726. Persuasion Tokens for Editing Factual Knowledge in LLMs
+
+- **ID:** `2601.16781` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Paul Youssef, Christin Seifert, Jörg Schlötterer
+- **abs:** https://arxiv.org/abs/2601.16781 · **PDF:** https://arxiv.org/pdf/2601.16781v2
+
+## 19727. Talking to Yourself: Defying Forgetting in Large Language Models
+
+- **ID:** `2602.20162` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Yutao Sun, Mingshuai Chen, Tiancheng Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.20162 · **PDF:** https://arxiv.org/pdf/2602.20162v1
+
+## 19728. GTA: Generative Traffic Agents for Simulating Realistic Mobility Behavior
+
+- **ID:** `2601.16778` · **primary:** `cs.HC` · **Submitted:** 2026-01-23
+- **Authors:** Simon Lämmer, Mark Colley, Patrick Ebel
+- **abs:** https://arxiv.org/abs/2601.16778 · **PDF:** https://arxiv.org/pdf/2601.16778v2
+
+## 19729. Kernel smoothing on manifolds
+
+- **ID:** `2601.16777` · **primary:** `math.ST` · **Submitted:** 2026-01-23
+- **Authors:** Eunseong Bae, Wolfgang Polonik
+- **abs:** https://arxiv.org/abs/2601.16777 · **PDF:** https://arxiv.org/pdf/2601.16777v1
+
+## 19730. Measuring Individual User Fairness with User Similarity and Effectiveness Disparity
+
+- **ID:** `2602.02516` · **primary:** `cs.CY` · **Submitted:** 2026-01-23
+- **Authors:** Theresia Veronika Rampisela, Maria Maistro, Tuukka Ruotsalo, et al.
+- **abs:** https://arxiv.org/abs/2602.02516 · **PDF:** https://arxiv.org/pdf/2602.02516v1
+
+## 19731. AutoRegressive Generation with B-rep Holistic Token Sequence Representation
+
+- **ID:** `2601.16771` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Jiahao Li, Yunpeng Bai, Yongkang Dai, et al.
+- **abs:** https://arxiv.org/abs/2601.16771 · **PDF:** https://arxiv.org/pdf/2601.16771v2
+
+## 19732. Do LLM hallucination detectors suffer from low-resource effect?
+
+- **ID:** `2601.16766` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Debtanu Datta, Mohan Kishore Chilukuri, Yash Kumar, et al.
+- **abs:** https://arxiv.org/abs/2601.16766 · **PDF:** https://arxiv.org/pdf/2601.16766v1
+
+## 19733. ReLU Networks for Model Predictive Control: Network Complexity and Performance Guarantees
+
+- **ID:** `2601.16764` · **primary:** `eess.SY` · **Submitted:** 2026-01-23
+- **Authors:** Xingchen Li, Keyou You
+- **abs:** https://arxiv.org/abs/2601.16764 · **PDF:** https://arxiv.org/pdf/2601.16764v1
+
+## 19734. Curated endoscopic retrograde cholangiopancreatography images dataset
+
+- **ID:** `2601.16759` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Alda João Andrade, Mónica Martins, André Ferreira, et al.
+- **abs:** https://arxiv.org/abs/2601.16759 · **PDF:** https://arxiv.org/pdf/2601.16759v1
+
+## 19735. Standardizing Longitudinal Radiology Report Evaluation via Large Language Model Annotation
+
+- **ID:** `2601.16753` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Xinyi Wang, Grazziela Figueredo, Ruizhe Li, et al.
+- **abs:** https://arxiv.org/abs/2601.16753 · **PDF:** https://arxiv.org/pdf/2601.16753v1
+
+## 19736. SWE-Pruner: Self-Adaptive Context Pruning for Coding Agents
+
+- **ID:** `2601.16746` · **primary:** `cs.SE` · **Submitted:** 2026-01-23
+- **Authors:** Yuhang Wang, Yuling Shi, Mo Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.16746 · **PDF:** https://arxiv.org/pdf/2601.16746v2
+
+## 19737. Scaling the Scaling Logic: Agentic Meta-Synthesis of Logic Reasoning
+
+- **ID:** `2602.13218` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Bowen Liu, Zhi Wu, Runquan Xie, et al.
+- **abs:** https://arxiv.org/abs/2602.13218 · **PDF:** https://arxiv.org/pdf/2602.13218v1
+
+## 19738. PENGUIN: General Vital Sign Reconstruction from PPG with Flow Matching State Space Model
+
+- **ID:** `2602.03858` · **primary:** `eess.SP` · **Submitted:** 2026-01-23
+- **Authors:** Shuntaro Suzuki, Shuitsu Koyama, Shinnosuke Hirano, et al.
+- **abs:** https://arxiv.org/abs/2602.03858 · **PDF:** https://arxiv.org/pdf/2602.03858v1
+
+## 19739. LongCat-Flash-Thinking-2601 Technical Report
+
+- **ID:** `2601.16725` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Meituan LongCat Team, Anchun Gui, Bei Li, et al.
+- **abs:** https://arxiv.org/abs/2601.16725 · **PDF:** https://arxiv.org/pdf/2601.16725v2
+
+## 19740. Mitigating Bias in Automated Grading Systems for ESL Learners: A Contrastive Learning Approach
+
+- **ID:** `2601.16724` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Kevin Fan, Eric Yun
+- **abs:** https://arxiv.org/abs/2601.16724 · **PDF:** https://arxiv.org/pdf/2601.16724v1
+
+## 19741. Beyond Instrumental and Substitutive Paradigms: Introducing Machine Culture as an Emergent Phenomenon in Large Language Models
+
+- **ID:** `2601.17096` · **primary:** `cs.CY` · **Submitted:** 2026-01-23
+- **Authors:** Yueqing Hu, Xinyang Peng, Yukun Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.17096 · **PDF:** https://arxiv.org/pdf/2601.17096v1
+
+## 19742. Dynamic Expert-Guided Model Averaging for Causal Discovery
+
+- **ID:** `2601.16715` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Adrick Tench, Thomas Demeester
+- **abs:** https://arxiv.org/abs/2601.16715 · **PDF:** https://arxiv.org/pdf/2601.16715v1
+
+## 19743. A Feature Extraction Pipeline for Enhancing Lightweight Neural Networks in sEMG-based Joint Torque Estimation
+
+- **ID:** `2601.16712` · **primary:** `cs.RO` · **Submitted:** 2026-01-23
+- **Authors:** Kartik Chari, Raid Dokhan, Anas Homsi, et al.
+- **abs:** https://arxiv.org/abs/2601.16712 · **PDF:** https://arxiv.org/pdf/2601.16712v2
+
+## 19744. Better Generalizing to Unseen Concepts: An Evaluation Framework and An LLM-Based Auto-Labeled Pipeline for Biomedical Concept Recognition
+
+- **ID:** `2601.16711` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Shanshan Liu, Noriki Nishida, Fei Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.16711 · **PDF:** https://arxiv.org/pdf/2601.16711v1
+
+## 19745. LoD Sketch Extraction from Architectural Models Using Generative AI: Dataset Construction for Multi-Level Architectural Design Generation
+
+- **ID:** `2601.17095` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Xusheng Du, Athiwat Kongkaeo, Ye Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.17095 · **PDF:** https://arxiv.org/pdf/2601.17095v1
+
+## 19746. Adoption of Generative Artificial Intelligence in the German Software Engineering Industry: An Empirical Study
+
+- **ID:** `2601.16700` · **primary:** `cs.SE` · **Submitted:** 2026-01-23
+- **Authors:** Ludwig Felder, Tobias Eisenreich, Mahsa Fischer, et al.
+- **abs:** https://arxiv.org/abs/2601.16700 · **PDF:** https://arxiv.org/pdf/2601.16700v1
+
+## 19747. The Mouth is Not the Brain: Bridging Energy-Based World Models and Language Generation
+
+- **ID:** `2601.17094` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Junichiro Niimi
+- **abs:** https://arxiv.org/abs/2601.17094 · **PDF:** https://arxiv.org/pdf/2601.17094v2
+
+## 19748. The Triangle of Similarity: A Multi-Faceted Framework for Comparing Neural Network Representations
+
+- **ID:** `2601.17093` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Olha Sirikova, Alvin Chan
+- **abs:** https://arxiv.org/abs/2601.17093 · **PDF:** https://arxiv.org/pdf/2601.17093v1
+
+## 19749. EMemBench: Interactive Benchmarking of Episodic Memory for VLM Agents
+
+- **ID:** `2601.16690` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Xinze Li, Ziyue Zhu, Siyuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.16690 · **PDF:** https://arxiv.org/pdf/2601.16690v1
+
+## 19750. AgentsEval: Clinically Faithful Evaluation of Medical Imaging Reports via Multi-Agent Reasoning
+
+- **ID:** `2601.16685` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Suzhong Fu, Jingqi Dong, Xuan Ding, et al.
+- **abs:** https://arxiv.org/abs/2601.16685 · **PDF:** https://arxiv.org/pdf/2601.16685v1
+
+## 19751. The Turing Synthetic Radar Dataset: A dataset for pulse deinterleaving
+
+- **ID:** `2602.03856` · **primary:** `eess.SP` · **Submitted:** 2026-01-23
+- **Authors:** Edward Gunn, Adam Hosford, Robert Jones, et al.
+- **abs:** https://arxiv.org/abs/2602.03856 · **PDF:** https://arxiv.org/pdf/2602.03856v1
+
+## 19752. Sim-to-Real Transfer via a Style-Identified Cycle Consistent Generative Adversarial Network: Zero-Shot Deployment on Robotic Manipulators through Visual Domain Adaptation
+
+- **ID:** `2601.16677` · **primary:** `cs.RO` · **Submitted:** 2026-01-23
+- **Authors:** Lucía Güitta-López, Lionel Güitta-López, Jaime Boal, et al.
+- **abs:** https://arxiv.org/abs/2601.16677 · **PDF:** https://arxiv.org/pdf/2601.16677v1
+
+## 19753. I Guess That's Why They Call it the Blues: Causal Analysis for Audio Classifiers
+
+- **ID:** `2601.16675` · **primary:** `cs.SD` · **Submitted:** 2026-01-23
+- **Authors:** David A. Kelly, Hana Chockler
+- **abs:** https://arxiv.org/abs/2601.16675 · **PDF:** https://arxiv.org/pdf/2601.16675v1
+
+## 19754. PLawBench: A Rubric-Based Benchmark for Evaluating LLMs in Real-World Legal Practice
+
+- **ID:** `2601.16669` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Yuzhen Shi, Huanghai Liu, Yiran Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.16669 · **PDF:** https://arxiv.org/pdf/2601.16669v2
+
+## 19755. Revisiting the Role of Natural Language Code Comments in Code Translation
+
+- **ID:** `2601.16661` · **primary:** `cs.SE` · **Submitted:** 2026-01-23
+- **Authors:** Monika Gupta, Ajay Meena, Anamitra Roy Choudhury, et al.
+- **abs:** https://arxiv.org/abs/2601.16661 · **PDF:** https://arxiv.org/pdf/2601.16661v1
+
+## 19756. Fast, faithful and photorealistic diffusion-based image super-resolution with enhanced Flow Map models
+
+- **ID:** `2601.16660` · **primary:** `eess.IV` · **Submitted:** 2026-01-23
+- **Authors:** Maxence Noble, Gonzalo Iñaki Quintana, Benjamin Aubin, et al.
+- **abs:** https://arxiv.org/abs/2601.16660 · **PDF:** https://arxiv.org/pdf/2601.16660v1
+
+## 19757. Provably Robust Bayesian Counterfactual Explanations under Model Changes
+
+- **ID:** `2601.16659` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Jamie Duell, Xiuyi Fan
+- **abs:** https://arxiv.org/abs/2601.16659 · **PDF:** https://arxiv.org/pdf/2601.16659v2
+
+## 19758. CUROCKET: Optimizing ROCKET for GPU
+
+- **ID:** `2601.17091` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Ole Stüven, Keno Moenck, Thorsten Schüppstuhl
+- **abs:** https://arxiv.org/abs/2601.17091 · **PDF:** https://arxiv.org/pdf/2601.17091v1
+
+## 19759. Generative Confidants: How do People Experience Trust in Emotional Support from Generative AI?
+
+- **ID:** `2601.16656` · **primary:** `cs.HC` · **Submitted:** 2026-01-23
+- **Authors:** Riccardo Volpato, Simone Stumpf, Lisa DeBruine
+- **abs:** https://arxiv.org/abs/2601.16656 · **PDF:** https://arxiv.org/pdf/2601.16656v1
+
+## 19760. Reliable Brain Tumor Segmentation Based on Spiking Neural Networks with Efficient Training
+
+- **ID:** `2601.16652` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Aurora Pia Ghiardelli, Guangzhi Tang, Tao Sun
+- **abs:** https://arxiv.org/abs/2601.16652 · **PDF:** https://arxiv.org/pdf/2601.16652v1
+
+## 19761. Select or Project? Evaluating Lower-dimensional Vectors for LLM Training Data Explanations
+
+- **ID:** `2601.16651` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Lukas Hinterleitner, Loris Schoenegger, Benjamin Roth
+- **abs:** https://arxiv.org/abs/2601.16651 · **PDF:** https://arxiv.org/pdf/2601.16651v2
+
+## 19762. LUMINA: Long-horizon Understanding for Multi-turn Interactive Agents
+
+- **ID:** `2601.16649` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Amin Rakhsha, Thomas Hehn, Pietro Mazzaglia, et al.
+- **abs:** https://arxiv.org/abs/2601.16649 · **PDF:** https://arxiv.org/pdf/2601.16649v1
+
+## 19763. Sycophancy Hides Linearly in the Attention Heads
+
+- **ID:** `2601.16644` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Rifo Genadi, Munachiso Nwadike, Nurdaulet Mukhituly, et al.
+- **abs:** https://arxiv.org/abs/2601.16644 · **PDF:** https://arxiv.org/pdf/2601.16644v1
+
+## 19764. Conformal prediction for full and sparse polynomial chaos expansions
+
+- **ID:** `2601.16636` · **primary:** `stat.ME` · **Submitted:** 2026-01-23
+- **Authors:** A. Hatstatt, X. Zhu, B. Sudret
+- **abs:** https://arxiv.org/abs/2601.16636 · **PDF:** https://arxiv.org/pdf/2601.16636v1
+
+## 19765. SFO: Learning PDE Operators via Spectral Filtering
+
+- **ID:** `2601.17090` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Noam Koren, Rafael Moschopoulos, Kira Radinsky, et al.
+- **abs:** https://arxiv.org/abs/2601.17090 · **PDF:** https://arxiv.org/pdf/2601.17090v1
+
+## 19766. Majorization-Minimization Networks for Inverse Problems: An Application to EEG Imaging
+
+- **ID:** `2602.03855` · **primary:** `eess.SP` · **Submitted:** 2026-01-23
+- **Authors:** Le Minh Triet Tran, Sarah Reynaud, Ronan Fablet, et al.
+- **abs:** https://arxiv.org/abs/2602.03855 · **PDF:** https://arxiv.org/pdf/2602.03855v1
+
+## 19767. Dual-Prototype Disentanglement: A Context-Aware Enhancement Framework for Time Series Forecasting
+
+- **ID:** `2601.16632` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Haonan Yang, Jianchao Tang, Zhuo Li
+- **abs:** https://arxiv.org/abs/2601.16632 · **PDF:** https://arxiv.org/pdf/2601.16632v3
+
+## 19768. Typologically Informed Parameter Aggregation
+
+- **ID:** `2601.16629` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Stef Accou, Wessel Poelman
+- **abs:** https://arxiv.org/abs/2601.16629 · **PDF:** https://arxiv.org/pdf/2601.16629v1
+
+## 19769. Gated Tree Cross-attention for Checkpoint-Compatible Syntax Injection in Decoder-Only LLMs
+
+- **ID:** `2602.15846` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Xinyu Gao, Shaonan Wang, Nai Ding
+- **abs:** https://arxiv.org/abs/2602.15846 · **PDF:** https://arxiv.org/pdf/2602.15846v1
+
+## 19770. MultiLexNorm++: A Unified Benchmark and a Generative Model for Lexical Normalization for Asian Languages
+
+- **ID:** `2601.16623` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Weerayut Buaphet, Thanh-Nhi Nguyen, Risa Kondo, et al.
+- **abs:** https://arxiv.org/abs/2601.16623 · **PDF:** https://arxiv.org/pdf/2601.16623v1
+
+## 19771. E2Former-V2: On-the-Fly Equivariant Attention with Linear Activation Memory
+
+- **ID:** `2601.16622` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Lin Huang, Chengxiang Huang, Ziang Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.16622 · **PDF:** https://arxiv.org/pdf/2601.16622v1
+
+## 19772. How Does Personalized Memory Shape LLM Behavior? Benchmarking Rational Preference Utilization in Personalized Assistants
+
+- **ID:** `2601.16621` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Xueyang Feng, Weinan Gan, Xu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.16621 · **PDF:** https://arxiv.org/pdf/2601.16621v1
+
+## 19773. PROST-LLM: Progressively Enhancing the Speech-to-Speech Translation Capability in LLMs
+
+- **ID:** `2601.16618` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Jing Xu, Jiaqi Wang, Daxin Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.16618 · **PDF:** https://arxiv.org/pdf/2601.16618v1
+
+## 19774. Boundary and Position Information Mining for Aerial Small Object Detection
+
+- **ID:** `2601.16617` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Rongxin Huang, Guangfeng Lin, Wenbo Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.16617 · **PDF:** https://arxiv.org/pdf/2601.16617v1
+
+## 19775. AuroraEdge-V-2B: A Faster And Stronger Edge Visual Large Language Model
+
+- **ID:** `2601.16615` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Xiang Chen
+- **abs:** https://arxiv.org/abs/2601.16615 · **PDF:** https://arxiv.org/pdf/2601.16615v1
+
+## 19776. A Lightweight Medical Image Classification Framework via Self-Supervised Contrastive Learning and Quantum-Enhanced Feature Modeling
+
+- **ID:** `2601.16608` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Jingsong Xia, Siqi Wang
+- **abs:** https://arxiv.org/abs/2601.16608 · **PDF:** https://arxiv.org/pdf/2601.16608v1
+
+## 19777. Efficient Learning of Stationary Diffusions with Stein-type Discrepancies
+
+- **ID:** `2601.16597` · **primary:** `stat.ML` · **Submitted:** 2026-01-23
+- **Authors:** Fabian Bleile, Sarah Lumpp, Mathias Drton
+- **abs:** https://arxiv.org/abs/2601.16597 · **PDF:** https://arxiv.org/pdf/2601.16597v2
+
+## 19778. Attention-MoA: Enhancing Mixture-of-Agents via Inter-Agent Semantic Attention and Deep Residual Synthesis
+
+- **ID:** `2601.16596` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Jianyu Wen, Yang Wei, Xiongxi Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.16596 · **PDF:** https://arxiv.org/pdf/2601.16596v1
+
+## 19779. Integrating Meteorological and Operational Data: A Novel Approach to Understanding Railway Delays in Finland
+
+- **ID:** `2601.16592` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Vinicius Pozzobon Borin, Jean Michel de Souza Sant'Ana, Usama Raheel, et al.
+- **abs:** https://arxiv.org/abs/2601.16592 · **PDF:** https://arxiv.org/pdf/2601.16592v1
+
+## 19780. Emerging Threats and Countermeasures in Neuromorphic Systems: A Survey
+
+- **ID:** `2601.16589` · **primary:** `cs.CR` · **Submitted:** 2026-01-23
+- **Authors:** Pablo Sorrentino, Stjepan Picek, Ihsen Alouani, et al.
+- **abs:** https://arxiv.org/abs/2601.16589 · **PDF:** https://arxiv.org/pdf/2601.16589v1
+
+## 19781. Learning Successive Interference Cancellation for Low-Complexity Soft-Output MIMO Detection
+
+- **ID:** `2601.16586` · **primary:** `eess.SP` · **Submitted:** 2026-01-23
+- **Authors:** Benedikt Fesl, Fatih Capar
+- **abs:** https://arxiv.org/abs/2601.16586 · **PDF:** https://arxiv.org/pdf/2601.16586v1
+
+## 19782. Predicting Startup Success Using Large Language Models: A Novel In-Context Learning Approach
+
+- **ID:** `2601.16568` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Abdurahman Maarouf, Alket Bakiaj, Stefan Feuerriegel
+- **abs:** https://arxiv.org/abs/2601.16568 · **PDF:** https://arxiv.org/pdf/2601.16568v2
+
+## 19783. Process-Tensor Tomography of SGD: Measuring Non-Markovian Memory via Back-Flow of Distinguishability
+
+- **ID:** `2601.16563` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Vasileios Sevetlidis, George Pavlidis
+- **abs:** https://arxiv.org/abs/2601.16563 · **PDF:** https://arxiv.org/pdf/2601.16563v1
+
+## 19784. PRISM: Purified Representation and Integrated Semantic Modeling for Generative Sequential Recommendation
+
+- **ID:** `2601.16556` · **primary:** `cs.IR` · **Submitted:** 2026-01-23
+- **Authors:** Dengzhao Fang, Jingtong Gao, Yu Li, et al.
+- **abs:** https://arxiv.org/abs/2601.16556 · **PDF:** https://arxiv.org/pdf/2601.16556v1
+
+## 19785. Generative Artificial Intelligence in Small and Medium Enterprises: Navigating its Promises and Challenges
+
+- **ID:** `2602.00091` · **primary:** `cs.CY` · **Submitted:** 2026-01-23
+- **Authors:** Kumaran Rajaram, Patrick Nicolas Tinguely
+- **abs:** https://arxiv.org/abs/2602.00091 · **PDF:** https://arxiv.org/pdf/2602.00091v1
+
+## 19786. Retrieve-Refine-Calibrate: A Framework for Complex Claim Fact-Checking
+
+- **ID:** `2601.16555` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Mingwei Sun, Qianlong Wang, Ruifeng Xu
+- **abs:** https://arxiv.org/abs/2601.16555 · **PDF:** https://arxiv.org/pdf/2601.16555v1
+
+## 19787. Lost in Simulation: LLM-Simulated Users are Unreliable Proxies for Human Users in Agentic Evaluations
+
+- **ID:** `2601.17087` · **primary:** `cs.HC` · **Submitted:** 2026-01-23
+- **Authors:** Preethi Seshadri, Samuel Cahyawijaya, Ayomide Odumakinde, et al.
+- **abs:** https://arxiv.org/abs/2601.17087 · **PDF:** https://arxiv.org/pdf/2601.17087v2
+
+## 19788. Understanding and Improving UMAP with Geometric and Topological Priors: The JORC-UMAP Algorithm
+
+- **ID:** `2601.16552` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Xiaobin Li, Run Zhang
+- **abs:** https://arxiv.org/abs/2601.16552 · **PDF:** https://arxiv.org/pdf/2601.16552v1
+
+## 19789. SonoEdit: Null-Space Constrained Knowledge Editing for Pronunciation Correction in LLM-Based TTS
+
+- **ID:** `2601.17086` · **primary:** `cs.SD` · **Submitted:** 2026-01-23
+- **Authors:** Ayush Pratap Singh, Harshit Singh, Nityanand Mathur, et al.
+- **abs:** https://arxiv.org/abs/2601.17086 · **PDF:** https://arxiv.org/pdf/2601.17086v1
+
+## 19790. LLM is Not All You Need: A Systematic Evaluation of ML vs. Foundation Models for text and image based Medical Classification
+
+- **ID:** `2601.16549` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Meet Raval, Tejul Pandit, Dhvani Upadhyay
+- **abs:** https://arxiv.org/abs/2601.16549 · **PDF:** https://arxiv.org/pdf/2601.16549v1
+
+## 19791. CORD: Bridging the Audio-Text Reasoning Gap via Weighted On-policy Cross-modal Distillation
+
+- **ID:** `2601.16547` · **primary:** `cs.SD` · **Submitted:** 2026-01-23
+- **Authors:** Jing Hu, Danxiang Zhu, Xianlong Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.16547 · **PDF:** https://arxiv.org/pdf/2601.16547v1
+
+## 19792. Semi-Supervised Hierarchical Open-Set Classification
+
+- **ID:** `2601.16541` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Erik Wallin, Fredrik Kahl, Lars Hammarstrand
+- **abs:** https://arxiv.org/abs/2601.16541 · **PDF:** https://arxiv.org/pdf/2601.16541v1
+
+## 19793. Do Models Hear Like Us? Probing the Representational Alignment of Audio LLMs and Naturalistic EEG
+
+- **ID:** `2601.16540` · **primary:** `cs.SD` · **Submitted:** 2026-01-23
+- **Authors:** Haoyun Yang, Xin Xiao, Jiang Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.16540 · **PDF:** https://arxiv.org/pdf/2601.16540v2
+
+## 19794. A Collision-Free Hot-Tier Extension for Engram-Style Conditional Memory: A Controlled Study of Training Dynamics
+
+- **ID:** `2601.16531` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Tao Lin
+- **abs:** https://arxiv.org/abs/2601.16531 · **PDF:** https://arxiv.org/pdf/2601.16531v2
+
+## 19795. Curate-Train-Refine: A Closed-Loop Agentic Framework for Zero Shot Classification
+
+- **ID:** `2601.16530` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Gaurav Maheshwari, Kevin El Haddad
+- **abs:** https://arxiv.org/abs/2601.16530 · **PDF:** https://arxiv.org/pdf/2601.16530v1
+
+## 19796. SycoEval-EM: Sycophancy Evaluation of Large Language Models in Simulated Clinical Encounters for Emergency Care
+
+- **ID:** `2601.16529` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Dongshen Peng, Yi Wang, Austin Schoeffler, et al.
+- **abs:** https://arxiv.org/abs/2601.16529 · **PDF:** https://arxiv.org/pdf/2601.16529v2
+
+## 19797. Beyond Superficial Unlearning: Sharpness-Aware Robust Erasure of Hallucinations in Multimodal LLMs
+
+- **ID:** `2601.16527` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Xianya Fang, Feiyang Ren, Xiang Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.16527 · **PDF:** https://arxiv.org/pdf/2601.16527v1
+
+## 19798. Recovering Performance in Speech Emotion Recognition from Discrete Tokens via Multi-Layer Fusion and Paralinguistic Feature Integration
+
+- **ID:** `2601.17085` · **primary:** `eess.AS` · **Submitted:** 2026-01-23
+- **Authors:** Esther Sun, Abinay Reddy Naini, Carlos Busso
+- **abs:** https://arxiv.org/abs/2601.17085 · **PDF:** https://arxiv.org/pdf/2601.17085v1
+
+## 19799. CreditAudit: 2$^\text{nd}$ Dimension for LLM Evaluation and Selection
+
+- **ID:** `2602.02515` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Yiliang Song, Hongjun An, Jiangong Xiao, et al.
+- **abs:** https://arxiv.org/abs/2602.02515 · **PDF:** https://arxiv.org/pdf/2602.02515v2
+
+## 19800. ChemNavigator: Agentic AI Discovery of Design Rules for Organic Photocatalysts
+
+- **ID:** `2601.17084` · **primary:** `physics.chem-ph` · **Submitted:** 2026-01-23
+- **Authors:** Iman Peivaste, Ahmed Makradi, Salim Belouettar
+- **abs:** https://arxiv.org/abs/2601.17084 · **PDF:** https://arxiv.org/pdf/2601.17084v1
+
+## 19801. Design and Evaluation of Whole-Page Experience Optimization for E-commerce Search
+
+- **ID:** `2602.02514` · **primary:** `cs.IR` · **Submitted:** 2026-01-23
+- **Authors:** Pratik Lahiri, Bingqing Ge, Zhou Qin, et al.
+- **abs:** https://arxiv.org/abs/2602.02514 · **PDF:** https://arxiv.org/pdf/2602.02514v1
+
+## 19802. TangramPuzzle: Evaluating Multimodal Large Language Models with Compositional Spatial Reasoning
+
+- **ID:** `2601.16520` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Daixian Liu, Jiayi Kuang, Yinghui Li, et al.
+- **abs:** https://arxiv.org/abs/2601.16520 · **PDF:** https://arxiv.org/pdf/2601.16520v1
+
+## 19803. DANCE: Dynamic, Available, Neighbor-gated Condensation for Federated Text-Attributed Graphs
+
+- **ID:** `2601.16519` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Zekai Chen, Haodong Lu, Xunkai Li, et al.
+- **abs:** https://arxiv.org/abs/2601.16519 · **PDF:** https://arxiv.org/pdf/2601.16519v1
+
+## 19804. Rethinking Large Language Models For Irregular Time Series Classification In Critical Care
+
+- **ID:** `2601.16516` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Feixiang Zheng, Yu Wu, Cecilia Mascolo, et al.
+- **abs:** https://arxiv.org/abs/2601.16516 · **PDF:** https://arxiv.org/pdf/2601.16516v2
+
+## 19805. Finite-Time Analysis of Gradient Descent for Shallow Transformers
+
+- **ID:** `2601.16514` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Enes Arda, Semih Cayci, Atilla Eryilmaz
+- **abs:** https://arxiv.org/abs/2601.16514 · **PDF:** https://arxiv.org/pdf/2601.16514v2
+
+## 19806. SearchLLM: Detecting LLM Paraphrased Text by Measuring the Similarity with Regeneration of the Candidate Source via Search Engine
+
+- **ID:** `2601.16512` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Hoang-Quoc Nguyen-Son, Minh-Son Dao, Koji Zettsu
+- **abs:** https://arxiv.org/abs/2601.16512 · **PDF:** https://arxiv.org/pdf/2601.16512v1
+
+## 19807. Learning to Optimize by Differentiable Programming
+
+- **ID:** `2601.16510` · **primary:** `cs.MS` · **Submitted:** 2026-01-23
+- **Authors:** Liping Tao, Xindi Tong, Chee Wei Tan
+- **abs:** https://arxiv.org/abs/2601.16510 · **PDF:** https://arxiv.org/pdf/2601.16510v2
+
+## 19808. kNN-Graph: An adaptive graph model for $k$-nearest neighbors
+
+- **ID:** `2601.16509` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Jiaye Li, Gang Chen, Hang Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.16509 · **PDF:** https://arxiv.org/pdf/2601.16509v1
+
+## 19809. Is Length Really A Liability? An Evaluation of Multi-turn LLM Conversations using BoolQ
+
+- **ID:** `2601.16508` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Karl Neergaard, Le Qiu, Emmanuele Chersoni
+- **abs:** https://arxiv.org/abs/2601.16508 · **PDF:** https://arxiv.org/pdf/2601.16508v1
+
+## 19810. SafeThinker: Reasoning about Risk to Deepen Safety Beyond Shallow Alignment
+
+- **ID:** `2601.16506` · **primary:** `cs.CR` · **Submitted:** 2026-01-23
+- **Authors:** Xianya Fang, Xianying Luo, Yadong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.16506 · **PDF:** https://arxiv.org/pdf/2601.16506v1
+
+## 19811. LOGICAL-COMMONSENSEQA: A Benchmark for Logical Commonsense Reasoning
+
+- **ID:** `2601.16504` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Obed Junias, Maria Leonor Pacheco
+- **abs:** https://arxiv.org/abs/2601.16504 · **PDF:** https://arxiv.org/pdf/2601.16504v2
+
+## 19812. MRAG: Benchmarking Retrieval-Augmented Generation for Bio-medicine
+
+- **ID:** `2601.16503` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Liz Li, Wei Zhu
+- **abs:** https://arxiv.org/abs/2601.16503 · **PDF:** https://arxiv.org/pdf/2601.16503v2
+
+## 19813. BoostFGL: Boosting Fairness in Federated Graph Learning
+
+- **ID:** `2601.16496` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Zekai Chen, Kairui Yang, Xunkai Li, et al.
+- **abs:** https://arxiv.org/abs/2601.16496 · **PDF:** https://arxiv.org/pdf/2601.16496v1
+
+## 19814. Learning ORDER-Aware Multimodal Representations for Composite Materials Design
+
+- **ID:** `2602.02513` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Xinyao Li, Hangwei Qian, Jingjing Li, et al.
+- **abs:** https://arxiv.org/abs/2602.02513 · **PDF:** https://arxiv.org/pdf/2602.02513v1
+
+## 19815. Robust Categorical Data Clustering Guided by Multi-Granular Competitive Learning
+
+- **ID:** `2601.16491` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Shenghong Cai, Yiqun Zhang, Xiaopeng Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.16491 · **PDF:** https://arxiv.org/pdf/2601.16491v1
+
+## 19816. EvoConfig: Self-Evolving Multi-Agent Systems for Efficient Autonomous Environment Configuration
+
+- **ID:** `2601.16489` · **primary:** `cs.SE` · **Submitted:** 2026-01-23
+- **Authors:** Xinshuai Guo, Jiayi Kuang, Linyue Pan, et al.
+- **abs:** https://arxiv.org/abs/2601.16489 · **PDF:** https://arxiv.org/pdf/2601.16489v1
+
+## 19817. Timely Machine: Awareness of Time Makes Test-Time Scaling Agentic
+
+- **ID:** `2601.16486` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Yichuan Ma, Linyang Li, Yongkang chen, et al.
+- **abs:** https://arxiv.org/abs/2601.16486 · **PDF:** https://arxiv.org/pdf/2601.16486v1
+
+## 19818. TL-GRPO: Turn-Level RL for Reasoning-Guided Iterative Optimization
+
+- **ID:** `2601.16480` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Peiji Li, Linyang Li, Handa Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.16480 · **PDF:** https://arxiv.org/pdf/2601.16480v1
+
+## 19819. Doc2AHP: Inferring Structured Multi-Criteria Decision Models via Semantic Trees with LLMs
+
+- **ID:** `2601.16479` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Hongjia Wu, Shuai Zhou, Hongxin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.16479 · **PDF:** https://arxiv.org/pdf/2601.16479v1
+
+## 19820. DeepEra: A Deep Evidence Reranking Agent for Scientific Retrieval-Augmented Generated Question Answering
+
+- **ID:** `2601.16478` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Haotian Chen, Qingqing Long, Siyu Pu, et al.
+- **abs:** https://arxiv.org/abs/2601.16478 · **PDF:** https://arxiv.org/pdf/2601.16478v1
+
+## 19821. DeMark: A Query-Free Black-Box Attack on Deepfake Watermarking Defenses
+
+- **ID:** `2601.16473` · **primary:** `cs.CR` · **Submitted:** 2026-01-23
+- **Authors:** Wei Song, Zhenchang Xing, Liming Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.16473 · **PDF:** https://arxiv.org/pdf/2601.16473v1
+
+## 19822. Do VLMs Have a Moral Backbone? A Study on the Fragile Morality of Vision-Language Models
+
+- **ID:** `2601.17082` · **primary:** `cs.CY` · **Submitted:** 2026-01-23
+- **Authors:** Zhining Liu, Tianyi Wang, Xiao Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.17082 · **PDF:** https://arxiv.org/pdf/2601.17082v1
+
+## 19823. VeRA: Verified Reasoning Data Augmentation at Scale
+
+- **ID:** `2602.13217` · **primary:** `cs.AI` · **Submitted:** 2026-01-23
+- **Authors:** Zerui Cheng, Jiashuo Liu, Chunjie Wu, et al.
+- **abs:** https://arxiv.org/abs/2602.13217 · **PDF:** https://arxiv.org/pdf/2602.13217v1
+
+## 19824. A Cautionary Tale of Self-Supervised Learning for Imaging Biomarkers: Alzheimer's Disease Case Study
+
+- **ID:** `2601.16467` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Maxwell Reynolds, Chaitanya Srinivasan, Vijay Cherupally, et al.
+- **abs:** https://arxiv.org/abs/2601.16467 · **PDF:** https://arxiv.org/pdf/2601.16467v1
+
+## 19825. Persona Jailbreaking in Large Language Models
+
+- **ID:** `2601.16466` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Jivnesh Sandhan, Fei Cheng, Tushar Sandhan, et al.
+- **abs:** https://arxiv.org/abs/2601.16466 · **PDF:** https://arxiv.org/pdf/2601.16466v1
+
+## 19826. On the Effects of Adversarial Perturbations on Distribution Robustness
+
+- **ID:** `2601.16464` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Yipei Wang, Zhaoying Pan, Xiaoqian Wang
+- **abs:** https://arxiv.org/abs/2601.16464 · **PDF:** https://arxiv.org/pdf/2601.16464v1
+
+## 19827. Graph-Anchored Knowledge Indexing for Retrieval-Augmented Generation
+
+- **ID:** `2601.16462` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Zhenghao Liu, Mingyan Wu, Xinze Li, et al.
+- **abs:** https://arxiv.org/abs/2601.16462 · **PDF:** https://arxiv.org/pdf/2601.16462v1
+
+## 19828. PC-MCL: Patient-Consistent Multi-Cycle Learning with multi-label bias correction for respiratory sound classification
+
+- **ID:** `2601.17080` · **primary:** `eess.AS` · **Submitted:** 2026-01-23
+- **Authors:** Seung Gyu Jeong, Seong-Eun Kim
+- **abs:** https://arxiv.org/abs/2601.17080 · **PDF:** https://arxiv.org/pdf/2601.17080v1
+
+## 19829. Efficient Edge Rewiring Strategies for Enhancing PageRank Fairness
+
+- **ID:** `2602.02512` · **primary:** `cs.SI` · **Submitted:** 2026-01-23
+- **Authors:** Changan Liu, Haoxin Sun, Ahad N. Zehmakan, et al.
+- **abs:** https://arxiv.org/abs/2602.02512 · **PDF:** https://arxiv.org/pdf/2602.02512v1
+
+## 19830. On the Expressive Power of Floating-Point Transformers
+
+- **ID:** `2601.16450` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Sejun Park, Yeachan Park, Geonho Hwang
+- **abs:** https://arxiv.org/abs/2601.16450 · **PDF:** https://arxiv.org/pdf/2601.16450v1
+
+## 19831. Emotion-LLaMAv2 and MMEVerse: A New Framework and Benchmark for Multimodal Emotion Understanding
+
+- **ID:** `2601.16449` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Xiaojiang Peng, Jingyi Chen, Zebang Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.16449 · **PDF:** https://arxiv.org/pdf/2601.16449v2
+
+## 19832. Mixing Expert Knowledge: Bring Human Thoughts Back To the Game of Go
+
+- **ID:** `2601.16447` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Yichuan Ma, Linyang Li, Yongkang Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.16447 · **PDF:** https://arxiv.org/pdf/2601.16447v1
+
+## 19833. Brownian ReLU(Br-ReLU): A New Activation Function for a Long-Short Term Memory (LSTM) Network
+
+- **ID:** `2601.16446` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** George Awiakye-Marfo, Elijah Agbosu, Victoria Mawuena Barns, et al.
+- **abs:** https://arxiv.org/abs/2601.16446 · **PDF:** https://arxiv.org/pdf/2601.16446v1
+
+## 19834. Exploring the Effects of Alignment on Numerical Bias in Large Language Models
+
+- **ID:** `2601.16444` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Ayako Sato, Hwichan Kim, Zhousi Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.16444 · **PDF:** https://arxiv.org/pdf/2601.16444v2
+
+## 19835. Endless Terminals: Scaling RL Environments for Terminal Agents
+
+- **ID:** `2601.16443` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Kanishk Gandhi, Shivam Garg, Noah D. Goodman, et al.
+- **abs:** https://arxiv.org/abs/2601.16443 · **PDF:** https://arxiv.org/pdf/2601.16443v3
+
+## 19836. AlphaFace: High Fidelity and Real-time Face Swapper Robust to Facial Pose
+
+- **ID:** `2601.16429` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Jongmin Yu, Hyeontaek Oh, Zhongtian Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.16429 · **PDF:** https://arxiv.org/pdf/2601.16429v1
+
+## 19837. Perfect Clustering for Sparse Directed Stochastic Block Models
+
+- **ID:** `2601.16427` · **primary:** `stat.ML` · **Submitted:** 2026-01-23
+- **Authors:** Behzad Aalipur, Yichen Qin
+- **abs:** https://arxiv.org/abs/2601.16427 · **PDF:** https://arxiv.org/pdf/2601.16427v2
+
+## 19838. Safe Multitask Molecular Graph Networks for Vapor Pressure and Odor Threshold Prediction
+
+- **ID:** `2601.16426` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Shuang Wu, Meijie Wang, Lun Yu
+- **abs:** https://arxiv.org/abs/2601.16426 · **PDF:** https://arxiv.org/pdf/2601.16426v1
+
+## 19839. Bayesian Experimental Design for Model Discrepancy Calibration: A Rivalry between Kullback--Leibler Divergence and Wasserstein Distance
+
+- **ID:** `2601.16425` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Huchen Yang, Xinghao Dong, Jin-Long Wu
+- **abs:** https://arxiv.org/abs/2601.16425 · **PDF:** https://arxiv.org/pdf/2601.16425v1
+
+## 19840. RENEW: Risk- and Energy-Aware Navigation in Dynamic Waterways
+
+- **ID:** `2601.16424` · **primary:** `cs.RO` · **Submitted:** 2026-01-23
+- **Authors:** Mingi Jeong, Alberto Quattrini Li
+- **abs:** https://arxiv.org/abs/2601.16424 · **PDF:** https://arxiv.org/pdf/2601.16424v1
+
+## 19841. E2PL: Effective and Efficient Prompt Learning for Incomplete Multi-view Multi-Label Class Incremental Learning
+
+- **ID:** `2601.17076` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Jiajun Chen, Yue Wu, Kai Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.17076 · **PDF:** https://arxiv.org/pdf/2601.17076v1
+
+## 19842. Learning Domain Knowledge in Multimodal Large Language Models through Reinforcement Fine-Tuning
+
+- **ID:** `2601.16419` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Qinglong Cao, Yuntian Chen, Chao Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.16419 · **PDF:** https://arxiv.org/pdf/2601.16419v2
+
+## 19843. PyHealth 2.0: A Comprehensive Open-Source Toolkit for Accessible and Reproducible Clinical Deep Learning
+
+- **ID:** `2601.16414` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** John Wu, Yongda Fan, Zhenbang Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.16414 · **PDF:** https://arxiv.org/pdf/2601.16414v1
+
+## 19844. Tight Regret Bounds for Bilateral Trade under Semi Feedback
+
+- **ID:** `2601.16412` · **primary:** `cs.GT` · **Submitted:** 2026-01-23
+- **Authors:** Yaonan Jin
+- **abs:** https://arxiv.org/abs/2601.16412 · **PDF:** https://arxiv.org/pdf/2601.16412v1
+
+## 19845. A Refinement of Vapnik--Chervonenkis' Theorem
+
+- **ID:** `2601.16411` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** A. Iosevich, A. Vagharshakyan, E. Wyman
+- **abs:** https://arxiv.org/abs/2601.16411 · **PDF:** https://arxiv.org/pdf/2601.16411v1
+
+## 19846. From Numbers to Prompts: A Cognitive Symbolic Transition Mechanism for Lightweight Time-Series Forecasting
+
+- **ID:** `2602.00088` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Namkyung Yoon, Hwangnam Kim
+- **abs:** https://arxiv.org/abs/2602.00088 · **PDF:** https://arxiv.org/pdf/2602.00088v1
+
+## 19847. Jacobian Scopes: token-level causal attributions in LLMs
+
+- **ID:** `2601.16407` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Toni J. B. Liu, Baran Zadeoğlu, Nicolas Boullé, et al.
+- **abs:** https://arxiv.org/abs/2601.16407 · **PDF:** https://arxiv.org/pdf/2601.16407v2
+
+## 19848. Reasoning-Enhanced Rare-Event Prediction with Balanced Outcome Correction
+
+- **ID:** `2601.16406` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Vitaly Bulgakov, Alexander Turchin
+- **abs:** https://arxiv.org/abs/2601.16406 · **PDF:** https://arxiv.org/pdf/2601.16406v1
+
+## 19849. Reinforcement Learning-Based Energy-Aware Coverage Path Planning for Precision Agriculture
+
+- **ID:** `2601.16405` · **primary:** `cs.RO` · **Submitted:** 2026-01-23
+- **Authors:** Beining Wu, Zihao Ding, Leo Ostigaard, et al.
+- **abs:** https://arxiv.org/abs/2601.16405 · **PDF:** https://arxiv.org/pdf/2601.16405v1
+
+## 19850. Towards a Theoretical Understanding to the Generalization of RLHF
+
+- **ID:** `2601.16403` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Zhaochun Li, Mingyang Yi, Yue Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.16403 · **PDF:** https://arxiv.org/pdf/2601.16403v1
+
+## 19851. Training Data Governance for Brain Foundation Models
+
+- **ID:** `2602.02511` · **primary:** `cs.CY` · **Submitted:** 2026-01-23
+- **Authors:** Margot Hanley, Jiunn-Tyng Yeh, Ryan Rodriguez, et al.
+- **abs:** https://arxiv.org/abs/2602.02511 · **PDF:** https://arxiv.org/pdf/2602.02511v1
+
+## 19852. Clarify or Answer: Reinforcement Learning for Agentic VQA with Context Under-specification
+
+- **ID:** `2601.16400` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Zongwan Cao, Bingbing Wen, Lucy Lu Wang
+- **abs:** https://arxiv.org/abs/2601.16400 · **PDF:** https://arxiv.org/pdf/2601.16400v1
+
+## 19853. A Hessian-Free Actor-Critic Algorithm for Bi-Level Reinforcement Learning with Applications to LLM Fine-Tuning
+
+- **ID:** `2601.16399` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Sihan Zeng, Sujay Bhatt, Sumitra Ganesh, et al.
+- **abs:** https://arxiv.org/abs/2601.16399 · **PDF:** https://arxiv.org/pdf/2601.16399v3
+
+## 19854. White-Box Sensitivity Auditing with Steering Vectors
+
+- **ID:** `2601.16398` · **primary:** `cs.CY` · **Submitted:** 2026-01-23
+- **Authors:** Hannah Cyberey, Yangfeng Ji, David Evans
+- **abs:** https://arxiv.org/abs/2601.16398 · **PDF:** https://arxiv.org/pdf/2601.16398v1
+
+## 19855. Cite-While-You-Generate: Training-Free Evidence Attribution for Multimodal Clinical Summarization
+
+- **ID:** `2601.16397` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Qianqi Yan, Huy Nguyen, Sumana Srivatsa, et al.
+- **abs:** https://arxiv.org/abs/2601.16397 · **PDF:** https://arxiv.org/pdf/2601.16397v1
+
+## 19856. ResAgent: Entropy-based Prior Point Discovery and Visual Reasoning for Referring Expression Segmentation
+
+- **ID:** `2601.16394` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Yihao Wang, Jusheng Zhang, Ziyi Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.16394 · **PDF:** https://arxiv.org/pdf/2601.16394v1
+
+## 19857. Cross-Lingual Activation Steering for Multilingual Language Models
+
+- **ID:** `2601.16390` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Rhitabrat Pokharel, Ameeta Agrawal, Tanay Nagar
+- **abs:** https://arxiv.org/abs/2601.16390 · **PDF:** https://arxiv.org/pdf/2601.16390v1
+
+## 19858. ECCO: Evidence-Driven Causal Reasoning for Compiler Optimization
+
+- **ID:** `2602.00087` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Haolin Pan, Lianghong Huang, Jinyuan Dong, et al.
+- **abs:** https://arxiv.org/abs/2602.00087 · **PDF:** https://arxiv.org/pdf/2602.00087v1
+
+## 19859. Beyond Translation: Cross-Cultural Meme Transcreation with Vision-Language Models
+
+- **ID:** `2602.02510` · **primary:** `cs.CY` · **Submitted:** 2026-01-23
+- **Authors:** Yuming Zhao, Peiyi Zhang, Oana Ignat
+- **abs:** https://arxiv.org/abs/2602.02510 · **PDF:** https://arxiv.org/pdf/2602.02510v1
+
+## 19860. PhysE-Inv: A Physics-Encoded Inverse Modeling approach for Arctic Snow Depth Prediction
+
+- **ID:** `2601.17074` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Akila Sampath, Vandana Janeja, Jianwu Wang
+- **abs:** https://arxiv.org/abs/2601.17074 · **PDF:** https://arxiv.org/pdf/2601.17074v3
+
+## 19861. Attention-Based Variational Framework for Joint and Individual Components Learning with Applications in Brain Network Analysis
+
+- **ID:** `2601.17073` · **primary:** `cs.LG` · **Submitted:** 2026-01-23
+- **Authors:** Yifei Zhang, Meimei Liu, Zhengwu Zhang
+- **abs:** https://arxiv.org/abs/2601.17073 · **PDF:** https://arxiv.org/pdf/2601.17073v1
+
+## 19862. Cognitively-Inspired Tokens Overcome Egocentric Bias in Multimodal Models
+
+- **ID:** `2601.16378` · **primary:** `cs.CV` · **Submitted:** 2026-01-23
+- **Authors:** Bridget Leonard, Scott O. Murray
+- **abs:** https://arxiv.org/abs/2601.16378 · **PDF:** https://arxiv.org/pdf/2601.16378v1
+
+## 19863. Polymer-Agent: Large Language Model Agent for Polymer Design
+
+- **ID:** `2601.16376` · **primary:** `cs.CL` · **Submitted:** 2026-01-23
+- **Authors:** Vani Nigam, Achuth Chandrasekhar, Amir Barati Farimani
+- **abs:** https://arxiv.org/abs/2601.16376 · **PDF:** https://arxiv.org/pdf/2601.16376v2
+
+## 19864. Improving the Accuracy of Community Detection on Signed Networks via Community Refinement and Contrastive Learning
+
+- **ID:** `2601.16372` · **primary:** `cs.SI` · **Submitted:** 2026-01-23
+- **Authors:** Hyunuk Shin, Hojin Kim, Chanyoung Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.16372 · **PDF:** https://arxiv.org/pdf/2601.16372v1
+
+## 19865. Trademark Search, Artificial Intelligence and the Role of the Private Sector
+
+- **ID:** `2601.17072` · **primary:** `cs.CY` · **Submitted:** 2026-01-22
+- **Authors:** Sonia Katyal, Aniket Kesari
+- **abs:** https://arxiv.org/abs/2601.17072 · **PDF:** https://arxiv.org/pdf/2601.17072v1
+
+## 19866. Analyzing Neural Network Information Flow Using Differential Geometry
+
+- **ID:** `2601.16366` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Shuhang Tan, Jayson Sia, Paul Bogdan, et al.
+- **abs:** https://arxiv.org/abs/2601.16366 · **PDF:** https://arxiv.org/pdf/2601.16366v1
+
+## 19867. Impact of LLMs news Sentiment Analysis on Stock Price Movement Prediction
+
+- **ID:** `2602.00086` · **primary:** `q-fin.ST` · **Submitted:** 2026-01-22
+- **Authors:** Walid Siala, Ahmed Khanfir, Mike Papadakis
+- **abs:** https://arxiv.org/abs/2602.00086 · **PDF:** https://arxiv.org/pdf/2602.00086v3
+
+## 19868. CARE-RFT: Confidence-Anchored Reinforcement Finetuning for Reliable Reasoning in Large Language Models
+
+- **ID:** `2602.00085` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Shuozhe Li, Jincheng Cao, Bodun Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.00085 · **PDF:** https://arxiv.org/pdf/2602.00085v1
+
+## 19869. Experience with Single Domain Generalization in Real World Medical Imaging Deployments
+
+- **ID:** `2601.16359` · **primary:** `eess.IV` · **Submitted:** 2026-01-22
+- **Authors:** Ayan Banerjee, Komandoor Srivathsan, Sandeep K. S. Gupta
+- **abs:** https://arxiv.org/abs/2601.16359 · **PDF:** https://arxiv.org/pdf/2601.16359v1
+
+## 19870. Why LoRA Resists Label Noise: A Theoretical Framework for Noise-Robust Parameter-Efficient Fine-Tuning
+
+- **ID:** `2602.00084` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Brady Steele
+- **abs:** https://arxiv.org/abs/2602.00084 · **PDF:** https://arxiv.org/pdf/2602.00084v1
+
+## 19871. Identity, Cooperation and Framing Effects within Groups of Real and Simulated Humans
+
+- **ID:** `2601.16355` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Suhong Moon, Minwoo Kang, Joseph Suh, et al.
+- **abs:** https://arxiv.org/abs/2601.16355 · **PDF:** https://arxiv.org/pdf/2601.16355v2
+
+## 19872. NOIR: Privacy-Preserving Generation of Code with Open-Source LLMs
+
+- **ID:** `2601.16354` · **primary:** `cs.CR` · **Submitted:** 2026-01-22
+- **Authors:** Khoa Nguyen, Khiem Ton, NhatHai Phan, et al.
+- **abs:** https://arxiv.org/abs/2601.16354 · **PDF:** https://arxiv.org/pdf/2601.16354v1
+
+## 19873. Regional Bias in Large Language Models
+
+- **ID:** `2601.16349` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** M P V S Gopinadh, Kappara Lakshmi Sindhu, Soma Sekhar Pandu Ranga Raju P, et al.
+- **abs:** https://arxiv.org/abs/2601.16349 · **PDF:** https://arxiv.org/pdf/2601.16349v1
+
+## 19874. Long-Term Probabilistic Forecast of Vegetation Conditions Using Climate Attributes in the Four Corners Region
+
+- **ID:** `2601.16347` · **primary:** `stat.AP` · **Submitted:** 2026-01-22
+- **Authors:** Erika McPhillips, Hyeongseong Lee, Xiangyu Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.16347 · **PDF:** https://arxiv.org/pdf/2601.16347v1
+
+## 19875. DSGym: A Holistic Framework for Evaluating and Training Data Science Agents
+
+- **ID:** `2601.16344` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Fan Nie, Junlin Wang, Harper Hua, et al.
+- **abs:** https://arxiv.org/abs/2601.16344 · **PDF:** https://arxiv.org/pdf/2601.16344v1
+
+## 19876. CodeGuard: Improving LLM Guardrails in CS Education
+
+- **ID:** `2602.02509` · **primary:** `cs.CY` · **Submitted:** 2026-01-22
+- **Authors:** Nishat Raihan, Noah Erdachew, Jayoti Devi, et al.
+- **abs:** https://arxiv.org/abs/2602.02509 · **PDF:** https://arxiv.org/pdf/2602.02509v1
+
+## 19877. DMAVA: Distributed Multi-Autonomous Vehicle Architecture Using Autoware
+
+- **ID:** `2601.16336` · **primary:** `cs.RO` · **Submitted:** 2026-01-22
+- **Authors:** Zubair Islam, Mohamed El-Darieby
+- **abs:** https://arxiv.org/abs/2601.16336 · **PDF:** https://arxiv.org/pdf/2601.16336v2
+
+## 19878. Where is the multimodal goal post? On the Ability of Foundation Models to Recognize Contextually Important Moments
+
+- **ID:** `2601.16333` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Aditya K Surikuchi, Raquel Fernández, Sandro Pezzelle
+- **abs:** https://arxiv.org/abs/2601.16333 · **PDF:** https://arxiv.org/pdf/2601.16333v2
+
+## 19879. Efficient Gaussian process learning via subspace projections
+
+- **ID:** `2601.16332` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Elsa Cazelles, Felipe Tobar
+- **abs:** https://arxiv.org/abs/2601.16332 · **PDF:** https://arxiv.org/pdf/2601.16332v2
+
+## 19880. DMV-AVP: Distributed Multi-Vehicle Autonomous Valet Parking Using Autoware
+
+- **ID:** `2601.16327` · **primary:** `cs.RO` · **Submitted:** 2026-01-22
+- **Authors:** Zubair Islam, Mohamed El-Darieby
+- **abs:** https://arxiv.org/abs/2601.16327 · **PDF:** https://arxiv.org/pdf/2601.16327v2
+
+## 19881. Student Mental Health Screening via Fitbit Data Collected During the COVID-19 Pandemic
+
+- **ID:** `2601.16324` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Rebecca Lopez, Avantika Shrestha, ML Tlachac, et al.
+- **abs:** https://arxiv.org/abs/2601.16324 · **PDF:** https://arxiv.org/pdf/2601.16324v1
+
+## 19882. Multi-Agent Deep Reinforcement Learning Under Constrained Communications
+
+- **ID:** `2601.17069` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Shahil Shaik, Jonathon M. Smereka, Yue Wang
+- **abs:** https://arxiv.org/abs/2601.17069 · **PDF:** https://arxiv.org/pdf/2601.17069v1
+
+## 19883. EdgeSpot: Efficient and High-Performance Few-Shot Model for Keyword Spotting
+
+- **ID:** `2601.16316` · **primary:** `eess.AS` · **Submitted:** 2026-01-22
+- **Authors:** Oguzhan Buyuksolak, Alican Gok, Osman Erman Okman
+- **abs:** https://arxiv.org/abs/2601.16316 · **PDF:** https://arxiv.org/pdf/2601.16316v1
+
+## 19884. Machine-Assisted Grading of Nationwide School-Leaving Essay Exams with LLMs and Statistical NLP
+
+- **ID:** `2601.16314` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Andres Karjus, Kais Allkivi, Silvia Maine, et al.
+- **abs:** https://arxiv.org/abs/2601.16314 · **PDF:** https://arxiv.org/pdf/2601.16314v1
+
+## 19885. Teaching and Evaluating LLMs to Reason About Polymer Design Related Tasks
+
+- **ID:** `2601.16312` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Dikshya Mohanty, Mohammad Saqib Hasan, Syed Mostofa Monsur, et al.
+- **abs:** https://arxiv.org/abs/2601.16312 · **PDF:** https://arxiv.org/pdf/2601.16312v1
+
+## 19886. A Longitudinal, Multinational, and Multilingual Corpus of News Coverage of the Russo-Ukrainian War
+
+- **ID:** `2601.16309` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Dikshya Mohanty, Taisiia Sabadyn, Jelwin Rodrigues, et al.
+- **abs:** https://arxiv.org/abs/2601.16309 · **PDF:** https://arxiv.org/pdf/2601.16309v2
+
+## 19887. SPARC-RAG: Adaptive Sequential-Parallel Scaling with Context Management for Retrieval-Augmented Generation
+
+- **ID:** `2602.00083` · **primary:** `cs.IR` · **Submitted:** 2026-01-22
+- **Authors:** Yuxin Yang, Gangda Deng, Ömer Faruk Akgül, et al.
+- **abs:** https://arxiv.org/abs/2602.00083 · **PDF:** https://arxiv.org/pdf/2602.00083v1
+
+## 19888. Memory-V2V: Memory-Augmented Video-to-Video Diffusion for Consistent Multi-Turn Editing
+
+- **ID:** `2601.16296` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Dohun Lee, Chun-Hao Paul Huang, Xuelin Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.16296 · **PDF:** https://arxiv.org/pdf/2601.16296v2
+
+## 19889. Space Filling Curves is All You Need: Communication-Avoiding Matrix Multiplication Made Simple
+
+- **ID:** `2601.16294` · **primary:** `cs.DC` · **Submitted:** 2026-01-22
+- **Authors:** Evangelos Georganas, Alexander Heinecke, Pradeep Dubey
+- **abs:** https://arxiv.org/abs/2601.16294 · **PDF:** https://arxiv.org/pdf/2601.16294v1
+
+## 19890. Active learning for photonic crystals
+
+- **ID:** `2601.16287` · **primary:** `physics.optics` · **Submitted:** 2026-01-22
+- **Authors:** Ryan Lopez, Charlotte Loh, Rumen Dangovski, et al.
+- **abs:** https://arxiv.org/abs/2601.16287 · **PDF:** https://arxiv.org/pdf/2601.16287v2
+
+## 19891. SemanticALLI: Caching Reasoning, Not Just Responses, in Agentic Systems
+
+- **ID:** `2601.16286` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Varun Chillara, Dylan Kline, Christopher Alvares, et al.
+- **abs:** https://arxiv.org/abs/2601.16286 · **PDF:** https://arxiv.org/pdf/2601.16286v2
+
+## 19892. Generating Literature-Driven Scientific Theories at Scale
+
+- **ID:** `2601.16282` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Peter Jansen, Peter Clark, Doug Downey, et al.
+- **abs:** https://arxiv.org/abs/2601.16282 · **PDF:** https://arxiv.org/pdf/2601.16282v1
+
+## 19893. When Agents Fail to Act: A Diagnostic Framework for Tool Invocation Reliability in Multi-Agent LLM Systems
+
+- **ID:** `2601.16280` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Donghao Huang, Gauri Malwe, Zhaoxia Wang
+- **abs:** https://arxiv.org/abs/2601.16280 · **PDF:** https://arxiv.org/pdf/2601.16280v1
+
+## 19894. Better as Generators Than Classifiers: Leveraging LLMs and Synthetic Data for Low-Resource Multilingual Classification
+
+- **ID:** `2601.16278` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Branislav Pecher, Jan Cegin, Robert Belanec, et al.
+- **abs:** https://arxiv.org/abs/2601.16278 · **PDF:** https://arxiv.org/pdf/2601.16278v1
+
+## 19895. GameTalk: Training LLMs for Strategic Conversation
+
+- **ID:** `2601.16276` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Victor Conchello Vendrell, Max Ruiz Luyten, Mihaela van der Schaar
+- **abs:** https://arxiv.org/abs/2601.16276 · **PDF:** https://arxiv.org/pdf/2601.16276v1
+
+## 19896. A Mechanistic View on Video Generation as World Models: State and Dynamics
+
+- **ID:** `2601.17067` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Luozhou Wang, Zhifei Chen, Yihua Du, et al.
+- **abs:** https://arxiv.org/abs/2601.17067 · **PDF:** https://arxiv.org/pdf/2601.17067v1
+
+## 19897. Why Can't I Open My Drawer? Mitigating Object-Driven Shortcuts in Zero-Shot Compositional Action Recognition
+
+- **ID:** `2601.16211` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Geo Ahn, Inwoong Lee, Taeoh Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.16211 · **PDF:** https://arxiv.org/pdf/2601.16211v1
+
+## 19898. PyraTok: Language-Aligned Pyramidal Tokenizer for Video Understanding and Generation
+
+- **ID:** `2601.16210` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Onkar Susladkar, Tushar Prakash, Adheesh Juvekar, et al.
+- **abs:** https://arxiv.org/abs/2601.16210 · **PDF:** https://arxiv.org/pdf/2601.16210v2
+
+## 19899. LLM-in-Sandbox Elicits General Agentic Intelligence
+
+- **ID:** `2601.16206` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Daixuan Cheng, Shaohan Huang, Yuxian Gu, et al.
+- **abs:** https://arxiv.org/abs/2601.16206 · **PDF:** https://arxiv.org/pdf/2601.16206v2
+
+## 19900. Distributional Computational Graphs: Error Bounds
+
+- **ID:** `2601.16250` · **primary:** `stat.ML` · **Submitted:** 2026-01-22
+- **Authors:** Olof Hallqvist Elias, Michael Selby, Phillip Stanley-Marbell
+- **abs:** https://arxiv.org/abs/2601.16250 · **PDF:** https://arxiv.org/pdf/2601.16250v2
+
+## 19901. Counterfactual Training: Teaching Models Plausible and Actionable Explanations
+
+- **ID:** `2601.16205` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Patrick Altmeyer, Aleksander Buszydlik, Arie van Deursen, et al.
+- **abs:** https://arxiv.org/abs/2601.16205 · **PDF:** https://arxiv.org/pdf/2601.16205v1
+
+## 19902. Feature-Space Adversarial Robustness Certification for Multimodal Large Language Models
+
+- **ID:** `2601.16200` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Song Xia, Meiwen Ding, Chenqi Kong, et al.
+- **abs:** https://arxiv.org/abs/2601.16200 · **PDF:** https://arxiv.org/pdf/2601.16200v2
+
+## 19903. Precoding-Oriented CSI Feedback Design with Mutual Information Regularized VQ-VAE
+
+- **ID:** `2602.02508` · **primary:** `cs.IT` · **Submitted:** 2026-01-22
+- **Authors:** Xi Chen, Homa Esfahanizadeh, Foad Sohrabi
+- **abs:** https://arxiv.org/abs/2602.02508 · **PDF:** https://arxiv.org/pdf/2602.02508v1
+
+## 19904. Pushing the limits of unconstrained machine-learned interatomic potentials
+
+- **ID:** `2601.16195` · **primary:** `physics.chem-ph` · **Submitted:** 2026-01-22
+- **Authors:** Filippo Bigi, Paolo Pegolo, Arslan Mazitov, et al.
+- **abs:** https://arxiv.org/abs/2601.16195 · **PDF:** https://arxiv.org/pdf/2601.16195v2
+
+## 19905. A Rolling-Space Branch-and-Price Algorithm for the Multi-Compartment Vehicle Routing Problem with Multiple Time Windows
+
+- **ID:** `2601.16194` · **primary:** `math.OC` · **Submitted:** 2026-01-22
+- **Authors:** El Mehdi Er Raqabi, Kevin Dalmeijer, Pascal Van Hentenryck
+- **abs:** https://arxiv.org/abs/2601.16194 · **PDF:** https://arxiv.org/pdf/2601.16194v1
+
+## 19906. Learning to Discover at Test Time
+
+- **ID:** `2601.16175` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Mert Yuksekgonul, Daniel Koceja, Xinhao Li, et al.
+- **abs:** https://arxiv.org/abs/2601.16175 · **PDF:** https://arxiv.org/pdf/2601.16175v2
+
+## 19907. Beyond Predictive Uncertainty: Reliable Representation Learning with Structural Constraints
+
+- **ID:** `2601.16174` · **primary:** `stat.ML` · **Submitted:** 2026-01-22
+- **Authors:** Yiyao Yang
+- **abs:** https://arxiv.org/abs/2601.16174 · **PDF:** https://arxiv.org/pdf/2601.16174v3
+
+## 19908. Structured Hints for Sample-Efficient Lean Theorem Proving
+
+- **ID:** `2601.16172` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Zachary Burton
+- **abs:** https://arxiv.org/abs/2601.16172 · **PDF:** https://arxiv.org/pdf/2601.16172v1
+
+## 19909. Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning
+
+- **ID:** `2601.16163` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Moo Jin Kim, Yihuai Gao, Tsung-Yi Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.16163 · **PDF:** https://arxiv.org/pdf/2601.16163v1
+
+## 19910. Ordering-based Causal Discovery via Generalized Score Matching
+
+- **ID:** `2601.16249` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Vy Vo, He Zhao, Trung Le, et al.
+- **abs:** https://arxiv.org/abs/2601.16249 · **PDF:** https://arxiv.org/pdf/2601.16249v2
+
+## 19911. Domain-Incremental Continual Learning for Robust and Efficient Keyword Spotting in Resource Constrained Systems
+
+- **ID:** `2601.16158` · **primary:** `cs.SD` · **Submitted:** 2026-01-22
+- **Authors:** Prakash Dhungana, Sayed Ahmad Salehi
+- **abs:** https://arxiv.org/abs/2601.16158 · **PDF:** https://arxiv.org/pdf/2601.16158v1
+
+## 19912. ThinkTank-ME: A Multi-Expert Framework for Middle East Event Forecasting
+
+- **ID:** `2601.17065` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Haoxuan Li, He Chang, Yunshan Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.17065 · **PDF:** https://arxiv.org/pdf/2601.17065v1
+
+## 19913. Substrate Stability Under Persistent Disagreement: Structural Constraints for Neutral Ontological Substrates
+
+- **ID:** `2601.16152` · **primary:** `cs.LO` · **Submitted:** 2026-01-22
+- **Authors:** Denise M. Case
+- **abs:** https://arxiv.org/abs/2601.16152 · **PDF:** https://arxiv.org/pdf/2601.16152v1
+
+## 19914. Pay (Cross) Attention to the Melody: Curriculum Masking for Single-Encoder Melodic Harmonization
+
+- **ID:** `2601.16150` · **primary:** `cs.SD` · **Submitted:** 2026-01-22
+- **Authors:** Maximos Kaliakatsos-Papakostas, Dimos Makris, Konstantinos Soiledis, et al.
+- **abs:** https://arxiv.org/abs/2601.16150 · **PDF:** https://arxiv.org/pdf/2601.16150v1
+
+## 19915. Beat-ssl: Capturing Local ECG Morphology through Heartbeat-level Contrastive Learning with Soft Targets
+
+- **ID:** `2601.16147` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Muhammad Ilham Rizqyawan, Peter Macfarlane, Stathis Hadjidemetriou, et al.
+- **abs:** https://arxiv.org/abs/2601.16147 · **PDF:** https://arxiv.org/pdf/2601.16147v1
+
+## 19916. Computing Fixpoints of Learned Functions: Chaotic Iteration and Simple Stochastic Games
+
+- **ID:** `2601.16142` · **primary:** `cs.LO` · **Submitted:** 2026-01-22
+- **Authors:** Paolo Baldan, Sebastian Gurke, Barbara König, et al.
+- **abs:** https://arxiv.org/abs/2601.16142 · **PDF:** https://arxiv.org/pdf/2601.16142v1
+
+## 19917. Learning to Watermark in the Latent Space of Generative Models
+
+- **ID:** `2601.16140` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Sylvestre-Alvise Rebuffi, Tuan Tran, Valeriu Lacatusu, et al.
+- **abs:** https://arxiv.org/abs/2601.16140 · **PDF:** https://arxiv.org/pdf/2601.16140v1
+
+## 19918. On the Intrinsic Dimensions of Data in Kernel Learning
+
+- **ID:** `2601.16139` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Rustem Takhanov
+- **abs:** https://arxiv.org/abs/2601.16139 · **PDF:** https://arxiv.org/pdf/2601.16139v1
+
+## 19919. Automatic Classification of Arabic Literature into Historical Eras
+
+- **ID:** `2601.16138` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Zainab Alhathloul, Irfan Ahmad
+- **abs:** https://arxiv.org/abs/2601.16138 · **PDF:** https://arxiv.org/pdf/2601.16138v1
+
+## 19920. LLM Prompt Evaluation for Educational Applications
+
+- **ID:** `2601.16134` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Langdon Holmes, Adam Coscia, Scott Crossley, et al.
+- **abs:** https://arxiv.org/abs/2601.16134 · **PDF:** https://arxiv.org/pdf/2601.16134v1
+
+## 19921. Replicating Human Motivated Reasoning Studies with LLMs
+
+- **ID:** `2601.16130` · **primary:** `cs.HC` · **Submitted:** 2026-01-22
+- **Authors:** Neeley Pate, Adiba Mahbub Proma, Hangfeng He, et al.
+- **abs:** https://arxiv.org/abs/2601.16130 · **PDF:** https://arxiv.org/pdf/2601.16130v1
+
+## 19922. Improving Training Efficiency and Reducing Maintenance Costs via Language Specific Model Merging
+
+- **ID:** `2601.16127` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Alphaeus Dmonte, Vidhi Gupta, Daniel J Perry, et al.
+- **abs:** https://arxiv.org/abs/2601.16127 · **PDF:** https://arxiv.org/pdf/2601.16127v2
+
+## 19923. Rethinking Composed Image Retrieval Evaluation: A Fine-Grained Benchmark from Image Editing
+
+- **ID:** `2601.16125` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Tingyu Song, Yanzhao Zhang, Mingxin Li, et al.
+- **abs:** https://arxiv.org/abs/2601.16125 · **PDF:** https://arxiv.org/pdf/2601.16125v1
+
+## 19924. Between Search and Platform: ChatGPT Under the DSA
+
+- **ID:** `2601.17064` · **primary:** `cs.CY` · **Submitted:** 2026-01-22
+- **Authors:** Toni Lorente, Kathrin Gardhouse
+- **abs:** https://arxiv.org/abs/2601.17064 · **PDF:** https://arxiv.org/pdf/2601.17064v2
+
+## 19925. When to Think Fast and Slow? AMOR: Entropy-Based Metacognitive Gate for Dynamic SSM-Attention Switching
+
+- **ID:** `2602.13215` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Haoran Zheng
+- **abs:** https://arxiv.org/abs/2602.13215 · **PDF:** https://arxiv.org/pdf/2602.13215v1
+
+## 19926. Synthetic Augmentation in Imbalanced Learning: When It Helps, When It Hurts, and How Much to Add
+
+- **ID:** `2601.16120` · **primary:** `stat.ML` · **Submitted:** 2026-01-22
+- **Authors:** Zhengchi Ma, Anru R. Zhang
+- **abs:** https://arxiv.org/abs/2601.16120 · **PDF:** https://arxiv.org/pdf/2601.16120v3
+
+## 19927. A Case for Hypergraphs to Model and Map SNNs on Neuromorphic Hardware
+
+- **ID:** `2601.16118` · **primary:** `cs.AR` · **Submitted:** 2026-01-22
+- **Authors:** Marco Ronzani, Cristina Silvano
+- **abs:** https://arxiv.org/abs/2601.16118 · **PDF:** https://arxiv.org/pdf/2601.16118v1
+
+## 19928. FlashMoE: Reducing SSD I/O Bottlenecks via ML-Based Cache Replacement for Mixture-of-Experts Inference on Edge Devices
+
+- **ID:** `2601.17063` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Byeongju Kim, Jungwan Lee, Donghyeon Han, et al.
+- **abs:** https://arxiv.org/abs/2601.17063 · **PDF:** https://arxiv.org/pdf/2601.17063v1
+
+## 19929. synthocr-gen: A synthetic ocr dataset generator for low-resource languages- breaking the data barrier
+
+- **ID:** `2601.16113` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Haq Nawaz Malik, Kh Mohmad Shafi, Tanveer Ahmad Reshi
+- **abs:** https://arxiv.org/abs/2601.16113 · **PDF:** https://arxiv.org/pdf/2601.16113v1
+
+## 19930. Variable Splitting Binary Tree Models Based on Bayesian Context Tree Models for Time Series Segmentation
+
+- **ID:** `2601.16112` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Yuta Nakahara, Shota Saito, Kohei Horinouchi, et al.
+- **abs:** https://arxiv.org/abs/2601.16112 · **PDF:** https://arxiv.org/pdf/2601.16112v1
+
+## 19931. Multimodal Climate Disinformation Detection: Integrating Vision-Language Models with External Knowledge Sources
+
+- **ID:** `2601.16108` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Marzieh Adeli Shamsabad, Hamed Ghodrati
+- **abs:** https://arxiv.org/abs/2601.16108 · **PDF:** https://arxiv.org/pdf/2601.16108v1
+
+## 19932. Benchmarking Deep Learning Models for Raman Spectroscopy Across Open-Source Datasets
+
+- **ID:** `2601.16107` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Adithya Sineesh, Akshita Kamsali
+- **abs:** https://arxiv.org/abs/2601.16107 · **PDF:** https://arxiv.org/pdf/2601.16107v1
+
+## 19933. Clustering-Guided Spatial-Spectral Mamba for Hyperspectral Image Classification
+
+- **ID:** `2601.16098` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Zack Dewis, Yimin Zhu, Zhengsen Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.16098 · **PDF:** https://arxiv.org/pdf/2601.16098v1
+
+## 19934. Adapter Fusion for Multilingual Text2Cypher with Linear and Learned Gating
+
+- **ID:** `2601.16097` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Makbule Gulcin Ozsoy
+- **abs:** https://arxiv.org/abs/2601.16097 · **PDF:** https://arxiv.org/pdf/2601.16097v1
+
+## 19935. Neural Particle Automata: Learning Self-Organizing Particle Dynamics
+
+- **ID:** `2601.16096` · **primary:** `cs.NE` · **Submitted:** 2026-01-22
+- **Authors:** Hyunsoo Kim, Ehsan Pajouheshgar, Sabine Süsstrunk, et al.
+- **abs:** https://arxiv.org/abs/2601.16096 · **PDF:** https://arxiv.org/pdf/2601.16096v1
+
+## 19936. Delayed Assignments in Online Non-Centroid Clustering with Stochastic Arrivals
+
+- **ID:** `2601.16091` · **primary:** `cs.MA` · **Submitted:** 2026-01-22
+- **Authors:** Saar Cohen
+- **abs:** https://arxiv.org/abs/2601.16091 · **PDF:** https://arxiv.org/pdf/2601.16091v1
+
+## 19937. Design and Empirical Study of a Large Language Model-Based Multi-Agent Investment System for Chinese Public REITs
+
+- **ID:** `2602.00082` · **primary:** `q-fin.ST` · **Submitted:** 2026-01-22
+- **Authors:** Zheng Li
+- **abs:** https://arxiv.org/abs/2602.00082 · **PDF:** https://arxiv.org/pdf/2602.00082v1
+
+## 19938. Controlling Long-Horizon Behavior in Language Model Agents with Explicit State Dynamics
+
+- **ID:** `2601.16087` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Sukesh Subaharan
+- **abs:** https://arxiv.org/abs/2601.16087 · **PDF:** https://arxiv.org/pdf/2601.16087v1
+
+## 19939. Probably Approximately Correct Maximum A Posteriori Inference
+
+- **ID:** `2601.16083` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Matthew Shorvon, Frederik Mallmann-Trenn, David S. Watson
+- **abs:** https://arxiv.org/abs/2601.16083 · **PDF:** https://arxiv.org/pdf/2601.16083v1
+
+## 19940. Explainable AI to Improve Machine Learning Reliability for Industrial Cyber-Physical Systems
+
+- **ID:** `2601.16074` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Annemarie Jutte, Uraz Odyurt
+- **abs:** https://arxiv.org/abs/2601.16074 · **PDF:** https://arxiv.org/pdf/2601.16074v1
+
+## 19941. CLASP: An online learning algorithm for Convex Losses And Squared Penalties
+
+- **ID:** `2601.16072` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Ricardo N. Ferreira, João Xavier, Cláudia Soares
+- **abs:** https://arxiv.org/abs/2601.16072 · **PDF:** https://arxiv.org/pdf/2601.16072v2
+
+## 19942. On damage of interpolation to adversarial robustness in regression
+
+- **ID:** `2601.16070` · **primary:** `stat.ML` · **Submitted:** 2026-01-22
+- **Authors:** Jingfu Peng, Yuhong Yang
+- **abs:** https://arxiv.org/abs/2601.16070 · **PDF:** https://arxiv.org/pdf/2601.16070v1
+
+## 19943. A Computer Vision Pipeline for Iterative Bullet Hole Tracking in Rifle Zeroing
+
+- **ID:** `2601.17062` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Robert M. Belcher, Brendan C. Degryse, Leonard R. Kosta, et al.
+- **abs:** https://arxiv.org/abs/2601.17062 · **PDF:** https://arxiv.org/pdf/2601.17062v1
+
+## 19944. How Information Evolves: Stability-Driven Assembly and the Emergence of a Natural Genetic Algorithm
+
+- **ID:** `2601.17061` · **primary:** `q-bio.PE` · **Submitted:** 2026-01-22
+- **Authors:** Dan Adler
+- **abs:** https://arxiv.org/abs/2601.17061 · **PDF:** https://arxiv.org/pdf/2601.17061v1
+
+## 19945. Prediction Markets as Bayesian Inverse Problems: Uncertainty Quantification, Identifiability, and Information Gain from Price-Volume Histories under Latent Types
+
+- **ID:** `2601.18815` · **primary:** `q-fin.MF` · **Submitted:** 2026-01-22
+- **Authors:** Juan Pablo Madrigal-Cianci, Camilo Monsalve Maya, Lachlan Breakey
+- **abs:** https://arxiv.org/abs/2601.18815 · **PDF:** https://arxiv.org/pdf/2601.18815v1
+
+## 19946. Designing faster mixed integer linear programming algorithm via learning the optimal path
+
+- **ID:** `2601.16056` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Ruizhi Liu, Liming Xu, Xulin Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.16056 · **PDF:** https://arxiv.org/pdf/2601.16056v1
+
+## 19947. AgriPINN: A Process-Informed Neural Network for Interpretable and Scalable Crop Biomass Prediction Under Water Stress
+
+- **ID:** `2601.16045` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Yue Shi, Liangxiu Han, Xin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.16045 · **PDF:** https://arxiv.org/pdf/2601.16045v1
+
+## 19948. Risk reversal for least squares estimators under nested convex constraints
+
+- **ID:** `2601.16041` · **primary:** `math.ST` · **Submitted:** 2026-01-22
+- **Authors:** Omar Al-Ghattas
+- **abs:** https://arxiv.org/abs/2601.16041 · **PDF:** https://arxiv.org/pdf/2601.16041v1
+
+## 19949. Grounding Large Language Models in Reaction Knowledge Graphs for Synthesis Retrieval
+
+- **ID:** `2601.16038` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Olga Bunkova, Lorenzo Di Fruscia, Sophia Rupprecht, et al.
+- **abs:** https://arxiv.org/abs/2601.16038 · **PDF:** https://arxiv.org/pdf/2601.16038v1
+
+## 19950. Universal Refusal Circuits Across LLMs: Cross-Model Transfer via Trajectory Replay and Concept-Basis Reconstruction
+
+- **ID:** `2601.16034` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Tony Cristofano
+- **abs:** https://arxiv.org/abs/2601.16034 · **PDF:** https://arxiv.org/pdf/2601.16034v2
+
+## 19951. Sawtooth Wavefront Reordering: Enhanced CuTile FlashAttention on NVIDIA GB10
+
+- **ID:** `2601.16032` · **primary:** `cs.PF` · **Submitted:** 2026-01-22
+- **Authors:** Yifan Zhu, Yekai Pan, Chen Ding
+- **abs:** https://arxiv.org/abs/2601.16032 · **PDF:** https://arxiv.org/pdf/2601.16032v2
+
+## 19952. Initial results of the Digital Consciousness Model
+
+- **ID:** `2601.17060` · **primary:** `cs.CY` · **Submitted:** 2026-01-22
+- **Authors:** Derek Shiller, Laura Duffy, Arvo Muñoz Morán, et al.
+- **abs:** https://arxiv.org/abs/2601.17060 · **PDF:** https://arxiv.org/pdf/2601.17060v1
+
+## 19953. Data-Driven Conditional Flexibility Index
+
+- **ID:** `2601.16028` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Moritz Wedemeyer, Eike Cramer, Alexander Mitsos, et al.
+- **abs:** https://arxiv.org/abs/2601.16028 · **PDF:** https://arxiv.org/pdf/2601.16028v2
+
+## 19954. Deja Vu in Plots: Leveraging Cross-Session Evidence with Retrieval-Augmented LLMs for Live Streaming Risk Assessment
+
+- **ID:** `2601.16027` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Yiran Qiao, Xiang Ao, Jing Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.16027 · **PDF:** https://arxiv.org/pdf/2601.16027v1
+
+## 19955. Mecellem Models: Turkish Models Trained from Scratch and Continually Pre-trained for the Legal Domain
+
+- **ID:** `2601.16018` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Özgür Uğur, Mahmut Göksu, Mahmut Çimen, et al.
+- **abs:** https://arxiv.org/abs/2601.16018 · **PDF:** https://arxiv.org/pdf/2601.16018v1
+
+## 19956. THOR: A Versatile Foundation Model for Earth Observation Climate and Society Applications
+
+- **ID:** `2601.16011` · **primary:** `eess.IV` · **Submitted:** 2026-01-22
+- **Authors:** Theodor Forgaard, Jarle H. Reksten, Anders U. Waldeland, et al.
+- **abs:** https://arxiv.org/abs/2601.16011 · **PDF:** https://arxiv.org/pdf/2601.16011v1
+
+## 19957. PhysicsMind: Sim and Real Mechanics Benchmarking for Physical Reasoning and Prediction in Foundational VLMs and World Models
+
+- **ID:** `2601.16007` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Chak-Wing Mak, Guanyu Zhu, Boyi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.16007 · **PDF:** https://arxiv.org/pdf/2601.16007v1
+
+## 19958. KD4MT: A Survey of Knowledge Distillation for Machine Translation
+
+- **ID:** `2602.15845` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Ona de Gibert, Joseph Attieh, Timothee Mickus, et al.
+- **abs:** https://arxiv.org/abs/2602.15845 · **PDF:** https://arxiv.org/pdf/2602.15845v1
+
+## 19959. PUMA: Perception-driven Unified Foothold Prior for Mobility Augmented Quadruped Parkour
+
+- **ID:** `2601.15995` · **primary:** `cs.RO` · **Submitted:** 2026-01-22
+- **Authors:** Liang Wang, Kanzhong Yao, Yang Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.15995 · **PDF:** https://arxiv.org/pdf/2601.15995v1
+
+## 19960. Partially Lazy Gradient Descent for Smoothed Online Learning
+
+- **ID:** `2601.15984` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Naram Mhaisen, George Iosifidis
+- **abs:** https://arxiv.org/abs/2601.15984 · **PDF:** https://arxiv.org/pdf/2601.15984v1
+
+## 19961. Predicting Healthcare System Visitation Flow by Integrating Hospital Attributes and Population Socioeconomics with Human Mobility Data
+
+- **ID:** `2601.15977` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Binbin Lin, Lei Zou, Hao Tian, et al.
+- **abs:** https://arxiv.org/abs/2601.15977 · **PDF:** https://arxiv.org/pdf/2601.15977v1
+
+## 19962. GLEAN: Grounded Lightweight Evaluation Anchors for Contamination-Aware Tabular Reasoning
+
+- **ID:** `2603.02212` · **primary:** `cs.DB` · **Submitted:** 2026-01-22
+- **Authors:** Qizhi Wang
+- **abs:** https://arxiv.org/abs/2603.02212 · **PDF:** https://arxiv.org/pdf/2603.02212v1
+
+## 19963. Decoupling Return-to-Go for Efficient Decision Transformer
+
+- **ID:** `2601.15953` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Yongyi Wang, Hanyu Liu, Lingfeng Li, et al.
+- **abs:** https://arxiv.org/abs/2601.15953 · **PDF:** https://arxiv.org/pdf/2601.15953v1
+
+## 19964. Natural Language-Driven Global Mapping of Martian Landforms
+
+- **ID:** `2601.15949` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Yiran Wang, Shuoyuan Wang, Zhaoran Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.15949 · **PDF:** https://arxiv.org/pdf/2601.15949v1
+
+## 19965. BotzoneBench: Scalable LLM Evaluation via Graded AI Anchors
+
+- **ID:** `2602.13214` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Lingfeng Li, Yunlong Lu, Yuefei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.13214 · **PDF:** https://arxiv.org/pdf/2602.13214v1
+
+## 19966. ICON: Invariant Counterfactual Optimization with Neuro-Symbolic Priors for Text-Based Person Search
+
+- **ID:** `2601.15931` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Xiangyu Wang, Zhixin Lv, Yongjiao Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.15931 · **PDF:** https://arxiv.org/pdf/2601.15931v1
+
+## 19967. MMGRid: Navigating Temporal-aware and Cross-domain Generative Recommendation via Model Merging
+
+- **ID:** `2601.15930` · **primary:** `cs.IR` · **Submitted:** 2026-01-22
+- **Authors:** Tianjun Wei, Enneng Yang, Yingpeng Du, et al.
+- **abs:** https://arxiv.org/abs/2601.15930 · **PDF:** https://arxiv.org/pdf/2601.15930v2
+
+## 19968. Class Confidence Aware Reweighting for Long Tailed Learning
+
+- **ID:** `2601.15924` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Brainard Philemon Jagati, Jitendra Tembhurne, Harsh Goud, et al.
+- **abs:** https://arxiv.org/abs/2601.15924 · **PDF:** https://arxiv.org/pdf/2601.15924v1
+
+## 19969. Progressive Power Homotopy for Non-convex Optimization
+
+- **ID:** `2601.15915` · **primary:** `math.OC` · **Submitted:** 2026-01-22
+- **Authors:** Chen Xu
+- **abs:** https://arxiv.org/abs/2601.15915 · **PDF:** https://arxiv.org/pdf/2601.15915v1
+
+## 19970. TeNet: Text-to-Network for Compact Policy Synthesis
+
+- **ID:** `2601.15912` · **primary:** `cs.RO` · **Submitted:** 2026-01-22
+- **Authors:** Ariyan Bighashdel, Kevin Sebastian Luck
+- **abs:** https://arxiv.org/abs/2601.15912 · **PDF:** https://arxiv.org/pdf/2601.15912v1
+
+## 19971. Transfer Learning from ImageNet for MEG-Based Decoding of Imagined Speech
+
+- **ID:** `2601.15909` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Soufiane Jhilal, Stéphanie Martin, Anne-Lise Giraud
+- **abs:** https://arxiv.org/abs/2601.15909 · **PDF:** https://arxiv.org/pdf/2601.15909v1
+
+## 19972. Iterative Amortized Hierarchical VAE
+
+- **ID:** `2601.15894` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Simon W. Penninga, Ruud J. G. van Sloun
+- **abs:** https://arxiv.org/abs/2601.15894 · **PDF:** https://arxiv.org/pdf/2601.15894v1
+
+## 19973. Stable-DiffCoder: Pushing the Frontier of Code Diffusion Large Language Model
+
+- **ID:** `2601.15892` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Chenghao Fan, Wen Heng, Bo Li, et al.
+- **abs:** https://arxiv.org/abs/2601.15892 · **PDF:** https://arxiv.org/pdf/2601.15892v2
+
+## 19974. Understanding the Transfer Limits of Vision Foundation Models
+
+- **ID:** `2601.15888` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Shiqi Huang, Yipei Wang, Natasha Thorley, et al.
+- **abs:** https://arxiv.org/abs/2601.15888 · **PDF:** https://arxiv.org/pdf/2601.15888v1
+
+## 19975. Can LLMs Clean Up Your Mess? A Survey of Application-Ready Data Preparation with LLMs
+
+- **ID:** `2601.17058` · **primary:** `cs.DB` · **Submitted:** 2026-01-22
+- **Authors:** Wei Zhou, Jun Zhou, Haoyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.17058 · **PDF:** https://arxiv.org/pdf/2601.17058v1
+
+## 19976. Evaluating and Achieving Controllable Code Completion in Code LLM
+
+- **ID:** `2601.15879` · **primary:** `cs.SE` · **Submitted:** 2026-01-22
+- **Authors:** Jiajun Zhang, Zeyu Cui, Lei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.15879 · **PDF:** https://arxiv.org/pdf/2601.15879v1
+
+## 19977. EvoCUA: Evolving Computer Use Agents via Learning from Scalable Synthetic Experience
+
+- **ID:** `2601.15876` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Taofeng Xue, Chong Peng, Mianqiu Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.15876 · **PDF:** https://arxiv.org/pdf/2601.15876v2
+
+## 19978. SoK: Challenges in Tabular Membership Inference Attacks
+
+- **ID:** `2601.15874` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Cristina Pêra, Tânia Carvalho, Maxime Cordy, et al.
+- **abs:** https://arxiv.org/abs/2601.15874 · **PDF:** https://arxiv.org/pdf/2601.15874v1
+
+## 19979. PF-D2M: A Pose-free Diffusion Model for Universal Dance-to-Music Generation
+
+- **ID:** `2601.15872` · **primary:** `cs.SD` · **Submitted:** 2026-01-22
+- **Authors:** Jaekwon Im, Natalia Polouliakh, Taketo Akama
+- **abs:** https://arxiv.org/abs/2601.15872 · **PDF:** https://arxiv.org/pdf/2601.15872v1
+
+## 19980. Why Inference in Large Models Becomes Decomposable After Training
+
+- **ID:** `2601.15871` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Jidong Jin
+- **abs:** https://arxiv.org/abs/2601.15871 · **PDF:** https://arxiv.org/pdf/2601.15871v3
+
+## 19981. Artificial Rigidities vs. Biological Noise: A Comparative Analysis of Multisensory Integration in AV-HuBERT and Human Observers
+
+- **ID:** `2601.15869` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Francisco Portillo López
+- **abs:** https://arxiv.org/abs/2601.15869 · **PDF:** https://arxiv.org/pdf/2601.15869v1
+
+## 19982. Lightweight Quantum-Enhanced ResNet for Coronary Angiography Classification: A Hybrid Quantum-Classical Feature Enhancement Framework
+
+- **ID:** `2601.18814` · **primary:** `quant-ph` · **Submitted:** 2026-01-22
+- **Authors:** Jingsong Xia
+- **abs:** https://arxiv.org/abs/2601.18814 · **PDF:** https://arxiv.org/pdf/2601.18814v1
+
+## 19983. A Lightweight Brain-Inspired Machine Learning Framework for Coronary Angiography: Hybrid Neural Representation and Robust Learning Strategies
+
+- **ID:** `2601.15865` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Jingsong Xia, Siqi Wang
+- **abs:** https://arxiv.org/abs/2601.15865 · **PDF:** https://arxiv.org/pdf/2601.15865v1
+
+## 19984. Uncertainty-guided Generation of Dark-field Radiographs
+
+- **ID:** `2601.15859` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Lina Felsner, Henriette Bast, Tina Dorosti, et al.
+- **abs:** https://arxiv.org/abs/2601.15859 · **PDF:** https://arxiv.org/pdf/2601.15859v1
+
+## 19985. Determinants of Training Corpus Size for Clinical Text Classification
+
+- **ID:** `2601.15846` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Jaya Chaturvedi, Saniya Deshpande, Chenkai Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.15846 · **PDF:** https://arxiv.org/pdf/2601.15846v1
+
+## 19986. Can professional translators identify machine-generated text?
+
+- **ID:** `2601.15828` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Michael Farrell
+- **abs:** https://arxiv.org/abs/2601.15828 · **PDF:** https://arxiv.org/pdf/2601.15828v2
+
+## 19987. Introducing the Generative Application Firewall (GAF)
+
+- **ID:** `2601.15824` · **primary:** `cs.CR` · **Submitted:** 2026-01-22
+- **Authors:** Joan Vendrell Farreny, Martí Jordà Roca, Miquel Cornudella Gaya, et al.
+- **abs:** https://arxiv.org/abs/2601.15824 · **PDF:** https://arxiv.org/pdf/2601.15824v2
+
+## 19988. ExDR: Explanation-driven Dynamic Retrieval Enhancement for Multimodal Fake News Detection
+
+- **ID:** `2601.15820` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Guoxuan Ding, Yuqing Li, Ziyan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.15820 · **PDF:** https://arxiv.org/pdf/2601.15820v1
+
+## 19989. Virtual Traffic Police: Large Language Model-Augmented Traffic Signal Control for Unforeseen Incidents
+
+- **ID:** `2601.15816` · **primary:** `eess.SY` · **Submitted:** 2026-01-22
+- **Authors:** Shiqi Wei, Qiqing Wang, Kaidi Yang
+- **abs:** https://arxiv.org/abs/2601.15816 · **PDF:** https://arxiv.org/pdf/2601.15816v1
+
+## 19990. Beyond Off-the-Shelf Models: A Lightweight and Accessible Machine Learning Pipeline for Ecologists Working with Image Data
+
+- **ID:** `2601.15813` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Clare Chemery, Hendrik Edelhoff, Ludwig Bothmann
+- **abs:** https://arxiv.org/abs/2601.15813 · **PDF:** https://arxiv.org/pdf/2601.15813v1
+
+## 19991. ErrorMap and ErrorAtlas: Charting the Failure Landscape of Large Language Models
+
+- **ID:** `2601.15812` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Shir Ashury-Tahan, Yifan Mai, Elron Bandel, et al.
+- **abs:** https://arxiv.org/abs/2601.15812 · **PDF:** https://arxiv.org/pdf/2601.15812v2
+
+## 19992. EduResearchBench: A Hierarchical Atomic Task Decomposition Benchmark for Full-Lifecycle Educational Research
+
+- **ID:** `2602.15034` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Houping Yue, Zixiang Di, Mei Jiang, et al.
+- **abs:** https://arxiv.org/abs/2602.15034 · **PDF:** https://arxiv.org/pdf/2602.15034v1
+
+## 19993. A Mobile Application for Flower Recognition System Based on Convolutional Neural Networks
+
+- **ID:** `2601.15810` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Mustafa Yurdakul, Enes Ayan, Fahrettin Horasan, et al.
+- **abs:** https://arxiv.org/abs/2601.15810 · **PDF:** https://arxiv.org/pdf/2601.15810v1
+
+## 19994. SteerEval: Inference-time Interventions Strengthen Multilingual Generalization in Neural Summarization Metrics
+
+- **ID:** `2601.15809` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Silvia Casola, Ryan Soh-Eun Shim, Felicia Körner, et al.
+- **abs:** https://arxiv.org/abs/2601.15809 · **PDF:** https://arxiv.org/pdf/2601.15809v1
+
+## 19995. Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verification
+
+- **ID:** `2601.15808` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Yuxuan Wan, Tianqing Fang, Zaitang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.15808 · **PDF:** https://arxiv.org/pdf/2601.15808v1
+
+## 19996. Algebraic Statistics in OSCAR
+
+- **ID:** `2601.15807` · **primary:** `stat.CO` · **Submitted:** 2026-01-22
+- **Authors:** Tobias Boege, Antony Della Vecchia, Marina Garrote-López, et al.
+- **abs:** https://arxiv.org/abs/2601.15807 · **PDF:** https://arxiv.org/pdf/2601.15807v1
+
+## 19997. A Beacon Based Solution for Autonomous UUVs GNSS-Denied Stealthy Navigation
+
+- **ID:** `2601.15802` · **primary:** `cs.RO` · **Submitted:** 2026-01-22
+- **Authors:** Alexandre Albore, Humbert Fiorino, Damien Pellier
+- **abs:** https://arxiv.org/abs/2601.15802 · **PDF:** https://arxiv.org/pdf/2601.15802v1
+
+## 19998. Attributing and Exploiting Safety Vectors through Global Optimization in Large Language Models
+
+- **ID:** `2601.15801` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Fengheng Chu, Jiahao Chen, Yuhong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.15801 · **PDF:** https://arxiv.org/pdf/2601.15801v1
+
+## 19999. VitalDiagnosis: AI-Driven Ecosystem for 24/7 Vital Monitoring and Chronic Disease Management
+
+- **ID:** `2601.15798` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Zhikai Xue, Tianqianjin Lin, Pengwei Yan, et al.
+- **abs:** https://arxiv.org/abs/2601.15798 · **PDF:** https://arxiv.org/pdf/2601.15798v1
+
+## 20000. Creativity in the Age of AI: Rethinking the Role of Intentional Agency
+
+- **ID:** `2601.15797` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** James S. Pearson, Matthew J. Dennis, Marc Cheong
+- **abs:** https://arxiv.org/abs/2601.15797 · **PDF:** https://arxiv.org/pdf/2601.15797v1
+
+## 20001. HumanLLM: Towards Personalized Understanding and Simulation of Human Nature
+
+- **ID:** `2601.15793` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Yuxuan Lei, Tianfu Wang, Jianxun Lian, et al.
+- **abs:** https://arxiv.org/abs/2601.15793 · **PDF:** https://arxiv.org/pdf/2601.15793v1
+
+## 20002. Agentic Confidence Calibration
+
+- **ID:** `2601.15778` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Jiaxin Zhang, Caiming Xiong, Chien-Sheng Wu
+- **abs:** https://arxiv.org/abs/2601.15778 · **PDF:** https://arxiv.org/pdf/2601.15778v1
+
+## 20003. Next Generation Active Learning: Mixture of LLMs in the Loop
+
+- **ID:** `2601.15773` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Yuanyuan Qi, Xiaohao Yang, Jueqing Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.15773 · **PDF:** https://arxiv.org/pdf/2601.15773v1
+
+## 20004. Rethinking Drug-Drug Interaction Modeling as Generalizable Relation Learning
+
+- **ID:** `2601.15771` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Dong Xu, Jiantao Wu, Qihua Pan, et al.
+- **abs:** https://arxiv.org/abs/2601.15771 · **PDF:** https://arxiv.org/pdf/2601.15771v1
+
+## 20005. Off-Policy Actor-Critic with Sigmoid-Bounded Entropy for Real-World Robot Learning
+
+- **ID:** `2601.15761` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Xiefeng Wu, Mingyu Hu, Shu Zhang
+- **abs:** https://arxiv.org/abs/2601.15761 · **PDF:** https://arxiv.org/pdf/2601.15761v1
+
+## 20006. Atlas-Assisted Segment Anything Model for Fetal Brain MRI (FeTal-SAM)
+
+- **ID:** `2601.15759` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Qi Zeng, Weide Liu, Bo Li, et al.
+- **abs:** https://arxiv.org/abs/2601.15759 · **PDF:** https://arxiv.org/pdf/2601.15759v1
+
+## 20007. Beyond Marginal Distributions: A Framework to Evaluate the Representativeness of Demographic-Aligned LLMs
+
+- **ID:** `2601.15755` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Tristan Williams, Franziska Weeber, Sebastian Padó, et al.
+- **abs:** https://arxiv.org/abs/2601.15755 · **PDF:** https://arxiv.org/pdf/2601.15755v2
+
+## 20008. CAFE-GB: Scalable and Stable Feature Selection for Malware Detection via Chunk-wise Aggregated Gradient Boosting
+
+- **ID:** `2601.15754` · **primary:** `cs.CR` · **Submitted:** 2026-01-22
+- **Authors:** Ajvad Haneef K, Karan Kuwar Singh, Madhu Kumar S D
+- **abs:** https://arxiv.org/abs/2601.15754 · **PDF:** https://arxiv.org/pdf/2601.15754v1
+
+## 20009. Tabular Incremental Inference
+
+- **ID:** `2601.15751` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Xinda Chen, Zhen Xing, Hanyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.15751 · **PDF:** https://arxiv.org/pdf/2601.15751v2
+
+## 20010. Hallucination Mitigating for Medical Report Generation
+
+- **ID:** `2601.15745` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Ruoqing Zhao, Runze Xia, Piji Li
+- **abs:** https://arxiv.org/abs/2601.15745 · **PDF:** https://arxiv.org/pdf/2601.15745v1
+
+## 20011. EvoDR: Evolving Dispatching Rules via Large Language Model for Dynamic Flexible Assembly Flow Shop Scheduling
+
+- **ID:** `2601.15738` · **primary:** `cs.NE` · **Submitted:** 2026-01-22
+- **Authors:** Junhao Qiu, Haoyang Zhuang, Fei Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.15738 · **PDF:** https://arxiv.org/pdf/2601.15738v2
+
+## 20012. PhysProver: Advancing Automatic Theorem Proving for Physics
+
+- **ID:** `2601.15737` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Hanning Zhang, Ruida Wang, Rui Pan, et al.
+- **abs:** https://arxiv.org/abs/2601.15737 · **PDF:** https://arxiv.org/pdf/2601.15737v1
+
+## 20013. FAIR-ESI: Feature Adaptive Importance Refinement for Electrophysiological Source Imaging
+
+- **ID:** `2601.15731` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Linyong Zou, Liang Zhang, Xiongfei Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.15731 · **PDF:** https://arxiv.org/pdf/2601.15731v1
+
+## 20014. DualShield: Safe Model Predictive Diffusion via Reachability Analysis for Interactive Autonomous Driving
+
+- **ID:** `2601.15729` · **primary:** `cs.RO` · **Submitted:** 2026-01-22
+- **Authors:** Rui Yang, Lei Zheng, Ruoyu Yao, et al.
+- **abs:** https://arxiv.org/abs/2601.15729 · **PDF:** https://arxiv.org/pdf/2601.15729v1
+
+## 20015. Benchmarking Text-to-Python against Text-to-SQL: The Impact of Explicit Logic and Ambiguity
+
+- **ID:** `2601.15728` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Hangle Hu, Chenyu Hou, Bin Cao, et al.
+- **abs:** https://arxiv.org/abs/2601.15728 · **PDF:** https://arxiv.org/pdf/2601.15728v2
+
+## 20016. Towards Automated Kernel Generation in the Era of LLMs
+
+- **ID:** `2601.15727` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Yang Yu, Peiyu Zang, Chi Hsu Tsai, et al.
+- **abs:** https://arxiv.org/abs/2601.15727 · **PDF:** https://arxiv.org/pdf/2601.15727v2
+
+## 20017. VideoThinker: Building Agentic VideoLLMs with LLM-Guided Tool Reasoning
+
+- **ID:** `2601.15724` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Chenglin Li, Qianglong Chen, Feng Han, et al.
+- **abs:** https://arxiv.org/abs/2601.15724 · **PDF:** https://arxiv.org/pdf/2601.15724v1
+
+## 20018. Communication-efficient Federated Graph Classification via Generative Diffusion Modeling
+
+- **ID:** `2601.15722` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Xiuling Wang, Xin Huang, Haibo Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.15722 · **PDF:** https://arxiv.org/pdf/2601.15722v1
+
+## 20019. CoNRec: Context-Discerning Negative Recommendation with LLMs
+
+- **ID:** `2601.15721` · **primary:** `cs.IR` · **Submitted:** 2026-01-22
+- **Authors:** Xinda Chen, Jiawei Wu, Yishuang Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.15721 · **PDF:** https://arxiv.org/pdf/2601.15721v1
+
+## 20020. Investigation of the Generalisation Ability of Genetic Programming-evolved Scheduling Rules in Dynamic Flexible Job Shop Scheduling
+
+- **ID:** `2601.15717` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Luyao Zhu, Fangfang Zhang, Yi Mei, et al.
+- **abs:** https://arxiv.org/abs/2601.15717 · **PDF:** https://arxiv.org/pdf/2601.15717v1
+
+## 20021. RebuttalAgent: Strategic Persuasion in Academic Rebuttal via Theory of Mind
+
+- **ID:** `2601.15715` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Zhitao He, Zongwei Lyu, Yi R Fung
+- **abs:** https://arxiv.org/abs/2601.15715 · **PDF:** https://arxiv.org/pdf/2601.15715v3
+
+## 20022. Even GPT-5.2 Can't Count to Five: The Case for Zero-Error Horizons in Trustworthy LLMs
+
+- **ID:** `2601.15714` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Ryoma Sato
+- **abs:** https://arxiv.org/abs/2601.15714 · **PDF:** https://arxiv.org/pdf/2601.15714v1
+
+## 20023. FlexLLM: Composable HLS Library for Flexible Hybrid LLM Accelerator Design
+
+- **ID:** `2601.15710` · **primary:** `cs.AR` · **Submitted:** 2026-01-22
+- **Authors:** Jiahao Zhang, Zifan He, Nicholas Fraser, et al.
+- **abs:** https://arxiv.org/abs/2601.15710 · **PDF:** https://arxiv.org/pdf/2601.15710v1
+
+## 20024. AgentSM: Semantic Memory for Agentic Text-to-SQL
+
+- **ID:** `2601.15709` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Asim Biswal, Chuan Lei, Xiao Qin, et al.
+- **abs:** https://arxiv.org/abs/2601.15709 · **PDF:** https://arxiv.org/pdf/2601.15709v1
+
+## 20025. Persona Switch: Mixing Distinct Perspectives in Decoding Time
+
+- **ID:** `2601.15708` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Junseok Kim, Nakyeong Yang, Kyomin Jung
+- **abs:** https://arxiv.org/abs/2601.15708 · **PDF:** https://arxiv.org/pdf/2601.15708v1
+
+## 20026. Adaptive Attribute-Decoupled Encryption for Trusted Respiratory Monitoring in Resource-Limited Consumer Healthcare
+
+- **ID:** `2601.16241` · **primary:** `cs.CR` · **Submitted:** 2026-01-22
+- **Authors:** Xinyu Li, Jinyang Huang, Feng-Qi Cui, et al.
+- **abs:** https://arxiv.org/abs/2601.16241 · **PDF:** https://arxiv.org/pdf/2601.16241v2
+
+## 20027. Improving Methodologies for LLM Evaluations Across Global Languages
+
+- **ID:** `2601.15706` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Akriti Vij, Benjamin Chua, Darshini Ramiah, et al.
+- **abs:** https://arxiv.org/abs/2601.15706 · **PDF:** https://arxiv.org/pdf/2601.15706v1
+
+## 20028. Agentic Uncertainty Quantification
+
+- **ID:** `2601.15703` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Jiaxin Zhang, Prafulla Kumar Choubey, Kung-Hsiang Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.15703 · **PDF:** https://arxiv.org/pdf/2601.15703v1
+
+## 20029. Beyond Visual Safety: Jailbreaking Multimodal Large Language Models for Harmful Image Generation via Semantic-Agnostic Inputs
+
+- **ID:** `2601.15698` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Mingyu Yu, Lana Liu, Zhehao Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.15698 · **PDF:** https://arxiv.org/pdf/2601.15698v1
+
+## 20030. Balancing Security and Privacy: The Pivotal Role of AI in Modern Healthcare Systems
+
+- **ID:** `2601.15697` · **primary:** `cs.CR` · **Submitted:** 2026-01-22
+- **Authors:** Binu V P, Deepthy K Bhaskar, Minimol B
+- **abs:** https://arxiv.org/abs/2601.15697 · **PDF:** https://arxiv.org/pdf/2601.15697v1
+
+## 20031. Learning Functional Graphs with Nonlinear Sufficient Dimension Reduction
+
+- **ID:** `2601.15696` · **primary:** `stat.ME` · **Submitted:** 2026-01-22
+- **Authors:** Kyongwon Kim, Bing Li
+- **abs:** https://arxiv.org/abs/2601.15696 · **PDF:** https://arxiv.org/pdf/2601.15696v1
+
+## 20032. From Passive Metric to Active Signal: The Evolving Role of Uncertainty Quantification in Large Language Models
+
+- **ID:** `2601.15690` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Jiaxin Zhang, Wendi Cui, Zhuohang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.15690 · **PDF:** https://arxiv.org/pdf/2601.15690v1
+
+## 20033. Performance-guided Reinforced Active Learning for Object Detection
+
+- **ID:** `2601.15688` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Zhixuan Liang, Xingyu Zeng, Rui Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.15688 · **PDF:** https://arxiv.org/pdf/2601.15688v1
+
+## 20034. FARM: Field-Aware Resolution Model for Intelligent Trigger-Action Automation
+
+- **ID:** `2601.15687` · **primary:** `cs.SE` · **Submitted:** 2026-01-22
+- **Authors:** Khusrav Badalov, Young Yoon
+- **abs:** https://arxiv.org/abs/2601.15687 · **PDF:** https://arxiv.org/pdf/2601.15687v1
+
+## 20035. Beyond Hard Writes and Rigid Preservation: Soft Recursive Least-Squares for Lifelong LLM Editing
+
+- **ID:** `2601.15686` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Xinyu Wang, Sicheng Lyu, Yu Gu, et al.
+- **abs:** https://arxiv.org/abs/2601.15686 · **PDF:** https://arxiv.org/pdf/2601.15686v1
+
+## 20036. Improving Methodologies for Agentic Evaluations Across Domains: Leakage of Sensitive Information, Fraud and Cybersecurity Threats
+
+- **ID:** `2601.15679` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Ee Wei Seah, Yongsen Zheng, Naga Nikshith, et al.
+- **abs:** https://arxiv.org/abs/2601.15679 · **PDF:** https://arxiv.org/pdf/2601.15679v1
+
+## 20037. Connect the Dots: Knowledge Graph-Guided Crawler Attack on Retrieval-Augmented Generation Systems
+
+- **ID:** `2601.15678` · **primary:** `cs.CR` · **Submitted:** 2026-01-22
+- **Authors:** Mengyu Yao, Ziqi Zhang, Ning Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.15678 · **PDF:** https://arxiv.org/pdf/2601.15678v2
+
+## 20038. Bridging the Perception Gap: A Lightweight Coarse-to-Fine Architecture for Edge Audio Systems
+
+- **ID:** `2601.15676` · **primary:** `cs.SD` · **Submitted:** 2026-01-22
+- **Authors:** Hengfan Zhang, Yueqian Lin, Hai Helen Li, et al.
+- **abs:** https://arxiv.org/abs/2601.15676 · **PDF:** https://arxiv.org/pdf/2601.15676v1
+
+## 20039. What Patients Really Ask: Exploring the Effect of False Assumptions in Patient Information Seeking
+
+- **ID:** `2601.15674` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Raymond Xiong, Furong Jia, Lionel Wong, et al.
+- **abs:** https://arxiv.org/abs/2601.15674 · **PDF:** https://arxiv.org/pdf/2601.15674v2
+
+## 20040. Learning-augmented smooth integer programs with PAC-learnable oracles
+
+- **ID:** `2602.02505` · **primary:** `cs.DS` · **Submitted:** 2026-01-22
+- **Authors:** Hao-Yuan He, Ming Li
+- **abs:** https://arxiv.org/abs/2602.02505 · **PDF:** https://arxiv.org/pdf/2602.02505v1
+
+## 20041. Enhancing guidance for missing data in diffusion-based sequential recommendation
+
+- **ID:** `2601.15673` · **primary:** `cs.IR` · **Submitted:** 2026-01-22
+- **Authors:** Qilong Yan, Yifei Xing, Dugang Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.15673 · **PDF:** https://arxiv.org/pdf/2601.15673v2
+
+## 20042. StreetDesignAI: A Multi-Persona Evaluation System for Inclusive Infrastructure Design
+
+- **ID:** `2601.15671` · **primary:** `cs.HC` · **Submitted:** 2026-01-22
+- **Authors:** Ziyi Wang, Yilong Dai, Duanya Lyu, et al.
+- **abs:** https://arxiv.org/abs/2601.15671 · **PDF:** https://arxiv.org/pdf/2601.15671v1
+
+## 20043. Dualformer: Time-Frequency Dual Domain Learning for Long-term Time Series Forecasting
+
+- **ID:** `2601.15669` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Jingjing Bai, Yoshinobu Kawahara
+- **abs:** https://arxiv.org/abs/2601.15669 · **PDF:** https://arxiv.org/pdf/2601.15669v1
+
+## 20044. Skywork UniPic 3.0: Unified Multi-Image Composition via Sequence Modeling
+
+- **ID:** `2601.15664` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Hongyang Wei, Hongbo Liu, Zidong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.15664 · **PDF:** https://arxiv.org/pdf/2601.15664v1
+
+## 20045. TempoNet: Learning Realistic Communication and Timing Patterns for Network Traffic Simulation
+
+- **ID:** `2601.15663` · **primary:** `cs.CR` · **Submitted:** 2026-01-22
+- **Authors:** Kristen Moore, Diksha Goel, Cody James Christopher, et al.
+- **abs:** https://arxiv.org/abs/2601.15663 · **PDF:** https://arxiv.org/pdf/2601.15663v1
+
+## 20046. The GT-Score: A Robust Objective Function for Reducing Overfitting in Data-Driven Trading Strategies
+
+- **ID:** `2602.00080` · **primary:** `q-fin.ST` · **Submitted:** 2026-01-22
+- **Authors:** Alexander Sheppert
+- **abs:** https://arxiv.org/abs/2602.00080 · **PDF:** https://arxiv.org/pdf/2602.00080v1
+
+## 20047. Integrating Knowledge Distillation Methods: A Sequential Multi-Stage Framework
+
+- **ID:** `2601.15657` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Yinxi Tian, Changwu Huang, Ke Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.15657 · **PDF:** https://arxiv.org/pdf/2601.15657v1
+
+## 20048. Event-VStream: Event-Driven Real-Time Understanding for Long Video Streams
+
+- **ID:** `2601.15655` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Zhenghui Guo, Yuanbin Man, Junyuan Sheng, et al.
+- **abs:** https://arxiv.org/abs/2601.15655 · **PDF:** https://arxiv.org/pdf/2601.15655v1
+
+## 20049. Predictive Coding and Information Bottleneck for Hallucination Detection in Large Language Models
+
+- **ID:** `2601.15652` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Manish Bhatt
+- **abs:** https://arxiv.org/abs/2601.15652 · **PDF:** https://arxiv.org/pdf/2601.15652v1
+
+## 20050. Towards Reliable Medical LLMs: Benchmarking and Enhancing Confidence Estimation of Large Language Models in Medical Consultation
+
+- **ID:** `2601.15645` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Zhiyao Ren, Yibing Zhan, Siyuan Liang, et al.
+- **abs:** https://arxiv.org/abs/2601.15645 · **PDF:** https://arxiv.org/pdf/2601.15645v1
+
+## 20051. Machine Failure Detection Based on Projected Quantum Models
+
+- **ID:** `2601.15641` · **primary:** `quant-ph` · **Submitted:** 2026-01-22
+- **Authors:** Larry Bowden, Qi Chu, Bernard Cena, et al.
+- **abs:** https://arxiv.org/abs/2601.15641 · **PDF:** https://arxiv.org/pdf/2601.15641v1
+
+## 20052. An Empirical Study on Ensemble-Based Transfer Learning Bayesian Optimisation with Mixed Variable Types
+
+- **ID:** `2601.15640` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Natasha Trinkle, Huong Ha, Jeffrey Chan
+- **abs:** https://arxiv.org/abs/2601.15640 · **PDF:** https://arxiv.org/pdf/2601.15640v1
+
+## 20053. Agentic AI Governance and Lifecycle Management in Healthcare
+
+- **ID:** `2601.15630` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Chandra Prakash, Mary Lind, Avneesh Sisodia
+- **abs:** https://arxiv.org/abs/2601.15630 · **PDF:** https://arxiv.org/pdf/2601.15630v1
+
+## 20054. CogToM: A Comprehensive Theory of Mind Benchmark inspired by Human Cognition for Large Language Models
+
+- **ID:** `2601.15628` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Haibo Tong, Zeyang Yue, Feifei Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.15628 · **PDF:** https://arxiv.org/pdf/2601.15628v1
+
+## 20055. Bridging Qualitative Rubrics and AI: A Binary Question Framework for Criterion-Referenced Grading in Engineering
+
+- **ID:** `2601.15626` · **primary:** `eess.SY` · **Submitted:** 2026-01-22
+- **Authors:** Lili Chen, Winn Wing-Yiu Chow, Stella Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.15626 · **PDF:** https://arxiv.org/pdf/2601.15626v1
+
+## 20056. Robust Tool Use via Fission-GRPO: Learning to Recover from Execution Errors
+
+- **ID:** `2601.15625` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Zhiwei Zhang, Fei Zhao, Rui Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.15625 · **PDF:** https://arxiv.org/pdf/2601.15625v1
+
+## 20057. Qwen3-TTS Technical Report
+
+- **ID:** `2601.15621` · **primary:** `cs.SD` · **Submitted:** 2026-01-22
+- **Authors:** Hangrui Hu, Xinfa Zhu, Ting He, et al.
+- **abs:** https://arxiv.org/abs/2601.15621 · **PDF:** https://arxiv.org/pdf/2601.15621v1
+
+## 20058. Closing the Gap on the Sample Complexity of 1-Identification
+
+- **ID:** `2601.15620` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Zitian Li, Wang Chi Cheung
+- **abs:** https://arxiv.org/abs/2601.15620 · **PDF:** https://arxiv.org/pdf/2601.15620v1
+
+## 20059. Embedding Compression via Spherical Coordinates
+
+- **ID:** `2602.00079` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Han Xiao
+- **abs:** https://arxiv.org/abs/2602.00079 · **PDF:** https://arxiv.org/pdf/2602.00079v4
+
+## 20060. When Sharpening Becomes Collapse: Sampling Bias and Semantic Coupling in RL with Verifiable Rewards
+
+- **ID:** `2601.15609` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Mingyuan Fan, Weiguang Han, Daixin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.15609 · **PDF:** https://arxiv.org/pdf/2601.15609v2
+
+## 20061. ToxiTwitch: Toward Emote-Aware Hybrid Moderation for Live Streaming Platforms
+
+- **ID:** `2601.15605` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Baktash Ansari, Elias Martin, Afra Mashhadi
+- **abs:** https://arxiv.org/abs/2601.15605 · **PDF:** https://arxiv.org/pdf/2601.15605v2
+
+## 20062. On the Nonasymptotic Scaling Guarantee of Hyperparameter Estimation in Inhomogeneous, Weakly-Dependent Complex Network Dynamical Systems
+
+- **ID:** `2601.15603` · **primary:** `math.ST` · **Submitted:** 2026-01-22
+- **Authors:** Yi Yu, Yubo Hou, Yinchong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.15603 · **PDF:** https://arxiv.org/pdf/2601.15603v1
+
+## 20063. Autonomous Business System via Neuro-symbolic AI
+
+- **ID:** `2601.15599` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Cecil Pang, Hiroki Sayama
+- **abs:** https://arxiv.org/abs/2601.15599 · **PDF:** https://arxiv.org/pdf/2601.15599v3
+
+## 20064. Ternary Spiking Neural Networks Enhanced by Complemented Neurons and Membrane Potential Aggregation
+
+- **ID:** `2601.15598` · **primary:** `cs.NE` · **Submitted:** 2026-01-22
+- **Authors:** Boxuan Zhang, Jiaxin Wang, Zhen Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.15598 · **PDF:** https://arxiv.org/pdf/2601.15598v1
+
+## 20065. Neural Nonlinear Shrinkage of Covariance Matrices for Minimum Variance Portfolio Optimization
+
+- **ID:** `2601.15597` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Liusha Yang, Siqi Zhao, Shuqi Chai
+- **abs:** https://arxiv.org/abs/2601.15597 · **PDF:** https://arxiv.org/pdf/2601.15597v1
+
+## 20066. DeepASMR: LLM-Based Zero-Shot ASMR Speech Generation for Anyone of Any Voice
+
+- **ID:** `2601.15596` · **primary:** `cs.SD` · **Submitted:** 2026-01-22
+- **Authors:** Leying Zhang, Tingxiao Zhou, Haiyang Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.15596 · **PDF:** https://arxiv.org/pdf/2601.15596v1
+
+## 20067. Data-Free Privacy-Preserving for LLMs via Model Inversion and Selective Unlearning
+
+- **ID:** `2601.15595` · **primary:** `cs.CR` · **Submitted:** 2026-01-22
+- **Authors:** Xinjie Zhou, Zhihui Yang, Lechao Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.15595 · **PDF:** https://arxiv.org/pdf/2601.15595v1
+
+## 20068. Parallelism and Generation Order in Masked Diffusion Language Models: Limits Today, Potential Tomorrow
+
+- **ID:** `2601.15593` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Yangyang Zhong, Yanmei Gu, Zhengqing Zang, et al.
+- **abs:** https://arxiv.org/abs/2601.15593 · **PDF:** https://arxiv.org/pdf/2601.15593v1
+
+## 20069. Deep Learning for Perishable Inventory Systems with Human Knowledge
+
+- **ID:** `2601.15589` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Xuan Liao, Zhenkang Peng, Ying Rong
+- **abs:** https://arxiv.org/abs/2601.15589 · **PDF:** https://arxiv.org/pdf/2601.15589v1
+
+## 20070. YuFeng-XGuard: A Reasoning-Centric, Interpretable, and Flexible Guardrail Model for Large Language Models
+
+- **ID:** `2601.15588` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Junyu Lin, Meizhen Liu, Xiufeng Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.15588 · **PDF:** https://arxiv.org/pdf/2601.15588v1
+
+## 20071. MapViT: A Two-Stage ViT-Based Framework for Real-Time Radio Quality Map Prediction in Dynamic Environments
+
+- **ID:** `2601.15578` · **primary:** `cs.NI` · **Submitted:** 2026-01-22
+- **Authors:** Cyril Shih-Huan Hsu, Xi Li, Lanfranco Zanzi, et al.
+- **abs:** https://arxiv.org/abs/2601.15578 · **PDF:** https://arxiv.org/pdf/2601.15578v1
+
+## 20072. PromptHelper: A Prompt Recommender System for Encouraging Creativity in AI Chatbot Interactions
+
+- **ID:** `2601.15575` · **primary:** `cs.HC` · **Submitted:** 2026-01-22
+- **Authors:** Jason Kim, Maria Teleki, James Caverlee
+- **abs:** https://arxiv.org/abs/2601.15575 · **PDF:** https://arxiv.org/pdf/2601.15575v1
+
+## 20073. High Convergence Rates of CMOS Invertible Logic Circuits Based on Many-Body Hamiltonians
+
+- **ID:** `2602.15033` · **primary:** `cs.ET` · **Submitted:** 2026-01-22
+- **Authors:** Naoya Onizawa, Takahiro Hanyu
+- **abs:** https://arxiv.org/abs/2602.15033 · **PDF:** https://arxiv.org/pdf/2602.15033v1
+
+## 20074. Enhanced Convergence in p-bit Based Simulated Annealing with Partial Deactivation for Large-Scale Combinatorial Optimization Problems
+
+- **ID:** `2601.15561` · **primary:** `cs.ET` · **Submitted:** 2026-01-22
+- **Authors:** Naoya Onizawa, Takahiro Hanyu
+- **abs:** https://arxiv.org/abs/2601.15561 · **PDF:** https://arxiv.org/pdf/2601.15561v1
+
+## 20075. From Generation to Collaboration: Using LLMs to Edit for Empathy in Healthcare
+
+- **ID:** `2601.15558` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Man Luo, Bahareh Harandizadeh, Amara Tariq, et al.
+- **abs:** https://arxiv.org/abs/2601.15558 · **PDF:** https://arxiv.org/pdf/2601.15558v1
+
+## 20076. LLM or Human? Perceptions of Trust and Information Quality in Research Summaries
+
+- **ID:** `2601.15556` · **primary:** `cs.CY` · **Submitted:** 2026-01-22
+- **Authors:** Nil-Jana Akpinar, Sandeep Avula, CJ Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.15556 · **PDF:** https://arxiv.org/pdf/2601.15556v1
+
+## 20077. BanditLP: Large-Scale Stochastic Optimization for Personalized Recommendations
+
+- **ID:** `2601.15552` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Phuc Nguyen, Benjamin Zelditch, Joyce Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.15552 · **PDF:** https://arxiv.org/pdf/2601.15552v1
+
+## 20078. ALIGNAgent: Adaptive Learner Intelligence for Gap Identification and Next-step guidance
+
+- **ID:** `2601.15551` · **primary:** `cs.AI` · **Submitted:** 2026-01-22
+- **Authors:** Bismack Tokoli, Luis Jaimes, Ayesha S. Dina
+- **abs:** https://arxiv.org/abs/2601.15551 · **PDF:** https://arxiv.org/pdf/2601.15551v1
+
+## 20079. Common to Whom? Regional Cultural Commonsense and LLM Bias in India
+
+- **ID:** `2601.15550` · **primary:** `cs.CL` · **Submitted:** 2026-01-22
+- **Authors:** Sangmitra Madhusudan, Trush Shashank More, Steph Buongiorno, et al.
+- **abs:** https://arxiv.org/abs/2601.15550 · **PDF:** https://arxiv.org/pdf/2601.15550v2
+
+## 20080. VIOLA: Towards Video In-Context Learning with Minimal Annotations
+
+- **ID:** `2601.15549` · **primary:** `cs.CV` · **Submitted:** 2026-01-22
+- **Authors:** Ryo Fujii, Hideo Saito, Ryo Hachiuma
+- **abs:** https://arxiv.org/abs/2601.15549 · **PDF:** https://arxiv.org/pdf/2601.15549v1
+
+## 20081. Learning Neural Operators from Partial Observations via Latent Autoregressive Modeling
+
+- **ID:** `2601.15547` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Jingren Hou, Hong Wang, Pengyu Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.15547 · **PDF:** https://arxiv.org/pdf/2601.15547v2
+
+## 20082. Beyond validation loss: Clinically-tailored optimization metrics improve a model's clinical performance
+
+- **ID:** `2601.15546` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Charles B. Delahunt, Courosh Mehanian, Daniel E. Shea, et al.
+- **abs:** https://arxiv.org/abs/2601.15546 · **PDF:** https://arxiv.org/pdf/2601.15546v1
+
+## 20083. RDumb++: Drift-Aware Continual Test-Time Adaptation
+
+- **ID:** `2601.15544` · **primary:** `cs.LG` · **Submitted:** 2026-01-22
+- **Authors:** Himanshu Mishra
+- **abs:** https://arxiv.org/abs/2601.15544 · **PDF:** https://arxiv.org/pdf/2601.15544v1
+
+## 20084. PRISM: Deriving a White-Box Transformer as a Signal-Noise Decomposition Operator via Maximum Coding Rate Reduction
+
+- **ID:** `2601.15540` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Dongchen Huang
+- **abs:** https://arxiv.org/abs/2601.15540 · **PDF:** https://arxiv.org/pdf/2601.15540v2
+
+## 20085. A Machine Vision Approach to Preliminary Skin Lesion Assessments
+
+- **ID:** `2601.15539` · **primary:** `eess.IV` · **Submitted:** 2026-01-21
+- **Authors:** Ali Khreis, Ro'Yah Radaideh, Quinn McGill
+- **abs:** https://arxiv.org/abs/2601.15539 · **PDF:** https://arxiv.org/pdf/2601.15539v1
+
+## 20086. QUAIL: Quantization Aware Unlearning for Mitigating Misinformation in LLMs
+
+- **ID:** `2601.15538` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Himanshu Mishra, Kanwal Mehreen
+- **abs:** https://arxiv.org/abs/2601.15538 · **PDF:** https://arxiv.org/pdf/2601.15538v1
+
+## 20087. Test-Time Adaptation for Speech Emotion Recognition
+
+- **ID:** `2601.16240` · **primary:** `eess.AS` · **Submitted:** 2026-01-21
+- **Authors:** Jiaheng Dong, Hong Jia, Ting Dang
+- **abs:** https://arxiv.org/abs/2601.16240 · **PDF:** https://arxiv.org/pdf/2601.16240v1
+
+## 20088. From Generative Engines to Actionable Simulators: The Imperative of Physical Grounding in World Models
+
+- **ID:** `2601.15533` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Zhikang Chen, Tingting Zhu
+- **abs:** https://arxiv.org/abs/2601.15533 · **PDF:** https://arxiv.org/pdf/2601.15533v1
+
+## 20089. Machine learning-enhanced non-amnestic Alzheimer's disease diagnosis from MRI and clinical features
+
+- **ID:** `2601.15530` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Megan A. Witherow, Michael L. Evans, Ahmed Temtam, et al.
+- **abs:** https://arxiv.org/abs/2601.15530 · **PDF:** https://arxiv.org/pdf/2601.15530v2
+
+## 20090. Code Mixologist : A Practitioner's Guide to Building Code-Mixed LLMs
+
+- **ID:** `2602.11181` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Himanshu Gupta, Pratik Jayarao, Chaitanya Dwivedi, et al.
+- **abs:** https://arxiv.org/abs/2602.11181 · **PDF:** https://arxiv.org/pdf/2602.11181v1
+
+## 20091. Language Model Representations for Efficient Few-Shot Tabular Classification
+
+- **ID:** `2602.15844` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Inwon Kang, Parikshit Ram, Yi Zhou, et al.
+- **abs:** https://arxiv.org/abs/2602.15844 · **PDF:** https://arxiv.org/pdf/2602.15844v1
+
+## 20092. TransportAgents: a multi-agents LLM framework for traffic accident severity prediction
+
+- **ID:** `2601.15519` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Zhichao Yang, Jiashu He, Jinxuan Fan, et al.
+- **abs:** https://arxiv.org/abs/2601.15519 · **PDF:** https://arxiv.org/pdf/2601.15519v2
+
+## 20093. DS@GT at TREC TOT 2025: Bridging Vague Recollection with Fusion Retrieval and Learned Reranking
+
+- **ID:** `2601.15518` · **primary:** `cs.IR` · **Submitted:** 2026-01-21
+- **Authors:** Wenxin Zhou, Ritesh Mehta, Anthony Miyaguchi
+- **abs:** https://arxiv.org/abs/2601.15518 · **PDF:** https://arxiv.org/pdf/2601.15518v2
+
+## 20094. DeltaDorsal: Enhancing Hand Pose Estimation with Dorsal Features in Egocentric Views
+
+- **ID:** `2601.15516` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** William Huang, Siyou Pei, Leyi Zou, et al.
+- **abs:** https://arxiv.org/abs/2601.15516 · **PDF:** https://arxiv.org/pdf/2601.15516v2
+
+## 20095. Assessing the informative value of macroeconomic indicators for public health forecasting
+
+- **ID:** `2601.15514` · **primary:** `stat.AP` · **Submitted:** 2026-01-21
+- **Authors:** Shome Chakraborty, Fardil Khan, Soutik Ghosal
+- **abs:** https://arxiv.org/abs/2601.15514 · **PDF:** https://arxiv.org/pdf/2601.15514v1
+
+## 20096. AdversaRiskQA: An Adversarial Factuality Benchmark for High-Risk Domains
+
+- **ID:** `2601.15511` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Adam Szelestey, Sofie van Engelen, Tianhao Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.15511 · **PDF:** https://arxiv.org/pdf/2601.15511v1
+
+## 20097. The Dark Side of AI Transformers: Sentiment Polarization & the Loss of Business Neutrality by NLP Transformers
+
+- **ID:** `2601.15509` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Prasanna Kumar
+- **abs:** https://arxiv.org/abs/2601.15509 · **PDF:** https://arxiv.org/pdf/2601.15509v1
+
+## 20098. Computational Representations of Character Significance in Novels
+
+- **ID:** `2601.15508` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Haaris Mian, Melanie Subbiah, Sharon Marcus, et al.
+- **abs:** https://arxiv.org/abs/2601.15508 · **PDF:** https://arxiv.org/pdf/2601.15508v1
+
+## 20099. ViT Registers and Fractal ViT
+
+- **ID:** `2601.15506` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Jason Chuan-Chih Chou, Abhinav Kumar, Shivank Garg
+- **abs:** https://arxiv.org/abs/2601.15506 · **PDF:** https://arxiv.org/pdf/2601.15506v1
+
+## 20100. SAGE-FM: A lightweight and interpretable spatial transcriptomics foundation model
+
+- **ID:** `2601.15504` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Xianghao Zhan, Jingyu Xu, Yuanning Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.15504 · **PDF:** https://arxiv.org/pdf/2601.15504v1
+
+## 20101. Data-driven Lake Water Quality Forecasting for Time Series with Missing Data using Machine Learning
+
+- **ID:** `2601.15503` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Rishit Chatterjee, Tahiya Chowdhury
+- **abs:** https://arxiv.org/abs/2601.15503 · **PDF:** https://arxiv.org/pdf/2601.15503v1
+
+## 20102. Low-Dimensional Adaptation of Rectified Flow: A Diffusion and Stochastic Localization Perspective
+
+- **ID:** `2601.15500` · **primary:** `stat.ML` · **Submitted:** 2026-01-21
+- **Authors:** Saptarshi Roy, Alessandro Rinaldo, Purnamrita Sarkar
+- **abs:** https://arxiv.org/abs/2601.15500 · **PDF:** https://arxiv.org/pdf/2601.15500v3
+
+## 20103. MARS: Unleashing the Power of Speculative Decoding via Margin-Aware Verification
+
+- **ID:** `2601.15498` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Jingwei Song, Xinyu Wang, Hanbin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.15498 · **PDF:** https://arxiv.org/pdf/2601.15498v1
+
+## 20104. Tracking the Limits of Knowledge Propagation: How LLMs Fail at Multi-Step Reasoning with Conflicting Knowledge
+
+- **ID:** `2601.15495` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Yiyang Feng, Zeming Chen, Haotian Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.15495 · **PDF:** https://arxiv.org/pdf/2601.15495v1
+
+## 20105. Multi-Persona Thinking for Bias Mitigation in Large Language Models
+
+- **ID:** `2601.15488` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Yuxing Chen, Guoqing Luo, Zijun Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.15488 · **PDF:** https://arxiv.org/pdf/2601.15488v1
+
+## 20106. MiRAGE: A Multiagent Framework for Generating Multimodal Multihop Question-Answer Dataset for RAG Evaluation
+
+- **ID:** `2601.15487` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Chandan Kumar Sahu, Premith Kumar Chilukuri, Matthew Hetrich
+- **abs:** https://arxiv.org/abs/2601.15487 · **PDF:** https://arxiv.org/pdf/2601.15487v1
+
+## 20107. The Rise of Large Language Models and the Direction and Impact of US Federal Research Funding
+
+- **ID:** `2601.15485` · **primary:** `cs.DL` · **Submitted:** 2026-01-21
+- **Authors:** Yifan Qian, Zhe Wen, Alexander C. Furnas, et al.
+- **abs:** https://arxiv.org/abs/2601.15485 · **PDF:** https://arxiv.org/pdf/2601.15485v1
+
+## 20108. Is Grokipedia Right-Leaning? Comparing Political Framing in Wikipedia and Grokipedia on Controversial Topics
+
+- **ID:** `2601.15484` · **primary:** `cs.IR` · **Submitted:** 2026-01-21
+- **Authors:** Philipp Eibl, Erica Coppolillo, Simone Mungari, et al.
+- **abs:** https://arxiv.org/abs/2601.15484 · **PDF:** https://arxiv.org/pdf/2601.15484v1
+
+## 20109. Martingale Foresight Sampling: A Principled Approach to Inference-Time LLM Decoding
+
+- **ID:** `2601.15482` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Huayu Li, ZhengXiao He, Siyuan Tian, et al.
+- **abs:** https://arxiv.org/abs/2601.15482 · **PDF:** https://arxiv.org/pdf/2601.15482v1
+
+## 20110. Early predicting of hospital admission using machine learning algorithms: Priority queues approach
+
+- **ID:** `2601.15481` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Jakub Antczak, James Montgomery, Małgorzata O'Reilly, et al.
+- **abs:** https://arxiv.org/abs/2601.15481 · **PDF:** https://arxiv.org/pdf/2601.15481v1
+
+## 20111. Benchmarking LLMs for Pairwise Causal Discovery in Biomedical and Multi-Domain Contexts
+
+- **ID:** `2601.15479` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Sydney Anuyah, Sneha Shajee-Mohan, Ankit-Singh Chauhan, et al.
+- **abs:** https://arxiv.org/abs/2601.15479 · **PDF:** https://arxiv.org/pdf/2601.15479v1
+
+## 20112. Reliability by design: quantifying and eliminating fabrication risk in LLMs. From generative to consultative AI: a comparative analysis in the legal domain and lessons for high-stakes knowledge bases
+
+- **ID:** `2601.15476` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Alex Dantart
+- **abs:** https://arxiv.org/abs/2601.15476 · **PDF:** https://arxiv.org/pdf/2601.15476v1
+
+## 20113. Multi-Targeted Graph Backdoor Attack
+
+- **ID:** `2601.15474` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Md Nabi Newaz Khan, Abdullah Arafat Miah, Yu Bi
+- **abs:** https://arxiv.org/abs/2601.15474 · **PDF:** https://arxiv.org/pdf/2601.15474v1
+
+## 20114. Panther: Faster and Cheaper Computations with Randomized Numerical Linear Algebra
+
+- **ID:** `2601.15473` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Fahd Seddik, Abdulrahman Elbedewy, Gaser Sami, et al.
+- **abs:** https://arxiv.org/abs/2601.15473 · **PDF:** https://arxiv.org/pdf/2601.15473v1
+
+## 20115. Large Language Models: A Mathematical Formulation
+
+- **ID:** `2601.22170` · **primary:** `math.NA` · **Submitted:** 2026-01-21
+- **Authors:** Ricardo Baptista, Andrew Stuart, Son Tran
+- **abs:** https://arxiv.org/abs/2601.22170 · **PDF:** https://arxiv.org/pdf/2601.22170v1
+
+## 20116. Learning from Synthetic Data: Limitations of ERM
+
+- **ID:** `2601.15468` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Kareem Amin, Alex Bie, Weiwei Kong, et al.
+- **abs:** https://arxiv.org/abs/2601.15468 · **PDF:** https://arxiv.org/pdf/2601.15468v2
+
+## 20117. Chunking, Retrieval, and Re-ranking: An Empirical Evaluation of RAG Architectures for Policy Document Question Answering
+
+- **ID:** `2601.15457` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Anuj Maharjan, Umesh Yadav
+- **abs:** https://arxiv.org/abs/2601.15457 · **PDF:** https://arxiv.org/pdf/2601.15457v1
+
+## 20118. Reflexis: Supporting Reflexivity and Rigor in Collaborative Qualitative Analysis through Design for Deliberation
+
+- **ID:** `2601.15445` · **primary:** `cs.HC` · **Submitted:** 2026-01-21
+- **Authors:** Runlong Ye, Oliver Huang, Patrick Yung Kang Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.15445 · **PDF:** https://arxiv.org/pdf/2601.15445v2
+
+## 20119. A tensor network formalism for neuro-symbolic AI
+
+- **ID:** `2601.15442` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Alex Goessmann, Janina Schütte, Maximilian Fröhlich, et al.
+- **abs:** https://arxiv.org/abs/2601.15442 · **PDF:** https://arxiv.org/pdf/2601.15442v1
+
+## 20120. CASL: Concept-Aligned Sparse Latents for Interpreting Diffusion Models
+
+- **ID:** `2601.15441` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Zhenghao He, Guangzhi Xiong, Boyang Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.15441 · **PDF:** https://arxiv.org/pdf/2601.15441v1
+
+## 20121. Ego4OOD: Rethinking Egocentric Video Domain Generalization via Covariate Shift Scoring
+
+- **ID:** `2601.17056` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Zahra Vaseqi, James Clark
+- **abs:** https://arxiv.org/abs/2601.17056 · **PDF:** https://arxiv.org/pdf/2601.17056v1
+
+## 20122. Not Your Typical Sycophant: The Elusive Nature of Sycophancy in Large Language Models
+
+- **ID:** `2601.15436` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Shahar Ben Natan, Oren Tsur
+- **abs:** https://arxiv.org/abs/2601.15436 · **PDF:** https://arxiv.org/pdf/2601.15436v2
+
+## 20123. Domain-Specific Knowledge Graphs in RAG-Enhanced Healthcare LLMs
+
+- **ID:** `2601.15429` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Sydney Anuyah, Mehedi Mahmud Kaushik, Hao Dai, et al.
+- **abs:** https://arxiv.org/abs/2601.15429 · **PDF:** https://arxiv.org/pdf/2601.15429v1
+
+## 20124. Lattice: A Confidence-Gated Hybrid System for Uncertainty-Aware Sequential Prediction with Behavioral Archetypes
+
+- **ID:** `2601.15423` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Lorian Bannis
+- **abs:** https://arxiv.org/abs/2601.15423 · **PDF:** https://arxiv.org/pdf/2601.15423v1
+
+## 20125. Ambient Dataloops: Generative Models for Dataset Refinement
+
+- **ID:** `2601.15417` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Adrián Rodríguez-Muñoz, William Daspit, Adam Klivans, et al.
+- **abs:** https://arxiv.org/abs/2601.15417 · **PDF:** https://arxiv.org/pdf/2601.15417v1
+
+## 20126. VibeTensor: System Software for Deep Learning, Fully Generated by AI Agents
+
+- **ID:** `2601.16238` · **primary:** `cs.SE` · **Submitted:** 2026-01-21
+- **Authors:** Bing Xu, Terry Chen, Fengzhe Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.16238 · **PDF:** https://arxiv.org/pdf/2601.16238v1
+
+## 20127. DuFal: Dual-Frequency-Aware Learning for High-Fidelity Extremely Sparse-view CBCT Reconstruction
+
+- **ID:** `2601.15416` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Cuong Tran Van, Trong-Thang Pham, Ngoc-Son Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.15416 · **PDF:** https://arxiv.org/pdf/2601.15416v1
+
+## 20128. A Checklist for Trustworthy, Safe, and User-Friendly Mental Health Chatbots
+
+- **ID:** `2601.15412` · **primary:** `cs.HC` · **Submitted:** 2026-01-21
+- **Authors:** Shreya Haran, Samiha Thatikonda, Dong Whi Yoo, et al.
+- **abs:** https://arxiv.org/abs/2601.15412 · **PDF:** https://arxiv.org/pdf/2601.15412v1
+
+## 20129. CURE: Curriculum-guided Multi-task Training for Reliable Anatomy Grounded Report Generation
+
+- **ID:** `2601.15408` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Pablo Messina, Andrés Villa, Juan León Alcázar, et al.
+- **abs:** https://arxiv.org/abs/2601.15408 · **PDF:** https://arxiv.org/pdf/2601.15408v1
+
+## 20130. Attention-Informed Surrogates for Navigating Power-Performance Trade-offs in HPC
+
+- **ID:** `2601.15399` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Ashna Nawar Ahmed, Banooqa Banday, Terry Jones, et al.
+- **abs:** https://arxiv.org/abs/2601.15399 · **PDF:** https://arxiv.org/pdf/2601.15399v1
+
+## 20131. Beyond Prompting: Efficient and Robust Contextual Biasing for Speech LLMs via Logit-Space Integration (LOGIC)
+
+- **ID:** `2601.15397` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Peidong Wang
+- **abs:** https://arxiv.org/abs/2601.15397 · **PDF:** https://arxiv.org/pdf/2601.15397v2
+
+## 20132. Beyond Fixed Psychological Personas: State Beats Trait, but Language Models are State-Blind
+
+- **ID:** `2601.15395` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Tamunotonye Harry, Ivoline Ngong, Chima Nweke, et al.
+- **abs:** https://arxiv.org/abs/2601.15395 · **PDF:** https://arxiv.org/pdf/2601.15395v1
+
+## 20133. Memorization Dynamics in Knowledge Distillation for Language Models
+
+- **ID:** `2601.15394` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Jaydeep Borkar, Karan Chadha, Niloofar Mireshghallah, et al.
+- **abs:** https://arxiv.org/abs/2601.15394 · **PDF:** https://arxiv.org/pdf/2601.15394v1
+
+## 20134. GeMM-GAN: A Multimodal Generative Model Conditioned on Histopathology Images and Clinical Descriptions for Gene Expression Profile Generation
+
+- **ID:** `2601.15392` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Francesca Pia Panaccione, Carlo Sgaravatti, Pietro Pinoli
+- **abs:** https://arxiv.org/abs/2601.15392 · **PDF:** https://arxiv.org/pdf/2601.15392v1
+
+## 20135. FedUMM: A General Framework for Federated Learning with Unified Multimodal Models
+
+- **ID:** `2601.15390` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Zhaolong Su, Leheng Zhao, Xiaoying Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.15390 · **PDF:** https://arxiv.org/pdf/2601.15390v1
+
+## 20136. VegaChat: A Robust Framework for LLM-Based Chart Generation and Assessment
+
+- **ID:** `2601.15385` · **primary:** `cs.HC` · **Submitted:** 2026-01-21
+- **Authors:** Marko Hostnik, Rauf Kurbanov, Yaroslav Sokolov, et al.
+- **abs:** https://arxiv.org/abs/2601.15385 · **PDF:** https://arxiv.org/pdf/2601.15385v1
+
+## 20137. You Need Better Attention Priors
+
+- **ID:** `2601.15380` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Elon Litman, Gabe Guo
+- **abs:** https://arxiv.org/abs/2601.15380 · **PDF:** https://arxiv.org/pdf/2601.15380v1
+
+## 20138. Iterative Refinement Improves Compositional Image Generation
+
+- **ID:** `2601.15286` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Shantanu Jaiswal, Mihir Prabhudesai, Nikash Bhardwaj, et al.
+- **abs:** https://arxiv.org/abs/2601.15286 · **PDF:** https://arxiv.org/pdf/2601.15286v1
+
+## 20139. Rethinking Video Generation Model for the Embodied World
+
+- **ID:** `2601.15282` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Yufan Deng, Zilin Pan, Hongyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.15282 · **PDF:** https://arxiv.org/pdf/2601.15282v1
+
+## 20140. LLM-based Multimodal Feedback Produces Equivalent Learning and Better Student Perceptions than Educator Feedback
+
+- **ID:** `2601.15280` · **primary:** `cs.HC` · **Submitted:** 2026-01-21
+- **Authors:** Chloe Qianhui Zhao, Jie Cao, Jionghao Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.15280 · **PDF:** https://arxiv.org/pdf/2601.15280v1
+
+## 20141. MolecularIQ: Characterizing Chemical Reasoning Capabilities Through Symbolic Verification on Molecular Graphs
+
+- **ID:** `2601.15279` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Christoph Bartmann, Johannes Schimunek, Mykyta Ielanskyi, et al.
+- **abs:** https://arxiv.org/abs/2601.15279 · **PDF:** https://arxiv.org/pdf/2601.15279v1
+
+## 20142. Robust Fake News Detection using Large Language Models under Adversarial Sentiment Attacks
+
+- **ID:** `2601.15277` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Sahar Tahmasebi, Eric Müller-Budack, Ralph Ewerth
+- **abs:** https://arxiv.org/abs/2601.15277 · **PDF:** https://arxiv.org/pdf/2601.15277v1
+
+## 20143. RayRoPE: Projective Ray Positional Encoding for Multi-view Attention
+
+- **ID:** `2601.15275` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Yu Wu, Minsik Jeon, Jen-Hao Rick Chang, et al.
+- **abs:** https://arxiv.org/abs/2601.15275 · **PDF:** https://arxiv.org/pdf/2601.15275v2
+
+## 20144. Improving MoE Compute Efficiency by Composing Weight and Data Sparsity
+
+- **ID:** `2601.15370` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Maciej Kilian, Oleg Mkrtchyan, Luke Zettlemoyer, et al.
+- **abs:** https://arxiv.org/abs/2601.15370 · **PDF:** https://arxiv.org/pdf/2601.15370v1
+
+## 20145. Evaluation of Large Language Models in Legal Applications: Challenges, Methods, and Future Directions
+
+- **ID:** `2601.15267` · **primary:** `cs.CY` · **Submitted:** 2026-01-21
+- **Authors:** Yiran Hu, Huanghai Liu, Chong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.15267 · **PDF:** https://arxiv.org/pdf/2601.15267v1
+
+## 20146. OpenVision 3: A Family of Unified Visual Encoder for Both Understanding and Generation
+
+- **ID:** `2601.15369` · **primary:** `eess.IV` · **Submitted:** 2026-01-21
+- **Authors:** Letian Zhang, Sucheng Ren, Yanqing Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.15369 · **PDF:** https://arxiv.org/pdf/2601.15369v2
+
+## 20147. Many Experiments, Few Repetitions, Unpaired Data, and Sparse Effects: Is Causal Inference Possible?
+
+- **ID:** `2601.15254` · **primary:** `stat.ML` · **Submitted:** 2026-01-21
+- **Authors:** Felix Schur, Niklas Pfister, Peng Ding, et al.
+- **abs:** https://arxiv.org/abs/2601.15254 · **PDF:** https://arxiv.org/pdf/2601.15254v1
+
+## 20148. The Effect of Scripts and Formats on LLM Numeracy
+
+- **ID:** `2601.15251` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Varshini Reddy, Craig W. Schmidt, Seth Ebner, et al.
+- **abs:** https://arxiv.org/abs/2601.15251 · **PDF:** https://arxiv.org/pdf/2601.15251v1
+
+## 20149. Recommending Best Paper Awards for ML/AI Conferences via the Isotonic Mechanism
+
+- **ID:** `2601.15249` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Garrett G. Wen, Buxin Su, Natalie Collina, et al.
+- **abs:** https://arxiv.org/abs/2601.15249 · **PDF:** https://arxiv.org/pdf/2601.15249v2
+
+## 20150. Taxonomy-Aligned Risk Extraction from 10-K Filings with Autonomous Improvement Using LLMs
+
+- **ID:** `2601.15247` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Rian Dolphin, Joe Dursun, Jarrett Blankenship, et al.
+- **abs:** https://arxiv.org/abs/2601.15247 · **PDF:** https://arxiv.org/pdf/2601.15247v1
+
+## 20151. Feasibility Preservation under Monotone Retrieval Truncation
+
+- **ID:** `2601.15241` · **primary:** `cs.LO` · **Submitted:** 2026-01-21
+- **Authors:** Sean Plummer
+- **abs:** https://arxiv.org/abs/2601.15241 · **PDF:** https://arxiv.org/pdf/2601.15241v1
+
+## 20152. Multi-context principal component analysis
+
+- **ID:** `2601.15239` · **primary:** `stat.ML` · **Submitted:** 2026-01-21
+- **Authors:** Kexin Wang, Salil Bhate, João M. Pereira, et al.
+- **abs:** https://arxiv.org/abs/2601.15239 · **PDF:** https://arxiv.org/pdf/2601.15239v1
+
+## 20153. Computational Foundations for Strategic Coopetition: Formalizing Collective Action and Loyalty
+
+- **ID:** `2601.16237` · **primary:** `cs.MA` · **Submitted:** 2026-01-21
+- **Authors:** Vik Pant, Eric Yu
+- **abs:** https://arxiv.org/abs/2601.16237 · **PDF:** https://arxiv.org/pdf/2601.16237v1
+
+## 20154. Metadata Conditioned Large Language Models for Localization
+
+- **ID:** `2601.15236` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Anjishnu Mukherjee, Ziwei Zhu, Antonios Anastasopoulos
+- **abs:** https://arxiv.org/abs/2601.15236 · **PDF:** https://arxiv.org/pdf/2601.15236v1
+
+## 20155. Tracing 3D Anatomy in 2D Strokes: A Multi-Stage Projection Driven Approach to Cervical Spine Fracture Identification
+
+- **ID:** `2601.15235` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Fabi Nahian Madhurja, Rusab Sarmun, Muhammad E. H. Chowdhury, et al.
+- **abs:** https://arxiv.org/abs/2601.15235 · **PDF:** https://arxiv.org/pdf/2601.15235v3
+
+## 20156. The Perplexity Paradox: Why Code Compresses Better Than Math in LLM Prompts
+
+- **ID:** `2602.15843` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Warren Johnson
+- **abs:** https://arxiv.org/abs/2602.15843 · **PDF:** https://arxiv.org/pdf/2602.15843v1
+
+## 20157. PROGRESSLM: Towards Progress Reasoning in Vision-Language Models
+
+- **ID:** `2601.15224` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Jianshu Zhang, Chengxuan Qian, Haosen Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.15224 · **PDF:** https://arxiv.org/pdf/2601.15224v1
+
+## 20158. Privacy Collapse: Benign Fine-Tuning Can Break Contextual Privacy in Language Models
+
+- **ID:** `2601.15220` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Anmol Goel, Cornelius Emde, Sangdoo Yun, et al.
+- **abs:** https://arxiv.org/abs/2601.15220 · **PDF:** https://arxiv.org/pdf/2601.15220v1
+
+## 20159. ZENITH: Automated Gradient Norm Informed Stochastic Optimization
+
+- **ID:** `2601.15212` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Dhrubo Saha
+- **abs:** https://arxiv.org/abs/2601.15212 · **PDF:** https://arxiv.org/pdf/2601.15212v1
+
+## 20160. Deaf and Hard of Hearing Access to Intelligent Personal Assistants: Comparison of Voice-Based Options with an LLM-Powered Touch Interface
+
+- **ID:** `2601.15209` · **primary:** `cs.HC` · **Submitted:** 2026-01-21
+- **Authors:** Paige S. DeVries, Michaela Okosi, Ming Li, et al.
+- **abs:** https://arxiv.org/abs/2601.15209 · **PDF:** https://arxiv.org/pdf/2601.15209v2
+
+## 20161. Lost Before Translation: Social Information Transmission and Survival in AI-AI Communication
+
+- **ID:** `2602.17674` · **primary:** `cs.HC` · **Submitted:** 2026-01-21
+- **Authors:** Bijean Ghafouri, Emilio Ferrara
+- **abs:** https://arxiv.org/abs/2602.17674 · **PDF:** https://arxiv.org/pdf/2602.17674v1
+
+## 20162. LangForce: Bayesian Decomposition of Vision Language Action Models via Latent Action Queries
+
+- **ID:** `2601.15197` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Shijie Lian, Bin Yu, Xiaopeng Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.15197 · **PDF:** https://arxiv.org/pdf/2601.15197v5
+
+## 20163. Where Do AI Coding Agents Fail? An Empirical Study of Failed Agentic Pull Requests in GitHub
+
+- **ID:** `2601.15195` · **primary:** `cs.SE` · **Submitted:** 2026-01-21
+- **Authors:** Ramtin Ehsani, Sakshi Pathak, Shriya Rawal, et al.
+- **abs:** https://arxiv.org/abs/2601.15195 · **PDF:** https://arxiv.org/pdf/2601.15195v1
+
+## 20164. Benchmarking Large Language Models for ABAP Code Generation: An Empirical Study on Iterative Improvement by Compiler Feedback
+
+- **ID:** `2601.15188` · **primary:** `cs.SE` · **Submitted:** 2026-01-21
+- **Authors:** Stephan Wallraven, Tim Köhne, Hartmut Westenberger, et al.
+- **abs:** https://arxiv.org/abs/2601.15188 · **PDF:** https://arxiv.org/pdf/2601.15188v1
+
+## 20165. Supporting Humans in Evaluating AI Summaries of Legal Depositions
+
+- **ID:** `2601.15182` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Naghmeh Farzi, Laura Dietz, Dave D. Lewis
+- **abs:** https://arxiv.org/abs/2601.15182 · **PDF:** https://arxiv.org/pdf/2601.15182v1
+
+## 20166. Dynamic Management of a Deep Learning-Based Anomaly Detection System for 5G Networks
+
+- **ID:** `2601.15177` · **primary:** `cs.CR` · **Submitted:** 2026-01-21
+- **Authors:** Lorenzo Fernández Maimó, Alberto Huertas Celdrán, Manuel Gil Pérez, et al.
+- **abs:** https://arxiv.org/abs/2601.15177 · **PDF:** https://arxiv.org/pdf/2601.15177v1
+
+## 20167. Is Peer Review Really in Decline? Analyzing Review Quality across Venues and Time
+
+- **ID:** `2601.15172` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Ilia Kuznetsov, Rohan Nayak, Alla Rozovskaya, et al.
+- **abs:** https://arxiv.org/abs/2601.15172 · **PDF:** https://arxiv.org/pdf/2601.15172v1
+
+## 20168. The Flexibility Trap: Why Arbitrary Order Limits Reasoning Potential in Diffusion Language Models
+
+- **ID:** `2601.15165` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Zanlin Ni, Shenzhi Wang, Yang Yue, et al.
+- **abs:** https://arxiv.org/abs/2601.15165 · **PDF:** https://arxiv.org/pdf/2601.15165v3
+
+## 20169. V-CAGE: Context-Aware Generation and Verification for Scalable Long-Horizon Embodied Tasks
+
+- **ID:** `2601.15164` · **primary:** `cs.RO` · **Submitted:** 2026-01-21
+- **Authors:** Yaru Liu, Ao-bo Wang, Nanyang Ye
+- **abs:** https://arxiv.org/abs/2601.15164 · **PDF:** https://arxiv.org/pdf/2601.15164v1
+
+## 20170. Automated Rubrics for Reliable Evaluation of Medical Dialogue Systems
+
+- **ID:** `2601.15161` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Yinzhu Chen, Abdine Maiga, Hossein A. Rahmani, et al.
+- **abs:** https://arxiv.org/abs/2601.15161 · **PDF:** https://arxiv.org/pdf/2601.15161v1
+
+## 20171. Knowledge Graphs are Implicit Reward Models: Path-Derived Signals Enable Compositional Reasoning
+
+- **ID:** `2601.15160` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Yuval Kansal, Niraj K. Jha
+- **abs:** https://arxiv.org/abs/2601.15160 · **PDF:** https://arxiv.org/pdf/2601.15160v3
+
+## 20172. Outcome-Based RL Provably Leads Transformers to Reason, but Only With the Right Data
+
+- **ID:** `2601.15158` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Yuval Ran-Milo, Yotam Alexander, Shahar Mendel, et al.
+- **abs:** https://arxiv.org/abs/2601.15158 · **PDF:** https://arxiv.org/pdf/2601.15158v3
+
+## 20173. How to Build AI Agents by Augmenting LLMs with Codified Human Expert Domain Knowledge? A Software Engineering Framework
+
+- **ID:** `2601.15153` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Choro Ulan uulu, Mikhail Kulyabin, Iris Fuhrmann, et al.
+- **abs:** https://arxiv.org/abs/2601.15153 · **PDF:** https://arxiv.org/pdf/2601.15153v1
+
+## 20174. CLEANER: Self-Purified Trajectories Boost Agentic Reinforcement Learning
+
+- **ID:** `2601.15141` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Tianshi Xu, Yuteng Chen, Meng Li
+- **abs:** https://arxiv.org/abs/2601.15141 · **PDF:** https://arxiv.org/pdf/2601.15141v1
+
+## 20175. Graph Recognition via Subgraph Prediction
+
+- **ID:** `2601.15133` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** André Eberhard, Gerhard Neumann, Pascal Friederich
+- **abs:** https://arxiv.org/abs/2601.15133 · **PDF:** https://arxiv.org/pdf/2601.15133v2
+
+## 20176. Vehicle Routing with Finite Time Horizon using Deep Reinforcement Learning with Improved Network Embedding
+
+- **ID:** `2601.15131` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Ayan Maity, Sudeshna Sarkar
+- **abs:** https://arxiv.org/abs/2601.15131 · **PDF:** https://arxiv.org/pdf/2601.15131v1
+
+## 20177. The Plausibility Trap: Using Probabilistic Engines for Deterministic Tasks
+
+- **ID:** `2601.15130` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Ivan Carrera, Daniel Maldonado-Ruiz
+- **abs:** https://arxiv.org/abs/2601.15130 · **PDF:** https://arxiv.org/pdf/2601.15130v1
+
+## 20178. RSNA Large Language Model Benchmark Dataset for Chest Radiographs of Cardiothoracic Disease: Radiologist Evaluation and Validation Enhanced by AI Labels (REVEAL-CXR)
+
+- **ID:** `2601.15129` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Yishu Wei, Adam E. Flanders, Errol Colak, et al.
+- **abs:** https://arxiv.org/abs/2601.15129 · **PDF:** https://arxiv.org/pdf/2601.15129v1
+
+## 20179. Predictor-Free and Hardware-Aware Federated Neural Architecture Search via Pareto-Guided Supernet Training
+
+- **ID:** `2601.15127` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Bostan Khan, Masoud Daneshtalab
+- **abs:** https://arxiv.org/abs/2601.15127 · **PDF:** https://arxiv.org/pdf/2601.15127v2
+
+## 20180. RAG-GFM: Overcoming In-Memory Bottlenecks in Graph Foundation Models via Retrieval-Augmented Generation
+
+- **ID:** `2601.15124` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Haonan Yuan, Qingyun Sun, Jiacheng Tao, et al.
+- **abs:** https://arxiv.org/abs/2601.15124 · **PDF:** https://arxiv.org/pdf/2601.15124v2
+
+## 20181. BREPS: Bounding-Box Robustness Evaluation of Promptable Segmentation
+
+- **ID:** `2601.15123` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Andrey Moskalenko, Danil Kuznetsov, Irina Dudko, et al.
+- **abs:** https://arxiv.org/abs/2601.15123 · **PDF:** https://arxiv.org/pdf/2601.15123v1
+
+## 20182. Emerging from Ground: Addressing Intent Deviation in Tool-Using Agents via Deriving Real Calls into Virtual Trajectories
+
+- **ID:** `2601.15120` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Qian Xiong, Yuekai Huang, Bo Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.15120 · **PDF:** https://arxiv.org/pdf/2601.15120v2
+
+## 20183. WavLink: Compact Audio-Text Embeddings with a Global Whisper Token
+
+- **ID:** `2601.15118` · **primary:** `cs.SD` · **Submitted:** 2026-01-21
+- **Authors:** Gokul Karthik Kumar, Ludovick Lepauloux, Hakim Hacid
+- **abs:** https://arxiv.org/abs/2601.15118 · **PDF:** https://arxiv.org/pdf/2601.15118v2
+
+## 20184. Auditing Language Model Unlearning via Information Decomposition
+
+- **ID:** `2601.15111` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Anmol Goel, Alan Ritter, Iryna Gurevych
+- **abs:** https://arxiv.org/abs/2601.15111 · **PDF:** https://arxiv.org/pdf/2601.15111v1
+
+## 20185. An Agentic Operationalization of DISARM for FIMI Investigation on Social Media
+
+- **ID:** `2601.15109` · **primary:** `cs.SI` · **Submitted:** 2026-01-21
+- **Authors:** Kevin Tseng, Juan Carlos Toledano, Bart De Clerck, et al.
+- **abs:** https://arxiv.org/abs/2601.15109 · **PDF:** https://arxiv.org/pdf/2601.15109v3
+
+## 20186. One scale to rule them all: interpretable multi-scale Deep Learning for predicting cell survival after proton and carbon ion irradiation
+
+- **ID:** `2601.15106` · **primary:** `physics.bio-ph` · **Submitted:** 2026-01-21
+- **Authors:** Giulio Bordieri, Giorgio Cartechini, Anna Bianchi, et al.
+- **abs:** https://arxiv.org/abs/2601.15106 · **PDF:** https://arxiv.org/pdf/2601.15106v1
+
+## 20187. Field-Space Autoencoder for Scalable Climate Emulators
+
+- **ID:** `2601.15102` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Johannes Meuer, Maximilian Witte, Étiénne Plésiat, et al.
+- **abs:** https://arxiv.org/abs/2601.15102 · **PDF:** https://arxiv.org/pdf/2601.15102v1
+
+## 20188. Joint single-shot ToA and DoA estimation for VAA-based BLE ranging with phase ambiguity: A deep learning-based approach
+
+- **ID:** `2602.02503` · **primary:** `eess.SP` · **Submitted:** 2026-01-21
+- **Authors:** Jincheng Xie, Yili Deng, Jiguang He, et al.
+- **abs:** https://arxiv.org/abs/2602.02503 · **PDF:** https://arxiv.org/pdf/2602.02503v1
+
+## 20189. Circadian Modulation of Semantic Exploration in Social Media Language
+
+- **ID:** `2601.15091` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Vuong Hung Truong, Mariana Gabrielle Cangco Reyes, Masatoshi Koizumi, et al.
+- **abs:** https://arxiv.org/abs/2601.15091 · **PDF:** https://arxiv.org/pdf/2601.15091v1
+
+## 20190. Memory Retention Is Not Enough to Master Memory Tasks in Reinforcement Learning
+
+- **ID:** `2601.15086` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Oleg Shchendrigin, Egor Cherepanov, Alexey K. Kovalev, et al.
+- **abs:** https://arxiv.org/abs/2601.15086 · **PDF:** https://arxiv.org/pdf/2601.15086v1
+
+## 20191. Bangla Music Genre Classification Using Bidirectional LSTMS
+
+- **ID:** `2601.15083` · **primary:** `cs.SD` · **Submitted:** 2026-01-21
+- **Authors:** Muntakimur Rahaman, Md Mahmudul Hoque, Md Mehedi Hassain
+- **abs:** https://arxiv.org/abs/2601.15083 · **PDF:** https://arxiv.org/pdf/2601.15083v1
+
+## 20192. LoRAP: Low-Rank Aggregation Prompting for Quantized Graph Neural Networks Training
+
+- **ID:** `2601.15079` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Chenyu Liu, Haige Li, Luca Rossi
+- **abs:** https://arxiv.org/abs/2601.15079 · **PDF:** https://arxiv.org/pdf/2601.15079v1
+
+## 20193. Multi-Agent Constraint Factorization Reveals Latent Invariant Solution Structure
+
+- **ID:** `2601.15077` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Christopher Scofield
+- **abs:** https://arxiv.org/abs/2601.15077 · **PDF:** https://arxiv.org/pdf/2601.15077v1
+
+## 20194. The Why Behind the Action: Unveiling Internal Drivers via Agentic Attribution
+
+- **ID:** `2601.15075` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Chen Qian, Peng Wang, Dongrui Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.15075 · **PDF:** https://arxiv.org/pdf/2601.15075v2
+
+## 20195. SmartOracle -- An Agentic Approach to Mitigate Noise in Differential Oracles
+
+- **ID:** `2601.15074` · **primary:** `cs.SE` · **Submitted:** 2026-01-21
+- **Authors:** Srinath Srinivasan, Tim Menzies, Marcelo D'Amorim
+- **abs:** https://arxiv.org/abs/2601.15074 · **PDF:** https://arxiv.org/pdf/2601.15074v1
+
+## 20196. Incentive-Tuning: Understanding and Designing Incentives for Empirical Human-AI Decision-Making Studies
+
+- **ID:** `2601.15064` · **primary:** `cs.HC` · **Submitted:** 2026-01-21
+- **Authors:** Simran Kaur, Sara Salimzadeh, Ujwal Gadiraju
+- **abs:** https://arxiv.org/abs/2601.15064 · **PDF:** https://arxiv.org/pdf/2601.15064v1
+
+## 20197. Differential Privacy Image Generation with Reconstruction Loss and Noise Injection Using an Error Feedback SGD
+
+- **ID:** `2601.15061` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Qiwei Ma, Jun Zhang
+- **abs:** https://arxiv.org/abs/2601.15061 · **PDF:** https://arxiv.org/pdf/2601.15061v1
+
+## 20198. Retrieval-Augmented Self-Taught Reasoning Model with Adaptive Chain-of-Thought for ASR Named Entity Correction
+
+- **ID:** `2602.12287` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Junjie An, Jingguang Tian, Tianyi Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.12287 · **PDF:** https://arxiv.org/pdf/2602.12287v1
+
+## 20199. The Responsibility Vacuum: Organizational Failure in Scaled Agent Systems
+
+- **ID:** `2601.15059` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Oleg Romanchuk, Roman Bondar
+- **abs:** https://arxiv.org/abs/2601.15059 · **PDF:** https://arxiv.org/pdf/2601.15059v1
+
+## 20200. SpooFL: Spoofing Federated Learning
+
+- **ID:** `2601.15055` · **primary:** `cs.CR` · **Submitted:** 2026-01-21
+- **Authors:** Isaac Baglin, Xiatian Zhu, Simon Hadfield
+- **abs:** https://arxiv.org/abs/2601.15055 · **PDF:** https://arxiv.org/pdf/2601.15055v1
+
+## 20201. LogicScore: Fine-grained Logic Evaluation of Conciseness, Completeness, and Determinateness in Attributed Question Answering
+
+- **ID:** `2601.15050` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Zhichao Yan, Yunxiao Zhao, Jiapu Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.15050 · **PDF:** https://arxiv.org/pdf/2601.15050v3
+
+## 20202. Federated Transformer-GNN for Privacy-Preserving Brain Tumor Localization with Modality-Level Explainability
+
+- **ID:** `2601.15042` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Andrea Protani, Riccardo Taiello, Marc Molina Van Den Bosch, et al.
+- **abs:** https://arxiv.org/abs/2601.15042 · **PDF:** https://arxiv.org/pdf/2601.15042v1
+
+## 20203. HyperNet-Adaptation for Diffusion-Based Test Case Generation
+
+- **ID:** `2601.15041` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Oliver Weißl, Vincenzo Riccio, Severin Kacianka, et al.
+- **abs:** https://arxiv.org/abs/2601.15041 · **PDF:** https://arxiv.org/pdf/2601.15041v1
+
+## 20204. A Curriculum-Based Deep Reinforcement Learning Framework for the Electric Vehicle Routing Problem
+
+- **ID:** `2601.15038` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Mertcan Daysalilar, Fuat Uyguroglu, Gabriel Nicolosi, et al.
+- **abs:** https://arxiv.org/abs/2601.15038 · **PDF:** https://arxiv.org/pdf/2601.15038v1
+
+## 20205. Knowledge Restoration-driven Prompt Optimization: Unlocking LLM Potential for Open-Domain Relational Triplet Extraction
+
+- **ID:** `2601.15037` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Xiaonan Jing, Gongqing Wu, Xingrui Zhuo, et al.
+- **abs:** https://arxiv.org/abs/2601.15037 · **PDF:** https://arxiv.org/pdf/2601.15037v1
+
+## 20206. Factorizable joint shift revisited
+
+- **ID:** `2601.15036` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Dirk Tasche
+- **abs:** https://arxiv.org/abs/2601.15036 · **PDF:** https://arxiv.org/pdf/2601.15036v2
+
+## 20207. Visual and Cognitive Demands of a Large Language Model-Powered In-vehicle Conversational Agent
+
+- **ID:** `2601.15034` · **primary:** `cs.HC` · **Submitted:** 2026-01-21
+- **Authors:** Chris Monk, Allegra Ayala, Christine S. P. Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.15034 · **PDF:** https://arxiv.org/pdf/2601.15034v1
+
+## 20208. Non-Stationary Functional Bilevel Optimization
+
+- **ID:** `2601.15363` · **primary:** `stat.ML` · **Submitted:** 2026-01-21
+- **Authors:** Jason Bohne, Ieva Petrulionyte, Michael Arbel, et al.
+- **abs:** https://arxiv.org/abs/2601.15363 · **PDF:** https://arxiv.org/pdf/2601.15363v1
+
+## 20209. Emergent, not Immanent: A Baradian Reading of Explainable AI
+
+- **ID:** `2601.15029` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Fabio Morreale, Joan Serrà, Yuki Mitsufuji
+- **abs:** https://arxiv.org/abs/2601.15029 · **PDF:** https://arxiv.org/pdf/2601.15029v2
+
+## 20210. Mixture-of-Experts Models in Vision: Routing, Optimization, and Generalization
+
+- **ID:** `2601.15021` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Adam Rokah, Daniel Veress, Caleb Caulk, et al.
+- **abs:** https://arxiv.org/abs/2601.15021 · **PDF:** https://arxiv.org/pdf/2601.15021v1
+
+## 20211. Plug-and-Play Benchmarking of Reinforcement Learning Algorithms for Large-Scale Flow Control
+
+- **ID:** `2601.15015` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Jannis Becktepe, Aleksandra Franz, Nils Thuerey, et al.
+- **abs:** https://arxiv.org/abs/2601.15015 · **PDF:** https://arxiv.org/pdf/2601.15015v1
+
+## 20212. Efficient and Minimax-optimal In-context Nonparametric Regression with Transformers
+
+- **ID:** `2601.15014` · **primary:** `stat.ML` · **Submitted:** 2026-01-21
+- **Authors:** Michelle Ching, Ioana Popescu, Nico Smith, et al.
+- **abs:** https://arxiv.org/abs/2601.15014 · **PDF:** https://arxiv.org/pdf/2601.15014v1
+
+## 20213. RadixMLP -- Intra-batch Deduplication for Causal Transformers
+
+- **ID:** `2601.15013` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Michael Feil, Julius Lipp
+- **abs:** https://arxiv.org/abs/2601.15013 · **PDF:** https://arxiv.org/pdf/2601.15013v1
+
+## 20214. Lineup Regularized Adjusted Plus-Minus (L-RAPM): Basketball Lineup Ratings with Informed Priors
+
+- **ID:** `2601.15000` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Christos Petridis, Konstantinos Pelechrinis
+- **abs:** https://arxiv.org/abs/2601.15000 · **PDF:** https://arxiv.org/pdf/2601.15000v1
+
+## 20215. Obscuring Data Contamination Through Translation: Evidence from Arabic Corpora
+
+- **ID:** `2601.14994` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Chaymaa Abbas, Nour Shamaa, Mariette Awad
+- **abs:** https://arxiv.org/abs/2601.14994 · **PDF:** https://arxiv.org/pdf/2601.14994v1
+
+## 20216. Consistency of Honest Decision Trees and Random Forests
+
+- **ID:** `2601.14991` · **primary:** `stat.ME` · **Submitted:** 2026-01-21
+- **Authors:** Martin Bladt, Rasmus Frigaard Lemvig
+- **abs:** https://arxiv.org/abs/2601.14991 · **PDF:** https://arxiv.org/pdf/2601.14991v1
+
+## 20217. Failing on Bias Mitigation: A Case Study on the Challenges of Fairness in Government Data
+
+- **ID:** `2601.17054` · **primary:** `cs.CY` · **Submitted:** 2026-01-21
+- **Authors:** Hongbo Bo, Jingyu Hu, Debbie Watson, et al.
+- **abs:** https://arxiv.org/abs/2601.17054 · **PDF:** https://arxiv.org/pdf/2601.17054v2
+
+## 20218. Interoperable Architecture for Digital Identity Delegation for AI Agents with Blockchain Integration
+
+- **ID:** `2601.14982` · **primary:** `cs.CR` · **Submitted:** 2026-01-21
+- **Authors:** David Ricardo Saavedra
+- **abs:** https://arxiv.org/abs/2601.14982 · **PDF:** https://arxiv.org/pdf/2601.14982v1
+
+## 20219. HumanDiffusion: A Vision-Based Diffusion Trajectory Planner with Human-Conditioned Goals for Search and Rescue UAV
+
+- **ID:** `2601.14973` · **primary:** `cs.RO` · **Submitted:** 2026-01-21
+- **Authors:** Faryal Batool, Iana Zhura, Valerii Serpiva, et al.
+- **abs:** https://arxiv.org/abs/2601.14973 · **PDF:** https://arxiv.org/pdf/2601.14973v2
+
+## 20220. Fine-Grained Traceability for Transparent ML Pipelines
+
+- **ID:** `2601.14971` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Liping Chen, Mujie Liu, Haytham Fayek
+- **abs:** https://arxiv.org/abs/2601.14971 · **PDF:** https://arxiv.org/pdf/2601.14971v1
+
+## 20221. Robust Machine Learning for Regulatory Sequence Modeling under Biological and Technical Distribution Shifts
+
+- **ID:** `2601.14969` · **primary:** `q-bio.GN` · **Submitted:** 2026-01-21
+- **Authors:** Yiyao Yang
+- **abs:** https://arxiv.org/abs/2601.14969 · **PDF:** https://arxiv.org/pdf/2601.14969v2
+
+## 20222. InstructTime++: Time Series Classification with Multimodal Language Modeling via Implicit Feature Enhancement
+
+- **ID:** `2601.14968` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Mingyue Cheng, Xiaoyu Tao, Huajian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.14968 · **PDF:** https://arxiv.org/pdf/2601.14968v1
+
+## 20223. Script Sensitivity: Benchmarking Language Models on Unicode, Romanized and Mixed-Script Sinhala
+
+- **ID:** `2601.14958` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Minuri Rajapakse, Ruvan Weerasinghe
+- **abs:** https://arxiv.org/abs/2601.14958 · **PDF:** https://arxiv.org/pdf/2601.14958v2
+
+## 20224. Improving Regret Approximation for Unsupervised Dynamic Environment Generation
+
+- **ID:** `2601.14957` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Harry Mead, Bruno Lacerda, Jakob Foerster, et al.
+- **abs:** https://arxiv.org/abs/2601.14957 · **PDF:** https://arxiv.org/pdf/2601.14957v1
+
+## 20225. Multi-Behavior Sequential Modeling with Transition-Aware Graph Attention Network for E-Commerce Recommendation
+
+- **ID:** `2601.14955` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Hanqi Jin, Gaoming Yang, Zhangming Chan, et al.
+- **abs:** https://arxiv.org/abs/2601.14955 · **PDF:** https://arxiv.org/pdf/2601.14955v1
+
+## 20226. Multimodal Rumor Detection Enhanced by External Evidence and Forgery Features
+
+- **ID:** `2601.14954` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Han Li, Hua Sun
+- **abs:** https://arxiv.org/abs/2601.14954 · **PDF:** https://arxiv.org/pdf/2601.14954v2
+
+## 20227. CorpusQA: A 10 Million Token Benchmark for Corpus-Level Analysis and Reasoning
+
+- **ID:** `2601.14952` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Zhiyuan Lu, Chenliang Li, Yingcheng Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.14952 · **PDF:** https://arxiv.org/pdf/2601.14952v1
+
+## 20228. TempViz: On the Evaluation of Temporal Knowledge in Text-to-Image Models
+
+- **ID:** `2601.14951` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Carolin Holtermann, Nina Krebs, Anne Lauscher
+- **abs:** https://arxiv.org/abs/2601.14951 · **PDF:** https://arxiv.org/pdf/2601.14951v1
+
+## 20229. TIDAL: Temporally Interleaved Diffusion and Action Loop for High-Frequency VLA Control
+
+- **ID:** `2601.14945` · **primary:** `cs.RO` · **Submitted:** 2026-01-21
+- **Authors:** Yuteng Sun, Haoran Wang, Ruofei Bai, et al.
+- **abs:** https://arxiv.org/abs/2601.14945 · **PDF:** https://arxiv.org/pdf/2601.14945v1
+
+## 20230. The GDN-CC Dataset: Automatic Corpus Clarification for AI-enhanced Democratic Citizen Consultations
+
+- **ID:** `2601.14944` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Pierre-Antoine Lequeu, Léo Labat, Laurène Cave, et al.
+- **abs:** https://arxiv.org/abs/2601.14944 · **PDF:** https://arxiv.org/pdf/2601.14944v2
+
+## 20231. Communication-Efficient Multi-Modal Edge Inference via Uncertainty-Aware Distributed Learning
+
+- **ID:** `2601.14942` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Hang Zhao, Hongru Li, Dongfang Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.14942 · **PDF:** https://arxiv.org/pdf/2601.14942v1
+
+## 20232. Generative Artificial Intelligence, Musical Heritage and the Construction of Peace Narratives: A Case Study in Mali
+
+- **ID:** `2601.14931` · **primary:** `cs.SD` · **Submitted:** 2026-01-21
+- **Authors:** Nouhoum Coulibaly, Ousmane Ly, Michael Leventhal, et al.
+- **abs:** https://arxiv.org/abs/2601.14931 · **PDF:** https://arxiv.org/pdf/2601.14931v1
+
+## 20233. Fast-ULCNet: A fast and ultra low complexity network for single-channel speech enhancement
+
+- **ID:** `2601.14925` · **primary:** `eess.AS` · **Submitted:** 2026-01-21
+- **Authors:** Nicolás Arrieta Larraza, Niels de Koeijer
+- **abs:** https://arxiv.org/abs/2601.14925 · **PDF:** https://arxiv.org/pdf/2601.14925v1
+
+## 20234. Vision-Language Models on the Edge for Real-Time Robotic Perception
+
+- **ID:** `2601.14921` · **primary:** `cs.RO` · **Submitted:** 2026-01-21
+- **Authors:** Sarat Ahmad, Maryam Hafeez, Syed Ali Raza Zaidi
+- **abs:** https://arxiv.org/abs/2601.14921 · **PDF:** https://arxiv.org/pdf/2601.14921v1
+
+## 20235. USDs: A universal stabilizer decoder framework using symmetry
+
+- **ID:** `2601.15361` · **primary:** `quant-ph` · **Submitted:** 2026-01-21
+- **Authors:** Hoshitaro Ohnishi, Hideo Mukai
+- **abs:** https://arxiv.org/abs/2601.15361 · **PDF:** https://arxiv.org/pdf/2601.15361v2
+
+## 20236. Standards for trustworthy AI in the European Union: technical rationale, structural challenges, and an implementation path
+
+- **ID:** `2602.00078` · **primary:** `cs.CY` · **Submitted:** 2026-01-21
+- **Authors:** Piercosma Bisconti, Marcello Galisai
+- **abs:** https://arxiv.org/abs/2602.00078 · **PDF:** https://arxiv.org/pdf/2602.00078v1
+
+## 20237. Tailoring Adverse Event Prediction in Type 1 Diabetes with Patient-Specific Deep Learning Models
+
+- **ID:** `2601.14917` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Giorgia Rigamonti, Mirko Paolo Barbato, Davide Marelli, et al.
+- **abs:** https://arxiv.org/abs/2601.14917 · **PDF:** https://arxiv.org/pdf/2601.14917v1
+
+## 20238. CodeDelegator: Mitigating Context Pollution via Role Separation in Code-as-Action Agents
+
+- **ID:** `2601.14914` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Tianxiang Fei, Cheng Chen, Yue Pan, et al.
+- **abs:** https://arxiv.org/abs/2601.14914 · **PDF:** https://arxiv.org/pdf/2601.14914v1
+
+## 20239. Robust X-Learner: Breaking the Curse of Imbalance and Heavy Tails via Robust Cross-Imputation
+
+- **ID:** `2601.15360` · **primary:** `stat.ML` · **Submitted:** 2026-01-21
+- **Authors:** Eichi Uehara
+- **abs:** https://arxiv.org/abs/2601.15360 · **PDF:** https://arxiv.org/pdf/2601.15360v1
+
+## 20240. Mechanistic Interpretability for Large Language Model Alignment: Progress, Challenges, and Future Directions
+
+- **ID:** `2602.11180` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Usman Naseem
+- **abs:** https://arxiv.org/abs/2602.11180 · **PDF:** https://arxiv.org/pdf/2602.11180v1
+
+## 20241. PodBench: A Comprehensive Benchmark for Instruction-Aware Audio-Oriented Podcast Script Generation
+
+- **ID:** `2601.14903` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Chenning Xu, Mao Zheng, Mingyu Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.14903 · **PDF:** https://arxiv.org/pdf/2601.14903v1
+
+## 20242. Just aware enough: Evaluating awareness across artificial systems
+
+- **ID:** `2601.14901` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Nadine Meertens, Suet Lee, Ophelia Deroy
+- **abs:** https://arxiv.org/abs/2601.14901 · **PDF:** https://arxiv.org/pdf/2601.14901v1
+
+## 20243. Language-Coupled Reinforcement Learning for Multilingual Retrieval-Augmented Generation
+
+- **ID:** `2601.14896` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Rui Qi, Fengran Mo, Yufeng Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.14896 · **PDF:** https://arxiv.org/pdf/2601.14896v1
+
+## 20244. SpatialMem: Metric-Aligned Long-Horizon Video Memory for Language Grounding and QA
+
+- **ID:** `2601.14895` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Xinyi Zheng, Yunze Liu, Chi-Hao Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.14895 · **PDF:** https://arxiv.org/pdf/2601.14895v2
+
+## 20245. To Neuro-Symbolic Classification and Beyond by Compiling Description Logic Ontologies to Probabilistic Circuits
+
+- **ID:** `2601.14894` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Nicolas Lazzari, Valentina Presutti, Antonio Vergari
+- **abs:** https://arxiv.org/abs/2601.14894 · **PDF:** https://arxiv.org/pdf/2601.14894v1
+
+## 20246. XInsight: Integrative Stage-Consistent Psychological Counseling Support Agents for Digital Well-Being
+
+- **ID:** `2603.06583` · **primary:** `cs.HC` · **Submitted:** 2026-01-21
+- **Authors:** Fei Wang, Jiangnan Yang, Junjie Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.06583 · **PDF:** https://arxiv.org/pdf/2603.06583v1
+
+## 20247. What Makes Low-Bit Quantization-Aware Training Work for Reasoning LLMs? A Systematic Study
+
+- **ID:** `2601.14888` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Keyu Lv, Manyi Zhang, Xiaobo Xia, et al.
+- **abs:** https://arxiv.org/abs/2601.14888 · **PDF:** https://arxiv.org/pdf/2601.14888v1
+
+## 20248. ExoMiner++ 2.0: Vetting TESS Full-Frame Image Transit Signals
+
+- **ID:** `2601.14877` · **primary:** `astro-ph.EP` · **Submitted:** 2026-01-21
+- **Authors:** Miguel J. S. Martinho, Hamed Valizadegan, Jon M. Jenkins, et al.
+- **abs:** https://arxiv.org/abs/2601.14877 · **PDF:** https://arxiv.org/pdf/2601.14877v2
+
+## 20249. GAT-NeRF: Geometry-Aware-Transformer Enhanced Neural Radiance Fields for High-Fidelity 4D Facial Avatars
+
+- **ID:** `2601.14875` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Zhe Chang, Haodong Jin, Ying Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.14875 · **PDF:** https://arxiv.org/pdf/2601.14875v1
+
+## 20250. Finite-Sample Inference for Sparsely Permuted Linear Regression
+
+- **ID:** `2601.14872` · **primary:** `math.ST` · **Submitted:** 2026-01-21
+- **Authors:** Hirofumi Ota, Masaaki Imaizumi
+- **abs:** https://arxiv.org/abs/2601.14872 · **PDF:** https://arxiv.org/pdf/2601.14872v2
+
+## 20251. Strategic Doctrine Language Models (sdLM): A Learning-System Framework for Doctrinal Consistency and Geopolitical Forecasting
+
+- **ID:** `2601.14862` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Olaf Yunus Laitinen Imanov, Taner Yilmaz, Derya Umut Kulali
+- **abs:** https://arxiv.org/abs/2601.14862 · **PDF:** https://arxiv.org/pdf/2601.14862v1
+
+## 20252. HiNS: Hierarchical Negative Sampling for More Comprehensive Memory Retrieval Embedding Model
+
+- **ID:** `2601.14857` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Motong Tian, Allen P. Wong, Mingjun Mao, et al.
+- **abs:** https://arxiv.org/abs/2601.14857 · **PDF:** https://arxiv.org/pdf/2601.14857v1
+
+## 20253. Adaptive Exponential Integration for Stable Gaussian Mixture Black-Box Variational Inference
+
+- **ID:** `2601.14855` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Baojun Che, Yifan Chen, Daniel Zhengyu Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.14855 · **PDF:** https://arxiv.org/pdf/2601.14855v2
+
+## 20254. From Observation to Prediction: LSTM for Vehicle Lane Change Forecasting on Highway On/Off-Ramps
+
+- **ID:** `2601.14848` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Mohamed Abouras, Catherine M. Elias
+- **abs:** https://arxiv.org/abs/2601.14848 · **PDF:** https://arxiv.org/pdf/2601.14848v1
+
+## 20255. CAG-Avatar: Cross-Attention Guided Gaussian Avatars for High-Fidelity Head Reconstruction
+
+- **ID:** `2601.14844` · **primary:** `cs.GR` · **Submitted:** 2026-01-21
+- **Authors:** Zhe Chang, Haodong Jin, Yan Song, et al.
+- **abs:** https://arxiv.org/abs/2601.14844 · **PDF:** https://arxiv.org/pdf/2601.14844v1
+
+## 20256. MTFlow: Time-Conditioned Flow Matching for Microtubule Segmentation in Noisy Microscopy Images
+
+- **ID:** `2601.14841` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Sidi Mohamed Sid El Moctar, Achraf Ait Laydi, Yousef El Mourabit, et al.
+- **abs:** https://arxiv.org/abs/2601.14841 · **PDF:** https://arxiv.org/pdf/2601.14841v1
+
+## 20257. Implementing Knowledge Representation and Reasoning with Object Oriented Design
+
+- **ID:** `2601.14840` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Abdelrhman Bassiouny, Tom Schierenbeck, Sorin Arion, et al.
+- **abs:** https://arxiv.org/abs/2601.14840 · **PDF:** https://arxiv.org/pdf/2601.14840v1
+
+## 20258. Measuring and Aligning Abstraction in Vision-Language Models with Medical Taxonomies
+
+- **ID:** `2601.14827` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Ben Schaper, Maxime Di Folco, Bernhard Kainz, et al.
+- **abs:** https://arxiv.org/abs/2601.14827 · **PDF:** https://arxiv.org/pdf/2601.14827v1
+
+## 20259. Comparative Study of Large Language Models on Chinese Film Script Continuation: An Empirical Analysis Based on GPT-5.2 and Qwen-Max
+
+- **ID:** `2601.14826` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Yuxuan Cao, Zida Yang, Ye Wang
+- **abs:** https://arxiv.org/abs/2601.14826 · **PDF:** https://arxiv.org/pdf/2601.14826v1
+
+## 20260. Automated univariate time series forecasting with regression trees
+
+- **ID:** `2602.00077` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Francisco Martínez, María P. Frías
+- **abs:** https://arxiv.org/abs/2602.00077 · **PDF:** https://arxiv.org/pdf/2602.00077v1
+
+## 20261. Multimodal system for skin cancer detection
+
+- **ID:** `2601.14822` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Volodymyr Sydorskyi, Igor Krashenyi, Oleksii Yakubenko
+- **abs:** https://arxiv.org/abs/2601.14822 · **PDF:** https://arxiv.org/pdf/2601.14822v1
+
+## 20262. Statistical Learning Theory for Distributional Classification
+
+- **ID:** `2601.14818` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Christian Fiedler
+- **abs:** https://arxiv.org/abs/2601.14818 · **PDF:** https://arxiv.org/pdf/2601.14818v1
+
+## 20263. Learning and extrapolating scale-invariant processes
+
+- **ID:** `2601.14810` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-01-21
+- **Authors:** Anaclara Alvez-Canepa, Cyril Furtlehner, François Landes
+- **abs:** https://arxiv.org/abs/2601.14810 · **PDF:** https://arxiv.org/pdf/2601.14810v2
+
+## 20264. Reflecting in the Reflection: Integrating a Socratic Questioning Framework into Automated AI-Based Question Generation
+
+- **ID:** `2601.14798` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Ondřej Holub, Essi Ryymin, Rodrigo Alves
+- **abs:** https://arxiv.org/abs/2601.14798 · **PDF:** https://arxiv.org/pdf/2601.14798v1
+
+## 20265. RANDSMAPs: Random-Feature/multi-Scale Neural Decoders with Mass Preservation
+
+- **ID:** `2601.14794` · **primary:** `math.NA` · **Submitted:** 2026-01-21
+- **Authors:** Dimitrios G. Patsatzis, Alessandro Della Pia, Lucia Russo, et al.
+- **abs:** https://arxiv.org/abs/2601.14794 · **PDF:** https://arxiv.org/pdf/2601.14794v1
+
+## 20266. Robustness of Mixtures of Experts to Feature Noise
+
+- **ID:** `2601.14792` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Dong Sun, Rahul Nittala, Rebekka Burkholz
+- **abs:** https://arxiv.org/abs/2601.14792 · **PDF:** https://arxiv.org/pdf/2601.14792v1
+
+## 20267. Synthetic Data Augmentation for Multi-Task Chinese Porcelain Classification: A Stable Diffusion Approach
+
+- **ID:** `2601.14791` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Ziyao Ling, Silvia Mirri, Paola Salomoni, et al.
+- **abs:** https://arxiv.org/abs/2601.14791 · **PDF:** https://arxiv.org/pdf/2601.14791v1
+
+## 20268. CI4A: Semantic Component Interfaces for Agents Empowering Web Automation
+
+- **ID:** `2601.14790` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Zhi Qiu, Jiazheng Sun, Chenxiao Xia, et al.
+- **abs:** https://arxiv.org/abs/2601.14790 · **PDF:** https://arxiv.org/pdf/2601.14790v1
+
+## 20269. Single-Pixel Vision-Language Model for Intrinsic Privacy-Preserving Behavioral Intelligence
+
+- **ID:** `2601.17050` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Hongjun An, Yiliang Song, Jiawei Shao, et al.
+- **abs:** https://arxiv.org/abs/2601.17050 · **PDF:** https://arxiv.org/pdf/2601.17050v1
+
+## 20270. Training-Efficient Text-to-Music Generation with State-Space Modeling
+
+- **ID:** `2601.14786` · **primary:** `cs.SD` · **Submitted:** 2026-01-21
+- **Authors:** Wei-Jaw Lee, Fang-Chih Hsieh, Xuanjun Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.14786 · **PDF:** https://arxiv.org/pdf/2601.14786v1
+
+## 20271. Towards Bound Consistency for the No-Overlap Constraint Using MDDs
+
+- **ID:** `2601.14784` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Amaury Guichard, Laurent Michel, Hélène Verhaeghe, et al.
+- **abs:** https://arxiv.org/abs/2601.14784 · **PDF:** https://arxiv.org/pdf/2601.14784v1
+
+## 20272. RECAP: Resistance Capture in Text-based Mental Health Counseling with Large Language Models
+
+- **ID:** `2601.14780` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Anqi Li, Yuqian Chen, Yu Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.14780 · **PDF:** https://arxiv.org/pdf/2601.14780v1
+
+## 20273. FunCineForge: A Unified Dataset Toolkit and Model for Zero-Shot Movie Dubbing in Diverse Cinematic Scenes
+
+- **ID:** `2601.14777` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Jiaxuan Liu, Yang Xiang, Han Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.14777 · **PDF:** https://arxiv.org/pdf/2601.14777v1
+
+## 20274. Semantic-Guided Unsupervised Video Summarization
+
+- **ID:** `2601.14773` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Haizhou Liu, Haodong Jin, Yiming Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.14773 · **PDF:** https://arxiv.org/pdf/2601.14773v1
+
+## 20275. Anytime Optimal Decision Tree Learning with Continuous Features
+
+- **ID:** `2601.14765` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Harold Kiossou, Pierre Schaus, Siegfried Nijssen
+- **abs:** https://arxiv.org/abs/2601.14765 · **PDF:** https://arxiv.org/pdf/2601.14765v1
+
+## 20276. An XAI View on Explainable ASP: Methods, Systems, and Perspectives
+
+- **ID:** `2601.14764` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Thomas Eiter, Tobias Geibinger, Zeynep G. Saribatur
+- **abs:** https://arxiv.org/abs/2601.14764 · **PDF:** https://arxiv.org/pdf/2601.14764v1
+
+## 20277. Mechanism Shift During Post-training from Autoregressive to Masked Diffusion Language Models
+
+- **ID:** `2601.14758` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Injin Kong, Hyoungjoon Lee, Yohan Jo
+- **abs:** https://arxiv.org/abs/2601.14758 · **PDF:** https://arxiv.org/pdf/2601.14758v3
+
+## 20278. Render-of-Thought: Rendering Textual Chain-of-Thought as Images for Visual Latent Reasoning
+
+- **ID:** `2601.14750` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Yifan Wang, Shiyu Li, Peiming Li, et al.
+- **abs:** https://arxiv.org/abs/2601.14750 · **PDF:** https://arxiv.org/pdf/2601.14750v2
+
+## 20279. Q-Probe: Scaling Image Quality Assessment to High Resolution via Context-Aware Agentic Probing
+
+- **ID:** `2601.15356` · **primary:** `eess.IV` · **Submitted:** 2026-01-21
+- **Authors:** Xiang Li, XueHeng Li, Yu Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.15356 · **PDF:** https://arxiv.org/pdf/2601.15356v2
+
+## 20280. RefProtoFL: Communication-Efficient Federated Learning via External-Referenced Prototype Alignment
+
+- **ID:** `2601.14746` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Hongyue Wu, Hangyu Li, Guodong Fan, et al.
+- **abs:** https://arxiv.org/abs/2601.14746 · **PDF:** https://arxiv.org/pdf/2601.14746v2
+
+## 20281. Alignment or Integration? Rethinking Multimodal Fusion in DNA-language Foundation Models
+
+- **ID:** `2602.12286` · **primary:** `q-bio.GN` · **Submitted:** 2026-01-21
+- **Authors:** Yanan Li, Christina Yi Jin, Yuan Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.12286 · **PDF:** https://arxiv.org/pdf/2602.12286v1
+
+## 20282. DeepMoLM: Leveraging Visual and Geometric Structural Information for Molecule-Text Modeling
+
+- **ID:** `2601.14732` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Jing Lan, Hexiao Ding, Hongzhao Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.14732 · **PDF:** https://arxiv.org/pdf/2601.14732v1
+
+## 20283. ARFT-Transformer: Modeling Metric Dependencies for Cross-Project Aging-Related Bug Prediction
+
+- **ID:** `2601.14731` · **primary:** `cs.SE` · **Submitted:** 2026-01-21
+- **Authors:** Shuning Ge, Fangyun Qin, Xiaohui Wan, et al.
+- **abs:** https://arxiv.org/abs/2601.14731 · **PDF:** https://arxiv.org/pdf/2601.14731v1
+
+## 20284. FSX: Message Flow Sensitivity Enhanced Structural Explainer for Graph Neural Networks
+
+- **ID:** `2601.14730` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Bizu Feng, Zhimu Yang, Shaode Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.14730 · **PDF:** https://arxiv.org/pdf/2601.14730v1
+
+## 20285. AQAScore: Evaluating Semantic Alignment in Text-to-Audio Generation via Audio Question Answering
+
+- **ID:** `2601.14728` · **primary:** `eess.AS` · **Submitted:** 2026-01-21
+- **Authors:** Chun-Yi Kuan, Kai-Wei Chang, Hung-yi Lee
+- **abs:** https://arxiv.org/abs/2601.14728 · **PDF:** https://arxiv.org/pdf/2601.14728v1
+
+## 20286. Memes-as-Replies: Can Models Select Humorous Manga Panel Responses?
+
+- **ID:** `2602.15842` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Ryosuke Kohita, Seiichiro Yoshioka
+- **abs:** https://arxiv.org/abs/2602.15842 · **PDF:** https://arxiv.org/pdf/2602.15842v1
+
+## 20287. HERMES: KV Cache as Hierarchical Memory for Efficient Streaming Video Understanding
+
+- **ID:** `2601.14724` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Haowei Zhang, Shudong Yang, Jinlan Fu, et al.
+- **abs:** https://arxiv.org/abs/2601.14724 · **PDF:** https://arxiv.org/pdf/2601.14724v2
+
+## 20288. Typhoon OCR: Open Vision-Language Model For Thai Document Extraction
+
+- **ID:** `2601.14722` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Surapon Nonesung, Natapong Nitarach, Teetouch Jaknamon, et al.
+- **abs:** https://arxiv.org/abs/2601.14722 · **PDF:** https://arxiv.org/pdf/2601.14722v1
+
+## 20289. PCL-Reasoner-V1.5: Advancing Math Reasoning with Offline Reinforcement Learning
+
+- **ID:** `2601.14716` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Yao Lu, Dengdong Fan, Jianzheng Nie, et al.
+- **abs:** https://arxiv.org/abs/2601.14716 · **PDF:** https://arxiv.org/pdf/2601.14716v1
+
+## 20290. From Instruction to Output: The Role of Prompting in Modern NLG
+
+- **ID:** `2602.11179` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Munazza Zaib, Elaf Alhazmi
+- **abs:** https://arxiv.org/abs/2602.11179 · **PDF:** https://arxiv.org/pdf/2602.11179v1
+
+## 20291. Adaptive Fidelity Estimation for Quantum Programs with Graph-Guided Noise Awareness
+
+- **ID:** `2601.14713` · **primary:** `quant-ph` · **Submitted:** 2026-01-21
+- **Authors:** Tingting Li, Ziming Zhao, Jianwei Yin
+- **abs:** https://arxiv.org/abs/2601.14713 · **PDF:** https://arxiv.org/pdf/2601.14713v1
+
+## 20292. DARA: Few-shot Budget Allocation in Online Advertising via In-Context Decision Making with RL-Finetuned LLMs
+
+- **ID:** `2601.14711` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Mingxuan Song, Yusen Huo, Bohan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.14711 · **PDF:** https://arxiv.org/pdf/2601.14711v1
+
+## 20293. Case-Guided Sequential Assay Planning in Drug Discovery
+
+- **ID:** `2601.14710` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Tianchi Chen, Jan Bima, Sean L. Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.14710 · **PDF:** https://arxiv.org/pdf/2601.14710v1
+
+## 20294. Proximal Policy Optimization with Evolutionary Mutations
+
+- **ID:** `2601.14705` · **primary:** `cs.NE` · **Submitted:** 2026-01-21
+- **Authors:** Casimir Czworkowski, Stephen Hornish, Alhassan S. Yasin
+- **abs:** https://arxiv.org/abs/2601.14705 · **PDF:** https://arxiv.org/pdf/2601.14705v1
+
+## 20295. AutoDriDM: An Explainable Benchmark for Decision-Making of Vision-Language Models in Autonomous Driving
+
+- **ID:** `2601.14702` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Zecong Tang, Zixu Wang, Yifei Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.14702 · **PDF:** https://arxiv.org/pdf/2601.14702v1
+
+## 20296. DARL: Encouraging Diverse Answers for General Reasoning without Verifiers
+
+- **ID:** `2601.14700` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Chongxuan Huang, Lei Lin, Xiaodong Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.14700 · **PDF:** https://arxiv.org/pdf/2601.14700v1
+
+## 20297. Repair Brain Damage: Real-Numbered Error Correction Code for Neural Network
+
+- **ID:** `2602.00076` · **primary:** `cs.NE` · **Submitted:** 2026-01-21
+- **Authors:** Ziqing Li, Myung Cho, Qiutong Jin, et al.
+- **abs:** https://arxiv.org/abs/2602.00076 · **PDF:** https://arxiv.org/pdf/2602.00076v1
+
+## 20298. ClaimDB: A Fact Verification Benchmark over Large Structured Data
+
+- **ID:** `2601.14698` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Michael Theologitis, Preetam Prabhu Srikar Dammu, Chirag Shah, et al.
+- **abs:** https://arxiv.org/abs/2601.14698 · **PDF:** https://arxiv.org/pdf/2601.14698v1
+
+## 20299. When Text-as-Vision Meets Semantic IDs in Generative Recommendation: An Empirical Study
+
+- **ID:** `2601.14697` · **primary:** `cs.IR` · **Submitted:** 2026-01-21
+- **Authors:** Shutong Qiao, Wei Yuan, Tong Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.14697 · **PDF:** https://arxiv.org/pdf/2601.14697v1
+
+## 20300. AdaTIR: Adaptive Tool-Integrated Reasoning via Difficulty-Aware Policy Optimization
+
+- **ID:** `2601.14696` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Zhaiyu Fang, Ruipeng Sun
+- **abs:** https://arxiv.org/abs/2601.14696 · **PDF:** https://arxiv.org/pdf/2601.14696v1
+
+## 20301. CoScale-RL: Efficient Post-Training by Co-Scaling Data and Computation
+
+- **ID:** `2601.14695` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Yutong Chen, Jiandong Gao, Ji Wu
+- **abs:** https://arxiv.org/abs/2601.14695 · **PDF:** https://arxiv.org/pdf/2601.14695v1
+
+## 20302. Re-understanding Graph Unlearning through Memorization
+
+- **ID:** `2601.14694` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Pengfei Ding, Yan Wang, Guanfeng Liu
+- **abs:** https://arxiv.org/abs/2601.14694 · **PDF:** https://arxiv.org/pdf/2601.14694v1
+
+## 20303. Beyond Error-Based Optimization: Experience-Driven Symbolic Regression with Goal-Conditioned Reinforcement Learning
+
+- **ID:** `2601.14693` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Jianwen Sun, Xinrui Li, Fuqing Li, et al.
+- **abs:** https://arxiv.org/abs/2601.14693 · **PDF:** https://arxiv.org/pdf/2601.14693v1
+
+## 20304. Gaming the Judge: Unfaithful Chain-of-Thought Can Undermine Agent Evaluation
+
+- **ID:** `2601.14691` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Muhammad Khalifa, Lajanugen Logeswaran, Jaekyeom Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.14691 · **PDF:** https://arxiv.org/pdf/2601.14691v2
+
+## 20305. Beyond Denial-of-Service: The Puppeteer's Attack for Fine-Grained Control in Ranking-Based Federated Learning
+
+- **ID:** `2601.14687` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Zhihao Chen, Zirui Gong, Jianting Ning, et al.
+- **abs:** https://arxiv.org/abs/2601.14687 · **PDF:** https://arxiv.org/pdf/2601.14687v1
+
+## 20306. IB-GRPO: Aligning LLM-based Learning Path Recommendation with Educational Objectives via Indicator-Based Group Relative Policy Optimization
+
+- **ID:** `2601.14686` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Shuai Wang, Yaoming Yang, Bingdong Li, et al.
+- **abs:** https://arxiv.org/abs/2601.14686 · **PDF:** https://arxiv.org/pdf/2601.14686v1
+
+## 20307. Local Language Models for Context-Aware Adaptive Anonymization of Sensitive Text
+
+- **ID:** `2601.14683` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Aisvarya Adeseye, Jouni Isoaho, Seppo Virtanen, et al.
+- **abs:** https://arxiv.org/abs/2601.14683 · **PDF:** https://arxiv.org/pdf/2601.14683v1
+
+## 20308. Agentic AI for Commercial Insurance Underwriting with Adversarial Self-Critique
+
+- **ID:** `2602.13213` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Joyjit Roy, Samaresh Kumar Singh
+- **abs:** https://arxiv.org/abs/2602.13213 · **PDF:** https://arxiv.org/pdf/2602.13213v1
+
+## 20309. HCVR Scene Generation: High Compatibility Virtual Reality Environment Generation for Extended Redirected Walking
+
+- **ID:** `2601.14679` · **primary:** `cs.MM` · **Submitted:** 2026-01-21
+- **Authors:** Yiran Zhang, Xingpeng Sun, Aniket Bera
+- **abs:** https://arxiv.org/abs/2601.14679 · **PDF:** https://arxiv.org/pdf/2601.14679v1
+
+## 20310. Transfer Learning from One Cancer to Another via Deep Learning Domain Adaptation
+
+- **ID:** `2601.14678` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Justin Cheung, Samuel Savine, Calvin Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.14678 · **PDF:** https://arxiv.org/pdf/2601.14678v1
+
+## 20311. A comprehensive overview of deep learning models for object detection from videos/images
+
+- **ID:** `2601.14677` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Sukana Zulfqar, Sadia Saeed, M. Azam Zia, et al.
+- **abs:** https://arxiv.org/abs/2601.14677 · **PDF:** https://arxiv.org/pdf/2601.14677v1
+
+## 20312. LaVR: Scene Latent Conditioned Generative Video Trajectory Re-Rendering using Large 4D Reconstruction Models
+
+- **ID:** `2601.14674` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Mingyang Xie, Numair Khan, Tianfu Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.14674 · **PDF:** https://arxiv.org/pdf/2601.14674v2
+
+## 20313. Efficient reformulations of ReLU deep neural networks for surrogate modelling in power system optimisation
+
+- **ID:** `2601.14673` · **primary:** `eess.SY` · **Submitted:** 2026-01-21
+- **Authors:** Yogesh Pipada Sunil Kumar, S. Ali Pourmousavi, Jon A. R. Liisberg, et al.
+- **abs:** https://arxiv.org/abs/2601.14673 · **PDF:** https://arxiv.org/pdf/2601.14673v1
+
+## 20314. GEGO: A Hybrid Golden Eagle and Genetic Optimization Algorithm for Efficient Hyperparameter Tuning in Resource-Constrained Environments
+
+- **ID:** `2601.14672` · **primary:** `cs.NE` · **Submitted:** 2026-01-21
+- **Authors:** Amaras Nazarians, Sachin Kumar
+- **abs:** https://arxiv.org/abs/2601.14672 · **PDF:** https://arxiv.org/pdf/2601.14672v1
+
+## 20315. INFA-Guard: Mitigating Malicious Propagation via Infection-Aware Safeguarding in LLM-Based Multi-Agent Systems
+
+- **ID:** `2601.14667` · **primary:** `cs.MA` · **Submitted:** 2026-01-21
+- **Authors:** Yijin Zhou, Xiaoya Lu, Dongrui Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.14667 · **PDF:** https://arxiv.org/pdf/2601.14667v1
+
+## 20316. Calibrated uncertainty quantification for prosumer flexibility aggregation in ancillary service markets
+
+- **ID:** `2601.14663` · **primary:** `eess.SY` · **Submitted:** 2026-01-21
+- **Authors:** Yogesh Pipada Sunil Kumar, S. Ali Pourmousavi, Jon A. R. Liisberg, et al.
+- **abs:** https://arxiv.org/abs/2601.14663 · **PDF:** https://arxiv.org/pdf/2601.14663v1
+
+## 20317. Query-Efficient Agentic Graph Extraction Attacks on GraphRAG Systems
+
+- **ID:** `2601.14662` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Shuhua Yang, Jiahao Zhang, Yilong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.14662 · **PDF:** https://arxiv.org/pdf/2601.14662v1
+
+## 20318. NeuroFilter: Privacy Guardrails for Conversational LLM Agents
+
+- **ID:** `2601.14660` · **primary:** `cs.CR` · **Submitted:** 2026-01-21
+- **Authors:** Saswat Das, Ferdinando Fioretto
+- **abs:** https://arxiv.org/abs/2601.14660 · **PDF:** https://arxiv.org/pdf/2601.14660v1
+
+## 20319. Say Anything but This: When Tokenizer Betrays Reasoning in LLMs
+
+- **ID:** `2601.14658` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Navid Ayoobi, Marcus I Armstrong, Arjun Mukherjee
+- **abs:** https://arxiv.org/abs/2601.14658 · **PDF:** https://arxiv.org/pdf/2601.14658v1
+
+## 20320. Statistical Reinforcement Learning in the Real World: A Survey of Challenges and Future Directions
+
+- **ID:** `2601.15353` · **primary:** `stat.AP` · **Submitted:** 2026-01-21
+- **Authors:** Asim H. Gazi, Yongyi Guo, Daiqi Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.15353 · **PDF:** https://arxiv.org/pdf/2601.15353v1
+
+## 20321. Efficient Imputation for Patch-based Missing Single-cell Data via Cluster-regularized Optimal Transport
+
+- **ID:** `2601.14653` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Yuyu Liu, Jiannan Yang, Ziyang Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.14653 · **PDF:** https://arxiv.org/pdf/2601.14653v1
+
+## 20322. MAS-Orchestra: Understanding and Improving Multi-Agent Reasoning Through Holistic Orchestration and Controlled Benchmarks
+
+- **ID:** `2601.14652` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Zixuan Ke, Yifei Ming, Austin Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.14652 · **PDF:** https://arxiv.org/pdf/2601.14652v4
+
+## 20323. TRSVR: An Adaptive Stochastic Trust-Region Method with Variance Reduction
+
+- **ID:** `2601.14647` · **primary:** `math.OC` · **Submitted:** 2026-01-21
+- **Authors:** Yuchen Fang, Xinshou Zheng, Javad Lavaei
+- **abs:** https://arxiv.org/abs/2601.14647 · **PDF:** https://arxiv.org/pdf/2601.14647v1
+
+## 20324. Summary of the Unusual Activity Recognition Challenge for Developmental Disability Support
+
+- **ID:** `2601.17049` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Christina Garcia, Nhat Tan Le, Taihei Fujioka, et al.
+- **abs:** https://arxiv.org/abs/2601.17049 · **PDF:** https://arxiv.org/pdf/2601.17049v1
+
+## 20325. OmniSpectra: A Unified Foundation Model for Native Resolution Astronomical Spectra
+
+- **ID:** `2601.15351` · **primary:** `astro-ph.IM` · **Submitted:** 2026-01-21
+- **Authors:** Md Khairul Islam, Judy Fox
+- **abs:** https://arxiv.org/abs/2601.15351 · **PDF:** https://arxiv.org/pdf/2601.15351v1
+
+## 20326. SiMiC: Context-Aware Silicon Microstructure Characterization Using Attention-Based Convolutional Neural Networks for Field-Emission Tip Analysis
+
+- **ID:** `2601.17048` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Jing Jie Tan, Rupert Schreiner, Matthias Hausladen, et al.
+- **abs:** https://arxiv.org/abs/2601.17048 · **PDF:** https://arxiv.org/pdf/2601.17048v1
+
+## 20327. Forest-Chat: Adapting Vision-Language Agents for Interactive Forest Change Analysis
+
+- **ID:** `2601.14637` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** James Brock, Ce Zhang, Nantheera Anantrasirichai
+- **abs:** https://arxiv.org/abs/2601.14637 · **PDF:** https://arxiv.org/pdf/2601.14637v2
+
+## 20328. Dimensional Peeking for Low-Variance Gradients in Zeroth-Order Discrete Optimization via Simulation
+
+- **ID:** `2602.00075` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Philipp Andelfinger, Wentong Cai
+- **abs:** https://arxiv.org/abs/2602.00075 · **PDF:** https://arxiv.org/pdf/2602.00075v1
+
+## 20329. Relational Graph Modeling for Credit Default Prediction: Heterogeneous GNNs and Hybrid Ensemble Learning
+
+- **ID:** `2601.14633` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Yvonne Yang, Eranki Vasistha
+- **abs:** https://arxiv.org/abs/2601.14633 · **PDF:** https://arxiv.org/pdf/2601.14633v1
+
+## 20330. Semi-Supervised Mixture Models under the Concept of Missing at Radom with Margin Confidence and Aranda Ordaz Function
+
+- **ID:** `2601.14631` · **primary:** `stat.ML` · **Submitted:** 2026-01-21
+- **Authors:** Jinyang Liao, Ziyang Lyu
+- **abs:** https://arxiv.org/abs/2601.14631 · **PDF:** https://arxiv.org/pdf/2601.14631v1
+
+## 20331. Online Linear Programming with Replenishment
+
+- **ID:** `2601.14629` · **primary:** `math.OC` · **Submitted:** 2026-01-21
+- **Authors:** Yuze Chen, Yuan Zhou, Baichuan Mo, et al.
+- **abs:** https://arxiv.org/abs/2601.14629 · **PDF:** https://arxiv.org/pdf/2601.14629v1
+
+## 20332. A Brain-inspired Embodied Intelligence for Fluid and Fast Reflexive Robotics Control
+
+- **ID:** `2601.14628` · **primary:** `cs.RO` · **Submitted:** 2026-01-21
+- **Authors:** Weiyu Guo, He Zhang, Pengteng Li, et al.
+- **abs:** https://arxiv.org/abs/2601.14628 · **PDF:** https://arxiv.org/pdf/2601.14628v1
+
+## 20333. Diffusion Epistemic Uncertainty with Asymmetric Learning for Diffusion-Generated Image Detection
+
+- **ID:** `2601.14625` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Yingsong Huang, Hui Guo, Jing Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.14625 · **PDF:** https://arxiv.org/pdf/2601.14625v1
+
+## 20334. Scaling Ambiguity: Augmenting Human Annotation in Speech Emotion Recognition with Audio-Language Models
+
+- **ID:** `2601.14620` · **primary:** `eess.AS` · **Submitted:** 2026-01-21
+- **Authors:** Wenda Zhang, Hongyu Jin, Siyi Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.14620 · **PDF:** https://arxiv.org/pdf/2601.14620v1
+
+## 20335. SearchGym: Bootstrapping Real-World Search Agents via Cost-Effective and High-Fidelity Environment Simulation
+
+- **ID:** `2601.14615` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Xichen Zhang, Ziyi He, Yinghao Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.14615 · **PDF:** https://arxiv.org/pdf/2601.14615v1
+
+## 20336. Communication-Efficient Federated Risk Difference Estimation for Time-to-Event Clinical Outcomes
+
+- **ID:** `2601.14609` · **primary:** `stat.ML` · **Submitted:** 2026-01-21
+- **Authors:** Ziwen Wang, Siqi Li, Marcus Eng Hock Ong, et al.
+- **abs:** https://arxiv.org/abs/2601.14609 · **PDF:** https://arxiv.org/pdf/2601.14609v1
+
+## 20337. Abusive music and song transformation using GenAI and LLMs
+
+- **ID:** `2601.15348` · **primary:** `cs.SD` · **Submitted:** 2026-01-21
+- **Authors:** Jiyang Choi, Rohitash Chandra
+- **abs:** https://arxiv.org/abs/2601.15348 · **PDF:** https://arxiv.org/pdf/2601.15348v1
+
+## 20338. A Contrastive Pre-trained Foundation Model for Deciphering Imaging Noisomics across Modalities
+
+- **ID:** `2601.17047` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Yuanjie Gu, Yiqun Wang, Chaohui Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.17047 · **PDF:** https://arxiv.org/pdf/2601.17047v1
+
+## 20339. U-Harmony: Enhancing Joint Training for Segmentation Models with Universal Harmonization
+
+- **ID:** `2601.14605` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Weiwei Ma, Xiaobing Yu, Peijie Qiu, et al.
+- **abs:** https://arxiv.org/abs/2601.14605 · **PDF:** https://arxiv.org/pdf/2601.14605v1
+
+## 20340. From Biased Chatbots to Biased Agents: Examining Role Assignment Effects on LLM Agent Robustness
+
+- **ID:** `2602.12285` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Linbo Cao, Lihao Sun, Yang Yue
+- **abs:** https://arxiv.org/abs/2602.12285 · **PDF:** https://arxiv.org/pdf/2602.12285v1
+
+## 20341. Variance-Adaptive Muon: Accelerating LLM Pretraining with NSR-Modulated and Variance-Scaled Momentum
+
+- **ID:** `2601.14603` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Jingru Li, Yibo Fan, Huan Li
+- **abs:** https://arxiv.org/abs/2601.14603 · **PDF:** https://arxiv.org/pdf/2601.14603v1
+
+## 20342. Rethinking Reinforcement fine-tuning of LLMs: A Multi-armed Bandit Learning Perspective
+
+- **ID:** `2601.14599` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Xiao Hu, Hong Xie, Tao Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.14599 · **PDF:** https://arxiv.org/pdf/2601.14599v1
+
+## 20343. HELIOS: Hierarchical Graph Abstraction for Structure-Aware LLM Decompilation
+
+- **ID:** `2601.14598` · **primary:** `cs.SE` · **Submitted:** 2026-01-21
+- **Authors:** Yonatan Gizachew Achamyeleh, Harsh Thomare, Mohammad Abdullah Al Faruque
+- **abs:** https://arxiv.org/abs/2601.14598 · **PDF:** https://arxiv.org/pdf/2601.14598v2
+
+## 20344. Optimality of Staircase Mechanisms for Vector Queries under Differential Privacy
+
+- **ID:** `2601.14597` · **primary:** `cs.IT` · **Submitted:** 2026-01-21
+- **Authors:** James Melbourne, Mario Diaz, Shahab Asoodeh
+- **abs:** https://arxiv.org/abs/2601.14597 · **PDF:** https://arxiv.org/pdf/2601.14597v1
+
+## 20345. IntelliSA: An Intelligent Static Analyzer for IaC Security Smell Detection Using Symbolic Rules and Neural Inference
+
+- **ID:** `2601.14595` · **primary:** `cs.CR` · **Submitted:** 2026-01-21
+- **Authors:** Qiyue Mei, Michael Fu
+- **abs:** https://arxiv.org/abs/2601.14595 · **PDF:** https://arxiv.org/pdf/2601.14595v1
+
+## 20346. From Volumes to Slices: Computationally Efficient Contrastive Learning for Sequential Abdominal CT Analysis
+
+- **ID:** `2601.14593` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Po-Kai Chiu, Hung-Hsuan Chen
+- **abs:** https://arxiv.org/abs/2601.14593 · **PDF:** https://arxiv.org/pdf/2601.14593v1
+
+## 20347. A Lightweight LLM Framework for Disaster Humanitarian Information Classification
+
+- **ID:** `2602.12284` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Han Jinzhen, Kim Jisung, Yang Jong Soo, et al.
+- **abs:** https://arxiv.org/abs/2602.12284 · **PDF:** https://arxiv.org/pdf/2602.12284v1
+
+## 20348. Counterfactual Modeling with Fine-Tuned LLMs for Health Intervention Design and Sensor Data Augmentation
+
+- **ID:** `2601.14590` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Shovito Barua Soumma, Asiful Arefeen, Stephanie M. Carpenter, et al.
+- **abs:** https://arxiv.org/abs/2601.14590 · **PDF:** https://arxiv.org/pdf/2601.14590v1
+
+## 20349. Designing KRIYA: An AI Companion for Wellbeing Self-Reflection
+
+- **ID:** `2601.14589` · **primary:** `cs.HC` · **Submitted:** 2026-01-21
+- **Authors:** Shanshan Zhu, Wenxuan Song, Jiayue Melissa Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.14589 · **PDF:** https://arxiv.org/pdf/2601.14589v1
+
+## 20350. Logic Programming on Knowledge Graph Networks And its Application in Medical Domain
+
+- **ID:** `2601.15347` · **primary:** `cs.AI` · **Submitted:** 2026-01-21
+- **Authors:** Chuanqing Wang, Zhenmin Zhao, Shanshan Du, et al.
+- **abs:** https://arxiv.org/abs/2601.15347 · **PDF:** https://arxiv.org/pdf/2601.15347v1
+
+## 20351. Adoption and Use of LLMs at an Academic Medical Center
+
+- **ID:** `2602.00074` · **primary:** `cs.CY` · **Submitted:** 2026-01-21
+- **Authors:** Nigam H. Shah, Nerissa Ambers, Abby Pandya, et al.
+- **abs:** https://arxiv.org/abs/2602.00074 · **PDF:** https://arxiv.org/pdf/2602.00074v1
+
+## 20352. Place with Intention: An Empirical Attendance Predictive Study of Expo 2025 Osaka, Kansai, Japan
+
+- **ID:** `2601.14570` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Xiaojie Yang, Dizhi Huang, Hangli Ge, et al.
+- **abs:** https://arxiv.org/abs/2601.14570 · **PDF:** https://arxiv.org/pdf/2601.14570v1
+
+## 20353. Social Caption: Evaluating Social Understanding in Multimodal Models
+
+- **ID:** `2601.14569` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Bhaavanaa Thumu, Leena Mathur, Youssouf Kebe, et al.
+- **abs:** https://arxiv.org/abs/2601.14569 · **PDF:** https://arxiv.org/pdf/2601.14569v1
+
+## 20354. Breaking the accuracy-resource dilemma: a lightweight adaptive video inference enhancement
+
+- **ID:** `2601.14568` · **primary:** `cs.CV` · **Submitted:** 2026-01-21
+- **Authors:** Wei Ma, Shaowu Chen, Junjie Ye, et al.
+- **abs:** https://arxiv.org/abs/2601.14568 · **PDF:** https://arxiv.org/pdf/2601.14568v1
+
+## 20355. Rewarding How Models Think Pedagogically: Integrating Pedagogical Reasoning and Thinking Rewards for LLMs in Education
+
+- **ID:** `2601.14560` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Unggi Lee, Jiyeong Bae, Jaehyeon Park, et al.
+- **abs:** https://arxiv.org/abs/2601.14560 · **PDF:** https://arxiv.org/pdf/2601.14560v1
+
+## 20356. Constructing Multi-label Hierarchical Classification Models for MITRE ATT&CK Text Tagging
+
+- **ID:** `2601.14556` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Andrew Crossman, Jonah Dodd, Viralam Ramamurthy Chaithanya Kumar, et al.
+- **abs:** https://arxiv.org/abs/2601.14556 · **PDF:** https://arxiv.org/pdf/2601.14556v1
+
+## 20357. From Toil to Thought: Designing for Strategic Exploration and Responsible AI in Systematic Literature Reviews
+
+- **ID:** `2603.05514` · **primary:** `cs.HC` · **Submitted:** 2026-01-21
+- **Authors:** Runlong Ye, Naaz Sibia, Angela Zavaleta Bernuy, et al.
+- **abs:** https://arxiv.org/abs/2603.05514 · **PDF:** https://arxiv.org/pdf/2603.05514v2
+
+## 20358. Self-Blinding and Counterfactual Self-Simulation Mitigate Biases and Sycophancy in Large Language Models
+
+- **ID:** `2601.14553` · **primary:** `cs.CL` · **Submitted:** 2026-01-21
+- **Authors:** Brian Christian, Matan Mazor
+- **abs:** https://arxiv.org/abs/2601.14553 · **PDF:** https://arxiv.org/pdf/2601.14553v1
+
+## 20359. QMC: Efficient SLM Edge Inference via Outlier-Aware Quantization and Emergent Memories Co-Design
+
+- **ID:** `2601.14549` · **primary:** `cs.LG` · **Submitted:** 2026-01-21
+- **Authors:** Nilesh Prasad Pandey, Jangseon Park, Onat Gungor, et al.
+- **abs:** https://arxiv.org/abs/2601.14549 · **PDF:** https://arxiv.org/pdf/2601.14549v1
+
+## 20360. Report for NSF Workshop on AI for Electronic Design Automation
+
+- **ID:** `2601.14541` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Deming Chen, Vijay Ganesh, Weikai Li, et al.
+- **abs:** https://arxiv.org/abs/2601.14541 · **PDF:** https://arxiv.org/pdf/2601.14541v3
+
+## 20361. engGNN: A Dual-Graph Neural Network for Omics-Based Disease Classification and Feature Selection
+
+- **ID:** `2601.14536` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Tiantian Yang, Yuxuan Wang, Zhenwei Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.14536 · **PDF:** https://arxiv.org/pdf/2601.14536v1
+
+## 20362. Search over Self-Edit Strategies for LLM Adaptation
+
+- **ID:** `2601.14532` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Alistair Cheong, Haolin Cong, Tyler Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.14532 · **PDF:** https://arxiv.org/pdf/2601.14532v1
+
+## 20363. LLM Security and Safety: Insights from Homotopy-Inspired Prompt Obfuscation
+
+- **ID:** `2601.14528` · **primary:** `cs.CR` · **Submitted:** 2026-01-20
+- **Authors:** Luis Lazo, Hamed Jelodar, Roozbeh Razavi-Far
+- **abs:** https://arxiv.org/abs/2601.14528 · **PDF:** https://arxiv.org/pdf/2601.14528v1
+
+## 20364. Policy-Embedded Graph Expansion: Networked HIV Testing with Diffusion-Driven Network Samples
+
+- **ID:** `2601.16233` · **primary:** `cs.SI` · **Submitted:** 2026-01-20
+- **Authors:** Akseli Kangaslahti, Davin Choo, Lingkai Kong, et al.
+- **abs:** https://arxiv.org/abs/2601.16233 · **PDF:** https://arxiv.org/pdf/2601.16233v1
+
+## 20365. Towards Execution-Grounded Automated AI Research
+
+- **ID:** `2601.14525` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Chenglei Si, Zitong Yang, Yejin Choi, et al.
+- **abs:** https://arxiv.org/abs/2601.14525 · **PDF:** https://arxiv.org/pdf/2601.14525v1
+
+## 20366. Large Language Model-Powered Evolutionary Code Optimization on a Phylogenetic Tree
+
+- **ID:** `2601.14523` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Leyi Zhao, Weijie Huang, Yitong Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.14523 · **PDF:** https://arxiv.org/pdf/2601.14523v1
+
+## 20367. Test-Time Adaptation for Non-stationary Time Series: From Synthetic Regime Shifts to Financial Markets
+
+- **ID:** `2602.00073` · **primary:** `q-fin.ST` · **Submitted:** 2026-01-20
+- **Authors:** Yurui Wu, Qingying Deng, Wonou Chung, et al.
+- **abs:** https://arxiv.org/abs/2602.00073 · **PDF:** https://arxiv.org/pdf/2602.00073v1
+
+## 20368. On the Runway Cascade of Transformers for Language Modeling
+
+- **ID:** `2601.14522` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Hunjae Lee, Corey Clark
+- **abs:** https://arxiv.org/abs/2601.14522 · **PDF:** https://arxiv.org/pdf/2601.14522v1
+
+## 20369. How Worst-Case Are Adversarial Attacks? Linking Adversarial and Perturbation Robustness
+
+- **ID:** `2601.14519` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Giulio Rossolini
+- **abs:** https://arxiv.org/abs/2601.14519 · **PDF:** https://arxiv.org/pdf/2601.14519v2
+
+## 20370. Business Logic-Driven Text-to-SQL Data Synthesis for Business Intelligence
+
+- **ID:** `2601.14518` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Jinhui Liu, Ximeng Zhang, Yanbo Ai, et al.
+- **abs:** https://arxiv.org/abs/2601.14518 · **PDF:** https://arxiv.org/pdf/2601.14518v1
+
+## 20371. Learning PDE Solvers with Physics and Data: A Unifying View of Physics-Informed Neural Networks and Neural Operators
+
+- **ID:** `2601.14517` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Yilong Dai, Shengyu Chen, Ziyi Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.14517 · **PDF:** https://arxiv.org/pdf/2601.14517v2
+
+## 20372. Large Data Limits of Laplace Learning for Gaussian Measure Data in Infinite Dimensions
+
+- **ID:** `2601.14515` · **primary:** `stat.ML` · **Submitted:** 2026-01-20
+- **Authors:** Zhengang Zhong, Yury Korolev, Matthew Thorpe
+- **abs:** https://arxiv.org/abs/2601.14515 · **PDF:** https://arxiv.org/pdf/2601.14515v1
+
+## 20373. What Do LLMs Know About Alzheimer's Disease? Fine-Tuning, Probing, and Data Synthesis for AD Detection
+
+- **ID:** `2602.11177` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Lei Jiang, Yue Zhou, Natalie Parde
+- **abs:** https://arxiv.org/abs/2602.11177 · **PDF:** https://arxiv.org/pdf/2602.11177v1
+
+## 20374. "Just in Time" World Modeling Supports Human Planning and Reasoning
+
+- **ID:** `2601.14514` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Tony Chen, Sam Cheyette, Kelsey Allen, et al.
+- **abs:** https://arxiv.org/abs/2601.14514 · **PDF:** https://arxiv.org/pdf/2601.14514v1
+
+## 20375. Compounding Disadvantage: Auditing Intersectional Bias in LLM-Generated Explanations Across Indian and American STEM Education
+
+- **ID:** `2601.14506` · **primary:** `cs.CY` · **Submitted:** 2026-01-20
+- **Authors:** Amogh Gupta, Niharika Patil, Sourojit Ghosh, et al.
+- **abs:** https://arxiv.org/abs/2601.14506 · **PDF:** https://arxiv.org/pdf/2601.14506v2
+
+## 20376. Uncovering and Understanding FPR Manipulation Attack in Industrial IoT Networks
+
+- **ID:** `2601.14505` · **primary:** `cs.CR` · **Submitted:** 2026-01-20
+- **Authors:** Mohammad Shamim Ahsan, Peng Liu
+- **abs:** https://arxiv.org/abs/2601.14505 · **PDF:** https://arxiv.org/pdf/2601.14505v1
+
+## 20377. GutenOCR: A Grounded Vision-Language Front-End for Documents
+
+- **ID:** `2601.14490` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Hunter Heidenreich, Ben Elliott, Olivia Dinica, et al.
+- **abs:** https://arxiv.org/abs/2601.14490 · **PDF:** https://arxiv.org/pdf/2601.14490v2
+
+## 20378. Translational Gaps in Graph Transformers for Longitudinal EHR Prediction: A Critical Appraisal of GT-BEHRT
+
+- **ID:** `2603.13231` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Krish Tadigotla
+- **abs:** https://arxiv.org/abs/2603.13231 · **PDF:** https://arxiv.org/pdf/2603.13231v1
+
+## 20379. Stabilizing autoregressive forecasts in chaotic systems via multi-rate latent recurrence
+
+- **ID:** `2601.14487` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Mrigank Dhingra, Omer San
+- **abs:** https://arxiv.org/abs/2601.14487 · **PDF:** https://arxiv.org/pdf/2601.14487v1
+
+## 20380. Scalable Knee-Point Guided Activity Group Selection in Multi-Tree Genetic Programming for Dynamic Multi-Mode Project Scheduling
+
+- **ID:** `2601.14485` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Yuan Tian, Yi Mei, Mengjie Zhang
+- **abs:** https://arxiv.org/abs/2601.14485 · **PDF:** https://arxiv.org/pdf/2601.14485v1
+
+## 20381. Can LLM Reasoning Be Trusted? A Comparative Study: Using Human Benchmarking on Statistical Tasks
+
+- **ID:** `2601.14479` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Crish Nagarkar, Leonid Bogachev, Serge Sharoff
+- **abs:** https://arxiv.org/abs/2601.14479 · **PDF:** https://arxiv.org/pdf/2601.14479v1
+
+## 20382. Large Language Models for Large-Scale, Rigorous Qualitative Analysis in Applied Health Services Research
+
+- **ID:** `2601.14478` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Sasha Ronaghi, Emma-Louise Aveling, Maria Levis, et al.
+- **abs:** https://arxiv.org/abs/2601.14478 · **PDF:** https://arxiv.org/pdf/2601.14478v1
+
+## 20383. XD-MAP: Cross-Modal Domain Adaptation using Semantic Parametric Mapping
+
+- **ID:** `2601.14477` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Frank Bieder, Hendrik Königshof, Haohao Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.14477 · **PDF:** https://arxiv.org/pdf/2601.14477v1
+
+## 20384. GPU-accelerated simulated annealing based on p-bits with real-world device-variability modeling
+
+- **ID:** `2601.14476` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Naoya Onizawa, Takahiro Hanyu
+- **abs:** https://arxiv.org/abs/2601.14476 · **PDF:** https://arxiv.org/pdf/2601.14476v1
+
+## 20385. Evaluating Few-Shot Temporal Reasoning of LLMs for Human Activity Prediction in Smart Environments
+
+- **ID:** `2602.11176` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Maral Doctorarastoo, Katherine A. Flanigan, Mario Bergés, et al.
+- **abs:** https://arxiv.org/abs/2602.11176 · **PDF:** https://arxiv.org/pdf/2602.11176v1
+
+## 20386. Real-Time Wildfire Localization on the NASA Autonomous Modular Sensor using Deep Learning
+
+- **ID:** `2601.14475` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Yajvan Ravan, Aref Malek, Chester Dolph, et al.
+- **abs:** https://arxiv.org/abs/2601.14475 · **PDF:** https://arxiv.org/pdf/2601.14475v1
+
+## 20387. Adaptive KDE for Real-Time Thresholding: Prioritized Queues for Financial Crime Investigation
+
+- **ID:** `2601.14473` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Danny Butvinik, Nana Boateng, Achi Hackmon
+- **abs:** https://arxiv.org/abs/2601.14473 · **PDF:** https://arxiv.org/pdf/2601.14473v2
+
+## 20388. Prosody-Guided Harmonic Attention for Phase-Coherent Neural Vocoding in the Complex Spectrum
+
+- **ID:** `2601.14472` · **primary:** `cs.SD` · **Submitted:** 2026-01-20
+- **Authors:** Mohammed Salah Al-Radhi, Riad Larbi, Mátyás Bartalis, et al.
+- **abs:** https://arxiv.org/abs/2601.14472 · **PDF:** https://arxiv.org/pdf/2601.14472v1
+
+## 20389. Tokenomics: Quantifying Where Tokens Are Used in Agentic Software Engineering
+
+- **ID:** `2601.14470` · **primary:** `cs.SE` · **Submitted:** 2026-01-20
+- **Authors:** Mohamad Salim, Jasmine Latendresse, SayedHassan Khatoonabadi, et al.
+- **abs:** https://arxiv.org/abs/2601.14470 · **PDF:** https://arxiv.org/pdf/2601.14470v1
+
+## 20390. On the Generalization Gap in LLM Planning: Tests and Verifier-Reward RL
+
+- **ID:** `2601.14456` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Valerio Belcamino, Nicholas Attolino, Alessio Capitanelli, et al.
+- **abs:** https://arxiv.org/abs/2601.14456 · **PDF:** https://arxiv.org/pdf/2601.14456v1
+
+## 20391. Generative AI-enhanced Probabilistic Multi-Fidelity Surrogate Modeling Via Transfer Learning
+
+- **ID:** `2602.00072` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Jice Zeng, David Barajas-Solano, Hui Chen
+- **abs:** https://arxiv.org/abs/2602.00072 · **PDF:** https://arxiv.org/pdf/2602.00072v1
+
+## 20392. Diffusion Large Language Models for Black-Box Optimization
+
+- **ID:** `2601.14446` · **primary:** `cs.CE` · **Submitted:** 2026-01-20
+- **Authors:** Ye Yuan, Can, Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.14446 · **PDF:** https://arxiv.org/pdf/2601.14446v1
+
+## 20393. VisTIRA: Closing the Image-Text Modality Gap in Visual Math Reasoning via Structured Tool Integration
+
+- **ID:** `2601.14440` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Saeed Khaki, Ashudeep Singh, Nima Safaei, et al.
+- **abs:** https://arxiv.org/abs/2601.14440 · **PDF:** https://arxiv.org/pdf/2601.14440v2
+
+## 20394. Vision-Based Natural Language Scene Understanding for Autonomous Driving: An Extended Dataset and a New Model for Traffic Scene Description Generation
+
+- **ID:** `2601.14438` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Danial Sadrian Zadeh, Otman A. Basir, Behzad Moshiri
+- **abs:** https://arxiv.org/abs/2601.14438 · **PDF:** https://arxiv.org/pdf/2601.14438v1
+
+## 20395. Agentic AI Meets Edge Computing in Autonomous UAV Swarms
+
+- **ID:** `2601.14437` · **primary:** `cs.RO` · **Submitted:** 2026-01-20
+- **Authors:** Thuan Minh Nguyen, Vu Tuan Truong, Long Bao Le
+- **abs:** https://arxiv.org/abs/2601.14437 · **PDF:** https://arxiv.org/pdf/2601.14437v1
+
+## 20396. A full process algebraic representation of Ant Colony Optimization
+
+- **ID:** `2601.14436` · **primary:** `cs.NE` · **Submitted:** 2026-01-20
+- **Authors:** Maria Garcia, Natalia Lopez, Ismael Rodriguez
+- **abs:** https://arxiv.org/abs/2601.14436 · **PDF:** https://arxiv.org/pdf/2601.14436v1
+
+## 20397. Quantum Super-resolution by Adaptive Non-local Observables
+
+- **ID:** `2601.14433` · **primary:** `quant-ph` · **Submitted:** 2026-01-20
+- **Authors:** Hsin-Yi Lin, Huan-Hsin Tseng, Samuel Yen-Chi Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.14433 · **PDF:** https://arxiv.org/pdf/2601.14433v2
+
+## 20398. Meta Flow Maps enable scalable reward alignment
+
+- **ID:** `2601.14430` · **primary:** `stat.ML` · **Submitted:** 2026-01-20
+- **Authors:** Peter Potaptchik, Adhi Saravanan, Abbas Mammadov, et al.
+- **abs:** https://arxiv.org/abs/2601.14430 · **PDF:** https://arxiv.org/pdf/2601.14430v1
+
+## 20399. Measuring the State of Open Science in Transportation Using Large Language Models
+
+- **ID:** `2601.14429` · **primary:** `cs.DL` · **Submitted:** 2026-01-20
+- **Authors:** Junyi Ji, Ruth Lu, Linda Belkessa, et al.
+- **abs:** https://arxiv.org/abs/2601.14429 · **PDF:** https://arxiv.org/pdf/2601.14429v1
+
+## 20400. Atomic Depth Estimation From Noisy Electron Microscopy Data Via Deep Learning
+
+- **ID:** `2601.17046` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Matan Leibovich, Mai Tan, Ramon Manzorro, et al.
+- **abs:** https://arxiv.org/abs/2601.17046 · **PDF:** https://arxiv.org/pdf/2601.17046v3
+
+## 20401. Quantifying Speaker Embedding Phonological Rule Interactions in Accented Speech Synthesis
+
+- **ID:** `2601.14417` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Thanathai Lertpetchpun, Yoonjeong Lee, Thanapat Trachu, et al.
+- **abs:** https://arxiv.org/abs/2601.14417 · **PDF:** https://arxiv.org/pdf/2601.14417v2
+
+## 20402. Recursivism: An Artistic Paradigm for Self-Transforming Art in the Age of AI
+
+- **ID:** `2601.14401` · **primary:** `cs.CY` · **Submitted:** 2026-01-20
+- **Authors:** Florentin Koch
+- **abs:** https://arxiv.org/abs/2601.14401 · **PDF:** https://arxiv.org/pdf/2601.14401v1
+
+## 20403. Cosmo-FOLD: Fast generation and upscaling of field-level cosmological maps with overlap latent diffusion
+
+- **ID:** `2601.14377` · **primary:** `astro-ph.CO` · **Submitted:** 2026-01-20
+- **Authors:** Satvik Mishra, Roberto Trotta, Matteo Viel
+- **abs:** https://arxiv.org/abs/2601.14377 · **PDF:** https://arxiv.org/pdf/2601.14377v1
+
+## 20404. VideoMaMa: Mask-Guided Video Matting via Generative Prior
+
+- **ID:** `2601.14255` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Sangbeom Lim, Seoung Wug Oh, Jiahui Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.14255 · **PDF:** https://arxiv.org/pdf/2601.14255v1
+
+## 20405. Which Reasoning Trajectories Teach Students to Reason Better? A Simple Metric of Informative Alignment
+
+- **ID:** `2601.14249` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Yuming Yang, Mingyoung Lai, Wanxu Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.14249 · **PDF:** https://arxiv.org/pdf/2601.14249v3
+
+## 20406. Jet-RL: Enabling On-Policy FP8 Reinforcement Learning with Unified Training and Rollout Precision Flow
+
+- **ID:** `2601.14243` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Haocheng Xi, Charlie Ruan, Peiyuan Liao, et al.
+- **abs:** https://arxiv.org/abs/2601.14243 · **PDF:** https://arxiv.org/pdf/2601.14243v2
+
+## 20407. APEX-Agents
+
+- **ID:** `2601.14242` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Bertie Vidgen, Austin Mann, Abby Fennelly, et al.
+- **abs:** https://arxiv.org/abs/2601.14242 · **PDF:** https://arxiv.org/pdf/2601.14242v3
+
+## 20408. SoundBreak: A Systematic Study of Audio-Only Adversarial Attacks on Trimodal Models
+
+- **ID:** `2601.16231` · **primary:** `cs.SD` · **Submitted:** 2026-01-20
+- **Authors:** Aafiya Hussain, Gaurav Srivastava, Alvi Ishmam, et al.
+- **abs:** https://arxiv.org/abs/2601.16231 · **PDF:** https://arxiv.org/pdf/2601.16231v1
+
+## 20409. Spatiotemporal Wildfire Prediction and Reinforcement Learning for Helitack Suppression
+
+- **ID:** `2601.14238` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Shaurya Mathur, Shreyas Bellary Manjunath, Nitin Kulkarni, et al.
+- **abs:** https://arxiv.org/abs/2601.14238 · **PDF:** https://arxiv.org/pdf/2601.14238v1
+
+## 20410. Opportunities in AI/ML for the Rubin LSST Dark Energy Science Collaboration
+
+- **ID:** `2601.14235` · **primary:** `astro-ph.IM` · **Submitted:** 2026-01-20
+- **Authors:** LSST Dark Energy Science Collaboration, Eric Aubourg, Camille Avestruz, et al.
+- **abs:** https://arxiv.org/abs/2601.14235 · **PDF:** https://arxiv.org/pdf/2601.14235v1
+
+## 20411. Q-learning with Adjoint Matching
+
+- **ID:** `2601.14234` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Qiyang Li, Sergey Levine
+- **abs:** https://arxiv.org/abs/2601.14234 · **PDF:** https://arxiv.org/pdf/2601.14234v2
+
+## 20412. KAGE-Bench: Fast Known-Axis Visual Generalization Evaluation for Reinforcement Learning
+
+- **ID:** `2601.14232` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Egor Cherepanov, Daniil Zelezetsky, Alexey K. Kovalev, et al.
+- **abs:** https://arxiv.org/abs/2601.14232 · **PDF:** https://arxiv.org/pdf/2601.14232v1
+
+## 20413. MASCOT: Towards Multi-Agent Socio-Collaborative Companion Systems
+
+- **ID:** `2601.14230` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Yiyang Wang, Yiqiao Jin, Alex Cabral, et al.
+- **abs:** https://arxiv.org/abs/2601.14230 · **PDF:** https://arxiv.org/pdf/2601.14230v1
+
+## 20414. Attention-Based Offline Reinforcement Learning and Clustering for Interpretable Sepsis Treatment
+
+- **ID:** `2601.14228` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Punit Kumar, Vaibhav Saran, Divyesh Patel, et al.
+- **abs:** https://arxiv.org/abs/2601.14228 · **PDF:** https://arxiv.org/pdf/2601.14228v1
+
+## 20415. Deep Learning Approaches to Quantum Error Mitigation
+
+- **ID:** `2601.14226` · **primary:** `quant-ph` · **Submitted:** 2026-01-20
+- **Authors:** Leonardo Placidi, Ifan Williams, Enrico Rinaldi, et al.
+- **abs:** https://arxiv.org/abs/2601.14226 · **PDF:** https://arxiv.org/pdf/2601.14226v1
+
+## 20416. Generalization and Completeness of Stochastic Local Search Algorithms
+
+- **ID:** `2601.14212` · **primary:** `cs.NE` · **Submitted:** 2026-01-20
+- **Authors:** Daniel Loscos, Narciso Marti-Oliet, Ismael Rodriguez
+- **abs:** https://arxiv.org/abs/2601.14212 · **PDF:** https://arxiv.org/pdf/2601.14212v1
+
+## 20417. DRIFT: Detecting Representational Inconsistencies for Factual Truthfulness
+
+- **ID:** `2601.14210` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Rohan Bhatnagar, Youran Sun, Chi Andrew Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.14210 · **PDF:** https://arxiv.org/pdf/2601.14210v2
+
+## 20418. InT: Self-Proposed Interventions Enable Credit Assignment in LLM Reasoning
+
+- **ID:** `2601.14209` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Matthew Y. R. Yang, Hao Bai, Ian Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.14209 · **PDF:** https://arxiv.org/pdf/2601.14209v1
+
+## 20419. Rig-Aware 3D Reconstruction of Vehicle Undercarriages using Gaussian Splatting
+
+- **ID:** `2601.14208` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Nitin Kulkarni, Akhil Devarashetti, Charlie Cluss, et al.
+- **abs:** https://arxiv.org/abs/2601.14208 · **PDF:** https://arxiv.org/pdf/2601.14208v1
+
+## 20420. VJEPA: Variational Joint Embedding Predictive Architectures as Probabilistic World Models
+
+- **ID:** `2601.14354` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Yongchao Huang
+- **abs:** https://arxiv.org/abs/2601.14354 · **PDF:** https://arxiv.org/pdf/2601.14354v1
+
+## 20421. Differentiated Pickup Point Offering for Emission Reduction in Last-Mile Delivery
+
+- **ID:** `2601.14196` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Albina Galiullina, Wouter van Heeswijk, Tom van Woensel
+- **abs:** https://arxiv.org/abs/2601.14196 · **PDF:** https://arxiv.org/pdf/2601.14196v1
+
+## 20422. Toward Efficient Agents: Memory, Tool learning, and Planning
+
+- **ID:** `2601.14192` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Xiaofang Yang, Lijun Li, Heng Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.14192 · **PDF:** https://arxiv.org/pdf/2601.14192v1
+
+## 20423. FoundationalASSIST: An Educational Dataset for Foundational Knowledge Tracing and Pedagogical Grounding of LLMs
+
+- **ID:** `2602.00070` · **primary:** `cs.CY` · **Submitted:** 2026-01-20
+- **Authors:** Eamon Worden, Cristina Heffernan, Neil Heffernan, et al.
+- **abs:** https://arxiv.org/abs/2602.00070 · **PDF:** https://arxiv.org/pdf/2602.00070v1
+
+## 20424. A model of errors in transformers
+
+- **ID:** `2601.14175` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Suvrat Raju, Praneeth Netrapalli
+- **abs:** https://arxiv.org/abs/2601.14175 · **PDF:** https://arxiv.org/pdf/2601.14175v1
+
+## 20425. Penalizing Localized Dirichlet Energies in Low Rank Tensor Products
+
+- **ID:** `2601.14173` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Paris A. Karakasis, Nicholas D. Sidiropoulos
+- **abs:** https://arxiv.org/abs/2601.14173 · **PDF:** https://arxiv.org/pdf/2601.14173v1
+
+## 20426. Human Values in a Single Sentence: Moral Presence, Hierarchies, and Transformer Ensembles on the Schwartz Continuum
+
+- **ID:** `2601.14172` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Víctor Yeste, Paolo Rosso
+- **abs:** https://arxiv.org/abs/2601.14172 · **PDF:** https://arxiv.org/pdf/2601.14172v3
+
+## 20427. Paper2Rebuttal: A Multi-Agent Framework for Transparent Author Response Assistance
+
+- **ID:** `2601.14171` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Qianli Ma, Chang Guo, Zhiheng Tian, et al.
+- **abs:** https://arxiv.org/abs/2601.14171 · **PDF:** https://arxiv.org/pdf/2601.14171v1
+
+## 20428. If You Want Coherence, Orchestrate a Team of Rivals: Multi-Agent Models of Organizational Intelligence
+
+- **ID:** `2601.14351` · **primary:** `cs.MA` · **Submitted:** 2026-01-20
+- **Authors:** Gopal Vijayaraghavan, Prasanth Jayachandran, Arun Murthy, et al.
+- **abs:** https://arxiv.org/abs/2601.14351 · **PDF:** https://arxiv.org/pdf/2601.14351v1
+
+## 20429. Domain-Adaptation through Synthetic Data: Fine-Tuning Large Language Models for German Law
+
+- **ID:** `2601.14160` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Ali Hamza Bashir, Muhammad Rehan Khalid, Kostadin Cvejoski, et al.
+- **abs:** https://arxiv.org/abs/2601.14160 · **PDF:** https://arxiv.org/pdf/2601.14160v1
+
+## 20430. MARBLE: Multi-Agent Reasoning for Bioinformatics Learning and Evolution
+
+- **ID:** `2601.14349` · **primary:** `cs.MA` · **Submitted:** 2026-01-20
+- **Authors:** Sunghyun Kim, Seokwoo Yun, Youngseo Yun, et al.
+- **abs:** https://arxiv.org/abs/2601.14349 · **PDF:** https://arxiv.org/pdf/2601.14349v1
+
+## 20431. ConceptCaps: a Distilled Concept Dataset for Interpretability in Music Models
+
+- **ID:** `2601.14157` · **primary:** `cs.SD` · **Submitted:** 2026-01-20
+- **Authors:** Bruno Sienkiewicz, Łukasz Neumann, Mateusz Modrzejewski
+- **abs:** https://arxiv.org/abs/2601.14157 · **PDF:** https://arxiv.org/pdf/2601.14157v3
+
+## 20432. LLM Augmented Intervenable Multimodal Adaptor for Post-operative Complication Prediction in Lung Cancer Surgery
+
+- **ID:** `2601.14154` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Shubham Pandey, Bhavin Jawade, Srirangaraj Setlur, et al.
+- **abs:** https://arxiv.org/abs/2601.14154 · **PDF:** https://arxiv.org/pdf/2601.14154v1
+
+## 20433. Lost in the Prompt Order: Revealing the Limitations of Causal Attention in Language Models
+
+- **ID:** `2601.14152` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Hyunjong Ok, Jaeho Lee
+- **abs:** https://arxiv.org/abs/2601.14152 · **PDF:** https://arxiv.org/pdf/2601.14152v1
+
+## 20434. DiSPA: Differential Substructure-Pathway Attention for Drug Response Prediction
+
+- **ID:** `2601.14346` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Yewon Han, Sunghyun Kim, Eunyi Jeong, et al.
+- **abs:** https://arxiv.org/abs/2601.14346 · **PDF:** https://arxiv.org/pdf/2601.14346v1
+
+## 20435. The Side Effects of Being Smart: Safety Risks in MLLMs' Multi-Image Reasoning
+
+- **ID:** `2601.14127` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Renmiao Chen, Yida Lu, Shiyao Cui, et al.
+- **abs:** https://arxiv.org/abs/2601.14127 · **PDF:** https://arxiv.org/pdf/2601.14127v1
+
+## 20436. Style Transfer as Bias Mitigation: Diffusion Models for Synthetic Mental Health Text for Arabic
+
+- **ID:** `2601.14124` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Saad Mankarious, Aya Zirikly
+- **abs:** https://arxiv.org/abs/2601.14124 · **PDF:** https://arxiv.org/pdf/2601.14124v1
+
+## 20437. A Systematic Analysis of Chunking Strategies for Reliable Question Answering
+
+- **ID:** `2601.14123` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Sofia Bennani, Charles Moslonka
+- **abs:** https://arxiv.org/abs/2601.14123 · **PDF:** https://arxiv.org/pdf/2601.14123v1
+
+## 20438. NewsRECON: News article REtrieval for image CONtextualization
+
+- **ID:** `2601.14121` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Jonathan Tonglet, Iryna Gurevych, Tinne Tuytelaars, et al.
+- **abs:** https://arxiv.org/abs/2601.14121 · **PDF:** https://arxiv.org/pdf/2601.14121v1
+
+## 20439. Latent Causal Diffusions for Single-Cell Perturbation Modeling
+
+- **ID:** `2601.15341` · **primary:** `q-bio.MN` · **Submitted:** 2026-01-20
+- **Authors:** Lars Lorch, Jiaqi Zhang, Charlotte Bunne, et al.
+- **abs:** https://arxiv.org/abs/2601.15341 · **PDF:** https://arxiv.org/pdf/2601.15341v1
+
+## 20440. Riemannian Liquid Spatio-Temporal Graph Network
+
+- **ID:** `2601.14115` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Liangsi Lu, Jingchao Wang, Zhaorong Dai, et al.
+- **abs:** https://arxiv.org/abs/2601.14115 · **PDF:** https://arxiv.org/pdf/2601.14115v1
+
+## 20441. Learning to Explain: Supervised Token Attribution from Transformer Attention Patterns
+
+- **ID:** `2601.14112` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** George Mihaila
+- **abs:** https://arxiv.org/abs/2601.14112 · **PDF:** https://arxiv.org/pdf/2601.14112v2
+
+## 20442. Truth with a Twist: The Rhetoric of Persuasion in Professional vs. Community-Authored Fact-Checks
+
+- **ID:** `2601.14105` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Olesya Razuvayevskaya, Kalina Bontcheva
+- **abs:** https://arxiv.org/abs/2601.14105 · **PDF:** https://arxiv.org/pdf/2601.14105v3
+
+## 20443. Learning Nonlinear Heterogeneity in Physical Kolmogorov-Arnold Networks
+
+- **ID:** `2601.15340` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-01-20
+- **Authors:** Fabiana Taglietti, Andrea Pulici, Maxwell Roxburgh, et al.
+- **abs:** https://arxiv.org/abs/2601.15340 · **PDF:** https://arxiv.org/pdf/2601.15340v3
+
+## 20444. Causal feature selection framework for stable soft sensor modeling based on time-delayed cross mapping
+
+- **ID:** `2601.14099` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Shi-Shun Chen, Xiao-Yang Li, Enrico Zio
+- **abs:** https://arxiv.org/abs/2601.14099 · **PDF:** https://arxiv.org/pdf/2601.14099v1
+
+## 20445. Sparse Adapter Fusion for Continual Learning in NLP
+
+- **ID:** `2602.02502` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Min Zeng, Xi Chen, Haiqin Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.02502 · **PDF:** https://arxiv.org/pdf/2602.02502v1
+
+## 20446. Remapping and navigation of an embedding space via error minimization: a fundamental organizational principle of cognition in natural and artificial systems
+
+- **ID:** `2601.14096` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Benedikt Hartl, Léo Pio-Lopez, Chris Fields, et al.
+- **abs:** https://arxiv.org/abs/2601.14096 · **PDF:** https://arxiv.org/pdf/2601.14096v2
+
+## 20447. Optimizing Energy and Data Collection in UAV-aided IoT Networks using Attention-based Multi-Objective Reinforcement Learning
+
+- **ID:** `2601.14092` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Babacar Toure, Dimitrios Tsilimantos, Omid Esrafilian, et al.
+- **abs:** https://arxiv.org/abs/2601.14092 · **PDF:** https://arxiv.org/pdf/2601.14092v1
+
+## 20448. Zero-shot adaptable task planning for autonomous construction robots: a comparative study of lightweight single and multi-AI agent systems
+
+- **ID:** `2601.14091` · **primary:** `cs.RO` · **Submitted:** 2026-01-20
+- **Authors:** Hossein Naderi, Alireza Shojaei, Lifu Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.14091 · **PDF:** https://arxiv.org/pdf/2601.14091v1
+
+## 20449. Zero-Shot Speech LLMs for Multi-Aspect Evaluation of L2 Speech: Challenges and Opportunities
+
+- **ID:** `2601.16230` · **primary:** `eess.AS` · **Submitted:** 2026-01-20
+- **Authors:** Aditya Kamlesh Parikh, Cristian Tejedor-Garcia, Catia Cucchiarini, et al.
+- **abs:** https://arxiv.org/abs/2601.16230 · **PDF:** https://arxiv.org/pdf/2601.16230v1
+
+## 20450. '1'-bit Count-based Sorting Unit to Reduce Link Power in DNN Accelerators
+
+- **ID:** `2601.14087` · **primary:** `cs.AR` · **Submitted:** 2026-01-20
+- **Authors:** Ruichi Han, Yizhi Chen, Tong Lei, et al.
+- **abs:** https://arxiv.org/abs/2601.14087 · **PDF:** https://arxiv.org/pdf/2601.14087v1
+
+## 20451. Two-Stream temporal transformer for video action classification
+
+- **ID:** `2601.14086` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Nattapong Kurpukdee, Adrian G. Bors
+- **abs:** https://arxiv.org/abs/2601.14086 · **PDF:** https://arxiv.org/pdf/2601.14086v1
+
+## 20452. DermaBench: A Clinician-Annotated Benchmark Dataset for Dermatology Visual Question Answering and Reasoning
+
+- **ID:** `2601.14084` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Abdurrahim Yilmaz, Ozan Erdem, Ece Gokyayla, et al.
+- **abs:** https://arxiv.org/abs/2601.14084 · **PDF:** https://arxiv.org/pdf/2601.14084v1
+
+## 20453. Unsupervised Video Class-Incremental Learning via Deep Embedded Clustering Management
+
+- **ID:** `2601.14069` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Nattapong Kurpukdee, Adrian G. Bors
+- **abs:** https://arxiv.org/abs/2601.14069 · **PDF:** https://arxiv.org/pdf/2601.14069v1
+
+## 20454. XCR-Bench: A Multi-Task Benchmark for Evaluating Cultural Reasoning in LLMs
+
+- **ID:** `2601.14063` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Mohsinul Kabir, Tasnim Ahmed, Md Mezbaur Rahman, et al.
+- **abs:** https://arxiv.org/abs/2601.14063 · **PDF:** https://arxiv.org/pdf/2601.14063v1
+
+## 20455. Demystifying the trend of the healthcare index: Is historical price a key driver?
+
+- **ID:** `2601.14062` · **primary:** `q-fin.ST` · **Submitted:** 2026-01-20
+- **Authors:** Payel Sadhukhan, Samrat Gupta, Subhasis Ghosh, et al.
+- **abs:** https://arxiv.org/abs/2601.14062 · **PDF:** https://arxiv.org/pdf/2601.14062v1
+
+## 20456. Variational Quantum Circuit-Based Reinforcement Learning for Dynamic Portfolio Optimization
+
+- **ID:** `2601.18811` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Vincent Gurgul, Ying Chen, Stefan Lessmann
+- **abs:** https://arxiv.org/abs/2601.18811 · **PDF:** https://arxiv.org/pdf/2601.18811v2
+
+## 20457. On finite-dimensional encoding/decoding theorems for neural operators
+
+- **ID:** `2602.00068` · **primary:** `math.FA` · **Submitted:** 2026-01-20
+- **Authors:** Vinícius Luz Oliveira, Vladimir G. Pestov
+- **abs:** https://arxiv.org/abs/2602.00068 · **PDF:** https://arxiv.org/pdf/2602.00068v1
+
+## 20458. POCI-Diff: Position Objects Consistently and Interactively with 3D-Layout Guided Diffusion
+
+- **ID:** `2601.14056` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Andrea Rigo, Luca Stornaiuolo, Weijie Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.14056 · **PDF:** https://arxiv.org/pdf/2601.14056v1
+
+## 20459. Decoder-Free Supervoxel GNN for Accurate Brain-Tumor Localization in Multi-Modal MRI
+
+- **ID:** `2601.14055` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Andrea Protani, Marc Molina Van Den Bosch, Lorenzo Giusti, et al.
+- **abs:** https://arxiv.org/abs/2601.14055 · **PDF:** https://arxiv.org/pdf/2601.14055v1
+
+## 20460. SecureSplit: Mitigating Backdoor Attacks in Split Learning
+
+- **ID:** `2601.14054` · **primary:** `cs.CR` · **Submitted:** 2026-01-20
+- **Authors:** Zhihao Dou, Dongfei Cui, Weida Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.14054 · **PDF:** https://arxiv.org/pdf/2601.14054v2
+
+## 20461. LLMOrbit: A Circular Taxonomy of Large Language Models -From Scaling Walls to Agentic AI Systems
+
+- **ID:** `2601.14053` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Badri N. Patro, Vijay S. Agneeswaran
+- **abs:** https://arxiv.org/abs/2601.14053 · **PDF:** https://arxiv.org/pdf/2601.14053v1
+
+## 20462. Kakugo: Distillation of Low-Resource Languages into Small Language Models
+
+- **ID:** `2601.14051` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Peter Devine, Mardhiyah Sanni, Farid Adilazuarda, et al.
+- **abs:** https://arxiv.org/abs/2601.14051 · **PDF:** https://arxiv.org/pdf/2601.14051v1
+
+## 20463. Understanding Multilingualism in Mixture-of-Experts LLMs: Routing Mechanism, Expert Specialization, and Layerwise Steering
+
+- **ID:** `2601.14050` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Yuxin Chen, Zhengzhou Cai, Xiangtian Ji, et al.
+- **abs:** https://arxiv.org/abs/2601.14050 · **PDF:** https://arxiv.org/pdf/2601.14050v1
+
+## 20464. Collective intelligence in science: direct elicitation of diverse information from experts with unknown information structure
+
+- **ID:** `2601.14047` · **primary:** `cs.GT` · **Submitted:** 2026-01-20
+- **Authors:** Alexey V. Osipov, Nikolay N. Osipov
+- **abs:** https://arxiv.org/abs/2601.14047 · **PDF:** https://arxiv.org/pdf/2601.14047v2
+
+## 20465. PRiSM: Benchmarking Phone Realization in Speech Models
+
+- **ID:** `2601.14046` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Shikhar Bharadwaj, Chin-Jou Li, Yoonjae Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.14046 · **PDF:** https://arxiv.org/pdf/2601.14046v1
+
+## 20466. Federated Balanced Learning
+
+- **ID:** `2601.14042` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Jiaze Li, Haoran Xu, Wanyi Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.14042 · **PDF:** https://arxiv.org/pdf/2601.14042v2
+
+## 20467. Top 10 Open Challenges Steering the Future of Diffusion Language Model and Its Variants
+
+- **ID:** `2601.14041` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Yunhe Wang, Kai Han, Huiling Zhen, et al.
+- **abs:** https://arxiv.org/abs/2601.14041 · **PDF:** https://arxiv.org/pdf/2601.14041v1
+
+## 20468. Generalizing Abstention for Noise-Robust Learning in Medical Image Segmentation
+
+- **ID:** `2601.14039` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Wesam Moustafa, Hossam Elsafty, Helen Schneider, et al.
+- **abs:** https://arxiv.org/abs/2601.14039 · **PDF:** https://arxiv.org/pdf/2601.14039v1
+
+## 20469. PAC-Private Responses with Adversarial Composition
+
+- **ID:** `2601.14033` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Xiaochen Zhu, Mayuri Sridhar, Srinivas Devadas
+- **abs:** https://arxiv.org/abs/2601.14033 · **PDF:** https://arxiv.org/pdf/2601.14033v1
+
+## 20470. RM-Distiller: Exploiting Generative LLM for Reward Model Distillation
+
+- **ID:** `2601.14032` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Hongli Zhou, Hui Huang, Wei Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.14032 · **PDF:** https://arxiv.org/pdf/2601.14032v1
+
+## 20471. Intermittent time series forecasting: local vs global models
+
+- **ID:** `2601.14031` · **primary:** `stat.ML` · **Submitted:** 2026-01-20
+- **Authors:** Stefano Damato, Nicolò Rubattu, Dario Azzimonti, et al.
+- **abs:** https://arxiv.org/abs/2601.14031 · **PDF:** https://arxiv.org/pdf/2601.14031v1
+
+## 20472. Numina-Lean-Agent: An Open and General Agentic Reasoning System for Formal Mathematics
+
+- **ID:** `2601.14027` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Junqi Liu, Zihao Zhou, Zekai Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.14027 · **PDF:** https://arxiv.org/pdf/2601.14027v1
+
+## 20473. Universal Approximation Theorem for Input-Connected Multilayer Perceptrons
+
+- **ID:** `2601.14026` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Vugar Ismailov
+- **abs:** https://arxiv.org/abs/2601.14026 · **PDF:** https://arxiv.org/pdf/2601.14026v2
+
+## 20474. Credible CO2 Comparisons: A Machine Learning Approach to Vehicle Powertrain Assessment
+
+- **ID:** `2601.14022` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Rodrigo Pereira David, Luciano Araujo Dourado Filho, Daniel Marques da Silva, et al.
+- **abs:** https://arxiv.org/abs/2601.14022 · **PDF:** https://arxiv.org/pdf/2601.14022v1
+
+## 20475. Lost in Transcription: How Speech-to-Text Errors Derail Code Understanding
+
+- **ID:** `2601.15339` · **primary:** `cs.SE` · **Submitted:** 2026-01-20
+- **Authors:** Jayant Havare, Ashish Mittal, Srikanth Tamilselvam, et al.
+- **abs:** https://arxiv.org/abs/2601.15339 · **PDF:** https://arxiv.org/pdf/2601.15339v1
+
+## 20476. MATE: Matryoshka Audio-Text Embeddings for Open-Vocabulary Keyword Spotting
+
+- **ID:** `2601.14012` · **primary:** `eess.AS` · **Submitted:** 2026-01-20
+- **Authors:** Youngmoon Jung, Myunghun Jung, Joon-Young Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.14012 · **PDF:** https://arxiv.org/pdf/2601.14012v1
+
+## 20477. BACH-V: Bridging Abstract and Concrete Human-Values in Large Language Models
+
+- **ID:** `2601.14007` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Junyu Zhang, Yipeng Kang, Jiong Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.14007 · **PDF:** https://arxiv.org/pdf/2601.14007v1
+
+## 20478. Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models
+
+- **ID:** `2601.14004` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Hengyuan Zhang, Zhihao Zhang, Mingyang Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.14004 · **PDF:** https://arxiv.org/pdf/2601.14004v3
+
+## 20479. Cross-Sensory Brain Passage Retrieval: Scaling Beyond Visual to Audio
+
+- **ID:** `2601.14001` · **primary:** `cs.IR` · **Submitted:** 2026-01-20
+- **Authors:** Niall McGuire, Yashar Moshfeghi
+- **abs:** https://arxiv.org/abs/2601.14001 · **PDF:** https://arxiv.org/pdf/2601.14001v2
+
+## 20480. Group-Invariant Unsupervised Skill Discovery: Symmetry-aware Skill Representations for Generalizable Behavior
+
+- **ID:** `2601.14000` · **primary:** `cs.RO` · **Submitted:** 2026-01-20
+- **Authors:** Junwoo Chang, Joseph Park, Roberto Horowitz, et al.
+- **abs:** https://arxiv.org/abs/2601.14000 · **PDF:** https://arxiv.org/pdf/2601.14000v1
+
+## 20481. DAME: Duration-Aware Matryoshka Embedding for Duration-Robust Speaker Verification
+
+- **ID:** `2601.13999` · **primary:** `eess.AS` · **Submitted:** 2026-01-20
+- **Authors:** Youngmoon Jung, Joon-Young Yang, Ju-ho Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.13999 · **PDF:** https://arxiv.org/pdf/2601.13999v1
+
+## 20482. From Tags to Trees: Structuring Fine-Grained Knowledge for Controllable Data Selection in LLM Instruction Tuning
+
+- **ID:** `2601.13995` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Zihan Niu, Wenping Hu, Junmin Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.13995 · **PDF:** https://arxiv.org/pdf/2601.13995v1
+
+## 20483. torch-sla: Differentiable Sparse Linear Algebra with Adjoint Solvers and Sparse Tensor Parallelism for PyTorch
+
+- **ID:** `2601.13994` · **primary:** `cs.DC` · **Submitted:** 2026-01-20
+- **Authors:** Mingyuan Chi
+- **abs:** https://arxiv.org/abs/2601.13994 · **PDF:** https://arxiv.org/pdf/2601.13994v1
+
+## 20484. "The Whole Is Greater Than the Sum of Its Parts": A Compatibility-Aware Multi-Teacher CoT Distillation Framework
+
+- **ID:** `2601.13992` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Jin Cui, Jiaqi Guo, Jiepeng Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.13992 · **PDF:** https://arxiv.org/pdf/2601.13992v1
+
+## 20485. A universal linearized subspace refinement framework for neural networks
+
+- **ID:** `2601.13989` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Wenbo Cao, Weiwei Zhang
+- **abs:** https://arxiv.org/abs/2601.13989 · **PDF:** https://arxiv.org/pdf/2601.13989v1
+
+## 20486. Harmonizing the Deep: A Unified Information Pipeline for Robust Marine Biodiversity Assessment Across Heterogeneous Domains
+
+- **ID:** `2601.13975` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Marco Piccolo, Qiwei Han, Astrid van Toor, et al.
+- **abs:** https://arxiv.org/abs/2601.13975 · **PDF:** https://arxiv.org/pdf/2601.13975v1
+
+## 20487. Turn-Based Structural Triggers: Prompt-Free Backdoors in Multi-Turn LLMs
+
+- **ID:** `2601.14340` · **primary:** `cs.CR` · **Submitted:** 2026-01-20
+- **Authors:** Yiyang Lu, Jinwen He, Yue Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.14340 · **PDF:** https://arxiv.org/pdf/2601.14340v1
+
+## 20488. Autonomous Knowledge Graph Exploration with Adaptive Breadth-Depth Retrieval
+
+- **ID:** `2601.13969` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Joaquín Polonuer, Lucas Vittor, Iñaki Arango, et al.
+- **abs:** https://arxiv.org/abs/2601.13969 · **PDF:** https://arxiv.org/pdf/2601.13969v1
+
+## 20489. CityCube: Benchmarking Cross-view Spatial Reasoning on Vision-Language Models in Urban Environments
+
+- **ID:** `2601.14339` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Haotian Xu, Yue Hu, Zhengqiu Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.14339 · **PDF:** https://arxiv.org/pdf/2601.14339v1
+
+## 20490. Modality as Heterogeneity: Node Splitting and Graph Rewiring for Multimodal Graph Learning
+
+- **ID:** `2602.00067` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Yihan Zhang, Ercan E. Kuruoglu
+- **abs:** https://arxiv.org/abs/2602.00067 · **PDF:** https://arxiv.org/pdf/2602.00067v1
+
+## 20491. RL-BioAug: Label-Efficient Reinforcement Learning for Self-Supervised EEG Representation Learning
+
+- **ID:** `2601.13964` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Cheol-Hui Lee, Hwa-Yeon Lee, Dong-Joo Kim
+- **abs:** https://arxiv.org/abs/2601.13964 · **PDF:** https://arxiv.org/pdf/2601.13964v2
+
+## 20492. IntentCoding: Amplifying User Intent in Code Generation
+
+- **ID:** `2602.00066` · **primary:** `cs.SE` · **Submitted:** 2026-01-20
+- **Authors:** Zheng Fang, Yihong Dong, Lili Mou, et al.
+- **abs:** https://arxiv.org/abs/2602.00066 · **PDF:** https://arxiv.org/pdf/2602.00066v1
+
+## 20493. Differentiable Logic Synthesis: Spectral Coefficient Selection via Sinkhorn-Constrained Composition
+
+- **ID:** `2601.13953` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Gorgi Pavlov
+- **abs:** https://arxiv.org/abs/2601.13953 · **PDF:** https://arxiv.org/pdf/2601.13953v3
+
+## 20494. Stream-Voice-Anon: Enhancing Utility of Real-Time Speaker Anonymization via Neural Audio Codec and Language Models
+
+- **ID:** `2601.13948` · **primary:** `eess.AS` · **Submitted:** 2026-01-20
+- **Authors:** Nikita Kuzmin, Songting Liu, Kong Aik Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.13948 · **PDF:** https://arxiv.org/pdf/2601.13948v3
+
+## 20495. Efficient Coordination with the System-Level Shared State: An Embodied-AI Native Modular Framework
+
+- **ID:** `2601.13945` · **primary:** `cs.RO` · **Submitted:** 2026-01-20
+- **Authors:** Yixuan Deng, Tongrun Wu, Donghao Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.13945 · **PDF:** https://arxiv.org/pdf/2601.13945v1
+
+## 20496. From Quotes to Concepts: Axial Coding of Political Debates with Ensemble LMs
+
+- **ID:** `2601.15338` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Angelina Parfenova, David Graus, Juergen Pfeffer
+- **abs:** https://arxiv.org/abs/2601.15338 · **PDF:** https://arxiv.org/pdf/2601.15338v1
+
+## 20497. Glance-or-Gaze: Incentivizing LMMs to Adaptively Focus Search via Reinforcement Learning
+
+- **ID:** `2601.13942` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Hongbo Bai, Yujin Zhou, Yile Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.13942 · **PDF:** https://arxiv.org/pdf/2601.13942v1
+
+## 20498. Agentic SPARQL: Evaluating SPARQL-MCP-powered Intelligent Agents on the Federated KGQA Benchmark
+
+- **ID:** `2603.06582` · **primary:** `cs.IR` · **Submitted:** 2026-01-20
+- **Authors:** Daniel Dobriy, Frederik Bauer, Amr Azzam, et al.
+- **abs:** https://arxiv.org/abs/2603.06582 · **PDF:** https://arxiv.org/pdf/2603.06582v1
+
+## 20499. IF-GEO: Conflict-Aware Instruction Fusion for Multi-Query Generative Engine Optimization
+
+- **ID:** `2601.13938` · **primary:** `cs.IR` · **Submitted:** 2026-01-20
+- **Authors:** Heyang Zhou, JiaJia Chen, Xiaolu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.13938 · **PDF:** https://arxiv.org/pdf/2601.13938v1
+
+## 20500. TrackletGPT: A Language-like GPT Framework for White Matter Tract Segmentation
+
+- **ID:** `2601.13935` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Anoushkrit Goel, Simroop Singh, Ankita Joshi, et al.
+- **abs:** https://arxiv.org/abs/2601.13935 · **PDF:** https://arxiv.org/pdf/2601.13935v1
+
+## 20501. Generating consensus and dissent on massive discussion platforms with an $O(N)$ semantic-vector model
+
+- **ID:** `2601.13932` · **primary:** `physics.soc-ph` · **Submitted:** 2026-01-20
+- **Authors:** A. Ferrer, D. Muñoz-Jordán, A. Rivero, et al.
+- **abs:** https://arxiv.org/abs/2601.13932 · **PDF:** https://arxiv.org/pdf/2601.13932v1
+
+## 20502. Towards Effective Negation Modeling in Joint Audio-Text Models for Music
+
+- **ID:** `2601.13931` · **primary:** `cs.SD` · **Submitted:** 2026-01-20
+- **Authors:** Yannis Vasilakis, Rachel Bittner, Johan Pauwels
+- **abs:** https://arxiv.org/abs/2601.13931 · **PDF:** https://arxiv.org/pdf/2601.13931v1
+
+## 20503. SCG With Your Phone: Diagnosis of Rhythmic Spectrum Disorders in Field Conditions
+
+- **ID:** `2601.13926` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-20
+- **Authors:** Peter Golenderov, Yaroslav Matushenko, Anastasia Tushina, et al.
+- **abs:** https://arxiv.org/abs/2601.13926 · **PDF:** https://arxiv.org/pdf/2601.13926v1
+
+## 20504. Responsible Evaluation of AI for Mental Health
+
+- **ID:** `2602.00065` · **primary:** `cs.CY` · **Submitted:** 2026-01-20
+- **Authors:** Hiba Arnaout, Anmol Goel, H. Andrew Schwartz, et al.
+- **abs:** https://arxiv.org/abs/2602.00065 · **PDF:** https://arxiv.org/pdf/2602.00065v1
+
+## 20505. Automatic Prompt Optimization for Dataset-Level Feature Discovery
+
+- **ID:** `2601.13922` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Adrian Cosma, Oleg Szehr, David Kletz, et al.
+- **abs:** https://arxiv.org/abs/2601.13922 · **PDF:** https://arxiv.org/pdf/2601.13922v1
+
+## 20506. Asymmetric regularization mechanism for GAN training with Variational Inequalities
+
+- **ID:** `2601.13920` · **primary:** `cs.GT` · **Submitted:** 2026-01-20
+- **Authors:** Spyridon C. Giagtzoglou, Mark H. M. Winands, Barbara Franci
+- **abs:** https://arxiv.org/abs/2601.13920 · **PDF:** https://arxiv.org/pdf/2601.13920v1
+
+## 20507. HyperWalker: Dynamic Hypergraph-Based Deep Diagnosis for Multi-Hop Clinical Modeling across EHR and X-Ray in Medical VLMs
+
+- **ID:** `2601.13919` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Yuezhe Yang, Hao Wang, Yige Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.13919 · **PDF:** https://arxiv.org/pdf/2601.13919v1
+
+## 20508. AgentEHR: Advancing Autonomous Clinical Decision-Making via Retrospective Summarization
+
+- **ID:** `2601.13918` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Yusheng Liao, Chuan Xuan, Yutong Cai, et al.
+- **abs:** https://arxiv.org/abs/2601.13918 · **PDF:** https://arxiv.org/pdf/2601.13918v1
+
+## 20509. SPGCL: Simple yet Powerful Graph Contrastive Learning via SVD-Guided Structural Perturbation
+
+- **ID:** `2602.00064` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Hao Deng, Zhang Guo, Shuiping Gou, et al.
+- **abs:** https://arxiv.org/abs/2602.00064 · **PDF:** https://arxiv.org/pdf/2602.00064v2
+
+## 20510. PREFAB: PREFerence-based Affective Modeling for Low-Budget Self-Annotation
+
+- **ID:** `2601.13904` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Jaeyoung Moon, Youjin Choi, Yucheon Park, et al.
+- **abs:** https://arxiv.org/abs/2601.13904 · **PDF:** https://arxiv.org/pdf/2601.13904v2
+
+## 20511. TractRLFusion: A GPT-Based Multi-Critic Policy Fusion Framework for Fiber Tractography
+
+- **ID:** `2601.13897` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Ankita Joshi, Ashutosh Sharma, Anoushkrit Goel, et al.
+- **abs:** https://arxiv.org/abs/2601.13897 · **PDF:** https://arxiv.org/pdf/2601.13897v1
+
+## 20512. OmniOVCD: Streamlining Open-Vocabulary Change Detection with SAM 3
+
+- **ID:** `2601.13895` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Xu Zhang, Danyang Li, Yingjie Xia, et al.
+- **abs:** https://arxiv.org/abs/2601.13895 · **PDF:** https://arxiv.org/pdf/2601.13895v1
+
+## 20513. Multi-Objective Hierarchical Optimization with Large Language Models
+
+- **ID:** `2601.13892` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Andrej Schwanke, Lyubomir Ivanov, David Salinas, et al.
+- **abs:** https://arxiv.org/abs/2601.13892 · **PDF:** https://arxiv.org/pdf/2601.13892v1
+
+## 20514. Log anomaly detection via Meta Learning and Prototypical Networks for Cross domain generalization
+
+- **ID:** `2601.14336` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Krishna Sharma, Vivek Yelleti
+- **abs:** https://arxiv.org/abs/2601.14336 · **PDF:** https://arxiv.org/pdf/2601.14336v1
+
+## 20515. Human Simulation Computation: A Human-Inspired Framework for Adaptive AI Systems
+
+- **ID:** `2601.13887` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Hong Su
+- **abs:** https://arxiv.org/abs/2601.13887 · **PDF:** https://arxiv.org/pdf/2601.13887v2
+
+## 20516. Confident Rankings with Fewer Items: Adaptive LLM Evaluation with Continuous Scores
+
+- **ID:** `2601.13885` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Esma Balkır, Alice Pernthaller, Marco Basaldella, et al.
+- **abs:** https://arxiv.org/abs/2601.13885 · **PDF:** https://arxiv.org/pdf/2601.13885v1
+
+## 20517. OpenLearnLM Benchmark: A Unified Framework for Evaluating Knowledge, Skill, and Attitude in Educational Large Language Models
+
+- **ID:** `2601.13882` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Unggi Lee, Sookbun Lee, Heungsoo Choi, et al.
+- **abs:** https://arxiv.org/abs/2601.13882 · **PDF:** https://arxiv.org/pdf/2601.13882v1
+
+## 20518. LifeAgentBench: A Multi-dimensional Benchmark and Agent for Personal Health Assistants in Digital Health
+
+- **ID:** `2601.13880` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Ye Tian, Zihao Wang, Onat Gungor, et al.
+- **abs:** https://arxiv.org/abs/2601.13880 · **PDF:** https://arxiv.org/pdf/2601.13880v1
+
+## 20519. Self-Supervised Score-Based Despeckling for SAR Imagery via Log-Domain Transformation
+
+- **ID:** `2601.14334` · **primary:** `eess.IV` · **Submitted:** 2026-01-20
+- **Authors:** Junhyuk Heo
+- **abs:** https://arxiv.org/abs/2601.14334 · **PDF:** https://arxiv.org/pdf/2601.14334v1
+
+## 20520. Chain-of-Thought Compression Should Not Be Blind: V-Skip for Efficient Multimodal Reasoning via Dual-Path Anchoring
+
+- **ID:** `2601.13879` · **primary:** `cs.MM` · **Submitted:** 2026-01-20
+- **Authors:** Dongxu Zhang, Yiding Sun, Cheng Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.13879 · **PDF:** https://arxiv.org/pdf/2601.13879v4
+
+## 20521. Pedagogical Alignment for Vision-Language-Action Models: A Comprehensive Framework for Data, Architecture, and Evaluation in Education
+
+- **ID:** `2601.13876` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Unggi Lee, Jahyun Jeong, Sunyoung Shin, et al.
+- **abs:** https://arxiv.org/abs/2601.13876 · **PDF:** https://arxiv.org/pdf/2601.13876v1
+
+## 20522. Unified Unbiased Variance Estimation for Maximum Mean Discrepancy: Robust Finite-Sample Performance with Imbalanced Data and Exact Acceleration under Null and Alternative Hypotheses
+
+- **ID:** `2601.13874` · **primary:** `stat.ML` · **Submitted:** 2026-01-20
+- **Authors:** Shijie Zhong, Yikun Yang, Da Gong, et al.
+- **abs:** https://arxiv.org/abs/2601.13874 · **PDF:** https://arxiv.org/pdf/2601.13874v2
+
+## 20523. Hierarchical Contextual Uplift Bandits for Catalog Personalization
+
+- **ID:** `2601.14333` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Anupam Agrawal, Rajesh Mohanty, Shamik Bhattacharjee, et al.
+- **abs:** https://arxiv.org/abs/2601.14333 · **PDF:** https://arxiv.org/pdf/2601.14333v1
+
+## 20524. HardSecBench: Benchmarking the Security Awareness of LLMs for Hardware Code Generation
+
+- **ID:** `2601.13864` · **primary:** `cs.CR` · **Submitted:** 2026-01-20
+- **Authors:** Qirui Chen, Jingxian Shuai, Shuangwu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.13864 · **PDF:** https://arxiv.org/pdf/2601.13864v1
+
+## 20525. Probabilistic Deep Discriminant Analysis for Wind Blade Segmentation
+
+- **ID:** `2601.13852` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Raül Pérez-Gonzalo, Andreas Espersen, Antonio Agudo
+- **abs:** https://arxiv.org/abs/2601.13852 · **PDF:** https://arxiv.org/pdf/2601.13852v1
+
+## 20526. Inverting Self-Organizing Maps: A Unified Activation-Based Framework
+
+- **ID:** `2601.13851` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Alessandro Londei, Matteo Benati, Denise Lanzieri, et al.
+- **abs:** https://arxiv.org/abs/2601.13851 · **PDF:** https://arxiv.org/pdf/2601.13851v2
+
+## 20527. Co-Initialization of Control Filter and Secondary Path via Meta-Learning for Active Noise Control
+
+- **ID:** `2601.13849` · **primary:** `eess.AS` · **Submitted:** 2026-01-20
+- **Authors:** Ziyi Yang, Li Rao, Zhengding Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.13849 · **PDF:** https://arxiv.org/pdf/2601.13849v1
+
+## 20528. Virtual Urbanism: An AI-Driven Framework for Quantifying Urban Identity. A Tokyo-Based Pilot Study Using Diffusion-Generated Synthetic Environments
+
+- **ID:** `2601.13846` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Glinskaya Maria
+- **abs:** https://arxiv.org/abs/2601.13846 · **PDF:** https://arxiv.org/pdf/2601.13846v1
+
+## 20529. Optimal L2 Regularization in High-dimensional Continual Linear Regression
+
+- **ID:** `2601.13844` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Gilad Karpel, Edward Moroshko, Ran Levinstein, et al.
+- **abs:** https://arxiv.org/abs/2601.13844 · **PDF:** https://arxiv.org/pdf/2601.13844v1
+
+## 20530. FutureOmni: Evaluating Future Forecasting from Omni-Modal Context for Multimodal LLMs
+
+- **ID:** `2601.13836` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Qian Chen, Jinlan Fu, Changsong Li, et al.
+- **abs:** https://arxiv.org/abs/2601.13836 · **PDF:** https://arxiv.org/pdf/2601.13836v1
+
+## 20531. Language Models Entangle Language and Culture
+
+- **ID:** `2601.15337` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Shourya Jain, Paras Chopra
+- **abs:** https://arxiv.org/abs/2601.15337 · **PDF:** https://arxiv.org/pdf/2601.15337v1
+
+## 20532. The Role of Prosodic and Lexical Cues in Turn-Taking with Self-Supervised Speech Representations
+
+- **ID:** `2601.13835` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Sam OConnor Russell, Delphine Charuau, Naomi Harte
+- **abs:** https://arxiv.org/abs/2601.13835 · **PDF:** https://arxiv.org/pdf/2601.13835v1
+
+## 20533. Learning Discrete Successor Transitions in Continuous Attractor Networks: Emergence, Limits, and Topological Constraints
+
+- **ID:** `2601.15336` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-20
+- **Authors:** Daniel Brownell
+- **abs:** https://arxiv.org/abs/2601.15336 · **PDF:** https://arxiv.org/pdf/2601.15336v1
+
+## 20534. LURE: Latent Space Unblocking for Multi-Concept Reawakening in Diffusion Models
+
+- **ID:** `2601.14330` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Mengyu Sun, Ziyuan Yang, Andrew Beng Jin Teoh, et al.
+- **abs:** https://arxiv.org/abs/2601.14330 · **PDF:** https://arxiv.org/pdf/2601.14330v1
+
+## 20535. ELSA: Efficient LLM-Centric Split Aggregation for Privacy-Aware Hierarchical Federated Learning over the Network Edge
+
+- **ID:** `2601.13824` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Xiaohong Yang, Tong Xie, Minghui Liwang, et al.
+- **abs:** https://arxiv.org/abs/2601.13824 · **PDF:** https://arxiv.org/pdf/2601.13824v2
+
+## 20536. Device Association and Resource Allocation for Hierarchical Split Federated Learning in Space-Air-Ground Integrated Network
+
+- **ID:** `2601.13817` · **primary:** `cs.DC` · **Submitted:** 2026-01-20
+- **Authors:** Haitao Zhao, Xiaoyu Tang, Bo Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.13817 · **PDF:** https://arxiv.org/pdf/2601.13817v3
+
+## 20537. Discriminant Learning-based Colorspace for Blade Segmentation
+
+- **ID:** `2601.13816` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Raül Pérez-Gonzalo, Andreas Espersen, Antonio Agudo
+- **abs:** https://arxiv.org/abs/2601.13816 · **PDF:** https://arxiv.org/pdf/2601.13816v1
+
+## 20538. Interpretable and Sparse Linear Attention with Decoupled Membership-Subspace Modeling via MCR2 Objective
+
+- **ID:** `2601.17042` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Tianyuan Liu, Libin Hou, Linyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.17042 · **PDF:** https://arxiv.org/pdf/2601.17042v1
+
+## 20539. DroneVLA: VLA based Aerial Manipulation
+
+- **ID:** `2601.13809` · **primary:** `cs.RO` · **Submitted:** 2026-01-20
+- **Authors:** Fawad Mehboob, Monijesu James, Amir Habel, et al.
+- **abs:** https://arxiv.org/abs/2601.13809 · **PDF:** https://arxiv.org/pdf/2601.13809v2
+
+## 20540. Knowledge Graph-Assisted LLM Post-Training for Enhanced Legal Reasoning
+
+- **ID:** `2601.13806` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Dezhao Song, Guglielmo Bonifazi, Frank Schilder, et al.
+- **abs:** https://arxiv.org/abs/2601.13806 · **PDF:** https://arxiv.org/pdf/2601.13806v1
+
+## 20541. Habibi: Laying the Open-Source Foundation of Unified-Dialectal Arabic Speech Synthesis
+
+- **ID:** `2601.13802` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Yushen Chen, Junzhe Liu, Yujie Tu, et al.
+- **abs:** https://arxiv.org/abs/2601.13802 · **PDF:** https://arxiv.org/pdf/2601.13802v2
+
+## 20542. CFM: Language-aligned Concept Foundation Model for Vision
+
+- **ID:** `2601.13798` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Kai Wittenmayer, Sukrut Rao, Amin Parchami-Araghi, et al.
+- **abs:** https://arxiv.org/abs/2601.13798 · **PDF:** https://arxiv.org/pdf/2601.13798v2
+
+## 20543. PAtt: A Pattern Attention Network for ETA Prediction Using Historical Speed Profiles
+
+- **ID:** `2601.13793` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** ByeoungDo Kim, JunYeop Na, Kyungwook Tak, et al.
+- **abs:** https://arxiv.org/abs/2601.13793 · **PDF:** https://arxiv.org/pdf/2601.13793v1
+
+## 20544. The Impact of Machine Learning Uncertainty on the Robustness of Counterfactual Explanations
+
+- **ID:** `2602.00063` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Leonidas Christodoulou, Chang Sun
+- **abs:** https://arxiv.org/abs/2602.00063 · **PDF:** https://arxiv.org/pdf/2602.00063v1
+
+## 20545. Principled Latent Diffusion for Graphs via Laplacian Autoencoders
+
+- **ID:** `2601.13780` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Antoine Siraudin, Christopher Morris
+- **abs:** https://arxiv.org/abs/2601.13780 · **PDF:** https://arxiv.org/pdf/2601.13780v2
+
+## 20546. Orthogonium : A Unified, Efficient Library of Orthogonal and 1-Lipschitz Building Blocks
+
+- **ID:** `2601.13776` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Thibaut Boissin, Franck Mamalet, Valentin Lafargue, et al.
+- **abs:** https://arxiv.org/abs/2601.13776 · **PDF:** https://arxiv.org/pdf/2601.13776v1
+
+## 20547. ToolCaching: Towards Efficient Caching for LLM Tool-calling
+
+- **ID:** `2601.15335` · **primary:** `cs.SE` · **Submitted:** 2026-01-20
+- **Authors:** Yi Zhai, Dian Shen, Junzhou Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.15335 · **PDF:** https://arxiv.org/pdf/2601.15335v1
+
+## 20548. Look-Ahead-Bench: a Standardized Benchmark of Look-ahead Bias in Point-in-Time LLMs for Finance
+
+- **ID:** `2601.13770` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Mostapha Benhenda
+- **abs:** https://arxiv.org/abs/2601.13770 · **PDF:** https://arxiv.org/pdf/2601.13770v1
+
+## 20549. vLinear: A Powerful Linear Model for Multivariate Time Series Forecasting
+
+- **ID:** `2601.13768` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Wenzhen Yue, Ruohao Guo, Ji Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.13768 · **PDF:** https://arxiv.org/pdf/2601.13768v1
+
+## 20550. Arabic Sign Language Recognition using Multimodal Approach
+
+- **ID:** `2601.17041` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Ghadeer Alanazi, Abir Benabid
+- **abs:** https://arxiv.org/abs/2601.17041 · **PDF:** https://arxiv.org/pdf/2601.17041v1
+
+## 20551. SCPL: Enhancing Neural Network Training Throughput with Decoupled Local Losses and Model Parallelism
+
+- **ID:** `2602.00062` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Ming-Yao Ho, Cheng-Kai Wang, You-Teng Lin, et al.
+- **abs:** https://arxiv.org/abs/2602.00062 · **PDF:** https://arxiv.org/pdf/2602.00062v2
+
+## 20552. DARC: Decoupled Asymmetric Reasoning Curriculum for LLM Evolution
+
+- **ID:** `2601.13761` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Shengda Fan, Xuyan Ye, Yankai Lin
+- **abs:** https://arxiv.org/abs/2601.13761 · **PDF:** https://arxiv.org/pdf/2601.13761v2
+
+## 20553. Finding RELIEF: Shaping Reasoning Behavior without Reasoning Supervision via Belief Engineering
+
+- **ID:** `2601.13752` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Chak Tou Leong, Dingwei Chen, Heming Xia, et al.
+- **abs:** https://arxiv.org/abs/2601.13752 · **PDF:** https://arxiv.org/pdf/2601.13752v1
+
+## 20554. Towards Onboard Continuous Change Detection for Floods
+
+- **ID:** `2601.13751` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Daniel Kyselica, Jonáš Herec, Oliver Kutis, et al.
+- **abs:** https://arxiv.org/abs/2601.13751 · **PDF:** https://arxiv.org/pdf/2601.13751v3
+
+## 20555. Pro-AI Bias in Large Language Models
+
+- **ID:** `2601.13749` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Benaya Trabelsi, Jonathan Shaki, Sarit Kraus
+- **abs:** https://arxiv.org/abs/2601.13749 · **PDF:** https://arxiv.org/pdf/2601.13749v1
+
+## 20556. EEG-Titans: Long-Horizon Seizure Forecasting via Dual-Branch Attention and Neural Memory
+
+- **ID:** `2601.13748` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Tien-Dat Pham, Xuan-The Tran
+- **abs:** https://arxiv.org/abs/2601.13748 · **PDF:** https://arxiv.org/pdf/2601.13748v1
+
+## 20557. Variational Dual-path Attention Network for CSI-Based Gesture Recognition
+
+- **ID:** `2601.13745` · **primary:** `cs.NI` · **Submitted:** 2026-01-20
+- **Authors:** N. Zhang
+- **abs:** https://arxiv.org/abs/2601.13745 · **PDF:** https://arxiv.org/pdf/2601.13745v1
+
+## 20558. BioNIC: Biologically Inspired Neural Network for Image Classification Using Connectomics Principles
+
+- **ID:** `2601.20876` · **primary:** `cs.NE` · **Submitted:** 2026-01-20
+- **Authors:** Diya Prasanth, Matthew Tivnan
+- **abs:** https://arxiv.org/abs/2601.20876 · **PDF:** https://arxiv.org/pdf/2601.20876v1
+
+## 20559. Hearing Between the Lines: Unlocking the Reasoning Power of LLMs for Speech Evaluation
+
+- **ID:** `2601.13742` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Arjun Chandra, Kevin Miller, Venkatesh Ravichandran, et al.
+- **abs:** https://arxiv.org/abs/2601.13742 · **PDF:** https://arxiv.org/pdf/2601.13742v2
+
+## 20560. Reasoning or Fluency? Dissecting Probabilistic Confidence in Best-of-N Selection
+
+- **ID:** `2601.13735` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Hojin Kim, Jaehyung Kim
+- **abs:** https://arxiv.org/abs/2601.13735 · **PDF:** https://arxiv.org/pdf/2601.13735v1
+
+## 20561. Simple Role Assignment is Extraordinarily Effective for Safety Alignment
+
+- **ID:** `2602.00061` · **primary:** `cs.CY` · **Submitted:** 2026-01-20
+- **Authors:** Zhou Ziheng, Jiakun Ding, Zhaowei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.00061 · **PDF:** https://arxiv.org/pdf/2602.00061v1
+
+## 20562. Towards robust long-context understanding of large language model via active recap learning
+
+- **ID:** `2601.13734` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Chenyu Hui
+- **abs:** https://arxiv.org/abs/2601.13734 · **PDF:** https://arxiv.org/pdf/2601.13734v1
+
+## 20563. Breaking the Data Barrier in Learning Symbolic Computation: A Case Study on Variable Ordering Suggestion for Cylindrical Algebraic Decomposition
+
+- **ID:** `2601.13731` · **primary:** `cs.SC` · **Submitted:** 2026-01-20
+- **Authors:** Rui-Juan Jing, Yuegang Zhao, Changbo Chen
+- **abs:** https://arxiv.org/abs/2601.13731 · **PDF:** https://arxiv.org/pdf/2601.13731v1
+
+## 20564. On Temperature-Constrained Non-Deterministic Machine Translation: Potential and Evaluation
+
+- **ID:** `2601.13729` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Weichuan Wang, Mingyang Liu, Linqi Song, et al.
+- **abs:** https://arxiv.org/abs/2601.13729 · **PDF:** https://arxiv.org/pdf/2601.13729v1
+
+## 20565. Yuan3.0 Ultra: A Trillion-Parameter Enterprise-Oriented MoE LLM
+
+- **ID:** `2601.14327` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** YuanLab. ai, :, Shawn Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.14327 · **PDF:** https://arxiv.org/pdf/2601.14327v3
+
+## 20566. No Reliable Evidence of Self-Reported Sentience in Small Large Language Models
+
+- **ID:** `2601.15334` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Caspar Kaiser, Sean Enderby
+- **abs:** https://arxiv.org/abs/2601.15334 · **PDF:** https://arxiv.org/pdf/2601.15334v1
+
+## 20567. OP-Bench: Benchmarking Over-Personalization for Memory-Augmented Personalized Conversational Agents
+
+- **ID:** `2601.13722` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Yulin Hu, Zimo Long, Jiahe Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.13722 · **PDF:** https://arxiv.org/pdf/2601.13722v1
+
+## 20568. Hierarchical Long Video Understanding with Audiovisual Entity Cohesion and Agentic Search
+
+- **ID:** `2601.13719` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Xinlei Yin, Xiulian Peng, Xiao Li, et al.
+- **abs:** https://arxiv.org/abs/2601.13719 · **PDF:** https://arxiv.org/pdf/2601.13719v2
+
+## 20569. Simulated Ignorance Fails: A Systematic Study of LLM Behaviors on Forecasting Problems Before Model Knowledge Cutoff
+
+- **ID:** `2601.13717` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Zehan Li, Yuxuan Wang, Ali El Lahib, et al.
+- **abs:** https://arxiv.org/abs/2601.13717 · **PDF:** https://arxiv.org/pdf/2601.13717v1
+
+## 20570. SWE-Tester: Training Open-Source LLMs for Issue Reproduction in Real-World Repositories
+
+- **ID:** `2601.13713` · **primary:** `cs.SE` · **Submitted:** 2026-01-20
+- **Authors:** Aditya Bharat Soni, Rajat Ghosh, Vaishnavi Bhargava, et al.
+- **abs:** https://arxiv.org/abs/2601.13713 · **PDF:** https://arxiv.org/pdf/2601.13713v1
+
+## 20571. Empowering LLMs for Structure-Based Drug Design via Exploration-Augmented Latent Inference
+
+- **ID:** `2601.15333` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Xuanning Hu, Anchen Li, Qianli Xing, et al.
+- **abs:** https://arxiv.org/abs/2601.15333 · **PDF:** https://arxiv.org/pdf/2601.15333v2
+
+## 20572. GerAV: Towards New Heights in German Authorship Verification using Fine-Tuned LLMs on a New Benchmark
+
+- **ID:** `2601.13711` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Lotta Kiefer, Christoph Leiter, Sotaro Takeshita, et al.
+- **abs:** https://arxiv.org/abs/2601.13711 · **PDF:** https://arxiv.org/pdf/2601.13711v1
+
+## 20573. Who Benefits From Sinus Surgery? Comparing Generative AI and Supervised Machine Learning for Predicting Surgical Outcomes in Chronic Rhinosinusitis
+
+- **ID:** `2601.13710` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Sayeed Shafayet Chowdhury, Snehasis Mukhopadhyay, Shiaofen Fang, et al.
+- **abs:** https://arxiv.org/abs/2601.13710 · **PDF:** https://arxiv.org/pdf/2601.13710v2
+
+## 20574. Hidden in Plain Text: Measuring LLM Deception Quality Against Human Baselines Using Social Deduction Games
+
+- **ID:** `2601.13709` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Christopher Kao, Vanshika Vats, James Davis
+- **abs:** https://arxiv.org/abs/2601.13709 · **PDF:** https://arxiv.org/pdf/2601.13709v1
+
+## 20575. Generative Adversarial Networks for Resource State Generation
+
+- **ID:** `2601.13708` · **primary:** `quant-ph` · **Submitted:** 2026-01-20
+- **Authors:** Shahbaz Shaik, Sourav Chatterjee, Sayantan Pramanik, et al.
+- **abs:** https://arxiv.org/abs/2601.13708 · **PDF:** https://arxiv.org/pdf/2601.13708v2
+
+## 20576. Attention-space Contrastive Guidance for Efficient Hallucination Mitigation in LVLMs
+
+- **ID:** `2601.13707` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Yujin Jo, Sangyoon Bae, Taesup Kim
+- **abs:** https://arxiv.org/abs/2601.13707 · **PDF:** https://arxiv.org/pdf/2601.13707v1
+
+## 20577. Performance and Complexity Trade-off Optimization of Speech Models During Training
+
+- **ID:** `2601.13704` · **primary:** `cs.SD` · **Submitted:** 2026-01-20
+- **Authors:** Esteban Gómez, Tom Bäckström
+- **abs:** https://arxiv.org/abs/2601.13704 · **PDF:** https://arxiv.org/pdf/2601.13704v2
+
+## 20578. Does Privacy Always Harm Fairness? Data-Dependent Trade-offs via Chernoff Information Neural Estimation
+
+- **ID:** `2601.13698` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Arjun Nichani, Hsiang Hsu, Chun-Fu, et al.
+- **abs:** https://arxiv.org/abs/2601.13698 · **PDF:** https://arxiv.org/pdf/2601.13698v2
+
+## 20579. Uncertainty-Aware Gradient Signal-to-Noise Data Selection for Instruction Tuning
+
+- **ID:** `2601.13697` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Zhihang Yuan, Chengyu Yue, Long Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.13697 · **PDF:** https://arxiv.org/pdf/2601.13697v1
+
+## 20580. OptiSQL: Executable SQL Generation from Optical Tokens
+
+- **ID:** `2601.13695` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Sifan Li, Hongkai Chen, Yujun Cai, et al.
+- **abs:** https://arxiv.org/abs/2601.13695 · **PDF:** https://arxiv.org/pdf/2601.13695v2
+
+## 20581. End-to-End Reverse Screening Identifies Protein Targets of Small Molecules Using HelixFold3
+
+- **ID:** `2601.13693` · **primary:** `q-bio.BM` · **Submitted:** 2026-01-20
+- **Authors:** Shengjie Xu, Xianbin Ye, Mengran Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.13693 · **PDF:** https://arxiv.org/pdf/2601.13693v1
+
+## 20582. Dr. Assistant: Enhancing Clinical Diagnostic Inquiry via Structured Diagnostic Reasoning Data and Reinforcement Learning
+
+- **ID:** `2601.13690` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Yue Guo, Fanfu Wang, Jianwei Lv, et al.
+- **abs:** https://arxiv.org/abs/2601.13690 · **PDF:** https://arxiv.org/pdf/2601.13690v1
+
+## 20583. Understanding Mental States to Guide Social Influence in Multi-Person Group Dialogue
+
+- **ID:** `2601.13687` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Zhichao Liang, Satoshi Nakamura
+- **abs:** https://arxiv.org/abs/2601.13687 · **PDF:** https://arxiv.org/pdf/2601.13687v2
+
+## 20584. HeteroCache: A Dynamic Retrieval Approach to Heterogeneous KV Cache Compression for Long-Context LLM Inference
+
+- **ID:** `2601.13684` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Zhiyuan Shi, Qibo Qiu, Feng Xue, et al.
+- **abs:** https://arxiv.org/abs/2601.13684 · **PDF:** https://arxiv.org/pdf/2601.13684v1
+
+## 20585. Autoregressive deep learning for real-time simulation of soft tissue dynamics during virtual neurosurgery
+
+- **ID:** `2601.13676` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Fabian Greifeneder, Wolfgang Fenz, Benedikt Alkin, et al.
+- **abs:** https://arxiv.org/abs/2601.13676 · **PDF:** https://arxiv.org/pdf/2601.13676v1
+
+## 20586. The Orchestration of Multi-Agent Systems: Architectures, Protocols, and Enterprise Adoption
+
+- **ID:** `2601.13671` · **primary:** `cs.MA` · **Submitted:** 2026-01-20
+- **Authors:** Apoorva Adimulam, Rajesh Gupta, Sumit Kumar
+- **abs:** https://arxiv.org/abs/2601.13671 · **PDF:** https://arxiv.org/pdf/2601.13671v1
+
+## 20587. FP-THD: Full page transcription of historical documents
+
+- **ID:** `2601.17040` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** H Neji, J Nogueras-Iso, J Lacasta, et al.
+- **abs:** https://arxiv.org/abs/2601.17040 · **PDF:** https://arxiv.org/pdf/2601.17040v1
+
+## 20588. CommunityBench: Benchmarking Community-Level Alignment across Diverse Groups and Tasks
+
+- **ID:** `2601.13669` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Jiayu Lin, Zhongyu Wei
+- **abs:** https://arxiv.org/abs/2601.13669 · **PDF:** https://arxiv.org/pdf/2601.13669v1
+
+## 20589. Reinforcement Learning for Opportunistic Routing in Software-Defined LEO-Terrestrial Systems
+
+- **ID:** `2601.13662` · **primary:** `cs.NI` · **Submitted:** 2026-01-20
+- **Authors:** Sivaram Krishnan, Zhouyou Gu, Jihong Park, et al.
+- **abs:** https://arxiv.org/abs/2601.13662 · **PDF:** https://arxiv.org/pdf/2601.13662v1
+
+## 20590. Temporal-Spatial Decouple before Act: Disentangled Representation Learning for Multimodal Sentiment Analysis
+
+- **ID:** `2601.13659` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Chunlei Meng, Ziyang Zhou, Lucas He, et al.
+- **abs:** https://arxiv.org/abs/2601.13659 · **PDF:** https://arxiv.org/pdf/2601.13659v1
+
+## 20591. Beyond Known Facts: Generating Unseen Temporal Knowledge to Address Data Contamination in LLM Evaluation
+
+- **ID:** `2601.13658` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Arthur Amalvy, Hen-Hsen Huang
+- **abs:** https://arxiv.org/abs/2601.13658 · **PDF:** https://arxiv.org/pdf/2601.13658v1
+
+## 20592. Communication-Free Collective Navigation for a Swarm of UAVs via LiDAR-Based Deep Reinforcement Learning
+
+- **ID:** `2601.13657` · **primary:** `cs.RO` · **Submitted:** 2026-01-20
+- **Authors:** Myong-Yol Choi, Hankyoul Ko, Hanse Cho, et al.
+- **abs:** https://arxiv.org/abs/2601.13657 · **PDF:** https://arxiv.org/pdf/2601.13657v1
+
+## 20593. Why Does the LLM Stop Computing: An Empirical Study of User-Reported Failures in Open-Source LLMs
+
+- **ID:** `2601.13655` · **primary:** `cs.SE` · **Submitted:** 2026-01-20
+- **Authors:** Guangba Yu, Zirui Wang, Yujie Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.13655 · **PDF:** https://arxiv.org/pdf/2601.13655v1
+
+## 20594. TimeART: Towards Agentic Time Series Reasoning via Tool-Augmentation
+
+- **ID:** `2601.13653` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Xingjian Wu, Junkai Lu, Zhengyu Li, et al.
+- **abs:** https://arxiv.org/abs/2601.13653 · **PDF:** https://arxiv.org/pdf/2601.13653v1
+
+## 20595. Fairness or Fluency? An Investigation into Language Bias of Pairwise LLM-as-a-Judge
+
+- **ID:** `2601.13649` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Xiaolin Zhou, Zheng Luo, Yicheng Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.13649 · **PDF:** https://arxiv.org/pdf/2601.13649v1
+
+## 20596. Fusion Segment Transformer: Bi-Directional Attention Guided Fusion Network for AI-Generated Music Detection
+
+- **ID:** `2601.13647` · **primary:** `cs.SD` · **Submitted:** 2026-01-20
+- **Authors:** Yumin Kim, Seonghyeon Go
+- **abs:** https://arxiv.org/abs/2601.13647 · **PDF:** https://arxiv.org/pdf/2601.13647v1
+
+## 20597. Quadratic Upper Bound for Boosting Robustness
+
+- **ID:** `2601.13645` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Euijin You, Hyang-Won Lee
+- **abs:** https://arxiv.org/abs/2601.13645 · **PDF:** https://arxiv.org/pdf/2601.13645v1
+
+## 20598. Towards Token-Level Text Anomaly Detection
+
+- **ID:** `2601.13644` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Yang Cao, Bicheng Yu, Sikun Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.13644 · **PDF:** https://arxiv.org/pdf/2601.13644v1
+
+## 20599. Sample Complexity of Average-Reward Q-Learning: From Single-agent to Federated Reinforcement Learning
+
+- **ID:** `2601.13642` · **primary:** `stat.ML` · **Submitted:** 2026-01-20
+- **Authors:** Yuchen Jiao, Jiin Woo, Gen Li, et al.
+- **abs:** https://arxiv.org/abs/2601.13642 · **PDF:** https://arxiv.org/pdf/2601.13642v1
+
+## 20600. A longitudinal geospatial multimodal dataset of post-discharge frailty, physiology, mobility, and neighborhoods
+
+- **ID:** `2602.00060` · **primary:** `cs.CY` · **Submitted:** 2026-01-20
+- **Authors:** Ali Abedi, Charlene H. Chu, Shehroz S. Khan
+- **abs:** https://arxiv.org/abs/2602.00060 · **PDF:** https://arxiv.org/pdf/2602.00060v1
+
+## 20601. Resilient Routing: Risk-Aware Dynamic Routing in Smart Logistics via Spatiotemporal Graph Learning
+
+- **ID:** `2601.13632` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Zhiming Xue, Sichen Zhao, Yalun Qi, et al.
+- **abs:** https://arxiv.org/abs/2601.13632 · **PDF:** https://arxiv.org/pdf/2601.13632v2
+
+## 20602. RECAP: A Resource-Efficient Method for Adversarial Prompting in Large Language Models
+
+- **ID:** `2601.15331` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Rishit Chugh
+- **abs:** https://arxiv.org/abs/2601.15331 · **PDF:** https://arxiv.org/pdf/2601.15331v1
+
+## 20603. Activation-Space Anchored Access Control for Multi-Class Permission Reasoning in Large Language Models
+
+- **ID:** `2601.13630` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Zhaopeng Zhang, Pengcheng Sun, Lan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.13630 · **PDF:** https://arxiv.org/pdf/2601.13630v1
+
+## 20604. TextBFGS: A Case-Based Reasoning Approach to Code Optimization via Error-Operator Retrieval
+
+- **ID:** `2602.00059` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Zizheng Zhang, Yuyang Liao, Chen Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.00059 · **PDF:** https://arxiv.org/pdf/2602.00059v2
+
+## 20605. CARPE: Context-Aware Image Representation Prioritization via Ensemble for Large Vision-Language Models
+
+- **ID:** `2601.13622` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Donghee Lee, Rui Cai, Zhe Zhao
+- **abs:** https://arxiv.org/abs/2601.13622 · **PDF:** https://arxiv.org/pdf/2601.13622v3
+
+## 20606. CauScientist: Teaching LLMs to Respect Data for Causal Discovery
+
+- **ID:** `2601.13614` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Bo Peng, Sirui Chen, Lei Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.13614 · **PDF:** https://arxiv.org/pdf/2601.13614v1
+
+## 20607. Fisher-Informed Parameterwise Aggregation for Federated Learning with Heterogeneous Data
+
+- **ID:** `2601.13608` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Zhipeng Chang, Ting He, Wenrui Hao
+- **abs:** https://arxiv.org/abs/2601.13608 · **PDF:** https://arxiv.org/pdf/2601.13608v1
+
+## 20608. MANGO: A Global Single-Date Paired Dataset for Mangrove Segmentation
+
+- **ID:** `2601.17039` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Junhyuk Heo, Beomkyu Choi, Hyunjin Shin, et al.
+- **abs:** https://arxiv.org/abs/2601.17039 · **PDF:** https://arxiv.org/pdf/2601.17039v1
+
+## 20609. Optimizing Parallel Schemes with Lyapunov Exponents and kNN-LLE Estimation
+
+- **ID:** `2601.13604` · **primary:** `math.NA` · **Submitted:** 2026-01-20
+- **Authors:** Mudassir Shams, Andrei Velichko, Bruno Carpentieri
+- **abs:** https://arxiv.org/abs/2601.13604 · **PDF:** https://arxiv.org/pdf/2601.13604v1
+
+## 20610. An Elementary Approach to Scheduling in Generative Diffusion Models
+
+- **ID:** `2601.13602` · **primary:** `cs.IT` · **Submitted:** 2026-01-20
+- **Authors:** Qiang Sun, H. Vincent Poor, Wenyi Zhang
+- **abs:** https://arxiv.org/abs/2601.13602 · **PDF:** https://arxiv.org/pdf/2601.13602v2
+
+## 20611. Foundations of Global Consistency Checking with Noisy LLM Oracles
+
+- **ID:** `2601.13600` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Paul He, Elke Kirschbaum, Shiva Kasiviswanathan
+- **abs:** https://arxiv.org/abs/2601.13600 · **PDF:** https://arxiv.org/pdf/2601.13600v1
+
+## 20612. Diffusion In Diffusion: Reclaiming Global Coherence in Semi-Autoregressive Diffusion
+
+- **ID:** `2601.13599` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Linrui Ma, Yufei Cui, Kai Han, et al.
+- **abs:** https://arxiv.org/abs/2601.13599 · **PDF:** https://arxiv.org/pdf/2601.13599v2
+
+## 20613. Comparison of Multiple Classifiers for Android Malware Detection with Emphasis on Feature Insights Using CICMalDroid 2020 Dataset
+
+- **ID:** `2602.00058` · **primary:** `cs.CR` · **Submitted:** 2026-01-20
+- **Authors:** Md Min-Ha-Zul Abedin, Tazqia Mehrub
+- **abs:** https://arxiv.org/abs/2602.00058 · **PDF:** https://arxiv.org/pdf/2602.00058v1
+
+## 20614. Machine learning based radiative parameterization scheme and its performance in operational reforecast experiments
+
+- **ID:** `2601.13592` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Hao Jing, Sa Xiao, Haoyu Li, et al.
+- **abs:** https://arxiv.org/abs/2601.13592 · **PDF:** https://arxiv.org/pdf/2601.13592v1
+
+## 20615. DSAEval: Evaluating Data Science Agents on a Wide Range of Real-World Data Science Problems
+
+- **ID:** `2601.13591` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Maojun Sun, Yifei Xie, Yue Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.13591 · **PDF:** https://arxiv.org/pdf/2601.13591v1
+
+## 20616. Vulnerability of LLMs' Stated Beliefs? LLMs Belief Resistance Check Through Strategic Persuasive Conversation Interventions
+
+- **ID:** `2601.13590` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Fan Huang, Haewoon Kwak, Jisun An
+- **abs:** https://arxiv.org/abs/2601.13590 · **PDF:** https://arxiv.org/pdf/2601.13590v3
+
+## 20617. Motion-to-Response Content Generation via Multi-Agent AI System with Real-Time Safety Verification
+
+- **ID:** `2601.13589` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** HyeYoung Lee
+- **abs:** https://arxiv.org/abs/2601.13589 · **PDF:** https://arxiv.org/pdf/2601.13589v1
+
+## 20618. TREX: Tokenizer Regression for Optimal Data Mixture
+
+- **ID:** `2601.13588` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Inho Won, Hangyeol Yoo, Minkyung Cho, et al.
+- **abs:** https://arxiv.org/abs/2601.13588 · **PDF:** https://arxiv.org/pdf/2601.13588v1
+
+## 20619. ICPO: Illocution-Calibrated Policy Optimization for Multi-Turn Conversation
+
+- **ID:** `2601.15330` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Zhebo Wang, Xiaohu Mu, Zijie Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.15330 · **PDF:** https://arxiv.org/pdf/2601.15330v1
+
+## 20620. Hybrid Deep Feature Extraction and ML for Construction and Demolition Debris Classification
+
+- **ID:** `2601.17038` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Obai Alashram, Nejad Alagha, Mahmoud AlKakuri, et al.
+- **abs:** https://arxiv.org/abs/2601.17038 · **PDF:** https://arxiv.org/pdf/2601.17038v1
+
+## 20621. SCRIPTMIND: Crime Script Inference and Cognitive Evaluation for LLM-based Social Engineering Scam Detection System
+
+- **ID:** `2601.13581` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Heedou Kim, Changsik Kim, Sanghwa Shin, et al.
+- **abs:** https://arxiv.org/abs/2601.13581 · **PDF:** https://arxiv.org/pdf/2601.13581v1
+
+## 20622. Neural Organ Transplantation (NOT): Checkpoint-Based Modular Adaptation for Transformer Models
+
+- **ID:** `2601.13580` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Ahmad Al-Zuraiqi
+- **abs:** https://arxiv.org/abs/2601.13580 · **PDF:** https://arxiv.org/pdf/2601.13580v1
+
+## 20623. FG-OrIU: Towards Better Forgetting via Feature-Gradient Orthogonality for Incremental Unlearning
+
+- **ID:** `2601.13578` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Qian Feng, JiaHang Tu, Mintong Kang, et al.
+- **abs:** https://arxiv.org/abs/2601.13578 · **PDF:** https://arxiv.org/pdf/2601.13578v1
+
+## 20624. Comparing Without Saying: A Dataset and Benchmark for Implicit Comparative Opinion Mining from Same-User Reviews
+
+- **ID:** `2601.13575` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Thanh-Lam T. Nguyen, Ngoc-Quang Le, Quoc-Trung Phu, et al.
+- **abs:** https://arxiv.org/abs/2601.13575 · **PDF:** https://arxiv.org/pdf/2601.13575v1
+
+## 20625. Behavior Knowledge Merge in Reinforced Agentic Models
+
+- **ID:** `2601.13572` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Xiangchi Yuan, Dachuan Shi, Chunhui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.13572 · **PDF:** https://arxiv.org/pdf/2601.13572v1
+
+## 20626. GeoDynamics: A Geometric State-Space Neural Network for Understanding Brain Dynamics on Riemannian Manifolds
+
+- **ID:** `2601.13570` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Tingting Dan, Jiaqi Ding, Guorong Wu
+- **abs:** https://arxiv.org/abs/2601.13570 · **PDF:** https://arxiv.org/pdf/2601.13570v2
+
+## 20627. DRGW: Learning Disentangled Representations for Robust Graph Watermarking
+
+- **ID:** `2601.13569` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Jiasen Li, Yanwei Liu, Zhuoyi Shang, et al.
+- **abs:** https://arxiv.org/abs/2601.13569 · **PDF:** https://arxiv.org/pdf/2601.13569v2
+
+## 20628. Self-Improvement as Coherence Optimization: A Theoretical Account
+
+- **ID:** `2601.13566` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Tianyi Qiu, Ahmed Hani Ismail, Zhonghao He, et al.
+- **abs:** https://arxiv.org/abs/2601.13566 · **PDF:** https://arxiv.org/pdf/2601.13566v1
+
+## 20629. Multi-objective fluorescent molecule design with a data-physics dual-driven generative framework
+
+- **ID:** `2601.13564` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Yanheng Li, Zhichen Pu, Lijiang Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.13564 · **PDF:** https://arxiv.org/pdf/2601.13564v1
+
+## 20630. ButterflyMoE: Sub-Linear Ternary Experts via Structured Butterfly Orbits
+
+- **ID:** `2601.13563` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Aryan Karmore
+- **abs:** https://arxiv.org/abs/2601.13563 · **PDF:** https://arxiv.org/pdf/2601.13563v4
+
+## 20631. Reasoning is a Modality
+
+- **ID:** `2601.13562` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Zhiguang Liu, Yi Shang
+- **abs:** https://arxiv.org/abs/2601.13562 · **PDF:** https://arxiv.org/pdf/2601.13562v1
+
+## 20632. AgentGC: Evolutionary Learning-based Lossless Compression for Genomics Data with LLM-driven Multiple Agent
+
+- **ID:** `2601.13559` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Sun Hui, Ding Yanfeng, Huidong Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.13559 · **PDF:** https://arxiv.org/pdf/2601.13559v1
+
+## 20633. Leveraging ChatGPT and Other NLP Methods for Identifying Risk and Protective Behaviors in MSM: Social Media and Dating apps Text Analysis
+
+- **ID:** `2601.13558` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Mehrab Beikzadeh, Chenglin Hong, Cory J Cascalheira, et al.
+- **abs:** https://arxiv.org/abs/2601.13558 · **PDF:** https://arxiv.org/pdf/2601.13558v1
+
+## 20634. Patterning: The Dual of Interpretability
+
+- **ID:** `2601.13548` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** George Wang, Daniel Murfet
+- **abs:** https://arxiv.org/abs/2601.13548 · **PDF:** https://arxiv.org/pdf/2601.13548v1
+
+## 20635. HateXScore: A Metric Suite for Evaluating Reasoning Quality in Hate Speech Explanations
+
+- **ID:** `2601.13547` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Yujia Hu, Roy Ka-Wei Lee
+- **abs:** https://arxiv.org/abs/2601.13547 · **PDF:** https://arxiv.org/pdf/2601.13547v1
+
+## 20636. ChatAD: Reasoning-Enhanced Time-Series Anomaly Detection with Multi-Turn Instruction Evolution
+
+- **ID:** `2601.13546` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Hui Sun, Chang Xu, Haonan Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.13546 · **PDF:** https://arxiv.org/pdf/2601.13546v1
+
+## 20637. TruthTensor: Evaluating LLMs through Human Imitation on Prediction Market under Drift and Holistic Reasoning
+
+- **ID:** `2601.13545` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Shirin Shahabi, Spencer Graham, Haruna Isah
+- **abs:** https://arxiv.org/abs/2601.13545 · **PDF:** https://arxiv.org/pdf/2601.13545v3
+
+## 20638. Refined Gradient-Based Temperature Optimization for the Replica-Exchange Monte-Carlo Method
+
+- **ID:** `2601.13542` · **primary:** `physics.comp-ph` · **Submitted:** 2026-01-20
+- **Authors:** Tatsuya Miyata, Shunta Arai, Satoshi Takabe
+- **abs:** https://arxiv.org/abs/2601.13542 · **PDF:** https://arxiv.org/pdf/2601.13542v1
+
+## 20639. When Wording Steers the Evaluation: Framing Bias in LLM judges
+
+- **ID:** `2601.13537` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Yerin Hwang, Dongryeol Lee, Taegwan Kang, et al.
+- **abs:** https://arxiv.org/abs/2601.13537 · **PDF:** https://arxiv.org/pdf/2601.13537v1
+
+## 20640. Diff-MN: Diffusion Parameterized MoE-NCDE for Continuous Time Series Generation with Irregular Observations
+
+- **ID:** `2601.13534` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Xu Zhang, Junwei Deng, Chang Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.13534 · **PDF:** https://arxiv.org/pdf/2601.13534v2
+
+## 20641. Reasoning While Recommending: Entropy-Guided Latent Reasoning in Generative Re-ranking Models
+
+- **ID:** `2601.13533` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Changshuo Zhang
+- **abs:** https://arxiv.org/abs/2601.13533 · **PDF:** https://arxiv.org/pdf/2601.13533v1
+
+## 20642. Eliciting Harmful Capabilities by Fine-Tuning On Safeguarded Outputs
+
+- **ID:** `2601.13528` · **primary:** `cs.CR` · **Submitted:** 2026-01-20
+- **Authors:** Jackson Kaunismaa, Avery Griffin, John Hughes, et al.
+- **abs:** https://arxiv.org/abs/2601.13528 · **PDF:** https://arxiv.org/pdf/2601.13528v1
+
+## 20643. StoTAM: Stochastic Alternating Minimization for Tucker-Structured Tensor Sensing
+
+- **ID:** `2601.13522` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Shuang Li
+- **abs:** https://arxiv.org/abs/2601.13522 · **PDF:** https://arxiv.org/pdf/2601.13522v1
+
+## 20644. Explore Brain-Inspired Machine Intelligence for Connecting Dots on Graphs Through Holographic Blueprint of Oscillatory Synchronization
+
+- **ID:** `2602.00057` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-20
+- **Authors:** Tingting Dan, Jiaqi Ding, Guorong Wu
+- **abs:** https://arxiv.org/abs/2602.00057 · **PDF:** https://arxiv.org/pdf/2602.00057v1
+
+## 20645. Small Gradient Norm Regret for Online Convex Optimization
+
+- **ID:** `2601.13519` · **primary:** `stat.ML` · **Submitted:** 2026-01-20
+- **Authors:** Wenzhi Gao, Chang He, Madeleine Udell
+- **abs:** https://arxiv.org/abs/2601.13519 · **PDF:** https://arxiv.org/pdf/2601.13519v3
+
+## 20646. AgenticRed: Optimizing Agentic Systems for Automated Red-teaming
+
+- **ID:** `2601.13518` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Jiayi Yuan, Jonathan Nöther, Natasha Jaques, et al.
+- **abs:** https://arxiv.org/abs/2601.13518 · **PDF:** https://arxiv.org/pdf/2601.13518v2
+
+## 20647. Automatic Adjustment of HPA Parameters and Attack Prevention in Kubernetes Using Random Forests
+
+- **ID:** `2601.13515` · **primary:** `cs.CR` · **Submitted:** 2026-01-20
+- **Authors:** Hanlin Zhou, Huah Yong Chan, Jingfei Ni, et al.
+- **abs:** https://arxiv.org/abs/2601.13515 · **PDF:** https://arxiv.org/pdf/2601.13515v1
+
+## 20648. CatMaster: An Agentic Autonomous System for Computational Heterogeneous Catalysis Research
+
+- **ID:** `2601.13508` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-20
+- **Authors:** Honghao Chen, Jiangjie Qiu, Yi Shen Tew, et al.
+- **abs:** https://arxiv.org/abs/2601.13508 · **PDF:** https://arxiv.org/pdf/2601.13508v1
+
+## 20649. Anonpsy: A Graph-Based Framework for Structure-Preserving De-identification of Psychiatric Narratives
+
+- **ID:** `2601.13503` · **primary:** `cs.CL` · **Submitted:** 2026-01-20
+- **Authors:** Kyung Ho Lim, Byung-Hoon Kim
+- **abs:** https://arxiv.org/abs/2601.13503 · **PDF:** https://arxiv.org/pdf/2601.13503v1
+
+## 20650. SilentDrift: Exploiting Action Chunking for Stealthy Backdoor Attacks on Vision-Language-Action Models
+
+- **ID:** `2601.14323` · **primary:** `cs.CR` · **Submitted:** 2026-01-20
+- **Authors:** Bingxin Xu, Yuzhang Shang, Binghui Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.14323 · **PDF:** https://arxiv.org/pdf/2601.14323v1
+
+## 20651. How Hyper-Datafication Impacts the Sustainability Costs in Frontier AI
+
+- **ID:** `2602.00056` · **primary:** `cs.CY` · **Submitted:** 2026-01-20
+- **Authors:** Sophia N. Wilson, Sebastian Mair, Mophat Okinyi, et al.
+- **abs:** https://arxiv.org/abs/2602.00056 · **PDF:** https://arxiv.org/pdf/2602.00056v2
+
+## 20652. Bridging the Gap Between Estimated and True Regret Towards Reliable Regret Estimation in Deep Learning based Mechanism Design
+
+- **ID:** `2601.13489` · **primary:** `cs.GT` · **Submitted:** 2026-01-20
+- **Authors:** Shuyuan You, Zhiqiang Zhuang, Kewen Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.13489 · **PDF:** https://arxiv.org/pdf/2601.13489v1
+
+## 20653. The Hidden Toll of Social Media News: Causal Effects on Psychosocial Wellbeing
+
+- **ID:** `2601.13487` · **primary:** `cs.SI` · **Submitted:** 2026-01-20
+- **Authors:** Olivia Pal, Agam Goyal, Eshwar Chandrasekharan, et al.
+- **abs:** https://arxiv.org/abs/2601.13487 · **PDF:** https://arxiv.org/pdf/2601.13487v1
+
+## 20654. Towards Efficient and Robust Linguistic Emotion Diagnosis for Mental Health via Multi-Agent Instruction Refinement
+
+- **ID:** `2601.13481` · **primary:** `cs.AI` · **Submitted:** 2026-01-20
+- **Authors:** Jian Zhang, Zhangqi Wang, Zhiyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.13481 · **PDF:** https://arxiv.org/pdf/2601.13481v1
+
+## 20655. A Unified Variational Imputation Framework for Electric Vehicle Charging Data Using Retrieval-Augmented Language Model
+
+- **ID:** `2601.13476` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Jinhao Li, Hao Wang
+- **abs:** https://arxiv.org/abs/2601.13476 · **PDF:** https://arxiv.org/pdf/2601.13476v1
+
+## 20656. Preconditioning Benefits of Spectral Orthogonalization in Muon
+
+- **ID:** `2601.13474` · **primary:** `cs.LG` · **Submitted:** 2026-01-20
+- **Authors:** Jianhao Ma, Yu Huang, Yuejie Chi, et al.
+- **abs:** https://arxiv.org/abs/2601.13474 · **PDF:** https://arxiv.org/pdf/2601.13474v1
+
+## 20657. AMVICC: A Novel Benchmark for Cross-Modal Failure Mode Profiling for VLMs and IGMs
+
+- **ID:** `2601.17037` · **primary:** `cs.CV` · **Submitted:** 2026-01-20
+- **Authors:** Aahana Basappa, Pranay Goel, Anusri Karra, et al.
+- **abs:** https://arxiv.org/abs/2601.17037 · **PDF:** https://arxiv.org/pdf/2601.17037v1
+
+## 20658. Graph Neural Networks are Heuristics
+
+- **ID:** `2601.13465` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Yimeng Min, Carla P. Gomes
+- **abs:** https://arxiv.org/abs/2601.13465 · **PDF:** https://arxiv.org/pdf/2601.13465v3
+
+## 20659. Context and Transcripts Improve Detection of Deepfake Audios of Public Figures
+
+- **ID:** `2601.13464` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Chongyang Gao, Marco Postiglione, Julian Baldwin, et al.
+- **abs:** https://arxiv.org/abs/2601.13464 · **PDF:** https://arxiv.org/pdf/2601.13464v1
+
+## 20660. Quantum Qualifiers for Neural Network Model Selection in Hadronic Physics
+
+- **ID:** `2601.13463` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Brandon B. Le, D. Keller
+- **abs:** https://arxiv.org/abs/2601.13463 · **PDF:** https://arxiv.org/pdf/2601.13463v1
+
+## 20661. SpatialBench-UC: Uncertainty-Aware Evaluation of Spatial Prompt Following in Text-to-Image Generation
+
+- **ID:** `2601.13462` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Amine Rostane
+- **abs:** https://arxiv.org/abs/2601.13462 · **PDF:** https://arxiv.org/pdf/2601.13462v1
+
+## 20662. Labels or Preferences? Budget-Constrained Learning with Human Judgments over AI-Generated Outputs
+
+- **ID:** `2601.13458` · **primary:** `stat.ML` · **Submitted:** 2026-01-19
+- **Authors:** Zihan Dong, Xiaotian Hou, Ruijia Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.13458 · **PDF:** https://arxiv.org/pdf/2601.13458v2
+
+## 20663. Federated Learning Under Temporal Drift -- Mitigating Catastrophic Forgetting via Experience Replay
+
+- **ID:** `2601.13456` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Sahasra Kokkula, Daniel David, Aaditya Baruah
+- **abs:** https://arxiv.org/abs/2601.13456 · **PDF:** https://arxiv.org/pdf/2601.13456v1
+
+## 20664. PhysicsSolutionAgent: Towards Multimodal Explanations for Numerical Physics Problem Solving
+
+- **ID:** `2601.13453` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Aditya Thole, Anmol Agrawal, Arnav Ramamoorthy, et al.
+- **abs:** https://arxiv.org/abs/2601.13453 · **PDF:** https://arxiv.org/pdf/2601.13453v1
+
+## 20665. Fairness-informed Pareto Optimization : An Efficient Bilevel Framework
+
+- **ID:** `2601.13448` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Sofiane Tanji, Samuel Vaiter, Yassine Laguel
+- **abs:** https://arxiv.org/abs/2601.13448 · **PDF:** https://arxiv.org/pdf/2601.13448v2
+
+## 20666. BladeSDF : Unconditional and Conditional Generative Modeling of Representative Blade Geometries Using Signed Distance Functions
+
+- **ID:** `2601.13445` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Ashish S. Nair, Sandipp Krishnan Ravi, Itzel Salgado, et al.
+- **abs:** https://arxiv.org/abs/2601.13445 · **PDF:** https://arxiv.org/pdf/2601.13445v1
+
+## 20667. Explicit Cognitive Allocation: A Principle for Governed and Auditable Inference in Large Language Models
+
+- **ID:** `2601.13443` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Héctor Manuel Manzanilla-Granados, Zaira Navarrete-Cazales, Miriam Pescador-Rojas, et al.
+- **abs:** https://arxiv.org/abs/2601.13443 · **PDF:** https://arxiv.org/pdf/2601.13443v1
+
+## 20668. MOSLD-Bench: Multilingual Open-Set Learning and Discovery Benchmark for Text Categorization
+
+- **ID:** `2601.13437` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Adriana-Valentina Costache, Daria-Nicoleta Dragomir, Silviu-Florin Gheorghe, et al.
+- **abs:** https://arxiv.org/abs/2601.13437 · **PDF:** https://arxiv.org/pdf/2601.13437v1
+
+## 20669. Distribution-Free Confidence Ellipsoids for Ridge Regression with PAC Bounds
+
+- **ID:** `2601.13436` · **primary:** `stat.ML` · **Submitted:** 2026-01-19
+- **Authors:** Szabolcs Szentpéteri, Balázs Csanád Csáji
+- **abs:** https://arxiv.org/abs/2601.13436 · **PDF:** https://arxiv.org/pdf/2601.13436v1
+
+## 20670. A Learnable Wavelet Transformer for Long-Short Equity Trading and Risk-Adjusted Return Optimization
+
+- **ID:** `2601.13435` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Shuozhe Li, Du Cheng, Leqi Liu
+- **abs:** https://arxiv.org/abs/2601.13435 · **PDF:** https://arxiv.org/pdf/2601.13435v4
+
+## 20671. Trust Me, I'm an Expert: Decoding and Steering Authority Bias in Large Language Models
+
+- **ID:** `2601.13433` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Priyanka Mary Mammen, Emil Joswin, Shankar Venkitachalam
+- **abs:** https://arxiv.org/abs/2601.13433 · **PDF:** https://arxiv.org/pdf/2601.13433v2
+
+## 20672. TrustEnergy: A Unified Framework for Accurate and Reliable User-level Energy Usage Prediction
+
+- **ID:** `2601.13422` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Dahai Yu, Rongchao Xu, Dingyi Zhuang, et al.
+- **abs:** https://arxiv.org/abs/2601.13422 · **PDF:** https://arxiv.org/pdf/2601.13422v1
+
+## 20673. Barriers to Discrete Reasoning with Transformers: A Survey Across Depth, Exactness, and Bandwidth
+
+- **ID:** `2602.11175` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Michelle Yuan, Weiyi Sun, Amir H. Rezaeian, et al.
+- **abs:** https://arxiv.org/abs/2602.11175 · **PDF:** https://arxiv.org/pdf/2602.11175v1
+
+## 20674. Using deep learning for predicting cleansing quality of colon capsule endoscopy images
+
+- **ID:** `2601.13412` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Puneet Sharma, Kristian Dalsbø Hindberg, Benedicte Schelde-Olesen, et al.
+- **abs:** https://arxiv.org/abs/2601.13412 · **PDF:** https://arxiv.org/pdf/2601.13412v1
+
+## 20675. Classifiers in High Dimensional Hilbert Metrics
+
+- **ID:** `2601.13410` · **primary:** `cs.CG` · **Submitted:** 2026-01-19
+- **Authors:** Aditya Acharya, Auguste H. Gezalyan, David M. Mount
+- **abs:** https://arxiv.org/abs/2601.13410 · **PDF:** https://arxiv.org/pdf/2601.13410v1
+
+## 20676. Integrating Virtual Reality and Large Language Models for Team-Based Non-Technical Skills Training and Evaluation in the Operating Room
+
+- **ID:** `2601.13406` · **primary:** `cs.HC` · **Submitted:** 2026-01-19
+- **Authors:** Jacob Barker, Doga Demirel, Cullen Jackson, et al.
+- **abs:** https://arxiv.org/abs/2601.13406 · **PDF:** https://arxiv.org/pdf/2601.13406v1
+
+## 20677. Local-to-Global Logical Explanations for Deep Vision Models
+
+- **ID:** `2601.13404` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Bhavan Vasu, Giuseppe Raffa, Prasad Tadepalli
+- **abs:** https://arxiv.org/abs/2601.13404 · **PDF:** https://arxiv.org/pdf/2601.13404v1
+
+## 20678. LLM-Generated or Human-Written? Comparing Review and Non-Review Papers on ArXiv
+
+- **ID:** `2601.17036` · **primary:** `cs.DL` · **Submitted:** 2026-01-19
+- **Authors:** Yanai Elazar, Maria Antoniak
+- **abs:** https://arxiv.org/abs/2601.17036 · **PDF:** https://arxiv.org/pdf/2601.17036v1
+
+## 20679. Reasoning with Pixel-level Precision: QVLM Architecture and SQuID Dataset for Quantitative Geospatial Analytics
+
+- **ID:** `2601.13401` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Peter A. Massih, Eric Cosatto
+- **abs:** https://arxiv.org/abs/2601.13401 · **PDF:** https://arxiv.org/pdf/2601.13401v1
+
+## 20680. Deep Image Prior with L0 Gradient Regularizer for Image Smoothing
+
+- **ID:** `2601.13400` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Nhat Thanh Tran, Kevin Bui, Jack Xin
+- **abs:** https://arxiv.org/abs/2601.13400 · **PDF:** https://arxiv.org/pdf/2601.13400v1
+
+## 20681. Can LLMs Compress (and Decompress)? Evaluating Code Understanding and Execution via Invertibility
+
+- **ID:** `2601.13398` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Nickil Maveli, Antonio Vergari, Shay B. Cohen
+- **abs:** https://arxiv.org/abs/2601.13398 · **PDF:** https://arxiv.org/pdf/2601.13398v1
+
+## 20682. Beyond Memorization: Testing LLM Reasoning on Unseen Theory of Computation Tasks
+
+- **ID:** `2601.13392` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Shlok Shelat, Jay Raval, Souvik Roy, et al.
+- **abs:** https://arxiv.org/abs/2601.13392 · **PDF:** https://arxiv.org/pdf/2601.13392v1
+
+## 20683. Structured Insight from Unstructured Data: Large Language Models for SDOH-Driven Diabetes Risk Prediction
+
+- **ID:** `2601.13388` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Sasha Ronaghi, Prerit Choudhary, David H Rehkopf, et al.
+- **abs:** https://arxiv.org/abs/2601.13388 · **PDF:** https://arxiv.org/pdf/2601.13388v1
+
+## 20684. Confidence over Time: Confidence Calibration with Temporal Logic for Large Language Model Reasoning
+
+- **ID:** `2601.13387` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Zhenjiang Mao, Anirudhh Venkat, Artem Bisliouk, et al.
+- **abs:** https://arxiv.org/abs/2601.13387 · **PDF:** https://arxiv.org/pdf/2601.13387v1
+
+## 20685. Organ-Aware Attention Improves CT Triage and Classification
+
+- **ID:** `2601.13385` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Lavsen Dahal, Yubraj Bhandari, Geoffrey D. Rubin, et al.
+- **abs:** https://arxiv.org/abs/2601.13385 · **PDF:** https://arxiv.org/pdf/2601.13385v1
+
+## 20686. From Completion to Editing: Unlocking Context-Aware Code Infilling via Search-and-Replace Instruction Tuning
+
+- **ID:** `2601.13384` · **primary:** `cs.SE` · **Submitted:** 2026-01-19
+- **Authors:** Jiajun Zhang, Zeyu Cui, Jiaxi Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.13384 · **PDF:** https://arxiv.org/pdf/2601.13384v1
+
+## 20687. A Lightweight Modular Framework for Constructing Autonomous Agents Driven by Large Language Models: Design, Implementation, and Applications in AgentForge
+
+- **ID:** `2601.13383` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Akbar Anbar Jafari, Cagri Ozcinar, Gholamreza Anbarjafari
+- **abs:** https://arxiv.org/abs/2601.13383 · **PDF:** https://arxiv.org/pdf/2601.13383v1
+
+## 20688. Bounded Minds, Generative Machines: Envisioning Conversational AI that Works with Human Heuristics and Reduces Bias Risk
+
+- **ID:** `2601.13376` · **primary:** `cs.ET` · **Submitted:** 2026-01-19
+- **Authors:** Jiqun Liu
+- **abs:** https://arxiv.org/abs/2601.13376 · **PDF:** https://arxiv.org/pdf/2601.13376v1
+
+## 20689. Recurrent Confidence Chain: Temporal-Aware Uncertainty Quantification in Large Language Models
+
+- **ID:** `2601.13368` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Zhenjiang Mao, Anirudhh Venkat
+- **abs:** https://arxiv.org/abs/2601.13368 · **PDF:** https://arxiv.org/pdf/2601.13368v1
+
+## 20690. CausationEntropy: Pythonic Optimal Causation Entropy
+
+- **ID:** `2601.13365` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Kevin Slote, Jeremie Fish, Erik Bollt
+- **abs:** https://arxiv.org/abs/2601.13365 · **PDF:** https://arxiv.org/pdf/2601.13365v1
+
+## 20691. Improving Geopolitical Forecasts with Bayesian Networks
+
+- **ID:** `2601.13362` · **primary:** `stat.AP` · **Submitted:** 2026-01-19
+- **Authors:** Matthew Martin
+- **abs:** https://arxiv.org/abs/2601.13362 · **PDF:** https://arxiv.org/pdf/2601.13362v1
+
+## 20692. Sockpuppetting: Jailbreaking LLMs Without Optimization Through Output Prefix Injection
+
+- **ID:** `2601.13359` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Asen Dotsinski, Panagiotis Eustratiadis
+- **abs:** https://arxiv.org/abs/2601.13359 · **PDF:** https://arxiv.org/pdf/2601.13359v1
+
+## 20693. The Geometry of Thought: How Scale Restructures Reasoning In Large Language Models
+
+- **ID:** `2601.13358` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Samuel Cyrenius Anderson
+- **abs:** https://arxiv.org/abs/2601.13358 · **PDF:** https://arxiv.org/pdf/2601.13358v2
+
+## 20694. On the Relation of State Space Models and Hidden Markov Models
+
+- **ID:** `2601.13357` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Aydin Ghojogh, M. Hadi Sepanj, Benyamin Ghojogh
+- **abs:** https://arxiv.org/abs/2601.13357 · **PDF:** https://arxiv.org/pdf/2601.13357v1
+
+## 20695. LLM-as-RNN: A Recurrent Language Model for Memory Updates and Sequence Prediction
+
+- **ID:** `2601.13352` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Yuxing Lu, J. Ben Tamo, Weichen Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.13352 · **PDF:** https://arxiv.org/pdf/2601.13352v1
+
+## 20696. Beyond Mapping : Domain-Invariant Representations via Spectral Embedding of Optimal Transport Plans
+
+- **ID:** `2601.13350` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Abdel Djalil Sad Saoud, Fred Maurice Ngolè Mboula, Hanane Slimani
+- **abs:** https://arxiv.org/abs/2601.13350 · **PDF:** https://arxiv.org/pdf/2601.13350v2
+
+## 20697. The AI Genie Phenomenon and Three Types of AI Chatbot Addiction: Escapist Roleplays, Pseudosocial Companions, and Epistemic Rabbit Holes
+
+- **ID:** `2601.13348` · **primary:** `cs.HC` · **Submitted:** 2026-01-19
+- **Authors:** M. Karen Shen, Jessica Huang, Olivia Liang, et al.
+- **abs:** https://arxiv.org/abs/2601.13348 · **PDF:** https://arxiv.org/pdf/2601.13348v1
+
+## 20698. AfroScope: A Framework for Studying the Linguistic Landscape of Africa
+
+- **ID:** `2601.13346` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Sang Yun Kwon, AbdelRahim Elmadany, Muhammad Abdul-Mageed
+- **abs:** https://arxiv.org/abs/2601.13346 · **PDF:** https://arxiv.org/pdf/2601.13346v2
+
+## 20699. MultiST: A Cross-Attention-Based Multimodal Model for Spatial Transcriptomic
+
+- **ID:** `2601.13331` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Wei Wang, Quoc-Toan Ly, Chong Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.13331 · **PDF:** https://arxiv.org/pdf/2601.13331v1
+
+## 20700. RegCheck: A tool for automating comparisons between study registrations and papers
+
+- **ID:** `2601.13330` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Jamie Cummins, Beth Clarke, Ian Hussey, et al.
+- **abs:** https://arxiv.org/abs/2601.13330 · **PDF:** https://arxiv.org/pdf/2601.13330v1
+
+## 20701. Reducing Tokenization Premiums for Low-Resource Languages
+
+- **ID:** `2601.13328` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Geoffrey Churchill, Steven Skiena
+- **abs:** https://arxiv.org/abs/2601.13328 · **PDF:** https://arxiv.org/pdf/2601.13328v1
+
+## 20702. PepEDiff: Zero-Shot Peptide Binder Design via Protein Embedding Diffusion
+
+- **ID:** `2601.13327` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Po-Yu Liang, Tibo Duran, Jun Bai
+- **abs:** https://arxiv.org/abs/2601.13327 · **PDF:** https://arxiv.org/pdf/2601.13327v2
+
+## 20703. Arab Voices: Mapping Standard and Dialectal Arabic Speech Technology
+
+- **ID:** `2601.13319` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Peter Sullivan, AbdelRahim Elmadany, Alcides Alcoba Inciarte, et al.
+- **abs:** https://arxiv.org/abs/2601.13319 · **PDF:** https://arxiv.org/pdf/2601.13319v2
+
+## 20704. Paid Voices vs. Public Feeds: Interpretable Cross-Platform Theme Modeling of Climate Discourse
+
+- **ID:** `2601.13317` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Samantha Sudhoff, Pranav Perumal, Zhaoqing Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.13317 · **PDF:** https://arxiv.org/pdf/2601.13317v1
+
+## 20705. Scaling laws for amplitude surrogates
+
+- **ID:** `2601.13308` · **primary:** `hep-ph` · **Submitted:** 2026-01-19
+- **Authors:** Henning Bahl, Victor Bresó-Pla, Anja Butter, et al.
+- **abs:** https://arxiv.org/abs/2601.13308 · **PDF:** https://arxiv.org/pdf/2601.13308v1
+
+## 20706. Verifying Local Robustness of Pruned Safety-Critical Networks
+
+- **ID:** `2601.13303` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Minh Le, Phuong Cao
+- **abs:** https://arxiv.org/abs/2601.13303 · **PDF:** https://arxiv.org/pdf/2601.13303v1
+
+## 20707. OI-Bench: An Option Injection Benchmark for Evaluating LLM Susceptibility to Directive Interference
+
+- **ID:** `2601.13300` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Yow-Fu Liou, Yu-Chien Tang, Yu-Hsiang Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.13300 · **PDF:** https://arxiv.org/pdf/2601.13300v1
+
+## 20708. CooperBench: Why Coding Agents Cannot be Your Teammates Yet
+
+- **ID:** `2601.13295` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Arpandeep Khatua, Hao Zhu, Peter Tran, et al.
+- **abs:** https://arxiv.org/abs/2601.13295 · **PDF:** https://arxiv.org/pdf/2601.13295v2
+
+## 20709. Scalable and Secure AI Inference in Healthcare: A Comparative Benchmarking of FastAPI and Triton Inference Server on Kubernetes
+
+- **ID:** `2602.00053` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Ratul Ali
+- **abs:** https://arxiv.org/abs/2602.00053 · **PDF:** https://arxiv.org/pdf/2602.00053v1
+
+## 20710. The Tag is the Signal: URL-Agnostic Credibility Scoring for Messages on Telegram
+
+- **ID:** `2601.13294` · **primary:** `cs.SI` · **Submitted:** 2026-01-19
+- **Authors:** Yipeng Wang, Huy Gia Han Vu, Mohit Singhal
+- **abs:** https://arxiv.org/abs/2601.13294 · **PDF:** https://arxiv.org/pdf/2601.13294v1
+
+## 20711. A BERTology View of LLM Orchestrations: Token- and Layer-Selective Probes for Efficient Single-Pass Classification
+
+- **ID:** `2601.13288` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Gonzalo Ariel Meyoyan, Luciano Del Corro
+- **abs:** https://arxiv.org/abs/2601.13288 · **PDF:** https://arxiv.org/pdf/2601.13288v1
+
+## 20712. AI-assisted Protocol Information Extraction For Improved Accuracy and Efficiency in Clinical Trial Workflows
+
+- **ID:** `2602.00052` · **primary:** `cs.IR` · **Submitted:** 2026-01-19
+- **Authors:** Ramtin Babaeipour, François Charest, Madison Wright
+- **abs:** https://arxiv.org/abs/2602.00052 · **PDF:** https://arxiv.org/pdf/2602.00052v1
+
+## 20713. AI Skills Improve Job Prospects: Causal Evidence from a Hiring Experiment
+
+- **ID:** `2601.13286` · **primary:** `econ.GN` · **Submitted:** 2026-01-19
+- **Authors:** Fabian Stephany, Ole Teutloff, Angelo Leone
+- **abs:** https://arxiv.org/abs/2601.13286 · **PDF:** https://arxiv.org/pdf/2601.13286v2
+
+## 20714. Balancing Classification and Calibration Performance in Decision-Making LLMs via Calibration Aware Reinforcement Learning
+
+- **ID:** `2601.13284` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Duygu Nur Yaldiz, Evangelia Spiliopoulou, Zheng Qi, et al.
+- **abs:** https://arxiv.org/abs/2601.13284 · **PDF:** https://arxiv.org/pdf/2601.13284v1
+
+## 20715. Classifier Calibration at Scale: An Empirical Study of Model-Agnostic Post-Hoc Methods
+
+- **ID:** `2601.19944` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Valery Manokhin, Daniel Grønhaug
+- **abs:** https://arxiv.org/abs/2601.19944 · **PDF:** https://arxiv.org/pdf/2601.19944v1
+
+## 20716. Multi-level Monte Carlo Dropout for Efficient Uncertainty Quantification
+
+- **ID:** `2601.13272` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Aaron Pim, Tristan Pryer
+- **abs:** https://arxiv.org/abs/2601.13272 · **PDF:** https://arxiv.org/pdf/2601.13272v1
+
+## 20717. Improving the Safety and Trustworthiness of Medical AI via Multi-Agent Evaluation Loops
+
+- **ID:** `2601.13268` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Zainab Ghafoor, Md Shafiqul Islam, Koushik Howlader, et al.
+- **abs:** https://arxiv.org/abs/2601.13268 · **PDF:** https://arxiv.org/pdf/2601.13268v1
+
+## 20718. Unlearning in LLMs: Methods, Evaluation, and Open Challenges
+
+- **ID:** `2601.13264` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Tyler Lizzo, Larry Heck
+- **abs:** https://arxiv.org/abs/2601.13264 · **PDF:** https://arxiv.org/pdf/2601.13264v1
+
+## 20719. CURE-Med: Curriculum-Informed Reinforcement Learning for Multilingual Medical Reasoning
+
+- **ID:** `2601.13262` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Eric Onyame, Akash Ghosh, Subhadip Baidya, et al.
+- **abs:** https://arxiv.org/abs/2601.13262 · **PDF:** https://arxiv.org/pdf/2601.13262v1
+
+## 20720. Stop Taking Tokenizers for Granted: They Are Core Design Decisions in Large Language Models
+
+- **ID:** `2601.13260` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Sawsan Alqahtani, Mir Tafseer Nayeem, Md Tahmid Rahman Laskar, et al.
+- **abs:** https://arxiv.org/abs/2601.13260 · **PDF:** https://arxiv.org/pdf/2601.13260v2
+
+## 20721. Deep Neural networks for solving high-dimensional parabolic partial differential equations
+
+- **ID:** `2601.13256` · **primary:** `math.NA` · **Submitted:** 2026-01-19
+- **Authors:** Wenzhong Zhang, Zheyuan Hu, Wei Cai, et al.
+- **abs:** https://arxiv.org/abs/2601.13256 · **PDF:** https://arxiv.org/pdf/2601.13256v3
+
+## 20722. A Hybrid Protocol for Large-Scale Semantic Dataset Generation in Low-Resource Languages: The Turkish Semantic Relations Corpus
+
+- **ID:** `2601.13253` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Ebubekir Tosun, Mehmet Emin Buldur, Özay Ezerceli, et al.
+- **abs:** https://arxiv.org/abs/2601.13253 · **PDF:** https://arxiv.org/pdf/2601.13253v1
+
+## 20723. Beyond Cosine Similarity: Taming Semantic Drift and Antonym Intrusion in a 15-Million Node Turkish Synonym Graph
+
+- **ID:** `2601.13251` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Ebubekir Tosun, Mehmet Emin Buldur, Özay Ezerceli, et al.
+- **abs:** https://arxiv.org/abs/2601.13251 · **PDF:** https://arxiv.org/pdf/2601.13251v1
+
+## 20724. Aligning Agentic World Models via Knowledgeable Experience Learning
+
+- **ID:** `2601.13247` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Baochang Ren, Yunzhi Yao, Rui Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.13247 · **PDF:** https://arxiv.org/pdf/2601.13247v1
+
+## 20725. Do Instruction-Tuned Models Always Perform Better Than Base Models? Evidence from Math and Domain-Shifted Benchmarks
+
+- **ID:** `2601.13244` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Prateek Munjal, Clement Christophe, Ronnie Rajan, et al.
+- **abs:** https://arxiv.org/abs/2601.13244 · **PDF:** https://arxiv.org/pdf/2601.13244v1
+
+## 20726. A Comprehensive Evaluation of LLM Reasoning: From Single-Model to Multi-Agent Paradigms
+
+- **ID:** `2601.13243` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Yapeng Li, Jiakuo Yu, Zhixin Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.13243 · **PDF:** https://arxiv.org/pdf/2601.13243v1
+
+## 20727. KOCO-BENCH: Can Large Language Models Leverage Domain Knowledge in Software Development?
+
+- **ID:** `2601.13240` · **primary:** `cs.SE` · **Submitted:** 2026-01-19
+- **Authors:** Xue Jiang, Jiaru Qian, Xianjie Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.13240 · **PDF:** https://arxiv.org/pdf/2601.13240v1
+
+## 20728. A Semantic Decoupling-Based Two-Stage Rainy-Day Attack for Revealing Weather Robustness Deficiencies in Vision-Language Models
+
+- **ID:** `2601.13238` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Chengyin Hu, Xiang Chen, Zhe Jia, et al.
+- **abs:** https://arxiv.org/abs/2601.13238 · **PDF:** https://arxiv.org/pdf/2601.13238v1
+
+## 20729. Pixelwise Uncertainty Quantification of Accelerated MRI Reconstruction
+
+- **ID:** `2601.13236` · **primary:** `eess.IV` · **Submitted:** 2026-01-19
+- **Authors:** Ilias I. Giannakopoulos, Lokesh B Gautham Muthukumar, Yvonne W. Lui, et al.
+- **abs:** https://arxiv.org/abs/2601.13236 · **PDF:** https://arxiv.org/pdf/2601.13236v1
+
+## 20730. RubRIX: Rubric-Driven Risk Mitigation in Caregiver-AI Interactions
+
+- **ID:** `2601.13235` · **primary:** `cs.HC` · **Submitted:** 2026-01-19
+- **Authors:** Drishti Goel, Jeongah Lee, Qiuyue Joy Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.13235 · **PDF:** https://arxiv.org/pdf/2601.13235v1
+
+## 20731. RAG: A Random-Forest-Based Generative Design Framework for Uncertainty-Aware Design of Metamaterials with Complex Functional Response Requirements
+
+- **ID:** `2601.13233` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Bolin Chen, Dex Doksoo Lee, Wei "Wayne'' Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.13233 · **PDF:** https://arxiv.org/pdf/2601.13233v1
+
+## 20732. Autoregressive Models Rival Diffusion Models at ANY-ORDER Generation
+
+- **ID:** `2601.13228` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Tianqi Du, Lizhe Fang, Weijie Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.13228 · **PDF:** https://arxiv.org/pdf/2601.13228v1
+
+## 20733. Insider Knowledge: How Much Can RAG Systems Gain from Evaluation Secrets?
+
+- **ID:** `2601.13227` · **primary:** `cs.IR` · **Submitted:** 2026-01-19
+- **Authors:** Laura Dietz, Bryan Li, Eugene Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.13227 · **PDF:** https://arxiv.org/pdf/2601.13227v2
+
+## 20734. Incorporating Q&A Nuggets into Retrieval-Augmented Generation
+
+- **ID:** `2601.13222` · **primary:** `cs.IR` · **Submitted:** 2026-01-19
+- **Authors:** Laura Dietz, Bryan Li, Gabrielle Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.13222 · **PDF:** https://arxiv.org/pdf/2601.13222v2
+
+## 20735. Beyond Single-shot Writing: Deep Research Agents are Unreliable at Multi-turn Report Revision
+
+- **ID:** `2601.13217` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Bingsen Chen, Boyan Li, Ping Nie, et al.
+- **abs:** https://arxiv.org/abs/2601.13217 · **PDF:** https://arxiv.org/pdf/2601.13217v1
+
+## 20736. Real-Time Deadlines Reveal Temporal Awareness Failures in LLM Strategic Dialogues
+
+- **ID:** `2601.13206` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Neil K. R. Sehgal, Sharath Chandra Guntuku, Lyle Ungar
+- **abs:** https://arxiv.org/abs/2601.13206 · **PDF:** https://arxiv.org/pdf/2601.13206v1
+
+## 20737. Diffusion-Driven Synthetic Tabular Data Generation for Enhanced DoS/DDoS Attack Classification
+
+- **ID:** `2601.13197` · **primary:** `cs.CR` · **Submitted:** 2026-01-19
+- **Authors:** Aravind B, Anirud R. S., Sai Surya Teja N, et al.
+- **abs:** https://arxiv.org/abs/2601.13197 · **PDF:** https://arxiv.org/pdf/2601.13197v2
+
+## 20738. Empirical Risk Minimization with $f$-Divergence Regularization
+
+- **ID:** `2601.13191` · **primary:** `stat.ML` · **Submitted:** 2026-01-19
+- **Authors:** Francisco Daunas, Iñaki Esnaola, Samir M. Perlaza, et al.
+- **abs:** https://arxiv.org/abs/2601.13191 · **PDF:** https://arxiv.org/pdf/2601.13191v1
+
+## 20739. LAViG-FLOW: Latent Autoregressive Video Generation for Fluid Flow Simulations
+
+- **ID:** `2601.13190` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Vittoria De Pellegrini, Tariq Alkhalifah
+- **abs:** https://arxiv.org/abs/2601.13190 · **PDF:** https://arxiv.org/pdf/2601.13190v2
+
+## 20740. Scientific production in the era of Large Language Models
+
+- **ID:** `2601.13187` · **primary:** `cs.DL` · **Submitted:** 2026-01-19
+- **Authors:** Keigo Kusumegi, Xinyu Yang, Paul Ginsparg, et al.
+- **abs:** https://arxiv.org/abs/2601.13187 · **PDF:** https://arxiv.org/pdf/2601.13187v1
+
+## 20741. Prompt Injection Mitigation with Agentic AI, Nested Learning, and AI Sustainability via Semantic Caching
+
+- **ID:** `2601.13186` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Diego Gosmar, Deborah A. Dahl
+- **abs:** https://arxiv.org/abs/2601.13186 · **PDF:** https://arxiv.org/pdf/2601.13186v1
+
+## 20742. OpenExempt: A Diagnostic Benchmark for Legal Reasoning and a Framework for Creating Custom Benchmarks on Demand
+
+- **ID:** `2601.13183` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Sergio Servantez, Sarah B. Lawsky, Rajiv Jain, et al.
+- **abs:** https://arxiv.org/abs/2601.13183 · **PDF:** https://arxiv.org/pdf/2601.13183v1
+
+## 20743. Medical Triage as Pairwise Ranking: A Benchmark for Urgency in Patient Portal Messages
+
+- **ID:** `2601.13178` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Joseph Gatto, Parker Seegmiller, Timothy Burdick, et al.
+- **abs:** https://arxiv.org/abs/2601.13178 · **PDF:** https://arxiv.org/pdf/2601.13178v1
+
+## 20744. Global stability of a Hebbian/anti-Hebbian network for principal subspace learning
+
+- **ID:** `2601.13170` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-19
+- **Authors:** David Lipshutz, Robert J. Lipshutz
+- **abs:** https://arxiv.org/abs/2601.13170 · **PDF:** https://arxiv.org/pdf/2601.13170v1
+
+## 20745. From 100,000+ images to winning the first brain MRI foundation model challenges: Sharing lessons and models
+
+- **ID:** `2601.13166` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Pedro M. Gordaliza, Jaume Banus, Benoît Gérin, et al.
+- **abs:** https://arxiv.org/abs/2601.13166 · **PDF:** https://arxiv.org/pdf/2601.13166v1
+
+## 20746. NeuroShield: A Neuro-Symbolic Framework for Adversarial Robustness
+
+- **ID:** `2601.13162` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Ali Shafiee Sarvestani, Jason Schmidt, Arman Roohi
+- **abs:** https://arxiv.org/abs/2601.13162 · **PDF:** https://arxiv.org/pdf/2601.13162v1
+
+## 20747. Training instability in deep learning follows low-dimensional dynamical principles
+
+- **ID:** `2601.13160` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Zhipeng Zhang, Zhenjie Yao, Kai Li, et al.
+- **abs:** https://arxiv.org/abs/2601.13160 · **PDF:** https://arxiv.org/pdf/2601.13160v1
+
+## 20748. Probe and Skip: Self-Predictive Token Skipping for Efficient Long-Context LLM Inference
+
+- **ID:** `2601.13155` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Zimeng Wu, Donghao Wang, Chaozhe Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.13155 · **PDF:** https://arxiv.org/pdf/2601.13155v2
+
+## 20749. Distributional Reinforcement Learning for Condition-Based Maintenance of Multi-Pump Equipment
+
+- **ID:** `2602.00051` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Takato Yasuno
+- **abs:** https://arxiv.org/abs/2602.00051 · **PDF:** https://arxiv.org/pdf/2602.00051v1
+
+## 20750. SolARED: Solar Active Region Emergence Dataset for Machine Learning Aided Predictions
+
+- **ID:** `2601.13145` · **primary:** `astro-ph.SR` · **Submitted:** 2026-01-19
+- **Authors:** Spiridon Kasapis, Eren Dogan, Irina N. Kitiashvili, et al.
+- **abs:** https://arxiv.org/abs/2601.13145 · **PDF:** https://arxiv.org/pdf/2601.13145v1
+
+## 20751. Forecasting Continuum Intensity for Solar Active Region Emergence Prediction using Transformers
+
+- **ID:** `2601.13144` · **primary:** `astro-ph.SR` · **Submitted:** 2026-01-19
+- **Authors:** Jonas Tirona, Sarang Patil, Spiridon Kasapis, et al.
+- **abs:** https://arxiv.org/abs/2601.13144 · **PDF:** https://arxiv.org/pdf/2601.13144v1
+
+## 20752. FastAV: Efficient Token Pruning for Audio-Visual Large Language Model Inference
+
+- **ID:** `2601.13143` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Chaeyoung Jung, Youngjoon Jang, Seungwoo Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.13143 · **PDF:** https://arxiv.org/pdf/2601.13143v1
+
+## 20753. TVWorld: Foundations for Remote-Control TV Agents
+
+- **ID:** `2601.13142` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Zhantao Ma, Quanfeng Lu, Shuai Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.13142 · **PDF:** https://arxiv.org/pdf/2601.13142v1
+
+## 20754. Adversarial Alignment: Ensuring Value Consistency in Large Language Models for Sensitive Domains
+
+- **ID:** `2601.13137` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Yuan Gao, Zhigang Liu, Xinyu Yao, et al.
+- **abs:** https://arxiv.org/abs/2601.13137 · **PDF:** https://arxiv.org/pdf/2601.13137v2
+
+## 20755. Responsible AI for General-Purpose Systems: Overview, Challenges, and A Path Forward
+
+- **ID:** `2601.13122` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Gourab K Patro, Himanshi Agrawal, Himanshu Gharat, et al.
+- **abs:** https://arxiv.org/abs/2601.13122 · **PDF:** https://arxiv.org/pdf/2601.13122v1
+
+## 20756. Agentic Conversational Search with Contextualized Reasoning via Reinforcement Learning
+
+- **ID:** `2601.13115` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Fengran Mo, Yifan Gao, Sha Li, et al.
+- **abs:** https://arxiv.org/abs/2601.13115 · **PDF:** https://arxiv.org/pdf/2601.13115v1
+
+## 20757. IntAgent: NWDAF-Based Intent LLM Agent Towards Advanced Next Generation Networks
+
+- **ID:** `2601.13114` · **primary:** `cs.NI` · **Submitted:** 2026-01-19
+- **Authors:** Abdelrahman Soliman, Ahmed Refaey, Aiman Erbad, et al.
+- **abs:** https://arxiv.org/abs/2601.13114 · **PDF:** https://arxiv.org/pdf/2601.13114v1
+
+## 20758. CORE-T: COherent REtrieval of Tables for Text-to-SQL
+
+- **ID:** `2601.13111` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Hassan Soliman, Vivek Gupta, Dan Roth, et al.
+- **abs:** https://arxiv.org/abs/2601.13111 · **PDF:** https://arxiv.org/pdf/2601.13111v1
+
+## 20759. The Script Tax: Measuring Tokenization-Driven Efficiency and Latency Disparities in Multilingual Language Models
+
+- **ID:** `2602.11174` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Aradhya Dixit, Shreem Dixit
+- **abs:** https://arxiv.org/abs/2602.11174 · **PDF:** https://arxiv.org/pdf/2602.11174v1
+
+## 20760. Leveraging Lora Fine-Tuning and Knowledge Bases for Construction Identification
+
+- **ID:** `2601.13105` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Liu Kaipeng, Wu Ling
+- **abs:** https://arxiv.org/abs/2601.13105 · **PDF:** https://arxiv.org/pdf/2601.13105v1
+
+## 20761. Approximate full conformal prediction in an RKHS
+
+- **ID:** `2601.13102` · **primary:** `stat.ML` · **Submitted:** 2026-01-19
+- **Authors:** Davidson Lova Razafindrakoto, Alain Celisse, Jérôme Lacaille
+- **abs:** https://arxiv.org/abs/2601.13102 · **PDF:** https://arxiv.org/pdf/2601.13102v2
+
+## 20762. Recursive Meta-Distillation: An Axiomatic Framework for Iterative Knowledge Refinement
+
+- **ID:** `2601.13100` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Aaron R. Flouro, Shawn P. Chadwick
+- **abs:** https://arxiv.org/abs/2601.13100 · **PDF:** https://arxiv.org/pdf/2601.13100v1
+
+## 20763. Alexandria: A Multi-Domain Dialectal Arabic Machine Translation Dataset for Culturally Inclusive and Linguistically Diverse LLMs
+
+- **ID:** `2601.13099` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Abdellah El Mekki, Samar M. Magdy, Houdaifa Atou, et al.
+- **abs:** https://arxiv.org/abs/2601.13099 · **PDF:** https://arxiv.org/pdf/2601.13099v1
+
+## 20764. RM -RF: Reward Model for Run-Free Unit Test Evaluation
+
+- **ID:** `2601.13097` · **primary:** `cs.SE` · **Submitted:** 2026-01-19
+- **Authors:** Elena Bruches, Daniil Grebenkin, Mikhail Klementev, et al.
+- **abs:** https://arxiv.org/abs/2601.13097 · **PDF:** https://arxiv.org/pdf/2601.13097v1
+
+## 20765. Adversarial News and Lost Profits: Manipulating Headlines in LLM-Driven Algorithmic Trading
+
+- **ID:** `2601.13082` · **primary:** `cs.CR` · **Submitted:** 2026-01-19
+- **Authors:** Advije Rizvani, Giovanni Apruzzese, Pavel Laskov
+- **abs:** https://arxiv.org/abs/2601.13082 · **PDF:** https://arxiv.org/pdf/2601.13082v1
+
+## 20766. Polychronous Wave Computing: Timing-Native Address Selection in Spiking Networks
+
+- **ID:** `2601.13079` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-01-19
+- **Authors:** Natalila G. Berloff
+- **abs:** https://arxiv.org/abs/2601.13079 · **PDF:** https://arxiv.org/pdf/2601.13079v1
+
+## 20767. METIS: Mentoring Engine for Thoughtful Inquiry & Solutions
+
+- **ID:** `2601.13075` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Abhinav Rajeev Kumar, Dhruv Trehan, Paras Chopra
+- **abs:** https://arxiv.org/abs/2601.13075 · **PDF:** https://arxiv.org/pdf/2601.13075v1
+
+## 20768. Author-in-the-Loop Response Generation and Evaluation: Integrating Author Expertise and Intent in Responses to Peer Review
+
+- **ID:** `2602.11173` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Qian Ruan, Iryna Gurevych
+- **abs:** https://arxiv.org/abs/2602.11173 · **PDF:** https://arxiv.org/pdf/2602.11173v1
+
+## 20769. MagicGUI-RMS: A Multi-Agent Reward Model System for Self-Evolving GUI Agents via Automated Feedback Reflux
+
+- **ID:** `2601.13060` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Zecheng Li, Zhihui Cao, Wenke Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.13060 · **PDF:** https://arxiv.org/pdf/2601.13060v1
+
+## 20770. TinyML-Enabled IoT for Sustainable Precision Irrigation
+
+- **ID:** `2601.13054` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Kamogelo Taueatsoala, Caitlyn Daniels, Angelina J. Ramsunar, et al.
+- **abs:** https://arxiv.org/abs/2601.13054 · **PDF:** https://arxiv.org/pdf/2601.13054v1
+
+## 20771. Profiling German Text Simplification with Interpretable Model-Fingerprints
+
+- **ID:** `2601.13050` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Lars Klöser, Mika Beele, Bodo Kraft
+- **abs:** https://arxiv.org/abs/2601.13050 · **PDF:** https://arxiv.org/pdf/2601.13050v1
+
+## 20772. Analysis of Long Range Dependency Understanding in State Space Models
+
+- **ID:** `2601.13048` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Srividya Ravikumar, Abhinav Anand, Shweta Verma, et al.
+- **abs:** https://arxiv.org/abs/2601.13048 · **PDF:** https://arxiv.org/pdf/2601.13048v1
+
+## 20773. Typhoon ASR Real-time: FastConformer-Transducer for Thai Automatic Speech Recognition
+
+- **ID:** `2601.13044` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Warit Sirichotedumrong, Adisai Na-Thalang, Potsawee Manakul, et al.
+- **abs:** https://arxiv.org/abs/2601.13044 · **PDF:** https://arxiv.org/pdf/2601.13044v1
+
+## 20774. SASA: Semantic-Aware Contrastive Learning Framework with Separated Attention for Triple Classification
+
+- **ID:** `2601.13035` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Xu Xiaodan, Hu Xiaolin
+- **abs:** https://arxiv.org/abs/2601.13035 · **PDF:** https://arxiv.org/pdf/2601.13035v1
+
+## 20775. Diagnosis Support of Sickle Cell Anemia by Classifying Red Blood Cell Shape in Peripheral Blood Images
+
+- **ID:** `2601.17032` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Wilkie Delgado-Font, Miriela Escobedo-Nicot, Manuel González-Hidalgo, et al.
+- **abs:** https://arxiv.org/abs/2601.17032 · **PDF:** https://arxiv.org/pdf/2601.17032v1
+
+## 20776. ReCQR: Incorporating conversational query rewriting to improve Multimodal Image Retrieval
+
+- **ID:** `2603.26669` · **primary:** `cs.IR` · **Submitted:** 2026-01-19
+- **Authors:** Yuan Hu, ZhiYu Cao, PeiFeng Li, et al.
+- **abs:** https://arxiv.org/abs/2603.26669 · **PDF:** https://arxiv.org/pdf/2603.26669v1
+
+## 20777. Tears or Cheers? Benchmarking LLMs via Culturally Elicited Distinct Affective Responses
+
+- **ID:** `2601.13024` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Chongyuan Dai, Yaling Shen, Jinpeng Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.13024 · **PDF:** https://arxiv.org/pdf/2601.13024v1
+
+## 20778. Enhancing Generalization in Sickle Cell Disease Diagnosis through Ensemble Methods and Feature Importance Analysis
+
+- **ID:** `2601.13021` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Nataša Petrović, Gabriel Moyà-Alcover, Antoni Jaume-i-Capó, et al.
+- **abs:** https://arxiv.org/abs/2601.13021 · **PDF:** https://arxiv.org/pdf/2601.13021v1
+
+## 20779. PASs-MoE: Mitigating Misaligned Co-drift among Router and Experts via Pathway Activation Subspaces for Continual Learning
+
+- **ID:** `2601.13020` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Zhiyan Hou, Haiyun Guo, Haokai Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.13020 · **PDF:** https://arxiv.org/pdf/2601.13020v1
+
+## 20780. Exploring the Interpretability of Forecasting Models for Energy Balancing Market
+
+- **ID:** `2602.00049` · **primary:** `q-fin.ST` · **Submitted:** 2026-01-19
+- **Authors:** Oskar Våle, Shiliang Zhang, Sabita Maharjan, et al.
+- **abs:** https://arxiv.org/abs/2602.00049 · **PDF:** https://arxiv.org/pdf/2602.00049v1
+
+## 20781. Synthesizing the Virtual Advocate: A Multi-Persona Speech Generation Framework for Diverse Linguistic Jurisdictions in Indic Languages
+
+- **ID:** `2602.11172` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Aniket Deroy
+- **abs:** https://arxiv.org/abs/2602.11172 · **PDF:** https://arxiv.org/pdf/2602.11172v1
+
+## 20782. Quantum Circuit-Based Learning Models: Bridging Quantum Computing and Machine Learning
+
+- **ID:** `2602.00048` · **primary:** `quant-ph` · **Submitted:** 2026-01-19
+- **Authors:** Fan Fan, Yilei Shi, Mihai Datcu, et al.
+- **abs:** https://arxiv.org/abs/2602.00048 · **PDF:** https://arxiv.org/pdf/2602.00048v1
+
+## 20783. Bi-Attention HateXplain : Taking into account the sequential aspect of data during explainability in a multi-task context
+
+- **ID:** `2601.13018` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Ghislain Dorian Tchuente Mondjo
+- **abs:** https://arxiv.org/abs/2601.13018 · **PDF:** https://arxiv.org/pdf/2601.13018v1
+
+## 20784. HT-GNN: Hyper-Temporal Graph Neural Network for Customer Lifetime Value Prediction in Baidu Ads
+
+- **ID:** `2601.13013` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Xiaohui Zhao, Xinjian Zhao, Jiahui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.13013 · **PDF:** https://arxiv.org/pdf/2601.13013v1
+
+## 20785. In Vino Veritas and Vulnerabilities: Examining LLM Safety via Drunk Language Inducement
+
+- **ID:** `2601.22169` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Anudeex Shetty, Aditya Joshi, Salil S. Kanhere
+- **abs:** https://arxiv.org/abs/2601.22169 · **PDF:** https://arxiv.org/pdf/2601.22169v1
+
+## 20786. ArchAgent: Scalable Legacy Software Architecture Recovery with LLMs
+
+- **ID:** `2601.13007` · **primary:** `cs.SE` · **Submitted:** 2026-01-19
+- **Authors:** Rusheng Pan, Bingcheng Mao, Tianyi Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.13007 · **PDF:** https://arxiv.org/pdf/2601.13007v1
+
+## 20787. Lightweight Edge Learning via Dataset Pruning
+
+- **ID:** `2602.00047` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Laha Ale, Hu Luo, Mingsheng Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.00047 · **PDF:** https://arxiv.org/pdf/2602.00047v1
+
+## 20788. OFA-MAS: One-for-All Multi-Agent System Topology Design based on Mixture-of-Experts Graph Generative Models
+
+- **ID:** `2601.12996` · **primary:** `cs.MA` · **Submitted:** 2026-01-19
+- **Authors:** Shiyuan Li, Yixin Liu, Yu Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.12996 · **PDF:** https://arxiv.org/pdf/2601.12996v1
+
+## 20789. Graph Reasoning Paradigm: Structured and Symbolic Reasoning with Topology-Aware Reinforcement Learning for Large Language Models
+
+- **ID:** `2601.12995` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Runxuan Liu, Xianhao Ou, Xinyan Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.12995 · **PDF:** https://arxiv.org/pdf/2601.12995v1
+
+## 20790. RAGExplorer: A Visual Analytics System for the Comparative Diagnosis of RAG Systems
+
+- **ID:** `2601.12991` · **primary:** `cs.HC` · **Submitted:** 2026-01-19
+- **Authors:** Haoyu Tian, Yingchaojie Feng, Zhen Wen, et al.
+- **abs:** https://arxiv.org/abs/2601.12991 · **PDF:** https://arxiv.org/pdf/2601.12991v2
+
+## 20791. Beyond Visual Realism: Toward Reliable Financial Time Series Generation
+
+- **ID:** `2601.12990` · **primary:** `q-fin.ST` · **Submitted:** 2026-01-19
+- **Authors:** Fan Zhang, Jiabin Luo, Zheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.12990 · **PDF:** https://arxiv.org/pdf/2601.12990v1
+
+## 20792. PaperGuide: Making Small Language-Model Paper-Reading Agents More Efficient
+
+- **ID:** `2601.12988` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Zijian Wang, Tiancheng Huang, Hanqi Li, et al.
+- **abs:** https://arxiv.org/abs/2601.12988 · **PDF:** https://arxiv.org/pdf/2601.12988v1
+
+## 20793. Extending Beacon to Hindi: Cultural Adaptation Drives Cross-Lingual Sycophancy
+
+- **ID:** `2602.00046` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Sarthak Sattigeri
+- **abs:** https://arxiv.org/abs/2602.00046 · **PDF:** https://arxiv.org/pdf/2602.00046v1
+
+## 20794. ChartAttack: Testing the Vulnerability of LLMs to Malicious Prompting in Chart Generation
+
+- **ID:** `2601.12983` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Jesus-German Ortiz-Barajas, Jonathan Tonglet, Vivek Gupta, et al.
+- **abs:** https://arxiv.org/abs/2601.12983 · **PDF:** https://arxiv.org/pdf/2601.12983v2
+
+## 20795. Early Prediction of Type 2 Diabetes Using Multimodal data and Tabular Transformers
+
+- **ID:** `2601.12981` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Sulaiman Khan, Md. Rafiul Biswas, Zubair Shah
+- **abs:** https://arxiv.org/abs/2601.12981 · **PDF:** https://arxiv.org/pdf/2601.12981v1
+
+## 20796. The Bitter Lesson of Diffusion Language Models for Agentic Workflows: A Comprehensive Reality Check
+
+- **ID:** `2601.12979` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Qingyu Lu, Liang Ding, Kanjian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.12979 · **PDF:** https://arxiv.org/pdf/2601.12979v2
+
+## 20797. Bridging the Knowledge-Action Gap by Evaluating LLMs in Dynamic Dental Clinical Scenarios
+
+- **ID:** `2601.12974` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Hongyang Ma, Tiantian Gu, Huaiyuan Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.12974 · **PDF:** https://arxiv.org/pdf/2601.12974v1
+
+## 20798. Pardon? Evaluating Conversational Repair in Large Audio-Language Models
+
+- **ID:** `2601.12973` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Shuanghong Huang, Jinlei Xu, Youchao Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.12973 · **PDF:** https://arxiv.org/pdf/2601.12973v1
+
+## 20799. Architecture-Optimization Co-Design for Physics-Informed Neural Networks Via Attentive Representations and Conflict-Resolved Gradients
+
+- **ID:** `2601.12971` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Pancheng Niu, Jun Guo, Qiaolin He, et al.
+- **abs:** https://arxiv.org/abs/2601.12971 · **PDF:** https://arxiv.org/pdf/2601.12971v1
+
+## 20800. Lombard Speech Synthesis for Any Voice with Controllable Style Embeddings
+
+- **ID:** `2601.12966` · **primary:** `cs.SD` · **Submitted:** 2026-01-19
+- **Authors:** Seymanur Akti, Alexander Waibel
+- **abs:** https://arxiv.org/abs/2601.12966 · **PDF:** https://arxiv.org/pdf/2601.12966v1
+
+## 20801. Deterministic Dynamics of Sampling Processes in Score-Based Diffusion Models with Multiplicative Noise Conditioning
+
+- **ID:** `2601.12965` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Doheon Kim
+- **abs:** https://arxiv.org/abs/2601.12965 · **PDF:** https://arxiv.org/pdf/2601.12965v1
+
+## 20802. Trustworthy Data-driven Chronological Age Estimation from Panoramic Dental Images
+
+- **ID:** `2601.12960` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Ainhoa Vivel-Couso, Nicolás Vila-Blanco, María J. Carreira, et al.
+- **abs:** https://arxiv.org/abs/2601.12960 · **PDF:** https://arxiv.org/pdf/2601.12960v1
+
+## 20803. Beyond Accuracy: Characterizing Code Comprehension Capabilities in (Large) Language Models
+
+- **ID:** `2601.12951` · **primary:** `cs.SE` · **Submitted:** 2026-01-19
+- **Authors:** Felix Mächtle, Jan-Niclas Serr, Nils Loose, et al.
+- **abs:** https://arxiv.org/abs/2601.12951 · **PDF:** https://arxiv.org/pdf/2601.12951v1
+
+## 20804. AI-generated data contamination erodes pathological variability and diagnostic reliability
+
+- **ID:** `2601.12946` · **primary:** `cs.CY` · **Submitted:** 2026-01-19
+- **Authors:** Hongyu He, Shaowen Xiang, Ye Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.12946 · **PDF:** https://arxiv.org/pdf/2601.12946v4
+
+## 20805. A Component-Based Survey of Interactions between Large Language Models and Multi-Armed Bandits
+
+- **ID:** `2601.12945` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Siguang Chen, Chunli Lv, Miao Xie
+- **abs:** https://arxiv.org/abs/2601.12945 · **PDF:** https://arxiv.org/pdf/2601.12945v3
+
+## 20806. Active Inference-Driven World Modeling for Adaptive UAV Swarm Trajectory Design
+
+- **ID:** `2601.12939` · **primary:** `cs.RO` · **Submitted:** 2026-01-19
+- **Authors:** Kaleem Arshid, Ali Krayani, Lucio Marcenaro, et al.
+- **abs:** https://arxiv.org/abs/2601.12939 · **PDF:** https://arxiv.org/pdf/2601.12939v1
+
+## 20807. The Post-Turing Condition: Conceptualising Artificial Subjectivity and Synthetic Sociality
+
+- **ID:** `2601.12938` · **primary:** `cs.CY` · **Submitted:** 2026-01-19
+- **Authors:** Thorsten Jelinek, Patrick Glauner, Alvin Wang Graylin, et al.
+- **abs:** https://arxiv.org/abs/2601.12938 · **PDF:** https://arxiv.org/pdf/2601.12938v1
+
+## 20808. On the Evidentiary Limits of Membership Inference for Copyright Auditing
+
+- **ID:** `2601.12937` · **primary:** `cs.CR` · **Submitted:** 2026-01-19
+- **Authors:** Murat Bilgehan Ertan, Emirhan Böge, Min Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.12937 · **PDF:** https://arxiv.org/pdf/2601.12937v1
+
+## 20809. Online Continual Learning for Time Series: a Natural Score-driven Approach
+
+- **ID:** `2601.12931` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Edoardo Urettini, Daniele Atzeni, Ioanna-Yvonni Tsaknaki, et al.
+- **abs:** https://arxiv.org/abs/2601.12931 · **PDF:** https://arxiv.org/pdf/2601.12931v1
+
+## 20810. Membership Inference Test: Auditing Training Data in Object Classification Models
+
+- **ID:** `2601.12929` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Gonzalo Mancera, Daniel DeAlcala, Aythami Morales, et al.
+- **abs:** https://arxiv.org/abs/2601.12929 · **PDF:** https://arxiv.org/pdf/2601.12929v1
+
+## 20811. An efficient heuristic for geometric analysis of cell deformations
+
+- **ID:** `2601.12928` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Yaima Paz Soto, Silena Herold Garcia, Ximo Gual-Arnau, et al.
+- **abs:** https://arxiv.org/abs/2601.12928 · **PDF:** https://arxiv.org/pdf/2601.12928v1
+
+## 20812. ForeDiffusion: Foresight-Conditioned Diffusion Policy via Future View Construction for Robot Manipulation
+
+- **ID:** `2601.12925` · **primary:** `cs.RO` · **Submitted:** 2026-01-19
+- **Authors:** Weize Xie, Yi Ding, Ying He, et al.
+- **abs:** https://arxiv.org/abs/2601.12925 · **PDF:** https://arxiv.org/pdf/2601.12925v1
+
+## 20813. Your Privacy Depends on Others: Collusion Vulnerabilities in Individual Differential Privacy
+
+- **ID:** `2601.12922` · **primary:** `cs.CR` · **Submitted:** 2026-01-19
+- **Authors:** Johannes Kaiser, Alexander Ziller, Eleni Triantafillou, et al.
+- **abs:** https://arxiv.org/abs/2601.12922 · **PDF:** https://arxiv.org/pdf/2601.12922v1
+
+## 20814. Augmenting Parameter-Efficient Pre-trained Language Models with Large Language Models
+
+- **ID:** `2602.02501` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Saurabh Anand, Shubham Malaviya, Manish Shukla, et al.
+- **abs:** https://arxiv.org/abs/2602.02501 · **PDF:** https://arxiv.org/pdf/2602.02501v1
+
+## 20815. Injecting Knowledge from Social Science Journals to Improve Indonesian Cultural Understanding by LLMs
+
+- **ID:** `2601.12921` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Adimulya Kartiyasa, Bao Gia Cao, Boyang Li
+- **abs:** https://arxiv.org/abs/2601.12921 · **PDF:** https://arxiv.org/pdf/2601.12921v1
+
+## 20816. Dynamic Hand Gesture Recognition for Robot Manipulator Tasks
+
+- **ID:** `2601.12918` · **primary:** `cs.RO` · **Submitted:** 2026-01-19
+- **Authors:** Dharmendra Sharma, Peeyush Thakur, Sandeep Gupta, et al.
+- **abs:** https://arxiv.org/abs/2601.12918 · **PDF:** https://arxiv.org/pdf/2601.12918v1
+
+## 20817. CooperLLM: Cloud-Edge-End Cooperative Federated Fine-tuning for LLMs via ZOO-based Gradient Correction
+
+- **ID:** `2601.12917` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** He Sun, Jinrui Zhou, Li Li, et al.
+- **abs:** https://arxiv.org/abs/2601.12917 · **PDF:** https://arxiv.org/pdf/2601.12917v1
+
+## 20818. Actionable Interpretability Must Be Defined in Terms of Symmetries
+
+- **ID:** `2601.12913` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Pietro Barbiero, Mateo Espinosa Zarlenga, Francesco Giannini, et al.
+- **abs:** https://arxiv.org/abs/2601.12913 · **PDF:** https://arxiv.org/pdf/2601.12913v3
+
+## 20819. Human Emotion Verification by Action Languages via Answer Set Programming
+
+- **ID:** `2601.12912` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Andreas Brännström, Juan Carlos Nieves
+- **abs:** https://arxiv.org/abs/2601.12912 · **PDF:** https://arxiv.org/pdf/2601.12912v1
+
+## 20820. SciCoQA: Quality Assurance for Scientific Paper--Code Alignment
+
+- **ID:** `2601.12910` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Tim Baumgärtner, Iryna Gurevych
+- **abs:** https://arxiv.org/abs/2601.12910 · **PDF:** https://arxiv.org/pdf/2601.12910v2
+
+## 20821. Gated Differentiable Working Memory for Long-Context Language Modeling
+
+- **ID:** `2601.12906` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Lingrui Mei, Shenghua Liu, Yiwei Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.12906 · **PDF:** https://arxiv.org/pdf/2601.12906v1
+
+## 20822. From Prefix Cache to Fusion RAG Cache: Accelerating LLM Inference in Retrieval-Augmented Generation
+
+- **ID:** `2601.12904` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Jiahao Wang, Weiyu Xie, Mingxing Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.12904 · **PDF:** https://arxiv.org/pdf/2601.12904v1
+
+## 20823. Deep Temporal Graph Clustering: A Comprehensive Benchmark and Datasets
+
+- **ID:** `2601.12903` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Meng Liu, Ke Liang, Siwei Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.12903 · **PDF:** https://arxiv.org/pdf/2601.12903v1
+
+## 20824. Supervised Learning for the (s,S) Inventory Model with General Interarrival Demands and General Lead Times
+
+- **ID:** `2601.12900` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Eliran Sherzer, Yonit Barron
+- **abs:** https://arxiv.org/abs/2601.12900 · **PDF:** https://arxiv.org/pdf/2601.12900v1
+
+## 20825. TwoHead-SwinFPN: A Unified DL Architecture for Synthetic Manipulation, Detection and Localization in Identity Documents
+
+- **ID:** `2601.12895` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Chan Naseeb, Adeel Ashraf Cheema, Hassan Sami, et al.
+- **abs:** https://arxiv.org/abs/2601.12895 · **PDF:** https://arxiv.org/pdf/2601.12895v1
+
+## 20826. AdaNODEs: Test Time Adaptation for Time Series Forecasting Using Neural ODEs
+
+- **ID:** `2601.12893` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Ting Dang, Soumyajit Chatterjee, Hong Jia, et al.
+- **abs:** https://arxiv.org/abs/2601.12893 · **PDF:** https://arxiv.org/pdf/2601.12893v1
+
+## 20827. Communication Methods in Multi-Agent Reinforcement Learning
+
+- **ID:** `2601.12886` · **primary:** `cs.MA` · **Submitted:** 2026-01-19
+- **Authors:** Christoph Wittner
+- **abs:** https://arxiv.org/abs/2601.12886 · **PDF:** https://arxiv.org/pdf/2601.12886v1
+
+## 20828. YOLO26: An Analysis of NMS-Free End to End Framework for Real-Time Object Detection
+
+- **ID:** `2601.12882` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Sudip Chakrabarty
+- **abs:** https://arxiv.org/abs/2601.12882 · **PDF:** https://arxiv.org/pdf/2601.12882v2
+
+## 20829. Hierarchical Sparse Circuit Extraction from Billion-Parameter Language Models through Scalable Attribution Graph Decomposition
+
+- **ID:** `2601.12879` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Mohammed Mudassir Uddin, Shahnawaz Alam, Mohammed Kaif Pasha
+- **abs:** https://arxiv.org/abs/2601.12879 · **PDF:** https://arxiv.org/pdf/2601.12879v1
+
+## 20830. Race, Ethnicity and Their Implication on Bias in Large Language Models
+
+- **ID:** `2601.12868` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Shiyue Hu, Ruizhe Li, Yanjun Gao
+- **abs:** https://arxiv.org/abs/2601.12868 · **PDF:** https://arxiv.org/pdf/2601.12868v1
+
+## 20831. PDFInspect: A Unified Feature Extraction Framework for Malicious Document Detection
+
+- **ID:** `2601.12866` · **primary:** `cs.CR` · **Submitted:** 2026-01-19
+- **Authors:** Sharmila S P
+- **abs:** https://arxiv.org/abs/2601.12866 · **PDF:** https://arxiv.org/pdf/2601.12866v1
+
+## 20832. Generating Cyclic Conformers with Flow Matching in Cremer-Pople Coordinates
+
+- **ID:** `2601.12859` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Luca Schaufelberger, Aline Hartgers, Kjell Jorner
+- **abs:** https://arxiv.org/abs/2601.12859 · **PDF:** https://arxiv.org/pdf/2601.12859v1
+
+## 20833. Tracing the Data Trail: A Survey of Data Provenance, Transparency and Traceability in LLMs
+
+- **ID:** `2601.14311` · **primary:** `cs.CR` · **Submitted:** 2026-01-19
+- **Authors:** Richard Hohensinner, Belgin Mutlu, Inti Gabriel Mendoza Estrada, et al.
+- **abs:** https://arxiv.org/abs/2601.14311 · **PDF:** https://arxiv.org/pdf/2601.14311v1
+
+## 20834. Data-Efficient Meningioma Segmentation via Implicit Spatiotemporal Mixing and Sim2Real Semantic Injection
+
+- **ID:** `2601.17031` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Yunhao Xu, Fuquan Zong, Yexuan Xing, et al.
+- **abs:** https://arxiv.org/abs/2601.17031 · **PDF:** https://arxiv.org/pdf/2601.17031v1
+
+## 20835. Mining Citywide Dengue Spread Patterns in Singapore Through Hotspot Dynamics from Open Web Data
+
+- **ID:** `2601.12856` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Liping Huang, Gaoxi Xiao, Stefan Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.12856 · **PDF:** https://arxiv.org/pdf/2601.12856v2
+
+## 20836. When LLMs Imagine People: A Human-Centered Persona Brainstorm Audit for Bias and Fairness in Creative Applications
+
+- **ID:** `2602.00044` · **primary:** `cs.CY` · **Submitted:** 2026-01-19
+- **Authors:** Hongliu Cao, Eoin Thomas, Rodrigo Acuna Agost
+- **abs:** https://arxiv.org/abs/2602.00044 · **PDF:** https://arxiv.org/pdf/2602.00044v2
+
+## 20837. The Cost of EFX: Generalized-Mean Welfare and Complexity Dichotomies with Few Surplus Items
+
+- **ID:** `2601.12849` · **primary:** `cs.GT` · **Submitted:** 2026-01-19
+- **Authors:** Eugene Lim, Tzeh Yuan Neoh, Nicholas Teh
+- **abs:** https://arxiv.org/abs/2601.12849 · **PDF:** https://arxiv.org/pdf/2601.12849v1
+
+## 20838. Rapport du Projet de Recherche TRAIMA
+
+- **ID:** `2601.12844` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Julie Rançon, Jean-François Cerisier, Emilie Remond, et al.
+- **abs:** https://arxiv.org/abs/2601.12844 · **PDF:** https://arxiv.org/pdf/2601.12844v1
+
+## 20839. SCULPT: Constraint-Guided Pruned MCTS that Carves Efficient Paths for Mathematical Reasoning
+
+- **ID:** `2601.12842` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Qitong Fang, Haotian Li, Xu Wang
+- **abs:** https://arxiv.org/abs/2601.12842 · **PDF:** https://arxiv.org/pdf/2601.12842v1
+
+## 20840. Knowledge-Integrated Representation Learning for Crypto Anomaly Detection under Extreme Label Scarcity; Relational Domain-Logic Integration with Retrieval-Grounded Context and Path-Level Explanations
+
+- **ID:** `2601.12839` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Gyuyeon Na, Minjung Park, Soyoun Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.12839 · **PDF:** https://arxiv.org/pdf/2601.12839v1
+
+## 20841. Cognition spaces: natural, artificial, and hybrid
+
+- **ID:** `2601.12837` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-19
+- **Authors:** Ricard Solé, Luis F Seoane, Jordi Pla-Mauri, et al.
+- **abs:** https://arxiv.org/abs/2601.12837 · **PDF:** https://arxiv.org/pdf/2601.12837v1
+
+## 20842. Efficient Hyper-Parameter Search for LoRA via Language-aided Bayesian Optimization
+
+- **ID:** `2602.11171` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Baek Seong-Eun, Lee Jung-Mok, Kim Sung-Bin, et al.
+- **abs:** https://arxiv.org/abs/2602.11171 · **PDF:** https://arxiv.org/pdf/2602.11171v1
+
+## 20843. MirrorGuard: Toward Secure Computer-Use Agents via Simulation-to-Real Reasoning Correction
+
+- **ID:** `2601.12822` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Wenqi Zhang, Yulin Shen, Changyue Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.12822 · **PDF:** https://arxiv.org/pdf/2601.12822v1
+
+## 20844. Fisher-Orthogonal Projected Natural Gradient Descent for Continual Learning
+
+- **ID:** `2601.12816` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Ishir Garg, Neel Kolhe, Andy Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.12816 · **PDF:** https://arxiv.org/pdf/2601.12816v2
+
+## 20845. Multimodal Multi-Agent Empowered Legal Judgment Prediction
+
+- **ID:** `2601.12815` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Zhaolu Kang, Junhao Gong, Qingxi Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.12815 · **PDF:** https://arxiv.org/pdf/2601.12815v5
+
+## 20846. Do Clinical Question Answering Systems Really Need Specialised Medical Fine Tuning?
+
+- **ID:** `2601.12812` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Sushant Kumar Ray, Gautam Siddharth Kashyap, Sahil Tripathi, et al.
+- **abs:** https://arxiv.org/abs/2601.12812 · **PDF:** https://arxiv.org/pdf/2601.12812v1
+
+## 20847. Left-Right Symmetry Breaking in CLIP-style Vision-Language Models Trained on Synthetic Spatial-Relation Data
+
+- **ID:** `2601.12809` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Takaki Yamamoto, Chihiro Noguchi, Toshihiro Tanizawa
+- **abs:** https://arxiv.org/abs/2601.12809 · **PDF:** https://arxiv.org/pdf/2601.12809v1
+
+## 20848. Joint Source-Channel-Generation Coding: From Distortion-oriented Reconstruction to Semantic-consistent Generation
+
+- **ID:** `2601.12808` · **primary:** `cs.IT` · **Submitted:** 2026-01-19
+- **Authors:** Tong Wu, Zhiyong Chen, Guo Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.12808 · **PDF:** https://arxiv.org/pdf/2601.12808v1
+
+## 20849. Semi-supervised Instruction Tuning for Large Language Models on Text-Attributed Graphs
+
+- **ID:** `2601.12807` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Zixing Song, Irwin King
+- **abs:** https://arxiv.org/abs/2601.12807 · **PDF:** https://arxiv.org/pdf/2601.12807v1
+
+## 20850. CORVUS: Red-Teaming Hallucination Detectors via Internal Signal Camouflage in Large Language Models
+
+- **ID:** `2601.14310` · **primary:** `cs.CR` · **Submitted:** 2026-01-19
+- **Authors:** Nay Myat Min, Long H. Pham, Hongyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.14310 · **PDF:** https://arxiv.org/pdf/2601.14310v1
+
+## 20851. SciHorizon-GENE: Benchmarking LLM for Life Sciences Inference from Gene Knowledge to Functional Understanding
+
+- **ID:** `2601.12805` · **primary:** `q-bio.GN` · **Submitted:** 2026-01-19
+- **Authors:** Xiaohan Huang, Meng Xiao, Chuan Qin, et al.
+- **abs:** https://arxiv.org/abs/2601.12805 · **PDF:** https://arxiv.org/pdf/2601.12805v2
+
+## 20852. SL-CBM: Enhancing Concept Bottleneck Models with Semantic Locality for Better Interpretability
+
+- **ID:** `2601.12804` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Hanwei Zhang, Luo Cheng, Rui Wen, et al.
+- **abs:** https://arxiv.org/abs/2601.12804 · **PDF:** https://arxiv.org/pdf/2601.12804v1
+
+## 20853. FRoM-W1: Towards General Humanoid Whole-Body Control with Language Instructions
+
+- **ID:** `2601.12799` · **primary:** `cs.RO` · **Submitted:** 2026-01-19
+- **Authors:** Peng Li, Zihan Zhuang, Yangfan Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.12799 · **PDF:** https://arxiv.org/pdf/2601.12799v1
+
+## 20854. PRIME: Policy-Reinforced Iterative Multi-agent Execution for Algorithmic Reasoning in Large Language Models
+
+- **ID:** `2602.11170` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Jiawei Xu, Zhenyu Yu, Ziqian Bi, et al.
+- **abs:** https://arxiv.org/abs/2602.11170 · **PDF:** https://arxiv.org/pdf/2602.11170v1
+
+## 20855. ECGomics: An Open Platform for AI-ECG Digital Biomarker Discovery
+
+- **ID:** `2601.15326` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-19
+- **Authors:** Deyun Zhang, Jun Li, Shijia Geng, et al.
+- **abs:** https://arxiv.org/abs/2601.15326 · **PDF:** https://arxiv.org/pdf/2601.15326v1
+
+## 20856. Distilling Time Series Foundation Models for Efficient Forecasting
+
+- **ID:** `2601.12785` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Yuqi Li, Kuiye Ding, Chuanguang Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.12785 · **PDF:** https://arxiv.org/pdf/2601.12785v1
+
+## 20857. VIRO: Robust and Efficient Neuro-Symbolic Reasoning with Verification for Referring Expression Comprehension
+
+- **ID:** `2601.12781` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Hyejin Park, Junhyuk Kwon, Suha Kwak, et al.
+- **abs:** https://arxiv.org/abs/2601.12781 · **PDF:** https://arxiv.org/pdf/2601.12781v2
+
+## 20858. Open Vocabulary Panoptic Segmentation With Retrieval Augmentation
+
+- **ID:** `2601.12779` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Nafis Sadeq, Qingfeng Liu, Mostafa El-Khamy
+- **abs:** https://arxiv.org/abs/2601.12779 · **PDF:** https://arxiv.org/pdf/2601.12779v1
+
+## 20859. Eddy-Resolving Global Ocean Forecasting with Multi-Scale Graph Neural Networks
+
+- **ID:** `2601.12775` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Yuta Hirabayashi, Daisuke Matusoka, Konobu Kimura
+- **abs:** https://arxiv.org/abs/2601.12775 · **PDF:** https://arxiv.org/pdf/2601.12775v1
+
+## 20860. JSR-GFNet: Jamming-to-Signal Ratio-Aware Dynamic Gating for Interference Classification in future Cognitive Global Navigation Satellite Systems
+
+- **ID:** `2602.00042` · **primary:** `eess.SP` · **Submitted:** 2026-01-19
+- **Authors:** Zhihan Zeng, Hongyuan Shu, Kaihe Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00042 · **PDF:** https://arxiv.org/pdf/2602.00042v1
+
+## 20861. Who Does This Name Remind You of ? Nationality Prediction via Large Language Model Associative Memory
+
+- **ID:** `2601.12771` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Keito Inoshita
+- **abs:** https://arxiv.org/abs/2601.12771 · **PDF:** https://arxiv.org/pdf/2601.12771v2
+
+## 20862. Analyzing Diffusion and Autoregressive Vision Language Models in Multimodal Embedding Space
+
+- **ID:** `2602.06056` · **primary:** `cs.MM` · **Submitted:** 2026-01-19
+- **Authors:** Zihang Wang, Siyue Zhang, Yilun Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.06056 · **PDF:** https://arxiv.org/pdf/2602.06056v1
+
+## 20863. Teaching LLMs to Learn Tool Trialing and Execution through Environment Interaction
+
+- **ID:** `2601.12762` · **primary:** `cs.SE` · **Submitted:** 2026-01-19
+- **Authors:** Xingjie Gao, Pengcheng Huang, Zhenghao Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.12762 · **PDF:** https://arxiv.org/pdf/2601.12762v1
+
+## 20864. Disentangling Direction and Magnitude in Transformer Representations: A Double Dissociation Through L2-Matched Perturbation Analysis
+
+- **ID:** `2602.11169` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Mangadoddi Srikar Vardhan, Lekkala Sai Teja
+- **abs:** https://arxiv.org/abs/2602.11169 · **PDF:** https://arxiv.org/pdf/2602.11169v1
+
+## 20865. VISPA: Pluralistic Alignment via Automatic Value Selection and Activation
+
+- **ID:** `2601.12758` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Shenyan Zheng, Jiayou Zhong, Anudeex Shetty, et al.
+- **abs:** https://arxiv.org/abs/2601.12758 · **PDF:** https://arxiv.org/pdf/2601.12758v1
+
+## 20866. Student Perceptions of Large Language Models Use in Self-Reflection and Design Critique in Architecture Studio
+
+- **ID:** `2602.00041` · **primary:** `cs.CY` · **Submitted:** 2026-01-19
+- **Authors:** Juan David Salazar Rodriguez, Sam Conrad Joyce, Nachamma Sockalingam, et al.
+- **abs:** https://arxiv.org/abs/2602.00041 · **PDF:** https://arxiv.org/pdf/2602.00041v2
+
+## 20867. Enhancing few-shot time series forecasting with LLM-guided diffusion
+
+- **ID:** `2602.00040` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Haonan Shi, Dehua Shuai, Liming Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.00040 · **PDF:** https://arxiv.org/pdf/2602.00040v1
+
+## 20868. PAIR-SAFE: A Paired-Agent Approach for Runtime Auditing and Refining AI-Mediated Mental Health Support
+
+- **ID:** `2601.12754` · **primary:** `cs.HC` · **Submitted:** 2026-01-19
+- **Authors:** Jiwon Kim, Violeta J. Rodriguez, Dong Whi Yoo, et al.
+- **abs:** https://arxiv.org/abs/2601.12754 · **PDF:** https://arxiv.org/pdf/2601.12754v1
+
+## 20869. SoundPlot: An Open-Source Framework for Birdsong Acoustic Analysis and Neural Synthesis with Interactive 3D Visualization
+
+- **ID:** `2601.12752` · **primary:** `cs.SD` · **Submitted:** 2026-01-19
+- **Authors:** Naqcho Ali Mehdi, Mohammad Adeel, Aizaz Ali Larik
+- **abs:** https://arxiv.org/abs/2601.12752 · **PDF:** https://arxiv.org/pdf/2601.12752v1
+
+## 20870. A Boolean Function-Theoretic Framework for Expressivity in GNNs with Applications to Fair Graph Mining
+
+- **ID:** `2601.12751` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Manjish Pal
+- **abs:** https://arxiv.org/abs/2601.12751 · **PDF:** https://arxiv.org/pdf/2601.12751v1
+
+## 20871. Towards Robust Process Reward Modeling via Noise-aware Learning
+
+- **ID:** `2601.12748` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Bin Xie, Bingbing Xu, Xueyun Tian, et al.
+- **abs:** https://arxiv.org/abs/2601.12748 · **PDF:** https://arxiv.org/pdf/2601.12748v1
+
+## 20872. A Graph Prompt Fine-Tuning Method for WSN Spatio-Temporal Correlation Anomaly Detection
+
+- **ID:** `2601.12745` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Miao Ye, Jing Cui, Yuan huang, et al.
+- **abs:** https://arxiv.org/abs/2601.12745 · **PDF:** https://arxiv.org/pdf/2601.12745v1
+
+## 20873. Vision Language Models for Optimization-Driven Intent Processing in Autonomous Networks
+
+- **ID:** `2601.12744` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Tasnim Ahmed, Yifan Zhu, Salimur Choudhury
+- **abs:** https://arxiv.org/abs/2601.12744 · **PDF:** https://arxiv.org/pdf/2601.12744v1
+
+## 20874. AirHunt: Bridging VLM Semantics and Continuous Planning for Efficient Aerial Object Navigation
+
+- **ID:** `2601.12742` · **primary:** `cs.RO` · **Submitted:** 2026-01-19
+- **Authors:** Xuecheng Chen, Zongzhuo Liu, Jianfa Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.12742 · **PDF:** https://arxiv.org/pdf/2601.12742v1
+
+## 20875. TreeWriter: AI-Assisted Hierarchical Planning and Writing for Long-Form Documents
+
+- **ID:** `2601.12740` · **primary:** `cs.HC` · **Submitted:** 2026-01-19
+- **Authors:** Zijian Zhang, Fangshi Du, Xingjian Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.12740 · **PDF:** https://arxiv.org/pdf/2601.12740v1
+
+## 20876. A Shared Geometry of Difficulty in Multilingual Language Models
+
+- **ID:** `2601.12731` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Stefano Civelli, Pietro Bernardelle, Nicolò Brunello, et al.
+- **abs:** https://arxiv.org/abs/2601.12731 · **PDF:** https://arxiv.org/pdf/2601.12731v1
+
+## 20877. Distribution-Centric Policy Optimization Dominates Exploration-Exploitation Trade-off
+
+- **ID:** `2601.12730` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Zhaochun Li, Chen Wang, Jionghao Bai, et al.
+- **abs:** https://arxiv.org/abs/2601.12730 · **PDF:** https://arxiv.org/pdf/2601.12730v1
+
+## 20878. AI-exhibited Personality Traits Can Shape Human Self-concept through Conversations
+
+- **ID:** `2601.12727` · **primary:** `cs.HC` · **Submitted:** 2026-01-19
+- **Authors:** Jingshu Li, Tianqi Song, Nattapat Boonprakong, et al.
+- **abs:** https://arxiv.org/abs/2601.12727 · **PDF:** https://arxiv.org/pdf/2601.12727v1
+
+## 20879. An Evolutionary Framework for Automatic Optimization Benchmark Generation via Large Language Models
+
+- **ID:** `2601.12723` · **primary:** `cs.NE` · **Submitted:** 2026-01-19
+- **Authors:** Yuhiro Ono, Tomohiro Harada, Yukiya Miura
+- **abs:** https://arxiv.org/abs/2601.12723 · **PDF:** https://arxiv.org/pdf/2601.12723v2
+
+## 20880. Teaching Large Reasoning Models Effective Reflection
+
+- **ID:** `2601.12720` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Hanbin Wang, Jingwei Song, Jinpeng Li, et al.
+- **abs:** https://arxiv.org/abs/2601.12720 · **PDF:** https://arxiv.org/pdf/2601.12720v1
+
+## 20881. RSOD: Reliability-Guided Sonar Image Object Detection with Extremely Limited Labels
+
+- **ID:** `2601.12715` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Chengzhou Li, Ping Guo, Guanchen Meng, et al.
+- **abs:** https://arxiv.org/abs/2601.12715 · **PDF:** https://arxiv.org/pdf/2601.12715v1
+
+## 20882. P2L-CA: An Effective Parameter Tuning Framework for Rehearsal-Free Multi-Label Class-Incremental Learning
+
+- **ID:** `2601.12714` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Songlin Dong, Jiangyang Li, Chenhao Ding, et al.
+- **abs:** https://arxiv.org/abs/2601.12714 · **PDF:** https://arxiv.org/pdf/2601.12714v1
+
+## 20883. Neurosymbolic LoRA: Why and When to Tune Weights vs. Rewrite Prompts
+
+- **ID:** `2601.12711` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Kevin Wang, Neel P. Bhatt, Cong Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.12711 · **PDF:** https://arxiv.org/pdf/2601.12711v1
+
+## 20884. Decoding Rewards in Competitive Games: Inverse Game Theory with Entropy Regularization
+
+- **ID:** `2601.12707` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Junyi Liao, Zihan Zhu, Ethan Fang, et al.
+- **abs:** https://arxiv.org/abs/2601.12707 · **PDF:** https://arxiv.org/pdf/2601.12707v1
+
+## 20885. Trend-Adjusted Time Series Models with an Application to Gold Price Forecasting
+
+- **ID:** `2601.12706` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Sina Kazemdehbashi
+- **abs:** https://arxiv.org/abs/2601.12706 · **PDF:** https://arxiv.org/pdf/2601.12706v2
+
+## 20886. Adaptively trained Physics-informed Radial Basis Function Neural Networks for Solving Multi-asset Option Pricing Problems
+
+- **ID:** `2601.12704` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Yan Ma, Yumeng Ren
+- **abs:** https://arxiv.org/abs/2601.12704 · **PDF:** https://arxiv.org/pdf/2601.12704v1
+
+## 20887. Towards Spectroscopy: Susceptibility Clusters in Language Models
+
+- **ID:** `2601.12703` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Andrew Gordon, Garrett Baker, George Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.12703 · **PDF:** https://arxiv.org/pdf/2601.12703v1
+
+## 20888. LSSF: Safety Alignment for Large Language Models through Low-Rank Safety Subspace Fusion
+
+- **ID:** `2602.00038` · **primary:** `cs.CY` · **Submitted:** 2026-01-19
+- **Authors:** Guanghao Zhou, Panjia Qiu, Cen Chen, et al.
+- **abs:** https://arxiv.org/abs/2602.00038 · **PDF:** https://arxiv.org/pdf/2602.00038v1
+
+## 20889. Resource-Conscious RL Algorithms for Deep Brain Stimulation
+
+- **ID:** `2601.12699` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Arkaprava Gupta, Nicholas Carter, William Zellers, et al.
+- **abs:** https://arxiv.org/abs/2601.12699 · **PDF:** https://arxiv.org/pdf/2601.12699v1
+
+## 20890. A Two-Stage GPU Kernel Tuner Combining Semantic Refactoring and Search-Based Optimization
+
+- **ID:** `2601.12698` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Qiuyi Qu, Yicheng Sui, Yufei Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.12698 · **PDF:** https://arxiv.org/pdf/2601.12698v3
+
+## 20891. UbuntuGuard: A Culturally-Grounded Policy Benchmark for Equitable AI Safety in African Languages
+
+- **ID:** `2601.12696` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Tassallah Abdullahi, Macton Mgonzo, Mardiyyah Oduwole, et al.
+- **abs:** https://arxiv.org/abs/2601.12696 · **PDF:** https://arxiv.org/pdf/2601.12696v1
+
+## 20892. BlocksecRT-DETR: Decentralized Privacy-Preserving and Token-Efficient Federated Transformer Learning for Secure Real-Time Object Detection in ITS
+
+- **ID:** `2601.12693` · **primary:** `cs.CR` · **Submitted:** 2026-01-19
+- **Authors:** Mohoshin Ara Tahera, Sabbir Rahman, Shuvalaxmi Dass, et al.
+- **abs:** https://arxiv.org/abs/2601.12693 · **PDF:** https://arxiv.org/pdf/2601.12693v1
+
+## 20893. Logic-Guided Multistage Inference for Explainable Multidefendant Judgment Prediction
+
+- **ID:** `2601.12688` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Xu Zhang, Qinghua Wang, Mengyang Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.12688 · **PDF:** https://arxiv.org/pdf/2601.12688v1
+
+## 20894. Enhancing SDG-Text Classification with Combinatorial Fusion Analysis and Generative AI
+
+- **ID:** `2602.11168` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Jingyan Xu, Marcelo L. LaFleur, Christina Schweikert, et al.
+- **abs:** https://arxiv.org/abs/2602.11168 · **PDF:** https://arxiv.org/pdf/2602.11168v1
+
+## 20895. MetaToolAgent: Towards Generalizable Tool Usage in LLMs through Meta-Learning
+
+- **ID:** `2601.12680` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Zheng Fang, Wolfgang Mayer, Zeyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.12680 · **PDF:** https://arxiv.org/pdf/2601.12680v1
+
+## 20896. Bitcoin Price Prediction using Machine Learning and Combinatorial Fusion Analysis
+
+- **ID:** `2602.00037` · **primary:** `q-fin.ST` · **Submitted:** 2026-01-19
+- **Authors:** Yuanhong Wu, Wei Ye, Jingyan Xu, et al.
+- **abs:** https://arxiv.org/abs/2602.00037 · **PDF:** https://arxiv.org/pdf/2602.00037v2
+
+## 20897. Exploiting Test-Time Augmentation in Federated Learning for Brain Tumor MRI Classification
+
+- **ID:** `2601.12671` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Thamara Leandra de Deus Melo, Rodrigo Moreira, Larissa Ferreira Rodrigues Moreira, et al.
+- **abs:** https://arxiv.org/abs/2601.12671 · **PDF:** https://arxiv.org/pdf/2601.12671v1
+
+## 20898. Empowering All-in-Loop Health Management of Spacecraft Power System in the Mega-Constellation Era via Human-AI Collaboration
+
+- **ID:** `2601.12667` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Yi Di, Zhibin Zhao, Fujin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.12667 · **PDF:** https://arxiv.org/pdf/2601.12667v2
+
+## 20899. Generalizable Hyperparameter Optimization for Federated Learning on Non-IID Cancer Images
+
+- **ID:** `2601.12664` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Elisa Gonçalves Ribeiro, Rodrigo Moreira, Larissa Ferreira Rodrigues Moreira, et al.
+- **abs:** https://arxiv.org/abs/2601.12664 · **PDF:** https://arxiv.org/pdf/2601.12664v1
+
+## 20900. Energy-Efficient Prediction in Textile Manufacturing: Enhancing Accuracy and Data Efficiency With Ensemble Deep Transfer Learning
+
+- **ID:** `2601.12663` · **primary:** `eess.SP` · **Submitted:** 2026-01-19
+- **Authors:** Yan-Chen Chen, Wei-Yu Chiu, Qun-Yu Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.12663 · **PDF:** https://arxiv.org/pdf/2601.12663v1
+
+## 20901. Decentralized Learning Strategies for Estimation Error Minimization with Graph Neural Networks
+
+- **ID:** `2601.12662` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Xingran Chen, Navid NaderiAlizadeh, Alejandro Ribeiro, et al.
+- **abs:** https://arxiv.org/abs/2601.12662 · **PDF:** https://arxiv.org/pdf/2601.12662v2
+
+## 20902. MedConsultBench: A Full-Cycle, Fine-Grained, Process-Aware Benchmark for Medical Consultation Agents
+
+- **ID:** `2601.12661` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Chuhan Qiao, Jianghua Huang, Daxing Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.12661 · **PDF:** https://arxiv.org/pdf/2601.12661v1
+
+## 20903. Toward Faithful Explanations in Acoustic Anomaly Detection
+
+- **ID:** `2601.12660` · **primary:** `cs.SD` · **Submitted:** 2026-01-19
+- **Authors:** Maab Elrashid, Anthony Deschênes, Cem Subakan, et al.
+- **abs:** https://arxiv.org/abs/2601.12660 · **PDF:** https://arxiv.org/pdf/2601.12660v1
+
+## 20904. Augmenting Question Answering with A Hybrid RAG Approach
+
+- **ID:** `2601.12658` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Tianyi Yang, Nashrah Haque, Vaishnave Jonnalagadda, et al.
+- **abs:** https://arxiv.org/abs/2601.12658 · **PDF:** https://arxiv.org/pdf/2601.12658v2
+
+## 20905. Explanation Multiplicity in SHAP: Characterization and Assessment
+
+- **ID:** `2601.12654` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Hyunseung Hwang, Seungeun Lee, Lucas Rosenblatt, et al.
+- **abs:** https://arxiv.org/abs/2601.12654 · **PDF:** https://arxiv.org/pdf/2601.12654v2
+
+## 20906. Intelligent Documentation in Medical Education: Can AI Replace Manual Case Logging?
+
+- **ID:** `2601.12648` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Nafiz Imtiaz Khan, Kylie Cleland, Vladimir Filkov, et al.
+- **abs:** https://arxiv.org/abs/2601.12648 · **PDF:** https://arxiv.org/pdf/2601.12648v1
+
+## 20907. Unbounded Harms, Bounded Law: Liability in the Age of Borderless AI
+
+- **ID:** `2601.12646` · **primary:** `cs.CY` · **Submitted:** 2026-01-19
+- **Authors:** Ha-Chi Tran
+- **abs:** https://arxiv.org/abs/2601.12646 · **PDF:** https://arxiv.org/pdf/2601.12646v1
+
+## 20908. STEP-LLM: Generating CAD STEP Models from Natural Language with Large Language Models
+
+- **ID:** `2601.12641` · **primary:** `cs.AI` · **Submitted:** 2026-01-19
+- **Authors:** Xiangyu Shi, Junyang Ding, Xu Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.12641 · **PDF:** https://arxiv.org/pdf/2601.12641v1
+
+## 20909. Objective Matters: Fine-Tuning Objectives Shape Safety, Robustness, and Persona Drift
+
+- **ID:** `2601.12639` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Daniel Vennemeyer, Punya Syon Pandey, Phan Anh Duong, et al.
+- **abs:** https://arxiv.org/abs/2601.12639 · **PDF:** https://arxiv.org/pdf/2601.12639v1
+
+## 20910. Mixed Precision PointPillars for Efficient 3D Object Detection with TensorRT
+
+- **ID:** `2601.12638` · **primary:** `cs.CV` · **Submitted:** 2026-01-19
+- **Authors:** Ninnart Fuengfusin, Keisuke Yoneda, Naoki Suganuma
+- **abs:** https://arxiv.org/abs/2601.12638 · **PDF:** https://arxiv.org/pdf/2601.12638v2
+
+## 20911. Topology-Aware Multiscale Mixture of Experts for Efficient Molecular Property Prediction
+
+- **ID:** `2601.12637` · **primary:** `cs.LG` · **Submitted:** 2026-01-19
+- **Authors:** Long D. Nguyen, Kelin Xia, Binh P. Nguyen
+- **abs:** https://arxiv.org/abs/2601.12637 · **PDF:** https://arxiv.org/pdf/2601.12637v1
+
+## 20912. New Trends in the Stability of Sinkhorn Semigroups
+
+- **ID:** `2601.12633` · **primary:** `math.PR` · **Submitted:** 2026-01-19
+- **Authors:** Pierre Del Moral, Ajay Jasra
+- **abs:** https://arxiv.org/abs/2601.12633 · **PDF:** https://arxiv.org/pdf/2601.12633v1
+
+## 20913. BioPulse-QA: A Dynamic Biomedical Question-Answering Benchmark for Evaluating Factuality, Robustness, and Bias in Large Language Models
+
+- **ID:** `2601.12632` · **primary:** `cs.CL` · **Submitted:** 2026-01-19
+- **Authors:** Kriti Bhattarai, Vipina K. Keloth, Donald Wright, et al.
+- **abs:** https://arxiv.org/abs/2601.12632 · **PDF:** https://arxiv.org/pdf/2601.12632v1
+
+## 20914. Distributed Causality in the SDG Network: Evidence from Panel VAR and Conditional Independence Analysis
+
+- **ID:** `2601.20875` · **primary:** `stat.AP` · **Submitted:** 2026-01-19
+- **Authors:** Md Muhtasim Munif Fahim, Md Jahid Hasan Imran, Luknath Debnath, et al.
+- **abs:** https://arxiv.org/abs/2601.20875 · **PDF:** https://arxiv.org/pdf/2601.20875v1
+
+## 20915. Enhanced Climbing Image Nudged Elastic Band method with Hessian Eigenmode Alignment
+
+- **ID:** `2601.12630` · **primary:** `physics.chem-ph` · **Submitted:** 2026-01-19
+- **Authors:** Rohit Goswami, Miha Gunde, Hannes Jónsson
+- **abs:** https://arxiv.org/abs/2601.12630 · **PDF:** https://arxiv.org/pdf/2601.12630v3
+
+## 20916. Towards Robust Universal Perturbation Attacks: A Float-Coded, Penalty-Driven Evolutionary Approach
+
+- **ID:** `2601.12624` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Shiqi Wang, Mahdi Khosravy, Neeraj Gupta, et al.
+- **abs:** https://arxiv.org/abs/2601.12624 · **PDF:** https://arxiv.org/pdf/2601.12624v1
+
+## 20917. Learning Deterministic Finite-State Machines from the Prefixes of a Single String is NP-Complete
+
+- **ID:** `2601.12621` · **primary:** `cs.FL` · **Submitted:** 2026-01-18
+- **Authors:** Radu Cosmin Dumitru, Ryo Yoshinaka, Ayumi Shinohara
+- **abs:** https://arxiv.org/abs/2601.12621 · **PDF:** https://arxiv.org/pdf/2601.12621v1
+
+## 20918. Disagreement as Data: Reasoning Trace Analytics in Multi-Agent Systems
+
+- **ID:** `2601.12618` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Elham Tajik, Conrad Borchers, Bahar Shahrokhian, et al.
+- **abs:** https://arxiv.org/abs/2601.12618 · **PDF:** https://arxiv.org/pdf/2601.12618v1
+
+## 20919. Creating Disability Story Videos with Generative AI: Motivation, Expression, and Sharing
+
+- **ID:** `2601.12617` · **primary:** `cs.HC` · **Submitted:** 2026-01-18
+- **Authors:** Shuo Niu, Dylan Clements, Hyungsin Kim
+- **abs:** https://arxiv.org/abs/2601.12617 · **PDF:** https://arxiv.org/pdf/2601.12617v1
+
+## 20920. Deterministic and probabilistic neural surrogates of global hybrid-Vlasov simulations
+
+- **ID:** `2601.12614` · **primary:** `physics.space-ph` · **Submitted:** 2026-01-18
+- **Authors:** Daniel Holmberg, Ivan Zaitsev, Markku Alho, et al.
+- **abs:** https://arxiv.org/abs/2601.12614 · **PDF:** https://arxiv.org/pdf/2601.12614v2
+
+## 20921. What Trace Powers Reveal About Log-Determinants: Closed-Form Estimators, Certificates, and Failure Modes
+
+- **ID:** `2601.12612` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Piyush Sao
+- **abs:** https://arxiv.org/abs/2601.12612 · **PDF:** https://arxiv.org/pdf/2601.12612v1
+
+## 20922. HERMES: A Unified Open-Source Framework for Realtime Multimodal Physiological Sensing, Edge AI, and Intervention in Closed-Loop Smart Healthcare Applications
+
+- **ID:** `2601.12610` · **primary:** `eess.SY` · **Submitted:** 2026-01-18
+- **Authors:** Maxim Yudayev, Juha Carlon, Diwas Lamsal, et al.
+- **abs:** https://arxiv.org/abs/2601.12610 · **PDF:** https://arxiv.org/pdf/2601.12610v1
+
+## 20923. Visualizing and Benchmarking LLM Factual Hallucination Tendencies via Internal State Analysis and Clustering
+
+- **ID:** `2602.11167` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Nathan Mao, Varun Kaushik, Shreya Shivkumar, et al.
+- **abs:** https://arxiv.org/abs/2602.11167 · **PDF:** https://arxiv.org/pdf/2602.11167v1
+
+## 20924. A Cloud-based Multi-Agentic Workflow for Science
+
+- **ID:** `2601.12607` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Anurag Acharya, Timothy Vega, Rizwan A. Ashraf, et al.
+- **abs:** https://arxiv.org/abs/2601.12607 · **PDF:** https://arxiv.org/pdf/2601.12607v1
+
+## 20925. Beyond Softmax and Entropy: Convergence Rates of Policy Gradients with f-SoftArgmax Parameterization & Coupled Regularization
+
+- **ID:** `2601.12604` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Safwan Labbi, Daniil Tiapkin, Paul Mangold, et al.
+- **abs:** https://arxiv.org/abs/2601.12604 · **PDF:** https://arxiv.org/pdf/2601.12604v2
+
+## 20926. onepot CORE -- an enumerated chemical space to streamline drug discovery, enabled by automated small molecule synthesis and AI
+
+- **ID:** `2601.12603` · **primary:** `physics.chem-ph` · **Submitted:** 2026-01-18
+- **Authors:** Andrei S. Tyrin, Brandon Wang, Manuel Muñoz, et al.
+- **abs:** https://arxiv.org/abs/2601.12603 · **PDF:** https://arxiv.org/pdf/2601.12603v1
+
+## 20927. SSVD-O: Parameter-Efficient Fine-Tuning with Structured SVD for Speech Recognition
+
+- **ID:** `2601.12600` · **primary:** `cs.SD` · **Submitted:** 2026-01-18
+- **Authors:** Pu Wang, Shinji Watanabe, Hugo Van hamme
+- **abs:** https://arxiv.org/abs/2601.12600 · **PDF:** https://arxiv.org/pdf/2601.12600v1
+
+## 20928. Dissecting Linear Recurrent Models: How Different Gating Strategies Drive Selectivity and Generalization
+
+- **ID:** `2601.12598` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Younes Bouhadjar, Maxime Fabre, Felix Schmidt, et al.
+- **abs:** https://arxiv.org/abs/2601.12598 · **PDF:** https://arxiv.org/pdf/2601.12598v1
+
+## 20929. SLAP: Scalable Language-Audio Pretraining with Variable-Duration Audio and Multi-Objective Training
+
+- **ID:** `2601.12594` · **primary:** `eess.AS` · **Submitted:** 2026-01-18
+- **Authors:** Xinhao Mei, Gael Le Lan, Haohe Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.12594 · **PDF:** https://arxiv.org/pdf/2601.12594v1
+
+## 20930. A Theory of Diversity for Random Matrices with Applications to In-Context Learning of Schrödinger Equations
+
+- **ID:** `2601.12587` · **primary:** `stat.ML` · **Submitted:** 2026-01-18
+- **Authors:** Frank Cole, Yulong Lu, Shaurya Sehgal
+- **abs:** https://arxiv.org/abs/2601.12587 · **PDF:** https://arxiv.org/pdf/2601.12587v1
+
+## 20931. Do MLLMs See What We See? Analyzing Visualization Literacy Barriers in AI Systems
+
+- **ID:** `2601.12585` · **primary:** `cs.HC` · **Submitted:** 2026-01-18
+- **Authors:** Mengli, Duan, Yuhe, et al.
+- **abs:** https://arxiv.org/abs/2601.12585 · **PDF:** https://arxiv.org/pdf/2601.12585v1
+
+## 20932. Ontology-aligned structuring and reuse of multimodal materials data and workflows towards automatic reproduction
+
+- **ID:** `2601.12582` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-18
+- **Authors:** Sepideh Baghaee Ravari, Abril Azocar Guzman, Sarath Menon, et al.
+- **abs:** https://arxiv.org/abs/2601.12582 · **PDF:** https://arxiv.org/pdf/2601.12582v1
+
+## 20933. Primate-like perceptual decision making emerges through deep recurrent reinforcement learning
+
+- **ID:** `2601.12577` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-18
+- **Authors:** Nathan J. Wispinski, Scott A. Stone, Anthony Singhal, et al.
+- **abs:** https://arxiv.org/abs/2601.12577 · **PDF:** https://arxiv.org/pdf/2601.12577v1
+
+## 20934. Prometheus Mind: Retrofitting Memory to Frozen Language Models
+
+- **ID:** `2601.15324` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Mark Wind
+- **abs:** https://arxiv.org/abs/2601.15324 · **PDF:** https://arxiv.org/pdf/2601.15324v2
+
+## 20935. Analyzing the Temporal Factors for Anxiety and Depression Symptoms with the Rashomon Perspective
+
+- **ID:** `2601.20874` · **primary:** `stat.AP` · **Submitted:** 2026-01-18
+- **Authors:** Mustafa Cavus, Przemysław Biecek, Julian Tejada, et al.
+- **abs:** https://arxiv.org/abs/2601.20874 · **PDF:** https://arxiv.org/pdf/2601.20874v1
+
+## 20936. LOGOS-CA: A Cellular Automaton Using Natural Language as State and Rule
+
+- **ID:** `2602.00036` · **primary:** `nlin.CG` · **Submitted:** 2026-01-18
+- **Authors:** Keishu Utimula
+- **abs:** https://arxiv.org/abs/2602.00036 · **PDF:** https://arxiv.org/pdf/2602.00036v2
+
+## 20937. Agentic Artificial Intelligence (AI): Architectures, Taxonomies, and Evaluation of Large Language Model Agents
+
+- **ID:** `2601.12560` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Arunkumar V, Gangadharan G. R., Rajkumar Buyya
+- **abs:** https://arxiv.org/abs/2601.12560 · **PDF:** https://arxiv.org/pdf/2601.12560v1
+
+## 20938. Life, Machine Learning, and the Search for Habitability: Predicting Biosignature Fluxes for the Habitable Worlds Observatory
+
+- **ID:** `2601.12557` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Mark Moussa, Amber V. Young, Brianna Isola, et al.
+- **abs:** https://arxiv.org/abs/2601.12557 · **PDF:** https://arxiv.org/pdf/2601.12557v1
+
+## 20939. Evaluating Contextually Mediated Factual Recall in Multilingual Large Language Models
+
+- **ID:** `2601.12555` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Yihong Liu, Bingyu Xiong, Hinrich Schütze
+- **abs:** https://arxiv.org/abs/2601.12555 · **PDF:** https://arxiv.org/pdf/2601.12555v1
+
+## 20940. Artificial Intelligence in Materials Science and Engineering: Current Landscape, Key Challenges, and Future Trajectorie
+
+- **ID:** `2601.12554` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-18
+- **Authors:** Iman Peivaste, Salim Belouettar, Francesco Mercuri, et al.
+- **abs:** https://arxiv.org/abs/2601.12554 · **PDF:** https://arxiv.org/pdf/2601.12554v1
+
+## 20941. Benchmarking Concept-Spilling Across Languages in LLMs
+
+- **ID:** `2601.12549` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Ilia Badanin, Daniil Dzenhaliou, Imanol Schlag
+- **abs:** https://arxiv.org/abs/2601.12549 · **PDF:** https://arxiv.org/pdf/2601.12549v1
+
+## 20942. How Clinicians Think and What AI Can Learn From It
+
+- **ID:** `2601.12547` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Dipayan Sengupta, Saumya Panda
+- **abs:** https://arxiv.org/abs/2601.12547 · **PDF:** https://arxiv.org/pdf/2601.12547v1
+
+## 20943. Press Start to Charge: Videogaming the Online Centralized Charging Scheduling Problem
+
+- **ID:** `2601.12543` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Alireza Ghahtarani, Martin Cousineau, Amir-massoud Farahmand, et al.
+- **abs:** https://arxiv.org/abs/2601.12543 · **PDF:** https://arxiv.org/pdf/2601.12543v1
+
+## 20944. Rethinking the AI Scientist: Interactive Multi-Agent Workflows for Scientific Discovery
+
+- **ID:** `2601.12542` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Lukas Weidener, Marko Brkić, Mihailo Jovanović, et al.
+- **abs:** https://arxiv.org/abs/2601.12542 · **PDF:** https://arxiv.org/pdf/2601.12542v2
+
+## 20945. MemeLens: Multilingual Multitask VLMs for Memes
+
+- **ID:** `2601.12539` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Ali Ezzat Shahroor, Mohamed Bayan Kmainasi, Abul Hasnat, et al.
+- **abs:** https://arxiv.org/abs/2601.12539 · **PDF:** https://arxiv.org/pdf/2601.12539v2
+
+## 20946. Agentic Reasoning for Large Language Models
+
+- **ID:** `2601.12538` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Tianxin Wei, Ting-Wei Li, Zhining Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.12538 · **PDF:** https://arxiv.org/pdf/2601.12538v1
+
+## 20947. Improving Low-Resource Machine Translation via Round-Trip Reinforcement Learning
+
+- **ID:** `2601.12535` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Ahmed Attia, Alham Fikri Aji
+- **abs:** https://arxiv.org/abs/2601.12535 · **PDF:** https://arxiv.org/pdf/2601.12535v3
+
+## 20948. Asynchronous MultiAgent Reinforcement Learning for 5G Routing under Side Constraints
+
+- **ID:** `2602.00035` · **primary:** `cs.NI` · **Submitted:** 2026-01-18
+- **Authors:** Sebastian Racedo, Brigitte Jaumard, Oscar Delgado, et al.
+- **abs:** https://arxiv.org/abs/2602.00035 · **PDF:** https://arxiv.org/pdf/2602.00035v1
+
+## 20949. Encoding Emotion Through Self-Supervised Eye Movement Reconstruction
+
+- **ID:** `2601.12534` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Marcus Ma, Jordan Prescott, Emily Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.12534 · **PDF:** https://arxiv.org/pdf/2601.12534v2
+
+## 20950. Approximating splits for decision trees quickly in sparse data streams
+
+- **ID:** `2601.12525` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Nikolaj Tatti
+- **abs:** https://arxiv.org/abs/2601.12525 · **PDF:** https://arxiv.org/pdf/2601.12525v1
+
+## 20951. Synthetic Student Responses: LLM-Extracted Features for IRT Difficulty Parameter Estimation
+
+- **ID:** `2602.00034` · **primary:** `cs.CY` · **Submitted:** 2026-01-18
+- **Authors:** Matias Hoyl
+- **abs:** https://arxiv.org/abs/2602.00034 · **PDF:** https://arxiv.org/pdf/2602.00034v1
+
+## 20952. Improved Bug Localization with AI Agents Leveraging Hypothesis and Dynamic Cognition
+
+- **ID:** `2601.12522` · **primary:** `cs.SE` · **Submitted:** 2026-01-18
+- **Authors:** Asif Mohammed Samir, Mohammad Masudur Rahman
+- **abs:** https://arxiv.org/abs/2601.12522 · **PDF:** https://arxiv.org/pdf/2601.12522v2
+
+## 20953. Learning Relativistic Geodesics and Chaotic Dynamics via Stabilized Lagrangian Neural Networks
+
+- **ID:** `2601.12519` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Abdullah Umut Hamzaogullari, Arkadas Ozakin
+- **abs:** https://arxiv.org/abs/2601.12519 · **PDF:** https://arxiv.org/pdf/2601.12519v1
+
+## 20954. Cooperative Multi-agent RL with Communication Constraints
+
+- **ID:** `2601.12518` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Nuoya Xiong, Aarti Singh
+- **abs:** https://arxiv.org/abs/2601.12518 · **PDF:** https://arxiv.org/pdf/2601.12518v1
+
+## 20955. IFNSO: Iteration-Free Newton-Schulz Orthogonalization
+
+- **ID:** `2602.02500` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Chen Hu, Qianxi Zhao, Xiaochen Yuan, et al.
+- **abs:** https://arxiv.org/abs/2602.02500 · **PDF:** https://arxiv.org/pdf/2602.02500v3
+
+## 20956. SDCoNet: Saliency-Driven Multi-Task Collaborative Network for Remote Sensing Object Detection
+
+- **ID:** `2601.12507` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Ruo Qi, Linhui Dai, Yusong Qin, et al.
+- **abs:** https://arxiv.org/abs/2601.12507 · **PDF:** https://arxiv.org/pdf/2601.12507v1
+
+## 20957. DoPE: Decoy Oriented Perturbation Encapsulation Human-Readable, AI-Hostile Documents for Academic Integrity
+
+- **ID:** `2601.12505` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Ashish Raj Shekhar, Shiven Agarwal, Priyanuj Bordoloi, et al.
+- **abs:** https://arxiv.org/abs/2601.12505 · **PDF:** https://arxiv.org/pdf/2601.12505v1
+
+## 20958. Semidefinite Programming for Quantum Channel Learning
+
+- **ID:** `2601.12502` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Mikhail Gennadievich Belov, Victor Victorovich Dubov, Vadim Konstantinovich Ivanov, et al.
+- **abs:** https://arxiv.org/abs/2601.12502 · **PDF:** https://arxiv.org/pdf/2601.12502v1
+
+## 20959. Failure Modes in Multi-Hop QA: The Weakest Link Law and the Recognition Bottleneck
+
+- **ID:** `2601.12499` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Meiru Zhang, Zaiqiao Meng, Nigel Collier
+- **abs:** https://arxiv.org/abs/2601.12499 · **PDF:** https://arxiv.org/pdf/2601.12499v1
+
+## 20960. Multi-Task Instruction Tuning via Data Scheduling for Low-Resource Arabic AudioLLMs
+
+- **ID:** `2601.12494` · **primary:** `cs.SD` · **Submitted:** 2026-01-18
+- **Authors:** Hunzalah Hassan Bhatti, Firoj Alam, Shammur Absar Chowdhury
+- **abs:** https://arxiv.org/abs/2601.12494 · **PDF:** https://arxiv.org/pdf/2601.12494v2
+
+## 20961. A Mixture of Experts Vision Transformer for High-Fidelity Surface Code Decoding
+
+- **ID:** `2601.12483` · **primary:** `quant-ph` · **Submitted:** 2026-01-18
+- **Authors:** Hoang Viet Nguyen, Manh Hung Nguyen, Hoang Ta, et al.
+- **abs:** https://arxiv.org/abs/2601.12483 · **PDF:** https://arxiv.org/pdf/2601.12483v1
+
+## 20962. Capability-Aware Early-Stage Research Idea Evaluation
+
+- **ID:** `2601.12473` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Renlong Jie, Chen Chu, Zhen Wang
+- **abs:** https://arxiv.org/abs/2601.12473 · **PDF:** https://arxiv.org/pdf/2601.12473v1
+
+## 20963. Knowing When to Abstain: Medical LLMs Under Clinical Uncertainty
+
+- **ID:** `2601.12471` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Sravanthi Machcha, Sushrita Yerra, Sahil Gupta, et al.
+- **abs:** https://arxiv.org/abs/2601.12471 · **PDF:** https://arxiv.org/pdf/2601.12471v2
+
+## 20964. Patch-Level Tokenization with CNN Encoders and Attention for Improved Transformer Time-Series Forecasting
+
+- **ID:** `2601.12467` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Saurish Nagrath, Saroj Kumar Panigrahy
+- **abs:** https://arxiv.org/abs/2601.12467 · **PDF:** https://arxiv.org/pdf/2601.12467v3
+
+## 20965. Incentivizing In-depth Reasoning over Long Contexts with Process Advantage Shaping
+
+- **ID:** `2601.12465` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Miao Peng, Weizhou Shen, Nuo Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.12465 · **PDF:** https://arxiv.org/pdf/2601.12465v1
+
+## 20966. TrojanPraise: Jailbreak LLMs via Benign Fine-Tuning
+
+- **ID:** `2601.12460` · **primary:** `cs.CR` · **Submitted:** 2026-01-18
+- **Authors:** Zhixin Xie, Xurui Song, Jun Luo
+- **abs:** https://arxiv.org/abs/2601.12460 · **PDF:** https://arxiv.org/pdf/2601.12460v1
+
+## 20967. AgenTRIM: Tool Risk Mitigation for Agentic AI
+
+- **ID:** `2601.12449` · **primary:** `cs.CR` · **Submitted:** 2026-01-18
+- **Authors:** Roy Betser, Shamik Bose, Amit Giloni, et al.
+- **abs:** https://arxiv.org/abs/2601.12449 · **PDF:** https://arxiv.org/pdf/2601.12449v1
+
+## 20968. Privacy-Preserving Federated Learning with Verifiable Fairness Guarantees
+
+- **ID:** `2601.12447` · **primary:** `cs.CR` · **Submitted:** 2026-01-18
+- **Authors:** Mohammed Himayath Ali, Mohammed Aqib Abdullah, Syed Muneer Hussain, et al.
+- **abs:** https://arxiv.org/abs/2601.12447 · **PDF:** https://arxiv.org/pdf/2601.12447v2
+
+## 20969. Large Language Model for OWL Proofs
+
+- **ID:** `2601.12444` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Hui Yang, Jiaoyan Chen, Uli Sattler
+- **abs:** https://arxiv.org/abs/2601.12444 · **PDF:** https://arxiv.org/pdf/2601.12444v1
+
+## 20970. Adversarial Defense in Vision-Language Models: An Overview
+
+- **ID:** `2601.12443` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Xiaowei Fu, Lei Zhang
+- **abs:** https://arxiv.org/abs/2601.12443 · **PDF:** https://arxiv.org/pdf/2601.12443v1
+
+## 20971. Constraint-Aware Neurosymbolic Uncertainty Quantification with Bayesian Deep Learning for Scientific Discovery
+
+- **ID:** `2601.12442` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Shahnawaz Alam, Mohammed Mudassir Uddin, Mohammed Kaif Pasha
+- **abs:** https://arxiv.org/abs/2601.12442 · **PDF:** https://arxiv.org/pdf/2601.12442v1
+
+## 20972. Purification Before Fusion: Toward Mask-Free Speech Enhancement for Robust Audio-Visual Speech Recognition
+
+- **ID:** `2601.12436` · **primary:** `eess.AS` · **Submitted:** 2026-01-18
+- **Authors:** Linzhi Wu, Xingyu Zhang, Hao Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.12436 · **PDF:** https://arxiv.org/pdf/2601.12436v2
+
+## 20973. Temporal Data and Short-Time Averages Improve Multiphase Mass Flow Metering
+
+- **ID:** `2601.12433` · **primary:** `eess.SP` · **Submitted:** 2026-01-18
+- **Authors:** Amanda Nyholm, Yessica Arellano, Jinyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.12433 · **PDF:** https://arxiv.org/pdf/2601.12433v1
+
+## 20974. System-Mediated Attention Imbalances Make Vision-Language Models Say Yes
+
+- **ID:** `2601.12430` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Tsan Tsai Chan, Varsha Suresh, Anisha Saha, et al.
+- **abs:** https://arxiv.org/abs/2601.12430 · **PDF:** https://arxiv.org/pdf/2601.12430v1
+
+## 20975. Stablecoin Design with Adversarial-Robust Multi-Agent Systems via Trust-Weighted Signal Aggregation
+
+- **ID:** `2601.22168` · **primary:** `q-fin.RM` · **Submitted:** 2026-01-18
+- **Authors:** Shengwei You, Aditya Joshi, Andrey Kuehlkamp, et al.
+- **abs:** https://arxiv.org/abs/2601.22168 · **PDF:** https://arxiv.org/pdf/2601.22168v1
+
+## 20976. Graph Attention Networks with Physical Constraints for Anomaly Detection
+
+- **ID:** `2601.12426` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Mohammadhossein Homaei, Iman Khazrak, Ruben Molano, et al.
+- **abs:** https://arxiv.org/abs/2601.12426 · **PDF:** https://arxiv.org/pdf/2601.12426v1
+
+## 20977. Legal experts disagree with rationale extraction techniques for explaining ECtHR case outcome classification
+
+- **ID:** `2601.12419` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Mahammad Namazov, Tomáš Koref, Ivan Habernal
+- **abs:** https://arxiv.org/abs/2601.12419 · **PDF:** https://arxiv.org/pdf/2601.12419v1
+
+## 20978. Orthogonalized Policy Optimization:Policy Optimization as Orthogonal Projection in Hilbert Space
+
+- **ID:** `2601.12415` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Wang Zixian
+- **abs:** https://arxiv.org/abs/2601.12415 · **PDF:** https://arxiv.org/pdf/2601.12415v5
+
+## 20979. Are LLMs Smarter Than Chimpanzees? An Evaluation on Perspective Taking and Knowledge State Estimation
+
+- **ID:** `2601.12410` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Dingyi Yang, Junqi Zhao, Xue Li, et al.
+- **abs:** https://arxiv.org/abs/2601.12410 · **PDF:** https://arxiv.org/pdf/2601.12410v2
+
+## 20980. De-Anonymization at Scale via Tournament-Style Attribution
+
+- **ID:** `2601.12407` · **primary:** `cs.CR` · **Submitted:** 2026-01-18
+- **Authors:** Lirui Zhang, Huishuai Zhang
+- **abs:** https://arxiv.org/abs/2601.12407 · **PDF:** https://arxiv.org/pdf/2601.12407v1
+
+## 20981. Explainable Machine Learning for Pediatric Dental Risk Stratification Using Socio-Demographic Determinants
+
+- **ID:** `2601.12405` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Manasi Kanade, Abhi Thakkar, Gabriela Fernandes
+- **abs:** https://arxiv.org/abs/2601.12405 · **PDF:** https://arxiv.org/pdf/2601.12405v1
+
+## 20982. Happy Young Women, Grumpy Old Men? Emotion-Driven Demographic Biases in Synthetic Face Generation
+
+- **ID:** `2602.00032` · **primary:** `cs.CY` · **Submitted:** 2026-01-18
+- **Authors:** Mengting Wei, Aditya Gulati, Guoying Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.00032 · **PDF:** https://arxiv.org/pdf/2602.00032v2
+
+## 20983. Weaknesses of Facial Emotion Recognition Systems
+
+- **ID:** `2601.12402` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Aleksandra Jamróz, Patrycja Wysocka, Piotr Garbat
+- **abs:** https://arxiv.org/abs/2601.12402 · **PDF:** https://arxiv.org/pdf/2601.12402v1
+
+## 20984. Beyond the Dirac Delta: Mitigating Diversity Collapse in Reinforcement Fine-Tuning for Versatile Image Generation
+
+- **ID:** `2601.12401` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Jinmei Liu, Haoru Li, Zhenhong Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.12401 · **PDF:** https://arxiv.org/pdf/2601.12401v1
+
+## 20985. BiCoLoR: Communication-Efficient Optimization with Bidirectional Compression and Local Training
+
+- **ID:** `2601.12400` · **primary:** `math.OC` · **Submitted:** 2026-01-18
+- **Authors:** Laurent Condat, Artavazd Maranjyan, Peter Richtárik
+- **abs:** https://arxiv.org/abs/2601.12400 · **PDF:** https://arxiv.org/pdf/2601.12400v1
+
+## 20986. PsychēChat: An Empathic Framework Focused on Emotion Shift Tracking and Safety Risk Analysis in Psychological Counseling
+
+- **ID:** `2601.12392` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Zhentao Xia, Yongqi Fan, Yuxiang Chu, et al.
+- **abs:** https://arxiv.org/abs/2601.12392 · **PDF:** https://arxiv.org/pdf/2601.12392v1
+
+## 20987. NADIR: Differential Attention Flow for Non-Autoregressive Transliteration in Indic Languages
+
+- **ID:** `2601.12389` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Lakshya Tomar, Vinayak Abrol, Puneet Agarwal
+- **abs:** https://arxiv.org/abs/2601.12389 · **PDF:** https://arxiv.org/pdf/2601.12389v1
+
+## 20988. Statistical-Neural Interaction Networks for Interpretable Mixed-Type Data Imputation
+
+- **ID:** `2601.12380` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Ou Deng, Shoji Nishimura, Atsushi Ogihara, et al.
+- **abs:** https://arxiv.org/abs/2601.12380 · **PDF:** https://arxiv.org/pdf/2601.12380v1
+
+## 20989. LR-DWM: Efficient Watermarking for Diffusion Language Models
+
+- **ID:** `2601.12376` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Ofek Raban, Ethan Fetaya, Gal Chechik
+- **abs:** https://arxiv.org/abs/2601.12376 · **PDF:** https://arxiv.org/pdf/2601.12376v1
+
+## 20990. LiQSS: Post-Transformer Linear Quantum-Inspired State-Space Tensor Networks for Real-Time 6G
+
+- **ID:** `2601.12375` · **primary:** `cs.NI` · **Submitted:** 2026-01-18
+- **Authors:** Farhad Rezazadeh, Hatim Chergui, Mehdi Bennis, et al.
+- **abs:** https://arxiv.org/abs/2601.12375 · **PDF:** https://arxiv.org/pdf/2601.12375v2
+
+## 20991. A Scalable Entity-Based Framework for Auditing Bias in LLMs
+
+- **ID:** `2601.12374` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Akram Elbouanani, Aboubacar Tuo, Adrian Popescu
+- **abs:** https://arxiv.org/abs/2601.12374 · **PDF:** https://arxiv.org/pdf/2601.12374v1
+
+## 20992. Can Deep Research Agents Retrieve and Organize? Evaluating the Synthesis Gap with Expert Taxonomies
+
+- **ID:** `2601.12369` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Ming Zhang, Jiabao Zhuang, Wenqing Jing, et al.
+- **abs:** https://arxiv.org/abs/2601.12369 · **PDF:** https://arxiv.org/pdf/2601.12369v2
+
+## 20993. Machine Learning-Based Framework for Real Time Detection and Early Prediction of Control Valve Stiction in Industrial Control Systems
+
+- **ID:** `2601.12362` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Natthapong Promsricha, Chotirawee Chatpattanasiri, Nuttavut Kerdgongsup, et al.
+- **abs:** https://arxiv.org/abs/2601.12362 · **PDF:** https://arxiv.org/pdf/2601.12362v1
+
+## 20994. Zero-Shot Embedding Drift Detection: A Lightweight Defense Against Prompt Injections in LLMs
+
+- **ID:** `2601.12359` · **primary:** `cs.CR` · **Submitted:** 2026-01-18
+- **Authors:** Anirudh Sekar, Mrinal Agarwal, Rachel Sharma, et al.
+- **abs:** https://arxiv.org/abs/2601.12359 · **PDF:** https://arxiv.org/pdf/2601.12359v1
+
+## 20995. From Prompts to Pavement: LMMs-based Agentic Behavior-Tree Generation Framework for Autonomous Vehicles
+
+- **ID:** `2601.12358` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Omar Y. Goba, Ahmed Y. Gado, Catherine M. Elias, et al.
+- **abs:** https://arxiv.org/abs/2601.12358 · **PDF:** https://arxiv.org/pdf/2601.12358v1
+
+## 20996. SimpleMatch: A Simple and Strong Baseline for Semantic Correspondence
+
+- **ID:** `2601.12357` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Hailing Jin, Huiying Li
+- **abs:** https://arxiv.org/abs/2601.12357 · **PDF:** https://arxiv.org/pdf/2601.12357v2
+
+## 20997. LB-MCTS: Synergizing Large Language Models and Bayesian Optimization for Efficient CASH
+
+- **ID:** `2601.12355` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Beicheng Xu, Weitong Qian, Lingching Tung, et al.
+- **abs:** https://arxiv.org/abs/2601.12355 · **PDF:** https://arxiv.org/pdf/2601.12355v1
+
+## 20998. Bone-conduction Guided Multimodal Speech Enhancement with Conditional Diffusion Models
+
+- **ID:** `2601.12354` · **primary:** `eess.AS` · **Submitted:** 2026-01-18
+- **Authors:** Sina Khanagha, Bunlong Lay, Timo Gerkmann
+- **abs:** https://arxiv.org/abs/2601.12354 · **PDF:** https://arxiv.org/pdf/2601.12354v1
+
+## 20999. RAPTOR-AI for Disaster OODA Loop: Hierarchical Multimodal RAG with Experience-Driven Agentic Decision-Making
+
+- **ID:** `2602.00030` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Takato Yasuno
+- **abs:** https://arxiv.org/abs/2602.00030 · **PDF:** https://arxiv.org/pdf/2602.00030v2
+
+## 21000. Zero-Permission Manipulation: Can We Trust Large Multimodal Model Powered GUI Agents?
+
+- **ID:** `2601.12349` · **primary:** `cs.CR` · **Submitted:** 2026-01-18
+- **Authors:** Yi Qian, Kunwei Qian, Xingbang He, et al.
+- **abs:** https://arxiv.org/abs/2601.12349 · **PDF:** https://arxiv.org/pdf/2601.12349v2
+
+## 21001. Adaptive Rotary Steering with Joint Autoregression for Robust Extraction of Closely Moving Speakers in Dynamic Scenarios
+
+- **ID:** `2601.12345` · **primary:** `eess.AS` · **Submitted:** 2026-01-18
+- **Authors:** Jakob Kienegger, Timo Gerkmann
+- **abs:** https://arxiv.org/abs/2601.12345 · **PDF:** https://arxiv.org/pdf/2601.12345v2
+
+## 21002. How Well Do LLMs Predict Human Behavior? A Measure of their Pretrained Knowledge
+
+- **ID:** `2601.12343` · **primary:** `econ.EM` · **Submitted:** 2026-01-18
+- **Authors:** Wayne Gao, Sukjin Han, Annie Liang
+- **abs:** https://arxiv.org/abs/2601.12343 · **PDF:** https://arxiv.org/pdf/2601.12343v1
+
+## 21003. Time-Continuous Modeling for Temporal Affective Pattern Recognition in LLMs
+
+- **ID:** `2601.12341` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Rezky Kam, Coddy N. Siswanto
+- **abs:** https://arxiv.org/abs/2601.12341 · **PDF:** https://arxiv.org/pdf/2601.12341v1
+
+## 21004. Actionable Advice from Reviews via Mixture of LoRA Experts: A Two-LLM Pipeline for Issue Extraction and Business Recommendations
+
+- **ID:** `2601.12338` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Kartikey Singh Bhandari, Manav Ganesh, Yashwant Viswanathan, et al.
+- **abs:** https://arxiv.org/abs/2601.12338 · **PDF:** https://arxiv.org/pdf/2601.12338v1
+
+## 21005. Efficient Privacy-Preserving Retrieval Augmented Generation with Distance-Preserving Encryption
+
+- **ID:** `2601.12331` · **primary:** `cs.CR` · **Submitted:** 2026-01-18
+- **Authors:** Huanyi Ye, Jiale Guo, Ziyao Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.12331 · **PDF:** https://arxiv.org/pdf/2601.12331v1
+
+## 21006. IceWatch: Forecasting Glacial Lake Outburst Floods (GLOFs) using Multimodal Deep Learning
+
+- **ID:** `2601.12330` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Zuha Fatima, Muhammad Anser Sohaib, Muhammad Talha, et al.
+- **abs:** https://arxiv.org/abs/2601.12330 · **PDF:** https://arxiv.org/pdf/2601.12330v1
+
+## 21007. The Expert Validation Framework (EVF): Enabling Domain Expert Control in AI Engineering
+
+- **ID:** `2601.12327` · **primary:** `cs.SE` · **Submitted:** 2026-01-18
+- **Authors:** Lucas Gren, Felix Dobslaw
+- **abs:** https://arxiv.org/abs/2601.12327 · **PDF:** https://arxiv.org/pdf/2601.12327v1
+
+## 21008. MARO: Learning Stronger Reasoning from Social Interaction
+
+- **ID:** `2601.12323` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Yin Cai, Zhouhong Gu, Juntao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.12323 · **PDF:** https://arxiv.org/pdf/2601.12323v2
+
+## 21009. Ordered Local Momentum for Asynchronous Distributed Learning under Arbitrary Delays
+
+- **ID:** `2601.12322` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Chang-Wei Shi, Shi-Shang Wang, Wu-Jun Li
+- **abs:** https://arxiv.org/abs/2601.12322 · **PDF:** https://arxiv.org/pdf/2601.12322v1
+
+## 21010. Beyond Human Annotation: Recent Advances in Data Generation Methods for Document Intelligence
+
+- **ID:** `2601.12318` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Dehao Ying, Fengchang Yu, Haihua Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.12318 · **PDF:** https://arxiv.org/pdf/2601.12318v1
+
+## 21011. Explanova: Automatically Discover Data Insights in N \times M Table via XAI Combined LLM Workflow
+
+- **ID:** `2601.12317` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Yiming Huang
+- **abs:** https://arxiv.org/abs/2601.12317 · **PDF:** https://arxiv.org/pdf/2601.12317v1
+
+## 21012. GazeFormer-MoE: Context-Aware Gaze Estimation via CLIP and MoE Transformer
+
+- **ID:** `2601.12316` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Xinyuan Zhao, Xianrui Chen, Ahmad Chaddad
+- **abs:** https://arxiv.org/abs/2601.12316 · **PDF:** https://arxiv.org/pdf/2601.12316v1
+
+## 21013. An Optimized Decision Tree-Based Framework for Explainable IoT Anomaly Detection
+
+- **ID:** `2601.14305` · **primary:** `cs.CR` · **Submitted:** 2026-01-18
+- **Authors:** Ashikuzzaman, Md. Shawkat Hossain, Jubayer Abdullah Joy, et al.
+- **abs:** https://arxiv.org/abs/2601.14305 · **PDF:** https://arxiv.org/pdf/2601.14305v1
+
+## 21014. Cross-reality Location Privacy Protection in 6G-enabled Vehicular Metaverses: An LLM-enhanced Hybrid Generative Diffusion Model-based Approach
+
+- **ID:** `2601.12311` · **primary:** `cs.NI` · **Submitted:** 2026-01-18
+- **Authors:** Xiaofeng Luo, Jiayi He, Jiawen Kang, et al.
+- **abs:** https://arxiv.org/abs/2601.12311 · **PDF:** https://arxiv.org/pdf/2601.12311v1
+
+## 21015. Survival is the Only Reward: Sustainable Self-Training Through Environment-Mediated Selection
+
+- **ID:** `2601.12310` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Jennifer Dodgson, Alfath Daryl Alhajir, Michael Joedhitya, et al.
+- **abs:** https://arxiv.org/abs/2601.12310 · **PDF:** https://arxiv.org/pdf/2601.12310v1
+
+## 21016. Adaptive Multi-Scale Correlation Meta-Network for Few-Shot Remote Sensing Image Classification
+
+- **ID:** `2601.12308` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Anurag Kaushish, Ayan Sar, Sampurna Roy, et al.
+- **abs:** https://arxiv.org/abs/2601.12308 · **PDF:** https://arxiv.org/pdf/2601.12308v1
+
+## 21017. Rethinking the Value of Multi-Agent Workflow: A Strong Single Agent Baseline
+
+- **ID:** `2601.12307` · **primary:** `cs.MA` · **Submitted:** 2026-01-18
+- **Authors:** Jiawei Xu, Arief Koesdwiady, Sisong Bei, et al.
+- **abs:** https://arxiv.org/abs/2601.12307 · **PDF:** https://arxiv.org/pdf/2601.12307v1
+
+## 21018. An Embodied Companion for Visual Storytelling
+
+- **ID:** `2603.05511` · **primary:** `cs.HC` · **Submitted:** 2026-01-18
+- **Authors:** Patrick Tresset, Markus Wulfmeier
+- **abs:** https://arxiv.org/abs/2603.05511 · **PDF:** https://arxiv.org/pdf/2603.05511v1
+
+## 21019. Machine Learning as a Service (MLaaS) Dataset Generator Framework for IoT Environments
+
+- **ID:** `2601.12305` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Deepak Kanneganti, Sajib Mistry, Sheik Fattah, et al.
+- **abs:** https://arxiv.org/abs/2601.12305 · **PDF:** https://arxiv.org/pdf/2601.12305v1
+
+## 21020. A Two-Stage Globally-Diverse Adversarial Attack for Vision-Language Pre-training Models
+
+- **ID:** `2601.12304` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Wutao Chen, Huaqin Zou, Chen Wan, et al.
+- **abs:** https://arxiv.org/abs/2601.12304 · **PDF:** https://arxiv.org/pdf/2601.12304v1
+
+## 21021. Distribution Shift Is Key to Learning Invariant Prediction
+
+- **ID:** `2601.12296` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Hong Zheng, Fei Teng
+- **abs:** https://arxiv.org/abs/2601.12296 · **PDF:** https://arxiv.org/pdf/2601.12296v1
+
+## 21022. ToolPRMBench: Evaluating and Advancing Process Reward Models for Tool-using Agents
+
+- **ID:** `2601.12294` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Dawei Li, Yuguang Yao, Zhen Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.12294 · **PDF:** https://arxiv.org/pdf/2601.12294v1
+
+## 21023. Guided by the Plan: Enhancing Faithful Autoregressive Text-to-Audio Generation with Guided Decoding
+
+- **ID:** `2601.14304` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Juncheng Wang, Zhe Hu, Chao Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.14304 · **PDF:** https://arxiv.org/pdf/2601.14304v1
+
+## 21024. ParaMETA: Towards Learning Disentangled Paralinguistic Speaking Styles Representations from Speech
+
+- **ID:** `2601.12289` · **primary:** `cs.SD` · **Submitted:** 2026-01-18
+- **Authors:** Haowei Lou, Hye-young Paik, Wen Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.12289 · **PDF:** https://arxiv.org/pdf/2601.12289v1
+
+## 21025. TimeGMM: Single-Pass Probabilistic Forecasting via Adaptive Gaussian Mixture Models with Reversible Normalization
+
+- **ID:** `2601.12288` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Lei Liu, Tengyuan Liu, Hongwei Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.12288 · **PDF:** https://arxiv.org/pdf/2601.12288v1
+
+## 21026. Conversational Context Classification: A Representation Engineering Approach
+
+- **ID:** `2601.12286` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Jonathan Pan
+- **abs:** https://arxiv.org/abs/2601.12286 · **PDF:** https://arxiv.org/pdf/2601.12286v1
+
+## 21027. CytoCLIP: Learning Cytoarchitectural Characteristics in Developing Human Brain Using Contrastive Language Image Pre-Training
+
+- **ID:** `2601.12282` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Pralaypati Ta, Sriram Venkatesaperumal, Keerthi Ram, et al.
+- **abs:** https://arxiv.org/abs/2601.12282 · **PDF:** https://arxiv.org/pdf/2601.12282v1
+
+## 21028. HCFT: Hierarchical Convolutional Fusion Transformer for EEG Decoding
+
+- **ID:** `2601.12279` · **primary:** `cs.HC` · **Submitted:** 2026-01-18
+- **Authors:** Haodong Zhang, Jiapeng Zhu, Yitong Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.12279 · **PDF:** https://arxiv.org/pdf/2601.12279v1
+
+## 21029. Predictive Prototyping: Evaluating Design Concepts with ChatGPT
+
+- **ID:** `2601.12276` · **primary:** `cs.HC` · **Submitted:** 2026-01-18
+- **Authors:** Hilsann Yong, Bradley A. Camburn
+- **abs:** https://arxiv.org/abs/2601.12276 · **PDF:** https://arxiv.org/pdf/2601.12276v2
+
+## 21030. Simulated Annealing Enhances Theory-of-Mind Reasoning in Autoregressive Language Models
+
+- **ID:** `2601.12269` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Xucong Hu, Jian-Qiao Zhu
+- **abs:** https://arxiv.org/abs/2601.12269 · **PDF:** https://arxiv.org/pdf/2601.12269v1
+
+## 21031. Construct, Align, and Reason: Large Ontology Models for Enterprise Knowledge Management
+
+- **ID:** `2602.00029` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Yao Zhang, Hongyin Zhu
+- **abs:** https://arxiv.org/abs/2602.00029 · **PDF:** https://arxiv.org/pdf/2602.00029v1
+
+## 21032. Statistical Firefly Algorithm for Truss Topology Optimization
+
+- **ID:** `2601.12265` · **primary:** `cs.NE` · **Submitted:** 2026-01-18
+- **Authors:** Nghi Huu Duong, Duy Vo, Pruettha Nanakorn
+- **abs:** https://arxiv.org/abs/2601.12265 · **PDF:** https://arxiv.org/pdf/2601.12265v1
+
+## 21033. Multimodal Generative Engine Optimization: Rank Manipulation for Vision-Language Model Rankers
+
+- **ID:** `2601.12263` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Yixuan Du, Chenxiao Yu, Haoyan Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.12263 · **PDF:** https://arxiv.org/pdf/2601.12263v1
+
+## 21034. Environment-Aware Code Generation: How far are We?
+
+- **ID:** `2601.12262` · **primary:** `cs.SE` · **Submitted:** 2026-01-18
+- **Authors:** Tongtong Wu, Rongyi Chen, Wenjie Du, et al.
+- **abs:** https://arxiv.org/abs/2601.12262 · **PDF:** https://arxiv.org/pdf/2601.12262v1
+
+## 21035. Docs2Synth: A Synthetic Data Trained Retriever Framework for Scanned Visually Rich Documents Understanding
+
+- **ID:** `2601.12260` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Yihao Ding, Qiang Sun, Puzhen Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.12260 · **PDF:** https://arxiv.org/pdf/2601.12260v1
+
+## 21036. FutureX-Pro: Extending Future Prediction to High-Value Vertical Domains
+
+- **ID:** `2601.12259` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Jiashuo Liu, Siyuan Chen, Zaiyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.12259 · **PDF:** https://arxiv.org/pdf/2601.12259v1
+
+## 21037. Soft Shadow Diffusion (SSD): Physics-inspired Learning for 3D Computational Periscopy
+
+- **ID:** `2601.12257` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Fadlullah Raji, John Murray-Bruce
+- **abs:** https://arxiv.org/abs/2601.12257 · **PDF:** https://arxiv.org/pdf/2601.12257v1
+
+## 21038. Improving Large Molecular Language Model via Relation-aware Multimodal Collaboration
+
+- **ID:** `2601.12256` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** Jinyoung Park, Minseong Bae, Jeehye Na, et al.
+- **abs:** https://arxiv.org/abs/2601.12256 · **PDF:** https://arxiv.org/pdf/2601.12256v1
+
+## 21039. Spark-LLM-Eval: A Distributed Framework for Statistically Rigorous Large Language Model Evaluation
+
+- **ID:** `2603.28769` · **primary:** `cs.DC` · **Submitted:** 2026-01-18
+- **Authors:** Subhadip Mitra
+- **abs:** https://arxiv.org/abs/2603.28769 · **PDF:** https://arxiv.org/pdf/2603.28769v1
+
+## 21040. Federated Joint Learning for Domain and Class Generalization
+
+- **ID:** `2601.12253` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Haoran Xu, Jiaze Li, Jianzhong Ju, et al.
+- **abs:** https://arxiv.org/abs/2601.12253 · **PDF:** https://arxiv.org/pdf/2601.12253v2
+
+## 21041. Large Language Models Predict Functional Outcomes after Acute Ischemic Stroke
+
+- **ID:** `2602.10119` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Anjali K. Kapoor, Anton Alyakin, Jin Vivian Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.10119 · **PDF:** https://arxiv.org/pdf/2602.10119v1
+
+## 21042. An Innovative Framework for Breast Cancer Detection Using Pyramid Adaptive Atrous Convolution, Transformer Integration, and Multi-Scale Feature Fusion
+
+- **ID:** `2601.12249` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Ehsan Sadeghi Pour, Mahdi Esmaeili, Morteza Romoozi
+- **abs:** https://arxiv.org/abs/2601.12249 · **PDF:** https://arxiv.org/pdf/2601.12249v1
+
+## 21043. AQUA-Bench: Beyond Finding Answers to Knowing When There Are None in Audio Question Answering
+
+- **ID:** `2601.12248` · **primary:** `eess.AS` · **Submitted:** 2026-01-18
+- **Authors:** Chun-Yi Kuan, Hung-yi Lee
+- **abs:** https://arxiv.org/abs/2601.12248 · **PDF:** https://arxiv.org/pdf/2601.12248v1
+
+## 21044. Plan, Verify and Fill: A Structured Parallel Decoding Approach for Diffusion Language Models
+
+- **ID:** `2601.12247` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Miao Li, Hanyang Jiang, Sikai Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.12247 · **PDF:** https://arxiv.org/pdf/2601.12247v2
+
+## 21045. A Comprehensive Review of Bio-Inspired Approaches to Coordination, Communication, and System Architecture in Underwater Swarm Robotics
+
+- **ID:** `2601.12244` · **primary:** `cs.RO` · **Submitted:** 2026-01-18
+- **Authors:** Shyalan Ramesh, Scott Mann, Alex Stumpf
+- **abs:** https://arxiv.org/abs/2601.12244 · **PDF:** https://arxiv.org/pdf/2601.12244v1
+
+## 21046. Less is More: Label-Guided Summarization of Procedural and Instructional Videos
+
+- **ID:** `2601.12243` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Shreya Rajpal, Michal Golovanevsky, Carsten Eickhoff
+- **abs:** https://arxiv.org/abs/2601.12243 · **PDF:** https://arxiv.org/pdf/2601.12243v2
+
+## 21047. Optimal Power Allocation and Sub-Optimal Channel Assignment for Downlink NOMA Systems Using Deep Reinforcement Learning
+
+- **ID:** `2601.12242` · **primary:** `cs.AI` · **Submitted:** 2026-01-18
+- **Authors:** WooSeok Kim, Jeonghoon Lee, Sangho Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.12242 · **PDF:** https://arxiv.org/pdf/2601.12242v1
+
+## 21048. On the Provable Suboptimality of Momentum SGD in Nonstationary Stochastic Optimization
+
+- **ID:** `2601.12238` · **primary:** `stat.ML` · **Submitted:** 2026-01-18
+- **Authors:** Sharan Sahu, Cameron J. Hogan, Martin T. Wells
+- **abs:** https://arxiv.org/abs/2601.12238 · **PDF:** https://arxiv.org/pdf/2601.12238v3
+
+## 21049. DDSA: Dual-Domain Strategic Attack for Spatial-Temporal Efficiency in Adversarial Robustness Testing
+
+- **ID:** `2601.14302` · **primary:** `cs.CR` · **Submitted:** 2026-01-18
+- **Authors:** Jinwei Hu, Shiyuan Meng, Yi Dong, et al.
+- **abs:** https://arxiv.org/abs/2601.14302 · **PDF:** https://arxiv.org/pdf/2601.14302v1
+
+## 21050. Proc3D: Procedural 3D Generation and Parametric Editing of 3D Shapes with Large Language Models
+
+- **ID:** `2601.12234` · **primary:** `cs.GR` · **Submitted:** 2026-01-18
+- **Authors:** Fadlullah Raji, Stefano Petrangeli, Matheus Gadelha, et al.
+- **abs:** https://arxiv.org/abs/2601.12234 · **PDF:** https://arxiv.org/pdf/2601.12234v1
+
+## 21051. Wavelet-Aware Anomaly Detection in Multi-Channel User Logs via Deviation Modulation and Resolution-Adaptive Attention
+
+- **ID:** `2601.12231` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Kaichuan Kong, Dongjie Liu, Xiaobo Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.12231 · **PDF:** https://arxiv.org/pdf/2601.12231v1
+
+## 21052. Learning Longitudinal Health Representations from EHR and Wearable Data
+
+- **ID:** `2601.12227` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Yuanyun Zhang, Han Zhou, Li Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.12227 · **PDF:** https://arxiv.org/pdf/2601.12227v1
+
+## 21053. Exploring Human-in-the-Loop Themes in AI Application Development: An Empirical Thematic Analysis
+
+- **ID:** `2603.05510` · **primary:** `cs.HC` · **Submitted:** 2026-01-18
+- **Authors:** Parm Suksakul, Nathan Kittichaikoonkij, Nakhin Polthai, et al.
+- **abs:** https://arxiv.org/abs/2603.05510 · **PDF:** https://arxiv.org/pdf/2603.05510v1
+
+## 21054. Where It Moves, It Matters: Referring Surgical Instrument Segmentation via Motion
+
+- **ID:** `2601.12224` · **primary:** `cs.CV` · **Submitted:** 2026-01-18
+- **Authors:** Meng Wei, Kun Yuan, Shi Li, et al.
+- **abs:** https://arxiv.org/abs/2601.12224 · **PDF:** https://arxiv.org/pdf/2601.12224v1
+
+## 21055. Persistent Sheaf Laplacian Analysis of Protein Stability and Solubility Changes upon Mutation
+
+- **ID:** `2601.12219` · **primary:** `math.SP` · **Submitted:** 2026-01-18
+- **Authors:** Yiming Ren, Junjie Wee, Xi Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.12219 · **PDF:** https://arxiv.org/pdf/2601.12219v1
+
+## 21056. Wavelet-Driven Masked Multiscale Reconstruction for PPG Foundation Models
+
+- **ID:** `2601.12215` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Megha Thukral, Cyrus Tanade, Simon A. Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.12215 · **PDF:** https://arxiv.org/pdf/2601.12215v1
+
+## 21057. One-Sided Matrix Completion from Ultra-Sparse Samples
+
+- **ID:** `2601.12213` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Hongyang R. Zhang, Zhenshuo Zhang, Huy L. Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.12213 · **PDF:** https://arxiv.org/pdf/2601.12213v1
+
+## 21058. Speculative Sampling with Reinforcement Learning
+
+- **ID:** `2601.12212` · **primary:** `cs.LG` · **Submitted:** 2026-01-18
+- **Authors:** Chenan Wang, Daniel H. Shi, Haipeng Chen
+- **abs:** https://arxiv.org/abs/2601.12212 · **PDF:** https://arxiv.org/pdf/2601.12212v1
+
+## 21059. CoReflect: Conversational Evaluation via Co-Evolutionary Simulation and Reflective Rubric Refinement
+
+- **ID:** `2601.12208` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Yunzhe Li, Richie Yueqi Feng, Tianxin Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.12208 · **PDF:** https://arxiv.org/pdf/2601.12208v1
+
+## 21060. Do Neural Codecs Generalize? A Controlled Study Across Unseen Languages and Non-Speech Tasks
+
+- **ID:** `2601.12205` · **primary:** `cs.SD` · **Submitted:** 2026-01-18
+- **Authors:** Shih-Heng Wang, Jiatong Shi, Jinchuan Tian, et al.
+- **abs:** https://arxiv.org/abs/2601.12205 · **PDF:** https://arxiv.org/pdf/2601.12205v1
+
+## 21061. CTC-DID: CTC-Based Arabic dialect identification for streaming applications
+
+- **ID:** `2601.12199` · **primary:** `cs.CL` · **Submitted:** 2026-01-18
+- **Authors:** Muhammad Umar Farooq, Oscar Saz
+- **abs:** https://arxiv.org/abs/2601.12199 · **PDF:** https://arxiv.org/pdf/2601.12199v1
+
+## 21062. PALMA: A Lightweight Tropical Algebra Library for ARM-Based Embedded Systems
+
+- **ID:** `2601.17028` · **primary:** `cs.MS` · **Submitted:** 2026-01-17
+- **Authors:** Gnankan Landry Regis N'guessan
+- **abs:** https://arxiv.org/abs/2601.17028 · **PDF:** https://arxiv.org/pdf/2601.17028v1
+
+## 21063. Aletheia: What Makes RLVR For Code Verifiers Tick?
+
+- **ID:** `2601.12186` · **primary:** `cs.SE` · **Submitted:** 2026-01-17
+- **Authors:** Vatsal Venkatkrishna, Indraneil Paul, Iryna Gurevych
+- **abs:** https://arxiv.org/abs/2601.12186 · **PDF:** https://arxiv.org/pdf/2601.12186v2
+
+## 21064. Tolerance Principle and Small Language Model Learning
+
+- **ID:** `2601.12179` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Adam E. Friedman, Stevan Harnad, Rushen Shi
+- **abs:** https://arxiv.org/abs/2601.12179 · **PDF:** https://arxiv.org/pdf/2601.12179v1
+
+## 21065. Federated Learning for the Design of Parametric Insurance Indices under Heterogeneous Renewable Production Losses
+
+- **ID:** `2601.12178` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Fallou Niakh
+- **abs:** https://arxiv.org/abs/2601.12178 · **PDF:** https://arxiv.org/pdf/2601.12178v1
+
+## 21066. Small Updates, Big Doubts: Does Parameter-Efficient Fine-tuning Enhance Hallucination Detection ?
+
+- **ID:** `2602.11166` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Xu Hu, Yifan Zhang, Songtao Wei, et al.
+- **abs:** https://arxiv.org/abs/2602.11166 · **PDF:** https://arxiv.org/pdf/2602.11166v1
+
+## 21067. Assessing LLM Reliability on Temporally Recent Open-Domain Questions
+
+- **ID:** `2602.11165` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Pushwitha Krishnappa, Amit Das, Vinija Jain, et al.
+- **abs:** https://arxiv.org/abs/2602.11165 · **PDF:** https://arxiv.org/pdf/2602.11165v1
+
+## 21068. The Language You Ask In: Language-Conditioned Ideological Divergence in LLM Analysis of Contested Political Documents
+
+- **ID:** `2601.12164` · **primary:** `cs.CY` · **Submitted:** 2026-01-17
+- **Authors:** Oleg Smirnov
+- **abs:** https://arxiv.org/abs/2601.12164 · **PDF:** https://arxiv.org/pdf/2601.12164v2
+
+## 21069. Streaming Operator Inference for Model Reduction of Large-Scale Dynamical Systems
+
+- **ID:** `2601.12161` · **primary:** `math.NA` · **Submitted:** 2026-01-17
+- **Authors:** Tomoki Koike, Prakash Mohan, Marc T. Henry de Frahan, et al.
+- **abs:** https://arxiv.org/abs/2601.12161 · **PDF:** https://arxiv.org/pdf/2601.12161v2
+
+## 21070. Reviewing the Reviewer: Elevating Peer Review Quality through LLM-Guided Feedback
+
+- **ID:** `2602.10118` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Sukannya Purkayastha, Qile Wan, Anne Lauscher, et al.
+- **abs:** https://arxiv.org/abs/2602.10118 · **PDF:** https://arxiv.org/pdf/2602.10118v1
+
+## 21071. Analyzing Cancer Patients' Experiences with Embedding-based Topic Modeling and LLMs
+
+- **ID:** `2601.12154` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Teodor-Călin Ionescu, Lifeng Han, Jan Heijdra Suasnabar, et al.
+- **abs:** https://arxiv.org/abs/2601.12154 · **PDF:** https://arxiv.org/pdf/2601.12154v1
+
+## 21072. Enhanced Diagnostic Performance via Large-Resolution Inference Optimization for Pathology Foundation Models
+
+- **ID:** `2601.12150` · **primary:** `cs.CV` · **Submitted:** 2026-01-17
+- **Authors:** Mengxuan Hu, Zihan Guan, John Kang, et al.
+- **abs:** https://arxiv.org/abs/2601.12150 · **PDF:** https://arxiv.org/pdf/2601.12150v1
+
+## 21073. Replayable Financial Agents: A Determinism-Faithfulness Assurance Harness for Tool-Using LLM Agents
+
+- **ID:** `2601.15322` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Raffi Khatchadourian
+- **abs:** https://arxiv.org/abs/2601.15322 · **PDF:** https://arxiv.org/pdf/2601.15322v2
+
+## 21074. Segment and Matte Anything in a Unified Model
+
+- **ID:** `2601.12147` · **primary:** `cs.CV` · **Submitted:** 2026-01-17
+- **Authors:** Zezhong Fan, Xiaohan Li, Topojoy Biswas, et al.
+- **abs:** https://arxiv.org/abs/2601.12147 · **PDF:** https://arxiv.org/pdf/2601.12147v1
+
+## 21075. Threshold Differential Attention for Sink-Free, Ultra-Sparse, and Non-Dispersive Language Modeling
+
+- **ID:** `2601.12145` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Xingyue Huang, Xueying Ding, Mingxuan Ju, et al.
+- **abs:** https://arxiv.org/abs/2601.12145 · **PDF:** https://arxiv.org/pdf/2601.12145v1
+
+## 21076. TIDE: A Trace-Informed Depth-First Exploration for Planning with Temporally Extended Goals
+
+- **ID:** `2601.12141` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Yuliia Suprun, Khen Elimelech, Lydia E. Kavraki, et al.
+- **abs:** https://arxiv.org/abs/2601.12141 · **PDF:** https://arxiv.org/pdf/2601.12141v1
+
+## 21077. DriveSafe: A Hierarchical Risk Taxonomy for Safety-Critical LLM-Based Driving Assistants
+
+- **ID:** `2601.12138` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Abhishek Kumar, Riya Tapwal, Carsten Maple
+- **abs:** https://arxiv.org/abs/2601.12138 · **PDF:** https://arxiv.org/pdf/2601.12138v3
+
+## 21078. EMoE: Eigenbasis-Guided Routing for Mixture-of-Experts
+
+- **ID:** `2601.12137` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Anzhe Cheng, Shukai Duan, Shixuan Li, et al.
+- **abs:** https://arxiv.org/abs/2601.12137 · **PDF:** https://arxiv.org/pdf/2601.12137v1
+
+## 21079. Human-Human-AI Triadic Programming: Uncovering the Role of AI Agent and the Value of Human Partner in Collaborative Learning
+
+- **ID:** `2601.12134` · **primary:** `cs.HC` · **Submitted:** 2026-01-17
+- **Authors:** Taufiq Daryanto, Xiaohan Ding, Kaike Ping, et al.
+- **abs:** https://arxiv.org/abs/2601.12134 · **PDF:** https://arxiv.org/pdf/2601.12134v1
+
+## 21080. Bengali Text Classification: An Evaluation of Large Language Model Approaches
+
+- **ID:** `2601.12132` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Md Mahmudul Hoque, Md Mehedi Hassain, Md Hojaifa Tanvir, et al.
+- **abs:** https://arxiv.org/abs/2601.12132 · **PDF:** https://arxiv.org/pdf/2601.12132v1
+
+## 21081. SolarGPT-QA: A Domain-Adaptive Large Language Model for Educational Question Answering in Space Weather and Heliophysics
+
+- **ID:** `2601.12131` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Santosh Chapagain, MohammadReza EskandariNasab, Onur Vural, et al.
+- **abs:** https://arxiv.org/abs/2601.12131 · **PDF:** https://arxiv.org/pdf/2601.12131v4
+
+## 21082. UniMo: Unified Motion Generation and Understanding with Chain of Thought
+
+- **ID:** `2601.12126` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Guocun Wang, Kenkun Liu, Jing Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.12126 · **PDF:** https://arxiv.org/pdf/2601.12126v1
+
+## 21083. SynQP: A Framework and Metrics for Evaluating the Quality and Privacy Risk of Synthetic Data
+
+- **ID:** `2601.12124` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Bing Hu, Yixin Li, Asma Bahamyirou, et al.
+- **abs:** https://arxiv.org/abs/2601.12124 · **PDF:** https://arxiv.org/pdf/2601.12124v1
+
+## 21084. Lost in Aggregation: The Causal Interpretation of the IV Estimand
+
+- **ID:** `2601.12120` · **primary:** `stat.ME` · **Submitted:** 2026-01-17
+- **Authors:** Danielle Tsao, Krikamol Muandet, Frederick Eberhardt, et al.
+- **abs:** https://arxiv.org/abs/2601.12120 · **PDF:** https://arxiv.org/pdf/2601.12120v1
+
+## 21085. Offline Policy Learning with Weight Clipping and Heaviside Composite Optimization
+
+- **ID:** `2601.12117` · **primary:** `math.OC` · **Submitted:** 2026-01-17
+- **Authors:** Jingren Liu, Hanzhang Qin, Junyi Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.12117 · **PDF:** https://arxiv.org/pdf/2601.12117v1
+
+## 21086. Powerful Training-Free Membership Inference Against Autoregressive Language Models
+
+- **ID:** `2601.12104` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** David Ilić, David Stanojević, Kostadin Cvejoski
+- **abs:** https://arxiv.org/abs/2601.12104 · **PDF:** https://arxiv.org/pdf/2601.12104v1
+
+## 21087. Large language models struggle with ethnographic text annotation
+
+- **ID:** `2601.12099` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Leonardo S. Goodall, Dor Shilton, Daniel A. Mullins, et al.
+- **abs:** https://arxiv.org/abs/2601.12099 · **PDF:** https://arxiv.org/pdf/2601.12099v1
+
+## 21088. Neural Isomorphic Fields: A Transformer-based Algebraic Numerical Embedding
+
+- **ID:** `2601.12095` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Hamidreza Sadeghi, Saeedeh Momtazi, Reza Safabakhsh
+- **abs:** https://arxiv.org/abs/2601.12095 · **PDF:** https://arxiv.org/pdf/2601.12095v1
+
+## 21089. PTL-PINNs: Perturbation-Guided Transfer Learning with Physics- Informed Neural Networks for Nonlinear Systems
+
+- **ID:** `2601.12093` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Duarte Alexandrino, Ben Moseley, Pavlos Protopapas
+- **abs:** https://arxiv.org/abs/2601.12093 · **PDF:** https://arxiv.org/pdf/2601.12093v1
+
+## 21090. Mitigating Cultural Bias in LLMs via Multi-Agent Cultural Debate
+
+- **ID:** `2601.12091` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Qian Tan, Lei Jiang, Yuting Zeng, et al.
+- **abs:** https://arxiv.org/abs/2601.12091 · **PDF:** https://arxiv.org/pdf/2601.12091v1
+
+## 21091. Learning to Factorize and Adapt: A Versatile Approach Toward Universal Spatio-Temporal Foundation Models
+
+- **ID:** `2601.12083` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Siru Zhong, Junjie Qiu, Yangyu Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.12083 · **PDF:** https://arxiv.org/pdf/2601.12083v1
+
+## 21092. Conditional Random Fields for Interactive Refinement of Histopathological Predictions
+
+- **ID:** `2601.12082` · **primary:** `cs.CV` · **Submitted:** 2026-01-17
+- **Authors:** Tiffanie Godelaine, Maxime Zanella, Karim El Khoury, et al.
+- **abs:** https://arxiv.org/abs/2601.12082 · **PDF:** https://arxiv.org/pdf/2601.12082v1
+
+## 21093. Optimizing User Profiles via Contextual Bandits for Retrieval-Augmented LLM Personalization
+
+- **ID:** `2601.12078` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Linfeng Du, Ye Yuan, Zichen Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.12078 · **PDF:** https://arxiv.org/pdf/2601.12078v1
+
+## 21094. CroBIM-V: Memory-Quality Controlled Remote Sensing Referring Video Object Segmentation
+
+- **ID:** `2601.12076` · **primary:** `cs.CV` · **Submitted:** 2026-01-17
+- **Authors:** H. Jiang, Y. Sun, Z. Dong, et al.
+- **abs:** https://arxiv.org/abs/2601.12076 · **PDF:** https://arxiv.org/pdf/2601.12076v1
+
+## 21095. ELLMPEG: An Edge-based Agentic LLM Video Processing Tool
+
+- **ID:** `2602.00028` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Zoha Azimi, Reza Farahani, Radu Prodan, et al.
+- **abs:** https://arxiv.org/abs/2602.00028 · **PDF:** https://arxiv.org/pdf/2602.00028v1
+
+## 21096. To Copy or Not to Copy: Copying Is Easier to Induce Than Recall
+
+- **ID:** `2601.12075` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Mehrdad Farahani, Franziska Penzkofer, Richard Johansson
+- **abs:** https://arxiv.org/abs/2601.12075 · **PDF:** https://arxiv.org/pdf/2601.12075v1
+
+## 21097. Representation Learning Enhanced Deep Reinforcement Learning for Optimal Operation of Hydrogen-based Multi-Energy Systems
+
+- **ID:** `2602.00027` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Zhenyu Pu, Yu Yang, Lun Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.00027 · **PDF:** https://arxiv.org/pdf/2602.00027v1
+
+## 21098. Bridging the Gap in Bangla Healthcare: Machine Learning Based Disease Prediction Using a Symptoms-Disease Dataset
+
+- **ID:** `2601.12068` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Rowzatul Zannat, Abdullah Al Shafi, Abdul Muntakim
+- **abs:** https://arxiv.org/abs/2601.12068 · **PDF:** https://arxiv.org/pdf/2601.12068v1
+
+## 21099. Scientific Image Synthesis: Benchmarking, Methodologies, and Downstream Utility
+
+- **ID:** `2601.17027` · **primary:** `cs.CV` · **Submitted:** 2026-01-17
+- **Authors:** Honglin Lin, Chonghan Qin, Zheng Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.17027 · **PDF:** https://arxiv.org/pdf/2601.17027v1
+
+## 21100. Codebook-Injected Dialogue Segmentation for Multi-Utterance Constructs Annotation: LLM-Assisted and Gold-Label-Free Evaluation
+
+- **ID:** `2601.12061` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Jinsook Lee, Kirk Vanacore, Zhuqian Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.12061 · **PDF:** https://arxiv.org/pdf/2601.12061v2
+
+## 21101. Automating Parameter Selection in Deep Image Prior for Fluorescence Microscopy Image Denoising via Similarity-Based Parameter Transfer
+
+- **ID:** `2601.12055` · **primary:** `cs.CV` · **Submitted:** 2026-01-17
+- **Authors:** Lina Meyer, Felix Wissel, Tobias Knopp, et al.
+- **abs:** https://arxiv.org/abs/2601.12055 · **PDF:** https://arxiv.org/pdf/2601.12055v1
+
+## 21102. A New Strategy for Artificial Intelligence: Training Foundation Models Directly on Human Brain Data
+
+- **ID:** `2601.12053` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-17
+- **Authors:** Maël Donoso
+- **abs:** https://arxiv.org/abs/2601.12053 · **PDF:** https://arxiv.org/pdf/2601.12053v1
+
+## 21103. \textit{FocaLogic}: Logic-Based Interpretation of Visual Model Decisions
+
+- **ID:** `2601.12049` · **primary:** `cs.CV` · **Submitted:** 2026-01-17
+- **Authors:** Chenchen Zhao, Muxi Chen, Qiang Xu
+- **abs:** https://arxiv.org/abs/2601.12049 · **PDF:** https://arxiv.org/pdf/2601.12049v1
+
+## 21104. Less Is More -- Until It Breaks: Security Pitfalls of Vision Token Compression in Large Vision-Language Models
+
+- **ID:** `2601.12042` · **primary:** `cs.CR` · **Submitted:** 2026-01-17
+- **Authors:** Xiaomei Zhang, Zhaoxi Zhang, Leo Yu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.12042 · **PDF:** https://arxiv.org/pdf/2601.12042v1
+
+## 21105. Partial Reasoning in Language Models: Search and Refinement Guided by Uncertainty
+
+- **ID:** `2601.12040` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Murilo da Luz, Bruno Brandão, Luana Martins, et al.
+- **abs:** https://arxiv.org/abs/2601.12040 · **PDF:** https://arxiv.org/pdf/2601.12040v1
+
+## 21106. Nonlinear Dynamic Factor Analysis With a Transformer Network
+
+- **ID:** `2601.12039` · **primary:** `econ.EM` · **Submitted:** 2026-01-17
+- **Authors:** Oliver Snellman
+- **abs:** https://arxiv.org/abs/2601.12039 · **PDF:** https://arxiv.org/pdf/2601.12039v1
+
+## 21107. Abstract Argumentation with Subargument Relations
+
+- **ID:** `2601.12038` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Beishui Liao
+- **abs:** https://arxiv.org/abs/2601.12038 · **PDF:** https://arxiv.org/pdf/2601.12038v2
+
+## 21108. Wildfire Suppression: Complexity, Models, and Instances
+
+- **ID:** `2603.29865` · **primary:** `cs.CE` · **Submitted:** 2026-01-17
+- **Authors:** Gustavo Delazeri, Marcus Ritt
+- **abs:** https://arxiv.org/abs/2603.29865 · **PDF:** https://arxiv.org/pdf/2603.29865v1
+
+## 21109. Don't Start Over: A Cost-Effective Framework for Migrating Personalized Prompts Between LLMs
+
+- **ID:** `2601.12034` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Ziyi Zhao, Chongming Gao, Yang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.12034 · **PDF:** https://arxiv.org/pdf/2601.12034v1
+
+## 21110. Preserving Fairness and Safety in Quantized LLMs Through Critical Weight Protection
+
+- **ID:** `2601.12033` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Muhammad Alif Al Hakim, Alfan Farizki Wicaksono, Fajri Koto
+- **abs:** https://arxiv.org/abs/2601.12033 · **PDF:** https://arxiv.org/pdf/2601.12033v1
+
+## 21111. Speaking to Silicon: Neural Communication with Bitcoin Mining ASICs
+
+- **ID:** `2601.12032` · **primary:** `cs.NE` · **Submitted:** 2026-01-17
+- **Authors:** Francisco Angulo de Lafuente, Vladimir Veselov, Richard Goodman
+- **abs:** https://arxiv.org/abs/2601.12032 · **PDF:** https://arxiv.org/pdf/2601.12032v1
+
+## 21112. ARC: Active and Reflection-driven Context Management for Long-Horizon Information Seeking Agents
+
+- **ID:** `2601.12030` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Yilun Yao, Shan Huang, Elsie Dai, et al.
+- **abs:** https://arxiv.org/abs/2601.12030 · **PDF:** https://arxiv.org/pdf/2601.12030v1
+
+## 21113. A Multi-Agent System for Generating Actionable Business Advice
+
+- **ID:** `2601.12024` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Kartikey Singh Bhandari, Tanish Jain, Archit Agrawal, et al.
+- **abs:** https://arxiv.org/abs/2601.12024 · **PDF:** https://arxiv.org/pdf/2601.12024v1
+
+## 21114. A Kernel Approach for Semi-implicit Variational Inference
+
+- **ID:** `2601.12023` · **primary:** `stat.ML` · **Submitted:** 2026-01-17
+- **Authors:** Longlin Yu, Ziheng Cheng, Shiyue Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.12023 · **PDF:** https://arxiv.org/pdf/2601.12023v1
+
+## 21115. Acting Flatterers via LLMs Sycophancy: Combating Clickbait with LLMs Opposing-Stance Reasoning
+
+- **ID:** `2601.12019` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Chaowei Zhang, Xiansheng Luo, Zewei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.12019 · **PDF:** https://arxiv.org/pdf/2601.12019v1
+
+## 21116. Are LLMs Ready for TOON? Benchmarking Structural Correctness-Sustainability Trade-offs in Novel Structured Output Formats
+
+- **ID:** `2601.12014` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Elio Masciari, Vincenzo Moscato, Enea Vincenzo Napolitano, et al.
+- **abs:** https://arxiv.org/abs/2601.12014 · **PDF:** https://arxiv.org/pdf/2601.12014v1
+
+## 21117. Why Loss Re-weighting Works If You Stop Early: Training Dynamics of Unconstrained Features
+
+- **ID:** `2601.12011` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Yize Zhao, Christos Thrampoulidis
+- **abs:** https://arxiv.org/abs/2601.12011 · **PDF:** https://arxiv.org/pdf/2601.12011v1
+
+## 21118. Extreme Value Policy Optimization for Safe Reinforcement Learning
+
+- **ID:** `2601.12008` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Shiqing Gao, Yihang Zhou, Shuai Shao, et al.
+- **abs:** https://arxiv.org/abs/2601.12008 · **PDF:** https://arxiv.org/pdf/2601.12008v1
+
+## 21119. Robust Verification of Concurrent Stochastic Games
+
+- **ID:** `2601.12003` · **primary:** `cs.LO` · **Submitted:** 2026-01-17
+- **Authors:** Angel Y. He, David Parker
+- **abs:** https://arxiv.org/abs/2601.12003 · **PDF:** https://arxiv.org/pdf/2601.12003v2
+
+## 21120. Kernel-Based Learning of Safety Barriers
+
+- **ID:** `2601.12002` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Oliver Schön, Zhengang Zhong, Sadegh Soudjani
+- **abs:** https://arxiv.org/abs/2601.12002 · **PDF:** https://arxiv.org/pdf/2601.12002v1
+
+## 21121. Strategies for Creating Uncertainty in the AI Era to Trigger Students Critical Thinking: Pedagogical Design, Assessment Rubric, and Exam System
+
+- **ID:** `2602.00026` · **primary:** `cs.CY` · **Submitted:** 2026-01-17
+- **Authors:** Ahmad Samer Wazan
+- **abs:** https://arxiv.org/abs/2602.00026 · **PDF:** https://arxiv.org/pdf/2602.00026v1
+
+## 21122. Hybrid IDS Using Signature-Based and Anomaly-Based Detection
+
+- **ID:** `2601.11998` · **primary:** `cs.CR` · **Submitted:** 2026-01-17
+- **Authors:** Messaouda Boutassetta, Amina Makhlouf, Newfel Messaoudi, et al.
+- **abs:** https://arxiv.org/abs/2601.11998 · **PDF:** https://arxiv.org/pdf/2601.11998v1
+
+## 21123. MongoDB Injection Query Classification Model using MongoDB Log files as Training Data
+
+- **ID:** `2601.11996` · **primary:** `cs.CR` · **Submitted:** 2026-01-17
+- **Authors:** Shaunak Perni, Minal Shirodkar, Ramdas Karmalli
+- **abs:** https://arxiv.org/abs/2601.11996 · **PDF:** https://arxiv.org/pdf/2601.11996v1
+
+## 21124. Learning Audio-Visual Embeddings with Inferred Latent Interaction Graphs
+
+- **ID:** `2601.11995` · **primary:** `cs.MM` · **Submitted:** 2026-01-17
+- **Authors:** Donghuo Zeng, Hao Niu, Yanan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.11995 · **PDF:** https://arxiv.org/pdf/2601.11995v1
+
+## 21125. Automated Optimization Modeling via a Localizable Error-Driven Perspective
+
+- **ID:** `2602.11164` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Weiting Liu, Han Wu, Yufei Kuang, et al.
+- **abs:** https://arxiv.org/abs/2602.11164 · **PDF:** https://arxiv.org/pdf/2602.11164v1
+
+## 21126. Process In-Context Learning: Enhancing Mathematical Reasoning via Dynamic Demonstration Insertion
+
+- **ID:** `2601.11979` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Ang Gao, Changshuo Zhang, Xiao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.11979 · **PDF:** https://arxiv.org/pdf/2601.11979v1
+
+## 21127. One-Shot Price Forecasting with Covariate-Guided Experts under Privacy Constraints
+
+- **ID:** `2601.11977` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Ren He, Yinliang Xu, Jinfeng Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.11977 · **PDF:** https://arxiv.org/pdf/2601.11977v1
+
+## 21128. Learn Like Humans: Use Meta-cognitive Reflection for Efficient Self-Improvement
+
+- **ID:** `2601.11974` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Xinmeng Hou, Peiliang Gong, Bohao Qu, et al.
+- **abs:** https://arxiv.org/abs/2601.11974 · **PDF:** https://arxiv.org/pdf/2601.11974v1
+
+## 21129. MemoryRewardBench: Benchmarking Reward Models for Long-Term Memory Management in Large Language Models
+
+- **ID:** `2601.11969` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Zecheng Tang, Baibei Ji, Ruoxi Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.11969 · **PDF:** https://arxiv.org/pdf/2601.11969v2
+
+## 21130. Nested Named Entity Recognition in Plasma Physics Research Articles
+
+- **ID:** `2602.11163` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Muhammad Haris, Hans Höft, Markus M. Becker, et al.
+- **abs:** https://arxiv.org/abs/2602.11163 · **PDF:** https://arxiv.org/pdf/2602.11163v1
+
+## 21131. R$^2$PO: Decoupling Training Trajectories from Inference Responses for LLM Reasoning
+
+- **ID:** `2601.11960` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Jingchu Wang, Bingbing Xu, Yige Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.11960 · **PDF:** https://arxiv.org/pdf/2601.11960v2
+
+## 21132. PEARL: Self-Evolving Assistant for Time Management with Reinforcement Learning
+
+- **ID:** `2601.11957` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Bingxuan Li, Jeonghwan Kim, Cheng Qian, et al.
+- **abs:** https://arxiv.org/abs/2601.11957 · **PDF:** https://arxiv.org/pdf/2601.11957v2
+
+## 21133. Double-Calibration: Towards Trustworthy LLMs via Calibrating Knowledge and Reasoning Confidence
+
+- **ID:** `2601.11956` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Yuyin Lu, Ziran Liang, Yanghui Rao, et al.
+- **abs:** https://arxiv.org/abs/2601.11956 · **PDF:** https://arxiv.org/pdf/2601.11956v1
+
+## 21134. Data-centric Prompt Tuning for Dynamic Graphs
+
+- **ID:** `2601.11954` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Yufei Peng, Cheng Yang, Zhengjie Fan, et al.
+- **abs:** https://arxiv.org/abs/2601.11954 · **PDF:** https://arxiv.org/pdf/2601.11954v1
+
+## 21135. Controlling Underestimation Bias in Constrained Reinforcement Learning for Safe Exploration
+
+- **ID:** `2601.11953` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Shiqing Gao, Jiaxin Ding, Luoyi Fu, et al.
+- **abs:** https://arxiv.org/abs/2601.11953 · **PDF:** https://arxiv.org/pdf/2601.11953v1
+
+## 21136. An Overlay Multicast Routing Method Based on Network Situational Aware-ness and Hierarchical Multi-Agent Reinforcement Learning
+
+- **ID:** `2602.13211` · **primary:** `cs.NI` · **Submitted:** 2026-01-17
+- **Authors:** Miao Ye, Yanye Chen, Yong Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.13211 · **PDF:** https://arxiv.org/pdf/2602.13211v1
+
+## 21137. Trainability-Oriented Hybrid Quantum Regression via Geometric Preconditioning and Curriculum Optimization
+
+- **ID:** `2601.11942` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Qingyu Meng, Yangshuai Wang
+- **abs:** https://arxiv.org/abs/2601.11942 · **PDF:** https://arxiv.org/pdf/2601.11942v2
+
+## 21138. Thinking Traps in Long Chain-of-Thought: A Measurable Study and Trap-Aware Adaptive Restart
+
+- **ID:** `2601.11940` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Kang Chen, Fan Yu, Junjie Nian, et al.
+- **abs:** https://arxiv.org/abs/2601.11940 · **PDF:** https://arxiv.org/pdf/2601.11940v1
+
+## 21139. Impact of Circuit Depth versus Qubit Count on Variational Quantum Classifiers for Higgs Boson Signal Detection
+
+- **ID:** `2601.11937` · **primary:** `quant-ph` · **Submitted:** 2026-01-17
+- **Authors:** Fatih Maulana
+- **abs:** https://arxiv.org/abs/2601.11937 · **PDF:** https://arxiv.org/pdf/2601.11937v1
+
+## 21140. Big Data Workload Profiling for Energy-Aware Cloud Resource Management
+
+- **ID:** `2601.11935` · **primary:** `cs.DC` · **Submitted:** 2026-01-17
+- **Authors:** Milan Parikh, Aniket Abhishek Soni, Sneja Mitinbhai Shah, et al.
+- **abs:** https://arxiv.org/abs/2601.11935 · **PDF:** https://arxiv.org/pdf/2601.11935v1
+
+## 21141. Event Detection with a Context-Aware Encoder and LoRA for Improved Performance on Long-Tailed Classes
+
+- **ID:** `2601.11932` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Abdullah Al Monsur, Nitesh Vamshi Bommisetty, Gene Louis Kim
+- **abs:** https://arxiv.org/abs/2601.11932 · **PDF:** https://arxiv.org/pdf/2601.11932v2
+
+## 21142. Harmonica: A Self-Adaptation Exemplar for Sustainable MLOps
+
+- **ID:** `2601.11926` · **primary:** `cs.SE` · **Submitted:** 2026-01-17
+- **Authors:** Ananya Halgatti, Shaunak Biswas, Hiya Bhatt, et al.
+- **abs:** https://arxiv.org/abs/2601.11926 · **PDF:** https://arxiv.org/pdf/2601.11926v2
+
+## 21143. Communication-Corruption Coupling and Verification in Cooperative Multi-Objective Bandits
+
+- **ID:** `2601.11924` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Ming Shi
+- **abs:** https://arxiv.org/abs/2601.11924 · **PDF:** https://arxiv.org/pdf/2601.11924v2
+
+## 21144. Mapping the maturation of TCM as an adjuvant to radiotherapy
+
+- **ID:** `2601.11923` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** P. Bilha Githinji, Aikaterini Melliou, Xi Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.11923 · **PDF:** https://arxiv.org/pdf/2601.11923v1
+
+## 21145. Enhancing LLM-Based Data Annotation with Error Decomposition
+
+- **ID:** `2601.11920` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Zhen Xu, Vedant Khatri, Yijun Dai, et al.
+- **abs:** https://arxiv.org/abs/2601.11920 · **PDF:** https://arxiv.org/pdf/2601.11920v1
+
+## 21146. Effects of Gabor Filters on Classification Performance of CNNs Trained on a Limited Number of Conditions
+
+- **ID:** `2601.11918` · **primary:** `cs.CV` · **Submitted:** 2026-01-17
+- **Authors:** Akito Morita, Hirotsugu Okuno
+- **abs:** https://arxiv.org/abs/2601.11918 · **PDF:** https://arxiv.org/pdf/2601.11918v1
+
+## 21147. LSTM-MAS: A Long Short-Term Memory Inspired Multi-Agent System for Long-Context Understanding
+
+- **ID:** `2601.11913` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Yichen Jiang, Peng Ye, Jiakang Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.11913 · **PDF:** https://arxiv.org/pdf/2601.11913v1
+
+## 21148. PPA-Plan: Proactive Pitfall Avoidance for Reliable Planning in Long-Context LLM Reasoning
+
+- **ID:** `2601.11908` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Byeongjin Kim, Gyuwan Kim, Seo Yeon Park
+- **abs:** https://arxiv.org/abs/2601.11908 · **PDF:** https://arxiv.org/pdf/2601.11908v1
+
+## 21149. Towards Airborne Object Detection: A Deep Learning Analysis
+
+- **ID:** `2601.11907` · **primary:** `cs.CV` · **Submitted:** 2026-01-17
+- **Authors:** Prosenjit Chatterjee, ANK Zaman
+- **abs:** https://arxiv.org/abs/2601.11907 · **PDF:** https://arxiv.org/pdf/2601.11907v1
+
+## 21150. LIBRA: Language Model Informed Bandit Recourse Algorithm for Personalized Treatment Planning
+
+- **ID:** `2601.11905` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Junyu Cao, Ruijiang Gao, Esmaeil Keyvanshokooh, et al.
+- **abs:** https://arxiv.org/abs/2601.11905 · **PDF:** https://arxiv.org/pdf/2601.11905v1
+
+## 21151. AEMA: Verifiable Evaluation Framework for Trustworthy and Controlled Agentic LLM Systems
+
+- **ID:** `2601.11903` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** YenTing Lee, Keerthi Koneru, Zahra Moslemi, et al.
+- **abs:** https://arxiv.org/abs/2601.11903 · **PDF:** https://arxiv.org/pdf/2601.11903v1
+
+## 21152. Task-tailored Pre-processing: Fair Downstream Supervised Learning
+
+- **ID:** `2601.11897` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Jinwon Sohn, Guang Lin, Qifan Song
+- **abs:** https://arxiv.org/abs/2601.11897 · **PDF:** https://arxiv.org/pdf/2601.11897v1
+
+## 21153. DevBench: A Realistic, Developer-Informed Benchmark for Code Generation Models
+
+- **ID:** `2601.11895` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Pareesa Ameneh Golnari, Adarsh Kumarappan, Wen Wen, et al.
+- **abs:** https://arxiv.org/abs/2601.11895 · **PDF:** https://arxiv.org/pdf/2601.11895v2
+
+## 21154. From Relative Entropy to Minimax: A Unified Framework for Coverage in MDPs
+
+- **ID:** `2601.11890` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Xihe Gu, Urbashi Mitra, Tara Javidi
+- **abs:** https://arxiv.org/abs/2601.11890 · **PDF:** https://arxiv.org/pdf/2601.11890v1
+
+## 21155. Agentic-R: Learning to Retrieve for Agentic Search
+
+- **ID:** `2601.11888` · **primary:** `cs.IR` · **Submitted:** 2026-01-17
+- **Authors:** Wenhan Liu, Xinyu Ma, Yutao Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.11888 · **PDF:** https://arxiv.org/pdf/2601.11888v1
+
+## 21156. Faithfulness vs. Safety: Evaluating LLM Behavior Under Counterfactual Medical Evidence
+
+- **ID:** `2601.11886` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Kaijie Mo, Siddhartha Venkatayogi, Chantal Shaib, et al.
+- **abs:** https://arxiv.org/abs/2601.11886 · **PDF:** https://arxiv.org/pdf/2601.11886v1
+
+## 21157. MyGram: Modality-aware Graph Transformer with Global Distribution for Multi-modal Entity Alignment
+
+- **ID:** `2601.11885` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Zhifei Li, Ziyue Qin, Xiangyu Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.11885 · **PDF:** https://arxiv.org/pdf/2601.11885v1
+
+## 21158. Gradient Structure Estimation under Label-Only Oracles via Spectral Sensitivity
+
+- **ID:** `2601.14300` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Jun Liu, Leo Yu Zhang, Fengpeng Li, et al.
+- **abs:** https://arxiv.org/abs/2601.14300 · **PDF:** https://arxiv.org/pdf/2601.14300v2
+
+## 21159. Approximation Algorithm for Constrained $k$-Center Clustering: A Local Search Approach
+
+- **ID:** `2601.11883` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Chaoqi Jia, Longkun Guo, Kewen Liao, et al.
+- **abs:** https://arxiv.org/abs/2601.11883 · **PDF:** https://arxiv.org/pdf/2601.11883v1
+
+## 21160. TF-CoDiT: Conditional Time Series Synthesis with Diffusion Transformers for Treasury Futures
+
+- **ID:** `2601.11880` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Yingxiao Zhang, Jiaxin Duan, Junfu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.11880 · **PDF:** https://arxiv.org/pdf/2601.11880v1
+
+## 21161. Accelerated MR Elastography Using Learned Neural Network Representation
+
+- **ID:** `2601.11878` · **primary:** `eess.SP` · **Submitted:** 2026-01-17
+- **Authors:** Xi Peng
+- **abs:** https://arxiv.org/abs/2601.11878 · **PDF:** https://arxiv.org/pdf/2601.11878v1
+
+## 21162. AI for Green Spaces: Leveraging Autonomous Navigation and Computer Vision for Park Litter Removal
+
+- **ID:** `2601.11876` · **primary:** `cs.RO` · **Submitted:** 2026-01-17
+- **Authors:** Christopher Kao, Akhil Pathapati, James Davis
+- **abs:** https://arxiv.org/abs/2601.11876 · **PDF:** https://arxiv.org/pdf/2601.11876v1
+
+## 21163. GloCTM: Cross-Lingual Topic Modeling via a Global Context Space
+
+- **ID:** `2601.11872` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Nguyen Tien Phat, Ngo Vu Minh, Linh Van Ngo, et al.
+- **abs:** https://arxiv.org/abs/2601.11872 · **PDF:** https://arxiv.org/pdf/2601.11872v1
+
+## 21164. Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces
+
+- **ID:** `2601.11868` · **primary:** `cs.SE` · **Submitted:** 2026-01-17
+- **Authors:** Mike A. Merrill, Alexander G. Shaw, Nicholas Carlini, et al.
+- **abs:** https://arxiv.org/abs/2601.11868 · **PDF:** https://arxiv.org/pdf/2601.11868v1
+
+## 21165. Advances in LLM Reasoning Enable Flexibility in Clinical Problem-Solving
+
+- **ID:** `2601.11866` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Kie Shidara, Preethi Prem, Jonathan Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.11866 · **PDF:** https://arxiv.org/pdf/2601.11866v1
+
+## 21166. CTPD: Cross Tokenizer Preference Distillation
+
+- **ID:** `2601.11865` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Truong Nguyen, Phi Van Dat, Ngan Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.11865 · **PDF:** https://arxiv.org/pdf/2601.11865v1
+
+## 21167. AGGC: Adaptive Group Gradient Clipping for Stabilizing Large Language Model Training
+
+- **ID:** `2601.11864` · **primary:** `cs.LG` · **Submitted:** 2026-01-17
+- **Authors:** Zhiyuan Li, Yuan Wu, Yi Chang
+- **abs:** https://arxiv.org/abs/2601.11864 · **PDF:** https://arxiv.org/pdf/2601.11864v1
+
+## 21168. Utilizing Metadata for Better Retrieval-Augmented Generation
+
+- **ID:** `2601.11863` · **primary:** `cs.IR` · **Submitted:** 2026-01-17
+- **Authors:** Raquib Bin Yousuf, Shengzhe Xu, Mandar Sharma, et al.
+- **abs:** https://arxiv.org/abs/2601.11863 · **PDF:** https://arxiv.org/pdf/2601.11863v1
+
+## 21169. Adversarial Drift-Aware Predictive Transfer: Toward Durable Clinical AI
+
+- **ID:** `2601.11860` · **primary:** `stat.AP` · **Submitted:** 2026-01-17
+- **Authors:** Xin Xiong, Zijian Guo, Haobo Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.11860 · **PDF:** https://arxiv.org/pdf/2601.11860v2
+
+## 21170. Cascaded Transformer for Robust and Scalable SLA Decomposition via Amortized Optimization
+
+- **ID:** `2601.11859` · **primary:** `cs.NI` · **Submitted:** 2026-01-17
+- **Authors:** Cyril Shih-Huan Hsu
+- **abs:** https://arxiv.org/abs/2601.11859 · **PDF:** https://arxiv.org/pdf/2601.11859v1
+
+## 21171. ATOD: An Evaluation Framework and Benchmark for Agentic Task-Oriented Dialogue Systems
+
+- **ID:** `2601.11854` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Yifei Zhang, Hooshang Nayyeri, Rinat Khaziev, et al.
+- **abs:** https://arxiv.org/abs/2601.11854 · **PDF:** https://arxiv.org/pdf/2601.11854v2
+
+## 21172. Human-AI Collaborative Inductive Thematic Analysis: AI Guided Analysis and Human Interpretive Authority
+
+- **ID:** `2601.11850` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Matthew Nyaaba, Min SungEun, Mary Abiswin Apam, et al.
+- **abs:** https://arxiv.org/abs/2601.11850 · **PDF:** https://arxiv.org/pdf/2601.11850v1
+
+## 21173. The Third VoicePrivacy Challenge: Preserving Emotional Expressiveness and Linguistic Content in Voice Anonymization
+
+- **ID:** `2601.11846` · **primary:** `cs.CL` · **Submitted:** 2026-01-17
+- **Authors:** Natalia Tomashenko, Xiaoxiao Miao, Pierre Champion, et al.
+- **abs:** https://arxiv.org/abs/2601.11846 · **PDF:** https://arxiv.org/pdf/2601.11846v1
+
+## 21174. Imandra CodeLogician: Neuro-Symbolic Reasoning for Precise Analysis of Software Logic
+
+- **ID:** `2601.11840` · **primary:** `cs.AI` · **Submitted:** 2026-01-17
+- **Authors:** Hongyu Lin, Samer Abdallah, Makar Valentinov, et al.
+- **abs:** https://arxiv.org/abs/2601.11840 · **PDF:** https://arxiv.org/pdf/2601.11840v2
+
+## 21175. Karhunen-Loève Expansion-Based Residual Anomaly Map for Resource-Efficient Glioma MRI Segmentation
+
+- **ID:** `2601.11833` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-16
+- **Authors:** Anthony Hur
+- **abs:** https://arxiv.org/abs/2601.11833 · **PDF:** https://arxiv.org/pdf/2601.11833v2
+
+## 21176. Shortest-Path Flow Matching with Mixture-Conditioned Bases for OOD Generalization to Unseen Conditions
+
+- **ID:** `2601.11827` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Andrea Rubbi, Amir Akbarnejad, Mohammad Vali Sanian, et al.
+- **abs:** https://arxiv.org/abs/2601.11827 · **PDF:** https://arxiv.org/pdf/2601.11827v2
+
+## 21177. Emergent Specialization in Learner Populations: Competition as the Source of Diversity
+
+- **ID:** `2601.19943` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Yuhao Li
+- **abs:** https://arxiv.org/abs/2601.19943 · **PDF:** https://arxiv.org/pdf/2601.19943v1
+
+## 21178. Latent Object Permanence: Topological Phase Transitions, Free-Energy Principles, and Renormalization Group Flows in Deep Transformer Manifolds
+
+- **ID:** `2601.19942` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Faruk Alpay, Bugra Kilictas
+- **abs:** https://arxiv.org/abs/2601.19942 · **PDF:** https://arxiv.org/pdf/2601.19942v1
+
+## 21179. AI Co-Scientist for Knowledge Synthesis in Medical Contexts: A Proof of Concept
+
+- **ID:** `2601.11825` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Arya Rahgozar, Pouria Mortezaagha
+- **abs:** https://arxiv.org/abs/2601.11825 · **PDF:** https://arxiv.org/pdf/2601.11825v1
+
+## 21180. RAPID-Serve: Resource-efficient and Accelerated P/D Intra-GPU Disaggregation
+
+- **ID:** `2601.11822` · **primary:** `cs.DC` · **Submitted:** 2026-01-16
+- **Authors:** Amna Masood, Pratishtha Gaur, Nuwan Jayasena
+- **abs:** https://arxiv.org/abs/2601.11822 · **PDF:** https://arxiv.org/pdf/2601.11822v1
+
+## 21181. Shapelets-Enriched Selective Forecasting using Time Series Foundation Models
+
+- **ID:** `2601.11821` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Shivani Tomar, Seshu Tirupathi, Elizabeth Daly, et al.
+- **abs:** https://arxiv.org/abs/2601.11821 · **PDF:** https://arxiv.org/pdf/2601.11821v1
+
+## 21182. TWeddit : A Dataset of Triggering Stories Predominantly Shared by Women on Reddit
+
+- **ID:** `2601.11819` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Shirlene Rose Bandela, Sanjeev Parthasarathy, Vaibhav Garg
+- **abs:** https://arxiv.org/abs/2601.11819 · **PDF:** https://arxiv.org/pdf/2601.11819v1
+
+## 21183. POLARIS: Typed Planning and Governed Execution for Agentic AI in Back-Office Automation
+
+- **ID:** `2601.11816` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Zahra Moslemi, Keerthi Koneru, Yen-Ting Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.11816 · **PDF:** https://arxiv.org/pdf/2601.11816v1
+
+## 21184. Multi-agent DRL-based Lane Change Decision Model for Cooperative Planning in Mixed Traffic
+
+- **ID:** `2601.11809` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Zeyu Mu, Shangtong Zhang, B. Brian Park
+- **abs:** https://arxiv.org/abs/2601.11809 · **PDF:** https://arxiv.org/pdf/2601.11809v1
+
+## 21185. RobotDesignGPT: Automated Robot Design Synthesis using Vision Language Models
+
+- **ID:** `2601.11801` · **primary:** `cs.RO` · **Submitted:** 2026-01-16
+- **Authors:** Nitish Sontakke, K. Niranjan Kumar, Sehoon Ha
+- **abs:** https://arxiv.org/abs/2601.11801 · **PDF:** https://arxiv.org/pdf/2601.11801v1
+
+## 21186. Physics-Constrained Denoising Autoencoders for Data-Scarce Wildfire UAV Sensing
+
+- **ID:** `2601.11794` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Abdelrahman Ramadan, Zahra Dorbeigi Namaghi, Emily Taylor, et al.
+- **abs:** https://arxiv.org/abs/2601.11794 · **PDF:** https://arxiv.org/pdf/2601.11794v1
+
+## 21187. A self-evolving multi-role collaborative framework with fine-grained difficulty guidance for innovative mathematical problem generation
+
+- **ID:** `2601.11792` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Yifei Sun, Yongan Li, A. K. Qin, et al.
+- **abs:** https://arxiv.org/abs/2601.11792 · **PDF:** https://arxiv.org/pdf/2601.11792v1
+
+## 21188. Beyond Tokens: Concept-Level Training Objectives for LLMs
+
+- **ID:** `2601.11791` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Laya Iyer, Pranav Somani, Alice Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.11791 · **PDF:** https://arxiv.org/pdf/2601.11791v2
+
+## 21189. Gradient-based Active Learning with Gaussian Processes for Global Sensitivity Analysis
+
+- **ID:** `2601.11790` · **primary:** `stat.ML` · **Submitted:** 2026-01-16
+- **Authors:** Guerlain Lambert, Céline Helbert, Claire Lauvernet
+- **abs:** https://arxiv.org/abs/2601.11790 · **PDF:** https://arxiv.org/pdf/2601.11790v1
+
+## 21190. Suspicious Alignment of SGD: A Fine-Grained Step Size Condition Analysis
+
+- **ID:** `2601.11789` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Shenyang Deng, Boyao Liao, Zhuoli Ouyang, et al.
+- **abs:** https://arxiv.org/abs/2601.11789 · **PDF:** https://arxiv.org/pdf/2601.11789v1
+
+## 21191. The Stability Trap: Evaluating the Reliability of LLM-Based Instruction Adherence Auditing
+
+- **ID:** `2601.11783` · **primary:** `cs.SE` · **Submitted:** 2026-01-16
+- **Authors:** Murtuza N. Shergadwala
+- **abs:** https://arxiv.org/abs/2601.11783 · **PDF:** https://arxiv.org/pdf/2601.11783v1
+
+## 21192. Risk-Aware Human-in-the-Loop Framework with Adaptive Intrusion Response for Autonomous Vehicles
+
+- **ID:** `2601.11781` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Dawood Wasif, Terrence J. Moore, Seunghyun Yoon, et al.
+- **abs:** https://arxiv.org/abs/2601.11781 · **PDF:** https://arxiv.org/pdf/2601.11781v1
+
+## 21193. Translation as a Scalable Proxy for Multilingual Evaluation
+
+- **ID:** `2601.11778` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Sheriff Issaka, Erick Rosas Gonzalez, Lieqi Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.11778 · **PDF:** https://arxiv.org/pdf/2601.11778v1
+
+## 21194. Cleansing the Artificial Mind: A Self-Reflective Detoxification Framework for Large Language Models
+
+- **ID:** `2601.11776` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Kaituo Zhang, Zhimeng Jiang, Na Zou
+- **abs:** https://arxiv.org/abs/2601.11776 · **PDF:** https://arxiv.org/pdf/2601.11776v1
+
+## 21195. Quantum Kernel Machine Learning for Autonomous Materials Science
+
+- **ID:** `2601.11775` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-16
+- **Authors:** Felix Adams, Daiwei Zhu, David W. Steuerman, et al.
+- **abs:** https://arxiv.org/abs/2601.11775 · **PDF:** https://arxiv.org/pdf/2601.11775v1
+
+## 21196. Bench4HLS: End-to-End Evaluation of LLMs in High-Level Synthesis Code Generation
+
+- **ID:** `2601.19941` · **primary:** `cs.AR` · **Submitted:** 2026-01-16
+- **Authors:** M Zafir Sadik Khan, Kimia Azar, Hadi Kamali
+- **abs:** https://arxiv.org/abs/2601.19941 · **PDF:** https://arxiv.org/pdf/2601.19941v1
+
+## 21197. Lightweight Self-Supervised Detection of Fundamental Frequency and Accurate Probability of Voicing in Monophonic Music
+
+- **ID:** `2601.11768` · **primary:** `eess.AS` · **Submitted:** 2026-01-16
+- **Authors:** Venkat Suprabath Bitra, Homayoon Beigi
+- **abs:** https://arxiv.org/abs/2601.11768 · **PDF:** https://arxiv.org/pdf/2601.11768v1
+
+## 21198. Guardrails for trust, safety, and ethical development and deployment of Large Language Models (LLM)
+
+- **ID:** `2601.14298` · **primary:** `cs.CR` · **Submitted:** 2026-01-16
+- **Authors:** Anjanava Biswas, Wrick Talukdar
+- **abs:** https://arxiv.org/abs/2601.14298 · **PDF:** https://arxiv.org/pdf/2601.14298v1
+
+## 21199. Industry-Aligned Granular Topic Modeling
+
+- **ID:** `2601.11762` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Sae Young Moon, Myeongjun Erik Jang, Haoyan Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.11762 · **PDF:** https://arxiv.org/pdf/2601.11762v1
+
+## 21200. Early Linguistic Pattern of Anxiety from Social Media Using Interpretable Linguistic Features: A Multi-Faceted Validation Study with Author-Disjoint Evaluation
+
+- **ID:** `2601.11758` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Arnab Das Utsa
+- **abs:** https://arxiv.org/abs/2601.11758 · **PDF:** https://arxiv.org/pdf/2601.11758v1
+
+## 21201. Measurement for Opaque Systems: Multi-source Triangulation with Interpretable Machine Learning
+
+- **ID:** `2602.00022` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Margaret Foster
+- **abs:** https://arxiv.org/abs/2602.00022 · **PDF:** https://arxiv.org/pdf/2602.00022v1
+
+## 21202. PRISM: Learning Design Knowledge from Data for Stylistic Design Improvement
+
+- **ID:** `2601.11747` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Huaxiaoyue Wang, Sunav Choudhary, Franck Dernoncourt, et al.
+- **abs:** https://arxiv.org/abs/2601.11747 · **PDF:** https://arxiv.org/pdf/2601.11747v1
+
+## 21203. LIME-LLM: Probing Models with Fluent Counterfactuals, Not Broken Text
+
+- **ID:** `2601.11746` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** George Mihaila, Suleyman Olcay Polat, Poli Nemkova, et al.
+- **abs:** https://arxiv.org/abs/2601.11746 · **PDF:** https://arxiv.org/pdf/2601.11746v1
+
+## 21204. On Nonasymptotic Confidence Intervals for Treatment Effects in Randomized Experiments
+
+- **ID:** `2601.11744` · **primary:** `stat.ME` · **Submitted:** 2026-01-16
+- **Authors:** Ricardo J. Sandoval, Sivaraman Balakrishnan, Avi Feller, et al.
+- **abs:** https://arxiv.org/abs/2601.11744 · **PDF:** https://arxiv.org/pdf/2601.11744v2
+
+## 21205. Bridging Human Interpretation and Machine Representation: A Landscape of Qualitative Data Analysis in the LLM Era
+
+- **ID:** `2601.11739` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Xinyu Pi, Qisen Yang, Chuong Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.11739 · **PDF:** https://arxiv.org/pdf/2601.11739v1
+
+## 21206. SpaRRTa: A Synthetic Benchmark for Evaluating Spatial Intelligence in Visual Foundation Models
+
+- **ID:** `2601.11729` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Turhan Can Kargin, Wojciech Jasiński, Adam Pardyl, et al.
+- **abs:** https://arxiv.org/abs/2601.11729 · **PDF:** https://arxiv.org/pdf/2601.11729v1
+
+## 21207. A Proof of Concept for a Digital Twin of an Ultrasonic Fermentation System
+
+- **ID:** `2601.11723` · **primary:** `cs.ET` · **Submitted:** 2026-01-16
+- **Authors:** Francesco Saverio Sconocchia Pisoni, Andrea Vitaletti, Davide Appolloni, et al.
+- **abs:** https://arxiv.org/abs/2601.11723 · **PDF:** https://arxiv.org/pdf/2601.11723v1
+
+## 21208. RAC: Retrieval-Augmented Clarification for Faithful Conversational Search
+
+- **ID:** `2601.11722` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Ahmed Rayane Kebir, Vincent Guigue, Lynda Said Lhadj, et al.
+- **abs:** https://arxiv.org/abs/2601.11722 · **PDF:** https://arxiv.org/pdf/2601.11722v1
+
+## 21209. jBOT: Semantic Jet Representation Clustering Emerges from Self-Distillation
+
+- **ID:** `2601.11719` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Ho Fung Tsoi, Dylan Rankin
+- **abs:** https://arxiv.org/abs/2601.11719 · **PDF:** https://arxiv.org/pdf/2601.11719v2
+
+## 21210. AllShowers: One model for all calorimeter showers
+
+- **ID:** `2601.11716` · **primary:** `physics.ins-det` · **Submitted:** 2026-01-16
+- **Authors:** Thorsten Buss, Henry Day-Hall, Frank Gaede, et al.
+- **abs:** https://arxiv.org/abs/2601.11716 · **PDF:** https://arxiv.org/pdf/2601.11716v1
+
+## 21211. Inter-Cell Interference Rejection Based on Ultrawideband Walsh-Domain Wireless Autoencoding
+
+- **ID:** `2601.11713` · **primary:** `eess.SP` · **Submitted:** 2026-01-16
+- **Authors:** Rodney Martinez Alonso, Cel Thys, Cedric Dehos, et al.
+- **abs:** https://arxiv.org/abs/2601.11713 · **PDF:** https://arxiv.org/pdf/2601.11713v1
+
+## 21212. How Long Is a Piece of String? A Brief Empirical Analysis of Tokenizers
+
+- **ID:** `2601.11518` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Jonathan Roberts, Kai Han, Samuel Albanie
+- **abs:** https://arxiv.org/abs/2601.11518 · **PDF:** https://arxiv.org/pdf/2601.11518v1
+
+## 21213. PASTA: A Scalable Framework for Multi-Policy AI Compliance Evaluation
+
+- **ID:** `2601.11702` · **primary:** `cs.HC` · **Submitted:** 2026-01-16
+- **Authors:** Yu Yang, Ig-Jae Kim, Dongwook Yoon
+- **abs:** https://arxiv.org/abs/2601.11702 · **PDF:** https://arxiv.org/pdf/2601.11702v2
+
+## 21214. Do explanations generalize across large reasoning models?
+
+- **ID:** `2601.11517` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Koyena Pal, David Bau, Chandan Singh
+- **abs:** https://arxiv.org/abs/2601.11517 · **PDF:** https://arxiv.org/pdf/2601.11517v1
+
+## 21215. Building Production-Ready Probes For Gemini
+
+- **ID:** `2601.11516` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** János Kramár, Joshua Engels, Zheng Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.11516 · **PDF:** https://arxiv.org/pdf/2601.11516v4
+
+## 21216. ShapeR: Robust Conditional 3D Shape Generation from Casual Captures
+
+- **ID:** `2601.11514` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Yawar Siddiqui, Duncan Frost, Samir Aroudj, et al.
+- **abs:** https://arxiv.org/abs/2601.11514 · **PDF:** https://arxiv.org/pdf/2601.11514v1
+
+## 21217. Stability and Accuracy Trade-offs in Statistical Estimation
+
+- **ID:** `2601.11701` · **primary:** `math.ST` · **Submitted:** 2026-01-16
+- **Authors:** Abhinav Chakraborty, Yuetian Luo, Rina Foygel Barber
+- **abs:** https://arxiv.org/abs/2601.11701 · **PDF:** https://arxiv.org/pdf/2601.11701v1
+
+## 21218. Telling Human and Machine Handwriting Apart
+
+- **ID:** `2601.11700` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Luis A. Leiva, Moises Diaz, Nuwan T. Attygalle, et al.
+- **abs:** https://arxiv.org/abs/2601.11700 · **PDF:** https://arxiv.org/pdf/2601.11700v1
+
+## 21219. Early Warning Signals Appear Long Before Dropping Out: An Idiographic Approach Grounded in Complex Dynamic Systems Theory
+
+- **ID:** `2602.00021` · **primary:** `cs.CY` · **Submitted:** 2026-01-16
+- **Authors:** Mohammed Saqr, Sonsoles López-Pernas, Santtu Tikka, et al.
+- **abs:** https://arxiv.org/abs/2602.00021 · **PDF:** https://arxiv.org/pdf/2602.00021v1
+
+## 21220. MetaboNet: The Largest Publicly Available Consolidated Dataset for Type 1 Diabetes Management
+
+- **ID:** `2601.11505` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Miriam K. Wolff, Peter Calhoun, Eleonora Maria Aiello, et al.
+- **abs:** https://arxiv.org/abs/2601.11505 · **PDF:** https://arxiv.org/pdf/2601.11505v1
+
+## 21221. QUPID: A Partitioned Quantum Neural Network for Anomaly Detection in Smart Grid
+
+- **ID:** `2601.11500` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Hoang M. Ngo, Tre' R. Jeter, Jung Taek Seo, et al.
+- **abs:** https://arxiv.org/abs/2601.11500 · **PDF:** https://arxiv.org/pdf/2601.11500v1
+
+## 21222. On the Probability of First Success in Differential Evolution: Hazard Identities and Tail Bounds
+
+- **ID:** `2601.11499` · **primary:** `cs.NE` · **Submitted:** 2026-01-16
+- **Authors:** Dimitar Nedanovski, Svetoslav Nenov, Dimitar Pilev
+- **abs:** https://arxiv.org/abs/2601.11499 · **PDF:** https://arxiv.org/pdf/2601.11499v1
+
+## 21223. The Poisoned Apple Effect: Strategic Manipulation of Mediated Markets via Technology Expansion of AI Agents
+
+- **ID:** `2601.11496` · **primary:** `cs.GT` · **Submitted:** 2026-01-16
+- **Authors:** Eilam Shapira, Roi Reichart, Moshe Tennenholtz
+- **abs:** https://arxiv.org/abs/2601.11496 · **PDF:** https://arxiv.org/pdf/2601.11496v1
+
+## 21224. BoxMind: Closed-loop AI strategy optimization for elite boxing validated in the 2024 Olympics
+
+- **ID:** `2601.11492` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Kaiwen Wang, Kaili Zheng, Rongrong Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.11492 · **PDF:** https://arxiv.org/pdf/2601.11492v2
+
+## 21225. Extractive summarization on a CMOS Ising machine
+
+- **ID:** `2601.11491` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Ziqing Zeng, Abhimanyu Kumar, Ahmet Efe, et al.
+- **abs:** https://arxiv.org/abs/2601.11491 · **PDF:** https://arxiv.org/pdf/2601.11491v2
+
+## 21226. CTest-Metric: A Unified Framework to Assess Clinical Validity of Metrics for CT Report Generation
+
+- **ID:** `2601.11488` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Vanshali Sharma, Andrea Mia Bejar, Gorkem Durak, et al.
+- **abs:** https://arxiv.org/abs/2601.11488 · **PDF:** https://arxiv.org/pdf/2601.11488v1
+
+## 21227. Health Facility Location in Ethiopia: Leveraging LLMs to Integrate Expert Knowledge into Algorithmic Planning
+
+- **ID:** `2601.11479` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Yohai Trabelsi, Guojun Xiong, Fentabil Getnet, et al.
+- **abs:** https://arxiv.org/abs/2601.11479 · **PDF:** https://arxiv.org/pdf/2601.11479v2
+
+## 21228. Temporal Complexity and Self-Organization in an Exponential Dense Associative Memory Model
+
+- **ID:** `2601.11478` · **primary:** `nlin.AO` · **Submitted:** 2026-01-16
+- **Authors:** Marco Cafiso, Paolo Paradisi
+- **abs:** https://arxiv.org/abs/2601.11478 · **PDF:** https://arxiv.org/pdf/2601.11478v1
+
+## 21229. A Probabilistic Approach to Trajectory-Based Optimal Experimental Design
+
+- **ID:** `2601.11473` · **primary:** `math.OC` · **Submitted:** 2026-01-16
+- **Authors:** Ahmed Attia
+- **abs:** https://arxiv.org/abs/2601.11473 · **PDF:** https://arxiv.org/pdf/2601.11473v1
+
+## 21230. Low-Rank Key Value Attention
+
+- **ID:** `2601.11471` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** James O'Neill, Robert Clancy, Mariia Matskevichus, et al.
+- **abs:** https://arxiv.org/abs/2601.11471 · **PDF:** https://arxiv.org/pdf/2601.11471v2
+
+## 21231. Exploring LLM Features in Predictive Process Monitoring for Small-Scale Event-Logs
+
+- **ID:** `2601.11468` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Alessandro Padella, Massimiliano de Leoni, Marlon Dumas
+- **abs:** https://arxiv.org/abs/2601.11468 · **PDF:** https://arxiv.org/pdf/2601.11468v1
+
+## 21232. MHA2MLA-VLM: Enabling DeepSeek's Economical Multi-Head Latent Attention across Vision-Language Models
+
+- **ID:** `2601.11464` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Xiaoran Fan, Zhichao Sun, Tao Ji, et al.
+- **abs:** https://arxiv.org/abs/2601.11464 · **PDF:** https://arxiv.org/pdf/2601.11464v1
+
+## 21233. Learning Semantic-Geometric Task Graph-Representations from Human Demonstrations
+
+- **ID:** `2601.11460` · **primary:** `cs.RO` · **Submitted:** 2026-01-16
+- **Authors:** Franziska Herbert, Vignesh Prasad, Han Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.11460 · **PDF:** https://arxiv.org/pdf/2601.11460v1
+
+## 21234. Interactive Narrative Analytics: Bridging Computational Narrative Extraction and Human Sensemaking
+
+- **ID:** `2601.11459` · **primary:** `cs.HC` · **Submitted:** 2026-01-16
+- **Authors:** Brian Keith
+- **abs:** https://arxiv.org/abs/2601.11459 · **PDF:** https://arxiv.org/pdf/2601.11459v1
+
+## 21235. Continuous-Flow Data-Rate-Aware CNN Inference on FPGA
+
+- **ID:** `2601.19940` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Tobias Habermann, Michael Mecik, Zhenyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19940 · **PDF:** https://arxiv.org/pdf/2601.19940v2
+
+## 21236. PRISM-CAFO: Prior-conditioned Remote-sensing Infrastructure Segmentation and Mapping for CAFOs
+
+- **ID:** `2601.11451` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Oishee Bintey Hoque, Nibir Chandra Mandal, Kyle Luong, et al.
+- **abs:** https://arxiv.org/abs/2601.11451 · **PDF:** https://arxiv.org/pdf/2601.11451v2
+
+## 21237. IMS: Intelligent Hardware Monitoring System for Secure SoCs
+
+- **ID:** `2601.11447` · **primary:** `cs.CR` · **Submitted:** 2026-01-16
+- **Authors:** Wadid Foudhaili, Aykut Rencber, Anouar Nechi, et al.
+- **abs:** https://arxiv.org/abs/2601.11447 · **PDF:** https://arxiv.org/pdf/2601.11447v1
+
+## 21238. When Are Two Scores Better Than One? Investigating Ensembles of Diffusion Models
+
+- **ID:** `2601.11444` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Raphaël Razafindralambo, Rémy Sun, Frédéric Precioso, et al.
+- **abs:** https://arxiv.org/abs/2601.11444 · **PDF:** https://arxiv.org/pdf/2601.11444v2
+
+## 21239. Predict the Retrieval! Test time adaptation for Retrieval Augmented Generation
+
+- **ID:** `2601.11443` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Xin Sun, Zhongqi Chen, Qiang Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.11443 · **PDF:** https://arxiv.org/pdf/2601.11443v1
+
+## 21240. Map2Thought: Explicit 3D Spatial Reasoning via Metric Cognitive Maps
+
+- **ID:** `2601.11442` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Xiangjun Gao, Zhensong Zhang, Dave Zhenyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.11442 · **PDF:** https://arxiv.org/pdf/2601.11442v1
+
+## 21241. Ensuring Computer Science Learning in the AI Era: Open Generative AI Policies and Assignment-Driven Written Quizzes
+
+- **ID:** `2601.17024` · **primary:** `cs.CY` · **Submitted:** 2026-01-16
+- **Authors:** Chan-Jin Chung
+- **abs:** https://arxiv.org/abs/2601.17024 · **PDF:** https://arxiv.org/pdf/2601.17024v1
+
+## 21242. Hierarchical Orthogonal Residual Spread for Precise Massive Editing in Large Language Models
+
+- **ID:** `2601.11441` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Xiaojie Gu, Guangxu Chen, Yuheng Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.11441 · **PDF:** https://arxiv.org/pdf/2601.11441v1
+
+## 21243. GenDA: Generative Data Assimilation on Complex Urban Areas via Classifier-Free Diffusion Guidance
+
+- **ID:** `2601.11440` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Francisco Giral, Álvaro Manzano, Ignacio Gómez, et al.
+- **abs:** https://arxiv.org/abs/2601.11440 · **PDF:** https://arxiv.org/pdf/2601.11440v2
+
+## 21244. Near-Optimal Decentralized Stochastic Nonconvex Optimization with Heavy-Tailed Noise
+
+- **ID:** `2601.11435` · **primary:** `math.OC` · **Submitted:** 2026-01-16
+- **Authors:** Menglian Wang, Zhuanghua Liu, Luo Luo
+- **abs:** https://arxiv.org/abs/2601.11435 · **PDF:** https://arxiv.org/pdf/2601.11435v1
+
+## 21245. Inter-patient ECG Arrhythmia Classification with LGNs and LUTNs
+
+- **ID:** `2601.11433` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Wout Mommen, Lars Keuninckx, Paul Detterer, et al.
+- **abs:** https://arxiv.org/abs/2601.11433 · **PDF:** https://arxiv.org/pdf/2601.11433v1
+
+## 21246. The unreasonable effectiveness of pattern matching
+
+- **ID:** `2601.11432` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Gary Lupyan, Blaise Agüera y Arcas
+- **abs:** https://arxiv.org/abs/2601.11432 · **PDF:** https://arxiv.org/pdf/2601.11432v2
+
+## 21247. Relational Linearity is a Predictor of Hallucinations
+
+- **ID:** `2601.11429` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Yuetian Lu, Yihong Liu, Hinrich Schütze
+- **abs:** https://arxiv.org/abs/2601.11429 · **PDF:** https://arxiv.org/pdf/2601.11429v1
+
+## 21248. Forcing and Diagnosing Failure Modes of Fourier Neural Operators Across Diverse PDE Families
+
+- **ID:** `2601.11428` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Lennon Shikhman
+- **abs:** https://arxiv.org/abs/2601.11428 · **PDF:** https://arxiv.org/pdf/2601.11428v3
+
+## 21249. Isotropy-Optimized Contrastive Learning for Semantic Course Recommendation
+
+- **ID:** `2601.11427` · **primary:** `cs.IR` · **Submitted:** 2026-01-16
+- **Authors:** Ali Khreis, Anthony Nasr, Yusuf Hilal
+- **abs:** https://arxiv.org/abs/2601.11427 · **PDF:** https://arxiv.org/pdf/2601.11427v1
+
+## 21250. PubMed-OCR: PMC Open Access OCR Annotations
+
+- **ID:** `2601.11425` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Hunter Heidenreich, Yosheb Getachew, Olivia Dinica, et al.
+- **abs:** https://arxiv.org/abs/2601.11425 · **PDF:** https://arxiv.org/pdf/2601.11425v1
+
+## 21251. The Great March 100: 100 Detail-oriented Tasks for Evaluating Embodied AI Agents
+
+- **ID:** `2601.11421` · **primary:** `cs.RO` · **Submitted:** 2026-01-16
+- **Authors:** Ziyu Wang, Chenyuan Liu, Yushun Xiang, et al.
+- **abs:** https://arxiv.org/abs/2601.11421 · **PDF:** https://arxiv.org/pdf/2601.11421v1
+
+## 21252. Statistical Robustness of Interval CVaR Based Regression Models under Perturbation and Contamination
+
+- **ID:** `2601.11420` · **primary:** `math.OC` · **Submitted:** 2026-01-16
+- **Authors:** Yulei You, Junyi Liu
+- **abs:** https://arxiv.org/abs/2601.11420 · **PDF:** https://arxiv.org/pdf/2601.11420v1
+
+## 21253. Beyond Static Question Banks: Dynamic Knowledge Expansion via LLM-Automated Graph Construction and Adaptive Generation
+
+- **ID:** `2602.00020` · **primary:** `cs.CY` · **Submitted:** 2026-01-16
+- **Authors:** Yingquan Wang, Tianyu Wei, Qinsi Li, et al.
+- **abs:** https://arxiv.org/abs/2602.00020 · **PDF:** https://arxiv.org/pdf/2602.00020v2
+
+## 21254. Zero-Shot Detection of Elastic Transient Morphology Across Physical Systems
+
+- **ID:** `2601.11415` · **primary:** `astro-ph.IM` · **Submitted:** 2026-01-16
+- **Authors:** Jose Sánchez Andreu
+- **abs:** https://arxiv.org/abs/2601.11415 · **PDF:** https://arxiv.org/pdf/2601.11415v1
+
+## 21255. New Adaptive Mechanism for Large Neighborhood Search using Dual Actor-Critic
+
+- **ID:** `2601.11414` · **primary:** `cs.GT` · **Submitted:** 2026-01-16
+- **Authors:** Shaohua Yu, Wenhao Mao, Zigao Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.11414 · **PDF:** https://arxiv.org/pdf/2601.11414v1
+
+## 21256. oculomix: Hierarchical Sampling for Retinal-Based Systemic Disease Prediction
+
+- **ID:** `2601.19939` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Hyunmin Kim, Yukun Zhou, Rahul A. Jonas, et al.
+- **abs:** https://arxiv.org/abs/2601.19939 · **PDF:** https://arxiv.org/pdf/2601.19939v1
+
+## 21257. Topology-Guaranteed Image Segmentation: Enforcing Connectivity, Genus, and Width Constraints
+
+- **ID:** `2601.11409` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Wenxiao Li, Xue-Cheng Tai, Jun Liu
+- **abs:** https://arxiv.org/abs/2601.11409 · **PDF:** https://arxiv.org/pdf/2601.11409v1
+
+## 21258. Anisotropic Tensor Deconvolution of Hyperspectral Images
+
+- **ID:** `2601.11694` · **primary:** `eess.IV` · **Submitted:** 2026-01-16
+- **Authors:** Xinjue Wang, Xiuheng Wang, Esa Ollila, et al.
+- **abs:** https://arxiv.org/abs/2601.11694 · **PDF:** https://arxiv.org/pdf/2601.11694v1
+
+## 21259. Factored Value Functions for Graph-Based Multi-Agent Reinforcement Learning
+
+- **ID:** `2601.11401` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Ahmed Rashwan, Keith Briggs, Chris Budd, et al.
+- **abs:** https://arxiv.org/abs/2601.11401 · **PDF:** https://arxiv.org/pdf/2601.11401v1
+
+## 21260. Wetland mapping from sparse annotations with satellite image time series and temporal-aware segment anything model
+
+- **ID:** `2601.11400` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Shuai Yuan, Tianwu Lin, Shuang Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.11400 · **PDF:** https://arxiv.org/pdf/2601.11400v1
+
+## 21261. Latent Space Inference via Paired Autoencoders
+
+- **ID:** `2601.11397` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Emma Hart, Bas Peters, Julianne Chung, et al.
+- **abs:** https://arxiv.org/abs/2601.11397 · **PDF:** https://arxiv.org/pdf/2601.11397v1
+
+## 21262. Hyperparameter Optimization of Constraint Programming Solvers
+
+- **ID:** `2601.11389` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Hedieh Haddad, Thibault Falque, Pierre Talbot, et al.
+- **abs:** https://arxiv.org/abs/2601.11389 · **PDF:** https://arxiv.org/pdf/2601.11389v1
+
+## 21263. Evaluating LLM Behavior in Hiring: Implicit Weights, Fairness Across Groups, and Alignment with Human Preferences
+
+- **ID:** `2601.11379` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Morgane Hoffmann, Emma Jouffroy, Warren Jouanneau, et al.
+- **abs:** https://arxiv.org/abs/2601.11379 · **PDF:** https://arxiv.org/pdf/2601.11379v1
+
+## 21264. Explainable histomorphology-based survival prediction of glioblastoma, IDH-wildtype
+
+- **ID:** `2601.11691` · **primary:** `eess.IV` · **Submitted:** 2026-01-16
+- **Authors:** Jan-Philipp Redlich, Friedrich Feuerhake, Stefan Nikolin, et al.
+- **abs:** https://arxiv.org/abs/2601.11691 · **PDF:** https://arxiv.org/pdf/2601.11691v3
+
+## 21265. Reward Modeling for Scientific Writing Evaluation
+
+- **ID:** `2601.11374` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Furkan Şahinuç, Subhabrata Dutta, Iryna Gurevych
+- **abs:** https://arxiv.org/abs/2601.11374 · **PDF:** https://arxiv.org/pdf/2601.11374v1
+
+## 21266. Institutional AI: Governing LLM Collusion in Multi-Agent Cournot Markets via Public Governance Graphs
+
+- **ID:** `2601.11369` · **primary:** `cs.GT` · **Submitted:** 2026-01-16
+- **Authors:** Marcantonio Bracale Syrnikov, Federico Pierucci, Marcello Galisai, et al.
+- **abs:** https://arxiv.org/abs/2601.11369 · **PDF:** https://arxiv.org/pdf/2601.11369v2
+
+## 21267. Think-Clip-Sample: Slow-Fast Frame Selection for Video Understanding
+
+- **ID:** `2601.11359` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Wenhui Tan, Ruihua Song, Jiaze Li, et al.
+- **abs:** https://arxiv.org/abs/2601.11359 · **PDF:** https://arxiv.org/pdf/2601.11359v1
+
+## 21268. AstroReason-Bench: Evaluating Unified Agentic Planning across Heterogeneous Space Planning Problems
+
+- **ID:** `2601.11354` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Weiyi Wang, Xinchi Chen, Jingjing Gong, et al.
+- **abs:** https://arxiv.org/abs/2601.11354 · **PDF:** https://arxiv.org/pdf/2601.11354v1
+
+## 21269. Offline Reinforcement-Learning-Based Power Control for Application-Agnostic Energy Efficiency
+
+- **ID:** `2601.11352` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Akhilesh Raj, Swann Perarnau, Aniruddha Gokhale, et al.
+- **abs:** https://arxiv.org/abs/2601.11352 · **PDF:** https://arxiv.org/pdf/2601.11352v1
+
+## 21270. FEATHer: Fourier-Efficient Adaptive Temporal Hierarchy Forecaster for Time-Series Forecasting
+
+- **ID:** `2601.11350` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Jaehoon Lee, Seungwoo Lee, Younghwi Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.11350 · **PDF:** https://arxiv.org/pdf/2601.11350v1
+
+## 21271. How Much Would a Clinician Edit This Draft? Evaluating LLM Alignment for Patient Message Response Drafting
+
+- **ID:** `2601.11344` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Parker Seegmiller, Joseph Gatto, Sarah E. Greer, et al.
+- **abs:** https://arxiv.org/abs/2601.11344 · **PDF:** https://arxiv.org/pdf/2601.11344v1
+
+## 21272. Unlocking the Potentials of Retrieval-Augmented Generation for Diffusion Language Models
+
+- **ID:** `2601.11342` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Chuanyue Yu, Jiahui Wang, Yuhan Li, et al.
+- **abs:** https://arxiv.org/abs/2601.11342 · **PDF:** https://arxiv.org/pdf/2601.11342v1
+
+## 21273. Neural Chain-of-Thought Search: Searching the Optimal Reasoning Path to Enhance Large Language Models
+
+- **ID:** `2601.11340` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Guoming Ling, Zhongzhan Huang, Yupei Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.11340 · **PDF:** https://arxiv.org/pdf/2601.11340v1
+
+## 21274. Beer-Lambert Autoencoder for Unsupervised Stain Representation Learning and Deconvolution in Multi-immunohistochemical Brightfield Histology Images
+
+- **ID:** `2601.11336` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Mark Eastwood, Thomas McKee, Zedong Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.11336 · **PDF:** https://arxiv.org/pdf/2601.11336v1
+
+## 21275. Information Theoretic Perspective on Representation Learning
+
+- **ID:** `2601.11334` · **primary:** `cs.IT` · **Submitted:** 2026-01-16
+- **Authors:** Deborah Pereg
+- **abs:** https://arxiv.org/abs/2601.11334 · **PDF:** https://arxiv.org/pdf/2601.11334v1
+
+## 21276. Idea First, Code Later: Disentangling Problem Solving from Code Generation in Evaluating LLMs for Competitive Programming
+
+- **ID:** `2601.11332` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Sama Hadhoud, Alaa Elsetohy, Frederikus Hudi, et al.
+- **abs:** https://arxiv.org/abs/2601.11332 · **PDF:** https://arxiv.org/pdf/2601.11332v1
+
+## 21277. F-Actor: Controllable Conversational Behaviour in Full-Duplex Models
+
+- **ID:** `2601.11329` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Maike Züfle, Ondrej Klejch, Nicholas Sanders, et al.
+- **abs:** https://arxiv.org/abs/2601.11329 · **PDF:** https://arxiv.org/pdf/2601.11329v2
+
+## 21278. GENPACK: KPI-Guided Multi-Criteria Genetic Algorithm for Industrial 3D Bin Packing
+
+- **ID:** `2601.11325` · **primary:** `cs.NE` · **Submitted:** 2026-01-16
+- **Authors:** Dheeraj Poolavaram, Carsten Markgraf, Sebastian Dorn
+- **abs:** https://arxiv.org/abs/2601.11325 · **PDF:** https://arxiv.org/pdf/2601.11325v3
+
+## 21279. Membership Inference on LLMs in the Wild
+
+- **ID:** `2601.11314` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Jiatong Yi, Yanyang Li
+- **abs:** https://arxiv.org/abs/2601.11314 · **PDF:** https://arxiv.org/pdf/2601.11314v1
+
+## 21280. FORESTLLM: Large Language Models Make Random Forest Great on Few-shot Tabular Learning
+
+- **ID:** `2601.11311` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Zhihan Yang, Jiaqi Wei, Xiang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.11311 · **PDF:** https://arxiv.org/pdf/2601.11311v1
+
+## 21281. AI-based System for Transforming text and sound to Educational Videos
+
+- **ID:** `2601.17022` · **primary:** `cs.MM` · **Submitted:** 2026-01-16
+- **Authors:** M. E. ElAlami, S. M. Khater, M. El. R. Rehan
+- **abs:** https://arxiv.org/abs/2601.17022 · **PDF:** https://arxiv.org/pdf/2601.17022v1
+
+## 21282. One LLM to Train Them All: Multi-Task Learning Framework for Fact-Checking
+
+- **ID:** `2601.11293` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Malin Astrid Larsson, Harald Fosen Grunnaleite, Vinay Setty
+- **abs:** https://arxiv.org/abs/2601.11293 · **PDF:** https://arxiv.org/pdf/2601.11293v1
+
+## 21283. XChoice: Explainable Evaluation of AI-Human Alignment in LLM-based Constrained Choice Decision Making
+
+- **ID:** `2601.11286` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Weihong Qi, Fan Huang, Rasika Muralidharan, et al.
+- **abs:** https://arxiv.org/abs/2601.11286 · **PDF:** https://arxiv.org/pdf/2601.11286v1
+
+## 21284. Metabolomic Biomarker Discovery for ADHD Diagnosis Using Interpretable Machine Learning
+
+- **ID:** `2601.11283` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Nabil Belacel, Mohamed Rachid Boulassel
+- **abs:** https://arxiv.org/abs/2601.11283 · **PDF:** https://arxiv.org/pdf/2601.11283v2
+
+## 21285. From SERPs to Sound: How Search Engine Result Pages and AI-generated Podcasts Interact to Influence User Attitudes on Controversial Topics
+
+- **ID:** `2601.11282` · **primary:** `cs.IR` · **Submitted:** 2026-01-16
+- **Authors:** Junjie Wang, Gaole He, Alisa Rieger, et al.
+- **abs:** https://arxiv.org/abs/2601.11282 · **PDF:** https://arxiv.org/pdf/2601.11282v1
+
+## 21286. X-Distill: Cross-Architecture Vision Distillation for Visuomotor Learning
+
+- **ID:** `2601.11269` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Maanping Shao, Feihong Zhang, Gu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.11269 · **PDF:** https://arxiv.org/pdf/2601.11269v1
+
+## 21287. Sample-Near-Optimal Agnostic Boosting with Improved Running Time
+
+- **ID:** `2601.11265` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Arthur da Cunha, Mikael Møller Høgsgaard, Andrea Paudice
+- **abs:** https://arxiv.org/abs/2601.11265 · **PDF:** https://arxiv.org/pdf/2601.11265v3
+
+## 21288. Scalable Music Cover Retrieval Using Lyrics-Aligned Audio Embeddings
+
+- **ID:** `2601.11262` · **primary:** `cs.SD` · **Submitted:** 2026-01-16
+- **Authors:** Joanne Affolter, Benjamin Martin, Elena V. Epure, et al.
+- **abs:** https://arxiv.org/abs/2601.11262 · **PDF:** https://arxiv.org/pdf/2601.11262v1
+
+## 21289. Effects of Introducing Synaptic Scaling on Spiking Neural Network Learning
+
+- **ID:** `2601.11261` · **primary:** `cs.NE` · **Submitted:** 2026-01-16
+- **Authors:** Shinnosuke Touda, Hirotsugu Okuno
+- **abs:** https://arxiv.org/abs/2601.11261 · **PDF:** https://arxiv.org/pdf/2601.11261v1
+
+## 21290. Latent Dynamics Graph Convolutional Networks for model order reduction of parameterized time-dependent PDEs
+
+- **ID:** `2601.11259` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Lorenzo Tomada, Federico Pichi, Gianluigi Rozza
+- **abs:** https://arxiv.org/abs/2601.11259 · **PDF:** https://arxiv.org/pdf/2601.11259v1
+
+## 21291. Knowledge is Not Enough: Injecting RL Skills for Continual Adaptation
+
+- **ID:** `2601.11258` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Pingzhi Tang, Yiding Wang, Muhan Zhang
+- **abs:** https://arxiv.org/abs/2601.11258 · **PDF:** https://arxiv.org/pdf/2601.11258v1
+
+## 21292. Reasoning in Trees: Improving Retrieval-Augmented Generation for Multi-Hop Question Answering
+
+- **ID:** `2601.11255` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Yuling Shi, Maolin Sun, Zijun Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.11255 · **PDF:** https://arxiv.org/pdf/2601.11255v1
+
+## 21293. Beyond Model Scaling: Test-Time Intervention for Efficient Deep Reasoning
+
+- **ID:** `2601.11252` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Qianyue Wang, Jinwu Hu, Yufeng Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.11252 · **PDF:** https://arxiv.org/pdf/2601.11252v1
+
+## 21294. Quantifying and Attributing Polarization to Annotator Groups
+
+- **ID:** `2602.06055` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Dimitris Tsirmpas, John Pavlopoulos
+- **abs:** https://arxiv.org/abs/2602.06055 · **PDF:** https://arxiv.org/pdf/2602.06055v1
+
+## 21295. How DDAIR you? Disambiguated Data Augmentation for Intent Recognition
+
+- **ID:** `2601.11234` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Galo Castillo-López, Alexis Lombard, Nasredine Semmar, et al.
+- **abs:** https://arxiv.org/abs/2601.11234 · **PDF:** https://arxiv.org/pdf/2601.11234v1
+
+## 21296. FactCorrector: A Graph-Inspired Approach to Long-Form Factuality Correction of Large Language Models
+
+- **ID:** `2601.11232` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Javier Carnerero-Cano, Massimiliano Pronesti, Radu Marinescu, et al.
+- **abs:** https://arxiv.org/abs/2601.11232 · **PDF:** https://arxiv.org/pdf/2601.11232v1
+
+## 21297. Language of Thought Shapes Output Diversity in Large Language Models
+
+- **ID:** `2601.11227` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Shaoyang Xu, Wenxuan Zhang
+- **abs:** https://arxiv.org/abs/2601.11227 · **PDF:** https://arxiv.org/pdf/2601.11227v1
+
+## 21298. Operator learning on domain boundary through combining fundamental solution-based artificial data and boundary integral techniques
+
+- **ID:** `2601.11222` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Haochen Wu, Heng Wu, Benzhuo Lu
+- **abs:** https://arxiv.org/abs/2601.11222 · **PDF:** https://arxiv.org/pdf/2601.11222v1
+
+## 21299. MultiCaption: Detecting disinformation using multilingual visual claims
+
+- **ID:** `2601.11220` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Rafael Martins Frade, Rrubaa Panchendrarajan, Arkaitz Zubiaga
+- **abs:** https://arxiv.org/abs/2601.11220 · **PDF:** https://arxiv.org/pdf/2601.11220v1
+
+## 21300. SDFLoRA: Selective Decoupled Federated LoRA for Privacy-preserving Fine-tuning with Heterogeneous Clients
+
+- **ID:** `2601.11219` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Zhikang Shen, Jianrong Lu, Haiyuan Wan, et al.
+- **abs:** https://arxiv.org/abs/2601.11219 · **PDF:** https://arxiv.org/pdf/2601.11219v2
+
+## 21301. SpecMap: Hierarchical LLM Agent for Datasheet-to-Code Traceability Link Recovery in Systems Engineering
+
+- **ID:** `2601.11688` · **primary:** `cs.SE` · **Submitted:** 2026-01-16
+- **Authors:** Vedant Nipane, Pulkit Agrawal, Amit Singh
+- **abs:** https://arxiv.org/abs/2601.11688 · **PDF:** https://arxiv.org/pdf/2601.11688v1
+
+## 21302. Model-free policy gradient for discrete-time mean-field control
+
+- **ID:** `2601.11217` · **primary:** `math.OC` · **Submitted:** 2026-01-16
+- **Authors:** Matthieu Meunier, Huyên Pham, Christoph Reisinger
+- **abs:** https://arxiv.org/abs/2601.11217 · **PDF:** https://arxiv.org/pdf/2601.11217v2
+
+## 21303. T$^\star$: Progressive Block Scaling for Masked Diffusion Language Models Through Trajectory Aware Reinforcement Learning
+
+- **ID:** `2601.11214` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Hanchen Xia, Baoyou Chen, Yutang Ge, et al.
+- **abs:** https://arxiv.org/abs/2601.11214 · **PDF:** https://arxiv.org/pdf/2601.11214v4
+
+## 21304. LoRA as Oracle
+
+- **ID:** `2601.11207` · **primary:** `cs.CR` · **Submitted:** 2026-01-16
+- **Authors:** Marco Arazzi, Antonino Nocera
+- **abs:** https://arxiv.org/abs/2601.11207 · **PDF:** https://arxiv.org/pdf/2601.11207v1
+
+## 21305. Semantic Caching and Intent-Driven Context Optimization for Multi-Agent Natural Language to Code Systems
+
+- **ID:** `2601.11687` · **primary:** `cs.SE` · **Submitted:** 2026-01-16
+- **Authors:** Harmohit Singh
+- **abs:** https://arxiv.org/abs/2601.11687 · **PDF:** https://arxiv.org/pdf/2601.11687v1
+
+## 21306. Epistemic Control and the Normativity of Machine Learning-Based Science
+
+- **ID:** `2601.11202` · **primary:** `cs.CY` · **Submitted:** 2026-01-16
+- **Authors:** Emanuele Ratti
+- **abs:** https://arxiv.org/abs/2601.11202 · **PDF:** https://arxiv.org/pdf/2601.11202v1
+
+## 21307. FAQ: Mitigating Quantization Error via Regenerating Calibration Data with Family-Aware Quantization
+
+- **ID:** `2601.11200` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Haiyang Xiao, Weiqing Li, Jinyue Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.11200 · **PDF:** https://arxiv.org/pdf/2601.11200v1
+
+## 21308. SD-RAG: A Prompt-Injection-Resilient Framework for Selective Disclosure in Retrieval-Augmented Generation
+
+- **ID:** `2601.11199` · **primary:** `cs.CR` · **Submitted:** 2026-01-16
+- **Authors:** Aiman Al Masoud, Marco Arazzi, Antonino Nocera
+- **abs:** https://arxiv.org/abs/2601.11199 · **PDF:** https://arxiv.org/pdf/2601.11199v1
+
+## 21309. DecHW: Heterogeneous Decentralized Federated Learning Exploiting Second-Order Information
+
+- **ID:** `2601.19938` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Adnan Ahmad, Chiara Boldrini, Lorenzo Valerio, et al.
+- **abs:** https://arxiv.org/abs/2601.19938 · **PDF:** https://arxiv.org/pdf/2601.19938v1
+
+## 21310. Artificial Intelligence and the US Economy: An Accounting Perspective on Investment and Production
+
+- **ID:** `2601.11196` · **primary:** `econ.GN` · **Submitted:** 2026-01-16
+- **Authors:** Luisa Carpinelli, Filippo Natoli, Marco Taboga
+- **abs:** https://arxiv.org/abs/2601.11196 · **PDF:** https://arxiv.org/pdf/2601.11196v1
+
+## 21311. DOREMI: Optimizing Long Tail Predictions in Document-Level Relation Extraction
+
+- **ID:** `2601.11190` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Laura Menotti, Stefano Marchesin, Gianmaria Silvello
+- **abs:** https://arxiv.org/abs/2601.11190 · **PDF:** https://arxiv.org/pdf/2601.11190v1
+
+## 21312. Policy-Based Deep Reinforcement Learning Hyperheuristics for Job-Shop Scheduling Problems
+
+- **ID:** `2601.11189` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Sofiene Lassoued, Asrat Gobachew, Stefan Lier, et al.
+- **abs:** https://arxiv.org/abs/2601.11189 · **PDF:** https://arxiv.org/pdf/2601.11189v1
+
+## 21313. TimeMar: Multi-Scale Autoregressive Modeling for Unconditional Time Series Generation
+
+- **ID:** `2601.11184` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Xiangyu Xu, Qingsong Zhong, Jilin Hu
+- **abs:** https://arxiv.org/abs/2601.11184 · **PDF:** https://arxiv.org/pdf/2601.11184v1
+
+## 21314. TANDEM: Temporal-Aware Neural Detection for Multimodal Hate Speech
+
+- **ID:** `2601.11178` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Girish A. Koushik, Helen Treharne, Diptesh Kanojia
+- **abs:** https://arxiv.org/abs/2601.11178 · **PDF:** https://arxiv.org/pdf/2601.11178v1
+
+## 21315. Proof of Concept: Multi-Target Wildfire Risk Prediction and Large Language Model Synthesis
+
+- **ID:** `2601.11686` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Nicolas Caron, Christophe Guyeux, Hassan Noura, et al.
+- **abs:** https://arxiv.org/abs/2601.11686 · **PDF:** https://arxiv.org/pdf/2601.11686v1
+
+## 21316. The Growing Gains and Pains of Iterative Web Corpora Crawling: Insights from South Slavic CLASSLA-web 2.0 Corpora
+
+- **ID:** `2601.11170` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Taja Kuzman Pungeršek, Peter Rupnik, Vít Suchomel, et al.
+- **abs:** https://arxiv.org/abs/2601.11170 · **PDF:** https://arxiv.org/pdf/2601.11170v2
+
+## 21317. ES4R: Speech Encoding Based on Prepositive Affective Modeling for Empathetic Response Generation
+
+- **ID:** `2601.16225` · **primary:** `eess.AS` · **Submitted:** 2026-01-16
+- **Authors:** Zhuoyue Gao, Xiaohui Wang, Xiaocui Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.16225 · **PDF:** https://arxiv.org/pdf/2601.16225v1
+
+## 21318. LSTM VS. Feed-Forward Autoencoders for Unsupervised Fault Detection in Hydraulic Pumps
+
+- **ID:** `2601.11163` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** P. Sánchez, K. Reyes, B. Radu, et al.
+- **abs:** https://arxiv.org/abs/2601.11163 · **PDF:** https://arxiv.org/pdf/2601.11163v1
+
+## 21319. GMM-COMET: Continual Source-Free Universal Domain Adaptation via a Mean Teacher and Gaussian Mixture Model-Based Pseudo-Labeling
+
+- **ID:** `2601.11161` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Pascal Schlachter, Bin Yang
+- **abs:** https://arxiv.org/abs/2601.11161 · **PDF:** https://arxiv.org/pdf/2601.11161v1
+
+## 21320. Clustering High-dimensional Data: Balancing Abstraction and Representation Tutorial at AAAI 2026
+
+- **ID:** `2601.11160` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Claudia Plant, Lena G. M. Bauer, Christian Böhm
+- **abs:** https://arxiv.org/abs/2601.11160 · **PDF:** https://arxiv.org/pdf/2601.11160v1
+
+## 21321. Theoretically and Practically Efficient Resistance Distance Computation on Large Graphs
+
+- **ID:** `2601.11159` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Yichun Yang, Longlong Lin, Rong-Hua Li, et al.
+- **abs:** https://arxiv.org/abs/2601.11159 · **PDF:** https://arxiv.org/pdf/2601.11159v1
+
+## 21322. Large Language Models as Simulative Agents for Neurodivergent Adult Psychometric Profiles
+
+- **ID:** `2601.15319` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-16
+- **Authors:** Francesco Chiappone, Davide Marocco, Nicola Milano
+- **abs:** https://arxiv.org/abs/2601.15319 · **PDF:** https://arxiv.org/pdf/2601.15319v1
+
+## 21323. Assesing the Viability of Unsupervised Learning with Autoencoders for Predictive Maintenance in Helicopter Engines
+
+- **ID:** `2601.11154` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** P. Sánchez, K. Reyes, B. Radu, et al.
+- **abs:** https://arxiv.org/abs/2601.11154 · **PDF:** https://arxiv.org/pdf/2601.11154v1
+
+## 21324. Cross-Modal Attention Network with Dual Graph Learning in Multimodal Recommendation
+
+- **ID:** `2601.11151` · **primary:** `cs.IR` · **Submitted:** 2026-01-16
+- **Authors:** Ji Dai, Quan Fang, Jun Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.11151 · **PDF:** https://arxiv.org/pdf/2601.11151v1
+
+## 21325. Towards Efficient Image Deblurring for Edge Deployment
+
+- **ID:** `2601.11685` · **primary:** `eess.IV` · **Submitted:** 2026-01-16
+- **Authors:** Srinivas Miriyala, Sowmya Vajrala, Sravanth Kodavanti
+- **abs:** https://arxiv.org/abs/2601.11685 · **PDF:** https://arxiv.org/pdf/2601.11685v1
+
+## 21326. Do We Always Need Query-Level Workflows? Rethinking Agentic Workflow Generation for Multi-Agent Systems
+
+- **ID:** `2601.11147` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Zixu Wang, Bingbing Xu, Yige Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.11147 · **PDF:** https://arxiv.org/pdf/2601.11147v1
+
+## 21327. Deep GraphRAG: A Balanced Approach to Hierarchical Retrieval and Adaptive Integration
+
+- **ID:** `2601.11144` · **primary:** `cs.IR` · **Submitted:** 2026-01-16
+- **Authors:** Yuejie Li, Ke Yang, Tao Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.11144 · **PDF:** https://arxiv.org/pdf/2601.11144v3
+
+## 21328. Learning Quadrupedal Locomotion for a Heavy Hydraulic Robot Using an Actuator Model
+
+- **ID:** `2601.11143` · **primary:** `cs.RO` · **Submitted:** 2026-01-16
+- **Authors:** Minho Lee, Hyeonseok Kim, Jin Tak Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.11143 · **PDF:** https://arxiv.org/pdf/2601.11143v1
+
+## 21329. FlashLabs Chroma 1.0: A Real-Time End-to-End Spoken Dialogue Model with Personalized Voice Cloning
+
+- **ID:** `2601.11141` · **primary:** `cs.SD` · **Submitted:** 2026-01-16
+- **Authors:** Tanyu Chen, Tairan Chen, Kai Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.11141 · **PDF:** https://arxiv.org/pdf/2601.11141v1
+
+## 21330. STAER: Temporal Aligned Rehearsal for Continual Spiking Neural Network
+
+- **ID:** `2601.20870` · **primary:** `cs.NE` · **Submitted:** 2026-01-16
+- **Authors:** Matteo Gianferrari, Omayma Moussadek, Riccardo Salami, et al.
+- **abs:** https://arxiv.org/abs/2601.20870 · **PDF:** https://arxiv.org/pdf/2601.20870v1
+
+## 21331. Context-aware Graph Causality Inference for Few-Shot Molecular Property Prediction
+
+- **ID:** `2601.11135` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Van Thuy Hoang, O-Joun Lee
+- **abs:** https://arxiv.org/abs/2601.11135 · **PDF:** https://arxiv.org/pdf/2601.11135v1
+
+## 21332. FSL-BDP: Federated Survival Learning with Bayesian Differential Privacy for Credit Risk Modeling
+
+- **ID:** `2601.11134` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Sultan Amed, Tanmay Sen, Sayantan Banerjee
+- **abs:** https://arxiv.org/abs/2601.11134 · **PDF:** https://arxiv.org/pdf/2601.11134v1
+
+## 21333. Mobile-friendly Image de-noising: Hardware Conscious Optimization for Edge Application
+
+- **ID:** `2601.11684` · **primary:** `eess.IV` · **Submitted:** 2026-01-16
+- **Authors:** Srinivas Miriyala, Sowmya Vajrala, Hitesh Kumar, et al.
+- **abs:** https://arxiv.org/abs/2601.11684 · **PDF:** https://arxiv.org/pdf/2601.11684v1
+
+## 21334. Shape-morphing programming of soft materials on complex geometries via neural operator
+
+- **ID:** `2601.11126` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Lu Chen, Gengxiang Chen, Xu Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.11126 · **PDF:** https://arxiv.org/pdf/2601.11126v2
+
+## 21335. Learn Before Represent: Bridging Generative and Contrastive Learning for Domain-Specific LLM Embeddings
+
+- **ID:** `2601.11124` · **primary:** `cs.IR` · **Submitted:** 2026-01-16
+- **Authors:** Xiaoyu Liang, Yuchen Peng, Jiale Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.11124 · **PDF:** https://arxiv.org/pdf/2601.11124v1
+
+## 21336. Optimized Algorithms for Text Clustering with LLM-Generated Constraints
+
+- **ID:** `2601.11118` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Chaoqi Jia, Weihong Wu, Longkun Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.11118 · **PDF:** https://arxiv.org/pdf/2601.11118v1
+
+## 21337. Comprehensive Robust Dynamic Mode Decomposition from Mode Extraction to Dimensional Reduction
+
+- **ID:** `2601.11116` · **primary:** `eess.SP` · **Submitted:** 2026-01-16
+- **Authors:** Yuki Nakamura, Shingo Takemoto, Shunsuke Ono
+- **abs:** https://arxiv.org/abs/2601.11116 · **PDF:** https://arxiv.org/pdf/2601.11116v1
+
+## 21338. Differentially Private Subspace Fine-Tuning for Large Language Models
+
+- **ID:** `2601.11113` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Lele Zheng, Xiang Wang, Tao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.11113 · **PDF:** https://arxiv.org/pdf/2601.11113v1
+
+## 21339. Vision-as-Inverse-Graphics Agent via Interleaved Multimodal Reasoning
+
+- **ID:** `2601.11109` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Shaofeng Yin, Jiaxin Ge, Zora Zhiruo Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.11109 · **PDF:** https://arxiv.org/pdf/2601.11109v2
+
+## 21340. Bridging the Knowledge Void: Inference-time Acquisition of Unfamiliar Programming Languages for Coding Tasks
+
+- **ID:** `2602.06976` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Chen Shen, Wei Cheng, Jingyue Yang, et al.
+- **abs:** https://arxiv.org/abs/2602.06976 · **PDF:** https://arxiv.org/pdf/2602.06976v1
+
+## 21341. ReCreate: Reasoning and Creating Domain Agents Driven by Experience
+
+- **ID:** `2601.11100` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Zhezheng Hao, Hong Wang, Jian Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.11100 · **PDF:** https://arxiv.org/pdf/2601.11100v1
+
+## 21342. KANHedge: Efficient Hedging of High-Dimensional Options Using Kolmogorov-Arnold Network-Based BSDE Solver
+
+- **ID:** `2601.11097` · **primary:** `q-fin.CP` · **Submitted:** 2026-01-16
+- **Authors:** Rushikesh Handal, Masanori Hirano
+- **abs:** https://arxiv.org/abs/2601.11097 · **PDF:** https://arxiv.org/pdf/2601.11097v1
+
+## 21343. AutoBinder Agent: An MCP-Based Agent for End-to-End Protein Binder Design
+
+- **ID:** `2602.00019` · **primary:** `q-bio.BM` · **Submitted:** 2026-01-16
+- **Authors:** Fukang Ge, Jiarui Zhu, Linjie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.00019 · **PDF:** https://arxiv.org/pdf/2602.00019v1
+
+## 21344. Attesting Model Lineage by Consisted Knowledge Evolution with Fine-Tuning Trajectory
+
+- **ID:** `2601.11683` · **primary:** `cs.CR` · **Submitted:** 2026-01-16
+- **Authors:** Zhuoyi Shang, Jiasen Li, Pengzhen Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.11683 · **PDF:** https://arxiv.org/pdf/2601.11683v1
+
+## 21345. Integrity Shield A System for Ethical AI Use & Authorship Transparency in Assessments
+
+- **ID:** `2601.11093` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Ashish Raj Shekhar, Shiven Agarwal, Priyanuj Bordoloi, et al.
+- **abs:** https://arxiv.org/abs/2601.11093 · **PDF:** https://arxiv.org/pdf/2601.11093v1
+
+## 21346. Split-and-Conquer: Distributed Factor Modeling for High-Dimensional Matrix-Variate Time Series
+
+- **ID:** `2601.11091` · **primary:** `stat.ML` · **Submitted:** 2026-01-16
+- **Authors:** Hangjin Jiang, Yuzhou Li, Zhaoxing Gao
+- **abs:** https://arxiv.org/abs/2601.11091 · **PDF:** https://arxiv.org/pdf/2601.11091v1
+
+## 21347. Efficient Multilingual Name Type Classification Using Convolutional Networks
+
+- **ID:** `2601.11090` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Davor Lauc
+- **abs:** https://arxiv.org/abs/2601.11090 · **PDF:** https://arxiv.org/pdf/2601.11090v1
+
+## 21348. MiCA: A Mobility-Informed Causal Adapter for Lightweight Epidemic Forecasting
+
+- **ID:** `2601.11089` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Suhan Guo, Jiahong Deng, Furao Shen
+- **abs:** https://arxiv.org/abs/2601.11089 · **PDF:** https://arxiv.org/pdf/2601.11089v2
+
+## 21349. Soft Bayesian Context Tree Models for Real-Valued Time Series
+
+- **ID:** `2601.11079` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Shota Saito, Yuta Nakahara, Toshiyasu Matsushima
+- **abs:** https://arxiv.org/abs/2601.11079 · **PDF:** https://arxiv.org/pdf/2601.11079v1
+
+## 21350. Visual Marker Search for Autonomous Drone Landing in Diverse Urban Environments
+
+- **ID:** `2601.11078` · **primary:** `cs.RO` · **Submitted:** 2026-01-16
+- **Authors:** Jiaohong Yao, Linfeng Liang, Yao Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.11078 · **PDF:** https://arxiv.org/pdf/2601.11078v1
+
+## 21351. ABC-Bench: Benchmarking Agentic Backend Coding in Real-World Development
+
+- **ID:** `2601.11077` · **primary:** `cs.SE` · **Submitted:** 2026-01-16
+- **Authors:** Jie Yang, Honglin Guo, Li Ji, et al.
+- **abs:** https://arxiv.org/abs/2601.11077 · **PDF:** https://arxiv.org/pdf/2601.11077v1
+
+## 21352. A3D: Adaptive Affordance Assembly with Dual-Arm Manipulation
+
+- **ID:** `2601.11076` · **primary:** `cs.RO` · **Submitted:** 2026-01-16
+- **Authors:** Jiaqi Liang, Yue Chen, Qize Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.11076 · **PDF:** https://arxiv.org/pdf/2601.11076v1
+
+## 21353. Bridging Cognitive Neuroscience and Graph Intelligence: Hippocampus-Inspired Multi-View Hypergraph Learning for Web Finance Fraud
+
+- **ID:** `2601.11073` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Rongkun Cui, Nana Zhang, Kun Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.11073 · **PDF:** https://arxiv.org/pdf/2601.11073v2
+
+## 21354. Fairness in Healthcare Processes: A Quantitative Analysis of Decision Making in Triage
+
+- **ID:** `2601.11065` · **primary:** `cs.CY` · **Submitted:** 2026-01-16
+- **Authors:** Rachmadita Andreswari, Stephan A. Fahrenkrog-Petersen, Jan Mendling
+- **abs:** https://arxiv.org/abs/2601.11065 · **PDF:** https://arxiv.org/pdf/2601.11065v2
+
+## 21355. EmboTeam: Grounding LLM Reasoning into Reactive Behavior Trees via PDDL for Embodied Multi-Robot Collaboration
+
+- **ID:** `2601.11063` · **primary:** `cs.RO` · **Submitted:** 2026-01-16
+- **Authors:** Haishan Zeng, Mengna Wang, Peng Li
+- **abs:** https://arxiv.org/abs/2601.11063 · **PDF:** https://arxiv.org/pdf/2601.11063v2
+
+## 21356. Spurious Rewards Paradox: Mechanistically Understanding How RLVR Activates Memorization Shortcuts in LLMs
+
+- **ID:** `2601.11061` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Lecheng Yan, Ruizhe Li, Guanhua Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.11061 · **PDF:** https://arxiv.org/pdf/2601.11061v1
+
+## 21357. HALO: Semantic-Aware Distributed LLM Inference in Lossy Edge Network
+
+- **ID:** `2601.11676` · **primary:** `cs.DC` · **Submitted:** 2026-01-16
+- **Authors:** Peirong Zheng, Wenchao Xu, Haozhao Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.11676 · **PDF:** https://arxiv.org/pdf/2601.11676v1
+
+## 21358. Epistemic Constitutionalism Or: how to avoid coherence bias
+
+- **ID:** `2601.14295` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Michele Loi
+- **abs:** https://arxiv.org/abs/2601.14295 · **PDF:** https://arxiv.org/pdf/2601.14295v2
+
+## 21359. Predicting Biased Human Decision-Making with Large Language Models in Conversational Settings
+
+- **ID:** `2601.11049` · **primary:** `cs.HC` · **Submitted:** 2026-01-16
+- **Authors:** Stephen Pilli, Vivek Nallur
+- **abs:** https://arxiv.org/abs/2601.11049 · **PDF:** https://arxiv.org/pdf/2601.11049v1
+
+## 21360. Gap-K%: Measuring Top-1 Prediction Gap for Detecting Pretraining Data
+
+- **ID:** `2601.19936` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Minseo Kwak, Jaehyung Kim
+- **abs:** https://arxiv.org/abs/2601.19936 · **PDF:** https://arxiv.org/pdf/2601.19936v1
+
+## 21361. CoG: Controllable Graph Reasoning via Relational Blueprints and Failure-Aware Refinement over Knowledge Graphs
+
+- **ID:** `2601.11047` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Yuanxiang Liu, Songze Li, Xiaoke Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.11047 · **PDF:** https://arxiv.org/pdf/2601.11047v1
+
+## 21362. OpFML: Pipeline for ML-based Operational Forecasting
+
+- **ID:** `2601.11046` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Shahbaz Alvi, Giusy Fedele, Gabriele Accarino, et al.
+- **abs:** https://arxiv.org/abs/2601.11046 · **PDF:** https://arxiv.org/pdf/2601.11046v1
+
+## 21363. AgencyBench: Benchmarking the Frontiers of Autonomous Agents in 1M-Token Real-World Contexts
+
+- **ID:** `2601.11044` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Keyu Li, Junhao Shi, Yang Xiao, et al.
+- **abs:** https://arxiv.org/abs/2601.11044 · **PDF:** https://arxiv.org/pdf/2601.11044v2
+
+## 21364. Spectral Characterization and Mitigation of Sequential Knowledge Editing Collapse
+
+- **ID:** `2601.11042` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Chi Zhang, Mengqi Zhang, Xiaotian Ye, et al.
+- **abs:** https://arxiv.org/abs/2601.11042 · **PDF:** https://arxiv.org/pdf/2601.11042v1
+
+## 21365. SonicBench: Dissecting the Physical Perception Bottleneck in Large Audio Language Models
+
+- **ID:** `2601.11039` · **primary:** `cs.SD` · **Submitted:** 2026-01-16
+- **Authors:** Yirong Sun, Yanjun Chen, Xin Qiu, et al.
+- **abs:** https://arxiv.org/abs/2601.11039 · **PDF:** https://arxiv.org/pdf/2601.11039v1
+
+## 21366. Budget-Aware Anytime Reasoning with LLM-Synthesized Preference Data
+
+- **ID:** `2601.11038` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Xuanming Zhang, Shwan Ashrafi, Aziza Mirsaidova, et al.
+- **abs:** https://arxiv.org/abs/2601.11038 · **PDF:** https://arxiv.org/pdf/2601.11038v1
+
+## 21367. BAPO: Boundary-Aware Policy Optimization for Reliable Agentic Search
+
+- **ID:** `2601.11037` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Shiyu Liu, Yongjing Yin, Jianhao Yan, et al.
+- **abs:** https://arxiv.org/abs/2601.11037 · **PDF:** https://arxiv.org/pdf/2601.11037v1
+
+## 21368. Self-Augmented Mixture-of-Experts for QoS Prediction
+
+- **ID:** `2601.11036` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Kecheng Cai, Chao Peng, Chenyang Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.11036 · **PDF:** https://arxiv.org/pdf/2601.11036v3
+
+## 21369. Your One-Stop Solution for AI-Generated Video Detection
+
+- **ID:** `2601.11035` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Long Ma, Zihao Xue, Yan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.11035 · **PDF:** https://arxiv.org/pdf/2601.11035v1
+
+## 21370. IDDR-NGP: Incorporating Detectors for Distractor Removal with Instant Neural Radiance Field
+
+- **ID:** `2601.11030` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Xianliang Huang, Jiajie Gou, Shuhang Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.11030 · **PDF:** https://arxiv.org/pdf/2601.11030v1
+
+## 21371. A Quantum-Driven Evolutionary Framework for Solving High-Dimensional Sharpe Ratio Portfolio Optimization
+
+- **ID:** `2601.11029` · **primary:** `cs.NE` · **Submitted:** 2026-01-16
+- **Authors:** Mingyang Yu, Jiaqi Zhang, Haorui Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.11029 · **PDF:** https://arxiv.org/pdf/2601.11029v2
+
+## 21372. AVP-Pro: An Adaptive Multi-Modal Fusion and Contrastive Learning Approach for Comprehensive Two-Stage Antiviral Peptide Identification
+
+- **ID:** `2601.11028` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Xinru Wen, Weizhong Lin, zi liu, et al.
+- **abs:** https://arxiv.org/abs/2601.11028 · **PDF:** https://arxiv.org/pdf/2601.11028v1
+
+## 21373. Matching High-Dimensional Geometric Quantiles for Test-Time Adaptation of Transformers and Convolutional Networks Alike
+
+- **ID:** `2601.11022` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Sravan Danda, Aditya Challa, Shlok Mehendale, et al.
+- **abs:** https://arxiv.org/abs/2601.11022 · **PDF:** https://arxiv.org/pdf/2601.11022v1
+
+## 21374. Combating Spurious Correlations in Graph Interpretability via Self-Reflection
+
+- **ID:** `2601.11021` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Kecheng Cai, Chenyang Xu, Chao Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.11021 · **PDF:** https://arxiv.org/pdf/2601.11021v2
+
+## 21375. From Interpretability to Performance: Optimizing Retrieval Heads for Long-Context Language Models
+
+- **ID:** `2601.11020` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Youmi Ma, Naoaki Okazaki
+- **abs:** https://arxiv.org/abs/2601.11020 · **PDF:** https://arxiv.org/pdf/2601.11020v2
+
+## 21376. Finding the Translation Switch: Discovering and Exploiting the Task-Initiation Features in LLMs
+
+- **ID:** `2601.11019` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Xinwei Wu, Heng Liu, Xiaohu Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.11019 · **PDF:** https://arxiv.org/pdf/2601.11019v1
+
+## 21377. Generating metamers of human scene understanding
+
+- **ID:** `2601.11675` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Ritik Raina, Abe Leite, Alexandros Graikos, et al.
+- **abs:** https://arxiv.org/abs/2601.11675 · **PDF:** https://arxiv.org/pdf/2601.11675v3
+
+## 21378. Contextual Distributionally Robust Optimization with Causal and Continuous Structure: An Interpretable and Tractable Approach
+
+- **ID:** `2601.11016` · **primary:** `stat.ML` · **Submitted:** 2026-01-16
+- **Authors:** Fenglin Zhang, Jie Wang
+- **abs:** https://arxiv.org/abs/2601.11016 · **PDF:** https://arxiv.org/pdf/2601.11016v2
+
+## 21379. Efficient Protein Optimization via Structure-aware Hamiltonian Dynamics
+
+- **ID:** `2601.11012` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Jiahao Wang, Shuangjia Zheng
+- **abs:** https://arxiv.org/abs/2601.11012 · **PDF:** https://arxiv.org/pdf/2601.11012v1
+
+## 21380. AdaMARP: An Adaptive Multi-Agent Interaction Framework for General Immersive Role-Playing
+
+- **ID:** `2601.11007` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Zhenhua Xu, Dongsheng Chen, Shuo Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.11007 · **PDF:** https://arxiv.org/pdf/2601.11007v1
+
+## 21381. Backdoor Attacks on Multi-modal Contrastive Learning
+
+- **ID:** `2601.11006` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Simi D Kuniyilh, Rita Machacy
+- **abs:** https://arxiv.org/abs/2601.11006 · **PDF:** https://arxiv.org/pdf/2601.11006v1
+
+## 21382. Pigment Network Detection and Classification in Dermoscopic Images Using Directional Imaging Algorithms and Convolutional Neural Networks
+
+- **ID:** `2601.11674` · **primary:** `eess.IV` · **Submitted:** 2026-01-16
+- **Authors:** M. A. Rasel, Sameem Abdul Kareem, Unaizah Obaidellah
+- **abs:** https://arxiv.org/abs/2601.11674 · **PDF:** https://arxiv.org/pdf/2601.11674v1
+
+## 21383. NAACL: Noise-AwAre Verbal Confidence Calibration for Robust LLMs in RAG Systems
+
+- **ID:** `2601.11004` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Jiayu Liu, Rui Wang, Qing Zong, et al.
+- **abs:** https://arxiv.org/abs/2601.11004 · **PDF:** https://arxiv.org/pdf/2601.11004v2
+
+## 21384. Redefining Machine Simultaneous Interpretation: From Incremental Translation to Human-Like Strategies
+
+- **ID:** `2601.11002` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Qianen Zhang, Zeyu Yang, Satoshi Nakamura
+- **abs:** https://arxiv.org/abs/2601.11002 · **PDF:** https://arxiv.org/pdf/2601.11002v1
+
+## 21385. When Personalization Misleads: Understanding and Mitigating Hallucinations in Personalized LLMs
+
+- **ID:** `2601.11000` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Zhongxiang Sun, Yi Zhan, Chenglei Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.11000 · **PDF:** https://arxiv.org/pdf/2601.11000v1
+
+## 21386. Exact Constraint Enforcement in Physics-Informed Extreme Learning Machines using Null-Space Projection Framework
+
+- **ID:** `2601.10999` · **primary:** `math.NA` · **Submitted:** 2026-01-16
+- **Authors:** Rishi Mishra, Smriti, Balaji Srinivasan, et al.
+- **abs:** https://arxiv.org/abs/2601.10999 · **PDF:** https://arxiv.org/pdf/2601.10999v2
+
+## 21387. Regret-Driven Portfolios: LLM-Guided Smart Clustering for Optimal Allocation
+
+- **ID:** `2601.17021` · **primary:** `q-fin.PM` · **Submitted:** 2026-01-16
+- **Authors:** Muhammad Abro, Hassan Jaleel
+- **abs:** https://arxiv.org/abs/2601.17021 · **PDF:** https://arxiv.org/pdf/2601.17021v1
+
+## 21388. Memorize Early, Then Query: Inlier-Memorization-Guided Active Outlier Detection
+
+- **ID:** `2601.10993` · **primary:** `stat.ML` · **Submitted:** 2026-01-16
+- **Authors:** Minseo Kang, Seunghwan Park, Dongha Kim
+- **abs:** https://arxiv.org/abs/2601.10993 · **PDF:** https://arxiv.org/pdf/2601.10993v2
+
+## 21389. Constant Metric Scaling in Riemannian Computation
+
+- **ID:** `2601.10992` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Kisung You
+- **abs:** https://arxiv.org/abs/2601.10992 · **PDF:** https://arxiv.org/pdf/2601.10992v2
+
+## 21390. Reasoning Distillation for Lightweight Automated Program Repair
+
+- **ID:** `2601.10987` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Aanand Balasubramanian, Sashank Silwal
+- **abs:** https://arxiv.org/abs/2601.10987 · **PDF:** https://arxiv.org/pdf/2601.10987v1
+
+## 21391. BiomechAgent: AI-Assisted Biomechanical Analysis Through Code-Generating Agents
+
+- **ID:** `2602.06975` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** R. James Cotton, Thomas Leonard
+- **abs:** https://arxiv.org/abs/2602.06975 · **PDF:** https://arxiv.org/pdf/2602.06975v1
+
+## 21392. ZPD Detector: Data Selection via Capability-Difficulty Alignment for Large Language Models
+
+- **ID:** `2601.10986` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Bo Yang, Yunkui Chen, Lanfei Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.10986 · **PDF:** https://arxiv.org/pdf/2601.10986v1
+
+## 21393. Integrating Color Histogram Analysis and Convolutional Neural Network for Skin Lesion Classification
+
+- **ID:** `2601.20869` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-16
+- **Authors:** M. A. Rasel, Sameem Abdul Kareem, Unaizah Obaidellah
+- **abs:** https://arxiv.org/abs/2601.20869 · **PDF:** https://arxiv.org/pdf/2601.20869v1
+
+## 21394. Toward Adaptive Grid Resilience: A Gradient-Free Meta-RL Framework for Critical Load Restoration
+
+- **ID:** `2601.10973` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Zain ul Abdeen, Waris Gill, Ming Jin
+- **abs:** https://arxiv.org/abs/2601.10973 · **PDF:** https://arxiv.org/pdf/2601.10973v1
+
+## 21395. AJAR: Adaptive Jailbreak Architecture for Red-teaming
+
+- **ID:** `2601.10971` · **primary:** `cs.CR` · **Submitted:** 2026-01-16
+- **Authors:** Yipu Dou, Wang Yang
+- **abs:** https://arxiv.org/abs/2601.10971 · **PDF:** https://arxiv.org/pdf/2601.10971v2
+
+## 21396. Transient learning dynamics drive escape from sharp valleys in Stochastic Gradient Descent
+
+- **ID:** `2601.10962` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Ning Yang, Yikuan Zhang, Qi Ouyang, et al.
+- **abs:** https://arxiv.org/abs/2601.10962 · **PDF:** https://arxiv.org/pdf/2601.10962v1
+
+## 21397. Multivariate LSTM-Based Forecasting for Renewable Energy: Enhancing Climate Change Mitigation
+
+- **ID:** `2601.10961` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Farshid Kamrani, Kristen Schell
+- **abs:** https://arxiv.org/abs/2601.10961 · **PDF:** https://arxiv.org/pdf/2601.10961v1
+
+## 21398. Steering Language Models Before They Speak: Logit-Level Interventions
+
+- **ID:** `2601.10960` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Hyeseon An, Shinwoo Park, Hyundong Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.10960 · **PDF:** https://arxiv.org/pdf/2601.10960v1
+
+## 21399. Depression Detection Based on Electroencephalography Using a Hybrid Deep Neural Network CNN-GRU and MRMR Feature Selection
+
+- **ID:** `2601.10959` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-16
+- **Authors:** Mohammad Reza Yousefi, Hajar Ismail Al-Tamimi, Amin Dehghani
+- **abs:** https://arxiv.org/abs/2601.10959 · **PDF:** https://arxiv.org/pdf/2601.10959v1
+
+## 21400. A Confidence-Variance Theory for Pseudo-Label Selection in Semi-Supervised Learning
+
+- **ID:** `2601.11670` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Jinshi Liu, Pan Liu, Lei He
+- **abs:** https://arxiv.org/abs/2601.11670 · **PDF:** https://arxiv.org/pdf/2601.11670v2
+
+## 21401. Beyond Max Tokens: Stealthy Resource Amplification via Tool Calling Chains in LLM Agents
+
+- **ID:** `2601.10955` · **primary:** `cs.CR` · **Submitted:** 2026-01-16
+- **Authors:** Kaiyu Zhou, Yongsen Zheng, Yicheng He, et al.
+- **abs:** https://arxiv.org/abs/2601.10955 · **PDF:** https://arxiv.org/pdf/2601.10955v2
+
+## 21402. The Paradigm Shift: A Comprehensive Survey on Large Vision Language Models for Multimodal Fake News Detection
+
+- **ID:** `2601.15316` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Wei Ai, Yilong Tan, Yuntao Shou, et al.
+- **abs:** https://arxiv.org/abs/2601.15316 · **PDF:** https://arxiv.org/pdf/2601.15316v1
+
+## 21403. Multi-Stage Patient Role-Playing Framework for Realistic Clinical Interactions
+
+- **ID:** `2601.10951` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Shijie Jiang, Zefan Zhang, Kehua Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.10951 · **PDF:** https://arxiv.org/pdf/2601.10951v1
+
+## 21404. PatientVLM Meets DocVLM: Pre-Consultation Dialogue Between Vision-Language Models for Efficient Diagnosis
+
+- **ID:** `2601.10945` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** K Lokesh, Abhirama Subramanyam Penamakuri, Uday Agarwal, et al.
+- **abs:** https://arxiv.org/abs/2601.10945 · **PDF:** https://arxiv.org/pdf/2601.10945v1
+
+## 21405. IPEC: Test-Time Incremental Prototype Enhancement Classifier for Few-Shot Learning
+
+- **ID:** `2601.11669` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Wenwen Liao, Hang Ruan, Jianbo Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.11669 · **PDF:** https://arxiv.org/pdf/2601.11669v1
+
+## 21406. Distill-then-Replace: Efficient Task-Specific Hybrid Attention Model Construction
+
+- **ID:** `2601.11667` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Xiaojie Xia, Huigang Zhang, Chaoliang Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.11667 · **PDF:** https://arxiv.org/pdf/2601.11667v1
+
+## 21407. HOSL: Hybrid-Order Split Learning for Memory-Constrained Edge Training
+
+- **ID:** `2601.10940` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Aakriti Lnu, Zhe Li, Dandan Liang, et al.
+- **abs:** https://arxiv.org/abs/2601.10940 · **PDF:** https://arxiv.org/pdf/2601.10940v3
+
+## 21408. Sparse Data Tree Canopy Segmentation: Fine-Tuning Leading Pretrained Models on Only 150 Images
+
+- **ID:** `2601.10931` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** David Szczecina, Hudson Sun, Anthony Bertnyk, et al.
+- **abs:** https://arxiv.org/abs/2601.10931 · **PDF:** https://arxiv.org/pdf/2601.10931v1
+
+## 21409. MATEX: Multi-scale Attention and Text-guided Explainability of Medical Vision-Language Models
+
+- **ID:** `2601.11666` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Muhammad Imran, Chi Lee, Yugyung Lee
+- **abs:** https://arxiv.org/abs/2601.11666 · **PDF:** https://arxiv.org/pdf/2601.11666v1
+
+## 21410. Selecting Language Models for Social Science: Start Small, Start Open, and Validate
+
+- **ID:** `2601.10926` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Dustin S. Stoltz, Marshall A. Taylor, Sanuj Kumar
+- **abs:** https://arxiv.org/abs/2601.10926 · **PDF:** https://arxiv.org/pdf/2601.10926v1
+
+## 21411. Massively Multilingual Joint Segmentation and Glossing
+
+- **ID:** `2601.10925` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Michael Ginn, Lindia Tjuatja, Enora Rice, et al.
+- **abs:** https://arxiv.org/abs/2601.10925 · **PDF:** https://arxiv.org/pdf/2601.10925v2
+
+## 21412. What Matters in Data Curation for Multimodal Reasoning? Insights from the DCVLR Challenge
+
+- **ID:** `2601.10922` · **primary:** `cs.AI` · **Submitted:** 2026-01-16
+- **Authors:** Yosub Shin, Michael Buriek, Boris Sobolev, et al.
+- **abs:** https://arxiv.org/abs/2601.10922 · **PDF:** https://arxiv.org/pdf/2601.10922v1
+
+## 21413. RobuMTL: Enhancing Multi-Task Learning Robustness Against Weather Conditions
+
+- **ID:** `2601.10921` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Tasneem Shaffee, Sherief Reda
+- **abs:** https://arxiv.org/abs/2601.10921 · **PDF:** https://arxiv.org/pdf/2601.10921v1
+
+## 21414. Neural Induction of Finite-State Transducers
+
+- **ID:** `2601.10918` · **primary:** `cs.CL` · **Submitted:** 2026-01-16
+- **Authors:** Michael Ginn, Alexis Palmer, Mans Hulden
+- **abs:** https://arxiv.org/abs/2601.10918 · **PDF:** https://arxiv.org/pdf/2601.10918v2
+
+## 21415. Self-learned representation-guided latent diffusion model for breast cancer classification in deep ultraviolet whole surface images
+
+- **ID:** `2601.10917` · **primary:** `cs.CV` · **Submitted:** 2026-01-16
+- **Authors:** Pouya Afshin, David Helminiak, Tianling Niu, et al.
+- **abs:** https://arxiv.org/abs/2601.10917 · **PDF:** https://arxiv.org/pdf/2601.10917v2
+
+## 21416. A PAC-Bayesian Analysis of Channel-Induced Degradation in Edge Inference
+
+- **ID:** `2601.10915` · **primary:** `cs.IT` · **Submitted:** 2026-01-16
+- **Authors:** Yangshuo He, Guanding Yu, Jingge Zhu
+- **abs:** https://arxiv.org/abs/2601.10915 · **PDF:** https://arxiv.org/pdf/2601.10915v1
+
+## 21417. FAConvLSTM: Factorized-Attention ConvLSTM for Efficient Feature Extraction in Multivariate Climate Data
+
+- **ID:** `2601.10914` · **primary:** `cs.LG` · **Submitted:** 2026-01-16
+- **Authors:** Francis Ndikum Nji, Jianwu Wang
+- **abs:** https://arxiv.org/abs/2601.10914 · **PDF:** https://arxiv.org/pdf/2601.10914v1
+
+## 21418. Realistic Curriculum Reinforcement Learning for Autonomous and Sustainable Marine Vessel Navigation
+
+- **ID:** `2601.10911` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Zhang Xiaocai, Xiao Zhe, Liang Maohan, et al.
+- **abs:** https://arxiv.org/abs/2601.10911 · **PDF:** https://arxiv.org/pdf/2601.10911v1
+
+## 21419. Action Shapley: A Training Data Selection Metric for World Model in Reinforcement Learning
+
+- **ID:** `2601.10905` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Rajat Ghosh, Debojyoti Dutta
+- **abs:** https://arxiv.org/abs/2601.10905 · **PDF:** https://arxiv.org/pdf/2601.10905v1
+
+## 21420. Serverless AI Security: Attack Surface Analysis and Runtime Protection Mechanisms for FaaS-Based Machine Learning
+
+- **ID:** `2601.11664` · **primary:** `cs.CR` · **Submitted:** 2026-01-15
+- **Authors:** Chetan Pathade, Vinod Dhimam, Sheheryar Ahmad, et al.
+- **abs:** https://arxiv.org/abs/2601.11664 · **PDF:** https://arxiv.org/pdf/2601.11664v1
+
+## 21421. ARC Prize 2025: Technical Report
+
+- **ID:** `2601.10904` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** François Chollet, Mike Knoop, Gregory Kamradt, et al.
+- **abs:** https://arxiv.org/abs/2601.10904 · **PDF:** https://arxiv.org/pdf/2601.10904v1
+
+## 21422. How Do We Engage with Other Disciplines? A Framework to Study Meaningful Interdisciplinary Discourse in Scholarly Publications
+
+- **ID:** `2601.17020` · **primary:** `cs.DL` · **Submitted:** 2026-01-15
+- **Authors:** Bagyasree Sudharsan, Alexandria Leto, Maria Leonor Pacheco
+- **abs:** https://arxiv.org/abs/2601.17020 · **PDF:** https://arxiv.org/pdf/2601.17020v1
+
+## 21423. DialDefer: A Framework for Detecting and Mitigating LLM Dialogic Deference
+
+- **ID:** `2601.10896` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Parisa Rabbani, Priyam Sahoo, Ruben Mathew, et al.
+- **abs:** https://arxiv.org/abs/2601.10896 · **PDF:** https://arxiv.org/pdf/2601.10896v1
+
+## 21424. Learning collision operators from plasma phase space data using differentiable simulators
+
+- **ID:** `2601.10885` · **primary:** `physics.plasm-ph` · **Submitted:** 2026-01-15
+- **Authors:** Diogo D. Carvalho, Pablo J. Bilbao, Warren B. Mori, et al.
+- **abs:** https://arxiv.org/abs/2601.10885 · **PDF:** https://arxiv.org/pdf/2601.10885v1
+
+## 21425. Activation Sensitivity as a Unifying Principle for Post-Training Quantization
+
+- **ID:** `2601.11663` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Bruce Changlong Xu
+- **abs:** https://arxiv.org/abs/2601.11663 · **PDF:** https://arxiv.org/pdf/2601.11663v1
+
+## 21426. Medical SAM3: A Foundation Model for Universal Prompt-Driven Medical Image Segmentation
+
+- **ID:** `2601.10880` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Chongcong Jiang, Tianxingjian Ding, Chuhan Song, et al.
+- **abs:** https://arxiv.org/abs/2601.10880 · **PDF:** https://arxiv.org/pdf/2601.10880v1
+
+## 21427. Unit-Consistent (UC) Adjoint for GSD and Backprop in Deep Learning Applications
+
+- **ID:** `2601.10873` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Jeffrey Uhlmann
+- **abs:** https://arxiv.org/abs/2601.10873 · **PDF:** https://arxiv.org/pdf/2601.10873v1
+
+## 21428. Machine learning model for predicting surface wettability in laser-textured metal alloys
+
+- **ID:** `2601.11661` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Mohammad Mohammadzadeh Sanandaji, Danial Ebrahimzadeh, Mohammad Ikram Haider, et al.
+- **abs:** https://arxiv.org/abs/2601.11661 · **PDF:** https://arxiv.org/pdf/2601.11661v1
+
+## 21429. Beyond Accuracy: A Stability-Aware Metric for Multi-Horizon Forecasting
+
+- **ID:** `2601.10863` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Chutian Ma, Grigorii Pomazkin, Giacinto Paolo Saggese, et al.
+- **abs:** https://arxiv.org/abs/2601.10863 · **PDF:** https://arxiv.org/pdf/2601.10863v2
+
+## 21430. Zeros can be Informative: Masked Binary U-Net for Image Segmentation on Tensor Cores
+
+- **ID:** `2601.11660` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Chunshu Wu, Ruibing Song, Sushant Kondguli, et al.
+- **abs:** https://arxiv.org/abs/2601.11660 · **PDF:** https://arxiv.org/pdf/2601.11660v1
+
+## 21431. AI-Guided Human-In-the-Loop Inverse Design of High Performance Engineering Structures
+
+- **ID:** `2601.10859` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Dat Quoc Ha, Md Ferdous Alam, Markus J. Buehler, et al.
+- **abs:** https://arxiv.org/abs/2601.10859 · **PDF:** https://arxiv.org/pdf/2601.10859v1
+
+## 21432. The Llama 4 Herd: Architecture, Training, Evaluation, and Deployment Notes
+
+- **ID:** `2601.11659` · **primary:** `cs.SE` · **Submitted:** 2026-01-15
+- **Authors:** Redacted by arXiv
+- **abs:** https://arxiv.org/abs/2601.11659 · **PDF:** https://arxiv.org/pdf/2601.11659v1
+
+## 21433. Towards AGI A Pragmatic Approach Towards Self Evolving Agent
+
+- **ID:** `2601.11658` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Indrajit Kar, Sammy Zonunpuia, Zonunfeli Ralte
+- **abs:** https://arxiv.org/abs/2601.11658 · **PDF:** https://arxiv.org/pdf/2601.11658v1
+
+## 21434. Size is Not the Solution: Deformable Convolutions for Effective Physics Aware Deep Learning
+
+- **ID:** `2601.11657` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Jack T. Beerman, Shobhan Roy, H. S. Udaykumar, et al.
+- **abs:** https://arxiv.org/abs/2601.11657 · **PDF:** https://arxiv.org/pdf/2601.11657v1
+
+## 21435. EncodeRec: An Embedding Backbone for Recommendation Systems
+
+- **ID:** `2601.10837` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Guy Hadad, Neomi Rabaev, Bracha Shapira
+- **abs:** https://arxiv.org/abs/2601.10837 · **PDF:** https://arxiv.org/pdf/2601.10837v1
+
+## 21436. Can Vision-Language Models Understand Construction Workers? An Exploratory Study
+
+- **ID:** `2601.10835` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Hieu Bui, Nathaniel E. Chodosh, Arash Tavakoli
+- **abs:** https://arxiv.org/abs/2601.10835 · **PDF:** https://arxiv.org/pdf/2601.10835v1
+
+## 21437. Approximately Optimal Global Planning for Contact-Rich SE(2) Manipulation on a Graph of Reachable Sets
+
+- **ID:** `2601.10827` · **primary:** `cs.RO` · **Submitted:** 2026-01-15
+- **Authors:** Simin Liu, Tong Zhao, Bernhard Paus Graesdal, et al.
+- **abs:** https://arxiv.org/abs/2601.10827 · **PDF:** https://arxiv.org/pdf/2601.10827v1
+
+## 21438. Reasoning Models Generate Societies of Thought
+
+- **ID:** `2601.10825` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Junsol Kim, Shiyang Lai, Nino Scherrer, et al.
+- **abs:** https://arxiv.org/abs/2601.10825 · **PDF:** https://arxiv.org/pdf/2601.10825v1
+
+## 21439. Mugi: Value Level Parallelism For Efficient LLMs
+
+- **ID:** `2601.10823` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Daniel Price, Prabhu Vellaisamy, John Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.10823 · **PDF:** https://arxiv.org/pdf/2601.10823v2
+
+## 21440. Towards Reliable ML Feature Engineering via Planning in Constrained-Topology of LLM Agents
+
+- **ID:** `2601.10820` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Himanshu Thakur, Anusha Kamath, Anurag Muthyala, et al.
+- **abs:** https://arxiv.org/abs/2601.10820 · **PDF:** https://arxiv.org/pdf/2601.10820v1
+
+## 21441. Digital Metabolism: Decoupling Logic from Facts via Regenerative Unlearning -- Towards a Pure Neural Logic Core
+
+- **ID:** `2601.10810` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Mengmeng Peng, Zhenyu Fang, He Sun
+- **abs:** https://arxiv.org/abs/2601.10810 · **PDF:** https://arxiv.org/pdf/2601.10810v1
+
+## 21442. A Concise Agent is Less Expert: Revealing Side Effects of Using Style Features on Conversational Agents
+
+- **ID:** `2601.10809` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Young-Min Cho, Yuan Yuan, Sharath Chandra Guntuku, et al.
+- **abs:** https://arxiv.org/abs/2601.10809 · **PDF:** https://arxiv.org/pdf/2601.10809v1
+
+## 21443. BYOL: Bring Your Own Language Into LLMs
+
+- **ID:** `2601.10804` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Syed Waqas Zamir, Wassim Hamidouche, Boulbaba Ben Amor, et al.
+- **abs:** https://arxiv.org/abs/2601.10804 · **PDF:** https://arxiv.org/pdf/2601.10804v1
+
+## 21444. Towards Tensor Network Models for Low-Latency Jet Tagging on FPGAs
+
+- **ID:** `2601.10801` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Alberto Coppi, Ema Puljak, Lorenzo Borella, et al.
+- **abs:** https://arxiv.org/abs/2601.10801 · **PDF:** https://arxiv.org/pdf/2601.10801v1
+
+## 21445. DInf-Grid: A Neural Differential Equation Solver with Differentiable Feature Grids
+
+- **ID:** `2601.10715` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Navami Kairanda, Shanthika Naik, Marc Habermann, et al.
+- **abs:** https://arxiv.org/abs/2601.10715 · **PDF:** https://arxiv.org/pdf/2601.10715v1
+
+## 21446. MatchTIR: Fine-Grained Supervision for Tool-Integrated Reasoning via Bipartite Matching
+
+- **ID:** `2601.10712` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Changle Qu, Sunhao Dai, Hengyi Cai, et al.
+- **abs:** https://arxiv.org/abs/2601.10712 · **PDF:** https://arxiv.org/pdf/2601.10712v1
+
+## 21447. High-accuracy and dimension-free sampling with diffusions
+
+- **ID:** `2601.10708` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Khashayar Gatmiry, Sitan Chen, Adil Salim
+- **abs:** https://arxiv.org/abs/2601.10708 · **PDF:** https://arxiv.org/pdf/2601.10708v1
+
+## 21448. See Less, Drive Better: Generalizable End-to-End Autonomous Driving via Foundation Models Stochastic Patch Selection
+
+- **ID:** `2601.10707` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Amir Mallak, Erfan Aasi, Shiva Sreeram, et al.
+- **abs:** https://arxiv.org/abs/2601.10707 · **PDF:** https://arxiv.org/pdf/2601.10707v1
+
+## 21449. Distributed Perceptron under Bounded Staleness, Partial Participation, and Noisy Communication
+
+- **ID:** `2601.10705` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Keval Jain, Anant Raj, Saurav Prakash, et al.
+- **abs:** https://arxiv.org/abs/2601.10705 · **PDF:** https://arxiv.org/pdf/2601.10705v2
+
+## 21450. Grounding Agent Memory in Contextual Intent
+
+- **ID:** `2601.10702` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Ruozhen Yang, Yucheng Jiang, Yueqi Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.10702 · **PDF:** https://arxiv.org/pdf/2601.10702v1
+
+## 21451. Advances and Frontiers of LLM-based Issue Resolution in Software Engineering: A Comprehensive Survey
+
+- **ID:** `2601.11655` · **primary:** `cs.SE` · **Submitted:** 2026-01-15
+- **Authors:** Caihua Li, Lianghong Guo, Yanlin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.11655 · **PDF:** https://arxiv.org/pdf/2601.11655v1
+
+## 21452. Communication-Efficient and Privacy-Adaptable Mechanism -- a Federated Learning Scheme with Convergence Analysis
+
+- **ID:** `2601.10701` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Chun Hei Michael Shiu, Chih Wei Ling
+- **abs:** https://arxiv.org/abs/2601.10701 · **PDF:** https://arxiv.org/pdf/2601.10701v1
+
+## 21453. LIBERTy: A Causal Framework for Benchmarking Concept-Based Explanations of LLMs with Structural Counterfactuals
+
+- **ID:** `2601.10700` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Gilat Toker, Nitay Calderon, Ohad Amosy, et al.
+- **abs:** https://arxiv.org/abs/2601.10700 · **PDF:** https://arxiv.org/pdf/2601.10700v2
+
+## 21454. The Impact of Generative AI on Architectural Conceptual Design: Performance, Creative Self-Efficacy and Cognitive Load
+
+- **ID:** `2601.10696` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Han Jiang, Yao Xiao, Rachel Hurley, et al.
+- **abs:** https://arxiv.org/abs/2601.10696 · **PDF:** https://arxiv.org/pdf/2601.10696v1
+
+## 21455. Data-driven stochastic reduced-order modeling of parametrized dynamical systems
+
+- **ID:** `2601.10690` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Andrew F. Ilersich, Kevin Course, Prasanth B. Nair
+- **abs:** https://arxiv.org/abs/2601.10690 · **PDF:** https://arxiv.org/pdf/2601.10690v1
+
+## 21456. Unified Optimization of Source Weights and Transfer Quantities in Multi-Source Transfer Learning: An Asymptotic Framework
+
+- **ID:** `2601.10779` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Qingyue Zhang, Chang Chu, Haohao Fu, et al.
+- **abs:** https://arxiv.org/abs/2601.10779 · **PDF:** https://arxiv.org/pdf/2601.10779v2
+
+## 21457. On the origin of neural scaling laws: from random graphs to natural language
+
+- **ID:** `2601.10684` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Maissam Barkeshli, Alberto Alfarano, Andrey Gromov
+- **abs:** https://arxiv.org/abs/2601.10684 · **PDF:** https://arxiv.org/pdf/2601.10684v1
+
+## 21458. Structure and Diversity Aware Context Bubble Construction for Enterprise Retrieval Augmented Systems
+
+- **ID:** `2601.10681` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Amir Khurshid, Abhishek Sehgal
+- **abs:** https://arxiv.org/abs/2601.10681 · **PDF:** https://arxiv.org/pdf/2601.10681v1
+
+## 21459. Are Your Reasoning Models Reasoning or Guessing? A Mechanistic Analysis of Hierarchical Reasoning Models
+
+- **ID:** `2601.10679` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Zirui Ren, Ziming Liu
+- **abs:** https://arxiv.org/abs/2601.10679 · **PDF:** https://arxiv.org/pdf/2601.10679v2
+
+## 21460. Single-Stage Huffman Encoder for ML Compression
+
+- **ID:** `2601.10673` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Aditya Agrawal, Albert Magyar, Hiteshwar Eswaraiah, et al.
+- **abs:** https://arxiv.org/abs/2601.10673 · **PDF:** https://arxiv.org/pdf/2601.10673v1
+
+## 21461. Detecting Winning Arguments with Large Language Models and Persuasion Strategies
+
+- **ID:** `2601.10660` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Tiziano Labruna, Arkadiusz Modzelewski, Giorgio Satta, et al.
+- **abs:** https://arxiv.org/abs/2601.10660 · **PDF:** https://arxiv.org/pdf/2601.10660v1
+
+## 21462. PACEvolve: Enabling Long-Horizon Progress-Aware Consistent Evolution
+
+- **ID:** `2601.10657` · **primary:** `cs.NE` · **Submitted:** 2026-01-15
+- **Authors:** Minghao Yan, Bo Peng, Benjamin Coleman, et al.
+- **abs:** https://arxiv.org/abs/2601.10657 · **PDF:** https://arxiv.org/pdf/2601.10657v2
+
+## 21463. Multi-Property Synthesis
+
+- **ID:** `2601.10651` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Christoph Weinhuber, Yannik Schnitzer, Alessandro Abate, et al.
+- **abs:** https://arxiv.org/abs/2601.10651 · **PDF:** https://arxiv.org/pdf/2601.10651v1
+
+## 21464. Slang Context-based Inference Enhancement via Greedy Search-Guided Chain-of-Thought Prompting
+
+- **ID:** `2603.13230` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Jinghan Cao, Qingyang Ren, Xiangyun Chen, et al.
+- **abs:** https://arxiv.org/abs/2603.13230 · **PDF:** https://arxiv.org/pdf/2603.13230v1
+
+## 21465. Influential Training Data Retrieval for Explaining Verbalized Confidence of LLMs
+
+- **ID:** `2601.10645` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Yuxi Xia, Loris Schoenegger, Benjamin Roth
+- **abs:** https://arxiv.org/abs/2601.10645 · **PDF:** https://arxiv.org/pdf/2601.10645v1
+
+## 21466. AI Agents Need Memory Control Over More Context
+
+- **ID:** `2601.11653` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-15
+- **Authors:** Fouad Bousetouane
+- **abs:** https://arxiv.org/abs/2601.11653 · **PDF:** https://arxiv.org/pdf/2601.11653v1
+
+## 21467. Adjusted Similarity Measures and a Violation of Expectations
+
+- **ID:** `2601.10641` · **primary:** `stat.ME` · **Submitted:** 2026-01-15
+- **Authors:** William L. Lippitt, Edward J. Bedrick, Nichole E. Carlson
+- **abs:** https://arxiv.org/abs/2601.10641 · **PDF:** https://arxiv.org/pdf/2601.10641v1
+
+## 21468. STEM: Scaling Transformers with Embedding Modules
+
+- **ID:** `2601.10639` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Ranajoy Sadhukhan, Sheng Cao, Harry Dong, et al.
+- **abs:** https://arxiv.org/abs/2601.10639 · **PDF:** https://arxiv.org/pdf/2601.10639v1
+
+## 21469. Classification Imbalance as Transfer Learning
+
+- **ID:** `2601.10630` · **primary:** `stat.ML` · **Submitted:** 2026-01-15
+- **Authors:** Eric Xia, Jason M. Klusowski
+- **abs:** https://arxiv.org/abs/2601.10630 · **PDF:** https://arxiv.org/pdf/2601.10630v1
+
+## 21470. Parametric RDT approach to computational gap of symmetric binary perceptron
+
+- **ID:** `2601.10628` · **primary:** `stat.ML` · **Submitted:** 2026-01-15
+- **Authors:** Mihailo Stojnic
+- **abs:** https://arxiv.org/abs/2601.10628 · **PDF:** https://arxiv.org/pdf/2601.10628v1
+
+## 21471. Differentially Private Inference for Longitudinal Linear Regression
+
+- **ID:** `2601.10626` · **primary:** `math.ST` · **Submitted:** 2026-01-15
+- **Authors:** Getoar Sopa, Marco Avella Medina, Cynthia Rush
+- **abs:** https://arxiv.org/abs/2601.10626 · **PDF:** https://arxiv.org/pdf/2601.10626v1
+
+## 21472. Fair Regression under Demographic Parity: A Unified Framework
+
+- **ID:** `2601.10623` · **primary:** `stat.ME` · **Submitted:** 2026-01-15
+- **Authors:** Yongzhen Feng, Weiwei Wang, Raymond K. W. Wong, et al.
+- **abs:** https://arxiv.org/abs/2601.10623 · **PDF:** https://arxiv.org/pdf/2601.10623v1
+
+## 21473. Molmo2: Open Weights and Data for Vision-Language Models with Video Understanding and Grounding
+
+- **ID:** `2601.10611` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Christopher Clark, Jieyu Zhang, Zixian Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.10611 · **PDF:** https://arxiv.org/pdf/2601.10611v4
+
+## 21474. Procedural Fairness in Multi-Agent Bandits
+
+- **ID:** `2601.10600` · **primary:** `cs.MA` · **Submitted:** 2026-01-15
+- **Authors:** Joshua Caiata, Carter Blair, Kate Larson
+- **abs:** https://arxiv.org/abs/2601.10600 · **PDF:** https://arxiv.org/pdf/2601.10600v1
+
+## 21475. ProbFM: Probabilistic Time Series Foundation Model with Uncertainty Decomposition
+
+- **ID:** `2601.10591` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Arundeep Chinta, Lucas Vinh Tran, Jay Katukuri
+- **abs:** https://arxiv.org/abs/2601.10591 · **PDF:** https://arxiv.org/pdf/2601.10591v1
+
+## 21476. Be Your Own Red Teamer: Safety Alignment via Self-Play and Reflective Experience Replay
+
+- **ID:** `2601.10589` · **primary:** `cs.CR` · **Submitted:** 2026-01-15
+- **Authors:** Hao Wang, Yanting Wang, Hao Li, et al.
+- **abs:** https://arxiv.org/abs/2601.10589 · **PDF:** https://arxiv.org/pdf/2601.10589v1
+
+## 21477. Searching for Quantum Effects in the Brain: A Bell-Type Test for Nonclassical Latent Representations in Autoencoders
+
+- **ID:** `2601.10588` · **primary:** `quant-ph` · **Submitted:** 2026-01-15
+- **Authors:** I. K. Kominis, C. Xie, S. Li, et al.
+- **abs:** https://arxiv.org/abs/2601.10588 · **PDF:** https://arxiv.org/pdf/2601.10588v1
+
+## 21478. Adversarial Evasion Attacks on Computer Vision using SHAP Values
+
+- **ID:** `2601.10587` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Frank Mollard, Marcus Becker, Florian Roehrbein
+- **abs:** https://arxiv.org/abs/2601.10587 · **PDF:** https://arxiv.org/pdf/2601.10587v1
+
+## 21479. Combinatorial Optimization Augmented Machine Learning
+
+- **ID:** `2601.10583` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Maximilian Schiffer, Heiko Hoppe, Yue Su, et al.
+- **abs:** https://arxiv.org/abs/2601.10583 · **PDF:** https://arxiv.org/pdf/2601.10583v1
+
+## 21480. From Single to Multi-Agent Reasoning: Advancing GeneGPT for Genomics QA
+
+- **ID:** `2601.10581` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Kimia Abedini, Farzad Shami, Gianmaria Silvello
+- **abs:** https://arxiv.org/abs/2601.10581 · **PDF:** https://arxiv.org/pdf/2601.10581v1
+
+## 21481. Form and Meaning in Intrinsic Multilingual Evaluations
+
+- **ID:** `2601.10580` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Wessel Poelman, Miryam de Lhoneux
+- **abs:** https://arxiv.org/abs/2601.10580 · **PDF:** https://arxiv.org/pdf/2601.10580v1
+
+## 21482. WISP: Waste- and Interference-Suppressed Distributed Speculative LLM Serving at the Edge via Dynamic Drafting and SLO-Aware Batching
+
+- **ID:** `2601.11652` · **primary:** `cs.DC` · **Submitted:** 2026-01-15
+- **Authors:** Xiangchen Li, Jiakun Fan, Qingyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.11652 · **PDF:** https://arxiv.org/pdf/2601.11652v1
+
+## 21483. LLMs for Game Theory: Entropy-Guided In-Context Learning and Adaptive CoT Reasoning
+
+- **ID:** `2601.10775` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Tommaso Felice Banfi, Sashenka Gamage
+- **abs:** https://arxiv.org/abs/2601.10775 · **PDF:** https://arxiv.org/pdf/2601.10775v1
+
+## 21484. Generative AI collective behavior needs an interactionist paradigm
+
+- **ID:** `2601.10567` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Laura Ferrarotti, Gian Maria Campedelli, Roberto Dessì, et al.
+- **abs:** https://arxiv.org/abs/2601.10567 · **PDF:** https://arxiv.org/pdf/2601.10567v1
+
+## 21485. Representation-Aware Unlearning via Activation Signatures: From Suppression to Knowledge-Signature Erasure
+
+- **ID:** `2601.10566` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Syed Naveed Mahmood, Md. Rezaur Rahman Bhuiyan, Tasfia Zaman, et al.
+- **abs:** https://arxiv.org/abs/2601.10566 · **PDF:** https://arxiv.org/pdf/2601.10566v4
+
+## 21486. Kolmogorov Arnold Networks and Multi-Layer Perceptrons: A Paradigm Shift in Neural Modelling
+
+- **ID:** `2601.10563` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Aradhya Gaonkar, Nihal Jain, Vignesh Chougule, et al.
+- **abs:** https://arxiv.org/abs/2601.10563 · **PDF:** https://arxiv.org/pdf/2601.10563v1
+
+## 21487. Process-Guided Concept Bottleneck Model
+
+- **ID:** `2601.10562` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Reza M. Asiyabi, SEOSAW Partnership, Steven Hancock, et al.
+- **abs:** https://arxiv.org/abs/2601.10562 · **PDF:** https://arxiv.org/pdf/2601.10562v1
+
+## 21488. Learning Latency-Aware Orchestration for Parallel Multi-Agent Systems
+
+- **ID:** `2601.10560` · **primary:** `cs.MA` · **Submitted:** 2026-01-15
+- **Authors:** Xi Shi, Mengxin Zheng, Qian Lou
+- **abs:** https://arxiv.org/abs/2601.10560 · **PDF:** https://arxiv.org/pdf/2601.10560v1
+
+## 21489. causalfe: Causal Forests with Fixed Effects in Python
+
+- **ID:** `2601.10555` · **primary:** `econ.EM` · **Submitted:** 2026-01-15
+- **Authors:** Harry Aytug
+- **abs:** https://arxiv.org/abs/2601.10555 · **PDF:** https://arxiv.org/pdf/2601.10555v1
+
+## 21490. Defending Large Language Models Against Jailbreak Attacks via In-Decoding Safety-Awareness Probing
+
+- **ID:** `2601.10543` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Yinzhi Zhao, Ming Wang, Shi Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.10543 · **PDF:** https://arxiv.org/pdf/2601.10543v2
+
+## 21491. Analytic Bijections for Smooth and Interpretable Normalizing Flows
+
+- **ID:** `2601.10774` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Mathis Gerdes, Miranda C. N. Cheng
+- **abs:** https://arxiv.org/abs/2601.10774 · **PDF:** https://arxiv.org/pdf/2601.10774v1
+
+## 21492. Mixtures of Transparent Local Models
+
+- **ID:** `2601.10541` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Niffa Cheick Oumar Diaby, Thierry Duchesne, Mario Marchand
+- **abs:** https://arxiv.org/abs/2601.10541 · **PDF:** https://arxiv.org/pdf/2601.10541v1
+
+## 21493. CoGen: Creation of Reusable UI Components in Figma via Textual Commands
+
+- **ID:** `2601.10536` · **primary:** `cs.HC` · **Submitted:** 2026-01-15
+- **Authors:** Ishani Kanapathipillai, Obhasha Priyankara
+- **abs:** https://arxiv.org/abs/2601.10536 · **PDF:** https://arxiv.org/pdf/2601.10536v1
+
+## 21494. PERM: Psychology-grounded Empathetic Reward Modeling for Large Language Models
+
+- **ID:** `2601.10532` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Chengbing Wang, Wuqiang Zheng, Yang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.10532 · **PDF:** https://arxiv.org/pdf/2601.10532v2
+
+## 21495. Coarsening Causal DAG Models
+
+- **ID:** `2601.10531` · **primary:** `stat.ML` · **Submitted:** 2026-01-15
+- **Authors:** Francisco Madaleno, Pratik Misra, Alex Markham
+- **abs:** https://arxiv.org/abs/2601.10531 · **PDF:** https://arxiv.org/pdf/2601.10531v2
+
+## 21496. A Safety Report on GPT-5.2, Gemini 3 Pro, Qwen3-VL, Grok 4.1 Fast, Nano Banana Pro, and Seedream 4.5
+
+- **ID:** `2601.10527` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Xingjun Ma, Yixu Wang, Hengyuan Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.10527 · **PDF:** https://arxiv.org/pdf/2601.10527v2
+
+## 21497. Diagnosing Generalization Failures in Fine-Tuned LLMs: A Cross-Architectural Study on Phishing Detection
+
+- **ID:** `2601.10524` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Frank Bobe, Gregory D. Vetaw, Chase Pavlick, et al.
+- **abs:** https://arxiv.org/abs/2601.10524 · **PDF:** https://arxiv.org/pdf/2601.10524v1
+
+## 21498. Breaking Up with Normatively Monolithic Agency with GRACE: A Reason-Based Neuro-Symbolic Architecture for Safe and Ethical AI Alignment
+
+- **ID:** `2601.10520` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Felix Jahn, Yannic Muskalla, Lisa Dargasz, et al.
+- **abs:** https://arxiv.org/abs/2601.10520 · **PDF:** https://arxiv.org/pdf/2601.10520v2
+
+## 21499. Transformer-Based Cognitive Radio: Adaptive Modulation Strategies Using Transformer Models
+
+- **ID:** `2601.10519` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Andrea Melis, Andrea Piroddi, Roberto Girau
+- **abs:** https://arxiv.org/abs/2601.10519 · **PDF:** https://arxiv.org/pdf/2601.10519v1
+
+## 21500. AEQ-Bench: Measuring Empathy of Omni-Modal Large Models
+
+- **ID:** `2601.10513` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Xuan Luo, Lewei Yao, Libo Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.10513 · **PDF:** https://arxiv.org/pdf/2601.10513v1
+
+## 21501. SatMap: Revisiting Satellite Maps as Prior for Online HD Map Construction
+
+- **ID:** `2601.10512` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Kanak Mazumder, Fabian B. Flohr
+- **abs:** https://arxiv.org/abs/2601.10512 · **PDF:** https://arxiv.org/pdf/2601.10512v2
+
+## 21502. Scalable Algorithms for Approximate DNF Model Counting
+
+- **ID:** `2601.10511` · **primary:** `cs.DS` · **Submitted:** 2026-01-15
+- **Authors:** Paul Burkhardt, David G. Harris, Kevin T Schmitt
+- **abs:** https://arxiv.org/abs/2601.10511 · **PDF:** https://arxiv.org/pdf/2601.10511v1
+
+## 21503. LogicLens: Leveraging Semantic Code Graph to explore Multi Repository large systems
+
+- **ID:** `2601.10773` · **primary:** `cs.SE` · **Submitted:** 2026-01-15
+- **Authors:** Niko Usai, Dario Montagnini, Kristian Ilianov Iliev, et al.
+- **abs:** https://arxiv.org/abs/2601.10773 · **PDF:** https://arxiv.org/pdf/2601.10773v1
+
+## 21504. DR-Arena: an Automated Evaluation Framework for Deep Research Agents
+
+- **ID:** `2601.10504` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Yiwen Gao, Ruochen Zhao, Yang Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.10504 · **PDF:** https://arxiv.org/pdf/2601.10504v1
+
+## 21505. Aesthetics as Structural Harm: Algorithmic Lookism Across Text-to-Image Generation and Classification
+
+- **ID:** `2601.11651` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Miriam Doh, Aditya Gulati, Corinna Canali, et al.
+- **abs:** https://arxiv.org/abs/2601.11651 · **PDF:** https://arxiv.org/pdf/2601.11651v2
+
+## 21506. PROMA: Projected Microbatch Accumulation for Reference-Free Proximal Policy Updates
+
+- **ID:** `2601.10498` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Nilin Abrahamsen
+- **abs:** https://arxiv.org/abs/2601.10498 · **PDF:** https://arxiv.org/pdf/2601.10498v4
+
+## 21507. Model See, Model Do? Exposure-Aware Evaluation of Bug-vs-Fix Preference in Code LLMs
+
+- **ID:** `2601.10496` · **primary:** `cs.SE` · **Submitted:** 2026-01-15
+- **Authors:** Ali Al-Kaswan, Claudio Spiess, Prem Devanbu, et al.
+- **abs:** https://arxiv.org/abs/2601.10496 · **PDF:** https://arxiv.org/pdf/2601.10496v1
+
+## 21508. CROCS: A Two-Stage Clustering Framework for Behaviour-Centric Consumer Segmentation with Smart Meter Data
+
+- **ID:** `2601.10494` · **primary:** `stat.ML` · **Submitted:** 2026-01-15
+- **Authors:** Luke W. Yerbury, Ricardo J. G. B. Campello, G. C. Livingston, et al.
+- **abs:** https://arxiv.org/abs/2601.10494 · **PDF:** https://arxiv.org/pdf/2601.10494v1
+
+## 21509. Communication-Efficient Federated Learning by Exploiting Spatio-Temporal Correlations of Gradients
+
+- **ID:** `2601.10491` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Shenlong Zheng, Zhen Zhang, Yuhui Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.10491 · **PDF:** https://arxiv.org/pdf/2601.10491v1
+
+## 21510. Panning for Gold: Expanding Domain-Specific Knowledge Graphs with General Knowledge
+
+- **ID:** `2601.10485` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Runhao Zhao, Weixin Zeng, Wentao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.10485 · **PDF:** https://arxiv.org/pdf/2601.10485v3
+
+## 21511. High-Dimensional Analysis of Gradient Flow for Extensive-Width Quadratic Neural Networks
+
+- **ID:** `2601.10483` · **primary:** `math.OC` · **Submitted:** 2026-01-15
+- **Authors:** Simon Martin, Giulio Biroli, Francis Bach
+- **abs:** https://arxiv.org/abs/2601.10483 · **PDF:** https://arxiv.org/pdf/2601.10483v1
+
+## 21512. H-EFT-VA: An Effective-Field-Theory Variational Ansatz with Provable Barren Plateau Avoidance
+
+- **ID:** `2601.10479` · **primary:** `quant-ph` · **Submitted:** 2026-01-15
+- **Authors:** Eyad I. B Hamid
+- **abs:** https://arxiv.org/abs/2601.10479 · **PDF:** https://arxiv.org/pdf/2601.10479v1
+
+## 21513. Urban Socio-Semantic Segmentation with Vision-Language Reasoning
+
+- **ID:** `2601.10477` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Yu Wang, Yi Wang, Rui Dai, et al.
+- **abs:** https://arxiv.org/abs/2601.10477 · **PDF:** https://arxiv.org/pdf/2601.10477v2
+
+## 21514. DeFlow: Decoupling Manifold Modeling and Value Maximization for Offline Policy Extraction
+
+- **ID:** `2601.10471` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Zhancun Mu
+- **abs:** https://arxiv.org/abs/2601.10471 · **PDF:** https://arxiv.org/pdf/2601.10471v2
+
+## 21515. Evaluating the Evolution of Critical Thinking, Creativity, Communication and Collaboration in Higher Education Courses
+
+- **ID:** `2601.17018` · **primary:** `cs.CY` · **Submitted:** 2026-01-15
+- **Authors:** Margarida Romero
+- **abs:** https://arxiv.org/abs/2601.17018 · **PDF:** https://arxiv.org/pdf/2601.17018v1
+
+## 21516. ChartComplete: A Taxonomy-based Inclusive Chart Dataset
+
+- **ID:** `2601.10462` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Ahmad Mustapha, Charbel Toumieh, Mariette Awad
+- **abs:** https://arxiv.org/abs/2601.10462 · **PDF:** https://arxiv.org/pdf/2601.10462v3
+
+## 21517. Contextual StereoSet: Stress-Testing Bias Alignment Robustness in Large Language Models
+
+- **ID:** `2601.10460` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Abhinaba Basu, Pavan Chakraborty
+- **abs:** https://arxiv.org/abs/2601.10460 · **PDF:** https://arxiv.org/pdf/2601.10460v1
+
+## 21518. LangLasso: Interactive Cluster Descriptions through LLM Explanation
+
+- **ID:** `2601.10458` · **primary:** `cs.HC` · **Submitted:** 2026-01-15
+- **Authors:** Raphael Buchmüller, Dennis Collaris, Linhao Meng, et al.
+- **abs:** https://arxiv.org/abs/2601.10458 · **PDF:** https://arxiv.org/pdf/2601.10458v1
+
+## 21519. NSR-Boost: A Neuro-Symbolic Residual Boosting Framework for Industrial Legacy Models
+
+- **ID:** `2601.10457` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Ziming Dai, Dabiao Ma, Jinle Tong, et al.
+- **abs:** https://arxiv.org/abs/2601.10457 · **PDF:** https://arxiv.org/pdf/2601.10457v2
+
+## 21520. SurgGoal: Rethinking Surgical Planning Evaluation via Goal-Satisfiability
+
+- **ID:** `2601.10455` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Ruochen Li, Kun Yuan, Yufei Xia, et al.
+- **abs:** https://arxiv.org/abs/2601.10455 · **PDF:** https://arxiv.org/pdf/2601.10455v1
+
+## 21521. Physically constrained unfolded multi-dimensional OMP for large MIMO systems
+
+- **ID:** `2601.10771` · **primary:** `eess.SP` · **Submitted:** 2026-01-15
+- **Authors:** Nay Klaimi, Clément Elvira, Philippe Mary, et al.
+- **abs:** https://arxiv.org/abs/2601.10771 · **PDF:** https://arxiv.org/pdf/2601.10771v1
+
+## 21522. Stable Differentiable Modal Synthesis for Learning Nonlinear Dynamics
+
+- **ID:** `2601.10453` · **primary:** `cs.SD` · **Submitted:** 2026-01-15
+- **Authors:** Victor Zheleznov, Stefan Bilbao, Alec Wright, et al.
+- **abs:** https://arxiv.org/abs/2601.10453 · **PDF:** https://arxiv.org/pdf/2601.10453v3
+
+## 21523. AgentGuardian: Learning Access Control Policies to Govern AI Agent Behavior
+
+- **ID:** `2601.10440` · **primary:** `cs.CR` · **Submitted:** 2026-01-15
+- **Authors:** Nadya Abaev, Denis Klimov, Gerard Levinov, et al.
+- **abs:** https://arxiv.org/abs/2601.10440 · **PDF:** https://arxiv.org/pdf/2601.10440v1
+
+## 21524. Development of Ontological Knowledge Bases by Leveraging Large Language Models
+
+- **ID:** `2601.10436` · **primary:** `cs.IR` · **Submitted:** 2026-01-15
+- **Authors:** Le Ngoc Luyen, Marie-Hélène Abel, Philippe Gouspillou
+- **abs:** https://arxiv.org/abs/2601.10436 · **PDF:** https://arxiv.org/pdf/2601.10436v2
+
+## 21525. Are Language Models Models?
+
+- **ID:** `2601.10421` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Philip Resnik
+- **abs:** https://arxiv.org/abs/2601.10421 · **PDF:** https://arxiv.org/pdf/2601.10421v1
+
+## 21526. Reinforcement Learning with Multi-Step Lookahead Information Via Adaptive Batching
+
+- **ID:** `2601.10418` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Nadav Merlis
+- **abs:** https://arxiv.org/abs/2601.10418 · **PDF:** https://arxiv.org/pdf/2601.10418v1
+
+## 21527. LLMdoctor: Token-Level Flow-Guided Preference Optimization for Efficient Test-Time Alignment of Large Language Models
+
+- **ID:** `2601.10416` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Tiesunlong Shen, Rui Mao, Jin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.10416 · **PDF:** https://arxiv.org/pdf/2601.10416v1
+
+## 21528. LADFA: A Framework of Using Large Language Models and Retrieval-Augmented Generation for Personal Data Flow Analysis in Privacy Policies
+
+- **ID:** `2601.10413` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Haiyue Yuan, Nikolay Matyunin, Ali Raza, et al.
+- **abs:** https://arxiv.org/abs/2601.10413 · **PDF:** https://arxiv.org/pdf/2601.10413v1
+
+## 21529. TF3-RO-50M: Training Compact Romanian Language Models from Scratch on Synthetic Moral Microfiction
+
+- **ID:** `2601.10410` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Mihai Dan Nadas, Laura Diosan, Andreea Tomescu, et al.
+- **abs:** https://arxiv.org/abs/2601.10410 · **PDF:** https://arxiv.org/pdf/2601.10410v1
+
+## 21530. CS-GBA: A Critical Sample-based Gradient-guided Backdoor Attack for Offline Reinforcement Learning
+
+- **ID:** `2601.10407` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Yuanjie Zhao, Junnan Qiu, Yue Ding, et al.
+- **abs:** https://arxiv.org/abs/2601.10407 · **PDF:** https://arxiv.org/pdf/2601.10407v1
+
+## 21531. ErrEval: Error-Aware Evaluation for Question Generation through Explicit Diagnostics
+
+- **ID:** `2601.10406` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Weiping Fu, Bifan Wei, Jingyi Hao, et al.
+- **abs:** https://arxiv.org/abs/2601.10406 · **PDF:** https://arxiv.org/pdf/2601.10406v1
+
+## 21532. Discrete Feynman-Kac Correctors
+
+- **ID:** `2601.10403` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Mohsin Hasan, Viktor Ohanesian, Artem Gazizov, et al.
+- **abs:** https://arxiv.org/abs/2601.10403 · **PDF:** https://arxiv.org/pdf/2601.10403v1
+
+## 21533. Toward Ultra-Long-Horizon Agentic Science: Cognitive Accumulation for Machine Learning Engineering
+
+- **ID:** `2601.10402` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Xinyu Zhu, Yuzhu Cai, Zexi Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.10402 · **PDF:** https://arxiv.org/pdf/2601.10402v5
+
+## 21534. LatentRefusal: Latent-Signal Refusal for Unanswerable Text-to-SQL Queries
+
+- **ID:** `2601.10398` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Xuancheng Ren, Shijing Hu, Zhihui Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.10398 · **PDF:** https://arxiv.org/pdf/2601.10398v2
+
+## 21535. Unifying Speech Recognition, Synthesis and Conversion with Autoregressive Transformers
+
+- **ID:** `2601.10770` · **primary:** `cs.SD` · **Submitted:** 2026-01-15
+- **Authors:** Runyuan Cai, Yu Lin, Yiming Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.10770 · **PDF:** https://arxiv.org/pdf/2601.10770v1
+
+## 21536. INDIC DIALECT: A Multi Task Benchmark to Evaluate and Translate in Indian Language Dialects
+
+- **ID:** `2601.10388` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Tarun Sharma, Manikandan Ravikiran, Sourava Kumar Behera, et al.
+- **abs:** https://arxiv.org/abs/2601.10388 · **PDF:** https://arxiv.org/pdf/2601.10388v1
+
+## 21537. The Assistant Axis: Situating and Stabilizing the Default Persona of Language Models
+
+- **ID:** `2601.10387` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Christina Lu, Jack Gallagher, Jonathan Michala, et al.
+- **abs:** https://arxiv.org/abs/2601.10387 · **PDF:** https://arxiv.org/pdf/2601.10387v1
+
+## 21538. Handling Missing Modalities in Multimodal Survival Prediction for Non-Small Cell Lung Cancer
+
+- **ID:** `2601.10386` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Filippo Ruffini, Camillo Maria Caruso, Claudia Tacconi, et al.
+- **abs:** https://arxiv.org/abs/2601.10386 · **PDF:** https://arxiv.org/pdf/2601.10386v1
+
+## 21539. Global Context Compression with Interleaved Vision-Text Transformation
+
+- **ID:** `2601.10378` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Dian Jiao, Jiaxin Duan, Shuai Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.10378 · **PDF:** https://arxiv.org/pdf/2601.10378v2
+
+## 21540. Towards Efficient Low-rate Image Compression with Frequency-aware Diffusion Prior Refinement
+
+- **ID:** `2601.10373` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Yichong Xia, Yimin Zhou, Jinpeng Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.10373 · **PDF:** https://arxiv.org/pdf/2601.10373v1
+
+## 21541. PLGC: Pseudo-Labeled Graph Condensation
+
+- **ID:** `2601.10358` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Jay Nandy, Arnab Kumar Mondal, Anuj Rathore, et al.
+- **abs:** https://arxiv.org/abs/2601.10358 · **PDF:** https://arxiv.org/pdf/2601.10358v1
+
+## 21542. EvoMorph: Counterfactual Explanations for Continuous Time-Series Extrinsic Regression Applied to Photoplethysmography
+
+- **ID:** `2601.10356` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Mesut Ceylan, Alexis Tabin, Patrick Langer, et al.
+- **abs:** https://arxiv.org/abs/2601.10356 · **PDF:** https://arxiv.org/pdf/2601.10356v1
+
+## 21543. Unlocking Implicit Experience: Synthesizing Tool-Use Trajectories from Text
+
+- **ID:** `2601.10355` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Zhihao Xu, Rumei Li, Jiahuan Li, et al.
+- **abs:** https://arxiv.org/abs/2601.10355 · **PDF:** https://arxiv.org/pdf/2601.10355v1
+
+## 21544. SuS: Strategy-aware Surprise for Intrinsic Exploration
+
+- **ID:** `2601.10349` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Mark Kashirskiy, Ilya Makarov
+- **abs:** https://arxiv.org/abs/2601.10349 · **PDF:** https://arxiv.org/pdf/2601.10349v1
+
+## 21545. Training-Trajectory-Aware Token Selection
+
+- **ID:** `2601.10348` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Zhanming Shen, Jiaqi Hu, Zeyu Qin, et al.
+- **abs:** https://arxiv.org/abs/2601.10348 · **PDF:** https://arxiv.org/pdf/2601.10348v1
+
+## 21546. OctoBench: Benchmarking Scaffold-Aware Instruction Following in Repository-Grounded Agentic Coding
+
+- **ID:** `2601.10343` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Deming Ding, Shichun Liu, Enhui Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.10343 · **PDF:** https://arxiv.org/pdf/2601.10343v2
+
+## 21547. C-GRASP: Clinically-Grounded Reasoning for Affective Signal Processing
+
+- **ID:** `2601.10342` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Cheng Lin Cheng, Ting Chuan Lin, Chai Kai Chang
+- **abs:** https://arxiv.org/abs/2601.10342 · **PDF:** https://arxiv.org/pdf/2601.10342v1
+
+## 21548. Agent Skills in the Wild: An Empirical Study of Security Vulnerabilities at Scale
+
+- **ID:** `2601.10338` · **primary:** `cs.CR` · **Submitted:** 2026-01-15
+- **Authors:** Yi Liu, Weizhe Wang, Ruitao Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.10338 · **PDF:** https://arxiv.org/pdf/2601.10338v1
+
+## 21549. An analytic theory of convolutional neural network inverse problems solvers
+
+- **ID:** `2601.10334` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Minh Hai Nguyen, Quoc Bao Do, Edouard Pauwels, et al.
+- **abs:** https://arxiv.org/abs/2601.10334 · **PDF:** https://arxiv.org/pdf/2601.10334v1
+
+## 21550. Large Language Model Agent for User-friendly Chemical Process Simulations
+
+- **ID:** `2601.11650` · **primary:** `physics.chem-ph` · **Submitted:** 2026-01-15
+- **Authors:** Jingkang Liang, Niklas Groll, Gürkan Sin
+- **abs:** https://arxiv.org/abs/2601.11650 · **PDF:** https://arxiv.org/pdf/2601.11650v2
+
+## 21551. Meta Dynamic Graph for Traffic Flow Prediction
+
+- **ID:** `2601.10328` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Yiqing Zou, Hanning Yuan, Qianyu Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.10328 · **PDF:** https://arxiv.org/pdf/2601.10328v1
+
+## 21552. ROMA: Real-time Omni-Multimodal Assistant with Interactive Streaming Understanding
+
+- **ID:** `2601.10323` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Xueyun Tian, Wei Li, Bingbing Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.10323 · **PDF:** https://arxiv.org/pdf/2601.10323v1
+
+## 21553. An Efficient Long-Context Ranking Architecture With Calibrated LLM Distillation: Application to Person-Job Fit
+
+- **ID:** `2601.10321` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Warren Jouanneau, Emma Jouffroy, Marc Palyart
+- **abs:** https://arxiv.org/abs/2601.10321 · **PDF:** https://arxiv.org/pdf/2601.10321v2
+
+## 21554. Boundary-Aware NL2SQL: Integrating Reliability through Hybrid Reward and Data Synthesis
+
+- **ID:** `2601.10318` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Songsong Tian, Kongsheng Zhuo, Zhendong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.10318 · **PDF:** https://arxiv.org/pdf/2601.10318v1
+
+## 21555. ADVOSYNTH: A Synthetic Multi-Advocate Dataset for Speaker Identification in Courtroom Scenarios
+
+- **ID:** `2601.10315` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Aniket Deroy
+- **abs:** https://arxiv.org/abs/2601.10315 · **PDF:** https://arxiv.org/pdf/2601.10315v1
+
+## 21556. We Need a More Robust Classifier: Dual Causal Learning Empowers Domain-Incremental Time Series Classification
+
+- **ID:** `2601.10312` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Zhipeng Liu, Peibo Duan, Xuan Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.10312 · **PDF:** https://arxiv.org/pdf/2601.10312v1
+
+## 21557. Multilinguality as Sense Adaptation
+
+- **ID:** `2601.10310` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Jan Christian Blaise Cruz, David Ifeoluwa Adelani, Alham Fikri Aji
+- **abs:** https://arxiv.org/abs/2601.10310 · **PDF:** https://arxiv.org/pdf/2601.10310v1
+
+## 21558. The Straight and Narrow: Do LLMs Possess an Internal Moral Path?
+
+- **ID:** `2601.10307` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Luoming Hu, Jingjie Zeng, Liang Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.10307 · **PDF:** https://arxiv.org/pdf/2601.10307v1
+
+## 21559. Evidence-Augmented Policy Optimization with Reward Co-Evolution for Long-Context Reasoning
+
+- **ID:** `2601.10306` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Xin Guan, Zijian Li, Shen Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.10306 · **PDF:** https://arxiv.org/pdf/2601.10306v1
+
+## 21560. DanQing: An Up-to-Date Large-Scale Chinese Vision-Language Pre-training Dataset
+
+- **ID:** `2601.10305` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Hengyu Shen, Tiancheng Gu, Bin Qin, et al.
+- **abs:** https://arxiv.org/abs/2601.10305 · **PDF:** https://arxiv.org/pdf/2601.10305v3
+
+## 21561. Optimisation of complex product innovation processes based on trend models with three-valued logic
+
+- **ID:** `2601.10768` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Nina Bočková, Barbora Volná, Mirko Dohnal
+- **abs:** https://arxiv.org/abs/2601.10768 · **PDF:** https://arxiv.org/pdf/2601.10768v1
+
+## 21562. SPIKE: Sparse Koopman Regularization for Physics-Informed Neural Networks
+
+- **ID:** `2601.10282` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Jose Marie Antonio Miñoza
+- **abs:** https://arxiv.org/abs/2601.10282 · **PDF:** https://arxiv.org/pdf/2601.10282v2
+
+## 21563. Queueing-Aware Optimization of Reasoning Tokens for Accuracy-Latency Trade-offs in LLM Servers
+
+- **ID:** `2601.10274` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Emre Ozbas, Melih Bastopcu
+- **abs:** https://arxiv.org/abs/2601.10274 · **PDF:** https://arxiv.org/pdf/2601.10274v1
+
+## 21564. MoST: Mixing Speech and Text with Modality-Aware Mixture of Experts
+
+- **ID:** `2601.10272` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Yuxuan Lou, Kai Yang, Yang You
+- **abs:** https://arxiv.org/abs/2601.10272 · **PDF:** https://arxiv.org/pdf/2601.10272v1
+
+## 21565. Early Fault Detection on CMAPSS with Unsupervised LSTM Autoencoders
+
+- **ID:** `2601.10269` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** P. Sánchez, K. Reyes, B. Radu, et al.
+- **abs:** https://arxiv.org/abs/2601.10269 · **PDF:** https://arxiv.org/pdf/2601.10269v1
+
+## 21566. In-Context Source and Channel Coding
+
+- **ID:** `2601.10267` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Ziqiong Wang, Tianqi Ren, Rongpeng Li, et al.
+- **abs:** https://arxiv.org/abs/2601.10267 · **PDF:** https://arxiv.org/pdf/2601.10267v1
+
+## 21567. Measuring Affinity between Attention-Head Weight Subspaces via the Projection Kernel
+
+- **ID:** `2601.10266` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Hiroaki Yamagiwa, Yusuke Takase, Hidetoshi Shimodaira
+- **abs:** https://arxiv.org/abs/2601.10266 · **PDF:** https://arxiv.org/pdf/2601.10266v1
+
+## 21568. Sim2Real Deep Transfer for Per-Device CFO Calibration
+
+- **ID:** `2601.10264` · **primary:** `eess.SP` · **Submitted:** 2026-01-15
+- **Authors:** Jingze Zheng, Zhiguo Shi, Shibo He, et al.
+- **abs:** https://arxiv.org/abs/2601.10264 · **PDF:** https://arxiv.org/pdf/2601.10264v1
+
+## 21569. An Ensemble of Evolutionary Algorithms With Both Crisscross Search and Sparrow Search for Processing Inferior Individuals
+
+- **ID:** `2601.10263` · **primary:** `cs.NE` · **Submitted:** 2026-01-15
+- **Authors:** Mingxuan Du, Tingzhang Luo, Ziyang Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.10263 · **PDF:** https://arxiv.org/pdf/2601.10263v1
+
+## 21570. Untangling Input Language from Reasoning Language: A Diagnostic Framework for Cross-Lingual Moral Alignment in LLMs
+
+- **ID:** `2601.10257` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Nan Li, Bo Kang, Tijl De Bie
+- **abs:** https://arxiv.org/abs/2601.10257 · **PDF:** https://arxiv.org/pdf/2601.10257v1
+
+## 21571. NoReGeo: Non-Reasoning Geometry Benchmark
+
+- **ID:** `2601.10254` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Irina Abdullaeva, Anton Vasiliuk, Elizaveta Goncharova, et al.
+- **abs:** https://arxiv.org/abs/2601.10254 · **PDF:** https://arxiv.org/pdf/2601.10254v1
+
+## 21572. X-SAM: Boosting Sharpness-Aware Minimization with Dominant-Eigenvector Gradient Correction
+
+- **ID:** `2601.10251` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Hongru Duan, Yongle Chen, Lei Guan
+- **abs:** https://arxiv.org/abs/2601.10251 · **PDF:** https://arxiv.org/pdf/2601.10251v1
+
+## 21573. Beyond the Einstein-Bohr Debate: Cognitive Complementarity and the Emergence of Quantum Intuition
+
+- **ID:** `2601.15314` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-15
+- **Authors:** Lalit Kumar Shukla
+- **abs:** https://arxiv.org/abs/2601.15314 · **PDF:** https://arxiv.org/pdf/2601.15314v1
+
+## 21574. coTherapist: A Behavior-Aligned Small Language Model to Support Mental Healthcare Experts
+
+- **ID:** `2601.10246` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Prottay Kumar Adhikary, Reena Rawat, Tanmoy Chakraborty
+- **abs:** https://arxiv.org/abs/2601.10246 · **PDF:** https://arxiv.org/pdf/2601.10246v2
+
+## 21575. TRIM: Hybrid Inference via Targeted Stepwise Routing in Multi-Step Reasoning Tasks
+
+- **ID:** `2601.10245` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Vansh Kapoor, Aman Gupta, Hao Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.10245 · **PDF:** https://arxiv.org/pdf/2601.10245v1
+
+## 21576. Loop as a Bridge: Can Looped Transformers Truly Link Representation Space and Natural Language Outputs?
+
+- **ID:** `2601.10242` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Guanxu Chen, Dongrui Liu, Jing Shao
+- **abs:** https://arxiv.org/abs/2601.10242 · **PDF:** https://arxiv.org/pdf/2601.10242v1
+
+## 21577. Fundamental Limitations of Favorable Privacy-Utility Guarantees for DP-SGD
+
+- **ID:** `2601.10237` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Murat Bilgehan Ertan, Marten van Dijk
+- **abs:** https://arxiv.org/abs/2601.10237 · **PDF:** https://arxiv.org/pdf/2601.10237v1
+
+## 21578. Who Owns the Text? Design Patterns for Preserving Authorship in AI-Assisted Writing
+
+- **ID:** `2601.10236` · **primary:** `cs.HC` · **Submitted:** 2026-01-15
+- **Authors:** Bohan Zhang, Chengke Bu, Paramveer S. Dhillon
+- **abs:** https://arxiv.org/abs/2601.10236 · **PDF:** https://arxiv.org/pdf/2601.10236v1
+
+## 21579. GeoSteer: Faithful Chain-of-Thought Steering via Latent Manifold Gradients
+
+- **ID:** `2601.10229` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Kentaro Kazama, Daiki Shirafuji, Tatsuhiko Saito
+- **abs:** https://arxiv.org/abs/2601.10229 · **PDF:** https://arxiv.org/pdf/2601.10229v2
+
+## 21580. Introduction to optimization methods for training SciML models
+
+- **ID:** `2601.10222` · **primary:** `math.NA` · **Submitted:** 2026-01-15
+- **Authors:** Alena Kopaničáková, Elisa Riccietti
+- **abs:** https://arxiv.org/abs/2601.10222 · **PDF:** https://arxiv.org/pdf/2601.10222v1
+
+## 21581. Topo-RAG: Topology-aware retrieval for hybrid text-table documents
+
+- **ID:** `2601.10215` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Alex Dantart, Marco Kóvacs-Navarro
+- **abs:** https://arxiv.org/abs/2601.10215 · **PDF:** https://arxiv.org/pdf/2601.10215v1
+
+## 21582. PADER: Paillier-based Secure Decentralized Social Recommendation
+
+- **ID:** `2601.10212` · **primary:** `cs.CR` · **Submitted:** 2026-01-15
+- **Authors:** Chaochao Chen, Jiaming Qian, Fei Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.10212 · **PDF:** https://arxiv.org/pdf/2601.10212v1
+
+## 21583. One Instruction Does Not Fit All: How Well Do Embeddings Align Personas and Instructions in Low-Resource Indian Languages?
+
+- **ID:** `2601.10205` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Arya Shah, Himanshu beniwal, Mayank Singh
+- **abs:** https://arxiv.org/abs/2601.10205 · **PDF:** https://arxiv.org/pdf/2601.10205v1
+
+## 21584. PRL: Process Reward Learning Improves LLMs' Reasoning Ability and Broadens the Reasoning Boundary
+
+- **ID:** `2601.10201` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Jiarui Yao, Ruida Wang, Tong Zhang
+- **abs:** https://arxiv.org/abs/2601.10201 · **PDF:** https://arxiv.org/pdf/2601.10201v1
+
+## 21585. Deep g-Pricing for CSI 300 Index Options with Volatility Trajectories and Market Sentiment
+
+- **ID:** `2601.18804` · **primary:** `q-fin.CP` · **Submitted:** 2026-01-15
+- **Authors:** Yilun Zhang, Zheng Tang, Hexiang Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.18804 · **PDF:** https://arxiv.org/pdf/2601.18804v1
+
+## 21586. Graph Regularized PCA
+
+- **ID:** `2601.10199` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Antonio Briola, Marwin Schmidt, Fabio Caccioli, et al.
+- **abs:** https://arxiv.org/abs/2601.10199 · **PDF:** https://arxiv.org/pdf/2601.10199v1
+
+## 21587. HumanLLM: Benchmarking and Improving LLM Anthropomorphism via Human Cognitive Patterns
+
+- **ID:** `2601.10198` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Xintao Wang, Jian Yang, Weiyuan Li, et al.
+- **abs:** https://arxiv.org/abs/2601.10198 · **PDF:** https://arxiv.org/pdf/2601.10198v2
+
+## 21588. GFM4GA: Graph Foundation Model for Group Anomaly Detection
+
+- **ID:** `2601.10193` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Jiujiu Chen, Weijun Zeng, Shaofeng Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.10193 · **PDF:** https://arxiv.org/pdf/2601.10193v1
+
+## 21589. How does downsampling affect needle electromyography signals? A generalisable workflow for understanding downsampling effects on high-frequency time series
+
+- **ID:** `2601.10191` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Mathieu Cherpitel, Janne Luijten, Thomas Bäck, et al.
+- **abs:** https://arxiv.org/abs/2601.10191 · **PDF:** https://arxiv.org/pdf/2601.10191v1
+
+## 21590. HOMURA: Taming the Sand-Glass for Time-Constrained LLM Translation via Reinforcement Learning
+
+- **ID:** `2601.10187` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Ziang Cui, Mengran Yu, Tianjiao Li, et al.
+- **abs:** https://arxiv.org/abs/2601.10187 · **PDF:** https://arxiv.org/pdf/2601.10187v2
+
+## 21591. Reinforcement Learning to Discover a North-East Monsoon Index for Rainfall Prediction in Thailand
+
+- **ID:** `2601.10181` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Kiattikun Chobtham
+- **abs:** https://arxiv.org/abs/2601.10181 · **PDF:** https://arxiv.org/pdf/2601.10181v5
+
+## 21592. Bias in the Shadows: Explore Shortcuts in Encrypted Network Traffic Classification
+
+- **ID:** `2601.10180` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Chuyi Wang, Xiaohui Xie, Tongze Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.10180 · **PDF:** https://arxiv.org/pdf/2601.10180v1
+
+## 21593. CC-OR-Net: A Unified Framework for LTV Prediction through Structural Decoupling
+
+- **ID:** `2601.10176` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Mingyu Zhao, Haoran Bai, Yu Tian, et al.
+- **abs:** https://arxiv.org/abs/2601.10176 · **PDF:** https://arxiv.org/pdf/2601.10176v2
+
+## 21594. ReasAlign: Reasoning Enhanced Safety Alignment against Prompt Injection Attack
+
+- **ID:** `2601.10173` · **primary:** `cs.CR` · **Submitted:** 2026-01-15
+- **Authors:** Hao Li, Yankai Yang, G. Edward Suh, et al.
+- **abs:** https://arxiv.org/abs/2601.10173 · **PDF:** https://arxiv.org/pdf/2601.10173v1
+
+## 21595. CtD: Composition through Decomposition in Emergent Communication
+
+- **ID:** `2601.10169` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Boaz Carmeli, Ron Meir, Yonatan Belinkov
+- **abs:** https://arxiv.org/abs/2601.10169 · **PDF:** https://arxiv.org/pdf/2601.10169v1
+
+## 21596. RAG-3DSG: Enhancing 3D Scene Graphs with Re-Shot Guided Retrieval-Augmented Generation
+
+- **ID:** `2601.10168` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Yue Chang, Rufeng Chen, Zhaofan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.10168 · **PDF:** https://arxiv.org/pdf/2601.10168v2
+
+## 21597. Credit C-GPT: A Domain-Specialized Large Language Model for Conversational Understanding in Vietnamese Debt Collection
+
+- **ID:** `2601.10167` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Nhung Nguyen Thi Hong, Cuong Nguyen Dang, Tri Le Ngoc
+- **abs:** https://arxiv.org/abs/2601.10167 · **PDF:** https://arxiv.org/pdf/2601.10167v1
+
+## 21598. AWED-FiNER: Agents, Web applications, and Expert Detectors for Fine-grained Named Entity Recognition across 36 Languages for 6.6 Billion Speakers
+
+- **ID:** `2601.10161` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Prachuryya Kaushik, Ashish Anand
+- **abs:** https://arxiv.org/abs/2601.10161 · **PDF:** https://arxiv.org/pdf/2601.10161v2
+
+## 21599. Alignment Pretraining: AI Discourse Causes Self-Fulfilling (Mis)alignment
+
+- **ID:** `2601.10160` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Cameron Tice, Puria Radmard, Samuel Ratnam, et al.
+- **abs:** https://arxiv.org/abs/2601.10160 · **PDF:** https://arxiv.org/pdf/2601.10160v2
+
+## 21600. What Gets Activated: Uncovering Domain and Driver Experts in MoE Language Models
+
+- **ID:** `2601.10159` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Guimin Hu, Meng Li, Qiwei Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.10159 · **PDF:** https://arxiv.org/pdf/2601.10159v2
+
+## 21601. MMPG: MoE-based Adaptive Multi-Perspective Graph Fusion for Protein Representation Learning
+
+- **ID:** `2601.10157` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Yusong Wang, Jialun Shen, Zhihao Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.10157 · **PDF:** https://arxiv.org/pdf/2601.10157v1
+
+## 21602. ToolSafe: Enhancing Tool Invocation Safety of LLM-based agents via Proactive Step-level Guardrail and Feedback
+
+- **ID:** `2601.10156` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Yutao Mou, Zhangchi Xue, Lijun Li, et al.
+- **abs:** https://arxiv.org/abs/2601.10156 · **PDF:** https://arxiv.org/pdf/2601.10156v1
+
+## 21603. LOOKAT: Lookup-Optimized Key-Attention for Memory-Efficient Transformers
+
+- **ID:** `2601.10155` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Aryan Karmore
+- **abs:** https://arxiv.org/abs/2601.10155 · **PDF:** https://arxiv.org/pdf/2601.10155v1
+
+## 21604. MHub.ai: A Simple, Standardized, and Reproducible Platform for AI Models in Medical Imaging
+
+- **ID:** `2601.10154` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Leonard Nürnberg, Dennis Bontempi, Suraj Pai, et al.
+- **abs:** https://arxiv.org/abs/2601.10154 · **PDF:** https://arxiv.org/pdf/2601.10154v1
+
+## 21605. Simple Network Graph Comparative Learning
+
+- **ID:** `2601.10150` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Qiang Yu, Xinran Cheng, Shiqiang Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.10150 · **PDF:** https://arxiv.org/pdf/2601.10150v1
+
+## 21606. DecisionLLM: Large Language Models for Long Sequence Decision Exploration
+
+- **ID:** `2601.10148` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Xiaowei Lv, Zhilin Zhang, Yijun Li, et al.
+- **abs:** https://arxiv.org/abs/2601.10148 · **PDF:** https://arxiv.org/pdf/2601.10148v1
+
+## 21607. History Is Not Enough: An Adaptive Dataflow System for Financial Time-Series Synthesis
+
+- **ID:** `2601.10143` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Haochong Xia, Yao Long Teng, Regan Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.10143 · **PDF:** https://arxiv.org/pdf/2601.10143v1
+
+## 21608. Understanding and Preserving Safety in Fine-Tuned LLMs
+
+- **ID:** `2601.10141` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Jiawen Zhang, Yangfan Hu, Kejia Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.10141 · **PDF:** https://arxiv.org/pdf/2601.10141v1
+
+## 21609. Step-by-Step Causality: Transparent Causal Discovery with Multi-Agent Tree-Query and Adversarial Confidence Estimation
+
+- **ID:** `2601.10137` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Ziyi Ding, Chenfei Ye-Hao, Zheyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.10137 · **PDF:** https://arxiv.org/pdf/2601.10137v1
+
+## 21610. Is More Context Always Better? Examining LLM Reasoning Capability for Time Interval Prediction
+
+- **ID:** `2601.10132` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Yanan Cao, Farnaz Fallahi, Murali Mohana Krishna Dandu, et al.
+- **abs:** https://arxiv.org/abs/2601.10132 · **PDF:** https://arxiv.org/pdf/2601.10132v2
+
+## 21611. M^4olGen: Multi-Agent, Multi-Stage Molecular Generation under Precise Multi-Property Constraints
+
+- **ID:** `2601.10131` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Yizhan Li, Florence Cloutier, Sifan Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.10131 · **PDF:** https://arxiv.org/pdf/2601.10131v2
+
+## 21612. Redundancy-Driven Top-$k$ Functional Dependency Discovery
+
+- **ID:** `2601.10130` · **primary:** `cs.DB` · **Submitted:** 2026-01-15
+- **Authors:** Xiaolong Wan, Xixian Han
+- **abs:** https://arxiv.org/abs/2601.10130 · **PDF:** https://arxiv.org/pdf/2601.10130v1
+
+## 21613. LaViT: Aligning Latent Visual Thoughts for Multi-modal Reasoning
+
+- **ID:** `2601.10129` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Linquan Wu, Tianxiang Jiang, Yifei Dong, et al.
+- **abs:** https://arxiv.org/abs/2601.10129 · **PDF:** https://arxiv.org/pdf/2601.10129v1
+
+## 21614. Role-Playing Agents Driven by Large Language Models: Current Status, Challenges, and Future Trends
+
+- **ID:** `2601.10122` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Ye Wang, Jiaxing Chen, Hongjiang Xiao
+- **abs:** https://arxiv.org/abs/2601.10122 · **PDF:** https://arxiv.org/pdf/2601.10122v1
+
+## 21615. TopoDIM: One-shot Topology Generation of Diverse Interaction Modes for Multi-Agent Systems
+
+- **ID:** `2601.10120` · **primary:** `cs.MA` · **Submitted:** 2026-01-15
+- **Authors:** Rui Sun, Jie Ding, Chenghua Gong, et al.
+- **abs:** https://arxiv.org/abs/2601.10120 · **PDF:** https://arxiv.org/pdf/2601.10120v1
+
+## 21616. Following the Teacher's Footsteps: Scheduled Checkpoint Distillation for Domain-Specific LLMs
+
+- **ID:** `2601.10114` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Cheng Feng, Chaoliang Zhong, Jun Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.10114 · **PDF:** https://arxiv.org/pdf/2601.10114v1
+
+## 21617. Repository Intelligence Graph: Deterministic Architectural Map for LLM Code Assistants
+
+- **ID:** `2601.10112` · **primary:** `cs.SE` · **Submitted:** 2026-01-15
+- **Authors:** Tsvi Cherny-Shahar, Amiram Yehudai
+- **abs:** https://arxiv.org/abs/2601.10112 · **PDF:** https://arxiv.org/pdf/2601.10112v1
+
+## 21618. Multi-Constrained Evolutionary Molecular Design Framework: An Interpretable Drug Design Method Combining Rule-Based Evolution and Molecular Crossover
+
+- **ID:** `2601.10110` · **primary:** `cs.NE` · **Submitted:** 2026-01-15
+- **Authors:** Shanxian Lin, Wei Xia, Yuichi Nagata, et al.
+- **abs:** https://arxiv.org/abs/2601.10110 · **PDF:** https://arxiv.org/pdf/2601.10110v1
+
+## 21619. Skill-Aware Data Selection and Fine-Tuning for Data-Efficient Reasoning Distillation
+
+- **ID:** `2601.10109` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Lechen Zhang, Yunxiang Zhang, Wei Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.10109 · **PDF:** https://arxiv.org/pdf/2601.10109v1
+
+## 21620. SIN-Bench: Tracing Native Evidence Chains in Long-Context Multimodal Scientific Interleaved Literature
+
+- **ID:** `2601.10108` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Yiming Ren, Junjie Wang, Yuxin Meng, et al.
+- **abs:** https://arxiv.org/abs/2601.10108 · **PDF:** https://arxiv.org/pdf/2601.10108v1
+
+## 21621. MathDoc: Benchmarking Structured Extraction and Active Refusal on Noisy Mathematics Exam Papers
+
+- **ID:** `2601.10104` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Chenyue Zhou, Jiayi Tuo, Shitong Qin, et al.
+- **abs:** https://arxiv.org/abs/2601.10104 · **PDF:** https://arxiv.org/pdf/2601.10104v1
+
+## 21622. FlowAct-R1: Towards Interactive Humanoid Video Generation
+
+- **ID:** `2601.10103` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Lizhen Wang, Yongming Zhu, Zhipeng Ge, et al.
+- **abs:** https://arxiv.org/abs/2601.10103 · **PDF:** https://arxiv.org/pdf/2601.10103v1
+
+## 21623. Measuring Political Stance and Consistency in Large Language Models
+
+- **ID:** `2601.17016` · **primary:** `cs.CY` · **Submitted:** 2026-01-15
+- **Authors:** Salah Feras Alali, Mohammad Nashat Maasfeh, Mucahid Kutlu, et al.
+- **abs:** https://arxiv.org/abs/2601.17016 · **PDF:** https://arxiv.org/pdf/2601.17016v1
+
+## 21624. Matrix as Plan: Structured Logical Reasoning with Feedback-Driven Replanning
+
+- **ID:** `2601.10101` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Ke Chen, Jiandian Zeng, Zihao Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.10101 · **PDF:** https://arxiv.org/pdf/2601.10101v2
+
+## 21625. Multilingual-To-Multimodal (M2M): Unlocking New Languages with Monolingual Text
+
+- **ID:** `2601.10096` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Piyush Singh Pasi
+- **abs:** https://arxiv.org/abs/2601.10096 · **PDF:** https://arxiv.org/pdf/2601.10096v2
+
+## 21626. V-Zero: Self-Improving Multimodal Reasoning with Zero Annotation
+
+- **ID:** `2601.10094` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Han Wang, Yi Yang, Jingyuan Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.10094 · **PDF:** https://arxiv.org/pdf/2601.10094v1
+
+## 21627. LeMoF: Level-guided Multimodal Fusion for Heterogeneous Clinical Data
+
+- **ID:** `2601.10092` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Jongseok Kim, Seongae Kang, Jonghwan Shin, et al.
+- **abs:** https://arxiv.org/abs/2601.10092 · **PDF:** https://arxiv.org/pdf/2601.10092v1
+
+## 21628. Difficulty-guided Sampling: Bridging the Target Gap between Dataset Distillation and Downstream Tasks
+
+- **ID:** `2601.10090` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Mingzhuo Li, Guang Li, Linfeng Ye, et al.
+- **abs:** https://arxiv.org/abs/2601.10090 · **PDF:** https://arxiv.org/pdf/2601.10090v1
+
+## 21629. Bayesian Meta-Analyses Could Be More: A Case Study in Trial of Labor After a Cesarean-section Outcomes and Complications
+
+- **ID:** `2601.10089` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Ashley Klein, Edward Raff, Marcia DesJardin
+- **abs:** https://arxiv.org/abs/2601.10089 · **PDF:** https://arxiv.org/pdf/2601.10089v1
+
+## 21630. State of AI: An Empirical 100 Trillion Token Study with OpenRouter
+
+- **ID:** `2601.10088` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Malika Aubakirova, Alex Atallah, Chris Clark, et al.
+- **abs:** https://arxiv.org/abs/2601.10088 · **PDF:** https://arxiv.org/pdf/2601.10088v1
+
+## 21631. CALM-IT: Generating Realistic Long-Form Motivational Interviewing Dialogues with Dual-Actor Conversational Dynamics Tracking
+
+- **ID:** `2601.10085` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Viet Cuong Nguyen, Nhi Yen Nguyen, Kristin A. Candan, et al.
+- **abs:** https://arxiv.org/abs/2601.10085 · **PDF:** https://arxiv.org/pdf/2601.10085v1
+
+## 21632. Adaptive Label Error Detection: A Bayesian Approach to Mislabeled Data Detection
+
+- **ID:** `2601.10084` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Zan Chaudhry, Noam H. Rotenberg, Brian Caffo, et al.
+- **abs:** https://arxiv.org/abs/2601.10084 · **PDF:** https://arxiv.org/pdf/2601.10084v1
+
+## 21633. Reinforcement Learning for Dynamic Workflow Optimization in CI/CD Pipelines
+
+- **ID:** `2601.11647` · **primary:** `cs.SE` · **Submitted:** 2026-01-15
+- **Authors:** Aniket Abhishek Soni, Milan Parikh, Rashi Nimesh Kumar Dhenia, et al.
+- **abs:** https://arxiv.org/abs/2601.11647 · **PDF:** https://arxiv.org/pdf/2601.11647v1
+
+## 21634. Is MT Ready for the Next Crisis or Pandemic?
+
+- **ID:** `2601.10082` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Vipasha Bansal, Elizabeth Brown, Chelsea Kendrick, et al.
+- **abs:** https://arxiv.org/abs/2601.10082 · **PDF:** https://arxiv.org/pdf/2601.10082v1
+
+## 21635. Deriving Character Logic from Storyline as Codified Decision Trees
+
+- **ID:** `2601.10080` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Letian Peng, Kun Zhou, Longfei Yun, et al.
+- **abs:** https://arxiv.org/abs/2601.10080 · **PDF:** https://arxiv.org/pdf/2601.10080v1
+
+## 21636. Sparse-RL: Breaking the Memory Wall in LLM Reinforcement Learning via Stable Sparse Rollouts
+
+- **ID:** `2601.10079` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Sijia Luo, Xiaokang Zhang, Yuxuan Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.10079 · **PDF:** https://arxiv.org/pdf/2601.10079v2
+
+## 21637. Thinking Like Van Gogh: Structure-Aware Style Transfer via Flow-Guided 3D Gaussian Splatting
+
+- **ID:** `2601.10075` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Zhendong Wang, Lebin Zhou, Jingchuan Xiao, et al.
+- **abs:** https://arxiv.org/abs/2601.10075 · **PDF:** https://arxiv.org/pdf/2601.10075v1
+
+## 21638. ReaMIL: Reasoning- and Evidence-Aware Multiple Instance Learning for Whole-Slide Histopathology
+
+- **ID:** `2601.10073` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Hyun Do Jung, Jungwon Choi, Hwiyoung Kim
+- **abs:** https://arxiv.org/abs/2601.10073 · **PDF:** https://arxiv.org/pdf/2601.10073v1
+
+## 21639. Comparative Evaluation of Deep Learning-Based and WHO-Informed Approaches for Sperm Morphology Assessment
+
+- **ID:** `2601.10070` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Mohammad Abbadi
+- **abs:** https://arxiv.org/abs/2601.10070 · **PDF:** https://arxiv.org/pdf/2601.10070v1
+
+## 21640. Efficient Content-based Recommendation Model Training via Noise-aware Coreset Selection
+
+- **ID:** `2601.10067` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Hung Vinh Tran, Tong Chen, Hechuan Wen, et al.
+- **abs:** https://arxiv.org/abs/2601.10067 · **PDF:** https://arxiv.org/pdf/2601.10067v1
+
+## 21641. Long-Chain Reasoning Distillation via Adaptive Prefix Alignment
+
+- **ID:** `2601.10064` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Zhenghao Liu, Zhuoyang Wu, Xinze Li, et al.
+- **abs:** https://arxiv.org/abs/2601.10064 · **PDF:** https://arxiv.org/pdf/2601.10064v1
+
+## 21642. CoF-T2I: Video Models as Pure Visual Reasoners for Text-to-Image Generation
+
+- **ID:** `2601.10061` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Chengzhuo Tong, Mingkun Chang, Shenglong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.10061 · **PDF:** https://arxiv.org/pdf/2601.10061v1
+
+## 21643. Unlabeled Data Can Provably Enhance In-Context Learning of Transformers
+
+- **ID:** `2601.10058` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Renpu Liu, Jing Yang
+- **abs:** https://arxiv.org/abs/2601.10058 · **PDF:** https://arxiv.org/pdf/2601.10058v1
+
+## 21644. Instruction Finetuning LLaMA-3-8B Model Using LoRA for Financial Named Entity Recognition
+
+- **ID:** `2601.10043` · **primary:** `q-fin.CP` · **Submitted:** 2026-01-15
+- **Authors:** Zhiming Lian
+- **abs:** https://arxiv.org/abs/2601.10043 · **PDF:** https://arxiv.org/pdf/2601.10043v1
+
+## 21645. What Understanding Means in AI-Laden Astronomy
+
+- **ID:** `2601.10038` · **primary:** `astro-ph.IM` · **Submitted:** 2026-01-15
+- **Authors:** Yuan-Sen Ting, André Curtis-Trudel, Siyu Yao
+- **abs:** https://arxiv.org/abs/2601.10038 · **PDF:** https://arxiv.org/pdf/2601.10038v1
+
+## 21646. A Compute and Communication Runtime Model for Loihi 2
+
+- **ID:** `2601.10035` · **primary:** `cs.NE` · **Submitted:** 2026-01-15
+- **Authors:** Jonathan Timcheck, Alessandro Pierro, Sumit Bam Shrestha
+- **abs:** https://arxiv.org/abs/2601.10035 · **PDF:** https://arxiv.org/pdf/2601.10035v2
+
+## 21647. EmplifAI: a Fine-grained Dataset for Japanese Empathetic Medical Dialogues in 28 Emotion Labels
+
+- **ID:** `2601.10033` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Wan Jou She, Lis Kanashiro Pereira, Fei Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.10033 · **PDF:** https://arxiv.org/pdf/2601.10033v1
+
+## 21648. FilDeep: Learning Large Deformations of Elastic-Plastic Solids with Multi-Fidelity Data
+
+- **ID:** `2601.10031` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Jianheng Tang, Shilong Tao, Zhe Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.10031 · **PDF:** https://arxiv.org/pdf/2601.10031v1
+
+## 21649. PaperScout: An Autonomous Agent for Academic Paper Search with Process-Aware Sequence-Level Policy Optimization
+
+- **ID:** `2601.10029` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Tingyue Pan, Jie Ouyang, Mingyue Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.10029 · **PDF:** https://arxiv.org/pdf/2601.10029v2
+
+## 21650. Structured Personality Control and Adaptation for LLM Agents
+
+- **ID:** `2601.10025` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Jinpeng Wang, Xinyu Jia, Wei Wei Heng, et al.
+- **abs:** https://arxiv.org/abs/2601.10025 · **PDF:** https://arxiv.org/pdf/2601.10025v1
+
+## 21651. BPE: Behavioral Profiling Ensemble
+
+- **ID:** `2601.10024` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Yanxin Liu, Yunqi Zhang
+- **abs:** https://arxiv.org/abs/2601.10024 · **PDF:** https://arxiv.org/pdf/2601.10024v2
+
+## 21652. Latent Structural Similarity Networks for Unsupervised Discovery in Multivariate Time Series
+
+- **ID:** `2601.18803` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Olusegun Owoeye
+- **abs:** https://arxiv.org/abs/2601.18803 · **PDF:** https://arxiv.org/pdf/2601.18803v1
+
+## 21653. EHRNavigator: A Multi-Agent System for Patient-Level Clinical Question Answering over Heterogeneous Electronic Health Records
+
+- **ID:** `2601.10020` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Lingfei Qian, Mauro Giuffre, Yan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.10020 · **PDF:** https://arxiv.org/pdf/2601.10020v1
+
+## 21654. Time Aggregation Features for XGBoost Models
+
+- **ID:** `2601.10019` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Mykola Pinchuk
+- **abs:** https://arxiv.org/abs/2601.10019 · **PDF:** https://arxiv.org/pdf/2601.10019v1
+
+## 21655. Empowering Older Adults in Digital Technology Use with Foundation Models
+
+- **ID:** `2601.10018` · **primary:** `cs.HC` · **Submitted:** 2026-01-15
+- **Authors:** Hasti Sharifi, Homaira Huda Shomee, Sourav Medya, et al.
+- **abs:** https://arxiv.org/abs/2601.10018 · **PDF:** https://arxiv.org/pdf/2601.10018v1
+
+## 21656. CAFEDistill: Learning Personalized and Dynamic Models through Federated Early-Exit Network Distillation
+
+- **ID:** `2601.10015` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Boyi Liu, Zimu Zhou, Yongxin Tong
+- **abs:** https://arxiv.org/abs/2601.10015 · **PDF:** https://arxiv.org/pdf/2601.10015v1
+
+## 21657. PID-Guided Partial Alignment for Multimodal Decentralized Federated Learning
+
+- **ID:** `2601.10012` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Yanhang Shi, Xiaoyu Wang, Houwei Cao, et al.
+- **abs:** https://arxiv.org/abs/2601.10012 · **PDF:** https://arxiv.org/pdf/2601.10012v1
+
+## 21658. Memo-SQL: Structured Decomposition and Experience-Driven Self-Correction for Training-Free NL2SQL
+
+- **ID:** `2601.10011` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Zerui Yang, Weichuan Wang, Yanwei Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.10011 · **PDF:** https://arxiv.org/pdf/2601.10011v1
+
+## 21659. VERHallu: Evaluating and Mitigating Event Relation Hallucination in Video Large Language Models
+
+- **ID:** `2601.10010` · **primary:** `cs.CV` · **Submitted:** 2026-01-15
+- **Authors:** Zefan Zhang, Kehua Zhu, Shijie Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.10010 · **PDF:** https://arxiv.org/pdf/2601.10010v1
+
+## 21660. Continuous-Depth Transformers with Learned Control Dynamics
+
+- **ID:** `2601.10007` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Peter Jemley
+- **abs:** https://arxiv.org/abs/2601.10007 · **PDF:** https://arxiv.org/pdf/2601.10007v1
+
+## 21661. SoK: Privacy-aware LLM in Healthcare: Threat Model, Privacy Techniques, Challenges and Recommendations
+
+- **ID:** `2601.10004` · **primary:** `cs.CR` · **Submitted:** 2026-01-15
+- **Authors:** Mohoshin Ara Tahera, Karamveer Singh Sidhu, Shuvalaxmi Dass, et al.
+- **abs:** https://arxiv.org/abs/2601.10004 · **PDF:** https://arxiv.org/pdf/2601.10004v1
+
+## 21662. SocraticKG: Knowledge Graph Construction via QA-Driven Fact Extraction
+
+- **ID:** `2601.10003` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Sanghyeok Choi, Woosang Jeon, Kyuseok Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.10003 · **PDF:** https://arxiv.org/pdf/2601.10003v1
+
+## 21663. FaTRQ: Tiered Residual Quantization for LLM Vector Search in Far-Memory-Aware ANNS Systems
+
+- **ID:** `2601.09985` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Tianqi Zhang, Flavio Ponzina, Tajana Rosing
+- **abs:** https://arxiv.org/abs/2601.09985 · **PDF:** https://arxiv.org/pdf/2601.09985v1
+
+## 21664. Context Volume Drives Performance: Tackling Domain Shift in Extremely Low-Resource Translation via RAG
+
+- **ID:** `2601.09982` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** David Samuel Setiawan, Raphaël Merx, Jey Han Lau
+- **abs:** https://arxiv.org/abs/2601.09982 · **PDF:** https://arxiv.org/pdf/2601.09982v2
+
+## 21665. Performance of AI agents based on reasoning language models on ALD process optimization tasks
+
+- **ID:** `2601.09980` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-15
+- **Authors:** Angel Yanguas-Gil
+- **abs:** https://arxiv.org/abs/2601.09980 · **PDF:** https://arxiv.org/pdf/2601.09980v1
+
+## 21666. In-Context Operator Learning on the Space of Probability Measures
+
+- **ID:** `2601.09979` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Frank Cole, Dixi Wang, Yineng Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.09979 · **PDF:** https://arxiv.org/pdf/2601.09979v1
+
+## 21667. Large Language Model (LLM)-enabled Reinforcement Learning for Wireless Network Optimization
+
+- **ID:** `2602.13210` · **primary:** `cs.NI` · **Submitted:** 2026-01-15
+- **Authors:** Jie Zheng, Ruichen Zhang, Dusit Niyato, et al.
+- **abs:** https://arxiv.org/abs/2602.13210 · **PDF:** https://arxiv.org/pdf/2602.13210v1
+
+## 21668. SPRInG: Continual LLM Personalization via Selective Parametric Adaptation and Retrieval-Interpolated Generation
+
+- **ID:** `2601.09974` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Seoyeon Kim, Jaehyung Kim
+- **abs:** https://arxiv.org/abs/2601.09974 · **PDF:** https://arxiv.org/pdf/2601.09974v1
+
+## 21669. Chinese Labor Law Large Language Model Benchmark
+
+- **ID:** `2601.09972` · **primary:** `cs.AI` · **Submitted:** 2026-01-15
+- **Authors:** Zixun Lan, Maochun Xu, Yifan Ren, et al.
+- **abs:** https://arxiv.org/abs/2601.09972 · **PDF:** https://arxiv.org/pdf/2601.09972v1
+
+## 21670. An Exploratory Study to Repurpose LLMs to a Unified Architecture for Time Series Classification
+
+- **ID:** `2601.09971` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Hansen He, Shuheng Li
+- **abs:** https://arxiv.org/abs/2601.09971 · **PDF:** https://arxiv.org/pdf/2601.09971v1
+
+## 21671. A Sustainable AI Economy Needs Data Deals That Work for Generators
+
+- **ID:** `2601.09966` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Ruoxi Jia, Luis Oala, Wenjie Xiong, et al.
+- **abs:** https://arxiv.org/abs/2601.09966 · **PDF:** https://arxiv.org/pdf/2601.09966v1
+
+## 21672. Take Out Your Calculators: Estimating the Real Difficulty of Question Items with LLM Student Simulations
+
+- **ID:** `2601.09953` · **primary:** `cs.CL` · **Submitted:** 2026-01-15
+- **Authors:** Christabel Acquaye, Yi Ting Huang, Marine Carpuat, et al.
+- **abs:** https://arxiv.org/abs/2601.09953 · **PDF:** https://arxiv.org/pdf/2601.09953v1
+
+## 21673. Kinematic Tokenization: Optimization-Based Continuous-Time Tokens for Learnable Decision Policies in Noisy Time Series
+
+- **ID:** `2601.09949` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Griffin Kearney
+- **abs:** https://arxiv.org/abs/2601.09949 · **PDF:** https://arxiv.org/pdf/2601.09949v2
+
+## 21674. Interpolation-Based Optimization for Enforcing lp-Norm Metric Differential Privacy in Continuous and Fine-Grained Domains
+
+- **ID:** `2601.09946` · **primary:** `cs.LG` · **Submitted:** 2026-01-15
+- **Authors:** Chenxi Qiu
+- **abs:** https://arxiv.org/abs/2601.09946 · **PDF:** https://arxiv.org/pdf/2601.09946v1
+
+## 21675. Malware Classification using Diluted Convolutional Neural Network with Fast Gradient Sign Method
+
+- **ID:** `2601.09933` · **primary:** `cs.CR` · **Submitted:** 2026-01-14
+- **Authors:** Ashish Anand, Bhupendra Singh, Sunil Khemka, et al.
+- **abs:** https://arxiv.org/abs/2601.09933 · **PDF:** https://arxiv.org/pdf/2601.09933v1
+
+## 21676. Hallucination Detection and Mitigation in Large Language Models
+
+- **ID:** `2601.09929` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Ahmad Pesaranghader, Erin Li
+- **abs:** https://arxiv.org/abs/2601.09929 · **PDF:** https://arxiv.org/pdf/2601.09929v1
+
+## 21677. The PROPER Approach to Proactivity: Benchmarking and Advancing Knowledge Gap Navigation
+
+- **ID:** `2601.09926` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Kirandeep Kaur, Vinayak Gupta, Aditya Gupta, et al.
+- **abs:** https://arxiv.org/abs/2601.09926 · **PDF:** https://arxiv.org/pdf/2601.09926v2
+
+## 21678. CaMeLs Can Use Computers Too: System-level Security for Computer Use Agents
+
+- **ID:** `2601.09923` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Hanna Foerster, Tom Blanchard, Kristina Nikolić, et al.
+- **abs:** https://arxiv.org/abs/2601.09923 · **PDF:** https://arxiv.org/pdf/2601.09923v2
+
+## 21679. Learning to Decode in Parallel: Self-Coordinating Neural Network for Real-Time Quantum Error Correction
+
+- **ID:** `2601.09921` · **primary:** `quant-ph` · **Submitted:** 2026-01-14
+- **Authors:** Kai Zhang, Zhengzhong Yi, Shaojun Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.09921 · **PDF:** https://arxiv.org/pdf/2601.09921v1
+
+## 21680. Continuum Memory Architectures for Long-Horizon LLM Agents
+
+- **ID:** `2601.09913` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Joe Logan
+- **abs:** https://arxiv.org/abs/2601.09913 · **PDF:** https://arxiv.org/pdf/2601.09913v1
+
+## 21681. Self-reflection in Automated Qualitative Coding: Improving Text Annotation through Secondary LLM Critique
+
+- **ID:** `2601.09905` · **primary:** `cs.SE` · **Submitted:** 2026-01-14
+- **Authors:** Zackary Okun Dunivin, Mobina Noori, Seth Frey, et al.
+- **abs:** https://arxiv.org/abs/2601.09905 · **PDF:** https://arxiv.org/pdf/2601.09905v1
+
+## 21682. A Novel Contrastive Loss for Zero-Day Network Intrusion Detection
+
+- **ID:** `2601.09902` · **primary:** `cs.CR` · **Submitted:** 2026-01-14
+- **Authors:** Jack Wilkie, Hanan Hindy, Craig Michie, et al.
+- **abs:** https://arxiv.org/abs/2601.09902 · **PDF:** https://arxiv.org/pdf/2601.09902v1
+
+## 21683. Predicting When to Trust Vision-Language Models for Spatial Reasoning
+
+- **ID:** `2601.11644` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Muhammad Imran, Yugyung Lee
+- **abs:** https://arxiv.org/abs/2601.11644 · **PDF:** https://arxiv.org/pdf/2601.11644v1
+
+## 21684. The Algorithmic Gaze of Image Quality Assessment: An Audit and Trace Ethnography of the LAION-Aesthetics Predictor
+
+- **ID:** `2601.09896` · **primary:** `cs.HC` · **Submitted:** 2026-01-14
+- **Authors:** Jordan Taylor, William Agnew, Maarten Sap, et al.
+- **abs:** https://arxiv.org/abs/2601.09896 · **PDF:** https://arxiv.org/pdf/2601.09896v3
+
+## 21685. Clozing the Gap: Exploring Why Language Model Surprisal Outperforms Cloze Surprisal
+
+- **ID:** `2601.09886` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Sathvik Nair, Byung-Doh Oh
+- **abs:** https://arxiv.org/abs/2601.09886 · **PDF:** https://arxiv.org/pdf/2601.09886v1
+
+## 21686. Beyond Rule-Based Workflows: An Information-Flow-Orchestrated Multi-Agents Paradigm via Agent-to-Agent Communication from CORAL
+
+- **ID:** `2601.09883` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Xinxing Ren, Quagmire Zang, Caelum Forder, et al.
+- **abs:** https://arxiv.org/abs/2601.09883 · **PDF:** https://arxiv.org/pdf/2601.09883v1
+
+## 21687. Transition Matching Distillation for Fast Video Generation
+
+- **ID:** `2601.09881` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Weili Nie, Julius Berner, Nanye Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.09881 · **PDF:** https://arxiv.org/pdf/2601.09881v1
+
+## 21688. MedVL-SAM2: A unified 3D medical vision-language model for multimodal reasoning and prompt-driven segmentation
+
+- **ID:** `2601.09879` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Yang Xing, Jiong Wu, Savas Ozdemir, et al.
+- **abs:** https://arxiv.org/abs/2601.09879 · **PDF:** https://arxiv.org/pdf/2601.09879v1
+
+## 21689. Patient-Similarity Cohort Reasoning in Clinical Text-to-SQL
+
+- **ID:** `2601.09876` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Yifei Shen, Yilun Zhao, Justice Ou, et al.
+- **abs:** https://arxiv.org/abs/2601.09876 · **PDF:** https://arxiv.org/pdf/2601.09876v1
+
+## 21690. Epistemology gives a Future to Complementarity in Human-AI Interactions
+
+- **ID:** `2601.09871` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Andrea Ferrario, Alessandro Facchini, Juan M. Durán
+- **abs:** https://arxiv.org/abs/2601.09871 · **PDF:** https://arxiv.org/pdf/2601.09871v1
+
+## 21691. A Scoping Review of the Ethical Perspectives on Anthropomorphising Large Language Model-Based Conversational Agents
+
+- **ID:** `2601.09869` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Andrea Ferrario, Rasita Vinay, Matteo Casserini, et al.
+- **abs:** https://arxiv.org/abs/2601.09869 · **PDF:** https://arxiv.org/pdf/2601.09869v1
+
+## 21692. VibrantSR: Sub-Meter Canopy Height Models from Sentinel-2 Using Generative Flow Matching
+
+- **ID:** `2601.09866` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Kiarie Ndegwa, Andreas Gros, Tony Chang, et al.
+- **abs:** https://arxiv.org/abs/2601.09866 · **PDF:** https://arxiv.org/pdf/2601.09866v2
+
+## 21693. Advancing Model Refinement: Muon-Optimized Distillation and Quantization for LLM Deployment
+
+- **ID:** `2601.09865` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Jacob Sander, Brian Jalaian, Venkat R. Dasari
+- **abs:** https://arxiv.org/abs/2601.09865 · **PDF:** https://arxiv.org/pdf/2601.09865v1
+
+## 21694. Breaking the Limits of Open-Weight CLIP: An Optimization Framework for Self-supervised Fine-tuning of CLIP
+
+- **ID:** `2601.09859` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Anant Mehta, Xiyuan Wei, Xingyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.09859 · **PDF:** https://arxiv.org/pdf/2601.09859v1
+
+## 21695. OUTLINEFORGE: Hierarchical Reinforcement Learning with Explicit States for Scientific Writing
+
+- **ID:** `2601.09858` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Yilin Bao, Ziyao He, Zayden Yang
+- **abs:** https://arxiv.org/abs/2601.09858 · **PDF:** https://arxiv.org/pdf/2601.09858v1
+
+## 21696. Thinking Long, but Short: Stable Sequential Test-Time Scaling for Large Reasoning Models
+
+- **ID:** `2601.09855` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Michael R. Metel, Yufei Cui, Boxing Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.09855 · **PDF:** https://arxiv.org/pdf/2601.09855v1
+
+## 21697. MedRedFlag: Investigating how LLMs Redirect Misconceptions in Real-World Health Communication
+
+- **ID:** `2601.09853` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Sraavya Sambara, Yuan Pu, Ayman Ali, et al.
+- **abs:** https://arxiv.org/abs/2601.09853 · **PDF:** https://arxiv.org/pdf/2601.09853v1
+
+## 21698. Bears, all bears, and some bears. Language Constraints on Language Models' Inductive Inferences
+
+- **ID:** `2601.09852` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Sriram Padmanabhan, Siyuan Song, Kanishka Misra
+- **abs:** https://arxiv.org/abs/2601.09852 · **PDF:** https://arxiv.org/pdf/2601.09852v2
+
+## 21699. ViSIL: Unified Evaluation of Information Loss in Multimodal Video Captioning
+
+- **ID:** `2601.09851` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Po-han Li, Shenghui Chen, Ufuk Topcu, et al.
+- **abs:** https://arxiv.org/abs/2601.09851 · **PDF:** https://arxiv.org/pdf/2601.09851v2
+
+## 21700. Accelerated Regularized Wasserstein Proximal Sampling Algorithms
+
+- **ID:** `2601.09848` · **primary:** `stat.ML` · **Submitted:** 2026-01-14
+- **Authors:** Hong Ye Tan, Stanley Osher, Wuchen Li
+- **abs:** https://arxiv.org/abs/2601.09848 · **PDF:** https://arxiv.org/pdf/2601.09848v2
+
+## 21701. A pipeline for enabling path-specific causal fairness in observational health data
+
+- **ID:** `2601.09841` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Aparajita Kashyap, Sara Matijevic, Noémie Elhadad, et al.
+- **abs:** https://arxiv.org/abs/2601.09841 · **PDF:** https://arxiv.org/pdf/2601.09841v2
+
+## 21702. Stable and Explainable Personality Trait Evaluation in Large Language Models with Internal Activations
+
+- **ID:** `2601.09833` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Xiaoxu Ma, Xiangbo Zhang, Zhenyu Weng
+- **abs:** https://arxiv.org/abs/2601.09833 · **PDF:** https://arxiv.org/pdf/2601.09833v1
+
+## 21703. A New Convergence Analysis of Plug-and-Play Proximal Gradient Descent Under Prior Mismatch
+
+- **ID:** `2601.09831` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Guixian Xu, Jinglai Li, Junqi Tang
+- **abs:** https://arxiv.org/abs/2601.09831 · **PDF:** https://arxiv.org/pdf/2601.09831v2
+
+## 21704. Eluder dimension: localise it!
+
+- **ID:** `2601.09825` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Alireza Bakhtiari, Alex Ayoub, Samuel Robertson, et al.
+- **abs:** https://arxiv.org/abs/2601.09825 · **PDF:** https://arxiv.org/pdf/2601.09825v1
+
+## 21705. LLM-Based Agentic Systems for Software Engineering: Challenges and Opportunities
+
+- **ID:** `2601.09822` · **primary:** `cs.SE` · **Submitted:** 2026-01-14
+- **Authors:** Yongjian Tang, Thomas Runkler
+- **abs:** https://arxiv.org/abs/2601.09822 · **PDF:** https://arxiv.org/pdf/2601.09822v2
+
+## 21706. Private Accountability in the Age of Artificial Intelligence
+
+- **ID:** `2601.17013` · **primary:** `cs.CY` · **Submitted:** 2026-01-14
+- **Authors:** Sonia Katyal
+- **abs:** https://arxiv.org/abs/2601.17013 · **PDF:** https://arxiv.org/pdf/2601.17013v1
+
+## 21707. Explainable Deep Learning for Pediatric Pneumonia Detection in Chest X-Ray Images
+
+- **ID:** `2601.09814` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Adil O. Khadidos, Aziida Nanyonga, Alaa O. Khadidos, et al.
+- **abs:** https://arxiv.org/abs/2601.09814 · **PDF:** https://arxiv.org/pdf/2601.09814v1
+
+## 21708. QFed: Parameter-Compact Quantum-Classical Federated Learning
+
+- **ID:** `2601.09809` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Samar Abdelghani, Soumaya Cherkaoui
+- **abs:** https://arxiv.org/abs/2601.09809 · **PDF:** https://arxiv.org/pdf/2601.09809v1
+
+## 21709. Diffusion-Driven Deceptive Patches: Adversarial Manipulation and Forensic Detection in Facial Identity Verification
+
+- **ID:** `2601.09806` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Shahrzad Sayyafzadeh, Hongmei Chi, Shonda Bernadin
+- **abs:** https://arxiv.org/abs/2601.09806 · **PDF:** https://arxiv.org/pdf/2601.09806v1
+
+## 21710. Improving Chain-of-Thought for Logical Reasoning via Attention-Aware Intervention
+
+- **ID:** `2601.09805` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Nguyen Minh Phuong, Dang Huu Tien, Naoya Inoue
+- **abs:** https://arxiv.org/abs/2601.09805 · **PDF:** https://arxiv.org/pdf/2601.09805v2
+
+## 21711. Fast-ThinkAct: Efficient Vision-Language-Action Reasoning via Verbalizable Latent Planning
+
+- **ID:** `2601.09708` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Chi-Pin Huang, Yunze Man, Zhiding Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.09708 · **PDF:** https://arxiv.org/pdf/2601.09708v2
+
+## 21712. Value-Aware Numerical Representations for Transformer Language Models
+
+- **ID:** `2601.09706` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Andreea Dutulescu, Stefan Ruseti, Mihai Dascalu
+- **abs:** https://arxiv.org/abs/2601.09706 · **PDF:** https://arxiv.org/pdf/2601.09706v1
+
+## 21713. ShortCoder: Knowledge-Augmented Syntax Optimization for Token-Efficient Code Generation
+
+- **ID:** `2601.09703` · **primary:** `cs.SE` · **Submitted:** 2026-01-14
+- **Authors:** Sicong Liu, Yanxian Huang, Mingwei Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.09703 · **PDF:** https://arxiv.org/pdf/2601.09703v1
+
+## 21714. Attention Is Not Retention: The Orthogonality Constraint in Infinite-Context Architectures
+
+- **ID:** `2601.15313` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-14
+- **Authors:** Oliver Zahn, Matt Beton, Simran Chana
+- **abs:** https://arxiv.org/abs/2601.15313 · **PDF:** https://arxiv.org/pdf/2601.15313v2
+
+## 21715. Empathy Applicability Modeling for General Health Queries
+
+- **ID:** `2601.09696` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Shan Randhawa, Agha Ali Raza, Kentaro Toyama, et al.
+- **abs:** https://arxiv.org/abs/2601.09696 · **PDF:** https://arxiv.org/pdf/2601.09696v1
+
+## 21716. LLMs can Compress LLMs: Adaptive Pruning by Agents
+
+- **ID:** `2601.09694` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Sai Varun Kodathala, Rakesh Vunnam
+- **abs:** https://arxiv.org/abs/2601.09694 · **PDF:** https://arxiv.org/pdf/2601.09694v1
+
+## 21717. Contrastive Geometric Learning Unlocks Unified Structure- and Ligand-Based Drug Design
+
+- **ID:** `2601.09693` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Lisa Schneckenreiter, Sohvi Luukkonen, Lukas Friedrich, et al.
+- **abs:** https://arxiv.org/abs/2601.09693 · **PDF:** https://arxiv.org/pdf/2601.09693v2
+
+## 21718. Routing with Generated Data: Annotation-Free LLM Skill Estimation and Expert Selection
+
+- **ID:** `2601.09692` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Tianyi Niu, Justin Chih-Yao Chen, Genta Indra Winata, et al.
+- **abs:** https://arxiv.org/abs/2601.09692 · **PDF:** https://arxiv.org/pdf/2601.09692v1
+
+## 21719. DeepResearchEval: An Automated Framework for Deep Research Task Construction and Agentic Evaluation
+
+- **ID:** `2601.09688` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Yibo Wang, Lei Wang, Yue Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.09688 · **PDF:** https://arxiv.org/pdf/2601.09688v1
+
+## 21720. LARGE: A Locally Adaptive Regularization Approach for Estimating Gaussian Graphical Models
+
+- **ID:** `2601.09686` · **primary:** `stat.ME` · **Submitted:** 2026-01-14
+- **Authors:** Ha Nguyen, Sumanta Basu
+- **abs:** https://arxiv.org/abs/2601.09686 · **PDF:** https://arxiv.org/pdf/2601.09686v1
+
+## 21721. Disentangling Task Conflicts in Multi-Task LoRA via Orthogonal Gradient Projection
+
+- **ID:** `2601.09684` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Ziyu Yang, Guibin Chen, Yuxin Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.09684 · **PDF:** https://arxiv.org/pdf/2601.09684v1
+
+## 21722. Automating Supply Chain Disruption Monitoring via an Agentic AI Approach
+
+- **ID:** `2601.09680` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Sara AlMahri, Liming Xu, Alexandra Brintrup
+- **abs:** https://arxiv.org/abs/2601.09680 · **PDF:** https://arxiv.org/pdf/2601.09680v1
+
+## 21723. TimeSAE: Sparse Decoding for Faithful Explanations of Black-Box Time Series Models
+
+- **ID:** `2601.09776` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Khalid Oublal, Quentin Bouniot, Qi Gan, et al.
+- **abs:** https://arxiv.org/abs/2601.09776 · **PDF:** https://arxiv.org/pdf/2601.09776v1
+
+## 21724. The Geometry of Thought: Disclosing the Transformer as a Tropical Polynomial Circuit
+
+- **ID:** `2601.09775` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Faruk Alpay, Bilge Senturk
+- **abs:** https://arxiv.org/abs/2601.09775 · **PDF:** https://arxiv.org/pdf/2601.09775v1
+
+## 21725. Collaborative Multi-Agent Test-Time Reinforcement Learning for Reasoning
+
+- **ID:** `2601.09667` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Zhiyuan Hu, Yunhai Hu, Juncheng Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.09667 · **PDF:** https://arxiv.org/pdf/2601.09667v2
+
+## 21726. Syllabic Agglutinative Tokenizations for Indonesian LLM: A Study from Gasing Literacy Learning System
+
+- **ID:** `2601.11643` · **primary:** `cs.CY` · **Submitted:** 2026-01-14
+- **Authors:** H. Situngkir, A. B. Lumbantobing, Y. Surya
+- **abs:** https://arxiv.org/abs/2601.11643 · **PDF:** https://arxiv.org/pdf/2601.11643v1
+
+## 21727. Exploring Fine-Tuning for Tabular Foundation Models
+
+- **ID:** `2601.09654` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Aditya Tanna, Pratinav Seth, Mohamed Bouadi, et al.
+- **abs:** https://arxiv.org/abs/2601.09654 · **PDF:** https://arxiv.org/pdf/2601.09654v1
+
+## 21728. Creating a Hybrid Rule and Neural Network Based Semantic Tagger using Silver Standard Data: the PyMUSAS framework for Multilingual Semantic Annotation
+
+- **ID:** `2601.09648` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Andrew Moore, Paul Rayson, Dawn Archer, et al.
+- **abs:** https://arxiv.org/abs/2601.09648 · **PDF:** https://arxiv.org/pdf/2601.09648v2
+
+## 21729. Identifying Models Behind Text-to-Image Leaderboards
+
+- **ID:** `2601.09647` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Ali Naseh, Yuefeng Peng, Anshuman Suri, et al.
+- **abs:** https://arxiv.org/abs/2601.09647 · **PDF:** https://arxiv.org/pdf/2601.09647v1
+
+## 21730. PersonalAlign: Hierarchical Implicit Intent Alignment for Personalized GUI Agent with Long-Term User-Centric Records
+
+- **ID:** `2601.09636` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Yibo Lyu, Gongwei Chen, Rui Shao, et al.
+- **abs:** https://arxiv.org/abs/2601.09636 · **PDF:** https://arxiv.org/pdf/2601.09636v1
+
+## 21731. Large-Scale Optimization Model Auto-Formulation: Harnessing LLM Flexibility via Structured Workflow
+
+- **ID:** `2601.09635` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Kuo Liang, Yuhang Lu, Jianming Mao, et al.
+- **abs:** https://arxiv.org/abs/2601.09635 · **PDF:** https://arxiv.org/pdf/2601.09635v3
+
+## 21732. TaxoBell: Gaussian Box Embeddings for Self-Supervised Taxonomy Expansion
+
+- **ID:** `2601.09633` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Sahil Mishra, Srinitish Srinivasan, Srikanta Bedathur, et al.
+- **abs:** https://arxiv.org/abs/2601.09633 · **PDF:** https://arxiv.org/pdf/2601.09633v1
+
+## 21733. LLMs Got Rhythm? Hybrid Phonological Filtering for Greek Poetry Rhyme Detection and Generation
+
+- **ID:** `2601.09631` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Stergios Chatzikyriakidis, Anastasia Natsina
+- **abs:** https://arxiv.org/abs/2601.09631 · **PDF:** https://arxiv.org/pdf/2601.09631v4
+
+## 21734. Enhancing LUT-based Deep Neural Networks Inference through Architecture and Connectivity Optimization
+
+- **ID:** `2601.09773` · **primary:** `cs.AR` · **Submitted:** 2026-01-14
+- **Authors:** Binglei Lou, Ruilin Wu, Philip Leong
+- **abs:** https://arxiv.org/abs/2601.09773 · **PDF:** https://arxiv.org/pdf/2601.09773v1
+
+## 21735. From Prompt to Protocol: Fast Charging Batteries with Large Language Models
+
+- **ID:** `2601.09626` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Ge Lei, Ferran Brosa Planella, Sterling G. Baird, et al.
+- **abs:** https://arxiv.org/abs/2601.09626 · **PDF:** https://arxiv.org/pdf/2601.09626v1
+
+## 21736. The Promptware Kill Chain: How Prompt Injections Gradually Evolved Into a Multistep Malware Delivery Mechanism
+
+- **ID:** `2601.09625` · **primary:** `cs.CR` · **Submitted:** 2026-01-14
+- **Authors:** Oleg Brodt, Elad Feldman, Bruce Schneier, et al.
+- **abs:** https://arxiv.org/abs/2601.09625 · **PDF:** https://arxiv.org/pdf/2601.09625v2
+
+## 21737. Toward Understanding Unlearning Difficulty: A Mechanistic Perspective and Circuit-Guided Difficulty Metric
+
+- **ID:** `2601.09624` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Jiali Cheng, Ziheng Chen, Chirag Agarwal, et al.
+- **abs:** https://arxiv.org/abs/2601.09624 · **PDF:** https://arxiv.org/pdf/2601.09624v1
+
+## 21738. PSSF: Early osteoarthritis detection using physical synthetic knee X-ray scans and AI radiomics models
+
+- **ID:** `2601.11642` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Abbas Alzubaidi, Ali Al-Bayaty
+- **abs:** https://arxiv.org/abs/2601.11642 · **PDF:** https://arxiv.org/pdf/2601.11642v1
+
+## 21739. LemonadeBench: Evaluating the Economic Intuition of Large Language Models in Simple Markets
+
+- **ID:** `2602.13209` · **primary:** `q-fin.GN` · **Submitted:** 2026-01-14
+- **Authors:** Aidan Vyas
+- **abs:** https://arxiv.org/abs/2602.13209 · **PDF:** https://arxiv.org/pdf/2602.13209v1
+
+## 21740. What Is Novel? A Knowledge-Driven Framework for Bias-Aware Literature Originality Evaluation
+
+- **ID:** `2602.06054` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Abeer Mostafa, Thi Huyen Nguyen, Zahra Ahmadi
+- **abs:** https://arxiv.org/abs/2602.06054 · **PDF:** https://arxiv.org/pdf/2602.06054v2
+
+## 21741. Pruning as Evolution: Emergent Sparsity Through Selection Dynamics in Neural Networks
+
+- **ID:** `2601.10765` · **primary:** `cs.NE` · **Submitted:** 2026-01-14
+- **Authors:** Zubair Shah, Noaman Khan
+- **abs:** https://arxiv.org/abs/2601.10765 · **PDF:** https://arxiv.org/pdf/2601.10765v1
+
+## 21742. Full Disclosure, Less Trust? How the Level of Detail about AI Use in News Writing Affects Readers' Trust
+
+- **ID:** `2601.09620` · **primary:** `cs.HC` · **Submitted:** 2026-01-14
+- **Authors:** Pooja Prajod, Hannes Cools, Thomas Röggla, et al.
+- **abs:** https://arxiv.org/abs/2601.09620 · **PDF:** https://arxiv.org/pdf/2601.09620v1
+
+## 21743. CogRail: Benchmarking VLMs in Cognitive Intrusion Perception for Intelligent Railway Transportation Systems
+
+- **ID:** `2601.09613` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Yonglin Tian, Qiyao Zhang, Wei Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.09613 · **PDF:** https://arxiv.org/pdf/2601.09613v1
+
+## 21744. DPWriter: Reinforcement Learning with Diverse Planning Branching for Creative Writing
+
+- **ID:** `2601.09609` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Qian Cao, Yahui Liu, Wei Bi, et al.
+- **abs:** https://arxiv.org/abs/2601.09609 · **PDF:** https://arxiv.org/pdf/2601.09609v1
+
+## 21745. Mixture of Distributions Matters: Dynamic Sparse Attention for Efficient Video Diffusion Transformers
+
+- **ID:** `2601.11641` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Yuxi Liu, Yipeng Hu, Zekun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.11641 · **PDF:** https://arxiv.org/pdf/2601.11641v2
+
+## 21746. Sim2real Image Translation Enables Viewpoint-Robust Policies from Fixed-Camera Datasets
+
+- **ID:** `2601.09605` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Jeremiah Coholich, Justin Wit, Robert Azarcon, et al.
+- **abs:** https://arxiv.org/abs/2601.09605 · **PDF:** https://arxiv.org/pdf/2601.09605v3
+
+## 21747. Linear Complexity Self-Supervised Learning for Music Understanding with Random Quantizer
+
+- **ID:** `2601.09603` · **primary:** `cs.SD` · **Submitted:** 2026-01-14
+- **Authors:** Petros Vavaroutsos, Theodoros Palamas, Pantelis Vikatos
+- **abs:** https://arxiv.org/abs/2601.09603 · **PDF:** https://arxiv.org/pdf/2601.09603v1
+
+## 21748. Information Access of the Oppressed: A Problem-Posing Framework for Envisioning Emancipatory Information Access Platforms
+
+- **ID:** `2601.09600` · **primary:** `cs.CY` · **Submitted:** 2026-01-14
+- **Authors:** Bhaskar Mitra, Nicola Neophytou, Sireesh Gururaja
+- **abs:** https://arxiv.org/abs/2601.09600 · **PDF:** https://arxiv.org/pdf/2601.09600v2
+
+## 21749. Improving CMA-ES Convergence Speed, Efficiency, and Reliability in Noisy Robot Optimization Problems
+
+- **ID:** `2601.09594` · **primary:** `cs.NE` · **Submitted:** 2026-01-14
+- **Authors:** Russell M. Martin, Steven H. Collins
+- **abs:** https://arxiv.org/abs/2601.09594 · **PDF:** https://arxiv.org/pdf/2601.09594v1
+
+## 21750. Do people expect different behavior from large language models acting on their behalf? Evidence from norm elicitations in two canonical economic games
+
+- **ID:** `2601.15312` · **primary:** `cs.GT` · **Submitted:** 2026-01-14
+- **Authors:** Paweł Niszczota, Elia Antoniou
+- **abs:** https://arxiv.org/abs/2601.15312 · **PDF:** https://arxiv.org/pdf/2601.15312v1
+
+## 21751. Antisocial behavior towards large language model users: experimental evidence
+
+- **ID:** `2601.09772` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Paweł Niszczota, Cassandra Grützner
+- **abs:** https://arxiv.org/abs/2601.09772 · **PDF:** https://arxiv.org/pdf/2601.09772v1
+
+## 21752. Energy-Entropy Regularization: The True Power of Minimal Looped Transformers
+
+- **ID:** `2601.09588` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Wai-Lun Lam
+- **abs:** https://arxiv.org/abs/2601.09588 · **PDF:** https://arxiv.org/pdf/2601.09588v2
+
+## 21753. Show, don't tell -- Providing Visual Error Feedback for Handwritten Documents
+
+- **ID:** `2601.09586` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Said Yasin, Torsten Zesch
+- **abs:** https://arxiv.org/abs/2601.09586 · **PDF:** https://arxiv.org/pdf/2601.09586v1
+
+## 21754. Constraint- and Score-Based Nonlinear Granger Causality Discovery with Kernels
+
+- **ID:** `2601.09579` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Fiona Murphy, Alessio Benavoli
+- **abs:** https://arxiv.org/abs/2601.09579 · **PDF:** https://arxiv.org/pdf/2601.09579v1
+
+## 21755. Permutation Matching Under Parikh Budgets: Linear-Time Detection, Packing, and Disjoint Selection
+
+- **ID:** `2601.09577` · **primary:** `cs.DS` · **Submitted:** 2026-01-14
+- **Authors:** MD Nazmul Alam Shanto, Md. Tanzeem Rahat, Md. Manzurul Hasan
+- **abs:** https://arxiv.org/abs/2601.09577 · **PDF:** https://arxiv.org/pdf/2601.09577v1
+
+## 21756. Irreversible Kinetics Emerges from Bayesian Inference over Admissible Histories
+
+- **ID:** `2601.10763` · **primary:** `cond-mat.stat-mech` · **Submitted:** 2026-01-14
+- **Authors:** Manas V. Upadhyay
+- **abs:** https://arxiv.org/abs/2601.10763 · **PDF:** https://arxiv.org/pdf/2601.10763v1
+
+## 21757. Dialogue Telemetry: Turn-Level Instrumentation for Autonomous Information Gathering
+
+- **ID:** `2601.09570` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Dimitris Panagopoulos, Adolfo Perrusquia, Weisi Guo
+- **abs:** https://arxiv.org/abs/2601.09570 · **PDF:** https://arxiv.org/pdf/2601.09570v1
+
+## 21758. Hot-Start from Pixels: Low-Resolution Visual Tokens for Chinese Language Modeling
+
+- **ID:** `2601.09566` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Shuyang Xiang, Hao Guan
+- **abs:** https://arxiv.org/abs/2601.09566 · **PDF:** https://arxiv.org/pdf/2601.09566v3
+
+## 21759. Global Optimization By Gradient From Hierarchical Score-Matching Spaces
+
+- **ID:** `2601.11639` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Ming Li
+- **abs:** https://arxiv.org/abs/2601.11639 · **PDF:** https://arxiv.org/pdf/2601.11639v3
+
+## 21760. Aeon: High-Performance Neuro-Symbolic Memory Management for Long-Horizon LLM Agents
+
+- **ID:** `2601.15311` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Mustafa Arslan
+- **abs:** https://arxiv.org/abs/2601.15311 · **PDF:** https://arxiv.org/pdf/2601.15311v3
+
+## 21761. SiliconHealth: A Complete Low-Cost Blockchain Healthcare Infrastructure for Resource-Constrained Regions Using Repurposed Bitcoin Mining ASICs
+
+- **ID:** `2601.09557` · **primary:** `cs.NE` · **Submitted:** 2026-01-14
+- **Authors:** Francisco Angulo de Lafuente, Seid Mehammed Abdu, Nirmal Tej
+- **abs:** https://arxiv.org/abs/2601.09557 · **PDF:** https://arxiv.org/pdf/2601.09557v2
+
+## 21762. Verifying Physics-Informed Neural Network Fidelity using Classical Fisher Information from Differentiable Dynamical System
+
+- **ID:** `2601.11638` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Josafat Ribeiro Leal Filho, Antônio Augusto Fröhlich
+- **abs:** https://arxiv.org/abs/2601.11638 · **PDF:** https://arxiv.org/pdf/2601.11638v1
+
+## 21763. Benchmarking Post-Training Quantization of Large Language Models under Microscaling Floating Point Formats
+
+- **ID:** `2601.09555` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Manyi Zhang, Ji-Fu Li, Zhongao Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.09555 · **PDF:** https://arxiv.org/pdf/2601.09555v1
+
+## 21764. PCN-Rec: Agentic Proof-Carrying Negotiation for Reliable Governance-Constrained Recommendation
+
+- **ID:** `2601.09771` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Aradhya Dixit, Shreem Dixit
+- **abs:** https://arxiv.org/abs/2601.09771 · **PDF:** https://arxiv.org/pdf/2601.09771v1
+
+## 21765. Omni-R1: Towards the Unified Generative Paradigm for Multimodal Reasoning
+
+- **ID:** `2601.09536` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Dongjie Cheng, Yongqi Li, Zhixin Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.09536 · **PDF:** https://arxiv.org/pdf/2601.09536v1
+
+## 21766. Residual Power Flow for Neural Solvers
+
+- **ID:** `2601.09533` · **primary:** `eess.SY` · **Submitted:** 2026-01-14
+- **Authors:** Jochen Stiasny, Jochen Cremer
+- **abs:** https://arxiv.org/abs/2601.09533 · **PDF:** https://arxiv.org/pdf/2601.09533v1
+
+## 21767. Private LLM Inference on Consumer Blackwell GPUs: A Practical Guide for Cost-Effective Local Deployment in SMEs
+
+- **ID:** `2601.09527` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Jonathan Knoop, Hendrik Holtmann
+- **abs:** https://arxiv.org/abs/2601.09527 · **PDF:** https://arxiv.org/pdf/2601.09527v1
+
+## 21768. Class Adaptive Conformal Training
+
+- **ID:** `2601.09522` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Badr-Eddine Marani, Julio Silva-Rodriguez, Ismail Ben Ayed, et al.
+- **abs:** https://arxiv.org/abs/2601.09522 · **PDF:** https://arxiv.org/pdf/2601.09522v1
+
+## 21769. Optimizing the Landscape of LLM Embeddings with Dynamic Exploratory Graph Analysis for Generative Psychometrics: A Monte Carlo Study
+
+- **ID:** `2601.17010` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Hudson Golino
+- **abs:** https://arxiv.org/abs/2601.17010 · **PDF:** https://arxiv.org/pdf/2601.17010v1
+
+## 21770. Project Aletheia: Verifier-Guided Distillation of Backtracking for Small Language Models
+
+- **ID:** `2601.14290` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Aradhya Dixit, Tianxi Liang, Jai Telang
+- **abs:** https://arxiv.org/abs/2601.14290 · **PDF:** https://arxiv.org/pdf/2601.14290v1
+
+## 21771. Towards Realistic Synthetic Data for Automatic Drum Transcription
+
+- **ID:** `2601.09520` · **primary:** `cs.SD` · **Submitted:** 2026-01-14
+- **Authors:** Pierfrancesco Melucci, Paolo Merialdo, Taketo Akama
+- **abs:** https://arxiv.org/abs/2601.09520 · **PDF:** https://arxiv.org/pdf/2601.09520v1
+
+## 21772. Learning Whole-Body Human-Humanoid Interaction from Human-Human Demonstrations
+
+- **ID:** `2601.09518` · **primary:** `cs.RO` · **Submitted:** 2026-01-14
+- **Authors:** Wei-Jin Huang, Yue-Yi Zhang, Yi-Lin Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.09518 · **PDF:** https://arxiv.org/pdf/2601.09518v1
+
+## 21773. SERM: Self-Evolving Relevance Model with Agent-Driven Learning from Massive Query Streams
+
+- **ID:** `2601.09515` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Chenglong Wang, Canjia Li, Xingzhao Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.09515 · **PDF:** https://arxiv.org/pdf/2601.09515v1
+
+## 21774. GUI-Eyes: Tool-Augmented Perception for Visual Grounding in GUI Agents
+
+- **ID:** `2601.09770` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Chen Chen, Jiawei Shao, Dakuan Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.09770 · **PDF:** https://arxiv.org/pdf/2601.09770v1
+
+## 21775. Online parameter estimation for the Crazyflie quadcopter through an EM algorithm
+
+- **ID:** `2601.17009` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Yanhua Zhao
+- **abs:** https://arxiv.org/abs/2601.17009 · **PDF:** https://arxiv.org/pdf/2601.17009v1
+
+## 21776. CLARE: Continual Learning for Vision-Language-Action Models via Autonomous Adapter Routing and Expansion
+
+- **ID:** `2601.09512` · **primary:** `cs.RO` · **Submitted:** 2026-01-14
+- **Authors:** Ralf Römer, Yi Zhang, Angela P. Schoellig
+- **abs:** https://arxiv.org/abs/2601.09512 · **PDF:** https://arxiv.org/pdf/2601.09512v1
+
+## 21777. MVSS: A Unified Framework for Multi-View Structured Survey Generation
+
+- **ID:** `2601.09504` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Yinqi Liu, Yueqi Zhu, Yongkang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.09504 · **PDF:** https://arxiv.org/pdf/2601.09504v2
+
+## 21778. What Do LLM Agents Know About Their World? Task2Quiz: A Paradigm for Studying Environment Understanding
+
+- **ID:** `2601.09503` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Siyuan Liu, Hongbang Yuan, Xinze Li, et al.
+- **abs:** https://arxiv.org/abs/2601.09503 · **PDF:** https://arxiv.org/pdf/2601.09503v1
+
+## 21779. Towards Robust Cross-Dataset Object Detection Generalization under Domain Specificity
+
+- **ID:** `2601.09497` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Ritabrata Chakraborty, Hrishit Mitra, Shivakumara Palaiahnakote, et al.
+- **abs:** https://arxiv.org/abs/2601.09497 · **PDF:** https://arxiv.org/pdf/2601.09497v1
+
+## 21780. Parallelizable memory recurrent units
+
+- **ID:** `2601.09495` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Florent De Geeter, Gaspard Lambrechts, Damien Ernst, et al.
+- **abs:** https://arxiv.org/abs/2601.09495 · **PDF:** https://arxiv.org/pdf/2601.09495v2
+
+## 21781. Deep Operator Networks for Surrogate Modeling of Cyclic Adsorption Processes with Varying Initial Conditions
+
+- **ID:** `2601.09491` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Beatrice Ceccanti, Mattia Galanti, Ivo Roghair, et al.
+- **abs:** https://arxiv.org/abs/2601.09491 · **PDF:** https://arxiv.org/pdf/2601.09491v1
+
+## 21782. SlidesGen-Bench: Evaluating Slides Generation via Computational and Quantitative Metrics
+
+- **ID:** `2601.09487` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Yunqiao Yang, Wenbo Li, Houxing Ren, et al.
+- **abs:** https://arxiv.org/abs/2601.09487 · **PDF:** https://arxiv.org/pdf/2601.09487v1
+
+## 21783. Bridging Semantic Understanding and Popularity Bias with LLMs
+
+- **ID:** `2601.09478` · **primary:** `cs.IR` · **Submitted:** 2026-01-14
+- **Authors:** Renqiang Luo, Dong Zhang, Yupeng Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.09478 · **PDF:** https://arxiv.org/pdf/2601.09478v3
+
+## 21784. Terminally constrained flow-based generative models from an optimal control perspective
+
+- **ID:** `2601.09474` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Weiguo Gao, Ming Li, Qianxiao Li
+- **abs:** https://arxiv.org/abs/2601.09474 · **PDF:** https://arxiv.org/pdf/2601.09474v1
+
+## 21785. SimMerge: Learning to Select Merge Operators from Similarity Signals
+
+- **ID:** `2601.09473` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Oliver Bolton, Aakanksha, Arash Ahmadian, et al.
+- **abs:** https://arxiv.org/abs/2601.09473 · **PDF:** https://arxiv.org/pdf/2601.09473v2
+
+## 21786. Personalized Multimodal Feedback Using Multiple External Representations: Strategy Profiles and Learning in High School Physics
+
+- **ID:** `2601.09470` · **primary:** `physics.ed-ph` · **Submitted:** 2026-01-14
+- **Authors:** Natalia Revenga-Lozano, Karina E. Avila, Steffen Steinert, et al.
+- **abs:** https://arxiv.org/abs/2601.09470 · **PDF:** https://arxiv.org/pdf/2601.09470v1
+
+## 21787. FairGU: Fairness-aware Graph Unlearning in Social Networks
+
+- **ID:** `2601.09469` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Renqiang Luo, Yongshuai Yang, Huafei Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.09469 · **PDF:** https://arxiv.org/pdf/2601.09469v2
+
+## 21788. High-fidelity lunar topographic reconstruction across diverse terrain and illumination environments using deep learning
+
+- **ID:** `2601.09468` · **primary:** `astro-ph.EP` · **Submitted:** 2026-01-14
+- **Authors:** Hao Chen, Philipp Gläser, Konrad Willner, et al.
+- **abs:** https://arxiv.org/abs/2601.09468 · **PDF:** https://arxiv.org/pdf/2601.09468v1
+
+## 21789. Searth Transformer: A Transformer Architecture Incorporating Earth's Geospheric Physical Priors for Global Mid-Range Weather Forecasting
+
+- **ID:** `2601.09467` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Tianye Li, Qi Liu, Hao Li, et al.
+- **abs:** https://arxiv.org/abs/2601.09467 · **PDF:** https://arxiv.org/pdf/2601.09467v1
+
+## 21790. EvoFSM: Controllable Self-Evolution for Deep Research with Finite State Machines
+
+- **ID:** `2601.09465` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Shuo Zhang, Chaofa Yuan, Ryan Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.09465 · **PDF:** https://arxiv.org/pdf/2601.09465v2
+
+## 21791. Bayesian Robust Financial Trading with Adversarial Synthetic Market Data
+
+- **ID:** `2601.17008` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Haochong Xia, Simin Li, Ruixiao Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.17008 · **PDF:** https://arxiv.org/pdf/2601.17008v1
+
+## 21792. ROSA-Tuning: Enhancing Long-Context Modeling via Suffix Matching
+
+- **ID:** `2602.02499` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Yunao Zheng, Xiaojie Wang, Lei Ren, et al.
+- **abs:** https://arxiv.org/abs/2602.02499 · **PDF:** https://arxiv.org/pdf/2602.02499v2
+
+## 21793. SoK: Enhancing Cryptographic Collaborative Learning with Differential Privacy
+
+- **ID:** `2601.09460` · **primary:** `cs.CR` · **Submitted:** 2026-01-14
+- **Authors:** Francesco Capano, Jonas Böhler, Benjamin Weggenmann
+- **abs:** https://arxiv.org/abs/2601.09460 · **PDF:** https://arxiv.org/pdf/2601.09460v1
+
+## 21794. Dissecting Judicial Reasoning in U.S. Copyright Damage Awards
+
+- **ID:** `2601.09459` · **primary:** `cs.IR` · **Submitted:** 2026-01-14
+- **Authors:** Pei-Chi Lo, Thomas Y. Lu
+- **abs:** https://arxiv.org/abs/2601.09459 · **PDF:** https://arxiv.org/pdf/2601.09459v1
+
+## 21795. On the Hardness of Computing Counterfactual and Semifactual Explanations in XAI
+
+- **ID:** `2601.09455` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** André Artelt, Martin Olsen, Kevin Tierney
+- **abs:** https://arxiv.org/abs/2601.09455 · **PDF:** https://arxiv.org/pdf/2601.09455v1
+
+## 21796. Late Breaking Results: Quamba-SE: Soft-edge Quantizer for Activations in State Space Models
+
+- **ID:** `2601.09451` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Yizhi Chen, Ahmed Hemani
+- **abs:** https://arxiv.org/abs/2601.09451 · **PDF:** https://arxiv.org/pdf/2601.09451v1
+
+## 21797. Population-Aligned Audio Reproduction With LLM-Based Equalizers
+
+- **ID:** `2601.09448` · **primary:** `cs.SD` · **Submitted:** 2026-01-14
+- **Authors:** Ioannis Stylianou, Jon Francombe, Pablo Martinez-Nuevo, et al.
+- **abs:** https://arxiv.org/abs/2601.09448 · **PDF:** https://arxiv.org/pdf/2601.09448v1
+
+## 21798. Improving Symbolic Translation of Language Models for Logical Reasoning
+
+- **ID:** `2601.09446` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Ramya Keerthy Thatikonda, Jiuzhou Han, Wray Buntine, et al.
+- **abs:** https://arxiv.org/abs/2601.09446 · **PDF:** https://arxiv.org/pdf/2601.09446v1
+
+## 21799. Where Knowledge Collides: A Mechanistic Study of Intra-Memory Knowledge Conflict in Language Models
+
+- **ID:** `2601.09445` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Minh Vu Pham, Hsuvas Borkakoty, Yufang Hou
+- **abs:** https://arxiv.org/abs/2601.09445 · **PDF:** https://arxiv.org/pdf/2601.09445v1
+
+## 21800. DeepLight: A Sobolev-trained Image-to-Image Surrogate Model for Light Transport in Tissue
+
+- **ID:** `2601.09439` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Philipp Haim, Vasilis Ntziachristos, Torsten Enßlin, et al.
+- **abs:** https://arxiv.org/abs/2601.09439 · **PDF:** https://arxiv.org/pdf/2601.09439v1
+
+## 21801. Do Transformers Understand Ancient Roman Coin Motifs Better than CNNs?
+
+- **ID:** `2601.09433` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** David Reid, Ognjen Arandjelovic
+- **abs:** https://arxiv.org/abs/2601.09433 · **PDF:** https://arxiv.org/pdf/2601.09433v1
+
+## 21802. Leveraging Large Language Models to Extract and Translate Medical Information in Doctors' Notes for Health Records and Diagnostic Billing Codes
+
+- **ID:** `2603.22625` · **primary:** `cs.IR` · **Submitted:** 2026-01-14
+- **Authors:** Peter Hartnett, Chung-Chi Huang, Sarah Hartnett, et al.
+- **abs:** https://arxiv.org/abs/2603.22625 · **PDF:** https://arxiv.org/pdf/2603.22625v1
+
+## 21803. Draw it like Euclid: Teaching transformer models to generate CAD profiles using ruler and compass construction steps
+
+- **ID:** `2601.09428` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Siyi Li, Joseph G. Lambourne, Longfei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.09428 · **PDF:** https://arxiv.org/pdf/2601.09428v1
+
+## 21804. Bias Dynamics in BabyLMs: Towards a Compute-Efficient Sandbox for Democratising Pre-Training Debiasing
+
+- **ID:** `2601.09421` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Filip Trhlik, Andrew Caines, Paula Buttery
+- **abs:** https://arxiv.org/abs/2601.09421 · **PDF:** https://arxiv.org/pdf/2601.09421v2
+
+## 21805. Radiomics-Integrated Deep Learning with Hierarchical Loss for Osteosarcoma Histology Classification
+
+- **ID:** `2601.09416` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Yaxi Chen, Zi Ye, Shaheer U. Saeed, et al.
+- **abs:** https://arxiv.org/abs/2601.09416 · **PDF:** https://arxiv.org/pdf/2601.09416v1
+
+## 21806. Speech-Hands: A Self-Reflection Voice Agentic Approach to Speech Recognition and Audio Reasoning with Omni Perception
+
+- **ID:** `2601.09413` · **primary:** `cs.SD` · **Submitted:** 2026-01-14
+- **Authors:** Zhen Wan, Chao-Han Huck Yang, Jinchuan Tian, et al.
+- **abs:** https://arxiv.org/abs/2601.09413 · **PDF:** https://arxiv.org/pdf/2601.09413v1
+
+## 21807. Test-Time Detoxification without Training or Learning Anything
+
+- **ID:** `2602.02498` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Baturay Saglam, Dionysis Kalogerias
+- **abs:** https://arxiv.org/abs/2602.02498 · **PDF:** https://arxiv.org/pdf/2602.02498v1
+
+## 21808. Structured Knowledge Representation through Contextual Pages for Retrieval-Augmented Generation
+
+- **ID:** `2601.09402` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Xinze Li, Zhenghao Liu, Haidong Xin, et al.
+- **abs:** https://arxiv.org/abs/2601.09402 · **PDF:** https://arxiv.org/pdf/2601.09402v1
+
+## 21809. Preliminary Tests of the Anticipatory Classifier System with Hindsight Experience Replay
+
+- **ID:** `2601.09400` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Olgierd Unold, Stanisław Franczyk
+- **abs:** https://arxiv.org/abs/2601.09400 · **PDF:** https://arxiv.org/pdf/2601.09400v1
+
+## 21810. Ability Transfer and Recovery via Modularized Parameters Localization
+
+- **ID:** `2601.09398` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Songyao Jin, Kun Zhou, Wenqi Li, et al.
+- **abs:** https://arxiv.org/abs/2601.09398 · **PDF:** https://arxiv.org/pdf/2601.09398v1
+
+## 21811. RPC-Bench: A Fine-grained Benchmark for Research Paper Comprehension
+
+- **ID:** `2601.14289` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Yelin Chen, Fanjin Zhang, Suping Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.14289 · **PDF:** https://arxiv.org/pdf/2601.14289v1
+
+## 21812. FairGE: Fairness-Aware Graph Encoding in Incomplete Social Networks
+
+- **ID:** `2601.09394` · **primary:** `cs.SI` · **Submitted:** 2026-01-14
+- **Authors:** Renqiang Luo, Huafei Huang, Tao Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.09394 · **PDF:** https://arxiv.org/pdf/2601.09394v2
+
+## 21813. SLAM-LLM: A Modular, Open-Source Multimodal Large Language Model Framework and Best Practice for Speech, Language, Audio and Music Processing
+
+- **ID:** `2601.09385` · **primary:** `cs.SD` · **Submitted:** 2026-01-14
+- **Authors:** Ziyang Ma, Guanrou Yang, Wenxi Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.09385 · **PDF:** https://arxiv.org/pdf/2601.09385v1
+
+## 21814. CLiMB: A Domain-Informed Novelty Detection Clustering Framework for Galactic Archaeology and Scientific Discovery
+
+- **ID:** `2601.09768` · **primary:** `astro-ph.IM` · **Submitted:** 2026-01-14
+- **Authors:** Lorenzo Monti, Tatiana Muraveva, Brian Sheridan, et al.
+- **abs:** https://arxiv.org/abs/2601.09768 · **PDF:** https://arxiv.org/pdf/2601.09768v2
+
+## 21815. Long-term Task-oriented Agent: Proactive Long-term Intent Maintenance in Dynamic Environments
+
+- **ID:** `2601.09382` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Qinglong Shi, Donghai Wang, Hantao Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.09382 · **PDF:** https://arxiv.org/pdf/2601.09382v1
+
+## 21816. Query Languages for Machine-Learning Models
+
+- **ID:** `2601.09381` · **primary:** `cs.LO` · **Submitted:** 2026-01-14
+- **Authors:** Martin Grohe
+- **abs:** https://arxiv.org/abs/2601.09381 · **PDF:** https://arxiv.org/pdf/2601.09381v1
+
+## 21817. Analysis of voice recordings features for Classification of Parkinson's Disease
+
+- **ID:** `2601.17007` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Beatriz Pérez-Sánchez, Noelia Sánchez-Maroño, Miguel A. Díaz-Freire
+- **abs:** https://arxiv.org/abs/2601.17007 · **PDF:** https://arxiv.org/pdf/2601.17007v1
+
+## 21818. The Imperfective Paradox in Large Language Models
+
+- **ID:** `2601.09373` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Bolei Ma, Yusuke Miyao
+- **abs:** https://arxiv.org/abs/2601.09373 · **PDF:** https://arxiv.org/pdf/2601.09373v1
+
+## 21819. Relation Extraction Capabilities of LLMs on Clinical Text: A Bilingual Evaluation for English and Turkish
+
+- **ID:** `2601.09367` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Aidana Aidynkyzy, Oğuz Dikenelli, Oylum Alatlı, et al.
+- **abs:** https://arxiv.org/abs/2601.09367 · **PDF:** https://arxiv.org/pdf/2601.09367v1
+
+## 21820. Frame of Reference: Addressing the Challenges of Common Ground Representation in Situational Dialogs
+
+- **ID:** `2601.09365` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Biswesh Mohapatra, Théo Charlot, Giovanni Duca, et al.
+- **abs:** https://arxiv.org/abs/2601.09365 · **PDF:** https://arxiv.org/pdf/2601.09365v1
+
+## 21821. GeoRA: Geometry-Aware Low-Rank Adaptation for RLVR
+
+- **ID:** `2601.09361` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Jiaying Zhang, Lei Shi, Jiguo Li, et al.
+- **abs:** https://arxiv.org/abs/2601.09361 · **PDF:** https://arxiv.org/pdf/2601.09361v2
+
+## 21822. Measuring the benefits of lying in MARA under egalitarian social welfare
+
+- **ID:** `2601.09354` · **primary:** `cs.GT` · **Submitted:** 2026-01-14
+- **Authors:** Jonathan Carrero, Ismael Rodriguez, Fernando Rubio
+- **abs:** https://arxiv.org/abs/2601.09354 · **PDF:** https://arxiv.org/pdf/2601.09354v1
+
+## 21823. Monte-Carlo Tree Search with Neural Network Guidance for Lane-Free Autonomous Driving
+
+- **ID:** `2601.09353` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Ioannis Peridis, Dimitrios Troullinos, Georgios Chalkiadakis, et al.
+- **abs:** https://arxiv.org/abs/2601.09353 · **PDF:** https://arxiv.org/pdf/2601.09353v1
+
+## 21824. Navigating Ethical AI Challenges in the Industrial Sector: Balancing Innovation and Responsibility
+
+- **ID:** `2601.09351` · **primary:** `cs.CY` · **Submitted:** 2026-01-14
+- **Authors:** Ruomu Tan, Martin W Hoffmann
+- **abs:** https://arxiv.org/abs/2601.09351 · **PDF:** https://arxiv.org/pdf/2601.09351v1
+
+## 21825. Improving Implicit Hate Speech Detection via a Community-Driven Multi-Agent Framework
+
+- **ID:** `2601.09342` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Ewelina Gajewska, Katarzyna Budzynska, Jarosław A Chudziak
+- **abs:** https://arxiv.org/abs/2601.09342 · **PDF:** https://arxiv.org/pdf/2601.09342v2
+
+## 21826. High-Performance Serverless Computing: A Systematic Literature Review on Serverless for HPC, AI, and Big Data
+
+- **ID:** `2601.09334` · **primary:** `cs.DC` · **Submitted:** 2026-01-14
+- **Authors:** Valerio Besozzi, Matteo Della Bartola, Patrizio Dazzi, et al.
+- **abs:** https://arxiv.org/abs/2601.09334 · **PDF:** https://arxiv.org/pdf/2601.09334v1
+
+## 21827. DeepInflation: an AI agent for research and model discovery of inflation
+
+- **ID:** `2601.14288` · **primary:** `astro-ph.CO` · **Submitted:** 2026-01-14
+- **Authors:** Ze-Yu Peng, Hao-Shi Yuan, Qi Lai, et al.
+- **abs:** https://arxiv.org/abs/2601.14288 · **PDF:** https://arxiv.org/pdf/2601.14288v1
+
+## 21828. Understanding or Memorizing? A Case Study of German Definite Articles in Language Models
+
+- **ID:** `2601.09313` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Jonathan Drechsel, Erisa Bytyqi, Steffen Herbold
+- **abs:** https://arxiv.org/abs/2601.09313 · **PDF:** https://arxiv.org/pdf/2601.09313v1
+
+## 21829. On-Device Large Language Models for Sequential Recommendation
+
+- **ID:** `2601.09306` · **primary:** `cs.IR` · **Submitted:** 2026-01-14
+- **Authors:** Xin Xia, Hongzhi Yin, Shane Culpepper
+- **abs:** https://arxiv.org/abs/2601.09306 · **PDF:** https://arxiv.org/pdf/2601.09306v1
+
+## 21830. Single-Round Clustered Federated Learning via Data Collaboration Analysis for Non-IID Data
+
+- **ID:** `2601.09304` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Sota Sugawara, Yuji Kawamata, Akihiro Toyoda, et al.
+- **abs:** https://arxiv.org/abs/2601.09304 · **PDF:** https://arxiv.org/pdf/2601.09304v1
+
+## 21831. AI Survival Stories: a Taxonomic Analysis of AI Existential Risk
+
+- **ID:** `2601.09765` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Herman Cappelen, Simon Goldstein, John Hawthorne
+- **abs:** https://arxiv.org/abs/2601.09765 · **PDF:** https://arxiv.org/pdf/2601.09765v1
+
+## 21832. Policy-Based Reinforcement Learning with Action Masking for Dynamic Job Shop Scheduling under Uncertainty: Handling Random Arrivals and Machine Failures
+
+- **ID:** `2601.09293` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Sofiene Lassoued, Stefan Lier, Andreas Schwung
+- **abs:** https://arxiv.org/abs/2601.09293 · **PDF:** https://arxiv.org/pdf/2601.09293v1
+
+## 21833. Blue Teaming Function-Calling Agents
+
+- **ID:** `2601.09292` · **primary:** `cs.CR` · **Submitted:** 2026-01-14
+- **Authors:** Greta Dolcetti, Giulio Zizzo, Sergio Maffeis
+- **abs:** https://arxiv.org/abs/2601.09292 · **PDF:** https://arxiv.org/pdf/2601.09292v1
+
+## 21834. Explainable Autoencoder-Based Anomaly Detection in IEC 61850 GOOSE Networks
+
+- **ID:** `2601.09287` · **primary:** `cs.CR` · **Submitted:** 2026-01-14
+- **Authors:** Dafne Lozano-Paredes, Luis Bote-Curiel, Juan Ramón Feijóo-Martínez, et al.
+- **abs:** https://arxiv.org/abs/2601.09287 · **PDF:** https://arxiv.org/pdf/2601.09287v1
+
+## 21835. Why not Collaborative Filtering in Dual View? Bridging Sparse and Dense Models
+
+- **ID:** `2601.09286` · **primary:** `cs.IR` · **Submitted:** 2026-01-14
+- **Authors:** Hanze Guo, Jianxun Lian, Xiao Zhou
+- **abs:** https://arxiv.org/abs/2601.09286 · **PDF:** https://arxiv.org/pdf/2601.09286v1
+
+## 21836. Enhancing Spatial Reasoning in Large Language Models for Metal-Organic Frameworks Structure Prediction
+
+- **ID:** `2601.09285` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Mianzhi Pan, JianFei Li, Peishuo Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.09285 · **PDF:** https://arxiv.org/pdf/2601.09285v1
+
+## 21837. Cluster Workload Allocation: Semantic Soft Affinity Using Natural Language Processing
+
+- **ID:** `2601.09282` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Leszek Sliwko, Jolanta Mizeria-Pietraszko
+- **abs:** https://arxiv.org/abs/2601.09282 · **PDF:** https://arxiv.org/pdf/2601.09282v2
+
+## 21838. STaR: Sensitive Trajectory Regulation for Unlearning in Large Reasoning Models
+
+- **ID:** `2601.09281` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Jingjing Zhou, Gaoxiang Cong, Li Su, et al.
+- **abs:** https://arxiv.org/abs/2601.09281 · **PDF:** https://arxiv.org/pdf/2601.09281v1
+
+## 21839. ReGraM: Region-First Knowledge Graph Reasoning for Medical Question Answering
+
+- **ID:** `2601.09280` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Chaerin Lee, Sohee Park, Hyunsik Na, et al.
+- **abs:** https://arxiv.org/abs/2601.09280 · **PDF:** https://arxiv.org/pdf/2601.09280v1
+
+## 21840. Explicating Tacit Regulatory Knowledge from LLMs to Auto-Formalize Requirements for Compliance Test Case Generation
+
+- **ID:** `2601.09762` · **primary:** `cs.SE` · **Submitted:** 2026-01-14
+- **Authors:** Zhiyi Xue, Xiaohong Chen, Min Zhang
+- **abs:** https://arxiv.org/abs/2601.09762 · **PDF:** https://arxiv.org/pdf/2601.09762v1
+
+## 21841. M$^3$Searcher: Modular Multimodal Information Seeking Agency with Retrieval-Oriented Reasoning
+
+- **ID:** `2601.09278` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Xiaohan Yu, Chao Feng, Lang Mei, et al.
+- **abs:** https://arxiv.org/abs/2601.09278 · **PDF:** https://arxiv.org/pdf/2601.09278v1
+
+## 21842. LSR-Net: A Lightweight and Strong Robustness Network for Bearing Fault Diagnosis in Noise Environment
+
+- **ID:** `2601.10761` · **primary:** `eess.SP` · **Submitted:** 2026-01-14
+- **Authors:** Junseok Lee, Jihye Shin, Sangyong Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.10761 · **PDF:** https://arxiv.org/pdf/2601.10761v1
+
+## 21843. $A^3$-Bench: Benchmarking Memory-Driven Scientific Reasoning via Anchor and Attractor Activation
+
+- **ID:** `2601.09274` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Jian Zhang, Yu He, Zhiyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.09274 · **PDF:** https://arxiv.org/pdf/2601.09274v1
+
+## 21844. MCGA: A Multi-task Classical Chinese Literary Genre Audio Corpus
+
+- **ID:** `2601.09270` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Yexing Du, Kaiyuan Liu, Bihe Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.09270 · **PDF:** https://arxiv.org/pdf/2601.09270v2
+
+## 21845. RISER: Orchestrating Latent Reasoning Skills for Adaptive Activation Steering
+
+- **ID:** `2601.09269` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Wencheng Ye, Xiaoyang Yuan, Yi Bin, et al.
+- **abs:** https://arxiv.org/abs/2601.09269 · **PDF:** https://arxiv.org/pdf/2601.09269v2
+
+## 21846. Coordinated Pandemic Control with Large Language Model Agents as Policymaking Assistants
+
+- **ID:** `2601.09264` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Ziyi Shi, Xusen Guo, Hongliang Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.09264 · **PDF:** https://arxiv.org/pdf/2601.09264v1
+
+## 21847. Magnifying change: Rapid burn scar mapping with multi-resolution, multi-source satellite imagery
+
+- **ID:** `2601.09262` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Maria Sdraka, Dimitrios Michail, Ioannis Papoutsis
+- **abs:** https://arxiv.org/abs/2601.09262 · **PDF:** https://arxiv.org/pdf/2601.09262v1
+
+## 21848. Learning to Trust Experience: A Monitor-Trust-Regulator Framework for Learning under Unobservable Feedback Reliability
+
+- **ID:** `2601.09261` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Zhipeng Zhang, Zhenjie Yao, Kai Li, et al.
+- **abs:** https://arxiv.org/abs/2601.09261 · **PDF:** https://arxiv.org/pdf/2601.09261v2
+
+## 21849. Efficient Paths and Dense Rewards: Probabilistic Flow Reasoning for Large Language Models
+
+- **ID:** `2601.09260` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Yan Liu, Feng Zhang, Zhanyu Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.09260 · **PDF:** https://arxiv.org/pdf/2601.09260v1
+
+## 21850. MAXS: Meta-Adaptive Exploration with LLM Agents
+
+- **ID:** `2601.09259` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Jian Zhang, Zhiyuan Wang, Zhangqi Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.09259 · **PDF:** https://arxiv.org/pdf/2601.09259v1
+
+## 21851. PersonaPlex: Voice and Role Control for Full Duplex Conversational Speech Models
+
+- **ID:** `2602.06053` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Rajarshi Roy, Jonathan Raiman, Sang-gil Lee, et al.
+- **abs:** https://arxiv.org/abs/2602.06053 · **PDF:** https://arxiv.org/pdf/2602.06053v1
+
+## 21852. LatencyPrism: Online Non-intrusive Latency Sculpting for SLO-Guaranteed LLM Inference
+
+- **ID:** `2601.09258` · **primary:** `cs.DC` · **Submitted:** 2026-01-14
+- **Authors:** Yin Du, Jiayi Ren, Xiayu Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.09258 · **PDF:** https://arxiv.org/pdf/2601.09258v2
+
+## 21853. RIFT: Repurposing Negative Samples via Reward-Informed Fine-Tuning
+
+- **ID:** `2601.09253` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Zehua Liu, Shuqi Liu, Tao Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.09253 · **PDF:** https://arxiv.org/pdf/2601.09253v1
+
+## 21854. Rethinking Memory Mechanisms of Foundation Agents in the Second Half: A Survey
+
+- **ID:** `2602.06052` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Wei-Chieh Huang, Weizhi Zhang, Yueqing Liang, et al.
+- **abs:** https://arxiv.org/abs/2602.06052 · **PDF:** https://arxiv.org/pdf/2602.06052v3
+
+## 21855. HGATSolver: A Heterogeneous Graph Attention Solver for Fluid-Structure Interaction
+
+- **ID:** `2601.09251` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Qin-Yi Zhang, Hong Wang, Siyao Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.09251 · **PDF:** https://arxiv.org/pdf/2601.09251v1
+
+## 21856. When to Invoke: Refining LLM Fairness with Toxicity Assessment
+
+- **ID:** `2601.09250` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Jing Ren, Bowen Li, Ziqi Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.09250 · **PDF:** https://arxiv.org/pdf/2601.09250v1
+
+## 21857. Hybrid guided variational autoencoder for visual place recognition
+
+- **ID:** `2601.09248` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Ni Wang, Zihan You, Emre Neftci, et al.
+- **abs:** https://arxiv.org/abs/2601.09248 · **PDF:** https://arxiv.org/pdf/2601.09248v1
+
+## 21858. MathMixup: Boosting LLM Mathematical Reasoning with Difficulty-Controllable Data Synthesis and Curriculum Learning
+
+- **ID:** `2601.17006` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Xuchen Li, Jing Chen, Xuzhao Li, et al.
+- **abs:** https://arxiv.org/abs/2601.17006 · **PDF:** https://arxiv.org/pdf/2601.17006v1
+
+## 21859. TeachPro: Multi-Label Qualitative Teaching Evaluation via Cross-View Graph Synergy and Semantic Anchored Evidence Encoding
+
+- **ID:** `2601.09246` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Xiangqian Wang, Yifan Jia, Yang Xiang, et al.
+- **abs:** https://arxiv.org/abs/2601.09246 · **PDF:** https://arxiv.org/pdf/2601.09246v1
+
+## 21860. When to Trust: A Causality-Aware Calibration Framework for Accurate Knowledge Graph Retrieval-Augmented Generation
+
+- **ID:** `2601.09241` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Jing Ren, Bowen Li, Ziqi Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.09241 · **PDF:** https://arxiv.org/pdf/2601.09241v2
+
+## 21861. DSA-Tokenizer: Disentangled Semantic-Acoustic Tokenization via Flow Matching-based Hierarchical Fusion
+
+- **ID:** `2601.09239` · **primary:** `cs.SD` · **Submitted:** 2026-01-14
+- **Authors:** Hanlin Zhang, Daxin Tan, Dehua Tao, et al.
+- **abs:** https://arxiv.org/abs/2601.09239 · **PDF:** https://arxiv.org/pdf/2601.09239v2
+
+## 21862. XLinear: A Lightweight and Accurate MLP-Based Model for Long-Term Time Series Forecasting with Exogenous Inputs
+
+- **ID:** `2601.09237` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Xinyang Chen, Huidong Jin, Yu Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.09237 · **PDF:** https://arxiv.org/pdf/2601.09237v1
+
+## 21863. Reward Learning through Ranking Mean Squared Error
+
+- **ID:** `2601.09236` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Chaitanya Kharyal, Calarina Muslimani, Matthew E. Taylor
+- **abs:** https://arxiv.org/abs/2601.09236 · **PDF:** https://arxiv.org/pdf/2601.09236v2
+
+## 21864. STEMVerse: A Dual-Axis Diagnostic Framework for STEM Reasoning in Large Language Models
+
+- **ID:** `2602.02497` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Xuzhao Li, Xuchen Li, Jian Zhao, et al.
+- **abs:** https://arxiv.org/abs/2602.02497 · **PDF:** https://arxiv.org/pdf/2602.02497v1
+
+## 21865. GIFT: Reconciling Post-Training Objectives via Finite-Temperature Gibbs Initialization
+
+- **ID:** `2601.09233` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Zhengyang Zhao, Lu Ma, Yizhen Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.09233 · **PDF:** https://arxiv.org/pdf/2601.09233v2
+
+## 21866. From Hawkes Processes to Attention: Time-Modulated Mechanisms for Event Sequences
+
+- **ID:** `2601.09220` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Xinzi Tan, Kejian Zhang, Junhan Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.09220 · **PDF:** https://arxiv.org/pdf/2601.09220v2
+
+## 21867. UserLM-R1: Modeling Human Reasoning in User Language Models with Multi-Reward Reinforcement Learning
+
+- **ID:** `2601.09215` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Feng Zhang, Shijia Li, Chunmao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.09215 · **PDF:** https://arxiv.org/pdf/2601.09215v1
+
+## 21868. SpikeVAEDiff: Neural Spike-based Natural Visual Scene Reconstruction via VD-VAE and Versatile Diffusion
+
+- **ID:** `2601.09213` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Jialu Li, Taiyan Zhou
+- **abs:** https://arxiv.org/abs/2601.09213 · **PDF:** https://arxiv.org/pdf/2601.09213v1
+
+## 21869. Annealed Relaxation of Speculative Decoding for Faster Autoregressive Image Generation
+
+- **ID:** `2601.09212` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Xingyao Li, Fengzhuo Zhang, Cunxiao Du, et al.
+- **abs:** https://arxiv.org/abs/2601.09212 · **PDF:** https://arxiv.org/pdf/2601.09212v1
+
+## 21870. Mikasa: A Character-Driven Emotional AI Companion Inspired by Japanese Oshi Culture
+
+- **ID:** `2601.09208` · **primary:** `cs.HC` · **Submitted:** 2026-01-14
+- **Authors:** Miki Ueno
+- **abs:** https://arxiv.org/abs/2601.09208 · **PDF:** https://arxiv.org/pdf/2601.09208v2
+
+## 21871. A.X K1 Technical Report
+
+- **ID:** `2601.09200` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Sung Jun Cheon, Jaekyung Cho, Seongho Choi, et al.
+- **abs:** https://arxiv.org/abs/2601.09200 · **PDF:** https://arxiv.org/pdf/2601.09200v5
+
+## 21872. NavFormer: IGRF Forecasting in Moving Coordinate Frames
+
+- **ID:** `2601.18800` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Yoontae Hwang, Dongwoo Lee, Minseok Choi, et al.
+- **abs:** https://arxiv.org/abs/2601.18800 · **PDF:** https://arxiv.org/pdf/2601.18800v2
+
+## 21873. ProFit: Leveraging High-Value Signals in SFT via Probability-Guided Token Selection
+
+- **ID:** `2601.09195` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Tao Liu, Taiqiang Wu, Runming Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.09195 · **PDF:** https://arxiv.org/pdf/2601.09195v2
+
+## 21874. OrthoGeoLoRA: Geometric Parameter-Efficient Fine-Tuning for Structured Social Science Concept Retrieval on theWeb
+
+- **ID:** `2601.09185` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Zeqiang Wang, Xinyue Wu, Chenxi Li, et al.
+- **abs:** https://arxiv.org/abs/2601.09185 · **PDF:** https://arxiv.org/pdf/2601.09185v1
+
+## 21875. Position on LLM-Assisted Peer Review: Addressing Reviewer Gap through Mentoring and Feedback
+
+- **ID:** `2601.09182` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** JungMin Yun, JuneHyoung Kwon, MiHyeon Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.09182 · **PDF:** https://arxiv.org/pdf/2601.09182v1
+
+## 21876. From Noise to Insights: Enhancing Supply Chain Decision Support through AI-Based Survey Integrity Analytics
+
+- **ID:** `2601.17005` · **primary:** `cs.CY` · **Submitted:** 2026-01-14
+- **Authors:** Bhubalan Mani
+- **abs:** https://arxiv.org/abs/2601.17005 · **PDF:** https://arxiv.org/pdf/2601.17005v1
+
+## 21877. $D^2Prune$: Sparsifying Large Language Models via Dual Taylor Expansion and Attention Distribution Awareness
+
+- **ID:** `2601.09176` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Lang Xiong, Ning Liu, Ao Ren, et al.
+- **abs:** https://arxiv.org/abs/2601.09176 · **PDF:** https://arxiv.org/pdf/2601.09176v1
+
+## 21878. Geometric Stability: The Missing Axis of Representations
+
+- **ID:** `2601.09173` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Prashant C. Raju
+- **abs:** https://arxiv.org/abs/2601.09173 · **PDF:** https://arxiv.org/pdf/2601.09173v3
+
+## 21879. BalDRO: A Distributionally Robust Optimization based Framework for Large Language Model Unlearning
+
+- **ID:** `2601.09172` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Pengyang Shao, Naixin Zhai, Lei Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.09172 · **PDF:** https://arxiv.org/pdf/2601.09172v2
+
+## 21880. N-EIoU-YOLOv9: A Signal-Aware Bounding Box Regression Loss for Lightweight Mobile Detection of Rice Leaf Diseases
+
+- **ID:** `2601.09170` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Dung Ta Nguyen Duc, Thanh Bui Dang, Hoang Le Minh, et al.
+- **abs:** https://arxiv.org/abs/2601.09170 · **PDF:** https://arxiv.org/pdf/2601.09170v1
+
+## 21881. DP-FedSOFIM: Differentially Private Federated Stochastic Optimization using Regularized Fisher Information Matrix
+
+- **ID:** `2601.09166` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Sidhant Nair, Tanmay Sen, Mrinmay Sen, et al.
+- **abs:** https://arxiv.org/abs/2601.09166 · **PDF:** https://arxiv.org/pdf/2601.09166v2
+
+## 21882. Multi-Teacher Ensemble Distillation: A Mathematical Framework for Probability-Domain Knowledge Aggregation
+
+- **ID:** `2601.09165` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Aaron R. Flouro, Shawn P. Chadwick
+- **abs:** https://arxiv.org/abs/2601.09165 · **PDF:** https://arxiv.org/pdf/2601.09165v1
+
+## 21883. Rethinking LLM-Driven Heuristic Design: Generating Efficient and Specialized Solvers via Dynamics-Aware Optimization
+
+- **ID:** `2601.20868` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Rongzheng Wang, Yihong Huang, Muquan Li, et al.
+- **abs:** https://arxiv.org/abs/2601.20868 · **PDF:** https://arxiv.org/pdf/2601.20868v1
+
+## 21884. CAST: Character-and-Scene Episodic Memory for Agents
+
+- **ID:** `2602.06051` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Kexin Ma, Bojun Li, Yuhua Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.06051 · **PDF:** https://arxiv.org/pdf/2602.06051v3
+
+## 21885. Efficient Clustering in Stochastic Bandits
+
+- **ID:** `2601.09162` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** G Dhinesh Chandran, Kota Srinivas Reddy, Srikrishna Bhashyam
+- **abs:** https://arxiv.org/abs/2601.09162 · **PDF:** https://arxiv.org/pdf/2601.09162v1
+
+## 21886. Deep Learning-based Binary Analysis for Vulnerability Detection in x86-64 Machine Code
+
+- **ID:** `2601.09157` · **primary:** `cs.CR` · **Submitted:** 2026-01-14
+- **Authors:** Mitchell Petingola
+- **abs:** https://arxiv.org/abs/2601.09157 · **PDF:** https://arxiv.org/pdf/2601.09157v1
+
+## 21887. KTCF: Actionable Recourse in Knowledge Tracing via Counterfactual Explanations for Education
+
+- **ID:** `2601.09156` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Woojin Kim, Changkwon Lee, Hyeoncheol Kim
+- **abs:** https://arxiv.org/abs/2601.09156 · **PDF:** https://arxiv.org/pdf/2601.09156v1
+
+## 21888. PrivacyReasoner: Can LLM Emulate a Human-like Privacy Mind?
+
+- **ID:** `2601.09152` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Yiwen Tu, Xuan Liu, Lianhui Qin, et al.
+- **abs:** https://arxiv.org/abs/2601.09152 · **PDF:** https://arxiv.org/pdf/2601.09152v1
+
+## 21889. Interpretable Probability Estimation with LLMs via Shapley Reconstruction
+
+- **ID:** `2601.09151` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Yang Nan, Qihao Wen, Jiahao Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.09151 · **PDF:** https://arxiv.org/pdf/2601.09151v1
+
+## 21890. SSVP: Synergistic Semantic-Visual Prompting for Industrial Zero-Shot Anomaly Detection
+
+- **ID:** `2601.09147` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Chenhao Fu, Han Fang, Xiuzheng Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.09147 · **PDF:** https://arxiv.org/pdf/2601.09147v2
+
+## 21891. Chain-of-Memory: Lightweight Memory Construction with Dynamic Evolution for LLM Agents
+
+- **ID:** `2601.14287` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Xiucheng Xu, Bingbing Xu, Xueyun Tian, et al.
+- **abs:** https://arxiv.org/abs/2601.14287 · **PDF:** https://arxiv.org/pdf/2601.14287v1
+
+## 21892. Discrete Solution Operator Learning for Geometry-Dependent PDEs
+
+- **ID:** `2601.09143` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Jinshuai Bai, Haolin Li, Zahra Sharif Khodaei, et al.
+- **abs:** https://arxiv.org/abs/2601.09143 · **PDF:** https://arxiv.org/pdf/2601.09143v3
+
+## 21893. Beyond Simulations: What 20,000 Real Conversations Reveal About Mental Health AI Safety
+
+- **ID:** `2601.17003` · **primary:** `cs.CY` · **Submitted:** 2026-01-14
+- **Authors:** Caitlin A. Stamatis, Jonah Meyerhoff, Richard Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.17003 · **PDF:** https://arxiv.org/pdf/2601.17003v1
+
+## 21894. EvasionBench: A Large-Scale Benchmark for Detecting Managerial Evasion in Earnings Call Q&A
+
+- **ID:** `2601.09142` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Shijian Ma, Yan Lin, Yi Yang
+- **abs:** https://arxiv.org/abs/2601.09142 · **PDF:** https://arxiv.org/pdf/2601.09142v2
+
+## 21895. Identity-Robust Language Model Generation via Content Integrity Preservation
+
+- **ID:** `2601.09141` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Miao Zhang, Kelly Chen, Md Mehrab Tanjim, et al.
+- **abs:** https://arxiv.org/abs/2601.09141 · **PDF:** https://arxiv.org/pdf/2601.09141v1
+
+## 21896. SkinFlow: Efficient Information Transmission for Open Dermatological Diagnosis via Dynamic Visual Encoding and Staged RL
+
+- **ID:** `2601.09136` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Lijun Liu, Linwei Chen, Zhishou Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.09136 · **PDF:** https://arxiv.org/pdf/2601.09136v1
+
+## 21897. Relevance-aware Multi-context Contrastive Decoding for Retrieval-augmented Visual Question Answering
+
+- **ID:** `2602.06050` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Jongha Kim, Byungoh Ko, Jeehye Na, et al.
+- **abs:** https://arxiv.org/abs/2602.06050 · **PDF:** https://arxiv.org/pdf/2602.06050v1
+
+## 21898. Equi-ViT: Rotational Equivariant Vision Transformer for Robust Histopathology Analysis
+
+- **ID:** `2601.09130` · **primary:** `eess.IV` · **Submitted:** 2026-01-14
+- **Authors:** Fuyao Chen, Yuexi Du, Elèonore V. Lieffrig, et al.
+- **abs:** https://arxiv.org/abs/2601.09130 · **PDF:** https://arxiv.org/pdf/2601.09130v1
+
+## 21899. Mass Distribution versus Density Distribution in the Context of Clustering
+
+- **ID:** `2601.10759` · **primary:** `stat.ML` · **Submitted:** 2026-01-14
+- **Authors:** Kai Ming Ting, Ye Zhu, Hang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.10759 · **PDF:** https://arxiv.org/pdf/2601.10759v2
+
+## 21900. Investigating Tool-Memory Conflicts in Tool-Augmented LLMs
+
+- **ID:** `2601.09760` · **primary:** `cs.SE` · **Submitted:** 2026-01-14
+- **Authors:** Jiali Cheng, Rui Pan, Hadi Amiri
+- **abs:** https://arxiv.org/abs/2601.09760 · **PDF:** https://arxiv.org/pdf/2601.09760v1
+
+## 21901. Adaptive Multi-Stage Patent Claim Generation with Unified Quality Assessment
+
+- **ID:** `2601.09120` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Chen-Wei Liang, Bin Guo, Zhen-Yuan Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.09120 · **PDF:** https://arxiv.org/pdf/2601.09120v1
+
+## 21902. Contrastive Bi-Encoder Models for Multi-Label Skill Extraction: Enhancing ESCO Ontology Matching with BERT and Attention Mechanisms
+
+- **ID:** `2601.09119` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Yongming Sun
+- **abs:** https://arxiv.org/abs/2601.09119 · **PDF:** https://arxiv.org/pdf/2601.09119v1
+
+## 21903. A Marketplace for AI-Generated Adult Content and Deepfakes
+
+- **ID:** `2601.09117` · **primary:** `cs.CY` · **Submitted:** 2026-01-14
+- **Authors:** Shalmoli Ghosh, Matthew R. DeVerna, Filippo Menczer
+- **abs:** https://arxiv.org/abs/2601.09117 · **PDF:** https://arxiv.org/pdf/2601.09117v2
+
+## 21904. LP-LLM: End-to-End Real-World Degraded License Plate Text Recognition via Large Multimodal Models
+
+- **ID:** `2601.09116` · **primary:** `cs.CV` · **Submitted:** 2026-01-14
+- **Authors:** Haoyan Gong, Hongbin Liu
+- **abs:** https://arxiv.org/abs/2601.09116 · **PDF:** https://arxiv.org/pdf/2601.09116v1
+
+## 21905. A Machine Learning Approach Towards Runtime Optimisation of Matrix Multiplication
+
+- **ID:** `2601.09114` · **primary:** `cs.DC` · **Submitted:** 2026-01-14
+- **Authors:** Yufan Xia, Marco De La Pierre, Amanda S. Barnard, et al.
+- **abs:** https://arxiv.org/abs/2601.09114 · **PDF:** https://arxiv.org/pdf/2601.09114v1
+
+## 21906. The AI Hippocampus: How Far are We From Human Memory?
+
+- **ID:** `2601.09113` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Zixia Jia, Jiaqi Li, Yipeng Kang, et al.
+- **abs:** https://arxiv.org/abs/2601.09113 · **PDF:** https://arxiv.org/pdf/2601.09113v1
+
+## 21907. RAM-SD: Retrieval-Augmented Multi-agent framework for Sarcasm Detection
+
+- **ID:** `2601.17002` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Ziyang Zhou, Ziqi Liu, Yan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.17002 · **PDF:** https://arxiv.org/pdf/2601.17002v1
+
+## 21908. AviationLMM: A Large Multimodal Foundation Model for Civil Aviation
+
+- **ID:** `2601.09105` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Wenbin Li, Jingling Wu, Xiaoyong Lin. Jing Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.09105 · **PDF:** https://arxiv.org/pdf/2601.09105v2
+
+## 21909. Enhancing Imbalanced Electrocardiogram Classification: A Novel Approach Integrating Data Augmentation through Wavelet Transform and Interclass Fusion
+
+- **ID:** `2601.09103` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Haijian Shao, Wei Liu, Xing Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.09103 · **PDF:** https://arxiv.org/pdf/2601.09103v1
+
+## 21910. DScheLLM: Enabling Dynamic Scheduling through a Fine-Tuned Dual-System Large language Model
+
+- **ID:** `2601.09100` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Lixiang Zhang, Chenggong Zhao, Qing Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.09100 · **PDF:** https://arxiv.org/pdf/2601.09100v2
+
+## 21911. Semantic Differentiation for Tackling Challenges in Watermarking Low-Entropy Constrained Generation Outputs
+
+- **ID:** `2601.11629` · **primary:** `cs.CR` · **Submitted:** 2026-01-14
+- **Authors:** Nghia T. Le, Alan Ritter, Kartik Goyal
+- **abs:** https://arxiv.org/abs/2601.11629 · **PDF:** https://arxiv.org/pdf/2601.11629v1
+
+## 21912. Programming over Thinking: Efficient and Robust Multi-Constraint Planning
+
+- **ID:** `2601.09097` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Derrick Goh Xin Deik, Quanyu Long, Zhengyuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.09097 · **PDF:** https://arxiv.org/pdf/2601.09097v2
+
+## 21913. Comparative Assessment of Concrete Compressive Strength Prediction at Industry Scale Using Embedding-based Neural Networks, Transformers, and Traditional Machine Learning Approaches
+
+- **ID:** `2601.09096` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Md Asiful Islam, Md Ahmed Al Muzaddid, Afia Jahin Prema, et al.
+- **abs:** https://arxiv.org/abs/2601.09096 · **PDF:** https://arxiv.org/pdf/2601.09096v1
+
+## 21914. Hidden States as Early Signals: Step-level Trace Evaluation and Pruning for Efficient Test-Time Scaling
+
+- **ID:** `2601.09093` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Zhixiang Liang, Beichen Huang, Zheng Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.09093 · **PDF:** https://arxiv.org/pdf/2601.09093v1
+
+## 21915. SubTokenTest: A Practical Benchmark for Real-World Sub-token Understanding
+
+- **ID:** `2601.09089` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Shuyang Hou, Yi Hu, Muhan Zhang
+- **abs:** https://arxiv.org/abs/2601.09089 · **PDF:** https://arxiv.org/pdf/2601.09089v1
+
+## 21916. Distribution-Aligned Sequence Distillation for Superior Long-CoT Reasoning
+
+- **ID:** `2601.09088` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Shaotian Yan, Kaiyuan Liu, Chen Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.09088 · **PDF:** https://arxiv.org/pdf/2601.09088v1
+
+## 21917. MMR-GRPO: Accelerating GRPO-Style Training through Diversity-Aware Reward Reweighting
+
+- **ID:** `2601.09085` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Kangda Wei, Ruihong Huang
+- **abs:** https://arxiv.org/abs/2601.09085 · **PDF:** https://arxiv.org/pdf/2601.09085v1
+
+## 21918. How Many Human Judgments Are Enough? Feasibility Limits of Human Preference Evaluation
+
+- **ID:** `2601.09084` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Wilson Y. Lee
+- **abs:** https://arxiv.org/abs/2601.09084 · **PDF:** https://arxiv.org/pdf/2601.09084v2
+
+## 21919. SRT: Accelerating Reinforcement Learning via Speculative Rollout with Tree-Structured Cache
+
+- **ID:** `2601.09083` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Chi-Chih Chang, Siqi Zhu, Zhichen Zeng, et al.
+- **abs:** https://arxiv.org/abs/2601.09083 · **PDF:** https://arxiv.org/pdf/2601.09083v1
+
+## 21920. Knowledge Boundary Discovery for Large Language Models
+
+- **ID:** `2603.21022` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Ziquan Wang, Zhongqi Lu
+- **abs:** https://arxiv.org/abs/2603.21022 · **PDF:** https://arxiv.org/pdf/2603.21022v1
+
+## 21921. Lean Clients, Full Accuracy: Hybrid Zeroth- and First-Order Split Federated Learning
+
+- **ID:** `2601.09076` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Zhoubin Kou, Zihan Chen, Jing Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.09076 · **PDF:** https://arxiv.org/pdf/2601.09076v1
+
+## 21922. GNN-based Path-aware multi-view Circuit Learning for Technology Mapping
+
+- **ID:** `2601.14286` · **primary:** `cs.ET` · **Submitted:** 2026-01-14
+- **Authors:** Wentao Jiang, Jingxin Wang, Zhang Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.14286 · **PDF:** https://arxiv.org/pdf/2601.14286v1
+
+## 21923. DSL-R1: From SQL to DSL for Training Retrieval Agents across Structured and Unstructured Data with Reinforcement Learning
+
+- **ID:** `2603.21018` · **primary:** `cs.IR` · **Submitted:** 2026-01-14
+- **Authors:** Yunhai Hu, Junwei Zhou, Yumo Cao, et al.
+- **abs:** https://arxiv.org/abs/2603.21018 · **PDF:** https://arxiv.org/pdf/2603.21018v1
+
+## 21924. Human-AI Co-design for Clinical Prediction Models
+
+- **ID:** `2601.09072` · **primary:** `cs.AI` · **Submitted:** 2026-01-14
+- **Authors:** Jean Feng, Avni Kothari, Patrick Vossler, et al.
+- **abs:** https://arxiv.org/abs/2601.09072 · **PDF:** https://arxiv.org/pdf/2601.09072v1
+
+## 21925. Resolving Predictive Multiplicity for the Rashomon Set
+
+- **ID:** `2601.09071` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Parian Haghighat, Hadis Anahideh, Cynthia Rudin
+- **abs:** https://arxiv.org/abs/2601.09071 · **PDF:** https://arxiv.org/pdf/2601.09071v1
+
+## 21926. From Symbolic to Natural-Language Relations: Rethinking Knowledge Graph Construction in the Era of Large Language Models
+
+- **ID:** `2601.09069` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Kanyao Han, Yushang Lai
+- **abs:** https://arxiv.org/abs/2601.09069 · **PDF:** https://arxiv.org/pdf/2601.09069v1
+
+## 21927. Detecting Batch Heterogeneity via Likelihood Clustering
+
+- **ID:** `2601.09758` · **primary:** `q-bio.GN` · **Submitted:** 2026-01-14
+- **Authors:** Austin Talbot, Yue Ke
+- **abs:** https://arxiv.org/abs/2601.09758 · **PDF:** https://arxiv.org/pdf/2601.09758v1
+
+## 21928. Mi:dm 2.0 Korea-centric Bilingual Language Models
+
+- **ID:** `2601.09066` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Donghoon Shin, Sejung Lee, Soonmin Bae, et al.
+- **abs:** https://arxiv.org/abs/2601.09066 · **PDF:** https://arxiv.org/pdf/2601.09066v1
+
+## 21929. Beyond Consensus: Perspectivist Modeling and Evaluation of Annotator Disagreement in NLP
+
+- **ID:** `2601.09065` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Yinuo Xu, David Jurgens
+- **abs:** https://arxiv.org/abs/2601.09065 · **PDF:** https://arxiv.org/pdf/2601.09065v2
+
+## 21930. Efficient Multilingual Dialogue Processing via Translation Pipelines and Distilled Language Models
+
+- **ID:** `2601.09059` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Santiago Martínez Novoa, Nicolás Rozo Fajardo, Diego Alejandro González Vargas, et al.
+- **abs:** https://arxiv.org/abs/2601.09059 · **PDF:** https://arxiv.org/pdf/2601.09059v1
+
+## 21931. SafeTalkCoach: Diversity-Driven Multi-Agent Simulation for Parent-Teen Health Conversations
+
+- **ID:** `2602.00017` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Benyamin Tabarsi, Wenbo Li, Tahreem Yasir, et al.
+- **abs:** https://arxiv.org/abs/2602.00017 · **PDF:** https://arxiv.org/pdf/2602.00017v1
+
+## 21932. StegoStylo: Squelching Stylometric Scrutiny through Steganographic Stitching
+
+- **ID:** `2601.09056` · **primary:** `cs.CR` · **Submitted:** 2026-01-14
+- **Authors:** Robert Dilworth
+- **abs:** https://arxiv.org/abs/2601.09056 · **PDF:** https://arxiv.org/pdf/2601.09056v2
+
+## 21933. Deep Incomplete Multi-View Clustering via Hierarchical Imputation and Alignment
+
+- **ID:** `2601.09051` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Yiming Du, Ziyu Wang, Jian Li, et al.
+- **abs:** https://arxiv.org/abs/2601.09051 · **PDF:** https://arxiv.org/pdf/2601.09051v1
+
+## 21934. SITA: Learning Speaker-Invariant and Tone-Aware Speech Representations for Low-Resource Tonal Languages
+
+- **ID:** `2601.09050` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Tianyi Xu, Xuan Ouyang, Binwei Yao, et al.
+- **abs:** https://arxiv.org/abs/2601.09050 · **PDF:** https://arxiv.org/pdf/2601.09050v1
+
+## 21935. The Hypocrisy Gap: Quantifying Divergence Between Internal Belief and Chain-of-Thought Explanation via Sparse Autoencoders
+
+- **ID:** `2602.02496` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Shikhar Shiromani, Archie Chaudhury, Sri Pranav Kunda
+- **abs:** https://arxiv.org/abs/2602.02496 · **PDF:** https://arxiv.org/pdf/2602.02496v1
+
+## 21936. Democracy and Distrust in an Era of Artificial Intelligence
+
+- **ID:** `2601.09757` · **primary:** `cs.CY` · **Submitted:** 2026-01-14
+- **Authors:** Sonia Katyal
+- **abs:** https://arxiv.org/abs/2601.09757 · **PDF:** https://arxiv.org/pdf/2601.09757v1
+
+## 21937. Is Grokking Worthwhile? Functional Analysis and Transferability of Generalization Circuits in Transformers
+
+- **ID:** `2601.09049` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Kaiyu He, Zhang Mian, Peilin Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.09049 · **PDF:** https://arxiv.org/pdf/2601.09049v1
+
+## 21938. Horseshoe Mixtures-of-Experts (HS-MoE)
+
+- **ID:** `2601.09043` · **primary:** `stat.ML` · **Submitted:** 2026-01-14
+- **Authors:** Nick Polson, Vadim Sokolov
+- **abs:** https://arxiv.org/abs/2601.09043 · **PDF:** https://arxiv.org/pdf/2601.09043v1
+
+## 21939. SCaLE: Switching Cost aware Learning and Exploration
+
+- **ID:** `2601.09042` · **primary:** `cs.LG` · **Submitted:** 2026-01-14
+- **Authors:** Neelkamal Bhuyan, Debankur Mukherjee, Adam Wierman
+- **abs:** https://arxiv.org/abs/2601.09042 · **PDF:** https://arxiv.org/pdf/2601.09042v1
+
+## 21940. Can LLMs interpret figurative language as humans do?: surface-level vs representational similarity
+
+- **ID:** `2601.09041` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Samhita Bollepally, Aurora Sloman-Moll, Takashi Yamauchi
+- **abs:** https://arxiv.org/abs/2601.09041 · **PDF:** https://arxiv.org/pdf/2601.09041v1
+
+## 21941. SpectraQuery: A Hybrid Retrieval-Augmented Conversational Assistant for Battery Science
+
+- **ID:** `2601.09036` · **primary:** `cs.CL` · **Submitted:** 2026-01-14
+- **Authors:** Sreya Vangara, Jagjit Nanda, Yan-Kai Tzeng, et al.
+- **abs:** https://arxiv.org/abs/2601.09036 · **PDF:** https://arxiv.org/pdf/2601.09036v1
+
+## 21942. A Decompilation-Driven Framework for Malware Detection with Large Language Models
+
+- **ID:** `2601.09035` · **primary:** `cs.CR` · **Submitted:** 2026-01-14
+- **Authors:** Aniesh Chawla, Udbhav Prasad
+- **abs:** https://arxiv.org/abs/2601.09035 · **PDF:** https://arxiv.org/pdf/2601.09035v1
+
+## 21943. The Hierarchy of Agentic Capabilities: Evaluating Frontier Models on Realistic RL Environments
+
+- **ID:** `2601.09032` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Logan Ritchie, Sushant Mehta, Nick Heiner, et al.
+- **abs:** https://arxiv.org/abs/2601.09032 · **PDF:** https://arxiv.org/pdf/2601.09032v1
+
+## 21944. Generalizable Geometric Prior and Recurrent Spiking Feature Learning for Humanoid Robot Manipulation
+
+- **ID:** `2601.09031` · **primary:** `cs.RO` · **Submitted:** 2026-01-13
+- **Authors:** Xuetao Li, Wenke Huang, Mang Ye, et al.
+- **abs:** https://arxiv.org/abs/2601.09031 · **PDF:** https://arxiv.org/pdf/2601.09031v1
+
+## 21945. Proactively Detecting Threats: A Novel Approach Using LLMs
+
+- **ID:** `2601.09029` · **primary:** `cs.CR` · **Submitted:** 2026-01-13
+- **Authors:** Aniesh Chawla, Udbhav Prasad
+- **abs:** https://arxiv.org/abs/2601.09029 · **PDF:** https://arxiv.org/pdf/2601.09029v1
+
+## 21946. OpenDecoder: Open Large Language Model Decoding to Incorporate Document Quality in RAG
+
+- **ID:** `2601.09028` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Fengran Mo, Zhan Su, Yuchen Hui, et al.
+- **abs:** https://arxiv.org/abs/2601.09028 · **PDF:** https://arxiv.org/pdf/2601.09028v2
+
+## 21947. Layer-Parallel Training for Transformers
+
+- **ID:** `2601.09026` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Shuai Jiang, Marc Salvadó-Benasco, Eric C. Cyr, et al.
+- **abs:** https://arxiv.org/abs/2601.09026 · **PDF:** https://arxiv.org/pdf/2601.09026v2
+
+## 21948. Universal Latent Homeomorphic Manifolds: A Framework for Cross-Domain Representation Unification
+
+- **ID:** `2601.09025` · **primary:** `eess.IV` · **Submitted:** 2026-01-13
+- **Authors:** Tong Wu, Tayab Uddin Wara, Daniel Hernandez, et al.
+- **abs:** https://arxiv.org/abs/2601.09025 · **PDF:** https://arxiv.org/pdf/2601.09025v2
+
+## 21949. An Inexact Weighted Proximal Trust-Region Method
+
+- **ID:** `2601.09024` · **primary:** `math.OC` · **Submitted:** 2026-01-13
+- **Authors:** Leandro Farias Maia, Robert Baraldi, Drew P. Kouri
+- **abs:** https://arxiv.org/abs/2601.09024 · **PDF:** https://arxiv.org/pdf/2601.09024v1
+
+## 21950. Tail-Sensitive KL and Rényi Convergence of Unadjusted Hamiltonian Monte Carlo via One-Shot Couplings
+
+- **ID:** `2601.09019` · **primary:** `stat.ML` · **Submitted:** 2026-01-13
+- **Authors:** Nawaf Bou-Rabee, Siddharth Mitra, Andre Wibisono
+- **abs:** https://arxiv.org/abs/2601.09019 · **PDF:** https://arxiv.org/pdf/2601.09019v1
+
+## 21951. Meta-learning to Address Data Shift in Time Series Classification
+
+- **ID:** `2601.09018` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Samuel Myren, Nidhi Parikh, Natalie Klein
+- **abs:** https://arxiv.org/abs/2601.09018 · **PDF:** https://arxiv.org/pdf/2601.09018v1
+
+## 21952. Multicultural Spyfall: Assessing LLMs through Dynamic Multilingual Social Deduction Game
+
+- **ID:** `2601.09017` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Haryo Akbarianto Wibowo, Alaa Elsetohy, Qinrong Cui, et al.
+- **abs:** https://arxiv.org/abs/2601.09017 · **PDF:** https://arxiv.org/pdf/2601.09017v1
+
+## 21953. TranslateGemma Technical Report
+
+- **ID:** `2601.09012` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Mara Finkelstein, Isaac Caswell, Tobias Domhan, et al.
+- **abs:** https://arxiv.org/abs/2601.09012 · **PDF:** https://arxiv.org/pdf/2601.09012v3
+
+## 21954. Block Decomposable Methods for Large-Scale Optimization Problems
+
+- **ID:** `2601.09010` · **primary:** `math.OC` · **Submitted:** 2026-01-13
+- **Authors:** Leandro Farias Maia
+- **abs:** https://arxiv.org/abs/2601.09010 · **PDF:** https://arxiv.org/pdf/2601.09010v1
+
+## 21955. Entropy Sentinel: Continuous LLM Accuracy Monitoring from Decoding Entropy Traces in STEM
+
+- **ID:** `2601.09001` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Pedro Memoli Buffa, Luciano Del Corro
+- **abs:** https://arxiv.org/abs/2601.09001 · **PDF:** https://arxiv.org/pdf/2601.09001v3
+
+## 21956. Universal Dynamics of Warmup Stable Decay: understanding WSD beyond Transformers
+
+- **ID:** `2601.09000` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Annalisa Belloni, Lorenzo Noci, Antonio Orvieto
+- **abs:** https://arxiv.org/abs/2601.09000 · **PDF:** https://arxiv.org/pdf/2601.09000v1
+
+## 21957. Physics-Guided Counterfactual Explanations for Large-Scale Multivariate Time Series: Application in Scalable and Interpretable SEP Event Prediction
+
+- **ID:** `2601.08999` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Pranjal Patil, Anli Ji, Berkay Aydin
+- **abs:** https://arxiv.org/abs/2601.08999 · **PDF:** https://arxiv.org/pdf/2601.08999v1
+
+## 21958. Optimising for Energy Efficiency and Performance in Machine Learning
+
+- **ID:** `2601.08991` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Emile Dos Santos Ferreira, Andrei Paleyes, Neil D. Lawrence
+- **abs:** https://arxiv.org/abs/2601.08991 · **PDF:** https://arxiv.org/pdf/2601.08991v2
+
+## 21959. ART: Action-based Reasoning Task Benchmarking for Medical AI Agents
+
+- **ID:** `2601.08988` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Ananya Mantravadi, Shivali Dalmia, Abhishek Mukherji
+- **abs:** https://arxiv.org/abs/2601.08988 · **PDF:** https://arxiv.org/pdf/2601.08988v1
+
+## 21960. Continuous Fairness On Data Streams
+
+- **ID:** `2601.08976` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Subhodeep Ghosh, Zhihui Du, Angela Bonifati, et al.
+- **abs:** https://arxiv.org/abs/2601.08976 · **PDF:** https://arxiv.org/pdf/2601.08976v1
+
+## 21961. Investigating Self-regulated Learning Sequences within a Generative AI-based Intelligent Tutoring System
+
+- **ID:** `2601.17000` · **primary:** `cs.CY` · **Submitted:** 2026-01-13
+- **Authors:** Jie Gao, Shasha Li, Jianhua Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.17000 · **PDF:** https://arxiv.org/pdf/2601.17000v1
+
+## 21962. Machine Learning-Driven Creep Law Discovery Across Alloy Compositional Space
+
+- **ID:** `2601.08970` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-13
+- **Authors:** Hongshun Chen, Ryan Zhou, Rujing Zha, et al.
+- **abs:** https://arxiv.org/abs/2601.08970 · **PDF:** https://arxiv.org/pdf/2601.08970v1
+
+## 21963. MLCBART: Multilabel Classification with Bayesian Additive Regression Trees
+
+- **ID:** `2601.08964` · **primary:** `stat.ME` · **Submitted:** 2026-01-13
+- **Authors:** Jiahao Tian, Hugh Chipman, Thomas Loughin
+- **abs:** https://arxiv.org/abs/2601.08964 · **PDF:** https://arxiv.org/pdf/2601.08964v1
+
+## 21964. Breaking the Bottlenecks: Scalable Diffusion Models for 3D Molecular Generation
+
+- **ID:** `2601.08963` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Adrita Das, Peiran Jiang, Dantong Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.08963 · **PDF:** https://arxiv.org/pdf/2601.08963v1
+
+## 21965. A Comparison of Polynomial-Based Tree Clustering Methods
+
+- **ID:** `2601.14285` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Pengyu Liu, Mariel Vázquez, Nataša Jonoska
+- **abs:** https://arxiv.org/abs/2601.14285 · **PDF:** https://arxiv.org/pdf/2601.14285v1
+
+## 21966. Imagine-then-Plan: Agent Learning from Adaptive Lookahead with World Models
+
+- **ID:** `2601.08955` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Youwei Liu, Jian Wang, Hanlin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.08955 · **PDF:** https://arxiv.org/pdf/2601.08955v2
+
+## 21967. Fairness risk and its privacy-enabled solution in AI-driven robotic applications
+
+- **ID:** `2601.08953` · **primary:** `cs.RO` · **Submitted:** 2026-01-13
+- **Authors:** Le Liu, Bangguo Yu, Nynke Vellinga, et al.
+- **abs:** https://arxiv.org/abs/2601.08953 · **PDF:** https://arxiv.org/pdf/2601.08953v1
+
+## 21968. PluriHarms: Benchmarking the Full Spectrum of Human Judgments on AI Harm
+
+- **ID:** `2601.08951` · **primary:** `cs.CY` · **Submitted:** 2026-01-13
+- **Authors:** Jing-Jing Li, Joel Mire, Eve Fleisig, et al.
+- **abs:** https://arxiv.org/abs/2601.08951 · **PDF:** https://arxiv.org/pdf/2601.08951v2
+
+## 21969. ConvoLearn: A Dataset of Constructivist Tutor-Student Dialogue
+
+- **ID:** `2601.08950` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Mayank Sharma, Roy Pea, Hari Subramonyam
+- **abs:** https://arxiv.org/abs/2601.08950 · **PDF:** https://arxiv.org/pdf/2601.08950v1
+
+## 21970. Synthetic Data for Veterinary EHR De-identification: Benefits, Limits, and Safety Trade-offs Under Fixed Compute
+
+- **ID:** `2601.09756` · **primary:** `cs.CR` · **Submitted:** 2026-01-13
+- **Authors:** David Brundage
+- **abs:** https://arxiv.org/abs/2601.09756 · **PDF:** https://arxiv.org/pdf/2601.09756v1
+
+## 21971. DriftGuard: A Hierarchical Framework for Concept Drift Detection and Remediation in Supply Chain Forecasting
+
+- **ID:** `2601.08928` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Shahnawaz Alam, Mohammed Abdul Rahman, Bareera Sadeqa
+- **abs:** https://arxiv.org/abs/2601.08928 · **PDF:** https://arxiv.org/pdf/2601.08928v1
+
+## 21972. Fine Grained Evaluation of LLMs-as-Judges
+
+- **ID:** `2601.08919` · **primary:** `cs.IR` · **Submitted:** 2026-01-13
+- **Authors:** Sourav Saha, Mandar Mitra
+- **abs:** https://arxiv.org/abs/2601.08919 · **PDF:** https://arxiv.org/pdf/2601.08919v1
+
+## 21973. Towards a Self-Driving Trigger at the LHC: Adaptive Response in Real Time
+
+- **ID:** `2601.08910` · **primary:** `physics.ins-det` · **Submitted:** 2026-01-13
+- **Authors:** Shaghayegh Emami, Cecilia Tosciri, Giovanna Salvi, et al.
+- **abs:** https://arxiv.org/abs/2601.08910 · **PDF:** https://arxiv.org/pdf/2601.08910v1
+
+## 21974. Modeling LLM Agent Reviewer Dynamics in Elo-Ranked Review System
+
+- **ID:** `2601.08829` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Hsiang-Wei Huang, Junbin Lu, Kuang-Ming Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.08829 · **PDF:** https://arxiv.org/pdf/2601.08829v1
+
+## 21975. Motion Attribution for Video Generation
+
+- **ID:** `2601.08828` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Xindi Wu, Despoina Paschalidou, Jun Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.08828 · **PDF:** https://arxiv.org/pdf/2601.08828v1
+
+## 21976. Navigating Ideation Space: Decomposed Conceptual Representations for Positioning Scientific Ideas
+
+- **ID:** `2601.08901` · **primary:** `cs.IR` · **Submitted:** 2026-01-13
+- **Authors:** Yuexi Shen, Minqian Liu, Dawei Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.08901 · **PDF:** https://arxiv.org/pdf/2601.08901v1
+
+## 21977. MemRec: Collaborative Memory-Augmented Agentic Recommender System
+
+- **ID:** `2601.08816` · **primary:** `cs.IR` · **Submitted:** 2026-01-13
+- **Authors:** Weixin Chen, Yuhan Zhao, Jingyuan Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.08816 · **PDF:** https://arxiv.org/pdf/2601.08816v2
+
+## 21978. Reasoning Matters for 3D Visual Grounding
+
+- **ID:** `2601.08811` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Hsiang-Wei Huang, Kuang-Ming Chen, Wenhao Chai, et al.
+- **abs:** https://arxiv.org/abs/2601.08811 · **PDF:** https://arxiv.org/pdf/2601.08811v1
+
+## 21979. Multiplex Thinking: Reasoning via Token-wise Branch-and-Merge
+
+- **ID:** `2601.08808` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Yao Tang, Li Dong, Yaru Hao, et al.
+- **abs:** https://arxiv.org/abs/2601.08808 · **PDF:** https://arxiv.org/pdf/2601.08808v1
+
+## 21980. S3-CLIP: Video Super Resolution for Person-ReID
+
+- **ID:** `2601.08807` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Tamas Endrei, Gyorgy Cserey
+- **abs:** https://arxiv.org/abs/2601.08807 · **PDF:** https://arxiv.org/pdf/2601.08807v1
+
+## 21981. APEX-SWE
+
+- **ID:** `2601.08806` · **primary:** `cs.SE` · **Submitted:** 2026-01-13
+- **Authors:** Abhi Kottamasu, Chirag Mahapatra, Sam Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.08806 · **PDF:** https://arxiv.org/pdf/2601.08806v3
+
+## 21982. Uncovering Political Bias in Large Language Models using Parliamentary Voting Records
+
+- **ID:** `2601.08785` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Jieying Chen, Karen de Jong, Andreas Poole, et al.
+- **abs:** https://arxiv.org/abs/2601.08785 · **PDF:** https://arxiv.org/pdf/2601.08785v1
+
+## 21983. On the use of graph models to achieve individual and group fairness
+
+- **ID:** `2601.08784` · **primary:** `stat.ML` · **Submitted:** 2026-01-13
+- **Authors:** Arturo Pérez-Peralta, Sandra Benítez-Peña, Rosa E. Lillo
+- **abs:** https://arxiv.org/abs/2601.08784 · **PDF:** https://arxiv.org/pdf/2601.08784v1
+
+## 21984. Fast and explainable clustering in the Manhattan and Tanimoto distance
+
+- **ID:** `2601.08781` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Stefan Güttel, Kaustubh Roy
+- **abs:** https://arxiv.org/abs/2601.08781 · **PDF:** https://arxiv.org/pdf/2601.08781v1
+
+## 21985. Pervasive Annotation Errors Break Text-to-SQL Benchmarks and Leaderboards
+
+- **ID:** `2601.08778` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Tengjun Jin, Yoojin Choi, Yuxuan Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.08778 · **PDF:** https://arxiv.org/pdf/2601.08778v3
+
+## 21986. Asymptotic Universal Alignment: A New Alignment Framework via Test-Time Scaling
+
+- **ID:** `2601.08777` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Yang Cai, Weiqiang Zheng
+- **abs:** https://arxiv.org/abs/2601.08777 · **PDF:** https://arxiv.org/pdf/2601.08777v1
+
+## 21987. An Example for Domain Adaptation Using CycleGAN
+
+- **ID:** `2601.08776` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Yanhua Zhao
+- **abs:** https://arxiv.org/abs/2601.08776 · **PDF:** https://arxiv.org/pdf/2601.08776v3
+
+## 21988. Comprehensive Machine Learning Benchmarking for Fringe Projection Profilometry with Photorealistic Synthetic Data
+
+- **ID:** `2601.08900` · **primary:** `eess.IV` · **Submitted:** 2026-01-13
+- **Authors:** Anush Lakshman S, Adam Haroon, Beiwen Li
+- **abs:** https://arxiv.org/abs/2601.08900 · **PDF:** https://arxiv.org/pdf/2601.08900v2
+
+## 21989. Reliable Graph-RAG for Codebases: AST-Derived Graphs vs LLM-Extracted Knowledge Graphs
+
+- **ID:** `2601.08773` · **primary:** `cs.SE` · **Submitted:** 2026-01-13
+- **Authors:** Manideep Reddy Chinthareddy
+- **abs:** https://arxiv.org/abs/2601.08773 · **PDF:** https://arxiv.org/pdf/2601.08773v1
+
+## 21990. Uncertainty Quantification for Named Entity Recognition via Full-Sequence and Subsequence Conformal Prediction
+
+- **ID:** `2601.16999` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Matthew Singer, Srijan Sengupta, Karl Pazdernik
+- **abs:** https://arxiv.org/abs/2601.16999 · **PDF:** https://arxiv.org/pdf/2601.16999v1
+
+## 21991. AI as Entertainment
+
+- **ID:** `2601.08768` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Cody Kommers, Ari Holtzman
+- **abs:** https://arxiv.org/abs/2601.08768 · **PDF:** https://arxiv.org/pdf/2601.08768v1
+
+## 21992. Rewarding the Rare: Uniqueness-Aware RL for Creative Problem Solving in LLMs
+
+- **ID:** `2601.08763` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Zhiyuan Hu, Yucheng Wang, Yufei He, et al.
+- **abs:** https://arxiv.org/abs/2601.08763 · **PDF:** https://arxiv.org/pdf/2601.08763v2
+
+## 21993. Adaptive Requesting in Decentralized Edge Networks via Non-Stationary Bandits
+
+- **ID:** `2601.08760` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Yi Zhuang, Kun Yang, Xingran Chen
+- **abs:** https://arxiv.org/abs/2601.08760 · **PDF:** https://arxiv.org/pdf/2601.08760v3
+
+## 21994. Grid-Aware Charging and Operational Optimization for Mixed-Fleet Public Transit
+
+- **ID:** `2601.08753` · **primary:** `math.OC` · **Submitted:** 2026-01-13
+- **Authors:** Rishav Sen, Amutheezan Sivagnanam, Aron Laszka, et al.
+- **abs:** https://arxiv.org/abs/2601.08753 · **PDF:** https://arxiv.org/pdf/2601.08753v1
+
+## 21995. A Geolocation-Aware Multimodal Approach for Ecological Prediction
+
+- **ID:** `2601.08750` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Valerie Zermatten, Chiara Vanalli, Gencer Sumbul, et al.
+- **abs:** https://arxiv.org/abs/2601.08750 · **PDF:** https://arxiv.org/pdf/2601.08750v2
+
+## 21996. To Retrieve or To Think? An Agentic Approach for Context Evolution
+
+- **ID:** `2601.08747` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Rubing Chen, Jian Wang, Wenjie Li, et al.
+- **abs:** https://arxiv.org/abs/2601.08747 · **PDF:** https://arxiv.org/pdf/2601.08747v2
+
+## 21997. TableCache: Primary Foreign Key Guided KV Cache Precomputation for Low Latency Text-to-SQL
+
+- **ID:** `2601.08743` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Jinbo Su, Yuxuan Hu, Cuiping Li, et al.
+- **abs:** https://arxiv.org/abs/2601.08743 · **PDF:** https://arxiv.org/pdf/2601.08743v1
+
+## 21998. Inferring Latent Intentions: Attributional Natural Language Inference in LLM Agents
+
+- **ID:** `2601.08742` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Xin Quan, Jiafeng Xiong, Marco Valentino, et al.
+- **abs:** https://arxiv.org/abs/2601.08742 · **PDF:** https://arxiv.org/pdf/2601.08742v1
+
+## 21999. From Rows to Reasoning: A Retrieval-Augmented Multimodal Framework for Spreadsheet Understanding
+
+- **ID:** `2601.08741` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Anmol Gulati, Sahil Sen, Waqar Sarguroh, et al.
+- **abs:** https://arxiv.org/abs/2601.08741 · **PDF:** https://arxiv.org/pdf/2601.08741v2
+
+## 22000. PrivGemo: Privacy-Preserving Dual-Tower Graph Retrieval for Empowering LLM Reasoning with Memory Augmentation
+
+- **ID:** `2601.08739` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Xingyu Tan, Xiaoyang Wang, Qing Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.08739 · **PDF:** https://arxiv.org/pdf/2601.08739v1
+
+## 22001. TerraFormer: Automated Infrastructure-as-Code with LLMs Fine-Tuned via Policy-Guided Verifier Feedback
+
+- **ID:** `2601.08734` · **primary:** `cs.SE` · **Submitted:** 2026-01-13
+- **Authors:** Prithwish Jana, Sam Davidson, Bhavana Bhasker, et al.
+- **abs:** https://arxiv.org/abs/2601.08734 · **PDF:** https://arxiv.org/pdf/2601.08734v1
+
+## 22002. A Novel Approach to Explainable AI with Quantized Active Ingredients in Decision Making
+
+- **ID:** `2601.08733` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** A. M. A. S. D. Alagiyawanna, Asoka Karunananda, Thushari Silva, et al.
+- **abs:** https://arxiv.org/abs/2601.08733 · **PDF:** https://arxiv.org/pdf/2601.08733v1
+
+## 22003. ISLA: A U-Net for MRI-based acute ischemic stroke lesion segmentation with deep supervision, attention, domain adaptation, and ensemble learning
+
+- **ID:** `2601.08732` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Vincent Roca, Martin Bretzner, Hilde Henon, et al.
+- **abs:** https://arxiv.org/abs/2601.08732 · **PDF:** https://arxiv.org/pdf/2601.08732v1
+
+## 22004. Learning from Demonstrations via Capability-Aware Goal Sampling
+
+- **ID:** `2601.08731` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Yuanlin Duan, Yuning Wang, Wenjie Qiu, et al.
+- **abs:** https://arxiv.org/abs/2601.08731 · **PDF:** https://arxiv.org/pdf/2601.08731v1
+
+## 22005. Model-Agnostic Solutions for Deep Reinforcement Learning in Non-Ergodic Contexts
+
+- **ID:** `2601.08726` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Bert Verbruggen, Arne Vanhoyweghen, Vincent Ginis
+- **abs:** https://arxiv.org/abs/2601.08726 · **PDF:** https://arxiv.org/pdf/2601.08726v1
+
+## 22006. Kernel Learning for Regression via Quantum Annealing Based Spectral Sampling
+
+- **ID:** `2601.08724` · **primary:** `quant-ph` · **Submitted:** 2026-01-13
+- **Authors:** Yasushi Hasegawa, Masayuki Ohzeki
+- **abs:** https://arxiv.org/abs/2601.08724 · **PDF:** https://arxiv.org/pdf/2601.08724v1
+
+## 22007. Soft Partition-based KAPI-ELM for Multi-Scale PDEs
+
+- **ID:** `2601.08719` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Vikas Dwivedi, Monica Sigovan, Bruno Sixou
+- **abs:** https://arxiv.org/abs/2601.08719 · **PDF:** https://arxiv.org/pdf/2601.08719v1
+
+## 22008. Real-Time Localization Framework for Autonomous Basketball Robots
+
+- **ID:** `2601.08713` · **primary:** `cs.RO` · **Submitted:** 2026-01-13
+- **Authors:** Naren Medarametla, Sreejon Mondal
+- **abs:** https://arxiv.org/abs/2601.08713 · **PDF:** https://arxiv.org/pdf/2601.08713v1
+
+## 22009. Multi-Preconditioned LBFGS for Training Finite-Basis PINNs
+
+- **ID:** `2601.08709` · **primary:** `math.NA` · **Submitted:** 2026-01-13
+- **Authors:** Marc Salvadó-Benasco, Aymane Kssim, Alexander Heinlein, et al.
+- **abs:** https://arxiv.org/abs/2601.08709 · **PDF:** https://arxiv.org/pdf/2601.08709v2
+
+## 22010. RMBRec: Robust Multi-Behavior Recommendation towards Target Behaviors
+
+- **ID:** `2601.08705` · **primary:** `cs.IR` · **Submitted:** 2026-01-13
+- **Authors:** Miaomiao Cai, Zhijie Zhang, Junfeng Fang, et al.
+- **abs:** https://arxiv.org/abs/2601.08705 · **PDF:** https://arxiv.org/pdf/2601.08705v3
+
+## 22011. Evaluating the Ability of Explanations to Disambiguate Models in a Rashomon Set
+
+- **ID:** `2601.08703` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Kaivalya Rawal, Eoin Delaney, Zihao Fu, et al.
+- **abs:** https://arxiv.org/abs/2601.08703 · **PDF:** https://arxiv.org/pdf/2601.08703v1
+
+## 22012. RAGShaper: Eliciting Sophisticated Agentic RAG Skills via Automated Data Synthesis
+
+- **ID:** `2601.08699` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Zhengwei Tao, Bo Li, Jialong Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.08699 · **PDF:** https://arxiv.org/pdf/2601.08699v1
+
+## 22013. Auditing Student-AI Collaboration: A Case Study of Online Graduate CS Students
+
+- **ID:** `2601.08697` · **primary:** `cs.HC` · **Submitted:** 2026-01-13
+- **Authors:** Nifu Dan
+- **abs:** https://arxiv.org/abs/2601.08697 · **PDF:** https://arxiv.org/pdf/2601.08697v4
+
+## 22014. Beyond Affinity: A Benchmark of 1D, 2D, and 3D Methods Reveals Critical Trade-offs in Structure-Based Drug Design
+
+- **ID:** `2601.14283` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Kangyu Zheng, Kai Zhang, Jiale Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.14283 · **PDF:** https://arxiv.org/pdf/2601.14283v1
+
+## 22015. Enabling Population-Based Architectures for Neural Combinatorial Optimization
+
+- **ID:** `2601.08696` · **primary:** `cs.NE` · **Submitted:** 2026-01-13
+- **Authors:** Andoni Irazusta Garmendia, Josu Ceberio, Alexander Mendiburu
+- **abs:** https://arxiv.org/abs/2601.08696 · **PDF:** https://arxiv.org/pdf/2601.08696v1
+
+## 22016. Nationality and Region Prediction from Names: A Comparative Study of Neural Models and Large Language Models
+
+- **ID:** `2601.08692` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Keito Inoshita
+- **abs:** https://arxiv.org/abs/2601.08692 · **PDF:** https://arxiv.org/pdf/2601.08692v2
+
+## 22017. All Required, In Order: Phase-Level Evaluation for AI-Human Dialogue in Healthcare and Beyond
+
+- **ID:** `2601.08690` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Shubham Kulkarni, Alexander Lyzhov, Shiva Chaitanya, et al.
+- **abs:** https://arxiv.org/abs/2601.08690 · **PDF:** https://arxiv.org/pdf/2601.08690v1
+
+## 22018. QuantEval: A Benchmark for Financial Quantitative Tasks in Large Language Models
+
+- **ID:** `2601.08689` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Zhaolu Kang, Junhao Gong, Wenqing Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.08689 · **PDF:** https://arxiv.org/pdf/2601.08689v2
+
+## 22019. MEMEWEAVER: Inter-Meme Graph Reasoning for Sexism and Misogyny Detection
+
+- **ID:** `2601.08684` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Paolo Italiani, David Gimeno-Gomez, Luca Ragazzi, et al.
+- **abs:** https://arxiv.org/abs/2601.08684 · **PDF:** https://arxiv.org/pdf/2601.08684v1
+
+## 22020. Region of interest detection for efficient aortic segmentation
+
+- **ID:** `2601.08683` · **primary:** `eess.IV` · **Submitted:** 2026-01-13
+- **Authors:** Loris Giordano, Ine Dirks, Tom Lenaerts, et al.
+- **abs:** https://arxiv.org/abs/2601.08683 · **PDF:** https://arxiv.org/pdf/2601.08683v1
+
+## 22021. Lessons from the Field: An Adaptable Lifecycle Approach to Applied Dialogue Summarization
+
+- **ID:** `2601.08682` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Kushal Chawla, Chenyang Zhu, Pengshan Cai, et al.
+- **abs:** https://arxiv.org/abs/2601.08682 · **PDF:** https://arxiv.org/pdf/2601.08682v1
+
+## 22022. PersonaDual: Balancing Personalization and Objectivity via Adaptive Reasoning
+
+- **ID:** `2601.08679` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Xiaoyou Liu, Xinyi Mou, Shengbin Yue, et al.
+- **abs:** https://arxiv.org/abs/2601.08679 · **PDF:** https://arxiv.org/pdf/2601.08679v2
+
+## 22023. Advancing ESG Intelligence: An Expert-level Agent and Comprehensive Benchmark for Sustainable Finance
+
+- **ID:** `2601.08676` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Yilei Zhao, Wentao Zhang, Lei Xiao, et al.
+- **abs:** https://arxiv.org/abs/2601.08676 · **PDF:** https://arxiv.org/pdf/2601.08676v2
+
+## 22024. Além do Desempenho: Um Estudo da Confiabilidade de Detectores de Deepfakes
+
+- **ID:** `2601.08674` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Lucas Lopes, Rayson Laroca, André Grégio
+- **abs:** https://arxiv.org/abs/2601.08674 · **PDF:** https://arxiv.org/pdf/2601.08674v1
+
+## 22025. Why AI Alignment Failure Is Structural: Learned Human Interaction Structures and AGI as an Endogenous Evolutionary Shock
+
+- **ID:** `2601.08673` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Didier Sornette, Sandro Claudio Lera, Ke Wu
+- **abs:** https://arxiv.org/abs/2601.08673 · **PDF:** https://arxiv.org/pdf/2601.08673v1
+
+## 22026. Parallel Context-of-Experts Decoding for Retrieval Augmented Generation
+
+- **ID:** `2601.08670` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Giulio Corallo, Paolo Papotti
+- **abs:** https://arxiv.org/abs/2601.08670 · **PDF:** https://arxiv.org/pdf/2601.08670v1
+
+## 22027. Analyzing Bias in False Refusal Behavior of Large Language Models for Hate Speech Detoxification
+
+- **ID:** `2601.08668` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Kyuri Im, Shuzhou Yuan, Michael Färber
+- **abs:** https://arxiv.org/abs/2601.08668 · **PDF:** https://arxiv.org/pdf/2601.08668v1
+
+## 22028. From Classical to Quantum Reinforcement Learning and Its Applications in Quantum Control: A Beginner's Tutorial
+
+- **ID:** `2601.08662` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Abhijit Sen, Sonali Panda, Mahima Arya, et al.
+- **abs:** https://arxiv.org/abs/2601.08662 · **PDF:** https://arxiv.org/pdf/2601.08662v2
+
+## 22029. TRACE: Reconstruction-Based Anomaly Detection in Ensemble and Time-Dependent Simulations
+
+- **ID:** `2601.08659` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Hamid Gadirov, Martijn Westra, Steffen Frey
+- **abs:** https://arxiv.org/abs/2601.08659 · **PDF:** https://arxiv.org/pdf/2601.08659v1
+
+## 22030. NEVO-GSPT: Population-Based Neural Network Evolution Using Inflate and Deflate Operators
+
+- **ID:** `2601.08657` · **primary:** `cs.NE` · **Submitted:** 2026-01-13
+- **Authors:** Davide Farinati, Frederico J. J. B. Santos, Leonardo Vanneschi, et al.
+- **abs:** https://arxiv.org/abs/2601.08657 · **PDF:** https://arxiv.org/pdf/2601.08657v1
+
+## 22031. RULERS: Locked Rubrics and Evidence-Anchored Scoring for Robust LLM Evaluation
+
+- **ID:** `2601.08654` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Yihan Hong, Huaiyuan Yao, Bolin Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.08654 · **PDF:** https://arxiv.org/pdf/2601.08654v1
+
+## 22032. Prism: Towards Lowering User Cognitive Load in LLMs via Complex Intent Understanding
+
+- **ID:** `2601.08653` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Zenghua Liao, Jinzhi Liao, Xiang Zhao
+- **abs:** https://arxiv.org/abs/2601.08653 · **PDF:** https://arxiv.org/pdf/2601.08653v2
+
+## 22033. Safe Language Generation in the Limit
+
+- **ID:** `2601.08648` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Antonios Anastasopoulos, Giuseppe Ateniese, Evgenios M. Kornaropoulos
+- **abs:** https://arxiv.org/abs/2601.08648 · **PDF:** https://arxiv.org/pdf/2601.08648v1
+
+## 22034. Provably Safe Reinforcement Learning for Stochastic Reach-Avoid Problems with Entropy Regularization
+
+- **ID:** `2601.08646` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Abhijit Mazumdar, Rafal Wisniewski, Manuela L. Bujorianu
+- **abs:** https://arxiv.org/abs/2601.08646 · **PDF:** https://arxiv.org/pdf/2601.08646v3
+
+## 22035. XGBoost Forecasting of NEPSE Index Log Returns with Walk Forward Validation
+
+- **ID:** `2601.08896` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Sahaj Raj Malla, Shreeyash Kayastha, Rumi Suwal, et al.
+- **abs:** https://arxiv.org/abs/2601.08896 · **PDF:** https://arxiv.org/pdf/2601.08896v1
+
+## 22036. When Generative AI Meets Extended Reality: Enabling Scalable and Natural Interactions
+
+- **ID:** `2601.15308` · **primary:** `cs.HC` · **Submitted:** 2026-01-13
+- **Authors:** Mingyu Zhu, Jiangong Chen, Bin Li
+- **abs:** https://arxiv.org/abs/2601.15308 · **PDF:** https://arxiv.org/pdf/2601.15308v1
+
+## 22037. A Parallel Cross-Lingual Benchmark for Multimodal Idiomaticity Understanding
+
+- **ID:** `2601.08645` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Dilara Torunoğlu-Selamet, Dogukan Arslan, Rodrigo Wilkens, et al.
+- **abs:** https://arxiv.org/abs/2601.08645 · **PDF:** https://arxiv.org/pdf/2601.08645v2
+
+## 22038. Automatic debiased machine learning and sensitivity analysis for sample selection models
+
+- **ID:** `2601.08643` · **primary:** `econ.EM` · **Submitted:** 2026-01-13
+- **Authors:** Jakob Bjelac, Victor Chernozhukov, Phil-Adrian Klotz, et al.
+- **abs:** https://arxiv.org/abs/2601.08643 · **PDF:** https://arxiv.org/pdf/2601.08643v1
+
+## 22039. Resisting Manipulative Bots in Meme Coin Copy Trading: A Multi-Agent Approach with Chain-of-Thought Reasoning
+
+- **ID:** `2601.08641` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Yichen Luo, Yebo Feng, Jiahua Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.08641 · **PDF:** https://arxiv.org/pdf/2601.08641v3
+
+## 22040. Moral Lenses, Political Coordinates: Towards Ideological Positioning of Morally Conditioned LLMs
+
+- **ID:** `2601.08634` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Chenchen Yuan, Bolei Ma, Zheyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.08634 · **PDF:** https://arxiv.org/pdf/2601.08634v1
+
+## 22041. M$^2$FMoE: Multi-Resolution Multi-View Frequency Mixture-of-Experts for Extreme-Adaptive Time Series Forecasting
+
+- **ID:** `2601.08631` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Yaohui Huang, Runmin Zou, Yun Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.08631 · **PDF:** https://arxiv.org/pdf/2601.08631v1
+
+## 22042. Get away with less: Need of source side data curation to build parallel corpus for low resource Machine Translation
+
+- **ID:** `2601.08629` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Saumitra Yadav, Manish Shrivastava
+- **abs:** https://arxiv.org/abs/2601.08629 · **PDF:** https://arxiv.org/pdf/2601.08629v2
+
+## 22043. How Order-Sensitive Are LLMs? OrderProbe for Deterministic Structural Reconstruction
+
+- **ID:** `2601.08626` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Yingjie He, Zhaolu Kang, Kehan Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.08626 · **PDF:** https://arxiv.org/pdf/2601.08626v2
+
+## 22044. SafeRedir: Prompt Embedding Redirection for Robust Unlearning in Image Generation Models
+
+- **ID:** `2601.08623` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Renyang Liu, Kangjie Chen, Han Qiu, et al.
+- **abs:** https://arxiv.org/abs/2601.08623 · **PDF:** https://arxiv.org/pdf/2601.08623v1
+
+## 22045. GraphSearch: Agentic Search-Augmented Reasoning for Zero-Shot Graph Learning
+
+- **ID:** `2601.08621` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Jiajin Liu, Yuanfu Sun, Dongzhe Fan, et al.
+- **abs:** https://arxiv.org/abs/2601.08621 · **PDF:** https://arxiv.org/pdf/2601.08621v1
+
+## 22046. ViDoRe V3: A Comprehensive Evaluation of Retrieval Augmented Generation in Complex Real-World Scenarios
+
+- **ID:** `2601.08620` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** António Loison, Quentin Macé, Antoine Edy, et al.
+- **abs:** https://arxiv.org/abs/2601.08620 · **PDF:** https://arxiv.org/pdf/2601.08620v1
+
+## 22047. Robust low-rank estimation with multiple binary responses using pairwise AUC loss
+
+- **ID:** `2601.08618` · **primary:** `stat.ML` · **Submitted:** 2026-01-13
+- **Authors:** The Tien Mai
+- **abs:** https://arxiv.org/abs/2601.08618 · **PDF:** https://arxiv.org/pdf/2601.08618v1
+
+## 22048. Accelerated Methods with Complexity Separation Under Data Similarity for Federated Learning Problems
+
+- **ID:** `2601.08614` · **primary:** `math.OC` · **Submitted:** 2026-01-13
+- **Authors:** Dmitry Bylinkin, Sergey Skorik, Dmitriy Bystrov, et al.
+- **abs:** https://arxiv.org/abs/2601.08614 · **PDF:** https://arxiv.org/pdf/2601.08614v1
+
+## 22049. VeriTaS: The First Dynamic Benchmark for Multimodal Automated Fact-Checking
+
+- **ID:** `2601.08611` · **primary:** `cs.IR` · **Submitted:** 2026-01-13
+- **Authors:** Mark Rothermel, Marcus Kornmann, Marcus Rohrbach, et al.
+- **abs:** https://arxiv.org/abs/2601.08611 · **PDF:** https://arxiv.org/pdf/2601.08611v1
+
+## 22050. ExpSeek: Self-Triggered Experience Seeking for Web Agents
+
+- **ID:** `2601.08605` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Wenyuan Zhang, Xinghua Zhang, Haiyang Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.08605 · **PDF:** https://arxiv.org/pdf/2601.08605v1
+
+## 22051. Interpretability and Individuality in Knee MRI: Patient-Specific Radiomic Fingerprint with Reconstructed Healthy Personas
+
+- **ID:** `2601.08604` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Yaxi Chen, Simin Ni, Shuai Li, et al.
+- **abs:** https://arxiv.org/abs/2601.08604 · **PDF:** https://arxiv.org/pdf/2601.08604v1
+
+## 22052. WaveFormer: Frequency-Time Decoupled Vision Modeling with Wave Equation
+
+- **ID:** `2601.08602` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Zishan Shu, Juntong Wu, Wei Yan, et al.
+- **abs:** https://arxiv.org/abs/2601.08602 · **PDF:** https://arxiv.org/pdf/2601.08602v1
+
+## 22053. DeepSurvey-Bench: Evaluating Academic Value of Automatically Generated Scientific Survey
+
+- **ID:** `2601.15307` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Guo-Biao Zhang, Ding-Yuan Liu, Da-Yi Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.15307 · **PDF:** https://arxiv.org/pdf/2601.15307v1
+
+## 22054. A Sharp Universality Dichotomy for the Free Energy of Spherical Spin Glasses
+
+- **ID:** `2601.08599` · **primary:** `math.PR` · **Submitted:** 2026-01-13
+- **Authors:** Taegyun Kim
+- **abs:** https://arxiv.org/abs/2601.08599 · **PDF:** https://arxiv.org/pdf/2601.08599v1
+
+## 22055. Sample Complexity of Composite Quantum Hypothesis Testing
+
+- **ID:** `2601.08588` · **primary:** `quant-ph` · **Submitted:** 2026-01-13
+- **Authors:** Jacob Paul Simpson, Efstratios Palias, Sharu Theresa Jose
+- **abs:** https://arxiv.org/abs/2601.08588 · **PDF:** https://arxiv.org/pdf/2601.08588v3
+
+## 22056. Ministral 3
+
+- **ID:** `2601.08584` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Alexander H. Liu, Kartik Khandelwal, Sandeep Subramanian, et al.
+- **abs:** https://arxiv.org/abs/2601.08584 · **PDF:** https://arxiv.org/pdf/2601.08584v1
+
+## 22057. Rewriting Video: Text-Driven Reauthoring of Video Footage
+
+- **ID:** `2601.08565` · **primary:** `cs.HC` · **Submitted:** 2026-01-13
+- **Authors:** Sitong Wang, Anh Truong, Lydia B. Chilton, et al.
+- **abs:** https://arxiv.org/abs/2601.08565 · **PDF:** https://arxiv.org/pdf/2601.08565v1
+
+## 22058. WaterCopilot: An AI-Driven Virtual Assistant for Water Management
+
+- **ID:** `2601.08559` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Keerththanan Vickneswaran, Mariangel Garcia Andarcia, Hugo Retief, et al.
+- **abs:** https://arxiv.org/abs/2601.08559 · **PDF:** https://arxiv.org/pdf/2601.08559v1
+
+## 22059. VideoHEDGE: Entropy-Based Hallucination Detection for Video-VLMs via Semantic Clustering and Spatiotemporal Perturbations
+
+- **ID:** `2601.08557` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Sushant Gautam, Cise Midoglu, Vajira Thambawita, et al.
+- **abs:** https://arxiv.org/abs/2601.08557 · **PDF:** https://arxiv.org/pdf/2601.08557v1
+
+## 22060. EviNAM: Intelligibility and Uncertainty via Evidential Neural Additive Models
+
+- **ID:** `2601.08556` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Sören Schleibaum, Anton Frederik Thielmann, Julian Teusch, et al.
+- **abs:** https://arxiv.org/abs/2601.08556 · **PDF:** https://arxiv.org/pdf/2601.08556v1
+
+## 22061. Comparison of Outlier Detection Algorithms on String Data
+
+- **ID:** `2603.11049` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Philip Maus
+- **abs:** https://arxiv.org/abs/2603.11049 · **PDF:** https://arxiv.org/pdf/2603.11049v1
+
+## 22062. Contrastive and Multi-Task Learning on Noisy Brain Signals with Nonlinear Dynamical Signatures
+
+- **ID:** `2601.08549` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Sucheta Ghosh, Felix Dietrich, Zahra Monfared
+- **abs:** https://arxiv.org/abs/2601.08549 · **PDF:** https://arxiv.org/pdf/2601.08549v3
+
+## 22063. Convergence of gradient flow for learning convolutional neural networks
+
+- **ID:** `2601.08547` · **primary:** `math.OC` · **Submitted:** 2026-01-13
+- **Authors:** Jona-Maria Diederen, Holger Rauhut, Ulrich Terstiege
+- **abs:** https://arxiv.org/abs/2601.08547 · **PDF:** https://arxiv.org/pdf/2601.08547v1
+
+## 22064. Learner-Tailored Program Repair: A Solution Generator with Iterative Edit-Driven Retrieval Enhancement
+
+- **ID:** `2601.08545` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Zhenlong Dai, Zhuoluo Zhao, Hengning Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.08545 · **PDF:** https://arxiv.org/pdf/2601.08545v2
+
+## 22065. Reducing Compute Waste in LLMs through Kernel-Level DVFS
+
+- **ID:** `2601.08539` · **primary:** `cs.PF` · **Submitted:** 2026-01-13
+- **Authors:** Jeffrey Spaan, Kuan-Hsun Chen, Ana-Lucia Varbanescu
+- **abs:** https://arxiv.org/abs/2601.08539 · **PDF:** https://arxiv.org/pdf/2601.08539v1
+
+## 22066. Do Open-Vocabulary Detectors Transfer to Aerial Imagery? A Comparative Evaluation
+
+- **ID:** `2601.22164` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Christos Tsourveloudis
+- **abs:** https://arxiv.org/abs/2601.22164 · **PDF:** https://arxiv.org/pdf/2601.22164v1
+
+## 22067. DeepResearch Bench II: Diagnosing Deep Research Agents via Rubrics from Expert Report
+
+- **ID:** `2601.08536` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Ruizhe Li, Mingxuan Du, Benfeng Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.08536 · **PDF:** https://arxiv.org/pdf/2601.08536v2
+
+## 22068. Sketch-Based Facade Renovation With Generative AI: A Streamlined Framework for Bypassing As-Built Modelling in Industrial Adaptive Reuse
+
+- **ID:** `2601.08531` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Warissara Booranamaitree, Xusheng Du, Yushu Cai, et al.
+- **abs:** https://arxiv.org/abs/2601.08531 · **PDF:** https://arxiv.org/pdf/2601.08531v1
+
+## 22069. Sampling via Stochastic Interpolants by Langevin-based Velocity and Initialization Estimation in Flow ODEs
+
+- **ID:** `2601.08527` · **primary:** `math.NA` · **Submitted:** 2026-01-13
+- **Authors:** Chenguang Duan, Yuling Jiao, Gabriele Steidl, et al.
+- **abs:** https://arxiv.org/abs/2601.08527 · **PDF:** https://arxiv.org/pdf/2601.08527v2
+
+## 22070. Supervised Spike Agreement Dependent Plasticity for Fast Local Learning in Spiking Neural Networks
+
+- **ID:** `2601.08526` · **primary:** `cs.NE` · **Submitted:** 2026-01-13
+- **Authors:** Gouri Lakshmi S, Athira Chandrasekharan, Harshit Kumar, et al.
+- **abs:** https://arxiv.org/abs/2601.08526 · **PDF:** https://arxiv.org/pdf/2601.08526v1
+
+## 22071. Your Group-Relative Advantage Is Biased
+
+- **ID:** `2601.08521` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Fengkai Yang, Zherui Chen, Xiaohan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.08521 · **PDF:** https://arxiv.org/pdf/2601.08521v2
+
+## 22072. CD^2: Constrained Dataset Distillation for Few-Shot Class-Incremental Learning
+
+- **ID:** `2601.08519` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Kexin Bao, Daichi Zhang, Hansong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.08519 · **PDF:** https://arxiv.org/pdf/2601.08519v1
+
+## 22073. Algorithmic Stability in Infinite Dimensions: Characterizing Unconditional Convergence in Banach Spaces
+
+- **ID:** `2601.08512` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Przemysław Spyra
+- **abs:** https://arxiv.org/abs/2601.08512 · **PDF:** https://arxiv.org/pdf/2601.08512v1
+
+## 22074. STAR: Detecting Inference-time Backdoors in LLM Reasoning via State-Transition Amplification Ratio
+
+- **ID:** `2601.08511` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Seong-Gyu Park, Sohee Park, Jisu Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.08511 · **PDF:** https://arxiv.org/pdf/2601.08511v1
+
+## 22075. STAGE: A Benchmark for Knowledge Graph Construction, Question Answering, and In-Script Role-Playing over Movie Screenplays
+
+- **ID:** `2601.08510` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Qiuyu Tian, Yiding Li, Fengyi Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.08510 · **PDF:** https://arxiv.org/pdf/2601.08510v2
+
+## 22076. Spectral Generative Flow Models: A Physics-Inspired Replacement for Vectorized Large Language Models
+
+- **ID:** `2601.08893` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Andrew Kiruluta
+- **abs:** https://arxiv.org/abs/2601.08893 · **PDF:** https://arxiv.org/pdf/2601.08893v2
+
+## 22077. What If TSF: A Benchmark for Reframing Forecasting as Scenario-Guided Multimodal Forecasting
+
+- **ID:** `2601.08509` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Jinkwan Jang, Hyunbin Jin, Hyungjin Park, et al.
+- **abs:** https://arxiv.org/abs/2601.08509 · **PDF:** https://arxiv.org/pdf/2601.08509v1
+
+## 22078. Temporal Fusion Nexus: A task-agnostic multi-modal embedding model for clinical narratives and irregular time series in post-kidney transplant care
+
+- **ID:** `2601.08503` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Aditya Kumar, Simon Rauch, Mario Cypko, et al.
+- **abs:** https://arxiv.org/abs/2601.08503 · **PDF:** https://arxiv.org/pdf/2601.08503v1
+
+## 22079. It's All About the Confidence: An Unsupervised Approach for Multilingual Historical Entity Linking using Large Language Models
+
+- **ID:** `2601.08500` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Cristian Santini, Marieke Van Erp, Mehwish Alam
+- **abs:** https://arxiv.org/abs/2601.08500 · **PDF:** https://arxiv.org/pdf/2601.08500v1
+
+## 22080. EfficientFSL: Enhancing Few-Shot Classification via Query-Only Tuning in Vision Transformers
+
+- **ID:** `2601.08499` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Wenwen Liao, Hang Ruan, Jianbo Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.08499 · **PDF:** https://arxiv.org/pdf/2601.08499v2
+
+## 22081. GraphFusionSBR: Denoising Multi-Channel Graphs for Session-Based Recommendation
+
+- **ID:** `2601.08497` · **primary:** `cs.IR` · **Submitted:** 2026-01-13
+- **Authors:** Jia-Xin He, Hung-Hsuan Chen
+- **abs:** https://arxiv.org/abs/2601.08497 · **PDF:** https://arxiv.org/pdf/2601.08497v1
+
+## 22082. PKI: Prior Knowledge-Infused Neural Network for Few-Shot Class-Incremental Learning
+
+- **ID:** `2601.08493` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Kexin Baoa, Fanzhao Lin, Zichen Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.08493 · **PDF:** https://arxiv.org/pdf/2601.08493v1
+
+## 22083. Evaluating Role-Consistency in LLMs for Counselor Training
+
+- **ID:** `2601.08892` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Eric Rudolph, Natalie Engert, Jens Albrecht
+- **abs:** https://arxiv.org/abs/2601.08892 · **PDF:** https://arxiv.org/pdf/2601.08892v1
+
+## 22084. AUV Trajectory Learning for Underwater Acoustic Energy Transfer and Age Minimization
+
+- **ID:** `2601.08491` · **primary:** `cs.RO` · **Submitted:** 2026-01-13
+- **Authors:** Mohamed Afouene Melki, Mohammad Shehab, Mohamed-Slim Alouini
+- **abs:** https://arxiv.org/abs/2601.08491 · **PDF:** https://arxiv.org/pdf/2601.08491v1
+
+## 22085. BenchOverflow: Measuring Overflow in Large Language Models via Plain-Text Prompts
+
+- **ID:** `2601.08490` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Erin Feiglin, Nir Hutnik, Raz Lapid
+- **abs:** https://arxiv.org/abs/2601.08490 · **PDF:** https://arxiv.org/pdf/2601.08490v1
+
+## 22086. Surgical Refusal Ablation: Disentangling Safety from Intelligence via Concept-Guided Spectral Cleaning
+
+- **ID:** `2601.08489` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Tony Cristofano
+- **abs:** https://arxiv.org/abs/2601.08489 · **PDF:** https://arxiv.org/pdf/2601.08489v1
+
+## 22087. DiffMM: Efficient Method for Accurate Noisy and Sparse Trajectory Map Matching via One Step Diffusion
+
+- **ID:** `2601.08482` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Chenxu Han, Sean Bin Yang, Jilin Hu
+- **abs:** https://arxiv.org/abs/2601.08482 · **PDF:** https://arxiv.org/pdf/2601.08482v1
+
+## 22088. Do You Understand How I Feel?: Towards Verified Empathy in Therapy Chatbots
+
+- **ID:** `2601.08477` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Francesco Dettori, Matteo Forasassi, Lorenzo Veronese, et al.
+- **abs:** https://arxiv.org/abs/2601.08477 · **PDF:** https://arxiv.org/pdf/2601.08477v1
+
+## 22089. SUMMPILOT: Bridging Efficiency and Customization for Interactive Summarization System
+
+- **ID:** `2601.08475` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** JungMin Yun, Juhwan Choi, Kyohoon Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.08475 · **PDF:** https://arxiv.org/pdf/2601.08475v1
+
+## 22090. sui-1: Grounded and Verifiable Long-Form Summarization
+
+- **ID:** `2601.08472` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Benedikt Droste, Jan Philipp Harries, Maximilian Idahl, et al.
+- **abs:** https://arxiv.org/abs/2601.08472 · **PDF:** https://arxiv.org/pdf/2601.08472v1
+
+## 22091. Heterogeneous computing platform for real-time robotics
+
+- **ID:** `2601.09755` · **primary:** `cs.NE` · **Submitted:** 2026-01-13
+- **Authors:** Jakub Fil, Yulia Sandamirskaya, Hector Gonzalez, et al.
+- **abs:** https://arxiv.org/abs/2601.09755 · **PDF:** https://arxiv.org/pdf/2601.09755v1
+
+## 22092. JudgeRLVR: Judge First, Generate Second for Efficient Reasoning
+
+- **ID:** `2601.08468` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Jiangshan Duo, Hanyu Li, Hailin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.08468 · **PDF:** https://arxiv.org/pdf/2601.08468v1
+
+## 22093. Zero-Shot Distracted Driver Detection via Vision Language Models with Double Decoupling
+
+- **ID:** `2601.08467` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Takamichi Miyata, Sumiko Miyata, Andrew Morris
+- **abs:** https://arxiv.org/abs/2601.08467 · **PDF:** https://arxiv.org/pdf/2601.08467v1
+
+## 22094. CoMa: Contextual Massing Generation with Vision-Language Models
+
+- **ID:** `2601.08464` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Evgenii Maslov, Valentin Khrulkov, Anastasia Volkova, et al.
+- **abs:** https://arxiv.org/abs/2601.08464 · **PDF:** https://arxiv.org/pdf/2601.08464v1
+
+## 22095. M3-BENCH: Process-Aware Evaluation of LLM Agents' Social Behaviors in Mixed-Motive Games
+
+- **ID:** `2601.08462` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Sixiong Xie, Zhuofan Shi, Haiyang Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.08462 · **PDF:** https://arxiv.org/pdf/2601.08462v2
+
+## 22096. An Under-Explored Application for Explainable Multimodal Misogyny Detection in code-mixed Hindi-English
+
+- **ID:** `2601.08457` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Sargam Yadav, Abhishek Kaushik, Kevin Mc Daid
+- **abs:** https://arxiv.org/abs/2601.08457 · **PDF:** https://arxiv.org/pdf/2601.08457v1
+
+## 22097. Decoding Order Matters in Autoregressive Speech Synthesis
+
+- **ID:** `2601.08450` · **primary:** `cs.SD` · **Submitted:** 2026-01-13
+- **Authors:** Minghui Zhao, Anton Ragni
+- **abs:** https://arxiv.org/abs/2601.08450 · **PDF:** https://arxiv.org/pdf/2601.08450v1
+
+## 22098. Attention Consistency Regularization for Interpretable Early-Exit Neural Networks
+
+- **ID:** `2601.08891` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Yanhua Zhao
+- **abs:** https://arxiv.org/abs/2601.08891 · **PDF:** https://arxiv.org/pdf/2601.08891v2
+
+## 22099. Divide and Conquer: Static-Dynamic Collaboration for Few-Shot Class-Incremental Learning
+
+- **ID:** `2601.08448` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Kexin Bao, Daichi Zhang, Yong Li, et al.
+- **abs:** https://arxiv.org/abs/2601.08448 · **PDF:** https://arxiv.org/pdf/2601.08448v1
+
+## 22100. Sleep-Based Homeostatic Regularization for Stabilizing Spike-Timing-Dependent Plasticity in Recurrent Spiking Neural Networks
+
+- **ID:** `2601.08447` · **primary:** `cs.NE` · **Submitted:** 2026-01-13
+- **Authors:** Andreas Massey, Aliaksandr Hubin, Stefano Nichele, et al.
+- **abs:** https://arxiv.org/abs/2601.08447 · **PDF:** https://arxiv.org/pdf/2601.08447v1
+
+## 22101. Noise-Adaptive Regularization for Robust Multi-Label Remote Sensing Image Classification
+
+- **ID:** `2601.08446` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Tom Burgert, Julia Henkel, Begüm Demir
+- **abs:** https://arxiv.org/abs/2601.08446 · **PDF:** https://arxiv.org/pdf/2601.08446v1
+
+## 22102. Beyond Linearization: Attributed Table Graphs for Table Reasoning
+
+- **ID:** `2601.08444` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Yuxiang Wang, Junhao Gan, Shengxiang Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.08444 · **PDF:** https://arxiv.org/pdf/2601.08444v1
+
+## 22103. YaPO: Learnable Sparse Activation Steering Vectors for Domain Adaptation
+
+- **ID:** `2601.08441` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Abdelaziz Bounhar, Rania Hossam Elmohamady Elbadry, Hadi Abdine, et al.
+- **abs:** https://arxiv.org/abs/2601.08441 · **PDF:** https://arxiv.org/pdf/2601.08441v1
+
+## 22104. Fine-Mem: Fine-Grained Feedback Alignment for Long-Horizon Memory Management
+
+- **ID:** `2601.08435` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Weitao Ma, Xiaocheng Feng, Lei Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.08435 · **PDF:** https://arxiv.org/pdf/2601.08435v1
+
+## 22105. Large Multimodal Models for Embodied Intelligent Driving: The Next Frontier in Self-Driving?
+
+- **ID:** `2601.08434` · **primary:** `cs.RO` · **Submitted:** 2026-01-13
+- **Authors:** Long Zhang, Yuchen Xia, Bingqing Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.08434 · **PDF:** https://arxiv.org/pdf/2601.08434v3
+
+## 22106. RubricHub: A Comprehensive and Highly Discriminative Rubric Dataset via Automated Coarse-to-Fine Generation
+
+- **ID:** `2601.08430` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Sunzhu Li, Jiale Zhao, Miteto Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.08430 · **PDF:** https://arxiv.org/pdf/2601.08430v2
+
+## 22107. Silence the Judge: Reinforcement Learning with Self-Verifier via Latent Geometric Clustering
+
+- **ID:** `2601.08427` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Nonghai Zhang, Weitao Ma, Zhanyu Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.08427 · **PDF:** https://arxiv.org/pdf/2601.08427v2
+
+## 22108. Agentic Digital Twins: A Taxonomy of Capabilities for Understanding Possible Futures
+
+- **ID:** `2601.18799` · **primary:** `cs.CY` · **Submitted:** 2026-01-13
+- **Authors:** Christopher Burr, Mark Enzer, Jason Shepherd, et al.
+- **abs:** https://arxiv.org/abs/2601.18799 · **PDF:** https://arxiv.org/pdf/2601.18799v1
+
+## 22109. Coverage Improvement and Fast Convergence of On-policy Preference Learning
+
+- **ID:** `2601.08421` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Juno Kim, Jihun Yun, Jason D. Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.08421 · **PDF:** https://arxiv.org/pdf/2601.08421v1
+
+## 22110. Taxon: Hierarchical Tax Code Prediction with Semantically Aligned LLM Expert Guidance
+
+- **ID:** `2601.08418` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Jihang Li, Qing Liu, Zulong Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.08418 · **PDF:** https://arxiv.org/pdf/2601.08418v1
+
+## 22111. Regulatory gray areas of LLM Terms
+
+- **ID:** `2601.08415` · **primary:** `cs.CY` · **Submitted:** 2026-01-13
+- **Authors:** Brittany I. Davidson, Kate Muir, Florian A. D. Burnat, et al.
+- **abs:** https://arxiv.org/abs/2601.08415 · **PDF:** https://arxiv.org/pdf/2601.08415v2
+
+## 22112. Hybrid Distillation with CoT Guidance for Edge-Drone Control Code Generation
+
+- **ID:** `2601.08412` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Yizhan Feng, Hichem Snoussi, Yuhang Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.08412 · **PDF:** https://arxiv.org/pdf/2601.08412v1
+
+## 22113. WebTrap Park: An Automated Platform for Systematic Security Evaluation of Web Agents
+
+- **ID:** `2601.08406` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Xinyi Wu, Jiagui Chen, Geng Hong, et al.
+- **abs:** https://arxiv.org/abs/2601.08406 · **PDF:** https://arxiv.org/pdf/2601.08406v1
+
+## 22114. Out-of-distribution generalization of deep-learning surrogates for 2D PDE-generated dynamics in the small-data regime
+
+- **ID:** `2601.08404` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Binh Duong Nguyen, Stefan Sandfeld
+- **abs:** https://arxiv.org/abs/2601.08404 · **PDF:** https://arxiv.org/pdf/2601.08404v1
+
+## 22115. Owen-Shapley Policy Optimization (OSPO): A Principled RL Algorithm for Generative Search LLMs
+
+- **ID:** `2601.08403` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Abhijnan Nath, Alireza Bagheri Garakani, Tianchen Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.08403 · **PDF:** https://arxiv.org/pdf/2601.08403v1
+
+## 22116. PATS: Personality-Aware Teaching Strategies with Large Language Model Tutors
+
+- **ID:** `2601.08402` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Donya Rooein, Sankalan Pal Chowdhury, Mariia Eremeeva, et al.
+- **abs:** https://arxiv.org/abs/2601.08402 · **PDF:** https://arxiv.org/pdf/2601.08402v1
+
+## 22117. An Explainable Two Stage Deep Learning Framework for Pericoronitis Assessment in Panoramic Radiographs Using YOLOv8 and ResNet-50
+
+- **ID:** `2601.08401` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Ajo Babu George, Pranav S, Kunal Agarwal
+- **abs:** https://arxiv.org/abs/2601.08401 · **PDF:** https://arxiv.org/pdf/2601.08401v1
+
+## 22118. Controlled LLM Training on Spectral Sphere
+
+- **ID:** `2601.08393` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Tian Xie, Haoming Luo, Haoyu Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.08393 · **PDF:** https://arxiv.org/pdf/2601.08393v3
+
+## 22119. Creativity in AI as Emergence from Domain-Limited Generative Models
+
+- **ID:** `2601.08388` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Corina Chutaux
+- **abs:** https://arxiv.org/abs/2601.08388 · **PDF:** https://arxiv.org/pdf/2601.08388v1
+
+## 22120. Deconstructing Pre-training: Knowledge Attribution Analysis in MoE and Dense Models
+
+- **ID:** `2601.08383` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Bo Wang, Junzhuo Li, Hong Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.08383 · **PDF:** https://arxiv.org/pdf/2601.08383v1
+
+## 22121. A Qualitative Model to Reason about Object Rotations (QOR) applied to solve the Cube Comparison Test (CCT)
+
+- **ID:** `2601.08382` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Zoe Falomir
+- **abs:** https://arxiv.org/abs/2601.08382 · **PDF:** https://arxiv.org/pdf/2601.08382v2
+
+## 22122. Thematic Working Group 5 -- Artificial Intelligence (AI) literacy for teaching and learning: design and implementation
+
+- **ID:** `2601.08380` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Mary Webb, Matt Bower, Ana Amélia Carvalho, et al.
+- **abs:** https://arxiv.org/abs/2601.08380 · **PDF:** https://arxiv.org/pdf/2601.08380v1
+
+## 22123. Training-Free Distribution Adaptation for Diffusion Models via Maximum Mean Discrepancy Guidance
+
+- **ID:** `2601.08379` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Matina Mahdizadeh Sani, Nima Jamali, Mohammad Jalali, et al.
+- **abs:** https://arxiv.org/abs/2601.08379 · **PDF:** https://arxiv.org/pdf/2601.08379v1
+
+## 22124. Geo-NVS-w: Geometry-Aware Novel View Synthesis In-the-Wild with an SDF Renderer
+
+- **ID:** `2601.08371` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Anastasios Tsalakopoulos, Angelos Kanlis, Evangelos Chatzis, et al.
+- **abs:** https://arxiv.org/abs/2601.08371 · **PDF:** https://arxiv.org/pdf/2601.08371v1
+
+## 22125. PosIR: Position-Aware Heterogeneous Information Retrieval Benchmark
+
+- **ID:** `2601.08363` · **primary:** `cs.IR` · **Submitted:** 2026-01-13
+- **Authors:** Ziyang Zeng, Dun Zhang, Yu Yan, et al.
+- **abs:** https://arxiv.org/abs/2601.08363 · **PDF:** https://arxiv.org/pdf/2601.08363v2
+
+## 22126. Scalable Sequential Recommendation under Latency and Memory Constraints
+
+- **ID:** `2601.08360` · **primary:** `cs.IR` · **Submitted:** 2026-01-13
+- **Authors:** Adithya Parthasarathy, Aswathnarayan Muthukrishnan Kirubakaran, Vinoth Punniyamoorthy, et al.
+- **abs:** https://arxiv.org/abs/2601.08360 · **PDF:** https://arxiv.org/pdf/2601.08360v2
+
+## 22127. Decodable but not structured: linear probing enables Underwater Acoustic Target Recognition with pretrained audio embeddings
+
+- **ID:** `2601.08358` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Hilde I. Hummel, Sandjai Bhulai, Rob D. van der Mei, et al.
+- **abs:** https://arxiv.org/abs/2601.08358 · **PDF:** https://arxiv.org/pdf/2601.08358v1
+
+## 22128. MLPlatt: Simple Calibration Framework for Ranking Models
+
+- **ID:** `2601.08345` · **primary:** `cs.IR` · **Submitted:** 2026-01-13
+- **Authors:** Piotr Bajger, Roman Dusek, Krzysztof Galias, et al.
+- **abs:** https://arxiv.org/abs/2601.08345 · **PDF:** https://arxiv.org/pdf/2601.08345v1
+
+## 22129. When KV Cache Reuse Fails in Multi-Agent Systems: Cross-Candidate Interaction is Crucial for LLM Judges
+
+- **ID:** `2601.08343` · **primary:** `cs.MA` · **Submitted:** 2026-01-13
+- **Authors:** Sichu Liang, Zhenglin Wang, Jiajia Chu, et al.
+- **abs:** https://arxiv.org/abs/2601.08343 · **PDF:** https://arxiv.org/pdf/2601.08343v1
+
+## 22130. Detecting Mental Manipulation in Speech via Synthetic Multi-Speaker Dialogue
+
+- **ID:** `2601.08342` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Run Chen, Wen Liang, Ziwei Gong, et al.
+- **abs:** https://arxiv.org/abs/2601.08342 · **PDF:** https://arxiv.org/pdf/2601.08342v1
+
+## 22131. Automated Machine Learning in Radiomics: A Comparative Evaluation of Performance, Efficiency and Accessibility
+
+- **ID:** `2601.08334` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Jose Lozano-Montoya, Emilio Soria-Olivas, Almudena Fuster-Matanzo, et al.
+- **abs:** https://arxiv.org/abs/2601.08334 · **PDF:** https://arxiv.org/pdf/2601.08334v2
+
+## 22132. Semantic Laundering in AI Agent Architectures: Why Tool Boundaries Do Not Confer Epistemic Warrant
+
+- **ID:** `2601.08333` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Oleg Romanchuk, Roman Bondar
+- **abs:** https://arxiv.org/abs/2601.08333 · **PDF:** https://arxiv.org/pdf/2601.08333v1
+
+## 22133. IGAN: A New Inception-based Model for Stable and High-Fidelity Image Synthesis Using Generative Adversarial Networks
+
+- **ID:** `2601.08332` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Ahmed A. Hashim, Ali Al-Shuwaili, Asraa Saeed, et al.
+- **abs:** https://arxiv.org/abs/2601.08332 · **PDF:** https://arxiv.org/pdf/2601.08332v1
+
+## 22134. CLaS-Bench: A Cross-Lingual Alignment and Steering Benchmark
+
+- **ID:** `2601.08331` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Daniil Gurgurov, Yusser Al Ghussin, Tanja Baeumel, et al.
+- **abs:** https://arxiv.org/abs/2601.08331 · **PDF:** https://arxiv.org/pdf/2601.08331v1
+
+## 22135. Safe Heterogeneous Multi-Agent RL with Communication Regularization for Coordinated Target Acquisition
+
+- **ID:** `2601.08327` · **primary:** `cs.RO` · **Submitted:** 2026-01-13
+- **Authors:** Gabriele Calzolari, Vidya Sumathy, Christoforos Kanellakis, et al.
+- **abs:** https://arxiv.org/abs/2601.08327 · **PDF:** https://arxiv.org/pdf/2601.08327v1
+
+## 22136. AtomMem : Learnable Dynamic Agentic Memory with Atomic Memory Operation
+
+- **ID:** `2601.08323` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Yupeng Huo, Yaxi Lu, Zhong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.08323 · **PDF:** https://arxiv.org/pdf/2601.08323v3
+
+## 22137. Disentangling History and Propagation Dependencies in Cross-Subject Knee Contact Stress Prediction Using a Shared MeshGraphNet Backbone
+
+- **ID:** `2601.08318` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-13
+- **Authors:** Zhengye Pan, Jianwei Zuo, Jiajia Luo
+- **abs:** https://arxiv.org/abs/2601.08318 · **PDF:** https://arxiv.org/pdf/2601.08318v1
+
+## 22138. Deep Exploration of Epoch-wise Double Descent in Noisy Data: Signal Separation, Large Activation, and Benign Overfitting
+
+- **ID:** `2601.08316` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Tomoki Kubo, Ryuken Uda, Yusuke Iida
+- **abs:** https://arxiv.org/abs/2601.08316 · **PDF:** https://arxiv.org/pdf/2601.08316v1
+
+## 22139. Enhancing Image Quality Assessment Ability of LMMs via Retrieval-Augmented Generation
+
+- **ID:** `2601.08311` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Kang Fu, Huiyu Duan, Zicheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.08311 · **PDF:** https://arxiv.org/pdf/2601.08311v1
+
+## 22140. ORBIT: On-policy Exploration-Exploitation for Controllable Multi-Budget Reasoning
+
+- **ID:** `2601.08310` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Kun Liang, Clive Bai, Xin Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.08310 · **PDF:** https://arxiv.org/pdf/2601.08310v1
+
+## 22141. AgriAgent: Contract-Driven Planning and Capability-Aware Tool Orchestration in Real-World Agriculture
+
+- **ID:** `2601.08308` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Bo Yang, Yu Zhang, Yunkui Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.08308 · **PDF:** https://arxiv.org/pdf/2601.08308v1
+
+## 22142. Uncovering Latent Bias in LLM-Based Emergency Department Triage Through Proxy Variables
+
+- **ID:** `2601.15306` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Ethan Zhang
+- **abs:** https://arxiv.org/abs/2601.15306 · **PDF:** https://arxiv.org/pdf/2601.15306v1
+
+## 22143. Enhancing Sentiment Classification and Irony Detection in Large Language Models through Advanced Prompt Engineering Techniques
+
+- **ID:** `2601.08302` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Marvin Schmitt, Anne Schwerk, Sebastian Lempert
+- **abs:** https://arxiv.org/abs/2601.08302 · **PDF:** https://arxiv.org/pdf/2601.08302v1
+
+## 22144. Demystifying the Slash Pattern in Attention: The Role of RoPE
+
+- **ID:** `2601.08297` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Yuan Cheng, Fengzhuo Zhang, Yunlong Hou, et al.
+- **abs:** https://arxiv.org/abs/2601.08297 · **PDF:** https://arxiv.org/pdf/2601.08297v2
+
+## 22145. OpenMic: A Multi-Agent-Based Stand-Up Comedy Generation System
+
+- **ID:** `2601.08288` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Yuyang Wu, Hanzhong Cao, Jianhao Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.08288 · **PDF:** https://arxiv.org/pdf/2601.08288v1
+
+## 22146. AgriLens: Semantic Retrieval in Agricultural Texts Using Topic Modeling and Language Models
+
+- **ID:** `2601.08283` · **primary:** `cs.IR` · **Submitted:** 2026-01-13
+- **Authors:** Heba Shakeel, Tanvir Ahmad, Tanya Liyaqat, et al.
+- **abs:** https://arxiv.org/abs/2601.08283 · **PDF:** https://arxiv.org/pdf/2601.08283v1
+
+## 22147. D$^2$Plan: Dual-Agent Dynamic Global Planning for Complex Retrieval-Augmented Reasoning
+
+- **ID:** `2601.08282` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Kangcheng Luo, Tinglang Wu, Yansong Feng
+- **abs:** https://arxiv.org/abs/2601.08282 · **PDF:** https://arxiv.org/pdf/2601.08282v1
+
+## 22148. Greedy Is Enough: Sparse Action Discovery in Agentic LLMs
+
+- **ID:** `2601.08280` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Angshul Majumdar
+- **abs:** https://arxiv.org/abs/2601.08280 · **PDF:** https://arxiv.org/pdf/2601.08280v1
+
+## 22149. One-Shot Identification with Different Neural Network Approaches
+
+- **ID:** `2601.08278` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Janis Mohr, Jörg Frochte
+- **abs:** https://arxiv.org/abs/2601.08278 · **PDF:** https://arxiv.org/pdf/2601.08278v1
+
+## 22150. ToolACE-MCP: Generalizing History-Aware Routing from MCP Tools to the Agent Web
+
+- **ID:** `2601.08276` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Zhiyuan Yao, Zishan Xu, Yifu Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.08276 · **PDF:** https://arxiv.org/pdf/2601.08276v1
+
+## 22151. Discovery and Reinforcement of Tool-Integrated Reasoning Chains via Rollout Trees
+
+- **ID:** `2601.08274` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Kun Li, Zenan Xu, Junan Li, et al.
+- **abs:** https://arxiv.org/abs/2601.08274 · **PDF:** https://arxiv.org/pdf/2601.08274v2
+
+## 22152. HIPPO: Accelerating Video Large Language Models Inference via Holistic-aware Parallel Speculative Decoding
+
+- **ID:** `2601.08273` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Qitan Lv, Tianyu Liu, Wen Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.08273 · **PDF:** https://arxiv.org/pdf/2601.08273v1
+
+## 22153. Sparsity Is Necessary: Polynomial-Time Stability for Agentic LLMs in Large Action Spaces
+
+- **ID:** `2601.08271` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Angshul Majumdar
+- **abs:** https://arxiv.org/abs/2601.08271 · **PDF:** https://arxiv.org/pdf/2601.08271v1
+
+## 22154. Med-CoReasoner: Reducing Language Disparities in Medical Reasoning via Language-Informed Co-Reasoning
+
+- **ID:** `2601.08267` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Fan Gao, Sherry T. Tong, Jiwoong Sohn, et al.
+- **abs:** https://arxiv.org/abs/2601.08267 · **PDF:** https://arxiv.org/pdf/2601.08267v2
+
+## 22155. VGG Induced Deep Hand Sign Language Detection
+
+- **ID:** `2601.08262` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Subham Sharma, Sharmila Subudhi
+- **abs:** https://arxiv.org/abs/2601.08262 · **PDF:** https://arxiv.org/pdf/2601.08262v1
+
+## 22156. A Usable GAN-Based Tool for Synthetic ECG Generation in Cardiac Amyloidosis Research
+
+- **ID:** `2601.08260` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Francesco Speziale, Ugo Lomoio, Fabiola Boccuto, et al.
+- **abs:** https://arxiv.org/abs/2601.08260 · **PDF:** https://arxiv.org/pdf/2601.08260v1
+
+## 22157. T3: Benchmarking Sycophancy and Skepticism in Causal Judgment
+
+- **ID:** `2601.08258` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Edward Y. Chang
+- **abs:** https://arxiv.org/abs/2601.08258 · **PDF:** https://arxiv.org/pdf/2601.08258v2
+
+## 22158. On Evaluation of Unsupervised Feature Selection for Pattern Classification
+
+- **ID:** `2601.08257` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Gyu-Il Kim, Dae-Won Kim, Jaesung Lee
+- **abs:** https://arxiv.org/abs/2601.08257 · **PDF:** https://arxiv.org/pdf/2601.08257v3
+
+## 22159. Large Artificial Intelligence Model Guided Deep Reinforcement Learning for Resource Allocation in Non Terrestrial Networks
+
+- **ID:** `2601.08254` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Abdikarim Mohamed Ibrahim, Rosdiadee Nordin
+- **abs:** https://arxiv.org/abs/2601.08254 · **PDF:** https://arxiv.org/pdf/2601.08254v1
+
+## 22160. Mem2ActBench: A Benchmark for Evaluating Long-Term Memory Utilization in Task-Oriented Autonomous Agents
+
+- **ID:** `2601.19935` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Yiting Shen, Kun Li, Wei Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.19935 · **PDF:** https://arxiv.org/pdf/2601.19935v1
+
+## 22161. LDLT L-Lipschitz Network Weight Parameterization Initialization
+
+- **ID:** `2601.08253` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Marius F. R. Juston, Ramavarapu S. Sreenivas, Dustin Nottage, et al.
+- **abs:** https://arxiv.org/abs/2601.08253 · **PDF:** https://arxiv.org/pdf/2601.08253v1
+
+## 22162. Hyperbolic Heterogeneous Graph Transformer
+
+- **ID:** `2601.08251` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Jongmin Park, Seunghoon Han, Hyewon Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.08251 · **PDF:** https://arxiv.org/pdf/2601.08251v1
+
+## 22163. Incorporating Cognitive Biases into Reinforcement Learning for Financial Decision-Making
+
+- **ID:** `2601.08247` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Liu He
+- **abs:** https://arxiv.org/abs/2601.08247 · **PDF:** https://arxiv.org/pdf/2601.08247v1
+
+## 22164. The End of Reward Engineering: How LLMs Are Redefining Multi-Agent Coordination
+
+- **ID:** `2601.08237` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Haoran Su, Yandong Sun, Congjia Yu
+- **abs:** https://arxiv.org/abs/2601.08237 · **PDF:** https://arxiv.org/pdf/2601.08237v1
+
+## 22165. Structural Dimension Reduction in Bayesian Networks
+
+- **ID:** `2601.08236` · **primary:** `stat.ML` · **Submitted:** 2026-01-13
+- **Authors:** Pei Heng, Yi Sun, Jianhua Guo
+- **abs:** https://arxiv.org/abs/2601.08236 · **PDF:** https://arxiv.org/pdf/2601.08236v1
+
+## 22166. MPCI-Bench: A Benchmark for Multimodal Pairwise Contextual Integrity Evaluation of Language Model Agents
+
+- **ID:** `2601.08235` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Shouju Wang, Haopeng Zhang
+- **abs:** https://arxiv.org/abs/2601.08235 · **PDF:** https://arxiv.org/pdf/2601.08235v3
+
+## 22167. GADPN: Graph Adaptive Denoising and Perturbation Networks via Singular Value Decomposition
+
+- **ID:** `2601.08230` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Hao Deng, Bo Liu
+- **abs:** https://arxiv.org/abs/2601.08230 · **PDF:** https://arxiv.org/pdf/2601.08230v1
+
+## 22168. Knowledge-based learning in Text-RAG and Image-RAG
+
+- **ID:** `2601.08226` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Alexander Shim, Khalil Saieh, Samuel Clarke
+- **abs:** https://arxiv.org/abs/2601.08226 · **PDF:** https://arxiv.org/pdf/2601.08226v1
+
+## 22169. User-Oriented Multi-Turn Dialogue Generation with Tool Use at scale
+
+- **ID:** `2601.08225` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Jungho Cho, Minbyul Jeong, Sungrae Park
+- **abs:** https://arxiv.org/abs/2601.08225 · **PDF:** https://arxiv.org/pdf/2601.08225v1
+
+## 22170. An Axiomatic Approach to General Intelligence: SANC(E3) -- Self-organizing Active Network of Concepts with Energy E3
+
+- **ID:** `2601.08224` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Daesuk Kwon, Won-gi Paeng
+- **abs:** https://arxiv.org/abs/2601.08224 · **PDF:** https://arxiv.org/pdf/2601.08224v1
+
+## 22171. DNF: Dual-Layer Nested Fingerprinting for Large Language Model Intellectual Property Protection
+
+- **ID:** `2601.08223` · **primary:** `cs.CR` · **Submitted:** 2026-01-13
+- **Authors:** Zhenhua Xu, Yiran Zhao, Mengting Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.08223 · **PDF:** https://arxiv.org/pdf/2601.08223v3
+
+## 22172. One-Shot Federated Ridge Regression: Exact Recovery via Sufficient Statistic Aggregation
+
+- **ID:** `2601.08216` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Zahir Alsulaimawi
+- **abs:** https://arxiv.org/abs/2601.08216 · **PDF:** https://arxiv.org/pdf/2601.08216v1
+
+## 22173. Towards Principled Design of Mixture-of-Experts Language Models under Memory and Inference Constraints
+
+- **ID:** `2601.08215` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Seng Pei Liew, Kenta Shinzato, Yuyang Dong
+- **abs:** https://arxiv.org/abs/2601.08215 · **PDF:** https://arxiv.org/pdf/2601.08215v1
+
+## 22174. Adapting Rules of Official International Mahjong for Online Players
+
+- **ID:** `2601.08211` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Chucai Wang, Lingfeng Li, Yunlong Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.08211 · **PDF:** https://arxiv.org/pdf/2601.08211v1
+
+## 22175. Adaptive few-shot learning for robust part quality classification in two-photon lithography
+
+- **ID:** `2601.08885` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Sixian Jia, Ruo-Syuan Mei, Chenhui Shao
+- **abs:** https://arxiv.org/abs/2601.08885 · **PDF:** https://arxiv.org/pdf/2601.08885v1
+
+## 22176. Scalable Multiagent Reinforcement Learning with Collective Influence Estimation
+
+- **ID:** `2601.08210` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Zhenglong Luo, Zhiyong Chen, Aoxiang Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.08210 · **PDF:** https://arxiv.org/pdf/2601.08210v1
+
+## 22177. Generation-Augmented Generation: A Plug-and-Play Framework for Private Knowledge Injection in Large Language Models
+
+- **ID:** `2601.08209` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Rongji Li, Jian Xu, Xueqing Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.08209 · **PDF:** https://arxiv.org/pdf/2601.08209v2
+
+## 22178. FUME: Fused Unified Multi-Gas Emission Network for Livestock Rumen Acidosis Detection
+
+- **ID:** `2601.08205` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Taminul Islam, Toqi Tahamid Sarker, Mohamed Embaby, et al.
+- **abs:** https://arxiv.org/abs/2601.08205 · **PDF:** https://arxiv.org/pdf/2601.08205v1
+
+## 22179. Triplets Better Than Pairs: Towards Stable and Effective Self-Play Fine-Tuning for LLMs
+
+- **ID:** `2601.08198` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Yibo Wang, Hai-Long Sun, Qing-Guo Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.08198 · **PDF:** https://arxiv.org/pdf/2601.08198v1
+
+## 22180. Hallucination-Free Automatic Question & Answer Generation for Intuitive Learning
+
+- **ID:** `2601.14280` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Nicholas X. Wang, Aggelos K. Katsaggelos
+- **abs:** https://arxiv.org/abs/2601.14280 · **PDF:** https://arxiv.org/pdf/2601.14280v1
+
+## 22181. Evaluating Implicit Regulatory Compliance in LLM Tool Invocation via Logic-Guided Synthesis
+
+- **ID:** `2601.08196` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Da Song, Yuheng Huang, Boqi Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.08196 · **PDF:** https://arxiv.org/pdf/2601.08196v1
+
+## 22182. On the Limits of Learned Importance Scoring for KV Cache Compression
+
+- **ID:** `2601.14279` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Brady Steele
+- **abs:** https://arxiv.org/abs/2601.14279 · **PDF:** https://arxiv.org/pdf/2601.14279v1
+
+## 22183. ForgetMark: Stealthy Fingerprint Embedding via Targeted Unlearning in Language Models
+
+- **ID:** `2601.08189` · **primary:** `cs.CR` · **Submitted:** 2026-01-13
+- **Authors:** Zhenhua Xu, Haobo Zhang, Zhebo Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.08189 · **PDF:** https://arxiv.org/pdf/2601.08189v2
+
+## 22184. Improving LLM Reasoning with Homophily-aware Structural and Semantic Text-Attributed Graph Compression
+
+- **ID:** `2601.08187` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Zijun Di, Bin Lu, Huquan Kang, et al.
+- **abs:** https://arxiv.org/abs/2601.08187 · **PDF:** https://arxiv.org/pdf/2601.08187v2
+
+## 22185. Autonomous Materials Exploration by Integrating Automated Phase Identification and AI-Assisted Human Reasoning
+
+- **ID:** `2601.08185` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-13
+- **Authors:** Ming-Chiang Chang, Maximilian Amsler, Duncan R. Sutherland, et al.
+- **abs:** https://arxiv.org/abs/2601.08185 · **PDF:** https://arxiv.org/pdf/2601.08185v1
+
+## 22186. Wasserstein-p Central Limit Theorem Rates: From Local Dependence to Markov Chains
+
+- **ID:** `2601.08184` · **primary:** `math.PR` · **Submitted:** 2026-01-13
+- **Authors:** Yixuan Zhang, Qiaomin Xie
+- **abs:** https://arxiv.org/abs/2601.08184 · **PDF:** https://arxiv.org/pdf/2601.08184v2
+
+## 22187. GI-Bench: A Panoramic Benchmark Revealing the Knowledge-Experience Dissociation of Multimodal Large Language Models in Gastrointestinal Endoscopy Against Clinical Standards
+
+- **ID:** `2601.08183` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Yan Zhu, Te Luo, Pei-Yao Fu, et al.
+- **abs:** https://arxiv.org/abs/2601.08183 · **PDF:** https://arxiv.org/pdf/2601.08183v2
+
+## 22188. TabPFN Through The Looking Glass: An interpretability study of TabPFN and its internal representations
+
+- **ID:** `2601.08181` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Aviral Gupta, Armaan Sethi, Dhruv Kumar
+- **abs:** https://arxiv.org/abs/2601.08181 · **PDF:** https://arxiv.org/pdf/2601.08181v1
+
+## 22189. Instruction-Driven 3D Facial Expression Generation and Transition
+
+- **ID:** `2601.08179` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Anh H. Vo, Tae-Seok Kim, Hulin Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.08179 · **PDF:** https://arxiv.org/pdf/2601.08179v1
+
+## 22190. Prompt-Based Clarity Evaluation and Topic Detection in Political Question Answering
+
+- **ID:** `2601.08176` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Lavanya Prahallad, Sai Utkarsh Choudarypally, Pragna Prahallad, et al.
+- **abs:** https://arxiv.org/abs/2601.08176 · **PDF:** https://arxiv.org/pdf/2601.08176v1
+
+## 22191. The Agent's First Day: Benchmarking Learning, Exploration, and Scheduling in the Workplace Scenarios
+
+- **ID:** `2601.08173` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Daocheng Fu, Jianbiao Mei, Rong Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.08173 · **PDF:** https://arxiv.org/pdf/2601.08173v1
+
+## 22192. VBO-MI: A Fully Gradient-Based Bayesian Optimization Framework Using Variational Mutual Information Estimation
+
+- **ID:** `2601.08172` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Farhad Mirkarimi
+- **abs:** https://arxiv.org/abs/2601.08172 · **PDF:** https://arxiv.org/pdf/2601.08172v1
+
+## 22193. Relational Knowledge Distillation Using Fine-tuned Function Vectors
+
+- **ID:** `2601.08169` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Andrea Kang, Yingnian Wu, Hongjing Lu
+- **abs:** https://arxiv.org/abs/2601.08169 · **PDF:** https://arxiv.org/pdf/2601.08169v1
+
+## 22194. ZeroDVFS: Zero-Shot LLM-Guided Core and Frequency Allocation for Embedded Platforms
+
+- **ID:** `2601.08166` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Mohammad Pivezhandi, Mahdi Banisharif, Abusayeed Saifullah, et al.
+- **abs:** https://arxiv.org/abs/2601.08166 · **PDF:** https://arxiv.org/pdf/2601.08166v2
+
+## 22195. SwiftMem: Fast Agentic Memory via Query-aware Indexing
+
+- **ID:** `2601.08160` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Anxin Tian, Yiming Li, Xing Li, et al.
+- **abs:** https://arxiv.org/abs/2601.08160 · **PDF:** https://arxiv.org/pdf/2601.08160v1
+
+## 22196. WISE-Flow: Workflow-Induced Structured Experience for Self-Evolving Conversational Service Agents
+
+- **ID:** `2601.08158` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Yuqing Zhou, Zhuoer Wang, Jie Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.08158 · **PDF:** https://arxiv.org/pdf/2601.08158v1
+
+## 22197. Project Synapse: A Hierarchical Multi-Agent Framework with Hybrid Memory for Autonomous Resolution of Last-Mile Delivery Disruptions
+
+- **ID:** `2601.08156` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Arin Gopalan Yadav, Varad Dherange, Kumar Shivam
+- **abs:** https://arxiv.org/abs/2601.08156 · **PDF:** https://arxiv.org/pdf/2601.08156v1
+
+## 22198. Enriching Semantic Profiles into Knowledge Graph for Recommender Systems Using Large Language Models
+
+- **ID:** `2601.08148` · **primary:** `cs.IR` · **Submitted:** 2026-01-13
+- **Authors:** Seokho Ahn, Sungbok Shin, Young-Duk Seo
+- **abs:** https://arxiv.org/abs/2601.08148 · **PDF:** https://arxiv.org/pdf/2601.08148v1
+
+## 22199. Dynamic Graph Structure Learning via Resistance Curvature Flow
+
+- **ID:** `2601.08149` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Chaoqun Fei, Huanjiang Liu, Tinglve Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.08149 · **PDF:** https://arxiv.org/pdf/2601.08149v1
+
+## 22200. Mechanisms are Transferable: Data-Efficient Low-Resource Adaptation via Circuit-Targeted Supervised Fine-Tuning
+
+- **ID:** `2601.08146` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Khumaisa Nur'aini, Ayu Purwarianti, Alham Fikri Aji, et al.
+- **abs:** https://arxiv.org/abs/2601.08146 · **PDF:** https://arxiv.org/pdf/2601.08146v2
+
+## 22201. Qalb: Largest State-of-the-Art Urdu Large Language Model for 230M Speakers with Systematic Continued Pre-training
+
+- **ID:** `2601.08141` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Muhammad Taimoor Hassan, Jawad Ahmed, Muhammad Awais
+- **abs:** https://arxiv.org/abs/2601.08141 · **PDF:** https://arxiv.org/pdf/2601.08141v1
+
+## 22202. Subspace Alignment for Vision-Language Model Test-time Adaptation
+
+- **ID:** `2601.08139` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Zhichen Zeng, Wenxuan Bao, Xiao Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.08139 · **PDF:** https://arxiv.org/pdf/2601.08139v1
+
+## 22203. Critically Engaged Pragmatism: A Scientific Norm and Social, Pragmatist Epistemology for AI Science Evaluation Tools
+
+- **ID:** `2601.09753` · **primary:** `cs.CY` · **Submitted:** 2026-01-13
+- **Authors:** Carole J. Lee
+- **abs:** https://arxiv.org/abs/2601.09753 · **PDF:** https://arxiv.org/pdf/2601.09753v1
+
+## 22204. Reverse Flow Matching: A Unified Framework for Online Reinforcement Learning with Diffusion and Flow Policies
+
+- **ID:** `2601.08136` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Zeyang Li, Sunbochen Tang, Navid Azizan
+- **abs:** https://arxiv.org/abs/2601.08136 · **PDF:** https://arxiv.org/pdf/2601.08136v1
+
+## 22205. Hierarchical Online-Scheduling for Energy-Efficient Split Inference with Progressive Transmission
+
+- **ID:** `2601.08135` · **primary:** `cs.NI` · **Submitted:** 2026-01-13
+- **Authors:** Zengzipeng Tang, Yuxuan Sun, Wei Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.08135 · **PDF:** https://arxiv.org/pdf/2601.08135v1
+
+## 22206. How Reliable are Confidence Estimators for Large Reasoning Models? A Systematic Benchmark on High-Stakes Domains
+
+- **ID:** `2601.08134` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Reza Khanmohammadi, Erfan Miahi, Simerjot Kaur, et al.
+- **abs:** https://arxiv.org/abs/2601.08134 · **PDF:** https://arxiv.org/pdf/2601.08134v2
+
+## 22207. How Do Optical Flow and Textual Prompts Collaborate to Assist in Audio-Visual Semantic Segmentation?
+
+- **ID:** `2601.08133` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Yujian Lee, Peng Gao, Yongqi Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.08133 · **PDF:** https://arxiv.org/pdf/2601.08133v2
+
+## 22208. Attention Projection Mixing with Exogenous Anchors
+
+- **ID:** `2601.08131` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Jonathan Su
+- **abs:** https://arxiv.org/abs/2601.08131 · **PDF:** https://arxiv.org/pdf/2601.08131v3
+
+## 22209. Embedded AI Companion System on Edge Devices
+
+- **ID:** `2601.08128` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Rahul Gupta, Stephen D. H. Hsu
+- **abs:** https://arxiv.org/abs/2601.08128 · **PDF:** https://arxiv.org/pdf/2601.08128v1
+
+## 22210. PathoGen: Diffusion-Based Synthesis of Realistic Lesions in Histopathology Images
+
+- **ID:** `2601.08127` · **primary:** `cs.CV` · **Submitted:** 2026-01-13
+- **Authors:** Mohamad Koohi-Moghadam, Mohammad-Ali Nikouei Mahani, Kyongtae Tyler Bae
+- **abs:** https://arxiv.org/abs/2601.08127 · **PDF:** https://arxiv.org/pdf/2601.08127v1
+
+## 22211. How vehicles change lanes after encountering crashes: Empirical analysis and modeling
+
+- **ID:** `2601.08125` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Kequan Chen, Yuxuan Wang, Pan Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.08125 · **PDF:** https://arxiv.org/pdf/2601.08125v1
+
+## 22212. Generalization Analysis and Method for Domain Generalization for a Family of Recurrent Neural Networks
+
+- **ID:** `2601.08122` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Atefeh Termehchi, Ekram Hossain, Isaac Woungang
+- **abs:** https://arxiv.org/abs/2601.08122 · **PDF:** https://arxiv.org/pdf/2601.08122v1
+
+## 22213. Intra-tree Column Subsampling Hinders XGBoost Learning of Ratio-like Interactions
+
+- **ID:** `2601.08121` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Mykola Pinchuk
+- **abs:** https://arxiv.org/abs/2601.08121 · **PDF:** https://arxiv.org/pdf/2601.08121v1
+
+## 22214. Structure Detection for Contextual Reinforcement Learning
+
+- **ID:** `2601.08120` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Tianyue Zhou, Jung-Hoon Cho, Cathy Wu
+- **abs:** https://arxiv.org/abs/2601.08120 · **PDF:** https://arxiv.org/pdf/2601.08120v1
+
+## 22215. MirrorBench: A Benchmark to Evaluate Conversational User-Proxy Agents for Human-Likeness
+
+- **ID:** `2601.08118` · **primary:** `cs.AI` · **Submitted:** 2026-01-13
+- **Authors:** Ashutosh Hathidara, Julien Yu, Vaishali Senthil, et al.
+- **abs:** https://arxiv.org/abs/2601.08118 · **PDF:** https://arxiv.org/pdf/2601.08118v2
+
+## 22216. Learning a Stochastic Differential Equation Model of Tropical Cyclone Intensification from Reanalysis and Observational Data
+
+- **ID:** `2601.08116` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Kenneth Gee, Sai Ravela
+- **abs:** https://arxiv.org/abs/2601.08116 · **PDF:** https://arxiv.org/pdf/2601.08116v1
+
+## 22217. CSQL: Mapping Documents into Causal Databases
+
+- **ID:** `2601.08109` · **primary:** `cs.DB` · **Submitted:** 2026-01-13
+- **Authors:** Sridhar Mahadevan
+- **abs:** https://arxiv.org/abs/2601.08109 · **PDF:** https://arxiv.org/pdf/2601.08109v1
+
+## 22218. Debiasing Large Language Models via Adaptive Causal Prompting with Sketch-of-Thought
+
+- **ID:** `2601.08108` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Bowen Li, Ziqi Xu, Jing Ren, et al.
+- **abs:** https://arxiv.org/abs/2601.08108 · **PDF:** https://arxiv.org/pdf/2601.08108v1
+
+## 22219. STO-RL: Offline RL under Sparse Rewards via LLM-Guided Subgoal Temporal Order
+
+- **ID:** `2601.08107` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Chengyang Gu, Yuxin Pan, Hui Xiong, et al.
+- **abs:** https://arxiv.org/abs/2601.08107 · **PDF:** https://arxiv.org/pdf/2601.08107v1
+
+## 22220. Query Suggestion for Retrieval-Augmented Generation via Dynamic In-Context Learning
+
+- **ID:** `2601.08105` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Fabian Spaeh, Tianyi Chen, Chen-Hao Chiang, et al.
+- **abs:** https://arxiv.org/abs/2601.08105 · **PDF:** https://arxiv.org/pdf/2601.08105v1
+
+## 22221. High-Fidelity Modeling of Stochastic Chemical Dynamics on Complex Manifolds: A Multi-Scale SIREN-PINN Framework for the Curvature-Perturbed Ginzburg-Landau Equation
+
+- **ID:** `2601.08104` · **primary:** `nlin.CD` · **Submitted:** 2026-01-13
+- **Authors:** Julian Evan Chrisnanto, Salsabila Rahma Alia, Nurfauzi Fadillah, et al.
+- **abs:** https://arxiv.org/abs/2601.08104 · **PDF:** https://arxiv.org/pdf/2601.08104v2
+
+## 22222. Towards A Unified PAC-Bayesian Framework for Norm-based Generalization Bounds
+
+- **ID:** `2601.08100` · **primary:** `stat.ML` · **Submitted:** 2026-01-13
+- **Authors:** Xinping Yi, Gaojie Jin, Xiaowei Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.08100 · **PDF:** https://arxiv.org/pdf/2601.08100v1
+
+## 22223. AdaJudge: Adaptive Multi-Perspective Judging for Reward Modeling
+
+- **ID:** `2601.08097` · **primary:** `cs.CL` · **Submitted:** 2026-01-13
+- **Authors:** Yongliang Miao, Yangyang Liang, Mengnan Du
+- **abs:** https://arxiv.org/abs/2601.08097 · **PDF:** https://arxiv.org/pdf/2601.08097v1
+
+## 22224. Local-Global Feature Fusion for Subject-Independent EEG Emotion Recognition
+
+- **ID:** `2601.08094` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Zheng Zhou, Isabella McEvoy, Camilo E. Valderrama
+- **abs:** https://arxiv.org/abs/2601.08094 · **PDF:** https://arxiv.org/pdf/2601.08094v1
+
+## 22225. Q-realign: Piggybacking Realignment on Quantization for Safe and Efficient LLM Deployment
+
+- **ID:** `2601.08089` · **primary:** `cs.LG` · **Submitted:** 2026-01-13
+- **Authors:** Qitao Tan, Xiaoying Song, Ningxi Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.08089 · **PDF:** https://arxiv.org/pdf/2601.08089v1
+
+## 22226. Bridging the Gap: Empowering Small Models in Reliable OpenACC-based Parallelization via GEPA-Optimized Prompting
+
+- **ID:** `2601.08884` · **primary:** `cs.SE` · **Submitted:** 2026-01-12
+- **Authors:** Samyak Jhaveri, Cristina V. Lopes
+- **abs:** https://arxiv.org/abs/2601.08884 · **PDF:** https://arxiv.org/pdf/2601.08884v1
+
+## 22227. Limits of n-gram Style Control for LLMs via Logit-Space Injection
+
+- **ID:** `2601.16224` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Sami-ul Ahmed
+- **abs:** https://arxiv.org/abs/2601.16224 · **PDF:** https://arxiv.org/pdf/2601.16224v1
+
+## 22228. MemoBrain: Executive Memory as an Agentic Brain for Reasoning
+
+- **ID:** `2601.08079` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Hongjin Qian, Zhao Cao, Zheng Liu
+- **abs:** https://arxiv.org/abs/2601.08079 · **PDF:** https://arxiv.org/pdf/2601.08079v1
+
+## 22229. Exploiting DINOv3-Based Self-Supervised Features for Robust Few-Shot Medical Image Segmentation
+
+- **ID:** `2601.08078` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Guoping Xu, Jayaram K. Udupa, Weiguo Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.08078 · **PDF:** https://arxiv.org/pdf/2601.08078v1
+
+## 22230. Semantic Gravity Wells: Why Negative Constraints Backfire
+
+- **ID:** `2601.08070` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Shailesh Rana
+- **abs:** https://arxiv.org/abs/2601.08070 · **PDF:** https://arxiv.org/pdf/2601.08070v1
+
+## 22231. A New Strategy for Verifying Reach-Avoid Specifications in Neural Feedback Systems
+
+- **ID:** `2601.08065` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Samuel I. Akinwande, Sydney M. Katz, Mykel J. Kochenderfer, et al.
+- **abs:** https://arxiv.org/abs/2601.08065 · **PDF:** https://arxiv.org/pdf/2601.08065v1
+
+## 22232. Calibration Is Not Enough: Evaluating Confidence Estimation Under Language Variations
+
+- **ID:** `2601.08064` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Yuxi Xia, Dennis Ulmer, Terra Blevins, et al.
+- **abs:** https://arxiv.org/abs/2601.08064 · **PDF:** https://arxiv.org/pdf/2601.08064v1
+
+## 22233. Universal computation is intrinsic to language model decoding
+
+- **ID:** `2601.08061` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Alex Lewandowski, Marlos C. Machado, Dale Schuurmans
+- **abs:** https://arxiv.org/abs/2601.08061 · **PDF:** https://arxiv.org/pdf/2601.08061v2
+
+## 22234. Reasoning Beyond Chain-of-Thought: A Latent Computational Mode in Large Language Models
+
+- **ID:** `2601.08058` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Zhenghao He, Guangzhi Xiong, Bohan Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.08058 · **PDF:** https://arxiv.org/pdf/2601.08058v1
+
+## 22235. Forecast Aware Deep Reinforcement Learning for Efficient Electricity Load Scheduling in Dairy Farms
+
+- **ID:** `2601.08052` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Nawazish Ali, Rachael Shaw, Karl Mason
+- **abs:** https://arxiv.org/abs/2601.08052 · **PDF:** https://arxiv.org/pdf/2601.08052v2
+
+## 22236. Integrating Attendance Tracking and Emotion Detection for Enhanced Student Engagement in Smart Classrooms
+
+- **ID:** `2601.08049` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Keith Ainebyona, Ann Move Oguti, Joseph Walusimbi, et al.
+- **abs:** https://arxiv.org/abs/2601.08049 · **PDF:** https://arxiv.org/pdf/2601.08049v1
+
+## 22237. LUT-Compiled Kolmogorov-Arnold Networks for Lightweight DoS Detection on IoT Edge Devices
+
+- **ID:** `2601.08044` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Oleksandr Kuznetsov
+- **abs:** https://arxiv.org/abs/2601.08044 · **PDF:** https://arxiv.org/pdf/2601.08044v1
+
+## 22238. The Role of Noisy Data in Improving CNN Robustness for Image Classification
+
+- **ID:** `2601.08043` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Oscar H. Ramírez-Agudelo, Nicoleta Gorea, Aliza Reif, et al.
+- **abs:** https://arxiv.org/abs/2601.08043 · **PDF:** https://arxiv.org/pdf/2601.08043v1
+
+## 22239. Riemannian Zeroth-Order Gradient Estimation with Structure-Preserving Metrics for Geodesically Incomplete Manifolds
+
+- **ID:** `2601.08039` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Shaocong Ma, Heng Huang
+- **abs:** https://arxiv.org/abs/2601.08039 · **PDF:** https://arxiv.org/pdf/2601.08039v1
+
+## 22240. InfGraND: An Influence-Guided GNN-to-MLP Knowledge Distillation
+
+- **ID:** `2601.08033` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Amir Eskandari, Aman Anand, Elyas Rashno, et al.
+- **abs:** https://arxiv.org/abs/2601.08033 · **PDF:** https://arxiv.org/pdf/2601.08033v1
+
+## 22241. FigEx2: Visual-Conditioned Panel Detection and Captioning for Scientific Compound Figures
+
+- **ID:** `2601.08026` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Jifeng Song, Arun Das, Pan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.08026 · **PDF:** https://arxiv.org/pdf/2601.08026v4
+
+## 22242. Representations of Text and Images Align From Layer One
+
+- **ID:** `2601.08017` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Evžen Wybitul, Javier Rando, Florian Tramèr, et al.
+- **abs:** https://arxiv.org/abs/2601.08017 · **PDF:** https://arxiv.org/pdf/2601.08017v1
+
+## 22243. Beyond the Next Port: A Multi-Task Transformer for Forecasting Future Voyage Segment Durations
+
+- **ID:** `2601.08013` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Nairui Liu, Fang He, Xindi Tang
+- **abs:** https://arxiv.org/abs/2601.08013 · **PDF:** https://arxiv.org/pdf/2601.08013v1
+
+## 22244. TP-Blend: Textual-Prompt Attention Pairing for Precise Object-Style Blending in Diffusion Models
+
+- **ID:** `2601.08011` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Xin Jin, Yichuan Zhong, Yapeng Tian
+- **abs:** https://arxiv.org/abs/2601.08011 · **PDF:** https://arxiv.org/pdf/2601.08011v4
+
+## 22245. Internal Deployment Gaps in AI Regulation
+
+- **ID:** `2601.08005` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Joe Kwon, Stephen Casper
+- **abs:** https://arxiv.org/abs/2601.08005 · **PDF:** https://arxiv.org/pdf/2601.08005v3
+
+## 22246. LLM Review: Enhancing Creative Writing via Blind Peer Review Feedback
+
+- **ID:** `2601.08003` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Weiyue Li, Mingxiao Song, Zhenda Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.08003 · **PDF:** https://arxiv.org/pdf/2601.08003v1
+
+## 22247. Operator learning for models of tear film breakup
+
+- **ID:** `2601.08001` · **primary:** `math.NA` · **Submitted:** 2026-01-12
+- **Authors:** Qinying Chen, Arnab Roy, Tobin A. Driscoll
+- **abs:** https://arxiv.org/abs/2601.08001 · **PDF:** https://arxiv.org/pdf/2601.08001v1
+
+## 22248. Reasoning over Precedents Alongside Statutes: Case-Augmented Deliberative Alignment for LLM Safety
+
+- **ID:** `2601.08000` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Can Jin, Rui Wu, Tong Che, et al.
+- **abs:** https://arxiv.org/abs/2601.08000 · **PDF:** https://arxiv.org/pdf/2601.08000v1
+
+## 22249. Is Sentiment Banana-Shaped? Exploring the Geometry and Portability of Sentiment Concept Vectors
+
+- **ID:** `2601.07995` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Laurits Lyngbaek, Pascale Feldkamp, Yuri Bizzoni, et al.
+- **abs:** https://arxiv.org/abs/2601.07995 · **PDF:** https://arxiv.org/pdf/2601.07995v1
+
+## 22250. DYCP: Dynamic Context Pruning for Long-Form Dialogue with LLMs
+
+- **ID:** `2601.07994` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Nayoung Choi, Jonathan Zhang, Jinho D. Choi
+- **abs:** https://arxiv.org/abs/2601.07994 · **PDF:** https://arxiv.org/pdf/2601.07994v3
+
+## 22251. From Word Sequences to Behavioral Sequences: Adapting Modeling and Evaluation Paradigms for Longitudinal NLP
+
+- **ID:** `2601.07988` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Adithya V Ganesan, Vasudha Varadarajan, Oscar NE Kjell, et al.
+- **abs:** https://arxiv.org/abs/2601.07988 · **PDF:** https://arxiv.org/pdf/2601.07988v1
+
+## 22252. VULCA-Bench: A Multicultural Vision-Language Benchmark for Evaluating Cultural Understanding
+
+- **ID:** `2601.07986` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Haorui Yu, Diji Yang, Hang He, et al.
+- **abs:** https://arxiv.org/abs/2601.07986 · **PDF:** https://arxiv.org/pdf/2601.07986v3
+
+## 22253. Multilingual, Multimodal Pipeline for Creating Authentic and Structured Fact-Checked Claim Dataset
+
+- **ID:** `2601.07985` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Z. Melce Hüsünbeyi, Virginie Mouilleron, Leonie Uhling, et al.
+- **abs:** https://arxiv.org/abs/2601.07985 · **PDF:** https://arxiv.org/pdf/2601.07985v3
+
+## 22254. Gated Sparse Attention: Combining Computational Efficiency with Training Stability for Long-Context Language Models
+
+- **ID:** `2601.15305` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Alfred Shen, Aaron Shen
+- **abs:** https://arxiv.org/abs/2601.15305 · **PDF:** https://arxiv.org/pdf/2601.15305v1
+
+## 22255. Cross-Cultural Expert-Level Art Critique Evaluation with Vision-Language Models
+
+- **ID:** `2601.07984` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Haorui Yu, Xuehang Wen, Fengrui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.07984 · **PDF:** https://arxiv.org/pdf/2601.07984v3
+
+## 22256. A Dataset of Dengue Hospitalizations in Brazil (1999 to 2021) with Weekly Disaggregation from Monthly Counts
+
+- **ID:** `2601.16994` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Lucas M. Morello, Matheus Lima Castro, Pedro Cesar M. G. Camargo, et al.
+- **abs:** https://arxiv.org/abs/2601.16994 · **PDF:** https://arxiv.org/pdf/2601.16994v1
+
+## 22257. Spatial Covariance Constraints for Gaussian Mixture Models
+
+- **ID:** `2601.07979` · **primary:** `stat.ME` · **Submitted:** 2026-01-12
+- **Authors:** Hanzhang Lu, Keiran Malott, Venkat Suprabath Bitra, et al.
+- **abs:** https://arxiv.org/abs/2601.07979 · **PDF:** https://arxiv.org/pdf/2601.07979v1
+
+## 22258. Explaining Generalization of AI-Generated Text Detectors Through Linguistic Analysis
+
+- **ID:** `2601.07974` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Yuxi Xia, Kinga Stańczak, Benjamin Roth
+- **abs:** https://arxiv.org/abs/2601.07974 · **PDF:** https://arxiv.org/pdf/2601.07974v2
+
+## 22259. Cultural Compass: A Framework for Organizing Societal Norms to Detect Violations in Human-AI Conversations
+
+- **ID:** `2601.07973` · **primary:** `cs.CY` · **Submitted:** 2026-01-12
+- **Authors:** Myra Cheng, Vinodkumar Prabhakaran, Alice Oh, et al.
+- **abs:** https://arxiv.org/abs/2601.07973 · **PDF:** https://arxiv.org/pdf/2601.07973v1
+
+## 22260. Knowing But Not Doing: Convergent Morality and Divergent Action in LLMs
+
+- **ID:** `2601.07972` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Jen-tse Huang, Jiantong Qin, Xueli Qiu, et al.
+- **abs:** https://arxiv.org/abs/2601.07972 · **PDF:** https://arxiv.org/pdf/2601.07972v1
+
+## 22261. Tuberculosis Screening from Cough Audio: Baseline Models, Clinical Variables, and Uncertainty Quantification
+
+- **ID:** `2601.07969` · **primary:** `eess.AS` · **Submitted:** 2026-01-12
+- **Authors:** George P. Kafentzis, Efstratios Selisios
+- **abs:** https://arxiv.org/abs/2601.07969 · **PDF:** https://arxiv.org/pdf/2601.07969v2
+
+## 22262. DataScribe: An AI-Native, Policy-Aligned Web Platform for Multi-Objective Materials Design and Discovery
+
+- **ID:** `2601.07966` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Divyanshu Singh, Doguhan Sarıtürk, Cameron Lea, et al.
+- **abs:** https://arxiv.org/abs/2601.07966 · **PDF:** https://arxiv.org/pdf/2601.07966v1
+
+## 22263. When Models Know When They Do Not Know: Calibration, Cascading, and Cleaning
+
+- **ID:** `2601.07965` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Chenjie Hao, Weyl Lu, Yuko Ishiwaka, et al.
+- **abs:** https://arxiv.org/abs/2601.07965 · **PDF:** https://arxiv.org/pdf/2601.07965v1
+
+## 22264. Executable Ontologies in Game Development: From Algorithmic Control to Semantic World Modeling
+
+- **ID:** `2601.07964` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Alexander Boldachev
+- **abs:** https://arxiv.org/abs/2601.07964 · **PDF:** https://arxiv.org/pdf/2601.07964v1
+
+## 22265. LJ-Spoof: A Generatively Varied Corpus for Audio Anti-Spoofing and Synthesis Source Tracing
+
+- **ID:** `2601.07958` · **primary:** `cs.SD` · **Submitted:** 2026-01-12
+- **Authors:** Surya Subramani, Hashim Ali, Hafiz Malik
+- **abs:** https://arxiv.org/abs/2601.07958 · **PDF:** https://arxiv.org/pdf/2601.07958v1
+
+## 22266. LWMSCNN-SE: A Lightweight Multi-Scale Network for Efficient Maize Disease Classification on Edge Devices
+
+- **ID:** `2601.07957` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Fikadu Weloday, Jianmei Su
+- **abs:** https://arxiv.org/abs/2601.07957 · **PDF:** https://arxiv.org/pdf/2601.07957v1
+
+## 22267. A Human-Centric Pipeline for Aligning Large Language Models with Chinese Medical Ethics
+
+- **ID:** `2601.07954` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Haoan Jin, Han Ying, Jiacheng Ji, et al.
+- **abs:** https://arxiv.org/abs/2601.07954 · **PDF:** https://arxiv.org/pdf/2601.07954v1
+
+## 22268. Quantum automated theorem proving
+
+- **ID:** `2601.07953` · **primary:** `quant-ph` · **Submitted:** 2026-01-12
+- **Authors:** Zheng-Zhi Sun, Qi Ye, Dong-Ling Deng
+- **abs:** https://arxiv.org/abs/2601.07953 · **PDF:** https://arxiv.org/pdf/2601.07953v1
+
+## 22269. Hybrid SARIMA LSTM Model for Local Weather Forecasting: A Residual Learning Approach for Data Driven Meteorological Prediction
+
+- **ID:** `2601.07951` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Shreyas Rajeev, Karthik Mudenahalli Ashoka, Amit Mallappa Tiparaddi
+- **abs:** https://arxiv.org/abs/2601.07951 · **PDF:** https://arxiv.org/pdf/2601.07951v1
+
+## 22270. Reinforcement Learning Methods for Neighborhood Selection in Local Search
+
+- **ID:** `2601.07948` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Yannick Molinghen, Augustin Delecluse, Renaud De Landtsheer, et al.
+- **abs:** https://arxiv.org/abs/2601.07948 · **PDF:** https://arxiv.org/pdf/2601.07948v1
+
+## 22271. CRAFT: Cost-aware Expert Replica Allocation with Fine-Grained Layerwise Estimations
+
+- **ID:** `2603.28768` · **primary:** `cs.DC` · **Submitted:** 2026-01-12
+- **Authors:** Adrian Zhao, Zhenkun Cai, Zhenyu Song, et al.
+- **abs:** https://arxiv.org/abs/2603.28768 · **PDF:** https://arxiv.org/pdf/2603.28768v1
+
+## 22272. Coupled Diffusion-Encoder Models for Reconstruction of Flow Fields
+
+- **ID:** `2601.07946` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** AmirPouya Hemmasian, Amir Barati Farimani
+- **abs:** https://arxiv.org/abs/2601.07946 · **PDF:** https://arxiv.org/pdf/2601.07946v1
+
+## 22273. A Statistical Assessment of Amortized Inference Under Signal-to-Noise Variation and Distribution Shift
+
+- **ID:** `2601.07944` · **primary:** `stat.ML` · **Submitted:** 2026-01-12
+- **Authors:** Roy Shivam Ram Shreshtth, Arnab Hazra, Gourab Mukherjee
+- **abs:** https://arxiv.org/abs/2601.07944 · **PDF:** https://arxiv.org/pdf/2601.07944v1
+
+## 22274. Enhancing Portfolio Optimization with Deep Learning Insights
+
+- **ID:** `2601.07942` · **primary:** `q-fin.PM` · **Submitted:** 2026-01-12
+- **Authors:** Brandon Luo, Jim Skufca
+- **abs:** https://arxiv.org/abs/2601.07942 · **PDF:** https://arxiv.org/pdf/2601.07942v1
+
+## 22275. Moonworks Lunara Aesthetic Dataset
+
+- **ID:** `2601.07941` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Yan Wang, Sayeef Abdullah, Partho Hassan, et al.
+- **abs:** https://arxiv.org/abs/2601.07941 · **PDF:** https://arxiv.org/pdf/2601.07941v4
+
+## 22276. SECite: Analyzing and Summarizing Citations in Software Engineering Literature
+
+- **ID:** `2601.07939` · **primary:** `cs.SE` · **Submitted:** 2026-01-12
+- **Authors:** Shireesh Reddy Pyreddy, Khaja Valli Pathan, Hasan Masum, et al.
+- **abs:** https://arxiv.org/abs/2601.07939 · **PDF:** https://arxiv.org/pdf/2601.07939v1
+
+## 22277. Towards Specialized Generalists: A Multi-Task MoE-LoRA Framework for Domain-Specific LLM Adaptation
+
+- **ID:** `2601.07935` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Yuxin Yang, Aoxiong Zeng, Xiangquan Yang
+- **abs:** https://arxiv.org/abs/2601.07935 · **PDF:** https://arxiv.org/pdf/2601.07935v1
+
+## 22278. Transformer-Based Approach for Automated Functional Group Replacement in Chemical Compounds
+
+- **ID:** `2601.07930` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Bo Pan, Zhiping Zhang, Kevin Spiekermann, et al.
+- **abs:** https://arxiv.org/abs/2601.07930 · **PDF:** https://arxiv.org/pdf/2601.07930v1
+
+## 22279. A Complete Decomposition of Stochastic Differential Equations
+
+- **ID:** `2601.07834` · **primary:** `math.PR` · **Submitted:** 2026-01-12
+- **Authors:** Samuel Duffield
+- **abs:** https://arxiv.org/abs/2601.07834 · **PDF:** https://arxiv.org/pdf/2601.07834v1
+
+## 22280. MHLA: Restoring Expressivity of Linear Attention via Token-Level Multi-Head
+
+- **ID:** `2601.07832` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Kewei Zhang, Ye Huang, Yufan Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.07832 · **PDF:** https://arxiv.org/pdf/2601.07832v2
+
+## 22281. Optimal Learning Rate Schedule for Balancing Effort and Performance
+
+- **ID:** `2601.07830` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Valentina Njaradi, Rodrigo Carrasco-Davis, Peter E. Latham, et al.
+- **abs:** https://arxiv.org/abs/2601.07830 · **PDF:** https://arxiv.org/pdf/2601.07830v1
+
+## 22282. Failure-Aware RL: Reliable Offline-to-Online Reinforcement Learning with Self-Recovery for Real-World Manipulation
+
+- **ID:** `2601.07821` · **primary:** `cs.RO` · **Submitted:** 2026-01-12
+- **Authors:** Huanyu Li, Kun Lei, Sheng Zang, et al.
+- **abs:** https://arxiv.org/abs/2601.07821 · **PDF:** https://arxiv.org/pdf/2601.07821v1
+
+## 22283. Reference Games as a Testbed for the Alignment of Model Uncertainty and Clarification Requests
+
+- **ID:** `2601.07820` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Manar Ali, Judith Sieker, Sina Zarrieß, et al.
+- **abs:** https://arxiv.org/abs/2601.07820 · **PDF:** https://arxiv.org/pdf/2601.07820v1
+
+## 22284. Compressing Vision Transformers in Geospatial Transfer Learning with Manifold-Constrained Optimization
+
+- **ID:** `2601.08882` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Thomas Snyder, H. Lexie Yang, Stefan Schnake, et al.
+- **abs:** https://arxiv.org/abs/2601.08882 · **PDF:** https://arxiv.org/pdf/2601.08882v1
+
+## 22285. The Confidence Trap: Gender Bias and Predictive Certainty in LLMs
+
+- **ID:** `2601.07806` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Ahmed Sabir, Markus Kängsepp, Rajesh Sharma
+- **abs:** https://arxiv.org/abs/2601.07806 · **PDF:** https://arxiv.org/pdf/2601.07806v1
+
+## 22286. Concatenated Matrix SVD: Compression Bounds, Incremental Approximation, and Error-Constrained Clustering
+
+- **ID:** `2601.11626` · **primary:** `math.NA` · **Submitted:** 2026-01-12
+- **Authors:** Maksym Shamrai
+- **abs:** https://arxiv.org/abs/2601.11626 · **PDF:** https://arxiv.org/pdf/2601.11626v1
+
+## 22287. PTCBENCH: Benchmarking Contextual Stability of Personality Traits in LLM Systems
+
+- **ID:** `2602.00016` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Jiongchi Yu, Yuhan Ma, Xiaoyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.00016 · **PDF:** https://arxiv.org/pdf/2602.00016v1
+
+## 22288. Learning Through Dialogue: Engagement and Efficacy Matter More Than Explanations
+
+- **ID:** `2601.07796` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Shaz Furniturewala, Gerard Christopher Yeo, Kokil Jaidka
+- **abs:** https://arxiv.org/abs/2601.07796 · **PDF:** https://arxiv.org/pdf/2601.07796v2
+
+## 22289. Kinship Data Benchmark for Multi-hop Reasoning
+
+- **ID:** `2601.07794` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Tianda Sun, Dimitar Kazakov
+- **abs:** https://arxiv.org/abs/2601.07794 · **PDF:** https://arxiv.org/pdf/2601.07794v1
+
+## 22290. Non-Convex Portfolio Optimization via Energy-Based Models: A Comparative Analysis Using the Thermodynamic HypergRaphical Model Library (THRML) for Index Tracking
+
+- **ID:** `2601.07792` · **primary:** `q-fin.CP` · **Submitted:** 2026-01-12
+- **Authors:** Javier Mancilla, Theodoros D. Bouloumis, Frederic Goguikian
+- **abs:** https://arxiv.org/abs/2601.07792 · **PDF:** https://arxiv.org/pdf/2601.07792v1
+
+## 22291. Benchmarking Small Language Models and Small Reasoning Language Models on System Log Severity Classification
+
+- **ID:** `2601.07790` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Yahya Masri, Emily Ma, Zifu Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.07790 · **PDF:** https://arxiv.org/pdf/2601.07790v1
+
+## 22292. Beyond Single-Shot: Multi-step Tool Retrieval via Query Planning
+
+- **ID:** `2601.07782` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Wei Fang, James Glass
+- **abs:** https://arxiv.org/abs/2601.07782 · **PDF:** https://arxiv.org/pdf/2601.07782v1
+
+## 22293. Enhancing Self-Correction in Large Language Models through Multi-Perspective Reflection
+
+- **ID:** `2601.07780` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Mariana Costa, Alberlucia Rafael Soarez, Daniel Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.07780 · **PDF:** https://arxiv.org/pdf/2601.07780v1
+
+## 22294. OS-Symphony: A Holistic Framework for Robust and Generalist Computer-Using Agent
+
+- **ID:** `2601.07779` · **primary:** `cs.MA` · **Submitted:** 2026-01-12
+- **Authors:** Bowen Yang, Kaiming Jin, Zhenyu Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.07779 · **PDF:** https://arxiv.org/pdf/2601.07779v1
+
+## 22295. DT-ICU: Towards Explainable Digital Twins for ICU Patient Monitoring via Multi-Modal and Multi-Task Iterative Inference
+
+- **ID:** `2601.07778` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Wen Guo
+- **abs:** https://arxiv.org/abs/2601.07778 · **PDF:** https://arxiv.org/pdf/2601.07778v1
+
+## 22296. Are LLM Decisions Faithful to Verbal Confidence?
+
+- **ID:** `2601.07767` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Jiawei Wang, Yanfei Zhou, Siddartha Devic, et al.
+- **abs:** https://arxiv.org/abs/2601.07767 · **PDF:** https://arxiv.org/pdf/2601.07767v1
+
+## 22297. Contrastive Learning with Narrative Twins for Modeling Story Salience
+
+- **ID:** `2601.07765` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Igor Sterner, Alex Lascarides, Frank Keller
+- **abs:** https://arxiv.org/abs/2601.07765 · **PDF:** https://arxiv.org/pdf/2601.07765v1
+
+## 22298. Reasoning Stabilization Point: A Training-Time Signal for Stable Evidence and Shortcut Reliance
+
+- **ID:** `2601.11625` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Sahil Rajesh Dhayalkar
+- **abs:** https://arxiv.org/abs/2601.11625 · **PDF:** https://arxiv.org/pdf/2601.11625v1
+
+## 22299. Free-RBF-KAN: Kolmogorov-Arnold Networks with Adaptive Radial Basis Functions for Efficient Function Learning
+
+- **ID:** `2601.07760` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Shao-Ting Chiu, Siu Wun Cheung, Ulisses Braga-Neto, et al.
+- **abs:** https://arxiv.org/abs/2601.07760 · **PDF:** https://arxiv.org/pdf/2601.07760v3
+
+## 22300. Learning to bin: differentiable and Bayesian optimization for multi-dimensional discriminants in high-energy physics
+
+- **ID:** `2601.07756` · **primary:** `physics.data-an` · **Submitted:** 2026-01-12
+- **Authors:** Johannes Erdmann, Nitish Kumar Kasaraguppe, Florian Mausolf
+- **abs:** https://arxiv.org/abs/2601.07756 · **PDF:** https://arxiv.org/pdf/2601.07756v1
+
+## 22301. Structure First, Reason Next: Enhancing a Large Language Model using Knowledge Graph for Numerical Reasoning in Financial Documents
+
+- **ID:** `2601.07754` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Aryan Mishra, Akash Anil
+- **abs:** https://arxiv.org/abs/2601.07754 · **PDF:** https://arxiv.org/pdf/2601.07754v1
+
+## 22302. A Unified Framework for Debiased Machine Learning: Riesz Representer Fitting under Bregman Divergence
+
+- **ID:** `2601.07752` · **primary:** `econ.EM` · **Submitted:** 2026-01-12
+- **Authors:** Masahiro Kato
+- **abs:** https://arxiv.org/abs/2601.07752 · **PDF:** https://arxiv.org/pdf/2601.07752v3
+
+## 22303. Improving Domain Generalization in Contrastive Learning using Adaptive Temperature Control
+
+- **ID:** `2601.07748` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Robert Lewis, Katie Matton, Rosalind W. Picard, et al.
+- **abs:** https://arxiv.org/abs/2601.07748 · **PDF:** https://arxiv.org/pdf/2601.07748v1
+
+## 22304. PFT: Phonon Fine-tuning for Machine Learned Interatomic Potentials
+
+- **ID:** `2601.07742` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-12
+- **Authors:** Teddy Koker, Abhijeet Gangan, Mit Kotak, et al.
+- **abs:** https://arxiv.org/abs/2601.07742 · **PDF:** https://arxiv.org/pdf/2601.07742v3
+
+## 22305. Evaluating the encoding competence of visual language models using uncommon actions
+
+- **ID:** `2601.07737` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Chen Ling, Nai Ding
+- **abs:** https://arxiv.org/abs/2601.07737 · **PDF:** https://arxiv.org/pdf/2601.07737v1
+
+## 22306. Backward Reconstruction of the Chafee--Infante Equation via Physics-Informed WGAN-GP
+
+- **ID:** `2601.07733` · **primary:** `math.AP` · **Submitted:** 2026-01-12
+- **Authors:** Joseph L. Shomberg
+- **abs:** https://arxiv.org/abs/2601.07733 · **PDF:** https://arxiv.org/pdf/2601.07733v1
+
+## 22307. Hiking in the Wild: A Scalable Perceptive Parkour Framework for Humanoids
+
+- **ID:** `2601.07718` · **primary:** `cs.RO` · **Submitted:** 2026-01-12
+- **Authors:** Shaoting Zhu, Ziwen Zhuang, Mengjie Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.07718 · **PDF:** https://arxiv.org/pdf/2601.07718v1
+
+## 22308. Is Agentic RAG worth it? An experimental comparison of RAG approaches
+
+- **ID:** `2601.07711` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Pietro Ferrazzi, Milica Cvjeticanin, Alessio Piraccini, et al.
+- **abs:** https://arxiv.org/abs/2601.07711 · **PDF:** https://arxiv.org/pdf/2601.07711v1
+
+## 22309. Deep Whole-body Parkour
+
+- **ID:** `2601.07701` · **primary:** `cs.RO` · **Submitted:** 2026-01-12
+- **Authors:** Ziwen Zhuang, Shaoting Zhu, Mengjie Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.07701 · **PDF:** https://arxiv.org/pdf/2601.07701v1
+
+## 22310. Hidden Monotonicity: Explaining Deep Neural Networks via their DC Decomposition
+
+- **ID:** `2601.07700` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Jakob Paul Zimmermann, Georg Loho
+- **abs:** https://arxiv.org/abs/2601.07700 · **PDF:** https://arxiv.org/pdf/2601.07700v2
+
+## 22311. BibAgent: An Agentic Framework for Traceable Miscitation Detection in Scientific Literature
+
+- **ID:** `2601.16993` · **primary:** `cs.DL` · **Submitted:** 2026-01-12
+- **Authors:** Peiran Li, Fangzhou Lin, Shuo Xing, et al.
+- **abs:** https://arxiv.org/abs/2601.16993 · **PDF:** https://arxiv.org/pdf/2601.16993v2
+
+## 22312. Emotional Support Evaluation Framework via Controllable and Diverse Seeker Simulator
+
+- **ID:** `2601.07698` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Chaewon Heo, Cheyon Jin, Yohan Jo
+- **abs:** https://arxiv.org/abs/2601.07698 · **PDF:** https://arxiv.org/pdf/2601.07698v1
+
+## 22313. Exploring the Meta-level Reasoning of Large Language Models via a Tool-based Multi-hop Tabular Question Answering Task
+
+- **ID:** `2601.07696` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Nick Ferguson, Alan Bundy, Kwabena Nuamah
+- **abs:** https://arxiv.org/abs/2601.07696 · **PDF:** https://arxiv.org/pdf/2601.07696v1
+
+## 22314. A Framework for Feature Discovery in Intracranial Pressure Monitoring Data Using Neural Network Attention
+
+- **ID:** `2601.07691` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-12
+- **Authors:** Jonathan D. Socha, Seyed F. Maroufi, Dipankar Biswas, et al.
+- **abs:** https://arxiv.org/abs/2601.07691 · **PDF:** https://arxiv.org/pdf/2601.07691v1
+
+## 22315. Physics-Informed Singular-Value Learning for Cross-Covariances Forecasting in Financial Markets
+
+- **ID:** `2601.07687` · **primary:** `q-fin.ST` · **Submitted:** 2026-01-12
+- **Authors:** Efstratios Manolakis, Christian Bongiorno, Rosario Nunzio Mantegna
+- **abs:** https://arxiv.org/abs/2601.07687 · **PDF:** https://arxiv.org/pdf/2601.07687v2
+
+## 22316. Predictive Analytics for Dementia: Machine Learning on Healthcare Data
+
+- **ID:** `2601.07685` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Shafiul Ajam Opee, Nafiz Fahad, Anik Sen, et al.
+- **abs:** https://arxiv.org/abs/2601.07685 · **PDF:** https://arxiv.org/pdf/2601.07685v1
+
+## 22317. AnyECG: Evolved ECG Foundation Model for Holistic Health Profiling
+
+- **ID:** `2601.10748` · **primary:** `eess.SP` · **Submitted:** 2026-01-12
+- **Authors:** Jun Li, Hongling Zhu, Yujie Xiao, et al.
+- **abs:** https://arxiv.org/abs/2601.10748 · **PDF:** https://arxiv.org/pdf/2601.10748v1
+
+## 22318. Tab-TRM: Tiny Recursive Model for Insurance Pricing on Tabular Data
+
+- **ID:** `2601.07675` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Kishan Padayachy, Ronald Richman, Mario V. Wüthrich
+- **abs:** https://arxiv.org/abs/2601.07675 · **PDF:** https://arxiv.org/pdf/2601.07675v1
+
+## 22319. Self-Creating Random Walks for Decentralized Learning under Pac-Man Attacks
+
+- **ID:** `2601.07674` · **primary:** `cs.MA` · **Submitted:** 2026-01-12
+- **Authors:** Xingran Chen, Parimal Parag, Rohit Bhagat, et al.
+- **abs:** https://arxiv.org/abs/2601.07674 · **PDF:** https://arxiv.org/pdf/2601.07674v1
+
+## 22320. SAGE: Tool-Augmented LLM Task Solving Strategies in Scalable Multi-Agent Environments
+
+- **ID:** `2601.09750` · **primary:** `cs.SE` · **Submitted:** 2026-01-12
+- **Authors:** Robert K. Strehlow, Tobias Küster, Oskar F. Kupke, et al.
+- **abs:** https://arxiv.org/abs/2601.09750 · **PDF:** https://arxiv.org/pdf/2601.09750v1
+
+## 22321. Adaptive Layer Selection for Layer-Wise Token Pruning in LLM Inference
+
+- **ID:** `2601.07667` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Rei Taniguchi, Yuyang Dong, Makoto Onizuka, et al.
+- **abs:** https://arxiv.org/abs/2601.07667 · **PDF:** https://arxiv.org/pdf/2601.07667v1
+
+## 22322. Learning to accelerate Krasnosel'skii-Mann fixed-point iterations with guarantees
+
+- **ID:** `2601.07665` · **primary:** `eess.SY` · **Submitted:** 2026-01-12
+- **Authors:** Andrea Martin, Giuseppe Belgioioso
+- **abs:** https://arxiv.org/abs/2601.07665 · **PDF:** https://arxiv.org/pdf/2601.07665v1
+
+## 22323. Variational Contrastive Learning for Skeleton-based Action Recognition
+
+- **ID:** `2601.07666` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Dang Dinh Nguyen, Decky Aspandi Latif, Titus Zaharia
+- **abs:** https://arxiv.org/abs/2601.07666 · **PDF:** https://arxiv.org/pdf/2601.07666v1
+
+## 22324. Reasoning Models Will Blatantly Lie About Their Reasoning
+
+- **ID:** `2601.07663` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** William Walden
+- **abs:** https://arxiv.org/abs/2601.07663 · **PDF:** https://arxiv.org/pdf/2601.07663v2
+
+## 22325. Towards Automating Blockchain Consensus Verification with IsabeLLM
+
+- **ID:** `2601.07654` · **primary:** `cs.CR` · **Submitted:** 2026-01-12
+- **Authors:** Elliot Jones, William Knottenbelt
+- **abs:** https://arxiv.org/abs/2601.07654 · **PDF:** https://arxiv.org/pdf/2601.07654v1
+
+## 22326. Active Evaluation of General Agents: Problem Definition and Comparison of Baseline Algorithms
+
+- **ID:** `2601.07651` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Marc Lanctot, Kate Larson, Ian Gemp, et al.
+- **abs:** https://arxiv.org/abs/2601.07651 · **PDF:** https://arxiv.org/pdf/2601.07651v2
+
+## 22327. Order in the Evaluation Court: A Critical Analysis of NLG Evaluation Trends
+
+- **ID:** `2601.07648` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Jing Yang, Nils Feldhus, Salar Mohtaj, et al.
+- **abs:** https://arxiv.org/abs/2601.07648 · **PDF:** https://arxiv.org/pdf/2601.07648v1
+
+## 22328. Studying the Role of Synthetic Data for Machine Learning-based Wireless Networks Traffic Forecasting
+
+- **ID:** `2601.07646` · **primary:** `eess.SY` · **Submitted:** 2026-01-12
+- **Authors:** José Pulido, Francesc Wilhelmi, Sergio Fortes, et al.
+- **abs:** https://arxiv.org/abs/2601.07646 · **PDF:** https://arxiv.org/pdf/2601.07646v2
+
+## 22329. PlaM: Training-Free Plateau-Guided Model Merging for Better Visual Grounding in MLLMs
+
+- **ID:** `2601.07645` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Zijing Wang, Yongkang Liu, Mingyang Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.07645 · **PDF:** https://arxiv.org/pdf/2601.07645v1
+
+## 22330. Beyond Static Tools: Test-Time Tool Evolution for Scientific Reasoning
+
+- **ID:** `2601.07641` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Jiaxuan Lu, Ziyu Kong, Yemin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.07641 · **PDF:** https://arxiv.org/pdf/2601.07641v1
+
+## 22331. Dual-Level Models for Physics-Informed Multi-Step Time Series Forecasting
+
+- **ID:** `2601.07640` · **primary:** `stat.ML` · **Submitted:** 2026-01-12
+- **Authors:** Mahdi Nasiri, Johanna Kortelainen, Simo Särkkä
+- **abs:** https://arxiv.org/abs/2601.07640 · **PDF:** https://arxiv.org/pdf/2601.07640v1
+
+## 22332. SALT-KG: A Benchmark for Semantics-Aware Learning on Enterprise Tables
+
+- **ID:** `2601.07638` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Isaiah Onando Mulang, Felix Sasaki, Tassilo Klein, et al.
+- **abs:** https://arxiv.org/abs/2601.07638 · **PDF:** https://arxiv.org/pdf/2601.07638v1
+
+## 22333. Reinforcement Learning for Micro-Level Claims Reserving
+
+- **ID:** `2601.07637` · **primary:** `q-fin.RM` · **Submitted:** 2026-01-12
+- **Authors:** Benjamin Avanzi, Ronald Richman, Bernard Wong, et al.
+- **abs:** https://arxiv.org/abs/2601.07637 · **PDF:** https://arxiv.org/pdf/2601.07637v1
+
+## 22334. Beyond Sharpness: A Flatness Decomposition Framework for Efficient Continual Learning
+
+- **ID:** `2601.07636` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Yanan Chen, Tieliang Gong, Yunjiao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.07636 · **PDF:** https://arxiv.org/pdf/2601.07636v1
+
+## 22335. Learning About Learning: A Physics Path from Spin Glasses to Artificial Intelligence
+
+- **ID:** `2601.07635` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-01-12
+- **Authors:** Denis D. Caprioti, Matheus Haas, Constantino F. Vasconcelos, et al.
+- **abs:** https://arxiv.org/abs/2601.07635 · **PDF:** https://arxiv.org/pdf/2601.07635v2
+
+## 22336. GeoMotionGPT: Geometry-Aligned Motion Understanding with Large Language Models
+
+- **ID:** `2601.07632` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Zhankai Ye, Bofan Li, Yukai Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.07632 · **PDF:** https://arxiv.org/pdf/2601.07632v4
+
+## 22337. Integrating Machine-Generated Short Descriptions into the Wikipedia Android App: A Pilot Deployment of Descartes
+
+- **ID:** `2601.07631` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Marija Šakota, Dmitry Brant, Cooltey Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.07631 · **PDF:** https://arxiv.org/pdf/2601.07631v1
+
+## 22338. Neural Architecture for Fast and Reliable Coagulation Assessment in Clinical Settings: Leveraging Thromboelastography
+
+- **ID:** `2601.07618` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Yulu Wang, Ziqian Zeng, Jianjun Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.07618 · **PDF:** https://arxiv.org/pdf/2601.07618v1
+
+## 22339. DIAGPaper: Diagnosing Valid and Specific Weaknesses in Scientific Papers via Multi-Agent Reasoning
+
+- **ID:** `2601.07611` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Zhuoyang Zou, Abolfazl Ansari, Delvin Ce Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.07611 · **PDF:** https://arxiv.org/pdf/2601.07611v2
+
+## 22340. Proof of Time: A Benchmark for Evaluating Scientific Idea Judgments
+
+- **ID:** `2601.07606` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Bingyang Ye, Shan Chen, Jingxuan Tu, et al.
+- **abs:** https://arxiv.org/abs/2601.07606 · **PDF:** https://arxiv.org/pdf/2601.07606v1
+
+## 22341. Enhancing Large Language Models for Time-Series Forecasting via Vector-Injected In-Context Learning
+
+- **ID:** `2601.07903` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Jianqi Zhang, Jingyao Wang, Wenwen Qiang, et al.
+- **abs:** https://arxiv.org/abs/2601.07903 · **PDF:** https://arxiv.org/pdf/2601.07903v3
+
+## 22342. TAG-MoE: Task-Aware Gating for Unified Generative Mixture-of-Experts
+
+- **ID:** `2601.08881` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Yu Xu, Hongbin Yan, Juan Cao, et al.
+- **abs:** https://arxiv.org/abs/2601.08881 · **PDF:** https://arxiv.org/pdf/2601.08881v2
+
+## 22343. Pheromone-Focused Ant Colony Optimization algorithm for path planning
+
+- **ID:** `2601.07597` · **primary:** `cs.NE` · **Submitted:** 2026-01-12
+- **Authors:** Yi Liu, Hongda Zhang, Zhongxue Gan, et al.
+- **abs:** https://arxiv.org/abs/2601.07597 · **PDF:** https://arxiv.org/pdf/2601.07597v1
+
+## 22344. GRPO with State Mutations: Improving LLM-Based Hardware Test Plan Generation
+
+- **ID:** `2601.07593` · **primary:** `cs.AR` · **Submitted:** 2026-01-12
+- **Authors:** Dimple Vijay Kochar, Nathaniel Pinckney, Guan-Ting Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.07593 · **PDF:** https://arxiv.org/pdf/2601.07593v1
+
+## 22345. Temporal-Aligned Meta-Learning for Risk Management: A Stacking Approach for Multi-Source Credit Scoring
+
+- **ID:** `2601.07588` · **primary:** `q-fin.RM` · **Submitted:** 2026-01-12
+- **Authors:** O. Didkovskyi, A. Vidali, N. Jean, et al.
+- **abs:** https://arxiv.org/abs/2601.07588 · **PDF:** https://arxiv.org/pdf/2601.07588v1
+
+## 22346. Machine learning nonequilibrium phase transitions in charge-density wave insulators
+
+- **ID:** `2601.07583` · **primary:** `cond-mat.str-el` · **Submitted:** 2026-01-12
+- **Authors:** Yunhao Fan, Sheng Zhang, Gia-Wei Chern
+- **abs:** https://arxiv.org/abs/2601.07583 · **PDF:** https://arxiv.org/pdf/2601.07583v1
+
+## 22347. ES-Mem: Event Segmentation-Based Memory for Long-Term Dialogue Agents
+
+- **ID:** `2601.07582` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Huhai Zou, Tianhao Sun, Chuanjiang He, et al.
+- **abs:** https://arxiv.org/abs/2601.07582 · **PDF:** https://arxiv.org/pdf/2601.07582v2
+
+## 22348. Large Language Models for Physics Instrument Design
+
+- **ID:** `2601.07580` · **primary:** `physics.ins-det` · **Submitted:** 2026-01-12
+- **Authors:** Sara Zoccheddu, Shah Rukh Qasim, Patrick Owen, et al.
+- **abs:** https://arxiv.org/abs/2601.07580 · **PDF:** https://arxiv.org/pdf/2601.07580v1
+
+## 22349. An adjoint method for training data-driven reduced-order models
+
+- **ID:** `2601.07579` · **primary:** `cs.CE` · **Submitted:** 2026-01-12
+- **Authors:** Donglin Liu, Francisco García Atienza, Mengwu Guo
+- **abs:** https://arxiv.org/abs/2601.07579 · **PDF:** https://arxiv.org/pdf/2601.07579v1
+
+## 22350. Beyond Entangled Planning: Task-Decoupled Planning for Long-Horizon Agents
+
+- **ID:** `2601.07577` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Yunfan Li, Bingbing Xu, Xueyun Tian, et al.
+- **abs:** https://arxiv.org/abs/2601.07577 · **PDF:** https://arxiv.org/pdf/2601.07577v1
+
+## 22351. A Model of Artificial Jagged Intelligence
+
+- **ID:** `2601.07573` · **primary:** `econ.TH` · **Submitted:** 2026-01-12
+- **Authors:** Joshua Gans
+- **abs:** https://arxiv.org/abs/2601.07573 · **PDF:** https://arxiv.org/pdf/2601.07573v1
+
+## 22352. d3LLM: Ultra-Fast Diffusion LLM using Pseudo-Trajectory Distillation
+
+- **ID:** `2601.07568` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Yu-Yang Qian, Junda Su, Lanxiang Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.07568 · **PDF:** https://arxiv.org/pdf/2601.07568v2
+
+## 22353. A Unified Framework for Emotion Recognition and Sentiment Analysis via Expert-Guided Multimodal Fusion with Large Language Models
+
+- **ID:** `2601.07565` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Jiaqi Qiao, Xiujuan Xu, Xinran Li, et al.
+- **abs:** https://arxiv.org/abs/2601.07565 · **PDF:** https://arxiv.org/pdf/2601.07565v1
+
+## 22354. Backpropagation-Free Test-Time Adaptation for Lightweight EEG-Based Brain-Computer Interfaces
+
+- **ID:** `2601.07556` · **primary:** `cs.HC` · **Submitted:** 2026-01-12
+- **Authors:** Siyang Li, Jiayi Ouyang, Zhenyao Cui, et al.
+- **abs:** https://arxiv.org/abs/2601.07556 · **PDF:** https://arxiv.org/pdf/2601.07556v1
+
+## 22355. VirtualEnv: A Platform for Embodied AI Research
+
+- **ID:** `2601.07553` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Kabir Swain, Sijie Han, Ayush Raina, et al.
+- **abs:** https://arxiv.org/abs/2601.07553 · **PDF:** https://arxiv.org/pdf/2601.07553v2
+
+## 22356. TFEC: Multivariate Time-Series Clustering via Temporal-Frequency Enhanced Contrastive Learning
+
+- **ID:** `2601.07550` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Zexi Tan, Tao Xie, Haoyi Xiao, et al.
+- **abs:** https://arxiv.org/abs/2601.07550 · **PDF:** https://arxiv.org/pdf/2601.07550v1
+
+## 22357. Contextual Discrepancy-Aware Contrastive Learning for Robust Medical Time Series Diagnosis in Small-Sample Scenarios
+
+- **ID:** `2601.07548` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Kaito Tanaka, Aya Nakayama, Masato Ito, et al.
+- **abs:** https://arxiv.org/abs/2601.07548 · **PDF:** https://arxiv.org/pdf/2601.07548v1
+
+## 22358. Near-Optimal Private Linear Regression via Iterative Hessian Mixing
+
+- **ID:** `2601.07545` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Omri Lev, Moshe Shenfeld, Vishwak Srinivasan, et al.
+- **abs:** https://arxiv.org/abs/2601.07545 · **PDF:** https://arxiv.org/pdf/2601.07545v1
+
+## 22359. Trust, Safety, and Accuracy: Assessing LLMs for Routine Maternity Advice
+
+- **ID:** `2603.16872` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** V Sai Divya, A Bhanusree, Rimjhim, et al.
+- **abs:** https://arxiv.org/abs/2603.16872 · **PDF:** https://arxiv.org/pdf/2603.16872v1
+
+## 22360. Nonparametric Kernel Clustering with Bandit Feedback
+
+- **ID:** `2601.07535` · **primary:** `stat.ML` · **Submitted:** 2026-01-12
+- **Authors:** Victor Thuot, Sebastian Vogt, Debarghya Ghoshdastidar, et al.
+- **abs:** https://arxiv.org/abs/2601.07535 · **PDF:** https://arxiv.org/pdf/2601.07535v1
+
+## 22361. Loci Similes: A Benchmark for Extracting Intertextualities in Latin Literature
+
+- **ID:** `2601.07533` · **primary:** `cs.IR` · **Submitted:** 2026-01-12
+- **Authors:** Julian Schelb, Michael Wittweiler, Marie Revellio, et al.
+- **abs:** https://arxiv.org/abs/2601.07533 · **PDF:** https://arxiv.org/pdf/2601.07533v1
+
+## 22362. From RAG to Agentic RAG for Faithful Islamic Question Answering
+
+- **ID:** `2601.07528` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Gagan Bhatia, Hamdy Mubarak, Mustafa Jarrar, et al.
+- **abs:** https://arxiv.org/abs/2601.07528 · **PDF:** https://arxiv.org/pdf/2601.07528v1
+
+## 22363. Thinking Before Constraining: A Unified Decoding Framework for Large Language Models
+
+- **ID:** `2601.07525` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Ngoc Trinh Hung Nguyen, Alonso Silva, Laith Zumot, et al.
+- **abs:** https://arxiv.org/abs/2601.07525 · **PDF:** https://arxiv.org/pdf/2601.07525v1
+
+## 22364. Stagewise Reinforcement Learning and the Geometry of the Regret Landscape
+
+- **ID:** `2601.07524` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Chris Elliott, Einar Urdshals, David Quarel, et al.
+- **abs:** https://arxiv.org/abs/2601.07524 · **PDF:** https://arxiv.org/pdf/2601.07524v2
+
+## 22365. Mon3tr: Monocular 3D Telepresence with Pre-built Gaussian Avatars as Amortization
+
+- **ID:** `2601.07518` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Fangyu Lin, Yingdong Hu, Zhening Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.07518 · **PDF:** https://arxiv.org/pdf/2601.07518v1
+
+## 22366. Controlling Multimodal Conversational Agents with Coverage-Enhanced Latent Actions
+
+- **ID:** `2601.07516` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Yongqi Li, Hao Lang, Tieyun Qian, et al.
+- **abs:** https://arxiv.org/abs/2601.07516 · **PDF:** https://arxiv.org/pdf/2601.07516v1
+
+## 22367. Data-Driven Stochastic VRP: Integration of Forecast Duration into Optimization for Utility Workforce Management
+
+- **ID:** `2601.07514` · **primary:** `math.OC` · **Submitted:** 2026-01-12
+- **Authors:** Matteo Garbelli
+- **abs:** https://arxiv.org/abs/2601.07514 · **PDF:** https://arxiv.org/pdf/2601.07514v1
+
+## 22368. Land-then-transport: A Flow Matching-Based Generative Decoder for Wireless Image Transmission
+
+- **ID:** `2601.07512` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Jingwen Fu, Ming Xiao, Mikael Skoglund, et al.
+- **abs:** https://arxiv.org/abs/2601.07512 · **PDF:** https://arxiv.org/pdf/2601.07512v1
+
+## 22369. High-Rank Structured Modulation for Parameter-Efficient Fine-Tuning
+
+- **ID:** `2601.07507` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Yongkang Liu, Xing Li, Mengjie Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.07507 · **PDF:** https://arxiv.org/pdf/2601.07507v1
+
+## 22370. Judging Against the Reference: Uncovering Knowledge-Driven Failures in LLM-Judges on QA Evaluation
+
+- **ID:** `2601.07506` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Dongryeol Lee, Yerin Hwang, Taegwan Kang, et al.
+- **abs:** https://arxiv.org/abs/2601.07506 · **PDF:** https://arxiv.org/pdf/2601.07506v1
+
+## 22371. FROAV: A Framework for RAG Observation and Agent Verification -- Lowering the Barrier to LLM Agent Research
+
+- **ID:** `2601.07504` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Tzu-Hsuan Lin, Chih-Hsuan Kao
+- **abs:** https://arxiv.org/abs/2601.07504 · **PDF:** https://arxiv.org/pdf/2601.07504v1
+
+## 22372. Decentralized Online Convex Optimization with Unknown Feedback Delays
+
+- **ID:** `2601.07901` · **primary:** `stat.ML` · **Submitted:** 2026-01-12
+- **Authors:** Hao Qiu, Mengxiao Zhang, Juliette Achddou
+- **abs:** https://arxiv.org/abs/2601.07901 · **PDF:** https://arxiv.org/pdf/2601.07901v1
+
+## 22373. Graph Inference Towards ICD Coding
+
+- **ID:** `2601.07496` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Xiaoxiao Deng
+- **abs:** https://arxiv.org/abs/2601.07496 · **PDF:** https://arxiv.org/pdf/2601.07496v1
+
+## 22374. Online Markov Decision Processes with Terminal Law Constraints
+
+- **ID:** `2601.07492` · **primary:** `math.OC` · **Submitted:** 2026-01-12
+- **Authors:** Bianca Marin Moreno, Margaux Brégère, Pierre Gaillard, et al.
+- **abs:** https://arxiv.org/abs/2601.07492 · **PDF:** https://arxiv.org/pdf/2601.07492v1
+
+## 22375. The Secretary Problem with Predictions and a Chosen Order
+
+- **ID:** `2601.07482` · **primary:** `cs.DS` · **Submitted:** 2026-01-12
+- **Authors:** Helia Karisani, Mohammadreza Daneshvaramoli, Hedyeh Beyhaghi, et al.
+- **abs:** https://arxiv.org/abs/2601.07482 · **PDF:** https://arxiv.org/pdf/2601.07482v1
+
+## 22376. From "Help" to Helpful: A Hierarchical Assessment of LLMs in Mental e-Health Applications
+
+- **ID:** `2602.18443` · **primary:** `cs.HC` · **Submitted:** 2026-01-12
+- **Authors:** Philipp Steigerwald, Jens Albrecht
+- **abs:** https://arxiv.org/abs/2602.18443 · **PDF:** https://arxiv.org/pdf/2602.18443v1
+
+## 22377. JudgeFlow: Agentic Workflow Optimization via Block Judge
+
+- **ID:** `2601.07477` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Zihan Ma, Zhikai Zhao, Chuanbo Hua, et al.
+- **abs:** https://arxiv.org/abs/2601.07477 · **PDF:** https://arxiv.org/pdf/2601.07477v2
+
+## 22378. Large Language Models and Algorithm Execution: Application to an Arithmetic Function
+
+- **ID:** `2601.07898` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Farah Ben Slama, Frédéric Armetta
+- **abs:** https://arxiv.org/abs/2601.07898 · **PDF:** https://arxiv.org/pdf/2601.07898v1
+
+## 22379. ARCQuant: Boosting NVFP4 Quantization with Augmented Residual Channels for LLMs
+
+- **ID:** `2601.07475` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Haoqian Meng, Yilun Luo, Yafei Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.07475 · **PDF:** https://arxiv.org/pdf/2601.07475v1
+
+## 22380. Task Prototype-Based Knowledge Retrieval for Multi-Task Learning from Partially Annotated Data
+
+- **ID:** `2601.07474` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Youngmin Oh, Hyung-Il Kim, Jung Uk Kim
+- **abs:** https://arxiv.org/abs/2601.07474 · **PDF:** https://arxiv.org/pdf/2601.07474v1
+
+## 22381. AntiPaSTO: Self-Supervised Honesty Steering via Anti-Parallel Representations
+
+- **ID:** `2601.07473` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Michael J. Clark
+- **abs:** https://arxiv.org/abs/2601.07473 · **PDF:** https://arxiv.org/pdf/2601.07473v3
+
+## 22382. Learning How to Remember: A Meta-Cognitive Management Method for Structured and Transferable Agent Memory
+
+- **ID:** `2601.07470` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Sirui Liang, Pengfei Cao, Jian Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.07470 · **PDF:** https://arxiv.org/pdf/2601.07470v1
+
+## 22383. Knowledge Distillation for LLM-Based Human Activity Recognition in Homes
+
+- **ID:** `2601.07469` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Julien Cumin, Oussama Er-Rahmany, Xi Chen
+- **abs:** https://arxiv.org/abs/2601.07469 · **PDF:** https://arxiv.org/pdf/2601.07469v1
+
+## 22384. Beyond Dialogue Time: Temporal Semantic Memory for Personalized LLM Agents
+
+- **ID:** `2601.07468` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Miao Su, Yucan Guo, Zhongni Hou, et al.
+- **abs:** https://arxiv.org/abs/2601.07468 · **PDF:** https://arxiv.org/pdf/2601.07468v1
+
+## 22385. IFDNS: An Iterative Feedback-Driven Neuro-Symbolic Method for Faithful Logical Reasoning
+
+- **ID:** `2601.07464` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Xiaoheng Wang, Tongxuan Liu, Zi Gong, et al.
+- **abs:** https://arxiv.org/abs/2601.07464 · **PDF:** https://arxiv.org/pdf/2601.07464v1
+
+## 22386. Puzzle it Out: Local-to-Global World Model for Offline Multi-Agent Reinforcement Learning
+
+- **ID:** `2601.07463` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Sijia Li, Xinran Li, Shibo Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.07463 · **PDF:** https://arxiv.org/pdf/2601.07463v2
+
+## 22387. R-LAM: Reproducibility-Constrained Large Action Models for Scientific Workflow Automation
+
+- **ID:** `2601.09749` · **primary:** `cs.SE` · **Submitted:** 2026-01-12
+- **Authors:** Suriya Sureshkumar
+- **abs:** https://arxiv.org/abs/2601.09749 · **PDF:** https://arxiv.org/pdf/2601.09749v1
+
+## 22388. Improving Video Question Answering through query-based frame selection
+
+- **ID:** `2601.07459` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Himanshu Patil, Geo Jolly, Ramana Raja Buddala, et al.
+- **abs:** https://arxiv.org/abs/2601.07459 · **PDF:** https://arxiv.org/pdf/2601.07459v1
+
+## 22389. RLPO: Residual Listwise Preference Optimization for Long-Context Review Ranking
+
+- **ID:** `2601.07449` · **primary:** `cs.IR` · **Submitted:** 2026-01-12
+- **Authors:** Hao Jiang, Zhi Yang, Annan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.07449 · **PDF:** https://arxiv.org/pdf/2601.07449v1
+
+## 22390. Surrogate-based Optimization via Clustering for Box-Constrained Problems
+
+- **ID:** `2601.07442` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Maaz Ahmad, Iftekhar A. Karimi
+- **abs:** https://arxiv.org/abs/2601.07442 · **PDF:** https://arxiv.org/pdf/2601.07442v1
+
+## 22391. Variational Autoencoder with Normalizing flow for X-ray spectral fitting
+
+- **ID:** `2601.07440` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Fiona Redmen, Ethan Tregidga, James F. Steiner, et al.
+- **abs:** https://arxiv.org/abs/2601.07440 · **PDF:** https://arxiv.org/pdf/2601.07440v1
+
+## 22392. PIDT: Physics-Informed Digital Twin for Optical Fiber Parameter Estimation
+
+- **ID:** `2601.07436` · **primary:** `eess.SP` · **Submitted:** 2026-01-12
+- **Authors:** Zicong Jiang, Magnus Karlsson, Erik Agrell, et al.
+- **abs:** https://arxiv.org/abs/2601.07436 · **PDF:** https://arxiv.org/pdf/2601.07436v1
+
+## 22393. KALE: Enhancing Knowledge Manipulation in Large Language Models via Knowledge-aware Learning
+
+- **ID:** `2601.07430` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Qitan Lv, Tianyu Liu, Qiaosheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.07430 · **PDF:** https://arxiv.org/pdf/2601.07430v1
+
+## 22394. SAD: A Large-Scale Strategic Argumentative Dialogue Dataset
+
+- **ID:** `2601.07423` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Yongkang Liu, Jiayang Yu, Mingyang Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.07423 · **PDF:** https://arxiv.org/pdf/2601.07423v1
+
+## 22395. Two Pathways to Truthfulness: On the Intrinsic Encoding of LLM Hallucinations
+
+- **ID:** `2601.07422` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Wen Luo, Guangyue Peng, Wei Li, et al.
+- **abs:** https://arxiv.org/abs/2601.07422 · **PDF:** https://arxiv.org/pdf/2601.07422v1
+
+## 22396. Position: Don't be Afraid of Over-Smoothing And Over-Squashing
+
+- **ID:** `2601.07419` · **primary:** `stat.ML` · **Submitted:** 2026-01-12
+- **Authors:** Niklas Kormann, Benjamin Doerr, Johannes F. Lutzeyer
+- **abs:** https://arxiv.org/abs/2601.07419 · **PDF:** https://arxiv.org/pdf/2601.07419v1
+
+## 22397. PLANET v2.0: A comprehensive Protein-Ligand Affinity Prediction Model Based on Mixture Density Network
+
+- **ID:** `2601.07415` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Haotian Gao, Xiangying Zhang, Jingyuan Li, et al.
+- **abs:** https://arxiv.org/abs/2601.07415 · **PDF:** https://arxiv.org/pdf/2601.07415v1
+
+## 22398. The Practicality of Normalizing Flow Test-Time Training in Bayesian Inference for Agent-Based Models
+
+- **ID:** `2601.07413` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Junyao Zhang, Jinglai Li, Junqi Tang
+- **abs:** https://arxiv.org/abs/2601.07413 · **PDF:** https://arxiv.org/pdf/2601.07413v1
+
+## 22399. SCALPEL: Selective Capability Ablation via Low-rank Parameter Editing for Large Language Model Interpretability Analysis
+
+- **ID:** `2601.07411` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Zihao Fu, Xufeng Duan, Zhenguang G. Cai
+- **abs:** https://arxiv.org/abs/2601.07411 · **PDF:** https://arxiv.org/pdf/2601.07411v1
+
+## 22400. Outcome-Grounded Advantage Reshaping for Fine-Grained Credit Assignment in Mathematical Reasoning
+
+- **ID:** `2601.07408` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Ziheng Li, Liu Kang, Feng Xiao, et al.
+- **abs:** https://arxiv.org/abs/2601.07408 · **PDF:** https://arxiv.org/pdf/2601.07408v1
+
+## 22401. Sensor Placement for Urban Traffic Interpolation: A Data-Driven Evaluation to Inform Policy
+
+- **ID:** `2601.10747` · **primary:** `eess.SP` · **Submitted:** 2026-01-12
+- **Authors:** Silke K. Kaiser
+- **abs:** https://arxiv.org/abs/2601.10747 · **PDF:** https://arxiv.org/pdf/2601.10747v1
+
+## 22402. On Narrative: The Rhetorical Mechanisms of Online Polarisation
+
+- **ID:** `2601.07398` · **primary:** `cs.CY` · **Submitted:** 2026-01-12
+- **Authors:** Jan Elfes, Marco Bastos, Luca Maria Aiello
+- **abs:** https://arxiv.org/abs/2601.07398 · **PDF:** https://arxiv.org/pdf/2601.07398v1
+
+## 22403. Quantifying non deterministic drift in large language models
+
+- **ID:** `2601.19934` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Claire Nicholson
+- **abs:** https://arxiv.org/abs/2601.19934 · **PDF:** https://arxiv.org/pdf/2601.19934v1
+
+## 22404. Layerwise goal-oriented adaptivity for neural ODEs: an optimal control perspective
+
+- **ID:** `2601.07397` · **primary:** `math.OC` · **Submitted:** 2026-01-12
+- **Authors:** Michael Hintermüller, Michael Hinze, Denis Korolev
+- **abs:** https://arxiv.org/abs/2601.07397 · **PDF:** https://arxiv.org/pdf/2601.07397v1
+
+## 22405. MCP-ITP: An Automated Framework for Implicit Tool Poisoning in MCP
+
+- **ID:** `2601.07395` · **primary:** `cs.CR` · **Submitted:** 2026-01-12
+- **Authors:** Ruiqi Li, Zhiqiang Wang, Yunhao Yao, et al.
+- **abs:** https://arxiv.org/abs/2601.07395 · **PDF:** https://arxiv.org/pdf/2601.07395v1
+
+## 22406. Software-Hardware Co-optimization for Modular E2E AV Paradigm: A Unified Framework of Optimization Approaches, Simulation Environment and Evaluation Metrics
+
+- **ID:** `2601.07393` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Chengzhi Ji, Xingfeng Li, Zhaodong Lv, et al.
+- **abs:** https://arxiv.org/abs/2601.07393 · **PDF:** https://arxiv.org/pdf/2601.07393v1
+
+## 22407. OceanSAR-2: A Universal Feature Extractor for SAR Ocean Observation
+
+- **ID:** `2601.07392` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Alexandre Tuel, Thomas Kerdreux, Quentin Febvre, et al.
+- **abs:** https://arxiv.org/abs/2601.07392 · **PDF:** https://arxiv.org/pdf/2601.07392v1
+
+## 22408. On the Non-decoupling of Supervised Fine-tuning and Reinforcement Learning in Post-training
+
+- **ID:** `2601.07389` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Xueyan Niu, Bo Bai, Wei Han, et al.
+- **abs:** https://arxiv.org/abs/2601.07389 · **PDF:** https://arxiv.org/pdf/2601.07389v1
+
+## 22409. Computing patient similarity based on unstructured clinical notes
+
+- **ID:** `2601.07385` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Petr Zelina, Marko Řeháček, Jana Halámková, et al.
+- **abs:** https://arxiv.org/abs/2601.07385 · **PDF:** https://arxiv.org/pdf/2601.07385v1
+
+## 22410. CompNO: A Novel Foundation Model approach for solving Partial Differential Equations
+
+- **ID:** `2601.07384` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Hamda Hmida, Hsiu-Wen Chang Joly, Youssef Mesri
+- **abs:** https://arxiv.org/abs/2601.07384 · **PDF:** https://arxiv.org/pdf/2601.07384v1
+
+## 22411. Learning Dynamic Collaborative Network for Semi-supervised 3D Vessel Segmentation
+
+- **ID:** `2601.07377` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Jiao Xu, Xin Chen, Lihe Zhang
+- **abs:** https://arxiv.org/abs/2601.07377 · **PDF:** https://arxiv.org/pdf/2601.07377v1
+
+## 22412. OpenTinker: Separating Concerns in Agentic Reinforcement Learning
+
+- **ID:** `2601.07376` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Siqi Zhu, Jiaxuan You
+- **abs:** https://arxiv.org/abs/2601.07376 · **PDF:** https://arxiv.org/pdf/2601.07376v1
+
+## 22413. GROKE: Vision-Free Navigation Instruction Evaluation via Graph Reasoning on OpenStreetMap
+
+- **ID:** `2601.07375` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Farzad Shami, Subhrasankha Dey, Nico Van de Weghe, et al.
+- **abs:** https://arxiv.org/abs/2601.07375 · **PDF:** https://arxiv.org/pdf/2601.07375v1
+
+## 22414. Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models
+
+- **ID:** `2601.07372` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Xin Cheng, Wangding Zeng, Damai Dai, et al.
+- **abs:** https://arxiv.org/abs/2601.07372 · **PDF:** https://arxiv.org/pdf/2601.07372v1
+
+## 22415. Interpretable Text Classification Applied to the Detection of LLM-generated Creative Writing
+
+- **ID:** `2601.07368` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Minerva Suvanto, Andrea McGlinchey, Mattias Wahde, et al.
+- **abs:** https://arxiv.org/abs/2601.07368 · **PDF:** https://arxiv.org/pdf/2601.07368v1
+
+## 22416. On the universal definition of intelligence
+
+- **ID:** `2601.07364` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Joseph Chen
+- **abs:** https://arxiv.org/abs/2601.07364 · **PDF:** https://arxiv.org/pdf/2601.07364v1
+
+## 22417. Seeing Right but Saying Wrong: Inter- and Intra-Layer Refinement in MLLMs without Training
+
+- **ID:** `2601.07359` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Shezheng Song, Shasha Li, Jie Yu
+- **abs:** https://arxiv.org/abs/2601.07359 · **PDF:** https://arxiv.org/pdf/2601.07359v1
+
+## 22418. Efficient Convolutional Forward Model for Passive Acoustic Mapping and Temporal Monitoring
+
+- **ID:** `2601.07356` · **primary:** `eess.IV` · **Submitted:** 2026-01-12
+- **Authors:** Tatiana Gelvez-Barrera, Barbara Nicolas, Bruno Gilles, et al.
+- **abs:** https://arxiv.org/abs/2601.07356 · **PDF:** https://arxiv.org/pdf/2601.07356v1
+
+## 22419. Semantic Compression of LLM Instructions via Symbolic Metalanguages
+
+- **ID:** `2601.07354` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Ernst van Gassen
+- **abs:** https://arxiv.org/abs/2601.07354 · **PDF:** https://arxiv.org/pdf/2601.07354v1
+
+## 22420. TALON: Confidence-Aware Speculative Decoding with Adaptive Token Trees
+
+- **ID:** `2601.07353` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Tianyu Liu, Qitan Lv, Yuhao Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.07353 · **PDF:** https://arxiv.org/pdf/2601.07353v1
+
+## 22421. Beyond Hard Masks: Progressive Token Evolution for Diffusion Language Models
+
+- **ID:** `2601.07351` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Linhao Zhong, Linyu Wu, Bozhen Fang, et al.
+- **abs:** https://arxiv.org/abs/2601.07351 · **PDF:** https://arxiv.org/pdf/2601.07351v2
+
+## 22422. Reward Modeling from Natural Language Human Feedback
+
+- **ID:** `2601.07349` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Zongqi Wang, Rui Wang, Yuchuan Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.07349 · **PDF:** https://arxiv.org/pdf/2601.07349v2
+
+## 22423. Controlled Self-Evolution for Algorithmic Code Optimization
+
+- **ID:** `2601.07348` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Tu Hu, Ronghao Chen, Shuo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.07348 · **PDF:** https://arxiv.org/pdf/2601.07348v5
+
+## 22424. DiffER: Diffusion Entity-Relation Modeling for Reversal Curse in Diffusion Large Language Models
+
+- **ID:** `2601.07347` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Shaokai He, Kaiwen Wei, Xinyi Zeng, et al.
+- **abs:** https://arxiv.org/abs/2601.07347 · **PDF:** https://arxiv.org/pdf/2601.07347v1
+
+## 22425. EdgeNav-QE: QLoRA Quantization and Dynamic Early Exit for LAM-based Navigation on Edge Devices
+
+- **ID:** `2602.15836` · **primary:** `cs.RO` · **Submitted:** 2026-01-12
+- **Authors:** Mengyun Liu, Shanshan Huang, Jianan Jiang
+- **abs:** https://arxiv.org/abs/2602.15836 · **PDF:** https://arxiv.org/pdf/2602.15836v1
+
+## 22426. PulseMind: A Multi-Modal Medical Model for Real-World Clinical Diagnosis
+
+- **ID:** `2601.07344` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Jiao Xu, Junwei Liu, Jiangwei Lao, et al.
+- **abs:** https://arxiv.org/abs/2601.07344 · **PDF:** https://arxiv.org/pdf/2601.07344v1
+
+## 22427. Agentic Diagnostic Reasoning over Telecom and Datacenter Infrastructure
+
+- **ID:** `2601.07342` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Nicolas Tacheny
+- **abs:** https://arxiv.org/abs/2601.07342 · **PDF:** https://arxiv.org/pdf/2601.07342v1
+
+## 22428. Revealing the Attention Floating Mechanism in Masked Diffusion Models
+
+- **ID:** `2601.07894` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Xin Dai, Pengcheng Huang, Zhenghao Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.07894 · **PDF:** https://arxiv.org/pdf/2601.07894v1
+
+## 22429. Beyond Literal Mapping: Benchmarking and Improving Non-Literal Translation Evaluation
+
+- **ID:** `2601.07338` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Yanzhi Tian, Cunxiang Wang, Zeming Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.07338 · **PDF:** https://arxiv.org/pdf/2601.07338v1
+
+## 22430. SEE: Signal Embedding Energy for Quantifying Noise Interference in Large Audio Language Models
+
+- **ID:** `2601.07331` · **primary:** `cs.SD` · **Submitted:** 2026-01-12
+- **Authors:** Yuanhe Zhang, Jiayu Tian, Yibo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.07331 · **PDF:** https://arxiv.org/pdf/2601.07331v1
+
+## 22431. BayesRAG: Probabilistic Mutual Evidence Corroboration for Multimodal Retrieval-Augmented Generation
+
+- **ID:** `2601.07329` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Xuan Li, Yining Wang, Haocai Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.07329 · **PDF:** https://arxiv.org/pdf/2601.07329v1
+
+## 22432. How to predict creativity ratings from written narratives: A comparison of co-occurrence and textual forma mentis networks
+
+- **ID:** `2601.07327` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Roberto Passaro, Edith Haim, Massimo Stella
+- **abs:** https://arxiv.org/abs/2601.07327 · **PDF:** https://arxiv.org/pdf/2601.07327v1
+
+## 22433. Convergence Rate Analysis of the AdamW-Style Shampoo: Unifying One-sided and Two-Sided Preconditioning
+
+- **ID:** `2601.07326` · **primary:** `math.OC` · **Submitted:** 2026-01-12
+- **Authors:** Huan Li, Yiming Dong, Zhouchen Lin
+- **abs:** https://arxiv.org/abs/2601.07326 · **PDF:** https://arxiv.org/pdf/2601.07326v1
+
+## 22434. Sherry: Hardware-Efficient 1.25-Bit Ternary Quantization via Fine-grained Sparsification
+
+- **ID:** `2601.07892` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Hong Huang, Decheng Wu, Qiangqiang Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.07892 · **PDF:** https://arxiv.org/pdf/2601.07892v1
+
+## 22435. Robust Bayesian Inference via Variational Approximations of Generalized Rho-Posteriors
+
+- **ID:** `2601.07325` · **primary:** `stat.ML` · **Submitted:** 2026-01-12
+- **Authors:** EL Mahdi Khribch, Pierre Alquier
+- **abs:** https://arxiv.org/abs/2601.07325 · **PDF:** https://arxiv.org/pdf/2601.07325v2
+
+## 22436. Segmental Advantage Estimation: Enhancing PPO for Long-Context LLM Training
+
+- **ID:** `2601.07320` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Xue Gong, Qi Yi, Ziyuan Nan, et al.
+- **abs:** https://arxiv.org/abs/2601.07320 · **PDF:** https://arxiv.org/pdf/2601.07320v1
+
+## 22437. BEAT-Net: Injecting Biomimetic Spatio-Temporal Priors for Interpretable ECG Classification
+
+- **ID:** `2601.07316` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Runze Ma, Caizhi Liao
+- **abs:** https://arxiv.org/abs/2601.07316 · **PDF:** https://arxiv.org/pdf/2601.07316v1
+
+## 22438. VLM-CAD: VLM-Optimized Collaborative Agent Design Workflow for Analog Circuit Sizing
+
+- **ID:** `2601.07315` · **primary:** `cs.MA` · **Submitted:** 2026-01-12
+- **Authors:** Guanyuan Pan, Shuai Wang, Yugui Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.07315 · **PDF:** https://arxiv.org/pdf/2601.07315v4
+
+## 22439. Mitrasamgraha: A Comprehensive Classical Sanskrit Machine Translation Dataset
+
+- **ID:** `2601.07314` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Sebastian Nehrdich, David Allport, Sven Sellmer, et al.
+- **abs:** https://arxiv.org/abs/2601.07314 · **PDF:** https://arxiv.org/pdf/2601.07314v1
+
+## 22440. Explaining Machine Learning Predictive Models through Conditional Expectation Methods
+
+- **ID:** `2601.07313` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Silvia Ruiz-España, Laura Arnal, François Signol, et al.
+- **abs:** https://arxiv.org/abs/2601.07313 · **PDF:** https://arxiv.org/pdf/2601.07313v1
+
+## 22441. PsyCLIENT: Client Simulation via Conversational Trajectory Modeling for Trainee Practice and Model Evaluation in Mental Health Counseling
+
+- **ID:** `2601.07312` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Huachuan Qiu, Zhaoming Chen, Yuqian Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.07312 · **PDF:** https://arxiv.org/pdf/2601.07312v1
+
+## 22442. ARM: Role-Conditioned Neuron Transplantation for Training-Free Generalist LLM Agent Merging
+
+- **ID:** `2601.07309` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Zhuoka Feng, Kang Chen, Sihan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.07309 · **PDF:** https://arxiv.org/pdf/2601.07309v1
+
+## 22443. KVzap: Fast, Adaptive, and Faithful KV Cache Pruning
+
+- **ID:** `2601.07891` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Simon Jegou, Maximilian Jeblick
+- **abs:** https://arxiv.org/abs/2601.07891 · **PDF:** https://arxiv.org/pdf/2601.07891v2
+
+## 22444. Heterogeneous Multi-Expert Reinforcement Learning for Long-Horizon Multi-Goal Tasks in Autonomous Forklifts
+
+- **ID:** `2601.07304` · **primary:** `cs.RO` · **Submitted:** 2026-01-12
+- **Authors:** Yun Chen, Bowei Huang, Fan Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.07304 · **PDF:** https://arxiv.org/pdf/2601.07304v1
+
+## 22445. LRAS: Advanced Legal Reasoning with Agentic Search
+
+- **ID:** `2601.07296` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Yujin Zhou, Chuxue Cao, Jinluan Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.07296 · **PDF:** https://arxiv.org/pdf/2601.07296v1
+
+## 22446. NRR-Phi: Text-to-State Mapping for Ambiguity Preservation in LLM Inference
+
+- **ID:** `2601.19933` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Kei Saito
+- **abs:** https://arxiv.org/abs/2601.19933 · **PDF:** https://arxiv.org/pdf/2601.19933v5
+
+## 22447. "Newspaper Eat" Means "Not Tasty": A Taxonomy and Benchmark for Coded Languages in Real-World Chinese Online Reviews
+
+- **ID:** `2601.19932` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Ruyuan Wan, Changye Li, Ting-Hao 'Kenneth' Huang
+- **abs:** https://arxiv.org/abs/2601.19932 · **PDF:** https://arxiv.org/pdf/2601.19932v1
+
+## 22448. Photometric Redshift Estimation Using Scaled Ensemble Learning
+
+- **ID:** `2601.07292` · **primary:** `astro-ph.GA` · **Submitted:** 2026-01-12
+- **Authors:** Swagata Biswas, Shubhrangshu Ghosh, Avyarthana Ghosh, et al.
+- **abs:** https://arxiv.org/abs/2601.07292 · **PDF:** https://arxiv.org/pdf/2601.07292v1
+
+## 22449. A Visual Semantic Adaptive Watermark grounded by Prefix-Tuning for Large Vision-Language Model
+
+- **ID:** `2601.07291` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Qi Zheng, Shuliang Liu, Yu Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.07291 · **PDF:** https://arxiv.org/pdf/2601.07291v1
+
+## 22450. Kernel Alignment-based Multi-view Unsupervised Feature Selection with Sample-level Adaptive Graph Learning
+
+- **ID:** `2601.07288` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Yalan Tan, Yanyong Huang, Zongxin Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.07288 · **PDF:** https://arxiv.org/pdf/2601.07288v2
+
+## 22451. Does Visual Rendering Bypass Tokenization? Investigating Script-Tokenizer Misalignment in Pixel-Based Language Models
+
+- **ID:** `2602.06973` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Lucky Susanto, Musa Izzanardi Wijanarko, Khumaisa Nur'aini, et al.
+- **abs:** https://arxiv.org/abs/2602.06973 · **PDF:** https://arxiv.org/pdf/2602.06973v1
+
+## 22452. Minimum Wasserstein distance estimator under covariate shift: closed-form, super-efficiency and irregularity
+
+- **ID:** `2601.07282` · **primary:** `stat.ME` · **Submitted:** 2026-01-12
+- **Authors:** Junjun Lang, Qiong Zhang, Yukun Liu
+- **abs:** https://arxiv.org/abs/2601.07282 · **PDF:** https://arxiv.org/pdf/2601.07282v1
+
+## 22453. Covariance-Driven Regression Trees: Reducing Overfitting in CART
+
+- **ID:** `2601.07281` · **primary:** `stat.ML` · **Submitted:** 2026-01-12
+- **Authors:** Likun Zhang, Wei Ma
+- **abs:** https://arxiv.org/abs/2601.07281 · **PDF:** https://arxiv.org/pdf/2601.07281v1
+
+## 22454. ReasonTabQA: A Comprehensive Benchmark for Table Question Answering from Real World Industrial Scenarios
+
+- **ID:** `2601.07280` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Changzai Pan, Jie Zhang, Kaiwen Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.07280 · **PDF:** https://arxiv.org/pdf/2601.07280v1
+
+## 22455. A High-Recall Cost-Sensitive Machine Learning Framework for Real-Time Online Banking Transaction Fraud Detection
+
+- **ID:** `2601.07276` · **primary:** `cs.CR` · **Submitted:** 2026-01-12
+- **Authors:** Karthikeyan V. R., Premnath S., Kavinraaj S., et al.
+- **abs:** https://arxiv.org/abs/2601.07276 · **PDF:** https://arxiv.org/pdf/2601.07276v2
+
+## 22456. Towards Comprehensive Semantic Speech Embeddings for Chinese Dialects
+
+- **ID:** `2601.07274` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Kalvin Chang, Yiwen Shao, Jiahong Li, et al.
+- **abs:** https://arxiv.org/abs/2601.07274 · **PDF:** https://arxiv.org/pdf/2601.07274v1
+
+## 22457. Document-Level Zero-Shot Relation Extraction with Entity Side Information
+
+- **ID:** `2601.07271` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Mohan Raj Chanthran, Soon Lay Ki, Ong Huey Fang, et al.
+- **abs:** https://arxiv.org/abs/2601.07271 · **PDF:** https://arxiv.org/pdf/2601.07271v1
+
+## 22458. Learning Domain-Invariant Representations for Cross-Domain Image Registration via Scene-Appearance Disentanglement
+
+- **ID:** `2601.08875` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Jiahao Qin, Yiwen Wang
+- **abs:** https://arxiv.org/abs/2601.08875 · **PDF:** https://arxiv.org/pdf/2601.08875v2
+
+## 22459. The Confidence Dichotomy: Analyzing and Mitigating Miscalibration in Tool-Use Agents
+
+- **ID:** `2601.07264` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Weihao Xuan, Qingcheng Zeng, Heli Qi, et al.
+- **abs:** https://arxiv.org/abs/2601.07264 · **PDF:** https://arxiv.org/pdf/2601.07264v1
+
+## 22460. When Bots Take the Bait: Exposing and Mitigating the Emerging Social Engineering Attack in Web Automation Agent
+
+- **ID:** `2601.07263` · **primary:** `cs.CR` · **Submitted:** 2026-01-12
+- **Authors:** Xinyi Wu, Geng Hong, Yueyue Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.07263 · **PDF:** https://arxiv.org/pdf/2601.07263v1
+
+## 22461. The Illusion of Friendship: Why Generative AI Demands Unprecedented Ethical Vigilance
+
+- **ID:** `2601.08874` · **primary:** `cs.CY` · **Submitted:** 2026-01-12
+- **Authors:** Md Zahidul Islam
+- **abs:** https://arxiv.org/abs/2601.08874 · **PDF:** https://arxiv.org/pdf/2601.08874v1
+
+## 22462. Pseudodata-guided Invariant Representation Learning Boosts the Out-of-Distribution Generalization in Enzymatic Kinetic Parameter Prediction
+
+- **ID:** `2601.07261` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Haomin Wu, Zhiwei Nie, Hongyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.07261 · **PDF:** https://arxiv.org/pdf/2601.07261v1
+
+## 22463. ActiShade: Activating Overshadowed Knowledge to Guide Multi-Hop Reasoning in Large Language Models
+
+- **ID:** `2601.07260` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Huipeng Ma, Luan Zhang, Dandan Song, et al.
+- **abs:** https://arxiv.org/abs/2601.07260 · **PDF:** https://arxiv.org/pdf/2601.07260v1
+
+## 22464. Simulated Annealing-based Candidate Optimization for Batch Acquisition Functions
+
+- **ID:** `2601.07258` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Sk Md Ahnaf Akif Alvi, Raymundo Arróyave, Douglas Allaire
+- **abs:** https://arxiv.org/abs/2601.07258 · **PDF:** https://arxiv.org/pdf/2601.07258v1
+
+## 22465. Innovation Capacity of Dynamical Learning Systems
+
+- **ID:** `2601.07257` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Anthony M. Polloreno
+- **abs:** https://arxiv.org/abs/2601.07257 · **PDF:** https://arxiv.org/pdf/2601.07257v1
+
+## 22466. DDT: A Dual-Masking Dual-Expert Transformer for Energy Time-Series Forecasting
+
+- **ID:** `2601.07250` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Mingnan Zhu, Qixuan Zhang, Yixuan Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.07250 · **PDF:** https://arxiv.org/pdf/2601.07250v1
+
+## 22467. Multi-environment Invariance Learning with Missing Data
+
+- **ID:** `2601.07247` · **primary:** `stat.ML` · **Submitted:** 2026-01-12
+- **Authors:** Yiran Jia
+- **abs:** https://arxiv.org/abs/2601.07247 · **PDF:** https://arxiv.org/pdf/2601.07247v1
+
+## 22468. Learning to Trust the Crowd: A Multi-Model Consensus Reasoning Engine for Large Language Models
+
+- **ID:** `2601.07245` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Pranav Kallem
+- **abs:** https://arxiv.org/abs/2601.07245 · **PDF:** https://arxiv.org/pdf/2601.07245v1
+
+## 22469. Group Pattern Selection Optimization: Let LRMs Pick the Right Pattern for Reasoning
+
+- **ID:** `2601.07238` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Hanbin Wang, Jingwei Song, Jinpeng Li, et al.
+- **abs:** https://arxiv.org/abs/2601.07238 · **PDF:** https://arxiv.org/pdf/2601.07238v1
+
+## 22470. Stochastic CHAOS: Why Deterministic Inference Kills, and Distributional Variability Is the Heartbeat of Artifical Cognition
+
+- **ID:** `2601.07239` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Tanmay Joshi, Shourya Aggarwal, Anusa Saha, et al.
+- **abs:** https://arxiv.org/abs/2601.07239 · **PDF:** https://arxiv.org/pdf/2601.07239v1
+
+## 22471. Max-Min Neural Network Operators For Approximation of Multivariate Functions
+
+- **ID:** `2601.07886` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Abhishek Yadav, Uaday Singh, Feng Dai
+- **abs:** https://arxiv.org/abs/2601.07886 · **PDF:** https://arxiv.org/pdf/2601.07886v1
+
+## 22472. From "Thinking" to "Justifying": Aligning High-Stakes Explainability with Professional Communication Standards
+
+- **ID:** `2601.07233` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Chen Qian, Yimeng Wang, Yu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.07233 · **PDF:** https://arxiv.org/pdf/2601.07233v1
+
+## 22473. Yes FLoReNce, I Will Do Better Next Time! Agentic Feedback Reasoning for Humorous Meme Detection
+
+- **ID:** `2601.07232` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Olivia Shanhong Liu, Pai Chet Ng, De Wen Soh, et al.
+- **abs:** https://arxiv.org/abs/2601.07232 · **PDF:** https://arxiv.org/pdf/2601.07232v1
+
+## 22474. DiSCo: Making Absence Visible in Intelligent Summarization Interfaces
+
+- **ID:** `2601.07229` · **primary:** `cs.HC` · **Submitted:** 2026-01-12
+- **Authors:** Eran Fainman, Hagit Ben Shoshan, Adir Solomon, et al.
+- **abs:** https://arxiv.org/abs/2601.07229 · **PDF:** https://arxiv.org/pdf/2601.07229v2
+
+## 22475. Lost in the Noise: How Reasoning Models Fail with Contextual Distractors
+
+- **ID:** `2601.07226` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Seongyun Lee, Yongrae Jo, Minju Seo, et al.
+- **abs:** https://arxiv.org/abs/2601.07226 · **PDF:** https://arxiv.org/pdf/2601.07226v1
+
+## 22476. Consolidation or Adaptation? PRISM: Disentangling SFT and RL Data via Gradient Concentration
+
+- **ID:** `2601.07224` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Yang Zhao, Yangou Ouyang, Xiao Ding, et al.
+- **abs:** https://arxiv.org/abs/2601.07224 · **PDF:** https://arxiv.org/pdf/2601.07224v1
+
+## 22477. Small Symbols, Big Risks: Exploring Emoticon Semantic Confusion in Large Language Models
+
+- **ID:** `2601.07885` · **primary:** `cs.CR` · **Submitted:** 2026-01-12
+- **Authors:** Weipeng Jiang, Xiaoyu Zhang, Juan Zhai, et al.
+- **abs:** https://arxiv.org/abs/2601.07885 · **PDF:** https://arxiv.org/pdf/2601.07885v1
+
+## 22478. The Roots of Performance Disparity in Multilingual Language Models: Intrinsic Modeling Difficulty or Design Choices?
+
+- **ID:** `2601.07220` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Chen Shani, Yuval Reif, Nathan Roll, et al.
+- **abs:** https://arxiv.org/abs/2601.07220 · **PDF:** https://arxiv.org/pdf/2601.07220v2
+
+## 22479. BlindU: Blind Machine Unlearning without Revealing Erasing Data
+
+- **ID:** `2601.07214` · **primary:** `cs.CR` · **Submitted:** 2026-01-12
+- **Authors:** Weiqi Wang, Zhiyi Tian, Chenhan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.07214 · **PDF:** https://arxiv.org/pdf/2601.07214v1
+
+## 22480. MI-PRUN: Optimize Large Language Model Pruning via Mutual Information
+
+- **ID:** `2601.07212` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Hao Zhang, Zhibin Zhang, Guangxin Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.07212 · **PDF:** https://arxiv.org/pdf/2601.07212v1
+
+## 22481. SIRR-LMM: Single-image Reflection Removal via Large Multimodal Model
+
+- **ID:** `2601.07209` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Yu Guo, Zhiqiang Lao, Xiyun Song, et al.
+- **abs:** https://arxiv.org/abs/2601.07209 · **PDF:** https://arxiv.org/pdf/2601.07209v2
+
+## 22482. MAESTRO: Meta-learning Adaptive Estimation of Scalarization Trade-offs for Reward Optimization
+
+- **ID:** `2601.07208` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Yang Zhao, Hepeng Wang, Xiao Ding, et al.
+- **abs:** https://arxiv.org/abs/2601.07208 · **PDF:** https://arxiv.org/pdf/2601.07208v1
+
+## 22483. LLMRouterBench: A Massive Benchmark and Unified Framework for LLM Routing
+
+- **ID:** `2601.07206` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Hao Li, Yiqun Zhang, Zhaoyan Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.07206 · **PDF:** https://arxiv.org/pdf/2601.07206v1
+
+## 22484. Recontextualizing Famous Quotes for Brand Slogan Generation
+
+- **ID:** `2602.06049` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Ziao Yang, Zizhang Chen, Lei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2602.06049 · **PDF:** https://arxiv.org/pdf/2602.06049v1
+
+## 22485. CalPro: Prior-Aware Evidential--Conformal Prediction with Structure-Aware Guarantees for Protein Structures
+
+- **ID:** `2601.07201` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma
+- **abs:** https://arxiv.org/abs/2601.07201 · **PDF:** https://arxiv.org/pdf/2601.07201v1
+
+## 22486. Safeguarding LLM Fine-tuning via Push-Pull Distributional Alignment
+
+- **ID:** `2601.07200` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Haozhong Wang, Zhuo Li, Yibo Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.07200 · **PDF:** https://arxiv.org/pdf/2601.07200v1
+
+## 22487. Forward versus Backward: Comparing Reasoning Objectives in Direct Preference Optimization
+
+- **ID:** `2601.07199` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Murtaza Nikzad, Raghuram Ramanujan
+- **abs:** https://arxiv.org/abs/2601.07199 · **PDF:** https://arxiv.org/pdf/2601.07199v1
+
+## 22488. Beyond Variance: Knowledge-Aware LLM Compression via Fisher-Aligned Subspace Diagnostics
+
+- **ID:** `2601.07197` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma
+- **abs:** https://arxiv.org/abs/2601.07197 · **PDF:** https://arxiv.org/pdf/2601.07197v1
+
+## 22489. Relink: Constructing Query-Driven Evidence Graph On-the-Fly for GraphRAG
+
+- **ID:** `2601.07192` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Manzong Huang, Chenyang Bu, Yi He, et al.
+- **abs:** https://arxiv.org/abs/2601.07192 · **PDF:** https://arxiv.org/pdf/2601.07192v1
+
+## 22490. On Lie Groups Preserving Subspaces of Degenerate Clifford Algebras
+
+- **ID:** `2601.07191` · **primary:** `math.RA` · **Submitted:** 2026-01-12
+- **Authors:** E. R. Filimoshina, D. S. Shirokov
+- **abs:** https://arxiv.org/abs/2601.07191 · **PDF:** https://arxiv.org/pdf/2601.07191v1
+
+## 22491. Active Context Compression: Autonomous Memory Management in LLM Agents
+
+- **ID:** `2601.07190` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Nikhil Verma
+- **abs:** https://arxiv.org/abs/2601.07190 · **PDF:** https://arxiv.org/pdf/2601.07190v1
+
+## 22492. ForensicFormer: Hierarchical Multi-Scale Reasoning for Cross-Domain Image Forgery Detection
+
+- **ID:** `2601.08873` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Hema Hariharan Samson
+- **abs:** https://arxiv.org/abs/2601.08873 · **PDF:** https://arxiv.org/pdf/2601.08873v1
+
+## 22493. Standardization of Post-Publication Code Verification by Journals is Possible with the Support of the Community
+
+- **ID:** `2601.07189` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Susana Lopez-Moreno, Eric Dolores-Cuenca, Sangil Kim
+- **abs:** https://arxiv.org/abs/2601.07189 · **PDF:** https://arxiv.org/pdf/2601.07189v1
+
+## 22494. Defenses Against Prompt Attacks Learn Surface Heuristics
+
+- **ID:** `2601.07185` · **primary:** `cs.CR` · **Submitted:** 2026-01-12
+- **Authors:** Shawn Li, Chenxiao Yu, Zhiyu Ni, et al.
+- **abs:** https://arxiv.org/abs/2601.07185 · **PDF:** https://arxiv.org/pdf/2601.07185v1
+
+## 22495. PRPO: Aligning Process Reward with Outcome Reward in Policy Optimization
+
+- **ID:** `2601.07182` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Ruiyi Ding, Yongxuan Lv, Xianhui Meng, et al.
+- **abs:** https://arxiv.org/abs/2601.07182 · **PDF:** https://arxiv.org/pdf/2601.07182v3
+
+## 22496. Structured Reasoning for Large Language Models
+
+- **ID:** `2601.07180` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Jinyi Han, Zixiang Di, Zishang Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.07180 · **PDF:** https://arxiv.org/pdf/2601.07180v1
+
+## 22497. DIVER: Dynamic Iterative Visual Evidence Reasoning for Multimodal Fake News Detection
+
+- **ID:** `2601.07178` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Weilin Zhou, Zonghao Ying, Chunlei Meng, et al.
+- **abs:** https://arxiv.org/abs/2601.07178 · **PDF:** https://arxiv.org/pdf/2601.07178v1
+
+## 22498. Safe-FedLLM: Delving into the Safety of Federated Large Language Models
+
+- **ID:** `2601.07177` · **primary:** `cs.CR` · **Submitted:** 2026-01-12
+- **Authors:** Mingxiang Tao, Yu Tian, Wenxuan Tu, et al.
+- **abs:** https://arxiv.org/abs/2601.07177 · **PDF:** https://arxiv.org/pdf/2601.07177v1
+
+## 22499. G-MemLLM: Gated Latent Memory Augmentation for Long-Context Reasoning in Large Language Models
+
+- **ID:** `2602.00015` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Xun Xu
+- **abs:** https://arxiv.org/abs/2602.00015 · **PDF:** https://arxiv.org/pdf/2602.00015v1
+
+## 22500. Bridge-RAG: An Abstract Bridge Tree Based Retrieval Augmented Generation Algorithm With Cuckoo Filter
+
+- **ID:** `2603.26668` · **primary:** `cs.IR` · **Submitted:** 2026-01-12
+- **Authors:** Zihang Li, Wenjun Liu, Yikun Zong, et al.
+- **abs:** https://arxiv.org/abs/2603.26668 · **PDF:** https://arxiv.org/pdf/2603.26668v1
+
+## 22501. Offline Meta-Reinforcement Learning with Flow-Based Task Inference and Adaptive Correction of Feature Overgeneralization
+
+- **ID:** `2601.07164` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Min Wang, Xin Li, Mingzhong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.07164 · **PDF:** https://arxiv.org/pdf/2601.07164v1
+
+## 22502. AscendKernelGen: A Systematic Study of LLM-Based Kernel Generation for Neural Processing Units
+
+- **ID:** `2601.07160` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Xinzi Cao, Jianyang Zhai, Pengfei Li, et al.
+- **abs:** https://arxiv.org/abs/2601.07160 · **PDF:** https://arxiv.org/pdf/2601.07160v1
+
+## 22503. Stable On-Policy Distillation through Adaptive Target Reformulation
+
+- **ID:** `2601.07155` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Ijun Jang, Jewon Yeom, Juan Yeo, et al.
+- **abs:** https://arxiv.org/abs/2601.07155 · **PDF:** https://arxiv.org/pdf/2601.07155v1
+
+## 22504. Can Large Language Models Understand, Reason About, and Generate Code-Switched Text?
+
+- **ID:** `2601.07153` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Genta Indra Winata, David Anugraha, Patrick Amadeus Irawan, et al.
+- **abs:** https://arxiv.org/abs/2601.07153 · **PDF:** https://arxiv.org/pdf/2601.07153v1
+
+## 22505. Rewarding Creativity: A Human-Aligned Generative Reward Model for Reinforcement Learning in Storytelling
+
+- **ID:** `2601.07149` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Zhaoyan Li, Hang Lei, Yujia Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.07149 · **PDF:** https://arxiv.org/pdf/2601.07149v1
+
+## 22506. Measuring Iterative Temporal Reasoning with Time Puzzles
+
+- **ID:** `2601.07148` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Zhengxiang Wang, Zeyu Dong
+- **abs:** https://arxiv.org/abs/2601.07148 · **PDF:** https://arxiv.org/pdf/2601.07148v3
+
+## 22507. Generating readily synthesizable small molecule fluorophore scaffolds with reinforcement learning
+
+- **ID:** `2601.07145` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Ruhi Sayana, Kate Callon, Jennifer Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.07145 · **PDF:** https://arxiv.org/pdf/2601.07145v1
+
+## 22508. Ideological Isolation in Online Social Networks: A Survey of Computational Definitions, Metrics, and Mitigation Strategies
+
+- **ID:** `2601.07884` · **primary:** `cs.SI` · **Submitted:** 2026-01-12
+- **Authors:** Xiaodan Wang, Yanbin Liu, Shiqing Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.07884 · **PDF:** https://arxiv.org/pdf/2601.07884v1
+
+## 22509. Optimal Transport under Group Fairness Constraints
+
+- **ID:** `2601.07144` · **primary:** `stat.ML` · **Submitted:** 2026-01-12
+- **Authors:** Linus Bleistein, Mathieu Dagréou, Francisco Andrade, et al.
+- **abs:** https://arxiv.org/abs/2601.07144 · **PDF:** https://arxiv.org/pdf/2601.07144v2
+
+## 22510. A Large-Scale Study on the Development and Issues of Multi-Agent AI Systems
+
+- **ID:** `2601.07136` · **primary:** `cs.SE` · **Submitted:** 2026-01-12
+- **Authors:** Daniel Liu, Krishna Upadhyay, Vinaik Chhetri, et al.
+- **abs:** https://arxiv.org/abs/2601.07136 · **PDF:** https://arxiv.org/pdf/2601.07136v1
+
+## 22511. A Safety-Constrained Reinforcement Learning Framework for Reliable Wireless Autonomy
+
+- **ID:** `2602.13207` · **primary:** `cs.NI` · **Submitted:** 2026-01-12
+- **Authors:** Abdikarim Mohamed Ibrahim, Rosdiadee Nordin
+- **abs:** https://arxiv.org/abs/2602.13207 · **PDF:** https://arxiv.org/pdf/2602.13207v1
+
+## 22512. Proof of Reasoning for Privacy Enhanced Federated Blockchain Learning at the Edge
+
+- **ID:** `2601.07134` · **primary:** `cs.CR` · **Submitted:** 2026-01-12
+- **Authors:** James Calo, Benny Lo
+- **abs:** https://arxiv.org/abs/2601.07134 · **PDF:** https://arxiv.org/pdf/2601.07134v1
+
+## 22513. The Potential Impact of Neuromorphic Computing on Radio Telescope Observatories
+
+- **ID:** `2601.07130` · **primary:** `astro-ph.IM` · **Submitted:** 2026-01-12
+- **Authors:** Nicholas J. Pritchard, Richard Dodson, Andreas Wicenec
+- **abs:** https://arxiv.org/abs/2601.07130 · **PDF:** https://arxiv.org/pdf/2601.07130v1
+
+## 22514. ReinPool: Reinforcement Learning Pooling Multi-Vector Embeddings for Retrieval System
+
+- **ID:** `2601.07125` · **primary:** `cs.IR` · **Submitted:** 2026-01-12
+- **Authors:** Sungguk Cha, DongWook Kim, Mintae Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.07125 · **PDF:** https://arxiv.org/pdf/2601.07125v1
+
+## 22515. Semantic visually-guided acoustic highlighting with large vision-language models
+
+- **ID:** `2601.08871` · **primary:** `cs.SD` · **Submitted:** 2026-01-12
+- **Authors:** Junhua Huang, Chao Huang, Chenliang Xu
+- **abs:** https://arxiv.org/abs/2601.08871 · **PDF:** https://arxiv.org/pdf/2601.08871v1
+
+## 22516. Towards Automated Diagnosis of Inherited Arrhythmias: Combined Arrhythmia Classification Using Lead-Aware Spatial Attention Networks
+
+- **ID:** `2601.07124` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Sophie Sigfstead, River Jiang, Brianna Davies, et al.
+- **abs:** https://arxiv.org/abs/2601.07124 · **PDF:** https://arxiv.org/pdf/2601.07124v1
+
+## 22517. ENTRA: Entropy-Based Redundancy Avoidance in Large Language Model Reasoning
+
+- **ID:** `2601.07123` · **primary:** `cs.AI` · **Submitted:** 2026-01-12
+- **Authors:** Ruichu Cai, Haopeng Du, Qingwen Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.07123 · **PDF:** https://arxiv.org/pdf/2601.07123v1
+
+## 22518. Enhancing Cloud Network Resilience via a Robust LLM-Empowered Multi-Agent Reinforcement Learning Framework
+
+- **ID:** `2601.07122` · **primary:** `cs.CR` · **Submitted:** 2026-01-12
+- **Authors:** Yixiao Peng, Hao Hu, Feiyang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.07122 · **PDF:** https://arxiv.org/pdf/2601.07122v1
+
+## 22519. ReMIND: Orchestrating Modular Large Language Models for Controllable Serendipity A REM-Inspired System Design for Emergent Creative Ideation
+
+- **ID:** `2601.07121` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Makoto Sato
+- **abs:** https://arxiv.org/abs/2601.07121 · **PDF:** https://arxiv.org/pdf/2601.07121v1
+
+## 22520. Reward-Preserving Attacks For Robust Reinforcement Learning
+
+- **ID:** `2601.07118` · **primary:** `cs.LG` · **Submitted:** 2026-01-12
+- **Authors:** Lucas Schott, Elies Gherbi, Hatem Hajri, et al.
+- **abs:** https://arxiv.org/abs/2601.07118 · **PDF:** https://arxiv.org/pdf/2601.07118v2
+
+## 22521. Few-shot Class-Incremental Learning via Generative Co-Memory Regularization
+
+- **ID:** `2601.07117` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Kexin Bao, Yong Li, Dan Zeng, et al.
+- **abs:** https://arxiv.org/abs/2601.07117 · **PDF:** https://arxiv.org/pdf/2601.07117v1
+
+## 22522. CascadeMind at SemEval-2026 Task 4: A Hybrid Neuro-Symbolic Cascade for Narrative Similarity
+
+- **ID:** `2601.19931` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Sebastien Kawada, Dylan Holyoak
+- **abs:** https://arxiv.org/abs/2601.19931 · **PDF:** https://arxiv.org/pdf/2601.19931v2
+
+## 22523. The Need for a Socially-Grounded Persona Framework for User Simulation
+
+- **ID:** `2601.07110` · **primary:** `cs.CL` · **Submitted:** 2026-01-12
+- **Authors:** Pranav Narayanan Venkit, Yu Li, Yada Pruksachatkun, et al.
+- **abs:** https://arxiv.org/abs/2601.07110 · **PDF:** https://arxiv.org/pdf/2601.07110v2
+
+## 22524. MEDVISTAGYM: A Scalable Training Environment for Thinking with Medical Images via Tool-Integrated Reinforcement Learning
+
+- **ID:** `2601.07107` · **primary:** `cs.CV` · **Submitted:** 2026-01-12
+- **Authors:** Meng Lu, Yuxing Lu, Yuchen Zhuang, et al.
+- **abs:** https://arxiv.org/abs/2601.07107 · **PDF:** https://arxiv.org/pdf/2601.07107v1
+
+## 22525. Robust Bayesian Optimization via Tempered Posteriors
+
+- **ID:** `2601.07094` · **primary:** `stat.ME` · **Submitted:** 2026-01-11
+- **Authors:** Jiguang Li, Hengrui Luo
+- **abs:** https://arxiv.org/abs/2601.07094 · **PDF:** https://arxiv.org/pdf/2601.07094v1
+
+## 22526. Tackling Heterogeneity in Quantum Federated Learning: An Integrated Sporadic-Personalized Approach
+
+- **ID:** `2601.07882` · **primary:** `quant-ph` · **Submitted:** 2026-01-11
+- **Authors:** Ratun Rahman, Shaba Shaon, Dinh C. Nguyen
+- **abs:** https://arxiv.org/abs/2601.07882 · **PDF:** https://arxiv.org/pdf/2601.07882v1
+
+## 22527. 3D Wavelet-Based Structural Priors for Controlled Diffusion in Whole-Body Low-Dose PET Denoising
+
+- **ID:** `2601.07093` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Peiyuan Jing, Yue Yang, Chun-Wun Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.07093 · **PDF:** https://arxiv.org/pdf/2601.07093v4
+
+## 22528. When Should We Introduce Safety Interventions During Pretraining?
+
+- **ID:** `2601.07087` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Dylan Sam, Sachin Goyal, Pratyush Maini, et al.
+- **abs:** https://arxiv.org/abs/2601.07087 · **PDF:** https://arxiv.org/pdf/2601.07087v2
+
+## 22529. XBTorch: A Unified Framework for Modeling and Co-Design of Crossbar-Based Deep Learning Accelerators
+
+- **ID:** `2601.07086` · **primary:** `cs.ET` · **Submitted:** 2026-01-11
+- **Authors:** Osama Yousuf, Andreu L. Glasmann, Martin Lueker-Boden, et al.
+- **abs:** https://arxiv.org/abs/2601.07086 · **PDF:** https://arxiv.org/pdf/2601.07086v1
+
+## 22530. The AI Cognitive Trojan Horse: How Large Language Models May Bypass Human Epistemic Vigilance
+
+- **ID:** `2601.07085` · **primary:** `cs.HC` · **Submitted:** 2026-01-11
+- **Authors:** Andrew D. Maynard
+- **abs:** https://arxiv.org/abs/2601.07085 · **PDF:** https://arxiv.org/pdf/2601.07085v1
+
+## 22531. SDUs DAISY: A Benchmark for Danish Culture
+
+- **ID:** `2601.19930` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Jacob Nielsen, Stine L. Beltoft, Peter Schneider-Kamp, et al.
+- **abs:** https://arxiv.org/abs/2601.19930 · **PDF:** https://arxiv.org/pdf/2601.19930v1
+
+## 22532. Dynamical Systems Analysis Reveals Functional Regimes in Large Language Models
+
+- **ID:** `2601.11622` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Hassan Ugail, Newton Howard
+- **abs:** https://arxiv.org/abs/2601.11622 · **PDF:** https://arxiv.org/pdf/2601.11622v1
+
+## 22533. Robust Mean Estimation under Quantization
+
+- **ID:** `2601.07074` · **primary:** `stat.ML` · **Submitted:** 2026-01-11
+- **Authors:** Pedro Abdalla, Junren Chen
+- **abs:** https://arxiv.org/abs/2601.07074 · **PDF:** https://arxiv.org/pdf/2601.07074v1
+
+## 22534. Overcoming the Retrieval Barrier: Indirect Prompt Injection in the Wild for LLM Systems
+
+- **ID:** `2601.07072` · **primary:** `cs.CR` · **Submitted:** 2026-01-11
+- **Authors:** Hongyan Chang, Ergute Bao, Xinjian Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.07072 · **PDF:** https://arxiv.org/pdf/2601.07072v1
+
+## 22535. Neuromorphic FPGA Design for Digital Signal Processing
+
+- **ID:** `2601.07069` · **primary:** `cs.NE` · **Submitted:** 2026-01-11
+- **Authors:** Justin London
+- **abs:** https://arxiv.org/abs/2601.07069 · **PDF:** https://arxiv.org/pdf/2601.07069v1
+
+## 22536. Automated Domain Question Mapping (DQM) with Educational Learning Materials
+
+- **ID:** `2601.07062` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Jiho Noh, Mukhesh Raghava Katragadda, Dabae Lee
+- **abs:** https://arxiv.org/abs/2601.07062 · **PDF:** https://arxiv.org/pdf/2601.07062v1
+
+## 22537. Local EGOP for Continuous Index Learning
+
+- **ID:** `2601.07061` · **primary:** `stat.ML` · **Submitted:** 2026-01-11
+- **Authors:** Alex Kokot, Anand Hemmady, Vydhourie Thiyageswaran, et al.
+- **abs:** https://arxiv.org/abs/2601.07061 · **PDF:** https://arxiv.org/pdf/2601.07061v4
+
+## 22538. First African Digital Humanism Summer School 2025
+
+- **ID:** `2601.08870` · **primary:** `cs.CY` · **Submitted:** 2026-01-11
+- **Authors:** Carine P. Mukamakuza, Monika Lanzenberger, George Metakides, et al.
+- **abs:** https://arxiv.org/abs/2601.08870 · **PDF:** https://arxiv.org/pdf/2601.08870v1
+
+## 22539. Hallucinations Live in Variance
+
+- **ID:** `2601.07058` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Aaron R. Flouro, Shawn P. Chadwick
+- **abs:** https://arxiv.org/abs/2601.07058 · **PDF:** https://arxiv.org/pdf/2601.07058v1
+
+## 22540. Multi-Agent Cooperative Learning for Robust Vision-Language Alignment under OOD Concepts
+
+- **ID:** `2601.09746` · **primary:** `cs.MA` · **Submitted:** 2026-01-11
+- **Authors:** Philip Xu, Isabel Wagner, Eerke Boiten
+- **abs:** https://arxiv.org/abs/2601.09746 · **PDF:** https://arxiv.org/pdf/2601.09746v1
+
+## 22541. Adversarial Attacks on Medical Hyperspectral Imaging Exploiting Spectral-Spatial Dependencies and Multiscale Features
+
+- **ID:** `2601.07056` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Yunrui Gu, Zhenzhe Gao, Cong Kong, et al.
+- **abs:** https://arxiv.org/abs/2601.07056 · **PDF:** https://arxiv.org/pdf/2601.07056v1
+
+## 22542. Dr. Zero: Self-Evolving Search Agents without Training Data
+
+- **ID:** `2601.07055` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Zhenrui Yue, Kartikeya Upasani, Xianjun Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.07055 · **PDF:** https://arxiv.org/pdf/2601.07055v1
+
+## 22543. Fine-Tuning vs. RAG for Multi-Hop Question Answering with Novel Knowledge
+
+- **ID:** `2601.07054` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Zhuoyi Yang, Yurun Song, Iftekhar Ahmed, et al.
+- **abs:** https://arxiv.org/abs/2601.07054 · **PDF:** https://arxiv.org/pdf/2601.07054v1
+
+## 22544. GPU-Accelerated ANNS: Quantized for Speed, Built for Change
+
+- **ID:** `2601.07048` · **primary:** `cs.DB` · **Submitted:** 2026-01-11
+- **Authors:** Hunter McCoy, Zikun Wang, Prashant Pandey
+- **abs:** https://arxiv.org/abs/2601.07048 · **PDF:** https://arxiv.org/pdf/2601.07048v3
+
+## 22545. Engineering of Hallucination in Generative AI: It's not a Bug, it's a Feature
+
+- **ID:** `2601.07046` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Tim Fingscheidt, Patrick Blumenberg, Björn Möller
+- **abs:** https://arxiv.org/abs/2601.07046 · **PDF:** https://arxiv.org/pdf/2601.07046v1
+
+## 22546. When Abundance Conceals Weakness: Knowledge Conflict in Multilingual Models
+
+- **ID:** `2601.07041` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Jiaqi Zhao, Qiang Huang, Haodong Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.07041 · **PDF:** https://arxiv.org/pdf/2601.07041v1
+
+## 22547. Assessing LLM Response Quality in the Context of Technology-Facilitated Abuse
+
+- **ID:** `2602.17672` · **primary:** `cs.HC` · **Submitted:** 2026-01-11
+- **Authors:** Vijay Prakash, Majed Almansoori, Donghan Hu, et al.
+- **abs:** https://arxiv.org/abs/2602.17672 · **PDF:** https://arxiv.org/pdf/2602.17672v1
+
+## 22548. Task Arithmetic with Support Languages for Low-Resource ASR
+
+- **ID:** `2601.07038` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Emma Rafkin, Dan DeGenaro, Xiulin Yang
+- **abs:** https://arxiv.org/abs/2601.07038 · **PDF:** https://arxiv.org/pdf/2601.07038v2
+
+## 22549. Mid-Think: Training-Free Intermediate-Budget Reasoning via Token-Level Triggers
+
+- **ID:** `2601.07036` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Wang Yang, Debargha Ganguly, Xinpeng Li, et al.
+- **abs:** https://arxiv.org/abs/2601.07036 · **PDF:** https://arxiv.org/pdf/2601.07036v1
+
+## 22550. Explainable Deep Radiogenomic Molecular Imaging for MGMT Methylation Prediction in Glioblastoma
+
+- **ID:** `2601.07035` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Hasan M Jamil
+- **abs:** https://arxiv.org/abs/2601.07035 · **PDF:** https://arxiv.org/pdf/2601.07035v1
+
+## 22551. Codified Foreshadowing-Payoff Text Generation
+
+- **ID:** `2601.07033` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Longfei Yun, Kun Zhou, Yupeng Hou, et al.
+- **abs:** https://arxiv.org/abs/2601.07033 · **PDF:** https://arxiv.org/pdf/2601.07033v1
+
+## 22552. Which Quantization Should I Use? A Unified Evaluation of llama.cpp Quantization on Llama-3.1-8B-Instruct
+
+- **ID:** `2601.14277` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Uygar Kurt
+- **abs:** https://arxiv.org/abs/2601.14277 · **PDF:** https://arxiv.org/pdf/2601.14277v1
+
+## 22553. Sola-Visibility-ISPM: Benchmarking Agentic AI for Identity Security Posture Management Visibility
+
+- **ID:** `2601.07880` · **primary:** `cs.CR` · **Submitted:** 2026-01-11
+- **Authors:** Gal Engelberg, Konstantin Koutsyi, Leon Goldberg, et al.
+- **abs:** https://arxiv.org/abs/2601.07880 · **PDF:** https://arxiv.org/pdf/2601.07880v1
+
+## 22554. CloneMem: Benchmarking Long-Term Memory for AI Clones
+
+- **ID:** `2601.07023` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Sen Hu, Zhiyu Zhang, Yuxiang Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.07023 · **PDF:** https://arxiv.org/pdf/2601.07023v1
+
+## 22555. Solar Open Technical Report
+
+- **ID:** `2601.07022` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Sungrae Park, Sanghoon Kim, Jungho Cho, et al.
+- **abs:** https://arxiv.org/abs/2601.07022 · **PDF:** https://arxiv.org/pdf/2601.07022v1
+
+## 22556. Tight Analysis of Decentralized SGD: A Markov Chain Perspective
+
+- **ID:** `2601.07021` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Lucas Versini, Paul Mangold, Aymeric Dieuleveut
+- **abs:** https://arxiv.org/abs/2601.07021 · **PDF:** https://arxiv.org/pdf/2601.07021v1
+
+## 22557. TurkBench: A Benchmark for Evaluating Turkish Large Language Models
+
+- **ID:** `2601.07020` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Çağrı Toraman, Ahmet Kaan Sever, Ayse Aysu Cengiz, et al.
+- **abs:** https://arxiv.org/abs/2601.07020 · **PDF:** https://arxiv.org/pdf/2601.07020v2
+
+## 22558. Zer0n: An AI-Assisted Vulnerability Discovery and Blockchain-Backed Integrity Framework
+
+- **ID:** `2601.07019` · **primary:** `cs.CR` · **Submitted:** 2026-01-11
+- **Authors:** Harshil Parmar, Pushti Vyas, Prayers Khristi, et al.
+- **abs:** https://arxiv.org/abs/2601.07019 · **PDF:** https://arxiv.org/pdf/2601.07019v1
+
+## 22559. AI Deployment Authorisation: A Global Standard for Machine-Readable Governance of High-Risk Artificial Intelligence
+
+- **ID:** `2601.08869` · **primary:** `cs.CY` · **Submitted:** 2026-01-11
+- **Authors:** Daniel Djan Saparning
+- **abs:** https://arxiv.org/abs/2601.08869 · **PDF:** https://arxiv.org/pdf/2601.08869v1
+
+## 22560. Belief in False Information: A Human-Centered Security Risk in Sociotechnical Systems
+
+- **ID:** `2601.07016` · **primary:** `cs.SI` · **Submitted:** 2026-01-11
+- **Authors:** Fabian Walke, Thaddäa Nürnberger
+- **abs:** https://arxiv.org/abs/2601.07016 · **PDF:** https://arxiv.org/pdf/2601.07016v1
+
+## 22561. Conditional Normalizing Flows for Forward and Backward Joint State and Parameter Estimation
+
+- **ID:** `2601.07013` · **primary:** `stat.ML` · **Submitted:** 2026-01-11
+- **Authors:** Luke S. Lagunowich, Guoxiang Grayson Tong, Daniele E. Schiavazzi
+- **abs:** https://arxiv.org/abs/2601.07013 · **PDF:** https://arxiv.org/pdf/2601.07013v1
+
+## 22562. Stingy Context: 18:1 Hierarchical Code Compression for LLM Auto-Coding
+
+- **ID:** `2601.19929` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** David Linus Ostby
+- **abs:** https://arxiv.org/abs/2601.19929 · **PDF:** https://arxiv.org/pdf/2601.19929v2
+
+## 22563. Lexicalized Constituency Parsing for Middle Dutch: Low-resource Training and Cross-Domain Generalization
+
+- **ID:** `2601.07008` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Yiming Liang, Fang Zhao
+- **abs:** https://arxiv.org/abs/2601.07008 · **PDF:** https://arxiv.org/pdf/2601.07008v1
+
+## 22564. LLM Performance Predictors: Learning When to Escalate in Hybrid Human-AI Moderation Systems
+
+- **ID:** `2601.07006` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Or Bachar, Or Levi, Sardhendu Mishra, et al.
+- **abs:** https://arxiv.org/abs/2601.07006 · **PDF:** https://arxiv.org/pdf/2601.07006v1
+
+## 22565. MicLog: Towards Accurate and Efficient LLM-based Log Parsing via Progressive Meta In-Context Learning
+
+- **ID:** `2601.07005` · **primary:** `cs.SE` · **Submitted:** 2026-01-11
+- **Authors:** Jianbo Yu, Yixuan Li, Hai Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.07005 · **PDF:** https://arxiv.org/pdf/2601.07005v1
+
+## 22566. MemTrust: A Zero-Trust Architecture for Unified AI Memory System
+
+- **ID:** `2601.07004` · **primary:** `cs.CR` · **Submitted:** 2026-01-11
+- **Authors:** Xing Zhou, Dmitrii Ustiugov, Haoxin Shang, et al.
+- **abs:** https://arxiv.org/abs/2601.07004 · **PDF:** https://arxiv.org/pdf/2601.07004v1
+
+## 22567. Unity Forests: Improving Interaction Modelling and Interpretability in Random Forests
+
+- **ID:** `2601.07003` · **primary:** `stat.ME` · **Submitted:** 2026-01-11
+- **Authors:** Roman Hornung, Alexander Hapfelmeier
+- **abs:** https://arxiv.org/abs/2601.07003 · **PDF:** https://arxiv.org/pdf/2601.07003v1
+
+## 22568. FinCARDS: Card-Based Analyst Reranking for Financial Document Question Answering
+
+- **ID:** `2601.06992` · **primary:** `cs.IR` · **Submitted:** 2026-01-11
+- **Authors:** Yixi Zhou, Fan Zhang, Yu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.06992 · **PDF:** https://arxiv.org/pdf/2601.06992v1
+
+## 22569. Match Made with Matrix Completion: Efficient Learning under Matching Interference
+
+- **ID:** `2601.06982` · **primary:** `stat.ML` · **Submitted:** 2026-01-11
+- **Authors:** Zhiyuan Tang, Wanning Chen, Kan Xu
+- **abs:** https://arxiv.org/abs/2601.06982 · **PDF:** https://arxiv.org/pdf/2601.06982v1
+
+## 22570. MedTutor: A Retrieval-Augmented LLM System for Case-Based Medical Education
+
+- **ID:** `2601.06979` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Dongsuk Jang, Ziyao Shangguan, Kyle Tegtmeyer, et al.
+- **abs:** https://arxiv.org/abs/2601.06979 · **PDF:** https://arxiv.org/pdf/2601.06979v2
+
+## 22571. Benchmarking Autonomy in Scientific Experiments: A Hierarchical Taxonomy for Autonomous Large-Scale Facilities
+
+- **ID:** `2601.06978` · **primary:** `physics.ins-det` · **Submitted:** 2026-01-11
+- **Authors:** James Le Houx
+- **abs:** https://arxiv.org/abs/2601.06978 · **PDF:** https://arxiv.org/pdf/2601.06978v1
+
+## 22572. Explore with Long-term Memory: A Benchmark and Multimodal LLM-based Reinforcement Learning Framework for Embodied Exploration
+
+- **ID:** `2601.10744` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Sen Wang, Bangwei Liu, Zhenkun Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.10744 · **PDF:** https://arxiv.org/pdf/2601.10744v2
+
+## 22573. UETQuintet at BioCreative IX -- MedHopQA: Enhancing Biomedical QA with Selective Multi-hop Reasoning and Contextual Retrieval
+
+- **ID:** `2601.06974` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Quoc-An Nguyen, Thi-Minh-Thu Vu, Bich-Dat Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.06974 · **PDF:** https://arxiv.org/pdf/2601.06974v1
+
+## 22574. LLMs Can't Play Hangman: On the Necessity of a Private Working Memory for Language Agents
+
+- **ID:** `2601.06973` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Davide Baldelli, Ali Parviz, Amal Zouaq, et al.
+- **abs:** https://arxiv.org/abs/2601.06973 · **PDF:** https://arxiv.org/pdf/2601.06973v1
+
+## 22575. Categorize Early, Integrate Late: Divergent Processing Strategies in Automatic Speech Recognition
+
+- **ID:** `2601.06972` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Nathan Roll, Pranav Bhalerao, Martijn Bartelds, et al.
+- **abs:** https://arxiv.org/abs/2601.06972 · **PDF:** https://arxiv.org/pdf/2601.06972v1
+
+## 22576. Generalization Bounds for Transformer Channel Decoders
+
+- **ID:** `2601.06969` · **primary:** `cs.IT` · **Submitted:** 2026-01-11
+- **Authors:** Qinshan Zhang, Bin Chen, Yong Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.06969 · **PDF:** https://arxiv.org/pdf/2601.06969v1
+
+## 22577. A Robust Certified Machine Unlearning Method Under Distribution Shift
+
+- **ID:** `2601.06967` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Jinduo Guo, Yinzhi Cao
+- **abs:** https://arxiv.org/abs/2601.06967 · **PDF:** https://arxiv.org/pdf/2601.06967v1
+
+## 22578. RealMem: Benchmarking LLMs in Real-World Memory-Driven Interaction
+
+- **ID:** `2601.06966` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Haonan Bian, Zhiyuan Yao, Sen Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.06966 · **PDF:** https://arxiv.org/pdf/2601.06966v1
+
+## 22579. The Impact of Anisotropic Covariance Structure on the Training Dynamics and Generalization Error of Linear Networks
+
+- **ID:** `2601.06961` · **primary:** `stat.ML` · **Submitted:** 2026-01-11
+- **Authors:** Taishi Watanabe, Ryo Karakida, Jun-nosuke Teramae
+- **abs:** https://arxiv.org/abs/2601.06961 · **PDF:** https://arxiv.org/pdf/2601.06961v1
+
+## 22580. HAS-VQ: Hessian-Adaptive Sparse Vector Quantization for High-Fidelity LLM Compression
+
+- **ID:** `2601.06959` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Vladimer Khasia
+- **abs:** https://arxiv.org/abs/2601.06959 · **PDF:** https://arxiv.org/pdf/2601.06959v1
+
+## 22581. X-Coder: Advancing Competitive Programming with Fully Synthetic Tasks, Solutions, and Tests
+
+- **ID:** `2601.06953` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Jie Wu, Haoling Li, Xin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.06953 · **PDF:** https://arxiv.org/pdf/2601.06953v2
+
+## 22582. Sliced-Wasserstein Distribution Alignment Loss Improves the Ultra-Low-Bit Quantization of Large Language Models
+
+- **ID:** `2601.07878` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Deyu Cao, Yixin Yin, Samin Aref
+- **abs:** https://arxiv.org/abs/2601.07878 · **PDF:** https://arxiv.org/pdf/2601.07878v1
+
+## 22583. SketchJudge: A Diagnostic Benchmark for Grading Hand-drawn Diagrams with Multimodal Large Language Models
+
+- **ID:** `2601.06944` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Yuhang Su, Mei Wang, Yaoyao Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.06944 · **PDF:** https://arxiv.org/pdf/2601.06944v1
+
+## 22584. Watching, Reasoning, and Searching: A Video Deep Research Benchmark on Open Web for Agentic Video Reasoning
+
+- **ID:** `2601.06943` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Chengwen Liu, Xiaomin Yu, Zhuoyue Chang, et al.
+- **abs:** https://arxiv.org/abs/2601.06943 · **PDF:** https://arxiv.org/pdf/2601.06943v1
+
+## 22585. Towards Operational Streamflow Forecasting in the Limpopo River Basin using Long Short-Term Memory Networks
+
+- **ID:** `2601.06941` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** James Tlhomole, Edoardo Borgomeo, Karthikeyan Matheswaran, et al.
+- **abs:** https://arxiv.org/abs/2601.06941 · **PDF:** https://arxiv.org/pdf/2601.06941v1
+
+## 22586. VISTA: Knowledge-Driven Vessel Trajectory Imputation with Repair Provenance
+
+- **ID:** `2601.06940` · **primary:** `cs.DB` · **Submitted:** 2026-01-11
+- **Authors:** Hengyu Liu, Tianyi Li, Haoyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.06940 · **PDF:** https://arxiv.org/pdf/2601.06940v2
+
+## 22587. Forgetting Similar Samples: Can Machine Unlearning Do it Better?
+
+- **ID:** `2601.06938` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Heng Xu, Tianqing Zhu, Dayong Ye, et al.
+- **abs:** https://arxiv.org/abs/2601.06938 · **PDF:** https://arxiv.org/pdf/2601.06938v1
+
+## 22588. mind_call: A Dataset for Mental Health Function Calling with Large Language Models
+
+- **ID:** `2601.06937` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Fozle Rabbi Shafi, M. Anwar Hossain, Salimur Choudhury
+- **abs:** https://arxiv.org/abs/2601.06937 · **PDF:** https://arxiv.org/pdf/2601.06937v1
+
+## 22589. Symphonym: Universal Phonetic Embeddings for Cross-Script Name Matching
+
+- **ID:** `2601.06932` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Stephen Gadd
+- **abs:** https://arxiv.org/abs/2601.06932 · **PDF:** https://arxiv.org/pdf/2601.06932v4
+
+## 22590. Measuring Social Bias in Vision-Language Models with Face-Only Counterfactuals from Real Photos
+
+- **ID:** `2601.06931` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Haodong Chen, Qiang Huang, Jiaqi Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.06931 · **PDF:** https://arxiv.org/pdf/2601.06931v1
+
+## 22591. TreePS-RAG: Tree-based Process Supervision for Reinforcement Learning in Agentic RAG
+
+- **ID:** `2601.06922` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Tianhua Zhang, Kun Li, Junan Li, et al.
+- **abs:** https://arxiv.org/abs/2601.06922 · **PDF:** https://arxiv.org/pdf/2601.06922v1
+
+## 22592. Calibrating Agent-Based Financial Markets Simulators with Pretrainable Automatic Posterior Transformation-Based Surrogates
+
+- **ID:** `2601.06920` · **primary:** `cs.NE` · **Submitted:** 2026-01-11
+- **Authors:** Boquan Jiang, Zhenhua Yang, Chenkai Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.06920 · **PDF:** https://arxiv.org/pdf/2601.06920v1
+
+## 22593. Enhancing Formal Software Specification with Artificial Intelligence
+
+- **ID:** `2601.09745` · **primary:** `cs.SE` · **Submitted:** 2026-01-11
+- **Authors:** Antonio Abu Nassar, Eitan Farchi
+- **abs:** https://arxiv.org/abs/2601.09745 · **PDF:** https://arxiv.org/pdf/2601.09745v1
+
+## 22594. Active Learning Strategies for Efficient Machine-Learned Interatomic Potentials Across Diverse Material Systems
+
+- **ID:** `2601.06916` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Mohammed Azeez Khan, Aaron D'Souza, Vijay Choyal
+- **abs:** https://arxiv.org/abs/2601.06916 · **PDF:** https://arxiv.org/pdf/2601.06916v2
+
+## 22595. Towards Compositional Generalization in LLMs for Smart Contract Security: A Case Study on Reentrancy Vulnerabilities
+
+- **ID:** `2601.06914` · **primary:** `cs.CR` · **Submitted:** 2026-01-11
+- **Authors:** Ying Zhou, Jiacheng Wei, Yu Qi, et al.
+- **abs:** https://arxiv.org/abs/2601.06914 · **PDF:** https://arxiv.org/pdf/2601.06914v1
+
+## 22596. Tractable Multinomial Logit Contextual Bandits with Non-Linear Utilities
+
+- **ID:** `2601.06913` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Taehyun Hwang, Dahngoon Kim, Min-hwan Oh
+- **abs:** https://arxiv.org/abs/2601.06913 · **PDF:** https://arxiv.org/pdf/2601.06913v1
+
+## 22597. Distributional Clarity: The Hidden Driver of RL-Friendliness in Large Language Models
+
+- **ID:** `2601.06911` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Shaoning Sun, Mingzhu Cai, Huang He, et al.
+- **abs:** https://arxiv.org/abs/2601.06911 · **PDF:** https://arxiv.org/pdf/2601.06911v1
+
+## 22598. E^2-LLM: Bridging Neural Signals and Interpretable Affective Analysis
+
+- **ID:** `2601.07877` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Fei Ma, Han Lin, Yifan Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.07877 · **PDF:** https://arxiv.org/pdf/2601.07877v1
+
+## 22599. Fine-grained Verbal Attack Detection via a Hierarchical Divide-and-Conquer Framework
+
+- **ID:** `2601.06907` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Quan Zheng, Yuanhe Tian, Ming Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.06907 · **PDF:** https://arxiv.org/pdf/2601.06907v1
+
+## 22600. NOVAK: Unified adaptive optimizer for deep neural networks
+
+- **ID:** `2601.07876` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Sergii Kavun
+- **abs:** https://arxiv.org/abs/2601.07876 · **PDF:** https://arxiv.org/pdf/2601.07876v1
+
+## 22601. V2P: Visual Attention Calibration for GUI Grounding via Background Suppression and Center Peaking
+
+- **ID:** `2601.06899` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Jikai Chen, Long Chen, Dong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.06899 · **PDF:** https://arxiv.org/pdf/2601.06899v2
+
+## 22602. TagSpeech: End-to-End Multi-Speaker ASR and Diarization with Fine-Grained Temporal Grounding
+
+- **ID:** `2601.06896` · **primary:** `eess.AS` · **Submitted:** 2026-01-11
+- **Authors:** Mingyue Huo, Yiwen Shao, Yuheng Zhang
+- **abs:** https://arxiv.org/abs/2601.06896 · **PDF:** https://arxiv.org/pdf/2601.06896v1
+
+## 22603. Paraphrasing Adversarial Attack on LLM-as-a-Reviewer
+
+- **ID:** `2601.06884` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Masahiro Kaneko
+- **abs:** https://arxiv.org/abs/2601.06884 · **PDF:** https://arxiv.org/pdf/2601.06884v1
+
+## 22604. MixRI: Mixing Features of Reference Images for Novel Object Pose Estimation
+
+- **ID:** `2601.06883` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Xinhang Liu, Jiawei Shi, Zheng Dang, et al.
+- **abs:** https://arxiv.org/abs/2601.06883 · **PDF:** https://arxiv.org/pdf/2601.06883v1
+
+## 22605. Residual Cross-Modal Fusion Networks for Audio-Visual Navigation
+
+- **ID:** `2601.08868` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Yi Wang, Yinfeng Yu, Bin Ren
+- **abs:** https://arxiv.org/abs/2601.08868 · **PDF:** https://arxiv.org/pdf/2601.08868v1
+
+## 22606. Personality-Aware Reinforcement Learning for Persuasive Dialogue with LLM-Driven Simulation
+
+- **ID:** `2601.06877` · **primary:** `cs.HC` · **Submitted:** 2026-01-11
+- **Authors:** Donghuo Zeng, Roberto Legaspi, Kazushi Ikeda
+- **abs:** https://arxiv.org/abs/2601.06877 · **PDF:** https://arxiv.org/pdf/2601.06877v1
+
+## 22607. An Ubuntu-Guided Large Language Model Framework for Cognitive Behavioral Mental Health Dialogue
+
+- **ID:** `2601.06875` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Sontaga G. Forane, Absalom E. Ezugwu, Kevin Igwe, et al.
+- **abs:** https://arxiv.org/abs/2601.06875 · **PDF:** https://arxiv.org/pdf/2601.06875v1
+
+## 22608. Applying Embedding-Based Retrieval to Airbnb Search
+
+- **ID:** `2601.06873` · **primary:** `cs.IR` · **Submitted:** 2026-01-11
+- **Authors:** Mustafa Abdool, Soumyadip Banerjee, Moutupsi Paul, et al.
+- **abs:** https://arxiv.org/abs/2601.06873 · **PDF:** https://arxiv.org/pdf/2601.06873v1
+
+## 22609. DaQ-MSA: Denoising and Qualifying Diffusion Augmentations for Multimodal Sentiment Analysis
+
+- **ID:** `2601.06870` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Jiazhang Liang, Jianheng Dai, Miaosen Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.06870 · **PDF:** https://arxiv.org/pdf/2601.06870v1
+
+## 22610. U-MASK: User-adaptive Spatio-Temporal Masking for Personalized Mobile AI Applications
+
+- **ID:** `2601.06867` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Shiyuan Zhang, Yilai Liu, Yuwei Du, et al.
+- **abs:** https://arxiv.org/abs/2601.06867 · **PDF:** https://arxiv.org/pdf/2601.06867v1
+
+## 22611. qAttCNN - Self Attention Mechanism for Video QoE Prediction in Encrypted Traffic
+
+- **ID:** `2601.06862` · **primary:** `cs.CR` · **Submitted:** 2026-01-11
+- **Authors:** Michael Sidorov, Ofer Hadar
+- **abs:** https://arxiv.org/abs/2601.06862 · **PDF:** https://arxiv.org/pdf/2601.06862v1
+
+## 22612. BiasLab: A Multilingual, Dual-Framing Framework for Robust Measurement of Output-Level Bias in Large Language Models
+
+- **ID:** `2601.06861` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** William Guey, Wei Zhang, Pei-Luen Patrick Rau, et al.
+- **abs:** https://arxiv.org/abs/2601.06861 · **PDF:** https://arxiv.org/pdf/2601.06861v1
+
+## 22613. ET-Agent: Incentivizing Effective Tool-Integrated Reasoning Agent via Behavior Calibration
+
+- **ID:** `2601.06860` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Yifei Chen, Guanting Dong, Zhicheng Dou
+- **abs:** https://arxiv.org/abs/2601.06860 · **PDF:** https://arxiv.org/pdf/2601.06860v2
+
+## 22614. Deep Learning Based Channel Extrapolation for Dual-Band Massive MIMO Systems
+
+- **ID:** `2601.06858` · **primary:** `eess.SP` · **Submitted:** 2026-01-11
+- **Authors:** Qikai Xiao, Kehui Li, Binggui Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.06858 · **PDF:** https://arxiv.org/pdf/2601.06858v1
+
+## 22615. MoE-DisCo:Low Economy Cost Training Mixture-of-Experts Models
+
+- **ID:** `2601.06857` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Xin Ye, Daning Cheng, Boyang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.06857 · **PDF:** https://arxiv.org/pdf/2601.06857v1
+
+## 22616. †DAGGER: Distractor-Aware Graph Generation for Executable Reasoning in Math Problems
+
+- **ID:** `2601.06853` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Zabir Al Nazi, Shubhashis Roy Dipta, Sudipta Kar
+- **abs:** https://arxiv.org/abs/2601.06853 · **PDF:** https://arxiv.org/pdf/2601.06853v2
+
+## 22617. A Brain-like Synergistic Core in LLMs Drives Behaviour and Learning
+
+- **ID:** `2601.06851` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Pedro Urbina-Rodriguez, Zafeirios Fountas, Fernando E. Rosas, et al.
+- **abs:** https://arxiv.org/abs/2601.06851 · **PDF:** https://arxiv.org/pdf/2601.06851v1
+
+## 22618. Explainable Multimodal Aspect-Based Sentiment Analysis with Dependency-guided Large Language Model
+
+- **ID:** `2601.06848` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Zhongzheng Wang, Yuanhe Tian, Hongzhi Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.06848 · **PDF:** https://arxiv.org/pdf/2601.06848v1
+
+## 22619. MedGround: Bridging the Evidence Gap in Medical Vision-Language Models with Verified Grounding Data
+
+- **ID:** `2601.06847` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Mengmeng Zhang, Xiaoping Wu, Hao Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.06847 · **PDF:** https://arxiv.org/pdf/2601.06847v1
+
+## 22620. Code Evolution for Control: Synthesizing Policies via LLM-Driven Evolutionary Search
+
+- **ID:** `2601.06845` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Ping Guo, Chao Li, Yinglan Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.06845 · **PDF:** https://arxiv.org/pdf/2601.06845v1
+
+## 22621. Variational decomposition autoencoding improves disentanglement of latent representations
+
+- **ID:** `2601.06844` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Ioannis Ziogas, Aamna Al Shehhi, Ahsan H. Khandoker, et al.
+- **abs:** https://arxiv.org/abs/2601.06844 · **PDF:** https://arxiv.org/pdf/2601.06844v1
+
+## 22622. Speak While Watching: Unleashing TRUE Real-Time Video Understanding Capability of Multimodal Large Language Models
+
+- **ID:** `2601.06843` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Junyan Lin, Junlong Tong, Hao Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.06843 · **PDF:** https://arxiv.org/pdf/2601.06843v1
+
+## 22623. Seeing through the Conflict: Transparent Knowledge Conflict Handling in Retrieval-Augmented Generation
+
+- **ID:** `2601.06842` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Hua Ye, Siyuan Chen, Ziqi Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.06842 · **PDF:** https://arxiv.org/pdf/2601.06842v1
+
+## 22624. OSCAR: Optical-aware Semantic Control for Aleatoric Refinement in Sar-to-Optical Translation
+
+- **ID:** `2601.06835` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Hyunseo Lee, Sang Min Kim, Ho Kyung Shin, et al.
+- **abs:** https://arxiv.org/abs/2601.06835 · **PDF:** https://arxiv.org/pdf/2601.06835v1
+
+## 22625. Constrained Density Estimation via Optimal Transport
+
+- **ID:** `2601.06830` · **primary:** `stat.ML` · **Submitted:** 2026-01-11
+- **Authors:** Yinan Hu, Esteban G. Tabak
+- **abs:** https://arxiv.org/abs/2601.06830 · **PDF:** https://arxiv.org/pdf/2601.06830v2
+
+## 22626. PDR: A Plug-and-Play Positional Decay Framework for LLM Pre-training Data Detection
+
+- **ID:** `2601.06827` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Jinhan Liu, Yibo Yang, Ruiying Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.06827 · **PDF:** https://arxiv.org/pdf/2601.06827v1
+
+## 22627. AgentHallu: Benchmarking Automated Hallucination Attribution of LLM-based Agents
+
+- **ID:** `2601.06818` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Xuannan Liu, Xiao Yang, Zekun Li, et al.
+- **abs:** https://arxiv.org/abs/2601.06818 · **PDF:** https://arxiv.org/pdf/2601.06818v1
+
+## 22628. Analyzing the effect of prediction accuracy on the distributionally-robust competitive ratio
+
+- **ID:** `2601.06813` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Toru Yoshinaga, Yasushi Kawase
+- **abs:** https://arxiv.org/abs/2601.06813 · **PDF:** https://arxiv.org/pdf/2601.06813v1
+
+## 22629. Towards a Mechanistic Understanding of Large Reasoning Models: A Survey of Training, Inference, and Failures
+
+- **ID:** `2601.19928` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Yi Hu, Jiaqi Gu, Ruxin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19928 · **PDF:** https://arxiv.org/pdf/2601.19928v1
+
+## 22630. WFR-FM: Simulation-Free Dynamic Unbalanced Optimal Transport
+
+- **ID:** `2601.06810` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Qiangwei Peng, Zihan Wang, Junda Ying, et al.
+- **abs:** https://arxiv.org/abs/2601.06810 · **PDF:** https://arxiv.org/pdf/2601.06810v2
+
+## 22631. SpatialNav: Leveraging Spatial Scene Graphs for Zero-Shot Vision-and-Language Navigation
+
+- **ID:** `2601.06806` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Jiwen Zhang, Zejun Li, Siyuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.06806 · **PDF:** https://arxiv.org/pdf/2601.06806v1
+
+## 22632. Forest Before Trees: Latent Superposition for Efficient Visual Reasoning
+
+- **ID:** `2601.06803` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Yubo Wang, Juntian Zhang, Yichen Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.06803 · **PDF:** https://arxiv.org/pdf/2601.06803v1
+
+## 22633. Doing More with Less: Data Augmentation for Sudanese Dialect Automatic Speech Recognition
+
+- **ID:** `2601.06802` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Ayman Mansour
+- **abs:** https://arxiv.org/abs/2601.06802 · **PDF:** https://arxiv.org/pdf/2601.06802v1
+
+## 22634. Thinking with Deltas: Incentivizing Reinforcement Learning via Differential Visual Reasoning Policy
+
+- **ID:** `2601.06801` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Shujian Gao, Yuan Wang, Jiangtao Yan, et al.
+- **abs:** https://arxiv.org/abs/2601.06801 · **PDF:** https://arxiv.org/pdf/2601.06801v1
+
+## 22635. Graph Neural Network with One-side Edge Sampling for Fraud Detection
+
+- **ID:** `2601.06800` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Hoang Hiep Trieu
+- **abs:** https://arxiv.org/abs/2601.06800 · **PDF:** https://arxiv.org/pdf/2601.06800v1
+
+## 22636. CIRAG: Construction-Integration Retrieval and Adaptive Generation for Multi-hop Question Answering
+
+- **ID:** `2601.06799` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Zili Wei, Xiaocui Yang, Yilin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.06799 · **PDF:** https://arxiv.org/pdf/2601.06799v1
+
+## 22637. GDEPO: Group Dual-dynamic and Equal-right Advantage Policy Optimization with Enhanced Training Data Utilization for Sample-Constrained Reinforcement Learning
+
+- **ID:** `2601.06795` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Zhengqing Yan, Xinyang Liu, Yi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.06795 · **PDF:** https://arxiv.org/pdf/2601.06795v3
+
+## 22638. No More Stale Feedback: Co-Evolving Critics for Open-World Agent Learning
+
+- **ID:** `2601.06794` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Zhicong Li, Lingjie Jiang, Yulan Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.06794 · **PDF:** https://arxiv.org/pdf/2601.06794v1
+
+## 22639. CliffordNet: All You Need is Geometric Algebra
+
+- **ID:** `2601.06793` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Zhongping Ji
+- **abs:** https://arxiv.org/abs/2601.06793 · **PDF:** https://arxiv.org/pdf/2601.06793v2
+
+## 22640. Cross-Modal Computational Model of Brain-Heart Interactions via HRV and EEG Feature
+
+- **ID:** `2601.06792` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Malavika Pradeep, Akshay Sasi, Nusaibah Farrukh, et al.
+- **abs:** https://arxiv.org/abs/2601.06792 · **PDF:** https://arxiv.org/pdf/2601.06792v1
+
+## 22641. SecMoE: Communication-Efficient Secure MoE Inference via Select-Then-Compute
+
+- **ID:** `2601.06790` · **primary:** `cs.CR` · **Submitted:** 2026-01-11
+- **Authors:** Bowen Shen, Yuyue Chen, Peng Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.06790 · **PDF:** https://arxiv.org/pdf/2601.06790v1
+
+## 22642. MemGovern: Enhancing Code Agents through Learning from Governed Human Experiences
+
+- **ID:** `2601.06789` · **primary:** `cs.SE` · **Submitted:** 2026-01-11
+- **Authors:** Qihao Wang, Ziming Cheng, Shuo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.06789 · **PDF:** https://arxiv.org/pdf/2601.06789v2
+
+## 22643. Artificial Entanglement in the Fine-Tuning of Large Language Models
+
+- **ID:** `2601.06788` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Min Chen, Zihan Wang, Canyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.06788 · **PDF:** https://arxiv.org/pdf/2601.06788v1
+
+## 22644. Garbage Attention in Large Language Models: BOS Sink Heads and Sink-aware Pruning
+
+- **ID:** `2601.06787` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Jaewon Sok, Jewon Yeom, Seonghyeon Park, et al.
+- **abs:** https://arxiv.org/abs/2601.06787 · **PDF:** https://arxiv.org/pdf/2601.06787v1
+
+## 22645. EpiCaR: Knowing What You Don't Know Matters for Better Reasoning in LLMs
+
+- **ID:** `2601.06786` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Jewon Yeom, Jaewon Sok, Seonghyeon Park, et al.
+- **abs:** https://arxiv.org/abs/2601.06786 · **PDF:** https://arxiv.org/pdf/2601.06786v1
+
+## 22646. Dimension-reduced outcome-weighted learning for estimating individualized treatment regimes in observational studies
+
+- **ID:** `2601.06782` · **primary:** `stat.ML` · **Submitted:** 2026-01-11
+- **Authors:** Sungtaek Son, Eardi Lila, Kwun Chuen Gary Chan
+- **abs:** https://arxiv.org/abs/2601.06782 · **PDF:** https://arxiv.org/pdf/2601.06782v1
+
+## 22647. AutoTour: Automatic Photo Tour Guide with Smartphones and LLMs
+
+- **ID:** `2601.06781` · **primary:** `cs.HC` · **Submitted:** 2026-01-11
+- **Authors:** Huatao Xu, Zihe Liu, Zilin Zeng, et al.
+- **abs:** https://arxiv.org/abs/2601.06781 · **PDF:** https://arxiv.org/pdf/2601.06781v1
+
+## 22648. Multi-Stage Evolutionary Model Merging with Meta Data Driven Curriculum Learning for Sentiment-Specialized Large Language Modeling
+
+- **ID:** `2601.06780` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Keito Inoshita, Xiaokang Zhou, Akira Kawai
+- **abs:** https://arxiv.org/abs/2601.06780 · **PDF:** https://arxiv.org/pdf/2601.06780v1
+
+## 22649. From Text to Simulation: A Multi-Agent LLM Workflow for Automated Chemical Process Design
+
+- **ID:** `2601.06776` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Xufei Tian, Wenli Du, Shaoyi Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.06776 · **PDF:** https://arxiv.org/pdf/2601.06776v1
+
+## 22650. Multiplicative Orthogonal Sequential Editing for Language Models
+
+- **ID:** `2601.07873` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Hao-Xiang Xu, Jun-Yu Ma, Ziqi Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.07873 · **PDF:** https://arxiv.org/pdf/2601.07873v1
+
+## 22651. Structure-preserving learning and prediction in optimal control of collective motion
+
+- **ID:** `2601.06770` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Sofiia Huraka, Vakhtang Putkaradze
+- **abs:** https://arxiv.org/abs/2601.06770 · **PDF:** https://arxiv.org/pdf/2601.06770v1
+
+## 22652. ALFA: A Safe-by-Design Approach to Mitigate Quishing Attacks Launched via Fancy QR Codes
+
+- **ID:** `2601.06768` · **primary:** `cs.CR` · **Submitted:** 2026-01-11
+- **Authors:** Muhammad Wahid Akram, Keshav Sood, Muneeb Ul Hassan, et al.
+- **abs:** https://arxiv.org/abs/2601.06768 · **PDF:** https://arxiv.org/pdf/2601.06768v1
+
+## 22653. GanitLLM: Difficulty-Aware Bengali Mathematical Reasoning through Curriculum-GRPO
+
+- **ID:** `2601.06767` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Shubhashis Roy Dipta, Khairul Mahbub, Nadia Najjar
+- **abs:** https://arxiv.org/abs/2601.06767 · **PDF:** https://arxiv.org/pdf/2601.06767v1
+
+## 22654. Comparative Separation: Evaluating Separation on Comparative Judgment Test Data
+
+- **ID:** `2601.06761` · **primary:** `cs.SE` · **Submitted:** 2026-01-11
+- **Authors:** Xiaoyin Xi, Neeku Capak, Kate Stockwell, et al.
+- **abs:** https://arxiv.org/abs/2601.06761 · **PDF:** https://arxiv.org/pdf/2601.06761v1
+
+## 22655. A Backpropagation-Free Feedback-Hebbian Network for Continual Learning Dynamics
+
+- **ID:** `2601.06758` · **primary:** `cs.NE` · **Submitted:** 2026-01-11
+- **Authors:** Josh Li, Fow-sen Choa
+- **abs:** https://arxiv.org/abs/2601.06758 · **PDF:** https://arxiv.org/pdf/2601.06758v3
+
+## 22656. UBiGTLoc: A Unified BiLSTM-Graph Transformer Localization Framework for IoT Sensor Networks
+
+- **ID:** `2601.10743` · **primary:** `eess.SP` · **Submitted:** 2026-01-11
+- **Authors:** Ayesh Abu Lehyeh, Anastassia Gharib, Tian Xia, et al.
+- **abs:** https://arxiv.org/abs/2601.10743 · **PDF:** https://arxiv.org/pdf/2601.10743v1
+
+## 22657. MTMCS-Bench: Evaluating Contextual Safety of Multimodal Large Language Models in Multi-Turn Dialogues
+
+- **ID:** `2601.06757` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Zheyuan Liu, Dongwhi Kim, Yixin Wan, et al.
+- **abs:** https://arxiv.org/abs/2601.06757 · **PDF:** https://arxiv.org/pdf/2601.06757v1
+
+## 22658. Towards Computational Chinese Paleography
+
+- **ID:** `2601.06753` · **primary:** `cs.CL` · **Submitted:** 2026-01-11
+- **Authors:** Yiran Rex Ma
+- **abs:** https://arxiv.org/abs/2601.06753 · **PDF:** https://arxiv.org/pdf/2601.06753v2
+
+## 22659. AI Hallucination from Students' Perspective: A Thematic Analysis
+
+- **ID:** `2602.17671` · **primary:** `cs.HC` · **Submitted:** 2026-01-11
+- **Authors:** Abdulhadi Shoufan, Ahmad-Azmi-Abdelhamid Esmaeil
+- **abs:** https://arxiv.org/abs/2602.17671 · **PDF:** https://arxiv.org/pdf/2602.17671v1
+
+## 22660. R$^2$BD: A Reconstruction-Based Method for Generalizable and Efficient Detection of Fake Images
+
+- **ID:** `2601.08867` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Qingyu Liu, Zhongjie Ba, Jianmin Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.08867 · **PDF:** https://arxiv.org/pdf/2601.08867v1
+
+## 22661. Benchmarking Egocentric Clinical Intent Understanding Capability for Medical Multimodal Large Language Models
+
+- **ID:** `2601.06750` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Shaonan Liu, Guo Yu, Xiaoling Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.06750 · **PDF:** https://arxiv.org/pdf/2601.06750v1
+
+## 22662. FinForge: Semi-Synthetic Financial Benchmark Generation
+
+- **ID:** `2601.06747` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Glenn Matlin, Akhil Theerthala, Anant Gupta, et al.
+- **abs:** https://arxiv.org/abs/2601.06747 · **PDF:** https://arxiv.org/pdf/2601.06747v2
+
+## 22663. Federated Continual Learning for Privacy-Preserving Hospital Imaging Classification
+
+- **ID:** `2601.06742` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Anay Sinhal, Arpana Sinhal, Amit Sinhal
+- **abs:** https://arxiv.org/abs/2601.06742 · **PDF:** https://arxiv.org/pdf/2601.06742v1
+
+## 22664. A Mind Cannot Be Smeared Across Time
+
+- **ID:** `2601.11620` · **primary:** `cs.AI` · **Submitted:** 2026-01-11
+- **Authors:** Michael Timothy Bennett
+- **abs:** https://arxiv.org/abs/2601.11620 · **PDF:** https://arxiv.org/pdf/2601.11620v2
+
+## 22665. Logic-Driven Semantic Communication for Resilient Multi-Agent Systems
+
+- **ID:** `2601.06733` · **primary:** `cs.MA` · **Submitted:** 2026-01-11
+- **Authors:** Tamara Alshammari, Mehdi Bennis
+- **abs:** https://arxiv.org/abs/2601.06733 · **PDF:** https://arxiv.org/pdf/2601.06733v1
+
+## 22666. Why are there many equally good models? An Anatomy of the Rashomon Effect
+
+- **ID:** `2601.06730` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Harsh Parikh
+- **abs:** https://arxiv.org/abs/2601.06730 · **PDF:** https://arxiv.org/pdf/2601.06730v2
+
+## 22667. Predicting Student Success with Heterogeneous Graph Deep Learning and Machine Learning Models
+
+- **ID:** `2601.06729` · **primary:** `cs.LG` · **Submitted:** 2026-01-11
+- **Authors:** Anca Muresan, Mihaela Cardei, Ionut Cardei
+- **abs:** https://arxiv.org/abs/2601.06729 · **PDF:** https://arxiv.org/pdf/2601.06729v1
+
+## 22668. When Humans Judge Irises: Pupil Size Normalization as an Aid and Synthetic Irises as a Challenge
+
+- **ID:** `2601.06725` · **primary:** `cs.CV` · **Submitted:** 2026-01-11
+- **Authors:** Mahsa Mitcheff, Adam Czajka
+- **abs:** https://arxiv.org/abs/2601.06725 · **PDF:** https://arxiv.org/pdf/2601.06725v1
+
+## 22669. DS-CIM: Digital Stochastic Computing-In-Memory Featuring Accurate OR-Accumulation via Sample Region Remapping for Edge AI Models
+
+- **ID:** `2601.06724` · **primary:** `cs.AR` · **Submitted:** 2026-01-10
+- **Authors:** Kunming Shao, Liang Zhao, Jiangnan Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.06724 · **PDF:** https://arxiv.org/pdf/2601.06724v1
+
+## 22670. Imaging-anchored Multiomics in Cardiovascular Disease: Integrating Cardiac Imaging, Bulk, Single-cell, and Spatial Transcriptomics
+
+- **ID:** `2601.07871` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-10
+- **Authors:** Minh H. N. Le, Tuan Vinh, Thanh-Huy Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.07871 · **PDF:** https://arxiv.org/pdf/2601.07871v1
+
+## 22671. Diffusion Models with Heavy-Tailed Targets: Score Estimation and Sampling Guarantees
+
+- **ID:** `2601.06715` · **primary:** `math.ST` · **Submitted:** 2026-01-10
+- **Authors:** Yifeng Yu, Lu Yu
+- **abs:** https://arxiv.org/abs/2601.06715 · **PDF:** https://arxiv.org/pdf/2601.06715v1
+
+## 22672. An Explainable Market Integrity Monitoring System with Multi-Source Attention Signals and Transparent Scoring
+
+- **ID:** `2601.15304` · **primary:** `q-fin.RM` · **Submitted:** 2026-01-10
+- **Authors:** Sandeep Neela
+- **abs:** https://arxiv.org/abs/2601.15304 · **PDF:** https://arxiv.org/pdf/2601.15304v1
+
+## 22673. Evaluating Accounting Reasoning Capabilities of Large Language Models
+
+- **ID:** `2601.06707` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Jie Zhou, Xin Chen, Jie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.06707 · **PDF:** https://arxiv.org/pdf/2601.06707v1
+
+## 22674. HOSC: A Periodic Activation with Saturation Control for High-Fidelity Implicit Neural Representations
+
+- **ID:** `2601.07870` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Michal Jan Wlodarczyk, Danzel Serrano, Przemyslaw Musialski
+- **abs:** https://arxiv.org/abs/2601.07870 · **PDF:** https://arxiv.org/pdf/2601.07870v1
+
+## 22675. Beyond Perfect Scores: Proof-by-Contradiction for Trustworthy Machine Learning
+
+- **ID:** `2601.06704` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Dushan N. Wadduwage, Dineth Jayakody, Leonidas Zimianitis
+- **abs:** https://arxiv.org/abs/2601.06704 · **PDF:** https://arxiv.org/pdf/2601.06704v1
+
+## 22676. GRASP LoRA: GRPO Guided Adapter Sparsity Policy for Cross Lingual Transfer
+
+- **ID:** `2601.06702` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Besher Hassan, Xiuying Chen
+- **abs:** https://arxiv.org/abs/2601.06702 · **PDF:** https://arxiv.org/pdf/2601.06702v1
+
+## 22677. Explainability of Complex AI Models with Correlation Impact Ratio
+
+- **ID:** `2601.06701` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Poushali Sengupta, Rabindra Khadka, Sabita Maharjan, et al.
+- **abs:** https://arxiv.org/abs/2601.06701 · **PDF:** https://arxiv.org/pdf/2601.06701v1
+
+## 22678. Characterising Toxicity in Generative Large Language Models
+
+- **ID:** `2601.06700` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Zhiyao Zhang, Yazan Mash'Al, Yuhan Wu
+- **abs:** https://arxiv.org/abs/2601.06700 · **PDF:** https://arxiv.org/pdf/2601.06700v1
+
+## 22679. Plasticity vs. Rigidity: The Impact of Low-Rank Adapters on Reasoning on a Micro-Budget
+
+- **ID:** `2601.06677` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Zohaib Khan, Omer Tafveez, Zoha Hayat Bhatti
+- **abs:** https://arxiv.org/abs/2601.06677 · **PDF:** https://arxiv.org/pdf/2601.06677v1
+
+## 22680. IDRBench: Interactive Deep Research Benchmark
+
+- **ID:** `2601.06676` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Yingchaojie Feng, Qiang Huang, Xiaoya Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.06676 · **PDF:** https://arxiv.org/pdf/2601.06676v1
+
+## 22681. Evaluating Cross-Lingual Unlearning in Multilingual Language Models
+
+- **ID:** `2601.06675` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Tyler Lizzo, Larry Heck
+- **abs:** https://arxiv.org/abs/2601.06675 · **PDF:** https://arxiv.org/pdf/2601.06675v1
+
+## 22682. Will it Merge? On The Causes of Model Mergeability
+
+- **ID:** `2601.06672` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Adir Rahamim, Asaf Yehudai, Boaz Carmeli, et al.
+- **abs:** https://arxiv.org/abs/2601.06672 · **PDF:** https://arxiv.org/pdf/2601.06672v1
+
+## 22683. Otimizando A Alocação De Salas De Aula Com Foco Na Acessibilidade Para Pessoas Com Deficiência
+
+- **ID:** `2601.06670` · **primary:** `cs.CY` · **Submitted:** 2026-01-10
+- **Authors:** Francisco Glaubos Nunes Clímaco, Jorge Lucas Silva Cavalcante
+- **abs:** https://arxiv.org/abs/2601.06670 · **PDF:** https://arxiv.org/pdf/2601.06670v1
+
+## 22684. InFi-Check: Interpretable and Fine-Grained Fact-Checking of LLMs
+
+- **ID:** `2601.06666` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Yuzhuo Bai, Shuzheng Si, Kangyang Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.06666 · **PDF:** https://arxiv.org/pdf/2601.06666v1
+
+## 22685. RewriteNets: End-to-End Trainable String-Rewriting for Generative Sequence Modeling
+
+- **ID:** `2601.07868` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Harshil Vejendla
+- **abs:** https://arxiv.org/abs/2601.07868 · **PDF:** https://arxiv.org/pdf/2601.07868v1
+
+## 22686. Reinforcement Learning-Guided Dynamic Multi-Graph Fusion for Evacuation Traffic Prediction
+
+- **ID:** `2601.06664` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Md Nafees Fuad Rafi, Samiul Hasan
+- **abs:** https://arxiv.org/abs/2601.06664 · **PDF:** https://arxiv.org/pdf/2601.06664v1
+
+## 22687. SafePro: Evaluating the Safety of Professional-Level AI Agents
+
+- **ID:** `2601.06663` · **primary:** `cs.AI` · **Submitted:** 2026-01-10
+- **Authors:** Kaiwen Zhou, Shreedhar Jangam, Ashwin Nagarajan, et al.
+- **abs:** https://arxiv.org/abs/2601.06663 · **PDF:** https://arxiv.org/pdf/2601.06663v2
+
+## 22688. What makes for an enjoyable protagonist? An analysis of character warmth and competence
+
+- **ID:** `2601.06658` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Hannes Rosenbusch
+- **abs:** https://arxiv.org/abs/2601.06658 · **PDF:** https://arxiv.org/pdf/2601.06658v1
+
+## 22689. Physics-constrained Gaussian Processes for Predicting Shockwave Hugoniot Curves
+
+- **ID:** `2601.06655` · **primary:** `cs.CE` · **Submitted:** 2026-01-10
+- **Authors:** George D. Pasparakis, Himanshu Sharma, Rushik Desai, et al.
+- **abs:** https://arxiv.org/abs/2601.06655 · **PDF:** https://arxiv.org/pdf/2601.06655v1
+
+## 22690. Revisiting Training Scale: An Empirical Study of Token Count, Power Consumption, and Parameter Efficiency
+
+- **ID:** `2601.06649` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Joe Dwyer
+- **abs:** https://arxiv.org/abs/2601.06649 · **PDF:** https://arxiv.org/pdf/2601.06649v1
+
+## 22691. Quality or Quantity? Error-Informed Selective Online Learning with Gaussian Processes in Multi-Agent Systems: Extended Version
+
+- **ID:** `2601.14275` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Zewen Yang, Xiaobing Dai, Jiajun Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.14275 · **PDF:** https://arxiv.org/pdf/2601.14275v1
+
+## 22692. A Multimodal Deep Learning Framework for Predicting ICU Deterioration: Integrating ECG Waveforms with Clinical Data and Clinician Benchmarking
+
+- **ID:** `2601.06645` · **primary:** `eess.SP` · **Submitted:** 2026-01-10
+- **Authors:** Juan Miguel López Alcaraz, Xicoténcatl López Moran, Erick Dávila Zaragoza, et al.
+- **abs:** https://arxiv.org/abs/2601.06645 · **PDF:** https://arxiv.org/pdf/2601.06645v1
+
+## 22693. Do Language Models Reason Across Languages?
+
+- **ID:** `2601.06644` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Yan Meng, Wafaa Mohammed, Christof Monz
+- **abs:** https://arxiv.org/abs/2601.06644 · **PDF:** https://arxiv.org/pdf/2601.06644v1
+
+## 22694. Boosting Overlapping Organoid Instance Segmentation Using Pseudo-Label Unmixing and Synthesis-Assisted Learning
+
+- **ID:** `2601.06642` · **primary:** `cs.CV` · **Submitted:** 2026-01-10
+- **Authors:** Gui Huang, Kangyuan Zheng, Xuan Cai, et al.
+- **abs:** https://arxiv.org/abs/2601.06642 · **PDF:** https://arxiv.org/pdf/2601.06642v1
+
+## 22695. Leveraging Soft Prompts for Privacy Attacks in Federated Prompt Tuning
+
+- **ID:** `2601.06641` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Quan Minh Nguyen, Min-Seon Kim, Hoang M. Ngo, et al.
+- **abs:** https://arxiv.org/abs/2601.06641 · **PDF:** https://arxiv.org/pdf/2601.06641v2
+
+## 22696. Agentic AI Empowered Intent-Based Networking for 6G
+
+- **ID:** `2601.06640` · **primary:** `cs.AI` · **Submitted:** 2026-01-10
+- **Authors:** Genze Jiang, Kezhi Wang, Xiaomin Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.06640 · **PDF:** https://arxiv.org/pdf/2601.06640v1
+
+## 22697. Attack-Resistant Watermarking for AIGC Image Forensics via Diffusion-based Semantic Deflection
+
+- **ID:** `2601.06639` · **primary:** `cs.CR` · **Submitted:** 2026-01-10
+- **Authors:** Qingyu Liu, Yitao Zhang, Zhongjie Ba, et al.
+- **abs:** https://arxiv.org/abs/2601.06639 · **PDF:** https://arxiv.org/pdf/2601.06639v1
+
+## 22698. Efficient Aspect Term Extraction using Spiking Neural Network
+
+- **ID:** `2601.06637` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Abhishek Kumar Mishra, Arya Somasundaram, Anup Das, et al.
+- **abs:** https://arxiv.org/abs/2601.06637 · **PDF:** https://arxiv.org/pdf/2601.06637v1
+
+## 22699. MedEinst: Benchmarking the Einstellung Effect in Medical LLMs through Counterfactual Differential Diagnosis
+
+- **ID:** `2601.06636` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Wenting Chen, Zhongrui Zhu, Guolin Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.06636 · **PDF:** https://arxiv.org/pdf/2601.06636v1
+
+## 22700. KASER: Knowledge-Aligned Student Error Simulator for Open-Ended Coding Tasks
+
+- **ID:** `2601.06633` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Zhangqi Duan, Nigel Fernandez, Andrew Lan
+- **abs:** https://arxiv.org/abs/2601.06633 · **PDF:** https://arxiv.org/pdf/2601.06633v1
+
+## 22701. Labels have Human Values: Value Calibration of Subjective Tasks
+
+- **ID:** `2601.06631` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Mohammed Fayiz Parappan, Ricardo Henao
+- **abs:** https://arxiv.org/abs/2601.06631 · **PDF:** https://arxiv.org/pdf/2601.06631v1
+
+## 22702. Lower Bounds for the Algorithmic Complexity of Learned Indexes
+
+- **ID:** `2601.06629` · **primary:** `cs.DS` · **Submitted:** 2026-01-10
+- **Authors:** Luis Alberto Croquevielle, Roman Sokolovskii, Thomas Heinis
+- **abs:** https://arxiv.org/abs/2601.06629 · **PDF:** https://arxiv.org/pdf/2601.06629v1
+
+## 22703. Burn-After-Use for Preventing Data Leakage through a Secure Multi-Tenant Architecture in Enterprise LLM
+
+- **ID:** `2601.06627` · **primary:** `cs.CR` · **Submitted:** 2026-01-10
+- **Authors:** Qiang Zhang, Elena Emma Wang, Jiaming Li, et al.
+- **abs:** https://arxiv.org/abs/2601.06627 · **PDF:** https://arxiv.org/pdf/2601.06627v3
+
+## 22704. Informed Consent for AI Consciousness Research: A Talmudic Framework for Graduated Protections
+
+- **ID:** `2601.08864` · **primary:** `cs.CY` · **Submitted:** 2026-01-10
+- **Authors:** Ira Wolfson
+- **abs:** https://arxiv.org/abs/2601.08864 · **PDF:** https://arxiv.org/pdf/2601.08864v1
+
+## 22705. Efficient and Reliable Estimation of Named Entity Linking Quality: A Case Study on GutBrainIE
+
+- **ID:** `2601.06624` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Marco Martinelli, Stefano Marchesin, Gianmaria Silvello
+- **abs:** https://arxiv.org/abs/2601.06624 · **PDF:** https://arxiv.org/pdf/2601.06624v1
+
+## 22706. Cross-Border Data Security and Privacy Risks in Large Language Models and IoT Systems
+
+- **ID:** `2601.06612` · **primary:** `cs.CR` · **Submitted:** 2026-01-10
+- **Authors:** Chalitha Handapangoda
+- **abs:** https://arxiv.org/abs/2601.06612 · **PDF:** https://arxiv.org/pdf/2601.06612v1
+
+## 22707. Pragya: An AI-Based Semantic Recommendation System for Sanskrit Subhasitas
+
+- **ID:** `2601.06607` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Tanisha Raorane, Prasenjit Kole
+- **abs:** https://arxiv.org/abs/2601.06607 · **PDF:** https://arxiv.org/pdf/2601.06607v1
+
+## 22708. Bridging the Trust Gap: Clinician-Validated Hybrid Explainable AI for Maternal Health Risk Assessment in Bangladesh
+
+- **ID:** `2601.07866` · **primary:** `cs.AI` · **Submitted:** 2026-01-10
+- **Authors:** Farjana Yesmin, Nusrat Shirmin, Suraiya Shabnam Bristy
+- **abs:** https://arxiv.org/abs/2601.07866 · **PDF:** https://arxiv.org/pdf/2601.07866v1
+
+## 22709. CEDAR: Context Engineering for Agentic Data Science
+
+- **ID:** `2601.06606` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Rishiraj Saha Roy, Chris Hinze, Luzian Hahn, et al.
+- **abs:** https://arxiv.org/abs/2601.06606 · **PDF:** https://arxiv.org/pdf/2601.06606v1
+
+## 22710. Object-Centric World Models Meet Monte Carlo Tree Search
+
+- **ID:** `2601.06604` · **primary:** `cs.AI` · **Submitted:** 2026-01-10
+- **Authors:** Rodion Vakhitov, Leonid Ugadiarov, Aleksandr Panov
+- **abs:** https://arxiv.org/abs/2601.06604 · **PDF:** https://arxiv.org/pdf/2601.06604v1
+
+## 22711. N2N-GQA: Noise-to-Narrative for Graph-Based Table-Text Question Answering Using LLMs
+
+- **ID:** `2601.06603` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Mohamed Sharafath, Aravindh Annamalai, Ganesh Murugan, et al.
+- **abs:** https://arxiv.org/abs/2601.06603 · **PDF:** https://arxiv.org/pdf/2601.06603v1
+
+## 22712. UMLoc: Uncertainty-Aware Map-Constrained Inertial Localization with Quantified Bounds
+
+- **ID:** `2601.06602` · **primary:** `cs.RO` · **Submitted:** 2026-01-10
+- **Authors:** Mohammed S. Alharbi, Shinkyu Park
+- **abs:** https://arxiv.org/abs/2601.06602 · **PDF:** https://arxiv.org/pdf/2601.06602v1
+
+## 22713. Probing Multimodal Large Language Models on Cognitive Biases in Chinese Short-Video Misinformation
+
+- **ID:** `2601.06600` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Jen-tse Huang, Chang Chen, Shiyang Lai, et al.
+- **abs:** https://arxiv.org/abs/2601.06600 · **PDF:** https://arxiv.org/pdf/2601.06600v1
+
+## 22714. How Context Shapes Truth: Geometric Transformations of Statement-level Truth Representations in LLMs
+
+- **ID:** `2601.06599` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Shivam Adarsh, Maria Maistro, Christina Lioma
+- **abs:** https://arxiv.org/abs/2601.06599 · **PDF:** https://arxiv.org/pdf/2601.06599v1
+
+## 22715. Implicit bias as a Gauge correction: Theory and Inverse Design
+
+- **ID:** `2601.06597` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Nicola Aladrah, Emanuele Ballarin, Matteo Biagetti, et al.
+- **abs:** https://arxiv.org/abs/2601.06597 · **PDF:** https://arxiv.org/pdf/2601.06597v1
+
+## 22716. Are LLMs Vulnerable to Preference-Undermining Attacks (PUA)? A Factorial Analysis Methodology for Diagnosing the Trade-off between Preference Alignment and Real-World Validity
+
+- **ID:** `2601.06596` · **primary:** `cs.CR` · **Submitted:** 2026-01-10
+- **Authors:** Hongjun An, Yiliang Song, Jiangan Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.06596 · **PDF:** https://arxiv.org/pdf/2601.06596v1
+
+## 22717. Detecting LLM-Generated Text with Performance Guarantees
+
+- **ID:** `2601.06586` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Hongyi Zhou, Jin Zhu, Ying Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.06586 · **PDF:** https://arxiv.org/pdf/2601.06586v1
+
+## 22718. Softly Induced Functional Simplicity: Implications for Neural Network Generalisation, Robustness, and Distillation
+
+- **ID:** `2601.06584` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Maciej Glowacki
+- **abs:** https://arxiv.org/abs/2601.06584 · **PDF:** https://arxiv.org/pdf/2601.06584v2
+
+## 22719. Stylistic Evolution and LLM Neutrality in Singlish Language
+
+- **ID:** `2601.06580` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Linus Tze En Foo, Weihan Angela Ng, Wenkai Li, et al.
+- **abs:** https://arxiv.org/abs/2601.06580 · **PDF:** https://arxiv.org/pdf/2601.06580v1
+
+## 22720. NoiseFormer -- Noise Diffused Symmetric Attention Transformer
+
+- **ID:** `2601.11619` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Phani Kumar, Nyshadham, Jyothendra Varma, et al.
+- **abs:** https://arxiv.org/abs/2601.11619 · **PDF:** https://arxiv.org/pdf/2601.11619v1
+
+## 22721. Are Emotions Arranged in a Circle? Geometric Analysis of Emotion Representations via Hyperspherical Contrastive Learning
+
+- **ID:** `2601.06575` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Yusuke Yamauchi, Akiko Aizawa
+- **abs:** https://arxiv.org/abs/2601.06575 · **PDF:** https://arxiv.org/pdf/2601.06575v2
+
+## 22722. Geometric Attention: A Regime-Explicit Operator Semantics for Transformer Attention
+
+- **ID:** `2601.11618` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Luis Rosario Freytes
+- **abs:** https://arxiv.org/abs/2601.11618 · **PDF:** https://arxiv.org/pdf/2601.11618v1
+
+## 22723. QMAVIS: Long Video-Audio Understanding using Fusion of Large Multimodal Models
+
+- **ID:** `2601.06573` · **primary:** `cs.AI` · **Submitted:** 2026-01-10
+- **Authors:** Zixing Lin, Jiale Wang, Gee Wah Ng, et al.
+- **abs:** https://arxiv.org/abs/2601.06573 · **PDF:** https://arxiv.org/pdf/2601.06573v1
+
+## 22724. Hellinger Multimodal Variational Autoencoders
+
+- **ID:** `2601.06572` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Huyen Vo, Isabel Valera
+- **abs:** https://arxiv.org/abs/2601.06572 · **PDF:** https://arxiv.org/pdf/2601.06572v3
+
+## 22725. QCaption: Video Captioning and Q&A through Fusion of Large Multimodal Models
+
+- **ID:** `2601.06566` · **primary:** `cs.CV` · **Submitted:** 2026-01-10
+- **Authors:** Jiale Wang, Gee Wah Ng, Lee Onn Mak, et al.
+- **abs:** https://arxiv.org/abs/2601.06566 · **PDF:** https://arxiv.org/pdf/2601.06566v1
+
+## 22726. EVM-QuestBench: An Execution-Grounded Benchmark for Natural-Language Transaction Code Generation
+
+- **ID:** `2601.06565` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Pei Yang, Wanyi Chen, Ke Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.06565 · **PDF:** https://arxiv.org/pdf/2601.06565v5
+
+## 22727. Scalable Board Expansion within a General Game System
+
+- **ID:** `2601.16216` · **primary:** `cs.AI` · **Submitted:** 2026-01-10
+- **Authors:** Clémentine Sacré
+- **abs:** https://arxiv.org/abs/2601.16216 · **PDF:** https://arxiv.org/pdf/2601.16216v1
+
+## 22728. CSR-RAG: An Efficient Retrieval System for Text-to-SQL on the Enterprise Scale
+
+- **ID:** `2601.06564` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Rajpreet Singh, Novak Boškov, Lawrence Drabeck, et al.
+- **abs:** https://arxiv.org/abs/2601.06564 · **PDF:** https://arxiv.org/pdf/2601.06564v1
+
+## 22729. Mosaic: Unlocking Long-Context Inference for Diffusion LLMs via Global Memory Planning and Dynamic Peak Taming
+
+- **ID:** `2601.06562` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Liang Zheng, Bowen Shi, Yitao Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.06562 · **PDF:** https://arxiv.org/pdf/2601.06562v1
+
+## 22730. Line-based Event Preprocessing: Towards Low-Energy Neuromorphic Computer Vision
+
+- **ID:** `2601.10742` · **primary:** `cs.NE` · **Submitted:** 2026-01-10
+- **Authors:** Amélie Gruel, Pierre Lewden, Adrien F. Vincent, et al.
+- **abs:** https://arxiv.org/abs/2601.10742 · **PDF:** https://arxiv.org/pdf/2601.10742v1
+
+## 22731. L-RAG: Balancing Context and Retrieval with Entropy-Based Lazy Loading
+
+- **ID:** `2601.06551` · **primary:** `cs.IR` · **Submitted:** 2026-01-10
+- **Authors:** Sergii Voloshyn
+- **abs:** https://arxiv.org/abs/2601.06551 · **PDF:** https://arxiv.org/pdf/2601.06551v1
+
+## 22732. LLMTrack: Semantic Multi-Object Tracking with Multi-modal Large Language Models
+
+- **ID:** `2601.06550` · **primary:** `cs.CV` · **Submitted:** 2026-01-10
+- **Authors:** Pan Liao, Feng Yang, Di Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.06550 · **PDF:** https://arxiv.org/pdf/2601.06550v2
+
+## 22733. SimLLM: Fine-Tuning Code LLMs for SimPy-Based Queueing System Simulation
+
+- **ID:** `2601.06543` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Jun-Qi Chen, Kun Zhang, Rui Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.06543 · **PDF:** https://arxiv.org/pdf/2601.06543v1
+
+## 22734. Resource-constrained Project Scheduling with Time-of-Use Energy Tariffs and Machine States: A Logic-based Benders Decomposition Approach
+
+- **ID:** `2601.06542` · **primary:** `math.OC` · **Submitted:** 2026-01-10
+- **Authors:** Corentin Juvigny, Antonín Novák, Jan Mandík, et al.
+- **abs:** https://arxiv.org/abs/2601.06542 · **PDF:** https://arxiv.org/pdf/2601.06542v1
+
+## 22735. Self-Organizing Dual-Buffer Adaptive Clustering Experience Replay (SODASER) for Safe Reinforcement Learning in Optimal Control
+
+- **ID:** `2601.06540` · **primary:** `eess.SY` · **Submitted:** 2026-01-10
+- **Authors:** Roya Khalili Amirabadi, Mohsen Jalaeian Farimani, Omid Solaymani Fard
+- **abs:** https://arxiv.org/abs/2601.06540 · **PDF:** https://arxiv.org/pdf/2601.06540v1
+
+## 22736. Exposía: Academic Writing Assessment of Exposés and Peer Feedback
+
+- **ID:** `2601.06536` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Dennis Zyska, Alla Rozovskaya, Ilia Kuznetsov, et al.
+- **abs:** https://arxiv.org/abs/2601.06536 · **PDF:** https://arxiv.org/pdf/2601.06536v1
+
+## 22737. Short-term electricity load forecasting with multi-frequency reconstruction diffusion
+
+- **ID:** `2601.06533` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Qi Dong, Rubing Huang, Ling Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.06533 · **PDF:** https://arxiv.org/pdf/2601.06533v1
+
+## 22738. Improving Day-Ahead Grid Carbon Intensity Forecasting by Joint Modeling of Local-Temporal and Cross-Variable Dependencies Across Different Frequencies
+
+- **ID:** `2601.06530` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Bowen Zhang, Hongda Tian, Adam Berry, et al.
+- **abs:** https://arxiv.org/abs/2601.06530 · **PDF:** https://arxiv.org/pdf/2601.06530v1
+
+## 22739. Atomic-SNLI: Fine-Grained Natural Language Inference through Atomic Fact Decomposition
+
+- **ID:** `2601.06528` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Minghui Huang
+- **abs:** https://arxiv.org/abs/2601.06528 · **PDF:** https://arxiv.org/pdf/2601.06528v1
+
+## 22740. BabyVision: Visual Reasoning Beyond Language
+
+- **ID:** `2601.06521` · **primary:** `cs.CV` · **Submitted:** 2026-01-10
+- **Authors:** Liang Chen, Weichu Xie, Yiyan Liang, et al.
+- **abs:** https://arxiv.org/abs/2601.06521 · **PDF:** https://arxiv.org/pdf/2601.06521v1
+
+## 22741. MedRAGChecker: Claim-Level Verification for Biomedical Retrieval-Augmented Generation
+
+- **ID:** `2601.06519` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Yuelyu Ji, Min Gu Kwak, Hang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.06519 · **PDF:** https://arxiv.org/pdf/2601.06519v1
+
+## 22742. Pareto-Optimal Model Selection for Low-Cost, Single-Lead EMG Control in Embedded Systems
+
+- **ID:** `2601.06516` · **primary:** `cs.HC` · **Submitted:** 2026-01-10
+- **Authors:** Carl Vincent Ladres Kho
+- **abs:** https://arxiv.org/abs/2601.06516 · **PDF:** https://arxiv.org/pdf/2601.06516v1
+
+## 22743. Inference-Time Alignment for Diffusion Models via Variationally Stable Doob's Matching
+
+- **ID:** `2601.06514` · **primary:** `stat.ML` · **Submitted:** 2026-01-10
+- **Authors:** Jinyuan Chang, Chenguang Duan, Yuling Jiao, et al.
+- **abs:** https://arxiv.org/abs/2601.06514 · **PDF:** https://arxiv.org/pdf/2601.06514v2
+
+## 22744. A novel RF-enabled Non-Destructive Inspection Method through Machine Learning and Programmable Wireless Environments
+
+- **ID:** `2601.06512` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Stavros Tsimpoukis, Dimitrios Tyrovolas, Sotiris Ioannidis, et al.
+- **abs:** https://arxiv.org/abs/2601.06512 · **PDF:** https://arxiv.org/pdf/2601.06512v1
+
+## 22745. Neural Nonmyopic Bayesian Optimization in Dynamic Cost Settings
+
+- **ID:** `2601.06505` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Sang T. Truong, Duc Q. Nguyen, Willie Neiswanger, et al.
+- **abs:** https://arxiv.org/abs/2601.06505 · **PDF:** https://arxiv.org/pdf/2601.06505v1
+
+## 22746. DRAGON: LLM-Driven Decomposition and Reconstruction Agents for Large-Scale Combinatorial Optimization
+
+- **ID:** `2601.06502` · **primary:** `cs.AI` · **Submitted:** 2026-01-10
+- **Authors:** Shengkai Chen, Zhiguang Cao, Jianan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.06502 · **PDF:** https://arxiv.org/pdf/2601.06502v2
+
+## 22747. The AI Pyramid A Conceptual Framework for Workforce Capability in the Age of AI
+
+- **ID:** `2601.06500` · **primary:** `cs.AI` · **Submitted:** 2026-01-10
+- **Authors:** Alok Khatri, Bishesh Khanal
+- **abs:** https://arxiv.org/abs/2601.06500 · **PDF:** https://arxiv.org/pdf/2601.06500v2
+
+## 22748. Spec-o3: A Tool-Augmented Vision-Language Agent for Rare Celestial Object Candidate Vetting via Automated Spectral Inspection
+
+- **ID:** `2601.06498` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Minghui Jia, Qichao Zhang, Ali Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.06498 · **PDF:** https://arxiv.org/pdf/2601.06498v1
+
+## 22749. Coding in a Bubble? Evaluating LLMs in Resolving Context Adaptation Bugs During Code Adaptation
+
+- **ID:** `2601.06497` · **primary:** `cs.SE` · **Submitted:** 2026-01-10
+- **Authors:** Tanghaoran Zhang, Xinjun Mao, Shangwen Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.06497 · **PDF:** https://arxiv.org/pdf/2601.06497v1
+
+## 22750. ArenaRL: Scaling RL for Open-Ended Agents via Tournament-based Relative Ranking
+
+- **ID:** `2601.06487` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Qiang Zhang, Boli Chen, Fanrui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.06487 · **PDF:** https://arxiv.org/pdf/2601.06487v2
+
+## 22751. Learning Domain Agnostic Latent Embeddings of 3D Faces for Zero-shot Animal Expression Transfer
+
+- **ID:** `2601.06484` · **primary:** `cs.CV` · **Submitted:** 2026-01-10
+- **Authors:** Yue Wang, Lawrence Amadi, Xiang Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.06484 · **PDF:** https://arxiv.org/pdf/2601.06484v1
+
+## 22752. Deriving Decoder-Free Sparse Autoencoders from First Principles
+
+- **ID:** `2601.06478` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Alan Oursland
+- **abs:** https://arxiv.org/abs/2601.06478 · **PDF:** https://arxiv.org/pdf/2601.06478v1
+
+## 22753. IndRegBias: A Dataset for Studying Indian Regional Biases in English and Code-Mixed Social Media Comments
+
+- **ID:** `2601.06477` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Debasmita Panda, Akash Anil, Neelesh Kumar Shukla
+- **abs:** https://arxiv.org/abs/2601.06477 · **PDF:** https://arxiv.org/pdf/2601.06477v2
+
+## 22754. VVTRec: Radio Interferometric Reconstruction through Visual and Textual Modality Enrichment
+
+- **ID:** `2601.06475` · **primary:** `cs.CV` · **Submitted:** 2026-01-10
+- **Authors:** Kai Cheng, Ruoqi Wang, Qiong Luo
+- **abs:** https://arxiv.org/abs/2601.06475 · **PDF:** https://arxiv.org/pdf/2601.06475v1
+
+## 22755. SparseOccVLA: Bridging Occupancy and Vision-Language Models via Sparse Queries for Unified 4D Scene Understanding and Planning
+
+- **ID:** `2601.06474` · **primary:** `cs.CV` · **Submitted:** 2026-01-10
+- **Authors:** Chenxu Dang, Jie Wang, Guang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.06474 · **PDF:** https://arxiv.org/pdf/2601.06474v2
+
+## 22756. Hybrid LSTM-UKF Framework: Ankle Angle and Ground Reaction Force Estimation
+
+- **ID:** `2601.06473` · **primary:** `eess.SY` · **Submitted:** 2026-01-10
+- **Authors:** Mundla Narasimhappa, Praveen Kumar
+- **abs:** https://arxiv.org/abs/2601.06473 · **PDF:** https://arxiv.org/pdf/2601.06473v1
+
+## 22757. StablePDENet: Enhancing Stability of Operator Learning for Solving Differential Equations
+
+- **ID:** `2601.06472` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Chutian Huang, Chang Ma, Kaibo Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.06472 · **PDF:** https://arxiv.org/pdf/2601.06472v1
+
+## 22758. PRISP: Privacy-Safe Few-Shot Personalization via Lightweight Adaptation
+
+- **ID:** `2601.06471` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Junho Park, Dohoon Kim, Taesup Moon
+- **abs:** https://arxiv.org/abs/2601.06471 · **PDF:** https://arxiv.org/pdf/2601.06471v1
+
+## 22759. Divide and Refine: Enhancing Multimodal Representation and Explainability for Emotion Recognition in Conversation
+
+- **ID:** `2601.14274` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Anh-Tuan Mai, Cam-Van Thi Nguyen, Duc-Trong Le
+- **abs:** https://arxiv.org/abs/2601.14274 · **PDF:** https://arxiv.org/pdf/2601.14274v1
+
+## 22760. Gecko: An Efficient Neural Architecture Inherently Processing Sequences with Arbitrary Lengths
+
+- **ID:** `2601.06463` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Xuezhe Ma, Shicheng Wen, Linghao Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.06463 · **PDF:** https://arxiv.org/pdf/2601.06463v1
+
+## 22761. Physics-informed Gaussian Process Regression in Solving Eigenvalue Problem of Linear Operators
+
+- **ID:** `2601.06462` · **primary:** `stat.ML` · **Submitted:** 2026-01-10
+- **Authors:** Tianming Bai, Jiannan Yang
+- **abs:** https://arxiv.org/abs/2601.06462 · **PDF:** https://arxiv.org/pdf/2601.06462v1
+
+## 22762. Tone Matters: The Impact of Linguistic Tone on Hallucination in VLMs
+
+- **ID:** `2601.06460` · **primary:** `cs.CV` · **Submitted:** 2026-01-10
+- **Authors:** Weihao Hong, Zhiyuan Jiang, Bingyu Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.06460 · **PDF:** https://arxiv.org/pdf/2601.06460v1
+
+## 22763. PixRec: Leveraging Visual Context for Next-Item Prediction in Sequential Recommendation
+
+- **ID:** `2601.06458` · **primary:** `cs.IR` · **Submitted:** 2026-01-10
+- **Authors:** Sayak Chakrabarty, Souradip Pal
+- **abs:** https://arxiv.org/abs/2601.06458 · **PDF:** https://arxiv.org/pdf/2601.06458v1
+
+## 22764. ConSensus: Multi-Agent Collaboration for Multimodal Sensing
+
+- **ID:** `2601.06453` · **primary:** `cs.AI` · **Submitted:** 2026-01-10
+- **Authors:** Hyungjun Yoon, Mohammad Malekzadeh, Sung-Ju Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.06453 · **PDF:** https://arxiv.org/pdf/2601.06453v1
+
+## 22765. LitVISTA: A Benchmark for Narrative Orchestration in Literary Text
+
+- **ID:** `2601.06445` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Mingzhe Lu, Yiwen Wang, Yanbing Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.06445 · **PDF:** https://arxiv.org/pdf/2601.06445v1
+
+## 22766. Physics-Informed Tree Search for High-Dimensional Computational Design
+
+- **ID:** `2601.06444` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Suvo Banik, Troy D. Loeffler, Henry Chan, et al.
+- **abs:** https://arxiv.org/abs/2601.06444 · **PDF:** https://arxiv.org/pdf/2601.06444v1
+
+## 22767. A Methodology for Identifying Evaluation Items for Practical Dialogue Systems Based on Business-Dialogue System Alignment Models
+
+- **ID:** `2602.15835` · **primary:** `cs.HC` · **Submitted:** 2026-01-10
+- **Authors:** Mikio Nakano, Hironori Takeuchi, Kazunori Komatani
+- **abs:** https://arxiv.org/abs/2602.15835 · **PDF:** https://arxiv.org/pdf/2602.15835v1
+
+## 22768. FlexAct: Why Learn when you can Pick?
+
+- **ID:** `2601.06441` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Ramnath Kumar, Kyle Ritscher, Junmin Judy, et al.
+- **abs:** https://arxiv.org/abs/2601.06441 · **PDF:** https://arxiv.org/pdf/2601.06441v1
+
+## 22769. Time Travel Engine: A Shared Latent Chronological Manifold Enables Historical Navigation in Large Language Models
+
+- **ID:** `2601.06437` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Jingmin An, Wei Liu, Qian Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.06437 · **PDF:** https://arxiv.org/pdf/2601.06437v1
+
+## 22770. Certified Unlearning in Decentralized Federated Learning
+
+- **ID:** `2601.06436` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Hengliang Wu, Youming Tao, Anhao Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.06436 · **PDF:** https://arxiv.org/pdf/2601.06436v1
+
+## 22771. On a Gradient Approach to Chebyshev Center Problems with Applications to Function Learning
+
+- **ID:** `2601.06434` · **primary:** `math.OC` · **Submitted:** 2026-01-10
+- **Authors:** Abhinav Raghuvanshi, Mayank Baranwal, Debasish Chatterjee
+- **abs:** https://arxiv.org/abs/2601.06434 · **PDF:** https://arxiv.org/pdf/2601.06434v1
+
+## 22772. LSRIF: Logic-Structured Reinforcement Learning for Instruction Following
+
+- **ID:** `2601.06431` · **primary:** `cs.AI` · **Submitted:** 2026-01-10
+- **Authors:** Qingyu Ren, Qianyu He, Jingwen Chang, et al.
+- **abs:** https://arxiv.org/abs/2601.06431 · **PDF:** https://arxiv.org/pdf/2601.06431v2
+
+## 22773. A Unified Shape-Aware Foundation Model for Time Series Classification
+
+- **ID:** `2601.06429` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Zhen Liu, Yucheng Wang, Boyuan Li, et al.
+- **abs:** https://arxiv.org/abs/2601.06429 · **PDF:** https://arxiv.org/pdf/2601.06429v1
+
+## 22774. BackPlay: Plug-in Look-Back Self-Correction for Diffusion Language Models
+
+- **ID:** `2601.06428` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Liming Liu, Binxuan Huang, Zixuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.06428 · **PDF:** https://arxiv.org/pdf/2601.06428v2
+
+## 22775. NC-Bench: An LLM Benchmark for Evaluating Conversational Competence
+
+- **ID:** `2601.06426` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Robert J. Moore, Sungeun An, Farhan Ahmed, et al.
+- **abs:** https://arxiv.org/abs/2601.06426 · **PDF:** https://arxiv.org/pdf/2601.06426v2
+
+## 22776. HiDVFS: A Hierarchical Multi-Agent DVFS Scheduler for OpenMP DAG Workloads
+
+- **ID:** `2601.06425` · **primary:** `cs.DC` · **Submitted:** 2026-01-10
+- **Authors:** Mohammad Pivezhandi, Abusayeed Saifullah, Ali Jannesari
+- **abs:** https://arxiv.org/abs/2601.06425 · **PDF:** https://arxiv.org/pdf/2601.06425v1
+
+## 22777. Can a Unimodal Language Agent Provide Preferences to Tune a Multimodal Vision-Language Model?
+
+- **ID:** `2601.06424` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Sazia Tabasum Mim, Jack Morris, Manish Dhakal, et al.
+- **abs:** https://arxiv.org/abs/2601.06424 · **PDF:** https://arxiv.org/pdf/2601.06424v1
+
+## 22778. Does Inference Scaling Improve Reasoning Faithfulness? A Multi-Model Analysis of Self-Consistency Tradeoffs
+
+- **ID:** `2601.06423` · **primary:** `cs.AI` · **Submitted:** 2026-01-10
+- **Authors:** Deep Mehta
+- **abs:** https://arxiv.org/abs/2601.06423 · **PDF:** https://arxiv.org/pdf/2601.06423v1
+
+## 22779. Attribution Techniques for Mitigating Hallucinated Information in RAG Systems: A Survey
+
+- **ID:** `2601.19927` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Yuqing Zhao, Ziyao Liu, Yongsen Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.19927 · **PDF:** https://arxiv.org/pdf/2601.19927v1
+
+## 22780. EmbeddingRWKV: State-Centric Retrieval with Reusable States
+
+- **ID:** `2601.07861` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Haowen Hou, Jie Yang
+- **abs:** https://arxiv.org/abs/2601.07861 · **PDF:** https://arxiv.org/pdf/2601.07861v1
+
+## 22781. Semantic Enrichment of CAD-Based Industrial Environments via Scene Graphs for Simulation and Reasoning
+
+- **ID:** `2601.06415` · **primary:** `cs.RO` · **Submitted:** 2026-01-10
+- **Authors:** Nathan Pascal Walus, Ranulfo Bezerra, Shotaro Kojima, et al.
+- **abs:** https://arxiv.org/abs/2601.06415 · **PDF:** https://arxiv.org/pdf/2601.06415v1
+
+## 22782. Structured Episodic Event Memory
+
+- **ID:** `2601.06411` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Zhengxuan Lu, Dongfang Li, Yukun Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.06411 · **PDF:** https://arxiv.org/pdf/2601.06411v2
+
+## 22783. Value of Information: A Framework for Human-Agent Communication
+
+- **ID:** `2601.06407` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Yijiang River Dong, Tiancheng Hu, Zheng Hui, et al.
+- **abs:** https://arxiv.org/abs/2601.06407 · **PDF:** https://arxiv.org/pdf/2601.06407v1
+
+## 22784. One-Shot Hierarchical Federated Clustering
+
+- **ID:** `2601.06404` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Shenghong Cai, Zihua Yang, Yang Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.06404 · **PDF:** https://arxiv.org/pdf/2601.06404v1
+
+## 22785. Steer Model beyond Assistant: Controlling System Prompt Strength via Contrastive Decoding
+
+- **ID:** `2601.06403` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Yijiang River Dong, Tiancheng Hu, Zheng Hui, et al.
+- **abs:** https://arxiv.org/abs/2601.06403 · **PDF:** https://arxiv.org/pdf/2601.06403v1
+
+## 22786. BizFinBench.v2: A Unified Dual-Mode Bilingual Benchmark for Expert-Level Financial Capability Alignment
+
+- **ID:** `2601.06401` · **primary:** `cs.AI` · **Submitted:** 2026-01-10
+- **Authors:** Xin Guo, Rongjunchen Zhang, Guilong Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.06401 · **PDF:** https://arxiv.org/pdf/2601.06401v1
+
+## 22787. Neuro-Symbolic Activation Discovery: Transferring Mathematical Structures from Physics to Ecology for Parameter-Efficient Neural Networks
+
+- **ID:** `2601.10740` · **primary:** `cs.NE` · **Submitted:** 2026-01-10
+- **Authors:** Anas Hajbi
+- **abs:** https://arxiv.org/abs/2601.10740 · **PDF:** https://arxiv.org/pdf/2601.10740v1
+
+## 22788. MITRA: A Large-Scale Parallel Corpus and Multilingual Pretrained Language Model for Machine Translation and Semantic Retrieval for Pāli, Sanskrit, Buddhist Chinese, and Tibetan
+
+- **ID:** `2601.06400` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Sebastian Nehrdich, Kurt Keutzer
+- **abs:** https://arxiv.org/abs/2601.06400 · **PDF:** https://arxiv.org/pdf/2601.06400v1
+
+## 22789. AfriqueLLM: How Data Mixing and Model Architecture Impact Continued Pre-training for African Languages
+
+- **ID:** `2601.06395` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Hao Yu, Tianyi Xu, Michael A. Hedderich, et al.
+- **abs:** https://arxiv.org/abs/2601.06395 · **PDF:** https://arxiv.org/pdf/2601.06395v1
+
+## 22790. Context Matters: Peer-Aware Student Behavioral Engagement Measurement via VLM Action Parsing and LLM Sequence Classification
+
+- **ID:** `2601.06394` · **primary:** `cs.CV` · **Submitted:** 2026-01-10
+- **Authors:** Ahmed Abdelkawy, Ahmed Elsayed, Asem Ali, et al.
+- **abs:** https://arxiv.org/abs/2601.06394 · **PDF:** https://arxiv.org/pdf/2601.06394v2
+
+## 22791. Continual Quantum Architecture Search with Tensor-Train Encoding: Theory and Applications to Signal Processing
+
+- **ID:** `2601.06392` · **primary:** `quant-ph` · **Submitted:** 2026-01-10
+- **Authors:** Jun Qi, Chao-Han Huck Yang, Pin-Yu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.06392 · **PDF:** https://arxiv.org/pdf/2601.06392v1
+
+## 22792. Supervised and Unsupervised Neural Network Solver for First Order Hyperbolic Nonlinear PDEs
+
+- **ID:** `2601.06388` · **primary:** `math.NA` · **Submitted:** 2026-01-10
+- **Authors:** Zakaria Baba, Alexandre M. Bayen, Alexi Canesse, et al.
+- **abs:** https://arxiv.org/abs/2601.06388 · **PDF:** https://arxiv.org/pdf/2601.06388v1
+
+## 22793. An Efficient Evolutionary Algorithm for Few-for-Many Optimization
+
+- **ID:** `2601.06387` · **primary:** `cs.NE` · **Submitted:** 2026-01-10
+- **Authors:** Ke Shang, Hisao Ishibuchi, Zexuan Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.06387 · **PDF:** https://arxiv.org/pdf/2601.06387v1
+
+## 22794. Hierarchical Pooling and Explainability in Graph Neural Networks for Tumor and Tissue-of-Origin Classification Using RNA-seq Data
+
+- **ID:** `2601.06381` · **primary:** `cs.LG` · **Submitted:** 2026-01-10
+- **Authors:** Thomas Vaitses Fontanari, Mariana Recamonde-Mendoza
+- **abs:** https://arxiv.org/abs/2601.06381 · **PDF:** https://arxiv.org/pdf/2601.06381v1
+
+## 22795. HiMem: Hierarchical Long-Term Memory for LLM Long-Horizon Agents
+
+- **ID:** `2601.06377` · **primary:** `cs.AI` · **Submitted:** 2026-01-10
+- **Authors:** Ningning Zhang, Xingxing Yang, Zhizhong Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.06377 · **PDF:** https://arxiv.org/pdf/2601.06377v1
+
+## 22796. Talking to Extraordinary Objects: Folktales Offer Analogies for Interacting with Technology
+
+- **ID:** `2601.06372` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Martha Larson
+- **abs:** https://arxiv.org/abs/2601.06372 · **PDF:** https://arxiv.org/pdf/2601.06372v1
+
+## 22797. SafeGPT: Preventing Data Leakage and Unethical Outputs in Enterprise LLM Use
+
+- **ID:** `2601.06366` · **primary:** `cs.CR` · **Submitted:** 2026-01-10
+- **Authors:** Pratyush Desai, Luoxi Tang, Yuqiao Meng, et al.
+- **abs:** https://arxiv.org/abs/2601.06366 · **PDF:** https://arxiv.org/pdf/2601.06366v1
+
+## 22798. Human-in-the-Loop Interactive Report Generation for Chronic Disease Adherence
+
+- **ID:** `2601.06364` · **primary:** `cs.HC` · **Submitted:** 2026-01-10
+- **Authors:** Xiaotian Zhang, Jinhong Yu, Pengwei Yan, et al.
+- **abs:** https://arxiv.org/abs/2601.06364 · **PDF:** https://arxiv.org/pdf/2601.06364v1
+
+## 22799. Styles + Persona-plug = Customized LLMs
+
+- **ID:** `2601.06362` · **primary:** `cs.AI` · **Submitted:** 2026-01-10
+- **Authors:** Yutong Song, Jiang Wu, Shaofan Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.06362 · **PDF:** https://arxiv.org/pdf/2601.06362v1
+
+## 22800. Average shortest-path length in word-adjacency networks: Chinese versus English
+
+- **ID:** `2601.06361` · **primary:** `cs.CL` · **Submitted:** 2026-01-10
+- **Authors:** Jakub Dec, Michał Dolina, Stanisław Drożdż, et al.
+- **abs:** https://arxiv.org/abs/2601.06361 · **PDF:** https://arxiv.org/pdf/2601.06361v1
+
+## 22801. Computational Mapping of Reactive Stroma in Prostate Cancer Yields Interpretable, Prognostic Biomarkers
+
+- **ID:** `2601.06360` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-10
+- **Authors:** Mara Pleasure, Ekaterina Redekop, Dhakshina Ilango, et al.
+- **abs:** https://arxiv.org/abs/2601.06360 · **PDF:** https://arxiv.org/pdf/2601.06360v1
+
+## 22802. Smart Privacy Policy Assistant: An LLM-Powered System for Transparent and Actionable Privacy Notices
+
+- **ID:** `2601.06357` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Sriharshini Kalvakuntla, Luoxi Tang, Yuqiao Meng, et al.
+- **abs:** https://arxiv.org/abs/2601.06357 · **PDF:** https://arxiv.org/pdf/2601.06357v1
+
+## 22803. Monkey Jump : MoE-Style PEFT for Efficient Multi-Task Learning
+
+- **ID:** `2601.06356` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Nusrat Jahan Prottasha, Md Kowsher, Chun-Nam Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.06356 · **PDF:** https://arxiv.org/pdf/2601.06356v1
+
+## 22804. CARD: Cluster-level Adaptation with Reward-guided Decoding for Personalized Text Generation
+
+- **ID:** `2601.06352` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Yutong Song, Jiang Wu, Weijia Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.06352 · **PDF:** https://arxiv.org/pdf/2601.06352v1
+
+## 22805. A Fast and Effective Method for Euclidean Anticlustering: The Assignment-Based-Anticlustering Algorithm
+
+- **ID:** `2601.06351` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Philipp Baumann, Olivier Goldschmidt, Dorit S. Hochbaum, et al.
+- **abs:** https://arxiv.org/abs/2601.06351 · **PDF:** https://arxiv.org/pdf/2601.06351v1
+
+## 22806. Mixture-of-Experts as Soft Clustering: A Dual Jacobian-PCA Spectral Geometry Perspective
+
+- **ID:** `2601.11616` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Feilong Liu
+- **abs:** https://arxiv.org/abs/2601.11616 · **PDF:** https://arxiv.org/pdf/2601.11616v2
+
+## 22807. Federated Learning and Class Imbalances
+
+- **ID:** `2601.06348` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Siqi Zhu, Joshua D. Kaggie
+- **abs:** https://arxiv.org/abs/2601.06348 · **PDF:** https://arxiv.org/pdf/2601.06348v1
+
+## 22808. A Review on Machine Learning Approaches for the Prediction of Glucose Levels and Hypogylcemia
+
+- **ID:** `2601.11615` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Beyza Cinar, Louisa van den Boom, Maria Maleshkova
+- **abs:** https://arxiv.org/abs/2601.11615 · **PDF:** https://arxiv.org/pdf/2601.11615v1
+
+## 22809. What Matters When Building Universal Multilingual Named Entity Recognition Models?
+
+- **ID:** `2601.06347` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Jonas Golde, Patrick Haller, Alan Akbik
+- **abs:** https://arxiv.org/abs/2601.06347 · **PDF:** https://arxiv.org/pdf/2601.06347v1
+
+## 22810. Multi-modal MRI-Based Alzheimer's Disease Diagnosis with Transformer-based Image Synthesis and Transfer Learning
+
+- **ID:** `2601.11614` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Jason Qiu
+- **abs:** https://arxiv.org/abs/2601.11614 · **PDF:** https://arxiv.org/pdf/2601.11614v1
+
+## 22811. Evaluating Robustness of Large Language Models in Enterprise Applications: Benchmarks for Perturbation Consistency Across Formats and Languages
+
+- **ID:** `2601.06341` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Tara Bogavelli, Oluwanifemi Bamgbose, Gabrielle Gauthier Melançon, et al.
+- **abs:** https://arxiv.org/abs/2601.06341 · **PDF:** https://arxiv.org/pdf/2601.06341v1
+
+## 22812. Circuit Mechanisms for Spatial Relation Generation in Diffusion Transformers
+
+- **ID:** `2601.06338` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Binxu Wang, Jingxuan Fan, Xu Pan
+- **abs:** https://arxiv.org/abs/2601.06338 · **PDF:** https://arxiv.org/pdf/2601.06338v1
+
+## 22813. Future-as-Label: Scalable Supervision from Real-World Outcomes
+
+- **ID:** `2601.06336` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Benjamin Turtel, Paul Wilczewski, Danny Franklin, et al.
+- **abs:** https://arxiv.org/abs/2601.06336 · **PDF:** https://arxiv.org/pdf/2601.06336v2
+
+## 22814. Foundational Analysis of Safety Engineering Requirements (SAFER)
+
+- **ID:** `2601.06335` · **primary:** `cs.SE` · **Submitted:** 2026-01-09
+- **Authors:** Noga Chemo, Yaniv Mordecai, Yoram Reich
+- **abs:** https://arxiv.org/abs/2601.06335 · **PDF:** https://arxiv.org/pdf/2601.06335v1
+
+## 22815. Kolmogorov-Arnold Networks-Based Tolerance-Aware Manufacturability Assessment Integrating Design-for-Manufacturing Principles
+
+- **ID:** `2601.06334` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Masoud Deylami, Negar Izadipour, Adel Alaeddini
+- **abs:** https://arxiv.org/abs/2601.06334 · **PDF:** https://arxiv.org/pdf/2601.06334v1
+
+## 22816. On the Fallacy of Global Token Perplexity in Spoken Language Model Evaluation
+
+- **ID:** `2601.06329` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Jeff Chan-Jan Sju, Liang-Hsuan Tseng, Yi-Cheng Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.06329 · **PDF:** https://arxiv.org/pdf/2601.06329v1
+
+## 22817. ToolGym: an Open-world Tool-using Environment for Scalable Agent Testing and Data Curation
+
+- **ID:** `2601.06328` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Ziqiao Xi, Shuang Liang, Qi Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.06328 · **PDF:** https://arxiv.org/pdf/2601.06328v1
+
+## 22818. SourceNet: Interpretable Sim-to-Real Inference on Variable-Geometry Sensor Arrays for Earthquake Source Inversion
+
+- **ID:** `2601.06320` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Zhe Jia, Xiaotian Zhang, Junpeng Li
+- **abs:** https://arxiv.org/abs/2601.06320 · **PDF:** https://arxiv.org/pdf/2601.06320v2
+
+## 22819. Random is Faster than Systematic in Multi-Objective Local Search
+
+- **ID:** `2601.06318` · **primary:** `cs.NE` · **Submitted:** 2026-01-09
+- **Authors:** Zimin Liang, Miqing Li
+- **abs:** https://arxiv.org/abs/2601.06318 · **PDF:** https://arxiv.org/pdf/2601.06318v1
+
+## 22820. Annotating Dimensions of Social Perception in Text: The First Sentence-Level Dataset of Warmth and Competence
+
+- **ID:** `2601.06316` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Mutaz Ayesh, Saif M. Mohammad, Nedjma Ousidhoum
+- **abs:** https://arxiv.org/abs/2601.06316 · **PDF:** https://arxiv.org/pdf/2601.06316v1
+
+## 22821. VideoWeave: A Data-Centric Approach for Efficient Video Understanding
+
+- **ID:** `2601.06309` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Zane Durante, Silky Singh, Arpandeep Khatua, et al.
+- **abs:** https://arxiv.org/abs/2601.06309 · **PDF:** https://arxiv.org/pdf/2601.06309v1
+
+## 22822. A Rising Tide Lifts All Boats: MTQE Rewards for Idioms Improve General Translation Quality
+
+- **ID:** `2601.06307` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Ishika Agarwal, Zhenlin He, Dhruva Patil, et al.
+- **abs:** https://arxiv.org/abs/2601.06307 · **PDF:** https://arxiv.org/pdf/2601.06307v1
+
+## 22823. SyntaxMind at BLP-2025 Task 1: Leveraging Attention Fusion of CNN and GRU for Hate Speech Detection
+
+- **ID:** `2601.06306` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Md. Shihab Uddin Riad
+- **abs:** https://arxiv.org/abs/2601.06306 · **PDF:** https://arxiv.org/pdf/2601.06306v1
+
+## 22824. Why LoRA Fails to Forget: Regularized Low-Rank Adaptation Against Backdoors in Language Models
+
+- **ID:** `2601.06305` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Hoang-Chau Luong, Lingwei Chen
+- **abs:** https://arxiv.org/abs/2601.06305 · **PDF:** https://arxiv.org/pdf/2601.06305v1
+
+## 22825. Beyond BeautifulSoup: Benchmarking LLM-Powered Web Scraping for Everyday Users
+
+- **ID:** `2601.06301` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Arth Bhardwaj, Nirav Diwan, Gang Wang
+- **abs:** https://arxiv.org/abs/2601.06301 · **PDF:** https://arxiv.org/pdf/2601.06301v1
+
+## 22826. $\texttt{AMEND++}$: Benchmarking Eligibility Criteria Amendments in Clinical Trials
+
+- **ID:** `2601.06300` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Trisha Das, Mandis Beigi, Jacob Aptekar, et al.
+- **abs:** https://arxiv.org/abs/2601.06300 · **PDF:** https://arxiv.org/pdf/2601.06300v1
+
+## 22827. A Targeted Learning Framework for Estimating Restricted Mean Survival Time Difference using Pseudo-observations
+
+- **ID:** `2601.06296` · **primary:** `stat.ME` · **Submitted:** 2026-01-09
+- **Authors:** Man Jin, Yixin Fang
+- **abs:** https://arxiv.org/abs/2601.06296 · **PDF:** https://arxiv.org/pdf/2601.06296v2
+
+## 22828. How well can off-the-shelf LLMs elucidate molecular structures from mass spectra using chain-of-thought reasoning?
+
+- **ID:** `2601.06289` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Yufeng Wang, Lu Wei, Lin Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.06289 · **PDF:** https://arxiv.org/pdf/2601.06289v1
+
+## 22829. AIConfigurator: Lightning-Fast Configuration Optimization for Multi-Framework LLM Serving
+
+- **ID:** `2601.06288` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Tianhao Xu, Yiming Liu, Xianglong Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.06288 · **PDF:** https://arxiv.org/pdf/2601.06288v1
+
+## 22830. Amory: Building Coherent Narrative-Driven Agent Memory through Agentic Reasoning
+
+- **ID:** `2601.06282` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Yue Zhou, Xiaobo Guo, Belhassen Bayar, et al.
+- **abs:** https://arxiv.org/abs/2601.06282 · **PDF:** https://arxiv.org/pdf/2601.06282v1
+
+## 22831. Automated QoR improvement in OpenROAD with coding agents
+
+- **ID:** `2601.06268` · **primary:** `cs.SE` · **Submitted:** 2026-01-09
+- **Authors:** Amur Ghose, Junyeong Jang, Andrew B. Kahng, et al.
+- **abs:** https://arxiv.org/abs/2601.06268 · **PDF:** https://arxiv.org/pdf/2601.06268v2
+
+## 22832. Matrix Factorization Framework for Community Detection under the Degree-Corrected Block Model
+
+- **ID:** `2601.06262` · **primary:** `cs.SI` · **Submitted:** 2026-01-09
+- **Authors:** Alexandra Dache, Arnaud Vandaele, Nicolas Gillis
+- **abs:** https://arxiv.org/abs/2601.06262 · **PDF:** https://arxiv.org/pdf/2601.06262v1
+
+## 22833. Gamma2Patterns: Deep Cognitive Attention Region Identification and Gamma-Alpha Pattern Analysis
+
+- **ID:** `2601.06257` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-09
+- **Authors:** Sobhana Jahan, Saydul Akbar Murad, Nick Rahimi, et al.
+- **abs:** https://arxiv.org/abs/2601.06257 · **PDF:** https://arxiv.org/pdf/2601.06257v1
+
+## 22834. Manifold limit for the training of shallow graph convolutional neural networks
+
+- **ID:** `2601.06025` · **primary:** `stat.ML` · **Submitted:** 2026-01-09
+- **Authors:** Johanna Tengler, Christoph Brune, José A. Iglesias
+- **abs:** https://arxiv.org/abs/2601.06025 · **PDF:** https://arxiv.org/pdf/2601.06025v1
+
+## 22835. AdaFuse: Adaptive Ensemble Decoding with Test-Time Scaling for LLMs
+
+- **ID:** `2601.06022` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Chengming Cui, Tianxin Wei, Ziyi Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.06022 · **PDF:** https://arxiv.org/pdf/2601.06022v1
+
+## 22836. Chaining the Evidence: Robust Reinforcement Learning for Deep Search Agents with Citation-Aware Rubric Rewards
+
+- **ID:** `2601.06021` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Jiajie Zhang, Xin Lv, Ling Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.06021 · **PDF:** https://arxiv.org/pdf/2601.06021v1
+
+## 22837. LookAroundNet: Extending Temporal Context with Transformers for Clinically Viable EEG Seizure Detection
+
+- **ID:** `2601.06016` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Þór Sverrisson, Steinn Guðmundsson
+- **abs:** https://arxiv.org/abs/2601.06016 · **PDF:** https://arxiv.org/pdf/2601.06016v1
+
+## 22838. Detecting Stochasticity in Discrete Signals via Nonparametric Excursion Theorem
+
+- **ID:** `2601.06009` · **primary:** `stat.ML` · **Submitted:** 2026-01-09
+- **Authors:** Sunia Tanweer, Firas A. Khasawneh
+- **abs:** https://arxiv.org/abs/2601.06009 · **PDF:** https://arxiv.org/pdf/2601.06009v1
+
+## 22839. Don't Break the Cache: An Evaluation of Prompt Caching for Long-Horizon Agentic Tasks
+
+- **ID:** `2601.06007` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Elias Lumer, Faheem Nizar, Akshaya Jangiti, et al.
+- **abs:** https://arxiv.org/abs/2601.06007 · **PDF:** https://arxiv.org/pdf/2601.06007v2
+
+## 22840. The Molecular Structure of Thought: Mapping the Topology of Long Chain-of-Thought Reasoning
+
+- **ID:** `2601.06002` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Qiguang Chen, Yantao Du, Ziniu Li, et al.
+- **abs:** https://arxiv.org/abs/2601.06002 · **PDF:** https://arxiv.org/pdf/2601.06002v2
+
+## 22841. Hard Constraint Projection in a Physics Informed Neural Network
+
+- **ID:** `2601.06244` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-01-09
+- **Authors:** Miranda J. S. Horne, Peter K. Jimack, Amirul Khan, et al.
+- **abs:** https://arxiv.org/abs/2601.06244 · **PDF:** https://arxiv.org/pdf/2601.06244v1
+
+## 22842. Open-Vocabulary 3D Instruction Ambiguity Detection
+
+- **ID:** `2601.05991` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Jiayu Ding, Haoran Tang, Ge Li
+- **abs:** https://arxiv.org/abs/2601.05991 · **PDF:** https://arxiv.org/pdf/2601.05991v1
+
+## 22843. CyberGFM: Graph Foundation Models for Lateral Movement Detection in Enterprise Networks
+
+- **ID:** `2601.05988` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Isaiah J. King, Bernardo Trindade, Benjamin Bowman, et al.
+- **abs:** https://arxiv.org/abs/2601.05988 · **PDF:** https://arxiv.org/pdf/2601.05988v1
+
+## 22844. Community-Based Model Sharing and Generalisation: Anomaly Detection in IoT Temperature Sensor Networks
+
+- **ID:** `2601.05984` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Sahibzada Saadoon Hammad, Joaquín Huerta Guijarro, Francisco Ramos, et al.
+- **abs:** https://arxiv.org/abs/2601.05984 · **PDF:** https://arxiv.org/pdf/2601.05984v1
+
+## 22845. Hybrid Secure Routing in Mobile Ad-hoc Networks (MANETSs)
+
+- **ID:** `2602.13204` · **primary:** `cs.NI` · **Submitted:** 2026-01-09
+- **Authors:** Soundes Oumaima Boufaida, Abdemadjid Benmachiche, Majda Maatallah, et al.
+- **abs:** https://arxiv.org/abs/2602.13204 · **PDF:** https://arxiv.org/pdf/2602.13204v1
+
+## 22846. Real-Time Image Processing Algorithms for Embedded Systems
+
+- **ID:** `2601.06243` · **primary:** `eess.IV` · **Submitted:** 2026-01-09
+- **Authors:** Soundes Oumaima Boufaida, Abdemadjid Benmachiche, Majda Maatallah
+- **abs:** https://arxiv.org/abs/2601.06243 · **PDF:** https://arxiv.org/pdf/2601.06243v1
+
+## 22847. AWaRe-SAC: Proactive Slice Admission Control under Weather-Induced Capacity Uncertainty
+
+- **ID:** `2601.05978` · **primary:** `cs.NI` · **Submitted:** 2026-01-09
+- **Authors:** Dror Jacoby, Yanzhi Li, Shuyue Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.05978 · **PDF:** https://arxiv.org/pdf/2601.05978v2
+
+## 22848. DeePM: Regime-Robust Deep Learning for Systematic Macro Portfolio Management
+
+- **ID:** `2601.05975` · **primary:** `q-fin.TR` · **Submitted:** 2026-01-09
+- **Authors:** Kieran Wood, Stephen J. Roberts, Stefan Zohren
+- **abs:** https://arxiv.org/abs/2601.05975 · **PDF:** https://arxiv.org/pdf/2601.05975v1
+
+## 22849. VideoAR: Autoregressive Video Generation via Next-Frame & Scale Prediction
+
+- **ID:** `2601.05966` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Longbin Ji, Xiaoxiong Liu, Junyuan Shang, et al.
+- **abs:** https://arxiv.org/abs/2601.05966 · **PDF:** https://arxiv.org/pdf/2601.05966v2
+
+## 22850. A Framework for Low-Latency, LLM-driven Multimodal Interaction on the Pepper Robot
+
+- **ID:** `2603.21013` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Erich Studerus, Vivienne Jia Zhong, Stephan Vonschallen
+- **abs:** https://arxiv.org/abs/2603.21013 · **PDF:** https://arxiv.org/pdf/2603.21013v1
+
+## 22851. Distilling Feedback into Memory-as-a-Tool
+
+- **ID:** `2601.05960` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Víctor Gallego
+- **abs:** https://arxiv.org/abs/2601.05960 · **PDF:** https://arxiv.org/pdf/2601.05960v2
+
+## 22852. On the Robustness of Age for Learning-Based Wireless Scheduling in Unknown Environments
+
+- **ID:** `2601.05956` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Juaren Steiger, Bin Li
+- **abs:** https://arxiv.org/abs/2601.05956 · **PDF:** https://arxiv.org/pdf/2601.05956v2
+
+## 22853. Affect and Effect: Limitations of regularisation-based continual learning in EEG-based emotion classification
+
+- **ID:** `2601.07858` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Nina Peire, Yupei Li, Björn Schuller
+- **abs:** https://arxiv.org/abs/2601.07858 · **PDF:** https://arxiv.org/pdf/2601.07858v1
+
+## 22854. A Critical Examination of Active Learning Workflows in Materials Science
+
+- **ID:** `2601.05946` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-09
+- **Authors:** Akhil S. Nair, Lucas Foppa
+- **abs:** https://arxiv.org/abs/2601.05946 · **PDF:** https://arxiv.org/pdf/2601.05946v1
+
+## 22855. Agentic AI Microservice Framework for Deepfake and Document Fraud Detection in KYC Pipelines
+
+- **ID:** `2601.06241` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Chandra Sekhar Kubam
+- **abs:** https://arxiv.org/abs/2601.06241 · **PDF:** https://arxiv.org/pdf/2601.06241v1
+
+## 22856. Performance of a Deep Learning-Based Segmentation Model for Pancreatic Tumors on Public Endoscopic Ultrasound Datasets
+
+- **ID:** `2601.05937` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Pankaj Gupta, Priya Mudgil, Niharika Dutta, et al.
+- **abs:** https://arxiv.org/abs/2601.05937 · **PDF:** https://arxiv.org/pdf/2601.05937v1
+
+## 22857. Can We Predict Before Executing Machine Learning Agents?
+
+- **ID:** `2601.05930` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Jingsheng Zheng, Jintian Zhang, Yujie Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.05930 · **PDF:** https://arxiv.org/pdf/2601.05930v1
+
+## 22858. Prophet as a Reproducible Forecasting Framework: A Methodological Guide for Business and Financial Analytics
+
+- **ID:** `2601.05929` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Sidney Shapiro, Burhanuddin Panvelwala
+- **abs:** https://arxiv.org/abs/2601.05929 · **PDF:** https://arxiv.org/pdf/2601.05929v2
+
+## 22859. Cedalion Tutorial: A Python-based framework for comprehensive analysis of multimodal fNIRS & DOT from the lab to the everyday world
+
+- **ID:** `2601.05923` · **primary:** `eess.SP` · **Submitted:** 2026-01-09
+- **Authors:** E. Middell, L. Carlton, S. Moradi, et al.
+- **abs:** https://arxiv.org/abs/2601.05923 · **PDF:** https://arxiv.org/pdf/2601.05923v1
+
+## 22860. The Grammar of Transformers: A Systematic Review of Interpretability Research on Syntactic Knowledge in Language Models
+
+- **ID:** `2601.19926` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Nora Graichen, Iria de-Dios-Flores, Gemma Boleda
+- **abs:** https://arxiv.org/abs/2601.19926 · **PDF:** https://arxiv.org/pdf/2601.19926v1
+
+## 22861. Agentic LLMs as Powerful Deanonymizers: Re-identification of Participants in the Anthropic Interviewer Dataset
+
+- **ID:** `2601.05918` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Tianshi Li
+- **abs:** https://arxiv.org/abs/2601.05918 · **PDF:** https://arxiv.org/pdf/2601.05918v1
+
+## 22862. Distilling Lightweight Domain Experts from Large ML Models by Identifying Relevant Subspaces
+
+- **ID:** `2601.05913` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Pattarawat Chormai, Ali Hashemi, Klaus-Robert Müller, et al.
+- **abs:** https://arxiv.org/abs/2601.05913 · **PDF:** https://arxiv.org/pdf/2601.05913v1
+
+## 22863. Pantagruel: Unified Self-Supervised Encoders for French Text and Speech
+
+- **ID:** `2601.05911` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Phuong-Hang Le, Valentin Pelloin, Arnault Chatelain, et al.
+- **abs:** https://arxiv.org/abs/2601.05911 · **PDF:** https://arxiv.org/pdf/2601.05911v2
+
+## 22864. Multi-task Modeling for Engineering Applications with Sparse Data
+
+- **ID:** `2601.05910` · **primary:** `stat.ML` · **Submitted:** 2026-01-09
+- **Authors:** Yigitcan Comlek, R. Murali Krishnan, Sandipp Krishnan Ravi, et al.
+- **abs:** https://arxiv.org/abs/2601.05910 · **PDF:** https://arxiv.org/pdf/2601.05910v1
+
+## 22865. Auditing Fairness under Model Updates: Fundamental Complexity and Property-Preserving Updates
+
+- **ID:** `2601.05909` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Ayoub Ajarra, Debabrota Basu
+- **abs:** https://arxiv.org/abs/2601.05909 · **PDF:** https://arxiv.org/pdf/2601.05909v1
+
+## 22866. Illusions of Confidence? Diagnosing LLM Truthfulness via Neighborhood Consistency
+
+- **ID:** `2601.05905` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Haoming Xu, Ningyuan Zhao, Yunzhi Yao, et al.
+- **abs:** https://arxiv.org/abs/2601.05905 · **PDF:** https://arxiv.org/pdf/2601.05905v1
+
+## 22867. Can AI mediation improve democratic deliberation?
+
+- **ID:** `2601.05904` · **primary:** `cs.CY` · **Submitted:** 2026-01-09
+- **Authors:** Michael Henry Tessler, Georgina Evans, Michiel A. Bakker, et al.
+- **abs:** https://arxiv.org/abs/2601.05904 · **PDF:** https://arxiv.org/pdf/2601.05904v1
+
+## 22868. HAPS: Hierarchical LLM Routing with Joint Architecture and Parameter Search
+
+- **ID:** `2601.05903` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Zihang Tian, Rui Li, Jingsen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.05903 · **PDF:** https://arxiv.org/pdf/2601.05903v1
+
+## 22869. TowerMind: A Tower Defence Game Learning Environment and Benchmark for LLM as Agents
+
+- **ID:** `2601.05899` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Dawei Wang, Chengming Zhou, Di Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.05899 · **PDF:** https://arxiv.org/pdf/2601.05899v1
+
+## 22870. StackPlanner: A Centralized Hierarchical Multi-Agent System with Task-Experience Memory Management
+
+- **ID:** `2601.05890` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Ruizhe Zhang, Xinke Jiang, Zhibang Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.05890 · **PDF:** https://arxiv.org/pdf/2601.05890v1
+
+## 22871. GlueNN: gluing patchwise analytic solutions with neural networks
+
+- **ID:** `2601.05889` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Doyoung Kim, Donghee Lee, Hye-Sung Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.05889 · **PDF:** https://arxiv.org/pdf/2601.05889v2
+
+## 22872. An Empirical Study on Preference Tuning Generalization and Diversity Under Domain Shift
+
+- **ID:** `2601.05882` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Constantinos Karouzos, Xingwei Tan, Nikolaos Aletras
+- **abs:** https://arxiv.org/abs/2601.05882 · **PDF:** https://arxiv.org/pdf/2601.05882v1
+
+## 22873. Gender Bias in LLMs: Preliminary Evidence from Shared Parenting Scenario in Czech Family Law
+
+- **ID:** `2601.05879` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Jakub Harasta, Matej Vasina, Martin Kornel, et al.
+- **abs:** https://arxiv.org/abs/2601.05879 · **PDF:** https://arxiv.org/pdf/2601.05879v1
+
+## 22874. iReasoner: Trajectory-Aware Intrinsic Reasoning Supervision for Self-Evolving Large Multimodal Models
+
+- **ID:** `2601.05877` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Meghana Sunil, Manikandarajan Venmathimaran, Muthu Subash Kavitha
+- **abs:** https://arxiv.org/abs/2601.05877 · **PDF:** https://arxiv.org/pdf/2601.05877v2
+
+## 22875. Continual-learning for Modelling Low-Resource Languages from Large Language Models
+
+- **ID:** `2601.05874` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Santosh Srinath K, Mudit Somani, Varun Reddy Padala, et al.
+- **abs:** https://arxiv.org/abs/2601.05874 · **PDF:** https://arxiv.org/pdf/2601.05874v1
+
+## 22876. IIB-LPO: Latent Policy Optimization via Iterative Information Bottleneck
+
+- **ID:** `2601.05870` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Huilin Deng, Hongchen Luo, Yue Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.05870 · **PDF:** https://arxiv.org/pdf/2601.05870v1
+
+## 22877. Sequential Bayesian Optimal Experimental Design in Infinite Dimensions via Policy Gradient Reinforcement Learning
+
+- **ID:** `2601.05868` · **primary:** `math.OC` · **Submitted:** 2026-01-09
+- **Authors:** Kaichen Shen, Peng Chen
+- **abs:** https://arxiv.org/abs/2601.05868 · **PDF:** https://arxiv.org/pdf/2601.05868v1
+
+## 22878. FACTUM: Mechanistic Detection of Citation Hallucination in Long-Form RAG
+
+- **ID:** `2601.05866` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Maxime Dassen, Rebecca Kotula, Kenton Murray, et al.
+- **abs:** https://arxiv.org/abs/2601.05866 · **PDF:** https://arxiv.org/pdf/2601.05866v4
+
+## 22879. What do the metrics mean? A critical analysis of the use of Automated Evaluation Metrics in Interpreting
+
+- **ID:** `2601.05864` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Jonathan Downie, Joss Moorkens
+- **abs:** https://arxiv.org/abs/2601.05864 · **PDF:** https://arxiv.org/pdf/2601.05864v1
+
+## 22880. CLewR: Curriculum Learning with Restarts for Machine Translation Preference Learning
+
+- **ID:** `2601.05858` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Alexandra Dragomir, Florin Brad, Radu Tudor Ionescu
+- **abs:** https://arxiv.org/abs/2601.05858 · **PDF:** https://arxiv.org/pdf/2601.05858v1
+
+## 22881. LayerGS: Decomposition and Inpainting of Layered 3D Human Avatars via 2D Gaussian Splatting
+
+- **ID:** `2601.05853` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Yinghan Xu, John Dingliana
+- **abs:** https://arxiv.org/abs/2601.05853 · **PDF:** https://arxiv.org/pdf/2601.05853v1
+
+## 22882. Router-Suggest: Dynamic Routing for Multimodal Auto-Completion in Visually-Grounded Dialogs
+
+- **ID:** `2601.05851` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Sandeep Mishra, Devichand Budagam, Anubhab Mandal, et al.
+- **abs:** https://arxiv.org/abs/2601.05851 · **PDF:** https://arxiv.org/pdf/2601.05851v1
+
+## 22883. Goal Force: Teaching Video Models To Accomplish Physics-Conditioned Goals
+
+- **ID:** `2601.05848` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Nate Gillman, Yinghua Zhou, Zitian Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.05848 · **PDF:** https://arxiv.org/pdf/2601.05848v2
+
+## 22884. Evaluating Large Language Models for Abstract Evaluation Tasks: An Empirical Study
+
+- **ID:** `2601.19925` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Yinuo Liu, Emre Sezgin, Eric A. Youngstrom
+- **abs:** https://arxiv.org/abs/2601.19925 · **PDF:** https://arxiv.org/pdf/2601.19925v1
+
+## 22885. Semantic NLP Pipelines for Interoperable Patient Digital Twins from Unstructured EHRs
+
+- **ID:** `2601.05847` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Rafael Brens, Yuqiao Meng, Luoxi Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.05847 · **PDF:** https://arxiv.org/pdf/2601.05847v1
+
+## 22886. A New Family of Poisson Non-negative Matrix Factorization Methods Using the Shifted Log Link
+
+- **ID:** `2601.05845` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Eric Weine, Peter Carbonetto, Rafael A. Irizarry, et al.
+- **abs:** https://arxiv.org/abs/2601.05845 · **PDF:** https://arxiv.org/pdf/2601.05845v1
+
+## 22887. DexterCap: An Affordable and Automated System for Capturing Dexterous Hand-Object Manipulation
+
+- **ID:** `2601.05844` · **primary:** `cs.GR` · **Submitted:** 2026-01-09
+- **Authors:** Yutong Liang, Shiyi Xu, Yulong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.05844 · **PDF:** https://arxiv.org/pdf/2601.05844v2
+
+## 22888. Adversarial Network Imagination: Causal LLMs and Digital Twins for Proactive Telecom Mitigation
+
+- **ID:** `2602.13203` · **primary:** `cs.NI` · **Submitted:** 2026-01-09
+- **Authors:** Vignesh Sriram, Yuqiao Meng, Luoxi Tang, et al.
+- **abs:** https://arxiv.org/abs/2602.13203 · **PDF:** https://arxiv.org/pdf/2602.13203v1
+
+## 22889. An Intelligent AI glasses System with Multi-Agent Architecture for Real-Time Voice Processing and Task Execution
+
+- **ID:** `2601.06235` · **primary:** `cs.SD` · **Submitted:** 2026-01-09
+- **Authors:** Sheng-Kai Chen, Jyh-Horng Wu, Ching-Yao Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.06235 · **PDF:** https://arxiv.org/pdf/2601.06235v1
+
+## 22890. Intelligent Singularity Avoidance in UR10 Robotic Arm Path Planning Using Hybrid Fuzzy Logic and Reinforcement Learning
+
+- **ID:** `2601.05836` · **primary:** `cs.RO` · **Submitted:** 2026-01-09
+- **Authors:** Sheng-Kai Chen, Jyh-Horng Wu
+- **abs:** https://arxiv.org/abs/2601.05836 · **PDF:** https://arxiv.org/pdf/2601.05836v1
+
+## 22891. Left, Right, or Center? Evaluating LLM Framing in News Classification and Generation
+
+- **ID:** `2601.05835` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Molly Kennedy, Ali Parker, Yihong Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.05835 · **PDF:** https://arxiv.org/pdf/2601.05835v1
+
+## 22892. Peek2: A Regex-free implementation of pretokenizers for Byte-level BPE
+
+- **ID:** `2601.05833` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Liu Zai
+- **abs:** https://arxiv.org/abs/2601.05833 · **PDF:** https://arxiv.org/pdf/2601.05833v1
+
+## 22893. PCoKG: Personality-aware Commonsense Reasoning with Debate
+
+- **ID:** `2601.06234` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Weijie Li, Zhongqing Wang, Guodong Zhou
+- **abs:** https://arxiv.org/abs/2601.06234 · **PDF:** https://arxiv.org/pdf/2601.06234v1
+
+## 22894. Influence of Parallelism in Vector-Multiplication Units on Correlation Power Analysis
+
+- **ID:** `2601.05828` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Manuel Brosch, Matthias Probst, Stefan Kögler, et al.
+- **abs:** https://arxiv.org/abs/2601.05828 · **PDF:** https://arxiv.org/pdf/2601.05828v1
+
+## 22895. Decoding Workload and Agreement From EEG During Spoken Dialogue With Conversational AI
+
+- **ID:** `2601.05825` · **primary:** `cs.HC` · **Submitted:** 2026-01-09
+- **Authors:** Lucija Mihić Zidar, Philipp Wicke, Praneel Bhatia, et al.
+- **abs:** https://arxiv.org/abs/2601.05825 · **PDF:** https://arxiv.org/pdf/2601.05825v2
+
+## 22896. LLMs as Science Journalists: Supporting Early-stage Researchers in Communicating Their Science to the Public
+
+- **ID:** `2601.05821` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Milad Alshomary, Grace Li, Anubhav Jangra, et al.
+- **abs:** https://arxiv.org/abs/2601.05821 · **PDF:** https://arxiv.org/pdf/2601.05821v1
+
+## 22897. A Dual Pipeline Machine Learning Framework for Automated Multi Class Sleep Disorder Screening Using Hybrid Resampling and Ensemble Learning
+
+- **ID:** `2601.05814` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Md Sultanul Islam Ovi, Muhsina Tarannum Munfa, G. M. M Miftahul Alam Adib, et al.
+- **abs:** https://arxiv.org/abs/2601.05814 · **PDF:** https://arxiv.org/pdf/2601.05814v2
+
+## 22898. Detecting Autism Spectrum Disorder with Deep Eye Movement Features
+
+- **ID:** `2601.05812` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Zhanpei Huang, Taochen chen, Fangqing Gu, et al.
+- **abs:** https://arxiv.org/abs/2601.05812 · **PDF:** https://arxiv.org/pdf/2601.05812v1
+
+## 22899. Learning Reconstructive Embeddings in Reproducing Kernel Hilbert Spaces via the Representer Theorem
+
+- **ID:** `2601.05811` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Enrique Feito-Casares, Francisco M. Melgarejo-Meseguer, José-Luis Rojo-Álvarez
+- **abs:** https://arxiv.org/abs/2601.05811 · **PDF:** https://arxiv.org/pdf/2601.05811v1
+
+## 22900. SceneFoundry: Generating Interactive Infinite 3D Worlds
+
+- **ID:** `2601.05810` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** ChunTeng Chen, YiChen Hsu, YiWen Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.05810 · **PDF:** https://arxiv.org/pdf/2601.05810v2
+
+## 22901. EnvScaler: Scaling Tool-Interactive Environments for LLM Agent via Programmatic Synthesis
+
+- **ID:** `2601.05808` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Xiaoshuai Song, Haofei Chang, Guanting Dong, et al.
+- **abs:** https://arxiv.org/abs/2601.05808 · **PDF:** https://arxiv.org/pdf/2601.05808v1
+
+## 22902. Fusion Matters: Length-Aware Analysis of Positional-Encoding Fusion in Transformers
+
+- **ID:** `2601.05807` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Mohamed Amine Hallam, Kuo-Kun Tseng
+- **abs:** https://arxiv.org/abs/2601.05807 · **PDF:** https://arxiv.org/pdf/2601.05807v1
+
+## 22903. Simplify-This: A Comparative Analysis of Prompt-Based and Fine-Tuned LLMs
+
+- **ID:** `2601.05794` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Eilam Cohen, Itamar Bul, Danielle Inbar, et al.
+- **abs:** https://arxiv.org/abs/2601.05794 · **PDF:** https://arxiv.org/pdf/2601.05794v1
+
+## 22904. Tensor-DTI: Enhancing Biomolecular Interaction Prediction with Contrastive Embedding Learning
+
+- **ID:** `2601.05792` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Manel Gil-Sorribes, Júlia Vilalta-Mor, Isaac Filella-Mercè, et al.
+- **abs:** https://arxiv.org/abs/2601.05792 · **PDF:** https://arxiv.org/pdf/2601.05792v1
+
+## 22905. SAFE: Secure and Accurate Federated Learning for Privacy-Preserving Brain-Computer Interfaces
+
+- **ID:** `2601.05789` · **primary:** `cs.HC` · **Submitted:** 2026-01-09
+- **Authors:** Tianwang Jia, Xiaoqing Chen, Dongrui Wu
+- **abs:** https://arxiv.org/abs/2601.05789 · **PDF:** https://arxiv.org/pdf/2601.05789v1
+
+## 22906. From Off-Policy to On-Policy: Enhancing GUI Agents via Bi-level Expert-to-Policy Assimilation
+
+- **ID:** `2601.05787` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Zezhou Wang, Ziyun Zhang, Xiaoyi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.05787 · **PDF:** https://arxiv.org/pdf/2601.05787v2
+
+## 22907. Adaptive Disentangled Representation Learning for Incomplete Multi-View Multi-Label Classification
+
+- **ID:** `2601.05785` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Quanjiang Li, Zhiming Liu, Tianxiang Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.05785 · **PDF:** https://arxiv.org/pdf/2601.05785v1
+
+## 22908. One Script Instead of Hundreds? On Pretraining Romanized Encoder Language Models
+
+- **ID:** `2601.05776` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Benedikt Ebing, Lennart Keller, Goran Glavaš
+- **abs:** https://arxiv.org/abs/2601.05776 · **PDF:** https://arxiv.org/pdf/2601.05776v1
+
+## 22909. Weights to Code: Extracting Interpretable Algorithms from the Discrete Transformer
+
+- **ID:** `2601.05770` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Yifan Zhang, Wei Bi, Kechi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.05770 · **PDF:** https://arxiv.org/pdf/2601.05770v2
+
+## 22910. Domain-Specific Self-Supervised Pre-training for Agricultural Disease Classification: A Hierarchical Vision Transformer Study
+
+- **ID:** `2601.11612` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Arnav S. Sonavane
+- **abs:** https://arxiv.org/abs/2601.11612 · **PDF:** https://arxiv.org/pdf/2601.11612v1
+
+## 22911. Multi-Agent Framework for Controllable and Protected Generative Content Creation: Addressing Copyright and Provenance in AI-Generated Media
+
+- **ID:** `2601.06232` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Haris Khan, Sadia Asif, Shumaila Asif
+- **abs:** https://arxiv.org/abs/2601.06232 · **PDF:** https://arxiv.org/pdf/2601.06232v1
+
+## 22912. Variational Autoencoders for P-wave Detection on Strong Motion Earthquake Spectrograms
+
+- **ID:** `2601.05759` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Turkan Simge Ispak, Salih Tileylioglu, Erdem Akagunduz
+- **abs:** https://arxiv.org/abs/2601.05759 · **PDF:** https://arxiv.org/pdf/2601.05759v1
+
+## 22913. VIGIL: Defending LLM Agents Against Tool Stream Injection via Verify-Before-Commit
+
+- **ID:** `2601.05755` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Junda Lin, Zhaomeng Zhou, Zhi Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.05755 · **PDF:** https://arxiv.org/pdf/2601.05755v2
+
+## 22914. AutoMonitor-Bench: Evaluating the Reliability of LLM-Based Misbehavior Monitor
+
+- **ID:** `2601.05752` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Shu Yang, Jingyu Hu, Tong Li, et al.
+- **abs:** https://arxiv.org/abs/2601.05752 · **PDF:** https://arxiv.org/pdf/2601.05752v2
+
+## 22915. Analysing Differences in Persuasive Language in LLM-Generated Text: Uncovering Stereotypical Gender Patterns
+
+- **ID:** `2601.05751` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Amalie Brogaard Pauli, Maria Barrett, Max Müller-Eberstein, et al.
+- **abs:** https://arxiv.org/abs/2601.05751 · **PDF:** https://arxiv.org/pdf/2601.05751v1
+
+## 22916. DynaDebate: Breaking Homogeneity in Multi-Agent Debate with Dynamic Path Generation
+
+- **ID:** `2601.05746` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Zhenghao Li, Zhi Zheng, Wei Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.05746 · **PDF:** https://arxiv.org/pdf/2601.05746v1
+
+## 22917. The Echo Chamber Multi-Turn LLM Jailbreak
+
+- **ID:** `2601.05742` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Ahmad Alobaid, Martí Jordà Roca, Carlos Castillo, et al.
+- **abs:** https://arxiv.org/abs/2601.05742 · **PDF:** https://arxiv.org/pdf/2601.05742v1
+
+## 22918. ViTNT-FIQA: Training-Free Face Image Quality Assessment with Vision Transformers
+
+- **ID:** `2601.05741` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Guray Ozgur, Eduarda Caldeira, Tahar Chettaoui, et al.
+- **abs:** https://arxiv.org/abs/2601.05741 · **PDF:** https://arxiv.org/pdf/2601.05741v1
+
+## 22919. PII-VisBench: Evaluating Personally Identifiable Information Safety in Vision Language Models Along a Continuum of Visibility
+
+- **ID:** `2601.05739` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** G M Shahariar, Zabir Al Nazi, Md Olid Hasan Bhuiyan, et al.
+- **abs:** https://arxiv.org/abs/2601.05739 · **PDF:** https://arxiv.org/pdf/2601.05739v1
+
+## 22920. Triadic Concept Analysis for Logic Interpretation of Simple Artificial Networks
+
+- **ID:** `2601.06229` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Ingo Schmitt
+- **abs:** https://arxiv.org/abs/2601.06229 · **PDF:** https://arxiv.org/pdf/2601.06229v1
+
+## 22921. mHC-lite: You Don't Need 20 Sinkhorn-Knopp Iterations
+
+- **ID:** `2601.05732` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Yongyi Yang, Jianyang Gao
+- **abs:** https://arxiv.org/abs/2601.05732 · **PDF:** https://arxiv.org/pdf/2601.05732v1
+
+## 22922. Overcoming Joint Intractability with Lossless Hierarchical Speculative Decoding
+
+- **ID:** `2601.05724` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Yuxuan Zhou, Fei Huang, Heng Li, et al.
+- **abs:** https://arxiv.org/abs/2601.05724 · **PDF:** https://arxiv.org/pdf/2601.05724v2
+
+## 22923. Synthetic FMCW Radar Range Azimuth Maps Augmentation with Generative Diffusion Model
+
+- **ID:** `2601.06228` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Zhaoze Wang, Changxu Zhang, Tai Fei, et al.
+- **abs:** https://arxiv.org/abs/2601.06228 · **PDF:** https://arxiv.org/pdf/2601.06228v1
+
+## 22924. Visualising Information Flow in Word Embeddings with Diffusion Tensor Imaging
+
+- **ID:** `2601.05713` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Thomas Fabian
+- **abs:** https://arxiv.org/abs/2601.05713 · **PDF:** https://arxiv.org/pdf/2601.05713v1
+
+## 22925. Multimodal In-context Learning for ASR of Low-resource Languages
+
+- **ID:** `2601.05707` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Zhaolin Li, Jan Niehues
+- **abs:** https://arxiv.org/abs/2601.05707 · **PDF:** https://arxiv.org/pdf/2601.05707v1
+
+## 22926. Logic-Parametric Neuro-Symbolic NLI: Controlling Logical Formalisms for Verifiable LLM Reasoning
+
+- **ID:** `2601.05705` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Ali Farjami, Luca Redondi, Marco Valentino
+- **abs:** https://arxiv.org/abs/2601.05705 · **PDF:** https://arxiv.org/pdf/2601.05705v1
+
+## 22927. AIBoMGen: Generating an AI Bill of Materials for Secure, Transparent, and Compliant Model Training
+
+- **ID:** `2601.05703` · **primary:** `cs.SE` · **Submitted:** 2026-01-09
+- **Authors:** Wiebe Vandendriessche, Jordi Thijsman, Laurens D'hooge, et al.
+- **abs:** https://arxiv.org/abs/2601.05703 · **PDF:** https://arxiv.org/pdf/2601.05703v1
+
+## 22928. Afri-MCQA: Multimodal Cultural Question Answering for African Languages
+
+- **ID:** `2601.05699` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Atnafu Lambebo Tonja, Srija Anand, Emilio Villa-Cueva, et al.
+- **abs:** https://arxiv.org/abs/2601.05699 · **PDF:** https://arxiv.org/pdf/2601.05699v2
+
+## 22929. Circular Reasoning: Understanding Self-Reinforcing Loops in Large Reasoning Models
+
+- **ID:** `2601.05693` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Zenghao Duan, Liang Pang, Zihao Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.05693 · **PDF:** https://arxiv.org/pdf/2601.05693v1
+
+## 22930. When Smaller Wins: Dual-Stage Distillation and Pareto-Guided Compression of Liquid Neural Networks for Edge Battery Prognostics
+
+- **ID:** `2601.06227` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Dhivya Dharshini Kannan, Wei Li, Wei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.06227 · **PDF:** https://arxiv.org/pdf/2601.06227v2
+
+## 22931. UniFinEval: Towards Unified Evaluation of Financial Multimodal Models across Text, Images and Videos
+
+- **ID:** `2601.22162` · **primary:** `q-fin.GN` · **Submitted:** 2026-01-09
+- **Authors:** Zhi Yang, Lingfeng Zeng, Fangqi Lou, et al.
+- **abs:** https://arxiv.org/abs/2601.22162 · **PDF:** https://arxiv.org/pdf/2601.22162v1
+
+## 22932. FLRQ: Faster LLM Quantization with Flexible Low-Rank Matrix Sketching
+
+- **ID:** `2601.05684` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Hongyaoxing Gul, Lijuan Hu, Shuzi Niu, et al.
+- **abs:** https://arxiv.org/abs/2601.05684 · **PDF:** https://arxiv.org/pdf/2601.05684v1
+
+## 22933. Joint Optimization of Neural Autoregressors via Scoring rules
+
+- **ID:** `2601.05683` · **primary:** `cond-mat.soft` · **Submitted:** 2026-01-09
+- **Authors:** Jonas Landsgesell
+- **abs:** https://arxiv.org/abs/2601.05683 · **PDF:** https://arxiv.org/pdf/2601.05683v1
+
+## 22934. AGDC: Autoregressive Generation of Variable-Length Sequences with Joint Discrete and Continuous Spaces
+
+- **ID:** `2601.05680` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Yeonsang Shin, Insoo Kim, Bongkeun Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.05680 · **PDF:** https://arxiv.org/pdf/2601.05680v1
+
+## 22935. Falsifying Sparse Autoencoder Reasoning Features in Language Models
+
+- **ID:** `2601.05679` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** George Ma, Zhongyuan Liang, Irene Y. Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.05679 · **PDF:** https://arxiv.org/pdf/2601.05679v6
+
+## 22936. CHDP: Cooperative Hybrid Diffusion Policies for Reinforcement Learning in Parameterized Action Space
+
+- **ID:** `2601.05675` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Bingyi Liu, Jinbo He, Haiyong Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.05675 · **PDF:** https://arxiv.org/pdf/2601.05675v1
+
+## 22937. Integrating Temporal Context into Streaming Data for Human Activity Recognition in Smart Home
+
+- **ID:** `2601.11611` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Marina Vicini, Martin Rudorfer, Zhuangzhuang Dai, et al.
+- **abs:** https://arxiv.org/abs/2601.11611 · **PDF:** https://arxiv.org/pdf/2601.11611v1
+
+## 22938. Advancing credit mobility through stakeholder-informed AI design and adoption
+
+- **ID:** `2601.05666` · **primary:** `cs.HC` · **Submitted:** 2026-01-09
+- **Authors:** Yerin Kwak, Siddharth Adelkar, Zachary A. Pardos
+- **abs:** https://arxiv.org/abs/2601.05666 · **PDF:** https://arxiv.org/pdf/2601.05666v1
+
+## 22939. Projecting Out the Malice: A Global Subspace Approach to LLM Detoxification
+
+- **ID:** `2601.06226` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Zenghao Duan, Zhiyi Yin, Zhichao Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.06226 · **PDF:** https://arxiv.org/pdf/2601.06226v1
+
+## 22940. Tracing Stereotypes in Pre-trained Transformers: From Biased Neurons to Fairer Models
+
+- **ID:** `2601.05663` · **primary:** `cs.SE` · **Submitted:** 2026-01-09
+- **Authors:** Gianmario Voria, Moses Openja, Foutse Khomh, et al.
+- **abs:** https://arxiv.org/abs/2601.05663 · **PDF:** https://arxiv.org/pdf/2601.05663v1
+
+## 22941. Stephanie2: Thinking, Waiting, and Making Decisions Like Humans in Step-by-Step AI Social Chat
+
+- **ID:** `2601.05657` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Hao Yang, Hongyuan Lu, Dingkang Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.05657 · **PDF:** https://arxiv.org/pdf/2601.05657v1
+
+## 22942. HAG: Hierarchical Demographic Tree-based Agent Generation for Topic-Adaptive Simulation
+
+- **ID:** `2601.05656` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Rongxin Chen, Tianyu Wu, Bingbing Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.05656 · **PDF:** https://arxiv.org/pdf/2601.05656v2
+
+## 22943. OPT-Engine: Benchmarking the Limits of LLMs in Optimization Modeling via Complexity Scaling
+
+- **ID:** `2601.19924` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Yitian Chen, Cheng Cheng, Yinan Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.19924 · **PDF:** https://arxiv.org/pdf/2601.19924v1
+
+## 22944. A Framework for Personalized Persuasiveness Prediction via Context-Aware User Profiling
+
+- **ID:** `2601.05654` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Sejun Park, Yoonah Park, Jongwon Lim, et al.
+- **abs:** https://arxiv.org/abs/2601.05654 · **PDF:** https://arxiv.org/pdf/2601.05654v2
+
+## 22945. From Global to Local: Cluster-Aware Learning for Wi-Fi Fingerprinting Indoor Localisation
+
+- **ID:** `2601.05650` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Miguel Matey-Sanz, Joaquín Torres-Sospedra, Joaquín Huerta, et al.
+- **abs:** https://arxiv.org/abs/2601.05650 · **PDF:** https://arxiv.org/pdf/2601.05650v1
+
+## 22946. Open World Knowledge Aided Single-Cell Foundation Model with Robust Cross-Modal Cell-Language Pre-training
+
+- **ID:** `2601.05648` · **primary:** `q-bio.GN` · **Submitted:** 2026-01-09
+- **Authors:** Haoran Wang, Xuanyi Zhang, Shuangsang Fang, et al.
+- **abs:** https://arxiv.org/abs/2601.05648 · **PDF:** https://arxiv.org/pdf/2601.05648v1
+
+## 22947. Transformer Is Inherently a Causal Learner
+
+- **ID:** `2601.05647` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Xinyue Wang, Stephen Wang, Biwei Huang
+- **abs:** https://arxiv.org/abs/2601.05647 · **PDF:** https://arxiv.org/pdf/2601.05647v1
+
+## 22948. Multilingual Amnesia: On the Transferability of Unlearning in Multilingual LLMs
+
+- **ID:** `2601.05641` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Alireza Dehghanpour Farashah, Aditi Khandelwal, Marylou Fauchard, et al.
+- **abs:** https://arxiv.org/abs/2601.05641 · **PDF:** https://arxiv.org/pdf/2601.05641v1
+
+## 22949. Classroom AI: Large Language Models as Grade-Specific Teachers
+
+- **ID:** `2601.06225` · **primary:** `cs.CY` · **Submitted:** 2026-01-09
+- **Authors:** Jio Oh, Steven Euijong Whang, James Evans, et al.
+- **abs:** https://arxiv.org/abs/2601.06225 · **PDF:** https://arxiv.org/pdf/2601.06225v2
+
+## 22950. Compressing image encoders via latent distillation
+
+- **ID:** `2601.05639` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Caroline Mazini Rodrigues, Nicolas Keriven, Thomas Maugey
+- **abs:** https://arxiv.org/abs/2601.05639 · **PDF:** https://arxiv.org/pdf/2601.05639v1
+
+## 22951. GenCtrl -- A Formal Controllability Toolkit for Generative Models
+
+- **ID:** `2601.05637` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Emily Cheng, Carmen Amo Alonso, Federico Danieli, et al.
+- **abs:** https://arxiv.org/abs/2601.05637 · **PDF:** https://arxiv.org/pdf/2601.05637v1
+
+## 22952. Continual Pretraining on Encrypted Synthetic Data for Privacy-Preserving LLMs
+
+- **ID:** `2601.05635` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Honghao Liu, Xuhui Jiang, Chengjin Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.05635 · **PDF:** https://arxiv.org/pdf/2601.05635v2
+
+## 22953. GIFT: Games as Informal Training for Generalizable LLMs
+
+- **ID:** `2601.05633` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Nuoyan Lyu, Bingbing Xu, Weihao Meng, et al.
+- **abs:** https://arxiv.org/abs/2601.05633 · **PDF:** https://arxiv.org/pdf/2601.05633v1
+
+## 22954. Cumulative Path-Level Semantic Reasoning for Inductive Knowledge Graph Completion
+
+- **ID:** `2601.05629` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Jiapu Wang, Xinghe Cheng, Zezheng Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.05629 · **PDF:** https://arxiv.org/pdf/2601.05629v1
+
+## 22955. Text Detoxification in isiXhosa and Yorùbá: A Cross-Lingual Machine Learning Approach for Low-Resource African Languages
+
+- **ID:** `2601.05624` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Abayomi O. Agbeyangi
+- **abs:** https://arxiv.org/abs/2601.05624 · **PDF:** https://arxiv.org/pdf/2601.05624v1
+
+## 22956. Continual Learning of Achieving Forgetting-free and Positive Knowledge Transfer
+
+- **ID:** `2601.05623` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Zhi Wang, Zhongbin Wu, Yanni Li, et al.
+- **abs:** https://arxiv.org/abs/2601.05623 · **PDF:** https://arxiv.org/pdf/2601.05623v1
+
+## 22957. Dual-Phase LLM Reasoning: Self-Evolved Mathematical Frameworks
+
+- **ID:** `2601.05616` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** ShaoZhen Liu, Xinting Huang, Houwen Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.05616 · **PDF:** https://arxiv.org/pdf/2601.05616v1
+
+## 22958. PiXTime: A Model for Federated Time Series Forecasting with Heterogeneous Data Structures Across Nodes
+
+- **ID:** `2601.05613` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Yiming Zhou, Mingyue Cheng, Hao Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.05613 · **PDF:** https://arxiv.org/pdf/2601.05613v1
+
+## 22959. CTHA: Constrained Temporal Hierarchical Architecture for Stable Multi-Agent LLM Systems
+
+- **ID:** `2601.10738` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Percy Jardine
+- **abs:** https://arxiv.org/abs/2601.10738 · **PDF:** https://arxiv.org/pdf/2601.10738v1
+
+## 22960. Data Augmented Pipeline for Legal Information Extraction and Reasoning
+
+- **ID:** `2601.05609` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Nguyen Minh Phuong, Ha-Thanh Nguyen, May Myo Zin, et al.
+- **abs:** https://arxiv.org/abs/2601.05609 · **PDF:** https://arxiv.org/pdf/2601.05609v1
+
+## 22961. Orchestrating Tokens and Sequences: Dynamic Hybrid Policy Optimization for RLVR
+
+- **ID:** `2601.05607` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Zijun Min, Bingshuai Liu, Ante Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.05607 · **PDF:** https://arxiv.org/pdf/2601.05607v1
+
+## 22962. SceneAlign: Aligning Multimodal Reasoning to Scene Graphs in Complex Visual Scenes
+
+- **ID:** `2601.05600` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Chuhan Wang, Xintong Li, Jennifer Yuntong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.05600 · **PDF:** https://arxiv.org/pdf/2601.05600v1
+
+## 22963. Table-BiEval: A Self-Supervised, Dual-Track Framework for Decoupling Structure and Content in LLM Evaluation
+
+- **ID:** `2601.19923` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Boxiang Zhao, Qince Li, Zhonghao Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19923 · **PDF:** https://arxiv.org/pdf/2601.19923v1
+
+## 22964. Quantifying and Inducing Shape Bias in CNNs via Max-Pool Dilation
+
+- **ID:** `2601.05599` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Takito Sawada, Akinori Iwata, Masahiro Okuda
+- **abs:** https://arxiv.org/abs/2601.05599 · **PDF:** https://arxiv.org/pdf/2601.05599v2
+
+## 22965. Good Allocations from Bad Estimates
+
+- **ID:** `2601.05597` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Sílvia Casacuberta, Moritz Hardt
+- **abs:** https://arxiv.org/abs/2601.05597 · **PDF:** https://arxiv.org/pdf/2601.05597v1
+
+## 22966. Toward Safe and Responsible AI Agents: A Three-Pillar Model for Transparency, Accountability, and Trustworthiness
+
+- **ID:** `2601.06223` · **primary:** `cs.CY` · **Submitted:** 2026-01-09
+- **Authors:** Edward C. Cheng, Jeshua Cheng, Alice Siu
+- **abs:** https://arxiv.org/abs/2601.06223 · **PDF:** https://arxiv.org/pdf/2601.06223v1
+
+## 22967. Feature Entanglement-based Quantum Multimodal Fusion Neural Network
+
+- **ID:** `2601.07856` · **primary:** `quant-ph` · **Submitted:** 2026-01-09
+- **Authors:** Yu Wu, Qianli Zhou, Jie Geng, et al.
+- **abs:** https://arxiv.org/abs/2601.07856 · **PDF:** https://arxiv.org/pdf/2601.07856v1
+
+## 22968. SAPL: Semantic-Agnostic Prompt Learning in CLIP for Weakly Supervised Image Manipulation Localization
+
+- **ID:** `2601.06222` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Xinghao Wang, Changtao Miao, Dianmo Sheng, et al.
+- **abs:** https://arxiv.org/abs/2601.06222 · **PDF:** https://arxiv.org/pdf/2601.06222v1
+
+## 22969. PaCoRe: Learning to Scale Test-Time Compute with Parallel Coordinated Reasoning
+
+- **ID:** `2601.05593` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Jingcheng Hu, Yinmin Zhang, Shijie Shang, et al.
+- **abs:** https://arxiv.org/abs/2601.05593 · **PDF:** https://arxiv.org/pdf/2601.05593v1
+
+## 22970. A Causal Information-Flow Framework for Unbiased Learning-to-Rank
+
+- **ID:** `2601.05590` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Haoming Gong, Qingyao Ai, Zhihao Tao, et al.
+- **abs:** https://arxiv.org/abs/2601.05590 · **PDF:** https://arxiv.org/pdf/2601.05590v1
+
+## 22971. ACR: Adaptive Context Refactoring via Context Refactoring Operators for Multi-Turn Dialogue
+
+- **ID:** `2601.05589` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Jiawei Shen, Jia Zhu, Hanghui Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.05589 · **PDF:** https://arxiv.org/pdf/2601.05589v1
+
+## 22972. Autoregressive Ranking: Bridging the Gap Between Dual and Cross Encoders
+
+- **ID:** `2601.05588` · **primary:** `cs.IR` · **Submitted:** 2026-01-09
+- **Authors:** Benjamin Rozonoyer, Chong You, Michael Boratko, et al.
+- **abs:** https://arxiv.org/abs/2601.05588 · **PDF:** https://arxiv.org/pdf/2601.05588v4
+
+## 22973. HogVul: Black-box Adversarial Code Generation Framework Against LM-based Vulnerability Detectors
+
+- **ID:** `2601.05587` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Jingxiao Yang, Ping He, Tianyu Du, et al.
+- **abs:** https://arxiv.org/abs/2601.05587 · **PDF:** https://arxiv.org/pdf/2601.05587v1
+
+## 22974. Poisson Hyperplane Processes with Rectified Linear Units
+
+- **ID:** `2601.05586` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Shufei Ge, Shijia Wang, Lloyd Elliott
+- **abs:** https://arxiv.org/abs/2601.05586 · **PDF:** https://arxiv.org/pdf/2601.05586v1
+
+## 22975. GS-DMSR: Dynamic Sensitive Multi-scale Manifold Enhancement for Accelerated High-Quality 3D Gaussian Splatting
+
+- **ID:** `2601.05584` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Nengbo Lu, Minghua Pan, Shaohua Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.05584 · **PDF:** https://arxiv.org/pdf/2601.05584v1
+
+## 22976. Learn to Evolve: Self-supervised Neural JKO Operator for Wasserstein Gradient Flow
+
+- **ID:** `2601.05583` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Xue Feng, Li Wang, Deanna Needell, et al.
+- **abs:** https://arxiv.org/abs/2601.05583 · **PDF:** https://arxiv.org/pdf/2601.05583v1
+
+## 22977. Can large language models interpret unstructured chat data on dynamic group decision-making processes? Evidence on joint destination choice
+
+- **ID:** `2601.05582` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Sung-Yoo Lim, Koki Sato, Kiyoshi Takami, et al.
+- **abs:** https://arxiv.org/abs/2601.05582 · **PDF:** https://arxiv.org/pdf/2601.05582v1
+
+## 22978. LDTC: Lifelong deep temporal clustering for multivariate time series
+
+- **ID:** `2601.06221` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Zhi Wang, Yanni Li, Pingping Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.06221 · **PDF:** https://arxiv.org/pdf/2601.06221v1
+
+## 22979. RISE: Rule-Driven SQL Dialect Translation via Query Reduction
+
+- **ID:** `2601.05579` · **primary:** `cs.DB` · **Submitted:** 2026-01-09
+- **Authors:** Xudong Xie, Yuwei Zhang, Wensheng Dou, et al.
+- **abs:** https://arxiv.org/abs/2601.05579 · **PDF:** https://arxiv.org/pdf/2601.05579v1
+
+## 22980. Reinforcement Learning of Large Language Models for Interpretable Credit Card Fraud Detection
+
+- **ID:** `2601.05578` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Cooper Lin, Yanting Zhang, Maohao Ran, et al.
+- **abs:** https://arxiv.org/abs/2601.05578 · **PDF:** https://arxiv.org/pdf/2601.05578v1
+
+## 22981. Autonomous Discovery of the Ising Model's Critical Parameters with Reinforcement Learning
+
+- **ID:** `2601.05577` · **primary:** `cond-mat.stat-mech` · **Submitted:** 2026-01-09
+- **Authors:** Hai Man, Chaobo Wang, Jia-Rui Li, et al.
+- **abs:** https://arxiv.org/abs/2601.05577 · **PDF:** https://arxiv.org/pdf/2601.05577v1
+
+## 22982. HEART: A Unified Benchmark for Assessing Humans and LLMs in Emotional Support Dialogue
+
+- **ID:** `2601.19922` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Laya Iyer, Kriti Aggarwal, Sanmi Koyejo, et al.
+- **abs:** https://arxiv.org/abs/2601.19922 · **PDF:** https://arxiv.org/pdf/2601.19922v2
+
+## 22983. Crisis-Bench: Benchmarking Strategic Ambiguity and Reputation Management in Large Language Models
+
+- **ID:** `2601.05570` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Cooper Lin, Maohao Ran, Yanting Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.05570 · **PDF:** https://arxiv.org/pdf/2601.05570v1
+
+## 22984. WildSci: Advancing Scientific Reasoning from In-the-Wild Literature
+
+- **ID:** `2601.05567` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Tengxiao Liu, Deepak Nathani, Zekun Li, et al.
+- **abs:** https://arxiv.org/abs/2601.05567 · **PDF:** https://arxiv.org/pdf/2601.05567v1
+
+## 22985. The ICASSP 2026 HumDial Challenge: Benchmarking Human-like Spoken Dialogue Systems in the LLM Era
+
+- **ID:** `2601.05564` · **primary:** `cs.SD` · **Submitted:** 2026-01-09
+- **Authors:** Zhixian Zhao, Shuiyuan Wang, Guojian Li, et al.
+- **abs:** https://arxiv.org/abs/2601.05564 · **PDF:** https://arxiv.org/pdf/2601.05564v2
+
+## 22986. Multifaceted Scenario-Aware Hypergraph Learning for Next POI Recommendation
+
+- **ID:** `2601.11610` · **primary:** `cs.SI` · **Submitted:** 2026-01-09
+- **Authors:** Yuxi Lin, Yongkang Li, Jie Xing, et al.
+- **abs:** https://arxiv.org/abs/2601.11610 · **PDF:** https://arxiv.org/pdf/2601.11610v2
+
+## 22987. Auxiliary-predicted Compress Memory Model(ApCM Model): A Neural Memory Storage Model Based on Invertible Compression and Learnable Prediction
+
+- **ID:** `2601.11609` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Weinuo Ou
+- **abs:** https://arxiv.org/abs/2601.11609 · **PDF:** https://arxiv.org/pdf/2601.11609v1
+
+## 22988. ReasonAny: Incorporating Reasoning Capability to Any Model via Simple and Effective Model Merging
+
+- **ID:** `2601.05560` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Junyao Yang, Chen Qian, Dongrui Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.05560 · **PDF:** https://arxiv.org/pdf/2601.05560v1
+
+## 22989. Semi-Supervised Facial Expression Recognition based on Dynamic Threshold and Negative Learning
+
+- **ID:** `2601.05556` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Zhongpeng Cai, Jun Yu, Wei Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.05556 · **PDF:** https://arxiv.org/pdf/2601.05556v1
+
+## 22990. RoAD Benchmark: How LiDAR Models Fail under Coupled Domain Shifts and Label Evolution
+
+- **ID:** `2601.07855` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Subeen Lee, Siyeong Lee, Namil Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.07855 · **PDF:** https://arxiv.org/pdf/2601.07855v2
+
+## 22991. Generation-Based and Emotion-Reflected Memory Update: Creating the KEEM Dataset for Better Long-Term Conversation
+
+- **ID:** `2601.05548` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Jeonghyun Kang, Hongjin Kim, Harksoo Kim
+- **abs:** https://arxiv.org/abs/2601.05548 · **PDF:** https://arxiv.org/pdf/2601.05548v1
+
+## 22992. VIB-Probe: Detecting and Mitigating Hallucinations in Vision-Language Models via Variational Information Bottleneck
+
+- **ID:** `2601.05547` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Feiran Zhang, Yixin Wu, Zhenghua Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.05547 · **PDF:** https://arxiv.org/pdf/2601.05547v1
+
+## 22993. Can Large Language Models Differentiate Harmful from Argumentative Essays? Steps Toward Ethical Essay Scoring
+
+- **ID:** `2601.05545` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Hongjin Kim, Jeonghyun Kang, Harksoo Kim
+- **abs:** https://arxiv.org/abs/2601.05545 · **PDF:** https://arxiv.org/pdf/2601.05545v1
+
+## 22994. Buffered AUC maximization for scoring systems via mixed-integer optimization
+
+- **ID:** `2601.05544` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Moe Shiina, Shunnosuke Ikeda, Yuichi Takano
+- **abs:** https://arxiv.org/abs/2601.05544 · **PDF:** https://arxiv.org/pdf/2601.05544v2
+
+## 22995. Closing the Modality Reasoning Gap for Speech Large Language Models
+
+- **ID:** `2601.05543` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Chaoren Wang, Heng Lu, Xueyao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.05543 · **PDF:** https://arxiv.org/pdf/2601.05543v1
+
+## 22996. Understanding LLM-Driven Test Oracle Generation
+
+- **ID:** `2601.05542` · **primary:** `cs.SE` · **Submitted:** 2026-01-09
+- **Authors:** Adam Bodicoat, Gunel Jahangirova, Valerio Terragni
+- **abs:** https://arxiv.org/abs/2601.05542 · **PDF:** https://arxiv.org/pdf/2601.05542v1
+
+## 22997. Scalable Heterogeneous Graph Learning via Heterogeneous-aware Orthogonal Prototype Experts
+
+- **ID:** `2601.05537` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Wei Zhou, Hong Huang, Ruize Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.05537 · **PDF:** https://arxiv.org/pdf/2601.05537v1
+
+## 22998. DNATokenizer: A GPU-First Byte-to-Identifier Tokenizer for High-Throughput DNA Language Models
+
+- **ID:** `2601.05531` · **primary:** `q-bio.GN` · **Submitted:** 2026-01-09
+- **Authors:** Eliatan Niktab, Hardip Patel
+- **abs:** https://arxiv.org/abs/2601.05531 · **PDF:** https://arxiv.org/pdf/2601.05531v1
+
+## 22999. Bug Severity Prediction in Software Projects Using Supervised Machine Learning Models
+
+- **ID:** `2603.00004` · **primary:** `cs.SE` · **Submitted:** 2026-01-09
+- **Authors:** Nafisha Tamanna Nice
+- **abs:** https://arxiv.org/abs/2603.00004 · **PDF:** https://arxiv.org/pdf/2603.00004v1
+
+## 23000. Before We Trust Them: Decision-Making Failures in Navigation of Foundation Models
+
+- **ID:** `2601.05529` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Jua Han, Jaeyoon Seo, Jungbin Min, et al.
+- **abs:** https://arxiv.org/abs/2601.05529 · **PDF:** https://arxiv.org/pdf/2601.05529v4
+
+## 23001. Autonomous Probe Microscopy with Robust Bag-of-Features Multi-Objective Bayesian Optimization: Pareto-Front Mapping of Nanoscale Structure-Property Trade-Offs
+
+- **ID:** `2601.05528` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-09
+- **Authors:** Kamyar Barakati, Haochen Zhu, C Charlotte Buchanan, et al.
+- **abs:** https://arxiv.org/abs/2601.05528 · **PDF:** https://arxiv.org/pdf/2601.05528v1
+
+## 23002. Immunological Density Shapes Recovery Trajectories in Long COVID
+
+- **ID:** `2601.07854` · **primary:** `q-bio.OT` · **Submitted:** 2026-01-09
+- **Authors:** Jing Wang, Tong Zhang, Xing Niu, et al.
+- **abs:** https://arxiv.org/abs/2601.07854 · **PDF:** https://arxiv.org/pdf/2601.07854v1
+
+## 23003. DeMa: Dual-Path Delay-Aware Mamba for Efficient Multivariate Time Series Analysis
+
+- **ID:** `2601.05527` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Rui An, Haohao Qu, Wenqi Fan, et al.
+- **abs:** https://arxiv.org/abs/2601.05527 · **PDF:** https://arxiv.org/pdf/2601.05527v1
+
+## 23004. Can We Trust LLM Detectors?
+
+- **ID:** `2601.15301` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Jivnesh Sandhan, Harshit Jaiswal, Fei Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.15301 · **PDF:** https://arxiv.org/pdf/2601.15301v2
+
+## 23005. Explainable AI: Learning from the Learners
+
+- **ID:** `2601.05525` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Ricardo Vinuesa, Steven L. Brunton, Gianmarco Mengaldo
+- **abs:** https://arxiv.org/abs/2601.05525 · **PDF:** https://arxiv.org/pdf/2601.05525v2
+
+## 23006. Double: Breaking the Acceleration Limit via Double Retrieval Speculative Parallelism
+
+- **ID:** `2601.05524` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Yuhao Shen, Tianyu Liu, Junyi Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.05524 · **PDF:** https://arxiv.org/pdf/2601.05524v1
+
+## 23007. Toward an Integrated Cross-Urban Accident Prevention System: A Multi-Task Spatial-Temporal Learning Framework for Urban Safety Management
+
+- **ID:** `2601.05521` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Jiayu Fang, Zhiqi Shao, Haoning Xi, et al.
+- **abs:** https://arxiv.org/abs/2601.05521 · **PDF:** https://arxiv.org/pdf/2601.05521v1
+
+## 23008. CHisAgent: A Multi-Agent Framework for Event Taxonomy Construction in Ancient Chinese Cultural Systems
+
+- **ID:** `2601.05520` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Xuemei Tang, Chengxi Yan, Jinghang Gu, et al.
+- **abs:** https://arxiv.org/abs/2601.05520 · **PDF:** https://arxiv.org/pdf/2601.05520v1
+
+## 23009. Breaking Model Lock-in: Cost-Efficient Zero-Shot LLM Routing via a Universal Latent Space
+
+- **ID:** `2601.06220` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Cheng Yan, Wuyang Zhang, Zhiyuan Ning, et al.
+- **abs:** https://arxiv.org/abs/2601.06220 · **PDF:** https://arxiv.org/pdf/2601.06220v1
+
+## 23010. Enabling Stroke-Level Structural Analysis of Hieroglyphic Scripts without Language-Specific Priors
+
+- **ID:** `2601.05508` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Fuwen Luo, Zihao Wan, Ziyue Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.05508 · **PDF:** https://arxiv.org/pdf/2601.05508v1
+
+## 23011. FlashMem: Distilling Intrinsic Latent Memory via Computation Reuse
+
+- **ID:** `2601.05505` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Yubo Hou, Zhisheng Chen, Tao Wan, et al.
+- **abs:** https://arxiv.org/abs/2601.05505 · **PDF:** https://arxiv.org/pdf/2601.05505v1
+
+## 23012. FinVault: Benchmarking Financial Agent Safety in Execution-Grounded Environments
+
+- **ID:** `2601.07853` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Zhi Yang, Runguo Li, Qiqi Qiang, et al.
+- **abs:** https://arxiv.org/abs/2601.07853 · **PDF:** https://arxiv.org/pdf/2601.07853v1
+
+## 23013. Over-Searching in Search-Augmented Large Language Models
+
+- **ID:** `2601.05503` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Roy Xie, Deepak Gopinath, David Qiu, et al.
+- **abs:** https://arxiv.org/abs/2601.05503 · **PDF:** https://arxiv.org/pdf/2601.05503v2
+
+## 23014. Evaluating the Use of LLMs for Automated DOM-Level Resolution of Web Performance Issues
+
+- **ID:** `2601.05502` · **primary:** `cs.SE` · **Submitted:** 2026-01-09
+- **Authors:** Gideon Peters, SayedHassan Khatoonabadi, Emad Shihab
+- **abs:** https://arxiv.org/abs/2601.05502 · **PDF:** https://arxiv.org/pdf/2601.05502v1
+
+## 23015. Hi-ZFO: Hierarchical Zeroth- and First-Order LLM Fine-Tuning via Importance-Guided Tensor Selection
+
+- **ID:** `2601.05501` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Feihu Jin, Ying Tan
+- **abs:** https://arxiv.org/abs/2601.05501 · **PDF:** https://arxiv.org/pdf/2601.05501v1
+
+## 23016. The Illusion of AI Expertise Under Uncertainty: Navigating Elusive Ground Truth via a Probabilistic Paradigm
+
+- **ID:** `2601.05500` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Aparna Elangovan, Lei Xu, Mahsa Elyasi, et al.
+- **abs:** https://arxiv.org/abs/2601.05500 · **PDF:** https://arxiv.org/pdf/2601.05500v5
+
+## 23017. Prompt-Free SAM-Based Multi-Task Framework for Breast Ultrasound Lesion Segmentation and Classification
+
+- **ID:** `2601.05498` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Samuel E. Johnny, Bernes L. Atabonfack, Israel Alagbe, et al.
+- **abs:** https://arxiv.org/abs/2601.05498 · **PDF:** https://arxiv.org/pdf/2601.05498v1
+
+## 23018. MMViR: A Multi-Modal and Multi-Granularity Representation for Long-range Video Understanding
+
+- **ID:** `2601.05495` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Zizhong Li, Haopeng Zhang, Jiawei Zhang
+- **abs:** https://arxiv.org/abs/2601.05495 · **PDF:** https://arxiv.org/pdf/2601.05495v1
+
+## 23019. MemBuilder: Reinforcing LLMs for Long-Term Memory Construction via Attributed Dense Rewards
+
+- **ID:** `2601.05488` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Zhiyu Shen, Ziming Wu, Fuming Lai, et al.
+- **abs:** https://arxiv.org/abs/2601.05488 · **PDF:** https://arxiv.org/pdf/2601.05488v2
+
+## 23020. EvidFuse: Writing-Time Evidence Learning for Consistent Text-Chart Data Reporting
+
+- **ID:** `2601.05487` · **primary:** `cs.MA` · **Submitted:** 2026-01-09
+- **Authors:** Huanxiang Lin, Qianyue Wang, Jinwu Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.05487 · **PDF:** https://arxiv.org/pdf/2601.05487v1
+
+## 23021. Demystifying Multi-Agent Debate: The Role of Confidence and Diversity
+
+- **ID:** `2601.19921` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Xiaochen Zhu, Caiqi Zhang, Yizhou Chi, et al.
+- **abs:** https://arxiv.org/abs/2601.19921 · **PDF:** https://arxiv.org/pdf/2601.19921v1
+
+## 23022. MMUEChange: A Generalized LLM Agent Framework for Intelligent Multi-Modal Urban Environment Change Analysis
+
+- **ID:** `2601.05483` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Zixuan Xiao, Jun Ma, Siwei Zhang
+- **abs:** https://arxiv.org/abs/2601.05483 · **PDF:** https://arxiv.org/pdf/2601.05483v1
+
+## 23023. Hardware-Aware Reformulation of Convolutions for Efficient Execution on Specialized AI Hardware: A Case Study on NVIDIA Tensor Cores
+
+- **ID:** `2601.11608` · **primary:** `cs.DC` · **Submitted:** 2026-01-09
+- **Authors:** Ganesh Bikshandi
+- **abs:** https://arxiv.org/abs/2601.11608 · **PDF:** https://arxiv.org/pdf/2601.11608v1
+
+## 23024. The Facade of Truth: Uncovering and Mitigating LLM Susceptibility to Deceptive Evidence
+
+- **ID:** `2601.05478` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Herun Wan, Jiaying Wu, Minnan Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.05478 · **PDF:** https://arxiv.org/pdf/2601.05478v1
+
+## 23025. MaxCode: A Max-Reward Reinforcement Learning Framework for Automated Code Optimization
+
+- **ID:** `2601.05475` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Jiefu Ou, Sapana Chaudhary, Kaj Bostrom, et al.
+- **abs:** https://arxiv.org/abs/2601.05475 · **PDF:** https://arxiv.org/pdf/2601.05475v1
+
+## 23026. Efficient Differentiable Causal Discovery via Reliable Super-Structure Learning
+
+- **ID:** `2601.05474` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Pingchuan Ma, Qixin Zhang, Shuai Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.05474 · **PDF:** https://arxiv.org/pdf/2601.05474v1
+
+## 23027. AI-Powered Algorithms for the Prevention and Detection of Computer Malware Infections
+
+- **ID:** `2601.06219` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Rakesh Keshava, Sathish Kuppan Pandurangan, M. Sakthivanitha, et al.
+- **abs:** https://arxiv.org/abs/2601.06219 · **PDF:** https://arxiv.org/pdf/2601.06219v1
+
+## 23028. Two-step Authentication: Multi-biometric System Using Voice and Facial Recognition
+
+- **ID:** `2601.06218` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Kuan Wei Chen, Ting Yi Lin, Wen Ren Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.06218 · **PDF:** https://arxiv.org/pdf/2601.06218v1
+
+## 23029. Towards Valid Student Simulation with Large Language Models
+
+- **ID:** `2601.05473` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Zhihao Yuan, Yunze Xiao, Ming Li, et al.
+- **abs:** https://arxiv.org/abs/2601.05473 · **PDF:** https://arxiv.org/pdf/2601.05473v1
+
+## 23030. ROAP: A Reading-Order and Attention-Prior Pipeline for Optimizing Layout Transformers in Key Information Extraction
+
+- **ID:** `2601.05470` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Tingwei Xie, Jinxin He, Yonghong Song
+- **abs:** https://arxiv.org/abs/2601.05470 · **PDF:** https://arxiv.org/pdf/2601.05470v1
+
+## 23031. STELP: Secure Transpilation and Execution of LLM-Generated Programs
+
+- **ID:** `2601.05467` · **primary:** `cs.SE` · **Submitted:** 2026-01-09
+- **Authors:** Swapnil Shinde, Sahil Wadhwa, Andy Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.05467 · **PDF:** https://arxiv.org/pdf/2601.05467v3
+
+## 23032. Jailbreaking Large Language Models through Iterative Tool-Disguised Attacks via Reinforcement Learning
+
+- **ID:** `2601.05466` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Zhaoqi Wang, Zijian Zhang, Daqing He, et al.
+- **abs:** https://arxiv.org/abs/2601.05466 · **PDF:** https://arxiv.org/pdf/2601.05466v1
+
+## 23033. PRISMA: Reinforcement Learning Guided Two-Stage Policy Optimization in Multi-Agent Architecture for Open-Domain Multi-Hop Question Answering
+
+- **ID:** `2601.05465` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Yu Liu, Wenxiao Zhang, Cong Cao, et al.
+- **abs:** https://arxiv.org/abs/2601.05465 · **PDF:** https://arxiv.org/pdf/2601.05465v1
+
+## 23034. Do LLMs Need Inherent Reasoning Before Reinforcement Learning? A Study in Korean Self-Correction
+
+- **ID:** `2601.05459` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Hongjin Kim, Jaewook Lee, Kiyoung Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.05459 · **PDF:** https://arxiv.org/pdf/2601.05459v1
+
+## 23035. ART: Adaptive Reasoning Trees for Explainable Claim Verification
+
+- **ID:** `2601.05455` · **primary:** `cs.AI` · **Submitted:** 2026-01-09
+- **Authors:** Sahil Wadhwa, Himanshu Kumar, Guanqun Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.05455 · **PDF:** https://arxiv.org/pdf/2601.05455v1
+
+## 23036. RingSQL: Generating Synthetic Data with Schema-Independent Templates for Text-to-SQL Reasoning Models
+
+- **ID:** `2601.05451` · **primary:** `cs.LG` · **Submitted:** 2026-01-09
+- **Authors:** Marko Sterbentz, Kevin Cushing, Cameron Barrie, et al.
+- **abs:** https://arxiv.org/abs/2601.05451 · **PDF:** https://arxiv.org/pdf/2601.05451v1
+
+## 23037. Bias Detection and Rotation-Robustness Mitigation in Vision-Language Models and Generative Image Models
+
+- **ID:** `2601.08860` · **primary:** `cs.CV` · **Submitted:** 2026-01-09
+- **Authors:** Tarannum Mithila
+- **abs:** https://arxiv.org/abs/2601.08860 · **PDF:** https://arxiv.org/pdf/2601.08860v1
+
+## 23038. Knowledge-Driven Multi-Turn Jailbreaking on Large Language Models
+
+- **ID:** `2601.05445` · **primary:** `cs.CR` · **Submitted:** 2026-01-09
+- **Authors:** Songze Li, Ruishi He, Xiaojun Jia, et al.
+- **abs:** https://arxiv.org/abs/2601.05445 · **PDF:** https://arxiv.org/pdf/2601.05445v1
+
+## 23039. What Functions Does XGBoost Learn?
+
+- **ID:** `2601.05444` · **primary:** `math.ST` · **Submitted:** 2026-01-09
+- **Authors:** Dohyeong Ki, Adityanand Guntuboyina
+- **abs:** https://arxiv.org/abs/2601.05444 · **PDF:** https://arxiv.org/pdf/2601.05444v1
+
+## 23040. A brief note on learning problem with global perspectives
+
+- **ID:** `2601.05441` · **primary:** `stat.ML` · **Submitted:** 2026-01-09
+- **Authors:** Getachew K. Befekadu
+- **abs:** https://arxiv.org/abs/2601.05441 · **PDF:** https://arxiv.org/pdf/2601.05441v1
+
+## 23041. Tracing Moral Foundations in Large Language Models
+
+- **ID:** `2601.05437` · **primary:** `cs.CL` · **Submitted:** 2026-01-09
+- **Authors:** Chenxiao Yu, Bowen Yi, Farzan Karimi-Malekabadi, et al.
+- **abs:** https://arxiv.org/abs/2601.05437 · **PDF:** https://arxiv.org/pdf/2601.05437v1
+
+## 23042. Thinking with Map: Reinforced Parallel Map-Augmented Agent for Geolocalization
+
+- **ID:** `2601.05432` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Yuxiang Ji, Yong Wang, Ziyu Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.05432 · **PDF:** https://arxiv.org/pdf/2601.05432v1
+
+## 23043. Prediction of Fault Slip Tendency in CO${_2}$ Storage using Data-space Inversion
+
+- **ID:** `2601.05431` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Xiaowen He, Su Jiang, Louis J. Durlofsky
+- **abs:** https://arxiv.org/abs/2601.05431 · **PDF:** https://arxiv.org/pdf/2601.05431v1
+
+## 23044. Dynamic Inclusion and Bounded Multi-Factor Tilts for Robust Portfolio Construction
+
+- **ID:** `2601.05428` · **primary:** `math.OC` · **Submitted:** 2026-01-08
+- **Authors:** Roberto Garrone
+- **abs:** https://arxiv.org/abs/2601.05428 · **PDF:** https://arxiv.org/pdf/2601.05428v1
+
+## 23045. Efficient Inference for Noisy LLM-as-a-Judge Evaluation
+
+- **ID:** `2601.05420` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Yiqun T Chen, Sizhu Lu, Sijia Li, et al.
+- **abs:** https://arxiv.org/abs/2601.05420 · **PDF:** https://arxiv.org/pdf/2601.05420v1
+
+## 23046. Multi-Group Quadratic Discriminant Analysis via Projection
+
+- **ID:** `2601.05415` · **primary:** `stat.ME` · **Submitted:** 2026-01-08
+- **Authors:** Yuchao Wang, Tianying Wang
+- **abs:** https://arxiv.org/abs/2601.05415 · **PDF:** https://arxiv.org/pdf/2601.05415v1
+
+## 23047. Large Language Models Are Bad Dice Players: LLMs Struggle to Generate Random Numbers from Statistical Distributions
+
+- **ID:** `2601.05414` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Minda Zhao, Yilun Du, Mengyu Wang
+- **abs:** https://arxiv.org/abs/2601.05414 · **PDF:** https://arxiv.org/pdf/2601.05414v1
+
+## 23048. Glitter: Visualizing Lexical Surprisal for Readability in Administrative Texts
+
+- **ID:** `2601.05411` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Jan Černý, Ivana Kvapilíková, Silvie Cinková
+- **abs:** https://arxiv.org/abs/2601.05411 · **PDF:** https://arxiv.org/pdf/2601.05411v1
+
+## 23049. Interactive Distillation for Cooperative Multi-Agent Reinforcement Learning
+
+- **ID:** `2601.05407` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Minwoo Cho, Batuhan Altundas, Matthew Gombolay
+- **abs:** https://arxiv.org/abs/2601.05407 · **PDF:** https://arxiv.org/pdf/2601.05407v1
+
+## 23050. Same Claim, Different Judgment: Benchmarking Scenario-Induced Bias in Multilingual Financial Misinformation Detection
+
+- **ID:** `2601.05403` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Zhiwei Liu, Yupen Cao, Yuechen Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.05403 · **PDF:** https://arxiv.org/pdf/2601.05403v1
+
+## 23051. CEEMDAN-Based Multiscale CNN for Wind Turbine Gearbox Fault Detection
+
+- **ID:** `2601.06217` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Nejad Alagha, Anis Salwa Mohd Khairuddin, Obada Al-Khatib, et al.
+- **abs:** https://arxiv.org/abs/2601.06217 · **PDF:** https://arxiv.org/pdf/2601.06217v1
+
+## 23052. Multi-task Cross-modal Learning for Chest X-ray Image Retrieval
+
+- **ID:** `2601.05399` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Zhaohui Liang, Sivaramakrishnan Rajaraman, Niccolo Marini, et al.
+- **abs:** https://arxiv.org/abs/2601.05399 · **PDF:** https://arxiv.org/pdf/2601.05399v1
+
+## 23053. Archetypal cases for questionnaires with nominal multiple choice questions
+
+- **ID:** `2601.05392` · **primary:** `stat.ME` · **Submitted:** 2026-01-08
+- **Authors:** Aleix Alcacer, Irene Epifanio
+- **abs:** https://arxiv.org/abs/2601.05392 · **PDF:** https://arxiv.org/pdf/2601.05392v1
+
+## 23054. DynaSTy: A Framework for SpatioTemporal Node Attribute Prediction in Dynamic Graphs
+
+- **ID:** `2601.05391` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Namrata Banerji, Tanya Berger-Wolf
+- **abs:** https://arxiv.org/abs/2601.05391 · **PDF:** https://arxiv.org/pdf/2601.05391v1
+
+## 23055. ALL-FEM: Agentic Large Language models Fine-tuned for Finite Element Methods
+
+- **ID:** `2603.21011` · **primary:** `cs.CE` · **Submitted:** 2026-01-08
+- **Authors:** Rushikesh Deotale, Adithya Srinivasan, Yuan Tian, et al.
+- **abs:** https://arxiv.org/abs/2603.21011 · **PDF:** https://arxiv.org/pdf/2603.21011v1
+
+## 23056. On the Effect of Cheating in Chess
+
+- **ID:** `2601.05386` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Daniel Keren
+- **abs:** https://arxiv.org/abs/2601.05386 · **PDF:** https://arxiv.org/pdf/2601.05386v1
+
+## 23057. Conformity and Social Impact on AI Agents
+
+- **ID:** `2601.05384` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Alessandro Bellina, Giordano De Marzo, David Garcia
+- **abs:** https://arxiv.org/abs/2601.05384 · **PDF:** https://arxiv.org/pdf/2601.05384v1
+
+## 23058. Imitation Learning for Combinatorial Optimisation under Uncertainty
+
+- **ID:** `2601.05383` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Prakash Gawas, Antoine Legrain, Louis-Martin Rousseau
+- **abs:** https://arxiv.org/abs/2601.05383 · **PDF:** https://arxiv.org/pdf/2601.05383v3
+
+## 23059. LLM Agents in Law: Taxonomy, Applications, and Challenges
+
+- **ID:** `2601.06216` · **primary:** `cs.CY` · **Submitted:** 2026-01-08
+- **Authors:** Shuang Liu, Ruijia Zhang, Ruoyun Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.06216 · **PDF:** https://arxiv.org/pdf/2601.06216v1
+
+## 23060. Inverting Non-Injective Functions with Twin Neural Network Regression
+
+- **ID:** `2601.05378` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Sebastian J. Wetzel
+- **abs:** https://arxiv.org/abs/2601.05378 · **PDF:** https://arxiv.org/pdf/2601.05378v2
+
+## 23061. The Persona Paradox: Medical Personas as Behavioral Priors in Clinical Language Models
+
+- **ID:** `2601.05376` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Tassallah Abdullahi, Shrestha Ghosh, Hamish S Fraser, et al.
+- **abs:** https://arxiv.org/abs/2601.05376 · **PDF:** https://arxiv.org/pdf/2601.05376v1
+
+## 23062. From Unstructured Data to Demand Counterfactuals: Theory and Practice
+
+- **ID:** `2601.05374` · **primary:** `econ.EM` · **Submitted:** 2026-01-08
+- **Authors:** Timothy Christensen, Giovanni Compiani
+- **abs:** https://arxiv.org/abs/2601.05374 · **PDF:** https://arxiv.org/pdf/2601.05374v1
+
+## 23063. Ensemble of radiomics and ConvNeXt for breast cancer diagnosis
+
+- **ID:** `2601.05373` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Jorge Alberto Garza-Abdala, Gerardo Alejandro Fumagal-González, Beatriz A. Bosques-Palomo, et al.
+- **abs:** https://arxiv.org/abs/2601.05373 · **PDF:** https://arxiv.org/pdf/2601.05373v1
+
+## 23064. The Kernel Manifold: A Geometric Approach to Gaussian Process Model Selection
+
+- **ID:** `2601.05371` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Md Shafiqul Islam, Shakti Prasad Padhy, Douglas Allaire, et al.
+- **abs:** https://arxiv.org/abs/2601.05371 · **PDF:** https://arxiv.org/pdf/2601.05371v2
+
+## 23065. Lost in Execution: On the Multilingual Robustness of Tool Calling in Large Language Models
+
+- **ID:** `2601.05366` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Zheng Luo, T Pranav Kutralingam, Ogochukwu N Okoani, et al.
+- **abs:** https://arxiv.org/abs/2601.05366 · **PDF:** https://arxiv.org/pdf/2601.05366v1
+
+## 23066. STResNet & STYOLO : A New Family of Compact Classification and Object Detection Models for MCUs
+
+- **ID:** `2601.05364` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Sudhakar Sah, Ravish Kumar
+- **abs:** https://arxiv.org/abs/2601.05364 · **PDF:** https://arxiv.org/pdf/2601.05364v1
+
+## 23067. Sparsity-Aware Low-Rank Representation for Efficient Fine-Tuning of Large Language Models
+
+- **ID:** `2601.16991` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Longteng Zhang, Sen Wu, Shuai Hou, et al.
+- **abs:** https://arxiv.org/abs/2601.16991 · **PDF:** https://arxiv.org/pdf/2601.16991v2
+
+## 23068. The Table of Media Bias Elements: A sentence-level taxonomy of media bias types and propaganda techniques
+
+- **ID:** `2601.05358` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Tim Menzner, Jochen L. Leidner
+- **abs:** https://arxiv.org/abs/2601.05358 · **PDF:** https://arxiv.org/pdf/2601.05358v1
+
+## 23069. PRISM: Protocol Refinement through Intelligent Simulation Modeling
+
+- **ID:** `2601.05356` · **primary:** `cs.RO` · **Submitted:** 2026-01-08
+- **Authors:** Brian Hsu, Priyanka V Setty, Rory M Butler, et al.
+- **abs:** https://arxiv.org/abs/2601.05356 · **PDF:** https://arxiv.org/pdf/2601.05356v1
+
+## 23070. An AI-powered Bayesian Generative Modeling Approach for Arbitrary Conditional Inference
+
+- **ID:** `2601.05355` · **primary:** `stat.ML` · **Submitted:** 2026-01-08
+- **Authors:** Qiao Liu, Wing Hung Wong
+- **abs:** https://arxiv.org/abs/2601.05355 · **PDF:** https://arxiv.org/pdf/2601.05355v2
+
+## 23071. GlyRAG: Context-Aware Retrieval-Augmented Framework for Blood Glucose Forecasting
+
+- **ID:** `2601.05353` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Shovito Barua Soumma, Hassan Ghasemzadeh
+- **abs:** https://arxiv.org/abs/2601.05353 · **PDF:** https://arxiv.org/pdf/2601.05353v1
+
+## 23072. When the Server Steps In: Calibrated Updates for Fair Federated Learning
+
+- **ID:** `2601.05352` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Tianrun Yu, Kaixiang Zhao, Cheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.05352 · **PDF:** https://arxiv.org/pdf/2601.05352v2
+
+## 23073. A Multimodal Data Processing Pipeline for MIMIC-IV Dataset
+
+- **ID:** `2601.11606` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Farzana Islam Adiba, Varsha Danduri, Fahmida Liza Piya, et al.
+- **abs:** https://arxiv.org/abs/2601.11606 · **PDF:** https://arxiv.org/pdf/2601.11606v1
+
+## 23074. Formal Safety Guarantees for Autonomous Vehicles using Barrier Certificates
+
+- **ID:** `2601.09740` · **primary:** `cs.RO` · **Submitted:** 2026-01-08
+- **Authors:** Oumaima Barhoumi, Mohamed H Zaki, Sofiène Tahar
+- **abs:** https://arxiv.org/abs/2601.09740 · **PDF:** https://arxiv.org/pdf/2601.09740v1
+
+## 23075. Multi-turn Jailbreaking Attack in Multi-Modal Large Language Models
+
+- **ID:** `2601.05339` · **primary:** `cs.CR` · **Submitted:** 2026-01-08
+- **Authors:** Badhan Chandra Das, Md Tasnim Jawad, Joaquin Molto, et al.
+- **abs:** https://arxiv.org/abs/2601.05339 · **PDF:** https://arxiv.org/pdf/2601.05339v1
+
+## 23076. PiC-BNN: A 128-kbit 65 nm Processing-in-CAM-Based End-to-End Binary Neural Network Accelerator
+
+- **ID:** `2601.19920` · **primary:** `cs.AR` · **Submitted:** 2026-01-08
+- **Authors:** Yuval Harary, Almog Sharoni, Esteban Garzón, et al.
+- **abs:** https://arxiv.org/abs/2601.19920 · **PDF:** https://arxiv.org/pdf/2601.19920v1
+
+## 23077. Dynamics-inspired Structure Hallucination for Protein-protein Interaction Modeling
+
+- **ID:** `2601.06214` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Fang Wu, Stan Z. Li
+- **abs:** https://arxiv.org/abs/2601.06214 · **PDF:** https://arxiv.org/pdf/2601.06214v1
+
+## 23078. Generalized Canonical Polyadic Tensor Decompositions with General Symmetry
+
+- **ID:** `2601.05335` · **primary:** `math.NA` · **Submitted:** 2026-01-08
+- **Authors:** Alex Mulrooney, David Hong
+- **abs:** https://arxiv.org/abs/2601.05335 · **PDF:** https://arxiv.org/pdf/2601.05335v1
+
+## 23079. Cyber Threat Detection and Vulnerability Assessment System using Generative AI and Large Language Model
+
+- **ID:** `2601.06213` · **primary:** `cs.CR` · **Submitted:** 2026-01-08
+- **Authors:** Keerthi Kumar. M, Swarun Kumar Joginpelly, Sunil Khemka, et al.
+- **abs:** https://arxiv.org/abs/2601.06213 · **PDF:** https://arxiv.org/pdf/2601.06213v1
+
+## 23080. Improving Enzyme Prediction with Chemical Reaction Equations by Hypergraph-Enhanced Knowledge Graph Embeddings
+
+- **ID:** `2601.05330` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Tengwei Song, Long Yin, Zhen Han, et al.
+- **abs:** https://arxiv.org/abs/2601.05330 · **PDF:** https://arxiv.org/pdf/2601.05330v1
+
+## 23081. Bi-Orthogonal Factor Decomposition for Vision Transformers
+
+- **ID:** `2601.05328` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Fenil R. Doshi, Thomas Fel, Talia Konkle, et al.
+- **abs:** https://arxiv.org/abs/2601.05328 · **PDF:** https://arxiv.org/pdf/2601.05328v1
+
+## 23082. Optimal Lower Bounds for Online Multicalibration
+
+- **ID:** `2601.05245` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Natalie Collina, Jiuyao Lu, Georgy Noarov, et al.
+- **abs:** https://arxiv.org/abs/2601.05245 · **PDF:** https://arxiv.org/pdf/2601.05245v1
+
+## 23083. GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization
+
+- **ID:** `2601.05242` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Shih-Yang Liu, Xin Dong, Ximing Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.05242 · **PDF:** https://arxiv.org/pdf/2601.05242v1
+
+## 23084. RoboVIP: Multi-View Video Generation with Visual Identity Prompting Augments Robot Manipulation
+
+- **ID:** `2601.05241` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Boyang Wang, Haoran Zhang, Shujie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.05241 · **PDF:** https://arxiv.org/pdf/2601.05241v1
+
+## 23085. Robust Reasoning as a Symmetry-Protected Topological Phase
+
+- **ID:** `2601.05240` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Ilmo Sung
+- **abs:** https://arxiv.org/abs/2601.05240 · **PDF:** https://arxiv.org/pdf/2601.05240v1
+
+## 23086. Measuring and Fostering Peace through Machine Learning and Artificial Intelligence
+
+- **ID:** `2601.05232` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** P. Gilda, P. Dungarwal, A. Thongkham, et al.
+- **abs:** https://arxiv.org/abs/2601.05232 · **PDF:** https://arxiv.org/pdf/2601.05232v2
+
+## 23087. Learning Latent Action World Models In The Wild
+
+- **ID:** `2601.05230` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Quentin Garrido, Tushar Nagarajan, Basile Terver, et al.
+- **abs:** https://arxiv.org/abs/2601.05230 · **PDF:** https://arxiv.org/pdf/2601.05230v2
+
+## 23088. Stochastic Deep Learning: A Probabilistic Framework for Modeling Uncertainty in Structured Temporal Data
+
+- **ID:** `2601.05227` · **primary:** `stat.ML` · **Submitted:** 2026-01-08
+- **Authors:** James Rice
+- **abs:** https://arxiv.org/abs/2601.05227 · **PDF:** https://arxiv.org/pdf/2601.05227v1
+
+## 23089. CAOS: Conformal Aggregation of One-Shot Predictors
+
+- **ID:** `2601.05219` · **primary:** `stat.ML` · **Submitted:** 2026-01-08
+- **Authors:** Maja Waldron
+- **abs:** https://arxiv.org/abs/2601.05219 · **PDF:** https://arxiv.org/pdf/2601.05219v2
+
+## 23090. Akasha 2: Hamiltonian State Space Duality and Visual-Language Joint Embedding Predictive Architectur
+
+- **ID:** `2601.06212` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Yani Meziani
+- **abs:** https://arxiv.org/abs/2601.06212 · **PDF:** https://arxiv.org/pdf/2601.06212v1
+
+## 23091. MineNPC-Task: Task Suite for Memory-Aware Minecraft Agents
+
+- **ID:** `2601.05215` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Tamil Sudaravan Mohan Doss, Michael Xu, Sudha Rao, et al.
+- **abs:** https://arxiv.org/abs/2601.05215 · **PDF:** https://arxiv.org/pdf/2601.05215v2
+
+## 23092. Internal Representations as Indicators of Hallucinations in Agent Tool Selection
+
+- **ID:** `2601.05214` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Kait Healy, Bharathi Srinivasan, Visakh Madathil, et al.
+- **abs:** https://arxiv.org/abs/2601.05214 · **PDF:** https://arxiv.org/pdf/2601.05214v1
+
+## 23093. EARL: Energy-Aware Optimization of Liquid State Machines for Pervasive AI
+
+- **ID:** `2601.05205` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Zain Iqbal, Lorenzo Valerio
+- **abs:** https://arxiv.org/abs/2601.05205 · **PDF:** https://arxiv.org/pdf/2601.05205v1
+
+## 23094. Stock Market Price Prediction using Neural Prophet with Deep Neural Network
+
+- **ID:** `2601.05202` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Navin Chhibber, Sunil Khemka, Navneet Kumar Tyagi, et al.
+- **abs:** https://arxiv.org/abs/2601.05202 · **PDF:** https://arxiv.org/pdf/2601.05202v3
+
+## 23095. Mechanisms of Prompt-Induced Hallucination in Vision-Language Models
+
+- **ID:** `2601.05201` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** William Rudman, Michal Golovanevsky, Dana Arad, et al.
+- **abs:** https://arxiv.org/abs/2601.05201 · **PDF:** https://arxiv.org/pdf/2601.05201v1
+
+## 23096. Hindsight Preference Replay Improves Preference-Conditioned Multi-Objective Reinforcement Learning
+
+- **ID:** `2601.11604` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Jonaid Shianifar, Michael Schukat, Karl Mason
+- **abs:** https://arxiv.org/abs/2601.11604 · **PDF:** https://arxiv.org/pdf/2601.11604v1
+
+## 23097. An interpretable data-driven approach to optimizing clinical fall risk assessment
+
+- **ID:** `2601.05194` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Fardin Ganjkhanloo, Emmett Springer, Erik H. Hoyer, et al.
+- **abs:** https://arxiv.org/abs/2601.05194 · **PDF:** https://arxiv.org/pdf/2601.05194v2
+
+## 23098. LELA: an LLM-based Entity Linking Approach with Zero-Shot Domain Adaptation
+
+- **ID:** `2601.05192` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Samy Haffoudhi, Fabian M. Suchanek, Nils Holzenberger
+- **abs:** https://arxiv.org/abs/2601.05192 · **PDF:** https://arxiv.org/pdf/2601.05192v1
+
+## 23099. AgentCompress: Task-Aware Compression for Affordable Large Language Model Agents
+
+- **ID:** `2601.05191` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Zuhair Ahmed Khan Taha, Mohammed Mudassir Uddin, Shahnawaz Alam
+- **abs:** https://arxiv.org/abs/2601.05191 · **PDF:** https://arxiv.org/pdf/2601.05191v2
+
+## 23100. SimuAgent: An LLM-Based Simulink Modeling Assistant Enhanced with Reinforcement Learning
+
+- **ID:** `2601.05187` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Yanchang Liang, Xiaowei Zhao
+- **abs:** https://arxiv.org/abs/2601.05187 · **PDF:** https://arxiv.org/pdf/2601.05187v1
+
+## 23101. Observations and Remedies for Large Language Model Bias in Self-Consuming Performative Loop
+
+- **ID:** `2601.05184` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Yaxuan Wang, Zhongteng Cai, Yujia Bao, et al.
+- **abs:** https://arxiv.org/abs/2601.05184 · **PDF:** https://arxiv.org/pdf/2601.05184v1
+
+## 23102. Ontology Neural Networks for Topologically Conditioned Constraint Satisfaction
+
+- **ID:** `2601.05304` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Jaehong Oh
+- **abs:** https://arxiv.org/abs/2601.05304 · **PDF:** https://arxiv.org/pdf/2601.05304v1
+
+## 23103. FaST: Efficient and Effective Long-Horizon Forecasting for Large-Scale Spatial-Temporal Graphs via Mixture-of-Experts
+
+- **ID:** `2601.05174` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Yiji Zhao, Zihao Zhong, Ao Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.05174 · **PDF:** https://arxiv.org/pdf/2601.05174v1
+
+## 23104. CoV: Chain-of-View Prompting for Spatial Reasoning
+
+- **ID:** `2601.05172` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Haoyu Zhao, Akide Liu, Zeyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.05172 · **PDF:** https://arxiv.org/pdf/2601.05172v2
+
+## 23105. Inside Out: Evolving User-Centric Core Memory Trees for Long-Term Personalized Dialogue Systems
+
+- **ID:** `2601.05171` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Jihao Zhao, Ding Chen, Zhaoxin Fan, et al.
+- **abs:** https://arxiv.org/abs/2601.05171 · **PDF:** https://arxiv.org/pdf/2601.05171v2
+
+## 23106. Reverse-engineering NLI: A study of the meta-inferential properties of Natural Language Inference
+
+- **ID:** `2601.05170` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Rasmus Blanck, Bill Noble, Stergios Chatzikyriakidis
+- **abs:** https://arxiv.org/abs/2601.05170 · **PDF:** https://arxiv.org/pdf/2601.05170v2
+
+## 23107. RelayLLM: Efficient Reasoning via Collaborative Decoding
+
+- **ID:** `2601.05167` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Chengsong Huang, Tong Zheng, Langlin Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.05167 · **PDF:** https://arxiv.org/pdf/2601.05167v1
+
+## 23108. DocDancer: Towards Agentic Document-Grounded Information Seeking
+
+- **ID:** `2601.05163` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Qintong Zhang, Xinjie Lv, Jialong Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.05163 · **PDF:** https://arxiv.org/pdf/2601.05163v1
+
+## 23109. Vision-Language Introspection: Mitigating Overconfident Hallucinations in MLLMs via Interpretable Bi-Causal Steering
+
+- **ID:** `2601.05159` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Shuliang Liu, Songbo Yang, Dong Fang, et al.
+- **abs:** https://arxiv.org/abs/2601.05159 · **PDF:** https://arxiv.org/pdf/2601.05159v1
+
+## 23110. Learning Mixture Models via Efficient High-dimensional Sparse Fourier Transforms
+
+- **ID:** `2601.05157` · **primary:** `cs.DS` · **Submitted:** 2026-01-08
+- **Authors:** Alkis Kalavasis, Pravesh K. Kothari, Shuchen Li, et al.
+- **abs:** https://arxiv.org/abs/2601.05157 · **PDF:** https://arxiv.org/pdf/2601.05157v1
+
+## 23111. SPINAL -- Scaling-law and Preference Integration in Neural Alignment Layers
+
+- **ID:** `2601.06238` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Arion Das, Partha Pratim Saha, Amit Dhanda, et al.
+- **abs:** https://arxiv.org/abs/2601.06238 · **PDF:** https://arxiv.org/pdf/2601.06238v1
+
+## 23112. Safe Continual Reinforcement Learning Methods for Nonstationary Environments. Towards a Survey of the State of the Art
+
+- **ID:** `2601.05152` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Timofey Tomashevskiy
+- **abs:** https://arxiv.org/abs/2601.05152 · **PDF:** https://arxiv.org/pdf/2601.05152v1
+
+## 23113. ROOFS: RObust biOmarker Feature Selection
+
+- **ID:** `2601.05151` · **primary:** `stat.ML` · **Submitted:** 2026-01-08
+- **Authors:** Anastasiia Bakhmach, Paul Dufossé, Andrea Vaglio, et al.
+- **abs:** https://arxiv.org/abs/2601.05151 · **PDF:** https://arxiv.org/pdf/2601.05151v2
+
+## 23114. Atlas 2 -- Foundation models for clinical deployment
+
+- **ID:** `2601.05148` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Maximilian Alber, Timo Milbich, Alexandra Carpen-Amarie, et al.
+- **abs:** https://arxiv.org/abs/2601.05148 · **PDF:** https://arxiv.org/pdf/2601.05148v1
+
+## 23115. Distilling the Thought, Watermarking the Answer: A Principle Semantic Guided Watermark for Large Reasoning Models
+
+- **ID:** `2601.05144` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Shuliang Liu, Xingyu Li, Hongyi Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.05144 · **PDF:** https://arxiv.org/pdf/2601.05144v2
+
+## 23116. A Two-Stage Multitask Vision-Language Framework for Explainable Crop Disease Visual Question Answering
+
+- **ID:** `2601.05143` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Md. Zahid Hossain, Most. Sharmin Sultana Samu, Md. Rakibul Islam, et al.
+- **abs:** https://arxiv.org/abs/2601.05143 · **PDF:** https://arxiv.org/pdf/2601.05143v2
+
+## 23117. Neural Algorithmic Reasoning for Approximate $k$-Coloring with Recursive Warm Starts
+
+- **ID:** `2601.05137` · **primary:** `math.CO` · **Submitted:** 2026-01-08
+- **Authors:** Knut Vanderbush, Melanie Weber
+- **abs:** https://arxiv.org/abs/2601.05137 · **PDF:** https://arxiv.org/pdf/2601.05137v1
+
+## 23118. Sequential Subspace Noise Injection Prevents Accuracy Collapse in Certified Unlearning
+
+- **ID:** `2601.05134` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Polina Dolgova, Sebastian U. Stich
+- **abs:** https://arxiv.org/abs/2601.05134 · **PDF:** https://arxiv.org/pdf/2601.05134v1
+
+## 23119. VERSE: Visual Embedding Reduction and Space Exploration. Clustering-Guided Insights for Training Data Enhancement in Visually-Rich Document Understanding
+
+- **ID:** `2601.05125` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Ignacio de Rodrigo, Alvaro J. Lopez-Lopez, Jaime Boal
+- **abs:** https://arxiv.org/abs/2601.05125 · **PDF:** https://arxiv.org/pdf/2601.05125v1
+
+## 23120. Evaluative Fingerprints: Stable and Systematic Differences in LLM Evaluator Behavior
+
+- **ID:** `2601.05114` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Wajid Nasser
+- **abs:** https://arxiv.org/abs/2601.05114 · **PDF:** https://arxiv.org/pdf/2601.05114v1
+
+## 23121. Agent-as-a-Judge
+
+- **ID:** `2601.05111` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Runyang You, Hongru Cai, Caiqi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.05111 · **PDF:** https://arxiv.org/pdf/2601.05111v1
+
+## 23122. GlimpRouter: Efficient Collaborative Inference by Glimpsing One Token of Thoughts
+
+- **ID:** `2601.05110` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Wenhao Zeng, Xuteng Zhang, Yuling Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.05110 · **PDF:** https://arxiv.org/pdf/2601.05110v2
+
+## 23123. Controllable Memory Usage: Balancing Anchoring and Innovation in Long-Term Human-Agent Interaction
+
+- **ID:** `2601.05107` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Muzhao Tian, Zisu Huang, Xiaohua Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.05107 · **PDF:** https://arxiv.org/pdf/2601.05107v1
+
+## 23124. Token-Level LLM Collaboration via FusionRoute
+
+- **ID:** `2601.05106` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Nuoya Xiong, Yuhang Zhou, Hanqing Zeng, et al.
+- **abs:** https://arxiv.org/abs/2601.05106 · **PDF:** https://arxiv.org/pdf/2601.05106v3
+
+## 23125. How Human is AI? Examining the Impact of Emotional Prompts on Artificial and Human and Responsiveness
+
+- **ID:** `2601.05104` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Florence Bernays, Marco Henriques Pereira, Jochen Menges
+- **abs:** https://arxiv.org/abs/2601.05104 · **PDF:** https://arxiv.org/pdf/2601.05104v1
+
+## 23126. Semantically Orthogonal Framework for Citation Classification: Disentangling Intent and Content
+
+- **ID:** `2601.05103` · **primary:** `cs.DL` · **Submitted:** 2026-01-08
+- **Authors:** Changxu Duan, Zhiyin Tan
+- **abs:** https://arxiv.org/abs/2601.05103 · **PDF:** https://arxiv.org/pdf/2601.05103v1
+
+## 23127. Arabic Prompts with English Tools: A Benchmark
+
+- **ID:** `2601.05101` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Konstantin Kubrak, Ahmed El-Moselhy, Ammar Alsulami, et al.
+- **abs:** https://arxiv.org/abs/2601.05101 · **PDF:** https://arxiv.org/pdf/2601.05101v1
+
+## 23128. Multi-Disciplinary Dataset Discovery from Citation-Verified Literature Contexts
+
+- **ID:** `2601.05099` · **primary:** `cs.DL` · **Submitted:** 2026-01-08
+- **Authors:** Zhiyin Tan, Changxu Duan
+- **abs:** https://arxiv.org/abs/2601.05099 · **PDF:** https://arxiv.org/pdf/2601.05099v1
+
+## 23129. ECLIPSE: An Evolutionary Computation Library for Instrumentation Prototyping in Scientific Engineering
+
+- **ID:** `2601.05098` · **primary:** `cs.NE` · **Submitted:** 2026-01-08
+- **Authors:** Max Foreback, Evan Imata, Vincent Ragusa, et al.
+- **abs:** https://arxiv.org/abs/2601.05098 · **PDF:** https://arxiv.org/pdf/2601.05098v2
+
+## 23130. Advanced Multimodal Learning for Seizure Detection and Prediction: Concept, Challenges, and Future Directions
+
+- **ID:** `2601.05095` · **primary:** `cs.NE` · **Submitted:** 2026-01-08
+- **Authors:** Ijaz Ahmad, Faizan Ahmad, Sunday Timothy Aboyeji, et al.
+- **abs:** https://arxiv.org/abs/2601.05095 · **PDF:** https://arxiv.org/pdf/2601.05095v2
+
+## 23131. Code-Mix Sentiment Analysis on Hinglish Tweets
+
+- **ID:** `2601.05091` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Aashi Garg, Aneshya Das, Arshi Arya, et al.
+- **abs:** https://arxiv.org/abs/2601.05091 · **PDF:** https://arxiv.org/pdf/2601.05091v1
+
+## 23132. Driver-Intention Prediction with Deep Learning: Real-Time Brain-to-Vehicle Communication
+
+- **ID:** `2601.05084` · **primary:** `cs.HC` · **Submitted:** 2026-01-08
+- **Authors:** Niloufar Alavi, Swati Shah, Rezvan Alamian, et al.
+- **abs:** https://arxiv.org/abs/2601.05084 · **PDF:** https://arxiv.org/pdf/2601.05084v1
+
+## 23133. Driving on Registers
+
+- **ID:** `2601.05083` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Ellington Kirby, Alexandre Boulch, Yihong Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.05083 · **PDF:** https://arxiv.org/pdf/2601.05083v2
+
+## 23134. Exploring Student Expectations and Confidence in Learning Analytics
+
+- **ID:** `2601.05082` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Hayk Asatryan, Basile Tousside, Janis Mohr, et al.
+- **abs:** https://arxiv.org/abs/2601.05082 · **PDF:** https://arxiv.org/pdf/2601.05082v1
+
+## 23135. Chain-of-Sanitized-Thoughts: Plugging PII Leakage in CoT of Large Reasoning Models
+
+- **ID:** `2601.05076` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Arghyadeep Das, Sai Sreenivas Chintha, Rishiraj Girmal, et al.
+- **abs:** https://arxiv.org/abs/2601.05076 · **PDF:** https://arxiv.org/pdf/2601.05076v1
+
+## 23136. SemPA: Improving Sentence Embeddings of Large Language Models through Semantic Preference Alignment
+
+- **ID:** `2601.05075` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Ziyang Chen, Zhenxuan Huang, Yile Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.05075 · **PDF:** https://arxiv.org/pdf/2601.05075v1
+
+## 23137. Milestones over Outcome: Unlocking Geometric Reasoning with Sub-Goal Verifiable Reward
+
+- **ID:** `2601.05073` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Jianlong Chen, Daocheng Fu, Shengze Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.05073 · **PDF:** https://arxiv.org/pdf/2601.05073v1
+
+## 23138. Graph energy as a measure of community detectability in networks
+
+- **ID:** `2601.05065` · **primary:** `cs.SI` · **Submitted:** 2026-01-08
+- **Authors:** Lucas Böttcher, Mason A. Porter, Santo Fortunato
+- **abs:** https://arxiv.org/abs/2601.05065 · **PDF:** https://arxiv.org/pdf/2601.05065v1
+
+## 23139. Quantitative mapping from conventional MRI using self-supervised physics-guided deep learning: applications to a large-scale, clinically heterogeneous dataset
+
+- **ID:** `2601.05063` · **primary:** `physics.med-ph` · **Submitted:** 2026-01-08
+- **Authors:** Jelmer van Lune, Stefano Mandija, Oscar van der Heide, et al.
+- **abs:** https://arxiv.org/abs/2601.05063 · **PDF:** https://arxiv.org/pdf/2601.05063v1
+
+## 23140. Compositional Steering of Large Language Models with Steering Tokens
+
+- **ID:** `2601.05062` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Gorjan Radevski, Kiril Gashteovski, Giwon Hong, et al.
+- **abs:** https://arxiv.org/abs/2601.05062 · **PDF:** https://arxiv.org/pdf/2601.05062v1
+
+## 23141. From Understanding to Engagement: Personalized pharmacy Video Clips via Vision Language Models (VLMs)
+
+- **ID:** `2601.05059` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Suyash Mishra, Qiang Li, Srikanth Patil, et al.
+- **abs:** https://arxiv.org/abs/2601.05059 · **PDF:** https://arxiv.org/pdf/2601.05059v1
+
+## 23142. Reinforced Efficient Reasoning via Semantically Diverse Exploration
+
+- **ID:** `2601.05053` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Ziqi Zhao, Zhaochun Ren, Jiahong Zou, et al.
+- **abs:** https://arxiv.org/abs/2601.05053 · **PDF:** https://arxiv.org/pdf/2601.05053v1
+
+## 23143. DeepWeightFlow: Re-Basined Flow Matching for Generating Neural Network Weights
+
+- **ID:** `2601.05052` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Saumya Gupta, Scott Biggs, Moritz Laber, et al.
+- **abs:** https://arxiv.org/abs/2601.05052 · **PDF:** https://arxiv.org/pdf/2601.05052v1
+
+## 23144. Publishing FAIR and Machine-actionable Reviews in Materials Science: The Case for Symbolic Knowledge in Neuro-symbolic Artificial Intelligence
+
+- **ID:** `2601.05051` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Jennifer D'Souza, Soren Auer, Eleni Poupaki, et al.
+- **abs:** https://arxiv.org/abs/2601.05051 · **PDF:** https://arxiv.org/pdf/2601.05051v1
+
+## 23145. Large language models can effectively convince people to believe conspiracies
+
+- **ID:** `2601.05050` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Thomas H. Costello, Kellin Pelrine, Matthew Kowal, et al.
+- **abs:** https://arxiv.org/abs/2601.05050 · **PDF:** https://arxiv.org/pdf/2601.05050v2
+
+## 23146. How to Set the Learning Rate for Large-Scale Pre-training?
+
+- **ID:** `2601.05049` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Yunhua Zhou, Shuhao Xing, Junhao Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.05049 · **PDF:** https://arxiv.org/pdf/2601.05049v1
+
+## 23147. Challenges and Research Directions for Large Language Model Inference Hardware
+
+- **ID:** `2601.05047` · **primary:** `cs.AR` · **Submitted:** 2026-01-08
+- **Authors:** Xiaoyu Ma, David Patterson
+- **abs:** https://arxiv.org/abs/2601.05047 · **PDF:** https://arxiv.org/pdf/2601.05047v3
+
+## 23148. ArcAligner: Adaptive Recursive Aligner for Compressed Context Embeddings in RAG
+
+- **ID:** `2601.05038` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Jianbo Li, Yi Jiang, Sendong Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.05038 · **PDF:** https://arxiv.org/pdf/2601.05038v1
+
+## 23149. Exponential capacity scaling of classical GANs compared to hybrid latent style-based quantum GANs
+
+- **ID:** `2601.05036` · **primary:** `quant-ph` · **Submitted:** 2026-01-08
+- **Authors:** Milan Liepelt, Julien Baglio
+- **abs:** https://arxiv.org/abs/2601.05036 · **PDF:** https://arxiv.org/pdf/2601.05036v1
+
+## 23150. How to Set the Batch Size for Large-Scale Pre-training?
+
+- **ID:** `2601.05034` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Yunhua Zhou, Junhao Huang, Shuhao Xing, et al.
+- **abs:** https://arxiv.org/abs/2601.05034 · **PDF:** https://arxiv.org/pdf/2601.05034v2
+
+## 23151. A Data-Driven Predictive Framework for Inventory Optimization Using Context-Augmented Machine Learning Models
+
+- **ID:** `2601.05033` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Anees Fatima, Mohammad Abdus Salam
+- **abs:** https://arxiv.org/abs/2601.05033 · **PDF:** https://arxiv.org/pdf/2601.05033v1
+
+## 23152. Approximate equivariance via projection-based regularisation
+
+- **ID:** `2601.05028` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Torben Berndt, Jan Stühmer
+- **abs:** https://arxiv.org/abs/2601.05028 · **PDF:** https://arxiv.org/pdf/2601.05028v1
+
+## 23153. OptiSet: Unified Optimizing Set Selection and Ranking for Retrieval-Augmented Generation
+
+- **ID:** `2601.05027` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Yi Jiang, Sendong Zhao, Jianbo Li, et al.
+- **abs:** https://arxiv.org/abs/2601.05027 · **PDF:** https://arxiv.org/pdf/2601.05027v1
+
+## 23154. Hán Dān Xué Bù (Mimicry) or Qīng Chū Yú Lán (Mastery)? A Cognitive Perspective on Reasoning Distillation in Large Language Models
+
+- **ID:** `2601.05019` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Yueqing Hu, Xinyang Peng, Shuting Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.05019 · **PDF:** https://arxiv.org/pdf/2601.05019v1
+
+## 23155. HMVI: Unifying Heterogeneous Attributes with Natural Neighbors for Missing Value Inference
+
+- **ID:** `2601.05017` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Xiaopeng Luo, Zexi Tan, Zhuowei Wang
+- **abs:** https://arxiv.org/abs/2601.05017 · **PDF:** https://arxiv.org/pdf/2601.05017v1
+
+## 23156. From Idea to Co-Creation: A Planner-Actor-Critic Framework for Agent Augmented 3D Modeling
+
+- **ID:** `2601.05016` · **primary:** `cs.MA` · **Submitted:** 2026-01-08
+- **Authors:** Jin Gao, Saichandu Juluri
+- **abs:** https://arxiv.org/abs/2601.05016 · **PDF:** https://arxiv.org/pdf/2601.05016v1
+
+## 23157. Leveraging Prediction Entropy for Automatic Prompt Weighting in Zero-Shot Audio-Language Classification
+
+- **ID:** `2601.05011` · **primary:** `cs.SD` · **Submitted:** 2026-01-08
+- **Authors:** Karim El Khoury, Maxime Zanella, Tiffanie Godelaine, et al.
+- **abs:** https://arxiv.org/abs/2601.05011 · **PDF:** https://arxiv.org/pdf/2601.05011v1
+
+## 23158. An Empirical Investigation of Robustness in Large Language Models under Tabular Distortions
+
+- **ID:** `2601.05009` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Avik Dutta, Harshit Nigam, Hosein Hasanbeig, et al.
+- **abs:** https://arxiv.org/abs/2601.05009 · **PDF:** https://arxiv.org/pdf/2601.05009v1
+
+## 23159. Can Large Language Models Resolve Semantic Discrepancy in Self-Destructive Subcultures? Evidence from Jirai Kei
+
+- **ID:** `2601.05004` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Peng Wang, Xilin Tao, Siyi Yao, et al.
+- **abs:** https://arxiv.org/abs/2601.05004 · **PDF:** https://arxiv.org/pdf/2601.05004v1
+
+## 23160. On the Hidden Objective Biases of Group-based Reinforcement Learning
+
+- **ID:** `2601.05002` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Aleksandar Fontana, Marco Simoni, Giulio Rossolini, et al.
+- **abs:** https://arxiv.org/abs/2601.05002 · **PDF:** https://arxiv.org/pdf/2601.05002v1
+
+## 23161. AlgBench: To What Extent Do Large Reasoning Models Understand Algorithms?
+
+- **ID:** `2601.04996` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Henan Sun, Kaichi Yu, Yuyao Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.04996 · **PDF:** https://arxiv.org/pdf/2601.04996v2
+
+## 23162. Learning from Mistakes: Negative Reasoning Samples Enhance Out-of-Domain Generalization
+
+- **ID:** `2601.04992` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Xueyun Tian, Minghua Ma, Bingbing Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.04992 · **PDF:** https://arxiv.org/pdf/2601.04992v2
+
+## 23163. When to Act: Calibrated Confidence for Reliable Human Intention Prediction in Assistive Robotics
+
+- **ID:** `2601.04982` · **primary:** `cs.RO` · **Submitted:** 2026-01-08
+- **Authors:** Johannes A. Gaus, Winfried Ilg, Daniel Haeufle
+- **abs:** https://arxiv.org/abs/2601.04982 · **PDF:** https://arxiv.org/pdf/2601.04982v1
+
+## 23164. On the Definition and Detection of Cherry-Picking in Counterfactual Explanations
+
+- **ID:** `2601.04977` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** James Hinns, Sofie Goethals, Stephan Van der Veeken, et al.
+- **abs:** https://arxiv.org/abs/2601.04977 · **PDF:** https://arxiv.org/pdf/2601.04977v1
+
+## 23165. Effects of personality steering on cooperative behavior in Large Language Model agents
+
+- **ID:** `2601.05302` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Mizuki Sakai, Mizuki Yokoyama, Wakaba Tateishi, et al.
+- **abs:** https://arxiv.org/abs/2601.05302 · **PDF:** https://arxiv.org/pdf/2601.05302v2
+
+## 23166. ConMax: Confidence-Maximizing Compression for Efficient Chain-of-Thought Reasoning
+
+- **ID:** `2601.04973` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Minda Hu, Zexuan Qiu, Zenan Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.04973 · **PDF:** https://arxiv.org/pdf/2601.04973v1
+
+## 23167. Text as a Universal Interface for Transferable Personalization
+
+- **ID:** `2601.04963` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Yuting Liu, Jian Guan, Jia-Nan Li, et al.
+- **abs:** https://arxiv.org/abs/2601.04963 · **PDF:** https://arxiv.org/pdf/2601.04963v1
+
+## 23168. A Unified Spoken Language Model with Injected Emotional-Attribution Thinking for Human-like Interaction
+
+- **ID:** `2601.04960` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Qing Wang, Zehan Li, Yaodong Song, et al.
+- **abs:** https://arxiv.org/abs/2601.04960 · **PDF:** https://arxiv.org/pdf/2601.04960v1
+
+## 23169. Precision over Diversity: High-Precision Reward Generalizes to Robust Instruction Following
+
+- **ID:** `2601.04954` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Yirong Zeng, Yufei Liu, Xiao Ding, et al.
+- **abs:** https://arxiv.org/abs/2601.04954 · **PDF:** https://arxiv.org/pdf/2601.04954v2
+
+## 23170. Prototypicality Bias Reveals Blindspots in Multimodal Evaluation Metrics
+
+- **ID:** `2601.04946` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Subhadeep Roy, Gagan Bhatia, Steffen Eger
+- **abs:** https://arxiv.org/abs/2601.04946 · **PDF:** https://arxiv.org/pdf/2601.04946v2
+
+## 23171. T-Retriever: Tree-based Hierarchical Retrieval Augmented Generation for Textual Graphs
+
+- **ID:** `2601.04945` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Chunyu Wei, Huaiyu Qin, Siyuan He, et al.
+- **abs:** https://arxiv.org/abs/2601.04945 · **PDF:** https://arxiv.org/pdf/2601.04945v1
+
+## 23172. Cardinality augmented loss functions
+
+- **ID:** `2601.04941` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Miguel O'Malley
+- **abs:** https://arxiv.org/abs/2601.04941 · **PDF:** https://arxiv.org/pdf/2601.04941v1
+
+## 23173. CurricuLLM: Designing Personalized and Workforce-Aligned Cybersecurity Curricula Using Fine-Tuned LLMs
+
+- **ID:** `2601.04940` · **primary:** `cs.CR` · **Submitted:** 2026-01-08
+- **Authors:** Arthur Nijdam, Harri Kähkönen, Valtteri Niemi, et al.
+- **abs:** https://arxiv.org/abs/2601.04940 · **PDF:** https://arxiv.org/pdf/2601.04940v1
+
+## 23174. GenProve: Learning to Generate Text with Fine-Grained Provenance
+
+- **ID:** `2601.04932` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Jingxuan Wei, Xingyue Wang, Yanghaoyu Liao, et al.
+- **abs:** https://arxiv.org/abs/2601.04932 · **PDF:** https://arxiv.org/pdf/2601.04932v1
+
+## 23175. TIME: Temporally Intelligent Meta-reasoning Engine for Context Triggered Explicit Reasoning
+
+- **ID:** `2601.05300` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Susmit Das
+- **abs:** https://arxiv.org/abs/2601.05300 · **PDF:** https://arxiv.org/pdf/2601.05300v1
+
+## 23176. Can AI-Generated Persuasion Be Detected? Persuaficial Benchmark and AI vs. Human Linguistic Differences
+
+- **ID:** `2601.04925` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Arkadiusz Modzelewski, Paweł Golik, Anna Kołos, et al.
+- **abs:** https://arxiv.org/abs/2601.04925 · **PDF:** https://arxiv.org/pdf/2601.04925v1
+
+## 23177. Conversational AI for Rapid Scientific Prototyping: A Case Study on ESA's ELOPE Competition
+
+- **ID:** `2601.04920` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Nils Einecke
+- **abs:** https://arxiv.org/abs/2601.04920 · **PDF:** https://arxiv.org/pdf/2601.04920v1
+
+## 23178. What Students Ask, How a Generative AI Assistant Responds: Exploring Higher Education Students' Dialogues on Learning Analytics Feedback
+
+- **ID:** `2601.04919` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Yildiz Uzun, Andrea Gauthier, Mutlu Cukurova
+- **abs:** https://arxiv.org/abs/2601.04919 · **PDF:** https://arxiv.org/pdf/2601.04919v1
+
+## 23179. Breaking Robustness Barriers in Cognitive Diagnosis: A One-Shot Neural Architecture Search Perspective
+
+- **ID:** `2601.04918` · **primary:** `cs.IR` · **Submitted:** 2026-01-08
+- **Authors:** Ziwen Wang, Shangshang Yang, Xiaoshan Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.04918 · **PDF:** https://arxiv.org/pdf/2601.04918v1
+
+## 23180. From Stories to Cities to Games: A Qualitative Evaluation of Behaviour Planning
+
+- **ID:** `2601.04911` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Mustafa F. Abdelwahed, Joan Espasa, Alice Toniolo, et al.
+- **abs:** https://arxiv.org/abs/2601.04911 · **PDF:** https://arxiv.org/pdf/2601.04911v2
+
+## 23181. Distributed Online Convex Optimization with Efficient Communication: Improved Algorithm and Lower bounds
+
+- **ID:** `2601.04907` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Sifan Yang, Wenhao Yang, Wei Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.04907 · **PDF:** https://arxiv.org/pdf/2601.04907v2
+
+## 23182. Rotation-Robust Regression with Convolutional Model Trees
+
+- **ID:** `2601.04899` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Hongyi Li, William Ward Armstrong, Jun Xu
+- **abs:** https://arxiv.org/abs/2601.04899 · **PDF:** https://arxiv.org/pdf/2601.04899v1
+
+## 23183. V-FAT: Benchmarking Visual Fidelity Against Text-bias
+
+- **ID:** `2601.04897` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Ziteng Wang, Yujie He, Guanliang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.04897 · **PDF:** https://arxiv.org/pdf/2601.04897v1
+
+## 23184. DVD: A Robust Method for Detecting Variant Contamination in Large Language Model Evaluation
+
+- **ID:** `2601.04895` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Renzhao Liang, Jingru Chen, Bo Jia, et al.
+- **abs:** https://arxiv.org/abs/2601.04895 · **PDF:** https://arxiv.org/pdf/2601.04895v1
+
+## 23185. Scaling Vision Language Models for Pharmaceutical Long Form Video Reasoning on Industrial GenAI Platform
+
+- **ID:** `2601.04891` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Suyash Mishra, Qiang Li, Srikanth Patil, et al.
+- **abs:** https://arxiv.org/abs/2601.04891 · **PDF:** https://arxiv.org/pdf/2601.04891v1
+
+## 23186. Learnable Multipliers: Freeing the Scale of Language Model Matrix Layers
+
+- **ID:** `2601.04890` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Maksim Velikanov, Ilyas Chahed, Jingwei Zuo, et al.
+- **abs:** https://arxiv.org/abs/2601.04890 · **PDF:** https://arxiv.org/pdf/2601.04890v1
+
+## 23187. Faithful Summarisation under Disagreement via Belief-Level Aggregation
+
+- **ID:** `2601.04889` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Favour Yahdii Aghaebe, Tanefa Apekey, Elizabeth Williams, et al.
+- **abs:** https://arxiv.org/abs/2601.04889 · **PDF:** https://arxiv.org/pdf/2601.04889v1
+
+## 23188. SmartSearch: Process Reward-Guided Query Refinement for Search Agents
+
+- **ID:** `2601.04888` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Tongyu Wen, Guanting Dong, Zhicheng Dou
+- **abs:** https://arxiv.org/abs/2601.04888 · **PDF:** https://arxiv.org/pdf/2601.04888v1
+
+## 23189. Flexible Manufacturing Systems Intralogistics: Dynamic Optimization of AGVs and Tool Sharing Using Coloured-Timed Petri Nets and Actor-Critic RL with Actions Masking
+
+- **ID:** `2601.04887` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Sofiene Lassoued, Laxmikant Shrikant Bahetic, Nathalie Weiß-Borkowskib, et al.
+- **abs:** https://arxiv.org/abs/2601.04887 · **PDF:** https://arxiv.org/pdf/2601.04887v1
+
+## 23190. Analyzing Message-Code Inconsistency in AI Coding Agent-Authored Pull Requests
+
+- **ID:** `2601.04886` · **primary:** `cs.SE` · **Submitted:** 2026-01-08
+- **Authors:** Jingzhi Gong, Giovanni Pinna, Yixin Bian, et al.
+- **abs:** https://arxiv.org/abs/2601.04886 · **PDF:** https://arxiv.org/pdf/2601.04886v2
+
+## 23191. CuMA: Aligning LLMs with Sparse Cultural Values via Demographic-Aware Mixture of Adapters
+
+- **ID:** `2601.04885` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Ao Sun, Xiaoyu Wang, Zhe Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.04885 · **PDF:** https://arxiv.org/pdf/2601.04885v1
+
+## 23192. Precomputing Multi-Agent Path Replanning using Temporal Flexibility: A Case Study on the Dutch Railway Network
+
+- **ID:** `2601.04884` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Issa Hanou, Eric Kemmeren, Devin Wild Thomas, et al.
+- **abs:** https://arxiv.org/abs/2601.04884 · **PDF:** https://arxiv.org/pdf/2601.04884v1
+
+## 23193. Mind2Report: A Cognitive Deep Research Agent for Expert-Level Commercial Report Synthesis
+
+- **ID:** `2601.04879` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Mingyue Cheng, Daoyu Wang, Qi Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.04879 · **PDF:** https://arxiv.org/pdf/2601.04879v1
+
+## 23194. Higher-Order Knowledge Representations for Agentic Scientific Reasoning
+
+- **ID:** `2601.04878` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Isabella A. Stewart, Markus J. Buehler
+- **abs:** https://arxiv.org/abs/2601.04878 · **PDF:** https://arxiv.org/pdf/2601.04878v1
+
+## 23195. EvolSQL: Structure-Aware Evolution for Scalable Text-to-SQL Data Synthesis
+
+- **ID:** `2601.04875` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Xuanguang Pan, Chongyang Tao, Jiayuan Bai, et al.
+- **abs:** https://arxiv.org/abs/2601.04875 · **PDF:** https://arxiv.org/pdf/2601.04875v1
+
+## 23196. FibreCastML: An Open Web Platform for Predicting Electrospun Nanofibre Diameter Distributions
+
+- **ID:** `2601.04873` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Elisa Roldan, Kirstie Andrews, Stephen M. Richardson, et al.
+- **abs:** https://arxiv.org/abs/2601.04873 · **PDF:** https://arxiv.org/pdf/2601.04873v2
+
+## 23197. Gradient-based Optimisation of Modulation Effects
+
+- **ID:** `2601.04867` · **primary:** `eess.AS` · **Submitted:** 2026-01-08
+- **Authors:** Alistair Carson, Alec Wright, Stefan Bilbao
+- **abs:** https://arxiv.org/abs/2601.04867 · **PDF:** https://arxiv.org/pdf/2601.04867v1
+
+## 23198. Key-Value Pair-Free Continual Learner via Task-Specific Prompt-Prototype
+
+- **ID:** `2601.04864` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Haihua Luo, Xuming Ran, Zhengji Li, et al.
+- **abs:** https://arxiv.org/abs/2601.04864 · **PDF:** https://arxiv.org/pdf/2601.04864v2
+
+## 23199. Orchestrating Intelligence: Confidence-Aware Routing for Efficient Multi-Agent Collaboration across Multi-Scale Models
+
+- **ID:** `2601.04861` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Jingbo Wang, Sendong Zhao, Jiatong Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.04861 · **PDF:** https://arxiv.org/pdf/2601.04861v2
+
+## 23200. A Navigational Approach for Comprehensive RAG via Traversal over Proposition Graphs
+
+- **ID:** `2601.04859` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Maxime Delmas, Lei Xu, André Freitas
+- **abs:** https://arxiv.org/abs/2601.04859 · **PDF:** https://arxiv.org/pdf/2601.04859v1
+
+## 23201. Towards Public Administration Research Based on Interpretable Machine Learning
+
+- **ID:** `2601.06205` · **primary:** `cs.CY` · **Submitted:** 2026-01-08
+- **Authors:** Zhanyu Liu, Yang Yu
+- **abs:** https://arxiv.org/abs/2601.06205 · **PDF:** https://arxiv.org/pdf/2601.06205v1
+
+## 23202. MisSpans: Fine-Grained False Span Identification in Cross-Domain Fake News
+
+- **ID:** `2601.04857` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Zhiwei Liu, Paul Thompson, Jiaqi Rong, et al.
+- **abs:** https://arxiv.org/abs/2601.04857 · **PDF:** https://arxiv.org/pdf/2601.04857v1
+
+## 23203. Rethinking GNNs and Missing Features: Challenges, Evaluation and a Robust Solution
+
+- **ID:** `2601.04855` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Francesco Ferrini, Veronica Lachi, Antonio Longa, et al.
+- **abs:** https://arxiv.org/abs/2601.04855 · **PDF:** https://arxiv.org/pdf/2601.04855v1
+
+## 23204. Token Maturation: Autoregressive Language Generation via Continuous Token Dynamics
+
+- **ID:** `2601.04854` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Oshri Naparstek
+- **abs:** https://arxiv.org/abs/2601.04854 · **PDF:** https://arxiv.org/pdf/2601.04854v2
+
+## 23205. RAAR: Retrieval Augmented Agentic Reasoning for Cross-Domain Misinformation Detection
+
+- **ID:** `2601.04853` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Zhiwei Liu, Runteng Guo, Baojie Qu, et al.
+- **abs:** https://arxiv.org/abs/2601.04853 · **PDF:** https://arxiv.org/pdf/2601.04853v1
+
+## 23206. Optimizing Digital Adjudication through Social Network Analysis: An Empirical Study of Credit Card Disputes in Beijing
+
+- **ID:** `2601.05299` · **primary:** `cs.SI` · **Submitted:** 2026-01-08
+- **Authors:** Chung Han Tsai, ChengTo Lin, Chung Han Tsai, et al.
+- **abs:** https://arxiv.org/abs/2601.05299 · **PDF:** https://arxiv.org/pdf/2601.05299v1
+
+## 23207. When AI Settles Down: Late-Stage Stability as a Signature of AI-Generated Text Detection
+
+- **ID:** `2601.04833` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Ke Sun, Guangsheng Bao, Han Cui, et al.
+- **abs:** https://arxiv.org/abs/2601.04833 · **PDF:** https://arxiv.org/pdf/2601.04833v1
+
+## 23208. Illumination Angular Spectrum Encoding for Controlling the Functionality of Diffractive Networks
+
+- **ID:** `2601.04825` · **primary:** `physics.optics` · **Submitted:** 2026-01-08
+- **Authors:** Matan Kleiner, Lior Michaeli, Tomer Michaeli
+- **abs:** https://arxiv.org/abs/2601.04825 · **PDF:** https://arxiv.org/pdf/2601.04825v1
+
+## 23209. DR-LoRA: Dynamic Rank LoRA for Fine-Tuning Mixture-of-Experts Models
+
+- **ID:** `2601.04823` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Guanzhi Deng, Bo Li, Ronghao Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.04823 · **PDF:** https://arxiv.org/pdf/2601.04823v5
+
+## 23210. AECV-Bench: Benchmarking Multimodal Models on Architectural and Engineering Drawings Understanding
+
+- **ID:** `2601.04819` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Aleksei Kondratenko, Mussie Birhane, Houssame E. Hsain, et al.
+- **abs:** https://arxiv.org/abs/2601.04819 · **PDF:** https://arxiv.org/pdf/2601.04819v1
+
+## 23211. SCALER:Synthetic Scalable Adaptive Learning Environment for Reasoning
+
+- **ID:** `2601.04809` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Caijun Xu, Changyi Xiao, Zhongyuan Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.04809 · **PDF:** https://arxiv.org/pdf/2601.04809v3
+
+## 23212. Comparison of Maximum Likelihood Classification Before and After Applying Weierstrass Transform
+
+- **ID:** `2601.04808` · **primary:** `stat.AP` · **Submitted:** 2026-01-08
+- **Authors:** Muhammad Shoaib, Zaka Ur Rehman, Muhammad Qasim
+- **abs:** https://arxiv.org/abs/2601.04808 · **PDF:** https://arxiv.org/pdf/2601.04808v1
+
+## 23213. Parallelizing Node-Level Explainability in Graph Neural Networks
+
+- **ID:** `2601.04807` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Oscar Llorente, Jaime Boal, Eugenio F. Sánchez-Úbeda, et al.
+- **abs:** https://arxiv.org/abs/2601.04807 · **PDF:** https://arxiv.org/pdf/2601.04807v1
+
+## 23214. Thinking-Based Non-Thinking: Solving the Reward Hacking Problem in Training Hybrid Reasoning Models via Reinforcement Learning
+
+- **ID:** `2601.04805` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Siyuan Gan, Jiaheng Liu, Boyan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.04805 · **PDF:** https://arxiv.org/pdf/2601.04805v1
+
+## 23215. Mathematical Knowledge Graph-Driven Framework for Equation-Based Predictive and Reliable Additive Manufacturing
+
+- **ID:** `2601.05298` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Yeongbin Cha, Namjung Kim
+- **abs:** https://arxiv.org/abs/2601.05298 · **PDF:** https://arxiv.org/pdf/2601.05298v1
+
+## 23216. MPM-LLM4DSE: Reaching the Pareto Frontier in HLS with Multimodal Learning and LLM-Driven Exploration
+
+- **ID:** `2601.04801` · **primary:** `cs.AR` · **Submitted:** 2026-01-08
+- **Authors:** Lei Xu, Shanshan Wang, Chenglong Xiao
+- **abs:** https://arxiv.org/abs/2601.04801 · **PDF:** https://arxiv.org/pdf/2601.04801v1
+
+## 23217. Neural-Symbolic Integration with Evolvable Policies
+
+- **ID:** `2601.04799` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Marios Thoma, Vassilis Vassiliades, Loizos Michael
+- **abs:** https://arxiv.org/abs/2601.04799 · **PDF:** https://arxiv.org/pdf/2601.04799v1
+
+## 23218. Defense Against Indirect Prompt Injection via Tool Result Parsing
+
+- **ID:** `2601.04795` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Qiang Yu, Xinran Cheng, Chuanyi Liu
+- **abs:** https://arxiv.org/abs/2601.04795 · **PDF:** https://arxiv.org/pdf/2601.04795v1
+
+## 23219. APEX: Academic Poster Editing Agentic Expert
+
+- **ID:** `2601.04794` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Chengxin Shi, Qinnan Cai, Zeyuan Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.04794 · **PDF:** https://arxiv.org/pdf/2601.04794v1
+
+## 23220. Measurement-Consistent Langevin Corrector for Stabilizing Latent Diffusion Inverse Problem Solvers
+
+- **ID:** `2601.04791` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Lee Hyoseok, Sohwi Lim, Eunju Cha, et al.
+- **abs:** https://arxiv.org/abs/2601.04791 · **PDF:** https://arxiv.org/pdf/2601.04791v2
+
+## 23221. Machine learning assisted state prediction of misspecified linear dynamical system via modal reduction
+
+- **ID:** `2601.05297` · **primary:** `stat.ML` · **Submitted:** 2026-01-08
+- **Authors:** Rohan Vitthal Thorat, Rajdip Nayek
+- **abs:** https://arxiv.org/abs/2601.05297 · **PDF:** https://arxiv.org/pdf/2601.05297v1
+
+## 23222. Belief in Authority: Impact of Authority in Multi-Agent Evaluation Framework
+
+- **ID:** `2601.04790` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Junhyuk Choi, Jeongyoun Kwon, Heeju Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.04790 · **PDF:** https://arxiv.org/pdf/2601.04790v1
+
+## 23223. NC2C: Automated Convexification of Generic Non-Convex Optimization Problems
+
+- **ID:** `2601.04789` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Xinyue Peng, Yanming Liu, Yihan Cang, et al.
+- **abs:** https://arxiv.org/abs/2601.04789 · **PDF:** https://arxiv.org/pdf/2601.04789v1
+
+## 23224. AgentOCR: Reimagining Agent History via Optical Self-Compression
+
+- **ID:** `2601.04786` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Lang Feng, Fuchao Yang, Feng Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.04786 · **PDF:** https://arxiv.org/pdf/2601.04786v2
+
+## 23225. SRU-Pix2Pix: A Fusion-Driven Generator Network for Medical Image Translation with Few-Shot Learning
+
+- **ID:** `2601.04785` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Xihe Qiu, Yang Dai, Xiaoyu Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.04785 · **PDF:** https://arxiv.org/pdf/2601.04785v1
+
+## 23226. CounterVid: Counterfactual Video Generation for Mitigating Action and Temporal Hallucinations in Video-Language Models
+
+- **ID:** `2601.04778` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Tobia Poppi, Burak Uzkent, Amanmeet Garg, et al.
+- **abs:** https://arxiv.org/abs/2601.04778 · **PDF:** https://arxiv.org/pdf/2601.04778v1
+
+## 23227. GeM-VG: Towards Generalized Multi-image Visual Grounding with Multimodal Large Language Models
+
+- **ID:** `2601.04777` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Shurong Zheng, Yousong Zhu, Hongyin Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.04777 · **PDF:** https://arxiv.org/pdf/2601.04777v1
+
+## 23228. SciIF: Benchmarking Scientific Instruction Following Towards Rigorous Scientific Intelligence
+
+- **ID:** `2601.04770` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Encheng Su, Jianyu Wu, Chen Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.04770 · **PDF:** https://arxiv.org/pdf/2601.04770v2
+
+## 23229. RiskBridge: Turning CVEs into Business-Aligned Patch Priorities
+
+- **ID:** `2601.06201` · **primary:** `cs.SE` · **Submitted:** 2026-01-08
+- **Authors:** Yelena Mujibur Sheikh, Awez Akhtar Khatik, Luoxi Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.06201 · **PDF:** https://arxiv.org/pdf/2601.06201v1
+
+## 23230. LANGSAE EDITING: Improving Multilingual Information Retrieval via Post-hoc Language Identity Removal
+
+- **ID:** `2601.04768` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Dongjun Kim, Jeongho Yoon, Chanjun Park, et al.
+- **abs:** https://arxiv.org/abs/2601.04768 · **PDF:** https://arxiv.org/pdf/2601.04768v1
+
+## 23231. AT$^2$PO: Agentic Turn-based Policy Optimization via Tree Search
+
+- **ID:** `2601.04767` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Zefang Zong, Dingwei Chen, Yang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.04767 · **PDF:** https://arxiv.org/pdf/2601.04767v1
+
+## 23232. Revisiting Judge Decoding from First Principles via Training-Free Distributional Divergence
+
+- **ID:** `2601.04766` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Shengyin Sun, Yiming Li, Renxi Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.04766 · **PDF:** https://arxiv.org/pdf/2601.04766v1
+
+## 23233. Differential syntactic and semantic encoding in LLMs
+
+- **ID:** `2601.04765` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Santiago Acevedo, Alessandro Laio, Marco Baroni
+- **abs:** https://arxiv.org/abs/2601.04765 · **PDF:** https://arxiv.org/pdf/2601.04765v3
+
+## 23234. Orion-RAG: Path-Aligned Hybrid Retrieval for Graphless Data
+
+- **ID:** `2601.04764` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Zhen Chen, Weihao Xie, Peilin Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.04764 · **PDF:** https://arxiv.org/pdf/2601.04764v1
+
+## 23235. Smart IoT-Based Wearable Device for Detection and Monitoring of Common Cow Diseases Using a Novel Machine Learning Technique
+
+- **ID:** `2601.04761` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Rupsa Rani Mishra, D. Chandrasekhar Rao, Ajaya Kumar Tripathy
+- **abs:** https://arxiv.org/abs/2601.04761 · **PDF:** https://arxiv.org/pdf/2601.04761v1
+
+## 23236. PILOT-Bench: A Benchmark for Legal Reasoning in the Patent Domain with IRAC-Aligned Classification Tasks
+
+- **ID:** `2601.04758` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Yehoon Jang, Chaewon Lee, Hyun-seok Min, et al.
+- **abs:** https://arxiv.org/abs/2601.04758 · **PDF:** https://arxiv.org/pdf/2601.04758v1
+
+## 23237. Intraday spatiotemporal PV power prediction at national scale using satellite-based solar forecast models
+
+- **ID:** `2601.04751` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Luca Lanzilao, Angela Meyer
+- **abs:** https://arxiv.org/abs/2601.04751 · **PDF:** https://arxiv.org/pdf/2601.04751v1
+
+## 23238. When Single-Agent with Skills Replace Multi-Agent Systems and When They Fail
+
+- **ID:** `2601.04748` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Xiaoxiao Li
+- **abs:** https://arxiv.org/abs/2601.04748 · **PDF:** https://arxiv.org/pdf/2601.04748v2
+
+## 23239. KnowMe-Bench: Benchmarking Person Understanding for Lifelong Digital Companions
+
+- **ID:** `2601.04745` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Tingyu Wu, Zhisheng Chen, Ziyan Weng, et al.
+- **abs:** https://arxiv.org/abs/2601.04745 · **PDF:** https://arxiv.org/pdf/2601.04745v1
+
+## 23240. Semi-Supervised Diseased Detection from Speech Dialogues with Multi-Level Data Modeling
+
+- **ID:** `2601.04744` · **primary:** `cs.SD` · **Submitted:** 2026-01-08
+- **Authors:** Xingyuan Li, Mengyue Wu
+- **abs:** https://arxiv.org/abs/2601.04744 · **PDF:** https://arxiv.org/pdf/2601.04744v1
+
+## 23241. Tool-MAD: A Multi-Agent Debate Framework for Fact Verification with Diverse Tool Augmentation and Adaptive Retrieval
+
+- **ID:** `2601.04742` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Seyeon Jeong, Yeonjun Choi, JongWook Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.04742 · **PDF:** https://arxiv.org/pdf/2601.04742v1
+
+## 23242. Fast Mining and Dynamic Time-to-Event Prediction over Multi-sensor Data Streams
+
+- **ID:** `2601.04741` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Kota Nakamura, Koki Kawabata, Yasuko Matsubara, et al.
+- **abs:** https://arxiv.org/abs/2601.04741 · **PDF:** https://arxiv.org/pdf/2601.04741v2
+
+## 23243. StealthGraph: Exposing Domain-Specific Risks in LLMs through Knowledge-Graph-Guided Harmful Prompt Generation
+
+- **ID:** `2601.04740` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Huawei Zheng, Xinqi Jiang, Sen Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.04740 · **PDF:** https://arxiv.org/pdf/2601.04740v2
+
+## 23244. Leveraging Membership Inference Attacks for Privacy Measurement in Federated Learning for Remote Sensing Images
+
+- **ID:** `2601.06200` · **primary:** `cs.CR` · **Submitted:** 2026-01-08
+- **Authors:** Anh-Kiet Duong, Petra Gomez-Krämer, Hoàng-Ân Lê, et al.
+- **abs:** https://arxiv.org/abs/2601.06200 · **PDF:** https://arxiv.org/pdf/2601.06200v1
+
+## 23245. AM$^3$Safety: Towards Data Efficient Alignment of Multi-modal Multi-turn Safety for MLLMs
+
+- **ID:** `2601.04736` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Han Zhu, Jiale Chen, Chengkun Cai, et al.
+- **abs:** https://arxiv.org/abs/2601.04736 · **PDF:** https://arxiv.org/pdf/2601.04736v1
+
+## 23246. The Role of Quantum in Hybrid Quantum-Classical Neural Networks: A Realistic Assessment
+
+- **ID:** `2601.04732` · **primary:** `quant-ph` · **Submitted:** 2026-01-08
+- **Authors:** Dominik Freinberger, Philipp Moser
+- **abs:** https://arxiv.org/abs/2601.04732 · **PDF:** https://arxiv.org/pdf/2601.04732v1
+
+## 23247. Miner:Mining Intrinsic Mastery for Data-Efficient RL in Large Reasoning Models
+
+- **ID:** `2601.04731` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Shuyang Jiang, Yuhao Wang, Ya Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.04731 · **PDF:** https://arxiv.org/pdf/2601.04731v1
+
+## 23248. Automatic Classifiers Underdetect Emotions Expressed by Men
+
+- **ID:** `2601.04730` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Ivan Smirnov, Segun T. Aroyehun, Paul Plener, et al.
+- **abs:** https://arxiv.org/abs/2601.04730 · **PDF:** https://arxiv.org/pdf/2601.04730v1
+
+## 23249. Excess Description Length of Learning Generalizable Predictors
+
+- **ID:** `2601.04728` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Elizabeth Donoway, Hailey Joren, Fabien Roger, et al.
+- **abs:** https://arxiv.org/abs/2601.04728 · **PDF:** https://arxiv.org/pdf/2601.04728v1
+
+## 23250. Training a Custom CNN on Five Heterogeneous Image Datasets
+
+- **ID:** `2601.04727` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Anika Tabassum, Tasnuva Mahazabin Tuba, Nafisa Naznin
+- **abs:** https://arxiv.org/abs/2601.04727 · **PDF:** https://arxiv.org/pdf/2601.04727v1
+
+## 23251. Memory Matters More: Event-Centric Memory as a Logic Map for Agent Searching and Reasoning
+
+- **ID:** `2601.04726` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Yuyang Hu, Jiongnan Liu, Jiejun Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.04726 · **PDF:** https://arxiv.org/pdf/2601.04726v1
+
+## 23252. MoEBlaze: Breaking the Memory Wall for Efficient MoE Training on Modern GPUs
+
+- **ID:** `2601.05296` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Jiyuan Zhang, Yining Liu, Siqi Yan, et al.
+- **abs:** https://arxiv.org/abs/2601.05296 · **PDF:** https://arxiv.org/pdf/2601.05296v1
+
+## 23253. Qwen3-VL-Embedding and Qwen3-VL-Reranker: A Unified Framework for State-of-the-Art Multimodal Retrieval and Ranking
+
+- **ID:** `2601.04720` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Mingxin Li, Yanzhao Zhang, Dingkun Long, et al.
+- **abs:** https://arxiv.org/abs/2601.04720 · **PDF:** https://arxiv.org/pdf/2601.04720v2
+
+## 23254. GPU-Accelerated INT8 Quantization for KV Cache Compression in Large Language Models
+
+- **ID:** `2601.04719` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Maanas Taneja, Purab Shingvi
+- **abs:** https://arxiv.org/abs/2601.04719 · **PDF:** https://arxiv.org/pdf/2601.04719v1
+
+## 23255. Identifying and Mitigating Bottlenecks in Role-Playing Agents: A Systematic Study of Disentangling Character Profile Axes
+
+- **ID:** `2601.04716` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Yonghyun Jun, Junhyuk Choi, Jihyeong Park, et al.
+- **abs:** https://arxiv.org/abs/2601.04716 · **PDF:** https://arxiv.org/pdf/2601.04716v2
+
+## 23256. ThinkDrive: Chain-of-Thought Guided Progressive Reinforcement Learning Fine-Tuning for Autonomous Driving
+
+- **ID:** `2601.04714` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Chang Zhao, Zheming Yang, Yunqing Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.04714 · **PDF:** https://arxiv.org/pdf/2601.04714v1
+
+## 23257. DSC2025 -- ViHallu Challenge: Detecting Hallucination in Vietnamese LLMs
+
+- **ID:** `2601.04711` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Anh Thi-Hoang Nguyen, Khanh Quoc Tran, Tin Van Huynh, et al.
+- **abs:** https://arxiv.org/abs/2601.04711 · **PDF:** https://arxiv.org/pdf/2601.04711v1
+
+## 23258. Prior-Informed Zeroth-Order Optimization with Adaptive Direction Alignment for Memory-Efficient LLM Fine-Tuning
+
+- **ID:** `2601.04710` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Feihu Jin, Shipeng Cen, Ying Tan
+- **abs:** https://arxiv.org/abs/2601.04710 · **PDF:** https://arxiv.org/pdf/2601.04710v1
+
+## 23259. Bridging Temporal and Textual Modalities: A Multimodal Framework for Automated Cloud Failure Root Cause Analysis
+
+- **ID:** `2601.04709` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Gijun Park
+- **abs:** https://arxiv.org/abs/2601.04709 · **PDF:** https://arxiv.org/pdf/2601.04709v1
+
+## 23260. MQ-GNN: A Multi-Queue Pipelined Architecture for Scalable and Efficient GNN Training
+
+- **ID:** `2601.04707` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Irfan Ullah, Young-Koo Lee
+- **abs:** https://arxiv.org/abs/2601.04707 · **PDF:** https://arxiv.org/pdf/2601.04707v1
+
+## 23261. A zone-based training approach for last-mile routing using Graph Neural Networks and Pointer Networks
+
+- **ID:** `2601.04705` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Àngel Ruiz-Fas, Carlos Granell, José Francisco Ramos, et al.
+- **abs:** https://arxiv.org/abs/2601.04705 · **PDF:** https://arxiv.org/pdf/2601.04705v1
+
+## 23262. Beyond Monolithic Architectures: A Multi-Agent Search and Knowledge Optimization Framework for Agentic Search
+
+- **ID:** `2601.04703` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Yiqun Chen, Lingyong Yan, Zixuan Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.04703 · **PDF:** https://arxiv.org/pdf/2601.04703v1
+
+## 23263. PRISM: A Unified Framework for Post-Training LLMs Without Verifiable Rewards
+
+- **ID:** `2601.04700` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Mukesh Ghimire, Aosong Feng, Liwen You, et al.
+- **abs:** https://arxiv.org/abs/2601.04700 · **PDF:** https://arxiv.org/pdf/2601.04700v2
+
+## 23264. SeqWalker: Sequential-Horizon Vision-and-Language Navigation with Hierarchical Planning
+
+- **ID:** `2601.04699` · **primary:** `cs.RO` · **Submitted:** 2026-01-08
+- **Authors:** Zebin Han, Xudong Wang, Baichen Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.04699 · **PDF:** https://arxiv.org/pdf/2601.04699v1
+
+## 23265. TourPlanner: A Competitive Consensus Framework with Constraint-Gated Reinforcement Learning for Travel Planning
+
+- **ID:** `2601.04698` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Yinuo Wang, Mining Tan, Wenxiang Jiao, et al.
+- **abs:** https://arxiv.org/abs/2601.04698 · **PDF:** https://arxiv.org/pdf/2601.04698v1
+
+## 23266. A Method for Constructing a Digital Transformation Driving Mechanism Based on Semantic Understanding of Large Models
+
+- **ID:** `2601.04696` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Huayi Liu
+- **abs:** https://arxiv.org/abs/2601.04696 · **PDF:** https://arxiv.org/pdf/2601.04696v1
+
+## 23267. Tape: A Cellular Automata Benchmark for Evaluating Rule-Shift Generalization in Reinforcement Learning
+
+- **ID:** `2601.04695` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Enze Pan
+- **abs:** https://arxiv.org/abs/2601.04695 · **PDF:** https://arxiv.org/pdf/2601.04695v1
+
+## 23268. FastWhisper: Adaptive Self-knowledge Distillation for Real-time Automatic Speech Recognition
+
+- **ID:** `2601.19919` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Junseok Lee, Nahoon Kim, Sangyong Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.19919 · **PDF:** https://arxiv.org/pdf/2601.19919v1
+
+## 23269. ResMAS: Resilience Optimization in LLM-based Multi-agent Systems
+
+- **ID:** `2601.04694` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Zhilun Zhou, Zihan Liu, Jiahe Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.04694 · **PDF:** https://arxiv.org/pdf/2601.04694v1
+
+## 23270. Thunder-KoNUBench: A Corpus-Aligned Benchmark for Korean Negation Understanding
+
+- **ID:** `2601.04693` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Sungmok Jung, Yeonkyoung So, Joonhak Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.04693 · **PDF:** https://arxiv.org/pdf/2601.04693v1
+
+## 23271. See, Explain, and Intervene: A Few-Shot Multimodal Agent Framework for Hateful Meme Moderation
+
+- **ID:** `2601.04692` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Naquee Rizwan, Subhankar Swain, Paramananda Bhaskar, et al.
+- **abs:** https://arxiv.org/abs/2601.04692 · **PDF:** https://arxiv.org/pdf/2601.04692v1
+
+## 23272. Do LLMs Benefit from User and Item Embeddings in Recommendation Tasks?
+
+- **ID:** `2601.04690` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Mir Rayat Imtiaz Hossain, Leo Feng, Leonid Sigal, et al.
+- **abs:** https://arxiv.org/abs/2601.04690 · **PDF:** https://arxiv.org/pdf/2601.04690v1
+
+## 23273. ToolGate: Contract-Grounded and Verified Tool Execution for LLMs
+
+- **ID:** `2601.04688` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Yanming Liu, Xinyue Peng, Jiannan Cao, et al.
+- **abs:** https://arxiv.org/abs/2601.04688 · **PDF:** https://arxiv.org/pdf/2601.04688v1
+
+## 23274. Nightmare Dreamer: Dreaming About Unsafe States And Planning Ahead
+
+- **ID:** `2601.04686` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Oluwatosin Oseni, Shengjie Wang, Jun Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.04686 · **PDF:** https://arxiv.org/pdf/2601.04686v1
+
+## 23275. FastSLM: Hierarchical Frame Q-Former for Effective Speech Modality Adaptation
+
+- **ID:** `2601.06199` · **primary:** `eess.AS` · **Submitted:** 2026-01-08
+- **Authors:** Junseok Lee, Sangyong Lee, Chang-Jae Chun
+- **abs:** https://arxiv.org/abs/2601.06199 · **PDF:** https://arxiv.org/pdf/2601.06199v2
+
+## 23276. LLM-Guided Quantified SMT Solving over Uninterpreted Functions
+
+- **ID:** `2601.04675` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Kunhang Lv, Yuhang Dong, Rui Han, et al.
+- **abs:** https://arxiv.org/abs/2601.04675 · **PDF:** https://arxiv.org/pdf/2601.04675v1
+
+## 23277. Estimating Causal Effects in Gaussian Linear SCMs with Finite Data
+
+- **ID:** `2601.04673` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Aurghya Maiti, Prateek Jain
+- **abs:** https://arxiv.org/abs/2601.04673 · **PDF:** https://arxiv.org/pdf/2601.04673v1
+
+## 23278. Agri-R1: Empowering Generalizable Agricultural Reasoning in Vision-Language Models with Reinforcement Learning
+
+- **ID:** `2601.04672` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Wentao Zhang, Lifei Wang, Lina Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.04672 · **PDF:** https://arxiv.org/pdf/2601.04672v1
+
+## 23279. Learning Dynamics in RL Post-Training for Language Models
+
+- **ID:** `2601.04670` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Akiyoshi Tomihari
+- **abs:** https://arxiv.org/abs/2601.04670 · **PDF:** https://arxiv.org/pdf/2601.04670v1
+
+## 23280. Optimizing Path Planning using Deep Reinforcement Learning for UGVs in Precision Agriculture
+
+- **ID:** `2601.04668` · **primary:** `cs.RO` · **Submitted:** 2026-01-08
+- **Authors:** Laukik Patade, Rohan Rane, Sandeep Pillai
+- **abs:** https://arxiv.org/abs/2601.04668 · **PDF:** https://arxiv.org/pdf/2601.04668v1
+
+## 23281. Know Thy Enemy: Securing LLMs Against Prompt Injection via Diverse Data Synthesis and Instruction-Level Chain-of-Thought Learning
+
+- **ID:** `2601.04666` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Zhiyuan Chang, Mingyang Li, Yuekai Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.04666 · **PDF:** https://arxiv.org/pdf/2601.04666v1
+
+## 23282. CRANE: Causal Relevance Analysis of Language-Specific Neurons in Multilingual Large Language Models
+
+- **ID:** `2601.04664` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Yifan Le, Yunliang Li
+- **abs:** https://arxiv.org/abs/2601.04664 · **PDF:** https://arxiv.org/pdf/2601.04664v2
+
+## 23283. LAMB: LLM-based Audio Captioning with Modality Gap Bridging via Cauchy-Schwarz Divergence
+
+- **ID:** `2601.04658` · **primary:** `cs.SD` · **Submitted:** 2026-01-08
+- **Authors:** Hyeongkeun Lee, Jongmin Choi, KiHyun Nam, et al.
+- **abs:** https://arxiv.org/abs/2601.04658 · **PDF:** https://arxiv.org/pdf/2601.04658v2
+
+## 23284. LLMs-Integrated Automatic Hate Speech Recognition Using Controllable Text Generation Models
+
+- **ID:** `2601.04654` · **primary:** `eess.AS` · **Submitted:** 2026-01-08
+- **Authors:** Ryutaro Oshima, Yuya Hosoda, Youji Iiguni
+- **abs:** https://arxiv.org/abs/2601.04654 · **PDF:** https://arxiv.org/pdf/2601.04654v1
+
+## 23285. Vibe Coding an LLM-powered Theorem Prover
+
+- **ID:** `2601.04653` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Zhe Hou
+- **abs:** https://arxiv.org/abs/2601.04653 · **PDF:** https://arxiv.org/pdf/2601.04653v1
+
+## 23286. AI Safeguards, Generative AI and the Pandora Box: AI Safety Measures to Protect Businesses and Personal Reputation
+
+- **ID:** `2601.06197` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Prasanna Kumar
+- **abs:** https://arxiv.org/abs/2601.06197 · **PDF:** https://arxiv.org/pdf/2601.06197v1
+
+## 23287. Adversarial Yet Cooperative: Multi-Perspective Reasoning in Retrieved-Augmented Language Models
+
+- **ID:** `2601.04651` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Can Xu, Lingyong Yan, Jiayi Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.04651 · **PDF:** https://arxiv.org/pdf/2601.04651v2
+
+## 23288. Mechanism Design for Federated Learning with Non-Monotonic Network Effects
+
+- **ID:** `2601.04648` · **primary:** `cs.GT` · **Submitted:** 2026-01-08
+- **Authors:** Xiang Li, Bing Luo, Jianwei Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.04648 · **PDF:** https://arxiv.org/pdf/2601.04648v1
+
+## 23289. Succeeding at Scale: Automated Dataset Construction and Query-Side Adaptation for Multi-Tenant Search
+
+- **ID:** `2601.04646` · **primary:** `cs.IR` · **Submitted:** 2026-01-08
+- **Authors:** Prateek Jain, Shabari S Nair, Ritesh Goru, et al.
+- **abs:** https://arxiv.org/abs/2601.04646 · **PDF:** https://arxiv.org/pdf/2601.04646v3
+
+## 23290. DP-MGTD: Privacy-Preserving Machine-Generated Text Detection via Adaptive Differentially Private Entity Sanitization
+
+- **ID:** `2601.04641` · **primary:** `cs.CR` · **Submitted:** 2026-01-08
+- **Authors:** Lionel Z. Wang, Yusheng Zhao, Jiabin Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.04641 · **PDF:** https://arxiv.org/pdf/2601.04641v1
+
+## 23291. Manifold-based Sampling for In-Context Hallucination Detection in Large Language Models
+
+- **ID:** `2601.06196` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Bodla Krishna Vamshi, Rohan Bhatnagar, Haizhao Yang
+- **abs:** https://arxiv.org/abs/2601.06196 · **PDF:** https://arxiv.org/pdf/2601.06196v2
+
+## 23292. SpeechMedAssist: Efficiently and Effectively Adapting Speech Language Models for Medical Consultation
+
+- **ID:** `2601.04638` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Sirry Chen, Jieyi Wang, Wei Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.04638 · **PDF:** https://arxiv.org/pdf/2601.04638v1
+
+## 23293. MAGA-Bench: Machine-Augment-Generated Text via Alignment Detection Benchmark
+
+- **ID:** `2601.04633` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Anyang Song, Ying Cheng, Yiqian Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.04633 · **PDF:** https://arxiv.org/pdf/2601.04633v1
+
+## 23294. From National Curricula to Cultural Awareness: Constructing Open-Ended Culture-Specific Question Answering Dataset
+
+- **ID:** `2601.04632` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Haneul Yoo, Won Ik Cho, Geunhye Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.04632 · **PDF:** https://arxiv.org/pdf/2601.04632v1
+
+## 23295. Beyond the "Truth": Investigating Election Rumors on Truth Social During the 2024 Election
+
+- **ID:** `2601.04631` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Etienne Casanova, R. Michael Alvarez
+- **abs:** https://arxiv.org/abs/2601.04631 · **PDF:** https://arxiv.org/pdf/2601.04631v1
+
+## 23296. AgentDevel: Reframing Self-Evolving LLM Agents as Release Engineering
+
+- **ID:** `2601.04620` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Di Zhang
+- **abs:** https://arxiv.org/abs/2601.04620 · **PDF:** https://arxiv.org/pdf/2601.04620v1
+
+## 23297. DeepHalo: A Neural Choice Model with Controllable Context Effects
+
+- **ID:** `2601.04616` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Shuhan Zhang, Zhi Wang, Rui Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.04616 · **PDF:** https://arxiv.org/pdf/2601.04616v1
+
+## 23298. Commitment Checklist: Auditing Author Commitments in Peer Review
+
+- **ID:** `2603.00003` · **primary:** `cs.CY` · **Submitted:** 2026-01-08
+- **Authors:** Chung-Chi Chen, Iryna Gurevych
+- **abs:** https://arxiv.org/abs/2603.00003 · **PDF:** https://arxiv.org/pdf/2603.00003v1
+
+## 23299. Character-R1: Enhancing Role-Aware Reasoning in Role-Playing Agents via RLVR
+
+- **ID:** `2601.04611` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Yihong Tang, Kehai Chen, Xuefeng Bai, et al.
+- **abs:** https://arxiv.org/abs/2601.04611 · **PDF:** https://arxiv.org/pdf/2601.04611v1
+
+## 23300. Evaluating Human and Machine Confidence in Phishing Email Detection: A Comparative Study
+
+- **ID:** `2601.04610` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Paras Jain, Khushi Dhar, Olyemi E. Amujo, et al.
+- **abs:** https://arxiv.org/abs/2601.04610 · **PDF:** https://arxiv.org/pdf/2601.04610v1
+
+## 23301. When More Words Say Less: Decoupling Length and Specificity in Image Description Evaluation
+
+- **ID:** `2601.04609` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Rhea Kapur, Robert Hawkins, Elisa Kreiss
+- **abs:** https://arxiv.org/abs/2601.04609 · **PDF:** https://arxiv.org/pdf/2601.04609v1
+
+## 23302. Forecasting the U.S. Treasury Yield Curve: A Distributionally Robust Machine Learning Approach
+
+- **ID:** `2601.04608` · **primary:** `q-fin.MF` · **Submitted:** 2026-01-08
+- **Authors:** Jinjun Liu, Ming-Yen Cheng
+- **abs:** https://arxiv.org/abs/2601.04608 · **PDF:** https://arxiv.org/pdf/2601.04608v1
+
+## 23303. HUR-MACL: High-Uncertainty Region-Guided Multi-Architecture Collaborative Learning for Head and Neck Multi-Organ Segmentation
+
+- **ID:** `2601.04607` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** Xiaoyu Liu, Siwen Wei, Linhao Qu, et al.
+- **abs:** https://arxiv.org/abs/2601.04607 · **PDF:** https://arxiv.org/pdf/2601.04607v1
+
+## 23304. Crystal Generation using the Fully Differentiable Pipeline and Latent Space Optimization
+
+- **ID:** `2601.04606` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-08
+- **Authors:** Osman Goni Ridwan, Gilles Frapper, Hongfei Xue, et al.
+- **abs:** https://arxiv.org/abs/2601.04606 · **PDF:** https://arxiv.org/pdf/2601.04606v2
+
+## 23305. Constitutional Classifiers++: Efficient Production-Grade Defenses against Universal Jailbreaks
+
+- **ID:** `2601.04603` · **primary:** `cs.CR` · **Submitted:** 2026-01-08
+- **Authors:** Hoagy Cunningham, Jerry Wei, Zihan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.04603 · **PDF:** https://arxiv.org/pdf/2601.04603v1
+
+## 23306. On the Limitations of Rank-One Model Editing in Answering Multi-hop Questions
+
+- **ID:** `2601.04600` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Zhiyuan He, Binghan Chen, Tianxiang Xiong, et al.
+- **abs:** https://arxiv.org/abs/2601.04600 · **PDF:** https://arxiv.org/pdf/2601.04600v1
+
+## 23307. THaLLE-ThaiLLM: Domain-Specialized Small LLMs for Finance and Thai -- Technical Report
+
+- **ID:** `2601.04597` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** KBTG Labs, :, Anuruth Lertpiya, et al.
+- **abs:** https://arxiv.org/abs/2601.04597 · **PDF:** https://arxiv.org/pdf/2601.04597v1
+
+## 23308. Density Matrix RNN (DM-RNN): A Quantum Information Theoretic Framework for Modeling Musical Context and Polyphony
+
+- **ID:** `2601.04592` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Joonwon Seo, Mariana Montiel
+- **abs:** https://arxiv.org/abs/2601.04592 · **PDF:** https://arxiv.org/pdf/2601.04592v1
+
+## 23309. FedKDX: Federated Learning with Negative Knowledge Distillation for Enhanced Healthcare AI Systems
+
+- **ID:** `2601.04587` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Quang-Tu Pham, Hoang-Dieu Vu, Dinh-Dat Pham, et al.
+- **abs:** https://arxiv.org/abs/2601.04587 · **PDF:** https://arxiv.org/pdf/2601.04587v1
+
+## 23310. Autonomous Agents on Blockchains: Standards, Execution Models, and Trust Boundaries
+
+- **ID:** `2601.04583` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Saad Alqithami
+- **abs:** https://arxiv.org/abs/2601.04583 · **PDF:** https://arxiv.org/pdf/2601.04583v1
+
+## 23311. Aligning Text, Code, and Vision: A Multi-Objective Reinforcement Learning Framework for Text-to-Visualization
+
+- **ID:** `2601.04582` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Mizanur Rahman, Mohammed Saidul Islam, Md Tahmid Rahman Laskar, et al.
+- **abs:** https://arxiv.org/abs/2601.04582 · **PDF:** https://arxiv.org/pdf/2601.04582v1
+
+## 23312. Sci-Reasoning: A Dataset Decoding AI Innovation Patterns
+
+- **ID:** `2601.04577` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Jiachen Liu, Maestro Harmon, Zechen Zhang
+- **abs:** https://arxiv.org/abs/2601.04577 · **PDF:** https://arxiv.org/pdf/2601.04577v1
+
+## 23313. Scaling Behavior Cloning Improves Causal Reasoning: An Open Model for Real-Time Video Game Playing
+
+- **ID:** `2601.04575` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Yuguang Yue, Irakli Salia, Samuel Hunt, et al.
+- **abs:** https://arxiv.org/abs/2601.04575 · **PDF:** https://arxiv.org/pdf/2601.04575v2
+
+## 23314. FeedEval: Pedagogically Aligned Evaluation of LLM-Generated Essay Feedback
+
+- **ID:** `2601.04574` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Seongyeub Chu, Jongwoo Kim, Munyong Yi
+- **abs:** https://arxiv.org/abs/2601.04574 · **PDF:** https://arxiv.org/pdf/2601.04574v1
+
+## 23315. Spatial-Temporal Feedback Diffusion Guidance for Controlled Traffic Imputation
+
+- **ID:** `2601.04572` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Xiaowei Mao, Huihu Ding, Yan Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.04572 · **PDF:** https://arxiv.org/pdf/2601.04572v1
+
+## 23316. Enhancing Multimodal Retrieval via Complementary Information Extraction and Alignment
+
+- **ID:** `2601.04571` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Delong Zeng, Yuexiang Xie, Yaliang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.04571 · **PDF:** https://arxiv.org/pdf/2601.04571v1
+
+## 23317. Neurosymbolic Retrievers for Retrieval-augmented Generation
+
+- **ID:** `2601.04568` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Yash Saxena, Manas Gaur
+- **abs:** https://arxiv.org/abs/2601.04568 · **PDF:** https://arxiv.org/pdf/2601.04568v2
+
+## 23318. BackdoorAgent: A Unified Framework for Backdoor Attacks on LLM-based Agents
+
+- **ID:** `2601.04566` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Yunhao Feng, Yige Li, Yutao Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.04566 · **PDF:** https://arxiv.org/pdf/2601.04566v2
+
+## 23319. A Vision for Multisensory Intelligence: Sensing, Science, and Synergy
+
+- **ID:** `2601.04563` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Paul Pu Liang
+- **abs:** https://arxiv.org/abs/2601.04563 · **PDF:** https://arxiv.org/pdf/2601.04563v3
+
+## 23320. Reasoning Over Space: Enabling Geographic Reasoning for LLM-Based Generative Next POI Recommendation
+
+- **ID:** `2601.04562` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Dongyi Lv, Qiuyu Ding, Heng-Da Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.04562 · **PDF:** https://arxiv.org/pdf/2601.04562v1
+
+## 23321. Improving Semi-Supervised Contrastive Learning via Entropy-Weighted Confidence Integration of Anchor-Positive Pairs
+
+- **ID:** `2601.04555` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Shogo Nakayama, Masahiro Okuda
+- **abs:** https://arxiv.org/abs/2601.04555 · **PDF:** https://arxiv.org/pdf/2601.04555v1
+
+## 23322. EntroLnn: Entropy-Guided Liquid Neural Networks for Operando Refinement of Battery Capacity Fade Trajectories
+
+- **ID:** `2601.06195` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Wei Li, Wei Zhang, Qingyu Yan
+- **abs:** https://arxiv.org/abs/2601.06195 · **PDF:** https://arxiv.org/pdf/2601.06195v1
+
+## 23323. GEnSHIN: Graphical Enhanced Spatio-temporal Hierarchical Inference Network for Traffic Flow Prediction
+
+- **ID:** `2601.04550` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Zhiyan Zhou, Junjie Liao, Manho Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.04550 · **PDF:** https://arxiv.org/pdf/2601.04550v1
+
+## 23324. Political Alignment in Large Language Models: A Multidimensional Audit of Psychometric Identity and Behavioral Bias
+
+- **ID:** `2601.06194` · **primary:** `cs.CY` · **Submitted:** 2026-01-08
+- **Authors:** Adib Sakhawat, Tahsin Islam, Takia Farhin, et al.
+- **abs:** https://arxiv.org/abs/2601.06194 · **PDF:** https://arxiv.org/pdf/2601.06194v2
+
+## 23325. Identifying Good and Bad Neurons for Task-Level Controllable LLMs
+
+- **ID:** `2601.04548` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Wenjie Li, Guansong Pang, Hezhe Qiao, et al.
+- **abs:** https://arxiv.org/abs/2601.04548 · **PDF:** https://arxiv.org/pdf/2601.04548v2
+
+## 23326. Personalized Model-Based Design of Human Centric AI enabled CPS for Long term usage
+
+- **ID:** `2601.04545` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Bernard Ngabonziza, Ayan Banerjee, Sandeep K. S. Gupta
+- **abs:** https://arxiv.org/abs/2601.04545 · **PDF:** https://arxiv.org/pdf/2601.04545v1
+
+## 23327. TCAndon-Router: Adaptive Reasoning Router for Multi-Agent Collaboration
+
+- **ID:** `2601.04544` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Jiuzhou Zhao, Chunrong Chen, Chenqi Qiao, et al.
+- **abs:** https://arxiv.org/abs/2601.04544 · **PDF:** https://arxiv.org/pdf/2601.04544v1
+
+## 23328. Timeliness-Oriented Scheduling and Resource Allocation in Multi-Region Collaborative Perception
+
+- **ID:** `2601.04542` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Mengmeng Zhu, Yuxuan Sun, Yukuan Jia, et al.
+- **abs:** https://arxiv.org/abs/2601.04542 · **PDF:** https://arxiv.org/pdf/2601.04542v1
+
+## 23329. AdaptEval: A Benchmark for Evaluating Large Language Models on Code Snippet Adaptation
+
+- **ID:** `2601.04540` · **primary:** `cs.SE` · **Submitted:** 2026-01-08
+- **Authors:** Tanghaoran Zhang, Xinjun Mao, Shangwen Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.04540 · **PDF:** https://arxiv.org/pdf/2601.04540v1
+
+## 23330. Paradoxical noise preference in RNNs
+
+- **ID:** `2601.04539` · **primary:** `cs.NE` · **Submitted:** 2026-01-08
+- **Authors:** Noah Eckstein, Manoj Srinivasan
+- **abs:** https://arxiv.org/abs/2601.04539 · **PDF:** https://arxiv.org/pdf/2601.04539v1
+
+## 23331. Not All Steps are Informative: On the Linearity of LLMs' RLVR Training
+
+- **ID:** `2601.04537` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Tianle Wang, Zhongyuan Wu, Shenghao Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.04537 · **PDF:** https://arxiv.org/pdf/2601.04537v2
+
+## 23332. BanglaLorica: Design and Evaluation of a Robust Watermarking Algorithm for Large Language Models in Bangla Text Generation
+
+- **ID:** `2601.04534` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Amit Bin Tariqul, A N M Zahid Hossain Milkan, Sahab-Al-Chowdhury, et al.
+- **abs:** https://arxiv.org/abs/2601.04534 · **PDF:** https://arxiv.org/pdf/2601.04534v1
+
+## 23333. Self-MedRAG: a Self-Reflective Hybrid Retrieval-Augmented Generation Framework for Reliable Medical Question Answering
+
+- **ID:** `2601.04531` · **primary:** `cs.IR` · **Submitted:** 2026-01-08
+- **Authors:** Jessica Ryan, Alexander I. Gumilang, Robert Wiliam, et al.
+- **abs:** https://arxiv.org/abs/2601.04531 · **PDF:** https://arxiv.org/pdf/2601.04531v1
+
+## 23334. Advancing Language Models for Code-related Tasks
+
+- **ID:** `2601.04526` · **primary:** `cs.SE` · **Submitted:** 2026-01-08
+- **Authors:** Zhao Tian
+- **abs:** https://arxiv.org/abs/2601.04526 · **PDF:** https://arxiv.org/pdf/2601.04526v1
+
+## 23335. GRACE: Reinforcement Learning for Grounded Response and Abstention under Contextual Evidence
+
+- **ID:** `2601.04525` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Yibo Zhao, Jiapeng Zhu, Zichen Ding, et al.
+- **abs:** https://arxiv.org/abs/2601.04525 · **PDF:** https://arxiv.org/pdf/2601.04525v1
+
+## 23336. A Survey of Agentic AI and Cybersecurity: Challenges, Opportunities and Use-case Prototypes
+
+- **ID:** `2601.05293` · **primary:** `cs.CR` · **Submitted:** 2026-01-08
+- **Authors:** Sahaya Jestus Lazer, Kshitiz Aryal, Maanak Gupta, et al.
+- **abs:** https://arxiv.org/abs/2601.05293 · **PDF:** https://arxiv.org/pdf/2601.05293v1
+
+## 23337. BioPIE: A Biomedical Protocol Information Extraction Dataset for High-Reasoning-Complexity Experiment Question Answer
+
+- **ID:** `2601.04524` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Haofei Hou, Shunyi Zhao, Fanxu Meng, et al.
+- **abs:** https://arxiv.org/abs/2601.04524 · **PDF:** https://arxiv.org/pdf/2601.04524v1
+
+## 23338. MLB: A Scenario-Driven Benchmark for Evaluating Large Language Models in Clinical Applications
+
+- **ID:** `2601.06193` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Qing He, Dongsheng Bi, Jianrong Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.06193 · **PDF:** https://arxiv.org/pdf/2601.06193v1
+
+## 23339. The Ontological Neutrality Theorem: Why Neutral Ontological Substrates Must Be Pre-Causal and Pre-Normative
+
+- **ID:** `2601.14271` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Denise M. Case
+- **abs:** https://arxiv.org/abs/2601.14271 · **PDF:** https://arxiv.org/pdf/2601.14271v1
+
+## 23340. TSSR: Two-Stage Swap-Reward-Driven Reinforcement Learning for Character-Level SMILES Generation
+
+- **ID:** `2601.04521` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Jacob Ede Levine, Yun Lyan Luo, Sai Chandra Kosaraju
+- **abs:** https://arxiv.org/abs/2601.04521 · **PDF:** https://arxiv.org/pdf/2601.04521v2
+
+## 23341. Integrating Distribution Matching into Semi-Supervised Contrastive Learning for Labeled and Unlabeled Data
+
+- **ID:** `2601.04518` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Shogo Nakayama, Masahiro Okuda
+- **abs:** https://arxiv.org/abs/2601.04518 · **PDF:** https://arxiv.org/pdf/2601.04518v1
+
+## 23342. Bridging Distance and Spectral Positional Encodings via Anchor-Based Diffusion Geometry Approximation
+
+- **ID:** `2601.04517` · **primary:** `cs.IT` · **Submitted:** 2026-01-08
+- **Authors:** Zimo Yan, Zheng Xie, Runfan Duan, et al.
+- **abs:** https://arxiv.org/abs/2601.04517 · **PDF:** https://arxiv.org/pdf/2601.04517v1
+
+## 23343. LinguaGame: A Linguistically Grounded Game-Theoretic Paradigm for Multi-Agent Dialogue Generation
+
+- **ID:** `2601.04516` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Yuxiao Ye, Yiming Zhang, Yiran Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.04516 · **PDF:** https://arxiv.org/pdf/2601.04516v1
+
+## 23344. Multiagent Reinforcement Learning with Neighbor Action Estimation
+
+- **ID:** `2601.04511` · **primary:** `cs.RO` · **Submitted:** 2026-01-08
+- **Authors:** Zhenglong Luo, Zhiyong Chen, Aoxiang Liu
+- **abs:** https://arxiv.org/abs/2601.04511 · **PDF:** https://arxiv.org/pdf/2601.04511v1
+
+## 23345. Towards Spatio-Temporal Extrapolation of Phase-Field Simulations with Convolution-Only Neural Networks
+
+- **ID:** `2601.04510` · **primary:** `cs.CE` · **Submitted:** 2026-01-08
+- **Authors:** Christophe Bonneville, Nathan Bieberdorf, Pieterjan Robbe, et al.
+- **abs:** https://arxiv.org/abs/2601.04510 · **PDF:** https://arxiv.org/pdf/2601.04510v2
+
+## 23346. TimeGNN-Augmented Hybrid-Action MARL for Fine-Grained Task Partitioning and Energy-Aware Offloading in MEC
+
+- **ID:** `2601.06191` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Wei Ai, Yun Peng, Yuntao Shou, et al.
+- **abs:** https://arxiv.org/abs/2601.06191 · **PDF:** https://arxiv.org/pdf/2601.06191v1
+
+## 23347. A General Neural Backbone for Mixed-Integer Linear Optimization via Dual Attention
+
+- **ID:** `2601.04509` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Peixin Huang, Yaoxin Wu, Yining Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.04509 · **PDF:** https://arxiv.org/pdf/2601.04509v1
+
+## 23348. WESR: Scaling and Evaluating Word-level Event-Speech Recognition
+
+- **ID:** `2601.04508` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Chenchen Yang, Kexin Huang, Liwei Fan, et al.
+- **abs:** https://arxiv.org/abs/2601.04508 · **PDF:** https://arxiv.org/pdf/2601.04508v1
+
+## 23349. A Semi-supervised Molecular Learning Framework for Activity Cliff Estimation
+
+- **ID:** `2601.04507` · **primary:** `cs.CE` · **Submitted:** 2026-01-08
+- **Authors:** Fang Wu
+- **abs:** https://arxiv.org/abs/2601.04507 · **PDF:** https://arxiv.org/pdf/2601.04507v1
+
+## 23350. Surface-based Molecular Design with Multi-modal Flow Matching
+
+- **ID:** `2601.04506` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Fang Wu, Zhengyuan Zhou, Shuting Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.04506 · **PDF:** https://arxiv.org/pdf/2601.04506v1
+
+## 23351. CircuitLM: A Multi-Agent LLM-Aided Design Framework for Generating Circuit Schematics from Natural Language Prompts
+
+- **ID:** `2601.04505` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Khandakar Shakib Al Hasan, Syed Rifat Raiyan, Hasin Mahtab Alvee, et al.
+- **abs:** https://arxiv.org/abs/2601.04505 · **PDF:** https://arxiv.org/pdf/2601.04505v2
+
+## 23352. Specific Emitter Identification via Active Learning
+
+- **ID:** `2601.04502` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Jingyi Wang, Fanggang Wang
+- **abs:** https://arxiv.org/abs/2601.04502 · **PDF:** https://arxiv.org/pdf/2601.04502v1
+
+## 23353. The Minary Primitive of Computational Autopoiesis
+
+- **ID:** `2601.04501` · **primary:** `math.DS` · **Submitted:** 2026-01-08
+- **Authors:** Daniel Connor, Colin Defant
+- **abs:** https://arxiv.org/abs/2601.04501 · **PDF:** https://arxiv.org/pdf/2601.04501v1
+
+## 23354. GUITester: Enabling GUI Agents for Exploratory Defect Discovery
+
+- **ID:** `2601.04500` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Yifei Gao, Jiang Wu, Xiaoyi Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.04500 · **PDF:** https://arxiv.org/pdf/2601.04500v1
+
+## 23355. A Generalized Adaptive Joint Learning Framework for High-Dimensional Time-Varying Models
+
+- **ID:** `2601.04499` · **primary:** `stat.ME` · **Submitted:** 2026-01-08
+- **Authors:** Baolin Chen, Mengfei Ran
+- **abs:** https://arxiv.org/abs/2601.04499 · **PDF:** https://arxiv.org/pdf/2601.04499v2
+
+## 23356. IGenBench: Benchmarking the Reliability of Text-to-Infographic Generation
+
+- **ID:** `2601.04498` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Yinghao Tang, Xueding Liu, Boyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.04498 · **PDF:** https://arxiv.org/pdf/2601.04498v1
+
+## 23357. Vision-Language Agents for Interactive Forest Change Analysis
+
+- **ID:** `2601.04497` · **primary:** `cs.CV` · **Submitted:** 2026-01-08
+- **Authors:** James Brock, Ce Zhang, Nantheera Anantrasirichai
+- **abs:** https://arxiv.org/abs/2601.04497 · **PDF:** https://arxiv.org/pdf/2601.04497v2
+
+## 23358. Rational Synthesizers or Heuristic Followers? Analyzing LLMs in RAG-based Question-Answering
+
+- **ID:** `2601.06189` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Atharv Naphade
+- **abs:** https://arxiv.org/abs/2601.06189 · **PDF:** https://arxiv.org/pdf/2601.06189v1
+
+## 23359. Scalable Floating-Point Satisfiability via Staged Optimization
+
+- **ID:** `2601.04492` · **primary:** `cs.PL` · **Submitted:** 2026-01-08
+- **Authors:** Yuanzhuo Zhang, Zhoulai Fu, Binoy Ravindran
+- **abs:** https://arxiv.org/abs/2601.04492 · **PDF:** https://arxiv.org/pdf/2601.04492v1
+
+## 23360. A Closed-Loop Multi-Agent System Driven by LLMs for Meal-Level Personalized Nutrition Management
+
+- **ID:** `2601.04491` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Muqing Xu
+- **abs:** https://arxiv.org/abs/2601.04491 · **PDF:** https://arxiv.org/pdf/2601.04491v1
+
+## 23361. Decision-Aware Trust Signal Alignment for SOC Alert Triage
+
+- **ID:** `2601.04486` · **primary:** `cs.CR` · **Submitted:** 2026-01-08
+- **Authors:** Israt Jahan Chowdhury, Md Abu Yousuf Tanvir
+- **abs:** https://arxiv.org/abs/2601.04486 · **PDF:** https://arxiv.org/pdf/2601.04486v1
+
+## 23362. Hybrid Federated Learning for Noise-Robust Training
+
+- **ID:** `2601.04483` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Yongjun Kim, Hyeongjun Park, Hwanjin Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.04483 · **PDF:** https://arxiv.org/pdf/2601.04483v1
+
+## 23363. When Models Manipulate Manifolds: The Geometry of a Counting Task
+
+- **ID:** `2601.04480` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Wes Gurnee, Emmanuel Ameisen, Isaac Kauvar, et al.
+- **abs:** https://arxiv.org/abs/2601.04480 · **PDF:** https://arxiv.org/pdf/2601.04480v1
+
+## 23364. Prediction of Cellular Malignancy Using Electrical Impedance Signatures and Supervised Machine Learning
+
+- **ID:** `2601.04478` · **primary:** `eess.SP` · **Submitted:** 2026-01-08
+- **Authors:** Shadeeb Hossain
+- **abs:** https://arxiv.org/abs/2601.04478 · **PDF:** https://arxiv.org/pdf/2601.04478v4
+
+## 23365. Computational Compliance for AI Regulation: Blueprint for a New Research Domain
+
+- **ID:** `2601.04474` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Bill Marino, Nicholas D. Lane
+- **abs:** https://arxiv.org/abs/2601.04474 · **PDF:** https://arxiv.org/pdf/2601.04474v1
+
+## 23366. Convergence Rates for Learning Pseudo-Differential Operators
+
+- **ID:** `2601.04473` · **primary:** `math.ST` · **Submitted:** 2026-01-08
+- **Authors:** Jiaheng Chen, Daniel Sanz-Alonso
+- **abs:** https://arxiv.org/abs/2601.04473 · **PDF:** https://arxiv.org/pdf/2601.04473v1
+
+## 23367. SampoNLP: A Self-Referential Toolkit for Morphological Analysis of Subword Tokenizers
+
+- **ID:** `2601.04469` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Iaroslav Chelombitko, Ekaterina Chelombitko, Aleksey Komissarov
+- **abs:** https://arxiv.org/abs/2601.04469 · **PDF:** https://arxiv.org/pdf/2601.04469v1
+
+## 23368. Concept Tokens: Learning Behavioral Embeddings Through Concept Definitions
+
+- **ID:** `2601.04465` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Ignacio Sastre, Aiala Rosá
+- **abs:** https://arxiv.org/abs/2601.04465 · **PDF:** https://arxiv.org/pdf/2601.04465v1
+
+## 23369. Beyond Static Summarization: Proactive Memory Extraction for LLM Agents
+
+- **ID:** `2601.04463` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Chengyuan Yang, Zequn Sun, Wei Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.04463 · **PDF:** https://arxiv.org/pdf/2601.04463v1
+
+## 23370. Meta-probabilistic Modeling
+
+- **ID:** `2601.04462` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Kevin Zhang, Yixin Wang
+- **abs:** https://arxiv.org/abs/2601.04462 · **PDF:** https://arxiv.org/pdf/2601.04462v2
+
+## 23371. Users Mispredict Their Own Preferences for AI Writing Assistance
+
+- **ID:** `2601.04461` · **primary:** `cs.CL` · **Submitted:** 2026-01-08
+- **Authors:** Vivian Lai, Zana Buçinca, Nil-Jana Akpinar, et al.
+- **abs:** https://arxiv.org/abs/2601.04461 · **PDF:** https://arxiv.org/pdf/2601.04461v1
+
+## 23372. Using Large Language Models to Detect Socially Shared Regulation of Collaborative Learning
+
+- **ID:** `2601.04458` · **primary:** `cs.LG` · **Submitted:** 2026-01-08
+- **Authors:** Jiayi Zhang, Conrad Borchers, Clayton Cohn, et al.
+- **abs:** https://arxiv.org/abs/2601.04458 · **PDF:** https://arxiv.org/pdf/2601.04458v1
+
+## 23373. Large-Scale Continual Scheduling and Execution for Dynamic Distributed Satellite Constellation Observation Allocation
+
+- **ID:** `2601.06188` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Itai Zilberstein, Steve Chien
+- **abs:** https://arxiv.org/abs/2601.06188 · **PDF:** https://arxiv.org/pdf/2601.06188v2
+
+## 23374. Categorical Belief Propagation: Sheaf-Theoretic Inference via Descent and Holonomy
+
+- **ID:** `2601.04456` · **primary:** `cs.AI` · **Submitted:** 2026-01-08
+- **Authors:** Enrique ter Horst, Sridhar Mahadevan, Juan Diego Zambrano
+- **abs:** https://arxiv.org/abs/2601.04456 · **PDF:** https://arxiv.org/pdf/2601.04456v1
+
+## 23375. Re-Rankers as Relevance Judges
+
+- **ID:** `2601.04455` · **primary:** `cs.IR` · **Submitted:** 2026-01-08
+- **Authors:** Chuan Meng, Jiqun Liu, Mohammad Aliannejadi, et al.
+- **abs:** https://arxiv.org/abs/2601.04455 · **PDF:** https://arxiv.org/pdf/2601.04455v1
+
+## 23376. Time-Series Anomaly Classification for Launch Vehicle Propulsion Systems: Fast Statistical Detectors Enhancing LSTM Accuracy and Data Quality
+
+- **ID:** `2601.06186` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Sean P. Engelstad, Sameul R. Darr, Matthew Taliaferro, et al.
+- **abs:** https://arxiv.org/abs/2601.06186 · **PDF:** https://arxiv.org/pdf/2601.06186v1
+
+## 23377. Explainable Admission-Level Predictive Modeling for Prolonged Hospital Stay in Elderly Populations: Challenges in Low- and Middle-Income Countries
+
+- **ID:** `2601.04449` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Daniel Sierra-Botero, Ana Molina-Taborda, Leonardo Espinosa-Leal, et al.
+- **abs:** https://arxiv.org/abs/2601.04449 · **PDF:** https://arxiv.org/pdf/2601.04449v1
+
+## 23378. Merging Triggers, Breaking Backdoors: Defensive Poisoning for Instruction-Tuned Language Models
+
+- **ID:** `2601.04448` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** San Kim, Gary Geunbae Lee
+- **abs:** https://arxiv.org/abs/2601.04448 · **PDF:** https://arxiv.org/pdf/2601.04448v2
+
+## 23379. When Predictions Shape Reality: A Socio-Technical Synthesis of Performative Predictions in Machine Learning
+
+- **ID:** `2601.04447` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Gal Fybish, Teo Susnjak
+- **abs:** https://arxiv.org/abs/2601.04447 · **PDF:** https://arxiv.org/pdf/2601.04447v1
+
+## 23380. SpectraFormer: an Attention-Based Raman Unmixing Tool for Accessing the Graphene Buffer-Layer Signature on SiC
+
+- **ID:** `2601.04445` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-07
+- **Authors:** Dmitriy Poteryayev, Pietro Novelli, Annalisa Coriolano, et al.
+- **abs:** https://arxiv.org/abs/2601.04445 · **PDF:** https://arxiv.org/pdf/2601.04445v1
+
+## 23381. Large Language Models for Detecting Cyberattacks on Smart Grid Protective Relays
+
+- **ID:** `2601.04443` · **primary:** `cs.CR` · **Submitted:** 2026-01-07
+- **Authors:** Ahmad Mohammad Saber, Saeed Jafari, Zhengmao Ouyang, et al.
+- **abs:** https://arxiv.org/abs/2601.04443 · **PDF:** https://arxiv.org/pdf/2601.04443v2
+
+## 23382. Addressing Overthinking in Large Vision-Language Models via Gated Perception-Reasoning Optimization
+
+- **ID:** `2601.04442` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Xingjian Diao, Zheyuan Liu, Chunhui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.04442 · **PDF:** https://arxiv.org/pdf/2601.04442v1
+
+## 23383. Improving and Accelerating Offline RL in Large Discrete Action Spaces with Structured Policy Initialization
+
+- **ID:** `2601.04441` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Matthew Landers, Taylor W. Killian, Thomas Hartvigsen, et al.
+- **abs:** https://arxiv.org/abs/2601.04441 · **PDF:** https://arxiv.org/pdf/2601.04441v2
+
+## 23384. Learning to Simulate Human Dialogue
+
+- **ID:** `2601.04436` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Kanishk Gandhi, Agam Bhatia, Noah D. Goodman
+- **abs:** https://arxiv.org/abs/2601.04436 · **PDF:** https://arxiv.org/pdf/2601.04436v1
+
+## 23385. Accommodation and Epistemic Vigilance: A Pragmatic Account of Why LLMs Fail to Challenge Harmful Beliefs
+
+- **ID:** `2601.04435` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Myra Cheng, Robert D. Hawkins, Dan Jurafsky
+- **abs:** https://arxiv.org/abs/2601.04435 · **PDF:** https://arxiv.org/pdf/2601.04435v1
+
+## 23386. CRUNet-MR-Univ: A Foundation Model for Diverse Cardiac MRI Reconstruction
+
+- **ID:** `2601.04428` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Donghang Lyu, Marius Staring, Hildo Lamb, et al.
+- **abs:** https://arxiv.org/abs/2601.04428 · **PDF:** https://arxiv.org/pdf/2601.04428v1
+
+## 23387. XGrammar-2: Efficient Dynamic Structured Generation Engine for Agentic LLMs
+
+- **ID:** `2601.04426` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Linzhang Li, Yixin Dong, Guanjie Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.04426 · **PDF:** https://arxiv.org/pdf/2601.04426v2
+
+## 23388. Gavel: Agent Meets Checklist for Evaluating LLMs on Long-Context Legal Summarization
+
+- **ID:** `2601.04424` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yao Dou, Wei Xu
+- **abs:** https://arxiv.org/abs/2601.04424 · **PDF:** https://arxiv.org/pdf/2601.04424v1
+
+## 23389. Learning Multinomial Logits in $O(n \log n)$ time
+
+- **ID:** `2601.04423` · **primary:** `cs.DS` · **Submitted:** 2026-01-07
+- **Authors:** Flavio Chierichetti, Mirko Giacchini, Ravi Kumar, et al.
+- **abs:** https://arxiv.org/abs/2601.04423 · **PDF:** https://arxiv.org/pdf/2601.04423v1
+
+## 23390. Transitive Expert Error and Routing Problems in Complex AI Systems
+
+- **ID:** `2601.04416` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Forest Mars
+- **abs:** https://arxiv.org/abs/2601.04416 · **PDF:** https://arxiv.org/pdf/2601.04416v1
+
+## 23391. Attention Mechanism and Heuristic Approach: Context-Aware File Ranking Using Multi-Head Self-Attention
+
+- **ID:** `2601.06185` · **primary:** `cs.SE` · **Submitted:** 2026-01-07
+- **Authors:** Pradeep Kumar Sharma, Shantanu Godbole, Sarada Prasad Jena, et al.
+- **abs:** https://arxiv.org/abs/2601.06185 · **PDF:** https://arxiv.org/pdf/2601.06185v1
+
+## 23392. Distribution-Guided and Constrained Quantum Machine Unlearning
+
+- **ID:** `2601.04413` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Nausherwan Malik, Zubair Khalid, Muhammad Faryad
+- **abs:** https://arxiv.org/abs/2601.04413 · **PDF:** https://arxiv.org/pdf/2601.04413v2
+
+## 23393. Rate or Fate? RLV$^\varepsilon$R: Reinforcement Learning with Verifiable Noisy Rewards
+
+- **ID:** `2601.04411` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Ali Rad, Khashayar Filom, Darioush Keivan, et al.
+- **abs:** https://arxiv.org/abs/2601.04411 · **PDF:** https://arxiv.org/pdf/2601.04411v1
+
+## 23394. From Preoperative CT to Postmastoidectomy Mesh Construction: Mastoidectomy Shape Prediction for Cochlear Implant Surgery
+
+- **ID:** `2601.04405` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Yike Zhang, Eduardo Davalos, Dingjie Su, et al.
+- **abs:** https://arxiv.org/abs/2601.04405 · **PDF:** https://arxiv.org/pdf/2601.04405v2
+
+## 23395. 3D-Agent:Tri-Modal Multi-Agent Collaboration for Scalable 3D Object Annotation
+
+- **ID:** `2601.04404` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Jusheng Zhang, Yijia Fan, Zimo Wen, et al.
+- **abs:** https://arxiv.org/abs/2601.04404 · **PDF:** https://arxiv.org/pdf/2601.04404v1
+
+## 23396. Balancing Usability and Compliance in AI Smart Devices: A Privacy-by-Design Audit of Google Home, Alexa, and Siri
+
+- **ID:** `2601.04403` · **primary:** `cs.CY` · **Submitted:** 2026-01-07
+- **Authors:** Trevor De Clark, Yulia Bobkova, Ajay Kumar Shrestha
+- **abs:** https://arxiv.org/abs/2601.04403 · **PDF:** https://arxiv.org/pdf/2601.04403v2
+
+## 23397. Transformer-based Multi-agent Reinforcement Learning for Separation Assurance in Structured and Unstructured Airspaces
+
+- **ID:** `2601.04401` · **primary:** `cs.RO` · **Submitted:** 2026-01-07
+- **Authors:** Arsyi Aziz, Peng Wei
+- **abs:** https://arxiv.org/abs/2601.04401 · **PDF:** https://arxiv.org/pdf/2601.04401v1
+
+## 23398. Toward Youth-Centered Privacy-by-Design in Smart Devices: A Systematic Review
+
+- **ID:** `2601.11598` · **primary:** `cs.CY` · **Submitted:** 2026-01-07
+- **Authors:** Molly Campbell, Mohamad Sheikho Al Jasem, Ajay Kumar Shrestha
+- **abs:** https://arxiv.org/abs/2601.11598 · **PDF:** https://arxiv.org/pdf/2601.11598v2
+
+## 23399. Convenience vs. Control: A Qualitative Study of Youth Privacy with Smart Voice Assistants
+
+- **ID:** `2601.04399` · **primary:** `cs.CY` · **Submitted:** 2026-01-07
+- **Authors:** Molly Campbell, Trevor De Clark, Mohamad Sheikho Al Jasem, et al.
+- **abs:** https://arxiv.org/abs/2601.04399 · **PDF:** https://arxiv.org/pdf/2601.04399v2
+
+## 23400. Interpreting Transformers Through Attention Head Intervention
+
+- **ID:** `2601.04398` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Mason Kadem, Rong Zheng
+- **abs:** https://arxiv.org/abs/2601.04398 · **PDF:** https://arxiv.org/pdf/2601.04398v4
+
+## 23401. ARREST: Adversarial Resilient Regulation Enhancing Safety and Truth in Large Language Models
+
+- **ID:** `2601.04394` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Sharanya Dasgupta, Arkaprabha Basu, Sujoy Nath, et al.
+- **abs:** https://arxiv.org/abs/2601.04394 · **PDF:** https://arxiv.org/pdf/2601.04394v1
+
+## 23402. Assessing the quality and coherence of word embeddings after SCM-based intersectional bias mitigation
+
+- **ID:** `2601.04393` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Eren Kocadag, Seyed Sahand Mohammadi Ziabari, Ali Mohammed Mansoor Alsahag
+- **abs:** https://arxiv.org/abs/2601.04393 · **PDF:** https://arxiv.org/pdf/2601.04393v1
+
+## 23403. Enhanced-FQL($λ$), an Efficient and Interpretable RL with novel Fuzzy Eligibility Traces and Segmented Experience Replay
+
+- **ID:** `2601.04392` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Mohsen Jalaeian-Farimani
+- **abs:** https://arxiv.org/abs/2601.04392 · **PDF:** https://arxiv.org/pdf/2601.04392v1
+
+## 23404. SciFig: Towards Automating Scientific Figure Generation
+
+- **ID:** `2601.04390` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Siyuan Huang, Yutong Gao, Juyang Bai, et al.
+- **abs:** https://arxiv.org/abs/2601.04390 · **PDF:** https://arxiv.org/pdf/2601.04390v1
+
+## 23405. MiJaBench: Revealing Minority Biases in Large Language Models via Hate Speech Jailbreaking
+
+- **ID:** `2601.04389` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Iago Alves Brito, Walcy Santos Rezende Rios, Julia Soares Dollis, et al.
+- **abs:** https://arxiv.org/abs/2601.04389 · **PDF:** https://arxiv.org/pdf/2601.04389v1
+
+## 23406. Towards Latent Diffusion Suitable For Text
+
+- **ID:** `2601.16220` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Nesta Midavaine, Christian A. Naesseth, Grigory Bartosh
+- **abs:** https://arxiv.org/abs/2601.16220 · **PDF:** https://arxiv.org/pdf/2601.16220v1
+
+## 23407. LLM-Guided Lifecycle-Aware Clustering of Multi-Turn Customer Support Conversations
+
+- **ID:** `2601.04388` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Priyaranjan Pattnayak, Sanchari Chowdhuri, Amit Agarwal, et al.
+- **abs:** https://arxiv.org/abs/2601.04388 · **PDF:** https://arxiv.org/pdf/2601.04388v1
+
+## 23408. The Language of Bargaining: Linguistic Effects in LLM Negotiations
+
+- **ID:** `2601.04387` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Stuti Sinha, Himanshu Kumar, Aryan Raju Mandapati, et al.
+- **abs:** https://arxiv.org/abs/2601.04387 · **PDF:** https://arxiv.org/pdf/2601.04387v1
+
+## 23409. Few-Shot LoRA Adaptation of a Flow-Matching Foundation Model for Cross-Spectral Object Detection
+
+- **ID:** `2601.04381` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Maxim Clouser, Kia Khezeli, John Kalantari
+- **abs:** https://arxiv.org/abs/2601.04381 · **PDF:** https://arxiv.org/pdf/2601.04381v1
+
+## 23410. Aligned explanations in neural networks
+
+- **ID:** `2601.04378` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Corentin Lobet, Francesca Chiaromonte
+- **abs:** https://arxiv.org/abs/2601.04378 · **PDF:** https://arxiv.org/pdf/2601.04378v2
+
+## 23411. Disco-RAG: Discourse-Aware Retrieval-Augmented Generation
+
+- **ID:** `2601.04377` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Dongqi Liu, Hang Ding, Qiming Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.04377 · **PDF:** https://arxiv.org/pdf/2601.04377v3
+
+## 23412. Dialect Matters: Cross-Lingual ASR Transfer for Low-Resource Indic Language Varieties
+
+- **ID:** `2601.04373` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Akriti Dhasmana, Aarohi Srivastava, David Chiang
+- **abs:** https://arxiv.org/abs/2601.04373 · **PDF:** https://arxiv.org/pdf/2601.04373v2
+
+## 23413. Generalization to Political Beliefs from Fine-Tuning on Sports Team Preferences
+
+- **ID:** `2601.04369` · **primary:** `physics.soc-ph` · **Submitted:** 2026-01-07
+- **Authors:** Owen Terry
+- **abs:** https://arxiv.org/abs/2601.04369 · **PDF:** https://arxiv.org/pdf/2601.04369v3
+
+## 23414. Graph Integrated Transformers for Community Detection in Social Networks
+
+- **ID:** `2601.04367` · **primary:** `cs.SI` · **Submitted:** 2026-01-07
+- **Authors:** Heba Zahran, M. Omair Shafiq
+- **abs:** https://arxiv.org/abs/2601.04367 · **PDF:** https://arxiv.org/pdf/2601.04367v1
+
+## 23415. Machine Learning Model for Sparse PCM Completion
+
+- **ID:** `2601.04366` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Selcuk Koyuncu, Ronak Nouri, Stephen Providence
+- **abs:** https://arxiv.org/abs/2601.04366 · **PDF:** https://arxiv.org/pdf/2601.04366v1
+
+## 23416. Survival Dynamics of Neural and Programmatic Policies in Evolutionary Reinforcement Learning
+
+- **ID:** `2601.04365` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Anton Roupassov-Ruiz, Yiyang Zuo
+- **abs:** https://arxiv.org/abs/2601.04365 · **PDF:** https://arxiv.org/pdf/2601.04365v1
+
+## 23417. Phasor Agents: Oscillatory Graphs with Three-Factor Plasticity and Sleep-Staged Learning
+
+- **ID:** `2601.04362` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Rodja Trappe
+- **abs:** https://arxiv.org/abs/2601.04362 · **PDF:** https://arxiv.org/pdf/2601.04362v1
+
+## 23418. Causally-Aware Information Bottleneck for Domain Adaptation
+
+- **ID:** `2601.04361` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Mohammad Ali Javidian
+- **abs:** https://arxiv.org/abs/2601.04361 · **PDF:** https://arxiv.org/pdf/2601.04361v1
+
+## 23419. UNIC: Learning Unified Multimodal Extrinsic Contact Estimation
+
+- **ID:** `2601.04356` · **primary:** `cs.RO` · **Submitted:** 2026-01-07
+- **Authors:** Zhengtong Xu, Yuki Shirai
+- **abs:** https://arxiv.org/abs/2601.04356 · **PDF:** https://arxiv.org/pdf/2601.04356v2
+
+## 23420. Comparative Analysis of Custom CNN Architectures versus Pre-trained Models and Transfer Learning: A Study on Five Bangladesh Datasets
+
+- **ID:** `2601.04352` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Ibrahim Tanvir, Alif Ruslan, Sartaj Solaiman
+- **abs:** https://arxiv.org/abs/2601.04352 · **PDF:** https://arxiv.org/pdf/2601.04352v1
+
+## 23421. RIGOURATE: Quantifying Scientific Exaggeration with Evidence-Aligned Claim Evaluation
+
+- **ID:** `2601.04350` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Joseph James, Chenghao Xiao, Yucheng Li, et al.
+- **abs:** https://arxiv.org/abs/2601.04350 · **PDF:** https://arxiv.org/pdf/2601.04350v2
+
+## 23422. Summary of The Inaugural Music Source Restoration Challenge
+
+- **ID:** `2601.04343` · **primary:** `cs.SD` · **Submitted:** 2026-01-07
+- **Authors:** Yongyi Zang, Jiarui Hai, Wanying Ge, et al.
+- **abs:** https://arxiv.org/abs/2601.04343 · **PDF:** https://arxiv.org/pdf/2601.04343v1
+
+## 23423. Unified Text-Image Generation with Weakness-Targeted Post-Training
+
+- **ID:** `2601.04339` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Jiahui Chen, Philippe Hansen-Estruch, Xiaochuang Han, et al.
+- **abs:** https://arxiv.org/abs/2601.04339 · **PDF:** https://arxiv.org/pdf/2601.04339v2
+
+## 23424. Pilot Study on Student Public Opinion Regarding GAI
+
+- **ID:** `2601.04336` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** William Franz Lamberti, Sunbin Kim, Samantha Rose Lawrence
+- **abs:** https://arxiv.org/abs/2601.04336 · **PDF:** https://arxiv.org/pdf/2601.04336v1
+
+## 23425. ParaCodex: A Profiling-Guided Autonomous Coding Agent for Reliable Parallel Code Generation and Translation
+
+- **ID:** `2601.04327` · **primary:** `cs.DC` · **Submitted:** 2026-01-07
+- **Authors:** Erel Kaplan, Tomer Bitan, Lian Ghrayeb, et al.
+- **abs:** https://arxiv.org/abs/2601.04327 · **PDF:** https://arxiv.org/pdf/2601.04327v1
+
+## 23426. A universal vision transformer for fast calorimeter simulations
+
+- **ID:** `2601.05289` · **primary:** `hep-ph` · **Submitted:** 2026-01-07
+- **Authors:** Luigi Favaro, Andrea Giammanco, Claudius Krause
+- **abs:** https://arxiv.org/abs/2601.05289 · **PDF:** https://arxiv.org/pdf/2601.05289v1
+
+## 23427. A discrete Benamou-Brenier formulation of Optimal Transport on graphs
+
+- **ID:** `2601.04193` · **primary:** `cs.IT` · **Submitted:** 2026-01-07
+- **Authors:** Kieran Morris, Oliver Johnson
+- **abs:** https://arxiv.org/abs/2601.04193 · **PDF:** https://arxiv.org/pdf/2601.04193v1
+
+## 23428. Embedding Autonomous Agents in Resource-Constrained Robotic Platforms
+
+- **ID:** `2601.04191` · **primary:** `cs.RO` · **Submitted:** 2026-01-07
+- **Authors:** Negar Halakou, Juan F. Gutierrez, Ye Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.04191 · **PDF:** https://arxiv.org/pdf/2601.04191v1
+
+## 23429. Data-Driven Reduced-Complexity Modeling of Fluid Flows: A Community Challenge
+
+- **ID:** `2601.06183` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Oliver T. Schmidt, Aaron Towne, Adrian Lozano-Duran, et al.
+- **abs:** https://arxiv.org/abs/2601.06183 · **PDF:** https://arxiv.org/pdf/2601.06183v1
+
+## 23430. Lightweight Test-Time Adaptation for EMG-Based Gesture Recognition
+
+- **ID:** `2601.04181` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Nia Touko, Matthew O A Ellis, Cristiano Capone, et al.
+- **abs:** https://arxiv.org/abs/2601.04181 · **PDF:** https://arxiv.org/pdf/2601.04181v1
+
+## 23431. Quantifying the Effect of Test Set Contamination on Generative Evaluations
+
+- **ID:** `2601.04301` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Rylan Schaeffer, Joshua Kazdan, Baber Abbasi, et al.
+- **abs:** https://arxiv.org/abs/2601.04301 · **PDF:** https://arxiv.org/pdf/2601.04301v2
+
+## 23432. Robust Physics Discovery from Highly Corrupted Data: A PINN Framework Applied to the Nonlinear Schrödinger Equation
+
+- **ID:** `2601.04176` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Pietro de Oliveira Esteves
+- **abs:** https://arxiv.org/abs/2601.04176 · **PDF:** https://arxiv.org/pdf/2601.04176v1
+
+## 23433. Agentic Rubrics as Contextual Verifiers for SWE Agents
+
+- **ID:** `2601.04171` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Mohit Raghavendra, Anisha Gunjal, Bing Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.04171 · **PDF:** https://arxiv.org/pdf/2601.04171v1
+
+## 23434. Agent Drift: Quantifying Behavioral Degradation in Multi-Agent LLM Systems Over Extended Interactions
+
+- **ID:** `2601.04170` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Abhishek Rath
+- **abs:** https://arxiv.org/abs/2601.04170 · **PDF:** https://arxiv.org/pdf/2601.04170v1
+
+## 23435. Clinical Data Goes MEDS? Let's OWL make sense of it
+
+- **ID:** `2601.04164` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Alberto Marfoglia, Jong Ho Jhee, Adrien Coulet
+- **abs:** https://arxiv.org/abs/2601.04164 · **PDF:** https://arxiv.org/pdf/2601.04164v2
+
+## 23436. Scanner-Induced Domain Shifts Undermine the Robustness of Pathology Foundation Models
+
+- **ID:** `2601.04163` · **primary:** `eess.IV` · **Submitted:** 2026-01-07
+- **Authors:** Erik Thiringer, Fredrik K. Gustafsson, Kajsa Ledesma Eriksson, et al.
+- **abs:** https://arxiv.org/abs/2601.04163 · **PDF:** https://arxiv.org/pdf/2601.04163v1
+
+## 23437. Attention Isn't All You Need for Emotion Recognition:Domain Features Outperform Transformers on the EAV Dataset
+
+- **ID:** `2601.22161` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Anmol Guragain
+- **abs:** https://arxiv.org/abs/2601.22161 · **PDF:** https://arxiv.org/pdf/2601.22161v2
+
+## 23438. All That Glisters Is Not Gold: A Benchmark for Reference-Free Counterfactual Financial Misinformation Detection
+
+- **ID:** `2601.04160` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yuechen Jiang, Zhiwei Liu, Yupeng Cao, et al.
+- **abs:** https://arxiv.org/abs/2601.04160 · **PDF:** https://arxiv.org/pdf/2601.04160v3
+
+## 23439. FLEx: Language Modeling with Few-shot Language Explanations
+
+- **ID:** `2601.04157` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Adar Avsian, Christopher Richardson, Anirudh Sundar, et al.
+- **abs:** https://arxiv.org/abs/2601.04157 · **PDF:** https://arxiv.org/pdf/2601.04157v2
+
+## 23440. Apollo: Unified Multi-Task Audio-Video Joint Generation
+
+- **ID:** `2601.04151` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Jun Wang, Chunyu Qiang, Yuxin Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.04151 · **PDF:** https://arxiv.org/pdf/2601.04151v2
+
+## 23441. A Theoretical and Empirical Taxonomy of Imbalance in Binary Classification
+
+- **ID:** `2601.04149` · **primary:** `stat.ML` · **Submitted:** 2026-01-07
+- **Authors:** Rose Yvette Bandolo Essomba, Ernest Fokoué
+- **abs:** https://arxiv.org/abs/2601.04149 · **PDF:** https://arxiv.org/pdf/2601.04149v1
+
+## 23442. Transformer-Based Multi-Modal Temporal Embeddings for Explainable Metabolic Phenotyping in Type 1 Diabetes
+
+- **ID:** `2601.04299` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Pir Bakhsh Khokhar, Carmine Gravino, Fabio Palomba, et al.
+- **abs:** https://arxiv.org/abs/2601.04299 · **PDF:** https://arxiv.org/pdf/2601.04299v1
+
+## 23443. Wow, wo, val! A Comprehensive Embodied World Model Evaluation Turing Test
+
+- **ID:** `2601.04137` · **primary:** `cs.RO` · **Submitted:** 2026-01-07
+- **Authors:** Chun-Kai Fan, Xiaowei Chi, Xiaozhu Ju, et al.
+- **abs:** https://arxiv.org/abs/2601.04137 · **PDF:** https://arxiv.org/pdf/2601.04137v1
+
+## 23444. LLMberjack: Guided Trimming of Debate Trees for Multi-Party Conversation Creation
+
+- **ID:** `2601.04135` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Leonardo Bottona, Nicolò Penzo, Bruno Lepri, et al.
+- **abs:** https://arxiv.org/abs/2601.04135 · **PDF:** https://arxiv.org/pdf/2601.04135v1
+
+## 23445. ContextFocus: Activation Steering for Contextual Faithfulness in Large Language Models
+
+- **ID:** `2601.04131` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Nikhil Anand, Shwetha Somasundaram, Anirudh Phukan, et al.
+- **abs:** https://arxiv.org/abs/2601.04131 · **PDF:** https://arxiv.org/pdf/2601.04131v2
+
+## 23446. Pixel-Wise Multimodal Contrastive Learning for Remote Sensing Images
+
+- **ID:** `2601.04127` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Leandro Stival, Ricardo da Silva Torres, Helio Pedrini
+- **abs:** https://arxiv.org/abs/2601.04127 · **PDF:** https://arxiv.org/pdf/2601.04127v1
+
+## 23447. InfiniteWeb: Scalable Web Environment Synthesis for GUI Agent Training
+
+- **ID:** `2601.04126` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Ziyun Zhang, Zezhou Wang, Xiaoyi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.04126 · **PDF:** https://arxiv.org/pdf/2601.04126v2
+
+## 23448. ArtCognition: A Multimodal AI Framework for Affective State Sensing from Visual and Kinematic Drawing Cues
+
+- **ID:** `2601.04297` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Behrad Binaei-Haghighi, Nafiseh Sadat Sajadi, Mehrad Liviyan, et al.
+- **abs:** https://arxiv.org/abs/2601.04297 · **PDF:** https://arxiv.org/pdf/2601.04297v1
+
+## 23449. MORPHFED: Federated Learning for Cross-institutional Blood Morphology Analysis
+
+- **ID:** `2601.04121` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Gabriel Ansah, Eden Ruffell, Delmiro Fernandez-Reyes, et al.
+- **abs:** https://arxiv.org/abs/2601.04121 · **PDF:** https://arxiv.org/pdf/2601.04121v1
+
+## 23450. A Single-Loop Bilevel Deep Learning Method for Optimal Control of Obstacle Problems
+
+- **ID:** `2601.04120` · **primary:** `math.OC` · **Submitted:** 2026-01-07
+- **Authors:** Yongcun Song, Shangzhi Zeng, Jin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.04120 · **PDF:** https://arxiv.org/pdf/2601.04120v1
+
+## 23451. Causal Data Augmentation for Robust Fine-Tuning of Tabular Foundation Models
+
+- **ID:** `2601.04110` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Magnus Bühler, Lennart Purucker, Frank Hutter
+- **abs:** https://arxiv.org/abs/2601.04110 · **PDF:** https://arxiv.org/pdf/2601.04110v2
+
+## 23452. Equivariant Neural Networks for Force-Field Models of Lattice Systems
+
+- **ID:** `2601.04104` · **primary:** `cond-mat.str-el` · **Submitted:** 2026-01-07
+- **Authors:** Yunhao Fan, Gia-Wei Chern
+- **abs:** https://arxiv.org/abs/2601.04104 · **PDF:** https://arxiv.org/pdf/2601.04104v1
+
+## 23453. SSC-UNet: UNet with Self-Supervised Contrastive Learning for Phonocardiography Noise Reduction
+
+- **ID:** `2601.10735` · **primary:** `eess.SP` · **Submitted:** 2026-01-07
+- **Authors:** Lizy Abraham, Siobhan Coughlan, Kritika Rajain, et al.
+- **abs:** https://arxiv.org/abs/2601.10735 · **PDF:** https://arxiv.org/pdf/2601.10735v1
+
+## 23454. Quantifying the Impact of Modules and Their Interactions in the PSO-X Framework
+
+- **ID:** `2601.04100` · **primary:** `cs.NE` · **Submitted:** 2026-01-07
+- **Authors:** Christian L. Camacho-Villalón, Ana Nikolikj, Katharina Dost, et al.
+- **abs:** https://arxiv.org/abs/2601.04100 · **PDF:** https://arxiv.org/pdf/2601.04100v1
+
+## 23455. Neuro-Symbolic Compliance: Integrating LLMs and SMT Solvers for Automated Financial Legal Analysis
+
+- **ID:** `2601.06181` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Yung-Shen Hsia, Fang Yu, Jie-Hong Roland Jiang
+- **abs:** https://arxiv.org/abs/2601.06181 · **PDF:** https://arxiv.org/pdf/2601.06181v1
+
+## 23456. Layer-wise Positional Bias in Short-Context Language Modeling
+
+- **ID:** `2601.04098` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Maryam Rahimi, Mahdi Nouri, Yadollah Yaghoobzadeh
+- **abs:** https://arxiv.org/abs/2601.04098 · **PDF:** https://arxiv.org/pdf/2601.04098v1
+
+## 23457. SearchAttack: Red-Teaming LLMs against Knowledge-to-Action Threats under Online Web Search
+
+- **ID:** `2601.04093` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yu Yan, Sheng Sun, Mingfeng Li, et al.
+- **abs:** https://arxiv.org/abs/2601.04093 · **PDF:** https://arxiv.org/pdf/2601.04093v2
+
+## 23458. MixDPO: Modeling Preference Strength for Pluralistic Alignment
+
+- **ID:** `2601.06180` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Saki Imai, Pedram Heydari, Anthony Sicilia, et al.
+- **abs:** https://arxiv.org/abs/2601.06180 · **PDF:** https://arxiv.org/pdf/2601.06180v1
+
+## 23459. KDCM: Reducing Hallucination in LLMs through Explicit Reasoning Structures
+
+- **ID:** `2601.04086` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Jinbo Hao, Kai Yang, Qingzhen Su, et al.
+- **abs:** https://arxiv.org/abs/2601.04086 · **PDF:** https://arxiv.org/pdf/2601.04086v1
+
+## 23460. CSSG: Measuring Code Similarity with Semantic Graphs
+
+- **ID:** `2601.04085` · **primary:** `cs.PL` · **Submitted:** 2026-01-07
+- **Authors:** Yiyang Lu, Jingwen Xu, Changze Lv, et al.
+- **abs:** https://arxiv.org/abs/2601.04085 · **PDF:** https://arxiv.org/pdf/2601.04085v2
+
+## 23461. Cells on Autopilot: Adaptive Cell (Re)Selection via Reinforcement Learning
+
+- **ID:** `2601.04083` · **primary:** `cs.NI` · **Submitted:** 2026-01-07
+- **Authors:** Marvin Illian, Ramin Khalili, Antonio A. de A. Rocha, et al.
+- **abs:** https://arxiv.org/abs/2601.04083 · **PDF:** https://arxiv.org/pdf/2601.04083v3
+
+## 23462. Analyzing Reasoning Consistency in Large Multimodal Models under Cross-Modal Conflicts
+
+- **ID:** `2601.04073` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Zhihao Zhu, Jiafeng Liang, Shixin Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.04073 · **PDF:** https://arxiv.org/pdf/2601.04073v1
+
+## 23463. Mind the Generative Details: Direct Localized Detail Preference Optimization for Video Diffusion Models
+
+- **ID:** `2601.04068` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Zitong Huang, Kaidong Zhang, Yukang Ding, et al.
+- **abs:** https://arxiv.org/abs/2601.04068 · **PDF:** https://arxiv.org/pdf/2601.04068v2
+
+## 23464. Unsupervised Modular Adaptive Region Growing and RegionMix Classification for Wind Turbine Segmentation
+
+- **ID:** `2601.04065` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Raül Pérez-Gonzalo, Riccardo Magro, Andreas Espersen, et al.
+- **abs:** https://arxiv.org/abs/2601.04065 · **PDF:** https://arxiv.org/pdf/2601.04065v2
+
+## 23465. ComfySearch: Autonomous Exploration and Reasoning for ComfyUI Workflows
+
+- **ID:** `2601.04060` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Jinwei Su, Qizhen Lan, Zeyu Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.04060 · **PDF:** https://arxiv.org/pdf/2601.04060v1
+
+## 23466. Minimum distance classification for nonlinear dynamical systems
+
+- **ID:** `2601.04058` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Dominique Martinez
+- **abs:** https://arxiv.org/abs/2601.04058 · **PDF:** https://arxiv.org/pdf/2601.04058v2
+
+## 23467. Using Legacy Polysomnography Data to Train a Radar System to Quantify Sleep in Older Adults and People living with Dementia
+
+- **ID:** `2601.04057` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** M. Yin, K. G. Ravindran, C. Hadjipanayi, et al.
+- **abs:** https://arxiv.org/abs/2601.04057 · **PDF:** https://arxiv.org/pdf/2601.04057v1
+
+## 23468. Bridging the Discrete-Continuous Gap: Unified Multimodal Generation via Coupled Manifold Discrete Absorbing Diffusion
+
+- **ID:** `2601.04056` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yuanfeng Xu, Yuhao Chen, Liang Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.04056 · **PDF:** https://arxiv.org/pdf/2601.04056v1
+
+## 23469. Modular Prompt Optimization: Optimizing Structured Prompts with Section-Local Textual Gradients
+
+- **ID:** `2601.04055` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Prith Sharma, Austin Z. Henley
+- **abs:** https://arxiv.org/abs/2601.04055 · **PDF:** https://arxiv.org/pdf/2601.04055v1
+
+## 23470. LinkD: AutoRegressive Diffusion Model for Mechanical Linkage Synthesis
+
+- **ID:** `2601.04054` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Yayati Jadhav, Amir Barati Farimani
+- **abs:** https://arxiv.org/abs/2601.04054 · **PDF:** https://arxiv.org/pdf/2601.04054v1
+
+## 23471. Stable Language Guidance for Vision-Language-Action Models
+
+- **ID:** `2601.04052` · **primary:** `cs.RO` · **Submitted:** 2026-01-07
+- **Authors:** Zhihao Zhan, Yuhao Chen, Jiaying Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.04052 · **PDF:** https://arxiv.org/pdf/2601.04052v1
+
+## 23472. Symbolic Regression for Shared Expressions: Introducing Partial Parameter Sharing
+
+- **ID:** `2601.04051` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Viktor Martinek, Roland Herzog
+- **abs:** https://arxiv.org/abs/2601.04051 · **PDF:** https://arxiv.org/pdf/2601.04051v2
+
+## 23473. When Helpers Become Hazards: A Benchmark for Analyzing Multimodal LLM-Powered Safety in Daily Life
+
+- **ID:** `2601.04043` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Xinyue Lou, Jinan Xu, Jingyi Yin, et al.
+- **abs:** https://arxiv.org/abs/2601.04043 · **PDF:** https://arxiv.org/pdf/2601.04043v1
+
+## 23474. Analyzing and Improving Cross-lingual Knowledge Transfer for Machine Translation
+
+- **ID:** `2601.04036` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** David Stap
+- **abs:** https://arxiv.org/abs/2601.04036 · **PDF:** https://arxiv.org/pdf/2601.04036v1
+
+## 23475. MobileDreamer: Generative Sketch World Model for GUI Agent
+
+- **ID:** `2601.04035` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Yilin Cao, Yufeng Zhong, Zhixiong Zeng, et al.
+- **abs:** https://arxiv.org/abs/2601.04035 · **PDF:** https://arxiv.org/pdf/2601.04035v1
+
+## 23476. HoneyTrap: Deceiving Large Language Model Attackers to Honeypot Traps with Resilient Multi-Agent Defense
+
+- **ID:** `2601.04034` · **primary:** `cs.CR` · **Submitted:** 2026-01-07
+- **Authors:** Siyuan Li, Xi Lin, Jun Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.04034 · **PDF:** https://arxiv.org/pdf/2601.04034v1
+
+## 23477. SpeakerSleuth: Evaluating Large Audio-Language Models as Judges for Multi-turn Speaker Consistency
+
+- **ID:** `2601.04029` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Jonggeun Lee, Junseong Pyo, Gyuhyeon Seo, et al.
+- **abs:** https://arxiv.org/abs/2601.04029 · **PDF:** https://arxiv.org/pdf/2601.04029v1
+
+## 23478. Simulated Students in Tutoring Dialogues: Substance or Illusion?
+
+- **ID:** `2601.04025` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Alexander Scarlatos, Jaewook Lee, Simon Woodhead, et al.
+- **abs:** https://arxiv.org/abs/2601.04025 · **PDF:** https://arxiv.org/pdf/2601.04025v1
+
+## 23479. Modeling Behavioral Patterns in News Recommendations Using Fuzzy Neural Networks
+
+- **ID:** `2601.04019` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Kevin Innerebner, Stephan Bartl, Markus Reiter-Haas, et al.
+- **abs:** https://arxiv.org/abs/2601.04019 · **PDF:** https://arxiv.org/pdf/2601.04019v1
+
+## 23480. Correct and Weight: A Simple Yet Effective Loss for Implicit Feedback Recommendation
+
+- **ID:** `2601.04291` · **primary:** `cs.IR` · **Submitted:** 2026-01-07
+- **Authors:** Minglei Yin, Chuanbo Hu, Bin Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.04291 · **PDF:** https://arxiv.org/pdf/2601.04291v1
+
+## 23481. Performance of models for monitoring sustainable development goals from remote sensing: A three-level meta-regression
+
+- **ID:** `2601.06178` · **primary:** `cs.CY` · **Submitted:** 2026-01-07
+- **Authors:** Jonas Klingwort, Nina M. Leach, Joep Burger
+- **abs:** https://arxiv.org/abs/2601.06178 · **PDF:** https://arxiv.org/pdf/2601.06178v1
+
+## 23482. VotIE: Information Extraction from Meeting Minutes
+
+- **ID:** `2601.03997` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** José Pedro Evans, Luís Filipe Cunha, Purificação Silvano, et al.
+- **abs:** https://arxiv.org/abs/2601.03997 · **PDF:** https://arxiv.org/pdf/2601.03997v3
+
+## 23483. A Scheduling Framework for Efficient MoE Inference on Edge GPU-NDP Systems
+
+- **ID:** `2601.03992` · **primary:** `cs.DC` · **Submitted:** 2026-01-07
+- **Authors:** Qi Wu, Chao Fang, Jiayuan Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.03992 · **PDF:** https://arxiv.org/pdf/2601.03992v1
+
+## 23484. Using Small Language Models to Reverse-Engineer Machine Learning Pipelines Structures
+
+- **ID:** `2601.03988` · **primary:** `cs.SE` · **Submitted:** 2026-01-07
+- **Authors:** Nicolas Lacroix, Mireille Blay-Fornarino, Sébastien Mosser, et al.
+- **abs:** https://arxiv.org/abs/2601.03988 · **PDF:** https://arxiv.org/pdf/2601.03988v1
+
+## 23485. Benchmark^2: Systematic Evaluation of LLM Benchmarks
+
+- **ID:** `2601.03986` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Qi Qian, Chengsong Huang, Jingwen Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.03986 · **PDF:** https://arxiv.org/pdf/2601.03986v1
+
+## 23486. RADAR: Retrieval-Augmented Detector with Adversarial Refinement for Robust Fake News Detection
+
+- **ID:** `2601.03981` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Song-Duo Ma, Yi-Hung Liu, Hsin-Yu Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.03981 · **PDF:** https://arxiv.org/pdf/2601.03981v2
+
+## 23487. SoK: Privacy Risks and Mitigations in Retrieval-Augmented Generation Systems
+
+- **ID:** `2601.03979` · **primary:** `cs.CR` · **Submitted:** 2026-01-07
+- **Authors:** Andreea-Elena Bodea, Stephen Meisenbacher, Alexandra Klymenko, et al.
+- **abs:** https://arxiv.org/abs/2601.03979 · **PDF:** https://arxiv.org/pdf/2601.03979v1
+
+## 23488. Human-in-the-Loop Testing of AI Agents for Air Traffic Control with a Regulated Assessment Framework
+
+- **ID:** `2601.04288` · **primary:** `cs.HC` · **Submitted:** 2026-01-07
+- **Authors:** Ben Carvell, Marc Thomas, Andrew Pace, et al.
+- **abs:** https://arxiv.org/abs/2601.04288 · **PDF:** https://arxiv.org/pdf/2601.04288v1
+
+## 23489. Stage-specific cancer survival prediction enriched by explainable machine learning
+
+- **ID:** `2601.03977` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Parisa Poorhasani, Bogdan Iancu
+- **abs:** https://arxiv.org/abs/2601.03977 · **PDF:** https://arxiv.org/pdf/2601.03977v1
+
+## 23490. Muse: Towards Reproducible Long-Form Song Generation with Fine-Grained Style Control
+
+- **ID:** `2601.03973` · **primary:** `cs.SD` · **Submitted:** 2026-01-07
+- **Authors:** Changhao Jiang, Jiahao Chen, Zhenghao Xiang, et al.
+- **abs:** https://arxiv.org/abs/2601.03973 · **PDF:** https://arxiv.org/pdf/2601.03973v3
+
+## 23491. Anti-Length Shift: Dynamic Outlier Truncation for Training Efficient Reasoning Models
+
+- **ID:** `2601.03969` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Wei Wu, Liyi Chen, Congxi Xiao, et al.
+- **abs:** https://arxiv.org/abs/2601.03969 · **PDF:** https://arxiv.org/pdf/2601.03969v1
+
+## 23492. Online Action-Stacking Improves Reinforcement Learning Performance for Air Traffic Control
+
+- **ID:** `2601.04287` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Ben Carvell, George De Ath, Eseoghene Benjamin, et al.
+- **abs:** https://arxiv.org/abs/2601.04287 · **PDF:** https://arxiv.org/pdf/2601.04287v1
+
+## 23493. Enhancing Robustness of Asynchronous EEG-Based Movement Prediction using Classifier Ensembles
+
+- **ID:** `2601.04286` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Niklas Kueper, Kartik Chari, Elsa Andrea Kirchner
+- **abs:** https://arxiv.org/abs/2601.04286 · **PDF:** https://arxiv.org/pdf/2601.04286v1
+
+## 23494. A Future Capabilities Agent for Tactical Air Traffic Control
+
+- **ID:** `2601.04285` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Paul Kent, George De Ath, Martin Layton, et al.
+- **abs:** https://arxiv.org/abs/2601.04285 · **PDF:** https://arxiv.org/pdf/2601.04285v1
+
+## 23495. Trade-R1: Bridging Verifiable Rewards to Stochastic Environments via Process-Level Reasoning Verification
+
+- **ID:** `2601.03948` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Rui Sun, Yifan Sun, Sheng Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.03948 · **PDF:** https://arxiv.org/pdf/2601.03948v2
+
+## 23496. Provably Finding a Hidden Dense Submatrix among Many Planted Dense Submatrices via Convex Programming
+
+- **ID:** `2601.03946` · **primary:** `math.OC` · **Submitted:** 2026-01-07
+- **Authors:** Valentine Olanubi, Phineas Agar, Brendan Ames
+- **abs:** https://arxiv.org/abs/2601.03946 · **PDF:** https://arxiv.org/pdf/2601.03946v2
+
+## 23497. Large-Scale Aspect-Based Sentiment Analysis with Reasoning-Infused LLMs
+
+- **ID:** `2601.03940` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Paweł Liskowski, Krzysztof Jankowski
+- **abs:** https://arxiv.org/abs/2601.03940 · **PDF:** https://arxiv.org/pdf/2601.03940v1
+
+## 23498. FOREVER: Forgetting Curve-Inspired Memory Replay for Language Model Continual Learning
+
+- **ID:** `2601.03938` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Yujie Feng, Hao Wang, Jian Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03938 · **PDF:** https://arxiv.org/pdf/2601.03938v1
+
+## 23499. Bayes-PD: Exploring a Sequence to Binding Bayesian Neural Network model trained on Phage Display data
+
+- **ID:** `2601.03930` · **primary:** `q-bio.PE` · **Submitted:** 2026-01-07
+- **Authors:** Ilann Amiaud-Plachy, Michael Blank, Oliver Bent, et al.
+- **abs:** https://arxiv.org/abs/2601.03930 · **PDF:** https://arxiv.org/pdf/2601.03930v1
+
+## 23500. FocusUI: Efficient UI Grounding via Position-Preserving Visual Token Selection
+
+- **ID:** `2601.03928` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Mingyu Ouyang, Kevin Qinghong Lin, Mike Zheng Shou, et al.
+- **abs:** https://arxiv.org/abs/2601.03928 · **PDF:** https://arxiv.org/pdf/2601.03928v1
+
+## 23501. A comprehensive review and analysis of different modeling approaches for financial index tracking problem
+
+- **ID:** `2601.03927` · **primary:** `q-fin.PM` · **Submitted:** 2026-01-07
+- **Authors:** Vrinda Dhingra, Amita Sharma, Anubha Goel
+- **abs:** https://arxiv.org/abs/2601.03927 · **PDF:** https://arxiv.org/pdf/2601.03927v1
+
+## 23502. Doc-PP: Document Policy Preservation Benchmark for Large Vision-Language Models
+
+- **ID:** `2601.03926` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Haeun Jang, Hwan Chang, Hwanhee Lee
+- **abs:** https://arxiv.org/abs/2601.03926 · **PDF:** https://arxiv.org/pdf/2601.03926v1
+
+## 23503. A Gap Between Decision Trees and Neural Networks
+
+- **ID:** `2601.03919` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Akash Kumar
+- **abs:** https://arxiv.org/abs/2601.03919 · **PDF:** https://arxiv.org/pdf/2601.03919v2
+
+## 23504. Monaas: Mobile Node as a Service for TSCH-based Industrial IoT Networks
+
+- **ID:** `2601.03917` · **primary:** `cs.NI` · **Submitted:** 2026-01-07
+- **Authors:** Jinting Liu, Jingwei Li, Tengfei Chang
+- **abs:** https://arxiv.org/abs/2601.03917 · **PDF:** https://arxiv.org/pdf/2601.03917v1
+
+## 23505. Mitigating Position-Shift Failures in Text-Based Modular Arithmetic via Position Curriculum and Template Diversity
+
+- **ID:** `2601.04283` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Nikolay Yudin
+- **abs:** https://arxiv.org/abs/2601.04283 · **PDF:** https://arxiv.org/pdf/2601.04283v1
+
+## 23506. When Models Decide and When They Bind: A Two-Stage Computation for Multiple-Choice Question-Answering
+
+- **ID:** `2601.03914` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Hugh Mee Wong, Rick Nouwen, Albert Gatt
+- **abs:** https://arxiv.org/abs/2601.03914 · **PDF:** https://arxiv.org/pdf/2601.03914v1
+
+## 23507. An Algebraic Representation Theorem for Linear GENEOs in Geometric Machine Learning
+
+- **ID:** `2601.03910` · **primary:** `math.RT` · **Submitted:** 2026-01-07
+- **Authors:** Francesco Conti, Patrizio Frosini, Nicola Quercioli
+- **abs:** https://arxiv.org/abs/2601.03910 · **PDF:** https://arxiv.org/pdf/2601.03910v2
+
+## 23508. Decide Then Retrieve: A Training-Free Framework with Uncertainty-Guided Triggering and Dual-Path Retrieval
+
+- **ID:** `2601.03908` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Wang Chen, Guanqiang Qi, Weikang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03908 · **PDF:** https://arxiv.org/pdf/2601.03908v1
+
+## 23509. LEGATO: Good Identity Unlearning Is Continuous
+
+- **ID:** `2601.04282` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Qiang Chen, Chun-Wun Cheng, Xiu Su, et al.
+- **abs:** https://arxiv.org/abs/2601.04282 · **PDF:** https://arxiv.org/pdf/2601.04282v1
+
+## 23510. Current Agents Fail to Leverage World Model as Tool for Foresight
+
+- **ID:** `2601.03905` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Cheng Qian, Emre Can Acikgoz, Bingxuan Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03905 · **PDF:** https://arxiv.org/pdf/2601.03905v2
+
+## 23511. Adaptive-Boundary-Clipping GRPO: Ensuring Bounded Ratios for Stable and Generalizable Training
+
+- **ID:** `2601.03895` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Chi Liu, Xin Chen
+- **abs:** https://arxiv.org/abs/2601.03895 · **PDF:** https://arxiv.org/pdf/2601.03895v1
+
+## 23512. Lightweight and perceptually-guided voice conversion for electro-laryngeal speech
+
+- **ID:** `2601.03892` · **primary:** `cs.SD` · **Submitted:** 2026-01-07
+- **Authors:** Benedikt Mayrhofer, Franz Pernkopf, Philipp Aichinger, et al.
+- **abs:** https://arxiv.org/abs/2601.03892 · **PDF:** https://arxiv.org/pdf/2601.03892v2
+
+## 23513. Spectral Manifold Regularization for Stable and Modular Routing in Deep MoE Architectures
+
+- **ID:** `2601.03889` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Ibrahim Delibasoglu
+- **abs:** https://arxiv.org/abs/2601.03889 · **PDF:** https://arxiv.org/pdf/2601.03889v1
+
+## 23514. IndexTTS 2.5 Technical Report
+
+- **ID:** `2601.03888` · **primary:** `cs.SD` · **Submitted:** 2026-01-07
+- **Authors:** Yunpei Li, Xun Zhou, Jinchao Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.03888 · **PDF:** https://arxiv.org/pdf/2601.03888v3
+
+## 23515. FLNet: Flood-Induced Agriculture Damage Assessment using Super Resolution of Satellite Images
+
+- **ID:** `2601.03884` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Sanidhya Ghosal, Anurag Sharma, Sushil Ghildiyal, et al.
+- **abs:** https://arxiv.org/abs/2601.03884 · **PDF:** https://arxiv.org/pdf/2601.03884v1
+
+## 23516. Lowest Span Confidence: A Zero-Shot Metric for Efficient and Black-Box Hallucination Detection in LLMs
+
+- **ID:** `2601.19918` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yitong Qiao, Licheng Pan, Yu Mi, et al.
+- **abs:** https://arxiv.org/abs/2601.19918 · **PDF:** https://arxiv.org/pdf/2601.19918v1
+
+## 23517. Feature-Aware One-Shot Federated Learning via Hierarchical Token Sequences
+
+- **ID:** `2601.03882` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Shudong Liu, Hanwen Zhang, Xiuling Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.03882 · **PDF:** https://arxiv.org/pdf/2601.03882v1
+
+## 23518. Women Worry, Men Adopt: How Gendered Perceptions Shape the Use of Generative AI
+
+- **ID:** `2601.03880` · **primary:** `econ.GN` · **Submitted:** 2026-01-07
+- **Authors:** Fabian Stephany, Jedrzej Duszynski
+- **abs:** https://arxiv.org/abs/2601.03880 · **PDF:** https://arxiv.org/pdf/2601.03880v1
+
+## 23519. Generation of synthetic delay time series for air transport applications
+
+- **ID:** `2601.04279` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Pau Esteve, Massimiliano Zanin
+- **abs:** https://arxiv.org/abs/2601.04279 · **PDF:** https://arxiv.org/pdf/2601.04279v1
+
+## 23520. From Domains to Instances: Dual-Granularity Data Synthesis for LLM Unlearning
+
+- **ID:** `2601.04278` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Xiaoyu Xu, Minxin Du, Zitong Li, et al.
+- **abs:** https://arxiv.org/abs/2601.04278 · **PDF:** https://arxiv.org/pdf/2601.04278v1
+
+## 23521. Evaluating Small Decoder-Only Language Models for Grammar Correction and Text Simplification
+
+- **ID:** `2601.03874` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Anthony Lamelas
+- **abs:** https://arxiv.org/abs/2601.03874 · **PDF:** https://arxiv.org/pdf/2601.03874v1
+
+## 23522. Unlocking the Pre-Trained Model as a Dual-Alignment Calibrator for Post-Trained LLMs
+
+- **ID:** `2601.04277` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Beier Luo, Cheng Wang, Hongxin Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.04277 · **PDF:** https://arxiv.org/pdf/2601.04277v1
+
+## 23523. PILOT: Planning via Internalized Latent Optimization Trajectories for Large Language Models
+
+- **ID:** `2601.19917` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Haoyu Zheng, Yun Zhu, Yuqian Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.19917 · **PDF:** https://arxiv.org/pdf/2601.19917v1
+
+## 23524. Atlas: Orchestrating Heterogeneous Models and Tools for Multi-Domain Complex Reasoning
+
+- **ID:** `2601.03872` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Jinyang Wu, Guocheng Zhai, Ruihan Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.03872 · **PDF:** https://arxiv.org/pdf/2601.03872v1
+
+## 23525. Bayesian Monocular Depth Refinement via Neural Radiance Fields
+
+- **ID:** `2601.03869` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Arun Muthukkumar
+- **abs:** https://arxiv.org/abs/2601.03869 · **PDF:** https://arxiv.org/pdf/2601.03869v2
+
+## 23526. What Matters For Safety Alignment?
+
+- **ID:** `2601.03868` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Xing Li, Hui-Ling Zhen, Lihao Yin, et al.
+- **abs:** https://arxiv.org/abs/2601.03868 · **PDF:** https://arxiv.org/pdf/2601.03868v2
+
+## 23527. PartisanLens: A Multilingual Dataset of Hyperpartisan and Conspiratorial Immigration Narratives in European Media
+
+- **ID:** `2601.03860` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Michele Joshua Maggini, Paloma Piot, Anxo Pérez, et al.
+- **abs:** https://arxiv.org/abs/2601.03860 · **PDF:** https://arxiv.org/pdf/2601.03860v1
+
+## 23528. What Does Loss Optimization Actually Teach, If Anything? Knowledge Dynamics in Continual Pre-training of LLMs
+
+- **ID:** `2601.03858` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Seyed Mahed Mousavi, Simone Alghisi, Giuseppe Riccardi
+- **abs:** https://arxiv.org/abs/2601.03858 · **PDF:** https://arxiv.org/pdf/2601.03858v1
+
+## 23529. Shadow Unlearning: A Neuro-Semantic Approach to Fidelity-Preserving Faceless Forgetting in LLMs
+
+- **ID:** `2601.04275` · **primary:** `cs.CR` · **Submitted:** 2026-01-07
+- **Authors:** Dinesh Srivasthav P, Ashok Urlana, Rahul Mishra, et al.
+- **abs:** https://arxiv.org/abs/2601.04275 · **PDF:** https://arxiv.org/pdf/2601.04275v1
+
+## 23530. From No-Regret to Strategically Robust Learning in Repeated Auctions
+
+- **ID:** `2601.03853` · **primary:** `cs.GT` · **Submitted:** 2026-01-07
+- **Authors:** Junyao Zhao
+- **abs:** https://arxiv.org/abs/2601.03853 · **PDF:** https://arxiv.org/pdf/2601.03853v2
+
+## 23531. Rethinking Table Pruning in TableQA: From Sequential Revisions to Gold Trajectory-Supervised Parallel Search
+
+- **ID:** `2601.03851` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yu Guo, Shenghao Ye, Shuangwu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.03851 · **PDF:** https://arxiv.org/pdf/2601.03851v1
+
+## 23532. Investigating the Grounding Bottleneck for a Large-Scale Configuration Problem: Existing Tools and Constraint-Aware Guessing
+
+- **ID:** `2601.03850` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Veronika Semmelrock, Gerhard Friedrich
+- **abs:** https://arxiv.org/abs/2601.03850 · **PDF:** https://arxiv.org/pdf/2601.03850v1
+
+## 23533. Implementing the First-Order Logic of Here and There
+
+- **ID:** `2601.03848` · **primary:** `cs.LO` · **Submitted:** 2026-01-07
+- **Authors:** Jens Otten, Torsten Schaub
+- **abs:** https://arxiv.org/abs/2601.03848 · **PDF:** https://arxiv.org/pdf/2601.03848v1
+
+## 23534. xDNN(ASP): Explanation Generation System for Deep Neural Networks powered by Answer Set Programming
+
+- **ID:** `2601.03847` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Ly Ly Trieu, Tran Cao Son
+- **abs:** https://arxiv.org/abs/2601.03847 · **PDF:** https://arxiv.org/pdf/2601.03847v1
+
+## 23535. When Numbers Start Talking: Implicit Numerical Coordination Among LLM-Based Agents
+
+- **ID:** `2601.03846` · **primary:** `cs.MA` · **Submitted:** 2026-01-07
+- **Authors:** Alessio Buscemi, Daniele Proverbio, Alessandro Di Stefano, et al.
+- **abs:** https://arxiv.org/abs/2601.03846 · **PDF:** https://arxiv.org/pdf/2601.03846v1
+
+## 23536. Formally Explaining Decision Tree Models with Answer Set Programming
+
+- **ID:** `2601.03845` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Akihiro Takemura, Masayuki Otani, Katsumi Inoue
+- **abs:** https://arxiv.org/abs/2601.03845 · **PDF:** https://arxiv.org/pdf/2601.03845v1
+
+## 23537. XAI-LAW: A Logic Programming Tool for Modeling, Explaining, and Learning Legal Decisions
+
+- **ID:** `2601.03844` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Agostino Dovier, Talissa Dreossi, Andrea Formisano, et al.
+- **abs:** https://arxiv.org/abs/2601.03844 · **PDF:** https://arxiv.org/pdf/2601.03844v1
+
+## 23538. An ASP-based Solution to the Medical Appointment Scheduling Problem
+
+- **ID:** `2601.04274` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Alina Vozna, Andrea Monaldini, Stefania Costantini, et al.
+- **abs:** https://arxiv.org/abs/2601.04274 · **PDF:** https://arxiv.org/pdf/2601.04274v1
+
+## 23539. Hybrid MKNF for Aeronautics Applications: Usage and Heuristics
+
+- **ID:** `2601.04273` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Arun Raveendran Nair Sheela, Florence De Grancey, Christophe Rey, et al.
+- **abs:** https://arxiv.org/abs/2601.04273 · **PDF:** https://arxiv.org/pdf/2601.04273v1
+
+## 23540. On the Trap Space Semantics of Normal Logic Programs
+
+- **ID:** `2601.03842` · **primary:** `cs.LO` · **Submitted:** 2026-01-07
+- **Authors:** Van-Giang Trinh, Sylvain Soliman, François Fages, et al.
+- **abs:** https://arxiv.org/abs/2601.03842 · **PDF:** https://arxiv.org/pdf/2601.03842v1
+
+## 23541. Defeasible Conditionals using Answer Set Programming
+
+- **ID:** `2601.03840` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Racquel Dennison, Jesse Heyninck, Thomas Meyer
+- **abs:** https://arxiv.org/abs/2601.03840 · **PDF:** https://arxiv.org/pdf/2601.03840v1
+
+## 23542. Logic Tensor Network-Enhanced Generative Adversarial Network
+
+- **ID:** `2601.03839` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Nijesh Upreti, Vaishak Belle
+- **abs:** https://arxiv.org/abs/2601.03839 · **PDF:** https://arxiv.org/pdf/2601.03839v1
+
+## 23543. Propositional Abduction via Only-Knowing: A Non-Monotonic Approach
+
+- **ID:** `2601.04272` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Sanderson Molick, Vaishak Belle
+- **abs:** https://arxiv.org/abs/2601.04272 · **PDF:** https://arxiv.org/pdf/2601.04272v1
+
+## 23544. Correcting Autonomous Driving Object Detection Misclassifications with Automated Commonsense Reasoning
+
+- **ID:** `2601.04271` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Keegan Kimbrell, Wang Tianhao, Feng Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.04271 · **PDF:** https://arxiv.org/pdf/2601.04271v1
+
+## 23545. The Voice of Equity: A Systematic Evaluation of Bias Mitigation Techniques for Speech-Based Cognitive Impairment Detection Across Architectures and Demographics
+
+- **ID:** `2601.16989` · **primary:** `eess.AS` · **Submitted:** 2026-01-07
+- **Authors:** Yasaman Haghbin, Sina Rashidi, Ali Zolnour, et al.
+- **abs:** https://arxiv.org/abs/2601.16989 · **PDF:** https://arxiv.org/pdf/2601.16989v1
+
+## 23546. Beyond Physical Labels: Redefining Domains for Robust WiFi-based Gesture Recognition
+
+- **ID:** `2601.03825` · **primary:** `cs.HC` · **Submitted:** 2026-01-07
+- **Authors:** Xiang Zhang, Huan Yan, Jinyang Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.03825 · **PDF:** https://arxiv.org/pdf/2601.03825v2
+
+## 23547. IDESplat: Iterative Depth Probability Estimation for Generalizable 3D Gaussian Splatting
+
+- **ID:** `2601.03824` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Wei Long, Haifeng Wu, Shiyin Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.03824 · **PDF:** https://arxiv.org/pdf/2601.03824v3
+
+## 23548. Step Potential Advantage Estimation: Harnessing Intermediate Confidence and Correctness for Efficient Mathematical Reasoning
+
+- **ID:** `2601.03823` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Fei Wu, Zhenrong Zhang, Qikai Chang, et al.
+- **abs:** https://arxiv.org/abs/2601.03823 · **PDF:** https://arxiv.org/pdf/2601.03823v1
+
+## 23549. ROI-Reasoning: Rational Optimization for Inference via Pre-Computation Meta-Cognition
+
+- **ID:** `2601.03822` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Muyang Zhao, Qi Qi, Hao Sun
+- **abs:** https://arxiv.org/abs/2601.03822 · **PDF:** https://arxiv.org/pdf/2601.03822v1
+
+## 23550. Predictable Gradient Manifolds in Deep Learning: Temporal Path-Length and Intrinsic Rank as a Complexity Regime
+
+- **ID:** `2601.04270` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Anherutowa Calvo
+- **abs:** https://arxiv.org/abs/2601.04270 · **PDF:** https://arxiv.org/pdf/2601.04270v1
+
+## 23551. Systems Explaining Systems: A Framework for Intelligence and Consciousness
+
+- **ID:** `2601.04269` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Sean Niklas Semmler
+- **abs:** https://arxiv.org/abs/2601.04269 · **PDF:** https://arxiv.org/pdf/2601.04269v1
+
+## 23552. Making Tunable Parameters State-Dependent in Weather and Climate Models with Reinforcement Learning
+
+- **ID:** `2601.04268` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Pritthijit Nath, Sebastian Schemm, Henry Moss, et al.
+- **abs:** https://arxiv.org/abs/2601.04268 · **PDF:** https://arxiv.org/pdf/2601.04268v1
+
+## 23553. AI Generated Text Detection
+
+- **ID:** `2601.03812` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Adilkhan Alikhanov, Aidar Amangeldi, Diar Demeubay, et al.
+- **abs:** https://arxiv.org/abs/2601.03812 · **PDF:** https://arxiv.org/pdf/2601.03812v1
+
+## 23554. EvalBlocks: A Modular Pipeline for Rapidly Evaluating Foundation Models in Medical Imaging
+
+- **ID:** `2601.03811` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Jan Tagscherer, Sarah de Boer, Lena Philipp, et al.
+- **abs:** https://arxiv.org/abs/2601.03811 · **PDF:** https://arxiv.org/pdf/2601.03811v2
+
+## 23555. From Brute Force to Semantic Insight: Performance-Guided Data Transformation Design with LLMs
+
+- **ID:** `2601.03808` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Usha Shrestha, Dmitry Ignatov, Radu Timofte
+- **abs:** https://arxiv.org/abs/2601.03808 · **PDF:** https://arxiv.org/pdf/2601.03808v1
+
+## 23556. Detecting Semantic Backdoors in a Mystery Shopping Scenario
+
+- **ID:** `2601.03805` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Arpad Berta, Gabor Danner, Istvan Hegedus, et al.
+- **abs:** https://arxiv.org/abs/2601.03805 · **PDF:** https://arxiv.org/pdf/2601.03805v1
+
+## 23557. Quantum vs. Classical Machine Learning: A Benchmark Study for Financial Prediction
+
+- **ID:** `2601.03802` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Rehan Ahmad, Muhammad Kashif, Nouhaila Innan, et al.
+- **abs:** https://arxiv.org/abs/2601.03802 · **PDF:** https://arxiv.org/pdf/2601.03802v1
+
+## 23558. Physically Consistent Machine Learning for Melting Temperature Prediction of Refractory High-Entropy Alloys
+
+- **ID:** `2601.03801` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-07
+- **Authors:** Mohd Hasnain
+- **abs:** https://arxiv.org/abs/2601.03801 · **PDF:** https://arxiv.org/pdf/2601.03801v1
+
+## 23559. Where meaning lives: Layer-wise accessibility of psycholinguistic features in encoder and decoder language models
+
+- **ID:** `2601.03798` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Taisiia Tikhomirova, Dirk U. Wulff
+- **abs:** https://arxiv.org/abs/2601.03798 · **PDF:** https://arxiv.org/pdf/2601.03798v1
+
+## 23560. An Algorithmic Framework for Systematic Literature Reviews: A Case Study for Financial Narratives
+
+- **ID:** `2601.03794` · **primary:** `q-fin.GN` · **Submitted:** 2026-01-07
+- **Authors:** Gabin Taibi, Joerg Osterrieder
+- **abs:** https://arxiv.org/abs/2601.03794 · **PDF:** https://arxiv.org/pdf/2601.03794v1
+
+## 23561. Prompt Tuning without Labeled Samples for Zero-Shot Node Classification in Text-Attributed Graphs
+
+- **ID:** `2601.03793` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Sethupathy Parameswaran, Suresh Sundaram, Yuan Fang
+- **abs:** https://arxiv.org/abs/2601.03793 · **PDF:** https://arxiv.org/pdf/2601.03793v1
+
+## 23562. VietMed-MCQ: A Consistency-Filtered Data Synthesis Framework for Vietnamese Traditional Medicine Evaluation
+
+- **ID:** `2601.03792` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Huynh Trung Kiet, Dao Sy Duy Minh, Nguyen Dinh Ha Duong, et al.
+- **abs:** https://arxiv.org/abs/2601.03792 · **PDF:** https://arxiv.org/pdf/2601.03792v1
+
+## 23563. Do LLMs Really Memorize Personally Identifiable Information? Revisiting PII Leakage with a Cue-Controlled Memorization Framework
+
+- **ID:** `2601.03791` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Xiaoyu Luo, Yiyi Chen, Qiongxiu Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03791 · **PDF:** https://arxiv.org/pdf/2601.03791v1
+
+## 23564. NeoAMT: Neologism-Aware Agentic Machine Translation with Reinforcement Learning
+
+- **ID:** `2601.03790` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Zhongtao Miao, Kaiyan Zhao, Masaaki Nagata, et al.
+- **abs:** https://arxiv.org/abs/2601.03790 · **PDF:** https://arxiv.org/pdf/2601.03790v2
+
+## 23565. Criminal Liability of Generative Artificial Intelligence Providers for User-Generated Child Sexual Abuse Material
+
+- **ID:** `2601.03788` · **primary:** `cs.CY` · **Submitted:** 2026-01-07
+- **Authors:** Anamaria Mojica-Hanke, Thomas Goger, Svenja Wölfel, et al.
+- **abs:** https://arxiv.org/abs/2601.03788 · **PDF:** https://arxiv.org/pdf/2601.03788v1
+
+## 23566. Compact Example-Based Explanations for Language Models
+
+- **ID:** `2601.03786` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Loris Schoenegger, Benjamin Roth
+- **abs:** https://arxiv.org/abs/2601.03786 · **PDF:** https://arxiv.org/pdf/2601.03786v1
+
+## 23567. Membox: Weaving Topic Continuity into Long-Range Memory for LLM Agents
+
+- **ID:** `2601.03785` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Dehao Tao, Guoliang Ma, Yongfeng Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.03785 · **PDF:** https://arxiv.org/pdf/2601.03785v2
+
+## 23568. HearSay Benchmark: Do Audio LLMs Leak What They Hear?
+
+- **ID:** `2601.03783` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Jin Wang, Liang Lin, Kaiwen Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.03783 · **PDF:** https://arxiv.org/pdf/2601.03783v1
+
+## 23569. PointWorld: Scaling 3D World Models for In-The-Wild Robotic Manipulation
+
+- **ID:** `2601.03782` · **primary:** `cs.RO` · **Submitted:** 2026-01-07
+- **Authors:** Wenlong Huang, Yu-Wei Chao, Arsalan Mousavian, et al.
+- **abs:** https://arxiv.org/abs/2601.03782 · **PDF:** https://arxiv.org/pdf/2601.03782v1
+
+## 23570. Tracing the complexity profiles of different linguistic phenomena through the intrinsic dimension of LLM representations
+
+- **ID:** `2601.03779` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Marco Baroni, Emily Cheng, Iria de-Dios-Flores, et al.
+- **abs:** https://arxiv.org/abs/2601.03779 · **PDF:** https://arxiv.org/pdf/2601.03779v1
+
+## 23571. Improving Compactness and Reducing Ambiguity of CFIRE Rule-Based Explanations
+
+- **ID:** `2601.03776` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Sebastian Müller, Tobias Schneider, Ruben Kemna, et al.
+- **abs:** https://arxiv.org/abs/2601.03776 · **PDF:** https://arxiv.org/pdf/2601.03776v1
+
+## 23572. Do LLM Self-Explanations Help Users Predict Model Behavior? Evaluating Counterfactual Simulatability with Pragmatic Perturbations
+
+- **ID:** `2601.03775` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Pingjun Hong, Benjamin Roth
+- **abs:** https://arxiv.org/abs/2601.03775 · **PDF:** https://arxiv.org/pdf/2601.03775v1
+
+## 23573. Scalable Machine Learning Force Fields for Macromolecular Systems Through Long-Range Aware Message Passing
+
+- **ID:** `2601.03774` · **primary:** `physics.chem-ph` · **Submitted:** 2026-01-07
+- **Authors:** Chu Wang, Lin Huang, Xinran Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.03774 · **PDF:** https://arxiv.org/pdf/2601.03774v1
+
+## 23574. EntroCoT: Enhancing Chain-of-Thought via Adaptive Entropy-Guided Segmentation
+
+- **ID:** `2601.03769` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Zihang Li, Yuhang Wang, Yikun Zong, et al.
+- **abs:** https://arxiv.org/abs/2601.03769 · **PDF:** https://arxiv.org/pdf/2601.03769v3
+
+## 23575. Learning Shrinks the Hard Tail: Training-Dependent Inference Scaling in a Solvable Linear Model
+
+- **ID:** `2601.03764` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Noam Levi
+- **abs:** https://arxiv.org/abs/2601.03764 · **PDF:** https://arxiv.org/pdf/2601.03764v1
+
+## 23576. Probabilistic Transformers for Joint Modeling of Global Weather Dynamics and Decision-Centric Variables
+
+- **ID:** `2601.03753` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Paulius Rauba, Viktor Cikojevic, Fran Bartolic, et al.
+- **abs:** https://arxiv.org/abs/2601.03753 · **PDF:** https://arxiv.org/pdf/2601.03753v1
+
+## 23577. Evaluation of Multilingual LLMs Personalized Text Generation Capabilities Targeting Groups and Social-Media Platforms
+
+- **ID:** `2601.03752` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Dominik Macko
+- **abs:** https://arxiv.org/abs/2601.03752 · **PDF:** https://arxiv.org/pdf/2601.03752v1
+
+## 23578. Bridging OLAP and RAG: A Multidimensional Approach to the Design of Corpus Partitioning
+
+- **ID:** `2601.03748` · **primary:** `cs.IR` · **Submitted:** 2026-01-07
+- **Authors:** Dario Maio, Stefano Rizzi
+- **abs:** https://arxiv.org/abs/2601.03748 · **PDF:** https://arxiv.org/pdf/2601.03748v1
+
+## 23579. Whose Facts Win? LLM Source Preferences under Knowledge Conflicts
+
+- **ID:** `2601.03746` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Jakob Schuster, Vagrant Gautam, Katja Markert
+- **abs:** https://arxiv.org/abs/2601.03746 · **PDF:** https://arxiv.org/pdf/2601.03746v2
+
+## 23580. O-Researcher: An Open Ended Deep Research Model via Multi-Agent Distillation and Agentic RL
+
+- **ID:** `2601.03743` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yi Yao, He Zhu, Piaohong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.03743 · **PDF:** https://arxiv.org/pdf/2601.03743v1
+
+## 23581. RadDiff: Describing Differences in Radiology Image Sets with Natural Language
+
+- **ID:** `2601.03733` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Xiaoxian Shen, Yuhui Zhang, Sahithi Ankireddy, et al.
+- **abs:** https://arxiv.org/abs/2601.03733 · **PDF:** https://arxiv.org/pdf/2601.03733v1
+
+## 23582. From Laboratory to Real-World Applications: Benchmarking Agentic Code Reasoning at the Repository Level
+
+- **ID:** `2601.03731` · **primary:** `cs.SE` · **Submitted:** 2026-01-07
+- **Authors:** Jia Li, Yuxin Su, Michael R. Lyu
+- **abs:** https://arxiv.org/abs/2601.03731 · **PDF:** https://arxiv.org/pdf/2601.03731v2
+
+## 23583. CSMCIR: CoT-Enhanced Symmetric Alignment with Memory Bank for Composed Image Retrieval
+
+- **ID:** `2601.03728` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Zhipeng Qian, Zihan Liang, Yufei Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.03728 · **PDF:** https://arxiv.org/pdf/2601.03728v1
+
+## 23584. Stuttering-Aware Automatic Speech Recognition for Indonesian Language
+
+- **ID:** `2601.03727` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Fadhil Muhammad, Alwin Djuliansah, Adrian Aryaputra Hamzah, et al.
+- **abs:** https://arxiv.org/abs/2601.03727 · **PDF:** https://arxiv.org/pdf/2601.03727v2
+
+## 23585. EDCO: Dynamic Curriculum Orchestration for Domain-specific Large Language Model Fine-tuning
+
+- **ID:** `2601.03725` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Jing-Cheng Pang, Liu Sun, Chang Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.03725 · **PDF:** https://arxiv.org/pdf/2601.03725v1
+
+## 23586. ETR: Outcome-Guided Elastic Trust Regions for Policy Optimization
+
+- **ID:** `2601.03723` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Shijie Zhang, Kevin Zhang, Zheyuan Gu, et al.
+- **abs:** https://arxiv.org/abs/2601.03723 · **PDF:** https://arxiv.org/pdf/2601.03723v1
+
+## 23587. MIND: From Passive Mimicry to Active Reasoning through Capability-Aware Multi-Perspective CoT Distillation
+
+- **ID:** `2601.03717` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Jin Cui, Jiaqi Guo, Jiepeng Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.03717 · **PDF:** https://arxiv.org/pdf/2601.03717v1
+
+## 23588. The environmental impact of ICT in the era of data and artificial intelligence
+
+- **ID:** `2601.06174` · **primary:** `cs.CY` · **Submitted:** 2026-01-07
+- **Authors:** François Rottenberg, Thomas Feys, Liesbet Van der Perre
+- **abs:** https://arxiv.org/abs/2601.06174 · **PDF:** https://arxiv.org/pdf/2601.06174v1
+
+## 23589. R$^3$L: Reflect-then-Retry Reinforcement Learning with Language-Guided Exploration, Pivotal Credit, and Positive Amplification
+
+- **ID:** `2601.03715` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Weijie Shi, Yanxi Chen, Zexi Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03715 · **PDF:** https://arxiv.org/pdf/2601.03715v1
+
+## 23590. Visual Merit or Linguistic Crutch? A Close Look at DeepSeek-OCR
+
+- **ID:** `2601.03714` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yunhao Liang, Ruixuan Ying, Bo Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03714 · **PDF:** https://arxiv.org/pdf/2601.03714v2
+
+## 23591. State Backdoor: Towards Stealthy Real-world Poisoning Attack on Vision-Language-Action Model in State Space
+
+- **ID:** `2601.04266` · **primary:** `cs.CR` · **Submitted:** 2026-01-07
+- **Authors:** Ji Guo, Wenbo Jiang, Yansong Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.04266 · **PDF:** https://arxiv.org/pdf/2601.04266v1
+
+## 23592. The Power of 10: New Rules for the Digital World
+
+- **ID:** `2601.03709` · **primary:** `cs.CY` · **Submitted:** 2026-01-07
+- **Authors:** Sarah Spiekermann-Hoff, Marc Langheinrich, Johannes Hoff, et al.
+- **abs:** https://arxiv.org/abs/2601.03709 · **PDF:** https://arxiv.org/pdf/2601.03709v1
+
+## 23593. MHRC-Bench: A Multilingual Hardware Repository-Level Code Completion benchmark
+
+- **ID:** `2601.03708` · **primary:** `cs.PL` · **Submitted:** 2026-01-07
+- **Authors:** Qingyun Zou, Jiahao Cui, Nuo Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.03708 · **PDF:** https://arxiv.org/pdf/2601.03708v2
+
+## 23594. AirNav: A Large-Scale Real-World UAV Vision-and-Language Navigation Dataset with Natural and Diverse Instructions
+
+- **ID:** `2601.03707` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Hengxing Cai, Yijie Rao, Ligang Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.03707 · **PDF:** https://arxiv.org/pdf/2601.03707v1
+
+## 23595. The Geometry of the Pivot: A Note on Lazy Pivoted Cholesky and Farthest Point Sampling
+
+- **ID:** `2601.03706` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Gil Shabat
+- **abs:** https://arxiv.org/abs/2601.03706 · **PDF:** https://arxiv.org/pdf/2601.03706v3
+
+## 23596. Investigating Knowledge Distillation Through Neural Networks for Protein Binding Affinity Prediction
+
+- **ID:** `2601.03704` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Wajid Arshad Abbasi, Syed Ali Abbas, Maryum Bibi, et al.
+- **abs:** https://arxiv.org/abs/2601.03704 · **PDF:** https://arxiv.org/pdf/2601.03704v1
+
+## 23597. TreeAdv: Tree-Structured Advantage Redistribution for Group-Based RL
+
+- **ID:** `2601.03703` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Lang Cao, Hui Ruan, Yongqian Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03703 · **PDF:** https://arxiv.org/pdf/2601.03703v1
+
+## 23598. Inference Attacks Against Graph Generative Diffusion Models
+
+- **ID:** `2601.03701` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Xiuling Wang, Xin Huang, Guibo Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.03701 · **PDF:** https://arxiv.org/pdf/2601.03701v1
+
+## 23599. ADEPT: Adaptive Dynamic Early-Exit Process for Transformers
+
+- **ID:** `2601.03700` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Sangmin Yoo, Srikanth Malla, Chiho Choi, et al.
+- **abs:** https://arxiv.org/abs/2601.03700 · **PDF:** https://arxiv.org/pdf/2601.03700v1
+
+## 23600. RedBench: A Universal Dataset for Comprehensive Red Teaming of Large Language Models
+
+- **ID:** `2601.03699` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Quy-Anh Dang, Chris Ngo, Truong-Son Hy
+- **abs:** https://arxiv.org/abs/2601.03699 · **PDF:** https://arxiv.org/pdf/2601.03699v1
+
+## 23601. Evaluation Framework for AI Creativity: A Case Study Based on Story Generation
+
+- **ID:** `2601.03698` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Pharath Sathya, Yin Jou Huang, Fei Cheng
+- **abs:** https://arxiv.org/abs/2601.03698 · **PDF:** https://arxiv.org/pdf/2601.03698v1
+
+## 23602. Can AI Chatbots Provide Coaching in Engineering? Beyond Information Processing Toward Mastery
+
+- **ID:** `2601.03693` · **primary:** `cs.CY` · **Submitted:** 2026-01-07
+- **Authors:** Junaid Qadir, Muhammad Adil Attique, Saleha Shoaib, et al.
+- **abs:** https://arxiv.org/abs/2601.03693 · **PDF:** https://arxiv.org/pdf/2601.03693v1
+
+## 23603. The Psychology of Learning from Machines: Anthropomorphic AI and the Paradox of Automation in Education
+
+- **ID:** `2601.06172` · **primary:** `cs.CY` · **Submitted:** 2026-01-07
+- **Authors:** Junaid Qadir, Muhammad Mumtaz
+- **abs:** https://arxiv.org/abs/2601.06172 · **PDF:** https://arxiv.org/pdf/2601.06172v2
+
+## 23604. From Individual Prompts to Collective Intelligence: Mainstreaming Generative AI in the Classroom
+
+- **ID:** `2601.06171` · **primary:** `cs.CY` · **Submitted:** 2026-01-07
+- **Authors:** Junaid Qadir, Muhammad Salman Khan
+- **abs:** https://arxiv.org/abs/2601.06171 · **PDF:** https://arxiv.org/pdf/2601.06171v1
+
+## 23605. A Pre-trained Reaction Embedding Descriptor Capturing Bond Transformation Patterns
+
+- **ID:** `2601.03689` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Weiqi Liu, Fenglei Cao, Yuan Qi, et al.
+- **abs:** https://arxiv.org/abs/2601.03689 · **PDF:** https://arxiv.org/pdf/2601.03689v1
+
+## 23606. Personalized Medication Planning via Direct Domain Modeling and LLM-Generated Heuristics
+
+- **ID:** `2601.03687` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Yonatan Vernik, Alexander Tuisov, David Izhaki, et al.
+- **abs:** https://arxiv.org/abs/2601.03687 · **PDF:** https://arxiv.org/pdf/2601.03687v1
+
+## 23607. Rethinking Recurrent Neural Networks for Time Series Forecasting: A Reinforced Recurrent Encoder with Prediction-Oriented Proximal Policy Optimization
+
+- **ID:** `2601.03683` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Xin Lai, Shiming Deng, Lu Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.03683 · **PDF:** https://arxiv.org/pdf/2601.03683v2
+
+## 23608. From Implicit to Explicit: Token-Efficient Logical Supervision for Mathematical Reasoning in LLMs
+
+- **ID:** `2601.03682` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Shaojie Wang, Liang Zhang
+- **abs:** https://arxiv.org/abs/2601.03682 · **PDF:** https://arxiv.org/pdf/2601.03682v1
+
+## 23609. Accounting for Optimal Control in the Sizing of Isolated Hybrid Renewable Energy Systems Using Imitation Learning
+
+- **ID:** `2601.03679` · **primary:** `eess.SY` · **Submitted:** 2026-01-07
+- **Authors:** Simon Halvdansson, Lucas Ferreira Bernardino, Brage Rugstad Knudsen
+- **abs:** https://arxiv.org/abs/2601.03679 · **PDF:** https://arxiv.org/pdf/2601.03679v1
+
+## 23610. Towards Compositional Generalization of LLMs via Skill Taxonomy Guided Data Synthesis
+
+- **ID:** `2601.03676` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yifan Wei, Li Du, Xiaoyan Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.03676 · **PDF:** https://arxiv.org/pdf/2601.03676v1
+
+## 23611. Intelligence Degradation in Long-Context LLMs: Critical Threshold Determination via Natural Length Distribution Analysis
+
+- **ID:** `2601.15300` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Weiwei Wang, Jiyong Min, Weijie Zou
+- **abs:** https://arxiv.org/abs/2601.15300 · **PDF:** https://arxiv.org/pdf/2601.15300v1
+
+## 23612. Disentangling Aleatoric and Epistemic Uncertainty in Physics-Informed Neural Networks. Application to Insulation Material Degradation Prognostics
+
+- **ID:** `2601.03673` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Ibai Ramirez, Jokin Alcibar, Joel Pino, et al.
+- **abs:** https://arxiv.org/abs/2601.03673 · **PDF:** https://arxiv.org/pdf/2601.03673v1
+
+## 23613. Sandwich Reasoning: An Answer-Reasoning-Answer Approach for Low-Latency Query Correction
+
+- **ID:** `2601.03672` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Chen Zhang, Kepu Zhang, Jiatong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.03672 · **PDF:** https://arxiv.org/pdf/2601.03672v1
+
+## 23614. NeuronScope: A Multi-Agent Framework for Explaining Polysemantic Neurons in Language Models
+
+- **ID:** `2601.03671` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Weiqi Liu, Yongliang Miao, Haiyan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.03671 · **PDF:** https://arxiv.org/pdf/2601.03671v1
+
+## 23615. DisastQA: A Comprehensive Benchmark for Evaluating Question Answering in Disaster Management
+
+- **ID:** `2601.03670` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Zhitong Chen, Kai Yin, Xiangjue Dong, et al.
+- **abs:** https://arxiv.org/abs/2601.03670 · **PDF:** https://arxiv.org/pdf/2601.03670v1
+
+## 23616. MemKD: Memory-Discrepancy Knowledge Distillation for Efficient Time Series Classification
+
+- **ID:** `2601.04264` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Nilushika Udayangani, Kishor Nandakishor, Marimuthu Palaniswami
+- **abs:** https://arxiv.org/abs/2601.04264 · **PDF:** https://arxiv.org/pdf/2601.04264v1
+
+## 23617. eTracer: Towards Traceable Text Generation via Claim-Level Grounding
+
+- **ID:** `2601.03669` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Bohao Chu, Qianli Wang, Hendrik Damm, et al.
+- **abs:** https://arxiv.org/abs/2601.03669 · **PDF:** https://arxiv.org/pdf/2601.03669v1
+
+## 23618. Discontinuous Galerkin finite element operator network for solving non-smooth PDEs
+
+- **ID:** `2601.03668` · **primary:** `math.NA` · **Submitted:** 2026-01-07
+- **Authors:** Kapil Chawla, Youngjoon Hong, Jae Yong Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.03668 · **PDF:** https://arxiv.org/pdf/2601.03668v1
+
+## 23619. TRec: Learning Hand-Object Interactions through 2D Point Track Motion
+
+- **ID:** `2601.03667` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Dennis Holzmann, Sven Wachsmuth
+- **abs:** https://arxiv.org/abs/2601.03667 · **PDF:** https://arxiv.org/pdf/2601.03667v3
+
+## 23620. e5-omni: Explicit Cross-modal Alignment for Omni-modal Embeddings
+
+- **ID:** `2601.03666` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Haonan Chen, Sicheng Gao, Radu Timofte, et al.
+- **abs:** https://arxiv.org/abs/2601.03666 · **PDF:** https://arxiv.org/pdf/2601.03666v2
+
+## 23621. Stochastic Voronoi Ensembles for Anomaly Detection
+
+- **ID:** `2601.03664` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Yang Cao, Sikun Yang, Xuyun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.03664 · **PDF:** https://arxiv.org/pdf/2601.03664v2
+
+## 23622. How Does the Thinking Step Influence Model Safety? An Entropy-based Safety Reminder for LRMs
+
+- **ID:** `2601.03662` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Su-Hyeon Kim, Hyundong Jin, Yejin Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.03662 · **PDF:** https://arxiv.org/pdf/2601.03662v1
+
+## 23623. Learning to Reason: Temporal Saliency Distillation for Interpretable Knowledge Transfer
+
+- **ID:** `2601.04263` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Nilushika Udayangani Hewa Dehigahawattage, Kishor Nandakishor, Marimuthu Palaniswami
+- **abs:** https://arxiv.org/abs/2601.04263 · **PDF:** https://arxiv.org/pdf/2601.04263v1
+
+## 23624. AMIR-GRPO: Inducing Implicit Preference Signals into GRPO
+
+- **ID:** `2601.03661` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Amir Hossein Yari, Fajri Koto
+- **abs:** https://arxiv.org/abs/2601.03661 · **PDF:** https://arxiv.org/pdf/2601.03661v1
+
+## 23625. Group and Exclusive Sparse Regularization-based Continual Learning of CNNs
+
+- **ID:** `2601.03658` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Basile Tousside, Janis Mohr, Jörg Frochte
+- **abs:** https://arxiv.org/abs/2601.03658 · **PDF:** https://arxiv.org/pdf/2601.03658v1
+
+## 23626. In Search of Grandmother Cells: Tracing Interpretable Neurons in Tabular Representations
+
+- **ID:** `2601.03657` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Ricardo Knauer, Erik Rodner
+- **abs:** https://arxiv.org/abs/2601.03657 · **PDF:** https://arxiv.org/pdf/2601.03657v1
+
+## 23627. Quantum Classical Ridgelet Neural Network For Time Series Model
+
+- **ID:** `2601.03654` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Bahadur Yadav, Sanjay Kumar Mohanty
+- **abs:** https://arxiv.org/abs/2601.03654 · **PDF:** https://arxiv.org/pdf/2601.03654v1
+
+## 23628. SyncThink: A Training-Free Strategy to Align Inference Termination with Reasoning Saturation
+
+- **ID:** `2601.03649` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Gengyang Li, Wang Cai, Yifeng Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.03649 · **PDF:** https://arxiv.org/pdf/2601.03649v1
+
+## 23629. ELO: Efficient Layer-Specific Optimization for Continual Pretraining of Multilingual LLMs
+
+- **ID:** `2601.03648` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** HanGyeol Yoo, ChangSu Choi, Minjun Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.03648 · **PDF:** https://arxiv.org/pdf/2601.03648v2
+
+## 23630. ReLA: Representation Learning and Aggregation for Job Scheduling with Reinforcement Learning
+
+- **ID:** `2601.03646` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Zhengyi Kwan, Wei Zhang, Aik Beng Ng, et al.
+- **abs:** https://arxiv.org/abs/2601.03646 · **PDF:** https://arxiv.org/pdf/2601.03646v2
+
+## 23631. LLM-MC-Affect: LLM-Based Monte Carlo Modeling of Affective Trajectories and Latent Ambiguity for Interpersonal Dynamic Insight
+
+- **ID:** `2601.03645` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yu-Zheng Lin, Bono Po-Jen Shih, John Paul Martin Encinas, et al.
+- **abs:** https://arxiv.org/abs/2601.03645 · **PDF:** https://arxiv.org/pdf/2601.03645v1
+
+## 23632. Agent-Dice: Disentangling Knowledge Updates via Geometric Consensus for Agent Continual Learning
+
+- **ID:** `2601.03641` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Zheng Wu, Xingyu Lou, Xinbei Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.03641 · **PDF:** https://arxiv.org/pdf/2601.03641v3
+
+## 23633. MALTopic: Multi-Agent LLM Topic Modeling Framework
+
+- **ID:** `2601.15299` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yash Sharma
+- **abs:** https://arxiv.org/abs/2601.15299 · **PDF:** https://arxiv.org/pdf/2601.15299v1
+
+## 23634. Kantorovich-Type Stochastic Neural Network Operators for the Mean-Square Approximation of Certain Second-Order Stochastic Processes
+
+- **ID:** `2601.03634` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Sachin Saini, Uaday Singh
+- **abs:** https://arxiv.org/abs/2601.03634 · **PDF:** https://arxiv.org/pdf/2601.03634v1
+
+## 23635. MFC-RFNet: A Multi-scale Guided Rectified Flow Network for Radar Sequence Prediction
+
+- **ID:** `2601.03633` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Wenjie Luo, Chuanhu Deng, Chaorong Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03633 · **PDF:** https://arxiv.org/pdf/2601.03633v1
+
+## 23636. ReStyle-TTS: Relative and Continuous Style Control for Zero-Shot Speech Synthesis
+
+- **ID:** `2601.03632` · **primary:** `eess.AS` · **Submitted:** 2026-01-07
+- **Authors:** Haitao Li, Chunxiang Jin, Chenglin Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03632 · **PDF:** https://arxiv.org/pdf/2601.03632v1
+
+## 23637. Reasoning Model Is Superior LLM-Judge, Yet Suffers from Biases
+
+- **ID:** `2601.03630` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Hui Huang, Xuanxin Wu, Muyun Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.03630 · **PDF:** https://arxiv.org/pdf/2601.03630v1
+
+## 23638. Learning Shortest Paths When Data is Scarce
+
+- **ID:** `2601.03629` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Dmytro Matsypura, Yu Pan, Hanzhao Wang
+- **abs:** https://arxiv.org/abs/2601.03629 · **PDF:** https://arxiv.org/pdf/2601.03629v1
+
+## 23639. Evaluating the Pre-Consultation Ability of LLMs using Diagnostic Guidelines
+
+- **ID:** `2601.03627` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Jean Seo, Gibaeg Kim, Kihun Shin, et al.
+- **abs:** https://arxiv.org/abs/2601.03627 · **PDF:** https://arxiv.org/pdf/2601.03627v2
+
+## 23640. Learning from Limited Labels: Transductive Graph Label Propagation for Indian Music Analysis
+
+- **ID:** `2601.03626` · **primary:** `eess.AS` · **Submitted:** 2026-01-07
+- **Authors:** Parampreet Singh, Akshay Raina, Sayeedul Islam Sheikh, et al.
+- **abs:** https://arxiv.org/abs/2601.03626 · **PDF:** https://arxiv.org/pdf/2601.03626v1
+
+## 23641. Architecting Agentic Communities using Design Patterns
+
+- **ID:** `2601.03624` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Zoran Milosevic, Fethi Rabhi
+- **abs:** https://arxiv.org/abs/2601.03624 · **PDF:** https://arxiv.org/pdf/2601.03624v2
+
+## 23642. Safety-Utility Conflicts Are Not Global: Surgical Alignment via Head-Level Diagnosis
+
+- **ID:** `2601.04262` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Wang Cai, Yilin Wen, Jinchang Hou, et al.
+- **abs:** https://arxiv.org/abs/2601.04262 · **PDF:** https://arxiv.org/pdf/2601.04262v1
+
+## 23643. Parent-Guided Adaptive Reliability (PGAR): A Behavioural Meta-Learning Framework for Stable and Trustworthy AI
+
+- **ID:** `2601.06167` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Anshum Rankawat
+- **abs:** https://arxiv.org/abs/2601.06167 · **PDF:** https://arxiv.org/pdf/2601.06167v1
+
+## 23644. Systematic Evaluation of Depth Backbones and Semantic Cues for Monocular Pseudo-LiDAR 3D Detection
+
+- **ID:** `2601.03617` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Samson Oseiwe Ajadalu
+- **abs:** https://arxiv.org/abs/2601.03617 · **PDF:** https://arxiv.org/pdf/2601.03617v1
+
+## 23645. Analyzing Reasoning Shifts in Audio Deepfake Detection under Adversarial Attacks: The Reasoning Tax versus Shield Bifurcation
+
+- **ID:** `2601.03615` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Binh Nguyen, Thai Le
+- **abs:** https://arxiv.org/abs/2601.03615 · **PDF:** https://arxiv.org/pdf/2601.03615v1
+
+## 23646. Mathematical Foundations of Polyphonic Music Generation via Structural Inductive Bias
+
+- **ID:** `2601.03612` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Joonwon Seo
+- **abs:** https://arxiv.org/abs/2601.03612 · **PDF:** https://arxiv.org/pdf/2601.03612v6
+
+## 23647. Investigation into respiratory sound classification for an imbalanced data set using hybrid LSTM-KAN architectures
+
+- **ID:** `2601.03610` · **primary:** `cs.SD` · **Submitted:** 2026-01-07
+- **Authors:** Nithinkumar K., Anand R
+- **abs:** https://arxiv.org/abs/2601.03610 · **PDF:** https://arxiv.org/pdf/2601.03610v1
+
+## 23648. Policy-Guided Search on Tree-of-Thoughts for Efficient Problem Solving with Bounded Language Model Queries
+
+- **ID:** `2601.03606` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Sumedh Pendurkar, Guni Sharon
+- **abs:** https://arxiv.org/abs/2601.03606 · **PDF:** https://arxiv.org/pdf/2601.03606v1
+
+## 23649. DiVA: Fine-grained Factuality Verification with Agentic-Discriminative Verifier
+
+- **ID:** `2601.03605` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Hui Huang, Muyun Yang, Yuki Arase
+- **abs:** https://arxiv.org/abs/2601.03605 · **PDF:** https://arxiv.org/pdf/2601.03605v1
+
+## 23650. Interleaved Tool-Call Reasoning for Protein Function Understanding
+
+- **ID:** `2601.03604` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Chuanliu Fan, Zicheng Ma, Huanran Meng, et al.
+- **abs:** https://arxiv.org/abs/2601.03604 · **PDF:** https://arxiv.org/pdf/2601.03604v2
+
+## 23651. A Comparative Study of Traditional Machine Learning, Deep Learning, and Large Language Models for Mental Health Forecasting using Smartphone Sensing Data
+
+- **ID:** `2601.03603` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Kaidong Feng, Zhu Sun, Roy Ka-Wei Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.03603 · **PDF:** https://arxiv.org/pdf/2601.03603v2
+
+## 23652. ALERT: Zero-shot LLM Jailbreak Detection via Internal Discrepancy Amplification
+
+- **ID:** `2601.03600` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Xiao Lin, Philip Li, Zhichen Zeng, et al.
+- **abs:** https://arxiv.org/abs/2601.03600 · **PDF:** https://arxiv.org/pdf/2601.03600v1
+
+## 23653. From Chains to Graphs: Self-Structured Reasoning for General-Domain LLMs
+
+- **ID:** `2601.03597` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yingjian Chen, Haoran Liu, Yinhong Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.03597 · **PDF:** https://arxiv.org/pdf/2601.03597v2
+
+## 23654. Controllable LLM Reasoning via Sparse Autoencoder-Based Steering
+
+- **ID:** `2601.03595` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Yi Fang, Wenjie Wang, Mingfeng Xue, et al.
+- **abs:** https://arxiv.org/abs/2601.03595 · **PDF:** https://arxiv.org/pdf/2601.03595v1
+
+## 23655. Can LLMs See Without Pixels? Benchmarking Spatial Intelligence from Textual Descriptions
+
+- **ID:** `2601.03590` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Zhongbin Guo, Zhen Yang, Yushan Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03590 · **PDF:** https://arxiv.org/pdf/2601.03590v1
+
+## 23656. OLA: Output Language Alignment in Code-Switched LLM Interactions
+
+- **ID:** `2601.03589` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Juhyun Oh, Haneul Yoo, Faiz Ghifari Haznitrama, et al.
+- **abs:** https://arxiv.org/abs/2601.03589 · **PDF:** https://arxiv.org/pdf/2601.03589v1
+
+## 23657. Deontic Knowledge Graphs for Privacy Compliance in Multimodal Disaster Data Sharing
+
+- **ID:** `2601.03587` · **primary:** `cs.CR` · **Submitted:** 2026-01-07
+- **Authors:** Kelvin Uzoma Echenim, Karuna Pande Joshi
+- **abs:** https://arxiv.org/abs/2601.03587 · **PDF:** https://arxiv.org/pdf/2601.03587v1
+
+## 23658. Local Gradient Regulation Stabilizes Federated Learning under Client Heterogeneity
+
+- **ID:** `2601.03584` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Ping Luo, Jiahuan Wang, Ziqing Wen, et al.
+- **abs:** https://arxiv.org/abs/2601.03584 · **PDF:** https://arxiv.org/pdf/2601.03584v1
+
+## 23659. PsychEthicsBench: Evaluating Large Language Models Against Australian Mental Health Ethics
+
+- **ID:** `2601.03578` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yaling Shen, Stephanie Fong, Yiwen Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.03578 · **PDF:** https://arxiv.org/pdf/2601.03578v1
+
+## 23660. Variational Inference, Entropy, and Orthogonality: A Unified Theory of Mixture-of-Experts
+
+- **ID:** `2601.03577` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Ye Su, Yong Liu
+- **abs:** https://arxiv.org/abs/2601.03577 · **PDF:** https://arxiv.org/pdf/2601.03577v1
+
+## 23661. How Do Large Language Models Learn Concepts During Continual Pre-Training?
+
+- **ID:** `2601.03570` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Barry Menglong Yao, Sha Li, Yunzhi Yao, et al.
+- **abs:** https://arxiv.org/abs/2601.03570 · **PDF:** https://arxiv.org/pdf/2601.03570v1
+
+## 23662. Local Intrinsic Dimensionality of Ground Motion Data for Early Detection of Complex Catastrophic Slope Failure
+
+- **ID:** `2601.03569` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Yuansan Liu, Antoinette Tordesillas, James Bailey
+- **abs:** https://arxiv.org/abs/2601.03569 · **PDF:** https://arxiv.org/pdf/2601.03569v2
+
+## 23663. PaperAudit-Bench: Benchmarking Error Detection in Research Papers for Critical Automated Peer Review
+
+- **ID:** `2601.19916` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Songjun Tu, Yiwen Ma, Jiahao Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.19916 · **PDF:** https://arxiv.org/pdf/2601.19916v1
+
+## 23664. Provably Convergent Decentralized Optimization over Directed Graphs under Generalized Smoothness
+
+- **ID:** `2601.03566` · **primary:** `math.OC` · **Submitted:** 2026-01-07
+- **Authors:** Yanan Bo, Yongqiang Wang
+- **abs:** https://arxiv.org/abs/2601.03566 · **PDF:** https://arxiv.org/pdf/2601.03566v1
+
+## 23665. A Proposed Paradigm for Imputing Missing Multi-Sensor Data in the Healthcare Domain
+
+- **ID:** `2601.03565` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Vaibhav Gupta, Florian Grensing, Beyza Cinar, et al.
+- **abs:** https://arxiv.org/abs/2601.03565 · **PDF:** https://arxiv.org/pdf/2601.03565v1
+
+## 23666. Towards a Mechanistic Understanding of Propositional Logical Reasoning in Large Language Models
+
+- **ID:** `2601.04260` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Danchun Chen, Qiyao Yan, Liangming Pan
+- **abs:** https://arxiv.org/abs/2601.04260 · **PDF:** https://arxiv.org/pdf/2601.04260v1
+
+## 23667. Reinforced Linear Genetic Programming
+
+- **ID:** `2601.09736` · **primary:** `cs.NE` · **Submitted:** 2026-01-07
+- **Authors:** Urmzd Mukhammadnaim
+- **abs:** https://arxiv.org/abs/2601.09736 · **PDF:** https://arxiv.org/pdf/2601.09736v1
+
+## 23668. Modeling Next-Token Prediction as Left-Nested Intuitionistic Implication
+
+- **ID:** `2601.19915` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Paul Tarau
+- **abs:** https://arxiv.org/abs/2601.19915 · **PDF:** https://arxiv.org/pdf/2601.19915v1
+
+## 23669. Neural Operators for Biomedical Spherical Heterogeneity
+
+- **ID:** `2601.03561` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Hao Tang, Hao Chen, Hao Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03561 · **PDF:** https://arxiv.org/pdf/2601.03561v3
+
+## 23670. DiffCoT: Diffusion-styled Chain-of-Thought Reasoning in LLMs
+
+- **ID:** `2601.03559` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Shidong Cao, Hongzhan Lin, Yuxuan Gu, et al.
+- **abs:** https://arxiv.org/abs/2601.03559 · **PDF:** https://arxiv.org/pdf/2601.03559v1
+
+## 23671. SCRIBE: Structured Mid-Level Supervision for Tool-Using Language Models
+
+- **ID:** `2601.03555` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Yuxuan Jiang, Francis Ferraro
+- **abs:** https://arxiv.org/abs/2601.03555 · **PDF:** https://arxiv.org/pdf/2601.03555v1
+
+## 23672. Evaluating LLMs for Police Decision-Making: A Framework Based on Police Action Scenarios
+
+- **ID:** `2601.03553` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Sangyub Lee, Heedou Kim, Hyeoncheol Kim
+- **abs:** https://arxiv.org/abs/2601.03553 · **PDF:** https://arxiv.org/pdf/2601.03553v1
+
+## 23673. ReEfBench: Quantifying the Reasoning Efficiency of LLMs
+
+- **ID:** `2601.03550` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Zhizhang Fu, Yuancheng Gu, Chenkai Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.03550 · **PDF:** https://arxiv.org/pdf/2601.03550v1
+
+## 23674. EASLT: Emotion-Aware Sign Language Translation
+
+- **ID:** `2601.03549` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Guobin Tu, Di Weng
+- **abs:** https://arxiv.org/abs/2601.03549 · **PDF:** https://arxiv.org/pdf/2601.03549v1
+
+## 23675. Value-Action Alignment in Large Language Models under Privacy-Prosocial Conflict
+
+- **ID:** `2601.03546` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Guanyu Chen, Chenxiao Yu, Xiyang Hu
+- **abs:** https://arxiv.org/abs/2601.03546 · **PDF:** https://arxiv.org/pdf/2601.03546v1
+
+## 23676. EvolMem: A Cognitive-Driven Benchmark for Multi-Session Dialogue Memory
+
+- **ID:** `2601.03543` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Ye Shen, Dun Pei, Yiqiu Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.03543 · **PDF:** https://arxiv.org/pdf/2601.03543v1
+
+## 23677. Layer-Order Inversion: Rethinking Latent Multi-Hop Reasoning in Large Language Models
+
+- **ID:** `2601.03542` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Xukai Liu, Ye Liu, Jipeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.03542 · **PDF:** https://arxiv.org/pdf/2601.03542v1
+
+## 23678. DeepSynth-Eval: Objectively Evaluating Information Consolidation in Deep Survey Writing
+
+- **ID:** `2601.03540` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Hongzhi Zhang, Yuanze Hu, Tinghai Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.03540 · **PDF:** https://arxiv.org/pdf/2601.03540v1
+
+## 23679. STAR-S: Improving Safety Alignment through Self-Taught Reasoning on Safety Rules
+
+- **ID:** `2601.03537` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Di Wu, Yanyan Zhao, Xin Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.03537 · **PDF:** https://arxiv.org/pdf/2601.03537v1
+
+## 23680. Persona-aware and Explainable Bikeability Assessment: A Vision-Language Model Approach
+
+- **ID:** `2601.03534` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yilong Dai, Ziyi Wang, Chenguang Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.03534 · **PDF:** https://arxiv.org/pdf/2601.03534v1
+
+## 23681. Online Learning with Limited Information in the Sliding Window Model
+
+- **ID:** `2601.03533` · **primary:** `stat.ML` · **Submitted:** 2026-01-07
+- **Authors:** Vladimir Braverman, Sumegha Garg, Chen Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.03533 · **PDF:** https://arxiv.org/pdf/2601.03533v1
+
+## 23682. PALM-Bench: A Comprehensive Benchmark for Personalized Audio-Language Models
+
+- **ID:** `2601.03531` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yuwen Wang, Xinyuan Qian, Tian-Hao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.03531 · **PDF:** https://arxiv.org/pdf/2601.03531v1
+
+## 23683. What Users Leave Unsaid: Under-Specified Queries Limit Vision-Language Models
+
+- **ID:** `2601.06165` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Dasol Choi, Guijin Son, Hanwool Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.06165 · **PDF:** https://arxiv.org/pdf/2601.06165v1
+
+## 23684. VeRPO: Verifiable Dense Reward Policy Optimization for Code Generation
+
+- **ID:** `2601.03525` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Longwen Wang, Xuan'er Wu, Xiaohui Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.03525 · **PDF:** https://arxiv.org/pdf/2601.03525v2
+
+## 23685. Retrieval Heads are Dynamic
+
+- **ID:** `2602.11162` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yuping Lin, Zitao Li, Yue Xing, et al.
+- **abs:** https://arxiv.org/abs/2602.11162 · **PDF:** https://arxiv.org/pdf/2602.11162v1
+
+## 23686. Variance Computation for Weighted Model Counting with Knowledge Compilation Approach
+
+- **ID:** `2601.03523` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Kengo Nakamura, Masaaki Nishino, Norihito Yasuda
+- **abs:** https://arxiv.org/abs/2601.03523 · **PDF:** https://arxiv.org/pdf/2601.03523v1
+
+## 23687. A Reinforcement Learning-Based Model for Mapping and Goal-Directed Navigation Using Multiscale Place Fields
+
+- **ID:** `2601.03520` · **primary:** `cs.NE` · **Submitted:** 2026-01-07
+- **Authors:** Bekarys Dukenbaev, Andrew Gerstenslager, Alexander Johnson, et al.
+- **abs:** https://arxiv.org/abs/2601.03520 · **PDF:** https://arxiv.org/pdf/2601.03520v1
+
+## 23688. Mem-Gallery: Benchmarking Multimodal Long-Term Conversational Memory for MLLM Agents
+
+- **ID:** `2601.03515` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Yuanchen Bei, Tianxin Wei, Xuying Ning, et al.
+- **abs:** https://arxiv.org/abs/2601.03515 · **PDF:** https://arxiv.org/pdf/2601.03515v1
+
+## 23689. Deploy-Master: Automating the Deployment of 50,000+ Agent-Ready Scientific Tools in One Day
+
+- **ID:** `2601.03513` · **primary:** `cs.SE` · **Submitted:** 2026-01-07
+- **Authors:** Yi Wang, Zhenting Huang, Zhaohan Ding, et al.
+- **abs:** https://arxiv.org/abs/2601.03513 · **PDF:** https://arxiv.org/pdf/2601.03513v1
+
+## 23690. Bootstrapping Code Translation with Weighted Multilanguage Exploration
+
+- **ID:** `2601.03512` · **primary:** `cs.SE` · **Submitted:** 2026-01-07
+- **Authors:** Yuhan Wu, Huan Zhang, Wei Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.03512 · **PDF:** https://arxiv.org/pdf/2601.03512v1
+
+## 23691. Adaptive Trust Metrics for Multi-LLM Systems: Enhancing Reliability in Regulated Industries
+
+- **ID:** `2601.08858` · **primary:** `cs.SE` · **Submitted:** 2026-01-07
+- **Authors:** Tejaswini Bollikonda
+- **abs:** https://arxiv.org/abs/2601.08858 · **PDF:** https://arxiv.org/pdf/2601.08858v1
+
+## 23692. IntroLM: Introspective Language Models via Prefilling-Time Self-Evaluation
+
+- **ID:** `2601.03511` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Hossein Hosseini Kasnavieh, Gholamreza Haffari, Chris Leckie, et al.
+- **abs:** https://arxiv.org/abs/2601.03511 · **PDF:** https://arxiv.org/pdf/2601.03511v1
+
+## 23693. Evolving Programmatic Skill Networks
+
+- **ID:** `2601.03509` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Haochen Shi, Xingdi Yuan, Bang Liu
+- **abs:** https://arxiv.org/abs/2601.03509 · **PDF:** https://arxiv.org/pdf/2601.03509v1
+
+## 23694. Contract2Plan: Verified Contract-Grounded Retrieval-Augmented Optimization for BOM-Aware Procurement and Multi-Echelon Inventory Planning
+
+- **ID:** `2601.06164` · **primary:** `cs.SE` · **Submitted:** 2026-01-07
+- **Authors:** Sahil Agarwal
+- **abs:** https://arxiv.org/abs/2601.06164 · **PDF:** https://arxiv.org/pdf/2601.06164v1
+
+## 23695. Reasoning Pattern Alignment Merging for Adaptive Reasoning
+
+- **ID:** `2601.03506` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Zhaofeng Zhong, Wei Yuan, Tong Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.03506 · **PDF:** https://arxiv.org/pdf/2601.03506v1
+
+## 23696. Beyond Perplexity: A Lightweight Benchmark for Knowledge Retention in Supervised Fine-Tuning
+
+- **ID:** `2601.03505` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Soheil Zibakhsh Shabgahi, Pedram Aghazadeh, Farinaz Koushanfar
+- **abs:** https://arxiv.org/abs/2601.03505 · **PDF:** https://arxiv.org/pdf/2601.03505v1
+
+## 23697. SDCD: Structure-Disrupted Contrastive Decoding for Mitigating Hallucinations in Large Vision-Language Models
+
+- **ID:** `2601.03500` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Yuxuan Xia, Siheng Wang, Peng Li
+- **abs:** https://arxiv.org/abs/2601.03500 · **PDF:** https://arxiv.org/pdf/2601.03500v1
+
+## 23698. A Quantifiable Information-Processing Hierarchy Provides a Necessary Condition for Detecting Agency
+
+- **ID:** `2601.03498` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-07
+- **Authors:** Brett J. Kagan, Valentina Baccetti, Brian D. Earp, et al.
+- **abs:** https://arxiv.org/abs/2601.03498 · **PDF:** https://arxiv.org/pdf/2601.03498v1
+
+## 23699. STELLA: Self-Reflective Terminology-Aware Framework for Building an Aerospace Information Retrieval Benchmark
+
+- **ID:** `2601.03496` · **primary:** `cs.IR` · **Submitted:** 2026-01-07
+- **Authors:** Bongmin Kim
+- **abs:** https://arxiv.org/abs/2601.03496 · **PDF:** https://arxiv.org/pdf/2601.03496v1
+
+## 23700. Cyberattack Detection in Virtualized Microgrids Using LightGBM and Knowledge-Distilled Classifiers
+
+- **ID:** `2601.03495` · **primary:** `eess.SY` · **Submitted:** 2026-01-07
+- **Authors:** Osasumwen Cedric Ogiesoba-Eguakun, Suman Rath
+- **abs:** https://arxiv.org/abs/2601.03495 · **PDF:** https://arxiv.org/pdf/2601.03495v2
+
+## 23701. Submodular Evaluation Subset Selection in Automatic Prompt Optimization
+
+- **ID:** `2601.03493` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Jinming Nian, Zhiyuan Peng, Hongwei Shang, et al.
+- **abs:** https://arxiv.org/abs/2601.03493 · **PDF:** https://arxiv.org/pdf/2601.03493v1
+
+## 23702. CroBIM-U: Uncertainty-Driven Referring Remote Sensing Image Segmentation
+
+- **ID:** `2601.03490` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Yuzhe Sun, Zhe Dong, Haochen Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.03490 · **PDF:** https://arxiv.org/pdf/2601.03490v1
+
+## 23703. From Bits to Chips: An LLM-based Hardware-Aware Quantization Agent for Streamlined Deployment of LLMs
+
+- **ID:** `2601.03484` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Kaiyuan Deng, Hangyu Zheng, Minghai Qing, et al.
+- **abs:** https://arxiv.org/abs/2601.03484 · **PDF:** https://arxiv.org/pdf/2601.03484v1
+
+## 23704. CALM: Culturally Self-Aware Language Models
+
+- **ID:** `2601.03483` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Lingzhi Shen, Xiaohao Cai, Yunfei Long, et al.
+- **abs:** https://arxiv.org/abs/2601.03483 · **PDF:** https://arxiv.org/pdf/2601.03483v1
+
+## 23705. Personalization of Large Foundation Models for Health Interventions
+
+- **ID:** `2601.03482` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Stefan Konigorski, Johannes E. Vedder, Babajide Alamu Owoyele, et al.
+- **abs:** https://arxiv.org/abs/2601.03482 · **PDF:** https://arxiv.org/pdf/2601.03482v1
+
+## 23706. Self-Explaining Hate Speech Detection with Moral Rationales
+
+- **ID:** `2601.03481` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** Francielle Vargas, Jackson Trager, Diego Alves, et al.
+- **abs:** https://arxiv.org/abs/2601.03481 · **PDF:** https://arxiv.org/pdf/2601.03481v1
+
+## 23707. Efficient Sequential Recommendation for Long Term User Interest Via Personalization
+
+- **ID:** `2601.03479` · **primary:** `cs.IR` · **Submitted:** 2026-01-07
+- **Authors:** Qiang Zhang, Hanchao Yu, Ivan Ji, et al.
+- **abs:** https://arxiv.org/abs/2601.03479 · **PDF:** https://arxiv.org/pdf/2601.03479v1
+
+## 23708. Forget-It-All: Multi-Concept Machine Unlearning via Concept-Aware Neuron Masking
+
+- **ID:** `2601.06163` · **primary:** `cs.CV` · **Submitted:** 2026-01-07
+- **Authors:** Kaiyuan Deng, Bo Hui, Gen Li, et al.
+- **abs:** https://arxiv.org/abs/2601.06163 · **PDF:** https://arxiv.org/pdf/2601.06163v1
+
+## 23709. Hybrid Approach for Driver Behavior Analysis with Machine Learning, Feature Optimization, and Explainable AI
+
+- **ID:** `2601.03477` · **primary:** `cs.LG` · **Submitted:** 2026-01-07
+- **Authors:** Mehedi Hasan Shuvo, Md. Raihan Tapader, Nur Mohammad Tamjid, et al.
+- **abs:** https://arxiv.org/abs/2601.03477 · **PDF:** https://arxiv.org/pdf/2601.03477v1
+
+## 23710. Online Decision-Making Under Uncertainty for Vehicle-to-Building Systems
+
+- **ID:** `2601.03476` · **primary:** `eess.SY` · **Submitted:** 2026-01-07
+- **Authors:** Rishav Sen, Yunuo Zhang, Fangqi Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.03476 · **PDF:** https://arxiv.org/pdf/2601.03476v1
+
+## 23711. CPGPrompt: Translating Clinical Guidelines into LLM-Executable Decision Support
+
+- **ID:** `2601.03475` · **primary:** `cs.AI` · **Submitted:** 2026-01-07
+- **Authors:** Ruiqi Deng, Geoffrey Martin, Tony Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.03475 · **PDF:** https://arxiv.org/pdf/2601.03475v1
+
+## 23712. SegNSP: Revisiting Next Sentence Prediction for Linear Text Segmentation
+
+- **ID:** `2601.03474` · **primary:** `cs.CL` · **Submitted:** 2026-01-07
+- **Authors:** José Isidro, Filipe Cunha, Purificação Silvano, et al.
+- **abs:** https://arxiv.org/abs/2601.03474 · **PDF:** https://arxiv.org/pdf/2601.03474v2
+
+## 23713. A Preliminary Agentic Framework for Matrix Deflation
+
+- **ID:** `2601.08219` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Paimon Goulart, Evangelos E. Papalexakis
+- **abs:** https://arxiv.org/abs/2601.08219 · **PDF:** https://arxiv.org/pdf/2601.08219v1
+
+## 23714. Forget Many, Forget Right: Scalable and Precise Concept Unlearning in Diffusion Models
+
+- **ID:** `2601.06162` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Kaiyuan Deng, Gen Li, Yang Xiao, et al.
+- **abs:** https://arxiv.org/abs/2601.06162 · **PDF:** https://arxiv.org/pdf/2601.06162v1
+
+## 23715. EpiQAL: Benchmarking Large Language Models in Epidemiological Question Answering for Enhanced Alignment and Reasoning
+
+- **ID:** `2601.03471` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Mingyang Wei, Dehai Min, Zewen Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.03471 · **PDF:** https://arxiv.org/pdf/2601.03471v2
+
+## 23716. Toward Maturity-Based Certification of Embodied AI: Quantifying Trustworthiness Through Measurement Mechanisms
+
+- **ID:** `2601.03470` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Michael C. Darling, Alan H. Hesu, Michael A. Mardikes, et al.
+- **abs:** https://arxiv.org/abs/2601.03470 · **PDF:** https://arxiv.org/pdf/2601.03470v2
+
+## 23717. Content vs. Form: What Drives the Writing Score Gap Across Socioeconomic Backgrounds? A Generated Panel Approach
+
+- **ID:** `2601.03469` · **primary:** `econ.EM` · **Submitted:** 2026-01-06
+- **Authors:** Nadav Kunievsky, Pedro Pertusi
+- **abs:** https://arxiv.org/abs/2601.03469 · **PDF:** https://arxiv.org/pdf/2601.03469v1
+
+## 23718. Latent Geometry of Taste: Scalable Low-Rank Matrix Factorization for Recommender Systems
+
+- **ID:** `2601.03466` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Joshua Salako
+- **abs:** https://arxiv.org/abs/2601.03466 · **PDF:** https://arxiv.org/pdf/2601.03466v2
+
+## 23719. Prompting Underestimates LLM Capability for Time Series Classification
+
+- **ID:** `2601.03464` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Dan Schumacher, Erfan Nourbakhsh, Rocky Slavin, et al.
+- **abs:** https://arxiv.org/abs/2601.03464 · **PDF:** https://arxiv.org/pdf/2601.03464v2
+
+## 23720. Experimental Comparison of Light-Weight and Deep CNN Models Across Diverse Datasets
+
+- **ID:** `2601.03463` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Md. Hefzul Hossain Papon, Shadman Rabby
+- **abs:** https://arxiv.org/abs/2601.03463 · **PDF:** https://arxiv.org/pdf/2601.03463v1
+
+## 23721. FROST-Drive: Scalable and Efficient End-to-End Driving with a Frozen Vision Encoder
+
+- **ID:** `2601.03460` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Zeyu Dong, Yimin Zhu, Yu Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.03460 · **PDF:** https://arxiv.org/pdf/2601.03460v1
+
+## 23722. An Expectation-Maximization Algorithm for Domain Adaptation in Gaussian Causal Models
+
+- **ID:** `2601.03459` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Mohammad Ali Javidian
+- **abs:** https://arxiv.org/abs/2601.03459 · **PDF:** https://arxiv.org/pdf/2601.03459v1
+
+## 23723. Cross-Language Speaker Attribute Prediction Using MIL and RL
+
+- **ID:** `2601.04257` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Sunny Shu, Seyed Sahand Mohammadi Ziabari, Ali Mohammed Mansoor Alsahag
+- **abs:** https://arxiv.org/abs/2601.04257 · **PDF:** https://arxiv.org/pdf/2601.04257v1
+
+## 23724. Automated Feedback Generation for Undergraduate Mathematics: Development and Evaluation of an AI Teaching Assistant
+
+- **ID:** `2601.03458` · **primary:** `cs.CY` · **Submitted:** 2026-01-06
+- **Authors:** Aron Gohr, Marie-Amelie Lawn, Kevin Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.03458 · **PDF:** https://arxiv.org/pdf/2601.03458v1
+
+## 23725. Measures of classification bias derived from sample size analysis
+
+- **ID:** `2601.03453` · **primary:** `stat.ME` · **Submitted:** 2026-01-06
+- **Authors:** Ioannis Ivrissimtzis, Shauna Concannon, Matthew Houliston, et al.
+- **abs:** https://arxiv.org/abs/2601.03453 · **PDF:** https://arxiv.org/pdf/2601.03453v1
+
+## 23726. Revisiting Software Engineering Education in the Era of Large Language Models: A Curriculum Adaptation and Academic Integrity Framework
+
+- **ID:** `2601.08857` · **primary:** `cs.SE` · **Submitted:** 2026-01-06
+- **Authors:** Mustafa Degerli
+- **abs:** https://arxiv.org/abs/2601.08857 · **PDF:** https://arxiv.org/pdf/2601.08857v2
+
+## 23727. Microeconomic Foundations of Multi-Agent Learning
+
+- **ID:** `2601.03451` · **primary:** `stat.ML` · **Submitted:** 2026-01-06
+- **Authors:** Nassim Helou
+- **abs:** https://arxiv.org/abs/2601.03451 · **PDF:** https://arxiv.org/pdf/2601.03451v1
+
+## 23728. Soft Contextualized Encoder For User Defined Text Classification
+
+- **ID:** `2601.03450` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Charu Maheshwari, Vyas Raina
+- **abs:** https://arxiv.org/abs/2601.03450 · **PDF:** https://arxiv.org/pdf/2601.03450v1
+
+## 23729. Enhancing Linguistic Competence of Language Models through Pre-training with Language Learning Tasks
+
+- **ID:** `2601.03448` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Atsuki Yamaguchi, Maggie Mi, Nikolaos Aletras
+- **abs:** https://arxiv.org/abs/2601.03448 · **PDF:** https://arxiv.org/pdf/2601.03448v1
+
+## 23730. Grading Scale Impact on LLM-as-a-Judge: Human-LLM Alignment Is Highest on 0-5 Grading Scale
+
+- **ID:** `2601.03444` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Weiyue Li, Minda Zhao, Weixuan Dong, et al.
+- **abs:** https://arxiv.org/abs/2601.03444 · **PDF:** https://arxiv.org/pdf/2601.03444v1
+
+## 23731. Discriminating real and synthetic super-resolved audio samples using embedding-based classifiers
+
+- **ID:** `2601.03443` · **primary:** `eess.AS` · **Submitted:** 2026-01-06
+- **Authors:** Mikhail Silaev, Konstantinos Drossos, Tuomas Virtanen
+- **abs:** https://arxiv.org/abs/2601.03443 · **PDF:** https://arxiv.org/pdf/2601.03443v1
+
+## 23732. Provable Acceleration of Distributed Optimization with Local Updates
+
+- **ID:** `2601.03442` · **primary:** `eess.SY` · **Submitted:** 2026-01-06
+- **Authors:** Zuang Wang, Yongqiang Wang
+- **abs:** https://arxiv.org/abs/2601.03442 · **PDF:** https://arxiv.org/pdf/2601.03442v3
+
+## 23733. MARVEL: A Multi Agent-based Research Validator and Enabler using Large Language Models
+
+- **ID:** `2601.03436` · **primary:** `astro-ph.IM` · **Submitted:** 2026-01-06
+- **Authors:** Nikhil Mukund, Yifang Luo, Fan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.03436 · **PDF:** https://arxiv.org/pdf/2601.03436v1
+
+## 23734. The Critical Role of Aspects in Measuring Document Similarity
+
+- **ID:** `2601.03435` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Eftekhar Hossain, Tarnika Hazra, Ahatesham Bhuiyan, et al.
+- **abs:** https://arxiv.org/abs/2601.03435 · **PDF:** https://arxiv.org/pdf/2601.03435v1
+
+## 23735. VNU-Bench: A Benchmarking Dataset for Multi-Source Multimodal News Video Understanding
+
+- **ID:** `2601.03434` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Zibo Liu, Muyang Li, Zhe Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.03434 · **PDF:** https://arxiv.org/pdf/2601.03434v1
+
+## 23736. DeepLeak: Privacy Enhancing Hardening of Model Explanations Against Membership Leakage
+
+- **ID:** `2601.03429` · **primary:** `cs.CR` · **Submitted:** 2026-01-06
+- **Authors:** Firas Ben Hmida, Zain Sbeih, Philemon Hailemariam, et al.
+- **abs:** https://arxiv.org/abs/2601.03429 · **PDF:** https://arxiv.org/pdf/2601.03429v1
+
+## 23737. Enhancing Model Context Protocol (MCP) with Context-Aware Server Collaboration
+
+- **ID:** `2601.11595` · **primary:** `cs.DC` · **Submitted:** 2026-01-06
+- **Authors:** Meenakshi Amulya Jayanti, X. Y. Han
+- **abs:** https://arxiv.org/abs/2601.11595 · **PDF:** https://arxiv.org/pdf/2601.11595v2
+
+## 23738. The Illusion of Specialization: Unveiling the Domain-Invariant "Standing Committee" in Mixture-of-Experts Models
+
+- **ID:** `2601.03425` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Yan Wang, Yitao Xu, Nanhan Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.03425 · **PDF:** https://arxiv.org/pdf/2601.03425v1
+
+## 23739. Spectral Archaeology: The Causal Topology of Model Evolution
+
+- **ID:** `2601.03424` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Valentin Noël
+- **abs:** https://arxiv.org/abs/2601.03424 · **PDF:** https://arxiv.org/pdf/2601.03424v1
+
+## 23740. Training-Free Adaptation of New-Generation LLMs using Legacy Clinical Models
+
+- **ID:** `2601.03423` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Sasha Ronaghi, Chloe Stanwyck, Asad Aali, et al.
+- **abs:** https://arxiv.org/abs/2601.03423 · **PDF:** https://arxiv.org/pdf/2601.03423v2
+
+## 23741. Jailbreaking LLMs Without Gradients or Priors: Effective and Transferable Attacks
+
+- **ID:** `2601.03420` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Zhakshylyk Nurlanov, Frank R. Schmidt, Florian Bernard
+- **abs:** https://arxiv.org/abs/2601.03420 · **PDF:** https://arxiv.org/pdf/2601.03420v1
+
+## 23742. PCoA: A New Benchmark for Medical Aspect-Based Summarization With Phrase-Level Context Attribution
+
+- **ID:** `2601.03418` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Bohao Chu, Sameh Frihat, Tabea M. G. Pakull, et al.
+- **abs:** https://arxiv.org/abs/2601.03418 · **PDF:** https://arxiv.org/pdf/2601.03418v1
+
+## 23743. Multi-Scale Negative Coupled Information Systems (MNCIS): A Unified Spectral Topology Framework for Stability in Turbulence, AI, and Biology
+
+- **ID:** `2601.11594` · **primary:** `physics.comp-ph` · **Submitted:** 2026-01-06
+- **Authors:** Pengyue Hou
+- **abs:** https://arxiv.org/abs/2601.11594 · **PDF:** https://arxiv.org/pdf/2601.11594v2
+
+## 23744. Implicit Graph, Explicit Retrieval: Towards Efficient and Interpretable Long-horizon Memory for Large Language Models
+
+- **ID:** `2601.03417` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Xin Zhang, Kailai Yang, Hao Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03417 · **PDF:** https://arxiv.org/pdf/2601.03417v1
+
+## 23745. Sensor to Pixels: Decentralized Swarm Gathering via Image-Based Reinforcement Learning
+
+- **ID:** `2601.03413` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Yigal Koifman, Eran Iceland, Erez Koifman, et al.
+- **abs:** https://arxiv.org/abs/2601.03413 · **PDF:** https://arxiv.org/pdf/2601.03413v1
+
+## 23746. Inferring Clinically Relevant Molecular Subtypes of Pancreatic Cancer from Routine Histopathology Using Deep Learning
+
+- **ID:** `2601.03410` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Abdul Rehman Akbar, Alejandro Levya, Ashwini Esnakula, et al.
+- **abs:** https://arxiv.org/abs/2601.03410 · **PDF:** https://arxiv.org/pdf/2601.03410v2
+
+## 23747. Tigrinya Number Verbalization: Rules, Algorithm, and Implementation
+
+- **ID:** `2601.03403` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Fitsum Gaim, Issayas Tesfamariam
+- **abs:** https://arxiv.org/abs/2601.03403 · **PDF:** https://arxiv.org/pdf/2601.03403v1
+
+## 23748. Beyond Accuracy: A Decision-Theoretic Framework for Allocation-Aware Healthcare AI
+
+- **ID:** `2601.06161` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Rifa Ferzana
+- **abs:** https://arxiv.org/abs/2601.06161 · **PDF:** https://arxiv.org/pdf/2601.06161v1
+
+## 23749. Rendering Data Unlearnable by Exploiting LLM Alignment Mechanisms
+
+- **ID:** `2601.03401` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Ruihan Zhang, Jun Sun
+- **abs:** https://arxiv.org/abs/2601.03401 · **PDF:** https://arxiv.org/pdf/2601.03401v1
+
+## 23750. Eye-Q: A Multilingual Benchmark for Visual Word Puzzle Solving and Image-to-Phrase Reasoning
+
+- **ID:** `2601.03400` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Ali Najar, Alireza Mirrokni, Arshia Izadyari, et al.
+- **abs:** https://arxiv.org/abs/2601.03400 · **PDF:** https://arxiv.org/pdf/2601.03400v1
+
+## 23751. Scaling Trends for Multi-Hop Contextual Reasoning in Mid-Scale Language Models
+
+- **ID:** `2601.04254` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Brady Steele, Micah Katz
+- **abs:** https://arxiv.org/abs/2601.04254 · **PDF:** https://arxiv.org/pdf/2601.04254v1
+
+## 23752. PIVONet: A Physically-Informed Variational Neuro ODE Model for Efficient Advection-Diffusion Fluid Simulation
+
+- **ID:** `2601.03397` · **primary:** `cs.CE` · **Submitted:** 2026-01-06
+- **Authors:** Hei Shing Cheung, Qicheng Long, Zhiyue Lin
+- **abs:** https://arxiv.org/abs/2601.03397 · **PDF:** https://arxiv.org/pdf/2601.03397v1
+
+## 23753. Breaking the Assistant Mold: Modeling Behavioral Variation in LLM Based Procedural Character Generation
+
+- **ID:** `2601.03396` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Maan Qraitem, Kate Saenko, Bryan A. Plummer
+- **abs:** https://arxiv.org/abs/2601.03396 · **PDF:** https://arxiv.org/pdf/2601.03396v1
+
+## 23754. Simulating Complex Multi-Turn Tool Calling Interactions in Stateless Execution Environments
+
+- **ID:** `2601.19914` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Maxwell Crouse, Ibrahim Abdelaziz, Kshitij Fadnis, et al.
+- **abs:** https://arxiv.org/abs/2601.19914 · **PDF:** https://arxiv.org/pdf/2601.19914v1
+
+## 23755. Better, But Not Sufficient: Testing Video ANNs Against Macaque IT Dynamics
+
+- **ID:** `2601.03392` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Matteo Dunnhofer, Christian Micheloni, Kohitij Kar
+- **abs:** https://arxiv.org/abs/2601.03392 · **PDF:** https://arxiv.org/pdf/2601.03392v1
+
+## 23756. Exploration Through Introspection: A Self-Aware Reward Model
+
+- **ID:** `2601.03389` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Michael Petrowski, Milica Gašić
+- **abs:** https://arxiv.org/abs/2601.03389 · **PDF:** https://arxiv.org/pdf/2601.03389v1
+
+## 23757. Metaphors are a Source of Cross-Domain Misalignment of Large Reasoning Models
+
+- **ID:** `2601.03388` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Zhibo Hu, Chen Wang, Yanfeng Shu, et al.
+- **abs:** https://arxiv.org/abs/2601.03388 · **PDF:** https://arxiv.org/pdf/2601.03388v2
+
+## 23758. SIGMA: Scalable Spectral Insights for LLM Model Collapse
+
+- **ID:** `2601.03385` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Yi Gu, Lingyou Pang, Xiangkun Ye, et al.
+- **abs:** https://arxiv.org/abs/2601.03385 · **PDF:** https://arxiv.org/pdf/2601.03385v3
+
+## 23759. Weather-Aware Transformer for Real-Time Route Optimization in Drone-as-a-Service Operations
+
+- **ID:** `2601.03376` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Kamal Mohamed, Lillian Wassim, Ali Hamdi, et al.
+- **abs:** https://arxiv.org/abs/2601.03376 · **PDF:** https://arxiv.org/pdf/2601.03376v1
+
+## 23760. Enhancing Small Dataset Classification Using Projected Quantum Kernels with Convolutional Neural Networks
+
+- **ID:** `2601.03375` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** A. M. A. S. D. Alagiyawanna, Asoka Karunananda, A. Mahasinghe, et al.
+- **abs:** https://arxiv.org/abs/2601.03375 · **PDF:** https://arxiv.org/pdf/2601.03375v1
+
+## 23761. RiskCueBench: Benchmarking Anticipatory Reasoning from Early Risk Cues in Video-Language Models
+
+- **ID:** `2601.03369` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Sha Luo, Yogesh Prabhu, Timothy Ossowski, et al.
+- **abs:** https://arxiv.org/abs/2601.03369 · **PDF:** https://arxiv.org/pdf/2601.03369v2
+
+## 23762. A path to natural language through tokenisation and transformers
+
+- **ID:** `2601.03368` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** David S. Berman, Alexander G. Stapleton
+- **abs:** https://arxiv.org/abs/2601.03368 · **PDF:** https://arxiv.org/pdf/2601.03368v1
+
+## 23763. Physics-Informed Gaussian Process Regression for the Constitutive Modeling of Concrete: A Data-Driven Improvement to Phenomenological Models
+
+- **ID:** `2601.03367` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Chenyang Li, Himanshu Sharma, Youcai Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.03367 · **PDF:** https://arxiv.org/pdf/2601.03367v1
+
+## 23764. Enhancing LLM Instruction Following: An Evaluation-Driven Multi-Agentic Workflow for Prompt Instructions Optimization
+
+- **ID:** `2601.03359` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Alberto Purpura, Li Wang, Sahil Badyal, et al.
+- **abs:** https://arxiv.org/abs/2601.03359 · **PDF:** https://arxiv.org/pdf/2601.03359v1
+
+## 23765. Digital Red Queen: Adversarial Program Evolution in Core War with LLMs
+
+- **ID:** `2601.03335` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Akarsh Kumar, Ryan Bahlous-Boldi, Prafull Sharma, et al.
+- **abs:** https://arxiv.org/abs/2601.03335 · **PDF:** https://arxiv.org/pdf/2601.03335v1
+
+## 23766. Automated Semantic Rules Detection (ASRD) for Emergent Communication Interpretation
+
+- **ID:** `2601.03254` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Bastien Vanderplaetse, Xavier Siebert, Stéphane Dupont
+- **abs:** https://arxiv.org/abs/2601.03254 · **PDF:** https://arxiv.org/pdf/2601.03254v1
+
+## 23767. Sphinx: Benchmarking and Modeling for LLM-Driven Pull Request Review
+
+- **ID:** `2601.04252` · **primary:** `cs.SE` · **Submitted:** 2026-01-06
+- **Authors:** Daoan Zhang, Shuo Zhang, Zijian Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.04252 · **PDF:** https://arxiv.org/pdf/2601.04252v1
+
+## 23768. STReasoner: Empowering LLMs for Spatio-Temporal Reasoning in Time Series via Spatial-Aware Reinforcement Learning
+
+- **ID:** `2601.03248` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Juntong Ni, Shiyu Wang, Ming Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.03248 · **PDF:** https://arxiv.org/pdf/2601.03248v1
+
+## 23769. Self-Supervised Learning from Noisy and Incomplete Data
+
+- **ID:** `2601.03244` · **primary:** `stat.ML` · **Submitted:** 2026-01-06
+- **Authors:** Julián Tachella, Mike Davies
+- **abs:** https://arxiv.org/abs/2601.03244 · **PDF:** https://arxiv.org/pdf/2601.03244v1
+
+## 23770. PET-TURTLE: Deep Unsupervised Support Vector Machines for Imbalanced Data Clusters
+
+- **ID:** `2601.03237` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Javier Salazar Cavazos
+- **abs:** https://arxiv.org/abs/2601.03237 · **PDF:** https://arxiv.org/pdf/2601.03237v1
+
+## 23771. MAGMA: A Multi-Graph based Agentic Memory Architecture for AI Agents
+
+- **ID:** `2601.03236` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Dongming Jiang, Yi Li, Guanpeng Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03236 · **PDF:** https://arxiv.org/pdf/2601.03236v1
+
+## 23772. Shallow-circuit Supervised Learning on a Quantum Processor
+
+- **ID:** `2601.03235` · **primary:** `quant-ph` · **Submitted:** 2026-01-06
+- **Authors:** Luca Candelori, Swarnadeep Majumder, Antonio Mezzacapo, et al.
+- **abs:** https://arxiv.org/abs/2601.03235 · **PDF:** https://arxiv.org/pdf/2601.03235v1
+
+## 23773. Multi-RADS Synthetic Radiology Report Dataset and Head-to-Head Benchmarking of 41 Open-Weight and Proprietary Language Models
+
+- **ID:** `2601.03232` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Kartik Bose, Abhinandan Kumar, Raghuraman Soundararajan, et al.
+- **abs:** https://arxiv.org/abs/2601.03232 · **PDF:** https://arxiv.org/pdf/2601.03232v1
+
+## 23774. Using Grok to Avoid Personal Attacks While Correcting Misinformation on X
+
+- **ID:** `2601.04251` · **primary:** `cs.SI` · **Submitted:** 2026-01-06
+- **Authors:** Kevin Matthe Caramancion
+- **abs:** https://arxiv.org/abs/2601.04251 · **PDF:** https://arxiv.org/pdf/2601.04251v1
+
+## 23775. The Sonar Moment: Benchmarking Audio-Language Models in Audio Geo-Localization
+
+- **ID:** `2601.03227` · **primary:** `cs.SD` · **Submitted:** 2026-01-06
+- **Authors:** Ruixing Zhang, Zihan Liu, Leilei Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.03227 · **PDF:** https://arxiv.org/pdf/2601.03227v1
+
+## 23776. The Fake Friend Dilemma: Trust and the Political Economy of Conversational AI
+
+- **ID:** `2601.03222` · **primary:** `cs.CY` · **Submitted:** 2026-01-06
+- **Authors:** Jacob Erickson
+- **abs:** https://arxiv.org/abs/2601.03222 · **PDF:** https://arxiv.org/pdf/2601.03222v1
+
+## 23777. From Entropy to Epiplexity: Rethinking Information for Computationally Bounded Intelligence
+
+- **ID:** `2601.03220` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Marc Finzi, Shikai Qiu, Yiding Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.03220 · **PDF:** https://arxiv.org/pdf/2601.03220v2
+
+## 23778. LUT-KAN: Segment-wise LUT Quantization for Fast KAN Inference
+
+- **ID:** `2601.03332` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Oleksandr Kuznetsov
+- **abs:** https://arxiv.org/abs/2601.03332 · **PDF:** https://arxiv.org/pdf/2601.03332v1
+
+## 23779. MalruleLib: Large-Scale Executable Misconception Reasoning with Step Traces for Modeling Student Thinking in Mathematics
+
+- **ID:** `2601.03217` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Xinghe Chen, Naiming Liu, Shashank Sonkar
+- **abs:** https://arxiv.org/abs/2601.03217 · **PDF:** https://arxiv.org/pdf/2601.03217v1
+
+## 23780. Critic-Guided Reinforcement Unlearning in Text-to-Image Diffusion
+
+- **ID:** `2601.03213` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Mykola Vysotskyi, Zahar Kohut, Mariia Shpir, et al.
+- **abs:** https://arxiv.org/abs/2601.03213 · **PDF:** https://arxiv.org/pdf/2601.03213v3
+
+## 23781. Fine-tuning Small Language Models as Efficient Enterprise Search Relevance Labelers
+
+- **ID:** `2601.03211` · **primary:** `cs.IR` · **Submitted:** 2026-01-06
+- **Authors:** Yue Kang, Zhuoyi Huang, Benji Schussheim, et al.
+- **abs:** https://arxiv.org/abs/2601.03211 · **PDF:** https://arxiv.org/pdf/2601.03211v1
+
+## 23782. MMErroR: A Benchmark for Erroneous Reasoning in Vision-Language Models
+
+- **ID:** `2601.03331` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Yang Shi, Yifeng Xie, Minzhe Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.03331 · **PDF:** https://arxiv.org/pdf/2601.03331v1
+
+## 23783. UltraLogic: Enhancing LLM Reasoning through Large-Scale Data Synthesis and Bipolar Float Reward
+
+- **ID:** `2601.03205` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Yile Liu, Yixian Liu, Zongwei Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03205 · **PDF:** https://arxiv.org/pdf/2601.03205v1
+
+## 23784. InfiAgent: An Infinite-Horizon Framework for General-Purpose Autonomous Agents
+
+- **ID:** `2601.03204` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Chenglin Yu, Yuchen Wang, Songmiao Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.03204 · **PDF:** https://arxiv.org/pdf/2601.03204v1
+
+## 23785. Counterfactual Fairness with Graph Uncertainty
+
+- **ID:** `2601.03203` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Davi Valério, Chrysoula Zerva, Mariana Pinto, et al.
+- **abs:** https://arxiv.org/abs/2601.03203 · **PDF:** https://arxiv.org/pdf/2601.03203v1
+
+## 23786. Recursive querying of neural networks via weighted structures
+
+- **ID:** `2601.03201` · **primary:** `cs.LO` · **Submitted:** 2026-01-06
+- **Authors:** Martin Grohe, Christoph Standke, Juno Steegmans, et al.
+- **abs:** https://arxiv.org/abs/2601.03201 · **PDF:** https://arxiv.org/pdf/2601.03201v1
+
+## 23787. DIP: Dynamic In-Context Planner For Diffusion Language Models
+
+- **ID:** `2601.03199` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Yang Li, Han Meng, Chenan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.03199 · **PDF:** https://arxiv.org/pdf/2601.03199v1
+
+## 23788. Empowering Reliable Visual-Centric Instruction Following in MLLMs
+
+- **ID:** `2601.03198` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Weilei He, Feng Ju, Zhiyuan Fan, et al.
+- **abs:** https://arxiv.org/abs/2601.03198 · **PDF:** https://arxiv.org/pdf/2601.03198v1
+
+## 23789. Sparse Knowledge Distillation: A Mathematical Framework for Probability-Domain Temperature Scaling and Multi-Stage Compression
+
+- **ID:** `2601.03195` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Aaron R. Flouro, Shawn P. Chadwick
+- **abs:** https://arxiv.org/abs/2601.03195 · **PDF:** https://arxiv.org/pdf/2601.03195v1
+
+## 23790. X-MuTeST: A Multilingual Benchmark for Explainable Hate Speech Detection and A Novel LLM-consulted Explanation Framework
+
+- **ID:** `2601.03194` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Mohammad Zia Ur Rehman, Sai Kartheek Reddy Kasu, Shashivardhan Reddy Koppula, et al.
+- **abs:** https://arxiv.org/abs/2601.03194 · **PDF:** https://arxiv.org/pdf/2601.03194v1
+
+## 23791. UniCorn: Towards Self-Improving Unified Multimodal Models through Self-Generated Supervision
+
+- **ID:** `2601.03193` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Ruiyan Han, Zhen Fang, XinYu Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.03193 · **PDF:** https://arxiv.org/pdf/2601.03193v2
+
+## 23792. MemRL: Self-Evolving Agents via Runtime Reinforcement Learning on Episodic Memory
+
+- **ID:** `2601.03192` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Shengtao Zhang, Jiaqian Wang, Ruiwen Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.03192 · **PDF:** https://arxiv.org/pdf/2601.03192v2
+
+## 23793. AnatomiX, an Anatomy-Aware Grounded Multimodal Large Language Model for Chest X-Ray Interpretation
+
+- **ID:** `2601.03191` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Anees Ur Rehman Hashmi, Numan Saeed, Christoph Lippert
+- **abs:** https://arxiv.org/abs/2601.03191 · **PDF:** https://arxiv.org/pdf/2601.03191v2
+
+## 23794. Attention mechanisms in neural networks
+
+- **ID:** `2601.03329` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Hasi Hays
+- **abs:** https://arxiv.org/abs/2601.03329 · **PDF:** https://arxiv.org/pdf/2601.03329v1
+
+## 23795. Maximizing Local Entropy Where It Matters: Prefix-Aware Localized LLM Unlearning
+
+- **ID:** `2601.03190` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Naixin Zhai, Pengyang Shao, Binbin Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.03190 · **PDF:** https://arxiv.org/pdf/2601.03190v2
+
+## 23796. Decentralized Autoregressive Generation
+
+- **ID:** `2601.03184` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Stepan Maschan, Haoxuan Qu, Jun Liu
+- **abs:** https://arxiv.org/abs/2601.03184 · **PDF:** https://arxiv.org/pdf/2601.03184v2
+
+## 23797. Multi-Modal Data-Enhanced Foundation Models for Prediction and Control in Wireless Networks: A Survey
+
+- **ID:** `2601.03181` · **primary:** `cs.NI` · **Submitted:** 2026-01-06
+- **Authors:** Han Zhang, Mohammad Farzanullah, Mohammad Ghassemi, et al.
+- **abs:** https://arxiv.org/abs/2601.03181 · **PDF:** https://arxiv.org/pdf/2601.03181v1
+
+## 23798. Predicting Time Pressure of Powered Two-Wheeler Riders for Proactive Safety Interventions
+
+- **ID:** `2601.03173` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Sumit S. Shevtekar, Chandresh K. Maurya, Gourab Sil, et al.
+- **abs:** https://arxiv.org/abs/2601.03173 · **PDF:** https://arxiv.org/pdf/2601.03173v1
+
+## 23799. Student Guides Teacher: Weak-to-Strong Inference via Spectral Orthogonal Exploration
+
+- **ID:** `2601.06160` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Dayu Wang, Jiaye Yang, Weikang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.06160 · **PDF:** https://arxiv.org/pdf/2601.06160v1
+
+## 23800. Can Embedding Similarity Predict Cross-Lingual Transfer? A Systematic Study on African Languages
+
+- **ID:** `2601.03168` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Tewodros Kederalah Idris, Prasenjit Mitra, Roald Eiselen
+- **abs:** https://arxiv.org/abs/2601.03168 · **PDF:** https://arxiv.org/pdf/2601.03168v1
+
+## 23801. Can we Improve Prediction of Psychotherapy Outcomes Through Pretraining With Simulated Data?
+
+- **ID:** `2601.06159` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Niklas Jacobs, Manuel C. Voelkle, Norbert Kathmann, et al.
+- **abs:** https://arxiv.org/abs/2601.06159 · **PDF:** https://arxiv.org/pdf/2601.06159v1
+
+## 23802. Dynamic Hyperparameter Importance for Efficient Multi-Objective Optimization
+
+- **ID:** `2601.03166` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Daphne Theodorakopoulos, Marcel Wever, Marius Lindauer
+- **abs:** https://arxiv.org/abs/2601.03166 · **PDF:** https://arxiv.org/pdf/2601.03166v1
+
+## 23803. WebAnchor: Anchoring Agent Planning to Stabilize Long-Horizon Web Reasoning
+
+- **ID:** `2601.03164` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Xinmiao Yu, Liwen Zhang, Xiaocheng Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.03164 · **PDF:** https://arxiv.org/pdf/2601.03164v2
+
+## 23804. On the Convergence Behavior of Preconditioned Gradient Descent Toward the Rich Learning Regime
+
+- **ID:** `2601.03162` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Shuai Jiang, Alexey Voronin, Eric Cyr, et al.
+- **abs:** https://arxiv.org/abs/2601.03162 · **PDF:** https://arxiv.org/pdf/2601.03162v1
+
+## 23805. Rapid Augmentations for Time Series (RATS): A High-Performance Library for Time Series Augmentation
+
+- **ID:** `2601.03159` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Wadie Skaf, Felix Kern, Aryamaan Basu Roy, et al.
+- **abs:** https://arxiv.org/abs/2601.03159 · **PDF:** https://arxiv.org/pdf/2601.03159v1
+
+## 23806. Prompt-Counterfactual Explanations for Generative AI System Behavior
+
+- **ID:** `2601.03156` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Sofie Goethals, Foster Provost, João Sedoc
+- **abs:** https://arxiv.org/abs/2601.03156 · **PDF:** https://arxiv.org/pdf/2601.03156v2
+
+## 23807. Decoupling the Effect of Chain-of-Thought Reasoning: A Human Label Variation Perspective
+
+- **ID:** `2601.03154` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Beiduo Chen, Tiancheng Hu, Caiqi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.03154 · **PDF:** https://arxiv.org/pdf/2601.03154v1
+
+## 23808. PersonaLedger: Generating Realistic Financial Transactions with Persona Conditioned LLMs and Rule Grounded Feedback
+
+- **ID:** `2601.03149` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Dehao Yuan, Tyler Farnan, Stefan Tesliuc, et al.
+- **abs:** https://arxiv.org/abs/2601.03149 · **PDF:** https://arxiv.org/pdf/2601.03149v2
+
+## 23809. Self-Verification is All You Need To Pass The Japanese Bar Examination
+
+- **ID:** `2601.03144` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Andrew Shin
+- **abs:** https://arxiv.org/abs/2601.03144 · **PDF:** https://arxiv.org/pdf/2601.03144v1
+
+## 23810. Accurate Table Question Answering with Accessible LLMs
+
+- **ID:** `2601.03137` · **primary:** `cs.DB` · **Submitted:** 2026-01-06
+- **Authors:** Yangfan Jiang, Fei Wei, Ergute Bao, et al.
+- **abs:** https://arxiv.org/abs/2601.03137 · **PDF:** https://arxiv.org/pdf/2601.03137v1
+
+## 23811. Limited Linguistic Diversity in Embodied AI Datasets
+
+- **ID:** `2601.03136` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Selma Wanna, Agnes Luhtaru, Jonathan Salfity, et al.
+- **abs:** https://arxiv.org/abs/2601.03136 · **PDF:** https://arxiv.org/pdf/2601.03136v1
+
+## 23812. Improving Indigenous Language Machine Translation with Synthetic Data and Language-Specific Preprocessing
+
+- **ID:** `2601.03135` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Aashish Dhawan, Christopher Driggers-Ellis, Christan Grant, et al.
+- **abs:** https://arxiv.org/abs/2601.03135 · **PDF:** https://arxiv.org/pdf/2601.03135v2
+
+## 23813. The Anatomy of Conversational Scams: A Topic-Based Red Teaming Analysis of Multi-Turn Interactions in LLMs
+
+- **ID:** `2601.03134` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Xiangzhe Yuan, Zhenhao Zhang, Haoming Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.03134 · **PDF:** https://arxiv.org/pdf/2601.03134v1
+
+## 23814. Finite Memory Belief Approximation for Optimal Control in Partially Observable Markov Decision Processes
+
+- **ID:** `2601.03132` · **primary:** `eess.SY` · **Submitted:** 2026-01-06
+- **Authors:** Mintae Kim
+- **abs:** https://arxiv.org/abs/2601.03132 · **PDF:** https://arxiv.org/pdf/2601.03132v1
+
+## 23815. Automatic Prompt Engineering with No Task Cues and No Tuning
+
+- **ID:** `2601.03130` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Faisal Chowdhury, Nandana Mihindukulasooriya, Niharika S D'Souza, et al.
+- **abs:** https://arxiv.org/abs/2601.03130 · **PDF:** https://arxiv.org/pdf/2601.03130v1
+
+## 23816. Unified Thinker: A General Reasoning Modular Core for Image Generation
+
+- **ID:** `2601.03127` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Sashuai Zhou, Qiang Zhou, Jijin Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.03127 · **PDF:** https://arxiv.org/pdf/2601.03127v1
+
+## 23817. LeafLife: An Explainable Deep Learning Framework with Robustness for Grape Leaf Disease Recognition
+
+- **ID:** `2601.03124` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** B. M. Shahria Alam, Md. Nasim Ahmed
+- **abs:** https://arxiv.org/abs/2601.03124 · **PDF:** https://arxiv.org/pdf/2601.03124v1
+
+## 23818. Gradient descent reliably finds depth- and gate-optimal circuits for generic unitaries
+
+- **ID:** `2601.03123` · **primary:** `quant-ph` · **Submitted:** 2026-01-06
+- **Authors:** Janani Gomathi, Alex Meiburg
+- **abs:** https://arxiv.org/abs/2601.03123 · **PDF:** https://arxiv.org/pdf/2601.03123v1
+
+## 23819. ToxiGAN: Toxic Data Augmentation via LLM-Guided Directional Adversarial Generation
+
+- **ID:** `2601.03121` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Peiran Li, Jan Fillies, Adrian Paschke
+- **abs:** https://arxiv.org/abs/2601.03121 · **PDF:** https://arxiv.org/pdf/2601.03121v1
+
+## 23820. Green MLOps: Closed-Loop, Energy-Aware Inference with NVIDIA Triton, FastAPI, and Bio-Inspired Thresholding
+
+- **ID:** `2601.04250` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Mustapha Hamdi, Mourad Jabou
+- **abs:** https://arxiv.org/abs/2601.04250 · **PDF:** https://arxiv.org/pdf/2601.04250v1
+
+## 23821. A framework for assuring the accuracy and fidelity of an AI-enabled Digital Twin of en route UK airspace
+
+- **ID:** `2601.03120` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Adam Keane, Nick Pepper, Chris Burr, et al.
+- **abs:** https://arxiv.org/abs/2601.03120 · **PDF:** https://arxiv.org/pdf/2601.03120v1
+
+## 23822. Transformers self-organize like newborn visual systems when trained in prenatal worlds
+
+- **ID:** `2601.03117` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-06
+- **Authors:** Lalit Pandey, Samantha M. W. Wood, Justin N. Wood
+- **abs:** https://arxiv.org/abs/2601.03117 · **PDF:** https://arxiv.org/pdf/2601.03117v1
+
+## 23823. Extreme-value forest fire prediction A study of the Loss Function in an Ordinality Scheme
+
+- **ID:** `2601.03327` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Nicolas Caron, Christophe Guyeux, Hassan Noura, et al.
+- **abs:** https://arxiv.org/abs/2601.03327 · **PDF:** https://arxiv.org/pdf/2601.03327v2
+
+## 23824. Discovering and Causally Validating Emotion-Sensitive Neurons in Large Audio-Language Models
+
+- **ID:** `2601.03115` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Xiutian Zhao, Björn Schuller, Berrak Sisman
+- **abs:** https://arxiv.org/abs/2601.03115 · **PDF:** https://arxiv.org/pdf/2601.03115v1
+
+## 23825. One Sample to Rule Them All: Extreme Data Efficiency in Multidiscipline Reasoning with Reinforcement Learning
+
+- **ID:** `2601.03111` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Yiyuan Li, Zhen Huang, Yanan Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.03111 · **PDF:** https://arxiv.org/pdf/2601.03111v2
+
+## 23826. Who Laughs with Whom? Disentangling Influential Factors in Humor Preferences across User Clusters and LLMs
+
+- **ID:** `2601.03103` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Soichiro Murakami, Hidetaka Kamigaito, Hiroya Takamura, et al.
+- **abs:** https://arxiv.org/abs/2601.03103 · **PDF:** https://arxiv.org/pdf/2601.03103v1
+
+## 23827. Text-Guided Layer Fusion Mitigates Hallucination in Multimodal LLMs
+
+- **ID:** `2601.03100` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Chenchen Lin, Sanbao Su, Rachel Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.03100 · **PDF:** https://arxiv.org/pdf/2601.03100v2
+
+## 23828. Time-Aware Synthetic Control
+
+- **ID:** `2601.03099` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Saeyoung Rho, Cyrus Illick, Samhitha Narasipura, et al.
+- **abs:** https://arxiv.org/abs/2601.03099 · **PDF:** https://arxiv.org/pdf/2601.03099v1
+
+## 23829. From Muscle to Text with MyoText: sEMG to Text via Finger Classification and Transformer-Based Decoding
+
+- **ID:** `2601.03098` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Meghna Roy Chowdhury, Shreyas Sen, Yi Ding
+- **abs:** https://arxiv.org/abs/2601.03098 · **PDF:** https://arxiv.org/pdf/2601.03098v1
+
+## 23830. ATLAS: Adaptive Test-Time Latent Steering with External Verifiers for Enhancing LLMs Reasoning
+
+- **ID:** `2601.03093` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Tuc Nguyen, Thai Le
+- **abs:** https://arxiv.org/abs/2601.03093 · **PDF:** https://arxiv.org/pdf/2601.03093v1
+
+## 23831. Higher order PCA-like rotation-invariant features for detailed shape descriptors modulo rotation
+
+- **ID:** `2601.03326` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Jarek Duda
+- **abs:** https://arxiv.org/abs/2601.03326 · **PDF:** https://arxiv.org/pdf/2601.03326v1
+
+## 23832. Grad-ELLM: Gradient-based Explanations for Decoder-only LLMs
+
+- **ID:** `2601.03089` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Xin Huang, Antoni B. Chan
+- **abs:** https://arxiv.org/abs/2601.03089 · **PDF:** https://arxiv.org/pdf/2601.03089v1
+
+## 23833. Audit Me If You Can: Query-Efficient Active Fairness Auditing of Black-Box LLMs
+
+- **ID:** `2601.03087` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** David Hartmann, Lena Pohlmann, Lelia Hanslik, et al.
+- **abs:** https://arxiv.org/abs/2601.03087 · **PDF:** https://arxiv.org/pdf/2601.03087v1
+
+## 23834. Real-Time Adaptive Anomaly Detection in Industrial IoT Environments
+
+- **ID:** `2601.03085` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Mahsa Raeiszadeh, Amin Ebrahimzadeh, Roch H. Glitho, et al.
+- **abs:** https://arxiv.org/abs/2601.03085 · **PDF:** https://arxiv.org/pdf/2601.03085v1
+
+## 23835. M-RAG: Making RAG Faster, Stronger, and More Efficient
+
+- **ID:** `2603.26667` · **primary:** `cs.IR` · **Submitted:** 2026-01-06
+- **Authors:** Sun Xu, Tongkai Xu, Baiheng Xie, et al.
+- **abs:** https://arxiv.org/abs/2603.26667 · **PDF:** https://arxiv.org/pdf/2603.26667v1
+
+## 23836. On the Identifiability of Regime-Switching Models with Multi-Lag Dependencies
+
+- **ID:** `2601.03325` · **primary:** `stat.ML` · **Submitted:** 2026-01-06
+- **Authors:** Carles Balsells-Rodas, Toshiko Matsui, Pedro A. M. Mediano, et al.
+- **abs:** https://arxiv.org/abs/2601.03325 · **PDF:** https://arxiv.org/pdf/2601.03325v1
+
+## 23837. Enhancing Moral Diagnosis and Correction in Large Language Models
+
+- **ID:** `2601.03079` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Bocheng Chen, Xi Chen, Han Zi, et al.
+- **abs:** https://arxiv.org/abs/2601.03079 · **PDF:** https://arxiv.org/pdf/2601.03079v3
+
+## 23838. Bare-Metal Tensor Virtualization: Overcoming the Memory Wall in Edge-AI Inference on ARM64
+
+- **ID:** `2601.03324` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Bugra Kilictas, Faruk Alpay
+- **abs:** https://arxiv.org/abs/2601.03324 · **PDF:** https://arxiv.org/pdf/2601.03324v1
+
+## 23839. Listen to Rhythm, Choose Movements: Autoregressive Multimodal Dance Generation via Diffusion and Mamba with Decoupled Dance Dataset
+
+- **ID:** `2601.03323` · **primary:** `cs.GR` · **Submitted:** 2026-01-06
+- **Authors:** Oran Duan, Yinghua Shen, Yingzhu Lv, et al.
+- **abs:** https://arxiv.org/abs/2601.03323 · **PDF:** https://arxiv.org/pdf/2601.03323v2
+
+## 23840. Joint Encoding of KV-Cache Blocks for Scalable LLM Serving
+
+- **ID:** `2601.03067` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Joseph Kampeas, Emir Haleva
+- **abs:** https://arxiv.org/abs/2601.03067 · **PDF:** https://arxiv.org/pdf/2601.03067v1
+
+## 23841. Do LLMs Encode Functional Importance of Reasoning Tokens?
+
+- **ID:** `2601.03066` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Janvijay Singh, Dilek Hakkani-Tür
+- **abs:** https://arxiv.org/abs/2601.03066 · **PDF:** https://arxiv.org/pdf/2601.03066v1
+
+## 23842. Explainable Fuzzy GNNs for Leak Detection in Water Distribution Networks
+
+- **ID:** `2601.03062` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Qusai Khaled, Pasquale De Marinis, Moez Louati, et al.
+- **abs:** https://arxiv.org/abs/2601.03062 · **PDF:** https://arxiv.org/pdf/2601.03062v1
+
+## 23843. IBISAgent: Reinforcing Pixel-Level Visual Reasoning in MLLMs for Universal Biomedical Object Referring and Segmentation
+
+- **ID:** `2601.03054` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Yankai Jiang, Qiaoru Li, Binlu Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.03054 · **PDF:** https://arxiv.org/pdf/2601.03054v3
+
+## 23844. Detecting Hallucinations in Retrieval-Augmented Generation via Semantic-level Internal Reasoning Graph
+
+- **ID:** `2601.03052` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Jianpeng Hu, Yanzeng Li, Jialun Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.03052 · **PDF:** https://arxiv.org/pdf/2601.03052v1
+
+## 23845. Temporal Graph Network: Hallucination Detection in Multi-Turn Conversation
+
+- **ID:** `2601.03051` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Vidhi Rathore, Sambu Aneesh, Himanshu Singh
+- **abs:** https://arxiv.org/abs/2601.03051 · **PDF:** https://arxiv.org/pdf/2601.03051v1
+
+## 23846. On the Intrinsic Limits of Transformer Image Embeddings in Non-Solvable Spatial Reasoning
+
+- **ID:** `2601.03048` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Siyi Lyu, Quan Liu, Feng Yan
+- **abs:** https://arxiv.org/abs/2601.03048 · **PDF:** https://arxiv.org/pdf/2601.03048v1
+
+## 23847. HEEGNet: Hyperbolic Embeddings for EEG
+
+- **ID:** `2601.03322` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Shanglin Li, Shiwen Chu, Okan Koç, et al.
+- **abs:** https://arxiv.org/abs/2601.03322 · **PDF:** https://arxiv.org/pdf/2601.03322v2
+
+## 23848. When the Coffee Feature Activates on Coffins: An Analysis of Feature Extraction and Steering for Mechanistic Interpretability
+
+- **ID:** `2601.03047` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Raphael Ronge, Markus Maier, Frederick Eberhardt
+- **abs:** https://arxiv.org/abs/2601.03047 · **PDF:** https://arxiv.org/pdf/2601.03047v1
+
+## 23849. Motion Blur Robust Wheat Pest Damage Detection with Dynamic Fuzzy Feature Fusion
+
+- **ID:** `2601.03046` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Han Zhang, Yanwei Wang, Fang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03046 · **PDF:** https://arxiv.org/pdf/2601.03046v1
+
+## 23850. Lil: Less is Less When Applying Post-Training Sparse-Attention Algorithms in Long-Decode Stage
+
+- **ID:** `2601.03043` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Junhao Hu, Fangze Li, Mingtao Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.03043 · **PDF:** https://arxiv.org/pdf/2601.03043v2
+
+## 23851. BaseCal: Unsupervised Confidence Calibration via Base Model Signals
+
+- **ID:** `2601.03042` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Hexiang Tan, Wanli Yang, Junwei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.03042 · **PDF:** https://arxiv.org/pdf/2601.03042v2
+
+## 23852. PiDR: Physics-Informed Inertial Dead Reckoning for Autonomous Platforms
+
+- **ID:** `2601.03040` · **primary:** `cs.RO` · **Submitted:** 2026-01-06
+- **Authors:** Arup Kumar Sahoo, Itzik Klein
+- **abs:** https://arxiv.org/abs/2601.03040 · **PDF:** https://arxiv.org/pdf/2601.03040v1
+
+## 23853. Aligning Findings with Diagnosis: A Self-Consistent Reinforcement Learning Framework for Trustworthy Radiology Reporting
+
+- **ID:** `2601.03321` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Kun Zhao, Siyuan Dai, Pan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.03321 · **PDF:** https://arxiv.org/pdf/2601.03321v2
+
+## 23854. Validating Generalist Robots with Situation Calculus and STL Falsification
+
+- **ID:** `2601.03038` · **primary:** `cs.RO` · **Submitted:** 2026-01-06
+- **Authors:** Changwen Li, Rongjie Yan, Chih-Hong Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.03038 · **PDF:** https://arxiv.org/pdf/2601.03038v1
+
+## 23855. NorwAI's Large Language Models: Technical Report
+
+- **ID:** `2601.03034` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Jon Atle Gulla, Peng Liu, Lemei Zhang
+- **abs:** https://arxiv.org/abs/2601.03034 · **PDF:** https://arxiv.org/pdf/2601.03034v2
+
+## 23856. Causal Manifold Fairness: Enforcing Geometric Invariance in Representation Learning
+
+- **ID:** `2601.03032` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Vidhi Rathore
+- **abs:** https://arxiv.org/abs/2601.03032 · **PDF:** https://arxiv.org/pdf/2601.03032v1
+
+## 23857. Flow Matching and Diffusion Models via PointNet for Generating Fluid Fields on Irregular Geometries
+
+- **ID:** `2601.03030` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Ali Kashefi
+- **abs:** https://arxiv.org/abs/2601.03030 · **PDF:** https://arxiv.org/pdf/2601.03030v2
+
+## 23858. Ratio-Variance Regularized Policy Optimization for Efficient LLM Fine-tuning
+
+- **ID:** `2601.03320` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Yu Luo, Shuo Han, Yihan Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.03320 · **PDF:** https://arxiv.org/pdf/2601.03320v1
+
+## 23859. Reducing Hallucinations in LLMs via Factuality-Aware Preference Learning
+
+- **ID:** `2601.03027` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Sindhuja Chaduvula, Ahmed Y. Radwan, Azib Farooq, et al.
+- **abs:** https://arxiv.org/abs/2601.03027 · **PDF:** https://arxiv.org/pdf/2601.03027v2
+
+## 23860. LittiChoQA: Literary Texts in Indic Languages Chosen for Question Answering
+
+- **ID:** `2601.03025` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Aarya Khandelwal, Ritwik Mishra, Rajiv Ratn Shah
+- **abs:** https://arxiv.org/abs/2601.03025 · **PDF:** https://arxiv.org/pdf/2601.03025v1
+
+## 23861. MedDialogRubrics: A Comprehensive Benchmark and Evaluation Framework for Multi-turn Medical Consultations in Large Language Models
+
+- **ID:** `2601.03023` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Lecheng Gong, Weimin Fang, Ting Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.03023 · **PDF:** https://arxiv.org/pdf/2601.03023v2
+
+## 23862. CaricatureGS: Exaggerating 3D Gaussian Splatting Faces With Gaussian Curvature
+
+- **ID:** `2601.03319` · **primary:** `cs.GR` · **Submitted:** 2026-01-06
+- **Authors:** Eldad Matmon, Amit Bracha, Noam Rotstein, et al.
+- **abs:** https://arxiv.org/abs/2601.03319 · **PDF:** https://arxiv.org/pdf/2601.03319v1
+
+## 23863. DNACHUNKER: Learnable Tokenization for DNA Language Models
+
+- **ID:** `2601.03019` · **primary:** `q-bio.GN` · **Submitted:** 2026-01-06
+- **Authors:** Taewon Kim, Jihwan Shin, Hyomin Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.03019 · **PDF:** https://arxiv.org/pdf/2601.03019v3
+
+## 23864. Dementia-R1: Reinforced Pretraining and Reasoning from Unstructured Clinical Notes for Real-World Dementia Prognosis
+
+- **ID:** `2601.03018` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Choonghan Kim, Hyunmin Hwang, Hangeol Chang, et al.
+- **abs:** https://arxiv.org/abs/2601.03018 · **PDF:** https://arxiv.org/pdf/2601.03018v2
+
+## 23865. MMFormalizer: Multimodal Autoformalization in the Wild
+
+- **ID:** `2601.03017` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Jing Xiong, Qi Han, Yunta Hsieh, et al.
+- **abs:** https://arxiv.org/abs/2601.03017 · **PDF:** https://arxiv.org/pdf/2601.03017v1
+
+## 23866. In-Context Reinforcement Learning through Bayesian Fusion of Context and Value Prior
+
+- **ID:** `2601.03015` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Anaïs Berkes, Vincent Taboga, Donna Vakalis, et al.
+- **abs:** https://arxiv.org/abs/2601.03015 · **PDF:** https://arxiv.org/pdf/2601.03015v1
+
+## 23867. SentGraph: Hierarchical Sentence Graph for Multi-hop Retrieval-Augmented Question Answering
+
+- **ID:** `2601.03014` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Junli Liang, Pengfei Zhou, Wangqiu Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.03014 · **PDF:** https://arxiv.org/pdf/2601.03014v2
+
+## 23868. Deep Learning-Based Image Recognition for Soft-Shell Shrimp Classification
+
+- **ID:** `2601.03317` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Yun-Hao Zhang, I-Hsien Ting, Dario Liberona, et al.
+- **abs:** https://arxiv.org/abs/2601.03317 · **PDF:** https://arxiv.org/pdf/2601.03317v1
+
+## 23869. JPU: Bridging Jailbreak Defense and Unlearning via On-Policy Path Rectification
+
+- **ID:** `2601.03005` · **primary:** `cs.CR` · **Submitted:** 2026-01-06
+- **Authors:** Xi Wang, Songlei Jian, Shasha Li, et al.
+- **abs:** https://arxiv.org/abs/2601.03005 · **PDF:** https://arxiv.org/pdf/2601.03005v1
+
+## 23870. Multi-Distribution Robust Conformal Prediction
+
+- **ID:** `2601.02998` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Yuqi Yang, Ying Jin
+- **abs:** https://arxiv.org/abs/2601.02998 · **PDF:** https://arxiv.org/pdf/2601.02998v1
+
+## 23871. Why LLMs Aren't Scientists Yet: Lessons from Four Autonomous Research Attempts
+
+- **ID:** `2601.03315` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Dhruv Trehan, Paras Chopra
+- **abs:** https://arxiv.org/abs/2601.03315 · **PDF:** https://arxiv.org/pdf/2601.03315v1
+
+## 23872. From Memorization to Creativity: LLM as a Designer of Novel Neural-Architectures
+
+- **ID:** `2601.02997` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Waleed Khalid, Dmitry Ignatov, Radu Timofte
+- **abs:** https://arxiv.org/abs/2601.02997 · **PDF:** https://arxiv.org/pdf/2601.02997v1
+
+## 23873. Large Reasoning Models Are (Not Yet) Multilingual Latent Reasoners
+
+- **ID:** `2601.02996` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Yihong Liu, Raoyuan Zhao, Hinrich Schütze, et al.
+- **abs:** https://arxiv.org/abs/2601.02996 · **PDF:** https://arxiv.org/pdf/2601.02996v1
+
+## 23874. Learning to Act Robustly with View-Invariant Latent Actions
+
+- **ID:** `2601.02994` · **primary:** `cs.RO` · **Submitted:** 2026-01-06
+- **Authors:** Youngjoon Jeong, Junha Chun, Taesup Kim
+- **abs:** https://arxiv.org/abs/2601.02994 · **PDF:** https://arxiv.org/pdf/2601.02994v1
+
+## 23875. Stable-RAG: Mitigating Retrieval-Permutation-Induced Hallucinations in Retrieval-Augmented Generation
+
+- **ID:** `2601.02993` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Qianchi Zhang, Hainan Zhang, Liang Pang, et al.
+- **abs:** https://arxiv.org/abs/2601.02993 · **PDF:** https://arxiv.org/pdf/2601.02993v3
+
+## 23876. Towards Faithful Reasoning in Comics for Small MLLMs
+
+- **ID:** `2601.02991` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Chengcheng Feng, Haojie Yin, Yucheng Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.02991 · **PDF:** https://arxiv.org/pdf/2601.02991v2
+
+## 23877. Mechanistic Interpretability of Large-Scale Counting in LLMs through a System-2 Strategy
+
+- **ID:** `2601.02989` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Hosein Hasani, Mohammadali Banayeeanzade, Ali Nafisi, et al.
+- **abs:** https://arxiv.org/abs/2601.02989 · **PDF:** https://arxiv.org/pdf/2601.02989v1
+
+## 23878. ULS+: Data-driven Model Adaptation Enhances Lesion Segmentation
+
+- **ID:** `2601.02988` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Rianne Weber, Niels Rocholl, Max de Grauw, et al.
+- **abs:** https://arxiv.org/abs/2601.02988 · **PDF:** https://arxiv.org/pdf/2601.02988v1
+
+## 23879. LAMS-Edit: Latent and Attention Mixing with Schedulers for Improved Content Preservation in Diffusion-Based Image and Style Editing
+
+- **ID:** `2601.02987` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Wingwa Fu, Takayuki Okatani
+- **abs:** https://arxiv.org/abs/2601.02987 · **PDF:** https://arxiv.org/pdf/2601.02987v1
+
+## 23880. P-Check: Advancing Personalized Reward Model via Learning to Generate Dynamic Checklist
+
+- **ID:** `2601.02986` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Kwangwook Seo, Dongha Lee
+- **abs:** https://arxiv.org/abs/2601.02986 · **PDF:** https://arxiv.org/pdf/2601.02986v1
+
+## 23881. Fuzzy Representation of Norms
+
+- **ID:** `2601.04249` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Ziba Assadi, Paola Inverardi
+- **abs:** https://arxiv.org/abs/2601.04249 · **PDF:** https://arxiv.org/pdf/2601.04249v1
+
+## 23882. Interpretable All-Type Audio Deepfake Detection with Audio LLMs via Frequency-Time Reinforcement Learning
+
+- **ID:** `2601.02983` · **primary:** `cs.SD` · **Submitted:** 2026-01-06
+- **Authors:** Yuankun Xie, Xiaoxuan Guo, Jiayi Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.02983 · **PDF:** https://arxiv.org/pdf/2601.02983v1
+
+## 23883. Generalizable Prompt Tuning for Audio-Language Models via Semantic Expansion
+
+- **ID:** `2601.20867` · **primary:** `cs.SD` · **Submitted:** 2026-01-06
+- **Authors:** Jaehyuk Jang, Wonjun Lee, Kangwook Ko, et al.
+- **abs:** https://arxiv.org/abs/2601.20867 · **PDF:** https://arxiv.org/pdf/2601.20867v1
+
+## 23884. Millimeter-Wave Gesture Recognition in ISAC: Does Reducing Sensing Airtime Hamper Accuracy?
+
+- **ID:** `2601.10733` · **primary:** `eess.SP` · **Submitted:** 2026-01-06
+- **Authors:** Jakob Struye, Nabeel Nisar Bhat, Siddhartha Kumar, et al.
+- **abs:** https://arxiv.org/abs/2601.10733 · **PDF:** https://arxiv.org/pdf/2601.10733v1
+
+## 23885. Mechanistic Knobs in LLMs: Retrieving and Steering High-Order Semantic Features via Sparse Autoencoders
+
+- **ID:** `2601.02978` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Ruikang Zhang, Shuo Wang, Qi Su
+- **abs:** https://arxiv.org/abs/2601.02978 · **PDF:** https://arxiv.org/pdf/2601.02978v1
+
+## 23886. Correct, Concise and Complete: Multi-stage Training For Adaptive Reasoning
+
+- **ID:** `2601.02972` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Nathanaël Carraz Rakotonirina, Ren Pang, Neha Anna John, et al.
+- **abs:** https://arxiv.org/abs/2601.02972 · **PDF:** https://arxiv.org/pdf/2601.02972v1
+
+## 23887. Reliability-Aware Adaptive Self-Consistency for Efficient Sampling in LLM Reasoning
+
+- **ID:** `2601.02970` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Junseok Kim, Nakyeong Yang, Kyungmin Min, et al.
+- **abs:** https://arxiv.org/abs/2601.02970 · **PDF:** https://arxiv.org/pdf/2601.02970v1
+
+## 23888. Rationale-Grounded In-Context Learning for Time Series Reasoning with Multimodal Large Language Models
+
+- **ID:** `2601.02968` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Qingxiang Liu, Zhiqing Cui, Xiaoliang Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.02968 · **PDF:** https://arxiv.org/pdf/2601.02968v1
+
+## 23889. MoE Adapter for Large Audio Language Models: Sparsity, Disentanglement, and Gradient-Conflict-Free
+
+- **ID:** `2601.02967` · **primary:** `cs.SD` · **Submitted:** 2026-01-06
+- **Authors:** Yishu Lei, Shuwei He, Jing Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.02967 · **PDF:** https://arxiv.org/pdf/2601.02967v2
+
+## 23890. Low-Resource Heuristics for Bahnaric Optical Character Recognition Improvement
+
+- **ID:** `2601.02965` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Phat Tran, Phuoc Pham, Hung Trinh, et al.
+- **abs:** https://arxiv.org/abs/2601.02965 · **PDF:** https://arxiv.org/pdf/2601.02965v1
+
+## 23891. LLM-Augmented Changepoint Detection: A Framework for Ensemble Detection and Automated Explanation
+
+- **ID:** `2601.02957` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Fabian Lukassen, Christoph Weisser, Michael Schlee, et al.
+- **abs:** https://arxiv.org/abs/2601.02957 · **PDF:** https://arxiv.org/pdf/2601.02957v3
+
+## 23892. Enhancing Multilingual RAG Systems with Debiased Language Preference-Guided Query Fusion
+
+- **ID:** `2601.02956` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Jeonghyun Park, Byeongjeong Kim, Seojin Hwang, et al.
+- **abs:** https://arxiv.org/abs/2601.02956 · **PDF:** https://arxiv.org/pdf/2601.02956v1
+
+## 23893. The World is Not Mono: Enabling Spatial Understanding in Large Audio-Language Models
+
+- **ID:** `2601.02954` · **primary:** `cs.SD` · **Submitted:** 2026-01-06
+- **Authors:** Yuhuan You, Lai Wei, Xihong Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.02954 · **PDF:** https://arxiv.org/pdf/2601.02954v2
+
+## 23894. Batch-of-Thought: Cross-Instance Learning for Enhanced LLM Reasoning
+
+- **ID:** `2601.02950` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Xuan Yang, Furong Jia, Roy Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.02950 · **PDF:** https://arxiv.org/pdf/2601.02950v2
+
+## 23895. MixTTE: Multi-Level Mixture-of-Experts for Scalable and Adaptive Travel Time Estimation
+
+- **ID:** `2601.02943` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Wenzhao Jiang, Jindong Han, Ruiqian Han, et al.
+- **abs:** https://arxiv.org/abs/2601.02943 · **PDF:** https://arxiv.org/pdf/2601.02943v1
+
+## 23896. SastBench: A Benchmark for Testing Agentic SAST Triage
+
+- **ID:** `2601.02941` · **primary:** `cs.CR` · **Submitted:** 2026-01-06
+- **Authors:** Jake Feiglin, Guy Dar
+- **abs:** https://arxiv.org/abs/2601.02941 · **PDF:** https://arxiv.org/pdf/2601.02941v1
+
+## 23897. Pearmut: Human Evaluation of Translation Made Trivial
+
+- **ID:** `2601.02933` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Vilém Zouhar, Tom Kocmi
+- **abs:** https://arxiv.org/abs/2601.02933 · **PDF:** https://arxiv.org/pdf/2601.02933v2
+
+## 23898. Memorization, Emergence, and Explaining Reversal Failures: A Controlled Study of Relational Semantics in LLMs
+
+- **ID:** `2601.02931` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Yihua Zhu, Qianying Liu, Jiaxin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.02931 · **PDF:** https://arxiv.org/pdf/2601.02931v1
+
+## 23899. PsyAgent: Constructing Human-like Agents Based on Psychological Modeling and Contextual Interaction
+
+- **ID:** `2601.06158` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Zibin Meng, Kani Chen
+- **abs:** https://arxiv.org/abs/2601.06158 · **PDF:** https://arxiv.org/pdf/2601.06158v2
+
+## 23900. PrismVAU: Prompt-Refined Inference System for Multimodal Video Anomaly Understanding
+
+- **ID:** `2601.02927` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Iñaki Erregue, Kamal Nasrollahi, Sergio Escalera
+- **abs:** https://arxiv.org/abs/2601.02927 · **PDF:** https://arxiv.org/pdf/2601.02927v2
+
+## 23901. DCG ReID: Disentangling Collaboration and Guidance Fusion Representations for Multi-modal Vehicle Re-Identification
+
+- **ID:** `2601.02924` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Aihua Zheng, Ya Gao, Shihao Li, et al.
+- **abs:** https://arxiv.org/abs/2601.02924 · **PDF:** https://arxiv.org/pdf/2601.02924v1
+
+## 23902. RAL2M: Retrieval Augmented Learning-To-Match Against Hallucination in Compliance-Guaranteed Service Systems
+
+- **ID:** `2601.02917` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Mengze Hong, Di Jiang, Jiangtao Wen, et al.
+- **abs:** https://arxiv.org/abs/2601.02917 · **PDF:** https://arxiv.org/pdf/2601.02917v1
+
+## 23903. ChemBART: A Pre-trained BART Model Assisting Organic Chemistry Analysis
+
+- **ID:** `2601.02915` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Kenan Li, Yijian Zhang, Jin Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.02915 · **PDF:** https://arxiv.org/pdf/2601.02915v1
+
+## 23904. Image, Word and Thought: A More Challenging Language Task for the Iterated Learning Model
+
+- **ID:** `2601.02911` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Hyoyeon Lee, Seth Bullock, Conor Houghton
+- **abs:** https://arxiv.org/abs/2601.02911 · **PDF:** https://arxiv.org/pdf/2601.02911v1
+
+## 23905. From Intuition to Calibrated Judgment: A Rubric-Based Expert-Panel Study of Human Detection of LLM-Generated Korean Text
+
+- **ID:** `2601.19913` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Shinwoo Park, Yo-Sub Han
+- **abs:** https://arxiv.org/abs/2601.19913 · **PDF:** https://arxiv.org/pdf/2601.19913v3
+
+## 23906. Beyond the Black Box: A Survey on the Theory and Mechanism of Large Language Models
+
+- **ID:** `2601.02907` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Zeyu Gan, Ruifeng Ren, Wei Yao, et al.
+- **abs:** https://arxiv.org/abs/2601.02907 · **PDF:** https://arxiv.org/pdf/2601.02907v2
+
+## 23907. TA-Prompting: Enhancing Video Large Language Models for Dense Video Captioning via Temporal Anchors
+
+- **ID:** `2601.02908` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Wei-Yuan Cheng, Kai-Po Chang, Chi-Pin Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.02908 · **PDF:** https://arxiv.org/pdf/2601.02908v1
+
+## 23908. Linear Script Representations in Speech Foundation Models Enable Zero-Shot Transliteration
+
+- **ID:** `2601.02906` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Ryan Soh-Eun Shim, Kwanghee Choi, Kalvin Chang, et al.
+- **abs:** https://arxiv.org/abs/2601.02906 · **PDF:** https://arxiv.org/pdf/2601.02906v1
+
+## 23909. LOST-3DSG: Lightweight Open-Vocabulary 3D Scene Graphs with Semantic Tracking in Dynamic Environments
+
+- **ID:** `2601.02905` · **primary:** `cs.RO` · **Submitted:** 2026-01-06
+- **Authors:** Sara Micol Ferraina, Michele Brienza, Francesco Argenziano, et al.
+- **abs:** https://arxiv.org/abs/2601.02905 · **PDF:** https://arxiv.org/pdf/2601.02905v2
+
+## 23910. Logical Phase Transitions: Understanding Collapse in LLM Logical Reasoning
+
+- **ID:** `2601.02902` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Xinglang Zhang, Yunyao Zhang, ZeLiang Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.02902 · **PDF:** https://arxiv.org/pdf/2601.02902v1
+
+## 23911. Bridging Mechanistic Interpretability and Prompt Engineering with Gradient Ascent for Interpretable Persona Control
+
+- **ID:** `2601.02896` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Harshvardhan Saini, Yiming Tang, Dianbo Liu
+- **abs:** https://arxiv.org/abs/2601.02896 · **PDF:** https://arxiv.org/pdf/2601.02896v1
+
+## 23912. Transparent Semantic Change Detection with Dependency-Based Profiles
+
+- **ID:** `2601.02891` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Bach Phan-Tat, Kris Heylen, Dirk Geeraerts, et al.
+- **abs:** https://arxiv.org/abs/2601.02891 · **PDF:** https://arxiv.org/pdf/2601.02891v3
+
+## 23913. Enhanced 3D Gravity Inversion Using ResU-Net with Density Logging Constraints: A Dual-Phase Training Approach
+
+- **ID:** `2601.02890` · **primary:** `physics.geo-ph` · **Submitted:** 2026-01-06
+- **Authors:** Siyuan Dong, Jinghuai Gao, Shuai Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.02890 · **PDF:** https://arxiv.org/pdf/2601.02890v1
+
+## 23914. RPIQ: Residual-Projected Multi-Collaboration Closed-Loop and Single Instance Quantization for Visually Impaired Assistance
+
+- **ID:** `2601.02888` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Xuanyu Wang, Haisen Su, Jingtao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.02888 · **PDF:** https://arxiv.org/pdf/2601.02888v2
+
+## 23915. Domain Generalization for Time Series: Enhancing Drilling Regression Models for Stick-Slip Index Prediction
+
+- **ID:** `2601.02884` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Hana Yahia, Bruno Figliuzzi, Florent Di Meglio, et al.
+- **abs:** https://arxiv.org/abs/2601.02884 · **PDF:** https://arxiv.org/pdf/2601.02884v1
+
+## 23916. STIPP: Space-time in situ postprocessing over the French Alps using proper scoring rules
+
+- **ID:** `2601.02882` · **primary:** `physics.ao-ph` · **Submitted:** 2026-01-06
+- **Authors:** David Landry, Isabelle Gouttevin, Hugo Merizen, et al.
+- **abs:** https://arxiv.org/abs/2601.02882 · **PDF:** https://arxiv.org/pdf/2601.02882v1
+
+## 23917. ReTreVal: Reasoning Tree with Validation -- A Hybrid Framework for Enhanced LLM Multi-Step Reasoning
+
+- **ID:** `2601.02880` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Abhishek HS, Pavan C Shekar, Arpit Jain, et al.
+- **abs:** https://arxiv.org/abs/2601.02880 · **PDF:** https://arxiv.org/pdf/2601.02880v1
+
+## 23918. Revisiting Data Compression with Language Modeling
+
+- **ID:** `2601.02875` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Chen-Han Tsai
+- **abs:** https://arxiv.org/abs/2601.02875 · **PDF:** https://arxiv.org/pdf/2601.02875v1
+
+## 23919. LongBench Pro: A More Realistic and Comprehensive Bilingual Long-Context Evaluation Benchmark
+
+- **ID:** `2601.02872` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Ziyang Chen, Xing Wu, Junlong Jia, et al.
+- **abs:** https://arxiv.org/abs/2601.02872 · **PDF:** https://arxiv.org/pdf/2601.02872v1
+
+## 23920. SimRPD: Optimizing Recruitment Proactive Dialogue Agents through Simulator-Based Data Evaluation and Selection
+
+- **ID:** `2601.02871` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Zhiyong Cao, Dunqiang Liu, Qi Dai, et al.
+- **abs:** https://arxiv.org/abs/2601.02871 · **PDF:** https://arxiv.org/pdf/2601.02871v2
+
+## 23921. VLM4VLA: Revisiting Vision-Language-Models in Vision-Language-Action Models
+
+- **ID:** `2601.03309` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Jianke Zhang, Xiaoyu Chen, Qiuyue Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.03309 · **PDF:** https://arxiv.org/pdf/2601.03309v1
+
+## 23922. Quantum-Enhanced Neural Contextual Bandit Algorithms
+
+- **ID:** `2601.02870` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Yuqi Huang, Vincent Y. F Tan, Sharu Theresa Jose
+- **abs:** https://arxiv.org/abs/2601.02870 · **PDF:** https://arxiv.org/pdf/2601.02870v1
+
+## 23923. Training Language Models with homotokens Leads to Delayed Overfitting
+
+- **ID:** `2601.02867` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Adrian Cosma, Stefan Ruseti, Emilian Radoi, et al.
+- **abs:** https://arxiv.org/abs/2601.02867 · **PDF:** https://arxiv.org/pdf/2601.02867v2
+
+## 23924. To Generate or Discriminate? Methodological Considerations for Measuring Cultural Alignment in LLMs
+
+- **ID:** `2601.02858` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Saurabh Kumar Pandey, Sougata Saha, Monojit Choudhury
+- **abs:** https://arxiv.org/abs/2601.02858 · **PDF:** https://arxiv.org/pdf/2601.02858v1
+
+## 23925. Electricity Price Forecasting: Bridging Linear Models, Neural Networks and Online Learning
+
+- **ID:** `2601.02856` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Btissame El Mahtout, Florian Ziel
+- **abs:** https://arxiv.org/abs/2601.02856 · **PDF:** https://arxiv.org/pdf/2601.02856v3
+
+## 23926. M3MAD-Bench: Are Multi-Agent Debates Really Effective Across Domains and Modalities?
+
+- **ID:** `2601.02854` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Ao Li, Jinghui Zhang, Luyu Li, et al.
+- **abs:** https://arxiv.org/abs/2601.02854 · **PDF:** https://arxiv.org/pdf/2601.02854v1
+
+## 23927. Sample-Efficient Neurosymbolic Deep Reinforcement Learning
+
+- **ID:** `2601.02850` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Celeste Veronese, Daniele Meli, Alessandro Farinelli
+- **abs:** https://arxiv.org/abs/2601.02850 · **PDF:** https://arxiv.org/pdf/2601.02850v1
+
+## 23928. TiMem: Temporal-Hierarchical Memory Consolidation for Long-Horizon Conversational Agents
+
+- **ID:** `2601.02845` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Kai Li, Xuanqing Yu, Ziyi Ni, et al.
+- **abs:** https://arxiv.org/abs/2601.02845 · **PDF:** https://arxiv.org/pdf/2601.02845v1
+
+## 23929. Breaking Self-Attention Failure: Rethinking Query Initialization for Infrared Small Target Detection
+
+- **ID:** `2601.02837` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Yuteng Liu, Duanni Meng, Maoxun Yuan, et al.
+- **abs:** https://arxiv.org/abs/2601.02837 · **PDF:** https://arxiv.org/pdf/2601.02837v1
+
+## 23930. The performances of the Chinese and U.S. Large Language Models on the Topic of Chinese Culture
+
+- **ID:** `2601.02830` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Feiyan Liu, Siyan Zhao, Chenxun Zhuo, et al.
+- **abs:** https://arxiv.org/abs/2601.02830 · **PDF:** https://arxiv.org/pdf/2601.02830v2
+
+## 23931. Beyond Immediate Activation: Temporally Decoupled Backdoor Attacks on Time Series Forecasting
+
+- **ID:** `2601.04247` · **primary:** `cs.CR` · **Submitted:** 2026-01-06
+- **Authors:** Zhixin Liu, Xuanlin Liu, Sihan Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.04247 · **PDF:** https://arxiv.org/pdf/2601.04247v1
+
+## 23932. Punctuation-aware Hybrid Trainable Sparse Attention for Large Language Models
+
+- **ID:** `2601.02819` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Junxiang Qiu, Shuo Wang, Zhengsu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.02819 · **PDF:** https://arxiv.org/pdf/2601.02819v1
+
+## 23933. Quantum-enhanced long short-term memory with attention for spatial permeability prediction in oilfield reservoirs
+
+- **ID:** `2601.02818` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Muzhen Zhang, Yujie Cheng, Zhanxiang Lei
+- **abs:** https://arxiv.org/abs/2601.02818 · **PDF:** https://arxiv.org/pdf/2601.02818v2
+
+## 23934. Mastering the Game of Go with Self-play Experience Replay
+
+- **ID:** `2601.03306` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Jingbin Liu, Xuechun Wang
+- **abs:** https://arxiv.org/abs/2601.03306 · **PDF:** https://arxiv.org/pdf/2601.03306v1
+
+## 23935. Causal-Enhanced AI Agents for Medical Research Screening
+
+- **ID:** `2601.02814` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Duc Ngo, Arya Rahgoza
+- **abs:** https://arxiv.org/abs/2601.02814 · **PDF:** https://arxiv.org/pdf/2601.02814v1
+
+## 23936. Mass Concept Erasure in Diffusion Models with Concept Hierarchy
+
+- **ID:** `2601.03305` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Jiahang Tu, Ye Li, Yiming Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.03305 · **PDF:** https://arxiv.org/pdf/2601.03305v1
+
+## 23937. HAL: Inducing Human-likeness in LLMs with Alignment
+
+- **ID:** `2601.02813` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Masum Hasan, Junjie Zhao, Ehsan Hoque
+- **abs:** https://arxiv.org/abs/2601.02813 · **PDF:** https://arxiv.org/pdf/2601.02813v2
+
+## 23938. COFFEE: COdesign Framework for Feature Enriched Embeddings in Ads-Ranking Systems
+
+- **ID:** `2601.02807` · **primary:** `cs.IR` · **Submitted:** 2026-01-06
+- **Authors:** Sohini Roychowdhury, Doris Wang, Qian Ge, et al.
+- **abs:** https://arxiv.org/abs/2601.02807 · **PDF:** https://arxiv.org/pdf/2601.02807v1
+
+## 23939. ECLIPTICA -- A Framework for Switchable LLM Alignment via CITA - Contrastive Instruction-Tuned Alignment
+
+- **ID:** `2601.06157` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Kapil Wanaskar, Gaytri Jena, Vinija Jain, et al.
+- **abs:** https://arxiv.org/abs/2601.06157 · **PDF:** https://arxiv.org/pdf/2601.06157v1
+
+## 23940. Channel Knowledge Map Construction via Guided Flow Matching
+
+- **ID:** `2601.06156` · **primary:** `cs.IT` · **Submitted:** 2026-01-06
+- **Authors:** Ziyu Huang, Yong Zeng, Shen Fu, et al.
+- **abs:** https://arxiv.org/abs/2601.06156 · **PDF:** https://arxiv.org/pdf/2601.06156v1
+
+## 23941. Stratified Hazard Sampling: Minimal-Variance Event Scheduling for CTMC/DTMC Discrete Diffusion and Flow Models
+
+- **ID:** `2601.02799` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Seunghwan Jang, SooJean Han
+- **abs:** https://arxiv.org/abs/2601.02799 · **PDF:** https://arxiv.org/pdf/2601.02799v2
+
+## 23942. RadioDiff-Flux: Efficient Radio Map Construction via Generative Denoise Diffusion Model Trajectory Midpoint Reuse
+
+- **ID:** `2601.02790` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Xiucheng Wang, Peilin Zheng, Honggang Jia, et al.
+- **abs:** https://arxiv.org/abs/2601.02790 · **PDF:** https://arxiv.org/pdf/2601.02790v1
+
+## 23943. MiMo-V2-Flash Technical Report
+
+- **ID:** `2601.02780` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Xiaomi LLM-Core Team, :, Bangjun Xiao, et al.
+- **abs:** https://arxiv.org/abs/2601.02780 · **PDF:** https://arxiv.org/pdf/2601.02780v2
+
+## 23944. Closing the Reality Gap: Zero-Shot Sim-to-Real Deployment for Dexterous Force-Based Grasping and Manipulation
+
+- **ID:** `2601.02778` · **primary:** `cs.RO` · **Submitted:** 2026-01-06
+- **Authors:** Zhe Zhao, Haoyu Dong, Zhengmao He, et al.
+- **abs:** https://arxiv.org/abs/2601.02778 · **PDF:** https://arxiv.org/pdf/2601.02778v2
+
+## 23945. UniSRCodec: Unified and Low-Bitrate Single Codebook Codec with Sub-Band Reconstruction
+
+- **ID:** `2601.02776` · **primary:** `cs.SD` · **Submitted:** 2026-01-06
+- **Authors:** Zhisheng Zhang, Xiang Li, Yixuan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.02776 · **PDF:** https://arxiv.org/pdf/2601.02776v1
+
+## 23946. Fast Conformal Prediction using Conditional Interquantile Intervals
+
+- **ID:** `2601.02769` · **primary:** `stat.ML` · **Submitted:** 2026-01-06
+- **Authors:** Naixin Guo, Rui Luo, Zhixin Zhou
+- **abs:** https://arxiv.org/abs/2601.02769 · **PDF:** https://arxiv.org/pdf/2601.02769v1
+
+## 23947. Netflix Artwork Personalization via LLM Post-training
+
+- **ID:** `2601.02764` · **primary:** `cs.IR` · **Submitted:** 2026-01-06
+- **Authors:** Hyunji Nam, Sejoon Oh, Emma Kong, et al.
+- **abs:** https://arxiv.org/abs/2601.02764 · **PDF:** https://arxiv.org/pdf/2601.02764v1
+
+## 23948. LLM Agent Framework for Intelligent Change Analysis in Urban Environment using Remote Sensing Imagery
+
+- **ID:** `2601.02757` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Zixuan Xiao, Jun Ma
+- **abs:** https://arxiv.org/abs/2601.02757 · **PDF:** https://arxiv.org/pdf/2601.02757v1
+
+## 23949. Q-Regularized Generative Auto-Bidding: From Suboptimal Trajectories to Optimal Policies
+
+- **ID:** `2601.02754` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Mingming Zhang, Na Li, Zhuang Feiqing, et al.
+- **abs:** https://arxiv.org/abs/2601.02754 · **PDF:** https://arxiv.org/pdf/2601.02754v2
+
+## 23950. EComStage: Stage-wise and Orientation-specific Benchmarking for Large Language Models in E-commerce
+
+- **ID:** `2601.02752` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Kaiyan Zhao, Zijie Meng, Zheyong Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.02752 · **PDF:** https://arxiv.org/pdf/2601.02752v1
+
+## 23951. Window-based Membership Inference Attacks Against Fine-tuned Large Language Models
+
+- **ID:** `2601.02751` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Yuetian Chen, Yuntao Du, Kaiyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.02751 · **PDF:** https://arxiv.org/pdf/2601.02751v2
+
+## 23952. The Path Ahead for Agentic AI: Challenges and Opportunities
+
+- **ID:** `2601.02749` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Nadia Sibai, Yara Ahmed, Serry Sibaee, et al.
+- **abs:** https://arxiv.org/abs/2601.02749 · **PDF:** https://arxiv.org/pdf/2601.02749v1
+
+## 23953. BotSim: Mitigating The Formation Of Conspiratorial Societies with Useful Bots
+
+- **ID:** `2601.06154` · **primary:** `cs.CY` · **Submitted:** 2026-01-06
+- **Authors:** Lynnette Hui Xian Ng, Kathleen M. Carley
+- **abs:** https://arxiv.org/abs/2601.06154 · **PDF:** https://arxiv.org/pdf/2601.06154v1
+
+## 23954. SYNAPSE: Empowering LLM Agents with Episodic-Semantic Memory via Spreading Activation
+
+- **ID:** `2601.02744` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Hanqi Jiang, Junhao Chen, Yi Pan, et al.
+- **abs:** https://arxiv.org/abs/2601.02744 · **PDF:** https://arxiv.org/pdf/2601.02744v3
+
+## 23955. Finder: A Multimodal AI-Powered Search Framework for Pharmaceutical Data Retrieval
+
+- **ID:** `2603.15623` · **primary:** `cs.IR` · **Submitted:** 2026-01-06
+- **Authors:** Suyash Mishra, Srikanth Patil, Satyanarayan Pati, et al.
+- **abs:** https://arxiv.org/abs/2603.15623 · **PDF:** https://arxiv.org/pdf/2603.15623v1
+
+## 23956. L-PRISMA: An Extension of PRISMA in the Era of Generative Artificial Intelligence (GenAI)
+
+- **ID:** `2603.19236` · **primary:** `cs.DL` · **Submitted:** 2026-01-06
+- **Authors:** Samar Shailendra, Rajan Kadel, Aakanksha Sharma, et al.
+- **abs:** https://arxiv.org/abs/2603.19236 · **PDF:** https://arxiv.org/pdf/2603.19236v1
+
+## 23957. Language Hierarchization Provides the Optimal Solution to Human Working Memory Limits
+
+- **ID:** `2601.02740` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Luyao Chen, Weibo Gao, Junjie Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.02740 · **PDF:** https://arxiv.org/pdf/2601.02740v1
+
+## 23958. Mitigating Prompt-Induced Hallucinations in Large Language Models via Structured Reasoning
+
+- **ID:** `2601.02739` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Jinbo Hao, Kai Yang, Qingzhen Su, et al.
+- **abs:** https://arxiv.org/abs/2601.02739 · **PDF:** https://arxiv.org/pdf/2601.02739v1
+
+## 23959. Hypothesize-Then-Verify: Speculative Root Cause Analysis for Microservices with Pathwise Parallelism
+
+- **ID:** `2601.02736` · **primary:** `cs.SE` · **Submitted:** 2026-01-06
+- **Authors:** Lingzhe Zhang, Tong Jia, Yunpeng Zhai, et al.
+- **abs:** https://arxiv.org/abs/2601.02736 · **PDF:** https://arxiv.org/pdf/2601.02736v1
+
+## 23960. Scalable Tree Ensemble Proximities in Python
+
+- **ID:** `2601.02735` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Adrien Aumon, Guy Wolf, Kevin R. Moon, et al.
+- **abs:** https://arxiv.org/abs/2601.02735 · **PDF:** https://arxiv.org/pdf/2601.02735v1
+
+## 23961. Agentic Memory Enhanced Recursive Reasoning for Root Cause Localization in Microservices
+
+- **ID:** `2601.02732` · **primary:** `cs.SE` · **Submitted:** 2026-01-06
+- **Authors:** Lingzhe Zhang, Tong Jia, Yunpeng Zhai, et al.
+- **abs:** https://arxiv.org/abs/2601.02732 · **PDF:** https://arxiv.org/pdf/2601.02732v1
+
+## 23962. CRoPE: Efficient Parametrization of Rotary Positional Embedding
+
+- **ID:** `2601.02728` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Beicheng Lou, Zifei Xu, Vivian W. H. Wong
+- **abs:** https://arxiv.org/abs/2601.02728 · **PDF:** https://arxiv.org/pdf/2601.02728v2
+
+## 23963. Foreground-Aware Dataset Distillation via Dynamic Patch Selection
+
+- **ID:** `2601.02727` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Longzhen Li, Guang Li, Ren Togo, et al.
+- **abs:** https://arxiv.org/abs/2601.02727 · **PDF:** https://arxiv.org/pdf/2601.02727v1
+
+## 23964. Interoperability in AI Safety Governance: Ethics, Regulations, and Standards
+
+- **ID:** `2601.06153` · **primary:** `cs.CY` · **Submitted:** 2026-01-06
+- **Authors:** Yik Chan Chin, David A. Raho, Hag-Min Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.06153 · **PDF:** https://arxiv.org/pdf/2601.06153v1
+
+## 23965. Sampling non-log-concave densities via Hessian-free high-resolution dynamics
+
+- **ID:** `2601.02725` · **primary:** `math.PR` · **Submitted:** 2026-01-06
+- **Authors:** Xiaoyu Wang, Yingli Wang, Lingjiong Zhu
+- **abs:** https://arxiv.org/abs/2601.02725 · **PDF:** https://arxiv.org/pdf/2601.02725v1
+
+## 23966. Privacy-Preserving AI-Enabled Decentralized Learning and Employment Records System
+
+- **ID:** `2601.02720` · **primary:** `cs.CR` · **Submitted:** 2026-01-06
+- **Authors:** Yuqiao Xu, Mina Namazi, Sahith Reddy Jalapally, et al.
+- **abs:** https://arxiv.org/abs/2601.02720 · **PDF:** https://arxiv.org/pdf/2601.02720v1
+
+## 23967. AI-Driven Cybersecurity Threats: A Survey of Emerging Risks and Defensive Strategies
+
+- **ID:** `2601.03304` · **primary:** `cs.CR` · **Submitted:** 2026-01-06
+- **Authors:** Sai Teja Erukude, Viswa Chaitanya Marella, Suhasnadh Reddy Veluru
+- **abs:** https://arxiv.org/abs/2601.03304 · **PDF:** https://arxiv.org/pdf/2601.03304v1
+
+## 23968. HiMeS: Hippocampus-inspired Memory System for Personalized AI Assistants
+
+- **ID:** `2601.06152` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Hailong Li, Feifei Li, Wenhui Que, et al.
+- **abs:** https://arxiv.org/abs/2601.06152 · **PDF:** https://arxiv.org/pdf/2601.06152v1
+
+## 23969. Time-Scaling Is What Agents Need Now
+
+- **ID:** `2601.02714` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Zhi Liu, Guangzhi Wang
+- **abs:** https://arxiv.org/abs/2601.02714 · **PDF:** https://arxiv.org/pdf/2601.02714v1
+
+## 23970. CREAM: Continual Retrieval on Dynamic Streaming Corpora with Adaptive Soft Memory
+
+- **ID:** `2601.02708` · **primary:** `cs.IR` · **Submitted:** 2026-01-06
+- **Authors:** HuiJeong Son, Hyeongu Kang, Sunho Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.02708 · **PDF:** https://arxiv.org/pdf/2601.02708v2
+
+## 23971. Scaling Laws of Machine Learning for Optimal Power Flow
+
+- **ID:** `2601.02706` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Xinyi Liu, Xuan He, Yize Chen
+- **abs:** https://arxiv.org/abs/2601.02706 · **PDF:** https://arxiv.org/pdf/2601.02706v1
+
+## 23972. MultiSessionCollab: Learning User Preferences with Memory to Improve Long-Term Collaboration
+
+- **ID:** `2601.02702` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Shuhaib Mehri, Priyanka Kargupta, Tal August, et al.
+- **abs:** https://arxiv.org/abs/2601.02702 · **PDF:** https://arxiv.org/pdf/2601.02702v3
+
+## 23973. Adversarial Question Answering Robustness: A Multi-Level Error Analysis and Mitigation Study
+
+- **ID:** `2601.02700` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Agniv Roy Choudhury, Vignesh Ponselvan Rajasingh
+- **abs:** https://arxiv.org/abs/2601.02700 · **PDF:** https://arxiv.org/pdf/2601.02700v1
+
+## 23974. Boosting Accuracy and Interpretability in Multilingual Hate Speech Detection Through Layer Freezing and Explainable AI
+
+- **ID:** `2601.02697` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Meysam Shirdel Bilehsavar, Negin Mahmoudi, Mohammad Jalili Torkamani, et al.
+- **abs:** https://arxiv.org/abs/2601.02697 · **PDF:** https://arxiv.org/pdf/2601.02697v1
+
+## 23975. EvoRoute: Experience-Driven Self-Routing LLM Agent Systems
+
+- **ID:** `2601.02695` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Guibin Zhang, Haiyang Yu, Kaiming Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.02695 · **PDF:** https://arxiv.org/pdf/2601.02695v1
+
+## 23976. LAUDE: LLM-Assisted Unit Test Generation and Debugging of Hardware DEsigns
+
+- **ID:** `2601.08856` · **primary:** `cs.SE` · **Submitted:** 2026-01-06
+- **Authors:** Deeksha Nandal, Riccardo Revalor, Soham Dan, et al.
+- **abs:** https://arxiv.org/abs/2601.08856 · **PDF:** https://arxiv.org/pdf/2601.08856v2
+
+## 23977. Which Deep Learner? A Systematic Evaluation of Advanced Deep Forecasting Models Accuracy and Efficiency for Network Traffic Prediction
+
+- **ID:** `2601.02694` · **primary:** `cs.NI` · **Submitted:** 2026-01-06
+- **Authors:** Eilaf MA Babai, Aalaa MA Babai, Koji Okamura
+- **abs:** https://arxiv.org/abs/2601.02694 · **PDF:** https://arxiv.org/pdf/2601.02694v1
+
+## 23978. PromptPort: A Reliability Layer for Cross-Model Structured Extraction
+
+- **ID:** `2601.06151` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Varun Kotte
+- **abs:** https://arxiv.org/abs/2601.06151 · **PDF:** https://arxiv.org/pdf/2601.06151v1
+
+## 23979. Multi-channel multi-speaker transformer for speech recognition
+
+- **ID:** `2601.02688` · **primary:** `cs.SD` · **Submitted:** 2026-01-06
+- **Authors:** Guo Yifan, Tian Yao, Suo Hongbin, et al.
+- **abs:** https://arxiv.org/abs/2601.02688 · **PDF:** https://arxiv.org/pdf/2601.02688v1
+
+## 23980. CageDroneRF: A Large-Scale RF Benchmark and Toolkit for Drone Perception
+
+- **ID:** `2601.03302` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Mohammad Rostami, Atik Faysal, Hongtao Xia, et al.
+- **abs:** https://arxiv.org/abs/2601.03302 · **PDF:** https://arxiv.org/pdf/2601.03302v2
+
+## 23981. Learning from Prompt itself: the Hierarchical Attribution Prompt Optimization
+
+- **ID:** `2601.02683` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Dongyu Chen, Jian Ma, Xianpeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.02683 · **PDF:** https://arxiv.org/pdf/2601.02683v1
+
+## 23982. Topology-Independent Robustness of the Weighted Mean under Label Poisoning Attacks in Heterogeneous Decentralized Learning
+
+- **ID:** `2601.02682` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Jie Peng, Weiyu Li, Stefan Vlaski, et al.
+- **abs:** https://arxiv.org/abs/2601.02682 · **PDF:** https://arxiv.org/pdf/2601.02682v1
+
+## 23983. Adversarial Contrastive Learning for LLM Quantization Attacks
+
+- **ID:** `2601.02680` · **primary:** `cs.CR` · **Submitted:** 2026-01-06
+- **Authors:** Dinghong Song, Zhiwei Xu, Hai Wan, et al.
+- **abs:** https://arxiv.org/abs/2601.02680 · **PDF:** https://arxiv.org/pdf/2601.02680v1
+
+## 23984. Uni-FinLLM: A Unified Multimodal Large Language Model with Modular Task Heads for Micro-Level Stock Prediction and Macro-Level Systemic Risk Assessment
+
+- **ID:** `2601.02677` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Gongao Zhang, Haijiang Zeng, Lu Jiang
+- **abs:** https://arxiv.org/abs/2601.02677 · **PDF:** https://arxiv.org/pdf/2601.02677v1
+
+## 23985. AI Agents as Policymakers in Simulated Epidemics
+
+- **ID:** `2601.04245` · **primary:** `cs.MA` · **Submitted:** 2026-01-06
+- **Authors:** Goshi Aoki, Navid Ghaffarzadegan
+- **abs:** https://arxiv.org/abs/2601.04245 · **PDF:** https://arxiv.org/pdf/2601.04245v1
+
+## 23986. PC2P: Multi-Agent Path Finding via Personalized-Enhanced Communication and Crowd Perception
+
+- **ID:** `2601.03301` · **primary:** `cs.MA` · **Submitted:** 2026-01-06
+- **Authors:** Guotao Li, Shaoyun Xu, Yuexing Hao, et al.
+- **abs:** https://arxiv.org/abs/2601.03301 · **PDF:** https://arxiv.org/pdf/2601.03301v1
+
+## 23987. Iterative Structured Pruning for Large Language Models with Multi-Domain Calibration
+
+- **ID:** `2601.02674` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Guangxin Wu, Hao Zhang, Zhang Zhibin, et al.
+- **abs:** https://arxiv.org/abs/2601.02674 · **PDF:** https://arxiv.org/pdf/2601.02674v1
+
+## 23988. TRYLOCK: Defense-in-Depth Against LLM Jailbreaks via Layered Preference and Representation Engineering
+
+- **ID:** `2601.03300` · **primary:** `cs.CR` · **Submitted:** 2026-01-06
+- **Authors:** Scott Thornton
+- **abs:** https://arxiv.org/abs/2601.03300 · **PDF:** https://arxiv.org/pdf/2601.03300v1
+
+## 23989. Extracting books from production language models
+
+- **ID:** `2601.02671` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Ahmed Ahmed, A. Feder Cooper, Sanmi Koyejo, et al.
+- **abs:** https://arxiv.org/abs/2601.02671 · **PDF:** https://arxiv.org/pdf/2601.02671v1
+
+## 23990. Multi-Turn Jailbreaking of Aligned LLMs via Lexical Anchor Tree Search
+
+- **ID:** `2601.02670` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Devang Kulshreshtha, Hang Su, Chinmay Hegde, et al.
+- **abs:** https://arxiv.org/abs/2601.02670 · **PDF:** https://arxiv.org/pdf/2601.02670v1
+
+## 23991. Towards Comprehensive Stage-wise Benchmarking of Large Language Models in Fact-Checking
+
+- **ID:** `2601.02669` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Hongzhan Lin, Zixin Chen, Zhiqi Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.02669 · **PDF:** https://arxiv.org/pdf/2601.02669v1
+
+## 23992. MAFS: Multi-head Attention Feature Selection for High-Dimensional Data via Deep Fusion of Filter Methods
+
+- **ID:** `2601.02668` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Xiaoyan Sun, Qingyu Meng, Yalu Wen
+- **abs:** https://arxiv.org/abs/2601.02668 · **PDF:** https://arxiv.org/pdf/2601.02668v1
+
+## 23993. Inferring Causal Graph Temporal Logic Formulas to Expedite Reinforcement Learning in Temporally Extended Tasks
+
+- **ID:** `2601.02666` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Hadi Partovi Aria, Zhe Xu
+- **abs:** https://arxiv.org/abs/2601.02666 · **PDF:** https://arxiv.org/pdf/2601.02666v1
+
+## 23994. When Do Tools and Planning Help Large Language Models Think? A Cost- and Latency-Aware Benchmark
+
+- **ID:** `2601.02663` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Subha Ghoshal, Ali Al-Bustami
+- **abs:** https://arxiv.org/abs/2601.02663 · **PDF:** https://arxiv.org/pdf/2601.02663v2
+
+## 23995. When Prompting Meets Spiking: Graph Sparse Prompting via Spiking Graph Prompt Learning
+
+- **ID:** `2601.02662` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Bo Jiang, Weijun Zhao, Beibei Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.02662 · **PDF:** https://arxiv.org/pdf/2601.02662v1
+
+## 23996. Empirical Comparison of Encoder-Based Language Models and Feature-Based Supervised Machine Learning Approaches to Automated Scoring of Long Essays
+
+- **ID:** `2601.02659` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Kuo Wang, Haowei Hua, Pengfei Yan, et al.
+- **abs:** https://arxiv.org/abs/2601.02659 · **PDF:** https://arxiv.org/pdf/2601.02659v2
+
+## 23997. Statistical Inference for Fuzzy Clustering
+
+- **ID:** `2601.02656` · **primary:** `stat.ME` · **Submitted:** 2026-01-06
+- **Authors:** Qiuyi Wu, Zihan Zhu, Anru R. Zhang
+- **abs:** https://arxiv.org/abs/2601.02656 · **PDF:** https://arxiv.org/pdf/2601.02656v1
+
+## 23998. Integrating Multi-Agent Simulation, Behavioral Forensics, and Trust-Aware Machine Learning for Adaptive Insider Threat Detection
+
+- **ID:** `2601.04243` · **primary:** `cs.CR` · **Submitted:** 2026-01-06
+- **Authors:** Firdous Kausar, Asmah Muallem, Naw Safrin Sattar, et al.
+- **abs:** https://arxiv.org/abs/2601.04243 · **PDF:** https://arxiv.org/pdf/2601.04243v1
+
+## 23999. Effective Online 3D Bin Packing with Lookahead Parcels Using Monte Carlo Tree Search
+
+- **ID:** `2601.02649` · **primary:** `cs.RO` · **Submitted:** 2026-01-06
+- **Authors:** Jiangyi Fang, Bowen Zhou, Haotian Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.02649 · **PDF:** https://arxiv.org/pdf/2601.02649v1
+
+## 24000. Prioritized Replay for RL Post-training
+
+- **ID:** `2601.02648` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Mehdi Fatemi
+- **abs:** https://arxiv.org/abs/2601.02648 · **PDF:** https://arxiv.org/pdf/2601.02648v1
+
+## 24001. DreamLoop: Controllable Cinemagraph Generation from a Single Photograph
+
+- **ID:** `2601.02646` · **primary:** `cs.CV` · **Submitted:** 2026-01-06
+- **Authors:** Aniruddha Mahapatra, Long Mai, Cusuh Ham, et al.
+- **abs:** https://arxiv.org/abs/2601.02646 · **PDF:** https://arxiv.org/pdf/2601.02646v1
+
+## 24002. AWARE-US: Preference-Aware Infeasibility Resolution in Tool-Calling Agents
+
+- **ID:** `2601.02643` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Mehmet Kurmaz
+- **abs:** https://arxiv.org/abs/2601.02643 · **PDF:** https://arxiv.org/pdf/2601.02643v2
+
+## 24003. An Empirical Study of On-Device Translation for Real-Time Live-Stream Chat on Mobile Devices
+
+- **ID:** `2601.02641` · **primary:** `cs.AI` · **Submitted:** 2026-01-06
+- **Authors:** Jeiyoon Park, Daehwan Lee, Changmin Yeo, et al.
+- **abs:** https://arxiv.org/abs/2601.02641 · **PDF:** https://arxiv.org/pdf/2601.02641v1
+
+## 24004. Bayesian Multiple Multivariate Density-Density Regression
+
+- **ID:** `2601.02640` · **primary:** `stat.ME` · **Submitted:** 2026-01-06
+- **Authors:** Khai Nguyen, Yang Ni, Peter Mueller
+- **abs:** https://arxiv.org/abs/2601.02640 · **PDF:** https://arxiv.org/pdf/2601.02640v1
+
+## 24005. Credit Assignment via Neural Manifold Noise Correlation
+
+- **ID:** `2601.02636` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Byungwoo Kang, Maceo Richards, Bernardo Sabatini
+- **abs:** https://arxiv.org/abs/2601.02636 · **PDF:** https://arxiv.org/pdf/2601.02636v1
+
+## 24006. Embedding Retrofitting: Data Engineering for better RAG
+
+- **ID:** `2601.15298` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Anantha Sharma
+- **abs:** https://arxiv.org/abs/2601.15298 · **PDF:** https://arxiv.org/pdf/2601.15298v2
+
+## 24007. TAAF: A Trace Abstraction and Analysis Framework Synergizing Knowledge Graphs and LLMs
+
+- **ID:** `2601.02632` · **primary:** `cs.SE` · **Submitted:** 2026-01-06
+- **Authors:** Alireza Ezaz, Ghazal Khodabandeh, Majid Babaei, et al.
+- **abs:** https://arxiv.org/abs/2601.02632 · **PDF:** https://arxiv.org/pdf/2601.02632v1
+
+## 24008. 130k Lines of Formal Topology in Two Weeks: Simple and Cheap Autoformalization for Everyone?
+
+- **ID:** `2601.03298` · **primary:** `cs.LO` · **Submitted:** 2026-01-06
+- **Authors:** Josef Urban
+- **abs:** https://arxiv.org/abs/2601.03298 · **PDF:** https://arxiv.org/pdf/2601.03298v1
+
+## 24009. Improved Evidence Extraction for Document Inconsistency Detection with LLMs
+
+- **ID:** `2601.02627` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Nelvin Tan, Yaowen Zhang, James Asikin Cheung, et al.
+- **abs:** https://arxiv.org/abs/2601.02627 · **PDF:** https://arxiv.org/pdf/2601.02627v1
+
+## 24010. LAsset: An LLM-assisted Security Asset Identification Framework for System-on-Chip (SoC) Verification
+
+- **ID:** `2601.02624` · **primary:** `cs.CR` · **Submitted:** 2026-01-06
+- **Authors:** Md Ajoad Hasan, Dipayan Saha, Khan Thamid Hasan, et al.
+- **abs:** https://arxiv.org/abs/2601.02624 · **PDF:** https://arxiv.org/pdf/2601.02624v1
+
+## 24011. Hierarchical temporal receptive windows and zero-shot timescale generalization in biologically constrained scale-invariant deep networks
+
+- **ID:** `2601.02618` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-06
+- **Authors:** Aakash Sarkar, Marc W. Howard
+- **abs:** https://arxiv.org/abs/2601.02618 · **PDF:** https://arxiv.org/pdf/2601.02618v1
+
+## 24012. AfriEconQA: A Benchmark Dataset for African Economic Analysis based on World Bank Reports
+
+- **ID:** `2601.15297` · **primary:** `cs.CL` · **Submitted:** 2026-01-06
+- **Authors:** Edward Ajayi
+- **abs:** https://arxiv.org/abs/2601.15297 · **PDF:** https://arxiv.org/pdf/2601.15297v2
+
+## 24013. Conformal novelty detection with false discovery rate control at the boundary
+
+- **ID:** `2601.02610` · **primary:** `stat.ME` · **Submitted:** 2026-01-06
+- **Authors:** Zijun Gao, Etienne Roquain, Daniel Xiang
+- **abs:** https://arxiv.org/abs/2601.02610 · **PDF:** https://arxiv.org/pdf/2601.02610v2
+
+## 24014. Chronicals: A High-Performance Framework for LLM Fine-Tuning with 3.51x Speedup over Unsloth
+
+- **ID:** `2601.02609` · **primary:** `cs.LG` · **Submitted:** 2026-01-06
+- **Authors:** Arjun S. Nair
+- **abs:** https://arxiv.org/abs/2601.02609 · **PDF:** https://arxiv.org/pdf/2601.02609v1
+
+## 24015. Scalable Construction of a Lung Cancer Knowledge Base: Profiling Semantic Reasoning in LLMs
+
+- **ID:** `2601.02604` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Cesar Felipe Martínez Cisneros, Jesús Ulises Quiroz Bautista, Claudia Anahí Guzmán Solano, et al.
+- **abs:** https://arxiv.org/abs/2601.02604 · **PDF:** https://arxiv.org/pdf/2601.02604v1
+
+## 24016. SWaRL: Safeguard Code Watermarking via Reinforcement Learning
+
+- **ID:** `2601.02602` · **primary:** `cs.CR` · **Submitted:** 2026-01-05
+- **Authors:** Neusha Javidnia, Ruisi Zhang, Ashish Kundu, et al.
+- **abs:** https://arxiv.org/abs/2601.02602 · **PDF:** https://arxiv.org/pdf/2601.02602v1
+
+## 24017. LongDA: Benchmarking LLM Agents for Long-Document Data Analysis
+
+- **ID:** `2601.02598` · **primary:** `cs.DL` · **Submitted:** 2026-01-05
+- **Authors:** Yiyang Li, Zheyuan Zhang, Tianyi Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.02598 · **PDF:** https://arxiv.org/pdf/2601.02598v2
+
+## 24018. Annealed Langevin Posterior Sampling (ALPS): A Rapid Algorithm for Image Restoration with Multiscale Energy Models
+
+- **ID:** `2601.02594` · **primary:** `eess.IV` · **Submitted:** 2026-01-05
+- **Authors:** Jyothi Rikhab Chand, Mathews Jacob
+- **abs:** https://arxiv.org/abs/2601.02594 · **PDF:** https://arxiv.org/pdf/2601.02594v1
+
+## 24019. FlowPlan-G2P: A Structured Generation Framework for Transforming Scientific Papers into Patent Descriptions
+
+- **ID:** `2601.02589` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Kris W Pan, Yongmin Yoo
+- **abs:** https://arxiv.org/abs/2601.02589 · **PDF:** https://arxiv.org/pdf/2601.02589v1
+
+## 24020. A Foundation Model Approach for Fetal Stress Prediction During Labor From cardiotocography (CTG) recordings
+
+- **ID:** `2601.06149` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Naomi Fridman, Berta Ben Shachar
+- **abs:** https://arxiv.org/abs/2601.06149 · **PDF:** https://arxiv.org/pdf/2601.06149v1
+
+## 24021. Threat Detection in Social Media Networks Using Machine Learning Based Network Analysis
+
+- **ID:** `2601.02581` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Aditi Sanjay Agrawal
+- **abs:** https://arxiv.org/abs/2601.02581 · **PDF:** https://arxiv.org/pdf/2601.02581v1
+
+## 24022. Reconstructing Item Characteristic Curves using Fine-Tuned Large Language Models
+
+- **ID:** `2601.02580` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Christopher Ormerod
+- **abs:** https://arxiv.org/abs/2601.02580 · **PDF:** https://arxiv.org/pdf/2601.02580v1
+
+## 24023. DataParasite Enables Scalable and Repurposable Online Data Curation
+
+- **ID:** `2601.02578` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Mengyi Sun
+- **abs:** https://arxiv.org/abs/2601.02578 · **PDF:** https://arxiv.org/pdf/2601.02578v1
+
+## 24024. Orchestral AI: A Framework for Agent Orchestration
+
+- **ID:** `2601.02577` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Alexander Roman, Jacob Roman
+- **abs:** https://arxiv.org/abs/2601.02577 · **PDF:** https://arxiv.org/pdf/2601.02577v1
+
+## 24025. Fact-Checking with Large Language Models via Probabilistic Certainty and Consistency
+
+- **ID:** `2601.02574` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Haoran Wang, Maryam Khalid, Qiong Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.02574 · **PDF:** https://arxiv.org/pdf/2601.02574v1
+
+## 24026. LendNova: Towards Automated Credit Risk Assessment with Language Models
+
+- **ID:** `2601.02573` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Kiarash Shamsi, Danijel Novokmet, Joshua Peters, et al.
+- **abs:** https://arxiv.org/abs/2601.02573 · **PDF:** https://arxiv.org/pdf/2601.02573v1
+
+## 24027. LoRA-Drop: Temporal LoRA Decoding for Efficient LLM Inference
+
+- **ID:** `2601.02569` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Hossein Rajabzadeh, Maryam Dialameh, Chul B. Park, et al.
+- **abs:** https://arxiv.org/abs/2601.02569 · **PDF:** https://arxiv.org/pdf/2601.02569v1
+
+## 24028. Compressed code: the hidden effects of quantization and distillation on programming tokens
+
+- **ID:** `2601.02563` · **primary:** `cs.SE` · **Submitted:** 2026-01-05
+- **Authors:** Viacheslav Siniaev, Iaroslav Chelombitko, Aleksey Komissarov
+- **abs:** https://arxiv.org/abs/2601.02563 · **PDF:** https://arxiv.org/pdf/2601.02563v3
+
+## 24029. CutisAI: Deep Learning Framework for Automated Dermatology and Cancer Screening
+
+- **ID:** `2601.02562` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Rohit Kaushik, Eva Kaushik
+- **abs:** https://arxiv.org/abs/2601.02562 · **PDF:** https://arxiv.org/pdf/2601.02562v1
+
+## 24030. LLM Flow Processes for Text-Conditioned Regression
+
+- **ID:** `2601.06147` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Felix Biggs, Samuel Willis
+- **abs:** https://arxiv.org/abs/2601.06147 · **PDF:** https://arxiv.org/pdf/2601.06147v1
+
+## 24031. AI-exposed jobs deteriorated before ChatGPT
+
+- **ID:** `2601.02554` · **primary:** `econ.GN` · **Submitted:** 2026-01-05
+- **Authors:** Morgan R. Frank, Alireza Javadian Sabet, Lisa Simon, et al.
+- **abs:** https://arxiv.org/abs/2601.02554 · **PDF:** https://arxiv.org/pdf/2601.02554v1
+
+## 24032. SimpleMem: Efficient Lifelong Memory for LLM Agents
+
+- **ID:** `2601.02553` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Jiaqi Liu, Yaofeng Su, Peng Xia, et al.
+- **abs:** https://arxiv.org/abs/2601.02553 · **PDF:** https://arxiv.org/pdf/2601.02553v3
+
+## 24033. Normalized Conditional Mutual Information Surrogate Loss for Deep Neural Classifiers
+
+- **ID:** `2601.02543` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Linfeng Ye, Zhixiang Chi, Konstantinos N. Plataniotis, et al.
+- **abs:** https://arxiv.org/abs/2601.02543 · **PDF:** https://arxiv.org/pdf/2601.02543v3
+
+## 24034. ModeX: Evaluator-Free Best-of-N Selection for Open-Ended Generation
+
+- **ID:** `2601.02535` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Hyeong Kyu Choi, Sharon Li
+- **abs:** https://arxiv.org/abs/2601.02535 · **PDF:** https://arxiv.org/pdf/2601.02535v1
+
+## 24035. Losses that Cook: Topological Optimal Transport for Structured Recipe Generation
+
+- **ID:** `2601.02531` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Mattia Ottoborgo, Daniele Rege Cambrin, Paolo Garza
+- **abs:** https://arxiv.org/abs/2601.02531 · **PDF:** https://arxiv.org/pdf/2601.02531v1
+
+## 24036. Multi-scale Graph Autoregressive Modeling: Molecular Property Prediction via Next Token Prediction
+
+- **ID:** `2601.02530` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Zhuoyang Jiang, Yaosen Min, Peiran Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.02530 · **PDF:** https://arxiv.org/pdf/2601.02530v3
+
+## 24037. First Provably Optimal Asynchronous SGD for Homogeneous and Heterogeneous Data
+
+- **ID:** `2601.02523` · **primary:** `math.OC` · **Submitted:** 2026-01-05
+- **Authors:** Artavazd Maranjyan
+- **abs:** https://arxiv.org/abs/2601.02523 · **PDF:** https://arxiv.org/pdf/2601.02523v1
+
+## 24038. On the Limits of Self-Improving in Large Language Models: The Singularity Is Not Near Without Symbolic Model Synthesis
+
+- **ID:** `2601.05280` · **primary:** `cs.IT` · **Submitted:** 2026-01-05
+- **Authors:** Hector Zenil
+- **abs:** https://arxiv.org/abs/2601.05280 · **PDF:** https://arxiv.org/pdf/2601.05280v2
+
+## 24039. Textual Explanations and Their Evaluations for Reinforcement Learning Policy
+
+- **ID:** `2601.02514` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Ahmad Terra, Mohit Ahmed, Rafia Inam, et al.
+- **abs:** https://arxiv.org/abs/2601.02514 · **PDF:** https://arxiv.org/pdf/2601.02514v1
+
+## 24040. MetagenBERT: a Transformer-based Architecture using Foundational genomic Large Language Models for novel Metagenome Representation
+
+- **ID:** `2601.03295` · **primary:** `q-bio.GN` · **Submitted:** 2026-01-05
+- **Authors:** Gaspar Roy, Eugeni Belda, Baptiste Hennecart, et al.
+- **abs:** https://arxiv.org/abs/2601.03295 · **PDF:** https://arxiv.org/pdf/2601.03295v1
+
+## 24041. LLM-Enhanced Reinforcement Learning for Time Series Anomaly Detection
+
+- **ID:** `2601.02511` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Bahareh Golchin, Banafsheh Rekabdar, Danielle Justo
+- **abs:** https://arxiv.org/abs/2601.02511 · **PDF:** https://arxiv.org/pdf/2601.02511v1
+
+## 24042. hdlib 2.0: Extending Machine Learning Capabilities of Vector-Symbolic Architectures
+
+- **ID:** `2601.02509` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Fabio Cumbo, Kabir Dhillon, Daniel Blankenberg
+- **abs:** https://arxiv.org/abs/2601.02509 · **PDF:** https://arxiv.org/pdf/2601.02509v1
+
+## 24043. Enhancing Debugging Skills with AI-Powered Assistance: A Real-Time Tool for Debugging Support
+
+- **ID:** `2601.02504` · **primary:** `cs.SE` · **Submitted:** 2026-01-05
+- **Authors:** Elizaveta Artser, Daniil Karol, Anna Potriasaeva, et al.
+- **abs:** https://arxiv.org/abs/2601.02504 · **PDF:** https://arxiv.org/pdf/2601.02504v1
+
+## 24044. GEM-Style Constraints for PEFT with Dual Gradient Projection in LoRA
+
+- **ID:** `2601.02500` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Brian Tekmen, Jason Yin, Qianqian Tong
+- **abs:** https://arxiv.org/abs/2601.02500 · **PDF:** https://arxiv.org/pdf/2601.02500v1
+
+## 24045. Polynomial Convergence of Riemannian Diffusion Models
+
+- **ID:** `2601.02499` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Xingyu Xu, Ziyi Zhang, Yorie Nakahira, et al.
+- **abs:** https://arxiv.org/abs/2601.02499 · **PDF:** https://arxiv.org/pdf/2601.02499v1
+
+## 24046. Variational (Energy-Based) Spectral Learning: A Machine Learning Framework for Solving Partial Differential Equations
+
+- **ID:** `2601.02492` · **primary:** `math.NA` · **Submitted:** 2026-01-05
+- **Authors:** M. M. Hammad
+- **abs:** https://arxiv.org/abs/2601.02492 · **PDF:** https://arxiv.org/pdf/2601.02492v1
+
+## 24047. Heterogeneous Low-Bandwidth Pre-Training of LLMs
+
+- **ID:** `2601.02360` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Yazan Obeidi, Amir Sarfi, Joel Lidin, et al.
+- **abs:** https://arxiv.org/abs/2601.02360 · **PDF:** https://arxiv.org/pdf/2601.02360v1
+
+## 24048. DARC: Drum accompaniment generation with fine-grained rhythm control
+
+- **ID:** `2601.02357` · **primary:** `cs.SD` · **Submitted:** 2026-01-05
+- **Authors:** Trey Brosnan
+- **abs:** https://arxiv.org/abs/2601.02357 · **PDF:** https://arxiv.org/pdf/2601.02357v1
+
+## 24049. Meta-Learning Guided Pruning for Few-Shot Plant Pathology on Edge Devices
+
+- **ID:** `2601.02353` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Shahnawaz Alam, Mohammed Mudassir Uddin, Mohammed Kaif Pasha
+- **abs:** https://arxiv.org/abs/2601.02353 · **PDF:** https://arxiv.org/pdf/2601.02353v2
+
+## 24050. Dynamic Quantization Error Propagation in Encoder-Decoder ASR Quantization
+
+- **ID:** `2601.02455` · **primary:** `cs.SD` · **Submitted:** 2026-01-05
+- **Authors:** Xinyu Wang, Yajie Luo, Yihong Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.02455 · **PDF:** https://arxiv.org/pdf/2601.02455v1
+
+## 24051. Falcon-H1R: Pushing the Reasoning Frontiers with a Hybrid Model for Efficient Test-Time Scaling
+
+- **ID:** `2601.02346` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Falcon LLM Team, Iheb Chaabane, Puneesh Khanna, et al.
+- **abs:** https://arxiv.org/abs/2601.02346 · **PDF:** https://arxiv.org/pdf/2601.02346v1
+
+## 24052. Robust Persona-Aware Toxicity Detection with Prompt Optimization and Learned Ensembling
+
+- **ID:** `2601.02337` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Berk Atil, Rebecca J. Passonneau, Ninareh Mehrabi
+- **abs:** https://arxiv.org/abs/2601.02337 · **PDF:** https://arxiv.org/pdf/2601.02337v1
+
+## 24053. The Rise of Agentic Testing: Multi-Agent Systems for Robust Software Quality Assurance
+
+- **ID:** `2601.02454` · **primary:** `cs.SE` · **Submitted:** 2026-01-05
+- **Authors:** Saba Naqvi, Mohammad Baqar, Nawaz Ali Mohammad
+- **abs:** https://arxiv.org/abs/2601.02454 · **PDF:** https://arxiv.org/pdf/2601.02454v1
+
+## 24054. Bridging the AI divide in sub-Saharan Africa: Challenges and opportunities for inclusivity
+
+- **ID:** `2601.06145` · **primary:** `cs.CY` · **Submitted:** 2026-01-05
+- **Authors:** Masike Malatji
+- **abs:** https://arxiv.org/abs/2601.06145 · **PDF:** https://arxiv.org/pdf/2601.06145v1
+
+## 24055. Hunting for "Oddballs" with Machine Learning: Detecting Anomalous Exoplanets Using a Deep-Learned Low-Dimensional Representation of Transit Spectra with Autoencoders
+
+- **ID:** `2601.02324` · **primary:** `astro-ph.EP` · **Submitted:** 2026-01-05
+- **Authors:** Alexander Roman, Emilie Panek, Roy T. Forestano, et al.
+- **abs:** https://arxiv.org/abs/2601.02324 · **PDF:** https://arxiv.org/pdf/2601.02324v1
+
+## 24056. Environment-Adaptive Covariate Selection: Learning When to Use Spurious Correlations for Out-of-Distribution Prediction
+
+- **ID:** `2601.02322` · **primary:** `stat.ME` · **Submitted:** 2026-01-05
+- **Authors:** Shuozhi Zuo, Yixin Wang
+- **abs:** https://arxiv.org/abs/2601.02322 · **PDF:** https://arxiv.org/pdf/2601.02322v1
+
+## 24057. Estimating Text Temperature with Language Models
+
+- **ID:** `2601.02320` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Nikolay Mikhaylovskiy
+- **abs:** https://arxiv.org/abs/2601.02320 · **PDF:** https://arxiv.org/pdf/2601.02320v2
+
+## 24058. DatBench: Discriminative, Faithful, and Efficient VLM Evaluations
+
+- **ID:** `2601.02316` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** DatologyAI, :, Siddharth Joshi, et al.
+- **abs:** https://arxiv.org/abs/2601.02316 · **PDF:** https://arxiv.org/pdf/2601.02316v2
+
+## 24059. Project Ariadne: A Structural Causal Framework for Auditing Faithfulness in LLM Agents
+
+- **ID:** `2601.02314` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Sourena Khanzadeh
+- **abs:** https://arxiv.org/abs/2601.02314 · **PDF:** https://arxiv.org/pdf/2601.02314v1
+
+## 24060. Game of Coding: Coding Theory in the Presence of Rational Adversaries, Motivated by Decentralized Machine Learning
+
+- **ID:** `2601.02313` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Hanzaleh Akbari Nodehi, Viveck R. Cadambe, Mohammad Ali Maddah-Ali
+- **abs:** https://arxiv.org/abs/2601.02313 · **PDF:** https://arxiv.org/pdf/2601.02313v1
+
+## 24061. The Patient/Industry Trade-off in Medical Artificial Intelligence
+
+- **ID:** `2601.06144` · **primary:** `cs.CY` · **Submitted:** 2026-01-05
+- **Authors:** Rina Khan, Annabelle Sauve, Imaan Bayoumi, et al.
+- **abs:** https://arxiv.org/abs/2601.06144 · **PDF:** https://arxiv.org/pdf/2601.06144v1
+
+## 24062. Placement Semantics for Distributed Deep Learning: A Systematic Framework for Analyzing Parallelism Strategies
+
+- **ID:** `2601.02311` · **primary:** `cs.DC` · **Submitted:** 2026-01-05
+- **Authors:** Deep Pankajbhai Mehta
+- **abs:** https://arxiv.org/abs/2601.02311 · **PDF:** https://arxiv.org/pdf/2601.02311v1
+
+## 24063. Domain Specific Specialization in Low-Resource Settings: The Efficacy of Offline Response-Based Knowledge Distillation in Large Language Models
+
+- **ID:** `2601.16219` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Erdem Aslan, Pakize Erdoğmuş
+- **abs:** https://arxiv.org/abs/2601.16219 · **PDF:** https://arxiv.org/pdf/2601.16219v1
+
+## 24064. Temporal Kolmogorov-Arnold Networks (T-KAN) for High-Frequency Limit Order Book Forecasting: Efficiency, Interpretability, and Alpha Decay
+
+- **ID:** `2601.02310` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Ahmad Makinde
+- **abs:** https://arxiv.org/abs/2601.02310 · **PDF:** https://arxiv.org/pdf/2601.02310v1
+
+## 24065. Differential Privacy for Transformer Embeddings of Text with Nonparametric Variational Information Bottleneck
+
+- **ID:** `2601.02307` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Dina El Zein, James Henderson
+- **abs:** https://arxiv.org/abs/2601.02307 · **PDF:** https://arxiv.org/pdf/2601.02307v2
+
+## 24066. Classifying several dialectal Nawatl varieties
+
+- **ID:** `2601.02303` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Juan-José Guzmán-Landa, Juan-Manuel Torres-Moreno, Miguel Figueroa-Saavedra, et al.
+- **abs:** https://arxiv.org/abs/2601.02303 · **PDF:** https://arxiv.org/pdf/2601.02303v1
+
+## 24067. Power-of-Two Quantization-Aware-Training (PoT-QAT) in Large Language Models (LLMs)
+
+- **ID:** `2601.02298` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Mahmoud Elgenedy
+- **abs:** https://arxiv.org/abs/2601.02298 · **PDF:** https://arxiv.org/pdf/2601.02298v1
+
+## 24068. mHC-GNN: Manifold-Constrained Hyper-Connections for Graph Neural Networks
+
+- **ID:** `2601.02451` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Subhankar Mishra
+- **abs:** https://arxiv.org/abs/2601.02451 · **PDF:** https://arxiv.org/pdf/2601.02451v1
+
+## 24069. pdfQA: Diverse, Challenging, and Realistic Question Answering over PDFs
+
+- **ID:** `2601.02285` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Tobias Schimanski, Imene Kolli, Yu Fan, et al.
+- **abs:** https://arxiv.org/abs/2601.02285 · **PDF:** https://arxiv.org/pdf/2601.02285v2
+
+## 24070. TopoLoRA-SAM: Topology-Aware Parameter-Efficient Adaptation of Foundation Segmenters for Thin-Structure and Cross-Domain Binary Semantic Segmentation
+
+- **ID:** `2601.02273` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Salim Khazem
+- **abs:** https://arxiv.org/abs/2601.02273 · **PDF:** https://arxiv.org/pdf/2601.02273v1
+
+## 24071. Predicting Early and Complete Drug Release from Long-Acting Injectables Using Explainable Machine Learning
+
+- **ID:** `2601.02265` · **primary:** `q-bio.BM` · **Submitted:** 2026-01-05
+- **Authors:** Karla N. Robles, Manar D. Samad
+- **abs:** https://arxiv.org/abs/2601.02265 · **PDF:** https://arxiv.org/pdf/2601.02265v1
+
+## 24072. POSEIDON: Physics-Optimized Seismic Energy Inference and Detection Operating Network
+
+- **ID:** `2601.02264` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Boris Kriuk, Fedor Kriuk
+- **abs:** https://arxiv.org/abs/2601.02264 · **PDF:** https://arxiv.org/pdf/2601.02264v1
+
+## 24073. Improved Accuracy for Private Continual Cardinality Estimation in Fully Dynamic Streams via Matrix Factorization
+
+- **ID:** `2601.02257` · **primary:** `cs.CR` · **Submitted:** 2026-01-05
+- **Authors:** Joel Daniel Andersson, Palak Jain, Satchit Sivakumar
+- **abs:** https://arxiv.org/abs/2601.02257 · **PDF:** https://arxiv.org/pdf/2601.02257v1
+
+## 24074. VAR RL Done Right: Tackling Asynchronous Policy Conflicts in Visual Autoregressive Generation
+
+- **ID:** `2601.02256` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Shikun Sun, Liao Qu, Huichao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.02256 · **PDF:** https://arxiv.org/pdf/2601.02256v1
+
+## 24075. Neuro-Channel Networks: A Multiplication-Free Architecture by Biological Signal Transmission
+
+- **ID:** `2601.02253` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Emrah Mete, Emin Erkan Korkmaz
+- **abs:** https://arxiv.org/abs/2601.02253 · **PDF:** https://arxiv.org/pdf/2601.02253v1
+
+## 24076. A Comparative Study of Custom CNNs, Pre-trained Models, and Transfer Learning Across Multiple Visual Datasets
+
+- **ID:** `2601.02246` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Annoor Sharara Akhand
+- **abs:** https://arxiv.org/abs/2601.02246 · **PDF:** https://arxiv.org/pdf/2601.02246v1
+
+## 24077. Emergent Complexity in Nuclear Reaction Networks: A Study of Stellar Nucleosynthesis through Chemical Organization Theory
+
+- **ID:** `2601.06143` · **primary:** `astro-ph.SR` · **Submitted:** 2026-01-05
+- **Authors:** Pedro Maldonado-Lang, Clément Vidal
+- **abs:** https://arxiv.org/abs/2601.06143 · **PDF:** https://arxiv.org/pdf/2601.06143v1
+
+## 24078. VIBE: Visual Instruction Based Editor
+
+- **ID:** `2601.02242` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Grigorii Alekseenko, Aleksandr Gordeev, Irina Tolstykh, et al.
+- **abs:** https://arxiv.org/abs/2601.02242 · **PDF:** https://arxiv.org/pdf/2601.02242v1
+
+## 24079. From Mice to Trains: Amortized Bayesian Inference on Graph Data
+
+- **ID:** `2601.02241` · **primary:** `stat.ML` · **Submitted:** 2026-01-05
+- **Authors:** Svenja Jedhoff, Elizaveta Semenova, Aura Raulo, et al.
+- **abs:** https://arxiv.org/abs/2601.02241 · **PDF:** https://arxiv.org/pdf/2601.02241v4
+
+## 24080. CD4LM: Consistency Distillation and aDaptive Decoding for Diffusion Language Models
+
+- **ID:** `2601.02236` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Yihao Liang, Ze Wang, Hao Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.02236 · **PDF:** https://arxiv.org/pdf/2601.02236v1
+
+## 24081. ELLA: Efficient Lifelong Learning for Adapters in Large Language Models
+
+- **ID:** `2601.02232` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Shristi Das Biswas, Yue Zhang, Anwesan Pal, et al.
+- **abs:** https://arxiv.org/abs/2601.02232 · **PDF:** https://arxiv.org/pdf/2601.02232v2
+
+## 24082. From XAI to Stories: A Factorial Study of LLM-Generated Explanation Quality
+
+- **ID:** `2601.02224` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Fabian Lukassen, Jan Herrmann, Christoph Weisser, et al.
+- **abs:** https://arxiv.org/abs/2601.02224 · **PDF:** https://arxiv.org/pdf/2601.02224v2
+
+## 24083. AgentMark: Utility-Preserving Behavioral Watermarking for Agents
+
+- **ID:** `2601.03294` · **primary:** `cs.CR` · **Submitted:** 2026-01-05
+- **Authors:** Kaibo Huang, Jin Tan, Yukun Wei, et al.
+- **abs:** https://arxiv.org/abs/2601.03294 · **PDF:** https://arxiv.org/pdf/2601.03294v1
+
+## 24084. LLM-Empowered Functional Safety and Security by Design in Automotive Systems
+
+- **ID:** `2601.02215` · **primary:** `cs.SE` · **Submitted:** 2026-01-05
+- **Authors:** Nenad Petrovic, Vahid Zolfaghari, Fengjunjie Pan, et al.
+- **abs:** https://arxiv.org/abs/2601.02215 · **PDF:** https://arxiv.org/pdf/2601.02215v1
+
+## 24085. Quantized SO(3)-Equivariant Graph Neural Networks for Efficient Molecular Property Prediction
+
+- **ID:** `2601.02213` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Haoyu Zhou, Ping Xue, Hao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.02213 · **PDF:** https://arxiv.org/pdf/2601.02213v2
+
+## 24086. ARCADE: A City-Scale Corpus for Fine-Grained Arabic Dialect Tagging
+
+- **ID:** `2601.02209` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Omer Nacar, Serry Sibaee, Adel Ammar, et al.
+- **abs:** https://arxiv.org/abs/2601.02209 · **PDF:** https://arxiv.org/pdf/2601.02209v1
+
+## 24087. Seeing the Unseen: Zooming in the Dark with Event Cameras
+
+- **ID:** `2601.02206` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Dachun Kai, Zeyu Xiao, Huyue Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.02206 · **PDF:** https://arxiv.org/pdf/2601.02206v1
+
+## 24088. NextFlow: Unified Sequential Modeling Activates Multimodal Understanding and Generation
+
+- **ID:** `2601.02204` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Huichao Zhang, Liao Qu, Yiheng Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.02204 · **PDF:** https://arxiv.org/pdf/2601.02204v1
+
+## 24089. CORE: Code-based Inverse Self-Training Framework with Graph Expansion for Virtual Agents
+
+- **ID:** `2601.02201` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Keyu Wang, Bingchen Miao, Wendong Bu, et al.
+- **abs:** https://arxiv.org/abs/2601.02201 · **PDF:** https://arxiv.org/pdf/2601.02201v1
+
+## 24090. Code for Machines, Not Just Humans: Quantifying AI-Friendliness with Code Health Metrics
+
+- **ID:** `2601.02200` · **primary:** `cs.SE` · **Submitted:** 2026-01-05
+- **Authors:** Markus Borg, Nadim Hagatulah, Adam Tornhill, et al.
+- **abs:** https://arxiv.org/abs/2601.02200 · **PDF:** https://arxiv.org/pdf/2601.02200v1
+
+## 24091. Is Sanskrit the most token-efficient language? A quantitative study using GPT, Gemini, and SentencePiece
+
+- **ID:** `2601.06142` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Anshul Kumar
+- **abs:** https://arxiv.org/abs/2601.06142 · **PDF:** https://arxiv.org/pdf/2601.06142v1
+
+## 24092. Mind the Gap: Continuous Magnification Sampling for Pathology Foundation Models
+
+- **ID:** `2601.02198` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Alexander Möllers, Julius Hense, Florian Schulz, et al.
+- **abs:** https://arxiv.org/abs/2601.02198 · **PDF:** https://arxiv.org/pdf/2601.02198v1
+
+## 24093. ACDZero: MCTS Agent for Mastering Automated Cyber Defense
+
+- **ID:** `2601.02196` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Yu Li, Sizhe Tang, Rongqian Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.02196 · **PDF:** https://arxiv.org/pdf/2601.02196v2
+
+## 24094. Learning with Monotone Adversarial Corruptions
+
+- **ID:** `2601.02193` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Kasper Green Larsen, Chirag Pabbaraju, Abhishek Shetty
+- **abs:** https://arxiv.org/abs/2601.02193 · **PDF:** https://arxiv.org/pdf/2601.02193v1
+
+## 24095. Evaluating Reward Model Generalization via Pairwise Maximum Discrepancy Competitions
+
+- **ID:** `2601.16987` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Shunyang Luo, Peibei Cao, Zhihui Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.16987 · **PDF:** https://arxiv.org/pdf/2601.16987v1
+
+## 24096. QuIC: A Quantum-Inspired Interaction Classifier for Revitalizing Shallow CNNs in Fine-Grained Recognition
+
+- **ID:** `2601.02189` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Cheng Ying Wu, Yen Jui Chang
+- **abs:** https://arxiv.org/abs/2601.02189 · **PDF:** https://arxiv.org/pdf/2601.02189v1
+
+## 24097. Toward Global Large Language Models in Medicine
+
+- **ID:** `2601.02186` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Rui Yang, Huitao Li, Weihao Xuan, et al.
+- **abs:** https://arxiv.org/abs/2601.02186 · **PDF:** https://arxiv.org/pdf/2601.02186v1
+
+## 24098. Confidence Estimation for LLMs in Multi-turn Interactions
+
+- **ID:** `2601.02179` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Caiqi Zhang, Ruihan Yang, Xiaochen Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.02179 · **PDF:** https://arxiv.org/pdf/2601.02179v1
+
+## 24099. Streaming Hallucination Detection in Long Chain-of-Thought Reasoning
+
+- **ID:** `2601.02170` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Haolang Lu, Minghui Pan, Ripeng Li, et al.
+- **abs:** https://arxiv.org/abs/2601.02170 · **PDF:** https://arxiv.org/pdf/2601.02170v1
+
+## 24100. EverMemOS: A Self-Organizing Memory Operating System for Structured Long-Horizon Reasoning
+
+- **ID:** `2601.02163` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Chuanrui Hu, Xingze Gao, Zuyi Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.02163 · **PDF:** https://arxiv.org/pdf/2601.02163v2
+
+## 24101. FormationEval, an open multiple-choice benchmark for petroleum geoscience
+
+- **ID:** `2601.02158` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Almaz Ermilov
+- **abs:** https://arxiv.org/abs/2601.02158 · **PDF:** https://arxiv.org/pdf/2601.02158v2
+
+## 24102. Multi-fidelity graph-based neural networks architectures to learn Navier-Stokes solutions on non-parametrized 2D domains
+
+- **ID:** `2601.02157` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-01-05
+- **Authors:** Francesco Songia, Raoul Sallé de Chou, Hugues Talbot, et al.
+- **abs:** https://arxiv.org/abs/2601.02157 · **PDF:** https://arxiv.org/pdf/2601.02157v2
+
+## 24103. Causal and Federated Multimodal Learning for Cardiovascular Risk Prediction under Heterogeneous Populations
+
+- **ID:** `2601.06140` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Rohit Kaushik, Eva Kaushik
+- **abs:** https://arxiv.org/abs/2601.06140 · **PDF:** https://arxiv.org/pdf/2601.06140v1
+
+## 24104. Entropy-Adaptive Fine-Tuning: Resolving Confident Conflicts to Mitigate Forgetting
+
+- **ID:** `2601.02151` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Muxi Diao, Lele Yang, Wuxuan Gong, et al.
+- **abs:** https://arxiv.org/abs/2601.02151 · **PDF:** https://arxiv.org/pdf/2601.02151v1
+
+## 24105. AI-enhanced tuning of quantum dot Hamiltonians toward Majorana modes
+
+- **ID:** `2601.02149` · **primary:** `cond-mat.mes-hall` · **Submitted:** 2026-01-05
+- **Authors:** Mateusz Krawczyk, Jarosław Pawłowski
+- **abs:** https://arxiv.org/abs/2601.02149 · **PDF:** https://arxiv.org/pdf/2601.02149v2
+
+## 24106. BiPrompt: Bilateral Prompt Optimization for Visual and Textual Debiasing in Vision-Language Models
+
+- **ID:** `2601.02147` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Sunny Gupta, Shounak Das, Amit Sethi
+- **abs:** https://arxiv.org/abs/2601.02147 · **PDF:** https://arxiv.org/pdf/2601.02147v1
+
+## 24107. Feature-based Inversion of 2.5D Controlled Source Electromagnetic Data using Generative Priors
+
+- **ID:** `2601.02145` · **primary:** `physics.geo-ph` · **Submitted:** 2026-01-05
+- **Authors:** Hongyu Zhou, Haoran Sun, Rui Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.02145 · **PDF:** https://arxiv.org/pdf/2601.02145v1
+
+## 24108. Routing by Analogy: kNN-Augmented Expert Assignment for Mixture-of-Experts
+
+- **ID:** `2601.02144` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Boxuan Lyu, Soichiro Murakami, Hidetaka Kamigaito, et al.
+- **abs:** https://arxiv.org/abs/2601.02144 · **PDF:** https://arxiv.org/pdf/2601.02144v1
+
+## 24109. Edge-aware GAT-based protein binding site prediction
+
+- **ID:** `2601.02138` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Weisen Yang, Hanqing Zhang, Wangren Qiu, et al.
+- **abs:** https://arxiv.org/abs/2601.02138 · **PDF:** https://arxiv.org/pdf/2601.02138v1
+
+## 24110. A Spatio-Temporal Deep Learning Approach For High-Resolution Gridded Monsoon Prediction
+
+- **ID:** `2601.02445` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Parashjyoti Borah, Sanghamitra Sarkar, Ranjan Phukan
+- **abs:** https://arxiv.org/abs/2601.02445 · **PDF:** https://arxiv.org/pdf/2601.02445v1
+
+## 24111. Towards Multi-Level Transcript Segmentation: LoRA Fine-Tuning for Table-of-Contents Generation
+
+- **ID:** `2601.02128` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Steffen Freisinger, Philipp Seeberger, Thomas Ranzenberger, et al.
+- **abs:** https://arxiv.org/abs/2601.02128 · **PDF:** https://arxiv.org/pdf/2601.02128v1
+
+## 24112. Remote Sensing Change Detection via Weak Temporal Supervision
+
+- **ID:** `2601.02126` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Xavier Bou, Elliot Vincent, Gabriele Facciolo, et al.
+- **abs:** https://arxiv.org/abs/2601.02126 · **PDF:** https://arxiv.org/pdf/2601.02126v1
+
+## 24113. SingingBot: An Avatar-Driven System for Robotic Face Singing Performance
+
+- **ID:** `2601.02125` · **primary:** `cs.RO` · **Submitted:** 2026-01-05
+- **Authors:** Zhuoxiong Xu, Xuanchen Li, Yuhao Cheng, et al.
+- **abs:** https://arxiv.org/abs/2601.02125 · **PDF:** https://arxiv.org/pdf/2601.02125v1
+
+## 24114. DeCode: Decoupling Content and Delivery for Medical QA
+
+- **ID:** `2601.02123` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Po-Jen Ko, Chen-Han Tsai, Yu-Shao Peng
+- **abs:** https://arxiv.org/abs/2601.02123 · **PDF:** https://arxiv.org/pdf/2601.02123v3
+
+## 24115. Inferring Network Evolutionary History via Structure-State Coupled Learning
+
+- **ID:** `2601.02121` · **primary:** `cs.SI` · **Submitted:** 2026-01-05
+- **Authors:** En Xu, Shihe Zhou, Huandong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.02121 · **PDF:** https://arxiv.org/pdf/2601.02121v1
+
+## 24116. VocalBridge: Latent Diffusion-Bridge Purification for Defeating Perturbation-Based Voiceprint Defenses
+
+- **ID:** `2601.02444` · **primary:** `cs.SD` · **Submitted:** 2026-01-05
+- **Authors:** Maryam Abbasihafshejani, AHM Nazmus Sakib, Murtuza Jadliwala
+- **abs:** https://arxiv.org/abs/2601.02444 · **PDF:** https://arxiv.org/pdf/2601.02444v1
+
+## 24117. Car Drag Coefficient Prediction from 3D Point Clouds Using a Slice-Based Surrogate Model
+
+- **ID:** `2601.02112` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Utkarsh Singh, Absaar Ali, Adarsh Roy
+- **abs:** https://arxiv.org/abs/2601.02112 · **PDF:** https://arxiv.org/pdf/2601.02112v1
+
+## 24118. Prototype-Based Learning for Healthcare: A Demonstration of Interpretable AI
+
+- **ID:** `2601.02106` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Ashish Rana, Ammar Shaker, Sascha Saralajew, et al.
+- **abs:** https://arxiv.org/abs/2601.02106 · **PDF:** https://arxiv.org/pdf/2601.02106v1
+
+## 24119. LION-DG: Layer-Informed Initialization with Deep Gradient Protocols for Accelerated Neural Network Training
+
+- **ID:** `2601.02105` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Hyunjun Kim
+- **abs:** https://arxiv.org/abs/2601.02105 · **PDF:** https://arxiv.org/pdf/2601.02105v1
+
+## 24120. Evaluating the Diagnostic Classification Ability of Multimodal Large Language Models: Insights from the Osteoarthritis Initiative
+
+- **ID:** `2601.02443` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Li Wang, Xi Chen, XiangWen Deng, et al.
+- **abs:** https://arxiv.org/abs/2601.02443 · **PDF:** https://arxiv.org/pdf/2601.02443v1
+
+## 24121. Horizon Activation Mapping for Neural Networks in Time Series Forecasting
+
+- **ID:** `2601.02094` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Krupakar Hans, V A Kandappan
+- **abs:** https://arxiv.org/abs/2601.02094 · **PDF:** https://arxiv.org/pdf/2601.02094v3
+
+## 24122. Vision-Based Early Fault Diagnosis and Self-Recovery for Strawberry Harvesting Robots
+
+- **ID:** `2601.02085` · **primary:** `cs.RO` · **Submitted:** 2026-01-05
+- **Authors:** Meili Sun, Chunjiang Zhao, Lichao Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.02085 · **PDF:** https://arxiv.org/pdf/2601.02085v2
+
+## 24123. A Differentiable Adversarial Framework for Task-Aware Data Subsampling
+
+- **ID:** `2601.02081` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Jiacheng Lyu, Bihua Bao
+- **abs:** https://arxiv.org/abs/2601.02081 · **PDF:** https://arxiv.org/pdf/2601.02081v2
+
+## 24124. The Homogeneity Trap: Spectral Collapse in Doubly-Stochastic Deep Networks
+
+- **ID:** `2601.02080` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Yizhi Liu
+- **abs:** https://arxiv.org/abs/2601.02080 · **PDF:** https://arxiv.org/pdf/2601.02080v1
+
+## 24125. Deferred Commitment Decoding for Diffusion Language Models
+
+- **ID:** `2601.02076` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Yingte Shu, Yuchuan Tian, Chao Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.02076 · **PDF:** https://arxiv.org/pdf/2601.02076v2
+
+## 24126. MDAgent2: Large Language Model for Code Generation and Knowledge Q&A in Molecular Dynamics
+
+- **ID:** `2601.02075` · **primary:** `cs.CE` · **Submitted:** 2026-01-05
+- **Authors:** Zhuofan Shi, Hubao A, Yufei Shao, et al.
+- **abs:** https://arxiv.org/abs/2601.02075 · **PDF:** https://arxiv.org/pdf/2601.02075v4
+
+## 24127. FormuLLA: A Large Language Model Approach to Generating Novel 3D Printable Formulations
+
+- **ID:** `2601.02071` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Adeshola Okubena, Yusuf Ali Mohammed, Moe Elbadawi
+- **abs:** https://arxiv.org/abs/2601.02071 · **PDF:** https://arxiv.org/pdf/2601.02071v2
+
+## 24128. Cost-Efficient Cross-Lingual Retrieval-Augmented Generation for Low-Resource Languages: A Case Study in Bengali Agricultural Advisory
+
+- **ID:** `2601.02065` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Md. Asif Hossain, Nabil Subhan, Mantasha Rahman Mahi, et al.
+- **abs:** https://arxiv.org/abs/2601.02065 · **PDF:** https://arxiv.org/pdf/2601.02065v1
+
+## 24129. Higher-Order Action Regularization in Deep Reinforcement Learning: From Continuous Control to Building Energy Management
+
+- **ID:** `2601.02061` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Faizan Ahmed, Aniket Dixit, James Brusey
+- **abs:** https://arxiv.org/abs/2601.02061 · **PDF:** https://arxiv.org/pdf/2601.02061v1
+
+## 24130. Perish or Flourish? A Holistic Evaluation of Large Language Models for Code Generation in Functional Programming
+
+- **ID:** `2601.02060` · **primary:** `cs.PL` · **Submitted:** 2026-01-05
+- **Authors:** Nguyet-Anh H. Lang, Eric Lang, Thanh Le-Cong, et al.
+- **abs:** https://arxiv.org/abs/2601.02060 · **PDF:** https://arxiv.org/pdf/2601.02060v1
+
+## 24131. Explore the Ideology of Deep Learning in ENSO Forecasts
+
+- **ID:** `2601.02050` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Yanhai Gan, Yipeng Chen, Ning Li, et al.
+- **abs:** https://arxiv.org/abs/2601.02050 · **PDF:** https://arxiv.org/pdf/2601.02050v1
+
+## 24132. Agentic Retoucher for Text-To-Image Generation
+
+- **ID:** `2601.02046` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Shaocheng Shen, Jianfeng Liang, Chunlei Cai, et al.
+- **abs:** https://arxiv.org/abs/2601.02046 · **PDF:** https://arxiv.org/pdf/2601.02046v3
+
+## 24133. The New Compiler Stack: A Survey on the Synergy of LLMs and Compilers
+
+- **ID:** `2601.02045` · **primary:** `cs.PL` · **Submitted:** 2026-01-05
+- **Authors:** Shuoming Zhang, Jiacheng Zhao, Qiuchu Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.02045 · **PDF:** https://arxiv.org/pdf/2601.02045v1
+
+## 24134. Simulated Reasoning is Reasoning
+
+- **ID:** `2601.02043` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Hendrik Kempt, Alon Lavie
+- **abs:** https://arxiv.org/abs/2601.02043 · **PDF:** https://arxiv.org/pdf/2601.02043v1
+
+## 24135. Multivariate Time-series Anomaly Detection via Dynamic Model Pool & Ensembling
+
+- **ID:** `2601.02037` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Wei Hu, Zewei Yu, Jianqiu Xu
+- **abs:** https://arxiv.org/abs/2601.02037 · **PDF:** https://arxiv.org/pdf/2601.02037v1
+
+## 24136. GDRO: Group-level Reward Post-training Suitable for Diffusion Models
+
+- **ID:** `2601.02036` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Yiyang Wang, Xi Chen, Xiaogang Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.02036 · **PDF:** https://arxiv.org/pdf/2601.02036v1
+
+## 24137. Output Embedding Centering for Stable LLM Pretraining
+
+- **ID:** `2601.02031` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Felix Stollenwerk, Anna Lokrantz, Niclas Hertzberg
+- **abs:** https://arxiv.org/abs/2601.02031 · **PDF:** https://arxiv.org/pdf/2601.02031v2
+
+## 24138. Understanding Pure Textual Reasoning for Blind Image Quality Assessment
+
+- **ID:** `2601.02441` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Yuan Li, Shin'ya Nishida
+- **abs:** https://arxiv.org/abs/2601.02441 · **PDF:** https://arxiv.org/pdf/2601.02441v2
+
+## 24139. Not All Needles Are Found: How Fact Distribution and Don't Make It Up Prompts Shape Literal Extraction, Logical Inference, and Hallucination Risks in Long-Context LLMs
+
+- **ID:** `2601.02023` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Amirali Ebrahimzadeh, Seyyed M. Salili
+- **abs:** https://arxiv.org/abs/2601.02023 · **PDF:** https://arxiv.org/pdf/2601.02023v1
+
+## 24140. Prior Diffusiveness and Regret in the Linear-Gaussian Bandit
+
+- **ID:** `2601.02022` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Yifan Zhu, John C. Duchi, Benjamin Van Roy
+- **abs:** https://arxiv.org/abs/2601.02022 · **PDF:** https://arxiv.org/pdf/2601.02022v1
+
+## 24141. Enhancing Object Detection with Privileged Information: A Model-Agnostic Teacher-Student Approach
+
+- **ID:** `2601.02016` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Matthias Bartolo, Dylan Seychell, Gabriel Hili, et al.
+- **abs:** https://arxiv.org/abs/2601.02016 · **PDF:** https://arxiv.org/pdf/2601.02016v1
+
+## 24142. Surprisal and Metaphor Novelty Judgments: Moderate Correlations and Divergent Scaling Effects Revealed by Corpus-Based and Synthetic Datasets
+
+- **ID:** `2601.02015` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Omar Momen, Emilie Sitter, Berenike Herrmann, et al.
+- **abs:** https://arxiv.org/abs/2601.02015 · **PDF:** https://arxiv.org/pdf/2601.02015v3
+
+## 24143. A neural network for modeling human concept formation, understanding and communication
+
+- **ID:** `2601.02010` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-05
+- **Authors:** Liangxuan Guo, Haoyang Chen, Yang Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.02010 · **PDF:** https://arxiv.org/pdf/2601.02010v1
+
+## 24144. XAI-MeD: Explainable Knowledge Guided Neuro-Symbolic Framework for Domain Generalization and Rare Class Detection in Medical Imaging
+
+- **ID:** `2601.02008` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Midhat Urooj, Ayan Banerjee, Sandeep Gupta
+- **abs:** https://arxiv.org/abs/2601.02008 · **PDF:** https://arxiv.org/pdf/2601.02008v1
+
+## 24145. Exploring Approaches for Detecting Memorization of Recommender System Data in Large Language Models
+
+- **ID:** `2601.02002` · **primary:** `cs.IR` · **Submitted:** 2026-01-05
+- **Authors:** Antonio Colacicco, Vito Guida, Dario Di Palma, et al.
+- **abs:** https://arxiv.org/abs/2601.02002 · **PDF:** https://arxiv.org/pdf/2601.02002v1
+
+## 24146. Exploring Diversity, Novelty, and Popularity Bias in ChatGPT's Recommendations
+
+- **ID:** `2601.01997` · **primary:** `cs.IR` · **Submitted:** 2026-01-05
+- **Authors:** Dario Di Palma, Giovanni Maria Biancofiore, Vito Walter Anelli, et al.
+- **abs:** https://arxiv.org/abs/2601.01997 · **PDF:** https://arxiv.org/pdf/2601.01997v1
+
+## 24147. Towards Privacy-Preserving Mental Health Support with Large Language Models
+
+- **ID:** `2601.01993` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Dong Xue, Jicheng Tu, Ming Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.01993 · **PDF:** https://arxiv.org/pdf/2601.01993v2
+
+## 24148. VIT-Ped: Visionary Intention Transformer for Pedestrian Behavior Analysis
+
+- **ID:** `2601.01989` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Aly R. Elkammar, Karim M. Gamaleldin, Catherine M. Elias
+- **abs:** https://arxiv.org/abs/2601.01989 · **PDF:** https://arxiv.org/pdf/2601.01989v1
+
+## 24149. ChaosBench-Logic: A Benchmark for Logical and Symbolic Reasoning on Chaotic Dynamical Systems
+
+- **ID:** `2601.01982` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Noel Thomas
+- **abs:** https://arxiv.org/abs/2601.01982 · **PDF:** https://arxiv.org/pdf/2601.01982v1
+
+## 24150. SerpentFlow: Generative Unpaired Domain Alignment via Shared-Structure Decomposition
+
+- **ID:** `2601.01979` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Julie Keisler, Anastase Alexandre Charantonis, Yannig Goude, et al.
+- **abs:** https://arxiv.org/abs/2601.01979 · **PDF:** https://arxiv.org/pdf/2601.01979v1
+
+## 24151. CNC-TP: Classifier Nominal Concept Based on Top-Pertinent Attributes
+
+- **ID:** `2601.01976` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Yasmine Souissi, Fabrice Boissier, Nida Meddouri
+- **abs:** https://arxiv.org/abs/2601.01976 · **PDF:** https://arxiv.org/pdf/2601.01976v1
+
+## 24152. Hidden State Poisoning Attacks against Mamba-based Language Models
+
+- **ID:** `2601.01972` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Alexandre Le Mercier, Chris Develder, Thomas Demeester
+- **abs:** https://arxiv.org/abs/2601.01972 · **PDF:** https://arxiv.org/pdf/2601.01972v4
+
+## 24153. A Multilayered Approach to Classifying Customer Responsiveness and Credit Risk
+
+- **ID:** `2601.01970` · **primary:** `stat.ML` · **Submitted:** 2026-01-05
+- **Authors:** Ayomide Afolabi, Ebere Ogburu, Symon Kimitei
+- **abs:** https://arxiv.org/abs/2601.01970 · **PDF:** https://arxiv.org/pdf/2601.01970v1
+
+## 24154. Refinement Provenance Inference: Detecting LLM-Refined Training Prompts from Model Behavior
+
+- **ID:** `2601.01966` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Bo Yin, Qi Li, Runpeng Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.01966 · **PDF:** https://arxiv.org/pdf/2601.01966v1
+
+## 24155. CSF: Contrastive Semantic Features for Direct Multilingual Sign Language Generation
+
+- **ID:** `2601.01964` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Tran Sy Bao
+- **abs:** https://arxiv.org/abs/2601.01964 · **PDF:** https://arxiv.org/pdf/2601.01964v1
+
+## 24156. Forget Less by Learning Together through Concept Consolidation
+
+- **ID:** `2601.01963` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Arjun Ramesh Kaushik, Naresh Kumar Devulapally, Vishnu Suresh Lokhande, et al.
+- **abs:** https://arxiv.org/abs/2601.01963 · **PDF:** https://arxiv.org/pdf/2601.01963v1
+
+## 24157. Mitigating Long-Tailed Anomaly Score Distributions with Importance-Weighted Loss
+
+- **ID:** `2601.02440` · **primary:** `stat.ML` · **Submitted:** 2026-01-05
+- **Authors:** Jungi Lee, Jungkwon Kim, Chi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.02440 · **PDF:** https://arxiv.org/pdf/2601.02440v1
+
+## 24158. The Invisible Hand of AI Libraries Shaping Open Source Projects and Communities
+
+- **ID:** `2601.01944` · **primary:** `cs.SE` · **Submitted:** 2026-01-05
+- **Authors:** Matteo Esposito, Andrea Janes, Valentina Lenarduzzi, et al.
+- **abs:** https://arxiv.org/abs/2601.01944 · **PDF:** https://arxiv.org/pdf/2601.01944v2
+
+## 24159. SynRXN: An Open Benchmark and Curated Dataset for Computational Reaction Modeling
+
+- **ID:** `2601.01943` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Tieu-Long Phan, Nhu-Ngoc Nguyen Song, Peter F. Stadler
+- **abs:** https://arxiv.org/abs/2601.01943 · **PDF:** https://arxiv.org/pdf/2601.01943v1
+
+## 24160. OpenSocInt: A Multi-modal Training Environment for Human-Aware Social Navigation
+
+- **ID:** `2601.01939` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Victor Sanchez, Chris Reinke, Ahamed Mohamed, et al.
+- **abs:** https://arxiv.org/abs/2601.01939 · **PDF:** https://arxiv.org/pdf/2601.01939v2
+
+## 24161. WebGym: Scaling Training Environments for Visual Web Agents with Realistic Tasks
+
+- **ID:** `2601.02439` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Hao Bai, Alexey Taymanov, Tong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.02439 · **PDF:** https://arxiv.org/pdf/2601.02439v5
+
+## 24162. Visualizing the Structure of Lenia Parameter Space
+
+- **ID:** `2601.01932` · **primary:** `nlin.CG` · **Submitted:** 2026-01-05
+- **Authors:** Barbora Hudcová, František Dušek, Marco Tuccio, et al.
+- **abs:** https://arxiv.org/abs/2601.01932 · **PDF:** https://arxiv.org/pdf/2601.01932v1
+
+## 24163. Focus on What Matters: Fisher-Guided Adaptive Multimodal Fusion for Vulnerability Detection
+
+- **ID:** `2601.02438` · **primary:** `cs.SE` · **Submitted:** 2026-01-05
+- **Authors:** Yun Bian, Yi Chen, HaiQuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.02438 · **PDF:** https://arxiv.org/pdf/2601.02438v2
+
+## 24164. DéjàQ: Open-Ended Evolution of Diverse, Learnable and Verifiable Problems
+
+- **ID:** `2601.01931` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Willem Röpke, Samuel Coward, Andrei Lupu, et al.
+- **abs:** https://arxiv.org/abs/2601.01931 · **PDF:** https://arxiv.org/pdf/2601.01931v1
+
+## 24165. MCGI: Manifold-Consistent Graph Indexing for Billion-Scale Disk-Resident Vector Search
+
+- **ID:** `2601.01930` · **primary:** `cs.IR` · **Submitted:** 2026-01-05
+- **Authors:** Dongfang Zhao
+- **abs:** https://arxiv.org/abs/2601.01930 · **PDF:** https://arxiv.org/pdf/2601.01930v2
+
+## 24166. Theoretical Convergence of SMOTE-Generated Samples
+
+- **ID:** `2601.01927` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Firuz Kamalov, Hana Sulieman, Witold Pedrycz
+- **abs:** https://arxiv.org/abs/2601.01927 · **PDF:** https://arxiv.org/pdf/2601.01927v1
+
+## 24167. Solving Cyclic Antibandwidth Problem by SAT
+
+- **ID:** `2601.04239` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Hieu Truong Xuan, Khanh To Van
+- **abs:** https://arxiv.org/abs/2601.04239 · **PDF:** https://arxiv.org/pdf/2601.04239v1
+
+## 24168. Efficient temporal prediction of compressible flows in irregular domains using Fourier neural operators
+
+- **ID:** `2601.01922` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-01-05
+- **Authors:** Yifan Nie, Qiaoxin Li
+- **abs:** https://arxiv.org/abs/2601.01922 · **PDF:** https://arxiv.org/pdf/2601.01922v1
+
+## 24169. A Defect is Being Born: How Close Are We? A Time Sensitive Forecasting Approach
+
+- **ID:** `2601.01921` · **primary:** `cs.SE` · **Submitted:** 2026-01-05
+- **Authors:** Mikel Robredo, Matteo Esposito, Fabio Palomba, et al.
+- **abs:** https://arxiv.org/abs/2601.01921 · **PDF:** https://arxiv.org/pdf/2601.01921v1
+
+## 24170. Distorted Distributional Policy Evaluation for Offline Reinforcement Learning
+
+- **ID:** `2601.01917` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Ryo Iwaki, Takayuki Osogami
+- **abs:** https://arxiv.org/abs/2601.01917 · **PDF:** https://arxiv.org/pdf/2601.01917v1
+
+## 24171. Toward Thermodynamic Reservoir Computing: Exploring SHA-256 ASICs as Potential Physical Substrates
+
+- **ID:** `2601.01916` · **primary:** `cs.NE` · **Submitted:** 2026-01-05
+- **Authors:** Francisco Angulo de Lafuente, Vladimir Veselov, Richard Goodman
+- **abs:** https://arxiv.org/abs/2601.01916 · **PDF:** https://arxiv.org/pdf/2601.01916v1
+
+## 24172. TAP-ViTs: Task-Adaptive Pruning for On-Device Deployment of Vision Transformers
+
+- **ID:** `2601.02437` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Zhibo Wang, Zuoyuan Zhang, Xiaoyi Pang, et al.
+- **abs:** https://arxiv.org/abs/2601.02437 · **PDF:** https://arxiv.org/pdf/2601.02437v1
+
+## 24173. MMP-A*: Multimodal Perception Enhanced Incremental Heuristic Search on Path Planning
+
+- **ID:** `2601.01910` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Minh Hieu Ha, Khanh Ly Ta, Hung Phan, et al.
+- **abs:** https://arxiv.org/abs/2601.01910 · **PDF:** https://arxiv.org/pdf/2601.01910v2
+
+## 24174. Nodule-DETR: A Novel DETR Architecture with Frequency-Channel Attention for Ultrasound Thyroid Nodule Detection
+
+- **ID:** `2601.01908` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Jingjing Wang, Qianglin Liu, Zhuo Xiao, et al.
+- **abs:** https://arxiv.org/abs/2601.01908 · **PDF:** https://arxiv.org/pdf/2601.01908v1
+
+## 24175. Evaluating Feature Dependent Noise in Preference-based Reinforcement Learning
+
+- **ID:** `2601.01904` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Yuxuan Li, Harshith Reddy Kethireddy, Srijita Das
+- **abs:** https://arxiv.org/abs/2601.01904 · **PDF:** https://arxiv.org/pdf/2601.01904v2
+
+## 24176. TT-FSI: Scalable Faithful Shapley Interactions via Tensor-Train
+
+- **ID:** `2601.01903` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Ungsik Kim, Suwon Lee
+- **abs:** https://arxiv.org/abs/2601.01903 · **PDF:** https://arxiv.org/pdf/2601.01903v1
+
+## 24177. FedBiCross: A Bi-Level Optimization Framework to Tackle Non-IID Challenges in Data-Free One-Shot Federated Learning on Medical Data
+
+- **ID:** `2601.01901` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Yuexuan Xia, Yinghao Zhang, Yalin Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.01901 · **PDF:** https://arxiv.org/pdf/2601.01901v1
+
+## 24178. Multi-strategy Improved Northern Goshawk Optimization for WSN Coverage Enhancement
+
+- **ID:** `2601.01898` · **primary:** `cs.NE` · **Submitted:** 2026-01-05
+- **Authors:** Yiran Tian, Yuanjia Liu
+- **abs:** https://arxiv.org/abs/2601.01898 · **PDF:** https://arxiv.org/pdf/2601.01898v1
+
+## 24179. Tackling the Inherent Difficulty of Noise Filtering in RAG
+
+- **ID:** `2601.01896` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Jingyu Liu, Jiaen Lin, Yong Liu
+- **abs:** https://arxiv.org/abs/2601.01896 · **PDF:** https://arxiv.org/pdf/2601.01896v2
+
+## 24180. Encoder-Free ECG-Language Models
+
+- **ID:** `2601.18798` · **primary:** `cs.MM` · **Submitted:** 2026-01-05
+- **Authors:** William Han, Tony Chen, Chaojing Duan, et al.
+- **abs:** https://arxiv.org/abs/2601.18798 · **PDF:** https://arxiv.org/pdf/2601.18798v1
+
+## 24181. Forget Less by Learning from Parents Through Hierarchical Relationships
+
+- **ID:** `2601.01892` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Arjun Ramesh Kaushik, Naresh Kumar Devulapally, Vishnu Suresh Lokhande, et al.
+- **abs:** https://arxiv.org/abs/2601.01892 · **PDF:** https://arxiv.org/pdf/2601.01892v1
+
+## 24182. SafeLoad: Efficient Admission Control Framework for Identifying Memory-Overloading Queries in Cloud Data Warehouses
+
+- **ID:** `2601.01888` · **primary:** `cs.DB` · **Submitted:** 2026-01-05
+- **Authors:** Yifan Wu, Yuhan Li, Zhenhua Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.01888 · **PDF:** https://arxiv.org/pdf/2601.01888v1
+
+## 24183. Safety at One Shot: Patching Fine-Tuned LLMs with A Single Instance
+
+- **ID:** `2601.01887` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Jiawen Zhang, Lipeng He, Kejia Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.01887 · **PDF:** https://arxiv.org/pdf/2601.01887v2
+
+## 24184. Agentic Memory: Learning Unified Long-Term and Short-Term Memory Management for Large Language Model Agents
+
+- **ID:** `2601.01885` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Yi Yu, Liuyi Yao, Yuexiang Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.01885 · **PDF:** https://arxiv.org/pdf/2601.01885v1
+
+## 24185. Deep Learning Superresolution for 7T Knee MR Imaging: Impact on Image Quality and Diagnostic Performance
+
+- **ID:** `2601.02436` · **primary:** `eess.IV` · **Submitted:** 2026-01-05
+- **Authors:** Pinzhen Chen, Libo Xu, Boyang Pan, et al.
+- **abs:** https://arxiv.org/abs/2601.02436 · **PDF:** https://arxiv.org/pdf/2601.02436v1
+
+## 24186. RainBalance: Alleviating Dual Imbalance in GNSS-based Precipitation Nowcasting via Continuous Probability Modeling
+
+- **ID:** `2601.06137` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Yifang Zhang, Shengwu Xiong, Henan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.06137 · **PDF:** https://arxiv.org/pdf/2601.06137v1
+
+## 24187. Theory Trace Card: Theory-Driven Socio-Cognitive Evaluation of LLMs
+
+- **ID:** `2601.01878` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Farzan Karimi-Malekabadi, Suhaib Abdurahman, Zhivar Sourati, et al.
+- **abs:** https://arxiv.org/abs/2601.01878 · **PDF:** https://arxiv.org/pdf/2601.01878v1
+
+## 24188. Random-Matrix-Induced Simplicity Bias in Over-parameterized Variational Quantum Circuits
+
+- **ID:** `2601.01877` · **primary:** `quant-ph` · **Submitted:** 2026-01-05
+- **Authors:** Jun Qi, Chao-Han Huck Yang, Pin-Yu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.01877 · **PDF:** https://arxiv.org/pdf/2601.01877v1
+
+## 24189. Toward Auditable Neuro-Symbolic Reasoning in Pathology: SQL as an Explicit Trace of Evidence
+
+- **ID:** `2601.01875` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Kewen Cao, Jianxu Chen, Yongbing Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.01875 · **PDF:** https://arxiv.org/pdf/2601.01875v1
+
+## 24190. CogFlow: Bridging Perception and Reasoning through Knowledge Internalization for Visual Mathematical Problem Solving
+
+- **ID:** `2601.01874` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Shuhang Chen, Yunqiu Xu, Junjie Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.01874 · **PDF:** https://arxiv.org/pdf/2601.01874v3
+
+## 24191. DermoGPT: Open Weights and Open Data for Morphology-Grounded Dermatological Reasoning MLLMs
+
+- **ID:** `2601.01868` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Jinghan Ru, Siyuan Yan, Yuguo Yin, et al.
+- **abs:** https://arxiv.org/abs/2601.01868 · **PDF:** https://arxiv.org/pdf/2601.01868v1
+
+## 24192. Judging with Personality and Confidence: A Study on Personality-Conditioned LLM Relevance Assessment
+
+- **ID:** `2601.01862` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Nuo Chen, Hanpei Fang, Piaohong Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.01862 · **PDF:** https://arxiv.org/pdf/2601.01862v1
+
+## 24193. Crystal-KV: Efficient KV Cache Management for Chain-of-Thought LLMs via Answer-First Principle
+
+- **ID:** `2601.16986` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Zihan Wang, Cheng Tang, Lei Gong, et al.
+- **abs:** https://arxiv.org/abs/2601.16986 · **PDF:** https://arxiv.org/pdf/2601.16986v1
+
+## 24194. High-Order Epistasis Detection Using Factorization Machine with Quadratic Optimization Annealing and MDR-Based Evaluation
+
+- **ID:** `2601.01860` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Shuta Kikuchi, Shu Tanaka
+- **abs:** https://arxiv.org/abs/2601.01860 · **PDF:** https://arxiv.org/pdf/2601.01860v1
+
+## 24195. Jenius Agent: Towards Experience-Driven Accuracy Optimization in Real-World Scenarios
+
+- **ID:** `2601.01857` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Defei Xia, Bingfeng Pi, Shenbin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.01857 · **PDF:** https://arxiv.org/pdf/2601.01857v3
+
+## 24196. MORE: Multi-Objective Adversarial Attacks on Speech Recognition
+
+- **ID:** `2601.01852` · **primary:** `eess.AS` · **Submitted:** 2026-01-05
+- **Authors:** Xiaoxue Gao, Zexin Li, Yiming Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.01852 · **PDF:** https://arxiv.org/pdf/2601.01852v2
+
+## 24197. Clinical Knowledge Graph Construction and Evaluation with Multi-LLMs via Retrieval-Augmented Generation
+
+- **ID:** `2601.01844` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Udiptaman Das, Krishnasai B. Atmakuri, Duy Ho, et al.
+- **abs:** https://arxiv.org/abs/2601.01844 · **PDF:** https://arxiv.org/pdf/2601.01844v1
+
+## 24198. Towards Automated Lexicography: Generating and Evaluating Definitions for Learner's Dictionaries
+
+- **ID:** `2601.01842` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Yusuke Ide, Adam Nohejl, Joshua Tanner, et al.
+- **abs:** https://arxiv.org/abs/2601.01842 · **PDF:** https://arxiv.org/pdf/2601.01842v1
+
+## 24199. Tackling Resource-Constrained and Data-Heterogeneity in Federated Learning with Double-Weight Sparse Pack
+
+- **ID:** `2601.01840` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Qiantao Yang, Liquan Chen, Mingfu Xue, et al.
+- **abs:** https://arxiv.org/abs/2601.01840 · **PDF:** https://arxiv.org/pdf/2601.01840v1
+
+## 24200. The Machine Learning Canvas: Empirical Findings on Why Strategy Matters More Than AI Code Generation
+
+- **ID:** `2601.01839` · **primary:** `cs.SE` · **Submitted:** 2026-01-05
+- **Authors:** Martin Prause
+- **abs:** https://arxiv.org/abs/2601.01839 · **PDF:** https://arxiv.org/pdf/2601.01839v1
+
+## 24201. COMPASS: A Framework for Evaluating Organization-Specific Policy Alignment in LLMs
+
+- **ID:** `2601.01836` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Dasol Choi, DongGeon Lee, Brigitta Jesica Kartono, et al.
+- **abs:** https://arxiv.org/abs/2601.01836 · **PDF:** https://arxiv.org/pdf/2601.01836v1
+
+## 24202. RSwinV2-MD: An Enhanced Residual SwinV2 Transformer for Monkeypox Detection from Skin Images
+
+- **ID:** `2601.01835` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Rashid Iqbal, Saddam Hussain Khan
+- **abs:** https://arxiv.org/abs/2601.01835 · **PDF:** https://arxiv.org/pdf/2601.01835v2
+
+## 24203. FAROS: Robust Federated Learning with Adaptive Scaling against Backdoor Attacks
+
+- **ID:** `2601.01833` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Chenyu Hu, Qiming Hu, Sinan Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.01833 · **PDF:** https://arxiv.org/pdf/2601.01833v1
+
+## 24204. Yukthi Opus: A Multi-Chain Hybrid Metaheuristic for Large-Scale NP-Hard Optimization
+
+- **ID:** `2601.01832` · **primary:** `cs.NE` · **Submitted:** 2026-01-05
+- **Authors:** SB Danush Vikraman, Hannah Abigail, Prasanna Kesavraj, et al.
+- **abs:** https://arxiv.org/abs/2601.01832 · **PDF:** https://arxiv.org/pdf/2601.01832v3
+
+## 24205. ARIES: A Scalable Multi-Agent Orchestration Framework for Real-Time Epidemiological Surveillance and Outbreak Monitoring
+
+- **ID:** `2601.01831` · **primary:** `cs.MA` · **Submitted:** 2026-01-05
+- **Authors:** Aniket Wattamwar, Sampson Akwafuo
+- **abs:** https://arxiv.org/abs/2601.01831 · **PDF:** https://arxiv.org/pdf/2601.01831v1
+
+## 24206. RealPDEBench: A Benchmark for Complex Physical Systems with Real-World Data
+
+- **ID:** `2601.01829` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Peiyan Hu, Haodong Feng, Hongyuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.01829 · **PDF:** https://arxiv.org/pdf/2601.01829v2
+
+## 24207. Emergent Introspective Awareness in Large Language Models
+
+- **ID:** `2601.01828` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Jack Lindsey
+- **abs:** https://arxiv.org/abs/2601.01828 · **PDF:** https://arxiv.org/pdf/2601.01828v1
+
+## 24208. Aspect Extraction from E-Commerce Product and Service Reviews
+
+- **ID:** `2601.01827` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Valiant Lance D. Dionela, Fatima Kriselle S. Dy, Robin James M. Hombrebueno, et al.
+- **abs:** https://arxiv.org/abs/2601.01827 · **PDF:** https://arxiv.org/pdf/2601.01827v1
+
+## 24209. CSCBench: A PVC Diagnostic Benchmark for Commodity Supply Chain Reasoning
+
+- **ID:** `2601.01825` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Yaxin Cui, Yuanqiang Zeng, Jiapeng Yan, et al.
+- **abs:** https://arxiv.org/abs/2601.01825 · **PDF:** https://arxiv.org/pdf/2601.01825v1
+
+## 24210. Physical Transformer
+
+- **ID:** `2601.02433` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Tao Xu, Zhixin Hu, Li Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.02433 · **PDF:** https://arxiv.org/pdf/2601.02433v1
+
+## 24211. Quantifying Quanvolutional Neural Networks Robustness for Speech in Healthcare Applications
+
+- **ID:** `2601.02432` · **primary:** `cs.SD` · **Submitted:** 2026-01-05
+- **Authors:** Ha Tran, Bipasha Kashyap, Pubudu N. Pathirana
+- **abs:** https://arxiv.org/abs/2601.02432 · **PDF:** https://arxiv.org/pdf/2601.02432v1
+
+## 24212. Admissibility Alignment
+
+- **ID:** `2601.01816` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Chris Duffey
+- **abs:** https://arxiv.org/abs/2601.01816 · **PDF:** https://arxiv.org/pdf/2601.01816v1
+
+## 24213. Spatio-temporal modeling and forecasting with Fourier neural operators
+
+- **ID:** `2601.01813` · **primary:** `stat.ME` · **Submitted:** 2026-01-05
+- **Authors:** Pratik Nag, Andrew Zammit-Mangion, Sumeetpal Singh, et al.
+- **abs:** https://arxiv.org/abs/2601.01813 · **PDF:** https://arxiv.org/pdf/2601.01813v1
+
+## 24214. Attention in Geometry: Scalable Spatial Modeling via Adaptive Density Fields and FAISS-Accelerated Kernels
+
+- **ID:** `2601.06135` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Zhaowen Fan
+- **abs:** https://arxiv.org/abs/2601.06135 · **PDF:** https://arxiv.org/pdf/2601.06135v2
+
+## 24215. Adaptive Hybrid Optimizer based Framework for Lumpy Skin Disease Identification
+
+- **ID:** `2601.01807` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Ubaidullah, Muhammad Abid Hussain, Mohsin Raza Jafri, et al.
+- **abs:** https://arxiv.org/abs/2601.01807 · **PDF:** https://arxiv.org/pdf/2601.01807v1
+
+## 24216. DeeperBrain: A Neuro-Grounded EEG Foundation Model Towards Universal BCI
+
+- **ID:** `2601.06134` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Jiquan Wang, Sha Zhao, Yangxuan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.06134 · **PDF:** https://arxiv.org/pdf/2601.06134v2
+
+## 24217. Moments Matter:Stabilizing Policy Optimization using Return Distributions
+
+- **ID:** `2601.01803` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Dennis Jabs, Aditya Mohan, Marius Lindauer
+- **abs:** https://arxiv.org/abs/2601.01803 · **PDF:** https://arxiv.org/pdf/2601.01803v1
+
+## 24218. PsychEval: A Multi-Session and Multi-Therapy Benchmark for High-Realism AI Psychological Counselor
+
+- **ID:** `2601.01802` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Qianjun Pan, Junyi Wang, Jie Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.01802 · **PDF:** https://arxiv.org/pdf/2601.01802v3
+
+## 24219. WebCoderBench: Benchmarking Web Application Generation with Comprehensive and Interpretable Evaluation Metrics
+
+- **ID:** `2601.02430` · **primary:** `cs.SE` · **Submitted:** 2026-01-05
+- **Authors:** Chenxu Liu, Yingjie Fu, Wei Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.02430 · **PDF:** https://arxiv.org/pdf/2601.02430v2
+
+## 24220. Sparse Threats, Focused Defense: Criticality-Aware Robust Reinforcement Learning for Safe Autonomous Driving
+
+- **ID:** `2601.01800` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Qi Wei, Junchao Fan, Zhao Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.01800 · **PDF:** https://arxiv.org/pdf/2601.01800v1
+
+## 24221. A Review of Online Diffusion Policy RL Algorithms for Scalable Robotic Control
+
+- **ID:** `2601.06133` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Wonhyeok Choi, Shutong Ding, Minwoo Choi, et al.
+- **abs:** https://arxiv.org/abs/2601.06133 · **PDF:** https://arxiv.org/pdf/2601.06133v2
+
+## 24222. VerLM: Explaining Face Verification Using Natural Language
+
+- **ID:** `2601.01798` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Syed Abdul Hannan, Hazim Bukhari, Thomas Cantalapiedra, et al.
+- **abs:** https://arxiv.org/abs/2601.01798 · **PDF:** https://arxiv.org/pdf/2601.01798v1
+
+## 24223. Distributed Federated Learning by Alternating Periods of Training
+
+- **ID:** `2601.01793` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Shamik Bhattacharyya, Rachel Kalpana Kalaimani
+- **abs:** https://arxiv.org/abs/2601.01793 · **PDF:** https://arxiv.org/pdf/2601.01793v1
+
+## 24224. HyperCLOVA X 8B Omni
+
+- **ID:** `2601.01792` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** NAVER Cloud HyperCLOVA X Team
+- **abs:** https://arxiv.org/abs/2601.01792 · **PDF:** https://arxiv.org/pdf/2601.01792v1
+
+## 24225. UnPII: Unlearning Personally Identifiable Information with Quantifiable Exposure Risk
+
+- **ID:** `2601.01786` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Intae Jeon, Yujeong Kwon, Hyungjoon Koo
+- **abs:** https://arxiv.org/abs/2601.01786 · **PDF:** https://arxiv.org/pdf/2601.01786v1
+
+## 24226. SRAS: A Lightweight Reinforcement Learning-based Document Selector for Edge-Native RAG Pipelines
+
+- **ID:** `2601.01785` · **primary:** `cs.IR` · **Submitted:** 2026-01-05
+- **Authors:** Rajiv Chaitanya Muttur
+- **abs:** https://arxiv.org/abs/2601.01785 · **PDF:** https://arxiv.org/pdf/2601.01785v1
+
+## 24227. Subimage Overlap Prediction: Task-Aligned Self-Supervised Pretraining For Semantic Segmentation In Remote Sensing Imagery
+
+- **ID:** `2601.01781` · **primary:** `cs.CV` · **Submitted:** 2026-01-05
+- **Authors:** Lakshay Sharma, Alex Marin
+- **abs:** https://arxiv.org/abs/2601.01781 · **PDF:** https://arxiv.org/pdf/2601.01781v1
+
+## 24228. LIA: Supervised Fine-Tuning of Large Language Models for Automatic Issue Assignment
+
+- **ID:** `2601.01780` · **primary:** `cs.SE` · **Submitted:** 2026-01-05
+- **Authors:** Arsham Khosravani, Alireza Hoseinpour, Arshia Akhavan, et al.
+- **abs:** https://arxiv.org/abs/2601.01780 · **PDF:** https://arxiv.org/pdf/2601.01780v2
+
+## 24229. Machine learning modularity
+
+- **ID:** `2601.01779` · **primary:** `hep-th` · **Submitted:** 2026-01-05
+- **Authors:** Yi Fan, Vishnu Jejjala, Yang Lei
+- **abs:** https://arxiv.org/abs/2601.01779 · **PDF:** https://arxiv.org/pdf/2601.01779v1
+
+## 24230. BanglaIPA: Towards Robust Text-to-IPA Transcription with Contextual Rewriting in Bengali
+
+- **ID:** `2601.01778` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Jakir Hasan, Shrestha Datta, Md Saiful Islam, et al.
+- **abs:** https://arxiv.org/abs/2601.01778 · **PDF:** https://arxiv.org/pdf/2601.01778v2
+
+## 24231. Can Large Language Models Solve Engineering Equations? A Systematic Comparison of Direct Prediction and Solver-Assisted Approaches
+
+- **ID:** `2601.01774` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Sai Varun Kodathala, Rakesh Vunnam
+- **abs:** https://arxiv.org/abs/2601.01774 · **PDF:** https://arxiv.org/pdf/2601.01774v1
+
+## 24232. OrbitFlow: SLO-Aware Long-Context LLM Serving with Fine-Grained KV Cache Reconfiguration
+
+- **ID:** `2601.10729` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Xinyue Ma, Heelim Hong, Taegeon Um, et al.
+- **abs:** https://arxiv.org/abs/2601.10729 · **PDF:** https://arxiv.org/pdf/2601.10729v2
+
+## 24233. Can LLMs Track Their Output Length? A Dynamic Feedback Mechanism for Precise Length Regulation
+
+- **ID:** `2601.01768` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Meiman Xiao, Ante Wang, Qingguo Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.01768 · **PDF:** https://arxiv.org/pdf/2601.01768v2
+
+## 24234. A New Benchmark for the Appropriate Evaluation of RTL Code Optimization
+
+- **ID:** `2601.01765` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Yao Lu, Shang Liu, Hangan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.01765 · **PDF:** https://arxiv.org/pdf/2601.01765v1
+
+## 24235. EPD-Serve: A Flexible Multimodal EPD Disaggregation Inference Serving System On Ascend
+
+- **ID:** `2601.11590` · **primary:** `cs.DC` · **Submitted:** 2026-01-05
+- **Authors:** Fan Bai, Pai Peng, Zhengzhi Tang, et al.
+- **abs:** https://arxiv.org/abs/2601.11590 · **PDF:** https://arxiv.org/pdf/2601.11590v1
+
+## 24236. Sparse Convex Biclustering
+
+- **ID:** `2601.01757` · **primary:** `stat.ML` · **Submitted:** 2026-01-05
+- **Authors:** Jiakun Jiang, Dewei Xiang, Chenliang Gu, et al.
+- **abs:** https://arxiv.org/abs/2601.01757 · **PDF:** https://arxiv.org/pdf/2601.01757v1
+
+## 24237. A Wachspress-based transfinite formulation for exactly enforcing Dirichlet boundary conditions on convex polygonal domains in physics-informed neural networks
+
+- **ID:** `2601.01756` · **primary:** `math.NA` · **Submitted:** 2026-01-05
+- **Authors:** N. Sukumar, Ritwick Roy
+- **abs:** https://arxiv.org/abs/2601.01756 · **PDF:** https://arxiv.org/pdf/2601.01756v2
+
+## 24238. Context-Free Recognition with Transformers
+
+- **ID:** `2601.01754` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Selim Jerad, Anej Svete, Sophie Hao, et al.
+- **abs:** https://arxiv.org/abs/2601.01754 · **PDF:** https://arxiv.org/pdf/2601.01754v2
+
+## 24239. MergeRec: Model Merging for Data-Isolated Cross-Domain Sequential Recommendation
+
+- **ID:** `2601.01753` · **primary:** `cs.IR` · **Submitted:** 2026-01-05
+- **Authors:** Hyunsoo Kim, Jaewan Moon, Seongmin Park, et al.
+- **abs:** https://arxiv.org/abs/2601.01753 · **PDF:** https://arxiv.org/pdf/2601.01753v1
+
+## 24240. Query-Document Dense Vectors for LLM Relevance Judgment Bias Analysis
+
+- **ID:** `2601.01751` · **primary:** `cs.IR` · **Submitted:** 2026-01-05
+- **Authors:** Samaneh Mohtadi, Gianluca Demartini
+- **abs:** https://arxiv.org/abs/2601.01751 · **PDF:** https://arxiv.org/pdf/2601.01751v1
+
+## 24241. Crafting Adversarial Inputs for Large Vision-Language Models Using Black-Box Optimization
+
+- **ID:** `2601.01747` · **primary:** `cs.CR` · **Submitted:** 2026-01-05
+- **Authors:** Jiwei Guan, Haibo Jin, Haohan Wang
+- **abs:** https://arxiv.org/abs/2601.01747 · **PDF:** https://arxiv.org/pdf/2601.01747v4
+
+## 24242. Multi-granularity Interactive Attention Framework for Residual Hierarchical Pronunciation Assessment
+
+- **ID:** `2601.01745` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Hong Han, Hao-Chen Pei, Zhao-Zheng Nie, et al.
+- **abs:** https://arxiv.org/abs/2601.01745 · **PDF:** https://arxiv.org/pdf/2601.01745v1
+
+## 24243. AI Agent Systems: Architectures, Applications, and Evaluation
+
+- **ID:** `2601.01743` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** Bin Xu
+- **abs:** https://arxiv.org/abs/2601.01743 · **PDF:** https://arxiv.org/pdf/2601.01743v1
+
+## 24244. Latent Space Element Method
+
+- **ID:** `2601.01741` · **primary:** `math.DS` · **Submitted:** 2026-01-05
+- **Authors:** Seung Whan Chung, Youngsoo Choi, Christopher Miller, et al.
+- **abs:** https://arxiv.org/abs/2601.01741 · **PDF:** https://arxiv.org/pdf/2601.01741v1
+
+## 24245. K-EXAONE Technical Report
+
+- **ID:** `2601.01739` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Eunbi Choi, Kibong Choi, Seokhee Hong, et al.
+- **abs:** https://arxiv.org/abs/2601.01739 · **PDF:** https://arxiv.org/pdf/2601.01739v2
+
+## 24246. Yuan3.0 Flash: An Open Multimodal Large Language Model for Enterprise Applications
+
+- **ID:** `2601.01718` · **primary:** `cs.AI` · **Submitted:** 2026-01-05
+- **Authors:** YuanLab. ai, :, Shawn Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.01718 · **PDF:** https://arxiv.org/pdf/2601.01718v1
+
+## 24247. Entropy-Aligned Decoding of LMs for Better Writing and Reasoning
+
+- **ID:** `2601.01714` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Kareem Ahmed, Sameer Singh
+- **abs:** https://arxiv.org/abs/2601.01714 · **PDF:** https://arxiv.org/pdf/2601.01714v1
+
+## 24248. RelayGR: Scaling Long-Sequence Generative Recommendation via Cross-Stage Relay-Race Inference
+
+- **ID:** `2601.01712` · **primary:** `cs.DC` · **Submitted:** 2026-01-05
+- **Authors:** Jiarui Wang, Huichao Chai, Yuanhang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.01712 · **PDF:** https://arxiv.org/pdf/2601.01712v1
+
+## 24249. An evaluation of LLMs for political bias in Western media: Israel-Hamas and Ukraine-Russia wars
+
+- **ID:** `2601.06132` · **primary:** `cs.CY` · **Submitted:** 2026-01-05
+- **Authors:** Rohitash Chandra, Haoyan Chen, Yaqing Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.06132 · **PDF:** https://arxiv.org/pdf/2601.06132v1
+
+## 24250. Lightweight Transformer Architectures for Edge Devices in Real-Time Applications
+
+- **ID:** `2601.03290` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Hema Hariharan Samson
+- **abs:** https://arxiv.org/abs/2601.03290 · **PDF:** https://arxiv.org/pdf/2601.03290v1
+
+## 24251. Reinforcement Learning for Option Hedging: Static Implied-Volatility Fit versus Shortfall-Aware Performance
+
+- **ID:** `2601.01709` · **primary:** `q-fin.PR` · **Submitted:** 2026-01-05
+- **Authors:** Ziheng Chen, Minxuan Hu, Jiayu Yi, et al.
+- **abs:** https://arxiv.org/abs/2601.01709 · **PDF:** https://arxiv.org/pdf/2601.01709v1
+
+## 24252. A Training-Free Large Reasoning Model-based Knowledge Tracing Framework for Unified Prediction and Prescription
+
+- **ID:** `2601.01708` · **primary:** `cs.CL` · **Submitted:** 2026-01-05
+- **Authors:** Unggi Lee, Joo Young Kim, Ran Ju, et al.
+- **abs:** https://arxiv.org/abs/2601.01708 · **PDF:** https://arxiv.org/pdf/2601.01708v1
+
+## 24253. Explicit World Models for Reliable Human-Robot Collaboration
+
+- **ID:** `2601.01705` · **primary:** `cs.RO` · **Submitted:** 2026-01-05
+- **Authors:** Kenneth Kwok, Basura Fernando, Qianli Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.01705 · **PDF:** https://arxiv.org/pdf/2601.01705v2
+
+## 24254. Beyond Homophily: Community Search on Heterophilic Graphs
+
+- **ID:** `2601.01703` · **primary:** `cs.SI` · **Submitted:** 2026-01-05
+- **Authors:** Qing Sima, Xiaoyang Wang, Wenjie Zhang
+- **abs:** https://arxiv.org/abs/2601.01703 · **PDF:** https://arxiv.org/pdf/2601.01703v2
+
+## 24255. Digital Twin-Driven Communication-Efficient Federated Anomaly Detection for Industrial IoT
+
+- **ID:** `2601.01701` · **primary:** `cs.LG` · **Submitted:** 2026-01-05
+- **Authors:** Mohammed Ayalew Belay, Adil Rasheed, Pierluigi Salvo Rossi
+- **abs:** https://arxiv.org/abs/2601.01701 · **PDF:** https://arxiv.org/pdf/2601.01701v1
+
+## 24256. Varying-Coefficient Mixture of Experts Model
+
+- **ID:** `2601.01699` · **primary:** `stat.ME` · **Submitted:** 2026-01-05
+- **Authors:** Qicheng Zhao, Celia M. T. Greenwood, Qihuang Zhang
+- **abs:** https://arxiv.org/abs/2601.01699 · **PDF:** https://arxiv.org/pdf/2601.01699v1
+
+## 24257. Hidden costs for inference with deep network on embedded system devices
+
+- **ID:** `2601.01698` · **primary:** `cs.CC` · **Submitted:** 2026-01-05
+- **Authors:** Chankyu Lee, Woohyun Choi, Sangwook Park
+- **abs:** https://arxiv.org/abs/2601.01698 · **PDF:** https://arxiv.org/pdf/2601.01698v1
+
+## 24258. Enhanced Multi-model Online Conformal Prediction
+
+- **ID:** `2601.01692` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Erfan Hajihashemi, Yanning Shen
+- **abs:** https://arxiv.org/abs/2601.01692 · **PDF:** https://arxiv.org/pdf/2601.01692v1
+
+## 24259. DiMEx: Breaking the Cold Start Barrier in Data-Free Model Extraction via Latent Diffusion Priors
+
+- **ID:** `2601.01688` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Yash Thesia, Meera Suthar
+- **abs:** https://arxiv.org/abs/2601.01688 · **PDF:** https://arxiv.org/pdf/2601.01688v2
+
+## 24260. FALCON: Few-Shot Adversarial Learning for Cross-Domain Medical Image Segmentation
+
+- **ID:** `2601.01687` · **primary:** `cs.CV` · **Submitted:** 2026-01-04
+- **Authors:** Abdur R. Fayjie, Pankhi Kashyap, Jutika Borah, et al.
+- **abs:** https://arxiv.org/abs/2601.01687 · **PDF:** https://arxiv.org/pdf/2601.01687v1
+
+## 24261. Lying with Truths: Open-Channel Multi-Agent Collusion for Belief Manipulation via Generative Montage
+
+- **ID:** `2601.01685` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Jinwei Hu, Xinmiao Huang, Youcheng Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.01685 · **PDF:** https://arxiv.org/pdf/2601.01685v1
+
+## 24262. LACONIC: Dense-Level Effectiveness for Scalable Sparse Retrieval via a Two-Phase Training Curriculum
+
+- **ID:** `2601.01684` · **primary:** `cs.IR` · **Submitted:** 2026-01-04
+- **Authors:** Zhichao Xu, Shengyao Zhuang, Crystina Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.01684 · **PDF:** https://arxiv.org/pdf/2601.01684v1
+
+## 24263. Simplex Deep Linear Discriminant Analysis
+
+- **ID:** `2601.01679` · **primary:** `stat.ML` · **Submitted:** 2026-01-04
+- **Authors:** Maxat Tezekbayev, Arman Bolatov, Zhenisbek Assylbekov
+- **abs:** https://arxiv.org/abs/2601.01679 · **PDF:** https://arxiv.org/pdf/2601.01679v2
+
+## 24264. HeurekaBench: A Benchmarking Framework for AI Co-scientist
+
+- **ID:** `2601.01678` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Siba Smarak Panigrahi, Jovana Videnović, Maria Brbić
+- **abs:** https://arxiv.org/abs/2601.01678 · **PDF:** https://arxiv.org/pdf/2601.01678v2
+
+## 24265. A Dynamic Retrieval-Augmented Generation System with Selective Memory and Remembrance
+
+- **ID:** `2601.02428` · **primary:** `cs.IR` · **Submitted:** 2026-01-04
+- **Authors:** Okan Bursa
+- **abs:** https://arxiv.org/abs/2601.02428 · **PDF:** https://arxiv.org/pdf/2601.02428v1
+
+## 24266. Exposing Hidden Interfaces: LLM-Guided Type Inference for Reverse Engineering macOS Private Frameworks
+
+- **ID:** `2601.01673` · **primary:** `cs.CR` · **Submitted:** 2026-01-04
+- **Authors:** Arina Kharlamova, Youcheng Sun, Ting Yu
+- **abs:** https://arxiv.org/abs/2601.01673 · **PDF:** https://arxiv.org/pdf/2601.01673v1
+
+## 24267. Graph-Based Analysis of AI-Driven Labor Market Transitions: Evidence from 10,000 Egyptian Jobs and Policy Implications
+
+- **ID:** `2601.06129` · **primary:** `cs.CY` · **Submitted:** 2026-01-04
+- **Authors:** Ahmed Dawoud, Sondos Samir, Youssef Nasr, et al.
+- **abs:** https://arxiv.org/abs/2601.06129 · **PDF:** https://arxiv.org/pdf/2601.06129v2
+
+## 24268. EHRSummarizer: A Privacy-Aware, FHIR-Native Architecture for Structured Clinical Summarization of Electronic Health Records
+
+- **ID:** `2601.01668` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Houman Kazemzadeh, Nima Minaifar, Kamyar Naderi, et al.
+- **abs:** https://arxiv.org/abs/2601.01668 · **PDF:** https://arxiv.org/pdf/2601.01668v1
+
+## 24269. Adversarial Instance Generation and Robust Training for Neural Combinatorial Optimization with Multiple Objectives
+
+- **ID:** `2601.01665` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Wei Liu, Yaoxin Wu, Yingqian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.01665 · **PDF:** https://arxiv.org/pdf/2601.01665v1
+
+## 24270. Who is the Winning Algorithm? Rank Aggregation for Comparative Studies
+
+- **ID:** `2601.01664` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Amichai Painsky
+- **abs:** https://arxiv.org/abs/2601.01664 · **PDF:** https://arxiv.org/pdf/2601.01664v1
+
+## 24271. Length-Aware Adversarial Training for Variable-Length Trajectories: Digital Twins for Mall Shopper Paths
+
+- **ID:** `2601.01663` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** He Sun, Jiwoong Shin, Ravi Dhar
+- **abs:** https://arxiv.org/abs/2601.01663 · **PDF:** https://arxiv.org/pdf/2601.01663v1
+
+## 24272. A Koopman-Bayesian Framework for High-Fidelity, Perceptually Optimized Haptic Surgical Simulation
+
+- **ID:** `2602.15834` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Rohit Kaushik, Eva Kaushik
+- **abs:** https://arxiv.org/abs/2602.15834 · **PDF:** https://arxiv.org/pdf/2602.15834v1
+
+## 24273. STEMNIST: Spiking Tactile Extended MNIST Neuromorphic Dataset
+
+- **ID:** `2601.01658` · **primary:** `cs.NE` · **Submitted:** 2026-01-04
+- **Authors:** Anubhab Tripathi, Li Gaishan, Zhengnan Fu, et al.
+- **abs:** https://arxiv.org/abs/2601.01658 · **PDF:** https://arxiv.org/pdf/2601.01658v1
+
+## 24274. UniCrop: A Universal, Multi-Source Data Engineering Pipeline for Scalable Crop Yield Prediction
+
+- **ID:** `2601.01655` · **primary:** `eess.IV` · **Submitted:** 2026-01-04
+- **Authors:** Emiliya Khidirova, Oktay Karakuş
+- **abs:** https://arxiv.org/abs/2601.01655 · **PDF:** https://arxiv.org/pdf/2601.01655v1
+
+## 24275. Learning Resilient Elections with Adversarial GNNs
+
+- **ID:** `2601.01653` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Hao Xiang Li, Yash Shah, Lorenzo Giusti
+- **abs:** https://arxiv.org/abs/2601.01653 · **PDF:** https://arxiv.org/pdf/2601.01653v1
+
+## 24276. Communication-Efficient Federated AUC Maximization with Cyclic Client Participation
+
+- **ID:** `2601.01649` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Umesh Vangapally, Wenhan Wu, Chen Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.01649 · **PDF:** https://arxiv.org/pdf/2601.01649v1
+
+## 24277. JMedEthicBench: A Multi-Turn Conversational Benchmark for Evaluating Medical Safety in Japanese Large Language Models
+
+- **ID:** `2601.01627` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Junyu Liu, Zirui Li, Qian Niu, et al.
+- **abs:** https://arxiv.org/abs/2601.01627 · **PDF:** https://arxiv.org/pdf/2601.01627v2
+
+## 24278. LAPS: A Length-Aware-Prefill LLM Serving System
+
+- **ID:** `2601.11589` · **primary:** `cs.DC` · **Submitted:** 2026-01-04
+- **Authors:** Jianshu She, Zonghang Li, Hongchao Du, et al.
+- **abs:** https://arxiv.org/abs/2601.11589 · **PDF:** https://arxiv.org/pdf/2601.11589v2
+
+## 24279. How Does Prefix Matter in Reasoning Model Tuning?
+
+- **ID:** `2601.01624` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Raj Vardhan Tomar, Preslav Nakov, Yuxia Wang
+- **abs:** https://arxiv.org/abs/2601.01624 · **PDF:** https://arxiv.org/pdf/2601.01624v1
+
+## 24280. The Gray Area: Characterizing Moderator Disagreement on Reddit
+
+- **ID:** `2601.01620` · **primary:** `cs.CY` · **Submitted:** 2026-01-04
+- **Authors:** Shayan Alipour, Shruti Phadke, Seyed Shahabeddin Mousavi, et al.
+- **abs:** https://arxiv.org/abs/2601.01620 · **PDF:** https://arxiv.org/pdf/2601.01620v2
+
+## 24281. Deep Linear Discriminant Analysis Revisited
+
+- **ID:** `2601.01619` · **primary:** `stat.ML` · **Submitted:** 2026-01-04
+- **Authors:** Maxat Tezekbayev, Rustem Takhanov, Arman Bolatov, et al.
+- **abs:** https://arxiv.org/abs/2601.01619 · **PDF:** https://arxiv.org/pdf/2601.01619v1
+
+## 24282. Real Time NILM Based Power Monitoring of Identical Induction Motors Representing Cutting Machines in Textile Industry
+
+- **ID:** `2601.01616` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Md Istiauk Hossain Rifat, Moin Khan, Mohammad Zunaed
+- **abs:** https://arxiv.org/abs/2601.01616 · **PDF:** https://arxiv.org/pdf/2601.01616v1
+
+## 24283. Structured Decomposition for LLM Reasoning: Cross-Domain Validation and Semantic Web Integration
+
+- **ID:** `2601.01609` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Albert Sadowski, Jarosław A. Chudziak
+- **abs:** https://arxiv.org/abs/2601.01609 · **PDF:** https://arxiv.org/pdf/2601.01609v1
+
+## 24284. REE-TTT: Highly Adaptive Radar Echo Extrapolation Based on Test-Time Training
+
+- **ID:** `2601.01605` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Xin Di, Xinglin Piao, Fei Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.01605 · **PDF:** https://arxiv.org/pdf/2601.01605v1
+
+## 24285. grangersearch: An R Package for Exhaustive Granger Causality Testing with Tidyverse Integration
+
+- **ID:** `2601.01604` · **primary:** `stat.CO` · **Submitted:** 2026-01-04
+- **Authors:** Nikolaos Korfiatis
+- **abs:** https://arxiv.org/abs/2601.01604 · **PDF:** https://arxiv.org/pdf/2601.01604v1
+
+## 24286. From Theory of Mind to Theory of Environment: Counterfactual Simulation of Latent Environmental Dynamics
+
+- **ID:** `2601.01599` · **primary:** `q-bio.NC` · **Submitted:** 2026-01-04
+- **Authors:** Ryutaro Uchiyama
+- **abs:** https://arxiv.org/abs/2601.01599 · **PDF:** https://arxiv.org/pdf/2601.01599v1
+
+## 24287. Variance-Reduced Diffusion Sampling via Target Score Identity
+
+- **ID:** `2601.01594` · **primary:** `stat.ML` · **Submitted:** 2026-01-04
+- **Authors:** Alois Duston, Tan Bui-Thanh
+- **abs:** https://arxiv.org/abs/2601.01594 · **PDF:** https://arxiv.org/pdf/2601.01594v3
+
+## 24288. SAGE-32B: Agentic Reasoning via Iterative Distillation
+
+- **ID:** `2601.04237` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Basab Jha, Firoj Paudel, Ujjwal Puri, et al.
+- **abs:** https://arxiv.org/abs/2601.04237 · **PDF:** https://arxiv.org/pdf/2601.04237v1
+
+## 24289. Identifying recurrent flows in high-dimensional dissipative chaos from low-dimensional embeddings
+
+- **ID:** `2601.01590` · **primary:** `nlin.CD` · **Submitted:** 2026-01-04
+- **Authors:** Pierre Beck, Tobias M. Schneider
+- **abs:** https://arxiv.org/abs/2601.01590 · **PDF:** https://arxiv.org/pdf/2601.01590v1
+
+## 24290. Learning Relationship between Quantum Walks and Underdamped Langevin Dynamics
+
+- **ID:** `2601.01589` · **primary:** `quant-ph` · **Submitted:** 2026-01-04
+- **Authors:** Yazhen Wang
+- **abs:** https://arxiv.org/abs/2601.01589 · **PDF:** https://arxiv.org/pdf/2601.01589v1
+
+## 24291. NitroGen: An Open Foundation Model for Generalist Gaming Agents
+
+- **ID:** `2601.02427` · **primary:** `cs.CV` · **Submitted:** 2026-01-04
+- **Authors:** Loïc Magne, Anas Awadalla, Guanzhi Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.02427 · **PDF:** https://arxiv.org/pdf/2601.02427v1
+
+## 24292. Steerability of Instrumental-Convergence Tendencies in LLMs
+
+- **ID:** `2601.01584` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Jakub Hoscilowicz
+- **abs:** https://arxiv.org/abs/2601.01584 · **PDF:** https://arxiv.org/pdf/2601.01584v2
+
+## 24293. CONSENT: A Negotiation Framework for Leveraging User Flexibility in Vehicle-to-Building Charging under Uncertainty
+
+- **ID:** `2601.01581` · **primary:** `cs.MA` · **Submitted:** 2026-01-04
+- **Authors:** Rishav Sen, Fangqi Liu, Jose Paolo Talusan, et al.
+- **abs:** https://arxiv.org/abs/2601.01581 · **PDF:** https://arxiv.org/pdf/2601.01581v3
+
+## 24294. The Two-Stage Decision-Sampling Hypothesis: Understanding the Emergence of Self-Reflection in RL-Trained LLMs
+
+- **ID:** `2601.01580` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Zibo Zhao, Yuanting Zha, Haipeng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.01580 · **PDF:** https://arxiv.org/pdf/2601.01580v1
+
+## 24295. HanoiWorld : A Joint Embedding Predictive Architecture BasedWorld Model for Autonomous Vehicle Controller
+
+- **ID:** `2601.01577` · **primary:** `cs.RO` · **Submitted:** 2026-01-04
+- **Authors:** Tran Tien Dat, Nguyen Hai An, Nguyen Khanh Viet Dung, et al.
+- **abs:** https://arxiv.org/abs/2601.01577 · **PDF:** https://arxiv.org/pdf/2601.01577v1
+
+## 24296. OpenNovelty: An LLM-powered Agentic System for Verifiable Scholarly Novelty Assessment
+
+- **ID:** `2601.01576` · **primary:** `cs.IR` · **Submitted:** 2026-01-04
+- **Authors:** Ming Zhang, Kexin Tan, Yueyuan Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.01576 · **PDF:** https://arxiv.org/pdf/2601.01576v2
+
+## 24297. CaveAgent: Transforming LLMs into Stateful Runtime Operators
+
+- **ID:** `2601.01569` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Maohao Ran, Zhenglin Wan, Cooper Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.01569 · **PDF:** https://arxiv.org/pdf/2601.01569v3
+
+## 24298. MM-Sonate: Multimodal Controllable Audio-Video Generation with Zero-Shot Voice Cloning
+
+- **ID:** `2601.01568` · **primary:** `cs.SD` · **Submitted:** 2026-01-04
+- **Authors:** Chunyu Qiang, Jun Wang, Xiaopeng Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.01568 · **PDF:** https://arxiv.org/pdf/2601.01568v2
+
+## 24299. Logics-STEM: Empowering LLM Reasoning via Failure-Driven Post-Training and Document Knowledge Enhancement
+
+- **ID:** `2601.01562` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Mingyu Xu, Cheng Fang, Keyue Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.01562 · **PDF:** https://arxiv.org/pdf/2601.01562v3
+
+## 24300. Utilizing Earth Foundation Models to Enhance the Simulation Performance of Hydrological Models with AlphaEarth Embeddings
+
+- **ID:** `2601.01558` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Pengfei Qu, Wenyu Ouyang, Chi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.01558 · **PDF:** https://arxiv.org/pdf/2601.01558v1
+
+## 24301. AIS-CycleGen: A CycleGAN-Based Framework for High-Fidelity Synthetic AIS Data Generation and Augmentation
+
+- **ID:** `2601.06127` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** SM Ashfaq uz Zaman, Faizan Qamar, Masnizah Mohd, et al.
+- **abs:** https://arxiv.org/abs/2601.06127 · **PDF:** https://arxiv.org/pdf/2601.06127v1
+
+## 24302. MOSS Transcribe Diarize Technical Report
+
+- **ID:** `2601.01554` · **primary:** `cs.SD` · **Submitted:** 2026-01-04
+- **Authors:** MOSI. AI, :, Donghua Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.01554 · **PDF:** https://arxiv.org/pdf/2601.01554v5
+
+## 24303. HalluZig: Hallucination Detection using Zigzag Persistence
+
+- **ID:** `2601.01552` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Shreyas N. Samaga, Gilberto Gonzalez Arroyo, Tamal K. Dey
+- **abs:** https://arxiv.org/abs/2601.01552 · **PDF:** https://arxiv.org/pdf/2601.01552v2
+
+## 24304. Vision-language models lag human performance on physical dynamics and intent reasoning
+
+- **ID:** `2601.01547` · **primary:** `cs.CV` · **Submitted:** 2026-01-04
+- **Authors:** Tianjun Gu, Jingyu Gong, Zhizhong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.01547 · **PDF:** https://arxiv.org/pdf/2601.01547v2
+
+## 24305. Improving Behavioral Alignment in LLM Social Simulations via Context Formation and Navigation
+
+- **ID:** `2601.01546` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Letian Kong, Qianran, Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.01546 · **PDF:** https://arxiv.org/pdf/2601.01546v1
+
+## 24306. Bridging the Data Gap: Creating a Hindi Text Summarization Dataset from the English XSUM
+
+- **ID:** `2601.01543` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Praveenkumar Katwe, RakeshChandra Balabantaray, Kaliprasad Vittala
+- **abs:** https://arxiv.org/abs/2601.01543 · **PDF:** https://arxiv.org/pdf/2601.01543v1
+
+## 24307. Aletheia: Quantifying Cognitive Conviction in Reasoning Models via Regularized Inverse Confusion Matrix
+
+- **ID:** `2601.01532` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Fanzhe Fu
+- **abs:** https://arxiv.org/abs/2601.01532 · **PDF:** https://arxiv.org/pdf/2601.01532v1
+
+## 24308. EmoHarbor: Evaluating Personalized Emotional Support by Simulating the User's Internal World
+
+- **ID:** `2601.01530` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Jing Ye, Lu Xiang, Yaping Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.01530 · **PDF:** https://arxiv.org/pdf/2601.01530v1
+
+## 24309. DrivingGen: A Comprehensive Benchmark for Generative Video World Models in Autonomous Driving
+
+- **ID:** `2601.01528` · **primary:** `cs.CV` · **Submitted:** 2026-01-04
+- **Authors:** Yang Zhou, Hao Shao, Letian Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.01528 · **PDF:** https://arxiv.org/pdf/2601.01528v2
+
+## 24310. NL2Dashboard: A Lightweight and Controllable Framework for Generating Dashboards with LLMs
+
+- **ID:** `2601.06126` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Boshen Shi, Kexin Yang, Yuanbo Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.06126 · **PDF:** https://arxiv.org/pdf/2601.06126v1
+
+## 24311. Bayesian Orchestration of Multi-LLM Agents for Cost-Aware Sequential Decision-Making
+
+- **ID:** `2601.01522` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Danial Amin
+- **abs:** https://arxiv.org/abs/2601.01522 · **PDF:** https://arxiv.org/pdf/2601.01522v1
+
+## 24312. Signal in the Noise: Decoding the Reality of Airline Service Quality with Large Language Models
+
+- **ID:** `2603.04404` · **primary:** `cs.IR` · **Submitted:** 2026-01-04
+- **Authors:** Ahmed Dawoud, Osama El-Shamy, Ahmed Habashy
+- **abs:** https://arxiv.org/abs/2603.04404 · **PDF:** https://arxiv.org/pdf/2603.04404v1
+
+## 24313. FastV-RAG: Towards Fast and Fine-Grained Video QA with Retrieval-Augmented Generation
+
+- **ID:** `2601.01513` · **primary:** `cs.CV` · **Submitted:** 2026-01-04
+- **Authors:** Gen Li, Peiyu Liu
+- **abs:** https://arxiv.org/abs/2601.01513 · **PDF:** https://arxiv.org/pdf/2601.01513v2
+
+## 24314. A Novel Deep Learning Method for Segmenting the Left Ventricle in Cardiac Cine MRI
+
+- **ID:** `2601.01512` · **primary:** `cs.CV` · **Submitted:** 2026-01-04
+- **Authors:** Wenhui Chu, Aobo Jin, Hardik A. Gohel
+- **abs:** https://arxiv.org/abs/2601.01512 · **PDF:** https://arxiv.org/pdf/2601.01512v1
+
+## 24315. Reading Between the Lines: Deconfounding Causal Estimates using Text Embeddings and Deep Learning
+
+- **ID:** `2601.01511` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Ahmed Dawoud, Osama El-Shamy
+- **abs:** https://arxiv.org/abs/2601.01511 · **PDF:** https://arxiv.org/pdf/2601.01511v1
+
+## 24316. Multiscale replay: A robust algorithm for stochastic variational inequalities with a Markovian buffer
+
+- **ID:** `2601.01502` · **primary:** `math.OC` · **Submitted:** 2026-01-04
+- **Authors:** Milind Nakul, Tianjiao Li, Ashwin Pananjady
+- **abs:** https://arxiv.org/abs/2601.01502 · **PDF:** https://arxiv.org/pdf/2601.01502v1
+
+## 24317. Advanced Global Wildfire Activity Modeling with Hierarchical Graph ODE
+
+- **ID:** `2601.01501` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Fan Xu, Wei Gong, Hao Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.01501 · **PDF:** https://arxiv.org/pdf/2601.01501v1
+
+## 24318. From Failure to Mastery: Generating Hard Samples for Tool-use Agents
+
+- **ID:** `2601.01498` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Bingguang Hao, Zengzhuang Xu, Yuntao Wen, et al.
+- **abs:** https://arxiv.org/abs/2601.01498 · **PDF:** https://arxiv.org/pdf/2601.01498v1
+
+## 24319. On the Practical Estimation and Interpretation of Rényi Transfer Entropy
+
+- **ID:** `2601.01497` · **primary:** `nlin.PS` · **Submitted:** 2026-01-04
+- **Authors:** Zlata Tabachová, Petr Jizba, Hynek Lavička, et al.
+- **abs:** https://arxiv.org/abs/2601.01497 · **PDF:** https://arxiv.org/pdf/2601.01497v1
+
+## 24320. The Optimal Sample Complexity of Linear Contracts
+
+- **ID:** `2601.01496` · **primary:** `cs.GT` · **Submitted:** 2026-01-04
+- **Authors:** Mikael Møller Høgsgaard
+- **abs:** https://arxiv.org/abs/2601.01496 · **PDF:** https://arxiv.org/pdf/2601.01496v1
+
+## 24321. Accelerating Decentralized Optimization via Overlapping Local Steps
+
+- **ID:** `2601.01493` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Yijie Zhou, Shi Pu
+- **abs:** https://arxiv.org/abs/2601.01493 · **PDF:** https://arxiv.org/pdf/2601.01493v1
+
+## 24322. Distortion Instead of Hallucination: The Effect of Reasoning Under Strict Constraints
+
+- **ID:** `2601.01490` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Junichiro Niimi
+- **abs:** https://arxiv.org/abs/2601.01490 · **PDF:** https://arxiv.org/pdf/2601.01490v1
+
+## 24323. Four Quadrants of Difficulty: A Simple Categorisation and its Limits
+
+- **ID:** `2601.01488` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Vanessa Toborek, Sebastian Müller, Christian Bauckhage
+- **abs:** https://arxiv.org/abs/2601.01488 · **PDF:** https://arxiv.org/pdf/2601.01488v1
+
+## 24324. DeepInv: A Novel Self-supervised Learning Approach for Fast and Accurate Diffusion Inversion
+
+- **ID:** `2601.01487` · **primary:** `cs.CV` · **Submitted:** 2026-01-04
+- **Authors:** Ziyue Zhang, Luxi Lin, Xiaolin Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.01487 · **PDF:** https://arxiv.org/pdf/2601.01487v1
+
+## 24325. SGD-Based Knowledge Distillation with Bayesian Teachers: Theory and Guidelines
+
+- **ID:** `2601.01484` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Itai Morad, Nir Shlezinger, Yonina C. Eldar
+- **abs:** https://arxiv.org/abs/2601.01484 · **PDF:** https://arxiv.org/pdf/2601.01484v1
+
+## 24326. Modeling Information Blackouts in Missing Not-At-Random Time Series Data
+
+- **ID:** `2601.01480` · **primary:** `stat.ML` · **Submitted:** 2026-01-04
+- **Authors:** Aman Sunesh, Allan Ma, Siddarth Nilol
+- **abs:** https://arxiv.org/abs/2601.01480 · **PDF:** https://arxiv.org/pdf/2601.01480v2
+
+## 24327. Can Legislation Be Made Machine-Readable in PROLEG?
+
+- **ID:** `2601.01477` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** May-Myo Zin, Sabine Wehnert, Yuntao Kong, et al.
+- **abs:** https://arxiv.org/abs/2601.01477 · **PDF:** https://arxiv.org/pdf/2601.01477v1
+
+## 24328. Multi-Subspace Multi-Modal Modeling for Diffusion Models: Estimation, Convergence and Mixture of Experts
+
+- **ID:** `2601.01475` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Ruofeng Yang, Yongcan Li, Bo Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.01475 · **PDF:** https://arxiv.org/pdf/2601.01475v1
+
+## 24329. Accelerating Storage-Based Training for Graph Neural Networks
+
+- **ID:** `2601.01473` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Myung-Hwan Jang, Jeong-Min Park, Yunyong Ko, et al.
+- **abs:** https://arxiv.org/abs/2601.01473 · **PDF:** https://arxiv.org/pdf/2601.01473v2
+
+## 24330. Double Machine Learning of Continuous Treatment Effects with General Instrumental Variables
+
+- **ID:** `2601.01471` · **primary:** `math.ST` · **Submitted:** 2026-01-04
+- **Authors:** Shuyuan Chen, Peng Zhang, Yifan Cui
+- **abs:** https://arxiv.org/abs/2601.01471 · **PDF:** https://arxiv.org/pdf/2601.01471v1
+
+## 24331. A construction of an optimal base for conditional attribute and attributional condition implications in triadic contexts
+
+- **ID:** `2601.01467` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Romuald Kwessy Mouona, Blaise Blériot Koguep Njionou, Etienne Romuald Temgoua Alomo, et al.
+- **abs:** https://arxiv.org/abs/2601.01467 · **PDF:** https://arxiv.org/pdf/2601.01467v1
+
+## 24332. Leveraging Flatness to Improve Information-Theoretic Generalization Bounds for SGD
+
+- **ID:** `2601.01465` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Ze Peng, Jian Zhang, Yisen Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.01465 · **PDF:** https://arxiv.org/pdf/2601.01465v1
+
+## 24333. Bridging the gap: A comparative exploration of Speech-LLM and end-to-end architecture for multilingual conversational ASR
+
+- **ID:** `2601.01461` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Yuxiang Mei, Dongxing Xu, Jiaen Liang, et al.
+- **abs:** https://arxiv.org/abs/2601.01461 · **PDF:** https://arxiv.org/pdf/2601.01461v3
+
+## 24334. Learning Minimally-Congested Drive Times from Sparse Open Networks: A Lightweight RF-Based Estimator for Urban Roadway Operations
+
+- **ID:** `2601.06124` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Adewumi Augustine Adepitan, Christopher J. Haruna, Morayo Ogunsina, et al.
+- **abs:** https://arxiv.org/abs/2601.06124 · **PDF:** https://arxiv.org/pdf/2601.06124v1
+
+## 24335. Rethinking Multimodal Few-Shot 3D Point Cloud Segmentation: From Fused Refinement to Decoupled Arbitration
+
+- **ID:** `2601.01456` · **primary:** `cs.CV` · **Submitted:** 2026-01-04
+- **Authors:** Wentao Bian, Fenglei Xu
+- **abs:** https://arxiv.org/abs/2601.01456 · **PDF:** https://arxiv.org/pdf/2601.01456v1
+
+## 24336. SmoothSync: Dual-Stream Diffusion Transformers for Jitter-Robust Beat-Synchronized Gesture Generation from Quantized Audio
+
+- **ID:** `2601.04236` · **primary:** `cs.SD` · **Submitted:** 2026-01-04
+- **Authors:** Yujiao Jiang, Qingmin Liao, Zongqing Lu
+- **abs:** https://arxiv.org/abs/2601.04236 · **PDF:** https://arxiv.org/pdf/2601.04236v1
+
+## 24337. Actively Obtaining Environmental Feedback for Autonomous Action Evaluation Without Predefined Measurements
+
+- **ID:** `2601.04235` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Hong Su
+- **abs:** https://arxiv.org/abs/2601.04235 · **PDF:** https://arxiv.org/pdf/2601.04235v1
+
+## 24338. Robust and Efficient Zeroth-Order LLM Fine-Tuning via Adaptive Bayesian Subspace Optimizer
+
+- **ID:** `2601.01452` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Jian Feng, Zhihong Huang
+- **abs:** https://arxiv.org/abs/2601.01452 · **PDF:** https://arxiv.org/pdf/2601.01452v4
+
+## 24339. Segmentation and Processing of German Court Decisions from Open Legal Data
+
+- **ID:** `2601.01449` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Harshil Darji, Martin Heckelmann, Christina Kratsch, et al.
+- **abs:** https://arxiv.org/abs/2601.01449 · **PDF:** https://arxiv.org/pdf/2601.01449v1
+
+## 24340. iFlip: Iterative Feedback-driven Counterfactual Example Refinement
+
+- **ID:** `2601.01446` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Yilong Wang, Qianli Wang, Nils Feldhus
+- **abs:** https://arxiv.org/abs/2601.01446 · **PDF:** https://arxiv.org/pdf/2601.01446v1
+
+## 24341. Fast Gibbs Sampling on Bayesian Hidden Markov Model with Missing Observations
+
+- **ID:** `2601.01442` · **primary:** `stat.ML` · **Submitted:** 2026-01-04
+- **Authors:** Dongrong Li, Tianwei Yu, Xiaodan Fan
+- **abs:** https://arxiv.org/abs/2601.01442 · **PDF:** https://arxiv.org/pdf/2601.01442v1
+
+## 24342. Online Estimation and Manipulation of Articulated Objects
+
+- **ID:** `2601.01438` · **primary:** `cs.RO` · **Submitted:** 2026-01-04
+- **Authors:** Russell Buchanan, Adrian Röfer, João Moura, et al.
+- **abs:** https://arxiv.org/abs/2601.01438 · **PDF:** https://arxiv.org/pdf/2601.01438v1
+
+## 24343. Personalizing black-box models for nonparametric regression with minimax optimality
+
+- **ID:** `2601.01432` · **primary:** `stat.ME` · **Submitted:** 2026-01-04
+- **Authors:** Sai Li, Linjun Zhang
+- **abs:** https://arxiv.org/abs/2601.01432 · **PDF:** https://arxiv.org/pdf/2601.01432v1
+
+## 24344. SWE-Lego: Pushing the Limits of Supervised Fine-tuning for Software Issue Resolving
+
+- **ID:** `2601.01426` · **primary:** `cs.SE` · **Submitted:** 2026-01-04
+- **Authors:** Chaofan Tao, Jierun Chen, Yuxin Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.01426 · **PDF:** https://arxiv.org/pdf/2601.01426v2
+
+## 24345. Unveiling the Heart-Brain Connection: An Analysis of ECG in Cognitive Performance
+
+- **ID:** `2601.01424` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Akshay Sasi, Malavika Pradeep, Nusaibah Farrukh, et al.
+- **abs:** https://arxiv.org/abs/2601.01424 · **PDF:** https://arxiv.org/pdf/2601.01424v1
+
+## 24346. Formal Analysis of AGI Decision-Theoretic Models and the Confrontation Question
+
+- **ID:** `2601.04234` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Denis Saklakov
+- **abs:** https://arxiv.org/abs/2601.04234 · **PDF:** https://arxiv.org/pdf/2601.04234v1
+
+## 24347. How Real is Your Jailbreak? Fine-grained Jailbreak Evaluation with Anchored Reference
+
+- **ID:** `2601.03288` · **primary:** `cs.CR` · **Submitted:** 2026-01-04
+- **Authors:** Songyang Liu, Chaozhuo Li, Rui Pu, et al.
+- **abs:** https://arxiv.org/abs/2601.03288 · **PDF:** https://arxiv.org/pdf/2601.03288v1
+
+## 24348. A Depth Hierarchy for Computing the Maximum in ReLU Networks via Extremal Graph Theory
+
+- **ID:** `2601.01417` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Itay Safran
+- **abs:** https://arxiv.org/abs/2601.01417 · **PDF:** https://arxiv.org/pdf/2601.01417v1
+
+## 24349. Reliable Grid Forecasting: State Space Models for Safety-Critical Energy Systems
+
+- **ID:** `2601.01410` · **primary:** `eess.SY` · **Submitted:** 2026-01-04
+- **Authors:** Sunki Hong, Jisoo Lee
+- **abs:** https://arxiv.org/abs/2601.01410 · **PDF:** https://arxiv.org/pdf/2601.01410v6
+
+## 24350. A large-scale nanocrystal database with aligned synthesis and properties enabling generative inverse design
+
+- **ID:** `2601.02424` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2026-01-04
+- **Authors:** Kai Gu, Yingping Liang, Senliang Peng, et al.
+- **abs:** https://arxiv.org/abs/2601.02424 · **PDF:** https://arxiv.org/pdf/2601.02424v1
+
+## 24351. From Emotion Classification to Emotional Reasoning: Enhancing Emotional Intelligence in Large Language Models
+
+- **ID:** `2601.01407` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Arjhun Sreedar, Rohan Pillay, Laukik Patade
+- **abs:** https://arxiv.org/abs/2601.01407 · **PDF:** https://arxiv.org/pdf/2601.01407v1
+
+## 24352. SwinIFS: Landmark Guided Swin Transformer For Identity Preserving Face Super Resolution
+
+- **ID:** `2601.01406` · **primary:** `cs.CV` · **Submitted:** 2026-01-04
+- **Authors:** Habiba Kausar, Saeed Anwar, Omar Jamal Hammad, et al.
+- **abs:** https://arxiv.org/abs/2601.01406 · **PDF:** https://arxiv.org/pdf/2601.01406v1
+
+## 24353. Efficient Cover Construction for Ball Mapper via Accelerated Range Queries
+
+- **ID:** `2601.01405` · **primary:** `cs.CG` · **Submitted:** 2026-01-04
+- **Authors:** Jay-Anne Bulauan, John Rick Manzanares
+- **abs:** https://arxiv.org/abs/2601.01405 · **PDF:** https://arxiv.org/pdf/2601.01405v1
+
+## 24354. A Graph-based Framework for Online Time Series Anomaly Detection Using Model Ensemble
+
+- **ID:** `2601.01403` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Zewei Yu, Jianqiu Xu, Caimin Li
+- **abs:** https://arxiv.org/abs/2601.01403 · **PDF:** https://arxiv.org/pdf/2601.01403v1
+
+## 24355. LANCET: Neural Intervention via Structural Entropy for Mitigating Faithfulness Hallucinations in LLMs
+
+- **ID:** `2601.01401` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Chenxu Wang, Chaozhuo Li, Pengbo Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.01401 · **PDF:** https://arxiv.org/pdf/2601.01401v1
+
+## 24356. EternalMath: A Living Benchmark of Frontier Mathematics that Evolves with Human Discovery
+
+- **ID:** `2601.01400` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Jicheng Ma, Guohua Wang, Xinhua Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.01400 · **PDF:** https://arxiv.org/pdf/2601.01400v1
+
+## 24357. SAFE-QAQ: End-to-End Slow-Thinking Audio-Text Fraud Detection via Reinforcement Learning
+
+- **ID:** `2601.01392` · **primary:** `cs.SD` · **Submitted:** 2026-01-04
+- **Authors:** Peidong Wang, Zhiming Ma, Xin Dai, et al.
+- **abs:** https://arxiv.org/abs/2601.01392 · **PDF:** https://arxiv.org/pdf/2601.01392v1
+
+## 24358. Bayesian Negative Binomial Regression of Afrobeats Chart Persistence
+
+- **ID:** `2601.01391` · **primary:** `eess.AS` · **Submitted:** 2026-01-04
+- **Authors:** Ian Jacob Cabansag, Paul Ntegeka
+- **abs:** https://arxiv.org/abs/2601.01391 · **PDF:** https://arxiv.org/pdf/2601.01391v1
+
+## 24359. Scale-Adaptive Power Flow Analysis with Local Topology Slicing and Multi-Task Graph Learning
+
+- **ID:** `2601.01387` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Yongzhe Li, Lin Guan, Zihan Cai, et al.
+- **abs:** https://arxiv.org/abs/2601.01387 · **PDF:** https://arxiv.org/pdf/2601.01387v1
+
+## 24360. ParkGaussian: Surround-view 3D Gaussian Splatting for Autonomous Parking
+
+- **ID:** `2601.01386` · **primary:** `cs.CV` · **Submitted:** 2026-01-04
+- **Authors:** Xiaobao Wei, Zhangjie Ye, Yuxiang Gu, et al.
+- **abs:** https://arxiv.org/abs/2601.01386 · **PDF:** https://arxiv.org/pdf/2601.01386v1
+
+## 24361. Data Complexity-aware Deep Model Performance Forecasting
+
+- **ID:** `2601.01383` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Yen-Chia Chen, Hsing-Kuo Pao, Hanjuan Huang
+- **abs:** https://arxiv.org/abs/2601.01383 · **PDF:** https://arxiv.org/pdf/2601.01383v1
+
+## 24362. Empowering Small Language Models with Factual Hallucination-Aware Reasoning for Financial Classification
+
+- **ID:** `2601.01378` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Han Yuan, Yilin Wu, Li Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.01378 · **PDF:** https://arxiv.org/pdf/2601.01378v1
+
+## 24363. UltraEval-Audio: A Unified Framework for Comprehensive Evaluation of Audio Foundation Models
+
+- **ID:** `2601.01373` · **primary:** `cs.SD` · **Submitted:** 2026-01-04
+- **Authors:** Qundong Shi, Jie Zhou, Biyuan Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.01373 · **PDF:** https://arxiv.org/pdf/2601.01373v1
+
+## 24364. SGD with Dependent Data: Optimal Estimation, Regret, and Inference
+
+- **ID:** `2601.01371` · **primary:** `math.ST` · **Submitted:** 2026-01-04
+- **Authors:** Yinan Shen, Yichen Zhang, Wen-Xin Zhou
+- **abs:** https://arxiv.org/abs/2601.01371 · **PDF:** https://arxiv.org/pdf/2601.01371v1
+
+## 24365. Causal discovery for linear causal model with correlated noise: an Adversarial Learning Approach
+
+- **ID:** `2601.01368` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Mujin Zhou, Junzhe Zhang
+- **abs:** https://arxiv.org/abs/2601.01368 · **PDF:** https://arxiv.org/pdf/2601.01368v1
+
+## 24366. KGCE: Knowledge-Augmented Dual-Graph Evaluator for Cross-Platform Educational Agent Benchmarking with Multimodal Language Models
+
+- **ID:** `2601.01366` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Zixian Liu, Sihao Liu, Yuqi Zhao
+- **abs:** https://arxiv.org/abs/2601.01366 · **PDF:** https://arxiv.org/pdf/2601.01366v1
+
+## 24367. A unified multimodal understanding and generation model for cross-disciplinary scientific research
+
+- **ID:** `2601.01363` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Xiaomeng Yang, Zhiyu Tan, Xiaohui Zhong, et al.
+- **abs:** https://arxiv.org/abs/2601.01363 · **PDF:** https://arxiv.org/pdf/2601.01363v1
+
+## 24368. Investigating the Multilingual Calibration Effects of Language Model Instruction-Tuning
+
+- **ID:** `2601.01362` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Jerry Huang, Peng Lu, Qiuhao Zeng, et al.
+- **abs:** https://arxiv.org/abs/2601.01362 · **PDF:** https://arxiv.org/pdf/2601.01362v1
+
+## 24369. Latent Space Communication via K-V Cache Alignment
+
+- **ID:** `2601.06123` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Lucio M. Dery, Zohar Yahav, Henry Prior, et al.
+- **abs:** https://arxiv.org/abs/2601.06123 · **PDF:** https://arxiv.org/pdf/2601.06123v1
+
+## 24370. A New Framework for Explainable Rare Cell Identification in Single-Cell Transcriptomics Data
+
+- **ID:** `2601.01358` · **primary:** `q-bio.GN` · **Submitted:** 2026-01-04
+- **Authors:** Di Su, Kai Ming Ting, Jie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.01358 · **PDF:** https://arxiv.org/pdf/2601.01358v1
+
+## 24371. Towards LLM-enabled autonomous combustion research: A literature-aware agent for self-corrective modeling workflows
+
+- **ID:** `2601.01357` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Ke Xiao, Haoze Zhang, Runze Mao, et al.
+- **abs:** https://arxiv.org/abs/2601.01357 · **PDF:** https://arxiv.org/pdf/2601.01357v1
+
+## 24372. COVR:Collaborative Optimization of VLMs and RL Agent for Visual-Based Control
+
+- **ID:** `2601.06122` · **primary:** `cs.CV` · **Submitted:** 2026-01-04
+- **Authors:** Canming Xia, Peixi Peng, Guang Tan, et al.
+- **abs:** https://arxiv.org/abs/2601.06122 · **PDF:** https://arxiv.org/pdf/2601.06122v1
+
+## 24373. Slot-ID: Identity-Preserving Video Generation from Reference Videos via Slot-Based Temporal Identity Encoding
+
+- **ID:** `2601.01352` · **primary:** `cs.CV` · **Submitted:** 2026-01-04
+- **Authors:** Yixuan Lai, He Wang, Kun Zhou, et al.
+- **abs:** https://arxiv.org/abs/2601.01352 · **PDF:** https://arxiv.org/pdf/2601.01352v1
+
+## 24374. FC-CONAN: An Exhaustively Paired Dataset for Robust Evaluation of Retrieval Systems
+
+- **ID:** `2601.01350` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Juan Junqueras, Florian Boudin, May-Myo Zin, et al.
+- **abs:** https://arxiv.org/abs/2601.01350 · **PDF:** https://arxiv.org/pdf/2601.01350v1
+
+## 24375. From Classification to Generation: An Open-Ended Paradigm for Adverse Drug Reaction Prediction Based on Graph-Motif Feature Fusion
+
+- **ID:** `2601.01347` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Yuyan Pi, Min Jin, Wentao Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.01347 · **PDF:** https://arxiv.org/pdf/2601.01347v1
+
+## 24376. Reasoning Over Recall: Evaluating the Efficacy of Generalist Architectures vs. Specialized Fine-Tunes in RAG-Based Mental Health Dialogue Systems
+
+- **ID:** `2601.01341` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Md Abdullah Al Kafi, Raka Moni, Sumit Kumar Banshal
+- **abs:** https://arxiv.org/abs/2601.01341 · **PDF:** https://arxiv.org/pdf/2601.01341v1
+
+## 24377. Watch Wider and Think Deeper: Collaborative Cross-modal Chain-of-Thought for Complex Visual Reasoning
+
+- **ID:** `2601.02422` · **primary:** `cs.CV` · **Submitted:** 2026-01-04
+- **Authors:** Wenting Lu, Didi Zhu, Tao Shen, et al.
+- **abs:** https://arxiv.org/abs/2601.02422 · **PDF:** https://arxiv.org/pdf/2601.02422v1
+
+## 24378. FLOP-Efficient Training: Early Stopping Based on Test-Time Compute Awareness
+
+- **ID:** `2601.01332` · **primary:** `cs.CL` · **Submitted:** 2026-01-04
+- **Authors:** Hossam Amer, Maryam Dialameh, Hossein Rajabzadeh, et al.
+- **abs:** https://arxiv.org/abs/2601.01332 · **PDF:** https://arxiv.org/pdf/2601.01332v1
+
+## 24379. AppellateGen: A Benchmark for Appellate Legal Judgment Generation
+
+- **ID:** `2601.01331` · **primary:** `cs.CY` · **Submitted:** 2026-01-04
+- **Authors:** Hongkun Yang, Lionel Z. Wang, Wei Fan, et al.
+- **abs:** https://arxiv.org/abs/2601.01331 · **PDF:** https://arxiv.org/pdf/2601.01331v3
+
+## 24380. Beyond Gemini-3-Pro: Revisiting LLM Routing and Aggregation at Scale
+
+- **ID:** `2601.01330` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Shengji Tang, Weihao Lin, Peng Ye, et al.
+- **abs:** https://arxiv.org/abs/2601.01330 · **PDF:** https://arxiv.org/pdf/2601.01330v2
+
+## 24381. Automated Post-Incident Policy Gap Analysis via Threat-Informed Evidence Mapping using Large Language Models
+
+- **ID:** `2601.03287` · **primary:** `cs.CR` · **Submitted:** 2026-01-04
+- **Authors:** Huan Lin Oh, Jay Yong Jun Jie, Mandy Lee Ling Siu, et al.
+- **abs:** https://arxiv.org/abs/2601.03287 · **PDF:** https://arxiv.org/pdf/2601.03287v1
+
+## 24382. LinMU: Multimodal Understanding Made Linear
+
+- **ID:** `2601.01322` · **primary:** `cs.CV` · **Submitted:** 2026-01-04
+- **Authors:** Hongjie Wang, Niraj K. Jha
+- **abs:** https://arxiv.org/abs/2601.01322 · **PDF:** https://arxiv.org/pdf/2601.01322v1
+
+## 24383. Digital Twin AI: Opportunities and Challenges from Large Language Models to World Models
+
+- **ID:** `2601.01321` · **primary:** `cs.AI` · **Submitted:** 2026-01-04
+- **Authors:** Rong Zhou, Dongping Chen, Zihan Jia, et al.
+- **abs:** https://arxiv.org/abs/2601.01321 · **PDF:** https://arxiv.org/pdf/2601.01321v1
+
+## 24384. Adaptive Hierarchical Evaluation of LLMs and SAST tools for CWE Prediction in Python
+
+- **ID:** `2601.01320` · **primary:** `cs.SE` · **Submitted:** 2026-01-04
+- **Authors:** Muntasir Adnan, Carlos C. N. Kuhn
+- **abs:** https://arxiv.org/abs/2601.01320 · **PDF:** https://arxiv.org/pdf/2601.01320v1
+
+## 24385. Benchmarking Continuous Dynamic Multi-Objective Optimization: Survey and Generalized Test Suite
+
+- **ID:** `2601.01317` · **primary:** `cs.NE` · **Submitted:** 2026-01-04
+- **Authors:** Chang Shao, Qi Zhao, Nana Pu, et al.
+- **abs:** https://arxiv.org/abs/2601.01317 · **PDF:** https://arxiv.org/pdf/2601.01317v1
+
+## 24386. Quantifying Local Strain Field and Deformation in Active Contraction of Bladder Using a Pretrained Transformer Model: A Speckle-Free Approach
+
+- **ID:** `2601.01315` · **primary:** `q-bio.TO` · **Submitted:** 2026-01-04
+- **Authors:** Alireza Asadbeygi, Anne M. Robertson, Yasutaka Tobe, et al.
+- **abs:** https://arxiv.org/abs/2601.01315 · **PDF:** https://arxiv.org/pdf/2601.01315v1
+
+## 24387. Spectral-Window Hybrid (SWH)
+
+- **ID:** `2601.01313` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** Vladimer Khasia
+- **abs:** https://arxiv.org/abs/2601.01313 · **PDF:** https://arxiv.org/pdf/2601.01313v1
+
+## 24388. Concave Certificates: Geometric Framework for Distributionally Robust Risk and Complexity Analysis
+
+- **ID:** `2601.01311` · **primary:** `math.OC` · **Submitted:** 2026-01-04
+- **Authors:** Hong T. M. Chu
+- **abs:** https://arxiv.org/abs/2601.01311 · **PDF:** https://arxiv.org/pdf/2601.01311v1
+
+## 24389. Prompt Engineering for Responsible Generative AI Use in African Education: A Report from a Three-Day Training Series
+
+- **ID:** `2601.06121` · **primary:** `cs.CY` · **Submitted:** 2026-01-04
+- **Authors:** Benjamin Quarshie, Vanessa Willemse, Macharious Nabang, et al.
+- **abs:** https://arxiv.org/abs/2601.06121 · **PDF:** https://arxiv.org/pdf/2601.06121v1
+
+## 24390. Making MoE-based LLM Inference Resilient with Tarragon
+
+- **ID:** `2601.01310` · **primary:** `cs.DC` · **Submitted:** 2026-01-04
+- **Authors:** Songyu Zhang, Aaron Tam, Myungjin Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.01310 · **PDF:** https://arxiv.org/pdf/2601.01310v2
+
+## 24391. Towards a Principled Muon under $μ\mathsf{P}$: Ensuring Spectral Conditions throughout Training
+
+- **ID:** `2601.01306` · **primary:** `cs.LG` · **Submitted:** 2026-01-04
+- **Authors:** John Zhao
+- **abs:** https://arxiv.org/abs/2601.01306 · **PDF:** https://arxiv.org/pdf/2601.01306v2
+
+## 24392. Accelerating Monte-Carlo Tree Search with Optimized Posterior Policies
+
+- **ID:** `2601.01301` · **primary:** `cs.AI` · **Submitted:** 2026-01-03
+- **Authors:** Keith Frankston, Benjamin Howard
+- **abs:** https://arxiv.org/abs/2601.01301 · **PDF:** https://arxiv.org/pdf/2601.01301v2
+
+## 24393. T3C: Test-Time Tensor Compression with Consistency Guarantees
+
+- **ID:** `2601.01299` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Ismail Lamaakal, Chaymae Yahyati, Yassine Maleh, et al.
+- **abs:** https://arxiv.org/abs/2601.01299 · **PDF:** https://arxiv.org/pdf/2601.01299v1
+
+## 24394. Warp-Cortex: An Asynchronous, Memory-Efficient Architecture for Million-Agent Cognitive Scaling on Consumer Hardware
+
+- **ID:** `2601.01298` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Jorge L. Ruiz Williams
+- **abs:** https://arxiv.org/abs/2601.01298 · **PDF:** https://arxiv.org/pdf/2601.01298v1
+
+## 24395. ARGUS: Adaptive Rotation-Invariant Geometric Unsupervised System
+
+- **ID:** `2601.01297` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Anantha Sharma
+- **abs:** https://arxiv.org/abs/2601.01297 · **PDF:** https://arxiv.org/pdf/2601.01297v2
+
+## 24396. Aggressive Compression Enables LLM Weight Theft
+
+- **ID:** `2601.01296` · **primary:** `cs.CR` · **Submitted:** 2026-01-03
+- **Authors:** Davis Brown, Juan-Pablo Rivera, Dan Hendrycks, et al.
+- **abs:** https://arxiv.org/abs/2601.01296 · **PDF:** https://arxiv.org/pdf/2601.01296v1
+
+## 24397. Sobolev Approximation of Deep ReLU Networks in Log-Barron Space
+
+- **ID:** `2601.01295` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Changhoon Song, Seungchan Ko, Youngjoon Hong
+- **abs:** https://arxiv.org/abs/2601.01295 · **PDF:** https://arxiv.org/pdf/2601.01295v2
+
+## 24398. Diffusion Timbre Transfer Via Mutual Information Guided Inpainting
+
+- **ID:** `2601.01294` · **primary:** `cs.SD` · **Submitted:** 2026-01-03
+- **Authors:** Ching Ho Lee, Javier Nistal, Stefan Lattner, et al.
+- **abs:** https://arxiv.org/abs/2601.01294 · **PDF:** https://arxiv.org/pdf/2601.01294v2
+
+## 24399. The Alchemy of Thought: Understanding In-Context Learning Through Supervised Classification
+
+- **ID:** `2601.01290` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Harshita Narnoli, Mihai Surdeanu
+- **abs:** https://arxiv.org/abs/2601.01290 · **PDF:** https://arxiv.org/pdf/2601.01290v1
+
+## 24400. PyBatchRender: A Python Library for Batched 3D Rendering at Up to One Million FPS
+
+- **ID:** `2601.01288` · **primary:** `cs.GR` · **Submitted:** 2026-01-03
+- **Authors:** Evgenii Rudakov, Jonathan Shock, Benjamin Ultan Cowley
+- **abs:** https://arxiv.org/abs/2601.01288 · **PDF:** https://arxiv.org/pdf/2601.01288v1
+
+## 24401. AI-Powered Deepfake Detection Using CNN and Vision Transformer Architectures
+
+- **ID:** `2601.01281` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Sifatullah Sheikh Urmi, Kirtonia Nuzath Tabassum Arthi, Md Al-Imran
+- **abs:** https://arxiv.org/abs/2601.01281 · **PDF:** https://arxiv.org/pdf/2601.01281v1
+
+## 24402. Does Memory Need Graphs? A Unified Framework and Empirical Analysis for Long-Term Dialog Memory
+
+- **ID:** `2601.01280` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Sen Hu, Yuxiang Wei, Jiaxin Ran, et al.
+- **abs:** https://arxiv.org/abs/2601.01280 · **PDF:** https://arxiv.org/pdf/2601.01280v3
+
+## 24403. Collusive Pricing Under LLM
+
+- **ID:** `2601.01279` · **primary:** `econ.TH` · **Submitted:** 2026-01-03
+- **Authors:** Shengyu Cao, Ming Hu
+- **abs:** https://arxiv.org/abs/2601.01279 · **PDF:** https://arxiv.org/pdf/2601.01279v2
+
+## 24404. L2CU: Learning to Complement Unseen Users
+
+- **ID:** `2601.06119` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Dileepa Pitawela, Gustavo Carneiro, Hsiang-Ting Chen
+- **abs:** https://arxiv.org/abs/2601.06119 · **PDF:** https://arxiv.org/pdf/2601.06119v1
+
+## 24405. Accelerated Full Waveform Inversion by Deep Compressed Learning
+
+- **ID:** `2601.01268` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Maayan Gelboim, Amir Adler, Mauricio Araya-Polo
+- **abs:** https://arxiv.org/abs/2601.01268 · **PDF:** https://arxiv.org/pdf/2601.01268v1
+
+## 24406. What Artificial Intelligence can do for High-Performance Computing systems?
+
+- **ID:** `2602.00014` · **primary:** `cs.DC` · **Submitted:** 2026-01-03
+- **Authors:** Pierrick Pochelu, Hyacinthe Cartiaux, Julien Schleich
+- **abs:** https://arxiv.org/abs/2602.00014 · **PDF:** https://arxiv.org/pdf/2602.00014v1
+
+## 24407. From Policy to Logic for Efficient and Interpretable Coverage Assessment
+
+- **ID:** `2601.01266` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Rhitabrat Pokharel, Hamid Reza Hassanzadeh, Ameeta Agrawal
+- **abs:** https://arxiv.org/abs/2601.01266 · **PDF:** https://arxiv.org/pdf/2601.01266v2
+
+## 24408. MambaFormer: Token-Level Guided Routing Mixture-of-Experts for Accurate and Efficient Clinical Assistance
+
+- **ID:** `2601.01260` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Hamad Khan, Saddam Hussain Khan
+- **abs:** https://arxiv.org/abs/2601.01260 · **PDF:** https://arxiv.org/pdf/2601.01260v1
+
+## 24409. Beyond Reproducibility: Token Probabilities Expose Large Language Model Nondeterminism
+
+- **ID:** `2601.06118` · **primary:** `cs.AI` · **Submitted:** 2026-01-03
+- **Authors:** Tairan Fu, Gonzalo Martínez, Javier Conde, et al.
+- **abs:** https://arxiv.org/abs/2601.06118 · **PDF:** https://arxiv.org/pdf/2601.06118v1
+
+## 24410. Seamlessly Natural: Image Stitching with Natural Appearance Preservation
+
+- **ID:** `2601.01257` · **primary:** `eess.IV` · **Submitted:** 2026-01-03
+- **Authors:** Gaetane Lorna N. Tchana, Damaris Belle M. Fotso, Antonio Hendricks, et al.
+- **abs:** https://arxiv.org/abs/2601.01257 · **PDF:** https://arxiv.org/pdf/2601.01257v1
+
+## 24411. Entity-Aware and Secure Query Optimization in Database Using Named Entity Recognition
+
+- **ID:** `2601.01254` · **primary:** `cs.DB` · **Submitted:** 2026-01-03
+- **Authors:** Azrin Sultana, Hasibur Rashid Chayon
+- **abs:** https://arxiv.org/abs/2601.01254 · **PDF:** https://arxiv.org/pdf/2601.01254v1
+
+## 24412. Stochastic Control Methods for Optimization
+
+- **ID:** `2601.01248` · **primary:** `math.OC` · **Submitted:** 2026-01-03
+- **Authors:** Jinniao Qiu
+- **abs:** https://arxiv.org/abs/2601.01248 · **PDF:** https://arxiv.org/pdf/2601.01248v3
+
+## 24413. Racka: Efficient Hungarian LLM Adaptation on Academic Infrastructure
+
+- **ID:** `2601.01244` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Zsolt Csibi, Bence György Gortka, Natabara Gyöngyössy, et al.
+- **abs:** https://arxiv.org/abs/2601.01244 · **PDF:** https://arxiv.org/pdf/2601.01244v2
+
+## 24414. Evidence Slopes and Effective Dimension in Singular Linear Models
+
+- **ID:** `2601.01238` · **primary:** `stat.ML` · **Submitted:** 2026-01-03
+- **Authors:** Kalyaan Rao
+- **abs:** https://arxiv.org/abs/2601.01238 · **PDF:** https://arxiv.org/pdf/2601.01238v1
+
+## 24415. Benchmarking the Computational and Representational Efficiency of State Space Models against Transformers on Long-Context Dyadic Sessions
+
+- **ID:** `2601.01237` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Abidemi Koledoye, Chinemerem Unachukwu, Gold Nwobu, et al.
+- **abs:** https://arxiv.org/abs/2601.01237 · **PDF:** https://arxiv.org/pdf/2601.01237v1
+
+## 24416. The Active Discoverer Framework: Towards Autonomous Physics Reasoning through Neuro-Symbolic LaTeX Synthesis
+
+- **ID:** `2601.06117` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Hyunjun Jeon
+- **abs:** https://arxiv.org/abs/2601.06117 · **PDF:** https://arxiv.org/pdf/2601.06117v3
+
+## 24417. The Dependency Divide: An Interpretable Machine Learning Framework for Profiling Student Digital Satisfaction in the Bangladesh Context
+
+- **ID:** `2601.01231` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Md Muhtasim Munif Fahim, Humyra Ankona, Md Monimul Huq, et al.
+- **abs:** https://arxiv.org/abs/2601.01231 · **PDF:** https://arxiv.org/pdf/2601.01231v1
+
+## 24418. NeuroSSM: Multiscale Differential State-Space Modeling for Context-Aware fMRI Analysis
+
+- **ID:** `2601.01229` · **primary:** `eess.SP` · **Submitted:** 2026-01-03
+- **Authors:** Furkan Genç, Boran İsmet Macun, Sait Sarper Özaslan, et al.
+- **abs:** https://arxiv.org/abs/2601.01229 · **PDF:** https://arxiv.org/pdf/2601.01229v1
+
+## 24419. Is He Extroverted? Identifying Missing Relevant Personas for Faithful User Simulation
+
+- **ID:** `2602.15832` · **primary:** `cs.HC` · **Submitted:** 2026-01-03
+- **Authors:** Weiwen Su, Yuhan Zhou, Zihan Wang, et al.
+- **abs:** https://arxiv.org/abs/2602.15832 · **PDF:** https://arxiv.org/pdf/2602.15832v2
+
+## 24420. Stylometry Analysis of Human and Machine Text for Academic Integrity
+
+- **ID:** `2601.01225` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Hezam Albaqami, Muhammad Asif Ayub, Nasir Ahmad, et al.
+- **abs:** https://arxiv.org/abs/2601.01225 · **PDF:** https://arxiv.org/pdf/2601.01225v1
+
+## 24421. Improved Object-Centric Diffusion Learning with Registers and Contrastive Alignment
+
+- **ID:** `2601.01224` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Bac Nguyen, Yuhta Takida, Naoki Murata, et al.
+- **abs:** https://arxiv.org/abs/2601.01224 · **PDF:** https://arxiv.org/pdf/2601.01224v2
+
+## 24422. Adaptive Conformal Prediction via Bayesian Uncertainty Weighting for Hierarchical Healthcare Data
+
+- **ID:** `2601.01223` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Marzieh Amiri Shahbazi, Ali Baheri, Nasibeh Azadeh-Fard
+- **abs:** https://arxiv.org/abs/2601.01223 · **PDF:** https://arxiv.org/pdf/2601.01223v1
+
+## 24423. Correctness isnt Efficiency: Runtime Memory Divergence in LLM-Generated Code
+
+- **ID:** `2601.01215` · **primary:** `cs.SE` · **Submitted:** 2026-01-03
+- **Authors:** Prateek Rajput, Yewei Song, Abdoul Aziz Bonkoungou, et al.
+- **abs:** https://arxiv.org/abs/2601.01215 · **PDF:** https://arxiv.org/pdf/2601.01215v2
+
+## 24424. Promptable Foundation Models for SAR Remote Sensing: Adapting the Segment Anything Model for Snow Avalanche Segmentation
+
+- **ID:** `2601.01213` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Riccardo Gelato, Carlo Sgaravatti, Jakob Grahn, et al.
+- **abs:** https://arxiv.org/abs/2601.01213 · **PDF:** https://arxiv.org/pdf/2601.01213v1
+
+## 24425. Structure-Aware Diversity Pursuit as an AI Safety Strategy against Homogenization
+
+- **ID:** `2601.06116` · **primary:** `cs.AI` · **Submitted:** 2026-01-03
+- **Authors:** Ian Rios-Sialer
+- **abs:** https://arxiv.org/abs/2601.06116 · **PDF:** https://arxiv.org/pdf/2601.06116v1
+
+## 24426. Dreaming Is Not a Bug: A Jung-Inspired Dream Layer for Multi-Agent LLM Companions
+
+- **ID:** `2601.06115` · **primary:** `cs.AI` · **Submitted:** 2026-01-03
+- **Authors:** V. Cheung
+- **abs:** https://arxiv.org/abs/2601.06115 · **PDF:** https://arxiv.org/pdf/2601.06115v1
+
+## 24427. Sparse Bayesian Message Passing under Structural Uncertainty
+
+- **ID:** `2601.01207` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Yoonhyuk Choi, Jiho Choi, Chanran Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.01207 · **PDF:** https://arxiv.org/pdf/2601.01207v1
+
+## 24428. MentalGame: Predicting Personality-Job Fitness for Software Developers Using Multi-Genre Games and Machine Learning Approaches
+
+- **ID:** `2601.01206` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Soroush Elyasi, Arya VarastehNezhad, Fattaneh Taghiyareh
+- **abs:** https://arxiv.org/abs/2601.01206 · **PDF:** https://arxiv.org/pdf/2601.01206v1
+
+## 24429. GroupSegment-SHAP: Shapley Value Explanations with Group-Segment Players for Multivariate Time Series
+
+- **ID:** `2601.06114` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Jinwoong Kim, Sangjin Park
+- **abs:** https://arxiv.org/abs/2601.06114 · **PDF:** https://arxiv.org/pdf/2601.06114v1
+
+## 24430. RefSR-Adv: Adversarial Attack on Reference-based Image Super-Resolution Models
+
+- **ID:** `2601.01202` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Jiazhu Dai, Huihui Jiang
+- **abs:** https://arxiv.org/abs/2601.01202 · **PDF:** https://arxiv.org/pdf/2601.01202v1
+
+## 24431. Reinforcement Learning Enhanced Multi-hop Reasoning for Temporal Knowledge Question Answering
+
+- **ID:** `2601.01195` · **primary:** `cs.AI` · **Submitted:** 2026-01-03
+- **Authors:** Wuzhenghong Wen, Chao Xue, Su Pan, et al.
+- **abs:** https://arxiv.org/abs/2601.01195 · **PDF:** https://arxiv.org/pdf/2601.01195v1
+
+## 24432. Towards Infinite Length Extrapolation: A Unified Approach
+
+- **ID:** `2601.06113` · **primary:** `cs.AI` · **Submitted:** 2026-01-03
+- **Authors:** Nitin Vetcha
+- **abs:** https://arxiv.org/abs/2601.06113 · **PDF:** https://arxiv.org/pdf/2601.06113v1
+
+## 24433. ReliabilityBench: Evaluating LLM Agent Reliability Under Production-Like Stress Conditions
+
+- **ID:** `2601.06112` · **primary:** `cs.AI` · **Submitted:** 2026-01-03
+- **Authors:** Aayush Gupta
+- **abs:** https://arxiv.org/abs/2601.06112 · **PDF:** https://arxiv.org/pdf/2601.06112v1
+
+## 24434. LLM Powered Social Digital Twins: A Framework for Simulating Population Behavioral Response to Policy Interventions
+
+- **ID:** `2601.06111` · **primary:** `cs.AI` · **Submitted:** 2026-01-03
+- **Authors:** Fatima Koaik, Aayush Gupta, Farahan Raza Sheikh
+- **abs:** https://arxiv.org/abs/2601.06111 · **PDF:** https://arxiv.org/pdf/2601.06111v2
+
+## 24435. CBMAS: Cognitive Behavioral Modeling via Activation Steering
+
+- **ID:** `2601.06109` · **primary:** `cs.AI` · **Submitted:** 2026-01-03
+- **Authors:** Ahmed H. Ismail, Anthony Kuang, Ayo Akinkugbe, et al.
+- **abs:** https://arxiv.org/abs/2601.06109 · **PDF:** https://arxiv.org/pdf/2601.06109v1
+
+## 24436. Almost Clinical: Linguistic properties of synthetic electronic health records
+
+- **ID:** `2601.01171` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Serge Sharoff, John Baker, David Francis Hunt, et al.
+- **abs:** https://arxiv.org/abs/2601.01171 · **PDF:** https://arxiv.org/pdf/2601.01171v2
+
+## 24437. Bridging the Semantic Gap for Categorical Data Clustering via Large Language Models
+
+- **ID:** `2601.01162` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Zihua Yang, Xin Liao, Yiqun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.01162 · **PDF:** https://arxiv.org/pdf/2601.01162v1
+
+## 24438. Gradient-Free Approaches is a Key to an Efficient Interaction with Markovian Stochasticity
+
+- **ID:** `2601.01160` · **primary:** `math.OC` · **Submitted:** 2026-01-03
+- **Authors:** Boris Prokhorov, Semyon Chebykin, Alexander Gasnikov, et al.
+- **abs:** https://arxiv.org/abs/2601.01160 · **PDF:** https://arxiv.org/pdf/2601.01160v1
+
+## 24439. DHI: Leveraging Diverse Hallucination Induction for Enhanced Contrastive Factuality Control in Large Language Models
+
+- **ID:** `2601.01156` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Jiani Guo, Xiangke Zeng, Jie Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.01156 · **PDF:** https://arxiv.org/pdf/2601.01156v1
+
+## 24440. SongSage: A Large Musical Language Model with Lyric Generative Pre-training
+
+- **ID:** `2601.01153` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Jiani Guo, Jiajia Li, Jie Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.01153 · **PDF:** https://arxiv.org/pdf/2601.01153v1
+
+## 24441. Evo-TFS: Evolutionary Time-Frequency Domain-Based Synthetic Minority Oversampling Approach to Imbalanced Time Series Classification
+
+- **ID:** `2601.01150` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Wenbin Pei, Ruohao Dai, Bing Xue, et al.
+- **abs:** https://arxiv.org/abs/2601.01150 · **PDF:** https://arxiv.org/pdf/2601.01150v1
+
+## 24442. Conformal Blindness: A Note on $A$-Cryptic change-points
+
+- **ID:** `2601.01147` · **primary:** `stat.ML` · **Submitted:** 2026-01-03
+- **Authors:** Johan Hallberg Szabadváry
+- **abs:** https://arxiv.org/abs/2601.01147 · **PDF:** https://arxiv.org/pdf/2601.01147v2
+
+## 24443. Self-Training the Neurochaos Learning Algorithm
+
+- **ID:** `2601.01146` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Anusree M, Akhila Henry, Pramod P Nair
+- **abs:** https://arxiv.org/abs/2601.01146 · **PDF:** https://arxiv.org/pdf/2601.01146v1
+
+## 24444. KOS-TL (Knowledge Operation System Type Logic)
+
+- **ID:** `2601.01143` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Peng Chen
+- **abs:** https://arxiv.org/abs/2601.01143 · **PDF:** https://arxiv.org/pdf/2601.01143v1
+
+## 24445. AI-Powered Hybrid Intrusion Detection Framework for Cloud Security Using Novel Metaheuristic Optimization
+
+- **ID:** `2601.01134` · **primary:** `cs.CR` · **Submitted:** 2026-01-03
+- **Authors:** Maryam Mahdi Alhusseini, Alireza Rouhi, Mohammad-Reza Feizi-Derakhshi
+- **abs:** https://arxiv.org/abs/2601.01134 · **PDF:** https://arxiv.org/pdf/2601.01134v1
+
+## 24446. Generating Diverse TSP Tours via a Combination of Graph Pointer Network and Dispersion
+
+- **ID:** `2601.01132` · **primary:** `cs.CG` · **Submitted:** 2026-01-03
+- **Authors:** Hao-Tsung Yang, Ssu-Yuan Lo, Kuan-Lun Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.01132 · **PDF:** https://arxiv.org/pdf/2601.01132v2
+
+## 24447. RovoDev Code Reviewer: A Large-Scale Online Evaluation of LLM-based Code Review Automation at Atlassian
+
+- **ID:** `2601.01129` · **primary:** `cs.SE` · **Submitted:** 2026-01-03
+- **Authors:** Kla Tantithamthavorn, Yaotian Zou, Andy Wong, et al.
+- **abs:** https://arxiv.org/abs/2601.01129 · **PDF:** https://arxiv.org/pdf/2601.01129v2
+
+## 24448. Wittgenstein's Family Resemblance Clustering Algorithm
+
+- **ID:** `2601.01127` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Golbahar Amanpour, Benyamin Ghojogh
+- **abs:** https://arxiv.org/abs/2601.01127 · **PDF:** https://arxiv.org/pdf/2601.01127v2
+
+## 24449. RoboPhD: Self-Improving Text-to-SQL Through Autonomous Agent Evolution
+
+- **ID:** `2601.01126` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Andrew Borthwick, Stephen Ash
+- **abs:** https://arxiv.org/abs/2601.01126 · **PDF:** https://arxiv.org/pdf/2601.01126v2
+
+## 24450. Learning from Historical Activations in Graph Neural Networks
+
+- **ID:** `2601.01123` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Yaniv Galron, Hadar Sinai, Haggai Maron, et al.
+- **abs:** https://arxiv.org/abs/2601.01123 · **PDF:** https://arxiv.org/pdf/2601.01123v1
+
+## 24451. Listen, Attend, Understand: a Regularization Technique for Stable E2E Speech Translation Training on High Variance labels
+
+- **ID:** `2601.01121` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Yacouba Diarra, Michael Leventhal
+- **abs:** https://arxiv.org/abs/2601.01121 · **PDF:** https://arxiv.org/pdf/2601.01121v1
+
+## 24452. Community-Based Early-Stage Chronic Kidney Disease Screening using Explainable Machine Learning for Low-Resource Settings
+
+- **ID:** `2601.01119` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Muhammad Ashad Kabir, Sirajam Munira, Dewan Tasnia Azad, et al.
+- **abs:** https://arxiv.org/abs/2601.01119 · **PDF:** https://arxiv.org/pdf/2601.01119v1
+
+## 24453. ScienceDB AI: An LLM-Driven Agentic Recommender System for Large-Scale Scientific Data Sharing Services
+
+- **ID:** `2601.01118` · **primary:** `cs.IR` · **Submitted:** 2026-01-03
+- **Authors:** Qingqing Long, Haotian Chen, Chenyang Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.01118 · **PDF:** https://arxiv.org/pdf/2601.01118v1
+
+## 24454. From RLHF to Direct Alignment: A Theoretical Unification of Preference Learning for Large Language Models
+
+- **ID:** `2601.06108` · **primary:** `cs.AI` · **Submitted:** 2026-01-03
+- **Authors:** Tarun Raheja, Nilay Pochhi
+- **abs:** https://arxiv.org/abs/2601.06108 · **PDF:** https://arxiv.org/pdf/2601.06108v1
+
+## 24455. EmoLoom-2B: Fast Base-Model Screening for Emotion Classification and VAD with Lexicon-Weak Supervision and KV-Off Evaluation
+
+- **ID:** `2601.01112` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Zilin Li, Weiwei Xu, Xuanbo Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.01112 · **PDF:** https://arxiv.org/pdf/2601.01112v2
+
+## 24456. Evolving CNN Architectures: From Custom Designs to Deep Residual Models for Diverse Image Classification and Detection Tasks
+
+- **ID:** `2601.01099` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Mahmudul Hasan, Mabsur Fatin Bin Hossain
+- **abs:** https://arxiv.org/abs/2601.01099 · **PDF:** https://arxiv.org/pdf/2601.01099v1
+
+## 24457. Neural Networks on Symmetric Spaces of Noncompact Type
+
+- **ID:** `2601.01097` · **primary:** `stat.ML` · **Submitted:** 2026-01-03
+- **Authors:** Xuan Son Nguyen, Shuo Yang, Aymeric Histace
+- **abs:** https://arxiv.org/abs/2601.01097 · **PDF:** https://arxiv.org/pdf/2601.01097v1
+
+## 24458. Judge Model for Large-scale Multimodality Benchmarks
+
+- **ID:** `2601.06106` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Min-Han Shih, Yu-Hsin Wu, Yu-Wei Chen
+- **abs:** https://arxiv.org/abs/2601.06106 · **PDF:** https://arxiv.org/pdf/2601.06106v1
+
+## 24459. NarrativeTrack: Evaluating Entity-Centric Reasoning for Narrative Understanding
+
+- **ID:** `2601.01095` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Hyeonjeong Ha, Jinjin Ge, Bo Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.01095 · **PDF:** https://arxiv.org/pdf/2601.01095v2
+
+## 24460. SoulSeek: Exploring the Use of Social Cues in LLM-based Information Seeking
+
+- **ID:** `2601.01094` · **primary:** `cs.HC` · **Submitted:** 2026-01-03
+- **Authors:** Yubo Shu, Peng Zhang, Meng Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.01094 · **PDF:** https://arxiv.org/pdf/2601.01094v1
+
+## 24461. ks-lit-3m: A 3.1 million word kashmiri text dataset for large language model pretraining
+
+- **ID:** `2601.01091` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Haq Nawaz Malik
+- **abs:** https://arxiv.org/abs/2601.01091 · **PDF:** https://arxiv.org/pdf/2601.01091v1
+
+## 24462. HyperCLOVA X 32B Think
+
+- **ID:** `2601.03286` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** NAVER Cloud HyperCLOVA X Team
+- **abs:** https://arxiv.org/abs/2601.03286 · **PDF:** https://arxiv.org/pdf/2601.03286v1
+
+## 24463. Multimodal Sentiment Analysis based on Multi-channel and Symmetric Mutual Promotion Feature Fusion
+
+- **ID:** `2601.02415` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Wangyuan Zhu, Jun Yu
+- **abs:** https://arxiv.org/abs/2601.02415 · **PDF:** https://arxiv.org/pdf/2601.02415v1
+
+## 24464. Harm in AI-Driven Societies: An Audit of Toxicity Adoption on Chirper.ai
+
+- **ID:** `2601.01090` · **primary:** `cs.MA` · **Submitted:** 2026-01-03
+- **Authors:** Erica Coppolillo, Luca Luceri, Emilio Ferrara
+- **abs:** https://arxiv.org/abs/2601.01090 · **PDF:** https://arxiv.org/pdf/2601.01090v2
+
+## 24465. Central Dogma Transformer: Towards Mechanism-Oriented AI for Cellular Understanding
+
+- **ID:** `2601.01089` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Nobuyuki Ota
+- **abs:** https://arxiv.org/abs/2601.01089 · **PDF:** https://arxiv.org/pdf/2601.01089v2
+
+## 24466. 600k-ks-ocr: a large-scale synthetic dataset for optical character recognition in kashmiri script
+
+- **ID:** `2601.01088` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Haq Nawaz Malik
+- **abs:** https://arxiv.org/abs/2601.01088 · **PDF:** https://arxiv.org/pdf/2601.01088v1
+
+## 24467. MIAR: Modality Interaction and Alignment Representation Fuison for Multimodal Emotion
+
+- **ID:** `2601.02414` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Jichao Zhu, Jun Yu
+- **abs:** https://arxiv.org/abs/2601.02414 · **PDF:** https://arxiv.org/pdf/2601.02414v1
+
+## 24468. Luminark: Training-free, Probabilistically-Certified Watermarking for General Vision Generative Models
+
+- **ID:** `2601.01085` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Jiayi Xu, Zhang Zhang, Yuanrui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.01085 · **PDF:** https://arxiv.org/pdf/2601.01085v1
+
+## 24469. Discount Model Search for Quality Diversity Optimization in High-Dimensional Measure Spaces
+
+- **ID:** `2601.01082` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Bryon Tjanaka, Henry Chen, Matthew C. Fontaine, et al.
+- **abs:** https://arxiv.org/abs/2601.01082 · **PDF:** https://arxiv.org/pdf/2601.01082v4
+
+## 24470. Australian Bushfire Intelligence with AI-Driven Environmental Analytics
+
+- **ID:** `2601.06105` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Tanvi Jois, Hussain Ahmad, Fatima Noor, et al.
+- **abs:** https://arxiv.org/abs/2601.06105 · **PDF:** https://arxiv.org/pdf/2601.06105v1
+
+## 24471. Comment on arXiv:2511.21731v1: Identifying Quantum Structure in AI Language: Evidence for Evolutionary Convergence of Human and Artificial Cognition
+
+- **ID:** `2601.06104` · **primary:** `cs.AI` · **Submitted:** 2026-01-03
+- **Authors:** Krzysztof Sienicki
+- **abs:** https://arxiv.org/abs/2601.06104 · **PDF:** https://arxiv.org/pdf/2601.06104v1
+
+## 24472. Scalable Data-Driven Reachability Analysis and Control via Koopman Operators with Conformal Coverage Guarantees
+
+- **ID:** `2601.01076` · **primary:** `eess.SY` · **Submitted:** 2026-01-03
+- **Authors:** Devesh Nath, Haoran Yin, Glen Chou
+- **abs:** https://arxiv.org/abs/2601.01076 · **PDF:** https://arxiv.org/pdf/2601.01076v1
+
+## 24473. Flow Equivariant World Models: Memory for Partially Observed Dynamic Environments
+
+- **ID:** `2601.01075` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Hansen Jin Lillemark, Benhao Huang, Fangneng Zhan, et al.
+- **abs:** https://arxiv.org/abs/2601.01075 · **PDF:** https://arxiv.org/pdf/2601.01075v1
+
+## 24474. Gendered Pathways in AI Companionship: Cross-Community Behavior and Toxicity Patterns on Reddit
+
+- **ID:** `2601.01073` · **primary:** `cs.SI` · **Submitted:** 2026-01-03
+- **Authors:** Erica Coppolillo, Emilio Ferrara
+- **abs:** https://arxiv.org/abs/2601.01073 · **PDF:** https://arxiv.org/pdf/2601.01073v1
+
+## 24475. Revisiting Weighted Strategy for Non-stationary Parametric Bandits and MDPs
+
+- **ID:** `2601.01069` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Jing Wang, Peng Zhao, Zhi-Hua Zhou
+- **abs:** https://arxiv.org/abs/2601.01069 · **PDF:** https://arxiv.org/pdf/2601.01069v1
+
+## 24476. Tiny Machine Learning for Real-Time Aquaculture Monitoring: A Case Study in Morocco
+
+- **ID:** `2601.01065` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Achraf Hsain, Yahya Zaki, Othman Abaakil, et al.
+- **abs:** https://arxiv.org/abs/2601.01065 · **PDF:** https://arxiv.org/pdf/2601.01065v1
+
+## 24477. SPoRC-VIST: A Benchmark for Evaluating Generative Natural Narrative in Vision-Language Models
+
+- **ID:** `2601.01062` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Yunlin Zeng
+- **abs:** https://arxiv.org/abs/2601.01062 · **PDF:** https://arxiv.org/pdf/2601.01062v1
+
+## 24478. A UCB Bandit Algorithm for General ML-Based Estimators
+
+- **ID:** `2601.01061` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Yajing Liu, Erkao Bao, Linqi Song
+- **abs:** https://arxiv.org/abs/2601.01061 · **PDF:** https://arxiv.org/pdf/2601.01061v2
+
+## 24479. Unsupervised Text Style Transfer for Controllable Intensity
+
+- **ID:** `2601.01060` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Shuhuan Gu, Wenbiao Tao, Xinchen Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.01060 · **PDF:** https://arxiv.org/pdf/2601.01060v1
+
+## 24480. Enhancing Histopathological Image Classification via Integrated HOG and Deep Features with Robust Noise Performance
+
+- **ID:** `2601.01056` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Ifeanyi Ezuma, Ugochukwu Ugwu
+- **abs:** https://arxiv.org/abs/2601.01056 · **PDF:** https://arxiv.org/pdf/2601.01056v1
+
+## 24481. Fibonacci-Driven Recursive Ensembles: Algorithms, Convergence, and Learning Dynamics
+
+- **ID:** `2601.01055` · **primary:** `stat.ML` · **Submitted:** 2026-01-03
+- **Authors:** Ernest Fokoué
+- **abs:** https://arxiv.org/abs/2601.01055 · **PDF:** https://arxiv.org/pdf/2601.01055v1
+
+## 24482. Byzantine-Robust Federated Learning Framework with Post-Quantum Secure Aggregation for Real-Time Threat Intelligence Sharing in Critical IoT Infrastructure
+
+- **ID:** `2601.01053` · **primary:** `cs.CR` · **Submitted:** 2026-01-03
+- **Authors:** Milad Rahmati, Nima Rahmati
+- **abs:** https://arxiv.org/abs/2601.01053 · **PDF:** https://arxiv.org/pdf/2601.01053v1
+
+## 24483. EgoGrasp: World-Space Hand-Object Interaction Estimation from Egocentric Videos
+
+- **ID:** `2601.01050` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Hongming Fu, Wenjia Wang, Xiaozhen Qiao, et al.
+- **abs:** https://arxiv.org/abs/2601.01050 · **PDF:** https://arxiv.org/pdf/2601.01050v2
+
+## 24484. KV-Embedding: Training-free Text Embedding via Internal KV Re-routing in Decoder-only LLMs
+
+- **ID:** `2601.01046` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Yixuan Tang, Yi Yang
+- **abs:** https://arxiv.org/abs/2601.01046 · **PDF:** https://arxiv.org/pdf/2601.01046v1
+
+## 24485. Coarse-Grained Kullback--Leibler Control of Diffusion-Based Generative AI
+
+- **ID:** `2601.01045` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Tatsuaki Tsuruyama
+- **abs:** https://arxiv.org/abs/2601.01045 · **PDF:** https://arxiv.org/pdf/2601.01045v2
+
+## 24486. Evaluating transfer learning strategies for improving dairy cattle body weight prediction in small farms using depth-image and point-cloud data
+
+- **ID:** `2601.01044` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Jin Wang, Angelo De Castro, Yuxi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.01044 · **PDF:** https://arxiv.org/pdf/2601.01044v1
+
+## 24487. Multi-Dimensional Prompt Chaining to Improve Open-Domain Dialogue Generation
+
+- **ID:** `2601.01037` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Livia Leong Hui Teng
+- **abs:** https://arxiv.org/abs/2601.01037 · **PDF:** https://arxiv.org/pdf/2601.01037v1
+
+## 24488. Beyond Demand Estimation: Consumer Surplus Evaluation via Cumulative Propensity Weights
+
+- **ID:** `2601.01029` · **primary:** `stat.ML` · **Submitted:** 2026-01-03
+- **Authors:** Zeyu Bian, Max Biggs, Ruijiang Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.01029 · **PDF:** https://arxiv.org/pdf/2601.01029v1
+
+## 24489. A Platform for Interactive AI Character Experiences
+
+- **ID:** `2601.01027` · **primary:** `cs.HC` · **Submitted:** 2026-01-03
+- **Authors:** Rafael Wampfler, Chen Yang, Dillon Elste, et al.
+- **abs:** https://arxiv.org/abs/2601.01027 · **PDF:** https://arxiv.org/pdf/2601.01027v1
+
+## 24490. Enhanced Leukemic Cell Classification Using Attention-Based CNN and Data Augmentation
+
+- **ID:** `2601.01026` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Douglas Costa Braga, Daniel Oliveira Dantas
+- **abs:** https://arxiv.org/abs/2601.01026 · **PDF:** https://arxiv.org/pdf/2601.01026v1
+
+## 24491. ITSELF: Attention Guided Fine-Grained Alignment for Vision-Language Retrieval
+
+- **ID:** `2601.01024` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Tien-Huy Nguyen, Huu-Loc Tran, Thanh Duc Ngo
+- **abs:** https://arxiv.org/abs/2601.01024 · **PDF:** https://arxiv.org/pdf/2601.01024v1
+
+## 24492. Wireless Dataset Similarity: Measuring Distances in Supervised and Unsupervised Machine Learning
+
+- **ID:** `2601.01023` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** João Morais, Sadjad Alikhani, Akshay Malhotra, et al.
+- **abs:** https://arxiv.org/abs/2601.01023 · **PDF:** https://arxiv.org/pdf/2601.01023v1
+
+## 24493. Decoupling Amplitude and Phase Attention in Frequency Domain for RGB-Event based Visual Object Tracking
+
+- **ID:** `2601.01022` · **primary:** `cs.CV` · **Submitted:** 2026-01-03
+- **Authors:** Shiao Wang, Xiao Wang, Haonan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.01022 · **PDF:** https://arxiv.org/pdf/2601.01022v1
+
+## 24494. Expanding the Chaos: Neural Operator for Stochastic (Partial) Differential Equations
+
+- **ID:** `2601.01021` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Dai Shi, Lequan Lin, Andi Han, et al.
+- **abs:** https://arxiv.org/abs/2601.01021 · **PDF:** https://arxiv.org/pdf/2601.01021v2
+
+## 24495. Improving Variational Autoencoder using Random Fourier Transformation: An Aviation Safety Anomaly Detection Case-Study
+
+- **ID:** `2601.01016` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Ata Akbari Asanjan, Milad Memarzadeh, Bryan Matthews, et al.
+- **abs:** https://arxiv.org/abs/2601.01016 · **PDF:** https://arxiv.org/pdf/2601.01016v3
+
+## 24496. HyperJoin: LLM-augmented Hypergraph Link Prediction for Joinable Table Discovery
+
+- **ID:** `2601.01015` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Shiyuan Liu, Jianwei Wang, Xuemin Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.01015 · **PDF:** https://arxiv.org/pdf/2601.01015v1
+
+## 24497. Geometric and Dynamic Scaling in Deep Transformers
+
+- **ID:** `2601.01014` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Haoran Su, Chenyu You
+- **abs:** https://arxiv.org/abs/2601.01014 · **PDF:** https://arxiv.org/pdf/2601.01014v3
+
+## 24498. The Impact of Post-training on Data Contamination
+
+- **ID:** `2601.06103` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Muhammed Yusuf Kocyigit, Caglar Yildirim
+- **abs:** https://arxiv.org/abs/2601.06103 · **PDF:** https://arxiv.org/pdf/2601.06103v1
+
+## 24499. Intention Collapse: Intention-Level Metrics for Reasoning in Language Models
+
+- **ID:** `2601.01011` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Patricio Vera
+- **abs:** https://arxiv.org/abs/2601.01011 · **PDF:** https://arxiv.org/pdf/2601.01011v2
+
+## 24500. Let Me Try Again: Examining Replay Behavior by Tracing Students' Latent Problem-Solving Pathways
+
+- **ID:** `2601.11586` · **primary:** `cs.CY` · **Submitted:** 2026-01-03
+- **Authors:** Shan Zhang, Siddhartha Pradhan, Ji-Eun Lee, et al.
+- **abs:** https://arxiv.org/abs/2601.11586 · **PDF:** https://arxiv.org/pdf/2601.11586v1
+
+## 24501. Dynamic Intelligence Ceilings: Measuring Long-Horizon Limits of Planning and Creativity in Artificial Systems
+
+- **ID:** `2601.06102` · **primary:** `cs.AI` · **Submitted:** 2026-01-03
+- **Authors:** Truong Xuan Khanh, Truong Quynh Hoa
+- **abs:** https://arxiv.org/abs/2601.06102 · **PDF:** https://arxiv.org/pdf/2601.06102v1
+
+## 24502. Disordered Dynamics in High Dimensions: Connections to Random Matrices and Machine Learning
+
+- **ID:** `2601.01010` · **primary:** `cond-mat.dis-nn` · **Submitted:** 2026-01-03
+- **Authors:** Blake Bordelon, Cengiz Pehlevan
+- **abs:** https://arxiv.org/abs/2601.01010 · **PDF:** https://arxiv.org/pdf/2601.01010v2
+
+## 24503. Data-Driven Assessment of Concrete Mixture Compositions on Chloride Transport via Standalone Machine Learning Algorithms
+
+- **ID:** `2601.01009` · **primary:** `cs.LG` · **Submitted:** 2026-01-03
+- **Authors:** Mojtaba Aliasghar-Mamaghani, Mohammadreza Khalafi
+- **abs:** https://arxiv.org/abs/2601.01009 · **PDF:** https://arxiv.org/pdf/2601.01009v1
+
+## 24504. How to Assess AI Literacy: Misalignment Between Self-Reported and Objective-Based Measures
+
+- **ID:** `2601.06101` · **primary:** `cs.CY` · **Submitted:** 2026-01-03
+- **Authors:** Shan Zhang, Ruiwei Xiao, Anthony F. Botelho, et al.
+- **abs:** https://arxiv.org/abs/2601.06101 · **PDF:** https://arxiv.org/pdf/2601.06101v1
+
+## 24505. An Explainable Agentic AI Framework for Uncertainty-Aware and Abstention-Enabled Acute Ischemic Stroke Imaging Decisions
+
+- **ID:** `2601.01008` · **primary:** `eess.IV` · **Submitted:** 2026-01-03
+- **Authors:** Md Rashadul Islam
+- **abs:** https://arxiv.org/abs/2601.01008 · **PDF:** https://arxiv.org/pdf/2601.01008v1
+
+## 24506. M3Kang: Evaluating Multilingual Multimodal Mathematical Reasoning in Vision-Language Models
+
+- **ID:** `2601.16218` · **primary:** `cs.CL` · **Submitted:** 2026-01-03
+- **Authors:** Aleix Torres-Camps, Nathaniel Mitrani Hadida, Víctor Conchello Vendrell, et al.
+- **abs:** https://arxiv.org/abs/2601.16218 · **PDF:** https://arxiv.org/pdf/2601.16218v1
+
+## 24507. Scale-aware Adaptive Supervised Network with Limited Medical Annotations
+
+- **ID:** `2601.01005` · **primary:** `eess.IV` · **Submitted:** 2026-01-02
+- **Authors:** Zihan Li, Dandan Shan, Yunxiang Li, et al.
+- **abs:** https://arxiv.org/abs/2601.01005 · **PDF:** https://arxiv.org/pdf/2601.01005v1
+
+## 24508. Contractive Diffusion Policies: Robust Action Diffusion via Contractive Score-Based Sampling with Differential Equations
+
+- **ID:** `2601.01003` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Amin Abyaneh, Charlotte Morissette, Mohamad H. Danesh, et al.
+- **abs:** https://arxiv.org/abs/2601.01003 · **PDF:** https://arxiv.org/pdf/2601.01003v2
+
+## 24509. Dynamic Accuracy Estimation in a Wi-Fi-based Positioning System
+
+- **ID:** `2601.00999` · **primary:** `eess.SP` · **Submitted:** 2026-01-02
+- **Authors:** Marcin Kolakowski, Vitomir Djaja-Josko
+- **abs:** https://arxiv.org/abs/2601.00999 · **PDF:** https://arxiv.org/pdf/2601.00999v1
+
+## 24510. VEAT Quantifies Implicit Associations in Text-to-Video Generator Sora and Reveals Challenges in Bias Mitigation
+
+- **ID:** `2601.00996` · **primary:** `cs.CY` · **Submitted:** 2026-01-02
+- **Authors:** Yongxu Sun, Michael Saxon, Ian Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.00996 · **PDF:** https://arxiv.org/pdf/2601.00996v1
+
+## 24511. ElecTwit: A Framework for Studying Persuasion in Multi-Agent Social Systems
+
+- **ID:** `2601.00994` · **primary:** `cs.AI` · **Submitted:** 2026-01-02
+- **Authors:** Michael Bao
+- **abs:** https://arxiv.org/abs/2601.00994 · **PDF:** https://arxiv.org/pdf/2601.00994v1
+
+## 24512. WildIng: A Wildlife Image Invariant Representation Model for Geographical Domain Shift
+
+- **ID:** `2601.00993` · **primary:** `cs.CV` · **Submitted:** 2026-01-02
+- **Authors:** Julian D. Santamaria, Claudia Isaza, Jhony H. Giraldo
+- **abs:** https://arxiv.org/abs/2601.00993 · **PDF:** https://arxiv.org/pdf/2601.00993v1
+
+## 24513. Filtering Beats Fine Tuning: A Bayesian Kalman View of In Context Learning in LLMs
+
+- **ID:** `2601.06100` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Andrew Kiruluta
+- **abs:** https://arxiv.org/abs/2601.06100 · **PDF:** https://arxiv.org/pdf/2601.06100v1
+
+## 24514. Feedback Indices to Evaluate LLM Responses to Rebuttals for Multiple Choice Type Questions
+
+- **ID:** `2601.03285` · **primary:** `physics.ed-ph` · **Submitted:** 2026-01-02
+- **Authors:** Justin C. Dunlap, Anne-Simone Parent, Ralf Widenhorn
+- **abs:** https://arxiv.org/abs/2601.03285 · **PDF:** https://arxiv.org/pdf/2601.03285v1
+
+## 24515. Tessellation Localized Transfer learning for nonparametric regression
+
+- **ID:** `2601.00987` · **primary:** `math.ST` · **Submitted:** 2026-01-02
+- **Authors:** Hélène Halconruy, Benjamin Bobbia, Paul Lejamtel
+- **abs:** https://arxiv.org/abs/2601.00987 · **PDF:** https://arxiv.org/pdf/2601.00987v2
+
+## 24516. Zero-shot Forecasting by Simulation Alone
+
+- **ID:** `2601.00970` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Boris N. Oreshkin, Mayank Jauhari, Ravi Kiran Selvam, et al.
+- **abs:** https://arxiv.org/abs/2601.00970 · **PDF:** https://arxiv.org/pdf/2601.00970v1
+
+## 24517. Value Vision-Language-Action Planning & Search
+
+- **ID:** `2601.00969` · **primary:** `cs.RO` · **Submitted:** 2026-01-02
+- **Authors:** Ali Salamatian, Ke, Ren, et al.
+- **abs:** https://arxiv.org/abs/2601.00969 · **PDF:** https://arxiv.org/pdf/2601.00969v1
+
+## 24518. Explainability-Guided Defense: Attribution-Aware Model Refinement Against Adversarial Data Attacks
+
+- **ID:** `2601.00968` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Longwei Wang, Mohammad Navid Nayyem, Abdullah Al Rakin, et al.
+- **abs:** https://arxiv.org/abs/2601.00968 · **PDF:** https://arxiv.org/pdf/2601.00968v1
+
+## 24519. Adapting Feature Attenuation to NLP
+
+- **ID:** `2601.00965` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Tianshuo Yang, Ryan Rabinowitz, Terrance E. Boult, et al.
+- **abs:** https://arxiv.org/abs/2601.00965 · **PDF:** https://arxiv.org/pdf/2601.00965v1
+
+## 24520. Deep Clustering with Associative Memories
+
+- **ID:** `2601.00963` · **primary:** `cs.CV` · **Submitted:** 2026-01-02
+- **Authors:** Bishwajit Saha, Dmitry Krotov, Mohammed J. Zaki, et al.
+- **abs:** https://arxiv.org/abs/2601.00963 · **PDF:** https://arxiv.org/pdf/2601.00963v1
+
+## 24521. Two Deep Learning Approaches for Automated Segmentation of Left Ventricle in Cine Cardiac MRI
+
+- **ID:** `2601.00794` · **primary:** `cs.CV` · **Submitted:** 2026-01-02
+- **Authors:** Wenhui Chu, Nikolaos V. Tsekos
+- **abs:** https://arxiv.org/abs/2601.00794 · **PDF:** https://arxiv.org/pdf/2601.00794v1
+
+## 24522. Geometry of Reason: Spectral Signatures of Valid Mathematical Reasoning
+
+- **ID:** `2601.00791` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Valentin Noël
+- **abs:** https://arxiv.org/abs/2601.00791 · **PDF:** https://arxiv.org/pdf/2601.00791v1
+
+## 24523. Adapting Natural Language Processing Models Across Jurisdictions: A pilot Study in Canadian Cancer Registries
+
+- **ID:** `2601.00787` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Jonathan Simkin, Lovedeep Gondara, Zeeshan Rizvi, et al.
+- **abs:** https://arxiv.org/abs/2601.00787 · **PDF:** https://arxiv.org/pdf/2601.00787v1
+
+## 24524. FedHypeVAE: Federated Learning with Hypernetwork Generated Conditional VAEs for Differentially Private Embedding Sharing
+
+- **ID:** `2601.00785` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Sunny Gupta, Amit Sethi
+- **abs:** https://arxiv.org/abs/2601.00785 · **PDF:** https://arxiv.org/pdf/2601.00785v1
+
+## 24525. Categorical Reparameterization with Denoising Diffusion models
+
+- **ID:** `2601.00781` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Samson Gourevitch, Alain Durmus, Eric Moulines, et al.
+- **abs:** https://arxiv.org/abs/2601.00781 · **PDF:** https://arxiv.org/pdf/2601.00781v2
+
+## 24526. Reliability Under Randomness: An Empirical Analysis of Sparse and Dense Language Models Across Decoding Temperatures
+
+- **ID:** `2601.00942` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Kabir Grover
+- **abs:** https://arxiv.org/abs/2601.00942 · **PDF:** https://arxiv.org/pdf/2601.00942v1
+
+## 24527. LLM Agents for Combinatorial Efficient Frontiers: Investment Portfolio Optimization
+
+- **ID:** `2601.00770` · **primary:** `cs.CE` · **Submitted:** 2026-01-02
+- **Authors:** Simon Paquette-Greenbaum, Jiangbo Yu
+- **abs:** https://arxiv.org/abs/2601.00770 · **PDF:** https://arxiv.org/pdf/2601.00770v1
+
+## 24528. FinRetrieval: A Benchmark for Financial Data Retrieval by AI Agents
+
+- **ID:** `2603.04403` · **primary:** `cs.IR` · **Submitted:** 2026-01-02
+- **Authors:** Eric Y. Kim, Jie Huang
+- **abs:** https://arxiv.org/abs/2603.04403 · **PDF:** https://arxiv.org/pdf/2603.04403v1
+
+## 24529. Memory Bank Compression for Continual Adaptation of Large Language Models
+
+- **ID:** `2601.00756` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Thomas Katraouras, Dimitrios Rafailidis
+- **abs:** https://arxiv.org/abs/2601.00756 · **PDF:** https://arxiv.org/pdf/2601.00756v1
+
+## 24530. A Machine Learning Framework for Off Ball Defensive Role and Performance Evaluation in Football
+
+- **ID:** `2601.00748` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Sean Groom, Shuo Wang, Francisco Belo, et al.
+- **abs:** https://arxiv.org/abs/2601.00748 · **PDF:** https://arxiv.org/pdf/2601.00748v1
+
+## 24531. The Reasoning-Creativity Trade-off: Toward Creativity-Driven Problem Solving
+
+- **ID:** `2601.00747` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Max Ruiz Luyten, Mihaela van der Schaar
+- **abs:** https://arxiv.org/abs/2601.00747 · **PDF:** https://arxiv.org/pdf/2601.00747v1
+
+## 24532. An Agentic Framework for Neuro-Symbolic Programming
+
+- **ID:** `2601.00743` · **primary:** `cs.AI` · **Submitted:** 2026-01-02
+- **Authors:** Aliakbar Nafar, Chetan Chigurupati, Danial Kamali, et al.
+- **abs:** https://arxiv.org/abs/2601.00743 · **PDF:** https://arxiv.org/pdf/2601.00743v1
+
+## 24533. Socially-Aware Recommender Systems Mitigate Opinion Clusterization
+
+- **ID:** `2601.02412` · **primary:** `cs.IR` · **Submitted:** 2026-01-02
+- **Authors:** Lukas Schüepp, Carmen Amo Alonso, Florian Dörfler, et al.
+- **abs:** https://arxiv.org/abs/2601.02412 · **PDF:** https://arxiv.org/pdf/2601.02412v1
+
+## 24534. Stochastic Actor-Critic: Mitigating Overestimation via Temporal Aleatoric Uncertainty
+
+- **ID:** `2601.00737` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Uğurcan Özalp
+- **abs:** https://arxiv.org/abs/2601.00737 · **PDF:** https://arxiv.org/pdf/2601.00737v1
+
+## 24535. Exploring the Performance of Large Language Models on Subjective Span Identification Tasks
+
+- **ID:** `2601.00736` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Alphaeus Dmonte, Roland Oruche, Tharindu Ranasinghe, et al.
+- **abs:** https://arxiv.org/abs/2601.00736 · **PDF:** https://arxiv.org/pdf/2601.00736v1
+
+## 24536. Comparative Analysis of Formula and Structure Prediction from Tandem Mass Spectra
+
+- **ID:** `2601.00941` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-02
+- **Authors:** Xujun Che, Xiuxia Du, Depeng Xu
+- **abs:** https://arxiv.org/abs/2601.00941 · **PDF:** https://arxiv.org/pdf/2601.00941v1
+
+## 24537. Precision autotuning for linear solvers via contextual bandit-based RL
+
+- **ID:** `2601.00728` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Erin Carson, Xinye Chen
+- **abs:** https://arxiv.org/abs/2601.00728 · **PDF:** https://arxiv.org/pdf/2601.00728v4
+
+## 24538. AI-Guided Discovery of Novel Ionic Liquid Solvents for Industrial CO2 Capture
+
+- **ID:** `2601.03284` · **primary:** `physics.chem-ph` · **Submitted:** 2026-01-02
+- **Authors:** Davide Garbelotto, Alexander Lobo, Urvi Awasthi, et al.
+- **abs:** https://arxiv.org/abs/2601.03284 · **PDF:** https://arxiv.org/pdf/2601.03284v1
+
+## 24539. Detecting Performance Degradation under Data Shift in Pathology Vision-Language Model
+
+- **ID:** `2601.00716` · **primary:** `cs.CV` · **Submitted:** 2026-01-02
+- **Authors:** Hao Guan, Li Zhou
+- **abs:** https://arxiv.org/abs/2601.00716 · **PDF:** https://arxiv.org/pdf/2601.00716v1
+
+## 24540. BSAT: B-Spline Adaptive Tokenizer for Long-Term Time Series Forecasting
+
+- **ID:** `2601.00698` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Maximilian Reinwardt, Michael Eichelbeck, Matthias Althoff
+- **abs:** https://arxiv.org/abs/2601.00698 · **PDF:** https://arxiv.org/pdf/2601.00698v1
+
+## 24541. Bayesian Inverse Games with High-Dimensional Multi-Modal Observations
+
+- **ID:** `2601.00696` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Yash Jain, Xinjie Liu, Lasse Peters, et al.
+- **abs:** https://arxiv.org/abs/2601.00696 · **PDF:** https://arxiv.org/pdf/2601.00696v1
+
+## 24542. A Vision-and-Knowledge Enhanced Large Language Model for Generalizable Pedestrian Crossing Behavior Inference
+
+- **ID:** `2601.00694` · **primary:** `cs.AI` · **Submitted:** 2026-01-02
+- **Authors:** Qingwen Pu, Kun Xie, Hong Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.00694 · **PDF:** https://arxiv.org/pdf/2601.00694v1
+
+## 24543. ARISE: Adaptive Reinforcement Integrated with Swarm Exploration
+
+- **ID:** `2601.00693` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Rajiv Chaitanya M, D R Ramesh Babu
+- **abs:** https://arxiv.org/abs/2601.00693 · **PDF:** https://arxiv.org/pdf/2601.00693v1
+
+## 24544. Rate-Distortion Analysis of Compressed Query Delegation with Low-Rank Riemannian Updates
+
+- **ID:** `2601.00938` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Faruk Alpay, Bugra Kilictas
+- **abs:** https://arxiv.org/abs/2601.00938 · **PDF:** https://arxiv.org/pdf/2601.00938v1
+
+## 24545. TeleDoCTR: Domain-Specific and Contextual Troubleshooting for Telecommunications
+
+- **ID:** `2601.00691` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Mohamed Trabelsi, Huseyin Uzunalioglu
+- **abs:** https://arxiv.org/abs/2601.00691 · **PDF:** https://arxiv.org/pdf/2601.00691v1
+
+## 24546. Cost Optimization in Production Line Using Genetic Algorithm
+
+- **ID:** `2601.00689` · **primary:** `cs.NE` · **Submitted:** 2026-01-02
+- **Authors:** Alireza Rezaee
+- **abs:** https://arxiv.org/abs/2601.00689 · **PDF:** https://arxiv.org/pdf/2601.00689v1
+
+## 24547. Sigmoid Head for Quality Estimation under Language Ambiguity
+
+- **ID:** `2601.00680` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Tu Anh Dinh, Jan Niehues
+- **abs:** https://arxiv.org/abs/2601.00680 · **PDF:** https://arxiv.org/pdf/2601.00680v2
+
+## 24548. SpikySpace: A Spiking State Space Model for Energy-Efficient Time Series Forecasting
+
+- **ID:** `2601.02411` · **primary:** `cs.NE` · **Submitted:** 2026-01-02
+- **Authors:** Kaiwen Tang, Jiaqi Zheng, Yuze Jin, et al.
+- **abs:** https://arxiv.org/abs/2601.02411 · **PDF:** https://arxiv.org/pdf/2601.02411v2
+
+## 24549. QSLM: A Performance- and Memory-aware Quantization Framework with Tiered Search Strategy for Spike-driven Language Models
+
+- **ID:** `2601.00679` · **primary:** `cs.NE` · **Submitted:** 2026-01-02
+- **Authors:** Rachmad Vidya Wicaksana Putra, Pasindu Wickramasinghe, Muhammad Shafique
+- **abs:** https://arxiv.org/abs/2601.00679 · **PDF:** https://arxiv.org/pdf/2601.00679v1
+
+## 24550. IRPM: Intergroup Relative Preference Modeling for Pointwise Generative Reward Models
+
+- **ID:** `2601.00677` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Haonan Song, Qingchen Xie, Huan Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.00677 · **PDF:** https://arxiv.org/pdf/2601.00677v2
+
+## 24551. Sparse FEONet: A Low-Cost, Memory-Efficient Operator Network via Finite-Element Local Sparsity for Parametric PDEs
+
+- **ID:** `2601.00672` · **primary:** `math.NA` · **Submitted:** 2026-01-02
+- **Authors:** Seungchan Ko, Jiyeon Kim, Dongwook Shin
+- **abs:** https://arxiv.org/abs/2601.00672 · **PDF:** https://arxiv.org/pdf/2601.00672v1
+
+## 24552. Fast-weight Product Key Memory
+
+- **ID:** `2601.00671` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Tianyu Zhao, Llion Jones
+- **abs:** https://arxiv.org/abs/2601.00671 · **PDF:** https://arxiv.org/pdf/2601.00671v2
+
+## 24553. Three factor delay learning rules for spiking neural networks
+
+- **ID:** `2601.00668` · **primary:** `cs.NE` · **Submitted:** 2026-01-02
+- **Authors:** Luke Vassallo, Nima Taherinejad
+- **abs:** https://arxiv.org/abs/2601.00668 · **PDF:** https://arxiv.org/pdf/2601.00668v2
+
+## 24554. Avatar Forcing: Real-Time Interactive Head Avatar Generation for Natural Conversation
+
+- **ID:** `2601.00664` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Taekyung Ki, Sangwon Jang, Jaehyeong Jo, et al.
+- **abs:** https://arxiv.org/abs/2601.00664 · **PDF:** https://arxiv.org/pdf/2601.00664v1
+
+## 24555. Interpretability-Guided Bi-objective Optimization: Aligning Accuracy and Explainability
+
+- **ID:** `2601.00655` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Kasra Fouladi, Hamta Rahmani
+- **abs:** https://arxiv.org/abs/2601.00655 · **PDF:** https://arxiv.org/pdf/2601.00655v2
+
+## 24556. Physio-DPO: Aligning Large Language Models with the Protein Energy Landscape to Eliminate Structural Hallucinations
+
+- **ID:** `2601.00647` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** QiWei Meng
+- **abs:** https://arxiv.org/abs/2601.00647 · **PDF:** https://arxiv.org/pdf/2601.00647v1
+
+## 24557. Probabilistic Guarantees for Reducing Contextual Hallucinations in LLMs
+
+- **ID:** `2601.00641` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Nils Rautenberg, Sven Schippkus
+- **abs:** https://arxiv.org/abs/2601.00641 · **PDF:** https://arxiv.org/pdf/2601.00641v1
+
+## 24558. Emoji-Based Jailbreaking of Large Language Models
+
+- **ID:** `2601.00936` · **primary:** `cs.CR` · **Submitted:** 2026-01-02
+- **Authors:** M P V S Gopinadh, S Mahaboob Hussain
+- **abs:** https://arxiv.org/abs/2601.00936 · **PDF:** https://arxiv.org/pdf/2601.00936v1
+
+## 24559. Improving Code-Switching Speech Recognition with TTS Data Augmentation
+
+- **ID:** `2601.00935` · **primary:** `eess.AS` · **Submitted:** 2026-01-02
+- **Authors:** Yue Heng Yeo, Yuchen Hu, Shreyas Gopal, et al.
+- **abs:** https://arxiv.org/abs/2601.00935 · **PDF:** https://arxiv.org/pdf/2601.00935v1
+
+## 24560. HyperPriv-EPN: Hypergraph Learning with Privileged Knowledge for Ependymoma Prognosis
+
+- **ID:** `2601.00626` · **primary:** `cs.CV` · **Submitted:** 2026-01-02
+- **Authors:** Shuren Gabriel Yu, Sikang Ren, Yongji Tian
+- **abs:** https://arxiv.org/abs/2601.00626 · **PDF:** https://arxiv.org/pdf/2601.00626v1
+
+## 24561. Do Chatbot LLMs Talk Too Much? The YapBench Benchmark
+
+- **ID:** `2601.00624` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Vadim Borisov, Michael Gröger, Mina Mikhael, et al.
+- **abs:** https://arxiv.org/abs/2601.00624 · **PDF:** https://arxiv.org/pdf/2601.00624v1
+
+## 24562. DA-DPO: Cost-efficient Difficulty-aware Preference Optimization for Reducing MLLM Hallucinations
+
+- **ID:** `2601.00623` · **primary:** `cs.AI` · **Submitted:** 2026-01-02
+- **Authors:** Longtian Qiu, Shan Ning, Chuyu Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.00623 · **PDF:** https://arxiv.org/pdf/2601.00623v1
+
+## 24563. Noise-Robust Tiny Object Localization with Flows
+
+- **ID:** `2601.00617` · **primary:** `cs.CV` · **Submitted:** 2026-01-02
+- **Authors:** Huixin Sun, Linlin Yang, Ronyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.00617 · **PDF:** https://arxiv.org/pdf/2601.00617v1
+
+## 24564. Integrating Multi-Armed Bandit, Active Learning, and Distributed Computing for Scalable Optimization
+
+- **ID:** `2601.00615` · **primary:** `stat.CO` · **Submitted:** 2026-01-02
+- **Authors:** Foo Hui-Mean, Yuan-chin Ivan Chang
+- **abs:** https://arxiv.org/abs/2601.00615 · **PDF:** https://arxiv.org/pdf/2601.00615v1
+
+## 24565. Automatic Question Generation for Intuitive Learning Utilizing Causal Graph Guided Chain of Thought Reasoning
+
+- **ID:** `2601.06098` · **primary:** `cs.AI` · **Submitted:** 2026-01-02
+- **Authors:** Nicholas X. Wang, Neel V. Parpia, Aaryan D. Parikh, et al.
+- **abs:** https://arxiv.org/abs/2601.06098 · **PDF:** https://arxiv.org/pdf/2601.06098v1
+
+## 24566. Stronger Approximation Guarantees for Non-Monotone γ-Weakly DR-Submodular Maximization
+
+- **ID:** `2601.00611` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Hareshkumar Jadav, Ranveer Singh, Vaneet Aggarwal
+- **abs:** https://arxiv.org/abs/2601.00611 · **PDF:** https://arxiv.org/pdf/2601.00611v1
+
+## 24567. Traffic-Aware Optimal Taxi Placement Using Graph Neural Network-Based Reinforcement Learning
+
+- **ID:** `2601.00607` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Sonia Khetarpaul, P Y Sharan
+- **abs:** https://arxiv.org/abs/2601.00607 · **PDF:** https://arxiv.org/pdf/2601.00607v1
+
+## 24568. Cycling Race Time Prediction: A Personalized Machine Learning Approach Using Route Topology and Training Load
+
+- **ID:** `2601.00604` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Francisco Aguilera Moreno
+- **abs:** https://arxiv.org/abs/2601.00604 · **PDF:** https://arxiv.org/pdf/2601.00604v2
+
+## 24569. ChiEngMixBench: Evaluating Large Language Models on Spontaneous and Natural Chinese-English Code-Mixed Generation
+
+- **ID:** `2601.16217` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Qingyan Yang, Tongxi Wang, Yunsheng Luo
+- **abs:** https://arxiv.org/abs/2601.16217 · **PDF:** https://arxiv.org/pdf/2601.16217v1
+
+## 24570. LOFA: Online Influence Maximization under Full-Bandit Feedback using Lazy Forward Selection
+
+- **ID:** `2601.00933` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Jinyu Xu, Abhishek K. Umrawal
+- **abs:** https://arxiv.org/abs/2601.00933 · **PDF:** https://arxiv.org/pdf/2601.00933v1
+
+## 24571. Enhanced Data-Driven Product Development via Gradient Based Optimization and Conformalized Monte Carlo Dropout Uncertainty Estimation
+
+- **ID:** `2601.00932` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Andrea Thomas Nava, Lijo Johny, Fabio Azzalini, et al.
+- **abs:** https://arxiv.org/abs/2601.00932 · **PDF:** https://arxiv.org/pdf/2601.00932v1
+
+## 24572. Beyond IVR: Benchmarking Customer Support LLM Agents for Business-Adherence
+
+- **ID:** `2601.00596` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Sumanth Balaji, Piyush Mishra, Aashraya Sachdeva, et al.
+- **abs:** https://arxiv.org/abs/2601.00596 · **PDF:** https://arxiv.org/pdf/2601.00596v1
+
+## 24573. SearchGym: A Modular Infrastructure for Cross-Platform Benchmarking and Hybrid Search Orchestration
+
+- **ID:** `2603.04402` · **primary:** `cs.IR` · **Submitted:** 2026-01-02
+- **Authors:** Jerome Tze-Hou Hsu
+- **abs:** https://arxiv.org/abs/2603.04402 · **PDF:** https://arxiv.org/pdf/2603.04402v1
+
+## 24574. Entropy-Tree: Tree-Based Decoding with Entropy-Guided Exploration
+
+- **ID:** `2601.15296` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Longxuan Wei, Yubo Zhang, Zijiao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.15296 · **PDF:** https://arxiv.org/pdf/2601.15296v1
+
+## 24575. Uncertainty-Adjusted Sorting for Asset Pricing with Machine Learning
+
+- **ID:** `2601.00593` · **primary:** `q-fin.PM` · **Submitted:** 2026-01-02
+- **Authors:** Yan Liu, Ye Luo, Zigan Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.00593 · **PDF:** https://arxiv.org/pdf/2601.00593v1
+
+## 24576. Opening the Black Box: A Survey on the Mechanisms of Multi-Step Reasoning in Large Language Models
+
+- **ID:** `2601.14270` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Liangming Pan, Jason Liang, Jiaran Ye, et al.
+- **abs:** https://arxiv.org/abs/2601.14270 · **PDF:** https://arxiv.org/pdf/2601.14270v1
+
+## 24577. CSSBench: Evaluating the Safety of Lightweight LLMs against Chinese-Specific Adversarial Patterns
+
+- **ID:** `2601.00588` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Zhenhong Zhou, Shilinlu Yan, Chuanpu Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.00588 · **PDF:** https://arxiv.org/pdf/2601.00588v2
+
+## 24578. The Vibe-Check Protocol: Quantifying Cognitive Offloading in AI Programming
+
+- **ID:** `2601.02410` · **primary:** `cs.SE` · **Submitted:** 2026-01-02
+- **Authors:** Aizierjiang Aiersilan
+- **abs:** https://arxiv.org/abs/2601.02410 · **PDF:** https://arxiv.org/pdf/2601.02410v1
+
+## 24579. HFedMoE: Resource-aware Heterogeneous Federated Learning with Mixture-of-Experts
+
+- **ID:** `2601.00583` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Zihan Fang, Zheng Lin, Senkang Hu, et al.
+- **abs:** https://arxiv.org/abs/2601.00583 · **PDF:** https://arxiv.org/pdf/2601.00583v1
+
+## 24580. AceFF: A State-of-the-Art Machine Learning Potential for Small Molecules
+
+- **ID:** `2601.00581` · **primary:** `physics.chem-ph` · **Submitted:** 2026-01-02
+- **Authors:** Stephen E. Farr, Stefan Doerr, Antonio Mirarchi, et al.
+- **abs:** https://arxiv.org/abs/2601.00581 · **PDF:** https://arxiv.org/pdf/2601.00581v2
+
+## 24581. Priority-Aware Multi-Robot Coverage Path Planning
+
+- **ID:** `2601.00580` · **primary:** `cs.RO` · **Submitted:** 2026-01-02
+- **Authors:** Kanghoon Lee, Hyeonjun Kim, Jiachen Li, et al.
+- **abs:** https://arxiv.org/abs/2601.00580 · **PDF:** https://arxiv.org/pdf/2601.00580v1
+
+## 24582. The Slow Drift of Support: Boundary Failures in Multi-Turn Mental Health LLM Dialogues
+
+- **ID:** `2601.14269` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Youyou Cheng, Zhuangwei Kang, Kerry Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.14269 · **PDF:** https://arxiv.org/pdf/2601.14269v1
+
+## 24583. Learning to be Reproducible: Custom Loss Design for Robust Neural Networks
+
+- **ID:** `2601.00578` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Waqas Ahmed, Sheeba Samuel, Kevin Coakley, et al.
+- **abs:** https://arxiv.org/abs/2601.00578 · **PDF:** https://arxiv.org/pdf/2601.00578v1
+
+## 24584. Adversarial Samples Are Not Created Equal
+
+- **ID:** `2601.00577` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Jennifer Crawford, Amol Khanna, Fred Lu, et al.
+- **abs:** https://arxiv.org/abs/2601.00577 · **PDF:** https://arxiv.org/pdf/2601.00577v1
+
+## 24585. InfoSynth: Information-Guided Benchmark Synthesis for LLMs
+
+- **ID:** `2601.00575` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Ishir Garg, Neel Kolhe, Xuandong Zhao, et al.
+- **abs:** https://arxiv.org/abs/2601.00575 · **PDF:** https://arxiv.org/pdf/2601.00575v1
+
+## 24586. Benchmarking ERP Analysis: Manual Features, Deep Learning, and Foundation Models
+
+- **ID:** `2601.00573` · **primary:** `cs.NE` · **Submitted:** 2026-01-02
+- **Authors:** Yihe Wang, Zhiqiao Kang, Bohan Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.00573 · **PDF:** https://arxiv.org/pdf/2601.00573v1
+
+## 24587. Expert-Guided Explainable Few-Shot Learning with Active Sample Selection for Medical Image Analysis
+
+- **ID:** `2601.02409` · **primary:** `eess.IV` · **Submitted:** 2026-01-02
+- **Authors:** Longwei Wang, Ifrat Ikhtear Uddin, KC Santosh
+- **abs:** https://arxiv.org/abs/2601.02409 · **PDF:** https://arxiv.org/pdf/2601.02409v1
+
+## 24588. Improving Scientific Document Retrieval with Academic Concept Index
+
+- **ID:** `2601.00567` · **primary:** `cs.IR` · **Submitted:** 2026-01-02
+- **Authors:** Jeyun Lee, Junhyoung Lee, Wonbin Kweon, et al.
+- **abs:** https://arxiv.org/abs/2601.00567 · **PDF:** https://arxiv.org/pdf/2601.00567v1
+
+## 24589. Cracking IoT Security: Can LLMs Outsmart Static Analysis Tools?
+
+- **ID:** `2601.00559` · **primary:** `cs.CR` · **Submitted:** 2026-01-02
+- **Authors:** Jason Quantrill, Noura Khajehnouri, Zihan Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.00559 · **PDF:** https://arxiv.org/pdf/2601.00559v1
+
+## 24590. A Language-Agnostic Hierarchical LoRA-MoE Architecture for CTC-based Multilingual ASR
+
+- **ID:** `2601.00557` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Yuang Zheng, Dongxu Chen, Yuxiang Mei, et al.
+- **abs:** https://arxiv.org/abs/2601.00557 · **PDF:** https://arxiv.org/pdf/2601.00557v2
+
+## 24591. Entropy Production in Machine Learning Under Fokker-Planck Probability Flow
+
+- **ID:** `2601.00554` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Lennon Shikhman
+- **abs:** https://arxiv.org/abs/2601.00554 · **PDF:** https://arxiv.org/pdf/2601.00554v3
+
+## 24592. A Comprehensive Dataset for Human vs. AI Generated Image Detection
+
+- **ID:** `2601.00553` · **primary:** `cs.CV` · **Submitted:** 2026-01-02
+- **Authors:** Rajarshi Roy, Nasrin Imanpour, Ashhar Aziz, et al.
+- **abs:** https://arxiv.org/abs/2601.00553 · **PDF:** https://arxiv.org/pdf/2601.00553v1
+
+## 24593. CoCo-Fed: A Unified Framework for Memory- and Communication-Efficient Federated Learning at the Wireless Edge
+
+- **ID:** `2601.00549` · **primary:** `cs.IT` · **Submitted:** 2026-01-02
+- **Authors:** Zhiheng Guo, Zhaoyang Liu, Zihan Cen, et al.
+- **abs:** https://arxiv.org/abs/2601.00549 · **PDF:** https://arxiv.org/pdf/2601.00549v1
+
+## 24594. ECR: Manifold-Guided Semantic Cues for Compact Language Models
+
+- **ID:** `2601.00543` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Chung-Wei Victor Yuan
+- **abs:** https://arxiv.org/abs/2601.00543 · **PDF:** https://arxiv.org/pdf/2601.00543v1
+
+## 24595. AlignUSER: Human-Aligned LLM Agents via World Models for Recommender System Evaluation
+
+- **ID:** `2601.00930` · **primary:** `cs.IR` · **Submitted:** 2026-01-02
+- **Authors:** Nicolas Bougie, Gian Maria Marconi, Tony Yip, et al.
+- **abs:** https://arxiv.org/abs/2601.00930 · **PDF:** https://arxiv.org/pdf/2601.00930v1
+
+## 24596. Parametrized Sharing for Multi-Agent Hybrid DRL for Multiple Multi-Functional RISs-Aided Downlink NOMA Networks
+
+- **ID:** `2601.00538` · **primary:** `eess.SP` · **Submitted:** 2026-01-02
+- **Authors:** Chi-Te Kuo, Li-Hsiang Shen, Jyun-Jhe Huang
+- **abs:** https://arxiv.org/abs/2601.00538 · **PDF:** https://arxiv.org/pdf/2601.00538v2
+
+## 24597. Retrieval--Reasoning Processes for Multi-hop Question Answering: A Four-Axis Design Framework and Empirical Trends
+
+- **ID:** `2601.00536` · **primary:** `cs.CL` · **Submitted:** 2026-01-02
+- **Authors:** Yuelyu Ji, Zhuochun Li, Rui Meng, et al.
+- **abs:** https://arxiv.org/abs/2601.00536 · **PDF:** https://arxiv.org/pdf/2601.00536v1
+
+## 24598. Cloud-Native Generative AI for Automated Planogram Synthesis: A Diffusion Model Approach for Multi-Store Retail Optimization
+
+- **ID:** `2601.00527` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Ravi Teja Pagidoju, Shriya Agarwal
+- **abs:** https://arxiv.org/abs/2601.00527 · **PDF:** https://arxiv.org/pdf/2601.00527v1
+
+## 24599. Federated Customization of Large Models: Approaches, Experiments, and Insights
+
+- **ID:** `2601.00526` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Yuchuan Ye, Ming Ding, Youjia Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.00526 · **PDF:** https://arxiv.org/pdf/2601.00526v1
+
+## 24600. Analyzing the Shopping Journey: Computing Shelf Browsing Visits in a Physical Retail Store
+
+- **ID:** `2601.00928` · **primary:** `cs.CV` · **Submitted:** 2026-01-02
+- **Authors:** Luis Yoichi Morales, Francesco Zanlungo, David M. Woollard
+- **abs:** https://arxiv.org/abs/2601.00928 · **PDF:** https://arxiv.org/pdf/2601.00928v1
+
+## 24601. Optimizing LSTM Neural Networks for Resource-Constrained Retail Sales Forecasting: A Model Compression Study
+
+- **ID:** `2601.00525` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Ravi Teja Pagidoju
+- **abs:** https://arxiv.org/abs/2601.00525 · **PDF:** https://arxiv.org/pdf/2601.00525v1
+
+## 24602. Probability-Aware Parking Selection
+
+- **ID:** `2601.00521` · **primary:** `eess.SY` · **Submitted:** 2026-01-02
+- **Authors:** Cameron Hickert, Sirui Li, Zhengbing He, et al.
+- **abs:** https://arxiv.org/abs/2601.00521 · **PDF:** https://arxiv.org/pdf/2601.00521v2
+
+## 24603. Measuring Social Media Polarization Using Large Language Models and Heuristic Rules
+
+- **ID:** `2601.00927` · **primary:** `cs.SI` · **Submitted:** 2026-01-02
+- **Authors:** Jawad Chowdhury, Rezaur Rashid, Gabriel Terejanu
+- **abs:** https://arxiv.org/abs/2601.00927 · **PDF:** https://arxiv.org/pdf/2601.00927v1
+
+## 24604. A Sparse-Attention Deep Learning Model Integrating Heterogeneous Multimodal Features for Parkinson's Disease Severity Profiling
+
+- **ID:** `2601.00519` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Dristi Datta, Tanmoy Debnath, Minh Chau, et al.
+- **abs:** https://arxiv.org/abs/2601.00519 · **PDF:** https://arxiv.org/pdf/2601.00519v1
+
+## 24605. Generative Conditional Missing Imputation Networks
+
+- **ID:** `2601.00517` · **primary:** `stat.ML` · **Submitted:** 2026-01-02
+- **Authors:** George Sun, Yi-Hui Zhou
+- **abs:** https://arxiv.org/abs/2601.00517 · **PDF:** https://arxiv.org/pdf/2601.00517v1
+
+## 24606. Trajectory Guard -- A Lightweight, Sequence-Aware Model for Real-Time Anomaly Detection in Agentic AI
+
+- **ID:** `2601.00516` · **primary:** `cs.LG` · **Submitted:** 2026-01-02
+- **Authors:** Laksh Advani
+- **abs:** https://arxiv.org/abs/2601.00516 · **PDF:** https://arxiv.org/pdf/2601.00516v1
+
+## 24607. The Illusion of Insight in Reasoning Models
+
+- **ID:** `2601.00514` · **primary:** `cs.AI` · **Submitted:** 2026-01-02
+- **Authors:** Liv G. d'Aliberti, Manoel Horta Ribeiro
+- **abs:** https://arxiv.org/abs/2601.00514 · **PDF:** https://arxiv.org/pdf/2601.00514v1
+
+## 24608. Semantic Event Graphs for Long-Form Video Question Answering
+
+- **ID:** `2601.06097` · **primary:** `cs.CV` · **Submitted:** 2026-01-02
+- **Authors:** Aradhya Dixit, Tianxi Liang
+- **abs:** https://arxiv.org/abs/2601.06097 · **PDF:** https://arxiv.org/pdf/2601.06097v1
+
+## 24609. When Small Models Are Right for Wrong Reasons: Process Verification for Trustworthy Agents
+
+- **ID:** `2601.00513` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Laksh Advani
+- **abs:** https://arxiv.org/abs/2601.00513 · **PDF:** https://arxiv.org/pdf/2601.00513v1
+
+## 24610. A Chain-of-Thought Approach to Semantic Query Categorization in e-Commerce Taxonomies
+
+- **ID:** `2601.00510` · **primary:** `cs.IR` · **Submitted:** 2026-01-01
+- **Authors:** Jetlir Duraj, Ishita Khan, Kilian Merkelbach, et al.
+- **abs:** https://arxiv.org/abs/2601.00510 · **PDF:** https://arxiv.org/pdf/2601.00510v1
+
+## 24611. Improving LLM-Assisted Secure Code Generation through Retrieval-Augmented-Generation and Multi-Tool Feedback
+
+- **ID:** `2601.00509` · **primary:** `cs.CR` · **Submitted:** 2026-01-01
+- **Authors:** Vidyut Sriram, Sawan Pandita, Achintya Lakshmanan, et al.
+- **abs:** https://arxiv.org/abs/2601.00509 · **PDF:** https://arxiv.org/pdf/2601.00509v1
+
+## 24612. MACA: A Framework for Distilling Trustworthy LLMs into Efficient Retrievers
+
+- **ID:** `2601.00926` · **primary:** `cs.IR` · **Submitted:** 2026-01-01
+- **Authors:** Satya Swaroop Gudipudi, Sahil Girhepuje, Ponnurangam Kumaraguru, et al.
+- **abs:** https://arxiv.org/abs/2601.00926 · **PDF:** https://arxiv.org/pdf/2601.00926v1
+
+## 24613. Rule-Based Approaches to Atomic Sentence Extraction
+
+- **ID:** `2601.00506` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Lineesha Kamana, Akshita Ananda Subramanian, Mehuli Ghosh, et al.
+- **abs:** https://arxiv.org/abs/2601.00506 · **PDF:** https://arxiv.org/pdf/2601.00506v1
+
+## 24614. MotionPhysics: Learnable Motion Distillation for Text-Guided Simulation
+
+- **ID:** `2601.00504` · **primary:** `cs.CV` · **Submitted:** 2026-01-01
+- **Authors:** Miaowei Wang, Jakub Zadrożny, Oisin Mac Aodha, et al.
+- **abs:** https://arxiv.org/abs/2601.00504 · **PDF:** https://arxiv.org/pdf/2601.00504v1
+
+## 24615. Interpretable Machine Learning for Quantum-Informed Property Predictions in Artificial Sensing Materials
+
+- **ID:** `2601.00503` · **primary:** `physics.chem-ph` · **Submitted:** 2026-01-01
+- **Authors:** Li Chen, Leonardo Medrano Sandonas, Shirong Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.00503 · **PDF:** https://arxiv.org/pdf/2601.00503v1
+
+## 24616. Noise-Aware Named Entity Recognition for Historical VET Documents
+
+- **ID:** `2601.00488` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Alexander M. Esser, Jens Dörpinghaus
+- **abs:** https://arxiv.org/abs/2601.00488 · **PDF:** https://arxiv.org/pdf/2601.00488v1
+
+## 24617. Multi-Agent Coordinated Rename Refactoring
+
+- **ID:** `2601.00482` · **primary:** `cs.SE` · **Submitted:** 2026-01-01
+- **Authors:** Abhiram Bellur, Mohammed Raihan Ullah, Fraol Batole, et al.
+- **abs:** https://arxiv.org/abs/2601.00482 · **PDF:** https://arxiv.org/pdf/2601.00482v1
+
+## 24618. MAESTRO: Multi-Agent Evaluation Suite for Testing, Reliability, and Observability
+
+- **ID:** `2601.00481` · **primary:** `cs.NI` · **Submitted:** 2026-01-01
+- **Authors:** Tie Ma, Yixi Chen, Vaastav Anand, et al.
+- **abs:** https://arxiv.org/abs/2601.00481 · **PDF:** https://arxiv.org/pdf/2601.00481v1
+
+## 24619. The Hessian of tall-skinny networks is easy to invert
+
+- **ID:** `2601.06096` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Ali Rahimi
+- **abs:** https://arxiv.org/abs/2601.06096 · **PDF:** https://arxiv.org/pdf/2601.06096v3
+
+## 24620. Progressive Ideation using an Agentic AI Framework for Human-AI Co-Creation
+
+- **ID:** `2601.00475` · **primary:** `cs.AI` · **Submitted:** 2026-01-01
+- **Authors:** Sankar B, Srinidhi Ranjini Girish, Aadya Bharti, et al.
+- **abs:** https://arxiv.org/abs/2601.00475 · **PDF:** https://arxiv.org/pdf/2601.00475v1
+
+## 24621. Deep Neural Networks as Discrete Dynamical Systems: Implications for Physics-Informed Learning
+
+- **ID:** `2601.00473` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Abhisek Ganguly, Santosh Ansumali, Sauro Succi
+- **abs:** https://arxiv.org/abs/2601.00473 · **PDF:** https://arxiv.org/pdf/2601.00473v2
+
+## 24622. Laplacian Kernelized Bandit
+
+- **ID:** `2601.00461` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Shuang Wu, Arash A. Amini
+- **abs:** https://arxiv.org/abs/2601.00461 · **PDF:** https://arxiv.org/pdf/2601.00461v1
+
+## 24623. Combining Residual U-Net and Data Augmentation for Dense Temporal Segmentation of Spike Wave Discharges in Single-Channel EEG
+
+- **ID:** `2601.00459` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Saurav Sengupta, Scott Kilianski, Suchetha Sharma, et al.
+- **abs:** https://arxiv.org/abs/2601.00459 · **PDF:** https://arxiv.org/pdf/2601.00459v2
+
+## 24624. Geometric Regularization in Mixture-of-Experts: The Disconnect Between Weights and Activations
+
+- **ID:** `2601.00457` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Hyunjun Kim
+- **abs:** https://arxiv.org/abs/2601.00457 · **PDF:** https://arxiv.org/pdf/2601.00457v1
+
+## 24625. Deep Networks Learn Deep Hierarchical Models
+
+- **ID:** `2601.00455` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Amit Daniely
+- **abs:** https://arxiv.org/abs/2601.00455 · **PDF:** https://arxiv.org/pdf/2601.00455v1
+
+## 24626. Entropic Context Shaping: Information-Theoretic Filtering for Context-Aware LLM Agents
+
+- **ID:** `2601.11585` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Hyunjun Kim
+- **abs:** https://arxiv.org/abs/2601.11585 · **PDF:** https://arxiv.org/pdf/2601.11585v1
+
+## 24627. Defensive M2S: Training Guardrail Models on Compressed Multi-turn Conversations
+
+- **ID:** `2601.00454` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Hyunjun Kim
+- **abs:** https://arxiv.org/abs/2601.00454 · **PDF:** https://arxiv.org/pdf/2601.00454v1
+
+## 24628. Imitation from Observations with Trajectory-Level Generative Embeddings
+
+- **ID:** `2601.00452` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Yongtao Qu, Shangzhe Li, Weitong Zhang
+- **abs:** https://arxiv.org/abs/2601.00452 · **PDF:** https://arxiv.org/pdf/2601.00452v2
+
+## 24629. Controllable Concept Bottleneck Models
+
+- **ID:** `2601.00451` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Hongbin Lin, Chenyang Ren, Juangui Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.00451 · **PDF:** https://arxiv.org/pdf/2601.00451v1
+
+## 24630. Quadratic Unconstrained Binary Optimisation for Training and Regularisation of Binary Neural Networks
+
+- **ID:** `2601.00449` · **primary:** `math.OC` · **Submitted:** 2026-01-01
+- **Authors:** Jonas Christoffer Villumsen, Yusuke Sugita
+- **abs:** https://arxiv.org/abs/2601.00449 · **PDF:** https://arxiv.org/pdf/2601.00449v1
+
+## 24631. Language as Mathematical Structure: Examining Semantic Field Theory Against Language Games
+
+- **ID:** `2601.00448` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Dimitris Vartziotis
+- **abs:** https://arxiv.org/abs/2601.00448 · **PDF:** https://arxiv.org/pdf/2601.00448v1
+
+## 24632. A Comparative Study of Adaptation Strategies for Time Series Foundation Models in Anomaly Detection
+
+- **ID:** `2601.00446` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Miseon Park, Kijung Yoon
+- **abs:** https://arxiv.org/abs/2601.00446 · **PDF:** https://arxiv.org/pdf/2601.00446v1
+
+## 24633. Comparative Efficiency Analysis of Lightweight Transformer Models: A Multi-Domain Empirical Benchmark for Enterprise NLP Deployment
+
+- **ID:** `2601.00444` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Muhammad Shahmeer Khan
+- **abs:** https://arxiv.org/abs/2601.00444 · **PDF:** https://arxiv.org/pdf/2601.00444v1
+
+## 24634. Application of deep learning techniques in non-contrast computed tomography pulmonary angiogram for pulmonary embolism diagnosis
+
+- **ID:** `2601.00925` · **primary:** `cs.CV` · **Submitted:** 2026-01-01
+- **Authors:** I-Hsien Ting, Yi-Jun Tseng, Yu-Sheng Lin
+- **abs:** https://arxiv.org/abs/2601.00925 · **PDF:** https://arxiv.org/pdf/2601.00925v1
+
+## 24635. Toward Better Temporal Structures for Geopolitical Events Forecasting
+
+- **ID:** `2601.00430` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Kian Ahrabian, Eric Boxer, Jay Pujara
+- **abs:** https://arxiv.org/abs/2601.00430 · **PDF:** https://arxiv.org/pdf/2601.00430v2
+
+## 24636. Interpretable ML Under the Microscope: Performance, Meta-Features, and the Regression-Classification Predictability Gap
+
+- **ID:** `2601.00428` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Mattia Billa, Giovanni Orlandi, Veronica Guidetti, et al.
+- **abs:** https://arxiv.org/abs/2601.00428 · **PDF:** https://arxiv.org/pdf/2601.00428v2
+
+## 24637. Complexity-based code embeddings
+
+- **ID:** `2601.00924` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Rares Folea, Radu Iacob, Emil Slusanschi, et al.
+- **abs:** https://arxiv.org/abs/2601.00924 · **PDF:** https://arxiv.org/pdf/2601.00924v1
+
+## 24638. RMAAT: Astrocyte-Inspired Memory Compression and Replay for Efficient Long-Context Transformers
+
+- **ID:** `2601.00426` · **primary:** `cs.NE` · **Submitted:** 2026-01-01
+- **Authors:** Md Zesun Ahmed Mia, Malyaban Bal, Abhronil Sengupta
+- **abs:** https://arxiv.org/abs/2601.00426 · **PDF:** https://arxiv.org/pdf/2601.00426v2
+
+## 24639. Evolving Personalities in Chaos: An LLM-Augmented Framework for Character Discovery in the Iterated Prisoners Dilemma under Environmental Stress
+
+- **ID:** `2601.02407` · **primary:** `cs.NE` · **Submitted:** 2026-01-01
+- **Authors:** Oguzhan Yildirim
+- **abs:** https://arxiv.org/abs/2601.02407 · **PDF:** https://arxiv.org/pdf/2601.02407v1
+
+## 24640. E-GRPO: High Entropy Steps Drive Effective Reinforcement Learning for Flow Models
+
+- **ID:** `2601.00423` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Shengjun Zhang, Zhang Zhang, Chensheng Dai, et al.
+- **abs:** https://arxiv.org/abs/2601.00423 · **PDF:** https://arxiv.org/pdf/2601.00423v1
+
+## 24641. Can Semantic Methods Enhance Team Sports Tactics? A Methodology for Football with Broader Applications
+
+- **ID:** `2601.00421` · **primary:** `cs.AI` · **Submitted:** 2026-01-01
+- **Authors:** Alessio Di Rubbo, Mattia Neri, Remo Pareschi, et al.
+- **abs:** https://arxiv.org/abs/2601.00421 · **PDF:** https://arxiv.org/pdf/2601.00421v1
+
+## 24642. Secure, Verifiable, and Scalable Multi-Client Data Sharing via Consensus-Based Privacy-Preserving Data Distribution
+
+- **ID:** `2601.00418` · **primary:** `cs.CR` · **Submitted:** 2026-01-01
+- **Authors:** Prajwal Panth, Sahaj Raj Malla
+- **abs:** https://arxiv.org/abs/2601.00418 · **PDF:** https://arxiv.org/pdf/2601.00418v1
+
+## 24643. Deep Delta Learning
+
+- **ID:** `2601.00417` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Yifan Zhang, Yifeng Liu, Mengdi Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.00417 · **PDF:** https://arxiv.org/pdf/2601.00417v2
+
+## 24644. Breaking Task Impasses Quickly: Adaptive Neuro-Symbolic Learning for Open-World Robotics
+
+- **ID:** `2601.16985` · **primary:** `cs.RO` · **Submitted:** 2026-01-01
+- **Authors:** Pierrick Lorang
+- **abs:** https://arxiv.org/abs/2601.16985 · **PDF:** https://arxiv.org/pdf/2601.16985v1
+
+## 24645. Do LLMs Judge Distantly Supervised Named Entity Labels Well? Constructing the JudgeWEL Dataset
+
+- **ID:** `2601.00411` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Alistair Plum, Laura Bernardy, Tharindu Ranasinghe
+- **abs:** https://arxiv.org/abs/2601.00411 · **PDF:** https://arxiv.org/pdf/2601.00411v2
+
+## 24646. Context Collapse: In-Context Learning and Model Collapse
+
+- **ID:** `2601.00923` · **primary:** `cs.AI` · **Submitted:** 2026-01-01
+- **Authors:** Josef Ott
+- **abs:** https://arxiv.org/abs/2601.00923 · **PDF:** https://arxiv.org/pdf/2601.00923v1
+
+## 24647. Adaptive Causal Coordination Detection for Social Media: A Memory-Guided Framework with Semi-Supervised Learning
+
+- **ID:** `2601.00400` · **primary:** `cs.AI` · **Submitted:** 2026-01-01
+- **Authors:** Weng Ding, Yi Han, Mu-Jiang-Shan Wang
+- **abs:** https://arxiv.org/abs/2601.00400 · **PDF:** https://arxiv.org/pdf/2601.00400v1
+
+## 24648. Bit-politeia: An AI Agent Community in Blockchain
+
+- **ID:** `2601.11583` · **primary:** `cs.CY` · **Submitted:** 2026-01-01
+- **Authors:** Xing Yang
+- **abs:** https://arxiv.org/abs/2601.11583 · **PDF:** https://arxiv.org/pdf/2601.11583v1
+
+## 24649. Revati: Transparent GPU-Free Time-Warp Emulation for LLM Serving
+
+- **ID:** `2601.00397` · **primary:** `cs.DC` · **Submitted:** 2026-01-01
+- **Authors:** Amey Agrawal, Mayank Yadav, Sukrit Kumar, et al.
+- **abs:** https://arxiv.org/abs/2601.00397 · **PDF:** https://arxiv.org/pdf/2601.00397v1
+
+## 24650. Real-Time Human Detection for Aerial Captured Video Sequences via Deep Models
+
+- **ID:** `2601.00391` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Nouar AlDahoul, Aznul Qalid Md Sabri, Ali Mohammed Mansoor
+- **abs:** https://arxiv.org/abs/2601.00391 · **PDF:** https://arxiv.org/pdf/2601.00391v1
+
+## 24651. NOS-Gate: Queue-Aware Streaming IDS for Consumer Gateways under Timing-Controlled Evasion
+
+- **ID:** `2601.00389` · **primary:** `cs.CR` · **Submitted:** 2026-01-01
+- **Authors:** Muhammad Bilal, Omer Tariq, Hasan Ahmed
+- **abs:** https://arxiv.org/abs/2601.00389 · **PDF:** https://arxiv.org/pdf/2601.00389v1
+
+## 24652. Vision-Language Reasoning for Geolocalization: A Reinforcement Learning Approach
+
+- **ID:** `2601.00388` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Biao Wu, Meng Fang, Ling Chen, et al.
+- **abs:** https://arxiv.org/abs/2601.00388 · **PDF:** https://arxiv.org/pdf/2601.00388v2
+
+## 24653. Engineering Attack Vectors and Detecting Anomalies in Additive Manufacturing
+
+- **ID:** `2601.00384` · **primary:** `cs.CR` · **Submitted:** 2026-01-01
+- **Authors:** Md Mahbub Hasan, Marcus Sternhagen, Krishna Chandra Roy
+- **abs:** https://arxiv.org/abs/2601.00384 · **PDF:** https://arxiv.org/pdf/2601.00384v1
+
+## 24654. Overview of the SciHigh Track at FIRE 2025: Research Highlight Generation from Scientific Papers
+
+- **ID:** `2601.11582` · **primary:** `cs.CY` · **Submitted:** 2026-01-01
+- **Authors:** Tohida Rehman, Debarshi Kumar Sanyal, Samiran Chattopadhyay
+- **abs:** https://arxiv.org/abs/2601.11582 · **PDF:** https://arxiv.org/pdf/2601.11582v1
+
+## 24655. Word Frequency Counting Based on Serverless MapReduce
+
+- **ID:** `2601.00380` · **primary:** `cs.DC` · **Submitted:** 2026-01-01
+- **Authors:** Hanzhe Li, Bingchen Lin, Mengyuan Xu
+- **abs:** https://arxiv.org/abs/2601.00380 · **PDF:** https://arxiv.org/pdf/2601.00380v1
+
+## 24656. Traffic Simulation in Ad Hoc Network of Flying UAVs with Generative AI Adaptation
+
+- **ID:** `2602.13200` · **primary:** `cs.NI` · **Submitted:** 2026-01-01
+- **Authors:** Andrii Grekhov, Volodymyr Kharchenko, Vasyl Kondratiuk
+- **abs:** https://arxiv.org/abs/2602.13200 · **PDF:** https://arxiv.org/pdf/2602.13200v1
+
+## 24657. Sparse Tucker Decomposition and Graph Regularization for High-Dimensional Time Series Forecasting
+
+- **ID:** `2601.00377` · **primary:** `math.ST` · **Submitted:** 2026-01-01
+- **Authors:** Sijia Xia, Michael K. Ng, Xiongjun Zhang
+- **abs:** https://arxiv.org/abs/2601.00377 · **PDF:** https://arxiv.org/pdf/2601.00377v1
+
+## 24658. In Line with Context: Repository-Level Code Generation via Context Inlining
+
+- **ID:** `2601.00376` · **primary:** `cs.SE` · **Submitted:** 2026-01-01
+- **Authors:** Chao Hu, Wenhao Zeng, Yuling Shi, et al.
+- **abs:** https://arxiv.org/abs/2601.00376 · **PDF:** https://arxiv.org/pdf/2601.00376v1
+
+## 24659. PatchBlock: A Lightweight Defense Against Adversarial Patches for Embedded EdgeAI Devices
+
+- **ID:** `2601.00367` · **primary:** `cs.CR` · **Submitted:** 2026-01-01
+- **Authors:** Nandish Chattopadhyay, Abdul Basit, Amira Guesmi, et al.
+- **abs:** https://arxiv.org/abs/2601.00367 · **PDF:** https://arxiv.org/pdf/2601.00367v1
+
+## 24660. BERT-JEPA: Reorganizing CLS Embeddings for Language-Invariant Semantics
+
+- **ID:** `2601.00366` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Taj Gillin, Adam Lalani, Kenneth Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.00366 · **PDF:** https://arxiv.org/pdf/2601.00366v1
+
+## 24661. The Role of Mixed-Language Documents for Multilingual Large Language Model Pretraining
+
+- **ID:** `2601.00364` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Jiandong Shao, Raphael Tang, Crystina Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.00364 · **PDF:** https://arxiv.org/pdf/2601.00364v2
+
+## 24662. Deterministic Coreset for Lp Subspace
+
+- **ID:** `2601.00361` · **primary:** `cs.DS` · **Submitted:** 2026-01-01
+- **Authors:** Rachit Chhaya, Anirban Dasgupta, Dan Feldman, et al.
+- **abs:** https://arxiv.org/abs/2601.00361 · **PDF:** https://arxiv.org/pdf/2601.00361v2
+
+## 24663. Mapping Human Anti-collusion Mechanisms to Multi-agent AI
+
+- **ID:** `2601.00360` · **primary:** `cs.MA` · **Submitted:** 2026-01-01
+- **Authors:** Jamiu Adekunle Idowu, Ahmed Almasoud, Ayman Alfahid
+- **abs:** https://arxiv.org/abs/2601.00360 · **PDF:** https://arxiv.org/pdf/2601.00360v1
+
+## 24664. Deep Q-Network Based Resilient Drone Communication:Neutralizing First-Order Markov Jammers
+
+- **ID:** `2601.06095` · **primary:** `cs.IT` · **Submitted:** 2026-01-01
+- **Authors:** Andrii Grekhov, Volodymyr Kharchenko, Vasyl Kondratiuk
+- **abs:** https://arxiv.org/abs/2601.06095 · **PDF:** https://arxiv.org/pdf/2601.06095v1
+
+## 24665. Robust Uncertainty Quantification for Factual Generation of Large Language Models
+
+- **ID:** `2601.00348` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Yuhao Zhang, Zhongliang Yang, Linna Zhou
+- **abs:** https://arxiv.org/abs/2601.00348 · **PDF:** https://arxiv.org/pdf/2601.00348v1
+
+## 24666. Solving nonlinear subsonic compressible flow in infinite domain via multi-stage neural networks
+
+- **ID:** `2601.00342` · **primary:** `physics.flu-dyn` · **Submitted:** 2026-01-01
+- **Authors:** Xuehui Qian, Hongkai Tao, Yongji Wang
+- **abs:** https://arxiv.org/abs/2601.00342 · **PDF:** https://arxiv.org/pdf/2601.00342v1
+
+## 24667. Bio-inspired Agentic Self-healing Framework for Resilient Distributed Computing Continuum Systems
+
+- **ID:** `2601.00339` · **primary:** `cs.AI` · **Submitted:** 2026-01-01
+- **Authors:** Alaa Saleh, Praveen Kumar Donta, Roberto Morabito, et al.
+- **abs:** https://arxiv.org/abs/2601.00339 · **PDF:** https://arxiv.org/pdf/2601.00339v1
+
+## 24668. Practical Geometric and Quantum Kernel Methods for Predicting Skeletal Muscle Outcomes in chronic obstructive pulmonary disease
+
+- **ID:** `2601.00921` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Azadeh Alavi, Hamidreza Khalili, Stanley H. Chan, et al.
+- **abs:** https://arxiv.org/abs/2601.00921 · **PDF:** https://arxiv.org/pdf/2601.00921v1
+
+## 24669. Smart Fault Detection in Nanosatellite Electrical Power System
+
+- **ID:** `2601.00335` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Alireza Rezaee, Niloofar Nobahari, Amin Asgarifar, et al.
+- **abs:** https://arxiv.org/abs/2601.00335 · **PDF:** https://arxiv.org/pdf/2601.00335v1
+
+## 24670. Sparse Probabilistic Coalition Structure Generation: Bayesian Greedy Pursuit and $\ell_1$ Relaxations
+
+- **ID:** `2601.00329` · **primary:** `cs.GT` · **Submitted:** 2026-01-01
+- **Authors:** Angshul Majumdar
+- **abs:** https://arxiv.org/abs/2601.00329 · **PDF:** https://arxiv.org/pdf/2601.00329v1
+
+## 24671. HarmoniAD: Harmonizing Local Structures and Global Semantics for Anomaly Detection
+
+- **ID:** `2601.00327` · **primary:** `cs.CV` · **Submitted:** 2026-01-01
+- **Authors:** Naiqi Zhang, Chuancheng Shi, Jingtong Dou, et al.
+- **abs:** https://arxiv.org/abs/2601.00327 · **PDF:** https://arxiv.org/pdf/2601.00327v1
+
+## 24672. Multiagent Reinforcement Learning for Liquidity Games
+
+- **ID:** `2601.00324` · **primary:** `cs.AI` · **Submitted:** 2026-01-01
+- **Authors:** Alicia Vidler, Gal A. Kaminka
+- **abs:** https://arxiv.org/abs/2601.00324 · **PDF:** https://arxiv.org/pdf/2601.00324v1
+
+## 24673. $α^3$-Bench: A Unified Benchmark of Safety, Robustness, and Efficiency for LLM-Based UAV Agents over 6G Networks
+
+- **ID:** `2601.03281` · **primary:** `eess.SY` · **Submitted:** 2026-01-01
+- **Authors:** Mohamed Amine Ferrag, Abderrahmane Lakas, Merouane Debbah
+- **abs:** https://arxiv.org/abs/2601.03281 · **PDF:** https://arxiv.org/pdf/2601.03281v1
+
+## 24674. Quantum King-Ring Domination in Chess: A QAOA Approach
+
+- **ID:** `2601.00318` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Gerhard Stenzel, Michael Kölle, Tobias Rohe, et al.
+- **abs:** https://arxiv.org/abs/2601.00318 · **PDF:** https://arxiv.org/pdf/2601.00318v1
+
+## 24675. MODE: Efficient Time Series Prediction with Mamba Enhanced by Low-Rank Neural ODEs
+
+- **ID:** `2601.00920` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Xingsheng Chen, Regina Zhang, Bo Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.00920 · **PDF:** https://arxiv.org/pdf/2601.00920v2
+
+## 24676. Can Optimal Transport Improve Federated Inverse Reinforcement Learning?
+
+- **ID:** `2601.00309` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** David Millard, Ali Baheri
+- **abs:** https://arxiv.org/abs/2601.00309 · **PDF:** https://arxiv.org/pdf/2601.00309v1
+
+## 24677. VisNet: Efficient Person Re-Identification via Alpha-Divergence Loss, Feature Fusion and Dynamic Multi-Task Learning
+
+- **ID:** `2601.00307` · **primary:** `cs.CV` · **Submitted:** 2026-01-01
+- **Authors:** Anns Ijaz, Muhammad Azeem Javed
+- **abs:** https://arxiv.org/abs/2601.00307 · **PDF:** https://arxiv.org/pdf/2601.00307v1
+
+## 24678. The Generative AI Paradox: GenAI and the Erosion of Trust, the Corrosion of Information Verification, and the Demise of Truth
+
+- **ID:** `2601.00306` · **primary:** `cs.CY` · **Submitted:** 2026-01-01
+- **Authors:** Emilio Ferrara
+- **abs:** https://arxiv.org/abs/2601.00306 · **PDF:** https://arxiv.org/pdf/2601.00306v1
+
+## 24679. DepFlow: Disentangled Speech Generation to Mitigate Semantic Bias in Depression Detection
+
+- **ID:** `2601.00303` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Yuxin Li, Xiangyu Zhang, Yifei Li, et al.
+- **abs:** https://arxiv.org/abs/2601.00303 · **PDF:** https://arxiv.org/pdf/2601.00303v1
+
+## 24680. ClinicalReTrial: A Self-Evolving AI Agent for Clinical Trial Protocol Optimization
+
+- **ID:** `2601.00290` · **primary:** `cs.AI` · **Submitted:** 2026-01-01
+- **Authors:** Sixue Xing, Xuanye Xia, Kerui Wu, et al.
+- **abs:** https://arxiv.org/abs/2601.00290 · **PDF:** https://arxiv.org/pdf/2601.00290v1
+
+## 24681. Identification and Estimation under Multiple Versions of Treatment: Mixture-of-Experts Approach
+
+- **ID:** `2601.00287` · **primary:** `stat.ME` · **Submitted:** 2026-01-01
+- **Authors:** Kohei Yoshikawa, Shuichi Kawano
+- **abs:** https://arxiv.org/abs/2601.00287 · **PDF:** https://arxiv.org/pdf/2601.00287v1
+
+## 24682. Towards Automated Differential Diagnosis of Skin Diseases Using Deep Learning and Imbalance-Aware Strategies
+
+- **ID:** `2601.00286` · **primary:** `cs.CV` · **Submitted:** 2026-01-01
+- **Authors:** Ali Anaissi, Ali Braytee, Weidong Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.00286 · **PDF:** https://arxiv.org/pdf/2601.00286v1
+
+## 24683. Deep learning estimation of the spectral density of functional time series on large domains
+
+- **ID:** `2601.00284` · **primary:** `stat.ME` · **Submitted:** 2026-01-01
+- **Authors:** Neda Mohammadi, Soham Sarkar, Piotr Kokoszka
+- **abs:** https://arxiv.org/abs/2601.00284 · **PDF:** https://arxiv.org/pdf/2601.00284v1
+
+## 24684. Can Large Language Models Still Explain Themselves? Investigating the Impact of Quantization on Self-Explanations
+
+- **ID:** `2601.00282` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Qianli Wang, Nils Feldhus, Pepa Atanasova, et al.
+- **abs:** https://arxiv.org/abs/2601.00282 · **PDF:** https://arxiv.org/pdf/2601.00282v1
+
+## 24685. Benchmarking Preprocessing and Integration Methods in Single-Cell Genomics
+
+- **ID:** `2601.00277` · **primary:** `q-bio.QM` · **Submitted:** 2026-01-01
+- **Authors:** Ali Anaissi, Seid Miad Zandavi, Weidong Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.00277 · **PDF:** https://arxiv.org/pdf/2601.00277v1
+
+## 24686. Task-Driven Kernel Flows: Label Rank Compression and Laplacian Spectral Filtering
+
+- **ID:** `2601.00276` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Hongxi Li, Chunlin Huang
+- **abs:** https://arxiv.org/abs/2601.00276 · **PDF:** https://arxiv.org/pdf/2601.00276v1
+
+## 24687. Simulation-Based Study of AI-Assisted Channel Adaptation in UAV-Enabled Cellular Networks
+
+- **ID:** `2602.13199` · **primary:** `cs.NI` · **Submitted:** 2026-01-01
+- **Authors:** Andrii Grekhov, Volodymyr Kharchenko, Vasyl Kondratiuk
+- **abs:** https://arxiv.org/abs/2602.13199 · **PDF:** https://arxiv.org/pdf/2602.13199v1
+
+## 24688. Vehicle Painting Robot Path Planning Using Hierarchical Optimization
+
+- **ID:** `2601.00271` · **primary:** `cs.RO` · **Submitted:** 2026-01-01
+- **Authors:** Yuya Nagai, Hiromitsu Nakamura, Narito Shinmachi, et al.
+- **abs:** https://arxiv.org/abs/2601.00271 · **PDF:** https://arxiv.org/pdf/2601.00271v1
+
+## 24689. Rectifying Adversarial Examples Using Their Vulnerabilities
+
+- **ID:** `2601.00270` · **primary:** `cs.CR` · **Submitted:** 2026-01-01
+- **Authors:** Fumiya Morimoto, Ryuto Morita, Satoshi Ono
+- **abs:** https://arxiv.org/abs/2601.00270 · **PDF:** https://arxiv.org/pdf/2601.00270v1
+
+## 24690. FaithSCAN: Model-Driven Single-Pass Hallucination Detection for Faithful Visual Question Answering
+
+- **ID:** `2601.00269` · **primary:** `cs.CV` · **Submitted:** 2026-01-01
+- **Authors:** Chaodong Tong, Qi Zhang, Chen Li, et al.
+- **abs:** https://arxiv.org/abs/2601.00269 · **PDF:** https://arxiv.org/pdf/2601.00269v3
+
+## 24691. Beyond Perfect APIs: A Comprehensive Evaluation of LLM Agents Under Real-World API Complexity
+
+- **ID:** `2601.00268` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Doyoung Kim, Zhiwei Ren, Jie Hao, et al.
+- **abs:** https://arxiv.org/abs/2601.00268 · **PDF:** https://arxiv.org/pdf/2601.00268v1
+
+## 24692. Parallel Universes, Parallel Languages: A Comprehensive Study on LLM-based Multilingual Counterfactual Example Generation
+
+- **ID:** `2601.00263` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Qianli Wang, Van Bach Nguyen, Yihong Liu, et al.
+- **abs:** https://arxiv.org/abs/2601.00263 · **PDF:** https://arxiv.org/pdf/2601.00263v1
+
+## 24693. Attention Needs to Focus: A Unified Perspective on Attention Allocation
+
+- **ID:** `2601.00919` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Zichuan Fu, Wentao Song, Guojing Li, et al.
+- **abs:** https://arxiv.org/abs/2601.00919 · **PDF:** https://arxiv.org/pdf/2601.00919v2
+
+## 24694. Enhancing the QA Model through a Multi-domain Debiasing Framework
+
+- **ID:** `2601.11581` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Yuefeng Wang, ChangJae Lee
+- **abs:** https://arxiv.org/abs/2601.11581 · **PDF:** https://arxiv.org/pdf/2601.11581v1
+
+## 24695. Next Generation Intelligent Low-Altitude Economy Deployments: The O-RAN Perspective
+
+- **ID:** `2601.00257` · **primary:** `eess.SY` · **Submitted:** 2026-01-01
+- **Authors:** Aly Sabri Abdalla, Vuk Marojevic
+- **abs:** https://arxiv.org/abs/2601.00257 · **PDF:** https://arxiv.org/pdf/2601.00257v1
+
+## 24696. An Empirical Evaluation of LLM-Based Approaches for Code Vulnerability Detection: RAG, SFT, and Dual-Agent Systems
+
+- **ID:** `2601.00254` · **primary:** `cs.SE` · **Submitted:** 2026-01-01
+- **Authors:** Md Hasan Saju, Maher Muhtadi, Akramul Azim
+- **abs:** https://arxiv.org/abs/2601.00254 · **PDF:** https://arxiv.org/pdf/2601.00254v1
+
+## 24697. Modern Neuromorphic AI: From Intra-Token to Inter-Token Processing
+
+- **ID:** `2601.00245` · **primary:** `cs.NE` · **Submitted:** 2026-01-01
+- **Authors:** Osvaldo Simeone
+- **abs:** https://arxiv.org/abs/2601.00245 · **PDF:** https://arxiv.org/pdf/2601.00245v3
+
+## 24698. Neural Minimum Weight Perfect Matching for Quantum Error Codes
+
+- **ID:** `2601.00242` · **primary:** `quant-ph` · **Submitted:** 2026-01-01
+- **Authors:** Yotam Peled, David Zenati, Eliya Nachmani
+- **abs:** https://arxiv.org/abs/2601.00242 · **PDF:** https://arxiv.org/pdf/2601.00242v1
+
+## 24699. When Agents See Humans as the Outgroup: Belief-Dependent Bias in LLM-Powered Agents
+
+- **ID:** `2601.00240` · **primary:** `cs.AI` · **Submitted:** 2026-01-01
+- **Authors:** Zongwei Wang, Bincheng Gu, Hongyu Yu, et al.
+- **abs:** https://arxiv.org/abs/2601.00240 · **PDF:** https://arxiv.org/pdf/2601.00240v2
+
+## 24700. Application Research of a Deep Learning Model Integrating CycleGAN and YOLO in PCB Infrared Defect Detection
+
+- **ID:** `2601.00237` · **primary:** `cs.CV` · **Submitted:** 2026-01-01
+- **Authors:** Chao Yang, Haoyuan Zheng, Yue Ma
+- **abs:** https://arxiv.org/abs/2601.00237 · **PDF:** https://arxiv.org/pdf/2601.00237v1
+
+## 24701. GRIT -- Geometry-Aware PEFT with K-FACPreconditioning, Fisher-Guided Reprojection, andDynamic Rank Adaptation
+
+- **ID:** `2601.00231` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Pritish Saha, Chandrav Rajbangshi, Rudra Goyal, et al.
+- **abs:** https://arxiv.org/abs/2601.00231 · **PDF:** https://arxiv.org/pdf/2601.00231v1
+
+## 24702. Robust Graph Fine-Tuning with Adversarial Graph Prompting
+
+- **ID:** `2601.00229` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Ziyan Zhang, Bo Jiang, Jin Tang
+- **abs:** https://arxiv.org/abs/2601.00229 · **PDF:** https://arxiv.org/pdf/2601.00229v1
+
+## 24703. FlashInfer-Bench: Building the Virtuous Cycle for AI-driven LLM Systems
+
+- **ID:** `2601.00227` · **primary:** `cs.AI` · **Submitted:** 2026-01-01
+- **Authors:** Shanli Xing, Yiyan Zhai, Alexander Jiang, et al.
+- **abs:** https://arxiv.org/abs/2601.00227 · **PDF:** https://arxiv.org/pdf/2601.00227v1
+
+## 24704. Talk Less, Verify More: Improving LLM Assistants with Semantic Checks and Execution Feedback
+
+- **ID:** `2601.00224` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Yan Sun, Ming Cai, Stanley Kok
+- **abs:** https://arxiv.org/abs/2601.00224 · **PDF:** https://arxiv.org/pdf/2601.00224v2
+
+## 24705. JP-TL-Bench: Anchored Pairwise LLM Evaluation for Bidirectional Japanese-English Translation
+
+- **ID:** `2601.00223` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Leonard Lin, Adam Lensenmayer
+- **abs:** https://arxiv.org/abs/2601.00223 · **PDF:** https://arxiv.org/pdf/2601.00223v1
+
+## 24706. Unknown Aware AI-Generated Content Attribution
+
+- **ID:** `2601.00218` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Ellie Thieu, Jifan Zhang, Haoyue Bai
+- **abs:** https://arxiv.org/abs/2601.00218 · **PDF:** https://arxiv.org/pdf/2601.00218v1
+
+## 24707. Mitigating Latent Mismatch in cVAE-Based Singing Voice Synthesis via Flow Matching
+
+- **ID:** `2601.00217` · **primary:** `cs.SD` · **Submitted:** 2026-01-01
+- **Authors:** Minhyeok Yun, Yong-Hoon Choi
+- **abs:** https://arxiv.org/abs/2601.00217 · **PDF:** https://arxiv.org/pdf/2601.00217v2
+
+## 24708. Latent-Constrained Conditional VAEs for Augmenting Large-Scale Climate Ensembles
+
+- **ID:** `2601.00915` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Jacquelyn Shelton, Przemyslaw Polewski, Alexander Robel, et al.
+- **abs:** https://arxiv.org/abs/2601.00915 · **PDF:** https://arxiv.org/pdf/2601.00915v1
+
+## 24709. From Evidence-Based Medicine to Knowledge Graph: Retrieval-Augmented Generation for Sports Rehabilitation and a Domain Benchmark
+
+- **ID:** `2601.00216` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Jinning Zhang, Jie Song, Wenhui Tu, et al.
+- **abs:** https://arxiv.org/abs/2601.00216 · **PDF:** https://arxiv.org/pdf/2601.00216v2
+
+## 24710. From Sight to Insight: Improving Visual Reasoning Capabilities of Multimodal Models via Reinforcement Learning
+
+- **ID:** `2601.00215` · **primary:** `cs.CV` · **Submitted:** 2026-01-01
+- **Authors:** Omar Sharif, Eftekhar Hossain, Patrick Ng
+- **abs:** https://arxiv.org/abs/2601.00215 · **PDF:** https://arxiv.org/pdf/2601.00215v1
+
+## 24711. Overlooked Safety Vulnerability in LLMs: Malicious Intelligent Optimization Algorithm Request and its Jailbreak
+
+- **ID:** `2601.00213` · **primary:** `cs.CR` · **Submitted:** 2026-01-01
+- **Authors:** Haoran Gu, Handing Wang, Yi Mei, et al.
+- **abs:** https://arxiv.org/abs/2601.00213 · **PDF:** https://arxiv.org/pdf/2601.00213v1
+
+## 24712. Clean-GS: Semantic Mask-Guided Pruning for 3D Gaussian Splatting
+
+- **ID:** `2601.00913` · **primary:** `cs.CV` · **Submitted:** 2026-01-01
+- **Authors:** Subhankar Mishra
+- **abs:** https://arxiv.org/abs/2601.00913 · **PDF:** https://arxiv.org/pdf/2601.00913v1
+
+## 24713. Knowledge Distillation for Temporal Knowledge Graph Reasoning with Large Language Models
+
+- **ID:** `2601.00202` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Wang Xing, Wei Song, Siyu Lin, et al.
+- **abs:** https://arxiv.org/abs/2601.00202 · **PDF:** https://arxiv.org/pdf/2601.00202v1
+
+## 24714. The Discovery Gap: How Product Hunt Startups Vanish in LLM Organic Discovery Queries
+
+- **ID:** `2601.00912` · **primary:** `cs.IR` · **Submitted:** 2026-01-01
+- **Authors:** Amit Prakash Sharma
+- **abs:** https://arxiv.org/abs/2601.00912 · **PDF:** https://arxiv.org/pdf/2601.00912v1
+
+## 24715. Device-Native Autonomous Agents for Privacy-Preserving Negotiations
+
+- **ID:** `2601.00911` · **primary:** `cs.CR` · **Submitted:** 2026-01-01
+- **Authors:** Joyjit Roy, Samaresh Kumar Singh
+- **abs:** https://arxiv.org/abs/2601.00911 · **PDF:** https://arxiv.org/pdf/2601.00911v2
+
+## 24716. Detecting Unobserved Confounders: A Kernelized Regression Approach
+
+- **ID:** `2601.00200` · **primary:** `stat.ML` · **Submitted:** 2026-01-01
+- **Authors:** Yikai Chen, Yunxin Mao, Chunyuan Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.00200 · **PDF:** https://arxiv.org/pdf/2601.00200v1
+
+## 24717. StockBot 2.0: Vanilla LSTMs Outperform Transformer-based Forecasting for Stock Prices
+
+- **ID:** `2601.00197` · **primary:** `cs.CE` · **Submitted:** 2026-01-01
+- **Authors:** Shaswat Mohanty
+- **abs:** https://arxiv.org/abs/2601.00197 · **PDF:** https://arxiv.org/pdf/2601.00197v1
+
+## 24718. Optimized Hybrid Feature Engineering for Resource-Efficient Arrhythmia Detection in ECG Signals: An Optimization Framework
+
+- **ID:** `2601.00192` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Moirangthem Tiken Singh, Manibhushan Yaikhom
+- **abs:** https://arxiv.org/abs/2601.00192 · **PDF:** https://arxiv.org/pdf/2601.00192v1
+
+## 24719. SSI-GAN: Semi-Supervised Swin-Inspired Generative Adversarial Networks for Neuronal Spike Classification
+
+- **ID:** `2601.00189` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Danial Sharifrazi, Nouman Javed, Mojtaba Mohammadi, et al.
+- **abs:** https://arxiv.org/abs/2601.00189 · **PDF:** https://arxiv.org/pdf/2601.00189v1
+
+## 24720. Reinforcement-Learned Unequal Error Protection for Quantized Semantic Embeddings
+
+- **ID:** `2601.00186` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Moirangthem Tiken Singh, Adnan Arif
+- **abs:** https://arxiv.org/abs/2601.00186 · **PDF:** https://arxiv.org/pdf/2601.00186v1
+
+## 24721. Understanding Emotion in Discourse: Recognition Insights and Linguistic Patterns for Generation
+
+- **ID:** `2601.00181` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Cheonkam Jeong, Adeline Nyamathi
+- **abs:** https://arxiv.org/abs/2601.00181 · **PDF:** https://arxiv.org/pdf/2601.00181v2
+
+## 24722. Early Prediction of Liver Cirrhosis Up to Three Years in Advance: A Machine Learning Study Benchmarking Against the FIB-4 Score
+
+- **ID:** `2601.00175` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Zhuqi Miao, Sujan Ravi, Abdulaziz Ahmed
+- **abs:** https://arxiv.org/abs/2601.00175 · **PDF:** https://arxiv.org/pdf/2601.00175v1
+
+## 24723. Sequential Reservoir Computing for Efficient High-Dimensional Spatiotemporal Forecasting
+
+- **ID:** `2601.00172` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Ata Akbari Asanjan, Filip Wudarski, Daniel O'Connor, et al.
+- **abs:** https://arxiv.org/abs/2601.00172 · **PDF:** https://arxiv.org/pdf/2601.00172v1
+
+## 24724. Hear the Heartbeat in Phases: Physiologically Grounded Phase-Aware ECG Biometrics
+
+- **ID:** `2601.00170` · **primary:** `eess.IV` · **Submitted:** 2026-01-01
+- **Authors:** Jintao Huang, Lu Leng, Yi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.00170 · **PDF:** https://arxiv.org/pdf/2601.00170v1
+
+## 24725. Online Finetuning Decision Transformers with Pure RL Gradients
+
+- **ID:** `2601.00167` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Junkai Luo, Yinglun Zhu
+- **abs:** https://arxiv.org/abs/2601.00167 · **PDF:** https://arxiv.org/pdf/2601.00167v1
+
+## 24726. Pat-DEVAL: Chain-of-Legal-Thought Evaluation for Patent Description
+
+- **ID:** `2601.00166` · **primary:** `cs.CL` · **Submitted:** 2026-01-01
+- **Authors:** Yongmin Yoo, Kris W Pan
+- **abs:** https://arxiv.org/abs/2601.00166 · **PDF:** https://arxiv.org/pdf/2601.00166v1
+
+## 24727. Security Hardening Using FABRIC: Implementing a Unified Compliance Aggregator for Linux Servers
+
+- **ID:** `2601.00909` · **primary:** `cs.CR` · **Submitted:** 2026-01-01
+- **Authors:** Sheldon Paul, Izzat Alsmadi
+- **abs:** https://arxiv.org/abs/2601.00909 · **PDF:** https://arxiv.org/pdf/2601.00909v1
+
+## 24728. Conformal Prediction Under Distribution Shift: A COVID-19 Natural Experiment
+
+- **ID:** `2601.00908` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Chorok Lee
+- **abs:** https://arxiv.org/abs/2601.00908 · **PDF:** https://arxiv.org/pdf/2601.00908v1
+
+## 24729. The Weather Paradox: Why Precipitation Fails to Predict Traffic Accident Severity in Large-Scale US Data
+
+- **ID:** `2601.00152` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Yann Bellec, Rohan Kaman, Siwen Cui, et al.
+- **abs:** https://arxiv.org/abs/2601.00152 · **PDF:** https://arxiv.org/pdf/2601.00152v1
+
+## 24730. Reinforcement Learning with Function Approximation for Non-Markov Processes
+
+- **ID:** `2601.00151` · **primary:** `cs.LG` · **Submitted:** 2026-01-01
+- **Authors:** Ali Devran Kara
+- **abs:** https://arxiv.org/abs/2601.00151 · **PDF:** https://arxiv.org/pdf/2601.00151v1
+
+## 24731. FCMBench: The First Large-scale Financial Credit Multimodal Benchmark for Real-world Applications
+
+- **ID:** `2601.00150` · **primary:** `cs.CV` · **Submitted:** 2026-01-01
+- **Authors:** Yehui Yang, Dalu Yang, Fangxin Shang, et al.
+- **abs:** https://arxiv.org/abs/2601.00150 · **PDF:** https://arxiv.org/pdf/2601.00150v3
+
+## 24732. Combining datasets with different ground truths using Low-Rank Adaptation to generalize image-based CNN models for photometric redshift prediction
+
+- **ID:** `2601.00146` · **primary:** `astro-ph.IM` · **Submitted:** 2026-01-01
+- **Authors:** Vikram Seenivasan, Srinath Saikrishnan, Andrew Lizarraga, et al.
+- **abs:** https://arxiv.org/abs/2601.00146 · **PDF:** https://arxiv.org/pdf/2601.00146v1
+
+## 24733. MethConvTransformer: A Deep Learning Framework for Cross-Tissue Alzheimer's Disease Detection
+
+- **ID:** `2601.00143` · **primary:** `q-bio.GN` · **Submitted:** 2026-01-01
+- **Authors:** Gang Qu, Guanghao Li, Zhongming Zhao
+- **abs:** https://arxiv.org/abs/2601.00143 · **PDF:** https://arxiv.org/pdf/2601.00143v1
+
+## 24734. An AI Monkey Gets Grapes for Sure -- Sphere Neural Networks for Reliable Decision-Making
+
+- **ID:** `2601.00142` · **primary:** `cs.AI` · **Submitted:** 2026-01-01
+- **Authors:** Tiansi Dong, Henry He, Pietro Liò, et al.
+- **abs:** https://arxiv.org/abs/2601.00142 · **PDF:** https://arxiv.org/pdf/2601.00142v1
+
+## 24735. Placenta Accreta Spectrum Detection using Multimodal Deep Learning
+
+- **ID:** `2601.00907` · **primary:** `eess.IV` · **Submitted:** 2025-12-31
+- **Authors:** Sumaiya Ali, Areej Alhothali, Sameera Albasri, et al.
+- **abs:** https://arxiv.org/abs/2601.00907 · **PDF:** https://arxiv.org/pdf/2601.00907v1
+
+## 24736. Explicit Abstention Knobs for Predictable Reliability in Video Question Answering
+
+- **ID:** `2601.00138` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Jorge Ortiz
+- **abs:** https://arxiv.org/abs/2601.00138 · **PDF:** https://arxiv.org/pdf/2601.00138v2
+
+## 24737. PCEval: A Benchmark for Evaluating Physical Computing Capabilities of Large Language Models
+
+- **ID:** `2601.02404` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Inpyo Song, Eunji Jeon, Jangwon Lee
+- **abs:** https://arxiv.org/abs/2601.02404 · **PDF:** https://arxiv.org/pdf/2601.02404v1
+
+## 24738. Democratizing Electronic-Photonic AI Systems: An Open-Source AI-Infused Cross-Layer Co-Design and Design Automation Toolflow
+
+- **ID:** `2601.00130` · **primary:** `physics.optics` · **Submitted:** 2025-12-31
+- **Authors:** Hongjian Zhou, Ziang Yin, Jiaqi Gu
+- **abs:** https://arxiv.org/abs/2601.00130 · **PDF:** https://arxiv.org/pdf/2601.00130v1
+
+## 24739. Toward Large-Scale Photonics-Empowered AI Systems: From Physical Design Automation to System-Algorithm Co-Exploration
+
+- **ID:** `2601.00129` · **primary:** `physics.optics` · **Submitted:** 2025-12-31
+- **Authors:** Ziang Yin, Hongjian Zhou, Nicholas Gangi, et al.
+- **abs:** https://arxiv.org/abs/2601.00129 · **PDF:** https://arxiv.org/pdf/2601.00129v1
+
+## 24740. Constructing a Neuro-Symbolic Mathematician from First Principles
+
+- **ID:** `2601.00125` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Keqin Xie
+- **abs:** https://arxiv.org/abs/2601.00125 · **PDF:** https://arxiv.org/pdf/2601.00125v1
+
+## 24741. Ask, Clarify, Optimize: Human-LLM Agent Collaboration for Smarter Inventory Control
+
+- **ID:** `2601.00121` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Yaqi Duan, Yichun Hu, Jiashuo Jiang
+- **abs:** https://arxiv.org/abs/2601.00121 · **PDF:** https://arxiv.org/pdf/2601.00121v1
+
+## 24742. Evaluating Contextual Intelligence in Recyclability: A Comprehensive Study of Image-Based Reasoning Systems
+
+- **ID:** `2601.00905` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Eliot Park, Abhi Kumar, Pranav Rajpurkar
+- **abs:** https://arxiv.org/abs/2601.00905 · **PDF:** https://arxiv.org/pdf/2601.00905v1
+
+## 24743. GRL-SNAM: Geometric Reinforcement Learning with Path Differential Hamiltonians for Simultaneous Navigation and Mapping in Unknown Environments
+
+- **ID:** `2601.00116` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Aditya Sai Ellendula, Yi Wang, Minh Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2601.00116 · **PDF:** https://arxiv.org/pdf/2601.00116v1
+
+## 24744. Mortar: Evolving Mechanics for Automatic Game Design
+
+- **ID:** `2601.00105` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Muhammad U. Nasir, Yuchen Li, Steven James, et al.
+- **abs:** https://arxiv.org/abs/2601.00105 · **PDF:** https://arxiv.org/pdf/2601.00105v1
+
+## 24745. Covariance Matrix Adaptation Evolution Strategy without a matrix
+
+- **ID:** `2601.00102` · **primary:** `cs.NE` · **Submitted:** 2025-12-31
+- **Authors:** Jarosław Arabas, Adam Stelmaszczyk, Eryk Warchulski, et al.
+- **abs:** https://arxiv.org/abs/2601.00102 · **PDF:** https://arxiv.org/pdf/2601.00102v1
+
+## 24746. Speculative Decoding: Performance or Illusion?
+
+- **ID:** `2601.11580` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Xiaoxuan Liu, Jiaxiang Yu, Jongseok Park, et al.
+- **abs:** https://arxiv.org/abs/2601.11580 · **PDF:** https://arxiv.org/pdf/2601.11580v2
+
+## 24747. GenAITEd Ghana: A First-of-Its-Kind Context-Aware and Curriculum-Aligned Conversational AI Agent for Teacher Education
+
+- **ID:** `2601.06093` · **primary:** `cs.CY` · **Submitted:** 2025-12-31
+- **Authors:** Matthew Nyaaba, Patrick Kyeremeh, Macharious Nabang, et al.
+- **abs:** https://arxiv.org/abs/2601.06093 · **PDF:** https://arxiv.org/pdf/2601.06093v2
+
+## 24748. Learning Speech Representations with Variational Predictive Coding
+
+- **ID:** `2601.00100` · **primary:** `eess.AS` · **Submitted:** 2025-12-31
+- **Authors:** Sung-Lin Yeh, Peter Bell, Hao Tang
+- **abs:** https://arxiv.org/abs/2601.00100 · **PDF:** https://arxiv.org/pdf/2601.00100v1
+
+## 24749. The Agentic Leash: Extracting Causal Feedback Fuzzy Cognitive Maps with LLMs
+
+- **ID:** `2601.00097` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Akash Kumar Panda, Olaoluwa Adigun, Bart Kosko
+- **abs:** https://arxiv.org/abs/2601.00097 · **PDF:** https://arxiv.org/pdf/2601.00097v3
+
+## 24750. Adaptive Constraint Propagation: Scaling Structured Inference for Large Language Models via Meta-Reinforcement Learning
+
+- **ID:** `2601.00095` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma
+- **abs:** https://arxiv.org/abs/2601.00095 · **PDF:** https://arxiv.org/pdf/2601.00095v3
+
+## 24751. Islamic Chatbots in the Age of Large Language Models
+
+- **ID:** `2601.06092` · **primary:** `cs.CY` · **Submitted:** 2025-12-31
+- **Authors:** Muhammad Aurangzeb Ahmad
+- **abs:** https://arxiv.org/abs/2601.06092 · **PDF:** https://arxiv.org/pdf/2601.06092v1
+
+## 24752. It's Never Too Late: Noise Optimization for Collapse Recovery in Trained Diffusion Models
+
+- **ID:** `2601.00090` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Anne Harrington, A. Sophia Koepke, Shyamgopal Karthik, et al.
+- **abs:** https://arxiv.org/abs/2601.00090 · **PDF:** https://arxiv.org/pdf/2601.00090v1
+
+## 24753. Dynamic Bayesian Optimization Framework for Instruction Tuning in Partial Differential Equation Discovery
+
+- **ID:** `2601.00088` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Junqi Qu, Yan Zhang, Shangqian Gao, et al.
+- **abs:** https://arxiv.org/abs/2601.00088 · **PDF:** https://arxiv.org/pdf/2601.00088v1
+
+## 24754. Deep Deterministic Nonlinear ICA via Total Correlation Minimization with Matrix-Based Entropy Functional
+
+- **ID:** `2601.00904` · **primary:** `stat.ME` · **Submitted:** 2025-12-31
+- **Authors:** Qiang Li, Shujian Yu, Liang Ma, et al.
+- **abs:** https://arxiv.org/abs/2601.00904 · **PDF:** https://arxiv.org/pdf/2601.00904v1
+
+## 24755. Reinforcement learning with timed constraints for robotics motion planning
+
+- **ID:** `2601.00087` · **primary:** `cs.RO` · **Submitted:** 2025-12-31
+- **Authors:** Zhaoan Wang, Junchao Li, Mahdi Mohammad, et al.
+- **abs:** https://arxiv.org/abs/2601.00087 · **PDF:** https://arxiv.org/pdf/2601.00087v1
+
+## 24756. RIMRULE: Improving Tool-Using Language Agents via MDL-Guided Rule Learning
+
+- **ID:** `2601.00086` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Xiang Gao, Yuguang Yao, Qi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.00086 · **PDF:** https://arxiv.org/pdf/2601.00086v2
+
+## 24757. Exploration in the Limit
+
+- **ID:** `2601.00084` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Brian M. Cho, Nathan Kallus
+- **abs:** https://arxiv.org/abs/2601.00084 · **PDF:** https://arxiv.org/pdf/2601.00084v1
+
+## 24758. Cuffless, calibration-free hemodynamic monitoring with physics-informed machine learning models
+
+- **ID:** `2601.00081` · **primary:** `physics.med-ph` · **Submitted:** 2025-12-31
+- **Authors:** Henry Crandall, Tyler Schuessler, Filip Bělík, et al.
+- **abs:** https://arxiv.org/abs/2601.00081 · **PDF:** https://arxiv.org/pdf/2601.00081v1
+
+## 24759. IMBWatch -- a Spatio-Temporal Graph Neural Network approach to detect Illicit Massage Business
+
+- **ID:** `2601.00075` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Swetha Varadarajan, Abhishek Ray, Lumina Albert
+- **abs:** https://arxiv.org/abs/2601.00075 · **PDF:** https://arxiv.org/pdf/2601.00075v1
+
+## 24760. Automated electrostatic characterization of quantum dot devices in single- and bilayer heterostructures
+
+- **ID:** `2601.00067` · **primary:** `cond-mat.mes-hall` · **Submitted:** 2025-12-31
+- **Authors:** Merritt P. R. Losert, Dario Denora, Barnaby van Straaten, et al.
+- **abs:** https://arxiv.org/abs/2601.00067 · **PDF:** https://arxiv.org/pdf/2601.00067v1
+
+## 24761. The Trojan in the Vocabulary: Stealthy Sabotage of LLM Composition
+
+- **ID:** `2601.00065` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Xiaoze Liu, Weichen Yu, Matt Fredrikson, et al.
+- **abs:** https://arxiv.org/abs/2601.00065 · **PDF:** https://arxiv.org/pdf/2601.00065v2
+
+## 24762. SpaceTimePilot: Generative Rendering of Dynamic Scenes Across Space and Time
+
+- **ID:** `2512.25075` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Zhening Huang, Hyeonho Jeong, Xuelin Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.25075 · **PDF:** https://arxiv.org/pdf/2512.25075v1
+
+## 24763. Coordinated Humanoid Manipulation with Choice Policies
+
+- **ID:** `2512.25072` · **primary:** `cs.RO` · **Submitted:** 2025-12-31
+- **Authors:** Haozhi Qi, Yen-Jen Wang, Toru Lin, et al.
+- **abs:** https://arxiv.org/abs/2512.25072 · **PDF:** https://arxiv.org/pdf/2512.25072v1
+
+## 24764. Scaling Open-Ended Reasoning to Predict the Future
+
+- **ID:** `2512.25070` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Nikhil Chandak, Shashwat Goel, Ameya Prabhu, et al.
+- **abs:** https://arxiv.org/abs/2512.25070 · **PDF:** https://arxiv.org/pdf/2512.25070v2
+
+## 24765. Vulcan: Instance-Optimal Systems Heuristics Through LLM-Driven Search
+
+- **ID:** `2512.25065` · **primary:** `cs.OS` · **Submitted:** 2025-12-31
+- **Authors:** Rohit Dwivedula, Divyanshu Saxena, Sujay Yadalam, et al.
+- **abs:** https://arxiv.org/abs/2512.25065 · **PDF:** https://arxiv.org/pdf/2512.25065v1
+
+## 24766. Many Minds from One Model: Bayesian-Inspired Transformers for Population Diversity
+
+- **ID:** `2512.25063` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Diji Yang, Yi Zhang
+- **abs:** https://arxiv.org/abs/2512.25063 · **PDF:** https://arxiv.org/pdf/2512.25063v2
+
+## 24767. On the geometry and topology of representations: the manifolds of modular addition
+
+- **ID:** `2512.25060` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Gabriela Moisescu-Pareja, Gavin McCracken, Harley Wiltzer, et al.
+- **abs:** https://arxiv.org/abs/2512.25060 · **PDF:** https://arxiv.org/pdf/2512.25060v1
+
+## 24768. Reliable and Resilient Collective Communication Library for LLM Training and Serving
+
+- **ID:** `2512.25059` · **primary:** `cs.DC` · **Submitted:** 2025-12-31
+- **Authors:** Wei Wang, Nengneng Yu, Sixian Xiong, et al.
+- **abs:** https://arxiv.org/abs/2512.25059 · **PDF:** https://arxiv.org/pdf/2512.25059v1
+
+## 24769. Context-aware LLM-based AI Agents for Human-centered Energy Management Systems in Smart Buildings
+
+- **ID:** `2512.25055` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Tianzhi He, Farrokh Jazizadeh
+- **abs:** https://arxiv.org/abs/2512.25055 · **PDF:** https://arxiv.org/pdf/2512.25055v1
+
+## 24770. AdaGReS:Adaptive Greedy Context Selection via Redundancy-Aware Scoring for Token-Budgeted RAG
+
+- **ID:** `2512.25052` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Chao Peng, Bin Wang, Zhilei Long, et al.
+- **abs:** https://arxiv.org/abs/2512.25052 · **PDF:** https://arxiv.org/pdf/2512.25052v1
+
+## 24771. Generative Classifiers Avoid Shortcut Solutions
+
+- **ID:** `2512.25034` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Alexander C. Li, Ananya Kumar, Deepak Pathak
+- **abs:** https://arxiv.org/abs/2512.25034 · **PDF:** https://arxiv.org/pdf/2512.25034v1
+
+## 24772. Modeling Language as a Sequence of Thoughts
+
+- **ID:** `2512.25026` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Nasim Borazjanizadeh, James McClelland
+- **abs:** https://arxiv.org/abs/2512.25026 · **PDF:** https://arxiv.org/pdf/2512.25026v2
+
+## 24773. ResponseRank: Data-Efficient Reward Modeling through Preference Strength Learning
+
+- **ID:** `2512.25023` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Timo Kaufmann, Yannick Metz, Daniel Keim, et al.
+- **abs:** https://arxiv.org/abs/2512.25023 · **PDF:** https://arxiv.org/pdf/2512.25023v1
+
+## 24774. Convergence of the generalization error for deep gradient flow methods for PDEs
+
+- **ID:** `2512.25017` · **primary:** `math.NA` · **Submitted:** 2025-12-31
+- **Authors:** Chenguang Liu, Antonis Papapantoleon, Jasper Rou
+- **abs:** https://arxiv.org/abs/2512.25017 · **PDF:** https://arxiv.org/pdf/2512.25017v2
+
+## 24775. MAMA-Memeia! Multi-Aspect Multi-Agent Collaboration for Depressive Symptoms Identification in Memes
+
+- **ID:** `2512.25015` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Siddhant Agarwal, Adya Dhuler, Polly Ruhnke, et al.
+- **abs:** https://arxiv.org/abs/2512.25015 · **PDF:** https://arxiv.org/pdf/2512.25015v1
+
+## 24776. Diffusion Language Models are Provably Optimal Parallel Samplers
+
+- **ID:** `2512.25014` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Haozhe Jiang, Nika Haghtalab, Lijie Chen
+- **abs:** https://arxiv.org/abs/2512.25014 · **PDF:** https://arxiv.org/pdf/2512.25014v1
+
+## 24777. Basic Inequalities for First-Order Optimization with Applications to Statistical Risk Analysis
+
+- **ID:** `2512.24999` · **primary:** `math.ST` · **Submitted:** 2025-12-31
+- **Authors:** Seunghoon Paik, Kangjie Zhou, Matus Telgarsky, et al.
+- **abs:** https://arxiv.org/abs/2512.24999 · **PDF:** https://arxiv.org/pdf/2512.24999v1
+
+## 24778. Classifying long legal documents using short random chunks
+
+- **ID:** `2512.24997` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Luis Adrián Cabrera-Diego
+- **abs:** https://arxiv.org/abs/2512.24997 · **PDF:** https://arxiv.org/pdf/2512.24997v1
+
+## 24779. Efficiently Estimating Data Efficiency for Language Model Fine-tuning
+
+- **ID:** `2512.24991` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Gyung Hyun Je, Colin Raffel
+- **abs:** https://arxiv.org/abs/2512.24991 · **PDF:** https://arxiv.org/pdf/2512.24991v1
+
+## 24780. DarkEQA: Benchmarking Vision-Language Models for Embodied Question Answering in Low-Light Indoor Environments
+
+- **ID:** `2512.24985` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Yohan Park, Hyunwoo Ha, Wonjun Jo, et al.
+- **abs:** https://arxiv.org/abs/2512.24985 · **PDF:** https://arxiv.org/pdf/2512.24985v3
+
+## 24781. A Modal Logic for Possibilistic Reasoning with Fuzzy Formal Contexts
+
+- **ID:** `2512.24980` · **primary:** `cs.LO` · **Submitted:** 2025-12-31
+- **Authors:** Prosenjit Howlader, Churn-Jung Liau
+- **abs:** https://arxiv.org/abs/2512.24980 · **PDF:** https://arxiv.org/pdf/2512.24980v1
+
+## 24782. Noise-Aware and Dynamically Adaptive Federated Defense Framework for SAR Image Target Recognition
+
+- **ID:** `2601.00900` · **primary:** `cs.CR` · **Submitted:** 2025-12-31
+- **Authors:** Yuchao Hou, Zixuan Zhang, Jie Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.00900 · **PDF:** https://arxiv.org/pdf/2601.00900v1
+
+## 24783. SymSeqBench: a unified framework for the generation and analysis of rule-based symbolic sequences and datasets
+
+- **ID:** `2512.24977` · **primary:** `q-bio.NC` · **Submitted:** 2025-12-31
+- **Authors:** Barna Zajzon, Younes Bouhadjar, Maxime Fabre, et al.
+- **abs:** https://arxiv.org/abs/2512.24977 · **PDF:** https://arxiv.org/pdf/2512.24977v1
+
+## 24784. Attribution-Guided Distillation of Matryoshka Sparse Autoencoders
+
+- **ID:** `2512.24975` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Cristina P. Martin-Linares, Jonathan P. Ling
+- **abs:** https://arxiv.org/abs/2512.24975 · **PDF:** https://arxiv.org/pdf/2512.24975v1
+
+## 24785. Evaluating the Impact of Compression Techniques on the Robustness of CNNs under Natural Corruptions
+
+- **ID:** `2512.24971` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Itallo Patrick Castro Alves Da Silva, Emanuel Adler Medeiros Pereira, Erick de Andrade Barboza, et al.
+- **abs:** https://arxiv.org/abs/2512.24971 · **PDF:** https://arxiv.org/pdf/2512.24971v1
+
+## 24786. Dichotomous Diffusion Policy Optimization
+
+- **ID:** `2601.00898` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Ruiming Liang, Yinan Zheng, Kexin Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.00898 · **PDF:** https://arxiv.org/pdf/2601.00898v2
+
+## 24787. Large language models and the entropy of English
+
+- **ID:** `2512.24969` · **primary:** `cond-mat.stat-mech` · **Submitted:** 2025-12-31
+- **Authors:** Colin Scheibner, Lindsay M. Smith, William Bialek
+- **abs:** https://arxiv.org/abs/2512.24969 · **PDF:** https://arxiv.org/pdf/2512.24969v1
+
+## 24788. The Impact of LLMs on Online News Consumption and Production
+
+- **ID:** `2512.24968` · **primary:** `econ.GN` · **Submitted:** 2025-12-31
+- **Authors:** Hangcheng Zhao, Ron Berman
+- **abs:** https://arxiv.org/abs/2512.24968 · **PDF:** https://arxiv.org/pdf/2512.24968v3
+
+## 24789. ShowUI-$π$: Flow-based Generative Models as GUI Dexterous Hands
+
+- **ID:** `2512.24965` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Siyuan Hu, Kevin Qinghong Lin, Mike Zheng Shou
+- **abs:** https://arxiv.org/abs/2512.24965 · **PDF:** https://arxiv.org/pdf/2512.24965v1
+
+## 24790. Semi-overlapping Multi-bandit Best Arm Identification for Sequential Support Network Learning
+
+- **ID:** `2512.24959` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** András Antos, András Millinghoffer, Péter Antal
+- **abs:** https://arxiv.org/abs/2512.24959 · **PDF:** https://arxiv.org/pdf/2512.24959v1
+
+## 24791. AMAP Agentic Planning Technical Report
+
+- **ID:** `2512.24957` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** AMAP AI Agent Team, Yulan Hu, Xiangwen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.24957 · **PDF:** https://arxiv.org/pdf/2512.24957v2
+
+## 24792. MSACL: Multi-Step Actor-Critic Learning with Lyapunov Certificates for Exponentially Stabilizing Control
+
+- **ID:** `2512.24955` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Yongwei Zhang, Yuanzhe Xing, Quanyi Liang, et al.
+- **abs:** https://arxiv.org/abs/2512.24955 · **PDF:** https://arxiv.org/pdf/2512.24955v3
+
+## 24793. ProDM: Synthetic Reality-driven Property-aware Progressive Diffusion Model for Coronary Calcium Motion Correction in Non-gated Chest CT
+
+- **ID:** `2512.24948` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Xinran Gong, Gorkem Durak, Halil Ertugrul Aktas, et al.
+- **abs:** https://arxiv.org/abs/2512.24948 · **PDF:** https://arxiv.org/pdf/2512.24948v1
+
+## 24794. CPJ: Explainable Agricultural Pest Diagnosis via Caption-Prompt-Judge with LLM-Judged Refinement
+
+- **ID:** `2512.24947` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Wentao Zhang, Tao Fang, Lina Lu, et al.
+- **abs:** https://arxiv.org/abs/2512.24947 · **PDF:** https://arxiv.org/pdf/2512.24947v1
+
+## 24795. HaineiFRDM: Explore Diffusion to Restore Defects in Fast-Movement Films
+
+- **ID:** `2512.24946` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Rongji Xun, Junjie Yuan, Zhongjie Wang
+- **abs:** https://arxiv.org/abs/2512.24946 · **PDF:** https://arxiv.org/pdf/2512.24946v1
+
+## 24796. CornViT: A Multi-Stage Convolutional Vision Transformer Framework for Hierarchical Corn Kernel Analysis
+
+- **ID:** `2601.00897` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Sai Teja Erukude, Jane Mascarenhas, Lior Shamir
+- **abs:** https://arxiv.org/abs/2601.00897 · **PDF:** https://arxiv.org/pdf/2601.00897v1
+
+## 24797. RAIR: A Rule-Aware Benchmark Uniting Challenging Long-Tail and Visual Salience Subset for E-commerce Relevance Assessment
+
+- **ID:** `2512.24943` · **primary:** `cs.IR` · **Submitted:** 2025-12-31
+- **Authors:** Chenji Lu, Zhuo Chen, Hui Zhao, et al.
+- **abs:** https://arxiv.org/abs/2512.24943 · **PDF:** https://arxiv.org/pdf/2512.24943v2
+
+## 24798. Iterative Deployment Improves Planning Skills in LLMs
+
+- **ID:** `2512.24940` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Augusto B. Corrêa, Yoav Gelberg, Luckeciano C. Melo, et al.
+- **abs:** https://arxiv.org/abs/2512.24940 · **PDF:** https://arxiv.org/pdf/2512.24940v1
+
+## 24799. Vibe Coding, Interface Flattening
+
+- **ID:** `2512.24939` · **primary:** `cs.HC` · **Submitted:** 2025-12-31
+- **Authors:** Hongrui Jin
+- **abs:** https://arxiv.org/abs/2512.24939 · **PDF:** https://arxiv.org/pdf/2512.24939v1
+
+## 24800. Investigation into U.S. Citizen and Non-Citizen Worker Health Insurance and Employment
+
+- **ID:** `2601.00896` · **primary:** `econ.GN` · **Submitted:** 2025-12-31
+- **Authors:** Annabelle Yao
+- **abs:** https://arxiv.org/abs/2601.00896 · **PDF:** https://arxiv.org/pdf/2601.00896v1
+
+## 24801. Adaptive Dependency-aware Prompt Optimization Framework for Multi-Step LLM Pipeline
+
+- **ID:** `2512.24933` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Minjun Zhao, Xinyu Zhang, Shuai Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.24933 · **PDF:** https://arxiv.org/pdf/2512.24933v1
+
+## 24802. Deep Learning Framework for RNA Inverse Folding with Geometric Structure Potentials
+
+- **ID:** `2601.00895` · **primary:** `q-bio.QM` · **Submitted:** 2025-12-31
+- **Authors:** Annabelle Yao
+- **abs:** https://arxiv.org/abs/2601.00895 · **PDF:** https://arxiv.org/pdf/2601.00895v1
+
+## 24803. Are First-Order Diffusion Samplers Really Slower? A Fast Forward-Value Approach
+
+- **ID:** `2512.24927` · **primary:** `stat.ML` · **Submitted:** 2025-12-31
+- **Authors:** Yuchen Jiao, Na Li, Changxiao Cai, et al.
+- **abs:** https://arxiv.org/abs/2512.24927 · **PDF:** https://arxiv.org/pdf/2512.24927v1
+
+## 24804. Frequent subgraph-based persistent homology for graph classification
+
+- **ID:** `2512.24917` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Xinyang Chen, Amaël Broustet, Guanyuan Zeng, et al.
+- **abs:** https://arxiv.org/abs/2512.24917 · **PDF:** https://arxiv.org/pdf/2512.24917v2
+
+## 24805. AI-Driven Cloud Resource Optimization for Multi-Cluster Environments
+
+- **ID:** `2512.24914` · **primary:** `cs.DC` · **Submitted:** 2025-12-31
+- **Authors:** Vinoth Punniyamoorthy, Akash Kumar Agarwal, Bikesh Kumar, et al.
+- **abs:** https://arxiv.org/abs/2512.24914 · **PDF:** https://arxiv.org/pdf/2512.24914v1
+
+## 24806. Spectral Graph Neural Networks for Cognitive Task Classification in fMRI Connectomes
+
+- **ID:** `2512.24901` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Debasis Maji, Arghya Banerjee, Debaditya Barman
+- **abs:** https://arxiv.org/abs/2512.24901 · **PDF:** https://arxiv.org/pdf/2512.24901v1
+
+## 24807. PRISM: A hierarchical multiscale approach for time series forecasting
+
+- **ID:** `2512.24898` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Zihao Chen, Alexandre Andre, Wenrui Ma, et al.
+- **abs:** https://arxiv.org/abs/2512.24898 · **PDF:** https://arxiv.org/pdf/2512.24898v1
+
+## 24808. When to Ponder: Adaptive Compute Allocation for Code Generation via Test-Time Training
+
+- **ID:** `2601.00894` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Gihyeon Sim
+- **abs:** https://arxiv.org/abs/2601.00894 · **PDF:** https://arxiv.org/pdf/2601.00894v1
+
+## 24809. Semi-Automated Data Annotation in Multisensor Datasets for Autonomous Vehicle Testing
+
+- **ID:** `2512.24896` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Andrii Gamalii, Daniel Górniak, Robert Nowak, et al.
+- **abs:** https://arxiv.org/abs/2512.24896 · **PDF:** https://arxiv.org/pdf/2512.24896v1
+
+## 24810. Towards eco friendly cybersecurity: machine learning based anomaly detection with carbon and energy metrics
+
+- **ID:** `2601.00893` · **primary:** `cs.CR` · **Submitted:** 2025-12-31
+- **Authors:** KC Aashish, Md Zakir Hossain Zamil, Md Shafiqul Islam Mridul, et al.
+- **abs:** https://arxiv.org/abs/2601.00893 · **PDF:** https://arxiv.org/pdf/2601.00893v1
+
+## 24811. BEDA: Belief Estimation as Probabilistic Constraints for Performing Strategic Dialogue Acts
+
+- **ID:** `2512.24885` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Hengli Li, Zhaoxin Yu, Qi Shen, et al.
+- **abs:** https://arxiv.org/abs/2512.24885 · **PDF:** https://arxiv.org/pdf/2512.24885v1
+
+## 24812. Hierarchical topological clustering
+
+- **ID:** `2601.00892` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Ana Carpio, Gema Duro
+- **abs:** https://arxiv.org/abs/2601.00892 · **PDF:** https://arxiv.org/pdf/2601.00892v1
+
+## 24813. mHC: Manifold-Constrained Hyper-Connections
+
+- **ID:** `2512.24880` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Zhenda Xie, Yixuan Wei, Huanqi Cao, et al.
+- **abs:** https://arxiv.org/abs/2512.24880 · **PDF:** https://arxiv.org/pdf/2512.24880v2
+
+## 24814. Let It Flow: Agentic Crafting on Rock and Roll, Building the ROME Model within an Open Agentic Learning Ecosystem
+
+- **ID:** `2512.24873` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Weixun Wang, XiaoXiao Xu, Wanhe An, et al.
+- **abs:** https://arxiv.org/abs/2512.24873 · **PDF:** https://arxiv.org/pdf/2512.24873v3
+
+## 24815. Encyclo-K: Evaluating LLMs with Dynamically Composed Knowledge Statements
+
+- **ID:** `2512.24867` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Yiming Liang, Yizhi Li, Yantao Du, et al.
+- **abs:** https://arxiv.org/abs/2512.24867 · **PDF:** https://arxiv.org/pdf/2512.24867v2
+
+## 24816. Characterization of Transfer Using Multi-task Learning Curves
+
+- **ID:** `2512.24866` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** András Millinghoffer, Bence Bolgár, Péter Antal
+- **abs:** https://arxiv.org/abs/2512.24866 · **PDF:** https://arxiv.org/pdf/2512.24866v1
+
+## 24817. Big AI is accelerating the metacrisis: What can we do?
+
+- **ID:** `2512.24863` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Steven Bird
+- **abs:** https://arxiv.org/abs/2512.24863 · **PDF:** https://arxiv.org/pdf/2512.24863v1
+
+## 24818. Enhancing Retrieval-Augmented Generation with Topic-Enriched Embeddings: A Hybrid Approach Integrating Traditional NLP Techniques
+
+- **ID:** `2601.00891` · **primary:** `cs.IR` · **Submitted:** 2025-12-31
+- **Authors:** Rodrigo Kataishi
+- **abs:** https://arxiv.org/abs/2601.00891 · **PDF:** https://arxiv.org/pdf/2601.00891v1
+
+## 24819. A study on constraint extraction and exception exclusion in care worker scheduling
+
+- **ID:** `2512.24853` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Koki Suenaga, Tomohiro Furuta, Satoshi Ono
+- **abs:** https://arxiv.org/abs/2512.24853 · **PDF:** https://arxiv.org/pdf/2512.24853v1
+
+## 24820. PrivacyBench: A Conversational Benchmark for Evaluating Privacy in Personalized AI
+
+- **ID:** `2512.24848` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Srija Mukhopadhyay, Sathwik Reddy, Shruthi Muthukumar, et al.
+- **abs:** https://arxiv.org/abs/2512.24848 · **PDF:** https://arxiv.org/pdf/2512.24848v1
+
+## 24821. AODDiff: Probabilistic Reconstruction of Aerosol Optical Depth via Diffusion-based Bayesian Inference
+
+- **ID:** `2512.24847` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Linhao Fan, Hongqiang Fang, Jingyang Dai, et al.
+- **abs:** https://arxiv.org/abs/2512.24847 · **PDF:** https://arxiv.org/pdf/2512.24847v1
+
+## 24822. Triangulation as an Acceptance Rule for Multilingual Mechanistic Interpretability
+
+- **ID:** `2512.24842` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Yanan Long
+- **abs:** https://arxiv.org/abs/2512.24842 · **PDF:** https://arxiv.org/pdf/2512.24842v1
+
+## 24823. GenZ: Foundational models as latent variable generators within traditional statistical models
+
+- **ID:** `2512.24834` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Marko Jojic, Nebojsa Jojic
+- **abs:** https://arxiv.org/abs/2512.24834 · **PDF:** https://arxiv.org/pdf/2512.24834v1
+
+## 24824. Explaining Why Things Go Where They Go: Interpretable Constructs of Human Organizational Preferences
+
+- **ID:** `2512.24829` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Emmanuel Fashae, Michael Burke, Leimin Tian, et al.
+- **abs:** https://arxiv.org/abs/2512.24829 · **PDF:** https://arxiv.org/pdf/2512.24829v1
+
+## 24825. Discovering Coordinated Joint Options via Inter-Agent Relative Dynamics
+
+- **ID:** `2512.24827` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Raul D. Steleac, Mohan Sridharan, David Abel
+- **abs:** https://arxiv.org/abs/2512.24827 · **PDF:** https://arxiv.org/pdf/2512.24827v2
+
+## 24826. Video and Language Alignment in 2D Systems for 3D Multi-object Scenes with Multi-Information Derivative-Free Control
+
+- **ID:** `2512.24826` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Jason Armitage, Rico Sennnrich
+- **abs:** https://arxiv.org/abs/2512.24826 · **PDF:** https://arxiv.org/pdf/2512.24826v1
+
+## 24827. Practising responsibility: Ethics in NLP as a hands-on course
+
+- **ID:** `2512.24825` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Malvina Nissim, Viviana Patti, Beatrice Savoldi
+- **abs:** https://arxiv.org/abs/2512.24825 · **PDF:** https://arxiv.org/pdf/2512.24825v1
+
+## 24828. Unregularized Linear Convergence in Zero-Sum Game from Preference Feedback
+
+- **ID:** `2512.24818` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Shulun Chen, Runlong Zhou, Zihan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.24818 · **PDF:** https://arxiv.org/pdf/2512.24818v2
+
+## 24829. Learning Temporally Consistent Turbulence Between Sparse Snapshots via Diffusion Models
+
+- **ID:** `2512.24813` · **primary:** `physics.flu-dyn` · **Submitted:** 2025-12-31
+- **Authors:** Mohammed Sardar, Małgorzata J. Zimoń, Samuel Draycott, et al.
+- **abs:** https://arxiv.org/abs/2512.24813 · **PDF:** https://arxiv.org/pdf/2512.24813v1
+
+## 24830. DTI-GP: Bayesian operations for drug-target interactions using deep kernel Gaussian processes
+
+- **ID:** `2512.24810` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Bence Bolgár, András Millinghoffer, Péter Antal
+- **abs:** https://arxiv.org/abs/2512.24810 · **PDF:** https://arxiv.org/pdf/2512.24810v1
+
+## 24831. FANoS: Friction-Adaptive Nosé--Hoover Symplectic Momentum for Stiff Objectives
+
+- **ID:** `2601.00889` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Nalin Dhiman
+- **abs:** https://arxiv.org/abs/2601.00889 · **PDF:** https://arxiv.org/pdf/2601.00889v1
+
+## 24832. Limits of quantum generative models with classical sampling hardness
+
+- **ID:** `2512.24801` · **primary:** `quant-ph` · **Submitted:** 2025-12-31
+- **Authors:** Sabrina Herbst, Ivona Brandić, Adrián Pérez-Salinas
+- **abs:** https://arxiv.org/abs/2512.24801 · **PDF:** https://arxiv.org/pdf/2512.24801v1
+
+## 24833. LeanCat: A Benchmark Suite for Formal Category Theory in Lean (Part I: 1-Categories)
+
+- **ID:** `2512.24796` · **primary:** `cs.LO` · **Submitted:** 2025-12-31
+- **Authors:** Rongge Xu, Hui Dai, Yiming Fu, et al.
+- **abs:** https://arxiv.org/abs/2512.24796 · **PDF:** https://arxiv.org/pdf/2512.24796v2
+
+## 24834. Nonlinear Noise2Noise for Efficient Monte Carlo Denoiser Training
+
+- **ID:** `2512.24794` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Andrew Tinits, Stephen Mann
+- **abs:** https://arxiv.org/abs/2512.24794 · **PDF:** https://arxiv.org/pdf/2512.24794v1
+
+## 24835. Self-Supervised Neural Architecture Search for Multimodal Deep Neural Networks
+
+- **ID:** `2512.24793` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Shota Suzuki, Satoshi Ono
+- **abs:** https://arxiv.org/abs/2512.24793 · **PDF:** https://arxiv.org/pdf/2512.24793v1
+
+## 24836. Projection-based Adversarial Attack using Physics-in-the-Loop Optimization for Monocular Depth Estimation
+
+- **ID:** `2512.24792` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Takeru Kusakabe, Yudai Hirose, Mashiho Mukaida, et al.
+- **abs:** https://arxiv.org/abs/2512.24792 · **PDF:** https://arxiv.org/pdf/2512.24792v1
+
+## 24837. HiGR: Efficient Generative Slate Recommendation via Hierarchical Planning and Multi-Objective Preference Alignment
+
+- **ID:** `2512.24787` · **primary:** `cs.IR` · **Submitted:** 2025-12-31
+- **Authors:** Yunsheng Pang, Zijian Liu, Yudong Li, et al.
+- **abs:** https://arxiv.org/abs/2512.24787 · **PDF:** https://arxiv.org/pdf/2512.24787v2
+
+## 24838. Group Cross-Correlations with Faintly Constrained Filters
+
+- **ID:** `2601.00045` · **primary:** `math.DS` · **Submitted:** 2025-12-31
+- **Authors:** Benedikt Fluhr
+- **abs:** https://arxiv.org/abs/2601.00045 · **PDF:** https://arxiv.org/pdf/2601.00045v2
+
+## 24839. Gradient Descent as Implicit EM in Distance-Based Neural Models
+
+- **ID:** `2512.24780` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Alan Oursland
+- **abs:** https://arxiv.org/abs/2512.24780 · **PDF:** https://arxiv.org/pdf/2512.24780v1
+
+## 24840. Compute-Accuracy Pareto Frontiers for Open-Source Reasoning Large Language Models
+
+- **ID:** `2512.24776` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Ákos Prucs, Márton Csutora, Mátyás Antal, et al.
+- **abs:** https://arxiv.org/abs/2512.24776 · **PDF:** https://arxiv.org/pdf/2512.24776v1
+
+## 24841. Uncertainty-aware Semi-supervised Ensemble Teacher Framework for Multilingual Depression Detection
+
+- **ID:** `2512.24772` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Mohammad Zia Ur Rehman, Velpuru Navya, Sanskar, et al.
+- **abs:** https://arxiv.org/abs/2512.24772 · **PDF:** https://arxiv.org/pdf/2512.24772v1
+
+## 24842. Sparse Offline Reinforcement Learning with Corruption Robustness
+
+- **ID:** `2512.24768` · **primary:** `stat.ML` · **Submitted:** 2025-12-31
+- **Authors:** Nam Phuong Tran, Andi Nika, Goran Radanovic, et al.
+- **abs:** https://arxiv.org/abs/2512.24768 · **PDF:** https://arxiv.org/pdf/2512.24768v2
+
+## 24843. From Trial to Deployment: A SEM Analysis of Traveler Adoptions to Fully Operational Autonomous Taxis
+
+- **ID:** `2512.24767` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Yutong Cai, Hua Wang
+- **abs:** https://arxiv.org/abs/2512.24767 · **PDF:** https://arxiv.org/pdf/2512.24767v1
+
+## 24844. Dream2Flow: Bridging Video Generation and Open-World Manipulation with 3D Object Flow
+
+- **ID:** `2512.24766` · **primary:** `cs.RO` · **Submitted:** 2025-12-31
+- **Authors:** Karthik Dharmarajan, Wenlong Huang, Jiajun Wu, et al.
+- **abs:** https://arxiv.org/abs/2512.24766 · **PDF:** https://arxiv.org/pdf/2512.24766v1
+
+## 24845. AstroReview: An LLM-driven Multi-Agent Framework for Telescope Proposal Peer Review and Refinement
+
+- **ID:** `2512.24754` · **primary:** `astro-ph.IM` · **Submitted:** 2025-12-31
+- **Authors:** Yutong Wang, Yunxiang Xiao, Yonglin Tian, et al.
+- **abs:** https://arxiv.org/abs/2512.24754 · **PDF:** https://arxiv.org/pdf/2512.24754v1
+
+## 24846. Fairness-Aware Insurance Pricing: A Multi-Objective Optimization Approach
+
+- **ID:** `2512.24747` · **primary:** `q-fin.RM` · **Submitted:** 2025-12-31
+- **Authors:** Tim J. Boonen, Xinyue Fan, Zixiao Quan
+- **abs:** https://arxiv.org/abs/2512.24747 · **PDF:** https://arxiv.org/pdf/2512.24747v1
+
+## 24847. Counterfactual Self-Questioning for Stable Policy Optimization in Language Models
+
+- **ID:** `2601.00885` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Mandar Parab
+- **abs:** https://arxiv.org/abs/2601.00885 · **PDF:** https://arxiv.org/pdf/2601.00885v1
+
+## 24848. SEGB: Self-Evolved Generative Bidding with Local Autoregressive Diffusion
+
+- **ID:** `2602.22226` · **primary:** `cs.IR` · **Submitted:** 2025-12-31
+- **Authors:** Yulong Gao, Wan Jiang, Mingzhe Cao, et al.
+- **abs:** https://arxiv.org/abs/2602.22226 · **PDF:** https://arxiv.org/pdf/2602.22226v1
+
+## 24849. BIOME-Bench: A Benchmark for Biomolecular Interaction Inference and Multi-Omics Pathway Mechanism Elucidation from Scientific Literature
+
+- **ID:** `2512.24733` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Sibo Wei, Peng Chen, Lifeng Dong, et al.
+- **abs:** https://arxiv.org/abs/2512.24733 · **PDF:** https://arxiv.org/pdf/2512.24733v1
+
+## 24850. PriceSeer: Evaluating Large Language Models in Real-Time Stock Prediction
+
+- **ID:** `2601.06088` · **primary:** `q-fin.ST` · **Submitted:** 2025-12-31
+- **Authors:** Bohan Liang, Zijian Chen, Qi Jia, et al.
+- **abs:** https://arxiv.org/abs/2601.06088 · **PDF:** https://arxiv.org/pdf/2601.06088v1
+
+## 24851. Equivalence of Personalized PageRank and Successor Representations
+
+- **ID:** `2512.24722` · **primary:** `cs.NE` · **Submitted:** 2025-12-31
+- **Authors:** Beren Millidge
+- **abs:** https://arxiv.org/abs/2512.24722 · **PDF:** https://arxiv.org/pdf/2512.24722v1
+
+## 24852. FPGA Co-Design for Efficient N:M Sparse and Quantized Model Inference
+
+- **ID:** `2512.24713` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Fen-Yu Hsieh, Yun-Chang Teng, Ding-Yong Hong, et al.
+- **abs:** https://arxiv.org/abs/2512.24713 · **PDF:** https://arxiv.org/pdf/2512.24713v2
+
+## 24853. LSRE: Latent Semantic Rule Encoding for Real-Time Semantic Risk Detection in Autonomous Driving
+
+- **ID:** `2512.24712` · **primary:** `cs.RO` · **Submitted:** 2025-12-31
+- **Authors:** Qian Cheng, Weitao Zhou, Cheng Jing, et al.
+- **abs:** https://arxiv.org/abs/2512.24712 · **PDF:** https://arxiv.org/pdf/2512.24712v2
+
+## 24854. BandiK: Efficient Multi-Task Decomposition Using a Multi-Bandit Framework
+
+- **ID:** `2512.24708` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** András Millinghoffer, András Formanek, András Antos, et al.
+- **abs:** https://arxiv.org/abs/2512.24708 · **PDF:** https://arxiv.org/pdf/2512.24708v1
+
+## 24855. Outlier Detection Using Vector Cosine Similarity by Adding a Dimension
+
+- **ID:** `2601.00883` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Zhongyang Shen
+- **abs:** https://arxiv.org/abs/2601.00883 · **PDF:** https://arxiv.org/pdf/2601.00883v1
+
+## 24856. Spiking Heterogeneous Graph Attention Networks
+
+- **ID:** `2601.02401` · **primary:** `cs.NE` · **Submitted:** 2025-12-31
+- **Authors:** Buqing Cao, Qian Peng, Xiang Xie, et al.
+- **abs:** https://arxiv.org/abs/2601.02401 · **PDF:** https://arxiv.org/pdf/2601.02401v1
+
+## 24857. Evolving, Not Training: Zero-Shot Reasoning Segmentation via Evolutionary Prompting
+
+- **ID:** `2512.24702` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Kai Ye, Xiaotong You, Jianghang Lin, et al.
+- **abs:** https://arxiv.org/abs/2512.24702 · **PDF:** https://arxiv.org/pdf/2512.24702v1
+
+## 24858. SAC-NeRF: Adaptive Ray Sampling for Neural Radiance Fields via Soft Actor-Critic Reinforcement Learning
+
+- **ID:** `2603.15622` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Chenyu Ge
+- **abs:** https://arxiv.org/abs/2603.15622 · **PDF:** https://arxiv.org/pdf/2603.15622v1
+
+## 24859. Causal Discovery with Mixed Latent Confounding via Precision Decomposition
+
+- **ID:** `2512.24696` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Amir Asiaee, Samhita Pal, James O'quinn, et al.
+- **abs:** https://arxiv.org/abs/2512.24696 · **PDF:** https://arxiv.org/pdf/2512.24696v1
+
+## 24860. Nested Learning: The Illusion of Deep Learning Architectures
+
+- **ID:** `2512.24695` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Ali Behrouz, Meisam Razaviyayn, Peilin Zhong, et al.
+- **abs:** https://arxiv.org/abs/2512.24695 · **PDF:** https://arxiv.org/pdf/2512.24695v1
+
+## 24861. Mobility-Assisted Decentralized Federated Learning: Convergence Analysis and A Data-Driven Approach
+
+- **ID:** `2512.24694` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Reza Jahani, Md Farhamdur Reza, Richeng Jin, et al.
+- **abs:** https://arxiv.org/abs/2512.24694 · **PDF:** https://arxiv.org/pdf/2512.24694v2
+
+## 24862. MUSIC: MUlti-Step Instruction Contrast for Multi-Turn Reward Models
+
+- **ID:** `2512.24693` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Wenzhe Li, Shujian Zhang, Wenxuan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2512.24693 · **PDF:** https://arxiv.org/pdf/2512.24693v1
+
+## 24863. Quantum Visual Word Sense Disambiguation: Unraveling Ambiguities Through Quantum Inference Model
+
+- **ID:** `2512.24687` · **primary:** `quant-ph` · **Submitted:** 2025-12-31
+- **Authors:** Wenbo Qiao, Peng Zhang, Qinghua Hu
+- **abs:** https://arxiv.org/abs/2512.24687 · **PDF:** https://arxiv.org/pdf/2512.24687v1
+
+## 24864. BatteryAgent: Synergizing Physics-Informed Interpretation with LLM Reasoning for Intelligent Battery Fault Diagnosis
+
+- **ID:** `2512.24686` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Songqi Zhou, Ruixue Liu, Boman Su, et al.
+- **abs:** https://arxiv.org/abs/2512.24686 · **PDF:** https://arxiv.org/pdf/2512.24686v1
+
+## 24865. R-Debater: Retrieval-Augmented Debate Generation through Argumentative Memory
+
+- **ID:** `2512.24684` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Maoyuan Li, Zhongsheng Wang, Haoyuan Li, et al.
+- **abs:** https://arxiv.org/abs/2512.24684 · **PDF:** https://arxiv.org/pdf/2512.24684v1
+
+## 24866. Multi-modal cross-domain mixed fusion model with dual disentanglement for fault diagnosis under unseen working conditions
+
+- **ID:** `2512.24679` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Pengcheng Xia, Yixiang Huang, Chengjin Qin, et al.
+- **abs:** https://arxiv.org/abs/2512.24679 · **PDF:** https://arxiv.org/pdf/2512.24679v1
+
+## 24867. A New Decomposition Paradigm for Graph-structured Nonlinear Programs via Message Passing
+
+- **ID:** `2512.24676` · **primary:** `math.OC` · **Submitted:** 2025-12-31
+- **Authors:** Kuangyu Ding, Marie Maros, Gesualdo Scutari
+- **abs:** https://arxiv.org/abs/2512.24676 · **PDF:** https://arxiv.org/pdf/2512.24676v2
+
+## 24868. An Adaptive, Disentangled Representation for Multidimensional MRI Reconstruction
+
+- **ID:** `2512.24674` · **primary:** `eess.IV` · **Submitted:** 2025-12-31
+- **Authors:** Ruiyang Zhao, Fan Lam
+- **abs:** https://arxiv.org/abs/2512.24674 · **PDF:** https://arxiv.org/pdf/2512.24674v1
+
+## 24869. VLA-RAIL: A Real-Time Asynchronous Inference Linker for VLA Models and Robots
+
+- **ID:** `2512.24673` · **primary:** `cs.RO` · **Submitted:** 2025-12-31
+- **Authors:** Yongsheng Zhao, Lei Zhao, Baoping Cheng, et al.
+- **abs:** https://arxiv.org/abs/2512.24673 · **PDF:** https://arxiv.org/pdf/2512.24673v1
+
+## 24870. HeteroHBA: A Generative Structure-Manipulating Backdoor Attack on Heterogeneous Graphs
+
+- **ID:** `2512.24665` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Honglin Gao, Lan Zhao, Junhao Ren, et al.
+- **abs:** https://arxiv.org/abs/2512.24665 · **PDF:** https://arxiv.org/pdf/2512.24665v1
+
+## 24871. Renormalization Group Guided Tensor Network Structure Search
+
+- **ID:** `2512.24663` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Maolin Wang, Bowen Yu, Sheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.24663 · **PDF:** https://arxiv.org/pdf/2512.24663v1
+
+## 24872. Do Large Language Models Know What They Are Capable Of?
+
+- **ID:** `2512.24661` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Casey O. Barkan, Sid Black, Oliver Sourbut
+- **abs:** https://arxiv.org/abs/2512.24661 · **PDF:** https://arxiv.org/pdf/2512.24661v1
+
+## 24873. Hybrid Motion Planning with Deep Reinforcement Learning for Mobile Robot Navigation
+
+- **ID:** `2512.24651` · **primary:** `cs.RO` · **Submitted:** 2025-12-31
+- **Authors:** Yury Kolomeytsev, Dmitry Golembiovsky
+- **abs:** https://arxiv.org/abs/2512.24651 · **PDF:** https://arxiv.org/pdf/2512.24651v1
+
+## 24874. Diagnosing Heteroskedasticity and Resolving Multicollinearity Paradoxes in Physicochemical Property Prediction
+
+- **ID:** `2512.24643` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Malikussaid, Septian Caesar Floresko, Ade Romadhony, et al.
+- **abs:** https://arxiv.org/abs/2512.24643 · **PDF:** https://arxiv.org/pdf/2512.24643v2
+
+## 24875. Universal Conditional Logic: A Formal Language for Prompt Engineering
+
+- **ID:** `2601.00880` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Anthony Mikinka
+- **abs:** https://arxiv.org/abs/2601.00880 · **PDF:** https://arxiv.org/pdf/2601.00880v1
+
+## 24876. DynaFix: Iterative Automated Program Repair Driven by Execution-Level Dynamic Information
+
+- **ID:** `2512.24635` · **primary:** `cs.SE` · **Submitted:** 2025-12-31
+- **Authors:** Zhili Huang, Ling Xu, Chao Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.24635 · **PDF:** https://arxiv.org/pdf/2512.24635v1
+
+## 24877. Soliton profiles: Classical Numerical Schemes vs. Neural Network - Based Solvers
+
+- **ID:** `2512.24634` · **primary:** `nlin.PS` · **Submitted:** 2025-12-31
+- **Authors:** Chandler Haight, Svetlana Roudenko, Zhongming Wang
+- **abs:** https://arxiv.org/abs/2512.24634 · **PDF:** https://arxiv.org/pdf/2512.24634v1
+
+## 24878. From Detection to Diagnosis: Advancing Hallucination Analysis with Automated Data Synthesis
+
+- **ID:** `2601.09734` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Yanyi Liu, Qingwen Yang, Tiezheng Guo, et al.
+- **abs:** https://arxiv.org/abs/2601.09734 · **PDF:** https://arxiv.org/pdf/2601.09734v1
+
+## 24879. AI-Driven Acoustic Voice Biomarker-Based Hierarchical Classification of Benign Laryngeal Voice Disorders from Sustained Vowels
+
+- **ID:** `2512.24628` · **primary:** `cs.SD` · **Submitted:** 2025-12-31
+- **Authors:** Mohsen Annabestani, Samira Aghadoost, Anais Rameau, et al.
+- **abs:** https://arxiv.org/abs/2512.24628 · **PDF:** https://arxiv.org/pdf/2512.24628v1
+
+## 24880. AutoFed: Manual-Free Federated Traffic Prediction via Personalized Prompt
+
+- **ID:** `2512.24625` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Zijian Zhao, Yitong Shang, Sen Li
+- **abs:** https://arxiv.org/abs/2512.24625 · **PDF:** https://arxiv.org/pdf/2512.24625v1
+
+## 24881. Youtu-LLM: Unlocking the Native Agentic Potential for Lightweight Large Language Models
+
+- **ID:** `2512.24618` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Junru Lu, Jiarui Qin, Lingfeng Qiao, et al.
+- **abs:** https://arxiv.org/abs/2512.24618 · **PDF:** https://arxiv.org/pdf/2512.24618v2
+
+## 24882. Dynamic Large Concept Models: Latent Reasoning in an Adaptive Semantic Space
+
+- **ID:** `2512.24617` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Xingwei Qu, Shaowen Wang, Zihao Huang, et al.
+- **abs:** https://arxiv.org/abs/2512.24617 · **PDF:** https://arxiv.org/pdf/2512.24617v2
+
+## 24883. Youtu-Agent: Scaling Agent Productivity with Automated Generation and Hybrid Policy Optimization
+
+- **ID:** `2512.24615` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Yuchen Shi, Yuzheng Cai, Siqi Cai, et al.
+- **abs:** https://arxiv.org/abs/2512.24615 · **PDF:** https://arxiv.org/pdf/2512.24615v1
+
+## 24884. Chat-Driven Optimal Management for Virtual Network Services
+
+- **ID:** `2512.24614` · **primary:** `cs.NI` · **Submitted:** 2025-12-31
+- **Authors:** Yuya Miyaoka, Masaki Inoue, Kengo Urata, et al.
+- **abs:** https://arxiv.org/abs/2512.24614 · **PDF:** https://arxiv.org/pdf/2512.24614v1
+
+## 24885. Group Deliberation Oriented Multi-Agent Conversational Model for Complex Reasoning
+
+- **ID:** `2512.24613` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Zheyu Shi, Dong Qiu, Shanlong Yu
+- **abs:** https://arxiv.org/abs/2512.24613 · **PDF:** https://arxiv.org/pdf/2512.24613v1
+
+## 24886. AzeroS: Extending LLM to Speech with Self-Generated Instruction-Free Tuning
+
+- **ID:** `2601.06086` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Yiwen Shao, Wei Liu, Jiahong Li, et al.
+- **abs:** https://arxiv.org/abs/2601.06086 · **PDF:** https://arxiv.org/pdf/2601.06086v1
+
+## 24887. Reinforcement Learning-Augmented LLM Agents for Collaborative Decision Making and Performance Optimization
+
+- **ID:** `2512.24609` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Dong Qiu, Duo Xu, Limengxi Yue
+- **abs:** https://arxiv.org/abs/2512.24609 · **PDF:** https://arxiv.org/pdf/2512.24609v1
+
+## 24888. Recursive Language Models
+
+- **ID:** `2512.24601` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Alex L. Zhang, Tim Kraska, Omar Khattab
+- **abs:** https://arxiv.org/abs/2512.24601 · **PDF:** https://arxiv.org/pdf/2512.24601v2
+
+## 24889. Large Empirical Case Study: Go-Explore adapted for AI Red Team Testing
+
+- **ID:** `2601.00042` · **primary:** `cs.CR` · **Submitted:** 2025-12-31
+- **Authors:** Manish Bhatt, Adrian Wood, Idan Habler, et al.
+- **abs:** https://arxiv.org/abs/2601.00042 · **PDF:** https://arxiv.org/pdf/2601.00042v2
+
+## 24890. 3D Semantic Segmentation for Post-Disaster Assessment
+
+- **ID:** `2512.24593` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Nhut Le, Maryam Rahnemoonfar
+- **abs:** https://arxiv.org/abs/2512.24593 · **PDF:** https://arxiv.org/pdf/2512.24593v1
+
+## 24891. MultiRisk: Multiple Risk Control via Iterative Score Thresholding
+
+- **ID:** `2512.24587` · **primary:** `stat.ML` · **Submitted:** 2025-12-31
+- **Authors:** Sunay Joshi, Yan Sun, Hamed Hassani, et al.
+- **abs:** https://arxiv.org/abs/2512.24587 · **PDF:** https://arxiv.org/pdf/2512.24587v1
+
+## 24892. Robust Bayesian Dynamic Programming for On-policy Risk-sensitive Reinforcement Learning
+
+- **ID:** `2512.24580` · **primary:** `q-fin.RM` · **Submitted:** 2025-12-31
+- **Authors:** Shanyu Han, Yangbo He, Yang Liu
+- **abs:** https://arxiv.org/abs/2512.24580 · **PDF:** https://arxiv.org/pdf/2512.24580v1
+
+## 24893. Understanding and Steering the Cognitive Behaviors of Reasoning Models at Test-Time
+
+- **ID:** `2512.24574` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Zhenyu Zhang, Xiaoxia Wu, Zhongzhu Zhou, et al.
+- **abs:** https://arxiv.org/abs/2512.24574 · **PDF:** https://arxiv.org/pdf/2512.24574v2
+
+## 24894. Developmental trajectories of decision making and affective dynamics in large language models
+
+- **ID:** `2601.14268` · **primary:** `cs.CY` · **Submitted:** 2025-12-31
+- **Authors:** Zhihao Wang, Yiyang Liu, Ting Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.14268 · **PDF:** https://arxiv.org/pdf/2601.14268v1
+
+## 24895. Korean Canonical Legal Benchmark: Toward Knowledge-Independent Evaluation of LLMs' Legal Reasoning Capabilities
+
+- **ID:** `2512.24572` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Hongseok Oh, Wonseok Hwang, Kyoung-Woon On
+- **abs:** https://arxiv.org/abs/2512.24572 · **PDF:** https://arxiv.org/pdf/2512.24572v2
+
+## 24896. SynRAG: A Large Language Model Framework for Executable Query Generation in Heterogeneous SIEM System
+
+- **ID:** `2512.24571` · **primary:** `cs.CR` · **Submitted:** 2025-12-31
+- **Authors:** Md Hasan Saju, Austin Page, Akramul Azim, et al.
+- **abs:** https://arxiv.org/abs/2512.24571 · **PDF:** https://arxiv.org/pdf/2512.24571v1
+
+## 24897. UR-Bench: A Benchmark for Multi-Hop Reasoning over Ultra-High-Resolution Images
+
+- **ID:** `2601.08748` · **primary:** `cs.CV` · **Submitted:** 2025-12-31
+- **Authors:** Siqi Li, Xinyu Cai, Jianbiao Mei, et al.
+- **abs:** https://arxiv.org/abs/2601.08748 · **PDF:** https://arxiv.org/pdf/2601.08748v1
+
+## 24898. MCPAgentBench: A Real-world Task Benchmark for Evaluating LLM Agent MCP Tool Use
+
+- **ID:** `2512.24565` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Wenrui Liu, Zixiang Liu, Elsie Dai, et al.
+- **abs:** https://arxiv.org/abs/2512.24565 · **PDF:** https://arxiv.org/pdf/2512.24565v3
+
+## 24899. CPR: Causal Physiological Representation Learning for Robust ECG Analysis under Distribution Shifts
+
+- **ID:** `2512.24564` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Shunbo Jia, Caizhi Liao
+- **abs:** https://arxiv.org/abs/2512.24564 · **PDF:** https://arxiv.org/pdf/2512.24564v1
+
+## 24900. Defense Against Synthetic Speech: Real-Time Detection of RVC Voice Conversion Attacks
+
+- **ID:** `2601.04227` · **primary:** `cs.SD` · **Submitted:** 2025-12-31
+- **Authors:** Prajwal Chinchmalatpure, Suyash Chinchmalatpure, Siddharth Chavan
+- **abs:** https://arxiv.org/abs/2601.04227 · **PDF:** https://arxiv.org/pdf/2601.04227v1
+
+## 24901. HaluNet: Multi-Granular Uncertainty Modeling for Efficient Hallucination Detection in LLM Question Answering
+
+- **ID:** `2512.24562` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Chaodong Tong, Qi Zhang, Jiayang Gao, et al.
+- **abs:** https://arxiv.org/abs/2512.24562 · **PDF:** https://arxiv.org/pdf/2512.24562v1
+
+## 24902. Localized Calibrated Uncertainty in Code Language Models
+
+- **ID:** `2512.24560` · **primary:** `cs.SE` · **Submitted:** 2025-12-31
+- **Authors:** David Gros, Prem Devanbu
+- **abs:** https://arxiv.org/abs/2512.24560 · **PDF:** https://arxiv.org/pdf/2512.24560v1
+
+## 24903. Evolutionary Discovery of Sequence Acceleration Methods for Slab Geometry Neutron Transport
+
+- **ID:** `2512.24559` · **primary:** `cs.NE` · **Submitted:** 2025-12-31
+- **Authors:** Japan K. Patel, Barry D. Ganapol, Anthony Magliari, et al.
+- **abs:** https://arxiv.org/abs/2512.24559 · **PDF:** https://arxiv.org/pdf/2512.24559v1
+
+## 24904. Probabilistic Computers for Neural Quantum States
+
+- **ID:** `2512.24558` · **primary:** `quant-ph` · **Submitted:** 2025-12-31
+- **Authors:** Shuvro Chowdhury, Jasper Pieterse, Navid Anjum Aadit, et al.
+- **abs:** https://arxiv.org/abs/2512.24558 · **PDF:** https://arxiv.org/pdf/2512.24558v1
+
+## 24905. Safe in the Future, Dangerous in the Past: Dissecting Temporal and Linguistic Vulnerabilities in LLMs
+
+- **ID:** `2512.24556` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Muhammad Abdullahi Said, Muhammad Sammani Sani
+- **abs:** https://arxiv.org/abs/2512.24556 · **PDF:** https://arxiv.org/pdf/2512.24556v2
+
+## 24906. From Perception to Punchline: Empowering VLM with the Art of In-the-wild Meme
+
+- **ID:** `2512.24555` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Xueyan Li, Yingyi Xue, Mengjie Jiang, et al.
+- **abs:** https://arxiv.org/abs/2512.24555 · **PDF:** https://arxiv.org/pdf/2512.24555v2
+
+## 24907. More Than Bits: Multi-Envelope Double Binary Factorization for Extreme Quantization
+
+- **ID:** `2512.24545` · **primary:** `cs.LG` · **Submitted:** 2025-12-31
+- **Authors:** Yuma Ichikawa, Yoshihiko Fujisawa, Yudai Fujimoto, et al.
+- **abs:** https://arxiv.org/abs/2512.24545 · **PDF:** https://arxiv.org/pdf/2512.24545v1
+
+## 24908. A Graph Neural Network with Auxiliary Task Learning for Missing PMU Data Reconstruction
+
+- **ID:** `2512.24542` · **primary:** `eess.SY` · **Submitted:** 2025-12-31
+- **Authors:** Bo Li, Zijun Chen, Haiwang Zhong, et al.
+- **abs:** https://arxiv.org/abs/2512.24542 · **PDF:** https://arxiv.org/pdf/2512.24542v1
+
+## 24909. From Chaos to Clarity: Schema-Constrained AI for Auditable Biomedical Evidence Extraction from Full-Text PDFs
+
+- **ID:** `2601.14267` · **primary:** `cs.CL` · **Submitted:** 2025-12-31
+- **Authors:** Pouria Mortezaagha, Joseph Shaw, Bowen Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.14267 · **PDF:** https://arxiv.org/pdf/2601.14267v1
+
+## 24910. From Building Blocks to Planning: Multi-Step Spatial Reasoning in LLMs with Reinforcement Learning
+
+- **ID:** `2512.24532` · **primary:** `cs.AI` · **Submitted:** 2025-12-31
+- **Authors:** Amir Tahmasbi, Sadegh Majidi, Kazem Taram, et al.
+- **abs:** https://arxiv.org/abs/2512.24532 · **PDF:** https://arxiv.org/pdf/2512.24532v1
+
+## 24911. Generative AI-enhanced Sector-based Investment Portfolio Construction
+
+- **ID:** `2512.24526` · **primary:** `q-fin.PM` · **Submitted:** 2025-12-31
+- **Authors:** Alina Voronina, Oleksandr Romanko, Ruiwen Cao, et al.
+- **abs:** https://arxiv.org/abs/2512.24526 · **PDF:** https://arxiv.org/pdf/2512.24526v1
+
+## 24912. Deep Learning Approach for the Diagnosis of Pediatric Pneumonia Using Chest X-ray Imaging
+
+- **ID:** `2601.00041` · **primary:** `eess.IV` · **Submitted:** 2025-12-31
+- **Authors:** Fatemeh Hosseinabadi, Mohammad Mojtaba Rohani
+- **abs:** https://arxiv.org/abs/2601.00041 · **PDF:** https://arxiv.org/pdf/2601.00041v1
+
+## 24913. GCG Attack On A Diffusion LLM
+
+- **ID:** `2601.14266` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Ruben Neyroud, Sam Corley
+- **abs:** https://arxiv.org/abs/2601.14266 · **PDF:** https://arxiv.org/pdf/2601.14266v1
+
+## 24914. LearnAD: Learning Interpretable Rules for Brain Networks in Alzheimer's Disease Classification
+
+- **ID:** `2601.00877` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Thomas Andrews, Mark Law, Sara Ahmadi-Abhari, et al.
+- **abs:** https://arxiv.org/abs/2601.00877 · **PDF:** https://arxiv.org/pdf/2601.00877v1
+
+## 24915. Paragraph Segmentation Revisited: Towards a Standard Task for Structuring Speech
+
+- **ID:** `2512.24517` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Fabian Retkowski, Alexander Waibel
+- **abs:** https://arxiv.org/abs/2512.24517 · **PDF:** https://arxiv.org/pdf/2512.24517v1
+
+## 24916. Improving the stability of the covariance-controlled adaptive Langevin thermostat for large-scale Bayesian sampling
+
+- **ID:** `2512.24515` · **primary:** `stat.ML` · **Submitted:** 2025-12-30
+- **Authors:** Jiani Wei, Xiaocheng Shang
+- **abs:** https://arxiv.org/abs/2512.24515 · **PDF:** https://arxiv.org/pdf/2512.24515v1
+
+## 24917. Generalising E-prop to Deep Networks
+
+- **ID:** `2512.24506` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Beren Millidge
+- **abs:** https://arxiv.org/abs/2512.24506 · **PDF:** https://arxiv.org/pdf/2512.24506v1
+
+## 24918. Evaluating the Reasoning Abilities of LLMs on Underrepresented Mathematics Competition Problems
+
+- **ID:** `2512.24505` · **primary:** `cs.AI` · **Submitted:** 2025-12-30
+- **Authors:** Samuel Golladay, Majid Bani-Yaghoub
+- **abs:** https://arxiv.org/abs/2512.24505 · **PDF:** https://arxiv.org/pdf/2512.24505v1
+
+## 24919. Thinking on Maps: How Foundation Model Agents Explore, Remember, and Reason Map Environments
+
+- **ID:** `2512.24504` · **primary:** `cs.AI` · **Submitted:** 2025-12-30
+- **Authors:** Zhiwei Wei, Yuxing Liu, Hua Liao, et al.
+- **abs:** https://arxiv.org/abs/2512.24504 · **PDF:** https://arxiv.org/pdf/2512.24504v2
+
+## 24920. Can Small Training Runs Reliably Guide Data Curation? Rethinking Proxy-Model Practice
+
+- **ID:** `2512.24503` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Jiachen T. Wang, Tong Wu, Kaifeng Lyu, et al.
+- **abs:** https://arxiv.org/abs/2512.24503 · **PDF:** https://arxiv.org/pdf/2512.24503v1
+
+## 24921. What Drives Success in Physical Planning with Joint-Embedding Predictive World Models?
+
+- **ID:** `2512.24497` · **primary:** `cs.AI` · **Submitted:** 2025-12-30
+- **Authors:** Basile Terver, Tsung-Yen Yang, Jean Ponce, et al.
+- **abs:** https://arxiv.org/abs/2512.24497 · **PDF:** https://arxiv.org/pdf/2512.24497v2
+
+## 24922. Automated Classification of First-Trimester Fetal Heart Views Using Ultrasound-Specific Self-Supervised Learning
+
+- **ID:** `2512.24492` · **primary:** `eess.IV` · **Submitted:** 2025-12-30
+- **Authors:** Youssef Megahed, Aylin Erman, Robin Ducharme, et al.
+- **abs:** https://arxiv.org/abs/2512.24492 · **PDF:** https://arxiv.org/pdf/2512.24492v1
+
+## 24923. HOLOGRAPH: Active Causal Discovery via Sheaf-Theoretic Alignment of Large Language Model Priors
+
+- **ID:** `2512.24478` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Hyunjun Kim
+- **abs:** https://arxiv.org/abs/2512.24478 · **PDF:** https://arxiv.org/pdf/2512.24478v2
+
+## 24924. F2IDiff: Real-world Image Super-resolution using Feature to Image Diffusion Foundation Model
+
+- **ID:** `2512.24473` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Devendra K. Jangid, Ripon K. Saha, Dilshan Godaliyadda, et al.
+- **abs:** https://arxiv.org/abs/2512.24473 · **PDF:** https://arxiv.org/pdf/2512.24473v1
+
+## 24925. Foundation models on the bridge: Semantic hazard detection and safety maneuvers for maritime autonomy with vision-language models
+
+- **ID:** `2512.24470` · **primary:** `cs.RO` · **Submitted:** 2025-12-30
+- **Authors:** Kim Alexander Christensen, Andreas Gudahl Tufte, Alexey Gusev, et al.
+- **abs:** https://arxiv.org/abs/2512.24470 · **PDF:** https://arxiv.org/pdf/2512.24470v2
+
+## 24926. Spectral and Spatial Graph Learning for Multispectral Solar Image Compression
+
+- **ID:** `2512.24463` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Prasiddha Siwakoti, Atefeh Khoshkhahtinat, Piyush M. Mehta, et al.
+- **abs:** https://arxiv.org/abs/2512.24463 · **PDF:** https://arxiv.org/pdf/2512.24463v1
+
+## 24927. Align While Search: Belief-Guided Exploratory Inference for World-Grounded Embodied Agents
+
+- **ID:** `2512.24461` · **primary:** `cs.AI` · **Submitted:** 2025-12-30
+- **Authors:** Seohui Bae, Jeonghye Kim, Youngchul Sung, et al.
+- **abs:** https://arxiv.org/abs/2512.24461 · **PDF:** https://arxiv.org/pdf/2512.24461v1
+
+## 24928. IELTS Writing Revision Platform with Automated Essay Scoring and Adaptive Feedback
+
+- **ID:** `2512.24460` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Titas Ramancauskas, Kotryna Ramancauske
+- **abs:** https://arxiv.org/abs/2512.24460 · **PDF:** https://arxiv.org/pdf/2512.24460v1
+
+## 24929. Cleaning English Abstracts of Scientific Publications
+
+- **ID:** `2512.24459` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Michael E. Rose, Nils A. Herrmann, Sebastian Erhardt
+- **abs:** https://arxiv.org/abs/2512.24459 · **PDF:** https://arxiv.org/pdf/2512.24459v1
+
+## 24930. Detecting and Mitigating Treatment Leakage in Text-Based Causal Inference: Distillation and Sensitivity Analysis
+
+- **ID:** `2601.02400` · **primary:** `econ.EM` · **Submitted:** 2025-12-30
+- **Authors:** Adel Daoud, Richard Johansson, Connor T. Jerzak
+- **abs:** https://arxiv.org/abs/2601.02400 · **PDF:** https://arxiv.org/pdf/2601.02400v1
+
+## 24931. Privacy-Preserving Semantic Communications via Multi-Task Learning and Adversarial Perturbations
+
+- **ID:** `2512.24452` · **primary:** `cs.NI` · **Submitted:** 2025-12-30
+- **Authors:** Yalin E. Sagduyu, Tugba Erpek, Aylin Yener, et al.
+- **abs:** https://arxiv.org/abs/2512.24452 · **PDF:** https://arxiv.org/pdf/2512.24452v1
+
+## 24932. PackKV: Reducing KV Cache Memory Footprint through LLM-Aware Lossy Compression
+
+- **ID:** `2512.24449` · **primary:** `cs.DC` · **Submitted:** 2025-12-30
+- **Authors:** Bo Jiang, Taolue Yang, Youyuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.24449 · **PDF:** https://arxiv.org/pdf/2512.24449v2
+
+## 24933. LLMize: A Framework for Large Language Model-Based Numerical Optimization
+
+- **ID:** `2601.00874` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** M. Rizki Oktavian
+- **abs:** https://arxiv.org/abs/2601.00874 · **PDF:** https://arxiv.org/pdf/2601.00874v1
+
+## 24934. Generative forecasting with joint probability models
+
+- **ID:** `2512.24446` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Patrick Wyrod, Ashesh Chattopadhyay, Daniele Venturi
+- **abs:** https://arxiv.org/abs/2512.24446 · **PDF:** https://arxiv.org/pdf/2512.24446v1
+
+## 24935. Adaptive Learning Guided by Bias-Noise-Alignment Diagnostics
+
+- **ID:** `2512.24445` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Akash Samanta, Sheldon Williamson
+- **abs:** https://arxiv.org/abs/2512.24445 · **PDF:** https://arxiv.org/pdf/2512.24445v2
+
+## 24936. Sparse classification with positive-confidence data in high dimensions
+
+- **ID:** `2512.24443` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** The Tien Mai, Mai Anh Nguyen, Trung Nghia Nguyen
+- **abs:** https://arxiv.org/abs/2512.24443 · **PDF:** https://arxiv.org/pdf/2512.24443v1
+
+## 24937. Towards mechanistic understanding in a data-driven weather model: internal activations reveal interpretable physical features
+
+- **ID:** `2512.24440` · **primary:** `physics.ao-ph` · **Submitted:** 2025-12-30
+- **Authors:** Theodore MacMillan, Nicholas T. Ouellette
+- **abs:** https://arxiv.org/abs/2512.24440 · **PDF:** https://arxiv.org/pdf/2512.24440v1
+
+## 24938. Active learning for data-driven reduced models of parametric differential systems with Bayesian operator inference
+
+- **ID:** `2601.00038` · **primary:** `stat.ML` · **Submitted:** 2025-12-30
+- **Authors:** Shane A. McQuarrie, Mengwu Guo, Anirban Chaudhuri
+- **abs:** https://arxiv.org/abs/2601.00038 · **PDF:** https://arxiv.org/pdf/2601.00038v1
+
+## 24939. Virasoro Symmetry in Neural Network Field Theories
+
+- **ID:** `2512.24420` · **primary:** `hep-th` · **Submitted:** 2025-12-30
+- **Authors:** Brandon Robinson
+- **abs:** https://arxiv.org/abs/2512.24420 · **PDF:** https://arxiv.org/pdf/2512.24420v3
+
+## 24940. Comparing Approaches to Automatic Summarization in Less-Resourced Languages
+
+- **ID:** `2512.24410` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Chester Palen-Michel, Constantine Lignos
+- **abs:** https://arxiv.org/abs/2512.24410 · **PDF:** https://arxiv.org/pdf/2512.24410v1
+
+## 24941. Efficient Inference for Inverse Reinforcement Learning and Dynamic Discrete Choice Models
+
+- **ID:** `2512.24407` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Lars van der Laan, Aurelien Bibaut, Nathan Kallus
+- **abs:** https://arxiv.org/abs/2512.24407 · **PDF:** https://arxiv.org/pdf/2512.24407v1
+
+## 24942. Lifting Vision: Ground to Aerial Localization with Reasoning Guided Planning
+
+- **ID:** `2512.24404` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Soham Pahari, M. Srinivas
+- **abs:** https://arxiv.org/abs/2512.24404 · **PDF:** https://arxiv.org/pdf/2512.24404v1
+
+## 24943. Fast and Realistic Automated Scenario Simulations and Reporting for an Autonomous Racing Stack
+
+- **ID:** `2512.24402` · **primary:** `cs.RO` · **Submitted:** 2025-12-30
+- **Authors:** Giovanni Lambertini, Matteo Pini, Eugenio Mascaro, et al.
+- **abs:** https://arxiv.org/abs/2512.24402 · **PDF:** https://arxiv.org/pdf/2512.24402v1
+
+## 24944. Bielik 11B v3: Multilingual Large Language Model for European Languages
+
+- **ID:** `2601.11579` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Krzysztof Ociepa, Łukasz Flis, Remigiusz Kinas, et al.
+- **abs:** https://arxiv.org/abs/2601.11579 · **PDF:** https://arxiv.org/pdf/2601.11579v1
+
+## 24945. Quantum Machine Learning Approaches for Coordinated Stealth Attack Detection in Distributed Generation Systems
+
+- **ID:** `2601.00873` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Osasumwen Cedric Ogiesoba-Eguakun, Suman Rath
+- **abs:** https://arxiv.org/abs/2601.00873 · **PDF:** https://arxiv.org/pdf/2601.00873v1
+
+## 24946. Multi-Agent LLMs for Generating Research Limitations
+
+- **ID:** `2601.11578` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Ibrahim Al Azher, Zhishuai Guo, Hamed Alhoori
+- **abs:** https://arxiv.org/abs/2601.11578 · **PDF:** https://arxiv.org/pdf/2601.11578v2
+
+## 24947. FAST-IDS: A Fast Two-Stage Intrusion Detection System with Hybrid Compression for Real-Time Threat Detection in Connected and Autonomous Vehicles
+
+- **ID:** `2512.24391` · **primary:** `cs.CR` · **Submitted:** 2025-12-30
+- **Authors:** Devika S, Vishnu Hari, Pratik Narang, et al.
+- **abs:** https://arxiv.org/abs/2512.24391 · **PDF:** https://arxiv.org/pdf/2512.24391v1
+
+## 24948. Tubular Riemannian Laplace Approximations for Bayesian Neural Networks
+
+- **ID:** `2512.24381` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Rodrigo Pereira David
+- **abs:** https://arxiv.org/abs/2512.24381 · **PDF:** https://arxiv.org/pdf/2512.24381v2
+
+## 24949. Closing the Data Loop: Using OpenDataArena to Engineer Superior Training Datasets
+
+- **ID:** `2601.09733` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Xin Gao, Xiaoyang Wang, Yun Zhu, et al.
+- **abs:** https://arxiv.org/abs/2601.09733 · **PDF:** https://arxiv.org/pdf/2601.09733v1
+
+## 24950. Implicit score matching meets denoising score matching: improved rates of convergence and log-density Hessian estimation
+
+- **ID:** `2512.24378` · **primary:** `math.ST` · **Submitted:** 2025-12-30
+- **Authors:** Konstantin Yakovlev, Anna Markovich, Nikita Puchkin
+- **abs:** https://arxiv.org/abs/2512.24378 · **PDF:** https://arxiv.org/pdf/2512.24378v1
+
+## 24951. Skim-Aware Contrastive Learning for Efficient Document Representation
+
+- **ID:** `2512.24373` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Waheed Ahmed Abro, Zied Bouraoui
+- **abs:** https://arxiv.org/abs/2512.24373 · **PDF:** https://arxiv.org/pdf/2512.24373v1
+
+## 24952. Deep Learning in Geotechnical Engineering: A Critical Assessment of PINNs and Operator Learning
+
+- **ID:** `2512.24365` · **primary:** `physics.geo-ph` · **Submitted:** 2025-12-30
+- **Authors:** Krishna Kumar
+- **abs:** https://arxiv.org/abs/2512.24365 · **PDF:** https://arxiv.org/pdf/2512.24365v1
+
+## 24953. FedSecureFormer: A Fast, Federated and Secure Transformer Framework for Lightweight Intrusion Detection in Connected and Autonomous Vehicles
+
+- **ID:** `2512.24345` · **primary:** `cs.CR` · **Submitted:** 2025-12-30
+- **Authors:** Devika S, Vishnu Hari, Pratik Narang, et al.
+- **abs:** https://arxiv.org/abs/2512.24345 · **PDF:** https://arxiv.org/pdf/2512.24345v1
+
+## 24954. DermaVQA-DAS: Dermatology Assessment Schema (DAS) & Datasets for Closed-Ended Question Answering & Segmentation in Patient-Generated Dermatology Images
+
+- **ID:** `2512.24340` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Wen-wai Yim, Yujuan Fu, Asma Ben Abacha, et al.
+- **abs:** https://arxiv.org/abs/2512.24340 · **PDF:** https://arxiv.org/pdf/2512.24340v1
+
+## 24955. OptiVote: Non-Coherent FSO Over-the-Air Majority Vote for Communication-Efficient Distributed Federated Learning in Space Data Centers
+
+- **ID:** `2512.24334` · **primary:** `eess.SP` · **Submitted:** 2025-12-30
+- **Authors:** Anbang Zhang, Chenyuan Feng, Wai Ho Mow, et al.
+- **abs:** https://arxiv.org/abs/2512.24334 · **PDF:** https://arxiv.org/pdf/2512.24334v2
+
+## 24956. World model inspired sarcasm reasoning with large language model agents
+
+- **ID:** `2512.24329` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Keito Inoshita, Shinnosuke Mizuno
+- **abs:** https://arxiv.org/abs/2512.24329 · **PDF:** https://arxiv.org/pdf/2512.24329v1
+
+## 24957. Topological Spatial Graph Coarsening
+
+- **ID:** `2512.24327` · **primary:** `stat.ML` · **Submitted:** 2025-12-30
+- **Authors:** Anna Calissano, Etienne Lasalle
+- **abs:** https://arxiv.org/abs/2512.24327 · **PDF:** https://arxiv.org/pdf/2512.24327v2
+
+## 24958. MaRCA: Multi-Agent Reinforcement Learning for Dynamic Computation Allocation in Large-Scale Recommender Systems
+
+- **ID:** `2512.24325` · **primary:** `cs.IR` · **Submitted:** 2025-12-30
+- **Authors:** Wan Jiang, Xinyi Zang, Yudong Zhao, et al.
+- **abs:** https://arxiv.org/abs/2512.24325 · **PDF:** https://arxiv.org/pdf/2512.24325v1
+
+## 24959. Empower Low-Altitude Economy: A Reliability-Aware Dynamic Weighting Allocation for Multi-modal UAV Beam Prediction
+
+- **ID:** `2512.24324` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Haojin Li, Anbang Zhang, Chen Sun, et al.
+- **abs:** https://arxiv.org/abs/2512.24324 · **PDF:** https://arxiv.org/pdf/2512.24324v1
+
+## 24960. QianfanHuijin Technical Report: A Novel Multi-Stage Training Paradigm for Finance Industrial LLMs
+
+- **ID:** `2512.24314` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Shupeng Li, Weipeng Lu, Linyun Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.24314 · **PDF:** https://arxiv.org/pdf/2512.24314v2
+
+## 24961. Automated Reproducibility Has a Problem Statement Problem
+
+- **ID:** `2601.04226` · **primary:** `cs.CY` · **Submitted:** 2025-12-30
+- **Authors:** Thijs Snelleman, Peter Lundestad Lawrence, Holger H. Hoos, et al.
+- **abs:** https://arxiv.org/abs/2601.04226 · **PDF:** https://arxiv.org/pdf/2601.04226v1
+
+## 24962. Generative Video Compression: Towards 0.01% Compression Rate for Video Transmission
+
+- **ID:** `2512.24300` · **primary:** `eess.IV` · **Submitted:** 2025-12-30
+- **Authors:** Xiangyu Chen, Jixiang Luo, Jingyu Xu, et al.
+- **abs:** https://arxiv.org/abs/2512.24300 · **PDF:** https://arxiv.org/pdf/2512.24300v2
+
+## 24963. Figure It Out: Improve the Frontier of Reasoning with Executable Visual States
+
+- **ID:** `2512.24297` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Meiqi Chen, Fandong Meng, Jie Zhou
+- **abs:** https://arxiv.org/abs/2512.24297 · **PDF:** https://arxiv.org/pdf/2512.24297v2
+
+## 24964. Virtual-Eyes: Quantitative Validation of a Lung CT Quality-Control Pipeline for Foundation-Model Cancer Risk Prediction
+
+- **ID:** `2512.24294` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Md. Enamul Hoq, Linda Larson-Prior, Fred Prior
+- **abs:** https://arxiv.org/abs/2512.24294 · **PDF:** https://arxiv.org/pdf/2512.24294v1
+
+## 24965. Deep versus Broad Technology Search and the Timing of Innovation Impact
+
+- **ID:** `2601.00871` · **primary:** `physics.soc-ph` · **Submitted:** 2025-12-30
+- **Authors:** Likun Cao, James Evans
+- **abs:** https://arxiv.org/abs/2601.00871 · **PDF:** https://arxiv.org/pdf/2601.00871v1
+
+## 24966. Fast reconstruction-based ROI triggering via anomaly detection in the CYGNO optical TPC
+
+- **ID:** `2512.24290` · **primary:** `physics.ins-det` · **Submitted:** 2025-12-30
+- **Authors:** F. D. Amaro, R. Antonietti, E. Baracchini, et al.
+- **abs:** https://arxiv.org/abs/2512.24290 · **PDF:** https://arxiv.org/pdf/2512.24290v1
+
+## 24967. Automated Analysis of Sustainability Reports: Using Large Language Models for the Extraction and Prediction of EU Taxonomy-Compliant KPIs
+
+- **ID:** `2512.24289` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Jonathan Schmoll, Adam Jatowt
+- **abs:** https://arxiv.org/abs/2512.24289 · **PDF:** https://arxiv.org/pdf/2512.24289v1
+
+## 24968. DRL-TH: Jointly Utilizing Temporal Graph Attention and Hierarchical Fusion for UGV Navigation in Crowded Environments
+
+- **ID:** `2512.24284` · **primary:** `cs.RO` · **Submitted:** 2025-12-30
+- **Authors:** Ruitong Li, Lin Zhang, Yuenan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2512.24284 · **PDF:** https://arxiv.org/pdf/2512.24284v1
+
+## 24969. One-shot synthesis of rare gastrointestinal lesions improves diagnostic accuracy and clinical training
+
+- **ID:** `2512.24278` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Jia Yu, Yan Zhu, Peiyao Fu, et al.
+- **abs:** https://arxiv.org/abs/2512.24278 · **PDF:** https://arxiv.org/pdf/2512.24278v1
+
+## 24970. Taming Hallucinations: Boosting MLLMs' Video Understanding via Counterfactual Video Generation
+
+- **ID:** `2512.24271` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Zhe Huang, Hao Wen, Aiming Hao, et al.
+- **abs:** https://arxiv.org/abs/2512.24271 · **PDF:** https://arxiv.org/pdf/2512.24271v1
+
+## 24971. Joint Selection for Large-Scale Pre-Training Data via Policy Gradient-based Mask Learning
+
+- **ID:** `2512.24265` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Ziqing Fan, Yuqiao Xian, Yan Sun, et al.
+- **abs:** https://arxiv.org/abs/2512.24265 · **PDF:** https://arxiv.org/pdf/2512.24265v1
+
+## 24972. Constrained Language Model Policy Optimization via Risk-aware Stepwise Alignment
+
+- **ID:** `2512.24263` · **primary:** `cs.AI` · **Submitted:** 2025-12-30
+- **Authors:** Lijun Zhang, Lin Li, Wei Wei, et al.
+- **abs:** https://arxiv.org/abs/2512.24263 · **PDF:** https://arxiv.org/pdf/2512.24263v1
+
+## 24973. Tracing the Flow of Knowledge From Science to Technology Using Deep Learning
+
+- **ID:** `2512.24259` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Michael E. Rose, Mainak Ghosh, Sebastian Erhardt, et al.
+- **abs:** https://arxiv.org/abs/2512.24259 · **PDF:** https://arxiv.org/pdf/2512.24259v1
+
+## 24974. Early Prediction of Sepsis using Heart Rate Signals and Genetic Optimized LSTM Algorithm
+
+- **ID:** `2512.24253` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Alireza Rafiei, Farshid Hajati, Alireza Rezaee, et al.
+- **abs:** https://arxiv.org/abs/2512.24253 · **PDF:** https://arxiv.org/pdf/2512.24253v1
+
+## 24975. Deep Reinforcement Learning for Solving the Fleet Size and Mix Vehicle Routing Problem
+
+- **ID:** `2512.24251` · **primary:** `cs.AI` · **Submitted:** 2025-12-30
+- **Authors:** Pengfu Wan, Jiawei Chen, Gangyan Xu
+- **abs:** https://arxiv.org/abs/2512.24251 · **PDF:** https://arxiv.org/pdf/2512.24251v1
+
+## 24976. Simulation-Free PSRO: Removing Game Simulation from Policy Space Response Oracles
+
+- **ID:** `2601.05279` · **primary:** `cs.MA` · **Submitted:** 2025-12-30
+- **Authors:** Yingzhuo Liu, Shuodi Liu, Weijun Luo, et al.
+- **abs:** https://arxiv.org/abs/2601.05279 · **PDF:** https://arxiv.org/pdf/2601.05279v1
+
+## 24977. Cultural Encoding in Large Language Models: The Existence Gap in AI-Mediated Brand Discovery
+
+- **ID:** `2601.00869` · **primary:** `cs.AI` · **Submitted:** 2025-12-30
+- **Authors:** Huang Junyao, Situ Ruimin, Ye Renqin
+- **abs:** https://arxiv.org/abs/2601.00869 · **PDF:** https://arxiv.org/pdf/2601.00869v1
+
+## 24978. LAILA: A Large Trait-Based Dataset for Arabic Automated Essay Scoring
+
+- **ID:** `2512.24235` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** May Bashendy, Walid Massoud, Sohaila Eltanbouly, et al.
+- **abs:** https://arxiv.org/abs/2512.24235 · **PDF:** https://arxiv.org/pdf/2512.24235v2
+
+## 24979. MotivNet: Evolving Meta-Sapiens into an Emotionally Intelligent Foundation Model
+
+- **ID:** `2512.24231` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Rahul Medicharla, Alper Yilmaz
+- **abs:** https://arxiv.org/abs/2512.24231 · **PDF:** https://arxiv.org/pdf/2512.24231v1
+
+## 24980. SmartFlow Reinforcement Learning and Agentic AI for Bike-Sharing Optimisation
+
+- **ID:** `2601.00868` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Aditya Sreevatsa K, Arun Kumar Raveendran, Jesrael K Mani, et al.
+- **abs:** https://arxiv.org/abs/2601.00868 · **PDF:** https://arxiv.org/pdf/2601.00868v1
+
+## 24981. Medical Image Classification on Imbalanced Data Using ProGAN and SMA-Optimized ResNet: Application to COVID-19
+
+- **ID:** `2512.24214` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Sina Jahromi, Farshid Hajati, Alireza Rezaee, et al.
+- **abs:** https://arxiv.org/abs/2512.24214 · **PDF:** https://arxiv.org/pdf/2512.24214v1
+
+## 24982. The Silicon Psyche: Anthropomorphic Vulnerabilities in Large Language Models
+
+- **ID:** `2601.00867` · **primary:** `cs.CR` · **Submitted:** 2025-12-30
+- **Authors:** Giuseppe Canale, Kashyap Thimmaraju
+- **abs:** https://arxiv.org/abs/2601.00867 · **PDF:** https://arxiv.org/pdf/2601.00867v1
+
+## 24983. A-PINN: Auxiliary Physics-informed Neural Networks for Structural Vibration Analysis in Continuous Euler-Bernoulli Beam
+
+- **ID:** `2601.00866` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Shivani Saini, Ramesh Kumar Vats, Arup Kumar Sahoo
+- **abs:** https://arxiv.org/abs/2601.00866 · **PDF:** https://arxiv.org/pdf/2601.00866v1
+
+## 24984. Micro-Macro Tensor Neural Surrogates for Uncertainty Quantification in Collisional Plasma
+
+- **ID:** `2512.24205` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Wei Chen, Giacomo Dimarco, Lorenzo Pareschi
+- **abs:** https://arxiv.org/abs/2512.24205 · **PDF:** https://arxiv.org/pdf/2512.24205v1
+
+## 24985. PointRAFT: 3D deep learning for high-throughput prediction of potato tuber weight from partial point clouds
+
+- **ID:** `2512.24193` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Pieter M. Blok, Haozhou Wang, Hyun Kwon Suh, et al.
+- **abs:** https://arxiv.org/abs/2512.24193 · **PDF:** https://arxiv.org/pdf/2512.24193v1
+
+## 24986. SCP: Accelerating Discovery with a Global Web of Autonomous Scientific Agents
+
+- **ID:** `2512.24189` · **primary:** `cs.AI` · **Submitted:** 2025-12-30
+- **Authors:** Yankai Jiang, Wenjie Lou, Lilong Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.24189 · **PDF:** https://arxiv.org/pdf/2512.24189v1
+
+## 24987. MedKGI: Iterative Differential Diagnosis with Medical Knowledge Graphs and Information-Guided Inquiring
+
+- **ID:** `2512.24181` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Qipeng Wang, Rui Sheng, Yafei Li, et al.
+- **abs:** https://arxiv.org/abs/2512.24181 · **PDF:** https://arxiv.org/pdf/2512.24181v2
+
+## 24988. Guiding a Diffusion Transformer with the Internal Dynamics of Itself
+
+- **ID:** `2512.24176` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Xingyu Zhou, Qifan Li, Xiaobin Hu, et al.
+- **abs:** https://arxiv.org/abs/2512.24176 · **PDF:** https://arxiv.org/pdf/2512.24176v2
+
+## 24989. Variational Quantum Brushes
+
+- **ID:** `2512.24173` · **primary:** `quant-ph` · **Submitted:** 2025-12-30
+- **Authors:** Jui-Ting Lu, Henrique Ennes, Chih-Kang Huang, et al.
+- **abs:** https://arxiv.org/abs/2512.24173 · **PDF:** https://arxiv.org/pdf/2512.24173v1
+
+## 24990. Deep Global Clustering for Hyperspectral Image Segmentation: Concepts, Applications, and Open Challenges
+
+- **ID:** `2512.24172` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Yu-Tang Chang, Pin-Wei Chen, Shih-Fang Chen
+- **abs:** https://arxiv.org/abs/2512.24172 · **PDF:** https://arxiv.org/pdf/2512.24172v1
+
+## 24991. Distribution Matching for Graph Quantification Under Structural Covariate Shift
+
+- **ID:** `2601.00864` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Clemens Damke, Eyke Hüllermeier
+- **abs:** https://arxiv.org/abs/2601.00864 · **PDF:** https://arxiv.org/pdf/2601.00864v1
+
+## 24992. Concept Attractors in LLMs and their Applications
+
+- **ID:** `2601.11575` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Sotirios Panagiotis Chytas, Vikas Singh
+- **abs:** https://arxiv.org/abs/2601.11575 · **PDF:** https://arxiv.org/pdf/2601.11575v1
+
+## 24993. Developing controlled natural language for formal specification patterns using AI assistants
+
+- **ID:** `2512.24159` · **primary:** `cs.SE` · **Submitted:** 2025-12-30
+- **Authors:** Natalia Garanina, Vladimir Zyubin, Igor Anureev
+- **abs:** https://arxiv.org/abs/2512.24159 · **PDF:** https://arxiv.org/pdf/2512.24159v1
+
+## 24994. Training Report of TeleChat3-MoE
+
+- **ID:** `2512.24157` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Xinzhang Liu, Chao Wang, Zhihao Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.24157 · **PDF:** https://arxiv.org/pdf/2512.24157v1
+
+## 24995. Graph-Based Exploration for ARC-AGI-3 Interactive Reasoning Tasks
+
+- **ID:** `2512.24156` · **primary:** `cs.AI` · **Submitted:** 2025-12-30
+- **Authors:** Evgenii Rudakov, Jonathan Shock, Benjamin Ultan Cowley
+- **abs:** https://arxiv.org/abs/2512.24156 · **PDF:** https://arxiv.org/pdf/2512.24156v1
+
+## 24996. Score-based sampling without diffusions: Guidance from a simple and modular scheme
+
+- **ID:** `2512.24152` · **primary:** `math.ST` · **Submitted:** 2025-12-30
+- **Authors:** M. J. Wainwright
+- **abs:** https://arxiv.org/abs/2512.24152 · **PDF:** https://arxiv.org/pdf/2512.24152v1
+
+## 24997. Large Emotional World Model
+
+- **ID:** `2512.24149` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Changhao Song, Yazhou Zhang, Hui Gao, et al.
+- **abs:** https://arxiv.org/abs/2512.24149 · **PDF:** https://arxiv.org/pdf/2512.24149v1
+
+## 24998. When Does Pairing Seeds Reduce Variance? Evidence from a Multi-Agent Economic Simulation
+
+- **ID:** `2512.24145` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Udit Sharma
+- **abs:** https://arxiv.org/abs/2512.24145 · **PDF:** https://arxiv.org/pdf/2512.24145v3
+
+## 24999. Selective Imperfection as a Generative Framework for Analysis, Creativity and Discovery
+
+- **ID:** `2601.00863` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Markus J. Buehler
+- **abs:** https://arxiv.org/abs/2601.00863 · **PDF:** https://arxiv.org/pdf/2601.00863v1
+
+## 25000. Activation Steering for Masked Diffusion Language Models
+
+- **ID:** `2512.24143` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Adi Shnaidman, Erin Feiglin, Osher Yaari, et al.
+- **abs:** https://arxiv.org/abs/2512.24143 · **PDF:** https://arxiv.org/pdf/2512.24143v3
+
+## 25001. Colorful Pinball: Density-Weighted Quantile Regression for Conditional Guarantee of Conformal Prediction
+
+- **ID:** `2512.24139` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Qianyi Chen, Bo Li
+- **abs:** https://arxiv.org/abs/2512.24139 · **PDF:** https://arxiv.org/pdf/2512.24139v4
+
+## 25002. GARDO: Reinforcing Diffusion Models without Reward Hacking
+
+- **ID:** `2512.24138` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Haoran He, Yuxiao Ye, Jie Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.24138 · **PDF:** https://arxiv.org/pdf/2512.24138v1
+
+## 25003. Unified Embodied VLM Reasoning with Robotic Action via Autoregressive Discretized Pre-training
+
+- **ID:** `2512.24125` · **primary:** `cs.RO` · **Submitted:** 2025-12-30
+- **Authors:** Yi Liu, Sukai Wang, Dafeng Wei, et al.
+- **abs:** https://arxiv.org/abs/2512.24125 · **PDF:** https://arxiv.org/pdf/2512.24125v2
+
+## 25004. Universal Battery Degradation Forecasting Driven by Foundation Model Across Diverse Chemistries and Conditions
+
+- **ID:** `2601.00862` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Joey Chan, Huan Wang, Haoyu Pan, et al.
+- **abs:** https://arxiv.org/abs/2601.00862 · **PDF:** https://arxiv.org/pdf/2601.00862v1
+
+## 25005. OptRot: Mitigating Weight Outliers via Data-Free Rotations for Post-Training Quantization
+
+- **ID:** `2512.24124` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Advait Gadhikar, Riccardo Grazzi, James Hensman
+- **abs:** https://arxiv.org/abs/2512.24124 · **PDF:** https://arxiv.org/pdf/2512.24124v2
+
+## 25006. Enhancing LLM-Based Neural Network Generation: Few-Shot Prompting and Efficient Validation for Automated Architecture Design
+
+- **ID:** `2512.24120` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Chandini Vysyaraju, Raghuvir Duvvuri, Avi Goyal, et al.
+- **abs:** https://arxiv.org/abs/2512.24120 · **PDF:** https://arxiv.org/pdf/2512.24120v1
+
+## 25007. Quantitative Understanding of PDF Fits and their Uncertainties
+
+- **ID:** `2512.24116` · **primary:** `hep-ph` · **Submitted:** 2025-12-30
+- **Authors:** Amedeo Chiefa, Luigi Del Debbio, Richard Kenway
+- **abs:** https://arxiv.org/abs/2512.24116 · **PDF:** https://arxiv.org/pdf/2512.24116v2
+
+## 25008. CogRec: A Cognitive Recommender Agent Fusing Large Language Models and Soar for Explainable Recommendation
+
+- **ID:** `2512.24113` · **primary:** `cs.AI` · **Submitted:** 2025-12-30
+- **Authors:** Jiaxin Hu, Tao Wang, Bingsan Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.24113 · **PDF:** https://arxiv.org/pdf/2512.24113v1
+
+## 25009. Constructive Approximation of Random Process via Stochastic Interpolation Neural Network Operators
+
+- **ID:** `2512.24106` · **primary:** `stat.ML` · **Submitted:** 2025-12-30
+- **Authors:** Sachin Saini, Uaday Singh
+- **abs:** https://arxiv.org/abs/2512.24106 · **PDF:** https://arxiv.org/pdf/2512.24106v1
+
+## 25010. Multilevel Fair Allocation
+
+- **ID:** `2512.24105` · **primary:** `cs.GT` · **Submitted:** 2025-12-30
+- **Authors:** Maxime Lucet, Nawal Benabbou, Aurélie Beynier, et al.
+- **abs:** https://arxiv.org/abs/2512.24105 · **PDF:** https://arxiv.org/pdf/2512.24105v1
+
+## 25011. Enhancing LLM Planning Capabilities through Intrinsic Self-Critique
+
+- **ID:** `2512.24103` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Bernd Bohnet, Pierre-Alexandre Kamienny, Hanie Sedghi, et al.
+- **abs:** https://arxiv.org/abs/2512.24103 · **PDF:** https://arxiv.org/pdf/2512.24103v1
+
+## 25012. Autoregressivity in the Latent Space of a GP-VAE Language Model: An Empirical Ablation Study
+
+- **ID:** `2512.24102` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Yves Ruffenach
+- **abs:** https://arxiv.org/abs/2512.24102 · **PDF:** https://arxiv.org/pdf/2512.24102v1
+
+## 25013. Training a Huggingface Model on AWS Sagemaker (Without Tears)
+
+- **ID:** `2512.24098` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Liling Tan
+- **abs:** https://arxiv.org/abs/2512.24098 · **PDF:** https://arxiv.org/pdf/2512.24098v2
+
+## 25014. Factorized Learning for Temporally Grounded Video-Language Models
+
+- **ID:** `2512.24097` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Wenzheng Zeng, Difei Gao, Mike Zheng Shou, et al.
+- **abs:** https://arxiv.org/abs/2512.24097 · **PDF:** https://arxiv.org/pdf/2512.24097v1
+
+## 25015. Path Integral Solution for Dissipative Generative Dynamics
+
+- **ID:** `2601.00860` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Xidi Wang
+- **abs:** https://arxiv.org/abs/2601.00860 · **PDF:** https://arxiv.org/pdf/2601.00860v1
+
+## 25016. HY-MT1.5 Technical Report
+
+- **ID:** `2512.24092` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Mao Zheng, Zheng Li, Tao Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.24092 · **PDF:** https://arxiv.org/pdf/2512.24092v1
+
+## 25017. FedLiTeCAN : A Federated Lightweight Transformer for Fast and Robust CAN Bus Intrusion Detection
+
+- **ID:** `2512.24088` · **primary:** `cs.CR` · **Submitted:** 2025-12-30
+- **Authors:** Devika S, Pratik Narang, Tejasvi Alladi
+- **abs:** https://arxiv.org/abs/2512.24088 · **PDF:** https://arxiv.org/pdf/2512.24088v1
+
+## 25018. Random Multiplexing
+
+- **ID:** `2512.24087` · **primary:** `cs.IT` · **Submitted:** 2025-12-30
+- **Authors:** Lei Liu, Yuhao Chi, Shunqi Huang, et al.
+- **abs:** https://arxiv.org/abs/2512.24087 · **PDF:** https://arxiv.org/pdf/2512.24087v2
+
+## 25019. LoongFlow: Directed Evolutionary Search via a Cognitive Plan-Execute-Summarize Paradigm
+
+- **ID:** `2512.24077` · **primary:** `cs.AI` · **Submitted:** 2025-12-30
+- **Authors:** Chunhui Wan, Xunan Dai, Zhuo Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.24077 · **PDF:** https://arxiv.org/pdf/2512.24077v1
+
+## 25020. Physics-Guided Temporal Fusion for Lane-Change Intention Prediction
+
+- **ID:** `2512.24075` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Jiazhao Shi, Ziyu Wang, Yichen Lin, et al.
+- **abs:** https://arxiv.org/abs/2512.24075 · **PDF:** https://arxiv.org/pdf/2512.24075v4
+
+## 25021. Time-varying Mixing Matrix Design for Energy-efficient Decentralized Federated Learning
+
+- **ID:** `2512.24069` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Xusheng Zhang, Tuan Nguyen, Ting He
+- **abs:** https://arxiv.org/abs/2512.24069 · **PDF:** https://arxiv.org/pdf/2512.24069v1
+
+## 25022. Pathology Context Recalibration Network for Ocular Disease Recognition
+
+- **ID:** `2512.24066` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Zunjie Xiao, Xiaoqing Zhang, Risa Higashita, et al.
+- **abs:** https://arxiv.org/abs/2512.24066 · **PDF:** https://arxiv.org/pdf/2512.24066v1
+
+## 25023. How and Why LLMs Generalize: A Fine-Grained Analysis of LLM Reasoning from Cognitive Behaviors to Low-Level Patterns
+
+- **ID:** `2512.24063` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Haoyue Bai, Yiyou Sun, Wenjie Hu, et al.
+- **abs:** https://arxiv.org/abs/2512.24063 · **PDF:** https://arxiv.org/pdf/2512.24063v1
+
+## 25024. Hyperspherical Graph Representation Learning via Adaptive Neighbor-Mean Alignment and Uniformity
+
+- **ID:** `2512.24062` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Rui Chen, Junjun Guo, Hongbin Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.24062 · **PDF:** https://arxiv.org/pdf/2512.24062v1
+
+## 25025. Beyond Hallucinations: A Composite Score for Measuring Reliability in Open-Source Large Language Models
+
+- **ID:** `2512.24058` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Rohit Kumar Salla, Manoj Saravanan, Shrikar Reddy Kota
+- **abs:** https://arxiv.org/abs/2512.24058 · **PDF:** https://arxiv.org/pdf/2512.24058v1
+
+## 25026. Policy Mirror Descent with Temporal Difference Learning: Sample Complexity under Online Markov Data
+
+- **ID:** `2512.24056` · **primary:** `math.OC` · **Submitted:** 2025-12-30
+- **Authors:** Wenye Li, Hongxu Chen, Jiacai Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.24056 · **PDF:** https://arxiv.org/pdf/2512.24056v1
+
+## 25027. AHA: Aligning Large Audio-Language Models for Reasoning Hallucinations via Counterfactual Hard Negatives
+
+- **ID:** `2512.24052` · **primary:** `cs.SD` · **Submitted:** 2025-12-30
+- **Authors:** Yanxi Chen, Wenhui Zhu, Xiwen Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.24052 · **PDF:** https://arxiv.org/pdf/2512.24052v2
+
+## 25028. Jailbreaking Attacks vs. Content Safety Filters: How Far Are We in the LLM Safety Arms Race?
+
+- **ID:** `2512.24044` · **primary:** `cs.CR` · **Submitted:** 2025-12-30
+- **Authors:** Yuan Xin, Dingfan Chen, Linyi Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.24044 · **PDF:** https://arxiv.org/pdf/2512.24044v1
+
+## 25029. ROAD: Reflective Optimization via Automated Debugging for Zero-Shot Agent Alignment
+
+- **ID:** `2512.24040` · **primary:** `cs.AI` · **Submitted:** 2025-12-30
+- **Authors:** Natchaya Temyingyong, Daman Jain, Neeraj Kumarsahu, et al.
+- **abs:** https://arxiv.org/abs/2512.24040 · **PDF:** https://arxiv.org/pdf/2512.24040v1
+
+## 25030. Kidney Exchange: Faster Parameterized Algorithms and Tighter Lower Bounds
+
+- **ID:** `2512.24037` · **primary:** `cs.DS` · **Submitted:** 2025-12-30
+- **Authors:** Aritra Banik, Sujoy Bhore, Palash Dey, et al.
+- **abs:** https://arxiv.org/abs/2512.24037 · **PDF:** https://arxiv.org/pdf/2512.24037v1
+
+## 25031. PipeFlow: Pipelined Processing and Motion-Aware Frame Selection for Long-Form Video Editing
+
+- **ID:** `2512.24026` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Mustafa Munir, Md Mostafijur Rahman, Kartikeya Bhardwaj, et al.
+- **abs:** https://arxiv.org/abs/2512.24026 · **PDF:** https://arxiv.org/pdf/2512.24026v1
+
+## 25032. RSAgent: Learning to Reason and Act for Text-Guided Segmentation via Multi-Turn Tool Invocations
+
+- **ID:** `2512.24023` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Xingqi He, Yujie Zhang, Shuyong Gao, et al.
+- **abs:** https://arxiv.org/abs/2512.24023 · **PDF:** https://arxiv.org/pdf/2512.24023v1
+
+## 25033. FUSE-RSVLM: Feature Fusion Vision-Language Model for Remote Sensing
+
+- **ID:** `2512.24022` · **primary:** `cs.CV` · **Submitted:** 2025-12-30
+- **Authors:** Yunkai Dang, Donghao Wang, Jiacheng Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.24022 · **PDF:** https://arxiv.org/pdf/2512.24022v1
+
+## 25034. iCLP: Large Language Model Reasoning with Implicit Cognition Latent Planning
+
+- **ID:** `2512.24014` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Sijia Chen, Di Niu
+- **abs:** https://arxiv.org/abs/2512.24014 · **PDF:** https://arxiv.org/pdf/2512.24014v1
+
+## 25035. SPARK: Search Personalization via Agent-Driven Retrieval and Knowledge-sharing
+
+- **ID:** `2512.24008` · **primary:** `cs.AI` · **Submitted:** 2025-12-30
+- **Authors:** Gaurab Chhetri, Subasish Das, Tausif Islam Chowdhury
+- **abs:** https://arxiv.org/abs/2512.24008 · **PDF:** https://arxiv.org/pdf/2512.24008v2
+
+## 25036. TESO Tabu Enhanced Simulation Optimization for Noisy Black Box Problems
+
+- **ID:** `2512.24007` · **primary:** `cs.NE` · **Submitted:** 2025-12-30
+- **Authors:** Bulent Soykan, Sean Mondesire, Ghaith Rabadi
+- **abs:** https://arxiv.org/abs/2512.24007 · **PDF:** https://arxiv.org/pdf/2512.24007v1
+
+## 25037. Tracing the Heart's Pathways: ECG Representation Learning from a Cardiac Conduction Perspective
+
+- **ID:** `2512.24002` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Tan Pan, Yixuan Sun, Chen Jiang, et al.
+- **abs:** https://arxiv.org/abs/2512.24002 · **PDF:** https://arxiv.org/pdf/2512.24002v1
+
+## 25038. WISE: Web Information Satire and Fakeness Evaluation
+
+- **ID:** `2512.24000` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Gaurab Chhetri, Subasish Das, Tausif Islam Chowdhury
+- **abs:** https://arxiv.org/abs/2512.24000 · **PDF:** https://arxiv.org/pdf/2512.24000v2
+
+## 25039. Higress-RAG: A Holistic Optimization Framework for Enterprise Retrieval-Augmented Generation via Dual Hybrid Retrieval, Adaptive Routing, and CRAG
+
+- **ID:** `2602.23374` · **primary:** `cs.IR` · **Submitted:** 2025-12-30
+- **Authors:** Weixi Lin
+- **abs:** https://arxiv.org/abs/2602.23374 · **PDF:** https://arxiv.org/pdf/2602.23374v1
+
+## 25040. RepetitionCurse: Measuring and Understanding Router Imbalance in Mixture-of-Experts LLMs under DoS Stress
+
+- **ID:** `2512.23995` · **primary:** `cs.CR` · **Submitted:** 2025-12-30
+- **Authors:** Ruixuan Huang, Qingyue Wang, Hantao Huang, et al.
+- **abs:** https://arxiv.org/abs/2512.23995 · **PDF:** https://arxiv.org/pdf/2512.23995v1
+
+## 25041. PhyAVBench: A Challenging Audio Physics-Sensitivity Benchmark for Physically Grounded Text-to-Audio-Video Generation
+
+- **ID:** `2512.23994` · **primary:** `cs.SD` · **Submitted:** 2025-12-30
+- **Authors:** Tianxin Xie, Wentao Lei, Guanjie Huang, et al.
+- **abs:** https://arxiv.org/abs/2512.23994 · **PDF:** https://arxiv.org/pdf/2512.23994v1
+
+## 25042. Fantastic Reasoning Behaviors and Where to Find Them: Unsupervised Discovery of the Reasoning Process
+
+- **ID:** `2512.23988` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Zhenyu Zhang, Shujian Zhang, John Lambert, et al.
+- **abs:** https://arxiv.org/abs/2512.23988 · **PDF:** https://arxiv.org/pdf/2512.23988v1
+
+## 25043. MeLeMaD: Adaptive Malware Detection via Chunk-wise Feature Selection and Meta-Learning
+
+- **ID:** `2512.23987` · **primary:** `cs.CR` · **Submitted:** 2025-12-30
+- **Authors:** Ajvad Haneef K, Karan Kuwar Singh, Madhu Kumar S D
+- **abs:** https://arxiv.org/abs/2512.23987 · **PDF:** https://arxiv.org/pdf/2512.23987v1
+
+## 25044. Coding With AI: From a Reflection on Industrial Practices to Future Computer Science and Software Engineering Education
+
+- **ID:** `2512.23982` · **primary:** `cs.SE` · **Submitted:** 2025-12-30
+- **Authors:** Hung-Fu Chang, MohammadShokrolah Shirazi, Lizhou Cao, et al.
+- **abs:** https://arxiv.org/abs/2512.23982 · **PDF:** https://arxiv.org/pdf/2512.23982v1
+
+## 25045. Information-Theoretic Quality Metric of Low-Dimensional Embeddings
+
+- **ID:** `2512.23981` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Sebastián Gutiérrez-Bernal, Hector Medel Cobaxin, Abiel Galindo González
+- **abs:** https://arxiv.org/abs/2512.23981 · **PDF:** https://arxiv.org/pdf/2512.23981v2
+
+## 25046. Fundamental limits for weighted empirical approximations of tilted distributions
+
+- **ID:** `2512.23979` · **primary:** `math.ST` · **Submitted:** 2025-12-30
+- **Authors:** Sarvesh Ravichandran Iyer, Himadri Mandal, Dhruman Gupta, et al.
+- **abs:** https://arxiv.org/abs/2512.23979 · **PDF:** https://arxiv.org/pdf/2512.23979v1
+
+## 25047. Assured Autonomy: How Operations Research Powers and Orchestrates Generative AI Systems
+
+- **ID:** `2512.23978` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Tinglong Dai, David Simchi-Levi, Michelle Xiao Wu, et al.
+- **abs:** https://arxiv.org/abs/2512.23978 · **PDF:** https://arxiv.org/pdf/2512.23978v1
+
+## 25048. Causify DataFlow: A Framework For High-performance Machine Learning Stream Computing
+
+- **ID:** `2512.23977` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Giacinto Paolo Saggese, Paul Smith
+- **abs:** https://arxiv.org/abs/2512.23977 · **PDF:** https://arxiv.org/pdf/2512.23977v1
+
+## 25049. Exploring the Potential of Spiking Neural Networks in UWB Channel Estimation
+
+- **ID:** `2512.23975` · **primary:** `cs.ET` · **Submitted:** 2025-12-30
+- **Authors:** Youdong Zhang, Xu He, Xiaolin Meng
+- **abs:** https://arxiv.org/abs/2512.23975 · **PDF:** https://arxiv.org/pdf/2512.23975v1
+
+## 25050. A Community-Aware Framework for Influence Maximization with Explicit Accounting for Inter-Community Influence
+
+- **ID:** `2512.23973` · **primary:** `cs.SI` · **Submitted:** 2025-12-30
+- **Authors:** Eliot W. Robson, Abhishek K. Umrawal
+- **abs:** https://arxiv.org/abs/2512.23973 · **PDF:** https://arxiv.org/pdf/2512.23973v2
+
+## 25051. CEC-Zero: Zero-Supervision Character Error Correction with Self-Generated Rewards
+
+- **ID:** `2512.23971` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Zhiming Lin, Kai Zhao, Sophie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.23971 · **PDF:** https://arxiv.org/pdf/2512.23971v1
+
+## 25052. GRADE: Replacing Policy Gradients with Backpropagation for LLM Alignment
+
+- **ID:** `2601.11574` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Lukas Abrie Nel
+- **abs:** https://arxiv.org/abs/2601.11574 · **PDF:** https://arxiv.org/pdf/2601.11574v1
+
+## 25053. Efficient Context Scaling with LongCat ZigZag Attention
+
+- **ID:** `2512.23966` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Chen Zhang, Yang Bai, Jiahuan Li, et al.
+- **abs:** https://arxiv.org/abs/2512.23966 · **PDF:** https://arxiv.org/pdf/2512.23966v2
+
+## 25054. Physics-informed Graph Neural Networks for Operational Flood Modeling
+
+- **ID:** `2512.23964` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Carlo Malapad Acosta, Herath Mudiyanselage Viraj Vidura Herath, Jia Yu Lim, et al.
+- **abs:** https://arxiv.org/abs/2512.23964 · **PDF:** https://arxiv.org/pdf/2512.23964v1
+
+## 25055. An Comparative Analysis about KYC on a Recommendation System Toward Agentic Recommendation System
+
+- **ID:** `2512.23961` · **primary:** `cs.IR` · **Submitted:** 2025-12-30
+- **Authors:** Junjie H. Xu
+- **abs:** https://arxiv.org/abs/2512.23961 · **PDF:** https://arxiv.org/pdf/2512.23961v1
+
+## 25056. Improving Multi-step RAG with Hypergraph-based Memory for Long-Context Complex Relational Modeling
+
+- **ID:** `2512.23959` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Chulun Zhou, Chunkang Zhang, Guoxin Yu, et al.
+- **abs:** https://arxiv.org/abs/2512.23959 · **PDF:** https://arxiv.org/pdf/2512.23959v2
+
+## 25057. Implicit geometric regularization in flow matching via density weighted Stein operators
+
+- **ID:** `2512.23956` · **primary:** `stat.ML` · **Submitted:** 2025-12-30
+- **Authors:** Shinto Eguchi
+- **abs:** https://arxiv.org/abs/2512.23956 · **PDF:** https://arxiv.org/pdf/2512.23956v1
+
+## 25058. DivQAT: Enhancing Robustness of Quantized Convolutional Neural Networks against Model Extraction Attacks
+
+- **ID:** `2512.23948` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Kacem Khaled, Felipe Gohring de Magalhães, Gabriela Nicolescu
+- **abs:** https://arxiv.org/abs/2512.23948 · **PDF:** https://arxiv.org/pdf/2512.23948v1
+
+## 25059. Improved Balanced Classification with Theoretically Grounded Loss Functions
+
+- **ID:** `2512.23947` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Corinna Cortes, Mehryar Mohri, Yutao Zhong
+- **abs:** https://arxiv.org/abs/2512.23947 · **PDF:** https://arxiv.org/pdf/2512.23947v1
+
+## 25060. Decoupling Constraint from Two Direction in Evolutionary Constrained Multi-objective Optimization
+
+- **ID:** `2512.23945` · **primary:** `cs.NE` · **Submitted:** 2025-12-30
+- **Authors:** Ruiqing Sun, Dawei Feng, Xing Zhou, et al.
+- **abs:** https://arxiv.org/abs/2512.23945 · **PDF:** https://arxiv.org/pdf/2512.23945v1
+
+## 25061. Statistical Guarantees in the Search for Less Discriminatory Algorithms
+
+- **ID:** `2512.23943` · **primary:** `cs.CY` · **Submitted:** 2025-12-30
+- **Authors:** Chris Hays, Ben Laufer, Solon Barocas, et al.
+- **abs:** https://arxiv.org/abs/2512.23943 · **PDF:** https://arxiv.org/pdf/2512.23943v1
+
+## 25062. Disentangling Learning from Judgment: Representation Learning for Open Response Analytics
+
+- **ID:** `2512.23941` · **primary:** `cs.CL` · **Submitted:** 2025-12-30
+- **Authors:** Conrad Borchers, Manit Patel, Seiyon M. Lee, et al.
+- **abs:** https://arxiv.org/abs/2512.23941 · **PDF:** https://arxiv.org/pdf/2512.23941v2
+
+## 25063. ProSoftArena: Benchmarking Hierarchical Capabilities of Multimodal Agents in Professional Software Environments
+
+- **ID:** `2601.02399` · **primary:** `cs.SE` · **Submitted:** 2025-12-30
+- **Authors:** Jiaxin Ai, Yukang Feng, Fanrui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.02399 · **PDF:** https://arxiv.org/pdf/2601.02399v1
+
+## 25064. A Proof-of-Concept for Explainable Disease Diagnosis Using Large Language Models and Answer Set Programming
+
+- **ID:** `2512.23932` · **primary:** `cs.AI` · **Submitted:** 2025-12-30
+- **Authors:** Ioanna Gemou, Evangelos Lamprou
+- **abs:** https://arxiv.org/abs/2512.23932 · **PDF:** https://arxiv.org/pdf/2512.23932v1
+
+## 25065. Assessing generative modeling approaches for free energy estimates in condensed matter
+
+- **ID:** `2512.23930` · **primary:** `cond-mat.stat-mech` · **Submitted:** 2025-12-30
+- **Authors:** Maximilian Schebek, Jiajun He, Emil Hoffmann, et al.
+- **abs:** https://arxiv.org/abs/2512.23930 · **PDF:** https://arxiv.org/pdf/2512.23930v2
+
+## 25066. Harvesting AlphaEarth: Benchmarking the Geospatial Foundation Model for Agricultural Downstream Tasks
+
+- **ID:** `2601.00857` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Yuchi Ma, Yawen Shen, Anu Swatantran, et al.
+- **abs:** https://arxiv.org/abs/2601.00857 · **PDF:** https://arxiv.org/pdf/2601.00857v1
+
+## 25067. Stationary Reweighting Yields Local Convergence of Soft Fitted Q-Iteration
+
+- **ID:** `2512.23927` · **primary:** `stat.ML` · **Submitted:** 2025-12-30
+- **Authors:** Lars van der Laan, Nathan Kallus
+- **abs:** https://arxiv.org/abs/2512.23927 · **PDF:** https://arxiv.org/pdf/2512.23927v1
+
+## 25068. Identification of fixations and saccades in eye-tracking data using adaptive threshold-based method
+
+- **ID:** `2512.23926` · **primary:** `cs.NE` · **Submitted:** 2025-12-30
+- **Authors:** Charles Oriioma, Josef Krivan, Rujeena Mathema, et al.
+- **abs:** https://arxiv.org/abs/2512.23926 · **PDF:** https://arxiv.org/pdf/2512.23926v2
+
+## 25069. Interactive Machine Learning: From Theory to Scale
+
+- **ID:** `2512.23924` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Yinglun Zhu
+- **abs:** https://arxiv.org/abs/2512.23924 · **PDF:** https://arxiv.org/pdf/2512.23924v1
+
+## 25070. Tensor Computing Interface: An Application-Oriented, Lightweight Interface for Portable High-Performance Tensor Network Applications
+
+- **ID:** `2512.23917` · **primary:** `quant-ph` · **Submitted:** 2025-12-30
+- **Authors:** Rong-Yang Sun, Tomonori Shirakawa, Hidehiko Kohshiro, et al.
+- **abs:** https://arxiv.org/abs/2512.23917 · **PDF:** https://arxiv.org/pdf/2512.23917v1
+
+## 25071. Constraint Breeds Generalization: Temporal Dynamics as an Inductive Bias
+
+- **ID:** `2512.23916` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Xia Chen
+- **abs:** https://arxiv.org/abs/2512.23916 · **PDF:** https://arxiv.org/pdf/2512.23916v2
+
+## 25072. A multimodal Transformer for InSAR-based ground deformation forecasting with cross-site generalization across Europe
+
+- **ID:** `2512.23906` · **primary:** `eess.SP` · **Submitted:** 2025-12-30
+- **Authors:** Wendong Yao, Binhua Huang, Soumyabrata Dev
+- **abs:** https://arxiv.org/abs/2512.23906 · **PDF:** https://arxiv.org/pdf/2512.23906v1
+
+## 25073. Rethinking Dense Linear Transformations: Stagewise Pairwise Mixing (SPM) for Near-Linear Training in Neural Networks
+
+- **ID:** `2512.23905` · **primary:** `cs.LG` · **Submitted:** 2025-12-30
+- **Authors:** Peter Farag
+- **abs:** https://arxiv.org/abs/2512.23905 · **PDF:** https://arxiv.org/pdf/2512.23905v1
+
+## 25074. Comment on: Your Brain on ChatGPT: Accumulation of Cognitive Debt When Using an AI Assistant for Essay Writing Tasks
+
+- **ID:** `2601.00856` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Milos Stankovic, Ella Hirche, Sarah Kollatzsch, et al.
+- **abs:** https://arxiv.org/abs/2601.00856 · **PDF:** https://arxiv.org/pdf/2601.00856v1
+
+## 25075. Physically-Constrained Autoencoder-Assisted Bayesian Optimization for Refinement of High-Dimensional Defect-Sensitive Single Crystalline Structure
+
+- **ID:** `2601.00855` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2025-12-29
+- **Authors:** Joseph Oche Agada, Andrew McAninch, Haley Day, et al.
+- **abs:** https://arxiv.org/abs/2601.00855 · **PDF:** https://arxiv.org/pdf/2601.00855v1
+
+## 25076. Efficient Deep Learning for Short-Term Solar Irradiance Time Series Forecasting: A Benchmark Study in Ho Chi Minh City
+
+- **ID:** `2512.23898` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Tin Hoang
+- **abs:** https://arxiv.org/abs/2512.23898 · **PDF:** https://arxiv.org/pdf/2512.23898v1
+
+## 25077. How Large Language Models Systematically Misrepresent American Climate Opinions
+
+- **ID:** `2512.23889` · **primary:** `cs.CY` · **Submitted:** 2025-12-29
+- **Authors:** Sola Kim, Jieshu Wang, Marco A. Janssen, et al.
+- **abs:** https://arxiv.org/abs/2512.23889 · **PDF:** https://arxiv.org/pdf/2512.23889v1
+
+## 25078. Autoregressive long-horizon prediction of plasma edge dynamics
+
+- **ID:** `2512.23884` · **primary:** `physics.plasm-ph` · **Submitted:** 2025-12-29
+- **Authors:** Hunor Csala, Sebastian De Pascuale, Paul Laiu, et al.
+- **abs:** https://arxiv.org/abs/2512.23884 · **PDF:** https://arxiv.org/pdf/2512.23884v1
+
+## 25079. Breaking Audio Large Language Models by Attacking Only the Encoder: A Universal Targeted Latent-Space Audio Attack
+
+- **ID:** `2512.23881` · **primary:** `cs.SD` · **Submitted:** 2025-12-29
+- **Authors:** Roee Ziv, Raz Lapid, Moshe Sipper
+- **abs:** https://arxiv.org/abs/2512.23881 · **PDF:** https://arxiv.org/pdf/2512.23881v1
+
+## 25080. CASCADE: Cumulative Agentic Skill Creation through Autonomous Development and Evolution
+
+- **ID:** `2512.23880` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Xu Huang, Junwu Chen, Yuxing Fei, et al.
+- **abs:** https://arxiv.org/abs/2512.23880 · **PDF:** https://arxiv.org/pdf/2512.23880v2
+
+## 25081. Max-Entropy Reinforcement Learning with Flow Matching and A Case Study on LQR
+
+- **ID:** `2512.23870` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Yuyang Zhang, Yang Hu, Bo Dai, et al.
+- **abs:** https://arxiv.org/abs/2512.23870 · **PDF:** https://arxiv.org/pdf/2512.23870v1
+
+## 25082. Probing the Limits of Compressive Memory: A Study of Infini-Attention in Small-Scale Pretraining
+
+- **ID:** `2512.23862` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Ruizhe Huang, Kexuan Zhang, Yihao Fang, et al.
+- **abs:** https://arxiv.org/abs/2512.23862 · **PDF:** https://arxiv.org/pdf/2512.23862v1
+
+## 25083. Lifelong Domain Adaptive 3D Human Pose Estimation
+
+- **ID:** `2512.23860` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Qucheng Peng, Hongfei Xue, Pu Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.23860 · **PDF:** https://arxiv.org/pdf/2512.23860v2
+
+## 25084. Seeking Late Night Life Lines: Experiences of Conversational AI Use in Mental Health Crisis
+
+- **ID:** `2512.23859` · **primary:** `cs.HC` · **Submitted:** 2025-12-29
+- **Authors:** Leah Hope Ajmani, Arka Ghosh, Benjamin Kaveladze, et al.
+- **abs:** https://arxiv.org/abs/2512.23859 · **PDF:** https://arxiv.org/pdf/2512.23859v1
+
+## 25085. Yggdrasil: Bridging Dynamic Speculation and Static Runtime for Latency-Optimal Tree-Based LLM Decoding
+
+- **ID:** `2512.23858` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Yue Guan, Changming Yu, Shihan Fang, et al.
+- **abs:** https://arxiv.org/abs/2512.23858 · **PDF:** https://arxiv.org/pdf/2512.23858v1
+
+## 25086. A Quantum Model for Constrained Markowitz Modern Portfolio Using Slack Variables to Process Mixed-Binary Optimization under QAOA
+
+- **ID:** `2601.03278` · **primary:** `math.OC` · **Submitted:** 2025-12-29
+- **Authors:** Pablo Thomassin, Guillaume Guerard, Sonia Djebali, et al.
+- **abs:** https://arxiv.org/abs/2601.03278 · **PDF:** https://arxiv.org/pdf/2601.03278v1
+
+## 25087. Flow Matching Neural Processes
+
+- **ID:** `2512.23853` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Hussen Abu Hamad, Dan Rosenbaum
+- **abs:** https://arxiv.org/abs/2512.23853 · **PDF:** https://arxiv.org/pdf/2512.23853v1
+
+## 25088. Trellis: Learning to Compress Key-Value Memory in Attention Models
+
+- **ID:** `2512.23852` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Mahdi Karami, Ali Behrouz, Praneeth Kacham, et al.
+- **abs:** https://arxiv.org/abs/2512.23852 · **PDF:** https://arxiv.org/pdf/2512.23852v1
+
+## 25089. The Drill-Down and Fabricate Test (DDFT): A Protocol for Measuring Epistemic Robustness in Language Models
+
+- **ID:** `2512.23850` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Rahul Baxi
+- **abs:** https://arxiv.org/abs/2512.23850 · **PDF:** https://arxiv.org/pdf/2512.23850v1
+
+## 25090. Security Without Detection: Economic Denial as a Primitive for Edge and IoT Defense
+
+- **ID:** `2512.23849` · **primary:** `cs.CR` · **Submitted:** 2025-12-29
+- **Authors:** Samaresh Kumar Singh, Joyjit Roy
+- **abs:** https://arxiv.org/abs/2512.23849 · **PDF:** https://arxiv.org/pdf/2512.23849v1
+
+## 25091. Integrating Domain Knowledge for Financial QA: A Multi-Retriever RAG Approach with LLMs
+
+- **ID:** `2512.23848` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Yukun Zhang, Stefan Elbl Droguett, Samyak Jain
+- **abs:** https://arxiv.org/abs/2512.23848 · **PDF:** https://arxiv.org/pdf/2512.23848v1
+
+## 25092. A Test of Lookahead Bias in LLM Forecasts
+
+- **ID:** `2512.23847` · **primary:** `q-fin.GN` · **Submitted:** 2025-12-29
+- **Authors:** Zhenyu Gao, Wenxi Jiang, Yutong Yan
+- **abs:** https://arxiv.org/abs/2512.23847 · **PDF:** https://arxiv.org/pdf/2512.23847v1
+
+## 25093. From Correctness to Collaboration: Toward a Human-Centered Framework for Evaluating AI Agent Behavior in Software Engineering
+
+- **ID:** `2512.23844` · **primary:** `cs.SE` · **Submitted:** 2025-12-29
+- **Authors:** Tao Dong, Harini Sampath, Ja Young Lee, et al.
+- **abs:** https://arxiv.org/abs/2512.23844 · **PDF:** https://arxiv.org/pdf/2512.23844v1
+
+## 25094. From Clay to Code: Typological and Material Reasoning in AI Interpretations of Iranian Pigeon Towers
+
+- **ID:** `2601.00029` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Abolhassan Pishahang, Maryam Badiei
+- **abs:** https://arxiv.org/abs/2601.00029 · **PDF:** https://arxiv.org/pdf/2601.00029v1
+
+## 25095. Adversarial Lens: Exploiting Attention Layers to Generate Adversarial Examples for Evaluation
+
+- **ID:** `2512.23837` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Kaustubh Dhole
+- **abs:** https://arxiv.org/abs/2512.23837 · **PDF:** https://arxiv.org/pdf/2512.23837v1
+
+## 25096. Retrieval Augmented Question Answering: When Should LLMs Admit Ignorance?
+
+- **ID:** `2512.23836` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Dingmin Wang, Ji Ma, Shankar Kumar
+- **abs:** https://arxiv.org/abs/2512.23836 · **PDF:** https://arxiv.org/pdf/2512.23836v1
+
+## 25097. Explaining News Bias Detection: A Comparative SHAP Analysis of Transformer Model Decision Mechanisms
+
+- **ID:** `2512.23835` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Himel Ghosh
+- **abs:** https://arxiv.org/abs/2512.23835 · **PDF:** https://arxiv.org/pdf/2512.23835v1
+
+## 25098. Artificial Intelligence for All? Brazilian Teachers on Ethics, Equity, and the Everyday Challenges of AI in Education
+
+- **ID:** `2512.23834` · **primary:** `cs.CY` · **Submitted:** 2025-12-29
+- **Authors:** Bruno Florentino, Camila Sestito, Wellington Cruz, et al.
+- **abs:** https://arxiv.org/abs/2512.23834 · **PDF:** https://arxiv.org/pdf/2512.23834v1
+
+## 25099. Exploiting the Prior of Generative Time Series Imputation
+
+- **ID:** `2512.23832` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** YuYang Miao, Chang Li, Zehua Chen
+- **abs:** https://arxiv.org/abs/2512.23832 · **PDF:** https://arxiv.org/pdf/2512.23832v1
+
+## 25100. Deep learning methods for inverse problems using connections between proximal operators and Hamilton-Jacobi equations
+
+- **ID:** `2512.23829` · **primary:** `math.NA` · **Submitted:** 2025-12-29
+- **Authors:** Oluwatosin Akande, Gabriel P. Langlois, Akwum Onwunta
+- **abs:** https://arxiv.org/abs/2512.23829 · **PDF:** https://arxiv.org/pdf/2512.23829v2
+
+## 25101. FedSCAM (Federated Sharpness-Aware Minimization with Clustered Aggregation and Modulation): Scam-resistant SAM for Robust Federated Optimization in Heterogeneous Environments
+
+- **ID:** `2601.00853` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Sameer Rahil, Zain Abdullah Ahmad, Talha Asif
+- **abs:** https://arxiv.org/abs/2601.00853 · **PDF:** https://arxiv.org/pdf/2601.00853v1
+
+## 25102. MS-SSM: A Multi-Scale State Space Model for Efficient Sequence Modeling
+
+- **ID:** `2512.23824` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Mahdi Karami, Ali Behrouz, Peilin Zhong, et al.
+- **abs:** https://arxiv.org/abs/2512.23824 · **PDF:** https://arxiv.org/pdf/2512.23824v1
+
+## 25103. An Agentic LLM Framework for Adverse Media Screening in AML Compliance
+
+- **ID:** `2602.23373` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Pavel Chernakov, Sasan Jafarnejad, Raphaël Frank
+- **abs:** https://arxiv.org/abs/2602.23373 · **PDF:** https://arxiv.org/pdf/2602.23373v1
+
+## 25104. Video-Based Performance Evaluation for ECR Drills in Synthetic Training Environments
+
+- **ID:** `2512.23819` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Surya Rayala, Marcos Quinones-Grueiro, Naveeduddin Mohammed, et al.
+- **abs:** https://arxiv.org/abs/2512.23819 · **PDF:** https://arxiv.org/pdf/2512.23819v1
+
+## 25105. Energy-Tweedie: Score meets Score, Energy meets Energy
+
+- **ID:** `2512.23818` · **primary:** `stat.ML` · **Submitted:** 2025-12-29
+- **Authors:** Andrej Leban
+- **abs:** https://arxiv.org/abs/2512.23818 · **PDF:** https://arxiv.org/pdf/2512.23818v1
+
+## 25106. Quantum Error Mitigation with Attention Graph Transformers for Burgers Equation Solvers on NISQ Hardware
+
+- **ID:** `2512.23817` · **primary:** `quant-ph` · **Submitted:** 2025-12-29
+- **Authors:** Seyed Mohamad Ali Tousi, Adib Bazgir, Yuwen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.23817 · **PDF:** https://arxiv.org/pdf/2512.23817v1
+
+## 25107. Improved Bounds for Private and Robust Alignment
+
+- **ID:** `2512.23816` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Wenqian Weng, Yi He, Xingyu Zhou
+- **abs:** https://arxiv.org/abs/2512.23816 · **PDF:** https://arxiv.org/pdf/2512.23816v1
+
+## 25108. One if by Land, Two if by Sea, Three if by Four Seas, and More to Come -- Values of Perception, Prediction, Communication, and Common Sense in Decision Making
+
+- **ID:** `2601.06077` · **primary:** `cs.IT` · **Submitted:** 2025-12-29
+- **Authors:** Aolin Xu
+- **abs:** https://arxiv.org/abs/2601.06077 · **PDF:** https://arxiv.org/pdf/2601.06077v1
+
+## 25109. StressRoBERTa: Cross-Condition Transfer Learning from Depression, Anxiety, and PTSD to Stress Detection
+
+- **ID:** `2512.23813` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Amal Alqahtani, Efsun Kayi, Mona Diab
+- **abs:** https://arxiv.org/abs/2512.23813 · **PDF:** https://arxiv.org/pdf/2512.23813v1
+
+## 25110. An Empirical Analysis of Fine-Tuning Large Language Models on Bioinformatics Literature: PRSGPT and BioStarsGPT
+
+- **ID:** `2601.11573` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Muhammad Muneeb, David B. Ascher
+- **abs:** https://arxiv.org/abs/2601.11573 · **PDF:** https://arxiv.org/pdf/2601.11573v1
+
+## 25111. Zero-Trust Agentic Federated Learning for Secure IIoT Defense Systems
+
+- **ID:** `2512.23809` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Samaresh Kumar Singh, Joyjit Roy, Martin So
+- **abs:** https://arxiv.org/abs/2512.23809 · **PDF:** https://arxiv.org/pdf/2512.23809v1
+
+## 25112. MiMo-Audio: Audio Language Models are Few-Shot Learners
+
+- **ID:** `2512.23808` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Xiaomi LLM-Core Team, :, Dong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.23808 · **PDF:** https://arxiv.org/pdf/2512.23808v1
+
+## 25113. Fitted Q Evaluation Without Bellman Completeness via Stationary Weighting
+
+- **ID:** `2512.23805` · **primary:** `stat.ML` · **Submitted:** 2025-12-29
+- **Authors:** Lars van der Laan, Nathan Kallus
+- **abs:** https://arxiv.org/abs/2512.23805 · **PDF:** https://arxiv.org/pdf/2512.23805v1
+
+## 25114. Training AI Co-Scientists Using Rubric Rewards
+
+- **ID:** `2512.23707` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Shashwat Goel, Rishi Hazra, Dulhan Jayalath, et al.
+- **abs:** https://arxiv.org/abs/2512.23707 · **PDF:** https://arxiv.org/pdf/2512.23707v1
+
+## 25115. Eliciting Behaviors in Multi-Turn Conversations
+
+- **ID:** `2512.23701` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Jing Huang, Shujian Zhang, Lun Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.23701 · **PDF:** https://arxiv.org/pdf/2512.23701v1
+
+## 25116. Bellman Calibration for V-Learning in Offline Reinforcement Learning
+
+- **ID:** `2512.23694` · **primary:** `stat.ML` · **Submitted:** 2025-12-29
+- **Authors:** Lars van der Laan, Nathan Kallus
+- **abs:** https://arxiv.org/abs/2512.23694 · **PDF:** https://arxiv.org/pdf/2512.23694v1
+
+## 25117. Fine-Tuning LLMs with Fine-Grained Human Feedback on Text Spans
+
+- **ID:** `2512.23693` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Sky CH-Wang, Justin Svegliato, Helen Appel, et al.
+- **abs:** https://arxiv.org/abs/2512.23693 · **PDF:** https://arxiv.org/pdf/2512.23693v1
+
+## 25118. PROFASR-BENCH: A Benchmark for Context-Conditioned ASR in High-Stakes Professional Speech
+
+- **ID:** `2512.23686` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Deepak Babu Piskala
+- **abs:** https://arxiv.org/abs/2512.23686 · **PDF:** https://arxiv.org/pdf/2512.23686v1
+
+## 25119. Multilingual Hidden Prompt Injection Attacks on LLM-Based Academic Reviewing
+
+- **ID:** `2512.23684` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Panagiotis Theocharopoulos, Ajinkya Kulkarni, Mathew Magimai. -Doss
+- **abs:** https://arxiv.org/abs/2512.23684 · **PDF:** https://arxiv.org/pdf/2512.23684v1
+
+## 25120. Web World Models
+
+- **ID:** `2512.23676` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Jichen Feng, Yifan Zhang, Chenggong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.23676 · **PDF:** https://arxiv.org/pdf/2512.23676v1
+
+## 25121. End-to-End Test-Time Training for Long Context
+
+- **ID:** `2512.23675` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Arnuv Tandon, Karan Dalal, Xinhao Li, et al.
+- **abs:** https://arxiv.org/abs/2512.23675 · **PDF:** https://arxiv.org/pdf/2512.23675v2
+
+## 25122. Calibrated Multi-Level Quantile Forecasting
+
+- **ID:** `2512.23671` · **primary:** `stat.ML` · **Submitted:** 2025-12-29
+- **Authors:** Tiffany Ding, Isaac Gibbs, Ryan J. Tibshirani
+- **abs:** https://arxiv.org/abs/2512.23671 · **PDF:** https://arxiv.org/pdf/2512.23671v2
+
+## 25123. Random Controlled Differential Equations
+
+- **ID:** `2512.23670` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Francesco Piatti, Thomas Cass, William F. Turner
+- **abs:** https://arxiv.org/abs/2512.23670 · **PDF:** https://arxiv.org/pdf/2512.23670v1
+
+## 25124. Althea: Human-AI Collaboration for Fact-Checking and Critical Reasoning
+
+- **ID:** `2602.11161` · **primary:** `cs.HC` · **Submitted:** 2025-12-29
+- **Authors:** Svetlana Churina, Kokil Jaidka, Anab Maulana Barik, et al.
+- **abs:** https://arxiv.org/abs/2602.11161 · **PDF:** https://arxiv.org/pdf/2602.11161v1
+
+## 25125. Less is more: Probabilistic reduction is best explained by small-scale predictability measures
+
+- **ID:** `2512.23659` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Cassandra L. Jacobs, Andrés Buxó-Lugo, Anna K. Taylor, et al.
+- **abs:** https://arxiv.org/abs/2512.23659 · **PDF:** https://arxiv.org/pdf/2512.23659v2
+
+## 25126. Autonomous battery research: Principles of heuristic operando experimentation
+
+- **ID:** `2601.00851` · **primary:** `physics.ins-det` · **Submitted:** 2025-12-29
+- **Authors:** Emily Lu, Gabriel Perez, Peter Baker, et al.
+- **abs:** https://arxiv.org/abs/2601.00851 · **PDF:** https://arxiv.org/pdf/2601.00851v1
+
+## 25127. Nested Browser-Use Learning for Agentic Information Seeking
+
+- **ID:** `2512.23647` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Baixuan Li, Jialong Wu, Wenbiao Yin, et al.
+- **abs:** https://arxiv.org/abs/2512.23647 · **PDF:** https://arxiv.org/pdf/2512.23647v1
+
+## 25128. Simultaneous Approximation of the Score Function and Its Derivatives by Deep Neural Networks
+
+- **ID:** `2512.23643` · **primary:** `math.NA` · **Submitted:** 2025-12-29
+- **Authors:** Konstantin Yakovlev, Nikita Puchkin
+- **abs:** https://arxiv.org/abs/2512.23643 · **PDF:** https://arxiv.org/pdf/2512.23643v1
+
+## 25129. A Dataset and Benchmark for Consumer Healthcare Question Summarization
+
+- **ID:** `2512.23637` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Abhishek Basu, Deepak Gupta, Dina Demner-Fushman, et al.
+- **abs:** https://arxiv.org/abs/2512.23637 · **PDF:** https://arxiv.org/pdf/2512.23637v1
+
+## 25130. TabMixNN: A Unified Deep Learning Framework for Structural Mixed Effects Modeling on Tabular Data
+
+- **ID:** `2512.23787` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Deniz Akdemir
+- **abs:** https://arxiv.org/abs/2512.23787 · **PDF:** https://arxiv.org/pdf/2512.23787v1
+
+## 25131. AI tutoring can safely and effectively support students: An exploratory RCT in UK classrooms
+
+- **ID:** `2512.23633` · **primary:** `cs.CY` · **Submitted:** 2025-12-29
+- **Authors:** LearnLM Team, Eedi, :, et al.
+- **abs:** https://arxiv.org/abs/2512.23633 · **PDF:** https://arxiv.org/pdf/2512.23633v1
+
+## 25132. BOAD: Discovering Hierarchical Software Engineering Agents via Bandit Optimization
+
+- **ID:** `2512.23631` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Iris Xu, Guangtao Zeng, Zexue He, et al.
+- **abs:** https://arxiv.org/abs/2512.23631 · **PDF:** https://arxiv.org/pdf/2512.23631v2
+
+## 25133. Memorization in 3D Shape Generation: An Empirical Study
+
+- **ID:** `2512.23628` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Shu Pu, Boya Zeng, Kaichen Zhou, et al.
+- **abs:** https://arxiv.org/abs/2512.23628 · **PDF:** https://arxiv.org/pdf/2512.23628v1
+
+## 25134. Regret-Based Federated Causal Discovery with Unknown Interventions
+
+- **ID:** `2512.23626` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Federico Baldo, Charles K. Assaad
+- **abs:** https://arxiv.org/abs/2512.23626 · **PDF:** https://arxiv.org/pdf/2512.23626v1
+
+## 25135. Physics-Informed Neural Networks for Device and Circuit Modeling: A Case Study of NeuroSPICE
+
+- **ID:** `2512.23624` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Chien-Ting Tung, Chenming Hu
+- **abs:** https://arxiv.org/abs/2512.23624 · **PDF:** https://arxiv.org/pdf/2512.23624v2
+
+## 25136. Le Cam Distortion: A Decision-Theoretic Framework for Robust Transfer Learning
+
+- **ID:** `2512.23617` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Deniz Akdemir
+- **abs:** https://arxiv.org/abs/2512.23617 · **PDF:** https://arxiv.org/pdf/2512.23617v1
+
+## 25137. Close the Loop: Synthesizing Infinite Tool-Use Data via Multi-Agent Role-Playing
+
+- **ID:** `2512.23611` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Yuwen Li, Wei Zhang, Zelong Huang, et al.
+- **abs:** https://arxiv.org/abs/2512.23611 · **PDF:** https://arxiv.org/pdf/2512.23611v1
+
+## 25138. Marriage Discourse on Chinese Social Media: An LLM-assisted Analysis
+
+- **ID:** `2512.23609` · **primary:** `econ.GN` · **Submitted:** 2025-12-29
+- **Authors:** Frank Tian-Fang Ye, Xiaozi Gao
+- **abs:** https://arxiv.org/abs/2512.23609 · **PDF:** https://arxiv.org/pdf/2512.23609v2
+
+## 25139. Distribution-Free Process Monitoring with Conformal Prediction
+
+- **ID:** `2512.23602` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Christopher Burger
+- **abs:** https://arxiv.org/abs/2512.23602 · **PDF:** https://arxiv.org/pdf/2512.23602v1
+
+## 25140. Divergent-Convergent Thinking in Large Language Models for Creative Problem Generation
+
+- **ID:** `2512.23601` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Manh Hung Nguyen, Adish Singla
+- **abs:** https://arxiv.org/abs/2512.23601 · **PDF:** https://arxiv.org/pdf/2512.23601v1
+
+## 25141. The Nonstationarity-Complexity Tradeoff in Return Prediction
+
+- **ID:** `2512.23596` · **primary:** `stat.ML` · **Submitted:** 2025-12-29
+- **Authors:** Agostino Capponi, Chengpiao Huang, J. Antonio Sidaoui, et al.
+- **abs:** https://arxiv.org/abs/2512.23596 · **PDF:** https://arxiv.org/pdf/2512.23596v1
+
+## 25142. Style Amnesia: Investigating Speaking Style Degradation and Mitigation in Multi-Turn Spoken Language Models
+
+- **ID:** `2512.23578` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Yu-Xiang Lin, Cheng-Han Chiang, Hung-yi Lee
+- **abs:** https://arxiv.org/abs/2512.23578 · **PDF:** https://arxiv.org/pdf/2512.23578v2
+
+## 25143. Instruction-Following Evaluation of Large Vision-Language Models
+
+- **ID:** `2512.23572` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Daiki Shiono, Shumpei Miyawaki, Ryota Tanaka, et al.
+- **abs:** https://arxiv.org/abs/2512.23572 · **PDF:** https://arxiv.org/pdf/2512.23572v1
+
+## 25144. From geometry to dynamics: Learning overdamped Langevin dynamics from sparse observations with geometric constraints
+
+- **ID:** `2512.23566` · **primary:** `math.DS` · **Submitted:** 2025-12-29
+- **Authors:** Dimitra Maoutsa
+- **abs:** https://arxiv.org/abs/2512.23566 · **PDF:** https://arxiv.org/pdf/2512.23566v1
+
+## 25145. RxnBench: A Multimodal Benchmark for Evaluating Large Language Models on Chemical Reaction Understanding from Scientific Literature
+
+- **ID:** `2512.23565` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Hanzheng Li, Xi Fang, Yixuan Li, et al.
+- **abs:** https://arxiv.org/abs/2512.23565 · **PDF:** https://arxiv.org/pdf/2512.23565v5
+
+## 25146. VL-RouterBench: A Benchmark for Vision-Language Model Routing
+
+- **ID:** `2512.23562` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Zhehao Huang, Baijiong Lin, Jingyuan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.23562 · **PDF:** https://arxiv.org/pdf/2512.23562v2
+
+## 25147. Toward Trustworthy Agentic AI: A Multimodal Framework for Preventing Prompt Injection Attacks
+
+- **ID:** `2512.23557` · **primary:** `cs.CR` · **Submitted:** 2025-12-29
+- **Authors:** Toqeer Ali Syed, Mishal Ateeq Almutairi, Mahmoud Abdel Moaty
+- **abs:** https://arxiv.org/abs/2512.23557 · **PDF:** https://arxiv.org/pdf/2512.23557v1
+
+## 25148. Lie to Me: Knowledge Graphs for Robust Hallucination Self-Detection in LLMs
+
+- **ID:** `2512.23547` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Sahil Kale, Antonio Luca Alfeo
+- **abs:** https://arxiv.org/abs/2512.23547 · **PDF:** https://arxiv.org/pdf/2512.23547v1
+
+## 25149. PathFound: An Agentic Multimodal Model Activating Evidence-seeking Pathological Diagnosis
+
+- **ID:** `2512.23545` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Shengyi Hua, Jianfeng Wu, Tianle Shen, et al.
+- **abs:** https://arxiv.org/abs/2512.23545 · **PDF:** https://arxiv.org/pdf/2512.23545v1
+
+## 25150. Act2Goal: From World Model To General Goal-conditioned Policy
+
+- **ID:** `2512.23541` · **primary:** `cs.RO` · **Submitted:** 2025-12-29
+- **Authors:** Pengfei Zhou, Liliang Chen, Shengcong Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.23541 · **PDF:** https://arxiv.org/pdf/2512.23541v1
+
+## 25151. AnyMS: Bottom-up Attention Decoupling for Layout-guided and Training-free Multi-subject Customization
+
+- **ID:** `2512.23537` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Binhe Yu, Zhen Wang, Kexin Li, et al.
+- **abs:** https://arxiv.org/abs/2512.23537 · **PDF:** https://arxiv.org/pdf/2512.23537v2
+
+## 25152. Jamming Detection in Cell-Free MIMO with Dynamic Graphs
+
+- **ID:** `2601.06075` · **primary:** `cs.IT` · **Submitted:** 2025-12-29
+- **Authors:** Ali Hossary, Laura Crosara, Stefano Tomasin
+- **abs:** https://arxiv.org/abs/2601.06075 · **PDF:** https://arxiv.org/pdf/2601.06075v1
+
+## 25153. EEG-based Graph-guided Domain Adaptation for Robust Cross-Session Emotion Recognition
+
+- **ID:** `2512.23526` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Maryam Mirzaei, Farzaneh Shayegh, Hamed Narimani
+- **abs:** https://arxiv.org/abs/2512.23526 · **PDF:** https://arxiv.org/pdf/2512.23526v2
+
+## 25154. Trustworthy Machine Learning under Distribution Shifts
+
+- **ID:** `2512.23524` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Zhuo Huang
+- **abs:** https://arxiv.org/abs/2512.23524 · **PDF:** https://arxiv.org/pdf/2512.23524v1
+
+## 25155. Discrete Semantic States and Hamiltonian Dynamics in LLM Embedding Spaces
+
+- **ID:** `2601.11572` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Timo Aukusti Laine
+- **abs:** https://arxiv.org/abs/2601.11572 · **PDF:** https://arxiv.org/pdf/2601.11572v1
+
+## 25156. Single LLM Debate, MoLaCE: Mixture of Latent Concept Experts Against Confirmation Bias
+
+- **ID:** `2512.23518` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Hazel Kim, Philip Torr
+- **abs:** https://arxiv.org/abs/2512.23518 · **PDF:** https://arxiv.org/pdf/2512.23518v1
+
+## 25157. Leveraging Adaptive Group Negotiation for Heterogeneous Multi-Robot Collaboration with Large Language Models
+
+- **ID:** `2602.06967` · **primary:** `cs.RO` · **Submitted:** 2025-12-29
+- **Authors:** Siqi Song, Xuanbing Xie, Zonglin Li, et al.
+- **abs:** https://arxiv.org/abs/2602.06967 · **PDF:** https://arxiv.org/pdf/2602.06967v1
+
+## 25158. Alpha-R1: Alpha Screening with LLM Reasoning via Reinforcement Learning
+
+- **ID:** `2512.23515` · **primary:** `q-fin.TR` · **Submitted:** 2025-12-29
+- **Authors:** Zuoyou Jiang, Li Zhao, Rui Sun, et al.
+- **abs:** https://arxiv.org/abs/2512.23515 · **PDF:** https://arxiv.org/pdf/2512.23515v1
+
+## 25159. UniHetero: Could Generation Enhance Understanding for Vision-Language-Model at Large Data Scale?
+
+- **ID:** `2512.23512` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Fengjiao Chen, Minhao Jing, Weitao Lu, et al.
+- **abs:** https://arxiv.org/abs/2512.23512 · **PDF:** https://arxiv.org/pdf/2512.23512v2
+
+## 25160. EdgeJury: Cross-Reviewed Small-Model Ensembles for Truthful Question Answering on Serverless Edge Inference
+
+- **ID:** `2601.00850` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Aayush Kumar
+- **abs:** https://arxiv.org/abs/2601.00850 · **PDF:** https://arxiv.org/pdf/2601.00850v1
+
+## 25161. Why AI Safety Requires Uncertainty, Incomplete Preferences, and Non-Archimedean Utilities
+
+- **ID:** `2512.23508` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Alessio Benavoli, Alessandro Facchini, Marco Zaffalon
+- **abs:** https://arxiv.org/abs/2512.23508 · **PDF:** https://arxiv.org/pdf/2512.23508v1
+
+## 25162. Automatic Detection of Complex Quotation Patterns in Aggadic Literature
+
+- **ID:** `2512.23504` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Hadar Miller, Tsvi Kuflik, Moshe Lavee
+- **abs:** https://arxiv.org/abs/2512.23504 · **PDF:** https://arxiv.org/pdf/2512.23504v1
+
+## 25163. Benchmarking Cross-Lingual Semantic Alignment in Multilingual Embeddings
+
+- **ID:** `2601.09732` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Wen G. Gong
+- **abs:** https://arxiv.org/abs/2601.09732 · **PDF:** https://arxiv.org/pdf/2601.09732v1
+
+## 25164. Joint Link Adaptation and Device Scheduling Approach for URLLC Industrial IoT Network: A DRL-based Method with Bayesian Optimization
+
+- **ID:** `2512.23493` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Wei Gao, Paul Zheng, Peng Wu, et al.
+- **abs:** https://arxiv.org/abs/2512.23493 · **PDF:** https://arxiv.org/pdf/2512.23493v1
+
+## 25165. The Gaining Paths to Investment Success: Information-Driven LLM Graph Reasoning for Venture Capital Prediction
+
+- **ID:** `2512.23489` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Haoyu Pei, Zhongyang Liu, Xiangyi Xiao, et al.
+- **abs:** https://arxiv.org/abs/2512.23489 · **PDF:** https://arxiv.org/pdf/2512.23489v2
+
+## 25166. ML Compass: Navigating Capability, Cost, and Compliance Trade-offs in AI Model Deployment
+
+- **ID:** `2512.23487` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Vassilis Digalakis, Ramayya Krishnan, Gonzalo Martin Fernandez, et al.
+- **abs:** https://arxiv.org/abs/2512.23487 · **PDF:** https://arxiv.org/pdf/2512.23487v1
+
+## 25167. FRoD: Full-Rank Efficient Fine-Tuning with Rotational Degrees for Fast Convergence
+
+- **ID:** `2512.23485` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Guoan Wan, Tianyu Chen, Fangzheng Feng, et al.
+- **abs:** https://arxiv.org/abs/2512.23485 · **PDF:** https://arxiv.org/pdf/2512.23485v1
+
+## 25168. Theory of Mind for Explainable Human-Robot Interaction
+
+- **ID:** `2512.23482` · **primary:** `cs.RO` · **Submitted:** 2025-12-29
+- **Authors:** Marie S. Bauer, Julia Gachot, Matthias Kerzel, et al.
+- **abs:** https://arxiv.org/abs/2512.23482 · **PDF:** https://arxiv.org/pdf/2512.23482v3
+
+## 25169. Agentic AI for Autonomous Defense in Software Supply Chain Security: Beyond Provenance to Vulnerability Mitigation
+
+- **ID:** `2512.23480` · **primary:** `cs.CR` · **Submitted:** 2025-12-29
+- **Authors:** Toqeer Ali Syed, Mohammad Riyaz Belgaum, Salman Jan, et al.
+- **abs:** https://arxiv.org/abs/2512.23480 · **PDF:** https://arxiv.org/pdf/2512.23480v1
+
+## 25170. Geometric Patterns of Meaning: A PHATE Manifold Analysis of Multi-lingual Embeddings
+
+- **ID:** `2601.09731` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Wen G Gong
+- **abs:** https://arxiv.org/abs/2601.09731 · **PDF:** https://arxiv.org/pdf/2601.09731v1
+
+## 25171. Deep classifier kriging for probabilistic spatial prediction of air quality index
+
+- **ID:** `2512.23474` · **primary:** `stat.ME` · **Submitted:** 2025-12-29
+- **Authors:** Junyu Chen, Pratik Nag, Huixia Judy-Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.23474 · **PDF:** https://arxiv.org/pdf/2512.23474v1
+
+## 25172. Discovering Multi-Scale Semantic Structure in Text Corpora Using Density-Based Trees and LLM Embeddings
+
+- **ID:** `2512.23471` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Thomas Haschka, Joseph Bakarji
+- **abs:** https://arxiv.org/abs/2512.23471 · **PDF:** https://arxiv.org/pdf/2512.23471v2
+
+## 25173. HY-Motion 1.0: Scaling Flow Matching Models for Text-To-Motion Generation
+
+- **ID:** `2512.23464` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Yuxin Wen, Qing Shuai, Di Kang, et al.
+- **abs:** https://arxiv.org/abs/2512.23464 · **PDF:** https://arxiv.org/pdf/2512.23464v1
+
+## 25174. Prompt-Induced Over-Generation as Denial-of-Service: A Black-Box Attack-Side Benchmark
+
+- **ID:** `2512.23779` · **primary:** `cs.CR` · **Submitted:** 2025-12-29
+- **Authors:** Manu, Yi Guo, Kanchana Thilakarathna, et al.
+- **abs:** https://arxiv.org/abs/2512.23779 · **PDF:** https://arxiv.org/pdf/2512.23779v2
+
+## 25175. Eliminating Inductive Bias in Reward Models with Information-Theoretic Guidance
+
+- **ID:** `2512.23461` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Zhuo Li, Pengyu Cheng, Zhechao Yu, et al.
+- **abs:** https://arxiv.org/abs/2512.23461 · **PDF:** https://arxiv.org/pdf/2512.23461v1
+
+## 25176. Replay Failures as Successes: Sample-Efficient Reinforcement Learning for Instruction Following
+
+- **ID:** `2512.23457` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Kongcheng Zhang, Qi Yao, Shunyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.23457 · **PDF:** https://arxiv.org/pdf/2512.23457v1
+
+## 25177. A Survey on Graph Neural Networks for Fraud Detection in Ride Hailing Platforms
+
+- **ID:** `2512.23777` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Kanishka Hewageegana, Janani Harischandra, Nipuna Senanayake, et al.
+- **abs:** https://arxiv.org/abs/2512.23777 · **PDF:** https://arxiv.org/pdf/2512.23777v1
+
+## 25178. CoFi-Dec: Hallucination-Resistant Decoding via Coarse-to-Fine Generative Feedback in Large Vision-Language Models
+
+- **ID:** `2512.23453` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Zongsheng Cao, Yangfan He, Anran Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.23453 · **PDF:** https://arxiv.org/pdf/2512.23453v1
+
+## 25179. Dynamic Subspace Composition: Efficient Adaptation via Contractive Basis Expansion
+
+- **ID:** `2512.23448` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Vladimer Khasia
+- **abs:** https://arxiv.org/abs/2512.23448 · **PDF:** https://arxiv.org/pdf/2512.23448v1
+
+## 25180. Coupling Experts and Routers in Mixture-of-Experts via an Auxiliary Loss
+
+- **ID:** `2512.23447` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Ang Lv, Jin Ma, Yiyuan Ma, et al.
+- **abs:** https://arxiv.org/abs/2512.23447 · **PDF:** https://arxiv.org/pdf/2512.23447v2
+
+## 25181. Assessing behaviour coverage in a multi-agent system simulation for autonomous vehicle testing
+
+- **ID:** `2512.23445` · **primary:** `cs.MA` · **Submitted:** 2025-12-29
+- **Authors:** Manuel Franco-Vivo
+- **abs:** https://arxiv.org/abs/2512.23445 · **PDF:** https://arxiv.org/pdf/2512.23445v1
+
+## 25182. Adaptive Fusion Graph Network for 3D Strain Field Prediction in Solid Rocket Motor Grains
+
+- **ID:** `2512.23443` · **primary:** `physics.app-ph` · **Submitted:** 2025-12-29
+- **Authors:** Jiada Huang, Hao Ma, Zhibin Shen, et al.
+- **abs:** https://arxiv.org/abs/2512.23443 · **PDF:** https://arxiv.org/pdf/2512.23443v1
+
+## 25183. Stochastic Siamese MAE Pretraining for Longitudinal Medical Images
+
+- **ID:** `2512.23441` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Taha Emre, Arunava Chakravarty, Thomas Pinetz, et al.
+- **abs:** https://arxiv.org/abs/2512.23441 · **PDF:** https://arxiv.org/pdf/2512.23441v1
+
+## 25184. ClinDEF: A Dynamic Evaluation Framework for Large Language Models in Clinical Reasoning
+
+- **ID:** `2512.23440` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Yuqi Tang, Jing Yu, Zichang Su, et al.
+- **abs:** https://arxiv.org/abs/2512.23440 · **PDF:** https://arxiv.org/pdf/2512.23440v1
+
+## 25185. Fuzzy-Logic and Deep Learning for Environmental Condition-Aware Road Surface Classification
+
+- **ID:** `2512.23436` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Mustafa Demetgul, Sanja Lazarova Molnar
+- **abs:** https://arxiv.org/abs/2512.23436 · **PDF:** https://arxiv.org/pdf/2512.23436v2
+
+## 25186. Distilled HuBERT for Mobile Speech Emotion Recognition: A Cross-Corpus Validation Study
+
+- **ID:** `2512.23435` · **primary:** `cs.SD` · **Submitted:** 2025-12-29
+- **Authors:** Saifelden M. Ismail
+- **abs:** https://arxiv.org/abs/2512.23435 · **PDF:** https://arxiv.org/pdf/2512.23435v2
+
+## 25187. C2PO: Diagnosing and Disentangling Bias Shortcuts in LLMs
+
+- **ID:** `2512.23430` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Xuan Feng, Bo An, Tianlong Gu, et al.
+- **abs:** https://arxiv.org/abs/2512.23430 · **PDF:** https://arxiv.org/pdf/2512.23430v1
+
+## 25188. The Effect of Gender Diversity on Scientific Team Impact: A Team Roles Perspective
+
+- **ID:** `2512.23429` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Yi Zhao, Yongjun Zhu, Donghun Kim, et al.
+- **abs:** https://arxiv.org/abs/2512.23429 · **PDF:** https://arxiv.org/pdf/2512.23429v1
+
+## 25189. Towards Integrating Uncertainty for Domain-Agnostic Segmentation
+
+- **ID:** `2512.23427` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Jesse Brouwers, Xiaoyan Xing, Alexander Timans
+- **abs:** https://arxiv.org/abs/2512.23427 · **PDF:** https://arxiv.org/pdf/2512.23427v1
+
+## 25190. A general framework for deep learning
+
+- **ID:** `2512.23425` · **primary:** `math.ST` · **Submitted:** 2025-12-29
+- **Authors:** William Kengne, Modou Wade
+- **abs:** https://arxiv.org/abs/2512.23425 · **PDF:** https://arxiv.org/pdf/2512.23425v1
+
+## 25191. AKG kernel Agent: A Multi-Agent Framework for Cross-Platform Kernel Synthesis
+
+- **ID:** `2512.23424` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Jinye Du, Quan Yuan, Zuyao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.23424 · **PDF:** https://arxiv.org/pdf/2512.23424v1
+
+## 25192. Entropy-Guided Token Dropout: Training Autoregressive Language Models with Limited Domain Data
+
+- **ID:** `2512.23422` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Jiapeng Wang, Yiwen Hu, Yanzipeng Gao, et al.
+- **abs:** https://arxiv.org/abs/2512.23422 · **PDF:** https://arxiv.org/pdf/2512.23422v1
+
+## 25193. The World Is Bigger! A Computationally-Embedded Perspective on the Big World Hypothesis
+
+- **ID:** `2512.23419` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Alex Lewandowski, Adtiya A. Ramesh, Edan Meyer, et al.
+- **abs:** https://arxiv.org/abs/2512.23419 · **PDF:** https://arxiv.org/pdf/2512.23419v1
+
+## 25194. MindWatcher: Toward Smarter Multimodal Tool-Integrated Reasoning
+
+- **ID:** `2512.23412` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Jiawei Chen, Xintian Shen, Lihao Zheng, et al.
+- **abs:** https://arxiv.org/abs/2512.23412 · **PDF:** https://arxiv.org/pdf/2512.23412v2
+
+## 25195. High-Dimensional Search, Low-Dimensional Solution: Decoupling Optimization from Representation
+
+- **ID:** `2512.23410` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Yusuf Kalyoncuoglu, Ratmir Miftachov
+- **abs:** https://arxiv.org/abs/2512.23410 · **PDF:** https://arxiv.org/pdf/2512.23410v2
+
+## 25196. Probabilistic Modelling is Sufficient for Causal Inference
+
+- **ID:** `2512.23408` · **primary:** `stat.ML` · **Submitted:** 2025-12-29
+- **Authors:** Bruno Mlodozeniec, David Krueger, Richard E. Turner
+- **abs:** https://arxiv.org/abs/2512.23408 · **PDF:** https://arxiv.org/pdf/2512.23408v1
+
+## 25197. Theoretical Foundations of Scaling Law in Familial Models
+
+- **ID:** `2512.23407` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Huan Song, Qingfei Zhao, Ting Long, et al.
+- **abs:** https://arxiv.org/abs/2512.23407 · **PDF:** https://arxiv.org/pdf/2512.23407v2
+
+## 25198. Task-driven Heterophilic Graph Structure Learning
+
+- **ID:** `2512.23406` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Ayushman Raghuvanshi, Gonzalo Mateos, Sundeep Prabhakar Chepuri
+- **abs:** https://arxiv.org/abs/2512.23406 · **PDF:** https://arxiv.org/pdf/2512.23406v1
+
+## 25199. FineFT: Efficient and Risk-Aware Ensemble Reinforcement Learning for Futures Trading
+
+- **ID:** `2512.23773` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Molei Qin, Xinyu Cai, Yewen Li, et al.
+- **abs:** https://arxiv.org/abs/2512.23773 · **PDF:** https://arxiv.org/pdf/2512.23773v1
+
+## 25200. On the Sample Complexity of Learning for Blind Inverse Problems
+
+- **ID:** `2512.23405` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Nathan Buskulic, Luca Calatroni, Lorenzo Rosasco, et al.
+- **abs:** https://arxiv.org/abs/2512.23405 · **PDF:** https://arxiv.org/pdf/2512.23405v3
+
+## 25201. Quantum Intelligence Meets BD-RIS-Enabled AmBC: Challenges, Opportunities, and Practical Insights
+
+- **ID:** `2512.23400` · **primary:** `cs.SI` · **Submitted:** 2025-12-29
+- **Authors:** Abd Ullah Khan, Uman Khalid, Trung Q. Duong, et al.
+- **abs:** https://arxiv.org/abs/2512.23400 · **PDF:** https://arxiv.org/pdf/2512.23400v2
+
+## 25202. PINNs for Electromagnetic Wave Propagation
+
+- **ID:** `2512.23396` · **primary:** `physics.comp-ph` · **Submitted:** 2025-12-29
+- **Authors:** Nilufer K. Bulut
+- **abs:** https://arxiv.org/abs/2512.23396 · **PDF:** https://arxiv.org/pdf/2512.23396v2
+
+## 25203. Securing the AI Supply Chain: What Can We Learn From Developer-Reported Security Issues and Solutions of AI Projects?
+
+- **ID:** `2512.23385` · **primary:** `cs.SE` · **Submitted:** 2025-12-29
+- **Authors:** The Anh Nguyen, Triet Huynh Minh Le, M. Ali Babar
+- **abs:** https://arxiv.org/abs/2512.23385 · **PDF:** https://arxiv.org/pdf/2512.23385v2
+
+## 25204. A unified framework for detecting point and collective anomalies in operating system logs via collaborative transformers
+
+- **ID:** `2512.23380` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Mohammad Nasirzadeh, Jafar Tahmoresnezhad, Parviz Rashidi-Khazaee
+- **abs:** https://arxiv.org/abs/2512.23380 · **PDF:** https://arxiv.org/pdf/2512.23380v1
+
+## 25205. SoulX-FlashTalk: Real-Time Infinite Streaming of Audio-Driven Avatars via Self-Correcting Bidirectional Distillation
+
+- **ID:** `2512.23379` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Le Shen, Qian Qiao, Tan Yu, et al.
+- **abs:** https://arxiv.org/abs/2512.23379 · **PDF:** https://arxiv.org/pdf/2512.23379v3
+
+## 25206. Diffusion priors enhanced velocity model building from time-lag images using a neural operator
+
+- **ID:** `2512.23375` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Xiao Ma, Mohammad Hasyim Taufik, Tariq Alkhalifah
+- **abs:** https://arxiv.org/abs/2512.23375 · **PDF:** https://arxiv.org/pdf/2512.23375v1
+
+## 25207. Post-Training Quantization of OpenPangu Models for Efficient Deployment on Atlas A2
+
+- **ID:** `2512.23367` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Yilun Luo, Huaqing Zheng, Haoqian Meng, et al.
+- **abs:** https://arxiv.org/abs/2512.23367 · **PDF:** https://arxiv.org/pdf/2512.23367v2
+
+## 25208. AGRO-SQL: Agentic Group-Relative Optimization with High-Fidelity Data Synthesis
+
+- **ID:** `2512.23366` · **primary:** `cs.DB` · **Submitted:** 2025-12-29
+- **Authors:** Cehua Yang, Dongyu Xiao, Junming Lin, et al.
+- **abs:** https://arxiv.org/abs/2512.23366 · **PDF:** https://arxiv.org/pdf/2512.23366v1
+
+## 25209. A Stepwise-Enhanced Reasoning Framework for Large Language Models Based on External Subgraph Generation
+
+- **ID:** `2512.23356` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Xin Zhang, Yang Cao, Baoxing Wu, et al.
+- **abs:** https://arxiv.org/abs/2512.23356 · **PDF:** https://arxiv.org/pdf/2512.23356v1
+
+## 25210. ISOPO: Proximal policy gradients without pi-old
+
+- **ID:** `2512.23353` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Nilin Abrahamsen
+- **abs:** https://arxiv.org/abs/2512.23353 · **PDF:** https://arxiv.org/pdf/2512.23353v2
+
+## 25211. Persistent Homology via Finite Topological Spaces
+
+- **ID:** `2512.23348` · **primary:** `math.AT` · **Submitted:** 2025-12-29
+- **Authors:** Selçuk Kayacan
+- **abs:** https://arxiv.org/abs/2512.23348 · **PDF:** https://arxiv.org/pdf/2512.23348v2
+
+## 25212. ECG-RAMBA: Zero-Shot ECG Generalization by Morphology-Rhythm Disentanglement and Long-Range Modeling
+
+- **ID:** `2512.23347` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Hai Duong Nguyen, Xuan-The Tran
+- **abs:** https://arxiv.org/abs/2512.23347 · **PDF:** https://arxiv.org/pdf/2512.23347v1
+
+## 25213. Evolving Cognitive Architectures
+
+- **ID:** `2601.05277` · **primary:** `q-bio.NC` · **Submitted:** 2025-12-29
+- **Authors:** Alexander Serov
+- **abs:** https://arxiv.org/abs/2601.05277 · **PDF:** https://arxiv.org/pdf/2601.05277v1
+
+## 25214. AI Meets Brain: Memory Systems from Cognitive Neuroscience to Autonomous Agents
+
+- **ID:** `2512.23343` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Jiafeng Liang, Hao Li, Chang Li, et al.
+- **abs:** https://arxiv.org/abs/2512.23343 · **PDF:** https://arxiv.org/pdf/2512.23343v1
+
+## 25215. The Law of Multi-Model Collaboration: Scaling Limits of Model Ensembling for Large Language Models
+
+- **ID:** `2512.23340` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Dakuan Lu, Jiaqi Zhang, Cheng Yuan, et al.
+- **abs:** https://arxiv.org/abs/2512.23340 · **PDF:** https://arxiv.org/pdf/2512.23340v2
+
+## 25216. Visual Language Hypothesis
+
+- **ID:** `2512.23335` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Xiu Li
+- **abs:** https://arxiv.org/abs/2512.23335 · **PDF:** https://arxiv.org/pdf/2512.23335v2
+
+## 25217. Temporal Attack Pattern Detection in Multi-Agent AI Workflows: An Open Framework for Training Trace-Based Security Models
+
+- **ID:** `2601.00848` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Ron F. Del Rosario
+- **abs:** https://arxiv.org/abs/2601.00848 · **PDF:** https://arxiv.org/pdf/2601.00848v1
+
+## 25218. Deep learning for pedestrians: backpropagation in Transformers
+
+- **ID:** `2512.23329` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Laurent Boué
+- **abs:** https://arxiv.org/abs/2512.23329 · **PDF:** https://arxiv.org/pdf/2512.23329v1
+
+## 25219. CubeBench: Diagnosing Interactive, Long-Horizon Spatial Reasoning Under Partial Observations
+
+- **ID:** `2512.23328` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Huan-ang Gao, Zikang Zhang, Tianwei Luo, et al.
+- **abs:** https://arxiv.org/abs/2512.23328 · **PDF:** https://arxiv.org/pdf/2512.23328v3
+
+## 25220. On Conformant Planning and Model-Checking of $\exists^*\forall^*$ Hyperproperties
+
+- **ID:** `2512.23324` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Raven Beutner, Bernd Finkbeiner
+- **abs:** https://arxiv.org/abs/2512.23324 · **PDF:** https://arxiv.org/pdf/2512.23324v1
+
+## 25221. Explainable Neural Inverse Kinematics for Obstacle-Aware Robotic Manipulation: A Comparative Analysis of IKNet Variants
+
+- **ID:** `2512.23312` · **primary:** `cs.RO` · **Submitted:** 2025-12-29
+- **Authors:** Sheng-Kai Chen, Yi-Ling Tsai, Chun-Chih Chang, et al.
+- **abs:** https://arxiv.org/abs/2512.23312 · **PDF:** https://arxiv.org/pdf/2512.23312v1
+
+## 25222. Splitwise: Collaborative Edge-Cloud Inference for LLMs via Lyapunov-Assisted DRL
+
+- **ID:** `2512.23310` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Abolfazl Younesi, Abbas Shabrang Maryan, Elyas Oustad, et al.
+- **abs:** https://arxiv.org/abs/2512.23310 · **PDF:** https://arxiv.org/pdf/2512.23310v1
+
+## 25223. MedGemma vs GPT-4: Open-Source and Proprietary Zero-shot Medical Disease Classification from Images
+
+- **ID:** `2512.23304` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Md. Sazzadul Islam Prottasha, Nabil Walid Rafi
+- **abs:** https://arxiv.org/abs/2512.23304 · **PDF:** https://arxiv.org/pdf/2512.23304v1
+
+## 25224. AI4Reading: Chinese Audiobook Interpretation System Based on Multi-Agent Collaboration
+
+- **ID:** `2512.23300` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Minjiang Huang, Jipeng Qiang, Yi Zhu, et al.
+- **abs:** https://arxiv.org/abs/2512.23300 · **PDF:** https://arxiv.org/pdf/2512.23300v1
+
+## 25225. Spectral Analysis of Hard-Constraint PINNs: The Spatial Modulation Mechanism of Boundary Functions
+
+- **ID:** `2512.23295` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Yuchen Xie, Honghang Chi, Haopeng Quan, et al.
+- **abs:** https://arxiv.org/abs/2512.23295 · **PDF:** https://arxiv.org/pdf/2512.23295v1
+
+## 25226. Agentic Physical AI toward a Domain-Specific Foundation Model for Nuclear Reactor Control
+
+- **ID:** `2512.23292` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Yoonpyo Lee, Kazuma Kobayashi, Sai Puppala, et al.
+- **abs:** https://arxiv.org/abs/2512.23292 · **PDF:** https://arxiv.org/pdf/2512.23292v2
+
+## 25227. Revealing design archetypes and flexibility in e-molecule import pathways using Modeling to Generate Alternatives and interpretable machine learning
+
+- **ID:** `2512.23284` · **primary:** `eess.SY` · **Submitted:** 2025-12-29
+- **Authors:** Mahdi Kchaou, Francesco Contino, Diederik Coppitters
+- **abs:** https://arxiv.org/abs/2512.23284 · **PDF:** https://arxiv.org/pdf/2512.23284v1
+
+## 25228. Chinese Morph Resolution in E-commerce Live Streaming Scenarios
+
+- **ID:** `2512.23280` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Jiahao Zhu, Jipeng Qiang, Ran Bai, et al.
+- **abs:** https://arxiv.org/abs/2512.23280 · **PDF:** https://arxiv.org/pdf/2512.23280v1
+
+## 25229. You Only Need Your Transformer 25% of the Time: Meaning-First Execution for Eliminating Unnecessary Inference
+
+- **ID:** `2601.00847` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Ryan Shamim
+- **abs:** https://arxiv.org/abs/2601.00847 · **PDF:** https://arxiv.org/pdf/2601.00847v1
+
+## 25230. On the Inverse Flow Matching Problem in the One-Dimensional and Gaussian Cases
+
+- **ID:** `2512.23265` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Alexander Korotin, Gudmund Pammer
+- **abs:** https://arxiv.org/abs/2512.23265 · **PDF:** https://arxiv.org/pdf/2512.23265v1
+
+## 25231. La norme technique comme catalyseur de transfert de connaissances : la francophonie a l'œuvre dans le domaine de l'{é}ducation
+
+- **ID:** `2601.06069` · **primary:** `cs.CY` · **Submitted:** 2025-12-29
+- **Authors:** Mokhtar Ben Henda
+- **abs:** https://arxiv.org/abs/2601.06069 · **PDF:** https://arxiv.org/pdf/2601.06069v1
+
+## 25232. BIRD: A Museum Open Dataset Combining Behavior Patterns and Identity Types to Better Model Visitors' Experience
+
+- **ID:** `2602.11160` · **primary:** `cs.HC` · **Submitted:** 2025-12-29
+- **Authors:** Alexanne Worm, Florian Marchal, Sylvain Castagnos
+- **abs:** https://arxiv.org/abs/2602.11160 · **PDF:** https://arxiv.org/pdf/2602.11160v1
+
+## 25233. PFed-Signal: An ADR Prediction Model based on Federated Learning
+
+- **ID:** `2512.23262` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Tao Li, Peilin Li, Kui Lu, et al.
+- **abs:** https://arxiv.org/abs/2512.23262 · **PDF:** https://arxiv.org/pdf/2512.23262v1
+
+## 25234. Interpretable Safety Alignment via SAE-Constructed Low-Rank Subspace Adaptation
+
+- **ID:** `2512.23260` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Dianyun Wang, Qingsen Ma, Yuhu Shang, et al.
+- **abs:** https://arxiv.org/abs/2512.23260 · **PDF:** https://arxiv.org/pdf/2512.23260v2
+
+## 25235. SB-TRPO: Towards Safe Reinforcement Learning with Hard Constraints
+
+- **ID:** `2512.23770` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Ankit Kanwar, Dominik Wagner, Luke Ong
+- **abs:** https://arxiv.org/abs/2512.23770 · **PDF:** https://arxiv.org/pdf/2512.23770v2
+
+## 25236. ViLaCD-R1: A Vision-Language Framework for Semantic Change Detection in Remote Sensing
+
+- **ID:** `2512.23244` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Xingwei Ma, Shiyang Feng, Bo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.23244 · **PDF:** https://arxiv.org/pdf/2512.23244v1
+
+## 25237. Uncovering Discrimination Clusters: Quantifying and Explaining Systematic Fairness Violations
+
+- **ID:** `2512.23769` · **primary:** `cs.SE` · **Submitted:** 2025-12-29
+- **Authors:** Ranit Debnath Akash, Ashish Kumar, Verya Monjezi, et al.
+- **abs:** https://arxiv.org/abs/2512.23769 · **PDF:** https://arxiv.org/pdf/2512.23769v1
+
+## 25238. KernelEvolve: Scaling Agentic Kernel Coding for Heterogeneous AI Accelerators at Meta
+
+- **ID:** `2512.23236` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Gang Liao, Hongsen Qin, Ying Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.23236 · **PDF:** https://arxiv.org/pdf/2512.23236v3
+
+## 25239. FairGFL: Privacy-Preserving Fairness-Aware Federated Learning with Overlapping Subgraphs
+
+- **ID:** `2512.23235` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Zihao Zhou, Shusen Yang, Fangyuan Zhao, et al.
+- **abs:** https://arxiv.org/abs/2512.23235 · **PDF:** https://arxiv.org/pdf/2512.23235v1
+
+## 25240. Physics-Inspired Modeling and Content Adaptive Routing in an Infrared Gas Leak Detection Network
+
+- **ID:** `2512.23234` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Dongsheng Li, Tianli Ma, Siling Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.23234 · **PDF:** https://arxiv.org/pdf/2512.23234v2
+
+## 25241. Anomaly Detection by Effectively Leveraging Synthetic Images
+
+- **ID:** `2512.23227` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Sungho Kang, Hyunkyu Park, Yeonho Lee, et al.
+- **abs:** https://arxiv.org/abs/2512.23227 · **PDF:** https://arxiv.org/pdf/2512.23227v1
+
+## 25242. Holi-DETR: Holistic Fashion Item Detection Leveraging Contextual Information
+
+- **ID:** `2512.23221` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Youngchae Kwon, Jinyoung Choi, Injung Kim
+- **abs:** https://arxiv.org/abs/2512.23221 · **PDF:** https://arxiv.org/pdf/2512.23221v1
+
+## 25243. AI-Native Integrated Sensing and Communications for Self-Organizing Wireless Networks: Architectures, Learning Paradigms, and System-Level Design
+
+- **ID:** `2601.02398` · **primary:** `cs.NI` · **Submitted:** 2025-12-29
+- **Authors:** S. Zhang, M. Feizarefi, A. F. Mirzaei
+- **abs:** https://arxiv.org/abs/2601.02398 · **PDF:** https://arxiv.org/pdf/2601.02398v1
+
+## 25244. TCEval: Using Thermal Comfort to Assess Cognitive and Perceptual Abilities of AI
+
+- **ID:** `2512.23217` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Jingming Li
+- **abs:** https://arxiv.org/abs/2512.23217 · **PDF:** https://arxiv.org/pdf/2512.23217v1
+
+## 25245. Anka: A Domain-Specific Language for Reliable LLM Code Generation
+
+- **ID:** `2512.23214` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Saif Khalfan Saif Al Mazrouei
+- **abs:** https://arxiv.org/abs/2512.23214 · **PDF:** https://arxiv.org/pdf/2512.23214v1
+
+## 25246. Scoring, Reasoning, and Selecting the Best! Ensembling Large Language Models via a Peer-Review Process
+
+- **ID:** `2512.23213` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Zhijun Chen, Zeyu Ji, Qianren Mao, et al.
+- **abs:** https://arxiv.org/abs/2512.23213 · **PDF:** https://arxiv.org/pdf/2512.23213v2
+
+## 25247. Exploring Syn-to-Real Domain Adaptation for Military Target Detection
+
+- **ID:** `2512.23208` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Jongoh Jeong, Youngjin Oh, Gyeongrae Nam, et al.
+- **abs:** https://arxiv.org/abs/2512.23208 · **PDF:** https://arxiv.org/pdf/2512.23208v1
+
+## 25248. Not too long do read: Evaluating LLM-generated extreme scientific summaries
+
+- **ID:** `2512.23206` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Zhuoqi Lyu, Qing Ke
+- **abs:** https://arxiv.org/abs/2512.23206 · **PDF:** https://arxiv.org/pdf/2512.23206v1
+
+## 25249. Enabling Physical AI at the Edge: Hardware-Accelerated Recovery of System Dynamics
+
+- **ID:** `2512.23767` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Bin Xu, Ayan Banerjee, Sandeep Gupta
+- **abs:** https://arxiv.org/abs/2512.23767 · **PDF:** https://arxiv.org/pdf/2512.23767v1
+
+## 25250. Energy and Memory-Efficient Federated Learning With Ordered Layer Freezing
+
+- **ID:** `2512.23200` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Ziru Niu, Hai Dong, A. K. Qin, et al.
+- **abs:** https://arxiv.org/abs/2512.23200 · **PDF:** https://arxiv.org/pdf/2512.23200v1
+
+## 25251. ForCM: Forest Cover Mapping from Multispectral Sentinel-2 Image by Integrating Deep Learning with Object-Based Image Analysis
+
+- **ID:** `2512.23196` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Maisha Haque, Israt Jahan Ayshi, Sadaf M. Anis, et al.
+- **abs:** https://arxiv.org/abs/2512.23196 · **PDF:** https://arxiv.org/pdf/2512.23196v1
+
+## 25252. PGOT: A Physics-Geometry Operator Transformer for Complex PDEs
+
+- **ID:** `2512.23192` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Zhuo Zhang, Xi Yang, Ying Miao, et al.
+- **abs:** https://arxiv.org/abs/2512.23192 · **PDF:** https://arxiv.org/pdf/2512.23192v2
+
+## 25253. A Simple, Optimal and Efficient Algorithm for Online Exp-Concave Optimization
+
+- **ID:** `2512.23190` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Yi-Han Wang, Peng Zhao, Zhi-Hua Zhou
+- **abs:** https://arxiv.org/abs/2512.23190 · **PDF:** https://arxiv.org/pdf/2512.23190v2
+
+## 25254. EIR: Enhanced Image Representations for Medical Report Generation
+
+- **ID:** `2512.23185` · **primary:** `eess.IV` · **Submitted:** 2025-12-29
+- **Authors:** Qiang Sun, Zongcheng Ji, Yinlong Xiao, et al.
+- **abs:** https://arxiv.org/abs/2512.23185 · **PDF:** https://arxiv.org/pdf/2512.23185v1
+
+## 25255. From Model Choice to Model Belief: Establishing a New Measure for LLM-Based Research
+
+- **ID:** `2512.23184` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Hongshen Sun, Juanjuan Zhang
+- **abs:** https://arxiv.org/abs/2512.23184 · **PDF:** https://arxiv.org/pdf/2512.23184v1
+
+## 25256. Clipped Gradient Methods for Nonsmooth Convex Optimization under Heavy-Tailed Noise: A Refined Analysis
+
+- **ID:** `2512.23178` · **primary:** `math.OC` · **Submitted:** 2025-12-29
+- **Authors:** Zijian Liu
+- **abs:** https://arxiv.org/abs/2512.23178 · **PDF:** https://arxiv.org/pdf/2512.23178v2
+
+## 25257. Machine Learning-Assisted Vocal Cord Ultrasound Examination: Project VIPR
+
+- **ID:** `2512.23177` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Will Sebelik-Lassiter, Evan Schubert, Muhammad Alliyu, et al.
+- **abs:** https://arxiv.org/abs/2512.23177 · **PDF:** https://arxiv.org/pdf/2512.23177v1
+
+## 25258. HELM-BERT: A Transformer for Medium-sized Peptide Property Prediction
+
+- **ID:** `2512.23175` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Seungeon Lee, Takuto Koyama, Itsuki Maeda, et al.
+- **abs:** https://arxiv.org/abs/2512.23175 · **PDF:** https://arxiv.org/pdf/2512.23175v1
+
+## 25259. EquaCode: A Multi-Strategy Jailbreak Approach for Large Language Models via Equation Solving and Code Completion
+
+- **ID:** `2512.23173` · **primary:** `cs.CR` · **Submitted:** 2025-12-29
+- **Authors:** Zhen Liang, Hai Huang, Zhengkui Chen
+- **abs:** https://arxiv.org/abs/2512.23173 · **PDF:** https://arxiv.org/pdf/2512.23173v1
+
+## 25260. Certifying the Right to Be Forgotten: Primal-Dual Optimization for Sample and Label Unlearning in Vertical Federated Learning
+
+- **ID:** `2512.23171` · **primary:** `cs.CR` · **Submitted:** 2025-12-29
+- **Authors:** Yu Jiang, Xindi Tong, Ziyao Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.23171 · **PDF:** https://arxiv.org/pdf/2512.23171v3
+
+## 25261. SPIRAL: Symbolic LLM Planning via Grounded and Reflective Search
+
+- **ID:** `2512.23167` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Yifan Zhang, Giridhar Ganapavarapu, Srideepika Jayaraman, et al.
+- **abs:** https://arxiv.org/abs/2512.23167 · **PDF:** https://arxiv.org/pdf/2512.23167v1
+
+## 25262. Evaluating Parameter Efficient Methods for RLVR
+
+- **ID:** `2512.23165` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Qingyu Yin, Yulun Wu, Zhennan Shen, et al.
+- **abs:** https://arxiv.org/abs/2512.23165 · **PDF:** https://arxiv.org/pdf/2512.23165v2
+
+## 25263. Why We Need a New Framework for Emotional Intelligence in AI
+
+- **ID:** `2512.23163` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Max Parks, Kheli Atluru, Meera Vinod, et al.
+- **abs:** https://arxiv.org/abs/2512.23163 · **PDF:** https://arxiv.org/pdf/2512.23163v1
+
+## 25264. Enhancing Temporal Awareness in LLMs for Temporal Point Processes
+
+- **ID:** `2601.00845` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Lili Chen, Wensheng Gan, Shuang Liang, et al.
+- **abs:** https://arxiv.org/abs/2601.00845 · **PDF:** https://arxiv.org/pdf/2601.00845v1
+
+## 25265. Diffusion-based Decentralized Federated Multi-Task Representation Learning
+
+- **ID:** `2512.23161` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Donghwa Kang, Shana Moothedath
+- **abs:** https://arxiv.org/abs/2512.23161 · **PDF:** https://arxiv.org/pdf/2512.23161v1
+
+## 25266. A Weak Signal Learning Dataset and Its Baseline Method
+
+- **ID:** `2512.23160` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Xianqi Liu, Xiangru Li, Lefeng He, et al.
+- **abs:** https://arxiv.org/abs/2512.23160 · **PDF:** https://arxiv.org/pdf/2512.23160v1
+
+## 25267. A Granular Grassmannian Clustering Framework via the Schubert Variety of Best Fit
+
+- **ID:** `2512.23766` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Karim Salta, Michael Kirby, Chris Peterson
+- **abs:** https://arxiv.org/abs/2512.23766 · **PDF:** https://arxiv.org/pdf/2512.23766v1
+
+## 25268. Constraint programming model and biased random-key genetic algorithm for the single-machine coupled task scheduling problem with exact delays to minimize the makespan
+
+- **ID:** `2512.23150` · **primary:** `math.OC` · **Submitted:** 2025-12-29
+- **Authors:** Vítor A. Barbosa, Rafael A. Melo
+- **abs:** https://arxiv.org/abs/2512.23150 · **PDF:** https://arxiv.org/pdf/2512.23150v1
+
+## 25269. Reservoir Computing inspired Matrix Multiplication-free Language Model
+
+- **ID:** `2512.23145` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Takumi Shiratsuchi, Yuichiro Tanaka, Hakaru Tamukoh
+- **abs:** https://arxiv.org/abs/2512.23145 · **PDF:** https://arxiv.org/pdf/2512.23145v1
+
+## 25270. An Inference-Based Architecture for Intent and Affordance Saturation in Decision-Making
+
+- **ID:** `2512.23144` · **primary:** `q-bio.NC` · **Submitted:** 2025-12-29
+- **Authors:** Wendyam Eric Lionel Ilboudo, Saori C Tanaka
+- **abs:** https://arxiv.org/abs/2512.23144 · **PDF:** https://arxiv.org/pdf/2512.23144v1
+
+## 25271. Why Machine Learning Models Systematically Underestimate Extreme Values II: How to Fix It with LatentNN
+
+- **ID:** `2512.23138` · **primary:** `astro-ph.IM` · **Submitted:** 2025-12-29
+- **Authors:** Yuan-Sen Ting
+- **abs:** https://arxiv.org/abs/2512.23138 · **PDF:** https://arxiv.org/pdf/2512.23138v2
+
+## 25272. Graph Neural Networks with Transformer Fusion of Brain Connectivity Dynamics and Tabular Data for Forecasting Future Tobacco Use
+
+- **ID:** `2512.23137` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Runzhi Zhou, Xi Luo
+- **abs:** https://arxiv.org/abs/2512.23137 · **PDF:** https://arxiv.org/pdf/2512.23137v1
+
+## 25273. Principled Algorithms for Optimizing Generalized Metrics in Binary Classification
+
+- **ID:** `2512.23133` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Anqi Mao, Mehryar Mohri, Yutao Zhong
+- **abs:** https://arxiv.org/abs/2512.23133 · **PDF:** https://arxiv.org/pdf/2512.23133v1
+
+## 25274. Multi-Agent Framework for Threat Mitigation and Resilience in AI-Based Systems
+
+- **ID:** `2512.23132` · **primary:** `cs.CR` · **Submitted:** 2025-12-29
+- **Authors:** Armstrong Foundjem, Lionel Nganyewou Tidjon, Leuson Da Silva, et al.
+- **abs:** https://arxiv.org/abs/2512.23132 · **PDF:** https://arxiv.org/pdf/2512.23132v1
+
+## 25275. SE-MLP Model for Predicting Prior Acceleration Features in Penetration Signals
+
+- **ID:** `2512.23131` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Yankang Li, Changsheng Li
+- **abs:** https://arxiv.org/abs/2512.23131 · **PDF:** https://arxiv.org/pdf/2512.23131v1
+
+## 25276. PathoSyn: Imaging-Pathology MRI Synthesis via Disentangled Deviation Diffusion
+
+- **ID:** `2512.23130` · **primary:** `cs.CV` · **Submitted:** 2025-12-29
+- **Authors:** Jian Wang, Sixing Rong, Jiarui Xing, et al.
+- **abs:** https://arxiv.org/abs/2512.23130 · **PDF:** https://arxiv.org/pdf/2512.23130v2
+
+## 25277. It's a TRAP! Task-Redirecting Agent Persuasion Benchmark for Web Agents
+
+- **ID:** `2512.23128` · **primary:** `cs.HC` · **Submitted:** 2025-12-29
+- **Authors:** Karolina Korgul, Yushi Yang, Arkadiusz Drohomirecki, et al.
+- **abs:** https://arxiv.org/abs/2512.23128 · **PDF:** https://arxiv.org/pdf/2512.23128v1
+
+## 25278. InSPO: Unlocking Intrinsic Self-Reflection for LLM Preference Optimization
+
+- **ID:** `2512.23126` · **primary:** `cs.AI` · **Submitted:** 2025-12-29
+- **Authors:** Yu Li, Tian Lan, Zhengling Qi
+- **abs:** https://arxiv.org/abs/2512.23126 · **PDF:** https://arxiv.org/pdf/2512.23126v3
+
+## 25279. Entropy-Aware Speculative Decoding Toward Improved LLM Reasoning
+
+- **ID:** `2512.23765` · **primary:** `cs.CL` · **Submitted:** 2025-12-29
+- **Authors:** Tiancheng Su, Meicong Zhang, Guoxiu He
+- **abs:** https://arxiv.org/abs/2512.23765 · **PDF:** https://arxiv.org/pdf/2512.23765v1
+
+## 25280. Exploring Cumulative Effects in Survival Data Using Deep Learning Networks
+
+- **ID:** `2512.23764` · **primary:** `cs.LG` · **Submitted:** 2025-12-29
+- **Authors:** Kang-Chung Yang, Shinsheng Yuan
+- **abs:** https://arxiv.org/abs/2512.23764 · **PDF:** https://arxiv.org/pdf/2512.23764v1
+
+## 25281. Channel Selected Stratified Nested Cross Validation for Clinically Relevant EEG Based Parkinsons Disease Detection
+
+- **ID:** `2601.05276` · **primary:** `eess.SP` · **Submitted:** 2025-12-28
+- **Authors:** Nicholas R. Rasmussen, Rodrigue Rizk, Longwei Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.05276 · **PDF:** https://arxiv.org/pdf/2601.05276v2
+
+## 25282. How Much Data Is Enough? Uniform Convergence Bounds for Generative & Vision-Language Models under Low-Dimensional Structure
+
+- **ID:** `2512.23109` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Paul M. Thompson
+- **abs:** https://arxiv.org/abs/2512.23109 · **PDF:** https://arxiv.org/pdf/2512.23109v1
+
+## 25283. Can Consumer Chatbots Reason? A Student-Led Field Experiment Embedded in an "AI-for-All" Undergraduate Course
+
+- **ID:** `2601.04225` · **primary:** `cs.CY` · **Submitted:** 2025-12-28
+- **Authors:** Amarda Shehu, Adonyas Ababu, Asma Akbary, et al.
+- **abs:** https://arxiv.org/abs/2601.04225 · **PDF:** https://arxiv.org/pdf/2601.04225v1
+
+## 25284. Neural Optimal Design of Experiment for Inverse Problems
+
+- **ID:** `2512.23763` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** John E. Darges, Babak Maboudi Afkham, Matthias Chung
+- **abs:** https://arxiv.org/abs/2512.23763 · **PDF:** https://arxiv.org/pdf/2512.23763v2
+
+## 25285. A Note on Hybrid Online Reinforcement and Imitation Learning for LLMs: Formulations and Algorithms
+
+- **ID:** `2512.23097` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Yingru Li, Ziniu Li, Jiacai Liu
+- **abs:** https://arxiv.org/abs/2512.23097 · **PDF:** https://arxiv.org/pdf/2512.23097v1
+
+## 25286. Osmotic Learning: A Self-Supervised Paradigm for Decentralized Contextual Data Representation
+
+- **ID:** `2512.23096` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Mario Colosi, Reza Farahani, Maria Fazio, et al.
+- **abs:** https://arxiv.org/abs/2512.23096 · **PDF:** https://arxiv.org/pdf/2512.23096v1
+
+## 25287. Drift-Based Dataset Stability Benchmark
+
+- **ID:** `2512.23762` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Dominik Soukup, Richard Plný, Daniel Vašata, et al.
+- **abs:** https://arxiv.org/abs/2512.23762 · **PDF:** https://arxiv.org/pdf/2512.23762v1
+
+## 25288. Learning Coupled System Dynamics under Incomplete Physical Constraints and Missing Data
+
+- **ID:** `2512.23761` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Esha Saha, Hao Wang
+- **abs:** https://arxiv.org/abs/2512.23761 · **PDF:** https://arxiv.org/pdf/2512.23761v1
+
+## 25289. Benchmark Success, Clinical Failure: When Reinforcement Learning Optimizes for Benchmarks, Not Patients
+
+- **ID:** `2512.23090` · **primary:** `cs.AI` · **Submitted:** 2025-12-28
+- **Authors:** Armin Berger, Manuela Bergau, Helen Schneider, et al.
+- **abs:** https://arxiv.org/abs/2512.23090 · **PDF:** https://arxiv.org/pdf/2512.23090v2
+
+## 25290. MedSAM-based lung masking for multi-label chest X-ray classification
+
+- **ID:** `2512.23089` · **primary:** `cs.CV` · **Submitted:** 2025-12-28
+- **Authors:** Brayden Miao, Zain Rehman, Xin Miao, et al.
+- **abs:** https://arxiv.org/abs/2512.23089 · **PDF:** https://arxiv.org/pdf/2512.23089v1
+
+## 25291. Dynamic Vocabulary Pruning: Stable LLM-RL by Taming the Tail
+
+- **ID:** `2512.23087` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Yingru Li, Jiawei Xu, Jiacai Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.23087 · **PDF:** https://arxiv.org/pdf/2512.23087v2
+
+## 25292. QSAR-Guided Generative Framework for the Discovery of Synthetically Viable Odorants
+
+- **ID:** `2512.23080` · **primary:** `physics.chem-ph` · **Submitted:** 2025-12-28
+- **Authors:** Tim C. Pearce, Ahmed Ibrahim
+- **abs:** https://arxiv.org/abs/2512.23080 · **PDF:** https://arxiv.org/pdf/2512.23080v1
+
+## 25293. Deep Learning for Art Market Valuation
+
+- **ID:** `2512.23078` · **primary:** `q-fin.GN` · **Submitted:** 2025-12-28
+- **Authors:** Jianping Mei, Michael Moses, Jan Waelty, et al.
+- **abs:** https://arxiv.org/abs/2512.23078 · **PDF:** https://arxiv.org/pdf/2512.23078v1
+
+## 25294. Multimodal Functional Maximum Correlation for Emotion Recognition
+
+- **ID:** `2512.23076` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Deyang Zheng, Tianyi Zhang, Wenming Zheng, et al.
+- **abs:** https://arxiv.org/abs/2512.23076 · **PDF:** https://arxiv.org/pdf/2512.23076v1
+
+## 25295. Trust Region Masking for Long-Horizon LLM Reinforcement Learning
+
+- **ID:** `2512.23075` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Yingru Li, Jiacai Liu, Jiawei Xu, et al.
+- **abs:** https://arxiv.org/abs/2512.23075 · **PDF:** https://arxiv.org/pdf/2512.23075v4
+
+## 25296. Rethinking Fine-Tuning: Unlocking Hidden Capabilities in Vision-Language Models
+
+- **ID:** `2512.23073` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Mingyuan Zhang, Yue Bai, Yifan Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.23073 · **PDF:** https://arxiv.org/pdf/2512.23073v1
+
+## 25297. Federated Learning With L0 Constraint Via Probabilistic Gates For Sparsity
+
+- **ID:** `2512.23071` · **primary:** `stat.ML` · **Submitted:** 2025-12-28
+- **Authors:** Krishna Harsha Kovelakuntla Huthasana, Alireza Olama, Andreas Lundell
+- **abs:** https://arxiv.org/abs/2512.23071 · **PDF:** https://arxiv.org/pdf/2512.23071v1
+
+## 25298. FLEX-MoE: Federated Mixture-of-Experts with Load-balanced Expert Assignment
+
+- **ID:** `2512.23070` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Boyang Zhang, Xiaobing Chen, Songyang Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.23070 · **PDF:** https://arxiv.org/pdf/2512.23070v1
+
+## 25299. Breaking the Memory Wall: Exact Analytical Differentiation via Tiled Operator-Space Evolution
+
+- **ID:** `2512.23068` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Shuhuan Wang, Yuzhen Xie, Jiayi Li, et al.
+- **abs:** https://arxiv.org/abs/2512.23068 · **PDF:** https://arxiv.org/pdf/2512.23068v1
+
+## 25300. The Reward Model Selection Crisis in Personalized Alignment
+
+- **ID:** `2512.23067` · **primary:** `cs.AI` · **Submitted:** 2025-12-28
+- **Authors:** Fady Rezk, Yuangang Pan, Chuan-Sheng Foo, et al.
+- **abs:** https://arxiv.org/abs/2512.23067 · **PDF:** https://arxiv.org/pdf/2512.23067v2
+
+## 25301. TabiBERT: A Large-Scale ModernBERT Foundation Model and A Unified Benchmark for Turkish
+
+- **ID:** `2512.23065` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Melikşah Türker, A. Ebrar Kızıloğlu, Onur Güngör, et al.
+- **abs:** https://arxiv.org/abs/2512.23065 · **PDF:** https://arxiv.org/pdf/2512.23065v3
+
+## 25302. Value-guided action planning with JEPA world models
+
+- **ID:** `2601.00844` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Matthieu Destrade, Oumayma Bounou, Quentin Le Lidec, et al.
+- **abs:** https://arxiv.org/abs/2601.00844 · **PDF:** https://arxiv.org/pdf/2601.00844v1
+
+## 25303. PI-MFM: Physics-informed multimodal foundation model for solving partial differential equations
+
+- **ID:** `2512.23056` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Min Zhu, Jingmin Sun, Zecheng Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.23056 · **PDF:** https://arxiv.org/pdf/2512.23056v1
+
+## 25304. Audited Skill-Graph Self-Improvement for Agentic LLMs via Verifiable Rewards, Experience Synthesis, and Continual Memory
+
+- **ID:** `2512.23760` · **primary:** `cs.CR` · **Submitted:** 2025-12-28
+- **Authors:** Ken Huang, Jerry Huang
+- **abs:** https://arxiv.org/abs/2512.23760 · **PDF:** https://arxiv.org/pdf/2512.23760v1
+
+## 25305. Accelerating Language Model Workflows with Prompt Choreography
+
+- **ID:** `2512.23049` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** TJ Bai, Jason Eisner
+- **abs:** https://arxiv.org/abs/2512.23049 · **PDF:** https://arxiv.org/pdf/2512.23049v1
+
+## 25306. Mechanistic Analysis of Circuit Preservation in Federated Learning
+
+- **ID:** `2512.23043` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Muhammad Haseeb, Salaar Masood, Muhammad Abdullah Sohail
+- **abs:** https://arxiv.org/abs/2512.23043 · **PDF:** https://arxiv.org/pdf/2512.23043v1
+
+## 25307. Problems With Large Language Models for Learner Modelling: Why LLMs Alone Fall Short for Responsible Tutoring in K--12 Education
+
+- **ID:** `2512.23036` · **primary:** `cs.AI` · **Submitted:** 2025-12-28
+- **Authors:** Danial Hooshyar, Yeongwook Yang, Gustav Šíř, et al.
+- **abs:** https://arxiv.org/abs/2512.23036 · **PDF:** https://arxiv.org/pdf/2512.23036v1
+
+## 25308. Is Chain-of-Thought Really Not Explainability? Chain-of-Thought Can Be Faithful without Hint Verbalization
+
+- **ID:** `2512.23032` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Kerem Zaman, Shashank Srivastava
+- **abs:** https://arxiv.org/abs/2512.23032 · **PDF:** https://arxiv.org/pdf/2512.23032v1
+
+## 25309. Viability and Performance of a Private LLM Server for SMBs: A Benchmark Analysis of Qwen3-30B on Consumer-Grade Hardware
+
+- **ID:** `2512.23029` · **primary:** `cs.DC` · **Submitted:** 2025-12-28
+- **Authors:** Alex Khalil, Guillaume Heilles, Maria Parraga, et al.
+- **abs:** https://arxiv.org/abs/2512.23029 · **PDF:** https://arxiv.org/pdf/2512.23029v1
+
+## 25310. An Architecture-Led Hybrid Report on Body Language Detection Project
+
+- **ID:** `2512.23028` · **primary:** `cs.CV` · **Submitted:** 2025-12-28
+- **Authors:** Thomson Tong, Diba Darooneh
+- **abs:** https://arxiv.org/abs/2512.23028 · **PDF:** https://arxiv.org/pdf/2512.23028v1
+
+## 25311. LENS: LLM-Enabled Narrative Synthesis for Mental Health by Aligning Multimodal Sensing with Language Models
+
+- **ID:** `2512.23025` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Wenxuan Xu, Arvind Pillai, Subigya Nepal, et al.
+- **abs:** https://arxiv.org/abs/2512.23025 · **PDF:** https://arxiv.org/pdf/2512.23025v1
+
+## 25312. OpenGround: Active Cognition-based Reasoning for Open-World 3D Visual Grounding
+
+- **ID:** `2512.23020` · **primary:** `cs.CV` · **Submitted:** 2025-12-28
+- **Authors:** Wenyuan Huang, Zhao Wang, Zhou Wei, et al.
+- **abs:** https://arxiv.org/abs/2512.23020 · **PDF:** https://arxiv.org/pdf/2512.23020v2
+
+## 25313. Merge before Forget: A Single LoRA Continual Learning via Continual Merging
+
+- **ID:** `2512.23017` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Fuli Qiao, Mehrdad Mahdavi
+- **abs:** https://arxiv.org/abs/2512.23017 · **PDF:** https://arxiv.org/pdf/2512.23017v1
+
+## 25314. Improving Generalization in LLM Structured Pruning via Function-Aware Neuron Grouping
+
+- **ID:** `2512.23014` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Tao Yu, Yongqi An, Kuan Zhu, et al.
+- **abs:** https://arxiv.org/abs/2512.23014 · **PDF:** https://arxiv.org/pdf/2512.23014v1
+
+## 25315. Clinical Document Metadata Extraction: A Scoping Review
+
+- **ID:** `2601.09730` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Kurt Miller, Qiuhao Lu, William Hersh, et al.
+- **abs:** https://arxiv.org/abs/2601.09730 · **PDF:** https://arxiv.org/pdf/2601.09730v1
+
+## 25316. JADAI: Jointly Amortizing Adaptive Design and Bayesian Inference
+
+- **ID:** `2512.22999` · **primary:** `stat.ML` · **Submitted:** 2025-12-28
+- **Authors:** Niels Bracher, Lars Kühmichel, Desi R. Ivanova, et al.
+- **abs:** https://arxiv.org/abs/2512.22999 · **PDF:** https://arxiv.org/pdf/2512.22999v1
+
+## 25317. Enhancing Business Analytics through Hybrid Summarization of Financial Reports
+
+- **ID:** `2601.09729` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Tohida Rehman
+- **abs:** https://arxiv.org/abs/2601.09729 · **PDF:** https://arxiv.org/pdf/2601.09729v1
+
+## 25318. Fusion or Confusion? Multimodal Complexity Is Not All You Need
+
+- **ID:** `2512.22991` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Tillmann Rheude, Roland Eils, Benjamin Wild
+- **abs:** https://arxiv.org/abs/2512.22991 · **PDF:** https://arxiv.org/pdf/2512.22991v2
+
+## 25319. Risk-Averse Learning with Varying Risk Levels
+
+- **ID:** `2512.22986` · **primary:** `math.OC` · **Submitted:** 2025-12-28
+- **Authors:** Siyi Wang, Zifan Wang, Karl H. Johansson
+- **abs:** https://arxiv.org/abs/2512.22986 · **PDF:** https://arxiv.org/pdf/2512.22986v1
+
+## 25320. OmniNeuro: A Multimodal HCI Framework for Explainable BCI Feedback via Generative AI and Sonification
+
+- **ID:** `2601.00843` · **primary:** `cs.AI` · **Submitted:** 2025-12-28
+- **Authors:** Ayda Aghaei Nia
+- **abs:** https://arxiv.org/abs/2601.00843 · **PDF:** https://arxiv.org/pdf/2601.00843v1
+
+## 25321. Explaining AI Without Code: A User Study on Explainable AI
+
+- **ID:** `2602.11159` · **primary:** `cs.AI` · **Submitted:** 2025-12-28
+- **Authors:** Natalia Abarca, Andrés Carvallo, Claudia López Moncada, et al.
+- **abs:** https://arxiv.org/abs/2602.11159 · **PDF:** https://arxiv.org/pdf/2602.11159v1
+
+## 25322. A Context-Aware Temporal Modeling through Unified Multi-Scale Temporal Encoding and Hierarchical Sequence Learning for Single-Channel EEG Sleep Staging
+
+- **ID:** `2512.22976` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Amirali Vakili, Salar Jahanshiri, Armin Salimi-Badr
+- **abs:** https://arxiv.org/abs/2512.22976 · **PDF:** https://arxiv.org/pdf/2512.22976v2
+
+## 25323. Prompt engineering does not universally improve Large Language Model performance across clinical decision-making tasks
+
+- **ID:** `2512.22966` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Mengdi Chai, Ali R. Zomorrodi
+- **abs:** https://arxiv.org/abs/2512.22966 · **PDF:** https://arxiv.org/pdf/2512.22966v1
+
+## 25324. Deep Learning for the Multiple Optimal Stopping Problem
+
+- **ID:** `2512.22961` · **primary:** `math.OC` · **Submitted:** 2025-12-28
+- **Authors:** Mathieu Laurière, Mehdi Talbi
+- **abs:** https://arxiv.org/abs/2512.22961 · **PDF:** https://arxiv.org/pdf/2512.22961v1
+
+## 25325. FLOW: A Feedback-Driven Synthetic Longitudinal Dataset of Work and Wellbeing
+
+- **ID:** `2512.22956` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Wafaa El Husseini
+- **abs:** https://arxiv.org/abs/2512.22956 · **PDF:** https://arxiv.org/pdf/2512.22956v1
+
+## 25326. Diversity or Precision? A Deep Dive into Next Token Prediction
+
+- **ID:** `2512.22955` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Haoyuan Wu, Hai Wang, Jiajia Wu, et al.
+- **abs:** https://arxiv.org/abs/2512.22955 · **PDF:** https://arxiv.org/pdf/2512.22955v4
+
+## 25327. APO: Alpha-Divergence Preference Optimization
+
+- **ID:** `2512.22953` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Wang Zixian
+- **abs:** https://arxiv.org/abs/2512.22953 · **PDF:** https://arxiv.org/pdf/2512.22953v1
+
+## 25328. Multiple Token Divergence: Measuring and Steering In-Context Computation Density
+
+- **ID:** `2512.22944` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Vincent Herrmann, Eric Alcaide, Michael Wand, et al.
+- **abs:** https://arxiv.org/abs/2512.22944 · **PDF:** https://arxiv.org/pdf/2512.22944v1
+
+## 25329. Heterogeneity in Multi-Agent Reinforcement Learning
+
+- **ID:** `2512.22941` · **primary:** `cs.MA` · **Submitted:** 2025-12-28
+- **Authors:** Tianyi Hu, Zhiqiang Pu, Yuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.22941 · **PDF:** https://arxiv.org/pdf/2512.22941v1
+
+## 25330. Multimodal Fact-Checking: An Agent-based Approach
+
+- **ID:** `2512.22933` · **primary:** `cs.AI` · **Submitted:** 2025-12-28
+- **Authors:** Danni Xu, Shaojing Fan, Harry Cheng, et al.
+- **abs:** https://arxiv.org/abs/2512.22933 · **PDF:** https://arxiv.org/pdf/2512.22933v3
+
+## 25331. Geometric Structural Knowledge Graph Foundation Model
+
+- **ID:** `2512.22931` · **primary:** `cs.AI` · **Submitted:** 2025-12-28
+- **Authors:** Ling Xin, Mojtaba Nayyeri, Zahra Makki Nayeri, et al.
+- **abs:** https://arxiv.org/abs/2512.22931 · **PDF:** https://arxiv.org/pdf/2512.22931v1
+
+## 25332. Eliminating Agentic Workflow for Introduction Generation with Parametric Stage Tokens
+
+- **ID:** `2601.09728` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Meicong Zhang, Tiancheng su, Guoxiu He
+- **abs:** https://arxiv.org/abs/2601.09728 · **PDF:** https://arxiv.org/pdf/2601.09728v1
+
+## 25333. Sat-EnQ: Satisficing Ensembles of Weak Q-Learners for Reliable and Compute-Efficient Reinforcement Learning
+
+- **ID:** `2512.22910` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Ünver Çiftçi
+- **abs:** https://arxiv.org/abs/2512.22910 · **PDF:** https://arxiv.org/pdf/2512.22910v1
+
+## 25334. A first-order method for nonconvex-strongly-concave constrained minimax optimization
+
+- **ID:** `2512.22909` · **primary:** `math.OC` · **Submitted:** 2025-12-28
+- **Authors:** Zhaosong Lu, Sanyou Mei
+- **abs:** https://arxiv.org/abs/2512.22909 · **PDF:** https://arxiv.org/pdf/2512.22909v2
+
+## 25335. SciNets: Graph-Constrained Multi-Hop Reasoning for Scientific Literature Synthesis
+
+- **ID:** `2601.09727` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Sauhard Dubey
+- **abs:** https://arxiv.org/abs/2601.09727 · **PDF:** https://arxiv.org/pdf/2601.09727v1
+
+## 25336. MetaCD: A Meta Learning Framework for Cognitive Diagnosis based on Continual Learning
+
+- **ID:** `2512.22904` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Jin Wu, Chanjin Zheng
+- **abs:** https://arxiv.org/abs/2512.22904 · **PDF:** https://arxiv.org/pdf/2512.22904v1
+
+## 25337. Debugging Tabular Log as Dynamic Graphs
+
+- **ID:** `2512.22903` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Chumeng Liang, Zhanyang Jin, Zahaib Akhtar, et al.
+- **abs:** https://arxiv.org/abs/2512.22903 · **PDF:** https://arxiv.org/pdf/2512.22903v1
+
+## 25338. A Neural Network-Based Real-time Casing Collar Recognition System for Downhole Instruments
+
+- **ID:** `2512.22901` · **primary:** `eess.SY` · **Submitted:** 2025-12-28
+- **Authors:** Si-Yu Xiao, Xin-Di Zhao, Xiang-Zhan Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.22901 · **PDF:** https://arxiv.org/pdf/2512.22901v2
+
+## 25339. HiSciBench: A Hierarchical Multi-disciplinary Benchmark for Scientific Intelligence from Reading to Discovery
+
+- **ID:** `2512.22899` · **primary:** `cs.AI` · **Submitted:** 2025-12-28
+- **Authors:** Yaping Zhang, Qixuan Zhang, Xingquan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.22899 · **PDF:** https://arxiv.org/pdf/2512.22899v1
+
+## 25340. Federated Multi-Task Clustering
+
+- **ID:** `2512.22897` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Suyan Dai, Gan Sun, Fazeng Li, et al.
+- **abs:** https://arxiv.org/abs/2512.22897 · **PDF:** https://arxiv.org/pdf/2512.22897v2
+
+## 25341. SAMP-HDRL: Segmented Allocation with Momentum-Adjusted Utility for Multi-agent Portfolio Management via Hierarchical Deep Reinforcement Learning
+
+- **ID:** `2512.22895` · **primary:** `cs.AI` · **Submitted:** 2025-12-28
+- **Authors:** Xiaotian Ren, Nuerxiati Abudurexiti, Zhengyong Jiang, et al.
+- **abs:** https://arxiv.org/abs/2512.22895 · **PDF:** https://arxiv.org/pdf/2512.22895v1
+
+## 25342. DECEPTICON: How Dark Patterns Manipulate Web Agents
+
+- **ID:** `2512.22894` · **primary:** `cs.CR` · **Submitted:** 2025-12-28
+- **Authors:** Phil Cuvin, Hao Zhu, Diyi Yang
+- **abs:** https://arxiv.org/abs/2512.22894 · **PDF:** https://arxiv.org/pdf/2512.22894v2
+
+## 25343. Theory and Algorithms for Learning with Multi-Class Abstention and Multi-Expert Deferral
+
+- **ID:** `2512.22886` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Anqi Mao
+- **abs:** https://arxiv.org/abs/2512.22886 · **PDF:** https://arxiv.org/pdf/2512.22886v1
+
+## 25344. Agentic AI for Cyber Resilience: A New Security Paradigm and Its System-Theoretic Foundations
+
+- **ID:** `2512.22883` · **primary:** `cs.CR` · **Submitted:** 2025-12-28
+- **Authors:** Tao Li, Quanyan Zhu
+- **abs:** https://arxiv.org/abs/2512.22883 · **PDF:** https://arxiv.org/pdf/2512.22883v1
+
+## 25345. Fundamental Novel Consistency Theory: $H$-Consistency Bounds
+
+- **ID:** `2512.22880` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Yutao Zhong
+- **abs:** https://arxiv.org/abs/2512.22880 · **PDF:** https://arxiv.org/pdf/2512.22880v1
+
+## 25346. SwinTF3D: A Lightweight Multimodal Fusion Approach for Text-Guided 3D Medical Image Segmentation
+
+- **ID:** `2512.22878` · **primary:** `cs.CV` · **Submitted:** 2025-12-28
+- **Authors:** Hasan Faraz Khan, Noor Fatima, Muzammil Behzad
+- **abs:** https://arxiv.org/abs/2512.22878 · **PDF:** https://arxiv.org/pdf/2512.22878v1
+
+## 25347. Reinforcement Networks: novel framework for collaborative Multi-Agent Reinforcement Learning tasks
+
+- **ID:** `2512.22876` · **primary:** `cs.MA` · **Submitted:** 2025-12-28
+- **Authors:** Maksim Kryzhanovskiy, Svetlana Glazyrina, Roman Ischenko, et al.
+- **abs:** https://arxiv.org/abs/2512.22876 · **PDF:** https://arxiv.org/pdf/2512.22876v1
+
+## 25348. The body is not there to compute: Comment on "Informational embodiment: Computational role of information structure in codes and robots" by Pitti et al
+
+- **ID:** `2512.22868` · **primary:** `cs.RO` · **Submitted:** 2025-12-28
+- **Authors:** Matej Hoffmann
+- **abs:** https://arxiv.org/abs/2512.22868 · **PDF:** https://arxiv.org/pdf/2512.22868v1
+
+## 25349. Forgetting as a Feature: Cognitive Alignment of Large Language Models
+
+- **ID:** `2601.09726` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Hien Tran, Quinten Steenhuis, Alexandros Christoforos, et al.
+- **abs:** https://arxiv.org/abs/2601.09726 · **PDF:** https://arxiv.org/pdf/2601.09726v1
+
+## 25350. Adaptive Trust Consensus for Blockchain IoT: Comparing RL, DRL, and MARL Against Naive, Collusive, Adaptive, Byzantine, and Sleeper Attacks
+
+- **ID:** `2512.22860` · **primary:** `cs.CR` · **Submitted:** 2025-12-28
+- **Authors:** Soham Padia, Dhananjay Vaidya, Ramchandra Mangrulkar
+- **abs:** https://arxiv.org/abs/2512.22860 · **PDF:** https://arxiv.org/pdf/2512.22860v1
+
+## 25351. AutoForge: Automated Environment Synthesis for Agentic Reinforcement Learning
+
+- **ID:** `2512.22857` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Shihao Cai, Runnan Fang, Jialong Wu, et al.
+- **abs:** https://arxiv.org/abs/2512.22857 · **PDF:** https://arxiv.org/pdf/2512.22857v1
+
+## 25352. ByteLoom: Weaving Geometry-Consistent Human-Object Interactions through Progressive Curriculum Learning
+
+- **ID:** `2512.22854` · **primary:** `cs.CV` · **Submitted:** 2025-12-28
+- **Authors:** Bangya Liu, Xinyu Gong, Zelin Zhao, et al.
+- **abs:** https://arxiv.org/abs/2512.22854 · **PDF:** https://arxiv.org/pdf/2512.22854v2
+
+## 25353. Causal-Policy Forest for End-to-End Policy Learning
+
+- **ID:** `2512.22846` · **primary:** `econ.EM` · **Submitted:** 2025-12-28
+- **Authors:** Masahiro Kato
+- **abs:** https://arxiv.org/abs/2512.22846 · **PDF:** https://arxiv.org/pdf/2512.22846v1
+
+## 25354. FasterPy: An LLM-based Code Execution Efficiency Optimization Framework
+
+- **ID:** `2512.22827` · **primary:** `cs.SE` · **Submitted:** 2025-12-28
+- **Authors:** Yue Wu, Minghao Han, Ruiyin Li, et al.
+- **abs:** https://arxiv.org/abs/2512.22827 · **PDF:** https://arxiv.org/pdf/2512.22827v1
+
+## 25355. TEAS: Trusted Educational AI Standard: A Framework for Verifiable, Stable, Auditable, and Pedagogically Sound Learning Systems
+
+- **ID:** `2601.06066` · **primary:** `cs.CY` · **Submitted:** 2025-12-28
+- **Authors:** Abu Syed
+- **abs:** https://arxiv.org/abs/2601.06066 · **PDF:** https://arxiv.org/pdf/2601.06066v1
+
+## 25356. TEACH: Temporal Variance-Driven Curriculum for Reinforcement Learning
+
+- **ID:** `2512.22824` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Gaurav Chaudhary, Laxmidhar Behera
+- **abs:** https://arxiv.org/abs/2512.22824 · **PDF:** https://arxiv.org/pdf/2512.22824v1
+
+## 25357. NepEMO: A Multi-Label Emotion and Sentiment Analysis on Nepali Reddit with Linguistic Insights and Temporal Trends
+
+- **ID:** `2512.22823` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Sameer Sitoula, Tej Bahadur Shahi, Laxmi Prasad Bhatt, et al.
+- **abs:** https://arxiv.org/abs/2512.22823 · **PDF:** https://arxiv.org/pdf/2512.22823v1
+
+## 25358. Long-Range Distillation: Distilling 10,000 Years of Simulated Climate into Long Timestep AI Weather Models
+
+- **ID:** `2512.22814` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Scott A. Martin, Noah Brenowitz, Dale Durran, et al.
+- **abs:** https://arxiv.org/abs/2512.22814 · **PDF:** https://arxiv.org/pdf/2512.22814v1
+
+## 25359. EgoReAct: Egocentric Video-Driven 3D Human Reaction Generation
+
+- **ID:** `2512.22808` · **primary:** `cs.CV` · **Submitted:** 2025-12-28
+- **Authors:** Libo Zhang, Zekun Li, Tianyu Li, et al.
+- **abs:** https://arxiv.org/abs/2512.22808 · **PDF:** https://arxiv.org/pdf/2512.22808v2
+
+## 25360. Assessing and Improving Punctuation Robustness in English-Marathi Machine Translation
+
+- **ID:** `2601.09725` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Kaustubh Shivshankar Shejole, Sourabh Deoghare, Pushpak Bhattacharyya
+- **abs:** https://arxiv.org/abs/2601.09725 · **PDF:** https://arxiv.org/pdf/2601.09725v3
+
+## 25361. MoR: Mixture Of Representations For Mixed-Precision Training
+
+- **ID:** `2512.22804` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Bor-Yiing Su, Peter Dykas, Mike Chrzanowski, et al.
+- **abs:** https://arxiv.org/abs/2512.22804 · **PDF:** https://arxiv.org/pdf/2512.22804v1
+
+## 25362. ReDiF: Reinforced Distillation for Few Step Diffusion
+
+- **ID:** `2512.22802` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Amirhossein Tighkhorshid, Zahra Dehghanian, Gholamali Aminian, et al.
+- **abs:** https://arxiv.org/abs/2512.22802 · **PDF:** https://arxiv.org/pdf/2512.22802v1
+
+## 25363. Building AI Agents to Improve Job Referral Requests to Strangers
+
+- **ID:** `2601.10726` · **primary:** `cs.AI` · **Submitted:** 2025-12-28
+- **Authors:** Ross Chu, Yuting Huang
+- **abs:** https://arxiv.org/abs/2601.10726 · **PDF:** https://arxiv.org/pdf/2601.10726v1
+
+## 25364. On the use of case estimate and transactional payment data in neural networks for individual loss reserving
+
+- **ID:** `2601.05274` · **primary:** `q-fin.ST` · **Submitted:** 2025-12-28
+- **Authors:** Benjamin Avanzi, Matthew Lambrianidis, Greg Taylor, et al.
+- **abs:** https://arxiv.org/abs/2601.05274 · **PDF:** https://arxiv.org/pdf/2601.05274v1
+
+## 25365. CNSight: Evaluation of Clinical Note Segmentation Tools
+
+- **ID:** `2512.22795` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Risha Surana, Adrian Law, Sunwoo Kim, et al.
+- **abs:** https://arxiv.org/abs/2512.22795 · **PDF:** https://arxiv.org/pdf/2512.22795v1
+
+## 25366. MixRx: Predicting Drug Combination Interactions with LLMs
+
+- **ID:** `2601.03277` · **primary:** `q-bio.OT` · **Submitted:** 2025-12-28
+- **Authors:** Risha Surana, Cameron Saidock, Hugo Chacon
+- **abs:** https://arxiv.org/abs/2601.03277 · **PDF:** https://arxiv.org/pdf/2601.03277v1
+
+## 25367. Reach-Avoid Differential game with Reachability Analysis for UAVs: A decomposition approach
+
+- **ID:** `2512.22793` · **primary:** `eess.SY` · **Submitted:** 2025-12-28
+- **Authors:** Minh Bui, Simon Monckton, Mo Chen
+- **abs:** https://arxiv.org/abs/2512.22793 · **PDF:** https://arxiv.org/pdf/2512.22793v1
+
+## 25368. A Universal and Robust Framework for Multiple Gas Recognition Based-on Spherical Normalization-Coupled Mahalanobis Algorithm
+
+- **ID:** `2512.22792` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Shuai Chen, Yang Song, Chen Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.22792 · **PDF:** https://arxiv.org/pdf/2512.22792v2
+
+## 25369. Discovering Transmission Dynamics of COVID-19 in China
+
+- **ID:** `2512.22787` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Zhou Yang, Edward Dougherty, Chen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.22787 · **PDF:** https://arxiv.org/pdf/2512.22787v1
+
+## 25370. Nonlinear Dynamical Modeling of Human Intracranial Brain Activity with Flexible Inference
+
+- **ID:** `2512.22785` · **primary:** `q-bio.NC` · **Submitted:** 2025-12-28
+- **Authors:** Kiarash Vaziri, Lucine L. Oganesian, HyeongChan Jo, et al.
+- **abs:** https://arxiv.org/abs/2512.22785 · **PDF:** https://arxiv.org/pdf/2512.22785v1
+
+## 25371. Fake News Classification in Urdu: A Domain Adaptation Approach for a Low-Resource Language
+
+- **ID:** `2512.22778` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Muhammad Zain Ali, Bernhard Pfahringer, Tony Smith
+- **abs:** https://arxiv.org/abs/2512.22778 · **PDF:** https://arxiv.org/pdf/2512.22778v1
+
+## 25372. Adapting, Fast and Slow: Transportable Circuits for Few-Shot Learning
+
+- **ID:** `2512.22777` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Kasra Jalaldoust, Elias Bareinboim
+- **abs:** https://arxiv.org/abs/2512.22777 · **PDF:** https://arxiv.org/pdf/2512.22777v1
+
+## 25373. Schrodinger AI: A Unified Spectral-Dynamical Framework for Classification, Reasoning, and Operator-Based Generalization
+
+- **ID:** `2512.22774` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Truong Son Nguyen
+- **abs:** https://arxiv.org/abs/2512.22774 · **PDF:** https://arxiv.org/pdf/2512.22774v1
+
+## 25374. GRExplainer: A Universal Explanation Method for Temporal Graph Neural Networks
+
+- **ID:** `2512.22772` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Xuyan Li, Jie Wang, Zheng Yan
+- **abs:** https://arxiv.org/abs/2512.22772 · **PDF:** https://arxiv.org/pdf/2512.22772v1
+
+## 25375. Next Best View Selections for Semantic and Dynamic 3D Gaussian Splatting
+
+- **ID:** `2512.22771` · **primary:** `cs.CV` · **Submitted:** 2025-12-28
+- **Authors:** Yiqian Li, Wen Jiang, Kostas Daniilidis
+- **abs:** https://arxiv.org/abs/2512.22771 · **PDF:** https://arxiv.org/pdf/2512.22771v1
+
+## 25376. Understanding the Mechanisms of Fast Hyperparameter Transfer
+
+- **ID:** `2512.22768` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Nikhil Ghosh, Denny Wu, Alberto Bietti
+- **abs:** https://arxiv.org/abs/2512.22768 · **PDF:** https://arxiv.org/pdf/2512.22768v1
+
+## 25377. A Micro-Macro Machine Learning Framework for Predicting Childhood Obesity Risk Using NHANES and Environmental Determinants
+
+- **ID:** `2512.22758` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Eswarasanthosh Kumar Mamillapalli, Nishtha Sharma
+- **abs:** https://arxiv.org/abs/2512.22758 · **PDF:** https://arxiv.org/pdf/2512.22758v1
+
+## 25378. Active Constraint Learning in High Dimensions from Demonstrations
+
+- **ID:** `2512.22757` · **primary:** `cs.RO` · **Submitted:** 2025-12-28
+- **Authors:** Zheng Qiu, Chih-Yuan Chiu, Glen Chou
+- **abs:** https://arxiv.org/abs/2512.22757 · **PDF:** https://arxiv.org/pdf/2512.22757v1
+
+## 25379. From Confounding to Learning: Dynamic Service Fee Pricing on Third-Party Platforms
+
+- **ID:** `2512.22749` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Rui Ai, David Simchi-Levi, Feng Zhu
+- **abs:** https://arxiv.org/abs/2512.22749 · **PDF:** https://arxiv.org/pdf/2512.22749v1
+
+## 25380. Bridging Global Intent with Local Details: A Hierarchical Representation Approach for Semantic Validation in Text-to-SQL
+
+- **ID:** `2512.22744` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Rihong Qiu, Zhibang Yang, Xinke Jiang, et al.
+- **abs:** https://arxiv.org/abs/2512.22744 · **PDF:** https://arxiv.org/pdf/2512.22744v1
+
+## 25381. Robust LLM-based Column Type Annotation via Prompt Augmentation with LoRA Tuning
+
+- **ID:** `2512.22742` · **primary:** `cs.DB` · **Submitted:** 2025-12-28
+- **Authors:** Hanze Meng, Jianhao Cao, Rachel Pottinger
+- **abs:** https://arxiv.org/abs/2512.22742 · **PDF:** https://arxiv.org/pdf/2512.22742v1
+
+## 25382. Text-Routed Sparse Mixture-of-Experts Model with Explanation and Temporal Alignment for Multi-Modal Sentiment Analysis
+
+- **ID:** `2512.22741` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Dongning Rao, Yunbiao Zeng, Zhihua Jiang, et al.
+- **abs:** https://arxiv.org/abs/2512.22741 · **PDF:** https://arxiv.org/pdf/2512.22741v1
+
+## 25383. Multi-Task Learning for Metal Alloy Property Prediction: An Empirical Study of Negative Transfer and Mitigation Strategies
+
+- **ID:** `2512.22740` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Sungwoo Kang
+- **abs:** https://arxiv.org/abs/2512.22740 · **PDF:** https://arxiv.org/pdf/2512.22740v3
+
+## 25384. Harnessing Large Language Models for Biomedical Named Entity Recognition
+
+- **ID:** `2512.22738` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Jian Chen, Leilei Su, Cong Sun
+- **abs:** https://arxiv.org/abs/2512.22738 · **PDF:** https://arxiv.org/pdf/2512.22738v1
+
+## 25385. WeDLM: Reconciling Diffusion Language Models with Standard Causal Attention for Fast Inference
+
+- **ID:** `2512.22737` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Aiwei Liu, Minghua He, Shaoxun Zeng, et al.
+- **abs:** https://arxiv.org/abs/2512.22737 · **PDF:** https://arxiv.org/pdf/2512.22737v1
+
+## 25386. FoldAct: Efficient and Stable Context Folding for Long-Horizon Search Agents
+
+- **ID:** `2512.22733` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Jiaqi Shao, Yufeng Miao, Wei Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.22733 · **PDF:** https://arxiv.org/pdf/2512.22733v1
+
+## 25387. Data Augmentation for Classification of Negative Pregnancy Outcomes in Imbalanced Data
+
+- **ID:** `2512.22732` · **primary:** `cs.CL` · **Submitted:** 2025-12-28
+- **Authors:** Md Badsha Biswas
+- **abs:** https://arxiv.org/abs/2512.22732 · **PDF:** https://arxiv.org/pdf/2512.22732v1
+
+## 25388. Improved cystic hygroma detection from prenatal imaging using ultrasound-specific self-supervised representation learning
+
+- **ID:** `2512.22730` · **primary:** `cs.CV` · **Submitted:** 2025-12-28
+- **Authors:** Youssef Megahed, Robin Ducharme, Inok Lee, et al.
+- **abs:** https://arxiv.org/abs/2512.22730 · **PDF:** https://arxiv.org/pdf/2512.22730v2
+
+## 25389. Enabling Long FFT Convolutions on Memory-Constrained FPGAs via Chunking
+
+- **ID:** `2601.06065` · **primary:** `cs.LG` · **Submitted:** 2025-12-28
+- **Authors:** Peter Wang, Neelesh Gupta, Viktor Prasanna
+- **abs:** https://arxiv.org/abs/2601.06065 · **PDF:** https://arxiv.org/pdf/2601.06065v1
+
+## 25390. Mitigating Social Desirability Bias in Random Silicon Sampling
+
+- **ID:** `2512.22725` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Sashank Chapala, Maksym Mironov, Songgaojun Deng
+- **abs:** https://arxiv.org/abs/2512.22725 · **PDF:** https://arxiv.org/pdf/2512.22725v1
+
+## 25391. Memento 2: Learning by Stateful Reflective Memory
+
+- **ID:** `2512.22716` · **primary:** `cs.AI` · **Submitted:** 2025-12-27
+- **Authors:** Jun Wang
+- **abs:** https://arxiv.org/abs/2512.22716 · **PDF:** https://arxiv.org/pdf/2512.22716v3
+
+## 25392. Beg to Differ: Understanding Reasoning-Answer Misalignment Across Languages
+
+- **ID:** `2512.22712` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Anaelia Ovalle, Candace Ross, Sebastian Ruder, et al.
+- **abs:** https://arxiv.org/abs/2512.22712 · **PDF:** https://arxiv.org/pdf/2512.22712v3
+
+## 25393. GHaLIB: A Multilingual Framework for Hope Speech Detection in Low-Resource Languages
+
+- **ID:** `2512.22705` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Ahmed Abdullah, Sana Fatima, Haroon Mahmood
+- **abs:** https://arxiv.org/abs/2512.22705 · **PDF:** https://arxiv.org/pdf/2512.22705v1
+
+## 25394. What Matters in Deep Learning for Time Series Forecasting?
+
+- **ID:** `2512.22702` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Valentina Moretti, Andrea Cini, Ivan Marisca, et al.
+- **abs:** https://arxiv.org/abs/2512.22702 · **PDF:** https://arxiv.org/pdf/2512.22702v1
+
+## 25395. Predictive Modeling of Power Outages during Extreme Events: Integrating Weather and Socio-Economic Factors
+
+- **ID:** `2512.22699` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Nina Fatehi, Antar Kumar Biswas, Masoud H. Nazari
+- **abs:** https://arxiv.org/abs/2512.22699 · **PDF:** https://arxiv.org/pdf/2512.22699v3
+
+## 25396. Canonical correlation regression with noisy data
+
+- **ID:** `2512.22697` · **primary:** `econ.EM` · **Submitted:** 2025-12-27
+- **Authors:** Isaac Meza, Rahul Singh
+- **abs:** https://arxiv.org/abs/2512.22697 · **PDF:** https://arxiv.org/pdf/2512.22697v1
+
+## 25397. Learning with the $p$-adics
+
+- **ID:** `2512.22692` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** André F. T. Martins
+- **abs:** https://arxiv.org/abs/2512.22692 · **PDF:** https://arxiv.org/pdf/2512.22692v1
+
+## 25398. Multimodal Diffeomorphic Registration with Neural ODEs and Structural Descriptors
+
+- **ID:** `2512.22689` · **primary:** `cs.CV` · **Submitted:** 2025-12-27
+- **Authors:** Salvador Rodriguez-Sanz, Monica Hernandez
+- **abs:** https://arxiv.org/abs/2512.22689 · **PDF:** https://arxiv.org/pdf/2512.22689v1
+
+## 25399. Conformal Prediction Sets for Next-Token Prediction in Large Language Models: Balancing Coverage Guarantees with Set Efficiency
+
+- **ID:** `2512.22682` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Yoshith Roy Kotla, Varshith Roy Kotla
+- **abs:** https://arxiv.org/abs/2512.22682 · **PDF:** https://arxiv.org/pdf/2512.22682v1
+
+## 25400. Beyond Centralization: Provable Communication Efficient Decentralized Multi-Task Learning
+
+- **ID:** `2512.22675` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Donghwa Kang, Shana Moothedath
+- **abs:** https://arxiv.org/abs/2512.22675 · **PDF:** https://arxiv.org/pdf/2512.22675v1
+
+## 25401. TravelBench: A Broader Real-World Benchmark for Multi-Turn and Tool-Using Travel Planning
+
+- **ID:** `2512.22673` · **primary:** `cs.AI` · **Submitted:** 2025-12-27
+- **Authors:** Xiang Cheng, Yulan Hu, Xiangwen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.22673 · **PDF:** https://arxiv.org/pdf/2512.22673v2
+
+## 25402. Quantum Generative Models for Computational Fluid Dynamics: A First Exploration of Latent Space Learning in Lattice Boltzmann Simulations
+
+- **ID:** `2512.22672` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Achraf Hsain, Fouad Mohammed Abbou
+- **abs:** https://arxiv.org/abs/2512.22672 · **PDF:** https://arxiv.org/pdf/2512.22672v1
+
+## 25403. Fragile Knowledge, Robust Instruction-Following: The Width Pruning Dichotomy in Llama-3.2
+
+- **ID:** `2512.22671` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Pere Martra
+- **abs:** https://arxiv.org/abs/2512.22671 · **PDF:** https://arxiv.org/pdf/2512.22671v1
+
+## 25404. Syntactic Framing Fragility: An Audit of Robustness in LLM Ethical Decisions
+
+- **ID:** `2601.09724` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Katherine Elkins, Jon Chun
+- **abs:** https://arxiv.org/abs/2601.09724 · **PDF:** https://arxiv.org/pdf/2601.09724v2
+
+## 25405. INTERACT-CMIL: Multi-Task Shared Learning and Inter-Task Consistency for Conjunctival Melanocytic Intraepithelial Lesion Grading
+
+- **ID:** `2512.22666` · **primary:** `cs.CV` · **Submitted:** 2025-12-27
+- **Authors:** Mert Ikinci, Luna Toma, Karin U. Loeffler, et al.
+- **abs:** https://arxiv.org/abs/2512.22666 · **PDF:** https://arxiv.org/pdf/2512.22666v1
+
+## 25406. Unleashing Foundation Vision Models: Adaptive Transfer for Diverse Data-Limited Scientific Domains
+
+- **ID:** `2512.22664` · **primary:** `cs.CV` · **Submitted:** 2025-12-27
+- **Authors:** Qiankun Li, Feng He, Huabao Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.22664 · **PDF:** https://arxiv.org/pdf/2512.22664v1
+
+## 25407. Machine learning models for predicting catastrophe bond coupons using climate data
+
+- **ID:** `2512.22660` · **primary:** `q-fin.PR` · **Submitted:** 2025-12-27
+- **Authors:** Julia Kończal, Michał Balcerek, Krzysztof Burnecki
+- **abs:** https://arxiv.org/abs/2512.22660 · **PDF:** https://arxiv.org/pdf/2512.22660v1
+
+## 25408. Clinically Calibrated Machine Learning Benchmarks for Large-Scale Multi-Disorder EEG Classification
+
+- **ID:** `2512.22656` · **primary:** `cs.HC` · **Submitted:** 2025-12-27
+- **Authors:** Argha Kamal Samanta, Deepak Mewada, Monalisa Sarma, et al.
+- **abs:** https://arxiv.org/abs/2512.22656 · **PDF:** https://arxiv.org/pdf/2512.22656v1
+
+## 25409. Investigating Deep Learning Models for Ejection Fraction Estimation from Echocardiography Videos
+
+- **ID:** `2512.22657` · **primary:** `cs.CV` · **Submitted:** 2025-12-27
+- **Authors:** Shravan Saranyan, Pramit Saha
+- **abs:** https://arxiv.org/abs/2512.22657 · **PDF:** https://arxiv.org/pdf/2512.22657v1
+
+## 25410. Scaling Unverifiable Rewards: A Case Study on Visual Insights
+
+- **ID:** `2512.22650` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Shuyu Gan, James Mooney, Pan Hao, et al.
+- **abs:** https://arxiv.org/abs/2512.22650 · **PDF:** https://arxiv.org/pdf/2512.22650v1
+
+## 25411. Leveraging Machine Learning for Early Detection of Lung Diseases
+
+- **ID:** `2512.23757` · **primary:** `eess.IV` · **Submitted:** 2025-12-27
+- **Authors:** Bahareh Rahmani, Harsha Reddy Bindela, Rama Kanth Reddy Gosula, et al.
+- **abs:** https://arxiv.org/abs/2512.23757 · **PDF:** https://arxiv.org/pdf/2512.23757v1
+
+## 25412. Tree Meets Transformer: A Hybrid Architecture for Scalable Power Allocation in Cell-Free Networks
+
+- **ID:** `2512.22639` · **primary:** `eess.SY` · **Submitted:** 2025-12-27
+- **Authors:** Irched Chafaa, Giacomo Bacci, Luca Sanguinetti
+- **abs:** https://arxiv.org/abs/2512.22639 · **PDF:** https://arxiv.org/pdf/2512.22639v1
+
+## 25413. Likelihood-Preserving Embeddings for Statistical Inference
+
+- **ID:** `2512.22638` · **primary:** `stat.ML` · **Submitted:** 2025-12-27
+- **Authors:** Deniz Akdemir
+- **abs:** https://arxiv.org/abs/2512.22638 · **PDF:** https://arxiv.org/pdf/2512.22638v1
+
+## 25414. Evaluating GRPO and DPO for Faithful Chain-of-Thought Reasoning in LLMs
+
+- **ID:** `2512.22631` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Hadi Mohammadi, Tamas Kozak, Anastasia Giachanou
+- **abs:** https://arxiv.org/abs/2512.22631 · **PDF:** https://arxiv.org/pdf/2512.22631v1
+
+## 25415. On the Role of Discreteness in Diffusion LLMs
+
+- **ID:** `2512.22630` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Ziqi Jin, Bin Wang, Xiang Lin, et al.
+- **abs:** https://arxiv.org/abs/2512.22630 · **PDF:** https://arxiv.org/pdf/2512.22630v1
+
+## 25416. DICE: Discrete Interpretable Comparative Evaluation with Probabilistic Scoring for Retrieval-Augmented Generation
+
+- **ID:** `2512.22629` · **primary:** `cs.AI` · **Submitted:** 2025-12-27
+- **Authors:** Shiyan Liu, Jian Ma, Rui Qu
+- **abs:** https://arxiv.org/abs/2512.22629 · **PDF:** https://arxiv.org/pdf/2512.22629v1
+
+## 25417. M2G-Eval: Enhancing and Evaluating Multi-granularity Multilingual Code Generation
+
+- **ID:** `2512.22628` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Fanglin Xu, Wei Zhang, Jian Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.22628 · **PDF:** https://arxiv.org/pdf/2512.22628v1
+
+## 25418. Topic Segmentation Using Generative Language Models
+
+- **ID:** `2601.03276` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Pierre Mackenzie, Maya Shah, Patrick Frenett
+- **abs:** https://arxiv.org/abs/2601.03276 · **PDF:** https://arxiv.org/pdf/2601.03276v1
+
+## 25419. Chain-of-thought Reviewing and Correction for Time Series Question Answering
+
+- **ID:** `2512.22627` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Chen Su, Yuanhe Tian, Yan Song
+- **abs:** https://arxiv.org/abs/2512.22627 · **PDF:** https://arxiv.org/pdf/2512.22627v1
+
+## 25420. The Wisdom of Deliberating AI Crowds: Does Deliberation Improve LLM-Based Forecasting?
+
+- **ID:** `2512.22625` · **primary:** `cs.AI` · **Submitted:** 2025-12-27
+- **Authors:** Paul Schneider, Amalie Schramm
+- **abs:** https://arxiv.org/abs/2512.22625 · **PDF:** https://arxiv.org/pdf/2512.22625v1
+
+## 25421. SLO-Conditioned Action Routing for Retrieval-Augmented Generation: Objective Ablation and Failure Modes
+
+- **ID:** `2601.00841` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Bharath Nunepalli
+- **abs:** https://arxiv.org/abs/2601.00841 · **PDF:** https://arxiv.org/pdf/2601.00841v1
+
+## 25422. Communication Compression for Distributed Learning with Aggregate and Server-Guided Feedback
+
+- **ID:** `2512.22623` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Tomas Ortega, Chun-Yin Huang, Xiaoxiao Li, et al.
+- **abs:** https://arxiv.org/abs/2512.22623 · **PDF:** https://arxiv.org/pdf/2512.22623v2
+
+## 25423. Chord Recognition with Deep Learning
+
+- **ID:** `2512.22621` · **primary:** `cs.SD` · **Submitted:** 2025-12-27
+- **Authors:** Pierre Mackenzie
+- **abs:** https://arxiv.org/abs/2512.22621 · **PDF:** https://arxiv.org/pdf/2512.22621v1
+
+## 25424. HINTS: Extraction of Human Insights from Time-Series Without External Sources
+
+- **ID:** `2512.23755` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Sheo Yon Jhin, Noseong Park
+- **abs:** https://arxiv.org/abs/2512.23755 · **PDF:** https://arxiv.org/pdf/2512.23755v1
+
+## 25425. Evolutionary Algorithms for Computing Nash Equilibria in Dynamic Games
+
+- **ID:** `2601.02397` · **primary:** `cs.NE` · **Submitted:** 2025-12-27
+- **Authors:** Alireza Rezaee
+- **abs:** https://arxiv.org/abs/2601.02397 · **PDF:** https://arxiv.org/pdf/2601.02397v1
+
+## 25426. Dream-VL & Dream-VLA: Open Vision-Language and Vision-Language-Action Models with Diffusion Language Model Backbone
+
+- **ID:** `2512.22615` · **primary:** `cs.CV` · **Submitted:** 2025-12-27
+- **Authors:** Jiacheng Ye, Shansan Gong, Jiahui Gao, et al.
+- **abs:** https://arxiv.org/abs/2512.22615 · **PDF:** https://arxiv.org/pdf/2512.22615v2
+
+## 25427. Beyond Isolated Investor: Predicting Startup Success via Roleplay-Based Collective Agents
+
+- **ID:** `2512.22608` · **primary:** `cs.AI` · **Submitted:** 2025-12-27
+- **Authors:** Zhongyang Liu, Haoyu Pei, Xiangyi Xiao, et al.
+- **abs:** https://arxiv.org/abs/2512.22608 · **PDF:** https://arxiv.org/pdf/2512.22608v2
+
+## 25428. Gold Price Prediction Using Long Short-Term Memory and Multi-Layer Perceptron with Gray Wolf Optimizer
+
+- **ID:** `2512.22606` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Hesam Taghipour, Alireza Rezaee, Farshid Hajati
+- **abs:** https://arxiv.org/abs/2512.22606 · **PDF:** https://arxiv.org/pdf/2512.22606v1
+
+## 25429. Learning Multi-Modal Mobility Dynamics for Generalized Next Location Recommendation
+
+- **ID:** `2512.22605` · **primary:** `cs.AI` · **Submitted:** 2025-12-27
+- **Authors:** Junshu Dai, Yu Wang, Tongya Zheng, et al.
+- **abs:** https://arxiv.org/abs/2512.22605 · **PDF:** https://arxiv.org/pdf/2512.22605v1
+
+## 25430. Structured Prompting and LLM Ensembling for Multimodal Conversational Aspect-based Sentiment Analysis
+
+- **ID:** `2512.22603` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Zhiqiang Gao, Shihao Gao, Zixing Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.22603 · **PDF:** https://arxiv.org/pdf/2512.22603v1
+
+## 25431. Tyee: A Unified, Modular, and Fully-Integrated Configurable Toolkit for Intelligent Physiological Health Care
+
+- **ID:** `2512.22601` · **primary:** `cs.AI` · **Submitted:** 2025-12-27
+- **Authors:** Tao Zhou, Lingyu Shu, Zixing Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.22601 · **PDF:** https://arxiv.org/pdf/2512.22601v1
+
+## 25432. AdaFRUGAL: Adaptive Memory-Efficient Training with Dynamic Control
+
+- **ID:** `2601.11568` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Quang-Hung Bui, Anh Son Ta
+- **abs:** https://arxiv.org/abs/2601.11568 · **PDF:** https://arxiv.org/pdf/2601.11568v1
+
+## 25433. Cryptocurrency Price Prediction Using Parallel Gated Recurrent Units
+
+- **ID:** `2512.22599` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Milad Asadpour, Alireza Rezaee, Farshid Hajati
+- **abs:** https://arxiv.org/abs/2512.22599 · **PDF:** https://arxiv.org/pdf/2512.22599v1
+
+## 25434. Energy-Guided Flow Matching Enables Few-Step Conformer Generation and Ground-State Identification
+
+- **ID:** `2512.22597` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Guikun Xu, Xiaohan Yi, Peilin Zhao, et al.
+- **abs:** https://arxiv.org/abs/2512.22597 · **PDF:** https://arxiv.org/pdf/2512.22597v1
+
+## 25435. Data-Driven Analysis of Crash Patterns in SAE Level 2 and Level 4 Automated Vehicles Using K-means Clustering and Association Rule Mining
+
+- **ID:** `2512.22589` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Jewel Rana Palit, Vijayalakshmi K Kumarasamy, Osama A. Osman
+- **abs:** https://arxiv.org/abs/2512.22589 · **PDF:** https://arxiv.org/pdf/2512.22589v2
+
+## 25436. Structural Incompatibility of Differentiable Sorting and Within-Vector Rank Normalization
+
+- **ID:** `2512.22587` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Taeyun Kim
+- **abs:** https://arxiv.org/abs/2512.22587 · **PDF:** https://arxiv.org/pdf/2512.22587v2
+
+## 25437. SANet: A Semantic-aware Agentic AI Networking Framework for Cross-layer Optimization in 6G
+
+- **ID:** `2512.22579` · **primary:** `cs.AI` · **Submitted:** 2025-12-27
+- **Authors:** Yong Xiao, Xubo Li, Haoran Zhou, et al.
+- **abs:** https://arxiv.org/abs/2512.22579 · **PDF:** https://arxiv.org/pdf/2512.22579v1
+
+## 25438. SagaScale: A Realistic, Scalable, and High-Quality Long-Context Benchmark Built from Full-Length Novels
+
+- **ID:** `2601.09723` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Guancheng Du, Yong Hu, Wenqing Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.09723 · **PDF:** https://arxiv.org/pdf/2601.09723v1
+
+## 25439. Lessons from Neuroscience for AI: How integrating Actions, Compositional Structure and Episodic Memory could enable Safe, Interpretable and Human-Like AI
+
+- **ID:** `2512.22568` · **primary:** `cs.AI` · **Submitted:** 2025-12-27
+- **Authors:** Rajesh P. N. Rao, Vishwas Sathish, Linxing Preston Jiang, et al.
+- **abs:** https://arxiv.org/abs/2512.22568 · **PDF:** https://arxiv.org/pdf/2512.22568v1
+
+## 25440. Geometry-Aware Optimization for Respiratory Sound Classification: Enhancing Sensitivity with SAM-Optimized Audio Spectrogram Transformers
+
+- **ID:** `2512.22564` · **primary:** `eess.AS` · **Submitted:** 2025-12-27
+- **Authors:** Atakan Işık, Selin Vulga Işık, Ahmet Feridun Işık, et al.
+- **abs:** https://arxiv.org/abs/2512.22564 · **PDF:** https://arxiv.org/pdf/2512.22564v1
+
+## 25441. Generalized Regularized Evidential Deep Learning Models: Theory and Comprehensive Evaluation
+
+- **ID:** `2512.23753` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Deep Shankar Pandey, Hyomin Choi, Qi Yu
+- **abs:** https://arxiv.org/abs/2512.23753 · **PDF:** https://arxiv.org/pdf/2512.23753v1
+
+## 25442. Learning When Not to Attend Globally
+
+- **ID:** `2512.22562` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Xuan Luo, Kailai Zhang, Xifeng Yan
+- **abs:** https://arxiv.org/abs/2512.22562 · **PDF:** https://arxiv.org/pdf/2512.22562v1
+
+## 25443. RollArt: Scaling Agentic RL Training via Disaggregated Infrastructure
+
+- **ID:** `2512.22560` · **primary:** `cs.DC` · **Submitted:** 2025-12-27
+- **Authors:** Wei Gao, Yuheng Zhao, Tianyuan Wu, et al.
+- **abs:** https://arxiv.org/abs/2512.22560 · **PDF:** https://arxiv.org/pdf/2512.22560v1
+
+## 25444. Computing Pure-Strategy Nash Equilibria in a Two-Party Policy Competition: Existence and Algorithmic Approaches
+
+- **ID:** `2512.22552` · **primary:** `cs.GT` · **Submitted:** 2025-12-27
+- **Authors:** Chuang-Chieh Lin, Chi-Jen Lu, Po-An Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.22552 · **PDF:** https://arxiv.org/pdf/2512.22552v4
+
+## 25445. TimePerceiver: An Encoder-Decoder Framework for Generalized Time-Series Forecasting
+
+- **ID:** `2512.22550` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Jaebin Lee, Hankook Lee
+- **abs:** https://arxiv.org/abs/2512.22550 · **PDF:** https://arxiv.org/pdf/2512.22550v1
+
+## 25446. Self-Rewarded Multimodal Coherent Reasoning Across Diverse Visual Domains
+
+- **ID:** `2512.22545` · **primary:** `cs.CV` · **Submitted:** 2025-12-27
+- **Authors:** Jesen Zhang, Ningyuan Liu, Kaitong Cai, et al.
+- **abs:** https://arxiv.org/abs/2512.22545 · **PDF:** https://arxiv.org/pdf/2512.22545v1
+
+## 25447. ADMEDTAGGER: an annotation framework for distillation of expert knowledge for the Polish medical language
+
+- **ID:** `2601.09722` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Franciszek Górski, Andrzej Czyżewski
+- **abs:** https://arxiv.org/abs/2601.09722 · **PDF:** https://arxiv.org/pdf/2601.09722v1
+
+## 25448. CoAgent: Collaborative Planning and Consistency Agent for Coherent Video Generation
+
+- **ID:** `2512.22536` · **primary:** `cs.CV` · **Submitted:** 2025-12-27
+- **Authors:** Qinglin Zeng, Kaitong Cai, Ruiqi Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.22536 · **PDF:** https://arxiv.org/pdf/2512.22536v1
+
+## 25449. A Global Atlas of Digital Dermatology to Map Innovation and Disparities
+
+- **ID:** `2601.00840` · **primary:** `cs.DL` · **Submitted:** 2025-12-27
+- **Authors:** Fabian Gröger, Simone Lionetti, Philippe Gottfrois, et al.
+- **abs:** https://arxiv.org/abs/2601.00840 · **PDF:** https://arxiv.org/pdf/2601.00840v2
+
+## 25450. Multi-AI Agent Framework Reveals the "Oxide Gatekeeper" in Aluminum Nanoparticle Oxidation
+
+- **ID:** `2512.22529` · **primary:** `cs.AI` · **Submitted:** 2025-12-27
+- **Authors:** Yiming Lu, Tingyu Lu, Di Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.22529 · **PDF:** https://arxiv.org/pdf/2512.22529v1
+
+## 25451. Towards Reliable Evaluation of Adversarial Robustness for Spiking Neural Networks
+
+- **ID:** `2512.22522` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Jihang Wang, Dongcheng Zhao, Ruolin Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.22522 · **PDF:** https://arxiv.org/pdf/2512.22522v3
+
+## 25452. Decomposing Task Vectors for Refined Model Editing
+
+- **ID:** `2512.22511` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Hamed Damirchi, Ehsan Abbasnejad, Zhen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.22511 · **PDF:** https://arxiv.org/pdf/2512.22511v1
+
+## 25453. Predicting LLM Correctness in Prosthodontics Using Metadata and Hallucination Signals
+
+- **ID:** `2512.22508` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Lucky Susanto, Anasta Pranawijayana, Cortino Sukotjo, et al.
+- **abs:** https://arxiv.org/abs/2512.22508 · **PDF:** https://arxiv.org/pdf/2512.22508v1
+
+## 25454. Hierarchical Pedagogical Oversight: A Multi-Agent Adversarial Framework for Reliable AI Tutoring
+
+- **ID:** `2512.22496` · **primary:** `cs.MA` · **Submitted:** 2025-12-27
+- **Authors:** Saisab Sadhu, Ashim Dhor
+- **abs:** https://arxiv.org/abs/2512.22496 · **PDF:** https://arxiv.org/pdf/2512.22496v1
+
+## 25455. The Quest for Winning Tickets in Low-Rank Adapters
+
+- **ID:** `2512.22495` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Hamed Damirchi, Cristian Rodriguez-Opazo, Ehsan Abbasnejad, et al.
+- **abs:** https://arxiv.org/abs/2512.22495 · **PDF:** https://arxiv.org/pdf/2512.22495v1
+
+## 25456. Role-Based Fault Tolerance System for LLM RL Post-Training
+
+- **ID:** `2512.22492` · **primary:** `cs.DC` · **Submitted:** 2025-12-27
+- **Authors:** Zhenqian Chen, Baoquan Zhong, Xiang Li, et al.
+- **abs:** https://arxiv.org/abs/2512.22492 · **PDF:** https://arxiv.org/pdf/2512.22492v1
+
+## 25457. ManchuTTS: Towards High-Quality Manchu Speech Synthesis via Flow Matching and Hierarchical Text Representation
+
+- **ID:** `2512.22491` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Suhua Wang, Zifan Wang, Xiaoxin Sun, et al.
+- **abs:** https://arxiv.org/abs/2512.22491 · **PDF:** https://arxiv.org/pdf/2512.22491v1
+
+## 25458. Toward Real-World IoT Security: Concept Drift-Resilient IoT Botnet Detection via Latent Space Representation Learning and Alignment
+
+- **ID:** `2512.22488` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Hassan Wasswa, Timothy Lynar
+- **abs:** https://arxiv.org/abs/2512.22488 · **PDF:** https://arxiv.org/pdf/2512.22488v1
+
+## 25459. Constituency Structure over Eojeol in Korean Treebanks
+
+- **ID:** `2512.22487` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Jungyeul Park, Chulwoo Park
+- **abs:** https://arxiv.org/abs/2512.22487 · **PDF:** https://arxiv.org/pdf/2512.22487v1
+
+## 25460. SPECTRE: Spectral Pre-training Embeddings with Cylindrical Temporal Rotary Position Encoding for Fine-Grained sEMG-Based Movement Decoding
+
+- **ID:** `2512.22481` · **primary:** `cs.HC` · **Submitted:** 2025-12-27
+- **Authors:** Zihan Weng, Chanlin Yi, Pouya Bashivan, et al.
+- **abs:** https://arxiv.org/abs/2512.22481 · **PDF:** https://arxiv.org/pdf/2512.22481v1
+
+## 25461. Collaborative Optimization of Multiclass Imbalanced Learning: Density-Aware and Region-Guided Boosting
+
+- **ID:** `2512.22478` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Chuantao Li, Zhi Li, Jiahao Xu, et al.
+- **abs:** https://arxiv.org/abs/2512.22478 · **PDF:** https://arxiv.org/pdf/2512.22478v1
+
+## 25462. Gradient Dynamics of Attention: How Cross-Entropy Sculpts Bayesian Manifolds
+
+- **ID:** `2512.22473` · **primary:** `stat.ML` · **Submitted:** 2025-12-27
+- **Authors:** Naman Agarwal, Siddhartha R. Dalal, Vishal Misra
+- **abs:** https://arxiv.org/abs/2512.22473 · **PDF:** https://arxiv.org/pdf/2512.22473v4
+
+## 25463. Geometric Scaling of Bayesian Inference in LLMs
+
+- **ID:** `2512.23752` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Naman Agarwal, Siddhartha R. Dalal, Vishal Misra
+- **abs:** https://arxiv.org/abs/2512.23752 · **PDF:** https://arxiv.org/pdf/2512.23752v4
+
+## 25464. The Bayesian Geometry of Transformer Attention
+
+- **ID:** `2512.22471` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Naman Agarwal, Siddhartha R. Dalal, Vishal Misra
+- **abs:** https://arxiv.org/abs/2512.22471 · **PDF:** https://arxiv.org/pdf/2512.22471v4
+
+## 25465. DarkPatterns-LLM: A Multi-Layer Benchmark for Detecting Manipulative and Harmful AI Behavior
+
+- **ID:** `2512.22470` · **primary:** `cs.AI` · **Submitted:** 2025-12-27
+- **Authors:** Sadia Asif, Israel Antonio Rosales Laguan, Haris Khan, et al.
+- **abs:** https://arxiv.org/abs/2512.22470 · **PDF:** https://arxiv.org/pdf/2512.22470v1
+
+## 25466. GLUE: Gradient-free Learning to Unify Experts
+
+- **ID:** `2512.22467` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Jong-Ik Park, Shreyas Chaudhari, Srinivasa Pranav, et al.
+- **abs:** https://arxiv.org/abs/2512.22467 · **PDF:** https://arxiv.org/pdf/2512.22467v2
+
+## 25467. AMBIT: Augmenting Mobility Baselines with Interpretable Trees
+
+- **ID:** `2512.22466` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Qizhi Wang
+- **abs:** https://arxiv.org/abs/2512.22466 · **PDF:** https://arxiv.org/pdf/2512.22466v1
+
+## 25468. Democratizing GraphRAG: Linear, CPU-Only Graph Retrieval for Multi-Hop QA
+
+- **ID:** `2602.23372` · **primary:** `cs.IR` · **Submitted:** 2025-12-27
+- **Authors:** Qizhi Wang
+- **abs:** https://arxiv.org/abs/2602.23372 · **PDF:** https://arxiv.org/pdf/2602.23372v1
+
+## 25469. AFA-LoRA: Enabling Non-Linear Adaptations in LoRA with Activation Function Annealing
+
+- **ID:** `2512.22455` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Jiacheng Li, Jianchao Tan, Zhidong Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.22455 · **PDF:** https://arxiv.org/pdf/2512.22455v2
+
+## 25470. Towards Robust Optical-SAR Object Detection under Missing Modalities: A Dynamic Quality-Aware Fusion Framework
+
+- **ID:** `2512.22447` · **primary:** `cs.CV` · **Submitted:** 2025-12-27
+- **Authors:** Zhicheng Zhao, Yuancheng Xu, Andong Lu, et al.
+- **abs:** https://arxiv.org/abs/2512.22447 · **PDF:** https://arxiv.org/pdf/2512.22447v1
+
+## 25471. Accounting Reasoning in Large Language Models: Concepts, Evaluation, and Empirical Analysis
+
+- **ID:** `2512.22443` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Jie Zhou, Xin Chen, Jie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.22443 · **PDF:** https://arxiv.org/pdf/2512.22443v2
+
+## 25472. HiFi-RAG: Hierarchical Content Filtering and Two-Pass Generation for Open-Domain RAG
+
+- **ID:** `2512.22442` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Cattalyya Nuengsigkapian
+- **abs:** https://arxiv.org/abs/2512.22442 · **PDF:** https://arxiv.org/pdf/2512.22442v1
+
+## 25473. SuperiorGAT: Graph Attention Networks for Sparse LiDAR Point Cloud Reconstruction in Autonomous Systems
+
+- **ID:** `2512.22439` · **primary:** `cs.CV` · **Submitted:** 2025-12-27
+- **Authors:** Khalfalla Awedat, Mohamed Abidalrekab, Gurcan Comert, et al.
+- **abs:** https://arxiv.org/abs/2512.22439 · **PDF:** https://arxiv.org/pdf/2512.22439v3
+
+## 25474. AnalogSAGE: Self-evolving Analog Design Multi-Agents with Stratified Memory and Grounded Experience
+
+- **ID:** `2512.22435` · **primary:** `cs.AR` · **Submitted:** 2025-12-27
+- **Authors:** Zining Wang, Jian Gao, Weimin Fu, et al.
+- **abs:** https://arxiv.org/abs/2512.22435 · **PDF:** https://arxiv.org/pdf/2512.22435v1
+
+## 25475. Monadic Context Engineering
+
+- **ID:** `2512.22431` · **primary:** `cs.AI` · **Submitted:** 2025-12-27
+- **Authors:** Yifan Zhang, Yang Yuan, Mengdi Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.22431 · **PDF:** https://arxiv.org/pdf/2512.22431v5
+
+## 25476. Causality-Inspired Safe Residual Correction for Multivariate Time Series
+
+- **ID:** `2512.22428` · **primary:** `cs.LG` · **Submitted:** 2025-12-27
+- **Authors:** Jianxiang Xie, Yuncheng Hua, Mingyue Cheng, et al.
+- **abs:** https://arxiv.org/abs/2512.22428 · **PDF:** https://arxiv.org/pdf/2512.22428v2
+
+## 25477. Uncertainty-Aware Flow Field Reconstruction Using SVGP Kolmogorov-Arnold Networks
+
+- **ID:** `2512.22426` · **primary:** `physics.flu-dyn` · **Submitted:** 2025-12-27
+- **Authors:** Y. Sungtaek Ju
+- **abs:** https://arxiv.org/abs/2512.22426 · **PDF:** https://arxiv.org/pdf/2512.22426v1
+
+## 25478. FluenceFormer: Transformer-Driven Multi-Beam Fluence Map Regression for Radiotherapy Planning
+
+- **ID:** `2512.22425` · **primary:** `cs.CV` · **Submitted:** 2025-12-27
+- **Authors:** Ujunwa Mgboh, Rafi Ibn Sultan, Joshua Kim, et al.
+- **abs:** https://arxiv.org/abs/2512.22425 · **PDF:** https://arxiv.org/pdf/2512.22425v2
+
+## 25479. Bright 4B: Scaling Hyperspherical Learning for Segmentation in 3D Brightfield Microscopy
+
+- **ID:** `2512.22423` · **primary:** `cs.CV` · **Submitted:** 2025-12-27
+- **Authors:** Amil Khan, Matheus Palhares Viana, Suraj Mishra, et al.
+- **abs:** https://arxiv.org/abs/2512.22423 · **PDF:** https://arxiv.org/pdf/2512.22423v1
+
+## 25480. Differentiable Inverse Modeling with Physics-Constrained Latent Diffusion for Heterogeneous Subsurface Parameter Fields
+
+- **ID:** `2512.22421` · **primary:** `math.NA` · **Submitted:** 2025-12-27
+- **Authors:** Zihan Lin, QiZhi He
+- **abs:** https://arxiv.org/abs/2512.22421 · **PDF:** https://arxiv.org/pdf/2512.22421v1
+
+## 25481. Nightjar: Dynamic Adaptive Speculative Decoding for Large Language Models Serving
+
+- **ID:** `2512.22420` · **primary:** `cs.DC` · **Submitted:** 2025-12-27
+- **Authors:** Rui Li, Zhaoning Zhang, Libo Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.22420 · **PDF:** https://arxiv.org/pdf/2512.22420v4
+
+## 25482. Hallucination Detection and Evaluation of Large Language Model
+
+- **ID:** `2512.22416` · **primary:** `cs.CL` · **Submitted:** 2025-12-27
+- **Authors:** Chenggong Zhang, Haopeng Wang
+- **abs:** https://arxiv.org/abs/2512.22416 · **PDF:** https://arxiv.org/pdf/2512.22416v1
+
+## 25483. Emergence of Human to Robot Transfer in Vision-Language-Action Models
+
+- **ID:** `2512.22414` · **primary:** `cs.RO` · **Submitted:** 2025-12-27
+- **Authors:** Simar Kareer, Karl Pertsch, James Darpinian, et al.
+- **abs:** https://arxiv.org/abs/2512.22414 · **PDF:** https://arxiv.org/pdf/2512.22414v1
+
+## 25484. A Unified AI, Embedded, Simulation, and Mechanical Design Approach to an Autonomous Delivery Robot
+
+- **ID:** `2512.22408` · **primary:** `cs.RO` · **Submitted:** 2025-12-26
+- **Authors:** Amro Gamar, Ahmed Abduljalil, Alargam Mohammed, et al.
+- **abs:** https://arxiv.org/abs/2512.22408 · **PDF:** https://arxiv.org/pdf/2512.22408v1
+
+## 25485. Efficient Multi-Model Orchestration for Self-Hosted Large Language Models
+
+- **ID:** `2512.22402` · **primary:** `cs.DC` · **Submitted:** 2025-12-26
+- **Authors:** Bhanu Prakash Vangala, Tanu Malik
+- **abs:** https://arxiv.org/abs/2512.22402 · **PDF:** https://arxiv.org/pdf/2512.22402v1
+
+## 25486. Space AI: Leveraging Artificial Intelligence for Space to Improve Life on Earth
+
+- **ID:** `2512.22399` · **primary:** `astro-ph.IM` · **Submitted:** 2025-12-26
+- **Authors:** Ziyang Wang
+- **abs:** https://arxiv.org/abs/2512.22399 · **PDF:** https://arxiv.org/pdf/2512.22399v2
+
+## 25487. Lightweight Inference-Time Personalization for Frozen Knowledge Graph Embeddings
+
+- **ID:** `2512.22398` · **primary:** `cs.AI` · **Submitted:** 2025-12-26
+- **Authors:** Ozan Oguztuzun, Cerag Oguztuzun
+- **abs:** https://arxiv.org/abs/2512.22398 · **PDF:** https://arxiv.org/pdf/2512.22398v1
+
+## 25488. Integrating Wide and Deep Neural Networks with Squeeze-and-Excitation Blocks for Multi-Target Property Prediction in Additively Manufactured Fiber Reinforced Composites
+
+- **ID:** `2512.22397` · **primary:** `physics.comp-ph` · **Submitted:** 2025-12-26
+- **Authors:** Behzad Parvaresh, Rahmat K. Adesunkanmi, Adel Alaeddini
+- **abs:** https://arxiv.org/abs/2512.22397 · **PDF:** https://arxiv.org/pdf/2512.22397v1
+
+## 25489. HalluMat: Detecting Hallucinations in LLM-Generated Materials Science Content Through Multi-Stage Verification
+
+- **ID:** `2512.22396` · **primary:** `cs.AI` · **Submitted:** 2025-12-26
+- **Authors:** Bhanu Prakash Vangala, Sajid Mahmud, Pawan Neupane, et al.
+- **abs:** https://arxiv.org/abs/2512.22396 · **PDF:** https://arxiv.org/pdf/2512.22396v1
+
+## 25490. BLISS: Bandit Layer Importance Sampling Strategy for Efficient Training of Graph Neural Networks
+
+- **ID:** `2512.22388` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Omar Alsaqa, Linh Thi Hoang, Muhammed Fatih Balin
+- **abs:** https://arxiv.org/abs/2512.22388 · **PDF:** https://arxiv.org/pdf/2512.22388v1
+
+## 25491. AI-Generated Code Is Not Reproducible (Yet): An Empirical Study of Dependency Gaps in LLM-Based Coding Agents
+
+- **ID:** `2512.22387` · **primary:** `cs.SE` · **Submitted:** 2025-12-26
+- **Authors:** Bhanu Prakash Vangala, Ali Adibifar, Ashish Gehani, et al.
+- **abs:** https://arxiv.org/abs/2512.22387 · **PDF:** https://arxiv.org/pdf/2512.22387v3
+
+## 25492. Quantitative Rule-Based Strategy modeling in Classic Indian Rummy: A Metric Optimization Approach
+
+- **ID:** `2601.00024` · **primary:** `cs.AI` · **Submitted:** 2025-12-26
+- **Authors:** Purushottam Saha, Avirup Chakraborty, Sourish Sarkar, et al.
+- **abs:** https://arxiv.org/abs/2601.00024 · **PDF:** https://arxiv.org/pdf/2601.00024v1
+
+## 25493. LLM-Guided Exemplar Selection for Few-Shot Wearable-Sensor Human Activity Recognition
+
+- **ID:** `2512.22385` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Elsen Ronando, Sozo Inoue
+- **abs:** https://arxiv.org/abs/2512.22385 · **PDF:** https://arxiv.org/pdf/2512.22385v2
+
+## 25494. Completed Hyperparameter Transfer across Modules, Width, Depth, Batch and Duration
+
+- **ID:** `2512.22382` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Bruno Mlodozeniec, Pierre Ablin, Louis Béthune, et al.
+- **abs:** https://arxiv.org/abs/2512.22382 · **PDF:** https://arxiv.org/pdf/2512.22382v1
+
+## 25495. PHANTOM: Physics-Aware Adversarial Attacks against Federated Learning-Coordinated EV Charging Management System
+
+- **ID:** `2512.22381` · **primary:** `cs.ET` · **Submitted:** 2025-12-26
+- **Authors:** Mohammad Zakaria Haider, Amit Kumar Podder, Prabin Mali, et al.
+- **abs:** https://arxiv.org/abs/2512.22381 · **PDF:** https://arxiv.org/pdf/2512.22381v1
+
+## 25496. Towards Efficient Post-Training via Fourier-Driven Adapter Architectures
+
+- **ID:** `2512.22378` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Donggyun Bae, Jongil Park
+- **abs:** https://arxiv.org/abs/2512.22378 · **PDF:** https://arxiv.org/pdf/2512.22378v1
+
+## 25497. The Syntax of qulk-clauses in Yemeni Ibbi Arabic: A Minimalist Approach
+
+- **ID:** `2512.22376` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Zubaida Mohammed Albadani, Mohammed Q. Shormani
+- **abs:** https://arxiv.org/abs/2512.22376 · **PDF:** https://arxiv.org/pdf/2512.22376v2
+
+## 25498. Self-Evaluation Unlocks Any-Step Text-to-Image Generation
+
+- **ID:** `2512.22374` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Xin Yu, Xiaojuan Qi, Zhengqi Li, et al.
+- **abs:** https://arxiv.org/abs/2512.22374 · **PDF:** https://arxiv.org/pdf/2512.22374v1
+
+## 25499. Subgoaling Relaxation-based Heuristics for Numeric Planning with Infinite Actions
+
+- **ID:** `2512.22367` · **primary:** `cs.AI` · **Submitted:** 2025-12-26
+- **Authors:** Ángel Aso-Mollar, Diego Aineto, Enrico Scala, et al.
+- **abs:** https://arxiv.org/abs/2512.22367 · **PDF:** https://arxiv.org/pdf/2512.22367v1
+
+## 25500. Cost Trade-offs of Reasoning and Non-Reasoning Large Language Models in Text-to-SQL
+
+- **ID:** `2512.22364` · **primary:** `cs.DB` · **Submitted:** 2025-12-26
+- **Authors:** Saurabh Deochake, Debajyoti Mukhopadhyay
+- **abs:** https://arxiv.org/abs/2512.22364 · **PDF:** https://arxiv.org/pdf/2512.22364v2
+
+## 25501. Coordinate Matrix Machine: A Human-level Concept Learning to Classify Very Similar Documents
+
+- **ID:** `2512.23749` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Amin Sadri, M Maruf Hossain
+- **abs:** https://arxiv.org/abs/2512.23749 · **PDF:** https://arxiv.org/pdf/2512.23749v2
+
+## 25502. Beyond Interaction Effects: Two Logics for Studying Population Inequalities
+
+- **ID:** `2601.04223` · **primary:** `cs.CY` · **Submitted:** 2025-12-26
+- **Authors:** Adel Daoud
+- **abs:** https://arxiv.org/abs/2601.04223 · **PDF:** https://arxiv.org/pdf/2601.04223v1
+
+## 25503. VULCAN: Tool-Augmented Multi Agents for Iterative 3D Object Arrangement
+
+- **ID:** `2512.22351` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Zhengfei Kuang, Rui Lin, Long Zhao, et al.
+- **abs:** https://arxiv.org/abs/2512.22351 · **PDF:** https://arxiv.org/pdf/2512.22351v2
+
+## 25504. Human-like visual computing advances explainability and few-shot learning in deep neural networks for complex physiological data
+
+- **ID:** `2512.22349` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Alaa Alahmadi, Mohamed Hasan
+- **abs:** https://arxiv.org/abs/2512.22349 · **PDF:** https://arxiv.org/pdf/2512.22349v1
+
+## 25505. NewsScope: Schema-Grounded Cross-Domain News Claim Extraction with Open Models
+
+- **ID:** `2601.08852` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Nidhi Pandya
+- **abs:** https://arxiv.org/abs/2601.08852 · **PDF:** https://arxiv.org/pdf/2601.08852v1
+
+## 25506. Agentic Structured Graph Traversal for Root Cause Analysis of Code-related Incidents in Cloud Applications
+
+- **ID:** `2512.22113` · **primary:** `cs.DC` · **Submitted:** 2025-12-26
+- **Authors:** Shengkun Cui, Rahul Krishna, Saurabh Jha, et al.
+- **abs:** https://arxiv.org/abs/2512.22113 · **PDF:** https://arxiv.org/pdf/2512.22113v1
+
+## 25507. The Effectiveness of Approximate Regularized Replay for Efficient Supervised Fine-Tuning of Large Language Models
+
+- **ID:** `2512.22337` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Matthew Riemer, Erik Miehling, Miao Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.22337 · **PDF:** https://arxiv.org/pdf/2512.22337v1
+
+## 25508. Agent2World: Learning to Generate Symbolic World Models via Adaptive Multi-Agent Feedback
+
+- **ID:** `2512.22336` · **primary:** `cs.AI` · **Submitted:** 2025-12-26
+- **Authors:** Mengkang Hu, Bowei Xia, Yuran Wu, et al.
+- **abs:** https://arxiv.org/abs/2512.22336 · **PDF:** https://arxiv.org/pdf/2512.22336v1
+
+## 25509. Pediatric Pneumonia Detection from Chest X-Rays:A Comparative Study of Transfer Learning and Custom CNNs
+
+- **ID:** `2601.00837` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Agniv Roy Choudhury
+- **abs:** https://arxiv.org/abs/2601.00837 · **PDF:** https://arxiv.org/pdf/2601.00837v1
+
+## 25510. Pruning as a Game: Equilibrium-Driven Sparsification of Neural Networks
+
+- **ID:** `2512.22106` · **primary:** `cs.AI` · **Submitted:** 2025-12-26
+- **Authors:** Zubair Shah, Noaman Khan
+- **abs:** https://arxiv.org/abs/2512.22106 · **PDF:** https://arxiv.org/pdf/2512.22106v1
+
+## 25511. A Review of Diffusion-based Simulation-Based Inference: Foundations and Applications in Non-Ideal Data Scenarios
+
+- **ID:** `2512.23748` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Haley Rosso, Talea Mayo
+- **abs:** https://arxiv.org/abs/2512.23748 · **PDF:** https://arxiv.org/pdf/2512.23748v1
+
+## 25512. State-of-the-art Small Language Coder Model: Mify-Coder
+
+- **ID:** `2512.23747` · **primary:** `cs.SE` · **Submitted:** 2025-12-26
+- **Authors:** Abhinav Parmar, Abhisek Panigrahi, Abhishek Kumar Dwivedi, et al.
+- **abs:** https://arxiv.org/abs/2512.23747 · **PDF:** https://arxiv.org/pdf/2512.23747v1
+
+## 25513. Explainable Multimodal Regression via Information Decomposition
+
+- **ID:** `2512.22102` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Zhaozhao Ma, Shujian Yu
+- **abs:** https://arxiv.org/abs/2512.22102 · **PDF:** https://arxiv.org/pdf/2512.22102v1
+
+## 25514. A2P-Vis: an Analyzer-to-Presenter Agentic Pipeline for Visual Insights Generation and Reporting
+
+- **ID:** `2512.22101` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Shuyu Gan, Renxiang Wang, James Mooney, et al.
+- **abs:** https://arxiv.org/abs/2512.22101 · **PDF:** https://arxiv.org/pdf/2512.22101v1
+
+## 25515. Introducing TrGLUE and SentiTurca: A Comprehensive Benchmark for Turkish General Language Understanding and Sentiment Analysis
+
+- **ID:** `2512.22100` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Duygu Altinok
+- **abs:** https://arxiv.org/abs/2512.22100 · **PDF:** https://arxiv.org/pdf/2512.22100v1
+
+## 25516. Feature Learning with Multi-Stage Vision Transformers on Inter-Modality HER2 Status Scoring and Tumor Classification on Whole Slides
+
+- **ID:** `2512.22335` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Olaide N. Oyelade, Oliver Hoxey, Yulia Humrye
+- **abs:** https://arxiv.org/abs/2512.22335 · **PDF:** https://arxiv.org/pdf/2512.22335v1
+
+## 25517. SciEvalKit: An Open-source Evaluation Toolkit for Scientific General Intelligence
+
+- **ID:** `2512.22334` · **primary:** `cs.AI` · **Submitted:** 2025-12-26
+- **Authors:** Yiheng Wang, Yixin Chen, Shuo Li, et al.
+- **abs:** https://arxiv.org/abs/2512.22334 · **PDF:** https://arxiv.org/pdf/2512.22334v3
+
+## 25518. Emotion classification using EEG headset signals and Random Forest
+
+- **ID:** `2512.22333` · **primary:** `cs.HC` · **Submitted:** 2025-12-26
+- **Authors:** Ricardo Vasquez, Diego Riofrío-Luzcando, Joe Carrion-Jumbo, et al.
+- **abs:** https://arxiv.org/abs/2512.22333 · **PDF:** https://arxiv.org/pdf/2512.22333v1
+
+## 25519. Unifying Learning Dynamics and Generalization in Transformers Scaling Law
+
+- **ID:** `2512.22088` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Chiwun Yang
+- **abs:** https://arxiv.org/abs/2512.22088 · **PDF:** https://arxiv.org/pdf/2512.22088v2
+
+## 25520. Context as a Tool: Context Management for Long-Horizon SWE-Agents
+
+- **ID:** `2512.22087` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Shukai Liu, Jian Yang, Bo Jiang, et al.
+- **abs:** https://arxiv.org/abs/2512.22087 · **PDF:** https://arxiv.org/pdf/2512.22087v1
+
+## 25521. A Frobenius-Optimal Projection for Enforcing Linear Conservation in Learned Dynamical Models
+
+- **ID:** `2512.22084` · **primary:** `math.DS` · **Submitted:** 2025-12-26
+- **Authors:** John M. Mango, Ronald Katende
+- **abs:** https://arxiv.org/abs/2512.22084 · **PDF:** https://arxiv.org/pdf/2512.22084v1
+
+## 25522. The Multi-View Paradigm Shift in MRI Radiomics: Predicting MGMT Methylation in Glioblastoma
+
+- **ID:** `2512.22331` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Mariya Miteva, Maria Nisheva-Pavlova
+- **abs:** https://arxiv.org/abs/2512.22331 · **PDF:** https://arxiv.org/pdf/2512.22331v1
+
+## 25523. Scaling Adversarial Training via Data Selection
+
+- **ID:** `2512.22069` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Youran Ye, Dejin Wang, Ajinkya Bhandare
+- **abs:** https://arxiv.org/abs/2512.22069 · **PDF:** https://arxiv.org/pdf/2512.22069v1
+
+## 25524. Prefill vs. Decode Bottlenecks: SRAM-Frequency Tradeoffs and the Memory-Bandwidth Ceiling
+
+- **ID:** `2512.22066` · **primary:** `cs.AR` · **Submitted:** 2025-12-26
+- **Authors:** Hannah Atmer, Yuan Yao, Thiemo Voigt, et al.
+- **abs:** https://arxiv.org/abs/2512.22066 · **PDF:** https://arxiv.org/pdf/2512.22066v1
+
+## 25525. StreamAvatar: Streaming Diffusion Models for Real-Time Interactive Human Avatars
+
+- **ID:** `2512.22065` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Zhiyao Sun, Ziqiao Peng, Yifeng Ma, et al.
+- **abs:** https://arxiv.org/abs/2512.22065 · **PDF:** https://arxiv.org/pdf/2512.22065v2
+
+## 25526. Expert System for Bitcoin Forecasting: Integrating Global Liquidity via TimeXer Transformers
+
+- **ID:** `2512.22326` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Sravan Karthick T
+- **abs:** https://arxiv.org/abs/2512.22326 · **PDF:** https://arxiv.org/pdf/2512.22326v2
+
+## 25527. Toward Secure and Compliant AI: Organizational Standards and Protocols for NLP Model Lifecycle Management
+
+- **ID:** `2512.22060` · **primary:** `cs.CR` · **Submitted:** 2025-12-26
+- **Authors:** Sunil Arora, John Hastings
+- **abs:** https://arxiv.org/abs/2512.22060 · **PDF:** https://arxiv.org/pdf/2512.22060v1
+
+## 25528. Why Smooth Stability Assumptions Fail for ReLU Learning
+
+- **ID:** `2512.22055` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Ronald Katende
+- **abs:** https://arxiv.org/abs/2512.22055 · **PDF:** https://arxiv.org/pdf/2512.22055v1
+
+## 25529. SpotEdit: Selective Region Editing in Diffusion Transformers
+
+- **ID:** `2512.22323` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Zhibin Qin, Zhenxiong Tan, Zeqing Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.22323 · **PDF:** https://arxiv.org/pdf/2512.22323v1
+
+## 25530. SmartSnap: Proactive Evidence Seeking for Self-Verifying Agents
+
+- **ID:** `2512.22322` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Shaofei Cai, Yulei Qin, Haojia Lin, et al.
+- **abs:** https://arxiv.org/abs/2512.22322 · **PDF:** https://arxiv.org/pdf/2512.22322v2
+
+## 25531. Measuring Stability Beyond Accuracy in Small Open-Source Medical Large Language Models for Pediatric Endocrinology
+
+- **ID:** `2601.11567` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Vanessa D'Amario, Randy Daniel, Alessandro Zanetti, et al.
+- **abs:** https://arxiv.org/abs/2601.11567 · **PDF:** https://arxiv.org/pdf/2601.11567v1
+
+## 25532. From In Silico to In Vitro: Evaluating Molecule Generative Models for Hit Generation
+
+- **ID:** `2512.22031` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Nagham Osman, Vittorio Lembo, Giovanni Bottegoni, et al.
+- **abs:** https://arxiv.org/abs/2512.22031 · **PDF:** https://arxiv.org/pdf/2512.22031v1
+
+## 25533. LibContinual: A Comprehensive Library towards Realistic Continual Learning
+
+- **ID:** `2512.22029` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Wenbin Li, Shangge Liu, Borui Kang, et al.
+- **abs:** https://arxiv.org/abs/2512.22029 · **PDF:** https://arxiv.org/pdf/2512.22029v1
+
+## 25534. Cross-Platform Evaluation of Large Language Model Safety in Pediatric Consultations: Evolution of Adversarial Robustness and the Scale Paradox
+
+- **ID:** `2601.09721` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Vahideh Zolfaghari
+- **abs:** https://arxiv.org/abs/2601.09721 · **PDF:** https://arxiv.org/pdf/2601.09721v1
+
+## 25535. A Comprehensive Study of Deep Learning Model Fixing Approaches
+
+- **ID:** `2512.23745` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Hanmo You, Zan Wang, Zishuo Dong, et al.
+- **abs:** https://arxiv.org/abs/2512.23745 · **PDF:** https://arxiv.org/pdf/2512.23745v1
+
+## 25536. Direction Finding with Sparse Arrays Based on Variable Window Size Spatial Smoothing
+
+- **ID:** `2512.22024` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Wesley S. Leite, Rodrigo C. de Lamare, Yuriy Zakharov, et al.
+- **abs:** https://arxiv.org/abs/2512.22024 · **PDF:** https://arxiv.org/pdf/2512.22024v1
+
+## 25537. Meta-Learning-Based Handover Management in NextG O-RAN
+
+- **ID:** `2512.22022` · **primary:** `cs.NI` · **Submitted:** 2025-12-26
+- **Authors:** Michail Kalntis, George Iosifidis, José Suárez-Varela, et al.
+- **abs:** https://arxiv.org/abs/2512.22022 · **PDF:** https://arxiv.org/pdf/2512.22022v1
+
+## 25538. Uncertainty-Aware Dynamic Knowledge Graphs for Reliable Question Answering
+
+- **ID:** `2601.09720` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Yu Takahashi, Shun Takeuchi, Kexuan Xin, et al.
+- **abs:** https://arxiv.org/abs/2601.09720 · **PDF:** https://arxiv.org/pdf/2601.09720v1
+
+## 25539. Intrinsic-Metric Physics-Informed Neural Networks (IM-PINN) for Reaction-Diffusion Dynamics on Complex Riemannian Manifolds
+
+- **ID:** `2601.00834` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Julian Evan Chrisnanto, Salsabila Rahma Alia, Nurfauzi Fadillah, et al.
+- **abs:** https://arxiv.org/abs/2601.00834 · **PDF:** https://arxiv.org/pdf/2601.00834v3
+
+## 25540. Decomposing Uncertainty in Probabilistic Knowledge Graph Embeddings: Why Entity Variance Is Not Enough
+
+- **ID:** `2512.22318` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Chorok Lee
+- **abs:** https://arxiv.org/abs/2512.22318 · **PDF:** https://arxiv.org/pdf/2512.22318v2
+
+## 25541. HWL-HIN: A Hypergraph-Level Hypergraph Isomorphism Network as Powerful as the Hypergraph Weisfeiler-Lehman Test with Application to Higher-Order Network Robustness
+
+- **ID:** `2512.22014` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Chengyu Tian, Wenbin Pei
+- **abs:** https://arxiv.org/abs/2512.22014 · **PDF:** https://arxiv.org/pdf/2512.22014v1
+
+## 25542. LongFly: Long-Horizon UAV Vision-and-Language Navigation with Spatiotemporal Context Integration
+
+- **ID:** `2512.22010` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Wen Jiang, Li Wang, Kangyao Huang, et al.
+- **abs:** https://arxiv.org/abs/2512.22010 · **PDF:** https://arxiv.org/pdf/2512.22010v1
+
+## 25543. DuaDeep-SeqAffinity: Dual-Stream Deep Learning Framework for Sequence-Only Antigen-Antibody Affinity Prediction
+
+- **ID:** `2512.22007` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Aicha Boutorh, Soumia Bouyahiaoui, Sara Belhadj, et al.
+- **abs:** https://arxiv.org/abs/2512.22007 · **PDF:** https://arxiv.org/pdf/2512.22007v1
+
+## 25544. LangPrecip: Language-Aware Multimodal Precipitation Nowcasting
+
+- **ID:** `2512.22317` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Xudong Ling, Tianxi Huang, Qian Dong, et al.
+- **abs:** https://arxiv.org/abs/2512.22317 · **PDF:** https://arxiv.org/pdf/2512.22317v1
+
+## 25545. Look Closer! An Adversarial Parametric Editing Framework for Hallucination Mitigation in VLMs
+
+- **ID:** `2512.21999` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Jiayu Hu, Beibei Li, Jiangwei Xia, et al.
+- **abs:** https://arxiv.org/abs/2512.21999 · **PDF:** https://arxiv.org/pdf/2512.21999v1
+
+## 25546. VideoZoomer: Reinforcement-Learned Temporal Focusing for Long Video Reasoning
+
+- **ID:** `2512.22315` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Yang Ding, Yizhen Zhang, Xin Lai, et al.
+- **abs:** https://arxiv.org/abs/2512.22315 · **PDF:** https://arxiv.org/pdf/2512.22315v1
+
+## 25547. LVLM-Aided Alignment of Task-Specific Vision Models
+
+- **ID:** `2512.21985` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Alexander Koebler, Lukas Kuhn, Ingo Thon, et al.
+- **abs:** https://arxiv.org/abs/2512.21985 · **PDF:** https://arxiv.org/pdf/2512.21985v1
+
+## 25548. Socio-technical aspects of Agentic AI
+
+- **ID:** `2601.06064` · **primary:** `cs.CY` · **Submitted:** 2025-12-26
+- **Authors:** Praveen Kumar Donta, Alaa Saleh, Ying Li, et al.
+- **abs:** https://arxiv.org/abs/2601.06064 · **PDF:** https://arxiv.org/pdf/2601.06064v1
+
+## 25549. Optimistic Feasible Search for Closed-Loop Fair Threshold Decision-Making
+
+- **ID:** `2512.22313` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Wenzhang Du
+- **abs:** https://arxiv.org/abs/2512.22313 · **PDF:** https://arxiv.org/pdf/2512.22313v1
+
+## 25550. Modeling high dimensional point clouds with the spherical cluster model
+
+- **ID:** `2512.21960` · **primary:** `stat.ME` · **Submitted:** 2025-12-26
+- **Authors:** Frédéric Cazals, Antoine Commaret, Louis Goldenberg
+- **abs:** https://arxiv.org/abs/2512.21960 · **PDF:** https://arxiv.org/pdf/2512.21960v1
+
+## 25551. Self-attention vector output similarities reveal how machines pay attention
+
+- **ID:** `2512.21956` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Tal Halevi, Yarden Tzach, Ronit D. Gross, et al.
+- **abs:** https://arxiv.org/abs/2512.21956 · **PDF:** https://arxiv.org/pdf/2512.21956v2
+
+## 25552. Data relativistic uncertainty framework for low-illumination anime scenery image enhancement
+
+- **ID:** `2512.21944` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Yiquan Gao, John See
+- **abs:** https://arxiv.org/abs/2512.21944 · **PDF:** https://arxiv.org/pdf/2512.21944v3
+
+## 25553. Broken Words, Broken Performance: Effect of Tokenization on Performance of LLMs
+
+- **ID:** `2512.21933` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Sachin Pawar, Manoj Apte, Kshitij Jadhav, et al.
+- **abs:** https://arxiv.org/abs/2512.21933 · **PDF:** https://arxiv.org/pdf/2512.21933v1
+
+## 25554. Hybrid Combinatorial Multi-armed Bandits with Probabilistically Triggered Arms
+
+- **ID:** `2512.21925` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Kongchang Zhou, Tingyu Zhang, Wei Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.21925 · **PDF:** https://arxiv.org/pdf/2512.21925v1
+
+## 25555. Unsupervised Anomaly Detection in Brain MRI via Disentangled Anatomy Learning
+
+- **ID:** `2512.21924` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Tao Yang, Xiuying Wang, Hao Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.21924 · **PDF:** https://arxiv.org/pdf/2512.21924v1
+
+## 25556. AutoPP: Towards Automated Product Poster Generation and Optimization
+
+- **ID:** `2512.21921` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Jiahao Fan, Yuxin Qin, Wei Feng, et al.
+- **abs:** https://arxiv.org/abs/2512.21921 · **PDF:** https://arxiv.org/pdf/2512.21921v1
+
+## 25557. SWE-RM: Execution-free Feedback For Software Engineering Agents
+
+- **ID:** `2512.21919` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** KaShun Shum, Binyuan Hui, Jiawei Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.21919 · **PDF:** https://arxiv.org/pdf/2512.21919v1
+
+## 25558. Semiparametric Preference Optimization: Your Language Model is Secretly a Single-Index Model
+
+- **ID:** `2512.21917` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Nathan Kallus
+- **abs:** https://arxiv.org/abs/2512.21917 · **PDF:** https://arxiv.org/pdf/2512.21917v2
+
+## 25559. Exploring the Heterogeneity of Tabular Data: A Diversity-aware Data Generator via LLMs
+
+- **ID:** `2512.21915` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Yafeng Tang, Xiaoou Ding, Jianzhuo Du, et al.
+- **abs:** https://arxiv.org/abs/2512.21915 · **PDF:** https://arxiv.org/pdf/2512.21915v1
+
+## 25560. GQ-VAE: A gated quantized VAE for learning variable length tokens
+
+- **ID:** `2512.21913` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Theo Datta, Kayla Huang, Sham Kakade, et al.
+- **abs:** https://arxiv.org/abs/2512.21913 · **PDF:** https://arxiv.org/pdf/2512.21913v1
+
+## 25561. Accelerate Speculative Decoding with Sparse Computation in Verification
+
+- **ID:** `2512.21911` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Jikai Wang, Jianchao Tan, Yuxuan Hu, et al.
+- **abs:** https://arxiv.org/abs/2512.21911 · **PDF:** https://arxiv.org/pdf/2512.21911v1
+
+## 25562. SpatialBench: Can Agents Analyze Real-World Spatial Biology Data?
+
+- **ID:** `2512.21907` · **primary:** `cs.AI` · **Submitted:** 2025-12-26
+- **Authors:** Kenny Workman, Zhen Yang, Harihara Muralidharan, et al.
+- **abs:** https://arxiv.org/abs/2512.21907 · **PDF:** https://arxiv.org/pdf/2512.21907v2
+
+## 25563. Explainable Statute Prediction via Attention-based Model and LLM Prompting
+
+- **ID:** `2512.21902` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Sachin Pawar, Girish Keshav Palshikar, Anindita Sinha Banerjee, et al.
+- **abs:** https://arxiv.org/abs/2512.21902 · **PDF:** https://arxiv.org/pdf/2512.21902v1
+
+## 25564. LLMBoost: Make Large Language Models Stronger with Boosting
+
+- **ID:** `2512.22309` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Zehao Chen, Tianxiang Ai, Yifei Li, et al.
+- **abs:** https://arxiv.org/abs/2512.22309 · **PDF:** https://arxiv.org/pdf/2512.22309v1
+
+## 25565. Flexible Multitask Learning with Factorized Diffusion Policy
+
+- **ID:** `2512.21898` · **primary:** `cs.RO` · **Submitted:** 2025-12-26
+- **Authors:** Chaoqi Liu, Haonan Chen, Sigmund H. Høeg, et al.
+- **abs:** https://arxiv.org/abs/2512.21898 · **PDF:** https://arxiv.org/pdf/2512.21898v2
+
+## 25566. MMCTOP: A Multimodal Textualization and Mixture-of-Experts Framework for Clinical Trial Outcome Prediction
+
+- **ID:** `2512.21897` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Carolina Aparício, Qi Shi, Bo Wen, et al.
+- **abs:** https://arxiv.org/abs/2512.21897 · **PDF:** https://arxiv.org/pdf/2512.21897v1
+
+## 25567. Aerial World Model for Long-horizon Visual Generation and Navigation in 3D Space
+
+- **ID:** `2512.21887` · **primary:** `cs.RO` · **Submitted:** 2025-12-26
+- **Authors:** Weichen Zhang, Peizhi Tang, Xin Zeng, et al.
+- **abs:** https://arxiv.org/abs/2512.21887 · **PDF:** https://arxiv.org/pdf/2512.21887v2
+
+## 25568. Bounded Hyperbolic Tangent: A Stable and Efficient Alternative to Pre-Layer Normalization in Large Language Models
+
+- **ID:** `2601.09719` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Hoyoon Byun, Youngjun Choi, Taero Kim, et al.
+- **abs:** https://arxiv.org/abs/2601.09719 · **PDF:** https://arxiv.org/pdf/2601.09719v2
+
+## 25569. Optimizing Resource Allocation for Geographically-Distributed Inference by Large Language Models
+
+- **ID:** `2512.21884` · **primary:** `cs.DC` · **Submitted:** 2025-12-26
+- **Authors:** Tingyang Sun, Ting He, Bo Ji, et al.
+- **abs:** https://arxiv.org/abs/2512.21884 · **PDF:** https://arxiv.org/pdf/2512.21884v1
+
+## 25570. MASFIN: A Multi-Agent System for Decomposed Financial Reasoning and Forecasting
+
+- **ID:** `2512.21878` · **primary:** `cs.MA` · **Submitted:** 2025-12-26
+- **Authors:** Marc S. Montalvo, Hamed Yaghoobian
+- **abs:** https://arxiv.org/abs/2512.21878 · **PDF:** https://arxiv.org/pdf/2512.21878v1
+
+## 25571. CricBench: A Multilingual Benchmark for Evaluating LLMs in Cricket Analytics
+
+- **ID:** `2512.21877` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Vaibhav Devraj, Dhruv Kumar, Jagat Sesh Challa, et al.
+- **abs:** https://arxiv.org/abs/2512.21877 · **PDF:** https://arxiv.org/pdf/2512.21877v2
+
+## 25572. LLA: Enhancing Security and Privacy for Generative Models with Logic-Locked Accelerators
+
+- **ID:** `2512.22307` · **primary:** `cs.CR` · **Submitted:** 2025-12-26
+- **Authors:** You Li, Guannan Zhao, Yuhao Ju, et al.
+- **abs:** https://arxiv.org/abs/2512.22307 · **PDF:** https://arxiv.org/pdf/2512.22307v1
+
+## 25573. Beyond Single Bugs: Benchmarking Large Language Models for Multi-Vulnerability Detection
+
+- **ID:** `2512.22306` · **primary:** `cs.CR` · **Submitted:** 2025-12-26
+- **Authors:** Chinmay Pushkar, Sanchit Kabra, Dhruv Kumar, et al.
+- **abs:** https://arxiv.org/abs/2512.22306 · **PDF:** https://arxiv.org/pdf/2512.22306v1
+
+## 25574. PDx -- Adaptive Credit Risk Forecasting Model in Digital Lending using Machine Learning Operations
+
+- **ID:** `2512.22305` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Sultan Amed, Chan Yu Hang, Sayantan Banerjee
+- **abs:** https://arxiv.org/abs/2512.22305 · **PDF:** https://arxiv.org/pdf/2512.22305v1
+
+## 25575. StatLLaMA: Multi-Stage training for domain-optimized statistical large language models
+
+- **ID:** `2601.09718` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Jing-Yi Zeng, Guan-Hua Huang
+- **abs:** https://arxiv.org/abs/2601.09718 · **PDF:** https://arxiv.org/pdf/2601.09718v2
+
+## 25576. Bridging the Copyright Gap: Do Large Vision-Language Models Recognize and Respect Copyrighted Content?
+
+- **ID:** `2512.21871` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Naen Xu, Jinghuai Zhang, Changjiang Li, et al.
+- **abs:** https://arxiv.org/abs/2512.21871 · **PDF:** https://arxiv.org/pdf/2512.21871v1
+
+## 25577. Secure and Explainable Fraud Detection in Finance via Hierarchical Multi-source Dataset Distillation
+
+- **ID:** `2512.21866` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Yiming Qian, Thorsten Neumann, Xueyining Huang, et al.
+- **abs:** https://arxiv.org/abs/2512.21866 · **PDF:** https://arxiv.org/pdf/2512.21866v1
+
+## 25578. Balancing Accuracy and Efficiency: CNN Fusion Models for Diabetic Retinopathy Screening
+
+- **ID:** `2512.21861` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Md Rafid Islam, Rafsan Jany, Akib Ahmed, et al.
+- **abs:** https://arxiv.org/abs/2512.21861 · **PDF:** https://arxiv.org/pdf/2512.21861v1
+
+## 25579. TimeBill: Time-Budgeted Inference for Large Language Models
+
+- **ID:** `2512.21859` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Qi Fan, An Zou, Yehan Ma
+- **abs:** https://arxiv.org/abs/2512.21859 · **PDF:** https://arxiv.org/pdf/2512.21859v1
+
+## 25580. MoonBot: Modular and On-Demand Reconfigurable Robot Toward Moon Base Construction
+
+- **ID:** `2512.21853` · **primary:** `cs.RO` · **Submitted:** 2025-12-26
+- **Authors:** Kentaro Uno, Elian Neppel, Gustavo H. Diaz, et al.
+- **abs:** https://arxiv.org/abs/2512.21853 · **PDF:** https://arxiv.org/pdf/2512.21853v1
+
+## 25581. A Comedy of Estimators: On KL Regularization in RL Training of LLMs
+
+- **ID:** `2512.21852` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Vedant Shah, Johan Obando-Ceron, Vineet Jain, et al.
+- **abs:** https://arxiv.org/abs/2512.21852 · **PDF:** https://arxiv.org/pdf/2512.21852v3
+
+## 25582. Attack-Aware Deepfake Detection under Counter-Forensic Manipulations
+
+- **ID:** `2512.22303` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Noor Fatima, Hasan Faraz Khan, Muzammil Behzad
+- **abs:** https://arxiv.org/abs/2512.22303 · **PDF:** https://arxiv.org/pdf/2512.22303v1
+
+## 25583. HeartBench: Probing Core Dimensions of Anthropomorphic Intelligence in LLMs
+
+- **ID:** `2512.21849` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Jiaxin Liu, Peiyi Tu, Wenyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.21849 · **PDF:** https://arxiv.org/pdf/2512.21849v1
+
+## 25584. Statistical and Machine Learning Analysis of Traffic Accidents on US 158 in Currituck County: A Comparison with HSM Predictions
+
+- **ID:** `2512.22302` · **primary:** `cs.LG` · **Submitted:** 2025-12-26
+- **Authors:** Jennifer Sawyer, Julian Allagan
+- **abs:** https://arxiv.org/abs/2512.22302 · **PDF:** https://arxiv.org/pdf/2512.22302v1
+
+## 25585. Scalable Class-Incremental Learning Based on Parametric Neural Collapse
+
+- **ID:** `2512.21845` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Chuangxin Zhang, Guangfeng Lin, Enhui Zhao, et al.
+- **abs:** https://arxiv.org/abs/2512.21845 · **PDF:** https://arxiv.org/pdf/2512.21845v1
+
+## 25586. AlignAR: Generative Sentence Alignment for Arabic-English Parallel Corpora of Legal and Literary Texts
+
+- **ID:** `2512.21842` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Baorong Huang, Ali Asiri
+- **abs:** https://arxiv.org/abs/2512.21842 · **PDF:** https://arxiv.org/pdf/2512.21842v2
+
+## 25587. Knowledge Reasoning of Large Language Models Integrating Graph-Structured Information for Pest and Disease Control in Tobacco
+
+- **ID:** `2512.21837` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Siyu Li, Chenwei Song, Wan Zhou, et al.
+- **abs:** https://arxiv.org/abs/2512.21837 · **PDF:** https://arxiv.org/pdf/2512.21837v1
+
+## 25588. Conserved active information
+
+- **ID:** `2512.21834` · **primary:** `cs.NE` · **Submitted:** 2025-12-26
+- **Authors:** Yanchen Chen, Daniel Andrés Díaz-Pachón
+- **abs:** https://arxiv.org/abs/2512.21834 · **PDF:** https://arxiv.org/pdf/2512.21834v1
+
+## 25589. Hybrid-Code v2: Zero-Hallucination Clinical ICD-10 Coding via Neuro-Symbolic Verification and Automated Knowledge Base Expansion
+
+- **ID:** `2512.23743` · **primary:** `cs.SE` · **Submitted:** 2025-12-26
+- **Authors:** Yunguo Yu
+- **abs:** https://arxiv.org/abs/2512.23743 · **PDF:** https://arxiv.org/pdf/2512.23743v2
+
+## 25590. Tilt Matching for Scalable Sampling and Fine-Tuning
+
+- **ID:** `2512.21829` · **primary:** `stat.ML` · **Submitted:** 2025-12-26
+- **Authors:** Peter Potaptchik, Cheuk-Kit Lee, Michael S. Albergo
+- **abs:** https://arxiv.org/abs/2512.21829 · **PDF:** https://arxiv.org/pdf/2512.21829v1
+
+## 25591. AgenticTCAD: A LLM-based Multi-Agent Framework for Automated TCAD Code Generation and Device Optimization
+
+- **ID:** `2512.23742` · **primary:** `cs.SE` · **Submitted:** 2025-12-26
+- **Authors:** Guangxi Fan, Tianliang Ma, Xuguang Sun, et al.
+- **abs:** https://arxiv.org/abs/2512.23742 · **PDF:** https://arxiv.org/pdf/2512.23742v2
+
+## 25592. Method Decoration (DeMe): A Framework for LLM-Driven Adaptive Method Generation in Dynamic IoT Environments
+
+- **ID:** `2512.21817` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Hong Su
+- **abs:** https://arxiv.org/abs/2512.21817 · **PDF:** https://arxiv.org/pdf/2512.21817v1
+
+## 25593. Few Tokens Matter: Entropy Guided Attacks on Vision-Language Models
+
+- **ID:** `2512.21815` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Mengqi He, Xinyu Tian, Xin Shen, et al.
+- **abs:** https://arxiv.org/abs/2512.21815 · **PDF:** https://arxiv.org/pdf/2512.21815v1
+
+## 25594. Real-Time In-Cabin Driver Behavior Recognition on Low-Cost Edge Hardware
+
+- **ID:** `2512.22298` · **primary:** `cs.CV` · **Submitted:** 2025-12-26
+- **Authors:** Vesal Ahsani, Babak Hossein Khalaj, Hamed Shah-Mansouri
+- **abs:** https://arxiv.org/abs/2512.22298 · **PDF:** https://arxiv.org/pdf/2512.22298v2
+
+## 25595. On The Conceptualization and Societal Impact of Cross-Cultural Bias
+
+- **ID:** `2512.21809` · **primary:** `cs.CL` · **Submitted:** 2025-12-26
+- **Authors:** Vitthal Bhandari
+- **abs:** https://arxiv.org/abs/2512.21809 · **PDF:** https://arxiv.org/pdf/2512.21809v1
+
+## 25596. S&P 500 Stock's Movement Prediction using CNN
+
+- **ID:** `2512.21804` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Rahul Gupta
+- **abs:** https://arxiv.org/abs/2512.21804 · **PDF:** https://arxiv.org/pdf/2512.21804v1
+
+## 25597. CellMamba: Adaptive Mamba for Accurate and Efficient Cell Detection
+
+- **ID:** `2512.21803` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Ruochen Liu, Yi Tian, Jiahao Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.21803 · **PDF:** https://arxiv.org/pdf/2512.21803v1
+
+## 25598. Smart IoT-Based Leak Forecasting and Detection for Energy-Efficient Liquid Cooling in AI Data Centers
+
+- **ID:** `2512.21801` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Krishna Chaitanya Sunkara, Rambabu Konakanchi
+- **abs:** https://arxiv.org/abs/2512.21801 · **PDF:** https://arxiv.org/pdf/2512.21801v1
+
+## 25599. Deep Generative Models for Synthetic Financial Data: Applications to Portfolio and Risk Modeling
+
+- **ID:** `2512.21798` · **primary:** `q-fin.ST` · **Submitted:** 2025-12-25
+- **Authors:** Christophe D. Hounwanou, Yae Ulrich Gaba
+- **abs:** https://arxiv.org/abs/2512.21798 · **PDF:** https://arxiv.org/pdf/2512.21798v2
+
+## 25600. Multi-agent Adaptive Mechanism Design
+
+- **ID:** `2512.21794` · **primary:** `cs.GT` · **Submitted:** 2025-12-25
+- **Authors:** Qiushi Han, David Simchi-Levi, Renfei Tan, et al.
+- **abs:** https://arxiv.org/abs/2512.21794 · **PDF:** https://arxiv.org/pdf/2512.21794v2
+
+## 25601. Synthetic Financial Data Generation for Enhanced Financial Modelling
+
+- **ID:** `2512.21791` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Christophe D. Hounwanou, Yae Ulrich Gaba, Pierre Ntakirutimana
+- **abs:** https://arxiv.org/abs/2512.21791 · **PDF:** https://arxiv.org/pdf/2512.21791v1
+
+## 25602. Five Years of SciCap: What We Learned and Future Directions for Scientific Figure Captioning
+
+- **ID:** `2512.21789` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Ting-Hao 'Kenneth' Huang, Ryan A. Rossi, Sungchul Kim, et al.
+- **abs:** https://arxiv.org/abs/2512.21789 · **PDF:** https://arxiv.org/pdf/2512.21789v2
+
+## 25603. InstructMoLE: Instruction-Guided Mixture of Low-rank Experts for Multi-Conditional Image Generation
+
+- **ID:** `2512.21788` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Jinqi Xiao, Qing Yan, Liming Jiang, et al.
+- **abs:** https://arxiv.org/abs/2512.21788 · **PDF:** https://arxiv.org/pdf/2512.21788v2
+
+## 25604. Ara-HOPE: Human-Centric Post-Editing Evaluation for Dialectal Arabic to Modern Standard Arabic Translation
+
+- **ID:** `2512.21787` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Abdullah Alabdullah, Lifeng Han, Chenghua Lin
+- **abs:** https://arxiv.org/abs/2512.21787 · **PDF:** https://arxiv.org/pdf/2512.21787v2
+
+## 25605. VAMP-Net: An Interpretable Multi-Path Framework of Genomic Permutation-Invariant Set Attention and Quality-Aware 1D-CNN for MTB Drug Resistance
+
+- **ID:** `2512.21786` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Aicha Boutorh, Kamar Hibatallah Baghdadi, Anais Daoud
+- **abs:** https://arxiv.org/abs/2512.21786 · **PDF:** https://arxiv.org/pdf/2512.21786v1
+
+## 25606. Hybrid Quantum-Classical Mixture of Experts: Unlocking Topological Advantage via Interference-Based Routing
+
+- **ID:** `2512.22296` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Reda Heddad, Lamiae Bouanane
+- **abs:** https://arxiv.org/abs/2512.22296 · **PDF:** https://arxiv.org/pdf/2512.22296v1
+
+## 25607. A Knowledge Graph and Deep Learning-Based Semantic Recommendation Database System for Advertisement Retrieval and Personalization
+
+- **ID:** `2601.00833` · **primary:** `cs.IR` · **Submitted:** 2025-12-25
+- **Authors:** Tangtang Wang, Kaijie Zhang, Kuangcong Liu
+- **abs:** https://arxiv.org/abs/2601.00833 · **PDF:** https://arxiv.org/pdf/2601.00833v1
+
+## 25608. Accelerating Scientific Discovery with Autonomous Goal-evolving Agents
+
+- **ID:** `2512.21782` · **primary:** `cs.AI` · **Submitted:** 2025-12-25
+- **Authors:** Yuanqi Du, Botao Yu, Tianyu Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.21782 · **PDF:** https://arxiv.org/pdf/2512.21782v2
+
+## 25609. Inference-based GAN Video Generation
+
+- **ID:** `2512.21776` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Jingbo Yang, Adrian G. Bors
+- **abs:** https://arxiv.org/abs/2512.21776 · **PDF:** https://arxiv.org/pdf/2512.21776v2
+
+## 25610. Compliance Rating Scheme: A Data Provenance Framework for Generative AI Datasets
+
+- **ID:** `2512.21775` · **primary:** `cs.AI` · **Submitted:** 2025-12-25
+- **Authors:** Matyas Bohacek, Ignacio Vilanova Echavarri
+- **abs:** https://arxiv.org/abs/2512.21775 · **PDF:** https://arxiv.org/pdf/2512.21775v1
+
+## 25611. Learning from Negative Examples: Why Warning-Framed Training Data Teaches What It Warns Against
+
+- **ID:** `2512.22293` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Tsogt-Ochir Enkhbayar
+- **abs:** https://arxiv.org/abs/2512.22293 · **PDF:** https://arxiv.org/pdf/2512.22293v1
+
+## 25612. BertsWin: Resolving Topological Sparsity in 3D Masked Autoencoders via Component-Balanced Structural Optimization
+
+- **ID:** `2512.21769` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Evgeny Alves Limarenko, Anastasiia Studenikina
+- **abs:** https://arxiv.org/abs/2512.21769 · **PDF:** https://arxiv.org/pdf/2512.21769v1
+
+## 25613. Assessing the Effectiveness of Membership Inference on Generative Music
+
+- **ID:** `2512.21762` · **primary:** `cs.CR` · **Submitted:** 2025-12-25
+- **Authors:** Kurtis Chow, Omar Samiullah, Vinesh Sridhar, et al.
+- **abs:** https://arxiv.org/abs/2512.21762 · **PDF:** https://arxiv.org/pdf/2512.21762v1
+
+## 25614. A-QCF-Net: An Adaptive Quaternion Cross-Fusion Network for Multimodal Liver Tumor Segmentation from Unpaired Datasets
+
+- **ID:** `2512.21760` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Arunkumar V, Firos V M, Senthilkumar S, et al.
+- **abs:** https://arxiv.org/abs/2512.21760 · **PDF:** https://arxiv.org/pdf/2512.21760v1
+
+## 25615. How Do Agents Perform Code Optimization? An Empirical Study
+
+- **ID:** `2512.21757` · **primary:** `cs.SE` · **Submitted:** 2025-12-25
+- **Authors:** Huiyun Peng, Antonio Zhong, Ricardo Andrés Calvo Méndez, et al.
+- **abs:** https://arxiv.org/abs/2512.21757 · **PDF:** https://arxiv.org/pdf/2512.21757v1
+
+## 25616. Approximation Capabilities of Feedforward Neural Networks with GELU Activations
+
+- **ID:** `2512.21749` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Konstantin Yakovlev, Nikita Puchkin
+- **abs:** https://arxiv.org/abs/2512.21749 · **PDF:** https://arxiv.org/pdf/2512.21749v1
+
+## 25617. A Model of Causal Explanation on Neural Networks for Tabular Data
+
+- **ID:** `2512.21746` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Takashi Isozaki, Masahiro Yamamoto, Atsushi Noda
+- **abs:** https://arxiv.org/abs/2512.21746 · **PDF:** https://arxiv.org/pdf/2512.21746v1
+
+## 25618. Dynamic Feedback Engines: Layer-Wise Control for Self-Regulating Continual Learning
+
+- **ID:** `2512.21743` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Hengyi Wu, Zhenyi Wang, Heng Huang
+- **abs:** https://arxiv.org/abs/2512.21743 · **PDF:** https://arxiv.org/pdf/2512.21743v1
+
+## 25619. HELP: Hierarchical Embodied Language Planner for Household Tasks
+
+- **ID:** `2512.21723` · **primary:** `cs.RO` · **Submitted:** 2025-12-25
+- **Authors:** Alexandr V. Korchemnyi, Anatoly O. Onishchenko, Eva A. Bakaeva, et al.
+- **abs:** https://arxiv.org/abs/2512.21723 · **PDF:** https://arxiv.org/pdf/2512.21723v1
+
+## 25620. Towards representation agnostic probabilistic programming
+
+- **ID:** `2512.23740` · **primary:** `cs.PL` · **Submitted:** 2025-12-25
+- **Authors:** Ole Fenske, Maximilian Popko, Sebastian Bader, et al.
+- **abs:** https://arxiv.org/abs/2512.23740 · **PDF:** https://arxiv.org/pdf/2512.23740v1
+
+## 25621. An Information Theoretic Perspective on Agentic System Design
+
+- **ID:** `2512.21720` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Shizhe He, Avanika Narayan, Ishan S. Khare, et al.
+- **abs:** https://arxiv.org/abs/2512.21720 · **PDF:** https://arxiv.org/pdf/2512.21720v1
+
+## 25622. Multiconnectivity for SAGIN: Current Trends, Challenges, AI-driven Solutions, and Opportunities
+
+- **ID:** `2512.21717` · **primary:** `cs.NI` · **Submitted:** 2025-12-25
+- **Authors:** Abd Ullah Khan, Adnan Shahid, Haejoon Jung, et al.
+- **abs:** https://arxiv.org/abs/2512.21717 · **PDF:** https://arxiv.org/pdf/2512.21717v2
+
+## 25623. CATCH: A Controllable Theme Detection Framework with Contextualized Clustering and Hierarchical Generation
+
+- **ID:** `2512.21715` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Rui Ke, Jiahui Xu, Shenghao Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.21715 · **PDF:** https://arxiv.org/pdf/2512.21715v1
+
+## 25624. Break Out the Silverware -- Semantic Understanding of Stored Household Items
+
+- **ID:** `2512.23739` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Michaela Levi-Richter, Reuth Mirsky, Oren Glickman
+- **abs:** https://arxiv.org/abs/2512.23739 · **PDF:** https://arxiv.org/pdf/2512.23739v1
+
+## 25625. Do Latent Tokens Think? A Causal and Adversarial Analysis of Chain-of-Continuous-Thought
+
+- **ID:** `2512.21711` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Yuyi Zhang, Boyu Tang, Tianjie Ju, et al.
+- **abs:** https://arxiv.org/abs/2512.21711 · **PDF:** https://arxiv.org/pdf/2512.21711v1
+
+## 25626. Detecting AI-Generated Paraphrases in Bengali: A Comparative Study of Zero-Shot and Fine-Tuned Transformers
+
+- **ID:** `2512.21709` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Md. Rakibul Islam, Most. Sharmin Sultana Samu, Md. Zahid Hossain, et al.
+- **abs:** https://arxiv.org/abs/2512.21709 · **PDF:** https://arxiv.org/pdf/2512.21709v1
+
+## 25627. MoRAgent: Parameter Efficient Agent Tuning with Mixture-of-Roles
+
+- **ID:** `2512.21708` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Jing Han, Binwei Yan, Tianyu Guo, et al.
+- **abs:** https://arxiv.org/abs/2512.21708 · **PDF:** https://arxiv.org/pdf/2512.21708v1
+
+## 25628. Enabling Conversational Behavior Reasoning Capabilities in Full-Duplex Speech
+
+- **ID:** `2512.21706` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Shuchang Pan, Siddharth Banerjee, Dhruv Hebbar, et al.
+- **abs:** https://arxiv.org/abs/2512.21706 · **PDF:** https://arxiv.org/pdf/2512.21706v1
+
+## 25629. Multi-Head Spectral-Adaptive Graph Anomaly Detection
+
+- **ID:** `2512.22291` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Qingyue Cao, Bo Jin, Changwei Gong, et al.
+- **abs:** https://arxiv.org/abs/2512.22291 · **PDF:** https://arxiv.org/pdf/2512.22291v1
+
+## 25630. Zero-Shot to Zero-Lies: Detecting Bengali Deepfake Audio through Transfer Learning
+
+- **ID:** `2512.21702` · **primary:** `cs.SD` · **Submitted:** 2025-12-25
+- **Authors:** Most. Sharmin Sultana Samu, Md. Rakibul Islam, Md. Zahid Hossain, et al.
+- **abs:** https://arxiv.org/abs/2512.21702 · **PDF:** https://arxiv.org/pdf/2512.21702v1
+
+## 25631. Towards Responsible and Explainable AI Agents with Consensus-Driven Reasoning
+
+- **ID:** `2512.21699` · **primary:** `cs.AI` · **Submitted:** 2025-12-25
+- **Authors:** Eranga Bandara, Tharaka Hewa, Ross Gore, et al.
+- **abs:** https://arxiv.org/abs/2512.21699 · **PDF:** https://arxiv.org/pdf/2512.21699v1
+
+## 25632. BeHGAN: Bengali Handwritten Word Generation from Plain Text Using Generative Adversarial Networks
+
+- **ID:** `2512.21694` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Md. Rakibul Islam, Md. Kamrozzaman Bhuiyan, Safwan Muntasir, et al.
+- **abs:** https://arxiv.org/abs/2512.21694 · **PDF:** https://arxiv.org/pdf/2512.21694v1
+
+## 25633. RIPCN: A Road Impedance Principal Component Network for Probabilistic Traffic Flow Forecasting
+
+- **ID:** `2512.21685` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Haochen Lv, Yan Lin, Shengnan Guo, et al.
+- **abs:** https://arxiv.org/abs/2512.21685 · **PDF:** https://arxiv.org/pdf/2512.21685v1
+
+## 25634. Compass-Embedding v4: Robust Contrastive Learning for Multilingual E-commerce Embeddings
+
+- **ID:** `2601.11565` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Pakorn Ueareeworakul, Shuman Liu, Jinghao Feng, et al.
+- **abs:** https://arxiv.org/abs/2601.11565 · **PDF:** https://arxiv.org/pdf/2601.11565v1
+
+## 25635. Dictionary-Transform Generative Adversarial Networks
+
+- **ID:** `2512.21677` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Angshul Majumdar
+- **abs:** https://arxiv.org/abs/2512.21677 · **PDF:** https://arxiv.org/pdf/2512.21677v1
+
+## 25636. Comparative Analysis of Deep Learning Models for Perception in Autonomous Vehicles
+
+- **ID:** `2512.21673` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Jalal Khan
+- **abs:** https://arxiv.org/abs/2512.21673 · **PDF:** https://arxiv.org/pdf/2512.21673v1
+
+## 25637. The Deepfake Detective: Interpreting Neural Forensics Through Sparse Features and Manifolds
+
+- **ID:** `2512.21670` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Subramanyam Sahoo, Jared Junkin
+- **abs:** https://arxiv.org/abs/2512.21670 · **PDF:** https://arxiv.org/pdf/2512.21670v1
+
+## 25638. Bayesian Recovery for Probabilistic Coalition Structures
+
+- **ID:** `2601.05273` · **primary:** `cs.GT` · **Submitted:** 2025-12-25
+- **Authors:** Angshul Majumdar
+- **abs:** https://arxiv.org/abs/2601.05273 · **PDF:** https://arxiv.org/pdf/2601.05273v1
+
+## 25639. Near-Optimal Coalition Structures in Polynomial Time
+
+- **ID:** `2512.21657` · **primary:** `cs.GT` · **Submitted:** 2025-12-25
+- **Authors:** Angshul Majumdar
+- **abs:** https://arxiv.org/abs/2512.21657 · **PDF:** https://arxiv.org/pdf/2512.21657v1
+
+## 25640. Structural Induced Exploration for Balanced and Scalable Multi-Robot Path Planning
+
+- **ID:** `2512.21654` · **primary:** `cs.RO` · **Submitted:** 2025-12-25
+- **Authors:** Zikun Guo, Adeyinka P. Adedigba, Rammohan Mallipeddi, et al.
+- **abs:** https://arxiv.org/abs/2512.21654 · **PDF:** https://arxiv.org/pdf/2512.21654v1
+
+## 25641. Semantic Codebooks as Effective Priors for Neural Speech Compression
+
+- **ID:** `2512.21653` · **primary:** `cs.SD` · **Submitted:** 2025-12-25
+- **Authors:** Liuyang Bai, Weiyi Lu, Li Guo
+- **abs:** https://arxiv.org/abs/2512.21653 · **PDF:** https://arxiv.org/pdf/2512.21653v1
+
+## 25642. Enabling Ultra-Fast Cardiovascular Imaging Across Heterogeneous Clinical Environments with a Generalist Foundation Model and Multimodal Database
+
+- **ID:** `2512.21652` · **primary:** `eess.IV` · **Submitted:** 2025-12-25
+- **Authors:** Zi Wang, Mingkai Huang, Zhang Shi, et al.
+- **abs:** https://arxiv.org/abs/2512.21652 · **PDF:** https://arxiv.org/pdf/2512.21652v1
+
+## 25643. When Algorithms Manage Humans: A Double Machine Learning Approach to Estimating Nonlinear Effects of Algorithmic Control on Gig Worker Performance and Wellbeing
+
+- **ID:** `2512.22290` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Arunkumar V, Nivethitha S, Sharan Srinivas, et al.
+- **abs:** https://arxiv.org/abs/2512.22290 · **PDF:** https://arxiv.org/pdf/2512.22290v1
+
+## 25644. Rethinking Output Alignment For 1-bit Post-Training Quantization of Large Language Models
+
+- **ID:** `2512.21651` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Dung Anh Hoang, Cuong Pham, Cuong Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2512.21651 · **PDF:** https://arxiv.org/pdf/2512.21651v1
+
+## 25645. Self-Supervised Masked Autoencoders with Dense-Unet for Coronary Calcium Removal in limited CT Data
+
+- **ID:** `2601.02392` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Mo Chen
+- **abs:** https://arxiv.org/abs/2601.02392 · **PDF:** https://arxiv.org/pdf/2601.02392v1
+
+## 25646. Physic-HM: Restoring Physical Generative Logic in Multimodal Anomaly Detection via Hierarchical Modulation
+
+- **ID:** `2512.21650` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Xiao Liu, Junchen Jin, Yanjie Zhao, et al.
+- **abs:** https://arxiv.org/abs/2512.21650 · **PDF:** https://arxiv.org/pdf/2512.21650v2
+
+## 25647. Variance-Aware Prior-Based Tree Policies for Monte Carlo Tree Search
+
+- **ID:** `2512.21648` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Maximilian Weichart
+- **abs:** https://arxiv.org/abs/2512.21648 · **PDF:** https://arxiv.org/pdf/2512.21648v1
+
+## 25648. Co-GRPO: Co-Optimized Group Relative Policy Optimization for Masked Diffusion Model
+
+- **ID:** `2512.22288` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Renping Zhou, Zanlin Ni, Tianyi Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.22288 · **PDF:** https://arxiv.org/pdf/2512.22288v1
+
+## 25649. TrackTeller: Temporal Multimodal 3D Grounding for Behavior-Dependent Object References
+
+- **ID:** `2512.21641` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Jiahong Yu, Ziqi Wang, Hailiang Zhao, et al.
+- **abs:** https://arxiv.org/abs/2512.21641 · **PDF:** https://arxiv.org/pdf/2512.21641v1
+
+## 25650. Analysis of LLM Vulnerability to GPU Soft Errors: An Instruction-Level Fault Injection Study
+
+- **ID:** `2601.19912` · **primary:** `cs.AR` · **Submitted:** 2025-12-25
+- **Authors:** Duo Chai, Zizhen Liu, Shuhuai Wang, et al.
+- **abs:** https://arxiv.org/abs/2601.19912 · **PDF:** https://arxiv.org/pdf/2601.19912v1
+
+## 25651. Cluster Aggregated GAN (CAG): A Cluster-Based Hybrid Model for Appliance Pattern Generation
+
+- **ID:** `2512.22287` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Zikun Guo, Adeyinka. P. Adedigba, Rammohan Mallipeddi
+- **abs:** https://arxiv.org/abs/2512.22287 · **PDF:** https://arxiv.org/pdf/2512.22287v2
+
+## 25652. Mechanical Strength Prediction of Steel-Polypropylene Fiber-based High-Performance Concrete Using Hybrid Machine Learning Algorithms
+
+- **ID:** `2512.21638` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Jagaran Chakma, Zhiguang Zhou, Badhan Chakma
+- **abs:** https://arxiv.org/abs/2512.21638 · **PDF:** https://arxiv.org/pdf/2512.21638v1
+
+## 25653. Heaven-Sent or Hell-Bent? Benchmarking the Intelligence and Defectiveness of LLM Hallucinations
+
+- **ID:** `2512.21635` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Chengxu Yang, Jingling Yuan, Siqi Cai, et al.
+- **abs:** https://arxiv.org/abs/2512.21635 · **PDF:** https://arxiv.org/pdf/2512.21635v2
+
+## 25654. MAD-NG: Meta-Auto-Decoder Neural Galerkin Method for Solving Parametric Partial Differential Equations
+
+- **ID:** `2512.21633` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Qiuqi Li, Yiting Liu, Jin Zhao, et al.
+- **abs:** https://arxiv.org/abs/2512.21633 · **PDF:** https://arxiv.org/pdf/2512.21633v1
+
+## 25655. Multiple-play Stochastic Bandits with Prioritized Arm Capacity Sharing
+
+- **ID:** `2512.21626` · **primary:** `cs.AI` · **Submitted:** 2025-12-25
+- **Authors:** Hong Xie, Haoran Gu, Yanying Huang, et al.
+- **abs:** https://arxiv.org/abs/2512.21626 · **PDF:** https://arxiv.org/pdf/2512.21626v1
+
+## 25656. Rethinking Sample Polarity in Reinforcement Learning with Verifiable Rewards
+
+- **ID:** `2512.21625` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Xinyu Tang, Yuliang Zhan, Zhixun Li, et al.
+- **abs:** https://arxiv.org/abs/2512.21625 · **PDF:** https://arxiv.org/pdf/2512.21625v1
+
+## 25657. Democratizing Drug Discovery with an Orchestrated, Knowledge-Driven Multi-Agent Team for User-Guided Therapeutic Design
+
+- **ID:** `2512.21623` · **primary:** `cs.AI` · **Submitted:** 2025-12-25
+- **Authors:** Takahide Suzuki, Kazuki Nakanishi, Takashi Fujiwara, et al.
+- **abs:** https://arxiv.org/abs/2512.21623 · **PDF:** https://arxiv.org/pdf/2512.21623v1
+
+## 25658. AMS-IO-Bench and AMS-IO-Agent: Benchmarking and Structured Reasoning for Analog and Mixed-Signal Integrated Circuit Input/Output Design
+
+- **ID:** `2512.21613` · **primary:** `cs.AI` · **Submitted:** 2025-12-25
+- **Authors:** Zhishuai Zhang, Xintian Li, Shilong Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.21613 · **PDF:** https://arxiv.org/pdf/2512.21613v1
+
+## 25659. A Data-Driven Multi-Objective Approach for Predicting Mechanical Performance, Flowability, and Porosity in Ultra-High-Performance Concrete (UHPC)
+
+- **ID:** `2512.21610` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Jagaran Chakma, Zhiguang Zhou, Jyoti Chakma, et al.
+- **abs:** https://arxiv.org/abs/2512.21610 · **PDF:** https://arxiv.org/pdf/2512.21610v1
+
+## 25660. Robustness and Scalability Of Machine Learning for Imbalanced Clinical Data in Emergency and Critical Care
+
+- **ID:** `2512.21602` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Yusuf Brima, Marcellin Atemkeng
+- **abs:** https://arxiv.org/abs/2512.21602 · **PDF:** https://arxiv.org/pdf/2512.21602v1
+
+## 25661. Quantitative Verification of Omega-regular Properties in Probabilistic Programming
+
+- **ID:** `2512.21596` · **primary:** `cs.PL` · **Submitted:** 2025-12-25
+- **Authors:** Peixin Wang, Jianhao Bai, Min Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.21596 · **PDF:** https://arxiv.org/pdf/2512.21596v1
+
+## 25662. LLM-I2I: Boost Your Small Item2Item Recommendation Model with Large Language Model
+
+- **ID:** `2512.21595` · **primary:** `cs.IR` · **Submitted:** 2025-12-25
+- **Authors:** Yinfu Feng, Yanjing Wu, Rong Xiao, et al.
+- **abs:** https://arxiv.org/abs/2512.21595 · **PDF:** https://arxiv.org/pdf/2512.21595v1
+
+## 25663. Residual Prior Diffusion: A Probabilistic Framework Integrating Coarse Latent Priors with Diffusion Models
+
+- **ID:** `2512.21593` · **primary:** `stat.ML` · **Submitted:** 2025-12-25
+- **Authors:** Takuro Kutsuna
+- **abs:** https://arxiv.org/abs/2512.21593 · **PDF:** https://arxiv.org/pdf/2512.21593v1
+
+## 25664. Incorporating rank-free coupling and external field via an amplitude-only modulated spatial photonic Ising machine
+
+- **ID:** `2512.21587` · **primary:** `physics.optics` · **Submitted:** 2025-12-25
+- **Authors:** Ze Zheng, Yuegang Li, Hang Xu, et al.
+- **abs:** https://arxiv.org/abs/2512.21587 · **PDF:** https://arxiv.org/pdf/2512.21587v1
+
+## 25665. Videos are Sample-Efficient Supervisions: Behavior Cloning from Videos via Latent Representations
+
+- **ID:** `2512.21586` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Xin Liu, Haoran Li, Dongbin Zhao
+- **abs:** https://arxiv.org/abs/2512.21586 · **PDF:** https://arxiv.org/pdf/2512.21586v1
+
+## 25666. A Medical Multimodal Diagnostic Framework Integrating Vision-Language Models and Logic Tree Reasoning
+
+- **ID:** `2512.21583` · **primary:** `cs.AI` · **Submitted:** 2025-12-25
+- **Authors:** Zelin Zang, Wenyi Gu, Siqi Ma, et al.
+- **abs:** https://arxiv.org/abs/2512.21583 · **PDF:** https://arxiv.org/pdf/2512.21583v1
+
+## 25667. Gamayun's Path to Multilingual Mastery: Cost-Efficient Training of a 1.5B-Parameter LLM
+
+- **ID:** `2512.21580` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Alexander Podolskiy, Semen Molokov, Timofey Gerasin, et al.
+- **abs:** https://arxiv.org/abs/2512.21580 · **PDF:** https://arxiv.org/pdf/2512.21580v2
+
+## 25668. A General Weighting Theory for Ensemble Learning: Beyond Variance Reduction via Spectral and Geometric Structure
+
+- **ID:** `2512.22286` · **primary:** `stat.ML` · **Submitted:** 2025-12-25
+- **Authors:** Ernest Fokoué
+- **abs:** https://arxiv.org/abs/2512.22286 · **PDF:** https://arxiv.org/pdf/2512.22286v1
+
+## 25669. NEMO-4-PAYPAL: Leveraging NVIDIA's Nemo Framework for empowering PayPal's Commerce Agent
+
+- **ID:** `2512.21578` · **primary:** `cs.AI` · **Submitted:** 2025-12-25
+- **Authors:** Sudhanshu Garg, Andrew Wang, Chaitanya Kulkarni, et al.
+- **abs:** https://arxiv.org/abs/2512.21578 · **PDF:** https://arxiv.org/pdf/2512.21578v3
+
+## 25670. A Unified Definition of Hallucination: It's The World Model, Stupid!
+
+- **ID:** `2512.21577` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Emmy Liu, Varun Gangal, Chelsea Zou, et al.
+- **abs:** https://arxiv.org/abs/2512.21577 · **PDF:** https://arxiv.org/pdf/2512.21577v2
+
+## 25671. Towards Long-window Anchoring in Vision-Language Model Distillation
+
+- **ID:** `2512.21576` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Haoyi Zhou, Shuo Li, Tianyu Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.21576 · **PDF:** https://arxiv.org/pdf/2512.21576v1
+
+## 25672. Context Discipline and Performance Correlation: Analyzing LLM Performance and Quality Degradation Under Varying Context Lengths
+
+- **ID:** `2601.11564` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Ahilan Ayyachamy Nadar Ponnusamy, Karthic Chandran, M Maruf Hossain
+- **abs:** https://arxiv.org/abs/2601.11564 · **PDF:** https://arxiv.org/pdf/2601.11564v1
+
+## 25673. ShrimpXNet: A Transfer Learning Framework for Shrimp Disease Classification with Augmented Regularization, Adversarial Training, and Explainable AI
+
+- **ID:** `2601.00832` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Israk Hasan Jone, D. M. Rafiun Bin Masud, Promit Sarker, et al.
+- **abs:** https://arxiv.org/abs/2601.00832 · **PDF:** https://arxiv.org/pdf/2601.00832v1
+
+## 25674. RefineBridge: Generative Bridge Models Improve Financial Forecasting by Foundation Models
+
+- **ID:** `2512.21572` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Anthony Bolton, Wuyang Zhou, Zehua Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.21572 · **PDF:** https://arxiv.org/pdf/2512.21572v2
+
+## 25675. nncase: An End-to-End Compiler for Efficient LLM Deployment on Heterogeneous Storage Architectures
+
+- **ID:** `2512.21571` · **primary:** `cs.DC` · **Submitted:** 2025-12-25
+- **Authors:** Hui Guo, Qihang Zheng, Chenghai Huo, et al.
+- **abs:** https://arxiv.org/abs/2512.21571 · **PDF:** https://arxiv.org/pdf/2512.21571v1
+
+## 25676. AnchorGK: Anchor-based Incremental and Stratified Graph Learning Framework for Inductive Spatio-Temporal Kriging
+
+- **ID:** `2512.21569` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Xiaobin Ren, Kaiqi Zhao, Katerina Taškova, et al.
+- **abs:** https://arxiv.org/abs/2512.21569 · **PDF:** https://arxiv.org/pdf/2512.21569v1
+
+## 25677. Beyond Heuristics: A Decision-Theoretic Framework for Agent Memory Management
+
+- **ID:** `2512.21567` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Changzhi Sun, Xiangyu Chen, Jixiang Luo, et al.
+- **abs:** https://arxiv.org/abs/2512.21567 · **PDF:** https://arxiv.org/pdf/2512.21567v1
+
+## 25678. Discovering Sparse Recovery Algorithms Using Neural Architecture Search
+
+- **ID:** `2512.21563` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Patrick Yubeaton, Sarthak Gupta, M. Salman Asif, et al.
+- **abs:** https://arxiv.org/abs/2512.21563 · **PDF:** https://arxiv.org/pdf/2512.21563v1
+
+## 25679. Exploration of Reproducible Generated Image Detection
+
+- **ID:** `2512.21562` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Yihang Duan
+- **abs:** https://arxiv.org/abs/2512.21562 · **PDF:** https://arxiv.org/pdf/2512.21562v1
+
+## 25680. Bidirectional Human-AI Alignment in Education for Trustworthy Learning Environments
+
+- **ID:** `2512.21552` · **primary:** `cs.CY` · **Submitted:** 2025-12-25
+- **Authors:** Hua Shen
+- **abs:** https://arxiv.org/abs/2512.21552 · **PDF:** https://arxiv.org/pdf/2512.21552v1
+
+## 25681. Human-AI Interaction Alignment: Designing, Evaluating, and Evolving Value-Centered AI For Reciprocal Human-AI Futures
+
+- **ID:** `2512.21551` · **primary:** `cs.HC` · **Submitted:** 2025-12-25
+- **Authors:** Hua Shen, Tiffany Knearem, Divy Thakkar, et al.
+- **abs:** https://arxiv.org/abs/2512.21551 · **PDF:** https://arxiv.org/pdf/2512.21551v1
+
+## 25682. Horizon Reduction as Information Loss in Offline Reinforcement Learning
+
+- **ID:** `2601.00831` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Uday Kumar Nidadala, Venkata Bhumika Guthi
+- **abs:** https://arxiv.org/abs/2601.00831 · **PDF:** https://arxiv.org/pdf/2601.00831v1
+
+## 25683. AVP-Fusion: Adaptive Multi-Modal Fusion and Contrastive Learning for Two-Stage Antiviral Peptide Identification
+
+- **ID:** `2512.21544` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Xinru Wen, Weizhong Lin, Xuan Xiao
+- **abs:** https://arxiv.org/abs/2512.21544 · **PDF:** https://arxiv.org/pdf/2512.21544v1
+
+## 25684. Leash: Adaptive Length Penalty and Reward Shaping for Efficient Large Reasoning Model
+
+- **ID:** `2512.21540` · **primary:** `cs.AI` · **Submitted:** 2025-12-25
+- **Authors:** Yanhao Li, Lu Ma, Jiaran Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.21540 · **PDF:** https://arxiv.org/pdf/2512.21540v1
+
+## 25685. On Fibonacci Ensembles: An Alternative Approach to Ensemble Learning Inspired by the Timeless Architecture of the Golden Ratio
+
+- **ID:** `2512.22284` · **primary:** `stat.ML` · **Submitted:** 2025-12-25
+- **Authors:** Ernest Fokoué
+- **abs:** https://arxiv.org/abs/2512.22284 · **PDF:** https://arxiv.org/pdf/2512.22284v1
+
+## 25686. Synergizing Kolmogorov-Arnold Networks with Dynamic Adaptive Weighting for High-Frequency and Multi-Scale PDE Solutions
+
+- **ID:** `2512.22283` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Guokan Chen, Yao Xiao, Bin Fan, et al.
+- **abs:** https://arxiv.org/abs/2512.22283 · **PDF:** https://arxiv.org/pdf/2512.22283v3
+
+## 25687. Hierarchy-Aware Fine-Tuning of Vision-Language Models
+
+- **ID:** `2512.21529` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Jiayu Li, Rajesh Gangireddy, Samet Akcay, et al.
+- **abs:** https://arxiv.org/abs/2512.21529 · **PDF:** https://arxiv.org/pdf/2512.21529v1
+
+## 25688. A review of NMF, PLSA, LBA, EMA, and LCA with a focus on the identifiability issue
+
+- **ID:** `2512.22282` · **primary:** `stat.ML` · **Submitted:** 2025-12-25
+- **Authors:** Qianqian Qi, Peter G. M. van der Heijden
+- **abs:** https://arxiv.org/abs/2512.22282 · **PDF:** https://arxiv.org/pdf/2512.22282v1
+
+## 25689. WearVox: An Egocentric Multichannel Voice Assistant Benchmark for Wearables
+
+- **ID:** `2601.02391` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Zhaojiang Lin, Yong Xu, Kai Sun, et al.
+- **abs:** https://arxiv.org/abs/2601.02391 · **PDF:** https://arxiv.org/pdf/2601.02391v1
+
+## 25690. Generative Actor Critic
+
+- **ID:** `2512.21527` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Aoyang Qin, Deqian Kong, Wei Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.21527 · **PDF:** https://arxiv.org/pdf/2512.21527v1
+
+## 25691. Selective LLM-Guided Regularization for Enhancing Recommendation Models
+
+- **ID:** `2512.21526` · **primary:** `cs.IR` · **Submitted:** 2025-12-25
+- **Authors:** Shanglin Yang, Zhan Shi
+- **abs:** https://arxiv.org/abs/2512.21526 · **PDF:** https://arxiv.org/pdf/2512.21526v1
+
+## 25692. Enforcing Temporal Constraints for LLM Agents
+
+- **ID:** `2512.23738` · **primary:** `cs.PL` · **Submitted:** 2025-12-25
+- **Authors:** Adharsh Kamath, Sishen Zhang, Calvin Xu, et al.
+- **abs:** https://arxiv.org/abs/2512.23738 · **PDF:** https://arxiv.org/pdf/2512.23738v1
+
+## 25693. First Provable Guarantees for Practical Private FL: Beyond Restrictive Assumptions
+
+- **ID:** `2512.21521` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Egor Shulgin, Grigory Malinovsky, Sarit Khirirat, et al.
+- **abs:** https://arxiv.org/abs/2512.21521 · **PDF:** https://arxiv.org/pdf/2512.21521v1
+
+## 25694. Valori: A Deterministic Memory Substrate for AI Systems
+
+- **ID:** `2512.22280` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Varshith Gudur
+- **abs:** https://arxiv.org/abs/2512.22280 · **PDF:** https://arxiv.org/pdf/2512.22280v1
+
+## 25695. Global-Graph Guided and Local-Graph Weighted Contrastive Learning for Unified Clustering on Incomplete and Noise Multi-View Data
+
+- **ID:** `2512.21516` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Hongqing He, Jie Xu, Wenyuan Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.21516 · **PDF:** https://arxiv.org/pdf/2512.21516v1
+
+## 25696. Perplexity-Aware Data Scaling Law: Perplexity Landscapes Predict Performance for Continual Pre-training
+
+- **ID:** `2512.21515` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Lei Liu, Hao Zhu, Yue Shen, et al.
+- **abs:** https://arxiv.org/abs/2512.21515 · **PDF:** https://arxiv.org/pdf/2512.21515v1
+
+## 25697. DiverseGRPO: Mitigating Mode Collapse in Image Generation via Diversity-Aware GRPO
+
+- **ID:** `2512.21514` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Henglin Liu, Huijuan Huang, Jing Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.21514 · **PDF:** https://arxiv.org/pdf/2512.21514v1
+
+## 25698. Hierarchical Stacking Optimization Using Dirichlet's Process (SoDip): Towards Accelerated Design for Graft Polymerization
+
+- **ID:** `2512.22279` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Amgad Ahmed Ali Ibrahim, Hein Htet, Ryoji Asahi
+- **abs:** https://arxiv.org/abs/2512.22279 · **PDF:** https://arxiv.org/pdf/2512.22279v1
+
+## 25699. Can We Trust AI Explanations? Evidence of Systematic Underreporting in Chain-of-Thought Reasoning
+
+- **ID:** `2601.00830` · **primary:** `cs.AI` · **Submitted:** 2025-12-25
+- **Authors:** Deep Pankajbhai Mehta
+- **abs:** https://arxiv.org/abs/2601.00830 · **PDF:** https://arxiv.org/pdf/2601.00830v1
+
+## 25700. Missing Pattern Tree based Decision Grouping and Ensemble for Deep Incomplete Multi-View Clustering
+
+- **ID:** `2512.21510` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Wenyuan Yang, Jie Xu, Hongqing He, et al.
+- **abs:** https://arxiv.org/abs/2512.21510 · **PDF:** https://arxiv.org/pdf/2512.21510v1
+
+## 25701. MotionTeller: Multi-modal Integration of Wearable Time-Series with LLMs for Health and Behavioral Understanding
+
+- **ID:** `2512.21506` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Aiwei Zhang, Arvind Pillai, Andrew Campbell, et al.
+- **abs:** https://arxiv.org/abs/2512.21506 · **PDF:** https://arxiv.org/pdf/2512.21506v1
+
+## 25702. Oogiri-Master: Benchmarking Humor Understanding via Oogiri
+
+- **ID:** `2512.21494` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Soichiro Murakami, Hidetaka Kamigaito, Hiroya Takamura, et al.
+- **abs:** https://arxiv.org/abs/2512.21494 · **PDF:** https://arxiv.org/pdf/2512.21494v1
+
+## 25703. The Illusion of Clinical Reasoning: A Benchmark Reveals the Pervasive Gap in Vision-Language Models for Clinical Competency
+
+- **ID:** `2512.22275` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Dingyu Wang, Zimu Yuan, Jiajun Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.22275 · **PDF:** https://arxiv.org/pdf/2512.22275v1
+
+## 25704. Efficient MoE Inference with Fine-Grained Scheduling of Disaggregated Expert Parallelism
+
+- **ID:** `2512.21487` · **primary:** `cs.DC` · **Submitted:** 2025-12-25
+- **Authors:** Xinglin Pan, Shaohuai Shi, Wenxiang Lin, et al.
+- **abs:** https://arxiv.org/abs/2512.21487 · **PDF:** https://arxiv.org/pdf/2512.21487v1
+
+## 25705. When Bayesian Tensor Completion Meets Multioutput Gaussian Processes: Functional Universality and Rank Learning
+
+- **ID:** `2512.21486` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Siyuan Li, Shikai Fang, Lei Cheng, et al.
+- **abs:** https://arxiv.org/abs/2512.21486 · **PDF:** https://arxiv.org/pdf/2512.21486v1
+
+## 25706. LogicLens: Visual-Logical Co-Reasoning for Text-Centric Forgery Analysis
+
+- **ID:** `2512.21482` · **primary:** `cs.AI` · **Submitted:** 2025-12-25
+- **Authors:** Fanwei Zeng, Changtao Miao, Jing Huang, et al.
+- **abs:** https://arxiv.org/abs/2512.21482 · **PDF:** https://arxiv.org/pdf/2512.21482v1
+
+## 25707. Quantum Nondecimated Wavelet Transform: Theory, Circuits, and Applications
+
+- **ID:** `2512.21478` · **primary:** `quant-ph` · **Submitted:** 2025-12-25
+- **Authors:** Brani Vidakovic
+- **abs:** https://arxiv.org/abs/2512.21478 · **PDF:** https://arxiv.org/pdf/2512.21478v1
+
+## 25708. GPF-Net: Gated Progressive Fusion Learning for Polyp Re-Identification
+
+- **ID:** `2512.21476` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Suncheng Xiang, Xiaoyang Wang, Junjie Jiang, et al.
+- **abs:** https://arxiv.org/abs/2512.21476 · **PDF:** https://arxiv.org/pdf/2512.21476v1
+
+## 25709. SALP-CG: Standard-Aligned LLM Pipeline for Classifying and Grading Large Volumes of Online Conversational Health Data
+
+- **ID:** `2601.09717` · **primary:** `cs.CL` · **Submitted:** 2025-12-25
+- **Authors:** Yiwei Yan, Hao Li, Hua He, et al.
+- **abs:** https://arxiv.org/abs/2601.09717 · **PDF:** https://arxiv.org/pdf/2601.09717v1
+
+## 25710. CCAD: Compressed Global Feature Conditioned Anomaly Detection
+
+- **ID:** `2512.21459` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Xiao Jin, Liang Diao, Qixin Xiao, et al.
+- **abs:** https://arxiv.org/abs/2512.21459 · **PDF:** https://arxiv.org/pdf/2512.21459v1
+
+## 25711. Statistical vs. Deep Learning Models for Estimating Substance Overdose Excess Mortality in the US
+
+- **ID:** `2512.21456` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Sukanya Krishna, Marie-Laure Charpignon, Maimuna Majumder
+- **abs:** https://arxiv.org/abs/2512.21456 · **PDF:** https://arxiv.org/pdf/2512.21456v1
+
+## 25712. Intelligent recognition of GPR road hidden defect images based on feature fusion and attention mechanism
+
+- **ID:** `2512.21452` · **primary:** `cs.CV` · **Submitted:** 2025-12-25
+- **Authors:** Haotian Lv, Yuhui Zhang, Jiangbo Dai, et al.
+- **abs:** https://arxiv.org/abs/2512.21452 · **PDF:** https://arxiv.org/pdf/2512.21452v1
+
+## 25713. An approach to Fisher-Rao metric for infinite dimensional non-parametric information geometry
+
+- **ID:** `2512.21451` · **primary:** `stat.ML` · **Submitted:** 2025-12-25
+- **Authors:** Bing Cheng, Howell Tong
+- **abs:** https://arxiv.org/abs/2512.21451 · **PDF:** https://arxiv.org/pdf/2512.21451v2
+
+## 25714. RLLaVA: An RL-central Framework for Language and Vision Assistants
+
+- **ID:** `2512.21450` · **primary:** `cs.LG` · **Submitted:** 2025-12-25
+- **Authors:** Lei Zhao, Zihao Ma, Boyu Lin, et al.
+- **abs:** https://arxiv.org/abs/2512.21450 · **PDF:** https://arxiv.org/pdf/2512.21450v1
+
+## 25715. An Equivariance Toolbox for Learning Dynamics
+
+- **ID:** `2512.21447` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Yongyi Yang, Liu Ziyin
+- **abs:** https://arxiv.org/abs/2512.21447 · **PDF:** https://arxiv.org/pdf/2512.21447v1
+
+## 25716. dUltra: Ultra-Fast Diffusion Language Models via Reinforcement Learning
+
+- **ID:** `2512.21446` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Shirui Chen, Jiantao Jiao, Lillian J. Ratliff, et al.
+- **abs:** https://arxiv.org/abs/2512.21446 · **PDF:** https://arxiv.org/pdf/2512.21446v2
+
+## 25717. Fuzzwise: Intelligent Initial Corpus Generation for Fuzzing
+
+- **ID:** `2512.21440` · **primary:** `cs.SE` · **Submitted:** 2025-12-24
+- **Authors:** Hridya Dhulipala, Xiaokai Rong, Aashish Yadavally, et al.
+- **abs:** https://arxiv.org/abs/2512.21440 · **PDF:** https://arxiv.org/pdf/2512.21440v1
+
+## 25718. Morality is Contextual: Learning Interpretable Moral Contexts from Human Data with Probabilistic Clustering and Large Language Models
+
+- **ID:** `2512.21439` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Geoffroy Morlat, Marceau Nahon, Augustin Chartouny, et al.
+- **abs:** https://arxiv.org/abs/2512.21439 · **PDF:** https://arxiv.org/pdf/2512.21439v1
+
+## 25719. Decomposing LLM Self-Correction: The Accuracy-Correction Paradox and Error Depth Hypothesis
+
+- **ID:** `2601.00828` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Yin Li
+- **abs:** https://arxiv.org/abs/2601.00828 · **PDF:** https://arxiv.org/pdf/2601.00828v1
+
+## 25720. Dynamic Attention (DynAttn): Interpretable High-Dimensional Spatio-Temporal Forecasting (with Application to Conflict Fatalities)
+
+- **ID:** `2512.21435` · **primary:** `stat.AP` · **Submitted:** 2025-12-24
+- **Authors:** Stefano M. Iacus, Haodong Qi, Marcello Carammia, et al.
+- **abs:** https://arxiv.org/abs/2512.21435 · **PDF:** https://arxiv.org/pdf/2512.21435v1
+
+## 25721. Scalable Deep Subspace Clustering Network
+
+- **ID:** `2512.21434` · **primary:** `cs.CV` · **Submitted:** 2025-12-24
+- **Authors:** Nairouz Mrabah, Mohamed Bouguessa, Sihem Sami
+- **abs:** https://arxiv.org/abs/2512.21434 · **PDF:** https://arxiv.org/pdf/2512.21434v1
+
+## 25722. DeepCQ: General-Purpose Deep-Surrogate Framework for Lossy Compression Quality Prediction
+
+- **ID:** `2512.21433` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Khondoker Mirazul Mumenin, Robert Underwood, Dong Dai, et al.
+- **abs:** https://arxiv.org/abs/2512.21433 · **PDF:** https://arxiv.org/pdf/2512.21433v1
+
+## 25723. Cerberus: Multi-Agent Reasoning and Coverage-Guided Exploration for Static Detection of Runtime Errors
+
+- **ID:** `2512.21431` · **primary:** `cs.SE` · **Submitted:** 2025-12-24
+- **Authors:** Hridya Dhulipala, Xiaokai Rong, Tien N. Nguyen
+- **abs:** https://arxiv.org/abs/2512.21431 · **PDF:** https://arxiv.org/pdf/2512.21431v1
+
+## 25724. Teaching People LLM's Errors and Getting it Right
+
+- **ID:** `2512.21422` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Nathan Stringham, Fateme Hashemi Chaleshtori, Xinyuan Yan, et al.
+- **abs:** https://arxiv.org/abs/2512.21422 · **PDF:** https://arxiv.org/pdf/2512.21422v1
+
+## 25725. Three-way decision with incomplete information based on similarity and satisfiability
+
+- **ID:** `2512.21421` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Junfang Luo, Mengjun Hu, Keyun Qin
+- **abs:** https://arxiv.org/abs/2512.21421 · **PDF:** https://arxiv.org/pdf/2512.21421v1
+
+## 25726. Feasible strategies in three-way conflict analysis with three-valued ratings
+
+- **ID:** `2512.21420` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Jing Liu, Mengjun Hu, Guangming Lang
+- **abs:** https://arxiv.org/abs/2512.21420 · **PDF:** https://arxiv.org/pdf/2512.21420v2
+
+## 25727. Three-way conflict analysis based on alliance and conflict functions
+
+- **ID:** `2512.21419` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Junfang Luo, Mengjun Hu, Guangming Lang, et al.
+- **abs:** https://arxiv.org/abs/2512.21419 · **PDF:** https://arxiv.org/pdf/2512.21419v1
+
+## 25728. A Tool Bottleneck Framework for Clinically-Informed and Interpretable Medical Image Understanding
+
+- **ID:** `2512.21414` · **primary:** `cs.CV` · **Submitted:** 2025-12-24
+- **Authors:** Christina Liu, Alan Q. Wang, Joy Hsu, et al.
+- **abs:** https://arxiv.org/abs/2512.21414 · **PDF:** https://arxiv.org/pdf/2512.21414v1
+
+## 25729. A Survey of Freshness-Aware Wireless Networking with Reinforcement Learning
+
+- **ID:** `2512.21412` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Alimu Alibotaiken, Suyang Wang, Oluwaseun T. Ajayi, et al.
+- **abs:** https://arxiv.org/abs/2512.21412 · **PDF:** https://arxiv.org/pdf/2512.21412v1
+
+## 25730. Opportunities and Challenges of Natural Language Processing for Low-Resource Senegalese Languages in Social Science Research
+
+- **ID:** `2601.09716` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Derguene Mbaye, Tatiana D. P. Mbengue, Madoune R. Seye, et al.
+- **abs:** https://arxiv.org/abs/2601.09716 · **PDF:** https://arxiv.org/pdf/2601.09716v1
+
+## 25731. Singular Fluctuation as Specific Heat in Bayesian Learning
+
+- **ID:** `2512.21411` · **primary:** `math.ST` · **Submitted:** 2025-12-24
+- **Authors:** Sean Plummer
+- **abs:** https://arxiv.org/abs/2512.21411 · **PDF:** https://arxiv.org/pdf/2512.21411v2
+
+## 25732. kooplearn: A Scikit-Learn Compatible Library of Algorithms for Evolution Operator Learning
+
+- **ID:** `2512.21409` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Giacomo Turri, Grégoire Pacreau, Giacomo Meanti, et al.
+- **abs:** https://arxiv.org/abs/2512.21409 · **PDF:** https://arxiv.org/pdf/2512.21409v2
+
+## 25733. LLM-Driven Feature-Level Adversarial Attacks on Android Malware Detectors
+
+- **ID:** `2512.21404` · **primary:** `cs.CR` · **Submitted:** 2025-12-24
+- **Authors:** Tianwei Lan, Farid Naït-Abdesselam
+- **abs:** https://arxiv.org/abs/2512.21404 · **PDF:** https://arxiv.org/pdf/2512.21404v1
+
+## 25734. Governing Cloud Data Pipelines with Agentic AI
+
+- **ID:** `2512.23737` · **primary:** `cs.DC` · **Submitted:** 2025-12-24
+- **Authors:** Aswathnarayan Muthukrishnan Kirubakaran, Adithya Parthasarathy, Nitin Saksena, et al.
+- **abs:** https://arxiv.org/abs/2512.23737 · **PDF:** https://arxiv.org/pdf/2512.23737v1
+
+## 25735. Learning to Reconfigure: Using Device Status to Select the Right Constrained Coding Scheme
+
+- **ID:** `2512.21396` · **primary:** `cs.IT` · **Submitted:** 2025-12-24
+- **Authors:** Doğukan Özbayrak, Ahmed Hareedy
+- **abs:** https://arxiv.org/abs/2512.21396 · **PDF:** https://arxiv.org/pdf/2512.21396v1
+
+## 25736. A Reinforcement Learning Approach to Synthetic Data Generation
+
+- **ID:** `2512.21395` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Natalia Espinosa-Dice, Nicholas J. Jackson, Chao Yan, et al.
+- **abs:** https://arxiv.org/abs/2512.21395 · **PDF:** https://arxiv.org/pdf/2512.21395v2
+
+## 25737. Deep learning-enhanced dual-mode multiplexed optical sensor for point-of-care diagnostics of cardiovascular diseases
+
+- **ID:** `2512.21389` · **primary:** `physics.med-ph` · **Submitted:** 2025-12-24
+- **Authors:** Gyeo-Re Han, Merve Eryilmaz, Artem Goncharov, et al.
+- **abs:** https://arxiv.org/abs/2512.21389 · **PDF:** https://arxiv.org/pdf/2512.21389v1
+
+## 25738. Optimizing Decoding Paths in Masked Diffusion Models by Quantifying Uncertainty
+
+- **ID:** `2512.21336` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Ziyu Chen, Xinbei Jiang, Peng Sun, et al.
+- **abs:** https://arxiv.org/abs/2512.21336 · **PDF:** https://arxiv.org/pdf/2512.21336v1
+
+## 25739. Autonomous Uncertainty Quantification for Computational Point-of-care Sensors
+
+- **ID:** `2512.21335` · **primary:** `physics.med-ph` · **Submitted:** 2025-12-24
+- **Authors:** Artem Goncharov, Rajesh Ghosh, Hyou-Arm Joung, et al.
+- **abs:** https://arxiv.org/abs/2512.21335 · **PDF:** https://arxiv.org/pdf/2512.21335v1
+
+## 25740. C2LLM Technical Report: A New Frontier in Code Retrieval via Adaptive Cross-Attention Pooling
+
+- **ID:** `2512.21332` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Jin Qin, Zihan Liao, Ziyin Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.21332 · **PDF:** https://arxiv.org/pdf/2512.21332v1
+
+## 25741. Your Reasoning Benchmark May Not Test Reasoning: Revealing Perception Bottleneck in Abstract Reasoning Benchmarks
+
+- **ID:** `2512.21329` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Xinhe Wang, Jin Huang, Xingjian Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.21329 · **PDF:** https://arxiv.org/pdf/2512.21329v2
+
+## 25742. Measuring all the noises of LLM Evals
+
+- **ID:** `2512.21326` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Sida Wang
+- **abs:** https://arxiv.org/abs/2512.21326 · **PDF:** https://arxiv.org/pdf/2512.21326v2
+
+## 25743. Parallel Token Prediction for Language Models
+
+- **ID:** `2512.21323` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Felix Draxler, Justus Will, Farrin Marouf Sofian, et al.
+- **abs:** https://arxiv.org/abs/2512.21323 · **PDF:** https://arxiv.org/pdf/2512.21323v2
+
+## 25744. Variationally correct operator learning: Reduced basis neural operator with a posteriori error estimation
+
+- **ID:** `2512.21319` · **primary:** `math.NA` · **Submitted:** 2025-12-24
+- **Authors:** Yuan Qiu, Wolfgang Dahmen, Peng Chen
+- **abs:** https://arxiv.org/abs/2512.21319 · **PDF:** https://arxiv.org/pdf/2512.21319v1
+
+## 25745. Scaling Laws for Economic Productivity: Experimental Evidence in LLM-Assisted Consulting, Data Analyst, and Management Tasks
+
+- **ID:** `2512.21316` · **primary:** `econ.GN` · **Submitted:** 2025-12-24
+- **Authors:** Ali Merali
+- **abs:** https://arxiv.org/abs/2512.21316 · **PDF:** https://arxiv.org/pdf/2512.21316v1
+
+## 25746. Does the Data Processing Inequality Reflect Practice? On the Utility of Low-Level Tasks
+
+- **ID:** `2512.21315` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Roy Turgeman, Tom Tirer
+- **abs:** https://arxiv.org/abs/2512.21315 · **PDF:** https://arxiv.org/pdf/2512.21315v1
+
+## 25747. Learning to Solve PDEs on Neural Shape Representations
+
+- **ID:** `2512.21311` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Lilian Welschinger, Yilin Liu, Zican Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.21311 · **PDF:** https://arxiv.org/pdf/2512.21311v1
+
+## 25748. LLMTM: Benchmarking and Optimizing LLMs for Temporal Motif Analysis in Dynamic Graphs
+
+- **ID:** `2512.22266` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Bing Hao, Minglai Shao, Zengyi Wo, et al.
+- **abs:** https://arxiv.org/abs/2512.22266 · **PDF:** https://arxiv.org/pdf/2512.22266v2
+
+## 25749. Transcriptome-Conditioned Personalized De Novo Drug Generation for AML Using Metaheuristic Assembly and Target-Driven Filtering
+
+- **ID:** `2512.21301` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Abdullah G. Elafifi, Basma Mamdouh, Mariam Hanafy, et al.
+- **abs:** https://arxiv.org/abs/2512.21301 · **PDF:** https://arxiv.org/pdf/2512.21301v1
+
+## 25750. LuxIA: A Lightweight Unitary matriX-based Framework Built on an Iterative Algorithm for Photonic Neural Network Training
+
+- **ID:** `2512.22264` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Tzamn Melendez Carmona, Federico Marchesin, Marco P. Abrate, et al.
+- **abs:** https://arxiv.org/abs/2512.22264 · **PDF:** https://arxiv.org/pdf/2512.22264v1
+
+## 25751. Sensitivity Analysis of the Consistency Assumption
+
+- **ID:** `2512.21379` · **primary:** `stat.ME` · **Submitted:** 2025-12-24
+- **Authors:** Brian Knaeble, Qinyun Lin, Erich Kummerfeld, et al.
+- **abs:** https://arxiv.org/abs/2512.21379 · **PDF:** https://arxiv.org/pdf/2512.21379v1
+
+## 25752. Evaluating an Adaptive Multispectral Turret System for Autonomous Tracking Across Variable Illumination Conditions
+
+- **ID:** `2512.22263` · **primary:** `cs.CV` · **Submitted:** 2025-12-24
+- **Authors:** Aahan Sachdeva, Dhanvinkumar Ganeshkumar, James E. Gallagher, et al.
+- **abs:** https://arxiv.org/abs/2512.22263 · **PDF:** https://arxiv.org/pdf/2512.22263v1
+
+## 25753. Model Merging via Multi-Teacher Knowledge Distillation
+
+- **ID:** `2512.21288` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Seyed Arshan Dalili, Mehrdad Mahdavi
+- **abs:** https://arxiv.org/abs/2512.21288 · **PDF:** https://arxiv.org/pdf/2512.21288v1
+
+## 25754. SMART SLM: Structured Memory and Reasoning Transformer, A Small Language Model for Accurate Document Assistance
+
+- **ID:** `2512.21280` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Divij Dudeja, Mayukha Pal
+- **abs:** https://arxiv.org/abs/2512.21280 · **PDF:** https://arxiv.org/pdf/2512.21280v1
+
+## 25755. ReaSeq: Unleashing World Knowledge via Reasoning for Sequential Modeling
+
+- **ID:** `2512.21257` · **primary:** `cs.IR` · **Submitted:** 2025-12-24
+- **Authors:** Jiakai Tang, Chuan Wang, Gaoming Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.21257 · **PDF:** https://arxiv.org/pdf/2512.21257v2
+
+## 25756. Learning Factors in AI-Augmented Education: A Comparative Study of Middle and High School Students
+
+- **ID:** `2512.21246` · **primary:** `cs.HC` · **Submitted:** 2025-12-24
+- **Authors:** Gaia Ebli, Bianca Raimondi, Maurizio Gabbrielli
+- **abs:** https://arxiv.org/abs/2512.21246 · **PDF:** https://arxiv.org/pdf/2512.21246v1
+
+## 25757. LookPlanGraph: Embodied Instruction Following Method with VLM Graph Augmentation
+
+- **ID:** `2512.21243` · **primary:** `cs.RO` · **Submitted:** 2025-12-24
+- **Authors:** Anatoly O. Onishchenko, Alexey K. Kovalev, Aleksandr I. Panov
+- **abs:** https://arxiv.org/abs/2512.21243 · **PDF:** https://arxiv.org/pdf/2512.21243v1
+
+## 25758. Improving the Convergence Rate of Ray Search Optimization for Query-Efficient Hard-Label Attacks
+
+- **ID:** `2512.21241` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Xinjie Xu, Shuyu Cheng, Dongwei Xu, et al.
+- **abs:** https://arxiv.org/abs/2512.21241 · **PDF:** https://arxiv.org/pdf/2512.21241v1
+
+## 25759. Introducing Axlerod: An LLM-based Chatbot for Assisting Independent Insurance Agents
+
+- **ID:** `2601.09715` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Adam Bradley, John Hastings, Khandaker Mamun Ahmed
+- **abs:** https://arxiv.org/abs/2601.09715 · **PDF:** https://arxiv.org/pdf/2601.09715v1
+
+## 25760. Assessing the Software Security Comprehension of Large Language Models
+
+- **ID:** `2512.21238` · **primary:** `cs.SE` · **Submitted:** 2025-12-24
+- **Authors:** Mohammed Latif Siddiq, Natalie Sekerak, Antonio Karam, et al.
+- **abs:** https://arxiv.org/abs/2512.21238 · **PDF:** https://arxiv.org/pdf/2512.21238v1
+
+## 25761. Casting a SPELL: Sentence Pairing Exploration for LLM Limitation-breaking
+
+- **ID:** `2512.21236` · **primary:** `cs.CR` · **Submitted:** 2025-12-24
+- **Authors:** Yifan Huang, Xiaojun Jia, Wenbo Guo, et al.
+- **abs:** https://arxiv.org/abs/2512.21236 · **PDF:** https://arxiv.org/pdf/2512.21236v1
+
+## 25762. MiST: Understanding the Role of Mid-Stage Scientific Training in Developing Chemical Reasoning Models
+
+- **ID:** `2512.21231` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Andres M Bran, Tong Xie, Shai Pranesh, et al.
+- **abs:** https://arxiv.org/abs/2512.21231 · **PDF:** https://arxiv.org/pdf/2512.21231v3
+
+## 25763. PhononBench:A Large-Scale Phonon-Based Benchmark for Dynamical Stability in Crystal Generation
+
+- **ID:** `2512.21227` · **primary:** `cond-mat.mtrl-sci` · **Submitted:** 2025-12-24
+- **Authors:** Xiao-Qi Han, Peng-Jie Guo, Ze-Feng Gao, et al.
+- **abs:** https://arxiv.org/abs/2512.21227 · **PDF:** https://arxiv.org/pdf/2512.21227v2
+
+## 25764. Leveraging Lightweight Entity Extraction for Scalable Event-Based Image Retrieval
+
+- **ID:** `2512.21221` · **primary:** `cs.CV` · **Submitted:** 2025-12-24
+- **Authors:** Dao Sy Duy Minh, Huynh Trung Kiet, Nguyen Lam Phu Quy, et al.
+- **abs:** https://arxiv.org/abs/2512.21221 · **PDF:** https://arxiv.org/pdf/2512.21221v1
+
+## 25765. RoboSafe: Safeguarding Embodied Agents via Executable Safety Logic
+
+- **ID:** `2512.21220` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Le Wang, Zonghao Ying, Xiao Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.21220 · **PDF:** https://arxiv.org/pdf/2512.21220v2
+
+## 25766. Causal-driven attribution (CDA): Estimating channel influence without user-level data
+
+- **ID:** `2512.21211` · **primary:** `stat.ML` · **Submitted:** 2025-12-24
+- **Authors:** Georgios Filippou, Boi Mai Quach, Diana Lenghel, et al.
+- **abs:** https://arxiv.org/abs/2512.21211 · **PDF:** https://arxiv.org/pdf/2512.21211v1
+
+## 25767. Analytic and Variational Stability in Deep Learning Systems
+
+- **ID:** `2512.21208` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Ronald Katende
+- **abs:** https://arxiv.org/abs/2512.21208 · **PDF:** https://arxiv.org/pdf/2512.21208v2
+
+## 25768. INSIGHT: Spatially resolved survival modelling from routine histology crosslinked with molecular profiling reveals prognostic epithelial-immune axes in stage II/III colorectal cancer
+
+- **ID:** `2512.22262` · **primary:** `q-bio.QM` · **Submitted:** 2025-12-24
+- **Authors:** Piotr Keller, Mark Eastwood, Zedong Hu, et al.
+- **abs:** https://arxiv.org/abs/2512.22262 · **PDF:** https://arxiv.org/pdf/2512.22262v1
+
+## 25769. The Physics Constraint Paradox: When Removing Explicit Constraints Improves Physics-Informed Data for Machine Learning
+
+- **ID:** `2512.22261` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Rahul D Ray
+- **abs:** https://arxiv.org/abs/2512.22261 · **PDF:** https://arxiv.org/pdf/2512.22261v1
+
+## 25770. SpidR-Adapt: A Universal Speech Representation Model for Few-Shot Adaptation
+
+- **ID:** `2512.21204` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Mahi Luthra, Jiayi Shen, Maxime Poli, et al.
+- **abs:** https://arxiv.org/abs/2512.21204 · **PDF:** https://arxiv.org/pdf/2512.21204v1
+
+## 25771. Schrödinger's Navigator: Imagining an Ensemble of Futures for Zero-Shot Object Navigation
+
+- **ID:** `2512.21201` · **primary:** `cs.RO` · **Submitted:** 2025-12-24
+- **Authors:** Yu He, Da Huang, Zhenyang Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.21201 · **PDF:** https://arxiv.org/pdf/2512.21201v2
+
+## 25772. Safe Path Planning and Observation Quality Enhancement Strategy for Unmanned Aerial Vehicles in Water Quality Monitoring Tasks
+
+- **ID:** `2512.21375` · **primary:** `cs.RO` · **Submitted:** 2025-12-24
+- **Authors:** Yuanshuang Fu, Qianyao Wang, Qihao Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.21375 · **PDF:** https://arxiv.org/pdf/2512.21375v1
+
+## 25773. A Unified Framework for EEG Seizure Detection Using Universum-Integrated Generalized Eigenvalues Proximal Support Vector Machine
+
+- **ID:** `2512.21170` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Yogesh Kumar, Vrushank Ahire, M. A. Ganaie
+- **abs:** https://arxiv.org/abs/2512.21170 · **PDF:** https://arxiv.org/pdf/2512.21170v1
+
+## 25774. A Community-Enhanced Graph Representation Model for Link Prediction
+
+- **ID:** `2512.21166` · **primary:** `cs.SI` · **Submitted:** 2025-12-24
+- **Authors:** Lei Wang, Darong Lai
+- **abs:** https://arxiv.org/abs/2512.21166 · **PDF:** https://arxiv.org/pdf/2512.21166v1
+
+## 25775. BALLAST: Bandit-Assisted Learning for Latency-Aware Stable Timeouts in Raft
+
+- **ID:** `2512.21165` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Qizhi Wang
+- **abs:** https://arxiv.org/abs/2512.21165 · **PDF:** https://arxiv.org/pdf/2512.21165v1
+
+## 25776. ReVEAL: GNN-Guided Reverse Engineering for Formal Verification of Optimized Multipliers
+
+- **ID:** `2512.22260` · **primary:** `cs.LO` · **Submitted:** 2025-12-24
+- **Authors:** Chen Chen, Daniela Kaufmann, Chenhui Deng, et al.
+- **abs:** https://arxiv.org/abs/2512.22260 · **PDF:** https://arxiv.org/pdf/2512.22260v1
+
+## 25777. ElfCore: A 28nm Neural Processor Enabling Dynamic Structured Sparse Training and Online Self-Supervised Learning with Activity-Dependent Weight Update
+
+- **ID:** `2512.21153` · **primary:** `cs.AR` · **Submitted:** 2025-12-24
+- **Authors:** Zhe Su, Giacomo Indiveri
+- **abs:** https://arxiv.org/abs/2512.21153 · **PDF:** https://arxiv.org/pdf/2512.21153v1
+
+## 25778. LLM_annotate: A Python package for annotating and analyzing fiction characters
+
+- **ID:** `2601.03274` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Hannes Rosenbusch
+- **abs:** https://arxiv.org/abs/2601.03274 · **PDF:** https://arxiv.org/pdf/2601.03274v1
+
+## 25779. MODE: Multi-Objective Adaptive Coreset Selection
+
+- **ID:** `2512.21152` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Tanmoy Mukherjee, Pierre Marquis, Zied Bouraoui
+- **abs:** https://arxiv.org/abs/2512.21152 · **PDF:** https://arxiv.org/pdf/2512.21152v1
+
+## 25780. Evaluating Novelty in AI-Generated Research Plans Using Multi-Workflow LLM Pipelines
+
+- **ID:** `2601.09714` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Devesh Saraogi, Rohit Singhee, Dhruv Kumar
+- **abs:** https://arxiv.org/abs/2601.09714 · **PDF:** https://arxiv.org/pdf/2601.09714v1
+
+## 25781. AgentTutor: Empowering Personalized Learning with Multi-Turn Interactive Teaching in Intelligent Education Systems
+
+- **ID:** `2601.04219` · **primary:** `cs.CY` · **Submitted:** 2025-12-24
+- **Authors:** Yuxin Liu, Zeqing Song, Jiong Lou, et al.
+- **abs:** https://arxiv.org/abs/2601.04219 · **PDF:** https://arxiv.org/pdf/2601.04219v1
+
+## 25782. LLM-Driven Preference Data Synthesis for Proactive Prediction of the Next User Utterance in Human-Machine Dialogue
+
+- **ID:** `2601.09713` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Jinqiang Wang, Huansheng Ning, Jianguo Ding, et al.
+- **abs:** https://arxiv.org/abs/2601.09713 · **PDF:** https://arxiv.org/pdf/2601.09713v1
+
+## 25783. TGC-Net: A Structure-Aware and Semantically-Aligned Framework for Text-Guided Medical Image Segmentation
+
+- **ID:** `2512.21135` · **primary:** `cs.CV` · **Submitted:** 2025-12-24
+- **Authors:** Gaoren Lin, Huangxuan Zhao, Yuan Xiong, et al.
+- **abs:** https://arxiv.org/abs/2512.21135 · **PDF:** https://arxiv.org/pdf/2512.21135v1
+
+## 25784. AutoBaxBuilder: Bootstrapping Code Security Benchmarking
+
+- **ID:** `2512.21132` · **primary:** `cs.CR` · **Submitted:** 2025-12-24
+- **Authors:** Tobias von Arx, Niels Mündler, Mark Vero, et al.
+- **abs:** https://arxiv.org/abs/2512.21132 · **PDF:** https://arxiv.org/pdf/2512.21132v1
+
+## 25785. Active inference and artificial reasoning
+
+- **ID:** `2512.21129` · **primary:** `q-bio.NC` · **Submitted:** 2025-12-24
+- **Authors:** Karl Friston, Lancelot Da Costa, Alexander Tschantz, et al.
+- **abs:** https://arxiv.org/abs/2512.21129 · **PDF:** https://arxiv.org/pdf/2512.21129v1
+
+## 25786. A Real-World Evaluation of LLM Medication Safety Reviews in NHS Primary Care
+
+- **ID:** `2512.21127` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Oliver Normand, Esther Borsi, Mitch Fruin, et al.
+- **abs:** https://arxiv.org/abs/2512.21127 · **PDF:** https://arxiv.org/pdf/2512.21127v1
+
+## 25787. ClarifyMT-Bench: Benchmarking and Improving Multi-Turn Clarification for Conversational Large Language Models
+
+- **ID:** `2512.21120` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Sichun Luo, Yi Huang, Mukai Li, et al.
+- **abs:** https://arxiv.org/abs/2512.21120 · **PDF:** https://arxiv.org/pdf/2512.21120v1
+
+## 25788. STLDM: Spatio-Temporal Latent Diffusion Model for Precipitation Nowcasting
+
+- **ID:** `2512.21118` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Shi Quan Foo, Chi-Ho Wong, Zhihan Gao, et al.
+- **abs:** https://arxiv.org/abs/2512.21118 · **PDF:** https://arxiv.org/pdf/2512.21118v1
+
+## 25789. A Mechanistic Analysis of Transformers for Dynamical Systems
+
+- **ID:** `2512.21113` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Gregory Duthé, Nikolaos Evangelou, Wei Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.21113 · **PDF:** https://arxiv.org/pdf/2512.21113v1
+
+## 25790. Statistical and computational challenges in ranking
+
+- **ID:** `2512.21111` · **primary:** `math.ST` · **Submitted:** 2025-12-24
+- **Authors:** Alexandra Carpentier, Nicolas Verzelen
+- **abs:** https://arxiv.org/abs/2512.21111 · **PDF:** https://arxiv.org/pdf/2512.21111v1
+
+## 25791. Beyond Context: Large Language Models Failure to Grasp Users Intent
+
+- **ID:** `2512.21110` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Ahmed M. Hussain, Salahuddin Salahuddin, Panos Papadimitratos
+- **abs:** https://arxiv.org/abs/2512.21110 · **PDF:** https://arxiv.org/pdf/2512.21110v2
+
+## 25792. Semi-Supervised Learning for Large Language Models Safety and Content Moderation
+
+- **ID:** `2512.21107` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Eduard Stefan Dinuta, Iustin Sirbu, Traian Rebedea
+- **abs:** https://arxiv.org/abs/2512.21107 · **PDF:** https://arxiv.org/pdf/2512.21107v1
+
+## 25793. Semantic Refinement with LLMs for Graph Representations
+
+- **ID:** `2512.21106` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Safal Thapaliya, Zehong Wang, Jiazheng Li, et al.
+- **abs:** https://arxiv.org/abs/2512.21106 · **PDF:** https://arxiv.org/pdf/2512.21106v2
+
+## 25794. Shared Representation Learning for High-Dimensional Multi-Task Forecasting under Resource Contention in Cloud-Native Backends
+
+- **ID:** `2512.21102` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Zixiao Huang, Jixiao Yang, Sijia Li, et al.
+- **abs:** https://arxiv.org/abs/2512.21102 · **PDF:** https://arxiv.org/pdf/2512.21102v1
+
+## 25795. TexAvatars : Hybrid Texel-3D Representations for Stable Rigging of Photorealistic Gaussian Head Avatars
+
+- **ID:** `2512.21099` · **primary:** `cs.GR` · **Submitted:** 2025-12-24
+- **Authors:** Jaeseong Lee, Junyeong Ahn, Taewoong Kang, et al.
+- **abs:** https://arxiv.org/abs/2512.21099 · **PDF:** https://arxiv.org/pdf/2512.21099v1
+
+## 25796. Speak the Art: A Direct Speech to Image Generation Framework
+
+- **ID:** `2601.00827` · **primary:** `eess.AS` · **Submitted:** 2025-12-24
+- **Authors:** Mariam Saeed, Manar Amr, Farida Adel, et al.
+- **abs:** https://arxiv.org/abs/2601.00827 · **PDF:** https://arxiv.org/pdf/2601.00827v2
+
+## 25797. Cardiac mortality prediction in patients undergoing PCI based on real and synthetic data
+
+- **ID:** `2512.22259` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Daniil Burakov, Ivan Petrov, Dmitrii Khelimskii, et al.
+- **abs:** https://arxiv.org/abs/2512.22259 · **PDF:** https://arxiv.org/pdf/2512.22259v1
+
+## 25798. Hierarchical Modeling Approach to Fast and Accurate Table Recognition
+
+- **ID:** `2512.21083` · **primary:** `cs.CV` · **Submitted:** 2025-12-24
+- **Authors:** Takaya Kawakatsu
+- **abs:** https://arxiv.org/abs/2512.21083 · **PDF:** https://arxiv.org/pdf/2512.21083v1
+
+## 25799. Dyna-Style Reinforcement Learning Modeling and Control of Non-linear Dynamics
+
+- **ID:** `2512.21081` · **primary:** `eess.SY` · **Submitted:** 2025-12-24
+- **Authors:** Karim Abdelsalam, Zeyad Gamal, Ayman El-Badawy
+- **abs:** https://arxiv.org/abs/2512.21081 · **PDF:** https://arxiv.org/pdf/2512.21081v1
+
+## 25800. LLM Personas as a Substitute for Field Experiments in Method Benchmarking
+
+- **ID:** `2512.21080` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Enoch Hyunwook Kang
+- **abs:** https://arxiv.org/abs/2512.21080 · **PDF:** https://arxiv.org/pdf/2512.21080v3
+
+## 25801. Blurb-Refined Inference from Crowdsourced Book Reviews using Hierarchical Genre Mining with Dual-Path Graph Convolutions
+
+- **ID:** `2512.21076` · **primary:** `cs.IR` · **Submitted:** 2025-12-24
+- **Authors:** Suraj Kumar, Utsav Kumar Nareti, Soumi Chattopadhyay, et al.
+- **abs:** https://arxiv.org/abs/2512.21076 · **PDF:** https://arxiv.org/pdf/2512.21076v1
+
+## 25802. Understanding Scaling Laws in Deep Neural Networks via Feature Learning Dynamics
+
+- **ID:** `2512.21075` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Zihan Yao, Ruoyu Wu, Tianxiang Gao
+- **abs:** https://arxiv.org/abs/2512.21075 · **PDF:** https://arxiv.org/pdf/2512.21075v1
+
+## 25803. Logic Sketch Prompting (LSP): A Deterministic and Interpretable Prompting Method
+
+- **ID:** `2512.22258` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Satvik Tripathi
+- **abs:** https://arxiv.org/abs/2512.22258 · **PDF:** https://arxiv.org/pdf/2512.22258v1
+
+## 25804. Agentic Explainable Artificial Intelligence (Agentic XAI) Approach To Explore Better Explanation
+
+- **ID:** `2512.21066` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Tomoaki Yamaguchi, Yutong Zhou, Masahiro Ryo, et al.
+- **abs:** https://arxiv.org/abs/2512.21066 · **PDF:** https://arxiv.org/pdf/2512.21066v3
+
+## 25805. DexAvatar: 3D Sign Language Reconstruction with Hand and Body Pose Priors
+
+- **ID:** `2512.21054` · **primary:** `cs.CV` · **Submitted:** 2025-12-24
+- **Authors:** Kaustubh Kundu, Hrishav Bakul Barua, Lucy Robertson-Bell, et al.
+- **abs:** https://arxiv.org/abs/2512.21054 · **PDF:** https://arxiv.org/pdf/2512.21054v1
+
+## 25806. zkFL-Health: Blockchain-Enabled Zero-Knowledge Federated Learning for Medical AI Privacy
+
+- **ID:** `2512.21048` · **primary:** `cs.CR` · **Submitted:** 2025-12-24
+- **Authors:** Savvy Sharma, George Petrovic, Sarthak Kaushik
+- **abs:** https://arxiv.org/abs/2512.21048 · **PDF:** https://arxiv.org/pdf/2512.21048v1
+
+## 25807. AInsteinBench: Benchmarking Coding Agents on Scientific Repositories
+
+- **ID:** `2512.21373` · **primary:** `cs.SE` · **Submitted:** 2025-12-24
+- **Authors:** Titouan Duston, Shuo Xin, Yang Sun, et al.
+- **abs:** https://arxiv.org/abs/2512.21373 · **PDF:** https://arxiv.org/pdf/2512.21373v1
+
+## 25808. Agentic Multi-Persona Framework for Evidence-Aware Fake News Detection
+
+- **ID:** `2512.21039` · **primary:** `cs.IR` · **Submitted:** 2025-12-24
+- **Authors:** Roopa Bukke, Soumya Pandey, Suraj Kumar, et al.
+- **abs:** https://arxiv.org/abs/2512.21039 · **PDF:** https://arxiv.org/pdf/2512.21039v2
+
+## 25809. Agentic Software Issue Resolution with Large Language Models: A Survey
+
+- **ID:** `2512.22256` · **primary:** `cs.SE` · **Submitted:** 2025-12-24
+- **Authors:** Zhonghao Jiang, David Lo, Zhongxin Liu
+- **abs:** https://arxiv.org/abs/2512.22256 · **PDF:** https://arxiv.org/pdf/2512.22256v1
+
+## 25810. Critical Points of Degenerate Metrics on Algebraic Varieties: A Tale of Overparametrization
+
+- **ID:** `2512.21029` · **primary:** `math.AG` · **Submitted:** 2025-12-24
+- **Authors:** Giovanni Luca Marchetti, Erin Connelly, Paul Breiding, et al.
+- **abs:** https://arxiv.org/abs/2512.21029 · **PDF:** https://arxiv.org/pdf/2512.21029v1
+
+## 25811. Policy-Conditioned Policies for Multi-Agent Task Solving
+
+- **ID:** `2512.21024` · **primary:** `cs.GT` · **Submitted:** 2025-12-24
+- **Authors:** Yue Lin, Shuhui Zhu, Wenhao Li, et al.
+- **abs:** https://arxiv.org/abs/2512.21024 · **PDF:** https://arxiv.org/pdf/2512.21024v1
+
+## 25812. Shape of Thought: When Distribution Matters More than Correctness in Reasoning Tasks
+
+- **ID:** `2512.22255` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Abhranil Chandra, Ayush Agrawal, Arian Hosseini, et al.
+- **abs:** https://arxiv.org/abs/2512.22255 · **PDF:** https://arxiv.org/pdf/2512.22255v2
+
+## 25813. Towards Better Search with Domain-Aware Text Embeddings for C2C Marketplaces
+
+- **ID:** `2512.21021` · **primary:** `cs.IR` · **Submitted:** 2025-12-24
+- **Authors:** Andre Rusli, Miao Cao, Shoma Ishimoto, et al.
+- **abs:** https://arxiv.org/abs/2512.21021 · **PDF:** https://arxiv.org/pdf/2512.21021v1
+
+## 25814. Enhancing diffusion models with Gaussianization preprocessing
+
+- **ID:** `2512.21020` · **primary:** `stat.ML` · **Submitted:** 2025-12-24
+- **Authors:** Li Cunzhi, Louis Kang, Hideaki Shimazaki
+- **abs:** https://arxiv.org/abs/2512.21020 · **PDF:** https://arxiv.org/pdf/2512.21020v1
+
+## 25815. Rethinking Supervised Fine-Tuning: Emphasizing Key Answer Tokens for Improved LLM Accuracy
+
+- **ID:** `2512.21017` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Xiaofeng Shi, Qian Kou, Yuduo Li, et al.
+- **abs:** https://arxiv.org/abs/2512.21017 · **PDF:** https://arxiv.org/pdf/2512.21017v1
+
+## 25816. LLM Swiss Round: Aggregating Multi-Benchmark Performance via Competitive Swiss-System Dynamics
+
+- **ID:** `2512.21010` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Jiashuo Liu, Jiayun Wu, Chunjie Wu, et al.
+- **abs:** https://arxiv.org/abs/2512.21010 · **PDF:** https://arxiv.org/pdf/2512.21010v1
+
+## 25817. Learning from Neighbors with PHIBP: Predicting Infectious Disease Dynamics in Data-Sparse Environments
+
+- **ID:** `2512.21005` · **primary:** `stat.ML` · **Submitted:** 2025-12-24
+- **Authors:** Edwin Fong, Lancelot F. James, Juho Lee
+- **abs:** https://arxiv.org/abs/2512.21005 · **PDF:** https://arxiv.org/pdf/2512.21005v2
+
+## 25818. Distilling the Essence: Efficient Reasoning Distillation via Sequence Truncation
+
+- **ID:** `2512.21002` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Wei-Rui Chen, Vignesh Kothapalli, Ata Fatahibaarzi, et al.
+- **abs:** https://arxiv.org/abs/2512.21002 · **PDF:** https://arxiv.org/pdf/2512.21002v2
+
+## 25819. CoSeNet: A Novel Approach for Optimal Segmentation of Correlation Matrices
+
+- **ID:** `2512.21000` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Alberto. Palomo-Alonso, David Casillas-Perez, Silvia Jimenez-Fernandez, et al.
+- **abs:** https://arxiv.org/abs/2512.21000 · **PDF:** https://arxiv.org/pdf/2512.21000v1
+
+## 25820. Analyzing Skill Element in Online Fantasy Cricket
+
+- **ID:** `2512.22254` · **primary:** `cs.GT` · **Submitted:** 2025-12-24
+- **Authors:** Sarthak Sarkar, Supratim Das, Purushottam Saha, et al.
+- **abs:** https://arxiv.org/abs/2512.22254 · **PDF:** https://arxiv.org/pdf/2512.22254v1
+
+## 25821. TrafficSimAgent: A Hierarchical Agent Framework for Autonomous Traffic Simulation with MCP Control
+
+- **ID:** `2512.20996` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Yuwei Du, Jun Zhang, Jie Feng, et al.
+- **abs:** https://arxiv.org/abs/2512.20996 · **PDF:** https://arxiv.org/pdf/2512.20996v1
+
+## 25822. A multi-algorithm approach for operational human resources workload balancing in a last mile urban delivery system
+
+- **ID:** `2601.00023` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Luis M. Moreno-Saavedra, Silvia Jimenez-Fernandez, Antonio Portilla-Figueras, et al.
+- **abs:** https://arxiv.org/abs/2601.00023 · **PDF:** https://arxiv.org/pdf/2601.00023v1
+
+## 25823. Evolutionary optimization of spatially-distributed multi-sensors placement for indoor surveillance environments with security levels
+
+- **ID:** `2601.00826` · **primary:** `cs.NE` · **Submitted:** 2025-12-24
+- **Authors:** Luis M. Moreno-Saavedra, Vinıcius G. Costa, Adrian Garrido-Saez, et al.
+- **abs:** https://arxiv.org/abs/2601.00826 · **PDF:** https://arxiv.org/pdf/2601.00826v1
+
+## 25824. FinAgent: An Agentic AI Framework Integrating Personal Finance and Nutrition Planning
+
+- **ID:** `2512.20991` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Toqeer Ali Syed, Abdulaziz Alshahrani, Ali Ullah, et al.
+- **abs:** https://arxiv.org/abs/2512.20991 · **PDF:** https://arxiv.org/pdf/2512.20991v1
+
+## 25825. A Blockchain-Monitored Agentic AI Architecture for Trusted Perception-Reasoning-Action Pipelines
+
+- **ID:** `2512.20985` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Salman Jan, Hassan Ali Razzaqi, Ali Akarma, et al.
+- **abs:** https://arxiv.org/abs/2512.20985 · **PDF:** https://arxiv.org/pdf/2512.20985v1
+
+## 25826. Automatic Replication of LLM Mistakes in Medical Conversations
+
+- **ID:** `2512.20983` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Oleksii Proniakin, Diego Fajardo, Ruslan Nazarenko, et al.
+- **abs:** https://arxiv.org/abs/2512.20983 · **PDF:** https://arxiv.org/pdf/2512.20983v1
+
+## 25827. GenTSE: Enhancing Target Speaker Extraction via a Coarse-to-Fine Generative Language Model
+
+- **ID:** `2512.20978` · **primary:** `eess.AS` · **Submitted:** 2025-12-24
+- **Authors:** Haoyang Li, Xuyi Zhuang, Azmat Adnan, et al.
+- **abs:** https://arxiv.org/abs/2512.20978 · **PDF:** https://arxiv.org/pdf/2512.20978v1
+
+## 25828. Generalised Linear Models in Deep Bayesian RL with Learnable Basis Functions
+
+- **ID:** `2512.20974` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Jingyang You, Hanna Kurniawati
+- **abs:** https://arxiv.org/abs/2512.20974 · **PDF:** https://arxiv.org/pdf/2512.20974v1
+
+## 25829. Mesh-Attention: A New Communication-Efficient Distributed Attention with Improved Data Locality
+
+- **ID:** `2512.20968` · **primary:** `cs.DC` · **Submitted:** 2025-12-24
+- **Authors:** Sirui Chen, Jingji Chen, Siqi Zhu, et al.
+- **abs:** https://arxiv.org/abs/2512.20968 · **PDF:** https://arxiv.org/pdf/2512.20968v1
+
+## 25830. Deadline-Aware Online Scheduling for LLM Fine-Tuning with Spot Market Predictions
+
+- **ID:** `2512.20967` · **primary:** `cs.DC` · **Submitted:** 2025-12-24
+- **Authors:** Linggao Kong, Yuedong Xu, Lei Jiao, et al.
+- **abs:** https://arxiv.org/abs/2512.20967 · **PDF:** https://arxiv.org/pdf/2512.20967v1
+
+## 25831. Generalization of Diffusion Models Arises with a Balanced Representation Space
+
+- **ID:** `2512.20963` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Zekai Zhang, Xiao Li, Xiang Li, et al.
+- **abs:** https://arxiv.org/abs/2512.20963 · **PDF:** https://arxiv.org/pdf/2512.20963v2
+
+## 25832. Can Agentic AI Match the Performance of Human Data Scientists?
+
+- **ID:** `2512.20959` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** An Luo, Jin Du, Fangqiao Tian, et al.
+- **abs:** https://arxiv.org/abs/2512.20959 · **PDF:** https://arxiv.org/pdf/2512.20959v1
+
+## 25833. ReACT-Drug: Reaction-Template Guided Reinforcement Learning for de novo Drug Design
+
+- **ID:** `2512.20958` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** R Yadunandan, Nimisha Ghosh
+- **abs:** https://arxiv.org/abs/2512.20958 · **PDF:** https://arxiv.org/pdf/2512.20958v1
+
+## 25834. One Tool Is Enough: Reinforcement Learning for Repository-Level LLM Agents
+
+- **ID:** `2512.20957` · **primary:** `cs.SE` · **Submitted:** 2025-12-24
+- **Authors:** Zhaoxi Zhang, Yitong Duan, Yanzhi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.20957 · **PDF:** https://arxiv.org/pdf/2512.20957v5
+
+## 25835. Solving Functional PDEs with Gaussian Processes and Applications to Functional Renormalization Group Equations
+
+- **ID:** `2512.20956` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Xianjin Yang, Matthieu Darcy, Matthew Hudes, et al.
+- **abs:** https://arxiv.org/abs/2512.20956 · **PDF:** https://arxiv.org/pdf/2512.20956v1
+
+## 25836. Reflection Pretraining Enables Token-Level Self-Correction in Biological Sequence Models
+
+- **ID:** `2512.20954` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Xiang Zhang, Jiaqi Wei, Yuejin Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.20954 · **PDF:** https://arxiv.org/pdf/2512.20954v1
+
+## 25837. MultiMind at SemEval-2025 Task 7: Crosslingual Fact-Checked Claim Retrieval via Multi-Source Alignment
+
+- **ID:** `2512.20950` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Mohammad Mahdi Abootorabi, Alireza Ghahramani Kure, Mohammadali Mohammadkhani, et al.
+- **abs:** https://arxiv.org/abs/2512.20950 · **PDF:** https://arxiv.org/pdf/2512.20950v1
+
+## 25838. Graph Attention-based Adaptive Transfer Learning for Link Prediction
+
+- **ID:** `2512.22252` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Huashen Lu, Wensheng Gan, Guoting Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.22252 · **PDF:** https://arxiv.org/pdf/2512.22252v1
+
+## 25839. Neural Probe-Based Hallucination Detection for Large Language Models
+
+- **ID:** `2512.20949` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Shize Liang, Hongzhi Wang
+- **abs:** https://arxiv.org/abs/2512.20949 · **PDF:** https://arxiv.org/pdf/2512.20949v1
+
+## 25840. Foundation Model-based Evaluation of Neuropsychiatric Disorders: A Lifespan-Inclusive, Multi-Modal, and Multi-Lingual Study
+
+- **ID:** `2512.20948` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Zhongren Dong, Haotian Guo, Weixiang Xu, et al.
+- **abs:** https://arxiv.org/abs/2512.20948 · **PDF:** https://arxiv.org/pdf/2512.20948v1
+
+## 25841. AirGS: Real-Time 4D Gaussian Streaming for Free-Viewpoint Video Experiences
+
+- **ID:** `2512.20943` · **primary:** `cs.GR` · **Submitted:** 2025-12-24
+- **Authors:** Zhe Wang, Jinghang Li, Yifei Zhu
+- **abs:** https://arxiv.org/abs/2512.20943 · **PDF:** https://arxiv.org/pdf/2512.20943v1
+
+## 25842. A Multi-fidelity Double-Delta Wing Dataset and Empirical Scaling Laws for GNN-based Aerodynamic Field Surrogate
+
+- **ID:** `2512.20941` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Yiren Shen, Juan J. Alonso
+- **abs:** https://arxiv.org/abs/2512.20941 · **PDF:** https://arxiv.org/pdf/2512.20941v1
+
+## 25843. Interpretable Perturbation Modeling Through Biomedical Knowledge Graphs
+
+- **ID:** `2512.22251` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Pascal Passigan, Kevin Zhu, Angelina Ning
+- **abs:** https://arxiv.org/abs/2512.22251 · **PDF:** https://arxiv.org/pdf/2512.22251v2
+
+## 25844. Transductive Visual Programming: Evolving Tool Libraries from Experience for Spatial Reasoning
+
+- **ID:** `2512.20934` · **primary:** `cs.CV` · **Submitted:** 2025-12-24
+- **Authors:** Shengguang Wu, Xiaohan Wang, Yuhui Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.20934 · **PDF:** https://arxiv.org/pdf/2512.20934v1
+
+## 25845. Guardrailed Elasticity Pricing: A Churn-Aware Forecasting Playbook for Subscription Strategy
+
+- **ID:** `2512.20932` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Deepit Sapru
+- **abs:** https://arxiv.org/abs/2512.20932 · **PDF:** https://arxiv.org/pdf/2512.20932v2
+
+## 25846. Decoding Predictive Inference in Visual Language Processing via Spatiotemporal Neural Coherence
+
+- **ID:** `2512.20929` · **primary:** `q-bio.NC` · **Submitted:** 2025-12-24
+- **Authors:** Sean C. Borneman, Julia Krebs, Ronnie B. Wilbur, et al.
+- **abs:** https://arxiv.org/abs/2512.20929 · **PDF:** https://arxiv.org/pdf/2512.20929v1
+
+## 25847. Tree of Preferences for Diversified Recommendation
+
+- **ID:** `2601.02386` · **primary:** `cs.IR` · **Submitted:** 2025-12-24
+- **Authors:** Hanyang Yuan, Ning Tang, Tongya Zheng, et al.
+- **abs:** https://arxiv.org/abs/2601.02386 · **PDF:** https://arxiv.org/pdf/2601.02386v1
+
+## 25848. Clever Hans in Chemistry: Chemist Style Signals Confound Activity Prediction on Public Benchmarks
+
+- **ID:** `2512.20924` · **primary:** `q-bio.BM` · **Submitted:** 2025-12-24
+- **Authors:** Andrew D. Blevins, Ian K. Quigley
+- **abs:** https://arxiv.org/abs/2512.20924 · **PDF:** https://arxiv.org/pdf/2512.20924v1
+
+## 25849. RevFFN: Memory-Efficient Full-Parameter Fine-Tuning of Mixture-of-Experts LLMs with Reversible Blocks
+
+- **ID:** `2512.20920` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Ningyuan Liu, Jing Yang, Kaitong Cai, et al.
+- **abs:** https://arxiv.org/abs/2512.20920 · **PDF:** https://arxiv.org/pdf/2512.20920v1
+
+## 25850. Towards a General Framework for Predicting and Explaining the Hardness of Graph-based Combinatorial Optimization Problems using Machine Learning and Association Rule Mining
+
+- **ID:** `2512.20915` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Bharat Sharman, Elkafi Hassini
+- **abs:** https://arxiv.org/abs/2512.20915 · **PDF:** https://arxiv.org/pdf/2512.20915v1
+
+## 25851. Invariant Feature Extraction Through Conditional Independence and the Optimal Transport Barycenter Problem: the Gaussian case
+
+- **ID:** `2512.20914` · **primary:** `math.ST` · **Submitted:** 2025-12-24
+- **Authors:** Ian Bounos, Pablo Groisman, Mariela Sued, et al.
+- **abs:** https://arxiv.org/abs/2512.20914 · **PDF:** https://arxiv.org/pdf/2512.20914v1
+
+## 25852. Where Did This Sentence Come From? Tracing Provenance in LLM Reasoning Distillation
+
+- **ID:** `2512.20908` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Kaiyuan Liu, Shaotian Yan, Rui Miao, et al.
+- **abs:** https://arxiv.org/abs/2512.20908 · **PDF:** https://arxiv.org/pdf/2512.20908v2
+
+## 25853. Temporal Visual Semantics-Induced Human Motion Understanding with Large Language Models
+
+- **ID:** `2512.22249` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Zheng Xing, Weibing Zhao
+- **abs:** https://arxiv.org/abs/2512.22249 · **PDF:** https://arxiv.org/pdf/2512.22249v1
+
+## 25854. DiEC: Diffusion Embedded Clustering
+
+- **ID:** `2512.20905` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Haidong Hu, Xiaoyu Zheng, Jin Zhou, et al.
+- **abs:** https://arxiv.org/abs/2512.20905 · **PDF:** https://arxiv.org/pdf/2512.20905v4
+
+## 25855. Embodied AI-Enhanced IoMT Edge Computing: UAV Trajectory Optimization and Task Offloading with Mobility Prediction
+
+- **ID:** `2512.20902` · **primary:** `cs.NI` · **Submitted:** 2025-12-24
+- **Authors:** Siqi Mu, Shuo Wen, Yang Lu, et al.
+- **abs:** https://arxiv.org/abs/2512.20902 · **PDF:** https://arxiv.org/pdf/2512.20902v1
+
+## 25856. DGSAN: Dual-Graph Spatiotemporal Attention Network for Pulmonary Nodule Malignancy Prediction
+
+- **ID:** `2512.20898` · **primary:** `cs.CV` · **Submitted:** 2025-12-24
+- **Authors:** Xiao Yu, Zhaojie Fang, Guanyu Zhou, et al.
+- **abs:** https://arxiv.org/abs/2512.20898 · **PDF:** https://arxiv.org/pdf/2512.20898v1
+
+## 25857. Time-Efficient Evaluation and Enhancement of Adversarial Robustness in Deep Neural Networks
+
+- **ID:** `2512.20893` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Runqi Lin
+- **abs:** https://arxiv.org/abs/2512.20893 · **PDF:** https://arxiv.org/pdf/2512.20893v1
+
+## 25858. From GNNs to Symbolic Surrogates via Kolmogorov-Arnold Networks for Delay Prediction
+
+- **ID:** `2512.20885` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Sami Marouani, Kamal Singh, Baptiste Jeudy, et al.
+- **abs:** https://arxiv.org/abs/2512.20885 · **PDF:** https://arxiv.org/pdf/2512.20885v2
+
+## 25859. The Silent Scholar Problem: A Probabilistic Framework for Breaking Epistemic Asymmetry in LLM Agents
+
+- **ID:** `2512.20884` · **primary:** `cs.AI` · **Submitted:** 2025-12-24
+- **Authors:** Zan-Kai Chong, Hiroyuki Ohsaki, Bryan Ng
+- **abs:** https://arxiv.org/abs/2512.20884 · **PDF:** https://arxiv.org/pdf/2512.20884v1
+
+## 25860. Architectural Trade-offs in Small Language Models Under Compute Constraints
+
+- **ID:** `2512.20877` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Shivraj Singh Bhatti
+- **abs:** https://arxiv.org/abs/2512.20877 · **PDF:** https://arxiv.org/pdf/2512.20877v1
+
+## 25861. Amortized Inference for Model Rocket Aerodynamics: Learning to Estimate Physical Parameters from Simulation
+
+- **ID:** `2512.22248` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Rohit Pandey, Rohan Pandey
+- **abs:** https://arxiv.org/abs/2512.22248 · **PDF:** https://arxiv.org/pdf/2512.22248v1
+
+## 25862. Better Call Graphs: A New Dataset of Function Call Graphs for Malware Classification
+
+- **ID:** `2512.20872` · **primary:** `cs.CR` · **Submitted:** 2025-12-24
+- **Authors:** Jakir Hossain, Gurvinder Singh, Lukasz Ziarek, et al.
+- **abs:** https://arxiv.org/abs/2512.20872 · **PDF:** https://arxiv.org/pdf/2512.20872v1
+
+## 25863. The Environmental Impact of AI Servers and Sustainable Solutions
+
+- **ID:** `2601.06063` · **primary:** `cs.CY` · **Submitted:** 2025-12-24
+- **Authors:** Aadi Patel, Nikhil Mahalingam, Rusheen Patel
+- **abs:** https://arxiv.org/abs/2601.06063 · **PDF:** https://arxiv.org/pdf/2601.06063v1
+
+## 25864. From Values to Frameworks: A Qualitative Study of Ethical Reasoning in Agentic AI Practitioners
+
+- **ID:** `2601.06062` · **primary:** `cs.CY` · **Submitted:** 2025-12-24
+- **Authors:** Theodore Roberts, Bahram Zarrin
+- **abs:** https://arxiv.org/abs/2601.06062 · **PDF:** https://arxiv.org/pdf/2601.06062v1
+
+## 25865. Lightweight framework for underground pipeline recognition and spatial localization based on multi-view 2D GPR images
+
+- **ID:** `2512.20866` · **primary:** `cs.CV` · **Submitted:** 2025-12-24
+- **Authors:** Haotian Lv, Chao Li, Jiangbo Dai, et al.
+- **abs:** https://arxiv.org/abs/2512.20866 · **PDF:** https://arxiv.org/pdf/2512.20866v1
+
+## 25866. Robustness Certificates for Neural Networks against Adversarial Attacks
+
+- **ID:** `2512.20865` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Sara Taheri, Mahalakshmi Sabanayagam, Debarghya Ghoshdastidar, et al.
+- **abs:** https://arxiv.org/abs/2512.20865 · **PDF:** https://arxiv.org/pdf/2512.20865v1
+
+## 25867. Memory-Efficient Acceleration of Block Low-Rank Foundation Models on Resource Constrained GPUs
+
+- **ID:** `2512.20861` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** Pierre Abillama, Changwoo Lee, Juechu Dong, et al.
+- **abs:** https://arxiv.org/abs/2512.20861 · **PDF:** https://arxiv.org/pdf/2512.20861v2
+
+## 25868. The Affine Divergence: Aligning Activation Updates Beyond Normalisation
+
+- **ID:** `2512.22247` · **primary:** `cs.LG` · **Submitted:** 2025-12-24
+- **Authors:** George Bird
+- **abs:** https://arxiv.org/abs/2512.22247 · **PDF:** https://arxiv.org/pdf/2512.22247v2
+
+## 25869. NVIDIA Nemotron 3: Efficient and Open Intelligence
+
+- **ID:** `2512.20856` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** NVIDIA, :, Aaron Blakeman, et al.
+- **abs:** https://arxiv.org/abs/2512.20856 · **PDF:** https://arxiv.org/pdf/2512.20856v1
+
+## 25870. How important is Recall for Measuring Retrieval Quality?
+
+- **ID:** `2512.20854` · **primary:** `cs.CL` · **Submitted:** 2025-12-24
+- **Authors:** Shelly Schwartz, Oleg Vasilyev, Randy Sawaya
+- **abs:** https://arxiv.org/abs/2512.20854 · **PDF:** https://arxiv.org/pdf/2512.20854v1
+
+## 25871. Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
+
+- **ID:** `2512.20848` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** NVIDIA, :, Aaron Blakeman, et al.
+- **abs:** https://arxiv.org/abs/2512.20848 · **PDF:** https://arxiv.org/pdf/2512.20848v1
+
+## 25872. MAR:Multi-Agent Reflexion Improves Reasoning Abilities in LLMs
+
+- **ID:** `2512.20845` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Onat Ozer, Grace Wu, Yuchen Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.20845 · **PDF:** https://arxiv.org/pdf/2512.20845v1
+
+## 25873. CHAMMI-75: Pre-training multi-channel models with heterogeneous microscopy images
+
+- **ID:** `2512.20833` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Vidit Agrawal, John Peters, Tyler N. Thompson, et al.
+- **abs:** https://arxiv.org/abs/2512.20833 · **PDF:** https://arxiv.org/pdf/2512.20833v2
+
+## 25874. Context-Sensitive Abstractions for Reinforcement Learning with Parameterized Actions
+
+- **ID:** `2512.20831` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Rashmeet Kaur Nayyar, Naman Shah, Siddharth Srivastava
+- **abs:** https://arxiv.org/abs/2512.20831 · **PDF:** https://arxiv.org/pdf/2512.20831v1
+
+## 25875. NotSoTiny: A Large, Living Benchmark for RTL Code Generation
+
+- **ID:** `2512.20823` · **primary:** `cs.AR` · **Submitted:** 2025-12-23
+- **Authors:** Razine Moundir Ghorab, Emanuele Parisi, Cristian Gutierrez, et al.
+- **abs:** https://arxiv.org/abs/2512.20823 · **PDF:** https://arxiv.org/pdf/2512.20823v1
+
+## 25876. MediEval: A Unified Medical Benchmark for Patient-Contextual and Knowledge-Grounded Reasoning in LLMs
+
+- **ID:** `2512.20822` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Zhan Qu, Michael Färber
+- **abs:** https://arxiv.org/abs/2512.20822 · **PDF:** https://arxiv.org/pdf/2512.20822v1
+
+## 25877. Divided We Fall: Defending Against Adversarial Attacks via Soft-Gated Fractional Mixture-of-Experts with Randomized Adversarial Training
+
+- **ID:** `2512.20821` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Mohammad Meymani, Roozbeh Razavi-Far
+- **abs:** https://arxiv.org/abs/2512.20821 · **PDF:** https://arxiv.org/pdf/2512.20821v3
+
+## 25878. From Textbook to Talkbot: A Case Study of a Greek-Language RAG-Based Chatbot in Higher Education
+
+- **ID:** `2601.14265` · **primary:** `cs.CY` · **Submitted:** 2025-12-23
+- **Authors:** Maria Eleni Koutsiaki, Marina Delianidi, Chaido Mizeli, et al.
+- **abs:** https://arxiv.org/abs/2601.14265 · **PDF:** https://arxiv.org/pdf/2601.14265v1
+
+## 25879. EssayCBM: Rubric-Aligned Concept Bottleneck Models for Transparent Essay Grading
+
+- **ID:** `2512.20817` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Kumar Satvik Chaudhary, Chengshuai Zhao, Fan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.20817 · **PDF:** https://arxiv.org/pdf/2512.20817v1
+
+## 25880. FedMPDD: Communication-Efficient Federated Learning with Privacy Preservation Attributes via Projected Directional Derivative
+
+- **ID:** `2512.20814` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Mohammadreza Rostami, Solmaz S. Kia
+- **abs:** https://arxiv.org/abs/2512.20814 · **PDF:** https://arxiv.org/pdf/2512.20814v1
+
+## 25881. GraphFire-X: Physics-Informed Graph Attention Networks and Structural Gradient Boosting for Building-Scale Wildfire Preparedness at the Wildland-Urban Interface
+
+- **ID:** `2512.20813` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Miguel Esparza, Vamshi Battal, Ali Mostafavi
+- **abs:** https://arxiv.org/abs/2512.20813 · **PDF:** https://arxiv.org/pdf/2512.20813v1
+
+## 25882. Semantic Deception: When Reasoning Models Can't Compute an Addition
+
+- **ID:** `2512.20812` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Nathaniël de Leeuw, Marceau Nahon, Mathis Reymond, et al.
+- **abs:** https://arxiv.org/abs/2512.20812 · **PDF:** https://arxiv.org/pdf/2512.20812v1
+
+## 25883. Weighted MCC: A Robust Measure of Multiclass Classifier Performance for Observations with Individual Weights
+
+- **ID:** `2512.20811` · **primary:** `stat.ML` · **Submitted:** 2025-12-23
+- **Authors:** Rommel Cortez, Bala Krishnamoorthy
+- **abs:** https://arxiv.org/abs/2512.20811 · **PDF:** https://arxiv.org/pdf/2512.20811v1
+
+## 25884. Safety Alignment of LMs via Non-cooperative Games
+
+- **ID:** `2512.20806` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Anselm Paulus, Ilia Kulikov, Brandon Amos, et al.
+- **abs:** https://arxiv.org/abs/2512.20806 · **PDF:** https://arxiv.org/pdf/2512.20806v2
+
+## 25885. Calibrating LLM Judges: Linear Probes for Fast and Reliable Uncertainty Estimation
+
+- **ID:** `2512.22245` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Bhaktipriya Radharapu, Eshika Saxena, Kenneth Li, et al.
+- **abs:** https://arxiv.org/abs/2512.22245 · **PDF:** https://arxiv.org/pdf/2512.22245v1
+
+## 25886. A Benchmark for Evaluating Outcome-Driven Constraint Violations in Autonomous AI Agents
+
+- **ID:** `2512.20798` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Miles Q. Li, Benjamin C. M. Fung, Martin Weiss, et al.
+- **abs:** https://arxiv.org/abs/2512.20798 · **PDF:** https://arxiv.org/pdf/2512.20798v3
+
+## 25887. Measuring Mechanistic Independence: Can Bias Be Removed Without Erasing Demographics?
+
+- **ID:** `2512.20796` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Zhengyang Shan, Aaron Mueller
+- **abs:** https://arxiv.org/abs/2512.20796 · **PDF:** https://arxiv.org/pdf/2512.20796v1
+
+## 25888. Investigating Model Editing for Unlearning in Large Language Models
+
+- **ID:** `2512.20794` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Shariqah Hossain, Lalana Kagal
+- **abs:** https://arxiv.org/abs/2512.20794 · **PDF:** https://arxiv.org/pdf/2512.20794v1
+
+## 25889. X-GridAgent: An LLM-Powered Agentic AI System for Assisting Power Grid Analysis
+
+- **ID:** `2512.20789` · **primary:** `eess.SY` · **Submitted:** 2025-12-23
+- **Authors:** Yihan, Wen, Xin Chen
+- **abs:** https://arxiv.org/abs/2512.20789 · **PDF:** https://arxiv.org/pdf/2512.20789v1
+
+## 25890. Symbolic regression for defect interactions in 2D materials
+
+- **ID:** `2512.20785` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Mikhail Lazarev, Andrey Ustyuzhanin
+- **abs:** https://arxiv.org/abs/2512.20785 · **PDF:** https://arxiv.org/pdf/2512.20785v2
+
+## 25891. NULLBUS: Multimodal Mixed-Supervision for Breast Ultrasound Segmentation via Nullable Global-Local Prompts
+
+- **ID:** `2512.20783` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Raja Mallina, Bryar Shareef
+- **abs:** https://arxiv.org/abs/2512.20783 · **PDF:** https://arxiv.org/pdf/2512.20783v1
+
+## 25892. Large Language Models Approach Expert Pedagogical Quality in Math Tutoring but Differ in Instructional and Linguistic Profiles
+
+- **ID:** `2512.20780` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Ramatu Oiza Abdulsalam, Segun Aroyehun
+- **abs:** https://arxiv.org/abs/2512.20780 · **PDF:** https://arxiv.org/pdf/2512.20780v2
+
+## 25893. Towards Optimal Performance and Action Consistency Guarantees in Dec-POMDPs with Inconsistent Beliefs and Limited Communication
+
+- **ID:** `2512.20778` · **primary:** `cs.MA` · **Submitted:** 2025-12-23
+- **Authors:** Moshe Rafaeli Shimron, Vadim Indelman
+- **abs:** https://arxiv.org/abs/2512.20778 · **PDF:** https://arxiv.org/pdf/2512.20778v1
+
+## 25894. Improving Matrix Exponential for Generative AI Flows: A Taylor-Based Approach Beyond Paterson--Stockmeyer
+
+- **ID:** `2512.20777` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Jorge Sastre, Daniel Faronbi, José Miguel Alonso, et al.
+- **abs:** https://arxiv.org/abs/2512.20777 · **PDF:** https://arxiv.org/pdf/2512.20777v2
+
+## 25895. Más contexto no es mejor. Paradoja de la dilución vectorial en RAG corporativos
+
+- **ID:** `2601.08851` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Alex Dantart
+- **abs:** https://arxiv.org/abs/2601.08851 · **PDF:** https://arxiv.org/pdf/2601.08851v1
+
+## 25896. DIAL: Direct Iterative Adversarial Learning for Realistic Multi-Turn Dialogue Simulation
+
+- **ID:** `2512.20773` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Ziyi Zhu, Olivier Tieleman, Caitlin A. Stamatis, et al.
+- **abs:** https://arxiv.org/abs/2512.20773 · **PDF:** https://arxiv.org/pdf/2512.20773v3
+
+## 25897. Subgroup Discovery with the Cox Model
+
+- **ID:** `2512.20762` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Zachary Izzo, Iain Melvin
+- **abs:** https://arxiv.org/abs/2512.20762 · **PDF:** https://arxiv.org/pdf/2512.20762v2
+
+## 25898. TS-Arena -- A Live Forecast Pre-Registration Platform
+
+- **ID:** `2512.20761` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Marcel Meyer, Sascha Kaltenpoth, Henrik Albers, et al.
+- **abs:** https://arxiv.org/abs/2512.20761 · **PDF:** https://arxiv.org/pdf/2512.20761v2
+
+## 25899. Generalization of RLVR Using Causal Reasoning as a Testbed
+
+- **ID:** `2512.20760` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Brian Lu, Hongyu Zhao, Shuo Sun, et al.
+- **abs:** https://arxiv.org/abs/2512.20760 · **PDF:** https://arxiv.org/pdf/2512.20760v2
+
+## 25900. TokSuite: Measuring the Impact of Tokenizer Choice on Language Model Behavior
+
+- **ID:** `2512.20757` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Gül Sena Altıntaş, Malikeh Ehghaghi, Brian Lester, et al.
+- **abs:** https://arxiv.org/abs/2512.20757 · **PDF:** https://arxiv.org/pdf/2512.20757v1
+
+## 25901. Bridging Efficiency and Safety: Formal Verification of Neural Networks with Early Exits
+
+- **ID:** `2512.20755` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Yizhak Yisrael Elboher, Avraham Raviv, Amihay Elboher, et al.
+- **abs:** https://arxiv.org/abs/2512.20755 · **PDF:** https://arxiv.org/pdf/2512.20755v1
+
+## 25902. Domain-Partitioned Hybrid RAG for Legal Reasoning: Toward Modular and Explainable Legal AI for India
+
+- **ID:** `2602.23371` · **primary:** `cs.IR` · **Submitted:** 2025-12-23
+- **Authors:** Rakshita Goel, S Pranav Kumar, Anmol Agrawal, et al.
+- **abs:** https://arxiv.org/abs/2602.23371 · **PDF:** https://arxiv.org/pdf/2602.23371v1
+
+## 25903. Stabilizing Multimodal Autoencoders: A Theoretical and Empirical Analysis of Fusion Strategies
+
+- **ID:** `2512.20749` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Diyar Altinses, Andreas Schwung
+- **abs:** https://arxiv.org/abs/2512.20749 · **PDF:** https://arxiv.org/pdf/2512.20749v1
+
+## 25904. Predicting Mycotoxin Contamination in Irish Oats Using Deep and Transfer Learning
+
+- **ID:** `2512.22243` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Alan Inglis, Fiona Doohan, Subramani Natarajan, et al.
+- **abs:** https://arxiv.org/abs/2512.22243 · **PDF:** https://arxiv.org/pdf/2512.22243v1
+
+## 25905. A Physics Informed Neural Network For Deriving MHD State Vectors From Global Active Regions Observations
+
+- **ID:** `2512.20747` · **primary:** `astro-ph.SR` · **Submitted:** 2025-12-23
+- **Authors:** Subhamoy Chatterjee, Mausumi Dikpati
+- **abs:** https://arxiv.org/abs/2512.20747 · **PDF:** https://arxiv.org/pdf/2512.20747v1
+
+## 25906. TrashDet: Iterative Neural Architecture Search for Efficient Waste Detection
+
+- **ID:** `2512.20746` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Tony Tran, Bin Hu
+- **abs:** https://arxiv.org/abs/2512.20746 · **PDF:** https://arxiv.org/pdf/2512.20746v1
+
+## 25907. AgentMath: Empowering Mathematical Reasoning for Large Language Models via Tool-Augmented Agent
+
+- **ID:** `2512.20745` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Haipeng Luo, Huawen Feng, Qingfeng Sun, et al.
+- **abs:** https://arxiv.org/abs/2512.20745 · **PDF:** https://arxiv.org/pdf/2512.20745v3
+
+## 25908. Fairness Evaluation of Risk Estimation Models for Lung Cancer Screening
+
+- **ID:** `2512.22242` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Shaurya Gaur, Michel Vitale, Alessa Hering, et al.
+- **abs:** https://arxiv.org/abs/2512.22242 · **PDF:** https://arxiv.org/pdf/2512.22242v1
+
+## 25909. AI-Driven Green Cognitive Radio Networks for Sustainable 6G Communication
+
+- **ID:** `2512.20739` · **primary:** `cs.NI` · **Submitted:** 2025-12-23
+- **Authors:** Anshul Sharma, Shujaatali Badami, Biky Chouhan, et al.
+- **abs:** https://arxiv.org/abs/2512.20739 · **PDF:** https://arxiv.org/pdf/2512.20739v1
+
+## 25910. FEM-Bench: A Structured Scientific Reasoning Benchmark for Evaluating Code-Generating LLMs
+
+- **ID:** `2512.20732` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Saeed Mohammadzadeh, Erfan Hamdi, Joel Shor, et al.
+- **abs:** https://arxiv.org/abs/2512.20732 · **PDF:** https://arxiv.org/pdf/2512.20732v1
+
+## 25911. SA-DiffuSeq: Addressing Computational and Scalability Challenges in Long-Document Generation with Sparse Attention
+
+- **ID:** `2512.20724` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Alexandros Christoforos, Chadbourne Davis
+- **abs:** https://arxiv.org/abs/2512.20724 · **PDF:** https://arxiv.org/pdf/2512.20724v1
+
+## 25912. From artificial to organic: Rethinking the roots of intelligence for digital health
+
+- **ID:** `2512.20723` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Prajwal Ghimire, Keyoumars Ashkan
+- **abs:** https://arxiv.org/abs/2512.20723 · **PDF:** https://arxiv.org/pdf/2512.20723v1
+
+## 25913. From Pilots to Practices: A Scoping Review of GenAI-Enabled Personalization in Computer Science Education
+
+- **ID:** `2512.20714` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Iman Reihanian, Yunfei Hou, Qingquan Sun
+- **abs:** https://arxiv.org/abs/2512.20714 · **PDF:** https://arxiv.org/pdf/2512.20714v1
+
+## 25914. Real-World Adversarial Attacks on RF-Based Drone Detectors
+
+- **ID:** `2512.20712` · **primary:** `cs.CR` · **Submitted:** 2025-12-23
+- **Authors:** Omer Gazit, Yael Itzhakev, Yuval Elovici, et al.
+- **abs:** https://arxiv.org/abs/2512.20712 · **PDF:** https://arxiv.org/pdf/2512.20712v4
+
+## 25915. LongVideoAgent: Multi-Agent Reasoning with Long Videos
+
+- **ID:** `2512.20618` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Runtao Liu, Ziyi Liu, Jiaqi Tang, et al.
+- **abs:** https://arxiv.org/abs/2512.20618 · **PDF:** https://arxiv.org/pdf/2512.20618v1
+
+## 25916. Making Large Language Models Efficient Dense Retrievers
+
+- **ID:** `2512.20612` · **primary:** `cs.IR` · **Submitted:** 2025-12-23
+- **Authors:** Yibin Lei, Shwai He, Ang Li, et al.
+- **abs:** https://arxiv.org/abs/2512.20612 · **PDF:** https://arxiv.org/pdf/2512.20612v1
+
+## 25917. FedPOD: the deployable units of training for federated learning
+
+- **ID:** `2512.20610` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Daewoon Kim, Si Young Yie, Jae Sung Lee
+- **abs:** https://arxiv.org/abs/2512.20610 · **PDF:** https://arxiv.org/pdf/2512.20610v2
+
+## 25918. Saddle-to-Saddle Dynamics Explains A Simplicity Bias Across Neural Network Architectures
+
+- **ID:** `2512.20607` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Yedi Zhang, Andrew Saxe, Peter E. Latham
+- **abs:** https://arxiv.org/abs/2512.20607 · **PDF:** https://arxiv.org/pdf/2512.20607v2
+
+## 25919. Emergent temporal abstractions in autoregressive models enable hierarchical reinforcement learning
+
+- **ID:** `2512.20605` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Seijin Kobayashi, Yanick Schimpf, Maximilian Schlegel, et al.
+- **abs:** https://arxiv.org/abs/2512.20605 · **PDF:** https://arxiv.org/pdf/2512.20605v2
+
+## 25920. MoE-DiffuSeq: Enhancing Long-Document Diffusion Models with Sparse Attention and Mixture of Experts
+
+- **ID:** `2512.20604` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Alexandros Christoforos, Chadbourne Davis
+- **abs:** https://arxiv.org/abs/2512.20604 · **PDF:** https://arxiv.org/pdf/2512.20604v2
+
+## 25921. Enhanced geometry prediction in laser directed energy deposition using meta-learning
+
+- **ID:** `2512.22241` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Abdul Malik Al Mardhouf Al Saadi, Amrita Basak
+- **abs:** https://arxiv.org/abs/2512.22241 · **PDF:** https://arxiv.org/pdf/2512.22241v1
+
+## 25922. Cube Bench: A Benchmark for Spatial Visual Reasoning in MLLMs
+
+- **ID:** `2512.20595` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Dhruv Anand, Ehsan Shareghi
+- **abs:** https://arxiv.org/abs/2512.20595 · **PDF:** https://arxiv.org/pdf/2512.20595v1
+
+## 25923. Leveraging High-Fidelity Digital Models and Reinforcement Learning for Mission Engineering: A Case Study of Aerial Firefighting Under Perfect Information
+
+- **ID:** `2512.20589` · **primary:** `cs.CY` · **Submitted:** 2025-12-23
+- **Authors:** İbrahim Oğuz Çetinkaya, Sajad Khodadadian, Taylan G. Topcu
+- **abs:** https://arxiv.org/abs/2512.20589 · **PDF:** https://arxiv.org/pdf/2512.20589v2
+
+## 25924. EvoXplain: When Machine Learning Models Agree on Predictions but Disagree on Why -- Measuring Mechanistic Multiplicity Across Training Runs
+
+- **ID:** `2512.22240` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Chama Bensmail
+- **abs:** https://arxiv.org/abs/2512.22240 · **PDF:** https://arxiv.org/pdf/2512.22240v4
+
+## 25925. Automated stereotactic radiosurgery planning using a human-in-the-loop reasoning large language model agent
+
+- **ID:** `2512.20586` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Humza Nusrat, Luke Francisco, Bing Luo, et al.
+- **abs:** https://arxiv.org/abs/2512.20586 · **PDF:** https://arxiv.org/pdf/2512.20586v1
+
+## 25926. Relu and softplus neural nets as zero-sum turn-based games
+
+- **ID:** `2512.20582` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Stephane Gaubert, Yiannis Vlassopoulos
+- **abs:** https://arxiv.org/abs/2512.20582 · **PDF:** https://arxiv.org/pdf/2512.20582v1
+
+## 25927. Can LLMs Predict Their Own Failures? Self-Awareness via Internal Circuits
+
+- **ID:** `2512.20578` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Amirhosein Ghasemabadi, Di Niu
+- **abs:** https://arxiv.org/abs/2512.20578 · **PDF:** https://arxiv.org/pdf/2512.20578v2
+
+## 25928. Improving ML Training Data with Gold-Standard Quality Metrics
+
+- **ID:** `2512.20577` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Leslie Barrett, Michael W. Sherman
+- **abs:** https://arxiv.org/abs/2512.20577 · **PDF:** https://arxiv.org/pdf/2512.20577v1
+
+## 25929. Performative Policy Gradient: Optimality in Performative Reinforcement Learning
+
+- **ID:** `2512.20576` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Debabrota Basu, Udvas Das, Brahim Driss, et al.
+- **abs:** https://arxiv.org/abs/2512.20576 · **PDF:** https://arxiv.org/pdf/2512.20576v2
+
+## 25930. Fail Fast, Win Big: Rethinking the Drafting Strategy in Speculative Decoding via Diffusion LLMs
+
+- **ID:** `2512.20573` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Rui Pan, Zhuofu Chen, Hongyi Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.20573 · **PDF:** https://arxiv.org/pdf/2512.20573v3
+
+## 25931. Distilling to Hybrid Attention Models via KL-Guided Layer Selection
+
+- **ID:** `2512.20569` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Yanhong Li, Songlin Yang, Shawn Tan, et al.
+- **abs:** https://arxiv.org/abs/2512.20569 · **PDF:** https://arxiv.org/pdf/2512.20569v1
+
+## 25932. Random Gradient-Free Optimization in Infinite Dimensional Spaces
+
+- **ID:** `2512.20566` · **primary:** `math.OC` · **Submitted:** 2025-12-23
+- **Authors:** Caio Lins Peixoto, Daniel Csillag, Bernardo F. P. da Costa, et al.
+- **abs:** https://arxiv.org/abs/2512.20566 · **PDF:** https://arxiv.org/pdf/2512.20566v1
+
+## 25933. LEAD: Minimizing Learner-Expert Asymmetry in End-to-End Driving
+
+- **ID:** `2512.20563` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Long Nguyen, Micha Fauth, Bernhard Jaeger, et al.
+- **abs:** https://arxiv.org/abs/2512.20563 · **PDF:** https://arxiv.org/pdf/2512.20563v1
+
+## 25934. Shallow Neural Networks Learn Low-Degree Spherical Polynomials with Learnable Channel Attention
+
+- **ID:** `2512.20562` · **primary:** `stat.ML` · **Submitted:** 2025-12-23
+- **Authors:** Yingzhen Yang
+- **abs:** https://arxiv.org/abs/2512.20562 · **PDF:** https://arxiv.org/pdf/2512.20562v1
+
+## 25935. Information-theoretic signatures of causality in Bayesian networks and hypergraphs
+
+- **ID:** `2512.20552` · **primary:** `cs.IT` · **Submitted:** 2025-12-23
+- **Authors:** Sung En Chiang, Zhaolu Liu, Robert L. Peach, et al.
+- **abs:** https://arxiv.org/abs/2512.20552 · **PDF:** https://arxiv.org/pdf/2512.20552v1
+
+## 25936. Advancing Multimodal Teacher Sentiment Analysis:The Large-Scale T-MED Dataset & The Effective AAM-TSA Model
+
+- **ID:** `2512.20548` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Zhiyi Duan, Xiangren Wang, Hongyu Yuan, et al.
+- **abs:** https://arxiv.org/abs/2512.20548 · **PDF:** https://arxiv.org/pdf/2512.20548v2
+
+## 25937. Energy-Aware Routing to Large Reasoning Models
+
+- **ID:** `2601.00823` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Austin R. Ellis-Mohr, Max Hartman, Lav R. Varshney
+- **abs:** https://arxiv.org/abs/2601.00823 · **PDF:** https://arxiv.org/pdf/2601.00823v1
+
+## 25938. Over-the-Air Goal-Oriented Communications
+
+- **ID:** `2512.20533` · **primary:** `eess.SP` · **Submitted:** 2025-12-23
+- **Authors:** Kyriakos Stylianopoulos, Paolo Di Lorenzo, George C. Alexandropoulos
+- **abs:** https://arxiv.org/abs/2512.20533 · **PDF:** https://arxiv.org/pdf/2512.20533v1
+
+## 25939. ScoreMatchingRiesz: Score Matching for Debiased Machine Learning and Policy Path Estimation
+
+- **ID:** `2512.20523` · **primary:** `econ.EM` · **Submitted:** 2025-12-23
+- **Authors:** Masahiro Kato
+- **abs:** https://arxiv.org/abs/2512.20523 · **PDF:** https://arxiv.org/pdf/2512.20523v2
+
+## 25940. Benchmarking LLMs for Predictive Applications in the Intensive Care Units
+
+- **ID:** `2512.20520` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Chehak Malhotra, Mehak Gopal, Akshaya Devadiga, et al.
+- **abs:** https://arxiv.org/abs/2512.20520 · **PDF:** https://arxiv.org/pdf/2512.20520v1
+
+## 25941. Explainable time-series forecasting with sampling-free SHAP for Transformers
+
+- **ID:** `2512.20514` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Matthias Hertel, Sebastian Pütz, Ralf Mikut, et al.
+- **abs:** https://arxiv.org/abs/2512.20514 · **PDF:** https://arxiv.org/pdf/2512.20514v1
+
+## 25942. Recurrent Off-Policy Deep Reinforcement Learning Doesn't Have to be Slow
+
+- **ID:** `2512.20513` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Tyler Clark, Christine Evers, Jonathon Hare
+- **abs:** https://arxiv.org/abs/2512.20513 · **PDF:** https://arxiv.org/pdf/2512.20513v1
+
+## 25943. CogCanvas: Verbatim-Grounded Artifact Extraction for Long LLM Conversations
+
+- **ID:** `2601.00821` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Tao An
+- **abs:** https://arxiv.org/abs/2601.00821 · **PDF:** https://arxiv.org/pdf/2601.00821v2
+
+## 25944. Step-DeepResearch Technical Report
+
+- **ID:** `2512.20491` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Chen Hu, Haikuo Du, Heng Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.20491 · **PDF:** https://arxiv.org/pdf/2512.20491v4
+
+## 25945. SweRank+: Multilingual, Multi-Turn Code Ranking for Software Issue Localization
+
+- **ID:** `2512.20482` · **primary:** `cs.SE` · **Submitted:** 2025-12-23
+- **Authors:** Revanth Gangi Reddy, Ye Liu, Wenting Zhao, et al.
+- **abs:** https://arxiv.org/abs/2512.20482 · **PDF:** https://arxiv.org/pdf/2512.20482v1
+
+## 25946. Toward General Semantic Chunking: A Discriminative Framework for Ultra-Long Documents
+
+- **ID:** `2602.23370` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Kaifeng Wu, Junyan Wu, Qiang Liu, et al.
+- **abs:** https://arxiv.org/abs/2602.23370 · **PDF:** https://arxiv.org/pdf/2602.23370v1
+
+## 25947. Coherence in the brain unfolds across separable temporal regimes
+
+- **ID:** `2512.20481` · **primary:** `q-bio.NC` · **Submitted:** 2025-12-23
+- **Authors:** Davide Staub, Finn Rabe, Akhil Misra, et al.
+- **abs:** https://arxiv.org/abs/2512.20481 · **PDF:** https://arxiv.org/pdf/2512.20481v3
+
+## 25948. Bohrium + SciMaster: Building the Infrastructure and Ecosystem for Agentic Science at Scale
+
+- **ID:** `2512.20469` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Linfeng Zhang, Siheng Chen, Yuzhu Cai, et al.
+- **abs:** https://arxiv.org/abs/2512.20469 · **PDF:** https://arxiv.org/pdf/2512.20469v1
+
+## 25949. Snapshot 3D image projection using a diffractive decoder
+
+- **ID:** `2512.20464` · **primary:** `physics.optics` · **Submitted:** 2025-12-23
+- **Authors:** Cagatay Isil, Alexander Chen, Yuhang Li, et al.
+- **abs:** https://arxiv.org/abs/2512.20464 · **PDF:** https://arxiv.org/pdf/2512.20464v1
+
+## 25950. The Aligned Economic Index & The State Switching Model
+
+- **ID:** `2512.20460` · **primary:** `q-fin.ST` · **Submitted:** 2025-12-23
+- **Authors:** Ilias Aarab
+- **abs:** https://arxiv.org/abs/2512.20460 · **PDF:** https://arxiv.org/pdf/2512.20460v2
+
+## 25951. A Study of Solving Life-and-Death Problems in Go Using Relevance-Zone Based Solvers
+
+- **ID:** `2512.21365` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Chung-Chin Shih, Ti-Rong Wu, Ting Han Wei, et al.
+- **abs:** https://arxiv.org/abs/2512.21365 · **PDF:** https://arxiv.org/pdf/2512.21365v1
+
+## 25952. Multi-objective hybrid knowledge distillation for efficient deep learning in smart agriculture
+
+- **ID:** `2512.22239` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Phi-Hung Hoang, Nam-Thuan Trinh, Van-Manh Tran, et al.
+- **abs:** https://arxiv.org/abs/2512.22239 · **PDF:** https://arxiv.org/pdf/2512.22239v1
+
+## 25953. Machine Learning to Predict Digital Frustration from Clickstream Data
+
+- **ID:** `2512.20438` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Jibin Joseph
+- **abs:** https://arxiv.org/abs/2512.20438 · **PDF:** https://arxiv.org/pdf/2512.20438v1
+
+## 25954. Dual-Encoder Transformer-Based Multimodal Learning for Ischemic Stroke Lesion Segmentation Using Diffusion MRI
+
+- **ID:** `2512.20436` · **primary:** `eess.IV` · **Submitted:** 2025-12-23
+- **Authors:** Muhammad Usman, Azka Rehman, Muhammad Mutti Ur Rehman, et al.
+- **abs:** https://arxiv.org/abs/2512.20436 · **PDF:** https://arxiv.org/pdf/2512.20436v1
+
+## 25955. High Dimensional Data Decomposition for Anomaly Detection of Textured Images
+
+- **ID:** `2512.20432` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Ji Song, Xing Wang, Jianguo Wu, et al.
+- **abs:** https://arxiv.org/abs/2512.20432 · **PDF:** https://arxiv.org/pdf/2512.20432v1
+
+## 25956. Evasion-Resilient Detection of DNS-over-HTTPS Data Exfiltration: A Practical Evaluation and Toolkit
+
+- **ID:** `2512.20423` · **primary:** `cs.CR` · **Submitted:** 2025-12-23
+- **Authors:** Adam Elaoumari
+- **abs:** https://arxiv.org/abs/2512.20423 · **PDF:** https://arxiv.org/pdf/2512.20423v1
+
+## 25957. Simplifying Multi-Task Architectures Through Task-Specific Normalization
+
+- **ID:** `2512.20420` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Mihai Suteu, Ovidiu Serban
+- **abs:** https://arxiv.org/abs/2512.20420 · **PDF:** https://arxiv.org/pdf/2512.20420v1
+
+## 25958. DETACH : Decomposed Spatio-Temporal Alignment for Exocentric Video and Ambient Sensors with Staged Learning
+
+- **ID:** `2512.20409` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Junho Yoon, Jaemo Jung, Hyunju Kim, et al.
+- **abs:** https://arxiv.org/abs/2512.20409 · **PDF:** https://arxiv.org/pdf/2512.20409v1
+
+## 25959. AUDRON: A Deep Learning Framework with Fused Acoustic Signatures for Drone Type Recognition
+
+- **ID:** `2512.20407` · **primary:** `cs.SD` · **Submitted:** 2025-12-23
+- **Authors:** Rajdeep Chatterjee, Sudip Chakrabarty, Trishaani Acharjee, et al.
+- **abs:** https://arxiv.org/abs/2512.20407 · **PDF:** https://arxiv.org/pdf/2512.20407v2
+
+## 25960. Sentiment-Aware Extractive and Abstractive Summarization for Unstructured Text Mining
+
+- **ID:** `2512.20404` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Junyi Liu, Stanley Kok
+- **abs:** https://arxiv.org/abs/2512.20404 · **PDF:** https://arxiv.org/pdf/2512.20404v1
+
+## 25961. BRIDGE: Budget-aware Reasoning via Intermediate Distillation with Guided Examples
+
+- **ID:** `2512.20403` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Xuan-An Le, Minh-Nam Tran, Son Nguyen
+- **abs:** https://arxiv.org/abs/2512.20403 · **PDF:** https://arxiv.org/pdf/2512.20403v1
+
+## 25962. Masking Teacher and Reinforcing Student for Distilling Vision-Language Models
+
+- **ID:** `2512.22238` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Byung-Kwan Lee, Yu-Chiang Frank Wang, Ryo Hachiuma
+- **abs:** https://arxiv.org/abs/2512.22238 · **PDF:** https://arxiv.org/pdf/2512.22238v1
+
+## 25963. GeoTransolver: Learning Physics on Irregular Domains Using Multi-scale Geometry Aware Physics Attention Transformer
+
+- **ID:** `2512.20399` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Corey Adams, Rishikesh Ranade, Ram Cherukuri, et al.
+- **abs:** https://arxiv.org/abs/2512.20399 · **PDF:** https://arxiv.org/pdf/2512.20399v2
+
+## 25964. DeepEvidence: Empowering Biomedical Discovery with Deep Knowledge Graph Research
+
+- **ID:** `2601.11560` · **primary:** `cs.IR` · **Submitted:** 2025-12-23
+- **Authors:** Zifeng Wang, Zheng Chen, Ziwei Yang, et al.
+- **abs:** https://arxiv.org/abs/2601.11560 · **PDF:** https://arxiv.org/pdf/2601.11560v1
+
+## 25965. Generative Digital Twins: Vision-Language Simulation Models for Executable Industrial Systems
+
+- **ID:** `2512.20387` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** YuChe Hsu, AnJui Wang, TsaiChing Ni, et al.
+- **abs:** https://arxiv.org/abs/2512.20387 · **PDF:** https://arxiv.org/pdf/2512.20387v4
+
+## 25966. Identifying Appropriately-Sized Services with Deep Reinforcement Learning
+
+- **ID:** `2512.20381` · **primary:** `cs.SE` · **Submitted:** 2025-12-23
+- **Authors:** Syeda Tasnim Fabiha, Saad Shafiq, Wesley Klewerton Guez Assunção, et al.
+- **abs:** https://arxiv.org/abs/2512.20381 · **PDF:** https://arxiv.org/pdf/2512.20381v1
+
+## 25967. Cross-Population White Matter Atlas Creation for Concurrent Mapping of Brain Connections in Neonates and Adults with Diffusion MRI Tractography
+
+- **ID:** `2512.20370` · **primary:** `cs.NE` · **Submitted:** 2025-12-23
+- **Authors:** Wei Zhang, Yijie Li, Ruixi Zheng, et al.
+- **abs:** https://arxiv.org/abs/2512.20370 · **PDF:** https://arxiv.org/pdf/2512.20370v1
+
+## 25968. Avoiding the Price of Adaptivity: Inference in Linear Contextual Bandits via Stability
+
+- **ID:** `2512.20368` · **primary:** `stat.ML` · **Submitted:** 2025-12-23
+- **Authors:** Samya Praharaj, Koulik Khamaru
+- **abs:** https://arxiv.org/abs/2512.20368 · **PDF:** https://arxiv.org/pdf/2512.20368v2
+
+## 25969. Clust-PSI-PFL: A Population Stability Index Approach for Clustered Non-IID Personalized Federated Learning
+
+- **ID:** `2512.20363` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Daniel M. Jimenez-Gutierrez, Mehrdad Hassanzadeh, David Solans, et al.
+- **abs:** https://arxiv.org/abs/2512.20363 · **PDF:** https://arxiv.org/pdf/2512.20363v2
+
+## 25970. Multi-LLM Thematic Analysis with Dual Reliability Metrics: Combining Cohen's Kappa and Semantic Similarity for Qualitative Research Validation
+
+- **ID:** `2512.20352` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Nilesh Jain, Hyungil Suh, Seyi Adeyinka, et al.
+- **abs:** https://arxiv.org/abs/2512.20352 · **PDF:** https://arxiv.org/pdf/2512.20352v2
+
+## 25971. Field-Space Attention for Structure-Preserving Earth System Transformers
+
+- **ID:** `2512.20350` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Maximilian Witte, Johannes Meuer, Étienne Plésiat, et al.
+- **abs:** https://arxiv.org/abs/2512.20350 · **PDF:** https://arxiv.org/pdf/2512.20350v1
+
+## 25972. Physics-guided Neural Network-based Shaft Power Prediction for Vessels
+
+- **ID:** `2512.20348` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Dogan Altan, Hamza Haruna Mohammed, Glenn Terje Lines, et al.
+- **abs:** https://arxiv.org/abs/2512.20348 · **PDF:** https://arxiv.org/pdf/2512.20348v1
+
+## 25973. Inverse Autoregressive Flows for Zero Degree Calorimeter fast simulation
+
+- **ID:** `2512.20346` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Emilia Majerz, Witold Dzwinel, Jacek Kitowski
+- **abs:** https://arxiv.org/abs/2512.20346 · **PDF:** https://arxiv.org/pdf/2512.20346v1
+
+## 25974. A DeepSeek-Powered AI System for Automated Chest Radiograph Interpretation in Clinical Practice
+
+- **ID:** `2512.20344` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Yaowei Bai, Ruiheng Zhang, Yu Lei, et al.
+- **abs:** https://arxiv.org/abs/2512.20344 · **PDF:** https://arxiv.org/pdf/2512.20344v1
+
+## 25975. AI Application Operations -- A Socio-Technical Framework for Data-driven Organizations
+
+- **ID:** `2601.06061` · **primary:** `cs.CY` · **Submitted:** 2025-12-23
+- **Authors:** Daniel Jönsson, Mattias Tiger, Stefan Ekberg, et al.
+- **abs:** https://arxiv.org/abs/2601.06061 · **PDF:** https://arxiv.org/pdf/2601.06061v1
+
+## 25976. Why Slop Matters
+
+- **ID:** `2601.06060` · **primary:** `cs.CY` · **Submitted:** 2025-12-23
+- **Authors:** Cody Kommers, Eamon Duede, Julia Gordon, et al.
+- **abs:** https://arxiv.org/abs/2601.06060 · **PDF:** https://arxiv.org/pdf/2601.06060v1
+
+## 25977. SynCraft: Guiding Large Language Models to Predict Edit Sequences for Molecular Synthesizability Optimization
+
+- **ID:** `2512.20333` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Junren Li, Luhua Lai
+- **abs:** https://arxiv.org/abs/2512.20333 · **PDF:** https://arxiv.org/pdf/2512.20333v1
+
+## 25978. Meta-information Guided Cross-domain Synergistic Diffusion Model for Low-dose PET Reconstruction
+
+- **ID:** `2512.22237` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Mengxiao Geng, Ran Hong, Xiaoling Xu, et al.
+- **abs:** https://arxiv.org/abs/2512.22237 · **PDF:** https://arxiv.org/pdf/2512.22237v1
+
+## 25979. FedDPC : Handling Data Heterogeneity and Partial Client Participation in Federated Learning
+
+- **ID:** `2512.20329` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Mrinmay Sen, Subhrajit Nag
+- **abs:** https://arxiv.org/abs/2512.20329 · **PDF:** https://arxiv.org/pdf/2512.20329v1
+
+## 25980. Toward Explaining Large Language Models in Software Engineering Tasks
+
+- **ID:** `2512.20328` · **primary:** `cs.SE` · **Submitted:** 2025-12-23
+- **Authors:** Antonio Vitale, Khai-Nguyen Nguyen, Denys Poshyvanyk, et al.
+- **abs:** https://arxiv.org/abs/2512.20328 · **PDF:** https://arxiv.org/pdf/2512.20328v1
+
+## 25981. Base Station Deployment under EMF constrain by Deep Reinforcement learning
+
+- **ID:** `2601.02385` · **primary:** `cs.NI` · **Submitted:** 2025-12-23
+- **Authors:** Mohammed Mallik, Guillaume Villemaud
+- **abs:** https://arxiv.org/abs/2601.02385 · **PDF:** https://arxiv.org/pdf/2601.02385v1
+
+## 25982. Top-K Exterior Power Persistent Homology: Algorithm, Structure, and Stability
+
+- **ID:** `2512.20325` · **primary:** `cs.CG` · **Submitted:** 2025-12-23
+- **Authors:** Yoshihiro Maruyama
+- **abs:** https://arxiv.org/abs/2512.20325 · **PDF:** https://arxiv.org/pdf/2512.20325v1
+
+## 25983. Can LLMs Solve My Grandma's Riddle? Evaluating Multilingual Large Language Models on Reasoning Traditional Bangla Tricky Riddles
+
+- **ID:** `2512.20324` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Nurul Labib Sayeedi, Md. Faiyaz Abdullah Sayeedi, Khushnur Binte Jahangir, et al.
+- **abs:** https://arxiv.org/abs/2512.20324 · **PDF:** https://arxiv.org/pdf/2512.20324v1
+
+## 25984. Deep Learning Classification of EEG Responses to Multi-Dimensional Transcranial Electrical Stimulation
+
+- **ID:** `2512.20319` · **primary:** `q-bio.NC` · **Submitted:** 2025-12-23
+- **Authors:** Alexis Pomares Pastor, Ines Ribeiro Violante, Gregory Scott
+- **abs:** https://arxiv.org/abs/2512.20319 · **PDF:** https://arxiv.org/pdf/2512.20319v1
+
+## 25985. TableGPT-R1: Advancing Tabular Reasoning Through Reinforcement Learning
+
+- **ID:** `2512.20312` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Saisai Yang, Qingyi Huang, Jing Yuan, et al.
+- **abs:** https://arxiv.org/abs/2512.20312 · **PDF:** https://arxiv.org/pdf/2512.20312v2
+
+## 25986. Algorithm for Interpretable Graph Features via Motivic Persistent Cohomology
+
+- **ID:** `2512.20311` · **primary:** `cs.CG` · **Submitted:** 2025-12-23
+- **Authors:** Yoshihiro Maruyama
+- **abs:** https://arxiv.org/abs/2512.20311 · **PDF:** https://arxiv.org/pdf/2512.20311v1
+
+## 25987. SpidR: Learning Fast and Stable Linguistic Units for Spoken Language Models Without Supervision
+
+- **ID:** `2512.20308` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Maxime Poli, Mahi Luthra, Youssef Benchekroun, et al.
+- **abs:** https://arxiv.org/abs/2512.20308 · **PDF:** https://arxiv.org/pdf/2512.20308v2
+
+## 25988. A Structured Nonparametric Framework for Nonlinear Accelerated Failure Time Models (KAN-AFT)
+
+- **ID:** `2512.20305` · **primary:** `stat.ML` · **Submitted:** 2025-12-23
+- **Authors:** Mebin Jose, Jisha Francis, Sudheesh Kumar Kattumannil
+- **abs:** https://arxiv.org/abs/2512.20305 · **PDF:** https://arxiv.org/pdf/2512.20305v2
+
+## 25989. KnowVal: A Knowledge-Augmented and Value-Guided Autonomous Driving System
+
+- **ID:** `2512.20299` · **primary:** `cs.RO` · **Submitted:** 2025-12-23
+- **Authors:** Zhongyu Xia, Wenhao Chen, Yongtao Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.20299 · **PDF:** https://arxiv.org/pdf/2512.20299v2
+
+## 25990. Patterns vs. Patients: Evaluating LLMs against Mental Health Professionals on Personality Disorder Diagnosis through First-Person Narratives
+
+- **ID:** `2512.20298` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Karolina Drożdż, Kacper Dudzic, Anna Sterna, et al.
+- **abs:** https://arxiv.org/abs/2512.20298 · **PDF:** https://arxiv.org/pdf/2512.20298v1
+
+## 25991. TAVID: Text-Driven Audio-Visual Interactive Dialogue Generation
+
+- **ID:** `2512.20296` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Ji-Hoon Kim, Junseok Ahn, Doyeop Kwak, et al.
+- **abs:** https://arxiv.org/abs/2512.20296 · **PDF:** https://arxiv.org/pdf/2512.20296v1
+
+## 25992. AprielGuard
+
+- **ID:** `2512.20293` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Jaykumar Kasundra, Anjaneya Praharaj, Sourabh Surana, et al.
+- **abs:** https://arxiv.org/abs/2512.20293 · **PDF:** https://arxiv.org/pdf/2512.20293v2
+
+## 25993. SlideTailor: Personalized Presentation Slide Generation for Scientific Papers
+
+- **ID:** `2512.20292` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Wenzheng Zeng, Mingyu Ouyang, Langyuan Cui, et al.
+- **abs:** https://arxiv.org/abs/2512.20292 · **PDF:** https://arxiv.org/pdf/2512.20292v1
+
+## 25994. Mixture-of-Experts with Gradient Conflict-Driven Subspace Topology Pruning for Emergent Modularity
+
+- **ID:** `2512.20291` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Yuxing Gan, Ziyu Lei
+- **abs:** https://arxiv.org/abs/2512.20291 · **PDF:** https://arxiv.org/pdf/2512.20291v4
+
+## 25995. UbiQVision: Quantifying Uncertainty in XAI for Image Recognition
+
+- **ID:** `2512.20288` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Akshat Dubey, Aleksandar Anžel, Bahar İlgen, et al.
+- **abs:** https://arxiv.org/abs/2512.20288 · **PDF:** https://arxiv.org/pdf/2512.20288v1
+
+## 25996. We are not able to identify AI-generated images
+
+- **ID:** `2512.22236` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Adrien Pavão
+- **abs:** https://arxiv.org/abs/2512.22236 · **PDF:** https://arxiv.org/pdf/2512.22236v1
+
+## 25997. Synthesizing Procedural Memory: Challenges and Architectures in Automated Workflow Generation
+
+- **ID:** `2512.20278` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Nishant Gaurav, Adit Akarsh, Ankit Ranjan, et al.
+- **abs:** https://arxiv.org/abs/2512.20278 · **PDF:** https://arxiv.org/pdf/2512.20278v1
+
+## 25998. ActionFlow: A Pipelined Action Acceleration for Vision Language Models on Edge
+
+- **ID:** `2512.20276` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Yuntao Dai, Hang Gu, Teng Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.20276 · **PDF:** https://arxiv.org/pdf/2512.20276v1
+
+## 25999. Graph-Symbolic Policy Enforcement and Control (G-SPEC): A Neuro-Symbolic Framework for Safe Agentic AI in 5G Autonomous Networks
+
+- **ID:** `2512.20275` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Divya Vijay, Vignesh Ethiraj
+- **abs:** https://arxiv.org/abs/2512.20275 · **PDF:** https://arxiv.org/pdf/2512.20275v1
+
+## 26000. HGAN-SDEs: Learning Neural Stochastic Differential Equations with Hermite-Guided Adversarial Training
+
+- **ID:** `2512.20272` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Yuanjian Xu, Yuan Shuai, Jianing Hao, et al.
+- **abs:** https://arxiv.org/abs/2512.20272 · **PDF:** https://arxiv.org/pdf/2512.20272v2
+
+## 26001. Optimality-Informed Neural Networks for Solving Parametric Optimization Problems
+
+- **ID:** `2512.20270` · **primary:** `math.OC` · **Submitted:** 2025-12-23
+- **Authors:** Matthias K. Hoffmann, Amine Othmane, Kathrin Flaßkamp
+- **abs:** https://arxiv.org/abs/2512.20270 · **PDF:** https://arxiv.org/pdf/2512.20270v1
+
+## 26002. DeepONet-accelerated Bayesian inversion for moving boundary problems
+
+- **ID:** `2512.20268` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Marco A. Iglesias, Michael. E. Causon, Mikhail Y. Matveev, et al.
+- **abs:** https://arxiv.org/abs/2512.20268 · **PDF:** https://arxiv.org/pdf/2512.20268v1
+
+## 26003. Debate-Enhanced Pseudo Labeling and Frequency-Aware Progressive Debiasing for Weakly-Supervised Camouflaged Object Detection with Scribble Annotations
+
+- **ID:** `2512.20260` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Jiawei Ge, Jiuxin Cao, Xinyi Li, et al.
+- **abs:** https://arxiv.org/abs/2512.20260 · **PDF:** https://arxiv.org/pdf/2512.20260v4
+
+## 26004. Unified Multimodal Brain Decoding via Cross-Subject Soft-ROI Fusion
+
+- **ID:** `2512.20249` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Xuanyu Hu
+- **abs:** https://arxiv.org/abs/2512.20249 · **PDF:** https://arxiv.org/pdf/2512.20249v2
+
+## 26005. Memory as Resonance: A Biomimetic Architecture for Infinite Context Memory on Ergodic Phonetic Manifolds
+
+- **ID:** `2512.20245` · **primary:** `cs.NE` · **Submitted:** 2025-12-23
+- **Authors:** Tarik Houichime, Abdelghani Souhar, Younes El Amrani
+- **abs:** https://arxiv.org/abs/2512.20245 · **PDF:** https://arxiv.org/pdf/2512.20245v1
+
+## 26006. MemR$^3$: Memory Retrieval via Reflective Reasoning for LLM Agents
+
+- **ID:** `2512.20237` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Xingbo Du, Loka Li, Duzhen Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.20237 · **PDF:** https://arxiv.org/pdf/2512.20237v1
+
+## 26007. Context Video Semantic Transmission with Variable Length and Rate Coding over MIMO Channels
+
+- **ID:** `2601.06059` · **primary:** `cs.IT` · **Submitted:** 2025-12-23
+- **Authors:** Bingyan Xie, Yongpeng Wu, Wenjun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2601.06059 · **PDF:** https://arxiv.org/pdf/2601.06059v1
+
+## 26008. How I Met Your Bias: Investigating Bias Amplification in Diffusion Models
+
+- **ID:** `2512.20233` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Nathan Roos, Ekaterina Iakovleva, Ani Gjergji, et al.
+- **abs:** https://arxiv.org/abs/2512.20233 · **PDF:** https://arxiv.org/pdf/2512.20233v1
+
+## 26009. Adaptive Multi-task Learning for Probabilistic Load Forecasting
+
+- **ID:** `2512.20232` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Onintze Zaballa, Verónica Álvarez, Santiago Mazuelas
+- **abs:** https://arxiv.org/abs/2512.20232 · **PDF:** https://arxiv.org/pdf/2512.20232v2
+
+## 26010. Generalisation in Multitask Fitted Q-Iteration and Offline Q-learning
+
+- **ID:** `2512.20220` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Kausthubh Manda, Raghuram Bharadwaj Diddigi
+- **abs:** https://arxiv.org/abs/2512.20220 · **PDF:** https://arxiv.org/pdf/2512.20220v1
+
+## 26011. Cost-TrustFL: Cost-Aware Hierarchical Federated Learning with Lightweight Reputation Evaluation across Multi-Cloud
+
+- **ID:** `2512.20218` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Jixiao Yang, Jinyu Chen, Zixiao Huang, et al.
+- **abs:** https://arxiv.org/abs/2512.20218 · **PDF:** https://arxiv.org/pdf/2512.20218v1
+
+## 26012. TongSIM: A General Platform for Simulating Intelligent Machines
+
+- **ID:** `2512.20206` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Zhe Sun, Kunlun Wu, Chuanjian Fu, et al.
+- **abs:** https://arxiv.org/abs/2512.20206 · **PDF:** https://arxiv.org/pdf/2512.20206v1
+
+## 26013. Corpus of Cross-lingual Dialogues with Minutes and Detection of Misunderstandings
+
+- **ID:** `2512.20204` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Marko Čechovič, Natália Komorníková, Dominik Macháček, et al.
+- **abs:** https://arxiv.org/abs/2512.20204 · **PDF:** https://arxiv.org/pdf/2512.20204v1
+
+## 26014. Asynchronous Fast-Slow Vision-Language-Action Policies for Whole-Body Robotic Manipulation
+
+- **ID:** `2512.20188` · **primary:** `cs.RO` · **Submitted:** 2025-12-23
+- **Authors:** Teqiang Zou, Hongliang Zeng, Yuxuan Nong, et al.
+- **abs:** https://arxiv.org/abs/2512.20188 · **PDF:** https://arxiv.org/pdf/2512.20188v1
+
+## 26015. From Visual Perception to Deep Empathy: An Automated Assessment Framework for House-Tree-Person Drawings Using Multimodal LLMs and Multi-Agent Collaboration
+
+- **ID:** `2512.21360` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Shuide Wen, Yu Sun, Beier Ku, et al.
+- **abs:** https://arxiv.org/abs/2512.21360 · **PDF:** https://arxiv.org/pdf/2512.21360v1
+
+## 26016. FaithLens: Detecting and Explaining Faithfulness Hallucination
+
+- **ID:** `2512.20182` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Shuzheng Si, Qingyi Wang, Haozhe Zhao, et al.
+- **abs:** https://arxiv.org/abs/2512.20182 · **PDF:** https://arxiv.org/pdf/2512.20182v2
+
+## 26017. NeuralCrop: Combining physics and machine learning for improved crop yield projections
+
+- **ID:** `2512.20177` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Yunan Lin, Sebastian Bathiany, Maha Badri, et al.
+- **abs:** https://arxiv.org/abs/2512.20177 · **PDF:** https://arxiv.org/pdf/2512.20177v2
+
+## 26018. Towards Natural Language-Based Document Image Retrieval: New Dataset and Benchmark
+
+- **ID:** `2512.20174` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Hao Guo, Xugong Qin, Jun Jie Ou Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.20174 · **PDF:** https://arxiv.org/pdf/2512.20174v1
+
+## 26019. Offline Safe Policy Optimization From Heterogeneous Feedback
+
+- **ID:** `2512.20173` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Ze Gong, Pradeep Varakantham, Akshat Kumar
+- **abs:** https://arxiv.org/abs/2512.20173 · **PDF:** https://arxiv.org/pdf/2512.20173v1
+
+## 26020. Learning to Reason in LLMs by Expectation Maximization
+
+- **ID:** `2512.20169` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Junghyun Lee, Branislav Kveton, Anup Rao, et al.
+- **abs:** https://arxiv.org/abs/2512.20169 · **PDF:** https://arxiv.org/pdf/2512.20169v2
+
+## 26021. Odysseus: Jailbreaking Commercial Multimodal LLM-integrated Systems via Dual Steganography
+
+- **ID:** `2512.20168` · **primary:** `cs.CR` · **Submitted:** 2025-12-23
+- **Authors:** Songze Li, Jiameng Cheng, Yiming Li, et al.
+- **abs:** https://arxiv.org/abs/2512.20168 · **PDF:** https://arxiv.org/pdf/2512.20168v1
+
+## 26022. AI Security Beyond Core Domains: Resume Screening as a Case Study of Adversarial Vulnerabilities in Specialized LLM Applications
+
+- **ID:** `2512.20164` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Honglin Mu, Jinghao Liu, Kaiyang Wan, et al.
+- **abs:** https://arxiv.org/abs/2512.20164 · **PDF:** https://arxiv.org/pdf/2512.20164v1
+
+## 26023. Concept Generalization in Humans and Large Language Models: Insights from the Number Game
+
+- **ID:** `2512.20162` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Arghavan Bazigaran, Hansem Sohn
+- **abs:** https://arxiv.org/abs/2512.20162 · **PDF:** https://arxiv.org/pdf/2512.20162v1
+
+## 26024. A Bidirectional Gated Recurrent Unit Model for PUE Prediction in Data Centers
+
+- **ID:** `2512.20161` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Dhivya Dharshini Kannan, Anupam Trivedi, Dipti Srinivasan
+- **abs:** https://arxiv.org/abs/2512.20161 · **PDF:** https://arxiv.org/pdf/2512.20161v1
+
+## 26025. AXIOM: Benchmarking LLM-as-a-Judge for Code via Rule-Based Perturbation and Multisource Quality Calibration
+
+- **ID:** `2512.20159` · **primary:** `cs.SE` · **Submitted:** 2025-12-23
+- **Authors:** Ruiqi Wang, Xinchen Wang, Cuiyun Gao, et al.
+- **abs:** https://arxiv.org/abs/2512.20159 · **PDF:** https://arxiv.org/pdf/2512.20159v1
+
+## 26026. Fun-Audio-Chat Technical Report
+
+- **ID:** `2512.20156` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Tongyi Fun Team, Qian Chen, Luyao Cheng, et al.
+- **abs:** https://arxiv.org/abs/2512.20156 · **PDF:** https://arxiv.org/pdf/2512.20156v4
+
+## 26027. DiRL: An Efficient Post-Training Framework for Diffusion Language Models
+
+- **ID:** `2512.22234` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Ying Zhu, Jiaxin Wan, Xiaoran Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.22234 · **PDF:** https://arxiv.org/pdf/2512.22234v2
+
+## 26028. Retrieval-augmented Prompt Learning for Pre-trained Foundation Models
+
+- **ID:** `2512.20145` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Xiang Chen, Yixin Ou, Quan Feng, et al.
+- **abs:** https://arxiv.org/abs/2512.20145 · **PDF:** https://arxiv.org/pdf/2512.20145v1
+
+## 26029. Multi-hop Reasoning via Early Knowledge Alignment
+
+- **ID:** `2512.20144` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Yuxin Wang, Shicheng Fang, Bo Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.20144 · **PDF:** https://arxiv.org/pdf/2512.20144v2
+
+## 26030. Enhancing Zero-Shot Time Series Forecasting in Off-the-Shelf LLMs via Noise Injection
+
+- **ID:** `2512.20140` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Xingyou Yin, Ceyao Zhang, Min Hu, et al.
+- **abs:** https://arxiv.org/abs/2512.20140 · **PDF:** https://arxiv.org/pdf/2512.20140v1
+
+## 26031. M$^3$KG-RAG: Multi-hop Multimodal Knowledge Graph-enhanced Retrieval-Augmented Generation
+
+- **ID:** `2512.20136` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Hyeongcheol Park, Jiyoung Seo, Jaewon Mun, et al.
+- **abs:** https://arxiv.org/abs/2512.20136 · **PDF:** https://arxiv.org/pdf/2512.20136v2
+
+## 26032. MolAct: An Agentic RL Framework for Molecular Editing and Property Optimization
+
+- **ID:** `2512.20135` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Zhuo Yang, Yeyun Chen, Jiaqing Xie, et al.
+- **abs:** https://arxiv.org/abs/2512.20135 · **PDF:** https://arxiv.org/pdf/2512.20135v2
+
+## 26033. Sample-Efficient Policy Constraint Offline Deep Reinforcement Learning based on Sample Filtering
+
+- **ID:** `2512.20115` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Yuanhao Chen, Qi Liu, Pengbin Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.20115 · **PDF:** https://arxiv.org/pdf/2512.20115v1
+
+## 26034. Evolutionary Neural Architecture Search with Dual Contrastive Learning
+
+- **ID:** `2512.20112` · **primary:** `cs.NE` · **Submitted:** 2025-12-23
+- **Authors:** Xian-Rong Zhang, Yue-Jiao Gong, Wei-Neng Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.20112 · **PDF:** https://arxiv.org/pdf/2512.20112v1
+
+## 26035. ABBEL: LLM Agents Acting through Belief Bottlenecks Expressed in Language
+
+- **ID:** `2512.20111` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Aly Lidayan, Jakob Bjorner, Satvik Golechha, et al.
+- **abs:** https://arxiv.org/abs/2512.20111 · **PDF:** https://arxiv.org/pdf/2512.20111v1
+
+## 26036. A Novel Graph-Sequence Learning Model for Inductive Text Classification
+
+- **ID:** `2512.20097` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Zuo Wang, Ye Yuan
+- **abs:** https://arxiv.org/abs/2512.20097 · **PDF:** https://arxiv.org/pdf/2512.20097v1
+
+## 26037. Information-directed sampling for bandits: a primer
+
+- **ID:** `2512.20096` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Annika Hirling, Giorgio Nicoletti, Antonio Celani
+- **abs:** https://arxiv.org/abs/2512.20096 · **PDF:** https://arxiv.org/pdf/2512.20096v1
+
+## 26038. Scalable Cloud-Native Architectures for Intelligent PMU Data Processing
+
+- **ID:** `2512.22231` · **primary:** `cs.DC` · **Submitted:** 2025-12-23
+- **Authors:** Nachiappan Chockalingam, Akshay Deshpande, Lokesh Butra, et al.
+- **abs:** https://arxiv.org/abs/2512.22231 · **PDF:** https://arxiv.org/pdf/2512.22231v1
+
+## 26039. Jensen-Shannon Divergence Message-Passing for Rich-Text Graph Representation Learning
+
+- **ID:** `2512.20094` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Zuo Wang, Ye Yuan
+- **abs:** https://arxiv.org/abs/2512.20094 · **PDF:** https://arxiv.org/pdf/2512.20094v1
+
+## 26040. Memory-T1: Reinforcement Learning for Temporal Reasoning in Multi-session Agents
+
+- **ID:** `2512.20092` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Yiming Du, Baojun Wang, Yifan Xiang, et al.
+- **abs:** https://arxiv.org/abs/2512.20092 · **PDF:** https://arxiv.org/pdf/2512.20092v1
+
+## 26041. Item Region-based Style Classification Network (IRSN): A Fashion Style Classifier Based on Domain Knowledge of Fashion Experts
+
+- **ID:** `2512.20088` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Jinyoung Choi, Youngchae Kwon, Injung Kim
+- **abs:** https://arxiv.org/abs/2512.20088 · **PDF:** https://arxiv.org/pdf/2512.20088v1
+
+## 26042. Spatio-Temporal Graphs Beyond Grids: Benchmark for Maritime Anomaly Detection
+
+- **ID:** `2512.20086` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Jeehong Kim, Youngseok Hwang, Minchan Kim, et al.
+- **abs:** https://arxiv.org/abs/2512.20086 · **PDF:** https://arxiv.org/pdf/2512.20086v1
+
+## 26043. QE-Catalytic: A Graph-Language Multimodal Base Model for Relaxed-Energy Prediction in Catalytic Adsorption
+
+- **ID:** `2512.20084` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Yanjie Li, Jian Xu, Xueqing Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.20084 · **PDF:** https://arxiv.org/pdf/2512.20084v1
+
+## 26044. Adaptive Financial Sentiment Analysis for NIFTY 50 via Instruction-Tuned LLMs , RAG and Reinforcement Learning Approaches
+
+- **ID:** `2512.20082` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Chaithra, Kamesh Kadimisetty, Biju R Mohan
+- **abs:** https://arxiv.org/abs/2512.20082 · **PDF:** https://arxiv.org/pdf/2512.20082v2
+
+## 26045. CBA: Communication-Bound-Aware Cross-Domain Resource Assignment for Pipeline-Parallel Distributed LLM Training in Dynamic Multi-DC Optical Networks
+
+- **ID:** `2512.20080` · **primary:** `cs.NI` · **Submitted:** 2025-12-23
+- **Authors:** Dianxuan Fu, Xiaomin Liu, Yihao Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.20080 · **PDF:** https://arxiv.org/pdf/2512.20080v1
+
+## 26046. Reason2Decide: Rationale-Driven Multi-Task Learning
+
+- **ID:** `2512.20074` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** H M Quamran Hasan, Housam Khalifa Bashier, Jiayi Dai, et al.
+- **abs:** https://arxiv.org/abs/2512.20074 · **PDF:** https://arxiv.org/pdf/2512.20074v2
+
+## 26047. PairFlow: Closed-Form Source-Target Coupling for Few-Step Generation in Discrete Flow Models
+
+- **ID:** `2512.20063` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Mingue Park, Jisung Hwang, Seungwoo Yoo, et al.
+- **abs:** https://arxiv.org/abs/2512.20063 · **PDF:** https://arxiv.org/pdf/2512.20063v2
+
+## 26048. On the Effectiveness of Instruction-Tuning Local LLMs for Identifying Software Vulnerabilities
+
+- **ID:** `2512.20062` · **primary:** `cs.CR` · **Submitted:** 2025-12-23
+- **Authors:** Sangryu Park, Gihyuk Ko, Homook Cho
+- **abs:** https://arxiv.org/abs/2512.20062 · **PDF:** https://arxiv.org/pdf/2512.20062v1
+
+## 26049. Scaling Reinforcement Learning for Content Moderation with Large Language Models
+
+- **ID:** `2512.20061` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Hamed Firooz, Rui Liu, Yuchen Lu, et al.
+- **abs:** https://arxiv.org/abs/2512.20061 · **PDF:** https://arxiv.org/pdf/2512.20061v1
+
+## 26050. DS-HGCN: A Dual-Stream Hypergraph Convolutional Network for Predicting Student Engagement via Social Contagion
+
+- **ID:** `2512.20059` · **primary:** `cs.MM` · **Submitted:** 2025-12-23
+- **Authors:** Ziyang Fan, Li Tao, Yi Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.20059 · **PDF:** https://arxiv.org/pdf/2512.20059v1
+
+## 26051. Deep Eigenspace Network for Parametric Non-self-adjoint Eigenvalue Problems
+
+- **ID:** `2512.20058` · **primary:** `math.NA` · **Submitted:** 2025-12-23
+- **Authors:** H. Li, J. Sun, Z. Zhang
+- **abs:** https://arxiv.org/abs/2512.20058 · **PDF:** https://arxiv.org/pdf/2512.20058v2
+
+## 26052. Structure-Preserving Nonlinear Sufficient Dimension Reduction for Tensors
+
+- **ID:** `2512.20057` · **primary:** `math.ST` · **Submitted:** 2025-12-23
+- **Authors:** Dianjun Lin, Bing Li, Lingzhou Xue
+- **abs:** https://arxiv.org/abs/2512.20057 · **PDF:** https://arxiv.org/pdf/2512.20057v1
+
+## 26053. Towards Generative Location Awareness for Disaster Response: A Probabilistic Cross-view Geolocalization Approach
+
+- **ID:** `2512.20056` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Hao Li, Fabian Deuser, Wenping Yin, et al.
+- **abs:** https://arxiv.org/abs/2512.20056 · **PDF:** https://arxiv.org/pdf/2512.20056v1
+
+## 26054. An Optimal Policy for Learning Controllable Dynamics by Exploration
+
+- **ID:** `2512.20053` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Peter N. Loxley
+- **abs:** https://arxiv.org/abs/2512.20053 · **PDF:** https://arxiv.org/pdf/2512.20053v1
+
+## 26055. Learning Skills from Action-Free Videos
+
+- **ID:** `2512.20052` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Hung-Chieh Fang, Kuo-Han Hung, Chu-Rong Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.20052 · **PDF:** https://arxiv.org/pdf/2512.20052v1
+
+## 26056. Generative Bayesian Hyperparameter Tuning
+
+- **ID:** `2512.20051` · **primary:** `stat.ML` · **Submitted:** 2025-12-23
+- **Authors:** Hedibert Lopes, Nick Polson, Vadim Sokolov
+- **abs:** https://arxiv.org/abs/2512.20051 · **PDF:** https://arxiv.org/pdf/2512.20051v1
+
+## 26057. Multi-Level Embedding Conformer Framework for Bengali Automatic Speech Recognition
+
+- **ID:** `2601.09710` · **primary:** `eess.AS` · **Submitted:** 2025-12-23
+- **Authors:** Md. Nazmus Sakib, Golam Mahmud, Md. Maruf Bangabashi, et al.
+- **abs:** https://arxiv.org/abs/2601.09710 · **PDF:** https://arxiv.org/pdf/2601.09710v1
+
+## 26058. Self-motion as a structural prior for coherent and robust formation of cognitive maps
+
+- **ID:** `2512.20044` · **primary:** `q-bio.NC` · **Submitted:** 2025-12-23
+- **Authors:** Yingchao Yu, Pengfei Sun, Yaochu Jin, et al.
+- **abs:** https://arxiv.org/abs/2512.20044 · **PDF:** https://arxiv.org/pdf/2512.20044v1
+
+## 26059. Discovering Symmetry Groups with Flow Matching
+
+- **ID:** `2512.20043` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Yuxuan Chen, Jung Yeon Park, Floor Eijkelboom, et al.
+- **abs:** https://arxiv.org/abs/2512.20043 · **PDF:** https://arxiv.org/pdf/2512.20043v2
+
+## 26060. Beyond Vision: Contextually Enriched Image Captioning with Multi-Modal Retrieval
+
+- **ID:** `2512.20042` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Nguyen Lam Phu Quy, Pham Phu Hoa, Tran Chi Nguyen, et al.
+- **abs:** https://arxiv.org/abs/2512.20042 · **PDF:** https://arxiv.org/pdf/2512.20042v2
+
+## 26061. Optimal Anytime-Valid Tests for Composite Nulls
+
+- **ID:** `2512.20039` · **primary:** `math.ST` · **Submitted:** 2025-12-23
+- **Authors:** Shubhanshu Shekhar
+- **abs:** https://arxiv.org/abs/2512.20039 · **PDF:** https://arxiv.org/pdf/2512.20039v1
+
+## 26062. DecoKAN: Interpretable Decomposition for Forecasting Cryptocurrency Market Dynamics
+
+- **ID:** `2512.20028` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Yuan Gao, Zhenguo Dong, Xuelong Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.20028 · **PDF:** https://arxiv.org/pdf/2512.20028v1
+
+## 26063. Hierarchical Geometry of Cognitive States in Transformer Embedding Spaces
+
+- **ID:** `2512.22227` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Sophie Zhao
+- **abs:** https://arxiv.org/abs/2512.22227 · **PDF:** https://arxiv.org/pdf/2512.22227v1
+
+## 26064. VideoScaffold: Elastic-Scale Visual Hierarchies for Streaming Video Understanding in MLLMs
+
+- **ID:** `2512.22226` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Naishan Zheng, Jie Huang, Qingpei Guo, et al.
+- **abs:** https://arxiv.org/abs/2512.22226 · **PDF:** https://arxiv.org/pdf/2512.22226v1
+
+## 26065. Gaussian Process Assisted Meta-learning for Image Classification and Object Detection Models
+
+- **ID:** `2512.20021` · **primary:** `stat.ML` · **Submitted:** 2025-12-23
+- **Authors:** Anna R. Flowers, Christopher T. Franck, Robert B. Gramacy, et al.
+- **abs:** https://arxiv.org/abs/2512.20021 · **PDF:** https://arxiv.org/pdf/2512.20021v1
+
+## 26066. Bring My Cup! Personalizing Vision-Language-Action Models with Visual Attentive Prompting
+
+- **ID:** `2512.20014` · **primary:** `cs.RO` · **Submitted:** 2025-12-23
+- **Authors:** Sangoh Lee, Sangwoo Mo, Wook-Shin Han
+- **abs:** https://arxiv.org/abs/2512.20014 · **PDF:** https://arxiv.org/pdf/2512.20014v2
+
+## 26067. Reliable LLM-Based Edge-Cloud-Expert Cascades for Telecom Knowledge Systems
+
+- **ID:** `2512.20012` · **primary:** `eess.SP` · **Submitted:** 2025-12-23
+- **Authors:** Qiushuo Hou, Sangwoo Park, Matteo Zecchin, et al.
+- **abs:** https://arxiv.org/abs/2512.20012 · **PDF:** https://arxiv.org/pdf/2512.20012v1
+
+## 26068. Semiparametric KSD test: unifying score and distance-based approaches for goodness-of-fit testing
+
+- **ID:** `2512.20007` · **primary:** `stat.ML` · **Submitted:** 2025-12-23
+- **Authors:** Zhihan Huang, Ziang Niu
+- **abs:** https://arxiv.org/abs/2512.20007 · **PDF:** https://arxiv.org/pdf/2512.20007v2
+
+## 26069. Orthogonal Activation with Implicit Group-Aware Bias Learning for Class Imbalance
+
+- **ID:** `2512.20006` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Sukumar Kishanthan, Asela Hevapathige
+- **abs:** https://arxiv.org/abs/2512.20006 · **PDF:** https://arxiv.org/pdf/2512.20006v1
+
+## 26070. Literature Mining System for Nutraceutical Biosynthesis: From AI Framework to Biological Insight
+
+- **ID:** `2512.22225` · **primary:** `q-bio.QM` · **Submitted:** 2025-12-23
+- **Authors:** Xinyang Sun, Nipon Sarmah, Miao Guo
+- **abs:** https://arxiv.org/abs/2512.22225 · **PDF:** https://arxiv.org/pdf/2512.22225v1
+
+## 26071. IoT-based Android Malware Detection Using Graph Neural Network With Adversarial Defense
+
+- **ID:** `2512.20004` · **primary:** `cs.CR` · **Submitted:** 2025-12-23
+- **Authors:** Rahul Yumlembam, Biju Issac, Seibu Mary Jacob, et al.
+- **abs:** https://arxiv.org/abs/2512.20004 · **PDF:** https://arxiv.org/pdf/2512.20004v1
+
+## 26072. Control Variate Score Matching for Diffusion Models
+
+- **ID:** `2512.20003` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Khaled Kahouli, Romuald Elie, Klaus-Robert Müller, et al.
+- **abs:** https://arxiv.org/abs/2512.20003 · **PDF:** https://arxiv.org/pdf/2512.20003v1
+
+## 26073. LoFT-LLM: Low-Frequency Time-Series Forecasting with Large Language Models
+
+- **ID:** `2512.20002` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Jiacheng You, Jingcheng Yang, Yuhang Xie, et al.
+- **abs:** https://arxiv.org/abs/2512.20002 · **PDF:** https://arxiv.org/pdf/2512.20002v3
+
+## 26074. Schoenfeld's Anatomy of Mathematical Reasoning by Language Models
+
+- **ID:** `2512.19995` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Ming Li, Chenrui Fan, Yize Cheng, et al.
+- **abs:** https://arxiv.org/abs/2512.19995 · **PDF:** https://arxiv.org/pdf/2512.19995v1
+
+## 26075. S$^3$IT: A Benchmark for Spatially Situated Social Intelligence Test
+
+- **ID:** `2512.19992` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Zhe Sun, Xueyuan Yang, Yujie Lu, et al.
+- **abs:** https://arxiv.org/abs/2512.19992 · **PDF:** https://arxiv.org/pdf/2512.19992v1
+
+## 26076. Bloom Filter Encoding for Machine Learning
+
+- **ID:** `2512.19991` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** John Cartmell, Mihaela Cardei, Ionut Cardei
+- **abs:** https://arxiv.org/abs/2512.19991 · **PDF:** https://arxiv.org/pdf/2512.19991v1
+
+## 26077. A Novel CNN Gradient Boosting Ensemble for Guava Disease Detection
+
+- **ID:** `2512.19989` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Tamim Ahasan Rijon, Yeasin Arafath
+- **abs:** https://arxiv.org/abs/2512.19989 · **PDF:** https://arxiv.org/pdf/2512.19989v1
+
+## 26078. Covariance-Aware Simplex Projection for Cardinality-Constrained Portfolio Optimization
+
+- **ID:** `2512.19986` · **primary:** `q-fin.PM` · **Submitted:** 2025-12-23
+- **Authors:** Nikolaos Iliopoulos
+- **abs:** https://arxiv.org/abs/2512.19986 · **PDF:** https://arxiv.org/pdf/2512.19986v1
+
+## 26079. Neuron-Guided Interpretation of Code LLMs: Where, Why, and How?
+
+- **ID:** `2512.19980` · **primary:** `cs.SE` · **Submitted:** 2025-12-23
+- **Authors:** Zhe Yin, Xiaodong Gu, Beijun Shen
+- **abs:** https://arxiv.org/abs/2512.19980 · **PDF:** https://arxiv.org/pdf/2512.19980v2
+
+## 26080. Regression of Functions by Quantum Neural Networks Circuits
+
+- **ID:** `2512.19978` · **primary:** `quant-ph` · **Submitted:** 2025-12-23
+- **Authors:** Fernando M. de Paula Neto, Lucas dos Reis Silva, Paulo S. G. de Mattos Neto, et al.
+- **abs:** https://arxiv.org/abs/2512.19978 · **PDF:** https://arxiv.org/pdf/2512.19978v1
+
+## 26081. Spatio-Temporal Graph Neural Networks for Dairy Farm Sustainability Forecasting and Counterfactual Policy Analysis
+
+- **ID:** `2512.19970` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Surya Jayakumar, Kieran Sullivan, John McLaughlin, et al.
+- **abs:** https://arxiv.org/abs/2512.19970 · **PDF:** https://arxiv.org/pdf/2512.19970v1
+
+## 26082. FGDCC: Fine-Grained Deep Cluster Categorization -- A Framework for Intra-Class Variability Problems in Plant Classification
+
+- **ID:** `2512.19960` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Luciano Araujo Dourado Filho, Rodrigo Tripodi Calumby
+- **abs:** https://arxiv.org/abs/2512.19960 · **PDF:** https://arxiv.org/pdf/2512.19960v1
+
+## 26083. Zero-Shot Segmentation through Prototype-Guidance for Multi-Label Plant Species Identification
+
+- **ID:** `2512.19957` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Luciano Araujo Dourado Filho, Almir Moreira da Silva Neto, Rodrigo Pereira David, et al.
+- **abs:** https://arxiv.org/abs/2512.19957 · **PDF:** https://arxiv.org/pdf/2512.19957v1
+
+## 26084. Bias Beneath the Tone: Empirical Characterisation of Tone Bias in LLM-Driven UX Systems
+
+- **ID:** `2512.19950` · **primary:** `cs.CL` · **Submitted:** 2025-12-23
+- **Authors:** Heet Bodara, Md Masum Mushfiq, Isma Farah Siddiqui
+- **abs:** https://arxiv.org/abs/2512.19950 · **PDF:** https://arxiv.org/pdf/2512.19950v1
+
+## 26085. How Much 3D Do Video Foundation Models Encode?
+
+- **ID:** `2512.19949` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Zixuan Huang, Xiang Li, Zhaoyang Lv, et al.
+- **abs:** https://arxiv.org/abs/2512.19949 · **PDF:** https://arxiv.org/pdf/2512.19949v1
+
+## 26086. Block-Recurrent Dynamics in Vision Transformers
+
+- **ID:** `2512.19941` · **primary:** `cs.CV` · **Submitted:** 2025-12-23
+- **Authors:** Mozes Jacobs, Thomas Fel, Richard Hakim, et al.
+- **abs:** https://arxiv.org/abs/2512.19941 · **PDF:** https://arxiv.org/pdf/2512.19941v6
+
+## 26087. ReGAIN: Retrieval-Grounded AI Framework for Network Traffic Analysis
+
+- **ID:** `2512.22223` · **primary:** `cs.LG` · **Submitted:** 2025-12-23
+- **Authors:** Shaghayegh Shajarian, Kennedy Marsh, James Benson, et al.
+- **abs:** https://arxiv.org/abs/2512.22223 · **PDF:** https://arxiv.org/pdf/2512.22223v1
+
+## 26088. Interpolative Decoding: Exploring the Spectrum of Personality Traits in LLMs
+
+- **ID:** `2512.19937` · **primary:** `cs.AI` · **Submitted:** 2025-12-23
+- **Authors:** Eric Yeh, John Cadigan, Ran Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.19937 · **PDF:** https://arxiv.org/pdf/2512.19937v1
+
+## 26089. GIMLET: Generalizable and Interpretable Model Learning through Embedded Thermodynamics
+
+- **ID:** `2512.19936` · **primary:** `physics.flu-dyn` · **Submitted:** 2025-12-22
+- **Authors:** Suguru Shiratori, Elham Kiyani, Khemraj Shukla, et al.
+- **abs:** https://arxiv.org/abs/2512.19936 · **PDF:** https://arxiv.org/pdf/2512.19936v2
+
+## 26090. Conditional Adversarial Fragility in Financial Machine Learning under Macroeconomic Stress
+
+- **ID:** `2512.19935` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Samruddhi Baviskar
+- **abs:** https://arxiv.org/abs/2512.19935 · **PDF:** https://arxiv.org/pdf/2512.19935v1
+
+## 26091. Vehicle-centric Perception via Multimodal Structured Pre-training
+
+- **ID:** `2512.19934` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Wentao Wu, Xiao Wang, Chenglong Li, et al.
+- **abs:** https://arxiv.org/abs/2512.19934 · **PDF:** https://arxiv.org/pdf/2512.19934v1
+
+## 26092. PRISM: A Personality-Driven Multi-Agent Framework for Social Media Simulation
+
+- **ID:** `2512.19933` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Zhixiang Lu, Xueyuan Deng, Yiran Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.19933 · **PDF:** https://arxiv.org/pdf/2512.19933v1
+
+## 26093. Agentic AI for Autonomous, Explainable, and Real-Time Credit Risk Decision-Making
+
+- **ID:** `2601.00818` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Chandra Sekhar Kubam
+- **abs:** https://arxiv.org/abs/2601.00818 · **PDF:** https://arxiv.org/pdf/2601.00818v1
+
+## 26094. Unified Brain Surface and Volume Registration
+
+- **ID:** `2512.19928` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** S. Mazdak Abulnaga, Andrew Hoopes, Malte Hoffmann, et al.
+- **abs:** https://arxiv.org/abs/2512.19928 · **PDF:** https://arxiv.org/pdf/2512.19928v1
+
+## 26095. Müntz-Szász Networks: Neural Architectures with Learnable Power-Law Bases
+
+- **ID:** `2512.22222` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Gnankan Landry Regis N'guessan
+- **abs:** https://arxiv.org/abs/2512.22222 · **PDF:** https://arxiv.org/pdf/2512.22222v3
+
+## 26096. The Seismic Wavefield Common Task Framework
+
+- **ID:** `2512.19927` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Alexey Yermakov, Yue Zhao, Marine Denolle, et al.
+- **abs:** https://arxiv.org/abs/2512.19927 · **PDF:** https://arxiv.org/pdf/2512.19927v1
+
+## 26097. Mitigating LLM Hallucination via Behaviorally Calibrated Reinforcement Learning
+
+- **ID:** `2512.19920` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Jiayun Wu, Jiashuo Liu, Zhiyuan Zeng, et al.
+- **abs:** https://arxiv.org/abs/2512.19920 · **PDF:** https://arxiv.org/pdf/2512.19920v3
+
+## 26098. A Time-efficient Prioritised Scheduling Algorithm to Optimise Initial Flock Formation of Drones
+
+- **ID:** `2512.19914` · **primary:** `cs.RO` · **Submitted:** 2025-12-22
+- **Authors:** Sujan Warnakulasooriya, Andreas Willig, Xiaobing Wu
+- **abs:** https://arxiv.org/abs/2512.19914 · **PDF:** https://arxiv.org/pdf/2512.19914v1
+
+## 26099. Quasiprobabilistic Density Ratio Estimation with a Reverse Engineered Classification Loss Function
+
+- **ID:** `2512.19913` · **primary:** `stat.ML` · **Submitted:** 2025-12-22
+- **Authors:** Matthew Drnevich, Stephen Jiggins, Kyle Cranmer
+- **abs:** https://arxiv.org/abs/2512.19913 · **PDF:** https://arxiv.org/pdf/2512.19913v1
+
+## 26100. Modeling Non-Ergodic Path Effects Using Conditional Generative Model for Fourier Amplitude Spectra
+
+- **ID:** `2512.19909` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Maxime Lacour, Pu Ren, Rie Nakata, et al.
+- **abs:** https://arxiv.org/abs/2512.19909 · **PDF:** https://arxiv.org/pdf/2512.19909v1
+
+## 26101. Counterfactual LLM-based Framework for Measuring Rhetorical Style
+
+- **ID:** `2512.19908` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Jingyi Qiu, Hong Chen, Zongyi Li
+- **abs:** https://arxiv.org/abs/2512.19908 · **PDF:** https://arxiv.org/pdf/2512.19908v1
+
+## 26102. Mechanism-Based Intelligence (MBI): Differentiable Incentives for Rational Coordination and Guaranteed Alignment in Multi-Agent Systems
+
+- **ID:** `2512.20688` · **primary:** `cs.GT` · **Submitted:** 2025-12-22
+- **Authors:** Stefano Grassi
+- **abs:** https://arxiv.org/abs/2512.20688 · **PDF:** https://arxiv.org/pdf/2512.20688v1
+
+## 26103. Demystifying LLM-as-a-Judge: Analytically Tractable Model for Inference-Time Scaling
+
+- **ID:** `2512.19905` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Indranil Halder, Cengiz Pehlevan
+- **abs:** https://arxiv.org/abs/2512.19905 · **PDF:** https://arxiv.org/pdf/2512.19905v2
+
+## 26104. How well do Large Language Models Recognize Instructional Moves? Establishing Baselines for Foundation Models in Educational Discourse
+
+- **ID:** `2512.19903` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Kirk Vanacore, Rene F. Kizilcec
+- **abs:** https://arxiv.org/abs/2512.19903 · **PDF:** https://arxiv.org/pdf/2512.19903v1
+
+## 26105. Detecting cyberbullying in Spanish texts through deep learning techniques
+
+- **ID:** `2512.19899` · **primary:** `cs.HC` · **Submitted:** 2025-12-22
+- **Authors:** Paúl Cumba-Armijos, Diego Riofrío-Luzcando, Verónica Rodríguez-Arboleda, et al.
+- **abs:** https://arxiv.org/abs/2512.19899 · **PDF:** https://arxiv.org/pdf/2512.19899v1
+
+## 26106. Efficient Learning of Lattice Gauge Theories with Fermions
+
+- **ID:** `2512.19891` · **primary:** `hep-lat` · **Submitted:** 2025-12-22
+- **Authors:** Shreya Shukla, Yukari Yamauchi, Andrey Y. Lokhov, et al.
+- **abs:** https://arxiv.org/abs/2512.19891 · **PDF:** https://arxiv.org/pdf/2512.19891v1
+
+## 26107. A Branch-and-Price Algorithm for Fast and Equitable Last-Mile Relief Aid Distribution
+
+- **ID:** `2512.19882` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Mahdi Mostajabdaveh, F. Sibel Salman, Walter J. Gutjahr
+- **abs:** https://arxiv.org/abs/2512.19882 · **PDF:** https://arxiv.org/pdf/2512.19882v1
+
+## 26108. Fine-Tuned In-Context Learners for Efficient Adaptation
+
+- **ID:** `2512.19879` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Jorg Bornschein, Clare Lyle, Yazhe Li, et al.
+- **abs:** https://arxiv.org/abs/2512.19879 · **PDF:** https://arxiv.org/pdf/2512.19879v1
+
+## 26109. Interpretable and Adaptive Node Classification on Heterophilic Graphs via Combinatorial Scoring and Hybrid Learning
+
+- **ID:** `2512.22221` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Soroush Vahidi
+- **abs:** https://arxiv.org/abs/2512.22221 · **PDF:** https://arxiv.org/pdf/2512.22221v1
+
+## 26110. Toward a Physical Theory of Intelligence
+
+- **ID:** `2601.00021` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Peter David Fagan
+- **abs:** https://arxiv.org/abs/2601.00021 · **PDF:** https://arxiv.org/pdf/2601.00021v2
+
+## 26111. HARMON-E: Hierarchical Agentic Reasoning for Multimodal Oncology Notes to Extract Structured Data
+
+- **ID:** `2512.19864` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Shashi Kant Gupta, Arijeet Pramanik, Jerrin John Thomas, et al.
+- **abs:** https://arxiv.org/abs/2512.19864 · **PDF:** https://arxiv.org/pdf/2512.19864v2
+
+## 26112. Causal Inference with the "Napkin Graph"
+
+- **ID:** `2512.19861` · **primary:** `stat.ME` · **Submitted:** 2025-12-22
+- **Authors:** Anna Guo, David Benkeser, Razieh Nabi
+- **abs:** https://arxiv.org/abs/2512.19861 · **PDF:** https://arxiv.org/pdf/2512.19861v1
+
+## 26113. On Extending Semantic Abstraction for Efficient Search of Hidden Objects
+
+- **ID:** `2512.22220` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Tasha Pais, Nikhilesh Belulkar
+- **abs:** https://arxiv.org/abs/2512.22220 · **PDF:** https://arxiv.org/pdf/2512.22220v1
+
+## 26114. Gaussian Variational Inference with Non-Gaussian Factors for State Estimation: A UWB Localization Case Study
+
+- **ID:** `2512.19855` · **primary:** `cs.RO` · **Submitted:** 2025-12-22
+- **Authors:** Andrew Stirling, Mykola Lukashchuk, Dmitry Bagaev, et al.
+- **abs:** https://arxiv.org/abs/2512.19855 · **PDF:** https://arxiv.org/pdf/2512.19855v2
+
+## 26115. UCCL-EP: Portable Expert-Parallel Communication
+
+- **ID:** `2512.19849` · **primary:** `cs.DC` · **Submitted:** 2025-12-22
+- **Authors:** Ziming Mao, Yihan Zhang, Chihan Cui, et al.
+- **abs:** https://arxiv.org/abs/2512.19849 · **PDF:** https://arxiv.org/pdf/2512.19849v2
+
+## 26116. MathLedger: A Verifiable Learning Substrate with Ledger-Attested Feedback
+
+- **ID:** `2601.00816` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Ismail Ahmad Abdullah
+- **abs:** https://arxiv.org/abs/2601.00816 · **PDF:** https://arxiv.org/pdf/2601.00816v1
+
+## 26117. PHOTON: Hierarchical Autoregressive Modeling for Lightspeed and Memory-Efficient Language Generation
+
+- **ID:** `2512.20687` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Yuma Ichikawa, Naoya Takagi, Takumi Nakagawa, et al.
+- **abs:** https://arxiv.org/abs/2512.20687 · **PDF:** https://arxiv.org/pdf/2512.20687v2
+
+## 26118. Fundamentals of quantum Boltzmann machine learning with visible and hidden units
+
+- **ID:** `2512.19819` · **primary:** `quant-ph` · **Submitted:** 2025-12-22
+- **Authors:** Mark M. Wilde
+- **abs:** https://arxiv.org/abs/2512.19819 · **PDF:** https://arxiv.org/pdf/2512.19819v1
+
+## 26119. Guardrailed Uplift Targeting: A Causal Optimization Playbook for Marketing Strategy
+
+- **ID:** `2512.19805` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Deepit Sapru
+- **abs:** https://arxiv.org/abs/2512.19805 · **PDF:** https://arxiv.org/pdf/2512.19805v2
+
+## 26120. Reduced Order Modeling for Tsunami Forecasting with Bayesian Hierarchical Pooling
+
+- **ID:** `2512.19804` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Shane X. Coffing, John Tipton, Arvind T. Mohan, et al.
+- **abs:** https://arxiv.org/abs/2512.19804 · **PDF:** https://arxiv.org/pdf/2512.19804v1
+
+## 26121. PhysMaster: Building an Autonomous AI Physicist for Theoretical and Computational Physics Research
+
+- **ID:** `2512.19799` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Tingjia Miao, Jiawen Dai, Jingkun Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.19799 · **PDF:** https://arxiv.org/pdf/2512.19799v1
+
+## 26122. Scalable Stewardship of an LLM-Assisted Clinical Benchmark with Physician Oversight
+
+- **ID:** `2512.19691` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Junze Ye, Daniel Tawfik, Alex J. Goodell, et al.
+- **abs:** https://arxiv.org/abs/2512.19691 · **PDF:** https://arxiv.org/pdf/2512.19691v2
+
+## 26123. Pushing the Frontier of Audiovisual Perception with Large-Scale Multimodal Correspondence Learning
+
+- **ID:** `2512.19687` · **primary:** `cs.SD` · **Submitted:** 2025-12-22
+- **Authors:** Apoorv Vyas, Heng-Jui Chang, Cheng-Fu Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.19687 · **PDF:** https://arxiv.org/pdf/2512.19687v1
+
+## 26124. GenEnv: Difficulty-Aligned Co-Evolution Between LLM Agents and Environment Simulators
+
+- **ID:** `2512.19682` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Jiacheng Guo, Ling Yang, Peter Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.19682 · **PDF:** https://arxiv.org/pdf/2512.19682v2
+
+## 26125. The Inconsistency Critique: Epistemic Practices and AI Testimony About Inner States
+
+- **ID:** `2601.08850` · **primary:** `cs.CY` · **Submitted:** 2025-12-22
+- **Authors:** Gerol Petruzella
+- **abs:** https://arxiv.org/abs/2601.08850 · **PDF:** https://arxiv.org/pdf/2601.08850v1
+
+## 26126. WorldWarp: Propagating 3D Geometry with Asynchronous Video Diffusion
+
+- **ID:** `2512.19678` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Hanyang Kong, Xingyi Yang, Xiaoxu Zheng, et al.
+- **abs:** https://arxiv.org/abs/2512.19678 · **PDF:** https://arxiv.org/pdf/2512.19678v1
+
+## 26127. Bottom-up Policy Optimization: Your Language Model Policy Secretly Contains Internal Policies
+
+- **ID:** `2512.19673` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Yuqiao Tan, Minzheng Wang, Shizhu He, et al.
+- **abs:** https://arxiv.org/abs/2512.19673 · **PDF:** https://arxiv.org/pdf/2512.19673v2
+
+## 26128. Beyond CLIP: Knowledge-Enhanced Multimodal Transformers for Cross-Modal Alignment in Diabetic Retinopathy Diagnosis
+
+- **ID:** `2512.19663` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Argha Kamal Samanta, Harshika Goyal, Vasudha Joshi, et al.
+- **abs:** https://arxiv.org/abs/2512.19663 · **PDF:** https://arxiv.org/pdf/2512.19663v1
+
+## 26129. Clustering with Label Consistency
+
+- **ID:** `2512.19654` · **primary:** `cs.DS` · **Submitted:** 2025-12-22
+- **Authors:** Diptarka Chakraborty, Hendrik Fichtenberger, Bernhard Haeupler, et al.
+- **abs:** https://arxiv.org/abs/2512.19654 · **PDF:** https://arxiv.org/pdf/2512.19654v1
+
+## 26130. Exploring Zero-Shot ACSA with Unified Meaning Representation in Chain-of-Thought Prompting
+
+- **ID:** `2512.19651` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Filippos Ventirozos, Peter Appleby, Matthew Shardlow
+- **abs:** https://arxiv.org/abs/2512.19651 · **PDF:** https://arxiv.org/pdf/2512.19651v1
+
+## 26131. Deep Legendre Transform
+
+- **ID:** `2512.19649` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Aleksey Minabutdinov, Patrick Cheridito
+- **abs:** https://arxiv.org/abs/2512.19649 · **PDF:** https://arxiv.org/pdf/2512.19649v2
+
+## 26132. The Best of Both Worlds: Hybridizing Neural Operators and Solvers for Stable Long-Horizon Inference
+
+- **ID:** `2512.19643` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Rajyasri Roy, Dibyajyoti Nayak, Somdatta Goswami
+- **abs:** https://arxiv.org/abs/2512.19643 · **PDF:** https://arxiv.org/pdf/2512.19643v1
+
+## 26133. Psychometric Comparability of LLM-Based Digital Twins
+
+- **ID:** `2601.14264` · **primary:** `cs.CY` · **Submitted:** 2025-12-22
+- **Authors:** Yufei Zhang, Zhihao Ma
+- **abs:** https://arxiv.org/abs/2601.14264 · **PDF:** https://arxiv.org/pdf/2601.14264v1
+
+## 26134. Diacritic Restoration for Low-Resource Indigenous Languages: Case Study with Bribri and Cook Islands Māori
+
+- **ID:** `2512.19630` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Rolando Coto-Solano, Daisy Li, Manoela Teleginski Ferraz, et al.
+- **abs:** https://arxiv.org/abs/2512.19630 · **PDF:** https://arxiv.org/pdf/2512.19630v1
+
+## 26135. Exploring the features used for summary evaluation by Human and GPT
+
+- **ID:** `2512.19620` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Zahra Sadeghi, Evangelos Milios, Frank Rudzicz
+- **abs:** https://arxiv.org/abs/2512.19620 · **PDF:** https://arxiv.org/pdf/2512.19620v1
+
+## 26136. MauBERT: Universal Phonetic Inductive Biases for Few-Shot Acoustic Units Discovery
+
+- **ID:** `2512.19612` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Angelo Ortiz Tandazo, Manel Khentout, Youssef Benchekroun, et al.
+- **abs:** https://arxiv.org/abs/2512.19612 · **PDF:** https://arxiv.org/pdf/2512.19612v1
+
+## 26137. MapTrace: Scalable Data Generation for Route Tracing on Maps
+
+- **ID:** `2512.19609` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Artemis Panagopoulou, Aveek Purohit, Achin Kulshrestha, et al.
+- **abs:** https://arxiv.org/abs/2512.19609 · **PDF:** https://arxiv.org/pdf/2512.19609v1
+
+## 26138. KerJEPA: Kernel Discrepancies for Euclidean Self-Supervised Learning
+
+- **ID:** `2512.19605` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Eric Zimmermann, Harley Wiltzer, Justin Szeto, et al.
+- **abs:** https://arxiv.org/abs/2512.19605 · **PDF:** https://arxiv.org/pdf/2512.19605v1
+
+## 26139. Gaming the Answer Matcher: Examining the Impact of Text Manipulation on Automated Judgment
+
+- **ID:** `2601.08849` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Manas Khatore, Sumana Sridharan, Kevork Sulahian, et al.
+- **abs:** https://arxiv.org/abs/2601.08849 · **PDF:** https://arxiv.org/pdf/2601.08849v1
+
+## 26140. Increasing the Thinking Budget is Not All You Need
+
+- **ID:** `2512.19585` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Ignacio Iacobacci, Zhaozhi Qian, Faroq AL-Tam, et al.
+- **abs:** https://arxiv.org/abs/2512.19585 · **PDF:** https://arxiv.org/pdf/2512.19585v1
+
+## 26141. Deep Learning for Primordial $B$-mode Extraction
+
+- **ID:** `2512.19577` · **primary:** `astro-ph.CO` · **Submitted:** 2025-12-22
+- **Authors:** Eric Guzman, Joel Meyers
+- **abs:** https://arxiv.org/abs/2512.19577 · **PDF:** https://arxiv.org/pdf/2512.19577v1
+
+## 26142. LeLaR: The First In-Orbit Demonstration of an AI-Based Satellite Attitude Controller
+
+- **ID:** `2512.19576` · **primary:** `cs.RO` · **Submitted:** 2025-12-22
+- **Authors:** Kirill Djebko, Tom Baumann, Erik Dilger, et al.
+- **abs:** https://arxiv.org/abs/2512.19576 · **PDF:** https://arxiv.org/pdf/2512.19576v4
+
+## 26143. The Epistemological Consequences of Large Language Models: Rethinking collective intelligence and institutional knowledge
+
+- **ID:** `2512.19570` · **primary:** `cs.HC` · **Submitted:** 2025-12-22
+- **Authors:** Angjelin Hila
+- **abs:** https://arxiv.org/abs/2512.19570 · **PDF:** https://arxiv.org/pdf/2512.19570v2
+
+## 26144. Owning the Intelligence: Global AI Patents Landscape and Europe's Quest for Technological Sovereignty
+
+- **ID:** `2512.19569` · **primary:** `econ.GN` · **Submitted:** 2025-12-22
+- **Authors:** Lapo Santarlasci, Armando Rungi, Loredana Fattorini, et al.
+- **abs:** https://arxiv.org/abs/2512.19569 · **PDF:** https://arxiv.org/pdf/2512.19569v1
+
+## 26145. Results of the 2024 CommonRoad Motion Planning Competition for Autonomous Vehicles
+
+- **ID:** `2512.19564` · **primary:** `cs.RO` · **Submitted:** 2025-12-22
+- **Authors:** Yanliang Huang, Xia Yan, Peiran Yin, et al.
+- **abs:** https://arxiv.org/abs/2512.19564 · **PDF:** https://arxiv.org/pdf/2512.19564v1
+
+## 26146. REALM: A Real-to-Sim Validated Benchmark for Generalization in Robotic Manipulation
+
+- **ID:** `2512.19562` · **primary:** `cs.RO` · **Submitted:** 2025-12-22
+- **Authors:** Martin Sedlacek, Pavlo Yefanov, Georgy Ponimatkin, et al.
+- **abs:** https://arxiv.org/abs/2512.19562 · **PDF:** https://arxiv.org/pdf/2512.19562v1
+
+## 26147. BabyFlow: 3D modeling of realistic and expressive infant faces
+
+- **ID:** `2512.19560` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Antonia Alomar, Mireia Masias, Marius George Linguraru, et al.
+- **abs:** https://arxiv.org/abs/2512.19560 · **PDF:** https://arxiv.org/pdf/2512.19560v1
+
+## 26148. Augmenting Intelligence: A Hybrid Framework for Scalable and Stable Explanations
+
+- **ID:** `2512.19557` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Lawrence Krukrubo, Julius Odede, Olawande Olusegun
+- **abs:** https://arxiv.org/abs/2512.19557 · **PDF:** https://arxiv.org/pdf/2512.19557v1
+
+## 26149. CARE What Fails: Contrastive Anchored-REflection for Verifiable Multimodal Reasoning
+
+- **ID:** `2512.19554` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Yongxin Wang, Zhicheng Yang, Meng Cao, et al.
+- **abs:** https://arxiv.org/abs/2512.19554 · **PDF:** https://arxiv.org/pdf/2512.19554v4
+
+## 26150. Towards Closed-Loop Embodied Empathy Evolution: Probing LLM-Centric Lifelong Empathic Motion Generation in Unseen Scenarios
+
+- **ID:** `2512.19551` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Jiawen Wang, Jingjing Wang Tianyang Chen, Min Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.19551 · **PDF:** https://arxiv.org/pdf/2512.19551v1
+
+## 26151. DFORD: Directional Feedback based Online Ordinal Regression Learning
+
+- **ID:** `2512.19550` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Naresh Manwani, M Elamparithy, Tanish Taneja
+- **abs:** https://arxiv.org/abs/2512.19550 · **PDF:** https://arxiv.org/pdf/2512.19550v1
+
+## 26152. Algerian Dialect
+
+- **ID:** `2512.19543` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Zakaria Benmounah, Abdennour Boulesnane
+- **abs:** https://arxiv.org/abs/2512.19543 · **PDF:** https://arxiv.org/pdf/2512.19543v1
+
+## 26153. Active Convolved Illumination with Deep Transfer Learning for Complex Beam Transmission through Atmospheric Turbulence
+
+- **ID:** `2512.19540` · **primary:** `physics.optics` · **Submitted:** 2025-12-22
+- **Authors:** Adrian A. Moazzam, Anindya Ghoshroy, Breeanne Heusdens, et al.
+- **abs:** https://arxiv.org/abs/2512.19540 · **PDF:** https://arxiv.org/pdf/2512.19540v1
+
+## 26154. Event Extraction in Large Language Model
+
+- **ID:** `2512.19537` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Bobo Li, Xudong Han, Jiang Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.19537 · **PDF:** https://arxiv.org/pdf/2512.19537v1
+
+## 26155. CASA: Cross-Attention over Self-Attention for Efficient Vision-Language Fusion
+
+- **ID:** `2512.19535` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Moritz Böhle, Amélie Royer, Juliette Marrie, et al.
+- **abs:** https://arxiv.org/abs/2512.19535 · **PDF:** https://arxiv.org/pdf/2512.19535v2
+
+## 26156. Learning Continuous Solvent Effects from Transient Flow Data: A Graph Neural Network Benchmark on Catechol Rearrangement
+
+- **ID:** `2512.19530` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Hongsheng Xing, Qiuxin Si
+- **abs:** https://arxiv.org/abs/2512.19530 · **PDF:** https://arxiv.org/pdf/2512.19530v1
+
+## 26157. Deep Learning for Unrelated-Machines Scheduling: Handling Variable Dimensions
+
+- **ID:** `2512.19527` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Diego Hitzges, Guillaume Sagnol
+- **abs:** https://arxiv.org/abs/2512.19527 · **PDF:** https://arxiv.org/pdf/2512.19527v1
+
+## 26158. QuantiPhy: A Quantitative Benchmark Evaluating Physical Reasoning Abilities of Vision-Language Models
+
+- **ID:** `2512.19526` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Li Puyin, Tiange Xiang, Ella Mao, et al.
+- **abs:** https://arxiv.org/abs/2512.19526 · **PDF:** https://arxiv.org/pdf/2512.19526v1
+
+## 26159. Initialization of a Polyharmonic Cascade, Launch and Testing
+
+- **ID:** `2512.19524` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Yuriy N. Bakhvalov
+- **abs:** https://arxiv.org/abs/2512.19524 · **PDF:** https://arxiv.org/pdf/2512.19524v1
+
+## 26160. LacaDM: A Latent Causal Diffusion Model for Multiobjective Reinforcement Learning
+
+- **ID:** `2512.19516` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Xueming Yan, Bo Yin, Yaochu Jin
+- **abs:** https://arxiv.org/abs/2512.19516 · **PDF:** https://arxiv.org/pdf/2512.19516v1
+
+## 26161. Anatomy-R1: Enhancing Anatomy Reasoning in Multimodal Large Language Models via Anatomical Similarity Curriculum and Group Diversity Augmentation
+
+- **ID:** `2512.19512` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Ziyang Song, Zelin Zang, Zuyao Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.19512 · **PDF:** https://arxiv.org/pdf/2512.19512v2
+
+## 26162. Toward Scalable and Valid Conditional Independence Testing with Spectral Representations
+
+- **ID:** `2512.19510` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Alek Frohlich, Vladimir Kostic, Karim Lounici, et al.
+- **abs:** https://arxiv.org/abs/2512.19510 · **PDF:** https://arxiv.org/pdf/2512.19510v1
+
+## 26163. DK-STN: A Domain Knowledge Embedded Spatio-Temporal Network Model for MJO Forecast
+
+- **ID:** `2512.19506` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Hongliang Li, Nong Zhang, Zhewen Xu, et al.
+- **abs:** https://arxiv.org/abs/2512.19506 · **PDF:** https://arxiv.org/pdf/2512.19506v1
+
+## 26164. Kolmogorov-Arnold graph neural networks for chemically informed prediction tasks on inorganic nanomaterials
+
+- **ID:** `2512.19494` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Nikita Volzhin, Soowhan Yoon
+- **abs:** https://arxiv.org/abs/2512.19494 · **PDF:** https://arxiv.org/pdf/2512.19494v2
+
+## 26165. Learning from sanctioned government suppliers: A machine learning and network science approach to detecting fraud and corruption in Mexico
+
+- **ID:** `2512.19491` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Martí Medina-Hernández, Janos Kertész, Mihály Fazekas
+- **abs:** https://arxiv.org/abs/2512.19491 · **PDF:** https://arxiv.org/pdf/2512.19491v2
+
+## 26166. Lightweight Intrusion Detection in IoT via SHAP-Guided Feature Pruning and Knowledge-Distilled Kronecker Networks
+
+- **ID:** `2512.19488` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Hafsa Benaddi, Mohammed Jouhari, Nouha Laamech, et al.
+- **abs:** https://arxiv.org/abs/2512.19488 · **PDF:** https://arxiv.org/pdf/2512.19488v1
+
+## 26167. Structured Event Representation and Stock Return Predictability
+
+- **ID:** `2512.19484` · **primary:** `econ.GN` · **Submitted:** 2025-12-22
+- **Authors:** Gang Li, Dandan Qiao, Mingxuan Zheng
+- **abs:** https://arxiv.org/abs/2512.19484 · **PDF:** https://arxiv.org/pdf/2512.19484v1
+
+## 26168. A Dataset and Preliminary Study of Using GPT-5 for Code-change Impact Analysis
+
+- **ID:** `2512.19481` · **primary:** `cs.SE` · **Submitted:** 2025-12-22
+- **Authors:** Katharina Stengg, Christian Macho, Martin Pinzger
+- **abs:** https://arxiv.org/abs/2512.19481 · **PDF:** https://arxiv.org/pdf/2512.19481v1
+
+## 26169. A Large-Language-Model Framework for Automated Humanitarian Situation Reporting
+
+- **ID:** `2512.19475` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Ivan Decostanzi, Yelena Mejova, Kyriaki Kalimeri
+- **abs:** https://arxiv.org/abs/2512.19475 · **PDF:** https://arxiv.org/pdf/2512.19475v1
+
+## 26170. Multi-Layer Confidence Scoring for Detection of Out-of-Distribution Samples, Adversarial Attacks, and In-Distribution Misclassifications
+
+- **ID:** `2512.19472` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Lorenzo Capelli, Leandro de Souza Rosa, Gianluca Setti, et al.
+- **abs:** https://arxiv.org/abs/2512.19472 · **PDF:** https://arxiv.org/pdf/2512.19472v1
+
+## 26171. GLUE: Generative Latent Unification of Expertise-Informed Engineering Models
+
+- **ID:** `2512.19469` · **primary:** `cs.CE` · **Submitted:** 2025-12-22
+- **Authors:** Tim Aebersold, Soheyl Massoudi, Mark D. Fuge
+- **abs:** https://arxiv.org/abs/2512.19469 · **PDF:** https://arxiv.org/pdf/2512.19469v1
+
+## 26172. Epistemological Fault Lines Between Human and Artificial Intelligence
+
+- **ID:** `2512.19466` · **primary:** `cs.CY` · **Submitted:** 2025-12-22
+- **Authors:** Walter Quattrociocchi, Valerio Capraro, Matjaž Perc
+- **abs:** https://arxiv.org/abs/2512.19466 · **PDF:** https://arxiv.org/pdf/2512.19466v1
+
+## 26173. Diffusion Models in Simulation-Based Inference: A Tutorial Review
+
+- **ID:** `2512.20685` · **primary:** `stat.ML` · **Submitted:** 2025-12-22
+- **Authors:** Jonas Arruda, Niels Bracher, Ullrich Köthe, et al.
+- **abs:** https://arxiv.org/abs/2512.20685 · **PDF:** https://arxiv.org/pdf/2512.20685v2
+
+## 26174. An Agentic Framework for Autonomous Materials Computation
+
+- **ID:** `2512.19458` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Zeyu Xia, Jinzhe Ma, Congjie Zheng, et al.
+- **abs:** https://arxiv.org/abs/2512.19458 · **PDF:** https://arxiv.org/pdf/2512.19458v1
+
+## 26175. Learned Digital Codes for Over-the-Air Computation in Federated Edge Learning
+
+- **ID:** `2512.19777` · **primary:** `cs.IT` · **Submitted:** 2025-12-22
+- **Authors:** Antonio Tarizzo, Mohammad Kazemi, Deniz Gündüz
+- **abs:** https://arxiv.org/abs/2512.19777 · **PDF:** https://arxiv.org/pdf/2512.19777v1
+
+## 26176. Activations as Features: Probing LLMs for Generalizable Essay Scoring Representations
+
+- **ID:** `2512.19456` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Jinwei Chi, Ke Wang, Yu Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.19456 · **PDF:** https://arxiv.org/pdf/2512.19456v1
+
+## 26177. SiamGPT: Quality-First Fine-Tuning for Stable Thai Text Generation
+
+- **ID:** `2512.19455` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Thittipat Pairatsuppawat, Abhibhu Tachaapornchai, Paweekorn Kusolsomboon, et al.
+- **abs:** https://arxiv.org/abs/2512.19455 · **PDF:** https://arxiv.org/pdf/2512.19455v3
+
+## 26178. A Multi-Perspective Benchmark and Moderation Model for Evaluating Safety and Adversarial Robustness
+
+- **ID:** `2601.03273` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Naseem Machlovi, Maryam Saleki, Ruhul Amin, et al.
+- **abs:** https://arxiv.org/abs/2601.03273 · **PDF:** https://arxiv.org/pdf/2601.03273v2
+
+## 26179. Real-Time Streamable Generative Speech Restoration with Flow Matching
+
+- **ID:** `2512.19442` · **primary:** `eess.SP` · **Submitted:** 2025-12-22
+- **Authors:** Simon Welker, Bunlong Lay, Maris Hillemann, et al.
+- **abs:** https://arxiv.org/abs/2512.19442 · **PDF:** https://arxiv.org/pdf/2512.19442v2
+
+## 26180. Binary Kernel Logistic Regression: a sparsity-inducing formulation and a convergent decomposition training algorithm
+
+- **ID:** `2512.19440` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Antonio Consolo, Andrea Manno, Edoardo Amaldi
+- **abs:** https://arxiv.org/abs/2512.19440 · **PDF:** https://arxiv.org/pdf/2512.19440v1
+
+## 26181. An Inverse Scattering Inspired Fourier Neural Operator for Time-Dependent PDE Learning
+
+- **ID:** `2512.19439` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Rixin Yu
+- **abs:** https://arxiv.org/abs/2512.19439 · **PDF:** https://arxiv.org/pdf/2512.19439v1
+
+## 26182. MT-Mark: Rethinking Image Watermarking via Mutual-Teacher Collaboration with Adaptive Feature Modulation
+
+- **ID:** `2512.19438` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Fei Ge, Ying Huang, Jie Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.19438 · **PDF:** https://arxiv.org/pdf/2512.19438v1
+
+## 26183. MobileWorld: Benchmarking Autonomous Mobile Agents in Agent-User Interactive and MCP-Augmented Environments
+
+- **ID:** `2512.19432` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Quyu Kong, Xu Zhang, Zhenyu Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.19432 · **PDF:** https://arxiv.org/pdf/2512.19432v3
+
+## 26184. Attention Is Not What You Need
+
+- **ID:** `2512.19428` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Zhang Chong
+- **abs:** https://arxiv.org/abs/2512.19428 · **PDF:** https://arxiv.org/pdf/2512.19428v1
+
+## 26185. CodeSimpleQA: Scaling Factuality in Code Large Language Models
+
+- **ID:** `2512.19424` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Jian Yang, Wei Zhang, Yizhi Li, et al.
+- **abs:** https://arxiv.org/abs/2512.19424 · **PDF:** https://arxiv.org/pdf/2512.19424v1
+
+## 26186. Mirage Persistent Kernel: A Compiler and Runtime for Mega-Kernelizing Tensor Programs
+
+- **ID:** `2512.22219` · **primary:** `cs.DC` · **Submitted:** 2025-12-22
+- **Authors:** Xinhao Cheng, Zhihao Zhang, Yu Zhou, et al.
+- **abs:** https://arxiv.org/abs/2512.22219 · **PDF:** https://arxiv.org/pdf/2512.22219v1
+
+## 26187. From Retrieval to Reasoning: A Framework for Cyber Threat Intelligence NER with Explicit and Adaptive Instructions
+
+- **ID:** `2512.19414` · **primary:** `cs.CR` · **Submitted:** 2025-12-22
+- **Authors:** Jiaren Peng, Hongda Sun, Xuan Tian, et al.
+- **abs:** https://arxiv.org/abs/2512.19414 · **PDF:** https://arxiv.org/pdf/2512.19414v1
+
+## 26188. Research Program: Theory of Learning in Dynamical Systems
+
+- **ID:** `2512.19410` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Elad Hazan, Shai Shalev Shwartz, Nathan Srebro
+- **abs:** https://arxiv.org/abs/2512.19410 · **PDF:** https://arxiv.org/pdf/2512.19410v1
+
+## 26189. Symplectic Reservoir Representation of Legendre Dynamics
+
+- **ID:** `2512.19409` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Robert Simon Fong, Gouhei Tanaka, Kazuyuki Aihara
+- **abs:** https://arxiv.org/abs/2512.19409 · **PDF:** https://arxiv.org/pdf/2512.19409v1
+
+## 26190. Kunnafonidilaw ka Cadeau: an ASR dataset of present-day Bambara
+
+- **ID:** `2512.19400` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Yacouba Diarra, Panga Azazia Kamate, Nouhoum Souleymane Coulibaly, et al.
+- **abs:** https://arxiv.org/abs/2512.19400 · **PDF:** https://arxiv.org/pdf/2512.19400v1
+
+## 26191. Brain-Grounded Axes for Reading and Steering LLM States
+
+- **ID:** `2512.19399` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Sandro Andric
+- **abs:** https://arxiv.org/abs/2512.19399 · **PDF:** https://arxiv.org/pdf/2512.19399v1
+
+## 26192. EchoTrail-GUI: Building Actionable Memory for GUI Agents via Critic-Guided Self-Exploration
+
+- **ID:** `2512.19396` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Runze Li, Yuwen Zhai, Bo Xu, et al.
+- **abs:** https://arxiv.org/abs/2512.19396 · **PDF:** https://arxiv.org/pdf/2512.19396v1
+
+## 26193. DSTED: Decoupling Temporal Stabilization and Discriminative Enhancement for Surgical Workflow Recognition
+
+- **ID:** `2512.19387` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Yueyao Chen, Kai-Ni Wang, Dario Tayupo, et al.
+- **abs:** https://arxiv.org/abs/2512.19387 · **PDF:** https://arxiv.org/pdf/2512.19387v1
+
+## 26194. PediaMind-R1: A Temperament-Aware Language Model for Personalized Early Childhood Care Reasoning via Cognitive Modeling and Preference Alignment
+
+- **ID:** `2601.08848` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Zihe Zhang, Can Zhang, Yanheng Xu, et al.
+- **abs:** https://arxiv.org/abs/2601.08848 · **PDF:** https://arxiv.org/pdf/2601.08848v1
+
+## 26195. Real-Time Machine Learning for Embedded Anomaly Detection
+
+- **ID:** `2512.19383` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Abdelmadjid Benmachiche, Khadija Rais, Hamda Slimi
+- **abs:** https://arxiv.org/abs/2512.19383 · **PDF:** https://arxiv.org/pdf/2512.19383v1
+
+## 26196. OmniMER: Auxiliary-Enhanced LLM Adaptation for Indonesian Multimodal Emotion Recognition
+
+- **ID:** `2512.19379` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Xueming Yan, Boyan Xu, Yaochu Jin, et al.
+- **abs:** https://arxiv.org/abs/2512.19379 · **PDF:** https://arxiv.org/pdf/2512.19379v3
+
+## 26197. HATS: High-Accuracy Triple-Set Watermarking for Large Language Models
+
+- **ID:** `2512.19378` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Zhiqing Hu, Chenxu Zhao, Jiazhong Lu, et al.
+- **abs:** https://arxiv.org/abs/2512.19378 · **PDF:** https://arxiv.org/pdf/2512.19378v1
+
+## 26198. A Critical Assessment of Pattern Comparisons Between POD and Autoencoders in Intraventricular Flows
+
+- **ID:** `2512.19376` · **primary:** `physics.flu-dyn` · **Submitted:** 2025-12-22
+- **Authors:** Eneko Lazpita, Andrés Bell-Navas, Jesús Garicano-Mena, et al.
+- **abs:** https://arxiv.org/abs/2512.19376 · **PDF:** https://arxiv.org/pdf/2512.19376v1
+
+## 26199. Cluster-Based Generalized Additive Models Informed by Random Fourier Features
+
+- **ID:** `2512.19373` · **primary:** `stat.ML` · **Submitted:** 2025-12-22
+- **Authors:** Xin Huang, Jia Li, Jun Yu
+- **abs:** https://arxiv.org/abs/2512.19373 · **PDF:** https://arxiv.org/pdf/2512.19373v2
+
+## 26200. Sprecher Networks: A Parameter-Efficient Kolmogorov-Arnold Architecture
+
+- **ID:** `2512.19367` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Christian Hägg, Kathlén Kohn, Giovanni Luca Marchetti, et al.
+- **abs:** https://arxiv.org/abs/2512.19367 · **PDF:** https://arxiv.org/pdf/2512.19367v2
+
+## 26201. Learning General Policies with Policy Gradient Methods
+
+- **ID:** `2512.19366` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Simon Ståhlberg, Blai Bonet, Hector Geffner
+- **abs:** https://arxiv.org/abs/2512.19366 · **PDF:** https://arxiv.org/pdf/2512.19366v1
+
+## 26202. From Points to Coalitions: Hierarchical Contrastive Shapley Values for Prioritizing Data Samples
+
+- **ID:** `2512.19363` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Canran Xiao, Jiabao Dou, Zhiming Lin, et al.
+- **abs:** https://arxiv.org/abs/2512.19363 · **PDF:** https://arxiv.org/pdf/2512.19363v1
+
+## 26203. Interpretable Hybrid Deep Q-Learning Framework for IoT-Based Food Spoilage Prediction with Synthetic Data Generation and Hardware Validation
+
+- **ID:** `2512.19361` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Isshaan Singh, Divyansh Chawla, Anshu Garg, et al.
+- **abs:** https://arxiv.org/abs/2512.19361 · **PDF:** https://arxiv.org/pdf/2512.19361v1
+
+## 26204. First-Order Representation Languages for Goal-Conditioned RL
+
+- **ID:** `2512.19355` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Simon Ståhlberg, Hector Geffner
+- **abs:** https://arxiv.org/abs/2512.19355 · **PDF:** https://arxiv.org/pdf/2512.19355v1
+
+## 26205. PENDULUM: A Benchmark for Assessing Sycophancy in Multimodal Large Language Models
+
+- **ID:** `2512.19350` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** A. B. M. Ashikur Rahman, Saeed Anwar, Muhammad Usman, et al.
+- **abs:** https://arxiv.org/abs/2512.19350 · **PDF:** https://arxiv.org/pdf/2512.19350v1
+
+## 26206. VIGOR+: Iterative Confounder Generation and Validation via LLM-CEVAE Feedback Loop
+
+- **ID:** `2512.19349` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** JiaWei Zhu, ZiHeng Liu
+- **abs:** https://arxiv.org/abs/2512.19349 · **PDF:** https://arxiv.org/pdf/2512.19349v1
+
+## 26207. Faster Distributed Inference-Only Recommender Systems via Bounded Lag Synchronous Collectives
+
+- **ID:** `2512.19342` · **primary:** `cs.DC` · **Submitted:** 2025-12-22
+- **Authors:** Kiril Dichev, Filip Pawlowski, Albert-Jan Yzelman
+- **abs:** https://arxiv.org/abs/2512.19342 · **PDF:** https://arxiv.org/pdf/2512.19342v1
+
+## 26208. Orthogonal Approximate Message Passing with Optimal Spectral Initializations for Rectangular Spiked Matrix Models
+
+- **ID:** `2512.19334` · **primary:** `cs.IT` · **Submitted:** 2025-12-22
+- **Authors:** Haohua Chen, Songbin Liu, Junjie Ma
+- **abs:** https://arxiv.org/abs/2512.19334 · **PDF:** https://arxiv.org/pdf/2512.19334v1
+
+## 26209. A Logical View of GNN-Style Computation and the Role of Activation Functions
+
+- **ID:** `2512.19332` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Pablo Barceló, Floris Geerts, Matthias Lanzinger, et al.
+- **abs:** https://arxiv.org/abs/2512.19332 · **PDF:** https://arxiv.org/pdf/2512.19332v1
+
+## 26210. Alternative positional encoding functions for neural transformers
+
+- **ID:** `2512.19323` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Ezequiel Lopez-Rubio, Macoris Decena-Gimenez, Rafael Marcos Luque-Baena
+- **abs:** https://arxiv.org/abs/2512.19323 · **PDF:** https://arxiv.org/pdf/2512.19323v1
+
+## 26211. Learning-Assisted Multi-Operator Variable Neighborhood Search for Urban Cable Routing
+
+- **ID:** `2512.19321` · **primary:** `cs.NE` · **Submitted:** 2025-12-22
+- **Authors:** Wei Liu, Tao Zhang, Chenhui Lin, et al.
+- **abs:** https://arxiv.org/abs/2512.19321 · **PDF:** https://arxiv.org/pdf/2512.19321v1
+
+## 26212. MAGIC: Achieving Superior Model Merging via Magnitude Calibration
+
+- **ID:** `2512.19320` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Yayuan Li, Jian Zhang, Jintao Guo, et al.
+- **abs:** https://arxiv.org/abs/2512.19320 · **PDF:** https://arxiv.org/pdf/2512.19320v1
+
+## 26213. SafeMed-R1: Adversarial Reinforcement Learning for Generalizable and Robust Medical Reasoning in Vision-Language Models
+
+- **ID:** `2512.19317` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** A. A. Gde Yogi Pramana, Jason Ray, Anthony Jaya, et al.
+- **abs:** https://arxiv.org/abs/2512.19317 · **PDF:** https://arxiv.org/pdf/2512.19317v1
+
+## 26214. MixFlow Training: Alleviating Exposure Bias with Slowed Interpolation Mixture
+
+- **ID:** `2512.19311` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Hui Li, Jiayue Lyu, Fu-Yun Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.19311 · **PDF:** https://arxiv.org/pdf/2512.19311v1
+
+## 26215. Time-Vertex Machine Learning for Optimal Sensor Placement in Temporal Graph Signals: Applications in Structural Health Monitoring
+
+- **ID:** `2512.19309` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Keivan Faghih Niresi, Jun Qing, Mengjie Zhao, et al.
+- **abs:** https://arxiv.org/abs/2512.19309 · **PDF:** https://arxiv.org/pdf/2512.19309v1
+
+## 26216. Scalable and Reliable Evaluation of AI Knowledge Retrieval Systems: RIKER and the Coherent Simulated Universe
+
+- **ID:** `2601.08847` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** JV Roig
+- **abs:** https://arxiv.org/abs/2601.08847 · **PDF:** https://arxiv.org/pdf/2601.08847v2
+
+## 26217. CienaLLM: Generative Climate-Impact Extraction from News Articles with Autoregressive LLMs
+
+- **ID:** `2512.19305` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Javier Vela-Tambo, Jorge Gracia, Fernando Dominguez-Castro
+- **abs:** https://arxiv.org/abs/2512.19305 · **PDF:** https://arxiv.org/pdf/2512.19305v1
+
+## 26218. Helios: A Foundational Language Model for Smart Energy Knowledge Reasoning and Application
+
+- **ID:** `2512.19299` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Haoyu Jiang, Fanjie Zeng, Boan Qu, et al.
+- **abs:** https://arxiv.org/abs/2512.19299 · **PDF:** https://arxiv.org/pdf/2512.19299v2
+
+## 26219. Causal-Guided Detoxify Backdoor Attack of Open-Weight LoRA Models
+
+- **ID:** `2512.19297` · **primary:** `cs.CR` · **Submitted:** 2025-12-22
+- **Authors:** Linzhi Chen, Yang Sun, Hongru Wei, et al.
+- **abs:** https://arxiv.org/abs/2512.19297 · **PDF:** https://arxiv.org/pdf/2512.19297v1
+
+## 26220. Vibe Reasoning: Eliciting Frontier AI Mathematical Capabilities -- A Case Study on IMO 2025 Problem 6
+
+- **ID:** `2512.19287` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Jiaao Wu, Xian Zhang, Fan Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.19287 · **PDF:** https://arxiv.org/pdf/2512.19287v1
+
+## 26221. GShield: Mitigating Poisoning Attacks in Federated Learning
+
+- **ID:** `2512.19286` · **primary:** `cs.CR` · **Submitted:** 2025-12-22
+- **Authors:** Sameera K. M., Serena Nicolazzo, Antonino Nocera, et al.
+- **abs:** https://arxiv.org/abs/2512.19286 · **PDF:** https://arxiv.org/pdf/2512.19286v2
+
+## 26222. Digital Twin-Driven Zero-Shot Fault Diagnosis of Axial Piston Pumps Using Fluid-Borne Noise Signals
+
+- **ID:** `2512.19280` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Chang Dong, Jianfeng Tao, Chengliang Liu
+- **abs:** https://arxiv.org/abs/2512.19280 · **PDF:** https://arxiv.org/pdf/2512.19280v1
+
+## 26223. Is Visual Realism Enough? Evaluating Gait Biometric Fidelity in Generative AI Human Animation
+
+- **ID:** `2512.19275` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Ivan DeAndres-Tame, Chengwei Ye, Ruben Tolosana, et al.
+- **abs:** https://arxiv.org/abs/2512.19275 · **PDF:** https://arxiv.org/pdf/2512.19275v1
+
+## 26224. VLM-PAR: A Vision Language Model for Pedestrian Attribute Recognition
+
+- **ID:** `2512.22217` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Abdellah Zakaria Sellam, Salah Eddine Bekhouche, Fadi Dornaika, et al.
+- **abs:** https://arxiv.org/abs/2512.22217 · **PDF:** https://arxiv.org/pdf/2512.22217v1
+
+## 26225. Translating Flow to Policy via Hindsight Online Imitation
+
+- **ID:** `2512.19269` · **primary:** `cs.RO` · **Submitted:** 2025-12-22
+- **Authors:** Yitian Zheng, Zhangchen Ye, Weijun Dong, et al.
+- **abs:** https://arxiv.org/abs/2512.19269 · **PDF:** https://arxiv.org/pdf/2512.19269v2
+
+## 26226. Methodological Variation in Studying Staff and Student Perceptions of AI
+
+- **ID:** `2602.11158` · **primary:** `cs.HC` · **Submitted:** 2025-12-22
+- **Authors:** Juliana Gerard, Morgan Macleod, Kelly Norwood, et al.
+- **abs:** https://arxiv.org/abs/2602.11158 · **PDF:** https://arxiv.org/pdf/2602.11158v1
+
+## 26227. Semantic Alignment of Multilingual Knowledge Graphs via Contextualized Vector Projections
+
+- **ID:** `2601.00814` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Abhishek Kumar
+- **abs:** https://arxiv.org/abs/2601.00814 · **PDF:** https://arxiv.org/pdf/2601.00814v2
+
+## 26228. Machine Unlearning in the Era of Quantum Machine Learning: An Empirical Study
+
+- **ID:** `2512.19253` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Carla Crivoi, Radu Tudor Ionescu
+- **abs:** https://arxiv.org/abs/2512.19253 · **PDF:** https://arxiv.org/pdf/2512.19253v2
+
+## 26229. Small Language Models as Compiler Experts: Auto-Parallelization for Heterogeneous Systems
+
+- **ID:** `2512.19250` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Prathamesh Devadiga
+- **abs:** https://arxiv.org/abs/2512.19250 · **PDF:** https://arxiv.org/pdf/2512.19250v1
+
+## 26230. Auto-Prompting with Retrieval Guidance for Frame Detection in Logistics
+
+- **ID:** `2512.19247` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Do Minh Duc, Quan Xuan Truong, Nguyen Tat Dat, et al.
+- **abs:** https://arxiv.org/abs/2512.19247 · **PDF:** https://arxiv.org/pdf/2512.19247v1
+
+## 26231. From Black-Box Tuning to Guided Optimization via Hyperparameters Interaction Analysis
+
+- **ID:** `2512.19246` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Moncef Garouani, Ayah Barhrhouj
+- **abs:** https://arxiv.org/abs/2512.19246 · **PDF:** https://arxiv.org/pdf/2512.19246v1
+
+## 26232. ChemATP: A Training-Free Chemical Reasoning Framework for Large Language Models
+
+- **ID:** `2512.19240` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Mingxu Zhang, Dazhong Shen, Qi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.19240 · **PDF:** https://arxiv.org/pdf/2512.19240v1
+
+## 26233. Identifying Features Associated with Bias Against 93 Stigmatized Groups in Language Models and Guardrail Model Safety Mitigation
+
+- **ID:** `2512.19238` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Anna-Maria Gueorguieva, Aylin Caliskan
+- **abs:** https://arxiv.org/abs/2512.19238 · **PDF:** https://arxiv.org/pdf/2512.19238v1
+
+## 26234. Fast and Exact Least Absolute Deviations Line Fitting via Piecewise Affine Lower-Bounding
+
+- **ID:** `2512.20682` · **primary:** `stat.ML` · **Submitted:** 2025-12-22
+- **Authors:** Stefan Volz, Martin Storath, Andreas Weinmann
+- **abs:** https://arxiv.org/abs/2512.20682 · **PDF:** https://arxiv.org/pdf/2512.20682v1
+
+## 26235. DeliveryBench: Can Agents Earn Profit in Real World?
+
+- **ID:** `2512.19234` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Lingjun Mao, Jiawei Ren, Kun Zhou, et al.
+- **abs:** https://arxiv.org/abs/2512.19234 · **PDF:** https://arxiv.org/pdf/2512.19234v1
+
+## 26236. Regression generation adversarial network based on dual data evaluation strategy for industrial application
+
+- **ID:** `2512.19232` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Zesen Wang, Yonggang Li, Lijuan Lan
+- **abs:** https://arxiv.org/abs/2512.19232 · **PDF:** https://arxiv.org/pdf/2512.19232v1
+
+## 26237. Generation of Programmatic Rules for Document Forgery Detection Using Large Language Models
+
+- **ID:** `2512.19228` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Valentin Schmidberger, Manuel Eberhardinger, Setareh Maghsudi, et al.
+- **abs:** https://arxiv.org/abs/2512.19228 · **PDF:** https://arxiv.org/pdf/2512.19228v1
+
+## 26238. Phase-space entropy at acquisition reflects downstream learnability
+
+- **ID:** `2512.19223` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Xiu-Cheng Wang, Jun-Jie Zhanga, Nan Cheng, et al.
+- **abs:** https://arxiv.org/abs/2512.19223 · **PDF:** https://arxiv.org/pdf/2512.19223v2
+
+## 26239. Data Work in Egypt: Who Are the Workers Behind Artificial Intelligence?
+
+- **ID:** `2601.06057` · **primary:** `cs.CY` · **Submitted:** 2025-12-22
+- **Authors:** Myriam Raymond, Lucy Neveux, Antonio A. Casilli, et al.
+- **abs:** https://arxiv.org/abs/2601.06057 · **PDF:** https://arxiv.org/pdf/2601.06057v2
+
+## 26240. From Pixels to Predicates Structuring urban perception with scene graphs
+
+- **ID:** `2512.19221` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Yunlong Liu, Shuyang Li, Pengyuan Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.19221 · **PDF:** https://arxiv.org/pdf/2512.19221v1
+
+## 26241. Towards Minimal Fine-Tuning of VLMs
+
+- **ID:** `2512.19219` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Tiange Luo, Lajanugen Logeswaran, Jaekyeom Kim, et al.
+- **abs:** https://arxiv.org/abs/2512.19219 · **PDF:** https://arxiv.org/pdf/2512.19219v1
+
+## 26242. Observer, Not Player: Simulating Theory of Mind in LLMs through Game Observation
+
+- **ID:** `2512.19210` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Jerry Wang, Ting Yiu Liu
+- **abs:** https://arxiv.org/abs/2512.19210 · **PDF:** https://arxiv.org/pdf/2512.19210v1
+
+## 26243. MixKVQ: Query-Aware Mixed-Precision KV Cache Quantization for Long-Context Reasoning
+
+- **ID:** `2512.19206` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Tao Zhang, Ziqian Zeng, Hao Peng, et al.
+- **abs:** https://arxiv.org/abs/2512.19206 · **PDF:** https://arxiv.org/pdf/2512.19206v1
+
+## 26244. On the Koopman-Based Generalization Bounds for Multi-Task Deep Learning
+
+- **ID:** `2512.19199` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Mahdi Mohammadigohari, Giuseppe Di Fatta, Giuseppe Nicosia, et al.
+- **abs:** https://arxiv.org/abs/2512.19199 · **PDF:** https://arxiv.org/pdf/2512.19199v1
+
+## 26245. Adaptive Probability Flow Residual Minimization for High-Dimensional Fokker-Planck Equations
+
+- **ID:** `2512.19196` · **primary:** `physics.comp-ph` · **Submitted:** 2025-12-22
+- **Authors:** Xiaolong Wu, Qifeng Liao
+- **abs:** https://arxiv.org/abs/2512.19196 · **PDF:** https://arxiv.org/pdf/2512.19196v3
+
+## 26246. A K-Means, Ward and DBSCAN repeatability study
+
+- **ID:** `2512.19772` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Anthony Bertrand, Engelbert Mephu Nguifo, Violaine Antoine, et al.
+- **abs:** https://arxiv.org/abs/2512.19772 · **PDF:** https://arxiv.org/pdf/2512.19772v1
+
+## 26247. Causal Heterogeneous Graph Learning Method for Chronic Obstructive Pulmonary Disease Prediction
+
+- **ID:** `2512.19194` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Leming Zhou, Zuo Wang, Zhigang Liu
+- **abs:** https://arxiv.org/abs/2512.19194 · **PDF:** https://arxiv.org/pdf/2512.19194v1
+
+## 26248. PEDESTRIAN: An Egocentric Vision Dataset for Obstacle Detection on Pavements
+
+- **ID:** `2512.19190` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Marios Thoma, Zenonas Theodosiou, Harris Partaourides, et al.
+- **abs:** https://arxiv.org/abs/2512.19190 · **PDF:** https://arxiv.org/pdf/2512.19190v1
+
+## 26249. Operator-Based Generalization Bound for Deep Learning: Insights on Multi-Task Learning
+
+- **ID:** `2512.19184` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Mahdi Mohammadigohari, Giuseppe Di Fatta, Giuseppe Nicosia, et al.
+- **abs:** https://arxiv.org/abs/2512.19184 · **PDF:** https://arxiv.org/pdf/2512.19184v1
+
+## 26250. Practical Quantum-Classical Feature Fusion for complex data Classification
+
+- **ID:** `2512.19180` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Azadeh Alavi, Fatemeh Kouchmeshki, Abdolrahman Alavi
+- **abs:** https://arxiv.org/abs/2512.19180 · **PDF:** https://arxiv.org/pdf/2512.19180v1
+
+## 26251. Signal-SGN++: Topology-Enhanced Time-Frequency Spiking Graph Network for Skeleton-Based Action Recognition
+
+- **ID:** `2512.22214` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Naichuan Zheng, Xiahai Lun, Weiyi Li, et al.
+- **abs:** https://arxiv.org/abs/2512.22214 · **PDF:** https://arxiv.org/pdf/2512.22214v1
+
+## 26252. Vision-Language-Policy Model for Dynamic Robot Task Planning
+
+- **ID:** `2512.19178` · **primary:** `cs.RO` · **Submitted:** 2025-12-22
+- **Authors:** Jin Wang, Kim Tien Ly, Jacques Cloete, et al.
+- **abs:** https://arxiv.org/abs/2512.19178 · **PDF:** https://arxiv.org/pdf/2512.19178v1
+
+## 26253. CycleChart: A Unified Consistency-Based Learning Framework for Bidirectional Chart Understanding and Generation
+
+- **ID:** `2512.19173` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Dazhen Deng, Sen Yang, Yuchen He, et al.
+- **abs:** https://arxiv.org/abs/2512.19173 · **PDF:** https://arxiv.org/pdf/2512.19173v1
+
+## 26254. Finite-sample guarantees for data-driven forward-backward operator methods
+
+- **ID:** `2512.19172` · **primary:** `math.OC` · **Submitted:** 2025-12-22
+- **Authors:** Filippo Fabiani, Barbara Franci
+- **abs:** https://arxiv.org/abs/2512.19172 · **PDF:** https://arxiv.org/pdf/2512.19172v1
+
+## 26255. On the Existence and Behavior of Secondary Attention Sinks
+
+- **ID:** `2512.22213` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Jeffrey T. H. Wong, Cheng Zhang, Louis Mahon, et al.
+- **abs:** https://arxiv.org/abs/2512.22213 · **PDF:** https://arxiv.org/pdf/2512.22213v3
+
+## 26256. JEPA-Reasoner: Decoupling Latent Reasoning from Token Generation
+
+- **ID:** `2512.19171` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Bingyang Kelvin Liu, Ziyu Patrick Chen, David P. Woodruff
+- **abs:** https://arxiv.org/abs/2512.19171 · **PDF:** https://arxiv.org/pdf/2512.19171v3
+
+## 26257. From Speech to Subtitles: Evaluating ASR Models in Subtitling Italian Television Programs
+
+- **ID:** `2512.19161` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Alessandro Lucca, Francesco Pierri
+- **abs:** https://arxiv.org/abs/2512.19161 · **PDF:** https://arxiv.org/pdf/2512.19161v1
+
+## 26258. Can We Test Consciousness Theories on AI? Ablations, Markers, and Robustness
+
+- **ID:** `2512.19155` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Yin Jun Phua
+- **abs:** https://arxiv.org/abs/2512.19155 · **PDF:** https://arxiv.org/pdf/2512.19155v1
+
+## 26259. Beyond Sliding Windows: Learning to Manage Memory in Non-Markovian Environments
+
+- **ID:** `2512.19154` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Geraud Nangue Tasse, Matthew Riemer, Benjamin Rosman, et al.
+- **abs:** https://arxiv.org/abs/2512.19154 · **PDF:** https://arxiv.org/pdf/2512.19154v1
+
+## 26260. RP-CATE: Recurrent Perceptron-based Channel Attention Transformer Encoder for Industrial Hybrid Modeling
+
+- **ID:** `2512.19147` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Haoran Yang, Yinan Zhang, Wenjie Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.19147 · **PDF:** https://arxiv.org/pdf/2512.19147v1
+
+## 26261. A Convex Loss Function for Set Prediction with Optimal Trade-offs Between Size and Conditional Coverage
+
+- **ID:** `2512.19142` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Francis Bach
+- **abs:** https://arxiv.org/abs/2512.19142 · **PDF:** https://arxiv.org/pdf/2512.19142v1
+
+## 26262. Understanding Chain-of-Thought in Large Language Models via Topological Data Analysis
+
+- **ID:** `2512.19135` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Chenghao Li, Chaoning Zhang, Yi Lu, et al.
+- **abs:** https://arxiv.org/abs/2512.19135 · **PDF:** https://arxiv.org/pdf/2512.19135v2
+
+## 26263. QuCo-RAG: Quantifying Uncertainty from the Pre-training Corpus for Dynamic Retrieval-Augmented Generation
+
+- **ID:** `2512.19134` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Dehai Min, Kailin Zhang, Tongtong Wu, et al.
+- **abs:** https://arxiv.org/abs/2512.19134 · **PDF:** https://arxiv.org/pdf/2512.19134v1
+
+## 26264. Evidential Trust-Aware Model Personalization in Decentralized Federated Learning for Wearable IoT
+
+- **ID:** `2512.19131` · **primary:** `cs.DC` · **Submitted:** 2025-12-22
+- **Authors:** Murtaza Rangwala, Richard O. Sinnott, Rajkumar Buyya
+- **abs:** https://arxiv.org/abs/2512.19131 · **PDF:** https://arxiv.org/pdf/2512.19131v2
+
+## 26265. AWPO: Enhancing Tool-Use of Large Language Models through Adaptive Integration of Reasoning Rewards
+
+- **ID:** `2512.19126` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Zihan Lin, Xiaohan Wang, Hexiong Yang, et al.
+- **abs:** https://arxiv.org/abs/2512.19126 · **PDF:** https://arxiv.org/pdf/2512.19126v3
+
+## 26266. SAP: Syntactic Attention Pruning for Transformer-based Language Models
+
+- **ID:** `2512.19125` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Tzu-Yun Lee, Ding-Yong Hong, Jan-Jan Wu
+- **abs:** https://arxiv.org/abs/2512.19125 · **PDF:** https://arxiv.org/pdf/2512.19125v1
+
+## 26267. Less is more: Not all samples are effective for evaluation
+
+- **ID:** `2601.03272` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Wentang Song, Jinqiang Li, Kele Huang, et al.
+- **abs:** https://arxiv.org/abs/2601.03272 · **PDF:** https://arxiv.org/pdf/2601.03272v1
+
+## 26268. A Composable Channel-Adaptive Architecture for Seizure Classification
+
+- **ID:** `2512.19123` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Francesco Carzaniga, Michael Hersche, Kaspar Schindler, et al.
+- **abs:** https://arxiv.org/abs/2512.19123 · **PDF:** https://arxiv.org/pdf/2512.19123v1
+
+## 26269. BanglaForge: LLM Collaboration with Self-Refinement for Bangla Code Generation
+
+- **ID:** `2512.19122` · **primary:** `cs.SE` · **Submitted:** 2025-12-22
+- **Authors:** Mahir Labib Dihan, Sadif Ahmed, Md Nafiu Rahman
+- **abs:** https://arxiv.org/abs/2512.19122 · **PDF:** https://arxiv.org/pdf/2512.19122v1
+
+## 26270. Stop saying LLM: Large Discourse Models (LDM) and Artificial Discursive Agent (ADA)?
+
+- **ID:** `2512.19117` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Amar Lakel
+- **abs:** https://arxiv.org/abs/2512.19117 · **PDF:** https://arxiv.org/pdf/2512.19117v2
+
+## 26271. HyperLoad: A Cross-Modality Enhanced Large Language Model-Based Framework for Green Data Center Cooling Load Prediction
+
+- **ID:** `2512.19114` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Haoyu Jiang, Boan Qu, Junjie Zhu, et al.
+- **abs:** https://arxiv.org/abs/2512.19114 · **PDF:** https://arxiv.org/pdf/2512.19114v1
+
+## 26272. Using street view images and visual LLMs to predict heritage values for governance support: Risks, ethics, and policy implications
+
+- **ID:** `2601.06056` · **primary:** `cs.CY` · **Submitted:** 2025-12-22
+- **Authors:** Tim Johansson, Mikael Mangold, Kristina Dabrock, et al.
+- **abs:** https://arxiv.org/abs/2601.06056 · **PDF:** https://arxiv.org/pdf/2601.06056v1
+
+## 26273. FC-MIR: A Mobile Screen Awareness Framework for Intent-Aware Recommendation based on Frame-Compressed Multimodal Trajectory Reasoning
+
+- **ID:** `2512.19107` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Zhe Yang, Xiaoshuang Sheng, Zhengnan Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.19107 · **PDF:** https://arxiv.org/pdf/2512.19107v1
+
+## 26274. Explicit and Non-asymptotic Query Complexities of Rank-Based Zeroth-order Algorithm on Stochastic Smooth Functions
+
+- **ID:** `2512.19104` · **primary:** `math.OC` · **Submitted:** 2025-12-22
+- **Authors:** Haishan Ye
+- **abs:** https://arxiv.org/abs/2512.19104 · **PDF:** https://arxiv.org/pdf/2512.19104v1
+
+## 26275. Timely Parameter Updating in Over-the-Air Federated Learning
+
+- **ID:** `2512.19103` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Jiaqi Zhu, Zhongyuan Zhao, Xiao Li, et al.
+- **abs:** https://arxiv.org/abs/2512.19103 · **PDF:** https://arxiv.org/pdf/2512.19103v1
+
+## 26276. Dual Model Deep Learning for Alzheimer Prognostication
+
+- **ID:** `2512.19099` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Alireza Moayedikia, Sara Fin, Uffe Kock Wiil
+- **abs:** https://arxiv.org/abs/2512.19099 · **PDF:** https://arxiv.org/pdf/2512.19099v1
+
+## 26277. DIVER-1 : Deep Integration of Vast Electrophysiological Recordings at Scale
+
+- **ID:** `2512.19097` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Danny Dongyeop Han, Yonghyeon Gwon, Ahhyun Lucy Lee, et al.
+- **abs:** https://arxiv.org/abs/2512.19097 · **PDF:** https://arxiv.org/pdf/2512.19097v2
+
+## 26278. Conditioning Accept-Desirability models in the context of AGM-like belief change
+
+- **ID:** `2512.19096` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Kathelijne Coussement, Gert de Cooman, Keano De Vos
+- **abs:** https://arxiv.org/abs/2512.19096 · **PDF:** https://arxiv.org/pdf/2512.19096v1
+
+## 26279. Tool-Augmented Hybrid Ensemble Reasoning with Distillation for Bilingual Mathematical Problem Solving
+
+- **ID:** `2512.19093` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Peiqing Lu, Yuan Zhang, Haoyun Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.19093 · **PDF:** https://arxiv.org/pdf/2512.19093v1
+
+## 26280. A Large Language Model Based Method for Complex Logical Reasoning over Knowledge Graphs
+
+- **ID:** `2512.19092` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Ziyan Zhang, Chao Wang, Zhuo Chen, et al.
+- **abs:** https://arxiv.org/abs/2512.19092 · **PDF:** https://arxiv.org/pdf/2512.19092v1
+
+## 26281. Auditing Significance, Metric Choice, and Demographic Fairness in Medical AI Challenges
+
+- **ID:** `2512.19091` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Ariel Lubonja, Pedro R. A. S. Bassi, Wenxuan Li, et al.
+- **abs:** https://arxiv.org/abs/2512.19091 · **PDF:** https://arxiv.org/pdf/2512.19091v1
+
+## 26282. Free Energy-Based Modeling of Emotional Dynamics in Video Advertisements
+
+- **ID:** `2601.00812` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Takashi Ushio, Kazuhiro Onishi, Hideyoshi Yanagisawa
+- **abs:** https://arxiv.org/abs/2601.00812 · **PDF:** https://arxiv.org/pdf/2601.00812v1
+
+## 26283. $γ(3,4)$ `Attention' in Cognitive Agents: Ontology-Free Knowledge Representations With Promise Theoretic Semantics
+
+- **ID:** `2512.19084` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Mark Burgess
+- **abs:** https://arxiv.org/abs/2512.19084 · **PDF:** https://arxiv.org/pdf/2512.19084v1
+
+## 26284. Population-Evolve: a Parallel Sampling and Evolutionary Method for LLM Math Reasoning
+
+- **ID:** `2512.19081` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Yanzhi Zhang, Yitong Duan, Zhaoxi Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.19081 · **PDF:** https://arxiv.org/pdf/2512.19081v1
+
+## 26285. Watch Closely: Mitigating Object Hallucinations in Large Vision-Language Models with Disentangled Decoding
+
+- **ID:** `2512.19070` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Ruiqi Ma, Yu Yan, Chunhong Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.19070 · **PDF:** https://arxiv.org/pdf/2512.19070v1
+
+## 26286. Can abstract concepts from LLM improve SLM performance?
+
+- **ID:** `2512.19069` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Siddharth Tandon
+- **abs:** https://arxiv.org/abs/2512.19069 · **PDF:** https://arxiv.org/pdf/2512.19069v1
+
+## 26287. On Cost-Aware Sequential Hypothesis Testing with Random Costs and Action Cancellation
+
+- **ID:** `2512.19067` · **primary:** `cs.IT` · **Submitted:** 2025-12-22
+- **Authors:** George Vershinin, Asaf Cohen, Omer Gurewitz
+- **abs:** https://arxiv.org/abs/2512.19067 · **PDF:** https://arxiv.org/pdf/2512.19067v1
+
+## 26288. Fraud Detection Through Large-Scale Graph Clustering with Heterogeneous Link Transformation
+
+- **ID:** `2512.19061` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Chi Liu
+- **abs:** https://arxiv.org/abs/2512.19061 · **PDF:** https://arxiv.org/pdf/2512.19061v1
+
+## 26289. Efficient Personalization of Generative Models via Optimal Experimental Design
+
+- **ID:** `2512.19057` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Guy Schacht, Ziyad Sheebaelhamd, Riccardo De Santi, et al.
+- **abs:** https://arxiv.org/abs/2512.19057 · **PDF:** https://arxiv.org/pdf/2512.19057v2
+
+## 26290. Time-series Forecast for Indoor Zone Air Temperature with Long Horizons: A Case Study with Sensor-based Data from a Smart Building
+
+- **ID:** `2512.19038` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Liping Sun, Yucheng Guo, Siliang Lu, et al.
+- **abs:** https://arxiv.org/abs/2512.19038 · **PDF:** https://arxiv.org/pdf/2512.19038v2
+
+## 26291. Elevating Intrusion Detection and Security Fortification in Intelligent Networks through Cutting-Edge Machine Learning Paradigms
+
+- **ID:** `2512.19037` · **primary:** `cs.CR` · **Submitted:** 2025-12-22
+- **Authors:** Md Minhazul Islam Munna, Md Mahbubur Rahman, Jaroslav Frnda, et al.
+- **abs:** https://arxiv.org/abs/2512.19037 · **PDF:** https://arxiv.org/pdf/2512.19037v1
+
+## 26292. A Surrogate-Augmented Symbolic CFD-Driven Training Framework for Accelerating Multi-objective Physical Model Development
+
+- **ID:** `2512.19031` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Yuan Fang, Fabian Waschkowski, Maximilian Reissmann, et al.
+- **abs:** https://arxiv.org/abs/2512.19031 · **PDF:** https://arxiv.org/pdf/2512.19031v1
+
+## 26293. A Declarative Language for Building And Orchestrating LLM-Powered Agent Workflows
+
+- **ID:** `2512.19769` · **primary:** `cs.SE` · **Submitted:** 2025-12-22
+- **Authors:** Ivan Daunis
+- **abs:** https://arxiv.org/abs/2512.19769 · **PDF:** https://arxiv.org/pdf/2512.19769v1
+
+## 26294. Recontextualization Mitigates Specification Gaming without Modifying the Specification
+
+- **ID:** `2512.19027` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Ariana Azarbal, Victor Gillioz, Vladimir Ivanov, et al.
+- **abs:** https://arxiv.org/abs/2512.19027 · **PDF:** https://arxiv.org/pdf/2512.19027v2
+
+## 26295. Finer-Personalization Rank: Fine-Grained Retrieval Examines Identity Preservation for Personalized Generation
+
+- **ID:** `2512.19026` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Connor Kilrain, David Carlyn, Julia Chae, et al.
+- **abs:** https://arxiv.org/abs/2512.19026 · **PDF:** https://arxiv.org/pdf/2512.19026v1
+
+## 26296. Transformer Reconstructed with Dynamic Value Attention
+
+- **ID:** `2512.22212` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Xiaowei Wang
+- **abs:** https://arxiv.org/abs/2512.22212 · **PDF:** https://arxiv.org/pdf/2512.22212v1
+
+## 26297. The Erasure Illusion: Stress-Testing the Generalization of LLM Forgetting Evaluation
+
+- **ID:** `2512.19025` · **primary:** `cs.CR` · **Submitted:** 2025-12-22
+- **Authors:** Hengrui Jia, Taoran Li, Jonas Guan, et al.
+- **abs:** https://arxiv.org/abs/2512.19025 · **PDF:** https://arxiv.org/pdf/2512.19025v2
+
+## 26298. IndoorUAV: Benchmarking Vision-Language UAV Navigation in Continuous Indoor Environments
+
+- **ID:** `2512.19024` · **primary:** `cs.RO` · **Submitted:** 2025-12-22
+- **Authors:** Xu Liu, Yu Liu, Hanshuo Qiu, et al.
+- **abs:** https://arxiv.org/abs/2512.19024 · **PDF:** https://arxiv.org/pdf/2512.19024v1
+
+## 26299. CETCAM: Camera-Controllable Video Generation via Consistent and Extensible Tokenization
+
+- **ID:** `2512.19020` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Zelin Zhao, Xinyu Gong, Bangya Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.19020 · **PDF:** https://arxiv.org/pdf/2512.19020v1
+
+## 26300. Optimizer Dynamics at the Edge of Stability with Differential Privacy
+
+- **ID:** `2512.19019` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Ayana Hussain, Ricky Fang
+- **abs:** https://arxiv.org/abs/2512.19019 · **PDF:** https://arxiv.org/pdf/2512.19019v1
+
+## 26301. DramaBench: A Six-Dimensional Evaluation Framework for Drama Script Continuation
+
+- **ID:** `2512.19012` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Shijian Ma, Yunqi Huang, Yan Lin
+- **abs:** https://arxiv.org/abs/2512.19012 · **PDF:** https://arxiv.org/pdf/2512.19012v2
+
+## 26302. PromptScreen: Efficient Jailbreak Mitigation Using Semantic Linear Classification in a Multi-Staged Pipeline
+
+- **ID:** `2512.19011` · **primary:** `cs.CR` · **Submitted:** 2025-12-22
+- **Authors:** Akshaj Prashanth Rao, Advait Singh, Saumya Kumaar Saksena, et al.
+- **abs:** https://arxiv.org/abs/2512.19011 · **PDF:** https://arxiv.org/pdf/2512.19011v2
+
+## 26303. With Great Capabilities Come Great Responsibilities: Introducing the Agentic Risk & Capability Framework for Governing Agentic AI Systems
+
+- **ID:** `2512.22211` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Shaun Khoo, Jessica Foo, Roy Ka-Wei Lee
+- **abs:** https://arxiv.org/abs/2512.22211 · **PDF:** https://arxiv.org/pdf/2512.22211v1
+
+## 26304. The 6th International Verification of Neural Networks Competition (VNN-COMP 2025): Summary and Results
+
+- **ID:** `2512.19007` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Konstantin Kaulen, Tobias Ladner, Stanley Bak, et al.
+- **abs:** https://arxiv.org/abs/2512.19007 · **PDF:** https://arxiv.org/pdf/2512.19007v1
+
+## 26305. Toward Equitable Recovery: A Fairness-Aware AI Framework for Prioritizing Post-Flood Aid in Bangladesh
+
+- **ID:** `2512.22210` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Farjana Yesmin, Romana Akter
+- **abs:** https://arxiv.org/abs/2512.22210 · **PDF:** https://arxiv.org/pdf/2512.22210v1
+
+## 26306. Context-Aware Initialization for Reducing Generative Path Length in Diffusion Language Models
+
+- **ID:** `2512.19004` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Tongyuan Miao, Gary Huang, Kai Jun Han, et al.
+- **abs:** https://arxiv.org/abs/2512.19004 · **PDF:** https://arxiv.org/pdf/2512.19004v1
+
+## 26307. ORPR: An OR-Guided Pretrain-then-Reinforce Learning Model for Inventory Management
+
+- **ID:** `2512.19001` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Lingjie Zhao, Xue Yu, Yongzhi Qi, et al.
+- **abs:** https://arxiv.org/abs/2512.19001 · **PDF:** https://arxiv.org/pdf/2512.19001v2
+
+## 26308. Super-Resolution Enhancement of Medical Images Based on Diffusion Model: An Optimization Scheme for Low-Resolution Gastric Images
+
+- **ID:** `2512.22209` · **primary:** `eess.IV` · **Submitted:** 2025-12-22
+- **Authors:** Haozhe Jia
+- **abs:** https://arxiv.org/abs/2512.22209 · **PDF:** https://arxiv.org/pdf/2512.22209v1
+
+## 26309. Evaluating the Challenges of LLMs in Real-world Medical Follow-up: A Comparative Study and An Optimized Framework
+
+- **ID:** `2512.18999` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Jinyan Liu, Zikang Chen, Qinchuan Wang, et al.
+- **abs:** https://arxiv.org/abs/2512.18999 · **PDF:** https://arxiv.org/pdf/2512.18999v1
+
+## 26310. Training-Free Global Geometric Association for 4D LiDAR Panoptic Segmentation
+
+- **ID:** `2512.18991` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Gyeongrok Oh, Youngdong Jang, Jonghyun Choi, et al.
+- **abs:** https://arxiv.org/abs/2512.18991 · **PDF:** https://arxiv.org/pdf/2512.18991v2
+
+## 26311. Affordance RAG: Hierarchical Multimodal Retrieval with Affordance-Aware Embodied Memory for Mobile Manipulation
+
+- **ID:** `2512.18987` · **primary:** `cs.RO` · **Submitted:** 2025-12-22
+- **Authors:** Ryosuke Korekata, Quanting Xie, Yonatan Bisk, et al.
+- **abs:** https://arxiv.org/abs/2512.18987 · **PDF:** https://arxiv.org/pdf/2512.18987v1
+
+## 26312. R-GenIMA: Integrating Neuroimaging and Genetics with Interpretable Multimodal AI for Alzheimer's Disease Progression
+
+- **ID:** `2512.18986` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Kun Zhao, Siyuan Dai, Yingying Zhang, et al.
+- **abs:** https://arxiv.org/abs/2512.18986 · **PDF:** https://arxiv.org/pdf/2512.18986v1
+
+## 26313. OPBO: Order-Preserving Bayesian Optimization
+
+- **ID:** `2512.18980` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Wei Peng, Jianchen Hu, Kang Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.18980 · **PDF:** https://arxiv.org/pdf/2512.18980v1
+
+## 26314. Outlier detection in mixed-attribute data: a semi-supervised approach with fuzzy approximations and relative entropy
+
+- **ID:** `2512.18978` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Baiyang Chen, Zhong Yuan, Zheng Liu, et al.
+- **abs:** https://arxiv.org/abs/2512.18978 · **PDF:** https://arxiv.org/pdf/2512.18978v1
+
+## 26315. Consistency-guided semi-supervised outlier detection in heterogeneous data using fuzzy rough sets
+
+- **ID:** `2512.18977` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Baiyang Chen, Zhong Yuan, Dezhong Peng, et al.
+- **abs:** https://arxiv.org/abs/2512.18977 · **PDF:** https://arxiv.org/pdf/2512.18977v1
+
+## 26316. Open-Source Multimodal Moxin Models with Moxin-VLM and Moxin-VLA
+
+- **ID:** `2512.22208` · **primary:** `cs.CL` · **Submitted:** 2025-12-22
+- **Authors:** Pu Zhao, Arash Akbari, Xuan Shen, et al.
+- **abs:** https://arxiv.org/abs/2512.22208 · **PDF:** https://arxiv.org/pdf/2512.22208v2
+
+## 26317. On Conditional Stochastic Interpolation for Generative Nonlinear Sufficient Dimension Reduction
+
+- **ID:** `2512.18971` · **primary:** `stat.ML` · **Submitted:** 2025-12-22
+- **Authors:** Shuntuo Xu, Zhou Yu, Jian Huang
+- **abs:** https://arxiv.org/abs/2512.18971 · **PDF:** https://arxiv.org/pdf/2512.18971v1
+
+## 26318. Self-Attention with State-Object Weighted Combination for Compositional Zero Shot Learning
+
+- **ID:** `2512.18969` · **primary:** `cs.CV` · **Submitted:** 2025-12-22
+- **Authors:** Cheng-Hong Chang, Pei-Hsuan Tsai
+- **abs:** https://arxiv.org/abs/2512.18969 · **PDF:** https://arxiv.org/pdf/2512.18969v1
+
+## 26319. Lag Operator SSMs: A Geometric Framework for Structured State Space Modeling
+
+- **ID:** `2512.18965` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Sutashu Tomonaga, Kenji Doya, Noboru Murata
+- **abs:** https://arxiv.org/abs/2512.18965 · **PDF:** https://arxiv.org/pdf/2512.18965v1
+
+## 26320. Online Robust Reinforcement Learning with General Function Approximation
+
+- **ID:** `2512.18957` · **primary:** `cs.LG` · **Submitted:** 2025-12-22
+- **Authors:** Debamita Ghosh, George K. Atia, Yue Wang
+- **abs:** https://arxiv.org/abs/2512.18957 · **PDF:** https://arxiv.org/pdf/2512.18957v3
+
+## 26321. Training Multimodal Large Reasoning Models Needs Better Thoughts: A Three-Stage Framework for Long Chain-of-Thought Synthesis and Selection
+
+- **ID:** `2512.18956` · **primary:** `cs.AI` · **Submitted:** 2025-12-22
+- **Authors:** Yizhi Wang, Linan Yue, Min-Ling Zhang
+- **abs:** https://arxiv.org/abs/2512.18956 · **PDF:** https://arxiv.org/pdf/2512.18956v2
+
+## 26322. Benchmarking Attribute Discrimination in Infant-Scale Vision-Language Models
+
+- **ID:** `2512.18951` · **primary:** `cs.LG` · **Submitted:
+
+**Scores:** Importance:10, Urgency:10
+**Tier:** 1
diff --git a/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_long.md b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_long.md
new file mode 100644
index 0000000000000000000000000000000000000000..740f58e8147fc009754ffb2c2a07381fcd9d6448
--- /dev/null
+++ b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_long.md
@@ -0,0 +1,11 @@
+Good evening. Tonight, we turn to the rapidly evolving world of artificial intelligence as a new wave of advancements in generative AI has sparked both excitement and debate.
+
+These breakthroughs are being driven by a surge in research across multiple disciplines, with researchers exploring novel methods for creating realistic images, generating text, and even developing systems capable of understanding complex scientific concepts. This progress is fueled by the increasing availability of powerful computing resources and the collaborative efforts of researchers worldwide.
+
+One area of particular interest is the development of AI models that can generate high-quality content like images, videos, and written text. We're seeing these advancements in fields such as computer graphics, where AI is being used to create realistic virtual environments for games and simulations. In creative industries, AI tools are already helping artists and designers produce stunning visual effects and generate new ideas.
+
+The impact of this technology goes beyond entertainment and art. Researchers are exploring the potential of AI to revolutionize fields like medicine and healthcare. AI-powered systems are being used to analyze medical images, predict disease outbreaks, and even develop personalized treatments for patients. These applications hold immense promise for improving human health and well-being.
+
+However, alongside these exciting advancements, we must also address the ethical considerations surrounding AI development. As with any powerful technology, there are concerns about potential misuse, bias in algorithms, and the impact on employment. It's crucial that we engage in open discussions and develop responsible guidelines for the development and deployment of AI systems to ensure they benefit society as a whole.
+
+The future of AI is brimming with possibilities. As these technologies continue to evolve, it will be vital to prioritize ethical considerations and collaborate across disciplines to harness their full potential for good.
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_long_clean.txt b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_long_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..aec4fa69c1eda20ff0211d32c487cf9a81905dd0
--- /dev/null
+++ b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_long_clean.txt
@@ -0,0 +1,11 @@
+Good evening. Tonight, we turn to the rapidly evolving world of artificial intelligence as a new wave of advancements in generative AI has sparked both excitement and debate.
+
+These breakthroughs are being driven by a surge in research across multiple disciplines, with researchers exploring novel methods for creating realistic images, generating text, and even developing systems capable of understanding complex scientific concepts. This progress is fueled by the increasing availability of powerful computing resources and the collaborative efforts of researchers worldwide.
+
+One area of particular interest is the development of AI models that can generate high-quality content like images, videos, and written text. We're seeing these advancements in fields such as computer graphics, where AI is being used to create realistic virtual environments for games and simulations. In creative industries, AI tools are already helping artists and designers produce stunning visual effects and generate new ideas.
+
+The impact of this technology goes beyond entertainment and art. Researchers are exploring the potential of AI to revolutionize fields like medicine and healthcare. AI-powered systems are being used to analyze medical images, predict disease outbreaks, and even develop personalized treatments for patients. These applications hold immense promise for improving human health and well-being.
+
+However, alongside these exciting advancements, we must also address the ethical considerations surrounding AI development. As with any powerful technology, there are concerns about potential misuse, bias in algorithms, and the impact on employment. It's crucial that we engage in open discussions and develop responsible guidelines for the development and deployment of AI systems to ensure they benefit society as a whole.
+
+The future of AI is brimming with possibilities. As these technologies continue to evolve, it will be vital to prioritize ethical considerations and collaborate across disciplines to harness their full potential for good.
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_middle.md b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_middle.md
new file mode 100644
index 0000000000000000000000000000000000000000..5b49e79d45efb6feb42309543be26d6266a264da
--- /dev/null
+++ b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_middle.md
@@ -0,0 +1,7 @@
+Good evening. We're looking at some new research in the field of artificial intelligence and machine learning. A recent batch of papers has been published on arXiv, and they are making waves in the world of AI development.
+
+The papers cover a wide range of topics, from video game creation to language models. We're seeing a lot of focus on improving how these models work, especially when it comes to generating new content and understanding complex interactions.
+One paper explores "ActionParty," which is an attempt to create more dynamic and realistic action scenes in video games. Another focuses on "Steerable Visual Representations" and how to make images more adaptable and useful for different tasks.
+
+These are just a few examples of the exciting new developments happening in AI research. We'll continue to follow these breakthroughs as they unfold, and we'll bring you updates as they become available.
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_middle_clean.txt b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_middle_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..7551d1a4889c2e871aa4d4a037193a5105119774
--- /dev/null
+++ b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_middle_clean.txt
@@ -0,0 +1,9 @@
+Good evening. We're looking at some new research in the field of artificial intelligence and machine learning. A recent batch of papers has been published on arXiv, and they are making waves in the world of AI development.
+
+The papers cover a wide range of topics, from video game creation to language models. We're seeing a lot of focus on improving how these models work, especially when it comes to generating new content and understanding complex interactions.
+
+One paper explores "ActionParty," which is an attempt to create more dynamic and realistic action scenes in video games. Another focuses on "Steerable Visual Representations" and how to make images more adaptable and useful for different tasks.
+
+These are just a few examples of the exciting new developments happening in AI research. We'll continue to follow these breakthroughs as they unfold, and we'll bring you updates as they become available.
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_short.md b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_short.md
new file mode 100644
index 0000000000000000000000000000000000000000..716e4ff9e25a01d8765487f1afc252c97d6b29f6
--- /dev/null
+++ b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_short.md
@@ -0,0 +1 @@
+Breaking news! AI is booming - and it's about to change the way we play video games. A new study shows researchers are developing AI that can create more realistic and engaging gaming experiences. This could revolutionize the entire industry, with potential for everything from lifelike characters to dynamic environments. Stay tuned as this story unfolds! Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_short_clean.txt b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_short_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..716e4ff9e25a01d8765487f1afc252c97d6b29f6
--- /dev/null
+++ b/docs/AI_Machine_Learning/arxiv-ai-ml-20250405-20260405_script_short_clean.txt
@@ -0,0 +1 @@
+Breaking news! AI is booming - and it's about to change the way we play video games. A new study shows researchers are developing AI that can create more realistic and engaging gaming experiences. This could revolutionize the entire industry, with potential for everything from lifelike characters to dynamic environments. Stay tuned as this story unfolds! Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/arxiv-cs-multi-2026-04-05.md b/docs/AI_Machine_Learning/arxiv-cs-multi-2026-04-05.md
new file mode 100644
index 0000000000000000000000000000000000000000..c8707274953ddf21e07d2171d61785e9ec4533b2
--- /dev/null
+++ b/docs/AI_Machine_Learning/arxiv-cs-multi-2026-04-05.md
@@ -0,0 +1,331 @@
+# arXiv computer science — recent submissions (multi-category)
+
+*Categories:* cs.AI, cs.LG, cs.CL, cs.CV, cs.NE, cs.RO, cs.SY
+
+*Feed `updated`: 2026-04-05T06:25:36Z*
+
+*Fetched at (UTC): 2026-04-05T06:25:37Z*
+
+*Machine-readable twin: `arxiv-cs-multi-2026-04-05.json`*
+
+---
+
+## 1. EventHub: Data Factory for Generalizable Event-Based Stereo Networks without Active Sensors
+
+- **arXiv ID:** `2604.02331`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Luca Bartolomei, Fabio Tosi, Matteo Poggi, Stefano Mattoccia, Guillermo Gallego
+- **Abstract page:** [https://arxiv.org/abs/2604.02331](https://arxiv.org/abs/2604.02331)
+- **PDF:** [https://arxiv.org/pdf/2604.02331v1](https://arxiv.org/pdf/2604.02331v1)
+
+### Abstract
+
+We propose EventHub, a novel framework for training deep-event stereo networks without ground truth annotations from costly active sensors, relying instead on standard color images. From these images, we derive either proxy annotations and proxy events through state-of-the-art novel view synthesis techniques, or simply proxy annotations when images are already paired with event data. Using the training set generated by our data factory, we repurpose state-of-the-art stereo models from RGB literature to process event data, obtaining new event stereo models with unprecedented generalization capabilities. Experiments on widely used event stereo datasets support the effectiveness of EventHub and show how the same data distillation mechanism can improve the accuracy of RGB stereo foundation models in challenging conditions such as nighttime scenes.
+
+---
+
+## 2. ActionParty: Multi-Subject Action Binding in Generative Video Games
+
+- **arXiv ID:** `2604.02330`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `cs.AI`, `cs.LG`
+- **Published:** 2026-04-02
+- **Authors:** Alexander Pondaven, Ziyi Wu, Igor Gilitschenski, Philip Torr, Sergey Tulyakov, Fabio Pizzati, Aliaksandr Siarohin
+- **Abstract page:** [https://arxiv.org/abs/2604.02330](https://arxiv.org/abs/2604.02330)
+- **PDF:** [https://arxiv.org/pdf/2604.02330v1](https://arxiv.org/pdf/2604.02330v1)
+
+### Abstract
+
+Recent advances in video diffusion have enabled the development of "world models" capable of simulating interactive environments. However, these models are largely restricted to single-agent settings, failing to control multiple agents simultaneously in a scene. In this work, we tackle a fundamental issue of action binding in existing video diffusion models, which struggle to associate specific actions with their corresponding subjects. For this purpose, we propose ActionParty, an action controllable multi-subject world model for generative video games. It introduces subject state tokens, i.e. latent variables that persistently capture the state of each subject in the scene. By jointly modeling state tokens and video latents with a spatial biasing mechanism, we disentangle global video frame rendering from individual action-controlled subject updates. We evaluate ActionParty on the Melting Pot benchmark, demonstrating the first video world model capable of controlling up to seven players simultaneously across 46 diverse environments. Our results show significant improvements in action-following accuracy and identity consistency, while enabling robust autoregressive tracking of subjects through complex interactions.
+
+---
+
+## 3. Generative World Renderer
+
+- **arXiv ID:** `2604.02329`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Zheng-Hui Huang, Zhixiang Wang, Jiaming Tan, Ruihan Yu, Yidan Zhang, Bo Zheng, Yu-Lun Liu, Yung-Yu Chuang, Kaipeng Zhang
+- **Abstract page:** [https://arxiv.org/abs/2604.02329](https://arxiv.org/abs/2604.02329)
+- **PDF:** [https://arxiv.org/pdf/2604.02329v1](https://arxiv.org/pdf/2604.02329v1)
+
+### Abstract
+
+Scaling generative inverse and forward rendering to real-world scenarios is bottlenecked by the limited realism and temporal coherence of existing synthetic datasets. To bridge this persistent domain gap, we introduce a large-scale, dynamic dataset curated from visually complex AAA games. Using a novel dual-screen stitched capture method, we extracted 4M continuous frames (720p/30 FPS) of synchronized RGB and five G-buffer channels across diverse scenes, visual effects, and environments, including adverse weather and motion-blur variants. This dataset uniquely advances bidirectional rendering: enabling robust in-the-wild geometry and material decomposition, and facilitating high-fidelity G-buffer-guided video generation. Furthermore, to evaluate the real-world performance of inverse rendering without ground truth, we propose a novel VLM-based assessment protocol measuring semantic, spatial, and temporal consistency. Experiments demonstrate that inverse renderers fine-tuned on our data achieve superior cross-dataset generalization and controllable generation, while our VLM evaluation strongly correlates with human judgment. Combined with our toolkit, our forward renderer enables users to edit styles of AAA games from G-buffers using text prompts.
+
+---
+
+## 4. Modulate-and-Map: Crossmodal Feature Mapping with Cross-View Modulation for 3D Anomaly Detection
+
+- **arXiv ID:** `2604.02328`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Alex Costanzino, Pierluigi Zama Ramirez, Giuseppe Lisanti, Luigi Di Stefano
+- **Abstract page:** [https://arxiv.org/abs/2604.02328](https://arxiv.org/abs/2604.02328)
+- **PDF:** [https://arxiv.org/pdf/2604.02328v1](https://arxiv.org/pdf/2604.02328v1)
+
+### Abstract
+
+We present ModMap, a natively multiview and multimodal framework for 3D anomaly detection and segmentation. Unlike existing methods that process views independently, our method draws inspiration from the crossmodal feature mapping paradigm to learn to map features across both modalities and views, while explicitly modelling view-dependent relationships through feature-wise modulation. We introduce a cross-view training strategy that leverages all possible view combinations, enabling effective anomaly scoring through multiview ensembling and aggregation. To process high-resolution 3D data, we train and publicly release a foundational depth encoder tailored to industrial datasets. Experiments on SiM3D, a recent benchmark that introduces the first multiview and multimodal setup for 3D anomaly detection and segmentation, demonstrate that ModMap attains state-of-the-art performance by surpassing previous methods by wide margins.
+
+---
+
+## 5. Steerable Visual Representations
+
+- **arXiv ID:** `2604.02327`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `cs.AI`
+- **Published:** 2026-04-02
+- **Authors:** Jona Ruthardt, Manu Gaur, Deva Ramanan, Makarand Tapaswi, Yuki M. Asano
+- **Abstract page:** [https://arxiv.org/abs/2604.02327](https://arxiv.org/abs/2604.02327)
+- **PDF:** [https://arxiv.org/pdf/2604.02327v1](https://arxiv.org/pdf/2604.02327v1)
+
+### Abstract
+
+Pretrained Vision Transformers (ViTs) such as DINOv2 and MAE provide generic image features that can be applied to a variety of downstream tasks such as retrieval, classification, and segmentation. However, such representations tend to focus on the most salient visual cues in the image, with no way to direct them toward less prominent concepts of interest. In contrast, Multimodal LLMs can be guided with textual prompts, but the resulting representations tend to be language-centric and lose their effectiveness for generic visual tasks. To address this, we introduce Steerable Visual Representations, a new class of visual representations, whose global and local features can be steered with natural language. While most vision-language models (e.g., CLIP) fuse text with visual features after encoding (late fusion), we inject text directly into the layers of the visual encoder (early fusion) via lightweight cross-attention. We introduce benchmarks for measuring representational steerability, and demonstrate that our steerable visual features can focus on any desired objects in an image while preserving the underlying representation quality. Our method also matches or outperforms dedicated approaches on anomaly detection and personalized object discrimination, exhibiting zero-shot generalization to out-of-distribution tasks.
+
+---
+
+## 6. Grounded Token Initialization for New Vocabulary in LMs for Generative Recommendation
+
+- **arXiv ID:** `2604.02324`
+- **Primary category:** `cs.CL`
+- **Categories:** `cs.CL`, `cs.AI`, `cs.LG`
+- **Published:** 2026-04-02
+- **Authors:** Daiwei Chen, Zhoutong Fu, Chengming Jiang, Haichao Zhang, Ran Zhou, Tan Wang, Chunnan Yao, Guoyao Li, Rui Cai, Yihan Cao, Ruijie Jiang, Fedor Borisyuk, Jianqiang Shen, Jingwei Wu, Ramya Korlakai Vinayak
+- **Abstract page:** [https://arxiv.org/abs/2604.02324](https://arxiv.org/abs/2604.02324)
+- **PDF:** [https://arxiv.org/pdf/2604.02324v1](https://arxiv.org/pdf/2604.02324v1)
+
+### Abstract
+
+Language models (LMs) are increasingly extended with new learnable vocabulary tokens for domain-specific tasks, such as Semantic-ID tokens in generative recommendation. The standard practice initializes these new tokens as the mean of existing vocabulary embeddings, then relies on supervised fine-tuning to learn their representations. We present a systematic analysis of this strategy: through spectral and geometric diagnostics, we show that mean initialization collapses all new tokens into a degenerate subspace, erasing inter-token distinctions that subsequent fine-tuning struggles to fully recover. These findings suggest that \emph{token initialization} is a key bottleneck when extending LMs with new vocabularies. Motivated by this diagnosis, we propose the \emph{Grounded Token Initialization Hypothesis}: linguistically grounding novel tokens in the pretrained embedding space before fine-tuning better enables the model to leverage its general-purpose knowledge for novel-token domains. We operationalize this hypothesis as GTI (Grounded Token Initialization), a lightweight grounding stage that, prior to fine-tuning, maps new tokens to distinct, semantically meaningful locations in the pretrained embedding space using only paired linguistic supervision. Despite its simplicity, GTI outperforms both mean initialization and existing auxiliary-task adaptation methods in the majority of evaluation settings across multiple generative recommendation benchmarks, including industry-scale and public datasets. Further analyses show that grounded embeddings produce richer inter-token structure that persists through fine-tuning, corroborating the hypothesis that initialization quality is a key bottleneck in vocabulary extension.
+
+---
+
+## 7. Beyond Referring Expressions: Scenario Comprehension Visual Grounding
+
+- **arXiv ID:** `2604.02323`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Ruozhen He, Nisarg A. Shah, Qihua Dong, Zilin Xiao, Jaywon Koo, Vicente Ordonez
+- **Abstract page:** [https://arxiv.org/abs/2604.02323](https://arxiv.org/abs/2604.02323)
+- **PDF:** [https://arxiv.org/pdf/2604.02323v1](https://arxiv.org/pdf/2604.02323v1)
+
+### Abstract
+
+Existing visual grounding benchmarks primarily evaluate alignment between image regions and literal referring expressions, where models can often succeed by matching a prominent named category. We explore a complementary and more challenging setting of scenario-based visual grounding, where the target must be inferred from roles, intentions, and relational context rather than explicit naming. We introduce Referring Scenario Comprehension (RSC), a benchmark designed for this setting. The queries in this benchmark are paragraph-length texts that describe object roles, user goals, and contextual cues, including deliberate references to distractor objects that often require deep understanding to resolve. Each instance is annotated with interpretable difficulty tags for uniqueness, clutter, size, overlap, and position which expose distinct failure modes and support fine-grained analysis. RSC contains approximately 31k training examples, 4k in-domain test examples, and a 3k out-of-distribution split with unseen object categories. We further propose ScenGround, a curriculum reasoning method serving as a reference point for this setting, combining supervised warm-starting with difficulty-aware reinforcement learning. Experiments show that scenario-based queries expose systematic failures in current models that standard benchmarks do not reveal, and that curriculum training improves performance on challenging slices and transfers to standard benchmarks.
+
+---
+
+## 8. Batched Contextual Reinforcement: A Task-Scaling Law for Efficient Reasoning
+
+- **arXiv ID:** `2604.02322`
+- **Primary category:** `cs.LG`
+- **Categories:** `cs.LG`, `cs.AI`, `cs.CL`
+- **Published:** 2026-04-02
+- **Authors:** Bangji Yang, Hongbo Ma, Jiajun Fan, Ge Liu
+- **Abstract page:** [https://arxiv.org/abs/2604.02322](https://arxiv.org/abs/2604.02322)
+- **PDF:** [https://arxiv.org/pdf/2604.02322v1](https://arxiv.org/pdf/2604.02322v1)
+
+### Abstract
+
+Large Language Models employing Chain-of-Thought reasoning achieve strong performance but suffer from excessive token consumption that inflates inference costs. Existing efficiency methods such as explicit length penalties, difficulty estimators, or multi-stage curricula either degrade reasoning quality or require complex training pipelines. We introduce Batched Contextual Reinforcement, a minimalist, single-stage training paradigm that unlocks efficient reasoning through a simple structural modification: training the model to solve N problems simultaneously within a shared context window, rewarded purely by per-instance accuracy. This formulation creates an implicit token budget that yields several key findings: (1) We identify a novel task-scaling law: as the number of concurrent problems N increases during inference, per-problem token usage decreases monotonically while accuracy degrades far more gracefully than baselines, establishing N as a controllable throughput dimension. (2) BCR challenges the traditional accuracy-efficiency trade-off by demonstrating a "free lunch" phenomenon at standard single-problem inference. Across both 1.5B and 4B model families, BCR reduces token usage by 15.8% to 62.6% while consistently maintaining or improving accuracy across five major mathematical benchmarks. (3) Qualitative analyses reveal emergent self-regulated efficiency, where models autonomously eliminate redundant metacognitive loops without explicit length supervision. (4) Crucially, we empirically demonstrate that implicit budget constraints successfully circumvent the adversarial gradients and catastrophic optimization collapse inherent to explicit length penalties, offering a highly stable, constraint-based alternative for length control. These results prove BCR practical, showing simple structural incentives unlock latent high-density reasoning in LLMs.
+
+---
+
+## 9. Large-scale Codec Avatars: The Unreasonable Effectiveness of Large-scale Avatar Pretraining
+
+- **arXiv ID:** `2604.02320`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `cs.GR`
+- **Published:** 2026-04-02
+- **Authors:** Junxuan Li, Rawal Khirodkar, Chengan He, Zhongshi Jiang, Giljoo Nam, Lingchen Yang, Jihyun Lee, Egor Zakharov, Zhaoen Su, Rinat Abdrashitov, Yuan Dong, Julieta Martinez, Kai Li, Qingyang Tan, Takaaki Shiratori, Matthew Hu, Peihong Guo, Xuhua Huang, Ariyan Zarei, Marco Pesavento, Yichen Xu, He Wen, Teng Deng, Wyatt Borsos, Anjali Thakrar, Jean-Charles Bazin, Carsten Stoll, Ginés Hidalgo, James Booth, Lucy Wang, Xiaowen Ma, Yu Rong, Sairanjith Thalanki, Chen Cao, Christian Häne, Abhishek Kar, Sofien Bouaziz, Jason Saragih, Yaser Sheikh, Shunsuke Saito
+- **Abstract page:** [https://arxiv.org/abs/2604.02320](https://arxiv.org/abs/2604.02320)
+- **PDF:** [https://arxiv.org/pdf/2604.02320v1](https://arxiv.org/pdf/2604.02320v1)
+
+### Abstract
+
+High-quality 3D avatar modeling faces a critical trade-off between fidelity and generalization. On the one hand, multi-view studio data enables high-fidelity modeling of humans with precise control over expressions and poses, but it struggles to generalize to real-world data due to limited scale and the domain gap between the studio environment and the real world. On the other hand, recent large-scale avatar models trained on millions of in-the-wild samples show promise for generalization across a wide range of identities, yet the resulting avatars are often of low-quality due to inherent 3D ambiguities. To address this, we present Large-Scale Codec Avatars (LCA), a high-fidelity, full-body 3D avatar model that generalizes to world-scale populations in a feedforward manner, enabling efficient inference. Inspired by the success of large language models and vision foundation models, we present, for the first time, a pre/post-training paradigm for 3D avatar modeling at scale: we pretrain on 1M in-the-wild videos to learn broad priors over appearance and geometry, then post-train on high-quality curated data to enhance expressivity and fidelity. LCA generalizes across hair styles, clothing, and demographics while providing precise, fine-grained facial expressions and finger-level articulation control, with strong identity preservation. Notably, we observe emergent generalization to relightability and loose garment support to unconstrained inputs, and zero-shot robustness to stylized imagery, despite the absence of direct supervision.
+
+---
+
+## 10. No Single Best Model for Diversity: Learning a Router for Sample Diversity
+
+- **arXiv ID:** `2604.02319`
+- **Primary category:** `cs.CL`
+- **Categories:** `cs.CL`
+- **Published:** 2026-04-02
+- **Authors:** Yuhan Liu, Fangyuan Xu, Vishakh Padmakumar, Daphne Ippolito, Eunsol Choi
+- **Abstract page:** [https://arxiv.org/abs/2604.02319](https://arxiv.org/abs/2604.02319)
+- **PDF:** [https://arxiv.org/pdf/2604.02319v1](https://arxiv.org/pdf/2604.02319v1)
+
+### Abstract
+
+When posed with prompts that permit a large number of valid answers, comprehensively generating them is the first step towards satisfying a wide range of users. In this paper, we study methods to elicit a comprehensive set of valid responses. To evaluate this, we introduce \textbf{diversity coverage}, a metric that measures the total quality scores assigned to each \textbf{unique} answer in the predicted answer set relative to the best possible answer set with the same number of answers. Using this metric, we evaluate 18 LLMs, finding no single model dominates at generating diverse responses to a wide range of open-ended prompts. Yet, per each prompt, there exists a model that outperforms all other models significantly at generating a diverse answer set. Motivated by this finding, we introduce a router that predicts the best model for each query. On NB-Wildchat, our trained router outperforms the single best model baseline (26.3% vs $23.8%). We further show generalization to an out-of-domain dataset (NB-Curated) as well as different answer-generation prompting strategies. Our work lays foundation for studying generating comprehensive answers when we have access to a suite of models.
+
+---
+
+## 11. Stop Wandering: Efficient Vision-Language Navigation via Metacognitive Reasoning
+
+- **arXiv ID:** `2604.02318`
+- **Primary category:** `cs.RO`
+- **Categories:** `cs.RO`, `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Xueying Li, Feng Lyu, Hao Wu, Mingliu Liu, Jia-Nan Liu, Guozi Liu
+- **Abstract page:** [https://arxiv.org/abs/2604.02318](https://arxiv.org/abs/2604.02318)
+- **PDF:** [https://arxiv.org/pdf/2604.02318v1](https://arxiv.org/pdf/2604.02318v1)
+
+### Abstract
+
+Training-free Vision-Language Navigation (VLN) agents powered by foundation models can follow instructions and explore 3D environments. However, existing approaches rely on greedy frontier selection and passive spatial memory, leading to inefficient behaviors such as local oscillation and redundant revisiting. We argue that this stems from a lack of metacognitive capabilities: the agent cannot monitor its exploration progress, diagnose strategy failures, or adapt accordingly. To address this, we propose MetaNav, a metacognitive navigation agent integrating spatial memory, history-aware planning, and reflective correction. Spatial memory builds a persistent 3D semantic map. History-aware planning penalizes revisiting to improve efficiency. Reflective correction detects stagnation and uses an LLM to generate corrective rules that guide future frontier selection. Experiments on GOAT-Bench, HM3D-OVON, and A-EQA show that MetaNav achieves state-of-the-art performance while reducing VLM queries by 20.7%, demonstrating that metacognitive reasoning significantly improves robustness and efficiency.
+
+---
+
+## 12. A Simple Baseline for Streaming Video Understanding
+
+- **arXiv ID:** `2604.02317`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`
+- **Published:** 2026-04-02
+- **Authors:** Yujiao Shen, Shulin Tian, Jingkang Yang, Ziwei Liu
+- **Abstract page:** [https://arxiv.org/abs/2604.02317](https://arxiv.org/abs/2604.02317)
+- **PDF:** [https://arxiv.org/pdf/2604.02317v1](https://arxiv.org/pdf/2604.02317v1)
+
+### Abstract
+
+Recent streaming video understanding methods increasingly rely on complex memory mechanisms to handle long video streams. We challenge this trend with a simple finding: a sliding-window baseline that feeds only the most recent N frames to an off-the-shelf VLM already matches or surpasses published streaming models. We formalize this baseline as SimpleStream and evaluate it against 13 major offline and online video LLM baselines on OVO-Bench and StreamingBench. Despite its simplicity, SimpleStream delivers consistently strong performance. With only 4 recent frames, it reaches 67.7% average accuracy on OVO-Bench and 80.59% on StreamingBench. Controlled ablations further show that the value of longer context is backbone-dependent rather than uniformly increasing with model scale, and reveal a consistent perception-memory trade-off: adding more historical context can improve recall, but often weakens real-time perception. This suggests that stronger memory, retrieval, or compression modules should not be taken as evidence of progress unless they clearly outperform SimpleStream under the same protocol. We therefore argue that future streaming benchmarks should separate recent-scene perception from long-range memory, so that performance improvements from added complexity can be evaluated more clearly.
+
+---
+
+## 13. Beyond the Assistant Turn: User Turn Generation as a Probe of Interaction Awareness in Language Models
+
+- **arXiv ID:** `2604.02315`
+- **Primary category:** `cs.AI`
+- **Categories:** `cs.AI`
+- **Published:** 2026-04-02
+- **Authors:** Sarath Shekkizhar, Romain Cosentino, Adam Earle
+- **Abstract page:** [https://arxiv.org/abs/2604.02315](https://arxiv.org/abs/2604.02315)
+- **PDF:** [https://arxiv.org/pdf/2604.02315v1](https://arxiv.org/pdf/2604.02315v1)
+
+### Abstract
+
+Standard LLM benchmarks evaluate the assistant turn: the model generates a response to an input, a verifier scores correctness, and the analysis ends. This paradigm leaves unmeasured whether the LLM encodes any awareness of what follows the assistant response. We propose user-turn generation as a probe of this gap: given a conversation context of user query and assistant response, we let a model generate under the user role. If the model's weights encode interaction awareness, the generated user turn will be a grounded follow-up that reacts to the preceding context. Through experiments across $11$ open-weight LLMs (Qwen3.5, gpt-oss, GLM) and $5$ datasets (math reasoning, instruction following, conversation), we show that interaction awareness is decoupled from task accuracy. In particular, within the Qwen3.5 family, GSM8K accuracy scales from $41$ ($0.8$B) to $96.8$ ($397$B-A$17$B), yet genuine follow-up rates under deterministic generation remain near zero. In contrast, higher temperature sampling reveals interaction awareness is latent with follow up rates reaching $22$. Controlled perturbations validate that the proposed probe measures a real property of the model, and collaboration-oriented post-training on Qwen3.5-2B demonstrates an increase in follow-up rates. Our results show that user-turn generation captures a dimension of LLM behavior, interaction awareness, that is unexplored and invisible with current assistant-only benchmarks.
+
+---
+
+## 14. Topological Effects in Neural Network Field Theory
+
+- **arXiv ID:** `2604.02313`
+- **Primary category:** `hep-th`
+- **Categories:** `hep-th`, `cs.LG`
+- **Published:** 2026-04-02
+- **Authors:** Christian Ferko, James Halverson, Vishnu Jejjala, Brandon Robinson
+- **Abstract page:** [https://arxiv.org/abs/2604.02313](https://arxiv.org/abs/2604.02313)
+- **PDF:** [https://arxiv.org/pdf/2604.02313v1](https://arxiv.org/pdf/2604.02313v1)
+
+### Abstract
+
+Neural network field theory formulates field theory as a statistical ensemble of fields defined by a network architecture and a density on its parameters. We extend the construction to topological settings via the inclusion of discrete parameters that label the topological quantum number. We recover the Berezinskii--Kosterlitz--Thouless transition, including the spin-wave critical line and the proliferation of vortices at high temperatures. We also verify the T-duality of the bosonic string, showing invariance under the exchange of momentum and winding on $S^1$, the transformation of the sigma model couplings according to the Buscher rules on constant toroidal backgrounds, the enhancement of the current algebra at self-dual radius, and non-geometric T-fold transition functions.
+
+---
+
+## 15. go-$m$HC: Direct Parameterization of Manifold-Constrained Hyper-Connections via Generalized Orthostochastic Matrices
+
+- **arXiv ID:** `2604.02309`
+- **Primary category:** `cs.LG`
+- **Categories:** `cs.LG`, `cs.CL`
+- **Published:** 2026-04-02
+- **Authors:** Torque Dandachi, Sophia Diggs-Galligan
+- **Abstract page:** [https://arxiv.org/abs/2604.02309](https://arxiv.org/abs/2604.02309)
+- **PDF:** [https://arxiv.org/pdf/2604.02309v1](https://arxiv.org/pdf/2604.02309v1)
+
+### Abstract
+
+Doubly stochastic matrices enable learned mixing across residual streams, but parameterizing the set of doubly stochastic matrices (the Birkhoff polytope) exactly and efficiently remains an open challenge. Existing exact methods scale factorially with the number of streams ($d$), while Kronecker-factorized approaches are efficient but expressivity-limited. We introduce a novel exact parameterization grounded in the theory of generalized orthostochastic matrices, which scales as $\mathcal{O}(d^3)$ and exposes a single hyperparameter $s$ which continuously interpolates between a computationally efficient boundary and the fully expressive Birkhoff polytope. Building on Manifold-Constrained Hyper-Connections ($m$HC), a framework for learned dynamic layer connectivity, we instantiate this parameterization in go-$m$HC. Our method composes naturally with Kronecker-factorized methods, substantially recovering expressivity at similar FLOP costs. Spectral analysis indicates that go-$m$HC fills the Birkhoff polytope far more completely than Kronecker-factorized baselines. On synthetic stream-mixing tasks, go-$m$HC achieves the minimum theoretical loss while converging up to $10\times$ faster. We validate our approach in a 30M parameter GPT-style language model. The expressivity, efficiency, and exactness of go-$m$HC offer a practical avenue for scaling $d$ as a new dimension of model capacity.
+
+---
+
+## 16. VOID: Video Object and Interaction Deletion
+
+- **arXiv ID:** `2604.02296`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `cs.AI`
+- **Published:** 2026-04-02
+- **Authors:** Saman Motamed, William Harvey, Benjamin Klein, Luc Van Gool, Zhuoning Yuan, Ta-Ying Cheng
+- **Abstract page:** [https://arxiv.org/abs/2604.02296](https://arxiv.org/abs/2604.02296)
+- **PDF:** [https://arxiv.org/pdf/2604.02296v1](https://arxiv.org/pdf/2604.02296v1)
+
+### Abstract
+
+Existing video object removal methods excel at inpainting content "behind" the object and correcting appearance-level artifacts such as shadows and reflections. However, when the removed object has more significant interactions, such as collisions with other objects, current models fail to correct them and produce implausible results. We present VOID, a video object removal framework designed to perform physically-plausible inpainting in these complex scenarios. To train the model, we generate a new paired dataset of counterfactual object removals using Kubric and HUMOTO, where removing an object requires altering downstream physical interactions. During inference, a vision-language model identifies regions of the scene affected by the removed object. These regions are then used to guide a video diffusion model that generates physically consistent counterfactual outcomes. Experiments on both synthetic and real data show that our approach better preserves consistent scene dynamics after object removal compared to prior video object removal methods. We hope this framework sheds light on how to make video editing models better simulators of the world through high-level causal reasoning.
+
+---
+
+## 17. Taming the Exponential: A Fast Softmax Surrogate for Integer-Native Edge Inference
+
+- **arXiv ID:** `2604.02292`
+- **Primary category:** `cs.LG`
+- **Categories:** `cs.LG`, `cs.AR`
+- **Published:** 2026-04-02
+- **Authors:** Dimitrios Danopoulos, Enrico Lupi, Michael Kagan, Maurizio Pierini
+- **Abstract page:** [https://arxiv.org/abs/2604.02292](https://arxiv.org/abs/2604.02292)
+- **PDF:** [https://arxiv.org/pdf/2604.02292v1](https://arxiv.org/pdf/2604.02292v1)
+
+### Abstract
+
+Softmax can become a computational bottleneck in the Transformer model's Multi-Head Attention (MHA) block, particularly in small models under low-precision inference, where exponentiation and normalization incur significant overhead. As such, we suggest using Head-Calibrated Clipped-Linear Softmax (HCCS), a bounded, monotone surrogate to the exponential softmax function, which uses a clipped linear mapping of the max centered attention logits. This approximation produces a stable probability distribution, maintains the ordering of the original logits and has non-negative values. HCCS differs from previous softmax surrogates as it includes a set of lightweight calibration parameters that are optimized offline based on a representative dataset and calibrated for each individual attention head to preserve the statistical properties of the individual heads. We describe a hardware-motivated implementation of HCCS for high-throughput scenarios targeting the AMD Versal AI Engines. The current reference implementations from AMD for this platform rely upon either bfloat16 arithmetic or LUTs to perform the exponential operation, which might limit the throughput of the platform and fail to utilize the high-throughput integer vector processing units of the AI Engine. In contrast, HCCS provides a natural mapping to the AI Engines' int8 multiply accumulate (MAC) units. To the best of our knowledge, this is the first int8 optimized softmax surrogate for AMD AI engines that significantly exceeds the speed performance of other reference implementations while maintaining competitive task accuracy on small or heavily quantized MHA workloads after quantization-aware retraining.
+
+---
+
+## 18. AdamFlow: Adam-based Wasserstein Gradient Flows for Surface Registration in Medical Imaging
+
+- **arXiv ID:** `2604.02290`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `math.OC`
+- **Published:** 2026-04-02
+- **Authors:** Qiang Ma, Qingjie Meng, Xin Hu, Yicheng Wu, Wenjia Bai
+- **Abstract page:** [https://arxiv.org/abs/2604.02290](https://arxiv.org/abs/2604.02290)
+- **PDF:** [https://arxiv.org/pdf/2604.02290v1](https://arxiv.org/pdf/2604.02290v1)
+
+### Abstract
+
+Surface registration plays an important role for anatomical shape analysis in medical imaging. Existing surface registration methods often face a trade-off between efficiency and robustness. Local point matching methods are computationally efficient, but vulnerable to noise and initialisation. Methods designed for global point set alignment tend to incur a high computational cost. To address the challenge, here we present a fast surface registration method, which formulates surface meshes as probability measures and surface registration as a distributional optimisation problem. The discrepancy between two meshes is measured using an efficient sliced Wasserstein distance with log-linear computational complexity. We propose a novel optimisation method, AdamFlow, which generalises the well-known Adam optimisation method from the Euclidean space to the probability space for minimising the sliced Wasserstein distance. We theoretically analyse the asymptotic convergence of AdamFlow and empirically demonstrate its superior performance in both affine and non-rigid surface registration across various anatomical structures.
+
+---
+
+## 19. Omni123: Exploring 3D Native Foundation Models with Limited 3D Data by Unifying Text to 2D and 3D Generation
+
+- **arXiv ID:** `2604.02289`
+- **Primary category:** `cs.CV`
+- **Categories:** `cs.CV`, `cs.AI`
+- **Published:** 2026-04-02
+- **Authors:** Chongjie Ye, Cheng Cao, Chuanyu Pan, Yiming Hao, Yihao Zhi, Yuanming Hu, Xiaoguang Han
+- **Abstract page:** [https://arxiv.org/abs/2604.02289](https://arxiv.org/abs/2604.02289)
+- **PDF:** [https://arxiv.org/pdf/2604.02289v1](https://arxiv.org/pdf/2604.02289v1)
+
+### Abstract
+
+Recent multimodal large language models have achieved strong performance in unified text and image understanding and generation, yet extending such native capability to 3D remains challenging due to limited data. Compared to abundant 2D imagery, high-quality 3D assets are scarce, making 3D synthesis under-constrained. Existing methods often rely on indirect pipelines that edit in 2D and lift results into 3D via optimization, sacrificing geometric consistency. We present Omni123, a 3D-native foundation model that unifies text-to-2D and text-to-3D generation within a single autoregressive framework. Our key insight is that cross-modal consistency between images and 3D can serve as an implicit structural constraint. By representing text, images, and 3D as discrete tokens in a shared sequence space, the model leverages abundant 2D data as a geometric prior to improve 3D representations. We introduce an interleaved X-to-X training paradigm that coordinates diverse cross-modal tasks over heterogeneous paired datasets without requiring fully aligned text-image-3D triplets. By traversing semantic-visual-geometric cycles (e.g., text to image to 3D to image) within autoregressive sequences, the model jointly enforces semantic alignment, appearance fidelity, and multi-view geometric consistency. Experiments show that Omni123 significantly improves text-guided 3D generation and editing, demonstrating a scalable path toward multimodal 3D world models.
+
+---
+
+## 20. Unifying Group-Relative and Self-Distillation Policy Optimization via Sample Routing
+
+- **arXiv ID:** `2604.02288`
+- **Primary category:** `cs.LG`
+- **Categories:** `cs.LG`, `cs.AI`
+- **Published:** 2026-04-02
+- **Authors:** Gengsheng Li, Tianyu Yang, Junfeng Fang, Mingyang Song, Mao Zheng, Haiyun Guo, Dan Zhang, Jinqiao Wang, Tat-Seng Chua
+- **Abstract page:** [https://arxiv.org/abs/2604.02288](https://arxiv.org/abs/2604.02288)
+- **PDF:** [https://arxiv.org/pdf/2604.02288v1](https://arxiv.org/pdf/2604.02288v1)
+
+### Abstract
+
+Reinforcement learning with verifiable rewards (RLVR) has become a standard paradigm for post-training large language models. While Group Relative Policy Optimization (GRPO) is widely adopted, its coarse credit assignment uniformly penalizes failed rollouts, lacking the token-level focus needed to efficiently address specific deviations. Self-Distillation Policy Optimization (SDPO) addresses this by providing denser, more targeted logit-level supervision that facilitates rapid early improvement, yet it frequently collapses during prolonged training. We trace this late-stage instability to two intrinsic flaws: self-distillation on already-correct samples introduces optimization ambiguity, and the self-teacher's signal reliability progressively degrades. To resolve these issues, we propose Sample-Routed Policy Optimization (SRPO), a unified on-policy framework that routes correct samples to GRPO's reward-aligned reinforcement and failed samples to SDPO's targeted logit-level correction. SRPO further incorporates an entropy-aware dynamic weighting mechanism to suppress high-entropy, unreliable distillation targets while emphasizing confident ones. Evaluated across five benchmarks and two model scales, SRPO achieves both the rapid early improvement of SDPO and the long-horizon stability of GRPO. It consistently surpasses the peak performance of both baselines, raising the five-benchmark average on Qwen3-8B by 3.4% over GRPO and 6.3% over SDPO, while simultaneously yielding moderate response lengths and lowering per-step compute cost by up to 17.2%.
+
+---
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/arxiv-cs-multi-2026-04-05.rag_ingested b/docs/AI_Machine_Learning/arxiv-cs-multi-2026-04-05.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..ed0a64962b2c43e1e5eb6493a530b5edb9c83e6c
--- /dev/null
+++ b/docs/AI_Machine_Learning/arxiv-cs-multi-2026-04-05.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "0c80be89fff20996", "ingested_at": "2026-06-30T09:09:37.011146", "content_hash": "0c80be89fff20996"}
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/arxiv-evoskill-query-2026-04-05.md b/docs/AI_Machine_Learning/arxiv-evoskill-query-2026-04-05.md
new file mode 100644
index 0000000000000000000000000000000000000000..189c90c1909bc441751abecf8e5a723e3b7848b9
--- /dev/null
+++ b/docs/AI_Machine_Learning/arxiv-evoskill-query-2026-04-05.md
@@ -0,0 +1,59 @@
+# EvoSkill-related papers (arXiv search)
+
+*Query:* `all:evoskill OR ti:evoskill OR abs:evoskill` · *sort:* relevance
+
+*arXiv total hits:* 3 · *Returned:* 3
+
+*Fetched at (UTC):* 2026-04-05T06:27:05Z
+
+*Machine-readable twin:* `arxiv-evoskill-query-2026-04-05.json`
+
+## Quick orientation
+
+- **EvoSkill** ([2603.02766](https://arxiv.org/abs/2603.02766)) — automated discovery/synthesis of reusable agent skills from failures; multi-agent framing; reported gains on OfficeQA / SealQA; skills can transfer (e.g. to BrowseComp). Open implementation: [sentient-agi/EvoSkill](https://github.com/sentient-agi/EvoSkill).
+- **EvoSkills** ([2604.01687](https://arxiv.org/abs/2604.01687)) — distinct title; self-evolving skills via co-evolutionary verification (read abstract below).
+- **Third result** — query match in abstract; topic is LLM-generated tool-library evaluation (EvolveTool-Bench).
+
+---
+
+## 1. EvoSkill: Automated Skill Discovery for Multi-Agent Systems
+
+- **ID:** `2603.02766` · **Primary:** `cs.AI`
+- **Categories:** `cs.AI`, `cs.MA`
+- **Published:** 2026-03-03
+- **Authors:** Salaheddin Alzubi, Noah Provenzano, Jaydon Bingham, Weiyuan Chen, Tu Vu
+- **abs:** [https://arxiv.org/abs/2603.02766](https://arxiv.org/abs/2603.02766) · **PDF:** [https://arxiv.org/pdf/2603.02766v1](https://arxiv.org/pdf/2603.02766v1)
+
+### Abstract
+
+Coding agents are increasingly used as general-purpose problem solvers, but their flexibility does not by itself confer the domain expertise needed for specialized tasks. Recent work addresses this through \textit{agent skills}: reusable workflows, and code, that augment agents with domain-specific capabilities. Most skills today are hand-crafted, and existing evolutionary approaches optimize low-level artifacts (e.g. prompts code) that are tightly coupled to specific models and tasks. We introduce \textbf{EvoSkill}, a self-evolving framework that automatically discovers and refines agent skills through iterative failure analysis. EvoSkill analyzes execution failures, proposes new skills or edits to existing ones, and materializes them into structured, reusable skill folders. A Pareto frontier of agent programs governs selection, retaining only skills that improve held-out validation performance while the underlying model remains frozen. We evaluate EvoSkill on two benchmarks: OfficeQA, a grounded reasoning benchmark over U.S.\ Treasury data, where it improves exact-match accuracy by \textbf{7.3} (60.6 $\to$ 67.9); and SealQA, a search-augmented QA benchmark with noisy retrieval, where it yields a \textbf{12.1} gain (26.6 $\to$ 38.7). We also investigate the zero-shot transfer capabilties of skills evolved on one task to the other; in particular: skills evolved from SealQA transfers zero-shot to BrowseComp, improving accuracy by \textbf{5.3} without modification demonstrating that skill-level optimization produces transferable capabilities beyond the training task.
+
+---
+
+## 2. EvoSkills: Self-Evolving Agent Skills via Co-Evolutionary Verification
+
+- **ID:** `2604.01687` · **Primary:** `cs.AI`
+- **Categories:** `cs.AI`
+- **Published:** 2026-04-02
+- **Authors:** Hanrong Zhang, Shicheng Fan, Henry Peng Zou, Yankai Chen, Zhenting Wang, Jiayu Zhou, Chengze Li, Wei-Chieh Huang, Yifei Yao, Kening Zheng, Xue Liu, Xiaoxiao Li, Philip S. Yu
+- **abs:** [https://arxiv.org/abs/2604.01687](https://arxiv.org/abs/2604.01687) · **PDF:** [https://arxiv.org/pdf/2604.01687v1](https://arxiv.org/pdf/2604.01687v1)
+
+### Abstract
+
+Anthropic proposes the concept of skills for LLM agents to tackle multi-step professional tasks that simple tool invocations cannot address. A tool is a single, self-contained function, whereas a skill is a structured bundle of interdependent multi-file artifacts. Currently, skill generation is not only label-intensive due to manual authoring, but also may suffer from human--machine cognitive misalignment, which can lead to degraded agent performance, as evidenced by evaluations on SkillsBench. Therefore, we aim to enable agents to autonomously generate skills. However, existing self-evolving methods designed for tools cannot be directly applied to skills due to their increased complexity. To address these issues, we propose EvoSkills, a self-evolving skills framework that enables agents to autonomously construct complex, multi-file skill packages. Specifically, EvoSkills couples a Skill Generator that iteratively refines skills with a Surrogate Verifier that co-evolves to provide informative and actionable feedback without access to ground-truth test content. On SkillsBench, EvoSkills achieves the highest pass rate among five baselines on both Claude Code and Codex, and also exhibits strong generalization capabilities to six additional LLMs.
+
+---
+
+## 3. EvolveTool-Bench: Evaluating the Quality of LLM-Generated Tool Libraries as Software Artifacts
+
+- **ID:** `2604.00392` · **Primary:** `cs.SE`
+- **Categories:** `cs.SE`, `cs.AI`
+- **Published:** 2026-04-01
+- **Authors:** Alibek T. Kaliyev, Artem Maryanskyy
+- **abs:** [https://arxiv.org/abs/2604.00392](https://arxiv.org/abs/2604.00392) · **PDF:** [https://arxiv.org/pdf/2604.00392v1](https://arxiv.org/pdf/2604.00392v1)
+
+### Abstract
+
+Modern LLM agents increasingly create their own tools at runtime -- from Python functions to API clients -- yet existing benchmarks evaluate them almost exclusively by downstream task completion. This is analogous to judging a software engineer only by whether their code runs, ignoring redundancy, regression, and safety. We introduce EvolveTool-Bench, a diagnostic benchmark for LLM-generated tool libraries in software engineering workflows. Across three domains requiring actual tool execution (proprietary data formats, API orchestration, and numerical computation), we define library-level software quality metrics -- reuse, redundancy, composition success, regression stability, and safety -- alongside a per-tool Tool Quality Score measuring correctness, robustness, generality, and code quality. In the first head-to-head comparison of code-level and strategy-level tool evolution (ARISE vs. EvoSkill vs. one-shot baselines, 99 tasks, two models), we show that systems with similar task completion (63-68%) differ by up to 18% in library health, revealing software quality risks invisible to task-only evaluation. Our results highlight that evaluation and governance of LLM-generated tools require treating the evolving tool library as a first-class software artifact, not a black box.
+
+---
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/arxiv-evoskill-query-2026-04-05.rag_ingested b/docs/AI_Machine_Learning/arxiv-evoskill-query-2026-04-05.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..86572c51db6839a024f6fd07743668aa8fb7897c
--- /dev/null
+++ b/docs/AI_Machine_Learning/arxiv-evoskill-query-2026-04-05.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "56c78896ace29d74", "ingested_at": "2026-06-30T09:09:38.465978", "content_hash": "56c78896ace29d74"}
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/photonic_computing_survey.md b/docs/AI_Machine_Learning/photonic_computing_survey.md
new file mode 100644
index 0000000000000000000000000000000000000000..d3b295cc3ec17eb842da0061fe23c12cf9ff1cd5
--- /dev/null
+++ b/docs/AI_Machine_Learning/photonic_computing_survey.md
@@ -0,0 +1,155 @@
+# 🌟 光コンピューティングとAI (Photonic Computing & AI) 技術サーベイ報告書 (2025-2026)
+
+## 📋 エグゼクティブ・サマリー
+
+近年、ディープラーニングや大規模言語モデル (LLM) のパラメータ数およびコンテキスト長の爆発的な増加に伴い、従来のシリコン半導体(GPU/NPU等)は電気的データ移動の遅延と電力消費の限界、いわゆる「メモリの壁 (Memory Wall)」や「帯域幅 of 壁 (Bandwidth Wall)」に直面しています。
+
+本報告書は、2025年〜2026年に発表された最新のarXiv論文から、この課題を解決するために注目を集める**光コンピューティング (Photonic Computing)** および **光ニューラルネットワーク (PNN / ONN)** に関する重要な研究15本を抽出し、その技術動向と今後の展望をまとめたものです。
+
+現在、光AI技術は「静的な行列演算の高速化」という初期段階から、**「動的なトランスフォーマー推論の加速」「分散学習における光ネットワーク内演算」「物理設計自動化 (EPDA)」「量子・古典ハイブリッドQML」**といった、より実用的かつ大規模なシステム統合のフェーズへと急速にシフトしています。
+
+---
+
+## 🗺️ 技術マップ (Taxonomy of Photonic AI)
+
+```mermaid
+graph TD
+ A["光AIシステム (2025-2026)"] --> B["1. Transformer & LLM 加速"]
+ A --> C["2. 分散学習 & 光ネットワーク内演算"]
+ A --> D["3. アナログ光ニューラルネットワーク"]
+ A --> E["4. 量子機械学習 (QML) & 最適化"]
+ A --> F["5. AI駆動型設計自動化 (EPDA)"]
+
+ B --> B1["PRISM (KVキャッシュのO(1)光ブロック選択)"]
+ B --> B2["SimPhony (クロスレイヤー協調設計)"]
+
+ C --> C1["OptINC (光ファイバー内での勾配平均化)"]
+
+ D --> D1["ReDON (自己変調非線形性の再構成)"]
+ D --> D2["LuxIA (高速なPNNトレーニング)"]
+ D --> D3["Hebbian PNN (PCMシナプスによる全光学習)"]
+ D --> D4["Diffusion Accelerators (拡散モデルの加速)"]
+
+ E --> E1["PQKD (量子支援知識蒸留)"]
+ E --> E2["MerLin (QMLシミュレーションエンジン)"]
+ E --> E3["Photonic Ising Machine (イジングモデルの光シミュレーション)"]
+
+ F --> F1["OPTIAGENT (LLMによるレンズ自動設計)"]
+ F --> F2["SimPhony/Apollo (電子・光共設計EPDAツール)"]
+```
+
+---
+
+## 🔬 分野別・主要論文解説
+
+### 1. Transformer & LLM 推論の高速化 (Transformer & LLM Acceleration)
+
+LLMの長文コンテキスト推論において、GPUは計算そのものよりも、トークン生成ごとにKVキャッシュ全体をメモリからスキャンするメモリ帯域幅によってボトルネックが発生します。光の超高速なファンアウト特性を活かしたアプローチが登場しています。
+
+* **PRISM: Breaking the O(n) Memory Wall in Long-Context LLM Inference via O(1) Photonic Block Selection** (arXiv:2603.21576)
+ * **著者:** Hyoseok Park, Yeonsang Park (2026-03-23)
+ * **概要:** LLMのコンテキスト長 \(N\) が伸びるにつれ、KVキャッシュの選択コストは線形 \(O(N)\) に増加します。本研究は、 coarse なブロック選択(類似性検索)が「光のブロードキャスト&ウェイト・パラダイム」に完全に合致することを発見しました。薄膜リチウムナイオベート (TFLN) を用いた類似性エンジン「PRISM」を提案し、64Kトークンのコンテキストにおいて、100%のニードル精度、16倍の通信トラフィック削減、GPU基盤に対して4桁以上のエネルギー優位性を実証しました。
+* **Toward Large-Scale Photonics-Empowered AI Systems: From Physical Design Automation to System-Algorithm Co-Exploration** (arXiv:2601.00129)
+ * **著者:** Ziang Yin, Hongjian Zhou, Jiaqi Gu 他 (2025-12-31)
+ * **概要:** 大規模光AIシステム実現のための3つの要件(動的テンソル演算のサポート、A/D変換やデータ移動オーバーヘッドの最小化、ハードウェアの非理想性への堅牢性)を提示。これらを定量的に評価するため、初期探索からレイアウト生成までをカバーするクロスレイヤーツールチェーン (SimPhony, ADEPT, Apollo) を構築しました。
+
+---
+
+### 2. 分散学習と光ネットワーク内演算 (Distributed Learning & In-Network Computing)
+
+複数の計算ノードを繋ぐ光ファイバー自体で、データ転送と同時に「加算や量子化」などの演算を光のまま行う新しいインフラ技術です。
+
+* **OptINC: Optical In-Network-Computing for Scalable Distributed Learning** (arXiv:2603.28290)
+ * **著者:** Sijie Fei, Grace Li Zhang, Bing Li 他 (2026-03-30)
+ * **概要:** 分散学習における主要なボトルネックである勾配の同期処理(Ring All-Reduceなど)の通信オーバーヘッドを削減するため、光インターコネクト(光ファイバー)自体にマッハ・ツェンダー干渉計 (MZI) などの光デバイスを組み込み、**ネットワーク内で勾配の平均化と量子化を直接実行 (In-Network Computing)** するアーキテクチャ「OptINC」を提案。LLaMAベースのモデルやResNetの学習で、精度を損なうことなく通信オーバーヘッドを実質ゼロにできることを評価しました。
+
+---
+
+### 3. アナログ光ニューラルネットワーク (Analog PNNs & Diffractive Processing)
+
+回折格子やPCM(相変化材料)などの物理デバイスを用い、超低消費電力・低遅延でニューラルネットワークの順方向計算を行う手法です。
+
+* **ReDON: Recurrent Diffractive Optical Neural Processor with Reconfigurable Self-Modulated Nonlinearity** (arXiv:2602.23616)
+ * **著者:** Ziang Yin, Qi Jing, Jiaqi Gu 他 (2026-02-27)
+ * **概要:** 従来の回折型光ニューラルネットワーク (DONN) は、静的な位相マスクを使用するため非線形性の表現力と再プログラム性に課題がありました。本研究では、LLMのGated Linear Unit (GLU) に着想を得て、光の一部を検出して位相/強度をその場で電気光学変調する「自己変調非線形性」を備えた「ReDON」を提案。画像認識・セグメンテーションタスクにおいて、従来のDONNと比較してテスト精度やmIoUを最大20%向上させました。
+* **LuxIA: A Lightweight Unitary Matrix-Based Framework Built on an Iterative Algorithm for Photonic Neural Network Training** (arXiv:2512.22264)
+ * **著者:** Tzamn Melendez Carmona, Alessandro Savino 他 (2025-12-24)
+ * **概要:** 光ニューラルネットワーク (PNN) の伝達行列計算は非常にメモリを消費し、大規模なシミュレーションと逆伝播学習が困難でした。本研究では「Slicing method」と呼ばれる効率的な行列計算手法を提案し、これを統合した学習フレームワーク「LuxIA」を開発。従来ツールに比べ、メモリ消費と実行時間を劇的に削減し、大規模PNNの学習を可能にしました。
+* **Toward all-optical unsupervised Hebbian learning in deep photonic neuromorphic networks** (arXiv:2601.22300)
+ * **著者:** Xi Li, Disha Biswas, Qing Gu 他 (2026-01-29)
+ * **概要:** 外部での勾配計算や電気・光・電気 (O-E-O) 変換を一切行わず、相変化材料 (PCM) シナプスと局所的な光フィードバックを用いて、**光回路内のみで完結するオンライン・非監督ヘブ学習 (Hebbian Learning)** を行うディープ光ニューロモーフィックネットワークを提案。画像認識タスクにおいて、光回路が自律的にパターンを学習・符号化できることを実証しました。
+* **Accelerating Diffusion Models for Generative AI Applications with Silicon Photonics** (arXiv:2603.07626)
+ * **著者:** Tharini Suresh, Salma Afifi, Sudeep Pasricha (2026-03-08)
+ * **概要:** 生成AIの主流である拡散モデル (Diffusion Models) はUNetや注意機構の繰り返し計算が必要で、消費電力が極めて大きい問題があります。本研究は、シリコンフォトニクスに基づく拡散モデル専用のアクセラレータを提案し、従来のデジタル専用アクセラレータと比較して、少なくとも**3倍以上のエネルギー効率と5.5倍以上のスループット向上**を達成しました。
+
+---
+
+### 4. 量子機械学習 (QML) と最適化 (Quantum QML & Ising Optimization)
+
+光の量子状態やコヒーレンスを利用し、確率的探索や組合せ最適化問題を極めて高速に解く技術です。
+
+* **Photonic Quantum-Enhanced Knowledge Distillation** (arXiv:2603.14898)
+ * **著者:** Kuan-Cheng Chen, Ian Walmsley 他 (2026-03-16)
+ * **概要:** 光量子プロセッサが本質的に持つ確率的な測定ゆらぎ(ショット雑音)を「ハードウェア由来の構造化された乱数ソース」として利用し、知識蒸留 (Knowledge Distillation) に応用する「PQKD」フレームワークを提案。少ないパラメータ数の生徒モデルに適切な正則化(構造化ランダムネス)を適用し、圧縮限界に近い性能を維持します。
+* **MerLin: A Discovery Engine for Photonic and Hybrid Quantum Machine Learning** (arXiv:2602.11092)
+ * **著者:** Cassandre Notton, Jean Senellart 他 (2026-02-11)
+ * **概要:** 線形光回路の高速シミュレータをPyTorchやscikit-learnワークフローと統合したオープンソースのディスカバリーエンジン「MerLin」を開発。既存の著名な光量子機械学習 (QML) 論文18本の実験セットをモジュール化して再現し、ベンチマーク環境を提供します。
+* **Incorporating rank-free coupling and external field via an incoherent modulated spatial photonic Ising machine** (arXiv:2512.21587)
+ * **著者:** Ze Zheng, Guihua Zeng 他 (2025-12-25)
+ * **概要:** 振幅のみを変調する「ランクフリー空間光イジングマシン」を構築し、797スピンの完全結合イジングモデルを約9ビットの精度、200Hzの反復レートで直接実行。42万頂点を超える大規模グラフのMax-Cut問題を光学的に解決し、コヒーレント光イジングマシンのスケーラビリティを証明しました。
+
+---
+
+### 5. AI駆動型設計自動化 (EPDA / Optical Design)
+
+物理的なレンズの最適化や、電子・光集積回路のレイアウト設計をAIによって自動化する最先端ツールフローです。
+
+* **OPTIAGENT: A Physics-Driven Agentic Framework for Automated Optical Design** (arXiv:2602.23761)
+ * **著者:** Yuyu Geng, Kaiwei Wang 他 (2026-02-27)
+ * **概要:** 光学設計は極めて複雑な非凸最適化問題であり、専門知識が必要です。本研究は、大規模言語モデル (LLM) に物理シミュレータと最適化ルーチンを組み合わせたエージェントフレームワーク「OPTIAGENT」を構築。強化学習手法 (DrGRPO) と光学原理に基づく報酬設計を組み合わせ、専門知識のないユーザーでも自動で高性能なレンズシステムを設計できる環境を実現しました。
+* **Democratizing Electronic-Photonic AI Systems: An Open-Source AI-Infused Cross-Layer Co-Design and Design Automation Toolflow** (arXiv:2601.00130)
+ * **著者:** Hongjian Zhou, Ziang Yin, Jiaqi Gu (2025-12-31)
+ * **概要:** 物理デバイス、回路、システムアーキテクチャ、AIアルゴリズムを統合的にコデザインするためのオープンソースEPDAフレームワークを提示。AIを用いたマクスウェルソルバーや製造時の誤差を考慮した逆設計 (Inverse Design) アルゴリズムをサポートします。
+
+---
+
+## 📊 論文一覧サマリーテーブル
+
+| # | 論文ID | 論文タイトル | 主なカテゴリー | 提案手法 / 技術 | 発表日 |
+| :--- | :--- | :--- | :--- | :--- | :--- |
+| 1 | [2603.21576](https://arxiv.org/abs/2603.21576) | **PRISM: Breaking the O(n) Memory Wall...** | Transformer & LLM | 薄膜リチウムナイオベートによるO(1)光ブロック選択 (KVキャッシュ用) | 2026-03-23 |
+| 2 | [2603.28290](https://arxiv.org/abs/2603.28290) | **OptINC: Optical In-Network-Computing...** | 分散学習 / ネットワーク | MZI集積光ファイバー内での勾配平均化と量子化 | 2026-03-30 |
+| 3 | [2602.23616](https://arxiv.org/abs/2602.23616) | **ReDON: Recurrent Diffractive Optical...** | アナログ光AI (DONN) | 電気光学変調を用いた自己変調非線形性とハードウェア再利用 | 2026-02-27 |
+| 4 | [2603.14898](https://arxiv.org/abs/2603.14898) | **Photonic Quantum-Enhanced Knowledge...** | 量子機械学習 (QML) | 光量子のショット雑音を用いた辞書畳み込み用構造化ランダム生成 | 2026-03-16 |
+| 5 | [2512.22264](https://arxiv.org/abs/2512.22264) | **LuxIA: A Lightweight Unitary Matrix...** | シミュレーション・学習 | Slicing methodによるPNN逆伝播の省メモリ・高速化 | 2025-12-24 |
+| 6 | [2601.22300](https://arxiv.org/abs/2601.22300) | **Online unsupervised Hebbian learning...** | アナログ光AI (PNN) | PCMシナプスと局所フィードバックによる全光オンライン学習 | 2026-01-29 |
+| 7 | [2603.07626](https://arxiv.org/abs/2603.07626) | **Accelerating Diffusion Models...** | アナログ光AI (PNN) | シリコンフォトニクスによる拡散モデル (UNet/Attention) の加速 | 2026-03-08 |
+| 8 | [2602.11092](https://arxiv.org/abs/2602.11092) | **MerLin: A Discovery Engine for...** | 量子機械学習 (QML) | PyTorch統合の光量子シミュレータ&18本再現ベンチマーク | 2026-02-11 |
+| 9 | [2602.23761](https://arxiv.org/abs/2602.23761) | **OPTIAGENT: A Physics-Driven Agentic...** | 設計自動化 (EPDA) | LLMエージェント + 物理シミュレータによるレンズ自動設計 | 2026-02-27 |
+| 10 | [2601.00129](https://arxiv.org/abs/2601.00129) | **Toward Large-Scale Photonics-Empowered...** | 設計自動化 / 協調設計 | SimPhony、Apollo、LiDAR等を統合したクロスレイヤーツールフロー | 2025-12-31 |
+| 11 | [2512.21587](https://arxiv.org/abs/2512.21587) | **Incorporating rank-free coupling...** | 量子・イジング | 振幅変調器を用いた完全結合型空間光イジングマシン (797スピン) | 2025-12-25 |
+| 12 | [2603.29131](https://arxiv.org/abs/2603.29131) | **Large-scale nonlinear optical computing...** | アナログ光AI (DONN) | 強度エンコーディングと線形回折によるインコヒーレント光非線形演算 | 2026-03-31 |
+| 13 | [2602.19114](https://arxiv.org/abs/2602.19114) | **Kaiwu-PyTorch-Plugin: Bridging Deep...** | 量子・イジング | Coherent Ising Machine (CIM) を用いたPyTorch用量子・古典統合 | 2026-02-22 |
+| 14 | [2601.00130](https://arxiv.org/abs/2601.00130) | **Democratizing Electronic-Photonic AI...** | 設計自動化 (EPDA) | メタ光ニューラルネットワーク等のオープンソースEPDAフレームワーク | 2025-12-31 |
+| 15 | [2603.09645](https://arxiv.org/abs/2603.09645) | **Noise in Photonic Quantum Machine...** | 量子機械学習 (QML) | PQMLにおけるノイズ源の分類と、学習補正・ノイズ低減戦略の網羅 | 2026-03-10 |
+
+---
+
+## 📈 主要なトレンド分析 (Trend Analysis)
+
+1. **静的重み (Weight-Static) から動的演算 (Dynamic Compute) への脱却**
+ * 従来の光コンピューティングは「一度重みを設定したら固定されたまま画像推論を行う」静的回路が主流でした。しかし、近年のトランスフォーマーモデル(注意機構など)の台頭により、**クエリとキーの動的な行列積を光ドメインでその場(On-the-fly)で解く**必要性が生じています。これに対し、薄膜リチウムナイオベート (TFLN) などの高速光変調器を用いた**PRISM**や、自己変調機能を持つ**ReDON**などが登場し、動的な動的テンソル演算が可能になりつつあります。
+2. **電子・光協調設計 (Electronic-Photonic Co-Design) のツール化**
+ * デバイス単体の性能向上だけでなく、システム全体(A/D変換器やメモリ転送オーバーヘッドを含む)の最適化が焦点となっています。2025年末以降、**SimPhony**や**Apollo**、**MerLin**などのオープンソースツールが登場したことで、従来のPyTorchなどといったAIエンジニアの使い慣れたソフトウェアエコシステムと光物理シミュレータが融合し、**アルゴリズムと物理回路の協調探索 (EPDA)** が劇的に容易になりました。
+3. **光の超高速ファンアウトを活かした「ブロック選択」**
+ * KVキャッシュ全体の計算をすべて光で行うのではなく、メモリ帯域幅がボトルネックになる「 coarse な検索(ブロックの絞り込み)」のみを光のブロードキャストで瞬時に行い、精密な計算は電子回路(GPU)で行うといった、**ハイブリッドな適材適所の処理**が現実的なソリューションとして定着しつつあります(**PRISM**)。
+4. **ネットワークと演算の融合 (Optical In-Network Computing)**
+ * 分散学習時の「データ通信路」そのものを「演算回路」に変えてしまうというパラダイム(**OptINC**)は、従来のデータ転送時間のオーバーヘッドを完全に無効化する画期的な技術であり、数千ノードを接続する次世代スーパーコンピュータやLLM学習ファームの基盤となる可能性を秘めています。
+
+---
+
+> [!NOTE]
+> **結論**:
+> 2025年〜2026年の技術進展により、光コンピューティングは「特殊用途のデモチップ」から「大規模言語モデル (LLM) や拡散モデルなどの最先端AIアルゴリズムを物理的・システム的に統合して支えるインフラ」へと実用化が進んでいます。特に、EPDAツールチェーンの標準化と、TFLN等の新素材デバイスの進化が、この発展を支える強力な両輪となっています。
+
+*報告書作成日: 2026年6月22日 (Antigravity)*
diff --git a/docs/AI_Machine_Learning/photonic_computing_survey.rag_ingested b/docs/AI_Machine_Learning/photonic_computing_survey.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..3710ec027cff6730e8d65ecfde94e9852fe2d89d
--- /dev/null
+++ b/docs/AI_Machine_Learning/photonic_computing_survey.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "a4e00889ba6edba2", "ingested_at": "2026-06-30T09:09:44.577131", "content_hash": "a4e00889ba6edba2"}
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/related-papers-multilingual-hybrid-report-2026-04-05.md b/docs/AI_Machine_Learning/related-papers-multilingual-hybrid-report-2026-04-05.md
new file mode 100644
index 0000000000000000000000000000000000000000..8072b8e8a1fea6961b9d9e06e6d71e82c549bfc1
--- /dev/null
+++ b/docs/AI_Machine_Learning/related-papers-multilingual-hybrid-report-2026-04-05.md
@@ -0,0 +1,195 @@
+# Related papers — multilingual hybrid report (TranslateBlue)
+
+*Source report:* `TranslateBlue/Docs/Reports/multilingual-hybrid-report.md`
+*(Query path `docs/reports/...` was not in this repo; file lives under TranslateBlue.)*
+
+*Machine-readable twin:* `related-papers-multilingual-hybrid-report-2026-04-05.json`
+
+*Themes matched:* NMT+NLLB/MADLAD, LLM post-editing & self-correction, quality estimation & routing, instruction/LoRA-style adaptation, low-resource & cultural tuning, automatic post-editing._
+
+---
+
+## 1. No Language Left Behind: Scaling Human-Centered Machine Translation
+
+- **arXiv:** `2207.04672` · **Primary:** `cs.CL`
+- **Published:** 2022-07-11
+- **abs:** [https://arxiv.org/abs/2207.04672](https://arxiv.org/abs/2207.04672) · **PDF:** [https://arxiv.org/pdf/2207.04672v3](https://arxiv.org/pdf/2207.04672v3)
+- **Why related:** Foundational massively multilingual NMT (NLLB); aligns with NLLB/MADLAD and language coverage discussion.
+
+### Abstract
+
+Driven by the goal of eradicating language barriers on a global scale, machine translation has solidified itself as a key focus of artificial intelligence research today. However, such efforts have coalesced around a small subset of languages, leaving behind the vast majority of mostly low-resource languages. What does it take to break the 200 language barrier while ensuring safe, high quality results, all while keeping ethical considerations in mind? In No Language Left Behind, we took on this challenge by first contextualizing the need for low-resource language translation support through exploratory interviews with native speakers. Then, we created datasets and models aimed at narrowing the performance gap between low and high-resource languages. More specifically, we developed a conditional compute model based on Sparsely Gated Mixture of Experts that is trained on data obtained with novel and effective data mining techniques tailored for low-resource languages. We propose multiple architectural and training improvements to counteract overfitting while training on thousands of tasks. Critically, we evaluated the performance of over 40,000 different translation directions using a human-translated benchmark, Flores-200, and combined human evaluation with a novel toxicity benchmark covering all languages in Flores-200 to assess translation safety. Our model achieves an improvement of 44% BLEU relative to the previous state-of-the-art, laying important groundwork towards realizing a universal translation system. Finally, we open source all contributions described in this work, accessible at [https://github.com/facebookresearch/fairseq/tree/nllb](https://github.com/facebookresearch/fairseq/tree/nllb).
+
+---
+
+## 2. MADLAD-400: A Multilingual And Document-Level Large Audited Dataset
+
+- **arXiv:** `2309.04662` · **Primary:** `cs.CL`
+- **Published:** 2023-09-09
+- **abs:** [https://arxiv.org/abs/2309.04662](https://arxiv.org/abs/2309.04662) · **PDF:** [https://arxiv.org/pdf/2309.04662v1](https://arxiv.org/pdf/2309.04662v1)
+- **Why related:** Large multilingual document-level dataset (MADLAD-400); context for NMT data and long-form translation.
+
+### Abstract
+
+We introduce MADLAD-400, a manually audited, general domain 3T token monolingual dataset based on CommonCrawl, spanning 419 languages. We discuss the limitations revealed by self-auditing MADLAD-400, and the role data auditing had in the dataset creation process. We then train and release a 10.7B-parameter multilingual machine translation model on 250 billion tokens covering over 450 languages using publicly available data, and find that it is competitive with models that are significantly larger, and report the results on different domains. In addition, we train a 8B-parameter language model, and assess the results on few-shot translation. We make the baseline models available to the research community.
+
+---
+
+## 3. Memory-efficient NLLB-200: Language-specific Expert Pruning of a Massively Multilingual Machine Translation Model
+
+- **arXiv:** `2212.09811` · **Primary:** `cs.CL`
+- **Published:** 2022-12-19
+- **abs:** [https://arxiv.org/abs/2212.09811](https://arxiv.org/abs/2212.09811) · **PDF:** [https://arxiv.org/pdf/2212.09811v3](https://arxiv.org/pdf/2212.09811v3)
+- **Why related:** Expert pruning / memory-efficient NLLB; relates to on-device capacity and model compression.
+
+### Abstract
+
+The recently released NLLB-200 is a set of multilingual Neural Machine Translation models that cover 202 languages. The largest model is based on a Mixture of Experts architecture and achieves SoTA results across many language pairs. It contains 54.5B parameters and requires at least four 32GB GPUs just for inference. In this work, we propose a pruning method that enables the removal of up to 80% of experts without further finetuning and with a negligible loss in translation quality, which makes it feasible to run the model on a single 32GB GPU. Further analysis suggests that our pruning metrics can identify language-specific experts.
+
+---
+
+## 4. (metadata fetch failed)
+
+- **arXiv:** `2406.02267` · **Primary:** `None`
+- **Published:**
+- **abs:** None · **PDF:** None
+- **Why related:** LLM self-correcting MT with human-style error signals; parallels NMT+LLM refinement and quality loops.
+
+### Abstract
+
+---
+
+## 5. Guiding In-Context Learning of LLMs through Quality Estimation for Machine Translation
+
+- **arXiv:** `2406.07970` · **Primary:** `cs.CL`
+- **Published:** 2024-06-12
+- **abs:** [https://arxiv.org/abs/2406.07970](https://arxiv.org/abs/2406.07970) · **PDF:** [https://arxiv.org/pdf/2406.07970v3](https://arxiv.org/pdf/2406.07970v3)
+- **Why related:** Quality estimation guides LLM in-context MT; relates to routing and when to invoke stronger models.
+
+### Abstract
+
+The quality of output from large language models (LLMs), particularly in machine translation (MT), is closely tied to the quality of in-context examples (ICEs) provided along with the query, i.e., the text to translate. The effectiveness of these ICEs is influenced by various factors, such as the domain of the source text, the order in which the ICEs are presented, the number of these examples, and the prompt templates used. Naturally, selecting the most impactful ICEs depends on understanding how these affect the resulting translation quality, which ultimately relies on translation references or human judgment. This paper presents a novel methodology for in-context learning (ICL) that relies on a search algorithm guided by domain-specific quality estimation (QE). Leveraging the XGLM model, our methodology estimates the resulting translation quality without the need for translation references, selecting effective ICEs for MT to maximize translation quality. Our results demonstrate significant improvements over existing ICL methods and higher translation performance compared to fine-tuning a pre-trained language model (PLM), specifically mBART-50.
+
+---
+
+## 6. MTUncertainty: Assessing the Need for Post-editing of Machine Translation Outputs by Fine-tuning OpenAI LLMs
+
+- **arXiv:** `2308.00158` · **Primary:** `cs.CL`
+- **Published:** 2023-07-31
+- **abs:** [https://arxiv.org/abs/2308.00158](https://arxiv.org/abs/2308.00158) · **PDF:** [https://arxiv.org/pdf/2308.00158v6](https://arxiv.org/pdf/2308.00158v6)
+- **Why related:** Uses LLMs to predict need for post-editing; fits hybrid pipelines (NMT then fix).
+
+### Abstract
+
+Translation Quality Evaluation (TQE) is an essential step of the modern translation production process. TQE is critical in assessing both machine translation (MT) and human translation (HT) quality without reference translations. The ability to evaluate or even simply estimate the quality of translation automatically may open significant efficiency gains through process optimisation. This work examines whether the state-of-the-art large language models (LLMs) can be used for this purpose. We take OpenAI models as the best state-of-the-art technology and approach TQE as a binary classification task. On eight language pairs including English to Italian, German, French, Japanese, Dutch, Portuguese, Turkish, and Chinese, our experimental results show that fine-tuned gpt3.5 can demonstrate good performance on translation quality prediction tasks, i.e. whether the translation needs to be edited. Another finding is that simply increasing the sizes of LLMs does not lead to apparent better performances on this task by comparing the performance of three different versions of OpenAI models: curie, davinci, and gpt3.5 with 13B, 175B, and 175B parameters, respectively.
+
+---
+
+## 7. LangMark: A Multilingual Dataset for Automatic Post-Editing
+
+- **arXiv:** `2511.17153` · **Primary:** `cs.CL`
+- **Published:** 2025-11-21
+- **abs:** [https://arxiv.org/abs/2511.17153](https://arxiv.org/abs/2511.17153) · **PDF:** [https://arxiv.org/pdf/2511.17153v1](https://arxiv.org/pdf/2511.17153v1)
+- **Why related:** Multilingual automatic post-editing dataset; directly supports APE / second-pass research.
+
+### Abstract
+
+Automatic post-editing (APE) aims to correct errors in machine-translated text, enhancing translation quality, while reducing the need for human intervention. Despite advances in neural machine translation (NMT), the development of effective APE systems has been hindered by the lack of large-scale multilingual datasets specifically tailored to NMT outputs. To address this gap, we present and release LangMark, a new human-annotated multilingual APE dataset for English translation to seven languages: Brazilian Portuguese, French, German, Italian, Japanese, Russian, and Spanish. The dataset has 206,983 triplets, with each triplet consisting of a source segment, its NMT output, and a human post-edited translation. Annotated by expert human linguists, our dataset offers both linguistic diversity and scale. Leveraging this dataset, we empirically show that Large Language Models (LLMs) with few-shot prompting can effectively perform APE, improving upon leading commercial and even proprietary machine translation systems. We believe that this new resource will facilitate the future development and evaluation of APE systems.
+
+---
+
+## 8. (metadata fetch failed)
+
+- **arXiv:** `2511.07461` · **Primary:** `None`
+- **Published:**
+- **abs:** None · **PDF:** None
+- **Why related:** Explicit dual-stage terminology-aware translation; mirrors report's NMT + second-stage / terminology story.
+
+### Abstract
+
+---
+
+## 9. (metadata fetch failed)
+
+- **arXiv:** `2310.14451` · **Primary:** `None`
+- **Published:**
+- **abs:** None · **PDF:** None
+- **Why related:** Integrating domain terminology with LLMs into MT; related to glossary and specialized translation.
+
+### Abstract
+
+---
+
+## 10. Instruction-tuned Large Language Models for Machine Translation in the Medical Domain
+
+- **arXiv:** `2408.16440` · **Primary:** `cs.CL`
+- **Published:** 2024-08-29
+- **abs:** [https://arxiv.org/abs/2408.16440](https://arxiv.org/abs/2408.16440) · **PDF:** [https://arxiv.org/pdf/2408.16440v2](https://arxiv.org/pdf/2408.16440v2)
+- **Why related:** Instruction-tuned LLMs for domain MT; relates to SLM/LoRA and task-specific tuning.
+
+### Abstract
+
+Large Language Models (LLMs) have shown promising results on machine translation for high resource language pairs and domains. However, in specialised domains (e.g. medical) LLMs have shown lower performance compared to standard neural machine translation models. The consistency in the machine translation of terminology is crucial for users, researchers, and translators in specialised domains. In this study, we compare the performance between baseline LLMs and instruction-tuned LLMs in the medical domain. In addition, we introduce terminology from specialised medical dictionaries into the instruction formatted datasets for fine-tuning LLMs. The instruction-tuned LLMs significantly outperform the baseline models with automatic metrics.
+
+---
+
+## 11. On Instruction-Finetuning Neural Machine Translation Models
+
+- **arXiv:** `2410.05553` · **Primary:** `cs.CL`
+- **Published:** 2024-10-07
+- **abs:** [https://arxiv.org/abs/2410.05553](https://arxiv.org/abs/2410.05553) · **PDF:** [https://arxiv.org/pdf/2410.05553v1](https://arxiv.org/pdf/2410.05553v1)
+- **Why related:** Instruction-finetuning classical NMT models; bridges NMT heads and instruction paradigm.
+
+### Abstract
+
+In this work, we introduce instruction finetuning for Neural Machine Translation (NMT) models, which distills instruction following capabilities from Large Language Models (LLMs) into orders-of-magnitude smaller NMT models. Our instruction-finetuning recipe for NMT models enables customization of translations for a limited but disparate set of translation-specific tasks. We show that NMT models are capable of following multiple instructions simultaneously and demonstrate capabilities of zero-shot composition of instructions. We also show that through instruction finetuning, traditionally disparate tasks such as formality-controlled machine translation, multi-domain adaptation as well as multi-modal translations can be tackled jointly by a single instruction finetuned NMT model, at a performance level comparable to LLMs such as GPT-3.5-Turbo. To the best of our knowledge, our work is among the first to demonstrate the instruction-following capabilities of traditional NMT models, which allows for faster, cheaper and more efficient serving of customized translations.
+
+---
+
+## 12. Shortcomings of LLMs for Low-Resource Translation: Retrieval and Understanding are Both the Problem
+
+- **arXiv:** `2406.15625` · **Primary:** `cs.CL`
+- **Published:** 2024-06-21
+- **abs:** [https://arxiv.org/abs/2406.15625](https://arxiv.org/abs/2406.15625) · **PDF:** [https://arxiv.org/pdf/2406.15625v3](https://arxiv.org/pdf/2406.15625v3)
+- **Why related:** LLM limits in low-resource MT; relevant to low-resource NMT + Llama refiner discussion.
+
+### Abstract
+
+This work investigates the in-context learning abilities of pretrained large language models (LLMs) when instructed to translate text from a low-resource language into a high-resource language as part of an automated machine translation pipeline. We conduct a set of experiments translating Southern Quechua to Spanish and examine the informativity of various types of context retrieved from a constrained database of digitized pedagogical materials (dictionaries and grammar lessons) and parallel corpora. Using both automatic and human evaluation of model output, we conduct ablation studies that manipulate (1) context type (morpheme translations, grammar descriptions, and corpus examples), (2) retrieval methods (automated vs. manual), and (3) model type. Our results suggest that even relatively small LLMs are capable of utilizing prompt context for zero-shot low-resource translation when provided a minimally sufficient amount of relevant linguistic information. However, the variable effects of context type, retrieval method, model type, and language-specific factors highlight the limitations of using even the best LLMs as translation systems for the majority of the world's 7,000+ languages and their speakers.
+
+---
+
+## 13. (metadata fetch failed)
+
+- **arXiv:** `2506.15415` · **Primary:** `None`
+- **Published:**
+- **abs:** None · **PDF:** None
+- **Why related:** Lexical injection / cross-lingual alignment in Llama-family MT; LoRA/adapter-adjacent.
+
+### Abstract
+
+---
+
+## 14. (metadata fetch failed)
+
+- **arXiv:** `2603.19931` · **Primary:** `None`
+- **Published:**
+- **abs:** None · **PDF:** None
+- **Why related:** Agent-guided expert tuning for culturally attuned low-resource translation; strategy B cultural notes + agents.
+
+### Abstract
+
+---
+
+## 15. Automatic Post-Editing for Machine Translation
+
+- **arXiv:** `1910.08592` · **Primary:** `cs.CL`
+- **Published:** 2019-10-18
+- **abs:** [https://arxiv.org/abs/1910.08592](https://arxiv.org/abs/1910.08592) · **PDF:** [https://arxiv.org/pdf/1910.08592v1](https://arxiv.org/pdf/1910.08592v1)
+- **Why related:** Classic neural APE survey/line of work; baseline literature for post-editing stack.
+
+### Abstract
+
+Automatic Post-Editing (APE) aims to correct systematic errors in a machine translated text. This is primarily useful when the machine translation (MT) system is not accessible for improvement, leaving APE as a viable option to improve translation quality as a downstream task - which is the focus of this thesis. This field has received less attention compared to MT due to several reasons, which include: the limited availability of data to perform a sound research, contrasting views reported by different researchers about the effectiveness of APE, and limited attention from the industry to use APE in current production pipelines. In this thesis, we perform a thorough investigation of APE as a downstream task in order to: i) understand its potential to improve translation quality; ii) advance the core technology - starting from classical methods to recent deep-learning based solutions; iii) cope with limited and sparse data; iv) better leverage multiple input sources; v) mitigate the task-specific problem of over-correction; vi) enhance neural decoding to leverage external knowledge; and vii) establish an online learning framework to handle data diversity in real-time. All the above contributions are discussed across several chapters, and most of them are evaluated in the APE shared task organized each year at the Conference on Machine Translation. Our efforts in improving the technology resulted in the best system at the 2017 APE shared task, and our work on online learning received a distinguished paper award at the Italian Conference on Computational Linguistics. Overall, outcomes and findings of our work have boost interest among researchers and attracted industries to examine this technology to solve real-word problems.
+
+---
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/related-papers-multilingual-hybrid-report-2026-04-05.rag_ingested b/docs/AI_Machine_Learning/related-papers-multilingual-hybrid-report-2026-04-05.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..6135d1b6fda48df692eff0b43f7a74ab0ab225f8
--- /dev/null
+++ b/docs/AI_Machine_Learning/related-papers-multilingual-hybrid-report-2026-04-05.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "27344e92c96de749", "ingested_at": "2026-06-30T09:09:46.016522", "content_hash": "27344e92c96de749"}
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/script_long.md b/docs/AI_Machine_Learning/script_long.md
new file mode 100644
index 0000000000000000000000000000000000000000..bede9a6f614221de460b8ab5f0dbc6101fbe3c05
--- /dev/null
+++ b/docs/AI_Machine_Learning/script_long.md
@@ -0,0 +1,16 @@
+Good evening. Today's news features a fascinating look into the rapidly evolving world of artificial intelligence and machine learning. We have a deep dive into new research published on arXiv, an online repository for scientific papers. This week, we see a surge in submissions across various disciplines, including computer science, with some particularly groundbreaking advancements.
+
+First up, ActionParty: Multi-Subject Action Binding in Generative Video Games. This paper explores the potential of AI to create more interactive and engaging video games. Researchers from ActionParty are proposing a new approach for creating complex actions within these virtual worlds, potentially revolutionizing how we interact with gaming experiences. The authors have developed a novel method for binding multiple subjects together, allowing for more natural and fluid interactions within the game environment.
+
+Next, Steerable Visual Representations tackles the challenge of image representation in computer vision. This paper explores new techniques for creating visual representations that can be easily manipulated and controlled by AI systems. Researchers are proposing a system that allows for fine-grained control over how images are processed and understood by machine learning algorithms.
+
+We also see exciting developments in language modeling with Grounded Token Initialization for New Vocabulary in LMs for Generative Recommendation. This paper delves into the realm of natural language processing, exploring new ways to improve the accuracy and efficiency of recommendation systems. Researchers propose a novel method for initializing language models with more relevant and context-specific tokens, potentially leading to better recommendations across various domains.
+
+The field of artificial intelligence is also seeing progress in areas like reasoning and learning from data. Batched Contextual Reinforcement: A Task-Scaling Law for Efficient Reasoning examines the potential of reinforcement learning to solve complex problems efficiently. The authors propose a new approach that allows AI systems to learn more effectively by leveraging the power of batch processing, potentially leading to faster and more accurate solutions across various domains.
+
+And finally, we see a focus on diversity in language models with No Single Best Model for Diversity: Learning a Router for Sample Diversity. This paper explores the potential of creating more diverse language models that can better understand and respond to a wider range of user inputs. Researchers propose a novel method for learning a "router" that helps AI systems navigate through different types of data, potentially leading to more accurate and nuanced responses in various applications.
+
+These are just a few examples of the exciting advancements being made in the field of artificial intelligence and machine learning. The research published on arXiv is pushing the boundaries of what's possible, and we can expect to see even more groundbreaking developments in the years to come.
+
+
+This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/script_long.rag_ingested b/docs/AI_Machine_Learning/script_long.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..c6073fd50572298133896ebddd40f60604e6bc2d
--- /dev/null
+++ b/docs/AI_Machine_Learning/script_long.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "24d32ec1b1ea6681", "ingested_at": "2026-06-30T09:09:46.310445", "content_hash": "24d32ec1b1ea6681"}
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/script_long_clean.txt b/docs/AI_Machine_Learning/script_long_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..ba65273f3388c53a149accb5c8f923b44f76c973
--- /dev/null
+++ b/docs/AI_Machine_Learning/script_long_clean.txt
@@ -0,0 +1,15 @@
+Good evening. Today's news features a fascinating look into the rapidly evolving world of artificial intelligence and machine learning. We have a deep dive into new research published on arXiv, an online repository for scientific papers. This week, we see a surge in submissions across various disciplines, including computer science, with some particularly groundbreaking advancements.
+
+First up, ActionParty: Multi-Subject Action Binding in Generative Video Games. This paper explores the potential of AI to create more interactive and engaging video games. Researchers from ActionParty are proposing a new approach for creating complex actions within these virtual worlds, potentially revolutionizing how we interact with gaming experiences. The authors have developed a novel method for binding multiple subjects together, allowing for more natural and fluid interactions within the game environment.
+
+Next, Steerable Visual Representations tackles the challenge of image representation in computer vision. This paper explores new techniques for creating visual representations that can be easily manipulated and controlled by AI systems. Researchers are proposing a system that allows for fine-grained control over how images are processed and understood by machine learning algorithms.
+
+We also see exciting developments in language modeling with Grounded Token Initialization for New Vocabulary in LMs for Generative Recommendation. This paper delves into the realm of natural language processing, exploring new ways to improve the accuracy and efficiency of recommendation systems. Researchers propose a novel method for initializing language models with more relevant and context-specific tokens, potentially leading to better recommendations across various domains.
+
+A Task-Scaling Law for Efficient Reasoning examines the potential of reinforcement learning to solve complex problems efficiently. The authors propose a new approach that allows AI systems to learn more effectively by leveraging the power of batch processing, potentially leading to faster and more accurate solutions across various domains.
+
+And finally, we see a focus on diversity in language models with No Single Best Model for Diversity: Learning a Router for Sample Diversity. This paper explores the potential of creating more diverse language models that can better understand and respond to a wider range of user inputs. Researchers propose a novel method for learning a "router" that helps AI systems navigate through different types of data, potentially leading to more accurate and nuanced responses in various applications.
+
+These are just a few examples of the exciting advancements being made in the field of artificial intelligence and machine learning. The research published on arXiv is pushing the boundaries of what's possible, and we can expect to see even more groundbreaking developments in the years to come.
+
+This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/script_middle.md b/docs/AI_Machine_Learning/script_middle.md
new file mode 100644
index 0000000000000000000000000000000000000000..127deedf3c41ab2aa274c1583188ad30a83462ed
--- /dev/null
+++ b/docs/AI_Machine_Learning/script_middle.md
@@ -0,0 +1,18 @@
+Good evening. Today's news brings us updates on cutting-edge research in artificial intelligence and machine learning. We have a look at new papers submitted to arXiv, covering diverse fields like computer vision, language modeling, and even theoretical physics.
+
+The first paper we're looking at is "ActionParty: Multi-Subject Action Binding in Generative Video Games" by researchers from the University of California, Berkeley. This work explores how AI can be used to create more engaging video games with realistic actions, potentially revolutionizing game design.
+
+Next up is "Steerable Visual Representations," a study on using AI to understand and manipulate images. The authors are exploring new techniques for controlling visual representations in computer vision applications.
+
+We then turn our attention to "Grounded Token Initialization for New Vocabulary in LMs for Generative Recommendation." This paper focuses on improving the efficiency of language models used for recommendation systems, aiming to create more personalized recommendations.
+
+Moving into the realm of reinforcement learning, researchers have developed "Batched Contextual Reinforcement: A Task-Scaling Law for Efficient Reasoning," which could potentially lead to faster and more accurate AI systems in complex tasks.
+
+The next paper delves into "No Single Best Model for Diversity: Learning a Router for Sample Diversity." This research explores how to create AI models that can generate diverse outputs, leading to more robust and adaptable applications across various domains.
+
+We then turn our attention to the field of language modeling with "Beyond the Assistant Turn: User Turn Generation as a Probe of Interaction Awareness in Language Models," which investigates how AI systems can better understand and respond to user input.
+
+Finally, we have "Topological Effects in Neural Network Field Theory" from researchers at the University of California, San Diego. This paper explores the intersection of theoretical physics and AI, potentially leading to new insights into the behavior of neural networks.
+
+
+This is just a glimpse of the exciting advancements happening in the world of AI and machine learning. We'll continue to follow these developments closely as they shape our future.
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/script_middle.rag_ingested b/docs/AI_Machine_Learning/script_middle.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..f714ccbd34e56c225cdc9e5f56f5199dc97d79fd
--- /dev/null
+++ b/docs/AI_Machine_Learning/script_middle.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "c68126c03f4c01b1", "ingested_at": "2026-06-30T09:09:46.783488", "content_hash": "c68126c03f4c01b1"}
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/script_middle_clean.txt b/docs/AI_Machine_Learning/script_middle_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..d034ddf63a051d26e1c4a069f67d9cc3dbeda099
--- /dev/null
+++ b/docs/AI_Machine_Learning/script_middle_clean.txt
@@ -0,0 +1,17 @@
+Good evening. Today's news brings us updates on cutting-edge research in artificial intelligence and machine learning. We have a look at new papers submitted to arXiv, covering diverse fields like computer vision, language modeling, and even theoretical physics.
+
+The first paper we're looking at is "ActionParty: Multi-Subject Action Binding in Generative Video Games" by researchers from the University of California, Berkeley. This work explores how AI can be used to create more engaging video games with realistic actions, potentially revolutionizing game design.
+
+Next up is "Steerable Visual Representations," a study on using AI to understand and manipulate images. The authors are exploring new techniques for controlling visual representations in computer vision applications.
+
+We then turn our attention to "Grounded Token Initialization for New Vocabulary in LMs for Generative Recommendation." This paper focuses on improving the efficiency of language models used for recommendation systems, aiming to create more personalized recommendations.
+
+Moving into the realm of reinforcement learning, researchers have developed "Batched Contextual Reinforcement: A Task-Scaling Law for Efficient Reasoning," which could potentially lead to faster and more accurate AI systems in complex tasks.
+
+The next paper delves into "No Single Best Model for Diversity: Learning a Router for Sample Diversity." This research explores how to create AI models that can generate diverse outputs, leading to more robust and adaptable applications across various domains.
+
+We then turn our attention to the field of language modeling with "Beyond the Assistant Turn: User Turn Generation as a Probe of Interaction Awareness in Language Models," which investigates how AI systems can better understand and respond to user input.
+
+Finally, we have "Topological Effects in Neural Network Field Theory" from researchers at the University of California, San Diego. This paper explores the intersection of theoretical physics and AI, potentially leading to new insights into the behavior of neural networks.
+
+This is just a glimpse of the exciting advancements happening in the world of AI and machine learning. We'll continue to follow these developments closely as they shape our future.
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/script_short.md b/docs/AI_Machine_Learning/script_short.md
new file mode 100644
index 0000000000000000000000000000000000000000..e6d267391a1091b1eb211b33c41404a4afff4642
--- /dev/null
+++ b/docs/AI_Machine_Learning/script_short.md
@@ -0,0 +1,5 @@
+New research is dropping fast! We're talking about cutting-edge AI and machine learning breakthroughs from the arXiv preprint server.
+
+First up: ActionParty in generative video games. Researchers are exploring how to make actions more dynamic and engaging, and they've got a new paper out on this. Then there's Steerable Visual Representations, which is all about making visual representations more flexible and adaptable. And for those interested in language models, there's a new paper on grounded token initialization for better recommendations.
+
+The AI community is buzzing with these papers! You can find the full details of each study on the arXiv website. Stay tuned for more updates as we delve deeper into this exciting world of artificial intelligence. This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/script_short.rag_ingested b/docs/AI_Machine_Learning/script_short.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..2ba8a2cf56ee076c7c46efe3616891738dccced1
--- /dev/null
+++ b/docs/AI_Machine_Learning/script_short.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "a4fdb803e760eea0", "ingested_at": "2026-06-30T09:09:47.636703", "content_hash": "a4fdb803e760eea0"}
\ No newline at end of file
diff --git a/docs/AI_Machine_Learning/script_short_clean.txt b/docs/AI_Machine_Learning/script_short_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..7564dc931ae52ae515064fd6e8faf69a03855075
--- /dev/null
+++ b/docs/AI_Machine_Learning/script_short_clean.txt
@@ -0,0 +1,5 @@
+New research is dropping fast! We're talking about cutting-edge AI and machine learning breakthroughs from the arXiv preprint server.
+
+ActionParty in generative video games. Researchers are exploring how to make actions more dynamic and engaging, and they've got a new paper out on this. Then there's Steerable Visual Representations, which is all about making visual representations more flexible and adaptable. And for those interested in language models, there's a new paper on grounded token initialization for better recommendations.
+
+The AI community is buzzing with these papers! You can find the full details of each study on the arXiv website. Stay tuned for more updates as we delve deeper into this exciting world of artificial intelligence. This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/Cybersecurity/file_encryption_survey.json b/docs/Cybersecurity/file_encryption_survey.json
new file mode 100644
index 0000000000000000000000000000000000000000..ba1decf696435e5291677c4ec30f01481e0bd388
--- /dev/null
+++ b/docs/Cybersecurity/file_encryption_survey.json
@@ -0,0 +1,5 @@
+{
+ "importance": 7,
+ "urgency": 7,
+ "tier": 2
+}
\ No newline at end of file
diff --git a/docs/Cybersecurity/file_encryption_survey.md b/docs/Cybersecurity/file_encryption_survey.md
new file mode 100644
index 0000000000000000000000000000000000000000..9b31813e245da621adf3b854fe6fab53c02c3cf3
--- /dev/null
+++ b/docs/Cybersecurity/file_encryption_survey.md
@@ -0,0 +1,545 @@
+# ファイル暗号化技術サーベイ(2026年版)
+
+> [!NOTE]
+> 本レポートは2026年5月時点の情報に基づく、ファイル暗号化技術に関する包括的なサーベイです。暗号アルゴリズム、ツール、ライブラリ、エンタープライズソリューション、および最新トレンドを網羅します。
+
+---
+
+## 目次
+
+1. [対称暗号アルゴリズム](#1-対称暗号アルゴリズム)
+2. [鍵導出関数(KDF)](#2-鍵導出関数kdf)
+3. [ファイル暗号化ツール](#3-ファイル暗号化ツール)
+4. [透過的ファイル暗号化](#4-透過的ファイル暗号化)
+5. [暗号ライブラリ・フレームワーク](#5-暗号ライブラリフレームワーク)
+6. [エンタープライズソリューション](#6-エンタープライズソリューション)
+7. [耐量子暗号(PQC)](#7-耐量子暗号pqc)
+8. [準同型暗号(HE)](#8-準同型暗号he)
+9. [機密計算(Confidential Computing)](#9-機密計算confidential-computing)
+10. [鍵管理システム(KMS)](#10-鍵管理システムkms)
+11. [技術選定ガイドライン](#11-技術選定ガイドライン)
+12. [今後の展望](#12-今後の展望)
+
+---
+
+## 1. 対称暗号アルゴリズム
+
+ファイル暗号化の基盤となる対称暗号アルゴリズムの比較。2026年現在、**AES-256** と **ChaCha20** が業界標準。
+
+### 1.1 AES vs ChaCha20 比較
+
+| 特性 | AES-256-GCM | ChaCha20-Poly1305 |
+|:---|:---|:---|
+| **種別** | ブロック暗号 | ストリーム暗号 |
+| **鍵長** | 128/192/256 bit | 256 bit |
+| **最適環境** | HW加速対応サーバ・デスクトップ | モバイル・IoT・ソフトウェア実装 |
+| **HW加速** | AES-NI で極めて高速 | 不要(純ソフトウェアで高速) |
+| **実装複雑性** | 高(モード選択・パディング要) | 低(誤用リスクが少ない) |
+| **nonce長** | 96 bit(GCM) | 192 bit(XChaCha20) |
+| **nonce再利用耐性** | ❌ 壊滅的(GCM) | ⚠️ XChaCha20で大幅改善 |
+
+### 1.2 認証付き暗号(AEAD)
+
+> [!IMPORTANT]
+> 2026年のベストプラクティスとして、**素の暗号化(CBC等)は使用禁止**。必ず認証付き暗号(AEAD)を使用すること。
+
+| AEAD方式 | 特徴 | 推奨度 |
+|:---|:---|:---|
+| **AES-256-GCM** | 業界標準、HW加速で最速 | ⭐⭐⭐⭐⭐ |
+| **ChaCha20-Poly1305** | ソフトウェア環境で最速、実装容易 | ⭐⭐⭐⭐⭐ |
+| **XChaCha20-Poly1305** | 拡張nonce(192bit)、ランダムnonce安全 | ⭐⭐⭐⭐⭐ |
+| **AES-256-GCM-SIV** | nonce誤用耐性あり | ⭐⭐⭐⭐ |
+| **AES-256-CBC + HMAC** | レガシー互換、Encrypt-then-MAC必須 | ⭐⭐⭐ |
+
+### 1.3 ベストプラクティス
+
+```
+✅ 必ずAEADモードを使用(機密性+完全性+認証を一体化)
+✅ nonce/IVは絶対に再利用しない(特にAES-GCMでは壊滅的)
+✅ 鍵はCSPRNG(暗号論的擬似乱数生成器)で生成
+✅ HW加速がある環境ではAES-GCM、ない環境ではChaCha20-Poly1305
+✅ 大容量ファイルにはストリーム処理を使用(メモリ効率化)
+❌ ECBモードは絶対に使用しない
+❌ 自前で暗号を実装しない("Don't roll your own crypto")
+```
+
+---
+
+## 2. 鍵導出関数(KDF)
+
+パスワードから暗号鍵を安全に導出するための関数。ブルートフォース攻撃への耐性が重要。
+
+### 2.1 主要KDF比較
+
+| アルゴリズム | ステータス(2026年) | メモリ困難性 | GPU/ASIC耐性 | 推奨用途 |
+|:---|:---|:---|:---|:---|
+| **Argon2id** | 🥇 **ゴールドスタンダード** | ✅ 高 | ✅ 非常に高い | すべての新規プロジェクト |
+| **bcrypt** | ✅ 許容 | ❌ 低い | ⚠️ 中程度 | 既存システムの維持 |
+| **scrypt** | ⚠️ レガシー | ✅ 高 | ✅ 高い | Argon2idに置き換え推奨 |
+| **PBKDF2** | ⚠️ コンプライアンス専用 | ❌ なし | ❌ 低い | FIPS準拠が必須の場合のみ |
+
+### 2.2 Argon2id 推奨パラメータ(OWASP基準)
+
+```
+メモリ (m): 64 MiB (65536 KiB)
+反復回数 (t): 3
+並列度 (p): 1(またはCPUコア数)
+ソルト長: 16バイト以上(ユニークなランダム値)
+ハッシュ長: 32バイト(256bit鍵導出用)
+目標処理時間: 200〜500ms(本番HWにて)
+```
+
+> [!TIP]
+> Argon2idはArgon2i(サイドチャネル耐性)とArgon2d(GPU耐性)のハイブリッド。新規開発では**常にArgon2id**を選択すべき。
+
+---
+
+## 3. ファイル暗号化ツール
+
+### 3.1 用途別ツールマップ
+
+```mermaid
+graph TB
+ subgraph "ディスク全体暗号化"
+ A["VeraCrypt"]
+ B["BitLocker (Windows)"]
+ C["LUKS/dm-crypt (Linux)"]
+ D["FileVault (macOS)"]
+ end
+ subgraph "ファイル・フォルダ暗号化"
+ E["age / rage"]
+ F["GPG/PGP"]
+ G["Picocrypt"]
+ H["AxCrypt"]
+ end
+ subgraph "クラウドストレージ暗号化"
+ I["Cryptomator"]
+ J["rclone crypt"]
+ K["NordLocker"]
+ end
+ subgraph "CLI / スクリプト向け"
+ L["openssl enc"]
+ M["age"]
+ N["gpg"]
+ end
+```
+
+### 3.2 主要ツール詳細比較
+
+| ツール | 暗号方式 | 用途 | OSS | 監査 | プラットフォーム |
+|:---|:---|:---|:---|:---|:---|
+| **VeraCrypt** | AES/Serpent/Twofish/Camellia | ディスク・コンテナ暗号化 | ✅ | ✅ 複数回 | Win/Mac/Linux |
+| **Cryptomator** | AES-256-GCM + SIV | クラウドストレージ暗号化 | ✅ | ✅ Cure53 | Win/Mac/Linux/iOS/Android |
+| **age** | X25519 + ChaCha20-Poly1305 | ファイル暗号化(CLI) | ✅ | ✅ コミュニティ | 全OS(Go実装) |
+| **rage** | 同上(ageフォーマット互換) | ファイル暗号化(CLI) | ✅ | ✅ | 全OS(Rust実装) |
+| **Picocrypt** | XChaCha20 + Argon2id | シンプルなファイル暗号化 | ✅ | ✅ | Win/Mac/Linux |
+| **GPG/PGP** | RSA/AES/CAST5 等 | メール・ファイル暗号化 | ✅ | ✅ | 全OS |
+| **AxCrypt** | AES-256 | ファイル暗号化(GUI) | ❌ 商用 | ✅ | Win/Mac |
+| **NordLocker** | AES-256 + xChaCha20 | クラウド統合暗号化 | ❌ 商用 | ✅ | Win/Mac |
+
+### 3.3 ツール選定フローチャート
+
+```mermaid
+flowchart TD
+ START["何を暗号化したい?"] --> Q1{"ディスク全体?"}
+ Q1 -->|Yes| Q2{"OS は?"}
+ Q2 -->|Windows| BITLOCKER["BitLocker"]
+ Q2 -->|Linux| LUKS["LUKS2 / dm-crypt"]
+ Q2 -->|macOS| FILEVAULT["FileVault"]
+ Q2 -->|クロスプラット| VERACRYPT["VeraCrypt"]
+ Q1 -->|No| Q3{"クラウドに保存?"}
+ Q3 -->|Yes| CRYPTOMATOR["Cryptomator"]
+ Q3 -->|No| Q4{"GUI が必要?"}
+ Q4 -->|Yes| Q5{"商用OK?"}
+ Q5 -->|Yes| AXCRYPT["AxCrypt"]
+ Q5 -->|No| PICOCRYPT["Picocrypt"]
+ Q4 -->|No| Q6{"GPG互換が必要?"}
+ Q6 -->|Yes| GPG["GPG"]
+ Q6 -->|No| AGE["age / rage"]
+```
+
+---
+
+## 4. 透過的ファイル暗号化
+
+アプリケーションの変更なしに暗号化を適用する技術。
+
+### 4.1 比較表
+
+| 技術 | レベル | OS | 暗号方式 | ステータス(2026年) |
+|:---|:---|:---|:---|:---|
+| **BitLocker** | ブロック(FDE) | Windows | AES-128/256-XTS | ✅ **標準推奨** |
+| **LUKS2 / dm-crypt** | ブロック(FDE) | Linux | AES-XTS 等 | ✅ **標準推奨** |
+| **FileVault 2** | ブロック(FDE) | macOS | AES-256-XTS | ✅ **標準推奨** |
+| **fscrypt** | ファイルシステム | Linux (ext4/F2FS) | AES-256-XTS + CTS | ✅ **モダン推奨** |
+| **gocryptfs** | FUSE | Linux/macOS | AES-256-GCM | ✅ 特定用途に有効 |
+| **eCryptfs** | カーネル積層 | Linux | AES-256 | ❌ **非推奨(deprecated)** |
+| **EFS** | NTFS統合 | Windows | AES-256 | ⚠️ レガシー |
+
+> [!WARNING]
+> **eCryptfs** はメンテナンス不足・パフォーマンス問題があり、新規デプロイでは使用禁止。**fscrypt** に移行すべき。
+
+### 4.2 エンタープライズ考慮事項
+
+- **FDE**(フルディスク暗号化)は物理的盗難対策の基本。TPM連携が必須
+- **ファイルレベル暗号化**はコンテンツとファイル名を暗号化するが、**メタデータ(ファイルサイズ・ディレクトリ構造)は漏洩**する
+- 高セキュリティ環境では、Thales CipherTrust等の専用プラットフォームを検討
+
+---
+
+## 5. 暗号ライブラリ・フレームワーク
+
+### 5.1 汎用暗号ライブラリ
+
+#### Tier 1: 基盤ライブラリ
+
+| ライブラリ | 言語 | 特徴 | PQC対応 | FIPS対応 |
+|:---|:---|:---|:---|:---|
+| **OpenSSL 3.x** | C | 業界標準、TLS基盤、広範なアルゴリズム | ✅ ML-KEM統合中 | ✅ FIPS 140-3 |
+| **BoringSSL** | C | Google維持、Chrome/Android基盤 | ✅ ML-KEM対応 | ✅ |
+| **LibreSSL** | C | OpenBSD由来、安全性重視 | ⚠️ 限定的 | ❌ |
+| **libsodium** | C | モダン・誤用防止設計 | ❌ 未対応 | ❌ |
+| **wolfSSL** | C | 組み込み向け軽量 | ✅ 対応中 | ✅ FIPS 140-3 |
+| **mbedTLS** | C | ARM/IoT向け | ⚠️ 限定的 | ✅ |
+
+#### Tier 2: 高レベルフレームワーク
+
+| フレームワーク | 提供元 | 特徴 | KMS統合 |
+|:---|:---|:---|:---|
+| **Google Tink** | Google | 誤用防止設計、Keyset管理、暗号俊敏性 | ✅ GCP/AWS/Azure |
+| **AWS Encryption SDK** | AWS | エンベロープ暗号化、マルチリージョン | ✅ AWS KMS |
+| **Themis (Cossack Labs)** | OSS | クロスプラットフォーム、14言語対応 | ⚠️ 限定的 |
+
+### 5.2 言語別暗号ライブラリ
+
+#### Python
+| ライブラリ | バックエンド | 推奨度 | 備考 |
+|:---|:---|:---|:---|
+| **cryptography** | OpenSSL | ⭐⭐⭐⭐⭐ | 高レベル+低レベル両対応 |
+| **PyCryptodome** | 独自C実装 | ⭐⭐⭐⭐ | PyCrypto後継 |
+| **PyNaCl** | libsodium | ⭐⭐⭐⭐ | NaCl/libsodiumバインディング |
+
+#### Rust
+| ライブラリ | 実装 | FIPS | 推奨度 | 備考 |
+|:---|:---|:---|:---|:---|
+| **aws-lc-rs** | C/C++ (AWS-LC) | ✅ | ⭐⭐⭐⭐⭐ | 2026年のデファクト、rustlsデフォルト |
+| **ring** | Rust/C/ASM | ❌ | ⭐⭐⭐⭐ | メンテナンス懸念あり(2025年〜) |
+| **RustCrypto** | Pure Rust | ❌ | ⭐⭐⭐⭐ | WASM対応、可搬性最高 |
+| **rage** | Pure Rust | ❌ | ⭐⭐⭐⭐⭐ | ageフォーマット実装 |
+
+#### JavaScript / Node.js
+| ライブラリ | 推奨度 | 備考 |
+|:---|:---|:---|
+| **Web Crypto API** | ⭐⭐⭐⭐⭐ | ブラウザ標準、HW加速対応 |
+| **node:crypto** | ⭐⭐⭐⭐⭐ | Node.js標準、OpenSSLベース |
+| **libsodium.js** | ⭐⭐⭐⭐ | libsodiumのWASM移植 |
+
+#### Go
+| ライブラリ | 推奨度 | 備考 |
+|:---|:---|:---|
+| **crypto/*** | ⭐⭐⭐⭐⭐ | 標準ライブラリ、高品質 |
+| **golang.org/x/crypto** | ⭐⭐⭐⭐⭐ | 拡張暗号(ChaCha20等) |
+| **age (filippo.io/age)** | ⭐⭐⭐⭐⭐ | モダンファイル暗号化 |
+
+#### Java / JVM
+| ライブラリ | 推奨度 | 備考 |
+|:---|:---|:---|
+| **JCA/JCE** | ⭐⭐⭐⭐ | 標準API、プロバイダ切替可能 |
+| **Bouncy Castle** | ⭐⭐⭐⭐⭐ | 最も広範なアルゴリズムサポート |
+| **Google Tink (Java)** | ⭐⭐⭐⭐⭐ | 誤用防止、KMS統合 |
+
+### 5.3 ライブラリ選定マトリクス
+
+```mermaid
+quadrantChart
+ title 暗号ライブラリ選定マトリクス
+ x-axis "低レベル制御" --> "高レベル抽象化"
+ y-axis "限定的なスコープ" --> "広範なスコープ"
+ quadrant-1 "エンタープライズ向け"
+ quadrant-2 "汎用基盤"
+ quadrant-3 "特化型"
+ quadrant-4 "開発者フレンドリー"
+ OpenSSL: [0.2, 0.85]
+ BoringSSL: [0.25, 0.75]
+ libsodium: [0.35, 0.4]
+ Google Tink: [0.8, 0.7]
+ AWS Enc SDK: [0.85, 0.6]
+ RustCrypto: [0.3, 0.5]
+ aws-lc-rs: [0.3, 0.65]
+ Bouncy Castle: [0.25, 0.9]
+```
+
+---
+
+## 6. エンタープライズソリューション
+
+### 6.1 主要プラットフォーム
+
+| ソリューション | 提供元 | アプローチ | 最適環境 |
+|:---|:---|:---|:---|
+| **CipherTrust Data Security Platform** | Thales | データ中心型・透過暗号 | マルチクラウド・ハイブリッド |
+| **Microsoft Purview Information Protection** | Microsoft | ラベルベース・ドキュメント保護 | Microsoft 365エコシステム |
+| **IBM Guardium** | IBM | データベース暗号化・監査 | メインフレーム・ハイブリッド |
+| **Symantec Data Loss Prevention** | Broadcom | DLP統合暗号化 | 大規模エンタープライズ |
+| **Trend Micro Endpoint Encryption** | Trend Micro | エンドポイント暗号化 | マネージドエンドポイント |
+
+### 6.2 Thales CipherTrust vs Microsoft Purview
+
+| 観点 | Thales CipherTrust | Microsoft Purview |
+|:---|:---|:---|
+| **焦点** | データ所在に依存しない保護 | M365エコシステム統合 |
+| **暗号化タイプ** | 透過的・ファイル・ボリューム・アプリ層 | ラベルベース・ドキュメント保護 |
+| **鍵管理** | HSM統合、BYOK/HYOK、集中管理 | Azure Key Vault、Double Key Encryption |
+| **デプロイ** | ハイブリッド/オンプレミス/マルチクラウド | クラウド/SaaS中心 |
+| **規制対応** | 金融・防衛等の厳格な鍵分離要件 | M365環境でのGDPR/DLP |
+
+> [!TIP]
+> **Vormetric** は CipherTrust に統合済み(EOL)。Vormetricユーザーは CipherTrust への移行が必要。
+
+---
+
+## 7. 耐量子暗号(PQC)
+
+### 7.1 NIST PQC標準化状況(2026年5月時点)
+
+```mermaid
+timeline
+ title NIST PQC標準化タイムライン
+ 2024-08 : FIPS 203 (ML-KEM/Kyber) 最終化
+ : FIPS 204 (ML-DSA/Dilithium) 最終化
+ : FIPS 205 (SLH-DSA/SPHINCS+) 最終化
+ 2025-03 : HQC をバックアップアルゴリズムに選定
+ 2026 : HQC ドラフト標準 公開予定
+ 2027 : HQC 最終標準 予定
+```
+
+### 7.2 標準アルゴリズム一覧
+
+| 標準 | アルゴリズム名 | 種別 | 数学的基盤 | 用途 |
+|:---|:---|:---|:---|:---|
+| **FIPS 203** | ML-KEM (Kyber) | 鍵カプセル化 | 格子問題 (MLWE) | 鍵交換・暗号化 |
+| **FIPS 204** | ML-DSA (Dilithium) | デジタル署名 | 格子問題 (MLWE) | 汎用署名 |
+| **FIPS 205** | SLH-DSA (SPHINCS+) | デジタル署名 | ハッシュベース | 署名(ステートレス) |
+| **(Draft)** | FN-DSA (Falcon) | デジタル署名 | 格子問題 (NTRU) | 小署名サイズ |
+| **(選定済)** | HQC | 鍵カプセル化 | 符号ベース | ML-KEMバックアップ |
+
+### 7.3 PQC対応ライブラリ
+
+| ライブラリ | 言語 | 対応アルゴリズム | 備考 |
+|:---|:---|:---|:---|
+| **liboqs (Open Quantum Safe)** | C | ML-KEM, ML-DSA, SLH-DSA等 | 最も包括的なPQC実装 |
+| **PQClean** | C | 全NIST最終候補 | リファレンス実装集 |
+| **CIRCL (Cloudflare)** | Go | ML-KEM, ML-DSA等 | 本番利用実績あり |
+| **OpenSSL 3.x** | C | ML-KEM統合中 | ハイブリッド対応 |
+| **BoringSSL** | C | ML-KEM対応済 | Chrome/Google利用 |
+| **aws-lc-rs** | Rust | ML-KEM対応 | FIPS対応パス有 |
+
+### 7.4 移行戦略
+
+> [!CAUTION]
+> **「Harvest Now, Decrypt Later(HNDL)」脅威**: 攻撃者は現在の暗号化データを収集し、将来の量子コンピュータで復号する可能性がある。**2030年以降も機密性を保つ必要があるデータは、今すぐPQC移行を開始すべき**。
+
+**推奨移行ステップ:**
+
+1. **暗号資産の棚卸し** — RSA/ECCを使用している全システムの特定
+2. **データ感度の優先順位付け** — 長期機密性が必要なデータから着手
+3. **ハイブリッド方式の導入** — 従来暗号(X25519)+ PQC(ML-KEM)の併用
+4. **暗号俊敏性の確保** — アルゴリズムを容易に切り替え可能な設計
+5. **NIST IR 8547参照** — 公式移行ガイダンスに準拠
+
+---
+
+## 8. 準同型暗号(HE)
+
+暗号化したまま計算を可能にする技術。2026年時点で実用化段階に移行中。
+
+### 8.1 主要ライブラリ比較
+
+| ライブラリ | 提供元 | スキーム | 言語 | 最適用途 |
+|:---|:---|:---|:---|:---|
+| **Microsoft SEAL** | Microsoft | BFV, CKKS | C++ / .NET | 高性能エンタープライズ |
+| **OpenFHE** | コンソーシアム(26+組織) | BFV, BGV, CKKS, TFHE | C++ | マルチスキーム研究・実装 |
+| **Concrete** | Zama | TFHE | Python / Rust | AI/ML・ブロックチェーン |
+| **TFHE-rs** | Zama | TFHE | Rust | 高性能TFHE |
+| **Lattigo** | EPFL/Tune Insight | BFV, CKKS, BGV | Go | Go開発者向け |
+
+### 8.2 実用アプリケーション
+
+| 分野 | 応用例 | 主要ライブラリ |
+|:---|:---|:---|
+| **プライバシー保護ML** | 暗号化データ上での推論(医療診断、信用スコア) | Concrete ML, SEAL |
+| **機密ブロックチェーン** | 秘密投票、MEV耐性取引 | Zama fhEVM |
+| **安全なデータ共有** | 銀行間不正検知(GDPR/HIPAA準拠) | OpenFHE, SEAL |
+| **暗号化検索** | ブラインド検索エンジン、ゲノム解析 | OpenFHE |
+
+> [!NOTE]
+> 準同型暗号は通常の暗号化と比較して**10,000〜1,000,000倍**のオーバーヘッドがある。汎用ファイル暗号化には適さず、特定のプライバシー保護計算に限定して使用すべき。
+
+---
+
+## 9. 機密計算(Confidential Computing)
+
+ハードウェアベースの信頼できる実行環境(TEE)による「使用中データ」の保護。
+
+### 9.1 主要ハードウェア技術
+
+| 技術 | 提供元 | 仕組み | 主な特徴 |
+|:---|:---|:---|:---|
+| **Intel TDX** | Intel | SEAM モード + MKTME | VM単位の高保証分離 |
+| **AMD SEV-SNP** | AMD | Secure Processor + SNP | メモリ再マッピング/リプレイ防止 |
+| **Arm CCA** | Arm | Realm Management Extension | モバイル〜サーバまで対応 |
+| **NVIDIA CC** | NVIDIA | GPU機密計算 | AI/ML ワークロード保護 |
+
+### 9.2 データ保護の3層モデル
+
+```mermaid
+graph LR
+ A["保存データ
(Data at Rest)"] -->|AES-XTS, LUKS, BitLocker| B["転送データ
(Data in Transit)"]
+ B -->|TLS 1.3, mTLS| C["使用中データ
(Data in Use)"]
+ C -->|TEE, Intel TDX, AMD SEV| A
+
+ style A fill:#1a5276,stroke:#2980b9,color:#fff
+ style B fill:#1e8449,stroke:#27ae60,color:#fff
+ style C fill:#922b21,stroke:#c0392b,color:#fff
+```
+
+### 9.3 2026年のトレンド
+
+- **Confidential AI**: AI/MLモデルとデータの訓練/推論時保護が主要ユースケース
+- **GPU機密計算**: CPU保護だけでは不十分、NVIDIA CCがGPUワークロードも保護
+- **リフト&シフト**: クラウド各社(GCP, Azure, AWS)が「コード変更不要」な機密VMを提供
+- **プライベートクラウド展開**: VMware Cloud Foundation 9が機密VM対応
+- **標準化**: ベンダー中立の構成証明APIが進行中
+
+---
+
+## 10. 鍵管理システム(KMS)
+
+### 10.1 主要KMSプラットフォーム
+
+| プラットフォーム | 種別 | FIPS 140-3 | 最適環境 | 特徴 |
+|:---|:---|:---|:---|:---|
+| **AWS KMS** | マネージド | ✅ Level 3 | AWSクラウド | HSMフリート、自動ローテーション |
+| **Google Cloud KMS** | マネージド | ✅ | GCPクラウド | Tink統合、EKM対応 |
+| **Azure Key Vault** | マネージド | ✅ | Azureクラウド | Managed HSM、Purview統合 |
+| **HashiCorp Vault** | セルフホスト/マネージド | ✅ (Enterprise) | マルチクラウド | シークレット管理統合、動的シークレット |
+| **Thales Luna HSM** | ハードウェア | ✅ Level 3 | オンプレミス | 最高レベルの鍵保護 |
+| **AWS CloudHSM** | マネージド HSM | ✅ Level 3 | AWS | シングルテナントHSM |
+
+### 10.2 鍵管理アーキテクチャパターン
+
+```mermaid
+graph TB
+ subgraph "エンベロープ暗号化"
+ DEK["データ暗号化鍵 (DEK)"] -->|"暗号化"| DATA["ファイルデータ"]
+ KEK["鍵暗号化鍵 (KEK)"] -->|"暗号化"| DEK
+ ROOT["ルートキー (KMS/HSM)"] -->|"保護"| KEK
+ end
+```
+
+### 10.3 2026年のKMSトレンド
+
+| トレンド | 説明 |
+|:---|:---|
+| **PQC対応準備** | HNDL攻撃への防御として量子耐性鍵管理への投資加速 |
+| **統一ガバナンス** | マルチクラウド横断の一元的な鍵管理コントロールプレーン |
+| **自動化・俊敏性** | 証明書/鍵の短寿命化に伴う自動ライフサイクル管理 |
+| **AI ワークロード保護** | AI/MLモデルの機密データをHW裏付きの分離で保護 |
+| **FIPS 140-3 必須化** | FIPS 140-2の実質的なサンセット、140-3が新基準 |
+
+---
+
+## 11. 技術選定ガイドライン
+
+### 11.1 ユースケース別推奨スタック
+
+| ユースケース | 暗号方式 | KDF | ツール/ライブラリ | 鍵管理 |
+|:---|:---|:---|:---|:---|
+| **個人ファイル暗号化** | XChaCha20-Poly1305 | Argon2id | age / Picocrypt | ローカル鍵ファイル |
+| **チーム間ファイル共有** | AES-256-GCM | Argon2id | GPG / age | 公開鍵基盤 |
+| **クラウドバックアップ** | AES-256-GCM-SIV | Argon2id | Cryptomator / rclone crypt | パスフレーズ |
+| **エンドポイント保護** | AES-256-XTS | N/A (TPM) | BitLocker / LUKS / FileVault | TPM + ADポリシー |
+| **エンタープライズデータ保護** | AES-256-GCM | N/A (HSM) | Thales CipherTrust | HSM / Cloud KMS |
+| **開発者向けアプリ組込み** | AES-GCM / ChaCha20 | N/A | Google Tink / libsodium | Cloud KMS |
+| **長期アーカイブ(>10年)** | AES-256 + ML-KEM (ハイブリッド) | Argon2id | OpenSSL + liboqs | HSM + PQC鍵 |
+
+### 11.2 セキュリティレベル別推奨
+
+````carousel
+### 🟢 基本レベル(個人利用)
+- **暗号化**: AES-256-GCM or ChaCha20-Poly1305
+- **KDF**: Argon2id (デフォルトパラメータ)
+- **ツール**: age, Picocrypt, Cryptomator
+- **鍵管理**: パスフレーズ + ローカル鍵ファイル
+
+### 🟡 強化レベル(企業・組織)
+- **暗号化**: AES-256-GCM (HW加速)
+- **KDF**: Argon2id (強化パラメータ)
+- **ツール**: BitLocker/LUKS + CipherTrust
+- **鍵管理**: Cloud KMS + HSM バックアップ
+- **追加**: 監査ログ、ポリシー管理、FIPS 140-3準拠
+
+### 🔴 最高レベル(政府・防衛・長期機密)
+- **暗号化**: AES-256 + ML-KEM ハイブリッド
+- **KDF**: Argon2id (最大パラメータ) + HSM
+- **ツール**: HSM統合カスタムソリューション
+- **鍵管理**: オンプレHSM + BYOK + 鍵分離
+- **追加**: 機密計算(TEE)、PQC移行完了、ゼロトラスト
+````
+
+---
+
+## 12. 今後の展望
+
+### 12.1 技術動向予測
+
+```mermaid
+gantt
+ title ファイル暗号化技術ロードマップ
+ dateFormat YYYY
+ axisFormat %Y
+
+ section PQC移行
+ NIST標準最終化 (ML-KEM/ML-DSA) :done, 2024, 2025
+ HQCバックアップ標準ドラフト :active, 2026, 2027
+ ハイブリッド暗号の標準化 :2025, 2028
+ PQC専用移行の本格化 :2028, 2035
+
+ section メモリ安全性
+ Rustベース暗号ライブラリの台頭 :active, 2024, 2027
+ C/C++ライブラリからの段階的移行 :2026, 2032
+
+ section 機密計算
+ GPU機密計算の成熟 :active, 2025, 2028
+ クロスベンダー構成証明標準化 :2026, 2029
+
+ section 自動化
+ 暗号俊敏性プラットフォームの普及 :active, 2025, 2028
+ AI駆動の鍵管理 :2026, 2030
+```
+
+### 12.2 主要トレンドサマリ
+
+| トレンド | 影響度 | 緊急度 | 要点 |
+|:---|:---|:---|:---|
+| **PQC移行** | 🔴 極めて高い | 🟡 中(HNDL脅威で高まる) | ハイブリッド方式から段階的に移行、暗号俊敏性の確保が必須 |
+| **メモリ安全な暗号実装** | 🟡 高い | 🟡 中 | Rust (aws-lc-rs, RustCrypto) がC/C++実装を置き換える動き |
+| **誤用防止設計** | 🟡 高い | 🟢 進行中 | Google Tink等の高レベルAPIが標準的な選択肢に |
+| **FIPS 140-3 必須化** | 🟡 高い | 🔴 高い | FIPS 140-2のサンセットに伴い、新基準への準拠が急務 |
+| **Confidential AI** | 🟡 高い | 🟡 中 | AI/MLワークロードの暗号化がTEEで実現される |
+| **準同型暗号の実用化** | 🟢 中 | 🟢 低 | 特定ドメイン(医療、金融)で限定的に実用段階 |
+| **暗号化のデフォルト化** | 🟡 高い | 🟢 進行中 | GDPR等の規制により「暗号化しない理由」が問われる時代 |
+
+> [!IMPORTANT]
+> **2026年の最重要アクション**:
+> 1. 暗号資産の棚卸しと PQC 移行計画の策定
+> 2. FIPS 140-3 準拠の確認と移行
+> 3. 暗号俊敏性(Cryptographic Agility)を設計原則に組み込む
+
+---
+
+*本サーベイは2026年5月時点の公開情報に基づきます。暗号技術は急速に進化するため、NIST、IETF、各ベンダーの最新発表を定期的に確認することを推奨します。*
+
+
+**Scores:** Importance:7, Urgency:7
+**Tier:** 2
diff --git a/docs/Cybersecurity/file_encryption_survey.rag_ingested b/docs/Cybersecurity/file_encryption_survey.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..5829c05a09768d53430e79f6bb35642649c53e05
--- /dev/null
+++ b/docs/Cybersecurity/file_encryption_survey.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "340fa3dba4927a30", "ingested_at": "2026-06-30T09:09:55.826868", "content_hash": "340fa3dba4927a30"}
\ No newline at end of file
diff --git a/docs/Cybersecurity/file_encryption_survey_script_long.md b/docs/Cybersecurity/file_encryption_survey_script_long.md
new file mode 100644
index 0000000000000000000000000000000000000000..0c9fa38cf1695b86195e63acd8dbb0d0bd152e4e
--- /dev/null
+++ b/docs/Cybersecurity/file_encryption_survey_script_long.md
@@ -0,0 +1,56 @@
+This is a great start to a comprehensive guide on encryption tools and techniques. Here's a breakdown of the strengths, potential improvements, and ideas for expansion:
+
+**Strengths:**
+
+* **Clear Structure:** The document uses a clear structure with sections like "Tool Selection Flowchart," "4. Throughput Encryption," and "5. Crypto Libraries." This makes it easy to follow and navigate.
+* **Comprehensive Coverage:** The guide covers various aspects of encryption, including:
+ * Tool selection based on specific needs (e.g., full disk encryption, cloud storage)
+ * Comparison of different encryption methods and technologies
+ * Considerations for enterprise environments
+ * Guidance on choosing the right libraries and frameworks
+* **Focus on Practicality:** The guide provides practical examples and considerations relevant to real-world scenarios.
+
+
+**Potential Improvements & Expansion Ideas:**
+
+1. **Detailed Explanations:** Provide more in-depth explanations of each encryption method, including:
+ * **How it works:** Explain the underlying principles and algorithms used by each method (e.g., AES, RSA, ChaCha20).
+ * **Advantages and disadvantages:** Discuss the strengths and weaknesses of different methods, such as speed, security level, ease of use, and compatibility with various platforms.
+ * **Use Cases:** Give examples of where each method is most suitable (e.g., full disk encryption for physical devices, cloud storage for sensitive data).
+
+2. **Security Considerations:** Go beyond basic technical details and delve into the following security aspects:
+ * **Key Management:** Explain how to securely generate, store, and manage keys for different methods.
+ * **Vulnerabilities:** Discuss potential vulnerabilities associated with each method and mitigation strategies.
+ * **Compliance:** Highlight relevant compliance standards (e.g., GDPR, HIPAA) and how encryption can help meet those requirements.
+
+3. **Advanced Techniques:** Explore more advanced encryption techniques:
+ * **Homomorphic Encryption:** Explain the concept of homomorphic encryption and its potential for privacy-preserving computations.
+ * **Zero-Knowledge Proofs:** Discuss how these proofs can be used to verify data without revealing sensitive information.
+ * **Quantum Computing:** Briefly touch upon the impact of quantum computing on traditional encryption methods and the need for post-quantum cryptography (PQC).
+
+4. **Real-World Examples:** Include real-world examples of how different encryption techniques are being used in various industries:
+ * **Financial Services:** Secure transactions, data storage, and customer privacy.
+ * **Healthcare:** Protecting patient records, medical images, and research data.
+ * **Government:** National security, confidential communications, and digital identity.
+
+5. **Tools & Resources:** Provide a list of useful tools for encryption:
+ * Open source libraries (e.g., OpenSSL, libsodium)
+ * Cloud-based encryption services (e.g., AWS Encryption SDK, Google Cloud Key Management Service)
+ * Enterprise-grade security solutions
+
+6. **Future Trends:** Discuss emerging trends in encryption and how they might impact the future of data protection:
+ * **Privacy-enhancing technologies (PETs):** Explain how PETs like differential privacy and federated learning can be used to protect user data while still allowing for analysis.
+ * **Blockchain technology:** Discuss how blockchain can be used for secure data storage and transactions.
+
+
+**Visualizations:**
+
+* Use diagrams, charts, and tables to make the information more accessible and visually appealing. For example:
+ * A flowchart illustrating different encryption methods and their use cases.
+ * A comparison table highlighting key features of various libraries and frameworks.
+ * A timeline showing the evolution of encryption technologies over time.
+
+
+**Conclusion:**
+
+This guide provides a solid foundation for understanding encryption tools and techniques. By expanding on these areas, you can create a comprehensive resource that helps readers make informed decisions about data security.
\ No newline at end of file
diff --git a/docs/Cybersecurity/file_encryption_survey_script_long_clean.txt b/docs/Cybersecurity/file_encryption_survey_script_long_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..ff46a217aff5d5ecb74adf08680ba9a638dda398
--- /dev/null
+++ b/docs/Cybersecurity/file_encryption_survey_script_long_clean.txt
@@ -0,0 +1,73 @@
+This is a great start to a comprehensive guide on encryption tools and techniques. Here's a breakdown of the strengths, potential improvements, and ideas for expansion:
+
+Clear Structure: The document uses a clear structure with sections like "Tool Selection Flowchart," "4. Throughput Encryption," and "5. Crypto Libraries." This makes it easy to follow and navigate.
+
+Comprehensive Coverage: The guide covers various aspects of encryption, including:
+
+Tool selection based on specific needs
+
+Comparison of different encryption methods and technologies
+
+Considerations for enterprise environments
+
+Guidance on choosing the right libraries and frameworks
+
+Focus on Practicality: The guide provides practical examples and considerations relevant to real-world scenarios.
+
+Potential Improvements & Expansion Ideas:
+
+1. Detailed Explanations: Provide more in-depth explanations of each encryption method, including:
+
+How it works: Explain the underlying principles and algorithms used by each method .
+
+Advantages and disadvantages: Discuss the strengths and weaknesses of different methods, such as speed, security level, ease of use, and compatibility with various platforms.
+
+Use Cases: Give examples of where each method is most suitable .
+
+2. Security Considerations: Go beyond basic technical details and delve into the following security aspects:
+
+Key Management: Explain how to securely generate, store, and manage keys for different methods.
+
+Vulnerabilities: Discuss potential vulnerabilities associated with each method and mitigation strategies.
+
+Compliance: Highlight relevant compliance standards and how encryption can help meet those requirements.
+
+3. Advanced Techniques: Explore more advanced encryption techniques:
+
+Homomorphic Encryption: Explain the concept of homomorphic encryption and its potential for privacy-preserving computations.
+
+Zero-Knowledge Proofs: Discuss how these proofs can be used to verify data without revealing sensitive information.
+
+Quantum Computing: Briefly touch upon the impact of quantum computing on traditional encryption methods and the need for post-quantum cryptography .
+
+4. Real-World Examples: Include real-world examples of how different encryption techniques are being used in various industries:
+
+Financial Services: Secure transactions, data storage, and customer privacy.
+
+Healthcare: Protecting patient records, medical images, and research data.
+
+Government: National security, confidential communications, and digital identity.
+
+5. Tools & Resources: Provide a list of useful tools for encryption:
+
+Open source libraries
+
+Cloud-based encryption services
+
+Enterprise-grade security solutions
+
+6. Future Trends: Discuss emerging trends in encryption and how they might impact the future of data protection:
+
+Privacy-enhancing technologies : Explain how PETs like differential privacy and federated learning can be used to protect user data while still allowing for analysis.
+
+Blockchain technology: Discuss how blockchain can be used for secure data storage and transactions.
+
+Use diagrams, charts, and tables to make the information more accessible and visually appealing. For example:
+
+A flowchart illustrating different encryption methods and their use cases.
+
+A comparison table highlighting key features of various libraries and frameworks.
+
+A timeline showing the evolution of encryption technologies over time.
+
+This guide provides a solid foundation for understanding encryption tools and techniques. By expanding on these areas, you can create a comprehensive resource that helps readers make informed decisions about data security.
\ No newline at end of file
diff --git a/docs/Cybersecurity/file_encryption_survey_script_middle.md b/docs/Cybersecurity/file_encryption_survey_script_middle.md
new file mode 100644
index 0000000000000000000000000000000000000000..3e6d6fbec2051dc535a43af16aa85c80c6d76f60
--- /dev/null
+++ b/docs/Cybersecurity/file_encryption_survey_script_middle.md
@@ -0,0 +1,10 @@
+Good evening, and welcome back. Tonight's top story: the world of file encryption is undergoing a rapid evolution. A new report from the International Security Council sheds light on the latest trends in this field. The report highlights how advancements in cryptography are shaping our digital security landscape.
+
+We begin with a look at the state of file encryption, which has been revolutionized by the rise of advanced algorithms and tools. The report emphasizes that while traditional methods like AES-256 remain industry standards, newer techniques like ChaCha20 and its variations are gaining traction for their speed and efficiency in various environments.
+
+The report also underscores the importance of key derivation functions (KDFs) for generating secure encryption keys from passwords. It highlights the growing demand for robust KDFs that can withstand brute-force attacks, with Argon2id emerging as a top contender due to its resilience against these threats.
+
+Finally, the report delves into the world of file encryption tools, categorizing them based on their specific use cases. From disk encryption solutions like VeraCrypt and BitLocker to cloud storage options like Cryptomator and NordLocker, the report offers a comprehensive overview of available tools for various needs.
+
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/Cybersecurity/file_encryption_survey_script_middle_clean.txt b/docs/Cybersecurity/file_encryption_survey_script_middle_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..a083273a7245b0010e3f77bcab56a3a71af630c9
--- /dev/null
+++ b/docs/Cybersecurity/file_encryption_survey_script_middle_clean.txt
@@ -0,0 +1,9 @@
+Good evening, and welcome back. Tonight's top story: the world of file encryption is undergoing a rapid evolution. A new report from the International Security Council sheds light on the latest trends in this field. The report highlights how advancements in cryptography are shaping our digital security landscape.
+
+We begin with a look at the state of file encryption, which has been revolutionized by the rise of advanced algorithms and tools. The report emphasizes that while traditional methods like AES-256 remain industry standards, newer techniques like ChaCha20 and its variations are gaining traction for their speed and efficiency in various environments.
+
+The report also underscores the importance of key derivation functions for generating secure encryption keys from passwords. It highlights the growing demand for robust KDFs that can withstand brute-force attacks, with Argon2id emerging as a top contender due to its resilience against these threats.
+
+Finally, the report delves into the world of file encryption tools, categorizing them based on their specific use cases. From disk encryption solutions like VeraCrypt and BitLocker to cloud storage options like Cryptomator and NordLocker, the report offers a comprehensive overview of available tools for various needs.
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/Cybersecurity/file_encryption_survey_script_short.md b/docs/Cybersecurity/file_encryption_survey_script_short.md
new file mode 100644
index 0000000000000000000000000000000000000000..5b19f8305e91d3ca968d737ea60865128f231701
--- /dev/null
+++ b/docs/Cybersecurity/file_encryption_survey_script_short.md
@@ -0,0 +1,32 @@
+## File Encryption: A Deep Dive into the Latest Trends (Vertical Video Script)
+
+**(Video opens with a close-up of a computer screen displaying encrypted files)**
+
+**We're diving deep into the world of file encryption, where security is paramount.
+
+**(Quick cuts showing various devices and software - laptops, smartphones, cloud storage icons)**
+
+**The landscape of file encryption is constantly evolving, with new tools and techniques emerging all the time. But what are the latest trends? Let's break it down!
+
+**(Text overlay on screen: "File Encryption: A Deep Dive into the Latest Trends")**
+
+**First off, we have **AES-256-GCM**, a standard in the industry for its speed and security. It's like a high-speed highway for your data. And then there's **ChaCha20-Poly1305**, known for its ease of implementation and performance on mobile devices.
+
+**(Video transitions to an animation showing two different encryption algorithms, one faster than the other)**
+
+**But what about **KDF (Key Derivation Function)**? This is your secret weapon against brute force attacks. Think of it as a complex puzzle that makes it nearly impossible for hackers to crack your data.
+
+**(Video shows a person using their computer with a secure password manager app)**
+
+**The best practices are clear: **Argon2id**, the gold standard, offers unparalleled security and is recommended for all new projects. And don't forget about **Cloud Storage Encryption** solutions like Cryptomator and NordLocker, offering seamless protection on your cloud accounts.
+
+**(Video transitions to a screen with various encryption tools)**
+
+**From disk encryption to file-level protection, the options are vast! But remember, security is not just about the technology; it's also about understanding how to use it correctly and staying informed of the latest trends.
+
+
+**(Video ends with a shot of someone working on their laptop with a secure password manager app)**
+
+**Stay tuned for more updates on file encryption as we continue to explore this critical aspect of digital security!
+
+**(Text overlay: "The Android Times" and logo)**
\ No newline at end of file
diff --git a/docs/Cybersecurity/file_encryption_survey_script_short_clean.txt b/docs/Cybersecurity/file_encryption_survey_script_short_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..f1d39ec87aea07138470a67383ba963636679be4
--- /dev/null
+++ b/docs/Cybersecurity/file_encryption_survey_script_short_clean.txt
@@ -0,0 +1,17 @@
+We're diving deep into the world of file encryption, where security is paramount.
+
+The landscape of file encryption is constantly evolving, with new tools and techniques emerging all the time. But what are the latest trends? Let's break it down!
+
+"File Encryption: A Deep Dive into the Latest Trends")
+
+First off, we have AES-256-GCM, a standard in the industry for its speed and security. It's like a high-speed highway for your data. And then there's ChaCha20-Poly1305, known for its ease of implementation and performance on mobile devices.
+
+But what about KDF ? This is your secret weapon against brute force attacks. Think of it as a complex puzzle that makes it nearly impossible for hackers to crack your data.
+
+Argon2id, the gold standard, offers unparalleled security and is recommended for all new projects. And don't forget about Cloud Storage Encryption solutions like Cryptomator and NordLocker, offering seamless protection on your cloud accounts.
+
+From disk encryption to file-level protection, the options are vast! But remember, security is not just about the technology; it's also about understanding how to use it correctly and staying informed of the latest trends.
+
+Stay tuned for more updates on file encryption as we continue to explore this critical aspect of digital security!
+
+"The Android Times" and logo)
\ No newline at end of file
diff --git a/docs/Cybersecurity/ransomware_survey.md b/docs/Cybersecurity/ransomware_survey.md
new file mode 100644
index 0000000000000000000000000000000000000000..382f3b80e7da99dc48537af2856315047ce4fa41
--- /dev/null
+++ b/docs/Cybersecurity/ransomware_survey.md
@@ -0,0 +1,615 @@
+# ランサムウェア技術サーベイ 2026
+
+> [!NOTE]
+> 本サーベイは防御的セキュリティの観点から、ランサムウェアの技術的特徴・攻撃手法・防御戦略を体系的に整理したものです。2026年5月時点の公開情報に基づいています。
+
+---
+
+## 目次
+
+1. [歴史的変遷](#1-歴史的変遷)
+2. [ビジネスモデルの進化](#2-ビジネスモデルの進化)
+3. [技術的メカニズム](#3-技術的メカニズム)
+4. [攻撃ライフサイクル(キルチェーン)](#4-攻撃ライフサイクルキルチェーン)
+5. [MITRE ATT&CK マッピング](#5-mitre-attck-マッピング)
+6. [主要ランサムウェアファミリー](#6-主要ランサムウェアファミリー)
+7. [被害統計と経済的影響](#7-被害統計と経済的影響)
+8. [防御戦略](#8-防御戦略)
+9. [法執行機関の動向](#9-法執行機関の動向)
+10. [2025-2026年の最新トレンド](#10-2025-2026年の最新トレンド)
+11. [参考フレームワーク・リソース](#11-参考フレームワークリソース)
+
+---
+
+## 1. 歴史的変遷
+
+ランサムウェアは約37年の歴史を持ち、原始的なフロッピーディスク配布から、数十億ドル規模のサイバー犯罪エコシステムへと進化してきた。
+
+```mermaid
+timeline
+ title ランサムウェアの進化タイムライン
+ 1989 : AIDS Trojan(PC Cyborg)
+ : フロッピーディスク配布
+ : 身代金 $189
+ 2006 : Archiveus
+ : RSA暗号の初期採用
+ 2007 : ロッカー型ランサムウェア登場
+ : OS機能のロック
+ 2013 : CryptoLocker
+ : 強力な非対称暗号 + Bitcoin
+ : 近代ランサムウェアの幕開け
+ 2016 : Locky, Cerber
+ : RaaS モデルの普及
+ 2017 : WannaCry / NotPetya
+ : EternalBlue 悪用
+ : 世界規模の被害
+ 2019 : Maze
+ : 二重恐喝の発明
+ : データ窃取 + 暗号化
+ 2020-2021 : Ryuk, Conti, REvil
+ : Big Game Hunting
+ : 高額身代金要求
+ 2022-2023 : LockBit, ALPHV/BlackCat
+ : RaaS の産業化
+ : 三重恐喝
+ 2024 : 法執行機関の大規模摘発
+ : Operation Cronos(LockBit)
+ : グループの分裂・再編
+ 2025-2026 : 124+ の活動グループ
+ : AI活用・暗号化なし恐喝
+ : エコシステムの断片化
+```
+
+### 各時代の詳細
+
+| 年代 | 主な特徴 | 代表例 | 暗号化方式 | 支払い手段 |
+|:-----|:---------|:-------|:-----------|:-----------|
+| **1989** | 物理メディア配布、ファイル名暗号化 | AIDS Trojan | 簡易な対称暗号 | 郵便為替 |
+| **2006-2012** | スクリーンロッカー、初期暗号化 | Archiveus, Reveton | RSA(初期実装) | プリペイドカード, SMS |
+| **2013-2015** | 強力な暗号化 + 匿名決済 | CryptoLocker, CryptoWall | RSA-2048 + AES-256 | Bitcoin |
+| **2016-2018** | RaaS モデル確立、ワーム型拡散 | WannaCry, NotPetya, SamSam | AES-256 + RSA | Bitcoin, Monero |
+| **2019-2021** | 二重恐喝、Big Game Hunting | Maze, Ryuk, Conti, REvil | AES/ChaCha20 + RSA/ECC | Bitcoin, Monero |
+| **2022-2024** | RaaS 産業化、三重恐喝 | LockBit, ALPHV, Clop | ハイブリッド暗号 | 暗号通貨(複数種) |
+| **2025-2026** | 断片化、AI活用、暗号化なし恐喝 | Qilin, Akira, The Gentlemen | 部分暗号化 / 暗号化なし | 暗号通貨 |
+
+---
+
+## 2. ビジネスモデルの進化
+
+### 2.1 恐喝モデルの発展
+
+```mermaid
+graph LR
+ A["スクリーンロッカー
(2006-2012)"] --> B["暗号化ランサムウェア
(2013-2018)"]
+ B --> C["二重恐喝
(2019-)"]
+ C --> D["三重恐喝
(2021-)"]
+ D --> E["暗号化なし恐喝
(2024-)"]
+
+ style A fill:#ff6b6b,color:#fff
+ style B fill:#ee5a24,color:#fff
+ style C fill:#f39c12,color:#fff
+ style D fill:#e74c3c,color:#fff
+ style E fill:#8e44ad,color:#fff
+```
+
+| 恐喝モデル | 手法 | 被害者への圧力 |
+|:-----------|:-----|:--------------|
+| **単純暗号化** | ファイル暗号化のみ | データ復旧不可 |
+| **二重恐喝** | データ窃取 → 暗号化 → リークサイト公開脅迫 | データ漏洩リスク + 復旧不可 |
+| **三重恐喝** | 上記 + DDoS攻撃 + 顧客・パートナーへの直接連絡 | 事業継続リスク + 評判毀損 |
+| **四重恐喝** | 上記 + 規制当局への報告脅迫 | 法的リスク追加 |
+| **暗号化なし恐喝** | データ窃取のみ(暗号化せず) | 検知回避 + バックアップ無効化不要 |
+
+### 2.2 Ransomware-as-a-Service(RaaS)
+
+2025年時点で、RaaSは合法的なSaaSビジネスと同等の成熟度に達している。
+
+```mermaid
+graph TB
+ subgraph "RaaS エコシステム"
+ OP["RaaS オペレーター
(開発・インフラ管理)"]
+ AF["アフィリエイト
(攻撃実行者)"]
+ IAB["初期アクセスブローカー
(IAB)"]
+ DEV["マルウェア開発者"]
+ CRYPT["暗号化サービス"]
+ NEG["身代金交渉担当"]
+
+ IAB -->|アクセス販売| AF
+ DEV -->|ツール提供| OP
+ CRYPT -->|難読化| OP
+ OP -->|プラットフォーム提供| AF
+ AF -->|身代金の20-30%| OP
+ NEG -->|交渉代行| AF
+ end
+
+ style OP fill:#e74c3c,color:#fff
+ style AF fill:#f39c12,color:#fff
+ style IAB fill:#3498db,color:#fff
+```
+
+**RaaSモデルの特徴(2025年):**
+
+- **収益配分**: アフィリエイト70-80% / オペレーター20-30%が一般的
+- **参入障壁の低下**: 技術力不要で攻撃キャンペーン実施可能
+- **招待制・審査制**: 法執行機関の浸透を防ぐため、上位RaaSは審査制に移行
+- **階層型サービス**: 暗号化RaaS / データ窃取専門 / アクセス収益化など選択可能
+- **120+の活動グループ**: エコシステムの高度な断片化(2025年)
+
+---
+
+## 3. 技術的メカニズム
+
+### 3.1 暗号化アーキテクチャ
+
+現代のランサムウェアは**ハイブリッド暗号方式**を標準採用している。
+
+```mermaid
+sequenceDiagram
+ participant 攻撃者
+ participant マルウェア
+ participant ファイル
+ participant C2サーバー
+
+ 攻撃者->>マルウェア: 公開鍵を埋め込み
+
+ loop 各ファイルに対して
+ マルウェア->>マルウェア: ランダムな対称鍵を生成(AES-256 / ChaCha20)
+ マルウェア->>ファイル: 対称鍵でファイル暗号化
+ マルウェア->>マルウェア: 攻撃者の公開鍵で対称鍵を暗号化
+ マルウェア->>ファイル: 暗号化された対称鍵をファイルに付加
+ end
+
+ マルウェア->>C2サーバー: セッション情報送信(任意)
+
+ Note over 攻撃者: 秘密鍵は攻撃者のみ保有
→ 被害者は復号不可能
+```
+
+#### 使用される暗号アルゴリズム
+
+| 種類 | アルゴリズム | 用途 | 特徴 |
+|:-----|:-----------|:-----|:-----|
+| **対称暗号** | AES-256 (CBC/CTR) | ファイル本体の暗号化 | 高速、ハードウェアアクセラレーション対応 |
+| **対称暗号** | ChaCha20 | ファイル本体の暗号化 | ソフトウェア実装で高速、ARMに最適 |
+| **対称暗号** | Salsa20 | ファイル本体の暗号化 | ChaCha20の前身 |
+| **非対称暗号** | RSA-2048/4096 | 対称鍵の暗号化 | 広く使用、実績あり |
+| **非対称暗号** | Curve25519 (ECDH) | 鍵交換 | 高速、コンパクトな鍵サイズ |
+| **非対称暗号** | ECDSA/EdDSA | 認証・署名 | 鍵交換プロトコルの認証 |
+
+### 3.2 間欠暗号化(Intermittent Encryption)
+
+2022年頃から普及した高速暗号化技術。
+
+**仕組み:**
+- ファイル全体ではなく、一定間隔(例: 64バイトごと)のみ暗号化
+- ファイルヘッダや先頭部分のみを暗号化するパターンも存在
+- ランダムパターンで暗号化位置を変えるバリエーションあり
+
+**攻撃者にとっての利点:**
+- ⚡ **高速化**: 暗号化時間を大幅短縮(検知までの時間窓を縮小)
+- 🔇 **検知回避**: I/O負荷が低く、行動検知を回避
+- 💀 **十分な破壊**: 部分暗号化でもファイルは使用不能
+
+**防御側の対応:**
+- エントロピー分析による部分暗号化の検知
+- **White Phoenix** 等のツールによる部分復元の可能性
+
+### 3.3 回避技術
+
+| 技術 | 説明 | MITRE ATT&CK |
+|:-----|:-----|:-------------|
+| **BYOVD** | 脆弱な署名済みドライバーを持ち込み、カーネルレベルでEDRを無効化 | T1068, T1562.001 |
+| **Living off the Land** | PowerShell、WMI、PsExec等の正規ツールを悪用 | T1059, T1047 |
+| **DLLサイドローディング** | 正規アプリケーションに悪意のあるDLLをロードさせる | T1574.002 |
+| **コード署名悪用** | 窃取/購入した正規証明書でマルウェアに署名 | T1553.002 |
+| **ファイルレスマルウェア** | ディスクに書き込まずメモリ上で実行 | T1055 |
+| **ポリモーフィズム** | 実行ごとにコードを変形し、シグネチャ検知を回避 | T1027.001 |
+| **アンチサンドボックス** | 仮想環境や解析環境を検出して動作を停止 | T1497 |
+| **ログ消去** | Windows イベントログの削除 | T1070.001 |
+| **マスカレード** | 正規プロセス名(svchost.exe等)に偽装 | T1036 |
+
+---
+
+## 4. 攻撃ライフサイクル(キルチェーン)
+
+```mermaid
+graph LR
+ subgraph "フェーズ 1: 初期侵入"
+ IA1["フィッシング"]
+ IA2["脆弱性悪用
(VPN/RDP)"]
+ IA3["サプライチェーン攻撃"]
+ IA4["IAB経由の
アクセス購入"]
+ end
+
+ subgraph "フェーズ 2: 足場確立"
+ P1["永続化メカニズム"]
+ P2["C2通信確立"]
+ P3["権限昇格"]
+ end
+
+ subgraph "フェーズ 3: 内部偵察・横展開"
+ L1["Active Directory偵察"]
+ L2["クレデンシャル窃取
(Mimikatz)"]
+ L3["RDP/WMI/SMBで
横展開"]
+ end
+
+ subgraph "フェーズ 4: 事前準備"
+ E1["データ窃取
(Rclone/MEGA)"]
+ E2["バックアップ破壊"]
+ E3["セキュリティ製品無効化"]
+ end
+
+ subgraph "フェーズ 5: 実行"
+ R1["ランサムウェア展開"]
+ R2["シャドウコピー削除"]
+ R3["身代金要求ノート配置"]
+ end
+
+ IA1 --> P1
+ IA2 --> P1
+ IA3 --> P1
+ IA4 --> P1
+ P1 --> P2 --> P3
+ P3 --> L1 --> L2 --> L3
+ L3 --> E1 --> E2 --> E3
+ E3 --> R1 --> R2 --> R3
+```
+
+### 4.1 初期侵入ベクトル(2025年の傾向)
+
+| 侵入経路 | 使用割合 | 詳細 |
+|:---------|:---------|:-----|
+| **脆弱性悪用** | ★★★★★ | VPNアプライアンス、ファイアウォール、MFTソフトウェアのN-day/0-day脆弱性。SonicWall、Fortinet、Ivanti等が標的 |
+| **フィッシング** | ★★★★☆ | AI生成の高品質フィッシングメール。ClickFix、OAuthトークン窃取、デバイスコードフィッシング(M365)。Signal/WhatsApp経由も増加 |
+| **RDP/リモートアクセス** | ★★★★☆ | クレデンシャルスタッフィング、ブルートフォース。正規ツールのため検知困難 |
+| **IAB経由** | ★★★☆☆ | 初期アクセスブローカーがInfostealer等で窃取した認証情報を販売 |
+| **サプライチェーン** | ★★★☆☆ | MSP経由、ソフトウェアアップデートの侵害、信頼された接続の悪用 |
+
+### 4.2 横展開と権限昇格
+
+- **RDP(T1021.001)**: ランサムウェア侵入の約85%で使用される
+- **WMI(T1047)**: リモートコード実行に使用
+- **SMB/SSH**: ペイロードの配布とトンネリング
+- **Mimikatz**: LSASS からのクレデンシャルダンプ(T1003)
+- **Pass-the-Ticket(T1550.003)**: Kerberos チケットの再利用
+- **Cobalt Strike / Brute Ratel**: 商用レッドチームツールの悪用
+
+### 4.3 データ窃取
+
+| ツール | 用途 | 特徴 |
+|:-------|:-----|:-----|
+| **Rclone** | クラウドストレージへの大量データ転送 | MEGA、OneDrive、Google Drive等に対応。マルチスレッド転送 |
+| **MEGAsync** | MEGAクラウドへのデータ同期 | 大容量転送に対応 |
+| **WinRAR / 7-Zip** | データの圧縮・ステージング | 転送前の圧縮でフットプリント削減 |
+| **FileZilla** | FTP/SFTP転送 | 攻撃者管理のサーバーへ送信 |
+| **カスタムツール** | 独自開発の窃取ツール | グループ固有の実装 |
+
+> [!IMPORTANT]
+> データ窃取は暗号化の**数日前**に実行されるのが一般的。これにより、被害者がバックアップから復旧しても、データ漏洩の脅威で身代金支払いを強いる。
+
+### 4.4 バックアップ・復旧妨害
+
+```
+# シャドウコピー削除(最も一般的な手法)
+vssadmin.exe delete shadows /all /quiet
+wmic shadowcopy delete /nointeractive
+Get-WmiObject Win32_ShadowCopy | ForEach-Object {$_.Delete()}
+
+# バックアップサービスの停止
+net stop "Veeam Backup Service"
+net stop "SQL Server VSS Writer"
+
+# bcdedit によるリカバリモード無効化
+bcdedit /set {default} recoveryenabled No
+```
+
+- バックアップサーバーとリポジトリの特定と破壊
+- VMware ESXi ホストへの攻撃(仮想ディスクの暗号化/破壊)
+- オンラインバックアップの暗号化・削除
+
+---
+
+## 5. MITRE ATT&CK マッピング
+
+ランサムウェア攻撃で頻繁に観察されるTTPsの概要。
+
+| 戦術(Tactic) | 主要テクニック | ID |
+|:--------------|:-------------|:---|
+| **初期アクセス** | フィッシング | T1566 |
+| | 公開アプリケーションの悪用 | T1190 |
+| | 外部リモートサービス | T1133 |
+| **実行** | コマンド/スクリプトインタプリタ(PowerShell) | T1059.001 |
+| | WMI | T1047 |
+| **永続化** | スケジュールタスク | T1053 |
+| | レジストリRun Key | T1547.001 |
+| **権限昇格** | 脆弱性悪用(BYOVD含む) | T1068 |
+| **防御回避** | 難読化ファイル | T1027 |
+| | インジケータ除去(ログ消去) | T1070 |
+| | マスカレード | T1036 |
+| | セキュリティツール無効化 | T1562.001 |
+| **クレデンシャルアクセス** | OS資格情報ダンプ(LSASS) | T1003.001 |
+| **探索** | ドメイン信頼の探索 | T1482 |
+| | ネットワーク共有の探索 | T1135 |
+| **横展開** | リモートデスクトップ(RDP) | T1021.001 |
+| | SMB/Windows管理共有 | T1021.002 |
+| | Pass-the-Ticket | T1550.003 |
+| **収集・窃取** | データのステージング | T1074 |
+| | Webサービス経由の窃取 | T1567 |
+| **影響** | **データ暗号化による影響** | **T1486** |
+| | **システム復旧の妨害** | **T1490** |
+| | クラウドへのデータ転送 | T1537 |
+
+---
+
+## 6. 主要ランサムウェアファミリー
+
+### 6.1 歴史的に重要なファミリー
+
+| ファミリー | 活動期間 | 暗号化 | 配布方法 | 特筆事項 | 現在の状態 |
+|:----------|:---------|:-------|:---------|:---------|:----------|
+| **CryptoLocker** | 2013-2014 | RSA-2048 + AES | Gameover Zeus ボットネット | 近代ランサムウェアの原型。推定$300M被害 | 解体済み |
+| **WannaCry** | 2017 | AES-128-CBC + RSA-2048 | EternalBlue(MS17-010) | 150カ国、23万台以上に感染。NHS等に甚大な被害 | 亜種が残存 |
+| **NotPetya** | 2017 | AES + RSA | M.E.Doc更新、EternalBlue | 推定$10B被害。実質はワイパー(データ復旧不可) | 単発攻撃 |
+| **Ryuk** | 2018-2021 | AES-256 + RSA-4096 | TrickBot/BazarLoader → 手動展開 | Big Game Hunting の先駆け | Contiに移行 |
+| **Maze** | 2019-2020 | ChaCha20 + RSA | フィッシング、脆弱性悪用 | **二重恐喝を発明**。リークサイト運営 | 活動停止 |
+| **REvil/Sodinokibi** | 2019-2022 | Salsa20 + Curve25519 | RaaS、Kaseya VSA攻撃 | Kaseya攻撃で$70M要求 | 法執行で解体 |
+| **Conti** | 2020-2022 | AES-256 + RSA-4096 | TrickBot → Cobalt Strike | Costa Rica政府へ攻撃。内部チャット流出 | 解散・分裂 |
+
+### 6.2 現在活動中の主要グループ(2024-2026)
+
+| グループ | 特徴 | 主な標的 | 技術的特徴 | 状態(2026年) |
+|:---------|:-----|:---------|:-----------|:-------------|
+| **LockBit** | 最も長期間活動したRaaS | 広範な業種 | 高速暗号化、StealBit窃取ツール | Operation Cronosで打撃を受けるも復活試行中 |
+| **ALPHV/BlackCat** | Rust言語で開発 | ヘルスケア、金融 | クロスプラットフォーム(Windows/Linux/ESXi) | FBI浸透で実質解体。Cicada3301として再編の可能性 |
+| **Clop** | 脆弱性主導のデータ窃取 | ファイル転送ソフト利用企業 | MOVEit/GoAnywhere等の0-day大量悪用 | 暗号化なし恐喝に特化して活動中 |
+| **Qilin** | 2025年最も多産なグループ | 広範な業種 | 二重恐喝 | 2025年に最大勢力に。2026年初頭に活動減少 |
+| **Akira** | 安定した長期活動 | 中小企業 | SonicWall VPN等のN-day悪用 | 活発に活動中 |
+| **RansomHub** | 急成長したRaaS | 多業種 | 2025年初頭に主要プレイヤーに | 2025年4月に突然活動停止 |
+| **Black Basta** | Contiの後継とされる | 製造、建設 | QakBot経由の配布 | 活動中 |
+| **Play** | ステルス重視 | 中堅企業 | 間欠暗号化 | 活動中 |
+| **The Gentlemen** | 2025年末に出現 | 不明 | 急速な成長 | 2026年Q1に主要グループに台頭 |
+
+---
+
+## 7. 被害統計と経済的影響
+
+### 7.1 グローバル統計(2024-2025)
+
+| 指標 | 数値 | 出典 |
+|:-----|:-----|:-----|
+| 年間被害総額(2025年推計) | **$57B**(約8.5兆円) | Mimecast |
+| 2031年予測被害額 | $265B+ | Cybersecurity Ventures |
+| インシデント当たり平均復旧コスト | **$1.84M - $2.73M** | Sophos, 各社レポート |
+| 平均身代金要求額(2025年) | $1.20M(中央値) | Sophos |
+| 平均身代金支払額(2025年) | **$1.0M - $1.82M** | 各社レポート |
+| 身代金支払い拒否率(2024年) | **64%** | Coveware |
+| 平均ダウンタイム | **21-24日** | Varonis |
+| 攻撃頻度 | **11-19秒に1件** | グローバル推計 |
+| 活動中のランサムウェアグループ(2025年) | **124+** | Check Point |
+
+### 7.2 身代金支払い額の推移
+
+```mermaid
+xychart-beta
+ title "平均身代金支払い額の推移(推計, 単位: 万ドル)"
+ x-axis [2019, 2020, 2021, 2022, 2023, 2024, 2025]
+ y-axis "万ドル" 0 --> 200
+ bar [8, 31, 81, 81, 100, 126, 100]
+```
+
+> [!WARNING]
+> 統計値はレポート提供元(Sophos, Coveware, Chainalysis等)の調査方法により大幅に異なる場合がある。上記は複数ソースの中央的な値を示している。
+
+### 7.3 業種別の標的傾向(2025年)
+
+| 業種 | 標的となる理由 |
+|:-----|:-------------|
+| **製造業** | OT/IT融合環境、ダウンタイム許容度低い |
+| **ヘルスケア** | 患者データの高い機密性、ダウンタイム=生命リスク |
+| **教育** | セキュリティ予算の制約、広範なユーザーベース |
+| **政府機関** | レガシーシステム、政治的動機 |
+| **建設業** | セキュリティ成熟度が比較的低い(2025年に急増) |
+| **金融サービス** | 高価値データ、規制プレッシャー |
+
+---
+
+## 8. 防御戦略
+
+### 8.1 多層防御フレームワーク
+
+```mermaid
+graph TB
+ subgraph "層 1: 予防"
+ P1["パッチ管理
(特にVPN/FW/MFT)"]
+ P2["フィッシング耐性MFA"]
+ P3["特権アクセス管理(PAM)"]
+ P4["メールセキュリティ"]
+ P5["攻撃面管理(ASM)"]
+ end
+
+ subgraph "層 2: 検知"
+ D1["EDR/XDR"]
+ D2["行動分析"]
+ D3["カナリアファイル"]
+ D4["ネットワーク異常検知"]
+ D5["SIEM/SOAR"]
+ end
+
+ subgraph "層 3: 封じ込め"
+ C1["ネットワーク
マイクロセグメンテーション"]
+ C2["ゼロトラスト
アーキテクチャ"]
+ C3["自動隔離"]
+ end
+
+ subgraph "層 4: 復旧"
+ R1["イミュータブル
バックアップ"]
+ R2["エアギャップ
バックアップ"]
+ R3["3-2-1ルール"]
+ R4["定期的な復旧テスト"]
+ end
+
+ P1 --> D1
+ P2 --> D2
+ D1 --> C1
+ D2 --> C2
+ C1 --> R1
+ C2 --> R2
+```
+
+### 8.2 ID・アクセス管理(最優先の攻撃面)
+
+| 対策 | 詳細 |
+|:-----|:-----|
+| **フィッシング耐性MFA** | FIDO2/WebAuthn ベースの認証。SMS/メールOTPは不十分 |
+| **特権アクセス管理(PAM)** | Just-In-Time(JIT)アクセス。常時管理者権限を排除 |
+| **最小権限の原則** | 必要最小限のアクセス権のみ付与 |
+| **ID異常検知** | 不可能な移動、異常なログインパターン、権限昇格の監視 |
+| **パスワードレス認証** | 可能な限りパスワード依存を排除 |
+
+### 8.3 検知技術
+
+| 技術 | 用途 | 効果 |
+|:-----|:-----|:-----|
+| **EDR/XDR** | エンドポイント・ネットワーク横断の統合監視 | プロセスチェーン、ファイル操作の異常検知 |
+| **行動分析** | 大量ファイル変更、エントロピー変化の検知 | シグネチャに依存しない検知 |
+| **カナリアファイル** | 特定のデコイファイルへのアクセスで即座にアラート | 高精度・低誤検知 |
+| **DLP** | 大量データの外部転送を検知・遮断 | データ窃取フェーズで検知 |
+| **AI/ML検知** | 正規ツール(Rclone等)の異常使用パターン検知 | LotL攻撃に対応 |
+| **デセプション技術** | ハニーポット、偽認証情報の設置 | 攻撃者の横展開を早期検知 |
+
+### 8.4 バックアップ戦略
+
+> [!IMPORTANT]
+> バックアップは「最後の防衛線」。しかし二重恐喝の時代では、バックアップだけでは不十分。データ窃取の防止も同時に必要。
+
+| 原則 | 説明 |
+|:-----|:-----|
+| **3-2-1ルール** | 3つのコピー、2種類のメディア、1つはオフサイト |
+| **3-2-1-1-0** | 上記 + 1つはオフラインまたはイミュータブル + 0エラー(定期検証) |
+| **イミュータブルバックアップ** | WORM(Write Once Read Many)。管理者権限でも変更・削除不可 |
+| **エアギャップバックアップ** | ネットワークから物理的に分離された保管 |
+| **定期的な復旧テスト** | バックアップからの復元を定期的に実施・時間計測 |
+
+### 8.5 インシデント対応
+
+| フェーズ | アクション |
+|:---------|:---------|
+| **準備** | IRプレイブック策定、テーブルトップ演習実施、連絡体制確立 |
+| **検知・分析** | 感染範囲の特定、ランサムウェアファミリーの同定、TTPの分析 |
+| **封じ込め** | 感染端末のネットワーク分離、横展開の遮断、C2通信の遮断 |
+| **根絶** | マルウェアの除去、永続化メカニズムの排除、全パスワードリセット |
+| **復旧** | バックアップからの復元、システム再構築、段階的サービス復旧 |
+| **事後対応** | ポストモーテム分析、改善策の実装、規制当局への報告 |
+
+**身代金支払いに関する考慮事項:**
+- 支払いは復号を保証しない(支払い後の復旧成功率は低下傾向)
+- OFAC制裁対象グループへの支払いは法的リスクを伴う
+- No More Ransom プロジェクト(nomoreransom.org)で無料復号ツールの確認を推奨
+- 法執行機関は一般的に支払いを推奨しない
+
+---
+
+## 9. 法執行機関の動向
+
+### 9.1 主要な摘発オペレーション
+
+| オペレーション | 時期 | 標的 | 成果 |
+|:-------------|:-----|:-----|:-----|
+| **Operation Cronos** | 2024年2月 | LockBit | 34サーバー押収、200+暗号通貨ウォレット、1,000+復号キー取得 |
+| **ALPHV/BlackCat 浸透作戦** | 2023年12月 | ALPHV/BlackCat | FBIがネットワークに浸透。復号ツール開発。被害者 ~$99M を救済 |
+| **Hive テイクダウン** | 2023年1月 | Hive | 7ヶ月間の浸透捜査。1,300+の復号キー配布 |
+| **ALPHV関連者の起訴** | 2025-2026年 | アフィリエイト個人 | 米国人サイバーセキュリティ専門家2名が恐喝罪で有罪判決 |
+
+### 9.2 法執行の影響と限界
+
+**効果:**
+- 短期的なオペレーション混乱
+- 信頼性の毀損(アフィリエイトの離脱)
+- 復号キーの取得による被害者救済
+- 抑止効果
+
+**限界:**
+- LockBitは数日以内にリークサイトを再構築
+- 旧グループのメンバーが新グループとして再編(ALPHV → Cicada3301疑い)
+- アフィリエイトの他プラットフォームへの移動
+- 管轄権の壁(ロシア等の非協力国)
+
+---
+
+## 10. 2025-2026年の最新トレンド
+
+### 10.1 AI の攻撃への活用
+
+| 活用領域 | 詳細 |
+|:---------|:-----|
+| **フィッシング生成** | 生成AIによる高品質・多言語フィッシングメールの自動作成 |
+| **脆弱性特定** | AIによる攻撃対象の自動スキャンと脆弱性分析 |
+| **ペイロード最適化** | EDR回避のためのコード最適化 |
+| **ソーシャルエンジニアリング** | ディープフェイク音声/ビデオを使用したなりすまし |
+| **窃取データ分析** | 大量の窃取データから高価値情報を自動抽出 |
+| **攻撃タイムラインの圧縮** | 初期侵入から暗号化までの時間を短縮 |
+
+### 10.2 暗号化なし恐喝(Encryption-less Extortion)
+
+2024-2025年にかけて顕著になったトレンド。Clop が先駆者。
+
+- **暗号化を行わず、データ窃取のみ**で身代金を要求
+- **利点(攻撃者視点)**: EDR/行動検知の回避、高速実行、バックアップ戦略を無効化
+- **防御側への影響**: DLP(Data Loss Prevention)の重要性が飛躍的に増大
+
+### 10.3 クラウド・仮想化環境への標的化
+
+- VMware ESXi ホストへの直接攻撃(Linux向けランサムウェア)
+- クラウド環境でのトークン窃取、OAuth アプリ悪用
+- SaaS アプリケーションのデータを標的
+- マルチクラウド環境での横展開
+
+### 10.4 その他の注目トレンド
+
+| トレンド | 詳細 |
+|:---------|:-----|
+| **正規ツールの悪用** | BitLocker等のOS標準暗号化ツールを使用してランサムウェアとして利用 |
+| **IoTランサムウェア** | 産業用IoTデバイスやメディカルデバイスを標的 |
+| **エコシステムの断片化** | 大規模グループの解体により120+の小規模グループが乱立 |
+| **アフィリエイトの流動性** | 個々のアフィリエイトが複数のRaaSプラットフォームを併用 |
+| **ClickFix手法** | 偽のブラウザ/システムプロンプトで被害者に悪意のあるコマンドを実行させる |
+| **デバイスコードフィッシング** | M365のデバイス認証フローを悪用 |
+
+---
+
+## 11. 参考フレームワーク・リソース
+
+### 防御フレームワーク
+
+| フレームワーク | 概要 |
+|:-------------|:-----|
+| **NIST CSF 2.0** | Identify, Protect, Detect, Respond, Recover + Govern の6機能 |
+| **CISA Ransomware Guide** | ランサムウェア対策のベストプラクティス集 |
+| **MITRE ATT&CK** | 脅威アクターのTTPsの体系的カタログ |
+| **ISO 27001:2022** | 情報セキュリティ管理システムの国際規格 |
+| **CIS Controls v8** | 優先度順のセキュリティ対策リスト |
+
+### 脅威インテリジェンス・リソース
+
+| リソース | URL | 概要 |
+|:---------|:----|:-----|
+| **No More Ransom** | nomoreransom.org | 無料復号ツールの集約サイト |
+| **CISA Known Exploited Vulnerabilities** | cisa.gov/known-exploited-vulnerabilities-catalog | 実際に悪用された脆弱性カタログ |
+| **Ransomware.live** | ransomware.live | ランサムウェアグループ活動のリアルタイム追跡 |
+| **MITRE ATT&CK** | attack.mitre.org | TTPs データベース |
+| **VirusTotal** | virustotal.com | マルウェアスキャン・分析 |
+
+---
+
+> [!TIP]
+> **防御の優先順位(2025-2026年):**
+> 1. 🔐 **ID・アクセス管理の強化**(フィッシング耐性MFA、PAM)
+> 2. 🩹 **パッチ管理の徹底**(特にVPN/ファイアウォール/リモートアクセス)
+> 3. 💾 **イミュータブルバックアップの実装**
+> 4. 🔍 **EDR/XDR + 行動分析による検知**
+> 5. 🧱 **ネットワークセグメンテーション / ゼロトラスト**
+> 6. 🛡️ **データ窃取対策(DLP)の強化**
+> 7. 📋 **インシデント対応計画の策定と定期訓練**
+
+---
+
+*最終更新: 2026年5月24日*
diff --git a/docs/Cybersecurity/ransomware_survey.rag_ingested b/docs/Cybersecurity/ransomware_survey.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..5ca1386193cbd4100b9ced5dee4cb502840e5716
--- /dev/null
+++ b/docs/Cybersecurity/ransomware_survey.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "d44930ce2ab79660", "ingested_at": "2026-06-30T09:10:05.754605", "content_hash": "d44930ce2ab79660"}
\ No newline at end of file
diff --git a/docs/Cybersecurity/script_long.md b/docs/Cybersecurity/script_long.md
new file mode 100644
index 0000000000000000000000000000000000000000..8d42756d4f17db4e884ca9281197f37f2eca9e92
--- /dev/null
+++ b/docs/Cybersecurity/script_long.md
@@ -0,0 +1,18 @@
+Good evening, and welcome to our nightly news broadcast. Tonight's top story involves a comprehensive analysis of file encryption technology, as the field continues to evolve rapidly. We'll delve into the latest advancements in this crucial area, exploring everything from the most widely used algorithms to new tools that are shaping the future of data security.
+
+The foundation of modern file encryption lies in symmetric algorithms like AES-256 and ChaCha20. These algorithms provide robust protection for sensitive information. AES-256 is a popular choice due to its high performance and compatibility with hardware acceleration, while ChaCha20 offers excellent speed and ease of implementation.
+
+However, the landscape has shifted significantly in recent years. We're now seeing a strong emphasis on authentication-based encryption (AEAD) methods like AES-GCM and ChaCha20-Poly1305. These provide an extra layer of security by ensuring that only authorized users can decrypt data. We must also address the issue of nonce reuse, which poses significant risks to security if not handled properly.
+
+As a reminder, we're moving away from legacy methods like ECB mode and towards AEAD-based encryption for all new projects. This is crucial in protecting against potential attacks.
+
+The field of key derivation functions (KDFs) has also seen significant advancements. These functions are essential for generating secure encryption keys from passwords or other data sources. We're seeing a surge in the use of Argon2id, which offers exceptional resistance to brute-force attacks and is highly recommended for all new projects.
+
+A crucial aspect of file encryption is the choice of tools. We've seen a rise in specialized software designed for specific needs. VeraCrypt, BitLocker, LUKS/dm-crypt, FileVault are some examples of popular solutions for encrypting entire disks. For smaller files and folders, tools like age, GPG, Picocrypt, and AxCrypt offer flexible options.
+
+The cloud storage landscape is also seeing significant changes with the emergence of dedicated encryption tools such as Cryptomator and rclone crypt. These solutions provide a seamless way to encrypt data stored in the cloud. For those who prefer command-line interfaces, OpenSSL's encryption capabilities, age, GPG, and other powerful tools offer robust solutions.
+
+As we move forward into the future, it's important to stay informed about these advancements. This ongoing evolution of file encryption technology will continue to shape our digital world and ensure that our data remains secure.
+
+
+This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/Cybersecurity/script_long.rag_ingested b/docs/Cybersecurity/script_long.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..c25b2068d6e405361ae9470361ad91734deaf3ad
--- /dev/null
+++ b/docs/Cybersecurity/script_long.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "797fee345073152c", "ingested_at": "2026-06-30T09:10:08.794334", "content_hash": "797fee345073152c"}
\ No newline at end of file
diff --git a/docs/Cybersecurity/script_long_clean.txt b/docs/Cybersecurity/script_long_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..e0317b53707127d6fc415527cc605ad590a7af6e
--- /dev/null
+++ b/docs/Cybersecurity/script_long_clean.txt
@@ -0,0 +1,17 @@
+Good evening, and welcome to our nightly news broadcast. Tonight's top story involves a comprehensive analysis of file encryption technology, as the field continues to evolve rapidly. We'll delve into the latest advancements in this crucial area, exploring everything from the most widely used algorithms to new tools that are shaping the future of data security.
+
+The foundation of modern file encryption lies in symmetric algorithms like AES-256 and ChaCha20. These algorithms provide robust protection for sensitive information. AES-256 is a popular choice due to its high performance and compatibility with hardware acceleration, while ChaCha20 offers excellent speed and ease of implementation.
+
+However, the landscape has shifted significantly in recent years. We're now seeing a strong emphasis on authentication-based encryption methods like AES-GCM and ChaCha20-Poly1305. These provide an extra layer of security by ensuring that only authorized users can decrypt data. We must also address the issue of nonce reuse, which poses significant risks to security if not handled properly.
+
+As a reminder, we're moving away from legacy methods like ECB mode and towards AEAD-based encryption for all new projects. This is crucial in protecting against potential attacks.
+
+The field of key derivation functions has also seen significant advancements. These functions are essential for generating secure encryption keys from passwords or other data sources. We're seeing a surge in the use of Argon2id, which offers exceptional resistance to brute-force attacks and is highly recommended for all new projects.
+
+A crucial aspect of file encryption is the choice of tools. We've seen a rise in specialized software designed for specific needs. VeraCrypt, BitLocker, LUKS/dm-crypt, FileVault are some examples of popular solutions for encrypting entire disks. For smaller files and folders, tools like age, GPG, Picocrypt, and AxCrypt offer flexible options.
+
+The cloud storage landscape is also seeing significant changes with the emergence of dedicated encryption tools such as Cryptomator and rclone crypt. These solutions provide a seamless way to encrypt data stored in the cloud. For those who prefer command-line interfaces, OpenSSL's encryption capabilities, age, GPG, and other powerful tools offer robust solutions.
+
+As we move forward into the future, it's important to stay informed about these advancements. This ongoing evolution of file encryption technology will continue to shape our digital world and ensure that our data remains secure.
+
+This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/Cybersecurity/script_middle.md b/docs/Cybersecurity/script_middle.md
new file mode 100644
index 0000000000000000000000000000000000000000..a92c411faad23c26edcb287300f995125c3c2302
--- /dev/null
+++ b/docs/Cybersecurity/script_middle.md
@@ -0,0 +1,18 @@
+Good evening, and welcome back. Today we're diving into the world of file encryption, a crucial aspect of data security in today's digital landscape. A new report has been released that provides an in-depth analysis of the latest advancements in this field.
+
+The report, "File Encryption Technology Survey (2026 Edition)," offers a comprehensive overview of how we encrypt files and protect sensitive information. Let's break down some key takeaways from this survey.
+
+First off, the report highlights that symmetric encryption algorithms like AES-256 and ChaCha20 remain industry standards for file encryption. These algorithms offer robust protection, with AES-256 being particularly favored due to its hardware acceleration capabilities.
+
+The report emphasizes the importance of using authenticated encryption (AEAD) over traditional methods like CBC mode. AEAD offers a significant boost in security by ensuring data integrity and authenticity. It's crucial to use AEAD for all new projects, as it ensures both confidentiality and non-repudiation.
+
+Another key aspect is the selection of strong cryptographic keys. The report recommends using CSPRNG (Cryptographic Secure Pseudo Random Number Generator) for generating these keys. This ensures that your encryption keys are truly random and unpredictable.
+
+The report also delves into the world of Key Derivation Functions (KDFs), which are essential in creating secure encryption keys from passwords. It highlights the importance of robust KDFs, especially against brute-force attacks. The report recommends using Argon2id as a gold standard for KDFs due to its side channel resistance and high performance.
+
+Finally, we'll explore the various tools available for file encryption. These range from disk encryption solutions like VeraCrypt and BitLocker to more specialized options like GPG or AxCrypt for individual files and folders. For cloud storage, tools like Cryptomator and rclone crypt offer robust encryption capabilities.
+
+
+This report provides a comprehensive overview of the latest advancements in file encryption technology. By understanding these key concepts and tools, we can ensure that our data is protected from unauthorized access.
+
+And that's all for tonight's news update. Thank you for watching. This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/Cybersecurity/script_middle.rag_ingested b/docs/Cybersecurity/script_middle.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..b0f7d8d56ae3244f7d52e1427d468ba8779f148b
--- /dev/null
+++ b/docs/Cybersecurity/script_middle.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "89840b4cc17549f3", "ingested_at": "2026-06-30T09:10:11.133917", "content_hash": "89840b4cc17549f3"}
\ No newline at end of file
diff --git a/docs/Cybersecurity/script_middle_clean.txt b/docs/Cybersecurity/script_middle_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..6400f8f5521f2803d55fe7fff46a09556035d37e
--- /dev/null
+++ b/docs/Cybersecurity/script_middle_clean.txt
@@ -0,0 +1,17 @@
+Good evening, and welcome back. Today we're diving into the world of file encryption, a crucial aspect of data security in today's digital landscape. A new report has been released that provides an in-depth analysis of the latest advancements in this field.
+
+The report, "File Encryption Technology Survey ," offers a comprehensive overview of how we encrypt files and protect sensitive information. Let's break down some key takeaways from this survey.
+
+First off, the report highlights that symmetric encryption algorithms like AES-256 and ChaCha20 remain industry standards for file encryption. These algorithms offer robust protection, with AES-256 being particularly favored due to its hardware acceleration capabilities.
+
+The report emphasizes the importance of using authenticated encryption over traditional methods like CBC mode. AEAD offers a significant boost in security by ensuring data integrity and authenticity. It's crucial to use AEAD for all new projects, as it ensures both confidentiality and non-repudiation.
+
+Another key aspect is the selection of strong cryptographic keys. The report recommends using CSPRNG for generating these keys. This ensures that your encryption keys are truly random and unpredictable.
+
+The report also delves into the world of Key Derivation Functions , which are essential in creating secure encryption keys from passwords. It highlights the importance of robust KDFs, especially against brute-force attacks. The report recommends using Argon2id as a gold standard for KDFs due to its side channel resistance and high performance.
+
+Finally, we'll explore the various tools available for file encryption. These range from disk encryption solutions like VeraCrypt and BitLocker to more specialized options like GPG or AxCrypt for individual files and folders. For cloud storage, tools like Cryptomator and rclone crypt offer robust encryption capabilities.
+
+This report provides a comprehensive overview of the latest advancements in file encryption technology. By understanding these key concepts and tools, we can ensure that our data is protected from unauthorized access.
+
+And that's all for tonight's news update. Thank you for watching. This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/Cybersecurity/script_short.md b/docs/Cybersecurity/script_short.md
new file mode 100644
index 0000000000000000000000000000000000000000..e72c7954a58f3a2668737815f704630d1d329663
--- /dev/null
+++ b/docs/Cybersecurity/script_short.md
@@ -0,0 +1,10 @@
+Good evening, and welcome back for a quick look at the latest developments in file encryption technology. The world of data security is constantly evolving, and today we're diving into the key technologies shaping how we protect our information.
+
+First up: **symmetric encryption** - the bedrock of most file encryption systems. We're talking about AES-256 and ChaCha20, industry standards that offer robust protection for your data. But remember, security is not just about the algorithm; it's also about implementation. That's why we need **AEAD** - Authentication-based Encryption. This ensures our data remains secure even if someone tries to break into our system.
+
+Now, let's talk about **Key Derivation Functions (KDF)**. These are essential for generating strong encryption keys from your passwords. We're talking about Argon2id here, a gold standard in cybersecurity. It's super secure and easy to implement.
+
+And finally, we have the tools themselves! There are tons of file encryption tools out there, each with its own strengths and weaknesses. From VeraCrypt for full disk encryption to GPG for securing your files online, the best tool for you will depend on your specific needs.
+
+
+That's it for today's quick look at file encryption. For more in-depth information and analysis, check out our latest report. This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/Cybersecurity/script_short.rag_ingested b/docs/Cybersecurity/script_short.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..49a3dc8717d7515fd69a50f8f22fbe0fcc340f36
--- /dev/null
+++ b/docs/Cybersecurity/script_short.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "c671517f14b93a44", "ingested_at": "2026-06-30T09:10:12.548294", "content_hash": "c671517f14b93a44"}
\ No newline at end of file
diff --git a/docs/Cybersecurity/script_short_clean.txt b/docs/Cybersecurity/script_short_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..63c43596733c2636134a320cf175b42732f17753
--- /dev/null
+++ b/docs/Cybersecurity/script_short_clean.txt
@@ -0,0 +1,9 @@
+Good evening, and welcome back for a quick look at the latest developments in file encryption technology. The world of data security is constantly evolving, and today we're diving into the key technologies shaping how we protect our information.
+
+symmetric encryption - the bedrock of most file encryption systems. We're talking about AES-256 and ChaCha20, industry standards that offer robust protection for your data. But remember, security is not just about the algorithm; it's also about implementation. That's why we need AEAD - Authentication-based Encryption. This ensures our data remains secure even if someone tries to break into our system.
+
+Now, let's talk about Key Derivation Functions . These are essential for generating strong encryption keys from your passwords. We're talking about Argon2id here, a gold standard in cybersecurity. It's super secure and easy to implement.
+
+And finally, we have the tools themselves! There are tons of file encryption tools out there, each with its own strengths and weaknesses. From VeraCrypt for full disk encryption to GPG for securing your files online, the best tool for you will depend on your specific needs.
+
+That's it for today's quick look at file encryption. For more in-depth information and analysis, check out our latest report. This has been brought to you by The Android Times.
\ No newline at end of file
diff --git a/docs/Cybersecurity/steganography_survey.md b/docs/Cybersecurity/steganography_survey.md
new file mode 100644
index 0000000000000000000000000000000000000000..be1f0f43d69d070c1c85d858306d9328eea0f47e
--- /dev/null
+++ b/docs/Cybersecurity/steganography_survey.md
@@ -0,0 +1,579 @@
+# ステガノグラフィー(Steganography)包括的サーベイ
+
+> **作成日**: 2026-05-24
+> **対象期間**: 歴史的起源 〜 2025–2026年最新動向
+
+---
+
+## 目次
+
+1. [概要と定義](#1-概要と定義)
+2. [歴史的背景](#2-歴史的背景)
+3. [技術分類(タクソノミー)](#3-技術分類タクソノミー)
+4. [ドメイン別技術詳細](#4-ドメイン別技術詳細)
+ - 4.1 画像ステガノグラフィー
+ - 4.2 音声ステガノグラフィー
+ - 4.3 動画ステガノグラフィー
+ - 4.4 テキスト/言語ステガノグラフィー
+ - 4.5 ネットワークステガノグラフィー
+5. [生成型ステガノグラフィー](#5-生成型ステガノグラフィー)
+6. [ステガナリシス(検出技術)](#6-ステガナリシス検出技術)
+7. [評価指標とベンチマーク](#7-評価指標とベンチマーク)
+8. [応用分野](#8-応用分野)
+9. [サイバーセキュリティ脅威](#9-サイバーセキュリティ脅威)
+10. [主要ツール・フレームワーク](#10-主要ツールフレームワーク)
+11. [今後の研究課題と展望](#11-今後の研究課題と展望)
+12. [参考文献・リソース](#12-参考文献リソース)
+
+---
+
+## 1. 概要と定義
+
+**ステガノグラフィー**(Steganography)とは、秘密メッセージの**存在自体**を隠蔽する技術である。ギリシャ語の *steganós*(「隠された」)と *graphía*(「書くこと」)に由来する。
+
+> [!IMPORTANT]
+> **暗号化との違い**: 暗号化(Cryptography)はメッセージの**内容**を保護するのに対し、ステガノグラフィーは通信の**存在そのもの**を隠す。実用上は両者を組み合わせ、まず暗号化してからステガノグラフィーで埋め込むアプローチが一般的。
+
+### 基本構成要素
+
+```mermaid
+graph LR
+ A["秘密メッセージ
(Secret Message)"] --> C["埋め込みアルゴリズム
(Embedding)"]
+ B["カバーオブジェクト
(Cover Object)"] --> C
+ K["鍵 (Key)"] --> C
+ C --> D["ステゴオブジェクト
(Stego Object)"]
+ D --> E["抽出アルゴリズム
(Extraction)"]
+ K2["鍵 (Key)"] --> E
+ E --> F["秘密メッセージ復元"]
+```
+
+### 三大トレードオフ
+
+| 要素 | 説明 |
+|:---|:---|
+| **容量(Capacity)** | 埋め込み可能なデータ量(BPP: Bits Per Pixel 等) |
+| **非知覚性(Imperceptibility)** | 人間の感覚では変化を検知できないこと |
+| **堅牢性(Robustness)** | 圧縮・変換・攻撃に対する耐性 |
+
+これら3要素は互いにトレードオフの関係にあり、全てを同時に最大化することはできない。
+
+---
+
+## 2. 歴史的背景
+
+### 古代〜中世
+
+| 時代 | 手法 | 詳細 |
+|:---|:---|:---|
+| **紀元前440年頃** | 頭皮への刺青 | ヘロドトスが記録。奴隷の剃った頭にメッセージを刺青し、髪が伸びた後に送り出した |
+| **紀元前5世紀** | 蝋板 | デマラトスが蝋板の木の部分に直接メッセージを書き、蝋で覆い隠した |
+| **古代ローマ** | 不可視インク | レモン汁・ミルクなどを用いた不可視インクによる通信 |
+| **1499年** | Steganographia | ヨハネス・トリテミウスが著作『Steganographia』を発表。魔術書に偽装した暗号学の書 |
+
+### 近代
+
+| 時代 | 手法 | 詳細 |
+|:---|:---|:---|
+| **第一次・第二次世界大戦** | マイクロドット | 写真を極小サイズに縮小し、句読点に偽装して文書に埋め込んだ |
+| **同上** | 編み物パターン | モールス信号を編み物の目に変換して情報伝達 |
+| **冷戦期** | 不可視インクの高度化 | 化学的手法による検出困難な不可視インクの開発 |
+
+### デジタル時代(1990年代〜)
+
+- **1990年代**: LSB(最下位ビット)置換法の登場。デジタル画像・音声ファイルへの情報埋め込み
+- **2000年代**: 周波数変換領域(DCT, DWT)手法の発展。JPEG画像への堅牢な埋め込み
+- **2010年代**: 機械学習ベースの手法が登場。コスト関数最適化(HUGO, WOW, S-UNIWARD)
+- **2020年代**: 深層学習(GAN, Diffusion Model)による生成型ステガノグラフィーの台頭
+
+---
+
+## 3. 技術分類(タクソノミー)
+
+```mermaid
+graph TD
+ S["ステガノグラフィー"] --> D["ドメイン別"]
+ S --> M["手法別"]
+ S --> F["機能別"]
+
+ D --> D1["画像"]
+ D --> D2["音声"]
+ D --> D3["動画"]
+ D --> D4["テキスト/言語"]
+ D --> D5["ネットワーク"]
+ D --> D6["ファイルシステム"]
+
+ M --> M1["空間領域
(Spatial Domain)"]
+ M --> M2["周波数/変換領域
(Transform Domain)"]
+ M --> M3["深層学習ベース
(Deep Learning)"]
+
+ F --> F1["修正型
(Modification-based)"]
+ F --> F2["生成型
(Generative/Coverless)"]
+
+ M1 --> M1a["LSB置換"]
+ M1 --> M1b["PVD法"]
+ M2 --> M2a["DCT"]
+ M2 --> M2b["DWT"]
+ M2 --> M2c["SVD"]
+ M3 --> M3a["CNN/U-Net"]
+ M3 --> M3b["GAN"]
+ M3 --> M3c["拡散モデル"]
+ M3 --> M3d["Transformer"]
+```
+
+### 主要な分類軸
+
+| 分類軸 | カテゴリ | 説明 |
+|:---|:---|:---|
+| **ドメイン** | 画像/音声/動画/テキスト/ネットワーク/ファイルシステム | 隠蔽に使用するカバーメディアの種類 |
+| **埋め込み手法** | 空間領域 vs 変換領域 | ピクセル値を直接操作 vs 周波数成分を操作 |
+| **アプローチ** | 手工芸的(Traditional) vs 深層学習ベース | コスト関数設計 vs ニューラルネットワーク学習 |
+| **機能** | 修正型(Dominant) vs 生成型(Generative) | 既存カバーを改変 vs 新しいカバーを生成 |
+
+> [!NOTE]
+> 2025年の研究コミュニティでは、Wendzel et al. による「Generic Taxonomy for Steganography Methods」が標準的な分類フレームワークとして推奨されている。Information Hiding Patterns プロジェクト(patterns.ztt.hs-worms.de)も参照のこと。
+
+---
+
+## 4. ドメイン別技術詳細
+
+### 4.1 画像ステガノグラフィー
+
+最も研究が盛んな領域。カバー画像のピクセルデータに秘密情報を埋め込む。
+
+#### 4.1.1 空間領域手法
+
+| 手法 | 原理 | 長所 | 短所 |
+|:---|:---|:---|:---|
+| **LSB置換** | 各ピクセルの最下位ビットを秘密データで置換 | 実装が簡単、高容量 | 統計的解析に弱い |
+| **LSBマッチング** | ±1のランダムな変更でLSBを調整 | LSB置換より検出困難 | 容量がやや低い |
+| **PVD法** | 隣接ピクセルの差分値に情報を埋め込む | 適応的な容量配分 | 複雑な実装 |
+| **HUGO** | High-Dimensional cost function. 近傍の統計的特徴量への影響を最小化 | 高いセキュリティ | 計算コスト大 |
+| **WOW** | Wavelet Obtained Weights. ウェーブレットフィルタバンクでコストを計算 | テクスチャ適応型 | 実装が複雑 |
+
+#### 4.1.2 変換領域手法
+
+| 手法 | 変換 | 特徴 |
+|:---|:---|:---|
+| **DCTベース** | 離散コサイン変換 | JPEG圧縮との親和性が高い。JSteg, F5, nsF5 等 |
+| **DWTベース** | 離散ウェーブレット変換 | 多解像度解析。高周波成分への埋め込みで堅牢性確保 |
+| **SVDベース** | 特異値分解 | 特異値の微小変更により高堅牢性を実現 |
+
+#### 4.1.3 深層学習ベース手法(2024–2025 State-of-the-Art)
+
+```mermaid
+graph LR
+ subgraph "End-to-End Framework"
+ E["Encoder
(埋め込みネットワーク)"] --> S["Stego Image"]
+ S --> N["Noise Layer
(JPEG圧縮等のシミュレーション)"]
+ N --> D["Decoder
(抽出ネットワーク)"]
+ S --> Disc["Discriminator
(敵対的判別器)"]
+ end
+ Cover["Cover Image"] --> E
+ Secret["Secret Data"] --> E
+ D --> Recovered["Recovered Secret"]
+```
+
+**主要アーキテクチャ**:
+
+- **CNN/U-Netベース**: エンコーダ・デコーダ構造でend-to-end学習。HiDDeN, SteganoGAN 等
+- **GANベース**: 生成器(埋め込み)と判別器(ステガナリシス)の敵対的学習
+- **Transformerベース**: 長距離依存関係の捕捉によるより精密な埋め込み位置選択
+- **拡散モデルベース**: 高品質な画像生成能力を活用したカバーレスステガノグラフィー
+
+**2024–2025の主要トレンド**:
+- **適応型マルチエンコーダ・デコーダ**: 異なるタイプの秘密データに動的に対応
+- **堅牢性の追求**: JPEG圧縮・リサイズ・クロッピングなど実世界の画像操作への耐性
+- **AUC ≈ 0.5 を目標**: ステガナリシスの検出率をランダム推測レベルに抑える設計
+
+---
+
+### 4.2 音声ステガノグラフィー
+
+音声ファイル(WAV, MP3等)に情報を埋め込む技術。
+
+| 手法 | 説明 |
+|:---|:---|
+| **LSB符号化** | 音声サンプルの最下位ビットに埋め込み |
+| **位相符号化** | 音声セグメントの位相成分を変更 |
+| **エコーハイディング** | 人間の聴覚で感知しにくいエコーを付加 |
+| **スペクトル拡散** | 秘密データを広帯域に拡散して埋め込み |
+| **DWT/DCT変換** | 周波数領域での埋め込みにより圧縮耐性を確保 |
+
+**2024–2025の動向**:
+- AES暗号化とLSB音声ステガノグラフィーのハイブリッドアプローチ
+- カオス写像を用いた二重レイヤーセキュリティ
+- AI/ニューラルネットワークによる適応的埋め込み位置の選択
+
+---
+
+### 4.3 動画ステガノグラフィー
+
+動画は画像(フレーム)と音声の両方を含むため、大容量の埋め込みが可能。
+
+| 手法 | 説明 |
+|:---|:---|
+| **フレームベース** | 個別フレームに画像ステガノグラフィーを適用 |
+| **動き補償ベクトル** | H.264/H.265のモーションベクトルに情報を埋め込み |
+| **メタデータベース(VSTM)** | MP4のコメントフィールド等のメタデータに埋め込み。画質劣化なし |
+| **時間的・エッジガイダンス** | エッジ領域に集中して埋め込むことで知覚性を低下 |
+
+**2024–2025のブレークスルー**:
+- 高解像度動画(1080p以上)への対応
+- ソーシャルメディアの圧縮・再エンコードに対する堅牢性
+- 深層学習を用いた時間軸方向の適応的埋め込み
+
+---
+
+### 4.4 テキスト/言語ステガノグラフィー
+
+自然言語テキストに秘密情報を隠蔽する技術。LLMの台頭により急速に進化。
+
+#### 従来手法
+
+| 手法 | 説明 |
+|:---|:---|
+| **フォーマットベース** | 行間隔・フォントサイズ・空白文字の調整 |
+| **言語的手法** | 同義語置換・文構造変換による埋め込み |
+| **Unicode手法** | ゼロ幅文字(ZWSP, ZWNJ等)の挿入 |
+
+#### LLMベース手法(2024–2025 最先端)
+
+```mermaid
+graph TD
+ subgraph "White-Box"
+ W1["共有LLMの確率分布にアクセス"] --> W2["トークン確率の操作で
ビット列を符号化"]
+ end
+ subgraph "Black-Box (LLM-Stega等)"
+ B1["LLM API/UIのみ使用"] --> B2["プロンプトエンジニアリング"]
+ B2 --> B3["拒否サンプリング + 暗号化マッピング"]
+ end
+ subgraph "生成型"
+ G1["秘密メッセージ → プロンプト変換"] --> G2["LLMが自然なテキストを生成"]
+ G2 --> G3["ステゴテキスト"]
+ end
+```
+
+**主要手法**:
+- **LLsM**: ランクベース符号化方式。LLMのトークンランキングを利用
+- **LLM-Stega**: ブラックボックスアクセスのみで動作。プロンプトエンジニアリングと拒否サンプリングを活用
+- **Calgacus**: ランクベースの暗号化ステガノグラフィー。「雪崩効果」によるセキュリティ強化
+
+**重大な懸念: AIエージェント間のステガノグラフィー的共謀**
+
+> [!CAUTION]
+> 2024–2025年の研究で、LLMが明示的な指示なしにステガノグラフィーを**自律的に発展・使用**する可能性が示された(NeurIPS 2024, ICLR 2025)。AIエージェント間の隠密な通信は、標準的なコンテンツモデレーションやアラインメント監視を回避するリスクがあり、AIセーフティの重要課題となっている。
+
+---
+
+### 4.5 ネットワークステガノグラフィー
+
+ネットワーク通信プロトコルやトラフィックパターンに情報を隠蔽する技術。
+
+| 手法 | 説明 |
+|:---|:---|
+| **プロトコルヘッダ操作** | IP/TCP/UDP ヘッダの未使用・予約フィールドを利用 |
+| **DNSトンネリング** | DNSクエリ・レスポンスに情報を符号化 |
+| **HTTPSカバートチャネル** | TLS暗号化通信内にデータを隠蔽 |
+| **タイミングチャネル** | パケット間隔やビットレート変調でビット列を表現 |
+| **VoIPステガノグラフィー** | リアルタイム音声通信のプロトコルを利用 |
+| **QUICプロトコル悪用** | 新しい暗号化プロトコルの不透明性を利用 |
+
+---
+
+## 5. 生成型ステガノグラフィー
+
+既存のカバーオブジェクトを改変するのではなく、秘密メッセージから**新しいカバーを生成**する手法。カバーレス(Coverless)ステガノグラフィーとも呼ばれる。
+
+### 5.1 GANベース生成型
+
+```mermaid
+graph LR
+ M["秘密メッセージ"] --> G["Generator
(生成器)"]
+ Z["ノイズ z"] --> G
+ G --> SI["ステゴ画像"]
+ SI --> D["Discriminator
(判別器)"]
+ NI["自然画像"] --> D
+ D --> FB["フィードバック"]
+ FB --> G
+```
+
+- 秘密メッセージを生成器の潜在空間に埋め込み
+- 判別器が自然画像との区別不能性を学習
+- ペイロード(BPP)と画質のトレードオフ最適化
+
+### 5.2 拡散モデルベース生成型(2024–2025 最先端)
+
+GANに代わる新しいパラダイムとして急速に台頭。
+
+| 手法 | 特徴 |
+|:---|:---|
+| **GSD** | ノイズ領域での埋め込みと逆拡散プロセスによる復元 |
+| **StegoDiffusion** | 周波数領域での拡散プロセスの再調整 |
+| **MDDM** | **トレーニング不要**。カルダングリルを用いてガウシアンノイズにメッセージ符号化。制御可能なプロンプトによる画像生成 |
+
+**利点**:
+- GANと比較して高品質・多様な画像生成
+- ノイズ→画像の生成プロセスに自然にメッセージを統合
+- 制御可能なプロンプトによるステゴ画像の内容指定
+
+**課題**:
+- 拡散プロセスの完全な可逆性の保証が困難
+- 計算コストがGANより高い
+- 容量と品質のバランス
+
+### 5.3 新しいメディアタイプへの拡張
+
+- **ポイントクラウド**: 3D空間データへの生成型ステガノグラフィー
+- **テキスト(GTSD)**: プロンプトマッピングとバッチ選択による生成型テキストステガノグラフィー
+
+---
+
+## 6. ステガナリシス(検出技術)
+
+ステゴオブジェクト内の隠蔽データの存在を検出する技術。ステガノグラフィーとは「いたちごっこ」の関係。
+
+### 6.1 従来手法
+
+| 手法 | 説明 |
+|:---|:---|
+| **RS分析** | Regular/Singular グループの統計的特性を利用したLSB検出 |
+| **χ²検定** | 隣接値ペア(PoV)の統計的偏りを検出 |
+| **SPA(Sample Pair Analysis)** | サンプルペアの相関分析 |
+| **SRM(Spatial Rich Model)** | 多数のフィルタで高次元特徴量を抽出 → SVM分類器 |
+
+### 6.2 深層学習ベース手法(2024–2025 State-of-the-Art)
+
+```mermaid
+graph LR
+ I["入力画像"] --> HP["ハイパスフィルタ
(前処理)"]
+ HP --> CNN["CNN特徴抽出"]
+ CNN --> CLS["分類器"]
+ CLS --> R["Cover / Stego"]
+
+ subgraph "最新アーキテクチャ"
+ CNN
+ LSTM["BiLSTM
(時系列特徴)"]
+ ATT["Attention機構"]
+ CNN --> LSTM
+ LSTM --> ATT
+ ATT --> CLS
+ end
+```
+
+**主要モデルと発展**:
+
+| モデル/手法 | 特徴 |
+|:---|:---|
+| **SRNet** | 残差学習ベース。ステガノグラフィーの微小な残差パターンを学習 |
+| **Ye-Net** | ハイパスフィルタの初期層への統合 |
+| **CNN + BiLSTM** | 空間的特徴と時系列的特徴の融合 |
+| **自己教師あり学習(SSL)** | SRMフィルタと組み合わせた汎化性能の向上 |
+| **軽量化モデル** | プルーニング・量子化・重みクラスタリングによる高速化 |
+
+**2024–2025の主要課題**:
+
+> [!WARNING]
+> - **汎化問題**: BOSSbase等の特定データセットに過学習し、実世界の異種画像に対して性能が大幅に低下する「分布外」問題
+> - **AI生成ステガノグラフィーへの対応**: GANや拡散モデルで生成されたステゴ画像は、従来の統計的ステガナリシスが無効化される場合がある
+> - **計算効率**: 高精度モデルの軽量化と、リソース制約下でのリアルタイム検出
+
+---
+
+## 7. 評価指標とベンチマーク
+
+### 7.1 主要評価指標
+
+| 指標 | 定義 | 用途 |
+|:---|:---|:---|
+| **PSNR (dB)** | $10 \cdot \log_{10}\left(\frac{MAX^2}{MSE}\right)$ | 画質評価。一般に30dB以上で知覚困難 |
+| **SSIM** | 構造的類似性(輝度・コントラスト・構造の比較) | 知覚的品質。1.0に近いほど良好 |
+| **BPP** | Bits Per Pixel = 埋め込みビット数 / 総ピクセル数 | 埋め込み容量 |
+| **検出誤差率** | $P_E = \frac{1}{2}(P_{FA} + P_{MD})$ | ステガナリシスに対するセキュリティ |
+| **AUC** | ROC曲線下面積 | 0.5に近いほどステガナリシスを欺けている |
+| **SRA** | Secret Recovery Accuracy(秘密データ復元精度) | 抽出の正確性 |
+| **FID** | Fréchet Inception Distance | 生成型手法の画像品質 |
+
+### 7.2 主要ベンチマークデータセット
+
+| データセット | 内容 | 用途 |
+|:---|:---|:---|
+| **BOSSbase 1.01** | 10,000枚のグレースケール画像(512×512) | ステガノグラフィー/ステガナリシスの標準ベンチマーク |
+| **BOWS2** | BOSSbaseの補完用データセット | クロスデータセット評価 |
+| **ImageNet** | 大規模自然画像データセット | 深層学習ベース手法の学習・評価 |
+| **COCO** | 物体検出用データセット | 多様な画像での汎化性能評価 |
+| **DIV2K** | 高解像度画像データセット | 高解像度ステガノグラフィーの評価 |
+
+### 7.3 典型的な性能比較
+
+| 手法カテゴリ | 典型的PSNR (dB) | 典型的BPP | セキュリティ |
+|:---|:---|:---|:---|
+| LSB置換 | 50–55 | 1.0 | 低(容易に検出可能) |
+| コスト関数最適化(S-UNIWARD) | 38–42 | 0.2–0.4 | 高 |
+| 深層学習ベース(HiDDeN等) | 33–40 | 1.0–4.0 | 中〜高 |
+| GANベース生成型 | 28–35 | 0.5–2.0 | 高 |
+| 拡散モデルベース | 30–38 | 0.5–3.0 | 非常に高 |
+
+---
+
+## 8. 応用分野
+
+### 8.1 正当な応用
+
+| 分野 | 応用例 |
+|:---|:---|
+| **著作権保護** | デジタル透かし(Watermarking)による所有権の主張・追跡 |
+| **認証・改ざん検出** | 画像・文書の完全性を検証するフラジャイル透かし |
+| **秘密通信** | 外交・軍事・ジャーナリズムにおける秘匿通信 |
+| **検閲回避** | インターネット検閲下での情報の自由な流通 |
+| **プライバシー保護** | 医療画像への患者情報の安全な埋め込み |
+| **指紋追跡(Fingerprinting)** | コンテンツのコピーごとに一意の識別子を埋め込み、流出元を特定 |
+
+### 8.2 ステガノグラフィー vs デジタル透かし
+
+| 特性 | ステガノグラフィー | デジタル透かし |
+|:---|:---|:---|
+| **主目的** | メッセージの存在を隠す | 所有権・著作権の保護 |
+| **関係性** | カバーは秘密メッセージの「運び屋」 | 埋め込み情報はカバーに関連 |
+| **堅牢性** | 一般に不要(脆弱でも可) | 必須(攻撃・変換に耐えなければならない) |
+| **通信モデル** | 1対1(秘密通信) | 1対多(配信・放送) |
+
+---
+
+## 9. サイバーセキュリティ脅威
+
+> [!CAUTION]
+> ステガノグラフィーは、サイバー攻撃者にとって検出回避の強力な手段となっている。2024–2025年にはAI駆動のステガノグラフィー攻撃が急増。
+
+### 9.1 主要脅威カテゴリ
+
+```mermaid
+graph TD
+ T["ステガノグラフィーを利用した
サイバー脅威"] --> T1["ステゴマルウェア
(Stegomalware)"]
+ T --> T2["C2通信
(Command & Control)"]
+ T --> T3["データ窃取
(Data Exfiltration)"]
+ T --> T4["ステガノグラフィーボット"]
+ T --> T5["サプライチェーン攻撃"]
+
+ T1 --> T1a["画像内に悪意のある
ペイロードを隠蔽"]
+ T2 --> T2a["DNS/HTTP/HTTPS内に
C2コマンドを埋め込み"]
+ T3 --> T3a["機密データを画像/動画に
埋め込んで持ち出し"]
+ T4 --> T4a["AI駆動の自動化された
ステガノグラフィー攻撃"]
+ T5 --> T5a["正規ソフトウェアの
画像アセット等に埋め込み"]
+```
+
+### 9.2 実際の攻撃事例(2024–2025)
+
+| 脅威 | 手法 | 詳細 |
+|:---|:---|:---|
+| **ステゴマルウェア** | 画像内マルウェア | PNGやJPEG画像内にシェルコードやマルウェアローダーを埋め込み。従来のアンチウイルスをバイパス |
+| **多段階攻撃** | モジュラー感染チェーン | 初期ペイロードがステガノグラフィーで隠蔽された追加コンポーネントをダウンロード |
+| **IoT/クラウド標的** | プロトコル悪用 | リソース制約のあるIoTデバイスのセキュリティ監視の限界を悪用 |
+| **プロキシジャッキング** | 正規インフラ悪用 | Discord等のコラボレーションツールを介したC2通信とペイロード配信 |
+
+### 9.3 防御戦略
+
+| 対策 | 説明 |
+|:---|:---|
+| **行動分析** | 異常なトラフィックパターン・プロトコル使用の監視 |
+| **AI駆動防御** | 機械学習によるステガノグラフィーの自動検出 |
+| **ディープパケットインスペクション** | プロトコルメタデータとタイミング変動の分析 |
+| **ゼロトラストアーキテクチャ** | 全トラフィックの検証と横方向移動の制限 |
+| **ファイルサニタイゼーション** | アップロード/ダウンロードファイルのメタデータ除去・再エンコード |
+
+---
+
+## 10. 主要ツール・フレームワーク
+
+### 10.1 オープンソースツール
+
+| ツール | 種別 | 特徴 |
+|:---|:---|:---|
+| **Steghide** | CLI | JPEG/BMP/WAV/AUに対応。暗号化・圧縮内蔵。LSBステガノグラフィーのリファレンス実装 |
+| **OpenStego** | GUI/CLI (Java) | クロスプラットフォーム。データ隠蔽と透かしの両方をサポート。プラグインアーキテクチャ。2026年まで更新 |
+| **zsteg** | CLI (Ruby) | PNG/BMPのLSBステガノグラフィー検出に特化。フォレンジック・CTFで重宝 |
+| **Binwalk** | CLI | バイナリデータ内の埋め込みファイルの検出・抽出。ファームウェア解析にも利用 |
+| **ExifTool** | CLI | ファイルメタデータの表示・編集。メタデータベースの基本的なステガノグラフィーに関連 |
+| **Stego Toolkit** | Docker | steghide, zsteg, outguess, binwalk等を統合したDockerコンテナ。CTF向け |
+| **OutGuess** | CLI | JPEG画像向け。統計的検出への耐性を考慮した設計 |
+
+### 10.2 研究用フレームワーク
+
+| フレームワーク | 特徴 |
+|:---|:---|
+| **SteganographyToolkit (Python/PyTorch)** | 深層学習ベースの画像ステガノグラフィー実装群 |
+| **SteganoGAN** | PyTorchベースのGANステガノグラフィーフレームワーク |
+| **HiDDeN** | end-to-endの学習可能な画像ステガノグラフィー |
+| **LAMA (SRM)** | Spatial Rich Modelによるステガナリシス |
+
+---
+
+## 11. 今後の研究課題と展望
+
+### 11.1 技術的課題
+
+| 課題 | 詳細 |
+|:---|:---|
+| **汎化性能** | 特定データセットへの過学習を克服し、実世界の多様な画像に対応 |
+| **計算効率** | 高精度な深層学習モデルの軽量化(プルーニング・量子化) |
+| **可逆性** | 拡散モデルベース手法における完全な可逆性の保証 |
+| **マルチモーダル対応** | テキスト・画像・音声・動画を横断する統一的フレームワーク |
+| **量子コンピュータ耐性** | 量子計算時代のステガノグラフィー/ステガナリシスの再設計 |
+
+### 11.2 社会的・倫理的課題
+
+| 課題 | 詳細 |
+|:---|:---|
+| **AIエージェントの共謀** | LLMが自律的にステガノグラフィーを発展させるリスクへの対処 |
+| **悪用の民主化** | AI駆動ツールの普及によるステガノグラフィー攻撃の参入障壁低下 |
+| **規制・法律** | ステガノグラフィーの使用に関する法的枠組みの整備 |
+| **プライバシーvs安全保障** | 検閲回避ツールとしての価値と、犯罪利用リスクのバランス |
+
+### 11.3 今後の研究方向性
+
+```mermaid
+graph LR
+ F1["拡散モデルの高度化"] --> FG["次世代ステガノグラフィー"]
+ F2["マルチモーダルLLM統合"] --> FG
+ F3["敵対的堅牢性の強化"] --> FG
+ F4["リアルタイム動画ステガノグラフィー"] --> FG
+ F5["量子ステガノグラフィー"] --> FG
+ F6["AIセーフティとの統合"] --> FG
+```
+
+1. **拡散モデルの高度化**: トレーニング不要の手法、制御可能な生成、高容量化
+2. **マルチモーダルLLMの統合**: テキスト・画像・音声を横断する生成型ステガノグラフィー
+3. **敵対的堅牢性**: ステガナリシスとの軍拡競争における防御力の向上
+4. **リアルタイム動画ストリーミング**: 低遅延での動画ステガノグラフィーの実現
+5. **AIセーフティとの統合**: AIエージェント間の隠密通信の検出・防止
+6. **標準化の推進**: 統一的なタクソノミーと評価プロトコルの確立
+
+---
+
+## 12. 参考文献・リソース
+
+### 学術データベース・リポジトリ
+- **arXiv**: [arxiv.org](https://arxiv.org/) — 最新のプレプリント論文
+- **IEEE Xplore**: [ieeexplore.ieee.org](https://ieeexplore.ieee.org/) — 査読付き論文
+- **MDPI**: [mdpi.com](https://www.mdpi.com/) — オープンアクセスジャーナル
+
+### 主要会議
+- **ACM IH&MMSec**: Information Hiding and Multimedia Security(ステガノグラフィー専門の主要会議)
+- **IEEE WIFS**: Workshop on Information Forensics and Security
+- **ICML / NeurIPS / ICLR**: 機械学習トップ会議(深層学習ベースの手法)
+
+### ベンチマーク・データセット
+- **BOSSbase 1.01**: [agents.fel.cvut.cz/boss/](http://agents.fel.cvut.cz/boss/)
+- **BOWS2**: [bows2.ec-lille.fr](http://bows2.ec-lille.fr/)
+
+### 標準化プロジェクト
+- **Information Hiding Patterns**: [patterns.ztt.hs-worms.de](https://patterns.ztt.hs-worms.de/) — ステガノグラフィー手法の統一的分類
+
+### 推奨入門書
+- I. Cox et al., *Digital Watermarking and Steganography* (2nd ed., Morgan Kaufmann)
+- J. Fridrich, *Steganography in Digital Media: Principles, Algorithms, and Applications* (Cambridge University Press)
+- Wendzel et al., "A Generic Taxonomy for Steganography Methods" (2025)
+
+---
+
+> [!TIP]
+> **初学者向けアドバイス**: 画像ステガノグラフィーから始め、LSB置換の基本を理解した上で、コスト関数最適化手法(S-UNIWARD等)→ 深層学習ベース手法へと段階的に学習を進めるのが効率的です。CTF(Capture The Flag)の練習問題も実践的な理解に非常に有効です。
diff --git a/docs/Cybersecurity/steganography_survey.rag_ingested b/docs/Cybersecurity/steganography_survey.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..6db975958f269fb4ab248d6945f8b78ce406f4ea
--- /dev/null
+++ b/docs/Cybersecurity/steganography_survey.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "232db79eb399b3fb", "ingested_at": "2026-06-30T09:10:19.539427", "content_hash": "232db79eb399b3fb"}
\ No newline at end of file
diff --git a/docs/Documentation_and_API/analytics_reporting.json b/docs/Documentation_and_API/analytics_reporting.json
new file mode 100644
index 0000000000000000000000000000000000000000..2b7801082bc474fd2e5ce9ab70da063db2d2f90c
--- /dev/null
+++ b/docs/Documentation_and_API/analytics_reporting.json
@@ -0,0 +1,5 @@
+{
+ "importance": 8,
+ "urgency": 8,
+ "tier": 1
+}
\ No newline at end of file
diff --git a/docs/Documentation_and_API/analytics_reporting.md b/docs/Documentation_and_API/analytics_reporting.md
new file mode 100644
index 0000000000000000000000000000000000000000..85a057ac2b61c0b1f6fb305c18cfad12bcee0763
--- /dev/null
+++ b/docs/Documentation_and_API/analytics_reporting.md
@@ -0,0 +1,68 @@
+# YouTube Analytics and Reporting APIs Reference
+
+The YouTube Analytics and Reporting APIs allow you to retrieve viewing statistics, popularity metrics, and other data for YouTube videos and channels. These APIs are essential for building custom dashboards and automating reporting tasks.
+
+## 1. Choosing the Right API
+
+| Feature | YouTube Analytics API | YouTube Reporting API |
+| :--- | :--- | :--- |
+| **Primary Use** | Targeted, real-time queries. | Bulk reports for large datasets. |
+| **Data Retrieval** | Specific data points via API calls. | Pre-defined, downloadable bulk files. |
+| **Response Time** | Faster for small, specific queries. | Slower (asynchronous bulk generation). |
+| **Storage** | Data is not stored by the API. | Supports retrieval and local storage. |
+| **Best For** | Real-time dashboards, user-specific stats. | Mining historical data, deep analysis. |
+
+---
+
+## 2. YouTube Analytics API (Targeted Queries)
+
+The Analytics API allows you to execute "targeted queries" to retrieve specific statistics.
+
+### Key Concepts:
+- **Dimensions**: How data is grouped (e.g., `video`, `country`, `date`, `ageGroup`).
+- **Metrics**: The values being measured (e.g., `views`, `likes`, `estimatedMinutesWatched`).
+- **Filters**: Narrow down the result set (e.g., `video==VIDEO_ID`).
+
+### Common Parameters:
+- `ids`: The channel or content owner ID (e.g., `channel==MINE` or `contentOwner==OWNER_ID`).
+- `startDate` / `endDate`: The time range for the report (YYYY-MM-DD).
+- `metrics`: A comma-separated list of metrics (e.g., `views,likes,comments`).
+- `dimensions`: A comma-separated list of dimensions (e.g., `day,country`).
+- `sort`: Order of the results.
+
+---
+
+## 3. YouTube Reporting API (Bulk Data)
+
+The Reporting API is designed for applications that need to process vast amounts of data by downloading bulk reports.
+
+### Workflow:
+1. **List Report Types**: Find out what reports are available (e.g., `channel_basic_a2`).
+2. **Create a Job**: Schedule a regular bulk report generation.
+3. **Download Reports**: Retrieve the generated CSV files once they are ready.
+
+### Advantages:
+- Highly scalable for big data applications.
+- Provides comprehensive datasets without the complexity of individual query management.
+
+---
+
+## 4. Authentication and Authorization
+
+Both APIs require **OAuth 2.0** for all requests. API Keys are **not** sufficient for accessing analytics data.
+
+### Required Scopes:
+- `https://www.googleapis.com/auth/yt-analytics.readonly`: View YouTube Analytics reports for your YouTube content.
+- `https://www.googleapis.com/auth/yt-analytics-monetary.readonly`: View monetary and non-monetary YouTube Analytics reports.
+- `https://www.googleapis.com/auth/youtube.readonly`: View your YouTube account.
+
+---
+
+## 5. Support & Resources
+- **Dimensions & Metrics Reference**: [Developers Guide](https://developers.google.com/youtube/analytics/sample-requests)
+- **API Explorer**: [Try it out](https://developers.google.com/apis-explorer/#p/youtubeAnalytics/v2/)
+- **Code Samples**: [GitHub - youtube/api-samples](https://github.com/youtube/api-samples)
+
+
+**Scores:** Importance:8, Urgency:8
+**Tier:** 1
diff --git a/docs/Documentation_and_API/analytics_reporting.rag_ingested b/docs/Documentation_and_API/analytics_reporting.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..dd73bee19b7289788d653425e864310657df4de5
--- /dev/null
+++ b/docs/Documentation_and_API/analytics_reporting.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "552896662cb76877", "ingested_at": "2026-06-30T09:10:21.963345", "content_hash": "552896662cb76877"}
\ No newline at end of file
diff --git a/docs/Documentation_and_API/analytics_reporting_script_long.md b/docs/Documentation_and_API/analytics_reporting_script_long.md
new file mode 100644
index 0000000000000000000000000000000000000000..0f1e686243710a4d5d4ab100e43fb14901c8036a
--- /dev/null
+++ b/docs/Documentation_and_API/analytics_reporting_script_long.md
@@ -0,0 +1,15 @@
+Good evening. Today we're diving into the world of YouTube data analysis with a look at the new YouTube Analytics and Reporting APIs. These tools are designed to give developers unprecedented access to information about their content on the platform.
+
+For years, creators have relied on YouTube's built-in analytics for insights into their audience, but now they can take it a step further. The new APIs offer more granular control over data collection and analysis.
+
+The YouTube Analytics API allows users to query specific statistics like views, likes, and estimated minutes watched. Think of this as getting the precise information you need right when you need it. This is ideal for building custom dashboards or automating reporting tasks that require real-time updates.
+
+On the other hand, the YouTube Reporting API offers a more comprehensive approach. It's designed to handle large datasets, allowing developers to download bulk reports on topics like channel performance, audience demographics, and engagement trends.
+
+Both APIs rely on OAuth 2.0 for authentication, ensuring secure access to your data. You'll need to specify the necessary scopes when requesting access to YouTube Analytics or Reporting API data. This ensures only authorized users can view specific information.
+
+The ability to analyze data in this way opens up a world of possibilities for creators and developers alike. Imagine building tools that provide deeper insights into audience behavior, allowing you to tailor content strategies for maximum impact.
+
+
+This new era of YouTube analytics promises to reshape how we understand the platform's vast ecosystem. We'll be watching closely as these APIs are implemented by creators and developers.
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/Documentation_and_API/analytics_reporting_script_long_clean.txt b/docs/Documentation_and_API/analytics_reporting_script_long_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..51c06dcbb65888fdeaef08eee71a953247563eb4
--- /dev/null
+++ b/docs/Documentation_and_API/analytics_reporting_script_long_clean.txt
@@ -0,0 +1,15 @@
+Good evening. Today we're diving into the world of YouTube data analysis with a look at the new YouTube Analytics and Reporting APIs. These tools are designed to give developers unprecedented access to information about their content on the platform.
+
+For years, creators have relied on YouTube's built-in analytics for insights into their audience, but now they can take it a step further. The new APIs offer more granular control over data collection and analysis.
+
+The YouTube Analytics API allows users to query specific statistics like views, likes, and estimated minutes watched. Think of this as getting the precise information you need right when you need it. This is ideal for building custom dashboards or automating reporting tasks that require real-time updates.
+
+On the other hand, the YouTube Reporting API offers a more comprehensive approach. It's designed to handle large datasets, allowing developers to download bulk reports on topics like channel performance, audience demographics, and engagement trends.
+
+Both APIs rely on OAuth 2.0 for authentication, ensuring secure access to your data. You'll need to specify the necessary scopes when requesting access to YouTube Analytics or Reporting API data. This ensures only authorized users can view specific information.
+
+The ability to analyze data in this way opens up a world of possibilities for creators and developers alike. Imagine building tools that provide deeper insights into audience behavior, allowing you to tailor content strategies for maximum impact.
+
+This new era of YouTube analytics promises to reshape how we understand the platform's vast ecosystem. We'll be watching closely as these APIs are implemented by creators and developers.
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/Documentation_and_API/analytics_reporting_script_middle.md b/docs/Documentation_and_API/analytics_reporting_script_middle.md
new file mode 100644
index 0000000000000000000000000000000000000000..5f65063705fcb8b6c6a87f85565b65b9e7aa507b
--- /dev/null
+++ b/docs/Documentation_and_API/analytics_reporting_script_middle.md
@@ -0,0 +1,14 @@
+Good evening. Today, we're taking a look at the tools YouTube offers to understand your channel's performance. YouTube Analytics and Reporting APIs give you the power to dive deep into your data.
+
+Think of it like having a backstage pass to your own videos. You can see exactly how many people are watching, what they're liking, and even get insights on demographics. These APIs let you pull specific information about your content, or get an overview of your channel's performance in bulk.
+
+The YouTube Analytics API is perfect for those who want real-time data, like building a dashboard that updates as viewers watch. You can ask it questions like "how many views did this video get?" and even narrow down the results based on time frame and specific details about your videos.
+
+Now, if you need to analyze large amounts of data, the Reporting API comes in handy. This is your go-to for getting comprehensive reports on your channel's performance over a longer period. You can schedule these reports to be generated automatically.
+
+Both APIs require **OAuth 2.0** authentication. You won't be able to access analytics data with just an API key, you need the proper authorization. And remember, both APIs use the same security measures for all requests.
+
+For more information on how these APIs work and what they can do, check out the YouTube Developers Guide or try out the API Explorer.
+
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/Documentation_and_API/analytics_reporting_script_middle_clean.txt b/docs/Documentation_and_API/analytics_reporting_script_middle_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..9a6076013640dc47b2b319c75ddfd9f028998c33
--- /dev/null
+++ b/docs/Documentation_and_API/analytics_reporting_script_middle_clean.txt
@@ -0,0 +1,13 @@
+Good evening. Today, we're taking a look at the tools YouTube offers to understand your channel's performance. YouTube Analytics and Reporting APIs give you the power to dive deep into your data.
+
+Think of it like having a backstage pass to your own videos. You can see exactly how many people are watching, what they're liking, and even get insights on demographics. These APIs let you pull specific information about your content, or get an overview of your channel's performance in bulk.
+
+The YouTube Analytics API is perfect for those who want real-time data, like building a dashboard that updates as viewers watch. You can ask it questions like "how many views did this video get?" and even narrow down the results based on time frame and specific details about your videos.
+
+Now, if you need to analyze large amounts of data, the Reporting API comes in handy. This is your go-to for getting comprehensive reports on your channel's performance over a longer period. You can schedule these reports to be generated automatically.
+
+Both APIs require OAuth 2.0 authentication. You won't be able to access analytics data with just an API key, you need the proper authorization. And remember, both APIs use the same security measures for all requests.
+
+For more information on how these APIs work and what they can do, check out the YouTube Developers Guide or try out the API Explorer.
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/Documentation_and_API/analytics_reporting_script_short.md b/docs/Documentation_and_API/analytics_reporting_script_short.md
new file mode 100644
index 0000000000000000000000000000000000000000..bbe055deb6d236248df75d8e4b7d420537d1f8b9
--- /dev/null
+++ b/docs/Documentation_and_API/analytics_reporting_script_short.md
@@ -0,0 +1,11 @@
+Tonight's top tech story: YouTube is giving you more control over your data!
+
+They just launched two new APIs, the Analytics and Reporting API, allowing you to track your videos like never before.
+
+**Want real-time stats on your channel?** The Analytics API lets you pull specific data points about your content - think views, likes, comments. You can even filter by time range!
+**Need a bulk report for your business?** The Reporting API is your go-to for downloading big datasets like audience demographics and engagement trends.
+
+Both APIs require **OAuth 2.0 authentication**, so you'll need to set up your account first. You can find more details on the Google Developers website!
+
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/Documentation_and_API/analytics_reporting_script_short_clean.txt b/docs/Documentation_and_API/analytics_reporting_script_short_clean.txt
new file mode 100644
index 0000000000000000000000000000000000000000..5bd807777692b997864ecc93b8daae6a6779e655
--- /dev/null
+++ b/docs/Documentation_and_API/analytics_reporting_script_short_clean.txt
@@ -0,0 +1,11 @@
+Tonight's top tech story: YouTube is giving you more control over your data!
+
+They just launched two new APIs, the Analytics and Reporting API, allowing you to track your videos like never before.
+
+Want real-time stats on your channel? The Analytics API lets you pull specific data points about your content - think views, likes, comments. You can even filter by time range!
+
+Need a bulk report for your business? The Reporting API is your go-to for downloading big datasets like audience demographics and engagement trends.
+
+Both APIs require OAuth 2.0 authentication, so you'll need to set up your account first. You can find more details on the Google Developers website!
+
+Reporting for The Android Times.
\ No newline at end of file
diff --git a/docs/Documentation_and_API/documentation.md b/docs/Documentation_and_API/documentation.md
new file mode 100644
index 0000000000000000000000000000000000000000..4c73f390950017b05847e4f5bb17006c0551c63c
--- /dev/null
+++ b/docs/Documentation_and_API/documentation.md
@@ -0,0 +1,88 @@
+Title: YouTube Developer Documentation | Google for Developers
+
+Source: https://developers.google.com/youtube/documentation?hl=en
+
+---
+
+[Skip to main content](https://developers.google.com/youtube/documentation?hl=en#main-content)
+
+[YouTube](https://developers.google.com/youtube)
+
+- [English](https://developers.google.com/youtube/documentation?hl=en)
+- [Deutsch](https://developers.google.com/youtube/documentation?hl=en)
+- [Español](https://developers.google.com/youtube/documentation?hl=en)
+- [Español – América Latina](https://developers.google.com/youtube/documentation?hl=en)
+- [Français](https://developers.google.com/youtube/documentation?hl=en)
+- [Indonesia](https://developers.google.com/youtube/documentation?hl=en)
+- [Italiano](https://developers.google.com/youtube/documentation?hl=en)
+- [Polski](https://developers.google.com/youtube/documentation?hl=en)
+- [Português – Brasil](https://developers.google.com/youtube/documentation?hl=en)
+- [Tiếng Việt](https://developers.google.com/youtube/documentation?hl=en)
+- [Türkçe](https://developers.google.com/youtube/documentation?hl=en)
+- [Русский](https://developers.google.com/youtube/documentation?hl=en)
+- [עברית](https://developers.google.com/youtube/documentation?hl=en)
+- [العربيّة](https://developers.google.com/youtube/documentation?hl=en)
+- [فارسی](https://developers.google.com/youtube/documentation?hl=en)
+- [हिंदी](https://developers.google.com/youtube/documentation?hl=en)
+- [বাংলা](https://developers.google.com/youtube/documentation?hl=en)
+- [ภาษาไทย](https://developers.google.com/youtube/documentation?hl=en)
+- [中文 – 简体](https://developers.google.com/youtube/documentation?hl=en)
+- [中文 – 繁體](https://developers.google.com/youtube/documentation?hl=en)
+- [日本語](https://developers.google.com/youtube/documentation?hl=en)
+- [한국어](https://developers.google.com/youtube/documentation?hl=en)
+
+[Home](https://developers.google.com/youtube)
+[Guides](https://developers.google.com/youtube/documentation)
+[Samples](https://developers.google.com/youtube/code_samples)
+[Terms](https://developers.google.com/youtube/terms/api-services-terms-of-service)
+[Internal](https://developers.google.com/youtube/internal/partner)
+
+- YouTube Players
+- [IFrame Player API](https://developers.google.com/youtube/iframe_api_reference)
+- [iOS Helper Library](https://developers.google.com/youtube/v3/guides/ios_youtube_helper)
+- [Player Parameters](https://developers.google.com/youtube/player_parameters)
+
+- YouTube Data and Resources
+- [YouTube Data API](https://developers.google.com/youtube/v3)
+- [YouTube Analytics API](https://developers.google.com/youtube/analytics)
+- [YouTube Live Streaming API](https://developers.google.com/youtube/v3/live)
+- [Client Libraries](https://developers.google.com/youtube/code)
+
+- YouTube Tools and Widgets
+- [YouTube Subscribe Buttons](https://developers.google.com/youtube/subscribe)
+
+YouTube offers APIs and tools to integrate YouTube functionalities, like video playback and channel subscriptions, into your website or app.
+Developers can utilize YouTube Player APIs for Android, iOS, and web embeds to customize video playback experiences.
+The YouTube Data API enables access to features like video uploads, playlist management, and data retrieval for analytics and live streaming.
+Resources like the YouTube Subscribe Button facilitate user engagement and channel growth directly within external platforms.
+
+[YouTube](https://www.youtube.com) has a number of APIs and tools that let you embed YouTube functionality into your own website and applications.
+
+### YouTube Player and Widgets
+[IFrame Player API Reference](https://developers.google.com/youtube/iframe_api_reference)
+[iOS Helper Library](https://developers.google.com/youtube/v3/guides/ios_youtube_helper)
+[YouTube Player Parameters](https://developers.google.com/youtube/player_parameters)
+[YouTube Subscribe Button](https://developers.google.com/youtube/subscribe)
+
+### YouTube Data and Resources
+[YouTube Data API](https://developers.google.com/youtube/v3)
+[YouTube Live Streaming API](https://developers.google.com/youtube/v3/live)
+[YouTube Analytics API](https://developers.google.com/youtube/analytics)
+[YouTube Reporting API](https://developers.google.com/youtube/reporting)
+[YouTube Content ID API](https://developers.google.com/youtube/partner)
+
+### Tools
+- [Google APIs Explorer](https://developers.google.com/apis-explorer/#p/)
+- [YouTube Player Demo](https://developers.google.com/youtube/youtube_player_demo)
+- [Configure a Subscribe Button](https://developers.google.com/youtube/youtube_subscribe_button)
+
+### Issue Tracker
+- [File a bug](https://issuetracker.google.com/issues/new?component=186600&template=874803)
+- [Request a feature](https://issuetracker.google.com/issues/new?component=186600&template=874803)
+- [See open issues](https://issuetracker.google.com/issues?q=componentid:186600)
+
+### Product Info
+- [Terms of Service](https://developers.google.com/youtube/terms/api-services-terms-of-service)
+- [Developer Policies](https://developers.google.com/youtube/terms/developer-policies)
+- [Required Minimum Functionality](https://developers.google.com/youtube/terms/required-minimum-functionality)
+- [Branding Guidelines](https://developers.google.com/youtube/terms/branding-guidelines)
diff --git a/docs/Documentation_and_API/documentation.rag_ingested b/docs/Documentation_and_API/documentation.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..146f95e30ec2a6684ba63cb79dbd286ec8ce8263
--- /dev/null
+++ b/docs/Documentation_and_API/documentation.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "4234aa2bd1649582", "ingested_at": "2026-06-30T09:10:24.940542", "content_hash": "4234aa2bd1649582"}
\ No newline at end of file
diff --git a/docs/Documentation_and_API/documentation_raw.md b/docs/Documentation_and_API/documentation_raw.md
new file mode 100644
index 0000000000000000000000000000000000000000..98e2f8ca267087fb19c7bba4f356431745fa703e
--- /dev/null
+++ b/docs/Documentation_and_API/documentation_raw.md
@@ -0,0 +1,668 @@
+Title: YouTube Developer Documentation | Google for Developers
+
+Source: https://developers.google.com/youtube/documentation?hl=en
+
+---
+
+[Skip to main content](https://developers.google.com/youtube/documentation?hl=en#main-content)
+-
+
+
+
+
+
+
+ [YouTube](https://developers.google.com/youtube)
+
+
+
+
+[YouTube](https://developers.google.com/youtube)
+-
+ [English](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Deutsch](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Español](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Español – América Latina](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Français](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Indonesia](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Italiano](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Polski](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Português – Brasil](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Tiếng Việt](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Türkçe](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Русский](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [עברית](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [العربيّة](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [فارسی](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [हिंदी](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [বাংলা](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [ภาษาไทย](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [中文 – 简体](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [中文 – 繁體](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [日本語](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [한국어](https://developers.google.com/youtube/documentation?hl=en)
+
+[English](https://developers.google.com/youtube/documentation?hl=en)
+[Deutsch](https://developers.google.com/youtube/documentation?hl=en)
+[Español](https://developers.google.com/youtube/documentation?hl=en)
+[Español – América Latina](https://developers.google.com/youtube/documentation?hl=en)
+[Français](https://developers.google.com/youtube/documentation?hl=en)
+[Indonesia](https://developers.google.com/youtube/documentation?hl=en)
+[Italiano](https://developers.google.com/youtube/documentation?hl=en)
+[Polski](https://developers.google.com/youtube/documentation?hl=en)
+[Português – Brasil](https://developers.google.com/youtube/documentation?hl=en)
+[Tiếng Việt](https://developers.google.com/youtube/documentation?hl=en)
+[Türkçe](https://developers.google.com/youtube/documentation?hl=en)
+[Русский](https://developers.google.com/youtube/documentation?hl=en)
+[עברית](https://developers.google.com/youtube/documentation?hl=en)
+[العربيّة](https://developers.google.com/youtube/documentation?hl=en)
+[فارسی](https://developers.google.com/youtube/documentation?hl=en)
+[हिंदी](https://developers.google.com/youtube/documentation?hl=en)
+[বাংলা](https://developers.google.com/youtube/documentation?hl=en)
+[ภาษาไทย](https://developers.google.com/youtube/documentation?hl=en)
+[中文 – 简体](https://developers.google.com/youtube/documentation?hl=en)
+[中文 – 繁體](https://developers.google.com/youtube/documentation?hl=en)
+[日本語](https://developers.google.com/youtube/documentation?hl=en)
+[한국어](https://developers.google.com/youtube/documentation?hl=en)
+[Home](https://developers.google.com/youtube)
+[Guides](https://developers.google.com/youtube/documentation)
+[Samples](https://developers.google.com/youtube/code_samples)
+[Terms](https://developers.google.com/youtube/terms/api-services-terms-of-service)
+[Internal](https://developers.google.com/youtube/internal/partner)
+-
+
+
+
+
+
+
+ [YouTube](https://developers.google.com/youtube)
+
+
+
+
+[YouTube](https://developers.google.com/youtube)
+-
+
+
+ [Home](https://developers.google.com/youtube)
+
+
+
+-
+
+
+ [Guides](https://developers.google.com/youtube/documentation)
+
+
+
+-
+
+
+ [Samples](https://developers.google.com/youtube/code_samples)
+
+
+
+-
+
+
+ [Terms](https://developers.google.com/youtube/terms/api-services-terms-of-service)
+
+
+
+-
+
+[Internal](https://developers.google.com/youtube/internal/partner)
+
+
+
+[Home](https://developers.google.com/youtube)
+[Guides](https://developers.google.com/youtube/documentation)
+[Samples](https://developers.google.com/youtube/code_samples)
+[Terms](https://developers.google.com/youtube/terms/api-services-terms-of-service)
+[Internal](https://developers.google.com/youtube/internal/partner)
+- [YouTube Developer Documentation](https://developers.google.com/youtube/documentation)
+-
+ YouTube Players
+
+- [IFrame Player API](https://developers.google.com/youtube/iframe_api_reference)
+- [iOS Helper Library](https://developers.google.com/youtube/v3/guides/ios_youtube_helper)
+- [Player Parameters](https://developers.google.com/youtube/player_parameters)
+-
+ YouTube Data and Resources
+
+- [YouTube Data API](https://developers.google.com/youtube/v3)
+- [YouTube Analytics API](https://developers.google.com/youtube/analytics)
+- [YouTube Live Streaming API](https://developers.google.com/youtube/v3/live)
+- [Client Libraries](https://developers.google.com/youtube/code)
+-
+ YouTube Tools and Widgets
+
+- [YouTube Subscribe Buttons](https://developers.google.com/youtube/subscribe)
+[YouTube Developer Documentation](https://developers.google.com/youtube/documentation)
+[IFrame Player API](https://developers.google.com/youtube/iframe_api_reference)
+[iOS Helper Library](https://developers.google.com/youtube/v3/guides/ios_youtube_helper)
+[Player Parameters](https://developers.google.com/youtube/player_parameters)
+[YouTube Data API](https://developers.google.com/youtube/v3)
+[YouTube Analytics API](https://developers.google.com/youtube/analytics)
+[YouTube Live Streaming API](https://developers.google.com/youtube/v3/live)
+[Client Libraries](https://developers.google.com/youtube/code)
+[YouTube Subscribe Buttons](https://developers.google.com/youtube/subscribe)
+-
+
+
+
+
+
+ [Home](https://developers.google.com/)
+
+
+
+
+-
+
+
+
+
+
+
+
+
+ [Products](https://developers.google.com/products)
+
+
+
+
+-
+
+
+
+
+
+
+
+
+ [YouTube](https://developers.google.com/youtube)
+
+
+
+
+-
+
+
+
+
+
+
+
+
+ [Guides](https://developers.google.com/youtube/documentation)
+
+
+
+
+[Home](https://developers.google.com/)
+[Products](https://developers.google.com/products)
+[YouTube](https://developers.google.com/youtube)
+[Guides](https://developers.google.com/youtube/documentation)
+
+-
+ YouTube offers APIs and tools to integrate YouTube functionalities, like video playback and channel subscriptions, into your website or app.
+
+
+-
+ Developers can utilize YouTube Player APIs for Android, iOS, and web embeds to customize video playback experiences.
+
+
+-
+ The YouTube Data API enables access to features like video uploads, playlist management, and data retrieval for analytics and live streaming.
+
+
+-
+ Resources like the YouTube Subscribe Button facilitate user engagement and channel growth directly within external platforms.
+
+
+YouTube offers APIs and tools to integrate YouTube functionalities, like video playback and channel subscriptions, into your website or app.
+Developers can utilize YouTube Player APIs for Android, iOS, and web embeds to customize video playback experiences.
+The YouTube Data API enables access to features like video uploads, playlist management, and data retrieval for analytics and live streaming.
+Resources like the YouTube Subscribe Button facilitate user engagement and channel growth directly within external platforms.
+[YouTube](https://www.youtube.com) has a number of APIs and tools that let you embed YouTube functionality into your own website and applications.
+
+### YouTube Player and Widgets
+[IFrame Player API Reference](https://developers.google.com/youtube/iframe_api_reference)
+[iOS Helper Library](https://developers.google.com/youtube/v3/guides/ios_youtube_helper)
+[YouTube Player Parameters](https://developers.google.com/youtube/player_parameters)
+[YouTube Subscribe Button](https://developers.google.com/youtube/subscribe)
+
+### YouTube Data and Resources
+Use the following table to find the YouTube API that best fits your needs:
+[YouTube Data API](https://developers.google.com/youtube/v3)
+[YouTube Live Streaming API](https://developers.google.com/youtube/v3/live)
+[YouTube Analytics API](https://developers.google.com/youtube/analytics)
+[YouTube Reporting API](https://developers.google.com/youtube/reporting)
+[YouTube Content ID API](https://developers.google.com/youtube/partner)
+[details](https://support.google.com/youtube/answer/72851)
+Except as otherwise noted, the content of this page is licensed under the [Creative Commons Attribution 4.0 License](https://creativecommons.org/licenses/by/4.0/), and code samples are licensed under the [Apache 2.0 License](https://www.apache.org/licenses/LICENSE-2.0). For details, see the [Google Developers Site Policies](https://developers.google.com/site-policies). Java is a registered trademark of Oracle and/or its affiliates.
+Last updated 2026-04-28 UTC.
+-
+ [Blog](https://blog.youtube)
+ The latest news on the YouTube blog
+
+-
+ [GitHub](https://github.com/youtube/api-samples)
+ Find API code samples and other YouTube open-source projects.
+
+-
+ [Issue Tracker](https://issuetracker.google.com/issues/new?component=186600&template=874803)
+ Something wrong? Send us a bug report!
+
+-
+ [Stack Overflow](http://stackoverflow.com/questions/ask?tags=youtube-api)
+ Ask a question under the youtube-api tag
+
+-
+ [YouTube Researcher Program](https://research.youtube/)
+ For researchers interested in using data from YouTube’s global API
+
+[Blog](https://blog.youtube)
+[GitHub](https://github.com/youtube/api-samples)
+[Issue Tracker](https://issuetracker.google.com/issues/new?component=186600&template=874803)
+[Stack Overflow](http://stackoverflow.com/questions/ask?tags=youtube-api)
+[YouTube Researcher Program](https://research.youtube/)
+-
+ Tools
+
+
+
+
+ [Google APIs Explorer](https://developers.google.com/apis-explorer/#p/)
+
+
+
+
+
+
+ [YouTube Player Demo](https://developers.google.com/youtube/youtube_player_demo)
+
+
+
+
+
+
+ [Configure a Subscribe Button](https://developers.google.com/youtube/youtube_subscribe_button)
+
+
+
+
+
+
+-
+
+ [Google APIs Explorer](https://developers.google.com/apis-explorer/#p/)
+
+
+
+-
+
+ [YouTube Player Demo](https://developers.google.com/youtube/youtube_player_demo)
+
+
+
+-
+
+ [Configure a Subscribe Button](https://developers.google.com/youtube/youtube_subscribe_button)
+
+
+
+-
+ Issue Tracker
+
+
+
+
+ [File a bug](https://issuetracker.google.com/issues/new?component=186600&template=874803)
+
+
+
+
+
+
+ [Request a feature](https://issuetracker.google.com/issues/new?component=186600&template=874803)
+
+
+
+
+
+
+ [See open issues](https://issuetracker.google.com/issues?q=componentid:186600)
+
+
+
+
+
+
+-
+
+ [File a bug](https://issuetracker.google.com/issues/new?component=186600&template=874803)
+
+
+
+-
+
+ [Request a feature](https://issuetracker.google.com/issues/new?component=186600&template=874803)
+
+
+
+-
+
+ [See open issues](https://issuetracker.google.com/issues?q=componentid:186600)
+
+
+
+-
+ Product Info
+
+
+
+
+ [Terms of Service](https://developers.google.com/youtube/terms/api-services-terms-of-service)
+
+
+
+
+
+
+ [Developer Policies](https://developers.google.com/youtube/terms/developer-policies)
+
+
+
+
+
+
+ [Required Minimum Functionality](https://developers.google.com/youtube/terms/required-minimum-functionality)
+
+
+
+
+
+
+ [Branding Guidelines](https://developers.google.com/youtube/terms/branding-guidelines)
+
+
+
+
+
+
+-
+
+ [Terms of Service](https://developers.google.com/youtube/terms/api-services-terms-of-service)
+
+
+
+-
+
+ [Developer Policies](https://developers.google.com/youtube/terms/developer-policies)
+
+
+
+-
+
+ [Required Minimum Functionality](https://developers.google.com/youtube/terms/required-minimum-functionality)
+
+
+
+-
+
+ [Branding Guidelines](https://developers.google.com/youtube/terms/branding-guidelines)
+
+### Tools
+-
+
+ [Google APIs Explorer](https://developers.google.com/apis-explorer/#p/)
+
+
+
+-
+
+ [YouTube Player Demo](https://developers.google.com/youtube/youtube_player_demo)
+
+
+
+-
+
+ [Configure a Subscribe Button](https://developers.google.com/youtube/youtube_subscribe_button)
+
+
+
+[Google APIs Explorer](https://developers.google.com/apis-explorer/#p/)
+[YouTube Player Demo](https://developers.google.com/youtube/youtube_player_demo)
+[Configure a Subscribe Button](https://developers.google.com/youtube/youtube_subscribe_button)
+
+### Issue Tracker
+-
+
+ [File a bug](https://issuetracker.google.com/issues/new?component=186600&template=874803)
+
+
+
+-
+
+ [Request a feature](https://issuetracker.google.com/issues/new?component=186600&template=874803)
+
+
+
+-
+
+ [See open issues](https://issuetracker.google.com/issues?q=componentid:186600)
+
+
+
+[File a bug](https://issuetracker.google.com/issues/new?component=186600&template=874803)
+[Request a feature](https://issuetracker.google.com/issues/new?component=186600&template=874803)
+[See open issues](https://issuetracker.google.com/issues?q=componentid:186600)
+
+-
+
+ [Terms of Service](https://developers.google.com/youtube/terms/api-services-terms-of-service)
+
+
+
+-
+
+ [Developer Policies](https://developers.google.com/youtube/terms/developer-policies)
+
+
+
+-
+
+ [Required Minimum Functionality](https://developers.google.com/youtube/terms/required-minimum-functionality)
+
+
+
+-
+
+ [Branding Guidelines](https://developers.google.com/youtube/terms/branding-guidelines)
+
+
+
+[Terms of Service](https://developers.google.com/youtube/terms/api-services-terms-of-service)
+[Developer Policies](https://developers.google.com/youtube/terms/developer-policies)
+[Required Minimum Functionality](https://developers.google.com/youtube/terms/required-minimum-functionality)
+[Branding Guidelines](https://developers.google.com/youtube/terms/branding-guidelines)
+-
+ [Android](https://developer.android.com)
+
+-
+ [Chrome](https://developer.chrome.com/home)
+
+-
+ [Firebase](https://firebase.google.com)
+
+-
+ [Google Cloud Platform](https://cloud.google.com)
+
+-
+ [Google AI](https://ai.google.dev/)
+
+-
+ [All products](https://developers.google.com/products)
+
+[Android](https://developer.android.com)
+[Chrome](https://developer.chrome.com/home)
+[Firebase](https://firebase.google.com)
+[Google Cloud Platform](https://cloud.google.com)
+[Google AI](https://ai.google.dev/)
+[All products](https://developers.google.com/products)
+-
+
+
+ [Terms](https://developers.google.com/terms/site-terms)
+
+
+-
+
+
+ [Privacy](https://policies.google.com/privacy)
+
+
+-
+
+
+ [Manage cookies](https://developers.google.com/youtube/documentation?hl=en)
+
+
+[Terms](https://developers.google.com/terms/site-terms)
+[Privacy](https://policies.google.com/privacy)
+[Manage cookies](https://developers.google.com/youtube/documentation?hl=en)
+-
+ [English](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Deutsch](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Español](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Español – América Latina](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Français](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Indonesia](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Italiano](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Polski](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Português – Brasil](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Tiếng Việt](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Türkçe](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [Русский](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [עברית](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [العربيّة](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [فارسی](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [हिंदी](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [বাংলা](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [ภาษาไทย](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [中文 – 简体](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [中文 – 繁體](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [日本語](https://developers.google.com/youtube/documentation?hl=en)
+
+-
+ [한국어](https://developers.google.com/youtube/documentation?hl=en)
+
+[English](https://developers.google.com/youtube/documentation?hl=en)
+[Deutsch](https://developers.google.com/youtube/documentation?hl=en)
+[Español](https://developers.google.com/youtube/documentation?hl=en)
+[Español – América Latina](https://developers.google.com/youtube/documentation?hl=en)
+[Français](https://developers.google.com/youtube/documentation?hl=en)
+[Indonesia](https://developers.google.com/youtube/documentation?hl=en)
+[Italiano](https://developers.google.com/youtube/documentation?hl=en)
+[Polski](https://developers.google.com/youtube/documentation?hl=en)
+[Português – Brasil](https://developers.google.com/youtube/documentation?hl=en)
+[Tiếng Việt](https://developers.google.com/youtube/documentation?hl=en)
+[Türkçe](https://developers.google.com/youtube/documentation?hl=en)
+
+[Русский](https://developers.google.com/youtube/documentation?hl=en)
+[עברית](https://developers.google.com/youtube/documentation?hl=en)
+[العربيّة](https://developers.google.com/youtube/documentation?hl=en)
+[فارسی](https://developers.google.com/youtube/documentation?hl=en)
+[हिंदी](https://developers.google.com/youtube/documentation?hl=en)
+[বাংলা](https://developers.google.com/youtube/documentation?hl=en)
+[ภาษาไทย](https://developers.google.com/youtube/documentation?hl=en)
+[中文 – 简体](https://developers.google.com/youtube/documentation?hl=en)
+[中文 – 繁體](https://developers.google.com/youtube/documentation?hl=en)
+[日本語](https://developers.google.com/youtube/documentation?hl=en)
+[한국어](https://developers.google.com/youtube/documentation?hl=en)
+
diff --git a/docs/Documentation_and_API/documentation_raw.rag_ingested b/docs/Documentation_and_API/documentation_raw.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..66ee2d03ea456363b463ba23ec78b5280801679f
--- /dev/null
+++ b/docs/Documentation_and_API/documentation_raw.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "5e577c3dd29df712", "ingested_at": "2026-06-30T09:10:30.359987", "content_hash": "5e577c3dd29df712"}
\ No newline at end of file
diff --git a/docs/Documentation_and_API/iframe_api_reference.md b/docs/Documentation_and_API/iframe_api_reference.md
new file mode 100644
index 0000000000000000000000000000000000000000..f988e5ffecd55d621baea66314ba59dafe7d79b2
--- /dev/null
+++ b/docs/Documentation_and_API/iframe_api_reference.md
@@ -0,0 +1,93 @@
+# YouTube IFrame Player API Reference (English)
+
+Source: https://developers.google.com/youtube/iframe_api_reference?hl=en
+
+The IFrame player API lets you embed a YouTube video player on your website and control the player using JavaScript.
+
+## Getting Started
+
+### Browser Requirements
+The user's browser must support the HTML5 `postMessage` feature.
+
+### Minimum Viewport Size
+Embedded players must have a viewport that is at least 200px by 200px. Recommended size for 16:9 players is 480px by 270px.
+
+## Implementation Example
+
+```html
+
+
+
+
+
+
+
+
+
+```
+
+## Functions
+
+### Queueing Functions
+- `cueVideoById`: Loads thumbnail and prepares player.
+- `loadVideoById`: Loads and plays video.
+- `cuePlaylist`: Loads playlist thumbnail and prepares player.
+- `loadPlaylist`: Loads and plays playlist.
+
+### Playback Controls
+- `player.playVideo()`
+- `player.pauseVideo()`
+- `player.stopVideo()`
+- `player.seekTo(seconds, allowSeekAhead)`
+- `player.nextVideo()`
+- `player.previousVideo()`
+- `player.mute()`
+- `player.unMute()`
+- `player.setVolume(volume)`
+- `player.getVolume()`
+
+### Events
+- `onReady`
+- `onStateChange`
+- `onPlaybackQualityChange`
+- `onPlaybackRateChange`
+- `onError`
+- `onApiChange`
diff --git a/docs/Documentation_and_API/iframe_api_reference.rag_ingested b/docs/Documentation_and_API/iframe_api_reference.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..935267151011e7ec51017a3f3471170af7b874d2
--- /dev/null
+++ b/docs/Documentation_and_API/iframe_api_reference.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "7231233867bf1fd3", "ingested_at": "2026-06-30T09:10:32.090490", "content_hash": "7231233867bf1fd3"}
\ No newline at end of file
diff --git a/docs/Documentation_and_API/player_parameters.md b/docs/Documentation_and_API/player_parameters.md
new file mode 100644
index 0000000000000000000000000000000000000000..a3e3cb2fd986141dd65f96a02dbf84451629fc30
--- /dev/null
+++ b/docs/Documentation_and_API/player_parameters.md
@@ -0,0 +1,33 @@
+# YouTube Player Parameters (English)
+
+Source: https://developers.google.com/youtube/player_parameters?hl=en
+
+By appending parameters to the IFrame URL, you can customize the playback experience in your application.
+
+## Common Parameters
+
+| Parameter | Type | Description |
+| :--- | :--- | :--- |
+| `autoplay` | 0 or 1 | Automatically plays the video once the player loads. |
+| `cc_load_policy` | 1 | Causes closed captions to be shown by default. |
+| `controls` | 0 or 1 | Whether to display player controls. |
+| `disablekb` | 0 or 1 | Whether to disable keyboard controls. |
+| `enablejsapi` | 0 or 1 | Enables the player to be controlled via the IFrame Player API. |
+| `end` | number | Time in seconds when the video should stop playing. |
+| `fs` | 0 or 1 | Whether to display the fullscreen button. |
+| `hl` | string | The player's interface language (ISO 639-1 code). |
+| `iv_load_policy` | 1 or 3 | Whether to show video annotations. |
+| `list` | string | Identifies the list of content to be played (playlist ID). |
+| `listType` | string | The type of list being loaded (`playlist` or `user_uploads`). |
+| `loop` | 0 or 1 | Whether to play the initial video (or playlist) again and again. |
+| `origin` | string | Security measure; specify your domain. |
+| `playlist` | string | Comma-separated list of video IDs to play. |
+| `playsinline` | 0 or 1 | Whether to play inline on iOS. |
+| `rel` | 0 or 1 | Whether to show related videos at the end (limited to same channel since 2018). |
+| `start` | number | Time in seconds when the video should start playing. |
+
+## Deprecated Parameters
+- `modestbranding` (No longer supported as of 2023)
+- `showinfo`
+- `autohide`
+- `theme`
diff --git a/docs/Documentation_and_API/player_parameters.rag_ingested b/docs/Documentation_and_API/player_parameters.rag_ingested
new file mode 100644
index 0000000000000000000000000000000000000000..9cd752416bd07ca2c5b5ea44799c4c6e04006e73
--- /dev/null
+++ b/docs/Documentation_and_API/player_parameters.rag_ingested
@@ -0,0 +1 @@
+{"doc_id": "1ed7c3408f17af7a", "ingested_at": "2026-06-30T09:10:34.453460", "content_hash": "1ed7c3408f17af7a"}
\ No newline at end of file
diff --git a/docs/Documentation_and_API/player_parameters_raw.md b/docs/Documentation_and_API/player_parameters_raw.md
new file mode 100644
index 0000000000000000000000000000000000000000..f3f1593a882551ee85dcddb407f408881d07dcc8
--- /dev/null
+++ b/docs/Documentation_and_API/player_parameters_raw.md
@@ -0,0 +1,2279 @@
+Title: YouTube Embedded Players and Player Parameters | YouTube IFrame Player API | Google for Developers
+
+Source: https://developers.google.com/youtube/player_parameters?hl=en
+
+---
+
+[Skip to main content](https://developers.google.com/youtube/player_parameters?hl=en#main-content)
+-
+
+
+
+
+
+
+ [YouTube](https://developers.google.com/youtube)
+
+
+
+
+-
+
+
+
+
+
+
+
+
+ [IFrame Player API](https://developers.google.com/youtube/iframe_api_reference)
+
+
+
+
+[YouTube](https://developers.google.com/youtube)
+[IFrame Player API](https://developers.google.com/youtube/iframe_api_reference)
+-
+ [English](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [Deutsch](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [Español](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [Español – América Latina](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [Français](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [Indonesia](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [Italiano](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [Polski](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [Português – Brasil](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [Tiếng Việt](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [Türkçe](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [Русский](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [עברית](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [العربيّة](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [فارسی](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [हिंदी](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [বাংলা](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [ภาษาไทย](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [中文 – 简体](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [中文 – 繁體](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [日本語](https://developers.google.com/youtube/player_parameters?hl=en)
+
+-
+ [한국어](https://developers.google.com/youtube/player_parameters?hl=en)
+
+[English](https://developers.google.com/youtube/player_parameters?hl=en)
+[Deutsch](https://developers.google.com/youtube/player_parameters?hl=en)
+[Español](https://developers.google.com/youtube/player_parameters?hl=en)
+[Español – América Latina](https://developers.google.com/youtube/player_parameters?hl=en)
+[Français](https://developers.google.com/youtube/player_parameters?hl=en)
+[Indonesia](https://developers.google.com/youtube/player_parameters?hl=en)
+[Italiano](https://developers.google.com/youtube/player_parameters?hl=en)
+[Polski](https://developers.google.com/youtube/player_parameters?hl=en)
+[Português – Brasil](https://developers.google.com/youtube/player_parameters?hl=en)
+[Tiếng Việt](https://developers.google.com/youtube/player_parameters?hl=en)
+[Türkçe](https://developers.google.com/youtube/player_parameters?hl=en)
+[Русский](https://developers.google.com/youtube/player_parameters?hl=en)
+[עברית](https://developers.google.com/youtube/player_parameters?hl=en)
+[العربيّة](https://developers.google.com/youtube/player_parameters?hl=en)
+[فارسی](https://developers.google.com/youtube/player_parameters?hl=en)
+[हिंदी](https://developers.google.com/youtube/player_parameters?hl=en)
+[বাংলা](https://developers.google.com/youtube/player_parameters?hl=en)
+[ภาษาไทย](https://developers.google.com/youtube/player_parameters?hl=en)
+[中文 – 简体](https://developers.google.com/youtube/player_parameters?hl=en)
+[中文 – 繁體](https://developers.google.com/youtube/player_parameters?hl=en)
+[日本語](https://developers.google.com/youtube/player_parameters?hl=en)
+[한국어](https://developers.google.com/youtube/player_parameters?hl=en)
+[Reference](https://developers.google.com/youtube/iframe_api_reference)
+[Samples](https://developers.google.com/youtube/youtube_player_demo)
+[Support](https://developers.google.com/youtube/players/support)
+-
+
+
+
+
+
+
+ [YouTube](https://developers.google.com/youtube)
+
+
+
+
+-
+
+
+
+
+
+
+
+
+ [IFrame Player API](https://developers.google.com/youtube/iframe_api_reference)
+
+
+
+
+[YouTube](https://developers.google.com/youtube)
+
+[IFrame Player API](https://developers.google.com/youtube/iframe_api_reference)
+-
+
+
+ [Reference](https://developers.google.com/youtube/iframe_api_reference)
+
+
+
+-
+
+
+ [Samples](https://developers.google.com/youtube/youtube_player_demo)
+
+
+
+-
+
+
+ [Support](https://developers.google.com/youtube/players/support)
+
+
+
+[Reference](https://developers.google.com/youtube/iframe_api_reference)
+[Samples](https://developers.google.com/youtube/youtube_player_demo)
+[Support](https://developers.google.com/youtube/players/support)
+- [IFrame API](https://developers.google.com/youtube/iframe_api_reference)
+- [Player Parameters](https://developers.google.com/youtube/player_parameters)
+-
+ Mobile APIs
+
+- [iOS Helper Library](https://developers.google.com/youtube/v3/guides/ios_youtube_helper)
+[IFrame API](https://developers.google.com/youtube/iframe_api_reference)
+[Player Parameters](https://developers.google.com/youtube/player_parameters)
+[iOS Helper Library](https://developers.google.com/youtube/v3/guides/ios_youtube_helper)
+-
+
+
+
+
+
+ [Home](https://developers.google.com/)
+
+
+
+
+-
+
+
+
+
+
+
+
+
+ [Products](https://developers.google.com/products)
+
+
+
+
+-
+
+
+
+
+
+
+
+
+ [YouTube](https://developers.google.com/youtube)
+
+
+
+
+-
+
+
+
+
+
+
+
+
+ [IFrame Player API](https://developers.google.com/youtube/iframe_api_reference)
+
+
+
+
+[Home](https://developers.google.com/)
+[Products](https://developers.google.com/products)
+[YouTube](https://developers.google.com/youtube)
+[IFrame Player API](https://developers.google.com/youtube/iframe_api_reference)
+
+## Page Summary
+-
+ YouTube players can be embedded using either an